0 files changed, 0 insertions, 0 deletions
diff --git a/.cvsignore b/.cvsignore
deleted file mode 100644
index 9fd96ca149..0000000000
--- a/.cvsignore
+++ /dev/null
@@ -1,58 +0,0 @@
-*.bak
-*.orig
-*.rej
-*.sav
-*~
-.ccmalloc
-.ppack
-.ext
-.git
-.svn
-.rbconfig.time
-COPYING.LIB
-ChangeLog.pre-alpha
-ChangeLog.pre1_1
-Makefile
-README.fat-patch
-README.v6
-README.atheos
-archive
-autom4te*.cache
-automake
-beos
-config.cache
-config.h
-config.h.in
-config.log
-config.status
-configure
-foo.rb
-libruby.so.*
-miniruby
-miniruby.elhash
-miniruby.elhash2
-miniruby.orig2
-miniruby.plhash
-miniruby.plhash2
-modex.rb
-newdate.rb
-newver.rb
-parse.c
-parse.y.try
-pitest.rb
-ppack
-preview
-rbconfig.rb
-rename2.h
-repack
-riscos
-rubicon
-ruby
-ruby-man.rd.gz
-rubyunit
-st.c.power
-this that
-tmp
-web
-y.output
-y.tab.c
diff --git a/.document b/.document
index 230c50e387..2116ca6d26 100644
--- a/.document
+++ b/.document
@@ -5,12 +5,32 @@
 
 # Process all the C source files
 *.c
+*.y
 
-# the lib/ directory (which has its own .document file)
+# prelude
+prelude.rb
 
-lib
+rbconfig.rb
 
+# the lib/ directory (which has its own .document file)
+lib
 
 # and some of the ext/ directory (which has its own .document file)
-
 ext
+
+# rdoc files
+NEWS
+
+README.md
+README.ja.md
+
+COPYING
+COPYING.ja
+CONTRIBUTING.md
+
+LEGAL
+
+# win32/README.win32 linked from README.md
+win32
+
+doc
diff --git a/.editorconfig b/.editorconfig
new file mode 100644
index 0000000000..cb2729fff9
--- /dev/null
+++ b/.editorconfig
@@ -0,0 +1,27 @@
+root = true
+
+[*]
+end_of_line = lf
+indent_size = 4
+indent_style = space
+insert_final_newline = true
+tab_width = 8
+trim_trailing_whitespace = true
+
+[*.bat]
+end_of_line = crlf
+
+[*.gemspec]
+indent_size = 2
+
+[*.rb]
+indent_size = 2
+
+[*.yml]
+indent_size = 2
+
+[{*[Mm]akefile*,*.mak,*.mk,depend}]
+indent_style = tab
+
+[reg*.[ch]]
+indent_size = 2
diff --git a/.gdbinit b/.gdbinit
new file mode 100644
index 0000000000..1806196a42
--- /dev/null
+++ b/.gdbinit
@@ -0,0 +1,1346 @@
+set startup-with-shell off
+
+define hook-run
+  set $color_type = 0
+  set $color_highlite = 0
+  set $color_end = 0
+end
+
+define ruby_gdb_init
+  if !$color_type
+    set $color_type = "\033[31m"
+  end
+  if !$color_highlite
+    set $color_highlite = "\033[36m"
+  end
+  if !$color_end
+    set $color_end = "\033[m"
+  end
+  if ruby_dummy_gdb_enums.special_consts
+  end
+end
+
+# set prompt \033[36m(gdb)\033[m\040
+
+define rp
+  ruby_gdb_init
+  if (VALUE)($arg0) & RUBY_FIXNUM_FLAG
+    printf "FIXNUM: %ld\n", (long)($arg0) >> 1
+  else
+  if ((VALUE)($arg0) & ~(~(VALUE)0<<RUBY_SPECIAL_SHIFT)) == RUBY_SYMBOL_FLAG
+    set $id = (($arg0) >> RUBY_SPECIAL_SHIFT)
+    printf "%sSYMBOL%s: ", $color_type, $color_end
+    rp_id $id
+  else
+  if ($arg0) == RUBY_Qfalse
+    echo false\n
+  else
+  if ($arg0) == RUBY_Qtrue
+    echo true\n
+  else
+  if ($arg0) == RUBY_Qnil
+    echo nil\n
+  else
+  if ($arg0) == RUBY_Qundef
+    echo undef\n
+  else
+  if (VALUE)($arg0) & RUBY_IMMEDIATE_MASK
+    if ((VALUE)($arg0) & RUBY_FLONUM_MASK) == RUBY_FLONUM_FLAG
+      printf "%sFLONUM%s: %g\n", $color_type, $color_end, (double)rb_float_value($arg0)
+    else
+      echo immediate\n
+    end
+  else
+  set $flags = ((struct RBasic*)($arg0))->flags
+  if ($flags & RUBY_FL_PROMOTED) == RUBY_FL_PROMOTED
+    printf "[PROMOTED] "
+  end
+  if ($flags & RUBY_T_MASK) == RUBY_T_NONE
+    printf "%sT_NONE%s: ", $color_type, $color_end
+    print (struct RBasic *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_NIL
+    printf "%sT_NIL%s: ", $color_type, $color_end
+    print (struct RBasic *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_OBJECT
+    printf "%sT_OBJECT%s: ", $color_type, $color_end
+    print ((struct RObject *)($arg0))->basic
+    if ($flags & ROBJECT_EMBED)
+      print/x *((VALUE*)((struct RObject*)($arg0))->as.ary) @ (ROBJECT_EMBED_LEN_MAX+0)
+    else
+      print (((struct RObject *)($arg0))->as.heap)
+      if (((struct RObject*)($arg0))->as.heap.numiv) > 0
+        print/x *(((struct RObject*)($arg0))->as.heap.ivptr) @ (((struct RObject*)($arg0))->as.heap.numiv)
+      end
+    end
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_CLASS
+    printf "%sT_CLASS%s%s: ", $color_type, ($flags & RUBY_FL_SINGLETON) ? "*" : "", $color_end
+    rp_class $arg0
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_ICLASS
+    printf "%sT_ICLASS%s: ", $color_type, $color_end
+    rp_class $arg0
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_MODULE
+    printf "%sT_MODULE%s: ", $color_type, $color_end
+    rp_class $arg0
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_FLOAT
+    printf "%sT_FLOAT%s: %.16g ", $color_type, $color_end, (((struct RFloat*)($arg0))->float_value)
+    print (struct RFloat *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_STRING
+    printf "%sT_STRING%s: ", $color_type, $color_end
+    rp_string $arg0 $flags
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_REGEXP
+    set $regsrc = ((struct RRegexp*)($arg0))->src
+    set $rsflags = ((struct RBasic*)$regsrc)->flags
+    printf "%sT_REGEXP%s: ", $color_type, $color_end
+    set $len = ($rsflags & RUBY_FL_USER1) ? \
+            ((struct RString*)$regsrc)->as.heap.len : \
+            (($rsflags & (RUBY_FL_USER2|RUBY_FL_USER3|RUBY_FL_USER4|RUBY_FL_USER5|RUBY_FL_USER6)) >> RUBY_FL_USHIFT+2)
+    set print address off
+    output *(char *)(($rsflags & RUBY_FL_USER1) ? \
+	    ((struct RString*)$regsrc)->as.heap.ptr : \
+	    ((struct RString*)$regsrc)->as.ary) @ $len
+    set print address on
+    printf " len:%ld ", $len
+    if $flags & RUBY_FL_USER6
+      printf "(none) "
+    end
+    if $flags & RUBY_FL_USER5
+      printf "(literal) "
+    end
+    if $flags & RUBY_FL_USER4
+      printf "(fixed) "
+    end
+    printf "encoding:%d ", ($flags & RUBY_ENCODING_MASK) >> RUBY_ENCODING_SHIFT
+    print (struct RRegexp *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_ARRAY
+    if ($flags & RUBY_FL_USER1)
+      set $len = (($flags & (RUBY_FL_USER3|RUBY_FL_USER4)) >> (RUBY_FL_USHIFT+3))
+      printf "%sT_ARRAY%s: len=%ld ", $color_type, $color_end, $len
+      printf "(embed) "
+      if ($len == 0)
+	printf "{(empty)} "
+      else
+	print/x *((VALUE*)((struct RArray*)($arg0))->as.ary) @ $len
+	printf " "
+      end
+    else
+      set $len = ((struct RArray*)($arg0))->as.heap.len
+      printf "%sT_ARRAY%s: len=%ld ", $color_type, $color_end, $len
+      if ($flags & RUBY_FL_USER2)
+	printf "(shared) shared="
+	output/x ((struct RArray*)($arg0))->as.heap.aux.shared
+	printf " "
+      else
+	printf "(ownership) capa=%ld ", ((struct RArray*)($arg0))->as.heap.aux.capa
+      end
+      if ($len == 0)
+	printf "{(empty)} "
+      else
+	print/x *((VALUE*)((struct RArray*)($arg0))->as.heap.ptr) @ $len
+	printf " "
+      end
+    end
+    print (struct RArray *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_FIXNUM
+    printf "%sT_FIXNUM%s: ", $color_type, $color_end
+    print (struct RBasic *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_HASH
+    printf "%sT_HASH%s: ", $color_type, $color_end,
+    if (((struct RHash *)($arg0))->basic->flags & RHASH_ST_TABLE_FLAG)
+      printf "st len=%ld ", ((struct RHash *)($arg0))->as.st->num_entries
+    else
+      printf "li len=%ld bound=%ld ", \
+        ((((struct RHash *)($arg0))->basic->flags & RHASH_ARRAY_LEN_MASK) >> RHASH_ARRAY_LEN_SHIFT), \
+        ((((struct RHash *)($arg0))->basic->flags & RHASH_ARRAY_BOUND_MASK) >> RHASH_ARRAY_BOUND_SHIFT)
+    end
+    print (struct RHash *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_STRUCT
+    set $len = (($flags & (RUBY_FL_USER1|RUBY_FL_USER2)) ? \
+       ($flags & (RUBY_FL_USER1|RUBY_FL_USER2)) >> (RUBY_FL_USHIFT+1) : \
+       ((struct RStruct *)($arg0))->as.heap.len)
+    printf "%sT_STRUCT%s: len=%ld ", $color_type, $color_end, $len
+    print (struct RStruct *)($arg0)
+    output/x *(($flags & (RUBY_FL_USER1|RUBY_FL_USER2)) ? \
+              ((struct RStruct *)($arg0))->as.ary : \
+              ((struct RStruct *)($arg0))->as.heap.ptr) @ $len
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_BIGNUM
+    rp_bignum $arg0
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_RATIONAL
+    printf "%sT_RATIONAL%s: ", $color_type, $color_end
+    print (struct RRational *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_COMPLEX
+    printf "%sT_COMPLEX%s: ", $color_type, $color_end
+    print (struct RComplex *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_FILE
+    printf "%sT_FILE%s: ", $color_type, $color_end
+    print (struct RFile *)($arg0)
+    output *((struct RFile *)($arg0))->fptr
+    printf "\n"
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_TRUE
+    printf "%sT_TRUE%s: ", $color_type, $color_end
+    print (struct RBasic *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_FALSE
+    printf "%sT_FALSE%s: ", $color_type, $color_end
+    print (struct RBasic *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_DATA
+    if ((struct RTypedData *)($arg0))->typed_flag == 1
+      printf "%sT_DATA%s(%s): ", $color_type, $color_end, ((struct RTypedData *)($arg0))->type->wrap_struct_name
+      print (struct RTypedData *)($arg0)
+    else
+      printf "%sT_DATA%s: ", $color_type, $color_end
+      print (struct RData *)($arg0)
+    end
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_MATCH
+    printf "%sT_MATCH%s: ", $color_type, $color_end
+    print (struct RMatch *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_SYMBOL
+    printf "%sT_SYMBOL%s: ", $color_type, $color_end
+    print (struct RSymbol *)($arg0)
+    set $id_type = ((struct RSymbol *)($arg0))->id & RUBY_ID_SCOPE_MASK
+    if $id_type == RUBY_ID_LOCAL
+      printf "l"
+    else
+    if $id_type == RUBY_ID_INSTANCE
+      printf "i"
+    else
+    if $id_type == RUBY_ID_GLOBAL
+      printf "G"
+    else
+    if $id_type == RUBY_ID_ATTRSET
+      printf "a"
+    else
+    if $id_type == RUBY_ID_CONST
+      printf "C"
+    else
+    if $id_type == RUBY_ID_CLASS
+      printf "c"
+    else
+      printf "j"
+    end
+    end
+    end
+    end
+    end
+    end
+    set $id_fstr = ((struct RSymbol *)($arg0))->fstr
+    rp_string $id_fstr
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_UNDEF
+    printf "%sT_UNDEF%s: ", $color_type, $color_end
+    print (struct RBasic *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_IMEMO
+    printf "%sT_IMEMO%s(", $color_type, $color_end
+    output (enum imemo_type)(($flags>>RUBY_FL_USHIFT)&RUBY_IMEMO_MASK)
+    printf "): "
+    rp_imemo $arg0
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_NODE
+    printf "%sT_NODE%s(", $color_type, $color_end
+    output (enum node_type)(($flags&RUBY_NODE_TYPEMASK)>>RUBY_NODE_TYPESHIFT)
+    printf "): "
+    print *(NODE *)($arg0)
+  else
+  if ($flags & RUBY_T_MASK) == RUBY_T_ZOMBIE
+    printf "%sT_ZOMBIE%s: ", $color_type, $color_end
+    print (struct RData *)($arg0)
+  else
+    printf "%sunknown%s: ", $color_type, $color_end
+    print (struct RBasic *)($arg0)
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+end
+document rp
+  Print a Ruby's VALUE.
+end
+
+define rp_id
+  set $id = (ID)$arg0
+  if $id == '!' || $id == '+' || $id == '-' || $id == '*' || $id == '/' || $id == '%' || $id == '<' || $id == '>' || $id == '`'
+    printf "(:%c)\n", $id
+  else
+  if $id == idDot2
+    printf "(:..)\n"
+  else
+  if $id == idDot3
+    printf "(:...)\n"
+  else
+  if $id == idUPlus
+    printf "(:+@)\n"
+  else
+  if $id == idUMinus
+    printf "(:-@)\n"
+  else
+  if $id == idPow
+    printf "(:**)\n"
+  else
+  if $id == idCmp
+    printf "(:<=>)\n"
+  else
+  if $id == idLTLT
+    printf "(:<<)\n"
+  else
+  if $id == idGTGT
+    printf "(:>>)\n"
+  else
+  if $id == idLE
+    printf "(:<=)\n"
+  else
+  if $id == idGE
+    printf "(:>=)\n"
+  else
+  if $id == idEq
+    printf "(:==)\n"
+  else
+  if $id == idEqq
+    printf "(:===)\n"
+  else
+  if $id == idNeq
+    printf "(:!=)\n"
+  else
+  if $id == idEqTilde
+    printf "(:=~)\n"
+  else
+  if $id == idNeqTilde
+    printf "(:!~)\n"
+  else
+  if $id == idAREF
+    printf "(:[])\n"
+  else
+  if $id == idASET
+    printf "(:[]=)\n"
+  else
+  if $id == idCOLON2
+    printf "(:'::')\n"
+  else
+  if $id == idANDOP
+    printf "(:&&)\n"
+  else
+  if $id == idOROP
+    printf "(:||)\n"
+  else
+  if $id == idANDDOT
+    printf "(:&.)\n"
+  else
+    if $id <= tLAST_OP_ID
+      printf "O"
+    else
+      set $id_type = $id & RUBY_ID_SCOPE_MASK
+      if $id_type == RUBY_ID_LOCAL
+        printf "l"
+      else
+      if $id_type == RUBY_ID_INSTANCE
+        printf "i"
+      else
+      if $id_type == RUBY_ID_GLOBAL
+        printf "G"
+      else
+      if $id_type == RUBY_ID_ATTRSET
+        printf "a"
+      else
+      if $id_type == RUBY_ID_CONST
+        printf "C"
+      else
+      if $id_type == RUBY_ID_CLASS
+        printf "c"
+      else
+        printf "j"
+      end
+      end
+      end
+      end
+      end
+      end
+    end
+    printf "(%ld): ", $id
+    print_id $id
+    echo \n
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+end
+document rp_id
+  Print an ID.
+end
+
+define output_string
+  set $flags = ((struct RBasic*)($arg0))->flags
+  set $len = ($flags & RUBY_FL_USER1) ? \
+          ((struct RString*)($arg0))->as.heap.len : \
+          (($flags & (RUBY_FL_USER2|RUBY_FL_USER3|RUBY_FL_USER4|RUBY_FL_USER5|RUBY_FL_USER6)) >> RUBY_FL_USHIFT+2)
+  if $len > 0
+    output *(char *)(($flags & RUBY_FL_USER1) ? \
+	    ((struct RString*)($arg0))->as.heap.ptr : \
+	    ((struct RString*)($arg0))->as.ary) @ $len
+  else
+    output ""
+  end
+end
+
+define print_string
+  set $flags = ((struct RBasic*)($arg0))->flags
+  set $len = ($flags & RUBY_FL_USER1) ? \
+          ((struct RString*)($arg0))->as.heap.len : \
+          (($flags & (RUBY_FL_USER2|RUBY_FL_USER3|RUBY_FL_USER4|RUBY_FL_USER5|RUBY_FL_USER6)) >> RUBY_FL_USHIFT+2)
+  if $len > 0
+    printf "%s", *(char *)(($flags & RUBY_FL_USER1) ? \
+	    ((struct RString*)($arg0))->as.heap.ptr : \
+	    ((struct RString*)($arg0))->as.ary) @ $len
+  end
+end
+
+define rp_string
+  output_string $arg0
+  printf " bytesize:%ld ", $len
+  if !($flags & RUBY_FL_USER1)
+    printf "(embed) "
+  else
+    if ($flags & RUBY_FL_USER2)
+      printf "(shared) "
+    end
+    if ($flags & RUBY_FL_USER3)
+      printf "(assoc) "
+    end
+  end
+  printf "encoding:%d ", ($flags & RUBY_ENCODING_MASK) >> RUBY_ENCODING_SHIFT
+  if ($flags & RUBY_ENC_CODERANGE_MASK) == 0
+    printf "coderange:unknown "
+  else
+  if ($flags & RUBY_ENC_CODERANGE_MASK) == RUBY_ENC_CODERANGE_7BIT
+    printf "coderange:7bit "
+  else
+  if ($flags & RUBY_ENC_CODERANGE_MASK) == RUBY_ENC_CODERANGE_VALID
+    printf "coderange:valid "
+  else
+    printf "coderange:broken "
+  end
+  end
+  end
+  print (struct RString *)($arg0)
+end
+document rp_string
+  Print the content of a String.
+end
+
+define rp_bignum
+  set $flags = ((struct RBignum*)($arg0))->basic.flags
+  set $len = (($flags & RUBY_FL_USER2) ? \
+       ($flags & (RUBY_FL_USER5|RUBY_FL_USER4|RUBY_FL_USER3)) >> (RUBY_FL_USHIFT+3) : \
+       ((struct RBignum*)($arg0))->as.heap.len)
+  printf "%sT_BIGNUM%s: sign=%d len=%ld ", $color_type, $color_end, \
+         (($flags & RUBY_FL_USER1) != 0), $len
+  if $flags & RUBY_FL_USER2
+    printf "(embed) "
+  end
+  print (struct RBignum *)($arg0)
+  set $ptr = (($flags & RUBY_FL_USER2) ? \
+              ((struct RBignum*)($arg0))->as.ary : \
+              ((struct RBignum*)($arg0))->as.heap.digits)
+  set $len = $len-1
+  printf "0x%x", $ptr[$len]
+  while $len > 0
+    set $len = $len-1
+    set $val = $ptr[$len]
+    set $w = sizeof($ptr[0])
+    printf "_"
+    if $w > 8
+      printf "%.32x", $val
+    else
+    if $w > 4
+      printf "%.16x", $val
+    else
+    if $w > 2
+      printf "%.8x", $val
+    else
+    if $w > 1
+      printf "%.4x", $val
+    else
+      printf "%.2x", $val
+    end
+    end
+    end
+    end
+  end
+  printf "\n"
+end
+document rp_bignum
+  Print the content of a Bignum.
+end
+
+define rp_class
+  printf "(struct RClass *) %p", (void*)$arg0
+  if ((struct RClass *)($arg0))->ptr.origin_ != $arg0
+    printf " -> %p", ((struct RClass *)($arg0))->ptr.origin_
+  end
+  printf "\n"
+  rb_classname $arg0
+  print/x *(struct RClass *)($arg0)
+  print *((struct RClass *)($arg0))->ptr
+end
+document rp_class
+  Print the content of a Class/Module.
+end
+
+define rp_imemo
+  set $flags = (enum imemo_type)((((struct RBasic *)($arg0))->flags >> RUBY_FL_USHIFT) & RUBY_IMEMO_MASK)
+  if $flags == imemo_cref
+    printf "(rb_cref_t *) %p\n", (void*)$arg0
+    print *(rb_cref_t *)$arg0
+  else
+  if $flags == imemo_svar
+    printf "(struct vm_svar *) %p\n", (void*)$arg0
+    print *(struct vm_svar *)$arg0
+  else
+  if $flags == imemo_throw_data
+    printf "(struct vm_throw_data *) %p\n", (void*)$arg0
+    print *(struct vm_throw_data *)$arg0
+  else
+  if $flags == imemo_ifunc
+    printf "(struct vm_ifunc *) %p\n", (void*)$arg0
+    print *(struct vm_ifunc *)$arg0
+  else
+  if $flags == imemo_memo
+    printf "(struct MEMO *) %p\n", (void*)$arg0
+    print *(struct MEMO *)$arg0
+  else
+  if $flags == imemo_ment
+    printf "(rb_method_entry_t *) %p\n", (void*)$arg0
+    print *(rb_method_entry_t *)$arg0
+  else
+  if $flags == imemo_iseq
+    printf "(rb_iseq_t *) %p\n", (void*)$arg0
+    print *(rb_iseq_t *)$arg0
+  else
+    printf "(struct RIMemo *) %p\n", (void*)$arg0
+    print *(struct RIMemo *)$arg0
+  end
+  end
+  end
+  end
+  end
+  end
+  end
+end
+document rp_imemo
+  Print the content of a memo
+end
+
+define nd_type
+  print (enum node_type)((((NODE*)($arg0))->flags&RUBY_NODE_TYPEMASK)>>RUBY_NODE_TYPESHIFT)
+end
+document nd_type
+  Print a Ruby' node type.
+end
+
+define nd_file
+  print ((NODE*)($arg0))->nd_file
+end
+document nd_file
+  Print the source file name of a node.
+end
+
+define nd_line
+  print ((unsigned int)((((NODE*)($arg0))->flags>>RUBY_NODE_LSHIFT)&RUBY_NODE_LMASK))
+end
+document nd_line
+  Print the source line number of a node.
+end
+
+# Print members of ruby node.
+
+define nd_head
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+define nd_alen
+  printf "%su2.argc%s: ", $color_highlite, $color_end
+  p ($arg0).u2.argc
+end
+
+define nd_next
+  printf "%su3.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.node
+end
+
+
+define nd_cond
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+define nd_body
+  printf "%su2.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.node
+end
+
+define nd_else
+  printf "%su3.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.node
+end
+
+
+define nd_orig
+  printf "%su3.value%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.value
+end
+
+
+define nd_resq
+  printf "%su2.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.node
+end
+
+define nd_ensr
+  printf "%su3.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.node
+end
+
+
+define nd_1st
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+define nd_2nd
+  printf "%su2.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.node
+end
+
+
+define nd_stts
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+
+define nd_entry
+  printf "%su3.entry%s: ", $color_highlite, $color_end
+  p ($arg0).u3.entry
+end
+
+define nd_vid
+  printf "%su1.id%s: ", $color_highlite, $color_end
+  p ($arg0).u1.id
+end
+
+define nd_cflag
+  printf "%su2.id%s: ", $color_highlite, $color_end
+  p ($arg0).u2.id
+end
+
+define nd_cval
+  printf "%su3.value%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.value
+end
+
+define nd_tbl
+  printf "%su1.tbl%s: ", $color_highlite, $color_end
+  p ($arg0).u1.tbl
+end
+
+
+define nd_var
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+define nd_ibdy
+  printf "%su2.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.node
+end
+
+define nd_iter
+  printf "%su3.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.node
+end
+
+
+define nd_value
+  printf "%su2.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.node
+end
+
+define nd_aid
+  printf "%su3.id%s: ", $color_highlite, $color_end
+  p ($arg0).u3.id
+end
+
+
+define nd_lit
+  printf "%su1.value%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.value
+end
+
+define nd_rest
+  printf "%su2.argc%s: ", $color_highlite, $color_end
+  p ($arg0).u2.argc
+end
+
+define nd_opt
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+
+define nd_recv
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+define nd_mid
+  printf "%su2.id%s: ", $color_highlite, $color_end
+  p ($arg0).u2.id
+end
+
+define nd_args
+  printf "%su3.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.node
+end
+
+define nd_defn
+  printf "%su3.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.node
+end
+
+
+define nd_old
+  printf "%su1.id%s: ", $color_highlite, $color_end
+  p ($arg0).u1.id
+end
+
+define nd_new
+  printf "%su2.id%s: ", $color_highlite, $color_end
+  p ($arg0).u2.id
+end
+
+
+define nd_cname
+  printf "%su1.id%s: ", $color_highlite, $color_end
+  p ($arg0).u1.id
+end
+
+define nd_super
+  printf "%su3.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u3.node
+end
+
+
+define nd_modl
+  printf "%su1.id%s: ", $color_highlite, $color_end
+  p ($arg0).u1.id
+end
+
+define nd_clss
+  printf "%su1.value%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.value
+end
+
+
+define nd_beg
+  printf "%su1.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u1.node
+end
+
+define nd_end
+  printf "%su2.node%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.node
+end
+
+define nd_state
+  printf "%su3.state%s: ", $color_highlite, $color_end
+  p ($arg0).u3.state
+end
+
+define nd_rval
+  printf "%su2.value%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.value
+end
+
+
+define nd_nth
+  printf "%su2.argc%s: ", $color_highlite, $color_end
+  p ($arg0).u2.argc
+end
+
+
+define nd_tag
+  printf "%su1.id%s: ", $color_highlite, $color_end
+  p ($arg0).u1.id
+end
+
+define nd_tval
+  printf "%su2.value%s: ", $color_highlite, $color_end
+  rp ($arg0).u2.value
+end
+
+define nd_tree
+  set $buf = (struct RString *)rb_str_buf_new(0)
+  call dump_node((VALUE)($buf), rb_str_tmp_new(0), 0, ($arg0))
+  printf "%s\n", $buf->as.heap.ptr
+end
+
+define rb_p
+  call rb_p($arg0)
+end
+
+define rb_numtable_entry
+  set $rb_numtable_tbl = $arg0
+  set $rb_numtable_id = (st_data_t)$arg1
+  set $rb_numtable_key = 0
+  set $rb_numtable_rec = 0
+  if $rb_numtable_tbl->entries_packed
+    set $rb_numtable_p = $rb_numtable_tbl->as.packed.bins
+    while $rb_numtable_p && $rb_numtable_p < $rb_numtable_tbl->as.packed.bins+$rb_numtable_tbl->num_entries
+      if $rb_numtable_p.k == $rb_numtable_id
+	set $rb_numtable_key = $rb_numtable_p.k
+	set $rb_numtable_rec = $rb_numtable_p.v
+	set $rb_numtable_p = 0
+      else
+	set $rb_numtable_p = $rb_numtable_p + 1
+      end
+    end
+  else
+    set $rb_numtable_p = $rb_numtable_tbl->as.big.bins[st_numhash($rb_numtable_id) % $rb_numtable_tbl->num_bins]
+    while $rb_numtable_p
+      if $rb_numtable_p->key == $rb_numtable_id
+	set $rb_numtable_key = $rb_numtable_p->key
+	set $rb_numtable_rec = $rb_numtable_p->record
+	set $rb_numtable_p = 0
+      else
+	set $rb_numtable_p = $rb_numtable_p->next
+      end
+    end
+  end
+end
+
+define rb_id2name
+  ruby_gdb_init
+  printf "%sID%s: ", $color_type, $color_end
+  rp_id $arg0
+end
+document rb_id2name
+  Print the name of id
+end
+
+define rb_method_entry
+  set $rb_method_entry_klass = (struct RClass *)$arg0
+  set $rb_method_entry_id = (ID)$arg1
+  set $rb_method_entry_me = (rb_method_entry_t *)0
+  while !$rb_method_entry_me && $rb_method_entry_klass
+    rb_numtable_entry $rb_method_entry_klass->m_tbl_wrapper->tbl $rb_method_entry_id
+    set $rb_method_entry_me = (rb_method_entry_t *)$rb_numtable_rec
+    if !$rb_method_entry_me
+      set $rb_method_entry_klass = (struct RClass *)RCLASS_SUPER($rb_method_entry_klass)
+    end
+  end
+  if $rb_method_entry_me
+    print *$rb_method_entry_klass
+    print *$rb_method_entry_me
+  else
+    echo method not found\n
+  end
+end
+document rb_method_entry
+  Search method entry by class and id
+end
+
+define rb_classname
+  # up to 128bit int
+  set $rb_classname = rb_mod_name($arg0)
+  if $rb_classname != RUBY_Qnil
+    rp $rb_classname
+  else
+    echo anonymous class/module\n
+  end
+end
+
+define rb_ancestors
+  set $rb_ancestors_module = $arg0
+  while $rb_ancestors_module
+    rp_class $rb_ancestors_module
+    set $rb_ancestors_module = RCLASS_SUPER($rb_ancestors_module)
+  end
+end
+document rb_ancestors
+  Print ancestors.
+end
+
+define rb_backtrace
+  call rb_backtrace()
+end
+
+define iseq
+  if ruby_dummy_gdb_enums.special_consts
+  end
+  if ($arg0)->type == ISEQ_ELEMENT_NONE
+    echo [none]\n
+  end
+  if ($arg0)->type == ISEQ_ELEMENT_LABEL
+    print *(LABEL*)($arg0)
+  end
+  if ($arg0)->type == ISEQ_ELEMENT_INSN
+    print *(INSN*)($arg0)
+    if ((INSN*)($arg0))->insn_id != YARVINSN_jump
+      set $i = 0
+      set $operand_size = ((INSN*)($arg0))->operand_size
+      set $operands = ((INSN*)($arg0))->operands
+      while $i < $operand_size
+	rp $operands[$i++]
+      end
+    end
+  end
+  if ($arg0)->type == ISEQ_ELEMENT_ADJUST
+    print *(ADJUST*)($arg0)
+  end
+end
+
+define rb_ps
+  rb_ps_vm ruby_current_vm_ptr
+end
+document rb_ps
+Dump all threads and their callstacks
+end
+
+define rb_ps_vm
+  print $ps_vm = (rb_vm_t*)$arg0
+  set $ps_thread_ln = $ps_vm->living_threads.n.next
+  set $ps_thread_ln_last = $ps_vm->living_threads.n.prev
+  while 1
+    set $ps_thread_th = (rb_thread_t *)$ps_thread_ln
+    set $ps_thread = (VALUE)($ps_thread_th->self)
+    rb_ps_thread $ps_thread
+    if $ps_thread_ln == $ps_thread_ln_last
+      loop_break
+    end
+    set $ps_thread_ln = $ps_thread_ln->next
+  end
+end
+document rb_ps_vm
+Dump all threads in a (rb_vm_t*) and their callstacks
+end
+
+define print_lineno
+  set $cfp = $arg0
+  set $iseq = $cfp->iseq
+  set $pos = $cfp->pc - $iseq->body->iseq_encoded
+  if $pos != 0
+    set $pos = $pos - 1
+  end
+
+  set $index = 0
+  set $size = $iseq->body->insns_info.size
+  set $table = $iseq->body->insns_info.body
+  set $positions = $iseq->body->insns_info.positions
+  #printf "size: %d\n", $size
+  if $size == 0
+  else
+  if $size == 1
+    printf "%d", $table[0].line_no
+  else
+    if $positions
+      # get_insn_info_linear_search
+      set $index = 1
+      while $index < $size
+        #printf "table[%d]: position: %d, line: %d, pos: %d\n", $i, $positions[$i], $table[$i].line_no, $pos
+        if $positions[$index] > $pos
+          loop_break
+        end
+        set $index = $index + 1
+        if $positions[$index] == $pos
+          loop_break
+        end
+      end
+    else
+      # get_insn_info_succinct_bitvector
+      set $sd = $iseq->body->insns_info.succ_index_table
+      set $immediate_table_size = sizeof($sd->imm_part) / sizeof(uint64_t) * 9
+      if $pos < $immediate_table_size
+        set $i = $pos / 9
+        set $j = $pos % 9
+        set $index = ((int)($sd->imm_part[$i] >> ($j * 7))) & 0x7f
+      else
+        set $block_index = ($pos - $immediate_table_size) / 512
+        set $block = &$sd->succ_part[$block_index]
+        set $block_bit_index = ($pos - $immediate_table_size) % 512
+        set $small_block_index = $block_bit_index / 64
+        set $small_block_popcount = $small_block_index == 0 ? 0 : (((int)($block->small_block_ranks >> (($small_block_index - 1) * 9))) & 0x1ff)
+        set $x = $block->bits[$small_block_index] << (63 - $block_bit_index % 64)
+        set $x = ($x & 0x5555555555555555) + ($x >> 1 & 0x5555555555555555)
+        set $x = ($x & 0x3333333333333333) + ($x >> 2 & 0x3333333333333333)
+        set $x = ($x & 0x0707070707070707) + ($x >> 4 & 0x0707070707070707)
+        set $x = ($x & 0x001f001f001f001f) + ($x >> 8 & 0x001f001f001f001f)
+        set $x = ($x & 0x0000003f0000003f) + ($x >>16 & 0x0000003f0000003f)
+        set $popcnt = ($x & 0x7f) + ($x >>32 & 0x7f)
+        set $index = $block->rank + $small_block_popcount + $popcnt
+      end
+    end
+    printf "%d", $table[$index-1].line_no
+  end
+  end
+end
+
+define check_method_entry
+  set $imemo = (struct RBasic *)$arg0
+  if $imemo != RUBY_Qfalse
+    set $type = ($imemo->flags >> 12) & 0x07
+    if $type == imemo_ment
+      set $me = (rb_callable_method_entry_t *)$imemo
+    else
+    if $type == imemo_svar
+      set $imemo = ((struct vm_svar *)$imemo)->cref_or_me
+      check_method_entry $imemo
+    end
+    end
+  end
+end
+
+define print_id
+  set $id = $arg0
+  # rb_id_to_serial
+  if $id > tLAST_OP_ID
+    set $serial = (rb_id_serial_t)($id >> RUBY_ID_SCOPE_SHIFT)
+  else
+    set $serial = (rb_id_serial_t)$id
+  end
+  if $serial && $serial <= global_symbols.last_id
+    set $idx = $serial / ID_ENTRY_UNIT
+    set $ids = (struct RArray *)global_symbols.ids
+    set $flags = $ids->basic.flags
+    if ($flags & RUBY_FL_USER1)
+      set $idsptr = $ids->as.ary
+      set $idslen = (($flags & (RUBY_FL_USER3|RUBY_FL_USER4)) >> (RUBY_FL_USHIFT+3))
+    else
+      set $idsptr = $ids->as.heap.ptr
+      set $idslen = $ids->as.heap.len
+    end
+    if $idx < $idslen
+      set $t = 0
+      set $ary = (struct RArray *)$idsptr[$idx]
+      if $ary != RUBY_Qnil
+        set $flags = $ary->basic.flags
+        if ($flags & RUBY_FL_USER1)
+          set $aryptr = $ary->as.ary
+          set $arylen = (($flags & (RUBY_FL_USER3|RUBY_FL_USER4)) >> (RUBY_FL_USHIFT+3))
+        else
+          set $aryptr = $ary->as.heap.ptr
+          set $arylen = $ary->as.heap.len
+        end
+        set $result = $aryptr[($serial % ID_ENTRY_UNIT) * ID_ENTRY_SIZE + $t]
+	if $result != RUBY_Qnil
+          print_string $result
+	else
+	  echo undef
+	end
+      end
+    end
+  end
+end
+
+define print_pathobj
+  set $flags = ((struct RBasic*)($arg0))->flags
+  if ($flags & RUBY_T_MASK) == RUBY_T_STRING
+    print_string $arg0
+  end
+  if ($flags & RUBY_T_MASK) == RUBY_T_ARRAY
+    if $flags & RUBY_FL_USER1
+      set $str = ((struct RArray*)($arg0))->as.ary[0]
+    else
+      set $str = ((struct RArray*)($arg0))->as.heap.ptr[0]
+    end
+    print_string $str
+  end
+end
+
+define rb_ps_thread
+  set $ps_thread = (struct RTypedData*)$arg0
+  set $ps_thread_th = (rb_thread_t*)$ps_thread->data
+  printf "* #<Thread:%p rb_thread_t:%p native_thread:%p>\n", \
+    $ps_thread, $ps_thread_th, $ps_thread_th->thread_id
+  set $cfp = $ps_thread_th->ec->cfp
+  set $cfpend = (rb_control_frame_t *)($ps_thread_th->ec->vm_stack + $ps_thread_th->ec->vm_stack_size)-1
+  while $cfp < $cfpend
+    if $cfp->iseq
+      if !((VALUE)$cfp->iseq & RUBY_IMMEDIATE_MASK) && (((imemo_ifunc << RUBY_FL_USHIFT) | RUBY_T_IMEMO)==$cfp->iseq->flags & ((RUBY_IMEMO_MASK << RUBY_FL_USHIFT) | RUBY_T_MASK))
+        printf "%d:ifunc ", $cfpend-$cfp
+        set print symbol-filename on
+        output/a $cfp->iseq.body
+        set print symbol-filename off
+        printf "\n"
+      else
+      if $cfp->pc
+        set $location = $cfp->iseq->body->location
+        printf "%d:", $cfpend-$cfp
+        print_pathobj $location.pathobj
+        printf ":"
+        print_lineno $cfp
+        printf ":in `"
+        print_string $location.label
+        printf "'\n"
+      else
+        printf "%d: ???.rb:???:in `???'\n", $cfpend-$cfp
+      end
+      end
+    else
+      # if VM_FRAME_TYPE($cfp->flag) == VM_FRAME_MAGIC_CFUNC
+      set $ep = $cfp->ep
+      if ($ep[0] & 0xffff0001) == 0x55550001
+        #define VM_ENV_FLAG_LOCAL 0x02
+        #define VM_ENV_PREV_EP(ep)   GC_GUARDED_PTR_REF(ep[VM_ENV_DATA_INDEX_SPECVAL])
+        set $me = 0
+        set $env_specval = $ep[-1]
+        set $env_me_cref = $ep[-2]
+        while ($env_specval & 0x02) != 0
+          check_method_entry $env_me_cref
+          if $me != 0
+            loop_break
+          end
+          set $ep = $ep[0]
+          set $env_specval = $ep[-1]
+          set $env_me_cref = $ep[-2]
+        end
+        if $me == 0
+          check_method_entry $env_me_cref
+        end
+        printf "%d:", $cfpend-$cfp
+        set print symbol-filename on
+        output/a $me->def->body.cfunc.func
+        set print symbol-filename off
+        set $mid = $me->def->original_id
+        printf ":in `"
+        print_id $mid
+        printf "'\n"
+      else
+        printf "%d:unknown_frame:???:in `???'\n", $cfpend-$cfp
+      end
+    end
+    set $cfp = $cfp + 1
+  end
+end
+
+define rb_count_objects
+  set $objspace = ruby_current_vm_ptr->objspace
+  set $counts_00 = 0
+  set $counts_01 = 0
+  set $counts_02 = 0
+  set $counts_03 = 0
+  set $counts_04 = 0
+  set $counts_05 = 0
+  set $counts_06 = 0
+  set $counts_07 = 0
+  set $counts_08 = 0
+  set $counts_09 = 0
+  set $counts_0a = 0
+  set $counts_0b = 0
+  set $counts_0c = 0
+  set $counts_0d = 0
+  set $counts_0e = 0
+  set $counts_0f = 0
+  set $counts_10 = 0
+  set $counts_11 = 0
+  set $counts_12 = 0
+  set $counts_13 = 0
+  set $counts_14 = 0
+  set $counts_15 = 0
+  set $counts_16 = 0
+  set $counts_17 = 0
+  set $counts_18 = 0
+  set $counts_19 = 0
+  set $counts_1a = 0
+  set $counts_1b = 0
+  set $counts_1c = 0
+  set $counts_1d = 0
+  set $counts_1e = 0
+  set $counts_1f = 0
+  set $total = 0
+  set $i = 0
+  while $i < $objspace->heap_pages.allocated_pages
+    printf "\rcounting... %d/%d", $i, $objspace->heap_pages.allocated_pages
+    set $page = $objspace->heap_pages.sorted[$i]
+    set $p = $page->start
+    set $pend = $p + $page->total_slots
+    while $p < $pend
+      set $flags = $p->as.basic.flags & 0x1f
+      eval "set $counts_%02x = $counts_%02x + 1", $flags, $flags
+      set $p = $p + 1
+    end
+    set $total = $total + $page->total_slots
+    set $i = $i + 1
+  end
+  printf "\rTOTAL: %d, FREE: %d\n", $total, $counts_00
+  printf "T_OBJECT: %d\n", $counts_01
+  printf "T_CLASS: %d\n", $counts_02
+  printf "T_MODULE: %d\n", $counts_03
+  printf "T_FLOAT: %d\n", $counts_04
+  printf "T_STRING: %d\n", $counts_05
+  printf "T_REGEXP: %d\n", $counts_06
+  printf "T_ARRAY: %d\n", $counts_07
+  printf "T_HASH: %d\n", $counts_08
+  printf "T_STRUCT: %d\n", $counts_09
+  printf "T_BIGNUM: %d\n", $counts_0a
+  printf "T_FILE: %d\n", $counts_0b
+  printf "T_DATA: %d\n", $counts_0c
+  printf "T_MATCH: %d\n", $counts_0d
+  printf "T_COMPLEX: %d\n", $counts_0e
+  printf "T_RATIONAL: %d\n", $counts_0f
+  #printf "UNKNOWN_10: %d\n", $counts_10
+  printf "T_NIL: %d\n", $counts_11
+  printf "T_TRUE: %d\n", $counts_12
+  printf "T_FALSE: %d\n", $counts_13
+  printf "T_SYMBOL: %d\n", $counts_14
+  printf "T_FIXNUM: %d\n", $counts_15
+  printf "T_UNDEF: %d\n", $counts_16
+  #printf "UNKNOWN_17: %d\n", $counts_17
+  #printf "UNKNOWN_18: %d\n", $counts_18
+  #printf "UNKNOWN_19: %d\n", $counts_19
+  printf "T_IMEMO: %d\n", $counts_1a
+  printf "T_NODE: %d\n", $counts_1b
+  printf "T_ICLASS: %d\n", $counts_1c
+  printf "T_ZOMBIE: %d\n", $counts_1d
+  #printf "UNKNOWN_1E: %d\n", $counts_1e
+  printf "T_MASK: %d\n", $counts_1f
+end
+document rb_count_objects
+  Counts all objects grouped by type.
+end
+
+# Details: https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/MachineInstructionsTraceWithGDB
+define trace_machine_instructions
+  set logging on
+  set height 0
+  set width 0
+  display/i $pc
+  while !$exit_code
+    info line *$pc
+    si
+  end
+end
+
+define SDR
+  call rb_vmdebug_stack_dump_raw_current()
+end
+
+define rbi
+  if ((LINK_ELEMENT*)$arg0)->type == ISEQ_ELEMENT_LABEL
+    p *(LABEL*)$arg0
+  else
+  if ((LINK_ELEMENT*)$arg0)->type == ISEQ_ELEMENT_INSN
+    p *(INSN*)$arg0
+  else
+  if ((LINK_ELEMENT*)$arg0)->type == ISEQ_ELEMENT_ADJUST
+    p *(ADJUST*)$arg0
+  else
+    print *$arg0
+  end
+  end
+  end
+end
+
+define dump_node
+  set $str = rb_parser_dump_tree($arg0, 0)
+  set $flags = ((struct RBasic*)($str))->flags
+  printf "%s", (char *)(($flags & RUBY_FL_USER1) ? \
+                        ((struct RString*)$str)->as.heap.ptr : \
+                        ((struct RString*)$str)->as.ary)
+end
+
+define print_flags
+  printf "RUBY_FL_WB_PROTECTED: %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_WB_PROTECTED ? "1" : "0"
+  printf "RUBY_FL_PROMOTED0   : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_PROMOTED0 ? "1" : "0"
+  printf "RUBY_FL_PROMOTED1   : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_PROMOTED1 ? "1" : "0"
+  printf "RUBY_FL_FINALIZE    : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_FINALIZE ? "1" : "0"
+  printf "RUBY_FL_TAINT       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_TAINT ? "1" : "0"
+  printf "RUBY_FL_UNTRUSTED   : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_UNTRUSTED ? "1" : "0"
+  printf "RUBY_FL_EXIVAR      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_EXIVAR ? "1" : "0"
+  printf "RUBY_FL_FREEZE      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_FREEZE ? "1" : "0"
+
+  printf "RUBY_FL_USER0       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER0 ? "1" : "0"
+  printf "RUBY_FL_USER1       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER1 ? "1" : "0"
+  printf "RUBY_FL_USER2       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER2 ? "1" : "0"
+  printf "RUBY_FL_USER3       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER3 ? "1" : "0"
+  printf "RUBY_FL_USER4       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER4 ? "1" : "0"
+  printf "RUBY_FL_USER5       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER5 ? "1" : "0"
+  printf "RUBY_FL_USER6       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER6 ? "1" : "0"
+  printf "RUBY_FL_USER7       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER7 ? "1" : "0"
+  printf "RUBY_FL_USER8       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER8 ? "1" : "0"
+  printf "RUBY_FL_USER9       : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER9 ? "1" : "0"
+  printf "RUBY_FL_USER10      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER10 ? "1" : "0"
+  printf "RUBY_FL_USER11      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER11 ? "1" : "0"
+  printf "RUBY_FL_USER12      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER12 ? "1" : "0"
+  printf "RUBY_FL_USER13      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER13 ? "1" : "0"
+  printf "RUBY_FL_USER14      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER14 ? "1" : "0"
+  printf "RUBY_FL_USER15      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER15 ? "1" : "0"
+  printf "RUBY_FL_USER16      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER16 ? "1" : "0"
+  printf "RUBY_FL_USER17      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER17 ? "1" : "0"
+  printf "RUBY_FL_USER18      : %s\n", ((struct RBasic*)($arg0))->flags & RUBY_FL_USER18 ? "1" : "0"
+end
diff --git a/.gitattributes b/.gitattributes
new file mode 100644
index 0000000000..6ca2f89462
--- /dev/null
+++ b/.gitattributes
@@ -0,0 +1,6 @@
+*.gemspec diff=ruby
+*.rb diff=ruby
+bin svn-properties=svn:ignore=ruby
+bin/* diff=ruby
+tool/update-deps diff=ruby
+tool/make-snapshot diff=ruby
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000000..50bf27f28a
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,212 @@
+*-*-*.def
+*-*-*.exp
+*-*-*.lib
+*.a
+*.bak
+*.bc
+*.bundle
+*.dSYM
+*.dmyh
+*.dylib
+*.elc
+*.i
+*.inc
+*.log
+*.o
+*.obj
+*.old
+*.orig
+*.pch
+*.pdb
+*.rej
+*.s
+*.sav
+*.sl
+*.so
+*.swp
+*.yarb
+*~
+.*-*
+.*.list
+.*.time
+.DS_Store
+.bundle
+.ccmalloc
+.ext
+.pc
+.ppack
+.svn
+Makefile
+cygruby*.def
+extconf.h
+y.output
+y.tab.c
+*.gcda
+*.gcno
+*.gcov
+lcov*.info
+
+# /
+/*-fake.rb
+/*.dll
+/*.exe
+/*.res
+/*.pc
+/*.rc
+/*_prelude.c
+/COPYING.LIB
+/Doxyfile
+/GNUmakefile
+/README.atheos
+/README.fat-patch
+/README.v6
+/TAGS
+/archive
+/autom4te*.cache
+/automake
+/benchmark/benchmark-driver
+/beos
+/bmlog-*
+/breakpoints.gdb
+/config.cache
+/config.h
+/config.h.in
+/config.status
+/config.status.lineno
+/configure
+/coverage/simplecov
+/coverage/simplecov-html
+/coverage/doclie
+/coverage/.last_run.json
+/coverage/.resultset.json*
+/coverage/assets
+/coverage/index.html
+/doc/capi
+/enc.mk
+/encdb.h
+/exts.mk
+/goruby
+/id.[ch]
+/largefile.h
+/lcov-c-out
+/lcov-rb-out
+/lcov-out
+/lex.c
+/libruby*.*
+/miniprelude.c
+/miniruby
+/newdate.rb
+/newline.c
+/newver.rb
+/parse.c
+/parse.h
+/patches
+/patches-master
+/pitest.rb
+/ppack
+/prelude.c
+/preview
+/probes.dmyh
+/probes.h
+/rbconfig.rb
+/rename2.h
+/repack
+/revision.h
+/riscos
+/rubicon
+/ruby
+/ruby-runner
+/ruby-runner.h
+/ruby-man.rd.gz
+/run.gdb
+/sizes.c
+/test.rb
+/test-coverage.dat
+/tmp
+/transdb.h
+/uncommon.mk
+/verconf.h
+/verconf.mk
+/web
+/yasmdata.rb
+
+# /benchmark/
+/benchmark/bm_require.data
+/benchmark/bmx_*.rb
+/benchmark/fasta.output.*
+/benchmark/wc.input
+
+/enc/*.def
+/enc/*.exp
+/enc/*.lib
+/enc/jis/props.h
+/enc/unicode/data
+
+# /coroutine/
+!/coroutine/**/*.s
+/coroutine/**/.time
+
+# /enc/trans/
+/enc/trans/*.c
+/enc/trans/*.def
+/enc/trans/*.exp
+/enc/trans/*.lib
+/enc/trans/.time
+
+# /exe/
+/exe/goruby
+/exe/ruby
+/exe/.time
+
+# /ext/
+/ext/extinit.c
+/ext/configure-ext.mk
+/ext/*/exts.mk
+
+# /ext/-test-/win32/dln/
+/ext/-test-/win32/dln/dlntest.dll
+/ext/-test-/win32/dln/dlntest.exp
+/ext/-test-/win32/dln/dlntest.lib
+
+# /ext/etc/
+/ext/etc/constdefs.h
+
+# /ext/fiddle/
+/ext/fiddle/libffi-*
+
+# /ext/rbconfig/
+/ext/rbconfig/sizeof/sizes.c
+/ext/rbconfig/sizeof/limits.c
+
+# /ext/ripper/
+/ext/ripper/eventids1.c
+/ext/ripper/eventids2table.c
+/ext/ripper/ripper.*
+/ext/ripper/ids1
+/ext/ripper/ids2
+
+# /ext/socket/
+/ext/socket/constants.h
+/ext/socket/constdefs.h
+/ext/socket/constdefs.c
+
+# /gems
+/gems/*.gem
+/gems/src
+/gems/*-*
+
+# /spec/bundler
+/.rspec_status
+
+# /tool/
+/tool/config.guess
+/tool/config.sub
+
+# /win32/
+/win32/*.ico
+/win32/.time
+
+# MJIT
+/rb_mjit_header.h
+/mjit_config.h
+/include/ruby-*/*/rb_mjit_min_header-*.h
diff --git a/.indent.pro b/.indent.pro
new file mode 100644
index 0000000000..6a207a0554
--- /dev/null
+++ b/.indent.pro
@@ -0,0 +1,21 @@
+-bap
+-nbbb
+-nbc
+-br
+-nbs
+-ncdb
+-ce
+-cli0.5
+-ndj
+-ei
+-nfc1
+-i4
+-l120
+-lp
+-npcs
+-psl
+-sc
+-sob
+
+-TID
+-TVALUE
diff --git a/.travis.yml b/.travis.yml
new file mode 100644
index 0000000000..6004cf50db
--- /dev/null
+++ b/.travis.yml
@@ -0,0 +1,412 @@
+# Copyright (C) 2011 Urabe, Shyouhei.  All rights reserved.
+#
+# This file is  a part of the programming language  Ruby.  Permission is hereby
+# granted,  to either  redistribute  or  modify this  file,  provided that  the
+# conditions  mentioned in  the file  COPYING are  met.  Consult  the  file for
+# details.
+
+# This is  a Travis-CI  build configuration file.   The list  of configurations
+# available is located in
+#
+#     http://about.travis-ci.org/docs/user/build-configuration/
+#
+# and as Ruby itself is a project written in C language,
+#
+#     http://about.travis-ci.org/docs/user/languages/c/
+#
+# is also a good place to look at.
+
+language: c
+
+dist: xenial
+
+osx_image: xcode10.1
+
+sudo: false
+
+git:
+  quiet: true
+
+addons:
+  apt:
+    config:
+      retries: true
+    update: true
+    sources:
+      - ubuntu-toolchain-r-test
+    packages:
+      - gcc-8
+      - libffi-dev
+      - libgdbm-dev
+      - libgmp-dev
+      - libjemalloc-dev
+      - libncurses5-dev
+      - libncursesw5-dev
+      - libreadline6-dev
+      - libssl-dev
+      - libyaml-dev
+      - openssl
+      - valgrind
+      - zlib1g-dev
+  homebrew:
+    packages:
+      - gdbm
+      - gmp
+      - libffi
+      - openssl@1.1
+      - zlib
+      - ccache
+
+cache:
+  ccache: true
+  directories:
+    - $HOME/config_2nd
+    - $HOME/.downloaded-cache
+
+env:
+  global:
+    - CONFIGURE_TTY=no
+    - CCACHE_COMPILERCHECK=none
+    - CCACHE_NOCOMPRESS=1
+    - CCACHE_MAXSIZE=512Mi
+    - >-
+      NPROC="`case ${TRAVIS_OS_NAME} in
+      osx)   sysctl -n hw.activecpu ;;
+      linux) nproc ;;
+      esac`"
+    # JOBS and SETARCH are overridden when necessary; see below.
+    - JOBS=-j$((1+${NPROC}))
+    - SETARCH=
+    - RUBY_PREFIX=/tmp/ruby-prefix
+    - GEMS_FOR_TEST='timezone tzinfo'
+
+.org.ruby-lang.ci.matrix-definitions:
+
+  - &cron-only
+    if: (type = cron) AND (branch = trunk) AND (fork = false)
+
+  - &make-test-only
+    script:
+      - $SETARCH make -s test TESTOPTS="${TESTOPTS=$JOBS -q --tty=no}"
+
+  - &linux
+    os: linux
+    compiler: gcc-8
+
+  - &osx
+    os: osx
+    compiler: clang
+    before_install:
+      - /usr/local/opt/openssl@1.1/bin/openssl version
+
+  # --------
+
+  - &x86_64-linux
+    name: x86_64-linux
+    <<: *linux
+
+  - &jemalloc
+    name: --with-jemalloc
+    <<: *linux
+    <<: *cron-only
+    env:
+      - CONFIG_FLAG='--with-gmp --with-jemalloc --with-valgrind'
+
+  - &VM_CHECK_MODE
+    name: VM_CHECK_MODE=3
+    <<: *linux
+    <<: *cron-only
+    <<: *make-test-only
+    env:
+      - cppflags=-DVM_CHECK_MODE=0x0003
+
+  - &FIBER_USE_sjlj
+    name: FIBER_USE_NATIVE=0
+    <<: *linux
+    <<: *cron-only
+    env:
+      - cppflags=-DFIBER_USE_NATIVE=0
+
+  - &TOKEN_THREADED_CODE
+    name: TOKEN_THREADED_CODE
+    <<: *linux
+    <<: *cron-only
+    <<: *make-test-only
+    env:
+      - cppflags=-DOPT_THREADED_CODE=1
+
+  - &CALL_THREADED_CODE
+    name: CALL_THREADED_CODE
+    <<: *linux
+    <<: *cron-only
+    <<: *make-test-only
+    env:
+      - cppflags=-DOPT_THREADED_CODE=2
+
+  - &NO_THREADED_CODE
+    name: NO_THREADED_CODE
+    <<: *linux
+    <<: *cron-only
+    <<: *make-test-only
+    env:
+      - cppflags=-DOPT_THREADED_CODE=3
+
+  - &ASAN
+    name: -fsanitize=address
+    <<: *linux
+    #<<: *cron-only
+    <<: *make-test-only
+    compiler: clang
+    env:
+      - ASAN_OPTIONS=detect_leaks=0
+      - cflags='-march=native -fsanitize=address -fno-omit-frame-pointer'
+      - debugflags=-ggdb3
+      - optflags=-O1
+      - LD=clang
+      - LDFLAGS=-fsanitize=address
+
+  - &MSAN
+    name: -fsanitize=memory
+    <<: *linux
+    #<<: *cron-only
+    <<: *make-test-only
+    compiler: clang
+    env:
+      - cflags='-fsanitize=memory -fsanitize-memory-track-origins=2 -fno-omit-frame-pointer'
+      - optflags=-O1
+      - LD=clang
+      - LDFLAGS=-fsanitize=memory
+      - CONFIG_FLAG=--with-out-ext=openssl
+
+  - &UBSAN
+    name: -fsanitize=undefined
+    <<: *linux
+    #<<: *cron-only
+    <<: *make-test-only
+    compiler: clang
+    env:
+      - cflags='-fsanitize=undefined,integer,nullability -fno-omit-frame-pointer'
+      - cppflags=-DUNALIGNED_WORD_ACCESS=0
+      - debugflags=-ggdb3
+      - optflags='-O1 -march=native'
+      - LD=clang
+      - LDFLAGS=-fsanitize=undefined,integer,nullability
+
+  - &i686-linux
+    name: i686-linux
+    <<: *linux
+    sudo: required
+    env:
+      - GCC_FLAGS=-m32
+      - debugflags=-g0
+      - SETARCH='setarch i686 --verbose --3gb'
+    addons:
+      apt:
+        config:
+          retries: true
+        update: true
+        sources:
+          - ubuntu-toolchain-r-test
+        packages:
+          - gcc-8-multilib
+          - libffi-dev:i386
+          - libffi6:i386
+          - libgdbm-dev:i386
+          - libgdbm3:i386
+          - libncurses5-dev:i386
+          - libncurses5:i386
+          - libncursesw5-dev:i386
+          - libreadline6-dev:i386
+          - libreadline6:i386
+          - libssl-dev:i386
+          - libssl1.0.0:i386
+          - linux-libc-dev:i386
+          - zlib1g-dev:i386
+          - zlib1g:i386
+
+  - &pedanticism
+    name: -ansi -pedantic
+    <<: *linux
+    <<: *make-test-only
+    compiler: clang
+    env:
+      - GCC_FLAGS='-ansi -Werror=pedantic -pedantic-errors -std=iso9899:1990'
+      - CONFIG_FLAG=
+      - JOBS=
+      - >-
+        warnflags='
+        -Wall
+        -Wextra
+        -Werror=declaration-after-statement
+        -Werror=deprecated-declarations
+        -Werror=division-by-zero
+        -Werror=extra-tokens
+        -Werror=implicit-function-declaration
+        -Werror=implicit-int
+        -Werror=long-long
+        -Werror=pointer-arith
+        -Werror=shorten-64-to-32
+        -Werror=write-strings
+        -Wmissing-noreturn
+        -Wno-constant-logical-operand
+        -Wno-missing-field-initializers
+        -Wno-overlength-strings
+        -Wno-parentheses-equality
+        -Wno-self-assign
+        -Wno-tautological-compare
+        -Wno-unused-local-typedef
+        -Wno-unused-parameter
+        -Wunused-variable
+        '
+      - LDFLAGS=-Wno-unused-command-line-argument
+
+  - &rubyspec
+    name: ruby/spec on Ruby 2.3 # to ensure version guards are correctly added
+    <<: *linux
+    language: ruby
+    rvm: 2.3.8
+    addons:
+      apt:
+        packages:
+    before_install:
+    install:
+    before_script: chmod -R u+w spec/ruby
+    # -j randomly hangs. Using -fs to make sure we can know problematic spec on failure.
+    script: ruby -C spec/ruby ../mspec/bin/mspec -fs .
+
+  - &x86_64-darwin17
+    name: x86_64-darwin17
+    <<: *osx
+    env:
+      - CONFIG_FLAG=--with-opt-dir=/usr/local/opt/openssl@1.1:/usr/local/opt/zlib
+      - TEST_ALL_OPTS="$JOBS -q --tty=no --excludes=\$(TESTSDIR)/excludes/_travis/osx"
+
+  - &universal-darwin17
+    name: uinversal.x86_64h-darwin17
+    <<: *osx
+    <<: *cron-only
+    <<: *make-test-only
+    env:
+      - CONFIG_FLAG=--with-arch=x86_64h,x86_64,i386
+      - TEST_ALL_OPTS="$JOBS -q --tty=no --excludes=\$(TESTSDIR)/excludes/_travis/osx"
+
+matrix:
+  include:
+    # to reduce time for finishing all jobs, run the slowest osx build first.
+    - <<: *x86_64-darwin17
+    - <<: *x86_64-linux
+    - <<: *i686-linux
+    - <<: *jemalloc
+    - <<: *pedanticism
+    - <<: *ASAN
+    - <<: *MSAN
+    - <<: *UBSAN
+    - <<: *VM_CHECK_MODE
+    - <<: *FIBER_USE_sjlj
+    - <<: *TOKEN_THREADED_CODE
+    - <<: *CALL_THREADED_CODE
+    - <<: *NO_THREADED_CODE
+    - <<: *rubyspec
+  allow_failures:
+    - name: uinversal.x86_64h-darwin17
+    - name: -fsanitize=address
+    - name: -fsanitize=memory
+    - name: -fsanitize=undefined
+  fast_finish: true
+
+before_script:
+  - echo JOBS=${JOBS} SETARCH=${SETARCH}
+  - $SETARCH uname -a
+  - $SETARCH uname -r
+  - rm -fr .ext autom4te.cache
+  - echo $TERM
+  - |-
+    [ -d ~/.downloaded-cache ] ||
+    mkdir ~/.downloaded-cache
+  - ln -s ~/.downloaded-cache
+  - "> config.status"
+  - sed -f tool/prereq.status Makefile.in common.mk > Makefile
+  - make -s $JOBS update-download
+  - make -s $JOBS srcs
+  - rm config.status Makefile rbconfig.rb .rbconfig.time
+  - |-
+    if [ -d ~/config_2nd ]; then
+      cp -pr ~/config_2nd build
+    else
+      mkdir build
+    fi
+  - mkdir config_1st config_2nd
+  - chmod -R a-w .
+  - chmod -R u+w build config_1st config_2nd
+  - cd build
+  - ccache --show-stats
+  - |-
+    case "$CC" in
+    gcc*)   CC="ccache $CC${GCC_FLAGS:+ }$GCC_FLAGS -fno-diagnostics-color";;
+    clang*) CC="ccache $CC${GCC_FLAGS:+ }$GCC_FLAGS -fno-color-diagnostics";;
+    esac
+  - |-
+    [ ! -f config.cache ] ||
+    [ "$CC" = "`sed -n s/^ac_cv_prog_CC=//p config.cache`" ] ||
+    (set -x; exec rm config.cache)
+  - $SETARCH ../configure -C --disable-install-doc --prefix=$RUBY_PREFIX $CONFIG_FLAG
+  - cp -pr config.cache config.status .ext/include ../config_1st
+  - $SETARCH make reconfig
+  - cp -pr config.cache config.status .ext/include ../config_2nd
+  - (cd .. && exec diff -ru config_1st config_2nd)
+  - chmod u+w ..
+  - rm -rf ~/config_2nd
+  - mv ../config_2nd ~
+  - chmod u-w ..
+  - $SETARCH make -s $JOBS && make install
+  - ccache --show-stats
+  - |-
+    [ -z "${GEMS_FOR_TEST}" ] ||
+    $RUBY_PREFIX/bin/gem install --no-document $GEMS_FOR_TEST
+
+script:
+  - $SETARCH make -s test TESTOPTS="${TESTOPTS=$JOBS -q --tty=no}"
+  - $SETARCH make -s test-all -o exts TESTOPTS="${TEST_ALL_OPTS=$TESTOPTS}"
+  - $SETARCH make -s test-spec MSPECOPT=-ff # not using `-j` because sometimes `mspec -j` silently dies
+
+# Branch matrix.  Not all branches are Travis-ready so we limit branches here.
+branches:
+  only:
+    - trunk
+    - ruby_2_3
+    - ruby_2_4
+    - ruby_2_5
+    - /^feature\//
+    - /^bug\//
+    - /^travis-/
+
+# We want to be notified when something happens.
+notifications:
+  irc:
+    channels:
+      - "chat.freenode.net#ruby-core"
+      - "chat.freenode.net#ruby-ja"
+    on_success: change # [always|never|change] # default: always
+    on_failure: always # [always|never|change] # default: always
+    template:
+      - "%{message} by @%{author}: See %{build_url}"
+
+  slack:
+    rooms:
+      - secure: i1GLETSKye85ea6dGNA3MxI/5myChmMFiZtBd5C69xK+s1sBFqEgOSbaSf9KHc0CYrHVyNhQMaZRruieV7xS+6Pfs0Zvxf1DO6QQTWC2KhkqwFDLvZncAzjoyASdR90hbr+iRPOngQ+HJuE94zemALAwEqNAinzA74PMiJXktqY= # ruby:<token>#commits
+      - secure: ah7UEHBvncXT7bM5mvYIQAO+tIyV/wl7nXLb7wQD16dO2v8Gragy0mWjB79Q09hrrMGmp6H9bCDpdGS80boIA5EHaHoG4QaP0i9bsSt8U2AMWgZtfyIgQKJ4H2kXkGlrjO+AXTgnIkP7LNjdgAVUUTGQPb26T3QmoN2Splt+fIQ= # ruby:<token>#alerts
+    on_pull_requests: false
+    on_success: change
+    on_failure: always
+
+  email:
+    - ko1c-failure@atdot.net
+
+# Local Variables:
+# mode: YAML
+# coding: utf-8-unix
+# indent-tabs-mode: nil
+# tab-width: 4
+# fill-column: 79
+# default-justification: full
+# End:
diff --git a/BSDL b/BSDL
new file mode 100644
index 0000000000..a009caefea
--- /dev/null
+++ b/BSDL
@@ -0,0 +1,22 @@
+Copyright (C) 1993-2013 Yukihiro Matsumoto. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions
+are met:
+1. Redistributions of source code must retain the above copyright
+notice, this list of conditions and the following disclaimer.
+2. Redistributions in binary form must reproduce the above copyright
+notice, this list of conditions and the following disclaimer in the
+documentation and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+SUCH DAMAGE.
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
new file mode 100644
index 0000000000..ffdf2dd4b8
--- /dev/null
+++ b/CONTRIBUTING.md
@@ -0,0 +1,4 @@
+Please see the [official issue tracker] and wiki [HowToContribute].
+
+[official issue tracker]: https://bugs.ruby-lang.org
+[HowToContribute]: https://bugs.ruby-lang.org/projects/ruby/wiki/HowToContribute
diff --git a/COPYING b/COPYING
index 9043404a83..f06056fb45 100644
--- a/COPYING
+++ b/COPYING
@@ -1,6 +1,6 @@
 Ruby is copyrighted free software by Yukihiro Matsumoto <matz@netlab.jp>.
-You can redistribute it and/or modify it under either the terms of the GPL
-version 2 (see the file GPL), or the conditions below:
+You can redistribute it and/or modify it under either the terms of the
+2-clause BSDL (see the file BSDL), or the conditions below:
 
   1. You may make and give away verbatim copies of the source form of the
      software without restriction, provided that you duplicate all of the
@@ -11,8 +11,8 @@ version 2 (see the file GPL), or the conditions below:
 
        a) place your modifications in the Public Domain or otherwise
           make them Freely Available, such as by posting said
-	  modifications to Usenet or an equivalent medium, or by allowing
-	  the author to include your modifications in the software.
+          modifications to Usenet or an equivalent medium, or by allowing
+          the author to include your modifications in the software.
 
        b) use the modified software only within your corporation or
           organization.
@@ -26,11 +26,11 @@ version 2 (see the file GPL), or the conditions below:
      provided that you do at least ONE of the following:
 
        a) distribute the binaries and library files of the software,
-	  together with instructions (in the manual page or equivalent)
-	  on where to get the original distribution.
+          together with instructions (in the manual page or equivalent)
+          on where to get the original distribution.
 
        b) accompany the distribution with the machine-readable source of
-	  the software.
+          the software.
 
        c) give non-standard binaries non-standard names, with
           instructions on where to get the original software distribution.
@@ -44,9 +44,9 @@ version 2 (see the file GPL), or the conditions below:
      For the list of those files and their copying conditions, see the
      file LEGAL.
 
-  5. The scripts and library files supplied as input to or produced as 
+  5. The scripts and library files supplied as input to or produced as
      output from the software do not automatically fall under the
-     copyright of the software, but belong to whomever generated them, 
+     copyright of the software, but belong to whomever generated them,
      and may be sold commercially, and may be aggregated with this
      software.
 
diff --git a/COPYING.ja b/COPYING.ja
index 933cc7cb9a..e50d01c8d1 100644
--- a/COPYING.ja
+++ b/COPYING.ja
@@ -1,51 +1,51 @@
-$BK\%W%m%0%i%`$O%U%j!<%=%U%H%&%'%"$G$9!%(BGPL(the GNU General
-Public License)$B$^$?$O0J2<$K<($9>r7o$GK\%W%m%0%i%`$r:FG[I[$G(B
-$B$-$^$9!%(BGPL$B$K$D$$$F$O(BGPL$B%U%!%$%k$r;2>H$7$F2<$5$$!%(B
+本プログラムはフリーソフトウェアです．2-clause BSDL
+または以下に示す条件で本プログラムを再配布できます
+2-clause BSDLについてはBSDLファイルを参照して下さい．
 
-  1. $BJ#@=$O@)8B$J$/<+M3$G$9!%(B
+  1. 複製は制限なく自由です．
 
-  2. $B0J2<$N>r7o$N$$$:$l$+$rK~$?$9;~$KK\%W%m%0%i%`$N%=!<%9$r(B
-     $B<+M3$KJQ99$G$-$^$9!%(B
+  2. 以下の条件のいずれかを満たす時に本プログラムのソースを
+     自由に変更できます．
 
-     (a) $B%M%C%H%K%e!<%:$K%]%9%H$7$?$j!$:n<T$KJQ99$rAwIU$9$k(B
-         $B$J$I$NJ}K!$G!$JQ99$r8x3+$9$k!%(B
+     (a) ネットニューズにポストしたり，作者に変更を送付する
+         などの方法で，変更を公開する．
 
-     (b) $BJQ99$7$?K\%W%m%0%i%`$r<+J,$N=jB0$9$kAH?%FbIt$@$1$G(B
-         $B;H$&!%(B
+     (b) 変更した本プログラムを自分の所属する組織内部だけで
+         使う．
 
-     (c) $BJQ99E@$rL@<($7$?$&$(!$%=%U%H%&%'%"$NL>A0$rJQ99$9$k!%(B
-         $B$=$N%=%U%H%&%'%"$rG[I[$9$k;~$K$OJQ99A0$NK\%W%m%0%i(B
-         $B%`$bF1;~$KG[I[$9$k!%$^$?$OJQ99A0$NK\%W%m%0%i%`$N%=!<(B
-         $B%9$NF~<jK!$rL@<($9$k!%(B
+     (c) 変更点を明示したうえ，ソフトウェアの名前を変更する．
+         そのソフトウェアを配布する時には変更前の本プログラ
+         ムも同時に配布する．または変更前の本プログラムのソー
+         スの入手法を明示する．
 
-     (d) $B$=$NB>$NJQ99>r7o$r:n<T$H9g0U$9$k!%(B
+     (d) その他の変更条件を作者と合意する．
 
-  3. $B0J2<$N>r7o$N$$$:$l$+$rK~$?$9;~$KK\%W%m%0%i%`$r%3%s%Q%$(B
-     $B%k$7$?%*%V%8%'%/%H%3!<%I$d<B9T7A<0$G$bG[I[$G$-$^$9!%(B
+  3. 以下の条件のいずれかを満たす時に本プログラムをコンパイ
+     ルしたオブジェクトコードや実行形式でも配布できます．
 
-     (a) $B%P%$%J%j$r<u$1<h$C$??M$,%=!<%9$rF~<j$G$-$k$h$&$K!$(B
-         $B%=!<%9$NF~<jK!$rL@<($9$k!%(B
+     (a) バイナリを受け取った人がソースを入手できるように，
+         ソースの入手法を明示する．
 
-     (b) $B5!3#2DFI$J%=!<%9%3!<%I$rE:IU$9$k!%(B
+     (b) 機械可読なソースコードを添付する．
 
-     (c) $BJQ99$r9T$C$?%P%$%J%j$OL>A0$rJQ99$7$?$&$(!$%*%j%8%J(B
-         $B%k$N%=!<%9%3!<%I$NF~<jK!$rL@<($9$k!%(B
+     (c) 変更を行ったバイナリは名前を変更したうえ，オリジナ
+         ルのソースコードの入手法を明示する．
 
-     (d) $B$=$NB>$NG[I[>r7o$r:n<T$H9g0U$9$k!%(B
+     (d) その他の配布条件を作者と合意する．
 
-  4. $BB>$N%W%m%0%i%`$X$N0zMQ$O$$$+$J$kL\E*$G$"$l<+M3$G$9!%$?(B
-     $B$@$7!$K\%W%m%0%i%`$K4^$^$l$kB>$N:n<T$K$h$k%3!<%I$O!$$=(B
-     $B$l$>$l$N:n<T$N0U8~$K$h$k@)8B$,2C$($i$l$k>l9g$,$"$j$^$9!%(B
+  4. 他のプログラムへの引用はいかなる目的であれ自由です．た
+     だし，本プログラムに含まれる他の作者によるコードは，そ
+     れぞれの作者の意向による制限が加えられる場合があります．
 
-     $B$=$l$i%U%!%$%k$N0lMw$H$=$l$>$l$NG[I[>r7o$J$I$KIU$$$F$O(B
-     LEGAL$B%U%!%$%k$r;2>H$7$F$/$@$5$$!%(B
+     それらファイルの一覧とそれぞれの配布条件などに付いては
+     LEGALファイルを参照してください．
 
-  5. $BK\%W%m%0%i%`$X$NF~NO$H$J$k%9%/%j%W%H$*$h$S!$K\%W%m%0%i(B
-     $B%`$+$i$N=PNO$N8"Mx$OK\%W%m%0%i%`$N:n<T$G$O$J$/!$$=$l$>(B
-     $B$l$NF~=PNO$r@8@.$7$??M$KB0$7$^$9!%$^$?!$K\%W%m%0%i%`$K(B
-     $BAH$_9~$^$l$k$?$a$N3HD%%i%$%V%i%j$K$D$$$F$bF1MM$G$9!%(B
+  5. 本プログラムへの入力となるスクリプトおよび，本プログラ
+     ムからの出力の権利は本プログラムの作者ではなく，それぞ
+     れの入出力を生成した人に属します．また，本プログラムに
+     組み込まれるための拡張ライブラリについても同様です．
 
-  6. $BK\%W%m%0%i%`$OL5J]>Z$G$9!%:n<T$OK\%W%m%0%i%`$r%5%]!<%H(B
-     $B$9$k0U;V$O$"$j$^$9$,!$%W%m%0%i%`<+?H$N%P%0$"$k$$$OK\%W(B
-     $B%m%0%i%`$N<B9T$J$I$+$iH/@8$9$k$$$+$J$kB;32$KBP$7$F$b@U(B
-     $BG$$r;}$A$^$;$s!%(B
+  6. 本プログラムは無保証です．作者は本プログラムをサポート
+     する意志はありますが，プログラム自身のバグあるいは本プ
+     ログラムの実行などから発生するいかなる損害に対しても責
+     任を持ちません．
diff --git a/ChangeLog b/ChangeLog
deleted file mode 100644
index fb25146591..0000000000
--- a/ChangeLog
+++ /dev/null
@@ -1,18598 +0,0 @@
-Sun Jun 15 19:19:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (iconv_convert): check upper bound.  a patch from
-	  Daniel Luz at [ruby-Bugs-17910].
-
-Sun Jun 15 19:07:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (ftruncate): check if available.
-
-	* file.c (rb_file_truncate): check if ftruncate instead of truncate.
-
-Sun Jun 15 18:59:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (sigsetmask): check when signal semantics is not POSIX.
-
-	* signal.c (USE_TRAP_MASK): set true if sigprocmask or sigsetmask is
-	  available.
-
-Sat Jun 14 16:12:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/timeout.rb (Timeout::timeout): made sensitive to location on the
-	  stack.  [ruby-core:15458]
-
-Fri Jun 13 13:10:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/dl/ptr.c (dlmem_each_i): typo fixed.  a patch from IKOMA
-	  Yoshiki <ikoma@mb.i-chubu.ne.jp> in [ruby-dev:33776].
-
-Fri Jun 13 12:10:58 2008  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
-
-	* gc.c (rb_newobj): prohibit call of rb_newobj() during gc.
-	  Submitted by Sylvain Joyeux [ruby-core:12099].
-
-	* ext/dl/ptr.c: do not use LONG2NUM() inside dlptr_free().
-	  Slightly modified fix bassed on a patch by Sylvain Joyeux
-	  [ruby-core:12099] [ ruby-bugs-11859 ] [ ruby-bugs-11882 ]
-	  [ ruby-patches-13151 ].
-
-Fri Jun 13 12:09:02 2008  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* lib/benchmark.rb (Job::Benchmark#item): fix typo.
-
-Fri Jun 13 12:03:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): use strtod() for more
-	  precision.  [ruby-talk:290296]
-
-	* ext/bigdecimal/bigdecimal.c (BASE_FIG): made constant.
-
-	* ext/bigdecimal/extconf.rb: ditto.  [ruby-dev:33658]
-
-Fri Jun 13 11:58:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/irb.rb (IRB::Irb::eval_input): rescues Interrupt and other than
-	  SystemExit and SignalException.  [ruby-core:15359]
-
-Fri Jun 13 11:56:28 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/benchmark.rb (Benchmark::realtime): make Benchmark#realtime
-	  a bit faster.  a patch from Alexander Dymo <dymo@ukrpost.ua> in
-	  [ruby-core:15337].
-
-Fri Jun 13 11:42:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_open_file): should check NUL in path.
-	  <http://www.rubyist.net/~matz/20080125.html#c01>.
-
-	* io.c (rb_io_s_popen): ditto.
-
-	* io.c (rb_io_reopen): ditto.
-
-	* io.c (next_argv): ditto.
-
-	* io.c (rb_io_s_foreach): ditto.
-
-	* io.c (rb_io_s_readlines): ditto.
-
-	* io.c (rb_io_s_read): ditto.
-
-Wed Jun 11 15:06:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/uri/generic.rb (URI::Generic::inspect): use Kernel#to_s instead
-	 object_id with printf.  [ruby-dev:33347]
-
-Wed Jun 11 14:58:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in: Remove wrong assumptions about Cygwin.  a patch from
-	  Corinna Vinschen in [ruby-Bugs-17018].
-
-Mon Jun  9 18:03:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (eval): check if backtrace is empty.  [ruby-core:15040]
-
-Sun Jun  8 06:07:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_define_alloc_func, rb_undef_alloc_func): should
-	  define/undef on a signleton class.  [ruby-core:09959]
-
-Sun Jun  8 06:03:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* time.c (time_arg): use converted object.  [ruby-core:14759]
-
-Sun Jun  8 06:01:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (fptr_finalize): clear errno first.  [ruby-talk:284492]
-
-Sun Jun  8 05:53:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (TIMEZONE_VOID): check whether timezone requires zero
-	  arguments.  [ruby-dev:32631]
-
-Sun Jun  8 05:35:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (f_rest_arg): check if duplicated.  [ruby-core:14140]
-
-Sun Jun  8 05:21:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (RUBY_CHECK_VARTYPE): check if a variable is defined
-	  and its type.
-
-	* configure.in (timezone, altzone): check for recent cygwin.
-
-	* missing/strftime.c (strftime): fix for timezone.  [ruby-dev:32536]
-
-	* lib/mkmf.rb (try_var): should fail for functions.
-
-	* ext/readline/extconf.rb: should use have_func for functions instead
-	  of have_var.
-
-Sun Jun  8 05:09:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/uri/common.rb (URI::REGEXP::PATTERN): typo in REG_NAME
-	  regular expression.  a patch from Ueda Satoshi
-	  <s-ueda AT livedoor.jp>.  [ruby-dev:32514]
-
-Sun Jun  8 05:05:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/cgi.rb (read_multipart): exclude blanks from header values.
-	  [ruby-list:44327]
-
-Sun Jun  8 05:00:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bignum.c (rb_cstr_to_inum): trailing spaces may exist at sqeezing
-	  preceeding 0s.  [ruby-core:13873]
-
-Sun Jun  8 04:56:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (error_print): put newline unless multiple line message ends
-	  with a newline.  [ruby-dev:32429]
-
-Sun Jun  8 04:54:36 2008  James Edward Gray II  <jeg2@ruby-lang.org>
-
-	Merged 14070 from trunk.
-	
-	* lib/xmlrpc/server.rb (XMLRPC::Server#server): Improve signal handling so
-	  pressing control-c in the controlling terminal or sending SIGTERM stops
-	  the XML-RPC server.
-
-Sun Jun  8 04:46:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (newline_node): set line from outermost node before removing
-	  NODE_BEGIN.  [ruby-dev:32406]
-
-Sun Jun  8 04:37:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (stmt): remove unnecessary NODE_BEGIN.  [ruby-core:13814]
-
-Sun Jun  8 04:18:50 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (nil_plus): remove unused function.  [ruby-core:13737]
-
-Sun Jun  8 04:15:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_alias): do not call hook functions until initialization
-	  finishes.  [ruby-talk:279538]
-
-Sun Jun  8 04:12:50 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_invoke): bug fix. [ruby-talk:279100]
-
-Sun Jun  8 03:54:52 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/curses/extconf.rb: check macro if cannot find func.
-	  [ruby-list:44224]
-
-Sun Jun  8 03:50:40 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session::FileStore::restore): use
-	  lockfile for exclusive locks.  a patch from <tommy AT tmtm.org>.
-	  [ruby-dev:32296]
-
-Sun Jun  8 03:46:55 2008  Tanaka Akira  <akr@fsij.org>
-
-	* missing/isinf.c (isinf): don't define if the macro is defined.
-
-Sun Jun  8 03:34:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* numeric.c (round): fallback definition.
-
-	* numeric.c (flo_divmod, flo_round): use round() always.
-	  [ruby-dev:32269]
-
-Sun Jun  8 03:34:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* numeric.c (flodivmod): work around for infinity.
-
-	* numeric.c (flo_divmod): work around for platforms have no round().
-	  [ruby-dev:32247]
-
-Sun Jun  8 03:34:23 2008  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
-
-	* numeric.c (flo_divmod): round to the nearest integer.
-	  [ ruby-Bugs-14540 ]
-
-Sun Jun  8 03:08:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rexml/encodings/SHIFT-JIS.rb (REXML::Encoding): place -x for
-	  nkf conversion.  a patch from <moonwolf AT moonwolf.com>.
-	  [ruby-dev:32183]
-
-Sun Jun  8 03:05:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Switch::summarize): fix for long form
-	  option with very long argument.  a patch from Kobayashi Noritada
-	  <nori1 AT dolphin.c.u-tokyo.ac.jp> in [ruby-list:44179].
-
-Sun Jun  8 03:03:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* numeric.c (fix_pow): returns infinity for 0**-1.  [ruby-dev:32084]
-
-Sun Jun  8 02:57:32 2008  James Edward Gray II  <jeg2@ruby-lang.org>
-
-	Merged 13781 from trunk.
-
-	* lib/net/telnet.rb (Net::Telnet#login): Allowing "passphrase" in
-	  addition to "password" for Telnet login prompts. [ruby-Bugs-10746]
-
-Wed Oct 25 06:46:21 2007  James Edward Gray II  <jeg2@ruby-lang.org>
-
-	Merged 13779 from trunk.
-
-	* lib/net/telnet.rb (Net::Telnet#login): Making the password prompt
-	  pattern case insensitive. [ruby-Bugs-10746]
-
-Sun Jun  8 02:54:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_tell, rb_io_seek): check errno too.  [ruby-dev:32093]
-
-Sun Jun  8 01:51:52 2008  James Edward Gray II  <jeg2@ruby-lang.org>
-
-	Merged 13767, 13768, 13769, and 13770 from trunk.
-
-	* lib/xmlrpc/parser.rb (XMLRPC::Convert::dateTime): Fixing a bug that
-	  caused time zone conversion to fail for some ISO 8601 date formats.
-	  [ruby-Bugs-12677]
-
-	* lib/xmlrpc/client.rb (XMLRPC::Client#do_rpc): Explicitly start
-	  the HTTP connection to support keepalive requests. [ruby-Bugs-9353]
-
-	* lib/xmlrpc/client.rb (XMLRPC::Client#do_rpc): Improving the error
-	  message for Content-Type check failures. [ruby-core:12163]
-
-	* lib/xmlrpc/utils.rb (XMLRPC::ParseContentType#parse_content_type):
-	  Making Content-Type checks case insensitive. [ruby-Bugs-3367]
-
-Sun Jun  8 01:45:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* marshal.c (r_bytes0): refined length check.  [ruby-dev:32059]
-
-Sun Jun  8 01:45:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* marshal.c (r_bytes0): check if source has enough data.
-	  [ruby-dev:32054]
-
-Sun Jun  8 01:37:58 2008  Tanaka Akira  <akr@fsij.org>
-
-	* ext/socket/socket.c (s_accept_nonblock): make accepted fd
-	  nonblocking.  [ruby-talk:274079]
-
-Sun Jun  8 01:28:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/mkexports.rb: deal with __fastcall name decorations.
-	  [ruby-list:44111]
-
-Sun Jun  8 01:23:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rexml/source.rb (REXML::SourceFactory::SourceFactory): typo
-	  fixed.  [ruby-list:44099]
-
-Sun Jun  8 01:17:45 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* {bcc32,win32}/Makefile.sub (COMMON_MACROS): workaround for old SDK's
-	  bug. [ruby-core:12584]
-
-Sun Jun  8 00:56:44 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/net/http.rb, lib/open-uri.rb: remove
-	  Net::HTTP#enable_post_connection_check.  [ruby-dev:31960]
-
-	* lib/net/imap.rb: hostname should be verified against server's
-	  indentity as persented in the server's certificate. [ruby-dev:31960]
-
-	* ext/openssl/lib/net/telnets.rb, ext/openssl/lib/net/ftptls.rb: ditto.
-
-Thu Jun  5 16:19:27 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (make_cmdvector): adjust escaped successive
-	  double-quote handling. (merge from trunk)
-
-Thu Jun  5 12:24:25 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (init_env): initialize HOME and USER environment
-	  variables unless set. [ruby-core:12328] (merge from trunk)
-
-	* win32/win32.c (NtInitialize, getlogin): ditto.
-
-	* configure.in, win32/Makefile.sub (LIBS): need to link shell32
-	  library for SH* functions on mswin32 and mingw32.
-
-Thu Jun  5 12:21:06 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (id2ref): valid id should not refer T_VALUE nor T_ICLASS.
-	  [ruby-dev:31911]
-
-Wed Jun  4 16:39:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* Makefile.in (ext/extinit.o): use $(OUTFLAG) as well as other
-	  objects.  [ruby-Bugs-14228]
-
-Tue Jun  3 16:13:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (yyerror): limit error message length.  [ruby-dev:31848]
-
-	* regex.c (re_mbc_startpos): separated from re_adjust_startpos.
-
-Tue Jun  3 15:27:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* gc.c (os_obj_of, os_each_obj): hide objects to be finalized.
-	  [ruby-dev:31810]
-
-Wed Jun  4 19:15:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (remove_method): should not remove undef place holder.
-	  [ruby-dev:31817]
-
-Tue Jun  3 15:05:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (struct rb_exec_arg): proc should be a VALUE.
-
-	* process.c (rb_f_exec): suppress a warning.
-
-	* process.c (rb_detach_process): cast for the platforms where size of
-	  pointer differs from size of int.
-
-	* process.c (rb_f_exec, rb_f_system): should not exceptions after
-	  fork.  [ruby-core:08262]
-
-Wed May 21 01:32:56 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpservlet/filehandler.rb: should normalize path
-	  name in path_info to prevent script disclosure vulnerability on
-	  DOSISH filesystems. (fix: CVE-2008-1891)
-	  Note: NTFS/FAT filesystem should not be published by the platforms
-	  other than Windows. Pathname interpretation (including short
-	  filename) is less than perfect.
-
-	* lib/webrick/httpservlet/abstract.rb
-	  (WEBrick::HTTPServlet::AbstracServlet#redirect_to_directory_uri):
-	  should escape the value of Location: header.
-
-	* lib/webrick/httpservlet/cgi_runner.rb: accept interpreter
-	  command line arguments.
-
-Sun May 18 01:57:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (isdirsep): backslash is valid path separator on cygwin too.
-	  backported from 1.8 HEAD.
-
-Sat May 17 23:53:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (file_expand_path): fix for short file name on Cygwin.
-
-Sat May 17 23:50:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (OpenFile): prevent conflict on Windows.
-
-	* file.c (USE_NTFS): fixed merge miss.
-
-Sat May 17 12:36:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (file_expand_path): rb_str_set_len is not backported.
-
-Sat May 17 12:15:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (file_expand_path): support for alternative data stream
-	  and ignored trailing garbages of NTFS.
-
-	* file.c (rb_file_s_basename): ditto.
-
-	* file.c (rb_file_s_extname): ditto.
-
-Mon Mar  3 23:36:41 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpservlet/filehandler.rb: should normalize path
-	  separators in path_info to prevent directory traversal attacks
-	  on DOSISH platforms.
-	  reported by Digital Security Research Group [DSECRG-08-026].
-
-	* lib/webrick/httpservlet/filehandler.rb: pathnames which have
-	  not to be published should be checked case-insensitively.
-
-Sun Sep 23 21:57:25 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>  
-
-	* lib/net/http.rb: an SSL verification (the server hostname should
-	  be matched with its certificate's commonName) is added.
-	  this verification can be skipped by
-	  "Net::HTTP#enable_post_connection_check=(false)".
-	  suggested by Chris Clark <cclark at isecpartners.com>
-
-	* lib/net/open-uri.rb: use Net::HTTP#enable_post_connection_check to
-	  perform SSL post connection check.
-
-	* ext/openssl/lib/openssl/ssl.c
-	(OpenSSL::SSL::SSLSocket#post_connection_check): refine error message.
-
-Sun Sep 23 06:08:38 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_init): separate from strio_initialize
-	  to share with strio_reopen properly.  [ruby-Bugs-13919]
-
-Sun Sep 23 05:42:35 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* lib/rdoc/options.rb (Options::check_diagram): dot -V output
-	  changed. [ ruby-Bugs-11978 ], Thanks Florian Frank.
-
-Mon Sep 17 04:47:02 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ruby.c (proc_options): -W should be allowed in RUBYOPT
-	  environment variable.  [ruby-core:12118]
-
-Mon Sep 17 04:31:46 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* range.c (range_step): fixed integer overflow.  [ruby-dev:31763]
-
-Fri Sep  7 14:57:36 2007  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
-
-	* ruby.c (rubylib_mangled_path): eliminate RSTRING_PTR
-	  [ruby-dev:31679]
-
-	* ruby.c(push_include_cygwin): ditto.
-
-Fri Sep  7 14:32:38 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_subseq): need integer overflow check.
-	  [ruby-dev:31736]
-
-	* array.c (rb_ary_splice): ditto.  [ruby-dev:31737]
-
-	* array.c (rb_ary_fill): ditto.  [ruby-dev:31738]
-
-	* string.c (rb_str_splice): integer overflow for length.
-	  [ruby-dev:31739]
-
-Fri Sep  7 14:27:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (mnew): should preserve noex as safe_level.
-
-	* eval.c (rb_call0): tighten security check condition..
-
-Fri Sep  7 14:19:16 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (group_member): check if presents.
-
-	* configure.in (XCFLAGS): add _GNU_SOURCE on linux.
-
-	* file.c (group_member): use system routine if available.
-
-Fri Sep  7 14:14:19 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* include/ruby/defines.h (flush_register_windows): call "ta 0x03"
-	  even on Linux/Sparc.  [ruby-dev:31674]
-
-Fri Sep  7 14:12:37 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_type_progid, reg_enum_key, 
-	  reg_get_val, ole_wc2mb): fix the bug. Thanks, arton. 
-	  [ruby-dev:31576]
-
-Fri Sep  7 14:02:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (detach_process_watcher): should not pass the pointer
-	  to an auto variable to the thread to be created.  pointed and
-	  fix by KUBO Takehiro <kubo at jiubao.org>  [ruby-dev:30618]
-
-Fri Sep  7 13:52:36 2007  Tanaka Akira  <akr@fsij.org>
-
-	* bignum.c (big_lshift): make shift offset long type.
-	  (big_rshift): ditto.
-	  (rb_big_lshift): ditto.
-	  (big_rshift): ditto.
-	  [ruby-dev:31434]
-
-Thu Aug 16 08:44:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* hash.c (rb_hash_delete_key): delete the entry without calling block.
-
-	* hash.c (rb_hash_shift): should consider iter_lev too.
-
-	* hash.c (delete_if_i): use rb_hash_delete_key() so that the block
-	  isn't called twice.  [ruby-core:11556]
-
-Thu Aug 16 08:43:50 2007  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/tuplespace.rb: fix Rinda::TupleSpace keeper thread bug.
-	  the thread is started too early. [ruby-talk:264062] 
-
-	* test/rinda/test_rinda.rb: ditto.
-
-Thu Aug 16 08:40:38 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/pty/pty.c (establishShell): handshaking before close slave
-	  device.  [ruby-talk:263410]
-
-Thu Aug 16 08:29:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (ac_cv_func_isinf): set yes also on OpenSolaris.
-	  [ruby-Bugs-12859]
-
-Thu Aug 16 08:28:16 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/rexml/encodings/{ISO-8859-15,CP-1252}.rb: fixed invalid syntax.
-
-Thu Aug 16 08:26:08 2007  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/README: fixed a typo.
-
-Thu Aug 16 08:20:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): save all CONFIG values.
-
-	* ext/extmk.rb (extmake): remove mkmf.log at clean, and extconf.h at
-	  distclean, respectively.
-
-	* ext/extmk.rb: remove rdoc at clean, and installed list file at
-	  distclean, respectively.
-
-Thu Aug 16 07:58:18 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): should not check positional number as
-	  width.  [ruby-core:11838]
-
-Thu Aug 16 07:52:24 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dln.c (conv_to_posix_path): removed.
-
-	* ruby.c (rubylib_mangled_path, rubylib_mangled_path2): return
-	  VALUE instead of a pointer to static buffer.
-
-	* ruby.c (push_include_cygwin): fixed buffer overflow.
-	  [ruby-dev:31297]
-
-	* ruby.c (ruby_init_loadpath): not convert built-in paths.
-
-Thu Aug 16 07:51:37 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* defines.h: Pull the RUBY_MBCHAR_MAXSIZE definition from trunk,
-	  which is necessary for dir.c to compile on djgpp and emx.
-
-Thu Aug 16 07:42:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* intern.h (is_ruby_native_thread): removed since declared as an int
-	  function in ruby.h already.
-
-Thu Aug 16 07:40:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_s_rename): deleted code to get rid of a bug of
-	  old Cygwin.
-
-	* file.c (rb_file_truncate): added prototype of GetLastError()
-	  on cygwin.  [ruby-dev:31239]
-
-	* intern.h (is_ruby_native_thread): prototype.
-
-	* missing/strftime.c (strftime): fix printf format and actual
-	  arguments.
-
-	* ext/Win32API/Win32API.c (Win32API_initialize): ditto.
-
-	* ext/tk/tcltklib.c (ip_finalize): ditto.
-
-	* ext/dl/ptr.c (rb_dlptr_inspect): ditto.  [ruby-dev:31268]
-
-	* ext/dl/sym.c (rb_dlsym_inspect): ditto.
-
-	* ext/socket/getnameinfo.c: include stdio.h always.
-
-	* ext/win32ole/win32ole.c (ole_hresult2msg, folevariable_name,
-	  folevariable_ole_type, folevariable_ole_type_detail,
-	  folevariable_value, folemethod_visible): missing return value.
-
-Thu Aug 16 07:32:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): make OBJS depend on RUBY_EXTCONF_H
-	  only if extconf.h is created.
-
-Thu Aug 16 07:29:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* {win32,wince,bcc32}/setup.mak (-version-): no RUBY_EXTERN magic.
-
-Thu Aug 16 06:43:03 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (init_mkmf): should remove mkmf.log too.
-
-Thu Aug 16 06:40:58 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/openssl/ossl_config.c (ossl_config_set_section): do not
-	  initialize aggregations with dynamic values.  [ruby-talk:259306]
-
-Thu Aug 16 06:39:19 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (get_backtrace): check the result more.
-	  [ruby-dev:31261] [ruby-bugs-12398]
-
-Thu Aug 16 06:32:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bignum.c (rb_big_lshift, rb_big_rshift): separated functions
-	  to get rid of infinite recursion.  fixed calculation in edge
-	  cases.  [ruby-dev:31244]
-
-	* numeric.c (rb_fix_lshift, rb_fix_rshift): ditto.
-
-Thu Aug 16 06:26:58 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bignum.c (rb_big_pow): refine overflow check.  [ruby-dev:31242]
-
-Thu Aug 16 06:25:48 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* time.c (time_succ): Time#succ should return a time object in the
-	  same timezone mode to the original.  [ruby-talk:260256]
-
-Thu Aug 16 06:24:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* numeric.c (fix_pow): integer power calculation: 0**n => 0, 
-	  1**n => 1, -1**n => 1 (n: even) / -1 (n: odd).
-
-Thu Aug 16 06:11:34 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/base64.rb (Base64::b64encode): should not specify /o option
-	  for regular expression.  [ruby-dev:31221]
-
-Thu Aug 16 06:08:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): more checks for format argument.
-	  [ruby-core:11569], [ruby-core:11570], [ruby-core:11571],
-	  [ruby-core:11573]
-
-Thu Aug 16 05:39:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bignum.c (bignorm): do not empty Bignum.  [ruby-dev:31229]
-
-Thu Aug 16 05:12:05 2007  pegacorn  <subscriber.jp AT gmail.com>
-
-	* ext/openssl/ossl.h: include ossl_pkcs5.h.  [ruby-dev:31231]
-
-	* ext/openssl/ossl_pkcs5.h: new file for PKCS5.  [ruby-dev:31231]
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_to_s): use ossl_raise()
-	  instead of rb_raise().  [ruby-dev:31222]
-
-	* ext/sdbm/_sdbm.c: DOSISH platforms need io.h.  [ruby-dev:31232]
-
-	* ext/syck/syck.h: include stdlib.h for malloc() and free().
-	  [ruby-dev:31232]
-
-	* ext/syck/syck.h (syck_parser_set_input_type): prototype added.
-	  [ruby-dev:31231]
-
-	* win32/win32.c: include mbstring.h for _mbspbrk().  [ruby-dev:31232]
-
-	* win32.h (rb_w32_getcwd): prototype added.  [ruby-dev:31232]
-
-Thu Aug 16 05:02:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bignum.c (rb_cstr_to_inum): check leading non-digits.
-	  [ruby-core:11691]
-
-Thu Aug 16 05:00:01 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* numeric.c (fix_pow): 0**2 should not raise floating point
-	  exception.  [ruby-dev:31216]
-
-Thu Aug 16 04:56:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/win32.c (CreateChild): enclose command line except for
-	  command.com which can not handle quotes.  [ruby-talk:258939]
-
-Thu Aug 16 04:54:45 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (link_command, cc_command, cpp_command): do not expand
-	  ::CONFIG which is an alias of MAKEFILE_CONFIG.
-
-Thu Aug 16 04:53:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* struct.c (rb_struct_init_copy): disallow changing the size.
-	  [ruby-dev:31168]
-
-Thu Aug 16 04:52:11 2007  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* random.c: documentation fix.  srand(0) initializes PRNG with '0',
-	  not with random_seed.
-
-Thu Aug 16 04:49:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bcc32/{Makefile.sub,setup.mak}: remove surplus slash from srcdir.
-
-Thu Aug 16 04:40:37 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): sign bit extension should not be done
-	  if FPLUS flag is specified.  [ruby-list:39224]
-
-Thu Aug 16 04:39:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_initialize): should call rb_ary_modify() first.
-	  [ruby-core:11562]
-
-Thu Aug 16 04:38:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (yylex): return non-valid token for an invalid
-	  instance/class variable name.  a patch from from Yusuke ENDOH
-	  <mame AT tsg.ne.jp>.  [ruby-dev:31095]
-
-Thu Aug 16 04:36:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (dsym): return non-null NODE even if yyerror().  based on a
-	  patch from from Yusuke ENDOH <mame AT tsg.ne.jp>.  [ruby-dev:31085]
-
-Thu Aug 16 04:34:56 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (proc_exec_v, rb_proc_exec): preserve errno.
-
-Thu Aug 16 04:30:45 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* variable.c (rb_path2class): get rid of dangling pointer caused by
-	  optimized out value.
-
-Thu Aug 16 04:24:28 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/dl/lib/dl/win32.rb: seems that dl doesn't accept void argument.
-	  fixed [ruby-bugs:PR#5489].
-
-Thu Aug 16 04:23:44 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser): handle more
-	  extensions.  [ruby-dev:30972]
-
-Thu Aug 16 04:14:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (ruby_setreuid, ruby_setregid): rename to get rid of name
-	  clash.
-Thu Aug 16 04:11:17 2007  Ryan Davis  <ryand@zenspider.com>
-
-	* lib/rexml/dtd/dtd.rb: Fixed typo in code. Fixes bug #10420
-
-Thu Aug 16 04:08:20 2007  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (ResponseParser#next_token): fixed
-	  error message. (backported from HEAD)
-
-	* lib/net/imap.rb (ResponseParser#parse_error): fixed
-	  the condition not to refer @token.symbol unexpectedly.
-	  Thanks, Dick Monahan. (backported from HEAD)
-
-Thu Aug 16 04:05:20 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_extended): erroneous check condition when dump
-	  method is defined.  [ruby-core:10646]
-
-Sat Jun  9 10:40:00 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* stable version 1.8.5-p52 released.
-
-Thu Jun  7 14:53:46 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* eval.c (method_inspect): show proper class name.
-	  [ruby-talk:248647], Thanks Calamitas.
-
-Mon May 28 19:37:24 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (move_to_next_entry): revert r12338. not necessary
-	  on ruby_1_8_5.
-
-Sun May 27 05:52:37 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* eval.c (mnew): call of super via a method object should work again.
-	  [ruby-talk:248647], Thanks Calamitas.
-
-	* test/ruby/test_method.rb (TestMethod::test_method_super): test for
-	  above fix.
-
-Wed May 23 06:51:46 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* lib/cgi.rb (CGI#[]): get rid of exceptions being raised.
-	  [ruby-dev:30740], Thanks Kentaro KAWAMOTO.
-
-Wed May 23 06:14:15 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/win32.c (move_to_next_entry): loc also must move forward.
-	  [ruby-talk:251987]
-
-Wed May 23 05:55:04 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (init_stdhandle): stderr should be without buffering,
-	  but mswin32 use buffering when stderr is not connected to tty.
-
-Wed May 23 03:33:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/monitor.rb (ConditionVariable#wait, mon_enter, mon_exit_for_cond):
-	  ensures Thread.critical to be false.  [ruby-talk:248300]
-
-Wed May 23 03:25:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* util.c (ruby_strtod): exponent is radix 10.  [ruby-talk:248272]
-
-Wed May 23 02:09:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_yield_0): should not clear state on TAG_NEXT when
-	  it's invoked from within lambda body.  [ruby-talk:248136]
-
-	* eval.c (proc_invoke): handle TAG_NEXT which would be caused by
-	  next in the lambda body as well.
-
-Wed May 23 01:55:49 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_fclose, rb_w32_close): need to save errno
-	  before calling original fclose()/close().
-
-Wed May 23 01:42:29 2007  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (disconnect): call shutdown for
-	  SSLSocket. Thanks, Technorama Ltd.
-
-Wed May 23 01:28:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* error.c (rb_notimplement), io.c (pipe_open): removed definite
-	  articles and UNIX manual section from messages.  [ruby-dev:30690]
-
-	* io.c (pipe_open): raise NotImplementedError for command "-" on
-	  platforms where fork(2) is not available.  [ruby-dev:30681]
-
-Wed May 23 00:06:19 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/socket/socket.c (s_recv, s_recvfrom): some systems (such as
-	  windows) doesn't set fromlen if the socket is connection-oriented.
-	  reported by Bram Whillock in [ruby-core:10512] [ruby-Bugs#9061]
-
-Tue Mar 20 15:37:24 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* distruby.rb: Add zip generation.
-
-Fri Mar 16 21:48:11 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* ext/dl/dl.c (rb_ary2cary): Fix a bug in type validation;
-	  submitted by sheepman <sheepman AT sheepman.sakura.ne.jp>
-	  in [ruby-dev:30554].
-
-Fri Mar 16 18:28:06 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* ext/etc/etc.c (etc_getgrgid): Fix a bug in Etc::getgrgid()
-	  always returning the (real) group entry of the running process;
-	  reported by: UEDA Hiroyuki <ueda AT netforest.ad.jp>
-	  in [ruby-dev:30586].
-
-Wed Mar 14 12:30:00 2007  Shigeo Kobayashi  <shigeo@tinyforest.jp>
-
-	* ext/bigdecimal/bigdecimal.c: BigDecimal("-.31") is now
-	  treated as ("-0.31") not as ("0.31").
-
-Fri Mar 16 18:05:40 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* lib/cgi.rb (CGI::header): IIS >= 5.0 does not need the nph
-	  assumption any more; submitted by MIYASAKA Masaru <alkaid AT
-	  coral.ocn.ne.jp> in [ruby-dev:30537].
-
-Tue Mar 13 17:29:43 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* stable version 1.8.5-p35 released.
-
-Tue Mar 13 14:42:10 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* test/fileutils/fileasserts.rb: Fix wrong error message.
-
-	* lib/fileutils.rb (FileUtils::mv): Type Error; should utilize
-	  Strings instead of Symbols here.
-
-Mon Mar 12 17:22:44 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/fileutils.rb (FileUtils::mv): fix incomplete backport of
-	  FileUtils.mv changes at r11988
-
-Mon Mar 12 16:09:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* mkconfig.rb (patchlevel): read from version.h.
-
-Sun Mar 11 18:57:50 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* misc/README: Add a note about ruby-electric.el. 
-
-	* misc/ruby-mode.el (ruby-non-block-do-re): Fix
-	  ruby-non-block-do-re. [ruby-core:03719]
-
-	* misc/inf-ruby.el: Synchronize the comment section with trunk.
-
-	* misc/README, misc/rdebug.el: Add rdebug.el, Emacs ruby-debug
-	  interface based on rubydb3x.el; submitted by Martin Nordholts
-	  <enselic AT gmail.com> in [ruby-bugs:PR#9023].
-
-Sun Mar 11 17:45:51 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* ext/dl/mkcallback.rb (mkfunc): Make sure that a callback
-	  function is found in the function table before trying to call
-	  it; submitted by sheepman <sheepman AT sheepman.sakura.ne.jp>
-	  in [ruby-dev:30524].
-
-Tue Mar  6 18:58:37 2007  Keiju Ishitsuka  <keiju@ruby-lang.org>
-	
-	* lib/shell/process-controller.rb: fix thread synchronization
-	  problem for [ruby-dev:30477].
-	
-Sun Mar  4 23:53:27 2007  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mv): could not move a directory between
-	  different filesystems. [ruby-dev:30411]
-
-Sat Mar  3 21:41:31 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* eval.c (stack_check): Unset inline to fix build with GCC 3.4.6;
-	  submitted by: NISHIMATSU Takeshi <t_nissie AT yahoo.co.jp> in
-	  [ruby-list:43218].
-	  cf. http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24556
-
-Sat Mar  3 16:30:39 2007  Akinori MUSHA  <knu@iDaemons.org>
-
-	* env.h (SCOPE_CLONE): Introduce a new scope flag to prevent a
-	  local_tbl region from getting freed many times; submitted by
-	  Chikanaga Tomoyuki <chikanag AT nippon-control-system.co.jp> in
-	  [ruby-dev:30460].
-
-	* eval.c (proc_invoke): Ditto.
-
-	* gc.c (obj_free): Ditto.
-
-	* parse.y (top_local_setup_gen): Ditto.
-
-Sat Mar  3 15:41:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (stmt, arg): should not omit lhs of OP_ASGN1 even if
-	  empty.  [ruby-dev:30452]
-
-Wed Feb 28 20:51:32 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* pack.c (pack_unpack): properly ignore non-base64 octets such as
-	  UTF-8 encoded BOMs; submitted by SOUMA Yutaka <holon@radastery.jp>
-	  to fix [ruby-core:10437]
-
-Wed Feb 28 00:08:11 2007  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
-
-	* mkconfig.rb (RbConfig): add CONFIG['PATCHLEVEL']
-
-	* common.mk: new target dist
-
-	* distruby.rb: new file
-
-Tue Feb 27 21:19:35 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (set_pioinfo_extra): new function for VC++8 SP1
-	  workaround. [ruby-core:10259]
-
-	* win32/win32.c (NtInitialize): call above function.
-
-Mon Feb 26 09:57:58 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* signal.c (ruby_signal): don't set SA_RESTART.  a backport from
-	  the HEAD.  [ruby-talk:220937]  [ruby-talk:147220]
-
-Tue Feb 13 02:21:12 2007  Sam Roberts  <sroberts@uniserve.com>
-
-	* io.c (rb_f_syscall): Fix buffer overflow with syscall
-	  arguments.  [ruby-bugs:PR#8541]
-
-Sat Feb 10 09:33:47 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_variant2val): sorry, fix the enbug.
-
-Sat Feb 10 09:08:01 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_variant2val): fix compile error
-	  on VC++.
-
-Sat Feb 10 08:38:30 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_variant2val): fix the bug when
-	  SAFEARRAY pointer is NULL.
-
-Tue Feb  6 20:41:39 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* string.c (rb_str_sub_bang): calling rb_str_modify() should be just
-	  before actually modifying the string.
-	  fixed: [ruby-dev:30211] (originally reported by zunda)
-
-Sat Jan 27 15:20:11 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (dyna_var_lookup): should not alter dvar->val not to
-	  destroy living value.  [ruby-core:10076]
-
-	* parse.y (dyna_init): ditto.
-
-Wed Jan 24 18:05:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-font-lock-syntactic-keywords): fix
-	  regexp font-lock bug.  [ruby-talk:235758]
-
-Sun Jan 14 07:26:44 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_free, ole_type_free, 
-	  olemethod_free, olevariable_free, oleparam_free, 
-	  ole_event_free): fix memory leak.  [ruby-core:09846]
-
-Tue Jan  9 12:29:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/etc/etc.c (etc_getpwuid, etc_getgrgid): fix to correctly
-	  convert uid/gid from VALUE.
-
-	* ext/etc/etc.c (etc_getpwuid): ditto.
-
-Mon Dec 25 20:08:28 2006  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* ext/openssl/ossl.h: fixed compilation problem on gcc 3.2.
-	  [ruby-talk:214786]
-
-Mon Dec 25 10:40:40 2006  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* stable version 1.8.5-p12 released.
-
-Sat Dec 16 04:02:10 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c: shouldn't run the killed thread at callback.
-	  [ruby-talk: 227408]
-
-Fri Dec 15 17:21:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rdoc/ri/ri_options.rb: prevent NameError.  [ruby-dev:29597]
-
-Thu Dec 14 23:37:38 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (glob_helper): get rid of possible memory leak.
-
-	* win32/win32.c (cmdglob, rb_w32_cmdvector, rb_w32_opendir,
-	  rb_w32_get_environ): not to use GC before initialization.
-
-Wed Dec  6 19:53:41 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (SITE_DIR): fixed to emtpy RUBY_SITE_LIB in config.h on
-	  NetBSD.  fixed: [ruby-dev:29358]
-
-Mon Dec  4 10:43:46 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (dyna_init_gen): dvar initialization only if dvar is
-	  assigned inner block.  [ruby-talk:227402]
-
-Mon Dec  4 10:22:26 2006  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
-
-	* stable version 1.8.5-p2 released.
-
-Sun Dec  3 17:11:12 2006  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::QueryExtension::read_multipart): should quote
-	  boundary. JVN#84798830
-
-Sun Nov 26 16:36:46 2006  URABE Shyouhei  <shyouhei@ruby-lang.org>
-
-	* version.h: addition of RUBY_PATCHLEVEL.
-	* version.c: ditto.
-
-Fri Nov 24 10:17:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (bignorm): avoid segmentation.  a patch from Hiroyuki
-	  Ito <ZXB01226@nifty.com>.  [ruby-list:43012]
-
-Thu Nov  2 15:43:39 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* parse.y (primary): should set NODE even when compstmt is NULL.
-	  merge from trunk. fixed: [ruby-dev:29732]
-
-Sat Sep 23 21:34:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::QueryExtension::read_multipart): CGI content
-	  may be empty.  a patch from Jamis Buck <jamis at 37signals.com>.
-
-Mon Sep  4 21:43:57 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/dbm/extconf.rb: create makefile according to the result of check
-	  for dbm header.  fixed: [ruby-dev:29445]
-
-Tue Aug 29 19:10:10 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* hash.c (rb_hash_s_create): fixed memory leak, based on the patch
-	  by Kent Sibilev <ksruby at gmail.com>.  fixed: [ruby-talk:211233]
-
-Fri Aug 25 17:15:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* stable version 1.8.5 released.
-
-Fri Aug 25 17:02:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (gc_sweep): typo fixed.
-
-Tue Aug 22 18:47:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::handle_method):
-	  rdoc documents C module methods as instance methods. a patch in
-	  [ruby-core:08536]. 
-
-Sat Aug 19 14:15:02 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/Makefile.sub (config.status): include winsock2.h instead of
-	  winsock.h when --with-winsock2 is specified.
-	  fixed: [ruby-dev:29296]
-
-Sat Aug 19 11:28:08 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_s_rename): use errno if set properly.
-	  fixed: [ruby-dev:29293]
-
-Sat Aug 19 11:09:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (then): remove semicolon warning.  [ruby-dev:29299]
-
-Thu Aug 17 19:15:16 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_stat_[rRwWxX]): check for super user.
-	  fixed: [ruby-core:08616]
-
-Thu Aug 17 14:47:06 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb: added rdoc by Daniel Berger.  [ruby-core:08177]
-
-Thu Aug 17 00:39:05 2006  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/ring.rb (do_reply): Fix for RingServer fails to find a
-	  TupleSpace when TupleSpace resides in the same ruby process with
-	  RingServer. a patch from Kent Sibilev. [ruby-core:08453]
-
-Wed Aug 16 11:45:36 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (proc_setuid, proc_setgid, proc_seteuid, proc_setegid):
-	  get rid of bogus implementations on Mac OS X.
-
-Tue Aug 15 19:10:18 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#find_class_comment): Fix
-	  broken class-level documentation.
-
-Wed Aug 16 11:09:26 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ruby.c (set_arg0): fill argv other than the first with an empty
-	  string instead of NULL.
-
-Wed Aug 16 11:08:00 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/win32.h: removed an excess macro.  fixed: [ruby-dev:29258]
-
-Tue Aug  8 23:49:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/irb/extend-command.rb (IRB::ExtendCommandBundle): pacify
-	  RDoc.  a patch from Eric Hodel <drbrain at segment7.net>.
-	  [ruby-core:08522]
-
-Tue Aug  8 11:32:54 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* Makefile.in, common.mk, configure.in: fix for platforms without
-	  rm. patches from Yutaka kanemoto <kinpoco at gmail.com>.
-	  [ruby-dev:29215]
-
-Mon Aug  7 17:56:59 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c, ext/digest/rmd160/rmd160ossl.c,
-	  ext/digest/sha1/sha1ossl.c, ext/readline/readline.c: move
-	  incluion of config.h to pacify AIX.  a patch from Yutaka
-	  Kanemoto <kinpoco at gmail.com>.  [ruby-dev:29197]
-
-Mon Aug  7 15:55:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/syck/syck.c (syck_move_tokens): should avoid negative
-	  memmove.  [ruby-list:42625]
-
-Mon Aug  7 14:37:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* configure.in, common.mk: AIX link issue.  a patch from Yutaka
-	  Kanemoto <kinpoco at gmail.com>.  [ruby-dev:29190]
-
-	* ext/socket/socket.c: AIX socket support.  [ruby-dev:29190]
-
-Mon Aug  7 12:05:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dln.c, eval.c, gc.c, regex.c, ruby.h: shut up AIX alloca
-	  warning.  a patch from Yutaka Kanemoto <kinpoco at gmail.com>.
-	  [ruby-dev:29191]
-
-Sun Aug  6 20:40:41 2006  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date/format.rb (str[fp]time): %[EO]U didn't denote %U.
-
-Sat Aug  5 17:07:43 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (top_local_setup): local_vars[-1] should point
-	  ruby_scope itself to protect local_tbl from garbage collection.
-	  [ruby-dev:29049]
-
-Sat Aug  5 13:54:03 2006  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date/format.rb (str[fp]time): "%\n" means "\n".
-
-Fri Aug  4 15:21:00 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib: Merge RDoc and .document from HEAD.
-	* lib/drb/ssl.rb: Close socket on SSLError [ruby-core:7197]
-
-Fri Aug  4 19:13:41 2006  Keiju Ishitsuka  <keiju@ruby-lang.org>
-
-	* lib/irb/{init.rb,ruby-lex.rb,slex.rb}: can't input '\c' for
-	  [ruby-core: 7122].
-
-Fri Aug  4 14:02:14 2006  James Edward Gray II  <james@grayproductions.net>
-
-	* lib/date/format.rb (__strptime, strftime): allow multi-line patterns
-	  in Date#strftime the same as Time#strftime accepts.
-	  fixed: [ruby-core:08466]
-
-Fri Aug  4 13:56:51 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* pack.c (pack_pack): check argument overrun for 'P'.  based on a
-	  patch by rucila <rucila at yahoo.cojp>.  fixed: [ruby-dev:29182]
-
-Tue Aug  1 17:44:03 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/win32.c (init_stdhandle): assign standard file handles.
-
-Tue Aug  1 12:24:58 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (Init_Binding): fix old commit miss.
-
-Mon Jul 31 17:08:20 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (exit_handler): new function; release winsock and
-	  environment work area.
-
-	* win32/win32.c (NTInitialize): setup exit_handler.
-
-	* win32/win32.c (StartSockets): use exit_handler.
-
-	* win32/win32.c (rb_w32_getenv): use GetEnvironmentStrings() instead
-	  of GetEnvironmentVariable(), because the latter cannot distinguish
-	  wheather a null environment variable exists or not.
-	  fixed: [ruby-talk:205123]
-
-Mon Jul 31 16:15:13 2006  Tanaka Akira  <akr@fsij.org>
-
-	* test/ruby/test_process.rb (TestProcess#test_rlimit_nofile):
-	  setrlimit may fail with EINVAL.
-	  reported by MIYAMUKO Katsuyuki.  [ruby-dev:29174]
-
-Mon Jul 31 13:38:22 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httprequest.rb (WEBrick::HTTPReuqest#parse_uri): improve
-	  for the value of IPv6 address in the Host: header field.
-
-Mon Jul 31 09:22:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ruby.h: use ifdef (or defined) for macro constants that may or
-	  may not be defined to shut up gcc's -Wundef warnings.
-	  [ruby-core:08447]
-
-Sun Jul 30 23:26:22 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_call0): trace call/return of method defined from block.
-	  fixed: [ruby-core:08329]
-
-	* eval.c (rb_trap_eval): make the current thread runnable to deal with
-	  exceptions which occurred within the trap.  fixed: [ruby-dev:27729]
-
-	* lib/cgi/session.rb, lib/cgi/session/pstore.rb: suppress warnings.
-	  fixed: [ruby-talk:204896]
-
-Sat Jul 29 06:12:06 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: freeze ip_name for security reason.
-
-Sat Jul 29 01:23:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/logger.rb: improves the amount of documentation that Rdoc
-	  picks up when processing logger.rb by moving the require
-	  statement back before the comment block.  a patch from Hugh
-	  Sasse <hgs at dmu.ac.uk>.  [ruby-core:08422]
-
-Thu Jul 27 22:21:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* time.c (time_to_s): fixed format mismatch.
-
-Thu Jul 27 21:19:54 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* math.c (domain_check): a new function to check domain error
-	  explicitly for systems that return NaN like FreeBSD.
-	  [ruby-core:07019]
-
-	* math.c (math_acos, math_asin, math_acosh, math_atanh, math_log,
-	  math_log10, math_sqrt): use domain_check().
-
-	* math.c (math_sqrt): fix documentation flaw.
-
-Thu Jul 27 18:12:12 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* time.c: need to declare time_utc_offset.
-
-Thu Jul 27 17:01:01 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_close): always calls "close" method of the receiver.
-	  [ruby-core:6911] [ruby-core:8112]
-
-Thu Jul 27 16:49:01 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* time.c (time_to_s): use +0900 style timezone string for local time.
-	  [ruby-dev:29143]
-
-Thu Jul 27 16:41:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/openssl/ossl.h: move <ruby.h> inclusion point to shut up
-	  Solaris compiler.  [ruby-core:08114]
-
-Wed Jul 26 22:20:59 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* configure.in: add support for as and ASFLAGS.  [ruby-dev:29138]
-
-Wed Jul 26 22:13:45 2006  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: sync with HEAD (rev 1.132).
-
-	* lib/net/http.rb (Net::HTTP#post, request_post, request): should
-	  set Content-Type: x-www-form-urlencoded by default.
-
-	* lib/net/http.rb (Net::HTTPHeader#content_type): should return
-	  nil when there's no Content-Type.
-
-	* lib/net/http.rb (Net::HTTPHeader#sub_type): should return nil
-	  when there's no sub Content-Type (e.g. "Content-Type: text").
-
-	* lib/net/http.rb (Net::HTTPHeader#type_params): wrongly failed
-	  when there's no Content-Type.
-
-Wed Jul 26 18:35:38 2006  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/strscan/strscan.c: sync with HEAD (rev 1.25).
-
-	* ext/strscan/strscan.c (strscan_do_scan):
-	  StringScanner.new("").scan(//) should return "". [ruby-Bugs:4361]
-
-Wed Jul 26 18:14:19 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/pty/pty.c (getDevice): retry once after GC on failure.
-	  [ruby-core:08282]
-
-Wed Jul 26 17:28:16 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): prepend ".." to %u for negative bignum,
-	  but not "-".  fixed: [ruby-core:08167]
-
-Wed Jul 26 16:39:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_scan): add string modification check.
-	  [ruby-core:7216]
-
-Wed Jul 26 16:06:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::QueryExtension::read_multipart): check
-	  multipart boundary end.  a patch from Fujioka <fuj at rabbix.jp>
-	  [ruby-dev:28470]
-
-Wed Jul 26 01:02:59 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in: suppress warnings by automake 1.8 or later.
-
-Tue Jul 25 00:30:06 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/prettyprint.rb: RD to RDoc conversion by Hugh Sasse.
-
-Tue Jul 25 14:49:51 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/mkmf.rb (configuration): typo.
-
-Tue Jul 25 13:14:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (rb_proc_times): rename hz to hertz to avoid name
-	  crash on AIX.  [ruby-dev:29126]
-
-Mon Jul 24 22:03:40 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (backtrace): skip frames successive on node and method name.
-
-Mon Jul 24 17:55:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (rb_f_system): add security check.  [ruby-talk:202947]
-
-	* process.c (rb_f_system): move signal right before fork to avoid
-	  signal handler intervention.
-
-Mon Jul 24 15:51:52 2006  Tanaka Akira  <akr@fsij.org>
-
-	* ext/readline/readline.c (readline_readline): rl_deprep_term_function
-	  may be NULL with libedit.  reported by Ryan Davis.  [ruby-dev:29070]
-
-Mon Jul 24 15:19:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): revert last change.  [ruby-dev:29112]
-	  [ruby-core:08374]
-
-Sun Jul 23 22:59:49 2006  Tanaka Akira  <akr@fsij.org>
-
-	* test/socket/test_unix.rb: disabled on cygwin.
-	  reported by Kouhei Yanagita.  [ruby-dev:29080]
-
-Fri Jul 21 21:21:08 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_call0): include funcalled methods in caller list.
-	  fixed: [ruby-core:08290]
-
-Fri Jul 21 12:11:00 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb, lib/mkmf.rb (with_destdir): remove drive letter before
-	  prepending destdir on DOSISH.
-
-Thu Jul 20 15:07:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ruby.h: export classes/modules to implement sandbox.
-	  [ruby-core:08283]
-
-Thu Jul 20 00:06:29 2006  Keiju Ishitsuka  <keiju@ishitsuka.com>
-
-	* lib/irb/completion.rb: support for completion of numeric
-    	  number. [ruby-dev: 29038]
-
-Wed Jul 19 23:53:05 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/parser.rb, lib/rss/utils.rb: added documents.
-
-Tue Jul 18 22:10:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (rb_f_system): block SIGCHLD during the process
-	  execution, like glibc system(3) does.  [ruby-talk:202361]
-
-Tue Jul 18 23:12:14 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (open_ifs_socket): should not use plain malloc.
-
-	* win32/win32.c (rb_w32_opendir): should not use plain realloc.
-
-Tue Jul 18 18:05:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* test/ruby/test_float.rb (TestFloat::test_strtod): update test to
-	  conform strtod change.
-
-Tue Jul 18 15:49:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (pack_unpack): propagate association array to copied
-	  string.  [ruby-core:08223]
-
-	* pack.c (pack_unpack): return referenced string itself if it has
-	  same length as specified.  a patch from <nobu at ruby-lang.org>
-	  in [ruby-core:08225].
-
-	* pack.c (pack_pack): taint 'p' packed strings.
-
-Tue Jul 18 14:03:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/webrick/httpserver.rb (WEBrick::HTTPServer::unmount): remove
-	  inpect argument from sprintf.  [ruby-dev:29039]
-
-Tue Jul 18 10:53:37 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* object.c (rb_cstr_to_dbl): limit out-of-range message.
-
-	* util.c (ruby_strtod): return end pointer even if ERANGE occurred.
-	  fixed: [ruby-dev:29041]
-
-Mon Jul 18 00:43:05 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* util.c (ruby_strtod): stop at dot not followed by digits.
-	  fixed: [ruby-dev:29035]
-
-Tue Jul 18 00:01:27 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: remove LIBRUBY_SO if static linked extensions exist.
-
-Mon Jul 17 23:30:46 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (rb_cv_msvcrt): defaulted to msvcrt.  Workaround for a
-	  bug of cygwin 1.5.20.
-
-Mon Jul 17 13:43:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (define_swapx): should not use plain malloc.
-
-Mon Jul 17 12:58:41 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: should use ac_cv_lib_dl_dlopen=no on MinGW.
-
-Sat Jul 15 23:50:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_require_safe): wait for another thread requiring the same
-	  feature.  fixed: [ruby-core:08229]
-
-Sat Jul 15 01:27:13 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (has_magic): glob names contain alphabets to enable case fold
-	  search also for directories.  fixed: [ruby-talk:201917]
-
-Sat Jul 15 01:09:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* st.c (malloc): use xmalloc/xcalloc instead of plain
-	  malloc/calloc, to detect memory allocation failure.  see
-	  <http://www.nongnu.org/failmalloc/>.
-
-	* gc.c (rb_memerror): should not raise empty nomem_error.
-
-Fri Jul 14 13:08:13 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: add methods for new features of latest Tcl/Tk8.5.
-
-	* ext/tk/lib/tk/namespace.rb: ditto.
-
-Fri Jul 14 02:30:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/monitor.rb: document patch from Hugh Sasse <hgs at dmu.ac.uk>.
-	  [ruby-core:08205]
-
-Fri Jul 14 01:09:46 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (then): error in warning action.
-
-Fri Jul 14 00:10:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_pop): may cause realloc oscillation.  a patch
-	  from MORITA Naoyuki <mlgetter at kidou.sakura.ne.jp>.
-	  [ruby-dev:29028]
-
-Thu Jul 13 22:23:56 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/composite.rb: improve handling of the classname on the
-	  option database for the widget class which includes TkComposite.
-
-Thu Jul 13 20:32:19 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/parser.rb: updated documents by a patch from
-	Hugh Sasse <hgs at dmu.ac.uk>. [ruby-core:8194]
-
-Wed Jul 12 13:54:09 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (then): we'd like to reserve colon here for the future.
-	  warning added.
-
-Tue Jul 11 20:58:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ruby.h: export rb_cMethod.  [ruby-talk:201259]
-
-Tue Jul 11 19:13:33 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: remove restriction on the class of
-	  pseudo-toplevel.
-
-Tue Jul 11 18:00:57 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: security fix.
-
-Tue Jul 11 17:33:39 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* string.c (rb_str_dump): need to extend len for \b.
-
-Mon Jul 10 22:00:00 2006  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c: Allows '_' to appear within
-	  digits.  [ruby-dev:28872]
-
-	* ext/bigdecimal/lib/bigdecimal/util.rb: Bug in to_r reported by
-	  [ruby-list:42533] fixed.
-
-Mon Jul 10 19:22:19 2006  Tanaka Akira  <akr@fsij.org>
-
-	* gc.c (gc_sweep): expand heap earlier.
-	  reported by MORITA Naoyuki.  [ruby-dev:28960]
-
-Mon Jul 10 18:59:34 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/font.rb: sorry. mistaken to patch.
-
-Mon Jul 10 18:46:52 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c: make SEGV risk lower at exit.
-
-	* ext/tk/lib/tk.rb: ditto.
-
-	* ext/tk/lib/multi-tk.rb: fail to call function-style methods on slave
-	  interpreters. The strategy (MultiTkIp_PseudoToplevel_Evaluable) to
-	  fix the problem is a little tricky. You may have to take care of
-	  conflicting with it.
-
-	* ext/tk/lib/tk.rb: a little change for the pseudo-toplevel strategy.
-
-	* ext/tk/lib/tk/font.rb: ditto.
-
-	* ext/tk/lib/tk/msgcat.rb: ditto.
-
-	* ext/tk/lib/tkextlib/itk/incr_tk.rb: ditto.
-
-	* ext/tk/sample/demos-en/widget: fail to call function-style methods
-	  on sample scripts. To fix it, a strategy which similar to the way
-	  on MultiTiIp is used. Please take care when re-write and re-run a
-	  demo script on the Widget-Demo code viewer.
-
-	* ext/tk/sample/demos-jp/widget: ditto.
-
-Mon Jul 10 13:58:40 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* signal.c (ruby_nativethread_signal, posix_nativethread_signal,
-	  sigsend_to_ruby_thread, install_nativethread_sighandler):
-	  nativethread-support on signal handler. RE-backport from 1.9.
-
-	* ruby.h (HAVE_NATIVETHREAD_KILL): ditto.
-
-	* eval.c (ruby_native_thread_kill): ditto.
-
-Mon Jul 10 10:54:14 2006  Ryan Davis  <ryand@zenspider.com>
-
-	* lib/rdoc/parsers/parse_f95.rb: massive overhaul from Yasuhiro
-	  Morikawa including new file suffixes, function support, public
-	  variables and constants, derived-types, defined operators and
-	  assignments, namelists, and subroutine and function
-	  arguments. Truly massive.
-
-	* lib/rdoc/diagram.rb: diagrams are now cached.
-
-	* lib/irb/completion.rb: fixed a crasher when completing against
-	  an unnamed class/module.
-
-	* lib/rdoc/parsers/parse_c.rb: private comment (--/++) support in
-	  C-file rdoc.
-
-	* lib/debug.rb: minor clarification in help.
-
-	* lib/pp.rb: minor clarification on exception.
-
-Mon Jul 10 09:29:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_clear_cache_for_undef): clear entries for included
-	  module.  fixed: [ruby-core:08180]
-
-Mon Jul 10 01:48:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* st.h (st_data_t): use pointer sized integer for st_data_t.
-	  [ruby-dev:28988]
-
-Sun Jul  9 18:06:47 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (try_constant): fix for value 1 at cross compiling.
-
-	* lib/mkmf.rb (create_makefile): prevent substitution of macro
-	  definition.  fixed: http://www.yotabanana.com/lab/20060624.html#p02
-
-Sun Jul  9 00:54:34 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (next_jump): deal with destination of next.
-	  fixed: [ruby-core:08169]
-
-Fri Jul  7 00:38:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* hash.c (rb_hash_default): should not call default procedure if
-	  no key is given.  [ruby-list:42541]
-
-Fri Jul  7 00:29:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* time.c (time_mload): a patch from Daniel Berger
-	  <Daniel.Berger at qwest.com>.  [ruby-core:08128]
-
-Thu Jul  6 22:21:57 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (rb_proc_times): use sysconf(_SC_CLK_TCK) value prior to
-	  HZ and CLK_TCK.  fixed: [ruby-talk:200293]
-
-Thu Jul  6 22:17:21 2006  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/racc/cparse/cparse.c: sync with original code, rev 1.8.
-
-	* ext/racc/cparse/cparse.c: should mark CparseParams objects.
-
-	* lib/racc/parser.rb: sync with original code, rev 1.8.
-
-	* lib/racc/parser.rb: update coding style.
-
-Mon Jul  3 19:04:38 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c (ip_make_menu_embeddable): help to make a menu
-	  widget embeddable (pack, grid, and so on) like as a general widget.
-	  However, an embeddable menu may require to be definied some event
-	  bindings for general use.
-
-	* ext/tk/lib/tk/event.rb: [bug fix] Tk.callback_break and
-	  Tk.callback_continue don't work on MultiTkIp.
-
-	* ext/tk/lib/multi-tk.rb: ditto.
-
-	* ext/tk/lib/tk.rb: lack of Tk.callback_return.
-
-	* ext/tk/lib/tk/menu.rb: improve creating clone menus.
-
-Mon Jul  3 14:42:06 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/etc/extconf.rb (PW_UID2VAL, PW_GID2VAL): defaulted to conversion
-	  from int, and sys/types.h needs to be included before grp.h.
-	  fixed: [ruby-dev:28938]
-
-Mon Jul  3 01:14:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_inspect): encode \b (\010) for escape.
-	  [ruby-dev:28927]
-
-	* string.c (rb_str_dump): ditto.
-
-Sun Jul  2 19:17:56 2006  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/racc/cparse/cparse.c: sync with original code (rev 1.7).
-
-	* ext/racc/cparse/cparse.c: use rb_catch instead of rb_iterate.
-	  Giving a block to a Ruby-level method by rb_iterate is obsolete on
-	  Ruby 1.9.  Note that current cparse.c still includes one
-	  rb_iterate call on Ruby 1.8, but it is not a problem (at least
-	  just now).
-
-Sat Jul  1 15:15:49 2006  Tanaka Akira  <akr@m17n.org>
-
-	* test/socket/test_nonblock.rb: add timeout to send/receive
-	  an empty UDP packet.
-	  [ruby-dev:28820]
-
-Fri Jun 30 23:46:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* configure.in: should test isinf for Solaris with GCC compiler.
-	  a patch from <ville.mattila at stonesoft.com>.  [ruby-core:07791]
-
-	* configure.in: -shared patch from Andrew Morrow
-	  <andrew.c.morrow at gmail.com>.  [ruby-core:08100]
-
-Thu Jun 29 18:58:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c (BigDecimal_version): fix patch
-	  failure.
-
-Thu Jun 29 18:00:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c: add RDoc document.  a patch from
-	  mathew <meta at pobox.com>.   [ruby-core:07050]
-
-Wed Jun 28 15:47:14 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/optparse.rb: RDoc patch from Robin Stocker <robin@nibor.org>
-	  [ruby-core:08087]
-
-Wed Jun 28 19:04:34 2006  Tanaka Akira  <akr@m17n.org>
-
-	* test/socket/test_unix.rb: test_seqpacket_pair removed.
-	  [ruby-dev:28846]
-
-Tue Jun 27 23:03:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c: RDoc update for =~ method.  a patch from Alex Young
-	  <alex at blackkettle.org>.  [ruby-core:08068]
-
-Tue Jun 27 22:47:18 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c: forgot to update TCLTKLIB_RELEASE_DATE.
-
-	* ext/tk/lib/tk.rb (tk_tcl2ruby): [bug fix] sometimes fail to convert
-	  a tcl string to a ruby object if the tcl string includes "\n".
-
-Tue Jun 27 16:04:05 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* win32/win32.h: define isascii on MinGW for msvcrt compatibility.
-
-	* configure.in: set ac_cv_header_sys_time_h=no on MinGW
-	  for msvcrt compatibility.
-
-Tue Jun 27 11:36:02 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/etc/etc.c (setup_passwd, setup_group): allow bignum uid, gid and
-	  so on.  [ruby-talk:199102]
-
-Mon Jun 26 13:37:27 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/rdoc: Merge from HEAD.
-	  Add options to limit the ri search path.
-
-Tue Jun 27 00:54:08 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* util.c (powersOf10): constified.
-
-Mon Jun 26 18:37:44 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c (ip_delete): fix SEGV when a slave-ip is
-	  deleted on callback.
-
-Mon Jun 26 10:47:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (pipe_open): avoid closing uninitialized file descriptors.
-	  a patch from <tommy at tmtm.org> [ruby-dev:28600]
-
-Mon Jun 26 09:56:22 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.[ch] (rb_w32_send, rb_w32_sendto): constified.
-
-Sun Jun 25 23:02:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* Makefile.in, mkconfig.rb: catch-up for latest autoconf.
-
-Sat Jun 24 06:35:00 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* signal.c: revert last change.
-
-	* ruby.h: ditto.
-
-	* eval.c: ditto.
-
-Thu Jun 22 11:52:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/net/http.rb (Net::HTTPResponse): duplicated error 501;
-	  HTTPInternalServerError should be error 500.  [ruby-core:08037]
-
-Thu Jun 22 05:15:58 2006  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/socket.c (sock_s_socketpair): try GC only once.
-	  [ruby-dev:28778]
-
-Wed Jun 21 21:28:32 2006  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date.rb (jd_to_commercial): now works fine even if in
-	  mathn-ized context.
-
-Wed Jun 21 17:32:31 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* signal.c (ruby_nativethread_signal, posix_nativethread_signal,
-	  sigsend_to_ruby_thread, install_nativethread_sighandler):
-	  nativethread-support on signal handler (backport from 1.9).
-
-	* ruby.h (HAVE_NATIVETHREAD_KILL): ditto.
-
-	* eval.c (ruby_native_thread_kill): ditto.
-
-Wed Jun 21 08:39:54 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/xmlrpc/create.rb (XMLRPC::Create::conv2value): merge Date
-	  and Time processing.  [ruby-core:08033]
-
-Wed Jun 21 01:40:25 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (yylex, reswords): modifier token is no longer returned in
-	  fname state.  [ruby-dev:28775]
-
-Wed Jun 21 01:12:46 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: RSS::Element.def_corresponded_attr_writer
-	  supported date type.
-
-Tue Jun 20 22:08:36 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* test/rss/test_parser.rb: split parser tests into ...
-	* test/rss/test_parser_1.0.rb: ... RSS 1.0 parsing tests and ...
-	* test/rss/test_parser_2.0.rb: ... RSS 2.0 parsing tests.
-
-Tue Jun 20 21:19:06 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: provided default RSS::Element#children.
-
-	* lib/rss/0.9.rb: used default RSS::Element#children.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-
-Tue Jun 20 21:04:33 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: provided default RSS::Element#_tags.
-
-	* lib/rss/0.9.rb: used default RSS::Element#_tags.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-
-Tue Jun 20 20:47:07 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: hide RSS::Element.install_model.
-	  (RSS::Element.install_have_child_element,
-	   RSS::Element.install_have_children_element,
-	   RSS::Element.install_text_element,
-	   RSS::Element.install_date_element): call
-	  RSS::Element.install_model internally.
-
-	* lib/rss/0.9.rb: followed new API.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/content.rb: ditto.
-	* lib/rss/dublincore.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/syndication.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-	* lib/rss/trackback.rb: ditto.
-
-Tue Jun 20 20:18:05 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: add check for OBJ_NAME_do_all_sorted.
-
-	* ext/openssl/ossl_cipher.c (ossl_s_ciphers): new method
-	  OpenSSL::Cipher.ciphers. it returns all the cipher names.
-
-	* ext/openssl/lib/openssl/cipher.rb:
-	  - add constants AES128, AES192, AES256. [ruby-dev:28610]
-	  - reimplement without eval()
-
-	* ext/openssl/lib/openssl/digest.rb: reimplement without eval().
-
-	* test/openssl/test_cipher.rb, test_digest: fix about reimplemented
-	  features.
-
-	* sample/openssl/cipher.rb: rewrite all.
-
-Sat Jun 19 11:21:46 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/test/unit/assertions.rb: Merge RDoc from HEAD.
-
-Tue Jun 20 01:06:57 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb:
-	  - cleanup validation mechanism. Now, #XXX_validation is
-	    needless.
-	  - changed internal variable name RSS::Element::MODEL to
-	    RSS::Element::MODELS.
-	  - RSS::Element.install_model requires uri.
-
-	* lib/rss/0.9.rb: followed new validation API.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/content.rb: ditto.
-	* lib/rss/dublincore.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/syndication.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-	* lib/rss/trackback.rb: ditto.
-
-Mon Jun 19 23:40:59 2006  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/lib/kconv.rb: remove default -m0 and fix document.
-
-	* ext/nkf/nkf-8/{nkf.c, config.h, utf8tbl.c, utf8tbl.h}:
-	  imported nkf 2.0.7.
-
-Mon Jun 19 22:31:59 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb:
-	  - provided default #to_s as RSS::Element#to_s.
-	  - removed RSS::Element#other_element.
-	  - RSS::Element#tag requires attributes as Hash instead of Array.
-
-	* lib/rss/0.9.rb: removed #to_s to use RSS::Element#to_s.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-	* lib/rss/trackback.rb: ditto.
-
-	* lib/rss/2.0.rb: removed #other_element.
-
-Mon Jun 19 22:09:16 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c(ole_invoke): support some kind of
-	  method of word. [ruby-Bugs#3237]
-
-	* ext/win32ole/tests/test_word.rb: ditto.
-
-	* ext/win32ole/tests/testall.rb: ditto.
-
-Mon Jun 19 00:02:17 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: automatically detected attributes.
-
-	* lib/rss/0.9.rb: removed #_attrs.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-	* lib/rss/trackback.rb: ditto.
-
-	* lib/rss/parser.rb: followed new internal API.
-
-Mon Jun 19 00:00:17 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: fix bug: initialize improper tables.
-
-Sun Jun 18 22:36:13 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: RSS::Element#initialize accepts initial
-	  attributes.
-	* lib/rss/0.9.rb: ditto.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/dublincore.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-	* lib/rss/trackback.rb: ditto.
-
-	* lib/rss/utils.rb: added Utils.element_initialize_arguments? to
-	  detect backward compatibility initial arguments.
-
-	* lib/rss/parser.rb: user initial attributes to initialize
-	  RSS::Element.
-
-Sun Jun 18 18:24:42 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/converter.rb: use NKF for Uconv fallback.
-
-Sun Jun 18 18:22:04 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* test/rss/test_image.rb: shared name space configuration.
-
-Sun Jun 18 18:13:25 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: improved ignore_unknown_element
-	  handling. RSS::NotExpectedTagError provides tag URI.
-	* lib/rss/parser.rb: ditto.
-	* lib/rss/0.9.rb: ditto.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/content.rb: ditto.
-	* lib/rss/dublincore.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/syndication.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-	* lib/rss/trackback.rb: ditto.
-
-	* test/rss/rss-assertions.rb: checked URI of not expected tag too.
-	* test/rss/test_parser.rb: ditto.
-
-Sun Jun 18 18:08:36 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: changed empty namespace URI representation to ""
-	  from nil.
-	* lib/rss/parser.rb: ditto.
-	* lib/rss/0.9.rb: ditto.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-
-Sun Jun 18 18:03:50 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/parser.rb: removed a guard for requiring open-uri.
-
-Sun Jun 18 18:01:26 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: fixed typo: except -> expect
-	* lib/rss/parser.rb: ditto.
-	* test/rss/rss-assertions.rb: ditto.
-	* test/rss/test_parser.rb: ditto.
-
-Sun Jun 18 17:52:39 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: RSS::Element#calc_indent became to be deprecated.
-	* lib/rss/0.9.rb: ditto.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/taxonomy.rb: ditto.
-	* lib/rss/trackback.rb: ditto.
-
-	* test/rss/test_1.0.rb: removed RSS::Element.indent_size tests.
-	* test/rss/test_2.0.rb: ditto.
-
-Sun Jun 18 00:49:11 2006  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/socket.c (bsock_recv_nonblock): new method
-	  BasicSocket#recv_nonblock.
-	  (udp_recvfrom_nonblock): renamed from ip_recvfrom_nonblock.
-	  IPSocket#recvfrom_nonblock is moved to UDPSocket#recvfrom_nonblock.
-	  (unix_recvfrom_nonblock): removed.
-	  UNIXSocket#recvfrom_nonblock is removed.
-
-Sat Jun 17 22:17:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/mathn.rb (Integer::prime_division): raise ZeroDivisionError
-	  on zeros.  [ruby-dev:28739]
-
-Sat Jun 17 14:53:32 2006  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb: backport from 1.9.
-	  (Kernel#Pathname): new method.
-
-Sat Jun 17 10:30:41 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb (Hash#merge, Enumerable#sort_by): removed.
-
-	* lib/rss/rss.rb (RSS::RootElementMixin#to_xml): added.
-	  [ruby-talk:197284]
-
-	  We can convert RSS version easily like the following:
-	    rss10 = RSS::Parser.parse(File.read("1.0.rdf"))
-	    File.open("2.0.rss", "w") {|f| f.print(rss10.to_xml("2.0"))}
-
-	* test/rss/test_1.0.rb: added #to_xml test.
-	* test/rss/test_2.0.rb: ditto.
-
-	* test/rss/rss-testcase.rb: added some helper methods that
-	  generates sample RSS 2.0.
-
-	* sample/rss/convert.rb: added a sample script to convert RSS format.
-
-Sat Jun 17 10:23:22 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb (Kernel#funcall): removed.
-	* lib/rss/parser.rb (Kernel.URI): removed.
-
-	* lib/rss/maker/: supported
-	    xxx.new_yyy do |yyy|
-	      yyy.zzz = zzz
-	      ...
-	    end
-	  style and this style became the style of the recommendation.
-
-	  Old style
-	    yyy = xxx.new_yyy
-	    yyy.zzz = zzz
-	    ...
-	  is supported too but this style isn't recommended.
-	  [ruby-talk:197284]
-
-	* test/rss/test_*maker*.rb: used new recommended style.
-
-Sat Jun 17 09:03:47 2006  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss, test/rss: backported from trunk. (2005-11-16 - now)
-
-	* lib/rss/rss.rb (RSS::VERSION): 0.1.5 -> 0.1.6.
-	* test/rss/test_version.rb (RSS::TestVersion#test_version): ditto.
-
-	* lib/rss/trackback.rb: added TrackBack prefix.
-	* lib/rss/maker/trackback.rb: ditto.
-
-	* lib/rss/rss.rb : removed needless argument 'prefix'.
-	* lib/rss/parser.rb: ditto.
-
-	* lib/rss/1.0.rb: added rdf:Bag.
-
-	* lib/rss/taxonomy.rb: implemented taxonomy module.
-	* test/rss/test_taxonomy.rb: added tests for taxonomy support.
-
-	* lib/rss/1.0.rb: added convenience method 'resources'.
-	* lib/rss/taxonomy.rb: ditto.
-	* test/rss/rss-assertions.rb: added test for 'resources'.
-	* test/rss/test_taxonomy.rb: ditto.
-
-	* lib/rss/rss.rb: fixed a indentation bug.
-	* lib/rss/taxonomy.rb: fixed <taxo:topic> #to_s bug.
-	* test/rss/test_taxonomy.rb: added a #to_s test.
-
-	* lib/rss/maker/taxonomy.rb: implemented taxonomy module for RSS
-	  Maker.
-	* lib/rss/taxonomy.rb: supported RSS Maker.
-	* lib/rss/maker.rb: added taxonomy module support.
-
-	* lib/rss/rss.rb: adjusted to other element API.
-	* lib/rss/1.0.rb: adjusted to other element API but backward
-	  compatibility is reserved.
-	* lib/rss/0.9.rb: ditto.
-
-	* test/rss/test_maker_taxo.rb: added test case for taxonomy module
-	  for RSS Maker.
-	* test/rss/test_setup_maker_1.0.rb: added tests for taxo:topic.
-
-	* test/rss/test_setup_maker_1.0.rb: added backward compatibility
-	  test.
-	* test/rss/test_setup_maker_0.9.rb: ditto.
-	* test/rss/test_setup_maker_2.0.rb: ditto.
-
-	* test/rss/rss-testcase.rb: added convenience method for setting
-	  up taxo:topic.
-	* test/rss/rss-assertions.rb: added assertion for taxo:topic.
-
-	* sample/rss/blend.rb: followed new API.
-
-	* lib/rss/taxonomy.rb: changed class or module prefix to
-	  Taxonomy from Taxo.
-	* lib/rss/maker/taxonomy.rb: ditto.
-
-	* test/rss/test_taxonomy.rb: use #reject directory.
-
-	* lib/rss/: use #__send__ instead of #send.
-	* test/rss/: ditto.
-
-	* lib/rss/parser.rb: added entity handling type predicate.
-	* lib/rss/rexmlparser.rb: ditto.
-	* lib/rss/xmlparser.rb: ditto.
-	* lib/rss/xmlscanner.rb: ditto.
-
-	* lib/rss/xmlscanner.rb: more robust entity handling.
-
-	* test/rss/test_parser.rb: added an entity handling test.
-
-	* test/rss/test_2.0.rb: added RSS 2.0 tests.
-	* test/rss/rss-assertions.rb: extended XML stylesheet assertion.
-	* lib/rss/0.9.rb: added initialize method.
-	* test/rss/test_1.0.rb: cleanup.
-
-	* lib/rss/image.rb: added Image prefix.
-	* lib/rss/maker/image.rb: ditto.
-
-	* lib/rss/rss.rb: improved type conversion.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/0.9.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/syndication.rb: ditto.
-
-	* test/rss/test_2.0.rb: added type conversion tests.
-	* test/rss/test_accessor.rb: ditto.
-	* test/rss/test_to_s.rb: ditto.
-	* test/rss/test_syndication.rb: ditto.
-	* test/rss/test_setup_maker_2.0.rb: ditto.
-	* test/rss/test_setup_maker_1.0.rb: ditto.
-	* test/rss/test_setup_maker_0.9.rb: ditto.
-	* test/rss/test_maker_sy.rb: ditto.
-	* test/rss/test_maker_image.rb: ditto.
-	* test/rss/test_maker_2.0.rb: ditto.
-	* test/rss/test_maker_0.9.rb: ditto.
-	* test/rss/test_image.rb: ditto.
-
-	* test/rss/test_maker_1.0.rb: use assert instead of assert_equal.
-
-	* test/rss/rss-assertions.rb: improved type conversion assertions.
-
-	* lib/rss/rss.rb: added backward compatibility codes.
-	* lib/rss/parser.rb: ditto.
-	* test/rss/test_parser.rb: ditto.
-	* test/rss/test_2.0.rb: ditto.
-
-Sat Jun 17 02:01:00 2006  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pp.rb (Kernel#pretty_inspect): defined for pretty printed
-	  string.
-
-Sat Jun 17 00:23:58 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (reswords): kDO_BLOCK was missing.  fixed: [ruby-core:7995]
-
-Sat Jun 17 00:02:15 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_propertyput): support
-	  PROPERTYPUTREF. [ruby-talk:183042]
-
-	* ext/win32ole/tests/test_propertyputref.rb: ditto.
-
-Thu Jun 15 23:02:47 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (fole_methods): The return value
-	  of WIN32OLE#ole_methods should include PROPERTYPUTREF methods.
-
-	* ext/win32ole/win32ole.c (fole_put_methods): The return value
-	  of WIN32OLE#ole_put_methods should include PROPERTYPUTREF methods.
-
-	* ext/win32ole/tests/test_ole_methods.rb: ditto.
-
-	* ext/win32ole/tests/testall.rb : ditto.
-
-Wed Jun 14 18:23:28 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* enum.c (enum_any): Documentation typo.
-
-Wed Jun 14 15:01:09 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser#warn): Don't print
-	  warnings when -q is set.
-
-Wed Jun 14 23:03:53 2006  Tanaka Akira  <akr@m17n.org>
-
-	* configure.in: check sizeof(rlim_t).
-	  check setrlimit.
-
-	* process.c (proc_getrlimit): new method Process.getrlimit.
-	  (proc_setrlimit): new method Process.setrlimit.
-
-	* ruby.h (NUM2ULL): new macro.
-
-Mon Jun 12 22:25:09 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): adjust precision length to prevent
-	  splitting multi-byte characters.  [ruby-list:42389]
-
-Sun Jun 11 23:20:07 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Arguable#getopts): pass self to the
-	  parser.
-
-Sun Jun 11 10:00:57 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.h (write): not need to define on bcc.
-
-Sun Jun 11 08:30:33 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser#getopts): new methods.
-
-Sat Jun 10 18:02:40 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/bigdecimal/lib/bigdecimal/newton.rb (Newton::nlsolve): typo
-	  fixed: raize -> raise.  [ruby-talk:196608]
-
-Thu Jun  8 14:19:17 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.[ch] (rb_w32_read, rb_w32_write): new functions.
-	  use recv() and send() when fd is socket. fixed: [ruby-dev:28694]
-
-Wed Jun  7 16:22:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/tempfile.rb (Tempfile::make_tmpname): put dot between
-	  basename and pid.  [ruby-talk:196272]
-
-Wed Jun  7 14:53:04 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (errmap): add some winsock errors.
-
-Wed Jun  7 11:34:38 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* configure.in: add new configure option `--with-winsock2' for mingw.
-
-	* win32/Makefile.sub (config.h): define USE_WINSOCK2 in config.h
-	  instead of in CPPFLAGS.
-
-	* ext/socket/extconf.rb: determine whether to use winsock2 or not
-	  by using with_config.
-
-Wed Jun  7 10:45:10 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/{configure.bat, setup.mak, Makefile.sub, win32.h}: add
-	  new configure option `--with-winsock2'.
-
-	* win32/win32.c (StartSockets): ditto.
-
-	* ext/socket/extconf.rb: ditto.
-
-	* win32/win32.c (open_ifs_socket): new function.
-
-	* win32/win32.c (StartSockets, rb_w32_socket): use open_ifs_socket()
-	  instead of socket().
-	  ifs socket support is backported from trunk.
-
-Wed Jun  7 09:14:44 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): binding for the return event hook should have
-	  consistent scope.  [ruby-core:07928]
-
-	* eval.c (EXEC_EVENT_HOOK): trace_func may remove itself from
-	  event_hooks.	no guarantee for arbitrary hook deletion.
-	  [ruby-dev:28632]
-
-Mon Jun  5 18:12:12 2006  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/socket.c (sock_s_unpack_sockaddr_in): reject
-	  non-AF_INET/AF_INET6 sockaddr.
-	  (sock_s_unpack_sockaddr_un): reject non-AF_UNIX sockaddr.
-	  [ruby-dev:28691]
-
-Sun Jun  4 20:40:19 2006  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/socket.c: fix sockaddr_un handling.
-	  [ruby-dev:28677]
-
-Fri Jun  2 22:08:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/forwardable.rb: RDoc typo fix from Jan Svitok
-	  <jan.svitok at gmail.com>.  [ruby-core:07943]
-
-Fri Jun  2 19:02:09 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: use create_header.
-
-	* ext/openssl/ossl.h, ext/openssl/openssl_missing.h:
-	  include RUBY_EXTCONF_H.
-
-Fri Jun  2 17:16:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (CLEANINGS): remove extconf.h by distclean if created.
-
-Fri Jun  2 00:11:19 2006  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/socket.c (s_recvfrom): alen may be zero with UNIXSocket
-	  too.  (tested on NetBSD 3.0)
-	  (s_recvfrom_nonblock): extracted from sock_recvfrom_nonblock.
-	  (sock_recvfrom_nonblock): use s_recvfrom_nonblock.
-	  (ip_recvfrom_nonblock): new method: IPSocket#recvfrom_nonblock
-	  (unix_recvfrom_nonblock): new method: UNIXSocket#recvfrom_nonblock
-	  (s_accept_nonblock): extracted from sock_accept_nonblock.
-	  (sock_accept_nonblock): use s_accept_nonblock.
-	  (tcp_accept_nonblock): new method: TCPServer#accept_nonblock
-	  (unix_accept_nonblock): new method: UNIXServer#accept_nonblock
-
-Thu Jun  1 19:12:37 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_cmdvector): backslashes inside single-quotes
-	  no longer has special meanings.  fixed: [ruby-list:42311]
-
-Thu Jun  1 16:14:41 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_getcwd): runtime's getcwd() will not success
-	  if the length of the cwd is longer than MAX_PATH.
-	  fixed [ruby-list:42335]
-
-Thu Jun  1 11:29:14 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_getcwd): set errno if not set.
-	  fixed [ruby-list:42346]
-
-Sat May 27 11:29:46 2006  nobuyoshi nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): remove extinit files if no statically linked
-	  extensions.
-
-Fri May 26 09:05:11 2006  nobuyoshi nakada  <nobu@ruby-lang.org>
-
-	* ruby.h, lib/mkmf.rb (create_header): clear command line options for
-	  macros moved to extconf.h.
-
-	* ext/extmk.rb (extract_makefile, extmk): made RUBY_EXTCONF_H and
-	  EXTSTATIC permanent.
-
-	* ext/{dbm,digest/*,socket,zlib}/extconf.rb: used $defs and $INCFLAGS.
-
-	* {bcc32,win32,wince}/Makefile.sub (COMPILE_C, COMPILE_CXX): added
-	  $(INCFLAGS).
-
-	* lib/mkmf.rb (configuration): add $defs unless extconf.h was created.
-
-Thu May 25 01:52:07 2006  nobuyoshi nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (pkg_config): particular config commands support.
-
-	* ext/extmk.rb: deal with $static set in extconf.rb.
-
-	* mkconfig.rb: merge multiple entries to an entry with multiple lines.
-
-	* lib/mkmf.rb: allow a series of commands to link.
-
-	* win32/Makefile.sub: embed manifests.
-
-	* win32/setup.mak: suffix OS name by runtime version.
-
-Wed May 24 23:52:11 2006  nobuyoshi nakada  <nobu@ruby-lang.org>
-
-	* configure.in (ac_install_sh): ignore dummy install-sh.
-	  [ruby-talk:193876]
-
-Wed May 24 03:10:48 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/ssl.rb
-	  (OpenSSL::SSL::SocketForwarder#setsockopt,getsockopt): typo fixed.
-
-Mon May 22 17:54:12 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/socket/socket.c (sock_recvfrom_nonblock): use rb_read_pending
-	  instead of rb_io_read_pending.
-	  [ruby-dev:28663]
-
-Mon May 22 17:30:04 2006  Tanaka Akira  <akr@m17n.org>
-
-	* rubyio.h (rb_io_set_nonblock): declared.
-
-	* io.c (rb_io_set_nonblock): new function.
-	  (io_getpartial): nonblocking read support.
-	  (io_read_nonblock): new method: IO#read_nonblock.
-	  (io_write_nonblock): new method: IO#write_nonblock.
-
-	* ext/socket/socket.c
-	  (sock_connect_nonblock): new method: Socket#connect_nonblock.
-	  (sock_accept_nonblock): new method: Socket#accept_nonblock.
-	  (sock_recvfrom_nonblock): new method: Socket#recvfrom_nonblock.
-
-	  [ruby-core:7917]
-
-Mon May 22 15:57:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (umethod_bind): should not update original class.
-	  [ruby-dev:28636]
-
-Mon May 22 13:38:57 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (ev_const_get): should support constant access from
-	  within instance_eval().  [ruby-dev:28327]
-
-Thu May 18 17:51:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* time.c (time_timeval): should round for usec floating
-	  number.  [ruby-core:07896]
-
-	* time.c (time_add): ditto.
-
-Thu May 18 17:11:45 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::out): support utf-8.  a patch from Fujioka
-	  <fuj at rabbix.jp>.  [ruby-dev:28649]
-
-Thu May 18 00:42:12 2006  nobuyoshi nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb, lib/mkmf.rb: use BUILD_FILE_SEPARATOR in Makefiles.
-
-Wed May 17 17:55:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (sys_warning): should not call a vararg function
-	  rb_sys_warning() indirectly.  [ruby-core:07886]
-
-Wed May 17 08:17:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* util.c (ruby_strtod): try to reduce errors using powersOf10
-	  table.  [ruby-dev:28644]
-
-Tue May 16 15:34:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* re.c (rb_reg_initialize): should not allow modifying literal
-	  regexps.  frozen check moved from rb_reg_initialize_m as well.
-
-Tue May 16 09:20:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* re.c (rb_reg_initialize): should not modify untainted objects in
-	  safe levels higher than 3.
-
-	* re.c (rb_memcmp): type change from char* to const void*.
-
-	* dir.c (dir_close): should not close untainted dir stream.
-
-	* dir.c (GetDIR): add tainted/frozen check for each dir operation.
-
-Mon May 15 17:42:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_symbol_arg):
-	  typo fixed.  a patch from Florian Gross <florg at florg.net>.
-
-Sat May 13 16:14:05 2006  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pp.rb (PP.mcall): new method.
-	  (Struct#pretty_print): call Kernel#class and Struct#members even if
-	  overridden.
-	  (Struct#pretty_print_cycle): ditto.
-	  [ruby-core:7865]
-
-Thu May 11 19:57:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* util.c (ruby_strtod): differ addition to minimize error.
-	  [ruby-dev:28619]
-
-Fri Aug 11 15:39:25 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/yaml/tag.rb: Replace nodoc with stopdoc so Module methods get
-	  documented.
-
-Thu May 11 18:10:43 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* util.c (ruby_strtod): should not raise ERANGE when the input
-	  string does not have any digits.  [ruby-dev:28629]
-
-Sun May  7 03:09:51 2006  Stephan Maka  <stephan@spaceboyz.net>
-
-	* lib/resolv.rb (Resolv::DNS::Requester::ConnectedUDP#initialize):
-	  Use AF_INET6 for nameservers containing colons.
-
-Sat May  6 00:38:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* signal.c (trap): sig should be less then NSIG.  Coverity found
-	  this bug.  a patch from Kevin Tew <tewk at tewk.com>.
-	  [ruby-core:07823]
-
-Thu May  4 02:24:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/syck/emitter.c (syck_scan_scalar): avoid accessing
-	  uninitialized array element.  a patch from Pat Eyler
-	  <rubypate at gmail.com>.  [ruby-core:07809]
-
-	* array.c (rb_ary_fill): initialize local variables first.  a
-	  patch from Pat Eyler <rubypate at gmail.com>.  [ruby-core:07810]
-
-	* ext/syck/yaml2byte.c (syck_yaml2byte_handler): need to free
-	  type_tag.  a patch from Pat Eyler <rubypate at gmail.com>.
-	  [ruby-core:07808]
-
-Wed May  3 02:12:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (make_hostent_internal): accept ai_family
-	  check from Sam Roberts <sroberts at uniserve.com>.
-	  [ruby-core:07691]
-
-Mon May  1 12:23:19 2006    <sinara@blade.nagaokaut.ac.jp>
-
-	* numeric.c (num_div): use floor rather than rb_Integer().
-	  [ruby-dev:28589]
-
-	* numeric.c (flo_divmod): the first element of Float#divmod should
-	  be an integer. [ruby-dev:28589]
-
-	* test/ruby/test_float.rb: add tests for divmod, div, modulo and remainder.
-
-Sat Apr 29 22:42:08 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1_decode0): should initialize
-	  flag. [ruby-core:07785]
-
-Fri Apr 28 10:53:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* util.c (ruby_strtod): should not cut off 18 digits for no
-	  reason.  [ruby-core:07796]
-
-	* util.c (ruby_strtod): fixed wrong conversion.
-
-Thu Apr 27 01:38:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_fill): internalize local variable "beg" to
-	  pacify Coverity.  [ruby-core:07770]
-
-Wed Apr 26 16:59:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (pack_unpack): now supports CRLF newlines.  a patch from
-	  <tommy at tmtm.org>.  [ruby-dev:28601]
-
-Tue Apr 25 18:00:05 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c (delete_slaves): maybe increment the reference
-	  count of a NULL Tcl_Obj [ruby-core:07759].
-
-Tue Apr 25 07:55:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/jcode.rb (String::tr_s): should have translated non
-	  squeezing character sequence (i.e. a character) as well.  thanks
-	  to Hiroshi Ichikawa <gimite at gimite.ddo.jp> [ruby-list:42090]
-
-Tue Apr 25 00:08:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* regex.c (re_compile_pattern): should check if c is not a
-	  multibyte character.  a patch from KIMURA Koichi
-	  <kimura.koichi at canon.co.jp>.  [ruby-dev:28598]
-
-Fri Apr 21 15:19:13 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c (lib_eventloop_ensure): refer freed pointer
-	  [ruby-core:07744] and memory leak.
-
-Fri Apr 21 12:14:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c: document update patch from Sam Roberts
-	  <sroberts at uniserve.com>.  [ruby-core:07701]
-
-Wed Apr 19 13:55:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (arg): too much NEW_LIST()
-
-	* eval.c (SETUP_ARGS0): remove unnecessary access to nd_alen.
-
-Wed Apr 19 11:57:04 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_eval): use ARGSCAT for NODE_OP_ASGN1.
-	  [ruby-dev:28585]
-
-	* parse.y (list_concat): revert last change.
-
-	* parse.y (arg): use NODE_ARGSCAT for placeholder.
-
-Wed Apr 19 11:13:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/getoptlong.rb (GetoptLong::get): RDoc update patch from
-	  mathew <meta at pobox.com>.  [ruby-core:07738]
-
-Wed Apr 19 10:13:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* variable.c (rb_const_set): raise error when no target klass is
-	  supplied.  [ruby-dev:28582]
-
-Wed Apr 19 09:49:36 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (list_concat): should not modify nodes other than
-	  NODE_ARRAY.  [ruby-dev:28583]
-
-Tue Apr 18 17:40:37 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: add a binding to a container for a slave IP.
-
-	* ext/tk/lib/tk.rb: update RELEASE_DATE.
-
-	* ext/tk/tcltklib.c: forget to reset a Tcl interpreter.
-
-	* ext/tk/stubs.c: fix potential bugs about handling rb_argv0.
-
-Tue Apr 18 00:11:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c: block_unique should be 1, not frame_unique.
-	  [ruby-dev:28577]
-
-Fri Aug 11 15:39:25 2006  Eric Hodel  <drbrain@segment7.net>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#find_body): Make RDoc
-	  ignore C function prototypes.  Patch by Tilman Sauerbeck
-	  <tilman at code-monkey.de>.  [ruby-core:8574]
-	* lib/yaml/tag.rb: Replace nodoc with stopdoc so Module methods get
-	  documented.
-
-Mon Apr 10 01:03:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* prec.c (prec_prec_f): documentation patch from
-	  <gerardo.santana at gmail.com>.  [ruby-core:07689]
-
-Sat Apr  8 02:34:34 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (rb_big_pow): second operand may be too big even if
-	  it's a Fixnum.  [ruby-talk:187984]
-
-Sat Apr  8 02:12:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* README.EXT: update symbol description.  [ruby-talk:188104]
-
-Thu Apr  6 23:28:47 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* COPYING: explicitly note GPLv2.  [ruby-talk:187922]
-
-Thu Apr  6 11:18:37 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/panedwindow.rb: lack of arguments. [ruby-core:7681]
-
-Thu Apr  6 01:04:47 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tcltklib.c: fix SEGV when embedding to an application.
-	  [ruby-core:7600]
-
-	* ext/tk/tcltklib.c: fix SEGV at exit. [ruby-talk:186489]
-
-	* ext/tk/tkutil/tkutil.c: follow to changing specification of
-	  instance_eval on ruby-1.9.x.
-
-	* ext/tk/lib/tk.rb: ditto.
-
-	* ext/tk/lib/multi-tk.rb: ditto.
-
-	* ext/tk/lib/tk.rb: remove warning about redefinition of methods.
-
-	* ext/tk/lib/tk/variable.rb: remove warning about unseting Tcl
-	  variables.
-
-Wed Mar 29 20:54:44 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (fole_getproperty): WIN32OLE#[] should accept
-	  multi arguments.
-
-	* ext/win32ole/tests/testWIN32OLE.rb (test_setproperty_bracket): ditto.
-
-Wed Mar 29 10:07:44 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/nkf.c (nkf_each_char_to_hex, encode_fallback_subchar,
-	  e2w_conv): support C90 compiler.
-
-Wed Mar 29 06:48:40 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (backtrace): reports aliased method names in a generated
-	  backtrace.  a patch from "U.Nakamura" <usa at garbagecollect.jp>.
-	  [ruby-dev:28471]
-
-Mon Mar 27 22:19:09 2006  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/{nkf.c, utf8tbl.c, config.h}: imported nkf 2.0.6.
-	  * Add --ic / --oc option and mapping tables.
-	  * Add fallback option.
-	  * Add --no-best-fit-chars option.
-	  * Fix some bugs.
-
-	* ext/nkf/nkf.c (nkf_split_options): added for parse option string.
-
-	* ext/nkf/lib/kconv.rb (Kconv.to*): add -m0.
-	  Note that Kconv.to* still imply -X.
-
-Mon Mar 27 03:17:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): insecure calling should be checked for non
-	  NODE_SCOPE method invocations too.
-
-	* eval.c (rb_alias): should preserve the current safe level as
-	  well as method definition.
-
-Fri Mar 24 23:14:30 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (yield_under_i): pass self again for instance_eval().
-	  [ruby-dev:28466]
-
-Fri Mar 24 17:20:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (rb_f_sleep): remove description about SIGALRM which
-	  is not valid on the current implementation.  [ruby-dev:28464]
-
-Thu Mar 23 10:47:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (method_missing): should support argument splat in
-	  super.  [ruby-talk:185438]
-
-Mon Mar 20 12:05:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* configure.in: Solaris SunPro compiler -rapth patch from
-	  <kuwa at labs.fujitsu.com>.  [ruby-dev:28443]
-
-Mon Mar 20 09:40:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* configure.in: remove enable_rpath=no for Solaris.
-	  [ruby-dev:28440]
-
-Fri Mar 17 19:08:49 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c, ext/openssl/ossl_nsspki.c: fix typo.
-	  [ruby-core:07571]
-
-Wed Mar 15 16:54:21 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): support libraries without *.so.
-
-Wed Mar 15 16:35:43 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c, ext/openssl/ossl_nsspki.c: should use
-	  "rb_str_new(0, 0)" to make empty string.
-
-Sat Mar 11 14:24:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::TextFormatter::wrap): removed
-	  space before argument parenthesis.  [ruby-talk:183630]
-
-	* ruby.1: a clarification patch from David Lutterkort
-	  <dlutter at redhat.com>.  [ruby-core:7508]
-
-Sat Mar  4 15:26:40 2006  Tanaka Akira  <akr@m17n.org>
-
-	* gc.c (id2ref): fix symbol test.
-
-Sat Mar  4 01:08:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rdoc/ri/ri_paths.rb (RI::Paths): adding paths from rubygems
-	  directories.  a patch from Eric Hodel <drbrain at segment7.net>.
-	  [ruby-core:07423]
-
-Thu Mar  2 19:44:18 2006  Tanaka Akira  <akr@m17n.org>
-
-	* gc.c: align VALUE with sizeof(RVALUE) globally.
-	  (is_pointer_to_heap): check alignment out of loop.
-	  (id2ref): avoid collision between symbols and objects.
-	  (rb_obj_id): ditto.  moved from object.c.
-	  [ruby-talk:178364] [ruby-core:7305]
-
-Thu Mar  2 18:58:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_thread_fd_writable): should not re-schedule output
-	  from KILLED thread (must be error printing).
-
-Thu Mar  2 17:57:49 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* gc.c: commited magic for reducing RVALUE size on windows. (24->20byte)
-	  [ruby-core:7474]
-
-Thu Mar  2 12:59:14 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (filetime_to_unixtime): should set tm_isdst to -1.
-	  stat() didn't treat daylight saving time property on WinNT.
-	  [ruby-talk:182100]
-
-Thu Mar  2 08:02:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (add_heap): heap_slots may overflow.  a patch from Stefan
-	  Weil <weil at mail.berlios.de>.
-
-Wed Mar  1 00:24:31 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/rdoc/parsers/parse_rb.rb (read_escape): could not handle /\^/.
-	  merged Mr. Ishizuka's lib/irb/ruby-lex.rb 's patch rev 1.29.
-	  [ruby-talk:181631] [ruby-dev:28404]
-
-Tue Feb 28 09:32:17 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/drb/extservm.rb (invoke_service_command): cannot invoke command
-	  if command name is quoted on mswin32. [ruby-dev:28400]
-
-Mon Feb 27 00:19:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ruby.h (SYM2ID): should not cast to signed long.
-	  [ruby-core:07414]
-
-Fri Feb 24 20:07:23 2006  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/drb/drbtest.rb (add_service_command): quote pathnames in the
-	  server's command line for space contained directory names.
-	  Thanks, arton. [ruby-dev:28386]
-
-Fri Feb 24 12:11:08 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* instruby.rb: install *.exe.manifest and *.dll.manifest if exist.
-	  It's for VC++8.
-
-Fri Feb 24 11:33:52 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub (HAVE_HYPOT): bcc32 has hypot().
-
-Fri Feb 24 11:19:58 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* time.c (time_new_internal): add prototype to tell the compiler
-	  arugments types.
-
-	* win32/win32.c (NtInitialize): need to set a handler for VC++8.
-
-Fri Feb 24 08:19:16 2006  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* test.rb: Removed.  Obsolete by test/nkf.
-
-	* ext/.document: enabled documents in nkf and kconv
-
-	* ext/nkf/nkf.c ext/nkf/lib/kconv.rb: Add rdoc.
-
-Thu Feb 23 22:39:59 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: use borlndmm.dll if possible. bcc32's RTL internal
-	  memory manager cannot handle large memory block properly.
-	  ex: 10000.times { "" << "." * 529671; GC.start } # crash
-	  [ruby-dev:28230]
-
-Thu Feb 23 13:20:28 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* eval.c (SETUP_ARGS0): fixed memory corruption. [ruby-dev:28360]
-
-Tue Feb 21 02:18:46 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* configure.in (mingw): have link.  [ruby-list:41838]
-
-	* win32/Makefile.sub (config.h): ditto.
-
-Tue Feb 21 02:07:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (f_arglist): should set command_start = Qtrue for
-	  command body.  [ruby-talk:180648]
-
-Mon Feb 20 17:37:26 2006  Tanaka Akira  <akr@m17n.org>
-
-	* mkconfig.rb: alias RbConfig for Config.
-
-Mon Feb 20 12:27:53 2006  Kent Sibilev  <ksruby@gmail.com>
-
-	* lib/rational.rb (Integer::gcd): small typo fix.
-	  [ruby-core:07395]
-
-Mon Feb 20 01:05:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rational.rb (Integer::gcd): replaced by gcd4 in
-	  [ruby-core:07390].  [ruby-core:07377]
-
-Mon Feb 20 00:57:02 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.h (OSSL_Debug): should not use __func__.
-	  [ruby-dev:28339]
-
-Sun Feb 19 04:46:29 2006  Guy Decoux  <ts@moulon.inra.fr>
-
-	* eval.c: initial value for block_unique must be 1.
-	  [ruby-talk:180420]
-
-Sat Feb 18 23:58:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/tracer.rb (Tracer::Tracer.add_filter): turn on tracer mode
-	  only when caller() level size is one.  [ruby-core:07389]
-
-	* lib/rdoc/parsers/parse_rb.rb: need not to require "tracer".
-	  [ruby-core:07389]
-
-	* sample/rtags.rb: ditto.
-
-Sat Feb 18 12:18:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/fileutils.rb (FileUtils::fu_world_writable): make it
-	  private.  [ruby-core:07383]
-
-Sat Feb 18 00:22:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/tracer.rb: merged a minor clarification patch from Daniel
-	  Berger <Daniel.Berger at qwest.com>.  [ruby-core:07376]
-
-Fri Feb 17 11:18:42 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* util.c (ruby_strtod): Float("1e") should fail. [ruby-core:7330]
-
-	* pack.c (EXTEND32): unpack("l") did not work where sizeof(long) != 4.
-	  [ruby-talk:180024]
-
-	* pack.c (pack_unpack): fixed integer overflow on template "w".
-	  [ruby-talk:180126]
-
-Fri Feb 17 09:39:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_thread_wait_for): sleep should always sleep for
-	  specified amount of time.  [ruby-talk:180067]
-
-Thu Feb 16 01:10:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (backtrace): frame->orig_func may not be initialized.
-	  [ruby-core:07367]
-
-Wed Feb 15 16:52:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_eval): NODE_OP_ASGN1 should allow splat in its
-	  argument list.  [ruby-core:07366]
-
-	* parse.y (arg): avoid unnecessary extra argument.
-	  [ruby-core:07366]
-
-	* eval.c (rb_eval): honor visibility on OP_ASGN1 and
-	  OP_ASGN2. [ruby-core:07366]
-
-Wed Feb 15 10:09:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (yield_under_i): should not pass self as an argument to
-	  the block for instance_eval.  [ruby-core:07364]
-
-Wed Feb 15 09:20:35 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_obj_instance_eval): should be no singleton classes for
-	  true, false, and nil.  [ruby-dev:28186]
-
-Tue Feb 14 18:48:33 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (DMETHOD_P): accessing wrong frame.  [ruby-dev:28181]
-
-	* eval.c (proc_invoke): preserve FRAME_DMETH flag.
-
-Tue Feb 14 15:13:51 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/zlib/zlib.c: supress warning on test/zlib. [ruby-dev:28323]
-
-Tue Feb 14 14:01:17 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* time.c (search_time_t): support non 32bit time_t environments.
-
-	* win32/Makefile.sub (config.h): VC++8 has ``long long'' type.
-
-	* win32/Makefile.sub (config.h): VC++8's time_t is 64bit value.
-
-	* win32/win32.c (rb_w32_utime): drop read-only attribute before
-	  changing file time.
-
-	  all changes are backported from CVS HEAD.
-
-Tue Feb 14 11:21:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (argf_forward): should not use frame->argv.
-	  [ruby-core:07358]
-
-Mon Feb 13 18:08:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): argument update propagation.  [ruby-dev:28044]
-
-	* env.h: remove argv member from struct FRAME.
-
-Mon Feb 13 13:27:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (eval): should push class from binding if supplied.
-	  [ruby-core:07347]
-
-Mon Feb 13 00:04:00 2006  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/erb.rb (ERB::Compiler): add instance variable @insert_cmd to
-	  change <%='s behavior. (backported 1.15 - 1.16)
-
-Sat Feb 11 02:04:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (eval): no need to push ruby_class.  [ruby-dev:28176]
-
-Sat Feb 11 01:57:44 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_f_autoload): check if ruby_cbase is nil (during
-	  instance_eval for objects cannot have singleton classes,
-	  e.g. fixnums and symbols).  [ruby-dev:28178]
-
-Tue Feb  7 23:03:24 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/zlib/zlib.c: should not access ruby objects in finalizer.
-	  [ruby-dev:28286]
-
-Mon Feb  6 16:02:51 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* file.c (rb_thread_flock): ERROR_NOT_LOCKED is not an error on Cygwin.
-	  In such situation, flock() should return 0.
-
-Mon Feb  6 00:41:08 2006  Tanaka Akira  <akr@m17n.org>
-
-	* ruby.h (RSTRUCT_LEN, RSTRUCT_PTR): defined for source level
-	  compatibility with ruby 1.9.
-
-Sun Feb  5 21:05:34 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* numeric.c (fix_to_s): removed workaround for radix 2. Historically,
-	  rb_fix2str could only handle radix 8, 10, 16. (Rev1.37) But for now,
-	  it can handle radix 2..36. [ruby-Bugs#3438] [ruby-core:7300]
-
-Sun Feb  5 18:55:08 2006  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: imported from trunk, rev 1.129
-
-	* lib/net/http.rb (add_field, get_fields): keep 1.8.2 backward
-	  compatibility.
-
-	* lib/net/https.rb: imported from trunk, rev 1.3.
-
-	* lib/net/https.rb: #use_ssl? definition moved from net/http.rb.
-
-Sun Feb  5 14:22:15 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/pstore.rb: should return default value if name is not found.
-	  [ruby-core:7304]
-
-	* lib/pstore.rb: should raise PStore::Error if not in transaction.
-
-Sat Feb  4 22:51:43 2006  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c: apply the FreeBSD getcontext/setcontext workaround
-	  only before FreeBSD 7-CURRENT.
-
-Sat Feb  4 21:19:23 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (LK_ERR): ERROR_NOT_LOCKED is not an error.
-	  In such situation, flock() should return 0.
-
-Sat Feb  4 15:56:37 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* numeric.c (fix_to_s): (2**32).to_s(2) fails with exception where
-	  sizeof(int) == 4 < sizeof(long). [ruby-core:7300]
-
-Fri Feb  3 15:06:50 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/syck/syck.c (syck_move_tokens): should reset p->cursor or etc
-	  even if skip == 0. This causes buffer overrun.
-	  (ex: YAML.load('--- "..' + '\x82\xA0' * 511 + '"'))
-
-Thu Feb  2 23:51:18 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/syck/emitter.c (syck_emitter_write): should not set '\0' on
-	  emitter's marker. if marker points to the end of buffer, this causes
-	  buffer overrun. (ex: YAML.dump("." * 12288))
-
-Thu Feb  2 16:01:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (eval): need not to protect $SAFE value.
-	  [ruby-core:07177]
-
-Thu Feb  2 14:45:53 2006  Ville Mattila  <ville.mattila@stonesoft.com>
-
-	* configure.in: The isinf is not regognized by autoconf
-	  library guesser on solaris 10. [ruby-core:7138]
-
-Wed Feb  1 22:01:47 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* configure.in, hash.c (ruby_setenv): use setenv(3) and unsetenv(3)
-	  where they are supported. modifing environ variable seems to
-	  segfault solaris 10. [ruby-core:7276] [ruby-dev:28270]
-
-	* ruby.c (set_arg0): if use setenv(3), environ space cannot be used
-	  for altering argv[0].
-
-Tue Jan 31 14:46:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* struct.c (rb_struct_select): update RDoc description.
-	  [ruby-core:7254]
-
-Tue Jan 31 11:58:51 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: add MultiTkIp#eval and bg_eval.
-
-	* ext/tk/lib/tk/namespace.rb: TkNamespace#eval was enbugged at the
-	  last commit. Now it will return a proper object.
-
-Tue Jan 31 00:10:26 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/syck/rubyext.c (syck_resolver_transfer): workaround for SEGV.
-	  ex: ruby -ryaml -e 'YAML.load("!map:B {}")' [ruby-core:7217]
-
-Sat Jan 28 07:56:57 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/rdoc/usage.rb: support "a:0:33" style caller[-1]. In this case
-	  file name is "a:0". I don't know this really happens though...
-	  [ruby-Bugs:3344]
-
-Wed Jan 25 22:29:04 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in, dln.c, file.c, intern.h, missing.h (eaccess): use
-	  system routine if provided.  fixed: [ruby-core:07195]
-
-Sun Jan 22 23:27:13 2006  Go Noguchi  <gonoguti@yahoo.co.jp>
-
-	* lib/test/unit/autorunner.rb (process_args): ignore arguments after
-	  '--' so that test scripts can handle them.  fixed: [ruby-dev:28258]
-
-Sun Jan 22 22:09:52 2006  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (POST_GETCONTEXT): define separately from PRE_GETCONTEXT on
-	  IA64 to avoid reusing variable address.
-
-Sun Jan 22 20:03:35 2006  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (ruby_setjmp): define PRE_GETCONTEXT and POST_GETCONTEXT
-	  instead of FUNCTION_CALL_MAY_RETURN_TWICE.
-	  define PRE_GETCONTEXT to clear carry flag for workaround of
-	  FreeBSD/i386 getcontext/setcontext bug.
-	  [ruby-dev:28263]
-
-Sat Jan 21 00:36:47 2006  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (FUNCTION_CALL_MAY_RETURN_TWICE): use only on SPARC and IA64
-	  before gcc 4.0.3.
-	  [ruby-dev:28247]
-
-Thu Jan 19 22:21:23 2006  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mv): should remove file after copying.
-	  [ruby-dev:28223]
-
-Wed Jan 18 23:37:06 2006  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (FUNCTION_CALL_MAY_RETURN_TWICE): don't clobber %l7 of SPARC
-	  if enable-shared.
-	  (ruby_setjmp): call FUNCTION_CALL_MAY_RETURN_TWICE after getcontext
-	  too.
-	  reported by Pav Lucistnik and Marius Strobl.
-	  http://lists.freebsd.org/pipermail/freebsd-sparc64/2006-January/003739.html
-
-Tue Jan 17 11:32:46 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/setup.mak (MAKE): workaround for nmake 8.
-
-Tue Jan 17 11:10:21 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/{Makefile.sub,setup.mak}: invoke .bat via shell. workaround
-	  for nmake 8.
-
-Mon Jan 16 10:26:23 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/syck/emitter.c (syck_emit_seq, syck_emit_map, syck_emit_item):
-	  should output complex key mark even if map's key is empty seq/map.
-	  [ruby-core:7129]
-
-Sat Jan 14 05:37:06 2006  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (READ_DATA_PENDING, READ_DATA_PENDING_COUNT): defined
-	  for DragonFly BSD 1.4.0.
-
-Sat Jan 14 03:43:24 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* file.c (rb_file_s_chmod): avoid warning where sizeof(int) !=
-	  sizeof(void*).
-
-Fri Jan 13 19:14:56 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/rdoc/diagram.rb:
-	    - properly quote bare element attributes
-	    - terminates dangling elements (e.g. <img>, <br>, <link>, etc)
-	    - converts "CVS" to the more HTML-friendly acronym element
-	    - adds missing type attributes to style elements
-
-	  based on Paul Duncan's patch <pabs@pablotron.org> [ruby-core:7028]
-
-	* lib/rdoc/generators/html_generator.rb: ditto.
-	* lib/rdoc/generators/template/html/hefss.rb: ditto.
-	* lib/rdoc/generators/template/html/html.rb: ditto.
-	* lib/rdoc/generators/template/html/kilmer.rb: ditto.
-
-Thu Jan 12 11:53:08 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tkballoonhelp.rb: [bug fix] couldn't add to a widget
-	  which is constructed with TkComposite module.
-	  [new feature] support 'command' option which is called just before
-	  popping up the balloon help.
-
-Wed Jan 11 15:00:00 2006  Ville Mattila  <mulperi@iki.fi>
-
-	* io.c (READ_PENDING*): Support solaris 64-bit environments.
-	  Solaris defines a opaque FILE struct when compiling 64 bit
-	  binaries. This means that we dont have access to _ptr etc.
-	  members anymore. The solution by Steven Lumos is to define
-	  FILE64 that has needed members available. I've modified
-	  the origanal patch a bit so that it compiles both with gcc
-	  and now free sun studio 11 compiler and both amd64 and sparc.
-	  NOTE! We have to 64 bit solaris FILE structure time to time
-	  otherwise we'll get breakage.
-	  [ruby-core:7106]
-
-Tue Jan 10 19:42:33 2006  Tanaka Akira  <akr@m17n.org>
-
-	* gc.c (garbage_collect): mark ruby_current_node.
-	  if an exception is raised in a finalizer called written in C by
-	  rb_gc_call_finalizer_at_exit, ruby_set_current_source may use
-	  collected ruby_current_node and mark_source_filename may corrupt
-	  memory.
-
-Tue Jan 10 13:30:34 2006  akira yamada  <akira@ruby-lang.org>
-
-	* ext/syck/rubyext.c (syck_resolver_transfer): should be able to load
-	  !ruby/object:Bignum syntax 1.8.3 dumped. [ruby-core:6159]
-
-Tue Jan 10 12:47:41 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/yaml/rubytypes.rb (Fixnum): Bignum could not be loaded in
-	  ruby 1.8.3/1.8.4. [ruby-core:6115]
-
-	* lib/yaml/rubytypes.rb (Numeric): Subclass of Numeric could not
-	  be dumped properly. [ruby-core:7047]
-
-Tue Jan 10 12:00:48 2006  Aaron Schrab  <aaron @nospam@ schrab.com>
-
-	* lib/yaml/rubytypes.rb (Symbol#yaml_new): YAML loading of quoted
-	  Symbols broken. [ruby-Bugs:2535]
-
-Mon Jan  9 19:54:35 2006  arton  <artonx@yahoo.co.jp>
-
-	* ext/zlib/extconf.rb: zlib compiled DLL version 1.2.3 distributed by
-	  http://www.zlib.net/ has zdll.lib. [ruby-dev:28209]
-
-Mon Jan  9 14:17:12 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/Makefile.sub (OPTFLAGS): I have experienced trouble on y- flag,
-	  (VisualC++6) so use -O2b2xg- if  $(MSC_VER) < 1400. [ruby-core:7040]
-
-Mon Jan  9 14:17:12 2006  Kero van Gelder  <rubyforge @nospam@ kero.tmfweb.nl>
-
-	* lib/webrick/httpservlet/filehandler.rb: fixed typo. [ruby-core:7075]
-
-Sat Jan  7 15:40:07 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (singleton): get rid of segfault on syntax error.
-	  fixed: [ruby-core:07070]
-
-Fri Jan  6 10:16:20 2006  Steven Lumos  <steven@lumos.us>
-
-	* io.c (READ_DATA_PENDING): defined for 64bit Solaris on SPARC.
-	  [ruby-core:7057]
-	  (READ_DATA_PENDING_COUNT): ditto.
-	  (READ_DATA_PENDING_PTR): ditto.
-
-Sun Jan  1 17:07:59 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (rb_w32_seekdir): should not segfault even if passed
-	  the location which rb_w32_telldir didn't return. [ruby-core:7035]
-	  (I think HEAD implementation is better. but binary compatibility)
-
-	* test/ruby/test_dir.rb: added.
-
-Sat Dec 31 22:57:00 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_save_context): should not recycle scope object used
-	  in a thread.  fixed: [ruby-dev:28177]
-
-Fri Dec 30 18:22:42 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* gc.c (garbage_collect): mark objects refered from aborting threads.
-	  [ruby-dev:28190]
-
-	* win32/Makefile.sub: VC++8 support.
-
-Fri Dec 30 14:24:53 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* dir.c (glob_helper): do not use TRUE for djgpp.
-
-Fri Dec 30 04:54:40 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* file.c (eaccess): workaround for VC++8 runtime.
-
-	* win32/win32.c (ioinfo): VC++8 support.
-
-Thu Dec 29 23:59:37 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_gc_mark_threads): leave unmarked threads which won't wake
-	  up alone, and mark threads in the loading table.  [ruby-dev:28154]
-
-	* eval.c (rb_gc_abort_threads), gc.c (gc_sweep): kill unmarked
-	  threads.  [ruby-dev:28172]
-
-Thu Dec 29 17:02:07 2005  Tanaka Akira  <akr@m17n.org>
-
-	* test/ruby/envutil.rb (EnvUtil.rubybin): search "ruby" instead of
-	  "miniruby".  [ruby-dev:28140]
-
-Tue Dec 27 16:59:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* test/drb/drbtest.rb (DRbService::self.ext_service): increase
-	  timeout limit.  a patch from Kazuhiro NISHIYAMA
-	  <zn at mbf.nifty.com>. [ruby-dev:28132]
-
-Tue Dec 27 08:29:18 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLSocket#post_connection_chech):
-	  treat wildcard character in commonName. [ruby-dev:28121]
-
-Mon Dec 26 22:32:47 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval), gc.c (gc_mark_children), node.h (NEW_ALIAS,
-	  NEW_VALIAS), parse.y (fitem): allow dynamic symbols to
-	  NODE_UNDEF and NODE_ALIAS.
-	  backported from trunk.  fixed: [ruby-dev:28105]
-
-Mon Dec 26 08:50:36 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (ev_const_get): fixed a bug in constant reference during
-	  instance_eval.  [yarv-dev:707]
-
-	* eval.c (ev_const_defined): ditto.
-
-	* lib/yaml.rb (YAML::add_domain_type): typo fixed.  a patch from
-	  Joel VanderWerf <vjoel at path.berkeley.edu>.
-	  [ruby-talk:165285] [ruby-core:6995]
-
-Sat Dec 24 18:58:14 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* stable version 1.8.4 released.
-
-Fri Dec 23 10:30:23 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/digest/sha2/sha2.c (ULL): support AIX C.  a patch from
-	  Kailden <kailden at gmail.com>.  [ruby-core:06984]
-
-Wed Dec 21 16:53:06 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* file.c (w32_io_info): should return handle because FileIndex is
-	  valid only while file is open. [ruby-dev:28088]
-
-Wed Dec 21 14:53:26 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (test_kernel_open): use File.identical?.
-	  [ruby-talk:171804]
-
-Tue Dec 20 22:41:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (eval_under_i): evaluate source in caller's frame.
-	  [ruby-dev:28076]
-
-	* eval.c (rb_call_super): use original method name on exception.
-	  [ruby-dev:28078]
-
-Tue Dec 20 13:11:59 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/syck/rubyext.c: fixed GC problem (backported HEAD 1.55 - 1.62)
-	  [ruby-dev:27839]
-
-	* ext/syck/syck.h (S_FREE): small hack. no need to check if pointer is
-	  NULL or not before S_FREE.
-
-	* st.c: uses malloc instead of xmalloc to avoid GC. syck uses st_insert
-	  in gram.c to insert node from rb_syck_bad_anchor_handler into
-	  SyckParser's hash table. if GC occurs in st_insert, it's not under
-	  SyckParser's mark system yet. so RString can be released wrongly.
-	  [ruby-dev:28057]
-
-Tue Dec 20 12:53:23 2005  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/rubyext.c (syck_emitter_reset): to ensure compatibility
-	  with previous Ruby versions, documents are no longer headless.
-
-Tue Dec 20 01:46:48 2005  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (rb_f_backquote): fix a GC problem on
-	  IA64 with gcc 4.0.3 20051216 (prerelease) -O3.
-
-Mon Dec 19 23:32:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (rb_symname_p): fixed wrong validation.  [ruby-dev:28047]
-
-Sat Dec 17 03:57:01 2005  Tanaka Akira  <akr@m17n.org>
-
-	* bignum.c (rb_big_rshift): fix a GC problem on
-	  IA64 with gcc 4.0.3 20051216 (prerelease).
-
-Sat Dec 17 03:30:23 2005  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (bmcall): fix a GC problem by tail call on
-	  IA64 with gcc 4.0.3 20051216 (prerelease).
-
-Fri Dec 16 00:54:06 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* signal.c (Init_signal): revert C++ style comment.
-	  [ruby-dev:28041]
-
-Thu Dec 15 12:35:14 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/tmpdir.rb: merged RDoc patch from Eric Hodel <drbrain at
-	  segment7.net>.  [ruby-core:06894]
-
-Thu Dec 15 01:33:31 2005  Tanaka Akira  <akr@m17n.org>
-
-	* ext/zlib/zlib.c (zstream_run): fix a GC problem by tail call on
-	  x86_64 with gcc 4.0.3 20051111 (prerelease) (Debian 4.0.2-4)
-
-Wed Dec 14 12:11:46 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* test/gdbm/test_gdbm.rb: specify pid for the argument of
-	  Process.wait.  workaround for Cygwin.
-
-Wed Dec 14 12:01:26 2005  Tanaka Akira  <akr@m17n.org>
-
-	* marshal.c (r_object0): fix a GC problem for reading a bignum on
-	  IA64 with gcc 3.3.5 (Debian 1:3.3.5-13).
-
-Tue Dec 13 12:23:47 2005  Tanaka Akira  <akr@m17n.org>
-
-	* re.c (rb_reg_regcomp): fix a GC problem on x86_64 with
-	  gcc 3.3.5 (Debian 1:3.3.5-13).
-
-Tue Dec 13 01:44:16 2005  Tanaka Akira  <akr@m17n.org>
-
-	* array.c (rb_ary_diff): fix a GC problem on IA64 with
-	  gcc 3.3.5 (Debian 1:3.3.5-13).
-	  When rb_ary_push is called, there was no register which contains
-	  `hash' but `&RHASH(hash)->tbl' instead.
-
-Tue Dec 13 00:08:09 2005  Tanaka Akira  <akr@m17n.org>
-
-	* sprintf.c (rb_str_format): fix a GC problem.
-	  [ruby-dev:28001]
-
-Mon Dec 12 15:54:56 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* test/openssl/test_ssl.rb (test_parallel): call GC.start to close
-	  unused files. [ruby-dev:27981]
-
-Mon Dec 12 00:33:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/digest/digest.c (rb_digest_base_s_digest): add volatile to
-	  protect temporary context object.  [ruby-dev:27979]
-
-	* ext/iconv/iconv.c (Init_iconv): rb_gc_register_address() should
-	  be called before actual variable initialization.
-	  [ruby-dev:27986]
-
-Fri Dec  9 23:31:02 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/rexml/encoding.rb (encoding=): give priority to particular
-	  conversion to iconv.  [ruby-core:06520]
-
-Thu Dec  8 02:07:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (umethod_bind): adjust invoking class for module method.
-	  [ruby-dev:27964]
-
-Thu Dec  8 00:40:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (call_trace_func): klass parameter should be a
-	  class/module that defines calling method.  [ruby-talk:169307]
-
-Wed Dec  7 17:10:27 2005  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
-
-	* sprintf.c (rb_f_sprintf): [ruby-dev:27967]
-
-Wed Dec  7 15:31:35 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* sprintf.c (rb_str_format): integer overflow check added.
-
-	* sprintf.c (GETASTER): ditto.
-
-Wed Dec  7 01:02:04 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/README.macosx-aqua: [new document] tips to avoid the known
-	  bug on platform specific dialogs of Tcl/Tk Aqua on MacOS X.
-
-	* ext/tk/tcltklib.c: fix bug on switching threads and waiting on the
-	  deleted interpreter on vwait and tkwait command.
-
-	* ext/tk/lib/multi-tk.rb: kill the meaningless loop for the deleted Tk
-	  interpreter.
-
-	* ext/tk/sample/demos-jp/image3.rb: [bug fix] wrong argument.
-
-	* ext/tk/sample/demos-en/image3.rb: ditto.
-
-	* ext/tk/sample/demos-jp/menu.rb: fix message for MacOS X.
-
-	* ext/tk/sample/demos-jp/menu8x.rb: ditto.
-
-	* ext/tk/sample/demos-en/menu.rb: ditto.
-
-Tue Dec  6 16:37:57 2005  Yuya Nishida  <yuya@j96.org>
-
-	* eval.c (exec_under): avoid accessing ruby_frame->prev.
-	  [ruby-dev:27948]
-
-Thu Dec  1 00:50:33 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_funcall2): allow to call protected methods.
-	  fixed: [ruby-dev:27890]
-
-Wed Nov 30 23:52:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (NEWHEAP, ADD2HEAP): set count after pointer was set.
-	  fixed: [ruby-dev:27896]
-
-Wed Nov 30 13:43:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-expr-beg): support $! at the end of
-	  expression.   [ruby-dev:27868]
-
-Mon Nov 28 18:55:43 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/socket/socket.c (init_inetsock_internal): remove setting
-	  SO_REUSEADDR option on server socket on Cygwin.
-	  fixed: [ruby-core:6765] ([ ruby-Bugs-2872 ])
-
-Mon Nov 28 13:08:54 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (rb_w32_strerror): remove all CR and LF. (avoid broken
-	  error message on bccwin32 + winsock)
-
-Mon Nov 28 09:21:49 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/mkmf.rb (create_makefile): should not change sodir with
-	  dir.gsub!. (bccwin32 failed to install third party exntesions)
-	  [ruby-dev:27834]
-
-Sun Nov 27 00:56:13 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/wsdl/xmlSchema/complexContent.rb: missing
-	  ComplexContent#elementformdefault method.
-
-Sat Nov 26 19:57:45 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* dln.c (conv_to_posix_path): should initialize posix.
-
-Thu Nov 24 21:05:58 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* configure.in (AC_CHECK_FUNCS): need to check link().
-	  fixed: [ruby-dev:27814]
-
-Thu Nov 24 01:22:25 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* file.c (w32_io_info): CreateFile failed on Win9x if file was already
-	  opened. (FILE_SHARE_READ was needed, but actually I don't understand
-	  the flags of CreateFile well...)
-
-Wed Nov 23 20:59:01 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: add Tk.pkgconfig_list and Tk.pkgconfig_get
-	  [Tk8.5 feature].
-
-	* ext/tk/lib/tk/text.rb: supports new indices modifires on a Text
-	  widget [Tk8.5 feature].
-
-	* ext/tk/lib/tk/virtevent.rb: add TkNamedVirtualEvent.
-
-	* ext/tk/lib/tk/autoload.rb: ditto.
-
-	* ext/tk/lib/tk/event.rb: add :data key for virtual events [Tk8.5
-	  feature].
-
-Wed Nov 23 18:55:31 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* file.c (w32_io_info): should not call GetFileInformationByHandle
-	  for pipe.
-
-	* file.c (w32_io_info): checks return value from rb_w32_get_osfhandle.
-
-	* file.c (w32_io_info): now can identify directory on WinNT.
-
-Wed Nov 23 03:40:49 2005  Guy Decoux  <ts@moulon.inra.fr>
-
-	* re.c (KR_REHASH): should cast to unsigned for 64bit CPU.
-	  [ruby-core:06721]
-
-Wed Nov 23 11:01:33 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* intern.h, file.c: failed to compile on windows.
-
-Wed Nov 23 07:26:44 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: check for X509V3_EXT_nconf_nid.
-
-	* ext/openssl/ossl_x509ext.c (MakeX509ExtFactory): should use
-	  OPENSSL_malloc to allocate X509V3_CTX.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_create_ext): use
-	  X509V3_EXT_nconf_nid to avoid SEGV (and to build extensions which
-	  values are placed in separate section).
-
-	* test/openssl/test_x509ext.rb: new file.
-
-Wed Nov 23 01:22:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (test_identical): test if two files are identical.
-
-	* file.c (rb_f_test): support DOSISH systems where st_ino is not
-	  reliable.  fixed: [ruby-core:06672]
-
-	* win32.h, win32.c (rb_w32_osid): check the running platform.
-
-Tue Nov 22 23:52:06 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb: match incomplete (in current enconding) multibyte
-	  string.  http://inamode6.tokuhirom.dnsalias.org/show/1551
-
-Tue Nov 22 18:36:11 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (winnt_stat): set mapped errno instead of ENOENT.
-
-Tue Nov 22 14:46:57 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* file.c (rb_file_s_basename): skip slashes just after UNC top slashes.
-
-	* test/ruby/test_path.rb (test_dirname, test_basename): follow new
-	  spec. and add new tests.
-
-Tue Nov 22 13:18:32 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (rb_w32_stat): Dir.chdir('//server/shared');
-	  p Dir.glob('*') should work on WinNT. (implemented our own
-	  stat(2) on WinNT) [ruby-list:41552] [ruby-dev:27711]
-
-Tue Nov 22 02:31:53 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tkextlib/tile.rb: bug fix (Tk::Tile::USE_TTK_NAMESPACE
-	  is not defined).
-
-Tue Nov 22 01:45:21 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_s_basename): DOSISH_UNC is defined on cygwin but
-	  DOSISH is not.  fixed: [ruby-dev:27797]
-
-Mon Nov 21 22:50:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_path_skip_prefix, rb_file_s_basename): UNC without path
-	  should not be splitted.  fixed: [ruby-dev:27776] [ruby-dev:27786]
-
-	* parse.y (dsym): prohibit empty symbol literal by interpolation.
-	  fixed: [ruby-talk:166529]
-
-Mon Nov 21 16:03:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/setup.mk: findstr doesn't exist on win9x.
-	  fixed: [ruby-dev:27756]
-
-Sun Nov 20 22:34:06 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (rb_symname_p): [ not followed by ] is not valid symbol.
-	  fixed: [ruby-talk:166520]
-
-Sat Nov 19 19:57:54 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/fileutils.rb (FileUtils::ln): ln documentation fix.
-	  [ruby-core:06661]
-
-Sat Nov 19 07:34:32 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/font.rb: remove dependency on Ruby's version (1.8
-	  or 1.9).
-
-	* ext/tk/lib/tkextlib/ICONS/icons.rb: ditto.
-
-	* ext/tk/sample/tkextlib/treectrl/demo.rb: ditto.
-
-Fri Nov 18 17:57:08 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* file.c (rb_file_s_dirname): should use skipprefix for UNC path.
-	  pointed out by nobu ([ruby-dev:27744]). fixed: [ruby-core:5076]
-
-Fri Nov 18 17:35:09 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: add restriction to access the entried
-	  command table and manipulate other IPs (for reason of security).
-	  Now, a IP object can be controlled by only its master IP or the
-	  default IP.
-
-	* ext/tk/lib/remote-tk.rb: add restriction to manipulate.
-
-	* ext/tk/tcltklib.c (ip_is_slave_of_p): add TclTkIp#slave_of?(ip)
-	  to check manipulability.
-
-	* ext/tk/lib/tk.rb: bug fix on handling of Tcl's namespaces.
-
-	* ext/tk/lib/tk/namespace.rb: ditto.
-
-Fri Nov 18 17:26:06 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* file.c (rb_file_s_dirname): added checks for some patterns with drive
-	  letter. fixed: [ruby-dev:27738]
-
-	* test/ruby/test_path.rb (test_dirname): added tests for above
-	  patterns.
-
-Fri Nov 18 12:18:02 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.h (S_IFIFO): r,w = IO.pipe; r.stat.pipe? now
-	  returns true on VisualC++6.
-
-Wed Nov 16 23:24:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* common.mk (static-ruby): overridable.
-
-	* ext/extmk.rb (parse_args): force to link extensions statically only
-	  if static is given for extstatic.
-
-	* ext/extmk.rb (RUBY, RUBYW): overridable.
-
-Tue Nov 15 23:46:35 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/find.rb (Find::find): should not ignore symbolic links to
-	  non-existing files.  [ruby-talk:165866]
-
-Tue Nov 15 16:23:26 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* array.c (rb_ary_fill): previous commit disabled this usage:
-
-	    a = [0,1,2,3,4,5,6,7,8,9]
-	    a.fill {|i| a[i] * 10} #=> [nil, nil, ...., nil]
-
-	  previous commit has the advantage of early garbage collection, but
-	  potensially this would break some script. so I reverted behavior.
-
-Tue Nov 15 16:04:10 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* array.c (rb_ary_fill): tail elements were vanished when the middle
-	  part of array was filled. (ie: [0,1,2,3,4].fill(-1,2,1) => [0,1,-1])
-
-	* test/ruby/test_array.rb (test_fill): added.
-
-Tue Nov 15 14:39:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_fill): should adjust array length correctly when
-	  an array is expanded in the fill process.  [ruby-core:06625]
-
-Mon Nov 14 23:49:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_s_readlink): ERANGE will occur only on GPFS.
-	  [ruby-dev:27699]
-
-Mon Nov 14 17:36:22 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_first): RDoc update from Daniel Berger
-	  <djberg96@yahoo.com>.  [ruby-core:06577].
-
-Fri Nov 11 10:31:44 2005  Zach Dennis  <zdennis@mktec.com>
-
-	* ext/socket/socket.c: Socket Documentation. [ruby-core:6552]
-
-Fri Nov 11 08:20:56 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* Makefile.in (OUTFLAG): keep trailing spaces.  [ruby-dev:27666]
-
-	* mkconfig.rb: substitution refereces added.
-
-Fri Nov 11 07:44:18 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* configure.in: undef HAVE_LINK on BeOS. (link(2) always returns
-	  EINVAL, and this causes error in test/fileutils.)
-
-	* file.c: overwride chown(2) and fchown(2) on BeOS. (these functions
-	  should not change user/group id if -1 is passed as corresponding
-	  argument, and this causes error in test/fileutils too)
-	  [ruby-dev:27672]
-
-	* file.c (rb_file_s_link): checks HAVE_LINK.
-
-Tue Nov  8 15:32:27 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/drb/ssl.rb (DRb::SSLConfig#accept): fixed typo.
-	  [ruby-dev:27560] [ruby-core:4627]
-
-Mon Nov  7 13:43:51 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/stubs.c (_nativethread_consistency_check): use simpler
-	  (low cost) way to check whether the Tcl interpreter was compiled
-	  with threads enabled of not.
-
-	* ext/tk/tcltklib.c: reduce warnings.
-
-	* ext/tk/tkutil/tkutil.c: ditto.
-
-Mon Nov  7 00:06:58 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/yaml.rb: removed :nodoc: to generate Kernel doc. [ruby-core:6324]
-
-Sun Nov  6 23:39:13 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (Iconv::BrokenLibrary): exception when detected a
-	  bug of underlying library.
-
-Sun Nov  6 21:46:59 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tk/stubs.c (ruby_tcl_create_ip_and_stubs_init): should touch
-	  interpreter after initialization is done. [ruby-dev:27638]
-
-Sun Nov  6 20:13:27 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_s_readlink): readlink(2) on AIX fails with ERANGE if
-	  buffer size is less than required.  fixed: [ruby-dev:27634]
-
-Wed Nov  2 20:25:28 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/extconf.rb: ext/tk/extconf.rb: change the check parameter
-	  for Win32.
-
-Wed Nov  2 20:14:53 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib: merge into ext/tk and remove.
-
-Wed Nov  2 19:03:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_rbUpdateObjCmd,
-	  ip_rb_threadUpdateObjCmd): passed improper flags to DoOneEvent().
-
-	* ext/tk/tkutil.c: use rb_obj_respond_to() instead of rb_respond_to().
-
-Tue Nov  1 14:20:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call_super): should call method_missing if super is
-	  called from Kernel method.
-
-	* eval.c (exec_under): frame during eval should preserve external
-	  information.
-
-Tue Nov  1 10:50:17 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: should check ERR_peek_last_error().
-	  [ruby-dev:27597]
-
-	* ext/openssl/ossl.c (ossl_raise): ditto.
-
-Mon Oct 31 17:34:46 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* configure.in: use proper option for Sun linker. A patch from
-	  Shinya Kuwamura <kuwa at labs.fujitsu.com>.  [ruby-dev:27603]
-
-Mon Oct 31 11:27:22 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/gdbm/test_gdbm.rb, test/sdbm/test_sdbm.rb (test_s_open_error):
-	  skip on Win32/DOS platforms.
-
-Mon Oct 31 05:49:23 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_cipher.c (ossl_cipher_update): input data must
-	  not be empty. [ruby-talk:161220]
-
-	* test/openssl/test_cipher.rb: add test for Cipher#update("").
-
-Mon Oct 31 05:37:20 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpservlet/cgihandler.rb
-	  (WEBrick::HTTPServlet::CGIHandler#do_GET): the value of Set-Cookie:
-	  header field should be splited into each cookie.  [ruby-Bugs:2199]
-
-	* lib/webrick/cookie.rb (WEBrick::Cookie.parse_set_cookie): new method
-	  to parse the value of Set-Cookie: header field.
-
-	* test/webrick/test_cookie.rb, test/webrick/test_cgi.rb,
-	  test/webrick/webrick.cgi: add some test for cookie.
-
-Mon Oct 31 03:19:36 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/readline/readline.c (readline_readline): type check.
-	  [ruby-core:6089]
-
-	* numeric.c (fix_rshift): RDoc fix.  [ruby-core:6351]
-
-	* util.h (strtod): add #undef for platforms defines strtod()
-	  macro.   [ruby-dev:27563]
-
-Mon Oct 31 02:35:59 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* test/ruby/test_float.rb (test_precision): test by assert_in_delta.
-	  [ruby-dev:27575]
-
-Sat Oct 29 01:58:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/etc/etc.c: document update from mathew <meta@pobox.com>.
-	  [ruby-core:06473]
-
-	* ext/fcntl/fcntl.c: ditto.
-
-Thu Oct 27 16:45:31 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (scan_once): wrong condition to use mbclen2().
-	  [ruby-dev:27535]
-
-Wed Oct 26 09:27:27 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/syck/implicit.c (syck_type_id_to_uri): should return
-	  newly allocated memory. otherwise, type_id will be freed
-	  twice. [ruby-dev:27384] [ruby-core:6385]
-
-Wed Oct 26 09:04:51 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ruby.h (Qfalse, Qtrue, Qnil, Qundef): make sure these immediate
-	  values have VALUE type. there is an environment where sizeof(VALUE)
-	  != sizeof(int) like IA64. if 32bit integer (Qtrue) is passed to ANYARGS
-	  and received by 64bit integer (VALUE), upper bits may have garbage value.
-	  [ruby-dev:27513]
-
-Wed Oct 26 01:58:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (RUBY_EXTERN): macro to export symbols in shared
-	  library.  [ruby-core:05528]
-
-	* defines.h, {bcc32,win32,wince}/Makefile.sub (RUBY_EXTERN): moved to
-	  configuration pass.
-
-	* ext/extmk.rb (extmake): RUBY_EXTERN for static linked extensions.
-
-Tue Oct 25 15:32:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/rational.rb: applied documentation patch from Gavin Sinclair
-	  <gsinclair@gmail.com>.  [ruby-core:06364]
-
-	* lib/irb.rb (IRB::Irb::eval_input): handle prompts with newlines
-	  in irb auto-indentation mode.  [ruby-core:06358]
-
-Tue Oct 25 02:12:08 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/rdoc/markup/simple_markup.rb (SM::SimpleMarkup::LABEL_LIST_RE):
-	  reduce redundant backtrack.  [ruby-talk:161771]
-
-Tue Oct 25 00:27:35 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/*: RDoc documentation from Eric Hodel
-	  <drbrain@segment7.net> added.
-
-Mon Oct 24 21:14:29 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in, io.c: use sys/syscall.h if syscall.h is not available.
-	  [ruby-core:06247]
-
-Mon Oct 24 20:49:45 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/Win32API/lib/win32/resolv.rb (get_info): support multiple DNS.
-	  fixed: [ruby-list:40058], [ruby-dev:27479]
-
-Mon Oct 24 07:57:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/tk/lib/tk/canvas.rb (TkCanvasItemConfig::__item_val2ruby_optkeys):
-	  typo fixed.  [ruby-talk:162187]
-
-	* ext/tk/lib/tk/menu.rb (TkMenuEntryConfig::__item_val2ruby_optkeys):
-	  ditto.  [ruby-core:06359]
-
-Sun Oct 23 21:50:15 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/enumerator/enumerator.c: applied documentation patch from
-	  James Edward Gray II <james@grayproductions.net>.
-	  [ruby-core:06348]
-
-Sun Oct 23 07:11:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/extconf.rb: improve messages [ruby-core:06325].
-
-	* ext/tk/lib/tk.rb, ext/tk/lib/tk/canvas.rb, ext/tk/lib/tk/entry.rb,
-	  ext/tk/lib/tk/frame.rb, ext/tk/lib/tk/image.rb,
-	  ext/tk/lib/tk/itemconfig.rb, ext/tk/lib/tk/labelframe.rb,
-	  ext/tk/lib/tk/listbox.rb, ext/tk/lib/tk/menu.rb,
-	  ext/tk/lib/tk/radiobutton.rb, ext/tk/lib/tk/scale.rb,
-	  ext/tk/lib/tk/spinbox.rb, ext/tk/lib/tk/text.rb,
-	  ext/tk/lib/tk/toplevel.rb: improve conversion of option values.
-
-	* ext/tk/lib/tkextlib/*: ditto.
-
-	* ext/tk/lib/tkextlib/*: update to support ActiveTcl8.4.11.2.
-
-	* ext/tk/lib/tkextlib/trofs/*: support Trofs 0.4.3.
-
-	* ext/tk/lib/tkextlib/tile/*: support Tile 0.7.2.
-
-	* ext/tk/lib/tkextlib/vu/*: support vu 2.3.0.
-
-	* ext/tk/lib/tkextlib/tcllib/*: support Tcllib 1.8 (Tklib 0.3).
-
-Sat Oct 22 23:54:07 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb, lib/mkmf.rb (with_config): support --with-extension
-	  options.  [ruby-dev:27449]
-
-Sat Oct 22 13:26:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* object.c (sym_inspect), parse.y (parser_yylex, rb_symname_p): check
-	  if valid as a symbol name more strictly.  [ruby-dev:27478]
-
-	* test/ruby/test_symbol.rb: tests for [ruby-core:03573].
-
-	* time.c (rb_strftime): removed meaningless volatile modifiers, and
-	  concatenate successive nul characters at once.  [ruby-dev:27472]
-
-Fri Oct 21 19:21:56 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* rubysig.h (CHECK_INTS): fixed typo. (I believe bit-or is improper)
-
-Fri Oct 21 17:49:32 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bin/erb (ERB::Main::run): typo fixed.  [ruby-core:06337]
-
-Fri Oct 21 15:27:17 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bignum.c (bignew_1): convertion from `int' to `char' discards
-	  upper bits, (ie. (char)0xff00 -> 0) so it's better to test if
-	  nonzero and set 0 or 1 instead of simply casting ... as a flag usage.
-	  (but I believe this won't cause actual bug in current implementation)
-	  [ruby-dev:27055]
-
-	* time.c: should use LONG_LONG instead of `long long'.
-
-Thu Oct 20 09:37:15 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/mkmf.rb (create_makefile): Borland make seems not to allow
-	  empty dependency list. If this change is not good, please correct
-	  it.
-
-Thu Oct 20 07:55:09 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): get rid of a restriction
-	  of Borland make.  fixed: [ruby-dev:27460]
-
-Thu Oct 20 00:13:18 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* rubysig.h (CHECK_INTS): fix typo.
-
-Wed Oct 19 23:58:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): do not create unnecessary empty
-	  directories.  fixed: [ruby-dev:27451]
-
-Wed Oct 19 19:26:15 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (rb_gc_mark_parser): get rid of segfault with old yacc.
-	  fixed: [ruby-dev:27439]
-
-Wed Oct 19 08:28:32 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_join): elements may contain null pointer strings.
-	  report and fixed by Lloyd Zusman (hippoman): [ruby-core:06326]
-
-Wed Oct 19 02:34:33 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c, gc.c, time.c: made internal symbols static.  [ruby-dev:27435]
-
-Wed Oct 19 01:27:07 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* regex.c (re_compile_pattern): numeric literal inside character class
-	  disabled succeeding backtrack.  fixed: [ruby-list:41328]
-
-Mon Oct 17 21:18:50 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (parser_heap): byacc never free parser stack.
-	  fixed: [ruby-dev:27428]
-
-Mon Oct 17 16:04:47 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* file.c (chmod_internal, lchmod_internal): fixed type of 2nd argument.
-
-Sun Oct 16 22:16:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: omit non-existing directories.
-
-Sun Oct 16 14:30:05 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/rinda.rb (Rinda::Tuple#initialize): check remote hash
-	  tuple. fixed: [ruby-list:41227]
-
-	* test/rinda/test_rinda.rb: test it.
-
-Sun Oct 16 03:38:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* rubysig.h (CHECK_INTS): prevent signal handler to run during
-	  critical section.  [ruby-core:04039]
-
-	* eval.c (load_wait): need not to call rb_thread_schedule()
-	  explicitly.  [ruby-core:04039]
-
-	* eval.c (rb_thread_schedule): clear rb_thread_critical.
-	  [ruby-core:04039]
-
-Sat Oct 15 19:56:38 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* bin/erb: typo fixed, again. thanks, Doug Kearns.
-
-Fri Oct 14 22:08:26 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (ioctl): should set errno.
-
-Fri Oct 14 16:57:32 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/config.rb (Config::FileHandler): :UserDir should be nil.
-	  It is harmful to permit the access to ~/public_html by default.
-	  suggested by Hiroyuki Iwatsuki.
-
-Thu Oct 13 23:29:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (HEAPCNT): bison allocates indivisible size.
-	  fixed: [ruby-core:06261]
-
-	* io.c, pack.c, ext/syck/rubyext.c, ext/syck/syck.h, missing/isinf.c:
-	  get rid of warnings.  fixed: [ruby-core:06247]
-
-Wed Oct 12 12:52:57 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.c (Init_openssl): should call
-	  OpenSSL_add_ssl_algorithms().
-
-Wed Oct 12 11:08:54 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* file.c (rb_f_test): typo in RDoc comments.
-
-Tue Oct 11 21:41:58 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_obj_respond_to): check if obj responds to the given
-	  method with the given visibility.  [ruby-dev:27408]
-
-	* eval.c (rb_respond_to): conform to Object#respond_to?.  [ruby-dev:27411]
-
-Tue Oct 11 00:01:21 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* st.c (st_free_table): do not call free() but xfree().
-	  [ruby-core:06205]
-
-Sat Oct  8 20:04:40 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (Init_Binding): add Binding#dup method.  [yarv-dev:666]
-
-	* parse.y (rb_parser_malloc, rb_parser_free): manage parser stack on
-	  heap.  [ruby-list:41199]
-
-	* ext/iconv/charset_alias.rb: parse config.charset_alias file directly.
-
-Fri Oct  7 09:54:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::Cookie::parse): Cookies from Nokia devices may
-	  not be parsed correctly.  A patch from August Z. Flatby
-	  (augustzf) in [ruby-Patches-2595].  [ruby-core:06183]
-
-Thu Oct  6 20:12:16 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/strscan/strscan.c (strscan_free): remove useless code.
-	  [ruby-dev:26368] [ruby-dev:27389]
-	  (backported from trunk, rev 1.22)
-
-Wed Oct  5 04:42:38 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/xmlrpc/server.rb (XMLRPC::Server#initialize): should mount the
-	  servlet on "/".
-
-Wed Oct  5 03:59:09 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/xmlrpc/server.rb (XMLRPC::Server#serve): delete wrong call
-	  of "join".
-
-Mon Oct  3 00:04:00 2005  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
-
-	* pack.c (EXTEND16): [ruby-dev:27383]
-
-Thu Sep 29 10:26:18 2005  Tanaka Akira  <akr@m17n.org>
-
-	* ext/dl/dl.c (rb_io_to_ptr): abolish sizeof(FILE).
-	  [ruby-dev:27317]
-
-Thu Sep 29 07:22:05 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* evalc. (rb_f_send): underscores need to be escaped.
-	  fixed by Doug Kearns.  [ruby-core:06053]
-
-Thu Sep 29 00:57:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (ev_const_get), variable.c (rb_const_get_0): retry only when
-	  autoload succeeded.
-
-	* variable.c (rb_autoload_load): now return true if autoload
-	  succeeded.  fixed: [ruby-dev:27331]
-
-Wed Sep 28 23:42:15 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (apply2files): add prototype.
-
-	* file.c (rb_stat_inspect): constified.
-
-	* class.c (rb_mod_init_copy, rb_class_init_copy), file.c (rb_stat_init_copy),
-	  numeric.c (num_init_copy), object.c (rb_obj_init_copy, Init_Object),
-	  re.c (match_init_copy, rb_reg_init_copy), time.c (time_init_copy):
-	  undocumented.
-
-Wed Sep 28 23:09:23 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/delegate.rb: document update from James Edward Gray II
-	  <james@grayproductions.net>.  [ruby-core:06027]
-
-Wed Sep 28 15:14:19 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (WEBrick::CGI#start): req.query_string should
-	  refer the value of QUERY_STRING. [ruby-list:41186]
-
-	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#query_string=):
-	  add new method.
-
-Wed Sep 28 10:45:44 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: cannot compile with Tcl/Tk8.0.x
-	  [ruby-dev:27335].
-
-Wed Sep 28 08:12:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (read_buffered_data): check if reached EOF.  fixed: [ruby-dev:27334]
-
-Wed Sep 28 07:56:52 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/yaml/basenode.rb (YAML::BaseNode::match_segment): fix typo.
-	  [ruby-dev:27237], [ruby-core:05854]
-
-	* lib/yaml/tag.rb (Module#yaml_as): suppress warnings.
-
-	* lib/yaml/types.rb (YAML::PrivateType, YAML::DomainType): ditto.
-
-Wed Sep 28 03:23:35 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* rubysig.h: fixed build problem with --enable-pthread on platforms
-	  which don't have setitimer().
-
-Mon Sep 26 22:32:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (set_trace_func): add rb_secure(4) to prevent adding
-	  tracing function.
-
-Sun Sep 25 12:05:10 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* bin/erb: typo fixed.
-
-Sun Sep 25 01:46:43 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-calculate-indent): arrange deep-indent
-	  closing parenthesis at same column as the opening.
-
-Sun Sep 25 00:42:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-expr-beg): deal with heredoc separately.
-	  fixed: [ruby-list:41168]
-
-	* misc/ruby-mode.el (ruby-calculate-indent): not to deepen indent
-	  level for continuous line inside parentheses.
-	  http://nabeken.tdiary.net/20050915.html#p02
-
-Sun Sep 25 00:18:11 2005  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (unknown_node): show more information.  [ruby-dev:26196]
-
-Sat Sep 24 08:56:01 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (cd): no longer accept :noop option, related
-	  code is useless (backported from trunk, rev 1.67).
-	  [ruby-core:05858] [ruby-Bugs:2494]
-
-Sat Sep 24 08:38:07 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb: fix visibility of FileUtils::NoWrite, Verbose,
-	  DryRun (backported from trunk, rev 1.66). [ruby-core:05954]
-
-	* test/fileutils/test_nowrite.rb: test it.
-
-	* test/fileutils/test_dryrun.rb: new file.
-
-	* test/fileutils/test_verbose.rb: new file.
-
-Sat Sep 24 02:40:20 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/delegate.rb: document update from James Edward Gray II
-	  <james@grayproductions.net>.  [ruby-core:05942]
-
-Thu Sep 22 23:36:24 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (find_executable0): default path if environment is not
-	  set.  [ruby-dev:27281]
-
-Thu Sep 22 16:33:12 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* test/readline/test_readline.rb (TestReadline::replace_stdio):
-	  merged the patch of [ruby-dev:25232] instead of [ruby-dev:25223].
-
-Wed Sep 21 23:30:44 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (configuration): generalized nmake dependent code.
-
-Wed Sep 21 09:07:55 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* stable version 1.8.3 released.
-
-Wed Sep 21 08:52:25 2005  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/token.c: correctly compute identation of a block
-	  scalar's parent node. [ruby-talk:150620]
-
-Wed Sep 21 08:20:24 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* README.EXT, README.EXT.ja: add new features.
-
-Wed Sep 21 07:43:58 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (default_argv, Arguable#options): defaults strings
-	  to be parsed to Arguable instance.
-
-Wed Sep 21 02:44:09 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* file.c (path_check_0): disallow sticky world writable directory
-	  in PATH (and $LOAD_PATH).  [ruby-dev:27226]
-
-	* file.c (fpath_check): typo fixed.
-
-Tue Sep 20 22:29:49 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/wsdl/simpletype/rpc/test_rpc.rb, test/wsdl/ref/test_ref.rb,
-	  test/wsdl/any/test_any.rb test/soap/wsdlDriver/test_calc.rb:
-	  suppress deliberate warnings with $VERBOSE = nil.
-
-Tue Sep 20 21:26:23 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/io/wait/lib/nonblock.rb: disable on platforms non-blocking flag
-	  is not available.  fixed: [ruby-dev:27187]
-
-Tue Sep 20 18:23:04 2005  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (thread_mark): mark th->last_status.  [ruby-dev:27179]
-
-Tue Sep 20 18:20:33 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/yaml.rb: require 'yaml/constants'.  [ruby-core:5776]
-
-Tue Sep 20 17:48:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/xmlrpc/client.rb (XMLRPC::Client::do_rpc): add charset
-	  information to content-type header.[ruby-core:5127]
-
-	* lib/xmlrpc/server.rb (CGIServer::serve): ditto.
-
-	* lib/xmlrpc/server.rb (ModRubyServer::serve): ditto.
-
-	* lib/xmlrpc/server.rb (WEBrickServlet::service): ditto.
-
-Tue Sep 20 17:34:46 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* test/webrick/test_cgi.rb: set ENV["PATH"] to CGIEnvPath on
-	  windows. bcc32's runtime is not installed into system directory,
-	  so it cannot be found without this setting. [ruby-dev:27166]
-
-Tue Sep 20 17:10:38 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* test/dbm/test_dbm.rb (TestDBM::test_s_open_error): remove
-	  test_s_open_error test to detect duplicate open.
-	  [ruby-dev:27202]
-
-Tue Sep 20 17:08:31 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* io.c: PIPE_BUF is not defined on BeOS. use _POSIX_PIPE_BUF instead.
-	  [ruby-dev:27185]
-
-Tue Sep 20 16:53:53 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* test/readline/test_readline.rb (TestReadline::replace_stdio):
-	  BSD seek support from [ruby-dev:25223].  fixed: [ruby-dev:27150]
-
-Tue Sep 20 15:39:40 2005  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/emitter.c (syck_scan_scalar): prevent indicators from
-	  appearing alone or at the end of plain scalars. [ruby-core:5826]
-
-	* ext/syck/emitter.c (syck_emit_scalar): treat typed scalar nodes
-	  as complex keys.
-
-	* lib/syck.h: version 0.60.
-
-	* lib/yaml/basenode.rb (YAML::BaseNode#at): transform keys during
-	  key searches.
-
-	* ext/syck/rubyext.c: loading of binary-typed nodes.  prevent
-	  emission of plain strings that look like symbols, but which aren't.
-
-Tue Sep 20 05:50:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* test/xmlrpc/test_webrick_server.rb (setup_http_server):
-	  should not include 'webrick/https' unless 'use_ssl' because
-	  it fails where openssl is not installed.
-
-Tue Sep 20 00:34:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_close): call rb_io_close() directly if io is a T_FILE
-	  object.  [ruby-dev:27156]
-
-Mon Sep 19 19:09:08 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* file.c (rb_file_chown): should accept nil. [ruby-dev:27171]
-	  (backport from trunk, rev 1.208)
-
-Mon Sep 19 18:35:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/dl/dl.c (rb_io_to_ptr): fix DragonFlyBSD support.
-	  [ruby-dev:27151]
-
-Mon Sep 19 14:17:04 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/syck/emitter.c (syck_emit): passing an int* value to the
-	  long* parameter causes unaligned access on LP64 systems.
-	  [ruby-dev:27161]
-
-Mon Sep 19 13:44:03 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c: avoid core dump with WIN32OLE_EVENT.
-	  [ruby-dev:27133]
-
-Mon Sep 19 10:36:06 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (cp_r): default is :dereference_root=>true for
-	  backward compatibility. [ruby-dev:27145]
-
-	* test/fileutils/test_fileutils.rb (test_cp_r): test it.
-
-Mon Sep 19 09:57:39 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb: backported from trunk (1.36).
-	  (again) [ruby-dev:27145]
-
-Mon Sep 19 07:45:37 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_pkey.h, ossl_pkey_rsa.c, ossl_pkey_dsa.c:
-	  an instance variable "private" is added to OpenSSL::PKey class.
-	  this ivar is a flag that shows whether there is a private key
-	  in the instance.
-
-	* ext/openssl/ossl_engine.c: (ossl_engine_load_privkey): set private
-	  key flag.
-
-Mon Sep 19 06:41:32 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb: backported from trunk (rev 1.65):
-
-	* lib/fileutils.rb (rm_r): new option :secure.
-
-	* lib/fileutils.rb (rm_rf): new option :secure.
-
-	* lib/fileutils.rb: new method #remove_entry_secure.
-
-	* lib/fileutils.rb (cd): remove option :noop.
-
-	* lib/fileutils.rb (cp_r): new option :dereference_root.
-
-	* lib/fileutils.rb (cp_r): new option :dereference_root.
-
-	* lib/fileutils.rb: new method #remove_entry.
-
-	* lib/fileutils.rb: new method #chmod_R.
-
-	* lib/fileutils.rb: new method #chown.
-
-	* lib/fileutils.rb: new method #chown_R.
-
-	* lib/fileutils.rb: new method .commands.
-
-	* lib/fileutils.rb: new method .options.
-
-	* lib/fileutils.rb: new method .have_option?.
-
-	* lib/fileutils.rb: new method .options_of.
-
-	* lib/fileutils.rb: new method .collect_method.
-
-	* lib/fileutils.rb: use module_function instead of single extend.
-
-	* test/fileutils/test_fileutils.rb: backported from trunk (1.36).
-
-Mon Sep 19 03:17:48 2005  Tanaka Akira  <akr@m17n.org>
-
-	* file.c (rb_thread_flock): wrap the flock system call by
-	  TRAP_BEG/TRAP_END to enable signals.  [ruby-dev:27122]
-
-	* ext/socket/socket.c (bsock_send): wrap the sendto and send system
-	  call by TRAP_BEG/TRAP_END to enable signals when writing to a socket
-	  which is full.  [ruby-dev:27132]
-
-	* io.c (rb_io_syswrite): wrap the write system call by
-	  TRAP_BEG/TRAP_END to enable signals when writing to a pipe which is
-	  full.  [ruby-dev:27134]
-
-Mon Sep 19 03:02:08 2005  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (io_fwrite): wrap the write system call by TRAP_BEG/TRAP_END to
-	  enable signals when writing to a pipe which is full.
-
-Sun Sep 18 02:10:47 2005  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml/rubytypes.rb: remove comments that are bungling up
-	  the rdoc and ri output.  output symbols as plain scalars.
-
-	* ext/syck/rubyext.c (syck_emitter_reset): emit headless
-	  documents always.
-
-	* ext/syck/emitter.c (syck_scan_scalar): quote scalars with any
-	  kind of surrounding line space, tabs or spaces alike.
-
-	* ext/syck/token.c: accept tabs as whitespace, not for indentation,
-	  but strip from plain scalars.
-
-	* test/yaml/test_yaml.rb: remove outdated tests.
-
-Sat Sep 17 23:25:04 2005  sheepman  <sheepman@sheepman.sakura.ne.jp>
-
-	* lib/mathn.rb (Rational::inspect): should preserve original
-	  operand.  [ruby-core:05806]
-
-Sat Sep 17 23:20:27 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::Cookie): should handle multiple values for a
-	  cookie name.  [ruby-talk:156140]
-
-Sat Sep 17 10:42:13 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string and bg_eval_string
-	  should call Kernel.eval on caller's safe-level instead of slave's
-	  safe-level (Of course, the given script should be evaluated on
-	  slave's safe-level).
-
-Sat Sep 17 09:45:26 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_substr): should propagate taintness even for
-	  empty strings.  [ruby-dev:27121]
-
-	* string.c (rb_str_aref): should infect result if range argument
-	  is tainted.  [ruby-dev:27121]
-
-Sat Sep 17 08:35:39 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/maker/base.rb (RSS::Maker::ItemsBase#normalize): fixed
-	strange RSS::Maker::Item#max_size behavior.
-	Thanks to Kazuhiko <kazuhiko@fdiary.net>.
-
-	* test/rss/test_maker_1.0.rb (RSS::TestMaker10#test_items): ditto.
-
-Fri Sep 16 23:09:20 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_search_event_at): bug fix
-	  in ext/win32ole/sample/ienavi.rb.
-
-	* ext/win32ole/win32ole/tests/testOLEEVENT.rb: ditto.
-
-Fri Sep 16 22:41:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_s_extname): empty string for path name ending with a
-	  period.  fixed: [ruby-core:05651]
-
-	* file.c (rb_file_join): smarter behavior at edge cases.
-	  fixed: [ruby-core:05706]
-
-Fri Sep 16 18:34:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/syck/node.c (syck_replace_str): was using return from the
-	  void function.  a patch from MIYAMUKO Katsuyuki
-	  <miyamuko at mtb.biglobe.ne.jp>.  [ruby-dev:27111]
-
-Fri Sep 16 14:48:48 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: fix typo on MultiTkIp#bg_eval_string
-
-Fri Sep 16 12:02:12 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/syck/rubyext.c (syck_resolver_transfer): remove C++ style
-	  comment (//).  [ruby-core:05793]
-
-Fri Sep 16 00:14:14 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/logger/test_logger.rb: unintentionally overwritten changes by
-	  Usa.  reverted.
-
-Fri Sep 16 00:06:18 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#initialize): should set
-	  $stdout.binmode.
-
-Thu Sep 15 23:25:21 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/{soap,wsdl,xsd}, test/{soap,wsdl,xsd}: imported soap4r/1.5.5.
-
-	  #nnn is a ticket number at http://dev.ctor.org/soap4r
-
-	  * SOAP
-
-	    * allow to configure an envelope namespace of SOAP request. (#124)
-	   	TemporaryNamespace = 'http://www.w3.org/2003/05/soap-envelope'
-	  	@client.options["soap.envelope.requestnamespace"] =
-		  TemporaryNamespace
-	 	@client.options["soap.envelope.responsenamespace"] =
-		  TemporaryNamespace
-		@client.do_proc(...)
-
-	    * let SOAP request XML indent space configuable.  see
-	      "soap.envelope.no_indent" option. (#130)
-
-	    * let external CES configuable.
-	      ex. client["soap.mapping.external_ces"] = 'SJIS'.  $KCODE is used
-	      by default. (#133)
-		external CES ::= CES used in Ruby object of client and server
-		internal CES ::= CES used in SOAP/OM
-
-	    * add iso-8859-1 external CES support. (#106)
-
-	    * fixed illegal 'qualified' handling of elements.  it caused
-	      ASP.NET inteoperability problem. (#144)
-
-	    * added 'soap.envelope.use_numeric_character_reference' (boolean)
-	      option to let query XML use numeric character reference in XML,
-	      not plain UTF-8 character.  !GoogleSearch server seems to not
-	      allow plain UTF-8 character since 2005-08-15 update. (#147)
-
-	    * SOAP::Header::SimpleHeader (de)serialization throws an exception
-	      on !SimpleHeader.on_(in|out)bound when header is a String.  so we
-	      could not use a simple single element headerItem.  fixed.  thanks
-	      to emil. (#129)
-
-	    * out parameter of rpc operation did not work.  (#132)
-
-	    * follow HTTP redirect only if using http-access2.  (#125) (#145)
-
-	    * add a workaround for importing an WSDL whose path begins with
-	      drive letter.  (#115)
-
-	  * WSDL
-
-	    * SOAP Data which is defined as a simpletype was not mapped
-	      correctly to Ruby obj when using wsdl2ruby.rb generated classdef
-	      file. (#123)
-
-	    * rpc/literal support. (#118)
-
-	    * re-implemented local element qualify/unqualify control.  handles
-	      elementFormDefault and form in WSDL.  (#119)
-
-	    * Array of an element which has simpleType causes a crash. (#128)
-
-	    * prarmeterOrder may not contain return part so it can be shorter
-	      than parts size.  Thanks to Hugh.  (#139)
-
-	  * Samples
-
-	    * added !BasicAuth client sample. (#117)
-
-	    * added Base64 client/server sample.
-
-	    * added Flickr SOAP interface client sample. (#122)
-
-	    * added !SalesForce client sample. (#135)
-
-	    * updated Thawte CA certificate for !GoogleAdWords sample.
-
-	    * updated a client script with the newer version made by Johan.
-	      thanks!
-
-	    * shortened long file names. (#120)
-
-	    * fixed typo in authheader sample. (#129)
-
-	    * updated deprecated method usage.  (#138)
-
-Thu Sep 15 23:02:57 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.h (rb_w32_stat): added prototype.
-
-Thu Sep 15 22:35:55 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/ruby/test_signal.rb (test_exit_action): skip the test using
-	  fork on fork-less platforms.
-
-Thu Sep 15 11:39:18 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/dialog.rb: If a dialog does not show up yet,
-	  TkDialogObj#name raises an exception. [ruby-talk:156109]
-
-Thu Sep 15 01:39:19 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/tuplespace.rb (Rinda::TemplateEntry::initialize): pull
-	  up method. Tabs converted to spaces.
-
-Thu Sep 15 00:18:24 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/net/telnet.rb (Net::Telnet::waitfor): replace sysread with
-	  readpartial.  [ruby-talk:127641]
-
-Wed Sep 14 22:40:26 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (ruby_glob): glob function not using ruby exception system.
-
-Wed Sep 14 01:26:03 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/https.rb: backported from trunk, rev 1.3.
-	  [ruby-dev:25673] (again), [ruby-dev:26617] (again),
-	  [ruby-dev:27062]
-
-	* ext/openssl/lib/net/https.rb: removed.
-
-	* ext/openssl/lib/net/protocols.rb: removed.
-
-	* lib/net/http.rb: #use_ssl?, #use_ssl are moved from net/https.
-
-Tue Sep 13 22:09:40 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb (Logger): added formatter accessor to logger for
-	  dictating the way in which the logger should format the messages it
-	  displays.  Thanks to Nicholas Seckar (cf. [ruby-talk:153391]) and
-	  Daniel Berger.
-
-	* lib/logger.rb (Logger): added VERSION constant.
-
-	* lib/logger.rb: removed document for LogDevice. It is an
-	  implementation detail and is not a public interface.
-
-	* test/logger/test_logger.rb: added tests.
-
-Tue Sep 13 21:47:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (BEGIN_CALLARGS): pop halfly pushed status.
-	  fixed: [ruby-dev:26881]
-
-Tue Sep 13 16:26:45 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: backported from trunk, rev 1.128.
-	  [ruby-dev:25673] [ruby-dev:26617]
-
-	* lib/net/protocol.rb: backported from trunk, rev 1.78.
-
-	* lib/net/protocol.rb: new method #old_open to support net/smtp
-	  and net/pop.
-
-	* lib/net/smtp.rb: use #old_open.
-
-	* lib/net/pop.rb: ditto.
-
-Tue Sep 13 12:33:05 2005  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml.rb: reworking YAML::Stream to use the new
-	  emitter.
-
-	* lib/yaml/stream.rb: ditto.
-
-	* lib/yaml/rubytypes.rb: added Object#yaml_new.
-
-	* lib/yaml/tag.rb: the tag_subclasses? method now
-	  shows up in the class.  allow taguri to be set using an accessor.
-	  continue support of Object#to_yaml_type.
-
-	* ext/syck/rubyext.c: new emitter code.  yaml_new and yaml_initialize
-	  get called, should they be present.  consolidated all the diaspora of internal
-	  node types into the family below YAML::Syck::Node -- Map,
-	  Seq, Scalar -- all of whom are SyckNode structs pointing to
-	  Ruby data.  moved Object#yaml_new into the node_import and made it the
-	  default behavior.  the target_class is always called wih yaml_new, prepended
-	  a parameter, which is the klass.  loaded nodes through GenericResolver show their style.
-	  new Resolver#tagurize converts type ids to taguris.
-
-	* ext/syck/implicit.re: were 'y' and 'n' seriously omitted??
-
-	* ext/syck/emitter.c: renovated emitter, walks the tree in advance.
-	  consolidated redundant block_styles struct into
-	  the scalar_style struct.  (this means loaded nodes can now
-	  be sent back to emitter and preserve at least its very basic
-	  formatting.)
-
-	* ext/syck/gram.c: headless documents of any kind allowed.
-
-	* ext/syck/node.c: new syck_replace_str methods and syck_empty_*
-	  methods for rewriting node contents, while keeping the ID
-	  and other setup info.  added syck_seq_assign.
-
-	* ext/syck/syck.h: reflect block_styles and new node functions.
-
-Mon Sep 12 20:53:06 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* test/openssl/test_pkcs7.rb (test_enveloped): skip this test
-	  to avoid a bug of PKCS7_enctypt() (only if ext/openssl is
-	  compiled with OpenSSL-0.9.7d or earlier versions).
-	  http://www.mail-archive.com/openssl-dev@openssl.org/msg17376.html
-
-Mon Sep 12 14:03:33 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* test/dbm/test_dbm.rb: remove locking test, which may not be
-	  supported on some platforms.  [ruby-dev:27030]
-
-Mon Sep 12 10:45:58 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/dl/dl.c (rb_io_to_ptr): merged a patch for DragonFly BSD
-	  from Takahiro Kambe <taca at back-street.net>.  [ruby-dev:27023]
-
-Sun Sep 11 22:05:51 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* bin/erb (ERB::Main#run): set ERB#filename so that it is used
-	  when reporting syntax/runtime errors. Tabs converted to spaces.
-
-Sat Sep 10 10:17:03 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_engine.c (ossl_engine_s_by_id):
-	  OpenSSL::Engine.by_id calls given block before calling
-	  ENGINE_init (block parameter is the return value of this method
-	  itself).  this functionality is useful to load dynamic shared
-	  engines. the following code is a sample of loading a key using
-	  OpenSC PKCS #11 module.
-
-		require "openssl"
-		pkcs11 = OpenSSL::Engine.by_id("dynamic"){|e|
-		  e.ctrl_cmd("SO_PATH", "/usr/lib/opensc/engine_pkcs11.so")
-		  e.ctrl_cmd("LIST_ADD", "1")
-		  e.ctrl_cmd("LOAD")
-		}
-		pkcs11.ctrl_cmd("PIN", "secret")
-		key = pkcs11.load_private_key
-
-	* ext/openssl/ossl_engine.c (ossl_engine_ctrl_cmd): new method
-	  OpenSSL::Engine#ctrl_cmd. it wraps ENGINE_ctrl_cmd_string.
-
-	* ext/openssl/ossl_engine.c (ossl_engine_get_cmds): new method
-	  OpenSSL::Engine#cmds. it returms engine command definitions.
-
-Sat Sep 10 10:09:47 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (asn1str_to_str): new function.
-
-	* ext/openssl/ossl_pkcs7.c: new class OpenSSL::PKCS7::RecipientInfo.
-	  this class wraps PKCS7_RECIP_INFO struct.
-
-	* ext/openssl/ossl_pkcs7.c: OpenSSL::PKCS7::Signer is renamed to
-	  OpenSSL::PKCS7::SignerInfo. ("Signer" remains as an alias of
-	  SignerInfo.)
-
-	* test/openssl/test_pkcs7.rb: new file.
-
-Sat Sep 10 10:05:51 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ns_spki.c (ossl_spki_initialize): assume that
-	  the argument is a DER string if Base64 decoding failed.
-
-	* ext/openssl/ossl_ns_pki.c (ossl_spki_to_der): new method.
-
-	* test/openssl/test_ns_spki.rb: add new file.
-
-Sat Sep 10 09:56:24 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/digest.rb: added SHA224, SHA256, SHA384 and SHA512.
-	  these features are enabled if this library is compiled with
-	  OpenSSL 0.9.8 or later.
-
-	* test/openssl/test_digest.rb: add test for new digests.
-
-Sat Sep 10 09:51:30 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.c (ossl_raise): should use ERR_peek_last_error
-	  to get last error on the current thread. And should report
-	  errors on the stack while OpenSSL.debug is true.
-
-	* ext/openssl/ossl.c (ossl_get_errors): new method for debugging
-	  this library.
-
-	* ext/openssl/ossl_ssl.c (ossl_sslctx_set_ciphers): fix error message.
-
-	* ext/openssl/ossl_x509req.c (ossl_x509req_set_attributes): get rid
-	  of unused variable.
-
-	* ext/openssl/ossl_x509store.c (ossl_x509store_initialize): should
-	  set @time to avoid warning.
-
-	* ext/openssl/ossl_x509store.c (ossl_x509store_set_default_paths,
-	  X509_STORE_add_cert, X509_STORE_add_crl): should raise error if
-	  wrapped functions failed.
-
-	* test/openssl/test_x509store.rb: add test for errors.
-
-Fri Sep  9 22:13:19 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): prohibit calling tainted method (>2) when
-	  $SAFE == 0.
-
-Fri Sep  9 16:45:25 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (rb_str_times): make empty strings to keep taintness,
-	  and a little improvement.  [ruby-dev:26900]
-
-	* ext/iconv/iconv.c (iconv_try), ext/iconv/extconf.rb: get rid of meta
-	  characters in command line option.  fixed: [ruby-talk:155369]
-
-Thu Sep  8 14:58:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* merged a patch from Takahiro Kambe <taca at back-street.net> to
-	  support DragonFly BSD.  [ruby-dev:26984]
-
-Wed Sep  7 12:55:08 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb: abolish mod === tempfile to avoid a problem
-	  [ruby-dev:26967].
-
-Wed Sep  7 10:45:15 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_switch): convert all exceptions to
-	  SystemExit.  fixed: [ruby-core:05724]
-
-	* eval.c (rb_thread_terminated): show backtrace before propagate
-	  exceptions to main thread.
-
-Wed Sep  7 08:35:04 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* Makefile.in, configure.in (MINIOBJS): miniruby on HP-UX can not load
-	  extension libraries.
-
-	* bignum.c (bignew_1, bigadd): K&R style argument actually can't be
-	  defined as char.
-
-	* missing/vsnprintf.c: ANSI compiler supports const keyword.
-
-	* ext/digest/sha2/extconf.rb: reject platforms which has inttypes.h
-	  but no 64bit integer.
-
-	* lib/mkmf.rb (what_type?): guesstimate type.
-
-	* ext/etc/etc.c (setup_passwd), ext/etc/extconf.rb: pw_age might be
-	  char*.  fixed: [ruby-core:05470]
-
-Wed Sep  7 08:32:47 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (rb_mod_cvar_get, rb_mod_cvar_set): document fix from
-	  sheepman <sheepman@sheepman.sakura.ne.jp>; a bug in visibility
-	  description.  [ruby-dev:26965]
-
-	* sprintf.c (rb_f_sprintf): warn "too many argument" on verbose
-	  mode (-v/-w); backported from 1.9.  [ruby-dev:26963]
-
-Mon Sep  5 17:03:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/ostruct.rb: a patch from Florian Gross <florgro at gmail.com>
-	  merged to allow recursive inspect (and to_s) for OpenStruct.
-	  [ruby-core:05532]
-
-Mon Sep  5 07:01:12 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/openssl/lib/openssl/buffering.rb (Buffering#do_write):
-	  should clear data from the buffer which already been output.
-
-Fri Sep  2 23:51:54 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib: do not use __send__ to access private methods.  [ruby-dev:26935]
-
-Fri Sep  2 03:29:00 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
-
-	* lib/irb/init.rb: make IRB -I option that is same befavior for ruby.
-	  [ruby-dev:26872], [ruby-dev: 26920]
-
-	* lib/irb/locale.rb: support to print help message when OS locale is
-	  ja_JP.utf-8. [ruby-dev:26872]
-
-Thu Sep  1 17:11:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): wrong condition for $SAFE restoration.
-
-Thu Sep  1 14:12:45 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: On Tcl8.5, MultiTkIp#invoke_hidden doesn't
-	  work (gives wrong order of arguments).
-
-	* ext/tk/lib/multi-tk.rb: add MultiTkIp#invoke_hidden_on_namespace
-	  to support '-namespace' option of 'interp invokehidden' command
-	  on Tcl8.5.
-
-Wed Aug 31 14:43:15 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/Makefile.sub (OPTFLAGS): default global optimization to
-	  disabled for all VC++ versions.  fixed: [ruby-dev:26897]
-
-Wed Aug 31 11:35:43 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/gdbm/test_gdbm.rb (teardown): should remove GDBM temporary
-	  file.
-
-Wed Aug 31 10:30:56 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* process.c (proc_detach, proc_setmaxgroups): missing argument type
-	  declaration. (I recommend ANSI-style function)
-
-Tue Aug 30 23:20:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_rescue2): initialization miss.  fixed: [ruby-dev:26917]
-
-	* lib/mkmf.rb (xsystem, xpopen): no longer expand by Config.
-
-	* lib/mkmf.rb (link_command, cc_command, cpp_command): expand
-	  variables at once, and quote hdrdir.  fixed: [ruby-core:05680]
-
-	* lib/mkmf.rb (libpathflag): quote paths.
-
-Tue Aug 30 19:34:27 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/digest/md5/md5ossl.h, ext/digest/rmd160/rmd160ossl.h,
-	  ext/digest/sha1/sha1ossl.h: include <stddef.h> to avoid
-	  error in compilation with OpenSSL-0.9.8. [ruby-list:41068]
-
-Mon Aug 29 19:54:21 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/rdoc/usage.rb: improper exceptions. [ruby-dev:26870]
-
-	* lib/rdoc/usage.rb: support the case when non-ruby code exists before
-	  shebang. (this is needed when ri.bat is executed on windows)
-
-Mon Aug 29 17:48:17 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (method_arity): should return proper arity value.
-	  [ruby-dev:26390]
-
-Mon Aug 29 01:19:57 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/time.rb (Time.parse): extract fractional seconds using
-	  Date._parse.  [ruby-talk:153859]
-
-Sat Aug 27 20:20:01 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/curses/curses.c ({curses,window}_clrtoeol): added. suggested
-	  by Reyn Vlietstra.
-
-	* ext/curses/curses.c: chtype in curses is not `char', rahter `long'.
-	  [ruby-Bugs:2298]
-
-	* ext/curses/view.rb: String =~ String is deprecated.
-
-Wed Aug 24 10:53:28 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/logger/test_logger.rb (test_shifting_size): should close log
-	  device before unlink, since some platform cannot unlink opened
-	  file.
-
-Sun Aug 21 00:13:27 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/wsdl/xmlSchema/importer.rb (WSDL::XMLSchema::Importer#fetch): add
-	  a workaround for importing an WSDL whose path begins with drive
-	  letter.  [ruby-dev:26242]
-
-Sat Aug 20 22:37:13 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb (write, shift_log?, shift_log): file shifting race
-	  condition bug fixed.  [ruby-dev:26764]
-
-	* test/logger/test_logger.rb: tests.
-
-Fri Aug 19 18:13:39 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/time.rb (Time.apply_offset): fix a problem with last day of
-	  month.  reported by Lucas Nussbaum.  [ruby-talk:152866]
-
-Thu Aug 18 12:46:28 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub (COMMON_HEADERS): reverted 1.42.2.24.
-	  I misunderstood, bccwin32 on ruby_1_8 uses winsock2 originally.
-	  [ruby-dev:26806]
-
-	* win32/win32.h: include winsock2.h instead of winsock.h. (bcc32)
-
-Wed Aug 17 23:58:05 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* object.c (rb_to_integer): argument constified.
-
-	* eval.c (terminate_process): take String message.
-
-	* eval.c (rb_thread_switch): propagate the exception caused thread
-	  termination directly.  fixed: [ruby-core:05552]
-
-Wed Aug 17 00:05:46 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_add_method): preserve safe level in the environment
-	  where a method is defined .
-
-	* eval.c (rb_call0): restore preserved safe level in the method
-	  execution.
-
-Mon Aug 15 00:38:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_rescue2): reduce PUSH_TAG() as well as NODE_RESCUE.
-	  [ruby-dev:26800]
-
-	* range.c (range_check, range_init): reduce useless exceptions.
-
-Sat Aug 13 18:51:26 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_block_pass): distinguish current block from others.
-	  fixed: [ruby-dev:26274]
-
-	* ext/stringio/stringio.c (strio_set_string): disallow nil.
-	  http://www.rubyist.net/~nobu/t/20050811.html#c05
-
-Thu Aug 11 23:29:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c: keep holding string after closed.
-
-Thu Aug 11 13:01:48 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss: fixed sort bug. [ruby-list:41018]
-
-	* lib/rss/1.0.rb (RSS::RDF::Channel#setup_maker_attributes):
-	  removed self.
-
-	* lib/rss/maker/base.rb (RSS::Maker::ItemsBase#<=>): use #date
-	  instead of @date.
-	  (RSS::Maker::Base::self.def_array_element): added #size.
-
-	* lib/rss/maker/1.0.rb
-	  (RSS::Maker::RSS10::Channel#to_rss,
-	   RSS::Maker::RSS10::Items::Item#to_rss): cleared dc_dates set
-	  upped by using #date.
-
-	* lib/rss/maker/dublincore.rb
-	  (RSS::Maker::ChannelBase, RSS::Maker::ItemsBase::ItemBase):
-	  fixed opposite alias.
-
-	* test/rss/test_setup_maker_1.0.rb
-	  (RSS::TestSetupMaker10::test_setup_maker_items_sort): added some
-	  tests for RSS::Maker::ItemsBase#do_sort.
-
-Wed Aug 10 10:29:40 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: fix bug on handling __ruby2val_optkeys().
-
-	* ext/tk/lib/tk/itemconfig.rb: fix bug on handling
-	  __item_ruby2val_optkeys().
-
-	* ext/tk/lib/tk/canvas.rb: didn't check __item_ruby2val_optkeys().
-
-	* ext/tk/lib/tkextlib/blt/component.rb: ditto.
-
-Tue Aug  9 15:12:04 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: remove dangerous 'rb_jump_tag's.
-
-	* ext/tk/lib/tk.rb: add __val2ruby_optkeys and __ruby2val_optkeys to
-	  help to convert option values between ruby and tcl.
-
-	* ext/tk/lib/tk/itemconfig.rb: add __item_val2ruby_optkeys and
-	  __item_ruby2val_optkeys to help to convert option values between
-	    ruby and tcl.
-
-	* ext/tk/lib/tk/radiobutton.rb: use __ruby2val_optkeys for 'variable'
-	  option (for the reason of backward compatibility).
-
-	* ext/tk/lib/tk/composite.rb: clarify the arguments of super().
-
-	* ext/tk/lib/tk/spinbox.rb: ditto.
-
-	* ext/tk/lib/tk/text.rb: ditto.
-
-	* ext/tk/lib/tk/validation.rb: ditto.
-
-	* ext/tk/lib/tkextlib/*: support to treat tkvariable-type
-	  configure options.
-
-Tue Aug  9 20:30:19 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
-
-	* bignum.c (rb_big_coerce): allow bignum x bignum coercing.
-	  [ruby-dev:26778]
-
-Mon Aug  8 20:43:02 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_method.rb: added.  [ruby-dev:26761]
-
-Sun Aug  7 23:50:14 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_super.rb: added from HEAD. [ruby-dev:26743]
-
-Sun Aug  7 01:31:15 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (WIN32OLE_EVENT#on_event): should set
-	  only one event handler.
-
-	* ext/win32ole/tests/testOLEEVENT.rb: ditto.
-
-	* ext/win32ole/tests/testOLEPARAM.rb: remove re-defined
-	  test_ole_type_detail method.
-
-Sat Aug  6 12:35:24 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/{tk.rb,tk/itemconfig.rb}: configure creates
-	  TkVariable if key name is 'variable' or 'textvariable'
-	  by default. [ruby-dev:26749]
-
-	* ext/tk/lib/tk/{label,radiobutton}.rb: removed its own
-	  {variable,textvariable} function.
-
-	* ext/tk/lib/tk/variable.rb: retains backward conpatibility.
-
-Fri Aug  5 12:50:32 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c: fixed memory leak when tk_funcall raised
-	  exception. (copies argv into heap in tk_funcall instead of
-	  caller)
-
-Fri Aug  5 12:42:57 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): need to convert path separetor
-	  before invoking install command.
-
-Fri Aug  5 00:27:04 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c: refactoring - extract ruby string <->
-	  tcl object conversion as get_str_from_obj and get_obj_from_str.
-
-Fri Aug  5 00:19:33 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* extmk.rb (extmake): needs to be wrapped in an Array.
-
-Thu Aug  4 18:38:36 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: cannot compile for Tcl7.6/Tk4.2.
-
-	* ext/tcltklib/tcltklib.c: add nativethread consistency check.
-
-	* ext/tcltklib/stubs.c: ditto.
-
-	* ext/tk/lib/tk.rb: forgot to define TclTkIp.encoding and encoding=
-	  when Tcl is 7.6 or 8.0.
-
-	* ext/tk/lib/tk/wm.rb: support to make some methods as options of
-	  root or toplevel widget. [ruby-talk:150336]
-
-	* ext/tk/lib/tk/root.rb: ditto.
-
-	* ext/tk/lib/tk/toplevel.rb: ditto.
-
-	* ext/tk/lib/tkextlib/SUPPRT_STATUS: update RELEASE_DATE
-
-Thu Aug  4 08:03:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): should not modify $mflags for each
-	  extentions.
-
-Thu Aug  4 00:25:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* common.mk, Makefile.in, {bcc32,win32,wince}/Makefile.sub: integrated
-	  macro definitions.
-
-	* bcc32/Makefile.sub: LIBRUBY_SO should use DLDOBJS, not EXTOBJS.
-
-	* {win32,wince}/Makefile.sub: separate config.h for compiler versions.
-
-Wed Aug  3 21:59:16 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/variable.rb: TkVariable#trace didn't work on
-	  TkVariable retrived from TkVariable.new_hash.ref. [ruby-dev:26721]
-
-Wed Aug  3 08:22:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (ruby_connect): revert [ruby-talk:111654]
-	  changes at 2004-09-07.  [ruby-dev:26656]
-
-Tue Aug  2 10:20:54 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c: use Tcl_[GS]etVar2Ex instead of
-	  Tcl_Obj[GS]etVar2. (avoid Tcl_NewStringObj on supported platforms)
-
-	* ext/tcltklib/tcltklib.c: use ip_{get,set,unset}_variable2_core from
-	  ip_{get,set,unset}_variable.
-
-	* ext/tcltklib/tcltklib.c: replaced Tcl_Panic with rb_bug.
-
-Tue Aug  2 01:41:28 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/ping.rb (Ping.pingecho): should rescue StandardError.
-	  [ruby-dev:26677]
-
-Mon Aug  1 19:09:41 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c: refactoring - replaced rb_ivar_defined &
-	  rb_ivar_get with single rb_attr_get call.
-
-Mon Aug  1 18:45:07 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c (Tcl_GetStringResult): refactoring - define
-	  alternative macro on Tcl7.x or earlier.
-
-Mon Aug  1 13:57:35 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c (deleted_ip): refactoring - interpreter
-	  deletion check. [ruby-dev:26664]
-
-Mon Aug  1 01:17:40 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb (check_insecure_method): use private_methods and
-	  protected_methods instead of respond_to? to check method visibility.
-	  [ruby-dev:26616]
-
-	* test/drb/drbtest.rb: ditto.
-
-	* test/drb/ut_drb.rb: ditto.
-
-Mon Aug  1 00:07:32 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
-	* lib/irb/context.rb: fix `irb --readline` option. [ruby-list:40955]
-
-Fri Jul 29 09:59:38 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_yield_0): push yielded node instead of yielding.
-	  fixed: [yarv-dev:549]
-
-Thu Jul 28 18:09:55 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/stubs.c: When --enable-tcltk-stubs, the initialize
-	  routine creates a Tcl/Tk interpreter and deletes it. However,
-	  init cost of Tk's MainWindow is not so small. And that makes it
-	  impossible to use libraries written with Tcl functions only on
-	  an environment without a graphical display. This changes support
-	  delaying initalization of Tk_Stubs until the script needs Tk.
-
-	* ext/tcltklib/stubs.h: New file. Define prototypes and return
-	  codes of functions on stubs.c.
-
-	* ext/tcltklib/tcltklib.c: Support delaying initalization of
-	  Tk_Stubs until the script needs Tk.
-
-	* ext/tcltklib/tcltklib.c: Show friendly error messages for errors
-	  on initialization.
-
-	* ext/tcltklib/tcltklib.c: Avoid SEGV on ip_finalize() when ruby is
-	  exiting and $DEBUG is true. (Not fix. If you know the reason of
-	  why, please fix it.)
-
-	* ext/tk/tkutil.c (ary2list, ary2list2): bug fix on handling of
-	  encoding.
-
-	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string and bg_eval_string
-	  don't work propery.
-
-	* ext/tk/lib/tk.rb: Forget extending Tk::Encoding module to Tk.
-	* ext/tk/lib/tk/variable.rb: TkVarAccess fails to initialize the
-	  object for an element of a Tcl's array variable.
-
-Wed Jul 27 23:23:54 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (obj_free): make message format consistent with one from
-	  gc_mark().  [ruby-talk:149668]
-
-Wed Jul 27 22:11:37 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* sample/rss/tdiary_plugin: removed. because the plugin
-	  is imported in the tDiary plugin packages.
-
-Wed Jul 27 10:59:02 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (dir_each): rewinddir(3) before iteration.
-	  [ruby-talk:149628]
-
-Tue Jul 26 12:57:49 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/openssl_missin.c: include <openssl/engine.h> before
-	  <openssl/x509_vfy.h> to avoid compilation error of mswin32.
-	  suggested by NAKAMURA Usaku.
-
-Mon Jul 25 21:30:46 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* {bcc32,win32,wince}/Makefile.sub: moved CPPFLAGS only for ruby
-	  source to XCFLAGS.
-
-Mon Jul 25 13:45:18 2005  NAJIMA Hiroki  <najima@mickey.ai.kyutech.ac.jp>
-
-	* io.c: check HAVE_SYS_IOCTL_H before including the header.
-	  [ruby-dev:26610]
-
-Mon Jul 25 14:10:02 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: fix en-bugged part in the last commit.
-
-Sat Jul 23 16:49:04 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_engine.c (ossl_engine_s_load): should check
-	  OPENSSL_NO_STATIC_ENGINE.
-
-Fri Jul 22 21:06:08 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
-
-	* bignum.c (rb_big_eq): reduce isnan().  [ruby-dev:26600]
-
-	* numeric.c (flo_eq, flo_gt, flo_ge, flo_lt, flo_le): ditto.
-
-Fri Jul 22 15:02:39 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: moved copyright description to lib/rss.rb.
-
-	* lib/rss.rb: added for convenience.
-
-	* sample/rss/re_read.rb: added #to_s sample.
-
-	* sample/rss/blend.rb: use 'require "rss"' instead of
-	  'require "rss/*"'.
-	* sample/rss/list_description.rb: ditto.
-	* sample/rss/rss_recent.rb: ditto.
-	* sample/rss/tdiary-plugin/rss-recent.rb: ditto.
-
-	* sample/rss/tdiary-plugin/rss-recent.rb: 0.0.6 -> 0.0.7.
-
-Fri Jul 22 14:37:43 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/parser.rb (RSS::Parser#initialize): accept HTTP/FTP
-	  URI and local file path too.
-
-	* test/rss/test_parser.rb (RSS::TestParser#test_parse): test
-	  for the above.
-
-Fri Jul 22 07:01:42 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c (tk_conv_args): forget to revert thread_critical
-	  and gc_disable when raise ArgumentError.
-
-	* ext/tk/lib/remote-tk.rb: RemoteTkIp doesn't need to include TkUtil.
-
-	* ext/tcltklib/tcltklib.c: add TclTkIp#has_mainwindow? method.
-
-	* ext/tk/lib/tk.rb: add Tk.has_mainwindow? method.
-
-	* ext/tk/lib/multi-tk.rb: add MultiTkIp#has_mainwindow? method.
-
-	* ext/tk/lib/remote-tk.rb: add RemoteTkIp#has_mainwindow? method.
-
-	* ext/tk/lib/multi-tk.rb: slave IP fail to exit itself when $SAFE==4.
-
-	* ext/tk/lib/multi-tk.rb: remove constants from MultiTkIp module to
-	  avoid access from external.
-
-	* ext/tk/lib/multi-tk.rb: check_root flag is ignored on slave IPs'
-	  mainloop.
-
-	* ext/tk/lib/multi-tk.rb: hang-up Tk.mainloop called on a slave IP
-	  with $SAFE==4.
-
-	* ext/tk/lib/multi-tk.rb: MultiTkIp#bg_eval_proc doesn't work
-	  properly.
-
-	* ext/tk/lib/multi-tk.rb: add MultiTkIp#set_cb_error(proc) and
-	  cb_error(exc) to log errors at callbacks on safe slave IPs.
-
-	* ext/tk/lib/multi-tk.rb: fail to get an available slave IP object
-	  when call Tk.mainloop in the block which is given to new_* method,
-	    because cannot finish initialize while the root widget is alive.
-
-	* ext/tk/lib/multi-tk.rb: fail to control a slave IP when Tk.mainloop
-	  runs on the IP.
-
-Wed Jul 20 19:20:37 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* io.c (S_ISREG): need to define S_ISREG before it is used first.
-
-Wed Jul 20 18:40:50 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* io.c (wsplit_p): patch for the environment where
-	  fcntl(F_GETFL, O_NONBLOCK) is not supported. in that case,
-	  set FMODE_WSPLIT without fcntl check. [ruby-dev:26566]
-
-Wed Jul 20 18:07:11 2005  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (rb_io_ctl): update FMODE_WSPLIT_INITIALIZED and FMODE_WSPLIT
-	  by F_SETFL.
-
-Wed Jul 20 10:04:51 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* variable.c (rb_class_path): need to adjust snprintf() len for
-	  teminating NUL.  [ruby-dev:26581]
-
-Wed Jul 20 04:01:55 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/socket/socket.c: sorry, BeOS also uses HAVE_CLOSESOCKET,
-	  so reverted.
-
-	* ext/socket/extconf.rb: should not define HAVE_CLOSESOCKET
-	  on windows.
-
-Wed Jul 20 03:16:43 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/socket/socket.c: should not undef close() on win32.
-	  it's defined to rb_w32_close(), otherwise handle leaks.
-	  [ruby-Bugs-2131]
-
-Wed Jul 20 00:48:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* error.c (syserr_initialize): don't use str before StringValue()
-	  check.  [ruby-dev:26579]
-
-Tue Jul 19 22:47:29 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* error.c (syserr_initialize): add 1 byte for snprintf() size for
-	  NUL at the end.  [ruby-dev:26574]
-
-Tue Jul 19 16:39:46 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_inspect): replace sprintf() with "%s" format all
-	  over the place by snprintf() to avoid integer overflow.
-
-Tue Jul 19 14:08:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c: rbtk_eventloop_depth is used as int.
-
-	* ext/tcltklib/tcltklib.c: rbtk_pending_exception is tested with
-	  NIL_P, so should assign Qnil instead of 0 (Qfalse).
-
-	* ext/tcltklib/tcltklib.c (ip_invoke_real): fixed memory leak when
-	  ip is deleted.
-
-Tue Jul 19 13:19:46 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/variable.rb: For symmetry, add TkVariable#string. It
-	  returns a string even if the default value type of the TkVariable
-	  object is not "string".
-
-Mon Jul 18 21:40:20 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* eval.c (rb_call0): make the pointer to NODE volatile
-	  instead of NODE itself.
-
-Mon Jul 18 14:32:21 2005  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (rb_call0): make body volatile to avoid optimization problem.
-	  [ruby-dev:26195]
-
-Mon Jul 18 12:23:27 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/io/wait/wait.c: wrong backport from trunk.  fixed: [ruby-dev:26562]
-
-Mon Jul 18 09:36:25 2005  Tanaka Akira  <akr@m17n.org>
-
-	* rubyio.h (FMODE_WSPLIT, FMODE_WSPLIT_INITIALIZED): new constant.
-
-	* io.c (wsplit_p): new function.
-	  (io_fwrite): split writing data by PIPE_BUF if wsplit_p is true in
-	  multi-threaded mode.
-	  [ruby-dev:26540]
-
-Sun Jul 17 13:46:54 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/io/wait/extconf.rb, ext/io/wait/wait.c: Win32 platforms support.
-
-Fri Jul 15 23:59:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/rdoc/parsers/parse_c.rb (handle_class_module): handle a
-	  module enclosed in a built-in module.  fixed: [ruby-talk:148239]
-
-	* lib/rdoc/parsers/parse_c.rb (find_body): allow macros as methods.
-
-	* lib/rdoc/parsers/parse_c.rb (find_call_seq): allow :nodoc: modifier
-	  in C.  [ruby-core:04572]
-
-Fri Jul 15 18:00:01 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub (COMMON_HEADERS): ruby_1_8 is using winsock.h.
-	  failed to compile ext/socket on bcc5.6.4. [ruby-dev:26193]
-
-Fri Jul 15 07:58:56 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/server.rb (WEBrick::GenericServer#accept_client):
-	  sockets should be non-blocking mode. [ruby-dev:26405]
-
-	* lib/webrick/utils.rb (WEBrick::Utils.set_non_blocking): new method.
-
-	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#read_chunked):
-	  should call sock.read repeatedly until the preferred size data
-	  is obtained.
-
-Thu Jul 14 18:27:16 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (rb_w32_strerror): should return correct message
-	  for ENAMETOOLONG and ENOTEMPTY. (bcc32) [ruby-dev:26533]
-
-	* win32/win32.c (rb_w32_strerror): stripped CR LF on the tail.
-	  (bcc32) [ruby-dev:26533]
-
-Thu Jul 14 00:45:42 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* LEGAL (ext/nkf/nkf-utf8): updated from nkf1.7 to nkf-utf8.
-
-Wed Jul 13 19:37:47 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (rb_w32_mkdir): should set EEXIST (not EACCES)
-	  if file or directory already exists. (bcc32) [ruby-dev:26508]
-
-	* win32/win32.c (rb_w32_rmdir): should set ENOTDIR (not EINVAL)
-	  if it is not directory. (bcc32, win32)
-
-	* win32/win32.c (rb_w32_rmdir, rb_w32_unlink): restore
-	  FILE_ATTRIBUTE_READONLY flag on function failure.
-
-Wed Jul 13 12:40:00 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: TclTkLib.do_one_event doesn't work.
-
-	* ext/tk/lib/tk.rb: Tk.thread_update is available.
-
-Tue Jul 12 23:32:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb: keep curdir unexpanded.
-
-Mon Jul 11 08:31:29 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* regex.c (read_special): fix parsing backslashes following \c in
-	  regexp.  fixed: [ruby-dev:26500]
-
-Mon Jul 11 02:53:00 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#request_line):
-	  mistook in merging the patch of [ruby-dev:26235] at
-	  revision 1.4.2.6.
-
-Sun Jul 10 23:58:04 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#unlink): try Dir.unlink first to
-	  avoid unlink a directory by root.
-	  cf. [ruby-dev:26237]
-
-Sun Jul 11 05:18:17 2005  Michael Neumann  <mneumann@ruby-lang.org>
-
-	* lib/xmlrpc/server.rb (XMLRPC::Server): Switch from GServer over to
-	  WEBrick. This makes file lib/xmlrpc/httpserver.rb obsolete (at least it is
-		no further used by the XML-RPC library).
-
-Sun Jul 10 12:47:01 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/debug.rb (debug_command): added a deficient format specifier.
-	  fixed: [ruby-core:05419]
-
-Sat Jul  9 21:28:46 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_method_dispid): convert dispid
-	  in Ruby and C by INT2NUM and NUM2INT.
-
-	* ext/win32ole/win32ole.c (ole_invoke2): ditto.
-
-	* ext/win32ole/test/testWIN32OLE.rb: ditto.
-
-	* ext/win32ole/test/testOLEMETHOD.rb: ditto.
-
-Fri Jul  8 15:45:04 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb (RSS::VERSION): 0.1.4 -> 0.1.5.
-
-	* test/rss/test_version.rb (RSS::TestVersion#test_version):
-	  ditto.
-
-	* lib/rss/0.9.rb (RSS::Rss::Channel::Item::Category):
-	  domain attribute of <category> is optional. Thanks to
-	  Chris Lee <clee@kde.org>.
-
-	* test/rss/test_parser.rb (RSS::TestParser#test_category20):
-	  adjusted test case.
-
-Tue Jul  5 23:44:06 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* instruby.rb: expand source library path.
-
-Tue Jul  5 23:27:14 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* array.c (sort_2): get rid of yet another bcc's bug.
-	  fixed: [ruby-core:05152]
-
-	* eval.c (rb_thread_save_context): must not switch contexts during
-	  re-allocating stack.  fixed: [ruby-core:05219]
-
-Tue Jul  5 15:15:10 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c: fix typo.
-
-Tue Jul  5 14:51:35 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: bug fix on treating Unicode strings.
-
-	* ext/tcltklib/tcltklib.c: add methods to treat encoding mode.
-
-	* ext/tcltklib/MANUAL.eng: add description of TclTkLib#encoding,
-	  encoding_system, and so on.
-
-	* ext/tcltklib/MANUAL.euc: ditto.
-
-	* ext/tk/tkutil.c: fail to create a Tcl's list string from an
-	  array including multiple kind of encoded strings.
-
-	* ext/tk/lib/tk.rb: ditto.
-
-	* ext/tk/lib/multi-tk.rb: 2nd arg of _{to|from}UTF8 is omissible.
-
-	* ext/tk/lib/remote-tk.rb: ditto.
-
-	* ext/tk/lib/tk.rb: override TclTkLib#encoding and encoding= to
-	  use TkCore::INTERP.encoding and encoding=.
-
-	* ext/tk/lib/tk.rb: when "require 'tk'" and $KCODE=='NONE', check
-	  DEFAULT_TK_ENCODING to decide Ruby/Tk's system encoding mode.
-
-	* ext/tk/lib/tk/encodedstr.rb: check both of Tk.encoding and
-	  Tk.encoding_system. Tk.encoding has higher priority.
-
-	* ext/tk/lib/tk/optiondb.rb: ditto.
-
-	* ext/tk/lib/tk/spinbox.rb: ditto.
-
-	* ext/tk/lib/tk/validation.rb: ditto.
-
-	* ext/tk/lib/tk/namespace.rb: arguemnts for TclTkIp#_merge_tklist
-	  should be UTF-8 strings.
-
-Mon Jul  4 14:35:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* sample/svr.rb: service can be stopped by ill-behaved client; use
-	  tsvr.rb instead.
-
-Mon Jul  4 13:25:21 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* missing/erf.c: original erf.c by prof. Okumura is confirmed to
-	  be public domain.  reverted BSD implementation.
-
-Mon Jul  4 11:15:37 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/{dbm,gdbm,sdbm}/test_{dbm,gdbm,sdbm}.rb: skip some tests
-	  which using fork on fork-less platforms.
-
-Sun Jul  3 23:26:30 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/wsdl/document/test_rpc.rb: compare formatted time string of
-	  Time objects instead of comparing Time objects itself to avoid
-	  unintended conflict of usec part.  [ruby-dev:26220]
-
-Sat Jul  2 22:41:04 2005  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/socket.c (unix_send_io, unix_recv_io): support x86-64 and
-	  IA64.
-
-Sat Jul  2 17:06:23 2005  Tanaka Akira  <akr@m17n.org>
-
-	* defines.h (FLUSH_REGISTER_WINDOWS): defined for IA64.
-	  (flush_register_windows): declare flush_register_windows.
-
-	* eval.c (flush_register_windows): new function.
-
-Fri Jul  1 17:48:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (get2comp): revert all prior changes, and calculate
-	  proper 2's complement for negative numbers.  backported from
-	  HEAD.
-
-Fri Jul  1 15:50:12 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* missing/erf.c: need to include some headers for some platforms.
-
-	* win32/win32.h (copysign, scalb): define for compatibility with
-	  other platforms. [ruby-dev:26430]
-
-Fri Jul  1 15:37:42 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* missing/crypt.c: modified to make it compilable on platforms
-	  other than BSD.  [ruby-dev:26430]
-
-	* missing/erf.c: ditto.  code from <exp.c> merged.
-
-Fri Jul  1 12:44:56 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (OpenURI.open_http): refine post_connection_check
-	  call.
-
-Fri Jul  1 11:34:08 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* missing/crypt.c: replaced with 4.4BSD version.
-
-	* missing/erf.c: ditto.
-
-	* missing/vsnprintf.c: removed the third provision from the old
-	  BSD license.  [ruby-core:05177]
-
-Fri Jul  1 01:45:21 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* enum.c (enum_min, enum_max): must not return Qundef.
-	  fixed: [ruby-core:05299]
-
-Fri Jul  1 00:18:40 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/delegate.rb (Delegator::respond_to): respond_to? must check
-	  destination object.  [ruby-talk:146894]
-
-Thu Jun 30 19:00:21 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
-	* lib/irb/ruby-lex.rb (RubyLex::identify_number): alternative implements
-	  for [ruby-dev:26410]. And support a numeric form of 0d99999.
-
-Thu Jun 30 17:28:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/irb/ruby-lex.rb (RubyLex::identify_number): should not treat
-	  plain zero as an octal number.  [ruby-dev:26410]
-
-Thu Jun 30 15:13:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_eval): pre-evaluate argument for unambiguous
-	  evaluation order.  [ruby-dev:26383]
-
-Thu Jun 30 09:53:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/delegate.rb (Delegator::method_missing): forward unknown
-	  method to the destination.  suggested by
-	  <christophe.poucet@gmail.com>.  [ruby-talk:146776]
-
-Tue Jun 28 21:59:29 2005  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
-
-	* dir.c, eval.c, hash.c, process.c, ruby.c: avoid warning "unused
-	  variable" [ruby-dev:26387]
-
-Sat Jun 25 17:15:23 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.parse_query): should
-	  discard if key=val pair is empty. patch from Gary Wright.
-
-Sat Jun 25 23:30:51 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (detach_process_watcher): terminate process watcher
-	  thread right after rb_waitpid() succeed.  [ruby-talk:146430]
-
-Sat Jun 25 15:49:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* enum.c (enum_min, enum_max): do not ignore nil as the first element.
-
-Sat Jun 25 14:40:17 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/sdbm/init.c (fsdbm_select): SDBM#select had returned the array
-	  which contained each elements twice. [ruby-dev:26358]
-
-Fri Jun 25 05:06:47 2005  Michael Neumann  <mneumann@ruby-lang.org>
-
-	* lib/xmlrpc/*, test/xmlrpc/*: backported changes from HEAD into 1.8
-
-Fri Jun 24 17:00:00 2005  Shigeo Kobayashi  <shigeo@tinyforest.jp>
-
-	* ext/bigdecimal/bigdecimal.c: patch from "NATORI Shin"
-	  (u-tokyo.ac.jp) applied to fix rounding bug.
-
-Fri Jun 24 13:06:45 2005  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/common.rb, lib/uri/generic.rb: fixed typo in documents and
-	  replaced some existent domain name with "example.com".
-
-Fri Jun 24 12:23:19 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: fix typo on Tk.grid_propagate.
-
-	* ext/tk/lib/tk.rb: Tk.event_generate and TkWindow#event_generate
-	  accept TkEvent::Event object as context argument.
-
-	* ext/tk/lib/tk/event.rb: add TkEvent::Event#valid_fields and
-	  valid_for_generate to get field parameters of event_generate.
-
-Thu Jun 23 23:55:59 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* runruby.rb: should load built rbconfig.rb.
-
-Thu Jun 23 16:53:15 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/canvastag.rb: TkcGroup.new cannot include given items.
-	  TkcGroup#exclude calls wrong method.
-	  Add alias TkcGroup#add [ruby-talk:146049].
-
-	* ext/tk/lib/tk/canvas.rb: TkCanvas#dtag and some subcommands of
-	  TkCanvas#addtag fail to treat a TkcTag argument.
-
-	* ext/tk/lib/tk/event.rb: add TkEvent::Event#generate to help to send
-	  current event to other widgets.
-
-Mon Jun 20 18:44:04 2005  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (FUNCTION_CALL_MAY_RETURN_TWICE): DUMMY_SETJMP is replaced
-	  because setjmp is not enough to fix getcontext and SPARC register
-	  window problem.
-
-Mon Jun 20 16:48:36 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/dbm/dbm.c (fdbm_closed): new method DBM#closed?
-
-	* ext/gdbm/gdbm.c (fgdbm_closed): new method GDBM#closed?
-
-	* ext/sdbm/init.c (fsdbm_closed): new method SDBM#closed?
-
-	* test/dbm/test_dbm.rb, test/gdbm/test_gdbm.rb, test/sdbm/test_sdbm.rb
-	  (teardown): close all db objects before deleting data files.
-
-	* win32/win32.{ch} (unlink): hook runtime function to change
-	  file attribute before unlinking.
-	  fixed: [ruby-dev:26360]
-
-Mon Jun 20 02:15:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* gc.c (define_final): document fix: finalizers never get called
-	  before target object is destroyed.
-
-Mon Jun 20 01:26:49 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/openssl_missing.c, ext/openssl/ossl.h,
-	  ext/openssl/ossl_asn1.c, ext/openssl/ossl_bio.c,
-	  ext/openssl/ossl_pkcs12.h, ext/openssl/ossl_x509req.c: avoid
-	  compiler warnings. suggested by Michal Rokos.
-
-Sun Jun 19 14:09:07 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* gc.c (run_final): reduce unnecessary object allocation during
-	  finalization.
-
-	* gc.c (rb_gc_call_finalizer_at_exit): deferred finalizers list should
-	  be cleared before calling them.  fixed: [ruby-talk:145790]
-
-Fri Jun 17 13:01:40 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/time.rb (Time.parse): fix previous leap seconds support.
-	  (Time.rfc2822): ditto.
-	  (Time.xmlschema): ditto.
-
-Thu Jun 16 15:06:55 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_rb_threadVwaitCommand): Tcl_Release
-	  was missing.
-
-Thu Jun 16 13:34:48 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: add Tk.getMultiple{Open|Save}File() which return
-	  an Array of selected files.
-
-Thu Jun 16 12:53:24 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/time.rb (Time.parse): "Fri Jan  1 08:59:60 +0900 1999" was
-	  parsed as "Fri Jan 01 09:00:00 JST 1999" even on an environment
-	  which supports leap seconds.
-	  (Time.rfc2822): ditto.
-	  (Time.xmlschema): ditto.
-
-Thu Jun 16 08:29:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/dl/sym.c (rb_dlsym_call): needs FREE_ARGS before return.
-	  fixed memory leak. [ruby-Bugs-2034]
-
-Wed Jun 15 18:26:39 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: support "tk inactive" sub-command [for Tcl/Tk8.5a3]
-
-	* ext/tk/lib/tk/namespace.rb: support "namespace path" sub-command and
-	  'namespace ensemble' sub-command [for Tcl/Tk8.5a3]
-
-Tue Jun 14 02:02:43 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c: add TkUtil::CallbackSubst.subst_arg(m, ...) &
-	  _define_attribute_aliases(hash) to get substitution-argument from
-	  attributes (e.g. subst_arg(:x,:y,:num,:button) --> "%x %y %b %b ").
-
-	* ext/tk/lib/tk/event.rb: use _define_attribute_aliases().
-
-Mon Jun 13 13:01:05 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* hash.c (ruby_setenv): fixed SEGV. [ruby-dev:26186]
-
-Mon Jun 13 01:54:20 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* signal.c (sigexit): call rb_thread_signal_exit() instead of
-	  rb_exit().  [ruby-dev:26347]
-
-	* eval.c (rb_thread_signal_exit): a new function to exit on main
-	  thread.
-
-	* eval.c (rb_thread_switch): exit status should be retrieved from
-	  ruby_errinfo.
-
-	* eval.c (rb_f_exit): ensure exit(0) should call
-	  exit(EXIT_SUCCESS).
-
-Mon Jun 13 01:20:02 2005  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (rb_gc_mark_threads): curr_thread may not be part of the
-	  thread list.  [ruby-dev:26312]
-
-Fri Jun 10 23:35:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* missing/mkdir.c: remove. [ruby-core:05177]
-
-Fri Jun 10 22:54:26 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* missing.h: fd_set stuffs need sys/types.h.  fixed: [ruby-core:05179]
-
-Thu Jun  9 23:58:12 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/Win32API/Win32API.c (Win32API_Call): disable global
-	  optimization. fixed: [ruby-core:05143]
-
-Thu Jun  9 23:35:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* enum.c (enum_inject): default the result value to Qundef to use
-	  first element as initial value if not given.
-
-Thu Jun  9 19:55:41 2005  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (ruby_longjmp): new macro to call longjmp, setcontext, etc.
-	  (ruby_setjmp): new macro to call setjmp, getcontext, etc.
-	  (ruby_setjmp): call setjmp before getcontext to avoid IA64 register
-	  stack problem.
-	  [ruby-talk:144939]
-
-	* gc.c (Init_stack): remove IA64_MAGIC_STACK_LIMIT.
-
-Thu Jun  9 11:55:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/delegate.rb (SimpleDelegator::__setobj__): need check for
-	  recursive delegation.  [ruby-core:04940]
-
-Wed Jun  8 18:47:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-expr-beg): fix looking point drift.
-
-Wed Jun  8 11:11:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (get2comp): calculate proper 2's complement for
-	  negative numbers.  a bug in normalizing negative numbers
-	  reported from Honda Hiroki <hhonda@ipflex.com>.
-
-Wed Jun  8 08:33:10 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* enum.c (enum_min_by, enum_max_by): return nil if no iteration.
-	  fixed: [ruby-dev:26245]
-
-	* eval.c (rb_need_block): ensure a block is given.
-
-	* eval.c (backtrace): skip successive frames sharing same node.
-
-Wed Jun  8 00:15:08 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (ruby_getaddrinfo__aix): merged a patch from
-	  KUBO Takehiro <kubo at jiubao.org> to support AIX.  [ruby-list:40832]
-
-Wed Jun  8 00:09:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/yaml/rubytypes.rb (Array::to_yaml): merged a patch from
-	  Tilman Sauerbeck <tilman at code-monkey.de>.  [ruby-core:05055]
-
-	* lib/yaml/rubytypes.rb (Hash::to_yaml): ditto.
-
-Wed Jun  8 00:00:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/curses/curses.c (curses_insertln): merged a patch from
-	  TAKAHASHI Tamotsu <ttakah at lapis.plala.or.jp>.  [ruby-ext:02305]
-
-Tue Jun  7 19:34:15 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/irb/init.rb (IRB::IRB.rc_file_generators): more flexible
-	  IRB.rc_file_generators.  [ruby-core:05163]
-
-Tue Jun  7 18:39:31 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/thread.rb: RDoc documentation from Eric Hodel
-	  <drbrain at segment7.net> added.  [ruby-core:05148]
-
-Tue Jun  7 18:30:04 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): add .SUFFIXES from depend file.
-	  fixed: [ruby-dev:26294]
-
-Tue Jun  7 17:39:54 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (rb_mod_cvar_get): Module#class_variable_get(): back
-	  ported from CVS HEAD.  [ruby-talk:144741]
-
-	* object.c (rb_mod_cvar_set): Module#class_variable_set().
-	  [ruby-talk:144741]
-
-Tue Jun  7 16:32:53 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): raise exception on debug mode (-d),
-	  not verbose mode (-v/-w).  [ruby-core:05123]
-
-Tue Jun  7 10:30:49 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: slave-ip fails to call procedures
-	  delegated by master-ip.
-
-Sun Jun  5 23:00:35 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/console.rb: create console when required
-
-	* ext/tk/sample/tkextlib/tile/demo.rb: fix TypeError & create Console
-
-Sat Jun  4 14:55:18 2005  Tanaka Akira  <akr@m17n.org>
-
-	* test/dbm/test_dbm.rb: merged from ext/dbm/testdbm.rb.
-
-	* test/gdbm/test_gdbm.rb: merged from ext/gdbm/testgdbm.rb.
-
-	* test/sdbm/test_sdbm.rb: renamed from ext/sdbm/testsdbm.rb with
-	  modification to use test/unit.
-
-Fri Jun  3 14:06:12 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: fix typo.
-
-Wed Jun  1 11:32:42 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: can use single quote character in DESTDIR.
-	  [ruby-dev:26205]
-
-Mon May 30 23:48:29 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/macpkg.rb: add PACKAGE_NAME information of Tcl/Tk
-	  Extension.
-
-	* ext/tk/lib/tk/msgcat.rb: ditto.
-
-	* ext/tk/lib/tk/winpkg.rb: ditto.
-
-	* ext/tk/lib/tkextlib/*: ditto.
-
-Sat May 28 16:40:15 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* test/openssl/test_x509store.rb: add test for expired CRL
-	  and refine some assertions.
-
-Sat May 28 05:15:51 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509store.c (ossl_x509stctx_set_time): should
-	  not set internal flag directry.
-
-Sat May 28 02:00:11 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#request_line):
-	  ENV["REQUEST_URI"] is better to get correct Request-URI
-	  than ENV["SCRIPT_NAME"] + ENV["PATH_INFO"].  [ruby-dev:26235]
-
-Fri May 27 16:32:04 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/mkmf.rb: use the semicolon as the path separator
-	  in the environment of MSYS.  fixed: [ruby-dev:26232]
-
-Thu May 26 06:08:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: add shortcut-methods of tk_call + tk_split_list
-
-Wed May 25 22:52:42 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/irb/input-method.rb: do not use Readline::HISTORY.pop.
-	  (backported from HEAD)
-
-Wed May 25 21:55:40 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* ext/readline/readline.c: supported libedit. (backported from HEAD)
-
-	* ext/readline/extconf.rb: ditto.
-
-	* test/readline/test_readline.rb: ditto.
-
-Wed May 25 20:06:27 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: TkComm#tk_split_*list fail to split a kind of SJIS
-	  strings. To avoid the trouble, add arguments to control converting
-	  encoding, and do split on a UTF8 string.
-
-	* ext/tk/lib/multi-tk.rb: modify to attend encoding.
-
-	* ext/tk/lib/remote-tk.rb: ditto.
-
-	* ext/tk/lib/tk/itemconfig.rb: ditto.
-
-	* ext/tk/lib/tk/listbox.rb: ditto.
-
-	* ext/tk/lib/tk/namespace.rb: ditto.
-
-	* ext/tk/lib/tk/panedwindow.rb: ditto.
-
-	* ext/tk/lib/tk/text.rb: ditto.
-
-	* ext/tk/lib/tk/textmark.rb: ditto.
-
-	* ext/tk/lib/tk/texttag.rb: ditto.
-
-	* ext/tk/lib/tk/variable.rb: ditto.
-
-	* ext/tk/lib/tk/winfo.rb: ditto.
-
-	* ext/tk/lib/tkextlib/iwidgets/scrolledlistbox.rb: ditto.
-
-	* ext/tk/lib/tkextlib/iwidgets/scrolledtext.rb: ditto.
-
-	* ext/tk/lib/tk.rb: add TkWindow#lower_window/raise_window and
-	  Tk#lower_window/raise_window by reason of method-name conflict
-
-	* ext/tk/lib/tk/canvas.rb: bug fix on TkCanvas#delete when given
-	  non-TkcItem arguments.
-
-	* ext/tk/lib/tkextlib/iwidgets/scrolledcanvas.rb: ditto.
-
-Wed May 25 12:59:48 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (OpenURI::Meta::RE_QUOTED_STRING): a content of
-	  quoted-string should be zero or more characters.
-
-Tue May 24 23:42:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* numeric.c (fix_pow): support Fixnum ** Float case directly
-	  without coercing.  [ruby-talk:142697] [ruby-talk:143054]
-
-Tue May 24 16:57:24 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ruby.c (require_libraries): caused SEGV when continuation jumped
-	  in to the required library code.
-
-Tue May 24 11:56:25 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/getopts.rb: should warn only if verbose mode.
-	  fixed: [ruby-dev:26201]
-
-Tue May 24 06:45:31 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-font-lock-syntactic-keywords): string
-	  literals to be matched non-greedy.
-
-Tue May 24 00:34:32 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/calc: method name 'set' was able to crash with a class Set.
-	  [ruby-dev:26210]
-
-	* test/wsdl/document/test_rpc.rb: dateTime comparison failed under
-	  TZ=right/Asia/Tokyo (with leap second.) [ruby-dev:26208]
-
-Mon May 23 16:24:05 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/extconf.rb: Framework support on MacOS X Tiger.
-
-	* ext/tcltklib/README.1st: add description of Framework support options.
-
-Mon May 23 12:21:37 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* re.c (make_regexp): should not return junk address during
-	  compile time.  [ruby-dev:26206]
-
-Sun May 22 21:54:06 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/{soap,wsdl,xsd}, test/{soap,wsdl,xsd}: imported soap4r/1.5.4.
-
-	  == SOAP client and server ==
-
-	  === for both client side and server side ===
-
-	  * improved document/literal service support.
-	    style(rpc,document)/use(encoding, literal) combination are all
-	    supported.  for the detail about combination, see
-	    test/soap/test_style.rb.
-
-	  * let WSDLEncodedRegistry#soap2obj map SOAP/OM to Ruby according to
-	    WSDL as well as obj2soap.  closes #70.
-
-	  * let SOAP::Mapping::Object handle XML attribute for doc/lit service.
-	    you can set/get XML attribute via accessor methods which as a name
-	    'xmlattr_' prefixed (<foo name="bar"/> -> Foo#xmlattr_name).
-
-	  === client side ===
-
-	  * WSDLDriver capitalized name operation bug fixed.  from
-	    1.5.3-ruby1.8.2, operation which has capitalized name (such as
-	    KeywordSearchRequest in AWS) is defined as a method having
-	    uncapitalized name. (converted with GenSupport.safemethodname
-	    to handle operation name 'foo-bar').  it introduced serious
-	    incompatibility; in the past, it was defined as a capitalized.
-	    define capitalized method as well under that circumstance.
-
-	  * added new factory interface 'WSDLDriverFactory#create_rpc_driver'
-	    to create RPC::Driver, not WSDLDriver (RPC::Driver and WSDLDriver
-	    are merged).  'WSDLDriverFactory#create_driver' still creates
-	    WSDLDriver for compatibility but it warns that the method is
-	    deprecated.  please use create_rpc_driver instead of create_driver.
-
-	  * allow to use an URI object as an endpoint_url even with net/http,
-	    not http-access2.
-
-	  === server side ===
-
-	  * added mod_ruby support to SOAP::CGIStub.  rename a CGI script
-	    server.cgi to server.rb and let mod_ruby's RubyHandler handles the
-	    script.  CGIStub detects if it's running under mod_ruby environment
-	    or not.
-
-	  * added fcgi support to SOAP::CGIStub.  see the sample at
-	    sample/soap/calc/server.fcgi.  (almost same as server.cgi but has
-	    fcgi handler at the bottom.)
-
-	  * allow to return a SOAPFault object to respond customized SOAP fault.
-
-	  * added the interface 'generate_explicit_type' for server side
-	    (CGIStub, HTTPServer).  call 'self.generate_explicit_type = true'
-	    if you want to return simplified XML even if it's rpc/encoded
-	    service.
-
-	  == WSDL ==
-
-	  === WSDL definition ===
-
-	  * improved XML Schema support such as extension, restriction,
-	    simpleType, complexType + simpleContent, ref, length, import,
-	    include.
-
-	  * reduced "unknown element/attribute" warnings (warn only 1 time for
-	    each QName).
-
-	  * importing XSD file at schemaLocation with xsd:import.
-
-	  === code generation from WSDL ===
-
-	  * generator crashed when there's '-' in defined element/attribute
-	    name.
-
-	  * added ApacheMap WSDL definition.
-
-	* sample/{soap,wsdl}: removed.
-
-Sun May 22 19:11:35 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLServer#intialize):
-	  should initialize session id context. [ruby-core:4663]
-
-	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): add session id support.
-
-Sat May 21 10:24:21 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: tds files were not deleted when DESTDIR
-	  included '\' path delimiter. [ruby-dev:26193]
-
-Thu May 19 19:04:29 2005  speakillof  <speakillof@yahoo.co.jp>
-
-	* lib/rexml/encodings/SHIFT-JIS.rb: encoding and decoding were
-	  swapped. [ruby-core:4772]
-
-Wed May 18 23:42:25 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* error.c (exc_exception): reverted to call Exception#initialize
-	  directly.  fixed: [ruby-dev:26177]
-
-Wed May 18 23:39:09 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (glob_helper): get rid of using String.  [ruby-dev:26180]
-
-	* dir.c (push_braces): should skip balanced braces.
-
-	* eval.c (ruby_options), win32/win32.c (NtInitialize): move argument
-	  intialization back.  [ruby-dev:26180]
-
-Tue May 17 15:31:31 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): should
-	  break the loop if the socket reached to EOF. [ruby-talk:142285]
-
-Tue May 17 11:52:18 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (unixtime_to_filetime): use localtime() instead of
-	  gmtime() when using FileLocalTimeToFileTime().
-
-Mon May 16 22:28:43 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/win32.h, {bcc32,win32,wince}/Makefile.sub: moved rb_[ugp]id_t
-	  to get rid of redefinition warnings on mingw.
-
-	* class.c (rb_class_init_copy): singleton class is disallowed to copy,
-	  from its definition.  fixed: [ruby-talk:142749]
-
-Mon May 16 08:52:29 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.{h,c}: define rb_[pgu]id_t.
-
-Mon May 16 00:21:02 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#unlink): use SystemCallError instead of
-	  Errno::EISDIR because EISDIR is not portable.
-	  [ruby-core:5001]
-
-Sun May 15 22:11:33 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb (DRbObject#method_missing): use raise(exception).
-	  [ruby-dev:26164]
-
-Sun May 15 18:56:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in, ruby.h: define rb_[pgu]id_t macros instead of typedefs
-	  to get rid of types which might not be defined yet.  [ruby-dev:26165]
-
-Sun May 15 14:35:46 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#unlink): unlink a symlink to a directory
-	  was failed.  [ruby-core:4992]
-
-Sun May 15 09:57:30 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/win32.c (unixtime_to_filetime): deal with DST.
-	  [ruby-talk:141817]
-
-Sat May 14 23:59:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* error.c (exc_exception, {exit,name_err,syserr}_initialize): call
-	  Execption#initialize.  fixed: [ruby-talk:142593]
-
-Sat May 14 23:57:26 2005  Erik Huelsmann  <ehuels@gmail.com>
-
-	* configure.in: Check for the availability of pid_t, gid_t and uid_t and
-	  remove AC_TYPE_UID_T.  fixed: [ruby-core:04745]
-
-	* defines.h: Remove pid_t typedef.
-
-	* ruby.h: Define rb_pid_t, rb_gid_t and rb_uid_t in accordance with
-	 the available system types.
-
-	* process.c: Change instances of pid_t and gid_t to their rb_*
-	 counterparts.
-
-	* ext/pty/pty.c: Change pid_t to rb_pid_t.
-
-	* vms/config.h: Define HAVE_{P,G,U}ID_T to 1.
-
-	* win32/Makefile.sub: Remove #define for {g,u}id_t.
-
-	* win32/win32.c: Change pid_t to rb_pid_t.
-
-	* wince/Makefile.sub: Remove #define for {g,u}id_t.
-
-	* wince/sys/types.h: Remove definitions of {p,g,u}id_t.
-
-Fri May 13 23:44:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: keep srcdir unexpanded.
-
-	* lib/mkmf.rb (create_makefile): quote topdir and hdrdir if necessary.
-	  fixed: [ruby-core:04932]
-
-	* lib/mkmf.rb (configuration), {bcc32,win32,wince}/Makefile.sub: make
-	  also INSTALL_PROG and INSTALL_DATA system dependent.
-	  fixed: [ruby-core:04931]
-
-Fri May 13 17:54:39 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* variable.c (generic_ivar_get): rb_attr_get should not warn.
-	  [ruby-dev:26010]
-
-Fri May 13 12:28:43 2005  Daniel Berger  <djberge@qwest.com>
-
-	* array.c (rb_ary_select): can remove argc check. [ruby-core:4911]
-
-	* test/ruby/test_array.rb: add test for find_all.
-
-Fri May 13 11:29:00 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* eval.c (unknown_node): add volatile directive to prototype.
-
-Thu May 12 17:08:48 2005  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (rb_io_eof, remain_size, read_all, io_read, appendline)
-	  (swallow, rb_io_each_byte, rb_io_getc): revert previous change.
-
-	* io.c (rb_io_eof, io_fread, appendline, swallow, rb_io_each_byte)
-	  (rb_io_getc, rb_getc): call clearerr before getc to avoid
-	  stdio incompatibility.
-
-Thu May 12 16:52:20 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/rdoc/parsers/parse_c.rb: more readability for mixing
-	  progress "c..." and warning message.
-
-Thu May 12 16:31:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/nkf.c: follow nkf 2.0.5
-
-Thu May 12 16:15:01 2005  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (rb_io_eof, remain_size, read_all, io_read, appendline)
-	  (swallow, rb_io_each_byte, rb_io_getc): don't rely EOF flag.
-	  [ruby-talk:141527]
-
-Thu May 12 15:56:20 2005  Tilman Sauerbeck  <tilman@code-monkey.de>
-
-	* lib/rdoc/parsers/parse_c.rb: show parsing progress for C files.
-	  [ruby-core:4341]
-
-Thu May 12 13:47:56 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* test/drb/test_drb{ssl,unix}.rb: can test drb
-	  before install. (backported from HEAD) [ruby-dev:26146]
-
-Thu May 12 09:53:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* version.c (ruby_show_version): flush for non-tty stdout.
-
-Thu May 12 09:07:07 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* test/ruby/envutil.rb, test/drb/drbtest.rb: can test drb
-	  before install. (backported from HEAD) [ruby-Bugs-1672]
-
-Thu May 12 01:23:55 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval), parse.y (arg): reduce fixnum range literal at
-	  parser.  fixed: [ruby-dev:26113]
-
-	* eval.c (unknown_node): ignore broken NODE to get rid of accessing
-	  possibly inaccessible address.  fixed: [ruby-dev:26122]
-	  should emit more useful information like [ruby-dev:26126], though.
-
-Wed May 11 16:20:01 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb: new methods WEBrick::CGI#[], WEBrick::CGI#logger
-	  and WEBrick::CGI#config. (backported from HEAD)
-
-	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.escape_path): should
-	  not use String#split("/"). (backported from HEAD)
-
-Wed May 11 15:58:39 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (break_jump): break should not cross functions.
-	  [ruby-list:40818]
-
-Wed May 11 10:39:37 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/tempfile.rb (Tempfile#unlink): fixed typo.
-
-Wed May 11 01:03:36 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (TMP_ALLOC): use macro NEW_NODE() to get rid of warnings on
-	  platforms which have no alloca().  fixed: [ruby-talk:141301]
-
-Sun May  8 23:17:47 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/timer.rb: fix typo.
-
-Sun May  8 16:52:56 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/profiler.rb: fixed "undefined method `[]' for nil:NilClass"
-	  [ruby-core:4775] [ruby-talk:140401] [ruby-dev:26118]
-
-Sat May  7 22:58:00 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (have_var): no libs argument is given.
-
-Sun May  1 09:58:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ruby.c (process_sflag): replace '-' in variable names with '_'.
-	  [ruby-dev:26107]
-
-	* ruby.c (set_arg0): use also environment variable space for setting
-	  $0.  [ruby-core:04774]
-
-Wed Apr 27 23:42:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/Makefile.sub (OPTFLAGS): default global optimization to
-	  disabled only for VC++6.
-
-Tue Apr 26 22:58:00 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_invoke_core): call Tcl's "::unknown"
-	  command when can't get information of target command.
-
-Mon Apr 25 01:18:43 2005  Tanaka Akira  <akr@m17n.org>
-
-	* regex.c: declare rb_warn to have variadic argument.  [ruby-core:4751]
-
-Sat Apr 23 19:45:59 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_RubyExitCommand): exit with status code
-	  via TclTkIp#_eval didn't work. [ruby-talk:139390]
-
-Fri Apr 22 16:41:50 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_set_exc_message): fixed memory leak.
-
-	* ext/tcltklib/tcltklib.c: eTkCallbackReturn was not initialized.
-
-Thu Apr 21 00:07:50 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): support platforms have file separator
-	  other than /.
-
-	* {bcc32,win32,wince}/Makefile.sub (BUILD_FILE_SEPARATOR): separator
-	  of building platform.
-
-	* {bcc32,win32,wince}/Makefile.sub (CP, INSTALL): use COPY command.
-
-Wed Apr 20 23:22:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* Makefile.in, common.mk: miniruby depens on MINIOBJS.
-
-	* dmydln.c (dln_load): dummy function to raise LoadError.
-
-	* cygwin/GNUmakefile.in, {bcc32,win32,wince}/Makefile.sub: miniruby
-	  can't load extensions on Windows.
-
-Wed Apr 20 23:01:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* win32/ifchange.bat: delete testing files.
-
-Wed Apr 20 07:27:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* {bcc32,win32,wince}/configure.bat, {bcc32,win32,wince}/setup.mak:
-	  add extout option.
-
-	* bcc32/setup.mak: make configuration variables overridable.
-
-Wed Apr 20 04:15:27 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
-
-	* lib/irb.rb lib/irb/* doc/irb: IRB 0.9.5
-
-Tue Apr 19 23:37:09 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/ftools.rb (File.safe_unlink): do not modify a symlinked file.
-
-Tue Apr 19 00:06:20 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: expand path for ext/**/extconf.rb.
-
-Mon Apr 18 11:25:14 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/zlib/zlib.c (zstream_run): fixed SEGV. [ruby-core:4712]
-
-Sun Apr 17 23:57:49 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake, parse_args): do not expand destdir.
-
-	* ext/extmk.rb (relative_from): treat mere drive letter as an absolute
-	  path.
-
-Sat Apr 16 17:01:16 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* sample/rss/tdiary_plugin/rss-recent.rb (rss_recent_cache_rss):
-	  use the first date information of items as site date information
-	  if channel doesn't have date information.
-
-Sat Apr 16 15:27:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (RUBY_PROG_INSTALL): not add -p option to INSTALL.
-	  files need timestamps to be kept are only ar-archive on a few
-	  platforms, and be installed by instruby.rb but not INSTALL.
-	  fixed: [ruby-core:04721]
-
-	* mkconfig.rb: purge autoconf value variables.
-
-Sat Apr 16 10:36:01 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: quick hack... prepend DESTDIR.
-	  still have restriction on DESTDIR ("", "/", "e:")
-
-Sat Apr 16 03:59:42 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: check for OPENSSL_cleanse.
-
-	* ext/openssl/openssl_missing.h: ditto.
-
-Thu Apr 14 19:18:30 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (remove_file): ignore exceptions caused by
-	  chmod.
-
-	* lib/fileutils.rb (remove_dir): try to get rights to rmdir.
-	  [ruby-Bugs:1502] (2 items backportted from HEAD, rev 1.53-54)
-
-Thu Apr 14 16:57:40 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bcc32/Makefile.sub: failed to remove debug information files.
-	  fixed: [ruby-dev:26034]
-
-Wed Apr 13 23:40:21 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb (RSS::VERSION): 0.1.3 -> 0.1.4.
-
-	* lib/rss/rss.rb (RSS::Element#converter): fixed converter
-	  transmission bug.
-
-Wed Apr 13 21:20:35 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (mingw32): extract msvcr*.dll from objdump result.
-
-Wed Apr 13 20:24:30 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (mingw32): use actual runtime DLL name as ruby DLL
-	  name and default load path.
-
-	* win32/Makefile.sub, win32/setup.mak: ditto.
-
-Tue Apr 12 15:33:09 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_finalize): better modification than the
-	  previous commit [ruby-dev:26029].
-
-Tue Apr 12 12:38:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_finalize): fix SEGV when Tcl_GlobalEval()
-	  modifies the argument string to eval.
-
-Tue Apr 12 02:21:55 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_finalize): add existence check of
-	  Tcl commands before calling Tcl_GlobalEval().
-
-Mon Apr 11 23:47:21 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: [druby-ja:123] fix: When reference of my object is
-	  loaded, the object is tainted.
-
-	* test/drb/test_drb.rb: ditto.
-
-Mon Apr 11 22:18:23 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* dir.c, file.c (lstat): avoid warnings for mingw.
-
-Mon Apr 11 20:11:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_finalize): adhoc patch to avoid SEGV
-	  when exit on Tcl/Tk8.3.x.
-
-Mon Apr 11 15:26:25 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/mkmf.rb (configuration): shouldn't output hdrdir twice.
-
-Mon Apr 11 12:09:05 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* {bcc32,win32,wince}/Makefile.sub: ri data was not installed
-	  into correct path. [ruby-dev:26011]
-
-	* bcc32/Makefile.sub: defaulted install-nodoc. [ruby-dev:26011]
-
-Sun Apr 10 10:12:42 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c(ole_invoke): retry after converting Qnil
-	  to VT_EMPTY.
-
-	* ext/win32ole/win32ole/tests/testWIN32OLE.rb: correct error
-	  message string "Unknown" => "unknown".
-
-Sat Apr  9 18:20:31 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/image.rb: support to create TkImage object without
-	  creating a new image object on Tk.
-
-	* ext/tk/lib/tk/menu.rb: use TkCommandNames on create_self()
-
-	* ext/tk/lib/tk/root.rb: TkRoot.to_eval() returns '.'.
-
-	* ext/tk/lib/tk/text.rb: add methods to create a TkText::IndexString
-	  from (x, y) coords.
-
-	* ext/tk/lib/tkextlib/tile/: add demo and update support status.
-
-Sat Apr  9 14:42:29 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* sample/rss/tdiary_plugin/rss-recent.rb: supported configuration
-	  via Web browser.
-
-Sat Apr  9 11:59:57 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss: backoported from HEAD.
-
-	* lib/rss: refactored.
-	  - gave a name to 'x'.
-	  - undef_method -> remove_method for avoiding a warning in ruby 1.6.
-
-
-	* lib/rss/parser.rb: @@setter -> @@setters.
-
-	* lib/rss/parser.rb
-	  (RSS::BaseListener.register_uri)
-	  (RSS::BaseListener.uri_registered?)
-	  (RSS::BaseListener.install_get_text_element):
-	  swapped the first argument and the second argument.
-
-	* lib/rss/taxonomy.rb: swapped the first argument and the second
-	  argument for RSS::BaseListener.install_get_text_element.
-	* lib/rss/image.rb: ditto.
-	* lib/rss/syndication.rb: ditto.
-	* lib/rss/dublincore.rb: ditto.
-	* lib/rss/parser.rb: ditto.
-	* lib/rss/1.0.rb: ditto.
-	* lib/rss/2.0.rb: ditto.
-	* lib/rss/0.9.rb: ditto.
-	* lib/rss/content.rb: ditto.
-
-	* lib/rss/parser.rb
-	  (RSS::BaseListener.install_setter)
-	  (RSS::BaseListener.register_uri): changed fallback way.
-
-	* lib/rss/parser.rb: added class name registry for complex model
-	  elements. (ex. have childlen elements, have some attributes and
-	  a child element and so on.)
-
-
-	* lib/rss/dublincore.rb: supported multiple Dublin Core items.
-	* lib/rss/maker/dublincore.rb: ditto.
-
-	* lib/rss/maker/image.rb: supproted new Dublin Core API.
-
-
-	* lib/rss/maker/base.rb: added default current_element implementation.
-
-
-	* lib/rss/trackback.rb (RSS::TrackBackUtils.new_with_value_if_need):
-	  moved to RSS::Utils.
-
-	* lib/rss/utils.rb (RSS::Utils.new_with_value_if_need):
-	  moved from RSS::TrackBackUtils.
-
-
-	* lib/rss/maker/image.rb: fixed invalid argument of
-	  add_need_initialize_variable bug.
-	* lib/rss/maker/trackback.rb: ditto.
-
-
-	* lib/rss/rss.rb (Hash#merge): added for ruby 1.6.
-
-	* lib/rss/rss.rb (RSS::BaseModel.date_writer): changed to accept nil
-	  for date value.
-
-
-	* test/test_dublincore.rb: added tests for plural accessor and
-	  multiple Dublin Core items.
-
-	* test/test_setup_maker_1.0.rb: fixed swapped actual and expected
-	  values.
-
-	* test/rss/rss-assertions.rb (assert_multiple_dublin_core): added
-	  an assertion for testing multiple Dublin Core items.
-
-	* test/rss/test_maker_dc.rb (test_rss10_multiple): added a test
-	  for making multiple Dublin Core items.
-
-	* test/rss/test_maker_dc.rb (test_date): added a test for #date=
-	  and #dc_date=.
-
-
-	* sample/rss/tdiary_plugin/rss-recent.rb:
-	  new option: @options['rss-recent.use-image-link']:
-	  use image as link instread of text if available.
-
-	* sample/rss/tdiary_plugin/rss-recent.rb (RSS_RECENT_VERSION):
-	  0.0.5 -> 0.0.6.
-
-Fri Apr  8 20:17:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): hdrdir needs to be defined also in
-	  Config::CONFIG.
-
-	* lib/mkmf.rb (configuration, create_makefile): get rid of recursive
-	  macro reference.
-
-Fri Apr  8 18:26:56 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c: add callbacks to OpenSSL::SSL::SSLContexts.
-	  - SSLContext#client_cert_cb=(aProc). it is called when a client
-	    certificate is requested by a server and no certificate was not
-	    set for the SSLContext. it must return an Array which includes
-	    OpenSSL::X509::Certificate and OpenSSL::PKey::RSA/DSA objects.
-	  - SSLContext#tmp_dh_callback=(aProc). it is called in key
-	    exchange with DH algorithm. it must return an OpenSSL::PKey::DH
-	    object.
-
-	* ext/openssl/ossl_ssl.c (ossl_sslctx_set_ciphers): ignore the
-	  argument if it's nil.
-
-	* ext/openssl/ossl_pkey.c
-	  (GetPrivPKeyPtr, ossl_pkey_sign): should call rb_funcall first.
-	  (DupPrivPKeyPtr): new function.
-
-	* ext/openssl/ossl_pkey_dh.c: add default DH parameters.
-
-	* ext/openssl/ossl_pkey.h: ditto.
-
-Fri Apr  8 01:55:20 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/demos-{en,jp}/goldberg.rb: reduced window size.
-	  [ruby-dev:25992]
-
-Thu Apr  7 23:58:40 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): keep directory names in Makefile as macros.
-
-	* lib/mkmf.rb (configuration, create_makefile): ditto.
-
-	* lib/mkmf.rb (CXX_EXT): separate C++ extensions.
-
-Thu Apr  7 17:43:25 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* eval.c (rb_call0): "return" event hook should be always executed
-	  if event_hooks is set.  fixed: [ruby-core:04662]
-	  (backported from HEAD)
-
-Mon Apr  4 23:17:52 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb (TkComm#array2tk_list): accept enc-mode argument to
-	  decide whether convert encoding of each element or not.
-
-	* ext/tk/lib/tk/variable.rb (TkVariable#value=): fail to convert the
-	  encoding of array elements when assign an array to an TkVariable
-	  object.
-
-Mon Apr  4 10:26:48 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tk/lib/tk/dialog.rb: fixed typo.
-
-Sun Apr  3 17:16:33 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.{h,c} (rb_w32_fdopen): avoid warning on bcc32.
-	  (backported from HEAD)
-
-Sat Apr  2 23:38:54 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (CP, INSTALL): get rid of less portable options.
-
-	* lib/mkmf.rb (configuration, create_makefile): correct configuration
-	  variable.
-
-	* {bcc32,win32,wince}/{Makefile.sub,setup.mak}: leave prefix empty in
-	  config.status for backward compatibility.  fixed: [ruby-core:04649]
-
-	* lib/mkmf.rb (create_makefile): ensure library directories get made
-	  before copying libraries there.
-
-Sat Apr  2 16:59:46 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: forgot to update RELEASE_DATE
-
-	* ext/tk/lib/tk/variable.rb: fix namespace trouble when autoloading
-
-	* ext/tk/lib/tk/palette.rb: define Tcl variable 'tkPalette' as global
-
-	* ext/tk/lib/tk/dialog.rb: use array2tk_list method when calling
-	  Tk.ip_eval.
-
-	* ext/tk/lib/tk/autoload.rb: add autoload entry 'TkDialogObj' and
-	  'TkWarningObj'
-
-Sat Apr  2 02:19:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb (TkWindow.initialize): accept 'without_creating'
-	  option without 'widgetname' option to allow creating a widget object
-	  which is used as an argument of Tcl/Tk's widget allocation commands.
-
-	* ext/tk/lib/tk/image.rb (TkImage.initialize): accept 'imagename'
-	  option to create a image object by the given name.
-
-Thu Mar 31 22:23:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (SRC_EXT): exclude just case different suffixes on case
-	  insensitive file system platforms.
-
-	* README.EXT, README.EXT.ja (Appendix C): utility functions.
-
-Thu Mar 31 14:15:44 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_engine.c (ossl_engine_s_load): should return
-	  value. [ruby-dev:25971]
-
-Thu Mar 31 08:25:50 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* common.mk (RUBYOPT): clear for the environment RubyGems installed.
-
-	* common.mk (clean-local): keep $(PREP) files till distclean.
-
-	* common.mk (check): do all tests.
-
-Thu Mar 31 06:00:20 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_engine.c (ossl_engine_s_load): should not raise
-	  error even if the specified engine could not be loaded. (Dynamic
-	  engines don't have fixed name to load.)
-
-Thu Mar 31 00:18:27 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/ifchange.bat, win32/rm.bat: backported from HEAD.
-
-Wed Mar 30 23:44:50 2005  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
-
-	* Makefile.in, */Makefile.sub, */configure.bat,
-	  cygwin/GNUmakefile.in, common.mk, configure.in, ext/extmk.rb,
-	  lib/mkmf.rb, instruby.rb, runruby.rb: backport extout.
-	  [ruby-dev:25963]
-
-Wed Mar 30 17:41:48 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: add TclTkIp#_create_console() method to
-	  create a Tcl/Tk's console window.
-
-	* ext/tk/lib/multi-tk.rb: support TclTkIp#_create_console() method.
-
-	* ext/tk/lib/remote-tk.rb: ditto.
-
-	* ext/tk/lib/tk/console.rb: ditto.
-
-	* ext/tk/lib/tk.rb: update RELEASE_DATE
-
-	* ext/tk/sample/demo-*/check2.rb: use 'return' in the Proc object.
-
-	* ext/tk/sample/tkextlib/**: ditto.
-
-Tue Mar 29 22:11:56 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/rinda/test_rinda.rb: use DRbObject.new_with instead of reinit.
-	  [ruby-dev:25961]
-
-Mon Mar 28 23:40:40 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: move method DRbObject#reinit to DRbObject.new_with.
-	  extract method DRbObject.prepare_backtrace. add DRb.regist_server,
-	  remove_server, fetch_server. change server in thread variable if
-	  in-proc server. [druby-ja:113]
-
-	* lib/drb/gw.rb: ditto.
-
-Mon Mar 28 20:43:34 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/syck/rubyext.c: get rid of warnings caused by a bug of VC.
-
-Mon Mar 28 08:39:49 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (iconv_create): Iconv::Failure requires 3
-	  arguments.  (pointed out by NaHi)
-
-Sat Mar 26 22:51:33 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb (_callback_entry_class?): add for checking whether
-	  a class is available for a callback entry.
-
-	* ext/tk/lib/tk.rb (after_cancel): add Tk.after_cancel(afterID) method.
-
-	* ext/tk/lib/tk.rb (array2tk_list): change from private module method
-	  of TkComm to public module method.
-
-	* ext/tk/lib/tk.rb (cget): add check that slot argument is not
-	  empty string.
-
-	* ext/tk/lib/tk.rb (configinfo): ditto.
-
-	* ext/tk/lib/tk/itemconfig.rb (itemcget): add check that slot argument
-	  is not empty string.
-
-	* ext/tk/lib/tk/itemconfig.rb (itemconfiginfo): ditto.
-
-	* ext/tk/lib/tk/entry.rb: add TkEntry#icursor and icursor= (alias of
-	  cursor and cursor= method).
-
-	* ext/tk/lib/tk/font.rb: improve font treatment when the font name is
-	  empty string.
-
-	* ext/tk/lib/tk/variable.rb: add :variable, :window and :procedure
-	  type.
-
-	* ext/tk/lib/tk/variable.rb: improve treatment of array-type
-	  tkvariable.
-
-	* ext/tk/lib/tkextlib/blt.rb: add commands for zooming.
-
-	* ext/tk/lib/tkextlib/blt/*: bug fix.
-
-	* ext/tk/lib/tkextlib/treectrl/tktreectrl.rb: bug fix and add methods
-	  to call TreeCtrl commands for bindings.
-
-	* ext/tk/sample/tkextlib/blt/*: new sample scripts.
-
-	* ext/tk/sample/tkextlib/treectrl/*: ditto.
-
-Fri Mar 25 10:53:16 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (WIN32_LEAN_AND_MEAN): removed because a lot of
-	  troubles.  [ruby-list:40721]
-
-Thu Mar 24 23:10:44 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (macro_defined?): try to compile for an old compiler
-	  which doesn't bail out at #error directive.  [ruby-dev:25818]
-
-	* lib/mkmf.rb (check_sizeof): refine logging messages.
-
-Thu Mar 24 03:57:48 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/utils.rb (WEBrick::Utils.create_listeners):
-	  - should raise ArgumentError if no port is specified.
-	  - even if the specified port is 0, all TCPServers should be
-	  initialized with the port given to the first one.
-
-	* lib/webrick/server.rb (WEBrick::GenericServer#initialize): if :Port
-	  parameter is 0, it should be updated with the port number which
-	  actually listened.
-
-Wed Mar 23 00:35:10 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* test/ruby/test_settracefunc.rb (test_event): added tests for
-	  "class" and "end" and "raise".
-
-Tue Mar 22 22:40:18 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* eval.c (rb_call0): check event_hooks instead of trace_func.
-
-Tue Mar 22 17:30:44 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* eval.c (rb_add_event_hook): new function to add a hook function for
-	  interpreter events. (backported form HEAD)
-
-Sun Mar 20 22:51:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (mkmf_failed): check if Makefile is created without
-	  create_makefile.
-
-Sat Mar 19 23:48:10 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-expr-beg): returned true always.
-	  fixed: [ruby-list:40683]
-
-Sat Mar 19 00:41:02 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/font.rb: add some TkFont class methods to get font
-	  information without creating a TkFont object.
-
-	* ext/tk/lib/tkextlib/treectrl/tktreectrl.rb: bug fix and define some
-	  classes for components of Tk::TreeCtrl
-
-Thu Mar 17 17:42:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* struct.c (make_struct): allow non local-id field
-	  names. [ruby-core:04575]
-
-	* struct.c (inspect_struct): ditto.
-
-Wed Mar 16 23:36:02 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* eval.c (rb_call0): call_cfunc() should be protected.
-
-	* test/ruby/test_settracefunc.rb: added test for c-return.
-
-Wed Mar 16 22:20:25 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* object.c (str_to_id): fixed typo.
-
-Wed Mar 16 18:08:32 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): reorganize "return" event post.
-
-Tue Mar 15 23:49:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (Init_iconv): InvalidEncoding also should include
-	  Iconv::Failure.
-
-Tue Mar 15 16:38:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c (ary2list): give wrong arguments to hash2kv()
-
-Mon Mar 14 19:39:33 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/timer.rb (TkTimer): forgot to clear @return_value
-	  when restarting
-
-	* ext/tk/lib/tk/sample/cd_timer.rb: new sample of TkRTTimer
-
-Mon Mar 14 12:21:03 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/timer.rb (TkRTTimer): forgot to reset the callback
-	  time. So, 'continue' do all callbacks between 'stop' and 'continue'.
-
-Mon Mar 14 08:14:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (str_to_id): warn for NUL containing strings.
-
-Mon Mar 14 00:13:49 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/timer.rb (TkRTTimer): correct calculation of offset
-	  value. get a little better accuracy.
-
-	* ext/tk/sample/demos-en/widget: use a binding with no local variables
-	  when eval a sample script.
-
-	* ext/tk/sample/demos-en/bind.rb: ditto.
-
-	* ext/tk/sample/demos-en/tcolor: ditto.
-
-	* ext/tk/sample/demos-jp/widget: ditto.
-
-	* ext/tk/sample/demos-jp/bind.rb: ditto.
-
-	* ext/tk/sample/demos-jp/tcolor: ditto.
-
-Sun Mar 13 10:04:17 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/rinda/test_rinda.rb: remove test_gc. [ruby-dev:25871]
-
-Thu Mar 10 19:12:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (lib_eventloop_ensure): mis-delete a timer
-	  handler when exit from a recursive called eventloop
-
-	* ext/tk/lib/tk/timer.rb: new TkRTTimer class, which can works for a
-	  realtime operation
-
-	* ext/tk/sample/tkrttimer.rb: sample of TkRTTimer class
-
-	* ext/tk/lib/tk/textmark.rb: move  TkTextMark#+ and TkTextMark#- to
-	  TkText::IndexModMethods
-
-	* ext/tk/lib/tk/text.rb: improve TkTextMark#+ and TkTextMark#-, and
-	  add them to TkText::IndexModMethods module
-
-	* ext/tk/sample/tktextio.rb: add test part of "seek by text index
-	  modifiers"
-
-Thu Mar 10 08:10:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* re.c (make_regexp): need to free internal regexp structure when
-	  compilation fails.  [ruby-talk:133228]
-
-Wed Mar  9 20:25:58 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c (ossl_start_ssl, ossl_ssl_write): call
-	  rb_sys_fail if errno isn't 0. [ruby-dev:25831]
-
-	* ext/openssl/lib/openssl/cipher.rb: fix typo. [ruby-dev:24285]
-
-Wed Mar  9 15:46:35 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/server.rb (WEBrick::GenericServer#start): should
-	  restore @token if accept failure. suggested by Dominique Brezinski.
-	  [ruby-core:04518]
-
-Wed Mar  9 13:37:57 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tktextio.rb: fix bug of handling 'end' position.
-	  support initial text, overwrite setting and pos_gravity control.
-
-Tue Mar  8 18:16:55 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tktextio.rb: New sample script. TkTextIO class in this
-	  sample supports to use a text widget as if it is a I/O stream (such
-	  like as StringIO class).
-
-Tue Mar  8 13:54:40 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/socket/socket.c: workaround for some of 4.4BSD-Lite derived OSs.
-
-Tue Mar  8 12:36:17 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c: document from Sam Roberts
-	  <sroberts@uniserve.com> for getsockopt and setsockopt is merged.
-	  [ruby-doc:824]
-
-Tue Mar  8 01:27:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/nkf.c: follow nkf 1.66
-	  fixed: [ruby-dev:25828]
-
-Mon Mar  7 21:35:02 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* sample/webrick/httpsd.rb: fix typo in comment. suggested by
-	  Kazuhiko Shiozaki.
-
-Mon Mar  7 14:55:43 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (block_pass): should not push unique number if a block is
-	  not an orphan.  [ruby-dev:25808]
-
-Wed Feb 16 02:55:21 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c (ossl_start_ssl, ossl_ssl_read,
-	  ossl_ssl_write):
-	  - need to set errno on Win32 platform.
-	  - should call rb_sys_fail instead of rasing SSLError if
-	    SSL_ERROR_SYSCALL occured.
-	  - should wait for that the underlying IO become readable or
-	    writable if the error was SSL_ERROR_WANT_READ or
-	    SSL_ERROR_WANT_WRITE. [ruby-dev:25795]
-
-	* ext/openssl/lib/openssl/buffering.rb
-	  (Buffering#initialize): should set @eof and @rbuffer.
-	  (Buffering#fill_rbuff): should rescue Errno::EAGAIN.
-	  (Buffering#consume_rbuf): pointless eof flag resetting is deleted.
-	  (Buffering#read): should return an empty string if the specified
-	  size is zero.
-	  (Buffering#readpartial): new method.
-	  (Buffering#readline): fix typo.
-	  (Buffering#getc): return the first character of string correctly.
-	  (Buffering#each): fix typo.  suggested by Brian Ollenberger.
-	  (Buffering#readchar): fix typo.
-	  (Buffering#eof?): should read again it the input buffer is empty.
-	  (Buffering#do_write): should rescue Errno::EAGAIN.
-	  (Buffering#puts): use "\n" as the output field separator.
-
-	* ext/openssl/lib/openssl/ssl.rb: set non-blocking flag to the
-	  underlying IO.
-
-	* ext/openssl/extconf.rb: get rid of GNUmakefile generation.
-
-	* text/openssl/test_pair.rb: test for IO like methods.
-
-	* test/ruby/ut_eof.rb: test about empty file.
-
-Mon Mar  7 10:22:06 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/un.rb: should use OptionParser. (backported form HEAD)
-
-Mon Mar  7 09:18:42 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_cmp_m): should not return false but nil.
-	  fixed: [ruby-dev:25811]
-
-Mon Mar  7 01:22:14 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c: remove the some codes which depend on the
-	  difference between Ruby1.8 and 1.9, because st.c on Ruby1.9
-	  was changed.
-
-Mon Mar  7 00:01:04 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fail to call TclTkLib.mainloop when $SAFE==4
-
-Sun Mar  6 16:41:33 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: HTTPHeader holds its header fields as an array
-	  (backport from CVS HEAD rev 1.112-1.123). [ruby-list:40629]
-
-	* test/net/http/test_httpheader.rb: new file.
-
-Sun Mar  6 11:47:10 2005  Sam Roberts  <sroberts@uniserve.com>
-
-	* lib/pp.rb: rdoced.  [ruby-core:4490]
-
-Sun Mar  6 11:36:37 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pp.rb (File::Stat#pretty_print): Etc.getpwuid and Etc.getgrgid
-	  may return nil.  [ruby-talk:129826]
-	  reported by Daniel Berger.
-
-Sat Mar  5 18:06:21 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* dir.c (fnmatch): removed unnecessary code. (ruby_1_8 didn't have
-	  String#clear, so [ruby-dev:24749] didn't affect it)
-
-	* win32/win32.c (NtInitialize): ditto. (by numeric.c 1.101.2.14)
-
-Sat Mar  5 16:29:26 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: freeze callback-entry objects
-
-	* ext/tk/lib/tkextlib/tile.rb: support tile-0.6
-
-Fri Mar  4 19:39:28 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#do_includes): replace
-	  also locally defined modules.
-
-	* ext/iconv/iconv.c: backport Iconv::InvalidEncoding from CVS HEAD.
-
-	* ext/strscan/strscan.c: moved misplaced rdoc.
-
-Fri Mar  4 15:58:12 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi-lib.rb: add deprecation warning. [ruby-dev:25499]
-	  getopts.rb, parsearg.rb, importenv.rb as well.
-
-Fri Mar  4 11:17:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_rbUpdateCommand): get rid of
-	  warnings with Tcl/Tk 8.3 or former (backport from CVS_HEAD).
-
-	* ext/tcltklib/tcltklib.c (ip_rb_threadUpdateCommand): ditto.
-
-Fri Mar  4 10:15:30 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/set.rb (SortedSet::setup): a hack to shut up warning.
-	  [ruby-talk:132866]
-
-Fri Mar  4 07:07:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/nkf.c: follow nkf 1.63
-
-Thu Mar  3 23:49:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/nkf.c: follow nkf 1.62
-
-Thu Mar  3 11:49:51 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* sample/rss/tdiary_plugin/rss-recent.rb: added site information.
-
-Wed Mar  2 19:53:07 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (parse_args): add DESTDIR only when not directed
-	  already.  fixed: [ruby-dev:25781]
-
-Wed Mar  2 17:14:18 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (lib_eventloop_core): fix typo
-
-Wed Mar  2 16:00:02 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: enforce thread-check and exception-handling
-	  to avoid SEGV trouble.
-	  [KNOWN BUG] When supports pthread and running multiple Tk
-	  interpreters, an interrupt signal causes SEGV frequently. That
-	  may be a trouble of Ruby's signal handler.
-
-	* ext/tk/tkutil/tkutil.c; fix a bug on converting a SJIS string array
-	  to a Tcl's list string.
-
-	* ext/tk/tcltklib.c: wrap Tcl's original "namespace" command to
-	  protect from namespace crash.
-
-	* ext/tk/lib/multi-tk.rb: enforce exception-handling.
-
-	* ext/tk/lib/multi-tk.rb: catch IRB_EXIT to work on irb.
-
-	* ext/tk/lib/tk.rb: ditto.
-
-	* ext/tk/tcltklib.c: add TclTkLib.mainloop_thread?
-
-	* ext/tk/lib/multi-tk.rb: (bug fix) callback returns a value.
-
-	* ext/tk/lib/tk/canvas.rb (delete): bug fix when multiple arguments.
-
-	* ext/tk/lib/clock.rb: fix 'no method error'.
-
-	* ext/tk/lib/clock.rb (self.clicks): accept a Symbol argument.
-
-	* ext/tk/lib/variable.rb: be able to set default_value_type; :numeric,
-	  :bool, :string, :symbol, :list, :numlist or nil (default; same to
-	  :string). If set a type, TkVariable#value returns a value of the
-	  type.
-
-	* ext/tk/lib/tkextlib/tclx/tclx.rb: add Tk::TclX.signal to warn the
-	  risk of using TclX extension's 'signal' command.
-
-	* ext/tk/sample/irbtk.rb: irb with Ruby/Tk.
-
-	* ext/tk/sample/demos-*/anilabel.rb: bug fix on 'show code'
-
-	* ext/tk/sample/demos-*/aniwave.rb: new Ruby/Tk animation demo.
-
-	* ext/tk/sample/demos-*/pendulum.rb: ditto.
-
-	* ext/tk/sample/demos-*/goldberg.rb: ditto.
-
-	* ext/tk/sample/demos-*/widget: add entries of animation demos.
-
-Tue Mar  1 00:47:43 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/rinda/test_rinda.rb: backport from CVS_HEAD. use
-	  MockClock.sleep instead of Kernel.sleep [ruby-dev:25387]
-
-Tue Mar  1 00:34:24 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/tuplespace.rb (Rinda::TupleSpace): improved keeper thread.
-
-	* test/rinda/test_rinda.rb: ditto.
-
-Mon Feb 28 11:42:23 2005  Ian Macdonald  <ian@caliban.org>
-
-	* exception error messages updated.  [ruby-core:04497]
-
-Mon Feb 28 09:03:09 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (Init_socket): add bunch of Socket
-	  constants.  Patch from Sam Roberts <sroberts@uniserve.com>.
-	  [ruby-core:04409]
-
-Wed Feb 23 15:04:32 2005  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/generic.rb (split_userinfo): should split ":pass" into ""
-	  and "pass".  [ruby-dev:25667]
-
-Wed Feb 23 08:00:18 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_s_create): no need for negative argc check.
-	  [ruby-core:04463]
-
-	* array.c (rb_ary_unshift_m): ditto.
-
-Wed Feb 23 01:57:46 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (initialize): handle certs correctly. Thanks,
-	  NABEYA Kenichi. (backported from CVS HEAD)
-
-Tue Feb 22 07:25:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (parser_yylex): identfier after dot must not be a variable.
-
-Mon Feb 21 10:04:49 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* {bcc32,win32,wince}/Makefile.sub (config.h): add fcntl.
-
-	* win32/win32.[ch] (fcntl): ditto.
-
-	* win32/win32.c (rb_w32_connect): support nonblocking mode.
-
-	* ext/socket/socket.c (wait_connectable, ruby_connect): support
-	  nonblocking connect on various platforms.
-	  all changes are backported from CVS HEAD. [ruby-core:3154],
-	  [ruby-core:4364].
-
-Sun Feb 20 00:48:48 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::FTP#buffer_open): access mechanism
-	  re-implemented according to RFC 1738.
-	  reported by Guillaume Marcais.  [ruby-talk:131650]
-
-Sat Feb 19 18:11:47 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb (DRbObject#respond_to?): take two arguments.
-	  [ruby-dev:25722]
-
-	* test/drb/drbtest.rb: ditto.
-
-Sat Feb 19 13:52:02 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb: call OpenSSL::SSL::SSLSocket#post_connection_check
-	  after connection is made.
-
-Sat Feb 19 01:32:03 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/bigdecimal/lib/bigdecimal/newton.rb: resolved LoadError.
-	  [ruby-dev:25685]
-
-	* ext/bigdecimal/sample/linear.rb: ditto.
-
-	* ext/bigdecimal/sample/nlsolve.rb: ditto.
-
-	* ext/bigdecimal/lib/bigdecimal/nlsolve.rb: removed because this file
-	  is sample script and same file exists in ext/bigdecimal/sample.
-
-Fri Feb 18 17:14:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/xmlrpc/parser.rb (XMLRPC::FaultException): make it subclass
-	  of StandardError class, not Exception class.  [ruby-core:04429]
-
-Thu Feb 17 20:11:18 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/drb/drb.rb (DRbServer.default_safe_level): fix typo.
-
-Thu Feb 17 20:11:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/digest/test_digest.rb: separate test case for each algorithms.
-	  [ruby-dev:25412]
-
-Thu Feb 17 11:54:00 2005  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/collector.rb (collect_file): now deletes paths added
-	  to $LOAD_PATH instead of restoring it verbatim.
-
-	* lib/test/unit/autorunner.rb (AutoRunner.run): fixed so that
-	  'ruby -rtest/unit -rtest1 -rtest2 -e0' will use the objectspace
-	  collector again. Also tried to simplify the calling convention.
-
-	* test/runner.rb: adjusted for new AutoRunner semantics.
-
-	* lib/test/unit.rb: ditto.
-
-Thu Feb 17 04:21:47 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/open3.rb (Open3::popen3): $? should not be EXIT_FAILURE.
-	  fixed: [ruby-core:04444]
-
-Thu Feb 17 00:09:45 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/drb/ignore_test_drb.rb: move TestDRbReusePort to new file
-	  [ruby-dev:25238]
-
-	* test/drb/test_drb.rb: add method DRbService.ext_service, move
-	  TestDRbReusePort to new file [ruby-dev:25238]
-
-	* test/drb/test_drb.rb: ditto.
-
-	* test/drb/test_drbssl.rb: ditto.
-
-	* test/drb/test_drbunix.rb: ditto.
-
-	* test/drb/ut_drb.rb: reduce sleep.
-
-Thu Feb 17 00:02:27 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (is_defined): NODE_IASGN is an assignment.
-
-Wed Feb 16 23:34:30 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: add lazy stop_service. ([druby-ja:109])
-
-	* lib/drb/extserv.rb: ditto.
-
-Wed Feb 16 17:07:57 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c: Follow the change of st.c (st_foreach)
-	  [ruby-list:40623].
-	  Sometimes mis-convert from a Ruby's Array of SJIS Strings, which
-	  includes some kind of SJIS characters, to a Tcl's UTF8 list string.
-
-Mon Feb 14 23:58:17 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/parser.rb (RSS::ListenerMixin::tag_end):
-	  fixed invalid namespace handling bug.
-
-Mon Feb 14 13:12:38 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/ssl.rb
-	 (OpenSSL::SSL::SSLSocket#post_connection_check): new method.
-
-Mon Feb 14 00:40:49 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb (InvokeMethod.perform): pass DRb info to sub thread.
-
-	* test/drb/test_drb.rb (test_01_safe1_safe4_eval): fix test case.
-
-Sun Feb 13 23:13:46 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/dublincore.rb (RSS::DublicCoreModel#date{,=}): added
-	  convenient methods.
-
-	* lib/rss/0.9.rb (RSS::Rss::Channel#date{,=}): ditto.
-
-	* lib/rss/2.0.rb (RSS::Rss::Channel::Item#date{,=}): ditto.
-
-	* test/rss/: added tests for the convenient methods.
-
-Sun Feb 13 22:43:03 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb (DRbServer): add default_safe_level, safe_level,
-	  config[:safe_level] ([druby-ja:120])
-
-	* test/drb/test_drb.rb, ut_eval.rb, ut_safe1.rb: ditto.
-
-Sun Feb 13 16:56:52 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (WEBrick::CGI.start): should set reason-phrase
-	  to the value of status header field. ([ruby-dev:40617])
-
-Sun Feb 13 00:52:33 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/erb.rb (ERB::Util.h, u): make it module_function.
-
-Sat Feb 12 17:29:19 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (OpenURI.open_loop): send authentication only for
-	  the URI directly specified.
-
-Sat Feb 12 15:07:23 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* random.c (rand_init): suppress warning.
-
-Sat Feb 12 13:54:03 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb: support https if the platform provides CA
-	  certificates.
-
-Sat Feb 12 06:18:28 2005  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
-
-	* ext/etc/etc.c (Init_etc): sGroup needs HAVE_ST_GR_PASSWD check.
-	  [ruby-dev:25675]
-
-Fri Feb 11 17:40:42 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509store.c (ossl_x509store_set_default_paths):
-	  new method OpenSSL::X509::Store#set_default_paths.
-
-Fri Feb 11 11:33:53 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::HTTP#proxy_open): new option supported:
-	  :http_basic_authentication.
-	  suggested by Kent Sibilev.  [ruby-core:4392]
-
-Fri Feb 11 06:30:07 2005  George Ogata  <g_ogata@optushome.com.au>
-
-	* misc/ruby-mode.el: ignore parenthesis inside heredoc.
-	  [ruby-core:04415]
-
-Fri Feb 11 04:54:13 2005  Tilman Sauerbeck  <tilman@code-monkey.de>
-
-	* lib/rdoc/generators/html_generator.rb: [ruby-core:04412]
-
-	* lib/rdoc/generators/ri_generator.rb: ditto.
-
-Thu Feb 10 11:14:17 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/Makefile.sub (COMMON_HEADERS): shouldn't include winsock2.h.
-
-	* ext/socket/extconf.rb (sockaddr_storage): remove workaround for
-	  mswin32.
-
-Thu Feb 10 10:29:16 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/curses/curses.c: don't need to check HAVE_WCOLOR_SET excluding
-	  window_color_set().
-
-Thu Feb 10 00:47:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* struct.c (make_struct): fixed: [ruby-core:04402]
-
-Wed Feb  9 08:07:08 2005  Paul Duncan  <pabs@pablotron.org>
-
-	* ext/curses/curses.c (window_color_set): [ruby-core:04393]
-
-Tue Feb  8 23:51:47 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: reject :instance_eval, :class_eval, :module_eval
-	  [druby-ja:117]
-
-Tue Feb  8 13:06:12 2005  Sam Roberts  <sroberts@uniserve.com>
-
-	* ext/socket/socket.c (Init_socket): SO_REUSEPORT added.
-	  [ruby-talk:130092]
-
-Tue Feb  8 09:30:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::Cookie): [ruby-talk:130040]
-
-Tue Feb  8 00:19:02 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/resolv.rb (Resolv::DNS::Name#subdomain_of?): new method.
-	  (Resolv::DNS::Name#inspect): ditto.
-	  Suggested by Sam Roberts.  [ruby-talk:129086]
-
-Mon Feb  7 10:06:30 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c: [ruby-doc:818]
-
-Mon Feb  7 01:56:20 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* instruby.rb, rubytest.rb (srcdir): no longer embed srcdir into
-	  rbconfig.rb. (backported from CVS HEAD)
-
-	* ext/socket/extconf.rb (sockaddr_storage): winsock2.h have the
-	  definition of struct sockaddr_storage, but socket.c doesn't
-	  include it because this version of ruby still has binary level
-	  compatibility with winsock1.
-
-	* lib/mkmf.rb (create_makefile): should support header files in
-	  depend file.
-
-Mon Feb  7 01:21:50 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/socket/extconf.rb: check if getaddrinfo() works fine only when
-	  wide-getaddrinfo option is not given.  fixed: [ruby-dev:25422]
-
-	* lib/mkmf.rb ($extmk): check if under ext directory.
-
-	* lib/mkmf.rb (Logging.postpone): allow recursive operation.
-
-	* lib/mkmf.rb (try_constant): make sure if really a constant, reduce
-	  the number of times of compile.
-
-	* lib/mkmf.rb (have_macro, have_var, byte_order): new functions.
-
-	* lib/mkmf.rb (find_library): allow directory list with separators.
-
-	* lib/mkmf.rb (arg_config): manage provided configuration options.
-
-	* lib/mkmf.rb (dir_config): accept arrays of directory names as
-	  default values.
-
-	* mkconfig.rb: no longer embed srcdir and compile_dir into
-	  rbconfig.rb.
-
-	* lib/mkmf.rb (create_makefile): fix unbalanced parens.
-
-Sun Feb  6 19:23:01 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* eval.c (stack_extend): add prototype because VC++8 doesn't
-	  accept __declspec(noinline) with K&R style function definitions.
-	  (backported from CVS HEAD)
-
-Sun Feb  6 14:14:26 2005  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date.rb (new_with_hash): changed messages of exception.
-
-	* lib/date/format.rb (str[fp]time): undocumented conversion
-	  specifications %[1-3] are now deprecated.
-
-Sun Feb  6 12:20:11 2005  Akinori MUSHA  <knu@iDaemons.org>
-
-	* bignum.c (rb_big2ulong_pack): One too many arguments are passed
-	  to big2ulong().
-
-	* re.c (rb_reg_init_copy, rb_reg_initialize_m): One too many
-	  arguments are passed to rb_reg_initialize().
-
-Sun Feb  6 03:24:20 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/resolv.rb (Resolv::DNS::Resource::TXT): multiple strings was not
-	  handled.
-	  (Resolv::DNS::Resource::TXT#strings): new method to return all
-	  strings.
-	  (Resolv::DNS::Message::MessageEncoder#put_string_list): new method.
-	  (Resolv::DNS::Message::MessageDecoder#get_string_list): ditto.
-	  based on [ruby-talk:129732] by Sam Roberts.
-
-Fri Feb  4 00:30:45 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss: supported Image module.
-	  http://web.resource.org/rss/1.0/modules/image/
-
-Thu Feb  3 23:42:36 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_close, strio_close_read, strio_close_write):
-	  should return nil instead of self as well as IO.  [ruby-dev:25623]
-
-	* ext/stringio/stringio.c (strio_extend, strio_putc): fill with zero
-	  extended portion.  [ruby-dev:25626]
-
-Wed Feb  2 23:52:53 2005  sheepman  <sheepman@tcn.zaq.ne.jp>
-
-	* ext/stringio/stringio.c (strio_truncate): should MEMZERO an extended
-	  part.  [ruby-dev:25618]
-
-Wed Feb  2 21:56:01 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb (RSS::Element#convert): added.
-
-	* lib/rss/rss.rb: convert -> need_convert.
-
-	* lib/rss/1.0.rb: ditto.
-
-	* lib/rss/0.9.rb: ditto.
-
-	* lib/rss/2.0.rb: ditto.
-
-	* lib/rss/trackback.rb: ditto.
-
-Tue Feb  1 22:48:48 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb (DRb::DRbObject#respond_to?): check marshal_dump and
-	_dump.
-
-Tue Feb  1 00:20:23 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (aix): fix linker flags on AIX.  [ruby-talk:125460]
-
-Mon Jan 31 13:33:21 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: add invalid namespace check
-
-	* ext/tk/lib/multi-tk.rb: add invalid_namespace? method
-
-	* ext/tk/lib/remote-tk.rb: ditto
-
-Mon Jan 31 10:29:18 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/irb/context.rb (IRB::Context::initialize): [ruby-core:04330]
-
-Sat Jan 29 09:42:12 2005  Sam Roberts  <sroberts@uniserve.com>
-
-	* lib/resolv.rb (Resolv::DNS::Resource::IN::SRV): Added RFC2782 SRV
-	  resource record for specifying location of services.
-
-Fri Jan 28 17:16:55 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/resolv.rb (Resolv::DNS::Config.parse_resolv_conf):
-	  parse options line for ndots option.
-	  (Resolv::Hosts#lazy_initialize): return self.
-	  (Resolv::DNS#lazy_initialize): ditto.
-	  (Resolv::DNS::Config#lazy_initialize): ditto.
-	  Suggested by Sam Roberts.
-
-Thu Jan 27 13:18:03 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* st.c (st_foreach): report success/failure by return value.
-	  [ruby-Bugs-1396]
-
-Thu Jan 27 00:15:29 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb (setup): support BSD-style
-	  directory group inheritance. (backport from HEAD, rev 1.32)
-
-	* test/fileutils/fileasserts.rb (assert_same_entry): show entry
-	  difference. (backport from HEAD, rev 1.4)
-
-Wed Jan 26 23:09:11 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/protocol.rb (WriteAdapter#puts): should append \n, not
-	  prepend. [ruby-talk:128302] (backport from HEAD, rev 1.75)
-
-Wed Jan 26 10:51:50 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (flock_winnt, flock_win95): unlock file even if
-	  LOCK_NB is specified. (backported from CVS HEAD)
-
-Tue Jan 25 17:11:51 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ruby.c (proc_options): correct -T option in RUBYOPT. (backported
-	  from CVS HEAD)
-
-Tue Jan 25 14:05:52 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fix SEGV bug; trouble on canceling remained
-	  after scripts [ruby-dev:25479]: NULL current namespace when deleting
-	    Tk interpreter [ruby-talk:126225]
-
-	* ext/tcltklib/extconf.rb: bug fix; TCL_ENABLE_THREAD flag is inverted
-	  [ruby-talk:126360]
-
-	* ext/tcltklib/extconf.rb: add yet another native-thread check
-
-	* ext/tk/tkutil.c: fix SEGV bug; NULL string pointer when finalize
-	  Ruby interpreter
-
-	* ext/tk/lib/multi-tk.rb: avoid warning for deleted safeTk ip frame
-
-	* ext/tk/lib/tk/bindtag.rb: bug fix; new method of named bindtag
-	  doesn't return the created object [ruby-dev:25479]
-
-	* ext/tk/lib/tk/menu.rb: bug on treating arguments [ruby-dev:25479]
-
-	* ext/tk/lib/tk.rb: bug fix; cannot accept a callback ID string for
-	  a command argument [ruby-dev:25479]
-
-	* ext/tk/lib/multi-tk.rb: ditto
-
-	* ext/tk/lib/tk/*.rb: ditto
-
-	* ext/tk/lib/tkextlib/*.rb: ditto
-
-	* ext/tk/sample/demos-jp/anilabel.rb: new demo script
-
-	* ext/tk/sample/demos-en/anilabel.rb: ditto
-
-	* ext/tk/sample/tkHTML/ss.rb: local variable scope bug fix
-	  [ruby-dev:25479]
-
-Mon Jan 24 15:44:25 2005  Tilman Sauerbeck  <tilman@code-monkey.de>
-
-	* lib/rdoc/parsers/parse_c.rb: allow whitespace after function names.
-	  [ruby-core:4296]
-
-	* lib/rdoc/parsers/parse_simple.rb: adds support for private comments
-	  in the "simple" parser. [ruby-core:4301]
-
-Mon Jan 24 15:44:25 2005  Charles Mills  <cmills@freeshell.org>
-
-	* lib/rdoc/parsers/parse_c.rb: adds support for constants
-	  (rb_define_const), accessors (rb_define_attr), and makes a
-	  couple fixes. [ruby-core:4307]
-
-Mon Jan 24 15:44:25 2005  Florian Gro  <florgro@gmail.com>
-
-	* lib/rdoc/parsers/parse_rb.rb: Logic for def Builtin.method() end
-	  [ruby-core:4302]
-
-Mon Jan 24 15:44:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* document updates - [ruby-core:04296], [ruby-core:04301],
-	  [ruby-core:04302], [ruby-core:04307]
-
-Sun Jan 23 12:41:16 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/wsdlDriver.rb: from 1.5.3-ruby1.8.2, operation which has
-	  capitalized name (such as KeywordSearchRequest in AWS) is defined as
-	  a method having uncapitalized name. (converted with
-	  GenSupport.safemethodname to handle operation name 'foo-bar').  it
-	  introduced serious incompatibility; in the past, it was defined as a
-	  capitalized.
-
-	  define capitalized method as well under that circumstance.
-
-Sun Jan 23 05:24:42 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_to_der): should call
-	  GetOCSPReq at first.
-
-Sat Jan 22 23:09:47 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/ssl.rb (accept): rescue SSLError. [druby-ja:110]
-
-Sat Jan 22 22:35:03 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/unix.rb: fail if UNIXFileOwner is set. [druby-ja:111]
-
-Fri Jan 21 23:58:42 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_set_pos): clear EOF flag.
-	  [ruby-talk:127511]
-
-Fri Jan 21 20:07:02 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/resolv.rb (Resolv::DNS::Config.resolv): don't raise ResolvError.
-	  reported by Sam Roberts.  [ruby-talk:127133]
-
-Fri Jan 21 16:58:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (rb_push_glob): should work for NUL delimited patterns.
-
-Fri Jan 21 13:58:37 2005  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (u8tou16): fixed typo. fixed: [ruby-list:40546]
-	  (backported from CVS HEAD)
-
-Fri Jan 21 09:30:16 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* rubyio.h (rb_eof_error): should mark as NORETURN. (backported
-	  from CVS HEAD)
-
-Fri Jan 21 00:31:36 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/syck/rubyext.c (syck_parser_bufsize_set): avoid VC++ warning
-	  "local variable 'size' used without having been initialized".
-
-Thu Jan 20 19:03:24 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): shouldn't set $extflags on mswin32.
-
-	* win32/Makefile.sub (LIBRUBY_SO): should use $DLDOBJS instead of
-	  $EXTOBJS.
-	  fixed: [ruby-core:04290] (backported from CVS HEAD)
-
-Thu Jan 20 11:42:02 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_new4): should propagate taintedness.
-
-	* struct.c (rb_struct_set): use original method name, not callee
-	  name, to retrieve member slot.  [ruby-core:04268]
-
-	* time.c (time_strftime): protect from format modification from GC
-	  finalizers.
-
-Wed Jan 19 18:06:40 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/ipaddr.rb (to_s, test_to_s): too many colons with some cases.
-	  (backported from CVS HEAD)
-
-Wed Jan 19 01:16:30 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/resolv.rb (Resolv::DNS::Config.parse_resolv_conf): ignore
-	  domain and search directive without an argument.
-	  reported by Sam Roberts.  [ruby-talk:126781]
-
-Tue Jan 18 15:03:05 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/ssl.rb (WEBrick::Config::SSL): the default value
-	  of :SSLEnable is false.
-
-	* lib/webrick/server.rb (WEBrick::Daemon.start): prepared stdio
-	  don't allow changing its mode.
-
-	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_service):
-	  should delete trailing LF from the result of pack("m*").
-
-	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_connect):
-	  - should delete trailing LF from the result of pack("m*").
-	  - clear Request-Line not to send the response by HTTPServer#run.
-
-	* lib/webrick/httputils (WEBrick::HTTPUtils.parse_qvalues):
-	  refine regexp (and change the name of a local variable).
-
-	* lib/webrick/httputils.rb (WEBrick::HTTPUtils#escape_path): add
-	  new method to escape URI path component.
-
-	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#request_line): should
-	  escape SCRIPT_NAME and PATH_INFO before being parsed as a URI.
-
-	* test/webrick/*, sample/webrick/httpproxy.rb: add new file.
-
-Mon Jan 17 23:33:46 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (aix): fix typo.  [ruby-talk:126401]
-
-Mon Jan 17 07:08:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/readline/readline.c: suppress warnings.
-
-	* lib/irb/extend-command.rb (IRB::ContextExtender.def_extend_command):
-	  ditto.
-
-	* lib/irb/ext/history.rb (IRB::Context::set_last_value): ditto.
-
-	* lib/irb/ext/history.rb (IRB::Context::eval_history): ditto.
-
-	* lib/irb/locale.rb (IRB::Locale::real_load): ditto.
-
-	* lib/irb/slex.rb (SLex::Node::create_subnode): remove garbage.
-
-Mon Jan 17 00:09:42 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/uri/common.rb (PORT): typo fix. fixed: [ruby-core:04256]
-
-Sat Jan 15 14:57:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ruby.c (proc_options): ignore trailing CRs at the end of short
-	  options as well as long options.  fixed: [ruby-core:04232]
-
-Sat Jan 15 13:35:16 2005  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb (RSS::VERSION): 0.1.2 -> 0.1.3.
-
-	* lib/rss/rss.rb: accept inheritance. [ruby-talk:126104]
-
-Thu Jan 13 04:48:53 2005  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (io_fread): don't warn nonblocking behavior by default.
-
-Wed Jan 12 00:36:29 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* object.c (rb_class_superclass): superclass of singleton class also
-	  should be a singleton class.  fixed: [ruby-list:40519]
-
-Tue Jan 11 09:44:40 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* numeric.c (Init_Numeric): turn off floating point exceptions
-	  on bcc32. "1e300".to_f had crashed by overflow.
-
-Tue Jan 11 03:10:10 2005  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (copy_entry): could not copy symbolic link.
-	  [ruby-talk:125733]
-
-	* lib/fileutils.rb (copy_stream): use read/write instead of
-	  sysread/syswrite.
-
-Mon Jan 10 23:08:15 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* variable.c (rb_autoload): hide internal data from ruby level.
-	  fixed: [ruby-dev:25435], [ruby-list:40498]
-
-Mon Jan 10 01:22:55 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (rb_data_object_alloc): klass may be NULL.
-	  [ruby-list:40498]
-
-Sun Jan  9 03:12:58 2005  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (io_fread): warn nonblocking behavior.
-	  (io_readpartial): new method IO#readpartial.
-
-Sat Jan  8 04:38:47 2005  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml.rb: Kernel#y requires an argument.
-
-Fri Jan  7 21:12:29 2005  TAMURA Takashi  <sheepman@tcn.zaq.ne.jp>
-
-	* random.c (rand_init): use ALLOC_N instead of ALLOCA_N
-	  [ruby-dev:25426]
-
-Fri Jan  7 18:03:35 2005  Tanaka Akira  <akr@m17n.org>
-
-	* gc.c (mark_locations_array): avoid core dump with -O3.
-	  [ruby-dev:25424]
-
-Thu Jan  6 20:31:07 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/zlib/zlib.c (zstream_end): should return value. (backported
-	  from CVS HEAD)
-
-Thu Jan  6 19:55:13 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.c (rb_w32_close): didn't close socket handle.
-	  [ruby-dev:25414]
-
-	* win32/win32.c (rb_w32_open_osfhandle): bcc32's _open_osfhandle
-	  never set EMFILE.
-
-Thu Jan  6 17:14:31 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* random.c (random_seed): O_NONBLOCK isn't defined on some
-	  platforms. [ruby-dev:25417]
-
-Thu Jan  6 13:45:35 2005  Tanaka Akira  <akr@m17n.org>
-
-	* lib/time.rb: recognize +00:00 and GMT as a localtime.
-
-Thu Jan  6 07:58:28 2005  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/usage.rb (RDoc::RDoc.usage_no_exit): Allow for colons
-	  in path names on DOS machines. (thanks to Johan Nilsson)
-
-Wed Jan  5 20:16:32 2005  Tanaka Akira  <akr@m17n.org>
-
-	* random.c (limited_big_rand): didn't work if SIZEOF_BDIGITS == 2.
-	  [ruby-dev:25408]
-
-	* random.c (random_seed): refined.
-
-Wed Jan  5 12:49:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_initialize): Thread objects cannot be initialized
-	  again.  fixed: [ruby-core:04067]
-
-Wed Jan  5 10:48:16 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* dir.c (dir_s_mkdir): win32 special processing doesn't need any
-	  longer. (backported from CVS HEAD)
-
-	* win32/win32.[ch] (rb_w32_mkdir): new function. POSIX.1 compatible
-	  interface. (backported from CVS HEAD)
-
-	* win32/win32.[ch] (rb_w32_rmdir): new function. (backported from CVS
-	  HEAD)
-
-Wed Jan  5 02:30:11 2005  Tanaka Akira  <akr@m17n.org>
-
-	* random.c (init_by_array): imported from mt19937ar-cok.tgz.
-	  (genrand_int32): ditto.
-	  (genrand_real): replaced with genrand_res53 in mt19937ar-cok.
-	  (rand_init): support bignum for longer seed.
-	  (random_seed): generate longer seed.
-	  (make_mask): new function.
-	  (limited_rand): ditto.
-	  (limited_big_rand): ditto.
-	  (rb_f_rand): call limited_rand and limited_big_rand.
-	  [ruby-dev:25403]
-
-Tue Jan  4 23:25:29 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (rb_big_rand): should return positive random number.
-	  [ruby-dev:25401]
-
-Tue Jan  4 11:15:29 2005  TAMURA Takashi  <sheepman@tcn.zaq.ne.jp>
-
-	* bignum.c (rb_big_rand): do not use rb_big_modulo to generate
-	  random bignums.  [ruby-dev:25396]
-
-Mon Jan  3 14:01:54 2005  Tanaka Akira  <akr@m17n.org>
-
-	* random.c (random_seed): don't use /dev/urandom if it is not
-	  character device.
-
-Mon Jan  3 11:37:42 2005  Tanaka Akira  <akr@m17n.org>
-
-	* random.c (random_seed): use /dev/urandom if available.
-	  [ruby-dev:25392]
-
-Mon Jan  3 07:46:42 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpauth/htpasswd.rb (WEBrick::Htpasswd#reload):
-	  raise NotImplementedError if password is encrypted by digest
-	  algorithms. This patch is contributed by sheepman. [ruby-list:40467]
-
-	* lib/webrick/httpauth/digestauth.rb
-	  (WEBrick::HTTPAuth::DigestAuth#_authenticate): fix digest calculation.
-	  This patch is contributed by sheepman. [ruby-list:40482]
-
-	* lib/webrick/{httpauth.rb,httpauth/basicauth.rb,httpproxy.rb}: use
-	  pack/unpack-template char "m" instead of lib/base64.rb to do base64
-	  encoding/decoding. fixed: [ruby-dev:25336]
-
-	* test/webrick/test_httpauth.rb: new file.
-
-Sat Jan  1 04:20:23 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ns_spki.c (ossl_spki_set_challenge): should call
-	  StringValue before GetSPKI. fixed: [ruby-dev:25359].
-
-Sat Jan  1 01:13:28 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* variable.c (rb_autoload): [ruby-dev:25373]
-
-Fri Dec 31 14:10:43 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::TextFormatter::display_flow_item): Fix problem
-	  if heading contains formatting.
-
-Thu Dec 30 00:41:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (svalue_to_avalue): [ruby-dev:25366]
-
-	* string.c (rb_str_justify): [ruby-dev:25367]
-
-Wed Dec 29 11:07:07 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/template/html/kilmer.rb: Update to use new
-	  sections.
-
-Tue Dec 28 22:31:46 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (rb_str_justify): create buffer string after argument type
-	  conversion.  fixed: [ruby-dev:25341]
-
-Tue Dec 28 15:41:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/nkf.c (reinit): should initialize all static
-	  variables.  fixed: [ruby-list:40445]
-
-Tue Dec 28 15:25:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/nkf/lib/kconv.rb (Kconv::RegexpEucjp): second byte is up to
-	  0xfe.
-
-	* ext/nkf/lib/kconv.rb (Kconv#kconv): should handle UTF8 and UTF16
-	  properly.
-
-Tue Dec 28 13:35:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/zlib/zlib.c (rb_deflate_s_deflate, rb_inflate_s_inflate): ensure
-	  freeing internal zstreams.  fixed: [ruby-dev:25309]
-
-	* ext/zlib/zlib.c (rb_deflate_init_copy): replace rb_deflate_clone.
-
-Tue Dec 28 12:26:45 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/Makefile.sub, win32/setup.mak (RDOCTARGET, install,
-	  install-nodoc, install-doc): rdoc support for mswin32.
-
-	* win32/configure.bat (--enable-install-doc, --disable-install-doc):
-	  ditto.
-
-Mon Dec 27 20:02:14 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fix SEGV bug when deleting Tk interp
-
-	* ext/tk/lib/multi-tk.rb: ditto
-
-Mon Dec 27 16:55:17 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509name.c (Init_ossl_x509name): should use
-	  rb_hash_new to get exactly a Hash. fix [ruby-dev:25325].
-
-Mon Dec 27 16:29:56 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_justify): [ruby-dev:25341]
-
-Mon Dec 27 15:47:48 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/fileasserts.rb: sync with HEAD.
-
-	* test/fileutils/test_fileutils.rb: ditto.
-
-	* test/fileutils/test_nowrite.rb: ditto.
-
-Mon Dec 27 15:21:07 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mv): should raise error when moving a
-	  directory to the (empty) directory. [ruby-talk:124368]
-	  (backport from HEAD 1.48)
-
-	* lib/fileutils.rb (mv): wrongly did not overwrite file on Win32
-	  platforms. (backport from HEAD 1.48)
-
-Sat Dec 25 11:11:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* stable version 1.8.2 released.
-
-Sat Dec 25 04:23:49 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mkdir, mkdir_p): should ensure directory
-	  permission. (backportted from HEAD, 1.47)
-
-	* lib/fileutils.rb (traverse, remove_dir): untaint trasted
-	  objects. (backportted from HEAD, 1.46)
-
-Sat Dec 25 01:28:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c: cancel io_reopen() change on Dec. 24th.
-
-	* dln.c: use <dlfcn.h> for NetBSD.  [ruby-dev:25313]
-
-	* io.c (rb_f_select): IO list could be altered.  [ruby-dev:25312]
-
-Fri Dec 24 23:51:48 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: bcc32 should use RTL dll (backport from HEAD)
-	  [ruby-dev:25306]
-
-	* win32/win32.[ch]: ditto.
-
-Fri Dec 24 23:27:18 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/image.rb: TkPhotoImage#cget bug fix
-
-Fri Dec 24 18:39:25 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.[ch]: failed to compile on bcc32 (and probably wince)
-	  [ruby-dev:25306]
-
-Fri Dec 24 02:52:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (io_reopen, rb_io_reopen): prohibit to change access mode for
-	  special IO ports.  [ruby-dev:25225]
-
-Fri Dec 24 02:22:53 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/syck/rubyext.c (rb_syck_io_str_read): [ruby-core:03973]
-
-	* ext/syck/rubyext.c (syck_loader_transfer): check type conversion.
-
-	* ext/syck/rubyext.c (syck_parser_assign_io, rb_new_syck_node): duck
-	  typing.
-
-	* ext/syck/rubyext.c (syck_parser_s_alloc, syck_parser_initialize):
-	  allocation framework.
-
-	* ext/syck/rubyext.c (syck_emitter_s_alloc, syck_emitter_initialize):
-	  ditto.
-
-Fri Dec 24 01:21:00 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tkextlib/blt.rb: add BLT extension support
-
-	* ext/tk/lib/tkextlib/blt/*.rb: ditto
-
-	* ext/tk/lib/tkextlib/blt/tile/*.rb: ditto
-
-Thu Dec 23 23:36:28 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (proc_setgroups): check if the argument lenght is
-	  modified.  fixed: [ruby-dev:25285]
-
-Thu Dec 23 13:13:33 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: define TclTkLib::COMPILE_INFO and
-	  RELEASE_DATE
-
-	* ext/tcltklib/extconf.rb: ditto
-
-	* ext/tk/tkutil.c: define TkUtil::RELEASE_DATE
-
-	* ext/tk/lib/tk.rb: define Tk::RELEASE_DATE
-
-Thu Dec 23 09:38:31 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (io_reopen): restore exact mode.  fixed: [ruby-core:04003]
-
-Thu Dec 23 00:16:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (bsdi): use $(CC) for LDSHARED.  fixed [ruby-dev:25270]
-
-Wed Dec 22 11:14:55 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_mode_modenum): replace O_ACCMODE with O_RDWR.
-	  fixed: [ruby-dev:25273]
-
-Wed Dec 22 08:34:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/dl/sym.c (rb_dlsym_initialize): extract internal pointers after
-	  all argument conversion.  fixed: [ruby-dev:25271]
-
-Wed Dec 22 00:08:01 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/*, test/soap/*, sample/soap/authheader/*: eval cleanup.
-
-Tue Dec 21 22:07:33 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1_traverse, ossl_asn1_decode,
-	  ossl_asn1_decode_all): temporary value should be marked volatile.
-
-Tue Dec 21 14:40:02 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1_traverse, ossl_asn1_decode,
-	  ossl_asn1_decode_all): use rb_str_new4 to avoid SEGV.
-	  fix [ruby-dev:25261]
-
-	* test/openssl/test_asn1.rb: add tests for OpenSSL::ASN1.
-
-Tue Dec 21 12:22:40 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (io_reopen): keep duplex pipe in correct mode for exception
-	  safeness.  fixed: [ruby-dev:25152]
-
-Tue Dec 21 12:10:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/grid.rb: rescue bug of 'grid configure' on Tcl/Tk8.3-
-
-Tue Dec 21 00:53:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1_traverse): [ruby-dev:25261]
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1_decode): ditto.
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1_decode_all): ditto.
-
-Mon Dec 20 23:22:26 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* added files:
-	  * lib/soap/mapping/wsdl*.rb
-	  * lib/wsdl/soap/element.rb
-	  * lib/wsdl/xmlSchema/simpleContent.rb
-
-	* modified files:
-	  * lib/soap/*
-	  * lib/wsdl/*
-	  * lib/xsd/*
-	  * test/soap/*
-	  * test/wsdl/*
-	  * test/xsd/*
-	  * sample/soap/*
-	  * sample/sdl/*
-
-	* summary
-	  * imported from the soap4r repository.  Version: 1.5.3-ruby1.8.2
-
-	  * added several XSD basetype support: nonPositiveInteger,
-	    negativeInteger, nonNegativeInteger, unsignedLong, unsignedInt,
-	    unsignedShort, unsignedByte, positiveInteger
-
-	  * HTTP client connection/send/receive timeout support.
-
-	  * HTTP client/server gzipped content encoding support.
-
-	  * improved WSDL schema definition support; still is far from
-	    complete, but is making step by step improovement.
-
-Mon Dec 20 22:56:39 2004  Tanaka Akira  <akr@m17n.org>
-
-	* gc.c (stack_end_address): gcc noinline attribute is available since
-	  gcc-3.1.
-
-Mon Dec 20 14:07:02 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: supports new features of Tcl/Tk8.5a2
-
-	* ext/tk/lib/tk/clock.rb: ditto
-
-	* ext/tk/lib/tk/text.rb: ditto
-
-	* ext/tk/lib/tk/panedwindow.rb: ditto
-
-Mon Dec 20 12:47:13 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/net/https.rb,protocols.rb,telnets.rb: delete
-	  doc and code about SSLContext#{key_file,cert_file}.
-	  fixed: [ruby-dev:25243]
-
-Mon Dec 20 12:42:17 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* io.c (io_fwrite): workaround for MSVCRT's bug.
-	  fixed: [ruby-core:03982]
-
-Mon Dec 20 11:21:04 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_eof): check if closed before clearerr().
-	  fixed: [ruby-dev:25251]
-
-Mon Dec 20 03:30:40 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session#initialize): empty session id was
-	  used if request had no session key.  fixed: [ruby-core:03981]
-
-Mon Dec 20 01:51:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* struct.c (make_struct): [ruby-dev:25249]
-
-Mon Dec 20 00:28:20 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rexml/encodings/SHIFT-JIS.rb: backported from CVS HEAD.
-
-	* lib/rexml/encodings/SHIFT_JIS.rb: ditto.
-
-Sun Dec 19 17:19:48 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509store.c
-	  (ossl_x509store_set_time): add OpenSSL::X509::Store#time=.
-	  (ossl_x509stctx_set_time): add OpenSSL::X509::StoreContext#time=.
-
-	* test/openssl/ossl_x509store.rb: test certificate validity times.
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_to_s): add optional
-	  second argument to specify the output format (see also
-	  X509_NAME_print_ex).
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_init): new constants:
-	  OpenSSL::X509::Name::COMPAT, OpenSSL::X509::Name::RFC2253,
-	  OpenSSL::X509::ONELINE, OpenSSL::X509::MULTILINE.
-
-	* ext/openssl/lib/openssl/x509.rb (OpenSSL::X509::Name::RFC2253DN):
-	  new module to provide the parse for RFC2253 DN format.
-
-	* ext/openssl/lib/openssl/x509.rb (OpenSSL::X509::Name.parse_rfc2253):
-	  new method to parse RFC2253 DN format.
-
-	* test/openssl/ossl_x509name.rb: add tests about RFC2253 DN.
-
-	* text/openssl/ssl_server.rb: try to listen ports from 20443 to 20542
-	  while EADDRINUSE is raised.
-
-	* all changes in this entry are backport from 1.9.
-
-Sun Dec 19 17:24:59 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (enable_rpath): use rpath flag to embed the library
-	  path into extensions on ELF environment.  [ruby-dev:25035]
-
-Sun Dec 19 11:01:25 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit.rb: use standalone runner for -e.
-
-	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner#options): accept
-	  multiple -p and -x options.
-
-	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir#recursive_collect):
-	  ditto.
-
-Sat Dec 18 16:36:23 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/zlib/zlib.c (rb_deflate_s_deflate, rb_inflate_s_inflate):
-	  disallow interrupt by type conversion.  fixed: [ruby-dev:25226]
-
-Sat Dec 18 15:16:41 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/webrick/httpauth.rb,
-	  lib/webrick/httpauth/{basicauth.rb,digestauth.rb}: use
-	  pack/unpack-template char "m" instead of lib/base64.rb to do base64
-	  encoding/decoding.
-
-Sat Dec 18 10:51:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (dir_open_dir): new function.  [ruby-dev:25242]
-
-Fri Dec 17 18:07:01 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* test/readline/test_readline.rb: fix for BSD. Thanks, GOTOU Yuuzou.
-	  fixed: [ruby-dev:25218]
-
-Fri Dec 17 16:28:12 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: fix bug on setting up system encoding
-
-	* ext/tk/lib/tk/event.rb: fix error on require process
-
-	* ext/tk/lib/tk/font.rb: fix abnormal termination error on Windows
-
-	* ext/tk/lib/tk/virtevent.rb: TkVirtualEvent::PreDefVirtEvent.new()
-	  accepts event-sequence arguments
-
-	* ext/tk/lib/tk/text.rb: fail to dump embedded images
-
-	* ext/tk/lib/tk/text.rb: tag_nextrange and tag_prevrange returns wrong
-	  types of values
-
-	* ext/tk/lib/tk/texttag.rb: nextrange and prevrange returns wrong
-	  types of values
-
-	* ext/tk/lib/tk/text.rb: add TkText::IndexModMethods module and
-	  TkText::IndexString class to treat text index modifiers
-
-	* ext/tk/lib/tk/texttag.rb: use TkText::IndexModMethods module
-
-	* ext/tk/lib/tk/textmark.rb: ditto
-
-	* ext/tk/lib/tk/textimage.rb: ditto
-
-	* ext/tk/lib/tk/textwindow.rb: ditto
-
-	* ext/tk/lib/tk/textimage.rb: wrong gravity of text mark for embedded
-	  image
-
-	* ext/tk/lib/tk/textwindow.rb: wrong gravity of text mark for
-	  embedded window
-
-Fri Dec 17 13:50:00 2004  Akiyoshi, Masamichi  <akiyoshi@hp.com>
-
-	* vms/vmsruby_private.c, vms/vmsruby_private.h: private routines
-	  for VMS port are added.
-
-	* eval.c (ruby_init): change to call VMS private intialization routine.
-
-Fri Dec 17 13:33:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session#initialize): control adding
-	  session_id hidden fields.  fixed: [ruby-talk:123850]
-
-Thu Dec 16 23:25:25 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb, lib/drb/ssl.rb: backported from CVS HEAD.
-	  [druby-ja:101]
-
-	* test/drb/test_drb.rb: adjust and reduce sleep (backported from
-	  CVS HEAD.)
-
-Thu Dec 16 18:44:58 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): should wait
-	  for reading request till data arrive. [ruby-talk:121068]
-
-	* lib/webrick/server.rb (WEBrick::GenericServer#start_thread):
-	  should log about all accepted socket. [ruby-core:03962]
-
-	* lib/webrick/accesslog.rb (WEBrick::AccessLog#setup_params):
-	  "%%" and "%u" are supported. [webricken:135]
-
-	* lib/webrick/httpservlet/filehandler.rb
-	  (WEBrick::HTTPServlet::FileHandler#check_filename):
-	  :NondisclosureName is acceptable if it is Enumerable.
-
-	* lib/webrick/config.rb (WEBrick::Config::FileHandler):
-	  default value of :NondisclosureName is [".ht*", "*~"].
-
-Thu Dec 16 18:36:52 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.c (ossl_raise): refine message format.
-
-Thu Dec 16 16:29:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/demos-en/widget: modify version check for
-	  supporting features
-
-Thu Dec 16 16:03:50 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/bindtag.rb: bug fix [ruby-talk: 123667]
-
-	* ext/tk/lib/tk/timer.rb: accept :idle for the interval argument
-
-	* ext/tk/lib/tk.rb: add TkComm._callback_entry?()
-
-	* ext/tk/lib/multi-tk.rb: add MultiTkIp.cb_entry_class
-
-	* ext/tk/lib/tk/canvas.rb: use TkComm._callback_entry?()
-
-	* ext/tk/lib/tk/canvastag.rb: ditto
-
-	* ext/tk/lib/tk/dialog.rb: ditto
-
-	* ext/tk/lib/tk/optiondb.rb: ditto
-
-	* ext/tk/lib/tk/text.rb: ditto
-
-	* ext/tk/lib/tk/texttag.rb: ditto
-
-	* ext/tk/lib/tk/textwindow.rb: ditto
-
-	* ext/tk/lib/tk/timer.rb: ditto
-
-	* ext/tk/lib/tk/validation.rb: ditto
-
-	* ext/tk/lib/tkextlib/*: ditto
-
-Thu Dec 16 03:14:28 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb (basic_encode): return value of pack('m') may
-	  include multiple CR/LFs.  Backported from main trunk (rev 1.112).
-	  [ruby-dev:25212]
-
-Thu Dec 16 00:33:37 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* hash.c (Init_Hash): remove custom "hash" and "eql?".
-
-Wed Dec 15 18:57:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/set.rb (Set::eql): wrong definition.  [ruby-dev:25207]
-
-Wed Dec 15 18:48:42 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* ext/curses/curses.c (window_subwin): call NUM2INT() before
-	GetWINDOW(). (backported from CVS HEAD)
-
-Wed Dec 15 17:03:50 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.[ch] (rb_w32_isatty): new function to replace MSVCRT's
-	  isatty because it never sets errno. (backported from CVS HEAD)
-
-Wed Dec 15 15:39:32 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_to_a): avoid SEGV
-	  (rollback the previous commit).
-
-Wed Dec 15 16:10:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (rb_obj_id_obsolete): warn always.
-
-	* eval.c (rb_enable_super): ditto.
-
-Wed Dec 15 15:31:02 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/set.rb (Set#==): [ruby-dev:25206]
-
-Wed Dec 15 14:22:10 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_fdisset): check whether the handle is valid.
-	  fixed: [ruby-core:03959]
-
-Wed Dec 15 10:30:37 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/openssl/ossl_digest.c (ossl_digest_initialize): [ruby-dev:25198]
-
-Tue Dec 14 17:10:09 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_close): need to reset osfhnd().
-
-Tue Dec 14 14:03:57 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.c (ossl_raise): avoid buffer overrun.
-	  [ruby-dev:25187]
-
-Tue Dec 14 12:36:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session::initialize): generate new
-	  session if given session_id does not exist.  [ruby-list:40368]
-
-Mon Dec 13 18:13:52 2004  Tanaka Akira  <akr@m17n.org>
-
-	* gc.c (stack_end_address): new function to obtain stack end address.
-	  stack_end_address calls __builtin_frame_address(0) to obtain the
-	  frame pointer of a stack frame of stack_end_address.  The address
-	  is the stack pointer of the caller's stack frame.
-	  (SET_STACK_END): use stack_end_address.
-	  This makes the conservative garbage collector to scan a stack frame
-	  of the garbage_collect function itself.  This is required because
-	  callee-save registers may be stored in the frame.
-	  [ruby-dev:25158]
-
-Mon Dec 13 00:58:02 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (cleanpath_aggressive): make it private.
-	  (cleanpath_conservative): ditto.
-	  Suggested by Daniel Berger.  [ruby-core:3914]
-
-Sun Dec 12 20:06:38 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: backported from CVS HEAD.
-
-Sun Dec 12 10:35:10 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/template/html/html.rb (RDoc::Page): Don't
-	  show an accessor's r/w flag if none was specified
-
-Sun Dec 12 10:14:03 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/rdoc.rb (RDoc::RDoc::parse_files): Never exclude files
-	  explicitly given on the command line.
-
-Sun Dec 11 23:54:07 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/*: update to support libraries in ActiveTcl8.4.12.0
-	  (see ext/tk/ChangeLog.tkextlib).
-
-	* ext/tk/sample/scrollframe.rb: add a new sample.
-
-Sat Dec 11 20:12:21 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: add DRbRemoteError. [ruby-list:40348],
-	  [ruby-list:40390]
-
-	* test/drb/drbtest.rb: ditto.
-
-	* test/drb/ut_drb.rb: ditto.
-
-Sat Dec 11 15:38:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/jcode.rb (String::succ): [ruby-dev:25156]
-
-Sat Dec 11 12:41:55 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* eval.c (run_trap_eval): prototype; avoid VC++ warnings.
-
-	* ext/socket/getaddrinfo.c: fix typo. fixed: [ruby-core:03947]
-
-	* win32/win32.c: need to include dln.h.
-
-Sat Dec 11 00:10:18 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_reopen): [ruby-dev:25150]
-
-Fri Dec 10 08:39:27 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/socket/socket.c (sock_listen): get OpenFile just before calling
-	  listen(2).  fixed: [ruby-dev:25149]
-
-Thu Dec  9 17:00:00 2004  Akiyoshi, Masamichi  <akiyoshi@hp.com>
-
-	* ext/socket/socket.c, ext/socket/getaddrinfo.c: port to VMS
-
-Thu Dec  9 16:31:02 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/sdbm/init.c (GetDBM): typo.
-
-Thu Dec  9 16:05:00 2004  Akiyoshi, Masamichi  <akiyoshi@hp.com>
-
-	* defines.h: change path of vms.h
-	* vms/vms.h: delete reference for snprintf()
-	* vms/config.h: new file
-	* vms/config.h_in: deleted
-
-Thu Dec  9 14:38:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (rb_str_inspect): escape # which starts an expression
-	  substitution.  fixed: [ruby-core:03922]
-
-	* string.c (rb_str_dump): not escape # which isn't a substitution.
-
-Thu Dec  9 10:54:36 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/dbm/dbm.c (fdbm_select): [ruby-dev:25132]
-
-	* ext/sdbm/init.c: ditto.
-
-	* ext/gdbm/gdbm.c: ditto.
-
-Thu Dec  9 03:08:36 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_init): set root-win title to "ruby" when
-	  the running script is '-e one-liner' or '-' (stdin).
-
-	* ext/tcltklib/extconf.rb: add find_library("#{lib}#{ver}",..) for
-	  stub libs
-
-	* ext/tk/lib/tk/textmark.rb: TkTextMarkCurrent and TkTextMarkAnchor
-	  have a wrong parent class.
-
-	* ext/tk/lib/tk/dialog.rb: rename TkDialog2 --> TkDialogObj and
-	  TkWarning2 --> TkWarningObj (old names are changed to alias names)
-
-	* ext/tk/lib/tk/dialog.rb: bug fix of treatment of 'prev_command'
-	  option and hashes for configuration
-
-	* ext/tk/lib/tk/dialog.rb: add TkDialogObj#name to return the
-	  button name
-
-	* ext/tk/lib/tk/radiobutton.rb: rename enbugged method value() ==>
-	  get_value() and value=(val) ==> set_value(val).
-
-	* ext/tk/lib/tk/menu.rb: add TkMenu.new_menuspec
-
-	* ext/tk/lib/tk/menu.rb: add alias (TkMenuButton = TkMenubutton,
-	  TkOptionMenuButton = TkOptionMenubutton)
-
-	* ext/tk/lib/tk/event.rb: new method aliases (same as option keys of
-	  event_generate) for Event object
-
-	* ext/tk/lib/tk/font.rb: configinfo returns proper types of values
-
-	* ext/tk/lib/tk.rb: bind methods accept subst_args + block
-
-	* ext/tk/lib/tk/canvas.rb: ditto
-
-	* ext/tk/lib/tk/canvastag.rb: ditto
-
-	* ext/tk/lib/tk/frame.rb: ditto
-
-	* ext/tk/lib/tk/text.rb: ditto
-
-	* ext/tk/lib/tk/texttag.rb: ditto
-
-	* ext/tk/lib/tk/toplevel.rb: ditto
-
-	* ext/tk/lib/tkextlib/*: ditto and bug fix
-
-Wed Dec  8 23:54:29 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/template/html/html.rb (RDoc::Page): Typo
-	  meant that h2 tag was invisible.
-
-Wed Dec  8 21:56:31 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss, test/rss, sample/rss: backported from CVS HEAD.
-
-Wed Dec  8 14:31:36 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_fwrite): change dereference for cosmetic reason.
-
-	* sprintf.c (rb_f_sprintf): [ruby-dev:25104]
-
-Tue Dec  7 19:08:00 2004  Akiyoshi, Masamichi  <akiyoshi@hp.com>
-
-	* io.c (io_fwrite): fix offset incrementation (for VMS and Human68k)
-
-Tue Dec  7 00:27:37 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (proc_setgroups): [ruby-dev:25081]
-
-Mon Dec  6 18:08:10 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* re.c (rb_reg_eqq): document fix.  [ruby-talk:122541]
-
-Mon Dec  6 17:19:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* rubysig.h (TRAP_BEG, TRAP_END): safe errno around CHECK_INTS.
-	  (backported from CVS HEAD)  [ruby-dev:24993]
-
-Mon Dec  6 10:18:17 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::look_for_directives_in):
-	  Oops - 1.8 doesn't have String#clear
-
-Mon Dec  6 09:59:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (sock_connect): use rb_str_new4().
-	  [ruby-dev:25052]
-
-Mon Dec  6 01:42:08 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_public_encrypt,
-	  ossl_rsa_public_decrypt, ossl_rsa_private_encrypt,
-	  ossl_rsa_private_decrypt): should take an optional argument
-	  to specify padding mode. [ruby-talk:122539]
-
-	* ext/openssl/ossl_pkey_rsa.c (Init_ossl_rsa): add new constants
-	  PKCS1_PADDING, SSLV23_PADDING, NO_PADDING and PKCS1_OAEP_PADDING
-	  under OpenSSL::PKey::RSA.
-
-	* test/openssl/test_pkey_rsa.rb: new file.
-
-Sun Dec  5 19:39:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Completion#complete): new parameter
-	  to direct case insensitiveness.
-
-	* lib/optparse.rb (OptionParser#order!): ignore case only for long
-	  option.  [ruby-dev:25048]
-
-Sat Dec  4 22:54:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_write): remove rb_str_locktmp().  [ruby-dev:25050]
-
-	* io.c (io_fwrite): takes VALUE string as an argument.
-	  [ruby-dev:25050]
-
-	* ext/socket/socket.c (sock_connect): remove rb_str_locktmp().
-	  [ruby-dev:25050]
-
-	* ext/socket/socket.c (udp_connect): [ruby-dev:25045]
-
-	* ext/socket/socket.c (udp_bind): ditto.
-
-	* ext/socket/socket.c (udp_send): ditto.
-
-	* ext/socket/socket.c (bsock_send): ditto.
-
-	* ext/socket/socket.c (s_recvfrom): ditto.
-
-	* hash.c (rb_hash_hash): should provide "hash" method where "eql?"
-	  is redefined.  [ruby-talk:122482]
-
-Sat Dec  4 14:54:52 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* eval.c (proc_invoke): use volatile `tmp' rather than `args'.
-	  [ruby-core:03882]
-
-Sat Dec  4 14:28:56 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/code_objects.rb (RDoc::Context::Section::set_comment):
-	  Section comments may now be bracketed by lines which are
-	  ignored. You can now write
-	      # -----------
-	      # :section: Dave's Section
-	      # comment material
-	      # -----------
-	   The lines before :section: are removed, and identical lines at the end are
-	   also removed if present.
-
-Sat Dec  4 03:33:45 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* ext/readline/readline.c: check $SAFE. (backported from CVS HEAD)
-
-	* test/readline/test_readline.rb: added tests for readline.
-	  (backported from CVS HEAD)
-
-Sat Dec  4 02:24:00 2004  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf.c: add constant NKF::VERSION
-
-	* ext/nkf/nkf.c(guess): this becomes an alias of guess2
-
-	* ext/nkf/test.rb(mime_out2): add --no-cp932
-
-	* ext/nkf/nkf-utf8/nkf.c: original nkf2 revision 1.47
-
-Sat Dec  4 00:35:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (bsock_setsockopt): [ruby-dev:25039]
-
-Fri Dec  3 18:57:03 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/ostruct.rb: 1.9 marshaling support back-ported.
-	  [ruby-core:03871]
-
-Fri Dec  3 13:45:20 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_invoke): copy arguments to frame.argv.
-	  [ruby-core:03861]
-
-Fri Dec  3 12:25:41 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* st.h: fix prototypes.
-
-Fri Dec  3 00:21:05 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (convert_type): use rb_respond_to() again.
-	  [ruby-dev:25021]
-
-	* eval.c (rb_respond_to): funcall respond_to? if it's redefined.
-	  [ruby-dev:25021]
-
-Fri Dec  3 01:55:24 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: widget configuration by TkWindow#method_missing
-	  returns proper object. "widget.option = val" returns val, and
-	  "widget.option(val)" returns self.
-
-	* ext/tk/lib/tk/font.rb: TkFont#replace accepts only one font argument.
-
-	* ext/tk/lib/tk/radiobutton.rb: add TkRadiobutton#value and
-	  TkRadiobutton#value=(val).
-
-	* ext/tk/lib/tk/spinbox.rb: callback substitution support on
-	  command option.
-
-	* ext/tk/sample/demos-en/widget: bug fix (wrong image height)
-
-	* ext/tk/sample/demos-jp/widget: ditto.
-
-Fri Dec  3 00:11:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_file_initialize): [ruby-dev:25032]
-
-Thu Dec  2 16:41:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_protect): prevent continuations created inside from being
-	  called from the outside.  [ruby-dev:25003]
-
-	* eval.c (rb_callcc, rb_cont_call): prohibit calling from different
-	  signal contexts.  [ruby-dev:25022]
-
-Thu Dec  2 09:57:24 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/ostruct.rb (OpenStruct::Marshaler): OpenStruct can be
-	  marshaled again.  [ruby-core:03862]
-
-Thu Dec  2 09:30:06 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (thread_mark): mark thread group.  [ruby-dev:25020]
-
-	* eval.c (thgroup_add): check whether the argument is really a Thread.
-
-Thu Dec  2 07:57:16 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_ctl): [ruby-dev:25019]
-
-Wed Dec  1 02:21:02 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* signal.c (sighandler): call handler immediately only for default
-	  handlers.  [ruby-dev:25003]
-
-Tue Nov 30 23:38:18 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (io_fread): need not to null terminate.  [ruby-dev:24998]
-
-	* io.c (read_all): remove unnecessary rb_str_resize().
-	  [ruby-dev:24996]  (backported from CVS HEAD)
-
-	* io.c (io_readpartial): ditto.
-
-	* io.c (io_read): ditto.
-
-Tue Nov 30 16:18:50 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_fread): need not to null terminate.  [ruby-dev:24998]
-
-	* io.c (read_all): remove unnecessary rb_str_resize().
-	  [ruby-dev:24996]
-
-	* io.c (io_read): ditto.
-
-Tue Nov 30 00:49:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_sysread): use temporary lock.  [ruby-dev:24992]
-
-Mon Nov 29 16:06:04 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_write): insufficiently filled string
-	  being extended when overwriting.  [ruby-core:03836]
-
-Mon Nov 29 15:59:05 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/ostruct.rb (OpenStruct::method_missing): check method
-	  duplication for -d.
-
-	* lib/ostruct.rb (OpenStruct::initialize): ditto.
-
-Mon Nov 29 15:22:28 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/io/nonblock/test_flush.rb: abandon tests when io/nonblock is
-	  not supported.
-
-Mon Nov 29 03:08:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (convert_type): direct call conversion methods for the
-	  performance.  [ruby-core:03845]
-
-	* eval.c (rb_funcall_rescue): new function.
-
-	* object.c (rb_Array): avoid using rb_respond_to().
-
-	* object.c (rb_Integer): ditto.
-
-	* parse.y (reduce_nodes): empty body should return nil.
-
-	* string.c (rb_str_aset): the original string should not be
-	  affected by modifying duplicated string.  [ruby-dev:24981]
-
-Mon Nov 29 13:57:38 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (CreateChild): search executable file if no program
-	  name given. (backported from CVS HEAD)
-
-Mon Nov 29 13:37:54 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (fptr_finalize): must not use FILE after fclose().
-	  [ruby-dev:24985]
-
-Mon Nov 29 13:16:31 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (CreateChild): push back the last space before next
-	  loop because CharNext() eats it.
-
-Mon Nov 29 01:18:18 2004  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (rb_io_check_writable): call io_seek regardless of
-	  NEED_IO_SEEK_BETWEEN_RW.  [ruby-dev:24986]
-
-Sat Nov 27 21:43:39 2004  Tanaka Akira  <akr@m17n.org>
-
-	* io.c: avoid data lost with nonblocking fd and
-	  stdio buffering in sync mode.  [ruby-dev:24966]
-	  based on matz's patch [ruby-dev:24967]
-	  (io_fwrite): new primitive writing function which writes
-	  directly if sync mode.
-	  (rb_io_fwrite): wrapper for io_fwrite now.
-	  (io_write): call io_fwrite instead of rb_io_fwrite.
-
-Sat Nov 27 14:44:15 2004  Kent Sibilev  <ksibilev@bellsouth.net>
-
-	* lib/cgi/session.rb (CGI::Session::initialize): create_new_id is
-	  now a instance method.  [ruby-core:03832]
-
-Sat Nov 27 09:41:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_fread): old rb_io_fread with file closing checking.
-	  (rb_io_fread): wrapper for io_fread now.
-	  [ruby-dev:24964]
-
-Fri Nov 26 18:02:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: Tk.destroy uses TkWindow#epath
-
-	* ext/tk/lib/tk/image.rb: bug fix
-
-	* ext/tk/lib/tk/wm.rb: add 'iconphoto' method(Windows only)
-
-	* ext/tk/lib/tkextlib/*: some methods uses TkWindow#epath
-
-Fri Nov 26 13:49:06 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (method_missing): raise TypeError for classes do not
-	  have allocators.  [ruby-core:03752]
-
-	* lib/erb.rb: add RDoc by James Edward Gray II.  [ruby-core:03786]
-
-Fri Nov 26 13:29:02 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::look_for_directives_in): Break
-	  out of preprocessing when we find a :section: directive (previously cleared out the
-	  comment, but this apparently now generates an error in gsub!)
-
-Fri Nov 26 00:17:40 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_read): move StringValue() check before GetOpenFile().
-	  [ruby-dev:24959]
-
-Thu Nov 25 20:14:57 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/thwait.rb (ThreadsWait#join_nowait): abnormally terminated
-	  threads should be also processed.  [ruby-talk:121320]
-
-Thu Nov 25 10:14:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (push_braces): do not reuse buffer strings.  [ruby-core:03806]
-
-Thu Nov 25 07:59:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (read_all): stringify non-nil buffer argument, and always
-	  taint the result.  [ruby-dev:24955]
-
-Wed Nov 24 01:01:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_read): integer conversion should be prior to
-	  GetOpenFile().  [ruby-dev:24952]
-
-	* configure.in, io.c: cancel [ ruby-Patches-1074 ].
-
-Tue Nov 23 08:09:50 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/menu.rb: improve usability of TkOptionMenubutton
-
-Tue Nov 23 02:00:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* file.c (rb_file_chown): integer conversion should be prior to
-	  GetOpenFile().  [ruby-dev:24949]
-
-Tue Nov 23 00:10:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* file.c (rb_file_chown): integer conversion should be prior to
-	  GetOpenFile().  [ruby-dev:24947]
-
-	* file.c (rb_file_truncate): ditto.
-
-	* file.c (rb_file_s_truncate): ditto.
-
-	* dir.c (dir_seek): use NUM2OFFT().
-
-	* misc/ruby-mode.el (ruby-non-block-do-re): should not match words
-	  start with block keyword and underscore.  [ruby-core:03719]
-
-Mon Nov 22 22:33:02 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::parse_require): Don't use names
-	  of variables or constants when oarsing 'require'
-
-Mon Nov 22 00:13:35 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (dir_seek): should retrieve dir_data after NUM2INT().
-	  [ruby-dev:24941]
-
-Sat Nov 20 23:57:33 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/README (et al): Add a new directive, :section:, and
-	  change the output format to accomodate. :section: allows to to
-	  group together methods, attributes, constants, etc under
-	  headings in the output. If used, a table of contents is
-	  generated.
-
-Sat Nov 20 23:56:54 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/options.rb (Options::parse): Force --inline-source if
-	  --one-file option given
-
-Sat Nov 20 23:55:19 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_splice): should place index wrapping after
-	  possible modification.  [ruby-dev:24940]
-
-Sat Nov 20 13:26:03 2004  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/utf8tbl.c: original revision 1.7
-
-Sat Nov 20 05:34:24 2004  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf/nkf-utf8/nkf.c: original nkf.c rev:1.40
-
-	* ext/nkf/test.rb: add test for mime encode/decode
-
-Sat Nov 20 01:37:34 2004  Johan Holmberg  <holmberg@iar.se>
-
-	* eval.c (error_print): nicer traceback at interrupt.
-	  [ruby-core:03774]
-
-Sat Nov 20 00:07:16 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_gsub): internal buffer should not be listed by
-	  ObjectSpace.each_object() by String#gsub.  [ruby-dev:24931]
-
-Fri Nov 19 01:20:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session::FileStore::initialize): raise
-	  exception if data corresponding to session specified from the
-	  client does not exist.
-
-Fri Nov 19 00:59:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_gsub): internal buffer should not be listed by
-	  ObjectSpace.each_object().  [ruby-dev:24919]
-
-Thu Nov 18 18:41:08 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
-
-	* test/ruby/test_stringchar.rb (test_bang): added.
-
-	* string.c (rb_str_upcase_bang, rb_str_capitalize_bang)
-	(rb_str_swapcase_bang): missing rb_str_modify().  [ruby-dev:24915]
-
-Thu Nov 18 00:21:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (proc_getpgrp): prohibit for $SAFE=2.
-	  [ruby-dev:24899]
-
-	* process.c (get_pid): ditto.  [ruby-dev:24904]
-
-	* process.c (get_ppid): ditto.
-
-	* array.c (rb_ary_delete): defer rb_ary_modify() until actual
-	  modification.  [ruby-dev:24901]
-
-Thu Nov 18 10:10:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c, rubyio.h (rb_io_modenum_flags): exported.
-
-	* ext/stringio/stringio.c (strio_initialize): allow Fixnum as mode as
-	  well as IO.new does.  [ruby-dev:24896]
-
-Wed Nov 17 23:42:40 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_settracefunc.rb: added.  [ruby-dev:24884]
-
-Wed Nov 17 13:56:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (newline_node): should not use FL_SET. [ruby-dev:24874]
-
-	* parse.y (string_content): should not use FL_UNSET.
-
-	* node.h (NODE_NEWLINE): remove unused bit to utilize flag field
-	  in nodes.
-
-Wed Nov 17 13:09:40 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* {bcc32,win32,wince}/Makefile.sub (test): should build ruby.exe
-	  before running test. [ruby-core:03756]
-
-Wed Nov 17 04:33:01 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* pack.c: all features are backport from 1.9. [ruby-dev:24826]
-
-	* bignum.c (rb_big2ulong_pack): new function to pack Bignums.
-
-Wed Nov 17 03:42:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_splice): move rb_str_modify() after
-	  StringValue(), which may alter the receiver.  [ruby-dev:24878]
-
-Tue Nov 16 23:45:07 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* numeric.c (flo_divmod): protect float values from GC by
-	  assignment to local variables.  [ruby-dev:24873]
-
-Tue Nov 16 16:30:21 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* {bcc32,win32,wince}/setup.mak (-epilogue-): remove config.h and
-	  config.status to force updating them.
-
-Tue Nov 16 16:20:45 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): position was ignored when a
-	  buffer was passed.  http://www.yo.rim.or.jp/~nov/d/?date=20041116#p03
-
-Tue Nov 16 11:19:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner::options): use
-	  Regexp conversion.
-
-Tue Nov 16 01:41:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_mod_check): frozen check should be separated.
-	  [ruby-core:3742]
-
-	* array.c (rb_ary_update): pedantic check to detect
-	  rb_ary_to_ary() to modify the receiver.  [ruby-dev:24861]
-
-Mon Nov 15 13:50:52 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_justify): typo fixed.  [ruby-dev:24851]
-
-Mon Nov 15 11:50:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-special-char-p, ruby-parse-partial): handle
-	  operator symbols.  [ruby-talk:120177]
-
-Sun Nov 14 13:27:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/pp.rb (PP#object_address_group): remove odd number of 'f'
-	  prefixed to negative address.
-
-Sun Nov 14 08:51:04 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/logger/test_logger.rb: Logger just expects
-	  Logger#datetime_format to be used for Time#strftime independently of
-	  locale. [ruby-dev:24828]
-
-Fri Nov 12 15:03:26 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* eval.c (ruby_options): now we cannot call rb_glob() before
-	  ruby_init(), so call rb_w32_cmdvector() at ruby_options().
-
-	* win32.{c,h} (rb_w32_cmdvector): rename make_cmdvector() and
-	  export it.
-
-Fri Nov 12 14:08:01 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/event.rb: remove $LOADED_FEATURES trick
-
-	* ext/tk/lib/tk.rb: ditto
-
-Fri Nov 12 00:31:05 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/gdbm/gdbm.c (fgdbm_store): StringValue() may alter string
-	  pointer.  [ruby-dev:24783]
-
-Thu Nov 11 17:36:12 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (rb_globi): also should call back via rb_glob_caller().
-	  [ruby-dev:24775]
-
-Thu Nov 11 16:47:21 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/ruby/test_file.rb (test_truncate_wbuf): we want to test
-	  only File#truncate, not behaviour of seek(2).
-
-Thu Nov 11 09:41:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (push_braces): was confusing VALUE and char*.
-
-	* dir.c (rb_push_glob): Dir.glob should have called its block.
-
-Thu Nov 11 01:52:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* error.c (syserr_initialize): use stringified object.
-	  [ruby-dev:24768]
-
-Wed Nov 10 22:49:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/delegate.rb (SimpleDelegator::dup): wrong number of
-	  arguments.
-
-	* lib/delegate.rb (DelegateClass::dup): ditto.
-
-Wed Nov 10 12:31:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* README.EXT (Example): extconf.rb is indispensable now.
-
-Wed Nov 10 03:33:36 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fix SEGV when compiled with Tcl/Tk8.3.x
-	  or older
-
-	* ext/tk/lib/tkextlib/tile/style.rb: bug fix
-
-Tue Nov  9 14:27:18 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Officious): moved from DefaultList.
-
-Tue Nov  9 01:05:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (rb_glob2): do not allocate buffer from heap to avoid
-	  memory leaks.  use string object for buffering instead.
-	  [ruby-dev:24738]
-
-	* dir.c (join_path): ditto.
-
-	* io.c (io_read): external input buffer may be modified even after
-	  rb_str_locktmp().  [ruby-dev:24735]
-
-	* dir.c (fnmatch): p or s may be NULL.  [ruby-dev:24749]
-
-Tue Nov  9 00:53:53 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* regex.c (slow_match): avoid GCC 3.4.x warnings.
-
-Tue Nov  9 00:50:06 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/rdoc.rb: Change version numbering of RDoc and ri
-
-Mon Nov  8 23:38:35 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/extservm.rb: add DRb::ExtServManager#uri=.
-	  [ruby-dev:24743]
-
-Mon Nov  8 22:20:19 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_class):
-	  Fix bug where parent class wasn't being detected if the
-	  child class was defined using the A::B notation.
-
-Mon Nov  8 00:14:13 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: add setup for mignw32 cross compiling.
-	  [ruby-talk:119413]
-
-Sun Nov  7 23:49:26 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: bind-event methods accept multi substitution
-	  arguments.
-
-	* ext/tk/lib/tk/canvas.rb: ditto.
-
-	* ext/tk/lib/tk/canvastag.rb: ditto.
-
-	* ext/tk/lib/tk/text.rb: ditto.
-
-	* ext/tk/lib/tk/texttag.rb: ditto.
-
-	* ext/tk/lib/tkextlib: ditto.
-
-Sat Nov  6 14:58:44 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/server.rb (WEBrick::HTTPServer#start): remove
-	  :DoNotReverseLookup option. (Socket#do_not_reverse_lookup is a
-	  ruby 1.9 feature)
-
-Sat Nov  6 11:31:04 2004  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date.rb (_parse): checks whether zone was given.
-
-Sat Nov  6 00:46:27 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_locktmp): check STR_TMPLOCK flag before
-	  locking.  [ruby-dev:24727]
-
-Fri Nov  5 18:12:42 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/scrollable.rb: divide Scrollable module into
-	  X_Scrollable and Y_Scrollable
-
-	* ext/tk/lib/tk/entry.rb: include X_Scrollable instead of Scrollable
-
-	* ext/tk/lib/tk/autoload.rb: define autoload for X_Scrollable and
-	  Y_Scrollable
-
-Fri Nov  5 16:05:32 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: TkComm._at() supprts both of "@x,y" and "@x"
-
-Fri Nov  5 13:22:58 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/text.rb: sorry. bug fix again.
-
-Fri Nov  5 13:17:54 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/text.rb: bug fix
-
-Fri Nov  5 08:52:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (gc_mark): stricter GC stack check.
-
-Fri Nov  5 08:52:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (gc_mark): stricter GC stack check.
-
-Fri Nov  5 08:34:43 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_gsub): should have removed rb_str_unlocktmp(str).
-	  [ruby-dev:24708]
-
-Thu Nov  4 21:25:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_gsub): string modify check no longer based on
-	  tmplock.  [ruby-dev:24706]
-
-Thu Nov  4 19:27:46 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* io.c (rb_f_open): fix typo.
-
-Thu Nov  4 15:02:14 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/variable.rb: forget to initialize instance_variables
-	  of TkVarAccess objects
-
-Thu Nov  4 09:11:35 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (gc_mark): enable GC stack checking.
-
-Thu Nov  4 03:11:33 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_gsub): lock strings temporarily.  [ruby-dev:24687]
-
-	* ext/socket/socket.c (s_recvfrom): tmplock input buffer.
-	  [ruby-dev:24705]
-
-Wed Nov  3 22:32:12 2004  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* process.c: On NetBSD don't use setruid() and setrgid().
-
-Wed Nov  3 22:24:17 2004  Daigo Moriwaki  <techml@sgtpepper.net>
-
-	* lib/webrick/httpauth/digestauth.rb: use Base64.encode64 to
-	  avoid warnings.
-
-Wed Nov  3 17:19:59 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_uniq_bang): do not push frozen string from hash
-	  table.  [ruby-dev:24695]
-
-	* array.c (rb_ary_and): ditto.
-
-	* array.c (rb_ary_or): ditto.
-
-Wed Nov  3 17:13:02 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* io.c (pipe_open): fix compile error
-
-Wed Nov  3 16:58:07 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: support to use different Tcl commands between
-	  configure and configinfo
-
-	* ext/tk/lib/font.rb: ditto.
-
-	* ext/tk/lib/itemconfig.rb: support to use different Tcl commands
-	  between item_configure and item_configinfo
-
-	* ext/tk/lib/itemfont.rb: ditto.
-
-	* ext/tk/extconf.rb: install SUPPORT_STATUS
-
-	* ext/tk/lib/tkextlib: some bug fixes (see ext/tk/ChangeLog.tkextlib)
-
-Wed Nov  3 16:30:41 2004  NARUSE, Yui  <naruse@ruby-lang.org>
-
-	* ext/nkf: follow nkf 2.0.4
-
-Wed Nov  3 15:53:34 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* test/rss/test_maker_*.rb: added tests for RSS Maker.
-
-	* lib/rss/maker.rb: added RSS Maker.
-
-	* lib/rss/maker/*.rb: ditto.
-
-Tue Nov  2 16:35:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/enumerator/enumerator.c (each_cons_i): pass copy of an
-	  internal consequent array.  [ruby-talk:118691]
-
-Tue Nov  2 16:05:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (rb_f_fork): need to flush stdout and stderr before
-	  fork(2).  [ruby-talk:117715]
-
-Tue Nov  2 01:20:09 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_invoke): nail down dyna_var node when Proc object
-	  or continuation is created.  [ruby-dev:24671]
-
-Mon Nov  1 13:59:28 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/extmk.rb (MANIFEST): do not use anymore, use extconf.rb instead.
-
-	* ext/enumerator/extconf.rb, ext/fcntl/extconf.rb,
-	  ext/stringio/extconf.rb: added.
-
-	* MANIFEST, ext/**/MANIFEST: removed.
-
-	* README.EXT, README.EXT.ja: remove MANIFEST stuff.
-
-Mon Nov  1 01:14:52 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_f_open): create copy of popen specifier.  [ruby-dev:24656]
-
-Mon Nov  1 00:36:48 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* main.c (_stklen): move to gc.c.
-
-Sun Oct 31 00:22:28 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_locktmp): lock string temporarily.
-
-	* string.c (str_independent): add tmplock check.
-
-	* io.c (io_write): lock output string temporarily.
-	  [ruby-dev:24649]
-
-	* io.c (io_write): use rb_str_locktmp().
-
-	* io.c (read_all): ditto.
-
-Sat Oct 30 06:53:24 2004  Peter Vanbroekhoven  <peter.vanbroekhoven@cs.kuleuven.ac.be>
-
-	* eval.c (rb_eval): NODE_XSTR should pass copy of literal string.
-
-Sat Oct 30 00:19:40 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* enum.c (enum_sort_by): protect continuation jump in.
-	  [ruby-dev:24642]
-
-Fri Oct 29 21:27:51 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_check_initialized): new function to check uninitialized
-	  object.  [ruby-talk:118234]
-
-	* file.c (rb_file_path), io.c (rb_io_closed): check if initialized.
-
-Fri Oct 29 10:00:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_thread_start_0): forget to free some memory chunks.
-	  [ruby-core:03611]
-
-	* eval.c (ruby_cleanup): ruby_finalize_1 may cause exception,
-	  should be wrapped by PUSH_TAG/POP_TAG().  [ruby-dev:24627]
-
-Thu Oct 28 08:42:02 2004  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (argf_forward): use ANSI style.
-	  (argf_read): call argf_forward with argv argument.
-	  [ruby-dev:24624]
-
-Thu Oct 28 23:32:54 2004  akira yamada  <akira@ruby-lang.org>
-
-	* ext/zlib/zlib.c (zstream_detach_input): resets klass of z->input if
-	  z->input isn't nil.
-
-Thu Oct 28 23:19:31 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: prefer relative path.  [ruby-talk:93037]
-
-Wed Oct 27 18:49:11 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* gc.c: prototype; rb_io_fptr_finalize() doesn't return any value
-	  at this version.
-
-Wed Oct 27 17:27:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (gc_sweep): recover ruby_in_compile variable.
-
-Wed Oct 27 09:17:30 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (str_gsub): use a string object for exception safeness.
-	  [ruby-dev:24601]
-
-Tue Oct 26 23:52:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_getline): rs modification check should not interfere in the loop.
-
-Tue Oct 26 23:30:39 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/code_objects.rb (RDoc::Context::add_class_or_module):
-	  Restore correct :nopdoc: behavior with nested classes and modules.
-
-Tue Oct 26 18:21:29 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (RESIZE_CAPA): check string attribute before modifying
-	  capacity member of string structure.  [ruby-dev:24594]
-
-Tue Oct 26 11:33:26 2004  David G. Andersen  <dga@lcs.mit.edu>
-
-	* ext/zlib/zlib.c (gzreader_gets): use memchr() to to gain
-	  performance.  [ruby-talk:117701]
-
-Tue Oct 26 10:56:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): raise ArgumentError for extra
-	  arguments, unless (digit)$ style used.
-
-Tue Oct 26 11:33:26 2004  David G. Andersen  <dga@lcs.mit.edu>
-
-	* ext/zlib/zlib.c (gzreader_gets): use memchr() to to gain
-	  performance.  [ruby-talk:117701]
-
-Tue Oct 26 10:56:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): raise ArgumentError for extra
-	  arguments, unless (digit)$ style used.
-
-Mon Oct 25 18:35:39 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* win32/win32.c (isUNCRoot): should check NUL after '.'.
-	  [ruby-dev:24590]
-
-	* win32/win32.c (isUNCRoot): fixed buffer overrun.
-
-Mon Oct 25 08:03:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (get_backtrace): ignore illegal backtrace.  [ruby-dev:24587]
-
-Sun Oct 24 00:41:09 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_load, search_required, rb_require_safe, rb_require): use
-	  frozen shared string to avoid outside modification.  [ruby-dev:24580]
-
-Sat Oct 23 22:18:32 2004  Guy Decoux  <ts@moulon.inra.fr>
-
-	* eval.c (frame_free): Guy Decoux solved the leak problem.
-	  Thanks.  [ruby-core:03549]
-
-Sat Oct 23 00:20:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/zlib/zlib.c (zstream_append_input): clear klass for z->input
-	  to avoid potential vulnerability.
-
-	* ext/zlib/zlib.c (zstream_run): always use zstream_append_input()
-	  to avoid SEGV.  [ruby-dev:24568]
-
-Fri Oct 22 12:02:28 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_alias): was warning for wrong condition.
-	  [ruby-dev:24565]
-
-Fri Oct 22 10:36:37 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#meta_vars):
-	  should check if path_info is not nil.
-
-Fri Oct 22 00:22:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/zlib/zlib.c (zstream_shift_buffer): should restore class
-	  field of a buffer.  [ruby-dev:24562]
-
-Fri Oct 22 00:20:33 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (rb_str_include): should not treat char as negative value.
-	  [ruby-dev:24558]
-
-Thu Oct 21 21:32:30 2004  IWATSUKI Hiroyuki  <don@na.rim.or.jp>
-
-	* lib/pstore.rb (PStore#transaction): Use the empty content when a
-	  file is not found.  [ruby-dev:24561]
-
-Thu Oct 21 19:06:15 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpresponse.rb (WEBrick::HTTPResponse#send_body_io):
-	  ensure to close @body. (http://bugs.debian.org/277520)
-
-Thu Oct 21 00:36:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_alias): should warn on method discarding.
-	  [ruby-dev:24546]
-
-	* ext/zlib/zlib.c (zstream_expand_buffer_into): hide internal
-	  string buffer by clearing klass.  [ruby-dev:24548]
-
-Wed Oct 20 19:45:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_gsub): 	reentrant check.  [ruby-dev:24432]
-
-	* backport all SEGV bug fixes from CVS HEAD.  [ruby-dev:24536]
-
-Wed Oct 20 04:17:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/dbm/dbm.c (fdbm_delete_if): should check if deleting element
-	  is a string.  [ruby-dev:24490]
-
-	* ext/sdbm/init.c (fsdbm_delete_if): ditto.
-
-Wed Oct 20 01:37:18 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_times): Array#* should return an instance of
-	  the class of right operand.  [ruby-dev:24526]
-
-	* ext/zlib/zlib.c (zstream_detach_buffer): should not expose
-	  class-less object to Ruby world. [ruby-dev:24530]
-
-	* eval.c (proc_dup): provide Proc#dup as well.  [ruby-talk:116915]
-
-	* eval.c (ruby_exec): stack marking position may be higher than
-	  expected.  thanks to Guy Decoux.  [ruby-core:03527]
-
-Tue Oct 19 22:43:12 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_attr): If
-	  we come across 'attr' in a context where it isn't
-	  followed by a symbol, just issue a warning.
-
-Tue Oct 19 20:41:37 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole.c(ole_invoke): retrieve the result value when
-	  retrying the IDispatch::invoke.
-
-Tue Oct 19 17:24:11 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (read_all): block string buffer modification during
-	  rb_io_fread() by freezing it temporarily. [ruby-dev:24479]
-
-	* dir.c (rb_push_glob): block call at once the end of method.
-	  [ruby-dev:24487]
-
-	* ext/enumerator/enumerator.c (enum_each_slice): remove
-	  rb_gc_force_recycle() to prevent potential SEGV.
-	  [ruby-dev:24499]
-
-	* ext/zlib/zlib.c (zstream_expand_buffer): hide internal string
-	  buffer by clearing klass.  [ruby-dev:24510]
-
-Tue Oct 19 16:12:18 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c: backport from CVS HEAD
-
-Tue Oct 19 08:54:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* intern.h, object.c (rb_class_inherited_p): export.
-
-Tue Oct 19 08:46:57 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (rb_str_upto): method result must be checked.  [ruby-dev:24504]
-
-	* eval.c (error_print): ditto.  [ruby-dev:24519]
-
-Mon Oct 18 23:37:05 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* marshal.c (r_object0): check inheritance by the internal function.
-	  [ruby-dev:24515]
-
-Mon Oct 18 15:58:01 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* range.c (range_step, range_each): need cast.
-
-Fri Oct 29 16:34:19 2004  Daiki Ueno  <ueno@unixuser.org>
-
-	* misc/ruby-mode.el (ruby-parse-partial): Parse the rest of the
-	line after opening heredoc identifier.  [ruby-dev:24635]
-
-Mon Oct 18 07:26:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (rb_file_truncate): discard read buffer before truncation.
-	  [ruby-dev:24197]
-
-Mon Oct 18 02:11:21 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/config.rb (WEBrick::Config::General): add default values:
-	  - WEBrick::Config[:DoNotReverseLookup]
-	  - WEBrick::Config[:RequestCallback] (it used as an alias of
-	    :RequestHandler in WEBrick::HTTPServer#run)
-	  - WEBrick::Config::FileHandler[:AcceptableLanguages]
-
-	* lib/webrick/httpservlet/filehandler.rb
-	  (WEBrick::HTTPServlet::FileHandler#set_filename): search files
-	  having suffix of language-name which Accept-Language header field
-	  includes if :AcceptableLanguages options is present.
-
-	* lib/webrick/httpservlet/filehandler.rb
-	  (WEBrick::HTTPServlet::FileHandler#get_servlet): new method to
-	  search servlet correspond to the suffix of filename.
-
-	* lib/webrick/httprequest.rb: add attributes access methods: accept,
-	  accept_charset, accept_encoding, accept_language, content_length
-	  and content_type.
-
-	* lib/webrick/httpresponse.rb: add attribute access methods:
-	  content_length, content_length=, content_type and content_type=.
-
-	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.mime_types):
-	  use the second suffix to detect media type. (the first suffix
-	  may be a language name.)
-
-	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.parse_qvalues):
-	  add method to parse Accept header field. it returns an Array of
-	  values sorted by the qvalues.
-
-Mon Oct 18 02:04:11 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#virtual_host): new
-	  method to register virtual hosting servers.
-
-	* lib/webrick/server.rb (WEBrick::GenericServer#accept): call
-	  do_not_reverse_lookup for each socket if :DoNotReverseLookup
-	  is set.  [ruby-core:02357]
-
-Mon Oct 18 00:42:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (sock_s_getservbyaname): protocol string
-	  might be altered.  [ruby-dev:24503]
-
-	* string.c (rb_str_upto): check if return value from succ is a
-	  string.  [ruby-dev:24504]
-
-Sun Oct 17 23:03:48 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/timer.rb: TkTimer#start and restart accept a block
-
-Sun Oct 17 13:05:04 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (fole_func_methods): correct argument mismatch.
-	* ext/win32ole/win32ole.c (fole_get_methods): ditto.
-	* ext/win32ole/win32ole.c (fole_put_methods): ditto.
-	* ext/win32ole/tests/testWIN32OLE.rb: add test for WIN32OLE#ole_func_methods
-	  WIN32OLE#ole_get_methods, WIN32OLE#ole_put_methods
-
-Sat Oct 16 14:45:28 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/0.9.rb (RSS::Rss#to_s): removed garbage.
-
-Sat Oct 16 13:42:49 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/: untabified.
-	* test/rss/: untabified.
-	* lib/rss/0.9.rb (RSS::Rss#to_s): inent -> indent.
-
-Sat Oct 16 13:34:56 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss: supported prety print.
-	* test/rss/test_1.0.rb: added test for calculating default indent size.
-
-Fri Oct 15 18:04:35 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/timer.rb: TkTimer.new(interval, loop){ ... } is
-	  acceptable. Add TkTimer.start ( == new + start ).
-
-Fri Oct 15 12:43:09 2004  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (Init_stack): make prototype declaration consistent with
-	  the definition in gc.c.
-
-Thu Oct 14 14:34:01 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* io.c (MODE_BINMODE, MODE_BINARY): fixed reversed condition.
-
-Thu Oct 14 13:33:59 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: added link to Tutorial.
-
-Mon Oct 11 13:48:20 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/*: untabify
-
-Sun Oct 10 12:32:08 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::parse_require): Allow 'require'
-	  to be used as a variable name
-
-Sat Oct  9 21:23:37 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/converter.rb: changed to try to use Iconv for default
-	  conversion.
-
-	* lib/rss/rss.rb: 0.0.9 -> 0.1.0.
-
-Sat Oct  9 19:50:36 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_getline): should not treat char as negative value.
-	  [ruby-dev:24460]
-
-Fri Oct  8 09:49:32 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (pack_pack): pointer modification check before each
-	  iteration.  [ruby-dev:24445]
-
-Fri Oct  8 01:13:05 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/optiondb.rb: make it more secure
-
-Thu Oct  7 23:47:57 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/scrollbar.rb: When 'set' operation, a scrollbar
-	  cannot propagate view port information from the source widget
-	  (that calls 'set') to other assigned widgets.
-
-Thu Oct  7 17:36:25 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: When CHILDKILLED and so on, Tk.errorCode returns
-	  a Fixnum for 2nd element (it's pid) of the return value.
-
-Thu Oct  7 12:55:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (io_read): should freeze buffer before thread context
-	  switch. [ruby-dev:24442]
-
-	* pack.c (pack_unpack): string conversion should at the top of the
-	  method.  [ruby-dev:24439]
-
-	* io.c (io_read): buffer should be frozen only after the length
-	  check.  [ruby-dev:24440]
-
-Thu Oct  7 02:56:43 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c: use FMODE_APPEND.
-
-Thu Oct  7 01:05:33 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: add Tk.errorInfo and Tk.errorCode
-
-Thu Oct  7 00:08:37 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_s_sysopen): preserve path in the buffer allocated by
-	  ALLOCA_N() to prevent modification.  [ruby-dev:24438]
-
-Wed Oct  6 09:21:00 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_mode_flags): preserve append mode flag.
-	  [ruby-dev:24436]
-
-	* io.c (rb_io_modenum_mode): do not use external output buffer.
-
-	* string.c (rb_str_justify): differ pointer retrieval to prevent
-	  padding string modification.  [ruby-dev:24434]
-
-	* range.c (range_each_func): allow func to terminate loop by
-	  returning RANGE_EACH_BREAK.
-
-	* range.c (member_i): use RANGE_EACH_BREAK. [ruby-talk:114959]
-
-Mon Oct  4 14:04:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_file_open_internal, rb_io_reopen): fname might be altered
-	  while GC.  [ruby-dev:24408]
-
-Mon Oct  4 12:53:45 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/optiondb.rb: support definition of command
-	  resources on widgets
-
-	* ext/tk/lib/tk/image.rb: bug fix
-
-Sun Oct  3 21:20:03 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (TEXT_REGEXP): allow 8-bit characters for the german
-	version of Microsoft Exchange Server. (backported from HEAD)
-
-	* lib/net/imap.rb (RTEXT_REGEXP): ditto.
-
-	* lib/net/imap.rb (CTEXT_REGEXP): ditto.
-
-Sat Oct  2 20:34:22 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* node.h (NEW_DVAR): extra semicolon.
-
-Sat Oct  2 00:42:20 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (r_byte): retrieve pointer from string value for each
-	  time.  [ruby-dev:24404]
-
-	* marshal.c (r_bytes0): ditto.
-
-	* enum.c (sort_by_i): re-entrance check added.  [ruby-dev:24399]
-
-	* io.c (io_read): should freeze all reading buffer.
-	  [ruby-dev:24400]
-
-	* string.c (rb_str_sum): should use bignums when bits is greater
-	  than or equals to sizeof(long)*CHAR_BITS. [ruby-dev:24395]
-
-	* eval.c (specific_eval): defer pointer retrieval to prevent
-	  unsafe sourcefile string modification.  [ruby-dev:24382]
-
-	* eval.c (specific_eval): defer pointer retrieval to prevent
-	  unsafe sourcefile string modification.  [ruby-dev:24382]
-
-	* string.c (rb_str_sum): wrong cast caused wrong result.
-	  [ruby-dev:24385]
-
-	* enum.c (enum_sort_by): hide temporary array from
-	  ObjectSpace.each_object.  [ruby-dev:24386]
-
-	* string.c (rb_str_sum): check was done with false pointer.
-	  [ruby-dev:24383]
-
-	* string.c (rb_str_sum): string may be altered.  [ruby-dev:24381]
-
-Mon Oct 11 17:51:34 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_popen): get mode string via rb_io_flags_mode() to
-	  avoid mode string modification.  [ruby-dev:24454]
-
-	* io.c (rb_io_getline_fast): should take delim as unsigned char to
-	  distinguish EOF and '\377'.  [ruby-dev:24460]
-
-	* io.c (rb_io_getline): add check for RS modification.
-	  [ruby-dev:24461]
-
-	* enum.c (enum_sort_by): use qsort() directly instead using
-	  rb_iterate().  [ruby-dev:24462]
-
-	* enum.c (enum_each_with_index): remove rb_gc_force_recycle() to
-	  prevent access to recycled object (via continuation for
-	  example).  [ruby-dev:24463]
-
-Fri Oct  1 11:40:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_f_eval): defer pointer retrieval to prevent unsafe
-	  sourcefile string modification.  [ruby-dev:24373]
-
-	* io.c (io_read): block string buffer modification during
-	  rb_io_fread() by freezing it temporarily. [ruby-dev:24366]
-
-	* io.c (rb_io_s_popen): mode argument may be altered.
-	  [ruby-dev:24375]
-
-	* file.c (rb_file_s_basename): ext argument may be altered.
-	  [ruby-dev:24377]
-
-	* enum.c (enum_sort_by): use NODE instead of 2 element arrays.
-	  [ruby-dev:24378]
-
-	* string.c (rb_str_chomp_bang): StringValue() may change the
-	  receiver.  [ruby-dev:24371]
-
-Fri Oct  1 11:25:20 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/grid.rb: revive TkGrid.grid
-
-	* ext/tk/lib/tk/pack.rb: revive TkPack.pack
-
-	* ext/tk/lib/tk/place.rb: revive TkPlace.place
-
-Thu Sep 30 00:50:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_init): bug fix
-
-	* ext/tk/tkutil.c (get_eval_string_core): accept a Regexp object
-
-	* ext/tk/lib/multi-tk.rb: fix bug on 'exit' operation
-
-	* ext/tk/lib/tk/text.rb: 'tksearch' accepts a Regexp object as a
-	  matting pattern argument
-
-Wed Sep 29 10:58:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* enum.c (sort_by_i): internally used object must not be changed
-	  outside.  [ruby-dev:24368]
-
-Mon Sep 27 13:46:45 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* intern.h, struct.c (rb_struct_s_members, rb_struct_members): public
-	  accessors.  [ruby-dev:24342]
-
-	* marshal.c (w_object, r_object0): use accessors.
-
-Mon Sep 27 09:14:03 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (s_accept): don't retry for EWOULDBLOCK.
-	  [ruby-talk:113807]
-
-Fri Sep 24 16:09:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_invoke): propagate DVAR_DONT_RECYCLE on termination
-	  to avoid double call to rb_gc_force_recycle(). [ruby-dev:24311]
-
-Fri Sep 24 08:29:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_subseq): original object might be modified after
-	  sharing data creation.  [ruby-dev:24327]
-
-	* array.c (rb_ary_replace): ditto.
-
-	* array.c (ary_make_shared): freeze shared array. [ruby-dev:24325]
-
-	* struct.c (struct_members): always check struct size and size of
-	  members list in the class.  [ruby-dev:24320]
-
-Thu Sep 23 09:29:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_sub_bang): check if string is not modified
-	  during iteration.  [ruby-dev:24315]
-
-	* hash.c (rb_hash_rehash): replace st_foreach() by its deep
-	  checking counterpart.  [ruby-dev:24310]
-
-Wed Sep 22 13:38:12 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* hash.c (rb_hash_rehash): add iteration check.  [ruby-dev:24301]
-
-	* st.c (st_foreach): add deep check.
-
-Wed Sep 22 13:06:14 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_call_handler): workaround for Ctrl-C.
-	  merge from HEAD.
-
-Wed Sep 22 00:11:12 2004  Dave Thomas  <dave@pragprog.com>
-
-	* process.c: Add documentation for fork()
-
-Wed Sep 22 09:04:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_collect_bang): element size might change during
-	  comparison.  [ruby-dev:24300]
-
-	* array.c (rb_ary_reject_bang): ditto. [ruby-dev:24300]
-
-	* array.c (rb_ary_eql): ditto. [ruby-dev:24300]
-
-Tue Sep 21 18:29:49 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_equal): merge miss.
-
-	* array.c (rb_ary_uniq_bang): element size might change during
-	  comparison.  [ruby-dev:24298]
-
-Mon Sep 20 00:24:19 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* enum.c (enum_sort_by): do not use qsort directly.  use
-	  rb_ary_sort_bang() instead.  [ruby-dev:24291]
-
-	* enum.c (enum_sort_by): pedantic type check added.
-	  [ruby-dev:24291]
-
-	* hash.c (rb_hash_foreach_iter): check iter_lev after each
-	  iteration.  [ruby-dev:24289]
-
-	* array.c (rb_ary_and): element size might change during
-	  comparison.  [ruby-dev:24290]
-
-	* array.c (rb_ary_or): ditto. [ruby-dev:24292]
-
-	* array.c (rb_ary_equal): wrong fix. [ruby-dev:24286]
-
-Sat Sep 18 15:02:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_equal): element size might change during
-	  comparison.  [ruby-dev:24254]
-
-	* array.c (rb_ary_diff): ditto. [ruby-dev:24274]
-
-	* array.c (rb_ary_select): ditto. [ruby-dev:24278]
-
-	* array.c (rb_ary_delete): ditto. [ruby-dev:24283]
-
-	* array.c (rb_ary_rindex): ditto. [ruby-dev:24275]
-
-	* array.c (rb_ary_initialize): element size might change during
-	  initializing block.  [ruby-dev:24284]
-
-Sat Sep 18 14:10:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (dir_s_chdir): avoid memory leak and unnecessary chdir to
-	  the original directory when exception has caused in changing
-	  direcotry or within block.  thanks to Johan Holmberg
-	  <holmberg@iar.se> [ruby-core:03446]
-
-Fri Sep 17 20:20:27 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mkdir_p): backport from CVS HEAD 1.45. [ruby-core:03420]
-
-Fri Sep 17 17:11:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_delete): element comparison might change array
-	  size. [ruby-dev:24273]
-
-	* file.c (rb_file_truncate): clear stdio buffer before truncating
-	  the file.  [ruby-dev:24191]
-
-	* ext/digest/digest.c: use rb_obj_class() instead of CLASS_OF
-	  which might return singleton class.  [ruby-dev:24202]
-
-Fri Sep 17 16:07:09 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: improve exit operation
-
-Fri Sep 17 15:01:57 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fix SEGV when (thread_)vwait or
-	  (thread_)tkwait
-
-	* ext/tk/lib/tk.rb: add alias wait_window to wait_destroy
-
-	* ext/tk/lib/multi-tk.rb: support calling 'mainloop' on slave
-	  interpreters (however, the 'real' eventloop must be run on the
-	  Default Master IP)
-
-	* ext/tk/lib/remote-tk.rb: follow the changes of ext/tk/lib/multi-tk.rb
-
-	* ext/tk/sample/remote-ip_sample2.rb: ditto
-
-	* ext/tk/sample/tkoptdb-safeTk.rb: ditto
-
-Thu Sep 16 18:12:32 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (WEBrick::CGI#start): should set REMOTE_USER
-	  to request.user attribute.
-
-	* lib/webrick/httpservlet/filehandler.rb
-	  (WEBrick::HTTPServlet::FileHandler#initialize): should expand
-	  the pathname of document root directory.
-
-Thu Sep 16 15:49:28 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_intern): protect string argument from GC.
-	  [ruby-core:03411]
-
-Wed Sep 15 20:22:23 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tkoptdb-safeTk.rb: fix a bug depend on the changes
-	  of MultiTkIp
-
-Tue Sep 14 23:54:11 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string was en-bugged by
-	  the previous changes.
-
-Tue Sep 14 23:45:44 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::TextFormatter::TextFormatter.for):
-	  Add Eric Hodel's simpleformatter.
-
-Tue Sep 14 16:59:37 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fix SEGV
-
-	* ext/tk/lib/multi-tk.rb: improve safe-level handling of argument proc
-
-	* ext/tk/sample/multi-ip_sample.rb: rename of old 'safe-tk.rb'
-
-	* ext/tk/sample/safe-tk.rb: new sample script
-
-Tue Sep 14 00:15:15 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/zlib/zlib.c: backported from HEAD.
-
-Mon Sep 13 19:16:33 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* eval.c (blk_copy_prev): need frame_dup().  [ruby-dev:24103]
-
-Mon Sep 13 16:23:27 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: MultiTkIp.new_master and new_slave accept
-	  safe-level value argument
-
-Mon Sep 13 10:20:45 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* object.c (nil_inspect): fix typo.
-
-Mon Sep 13 01:03:02 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: improve control of preserv/release tcltkip
-
-	* ext/tcltklib/tcltklib.c: store original 'exit' command
-
-	* ext/tk/tkutil.c: fix(?) SEGV
-
-Sun Sep 12 23:46:23 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* util.c (ruby_strdup): remove unnecessary code. (xmalloc never
-	  returns NULL.)
-
-	* util.c (ruby_getcwd): fix memory leak on failure.
-
-Sun Sep 12 02:41:58 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: add TclTkIp#allow_ruby_exit? and
-	  allow_ruby_exit=
-
-
-	* ext/tk/lib/multi-tk.rb: ditto.
-
-	* ext/tk/lib/remote-tk.rb: ditto.
-
-	* ext/tcltklib/MANUAL.euc: ditto.
-
-	* ext/tcltklib/MANUAL.eng: ditto.
-
-	* ext/tcltklib/tcltklib.c: fix some reasons of SEGV
-
-	* ext/tk/tkutil.c: ditto.
-
-	* ext/tk/lib/multi-tk.rb: ditto.
-
-	* ext/tk/lib/tk/timer.rb: ditto.
-
-Sat Sep 11 16:09:46 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb: Fix up cross-file class merging.
-
-Fri Sep 10 20:20:53 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tcltklib/tcltklib.c (lib_merge_tklist): fix suspicious
-	  pointer conversion.
-
-Fri Sep 10 02:43:54 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/template/kilmer.rb: James Buck's
-	  patch for call-seq.
-
-Thu Sep  9 13:58:56 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_init): change flag value for setting
-	  'argv' and 'argv0' variable
-
-	* ext/tk/lib/remote-tk.rb: follow changes of multi-tk.rb
-
-Thu Sep  9 11:46:18 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_classes): Allow
-	  spaces aroun parameter to define_method_under (James Buck)
-
-Wed Sep  8 18:44:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_write): zero fill a gap if exsts.
-	  [ruby-dev:24190]
-
-Wed Sep  8 15:19:49 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_init): cannot create a IP at level 4
-
-	* ext/tk/lib/multi-tk.rb: improve 'exit' operation, security check,
-	  and error treatment
-
-	* ext/tk/lib/multi-tk.rb: allow a trusted slave IP to create slave IPs
-
-	* ext/tk/lib/tk/listbox.rb: add TkListbox#value, value=, clear,	and
-	  erase
-
-	* ext/tk/lib/tk/text.rb: add TkText#clear and erase
-
-Tue Sep  7 15:17:49 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/socket/socket.c (ruby_connect): break immediately if a
-	  socket is non-blocking.  [ruby-talk:111654]
-
-Mon Sep  6 11:08:50 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tk/lib/tk/menu.rb(TkOptionMenubutton#insert): call correct method
-
-Mon Sep  6 11:00:47 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (dir_s_chdir): the patch to shut up false warning when
-	  exception occurred within a block.  a patch was given from Johan
-	  Holmberg <holmberg at iar.se>.  [ruby-core:03292]
-
-Mon Sep  6 07:51:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (cvar_cbase): singletons should refer outer cvar scope.
-	  [ruby-dev:24223]
-
-	* eval.c (rb_load): should preserve previous ruby_wrapper value.
-	  [ruby-dev:24226]
-
-Sat Sep  4 01:14:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (cvar_cbase): class variables cause SEGV in
-	  instance_eval() for fixnums and symbols. [ruby-dev:24213]
-
-Fri Sep  3 17:47:58 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* struct.c (make_struct): remove redefining constant when
-	  conflict.  [ruby-dev:24210]
-
-Fri Sep  3 11:31:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: Tk.after makes TkCore::INTERP.tk_cmd_tbl grow
-	  [ruby-dev:24207]
-
-Fri Sep  3 02:12:48 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fix typo [ruby-talk:111266]
-
-	* ext/tk/lib/tk/text.rb: fix typo
-
-	* ext/tk/lib/multi-tk.rb: improve safe-level treatment on slave IPs
-
-Fri Sep  3 01:54:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: already built-in libraries satisfy dependencies.
-	  [ruby-dev:24028]
-
-Thu Sep  2 11:36:20 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* eval.c (rb_obj_instance_eval): backported from HEAD.
-
-Wed Sep  1 21:18:25 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tk/lib/tk/spinbox.rb: fix typo
-
-Tue Aug 31 18:24:04 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/tk/tkutil.c (cbsubst_init): fix memory leak
-
-	* ext/tk/tkutil.c (cbsubst_get_all_subst_keys): fix SEGV
-
-Tue Aug 31 16:04:22 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_delete): when a tcltkip is deleted,
-	  destroy its root widget
-
-Tue Aug 31 12:30:36 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (del_root): fix SEGV
-
-Mon Aug 30 23:11:06 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_driver.rb (and others): ri now merges documentation
-	  if it finds the same class in multiple places.
-
-Mon Aug 30 22:40:30 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: 'restart' method accepts arguments
-
-Mon Aug 30 21:50:14 2004  Dave Thomas  <dave@pragprog.com>
-
-	* object.c: Add RDoc for Module.included.
-
-Mon Aug 30 15:10:46 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (GNU/k*BSD): fixed FTBFS on GNU/k*BSD. [ruby-dev:24051]
-
-Mon Aug 30 11:29:35 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (CreateChild): strip trailing spaces. [ruby-dev:24143]
-	  merge from HEAD.
-
-Sun Aug 29 14:08:56 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: compile error on bcc32 [ruby-dev:24081]
-
-	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string does not work
-
-Sat Aug 28 23:04:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (rb_big_and): protect parameters from GC.
-	  [ruby-talk:110664]
-
-Thu Aug 26 04:38:29 2004  Dave Thomas  <dave@pragprog.com>
-
-	* eval.c (return_jump): Minor typo in error message. Now reads
-	  "return can't jump across threads".
-
-Tue Aug 24 17:30:00 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session::FileStore#initialize): do not
-	  use a session id as a filename. (backported from HEAD)
-
-	* lib/cgi/session/pstore.rb (CGI::Session::PStore#initialize): ditto.
-
-	* lib/cgi/session/pstore.rb (CGI::Session::PStore#initialize): use
-	  Dir::tmpdir. (backported from HEAD)
-
-Tue Aug 24 14:40:16 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session::FileStore#initialize): untaint
-	  session id after check. (backported from HEAD)
-
-Tue Aug 24 09:09:01 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509attr.c (ossl_x509attr_initialize): d2i
-	  functions may replace the pointer indicated by the first argument.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509ext_initialize): ditto.
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_initialize): ditto.
-
-Mon Aug 23 14:04:51 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_read):
-	  - should return an empty string if specified length to read is 0.
-	  - should check for pending data and wait for fd before reading.
-	  - call underlying IO's sysread if SSL session is not started.
-	  [ruby-dev:24072], [ruby-dev:24075]
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_write):
-	  - call underlying IO's syswrite if SSL session is not started.
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_pending): new method
-	  OpenSSL::SSL#pending.
-
-	* ext/openssl/lib/openssl/buffering.rb: should not use select.
-
-Mon Aug 23 12:40:56 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/resolv.rb (Config.default_config_hash): when multiple domains
-	  are set, Win32::Resolv.get_resolv_info returns Array.
-
-Sun Aug 22 01:15:31 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_connect):
-	  should call :ProxyContentHandler before finishing CONNECT.
-
-Sat Aug 21 06:41:16 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/tcltklib/extconf.rb (find_tcl, find_tk): find stub library.
-
-	* lib/mkmf.rb (arg_config, with_config): deal with '-' and '_'
-	  uniformly.  [ruby-dev:24118]
-
-Thu Aug 19 16:29:45 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: Fail to treat a hash value of 'font' option.
-
-	* ext/tk/lib/tk.rb: bindinfo cannot return '%' substiturion infomation.
-
-	* ext/tk/lib/menu.rb: typo bug.
-
-Thu Aug 19 15:15:24 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* dir.c (free_dir): fix memory leak.  reported by yamamoto
-	  madoka.
-
-Thu Aug 19 11:00:00 2004  Akiyoshi, Masamichi  <masamichi.akiyoshi@hp.com>
-
-	* dln.c (dln_load): Modify to call lib$find_image_symbol for VMS.
-	* io.c (rb_io_fwrite): Use fputc() for VMS non-stream file.
-
-Thu Aug 19 06:07:45 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/token.c: re2c no longer compiled with bit vectors.  caused
-	  problems for non-ascii characters. [ruby-core:03280]
-	* ext/syck/implicit.c: ditto.
-	* ext/syck/bytecode.c: ditto.
-
-	* lib/yaml/baseemitter.rb: folding now handles double-quoted strings,
-	  fixed problem with extra line feeds at end of folding, whitespace
-	  opening scalar blocks.
-
-	* lib/yaml/rubytypes.rb: subtelties in handling strings with
-	  non-printable characters and odd whitespace patterns.
-
-Wed Aug 18 23:41:33 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/protocol.rb (rbuf_fill): OpenSSL::SSL::SSLSocket has its own
-	  buffer, select(2) might not work. [ruby-dev:24072]
-
-Wed Aug 18 17:10:12 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/tcltklib/stubs.c (ruby_tcltk_stubs): need to call
-	  Tcl_FindExecutable() for Tcl/Tk 8.4.
-
-Wed Aug 18 12:52:55 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_obj_instance_eval): evaluates under special singleton
-	  classes as for special constants.
-
-Tue Aug 17 17:20:59 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_reopen): should clear allocated OpenFile.  pointed
-	  out by Guy Decoux. [ruby-core:03288]
-
-Tue Aug 17 01:36:32 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/usage.rb: Remove extra indent. Tidy 'ri' option
-	  parsing so RDoc::usage plays better with OptionParser.
-
-Sat Aug 14 13:09:10 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb: backport from CVS HEAD (rev1.44).
-
-	* lib/fileutils.rb: cp_r should copy symlink itself, except cp_r
-	  root.
-
-	* lib/fileutils.rb: new option mv :force.
-
-	* lib/fileutils.rb: new module FileUtils::DryRun.
-
-Sat Aug 14 02:48:16 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/usage.rb: Added. Allows command line programs
-	  to report usage using their initial RDoc comment.
-
-Fri Aug 13 13:23:17 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.parse_range_header):
-	  fix regex for range-spec.
-
-	* lib/webrick/httpservlet/filehandler.rb
-	  (WEBrick::HTTPServlet::DefaultFileHandler#make_partial_content):
-	  multipart/byteranges response was broken.
-
-	* lib/webrick/httpservlet/erbhandler.rb
-	  (WEBrick::HTTPServlet::ERBHandler#do_GET): should select media type
-	  by suffix of script filename.
-
-	* lib/xmlrpc/server.rb: refine example code.
-
-Wed Aug 11 17:17:50 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (RPATHFLAG): stop setting RPATHFLAG on Interix.
-
-Sun Aug  8 00:43:31 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/implicit.c: added sexagecimal float#base60.
-
-	* ext/syck/rubyext.c (yaml_org_handler): ditto.
-
-	* lib/token.c: indentation absolutely ignored when processing flow
-	  collections.  plain scalars are trimmed if indentation follows in
-	  an ambiguous flow collection.
-
-Sat Aug  7 00:50:01 2004  Tanaka Akira  <akr@m17n.org>
-
-	* ext/zlib/zlib.c: Zlib::GzipReader#read(0) returns "" instead of nil.
-
-Tue Aug  3 13:49:20 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/namespace.rb: bug fix
-
-	* ext/tk/lib/tkextlib/treectrl/tktreectrl.rb: add Tk::TreeCtrl.loupe
-
-Mon Aug  2 18:04:21 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/msgcat.rb (set_translation): bug fix (fail to set
-	  trans_str to the same as src_str when trans_str is not given.)
-
-Mon Aug  2 11:53:06 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/code_objects.rb (RDoc::Context::find_symbol): Fix infinite recursion
-	  looking up some top level symbols (batsman)
-
-Mon Aug  2 11:48:29 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_methods): Allow '.'s in
-	  variable names to support SWIG generated files (Hans Fugal)
-
-Sat Jul 31 17:40:16 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-expr-beg, ruby-parse-partial,
-	  ruby-calculate-indent, ruby-move-to-block, ruby-forward-sexp,
-	  ruby-backward-sexp): keywords must match word-wise.
-
-Sat Jul 31 05:47:37 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml.rb (YAML::load_file, YAML::parse_file): added.
-
-	* lib/yaml/rubytypes.rb: exceptions were using an older
-	  YAML.object_maker. [ruby-core:03080]
-
-	* ext/syck/token.c (sycklex_yaml_utf8): using newline_len to
-	  handline CR-LFs.  "\000" was showing up on folded blocks which
-	  stopped at EOF.
-
-	* ext/syck/token.c: re2c compiled with bit vectors now.
-	* ext/syck/implicit.c: ditto.
-	* ext/syck/bytecode.c: ditto.
-
-Fri Jul 30 16:10:54 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (lib_fromUTF8_core): raise ArgumentError when
-	  the unknown encoding name is given.
-
-	* ext/tcltklib/tcltklib.c (lib_toUTF8_core): ditto.
-
-	* ext/tk/lib/tk.rb (Tk::Encoding.encoding_convertfrom): bug fix.
-
-	* ext/tk/lib/tk.rb (Tk::Encoding.encoding_convertto): ditto.
-
-Wed Jul 28 18:59:17 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::initialize): remove at_exit code for CGI_PARAMS
-	  and CGI_COOKIES.  they will no longer be used.
-
-Wed Jul 28 01:04:44 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (run_final): wrong order of data. [ruby-dev:23984]
-
-Tue Jul 27 07:05:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_eval): copy on write for argument local variable
-	  assignment.
-
-	* eval.c (assign): ditto.
-
-	* eval.c (rb_call0): update ruby_frame->argv with the default
-	  value used for the optional arguments.
-
-	* object.c (Init_Object): "===" calls rb_obj_equal() directly.
-	  [ruby-list:39937]
-
-Mon Jul 26 11:22:55 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.escape): should
-	  escape space.
-
-Sun Jul 25 11:05:21 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* win32/win32.{h,c} (rb_w32_{f,fd,fs}open): workaround for bcc32's
-	  {f,fd,fs}open bug. set errno EMFILE and EBADF. [ruby-dev:23963]
-
-Sat Jul 24 13:32:47 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* range.c (rb_range_beg_len): returns Qnil only when "beg" points
-	  outside of a range.  No boundary check for "end".
-
-Fri Jul 23 16:40:25 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (define_final): should not disclose NODE* to Ruby world.
-	  [ruby-dev:23957]
-
-Fri Jul 23 09:03:16 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (disconnected?): new method. (backported from HEAD)
-
-Thu Jul 22 16:41:54 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi/session.rb (CGI::Session::FileStore#update): sets the
-	 permission of the session data file to 0600.
-
-	* lib/cgi/session/pstore.rb (CGI::Session::Pstore#initialize):
-	  ditto.
-
-Thu Jul 22 00:02:21 2004  Masahiro Kitajima  <katonbo@katontech.com>
-
-	* process.c (rb_f_system): not need to call last_status_set() any
-	  longer on _WIN32.
-
-Tue Jul 20 09:15:17 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* test/fileutils/test_fileutils.rb: File.link raises EINVAL on BeOS.
-
-Mon Jul 19 01:15:07 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpservlet/cgihandler.rb
-	  (WEBrick::HTTPServlet::CGIhandler#do_GET): set SystemRoot environment
-	  variable to CGI process on Windows native platforms. [ruby-dev:23936]
-
-	* lib/webrick/httpservlet/cgihandler.rb
-	  (WEBrick::HTTPServlet::CGIhandler#do_GET): use $?.exitstatus and
-	  refine log message.
-
-Sun Jul 18 16:14:29 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/msgcat.rb (TkMsgCatalog.callback): bug fix
-	  ( wrong number of argument )
-
-Sun Jul 18 08:13:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): remove extra sign digit.
-
-Sun Jul 18 03:21:42 2004  Akinori MUSHA  <knu@iDaemons.org>
-
-	* dir.c (range): use NULL instead of 0.
-
-	* dir.c (range): get rid of a gcc 3.4 warning.
-
-Sun Jul 18 03:12:11 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (receive_responses): return if a LOGOUT response
-	received. (backported from HEAD)
-	* lib/net/imap.rb (send_string_data): wait command continuation
-	requests before sending octet data of literals. (backported from HEAD)
-
-Sat Jul 17 23:54:59 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/variable.rb: TkVariable#ref returns a TkVariable object
-
-Sat Jul 17 22:04:44 2004  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/ldap.rb: method hierarchical? should be in URI::LDAP.
-
-Sat Jul 17 18:29:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (stmt): not to show same error messages twice.
-
-Sat Jul 17 13:13:32 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/irb/ruby-lex.rb (RubyLex::identify_string): %s string do not
-	  process expression interpolation.  [ruby-talk:106691]
-
-Sat Jul 17 05:26:27 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/diagram.rb: Incorporate Micheal Neuman's
-	  client-side imagemao patch
-
-Sat Jul 17 01:57:03 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (THREAD_ALLOC): th->thread should be initialized to NULL.
-	  [ruby-talk:106657]  The solution was found by Guy Decoux.
-
-Fri Jul 16 22:30:28 2004  Michael Neumann  <mneumann@ntecs.de>
-
-	* file.c (rb_stat_dev_major): new methods File::Stat#dev_major and
-	  #dev_minor. [ruby-core:03195]
-
-Fri Jul 16 15:23:53 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (return_jump, break_jump): raise unexpceted local jump
-	  exception directly.  [ruby-dev:23740]
-
-	* lib/base64.rb (Deprecated): super in bound method calls original
-	  name method in stable version.  [ruby-dev:23916]
-
-Fri Jul 16 11:31:49 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/test/unit/ui/{fox,gtk,gtk2}/testrunner.rb: remove
-	  garbage (patch from akira yamada) [ruby-dev:23911]
-
-Fri Jul 16 11:20:00 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): fix output of NaN, Inf and -Inf with
-	  "%f" or etc on MSVCRT platforms. (backported from HEAD)
-
-Fri Jul 16 11:17:38 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* error.c (exit_initialize): use EXIT_SUCCESS instead of 0.
-	  [ruby-dev:23913]
-
-	* error.c (exit_success_p): new method SystemExit#success?.
-	  [ruby-dev:23912]
-
-	* error.c (syserr_initialize): initialization for subclasses.
-	  [ruby-dev:23912]
-
-Thu Jul 15 23:53:38 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser#warn, OptionParser#abort): Exception
-	  no longer has to_str method.
-
-Thu Jul 15 22:59:48 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* ext/readline/extconf.rb: added dir_config for curses, ncurses,
-	termcap. (backported from HEAD)
-
-Thu Jul 15 20:29:15 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* class.c, error.c, eval.c, intern.h, object.c, variable.c:
-	  do not set path if it is a singleton class.  [ruby-dev:22588]
-	  (backport from 1.9)
-
-Thu Jul 15 10:15:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/, ext/tcltklib/: bug fix
-
-	* ext/tk/lib/tk.rb: better operation for SIGINT when processing
-	  callbacks.
-	* ext/tk/lib/tk/msgcat.rb: ditto.
-	* ext/tk/lib/tk/variable.rb: ditto.
-	* ext/tk/lib/tk/timer.rb: ditto.
-
-	* ext/tk/lib/tk/validation.rb: add Tk::ValidateConfigure.__def_validcmd
-	  to define validatecommand methods easier
-
-	* ext/tk/lib/tk.rb (_genobj_for_tkwidget): support autoload Tk ext
-	  classes
-
-	* ext/tk/lib/tk/canvas.rb and so on: remove the parent widget type
-	  check for items (e.g. canvas items; depends on the class) to
-	  avoid some troubles on Tk extension widget class definition.
-
-	* ext/tk/lib/tkextlib/: add Iwidget and TkTable extension support
-
-	* ext/tk/sample/tkextlib/: add samples of Iwidget and TkTable
-
-
-Wed Jul 14 18:08:37 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1cons_to_der): fix type of
-	  argument. [ruby-dev:23891]
-
-	* test/openssl/test_x509store.rb: prune tests for CRL checking
-	  unless X509::V_FLAG_CRL_CHECK is defined.
-
-Wed Jul 14 12:29:07 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* util.c (ruby_strtod): should not convert string in the form of
-	  "-I.FE-X" which both "I" and "F" are ommitted. [ruby-dev:23883]
-
-	* test/ruby/test_float.rb (test_strtod): add test for bug fix.
-
-Wed Jul 14 00:31:15 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* array.c: rdoc patch. merged patch from Johan Holmberg
-	  <holmberg@iar.se> [ruby-core:3170]
-
-Tue Jul 13 19:39:12 2004  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/generic.rb (URI::Generic#merge_path):
-	  "URI('http://www.example.com/foo/..') + './'" should return
-	  "URI('http://www.example.com/')".  [ruby-list:39838]
-	  "URI('http://www.example.com/') + './foo/bar/..'" should return
-	  "URI('http://www.example.com/foo/')".  [ruby-list:39844]
-
-	* test/uri/test_generic.rb (TestGeneric#test_merge): added tests.
-
-Tue Jul 13 15:51:45 2004  Akinori MUSHA  <knu@iDaemons.org>
-
-	* lib/mkmf.rb (init_mkmf): Do not add $(libdir) to $LIBPATH in
-	   extmk mode.
-
-	* lib/mkmf.rb (dir_config): Prepend a new library path instead of
-	  appending so it is tried first.
-
-Tue Jul 13 00:50:48 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb: Support call-seq: for Ruby files.
-
-Mon Jul 12 21:20:36 2004  Dave Thomas  <dave@pragprog.com>
-
-	* html_generator.rb: Support hyperlinks of the form {any text}[xxx]
-	  as well as stuff[xxx]
-
-Sat Jul 10 09:30:24 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/marshal/test_struct.rb: use qualified build-tin class name
-	  (::Struct) to avoid name crash.
-
-Sat Jul 10 04:21:56 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: better operation for SIGINT when processing
-	  callbacks.
-	* ext/tk/lib/tk/msgcat.rb: ditto.
-	* ext/tk/lib/tk/variable.rb: ditto.
-	* ext/tk/lib/tk/timer.rb: ditto.
-
-	* ext/tk/lib/tk/validation.rb (__def_validcmd):  add a module
-	  function of Tk::ValidateConfigure to define validatecommand
-	  methods easier
-
-Fri Jul  9 22:36:36 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* array.c, enum.c, pack.c: rdoc patch from Johan Holmberg
-	  <holmberg@iar.se> [ruby-core:3132] [ruby-core:3136]
-
-	* numeric.c: rdoc patch.
-
-Fri Jul  9 19:26:39 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::HTTPS#proxy_open): raise ArgumentError to
-	  notice https is not supported.
-
-Fri Jul  9 14:28:54 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_raise): accept third argument as well as
-	  Kernel#raise, and evaluate the arguments to create an exception in
-	  the caller's context.  [ruby-talk:105507]
-
-Fri Jul  9 01:47:08 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib : bug fix
-	* ext/tk/lib/tkextlib/itcl : add [incr Tcl] support
-	* ext/tk/lib/tkextlib/itk  : add [incr Tk] support
-	* ext/tk/lib/tkextlib/iwidgets : midway point of [incr Widgets] support
-	* ext/tk/sample/tkextlib/iwidgets : very simple examples of
-	  [incr Widgets]
-
-Thu Jul  8 22:52:19 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/{rss,parser,0.9,1.0,2.0}.rb: supported RSS 0.9x/2.0
-	  validation and validation which disregard order of elements.
-	* test/rss/test_parser.rb: added tests for RSS 0.9x/2.0
-	  validation.
-	* test/rss/{test_trackback,rss-testcase}.rb: fixed no good method
-	  name.
-
-Thu Jul  8 00:05:23 2004  akira yamada  <akira@ruby-lang.org>
-
-	* lib/tempfile.rb (Tempfile::initialize): got out code of
-	  generating tmpname.  [ruby-dev:23832][ruby-dev:23837]
-
-Wed Jul  7 15:53:14 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* string.c (rb_str_match): raise TypeError when both arguments are
-	  strings. [ruby-dev:22869] (backported from HEAD)
-
-	* string.c (rb_str_match2): removed.
-
-	* Makefile.in, bcc32/Makefile.sub, win32/Makefile.sub,
-	  wince/Makefile.sub (string.c): now not depend on version.h.
-
-Wed Jul  7 00:48:34 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/tk/lib/tkextlib/tktrans.rb,
-	  ext/tk/lib/tkextlib/treectrl.rb: fix syntax errors.
-
-Tue Jul  6 18:38:45 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib : improve framework of developping Tcl/Tk extension
-	  wrappers
-
-Mon Jul  5 23:56:42 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/{trackback,syndication,dublincore,content}.rb: worked
-	  with ruby 1.6 again.
-
-	* test/rss/rss-assertions.rb: ditto.
-
-Mon Jul  5 22:54:39 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/uri/common.rb (Kernel#URI): new global method for parsing URIs.
-
-Mon Jul  5 09:02:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_yield, rb_f_catch): 4th argument to rb_yield_0()
-	  is a set of bit flags.  [ruby-dev:23859]
-
-Mon Jul  5 01:27:32 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* lib/drb/drb.rb(DRbConn self.open): If socket pool is full, close
-	  the socket whose last-access-time is oldest. (and add new one)
-	  [ruby-dev:23860]
-
-Sun Jul  4 12:24:50 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* lib/rss/rss.rb: added copyright header.
-
-Sun Jul  4 00:24:40 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* added files
-	  * lib/soap/attachment.rb
-	  * lib/soap/header
-	  * lib/soap/mimemessage.rb
-	  * lib/soap/rpc/httpserver.rb
-	  * lib/wsdl/soap/cgiStubCreator.rb
-	  * lib/wsdl/soap/classDefCreator.rb
-	  * lib/wsdl/soap/classDefCreatorSupport.rb
-	  * lib/wsdl/soap/clientSkeltonCreator.rb
-	  * lib/wsdl/soap/driverCreator.rb
-	  * lib/wsdl/soap/mappingRegistryCreator.rb
-	  * lib/wsdl/soap/methodDefCreator.rb
-	  * lib/wsdl/soap/servantSkeltonCreator.rb
-	  * lib/wsdl/soap/standaloneServerStubCreator.rb
-	  * lib/wsdl/xmlSchema/enumeration.rb
-	  * lib/wsdl/xmlSchema/simpleRestriction.rb
-	  * lib/wsdl/xmlSchema/simpleType.rb
-	  * lib/xsd/codegen
-	  * lib/xsd/codegen.rb
-	  * sample/soap/authheader
-	  * sample/soap/raa2.4
-	  * sample/soap/ssl
-	  * sample/soap/swa
-	  * sample/soap/whois.rb
-	  * sample/soap/calc/samplehttpd.conf
-	  * sample/soap/exchange/samplehttpd.conf
-	  * sample/soap/sampleStruct/samplehttpd.conf
-	  * sample/wsdl/raa2.4
-	  * sample/wsdl/googleSearch/samplehttpd.conf
-	  * test/openssl/_test_ssl.rb
-	  * test/soap/header
-	  * test/soap/ssl
-	  * test/soap/struct
-	  * test/soap/swa
-	  * test/soap/wsdlDriver
-	  * test/wsdl/multiplefault.wsdl
-	  * test/wsdl/simpletype
-	  * test/wsdl/test_multiplefault.rb
-
-	* modified files
-	  * lib/soap/baseData.rb
-	  * lib/soap/element.rb
-	  * lib/soap/generator.rb
-	  * lib/soap/marshal.rb
-	  * lib/soap/netHttpClient.rb
-	  * lib/soap/parser.rb
-	  * lib/soap/processor.rb
-	  * lib/soap/property.rb
-	  * lib/soap/soap.rb
-	  * lib/soap/streamHandler.rb
-	  * lib/soap/wsdlDriver.rb
-	  * lib/soap/encodingstyle/handler.rb
-	  * lib/soap/encodingstyle/literalHandler.rb
-	  * lib/soap/encodingstyle/soapHandler.rb
-	  * lib/soap/mapping/factory.rb
-	  * lib/soap/mapping/mapping.rb
-	  * lib/soap/mapping/registry.rb
-	  * lib/soap/mapping/rubytypeFactory.rb
-	  * lib/soap/mapping/wsdlRegistry.rb
-	  * lib/soap/rpc/cgistub.rb
-	  * lib/soap/rpc/driver.rb
-	  * lib/soap/rpc/element.rb
-	  * lib/soap/rpc/proxy.rb
-	  * lib/soap/rpc/router.rb
-	  * lib/soap/rpc/soaplet.rb
-	  * lib/soap/rpc/standaloneServer.rb
-	  * lib/wsdl/data.rb
-	  * lib/wsdl/definitions.rb
-	  * lib/wsdl/operation.rb
-	  * lib/wsdl/parser.rb
-	  * lib/wsdl/soap/definitions.rb
-	  * lib/wsdl/xmlSchema/complexContent.rb
-	  * lib/wsdl/xmlSchema/complexType.rb
-	  * lib/wsdl/xmlSchema/data.rb
-	  * lib/wsdl/xmlSchema/parser.rb
-	  * lib/wsdl/xmlSchema/schema.rb
-	  * lib/xsd/datatypes.rb
-	  * lib/xsd/qname.rb
-	  * sample/soap/calc/httpd.rb
-	  * sample/soap/exchange/httpd.rb
-	  * sample/soap/sampleStruct/httpd.rb
-	  * sample/soap/sampleStruct/server.rb
-	  * sample/wsdl/amazon/AmazonSearch.rb
-	  * sample/wsdl/amazon/AmazonSearchDriver.rb
-	  * sample/wsdl/googleSearch/httpd.rb
-	  * test/soap/test_basetype.rb
-	  * test/soap/test_property.rb
-	  * test/soap/test_streamhandler.rb
-	  * test/soap/calc/test_calc.rb
-	  * test/soap/calc/test_calc2.rb
-	  * test/soap/calc/test_calc_cgi.rb
-	  * test/soap/helloworld/test_helloworld.rb
-	  * test/wsdl/test_emptycomplextype.rb
-	  * test/wsdl/axisArray/test_axisarray.rb
-	  * test/wsdl/datetime/test_datetime.rb
-	  * test/wsdl/raa/test_raa.rb
-	  * test/xsd/test_xmlschemaparser.rb
-	  * test/xsd/test_xsd.rb
-
-	* summary
-	  * add SOAP Header mustUnderstand support.
-
-	  * add HTTP client SSL configuration and Cookies support (works
-	    completely with http-access2).
-
-	  * add header handler for handling sending/receiving SOAP Header.
-
-	  * map Ruby's anonymous Struct to common SOAP Struct in SOAP Object
-	    Model.  it caused error.
-
-	  * add WSDL simpleType support to restrict lexical value space.
-
-	  * add SOAP with Attachment support.
-
-Sat Jul  3 17:19:44 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/tk/lib/tkextlib/tkDND.rb: fix syntax error.
-
-Thu Jul  1 23:15:29 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/pstore.rb (transaction): safer backup scheme.  [ruby-list:39102]
-
-	* lib/pstore.rb (commit_new): use FileUtils.copy_stream for Cygwin.
-	  [ruby-dev:23157]
-
-	* lib/pstore.rb (transaction): allow overriding dump and load.
-	  [ruby-dev:23567]
-
-	* lib/pstore.rb (PStore#transaction): get rid of opening in write mode
-	  when read only transaction.  [ruby-dev:23842]
-
-	* lib/yaml/store.rb: follow lib/pstore.rb's change.
-
-Thu Jul  1 18:36:08 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tcltklib : bug fix
-
-	* ext/tk/lib/tk : bug fix and add Tcl/Tk extension support libraries
-
-Thu Jul  1 11:59:45 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: check for EVP_CIPHER_CTX_copy, ENGINE_add,
-	  EVP_CIPHER_CTX_set_padding, EVP_CipherFinal_ex, EVP_CipherInit_ex,
-	  EVP_DigestFinal_ex and EVP_DigestInit_ex.
-
-	* ext/openssl/openssl_missing.c (EVP_CIPHER_CTX_copy): new function.
-
-	* ext/openssl/openssl_missing.h (EVP_DigestInit_ex, EVP_DigestFinal_ex,
-	  EVP_CipherInit_ex, EVP_CipherFinal_ex, HMAC_Init_ex): new macro for
-	  OpenSSL 0.9.6.
-
-	* ext/openssl/ossl_cipher.c (ossl_cipher_encrypt, ossl_cipher_decrypt):
-	  re-implemnt (the arguments for this method is ).
-
-	* ext/openssl/ossl_cipher.c (ossl_cipher_pkcs5_keyivgen): new method
-	  OpenSSL::Cipher::Cipher#pkcs5_keyivgen. it calls EVP_BytesToKey().
-
-	* ext/openssl/ossl_cipher.c (ossl_cipher_alloc, ossl_cipher_initialize,
-	  ossl_cipher_copy, ossl_cipher_reset ossl_cipher_final,
-	  ossl_cipher_set_key, ossl_cipher_set_iv): replace all EVP_CipherInit
-	  and EVP_CipherFinal into EVP_CipherInit_ex and EVP_CipherFinal_ex.
-	  and EVP_CIPHER_CTX_init should only be called once.
-
-	* ext/openssl/ossl_cipher.c (ossl_cipher_set_key_length): new method
-	  OpenSSL::Cipher::Cipher#key_len=.
-
-	* ext/openssl/ossl_cipher.c (ossl_cipher_init_deprecated): new
-	  finction; print warning for Cipher#<<.
-
-	* ext/openssl/ossl_digest.c: replace all EVP_DigestInit and
-	  EVP_DigestFinal into EVP_DigestInit_ex and EVP_DigestFinal_ex.
-	  and EVP_MD_CTX_init should only be called once.
-
-	* ext/openssl/ossl_digest.c (digest_final): should call
-	  EVP_MD_CTX_cleanup to avoid memory leak.
-
-	* ext/openssl/ossl_hmac.c (ossl_hmac_initialize): repalce HMAC_init
-	  into HMAC_init_ex. and HMAC_CTX_init is moved to ossl_hmac_alloc.
-
-	* ext/openssl/ossl_hmac.c (hmac_final): should call
-	  HMAC_CTX_cleanup to avoid memory leak.
-
-	* test/openssl/test_cipher.rb, test/openssl/test_digest.rb,
-	  test/openssl/test_hmac.rb: new file.
-
-Thu Jul  1 04:08:30 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (ossl_i2d_ASN1_TYPE, ossl_ASN1_TYPE_free):
-	  workaround for the versions earlier than OpenSSL-0.9.7.
-
-Thu Jul  1 03:33:55 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_pkey_dh.c (ossl_dh_initialize): should create
-	  empty pkey object if no argument is passed. [ruby-talk:103328]
-
-	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_initialize): ditto.
-
-	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_initialize): ditto.
-
-	* ext/openssl/ossl_pkey_dh.c: add new methods: OpenSSL::PKey::DH#p,
-	  OpenSSL::PKey::DH#p=, OpenSSL::PKey::DH#g, OpenSSL::PKey::DH#g=,
-	  OpenSSL::PKey::DH#pub_key, OpenSSL::PKey::DH#pub_key=,
-	  OpenSSL::PKey::DH#priv_key and OpenSSL::PKey::DH#priv_key=.
-
-	* ext/openssl/ossl_pkey_dsa.c: add new methods: OpenSSL::PKey::DSA#p,
-	  OpenSSL::PKey::DSA#p=, OpenSSL::PKey::DSA#q, OpenSSL::PKey::DSA#q=,
-	  OpenSSL::PKey::DSA#g, OpenSSL::PKey::DSA#g=,
-	  OpenSSL::PKey::DSA#pub_key, OpenSSL::PKey::DSA#pub_key=,
-	  OpenSSL::PKey::DSA#priv_key and OpenSSL::PKey::DSA#priv_key=.
-
-Thu Jul  1 03:16:09 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_read): take optional second argument
-	  to specify a string to be written.
-
-	* ext/openssl/lib/openssl/buffering.rb (OpenSSL::Buffering#read):
-	  take optional second argument to specify a string to be written.
-
-	* ext/openssl/lib/openssl/buffering.rb (OpenSSL::Buffering#gets):
-	  refine regexp for end-of-line.
-
-	* ext/opnessl/lib/openssl/ssl.rb
-	  (OpenSSL::SSL::SocketForwarder#listen): fix typo.
-
-Wed Jun 30 11:38:51 2004  Mikael Brockman  <phubuh@phubuh.org>
-
-	* parse.y (primary): should not be NULL.  [ruby-core:03098]
-
-Wed Jun 30 02:53:24 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/rubyext.c (syck_emitter_new): set buffer after
-	  Data_Wrap_Struct to avoid possible GC. [ruby-talk:104835]
-
-Tue Jun 29 10:31:19 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval_cmd, rb_thread_trap_eval): restore safe level.
-
-	* gc.c (define_final, run_final): preserve and restore safe level for
-	  finalizers.  [ruby-core:03058]
-
-	* signal.c (signal_exec, rb_trap_exit, trap): preserve and restore
-	  safe level for signal handlers.  [ruby-dev:23829]
-
-Mon Jun 28 14:57:56 2004  Jeff Mitchell  <quixoticsycophant@yahoo.com>
-
-	* configure.in, lib/mkmf.rb (LIBPATHFLAG): use double quotes due to
-	  DOSISH compilers.  [ruby-core:03107]
-
-Mon Jun 28 00:30:19 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* sample/drb/*.rb: using 'DRb.thread.join' instead of 'gets'
-
-Sun Jun 27 22:39:51 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* sample/rss/tdiary_plugin/rss-recent.rb: supported Hiki.
-
-Sun Jun 27 12:19:46 2004  Kouhei Sutou  <kou@cozmixng.org>
-
-	* {lib,sample,test}/rss: added RSS Parser. [ruby-dev:23780]
-
-Sat Jun 26 11:07:30 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (aix): -b must come at the start of the command line,
-	  and -e must not appear while testing libraries.  [ruby-talk:104501]
-
-	* lib/mkmf.rb (dir_config): quote directory names if necessary.
-	  [ruby-talk:104505]
-
-Fri Jun 25 15:33:19 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/extconf.rb: check stricter.  [ruby-talk:104501]
-
-	* ext/iconv/extconf.rb: include iconv.h for libiconv.  [ruby-dev:22715]
-
-Fri Jun 25 08:31:29 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_thread_atfork): remove "fork terminates thread"
-	  warning.  [ruby-dev:23768]
-
-	* object.c (rb_obj_clone): backport FL_FINALIZE patch from 1.9.
-	  [ruby-core:02786][ruby-core:03067]
-
-	* ext/socket/socket.c (sock_sockaddr): Socket#gethostbyname()
-	  should give us packed address, not struct sockaddr.
-	  [ruby-core:03053]
-
-Fri Jun 25 02:04:23 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* {bcc32,win32,wince}/setup.mak: remove RUBY_EXTERN lines when
-	  including version.h. [ruby-talk:104456] (backported from HEAD)
-
-Thu Jun 24 14:23:29 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_fread): return already read data when system call is
-	  interrupted.  [ruby-talk:97206]
-
-Thu Jun 24 01:32:43 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* version.h: added declarations of ruby_version,
-	  ruby_release_date, ruby_platform.
-	  (backported from HEAD)
-
-Wed Jun 23 22:23:37 2004  Dave Thomas  <dave@pragprog.com>
-
-	* ext/socket/socket.c (sock_s_gethostbyaddr): Work around problem
-	  with OS X not returning 'from' parameter to recvfrom for
-	  connection-oriented sockets.
-
-Wed Jun 23 01:45:27 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RubyLex::identify_quotation):
-	  Fix problem with the 'r' being dropped from %r{xxx}
-
-Wed Jun 23 00:20:20 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/win32ole/win32ole.c (ole_hresult2msg): remove trailing
-	  CRs and LFs. (doesn't depend on CR+LF) [ruby-dev:23749]
-
-Wed Jun 23 00:00:25 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_initialize): should check fcntl result.  [ruby-dev:23742]
-
-Tue Jun 22 21:11:36 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (OLE_FREE): should not call CoFreeUnuse-
-	  dLibraries().
-
-	* ext/win32ole/win32ole.c (ole_event_free): ditto.
-
-	* ext/win32ole/win32ole.c (ole_hresult2msg): truncate error message
-	  before CR.
-
-Tue Jun 22 16:47:42 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/ftp.rb (MDTM_REGEXP): fix for demon's ftp server.
-	Thanks, Rutger Nijlunsing.
-
-Mon Jun 21 10:19:23 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_opendir): use FindFirstFile()/FindNextFile()/
-	  FindClose() instead of _findfirst()/_findnext()/_findclose().
-	  merge from HEAD.
-
-Sat Jun 19 13:24:15 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (method_call): allow changing $SAFE.  [ruby-dev:23713]
-
-Fri Jun 18 23:12:22 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (proc_save_safe_level, rb_set_safe_level, safe_setter): limit
-	  safe level.
-
-Wed Jun 16 23:05:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* object.c (rb_mod_freeze): prepare string representation before
-	  freezing. [ruby-talk:103646]
-
-Wed Jun 16 16:04:40 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* object.c (rb_mod_le): singleton class inherits Class rather than its
-	  object's class.  [ruby-dev:23690]
-
-Wed Jun 16 16:01:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* gc.c (stack_grow_direction): memoize the direction.
-
-	* gc.c (Init_stack): should always move to end of VALUE.
-
-Tue Jun 15 12:10:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: bug fix (TkWindow#grab)
-
-Mon Jun 14 18:23:27 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/remote-tk.rb: bug fix
-
-Sun Jun 13 00:23:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/extconf.rb: [EXPERIMENTAL] MacOS X (darwin) support
-
-	* ext/tcltklib/tcltklib.c: fix thread trouble on callback proc, and
-	  eliminate warning about instance variable access
-
-	* ext/tk/lib/tk/menubar.rb: improve supported menu_spec
-
-	* ext/tk/lib/tk/menuspec.rb: [add] menu_spec support library
-
-	* ext/tk/lib/tk/root.rb: add menu_spec support
-
-	* ext/tk/lib/tk/text.rb: bug fix
-
-	* ext/tk/lib/tk/toplevel.rb: add menu_spec support
-
-	* ext/tk/sample/menubar?.rb: [add] sample of menu_spec usage
-
-Sat Jun 12 11:15:53 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (target_os): strip -gnu suffix on Linux.
-
-Fri Jun 11 17:08:21 2004  Akinori MUSHA  <knu@iDaemons.org>
-
-	* config.guess: Restore a wrongly removed hyphen.
-
-Fri Jun 11 14:30:08 2004  Akinori MUSHA  <knu@iDaemons.org>
-
-	* config.guess: Attempt to avoid system name change on
-	  Darwin platforms also.
-
-Fri Jun 11 14:22:45 2004  Akinori MUSHA  <knu@iDaemons.org>
-
-	* config.guess, config.sub: Attempt to avoid system name change on
-	  Linux platforms.  We have been using "linux" instead of
-	  "linux-gnu" on this branch.
-
-Thu Jun 10 19:19:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/sdbm/init.c (fsdbm_store): sdbm should use StringValue().
-	  [ruby-talk:103062]
-
-Wed Jun  9 18:04:14 2004  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/generic.rb (URI::Generic::merge,
-	  URI::Generic::route_from): accepts non-hierarchical URI.
-	  [ruby-dev:23631]
-
-	* test/uri/test_generic.rb (TestGeneric::test_route,
-	  TestGeneric::test_merge): added tests for above changes.
-
-Wed Jun  9 17:39:37 2004  Akinori MUSHA  <knu@iDaemons.org>
-
-	* config.guess, config.sub: Update to a more recent version as of
-	  2004-01-20.
-
-	* configure.in: Add support for DragonFly BSD.
-
-Wed Jun  2 20:16:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (str_new4): should share shared instance if it already
-	  exists.  [ruby-dev:23665]
-
-Wed Jun  2 12:41:53 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_gets_m): set lastline ($_) even when read line is
-	  nil.  [ruby-dev:23663]
-
-Fri May 28 11:20:31 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval): bad influence on frame node.
-
-	* eval.c (eval): reverted wrongly removed condition.  [ruby-dev:23638]
-
-Thu May 27 23:15:18 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb: leading 0 padding of timestamp usec part.
-
-	* lib/csv.rb (CSV.parse): [CAUTION] behavior changed.  in the past,
-	  CSV.parse accepts a filename to be read-opened (it was just a
-	  shortcut of CSV.open(filename, 'r')).  now CSV.parse accepts a
-	  string or a stream to be parsed e.g.
-	  CSV.parse("1,2\n3,r") #=> [['1', '2'], ['3', '4']]
-
-	* lib/csv.rb: CSV::Row and CSV::Cell are deprecated.  these classes
-	  are removed in the future.  in the new csv.rb, row is represented
-	  as just an Array.  since CSV::Row was a subclass of Array, it won't
-	  hurt almost all programs except one which depended CSV::Row#match.
-	  and a cell is represented as just a String or nil(NULL).  this
-	  change will cause widespread destruction.
-
-	      CSV.open("foo.csv", "r") do |row|
-		row.each do |cell|
-		  if cell.is_null       # using Cell#is_null
-		    p "(NULL)"
-		  else
-		    p cell.data         # using Cell#data
-		  end
-		end
-	      end
-
-	    must be just;
-
-	      CSV.open("foo.csv", "r") do |row|
-		row.each do |cell|
-		  if cell.nil?
-		    p "(NULL)"
-		  else
-		    p cell
-		  end
-		end
-	      end
-
-	* lib/csv.rb: [CAUTION] record separator(CR, LF, CR+LF) behavior
-	  change.  CSV.open, CSV.parse, and CSV,generate now do not force
-	  opened file binmode.  formerly it set binmode explicitly.
-
-	  with CSV.open, binmode of opened file depends the given mode
-	  parameter "r", "w", "rb", and "wb".  CSV.parse and CSV.generate open
-	  file with "r" and "w".
-
-	  setting mode properly is user's responsibility now.
-
-	* lib/csv.rb: accepts String as a fs (field separator/column separator)
-	  and rs (record separator/row separator)
-
-	* lib/csv.rb (CSV.read, CSV.readlines): added.  works as IO.read and
-	  IO.readlines in CSV format.
-
-	* lib/csv.rb: added CSV.foreach(path, rs = nil, &block).  CSV.foreach
-	  now does not handle "| cmd" as a path different from IO.foreach.
-	  needed?
-
-	* test/csv/test_csv.rb: updated.
-
-	* test/ruby/test_float.rb: added test_strtod to test Float("0").
-
-Thu May 27 21:37:50 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#initialize): refine pathname initialization
-	  by pathname.
-
-Thu May 27 20:22:05 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* io.c (rb_io_fwrite): check all case errno != 0 [ruby-dev:23648]
-
-Thu May 27 14:53:13 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* io.c (rb_io_fwrite): workaround for bcc32's fwrite bug.
-	  add errno checking.  [ruby-dev:23627]
-
-Wed May 26 14:19:42 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval, eval): make line number consistent on eval with
-	  Proc.  [ruby-talk:101253]
-
-Wed May 26 13:59:17 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::skip_for_variable): Allow for
-	  'do' after for statement
-
-Wed May 26 13:56:03 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb (Generators::MarkUp::style_url): Fix
-	  relative path to code CSS file
-
-Wed May 26 13:14:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_init_copy): copy also positions.  [ruby-talk:100910]
-
-Wed May 26 00:00:00 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/syck.c (syck_new_parser): clear parser on init.
-	  thanks, ts. [ruby-core:02931]
-
-	* ext/syck/token.c (sycklex_yaml_utf8): buffer underflow.
-	  thanks, ts. [ruby-core:02929]
-
-	* lib/yaml/baseemitter.rb (indent_text): simpler flow block code.
-
-	* lib/yaml.rb: added rdoc to beginning of lib.
-
-Mon May 24 10:46:26 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
-
-	* lib/rdoc/generators/template/html/html.rb: SYSTEM identifiers
-	  must be absolute URIs
-
-Sat May 22 12:00:04 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* MANIFEST: add new encodings in rexml.
-
-	* ext/tk/MANIFEST: add recent files.
-
-Sat May 22 05:37:11 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/remote-tk.rb: (NEW library) controll Tk interpreters
-	  on the other processes by Tcl/Tk's 'send' command
-
-Fri May 21 09:22:05 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method_parameters):
-	  Add ()'s around parameters that don't have them
-
-Thu May 20 17:02:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (check_sizeof): define result size.  [ruby-core:02911]
-
-	* lib/mkmf.rb (create_header): macro name should not include equal
-	  sign.
-
-Thu May 20 15:59:50 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* ext/socket/socket.c: fix SEGV. [ruby-dev:23550]
-
-Thu May 20 14:35:52 2004  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/socket.c: check SCM_RIGHTS macro addition to
-	  the msg_control field to test existence of file descriptor passing
-	  by msg_control.
-
-Thu May 20 12:38:06 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* numeric.c (flo_eq): always check if operands are NaN.
-	  [ruby-list:39685]
-
-Thu May 20 12:34:39 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_visibility):
-	  At Ryan Davis' suggestion, honor visibility modifers if guarded by a
-	  statement modifier
-
-Thu May 20 12:22:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (have_type): do not check pointer to incomplete type,
-	  which always get compiled.  [ruby-list:39683]
-
-Wed May 19 11:09:00 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: change permition of TkObject#tk_send from
-	  private to public
-
-Tue May 18 14:00:46 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* node.h (NEW_DSTR): adjust list length.
-
-	* parse.y (literal_concat): ditto.
-
-Mon May 17 16:14:25 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* numeric.c (flo_to_s): it's preferable that "p 0.0" outputs "0.0"
-	  instead of "0.0e+00". [ruby-dev:23480]
-
-	* numeric.c (flo_to_s): it's preferable that "p 0.00000000000000000001"
-	  outputs "1.0e-20" instead of "9.999999999999999e-21". (the precision
-	  is considered, but there is assumption DBL_DIG == 15 in current
-	  implementation)
-
-Mon May 17 10:13:33 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (setup_domain_and_type): honor duck typing.
-	  [ruby-dev:23522]
-
-	* ext/socket/socket.c (sock_s_getnameinfo): ditto.
-
-Mon May 17 01:15:23 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml.rb: removed fallback to pure Ruby parser.
-
-	* lib/yaml/baseemitter.rb (indent_text): was forcing a mod value
-	  of zero at times, which kept some blocks from getting indentation.
-
-	* lib/yaml/baseemitter.rb (node_text): rewriting folded scalars.
-
-	* ext/syck/syck.h: reports style of scalars now, be they plain, block
-	  single-, or double-quoted.
-
-	* ext/syck/syck.c: ditto.
-
-	* ext/syck/gram.c: ditto.
-
-	* ext/syck/node.c: ditto.
-
-	* ext/syck/token.c: ditto.
-
-	* ext/syck/rubyext.c (yaml_org_handler): symbols loaded only
-	  if scalar style is plain.
-
-	* ext/syck/rubyext.c (yaml_org_handler): some empty strings were
-	  loaded as symbols.
-
-	* test/yaml/test_yaml.rb (test_perl_regexp): updated test to
-	  match new regexp serialization.
-
-Mon May 17 00:03:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/drb/drb.rb: Cosmetic documentation changes.
-
-Sun May 16 22:36:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/test/unit.rb: Removed :nodoc: directive (it prevented effective
-	  RDoc operation), and added file-level comment.
-
-Sun May 16 20:55:49 2004  Tanaka Akira  <akr@m17n.org>
-
-	* ext/dbm/dbm.c (fdbm_initialize): accept optional 3rd argument to
-	  specify an open flag.
-	  (Init_dbm): define open flags: DBM::READER, DBM::WRITER, DBM::WRCREAT
-	  and DBM::NEWDB.
-
-Sun May 16 13:10:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/test/unit/**/*.rb: Removed :nodoc: directives (many were
-	  generating warnings, many were on private methods).
-
-Sat May 15 01:41:34 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (eval): forgot to restore $SAFE value before evaluating
-	  compiled node.  [ruby-core:02872]
-
-Sat May 15 01:33:12 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* range.c (range_each_func): terminates loop if generating value
-	  is same to @end.  [ruby-talk:100269]
-
-Fri May 14 22:08:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_new4): should not reuse frozen shared string if
-	  the original is not an instance of String. [ruby-talk:100193]
-
-Fri May 14 18:39:25 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/canvas.rb: improve coords support for canvas items.
-	  Now, supports all of the followings.
-	    TkcLine.new(c, 0, 0, 100, 100, :fill=>'red')
-	    TkcLine.new(c, [0, 0, 100, 100], :fill=>'red')
-	    TkcLine.new(c, [0, 0], [100, 100], :fill=>'red')
-	    TkcLine.new(c, [[0, 0], [100, 100]], :fill=>'red')
-	    TkcLine.new(c, :coords=>[0, 0, 100, 100], :fill=>'red')
-	    TkcLine.new(c, :coords=>[[0, 0], [100, 100]], :fill=>'red')
-
-Fri May 14 12:11:43 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* util.c (ruby_strtod): strtod("0", &end); => end should point '\0'.
-	 [ruby-dev:23498]
-
-Thu May 13 15:47:30 2004  akira yamada  <akira@ruby-lang.org>
-
-	* lib/net/telnet.rb (Net::Telnet::login): "options" can specify
-	  regexps for login prompt and/or password prompt.
-
-Thu May 13 14:23:45 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* hash.c (delete_if_i): use st_delete_safe() (via
-	  rb_hash_delete()) instead of returning ST_DELETE.
-	  backport from HEAD.  [ruby-dev:23487]
-
-Thu May 13 13:01:30 2004  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/mailto.rb (URI::MailTo::to_s): should include fragment.
-
-Thu May 13 11:04:08 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* pack.c (pack_pack): always add with null for 'Z'.
-
-	* pack.c (pack_unpack): terminated by null for 'Z'.  [ruby-talk:98281]
-
-Wed May 12 19:59:43 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (have_type, check_sizeof): replace unusable characters.
-	  [ruby-talk:99788]
-
-Wed May 12 17:41:42 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/resolv.rb (Resolv::DNS::Config): make it configurable without
-	  external file such as /etc/resolv.conf.
-
-Wed May 12 14:37:27 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509name.c: attribute value of DC (short name of
-	  domainComponent) should be IA5String.
-
-Wed May 12 13:20:19 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk/composite.rb: improve configure methods (based on
-	  the proposal of [ruby-talk:99671]).
-
-Wed May 12 11:51:08 2004  Dave Thomas  <dave@pragprog.com>
-
-	* class.c (rb_obj_singleton_methods): fix rdoc
-
-Mon May 10 21:44:42 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Change scheme for
-	  looking up symbols in  HTML generator.
-
-Mon May 10 16:45:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (eval): warning during eval should not cause deadlock.
-	  [ruby-talk:98651]
-
-	* eval.c (rb_eval): raise TypeError exception for superclass
-	  mismatch.  [ruby-list:39567]
-
-Mon May 10 12:11:37 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Hack to search parents
-	  for unqualified constant names.
-
-Mon May 10 12:11:37 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Hack to search parents
-	  for unqualified constant names.
-
-Sun May  9 22:37:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/net/ftp.rb: improved documentation
-	* lib/net/imap.rb: ditto
-	* lib/net/pop.rb: ditto
-	* lib/net/smtp.rb: ditto
-	* lib/net/telnet.rb: ditto
-
-Fri May  7 21:50:21 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::parse_include): Allow
-	  multiple arguments to 'include'
-
-Fri May  7 21:31:56 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (fu_list): Array() breaks pathes including "\n".
-	  [ruby-core:02843]
-
-Fri May  7 11:25:53 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* util.c (ruby_strtod): "0.0000000000000000001" should be converted
-	  to 1.0e-19 instead of 0.0. (leading zeros aren't significant digits)
-	  [ruby-talk:99318] [ruby-dev:23465]
-
-Fri May  7 10:00:05 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/tkutil.c (get_eval_string_core): bug fix. [ruby-dev:23466]
-
-Thu May  6 22:13:17 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* ext/socket/socket.c (ippaddr): use NUMERICHOST if can not resolve
-	  hostname.
-
-Thu May  6 14:22:29 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml/rubytypes.rb (to_yaml): added instance variable handling
-	  for Ranges, Strings, Structs, Regexps.
-
-	* lib/yaml/rubytypes.rb (to_yaml_fold): new method for setting a
-	  String's flow style.
-
-	* lib/yaml.rb (YAML::object_maker): now uses Object.allocate.
-
-	* ext/syck/gram.c: fixed transfer methods on structs, broke it
-	  last commit.
-
-Thu May  6 11:40:28 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (string): accept NIL.
-
-	* lib/net/imap.rb (body_type_basic): allow body-fields omissions.
-
-Thu May  6 01:59:04 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb (Generators::HtmlMethod::params):
-	  Don't include the &block parameter if we have explicit
-	  yield parameters.
-
-Wed May  5 03:40:29 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/ring.rb: use recv instead of recvfrom.
-
-Tue May  4 23:52:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/gserver.rb: documented
-
-Tue May  4 23:46:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/xmlrpc/README.txt: introduced for documentation purposes
-
-Mon May  3 09:47:24 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method_or_yield_parameters):
-	  Fix parsing bug if yield called within 1 line block
-
-Sun May  2 01:04:38 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib, ext/tk: renewal Ruby/Tk
-
-Fri Apr 30 20:08:41 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* time.c (SIZEOF_TIME_T): support SIZEOF_TIME_T == SIZEOF_INT.
-
-Tue Apr 27 13:12:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_eval): too many line trace call. (ruby-bugs PR#1320)
-
-Tue Apr 27 08:41:28 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml/rubytypes.rb: passing Range tests.
-
-	* ext/syck/syck.h: version 0.44.
-
-	* ext/syck/gram.c: transfers no longer open an indentation.
-	  fixed transfers which precede blocks.
-
-	* ext/syck/token.c: ditto.
-
-	* ext/syck/syck.c: fixed segfault if an anchor has been released already.
-
-	* ext/syck/node.c (syck_free_members): organized order of free'd nodes.
-
-	* ext/syck/rubyext.c (syck_emitter_write_m): test for proper string with
-	  StringValue.
-
-Mon Apr 26 23:56:54 2004  Daniel Kelley  <news-1082945587@dkelley.gmp.san-jose.ca.us>
-
-	* README.EXT, README.EXT.ja: fixed wrong function signature.
-	  [ruby-talk:98349]
-
-Mon Apr 26 21:40:09 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/code_objects.rb (RDoc::Context::add_alias): Only alias
-	  to instance methods.
-
-Sat Apr 24 10:38:31 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/markup/simple_markup.rb (SM::SimpleMarkup::group_lines):
-	  Fix bug where consecutive headings are merged.
-
-Fri Apr 23 23:26:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb: $hdrdir should not contain macros for backward
-	  compatibility.  [bruby-dev:28]
-
-	* version.c (ruby_show_copyright): obtain copyright year from
-	  RUBY_RELEASE_YEAR.
-
-	* win32/resource.rb: ditto.
-
-	* win32/resource.rb: default rubyw icon to ruby.ico, and let DLL also
-	  include them.
-
-	* win32/resource.rb: include winver.h for older WindowsCE.
-
-Fri Apr 23 16:38:46 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb: sync taint/freeze flag between
-	  a pathname object and its internal string object.
-
-Fri Apr 23 14:52:08 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (stmt, arg, aref_args): should not make sole splat into
-	  array, in aref_args other than aref with op_asgn.
-
-Fri Apr 23 14:14:38 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/resolv.rb: don't use Regexp#source to embed regexps.
-	  [ruby-dev:23432]
-
-Thu Apr 22 04:15:36 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (aref_args): should pass expanded list.  [ruby-core:02793]
-
-Thu Apr 22 01:12:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* numeric.c (flo_to_s): tweak output string based to preserve
-	  decimal point and to remove trailing zeros.  [ruby-talk:97891]
-
-	* string.c (rb_str_index_m): use unsigned comparison for T_FIXNUM
-	  search.  [ruby-talk:97342]
-
-Wed Apr 21 22:57:27 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/rinda.rb, test/rinda/test_rinda.rb: check Hash tuple size.
-
-Wed Apr 21 20:05:00 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::HTTP#proxy_open): set Host: field explicitly.
-	  [ruby-list:39542]
-
-Mon Apr 19 18:11:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* hash.c (rb_hash_equal): returns true if two hashes have same set
-	  of key-value set.  [ruby-talk:97559]
-
-	* hash.c (rb_hash_eql): returns true if two hashes are equal and
-	  have same default values.
-
-Mon Apr 19 08:19:58 2004  Doug Kearns  <djkea2@mugca.its.monash.edu.au>
-
-	* dln.c, io.c, lib/benchmark.rb, lib/cgi.rb, lib/csv.rb, lib/date.rb,
-	  lib/ftools.rb, lib/getoptlong.rb, lib/logger.rb, lib/matrix.rb,
-	  lib/monitor.rb, lib/set.rb, lib/thwait.rb, lib/timeout.rb,
-	  lib/yaml.rb, lib/drb/drb.rb, lib/irb/workspace.rb, lib/net/ftp.rb,
-	  lib/net/http.rb, lib/net/imap.rb, lib/net/telnet.rb,
-	  lib/racc/parser.rb, lib/rinda/rinda.rb, lib/rinda/tuplespace.rb,
-	  lib/shell/command-processor.rb, lib/soap/rpc/soaplet.rb,
-	  lib/test/unit/testcase.rb, lib/test/unit/testsuite.rb: typo fix.
-
-Mon Apr 19 08:14:18 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_body): Allow for
-	  #ifdef HAVE_PROTOTYPES
-
-Fri Apr 16 22:33:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* ext/iconv/iconv.c: nearly finished RDoc comments.
-
-Fri Apr 16 17:04:07 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_equal): always returns true or false, never
-	  returns nil. [ruby-dev:23404]
-
-Fri Apr 16 08:27:02 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: skip linking when libraries to be preloaded not
-	  compiled.  [ruby-list:39561]
-
-Thu Apr 15 23:21:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* process.c (pst_success_p): new method Process::Status#success?.
-	  [ruby-dev:23385]
-
-Thu Apr 15 17:12:13 2004  Tanaka Akira  <akr@m17n.org>
-
-	* ext/gdbm/gdbm.c (Init_gdbm): define GDBM::READER, GDBM::WRITER,
-	  GDBM::WRCREAT and GDBM::NEWDB.
-	  (fgdbm_initialize): use specified read/write flag.
-
-Wed Apr 14 11:29:56 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* numeric.c (flo_eq): workaround for bcc32's bug.
-	  (ruby-bugs-ja:PR#594)
-
-Wed Apr 14 13:06:35 2004  Doug Kearns  <djkea2@mugca.its.monash.edu.au>
-
-	* array.c, enum.c, eval.c, file.c, io.c, numeric.c, object.c, prec.c,
-	  process.c, re.c, string.c: typos in RDoc comments.  [ruby-core:02783]
-
-Wed Apr 14 11:06:38 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::scan): Changed
-	  behavior of :enddoc: -- it now unconditionally terminates
-	  processing of the current file.
-
-Wed Apr 14 11:03:22 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* defines.h: include <net/socket.h> to get fd_set definition in BeOS.
-
-Tue Apr 13 23:06:30 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/rinda.rb: change pattern matching.
-	  a === b -> a == b || a === b. [druby-ja:98]
-
-	* test/rinda/test_rinda.rb: ditto.
-
-Tue Apr 13 19:54:29 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: should not overwrite HTTP request header.
-	  [ruby-list:39543]
-
-Tue Apr 13 01:30:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* ext/iconv/iconv.c: RDoc documentation (from RD; nearly finished).
-	* ext/iconv/charset_alias.rb: Prevent from RDoc'ing.
-
-Mon Apr 12 19:11:29 2004  Eric Hodel  <drbrain@segment7.net>
-
-	* gc.c (rb_gc_copy_finalizer): typo.  [ruby-core:02774]
-
-Mon Apr 12 18:52:32 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_init_i): should return
-	  a value.
-
-Mon Apr 12 10:43:47 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* dir.c (rb_glob2, rb_glob, rb_globi, push_globs, push_braces,
-	  rb_push_glob): fix memory leak. (leaked when block was interrupted)
-
-Mon Apr 12 10:27:37 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: backport SIZEOF_TIME_T definition from 1.9.
-
-	* win32/Makefile.sub: ditto.
-
-	* wince/Makefile.sub: ditto.
-
-Sun Apr 11 19:12:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ruby.c (require_libraries): restore source file/line after
-	  statically linked extensions initialized.  [ruby-dev:23357]
-
-Sun Apr 11 10:47:04 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/code_objects.rb (RDoc::TopLevel::add_class_or_module): Toplevel
-	  classes and modules are a special case too... (handle extending existing
-	  classes with or without :enddoc:)
-
-Sat Apr 10 23:51:13 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/code_objects.rb (RDoc::Context::add_to): Implementation of :enddoc:
-	  made one too many assumptions...
-
-Sat Apr 10 00:00:19 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/markup/simple_markup/inline.rb: Fix problem
-	  with \_cat_<b>dog</b>
-
-Wed Apr  7 00:19:50 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/rinda.rb: fix hash tuple bug.
-
-	* lib/rinda/tuplespace.rb: ditto.
-
-	* test/rinda/test_rinda.rb
-
-Tue Apr  6 18:24:18 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_reopen): should use rb_io_check_io().
-
-Tue Apr  6 16:46:09 2004  Tanaka Akira  <akr@m17n.org>
-
-	* configure.in: check the size of time_t.
-
-	* time.c (time_add): new function.
-	  (time_plus): use time_add.
-	  (time_minus): use time_add.
-
-Tue Apr  6 13:21:30 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/socket/socket.c (make_hostent): must return value.
-
-Tue Apr  6 00:05:30 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/rinda.rb: add require 'drb/drb'
-
-Mon Apr  5 08:18:23 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/rdoc.rb: Remove leading ./ from file names so that cross
-	  references work properly.
-
-Sun Apr  4 20:33:42 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* eval.c (Init_load): make $LOADED_FEATURES built-in.
-	  [ruby-dev:23299]
-
-	* ruby.c (ruby_prog_init): make $PROGRAM_NAME built-in.
-
-	* lib/English.rb: remove $LOADED_FEATURES and $PROGRAM_NAME.
-
-Sun Apr  4 14:01:20 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/options.rb (Options::parse): Allow multiple -x options to RDoc.
-	  Fix bug where files weren't being excluded properly
-
-Sat Apr  3 17:11:05 2004  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/syck.h: version 0.43.
-
-	* ext/syck/lib/gram.c: allow root-level inline collections.
-	 [ruby-talk:94922]
-
-	* lib/yaml/rubytypes.rb (Symbol#to_yaml): emit symbols as implicits.
-	 [ruby-talk:94930]
-
-	* ext/syck/bytecode.c: turn off default implicit typing.
-
-	* ext/syck/implicit.c: detect base60 integers.
-
-	* ext/syck/rubyext.c: handle base60, as well as hex and octal
-	  with commas.  implicit typing of ruby symbols.
-
-Fri Apr  2 17:27:17 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (top_include): include in the wrapped load is done for
-	  the wrapper, not for a singleton class for wrapped main.
-	  [ruby-dev:23305]
-
-Fri Apr  2 15:13:44 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (rb_big_eq): use temporary double variable to save the
-	  result (internal float register may be bigger than 64 bits, for
-	  example, 80 bits on x86).  [ruby-dev:23311]
-
-Fri Apr  2 14:35:26 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (block_pass): should generate unique identifier of the
-	  pushing block.  [ruby-talk:96363]
-
-Fri Apr  2 07:31:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/socket.c (make_hostent): fix memory leak, based on
-	  the patch from HORIKAWA Hisashi <vzw00011@nifty.ne.jp>.
-
-Thu Apr  1 22:55:33 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb: Allow rdoc comments in
-	  =begin rdoc/=end
-
-	* lib/rdoc/parsers/parse_rb.rb: Fix problem with comment in
-	  top-level method being taken as file comment.
-
-Thu Apr  1 22:55:04 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_options.rb: Fix undefined variable warning.
-
-Thu Apr  1 19:58:37 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/mapping/{factory.rb,registry.rb}: fixed illegal mapped URI
-	  object with soap/marshal.
-	  added URIFactory class for URI mapping.  BasetypeFactory checks
-	  instance_variables when original mapping is not allowed (ivar must
-	  be empty).  Instance of URI have instance_variables but it must be
-	  llowed whenever original mapping is allowed or not.
-
-	* lib/xsd/datatypes.rb: check the smallest positive non-zero
-	  single-precision float exactly instead of packing with "f".
-	  [ruby-talk:88822]
-
-	* lib/soap/mapping/rubytypeFactory.rb: should not dump singleton class.
-	  [ruby-dev:22588]
-	  c = class << Object.new; class C; self; end; end; SOAPMarshal.dump(c)
-
-Wed Mar 31 19:06:23 2004  Tanaka Akira  <akr@m17n.org>
-
-	* time.c (year_leap_p): new function.
-	  (timegm_noleapsecond): ditto.
-	  (search_time_t): use timegm_noleapsecond instead of
-	  mktime for first guess.
-
-Wed Mar 31 12:04:04 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/delegate.rb (DelegateClass): define internal methods of the
-	  result class, but not metaclass of the caller.  [ruby-talk:96156]
-
-	* intern.h: provide proper prototypes.  [ruby-core:02724]
-
-	* ruby.h: missing.h is now prerequisite to intern.h.
-
-Tue Mar 30 20:25:34 2004  Tanaka Akira  <akr@m17n.org>
-
-	* time.c (search_time_t): limit guess range by mktime if it is
-	  available.  [ruby-dev:23274]
-
-Sun Mar 28 14:16:59 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/pop.rb (auth): failed when account/password include "%".
-	  [ruby-talk:95933]
-
-Sat Mar 27 21:40:41 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb: permit extra semicolon in content-type field.
-
-Sat Mar 27 10:40:48 2004  Tanaka Akira  <akr@m17n.org>
-
-	* (lib/pp.rb, lib/prettyprint.rb): define seplist in PP::PPMethods
-	  instead of PrettyPrint.
-
-Thu Mar 25 23:28:52 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* time.c (time_overflow_p): backport 1.9 usec overflow function.
-	  (ruby-bugs PR#1307)
-
-Thu Mar 25 23:15:24 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_options.rb (RI::Options::show_version):
-	  Add --version option
-
-Thu Mar 25 04:16:18 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_options.rb (RI::Options): Add the --list-names option,
-	  which dumps our all known names
-
-Thu Mar 25 03:57:47 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_util.rb (NameDescriptor::initialize): No longer
-	  allow nested classes to be designated using "."--you must
-	  now use "::"
-
-Thu Mar 25 02:00:18 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/template/html/one_page_html.rb (Page):
-	  Fix to work with C modules.
-
-Wed Mar 24 21:17:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/uri.rb: Documented (thanks Dmitry V. Sabanin).
-	* lib/uri/common.rb: Ditto.
-	* lib/uri/ftp.rb: Ditto.
-	* lib/uri/generic.rb: Ditto.
-	* lib/uri/http.rb: Ditto.
-	* lib/uri/https.rb: Ditto.
-	* lib/uri/ldap.rb: Ditto.
-	* lib/uri/mailto.rb: Ditto.
-	  (All backported from 1.9)
-
-Wed Mar 24 18:48:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb ($ruby, $topdir, $hdrdir): should not be affected by
-	  DESTDIR after installed.
-
-	* lib/mkmf.rb (RUBY): / is not recognized as path separator on
-	  nmake/bmake. [ruby-list:39388]
-
-	* lib/mkmf.rb (init_mkmf): $INCFLAGS also should be lazy-evaluated.
-
-Wed Mar 24 12:32:56 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::handle_class_module):
-	  Don't document methods if we don't know for sure the
-	  class or module.
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_class):
-	  Don't store documentation for singleton classes if we
-	  don't know the real class.
-
-Wed Mar 24 11:11:26 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb (Generators::HTMLGenerator::load_html_template):
-	  Allow non-RDoc templates by putting a slash in the template name
-
-Mon Mar 22 16:19:57 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ruby.1: add -width option to .Bl for old groff.
-
-Sun Mar 21 21:11:16 2004  Keiju Ishitsuka  <keiju@ishitsuka.com>
-
-	* lib/shell/*: bug fix for Shell#system(command_line_string).
-
-Sat Mar 20 20:57:10 2004  David Black  <dblack@wobblini.net>
-
-	* lib/scanf.rb: Backported 1.9 branch
-	  modifications/corrections to 1.8 branch
-
-Sat Mar 20 23:51:03 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* eval.c (rb_require_safe): preserve old ruby_errinfo.
-	  [ruby-talk:95409]
-
-	* eval.c (rb_f_raise): should not clear backtrace information if
-	  exception object already have one.
-
-Sat Mar 20 15:25:36 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/template/html/html.rb (RDoc::Page): Force
-	  page background to white.
-
-Sat Mar 20 09:52:33 2004  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date.rb, lib/date/format.rb: _parse() now accepts fractional
-	  part of second minute that follows a comma or a full stop.
-
-Fri Mar 19 01:55:57 2004  Mauricio Fernandez  <batsman.geo@yahoo.com>
-
-	* io.c (rb_io_sync): need not to check writable. [ruby-core:02674]
-
-Thu Mar 18 21:44:38 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: backport drb.rb 1.16.
-
-Fri Mar 18 17:49:51 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* struct.c (make_struct): allow const_id for accessor names.
-	  [ruby-core:04585]
-
-	* eval.c (rb_attr): check if attribute name is local_id or
-	  const_id.
-
-Thu Mar 18 16:22:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_eq): avoid false positive by using scope and
-	  dyna_vars.  no longer use frame.uniq.
-
-Wed Mar 17 14:44:43 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* dir.c (range): fix possible "\0" overrun. (in case of "\0-")
-
-Mon Mar 15 07:39:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_yield_0): should not re-submit TAG_BREAK if this
-	  yield is not break destination. [ruby-dev:23197]
-
-Sat Mar 13 14:28:16 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/drb/test_drbssl.rb: rescue LoadError. (Barkport from main
-	  trunk)
-
-	* test/drb/test_drbunix.rb: ditto.
-
-Wed Mar 10 22:28:09 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (remove_dir): should handle symlink correctly.
-	  This patch is contributed by Christian Loew.  [ruby-talk:94635]
-	  (Backport from main trunk)
-
-Wed Mar 10 16:28:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (return_jump): set return value to the return
-	  destination.  separated from localjump_destination().
-
-	* eval.c (break_jump): break innermost loop (or thread or proc).
-
-	* eval.c (rb_yield_0): set exit_value for block break.
-
-Wed Mar 10 15:58:43 2004  Ryan Davis  <ryand@zenspider.com>
-
-	* eval.c (eval): Only print backtrace if generating the backtrace
-	  doesn't generate an exception.  [ruby-core:02621]
-
-Tue Mar  9 13:04:26 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_ungetc): raise IOError instead of calling
-	  rb_sys_fail().  [ruby-talk:23181]
-
-Mon Mar  8 19:32:28 2004  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/common.rb (URI::REGEXP::PATTERN::HOSTPORT): (?:#{PORT})
-	  -> (?::#{PORT}).  [ruby-dev:23170]
-
-Mon Mar  8 15:31:41 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* dir.c (range): treat incomplete '[' as ordinary character (like
-	  has_magic does).
-
-	* dir.c (range):  Cancel above change. More discussion is needed.
-
-Sun Mar  7 22:37:46 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/drb/ut_drb.rb: use 'druby://localhost:0'. [ruby-dev:23078]
-
-	* test/drb/ut_eval.rb: ditto.
-
-	* test/drb/ut_large.rb: ditto.
-
-	* test/drb/ut_safe1.rb: ditto.
-
-	* test/drb/ut_drb_drbssl.rb: use 'drbssl://localhost:0'.
-
-Sun Mar  7 16:22:26 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* Makefile.in (lex.c): use $? instead of $<.
-
-Fri Mar  5 00:54:14 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/test/unit.rb: MOve RDoc documentation so that you can
-	  now say 'ri Test::Unit'
-
-Tue Mar  2 12:32:59 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/Makefile.sub, wince/Makefile.sub (config.h): shouldn't check
-	  defined? NORETURN. [ruby-dev:23100]
-
-Mon Mar  1 12:24:10 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_alias):
-	  Allow aliases to have parentheses
-
-Sun Feb 29 23:14:53 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_class):
-	  Handle :nodoc: on singleton classes.
-
-Sat Feb 28 10:58:49 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* MANIFEST: add test_erb.rb
-
-	* lib/erb.rb, test/erb/test_erb.rb: don't forget filename,
-	  if both filename and safe_level given. [ruby-dev:23050]
-
-Fri Feb 27 01:00:09 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb, test/drb/drbtest.rb: require drb/eq.rb by default
-
-Wed Feb 25 21:16:25 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* instruby.rb (with_destdir): should return the given argument if no
-	  DESTDIR is given.
-
-	* instruby.rb: use path name expansion of cmd.exe.
-
-Wed Feb 25 09:35:22 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* error.c (NameError::Message): new class for lazy evaluation of
-	  message to ensure replaced before marshalling. merge from HEAD.
-	  (ruby-bugs-ja:PR#588)
-
-	* eval.c (rb_method_missing): use NameError::Message. merge from
-	  HEAD. (ruby-bugs-ja:PR#588)
-
-Tue Feb 24 18:59:37 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* dir.c (glob_helper): '**/' should not match leading period
-	  unless File::FNM_DOTMATCH is set. (like '*/') [ruby-dev:23014]
-
-Tue Feb 24 13:22:21 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/rdoc.rb (RDoc::RDoc::normalized_file_list): Attempt to get better
-	  heuristics on which files to include and exclude. Now only include
-	  non-standard files if they are explicitly named in ARGV.
-
-Tue Feb 24 07:23:30 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Deal with :stopdoc: when
-	  choosing a default main page to display (ie. don't select a page
-	  if we don't have documentation for it).
-
-Tue Feb 24 06:40:14 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RubyLex::identify_identifier): Handle
-	  class variables in code listings
-
-Tue Feb 24 06:40:14 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RubyLex::identify_identifier): Handle
-	  class variables in code listings
-
-Tue Feb 24 06:32:27 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_aliases): Handle
-	  aliases in C files.
-
-Tue Feb 24 06:16:22 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/rdoc.rb (RDoc::RDoc::document): Now create op dir _before_
-	  parsing files.
-
-Tue Feb 24 06:08:47 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_constant):
-	  Start collecting text of constant values earlier: was missing
-	  values in output if there was no space after '='
-
-Tue Feb 24 06:08:25 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Escape contant values.
-
-Tue Feb 24 03:45:06 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_config.c (ossl_config_each): add new method
-	  OpenSSL::Config#each. it iterates with section name, field name
-	  and value.
-
-	* ext/openssl/ossl_config.c (Init_ossl_config): include Enumerable.
-
-Mon Feb 23 09:16:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* instruby.rb (DOSISH): embedded path in batch files should not be
-	  prefixed by DESTDIR.  [ruby-core:02186]
-
-Sun Feb 22 09:54:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* re.c: corrected documentation format (again)
-
-Sun Feb 22 09:43:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* re.c: corrected documentation format (rb_reg_initialize_m)
-
-Sat Feb 21 22:36:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* ext/zlib/zlib.c: documented, but needs more effort.
-
-Sat Feb 21 11:12:15 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* missing/os2.c, missing/x68.c: typo fix.  pointed out by greentea.
-
-Fri Feb 20 18:59:47 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/irb/init.rb (IRB::IRB.parse_opts): add -I option to
-	  irb. [ruby-dev:39243]
-
-Thu Feb 19 23:24:16 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb (Generators::HtmlClass::build_attribute_list):
-	  Support visibility modifiers for attributes
-
-Thu Feb 19 23:24:16 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb (Generators::HtmlClass::build_attribute_list):
-	  Support visibility modifiers for attributes
-
-Thu Feb 19 22:39:04 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/rinda/test_rinda.rb: DRb.start_service only once in testsuites.
-	  DRb.start_service could handle this.
-
-Thu Feb 19 22:19:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/ostruct.rb: documented
-
-Thu Feb 19 21:28:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* ext/strscan/strscan.c: improved documentation
-
-Thu Feb 19 03:10:52 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/strscan/strscan.c: synchronized with main trunk (rev 1.11).
-
-Thu Feb 19 02:30:34 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/strscan/strscan.c: documentation checked.
-
-Thu Feb 19 00:11:05 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/markup/simple_markup/preprocess.rb (SM::PreProcess::handle):
-	  Strip extraneous space from filenames in :include:
-
-Wed Feb 18 22:52:00 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/unix.rb: remove O_NONBLOCK, thanks \ay
-
-Wed Feb 18 22:47:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* ext/strscan/strscan.c: documented
-
-Wed Feb 18 22:03:11 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/*: should not depend on $KCODE.
-
-Wed Feb 18 17:18:01 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/win32ole/win32ole.c: need to include <olectl.h> on Cygwin.
-
-Wed Feb 18 10:40:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* sprintf.c (rb_f_sprintf): do not prepend dots for negative
-	  numbers if FZERO is specified.  [ruby-list:39218]
-
-Tue Feb 17 23:40:34 2004  Guy Decoux  <ts@moulon.inra.fr>
-
-	* sprintf.c (rb_f_sprintf): preserve original val for
-	  format_integer. [ruby-talk:92975]
-
-Tue Feb 17 23:28:45 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/marshaltestlib.rb: common marshal testcase added.
-
-	* test/ruby/test_marshal.rb: use above testsuite.
-
-	* test/soap/marshal/test_marshal.rb: ditto.
-
-	* test/soap/marshal/cmarshal.rb: removed (not used).
-
-Tue Feb 17 10:51:23 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/syck/rubyext.c (syck_emitter_end_object): takes only one arg.
-
-Tue Feb 17 01:35:28 2004  Tanaka Akira  <akr@m17n.org>
-
-	* eval.c (rb_eval): care that another thread replace NODE_DREGX_ONCE
-	  to NODE_LIT.  [ruby-dev:22920]
-
-Tue Feb 17 01:24:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bcc32/Makefile.sub, win32/Makefile.sub (config.h): define
-	  STACK_GROW_DIRECTION. [ruby-dev:22910]
-
-	* bcc32/Makefile.sub (config.h): add newer checks.
-
-	* wince/Makefile.sub (config.h): define NEED_IO_SEEK_BETWEEN_RW.
-
-Tue Feb 17 00:38:10 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/tuplespace.rb: TupleSpace#initialize, stop doubling timeout
-
-Tue Feb 17 00:18:03 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/rinda/test_rinda.rb: import test_rinda.rb
-
-Tue Feb 17 00:14:30 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: avoid warning "Redefinition of macro
-	  'HAVE_GETLOGIN'".
-
-	* vms/config.h_in: ditto.
-
-Mon Feb 16 23:28:14 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/csv.rb: document reduction.  [ruby-core:02429]
-
-Mon Feb 16 22:08:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/generator.rb: corrected doc format
-	* lib/rinda/rinda.rb: added documentation (from Hugh Sasse)
-	* lib/rinda/tuplespace.rb: ditto
-
-Mon Feb 16 20:41:32 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: show more warnings. (refering to mingw)
-
-	* bcc32/setup.mak: ditto.
-
-Mon Feb 16 13:39:44 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* dir.c (rb_glob, rb_globi): add const.
-
-	* ruby.h: ditto.
-
-Mon Feb 16 02:16:33 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* bcc32/Makefile.sub: should warn suspicious pointer conversion.
-
-	* bcc32/setup.mak: ditto.
-
-Sun Feb 15 19:06:42 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/tuplespace.rb: TupleSpace#read(tpl, 0), raise
-	  RequestExpiredError if not found.
-
-Sun Feb 15 15:56:46 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c: add IDispatch wrapper in val2variant.
-	  Thanks, arton.
-
-Sun Feb 15 01:46:05 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/mkmf.rb: absolute path of ruby is assigned to $(RUBY).
-	  [ruby-dev:22870]
-
-Sat Feb 14 11:29:41 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* sample/drb/*: import lib/drb/sample
-
-Sat Feb 14 11:08:23 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: add pretty_print, thanks gotoken.
-
-Fri Feb 13 12:35:08 2004  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb: File.link may raise EINVAL and
-	  EACCES on Windows.
-
-Thu Feb 12 21:45:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/ftools.rb: documented
-
-Thu Feb 12 21:25:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/base64.rb: backported from HEAD (modularised and documented)
-
-Thu Feb 12 20:31:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (create_tmpsrc): cpp32 of Borland C++ ignores #error
-	  directives in DOS line-ending files at all.
-
-Thu Feb 12 02:23:56 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb: use assert_raise instead of assert_raises.
-
-	* lib/pp.rb: ditto.
-
-	* lib/time.rb: ditto.
-
-	* lib/tsort.rb: ditto.
-	  use TSortHash and TSortArray instead of Hash and Array in test.
-
-Wed Feb 11 20:01:12 2004  akira yamada  <akira@ruby-lang.org>
-
-	* test/ruby/test_file.rb (TestFile::test_fnmatch): added tests for
-	  File.fnmatch. [ruby-dev:22815][ruby-dev:22819]
-
-	* test/ruby/test_proc.rb (TestProc::test_eq): added a
-	  test.  [ruby-dev:22599]
-
-	* test/ruby/test_proc.rb (TestProc::test_eq): added tests for
-	   Proc#==.  [ruby-dev:22592], [ruby-dev:22601]
-
-Tue Feb 10 16:43:56 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (umethod_bind): purge unused check.  [ruby-dev:22850]
-
-Mon Feb  9 17:16:00 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/rdoc/parsers/parse_c.rb: escape '{' and '}' to avoid warnings.
-
-Mon Feb  9 13:00:55 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
-
-	* dir.c (fnmatch): File.fnmatch('*?', 'a') should return true.
-	  [ruby-dev:22815]
-
-	* dir.c (fnmatch): File.fnmatch('\[1\]' , '[1]') should return true.
-	  [ruby-dev:22819]
-
-Sun Feb  8 16:46:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/pp.rb (PP::PPMethods::object_address_group): suppress negative
-	  sign for higher heap areas.
-
-Fri Feb  6 22:48:16 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb (gen_url): Support
-	  https in RDoc hyperlinks
-
-Fri Feb  6 22:41:22 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/pp.rb (PPInspectTest#test_to_s_with_iv): rollback the previous
-	  commit.  [ruby-dev:22813]
-
-Fri Feb  6 22:22:50 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/pp.rb (PPInspectTest#test_to_s_with_iv): remove instance
-	  variable which is defined in the test.
-
-Fri Feb  6 00:48:37 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/prettyprint.rb (PrettyPrint#first?): obsoleted.
-
-Thu Feb  5 23:56:55 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/prettyprint.rb (PrettyPrint#seplist): added.
-
-	* lib/pp.rb (PPMethods#pp_object): use seplist.
-	  (PPMethods#pp_hash): ditto.
-	  (Array#pretty_print): ditto.
-	  (Struct#pretty_print): ditto.
-	  (MatchData#pretty_print): ditto.
-
-	* lib/set.rb (Set#pretty_print): use seplist.
-
-Wed Feb  4 02:12:06 2004  Tanaka Akira  <akr@m17n.org>
-
-	* file.c (test_l): fix wrong method name in document.
-	  (test_S): ditto.
-	  (test_b): ditto.
-	  (test_c): ditto.
-	  (test_suid): ditto.
-	  (test_sgid): ditto.
-	  (test_sticky): ditto.
-
-Tue Feb  3 08:04:57 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pp.rb (Struct#pretty_print_cycle): follow 1.8 style.
-
-Mon Feb  2 19:33:49 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: backport from 1.9 for Interix.
-
-	* dln.c (dln_load): ditto.
-
-Mon Feb  2 13:31:51 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/net/http.rb (canonical_each): fix merge miss.
-
-Mon Feb  2 01:54:00 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pp.rb (Struct#pretty_print): make it 1.8 style.
-	  (Numeric#pretty_print, FalseClass#pretty_print)
-	  (TrueClass#pretty_print, Module#pretty_print): fix pp for objects
-	  with instance variables.  [ruby-talk:91157]
-
-	* lib/open-uri.rb (URI::Generic#find_proxy): return nil on loopback
-	  address.
-
-	* lib/resolv-replace.rb (BasicSocket#send): don't replace because
-	  it has no hostname argument.
-	  (IPSocket.getaddress): raise SocketError instead of
-	  Resolv::ResolvError for errors.
-	  (TCPSocket#initialize, UDPSocket#bind, UDPSocket#connect)
-	  (SOCKSSocket#initialize): use IPSocket.getaddress instead of
-	  Resolv.getaddress.
-	  (UDPSocket#send): recognize 3 arguments form.  try all addresses on
-	  4 arguments form.
-
-Sun Feb  1 18:17:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/net/http.rb: merged coding style changes from HEAD.
-
-Sun Feb  1 16:15:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/test/unit.rb: rearranged documentation for RDoc's sake.
-	* lib/matrix.rb: improved documentation.
-	* lib/net/http.rb: slight documentation formatting improvement.
-
-Sun Feb  1 05:30:06 2004  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::Generic#find_proxy): warn HTTP_PROXY.
-	 raise an errror on non-http proxy URI.
-	 (OpenURI::Buffer#<<): make a tempfile binmode.  [ruby-talk:90793]
-
-Sat Jan 31 09:20:32 2004  NAKAMURA, Hiroshi  <nakahiro@sairon.co.jp>
-
-	* sample/openssl/gen_csr.rb: wrong usage string.
-
-Sat Jan 31 01:00:32 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/wsdlDriver.rb, lib/wsdl/soap/operation.rb: add support of
-	  "parts" attribute of soap:body element in WSDL.
-
-	* lib/wsdl/xmlSchema/schema.rb: friendly warning message for
-	  simpleType element which is not supported for now.
-
-	* lib/soap/mapping/factory.rb: deleted unused methods.
-
-	* lib/soap/mapping/rubytypeFactory.rb: do no ignore case while xsi:type
-	  string <-> Ruby class name matching.
-
-	* test/wsdl/soap/{soapbodyparts.wsdl,test_soapbodyparts.wsdl}: new
-	  files.
-
-Thu Jan 29 23:56:00 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* util.c (mblen): fix overrun.  [ruby-dev:22672]
-
-Thu Jan 29 22:41:53 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Allow 'link:' in Tidylinks.
-	  THis means you can write "see f1[link:files/f1_rb.html]".
-
-Thu Jan 29 15:33:23 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509hame.c (ossl_x509name_initialize): change
-	  second argument. it expected to be a Hash not an Integer.
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_add_entry): add new
-	  function for OpenSSL::X509::Name#add_entry.
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_to_a): append ASN.1
-	  tag number to each element of return value.
-
-	* ext/openssl/ossl_x509name.c (Init_ossl_x509name): add constants
-	  OpenSSL::X509::Name::DEFAULT_OBJECT_TYPE and OBJECT_TYPE_TEMPLATE.
-
-	* ext/openssl/lib/openssl/x509.rb (OpenSSL::X509::Name#initialize):
-	  second argument takes OBJECT_TYPE_TEMPLATE by default.
-
-	* sample/openssl/gen_csr.rb: use OpenSSL::X509::Name.parse.
-
-Wed Jan 28 04:29:41 2004  Eric Schwartz  <emschwar@fc.hp.com>
-
-	* lib/cgi/session.rb: use LOCK_SH to read, and a few other
-	  improvements.  [ruby-core:02328]
-
-Tue Jan 27 11:09:29 2004  FUKUMOTO Atsushi  <fukumoto@nospam.imasy.or.jp>
-
-	* ext/socket/socket.c (s_recvfrom): sending length should be an
-	  invariant while retrying on EAGAIN.  [ruby-talk:89962]
-
-Tue Jan 27 10:35:18 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/win32ole/win32ole.c (set_argv): fix condition.
-
-Tue Jan 27 02:26:31 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httputils.rb (WEBrick:HTTPUtils::parse_header):
-	  refine regex for header-name.
-
-Tue Jan 27 00:30:11 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/Makefile.sub: rollback.
-
-Mon Jan 26 22:53:04 2004  Dave Thomas  <dave@pragprog.com>
-
-	* io.c: Remove documentation references to $defout.
-
-Mon Jan 26 15:11:47 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* sample/exyacc.rb: escape '}' to avoid warning.
-
-Mon Jan 26 14:41:46 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/delegate.rb (Delegator::initialize): preserve
-	  singleton_method_added method [ruby-dev:22685]
-
-	* lib/delegate.rb (Delegator::initialize): use Kernel::raise
-	  instead of mere raise.  [ruby-dev:22681]
-
-Mon Jan 26 12:47:17 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: define CONST84 when TCL_MAJOR_VERSION == 7
-
-Mon Jan 26 11:35:23 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb: Makefiles should depend on also rbconfig.rb.
-	  (ruby-bugs:PR#1256)
-
-	* ext/win32ole/win32ole.c (set_argv): set real arguments to
-	  WIN32OLE::ARGV.  [ruby-list:39073]
-
-Thu Jan 22 22:54:53 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (BEG_REGEXP): allow 8-bit characters in quoted
-	  strings for Novell GroupWise Internet Agent.
-	* lib/net/imap.rb (DATA_REGEXP): ditto.
-
-Thu Jan 22 16:21:33 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (string_content): reset lexical states at the beginning of
-	  string contents.  [ruby-list:39061]
-
-Wed Jan 21 21:55:51 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: remove O_NONBLOCK, thanks \ay
-	* lib/drb/extserv.rb: typo
-
-Wed Jan 21 17:57:56 2004  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (envelope): allow NIL.
-	* lib/net/imap.rb (body): ditto.
-	* lib/net/imap.rb (number): ditto.
-	* lib/net/imap.rb (ensure_nz_number): show a detailed error
-	  message.
-
-Wed Jan 21 16:44:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (merge_libs): squeeze successive same libraries.
-	  [ruby-dev:22652]
-
-Wed Jan 21 16:01:37 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/digest/rmd160/extconf.rb: have_library appends found library.
-
-Wed Jan 21 11:36:00 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (block_append): update nd_end for "real" head node.
-	  [ruby-list:39058]
-
-Tue Jan 20 14:48:13 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: should check <openssl/conf_api.h> instead
-	  of OPENSSL_VERSION_NUMBER. [ruby-list:39056]
-
-Tue Jan 20 14:43:17 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/base64.rb: Add RDoc
-
-Tue Jan 20 14:25:51 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/abbrev.rb: Add RDoc
-
-Tue Jan 20 13:22:39 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Document aliases at
-	  top-most level.
-
-	* lib/English.rb: Document English.rb.
-
-Tue Jan 20 02:49:22 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: add check for OpenSSL version.
-	  [ruby-list:39054]
-
-Tue Jan 20 02:38:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_class): should not dump singleton class.
-	  [ruby-dev:22631]
-
-Tue Jan 20 01:31:36 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* io.c (lineno): typo fix(FIX2INT -> INT2FIX).
-
-Mon Jan 19 21:53:38 2004  akira yamada  <akira@ruby-lang.org>
-
-	* io.c, re.c, string.c, time.c: fixed up positions of RDocs.
-
-Mon Jan 19 07:09:20 2004  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date.rb: zone was wrong when it was behind UTC.
-	  Thanks Mark J. Reed.
-
-	* lib/date/format.rb: %z is now always replaced by four digits
-	  with a leading plus or minus sign.
-
-	* sample/cal.rb: added a class, anyway.
-
-Sun Jan 18 20:47:35 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ruby.c: use translate_char() on Cygwin.
-
-Sun Jan 18 02:33:26 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* defines.h (_WIN32): undef _WIN32 on Cygwin before defining DOSISH.
-
-Sun Jan 18 00:23:55 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* marshal.c (class2path): check anonymous class/module before
-	  checking referable, and allow singleton classes.
-
-Fri Jan 16 14:33:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* marshal.c (class2path): get class path and check referable.
-	  [ruby-dev:22588]
-
-Fri Jan 16 09:52:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_eq): Proc with empty body may not be equal.
-	  [ruby-dev:22590]
-
-Thu Jan 15 13:03:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (argf_read): do not append EOF.  (ruby-bugs-ja:PR#585)
-
-	* io.c (rb_io_fwrite): ad-hockery hack to get rid of HP-UX stdio
-	  weird behavior.  [ruby-dev:22424]
-
-Wed Jan 14 13:31:06 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/extconf.rb: wrapper iconv.rb is dependent on platform.
-
-Tue Jan 13 18:54:28 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb(Logger#msg2str): no special treatment for the object
-	  which responds to :to_str.  commited at 2004-01-11T21:46:27 by
-	  gsinclair.
-
-	* lib/logger.rb(LogDevice#initialize): remove type checking if the
-	  given object is a String.  Kernel.open handles it correctly.
-	  commited at 2004-01-11T21:46:27 by gsinclair.
-
-	* test/logger/test_logger.rb: follow above change (ArgumentError ->
-	  TypeError.)  follow above commit.
-
-Tue Jan 13 14:27:13 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
-
-	* lib/test/unit/ui/testrunnerutilities.rb (TestRunnerUtilities):
-	  moved run method which allows output level.  [ruby-dev:22554]
-
-Tue Jan 13 04:29:52 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_driver.rb (RiDriver::report_method_stuff):
-	  Show fully-qualified class names in class list.
-
-Tue Jan 13 01:04:37 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_paths.rb (RI::Paths): First attempt at
-	  incorporating DESTDIR in the rdoc installation.
-
-Mon Jan 12 23:27:19 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (primary): fix position after FCALL.  [ruby-dev:22574]
-
-Mon Jan 12 12:07:22 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_methods):
-	  Someone changed the "// in eval.c" comments to "/*...*/" style,
-	  so the parsing of the source file name broke.
-
-	* object.c: Remove spurious space in TrueClass documentation.
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_body): Fix
-	  bad regexp: if the code before a documented method contained
-	  a comment that wasn't terminated by whitespace, that comment
-	  and all intervening code was included in the following
-	  method's documentation.
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::HtmlFormatter::break_to_newline):
-	  HTML formats need explicit line breaks.
-
-Mon Jan 12 11:46:30 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (LIBPATHFLAG, RPATHFLAG): enclose paths with single
-	  quotes.  [ruby-dev:22564]
-
-	* lib/mkmf.rb (libpathflag): do not enclose with quotes always.
-
-	* {bcc32,win32,wince}/Makefile.sub (LIBPATHFLAG): quoted.
-
-Mon Jan 12 02:24:07 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::HtmlFormatter): Add HTML
-	  generation support to ri (Elliot Hughes)
-
-Mon Jan 12 02:24:07 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::HtmlFormatter): Add HTML
-	  generation support to ri (Elliot Hughes)
-
-Sun Jan 11 02:07:47 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_options.rb (RI::Options::OptionList::OptionList):
-	  Also accept command line options via the 'RI' environment variable.
-
-Sun Jan 11 02:07:47 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_options.rb (RI::Options::OptionList::OptionList):
-	  Also accept command line options via the 'RI' environment variable.
-
-Sat Jan 10 21:27:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (eval): need to add message delimiter. [ruby-dev:22561]
-
-Sat Jan 10 01:54:50 2004  Eric Sunshine  <sunshine@sunshineco.com>
-
-	* defines.h (__NeXT__): Ensure that all standard S_IRUSR, S_IWGRP,
-	  S_IRWXO, etc. macros are defined since future code might require
-	  them (even though present code only requires a subset).
-
-	* defines.h (__NeXT__): Bug fix: WORDS_BIGENDIAN was not being set
-	  correctly on Rhapsody when -arch compiler flag was used (via
-	  configure's --enable-fat-binary option).
-
-Fri Jan  9 10:05:14 2004  Siena.  <siena@faculty.chiba-u.jp>
-
-	* lib/mkmf.rb (libpathflag): use single quotes.  [ruby-dev:22440]
-
-Thu Jan  8 23:49:21 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (RDOCTARGET): new macro.  if you want to install
-	  rdoc documentation, you need to run configure with
-	  --enable-install-doc.
-
-Thu Jan  8 21:29:43 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_pkey.c (ossl_pkey_to_der): removed; it returns
-	  public key only.
-
-	* ext/openssl/ossl_pkey_dh.c (ossl_dh_to_der): new function for
-	  OpenSSL::PKey::DH#to_der.
-
-	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_to_der): new function for
-	  OpenSSL::PKey::DSA#to_der.
-
-	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_to_der): new function for
-	  OpenSSL::PKey::RSA#to_der.
-
-Thu Jan  8 16:51:04 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/wsdl/datetime/test_datetime.rb: fixed a stupid testcase which
-	  dumps "E" at month-end.
-
-Thu Jan  8 11:20:01 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* eval.c, object.c, process.c, re.c: don't use C++ style comments.
-
-Thu Jan  8 04:36:21 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (WEBrick::CGI#initialize): should create
-	  @config[:Logger] if it was not given.
-
-	* sample/webrick/*: new files.
-
-	* MANIFEST: add sample/webrick/*
-
-Wed Jan  7 13:00:18 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_driver.rb: Fix problem where ri was
-	  being too eager to find matches of ambiguous method
-	  names (such as "ri Thread.join" would return both
-	  Thread.join and ThreadsWait.join)
-
-Wed Jan  7 12:35:41 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/debug.rb: revert command parse regexps.  [ruby-list:39014] by
-	  Shirai,Kaoru.
-
-Wed Jan  7 08:21:04 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parserfactory.rb: Check for shebang
-	  line in files that would otherwise be treated as
-	  plain text.
-
-Tue Jan  6 22:13:34 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_mod_modfunc): should break if m has no super class.
-	  [ruby-dev:22498]
-
-Tue Jan  6 21:55:02 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (fptr_finalize): should save errno just after failure.
-	  [ruby-dev:22492]
-
-Tue Jan  6 14:53:14 2004  Dave Thomas  <dave@pragprog.com>
-
-	* bin/ri: split out the display side, making it pluggable. Added
-	  new ri_driver and ri_display files in lib/rdoc/ri.
-
-Tue Jan  6 06:37:53 2004  Dave Thomas  <dave@pragprog.com>
-
-	* bin/rdoc: Add --ri-system switch
-
-	* lib/.document: Update with list of files that seem to have
-	  documentation
-
-	* lib/test/unit.rb: Reorder comment to make it RDoc friendly.
-
-	* Makefile.in: add install-nodoc target, and make it
-	  generate RDoc on default install.
-
-	* lib/rdoc/ri/ri_options.rb (RI::Options::parse): Add
-	  --doc-dir option to ri.
-
-Tue Jan  6 00:04:40 2004  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method_or_yield_parameters):
-	  fix parsing if there are braces in a method parameter list
-
-Fri Jan  2 14:54:11 2004  Dave Thomas  <dave@pragprog.com>
-
-	* bin/ri: Add new --classes option, and arrange for
-	  help messages to be paged too.
-
-	* bin/rdoc: Add statistics.
-
-	* process.c: (MG) Added Process documentation
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::AttributeFormatter::wrap):
-	  Fix problem with labels not displaying in RI labeled
-	  lists using BS and ANSI modes.
-
-Fri Jan  2 01:50:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (argf_eof): ARGF.eof? should not have any side effect.
-	  [ruby-dev:22469]
-
-Wed Dec 31 17:25:17 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (argf_each_byte): should return self.  [ruby-dev:22465]
-
-Wed Dec 31 11:20:34 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_methods): Make
-	  file referenced in "// in sss.c" relative to current file.
-
-Wed Dec 31 11:17:37 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/generators/html_generator.rb: Fix problem when
-	  a public method was aliased, but the alias is then
-	  made private, and hence doesn't appear in RDoc output.
-
-Wed Dec 31 01:33:05 2003  Dave Thomas  <dave@pragprog.com>
-
-	* array.c, error.c, eval.c, io.c, prec.c, range.c, re.c,
-	  string.c, time.c: Add RDoc for Kernel functions, and tidy.
-
-Tue Dec 30 19:39:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_f_readline): should raise EOFError at the end of
-	  files.  [ruby-dev:22458]
-
-	* io.c (argf_read): should concatenate input files when length
-	  argument is nil. [ruby-dev:22450]
-
-	* io.c (argf_read): should update supplied string buffer (2nd
-	  argument) even when IO#read is called multiple times.
-
-	* io.c: should initialize lineno by zero. [ruby-dev:22460]
-
-Tue Dec 30 12:30:30 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/code_objects.rb (RDoc::Context::find_symbol): If a
-	  class and a method have the same name, finding Xxx.abc was trying
-	  to find 'abc' in method 'Xxx', not class 'Xxx'.
-
-
-Tue Dec 30 08:32:32 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method):
-	  Handle undoing nesting of yield parameters correctly for:
-
-	  def each_entry(&b) Dir.foreach(@path) {|f| yield P.new(f) } end
-
-
-Tue Dec 30 08:32:32 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method):
-	  Handle undoing nesting of yield parameters correctly for:
-
-	    def each_entry(&block) Dir.foreach(@path) {|f| yield Pathname.new(f) } end
-
-Mon Dec 29 12:51:02 2003  Dave Thomas  <dave@pragprog.com>
-
-	* eval.c: Add RDoc for Kernel global functions.
-
-Mon Dec 29 11:00:16 2003  Dave Thomas  <dave@pragprog.com>
-
-	* array.c: Tidy up RDoc loose ends.
-
-Mon Dec 29 05:05:51 2003  Dave Thomas  <dave@pragprog.com>
-
-	* struct.c, random: Add RDoc comments
-
-Mon Dec 29 02:20:54 2003  Dave Thomas  <dave@pragprog.com>
-
-	* eval.c: Add RDoc for class Proc, Method, UnboundMethod
-
-Mon Dec 29 00:41:44 2003  Dave Thomas  <dave@pragprog.com>
-
-	* math.c: Add RDoc comments
-
-Sun Dec 28 20:19:11 2003  Tanaka Akira  <akr@m17n.org>
-
-	* ext/stringio/stringio.c (strio_sysread): StringIO.new.sysread didn't
-	  raise EOFError.
-
-	* ext/zlib/zlib.c (gzreader_gets): don't increment lineno when
-	  gzfile_read_all returns "".
-
-Sun Dec 28 15:25:08 2003  Dave Thomas  <dave@pragprog.com>
-
-	* class.c,object.c,parse.y,sprintf.c,variable.c: Document classes
-	  Object, Module, etc...
-
-Sun Dec 28 11:55:29 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/csv/test_csv.rb: generate bom.csv and mac.csv files on the fly.
-	  [ruby-talk:88852]
-
-	* test/csv/{bom.csv,mac.csv}: removed.
-
-Sun Dec 28 08:56:51 2003  Dave Thomas  <dave@pragprog.com>
-
-	* eval.c: Thead[Group] RDoc (thanks to MG)
-
-Sun Dec 28 03:50:05 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_override_comment):
-	  Escape method names used in regexp
-
-Sun Dec 28 01:46:02 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/ri/ri_formatter.rb (RI::TextFormatter::display_flow_item):
-	  Add support for rules in 'ri' output.
-
-Sun Dec 28 01:35:35 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_body):
-	  Sometimes the Ruby source aliases two otherwise
-	  unrelated methods (for example Kernel#object_id and
-	  Kernel#hash are both the same C function). Provide a
-	  facility to allow the methods to be documented
-	  separately.
-
-Sun Dec 28 01:05:31 2003  Dave Thomas  <dave@pragprog.com>
-
-	* marshal.c, signal.c: RDoc collemts added by Elliott Hughes
-
-Sun Dec 28 00:48:47 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_class_comment):
-	  Some source files use lower case class or module names
-	  when naming the Init_XXX function in C.
-
-Sat Dec 27 23:41:46 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: fix "test: too many arguments" error.
-
-Sat Dec 27 15:32:19 2003  Dave Thomas  <dave@wireless_3.local.thomases.com>
-
-	* time.c: RDoc comments added
-
-Sat Dec 27 15:07:57 2003  Dave Thomas  <dave@pragprog.com>
-
-	* object.c: Add RDoc comments for Symbol class.
-
-Sat Dec 27 14:42:30 2003  Dave Thomas  <dave@pragprog.com>
-
-	* numeric.c: Add RDoc comments.
-
-Sat Dec 27 00:44:00 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (next_argv): warn always for stdin on inplace edit mode.
-
-	* io.c (read_all): need to check string value.
-
-	* io.c (argf_read): allow ARGF.read(nil).  [ruby-dev:22433]
-
-Fri Dec 26 23:02:09 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_f_backquote): need not to check nil result.
-	  [ruby-core:02078]
-
-	* io.c (rb_io_getline): should return nil when read_all gives
-	  empty string, even when nil rs is specified. [ruby-core:02077]
-
-Fri Dec 26 18:50:59 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in: check if getcontext and setcontext are available.
-
-	* eval.c: use presence of getcontext/setcontext.
-
-Fri Dec 26 16:40:53 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (PathnameTest#test_plus): add 2 assertions.
-
-Fri Dec 26 09:26:58 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (pack_pack): add sign check for 'i', and 'l'.
-	  [ruby-dev:22427]
-
-	* bignum.c (rb_quad_pack): add range check for 'quad int'.
-
-Thu Dec 25 22:39:59 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* string.c (rb_str_update): don't return any value.
-
-Thu Dec 25 15:30:17 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_update): call rb_str_modify().
-
-Thu Dec 25 05:08:09 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (search_required): search actual file name once when no
-	  extension specified.
-
-Thu Dec 25 04:00:44 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* stable version 1.8.1 released.
-
-Thu Dec 25 00:17:53 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* configure.in: check for nanosleep, -lrt if required.
-	  [ruby-core:02059]
-
-	* eval.c (thread_timer): use select(2) if nanosleep(2) is not
-	  available.
-
-	* eval.c: check __stub_getcontext for glibc on some platforms.
-	  [ruby-list:38984]
-
-Wed Dec 24 23:48:04 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/test_basetype.rb, test/soap/marshal/test_marshal.rb
-	  test/xsd/test_xsd.rb: use "(-1.0 / (1.0 / 0.0))" instead of "-0.0"
-	  to express -0.0.  [ruby-talk:88786]
-
-Wed Dec 24 23:29:30 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/tsort.rb (test_orphaned_break): removed.
-
-Wed Dec 24 20:53:06 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tkmulticolumnlist.rb: new sample
-
-	* ext/tk/sample/tkmultilistframe.rb: bug fix
-
-Wed Dec 24 20:37:37 2003  Eric Sunshine  <sunshine@sunshineco.com>
-
-	* configure.in (LDSHARED): Fixed typographical error in assignment of
-	  LDSHARED for Rhapsody which caused linking of extension modules to
-	  fail.
-
-Wed Dec 24 17:51:18 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* file.c (rb_thread_flock): enable thread support again.
-
-Wed Dec 24 16:46:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (catch_timer): do not call rb_thread_schedule() inside to
-	  avoid pthread_mutex_lock() deadlock.  interrupts to system calls
-	  are detected by TRAP_END via EINTR error.
-
-	* eval.c (thread_timer): do not post signal unless it is
-	  absolutely necessary.
-
-	* rubysig.h (TRAP_END): add CHECK_INTS to switch thread.
-
-	* regex.c (re_compile_pattern): check if nextp is smaller than
-	  pend.  [ruby-dev:22372]
-
-	* eval.c (umethod_bind): remove method overridden check.
-	  [ruby-dev:22366]
-
-Wed Dec 24 16:13:05 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_read): should check for error
-	  status by SSL_get_error().
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_write): ditto.
-
-Wed Dec 24 14:23:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): clear the buffer argument
-	  when returning nil.  [ruby-dev:22363]
-
-	* test/ruby/ut_eof.rb (TestEOF::test_eof_0, TestEOF::test_eof_1):
-	  add buffer argument tests.
-
-Wed Dec 24 14:07:55 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: Modules are allowed to rescue.
-
-	* lib/test/unit/autorunner.rb: show output_level in order.
-
-	* lib/test/unit/collector/dir.rb: get rid of successive same
-	  directories in load path.
-
-	* test/testunit/test_assertions.rb (test_assert_nothing_raised,
-	  test_assert_raise): test for modules.
-
-Wed Dec 24 13:43:34 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/net/imap.rb (authenticate): remove "\n" from base64 encoded
-	  strings.
-
-Wed Dec 24 11:26:41 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/fileutils/test_fileutils.rb: should not create any
-	  files or directories in current directory.  [ruby-talk:88724]
-
-Wed Dec 24 10:29:53 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): never return nil at
-	  unlimited read.  [ruby-dev:22334]
-
-	* ext/stringio/stringio.c (strio_read): support second
-	  argument.  [ruby-dev:22350]
-
-Wed Dec 24 09:38:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (arg): should return 0 after error.  [ruby-dev:22360]
-
-Wed Dec 24 00:56:54 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (read_all): do not return nil at the end of file.
-	  [ruby-dev:22334]
-
-	* io.c (argf_read): do not depend on nil at eof behavior of
-	  IO#read().
-
-	* eval.c (rb_thread_join): dup exception before re-raising it.
-
-	* io.c (rb_io_eof): call clearerr() to prevent side effect.  this
-	  patch is supplied by Masahiro Sakai <sakai@tom.sfc.keio.ac.jp>.
-	  [ruby-dev:22234]
-
-	* pack.c (OFF16): get offset for big endian machines.
-
-	* pack.c (pack_pack): use OFF16 instead of OFF16B.
-	  [ruby-dev:22344]
-
-	* pack.c (pack_unpack): ditto.
-
-Tue Dec 23 22:47:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_io_check_readable): set FMODE_RBUF always, even if
-	  NEED_IO_SEEK_BETWEEN_RW is not defined. [ruby-dev:22340]
-
-	* io.c (rb_io_check_writable): clear FMODE_RBUF before writing
-	  something.
-
-Tue Dec 23 22:25:00 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/optparse.rb: incomplete RDoc documentation added in place of
-	  existing RD comments.  Tabs converted to spaces.
-
-Tue Dec 23 19:44:47 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/test_streamhandler.rb (test_basic_auth): removed.
-	  soap4r + basic_auth is not officially supported in ruby/1.8.1 even
-	  though soap4r + basic_auth + http-access2 should run fine.
-
-Tue Dec 23 19:42:59 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_ungetc): raise an exception at unread stream to
-	  avoid unspecified behavior.  [ruby-dev:22330]
-
-	* test/ruby/test_system.rb (test_syntax): glob relatively from
-	  __FILE__.
-
-Tue Dec 23 18:09:40 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (pack_pack): remove unnecessary negative value check.
-	  [ruby-dev:22329]
-
-Tue Dec 23 17:26:55 2003  KONISHI Hiromasa  <konishih@fd6.so-net.ne.jp>
-
-	* bcc32/Makefile.sub (config.h): bcc has finite(). [ruby-list:38940]
-
-Tue Dec 23 16:08:16 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/rexml/encodings/US-ASCII.rb: typo.  [ruby-talk:88650]
-
-	* test/ruby/test_system.rb: num of asserts depended on running dir.
-
-	* test/xsd/test_noencoding.rb: rexml + without iconv/uconv cannot
-	  handle euc-jp.  install iconv, uconv or xmlscan.
-
-Tue Dec 23 14:13:51 2003  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/generic.rb (URI::Generic::check_userinfo,
-	  URI::Generic::check_user, URI::Generic::check_password): tests
-	  conflicts/depends with other components closely.
-
-	* test/uri/test_generic.rb (TestGeneric::test_set_component):
-	  added tets.
-
-Tue Dec 23 11:08:34 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/xsd/test_noencoding.rb: rescue Errno::EINVAL and do not test.
-	  "euc-jp" might not be in supported encoding name list.
-	  [ruby-talk:88650]
-
-Tue Dec 23 06:10:31 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb (CGI): add support for mod_ruby.
-
-	* lib/webrick/cgi.rb (CGI::Socket): add check for existence of
-	  OpenSSL module in all HTTPS related methods.
-
-	* lib/webrick/cgi.rb (CGI::Socket#cipher): should create similar
-	  value to OpenSSL::SSLSocket#cipher.
-
-	* lib/webrick/httpresponse.rb (HTTPResponse#setup_header): should
-	  set "connection: close" if @keep_alive is false.
-
-	* lib/webrick/https.rb (HTTPrequest#meta_vars): add supprt for
-	  SSL_PROTOCOL, SSL_CIPHER_USEKEYSIZE and SSL_CIPHER_ALGKEYSIZE.
-
-Mon Dec 22 23:00:05 2003  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/generic.rb (URI::Generic::check_opaque): fixed typo.
-
-Mon Dec 22 21:59:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (map_charset): always ensure code is a String.
-
-Mon Dec 22 21:15:29 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* class.c (rb_mod_init_copy): always copy singleton class.
-	  [ruby-dev:22325]
-
-Mon Dec 22 20:44:36 2003  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/generic.rb (URI::Generic#route_from): accepts urls which
-	  has no host-part.
-
-	* test/uri/test_generic.rb (TestGeneric::test_route): added a test.
-
-Mon Dec 22 20:38:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/cgi.rb: reduce eval.
-
-	* lib/cgi.rb (CGI::QueryExtension::read_multipart): alias path to
-	  local_path.  [ruby-list:38883]
-
-Mon Dec 22 20:09:31 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/test_property.rb: remove duplicated test method.
-
-Mon Dec 22 18:22:04 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* bcc32/Makefile.sub, win32/Makefile.sub (config.h): remove
-	  HAVE_ISINF definition to follow previous commits of missing.h
-	  and win32/win32.h.
-
-Mon Dec 22 17:23:42 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (ac_cv_func_setitimer): moved from defines.h
-
-	* defines.h, rubysig.h, signal.c: removed macro handling which
-	  should be done in configure.
-
-	* configure.in (intrinsics.h): check if present.
-
-	* ruby.h: include intrinsics.h if available.
-
-	* bignum.c, marshal.c: include ieeefp.h if available.
-
-	* missing.h (isinf): define as a macro if finite() and isnan()
-	  are available.  [ruby-core:02032]
-
-Mon Dec 22 17:07:31 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in (mingw): set isnan, finite and isinf to yes.
-
-Mon Dec 22 13:40:19 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/property.rb: passing block by reference.
-
-Mon Dec 22 00:32:43 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_with_disable_interrupt): use ENABLE_INTS instead of
-	  ALLOW_INTS which may switch context.  [ruby-dev:22319]
-
-	* ext/syck/emitter.c (syck_emitter_write): str bigger than
-	  e->bufsize causes buffer overflow.  [ruby-dev:22307]
-
-Sun Dec 21 17:29:00 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* class.c (rb_check_inheritable): new function.  [ruby-dev:22316]
-
-	* intern.h: add prototype.
-
-	* eval.c (superclass): use rb_check_inheritable().
-
-	* object.c (rb_class_initialize): check argument validity.
-
-Sun Dec 21 16:25:10 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#+): re-implemented to resolve ".." in
-	  beginning of the argument.
-	  (Pathname#join): concatenate from the last argument.
-	  (Pathname#parent): just use Pathname#+.
-
-Sun Dec 21 00:12:37 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: add new methods (TkScrollbar#assign, assign_list)
-
-	* ext/tk/sample/tkmultilistframe.rb: use TkScrollbar#assign method
-
-Sat Dec 20 21:59:03 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httprequest.rb (HTTPRequest#meta_vars): refine regexp.
-
-	* lib/webrick/cgi.rb (CGI#start): NPH scripts return status line
-	  instead of Status: header field.
-
-	* lib/webrick/cgi.rb (CGI::Socket): refine some coditions.
-
-Sat Dec 20 16:07:14 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Completion::complete): wrong
-	  Regexp for word boundary.  pointed out by Gavin Sinclair.
-
-	* lib/optparse.rb (OptionParser::make_switch): [no-] prefix was
-	  missing.
-
-Sat Dec 20 11:40:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/yaml.rb (YAML::YAML): adjust Marshal version.
-
-Sat Dec 20 03:56:02 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_with_disable_interrupt): prohibit thread context
-	  switch during proc execution.  [ruby-dev:21899]
-
-Sat Dec 20 02:41:02 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/cgi.rb: add file. (yet another CGI library)
-
-	* MANIFEST: add lib/webrick/cgi.rb.
-
-Sat Dec 20 02:18:31 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-calculate-indent): proper indentation
-	  inside of parentheses.  [ruby-dev:22308]
-
-Fri Dec 19 21:24:22 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httprequest.rb (HTTPRequest#meta_vars): should not set
-	  HTTP_CONTENT_TYPE and HTTP_CONTENT_LENGTH.
-
-	* lib/webrick/https.rb (HTTPRequest#parse): should check presence
-	  of cert() method to detect SSLSocket.
-
-Fri Dec 19 22:56:46 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/property.rb (SOAP::Property#load): new method for loading
-	  property value into existing property tree.
-
-	* test/soap/test_property.rb: add test.
-
-Fri Dec 19 19:21:49 2003  akira yamada  <akira@ruby-lang.org>
-
-	* lib/runit/cui/testrunner.rb (RUNIT::CUI::TestRunner::run):
-	  should use Test::Unit::UI::{PROGRESS_ONLY,VERBOSE}.
-
-Fri Dec 19 17:36:49 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tkmultilistbox.rb: bug fix
-
-	* ext/tk/sample/tkmultilistframe.rb: new sample script
-
-Fri Dec 19 03:44:27 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httputils.rb (parse_form_data): should return an
-	  empty Hash if the body is empty.
-
-Thu Dec 18 21:47:35 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/mkmf.rb (create_makefile): should remove deffile if it's
-	  made by miniruby. based on nobu's patch.
-
-Thu Dec 18 21:44:21 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* eval.c (stack_extend): ignore inline optimization on VC7.
-
-	* win32/Makefile.sub (OS, RT): can override.
-
-	* win32/Makefile.sub (LDFLAGS): ditto. shouldn't use pdb:none
-	  option. based on Tietew's patch [ruby-dev:22289]
-
-Thu Dec 18 16:38:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (fnmatch): unlike find_dirsep(), rb_path_next() never
-	  return NULL.
-
-Thu Dec 18 15:27:59 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/ipaddr.rb (IPSocket::getaddress): merge usa's patch.
-	  [ruby-dev:21678]
-
-Wed Dec 17 15:15:30 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/cgi.rb (CGI::QueryExtension::Value::[]): should work like
-	  String#[] if more than one arguments are specified.
-
-	* lib/delegate.rb: avoid using common instance name as "@obj".
-
-	* lib/cgi.rb (CGI::QueryExtension::Value): Value is no longer
-	  subclass of String, but DelegateClass(String).
-
-	* ext/curses/extconf.rb: restore function check for init_color.
-	  [ruby-list:38905]
-
-	* Makefile.in: need to specify $(MAINLIBS) for the miniruby
-	  generation rule.
-
-	* configure.in: better FreeBSD -lc_r support.
-
-Wed Dec 17 00:16:14 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* ext/strscan/strscan.c: new method
-	  StringScanner#beginning_of_line? (alias #bol?)
-
-	* ext/strscan/strscan.c: new method StringScanner#concat and #<<.
-
-	* ext/strscan/strscan.c: StringScanner#new(str) does not duplicate
-	  nor freeze STR (allow destructive modification).
-
-	* test/strscan/test_stringscanner.rb: test new methods above.
-
-	* test/strscan/test_stringscanner.rb: test destructive string
-	  modification.
-
-Tue Dec 16 21:20:47 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pp.rb: don't use local variable `pp'.
-
-	* lib/prettyprint.rb: ditto.
-
-Tue Dec 16 13:20:43 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: condition bug of if statement on
-	  {pack,grid}_propagate methods
-
-Tue Dec 16 03:17:29 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml/rubytypes.rb: comments in strings. [ruby-talk:88012]
-
-	* test/yaml/test_yaml.rb: add test.
-
-Tue Dec 16 01:14:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (catch_timer): check rb_thread_crtical in main native
-	  thread.
-
-	* eval.c (thread_timer): just sends signals periodically, to
-	  prevent main native thread from receiving them in critical
-	  section.  [ruby-core:01959]
-
-Mon Dec 15 13:32:22 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (check_dirname): check string safety and remove extraneous
-	  trailing directory separators.  [ruby-dev:22279]
-
-	* file.c: renamed and externalized rb_path_next,
-	  rb_path_skip_prefix, rb_path_last_separator, rb_path_end.
-
-	* intern.h: prototypes for rb_path_next, rb_path_skip_prefix,
-	  rb_path_last_separator, rb_path_end.
-
-Mon Dec 15 09:27:46 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/openssl/ossl_pkcs12.c (ossl_pkcs12_initialize): first argument
-	  of rb_protect should take an argument of VALUE.
-
-Sun Dec 14 18:46:48 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/socket/socket.c (Init_socket): IPv6 is not supported although
-	  AF_INET6 is defined on MinGW.
-
-	* lib/ipaddr.rb (AF_INET6): workaround in the environment which does
-	  not support IPv6.
-
-Sat Dec 13 18:55:16 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/charset_alias.rb: preserve original order.
-
-	* ext/iconv/extconf.rb: remove wrapper file at clean.
-
-Sat Dec 13 18:09:42 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (thread_timer): use timer by sub-thread and nanosleep.
-	  [ruby-talk:87519]
-
-	* gc.c (Init_stack): no stack adjustment for THREAD_SAFE.
-
-Sat Dec 13 17:17:59 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (proc_alloc): cache the created object at first time.
-	  [ruby-talk:61288], [ruby-dev:22240]
-
-Sat Dec 13 09:01:23 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in: check ucontext.h.
-
-	* eval.c: use getcontext/setcontext() instead of setjmp/longjmp()
-	  on ia64 or with native thread enabled.  [ruby-core:01932]
-
-Sat Dec 13 03:09:14 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* lib/yaml/rubytypes.rb: anonymous struct fix. [ruby-core:01946]
-
-	* test/yaml/test_yaml.rb: add test.
-
-Fri Dec 12 22:36:44 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/csv.rb: add Cell#to_str and Cell#to_s for /.../ =~ aCell,
-	  "#{aCell}" and so on.
-
-	* test/csv/test_csv.rb: add tests.
-
-Fri Dec 12 19:33:06 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mkdir): remove trailing `/' from pathes.
-
-	* lib/fileutils.rb (rmdir): ditto. [ruby-dev:22238]
-
-	* lib/fileutils.rb (rmdir_r): ditto.
-
-	* lib/fileutils.rb (fu_copy_dir): check if it is a directory after
-	  mkdir(2).
-
-Fri Dec 12 06:06:09 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (proc_invoke): fix class name in warning message for
-	  define_method.  [ruby-dev:22235]
-
-Thu Dec 11 21:24:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_pkcs12.[ch]: new files. add OpenSSL::PKCS12.
-
-	* ext/openssl/ossl.[ch]: ditto.
-
-	* ext/openssl/MANIFEST: add ossl_pkcs12.[ch].
-
-Thu Dec 11 20:54:28 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mkdir_p): remove trailing `/' befere mkdir(2).
-	  mkdir("nonexistdir/") does not work on NetBSD/Alpha 1.6.1.
-
-	* lib/fileutils.rb (fu_list): call to_str for all arguments.
-
-Thu Dec 11 20:07:01 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/ftools.rb (makedirs): sync with fileutils.
-
-Thu Dec 11 19:53:03 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mkdir_p): catch all SystemCallErrors.
-	  (mkdir("C:\") causes EACCESS on Windows 2000/NTFS)
-
-Thu Dec 11 19:08:02 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (mkdir_p): check if it is a directory after
-	  mkdir(2) instead of before mkdir(2), to avoid race condition.
-	  [ruby-talk:87730]
-	  Refer: mkinstalldirs sh script, GNU mkdir(1) (coreutils 5.0)
-
-Thu Dec 11 18:49:30 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb: def m( arg ) -> def m(arg).
-
-Thu Dec 11 11:39:43 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (ieeefp.h), numeric.c: needed for finite() on
-	  Solaris.  [ruby-core:01921]
-
-	* file.c (rb_stat_inspect): adjust format specifier.
-
-	* parse.c (arg_prepend): nodetype() is for debug use.
-
-	* ruby.h (ISASCII, etc): cast to int to get rid of warning.
-
-	* ruby.h (alloca.h): include even in GCC.  [ruby-core:01925]
-
-	* ext/bigdecimal/bigdecimal.c (GetVpValue): adjust format
-	  specifier.
-
-	* ext/bigdecimal/bigdecimal.c (BigDecimal_prec, BigDecimal_coerce,
-	  BigDecimal_divmod): use rb_assoc_new() to suppress memory usage.
-
-	* ext/bigdecimal/bigdecimal.c (BigDecimal_split): ditto.
-
-	* ext/dl/sym.c (rb_dlsym_guardcall): guard itself should be
-	  volatile.
-
-	* ext/iconv/iconv.c (iconv_convert): ensure actual parameter with
-	  format specifier.
-
-	* ext/pty/pty.c (MasterDevice, SlaveDevice, deviceNo): do not
-	  define unless used.
-
-	* ext/pty/pty.c (getDevice): get rid of warning.
-
-	* ext/socket/socket.c (port_str, sock_s_getaddrinfo,
-	  sock_s_getnameinfo): FIX2INT() now returns long.
-
-	* ext/socket/socket.c (init_inetsock_internal): uninitialized
-	  variable.
-
-	* ext/syck/rubyext.c (syck_parser_assign_io): add prototype.
-
-	* ext/syck/rubyext.c (rb_syck_mktime, yaml_org_handler): use
-	  ISDIGIT() instead of isdigit() to avoid warnings and for
-	  platforms which don't support non-ascii charater.
-
-Wed Dec 10 19:28:56 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): set EOF flag at short read.
-	  [ruby-dev:22223], [ruby-dev:22224]
-
-Wed Dec 10 18:07:25 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/erb.rb: new method ERB#filename(=). [ruby-dev:22208]
-
-Wed Dec 10 17:54:51 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): do not set EOF flag when
-	  requested length is zero.  [ruby-dev:22214]
-
-Wed Dec 10 17:17:18 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (read_all): should return given string even if data read is
-	  empty.  [ruby-dev:22207]
-
-Wed Dec 10 17:16:06 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): adjust behavior at reading
-	  beyond EOF to IO.  [ruby-dev:22205]
-
-	* test/ruby/ut_eof.rb (TestEOF::Seek): test behaviors at reading
-	  beyond EOF.
-
-	* test/ruby/test_file.rb, test/stringio/test_stringio.rb: include
-	  TestEOF::Seek test case.
-
-Wed Dec 10 15:01:19 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* test/monitor/test_monitor.rb (test_cond): use Queue#deq
-	  instead of sleep.
-
-Wed Dec 10 14:45:39 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/pty/pty.c (HAVE_SYS_IOCTL_H): need to include <sys/ioctl.h>
-	  for TIOCSCTTY on *BSD.  based on gotoyuzo's patch.
-	  (ruby-bugs:PR#1211)
-
-	* ext/pty/pty.c (establishShell): should close descriptors if fork
-	  failed.
-
-Wed Dec 10 12:53:05 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* win32/win32.h: define execv() using do_aspawn().
-
-	* process.c (proc_exec_v): remove #ifdef's which stopped needing.
-
-Tue Dec  9 23:32:23 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb, ext/tk/lib/tkcanvas.rb, ext/tk/lib/tkdialog.rb,
-	  ext/tk/lib/tkentry.rb, ext/tk/lib/tkscrollbox.rb, ext/tk/lib/tktext.rb,
-	  ext/tk/sample/tkalignbox.rb, ext/tk/sample/tkcombobox.rb,
-	  ext/tk/sample/tkmultilistbox.rb, ext/tk/sample/tkoptdb.rb, ext/tk/sample/tktextframe.rb,
-	  ext/tk/sample/demos-en/dialog1.rb, ext/tk/sample/demos-en/dialog2.rb,
-	  ext/tk/sample/demos-jp/dialog1.rb, ext/tk/sample/demos-jp/dialog2.rb:
-	  overrided instance methods, which are private methods on the super
-	  class, are changed to 'private'
-
-Tue Dec  9 19:53:02 2003  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/generic.rb (URI::Generic#route_from0): make case insensitive
-	  for host-part.
-
-	* test/uri/test_generic.rb (test_route): added tests for the above
-	  change.
-
-Tue Dec  9 14:10:48 2003  Tanaka Akira  <akr@m17n.org>
-
-	* io.c (rb_io_check_readable): don't call io_seek if EOF flag is set,
-	  to avoid clearing EOF flag.
-	  (rb_io_check_writable): ditto.
-
-Tue Dec  9 02:53:55 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tkalignbox.rb: new sample script
-
-Tue Dec  9 00:45:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: renamed #assert_raises to #assert_raise
-	  and made the former call the latter. [ruby-core:01890]
-
-	* test/testunit/test_assertions.rb: ditto.
-
-Tue Dec  9 00:07:35 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/rpc/standaloneServer.rb: add 'shutdown' and 'status'
-	  methods as delegates to WEBrick.
-
-	* test/soap/calc/{test_calc.rb,test_calc2.rb},
-	  test/soap/helloworld/test_helloworld.rb,
-	  test/wsdl/datetime/test_datetime.rb, test/wsdl/raa/test_raa.rb:
-	  follow the change.
-
-Mon Dec  8 22:48:03 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/autorunner.rb: remove dependency to a particular
-	  runner.  [ruby-core:01901], [ruby-list:38869]
-
-	* lib/test/unit/ui/testrunnerutilities.rb: moved output level
-	  constants from Console.
-
-	* lib/test/unit/ui/console/testrunner.rb: ditto.
-
-	* lib/test/unit/ui/{fox,gtk,gtk2,tk}/testrunner.rb (initialize):
-	  accept output_level.
-
-Mon Dec  8 15:03:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/syck/syck.c (syck_io_str_read): get rid of buffer overflow.
-
-Mon Dec  8 13:02:11 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/uri/common.rb: new method URI.regexp. [ruby-dev:22121]
-
-	* test/uri/test_common.rb: add test for URI.regexp.
-
-Mon Dec  8 12:44:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c: define swap16 and swap32 only if they are not
-	  defined. OpenBSD defines these macros. [ruby-dev:22181]
-
-Sun Dec  7 20:54:17 2003  Tanaka Akira  <akr@m17n.org>
-
-	* ext/iconv/iconv.c (map_charset): make case sensitive.
-	  ext/iconv/charset_alias.rb (charset_alias): don't ignore
-	  config.charset's information.  sort aliases.
-
-Sat Dec  6 22:58:03 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c (ossl_start_ssl): new function to wrap
-	  SSL_connect and SSL_accept; if SSL_connect (or SSL_accept) returned
-	  but not finished the handshake process, we should retry it.
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_connect): call ossl_start_ssl.
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_accept): ditto.
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_read): allow signal traps.
-
-Sat Dec  6 21:45:10 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* io.c (flush_before_seek): flush before seek on any platform.
-
-	* configure.in: ditto.
-
-Sat Dec  6 17:23:00 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/soap.rb(SOAP::Env.getenv): allow upcase environment variable
-	  as well as downcase one.
-
-	* lib/soap/netHttpClient.rb(SOAP::NetHttpClient#proxy=): check URI.
-
-Fri Dec  5 23:22:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb (Test::Unit::Assertions::assert_raises,
-	  Test::Unit::Assertions::assert_nothing_raised): use the last
-	  argument as message unless class object.
-
-	* test/testunit/test_assertions.rb (test_assert_raises): test for
-	  multiple exception list.  [ruby-core:01891]
-
-	* test/testunit/test_assertions.rb (test_assert_nothing_raised): test
-	  for non-exception classes.
-
-Fri Dec  5 22:23:04 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/netHttpClient.rb: proxy support did not work.  fixed.
-
-	* lib/soap/property.rb: add class methods for loading property from
-	  stream/file/propertyfile.  propertyfile is a file which is located at
-	  somedir in $:.
-
-	* lib/soap/soap.rb, lib/soap/wsdlDriver.rb, lib/soap/rpc/driver.rb,
-	  lib/wsdl/importer.rb: load property from propertyfile 'soap/property'
-	  e.g. /usr/local/lib/ruby/site_ruby/1.8/soap/property.
-
-	* test/soap/test_property.rb, test/soap/test_streamhandler.rb: new file.
-
-Fri Dec  5 17:26:23 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_exec_end_proc): maintain tmp_end_procs.
-	  [ruby-dev:22154]
-
-Fri Dec  5 13:36:59 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_exec_end_proc): should not clear end_procs and
-	  ephemeral_end_procs before execution. [ruby-dev:22144]
-
-	* eval.c (rb_obj_extend): call Module#extended hook after
-	  extended_object.  [ruby-list:38866]
-
-	* object.c (Init_Object): Module#extended defined.
-
-Fri Dec  5 13:17:30 2003  Tanaka Akira  <akr@m17n.org>
-
-	* test/ruby/test_pipe.rb: use IO.pipe instead of IO.popen.
-
-Fri Dec  5 11:54:45 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): follow IO#read.
-
-	* test/ruby/ut_eof.rb, test/ruby/test_file.rb, test/ruby/test_pipe.rb,
-	  test/stringio/test_stringio.rb: add EOF test.
-
-Fri Dec  5 02:49:35 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb (Test::Unit::Assertions::assert_raises):
-	  allow multiple exception list.  [ruby-core:01884]
-
-	* lib/test/unit/assertions.rb (Test::Unit::Assertions::assert_nothing_raised):
-	  check whether arguments are subclass of Exception.
-
-Thu Dec  4 23:54:00 2003  Rick Ohnemus  <rick.ohnemus@systemware.com>
-
-	* dln.c (aix_loaderror): should not use member named 'errno' which
-	  might be a macro (e.g. on AIX).
-
-Thu Dec  4 23:32:26 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (read_all): do not depend on lseek position.
-	  [ruby-dev:22026]
-
-Thu Dec  4 22:37:26 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval): preserve $! value when retry happens in the
-	  rescue clause.  [ruby-talk:86697]
-
-Thu Dec  4 21:50:07 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/drb/drb.rb (DRb::DRbMessage::send_request, send_reply):
-	  should rescue errors and re-raise DRbConnError on write too.
-	  [ruby-dev:22132]
-
-Thu Dec  4 16:41:17 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (exc_list): allow expanding list.  [ruby-dev:22134]
-
-Thu Dec  4 14:09:24 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb (test_cp): test if the error is
-	  kind of SystemCallError.  It is needless details that which errno
-	  is set on each systems.
-
-Thu Dec  4 13:24:13 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/monitor.rb: use Object#__send__ instead of Object#send.
-
-Thu Dec  4 13:17:45 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/streamHandler.rb: support latest released version of
-	  http-access2.
-
-Thu Dec  4 13:04:44 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/soap.rb: add SOAP::Env module for environment repository
-	  such as HTTP_PROXY.
-
-	* lib/soap/property.rb: property implementation.
-
-	* lib/soap/streamHandler.rb, lib/soap/wsdlDriver.rb,
-	  lib/soap/rpc/driver.rb: use soap/property.rb.
-
-	* lib/wsdl/importer.rb, lib/soap/wsdlDriver.rb, lib/soap/rpc/driver.rb:
-	  use SOAP::Env.
-
-	* lib/soap/netHttpClient.rb: add basic_auth, ssl_config, and cookie
-	  management interface, but ignored for now.
-
-	* lib/xsd/charset.rb: add XSD::Charset.encoding= interface to set
-	  wiredump charset explicitly.  it was fixed to 'utf-8' when iconv or
-	  uconv module was found.
-
-Thu Dec  4 10:43:58 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/dl/sym.c (rb_dlsym_guardcall): __declspec(noinline) is VC7
-	  feature.
-
-Thu Dec  4 10:27:12 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: update hyperlink to the Japanese document.
-
-Thu Dec  4 09:12:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (asn1time_to_time): should check that
-	  the underlying value of ASN1_TIME isn't NULL. [ruby-core:01881]
-
-Thu Dec  4 08:29:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/server.rb (GenericServer#start): should rescue
-	  Exception to avoid unexpected aborting. [ruby-core:01853]
-
-	* lib/webrick/server.rb (GenericServer#start_thread): should check
-	  that peeraddr isn't nil before printing.
-
-	* lib/webrick/httpresponse.rb (HTTPResponse#start_thread): should
-	  rescue Exception to avoid unexpected aborting of thread.
-
-Thu Dec  4 03:48:59 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#link, Pathname#symlink): obsoleted.
-	  (Pathname#make_link, Pathname#make_symlink): new method.
-
-Thu Dec  4 01:45:24 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (argf_read): should not terminate on empty string; wait
-	  until real EOF.  [ruby-dev:21969]
-
-	* io.c (argf_read): should adjust length to read, when length is
-	  specified and read spans command line argument files.
-
-Wed Dec  3 19:38:36 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: correct fcntl parameter. [ruby-dev:22120]
-
-Wed Dec  3 13:49:07 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: 'format'==>'Kernel.format' (avoid override trouble)
-
-	* ext/tk/lib/tkafter.rb: ditto.
-
-	* ext/tk/lib/tkcanvas.rb: ditto.
-
-	* ext/tk/lib/tkdialog.rb: ditto.
-
-	* ext/tk/lib/tktext.rb: ditto.
-
-Wed Dec  3 13:28:13 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* Makefile.in (lex.c): try gperf first, and copy from the source
-	  directory if failed.  [ruby-dev:22123]
-
-	* ext/extmk.rb (MTIMES): let makefiles depend to mkmf.rb.
-
-	* lib/mkmf.rb (configuration): DLDFLAGS was duplicated.
-
-Tue Dec  2 23:18:12 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: wrote the warning about HTTP_PROXY environment
-	  variable.
-
-Tue Dec  2 21:31:42 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* bin/testrb: new test runner.  [ruby-core:01845]
-
-	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner.run,
-	  Test::Unit::AutoRunner#process_args): take test list to run and
-	  options.
-
-	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner::RUNNERS,
-	  Test::Unit::AutoRunner#run): should not exit inside a library,
-	  just return the result instead.
-
-	* lib/test/unit.rb: ditto.
-
-	* test/runner.rb: exit with the test result.
-
-Tue Dec  2 20:18:48 2003  Eric Sunshine  <sunshine@sunshineco.com>
-
-	* configure.in (AC_PROG_YACC): AC_DEFINE(OLD_YACC) if Yacc is found
-	  instead of Bison or byacc.
-
-	* parse.y: If OLD_YACC is defined, ensure that YYMAXDEPTH is at least
-	  10000 (Bison's default) since some old versions of Yacc define it as
-	  low as 150 by default, which is too low for Ruby to parse some files,
-	  such as date/format.rb.  Among other issues, the parse problem causes
-	  "make test" to fail.
-
-Tue Dec  2 20:03:20 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb: check if Pathnames are usable
-	  for arguments.
-
-Tue Dec  2 04:22:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: fixed #assert_no_match message.
-
-	* test/testunit/test_assertions.rb: ditto.
-
-Tue Dec  2 00:43:00 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/syck.c: string buffering bug.  decrementing by full
-	  max_size now. [ruby-core:01834]
-
-Mon Dec  1 21:33:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* numeric.c (num_sadded): prohibit singleton method definition for
-	  Numerics.  fill yet another gap between Fixnum and Bignum.
-
-Mon Dec  1 17:33:47 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (htov16): converts endian using swap16. htov32(), hton16,
-	  hton32 as well. [ruby-talk:85377]
-
-	* pack.c (swap16): swap 2 bytes no matter how big short is on the
-	  platform.  swap32() is also prepared.
-
-	* numeric.c (rb_num2int): returns long to preserve information.
-	  rb_fix2int(), rb_num2uint(), rb_fix2uint() as well.
-	  [ruby-talk:85377]
-
-	* numeric.c (rb_num2uint): should not check for value range if the
-	  source value is negative.
-
-Mon Dec  1 17:14:34 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* sample/optparse/opttest.rb: added.
-
-Mon Dec  1 16:10:52 2003  Dave Thomas  <dave@pragprog.com>
-
-	* lib/rdoc/rdoc.rb: (etc) initial merge into main tree.
-
-Mon Dec  1 14:17:49 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (fu_each_src_dest0): call #to_str to allow
-	  Pathname for arguments. [ruby-core:01795]
-
-	* test/fileutils/test_fileutils.rb: does much strict test on
-	  "same" files detecting.
-
-Mon Dec  1 09:28:14 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
-	  (XCFLAGS): re-export $(XCFLAGS).
-
-	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
-	  (ARCH_FLAG): export $(ARCH_FLAG) (perhaps empty value).
-
-Mon Dec  1 01:03:27 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/mkmf.rb (TRY_LINK, link_command): added support for DLDFLAGS
-	  and ARCH_FLAG.  [ruby-dev:22085]
-
-Sun Nov 30 20:18:07 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: keep ARCH_FLAG separate. export ARCH_FLAG.
-	  [ruby-core:01819]
-
-	* Makefile.in: add ARCH_FLAG to CFLAGS.
-
-	* Makefile.in: add @CPPFLAGS@ to CPPFLAGS.
-
-	* lib/mkmf.rb (link_command, cc_command): use ARCH_FLAG.
-
-	* lib/mkmf.rb (configuration): add ARCH_FLAG to DLDFLAGS.
-
-	* Makefile.in: add ARCH_FLAG to DLDFLAGS.
-
-	* configure.in: should put getcwd in AC_CHECK_FUNCS, not
-	  AC_REPLACE_FUNCS.  [ruby-core:01826]
-
-Sun Nov 30 18:22:48 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: do not override CCDLDFLAGS, LDFLAGS, XLDFLAGS,
-	  DLDFLAGS and LDSHARED.
-
-	* configure.in: XCFLAGS for compiling ruby itself.  ARCH_FLAG is
-	  reflected in CFLAGS.
-
-	* lib/mkmf.rb: ditto.  do not import XCFLAGS from config.status.
-
-Sun Nov 30 17:37:36 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: bug fix [ruby-talk:86746]
-
-Sun Nov 30 13:02:00 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/encodingstyle/soapHandler.rb: refactoring - Simplifying
-	  Conditional Expressions.
-
-	* lib/wsdl/soap/definitions.rb: refactoring - Move Method.
-
-	* test/xsd/{test_noencoding.rb,noencoding.xml}: new files.  test for
-	  encoding unspecified XML file parsing.
-
-	* test/wsdl/{test_fault.rb,map,datetime}: new files.  test of
-	  SOAPFault, dateTime and Apache's Map.
-
-Sun Nov 30 09:35:14 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* string.c (rb_str_update): get rid of SEGV at just allocated String.
-	  [ruby-core:01812]
-
-Fri Nov 28 23:19:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (gc_mark): explicitly check mark recursion levels, instead
-	  of unreliable stack length.
-
-Fri Nov 28 22:49:56 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/rinda/rinda.rb: fix TupleSpaceProxy#read, read_all.
-
-Fri Nov 28 21:44:40 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* test/fileutils/test_fileutils.rb (test_ln_s): should be a file, not
-	  a directory for FreeBSD.
-
-Fri Nov 28 19:37:56 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* hash.c (env_has_value, env_index): must match exactly.
-
-	* test/ruby/test_env.rb (test_has_value, test_index): condition for
-	  aboves.
-
-Fri Nov 28 17:59:20 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/ruby/test_env.rb: add tests for ENV.
-
-Fri Nov 28 17:47:46 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb (DRbMessage#load): rescue Errno::* and raise
-	  DRbConnError.
-
-Fri Nov 28 15:41:15 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#realpath): obsolete the force_absolute
-	  argument.
-
-Fri Nov 28 14:41:52 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/streamHandler.rb: drop unused http parameters.
-
-	* lib/soap/encodingstyle/soapHandler.rb, lib/soap/mapping/factory.rb,
-	  lib/soap/mapping/mapping.rb, lib/soap/mapping/registry.rb,
-	  lib/wsdl/soap/complexType.rb: ApacheSOAP's map support was broken
-	  under WSDL dynanic client environment.  fixed.
-
-	* test/wsdl/raa/*: add tests.
-
-	* lib/xsd/datatypes.rb: dateTime precision bug fix (at least, I hope.)
-	  bug of soap4r.  XSDDateTimeImple.to_time passed a Float to
-	  Time.local/Time.gm as an usec, and NUM2LONG(rb_num2long for Float)
-	  causes rounding error.
-
-	* test/soap/test_basetype.rb, test/xsd/test_xsd.rb: add tests.
-
-Fri Nov 28 04:15:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (method_arity): used wrong Proc object.  [ruby-talk:86504]
-
-Fri Nov 28 00:47:29 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_f_exit), process.c (rb_f_exit_bang): treat true as
-	  success, false as failure.  [ruby-dev:22067]
-
-	* eval.c (rb_f_abort, rb_thread_switch), process.c (rb_f_system): use
-	  ANSI macro instead of hard coded value.
-
-	* eval.c (rb_f_exit), process.c (rb_f_exit_bang): use VALUEs not but
-	  TYPEs.
-
-Thu Nov 27 22:05:48 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* eval.c, gc.c: FreeBSD/ia64 currently does not have a way for a
-	  process to get the base address for the RSE backing store, so
-	  hardcode it for the moment.
-	  [submitted by: Marcel Moolenaar <marcel@FreeBSD.org>]
-
-Thu Nov 27 17:36:42 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tkafter.rb: bug fix on TkTimer#cancel_on_exception=(mode).
-	  TkTimer#wait recieves the exception of the callback.
-	  The exception is kept on @return_value.
-
-Thu Nov 27 16:58:48 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* win32/win32.c (rb_w32_stat): remove _fullpath() for NUL: device.
-
-Wed Nov 26 15:38:47 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* test/fileutils/test_fileutils.rb (test_ln_s): should take the
-	  existing symbolic link for OpenBSD.
-
-Wed Nov 26 04:48:42 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/token.c: removed YYTOKTMP references which
-	  were causing buffer overflows on large block scalars,
-	  comments, quoted scalars and plain scalars.
-
-	* ext/syck/rubyext.c: dynamic changing of buffer size.
-
-	* ext/syck/syck.h: default buffer size of 4k.
-
-Wed Nov 26 00:55:30 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpresponse.rb: add HTTPResponse#keep_alive=.
-
-	* lib/webrick/httpserver.rb (HTTPServer#run): should pass the
-	  request's keep_alive flag to the response.
-
-Tue Nov 25 21:41:35 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* defines.h (ENV_IGNORECASE): should define when DOSISH without
-	  human68k. [ruby-dev:22047]
-
-	* hash.c (env_has_value, env_index): don't ignore case of value.
-	  [ruby-dev:22048]
-
-Tue Nov 25 21:39:37 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* file.c (path_check_1): honor sticky bits always.
-	  [ruby-talk:86273]
-
-Tue Nov 25 20:02:14 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb: do test in more deep
-	  directory.
-
-	* test/fileutils/test_nowrite.rb: ditto.
-
-Tue Nov 25 19:04:23 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::Generic#find_proxy): ENV case sensitivity test
-	  refined.
-
-Tue Nov 25 18:13:30 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb: chdir Dir.tmpdir before each
-	  test. [ruby-dev:22045]
-
-	* test/fileutils/test_nowrite.rb: ditto.
-
-Tue Nov 25 17:52:11 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::Generic#find_proxy): use http_proxy under CGI
-	  if the environment variable is case sensitive.
-
-Tue Nov 25 16:41:33 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/wsdl/multiplefault.wsdl, test/wsdl/test_multiplefault.rb:
-	  removed.  this test requires extra libraries in soap4r/1.5.*.
-
-Tue Nov 25 16:24:42 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/**/*.rb, lib/wsdl/**/*.rb, lib/xsd/**/*.rb: changed license;
-	  GPL2 -> Ruby's.
-
-	* lib/soap/rpc/driver.rb, lib/soap/wsdlDriver.rb,
-	  lib/soap/streamHandler.rb: add interface to streamhandler.
-
-	* lib/soap/marshal.rb: raise error if parse fails.
-
-	* lib/soap/netHttpClient.rb: add https support.  Patched by
-	  Oliver M. Bolzer.
-
-	* lib/soap/netHttpClient.rb: dump HTTP response message body by itself.
-
-	* lib/soap/rpc/driver.rb, lib/soap/rpc/proxy.rb,
-	  lib/soap/wsdlDriver.rb: add driver#mandatorycharset interface to foce
-	  using charset for parsing response from buggy server.
-
-	* lib/soap/encodingstyle/soapHandler.rb: support Apache Axis's half
-	  typed multi-ref array.
-
-	* lib/soap/mapping/factory.rb, lib/soap/mapping/registry.rb: map
-	  SOAPStruct which has multi-accessors which name are the same, to an
-	  array.
-
-	* lib/soap/rpc/element.rb: fixed illegal parameter order.
-
-	* lib/soap/rpc/element.rb: element name of response message could have
-	  the name other than 'return'.
-
-	* lib/wsdl/operation.rb, lib/wsdl/operationBinding.rb,
-	  lib/wsdl/soap/classDefCreator.rb, lib/wsdl/soap/methodDefCreator.rb,
-	  lib/wsdl/soap/methodDefCreatorSupport.rb: WSDL/1.1 allows plural
-	  fault definition in a operation. [ruby-talk:84948]
-
-	* test/wsdl/multiplefault.wsdl, test/wsdl/test_multiplefault.rb: add
-	  test for above fix.
-
-	* lib/wsdl/soap/complexType.rb: support WSDL array definition with
-	  maxOccures="unbound".
-
-	* lib/xsd/charset.rb: use cp932 under emx.  Patched by
-	  Siena. / SHINAGAWA, Norihide in [ruby-dev:21972]
-
-	* lib/xsd/xmlparser/parser.rb: set @charset nil by default.  Nil means
-	  'follow encoding declaration in XML'.
-
-	* sample/soap/digraph.rb, sample/wsdl/amazon/wsdlDriver.rb,
-	  sample/wsdl/googleSearch/sampleClient.rb,
-	  sample/wsdl/googleSearch/wsdlDriver.rb,
-	  test/wsdl/test_emptycomplextype.rb,
-	  test/wsdl/marshal/test_wsdlmarshal.rb,
-	  test/xsd/test_xmlschemaparser.rb: use File.open(...) { |f| f.read }
-	  instead of File.open(...).read. [ruby-dev:21964]
-
-	* test/wsdl/emptycomplextype.wsdl, test/wsdl/test_emptycomplextype.rb:
-	  simplify the test case.
-
-	* test/wsdl/axisArray/*: add tests for axis's array encoding.
-
-Tue Nov 25 16:15:29 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ruby.h: don't treat Cygwin as Windows.
-
-Tue Nov 25 15:18:28 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* configure.in: change default value of --enable-pthread (default: no)
-
-Tue Nov 25 07:31:16 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (primary): allow newlines just before right argument
-	  parenthesis.  (ruby-bugs:PR#1221)
-
-Mon Nov 24 23:32:06 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (OpenURI.open_loop, URI::HTTP#proxy_open): use
-	  catch/throw for redirection instead of exception.
-	  (OpenURI.open_loop, OpenURI.redirectable?): restrict redirection.
-
-Mon Nov 24 19:59:48 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (URI::Generic#find_proxy): use CGI_HTTP_PROXY
-	  instead of HTTP_PROXY in the CGI environment.
-
-Mon Nov 24 19:32:55 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/etc/extconf.rb: check for pw_passwd in struct passwd and
-	  gr_passwd in struct group for DJGPP.
-
-	* ext/etc/etc.c: ditto.
-
-	* ext/Setup.dj: support for curses, etc, zlib.
-
-Mon Nov 24 17:00:00 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb: validate option names.
-	  :content_length_proc and :progress_proc option implemented.
-
-Mon Nov 24 14:53:10 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
-	  (XCFLAGS): output empty value instead of `-DRUBY_EXPORT'.
-
-Sat Nov 22 23:09:45 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: set enable_pthread to no on MinGW.
-
-Sat Nov 22 22:56:20 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* configure.in: add --enable-pthread option (default: yes)
-
-Sat Nov 22 22:48:46 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: add Tk.grab_release and fix bug of TkComposite
-
-	* ext/tk/lib/tkafter.rb: bug fix of TkAfter#start
-
-	* ext/tk/sample/tkcombobox.rb: new sample script
-
-	* ext/tcltklib/tcltklib.c: add native thread check
-
-Sat Nov 22 18:49:47 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/curses/curses.c (window_nodelay): nodelay() of NetBSD's
-	  libcruses returns no value, just like keypad().
-
-Sat Nov 22 17:36:36 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
-	  (HAVE_GETCWD): output to config.h.
-
-	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
-	  (XCFLAGS): output to config.status.
-
-Sat Nov 22 13:10:10 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (have_st_ino?): djgpp has valid st_ino.
-
-Sat Nov 22 11:28:48 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (Init_stack): stack region is far smaller than usual if
-	  pthread is used.
-
-Sat Nov 22 07:30:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/util/backtracefilter.rb: fixed a bug that occurred
-	  when an exception had no backtrace.
-
-	* test/testunit/util/test_backtracefilter.rb: ditto.
-
-Fri Nov 21 16:44:18 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tkentry.rb: fix the encoding trouble of percent
-	  substitutions on validatecommand option of TkEntry widget
-
-	* ext/tk/lib/tk.rb: fix bug on {pack|grid}_propagate() method
-
-Fri Nov 21 16:12:11 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* ruby.1: Fix markups and grammar.
-
-Fri Nov 21 14:49:42 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* ruby.1: wrote about ruby related environment variables.
-
-Fri Nov 21 12:28:03 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_extended): singleton methods should not be checked
-	  when dumping via marshal_dump() or _dump(). [ruby-talk:85909]
-
-Fri Nov 21 01:40:00 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* configure.in: check <pthread.h>
-
-	* ruby.h: include pthread.h if existence.
-	  define is_ruby_native() macro when not HAVE_NATIVETHREAD
-
-	* eval.c: undef is_ruby_native() function when not HAVE_NATIVETHREAD
-
-Fri Nov 21 00:43:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: use #__send__ instead of #send.
-
-	* lib/test/unit/testcase.rb: ditto.
-
-Thu Nov 20 19:19:22 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in: don't find the Cygwin's pthread library on MinGW.
-
-Thu Nov 20 19:15:50 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (have_st_ino?): emx (OS/2 with EMX) does not
-	  have st_ino (always 0). [ruby-dev:21972]
-
-	* lib/fileutils.rb (rename_cannot_overwrite_file?): emx does not
-	  allow overwriting files by rename(2).
-
-	* test/fileutils/test_fileutils.rb: windows? ->
-	  have_drive_letter?, have_file_perm?
-
-Thu Nov 20 17:50:58 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/tkballoonhelp.rb: new sample script
-
-	* ext/tk/sample/tkmultilistbox.rb: ditto
-
-	* ext/tk/sample/tktextframe.rb: ditto
-
-Thu Nov 20 13:37:34 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ruby.h: define is_ruby_native_thread() for no native thread
-	  environment
-
-	* eval.c: ditto
-
-Thu Nov 20 12:42:47 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* configure.in: always check existence of the pthread library
-
-	* ruby.h: define macros for ruby's native thread check
-
-	* eval.c: add ruby's native thread check
-
-	* gc.c: ditto
-
-Wed Nov 19 14:45:18 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb (to_ary): print more friendly warning message.
-
-Wed Nov 19 14:32:08 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (fu_same?): add djgpp and wince.
-
-	* lib/fileutils.rb (cannot_overwrite_file?): add wince.
-
-Wed Nov 19 11:04:47 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/fileutils.rb (cannot_overwrite_file?, have_st_ino?): bccwin32
-	  is same as mswin32.
-
-Wed Nov 19 07:54:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit.rb: do not run tests if $! is set.
-
-	* lib/test/unit/assertionfailederror.rb: extend StandardError instead
-	  Exception (irb catches the former but not the latter).
-
-Tue Nov 18 23:31:36 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* missing/memmove.c (memmove): take void *, not char *.
-
-	* missing.h (memmove): ditto.
-
-	* missing.h (strchr, strrchr): return char *, not int.
-
-Tue Nov 18 22:20:10 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (fu_same?): temporal fix for windows.
-
-Tue Nov 18 19:05:04 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (fu_same?): check by inode instead of path
-	  name, to detect two hard links pointing to the same content.
-
-	* test/fileutils.rb: did not create correctly looped symlinks.
-
-Tue Nov 18 18:23:05 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_read): behave as IO at empty string.
-	  [ruby-dev:21939], [ruby-dev:21941]
-
-	* ext/stringio/stringio.c (strio_getc, strio_getline): set EOF flag.
-
-	* ext/stringio/stringio.c (strio_rewind, strio_seek, strio_ungetc):
-	  clear EOF flag.
-
-	* test/stringio/test_stringio.rb: imported from [ruby-dev:21941].
-
-Tue Nov 18 14:06:35 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/fileutils.rb (fu_each_src_dest): raise if src==dest.
-	  [ruby-talk:85344] [ruby-core:01699]
-
-	* lib/fileutils.rb: use Object#is_a? instead of Class#=== to allow
-	  e.g. remote objects for receivers.
-
-	* lib/fileutils.rb: FileTest -> File.
-
-	* lib/fileutils.rb: put parentheses for arguments of File.xxxx?
-
-	* test/fileutils/test_fileutils.rb (test_cp): test "cp a a".
-
-	* test/fileutils/test_fileutils.rb (test_mv): test "mv a a".
-
-	* test/fileutils/test_fileutils.rb (test_ln): test "ln a a".
-
-	* test/fileutils/test_fileutils.rb (test_ln_s): test "ln_s a a".
-
-	* test/fileutils/test_fileutils.rb (test_install): test "install a a".
-
-	* test/fileutils/fileasserts.rb: new method assert_symlink.
-
-	* test/fileutils/fileasserts.rb: assert_is_directory -> assert_directory.
-
-Mon Nov 17 19:38:49 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* file.c (getcwdofdrv): avoid using getcwd() directly, use
-	  my_getcwd() instead.
-
-	* merged NeXT, OpenStep, Rhapsody ports patch from Eric Sunshine
-	  <sunshine@sunshineco.com>.  [ruby-core:01596]
-
-Mon Nov 17 10:50:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Completion::complete): allow least
-	  common completion for three or more candidates.
-
-Mon Nov 17 09:41:38 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/ui/tk/testrunner.rb,
-	  lib/test/unit/ui/gtk/testrunner.rb:
-	  run GUI main loop in sub thread.
-
-	* lib/test/unit/ui/gtk2/testrunner.rb: imported from rough.
-
-	* lib/test/unit/autorunner.rb (keyword_display): sort keywords.
-
-Sun Nov 16 18:10:57 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval): iterator should return value from next inside
-	  begin/rescue/end.  (ruby-bugs:PR#1218)
-
-Sun Nov 16 13:26:07 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_object): LINK check earlier than anything else,
-	  i.e. do not dump TYPE_IVAR for already dumped objects.
-	  (ruby-bugs:PR#1220)
-
-	* eval.c (rb_eval): call "inherited" only when a new class is
-	  generated; not on reopening.
-
-	* eval.c (eval): prepend error position in evaluating string to
-	  "mesg" attribute string only when it's available and is a
-	  string.
-
-Sun Nov 16 12:16:10 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/protocol.rb: logging response body. [experimental]
-	  [ruby-list:38800]
-
-Sun Nov 16 10:49:38 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
-
-	* lib/thread.rb (Thread.exclusive): wrap method definition in
-	  class Thread to enable rdoc to process.
-
-Sun Nov 16 09:45:23 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb (set_debug_output): warn if method is called
-	  after #start.  [ruby-dev:38798]
-
-Sun Nov 16 04:41:33 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (eval): do not re-raise exception to avoid unnecessary
-	  exception copying, instead modify exception and internal
-	  information to adjust eval().
-
-	* eval.c (backtrace): can return the current frame information
-	  only if lev < -1.
-
-Sat Nov 15 22:16:42 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* /ext/openssl/ossl_x509ext.c (ossl_x509extfactory_create_ext):
-	  refine error message.
-
-Sat Nov 15 10:05:40 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (OpenURI.open_loop, OpenURI::HTTP#proxy_open):
-	  refactored to support options.
-	  (Buffer): maintain size by this class.
-
-Sat Nov 15 07:40:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_method_node): new API to retrieve method body.
-
-Fri Nov 14 13:21:30 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: fix (en-bugged at 2003/11/07)
-
-	* ext/tk/lib/tkdialog.rb: TkDialog.new accepts a parent widget
-	  argument [ruby-talk:85066]
-
-Thu Nov 13 20:53:35 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (Kernel[#.]open): hard coded URI schemes removed.
-	  [ruby-ext:02251]
-
-Thu Nov 13 19:17:00 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* lib/test/unit/ui/tk/testrunner.rb: use grid and panedwindow
-	  (if available)
-
-Thu Nov 13 17:56:41 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/open-uri.rb (OpenURI.open_uri): use File::RDONLY.
-	  reported by Take_tk <ggb03124@nifty.ne.jp>.
-	  [ruby-ext:02245]
-
-Thu Nov 13 16:45:53 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509req.c (ossl_x509req_to_der): add function for
-	  X509::Request#to_der.
-
-Thu Nov 13 11:31:14 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Completion#complete): prior shorter
-	  name to containing longer name.
-
-Thu Nov 13 06:08:54 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: stop freezing some classes
-
-	* ext/tk/lib/multi-tk.rb: ditto.
-
-Wed Nov 12 17:32:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb (assert_throws, assert_nothing_thrown):
-	  uncaught throw in sub thread raises ThreadError.
-
-	* lib/test/unit/ui/tk/testrunner.rb (setup_ui): "expand" is not
-	  necessary.
-
-Wed Nov 12 14:09:43 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* test/monitor/test_monitor.rb: fix the timing problem by Queue.
-
-Wed Nov 12 12:59:44 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* test/monitor/test_monitor.rb: added.
-
-Wed Nov 12 10:14:28 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/monitor.rb: refactored. Thanks, Gennady Bystritsky.
-
-Wed Nov 12 06:11:39 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.c (ossl_x509_sk2ary, ossl_x509crl_sk2ary):
-	  add functions to convert STACK into Array.
-
-	* ext/openssl/ossl.h: add prototypes.
-
-	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_set_certificates,
-	  ossl_pkcs7_get_certificates, ossl_pkcs7_get_crls,
-	  ossl_pkcs7_set_crls): add functions for PKCS7#certificates=
-	  PKCS7#certificates, PKCS7#crls= and PKCS7#crls.
-
-Wed Nov 12 00:47:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/ui/testrunnermediator.rb: should require 'test/unit'.
-
-Tue Nov 11 23:54:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/ui/gtk/testrunner.rb: added a rescue clause to handle
-	  the case when the requested font is not available.
-
-Tue Nov 11 22:44:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (appendline): file may not end with newline.  a bug if
-	  READ_DATA_PENDING_PTR is defined. [ruby-talk:84925]
-
-Tue Nov 11 10:42:41 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: raise an exception when creating TkWindow
-	  object, because TkWindow class is an abstract class.
-
-Tue Nov 11 03:30:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/ext/openssl/ossl_conf.c (ossl_config_get_value): return nil
-	  if the specified value doesn't exist.
-
-	* lib/ext/openssl/ossl_conf.c (ossl_config_get_section): return
-	  a empty hash if the specified section doesn't exist.
-
-Mon Nov 10 11:40:29 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/monitor.rb (wait): return true on signal/broadcastfalse and
-	  false on timeout. Thanks Gennady Bystritsky.
-
-Mon Nov 10 00:07:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (primary): primary_value may be 0 when syntax error.
-	  [ruby-talk:84893]
-
-Sun Nov  9 02:05:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: un-deprecated #assert_not_nil to
-	  maintain symmetry with #assert_nil. Also added better output for
-	  #assert_kind_of.
-
-	* test/testunit/tc_assertions.rb: ditto.
-
-Sat Nov  8 18:50:20 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/wsdl/raa/*: add new testcase for WSDL loading, parsing and
-	  reading.
-
-	* test/soap/marshal/*: backport from soap4r/1.5.1.  all differences are
-	  for ruby/1.6.
-
-	* lib/soap/*: backport from soap4r/1.5.1.  all differences are for
-	  ruby/1.6.
-
-	* lib/wsdl/data.rb, lib/wsdl/xmlSchema/data.rb: move definition of
-	  ArrayTypeAttrName from ::WSDL::XMLSchema::* to ::WSDL::*.
-	  [ruby-talk:84813]
-
-	* lib/wsdl/soap/definitions.rb: element name typo in custom exception
-	  struct definition which is needed for wsdlDriver; camelCase ->
-	  underscore_name.
-
-Sat Nov  8 13:49:50 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* configure.in: improvement of pthread check
-
-Sat Nov  8 13:28:46 2003  Takaaki Tateishi  <ttate@ttsky.net>
-	* ext/dl/sym.c: Add DL.win32_last_error and DL.last_error.
-	  Thanks, Kaoru Shirai.
-
-Sat Nov  8 06:19:38 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: To fix 'pthread-enabled Tcl/Tk' problem,
-	  TclTkIp#_eval calls Tcl_Eval() on the mainloop thread only
-	  (queueing a handler to the EventQueue).
-
-	* ext/tcltklib/README.1st: edit the description of '--with-pthread-ext'
-
-Fri Nov  7 23:23:04 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (Pathname#+): if self or the argument is `.', return
-	  another.
-	  (Pathname#parent): if self is `.', return `..'.
-	  (Pathname#children): if self is `.', don't prepend self for a
-	  pathname in a result.
-	  (Pathname#join): re-implemented using Pathname#+.
-	  (Pathname#find): if self is `.', remove `./' prefix of yielding
-	  pathname.
-
-Fri Nov  7 10:23:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/socket/socket.c (make_hostent): get rid of SEGV on aliases
-	  lookup failure.  (ruby-bugs:PR#1215)
-
-Fri Nov  7 04:08:05 2003  UENO Katsuhiro  <katsu@blue.sky.or.jp>
-
-	* ext/zlib/zlib.c (Init_zlib): define Zlib::GzipReader#each_line as
-	  an alias of Zlib::GzipReader#each.
-
-Fri Nov  7 01:03:16 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_load): save and restore rb_prohibit_interrupt.
-	  [ruby-dev:21857]
-
-Thu Nov  6 18:05:07 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* io.c (rb_io_inspect): show the path also at a closed file.
-	  [ruby-dev:21851]
-
-Thu Nov  6 11:42:07 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/stringio/stringio.c (strio_set_string, strio_reopen): check
-	  tainted.
-
-	* ext/stringio/stringio.c (strio_copy, strio_ungetc, strio_write,
-	  strio_putc): add infection.
-
-	* ext/stringio/stringio.c (strio_path): just nil.  [ruby-dev:21846]
-
-	* ruby.c (proc_options): reserve searched script path in the
-	  source file name table.  [ruby-list:38765]
-
-	* lib/optparse.rb (OptionParser::Completion#complete): default not to
-	  ignore case on completion.  [ruby-talk:84726]
-
-	* win32/win32.c (make_cmdvector): process backslashes even if a quote
-	  is not enclosed.
-
-Wed Nov  5 23:49:45 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* sample/openssl/gen_csr.rb: there (at least) is a CA which does not
-	  accept DN in UTF8STRING format.  it's a sample.
-
-Wed Nov  5 22:55:16 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* configure.in, eval.c, signal.c: : add '--with-pthread-ext'
-	  option to fix the pthread trouble on 'tcltklib'
-
-	* ext/tcltklib/README.1st: add the description of '--with-pthread-ext'
-
-	* ext/tk/lib/tktext.rb : add TkText#text_copy, text_cut, text_paste
-	  to support Tcl/Tk8.4's tk_textCopy, tk_textCut, tk_textPaste
-
-	* ext/tk/lib/tk.rb : add TkMenu#set_focus support Tcl/Tk's
-	  tk_menuSetFocus
-
-Wed Nov  5 17:33:45 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_load): allow interrupt during loaded program
-	  evaluation.  [ruby-dev:21834]
-
-	* hash.c (rb_hash_fetch): always warn if default argument and a
-	  block are supplied at the same time. [ruby-dev:21842]
-
-	* hash.c (env_fetch): ditto.
-
-	* array.c (rb_ary_fetch): ditto.
-
-Wed Nov  5 19:08:47 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (OptionParser::Switch::PlacedArgument::parse):
-	  do not remove next argument if empty value is placed.
-
-	* test/optparse: added.
-
-Wed Nov  5 17:05:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/ui/gtk/testrunner.rb: typo.
-
-Wed Nov  5 11:13:32 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* string.c: add #include "version.h". this file still depends on it.
-
-	* Makefile.in, bcc32/Makefile.sub, win32/Makefile.sub,
-	  wince/Makefile.sub: add version.h dependency to string.c.
-
-Wed Nov  5 09:14:23 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/monitor.rb: revert to the previous revision.
-
-Wed Nov  5 08:39:51 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/https.rb (HTTPRequest#parse): set @client_cert_chain.
-
-	* lib/webrick/https.rb (HTTPRequest#meta_vars): create
-	  SSL_CLIENT_CERT_CHAIN_n from @client_cert_chain.
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_get_peer_cert_chain): return nil
-	  if no cert-chain was given.
-
-Tue Nov  4 23:44:48 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub:
-	  remove needless version.h dependency.
-
-Tue Nov  4 23:38:43 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* class.c, hash.c, string.c: remove #include "version.h".
-
-	* Makefile.in: remove needless version.h dependency.
-
-Tue Nov  4 06:54:52 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (read_all): fptr->f may be NULL, if IO is closed in the
-	  signal handler.
-
-	* io.c (io_read): ditto.
-
-	* string.c (get_pat): remove 1.8.0 warning code.
-
-	* string.c (rb_str_match): extend warning until 1.8.2.
-
-	* string.c (rb_str_match2): ditto.
-
-	* class.c (class_instance_method_list): remove 1.8.0 warnings.
-	  method_list now recurs.  [ruby-dev:21816]
-
-	* class.c (rb_obj_singleton_methods): ditto.
-
-	* array.c (rb_ary_select): remove select with block.
-	  [ruby-dev:21824]
-
-	* hash.c (rb_hash_select): ditto.
-
-	* hash.c (env_select): ditto.
-
-	* re.c (match_select): ditto.
-
-	* struct.c (rb_struct_select): ditto.
-
-Mon Nov  3 22:53:21 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/racc/parser.rb: synchronize with Racc 1.4.4.
-
-	* ext/racc/cparse/cparse.c: ditto.
-
-	* ext/racc/cparse/cparse.c (parse_main): should abort when
-	  the length of LR state stack <=1, not ==0.
-
-Mon Nov  3 08:50:47 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* process.c (check_uid_switch): remove duplicated error messages.
-
-	* process.c (check_gid_switch): ditto.
-
-Sun Nov  2 02:28:33 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/ssl.rb: new option :SSLExtraChainCert.
-
-Sun Nov  2 01:02:04 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* string.c (rb_str_hash): Update the HASH_PERL alternative hash
-	  algorithm in sync with Perl 5.8.
-
-	* st.c (strhash): Ditto.
-
-Sat Nov  1 18:21:09 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c (ossl_ssl_peer_cert_chain): add new method
-	  SSLSocket#peer_cert_chain.
-
-	* ext/openssl/ossl_x509req.c (GetX509ReqPtr): new function
-	  which returns underlying X509_REQ.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_set_issuer_cert,
-	  ossl_x509extfactory_set_subject_cert, ossl_x509extfactory_set_crl,
-	  ossl_x509extfactory_set_subject_req, ossl_x509extfactory_set_config):
-	  use underlying C struct without duplication not to leak momory.
-
-Sat Nov  1 01:49:03 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/mapping/factory.rb: mark marshalled basetype objects when
-	  @allow_original_mapping is true.  multi-referencing basetype node is
-	  prohibited in SOAP/1.1 encoding but soap4r's original ruby object
-	  mapping requires basetype to be marked to detect self referencing
-	  loop.  e.g. o = 1; o.instance_eval { @iv = o }  soap4r's original
-	  mapping is only used through soap/marshal API.
-
-	* test/soap/marshal/test_marshal.rb: add tests for self referencing
-	  immutable objects.
-
-	* test/soap/calc/test_calc_cgi.rb: fix test name.
-
-Fri Oct 31 22:26:29 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
-
-	* wince/string_wce.c (strrchr): should decrement pointer.
-
-	* wince/Makefile.sub: correct a range of isdigit().
-
-Fri Oct 31 12:55:24 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* configure.in, lib/mkmf.rb: add RPATHFLAG for NetBSD.
-	  [ruby-dev:21791]
-
-	* bcc32/Makefile.sub, win32/Makefile.sub, win32/Makefile.sub: ditto.
-
-Fri Oct 31 01:38:14 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* wince/Makefile.sub, win32/Makefile.sub (.y.c): allow white spaces
-	  at the beginning of line to remove by sed. (ruby-bugs-ja:PR#580)
-
-Fri Oct 31 01:02:24 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* compar.c (cmp_equal): protect exceptions from <=> comparison
-	  again.  returns nil if any exception or error happened during
-	  comparison.
-
-	* eval.c (search_required): should update *featurep when DLEXT2 is
-	  defined. (ruby-bugs-ja:PR#581)
-
-Thu Oct 30 23:41:04 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/drb.rb: add DRbArray
-
-	* lib/drb/invokemethod.rb: fix Hash#each problem. [ruby-dev:21773]
-
-	* lib/drb/unix.rb: add LoadError. [ruby-dev:21743]
-
-Thu Oct 30 23:19:11 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/generator.rb: better XML pretty printing.
-
-	* lib/soap/encodingstyle/soapHandler.rb: remove unnecessary namespace
-	  assignment in the element which has "encodingStyle" attribute, and
-	  add necessary namespace assignment for "arrayType" attribute.
-
-	* test/soap/calc/test_calc_cgi.rb: take over $DEBUG to ruby process
-	  through CGI.
-
-Thu Oct 30 22:59:39 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/yaml2byte.c: HASH const too long.  Thanks, matz.
-
-Thu Oct 30 19:13:53 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* ext/syck/MANIFEST: Add yamlbyte.h.
-
-Thu Oct 30 14:25:31 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (READ_DATA_BUFFERED): new macro to detect whether stdio
-	  buffer filled.
-
-	* io.c (rb_io_fptr_cleanup): move path deallocation to
-	  rb_io_fptr_finalize (finalizer called by GC).
-
-Thu Oct 30 13:23:39 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (logop): left may be NULL. [ruby-talk:84539]
-
-	* eval.c (rb_eval): NODE_CASE nd_head may be NULL.
-
-Thu Oct 30 10:14:51 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/test/unit/autorunner.rb: make fox runner work.
-
-Thu Oct 30 09:32:26 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* process.c (rb_f_system): fixed lack of security check before
-	  calling do_spawn() on win32. [ruby-talk:84555]
-
-Thu Oct 30 02:46:35 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_invoke): single array value to normal Proc#call
-	  (i.e. not via lambda call), should be treated just like yield.
-	  [ruby-dev:21726]
-
-Thu Oct 30 02:25:48 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/buffering.rb (Buffering#initialize):
-	  add new method to inherit @sync from @io.sync.
-
-	* ext/openssl/lib/net/protocols.rb (SSLIO#ssl_connect): no need to
-	  set sync flag explicitly.
-
-	* ext/openssl/ossl_ssl.c (ossl_sslctx_initialize): call super.
-
-	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): set extra chain
-	  certificates in @extra_chain_cert.
-
-Wed Oct 29 22:02:04 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/drb/drbtest.rb: use rbconfig.rb to make the path of ruby
-	  interpreter to exec, instead of test/ruby/envutil.rb,
-
-Wed Oct 29 19:58:59 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/tcltklib/tcltklib.c (CONST84): define CONST84 when it is not
-	  defined and TCL_MAJOR_VERSION >= 8.
-
-	* ext/tcltklib/tcltklib.c (VwaitVarProc, WaitVariableProc,
-	  rb_threadVwaitProc): use CONST84 instead of CONST.
-
-	* ext/tcltklib/tcltklib.c (ip_rbTkWaitCommand,
-	  ip_rb_threadTkWaitCommand): use CONST84 always.
-
-Wed Oct 29 17:27:05 2003  Tanaka Akira  <akr@m17n.org>
-
-	* re.c (rb_reg_s_union, Init_Regexp): new method `Regexp.union'.
-
-	* lib/pathname.rb (realpath): examine Dir.pwd because it may have
-	  symlinks.
-
-Wed Oct 29 17:16:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_longjmp): must not disturb original jump.
-	  [ruby-dev:21733]
-
-Wed Oct 29 15:28:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (Init_Proc): taint preallocated exception object
-	  sysstack_error. [ruby-talk:84534]
-
-Wed Oct 29 11:27:39 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (ret_args): node may be NULL. [ruby-talk:84530]
-
-Tue Oct 28 15:20:12 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/tcltklib/tcltklib.c (VwaitVarProc, ip_rbVwaitObjCmd,
-	  WaitVariableProc, WaitVisibilityProc, WaitWindowProc,
-	  ip_rbTkWaitObjCmd, ip_rbTkWaitCommand, rb_threadVwaitProc,
-	  rb_threadWaitVisibilityProc, rb_threadWaitWindowProc,
-	  ip_rb_threadVwaitObjCmd, ip_rb_threadTkWaitObjCmd): prototype;
-	  avoid VC++ warnings.
-
-Mon Oct 27 19:19:55 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_longjmp): ignore reentering error while warning.
-	  [ruby-dev:21730]
-
-Mon Oct 27 00:23:50 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_ruby): bug fix on Win : hang-up when
-	  calling 'exit' in the Tk callback procedure. [ruby-list:38656]
-
-Sat Oct 25 09:18:04 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_method_missing): protect exception from within
-	  "inspect".  (ruby-bugs:PR#1204)
-
-Fri Oct 24 23:26:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* hash.c (rb_hash_each): Hash#each should yield single value.
-	  [ruby-talk:84420]
-
-	* hash.c (env_each): ditto for ENV.each.
-
-Thu Oct 23 20:25:32 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/server.rb (GenericServer#start): should rescue
-	  IOError from IO::accept. [ruby-dev:21692]
-
-Thu Oct 23 17:59:36 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (ruby_cleanup): initialize stack bottom for embedding.
-	  [ruby-dev:21686]
-
-	* ext/dl/extconf.rb: move list of files to clean from DEPEND file,
-	  to get rid of macro redefinitions.
-
-Thu Oct 23 13:44:00 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y: integrate operations for stack_type.  [ruby-dev:21681]
-
-Thu Oct 23 00:41:45 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/calc/*, test/soap/helloworld/*: set logging threshold
-	  to ERROR.
-
-Wed Oct 22 12:53:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir#collect_file):
-	  ignore tests which raised LoadError.
-
-	* test/drb/drbtest.rb, test/ruby/test_beginendblock.rb,
-	  test/ruby/test_system.rb: avoid requiring same file twice.
-
-	* test/drb/test_drbssl.rb, test/drb/test_drbunix.rb: should not use
-	  ARGV unless invoked directly.  do not create test cases unless
-	  required libraries are available.
-
-Wed Oct 22 02:31:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (ruby_cleanup): should not ignore exit_value in END
-	  execution. [ruby-dev:21670]
-
-Tue Oct 21 23:16:26 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (ruby_cleanup): call finalizers and exit procs before
-	  terminating threads.
-
-	* eval.c (ruby_cleanup): preserve ruby_errinfo before ruby_finalize_0().
-
-Tue Oct 21 15:57:11 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir#collect_file):
-	  prepend the directory of target file to the load path.
-
-Tue Oct 21 15:08:53 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (do_spawn, do_aspawn): should wait child process even
-	  if callded with P_OVERLAY.
-
-	* win32/win32.c (do_spawn, do_aspawn): should return child's exit
-	  status to parent.
-
-Tue Oct 21 00:35:02 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/calc/*, test/soap/helloworld/*: catch the exception from
-	  test server thread and recover.
-
-Tue Oct 21 00:22:57 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* test/drb/*: import drb/runit.
-
-Mon Oct 20 23:55:47 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval): set current node after arguments evaluation.
-	  [ruby-dev:21632]
-
-	* eval.c (rb_yield_0): set current node and keep it at local jump.
-
-Mon Oct 20 22:01:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_cleanup): keep thread group for main thread.
-	  [ruby-dev:21644]
-
-Mon Oct 20 18:28:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_catch): backout.
-
-Mon Oct 20 17:31:46 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (PUSH_FRAME): generate unique number to be TAG_JUMP()
-	  destination.
-
-	* eval.c (localjump_destination): use unique number in ruby_frame
-	  for localjump destination.
-
-Mon Oct 20 11:31:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/ruby/test_signal.rb (test_signal): restore old trap.
-
-Mon Oct 20 11:00:46 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (gc_sweep): loosen page free condition to avoid add_heap()
-	  race condition. [ruby-dev:21633]
-
-	* gc.c (gc_sweep): do not update malloc_limit when malloc_increase
-	  is smaller than malloc_limit.
-
-Mon Oct 20 09:45:12 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* lib/debug.rb (debug_command): remove debug print.
-
-Wed Oct 20 00:25:41 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (search_required): required name must not be changed before
-	  loading.  [ruby-dev:24492]
-
-Sun Oct 19 13:12:30 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (foreachline, dir_foreach): add obsolete warning.
-
-Sun Oct 19 00:14:22 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/calc/*, test/soap/helloworkd/*: changed port# of test
-	  server. (17171)
-
-Sat Oct 18 23:01:32 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* missing/acosh.c (DBL_MANT_DIG): typo fix(ifdef -> ifndef).
-
-Sat Oct 18 05:48:59 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/rubyext.c: YAML::Syck::compile method.
-
-	* ext/syck/syck.c: Buffer edge bug.
-
-	* ext/syck/yaml2byte.c: YAML to bytecode converter.
-
-	* ext/syck/yamlbyte.h: Ditto.
-
-	* ext/syck/bytecode.c: Bytecode parser fixes to empty collections
-	  and empty strings.
-
-	* ext/syck/token.c: Ditto.
-
-Fri Oct 17 23:07:38 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* ext/enumerator/enumerator.c, ext/enumerator/enumerator.txt:
-	  Provide Kernel#to_enum as an alias for Kernel#enum_for.  Maybe
-	  this is a better name.
-
-Fri Oct 17 23:00:30 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* lib/generator.rb: Add rdoc documentation.
-
-Fri Oct 17 22:16:42 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* lib/set.rb: Reword and fix Overview.
-
-	* lib/set.rb: It is not necessary to require
-	  'test/unit/ui/console/testrunner'.
-
-Fri Oct 17 11:15:22 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* test/ruby/test_range.rb: added.
-
-	* MANIFEST: add test/ruby/test_range.rb.
-
-Fri Oct 17 03:21:23 2003  William Sobel  <will.sobel@barra.com>
-
-	* ext/socket/socket.c (make_hostent): h_aliases may be NULL.
-	  (ruby-bugs:PR#1195)
-
-	* ext/socket/socket.c (sock_s_gethostbyaddr): ditto.
-
-Fri Oct 17 00:12:41 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: (bug fix) instance variable @frame was used
-	  without initializing on TkComposite module.
-
-Thu Oct 16 23:51:04 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: If $DEBUG == true and some exception is caused
-	  in a callback operation, Ruby/Tk shows a (verbose) backtrace
-	  information on the callback process.
-
-Thu Oct 16 17:09:19 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/debug.rb (DEBUGGER__::Context::debug_command): do not call
-	  debug_silent_eval() when $1 is not set. (ruby-bugs:PR#1194)
-
-Thu Oct 16 16:54:57 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (rb_str_upto): ("a"..."a").to_a should return [].
-	  [ruby-core:01634]
-
-Thu Oct 16 16:40:51 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb:
-	  Add Tk::EncodedString and Tk::UTF8_String class to support
-	  characters using the \uXXXX escape to the UNICODE string.
-
-	* ext/tk/sample/{demos-en,demos-jp}/unicodeout.rb
-	  new demo-scripts (samples of Tk::UTF8_String)
-
-	* ext/tk/sample/{demos-en,demos-jp}/widget
-	  add entries for 'unicodeout.rb'
-
-Thu Oct 16 08:38:06 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/digest/test_digest.rb (test_eq): show failed class.
-
-	* test/ruby/test_iterator.rb (test_break, test_return_trace_func):
-	  test localjump destination.
-
-Wed Oct 15 20:22:31 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/netHttpClient.rb: use URI::HTTP#request_uri instead of
-	  instance_eval('path_query').  [ruby-list:38575]
-
-Wed Oct 15 17:24:45 2003  URABE Shyouhei  <root@mput.dip.jp>
-
-	* lib/cgi.rb (CGI::Cookie): tiny typo fix.
-
-Wed Oct 15 15:00:54 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (ruby_run): just return FAILURE instead of parse error
-	  count.  [ruby-list:38569]
-
-Wed Oct 15 13:17:02 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/digest/digest.c (rb_digest_base_alloc): need to initialize
-	  buffer. [ruby-dev:21622]
-
-Wed Oct 15 11:23:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_object): dump extended modules as well.
-
-	* marshal.c (r_object0): TYPE_USRMARSHAL should restore extended
-	  modules before invoking marshal_load.  these two fixes are done
-	  by Masatoshi Seki <m_seki@mva.biglobe.ne.jp>.
-
-Wed Oct 15 09:30:34 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/enumerator/enumerator.c (enumerator_each): avoid VC++ warning.
-
-	* ext/syck/syck.h: include stdio.h for definition of FILE.
-
-Wed Oct 15 08:09:07 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/bytecode.c: Checkin of YAML bytecode support.
-
-	* ext/syck/gram.c: Ditto.
-
-	* ext/syck/syck.c: Ditto.
-
-	* ext/syck/token.c: Ditto.
-
-	* ext/syck/handler.c: Ditto.
-
-	* ext/syck/handler.c: Now using 'tag' rather than 'taguri' in type URIs.
-
-	* ext/syck/rubyext.c: Ditto (on both counts).
-
-Wed Oct 15 05:05:53 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* lib/generator.rb: A new library which converts an internal
-	  iterator to an external iterator.
-
-	* lib/abbrev.rb: A new library which creates an abbreviation table
-	  from a list.
-
-Wed Oct 15 04:31:51 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/sample/demos-en/entry3.rb, ext/tk/sample/demos-jp/entry3.rb :
-	  new demo-scripts
-
-	* ext/tk/sample/demos-en/widget, ext/tk/sample/demos-jp/widget :
-	  add entries for 'entry3.rb'
-
-Wed Oct 15 04:31:47 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* test/digest/test_digest.rb: Moved from ext/digest/test.rb.
-
-Wed Oct 15 03:53:20 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/tk.rb: fixed trouble on auto-load Tcl commands (enbug
-	  on the last commit).
-
-Wed Oct 15 00:25:00 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (yylex): argument parentheses preceded by spaces should
-	  be warned; not error.  [ruby-talk:84103]
-
-Wed Oct 15 00:20:15 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c: replace Tcl/Tk's vwait and tkwait to
-	  switch on threads smoothly and avoid seg-fault.
-
-	* ext/tcltklib/tcltklib.c: add TclTkIp._thread_vwait and
-	  _thread_tkwait for waiting on a thread. (Because Tcl/Tk's vwait
-	  and tkwait command wait on an eventloop.)
-
-	* ext/tk/lib/multi-tk.rb: support TclTkIp._thread_vwait and
-	  _thread_tkwait.
-
-	* ext/tk/lib/tk.rb: now, TkVariable#wait has 2 arguments.
-	  If 1st argument is true, waits on a thread. If false, waits on
-	  an eventloop. If 2nd argument is true, checks existence of
-	  rootwidgets. If false, doesn't. Default is wait(true, false).
-
-	* ext/tk/lib/tk.rb: add TkVariable#tkwait(arg) which is equal to
-	  TkVariable#wait(arg, true). wait_visibility and wait_destroy
-	  have an argument for waiting on a thread or an eventloop.
-
-	* ext/tk/lib/tk.rb: improve of accessing Tcl/Tk's special variables.
-
-	* ext/tk/lib/tkafter.rb: support 'wait on a thread' and 'wait on
-	  an eventloop'.
-
-Wed Oct 15 00:10:24 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/baseData.rb: Introduce SOAPType as the common ancestor of
-	  SOAPBasetype and SOAPCompoundtype.
-
-	* lib/soap/generator.rb, lib/soap/element.rb, lib/soap/encodingstyle/*:
-	  Encoding methods signature change.  Pass SOAPGenerator as a parameter.
-
-	* lib/soap/mapping/*, test/soap/marshal/test_marshal.rb: Refactoring
-	  for better marshalling/unmarshalling support.  Now I think SOAP
-	  marshaller supports all kind of object graph which is supported by
-	  Ruby's original marshaller.  Of course there could be bugs as always.
-	  Find it.  :-)
-
-	* lib/soap/rpc/standaloneServer.rb: Set severity threshould to INFO.
-	  DEBUG is too noisy.
-
-	* lib/xsd/datatypes.rb: DateTime#of is obsoleted.  Use DateTime#offset.
-
-	* test/wsdl/emptycomplextype.wsdl, test/xsd/xmlschema.xml: Avoid
-	  useless warning.
-
-Tue Oct 14 19:09:35 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (ruby_finalize_0): return the given exit status unless
-	  SystemExit got raised.
-
-Tue Oct 14 11:53:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* intern.h (ruby_stop): never return.
-
-	* ruby.h (ruby_run): ditto.
-
-Tue Oct 14 04:43:55 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (realpath): make ELOOP check bit more robust.
-	  (children): prepend self by default.
-	  (chroot): obsoleted.
-
-Tue Oct 14 02:29:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_require_safe): segfault after loading .so.
-
-Tue Oct 14 02:05:23 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* ext/Setup*, ext/enumerator/*: Add ext/enumerator, a helper
-	  module for the Enumerable interface.
-
-Mon Oct 13 23:55:59 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* test/ruby/envutil.rb: use Config::CONFIG["ruby_install_name"],
-	  not "ruby".
-
-Mon Oct 13 23:57:29 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_feature_p): match by classified suffix.
-
-	* eval.c (rb_require_safe): require library in the specified safe
-	  level.
-
-	* variable.c (rb_autoload, rb_autoload_load): restore safe level
-	  when autoload was called.  [ruby-dev:21338]
-
-	* intern.h: prototypes; rb_require_safe.
-
-	* test/runner.rb: accept non-option arguments.
-
-Mon Oct 13 20:49:51 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_new4): should not preserve FL_TAINT status in the
-	  internal shared string. [ruby-dev:21601]
-
-	* string.c (rb_str_new4): ditto.
-
-	* eval.c: use EXIT_SUCCESS and EXIT_FAILURE for exit values.
-
-	* process.c: ditto. [ruby-list:38521]
-
-Mon Oct 13 19:51:02 2003  Koji Arai  <jca02266@nifty.ne.jp>
-
-	* lib/debug.rb (debug_command): should enter emacs mode when
-	  assigned any value to the environment variable "EMACS".
-	  On Meadow, (getenv "EMACS") is "meadow".
-
-Sun Oct 12 14:45:03 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/win32ole/extconf.rb: check "windows.h", not "windows".
-	  [ruby-talk:84051]
-
-Sat Oct 11 20:41:03 2003  Corinna Vinschen  <corinna@vinschen.de>
-
-	* file.c (eaccess): Use access(2) on Cygwin.
-
-Sat Oct 11 17:09:21 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/rexml/quickpath.rb (REXML::QuickPath::match):
-	  escape '[' to avoid warning.
-
-Sat Oct 11 16:08:41 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (realpath): check existence of the file.
-
-	* lib/pathname.rb (realpath): re-implemented.
-	  (realpath_root?, realpath_rec): removed
-
-Sat Oct 11 10:19:39 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* lib/monitor.rb: handle exceptions correctly. Thanks, Gennady
-	  Bystritsky.
-
-Fri Oct 10 07:50:54 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (is_defined): inheritance line adjustment as like as
-	  rb_call_super().
-
-Fri Oct 10 01:19:00 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_initialize): add
-	  optional argument to specify the DirectoryString type
-	  (ASN1::UTF8STRING by default). RFC3280 deprecates PrintableString
-	  for DirectoryString, and strongly requires to use UTF8String for
-	  all certificates issued after December, 31 2003.
-
-	* ext/openssl/lib/openssl/x509.rb (X509::Name::parse): ditto.
-
-Thu Oct  9 23:50:21 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_start_0): prevent thread from GC.
-	  [ruby-dev:21572]
-
-Thu Oct  9 19:11:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_start_0): non-volatile should be restored from
-	  volatile.
-
-Thu Oct  9 17:43:36 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (proc_save_safe_level, proc_get_safe_level,
-	  proc_set_safe_level): save/restore safe level 1..4.
-
-Thu Oct  9 16:33:23 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (r_object0): remove unnecessary iv restoration for
-	  USRMARSHAL. [ruby-dev:21582]
-
-	* marshal.c (w_object): dump generic instance variables from
-	  a string from '_dump'.
-
-	* variable.c (rb_generic_ivar_table): return 0 if obj's FL_EXIVAR
-	  is not set.
-
-	* time.c (time_dump): copy instance variables to dumped string, to
-	  be included in the marshaled data.
-
-	* bignum.c (rb_big2ulong): add range check to ensure round trip.
-
-Thu Oct  9 15:45:27 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* pack.c (uv_to_utf8): change message to "out of range", since
-	  negative values are not "too big". [ruby-dev:21567]
-
-Thu Oct  9 14:05:38 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_set_end_proc, rb_exec_end_proc): restore safe level.
-	  [ruby-dev:21557]
-
-Thu Oct  9 10:51:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_yield_0): no error if block is empty.
-
-Thu Oct  9 06:43:33 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (localjump_error): id should be ID.
-
-	* eval.c (rb_eval): nd_rval is set in copy_node_scope().
-
-	* eval.c (rb_yield_0): unused variable.
-
-	* eval.c (rb_yield_0): nothing to do for empty node.
-
-	* eval.c (call_end_proc, proc_invoke): adjust backtrace in END.
-	  [ruby-dev:21551]
-
-	* eval.c (rb_thread_start_0): set the value by break as the result.
-	  [ruby-dev:21552]
-
-	* eval.c (rb_thread_start_0, rb_thread_raise, rb_callcc): save
-	  variables across THREAD_SAVE_CONTEXT.
-
-Thu Oct  9 12:05:46 2003  Eric Sunshine  <sunshine@sunshineco.com>
-
-	* configure.in: revived NextStep, OpenStep, and Rhapsody ports which
-	  had become unbuildable; enhanced --enable-fat-binary option so that
-	  it accepts a list of desired architectures (rather than assuming a
-	  fixed list), or defaults to a platform-appropriate list if user does
-	  not provide an explicit list; made the default list of architectures
-	  for MAB (fat binary) more comprehensive; now uses -fno-common even
-	  when building the interpreter (in addition to using it for
-	  extensions), thus allowing the interpreter to be embedded into a
-	  plugin module of an external project (in addition to allowing
-	  embedding directly into an application); added checks for
-	  <netinet/in_systm.h> (needed by `socket' extension) and getcwd(); now
-	  ensures that -I/usr/local/include is employed when extensions'
-	  extconf.rb scripts invoke have_header() since extension checks on
-	  NextStep and OpenStep will fail without it if the desired resource
-	  resides in the /usr/local tree; fixed formatting of --help message.
-
-	* Makefile.in: $(LIBRUBY_A) rule now deletes the archive before
-	  invoking $(AR) since `ar' on Apple/NeXT can not "update" MAB archives
-	  (see configure's --enable-fat-binary option); added rule for new
-	  missing/getcwd.c.
-
-	* defines.h: fixed endian handling during MAB build (see configure's
-	  --enable-fat-binary option) to ensure that all portions of the
-	  project see the correct WORDS_BIGENDIAN value (some extension modules
-	  were getting the wrong endian setting); added missing constants
-	  GETPGRP_VOID, WNOHANG, WUNTRACED, X_OK, and type pid_t for NextStep
-	  and OpenStep; removed unnecessary and problematic HAVE_SYS_WAIT_H
-	  define in NeXT section.
-
-	* dir.c: do not allow NAMLEN() macro to trust dirent::d_namlen on
-	  NextStep since, on some installations, this value always resolves
-	  uselessly to zero.
-
-	* dln.c: added error reporting to NextStep extension loader since the
-	  previous behavior of failing silently was not useful; now ensures
-	  that NSLINKMODULE_OPTION_BINDNOW compatibility constant is defined
-	  for OpenStep and Rhapsody; no longer includes <mach-o/dyld.h> twice
-	  on Rhapsody since this header lacks multiple-include protection,
-	  which resulted in "redefinition" compilation errors.
-
-	* main.c: also create hard reference to objc_msgSend() on NeXT
-	  platforms (in addition to Apple platforms).
-
-	* lib/mkmf.rb: now exports XCFLAGS from configure script to extension
-	  makefiles so that extensions can be built MAB (see configure's
-	  --enable-fat-binary option); also utilize XCFLAGS in cc_command()
-	  (but not cpp_command() because MAB flags are incompatible with
-	  direct invocation of `cpp').
-
-	* ext/curses/extconf.rb: now additionally checks for presence of these
-	  curses functions which are not present on NextStep or Openstep:
-	  bkgd(), bkgdset(), color(), curs(), getbkgd(), init(), scrl(), set(),
-	  setscrreg(), wattroff(), wattron(), wattrset(), wbkgd(), wbkgdset(),
-	  wscrl(), wsetscrreg()
-
-	* ext/curses/curses.c: added appropriate #ifdef's for additional set of
-	  curses functions now checked by extconf.rb; fixed curses_bkgd() and
-	  window_bkgd() to correctly return boolean result rather than numeric
-	  result; fixed window_getbkgd() to correctly signal an error by
-	  returning nil rather than -1.
-
-	* ext/etc/etc.c: setup_passwd() and setup_group() now check for null
-	  pointers before invoking rb_tainted_str_new2() upon fields extracted
-	  from `struct passwd' and `struct group' since null pointers in some
-	  fields are common on NextStep/OpenStep (especially so for the
-	  `pw_comment' field) and rb_tainted_str_new2() throws an exception
-	  when it receives a null pointer.
-
-	* ext/pty/pty.c: include "util.h" for strdup()/ruby_strdup() for
-	  platforms such as NextStep and OpenStep which lack strdup().
-
-	* ext/socket/getaddrinfo.c: cast first argument of getservbyname(),
-	  gethostbyaddr(), and gethostbyname() from (const char*) to non-const
-	  (char*) for older platforms such as NextStep and OpenStep.
-
-	* ext/socket/socket.c: include "util.h" for strdup()/ruby_strdup() for
-	  platforms such as NextStep and OpenStep which lack strdup(); include
-	  <netinet/in_systm.h> if present for NextStep and OpenStep; cast first
-	  argument of gethostbyaddr() and getservbyname() from (const char*) to
-	  non-const (char*) for older platforms.
-
-	* ext/syslog/syslog.c: include "util.h" for strdup()/ruby_strdup() for
-	  platforms such as NextStep and OpenStep which lack strdup().
-
-Wed Oct  8 22:19:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit.rb: removed installation instructions.
-
-	* lib/test/unit/ui/testrunnermediator.rb: moved the run flag to a more
-	  central location.
-
-	* lib/test/unit.rb: ditto.
-
-	* lib/test/unit.rb: extracted the running code in to AutoRunner.
-
-	* lib/test/unit/autorunner.rb: added.
-
-	* lib/test/unit/collector/objectspace.rb: extracted common test
-	  collection functionality in to a module.
-
-	* lib/test/unit/collector.rb: ditto; added.
-
-	* test/testunit/collector/test_objectspace.rb: ditto.
-
-	* lib/test/unit/collector/dir.rb: added. Supports collecting tests out
-	  of a directory structure.
-
-	* test/testunit/collector/test_dir.rb: added.
-
-	* test/runner.rb: simplified to use the new capabilities.
-
-Tue Oct  7 15:23:09 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_beginendblock.rb: add tests for nested BEGIN/END.
-
-	* test/ruby/beginmainend.rb: add tests for nested BEGIN/END.
-
-	* test/ruby/endblockwarn.rb: new file added to test of END-in-method
-	  warning.
-
-Tue Oct  7 12:23:47 2003  Tanaka Akira  <akr@m17n.org>
-
-	* ext/fcntl/fcntl.c (Init_fcntl): define Fcntl::O_ACCMODE.
-
-	* ext/socket/extconf.rb: useless assignment removed.
-
-Tue Oct  7 09:13:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/ruby/test_beginendblock.rb (test_endinmethod): END{} is now
-	  allowed in eval.
-
-Tue Oct  7 04:15:25 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (stmt): should not expand mrhs if lhs is solely starred.
-
-Tue Oct  7 02:57:53 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (stmt): rhs of multiple assignment should not be
-	  expanded using "to_a". [ruby-dev:21527]
-
-Tue Oct  7 01:42:34 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1_get_asn1type): use appropriate
-	  free function for ASN1_OBJECT.
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1obj_get_sn): add new function for
-	  ASN1::ObjectId#sn; it returns short name text representation of OID.
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1obj_get_ln): add new function for
-	  ASN1::ObjectId#ln; it returns long name text representation of OID.
-
-	* ext/openssl/ossl_asn1.c (ossl_asn1obj_get_oid): add new function for
-	  ASN1::ObjectId#oid; it returns numerical representation of OID.
-
-Mon Oct  6 22:59:46 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/csv.rb (IOReader, BasicWriter): call binmode when a given IO
-	  respond_to?(:binmode).  record separator was wrong when you gave
-	  text mode IO to Reader.parse and Writer.generate.
-
-	* test/csv/test_csv.rb: add tests for above change.
-
-Sun Oct  5 23:27:09 2003  Tanaka Akira  <akr@m17n.org>
-
-	* ext/socket/extconf.rb: check recvmsg even if sendmsg is exists.
-
-	* ext/socket/socket.c (thread_read_select): restored.
-
-Mon Oct  6 16:23:38 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* marshal.c (w_object): wrong method name in the message.
-
-Mon Oct  6 16:02:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (stmt): END in method should cause warning.
-	  [ruby-dev:21519]
-
-Mon Oct  6 15:17:23 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_iterator.rb (test_block_argument_without_paren):
-	  added. (follows sample/test.rb)
-
-Mon Oct  6 11:57:06 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_beginendblock.rb, test/ruby/beginmainend.rb: added
-	  test for eval-ed BEGIN END order.
-
-Mon Oct  6 09:19:54 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_object): should pass "weak" value to next level.
-	  [ruby-dev:21496]
-
-	* eval.c (proc_alloc): should not use cached object if klass is
-	  different. [ruby-talk:83685]
-
-Sun Oct  5 23:27:09 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb: version information is added in document.
-
-Sun Oct  5 23:07:03 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_f_END): block should be given.  [ruby-dev:21497]
-
-Sun Oct  5 22:51:23 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/ext/openssl/extconf.rb: add check for some engine functions
-	  unavailable in OpenSSL-0.9.6.
-
-	* lib/ext/openssl/ossl_engine.c: ditto.
-
-Sun Oct  5 17:56:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval): fix evaluation order.  [ruby-list:38431]
-
-Sun Oct  5 15:05:06 2003  akira yamada  <akira@ruby-lang.org>
-
-	* test/uri/*: translated RUNIT to Test::Unit.
-
-Sun Oct  5 14:37:39 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/xsd/datatypes.rb: Rational -> Decimal string bug fix.
-
-	* test/soap/marshal/test_marshal.rb: ditto.
-
-	* test/soap/calc/test_calc_cgi.rb: add Config::CONFIG["EXEEXT"] to
-	  RUBYBIN.
-
-Sun Oct  5 13:47:22 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_beginendblock.rb, test/ruby/beginmainend.rb: add tests
-	  about scope, order and allowed syntax.
-
-Sun Oct  5 11:54:29 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/envutil.rb: added.  split "rubybin" from test_system.rb.
-
-	* test/ruby/test_system.rb: use envutil.rb
-
-	* test/ruby/test_beginendblock.rb: added.
-
-	* test/ruby/beginmainend.rb: added.  used in test_beginendblock.rb.
-
-Sun Oct  5 11:23:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* test/testunit/runit/test_testresult.rb: removed some unnecessary
-	  cruft.
-
-Sun Oct  5 11:14:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/rubyunit.rb: aliasing TestCase into the top level is
-	  problematic.
-
-	* lib/runit/assert.rb: fixed a couple of bugs caused by recent
-	  refactoring in Test::Unit.
-
-	* test/testunit/runit/*: added.
-
-Sun Oct  5 10:55:29 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/open-uri.rb (URI::Generic#find_proxy): no_proxy support did not
-	  work.  [ruby-dev:21484]
-
-Sun Oct  5 09:52:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: will use pp for output if available.
-	  Can be disabled by setting Assertions.use_pp = false.
-
-	* test/testunit/test_assertions.rb: made a small change to exception
-	  formatting.
-
-Sun Oct  5 07:42:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: made small improvements to assertion
-	  messages. Deprecated Assertions#assert_not_nil; use #assert instead.
-
-	* test/testunit/test_assertions.rb: ditto.
-
-	* test/testunit/util/test_procwrapper.rb: use #assert instead of
-	  #assert_not_nil.
-
-Sun Oct  5 04:10:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: refactored message building.
-
-Sun Oct  5 03:40:22 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.h: global symbols should be declared
-	  as external.
-
-Sun Oct  5 03:03:20 2003  akira yamada  <akira@ruby-lang.org>
-
-	* test/ruby/test_exception.rb (test_else): added.
-
-Sun Oct  5 02:12:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: changed assertion messages to rely more
-	  heavily on #inspect. Added backtrace filtering for exceptions in
-	  assertion messages.
-
-	* test/testunit/test_assertions.rb: ditto.
-
-Sun Oct  5 02:12:00 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
-
-	* lib/drb/acl.rb, lib/drb/ssl.rb: added.
-
-	* lib/drb/drb.rb: exit from a thread using 'break'.
-
-Sat Oct  4 21:49:14 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* gc.c (Init_stack): the type of space is changed to unsigned int
-	  from double.  [ruby-dev:21483]
-
-Sat Oct  4 17:52:59 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/netHttpClient.rb: follow http-access2.  hosts which matches
-	  ENV['no_proxy'] or ENV['NO_PROXY'] are not proxyed.
-	  - [,:] separated. ("ruby-lang.org:rubyist.net")
-	  - no regexp. (give "ruby-lang.org", not "*.ruby-lang.org")
-	  - if you want specify host by IP address, give full address.
-	    ("192.168.1.1, 192.168.1.2")
-
-	* lib/soap/rpc/cgistub.rb: return "Status: XXX MMM" line.
-
-	* test/runner.rb: give testsuite name.
-
-Sat Oct  4 15:16:02 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_object): instance variable dump do not cause error
-	  for objects that cannot be dumped, if they traversed from
-	  marshal_dump.  they are just ignored.
-
-	* gc.c (Init_stack): cast "space" (doble value) into unsigned
-	  int.  should run on PowerPC.
-
-	* eval.c (rb_eval): should not execute else part if any exception
-	  is caught. [ruby-dev:21482]
-
-	* parse.y (f_args): should allow unparenthesized block argument.
-
-	* parse.y (f_rest_arg): should allow unparenthesized rest
-	  argument.
-
-Sat Oct  4 14:59:51 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (initialize): raise ArgumentError if argument has
-	  '\0' character.
-	  (relative_path_from): new method.
-	  (each_entry): new method for replacement of dir_foreach.
-	  (foreach, foreachline, dir_foreach, chdir): obsoleted.
-
-Sat Oct  4 12:58:48 2003  akira yamada  <akira@ruby-lang.org>
-
-	* test/uri/* (6 files): added.
-
-Sat Oct  4 12:44:45 2003  akira yamada  <akira@ruby-lang.org>
-
-	* lib/uri/ftp.rb, lib/uri/mailto.rb: renamed to #to_s from #to_str.
-
-Sat Oct  4 07:33:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/testsuite.rb: changed #<< to return self, and added
-	  #delete.
-
-	* test/testunit/test_testsuite.rb: ditto. Also slightly refactored
-	  #test_size.
-
-	* lib/test/unit/collector/objectspace.rb: collector now preserves the
-	  hierarchy of suites.
-
-	* test/testunit/collector/test_objectspace.rb: ditto.
-
-Sat Oct  4 04:48:49 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/rubyext.c: default keys handled.
-
-	* ext/syck/syck.h: lowered default buffer size to 16k for increased
-	  performance.
-
-	* test/yaml: checkin of basic unit tests.
-
-Sat Oct  4 04:24:19 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/extconf.rb: add check for X509V3_set_nconf.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_set_config):
-	  cannot implement if X509V3_set_nconf doesn't exist.
-
-Sat Oct  4 02:12:44 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/xsd/datatypes.rb: dump sign by itself.  under the problematic
-	  platform, sprintf("%+.10g", -0.0) => +0.  sigh.
-
-	* sample/wsdl/amazon/*: update schema ver2 to ver3.
-
-Sat Oct  4 01:33:46 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb (initialize): duplicate and freeze argument.
-	  (to_s): return duplicated string.
-	  (children): new method.
-	  (each_line): new alias to foreachline.
-
-Fri Oct  3 16:13:19 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_asn1.c: add DER encoder and decoder.
-
-	* ext/openssl/ossl_asn1.h: add OpenSSL::ASN1 module.
-
-	* ext/openssl/ossl.c (Init_openssl): call Init_ossl_asn1.
-
-	* ext/openssl/extconf.rb: check if X509_ATTRIBUTE has field "single".
-
-	* ext/openssl/ossl_x509attr.c (ossl_x509attr_set_value): accept
-	  DER encoded data argument.
-
-	* ext/openssl/ossl_x509attr.c (ossl_x509attr_get_value): return
-	  DER encoded data in OpenSSL::ASN1 types.
-
-Fri Oct  3 13:02:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit.rb: refactored to use optparse.
-
-	* lib/test/unit.rb: added support for selecting the output
-	  level from the command-line.
-
-	* lib/test/unit.rb: added a command-line switch to stop processing
-	  the command-line, allowing arguments to be passed to tests.
-
-	* lib/test/unit.rb: changed the method for specifying a runner or a
-	  filter from the command-line.
-
-	* lib/test/unit/collector/objectspace.rb: fixed a bug causing all
-	  tests to be excluded when the filter was set to an empty array.
-
-	* test/testunit/collector/test_objectspace.rb: ditto.
-
-Fri Oct  3 08:14:32 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/irb/ruby-lex.rb (RubyLex::identify_identifier): support
-	  'class ::Foo' syntax. [ruby-talk:83514]
-
-Fri Oct  3 08:01:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: added a default message for #assert,
-	  #assert_block, and #flunk.
-
-	* test/testunit/test_assertions.rb: ditto.
-
-	* lib/test/unit/failure.rb: failures now show a better trace of where
-	  they occurred.
-
-	* test/testunit/test_failure.rb: ditto (added).
-
-	* lib/test/unit/testcase.rb: ditto.
-
-	* test/testunit/test_testcase.rb: ditto.
-
-	* lib/test/unit/util/backtracefilter.rb: added.
-
-	* test/testunit/util/test_backtracefilter.rb: added.
-
-	* lib/test/unit/error.rb: changed to use BacktraceFilter and improved
-	  output.
-
-	* test/testunit/test_error.rb: ditto.
-
-Thu Oct  2 20:33:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (iconv_failure_initialize): conform with
-	  orthodox initialization method.
-
-	* ext/iconv/iconv.c (iconv_fail): initialize exception instance
-	  from the class, and do not share instance variables with the
-	  others.  [ruby-dev:21470]
-
-Thu Oct  2 18:20:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* time.c (Init_Time): define initialize.  [ruby-dev:21469]
-
-Thu Oct  2 17:39:38 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_engine.c: add a new module OpenSSL::Engine.
-	  it supports OpenSSL hardware cryptographic engine interface.
-
-	* ext/openssl/ossl_engine.h: ditto.
-
-	* ext/openssl/MANIFEST: add ossl_engine.c and ossl_engine.h.
-
-	* ext/openssl/extconf.rb: add check for openssl/engine.h.
-
-	* ext/openssl/ossl.c: call Init_ossl_engine().
-
-	* ext/openssl/ossl.h: include openssl/engine.h.
-
-	* ext/openssl/ossl_pkey_{rsa,dsa,dh}.c: check if underlying
-	  EVP_PKEY referes engine.
-
-Thu Oct  2 17:22:37 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* time.c (time_load): restore instance variables (if any) before
-	  loading from marshaled data.
-
-Thu Oct  2 14:19:15 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (iconv_fail): now yield erred substring, and
-	  set error object to $!.
-
-	* ext/iconv/iconv.c (iconv_convert): error handler block should
-	  return appended part and the rest.  if rest is nil, the
-	  conversion stops.
-
-Thu Oct  2 12:00:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* variable.c (rb_const_defined_0): look up constants in Object as
-	  well.  [ruby-dev:21458]
-
-	* test/ruby/test_defined.rb (TestDefined::test_defined): test for
-	  constants.
-
-Thu Oct  2 11:17:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/assertions.rb: should not capture an
-	  AssertionFailedError unless explicitly requested.
-
-	* test/testunit/test_assertions.rb: ditto.
-
-	* test/testunit/collector/test_objectspace.rb: fixed a test failure
-	  caused by methods being returned in different orders on different
-	  platforms by moving test sorting from TestSuite into the locations
-	  where suites are constructed. [ruby-talk:83156]
-
-	* lib/test/unit/testcase.rb: ditto.
-
-	* lib/test/unit/testsuite.rb: ditto.
-
-	* lib/test/unit/collector/objectspace.rb: ditto.
-
-Thu Oct  2 03:25:01 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* eval.c (rb_thread_raise): prototype; avoid VC++ warning.
-
-Thu Oct  2 01:37:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* time.c (time_mdump): new marshal dumper. _dump is still
-	  available for compatibility.
-
-	* time.c (time_mload): new marshal loader.
-
-	* marshal.c (w_object): preserve instance variables for objects
-	  with marshal_dump.
-
-	* marshal.c (r_object0): restore instance variables before calling
-	  marshal_load.
-
-	* error.c (rb_warn_m): always return nil.
-
-Thu Oct  2 01:32:46 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_f_block_given_p): real required condition is
-	  ruby_frame->prev->iter == ITER_CUR.
-
-	* eval.c (rb_block_given_p): ditto.
-
-	* eval.c (block_pass): update ruby_frame->iter only when previous
-	  value is ITER_NOT.
-
-Thu Oct  2 01:02:35 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* variable.c (rb_const_defined_at): should exclude constants from
-	  Object when TYPE(klass) == T_MODULE *and* exclude is on.
-	  [ruby-dev:21458]
-
-	* variable.c (rb_const_get_0): do not lookup constants from Object
-	  when TYPE(klass) == T_MODULE *and* exclude is on.
-
-Thu Oct  2 00:21:11 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/logger/test_logger.rb: unlinking file before close causes
-	  problem under win32 box.
-
-	* lib/xsd/datatypes.rb(XSDFloat, XSDDouble): add +/- sign explicitly
-	  when stringified and embedded into XML instance.  Ruby's sprintf may
-	  format -0.0 as "0.0" (no minus sign) depending on underlying C
-	  sprintf implementation.
-
-	* test/xsd/test_xsd.rb, test/soap/test_basetype.rb: follow above change.
-
-	* test/soap/calc/*: give httpd config param "CGIInterpreter".
-	  "/usr/bin/env ruby" thing does not work under non-Unix boxes.
-
-Sat Oct  2 00:42:20 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (r_byte): retrieve pointer from string value for each
-	  time.  [ruby-dev:24404]
-
-	* marshal.c (r_bytes0): ditto.
-
-	* enum.c (sort_by_i): re-entrance check added.  [ruby-dev:24399]
-
-	* io.c (io_read): should freeze all reading buffer.
-	  [ruby-dev:24400]
-
-	* string.c (rb_str_sum): should use bignums when bits is greater
-	  than or equals to sizeof(long)*CHAR_BITS. [ruby-dev:24395]
-
-	* eval.c (specific_eval): defer pointer retrieval to prevent
-	  unsafe sourcefile string modification.  [ruby-dev:24382]
-
-	* string.c (rb_str_sum): wrong cast caused wrong result.
-	  [ruby-dev:24385]
-
-	* enum.c (enum_sort_by): hide temporary array from
-	  ObjectSpace.each_object.  [ruby-dev:24386]
-
-	* string.c (rb_str_sum): check was done with false pointer.
-	  [ruby-dev:24383]
-
-	* string.c (rb_str_sum): string may be altered.  [ruby-dev:24381]
-
-Thu Oct  2 00:25:21 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* signal.c (ruby_signal_name): adjust to the prototype.
-
-	* process.c (pst_inspect): ditto.
-
-	* ext/etc/etc.c (etc_getgrent, Init_etc): typo.
-
-Wed Oct  1 20:49:41 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* gc.c (heaps): manage slots and limits together.  [ruby-dev:21453]
-
-	* gc.c (add_heap): should not clear heaps slot even if realloc()
-	  failed.
-
-Wed Oct  1 20:36:49 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* MANIFEST: add wince/mkconfig_wce.rb.
-
-Wed Oct  1 17:22:33 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/etc/etc.c: add new functions: setpwent, getpwent, endpwent,
-	  setgrent, getgrent, endgrent.
-
-	* ext/socket/socket.c (sock_s_gethostbyname): do not reverse lookup.
-
-Wed Oct  1 17:01:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_load): Object scope had priority over required file
-	  scope.  [ruby-dev:21415]
-
-Wed Oct  1 14:09:53 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
-
-	* wince/mkconfig_wce.rb: sorry, forget to commit.
-
-Wed Oct  1 10:08:42 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
-
-	* wince/setup.mak: add sigmarionIII SDK support.
-
-	* wince/Makefile.sub: ditto.
-
-	* wince/mkexports.rb: fix linker error in SH4.
-
-	* wince/mkconfig_wce.rb: camouflage RUBY_PLATFORM for compiling ext.
-
-Wed Oct  1 08:02:52 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
-
-	* wince/time_wce.c (time): add zero check.
-
-Tue Sep 30 16:11:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* Makefile.in: copy lex.c from $(srcdir) if it's not the current
-	  directory.  [ruby-dev:21437]
-
-Tue Sep 30 11:29:23 2003  Tanaka Akira  <akr@m17n.org>
-
-	* process.c (pst_inspect): describe stopped process "stopped".
-
-Tue Sep 30 09:31:56 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/runner.rb: glob for directories.
-
-Tue Sep 30 09:11:43 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_eval): while/until should not capture break unless
-	  they are destination of the break.
-
-Tue Sep 30 03:12:02 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb (finish): revert to 1.93.
-
-	* lib/net/pop.rb (finish): revert to 1.60.
-
-	* lib/net/smtp.rb (finish): revert to 1.67.
-
-	* lib/net/http.rb (do_start): ensure to close socket if failed to
-	  start session.
-
-	* lib/net/pop.rb (do_start): ditto.
-
-	* lib/net/smtp.rb (do_start): ditto.
-
-	* lib/net/smtp.rb: SMTP#started? wrongly returned false always.
-
-Tue Sep 30 02:54:49 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/ruby/test_iterator.rb: new test
-	  test_break__nested_loop[123].
-
-Mon Sep 29 23:39:13 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb (finish): does not raise IOError even if
-	  !started?, to allow closing socket which was opened before
-	  session started.
-
-	* lib/net/pop.rb (finish): ditto.
-
-	* lib/net/smtp.rb (finish): ditto.
-
-Mon Sep 29 19:06:51 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* ext/win32ole/extconf.rb: add windows.h checking.
-	  (ruby-bugs:PR#1185)
-
-Mon Sep 29 16:18:30 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb: check if the given logdevice object respond_to :write
-	  and :close, not is_a? IO.  duck duck.
-
-	* test/logger/test_logger.rb: self IO.pipe reading/writing may be
-	  locked by the flood.  use tempfile.
-
-	* lib/wsdl/xmlSchema/data.rb: wrong constant reference.
-
-Mon Sep 29 16:11:23 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb: clean up temporary symlink.
-	  Patched by NaHi.  [ruby-dev:21420]
-
-Mon Sep 29 11:16:55 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_thread_atfork): wrong format specifier.
-	  [ruby-dev:21428]
-
-	* process.c (pst_inspect): better description.
-
-Mon Sep 29 02:31:44 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/utils.rb (Utils::su): use setgid and setuid to
-	  set real and effective IDs. and setup group access list by
-	  initgroups.
-
-Sun Sep 28 11:14:19 2003  Koji Arai  <jca02266@nifty.ne.jp>
-
-	* ext/digest/digest.c (Init_digest): `copy_object' was deprecated.
-	  `initialize_copy' should be defined.
-
-	* ext/stringio/stringio.c (Init_stringio): ditto.
-
-Sat Sep 27 18:25:13 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/xsd/charset.rb: XSD::Charset.is_ces did return always true under
-	  $KCODE = "NONE" environment.  check added.
-
-	* test/xsd/test_xsd.rb: add tests for above fix.
-
-Sat Sep 27 15:58:50 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/rpc/cgistub.rb: make logging severity threshold higher.
-
-	* lib/soap/rpc/standaloneServer.rb: defer WEBrick server start to give
-	  a chance to reset logging severity threshold.
-
-	* test/soap/calc/test_*, test/soap/helloworld/test_helloworld.rb: run
-	  silent.
-
-Sat Sep 27 09:44:18 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/test_fileutils.rb: clear all errors on Windows.
-	  [ruby-dev:21417]
-
-	* test/fileutils/test_nowrite.rb: ditto.
-
-Mon Sep 27 09:14:03 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* array.c (rb_ary_delete): comparison may change the capacity.
-	  [ruby-dev:24348]
-
-	* array.c (rb_ary_fill): fill should honor length argument.
-	  [ruby-dev:24346]
-
-	* array.c (rb_ary_replace): should not use ptr from shared array.
-	  [ruby-dev:24345]
-
-	* ext/socket/socket.c (s_accept): don't retry for EWOULDBLOCK.
-	  [ruby-talk:113807]
-
-Sat Sep 27 04:57:07 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_file.rb: new file.  only asserts unlink-before-close
-	  behaviour now.
-
-	* test/soap/marshal/test_digraph.rb: should close before unlink.
-	  unlink-before-close pattern is not needed here.
-
-Sat Sep 27 03:32:37 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/soap/*, test/wsdl/*, test/xsd/*: move TestCase classes into
-	  each module namespace.  TestMarshal in
-	  test/soap/marshal/test_marshal.rb crashed with
-	  test/ruby/test_marshal.rb.
-
-Sat Sep 27 01:30:59 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/socket/socket.c (ruby_connect): on win32, type of the 4th
-	  argument of getsockopt is char *.
-
-Fri Sep 26 18:35:40 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/resolv-replace.rb: 1.8 compliance.  [ruby-talk:82946]
-
-Fri Sep 26 17:39:27 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_marshal.rb: add test for ruby's objects.
-
-Fri Sep 26 09:52:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* defines.h (flush_register_windows): use volatile only for gcc on
-	  Solaris.  [ruby-dev:21403]
-
-	* lib/mkmf.rb (xsystem): use system directly to honor shell meta
-	  charaters.
-
-Fri Sep 26 00:10:13 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/README: updated.
-
-Thu Sep 25 17:48:10 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/openssl/ossl.c (ossl_buf2str): fix type of 1st argument for
-	  rb_protect.
-
-	* ext/openssl/ossl_hmac.c (ossl_hmac_digest): should return meaningful
-	  value.
-
-Thu Sep 25 09:00:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/ostruct.rb: Added OpenStruct#==.
-
-	* test/ostruct/test_ostruct.rb: Added.
-
-Thu Sep 25 07:55:26 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/win32ole/win32ole.c, ext/openssl/ossl_pkey_dsa.c,
-	  ext/openssl/ossl_pkey_rsa.c, ext/bigdecimal/bigdecimal.h: must
-	  not use C++ or C99 style comment yet.  (ruby-bugs:PR#1184)
-
-Thu Sep 25 00:23:22 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* MANIFEST: add SOAP4R.
-
-Thu Sep 25 00:13:15 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/soap/* (29 files): SOAP4R added.
-
-	* lib/wsdl/* (42 files): WSDL4R added.
-
-	* lib/xsd/* (12 files): XSD4R added.
-
-	* test/soap/* (16 files): added.
-
-	* test/wsdl/* (2 files): added.
-
-	* test/xsd/* (3 files): added.
-
-	* sample/soap/* (27 files): added.
-
-	* sample/wsdl/* (13 files): added.
-
-Wed Sep 24 02:08:11 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpservlet/cgihandler.rb: conform to mswin32.
-	  [ruby-talk:82735], [ruby-talk:82748], [ruby-talk:82818]
-
-Tue Sep 23 23:10:16 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb: add Logger#<<(msg) for writing msg without any
-	  formatting.
-
-	* test/logger/test_logger.rb: ditto.
-
-Tue Sep 23 20:47:51 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* error.c (rb_warn_m): should not warn if -W0 is specified.
-	  [ruby-talk:82675]
-
-Mon Sep 22 21:28:57 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* MANIFEST: updated.
-
-Mon Sep 22 19:22:26 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* configure.in (AC_CHECK_FUNCS): add setuid and setgid.
-
-Mon Sep 22 12:34:55 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* util.c (ruby_strtod): skip preceding zeros before counting
-	  digits in the mantissa. (ruby-bugs:PR#1181)
-
-Sun Sep 21 04:12:36 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_initialize): the argument
-	  should be a String.
-
-	* ext/openssl/ossl_ocsp.c (ossl_ocspres_initialize): ditt.
-
-	* ext/openssl/ossl_x509attr.c (ossl_x509attr_initialize): ditto.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509ext_initialize): ditto.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509ext_set_value): ditto.
-
-Sat Sep 20 11:49:05 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb: typo fixed.
-
-	* test/logger/test_logger.rb: new file.
-
-Fri Sep 19 11:39:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* test/testunit/*: Added.
-
-	* lib/test/unit.rb: Documentation update.
-
-	* lib/test/unit/ui/console/testrunner.rb (TestRunner#initialize):
-	  Ditto.
-
-	* lib/test/unit.rb: Factored out an ObjectSpace collector.
-
-	* lib/test/unit/collector/objectspace.rb: Ditto.
-
-	* sample/testunit/*: Added.
-
-Fri Sep 19 01:00:48 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/log.rb (BasicLog#log): get rid of as ineffectual
-	  condition.
-
-	* lib/webrick/log.rb (BasicLog#format): add "\n" to message.
-
-Thu Sep 18 22:43:20 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_invoke): should push PROT_PCALL tag for orphans.
-
-	* eval.c (proc_invoke): should update "result" for orphans.
-
-Thu Sep 18 20:33:03 2003  Tietew  <tietew-ml-ruby-list@tietew.net>
-
-	* parse.y (str_xquote): do not prepend escapes in
-	  backqoute literals.  [ruby-list:38409]
-
-Thu Sep 18 20:30:17 2003  Tanaka Akira  <akr@m17n.org>
-
-	* lib/pathname.rb: update document.
-
-Thu Sep 18 15:27:05 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/logger.rb: new file.  Logger, formerly called devel-logger or
-	  Devel::Logger.
-
-	* sample/logger/*: new file.  samples of logger.rb.
-
-Wed Sep 17 23:41:45 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (localjump_destination): should not raise ThreadError
-	  exception for "break". [ruby-dev:21348]
-
-	* eval.c (proc_invoke): use result instead of prot_tag->retval.
-	  retval is no longer propagated to the ancestors.
-
-Wed Sep 17 20:34:00 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (tokadd_string, parse_string, yylex): escaped terminator
-	  is now interpreted as is.  [ruby-talk:82206]
-
-Wed Sep 17 18:52:36 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/fileutils/fileassertions.rb: new file.
-
-	* test/fileutils/test_fileutils.rb: new file.
-
-	* test/fileutils/test_nowrite.rb: new file.
-
-Wed Sep 17 18:51:02 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/strscan/test_stringscanner.rb: require test/unit.
-
-Wed Sep 17 18:35:34 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* test/strscan/test_stringscanner.rb: new file.
-
-Wed Sep 17 18:03:30 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl: all files are reviewed to simplify and avoid memory leak.
-
-	* ext/openssl/extconf.rb: add check for assert.h.
-
-	* ext/openssl/ossl.c (ossl_buf2str): new function to convert
-	  C buffer to String and free buffer.
-
-	* ext/openssl/ossl.c (ossl_x509_ary2sk): new function to convert
-	  Array of OpenSSL::X509 to STACK_OF(X509) with exception safe.
-
-	* ext/openssl/ossl.c (ossl_to_der, ossl_to_der_if_possible): new
-	  functions to convert object to DER string.
-
-	* ext/openssl/ossl.h: ditto.
-
-	* ext/openssl/ossl_bio.c (ossl_membio2str): new function to convert
-	  BIO to String object and free BIO.
-
-	* ext/openssl/ossl_bio.h: ditto.
-
-	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_to_der): add for "to_der".
-
-	* ext/openssl/ossl_x509name.c (ossl_x509name_to_der): ditto.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509ext_to_der): ditto.
-
-	* ext/openssl/ossl_x509ext.c (create_ext_from_array): removed
-	  and reimplement in openssl/x509.rb.
-
-	* ext/openssl/ossl_x509attr.c: reimplemented and disable some
-	  method temporarily. this class doesn't work fine without ASN.1
-	  data support;-) I'll rewrite in near future.
-
-	* ext/openssl/lib/openssl/x509.c (X509::Attribute): get rid off
-	  unused code.
-
-	* ext/openssl/lib/openssl/x509.c (X509::ExtensionFactory): refine all.
-
-Tue Sep 16 22:25:06 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/csv/test_csv.rb: add negative tests of row_sep.
-
-Tue Sep 16 18:02:36 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* regex.c (re_compile_pattern): should not translate character
-	  class range edge. [ruby-list:38393]
-
-Tue Sep 16 16:47:56 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* MANIFEST: add test/csv/mac.csv.
-
-	* win32/Makefile.sub, bcc32/Makefile.sub (test): add phony NUL target.
-
-Mon Sep 15 19:02:52 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* lib/csv.rb: add extra pamameter to specify row(record) separater
-	  character.  To parse Mac's CR separated CSV, do like this.
-	    CSV.open("mac.csv", "r", ?,, ?\r) { |row| p row.to_a }
-	  The 3rd parameter in this example ?, is for column separater and the
-	  4th ?\r is for row separater.  Row separater is nil by default.  Nil
-	  separater means "\r\n" or "\n".
-
-	* test/csv/test_csv.rb: add tests for above feature.
-
-	* test/csv/mac.csv: added.  Sample CR separated CSV file.
-
-Fri Sep 12 22:41:48 2003  Michal Rokos  <m.rokos@sh.cvut.cz>
-
-	* ext/openssl/ossl.c: move ASN.1 stuff to ossl_asn1.[ch]
-
-	* ext/openssl/ossl.c: move BIO stuff to ossl_bio.[ch]
-
-	* ext/openssl/ossl_asn1.[ch]: new files
-
-	* ext/openssl/ossl_bio.[ch]: new files
-
-Fri Sep 12 12:30:41 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* intern.h (rb_disable_super, rb_enable_super): replace with dummy
-	  expressions instead of prototypes.  the functions remain yet for
-	  binary compatibility.  [ruby-talk:81758]
-
-Fri Sep 12 12:09:54 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* bignum.c (rb_big_and): convert argument using 'to_int'.
-
-	* bignum.c (rb_big_or): ditto.
-
-	* bignum.c (rb_big_xor): ditto.
-
-Fri Sep 12 07:06:14 2003  David Black  <dblack@superlink.net>
-
-	* lib/scanf.rb: Took out useless @matched_item variable; some small
-	  refactoring.
-
-Thu Sep 11 08:43:44 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_f_require): allow "require" on $SAFE>0, if feature
-	  name is not tainted.
-
-	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser::stream):
-	  Supports StringIO.
-
-Wed Sep 10 22:47:30 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.h: add a workaround for win32 platform.
-	  libeay32.dll doesn't export functions defined in conf_api.h.
-
-	* ext/openssl/ossl_config.c (ossl_config_initialize): ditto.
-
-	* ext/openssl/ossl_config.c (ossl_config_add_value): ditto.
-
-	* ext/openssl/ossl_config.c (set_conf_section_i): should check
-	  if the argument is Array.
-
-Wed Sep 10 22:41:54 2003  Tietew  <tietew@tietew.net>
-
-	* eval.c (win32_get_exception_list): avoid VC7 warning.
-	  [ruby-win32:577]
-
-Tue Sep  9 10:39:51 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (struct tag): dst should be VALUE.
-
-Tue Sep  9 10:39:51 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (localjump_destination): stop at the scope where the current
-	  block was created.  [ruby-dev:21353]
-
-Tue Sep  9 05:17:04 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_config.rb: avoid compile error in OpenSSL-0.9.6.
-
-Tue Sep  9 02:41:35 2003  Michal Rokos  <m.rokos@sh.cvut.cz>
-
-	* ext/openssl/ossl_config.c: Refine compatibility.
-
-Tue Sep  9 01:50:45 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httpserver.rb (HTTPServer#access_log): add "\n" to
-	  the message.
-
-	* lib/webrick/log.rb (BasicLog#log): add "\n" only if needed.
-
-Mon Sep  8 22:15:33 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tk/lib/multi-tk.rb: modify security check at creating
-	  a new interpreter
-
-Mon Sep  8 20:00:12 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb, lib/optparse/version.rb: search also all
-	  capital versions.
-
-Mon Sep  8 19:26:33 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl.h: include openssl/conf.h and openssl/conf_api.h.
-
-	* ext/openssl/ossl_config.c: refine all with backward compatibility.
-
-	* ext/openssl/ossl_config.h: export GetConfigPtr() and DupConfigPtr().
-
-	* ext/openssl/ossl_x509.c: added new constants under X509 module.
-	  DEFAULT_CERT_AREA, DEFAULT_CERT_DIR, DEFAULT_CERT_FILE,
-	  DEFAULT_CERT_DIR_ENV, DEFAULT_CERT_FILE_ENV and DEFAULT_PRIVATE_DIR.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_free): don't free
-	  the members of the struct. it's left to GC.
-
-	* ext/openssl/ossl_x509ext.c (ossl_x509_set_config): add for config=.
-
-	* ext/openssl/ossl_x509ext.c (Xossl_x509extfactory_initialize):
-	  add attr readers: issuer_certificate, subject_certificate,
-	  subject_request, crl and config.
-
-Mon Sep  8 18:26:41 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/accesslog.rb (AccessLog::setup_params): use req.port
-	  instead of config[:Port] or req.request_uri.port.
-
-	* lib/webrick/httprequest.rb (HTTPRequest#meta_vars): ditto.
-
-	* lib/webrick/httpservlet/filehandler.rb (FileHandler#dir_list): ditto.
-
-	* lib/webrick/config.rb: :Listen option never be used.
-
-	* lib/webrick/server.rb (GenericServer#initialize): don't use :Listen
-	  option and add warning message.
-
-	* lib/webrick/log.rb (BasicLog#<<): shortcut of log(INFO, ...).
-
-	* lib/webrick/httpserver.rb (HTTPServer#accesslog): use << for logging.
-
-Sun Sep  7 16:08:28 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (lib_mainloop_core): fixed signal-trap bug
-
-	* ext/tk/lib/*.rb : Ruby/Tk works at $SAFE == 4
-
-Sat Sep  6 02:26:34 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_*.rb: assert_same, assert_match, and so on.
-
-Sat Sep  6 18:45:46 2003  Mauricio Fernandez  <batsman.geo@yahoo.com>
-
-	* parse.y (assignable): call rb_compile_error(), not rb_bug().
-	  [ruby-core:01523]
-
-Sat Sep  6 17:40:41 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ruby_missing.c: rid of unnecessary backward
-	  compatibility stuff. and remove DEFINE_ALLOC_WRAPPER from
-	  all sources.
-
-	* ext/openssl/ossl_x509ext.c (X509::Extension.new): new method.
-
-	* ext/openssl/ossl_x509ext.c (X509::Extension#oid=): new method.
-
-	* ext/openssl/ossl_x509ext.c (X509::Extension#value=): new method.
-
-	* ext/openssl/ossl_x509ext.c (X509::Extension#critical=): new method.
-
-Sat Sep  6 01:23:22 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (CreateChild): need to quote cmd if RUBYSHELL is set.
-
-	* win32/win32.c (CreateChild): fix condition about whether to call
-	  shell or not.
-
-Sat Sep  6 00:36:20 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
-
-	* Makefile.in (test): phony target.
-
-	* lib/mkmf.rb (have_library, find_library): configure by library
-	  name.
-
-	* lib/optparse.rb (OptionParser#order, #permute, #parse): allow an
-	  array as argument.
-
-	* test/ruby/test_*.rb: moved invariants to left side in
-	  assert_equal, and use assert_nil, assert_raises and so on.
-
-	* win32/win32.c (isInternalCmd): distinguish command.com and
-	  cmd.exe.
-
-	* win32/win32.c (make_cmdvector): a character just after wildcard
-	  was ignored.  [ruby-core:01518]
-
-Fri Sep  5 20:27:08 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_*.rb: replace 'assert(a == b)' with assert_equal(a, b)'
-
-Fri Sep  5 18:00:51 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/x509.rb: new method X509::Name::parse.
-
-	* ext/openssl/ossl_digest.c: add ossl_digest_new().
-
-	* ext/openssl/ossl_digest.h: ditto.
-
-	* ext/openssl/ossl_cipher.c: add ossl_cipher_new().
-
-	* ext/openssl/ossl_cipher.h: ditto.
-
-Fri Sep  5 15:32:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-font-lock-maybe-here-docs): should not
-	  search delimiter forward if found in backward.
-
-Fri Sep  5 13:32:48 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/runner.rb: arguments should be keys.
-
-Fri Sep  5 12:09:55 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* test/ruby/test_system.rb (test_system): check existence of ruby
-	  interpreter.
-
-Fri Sep  5 11:32:17 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/optparse.rb (--version): fix assignment/reference order.
-
-	* lib/optparse.rb (OptionParser#help): new; OptionParser#to_s may
-	  be deprecated in future.
-
-	* lib/optparse/version.rb (OptionParser#show_version): hide Object.
-
-	* test/runner.rb: fix optparse usage.
-
-	* test/runner.rb: glob all testsuits if no tests given.
-
-Fri Sep  5 10:42:58 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/runner.rb: added.  gets testcases from command line and runs it.
-
-	* test/ruby/test_gc.rb: remove useless part which was for dumping test
-	  result.
-
-Fri Sep  5 09:28:59 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby/test_gc.rb: added.  splitter.rb which I made to split
-	  sample/test.rb into test/ruby/test_* kindly removed GC test (the
-	  last section in the original test) to reduce things to be worried.
-
-Fri Sep  5 03:00:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* test/ruby/test_iterator.rb (test_block_in_arg): add no block
-	  given tests.
-
-	* test/ruby/test_iterator.rb (test_ljump): uncomment LocalJumpError
-	  test.
-
-Fri Sep  5 01:10:11 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/ruby: tests for ruby itself.
-
-	* test/ruby/test_*.rb: split sample/test.rb into 28 test/unit testcases.
-	  some tests could not be translates...  search '!!' mark to see it.
-
-	* test/csv/test_csv.rb: should require 'csv', not '../lib/csv'.  test
-	  runner should set load path correctly.
-
-Fri Sep  5 01:03:59 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/csv/test_csv.rb: close opened files for CSV::IOBuf explicitly.
-	  opened file cannot be removed under win32 box.
-
-Thu Sep  4 23:59:40 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* parse.y (tokadd_string): newlines have no special meanings in
-	  %w/%W, otherwise they are ignored only when interpolation is
-	  enabled.  [ruby-dev:21325]
-
-Thu Sep  4 19:38:25 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* ext/io/wait/.cvsignore: added.
-
-	* ext/openssl/.cvsignore: added.
-
-Thu Sep  4 19:28:24 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* sample/openssl: added.  Sample of standard distribution library
-	  should be locate in sample/{module_name}/*.
-
-	* ext/openssl/sample/*: removed.  move to sample/openssl/*.
-
-Thu Sep  4 18:02:15 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/csv/test_csv.rb: use remove_const to reduce warnings.  use
-	  Dir.tmpdir to locate working files.
-
-Thu Sep  4 17:41:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* misc/ruby-mode.el (ruby-here-doc-beg-re): underscore also is
-	  valid delimiter.
-
-	* misc/ruby-mode.el (ruby-here-doc-end-match): must quote
-	  arbitrary string to use as regexp.
-
-	* misc/ruby-mode.el (ruby-font-lock-maybe-here-docs): must not
-	  call `ruby-here-doc-end-match' unless `ruby-here-doc-beg-re'
-	  matched.
-
-Thu Sep  4 15:40:07 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test/csv/test_csv.rb: run on test/unit original layer.
-
-Thu Sep  4 12:54:50 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/token.c: headerless documents with root-level spacing now
-	  honored.
-
-Thu Sep  4 00:06:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (mark_frame_adj): need to adjust argv pointer if using
-	  system's alloca. [ruby-core:01503]
-
-Wed Sep  3 21:33:20 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-
-	* test: add test directory.  Test::Unit aware testcases and needed
-	  files should be located in this directory.  dir/file name convention;
-	    test/{module_name}/test_{testcase_name}.rb
-	    test/{module_name}/{needed_files}
-	  someday, someone will write testrunner which searches test_*.rb and
-	  run testcases automatically.
-
-	* test/csv/*: add testcase for lib/csv.rb.
-
-Wed Sep  3 01:37:09 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* io.c (rb_f_gets): should call next_argv() before type check
-	  current_file. [ruby-list:38336]
-
-Tue Sep  2 20:37:15 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/net/protocols.rb (SSLIO#ssl_connect): warning
-	  for skipping server verification.
-
-Tue Sep  2 23:36:57 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (proc_invoke): should retrieve retval when pcall is true.
-
-Tue Sep  2 14:09:20 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* ext/socket/extconf.rb: check s6_addr8 in in6_addr (Tru64 UNIX).
-	  the patch is submitted by nmu <nmu@users.sourceforge.jp>.
-
-	* ext/socket/getaddrinfo.c (getaddrinfo): should use in6_addr8 on
-	  some platforms.
-
-	* ext/socket/getnameinfo.c (getnameinfo): ditto.
-
-Tue Sep  2 14:02:19 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* ext/tcltklib/tcltklib.c (ip_invoke): fixed bug on passing a exception
-
-	* ext/tk/lib/{tk.rb, tkcanvas.rb, tkfont.rb, tktext.rb} :
-	  bug fix and improvement of font control
-
-Tue Sep  2 09:51:36 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval): should not handle exceptions within rescue
-	  argument.  [ruby-talk:80804]
-
-Tue Sep  2 00:44:37 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* re.c (rb_memsearch): fix overrun.  [ruby-talk:80759]
-
-Tue Sep  2 00:41:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/iconv/iconv.c (map_charset): use lower case keys.
-
-	* ext/iconv/iconv.c (iconv_fail): just yield error and return the
-	  result if a block is given.
-
-	* ext/iconv/iconv.c (iconv_convert): yield error and append the
-	  result if a block is given.
-
-	* ext/iconv/charset_alias.rb (charset_alias): optional third
-	  argument.
-
-	* ext/iconv/charset_alias.rb (charset_alias): use CP932 instead of
-	  SHIFT_JIS on cygwin.
-
-Mon Sep  1 18:34:25 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval): make tail recursion in ELSE clause of
-	  RESCUE a jump.
-
-Mon Sep  1 18:00:02 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (aref_args): forgot to call NEW_SPLAT(). reported by
-	  Dave Butcher.
-
-	* eval.c (Init_Thread): protect thgroup_default.  suggested by Guy
-	  Decoux in [ruby-talk:80623]
-
-Mon Sep  1 16:59:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_thread_switch): add RESTORE_EXIT; exit by another
-	  thread termination.
-
-	* eval.c (rb_thread_start_0): should not error_print() within
-	  terminated thread, because $stderr used by it might be
-	  overriden now.  [ruby-dev:21280]
-
-Sun Aug 31 22:46:55 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* eval.c (TAG_DST()): take no argument.
-
-	* process.c (p_gid_sw_ensure): return VALUE.
-
-Sun Aug 31 22:27:10 2003  Hidetoshi NAGAI  <nagai@dumbo.ai.kyutech.ac.jp>
-
-	* process.c (p_gid_sw_ensure): lack of function type
-
-Sun Aug 31 12:25:06 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
-
-	* lib/optparse.rb: --version takes an optional argument; "all" or
-	  a list of package names.
-
-Sun Aug 31 10:17:02 2003  Tadayoshi Funaba  <tadf@dotrb.org>
-
-	* lib/date/format.rb: yyyy/mm is not an acceptable format.
-
-	* lib/time.rb: follow above.
-
-Sat Aug 30 14:25:43 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_iter_break): should not call TAG_JUMP directly.
-
-Sat Aug 30 03:58:21 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (struct BLOCK): remove BLOCKTAG, use scope instead.
-
-	* eval.c (POP_TAG): no longer propagate retval.  retval is now set
-	  directly by localjump_destination().
-
-	* eval.c (localjump_destination): new function to cast
-	  return/break local jump.
-
-	* eval.c (rb_yield_0): stop TAG_RETURN/TAG_BREAK escaping.
-
-Fri Aug 29 22:35:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* bigdecimal.c *.html: The 2nd arg. for add,sub,mult, and div is 0,
-	  then result will be the same as +,-,*,/ respectively.
-
-Fri Aug 29 17:30:15 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
-
-	* process.c: bug fix
-
-	* process.c: add rb_secure(2) to methods of Process::{UID,GID,Sys}
-
-	* process.c: deny handling IDs during evaluating the block given to
-	  the Process::{UID,GID}.switch method
-
-	* ext/tcltklib/tcltklib.c : some methods have no effect if on slave-IP
-
-	* ext/tcltklib/tcltklib.c : can create a interpreter without Tk
-
-	* ext/tcltklib/tcltklib.c : bug fix on handling exceptions
-
-	* ext/tcltklib/MANUAL.euc : modify
-
-	* ext/tk/lib/tk.rb : freeze some core modules
-
-	* ext/tk/lib/multi-tk.rb : more secure
-
-	* ext/tk/lib/tk.rb: TkVariable.new(array) --> treat the array as the
-	  Tk's list
-
-	* ext/tk/lib/tk.rb: improve accessibility of TkVariable object
-
-	* ext/tk/lib/tk.rb, ext/tk/lib/tkfont.rb, ext/tk/lib/tkcanvas.rb,
-	  ext/tk/lib/tktext.rb : fix bug of font handling
-
-	* ext/tk/lib/tkfont.rb TkFont.new() accepts compound fonts
-
-Thu Aug 28 22:07:12 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* variable.c (rb_autoload_load): call const_missing if autoloading
-	  constant is not defined to allow hook.
-
-	* eval.c (rb_eval): use rb_const_get_from() instead of
-	  rb_const_get_at().
-
-	* eval.c (is_defined): forgot to check NODE_COLON3.
-
-Thu Aug 28 17:30:24 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* variable.c (rb_const_get_0): should check constants defined in
-	  included modules, if klass is Object. [ruby-talk:79302]
-
-	* numeric.c (check_uint): check should be done using UINT_MAX, not
-	  INT_MAX. this fix is submitted by Lyle Johnson
-	  <lyle@knology.net> in [ruby-core:01486]
-
-Thu Aug 28 05:02:52 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (singleton): typo fixed (ruby-bugs-ja:PR#562)
-
-Thu Aug 28 02:37:45 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_eval): *a = [1,2] now assigns [[1,2]] to a.
-	  consistent with *a = [1], which set [[1]] to a.
-
-	* node.h: merge NODE_RESTARY to NODE_SPLAT.
-
-	* parse.y: rules simplified a bit by removing NODE_RESTARY.
-
-	* sample/test.rb: updated for new assignment behavior.
-
-Wed Aug 27 22:33:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* error.c (rb_bug): should not use other methods; this function is
-	  not for ordinary use.  [ruby-dev:21259]
-
-Wed Aug 27 15:07:57 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/smtp.rb (check_response): AUTH CRAM-MD5 returns 334
-	  response. [ruby-list:38279]
-
-Wed Aug 27 05:10:15 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* win32/win32.c (map_errno): support winsock error.
-
-	* win32/win32.c (pipe_exec, CreateChild, poll_child_status, waitpid,
-	  kill, link, rb_w32_rename, unixtime_to_filetime, rb_w32_utime):
-	  pass errno to map_errno().
-
-	* win32/win32.c (rb_w32_select, rb_w32_accept, rb_w32_bind,
-	  rb_w32_connect, rb_w32_getpeername, rb_w32_getsockname,
-	  rb_w32_getsockopt, rb_w32_ioctlsocket, rb_w32_listen, rb_w32_recv,
-	  rb_w32_recvfrom, rb_w32_send, rb_w32_sendto, rb_w32_setsockopt,
-	  rb_w32_shutdown, rb_w32_socket, rb_w32_gethostbyaddr,
-	  rb_w32_gethostbyname, rb_w32_gethostname, rb_w32_getprotobyname,
-	  rb_w32_getprotobynumber, rb_w32_getservbyname, rb_w32_getservbyport,
-	  rb_w32_fclose, rb_w32_close): use map_errno().
-
-	* win32/win32.h: add winsock errors.
-
-Tue Aug 26 23:53:23 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* lib/ostruct.rb (OpenStruct::method_missing): prohibit modifying
-	  frozen OpenStruct. [ruby-talk:80214]
-
-Tue Aug 26 20:03:50 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* lib/mkmf.rb (create_tmpsrc): add the hook for source.
-	  [ruby-list:38122]
-
-Tue Aug 26 15:59:53 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* implicit.c (syck_type_id_to_taguri): corrected detection of
-	  x-private types.
-
-Sun Aug 24 01:02:48 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* file.c (file_expand_path): performance improvement.
-	  [ruby-talk:79748]
-
-Sat Aug 23 23:41:16 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* file.c (rb_file_s_expand_path): avoid calling rb_scan_args() for
-	  apparent cases. [ruby-talk:79748]
-
-Sat Aug 23 18:56:53 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/nkf/nkf.c (rb_nkf_putchar): should use rb_str_resize() to just
-	  resize a string, rb_str_cat() disallows NULL.  [ruby-dev:21237]
-
-Sat Aug 23 16:48:41 2003  Keiju Ishitsuka  <keiju@ishitsuka.com>
-
-	* lib/irb/ruby-lex.rb: bug fix for "foo" !~ /bar/. [ruby-talk:79942]
-
-Sat Aug 23 15:59:58 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_eval, rb_iterate, block_pass): reduce PUSH/POP_TAG and
-	  EXEC_TAG() for retry.  [ruby-dev:21216]
-
-Sat Aug 23 02:32:33 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_yield_splat): should check if "values" is array.
-
-	* enum.c (each_with_index_i): typo.
-
-Fri Aug 22 17:07:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* enum.c (inject_i): use rb_yield_values.
-
-	* enum.c (each_with_index_i): ditto.
-
-	* eval.c (rb_yield_splat): new function to call "yield *values".
-
-	* string.c (rb_str_scan): use rb_yield_splat().
-
-Fri Aug 22 06:13:22 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/rubyext.c: refactoring of the transfer method
-	  dispatch.  added yaml_org_handler for faster dispatch of
-	  transfers to base types.
-
-	* lib/yaml/rubytypes.rb: removed handling of builtins from
-	  Ruby library.
-
-	* ext/syck/token.c: quoted and block scalars are now implicit !str
-
-	* ext/syck/implicit.c: empty string detected as !null.
-
-Fri Aug 22 01:00:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (block_pass): improve passing current block.
-
-Fri Aug 22 00:13:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c: Int. overflow bug in multiplication
-	  fixed, and VpNmlz() speed up.
-
-Wed Aug 20 16:44:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/socket/socket.c (ruby_connect): many systems seem to have
-	  a problem in select() after EINPROGRESS.  [ruby-list:38080]
-
-Wed Aug 20 01:31:17 2003  why the lucky stiff  <why@ruby-lang.org>
-
-	* ext/syck/syck.h: Parser definition problems on HP-UX.
-	  [ruby-talk:79389]
-
-	* ext/syck/handler.c (syck_hdlr_get_anchor): Memory leak.
-
-	* ext/syck/syck.s (syck_io_file_read): Bad arguments to fread.
-
-	* ext/syck/rubyext.c: Tainting issues.
-
-Tue Aug 19 23:20:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c .h .html: to_s("+") implemented.
-
-	* ext/bigdecimal/lib/bigdecimal/math.rb: E implemented.
-
-Tue Aug 19 07:47:09 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/ssl.rb: new file; SSL/TLS enhancement for GenericServer.
-
-	* lib/webrick/https.rb: SSLSocket handling is moved to webrick/ssl.rb.
-
-	* lib/webrick/compat.rb (File::fnmatch): remove old migration code.
-
-	* lib/webrick/httpserver.rb (HTTPServer#run): ditto.
-
-	* lib/webrick/server.rb (GenericServer#listen): the body of this
-	  method is pull out as Utils::create_lisnteners.
-
-	* lib/webrick/utils.rb (Utils::create_lisnteners): new method.
-
-	* lib/webrick/server.rb (GenericServer#start): should rescue
-	  unknown errors. and refine comments.
-
-	* ext/openssl/lib/openssl/ssl.rb (SSLServer#accept): should close
-	  socket if SSLSocket raises error.
-
-Tue Aug 19 11:19:33 2003  Shugo Maeda  <shugo@ruby-lang.org>
-
-	* io.c (next_argv): should not call GetOpenFile() if rb_stdout is
-	  not a IO (T_FILE).
-
-Tue Aug 19 07:47:09 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/ossl_ssl.c: sync_close is moved to SSLSocket as
-	  a builtin.
-
-	* ext/openssl/lib/openssl/buffering.rb (Buffering#close): ditto.
-
-	* ext/openssl/lib/openssl/buffering.rb (Buffering#puts): should
-	  add a return to the tails of each line.
-
-	* ext/openssl/lib/openssl/ssl.rb: new class OpenSSL::SSL::SSLServer.
-
-	* ext/openssl/lib/net/protocols.rb (SSLIO#ssl_connect): use sync_close.
-
-	* ext/openssl/sample/echo_svr.rb: use SSLServer.
-
-	* ext/openssl/sample/echo_cli.rb: add example of SSLSocket#sync_close.
-
-Tue Aug 19 01:24:34 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/curses/curses.c (_XOPEN_SOURCE_EXTENDED): Mac OS X standard
-	  headers are inconsistent at this macro.  [ruby-core:01432]
-
-	* ext/curses/extconf.rb: check if _XOPEN_SOURCE_EXTENDED breaks.
-
-	* ext/tcltklib/stubs.c: Status macro in X11/Xthreads.h bothers
-	  winspool.h
-
-	* instruby.rb: make list at first instead of iterator.
-	  [ruby-talk:79347]
-
-Mon Aug 18 11:23:11 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* dir.c (glob_helper): preserve raw order for **.
-
-Sun Aug 17 23:39:55 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/openssl/extconf.rb (HAVE_VA_ARGS_MACRO): need to compile.
-
-Sun Aug 17 17:10:03 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/ssl.rb (SSLSocket#sync_close=): add a
-	  method to specify if the underlying IO will be closed in
-	  SSLSocket#close.
-
-	* ext/openssl/lib/openssl/buffering.rb: add forwarders to
-	  setsockopt, getsockopt and fcntl.
-
-	* ext/openssl/lib/net/protocols.rb: enable sync for SSLSocket.
-
-Sun Aug 17 11:32:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): should not force to remake Makefile when
-	  installation and so on.
-
-Sat Aug 16 23:58:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* marshal.c (w_symbol, w_object): get rid of warnings.
-
-	* re.c (rb_memsearch): ditto.
-
-	* time.c (time_dump): ditto.
-
-	* ext/extmk.rb (extmake): not continue making when extconf.rb
-	  failed.
-
-	* ext/openssl/extconf.rb: check __VA_ARGS__ macro more precisely.
-
-	* ext/openssl/ossl.h: remove version.h dependency.
-
-	* ext/openssl/ruby_missing.h: ditto.
-
-	* lib/mkmf.rb (pkg_config): use --libs output except with
-	  only-L for other options.  [ruby-list:38099]
-
-	* lib/mkmf.rb (create_makefile): separate rule for static
-	  library from shared object.
-
-	* win32/Makefile.sub, bcc32/Makefile.sub, wince/Makefile.sub:
-	  define exec_prefix and libdir.
-
-Fri Aug 15 23:15:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c .h: Bug in combination of limit & div
-	  method fixed.
-
-	* ext/bigdecimal/lib/bigdecimal/math.rb: atan() & sqrt() added.
-
-Fri Aug 15 12:01:43 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* configure.in (HUGE_ST_INO): check whether struct stat.st_ino
-	  is larger than long.  [ruby-dev:21194]
-	  http://www.geocities.co.jp/SiliconValley-PaloAlto/1409/ruby/beos.html
-
-	* error.c (syserr_eqq): errno might exceed Fixnum limit.
-
-	* error.c (Init_Exception): moved base initialization from
-	  init_syserr().
-
-	* inits.c (rb_call_inits): postpone initializing errnos until
-	  Bignum is available.
-
-Fri Aug 15 12:01:43 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/curses/curses.c (_XOPEN_SOURCE_EXTENDED): needed to let
-	  keyname() and so on be declared.
-
-	* ext/curses/curses.c (curses_resizeterm, window_resize):
-	  arguments conflicted with macros in term.h.
-
-	* ext/curses/curses.c (Curses module methods): ensure
-	  initialized.  [ruby-dev:21191]
-
-Fri Aug 15 02:08:53 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* gc.c (id2ref): recycle check should be done by klass == 0.
-	  [ruby-core:01408]
-
-Fri Aug 15 01:34:23 2003  Michal Rokos  <m.rokos@sh.cvut.cz>
-
-	* ext/openssl/ossl_pkey.c: move generate_cb here
-
-	* ext/openssl/ossl_pkey_{dh|dsa|rsa}.c: adapt to this cb
-
-	* ext/openssl/openssl_missing.[ch]: add (0.9.6x, x<j) missing BN funcs
-
-	* ext/openssl/ossl_bn.c: use supplied funcs from openssl_missing.c
-
-Fri Aug 15 00:38:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c: Bug in div method fixed.
-
-	* ext/bigdecimal/lib/bigdecimal/math.rb: Newly added.
-
-	* ext/bigdecimal/sample/pi.rb: Changed so as to use math.rb.
-
-Thu Aug 14 21:19:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (Init_Thread): Continuation#[] added.  [ruby-talk:79028]
-
-Thu Aug 14 20:03:34 2003  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
-
-	* ext/win32ole/win32ole.c (OLE_FREE): should not call
-	  ole_message_loop.
-
-	* ext/win32ole/win32ole.c (ole_event_free): ditto.
-
-	* ext/win32ole/win32ole.c (ole_initialize): stop calling
-	  OleUninitialize at exit.
-
-Thu Aug 14 11:27:37 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* gc.c (rb_data_object_alloc): check type of 1st argument.
-	  [ruby-dev:21192]
-
-Thu Aug 14 00:21:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* parse.y (mlhs_node): should allow "::Foo" (colon3) as lhs.
-
-	* parse.y (lhs): ditto.
-
-	* parse.y (yylex): should return tCOLON3 right after kCLASS.
-	  [ruby-talk:78918]
-
-	* error.c (exc_initialize): was converting argument to string too
-	  eagerly.  Only check was needed. [ruby-talk:78958]
-
-Wed Aug 13 23:31:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal/bigdecimal.c .h .html: Ambiguity of
-	  BigDecimal::limit removed.
-
-Wed Aug 13 19:21:34 2003  Christian Neukirchen  <chneukirchen@yahoo.de>
-
-	* lib/webrick/https.rb (HTTPServer#run): should set syncing-mode
-	  to SSLSocket. [ruby-talk:78919]
-
-Wed Aug 13 18:13:49 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (POP_BLOCK): turn on BLOCK_LEFT flag when leaving block.
-
-	* eval.c (proc_invoke): unpack return/break destination when block
-	  is already left.
-
-Wed Aug 13 15:58:31 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* object.c (rb_class_s_alloc): add function prototype to avoid VC++
-	  warning.
-
-Wed Aug 13 13:50:59 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/Win32API/Win32API.c (Win32API_initialize): should pass some
-	  class to first argument of Data_Wrap_Struct(). (ruby-bugs:PR#1109)
-
-Tue Aug 12 16:55:11 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* Makefile.in: static link libraries to LIBRUBY_SO with static linked
-	  ext.  [ruby-dev:21157]
-
-	* ext/extmk.rb (extmake): sort extension library initialization order.
-
-	* ext/extmk.rb (extmake): compact $extlibs.
-
-Tue Aug 12 02:48:56 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (THREAD_SAVE_CONTEXT): should explicitly turn off the
-	  flag before calling getcontext(2).
-
-	* eval.c (struct thread): add member to save backing store on
-	  IA64. (ruby-bugs PR1086)
-
-	* eval.c (thread_mark): mark IA64 backing store region.
-
-	* eval.c (thread_free): free saved IA64 backing store.
-
-	* eval.c (rb_thread_save_context): save IA64 backing store as well.
-
-	* eval.c (rb_thread_restore_context): restore IA64 backing store.
-
-	* eval.c (THREAD_ALLOC): initialize IA64 members.
-
-Mon Aug 11 22:31:50 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
-	* lib/debug.rb(debug_command): inspection command should inspect
-	  resulting value even if it's nil.  [ruby-dev:21180] by OMAE, jun
-	  <jun66j5@ybb.ne.jp>.
-
-	* lib/debug.rb(debug_command): incomplete regexp.
-
-Mon Aug 11 17:33:07 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call_super): do not use rb_block_given_p() for
-	  check. [ruby-talk:78656]
-
-	* eval.c (BEGIN_CALLARGS): push ITER_NOT only when ITER_PRE.
-
-Sun Aug 10 10:43:05 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* ext/openssl/lib/openssl/buffering.rb: increase BLOCK_SIZE
-	  from 1k to 16k bytes. [ruby-talk:78603]
-
-	* ext/openssl/ossl_ssl.c (ossl_sslctx_s_alloc): enable
-	  partial write to allow interruption in SSLSocket#write.
-
-Sun Aug 10 00:34:16 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* cygwin/GNUmakefile: remove unnecessary '--drive-name=$(CC)'
-	  for ccache.
-
-Sat Aug  9 10:36:21 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* marshal.c (w_object): do not dump generic instance variable when
-	  marshal_dump is defined.
-
-Sat Aug  9 00:35:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal.c: F style output(like 1234.56789) implemented
-	  to to_s method.
-	* ext/bigdecimal_??.html: F style output(like 1234.56789)
-	  implemented to to_s method.
-
-Fri Aug  8 12:33:17 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* bcc32/Makefile.sub: rubyw.exe should be a Windows GUI program.
-	  add the -aa option to WLDFLAGS.
-
-Fri Aug  8 11:29:26 2003  Koji Arai  <jca02266@nifty.ne.jp>
-
-	* marshal.c (w_object): should set `c_arg' at first.
-
-Fri Aug  8 03:22:28 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
-
-	* lib/webrick/httputils.rb (FormData#list): should not take
-	  a side effect for the receiver.
-
-Thu Aug  7 14:40:37 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* cygwin/GNUmakefile: better --disbale-shared option support.
-
-	* cygwin/GNUmakefile: add forwarding DLL target for cygwin.
-
-Thu Aug  7 14:21:05 2003  Corinna Vinschen  <vinschen@redhat.com>
-
-	* configure.in: Fix Cygwin specific naming of libraries to
-	  be net distribution compliant. (ruby-bugs:PR#1077)
-	  cygwin-ruby18.dll -> cygruby18.dll
-
-Thu Aug  7 12:51:38 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_f_at_exit): should not be called without a block.
-	  block_given check added.
-
-Thu Aug  7 06:46:06 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_call0): forgot to pop ruby_class.
-
-	* eval.c (rb_call0): update ruby_class as well as ruby_cref.
-	  (ruby-bugs-ja:PR#540)
-
-Thu Aug  7 04:52:50 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_yield_0): remove ruby_frame->cbase and unify to
-	  ruby_cref.  [ruby-talk:78141]
-
-Thu Aug  7 04:19:15 2003  Akinori MUSHA  <knu@iDaemons.org>
-
-	* gc.c: FreeBSD/ia64's mcontext_t is a bit different from that of
-	  Linux/ia64.  This makes gc.c compile but miniruby coredumps for
-	  the moment.
-
-Thu Aug  7 00:15:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
-
-	* ext/bigdecimal.c: Comparison results adjusted to Float's.
-	* ext/bigdecimal.c: Use rb_num_coerce_????(x,y) instead of own.
-
-Wed Aug  6 22:58:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
-
-	* lib/test/unit/testcase.rb: Added equality checking.
-	* lib/test/unit/testsuite.rb: Added equality checking.
-	* lib/test/unit/assertions.rb: Fixed a warning.
-
-Wed Aug  6 17:28:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* ext/extmk.rb (extmake): pass LIBPATH to make ruby.  [ruby-dev:21137]
-
-	* ext/extmk.rb (extmake): set library name as source file name in
-	  Init_ext().  [ruby-dev:21137]
-
-	* lib/mkmf.rb (Logging::postpone): postpone logging messages after
-	  heading message as the result of the block.
-
-	* lib/mkmf.rb (macro_defined?): append newline to src unless ended
-	  with it.
-
-	* lib/mkmf.rb (have_library): treat nil function name as "main".
-	  (ruby-bugs:PR#1083)
-
-	* lib/mkmf.rb (pkg_config): should append additional libraries to
-	  $libs but not $LIBS.  [ruby-dev:21137]
-
-	* ext/io/wait/extconf.rb: check DOSISH macro instead of platform.
-
-	* ext/digest/sha1/extconf.rb: have_library already appends library
-	  name.
-
-Wed Aug  6 17:23:57 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c: initialize /* OK */ variables by Qnil to stop warnings.
-
-Wed Aug  6 04:58:32 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
-
-	* ext/Setup*: add io/wait and openssl.
-
-Wed Aug  6 01:13:38 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* eval.c (rb_f_autoload): use ruby_cbase instead of ruby_class.
-
-	* eval.c (rb_f_autoload_p): ditto.
-
-	* class.c (rb_mod_init_copy): no longer implements independent
-	  clone and dup methods.  override "initialize_copy" instead.
-	  [ruby-core:01352]
-
-	* object.c (rb_class_s_alloc): define Class allocation function.
-	  this makes Classes to follow clone framework that uses
-	  initialize_copy.
-
-	* object.c (rb_class_initialize): separate instantiation and
-	  initialization.
-
-	* object.c (rb_obj_alloc): prohibit instantiation from
-	  uninitialized class.
-
-	* object.c (rb_class_superclass): check uninitialized class.
-
-	* array.c (rb_ary_fill): wrong index processing with block.  this
-	  fix was done by Koji Arai <JCA02266@nifty.ne.jp> [ruby-list:38029]
-
-	* marshal.c (w_object): should preserve generic ivar for nil,
-	  true, false, symbols, and fixnums.
-
-	* marshal.c (w_uclass): base_klass check should be done after
-	  rb_class_real().
-
-Wed Aug  6 01:18:50 2003  Minero Aoki  <aamine@loveruby.net>
-
-	* lib/net/http.rb: update document.
-
-	* lib/net/pop.rb: ditto.
-
-	* lib/net/protocol.rb: ditto.
-
-Wed Aug  6 00:48:37 2003  Koji Arai  <jca02266@nifty.ne.jp>
-
-	* marshal.c (w_object): should recommend marshal_dump rather than
-	  _dump_data.
-
-Tue Aug  5 17:58:57 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
-
-	* lib/fileutils.rb (install): should preserve timestamp only.
-
-Tue Aug  5 17:31:59 2003  Ian Macdonald  <ian@caliban.org>
-
-	* lib/shell/command-processor.rb (Shell::CommandProcessor::rmdir):
-	  simple typo.
-
-Tue Aug  5 15:47:34 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
-
-	* eval.c (rb_load): should preserve current source file/line.
-
-Tue Aug  5 10:04:42 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* string.c (str_new4): ptr may refer null_str.
-
-Mon Aug  4 17:25:18 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
-
-	* stable version 1.8.0 released.
-
-For the changes before 1.8.0, see doc/ChangeLog-1.8.0
-
-Local variables:
-add-log-time-format: (lambda ()
-  (let* ((time (current-time))
-	 (diff (+ (cadr time) 32400))
-	 (lo (% diff 65536))
-	 (hi  (+ (car time) (/ diff 65536))))
-  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
-indent-tabs-mode: t
-tab-width: 8
-end:
diff --git a/GPL b/GPL
index 5b6e7c66c2..d159169d10 100644
--- a/GPL
+++ b/GPL
@@ -1,12 +1,12 @@
-		    GNU GENERAL PUBLIC LICENSE
-		       Version 2, June 1991
+                    GNU GENERAL PUBLIC LICENSE
+                       Version 2, June 1991
 
- Copyright (C) 1989, 1991 Free Software Foundation, Inc.
-                       59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ Copyright (C) 1989, 1991 Free Software Foundation, Inc.,
+ 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  Everyone is permitted to copy and distribute verbatim copies
  of this license document, but changing it is not allowed.
 
-			    Preamble
+                            Preamble
 
   The licenses for most software are designed to take away your
 freedom to share and change it.  By contrast, the GNU General Public
@@ -15,7 +15,7 @@ software--to make sure the software is free for all its users.  This
 General Public License applies to most of the Free Software
 Foundation's software and to any other program whose authors commit to
 using it.  (Some other Free Software Foundation software is covered by
-the GNU Library General Public License instead.)  You can apply it to
+the GNU Lesser General Public License instead.)  You can apply it to
 your programs, too.
 
   When we speak of free software, we are referring to freedom, not
@@ -55,8 +55,8 @@ patent must be licensed for everyone's free use or not licensed at all.
 
   The precise terms and conditions for copying, distribution and
 modification follow.
-
-		    GNU GENERAL PUBLIC LICENSE
+
+                    GNU GENERAL PUBLIC LICENSE
    TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
 
   0. This License applies to any program or other work which contains
@@ -110,7 +110,7 @@ above, provided that you also meet all of these conditions:
     License.  (Exception: if the Program itself is interactive but
     does not normally print such an announcement, your work based on
     the Program is not required to print an announcement.)
-
+
 These requirements apply to the modified work as a whole.  If
 identifiable sections of that work are not derived from the Program,
 and can be reasonably considered independent and separate works in
@@ -168,7 +168,7 @@ access to copy from a designated place, then offering equivalent
 access to copy the source code from the same place counts as
 distribution of the source code, even though third parties are not
 compelled to copy the source along with the object code.
-
+
   4. You may not copy, modify, sublicense, or distribute the Program
 except as expressly provided under this License.  Any attempt
 otherwise to copy, modify, sublicense or distribute the Program is
@@ -225,7 +225,7 @@ impose that choice.
 
 This section is intended to make thoroughly clear what is believed to
 be a consequence of the rest of this License.
-
+
   8. If the distribution and/or use of the Program is restricted in
 certain countries either by patents or by copyrighted interfaces, the
 original copyright holder who places the Program under this License
@@ -255,7 +255,7 @@ make exceptions for this.  Our decision will be guided by the two goals
 of preserving the free status of all derivatives of our free software and
 of promoting the sharing and reuse of software generally.
 
-			    NO WARRANTY
+                            NO WARRANTY
 
   11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
 FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW.  EXCEPT WHEN
@@ -277,9 +277,9 @@ YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
 PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
 POSSIBILITY OF SUCH DAMAGES.
 
-		     END OF TERMS AND CONDITIONS
-
-	    How to Apply These Terms to Your New Programs
+                     END OF TERMS AND CONDITIONS
+
+            How to Apply These Terms to Your New Programs
 
   If you develop a new program, and you want it to be of the greatest
 possible use to the public, the best way to achieve this is to make it
@@ -303,10 +303,9 @@ the "copyright" line and a pointer to where the full notice is found.
     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
     GNU General Public License for more details.
 
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-
+    You should have received a copy of the GNU General Public License along
+    with this program; if not, write to the Free Software Foundation, Inc.,
+    51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
 
 Also add information on how to contact you by electronic and paper mail.
 
@@ -336,5 +335,5 @@ necessary.  Here is a sample; alter the names:
 This General Public License does not permit incorporating your program into
 proprietary programs.  If your program is a subroutine library, you may
 consider it more useful to permit linking proprietary applications with the
-library.  If this is what you want to do, use the GNU Library General
+library.  If this is what you want to do, use the GNU Lesser General
 Public License instead of this License.
diff --git a/KNOWNBUGS.rb b/KNOWNBUGS.rb
new file mode 100644
index 0000000000..35a8e75876
--- /dev/null
+++ b/KNOWNBUGS.rb
@@ -0,0 +1,7 @@
+#
+# IMPORTANT: Always keep the first 7 lines (comments),
+# even if this file is otherwise empty.
+#
+# This test file includes tests which point out known bugs.
+# So all tests will cause failure.
+#
diff --git a/LEGAL b/LEGAL
index 908eb270f5..0fe24dfc3c 100644
--- a/LEGAL
+++ b/LEGAL
@@ -1,61 +1,231 @@
-LEGAL NOTICE INFORMATION
-------------------------
+# -*- rdoc -*-
+
+= LEGAL NOTICE INFORMATION
+--------------------------
 
 All the files in this distribution are covered under either the Ruby's
 license (see the file COPYING) or public-domain except some files
 mentioned below.
 
-regex.[ch]:
+ccan/build_assert/build_assert.h::
+ccan/check_type/check_type.h::
+ccan/container_of/container_of.h::
+ccan/str/str.h::
+
+  These files are licensed under the CC0.
+
+  >>>
+    https://creativecommons.org/choose/zero/
+
+ccan/list/list.h::
+
+  This file is licensed under the MIT License.
+
+  >>>
+    Permission is hereby granted, free of charge, to any person obtaining a copy
+    of this software and associated documentation files (the "Software"), to deal
+    in the Software without restriction, including without limitation the rights
+    to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+    copies of the Software, and to permit persons to whom the Software is
+    furnished to do so, subject to the following conditions:
+
+    The above copyright notice and this permission notice shall be included in
+    all copies or substantial portions of the Software.
+
+    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+    THE SOFTWARE.
+
+include/ruby/onigmo.h::
+include/ruby/oniguruma.h::
+regcomp.c::
+regenc.c::
+regenc.h::
+regerror.c::
+regexec.c::
+regint.h::
+regparse.c::
+regparse.h::
+enc/ascii.c::
+enc/big5.c::
+enc/cp949.c::
+enc/emacs_mule.c::
+enc/encdb.c::
+enc/euc_jp.c::
+enc/euc_kr.c::
+enc/euc_tw.c::
+enc/gb18030.c::
+enc/gb2312.c::
+enc/gbk.c::
+enc/iso_8859_1.c::
+enc/iso_8859_10.c::
+enc/iso_8859_11.c::
+enc/iso_8859_13.c::
+enc/iso_8859_14.c::
+enc/iso_8859_15.c::
+enc/iso_8859_16.c::
+enc/iso_8859_2.c::
+enc/iso_8859_3.c::
+enc/iso_8859_4.c::
+enc/iso_8859_5.c::
+enc/iso_8859_6.c::
+enc/iso_8859_7.c::
+enc/iso_8859_8.c::
+enc/iso_8859_9.c::
+enc/koi8_r.c::
+enc/koi8_u.c::
+enc/shift_jis.c::
+enc/unicode.c::
+enc/us_ascii.c::
+enc/utf_16be.c::
+enc/utf_16le.c::
+enc/utf_32be.c::
+enc/utf_32le.c::
+enc/utf_8.c::
+enc/windows_1251.c::
+
+  Onigmo (Oniguruma-mod) LICENSE
+
+  >>>
+    Copyright (c) 2002-2009  K.Kosako  <sndgk393 AT ybb DOT ne DOT jp>  
+    Copyright (c) 2011-2014  K.Takata  <kentkt AT csc DOT jp>  
+    All rights reserved.
 
-  These files are under LGPL.  Treat them as LGPL says. (See the file
-  LGPL for details)
+    Redistribution and use in source and binary forms, with or without
+    modification, are permitted provided that the following conditions
+    are met:
+    1. Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
 
-    Extended regular expression matching and search library.
-    Copyright (C) 1993, 94, 95, 96, 97, 98 Free Software Foundation, Inc.
+    THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+    ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+    ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+    DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+    OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+    HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+    OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+    SUCH DAMAGE.
 
-    The GNU C Library is free software; you can redistribute it and/or
-    modify it under the terms of the GNU Library General Public License as
-    published by the Free Software Foundation; either version 2 of the
-    License, or (at your option) any later version.
+  Oniguruma LICENSE
 
-    The GNU C Library is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
-    Library General Public License for more details.
+  >>>
+    Copyright (c) 2002-2009  K.Kosako  <sndgk393 AT ybb DOT ne DOT jp>  
+    All rights reserved.
+
+    Redistribution and use in source and binary forms, with or without
+    modification, are permitted provided that the following conditions
+    are met:
+    1. Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
+
+    THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+    ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+    ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+    DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+    OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+    HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+    OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+    SUCH DAMAGE.
+
+  * https://github.com/k-takata/Onigmo/
+  * https://github.com/kkos/oniguruma
+  * https://svnweb.freebsd.org/ports/head/devel/oniguruma/
+
+    When this software is partly used or it is distributed with Ruby,
+    this of Ruby follows the license of Ruby.
+
+enc/trans/GB/GB12345%UCS.src::
+enc/trans/GB/UCS%GB12345.src::
+enc/trans/GB/GB2312%UCS.src::
+enc/trans/GB/UCS%GB2312.src::
 
-    You should have received a copy of the GNU Library General Public
-    License along with the GNU C Library; see the file LGPL.  If not,
-    write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
-    Boston, MA 02111-1307, USA.  */
+  These files have this explanatory texts.
 
-    Multi-byte extension added May, 1993 by t^2 (Takahiro Tanimoto)
-    Last change: May 21, 1993 by t^2
-    removed gapped buffer support, multiple syntax support by matz <matz@nts.co.jp>
-    Perl5 extension added by matz <matz@caelum.co.jp>
-    UTF-8 extension added Jan 16 1999 by Yoshida Masato  <yoshidam@tau.bekkoame.ne.jp>
+  >>>
+    This mapping data was created from files provided by Unicode, Inc.
+    (The Unicode Consortium). The files were used to create a product supporting
+    Unicode, as explicitly permitted in the files' copyright notices.
+    Please note that Unicode, Inc. never made any claims as to fitness of these
+    files for any particular purpose, and has ceased to publish the files many
+    years ago.
 
-configure:
+enc/trans/JIS/JISX0201-KANA%UCS.src::
+enc/trans/JIS/JISX0208\@1990%UCS.src::
+enc/trans/JIS/JISX0212%UCS.src::
+enc/trans/JIS/UCS%JISX0201-KANA.src::
+enc/trans/JIS/UCS%JISX0208@1990.src::
+enc/trans/JIS/UCS%JISX0212.src::
+
+  These files are copyrighted as the following.
+
+  >>>
+    © 2015 Unicode®, Inc.
+
+    For terms of use, see http://www.unicode.org/terms_of_use.html
+
+enc/trans/JIS/JISX0213-1%UCS@BMP.src::
+enc/trans/JIS/JISX0213-1%UCS@SIP.src::
+enc/trans/JIS/JISX0213-2%UCS@BMP.src::
+enc/trans/JIS/JISX0213-2%UCS@SIP.src::
+
+  These files are copyrighted as the following.
+
+  >>>
+    Copyright (C) 2001 earthian@tama.or.jp, All Rights Reserved.  
+    Copyright (C) 2001 I'O, All Rights Reserved.  
+    Copyright (C) 2006 Project X0213, All Rights Reserved.  
+    You can use, modify, distribute this table freely.
+
+enc/trans/JIS/UCS@BMP%JISX0213-1.src::
+enc/trans/JIS/UCS@BMP%JISX0213-2.src::
+enc/trans/JIS/UCS@SIP%JISX0213-1.src::
+enc/trans/JIS/UCS@SIP%JISX0213-2.src::
+
+  These files are copyrighted as the following.
+
+  >>>
+    Copyright (C) 2001 earthian@tama.or.jp, All Rights Reserved.  
+    Copyright (C) 2001 I'O, All Rights Reserved.  
+    You can use, modify, distribute this table freely.
+
+configure::
 
   This file is free software.
 
-    Copyright (C) 1992, 93, 94, 95, 96 Free Software Foundation, Inc.
+  >>>
+    Copyright (C) 1992-1996, 1998-2012 Free Software Foundation, Inc.
 
     This configure script is free software; the Free Software Foundation
     gives unlimited permission to copy, distribute and modify it.
 
-config.guess:
-config.sub:
-parse.c:
+tool/config.guess::
+tool/config.sub::
 
   As long as you distribute these files with the file configure, they
   are covered under the Ruby's license.
 
-      Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999
-      Free Software Foundation, Inc.
+  >>>
+    Copyright 1992-2018 Free Software Foundation, Inc.
 
     This file is free software; you can redistribute it and/or modify it
     under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
+    the Free Software Foundation; either version 3 of the License, or
     (at your option) any later version.
 
     This program is distributed in the hope that it will be useful, but
@@ -64,54 +234,134 @@ parse.c:
     General Public License for more details.
 
     You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+    along with this program; if not, see <https://www.gnu.org/licenses/>.
 
     As a special exception to the GNU General Public License, if you
     distribute this file as part of a program that contains a
     configuration script generated by Autoconf, you may include it under
-    the same distribution terms that you use for the rest of that program.
+    the same distribution terms that you use for the rest of that
+    program.  This Exception is an additional permission under section 7
+    of the GNU General Public License, version 3 ("GPLv3").
+
+parse.c::
+
+  This file is licensed under the GPL, but is incorporated into Ruby and
+  redistributed under the terms of the Ruby license, as permitted by the
+  exception to the GPL below.
+
+  >>>
+    Copyright (C) 1984, 1989-1990, 2000-2015, 2018 Free Software Foundation, Inc.
+
+    This program is free software: you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+    As a special exception, you may create a larger work that contains
+    part or all of the Bison parser skeleton and distribute that work
+    under terms of your choice, so long as that work isn't itself a
+    parser generator using the skeleton or a modified version thereof
+    as a parser skeleton.  Alternatively, if you modify or redistribute
+    the parser skeleton itself, you may (at your option) remove this
+    special exception, which will cause the skeleton and the resulting
+    Bison output files to be licensed under the GNU General Public
+    License without this special exception.
+
+    This special exception was added by the Free Software Foundation in
+    version 2.2 of Bison.
+
+util.c (partly)::
+
+  This file contains the source code under these licenses.
+
+  >>>
+    Copyright (c) 1991, 2000, 2001 by Lucent Technologies.
+
+    Permission to use, copy, modify, and distribute this software for any
+    purpose without fee is hereby granted, provided that this entire notice
+    is included in all copies of any software which is or includes a copy
+    or modification of this software and in all copies of the supporting
+    documentation for such software.
+
+    THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR IMPLIED
+    WARRANTY.  IN PARTICULAR, NEITHER THE AUTHOR NOR LUCENT MAKES ANY
+    REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE MERCHANTABILITY
+    OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR PURPOSE.
+
+  >>>
+    Copyright (c) 2004-2008 David Schultz <das@FreeBSD.ORG>  
+    All rights reserved.
+
+    Redistribution and use in source and binary forms, with or without
+    modification, are permitted provided that the following conditions
+    are met:
+    1. Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
+
+    THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+    ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+    ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+    DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+    OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+    HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+    OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+    SUCH DAMAGE.
 
-util.c (partly):
-win32/win32.[ch]:
+win32/win32.{c,h}::
 
   You can apply the Artistic License to these files. (or GPL,
   alternatively)
 
+  >>>
     Copyright (c) 1993, Intergraph Corporation
 
     You may distribute under the terms of either the GNU General Public
     License or the Artistic License, as specified in the perl README file.
 
-random.c
+random.c::
 
-  This file is under the new-style BSD license.
+  This file contains the source code under the new-style BSD license.
 
+  >>>
     A C-program for MT19937, with initialization improved 2002/2/10.
-    Coded by Takuji Nishimura and Makoto Matsumoto.
+    Coded by Takuji Nishimura and Makoto Matsumoto.  
     This is a faster version by taking Shawn Cokus's optimization,
     Matthe Bellew's simplification, Isaku Wada's real version.
 
-    Before using, initialize the state by using init_genrand(seed) 
+    Before using, initialize the state by using init_genrand(seed)
     or init_by_array(init_key, key_length).
 
     Copyright (C) 1997 - 2002, Makoto Matsumoto and Takuji Nishimura,
-    All rights reserved.                          
+    All rights reserved.
 
     Redistribution and use in source and binary forms, with or without
     modification, are permitted provided that the following conditions
     are met:
 
-      1. Redistributions of source code must retain the above copyright
-	 notice, this list of conditions and the following disclaimer.
+    1. Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
 
-      2. Redistributions in binary form must reproduce the above copyright
-	 notice, this list of conditions and the following disclaimer in the
-	 documentation and/or other materials provided with the distribution.
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
 
-      3. The names of its contributors may not be used to endorse or promote 
-	 products derived from this software without specific prior written 
-	 permission.
+    3. The names of its contributors may not be used to endorse or promote
+       products derived from this software without specific prior written
+       permission.
 
     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
@@ -130,69 +380,47 @@ random.c
     http://www.math.keio.ac.jp/matumoto/emt.html
     email: matumoto@math.keio.ac.jp
 
-st.[ch]:
-x68/*:
-missing/alloca.c:
-missing/dup2.c:
-missing/finite.c:
-missing/hypot.c:
-missing/isinf.c:
-missing/isnan.c:
-missing/memcmp.c:
-missing/memmove.c:
-missing/strcasecmp.c:
-missing/strchr.c:
-missing/streror.c:
-missing/strftime.c:
-missing/strncasecmp.c:
-missing/strstr.c:
-missing/strtol.c:
-ext/digest/sha1/sha1.[ch]:
-
-  These files are all under public domain.
+  The Wayback Machine url: http://web.archive.org/web/19990429082237/http://www.math.keio.ac.jp/matumoto/emt.html
 
-missing/strtod.c:
+vm_dump.c:procstat_vm::
 
-  This file will not be used on most platforms depending on how the
-  configure script results.  In any case you must not receive any fee
-  with the file itself.
+  This file contains the source code under the new-style BSD license.
 
-    Copyright (c) 1988-1993 The Regents of the University of California.
-    Copyright (c) 1994 Sun Microsystems, Inc.
-
-    Permission to use, copy, modify, and distribute this
-    software and its documentation for any purpose and without
-    fee is hereby granted, provided that the above copyright
-    notice appear in all copies.  The University of California
-    makes no representations about the suitability of this
-    software for any purpose.  It is provided "as is" without
-    express or implied warranty.
-
-missing/strtoul.c:
+  >>>
+    Copyright (c) 2007 Robert N. M. Watson  
+    All rights reserved.
 
-  This file will not be used on most platforms depending on how the
-  configure script results.  In any case you must not receive any fee
-  with the file itself.
+    Redistribution and use in source and binary forms, with or without
+    modification, are permitted provided that the following conditions
+    are met:
+    1. Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
 
-    Copyright 1988 Regents of the University of California
+    THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+    ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+    ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+    DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+    OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+    HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+    OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+    SUCH DAMAGE.
 
-    Permission to use, copy, modify, and distribute this
-    software and its documentation for any purpose and without
-    fee is hereby granted, provided that the above copyright
-    notice appear in all copies.  The University of California
-    makes no representations about the suitability of this
-    software for any purpose.  It is provided "as is" without
-    express or implied warranty.
+    $FreeBSD: head/usr.bin/procstat/procstat_vm.c 261780 2014-02-11 21:57:37Z jhb $
 
-missing/erf.c:
-missing/crypt.c:
-missing/vsnprintf.c:
+vsnprintf.c::
 
   This file is under the old-style BSD license.  Note that the
   paragraph 3 below is now null and void.
 
-    Copyright (c) 1990, 1993
-         The Regents of the University of California.  All rights reserved.
+  >>>
+    Copyright (c) 1990, 1993  
+    The Regents of the University of California.  All rights reserved.
 
     This code is derived from software contributed to Berkeley by
     Chris Torek.
@@ -226,11 +454,142 @@ missing/vsnprintf.c:
     From ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change
     paragraph 3 above is now null and void.
 
-ext/digest/md5/md5.[ch]:
+st.c::
+strftime.c::
+include/ruby/st.h::
+missing/acosh.c::
+missing/alloca.c::
+missing/dup2.c::
+missing/erf.c::
+missing/finite.c::
+missing/hypot.c::
+missing/isinf.c::
+missing/isnan.c::
+missing/lgamma_r.c::
+missing/memcmp.c::
+missing/memmove.c::
+missing/strchr.c::
+missing/strerror.c::
+missing/strstr.c::
+missing/tgamma.c::
+ext/date/date_strftime.c::
+ext/digest/sha1/sha1.c::
+ext/digest/sha1/sha1.h::
+ext/sdbm/_sdbm.c::
+ext/sdbm/sdbm.h::
+
+  These files are all under public domain.
+
+missing/crypt.c::
+
+  This file is under the old-style BSD license.  Note that the
+  paragraph 3 below is now null and void.
+
+  >>>
+    Copyright (c) 1989, 1993  
+    The Regents of the University of California.  All rights reserved.
+
+    This code is derived from software contributed to Berkeley by
+    Tom Truscott.
+
+    Redistribution and use in source and binary forms, with or without
+    modification, are permitted provided that the following conditions
+    are met:
+    1. Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
+    3. Neither the name of the University nor the names of its contributors
+       may be used to endorse or promote products derived from this software
+       without specific prior written permission.
+
+    THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+    ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+    ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+    DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+    OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+    HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+    OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+    SUCH DAMAGE.
+
+missing/setproctitle.c::
+
+  This file is under the old-style BSD license.  Note that the
+  paragraph 3 below is now null and void.
+
+  >>>
+    Copyright 2003 Damien Miller  
+    Copyright (c) 1983, 1995-1997 Eric P. Allman  
+    Copyright (c) 1988, 1993  
+    The Regents of the University of California.  All rights reserved.
+
+    Redistribution and use in source and binary forms, with or without
+    modification, are permitted provided that the following conditions
+    are met:
+    1. Redistributions of source code must retain the above copyright
+       notice, this list of conditions and the following disclaimer.
+    2. Redistributions in binary form must reproduce the above copyright
+       notice, this list of conditions and the following disclaimer in the
+       documentation and/or other materials provided with the distribution.
+    3. Neither the name of the University nor the names of its contributors
+       may be used to endorse or promote products derived from this software
+       without specific prior written permission.
+
+    THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+    ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+    IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+    ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+    FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+    DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+    OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+    HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+    LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+    OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+    SUCH DAMAGE.
+
+missing/strlcat.c::
+missing/strlcpy.c::
+
+  These files are under an ISC-style license.
+
+  >>>
+    Copyright (c) 1998, 2015 Todd C. Miller <Todd.Miller@courtesan.com>
+
+    Permission to use, copy, modify, and distribute this software for any
+    purpose with or without fee is hereby granted, provided that the above
+    copyright notice and this permission notice appear in all copies.
+
+    THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+    WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+    MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+    ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+    WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+    ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+    OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+
+missing/langinfo.c::
+
+  This file is from http://www.cl.cam.ac.uk/~mgk25/ucs/langinfo.c.
+  Ruby uses a modified version. The file contains the following
+  author/copyright notice:
+
+  >>>
+    Markus.Kuhn@cl.cam.ac.uk -- 2002-03-11  
+    Permission to use, copy, modify, and distribute this software
+    for any purpose and without fee is hereby granted. The author
+    disclaims all warranties with regard to this software.
+
+ext/digest/md5/md5.c::
+ext/digest/md5/md5.h::
 
   These files are under the following license.  Ruby uses modified
   versions of them.
 
+  >>>
     Copyright (C) 1999, 2000 Aladdin Enterprises.  All rights reserved.
 
     This software is provided 'as-is', without any express or implied
@@ -252,33 +611,26 @@ ext/digest/md5/md5.[ch]:
     L. Peter Deutsch
     ghost@aladdin.com
 
-ext/digest/rmd160/rmd160.[ch]:
+ext/digest/rmd160/rmd160.c::
+ext/digest/rmd160/rmd160.h::
 
   These files have the following copyright information, and by the
   author we are allowed to use it under the new-style BSD license.
 
-    AUTHOR:   Antoon Bosselaers, ESAT-COSIC
+  >>>
+    AUTHOR::  Antoon Bosselaers, ESAT-COSIC
               (Arranged for libc by Todd C. Miller)
-    DATE:     1 March 1996
+    DATE::    1 March 1996
 
     Copyright (c) Katholieke Universiteit Leuven
     1996, All Rights Reserved
 
-ext/digest/rmd160/rmd160hl.c:
-ext/digest/sha1/sha1hl.c:
-
-  These files are under the beer-ware license.
-
-    "THE BEER-WARE LICENSE" (Revision 42):
-    <phk@login.dkuug.dk> wrote this file.  As long as you retain this notice you
-    can do whatever you want with this stuff. If we meet some day, and you think
-    this stuff is worth it, you can buy me a beer in return.   Poul-Henning Kamp
-
-ext/digest/sha2/sha2.[ch]:
-ext/digest/sha2/sha2hl.c:
+ext/digest/sha2/sha2.c::
+ext/digest/sha2/sha2.h::
 
   These files are under the new-style BSD license.
 
+  >>>
     Copyright 2000 Aaron D. Gifford.  All rights reserved.
 
     Redistribution and use in source and binary forms, with or without
@@ -305,28 +657,108 @@ ext/digest/sha2/sha2hl.c:
     OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
     SUCH DAMAGE.
 
-ext/nkf/nkf-utf8/config.h:
-ext/nkf/nkf-utf8/nkf.c:
-ext/nkf/nkf-utf8/utf8tbl.c:
+ext/json/generator/generator.c::
+
+  The file contains the following copyright notice.
+
+  >>>
+    Copyright 2001-2004 Unicode, Inc.
+
+    Disclaimer::
+
+      This source code is provided as is by Unicode, Inc. No claims are
+      made as to fitness for any particular purpose. No warranties of any
+      kind are expressed or implied. The recipient agrees to determine
+      applicability of information provided. If this file has been
+      purchased on magnetic or optical media from Unicode, Inc., the
+      sole remedy for any claim will be exchange of defective media
+      within 90 days of receipt.
+
+    Limitations on Rights to Redistribute This Code::
+
+      Unicode, Inc. hereby grants the right to freely use the information
+      supplied in this file in the creation of products supporting the
+      Unicode Standard, and to make copies of this file in any form
+      for internal or external distribution as long as this notice
+      remains attached.
+
+ext/nkf/nkf-utf8/config.h::
+ext/nkf/nkf-utf8/nkf.c::
+ext/nkf/nkf-utf8/utf8tbl.c::
 
   These files are under the following license.  So to speak, it is
   copyrighted semi-public-domain software.
 
+  >>>
     Copyright (C) 1987, Fujitsu LTD. (Itaru ICHIKAWA)
-       Everyone is permitted to do anything on this program 
-       including copying, modifying, improving,
-       as long as you don't try to pretend that you wrote it.
-       i.e., the above copyright notice has to appear in all copies.
-       Binary distribution requires original version messages.
-       You don't have to ask before copying, redistribution or publishing.
-       THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE.
-
-ext/socket/addrinfo.h:
-ext/socket/getaddrinfo.c:
-ext/socket/getnameinfo.c:
+
+    Everyone is permitted to do anything on this program
+    including copying, modifying, improving,
+    as long as you don't try to pretend that you wrote it.
+    i.e., the above copyright notice has to appear in all copies.
+    Binary distribution requires original version messages.
+    You don't have to ask before copying, redistribution or publishing.
+    THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE.
+
+ext/psych::
+test/psych::
+
+  The files under these directories are under the following license, except for
+  ext/psych/yaml.
+
+  >>>
+    Copyright 2009 Aaron Patterson, et al.
+
+    Permission is hereby granted, free of charge, to any person obtaining a copy of
+    this software and associated documentation files (the 'Software'), to deal in
+    the Software without restriction, including without limitation the rights to
+    use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+    of the Software, and to permit persons to whom the Software is furnished to do
+    so, subject to the following conditions:
+
+    The above copyright notice and this permission notice shall be included in all
+    copies or substantial portions of the Software.
+
+    THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+    SOFTWARE.
+
+ext/psych/yaml::
+
+  The files under this directory are under the following license.
+
+  >>>
+    Copyright (c) 2006 Kirill Simonov
+
+    Permission is hereby granted, free of charge, to any person obtaining a copy of
+    this software and associated documentation files (the "Software"), to deal in
+    the Software without restriction, including without limitation the rights to
+    use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
+    of the Software, and to permit persons to whom the Software is furnished to do
+    so, subject to the following conditions:
+
+    The above copyright notice and this permission notice shall be included in all
+    copies or substantial portions of the Software.
+
+    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+    SOFTWARE.
+
+ext/socket/addrinfo.h::
+ext/socket/getaddrinfo.c::
+ext/socket/getnameinfo.c::
 
   These files are under the new-style BSD license.
 
+  >>>
     Copyright (C) 1995, 1996, 1997, 1998, and 1999 WIDE Project.
     All rights reserved.
 
@@ -354,17 +786,150 @@ ext/socket/getnameinfo.c:
     OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
     SUCH DAMAGE.
 
-ext/win32ole/win32ole.c:
+ext/win32ole/win32ole.c::
 
   You can apply the Artistic License to this file. (or GPL,
   alternatively)
 
+  >>>
     (c) 1995 Microsoft Corporation. All rights reserved.
     Developed by ActiveWare Internet Corp., http://www.ActiveWare.com
 
     Other modifications Copyright (c) 1997, 1998 by Gurusamy Sarathy
     <gsar@umich.edu> and Jan Dubois <jan.dubois@ibm.net>
- 
+
     You may distribute under the terms of either the GNU General Public
     License or the Artistic License, as specified in the README file
     of the Perl distribution.
+
+  The Wayback Machine url: http://web.archive.org/web/19970607104352/http://www.activeware.com:80/
+
+lib/rdoc/generator/template/darkfish/css/fonts.css::
+
+  This file is licensed under the SIL Open Font License.
+
+  >>>
+    http://scripts.sil.org/OFL
+
+spec/mspec::
+spec/ruby::
+
+  The files under these directories are under the following license.
+
+  >>>
+    Copyright (c) 2008 Engine Yard, Inc. All rights reserved.
+
+    Permission is hereby granted, free of charge, to any person
+    obtaining a copy of this software and associated documentation
+    files (the "Software"), to deal in the Software without
+    restriction, including without limitation the rights to use,
+    copy, modify, merge, publish, distribute, sublicense, and/or sell
+    copies of the Software, and to permit persons to whom the
+    Software is furnished to do so, subject to the following
+    conditions:
+
+    The above copyright notice and this permission notice shall be
+    included in all copies or substantial portions of the Software.
+
+    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+    EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
+    OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+    NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
+    HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
+    WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+    FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+    OTHER DEALINGS IN THE SOFTWARE.
+
+lib/rubygems.rb::
+lib/rubygems::
+test/rubygems::
+
+  RubyGems is under the following license.
+
+  >>>
+    RubyGems is copyrighted free software by Chad Fowler, Rich Kilmer, Jim
+    Weirich and others.  You can redistribute it and/or modify it under
+    either the terms of the MIT license (see the file MIT.txt), or the
+    conditions below:
+
+    1. You may make and give away verbatim copies of the source form of the
+       software without restriction, provided that you duplicate all of the
+       original copyright notices and associated disclaimers.
+
+    2. You may modify your copy of the software in any way, provided that
+       you do at least ONE of the following:
+
+       a. place your modifications in the Public Domain or otherwise
+          make them Freely Available, such as by posting said
+          modifications to Usenet or an equivalent medium, or by allowing
+          the author to include your modifications in the software.
+
+       b. use the modified software only within your corporation or
+          organization.
+
+       c. give non-standard executables non-standard names, with
+          instructions on where to get the original software distribution.
+
+       d. make other distribution arrangements with the author.
+
+    3. You may distribute the software in object code or executable
+       form, provided that you do at least ONE of the following:
+
+       a. distribute the executables and library files of the software,
+          together with instructions (in the manual page or equivalent)
+          on where to get the original distribution.
+
+       b. accompany the distribution with the machine-readable source of
+          the software.
+
+       c. give non-standard executables non-standard names, with
+          instructions on where to get the original software distribution.
+
+       d. make other distribution arrangements with the author.
+
+    4. You may modify and include the part of the software into any other
+       software (possibly commercial).
+
+    5. The scripts and library files supplied as input to or produced as
+       output from the software do not automatically fall under the
+       copyright of the software, but belong to whomever generated them,
+       and may be sold commercially, and may be aggregated with this
+       software.
+
+    6. THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR
+       IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
+       WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+       PURPOSE.
+
+lib/bundler::
+lib/bundler.rb::
+lib/bundler.gemspec::
+spec/bundler::
+man/bundle-*,gemfile.*::
+
+  Bundler is under the following license.
+
+  >>>
+    Portions copyright (c) 2010 Andre Arko  
+    Portions copyright (c) 2009 Engine Yard  
+
+    MIT License::
+
+      Permission is hereby granted, free of charge, to any person obtaining
+      a copy of this software and associated documentation files (the
+      "Software"), to deal in the Software without restriction, including
+      without limitation the rights to use, copy, modify, merge, publish,
+      distribute, sublicense, and/or sell copies of the Software, and to
+      permit persons to whom the Software is furnished to do so, subject to
+      the following conditions:
+
+      The above copyright notice and this permission notice shall be
+      included in all copies or substantial portions of the Software.
+
+      THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+      EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+      MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+      NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+      LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+      OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+      WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/LGPL b/LGPL
deleted file mode 100644
index b1e3f5a263..0000000000
--- a/LGPL
+++ /dev/null
@@ -1,504 +0,0 @@
-		  GNU LESSER GENERAL PUBLIC LICENSE
-		       Version 2.1, February 1999
-
- Copyright (C) 1991, 1999 Free Software Foundation, Inc.
-     59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
- Everyone is permitted to copy and distribute verbatim copies
- of this license document, but changing it is not allowed.
-
-[This is the first released version of the Lesser GPL.  It also counts
- as the successor of the GNU Library Public License, version 2, hence
- the version number 2.1.]
-
-			    Preamble
-
-  The licenses for most software are designed to take away your
-freedom to share and change it.  By contrast, the GNU General Public
-Licenses are intended to guarantee your freedom to share and change
-free software--to make sure the software is free for all its users.
-
-  This license, the Lesser General Public License, applies to some
-specially designated software packages--typically libraries--of the
-Free Software Foundation and other authors who decide to use it.  You
-can use it too, but we suggest you first think carefully about whether
-this license or the ordinary General Public License is the better
-strategy to use in any particular case, based on the explanations below.
-
-  When we speak of free software, we are referring to freedom of use,
-not price.  Our General Public Licenses are designed to make sure that
-you have the freedom to distribute copies of free software (and charge
-for this service if you wish); that you receive source code or can get
-it if you want it; that you can change the software and use pieces of
-it in new free programs; and that you are informed that you can do
-these things.
-
-  To protect your rights, we need to make restrictions that forbid
-distributors to deny you these rights or to ask you to surrender these
-rights.  These restrictions translate to certain responsibilities for
-you if you distribute copies of the library or if you modify it.
-
-  For example, if you distribute copies of the library, whether gratis
-or for a fee, you must give the recipients all the rights that we gave
-you.  You must make sure that they, too, receive or can get the source
-code.  If you link other code with the library, you must provide
-complete object files to the recipients, so that they can relink them
-with the library after making changes to the library and recompiling
-it.  And you must show them these terms so they know their rights.
-
-  We protect your rights with a two-step method: (1) we copyright the
-library, and (2) we offer you this license, which gives you legal
-permission to copy, distribute and/or modify the library.
-
-  To protect each distributor, we want to make it very clear that
-there is no warranty for the free library.  Also, if the library is
-modified by someone else and passed on, the recipients should know
-that what they have is not the original version, so that the original
-author's reputation will not be affected by problems that might be
-introduced by others.
-
-  Finally, software patents pose a constant threat to the existence of
-any free program.  We wish to make sure that a company cannot
-effectively restrict the users of a free program by obtaining a
-restrictive license from a patent holder.  Therefore, we insist that
-any patent license obtained for a version of the library must be
-consistent with the full freedom of use specified in this license.
-
-  Most GNU software, including some libraries, is covered by the
-ordinary GNU General Public License.  This license, the GNU Lesser
-General Public License, applies to certain designated libraries, and
-is quite different from the ordinary General Public License.  We use
-this license for certain libraries in order to permit linking those
-libraries into non-free programs.
-
-  When a program is linked with a library, whether statically or using
-a shared library, the combination of the two is legally speaking a
-combined work, a derivative of the original library.  The ordinary
-General Public License therefore permits such linking only if the
-entire combination fits its criteria of freedom.  The Lesser General
-Public License permits more lax criteria for linking other code with
-the library.
-
-  We call this license the "Lesser" General Public License because it
-does Less to protect the user's freedom than the ordinary General
-Public License.  It also provides other free software developers Less
-of an advantage over competing non-free programs.  These disadvantages
-are the reason we use the ordinary General Public License for many
-libraries.  However, the Lesser license provides advantages in certain
-special circumstances.
-
-  For example, on rare occasions, there may be a special need to
-encourage the widest possible use of a certain library, so that it becomes
-a de-facto standard.  To achieve this, non-free programs must be
-allowed to use the library.  A more frequent case is that a free
-library does the same job as widely used non-free libraries.  In this
-case, there is little to gain by limiting the free library to free
-software only, so we use the Lesser General Public License.
-
-  In other cases, permission to use a particular library in non-free
-programs enables a greater number of people to use a large body of
-free software.  For example, permission to use the GNU C Library in
-non-free programs enables many more people to use the whole GNU
-operating system, as well as its variant, the GNU/Linux operating
-system.
-
-  Although the Lesser General Public License is Less protective of the
-users' freedom, it does ensure that the user of a program that is
-linked with the Library has the freedom and the wherewithal to run
-that program using a modified version of the Library.
-
-  The precise terms and conditions for copying, distribution and
-modification follow.  Pay close attention to the difference between a
-"work based on the library" and a "work that uses the library".  The
-former contains code derived from the library, whereas the latter must
-be combined with the library in order to run.
-
-		  GNU LESSER GENERAL PUBLIC LICENSE
-   TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
-
-  0. This License Agreement applies to any software library or other
-program which contains a notice placed by the copyright holder or
-other authorized party saying it may be distributed under the terms of
-this Lesser General Public License (also called "this License").
-Each licensee is addressed as "you".
-
-  A "library" means a collection of software functions and/or data
-prepared so as to be conveniently linked with application programs
-(which use some of those functions and data) to form executables.
-
-  The "Library", below, refers to any such software library or work
-which has been distributed under these terms.  A "work based on the
-Library" means either the Library or any derivative work under
-copyright law: that is to say, a work containing the Library or a
-portion of it, either verbatim or with modifications and/or translated
-straightforwardly into another language.  (Hereinafter, translation is
-included without limitation in the term "modification".)
-
-  "Source code" for a work means the preferred form of the work for
-making modifications to it.  For a library, complete source code means
-all the source code for all modules it contains, plus any associated
-interface definition files, plus the scripts used to control compilation
-and installation of the library.
-
-  Activities other than copying, distribution and modification are not
-covered by this License; they are outside its scope.  The act of
-running a program using the Library is not restricted, and output from
-such a program is covered only if its contents constitute a work based
-on the Library (independent of the use of the Library in a tool for
-writing it).  Whether that is true depends on what the Library does
-and what the program that uses the Library does.
-  
-  1. You may copy and distribute verbatim copies of the Library's
-complete source code as you receive it, in any medium, provided that
-you conspicuously and appropriately publish on each copy an
-appropriate copyright notice and disclaimer of warranty; keep intact
-all the notices that refer to this License and to the absence of any
-warranty; and distribute a copy of this License along with the
-Library.
-
-  You may charge a fee for the physical act of transferring a copy,
-and you may at your option offer warranty protection in exchange for a
-fee.
-
-  2. You may modify your copy or copies of the Library or any portion
-of it, thus forming a work based on the Library, and copy and
-distribute such modifications or work under the terms of Section 1
-above, provided that you also meet all of these conditions:
-
-    a) The modified work must itself be a software library.
-
-    b) You must cause the files modified to carry prominent notices
-    stating that you changed the files and the date of any change.
-
-    c) You must cause the whole of the work to be licensed at no
-    charge to all third parties under the terms of this License.
-
-    d) If a facility in the modified Library refers to a function or a
-    table of data to be supplied by an application program that uses
-    the facility, other than as an argument passed when the facility
-    is invoked, then you must make a good faith effort to ensure that,
-    in the event an application does not supply such function or
-    table, the facility still operates, and performs whatever part of
-    its purpose remains meaningful.
-
-    (For example, a function in a library to compute square roots has
-    a purpose that is entirely well-defined independent of the
-    application.  Therefore, Subsection 2d requires that any
-    application-supplied function or table used by this function must
-    be optional: if the application does not supply it, the square
-    root function must still compute square roots.)
-
-These requirements apply to the modified work as a whole.  If
-identifiable sections of that work are not derived from the Library,
-and can be reasonably considered independent and separate works in
-themselves, then this License, and its terms, do not apply to those
-sections when you distribute them as separate works.  But when you
-distribute the same sections as part of a whole which is a work based
-on the Library, the distribution of the whole must be on the terms of
-this License, whose permissions for other licensees extend to the
-entire whole, and thus to each and every part regardless of who wrote
-it.
-
-Thus, it is not the intent of this section to claim rights or contest
-your rights to work written entirely by you; rather, the intent is to
-exercise the right to control the distribution of derivative or
-collective works based on the Library.
-
-In addition, mere aggregation of another work not based on the Library
-with the Library (or with a work based on the Library) on a volume of
-a storage or distribution medium does not bring the other work under
-the scope of this License.
-
-  3. You may opt to apply the terms of the ordinary GNU General Public
-License instead of this License to a given copy of the Library.  To do
-this, you must alter all the notices that refer to this License, so
-that they refer to the ordinary GNU General Public License, version 2,
-instead of to this License.  (If a newer version than version 2 of the
-ordinary GNU General Public License has appeared, then you can specify
-that version instead if you wish.)  Do not make any other change in
-these notices.
-
-  Once this change is made in a given copy, it is irreversible for
-that copy, so the ordinary GNU General Public License applies to all
-subsequent copies and derivative works made from that copy.
-
-  This option is useful when you wish to copy part of the code of
-the Library into a program that is not a library.
-
-  4. You may copy and distribute the Library (or a portion or
-derivative of it, under Section 2) in object code or executable form
-under the terms of Sections 1 and 2 above provided that you accompany
-it with the complete corresponding machine-readable source code, which
-must be distributed under the terms of Sections 1 and 2 above on a
-medium customarily used for software interchange.
-
-  If distribution of object code is made by offering access to copy
-from a designated place, then offering equivalent access to copy the
-source code from the same place satisfies the requirement to
-distribute the source code, even though third parties are not
-compelled to copy the source along with the object code.
-
-  5. A program that contains no derivative of any portion of the
-Library, but is designed to work with the Library by being compiled or
-linked with it, is called a "work that uses the Library".  Such a
-work, in isolation, is not a derivative work of the Library, and
-therefore falls outside the scope of this License.
-
-  However, linking a "work that uses the Library" with the Library
-creates an executable that is a derivative of the Library (because it
-contains portions of the Library), rather than a "work that uses the
-library".  The executable is therefore covered by this License.
-Section 6 states terms for distribution of such executables.
-
-  When a "work that uses the Library" uses material from a header file
-that is part of the Library, the object code for the work may be a
-derivative work of the Library even though the source code is not.
-Whether this is true is especially significant if the work can be
-linked without the Library, or if the work is itself a library.  The
-threshold for this to be true is not precisely defined by law.
-
-  If such an object file uses only numerical parameters, data
-structure layouts and accessors, and small macros and small inline
-functions (ten lines or less in length), then the use of the object
-file is unrestricted, regardless of whether it is legally a derivative
-work.  (Executables containing this object code plus portions of the
-Library will still fall under Section 6.)
-
-  Otherwise, if the work is a derivative of the Library, you may
-distribute the object code for the work under the terms of Section 6.
-Any executables containing that work also fall under Section 6,
-whether or not they are linked directly with the Library itself.
-
-  6. As an exception to the Sections above, you may also combine or
-link a "work that uses the Library" with the Library to produce a
-work containing portions of the Library, and distribute that work
-under terms of your choice, provided that the terms permit
-modification of the work for the customer's own use and reverse
-engineering for debugging such modifications.
-
-  You must give prominent notice with each copy of the work that the
-Library is used in it and that the Library and its use are covered by
-this License.  You must supply a copy of this License.  If the work
-during execution displays copyright notices, you must include the
-copyright notice for the Library among them, as well as a reference
-directing the user to the copy of this License.  Also, you must do one
-of these things:
-
-    a) Accompany the work with the complete corresponding
-    machine-readable source code for the Library including whatever
-    changes were used in the work (which must be distributed under
-    Sections 1 and 2 above); and, if the work is an executable linked
-    with the Library, with the complete machine-readable "work that
-    uses the Library", as object code and/or source code, so that the
-    user can modify the Library and then relink to produce a modified
-    executable containing the modified Library.  (It is understood
-    that the user who changes the contents of definitions files in the
-    Library will not necessarily be able to recompile the application
-    to use the modified definitions.)
-
-    b) Use a suitable shared library mechanism for linking with the
-    Library.  A suitable mechanism is one that (1) uses at run time a
-    copy of the library already present on the user's computer system,
-    rather than copying library functions into the executable, and (2)
-    will operate properly with a modified version of the library, if
-    the user installs one, as long as the modified version is
-    interface-compatible with the version that the work was made with.
-
-    c) Accompany the work with a written offer, valid for at
-    least three years, to give the same user the materials
-    specified in Subsection 6a, above, for a charge no more
-    than the cost of performing this distribution.
-
-    d) If distribution of the work is made by offering access to copy
-    from a designated place, offer equivalent access to copy the above
-    specified materials from the same place.
-
-    e) Verify that the user has already received a copy of these
-    materials or that you have already sent this user a copy.
-
-  For an executable, the required form of the "work that uses the
-Library" must include any data and utility programs needed for
-reproducing the executable from it.  However, as a special exception,
-the materials to be distributed need not include anything that is
-normally distributed (in either source or binary form) with the major
-components (compiler, kernel, and so on) of the operating system on
-which the executable runs, unless that component itself accompanies
-the executable.
-
-  It may happen that this requirement contradicts the license
-restrictions of other proprietary libraries that do not normally
-accompany the operating system.  Such a contradiction means you cannot
-use both them and the Library together in an executable that you
-distribute.
-
-  7. You may place library facilities that are a work based on the
-Library side-by-side in a single library together with other library
-facilities not covered by this License, and distribute such a combined
-library, provided that the separate distribution of the work based on
-the Library and of the other library facilities is otherwise
-permitted, and provided that you do these two things:
-
-    a) Accompany the combined library with a copy of the same work
-    based on the Library, uncombined with any other library
-    facilities.  This must be distributed under the terms of the
-    Sections above.
-
-    b) Give prominent notice with the combined library of the fact
-    that part of it is a work based on the Library, and explaining
-    where to find the accompanying uncombined form of the same work.
-
-  8. You may not copy, modify, sublicense, link with, or distribute
-the Library except as expressly provided under this License.  Any
-attempt otherwise to copy, modify, sublicense, link with, or
-distribute the Library is void, and will automatically terminate your
-rights under this License.  However, parties who have received copies,
-or rights, from you under this License will not have their licenses
-terminated so long as such parties remain in full compliance.
-
-  9. You are not required to accept this License, since you have not
-signed it.  However, nothing else grants you permission to modify or
-distribute the Library or its derivative works.  These actions are
-prohibited by law if you do not accept this License.  Therefore, by
-modifying or distributing the Library (or any work based on the
-Library), you indicate your acceptance of this License to do so, and
-all its terms and conditions for copying, distributing or modifying
-the Library or works based on it.
-
-  10. Each time you redistribute the Library (or any work based on the
-Library), the recipient automatically receives a license from the
-original licensor to copy, distribute, link with or modify the Library
-subject to these terms and conditions.  You may not impose any further
-restrictions on the recipients' exercise of the rights granted herein.
-You are not responsible for enforcing compliance by third parties with
-this License.
-
-  11. If, as a consequence of a court judgment or allegation of patent
-infringement or for any other reason (not limited to patent issues),
-conditions are imposed on you (whether by court order, agreement or
-otherwise) that contradict the conditions of this License, they do not
-excuse you from the conditions of this License.  If you cannot
-distribute so as to satisfy simultaneously your obligations under this
-License and any other pertinent obligations, then as a consequence you
-may not distribute the Library at all.  For example, if a patent
-license would not permit royalty-free redistribution of the Library by
-all those who receive copies directly or indirectly through you, then
-the only way you could satisfy both it and this License would be to
-refrain entirely from distribution of the Library.
-
-If any portion of this section is held invalid or unenforceable under any
-particular circumstance, the balance of the section is intended to apply,
-and the section as a whole is intended to apply in other circumstances.
-
-It is not the purpose of this section to induce you to infringe any
-patents or other property right claims or to contest validity of any
-such claims; this section has the sole purpose of protecting the
-integrity of the free software distribution system which is
-implemented by public license practices.  Many people have made
-generous contributions to the wide range of software distributed
-through that system in reliance on consistent application of that
-system; it is up to the author/donor to decide if he or she is willing
-to distribute software through any other system and a licensee cannot
-impose that choice.
-
-This section is intended to make thoroughly clear what is believed to
-be a consequence of the rest of this License.
-
-  12. If the distribution and/or use of the Library is restricted in
-certain countries either by patents or by copyrighted interfaces, the
-original copyright holder who places the Library under this License may add
-an explicit geographical distribution limitation excluding those countries,
-so that distribution is permitted only in or among countries not thus
-excluded.  In such case, this License incorporates the limitation as if
-written in the body of this License.
-
-  13. The Free Software Foundation may publish revised and/or new
-versions of the Lesser General Public License from time to time.
-Such new versions will be similar in spirit to the present version,
-but may differ in detail to address new problems or concerns.
-
-Each version is given a distinguishing version number.  If the Library
-specifies a version number of this License which applies to it and
-"any later version", you have the option of following the terms and
-conditions either of that version or of any later version published by
-the Free Software Foundation.  If the Library does not specify a
-license version number, you may choose any version ever published by
-the Free Software Foundation.
-
-  14. If you wish to incorporate parts of the Library into other free
-programs whose distribution conditions are incompatible with these,
-write to the author to ask for permission.  For software which is
-copyrighted by the Free Software Foundation, write to the Free
-Software Foundation; we sometimes make exceptions for this.  Our
-decision will be guided by the two goals of preserving the free status
-of all derivatives of our free software and of promoting the sharing
-and reuse of software generally.
-
-			    NO WARRANTY
-
-  15. BECAUSE THE LIBRARY IS LICENSED FREE OF CHARGE, THERE IS NO
-WARRANTY FOR THE LIBRARY, TO THE EXTENT PERMITTED BY APPLICABLE LAW.
-EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR
-OTHER PARTIES PROVIDE THE LIBRARY "AS IS" WITHOUT WARRANTY OF ANY
-KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE
-IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
-PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE
-LIBRARY IS WITH YOU.  SHOULD THE LIBRARY PROVE DEFECTIVE, YOU ASSUME
-THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
-
-  16. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN
-WRITING WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY
-AND/OR REDISTRIBUTE THE LIBRARY AS PERMITTED ABOVE, BE LIABLE TO YOU
-FOR DAMAGES, INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR
-CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OR INABILITY TO USE THE
-LIBRARY (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR DATA BEING
-RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD PARTIES OR A
-FAILURE OF THE LIBRARY TO OPERATE WITH ANY OTHER SOFTWARE), EVEN IF
-SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH
-DAMAGES.
-
-		     END OF TERMS AND CONDITIONS
-
-           How to Apply These Terms to Your New Libraries
-
-  If you develop a new library, and you want it to be of the greatest
-possible use to the public, we recommend making it free software that
-everyone can redistribute and change.  You can do so by permitting
-redistribution under these terms (or, alternatively, under the terms of the
-ordinary General Public License).
-
-  To apply these terms, attach the following notices to the library.  It is
-safest to attach them to the start of each source file to most effectively
-convey the exclusion of warranty; and each file should have at least the
-"copyright" line and a pointer to where the full notice is found.
-
-    <one line to give the library's name and a brief idea of what it does.>
-    Copyright (C) <year>  <name of author>
-
-    This library is free software; you can redistribute it and/or
-    modify it under the terms of the GNU Lesser General Public
-    License as published by the Free Software Foundation; either
-    version 2.1 of the License, or (at your option) any later version.
-
-    This library is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
-    Lesser General Public License for more details.
-
-    You should have received a copy of the GNU Lesser General Public
-    License along with this library; if not, write to the Free Software
-    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-
-Also add information on how to contact you by electronic and paper mail.
-
-You should also get your employer (if you work as a programmer) or your
-school, if any, to sign a "copyright disclaimer" for the library, if
-necessary.  Here is a sample; alter the names:
-
-  Yoyodyne, Inc., hereby disclaims all copyright interest in the
-  library `Frob' (a library for tweaking knobs) written by James Random Hacker.
-
-  <signature of Ty Coon>, 1 April 1990
-  Ty Coon, President of Vice
-
-That's all there is to it!
-
-
diff --git a/Makefile.in b/Makefile.in
index 8fd59babcd..fa1e19ef37 100644
--- a/Makefile.in
+++ b/Makefile.in
@@ -1,16 +1,37 @@
 SHELL = /bin/sh
+NULLCMD = @NULLCMD@
+n=$(NULLCMD)
+ECHO1 = $(V:1=$n)
+RUNCMD = $(SHELL)
+CDPATH = .
+CHDIR = @CHDIR@
+exec = exec
+NULL = /dev/null
+PATH_SEPARATOR = @PATH_SEPARATOR@
 
 #### Start of system configuration section. ####
 
 srcdir = @srcdir@
-VPATH = $(srcdir):$(srcdir)/missing
+top_srcdir = $(srcdir)
+hdrdir = $(srcdir)/include
+PLATFORM_DIR = @PLATFORM_DIR@
 
+CC_WRAPPER = @XCC_WRAPPER@
 CC = @CC@
-YACC = @YACC@
+CPP = @CPP@
+LD = @LD@
+YACC = bison
 PURIFY =
 AUTOCONF = autoconf
+ACLOCAL = aclocal
+CONFIGURE = @CONFIGURE@
 @SET_MAKE@
 MKFILES = @MAKEFILES@
+BASERUBY = @BASERUBY@
+HAVE_BASERUBY = @HAVE_BASERUBY@
+TEST_RUNNABLE = @TEST_RUNNABLE@
+CROSS_COMPILING = @CROSS_COMPILING@
+DOXYGEN = @DOXYGEN@
 
 prefix = @prefix@
 exec_prefix = @exec_prefix@
@@ -23,156 +44,590 @@ datadir = @datadir@
 arch = @arch@
 sitearch = @sitearch@
 sitedir = @sitedir@
+archlibdir = @archlibdir@
+includedir = @includedir@
+archincludedir = @archincludedir@
+rubylibprefix = @rubylibprefix@
+rubylibdir = @rubylibdir@
+rubyarchprefix = @rubyarchprefix@
+rubyarchdir = @rubyarchdir@
+rubyhdrdir = @rubyhdrdir@
+rubyarchhdrdir = @rubyarchhdrdir@
+ruby_version = @ruby_version@
+RUBY_VERSION_NAME = @RUBY_VERSION_NAME@
+UNIVERSAL_ARCHNAMES = @UNIVERSAL_ARCHNAMES@
 
 TESTUI = console
 TESTS =
-RDOCTARGET = @RDOCTARGET@
+INSTALLDOC = @INSTALLDOC@
+DOCTARGETS = @RDOCTARGET@ @CAPITARGET@
 
 EXTOUT = @EXTOUT@
-RIDATADIR = $(DESTDIR)$(datadir)/ri/$(MAJOR).$(MINOR)/system
+TIMESTAMPDIR  = $(EXTOUT)/.timestamp
+arch_hdrdir = $(EXTOUT)/include/$(arch)
+VPATH = $(arch_hdrdir)/ruby:$(hdrdir)/ruby:$(srcdir):$(srcdir)/missing
 
 empty =
+CC_VERSION = @CC_VERSION@
 OUTFLAG = @OUTFLAG@$(empty)
-CFLAGS = @CFLAGS@ @XCFLAGS@ @ARCH_FLAG@
-CPPFLAGS = -I. -I$(srcdir) @CPPFLAGS@
+COUTFLAG = @COUTFLAG@$(empty)
+CPPOUTFLAG = $(COUTFLAG)
+ARCH_FLAG = @ARCH_FLAG@
+CFLAGS_NO_ARCH = @CFLAGS@
+CFLAGS = $(CFLAGS_NO_ARCH) $(ARCH_FLAG)
+cflags = @cflags@
+optflags = @optflags@
+debugflags = @debugflags@
+warnflags = @warnflags@ @strict_warnflags@
+cppflags = @cppflags@
+MATHN = @MATHN@
+XCFLAGS = @XCFLAGS@ $(MATHN:yes=-DCANONICALIZATION_FOR_MATHN)
+USE_RUBYGEMS = @USE_RUBYGEMS@
+USE_RUBYGEMS_ = $(USE_RUBYGEMS:yes=)
+CPPFLAGS = @CPPFLAGS@ $(INCFLAGS) $(USE_RUBYGEMS_:no=-DDISABLE_RUBYGEMS=1)
+MJIT_SUPPORT = @MJIT_SUPPORT@
+MJIT_HEADER_FLAGS = @MJIT_HEADER_FLAGS@
+MJIT_HEADER_SUFFIX =
+MJIT_HEADER_ARCH =
+MJIT_HEADER_INSTALL_DIR = @MJIT_HEADER_INSTALL_DIR@
+MJIT_CC = @MJIT_CC@
+MJIT_CFLAGS = @MJIT_CFLAGS@
+MJIT_OPTFLAGS = @MJIT_OPTFLAGS@
+MJIT_DEBUGFLAGS = @MJIT_DEBUGFLAGS@
+MJIT_LDSHARED = @MJIT_LDSHARED@
+MJIT_DLDFLAGS = @DLDFLAGS@
+MJIT_HEADER   = rb_mjit_header.h
+MJIT_MIN_HEADER_NAME = rb_mjit_min_header-$(RUBY_PROGRAM_VERSION).h
+MJIT_MIN_HEADER = $(MJIT_HEADER_BUILD_DIR)/$(MJIT_MIN_HEADER_NAME)
+MJIT_HEADER_BUILD_DIR = $(EXTOUT)/include/$(arch)
 LDFLAGS = @STATIC@ $(CFLAGS) @LDFLAGS@
-EXTLDFLAGS = 
+EXTLDFLAGS = @EXTLDFLAGS@
 XLDFLAGS = @XLDFLAGS@ $(EXTLDFLAGS)
-EXTLIBS = 
+EXTLIBS =
 LIBS = @LIBS@ $(EXTLIBS)
 MISSING = @LIBOBJS@ @ALLOCA@
+ENABLE_SHARED = @ENABLE_SHARED@
 LDSHARED = @LIBRUBY_LDSHARED@
-DLDFLAGS = @LIBRUBY_DLDFLAGS@ $(EXTLDFLAGS) @ARCH_FLAG@
+DLDSHARED = @DLDSHARED@
+DLDFLAGS = @LIBRUBY_DLDFLAGS@ $(XLDFLAGS) $(ARCH_FLAG)
 SOLIBS = @SOLIBS@
+ENABLE_DEBUG_ENV = @ENABLE_DEBUG_ENV@
 MAINLIBS = @MAINLIBS@
-MINIOBJS = @MINIOBJS@
-
+ARCHMINIOBJS = @MINIOBJS@
+DLNOBJ = @DLNOBJ@
+ENCOBJS = @ENCOBJS@
+EXTOBJS = @EXTOBJS@
+BUILTIN_ENCOBJS = @BUILTIN_ENCOBJS@
+BUILTIN_TRANSSRCS = @BUILTIN_TRANSSRCS@
+BUILTIN_TRANSOBJS = @BUILTIN_TRANSOBJS@
+POSTLINK = @POSTLINK@
+
+RUBY_BASE_NAME=@RUBY_BASE_NAME@
+RUBY_PROGRAM_VERSION=@RUBY_PROGRAM_VERSION@
+RUBY_API_VERSION=@RUBY_API_VERSION@
 RUBY_INSTALL_NAME=@RUBY_INSTALL_NAME@
 RUBY_SO_NAME=@RUBY_SO_NAME@
 EXEEXT = @EXEEXT@
+LIBEXT = @LIBEXT@
 PROGRAM=$(RUBY_INSTALL_NAME)$(EXEEXT)
 RUBY = $(RUBY_INSTALL_NAME)
-MINIRUBY = @MINIRUBY@
-RUNRUBY = @RUNRUBY@
+MINIRUBY = @MINIRUBY@\
+	$(MINIRUBYOPT)
+# RUNRUBY_COMMAND:: runruby.rb or baseruby. do not append options directly
+RUNRUBY_COMMAND = @RUNRUBY_COMMAND@
+# RUNRUBY:: run ruby with RUN_OPTS which is passed to ruby
+RUNRUBY = @RUNRUBY@ $(RUN_OPTS)
+# RUNRUBY_DEBUGGER:: debugging option for runruby.rb
+RUNRUBY_DEBUGGER = --debugger='gdb -x run.gdb --quiet --args'
+XRUBY = @XRUBY@
+BTESTRUBY = @BTESTRUBY@\
+	$(MINIRUBYOPT)
+INSTALL = @INSTALL@
+INSTALL_DATA = @INSTALL_DATA@
+INSTALL_SCRIPT = @INSTALL_SCRIPT@
+INSTALL_PROGRAM = @INSTALL_PROGRAM@
+XRUBY_LIBDIR = @XRUBY_LIBDIR@
+XRUBY_RUBYLIBDIR = @XRUBY_RUBYLIBDIR@
+XRUBY_RUBYHDRDIR = @XRUBY_RUBYHDRDIR@
+BOOTSTRAPRUBY = @BOOTSTRAPRUBY@
+
+COROUTINE_H = @X_FIBER_COROUTINE_H@
+COROUTINE_OBJ = $(COROUTINE_H:.h=.@OBJEXT@)
 
 #### End of system configuration section. ####
 
 MAJOR=	@MAJOR@
 MINOR=	@MINOR@
 TEENY=	@TEENY@
+RUBY_PROGRAM_VERSION = @RUBY_PROGRAM_VERSION@
 
 LIBRUBY_A     = @LIBRUBY_A@
 LIBRUBY_SO    = @LIBRUBY_SO@
+LIBRUBY_SONAME= @LIBRUBY_SONAME@
 LIBRUBY_ALIASES= @LIBRUBY_ALIASES@
 LIBRUBY	      = @LIBRUBY@
 LIBRUBYARG    = @LIBRUBYARG@
 LIBRUBYARG_STATIC = @LIBRUBYARG_STATIC@
 LIBRUBYARG_SHARED = @LIBRUBYARG_SHARED@
+LIBRUBY_RELATIVE = @LIBRUBY_RELATIVE@
+LIBRUBY_A_OBJS = @LIBRUBY_A_OBJS@
+
+DTRACE_REBUILD_OBJS = $(DTRACE_REBUILD:yes=$(DTRACE_DEPENDENT_OBJS))
+
+DTRACE_DEPENDENT_OBJS = array.$(OBJEXT) \
+			eval.$(OBJEXT) \
+			gc.$(OBJEXT) \
+			hash.$(OBJEXT) \
+			load.$(OBJEXT) \
+			object.$(OBJEXT) \
+			parse.$(OBJEXT) \
+			string.$(OBJEXT) \
+			symbol.$(OBJEXT) \
+			vm.$(OBJEXT)
+
+THREAD_MODEL  = @THREAD_MODEL@
 
 PREP          = @PREP@
 ARCHFILE      = @ARCHFILE@
 SETUP         =
 EXTSTATIC     = @EXTSTATIC@
+ENCSTATIC     = @ENCSTATIC@
+SET_LC_MESSAGES = env LC_MESSAGES=C
 
+MAKEDIRS      = @MKDIR_P@
+CP            = cp
+MV            = mv
 RM            = rm -f
+RMDIR         = @RMDIR@
+RMDIRS        = @RMDIRS@
+RMALL         = @RMALL@
 NM            = @NM@
 AR            = @AR@
-ARFLAGS       = rcu
+ARFLAGS       = @ARFLAGS@$(empty)
 RANLIB        = @RANLIB@
 AS            = @AS@
-ASFLAGS       = @ASFLAGS@
+ASFLAGS       = @ASFLAGS@ $(INCFLAGS)
+IFCHANGE      = $(srcdir)/tool/ifchange
+OBJDUMP       = @OBJDUMP@
+OBJCOPY       = @OBJCOPY@
+HAVE_GIT      = @HAVE_GIT@
+GIT           = @GIT@
+VCS           = @VCS@
+VCSUP         = @VCSUP@
+DTRACE        = @DTRACE@ @DTRACE_OPT@
+DTRACE_EXT    = @DTRACE_EXT@
+DTRACE_OBJ    = @DTRACE_OBJ@
+DTRACE_REBUILD= @DTRACE_REBUILD@
+DTRACE_GLOMMED_OBJ = $(DTRACE_REBUILD:yes=ruby-glommed.$(OBJEXT))
 
 OBJEXT        = @OBJEXT@
+ASMEXT        = S
+SOEXT         = @SOEXT@
+DLEXT         = @DLEXT@
 MANTYPE	      = @MANTYPE@
+SYMBOL_PREFIX = @SYMBOL_PREFIX@
+
+INSTALLED_LIST= .installed.list
+
+NEWLINE_C     = enc/trans/newline.c
+MINIPRELUDE_C = miniprelude.c
+PRELUDE_C     = prelude.c
+GOLF_PRELUDE_C= golf_prelude.c
+RBCONFIG      = .rbconfig.time
 
-PREINSTALL    = @PREINSTALL@
+MAINSRC = $(MAINOBJ:@OBJEXT@=c)
+
+SRC_FILE      = $<
+OS_SRC_FILE   = $<
+DEST_FILE     = $@
+OS_DEST_FILE  = $@
+
+MESSAGE_BEGIN = @for line in
+MESSAGE_END = ; do echo "$$line"; done
+ECHO_BEGIN = @sep=''; for word in
+ECHO_END = ; do echo @ECHO_N@ "$$sep'$$word'@ECHO_C@"; sep=' '; done; echo
+
+DESTDIR       = @DESTDIR@
+
+configure_args = @configure_args@
 #### End of variables
 
-all:
+.SUFFIXES: .inc .h .c .y .i .$(ASMEXT) .$(DTRACE_EXT)
 
-.DEFAULT: all
+all:
 
 # Prevent GNU make v3 from overflowing arg limit on SysV.
 .NOEXPORT:
 
 miniruby$(EXEEXT):
-		@$(RM) $@
-		$(PURIFY) $(CC) $(MAINOBJ) $(MINIOBJS) $(LIBRUBY_A) $(LIBS) $(OUTFLAG)$@ $(LDFLAGS) $(MAINLIBS)
+		@-if test -f $@; then $(MV) -f $@ $@.old; $(RM) $@.old; fi
+		$(ECHO) linking $@
+		$(Q) $(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) $(NORMALMAINOBJ) $(MINIOBJS) $(COMMONOBJS) $(MAINLIBS) $(LIBS) $(OUTFLAG)$@
+		$(Q) $(POSTLINK)
 
 $(PROGRAM):
 		@$(RM) $@
-		$(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) $(MAINLIBS) $(MAINOBJ) $(EXTOBJS) $(LIBRUBYARG) $(LIBS) $(OUTFLAG)$@
+		$(ECHO) linking $@
+		$(Q) $(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) $(MAINOBJ) $(EXTOBJS) $(LIBRUBYARG) $(MAINLIBS) $(LIBS) $(EXTLIBS) $(OUTFLAG)$@
+		$(Q) $(POSTLINK)
+
+PRE_LIBRUBY_UPDATE = [ -n "$(LIBRUBY_SO_UPDATE)" ] || $(exec) $(RM) $(LIBRUBY_EXTS)
 
 # We must `rm' the library each time this rule is invoked because "updating" a
 # MAB library on Apple/NeXT (see --enable-fat-binary in configure) is not
 # supported.
 $(LIBRUBY_A):
 		@$(RM) $@
-		$(AR) $(ARFLAGS) $@ $(OBJS) $(DMYEXT)
+		@-[ -z "$(EXTSTATIC)" ] || $(PRE_LIBRUBY_UPDATE)
+		$(ECHO) linking static-library $@
+		$(Q) $(AR) $(ARFLAGS) $@ $(LIBRUBY_A_OBJS) $(INITOBJS)
 		@-$(RANLIB) $@ 2> /dev/null || true
 
+verify-static-library: $(LIBRUBY_A)
+		$(ECHO) verifying static-library $@
+		@$(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) $(MAINOBJ) $(LIBRUBY_A) $(MAINLIBS) $(EXTLIBS) $(LIBS) $(OUTFLAG)conftest$(EXEEXT)
+		@$(RMALL) conftest$(EXEEXT) conftest.c conftest.dSYM
+
 $(LIBRUBY_SO):
-		$(LDSHARED) $(DLDFLAGS) $(OBJS) $(DLDOBJS) $(SOLIBS) $(OUTFLAG)$@
-		@-$(MINIRUBY) -e 'ARGV.each{|link| File.delete link if File.exist? link; \
-						  File.symlink "$(LIBRUBY_SO)", link}' \
-				$(LIBRUBY_ALIASES) || true
-
-fake.rb:	Makefile
-		@echo ' \
-		class Object; \
-		  CROSS_COMPILING = RUBY_PLATFORM; \
-		  remove_const :RUBY_PLATFORM; \
-		  remove_const :RUBY_VERSION; \
-		  RUBY_PLATFORM = "@arch@"; \
-		  RUBY_VERSION = "@MAJOR@.@MINOR@.@TEENY@"; \
-		end; \
-		if RUBY_PLATFORM =~ /mswin|bccwin|mingw/; \
-		  class File; \
-		    remove_const :ALT_SEPARATOR; \
-		    ALT_SEPARATOR = "\\"; \
-		  end; \
-		end; \
-		' > $@
-
-Makefile:	$(srcdir)/Makefile.in
-
-$(MKFILES): config.status $(srcdir)/common.mk
-		MAKE=$(MAKE) $(SHELL) ./config.status
-		@{ \
+		@-[ -n "$(EXTSTATIC)" ] || $(PRE_LIBRUBY_UPDATE)
+		$(ECHO) linking shared-library $@
+		$(Q) $(LDSHARED) $(DLDFLAGS) $(OBJS) $(DLDOBJS) $(SOLIBS) $(EXTSOLIBS) $(OUTFLAG)$@
+		-$(Q) $(OBJCOPY) -w -L '$(SYMBOL_PREFIX)Init_*' -L '$(SYMBOL_PREFIX)ruby_static_id_*' \
+			-L '$(SYMBOL_PREFIX)*_threadptr_*' -L '$(SYMBOL_PREFIX)*_ec_*' $@
+		$(Q) $(POSTLINK)
+		@-$(MINIRUBY) -e 'so, *aliases = ARGV; aliases.uniq!; aliases.delete(File.basename(so)); \
+				  aliases.each { |link| File.delete link rescue nil; File.symlink so, link }' \
+				$(LIBRUBY_SO) $(LIBRUBY_ALIASES) || true
+
+LIBRUBY_WITH_EXT = @LIBRUBY_WITH_EXT@
+$(LIBRUBY_$(LIBRUBY_WITH_EXT)): $(LIBRUBY_SO_UPDATE)
+
+ruby_pc = @ruby_pc@
+ruby.pc: $(ruby_pc)
+$(ruby_pc): config.status
+		@./config.status --file=$@:$(srcdir)/template/ruby.pc.in
+
+ruby-runner.h: template/ruby-runner.h.in config.status
+		@./config.status --file=$@:$(srcdir)/template/$(@F).in
+
+$(RBCONFIG): $(PREP)
+
+rbconfig.rb: $(RBCONFIG)
+
+install-cross: $(arch)-fake.rb $(RBCONFIG) rbconfig.rb $(arch_hdrdir)/ruby/config.h \
+	$(LIBRUBY_A) $(LIBRUBY_SO) $(ARCHFILE)
+	$(ECHO) installing cross-compiling stuff
+	$(Q) $(MAKEDIRS) $(XRUBY_RUBYLIBDIR)/$(arch) $(XRUBY_RUBYHDRDIR)/$(arch)/ruby
+	$(Q) sed '/^\$$:\.unshift/q' $(arch)-fake.rb > fake.rb
+	$(Q) $(BASERUBY) -p \
+	-e '~/^\s*CONFIG\["LDFLAGS"\]/ and' \
+	-e '$$_[/(?=\s*"$$)/] = %q[ #{(CONFIG["LIBPATHFLAG"]%File.dirname(__FILE__)).strip}]' \
+	rbconfig.rb > fake-rbconfig.rb
+	$(INSTALL_SCRIPT) fake.rb $(XRUBY_RUBYLIBDIR)/$(arch)/fake.rb
+	$(INSTALL_SCRIPT) fake-rbconfig.rb $(XRUBY_RUBYLIBDIR)/$(arch)/rbconfig.rb
+	@$(RM) fake.rb fake-rbconfig.rb
+	$(INSTALL_DATA) $(arch_hdrdir)/ruby/config.h $(XRUBY_RUBYHDRDIR)/$(arch)/ruby
+	$(INSTALL_DATA) $(top_srcdir)/include/ruby/win32.h $(XRUBY_RUBYHDRDIR)/ruby
+	$(INSTALL_DATA) $(LIBRUBY) $(LIBRUBY_A) $(XRUBY_RUBYLIBDIR)/$(arch)
+	$(INSTALL_PROGRAM) $(LIBRUBY_SO) $(XRUBY_RUBYLIBDIR)/$(arch)
+
+Makefile:	$(srcdir)/Makefile.in $(srcdir)/enc/Makefile.in
+
+$(MKFILES): config.status $(srcdir)/version.h
+		@[ -f $@ ] && mv $@ $@.old
+		MAKE=$(MAKE) $(SHELL) ./config.status $@
+		@cmp $@ $@.old > /dev/null 2>&1 && echo $@ unchanged && exit 0; \
+		{ \
 		    echo "all:; -@rm -f conftest.mk"; \
 		    echo "conftest.mk: .force; @echo AUTO_REMAKE"; \
 		    echo ".force:"; \
 		} > conftest.mk || exit 1; \
 		$(MAKE) -f conftest.mk | grep '^AUTO_REMAKE$$' >/dev/null 2>&1 || \
-		{ echo "Makefile updated, restart."; exit 1; }
-
-config.status:	$(srcdir)/configure
-		MINIRUBY="$(MINIRUBY)" $(SHELL) ./config.status --recheck
-
-$(srcdir)/configure: $(srcdir)/configure.in
-		cd $(srcdir) && $(AUTOCONF)
-
-lex.c: keywords
-	@-$(RM) $@
-	gperf -p -j1 -i 1 -g -o -t -N rb_reserved_word -k1,3,$$ $? > $@ || \
-	cp "$(srcdir)/$@" .
-
-.y.c:
-	$(YACC) $<
-	sed '/^#/s|y\.tab\.c|$@|' y.tab.c > $@
-	rm -f y.tab.c
+		{ echo "$@ updated, restart."; exit 1; }
+
+uncommon.mk: $(srcdir)/common.mk
+		sed 's/{\$$([^(){}]*)[^{}]*}//g' $< > $@
+
+.PHONY: reconfig
+reconfig-args = $(srcdir)/$(CONFIGURE) $(configure_args)
+config.status-args = ./config.status --recheck
+reconfig-exec-0 = test -t 1 && { : $${CONFIGURE_TTY=yes}; export CONFIGURE_TTY; }; exec 3>&1; exit `exec 4>&1; { "$$@" 3>&- 4>&-; echo $$? 1>&4; } | fgrep -v '(cached)' 1>&3 3>&- 4>&-`
+reconfig-exec-1 = set -x; "$$@"
+
+reconfig config.status:	$(srcdir)/$(CONFIGURE) $(srcdir)/enc/Makefile.in \
+			$(srcdir)/include/ruby/version.h
+	@PWD= MINIRUBY="$(MINIRUBY)"; export MINIRUBY; \
+	warnflags="@warnflags@"; export warnflags; \
+	set $(SHELL) $($@-args); $(reconfig-exec-$(V))
+
+$(srcdir)/$(CONFIGURE): $(srcdir)/configure.ac $(srcdir)/aclocal.m4
+	$(CHDIR) $(srcdir) && exec $(AUTOCONF) -o $(@F)
+
+$(srcdir)/aclocal.m4:
+	$(CHDIR) $(srcdir) && \
+	type $(ACLOCAL) >/dev/null 2>&1 && exec $(ACLOCAL); \
+	touch $(@F)
+
+prereq: $(srcdir)/$(CONFIGURE)
+
+incs: id.h
+all-incs: probes.h
+
+# Things which should be considered:
+# * with gperf v.s. without gperf
+# * committers may have various versions of gperf
+# * ./configure v.s. ../ruby/configure
+# * GNU make v.s. HP-UX make	# HP-UX make invokes the action if lex.c and keywords has same mtime.
+# * svn checkout generate a file with mtime as current time
+# * ext4 and XFS has a mtime with fractional part
+lex.c: defs/keywords
+	@\
+	if cmp -s $(srcdir)/defs/lex.c.src $?; then \
+	  [ $(Q) ] && echo copying $@ || set -x; \
+	  $(CP) $(srcdir)/lex.c.blt $@; \
+	else \
+	  [ $(Q) ] && echo generating $@ || set -x; \
+	  gperf -C -P -p -j1 -i 1 -g -o -t -N rb_reserved_word -k1,3,$$ $? \
+	  | sed -f $(srcdir)/tool/gperf.sed \
+	  > $@.tmp && \
+	  $(MV) $@.tmp $@ && \
+	  $(CP) $? $(srcdir)/defs/lex.c.src && \
+	  $(CP) $@ $(srcdir)/lex.c.blt; \
+	fi
+
+JIS_PROPS_OPTIONS = -k1,3 -7 -c -j1 -i1 -t -C -P -t --ignore-case -H onig_jis_property_hash -Q onig_jis_property_pool -N onig_jis_property
+
+$(srcdir)/enc/jis/props.h: enc/jis/props.kwd
+	$(MAKEDIRS) $(@D)
+	@set +e; \
+	if cmp -s $(?:.kwd=.src) $?; then \
+	  set -x; \
+	  $(CP) $(?:.kwd=.h.blt) $@; \
+	else \
+	  set -x; \
+	  gperf $(JIS_PROPS_OPTIONS) $? | \
+	  sed -f $(srcdir)/tool/gperf.sed > $@ && \
+	  $(CP) $? $(?:.kwd=.src) && \
+	  $(CP) $@ $(?:.kwd=.h.blt); \
+	fi
 
 .c.@OBJEXT@:
-	$(CC) $(CFLAGS) $(CPPFLAGS) -c $<
-
-.s.@OBJEXT@:
-	$(AS) $(ASFLAGS) -o $@ $<
+	@$(ECHO) compiling $<
+	$(Q) $(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -c $<
+
+.$(ASMEXT).@OBJEXT@:
+	@$(ECHO) assembling $<
+	$(Q) $(CC) $(ASFLAGS) -o $@ -c $<
+
+.c.$(ASMEXT):
+	@$(ECHO) translating $<
+	$(Q) $(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -S $<
+
+.c.i:
+	@$(ECHO) preprocessing $<
+	$(Q) $(CPP) $(warnflags) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -E $< > $@
+
+.d.h:
+	@$(ECHO) translating probes $<
+	$(Q) $(DTRACE) -o $@.tmp -h -C $(INCFLAGS) -s $<
+	$(Q) sed -e 's/RUBY_/RUBY_DTRACE_/g' -e 's/PROBES_H_TMP/RUBY_PROBES_H/' -e 's/(char \*/(const char */g' -e 's/, char \*/, const char */g' $@.tmp > $@
+	$(Q) $(RM) $@.tmp
+
+.dmyh.h:
+	@$(ECHO) making dummy $(DEST_FILE)
+	$(Q)echo '#include "$(*F).dmyh"' > $@
+
+probes.stamp: $(DTRACE_REBUILD_OBJS)
+	$(Q) if test -f $@ -o -f probes.$(OBJEXT); then \
+	  $(RM) $(DTRACE_REBUILD_OBJS) $@; \
+	  $(ECHO0) "rebuilding objects which were modified by \"dtrace -G\""; \
+	  $(MAKE) $(DTRACE_REBUILD_OBJS); \
+	fi
+	$(Q) touch $@
+
+probes.$(OBJEXT): $(srcdir)/probes.d $(DTRACE_REBUILD:yes=probes.stamp)
+	@$(ECHO) processing probes in object files
+	$(Q) $(RM) $@
+	$(Q) $(DTRACE) -G -C $(INCFLAGS) -s $(srcdir)/probes.d -o $@ $(DTRACE_REBUILD_OBJS)
+
+main: mjit-headers
+yes-mjit-headers: $(MJIT_MIN_HEADER)
+clean-local::
+	$(Q)$(RM) $(MJIT_HEADER) $(MJIT_HEADER:.h=)$(MJIT_HEADER_SUFFIX:%=*).h
+	$(Q)$(RM) $(MJIT_MIN_HEADER) $(MJIT_MIN_HEADER:.h=)$(MJIT_HEADER_SUFFIX:%=*).h
+	$(Q)$(RM) $(MJIT_HEADER_INSTALL_DIR)/rb_mjit_min_header-*.h
+	$(Q)$(RM) $(TIMESTAMPDIR)/$(MJIT_HEADER:.h=)$(MJIT_HEADER_SUFFIX).time mjit_config.h
+	$(Q)$(RM) -r mjit_build_dir.*
+	-$(Q) $(RMDIRS) $(MJIT_HEADER_INSTALL_DIR) 2> $(NULL) || exit 0
+
+# DTrace static library hacks described here:
+# http://mail.opensolaris.org/pipermail/dtrace-discuss/2005-August/000207.html
+ruby-glommed.$(OBJEXT):
+	@$(ECHO) generating a glommed object with DTrace probes for static library
+	$(Q) $(LD) -r -o $@ $(OBJS)
 
 clean-local::
-	@$(RM) ext/extinit.c ext/extinit.$(OBJEXT)
+	$(Q)$(RM) ext/extinit.c ext/extinit.$(OBJEXT) ext/ripper/y.output \
+		enc/encinit.c enc/encinit.$(OBJEXT)
+	-$(Q)$(RM) $(pkgconfig_DATA)
+	-$(Q)$(RMALL) exe/ ruby-runner.$(OBJEXT) ruby-runner.h *.dSYM
 
 distclean-local::
-	@$(RM) ext/config.cache $(RBCONFIG)
+	$(Q)$(RM) ext/config.cache $(RBCONFIG) Doxyfile
+	-$(Q)$(RM) run.gdb
+	-$(Q)$(RM) $(INSTALLED_LIST) $(arch_hdrdir)/ruby/config.h verconf.h
+	-$(Q)$(RMDIRS) $(arch_hdrdir)/ruby 2> /dev/null || true
+
+ext/clean.sub gems/clean.sub:: ext/clean.mk
+ext/distclean.sub gems/distclean.sub:: ext/distclean.mk
+ext/realclean.sub gems/realclean.sub:: ext/realclean.mk
+
+ext/clean.mk ext/distclean.mk ext/realclean.mk::
+	-$(Q) if [ -f $(EXTS_MK) ]; then exec $(MAKE) -f $(EXTS_MK) $(@F:.mk=); fi
+
+ext/clean:: ext/clean.sub
+ext/distclean:: ext/distclean.sub
+ext/realclean:: ext/realclean.sub
+gems/clean:: gems/clean.sub
+gems/distclean:: gems/distclean.sub
+gems/realclean:: gems/realclean.sub
+
+ext/clean.sub ext/distclean.sub ext/realclean.sub \
+gems/clean.sub gems/distclean.sub gems/realclean.sub::
+	$(Q) set dummy `echo "${EXTS}" | tr , ' '`; shift; \
+	test "$$#" = 0 && set .; \
+	set dummy `\
+	    cd $(@D) 2>/dev/null && \
+	    find "$$@" \( -name Makefile -o -name exts.mk \) -print | \
+	    sed -n 's:^\./::;s:^:$(@D)/:;s:/[^/][^/]*$$::p' | sort -u; \
+	`; shift; \
+	for dir do \
+	    $(RM) "$$dir/exts.mk"; \
+	    if [ -f "$$dir/Makefile" ]; then \
+		echo $(@F:.sub=)ing "$$dir"; \
+		(cd "$$dir" && exec $(MAKE) $(mflags) $(@F:.sub=)); \
+	    fi; \
+	done || true
+
+ext/distclean ext/realclean gems/distclean gems/realclean::
+	$(Q) set dummy `echo "${EXTS}" | tr , ' '`; shift; \
+	test "$$#" = 0 && set .; \
+	cd $(@D) 2>/dev/null && \
+	find "$$@" -type d -empty -exec $(RMDIRS) {} + 2> /dev/null || true
+	$(Q) $(RMDIRS) $(@D) 2> /dev/null || true
+
+clean-enc distclean-enc realclean-enc:
+	@test -f "$(ENC_MK)" || exit 0; \
+	echo $(@:-enc=ing) encodings; \
+	exec $(MAKE) $(MAKE_ENC) $(@:-enc=)
 
 ext/extinit.$(OBJEXT): ext/extinit.c $(SETUP)
-	$(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(OUTFLAG)$@ -c ext/extinit.c
+	$(ECHO) compiling $@
+	$(Q) $(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -c ext/extinit.c
+
+enc/encinit.$(OBJEXT): enc/encinit.c $(SETUP)
+
+cont.$(OBJEXT): $(COROUTINE_H)
+
+test-bundled-gems-run:
+	$(Q) set -e; while read gem _; do \
+	  echo testing $$gem gem && \
+	  $(XRUBY) -C $(srcdir)/gems/src/$$gem -Ilib ../../../.bundle/bin/rake; \
+	done < $(srcdir)/gems/bundled_gems
+
+update-src::
+	@$(CHDIR) "$(srcdir)" && LC_TIME=C exec $(VCSUP)
+
+update-download:: update-config_files
+
+after-update:: prereq
+
+gcov:
+	$(Q) $(BASERUBY) $(srcdir)/tool/run-gcov.rb
+
+lcov:
+	$(Q) $(BASERUBY) $(srcdir)/tool/run-lcov.rb
+
+update-benchmark-driver:
+	$(Q) $(srcdir)/tool/git-refresh -C $(srcdir)/benchmark $(Q1:0=-q) \
+		--branch $(BENCHMARK_DRIVER_GIT_REF) \
+		$(BENCHMARK_DRIVER_GIT_URL) benchmark-driver $(GIT_OPTS)
+
+update-doclie:
+	$(Q) $(srcdir)/tool/git-refresh -C $(srcdir)/coverage $(Q1:0=-q) \
+		--branch $(DOCLIE_GIT_REF) \
+		$(DOCLIE_GIT_URL) doclie $(GIT_OPTS)
+
+update-simplecov-html:
+	$(Q) $(srcdir)/tool/git-refresh -C $(srcdir)/coverage $(Q1:0=-q) \
+		--branch $(SIMPLECOV_HTML_GIT_REF) \
+		$(SIMPLECOV_HTML_GIT_URL) simplecov-html $(GIT_OPTS)
+
+update-simplecov:
+	$(Q) $(srcdir)/tool/git-refresh -C $(srcdir)/coverage $(Q1:0=-q) \
+		--branch $(SIMPLECOV_GIT_REF)  \
+		$(SIMPLECOV_GIT_URL) simplecov $(GIT_OPTS)
+
+update-coverage: update-simplecov update-simplecov-html update-doclie
+
+update-known-errors:
+	errno --list | cut -d' ' -f1 | sort -u - $(srcdir)/defs/known_errors.def | \
+	$(IFCHANGE) $(srcdir)/defs/known_errors.def -
+
+INSNS	= opt_sc.inc optinsn.inc optunifs.inc insns.inc insns_info.inc \
+	  vmtc.inc vm.inc mjit_compile.inc
+
+$(INSNS): $(srcdir)/insns.def vm_opts.h \
+	  $(srcdir)/defs/opt_operand.def $(srcdir)/defs/opt_insn_unif.def \
+	  $(srcdir)/tool/insns2vm.rb
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) -Ku $(srcdir)/tool/insns2vm.rb $(INSNS2VMOPT) $@
+
+verconf.h: $(RBCONFIG)
+
+loadpath: verconf.h
+	@$(CPP) $(XCFLAGS) $(CPPFLAGS) $(srcdir)/loadpath.c | \
+	sed -e '1,/^const char ruby_initial_load_paths/d;/;/,$$d' \
+	    -e '/^ /!d;s/ *"\\0"$$//;s/" *"//g'
+
+un-runnable:
+	$(ECHO) cannot make runnable, configure with --enable-load-relative.
+	$(Q) exit 1
+
+mjit_config.h:
+	$(ECHO) making $@
+	@{ \
+	. $(srcdir)/tool/mjit_archflag.sh; \
+	parse_arch_flags "$(UNIVERSAL_ARCHNAMES)" $(ARCH_FLAG); \
+	test "$(Q)" = @ || set -x; \
+	echo '#ifndef RUBY_MJIT_CONFIG_H'; \
+	echo '#define RUBY_MJIT_CONFIG_H 1'; \
+	echo; \
+	sep=; \
+	echo '#ifdef LOAD_RELATIVE'; \
+	quote MJIT_HEADER_INSTALL_DIR "/$(MJIT_HEADER_INSTALL_DIR)"; \
+	echo '#else'; \
+	quote MJIT_HEADER_INSTALL_DIR "$(rubyarchhdrdir)"; \
+	echo '#endif'; \
+	quote MJIT_MIN_HEADER_NAME "$(MJIT_MIN_HEADER_NAME)"; \
+	sep=,; \
+	quote "MJIT_CC_COMMON  " $(MJIT_CC); \
+	quote "MJIT_CFLAGS      MJIT_ARCHFLAG" $(MJIT_CFLAGS); \
+	quote "MJIT_OPTFLAGS   " $(MJIT_OPTFLAGS); \
+	quote "MJIT_DEBUGFLAGS " $(MJIT_DEBUGFLAGS); \
+	quote "MJIT_LDSHARED   " $(MJIT_LDSHARED); \
+	quote "MJIT_DLDFLAGS    MJIT_ARCHFLAG" $(MJIT_DLDFLAGS); \
+	quote "MJIT_LIBS       " $(LIBRUBYARG_SHARED); \
+	quote 'PRELOADENV       "@PRELOADENV@"'; \
+	indent=$${archs:+'  '}; \
+	define_arch_flags; \
+	echo; \
+	echo '#endif /* RUBY_MJIT_CONFIG_H */'; \
+	} > $@
+
+yes-test-almost yes-test-all: mjit_build_dir.$(SOEXT)
+mjit_build_dir.$(SOEXT): $(MJIT_MIN_HEADER) $(srcdir)/ruby-runner.c ruby-runner.h
+	$(ECHO) making $@
+	$(Q) $(DLDSHARED) $(MJIT_DLDFLAGS) $(ARCH_FLAG) $(CFLAGS) $(CPPFLAGS) \
+		-DMAKE_MJIT_BUILD_DIR=1 -DMJIT_MIN_HEADER='"$(MJIT_MIN_HEADER)"' \
+		$(OUTFLAG)$@ $(srcdir)/ruby-runner.c
diff --git a/NEWS b/NEWS
new file mode 100644
index 0000000000..385c12bb70
--- /dev/null
+++ b/NEWS
@@ -0,0 +1,612 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 2.6.0
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or reference
+information is supplied with.  For a full list of changes with all
+sufficient information, see the ChangeLog file or Redmine
+(e.g. <tt>https://bugs.ruby-lang.org/issues/$FEATURE_OR_BUG_NUMBER</tt>)
+
+== Changes since the 2.5.0 release
+
+=== Language changes
+
+* <code>$SAFE</code> is a process global state and we can set 0 again. [Feature #14250]
+
+* refinements take place at block passing.  [Feature #14223]
+
+* refinements take place at Kernel#public_send.  [Feature #15326]
+
+* refinements take place at Kernel#respond_to?.  [Feature #15327]
+
+* +else+ without +rescue+ now causes a syntax error.  [EXPERIMENTAL] [Feature #14606]
+
+* constant names may start with a non-ASCII capital letter. [Feature #13770]
+
+* An endless range is introduced.  You can write a range that has no end,
+  like <code>(0..)</code> (or similarly <code>(0...)</code>).
+  The following shows typical use cases.  [Feature #12912]
+
+    ary[1..]                              # identical to ary[1..-1]
+    (1...).each {|index| block }          # infinite loop from index 1
+    ary.zip(1..) {|elem, index| block }   # ary.each.with_index(1) { }
+
+* Non-Symbol key in keyword arguments hash causes an exception.
+
+* "shadowing outer local variable" warning was removed.  [Feature #12490]
+  You can now write the following without warning.
+
+    user = users.find {|user| cond(user) }
+
+* Print exception backtrace and error message in reverse order when the
+  exception is not caught and STDOUT is unchanged and a tty. [Feature #8661]
+
+* Print +cause+ of the exception if the exception is not caught and printed
+  its backtraces and error message [Feature #8257]
+
+* Flip-flop syntax is deprecated. [Feature #5400]
+
+=== Core classes updates (outstanding ones only)
+
+[Array]
+
+  [New methods]
+
+    * Added Array#union and Array#difference instance methods.  [Feature #14097]
+
+  [Modified methods]
+
+    * Array#to_h now maps elements to new keys and values by the
+      block if given.  [Feature #15143]
+
+  [Aliased methods]
+
+    * Array#filter is a new alias for Array#select. [Feature #13784]
+    * Array#filter! is a new alias for Array#select!. [Feature #13784]
+
+[Binding]
+
+  [New methods]
+
+    * added Binding#source_location.  [Feature #14230]
+
+      This method returns the source location of the binding, a 2-element
+      array of <code>__FILE__</code> and <code>__LINE__</code>.
+      Traditionally, the same information could be retrieved by
+      <code>eval("[__FILE__, __LINE__]", binding)</code>, but we are
+      planning to change this behavior so that Kernel#eval ignores
+      binding's source location [Bug #4352].  So, users should use this
+      newly-introduced method instead of Kernel#eval.
+
+[Dir]
+
+  [New methods]
+
+    * added Dir#each_child and Dir#children instance methods. [Feature #13969]
+
+[Enumerable]
+
+  [New methods]
+
+    * Enumerable#chain returns an enumerator object that iterates over the
+      elements of the receiver and then those of each argument
+      in sequence.  [Feature #15144]
+
+  [Modified methods]
+
+    * Enumerable#to_h now maps elements to new keys and values by the block if
+      given.  [Feature #15143]
+
+  [Aliased methods]
+
+    * Enumerable#filter is a new alias for Enumerable#select. [Feature #13784]
+
+[Enumerator::ArithmeticSequence]
+
+  * This is a new class to represent a generator of an arithmetic sequence,
+    that is a number sequence defined by a common difference. It can be used
+    for representing what is similar to Python's slice. You can get an
+    instance of this class from Numeric#step and Range#step.
+
+[Enumerator::Chain]
+
+  * This is a new class to represent a chain of enumerables that works as a
+    single enumerator, generated by such methods as Enumerable#chain and
+    Enumerator#+.
+
+[Enumerator::Lazy]
+
+  [Aliased methods]
+
+    * Enumerator::Lazy#filter is a new alias for
+      Enumerator::Lazy#select.  [Feature #13784]
+
+[Enumerator]
+
+  [New methods]
+
+    * Enumerator#+ returns an enumerator object that iterates over the
+      elements of the receiver and then those of the other operand.  [Feature #15144]
+
+[ENV]
+
+  [Modified methods]
+
+    * ENV.to_h now maps names and values to new keys and values
+      by the block if given.  [Feature #15143]
+
+[Exception]
+
+  [New options]
+
+    * Exception#full_message takes +:highlight+ and +:order+
+      options. [Bug #14324]
+
+[Hash]
+
+  [Modified methods]
+
+    * Hash#merge, Hash#merge!, and Hash#update now accept multiple
+      arguments.  [Feature #15111]
+
+    * Hash#to_h now maps keys and values to new keys and values
+      by the block if given.  [Feature #15143]
+
+  [Aliased methods]
+
+    * Hash#filter is a new alias for Hash#select.  [Feature #13784]
+
+    * Hash#filter! is a new alias for Hash#select!. [Feature #13784]
+
+[IO]
+
+  [New options]
+
+    * new mode character <code>'x'</code> to open files for exclusive
+      access. [Feature #11258]
+
+[Kernel]
+
+  [Aliased methods]
+
+    * Kernel#then is a new alias for Kernel#yield_self. [Feature #14594]
+
+  [New options]
+
+    * Kernel#Complex, Kernel#Float, Kernel#Integer, and
+      Kernel#Rational take +:exception+ option to specify the way of
+      error handling.  [Feature #12732]
+
+    * Kernel#system takes +:exception+ option to raise an exception
+      on failure.  [Feature #14386]
+
+  [Incompatible changes]
+
+    * Kernel#system and Kernel#exec do not close non-standard file descriptors
+      (The default of the +:close_others+ option is changed to +false+,
+      but we still set the +FD_CLOEXEC+ flag on descriptors we
+      create).  [Misc #14907]
+
+[KeyError]
+
+  [New options]
+
+    * KeyError.new accepts +:receiver+ and +:key+ options to set receiver and
+      key in Ruby code.  [Feature #14313]
+
+[Method]
+
+  [New methods]
+
+    * added Method#<< and Method#>> for Proc composition.  [Feature #6284]
+
+[Module]
+
+  [New methods]
+
+    * Module#method_defined?, Module#private_method_defined?, and
+      Module#protected_method_defined? now accept the second
+      parameter as optional. If it's +true+ (=default), checks ancestor
+      modules/classes, or checks only the class itself. [Feature #14944]
+
+[NameError]
+
+  [New options]
+
+    * NameError.new accepts +:receiver+ option to set receiver in Ruby
+      code.  [Feature #14313]
+
+[NilClass]
+
+  [New methods]
+
+    * NilClass#=~ is added for compatibility.  [Feature #15231]
+
+[NoMethodError]
+
+  [New options]
+
+    * NoMethodError.new accepts +:receiver+ option to set receiver in Ruby
+      code.  [Feature #14313]
+
+[Numeric]
+
+  [Incompatible changes]
+
+    * Numeric#step now returns an instance of Enumerator::ArithmeticSequence
+      class rather than one of Enumerator class.
+
+[Proc]
+
+  [New methods]
+
+    * added Proc#<< and Proc#>> for Proc composition.  [Feature #6284]
+
+  [Incompatible changes]
+
+    * Proc#call doesn't change <code>$SAFE</code> any more.  [Feature #14250]
+
+[Random]
+
+  [New methods]
+
+    * added Random.bytes.  [Feature #4938]
+
+[Range]
+
+  [New methods]
+
+    * Added Range#% instance method.  [Feature #14697]
+
+  [Incompatible changes]
+
+    * Range#=== now uses +#cover?+ instead of +#include?+ method. [Feature #14575]
+    * Range#cover? now accepts Range object. [Feature #14473]
+    * Range#step now returns an instance of Enumerator::ArithmeticSequence
+      class rather than one of Enumerator class.
+
+[Regexp/String]
+
+    * Update Unicode version from 10.0.0 to 11.0.0 [Feature #14802]
+      This includes a rewrite of the grapheme cluster (/\X/) algorithm
+      and special-casing for Georgian MTAVRULI on String#downcase.
+
+    * Update Emoji version from 5.0 to 11.0.0 [Feature #14802]
+
+[RubyVM::AbstractSyntaxTree]
+
+  [New methods]
+
+    * RubyVM::AbstractSyntaxTree.parse parses a given string and returns AST
+      nodes. [experimental]
+
+    * RubyVM::AbstractSyntaxTree.parse_file parses a given file and returns AST
+      nodes.  [experimental]
+
+    * RubyVM::AbstractSyntaxTree.of returns AST nodes of the given proc or method.
+      [experimental]
+
+[RubyVM]
+
+  [New methods]
+
+    * RubyVM.resolve_feature_path identifies the file that will be loaded by
+      "require(feature)". [experimental] [Feature #15230]
+
+[String]
+
+  * String#crypt is now deprecated. [Feature #14915]
+
+  [New features]
+
+    * String#split yields each substring to the block if given. [Feature #4780]
+
+[Struct]
+
+  [Modified methods]
+
+    * Struct#to_h now maps keys and values to new keys and values
+      by the block if given.  [Feature #15143]
+
+  [Aliased method]
+
+    * Struct#filter is a new alias for Struct#select. [Feature #13784]
+
+[Time]
+
+  [New features]
+
+    * Time.new and Time#getlocal accept a timezone object as well as
+      an UTC offset string. Time#+, Time#- and Time#succ also preserve
+      the timezone.  [Feature #14850]
+
+[TracePoint]
+
+  [New features]
+
+    * "script_compiled" event is supported. [Feature #15287]
+
+  [New methods]
+
+    * TracePoint#parameters [Feature #14694]
+
+    * TracePoint#instruction_sequence [Feature #15287]
+
+    * TracePoint#eval_script [Feature #15287]
+
+  [Modified methods]
+
+    * TracePoint#enable accepts new keywords "target:" and "target_line:".
+      [Feature #15289]
+
+=== Stdlib updates (outstanding ones only)
+
+[BigDecimal]
+
+  Update to the version 1.4.0.  This version includes several compatibility
+  issues, see Compatibility issues section below for the detail.
+
+  [Modified methods]
+
+  * BigDecimal() accepts new keyword "exception:" similar to Float().
+
+[Bundler]
+
+  * Add Bundler to Standard Library. [Feature #12733]
+
+  * Use 1.17.2. It's the latest stable version.
+
+[Coverage]
+
+  A oneshot_lines mode is added.  [Feature #15022]
+  This mode checks "whether each line was executed at least once or not",
+  instead of "how many times each line was executed".
+  A hook for each line is fired at most once, and after it is fired
+  the hook flag is removed, i.e., it runs with zero overhead.
+
+  [New options]
+
+    * Add +:oneshot_lines+ keyword argument to Coverage.start.
+
+    * Add +:stop+ and +:clear+ keyword arguments to Coverage.result.
+      If +clear+ is true, it clears the counters to zero.
+      If +stop+ is true, it disables coverage measurement.
+
+  [New methods]
+
+    * Coverage.line_stub, which is a simple helper function that
+      creates the "stub" of line coverage from a given source code.
+
+[ERB]
+
+  [New options]
+
+    * Add +:trim_mode+ and +:eoutvar+ keyword arguments to ERB.new.
+      Now non-keyword arguments other than first one are softly deprecated
+      and will be removed when Ruby 2.5 becomes EOL. [Feature #14256]
+
+    * erb command's <tt>-S</tt> option is deprecated, which will be removed
+      in the next version.
+
+[FileUtils]
+
+  [New methods]
+
+    * FileUtils#cp_lr.  [Feature #4189]
+
+[Matrix]
+
+  [New methods]
+
+    * Matrix#antisymmetric?, Matrix#skew_symmetric?
+
+    * Matrix#map!, Matrix#collect! [Feature #14151]
+
+    * Matrix#[]=
+
+    * Vector#map!, Vector#collect!
+
+    * Vector#[]=
+
+[Net]
+
+  [New options]
+
+    * Add +:write_timeout+ keyword argument to Net::HTTP.new. [Feature #13396]
+
+  [New methods]
+
+    * Add Net::HTTP#write_timeout and Net::HTTP#write_timeout=.  [Feature #13396]
+
+  [New constant]
+
+    * Add Net::HTTPClientException to deprecate Net::HTTPServerException,
+      whose name is misleading.  [Bug #14688]
+
+[NKF]
+
+  * Upgrade to nkf v2.1.5
+
+[Psych]
+
+  * Upgrade to Psych 3.1.0.pre2
+
+[REXML]
+
+  [Improved some XPath implementations]
+
+    * <code>concat()</code> function: Stringify all arguments before concatenating
+
+    * <code>string()</code> function: Support context node
+
+    * <code>string()</code> function: Support processing instruction node
+
+    * Support <code>"*:#{ELEMENT_NAME}"</code> syntax in XPath 2.0
+
+  [Fixed some XPath implementations]
+
+    * <code>"//#{ELEMENT_NAME}[#{POSITION}]"</code> case
+
+    * <code>string()</code> function: Fix <code>function(document)</code>
+      returns nodes that are out of root elements.
+
+    * <code>"/ #{ELEMENT_NAME} "</code> case
+
+    * <code>"/ #{ELEMENT_NAME} [ #{PREDICATE} ]"</code> case
+
+    * <code>"/ #{AXIS}::#{ELEMENT_NAME}"</code> case
+
+    * <code>"#{N}-#{M}"</code> case: One or more white spaces were required
+      before <code>"-"</code>
+
+    * <code>"/child::node()"</code> case
+
+    * <code>"#{FUNCTION}()/#{PATH}"</code> case
+
+    * <code>"@#{ATTRIBUTE}/parent::"</code> case
+
+    * <code>"name(#{NODE_SET})"</code> case
+
+[RSS]
+
+  [New options]
+
+    * RSS::Parser.parse: Accept options as Hash. +:validate+ ,
+      +:ignore_unknown_element+ , +:parser_class+ options are available.
+
+[RubyGems]
+
+  * Upgrade to RubyGems 3.0.0.beta3
+
+[Set]
+
+  [Aliased methods]
+
+    * Set#filter! is a new alias for Set#select!.  [Feature #13784]
+
+[URI]
+
+  [New constant]
+
+    * Add URI::File to handle file URI scheme.  [Feature #14035]
+
+=== Compatibility issues (excluding feature bug fixes)
+
+[Dir]
+
+  * Dir.glob with <code>'\0'</code>-separated pattern list will be deprecated,
+    and is now warned.  [Feature #14643]
+
+[File]
+
+  * File.read, File.binread, File.write, File.binwrite, File.foreach, and
+    File.readlines do not invoke external commands even if the path starts
+    with the pipe character <code>'|'</code>. [Feature #14245]
+
+[Object]
+
+  * Object#=~ is deprecated.  [Feature #15231]
+
+=== Stdlib compatibility issues (excluding feature bug fixes)
+
+* These standard libraries have been promoted to default gems.
+
+  * e2mmap
+  * forwardable
+  * irb
+  * logger
+  * matrix
+  * mutex_m
+  * ostruct
+  * prime
+  * rexml
+  * rss
+  * shell
+  * sync
+  * thwait
+  * tracer
+
+[BigDecimal]
+
+  * The following methods are removed.
+
+    * BigDecimal.allocate
+    * BigDecimal.ver
+
+  * Every BigDecimal object is frozen. [Feature #13984]
+
+  * BigDecimal() parses the given string like Float().
+
+  * String#to_d parses the receiver string like String#to_f.
+
+  * BigDecimal.new will be removed on its version 1.5.
+
+[Pathname]
+
+  * Pathname#read, Pathname#binread, Pathname#write, Pathname#binwrite,
+    Pathname#each_line and Pathname#readlines do not invoke external
+    commands even if the path starts with the pipe character <code>'|'</code>.
+    This follows [Feature #14245].
+
+=== C API updates
+
+=== Implementation improvements
+
+* Speedup Proc#call because we don't need to care about <code>$SAFE</code>
+  any more. [Feature #14318]
+
+  With +lc_fizzbuzz+ benchmark which uses so many Proc#call we can
+  measure x1.4 improvements.  [Bug #10212]
+
+* Speedup block.call where +block+ is passed block parameter. [Feature #14330]
+
+  Ruby 2.5 improves block passing performance. [Feature #14045]
+
+  Additionally, Ruby 2.6 improves the performance of passed block calling.
+
+* Introduce an initial implementation of JIT (Just-in-time) compiler. [Feature #14235] [experimental]
+
+  * <tt>--jit</tt> command line option is added to enable JIT. <tt>--jit-verbose=1</tt>
+    is good for inspection.  See <tt>ruby --help</tt> for others.
+  * To generate machine code, this JIT compiler uses C compiler used for building
+    the interpreter. Currently GCC, Clang, and Microsoft Visual C++ are supported for it.
+  * <tt>--disable-mjit-support</tt> option is added to configure. This is added for JIT debugging,
+    but if you get an error on building a header file for JIT, you can use this option to skip
+    building it as a workaround.
+  * rb_waitpid reimplemented on Unix-like platforms to maintain
+    compatibility with processes created for JIT [Bug #14867]
+
+* VM generator script renewal; makes the generated VM more optimized. [GH-1779]
+
+* Thread cache enabled for pthreads platforms (for Thread.new and
+  Thread.start).  [Feature #14757]
+
+* timer thread is eliminated for platforms with POSIX timers [Misc #14937]
+
+* Transient Heap (theap) is supported. [Bug #14858] [Feature #14989]
+  theap is managed heap for short-living memory objects. For example,
+  making small and short-living Hash object is x2 faster. With rdoc benchmark,
+  we measured 6-7% performance improvement.
+
+* Native implementations (arm32, arm64, ppc64le, win32, win64, x86, amd64) of
+  coroutines to improve performance of Fiber significantly. [Feature #14739]
+
+=== Miscellaneous changes
+
+* On macOS, shared libraries no longer include a full version number of Ruby
+  in their names.  This eliminates the burden of each teeny upgrade on the
+  platform that users need to rebuild every extension library.
+
+  [Before]
+    * libruby.2.6.0.dylib
+    * libruby.2.6.dylib -> libruby.2.6.0.dylib
+    * libruby.dylib -> libruby.2.6.0.dylib
+
+  [After]
+    * libruby.2.6.dylib
+    * libruby.dylib -> libruby.2.6.dylib
+
+* Extracted misc/*.el files to https://github.com/ruby/elisp
diff --git a/README b/README
deleted file mode 100644
index cf836415ad..0000000000
--- a/README
+++ /dev/null
@@ -1,114 +0,0 @@
-* What's Ruby
-
-Ruby is the interpreted scripting language for quick and
-easy object-oriented programming.  It has many features to
-process text files and to do system management tasks (as in
-Perl).  It is simple, straight-forward, and extensible.
-
-
-* Features of Ruby
-
-  + Simple Syntax
-  + *Normal* Object-Oriented features(ex. class, method calls)
-  + *Advanced* Object-Oriented features(ex. Mix-in, Singleton-method)
-  + Operator Overloading
-  + Exception Handling
-  + Iterators and Closures
-  + Garbage Collection
-  + Dynamic Loading of Object files(on some architecture)
-  + Highly Portable(works on many UNIX machines, and on DOS,
-    Windows, Mac, BeOS etc.)
-
-
-* How to get Ruby
-
-The Ruby distribution files can be found in the following FTP site:
-
-  ftp://ftp.ruby-lang.org/pub/ruby/
-
-The latest source code of this version series can be checked out
-through SVN with the following command:
-
-  $ svn co http://svn.ruby-lang.org/repos/ruby/branches/ruby_1_8_6/
-
-The trunk of the Ruby source tree can be checked out with the
-following command:
-
-  $ svn co http://svn.ruby-lang.org/repos/ruby/trunk/ ruby
-
-There are some other branches under development.  Try the following
-command and see the list of branches:
-
-  $ svn ls http://svn.ruby-lang.org/repos/ruby/branches/
-
-
-* Ruby home-page
-
-The URL of the Ruby home-page is:
-
-   http://www.ruby-lang.org/
-
-
-* Mailing list
-
-There is a mailing list to talk about Ruby.
-To subscribe this list, please send the following phrase
-
-	subscribe YourFirstName YourFamilyName
-e.g.
-        subscribe Joseph Smith
-
-in the mail body (not subject) to the address <ruby-talk-ctl@ruby-lang.org>.
-
-
-* How to compile and install
-
-This is what you need to do to compile and install Ruby:
-
-  1. If ./configure does not exist or is older than configure.in,
-     run autoconf to (re)generate configure.
-
-  2. Run ./configure, which will generate config.h and Makefile.
-
-  3. Edit defines.h if you need.  Usually this step will not be needed.
-
-  4. Remove comment mark(#) before the module names from ext/Setup (or
-     add module names if not present), if you want to link modules
-     statically.
-
-     If you don't want to compile non static extension modules
-     (probably on architectures which does not allow dynamic loading),
-     remove comment mark from the line "#option nodynamic" in
-     ext/Setup.
-
-  5. Run make.
-
-  6. Optionally, run 'make test' to check whether the compiled Ruby
-     interpreter works well.  If you see the message "test succeeded",
-     your ruby works as it should (hopefully).
-
-  7. Run 'make install'
-
-     You may have to be a super user to install ruby.
-
-If you fail to compile ruby, please send the detailed error report with
-the error log and machine/OS type, to help others.
-
-
-* Copying
-
-See the file COPYING.
-
-
-* The Author
-
-Feel free to send comments and bug reports to the author.  Here is the 
-author's latest mail address:
-
-  matz@netlab.jp
-
--------------------------------------------------------
-created at: Thu Aug  3 11:57:36 JST 1995
-Local variables:
-mode: indented-text
-end:
diff --git a/README.EXT b/README.EXT
index e5d39911ca..48b8d964c4 100644
--- a/README.EXT
+++ b/README.EXT
@@ -1,1133 +1 @@
-.\" README.EXT -  -*- Text -*- created at: Mon Aug  7 16:45:54 JST 1995
-
-This document explains how to make extension libraries for Ruby.
-
-1. Basic knowledge
-
-In C, variables have types and data do not have types.  In contrast,
-Ruby variables do not have a static type, and data themselves have
-types, so data will need to be converted between the languages.
-
-Data in Ruby are represented by C type `VALUE'.  Each VALUE data has
-its data-type.
-
-To retrieve C data from a VALUE, you need to:
-
- (1) Identify the VALUE's data type
- (2) Convert the VALUE into C data
-
-Converting to the wrong data type may cause serious problems.
-
-
-1.1 Data-types
-
-The Ruby interpreter has the following data types:
-
-	T_NIL		nil
-	T_OBJECT	ordinary object
-	T_CLASS		class
-	T_MODULE	module
-	T_FLOAT		floating point number
-	T_STRING	string
-	T_REGEXP	regular expression
-	T_ARRAY		array
-	T_FIXNUM	Fixnum(31bit integer)
-	T_HASH		associative array
-	T_STRUCT	(Ruby) structure
-	T_BIGNUM	multi precision integer
-	T_FILE		IO
-	T_TRUE		true
-	T_FALSE		false
-	T_DATA		data
-	T_SYMBOL        symbol
-
-In addition, there are several other types used internally:
-
-	T_ICLASS
-	T_MATCH
-	T_UNDEF
-	T_VARMAP
-	T_SCOPE
-	T_NODE
-
-Most of the types are represented by C structures.
-
-1.2 Check Data Type of the VALUE
-
-The macro TYPE() defined in ruby.h shows the data type of the VALUE.
-TYPE() returns the constant number T_XXXX described above.  To handle
-data types, your code will look something like this:
-
-  switch (TYPE(obj)) {
-    case T_FIXNUM:
-      /* process Fixnum */
-      break;
-    case T_STRING:
-      /* process String */
-      break;
-    case T_ARRAY:
-      /* process Array */
-      break;
-    default:
-      /* raise exception */
-      rb_raise(rb_eTypeError, "not valid value");
-      break;
-  }
-
-There is the data-type check function
-
-  void Check_Type(VALUE value, int type)
-
-which raises an exception if the VALUE does not have the type specified.
-
-There are also faster check macros for fixnums and nil.
-
-  FIXNUM_P(obj)
-  NIL_P(obj)
-
-1.3 Convert VALUE into C data
-
-The data for type T_NIL, T_FALSE, T_TRUE are nil, true, false
-respectively.  They are singletons for the data type.
-
-The T_FIXNUM data is a 31bit length fixed integer (63bit length on
-some machines), which can be convert to a C integer by using the
-FIX2INT() macro.  There is also NUM2INT() which converts any Ruby
-numbers into C integers.  The NUM2INT() macro includes a type check, so
-an exception will be raised if the conversion failed.  NUM2DBL() can
-be used to retrieve the double float value in same way.
-
-To get char* from a VALUE, version 1.7 recommend to use new macros
-StringValue() and StringValuePtr().  StringValue(var) replaces var's
-value to the result of "var.to_str()".  StringValuePtr(var) does same
-replacement and returns char* representation of var.  These macros
-will skip the replacement if var is a String.  Notice that the macros
-requires to take only lvalue as their argument, to change the value
-of var in the replacement. 
-
-In version 1.6 or earlier, STR2CSTR() was used to do same thing
-but now it is obsoleted in version 1.7 because of STR2CSTR() has
-a risk of dangling pointer problem in to_str() impliclit conversion.
-
-Other data types have corresponding C structures, e.g. struct RArray
-for T_ARRAY etc. The VALUE of the type which has corresponding structure
-can be cast to retrieve the pointer to the struct.  The casting macro
-will be of the form RXXXX for each data type; for instance, RARRAY(obj). 
-See "ruby.h".
-
-For example, `RSTRING(str)->len' is the way to get the size of the
-Ruby String object.  The allocated region can be accessed by
-`RSTRING(str)->ptr'.  For arrays, use `RARRAY(ary)->len' and
-`RARRAY(ary)->ptr' respectively.
-
-Notice: Do not change the value of the structure directly, unless you
-are responsible for the result.  This ends up being the cause of interesting
-bugs.
-
-1.4 Convert C data into VALUE
-
-To convert C data to Ruby values:
-
-  * FIXNUM
-
-    left shift 1 bit, and turn on LSB.
-
-  * Other pointer values
-
-    cast to VALUE.
-
-You can determine whether a VALUE is pointer or not by checking its LSB.  
-
-Notice Ruby does not allow arbitrary pointer values to be a VALUE.  They
-should be pointers to the structures which Ruby knows about.  The known
-structures are defined in <ruby.h>.
-
-To convert C numbers to Ruby values, use these macros.
-
-  INT2FIX()	for integers within 31bits.
-  INT2NUM()	for arbitrary sized integer.
-
-INT2NUM() converts an integer into a Bignum if it is out of the FIXNUM
-range, but is a bit slower.
-
-1.5 Manipulating Ruby data
-
-As I already mentioned, it is not recommended to modify an object's internal
-structure.  To manipulate objects, use the functions supplied by the Ruby
-interpreter. Some (not all) of the useful functions are listed below:
-
- String functions
-
-  rb_str_new(const char *ptr, long len)
-
-    Creates a new Ruby string.
-
-  rb_str_new2(const char *ptr)
-
-    Creates a new Ruby string from a C string.  This is equivalent to
-    rb_str_new(ptr, strlen(ptr)).
-
-  rb_tainted_str_new(const char *ptr, long len)
-
-    Creates a new tainted Ruby string.  Strings from external data
-    sources should be tainted.
-
-  rb_tainted_str_new2(const char *ptr)
-
-    Creates a new tainted Ruby string from a C string.
-
-  rb_str_cat(VALUE str, const char *ptr, long len)
-
-    Appends len bytes of data from ptr to the Ruby string.
-
- Array functions
-
-  rb_ary_new()
-
-    Creates an array with no elements.
-
-  rb_ary_new2(long len)
-
-    Creates an array with no elements, allocating internal buffer
-    for len elements.
-
-  rb_ary_new3(long n, ...)
-
-    Creates an n-element array from the arguments.
-
-  rb_ary_new4(long n, VALUE *elts)
-
-    Creates an n-element array from a C array.
-
-  rb_ary_push(VALUE ary, VALUE val)
-  rb_ary_pop(VALUE ary)
-  rb_ary_shift(VALUE ary)
-  rb_ary_unshift(VALUE ary, VALUE val)
-
-    Array operations.  The first argument to each functions must be an 
-    array.  They may dump core if other types given.
-
-2. Extending Ruby with C
-
-2.1 Addding new features to Ruby
-
-You can add new features (classes, methods, etc.) to the Ruby
-interpreter.  Ruby provides APIs for defining the following things:
-
- * Classes, Modules
- * Methods, Singleton Methods
- * Constants
-
-2.1.1 Class/module definition
-
-To define a class or module, use the functions below:
-
-  VALUE rb_define_class(const char *name, VALUE super)
-  VALUE rb_define_module(const char *name)
-
-These functions return the newly created class or module.  You may
-want to save this reference into a variable to use later.
-
-To define nested classes or modules, use the functions below:
-
-  VALUE rb_define_class_under(VALUE outer, const char *name, VALUE super)
-  VALUE rb_define_module_under(VALUE outer, const char *name)
-
-2.1.2 Method/singleton method definition
-
-To define methods or singleton methods, use these functions:
-
-  void rb_define_method(VALUE klass, const char *name, 
-		        VALUE (*func)(), int argc)
-
-  void rb_define_singleton_method(VALUE object, const char *name, 
-			          VALUE (*func)(), int argc)
-
-The `argc' represents the number of the arguments to the C function,
-which must be less than 17.  But I believe you don't need that much. :-)
-
-If `argc' is negative, it specifies the calling sequence, not number of
-the arguments.  
-
-If argc is -1, the function will be called as:
-
-  VALUE func(int argc, VALUE *argv, VALUE obj)
-
-where argc is the actual number of arguments, argv is the C array of
-the arguments, and obj is the receiver.
-
-If argc is -2, the arguments are passed in a Ruby array. The function
-will be called like:
-
-  VALUE func(VALUE obj, VALUE args)
-
-where obj is the receiver, and args is the Ruby array containing
-actual arguments.
-
-There are two more functions to define methods.  One is to define
-private methods:
-
-  void rb_define_private_method(VALUE klass, const char *name, 
-			        VALUE (*func)(), int argc)
-
-The other is to define module functions, which are private AND singleton
-methods of the module.  For example, sqrt is the module function
-defined in Math module.  It can be call in the form like:
-
-  Math.sqrt(4)
-
-or
-
-  include Math
-  sqrt(4)
-
-To define module functions, use:
-
-  void rb_define_module_function(VALUE module, const char *name, 
-				 VALUE (*func)(), int argc)
-
-Oh, in addition, function-like methods, which are private methods defined
-in the Kernel module, can be defined using:
-
-  void rb_define_global_function(const char *name, VALUE (*func)(), int argc)
-
-To define alias to the method,
-
-  void rb_define_alias(VALUE module, const char* new, const char* old);
-
-To define and undefine the `allocate' class method,
-
-  void rb_define_alloc_func(VALUE klass, VALUE (*func)(VALUE klass));
-  void rb_undef_alloc_func(VALUE klass);
-
-func have to take the klass as the argument and return a newly
-allocated instance.  This instance should be empty as possible,
-without any expensive (including external) resources.
-
-2.1.3 Constant definition
-
-We have 2 functions to define constants:
-
-  void rb_define_const(VALUE klass, const char *name, VALUE val)
-  void rb_define_global_const(const char *name, VALUE val)
-
-The former is to define a constant under specified class/module.  The
-latter is to define a global constant.
-
-2.2 Use Ruby features from C
-
-There are several ways to invoke Ruby's features from C code.
-
-2.2.1 Evaluate Ruby Programs in a String
-
-The easiest way to use Ruby's functionality from a C program is to
-evaluate the string as Ruby program.  This function will do the job.
-
-  VALUE rb_eval_string(const char *str)
-
-Evaluation is done under the current context, thus current local variables
-of the innermost method (which is defined by Ruby) can be accessed.
-
-2.2.2 ID or Symbol
-
-You can invoke methods directly, without parsing the string.  First I need
-to explain about ID.  ID is the integer number to represent Ruby's
-identifiers such as variable names.  The Ruby data type corresponding to ID
-is Symbol.  It can be accessed from Ruby in the form:
-
- :Identifier
-
-You can get the ID value from a string within C code by using
-
-  rb_intern(const char *name)
-
-You can retrieve ID from Ruby object (Symbol or String) given as an
-argument by using
-
-  rb_to_id(VALUE symbol)
-
-You can convert C ID to Ruby Symbol by using
-
-  VALUE ID2SYM(ID id)
-
-and to convert Ruby Symbol object to ID, use
-
-  ID SYM2ID(VALUE symbol)
-
-2.2.3 Invoke Ruby method from C
-
-To invoke methods directly, you can use the function below
-
-  VALUE rb_funcall(VALUE recv, ID mid, int argc, ...)
-
-This function invokes a method on the recv, with the method name
-specified by the symbol mid.
-
-2.2.4 Accessing the variables and constants
-
-You can access class variables and instance variables using access
-functions.  Also, global variables can be shared between both environments.
-There's no way to access Ruby's local variables.
-
-The functions to access/modify instance variables are below:
-
-  VALUE rb_ivar_get(VALUE obj, ID id)
-  VALUE rb_ivar_set(VALUE obj, ID id, VALUE val)
-
-id must be the symbol, which can be retrieved by rb_intern().
-
-To access the constants of the class/module:
-
-  VALUE rb_const_get(VALUE obj, ID id)
-
-See 2.1.3 for defining new constant.
-
-3. Information sharing between Ruby and C
-
-3.1 Ruby constants that C can be accessed from C
-
-The following Ruby constants can be referred from C.
-
-  Qtrue
-  Qfalse
-
-Boolean values.  Qfalse is false in C also (i.e. 0).
-
-  Qnil
-
-Ruby nil in C scope.
-
-3.2 Global variables shared between C and Ruby
-
-Information can be shared between the two environments using shared global
-variables.  To define them, you can use functions listed below:
-
-  void rb_define_variable(const char *name, VALUE *var)
-
-This function defines the variable which is shared by both environments.
-The value of the global variable pointed to by `var' can be accessed
-through Ruby's global variable named `name'.
-
-You can define read-only (from Ruby, of course) variables using the
-function below.
-
-  void rb_define_readonly_variable(const char *name, VALUE *var)
-
-You can defined hooked variables.  The accessor functions (getter and
-setter) are called on access to the hooked variables.
-
-  void rb_define_hooked_variable(constchar *name, VALUE *var,
-				 VALUE (*getter)(), void (*setter)())
-
-If you need to supply either setter or getter, just supply 0 for the
-hook you don't need.  If both hooks are 0, rb_define_hooked_variable()
-works just like rb_define_variable().
-
-  void rb_define_virtual_variable(const char *name,
-				  VALUE (*getter)(), void (*setter)())
-
-This function defines a Ruby global variable without a corresponding C
-variable.  The value of the variable will be set/get only by hooks.
-
-The prototypes of the getter and setter functions are as follows:
-
-  (*getter)(ID id, void *data, struct global_entry* entry);
-  (*setter)(VALUE val, ID id, void *data, struct global_entry* entry);
-
-3.3 Encapsulate C data into Ruby object
-
-To wrap and objectify a C pointer as a Ruby object (so called
-DATA), use Data_Wrap_Struct().
-
-  Data_Wrap_Struct(klass, mark, free, ptr)
-
-Data_Wrap_Struct() returns a created DATA object.  The klass argument
-is the class for the DATA object.  The mark argument is the function
-to mark Ruby objects pointed by this data.  The free argument is the
-function to free the pointer allocation.  If this is -1, the pointer
-will be just freed.  The functions mark and free will be called from
-garbage collector.
-
-You can allocate and wrap the structure in one step.
-
-  Data_Make_Struct(klass, type, mark, free, sval)
-
-This macro returns an allocated Data object, wrapping the pointer to
-the structure, which is also allocated.  This macro works like:
-
-  (sval = ALLOC(type), Data_Wrap_Struct(klass, mark, free, sval))
-
-Arguments klass, mark, and free work like their counterparts in
-Data_Wrap_Struct().  A pointer to the allocated structure will be
-assigned to sval, which should be a pointer of the type specified.
-
-To retrieve the C pointer from the Data object, use the macro
-Data_Get_Struct().
-
-  Data_Get_Struct(obj, type, sval)
-
-A pointer to the structure will be assigned to the variable sval.
-
-See the example below for details. 
-
-4. Example - Creating dbm extension
-
-OK, here's the example of making an extension library.  This is the
-extension to access DBMs.  The full source is included in the ext/
-directory in the Ruby's source tree.
-
-(1) make the directory
-
-  % mkdir ext/dbm
-
-Make a directory for the extension library under ext directory.
-
-(2) design the library
-
-You need to design the library features, before making it.
-
-(3) write C code.
-
-You need to write C code for your extension library.  If your library
-has only one source file, choosing ``LIBRARY.c'' as a file name is
-preferred.  On the other hand, in case your library has multiple source
-files, avoid choosing ``LIBRARY.c'' for a file name.  It may conflict
-with an intermediate file ``LIBRARY.o'' on some platforms.
-
-Ruby will execute the initializing function named ``Init_LIBRARY'' in
-the library.  For example, ``Init_dbm()'' will be executed when loading
-the library.
-
-Here's the example of an initializing function.
-
---
-Init_dbm()
-{
-    /* define DBM class */
-    cDBM = rb_define_class("DBM", rb_cObject);
-    /* DBM includes Enumerate module */
-    rb_include_module(cDBM, rb_mEnumerable);
-
-    /* DBM has class method open(): arguments are received as C array */
-    rb_define_singleton_method(cDBM, "open", fdbm_s_open, -1);
-
-    /* DBM instance method close(): no args */
-    rb_define_method(cDBM, "close", fdbm_close, 0);
-    /* DBM instance method []: 1 argument */
-    rb_define_method(cDBM, "[]", fdbm_fetch, 1);
-		:
-
-    /* ID for a instance variable to store DBM data */
-    id_dbm = rb_intern("dbm");
-}
---
-
-The dbm extension wraps the dbm struct in the C environment using 
-Data_Make_Struct.
-
---
-struct dbmdata {
-    int  di_size;
-    DBM *di_dbm;
-};
-
-
-obj = Data_Make_Struct(klass, struct dbmdata, 0, free_dbm, dbmp);
---
-
-This code wraps the dbmdata structure into a Ruby object.  We avoid wrapping
-DBM* directly, because we want to cache size information.
-
-To retrieve the dbmdata structure from a Ruby object, we define the
-following macro:
-
---
-#define GetDBM(obj, dbmp) {\
-    Data_Get_Struct(obj, struct dbmdata, dbmp);\
-    if (dbmp->di_dbm == 0) closed_dbm();\
-}
---
-
-This sort of complicated macro does the retrieving and close checking for
-the DBM.
-
-There are three kinds of way to receive method arguments.  First,
-methods with a fixed number of arguments receive arguments like this:
-
---
-static VALUE
-fdbm_delete(obj, keystr)
-    VALUE obj, keystr;
-{
-	:
-}
---
-
-The first argument of the C function is the self, the rest are the
-arguments to the method.
-
-Second, methods with an arbitrary number of arguments receive
-arguments like this:
-
---
-static VALUE
-fdbm_s_open(argc, argv, klass)
-    int argc;
-    VALUE *argv;
-    VALUE klass;
-{
-	:
-    if (rb_scan_args(argc, argv, "11", &file, &vmode) == 1) {
-	mode = 0666;		/* default value */
-    }
-	:
-}
---
-
-The first argument is the number of method arguments, the second
-argument is the C array of the method arguments, and the third
-argument is the receiver of the method.
-
-You can use the function rb_scan_args() to check and retrieve the
-arguments.  For example, "11" means that the method requires at least one
-argument, and at most receives two arguments.
-
-Methods with an arbitrary number of arguments can receive arguments
-by Ruby's array, like this:
-
---
-static VALUE
-fdbm_indexes(obj, args)
-    VALUE obj, args;
-{
-	:
-}
---
-
-The first argument is the receiver, the second one is the Ruby array
-which contains the arguments to the method.
-
-** Notice
-
-GC should know about global variables which refer to Ruby's objects, but
-are not exported to the Ruby world.  You need to protect them by
-
-  void rb_global_variable(VALUE *var)
-
-(4) prepare extconf.rb
-
-If the file named extconf.rb exists, it will be executed to generate
-Makefile.
-
-extconf.rb is the file for check compilation conditions etc.  You
-need to put
-
-  require 'mkmf'
-
-at the top of the file.  You can use the functions below to check
-various conditions.
-
-  have_library(lib, func): check whether library containing function exists.
-  have_func(func, header): check whether function exists
-  have_header(header): check whether header file exists
-  create_makefile(target): generate Makefile
-
-The value of the variables below will affect the Makefile.
-
-  $CFLAGS: included in CFLAGS make variable (such as -O)
-  $CPPFLAGS: included in CPPFLAGS make variable (such as -I, -D)
-  $LDFLAGS: included in LDFLAGS make variable (such as -L)
-  $objs: list of object file names
-
-In normal, object files list is automatically generated by searching
-source files, but you need directs them explicitly if any sources will
-be generated while building.
-
-If a compilation condition is not fulfilled, you should not call
-``create_makefile''.  The Makefile will not generated, compilation will
-not be done.
-
-(5) prepare depend (optional)
-
-If the file named depend exists, Makefile will include that file to
-check dependencies.  You can make this file by invoking
-
-  % gcc -MM *.c > depend
-
-It's no harm.  Prepare it.
-
-(6) generate Makefile
-
-Try generating the Makefile by:
-
-  ruby extconf.rb
-
-You don't need this step if you put the extension library under the ext
-directory of the ruby source tree.  In that case, compilation of the
-interpreter will do this step for you.
-
-(7) make
-
-Type
-
-  make
-
-to compile your extension.  You don't need this step either if you have
-put extension library under the ext directory of the ruby source tree.
-
-(8) debug
-
-You may need to rb_debug the extension.  Extensions can be linked
-statically by the adding directory name in the ext/Setup file so that
-you can inspect the extension with the debugger.
-
-(9) done, now you have the extension library
-
-You can do anything you want with your library.  The author of Ruby
-will not claim any restrictions on your code depending on the Ruby API.
-Feel free to use, modify, distribute or sell your program.
-
-Appendix A. Ruby source files overview
-
-ruby language core
-
-  class.c
-  error.c
-  eval.c
-  gc.c
-  object.c
-  parse.y
-  variable.c
-
-utility functions
-
-  dln.c
-  regex.c
-  st.c
-  util.c
-
-ruby interpreter implementation
-
-  dmyext.c
-  inits.c
-  main.c
-  ruby.c
-  version.c
-
-class library
-
-  array.c
-  bignum.c
-  compar.c
-  dir.c
-  enum.c
-  file.c
-  hash.c
-  io.c
-  marshal.c
-  math.c
-  numeric.c
-  pack.c
-  prec.c
-  process.c
-  random.c
-  range.c
-  re.c
-  signal.c
-  sprintf.c
-  string.c
-  struct.c
-  time.c
-
-Appendix B. Ruby extension API reference
-
-** Types
-
- VALUE
-
-The type for the Ruby object.  Actual structures are defined in ruby.h,
-such as struct RString, etc.  To refer the values in structures, use
-casting macros like RSTRING(obj).
-
-** Variables and constants
-
- Qnil
-
-const: nil object
-
- Qtrue
-
-const: true object(default true value)
-
- Qfalse
-
-const: false object
-
-** C pointer wrapping
-
- Data_Wrap_Struct(VALUE klass, void (*mark)(), void (*free)(), void *sval)
-
-Wrap a C pointer into a Ruby object.  If object has references to other
-Ruby objects, they should be marked by using the mark function during
-the GC process.  Otherwise, mark should be 0.  When this object is no
-longer referred by anywhere, the pointer will be discarded by free
-function.
-
- Data_Make_Struct(klass, type, mark, free, sval)
-
-This macro allocates memory using malloc(), assigns it to the variable
-sval, and returns the DATA encapsulating the pointer to memory region.
-
- Data_Get_Struct(data, type, sval)
-
-This macro retrieves the pointer value from DATA, and assigns it to
-the variable sval. 
-
-** Checking data types
-
-TYPE(value)
-FIXNUM_P(value)
-NIL_P(value)
-void Check_Type(VALUE value, int type)
-void Check_SafeStr(VALUE value)
-
-** Data type conversion
-
-FIX2INT(value)
-INT2FIX(i)
-NUM2INT(value)
-INT2NUM(i)
-NUM2DBL(value)
-rb_float_new(f)
-StringValue(value)
-StringValuePtr(value)
-StringValueCStr(value)
-rb_str_new2(s)
-
-** defining class/module
-
- VALUE rb_define_class(const char *name, VALUE super)
-
-Defines a new Ruby class as a subclass of super.
-
- VALUE rb_define_class_under(VALUE module, const char *name, VALUE super)
-
-Creates a new Ruby class as a subclass of super, under the module's
-namespace.
-
- VALUE rb_define_module(const char *name)
-
-Defines a new Ruby module.
-
- VALUE rb_define_module_under(VALUE module, const char *name)
-
-Defines a new Ruby module under the module's namespace.
-
- void rb_include_module(VALUE klass, VALUE module)
-
-Includes module into class.  If class already includes it, just
-ignored.
-
- void rb_extend_object(VALUE object, VALUE module)
-
-Extend the object with the module's attributes.
-
-** Defining Global Variables
-
- void rb_define_variable(const char *name, VALUE *var)
-
-Defines a global variable which is shared between C and Ruby.  If name
-contains a character which is not allowed to be part of the symbol,
-it can't be seen from Ruby programs.
-
- void rb_define_readonly_variable(const char *name, VALUE *var)
-
-Defines a read-only global variable.  Works just like
-rb_define_variable(), except defined variable is read-only.
-
- void rb_define_virtual_variable(const char *name,
-				 VALUE (*getter)(), VALUE (*setter)())
-
-Defines a virtual variable, whose behavior is defined by a pair of C
-functions.  The getter function is called when the variable is
-referred.  The setter function is called when the value is set to the
-variable.  The prototype for getter/setter functions are:
-
-	VALUE getter(ID id)
-	void setter(VALUE val, ID id)
-
-The getter function must return the value for the access.
-
- void rb_define_hooked_variable(const char *name, VALUE *var,
-				VALUE (*getter)(), VALUE (*setter)())
-
-Defines hooked variable.  It's a virtual variable with a C variable.  
-The getter is called as
-
-	VALUE getter(ID id, VALUE *var)
-
-returning a new value.  The setter is called as
-
-	void setter(VALUE val, ID id, VALUE *var)
-
-GC requires C global variables which hold Ruby values to be marked.
-
- void rb_global_variable(VALUE *var)
-
-Tells GC to protect these variables.
-
-** Constant Definition
-
- void rb_define_const(VALUE klass, const char *name, VALUE val)
-
-Defines a new constant under the class/module.
-
- void rb_define_global_const(const char *name, VALUE val)
-
-Defines a global constant.  This is just the same as
-
-     rb_define_const(cKernal, name, val)
-
-** Method Definition
-
- rb_define_method(VALUE klass, const char *name, VALUE (*func)(), int argc)
-
-Defines a method for the class.  func is the function pointer.  argc
-is the number of arguments.  if argc is -1, the function will receive
-3 arguments: argc, argv, and self.  if argc is -2, the function will
-receive 2 arguments, self and args, where args is a Ruby array of
-the method arguments.
-
- rb_define_private_method(VALUE klass, const char *name, VALUE (*func)(), int argc)
-
-Defines a private method for the class.  Arguments are same as
-rb_define_method().
-
- rb_define_singleton_method(VALUE klass, const char *name, VALUE (*func)(), int argc)
-
-Defines a singleton method.  Arguments are same as rb_define_method().
-
- rb_scan_args(int argc, VALUE *argv, const char *fmt, ...)
-
-Retrieve argument from argc, argv.  The fmt is the format string for
-the arguments, such as "12" for 1 non-optional argument, 2 optional
-arguments.  If `*' appears at the end of fmt, it means the rest of
-the arguments are assigned to the corresponding variable, packed in
-an array.
-
-** Invoking Ruby method
-
- VALUE rb_funcall(VALUE recv, ID mid, int narg, ...)
-
-Invokes a method.  To retrieve mid from a method name, use rb_intern().
-
- VALUE rb_funcall2(VALUE recv, ID mid, int argc, VALUE *argv)
-
-Invokes a method, passing arguments by an array of values.
-
- VALUE rb_eval_string(const char *str)
-
-Compiles and executes the string as a Ruby program.
-
- ID rb_intern(const char *name)
-
-Returns ID corresponding to the name.
-
- char *rb_id2name(ID id)
-
-Returns the name corresponding ID.
-
- char *rb_class2name(VALUE klass)
-
-Returns the name of the class.
-
- int rb_respond_to(VALUE object, ID id)
-
-Returns true if the object responds to the message specified by id.
-
-** Instance Variables
-
- VALUE rb_iv_get(VALUE obj, const char *name)
-
-Retrieve the value of the instance variable.  If the name is not
-prefixed by `@', that variable shall be inaccessible from Ruby.
-
- VALUE rb_iv_set(VALUE obj, const char *name, VALUE val)
-
-Sets the value of the instance variable.
-
-** Control Structure
-
- VALUE rb_iterate(VALUE (*func1)(), void *arg1, VALUE (*func2)(), void *arg2)
-
-Calls the function func1, supplying func2 as the block.  func1 will be
-called with the argument arg1.  func2 receives the value from yield as
-the first argument, arg2 as the second argument.
- 
- VALUE rb_yield(VALUE val)
-
-Evaluates the block with value val.
-
- VALUE rb_rescue(VALUE (*func1)(), void *arg1, VALUE (*func2)(), void *arg2)
-
-Calls the function func1, with arg1 as the argument.  If an exception
-occurs during func1, it calls func2 with arg2 as the argument.  The
-return value of rb_rescue() is the return value from func1 if no
-exception occurs, from func2 otherwise.
-
- VALUE rb_ensure(VALUE (*func1)(), void *arg1, void (*func2)(), void *arg2)
-
-Calls the function func1 with arg1 as the argument, then calls func2
-with arg2 if execution terminated.  The return value from
-rb_ensure() is that of func1.
-
-** Exceptions and Errors
-
- void rb_warn(const char *fmt, ...)
-
-Prints a warning message according to a printf-like format.
-
- void rb_warning(const char *fmt, ...)
-
-Prints a warning message according to a printf-like format, if
-$VERBOSE is true.
-
-void rb_raise(rb_eRuntimeError, const char *fmt, ...)
-
-Raises RuntimeError.  The fmt is a format string just like printf().
-
- void rb_raise(VALUE exception, const char *fmt, ...)
-
-Raises a class exception.  The fmt is a format string just like printf().
-
- void rb_fatal(const char *fmt, ...)
-
-Raises a fatal error, terminates the interpreter.  No exception handling
-will be done for fatal errors, but ensure blocks will be executed.
-
- void rb_bug(const char *fmt, ...)
-
-Terminates the interpreter immediately.  This function should be
-called under the situation caused by the bug in the interpreter.  No
-exception handling nor ensure execution will be done.
-
-** Initialize and Starts the Interpreter
-
-The embedding API functions are below (not needed for extension libraries):
-
- void ruby_init()
-
-Initializes the interpreter.
-
- void ruby_options(int argc, char **argv)
-
-Process command line arguments for the interpreter.
-
- void ruby_run()
-
-Starts execution of the interpreter.
-
- void ruby_script(char *name)
-
-Specifies the name of the script ($0).
-
-Appendix C. Functions Available in extconf.rb
-
-These functions are available in extconf.rb:
-
- have_macro(macro, headers)
-
-Checks whether macro is defined with header.  Returns true if the macro
-is defined.
-
- have_library(lib, func)
-
-Checks whether the library exists, containing the specified function.
-Returns true if the library exists.
-
- find_library(lib, func, path...)
-
-Checks whether a library which contains the specified function exists in
-path.  Returns true if the library exists.
-
- have_func(func, header)
-
-Checks whether func exists with header.  Returns true if the function
-exists.  To check functions in an additional library, you need to
-check that library first using have_library().
-
- have_var(var, header)
-
-Checks whether var exists with header.  Returns true if the variable
-exists.  To check variables in an additional library, you need to
-check that library first using have_library().
-
- have_header(header)
-
-Checks whether header exists.  Returns true if the header file exists.
-
- find_header(header, path...)
-
-Checks whether header exists in path.  Returns true if the header file
-exists.
-
- have_struct_member(type, member, header)
-
-Checks whether type has member with header.  Returns true if the type
-is defined and has the member.
-
- have_type(type, header, opt)
-
-Checks whether type is defined with header.  Returns true if the type
-is defined.
-
- check_sizeof(type, header)
-
-Checks the size of type in char with header.  Returns the size if the
-type is defined, otherwise nil.
-
- create_makefile(target)
-
-Generates the Makefile for the extension library.  If you don't invoke
-this method, the compilation will not be done.
-
- find_executable(bin, path)
-
-Finds command in path, which is File::PATH_SEPARATOR-separated list of
-directories.  If path is nil or omitted, environment varialbe PATH
-will be used.  Returns the path name of the command if it is found,
-otherwise nil.
-
- with_config(withval[, default=nil])
-
-Parses the command line options and returns the value specified by
---with-<withval>.
-
- enable_config(config, *defaults)
- disable_config(config, *defaults)
-
-Parses the command line options for boolean.  Returns true if
---enable-<config> is given, or false if --disable-<config> is given.
-Otherwise, yields defaults to the given block and returns the result
-if it is called with a block, or returns defaults.
-
- dir_config(target[, default_dir])
- dir_config(target[, default_include, default_lib])
-
-Parses the command line options and adds the directories specified by
---with-<target>-dir, --with-<target>-include, and/or --with-<target>-lib
-to $CFLAGS and/or $LDFLAGS.  --with-<target>-dir=/path is equivalent to
---with-<target>-include=/path/include --with-<target>-lib=/path/lib.
-Returns an array of the added directories ([include_dir, lib_dir]).
-
- pkg_config(pkg)
-
-Obtains the information of pkg by pkg-config command.  The actual
-command name can be overriden by --with-pkg-config command line
-option.
-
-/*
- * Local variables:
- * fill-column: 70
- * end:
- */
+Moved to doc/extension.rdoc
diff --git a/README.EXT.ja b/README.EXT.ja
index 30c4d520ba..f884ecbb0e 100644
--- a/README.EXT.ja
+++ b/README.EXT.ja
@@ -1,1258 +1 @@
-.\" README.EXT.ja -  -*- Text -*- created at: Mon Aug  7 16:45:54 JST 1995
-
-Ruby�γ�ĥ�饤�֥��κ�������������ޤ���
-
-1�������μ�
-
-C���ѿ��ˤϷ������ꡤ�ǡ����ˤϷ�������ޤ��󡥤Ǥ����顤��
-�Ȥ��Хݥ��󥿤�int���ѿ�����������ȡ������ͤ������Ȥ��Ƽ�
-�갷���ޤ����դ�Ruby���ѿ��ˤϷ����ʤ����ǡ����˷��������
-�������ΰ㤤�Τ��ᡤC��Ruby����ߤ��Ѵ����ʤ���С����ߤ���
-�ǡ����򥢥������Ǥ��ޤ���
-
-Ruby�Υǡ�����VALUE�Ȥ���C�η���ɽ������ޤ���VALUE���Υǡ�
-���Ϥ��Υǡ��������פ�ʬ���ΤäƤ��ޤ������Υǡ��������פ�
-�����Τϥǡ���(���֥�������)�μºݤι�¤���̣���Ƥ��ơ�Ruby
-�Υ��饹�ȤϤޤ���ä���ΤǤ���
-
-VALUE����C�ˤȤäư�̣�Τ���ǡ�������Ф�����ˤ�
-
- (1) VALUE�Υǡ��������פ��Τ�
- (2) VALUE��C�Υǡ������Ѵ�����
-
-��ξ����ɬ�פǤ���(1)��˺���ȴְ�ä��ǡ������Ѵ����Ԥ��
-�ơ��ǰ��ץ�����बcore dump���ޤ���
-
-1.1 �ǡ���������
-
-Ruby�ˤϥ桼�����Ȥ���ǽ���Τ���ʲ��Υ����פ�����ޤ���
-
-	T_NIL		nil
-	T_OBJECT	�̾�Υ��֥�������
-	T_CLASS		���饹
-	T_MODULE	�⥸�塼��
-	T_FLOAT		��ư��������
-	T_STRING	ʸ����
-	T_REGEXP	����ɽ��
-	T_ARRAY		����
-	T_FIXNUM	Fixnum(31bitĹ����)
-	T_HASH		Ϣ������
-	T_STRUCT	(Ruby��)��¤��
-	T_BIGNUM	¿��Ĺ����
-	T_FILE		������
-	T_TRUE		��
-	T_FALSE		��
-	T_DATA		�ǡ���
-	T_SYMBOL	����ܥ�
-
-����¾�����������Ѥ���Ƥ���ʲ��Υ����פ�����ޤ���
-
-	T_ICLASS
-	T_MATCH
-	T_UNDEF
-	T_VARMAP
-	T_SCOPE
-	T_NODE
-
-�ۤȤ�ɤΥ����פ�C�ι�¤�ΤǼ�������Ƥ��ޤ���
-
-1.2 VALUE�Υǡ��������פ�����å�����
-
-ruby.h�Ǥ�TYPE()�Ȥ����ޥ������������Ƥ��ơ�VALUE�Υǡ���
-�����פ��Τ뤳�Ȥ�����ޤ���TYPE()�ޥ����Ͼ�ǾҲ𤷤�T_XXXX
-�η�����������֤��ޤ���VALUE�Υǡ��������פ˱����ƽ�������
-���ˤϡ�TYPE()���ͤ�ʬ�����뤳�Ȥˤʤ�ޤ���
-
-  switch (TYPE(obj)) {
-    case T_FIXNUM:
-      /* FIXNUM�ν��� */
-      break;
-    case T_STRING:
-      /* ʸ����ν��� */
-      break;
-    case T_ARRAY:
-      /* ����ν��� */
-      break;
-    default:
-      /* �㳰��ȯ�������� */
-      rb_raise(rb_eTypeError, "not valid value");
-      break;
-  }
-
-����ȥǡ��������פ�����å����ơ��������ʤ�����㳰��ȯ����
-��ؿ����Ѱդ���Ƥ��ޤ���
-
-  void Check_Type(VALUE value, int type)
-
-���δؿ���value��type��̵����С��㳰��ȯ�������ޤ���������
-����Ϳ����줿VALUE�Υǡ��������פ����������ɤ��������å���
-�뤿��ˤϡ����δؿ���Ȥ��ޤ���
-
-FIXNUM��NIL�˴ؤ��ƤϤ���®��Ƚ�̥ޥ������Ѱդ���Ƥ��ޤ���
-
-  FIXNUM_P(obj)
-  NIL_P(obj)
-
-1.3 VALUE��C�Υǡ������Ѵ�����
-
-�ǡ��������פ�T_NIL, T_FALSE, T_TRUE�Ǥ�������ǡ����Ϥ��줾
-��nil, false, true�Ǥ������Υǡ��������פΥ��֥������ȤϤҤ�
-�Ĥ��Ĥ���¸�ߤ��ޤ���
-
-�ǡ��������פ�T_FIXNUM�λ��������31bit�Υ����������������
-����FIXNUM��C���������Ѵ����뤿��ˤϥޥ�����FIX2INT()�פ��
-���ޤ������줫�顤FIXNUM�˸¤餺Ruby�Υǡ������������Ѵ�����
-��NUM2INT()�פȤ����ޥ���������ޤ������Υޥ����ϥǡ�������
-�פΥ����å�̵���ǻȤ��ޤ�(�������Ѵ��Ǥ��ʤ����ˤ��㳰��
-ȯ������)��Ʊ�ͤ˥����å�̵���ǻȤ����Ѵ��ޥ�����double��
-���Ф���NUM2DBL()�פ�����ޤ���
-
-char* ����Ф���硢version 1.6 �����Ǥϡ�STR2CSTR()�פ�
-�����ޥ�����ȤäƤ��ޤ������������ to_str() �ˤ����ۤ�
-���Ѵ���̤� GC ������ǽ�������뤿�ᡢversion 1.7 �ʹߤǤ�
-obsolete �Ȥʤꡢ����� StringValue() �� StringValuePtr()
-��Ȥ�����侩���Ƥ��ޤ���StringValue(var) �� var �� String
- �Ǥ���в��⤻���������Ǥʤ���� var �� var.to_str() �η�̤�
-�֤�������ޥ�����StringValuePtr(var) ��Ʊ�ͤ� var ���֤�����
-�Ƥ��� var ��ʸ����ɽ�����Ф��� char* ���֤��ޥ����Ǥ���var ��
-���Ƥ�ľ���֤����������������Τǡ�var �� lvalue �Ǥ���ɬ�פ�
-����ޤ���
-
-����ʳ��Υǡ��������פ��б�����C�ι�¤�Τ�����ޤ����б���
-�빽¤�ΤΤ���VALUE�Ϥ��Τޤޥ��㥹��(���Ѵ�)����й�¤�Τ�
-�ݥ��󥿤��Ѵ��Ǥ��ޤ���
-
-��¤�Τϡ�struct RXxxxx�פȤ���̾����ruby.h���������Ƥ���
-�����㤨��ʸ����ϡ�struct RString�פǤ����ºݤ˻Ȥ���ǽ����
-����Τ�ʸ��������󤯤餤���Ȼפ��ޤ���
-
-ruby.h�ǤϹ�¤�Τإ��㥹�Ȥ���ޥ������RXXXXX()��(������ʸ
-���ˤ������)�Ȥ���̾�����󶡤���Ƥ��ޤ�(��: RSTRING())��
-
-�㤨�С�ʸ����str��Ĺ�������뤿��ˤϡ�RSTRING(str)->len�פ�
-����ʸ����str��char*�Ȥ������뤿��ˤϡ�RSTRING(str)->ptr��
-�Ȥ��ޤ�������ξ��ˤϡ����줾���RARRAY(ary)->len�ס�
-��RARRAY(ary)->ptr�פȤʤ�ޤ���
-
-Ruby�ι�¤�Τ�ľ�ܥ�������������˵���Ĥ��ʤ���Фʤ�ʤ���
-�Ȥϡ������ʸ����ι�¤�Τ���Ȥϻ��Ȥ�������ǡ�ľ���ѹ���
-�ʤ����ȤǤ���ľ���ѹ�������硤���֥������Ȥ����Ƥ���������
-�Ȥ�ʤ��ʤäơ��פ�̥Х��θ����ˤʤ�ޤ���
-
-1.4 C�Υǡ�����VALUE���Ѵ�����
-
-VALUE�μºݤι�¤��
-
-  * FIXNUM�ξ��
-
-    1bit�����եȤ��ơ�LSB��Ω�Ƥ롥
-
-  * ����¾�Υݥ��󥿤ξ��
-
-    ���Τޤ�VALUE�˥��㥹�Ȥ��롥
-
-�ȤʤäƤ��ޤ�����äơ�LSB������å������VALUE��FIXNUM����
-�����狼��櫓�Ǥ�(�ݥ��󥿤�LSB��Ω�äƤ��ʤ����Ȥ��ꤷ��
-����)��
-
-�Ǥ����顤FIXNUM�ʳ���Ruby�Υ��֥������Ȥι�¤�Τ�ñ��VALUE
-�˥��㥹�Ȥ��������VALUE���Ѵ�����ޤ�����������Ǥ�դι�¤
-�Τ�VALUE�˥��㥹�Ƚ����櫓�ǤϤ���ޤ��󡥥��㥹�Ȥ����
-��Ruby���ΤäƤ��빽¤��(ruby.h���������Ƥ���struct RXxxx
-�Τ��)�����Ǥ���
-
-FIXNUM�˴ؤ��Ƥ��Ѵ��ޥ������ͳ����ɬ�פ�����ޤ���C������
-����VALUE���Ѵ�����ޥ����ϰʲ��Τ�Τ�����ޤ���ɬ�פ˱���
-�ƻȤ�ʬ���Ƥ���������
-
-  INT2FIX()	��Ȥ�������31bit����˼��ޤ뼫���������
-  INT2NUM()	Ǥ�դ���������VALUE��
-
-INT2NUM()��������FIXNUM���ϰϤ˼��ޤ�ʤ���硤Bignum���Ѵ�
-���Ƥ���ޤ�(���������٤�)��
-
-1.5 Ruby�Υǡ���������
-
-������Ҥ٤��̤ꡤRuby�ι�¤�Τ򥢥���������������Ƥι�����
-�Ԥ����Ȥϴ�����ޤ��󡥤ǡ�Ruby�Υǡ�����������ˤ�
-Ruby���Ѱդ��Ƥ���ؿ����Ѥ��Ƥ���������
-
-�����ǤϤ�äȤ�Ȥ���Ǥ�����ʸ��������������/�����
-���ؿ��򤢤��ޤ�(�����ǤϤʤ��Ǥ�)��
-
- ʸ������Ф���ؿ�
-
-  rb_str_new(const char *ptr, long len)
-
-    ������Ruby��ʸ������������롥
-
-  rb_str_new2(const char *ptr)
-
-    C��ʸ���󤫤�Ruby��ʸ������������롥���δؿ��ε�ǽ��
-    rb_str_new(ptr, strlen(ptr))��Ʊ���Ǥ��롥
-
-  rb_tainted_str_new(const char *ptr, long len)
-
-    �����ޡ������ղä��줿������Ruby��ʸ������������롥����
-    ����Υǡ����˴�Ť�ʸ����ˤϱ����ޡ������ղä����٤�
-    �Ǥ��롥
-
-  rb_tainted_str_new2(const char *ptr)
-
-    C��ʸ���󤫤�����ޡ������ղä��줿Ruby��ʸ������������롥
-
-  rb_str_cat(VALUE str, const char *ptr, long len)
-
-    Ruby��ʸ����str��len�Х��Ȥ�ʸ����ptr���ɲä��롥
-
- ������Ф���ؿ�
-
-  rb_ary_new()
-
-    ���Ǥ�0��������������롥
-
-  rb_ary_new2(long len)
-
-    ���Ǥ�0��������������롥len����ʬ���ΰ�򤢤餫������
-    ���ƤƤ�����
-
-  rb_ary_new3(long n, ...)
-
-    �����ǻ��ꤷ��n���Ǥ�ޤ�������������롥
-
-  rb_ary_new4(long n, VALUE *elts)
-
-    �����Ϳ����n���Ǥ�������������롥
-
-  rb_ary_push(VALUE ary, VALUE val)
-  rb_ary_pop(VALUE ary)
-  rb_ary_shift(VALUE ary)
-  rb_ary_unshift(VALUE ary, VALUE val)
-
-    Array��Ʊ̾�Υ᥽�åɤ�Ʊ��Ư���򤹤�ؿ�����1������ɬ��
-    ����Ǥʤ���Фʤ�ʤ���
-
-2��Ruby�ε�ǽ��Ȥ�
-
-����Ū��Ruby�ǽ񤱤뤳�Ȥ�C�Ǥ�񤱤ޤ���Ruby���Τ�Τ�C�ǵ�
-�Ҥ���Ƥ����Ǥ����顤�����Ȥ����������ʤ�Ǥ����ɡ�������
-��Ruby�γ�ĥ�˻Ȥ����Ȥ�¿����������ͽ¬����뵡ǽ���濴�˾�
-�𤷤ޤ���
-
-2.1 Ruby�˵�ǽ���ɲä���
-
-Ruby���󶡤���Ƥ���ؿ���Ȥ���Ruby���󥿥ץ꥿�˿�������ǽ
-���ɲä��뤳�Ȥ��Ǥ��ޤ���Ruby�Ǥϰʲ��ε�ǽ���ɲä���ؿ���
-�󶡤���Ƥ��ޤ���
-
- * ���饹���⥸�塼��
- * �᥽�åɡ��ðۥ᥽�åɤʤ�
- * ���
-
-�ǤϽ�˾Ҳ𤷤ޤ���
-
-2.1.1 ���饹/�⥸�塼�����
-
-���饹��⥸�塼���������뤿��ˤϡ��ʲ��δؿ���Ȥ��ޤ���
-
-  VALUE rb_define_class(const char *name, VALUE super)
-  VALUE rb_define_module(const char *name)
-
-�����δؿ��Ͽ�����������줿���饹��⥸�塼����֤��ޤ���
-�᥽�åɤ����������ˤ������ͤ�ɬ�פʤΤǡ��ۤȤ�ɤξ��
-������ͤ��ѿ��˳�Ǽ���Ƥ���ɬ�פ�����Ǥ��礦��
-
-���饹��⥸�塼���¾�Υ��饹�������˥ͥ��Ȥ�������������
-�ϰʲ��δؿ���Ȥ��ޤ���
-
-  VALUE rb_define_class_under(VALUE outer, const char *name, VALUE super)
-  VALUE rb_define_module_under(VALUE outer, const char *name)
-
-2.1.2 �᥽�å�/�ðۥ᥽�å����
-
-�᥽�åɤ��ðۥ᥽�åɤ��������ˤϰʲ��δؿ���Ȥ��ޤ���
-
-  void rb_define_method(VALUE klass, const char *name, 
-		        VALUE (*func)(), int argc)
-
-  void rb_define_singleton_method(VALUE object, const char *name, 
-			          VALUE (*func)(), int argc)
-
-
-ǰ�Τ�����������ȡ��ðۥ᥽�åɡפȤϡ���������Υ��֥�����
-�Ȥ��Ф��Ƥ���ͭ���ʥ᥽�åɤǤ���Ruby�ǤϤ褯Smalltalk�ˤ�
-���륯�饹�᥽�åɤȤ��ơ����饹���Ф����ðۥ᥽�åɤ��Ȥ��
-�ޤ���
-
-�����δؿ��� argc�Ȥ���������C�δؿ����Ϥ��������ο�(��
-����)����ޤ���argc��0�ʾ�λ��ϴؿ��˰����Ϥ������ο����
-̣���ޤ���16�İʾ�ΰ����ϻȤ��ޤ���(�����פ�ޤ����͡���
-��ʤ�)���ºݤδؿ��ˤ���Ƭ�ΰ����Ȥ���self��Ϳ�����ޤ���
-�ǡ����ꤷ�������1¿����������Ĥ��Ȥˤʤ�ޤ���
-
-argc����λ��ϰ����ο��ǤϤʤ�����������ꤷ�����Ȥˤʤ�ޤ���
-argc��-1�λ��ϰ����������������Ϥ���ޤ���argc��-2�λ��ϰ�
-����Ruby������Ȥ����Ϥ���ޤ���
-
-�᥽�åɤ��������ؿ��Ϥ⤦��Ĥ���ޤ����ҤȤĤ�private��
-���åɤ��������ؿ��ǡ�������rb_define_method()��Ʊ���Ǥ���
-
-  void rb_define_private_method(VALUE klass, const char *name, 
-				VALUE (*func)(), int argc)
-
-private�᥽�åɤȤϴؿ������Ǥ����ƤӽФ����Ȥν���ʤ��᥽��
-�ɤǤ���
-
-�⤦�ҤȤĤϥ⥸�塼��ؿ�����������ΤǤ����⥸�塼��ؿ�
-�Ȥϥ⥸�塼����ðۥ᥽�åɤǤ��ꡤƱ����private�᥽�åɤ�
-�⤢���ΤǤ�����򤢤����Math�⥸�塼���sqrt()�ʤɤ�����
-���ޤ������Υ᥽�åɤ�
-
-  Math.sqrt(4)
-
-�Ȥ��������Ǥ�
-
-  include Math
-  sqrt(4)
-
-�Ȥ��������Ǥ�Ȥ��ޤ����⥸�塼��ؿ����������ؿ��ϰʲ���
-�̤�Ǥ���
-
-  void rb_define_module_function(VALUE module, const char *name, 
-		                 VALUE (*func)(), int argc)
-
-�ؿ�Ū�᥽�å�(Kernel�⥸�塼���private method)��������뤿
-��δؿ��ϰʲ����̤�Ǥ���
-
-  void rb_define_global_function(const char *name, VALUE (*func)(), int argc)
-
-
-�᥽�åɤ���̾��������뤿��δؿ��ϰʲ����̤�Ǥ���
-
-  void rb_define_alias(VALUE module, const char* new, const char* old);
-
-���饹�᥽�å�allocate������������������ꤹ�뤿��δؿ���
-�ʲ����̤�Ǥ���
-
-  void rb_define_alloc_func(VALUE klass, VALUE (*func)(VALUE klass));
-  void rb_undef_alloc_func(VALUE klass);
-
-func�ϥ��饹������Ȥ��Ƽ�����äơ�������������Ƥ�줿����
-�����󥹤��֤��ʤ��ƤϤʤ�ޤ��󡣤��Υ��󥹥��󥹤ϡ�������
-�������ʤɤ�ޤޤʤ����Ǥ�������ֶ��פΤޤޤˤ��Ƥ������ۤ�
-���褤�Ǥ��礦��
-
-2.1.3 ������
-
-��ĥ�饤�֥�꤬ɬ�פ�����Ϥ��餫����������Ƥ����������ɤ�
-�Ǥ��礦��������������ؿ�����Ĥ���ޤ���
-
-  void rb_define_const(VALUE klass, const char *name, VALUE val)
-  void rb_define_global_const(const char *name, VALUE val)
-
-���Ԥ�����Υ��饹/�⥸�塼���°�����������������Ρ���
-�Ԥϥ������Х���������������ΤǤ���
-
-2.2 Ruby�ε�ǽ��C����ƤӽФ�
-
-���ˡ�1.5 Ruby�Υǡ���������٤ǰ����Ҳ𤷤��褦�ʴؿ���
-�Ȥ��С�Ruby�ε�ǽ��¸����Ƥ���ؿ���ľ�ܸƤӽФ����Ȥ�����
-�ޤ���
-
-# ���Τ褦�ʴؿ��ΰ���ɽ�Ϥ��ޤΤȤ�������ޤ��󡥥�������
-# �뤷���ʤ��Ǥ��͡�
-
-����ʳ��ˤ�Ruby�ε�ǽ��ƤӽФ���ˡ�Ϥ����Ĥ�����ޤ���
-
-2.2.1 Ruby�Υץ�������eval����
-
-C����Ruby�ε�ǽ��ƤӽФ���äȤ��ñ����ˡ�Ȥ��ơ�ʸ�����
-Ϳ����줿Ruby�Υץ�������ɾ������ʲ��δؿ�������ޤ���
-
-  VALUE rb_eval_string(const char *str)
-
-����ɾ���ϸ��ߤδĶ��ǹԤ��ޤ����Ĥޤꡤ���ߤΥ��������ѿ�
-�ʤɤ�����Ѥ��ޤ���
-
-2.2.2 ID�ޤ��ϥ���ܥ�
-
-C����ʸ������ͳ������Ruby�Υ᥽�åɤ�ƤӽФ����Ȥ�Ǥ���
-�����������ˡ�Ruby���󥿥ץ꥿��ǥ᥽�åɤ��ѿ�̾����ꤹ��
-���˻Ȥ��Ƥ���ID�ˤĤ����������Ƥ����ޤ��礦��
-
-ID�Ȥ��ѿ�̾���᥽�å�̾��ɽ�������Ǥ���Ruby�Ǥ�ID���б�����
-���֥������ȤȤ��ƥ���ܥ�(Symbol)�����ꡤ
-
- :���̻�
-
-�ǥ��������Ǥ��ޤ���C���餳�����������뤿��ˤϴؿ�
-
-  rb_intern(const char *name)
-
-��Ȥ��ޤ���Ruby��������Ȥ���Ϳ����줿����ܥ�(�ޤ���ʸ��
-��)��ID���Ѵ�����ˤϰʲ��δؿ���Ȥ��ޤ���
-
-  rb_to_id(VALUE symbol)
-
-ID���饷��ܥ�����뤿��ˤϰʲ��Υޥ�����Ȥ��ޤ���
-
-  VALUE ID2SYM(ID id)
-
-����ܥ뤫��ID�����뤿��ˤϰʲ��Υޥ�����Ȥ��ޤ���
-
-  ID SYM2ID(VALUE symbol)
-
-2.2.3 C����Ruby�Υ᥽�åɤ�ƤӽФ�
-
-C����ʸ������ͳ������Ruby�Υ᥽�åɤ�ƤӽФ�����ˤϰʲ�
-�δؿ���Ȥ��ޤ���
-
-  VALUE rb_funcall(VALUE recv, ID mid, int argc, ...)
-
-���δؿ��ϥ��֥�������recv��mid�ǻ��ꤵ���᥽�åɤ�Ƥӽ�
-���ޤ�������¾�˰����λ���λ������㤦�ʲ��δؿ��⤢��ޤ���
-
-  VALUE rb_funcall2(VALUE recv, ID mid, int argc, VALUE *argv)
-  VALUE rb_apply(VALUE recv, ID mid, VALUE args)
-
-apply�ˤϰ����Ȥ���Ruby�������Ϳ���ޤ���
-
-2.2.4 �ѿ�/����򻲾�/��������
-
-C����ؿ���Ȥäƻ��ȡ������Ǥ���Τϡ���������󥹥�����
-���Ǥ�������ѿ��ϰ����Τ�Τ�C������ѿ��Ȥ��ƥ��������Ǥ�
-�ޤ������������ѿ��򻲾Ȥ�����ˡ�ϸ������Ƥ��ޤ���
-
-���֥������ȤΥ��󥹥����ѿ��򻲾ȡ���������ؿ��ϰʲ�����
-��Ǥ���
-
-  VALUE rb_ivar_get(VALUE obj, ID id)
-  VALUE rb_ivar_set(VALUE obj, ID id, VALUE val)
-
-id��rb_intern()���������Τ�ȤäƤ���������
-
-����򻲾Ȥ���ˤϰʲ��δؿ���ȤäƤ���������
-
-  VALUE rb_const_get(VALUE obj, ID id)
-
-����򿷤���������뤿��ˤϡ�2.1.3 �������٤ǾҲ�
-��Ƥ���ؿ���ȤäƤ���������
-
-3��Ruby��C�Ȥξ���ͭ
-
-C�����Ruby�δ֤Ǿ����ͭ������ˡ�ˤĤ��Ʋ��⤷�ޤ���
-
-3.1 C���黲�ȤǤ���Ruby�����
-
-�ʲ���Ruby�������C�Υ�٥뤫�黲�ȤǤ��ޤ���
-
-  Qtrue
-  Qfalse
-
-    �����͡�Qfalse��C����Ǥ⵶�Ȥߤʤ���ޤ�(�Ĥޤ�0)��
-
-  Qnil
-
-    C���줫�鸫����nil�ס�
-
-3.2 C��Ruby�Ƕ�ͭ���������ѿ�
-
-C��Ruby������ѿ���Ȥäƾ����ͭ�Ǥ��ޤ�����ͭ�Ǥ������
-�ѿ��ˤϤ����Ĥ��μ��ब����ޤ������Τʤ��Ǥ�äȤ��ɤ��Ȥ�
-���Ȼפ���Τ�rb_define_variable()�Ǥ���
-
-  void rb_define_variable(const char *name, VALUE *var)
-
-���δؿ���Ruby��C�ȤǶ�ͭ��������ѿ���������ޤ����ѿ�̾��
-`$'�ǻϤޤ�ʤ����ˤϼ�ưŪ���ɲä���ޤ��������ѿ����ͤ���
-������ȼ�ưŪ��Ruby���б������ѿ����ͤ��Ѥ��ޤ���
-
-�ޤ�Ruby¦����Ϲ����Ǥ��ʤ��ѿ��⤢��ޤ�������read only��
-�ѿ��ϰʲ��δؿ���������ޤ���
-
-  void rb_define_readonly_variable(const char *name, VALUE *var)
-
-������ѿ���¾��hook��Ĥ�������ѿ�������Ǥ��ޤ���hook�դ�
-������ѿ��ϰʲ��δؿ����Ѥ���������ޤ���hook�դ�����ѿ���
-�ͤλ��Ȥ������hook�ǹԤ�ɬ�פ�����ޤ���
-
-  void rb_define_hooked_variable(const char *name, VALUE *var,
-				 VALUE (*getter)(), void (*setter)())
-
-���δؿ���C�δؿ��ˤ�ä�hook�ΤĤ���줿����ѿ����������
-�����ѿ������Ȥ��줿���ˤϴؿ�getter�����ѿ����ͤ����åȤ���
-�����ˤϴؿ�setter���ƤФ�롥hook����ꤷ�ʤ�����getter��
-setter��0����ꤷ�ޤ���
-
-# getter��setter��0�ʤ��rb_define_variable()��Ʊ���ˤʤ롥
-
-���줫�顤C�δؿ��ˤ�äƼ¸������Ruby������ѿ����������
-�ؿ�������ޤ���
-
-  void rb_define_virtual_variable(const char *name,
-				  VALUE (*getter)(), void (*setter)())
-
-���δؿ��ˤ�ä�������줿Ruby������ѿ������Ȥ��줿���ˤ�
-getter�����ѿ����ͤ����åȤ��줿���ˤ�setter���ƤФ�ޤ���
-
-getter��setter�λ��ͤϰʲ����̤�Ǥ���
-
-  (*getter)(ID id, void *data, struct global_entry* entry);
-  (*setter)(VALUE val, ID id, void *data, struct global_entry* entry);
-
-3.3 C�Υǡ�����Ruby���֥������Ȥˤ���
-
-C��������������줿�ǡ���(��¤��)��Ruby�Υ��֥������ȤȤ���
-��갷��������礬���ꤨ�ޤ������Τ褦�ʾ��ˤϡ�Data�Ȥ���
-Ruby���֥������Ȥ�C�ι�¤��(�ؤΥݥ���)�򤯤�ळ�Ȥ�Ruby
-���֥������ȤȤ��Ƽ�갷����褦�ˤʤ�ޤ���
-
-Data���֥������Ȥ��������ƹ�¤�Τ�Ruby���֥������Ȥ˥��ץ���
-�����뤿��ˤϡ��ʲ��Υޥ�����Ȥ��ޤ���
-
-  Data_Wrap_Struct(klass, mark, free, ptr)
-
-���Υޥ���������ͤ��������줿Data���֥������ȤǤ���
-
-klass�Ϥ���Data���֥������ȤΥ��饹�Ǥ���ptr�ϥ��ץ��벽����
-C�ι�¤�ΤؤΥݥ��󥿤Ǥ���mark�Ϥ��ι�¤�Τ�Ruby�Υ��֥���
-���Ȥؤλ��Ȥ�������˻Ȥ��ؿ��Ǥ������Τ褦�ʻ��Ȥ�ޤޤʤ�
-���ˤ�0����ꤷ�ޤ���
-
-# ���Τ褦�ʻ��Ȥϴ�����ޤ���
-
-free�Ϥ��ι�¤�Τ��⤦���פˤʤä����˸ƤФ��ؿ��Ǥ�������
-�ؿ��������١������쥯������ƤФ�ޤ������줬-1�ξ��ϡ�ñ
-��˳�������ޤ���
-
-C�ι�¤�Τγ�����Data���֥������Ȥ�������Ʊ���˹Ԥ��ޥ�����
-���ưʲ��Τ�Τ��󶡤���Ƥ��ޤ���
-
-  Data_Make_Struct(klass, type, mark, free, sval)
-
-���Υޥ���������ͤ��������줿Data���֥������ȤǤ���
-
-klass, mark, free��Data_Wrap_Struct��Ʊ��Ư���򤷤ޤ���type
-�ϳ�����Ƥ�C��¤�Τη��Ǥ���������Ƥ�줿��¤�Τ��ѿ�sval
-����������ޤ��������ѿ��η��� (type*) �Ǥ���ɬ�פ�����ޤ���
-
-Data���֥������Ȥ���ݥ��󥿤���Ф��Τϰʲ��Υޥ������Ѥ�
-�ޤ���
-
-  Data_Get_Struct(obj, type, sval)
-
-C�ι�¤�ΤؤΥݥ��󥿤��ѿ�sval����������ޤ���
-
-������Data�λȤ����Ϥ���ä�ʬ����ˤ����Τǡ������������
-����򻲾Ȥ��Ƥ���������
-
-4������ - dbm�ѥå���������
-
-�����ޤǤ������ǤȤꤢ������ĥ�饤�֥��Ϻ���Ϥ��Ǥ���
-Ruby��ext�ǥ��쥯�ȥ�ˤ��Ǥ˴ޤޤ�Ƥ���dbm�饤�֥������
-�����ʳ�Ū���������ޤ���
-
-(1) �ǥ��쥯�ȥ����
-
-  % mkdir ext/dbm
-
-Ruby 1.1�����Ǥ�դΥǥ��쥯�ȥ�ǥ����ʥߥå��饤�֥����
-�뤳�Ȥ��Ǥ���褦�ˤʤ�ޤ�����Ruby����Ū�˥�󥯤������
-��Ruby��Ÿ�������ǥ��쥯�ȥ�β���ext�ǥ��쥯�ȥ����˳�ĥ
-�饤�֥���ѤΥǥ��쥯�ȥ����ɬ�פ�����ޤ���̾����Ŭ����
-����ǹ����ޤ���
-
-(2) �߷פ���
-
-�ޤ��������ʤ�Ǥ����ɡ��ɤ�������ǽ��¸����뤫�ɤ����ޤ���
-�פ���ɬ�פ�����ޤ����ɤ�ʥ��饹��Ĥ��뤫�����Υ��饹�ˤ�
-�ɤ�ʥ᥽�åɤ����뤫�����饹���󶡤�������ʤɤˤĤ����߷�
-���ޤ���
-
-(3) C�����ɤ��
-
-��ĥ�饤�֥�����ΤȤʤ�C����Υ�������񤭤ޤ���C����Υ���
-�����ҤȤĤλ��ˤϡ֥饤�֥��̾.c�פ����֤��ɤ��Ǥ��礦��C
-����Υ�������ʣ���ξ��ˤϵդˡ֥饤�֥��̾.c�פȤ����ե�
-����̾���򤱤�ɬ�פ�����ޤ������֥������ȥե�����ȥ⥸�塼
-�������������Ū�����������֥饤�֥��̾.o�פȤ����ե�����
-�Ȥ����ͤ��뤫��Ǥ���
-
-Ruby�ϳ�ĥ�饤�֥�������ɤ�����ˡ�Init_�饤�֥��̾�פ�
-�����ؿ���ưŪ�˼¹Ԥ��ޤ���dbm�饤�֥��ξ���Init_dbm��
-�Ǥ������δؿ�����ǥ��饹���⥸�塼�롤�᥽�åɡ�����ʤɤ�
-�����Ԥ��ޤ���dbm.c����������Ѥ��ޤ���
-
---
-Init_dbm()
-{
-    /* DBM���饹��������� */
-    cDBM = rb_define_class("DBM", rb_cObject);
-    /* DBM��Enumerate�⥸�塼��򥤥󥯥롼�ɤ��� */
-    rb_include_module(cDBM, rb_mEnumerable);
-
-    /* DBM���饹�Υ��饹�᥽�å�open(): ������C������Ǽ����� */
-    rb_define_singleton_method(cDBM, "open", fdbm_s_open, -1);
-
-    /* DBM���饹�Υ᥽�å�close(): �����Ϥʤ� */
-    rb_define_method(cDBM, "close", fdbm_close, 0);
-    /* DBM���饹�Υ᥽�å�[]: ������1�� */
-    rb_define_method(cDBM, "[]", fdbm_fetch, 1);
-		:
-
-    /* DBM�ǡ������Ǽ���륤�󥹥����ѿ�̾�Τ����ID */
-    id_dbm = rb_intern("dbm");
-}
---
-
-DBM�饤�֥���dbm�Υǡ������б����륪�֥������Ȥˤʤ�Ϥ���
-�����顤C��������dbm��Ruby�������˼�����ɬ�פ�����ޤ���
-
-
-dbm.c�Ǥ�Data_Make_Struct��ʲ��Τ褦�˻ȤäƤ��ޤ���
-
---
-struct dbmdata {
-    int  di_size;
-    DBM *di_dbm;
-};
-
-
-obj = Data_Make_Struct(klass, struct dbmdata, 0, free_dbm, dbmp);
---
-
-�����Ǥ�dbmstruct��¤�ΤؤΥݥ��󥿤�Data�˥��ץ��벽���Ƥ�
-�ޤ���DBM*��ľ�ܥ��ץ��벽���ʤ��Τ�close()�������ν������
-���ƤΤ��ȤǤ���
-
-Data���֥������Ȥ���dbmstruct��¤�ΤΥݥ��󥿤���Ф�����
-�˰ʲ��Υޥ�����ȤäƤ��ޤ���
-
---
-#define GetDBM(obj, dbmp) {\
-    Data_Get_Struct(obj, struct dbmdata, dbmp);\
-    if (dbmp->di_dbm == 0) closed_dbm();\
-}
---
-
-����ä�ʣ���ʥޥ����Ǥ������פ����dbmdata��¤�ΤΥݥ���
-�μ��Ф��ȡ�close����Ƥ��뤫�ɤ����Υ����å���ޤȤ�Ƥ�
-������Ǥ���
-
-DBM���饹�ˤϤ�������᥽�åɤ�����ޤ�����ʬ�ह���3�����
-�����μ�����������ޤ����ҤȤĤϰ����ο�������Τ�Τǡ����
-���Ƥ�delete�᥽�åɤ�����ޤ���delete�᥽�åɤ�������Ƥ���
-fdbm_delete()�Ϥ��Τ褦�ˤʤäƤ��ޤ���
-
---
-static VALUE
-fdbm_delete(obj, keystr)
-    VALUE obj, keystr;
-{
-	:
-}
---
-
-�����ο�������Υ����פ���1������self����2�����ʹߤ��᥽�å�
-�ΰ����Ȥʤ�ޤ���
-
-�����ο�������Τ�Τ�C������Ǽ������Τ�Ruby������Ǽ���
-���ΤȤ�����ޤ���dbm�饤�֥�����ǡ�C������Ǽ�������
-��DBM�Υ��饹�᥽�åɤǤ���open()�Ǥ��������������Ƥ����
-��fdbm_s_open()�Ϥ����ʤäƤ��ޤ���
-
---
-static VALUE
-fdbm_s_open(argc, argv, klass)
-    int argc;
-    VALUE *argv;
-    VALUE klass;
-{
-	:
-    if (rb_scan_args(argc, argv, "11", &file, &vmode) == 1) {
-	mode = 0666;		/* default value */
-    }
-	:
-}
---
-
-���Υ����פδؿ�����1������Ϳ����줿�����ο�����2������Ϳ��
-��줿���������äƤ�������ˤʤ�ޤ���self����3�����Ȥ���Ϳ
-�����ޤ���
-
-���������Ϳ����줿��������Ϥ��뤿��δؿ���open()�Ǥ�Ȥ�
-��Ƥ���rb_scan_args()�Ǥ�����3�����˻��ꤷ���ե����ޥåȤ�
-��������4�ѿ��ʹߤ˻��ꤷ���ѿ����ͤ��������Ƥ���ޤ�������
-�ե����ޥåȤϡ���1ʸ���ܤ���ά�Ǥ��ʤ������ο�����2ʸ���ܤ�
-��ά�Ǥ�������ο�����3ʸ���ܤ��б�������̵꤬�����ޤ�ΰ�
-�������뤫�ɤ����򼨤�"*"�Ǥ���2ʸ���ܤ�3ʸ���ܤϾ�ά�Ǥ���
-����dbm.c����Ǥϡ��ե����ޥåȤ�"11"�Ǥ����顤�����Ϻ���1��
-�ǡ�2�Ĥޤǵ������Ȥ�����̣�ˤʤ�ޤ�����ά����Ƥ������
-�ѿ����ͤ�nil(C����Υ�٥�Ǥ�Qnil)�ˤʤ�ޤ���
-
-Ruby������ǰ������������Τ�indexes������ޤ��������Ϥ�
-���Ǥ���
-
---
-static VALUE
-fdbm_indexes(obj, args)
-    VALUE obj, args;
-{
-	:
-}
---
-
-��1������self����2������Ruby������Ǥ���
-
-** ���ջ���
-
-Ruby�ȶ�ͭ�Ϥ��ʤ���Ruby�Υ��֥������Ȥ��Ǽ�����ǽ���Τ���
-C������ѿ��ϰʲ��δؿ���Ȥä�Ruby���󥿥ץ꥿���ѿ���¸��
-�򶵤��Ƥ����Ƥ����������Ǥʤ���GC�ǥȥ�֥�򵯤����ޤ���
-
-  void rb_global_variable(VALUE *var)
-
-(4) extconf.rb���Ѱդ���
-
-Makefile������ο����ˤʤ�extconf.rb�Ȥ����ե��������
-�ޤ���extconf.rb�ϥ饤�֥��Υ���ѥ����ɬ�פʾ��Υ�����
-���ʤɤ�Ԥ����Ȥ���Ū�Ǥ����ޤ���
-
-  require 'mkmf'
-
-��extconf.rb����Ƭ���֤��ޤ���extconf.rb����Ǥϰʲ���Ruby��
-����Ȥ����Ȥ�����ޤ���
-
-  have_library(lib, func): �饤�֥���¸�ߥ����å�
-  have_func(func, header): �ؿ���¸�ߥ����å�
-  have_header(header): �إå��ե������¸�ߥ����å�
-  create_makefile(target): Makefile������
-
-�ʲ����ѿ���Ȥ����Ȥ��Ǥ��ޤ���
-
-  $CFLAGS: ����ѥ�������ɲ�Ū�˻��ꤹ��ե饰(-O�ʤ�)
-  $CPPFLAGS: �ץ�ץ����å����ɲ�Ū�˻��ꤹ��ե饰(-I��-D�ʤ�)
-  $LDFLAGS: ��󥯻����ɲ�Ū�˻��ꤹ��ե饰(-L�ʤ�)
-  $objs: ��󥯤���륪�֥������ȥե�����̾�Υꥹ��
-
-���֥������ȥե�����Υꥹ�Ȥϡ��̾�ϥ������ե�����򸡺���
-�Ƽ�ưŪ����������ޤ�����make������ǥ���������������褦��
-��������Ū�˻��ꤹ��ɬ�פ�����ޤ���
-
-�饤�֥��򥳥�ѥ��뤹���郎·�鷺�����Υ饤�֥��򥳥�
-�ѥ��뤷�ʤ����ˤ�create_makefile��ƤФʤ����Makefile����
-�����줺������ѥ����Ԥ��ޤ���
-
-(5) depend���Ѱդ���
-
-�⤷���ǥ��쥯�ȥ��depend�Ȥ����ե����뤬¸�ߤ���С�
-Makefile����¸�ط�������å����Ƥ���ޤ���
-
-  % gcc -MM *.c > depend
-
-�ʤɤǺ�뤳�Ȥ�����ޤ������ä�»��̵���Ǥ��礦��
-
-(6) Makefile����������
-
-Makefile��ºݤ��������뤿��ˤ�
-
-  ruby extconf.rb
-
-�Ȥ��ޤ���extconf.rb�� require 'mkmf' �ιԤ��ʤ����ˤϥ��顼
-�ˤʤ�ޤ��Τǡ��������ɲä���
-
-  ruby -r mkmf extconf.rb
-
-�Ȥ��Ƥ���������
-
-�ǥ��쥯�ȥ��ext�ʲ����Ѱդ������ˤ�Ruby���Τ�make�λ���
-��ưŪ��Makefile����������ޤ��Τǡ����Υ��ƥåפ����פǤ���
-
-(7) make����
-
-ưŪ��󥯥饤�֥�������������ˤϤ��ξ��make���Ƥ�����
-����ɬ�פǤ���� make install �ǥ��󥹥ȡ��뤵��ޤ���
-
-ext�ʲ��˥ǥ��쥯�ȥ���Ѱդ������ϡ�Ruby�Υǥ��쥯�ȥ��
-make��¹Ԥ����Makefile����������make��ɬ�פˤ�äƤϤ��Υ�
-���塼���Ruby�ؤΥ�󥯤ޤǼ�ưŪ�˼¹Ԥ��Ƥ���ޤ���
-extconf.rb��񤭴�����ʤɤ���Makefile�κ�������ɬ�פʻ��Ϥ�
-��Ruby�ǥ��쥯�ȥ��make���Ƥ���������
-
-��ĥ�饤�֥���make install��Ruby�饤�֥��Υǥ��쥯�ȥ��
-���˥��ԡ�����ޤ����⤷��ĥ�饤�֥��ȶ�Ĵ���ƻȤ�Ruby�ǵ�
-�Ҥ��줿�ץ�����ब���ꡤRuby�饤�֥����֤��������ˤϡ�
-��ĥ�饤�֥���ѤΥǥ��쥯�ȥ�β��� lib �Ȥ����ǥ��쥯�ȥ�
-���ꡤ������ ��ĥ�� .rb �Υե�������֤��Ƥ�����Ʊ���˥���
-���ȡ��뤵��ޤ���
-
-(8) �ǥХå�
-
-�ޤ����ǥХå����ʤ���ư���ʤ��Ǥ��礦�͡�ext/Setup�˥ǥ���
-���ȥ�̾��񤯤���Ū�˥�󥯤���ΤǥǥХå����Ȥ���褦�ˤ�
-��ޤ�������ʬ����ѥ��뤬�٤��ʤ�ޤ����ɡ�
-
-(9) �Ǥ�������
-
-��Ϥ��ä���Ȥ��ʤꡤ������������ʤꡤ���ʤꡤ����ͳ�ˤ�
-�Ȥ�����������Ruby�κ�Ԥϳ�ĥ�饤�֥��˴ؤ��ư��ڤθ�����
-��ĥ���ޤ���
-
-Appendix A. Ruby�Υ����������ɤ�ʬ��
-
-Ruby�Υ������Ϥ����Ĥ���ʬ�ह�뤳�Ȥ�����ޤ������Τ�������
-���饤�֥�����ʬ�ϴ���Ū�˳�ĥ�饤�֥���Ʊ��������ˤʤ�
-�Ƥ��ޤ��������Υ������Ϻ��ޤǤ������ǤۤȤ������Ǥ����
-�פ��ޤ���
-
-Ruby����Υ���
-
-  class.c
-  error.c
-  eval.c
-  gc.c
-  object.c
-  parse.y
-  variable.c
-
-�桼�ƥ���ƥ��ؿ�
-
-  dln.c
-  regex.c
-  st.c
-  util.c
-
-Ruby���ޥ�ɤμ���
-
-  dmyext.c
-  inits.c
-  main.c
-  ruby.c
-  version.c
-
-���饹�饤�֥��
-
-  array.c
-  bignum.c
-  compar.c
-  dir.c
-  enum.c
-  file.c
-  hash.c
-  io.c
-  marshal.c
-  math.c
-  numeric.c
-  pack.c
-  prec.c
-  process.c
-  random.c
-  range.c
-  re.c
-  signal.c
-  sprintf.c
-  string.c
-  struct.c
-  time.c
-
-Appendix B. ��ĥ�Ѵؿ���ե����
-
-C���줫��Ruby�ε�ǽ�����Ѥ���API�ϰʲ����̤�Ǥ��롥
-
-** ��
-
-VALUE
-
-  Ruby���֥������Ȥ�ɽ�����뷿��ɬ�פ˱����ƥ��㥹�Ȥ����Ѥ��롥
-  �Ȥ߹��߷���ɽ������C�η���ruby.h�˵��Ҥ��Ƥ���R�ǻϤޤ빽¤
-  �ΤǤ��롥VALUE���򤳤��˥��㥹�Ȥ��뤿���R�ǻϤޤ빽¤��
-  ̾��������ʸ���ˤ���̾���Υޥ������Ѱդ���Ƥ��롥
-
-** �ѿ������
-
-Qnil
-
-  ���: nil���֥�������
-
-Qtrue
-
-  ���: true���֥�������(���Υǥե������)
-
-Qfalse
-
-  ���: false���֥�������
-
-** C�ǡ����Υ��ץ��벽
-
-Data_Wrap_Struct(VALUE klass, void (*mark)(), void (*free)(), void *sval)
-
-  C��Ǥ�դΥݥ��󥿤򥫥ץ��벽����Ruby���֥������Ȥ��֤�����
-  �Υݥ��󥿤�Ruby���饢����������ʤ��ʤä�����free�ǻ��ꤷ��
-  �ؿ����ƤФ�롥�ޤ������Υݥ��󥿤λؤ��ǡ�����¾��Ruby����
-  �������Ȥ�ؤ��Ƥ����硤mark�˻��ꤹ��ؿ��ǥޡ�������ɬ��
-  �����롥
-
-Data_Make_Struct(klass, type, mark, free, sval)
-
-  type���Υ����malloc�����ѿ�sval�����������塤����򥫥ץ�
-  �벽�����ǡ������֤��ޥ�����
-
-Data_Get_Struct(data, type, sval)
-
-  data����type���Υݥ��󥿤���Ф��ѿ�sval����������ޥ�����
-
-** �������å�
-
-TYPE(value)
-FIXNUM_P(value)
-NIL_P(value)
-void Check_Type(VALUE value, int type)
-void Check_SafeStr(VALUE value)
-
-** ���Ѵ�
-
-FIX2INT(value)
-INT2FIX(i)
-NUM2INT(value)
-INT2NUM(i)
-NUM2DBL(value)
-rb_float_new(f)
-StringValue(value)
-StringValuePtr(value)
-StringValueCStr(value)
-rb_str_new2(s)
-
-** ���饹/�⥸�塼�����
-
-VALUE rb_define_class(const char *name, VALUE super)
-
-  super�Υ��֥��饹�Ȥ��ƿ�����Ruby���饹��������롥
-
-VALUE rb_define_class_under(VALUE module, const char *name, VALUE super)
-
-  super�Υ��֥��饹�Ȥ��ƿ�����Ruby���饹���������module��
-  ����Ȥ���������롥
-
-VALUE rb_define_module(const char *name)
-
-  ������Ruby�⥸�塼���������롥
-
-VALUE rb_define_module_under(VALUE module, const char *name)
-
-  ������Ruby�⥸�塼����������module������Ȥ���������롥
-
-void rb_include_module(VALUE klass, VALUE module)
-
-  �⥸�塼��򥤥󥯥롼�ɤ��롥class�����Ǥ�module�򥤥�
-  �롼�ɤ��Ƥ�����ˤϲ��⤷�ʤ�(¿�ť��󥯥롼�ɤζػ�)��
-
-void rb_extend_object(VALUE object, VALUE module)
-
-  ���֥������Ȥ�⥸�塼��(���������Ƥ���᥽�å�)�ǳ�ĥ���롥
-
-** ����ѿ����
-
-void rb_define_variable(const char *name, VALUE *var)
-
-  Ruby��C�ȤǶ�ͭ���륰�����Х��ѿ���������롥�ѿ�̾��`$'��
-  �Ϥޤ�ʤ����ˤϼ�ưŪ���ɲä���롥name�Ȥ���Ruby�μ��̻�
-  �Ȥ��Ƶ�����ʤ�ʸ��(�㤨��` ')��ޤ���ˤ�Ruby�ץ�����
-  �फ��ϸ����ʤ��ʤ롥
-
-void rb_define_readonly_variable(const char *name, VALUE *var)
-
-  Ruby��C�ȤǶ�ͭ����read only�Υ������Х��ѿ���������롥
-  read only�Ǥ��뤳�Ȱʳ���rb_define_variable()��Ʊ����
-
-void rb_define_virtual_variable(const char *name,
-				VALUE (*getter)(), void (*setter)())
-
-  �ؿ��ˤ�äƼ¸������Ruby�ѿ���������롥�ѿ������Ȥ��줿
-  ���ˤ�getter�����ѿ����ͤ����åȤ��줿���ˤ�setter���ƤФ�
-  �롥
-
-void rb_define_hooked_variable(const char *name, VALUE *var,
-			       VALUE (*getter)(), void (*setter)())
-
-  �ؿ��ˤ�ä�hook�ΤĤ���줿�������Х��ѿ���������롥�ѿ�
-  �����Ȥ��줿���ˤ�getter�����ؿ����ͤ����åȤ��줿���ˤ�
-  setter���ƤФ�롥getter��setter��0����ꤷ�����ˤ�hook��
-  ���ꤷ�ʤ��Τ�Ʊ�����ˤʤ롥
-
-void rb_global_variable(VALUE *var)
-
-  GC�Τ��ᡤRuby�ץ�����फ��ϥ�����������ʤ���, Ruby����
-  �������Ȥ�ޤ�����ѿ���ޡ������롥
-
-** ���
-
-void rb_define_const(VALUE klass, const char *name, VALUE val)
-
-  �����������롥
-
-void rb_define_global_const(const char *name, VALUE val)
-
-  ��������������롥
-
-     rb_define_const(rb_cObject, name, val)
-
-  ��Ʊ����̣��
-
-** �᥽�å����
-
-rb_define_method(VALUE klass, const char *name, VALUE (*func)(), int argc)
-
-  �᥽�åɤ�������롥argc��self����������ο���argc��-1�λ�, 
-  �ؿ��ˤϰ����ο�(self��ޤޤʤ�)����1����, �������������2
-  �����Ȥ��������Ϳ������(��3������self)��argc��-2�λ�, 
-  ��1������self, ��2������args(args�ϰ�����ޤ�Ruby������)��
-  ����������Ϳ�����롥
- 
-rb_define_private_method(VALUE klass, const char *name, VALUE (*func)(), int argc)
-
-  private�᥽�åɤ�������롥������rb_define_method()��Ʊ����
-
-rb_define_singleton_method(VALUE klass, const char *name, VALUE (*func)(), int argc)
-
-  �ðۥ᥽�åɤ�������롥������rb_define_method()��Ʊ����
-
-rb_scan_args(int argc, VALUE *argv, const char *fmt, ...)
-
-  argc, argv������Ϳ����줿������ʬ�򤹤롥fmt��ɬ�ܰ����ο�, 
-  �ղð����ο�, �Ĥ�ΰ��������뤫����ꤹ��ʸ�����, "����
-  ����*"�Ȥ��������Ǥ��롥 2 ���ܤο�����"*"�Ϥ��줾���ά��
-  ǽ�Ǥ��롥ɬ�ܰ�������Ĥ�ʤ�����0����ꤹ�롥��3������
-  �ߤ��ѿ��ؤΥݥ��󥿤�, �����������Ǥ������ѿ��˳�Ǽ����롥
-  �ղð������б����������Ϳ�����Ƥ��ʤ������ѿ���Qnil��
-  ��������롥
-
-** Ruby�᥽�åɸƤӽФ�
-
-VALUE rb_funcall(VALUE recv, ID mid, int narg, ...)
-
-  �᥽�åɸƤӽФ���ʸ���󤫤�mid�����뤿��ˤ�rb_intern()��
-  �Ȥ���
-
-VALUE rb_funcall2(VALUE recv, ID mid, int argc, VALUE *argv)
-
-  �᥽�åɸƤӽФ���������argc, argv�������Ϥ���
-
-VALUE rb_eval_string(const char *str)
-
-  ʸ�����Ruby������ץȤȤ��ƥ���ѥ��롦�¹Ԥ��롥
-
-ID rb_intern(const char *name)
-
-  ʸ������б�����ID���֤���
-
-char *rb_id2name(ID id)
-
-  ID���б�����ʸ������֤�(�ǥХå���)��
-
-char *rb_class2name(VALUE klass)
-
-  ���饹��̾�����֤�(�ǥХå���)�����饹��̾��������ʤ����ˤ�, 
-  ������̤ä�̾������ĥ��饹��̾�����֤���
-
-int rb_respond_to(VALUE obj, ID id)
-
-  obj��id�Ǽ������᥽�åɤ���Ĥ��ɤ������֤���
-
-** ���󥹥����ѿ�
-
-VALUE rb_iv_get(VALUE obj, const char *name)
-
-  obj�Υ��󥹥����ѿ����ͤ����롥`@'�ǻϤޤ�ʤ����󥹥���
-  ���ѿ��� Ruby�ץ�����फ�饢�������Ǥ��ʤ��ֱ��줿�ץ���
-  �������ѿ��ˤʤ롥�������ʸ����̾������ĥ��饹(�ޤ���
-  �⥸�塼��)�Υ��󥹥����ѿ��Ȥ��Ƽ�������Ƥ��롥
-
-VALUE rb_iv_set(VALUE obj, const char *name, VALUE val)
-
-  obj�Υ��󥹥����ѿ���val�˥��åȤ��롥
-
-** ���湽¤
-
-VALUE rb_iterate(VALUE (*func1)(), VALUE arg1, VALUE (*func2)(), VALUE arg2)
-
-  func2��֥��å��Ȥ������ꤷ, func1�򥤥ƥ졼���Ȥ��ƸƤ֡� 
-  func1�ˤ� arg1�������Ȥ����Ϥ���, func2�ˤ���1�����˥��ƥ졼
-  ������Ϳ����줿��, ��2������arg2���Ϥ���롥
- 
-VALUE rb_yield(VALUE val)
-
-  val���ͤȤ��ƥ��ƥ졼���֥��å���ƤӽФ���
-
-VALUE rb_rescue(VALUE (*func1)(), VALUE arg1, VALUE (*func2)(), VALUE arg2)
-
-  �ؿ�func1��arg1������˸ƤӽФ���func1�μ¹�����㳰��ȯ��
-  �������ˤ� func2��arg2������Ȥ��ƸƤ֡�����ͤ��㳰��ȯ��
-  ���ʤ��ä�����func1�������, �㳰��ȯ���������ˤ�func2����
-  ���ͤǤ��롥
-
-VALUE rb_ensure(VALUE (*func1)(), VALUE arg1, void (*func2)(), VALUE arg2)
-
-  �ؿ�func1��arg1������Ȥ��Ƽ¹Ԥ�, �¹Խ�λ��(���Ȥ��㳰��
-  ȯ�����Ƥ�) func2��arg2������Ȥ��Ƽ¹Ԥ��롥����ͤ�func1
-  ������ͤǤ���(�㳰��ȯ�������������ʤ�)��
-
-** �㳰�����顼
-
-void rb_warning(const char *fmt, ...)
-
-  rb_verbose����ɸ�२�顼���Ϥ˷ٹ�����ɽ�����롥������
-  printf()��Ʊ����
-
-void rb_raise(rb_eRuntimeError, const char *fmt, ...)
-
-  RuntimeError�㳰��ȯ�������롥������printf()��Ʊ����
-
-void rb_raise(VALUE exception, const char *fmt, ...)
-
-  exception�ǻ��ꤷ���㳰��ȯ�������롥fmt�ʲ��ΰ�����
-  printf()��Ʊ����
-
-void rb_fatal(const char *fmt, ...)
-
-  ��̿Ū�㳰��ȯ�������롥�̾���㳰�����ϹԤʤ�줺, ���󥿡�
-  �ץ꥿����λ����(������ensure�ǻ��ꤵ�줿�����ɤϽ�λ����
-  �¹Ԥ����)��
-
-void rb_bug(const char *fmt, ...)
-
-  ���󥿡��ץ꥿�ʤɥץ������ΥХ��Ǥ���ȯ������Ϥ��Τʤ�
-  �����λ��Ƥ֡����󥿡��ץ꥿�ϥ�������פ�ľ���˽�λ���롥
-  �㳰�����ϰ��ڹԤʤ��ʤ���
-
-** Ruby�ν�������¹�
-
-Ruby�򥢥ץꥱ����������������ˤϰʲ��Υ��󥿥ե�����
-��Ȥ����̾�γ�ĥ�饤�֥��ˤ�ɬ�פʤ���
-
-void ruby_init()
-
-  Ruby���󥿥ץ꥿�ν������Ԥʤ���
-
-void ruby_options(int argc, char **argv)
-
-  Ruby���󥿥ץ꥿�Υ��ޥ�ɥ饤������ν�����Ԥʤ���
-
-void ruby_run()
-
-  Ruby���󥿥ץ꥿��¹Ԥ��롥
-
-void ruby_script(char *name)
-
-  Ruby�Υ�����ץ�̾($0)�����ꤹ�롥
-
-
-Appendix C. extconf.rb�ǻȤ���ؿ�����
-
-extconf.rb����Ǥ����Ѳ�ǽ�ʥ���ѥ���������å��δؿ��ϰ�
-�����̤�Ǥ��롥
-
-have_macro(macro, headers)
-
-  �إå��ե�����header�򥤥󥯥롼�ɤ��ƥޥ���macro�������
-  ��Ƥ��뤫�ɤ��������å����롥�ޥ������������Ƥ����true
-  ���֤���
-
-have_library(lib, func)
-
-  �ؿ�func��������Ƥ���饤�֥��lib��¸�ߤ�����å����롥
-  �饤�֥�꤬¸�ߤ������true���֤���
-
-find_library(lib, func, path...)
-
-  �ؿ�func��������Ƥ���饤�֥��lib��¸�ߤ� -Lpath ���ɲ�
-  ���ʤ�������å����롥�饤�֥�꤬���դ��ä�����true���֤���
-
-have_func(func, header)
-
-  �إå��ե�����header�򥤥󥯥롼�ɤ��ƴؿ�func��¸�ߤ������
-  �����롥func��ɸ��Ǥϥ�󥯤���ʤ��饤�֥����Τ�ΤǤ�
-  ����ˤ����have_library�Ǥ��Υ饤�֥�������å����Ƥ���
-  �����ؿ���¸�ߤ����true���֤���
-
-have_var(var, header)
-
-  �إå��ե�����header�򥤥󥯥롼�ɤ����ѿ�var��¸�ߤ������
-  �����롥var��ɸ��Ǥϥ�󥯤���ʤ��饤�֥����Τ�ΤǤ�
-  ����ˤ����have_library�Ǥ��Υ饤�֥�������å����Ƥ���
-  �����ѿ���¸�ߤ����true���֤���
-
-have_header(header)
-
-  �إå��ե������¸�ߤ�����å����롥�إå��ե����뤬¸�ߤ�
-  ���true���֤���
-
-find_header(header, path...)
-
-  �إå��ե�����header��¸�ߤ� -Ipath ���ɲä��ʤ�������å�
-  ���롥�إå��ե����뤬���դ��ä�����true���֤���
-
-have_struct_member(type, member, header)
-
-  �إå��ե�����header�򥤥󥯥롼�ɤ��Ʒ�type�˥���member
-  ��¸�ߤ��뤫������å����롥type���������Ƥ��ơ�member��
-  ���Ĥ����true���֤���
-
-have_type(type, header, opt)
-
-  �إå��ե�����header�򥤥󥯥롼�ɤ��Ʒ�type��¸�ߤ��뤫��
-  �����å����롥type���������Ƥ����true���֤���
-
-check_sizeof(type, header)
-
-  �إå��ե�����header�򥤥󥯥롼�ɤ��Ʒ�type��charñ�̥���
-  ����Ĵ�٤롥type���������Ƥ�������Υ��������֤��������
-  ��Ƥ��ʤ��Ȥ���nil���֤���
-
-create_makefile(target)
-
-  ��ĥ�饤�֥���Ѥ�Makefile���������롥���δؿ���ƤФʤ���
-  �Ф��Υ饤�֥��ϥ���ѥ��뤵��ʤ���target�ϥ⥸�塼��̾
-  ��ɽ����
-
-find_executable(command, path)
-
-  ���ޥ��command��File::PATH_SEPARATOR�Ƕ��ڤ�줿�ѥ�̾��
-  �ꥹ��path����õ����path��nil�ޤ��Ͼ�ά���줿���ϡ��Ķ�
-  �ѿ�PATH���ͤ���Ѥ��롥�¹Բ�ǽ�ʥ��ޥ�ɤ����Ĥ��ä����
-  �ϥѥ���ޤ�ե�����̾�����Ĥ���ʤ��ä�����nil���֤���
-
-with_config(withval[, default=nil])
-
-  ���ޥ�ɥ饤����--with-<withval>�ǻ��ꤵ�줿���ץ�����ͤ����롥
-
-enable_config(config, *defaults)
-disable_config(config, *defaults)
-
-  ���ޥ�ɥ饤����--enable-<config>�ޤ���
-  --disable-<config>�ǻ��ꤵ�줿�����ͤ����롥
-  --enable-<config>�����ꤵ��Ƥ�������true��
-  --disable-<config>�����ꤵ��Ƥ�������false���֤���
-  �ɤ������ꤵ��Ƥ��ʤ����ϡ��֥��å��Ĥ��ǸƤӽФ���Ƥ������
-  *defaults��yield������̡��֥��å��ʤ��ʤ�*defaults���֤���
-
-dir_config(target[, default_dir])
-dir_config(target[, default_include, default_lib])
-
-  ���ޥ�ɥ饤����--with-<target>-dir, --with-<target>-include,
-  --with-<target>-lib�Τ����줫�ǻ��ꤵ���ǥ��쥯�ȥ��
-  $CFLAGS �� $LDFLAGS ���ɲä��롥--with-<target>-dir=/path��
-  --with-<target>-include=/path/include --with-<target>-lib=/path/lib
-  �������Ǥ��롥�ɲä��줿 include �ǥ��쥯�ȥ�� lib �ǥ��쥯�ȥ��
-  ������֤��� ([include_dir, lib_dir])
-
-pkg_config(pkg)
-
-  pkg-config���ޥ�ɤ���ѥå�����pkg�ξ�������롥 
-  pkg-config�μºݤΥ��ޥ��̾�ϡ�--with-pkg-config���ޥ��
-  �饤�󥪥ץ����ǻ����ǽ��
-
-/*
- * Local variables:
- * fill-column: 60
- * end:
- */
+doc/extension.ja.rdocに移動しました
diff --git a/README.ja b/README.ja
deleted file mode 100644
index df8cfc03f4..0000000000
--- a/README.ja
+++ /dev/null
@@ -1,160 +0,0 @@
-* Ruby�Ȥ�
-
-Ruby�ϥ���ץ뤫�Ķ��Ϥʥ��֥������Ȼظ�������ץȸ���Ǥ���
-Ruby�Ϻǽ餫����ʥ��֥������Ȼظ�����Ȥ����߷פ���Ƥ���
-�����顤���֥������Ȼظ��ץ�����ߥ󥰤��ڤ˹Ԥ����������
-������������̾�μ�³�����Υץ�����ߥ󥰤��ǽ�Ǥ���
-
-Ruby�ϥƥ����Ƚ����ط���ǽ�Ϥʤɤ�ͥ�졤Perl��Ʊ�����餤����
-�Ǥ�������˥���ץ��ʸˡ�ȡ��㳰�����䥤�ƥ졼���ʤɤε���
-�ˤ�äơ����ʬ����䤹���ץ�����ߥ󥰤�����ޤ���
-
-
-* Ruby����Ĺ
-
-  + ����ץ��ʸˡ
-  + ���̤Υ��֥������Ȼظ���ǽ(���饹���᥽�åɥ�����ʤ�)
-  + �ü�ʥ��֥������Ȼظ���ǽ(Mixin, �ðۥ᥽�åɤʤ�)
-  + �黻�ҥ����С�������
-  + �㳰������ǽ
-  + ���ƥ졼���ȥ���������
-  + �����١������쥯��
-  + �����ʥߥå������ǥ��� (�������ƥ�����ˤ��)
-  + �ܿ������⤤��¿����UNIX���ư�������Ǥʤ���DOS��Windows��
-    Mac��BeOS�ʤɤξ�Ǥ�ư��
-
-
-* ����ˡ
-
-** FTP��
-
-�ʲ��ξ��ˤ����Ƥ���ޤ���
-
-  ftp://ftp.ruby-lang.org/pub/ruby/
-
-** Subversion��
-
-�ܥ֥�����Ruby�κǿ��Υ����������ɤϼ��Υ��ޥ�ɤǼ����Ǥ��ޤ���
-
-  $ svn co http://svn.ruby-lang.org/repos/ruby/branches/ruby_1_8_6/
-
-��ȯ��ü�Υ����������ɤϼ��Υ��ޥ�ɤǼ����Ǥ��ޤ���
-
-  $ svn co http://svn.ruby-lang.org/repos/ruby/trunk/ ruby
-
-¾�˳�ȯ��Υ֥����ΰ����ϼ��Υ��ޥ�ɤǸ����ޤ���
-
-  $ svn ls http://svn.ruby-lang.org/repos/ruby/branches/
-
-
-* �ۡ���ڡ���
-
-Ruby�Υۡ���ڡ�����URL��
-
-   http://www.ruby-lang.org/
-
-�Ǥ���
-
-
-* �᡼��󥰥ꥹ��
-
-Ruby�Υ᡼��󥰥ꥹ�Ȥ�����ޤ������ô�˾������
-
-   ruby-list-ctl@ruby-lang.org
-
-�ޤ���ʸ��
-
-   subscribe YourFirstName YourFamilyName
-   
-�Ƚ񤤤����äƲ������� 
-
-Ruby��ȯ�Ը����᡼��󥰥ꥹ�Ȥ⤢��ޤ���������Ǥ�ruby�Υ�
-��������λ��ͳ�ĥ�ʤɼ����������ˤĤ��Ƶ�������Ƥ��ޤ���
-���ô�˾������
-
-   ruby-dev-ctl@ruby-lang.org
-
-�ޤ�ruby-list��Ʊ�ͤ���ˡ�ǥ᡼�뤷�Ƥ��������� 
-
-Ruby��ĥ�⥸�塼��ˤĤ����ä��礦ruby-ext�᡼��󥰥ꥹ�Ȥ�
-���شط�������ˤĤ����ä��礦ruby-math�᡼��󥰥ꥹ�Ȥ�
-�Ѹ���ä��礦ruby-talk�᡼��󥰥ꥹ�Ȥ⤢��ޤ���������ˡ
-�Ϥɤ��Ʊ���Ǥ��� 
-
-
-* ����ѥ��롦���󥹥ȡ���
-
-�ʲ��μ��ǹԤäƤ���������
-
-  1. �⤷configure�ե����뤬���Ĥ���ʤ����⤷����
-     configure.in���Ť��褦�ʤ顢autoconf��¹Ԥ���
-     ������configure����������
-
-  2. configure��¹Ԥ���Makefile�ʤɤ���������
-
-  3. (ɬ�פʤ��)defines.h���Խ�����
-
-     ¿ʬ��ɬ��̵���Ȼפ��ޤ���
-
-  4. (ɬ�פʤ��)ext/Setup����Ū�˥�󥯤����ĥ�⥸�塼���
-     ���ꤹ��
-
-     ext/Setup�˵��Ҥ����⥸�塼�����Ū�˥�󥯤���ޤ���
-
-     �����ʥߥå������ǥ��󥰤򥵥ݡ��Ȥ��Ƥ��ʤ��������ƥ�
-     ����Ǥ�Setup��1���ܤΡ�option nodynamic�פȤ����ԤΥ�
-     ���Ȥ򳰤�ɬ�פ�����ޤ����ޤ������Υ������ƥ������
-     ��ĥ�⥸�塼������Ѥ��뤿��ˤϡ����餫������Ū�˥��
-     �����Ƥ���ɬ�פ�����ޤ���
-
-  5. make��¹Ԥ��ƥ���ѥ��뤹��
-
-  6. make test�ǥƥ��Ȥ�Ԥ���
-
-     ��test succeeded�פ�ɽ�������������Ǥ����������ƥ���
-     ���������Ƥⴰ�������ݾڤ���Ƥ������ǤϤ���ޤ���
-
-  7. make install
-
-     root�Ǻ�Ȥ���ɬ�פ����뤫�⤷��ޤ���
-
-�⤷������ѥ�����˥��顼��ȯ���������ˤϥ��顼�Υ����ȥ�
-����OS�μ����ޤ�Ǥ�������ܤ�����ݡ��Ȥ��Ԥ����äƤ�
-�������¾�����Τ���ˤ�ʤ�ޤ���
-
-
-* �ܿ�
-
-UNIX�Ǥ����configure���ۤȤ�ɤκ��ۤ�ۼ����Ƥ����Ϥ���
-�������פ�̸���Ȥ������ä����(����˰㤤�ʤ�)����Ԥˤ���
-���Ȥ��ݡ��Ȥ���С����Ǥ��뤫���Τ�ޤ���
-
-�������ƥ�����ˤ�äȤ��¸����Τ�GC���Ǥ���Ruby��GC���о�
-�Υ������ƥ����㤬setjmp()�ˤ�ä����ƤΥ쥸������ jmp_buf��
-��Ǽ���뤳�Ȥȡ�jmp_buf�ȥ����å���32bit���饤����Ȥ����
-���뤳�Ȥ��ꤷ�Ƥ��ޤ����ä����Ԥ���Ω���ʤ������б�����
-��˺���Ǥ��礦����Ԥβ������Ū��ñ�ǡ�gc.c�ǥ����å���
-�ޡ������Ƥ�����ʬ�˥��饤����ȤΥХ��ȿ��������餷�ƥޡ�
-�����륳���ɤ��ɲä�������ǺѤߤޤ�����defined(THINK_C)�פ�
-����Ƥ�����ʬ�򻲹ͤˤ��Ƥ�������
-
-# �ºݤˤ�Ruby��Think C�Ǥϥ���ѥ���Ǥ��ޤ���
-
-�쥸����������ɥ������CPU�Ǥϡ��쥸����������ɥ��򥹥���
-���˥ե�å��夹�륢����֥饳���ɤ��ɲä���ɬ�פ����뤫����
-��ޤ���
-
-
-* ���۾��
-
-COPYING.ja�ե�����򻲾Ȥ��Ƥ���������
-
-
-* ����
-
-�����ȡ��Х���ݡ��Ȥ���¾�� matz@netlab.jp �ޤǡ�
--------------------------------------------------------
-created at: Thu Aug  3 11:57:36 JST 1995
-Local variables:
-mode: indented-text
-end:
diff --git a/README.ja.md b/README.ja.md
new file mode 100644
index 0000000000..353f9b14cc
--- /dev/null
+++ b/README.ja.md
@@ -0,0 +1,174 @@
+[![Build Status](https://travis-ci.org/ruby/ruby.svg)](https://travis-ci.org/ruby/ruby)
+[![Build status](https://ci.appveyor.com/api/projects/status/0sy8rrxut4o0k960/branch/trunk?svg=true)](https://ci.appveyor.com/project/ruby/ruby/branch/trunk)
+
+# Rubyとは
+
+Rubyはシンプルかつ強力なオブジェクト指向スクリプト言語です． Rubyは純粋なオブジェクト指向言語として設計されているので，
+オブジェクト指向プログラミングを手軽に行う事が出来ます．もちろん普通の手続き型のプログラミングも可能です．
+
+Rubyはテキスト処理関係の能力などに優れ，Perlと同じくらい強力です．さらにシンプルな文法と，
+例外処理やイテレータなどの機構によって，より分かりやすいプログラミングが出来ます．
+
+## Rubyの特長
+
+*   シンプルな文法
+*   普通のオブジェクト指向機能(クラス，メソッドコールなど)
+*   特殊なオブジェクト指向機能(Mixin, 特異メソッドなど)
+*   演算子オーバーロード
+*   例外処理機能
+*   イテレータとクロージャ
+*   ガーベージコレクタ
+*   ダイナミックローディング (アーキテクチャによる)
+*   移植性が高い．多くのUnix-like/POSIX互換プラットフォーム上で動くだけでなく，Windows， macOS，
+    Haikuなどの上でも動く cf.
+    https://github.com/ruby/ruby/blob/trunk/doc/contributing.rdoc#platform-maintainers
+
+
+## 入手法
+
+サードパーティーツールを使った方法を含むRubyのインストール方法の一覧は
+
+https://www.ruby-lang.org/ja/downloads/
+
+を参照してください．
+
+### Subversionで
+
+開発先端のソースコードは次のコマンドで取得できます．
+
+    $ svn co https://svn.ruby-lang.org/repos/ruby/trunk/ ruby
+
+他に開発中のブランチの一覧は次のコマンドで見られます．
+
+    $ svn ls https://svn.ruby-lang.org/repos/ruby/branches/
+
+### Gitで
+
+SubversionのミラーをGitHubに公開しています． 以下のコマンドでリポジトリを取得できます．
+
+    $ git clone https://github.com/ruby/ruby.git
+
+## ホームページ
+
+RubyのホームページのURLは
+
+https://www.ruby-lang.org/
+
+です．
+
+## メーリングリスト
+
+Rubyのメーリングリストがあります．参加希望の方は
+
+mailto:ruby-list-request@ruby-lang.org
+
+まで本文に
+
+    subscribe
+
+と書いて送って下さい．
+
+Ruby開発者向けメーリングリストもあります．こちらではrubyのバグ，将来の仕様拡張など実装上の問題について議論されています． 参加希望の方は
+
+mailto:ruby-dev-request@ruby-lang.org
+
+までruby-listと同様の方法でメールしてください．
+
+Ruby拡張モジュールについて話し合うruby-extメーリングリストと数学関係の話題について話し合うruby-mathメーリングリストと
+英語でrubyについて話し合うruby-talkメーリングリストもあります．参加方法はどれも同じです．
+
+## コンパイル・インストール
+
+以下の手順で行ってください．
+
+1.  もし `configure` ファイルが見つからない，もしくは `configure.ac` より古いようなら， `autoconf` を実行して
+    新しく `configure` を生成する
+
+2.  `configure` を実行して `Makefile` などを生成する
+
+    環境によってはデフォルトのCコンパイラ用オプションが付きます． `configure` オプションで `optflags=..`
+    `warnflags=..` 等で上書きできます．
+
+3.  (必要ならば)`defines.h` を編集する
+
+    多分，必要無いと思います．
+
+4.  (必要ならば)`ext/Setup` に静的にリンクする拡張モジュールを指定する
+
+    `ext/Setup` に記述したモジュールは静的にリンクされます．
+
+    ダイナミックローディングをサポートしていないアーキテクチャでは `Setup` の1行目の「`option nodynamic`」という行のコ
+    メントを外す必要があります．また，このアーキテクチャで拡張モジュールを利用するためには，あらかじめ静的にリンクをしておく必要があります．
+
+5.  `make` を実行してコンパイルする
+
+6.  `make check`でテストを行う．
+
+    「`check succeeded`」と表示されれば成功です．ただしテストに成功しても完璧だと保証されている訳ではありません．
+
+7.  `make install`
+
+    以下のディレクトリを作って，そこにファイルをインストー ルします．
+
+    *   `${DESTDIR}${prefix}/bin`
+    *   `${DESTDIR}${prefix}/include/ruby-${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/include/ruby-${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib`
+    *   `${DESTDIR}${prefix}/lib/ruby`
+    *   `${DESTDIR}${prefix}/lib/ruby/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/lib/ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib/ruby/site_ruby`
+    *   `${DESTDIR}${prefix}/lib/ruby/site_ruby/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/lib/ruby/site_ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib/ruby/vendor_ruby`
+    *   `${DESTDIR}${prefix}/lib/ruby/vendor_ruby/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/lib/ruby/vendor_ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib/ruby/gems/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/share/man/man1`
+    *   `${DESTDIR}${prefix}/share/ri/${MAJOR}.${MINOR}.${TEENY}/system`
+
+
+    RubyのAPIバージョンが'*x.y.z*'であれば，`${MAJOR}`は
+    '*x*'で，`${MINOR}`は'*y*'，`${TEENY}`は'*z*'です．
+
+    **注意**: APIバージョンの `teeny` は，Rubyプログラムのバージョンとは異なることがあります．
+
+    `root` で作業する必要があるかもしれません．
+
+
+もし，コンパイル時にエラーが発生した場合にはエラーのログとマシン，OSの種類を含むできるだけ詳しいレポートを作者に送って下さると他の方のためにもなります．
+
+## 移植
+
+UNIXであれば `configure` がほとんどの差異を吸収してくれるはずですが，思わぬ見落としがあった場合(ある事が多い)，作者にその
+ことを報告すれば，解決できる可能性があります．
+
+アーキテクチャにもっとも依存するのはGC部です．RubyのGCは対象
+のアーキテクチャが`setjmp()`または`getcontext()`によって全てのレジスタを `jmp_buf` や `ucontext_t`
+に格納することと， `jmp_buf` や `ucontext_t` とスタックが32bitアラインメントされていることを仮定
+しています．特に前者が成立しない場合の対応は非常に困難でしょう． 後者の解決は比較的簡単で， `gc.c` でスタックをマークしている
+部分にアラインメントのバイト数だけずらしてマークするコードを追加するだけで済みます．`defined(__mc68000__)`で括られてい
+る部分を参考にしてください．
+
+レジスタウィンドウを持つCPUでは，レジスタウィンドウをスタックにフラッシュするアセンブラコードを追加する必要があるかもしれません．
+
+## 配布条件
+
+[COPYING.ja](COPYING.ja) ファイルを参照してください．
+
+## フィードバック
+
+Rubyに関する質問は Ruby-Talk（英語）や Ruby-List（日本語） (https://www.ruby-lang.org/ja/community/mailing-lists) や，
+stackoverflow (https://ja.stackoverflow.com/) などのWebサイトに投稿してください．
+
+バグ報告は https://bugs.ruby-lang.org で受け付けています．
+
+
+## 著者
+
+Rubyのオリジナル版は，1995年にまつもとゆきひろ氏によって設計・開発されました．
+
+<mailto:matz@ruby-lang.org>
+
+---
+created at: Thu Aug  3 11:57:36 JST 1995
diff --git a/README.md b/README.md
new file mode 100644
index 0000000000..1e76fda4f0
--- /dev/null
+++ b/README.md
@@ -0,0 +1,172 @@
+[![Build Status](https://travis-ci.org/ruby/ruby.svg?branch=trunk)](https://travis-ci.org/ruby/ruby)
+[![Build status](https://ci.appveyor.com/api/projects/status/0sy8rrxut4o0k960/branch/trunk?svg=true)](https://ci.appveyor.com/project/ruby/ruby/branch/trunk)
+[![wercker status](https://app.wercker.com/status/e5e7e1704f62b76525022aa424aef6ef/s/trunk "wercker status")](https://app.wercker.com/project/byKey/e5e7e1704f62b76525022aa424aef6ef)
+
+# What's Ruby
+
+Ruby is the interpreted scripting language for quick and easy object-oriented
+programming. It has many features to process text files and to do system
+management tasks (as in Perl). It is simple, straight-forward, and
+extensible.
+
+## Features of Ruby
+
+*   Simple Syntax
+*   **Normal** Object-oriented Features (e.g. class, method calls)
+*   **Advanced** Object-oriented Features (e.g. mix-in, singleton-method)
+*   Operator Overloading
+*   Exception Handling
+*   Iterators and Closures
+*   Garbage Collection
+*   Dynamic Loading of Object Files (on some architectures)
+*   Highly Portable (works on many Unix-like/POSIX compatible platforms as
+    well as Windows, macOS, Haiku, etc.) cf.
+    https://github.com/ruby/ruby/blob/trunk/doc/contributing.rdoc#platform-maintainers
+
+
+## How to get Ruby
+
+For a complete list of ways to install Ruby, including using third-party tools
+like rvm, see:
+
+https://www.ruby-lang.org/en/downloads/
+
+The trunk of the Ruby source tree can be checked out with the following
+command:
+
+    $ svn co https://svn.ruby-lang.org/repos/ruby/trunk/ ruby
+
+Or if you are using git then use the following command:
+
+    $ git clone https://github.com/ruby/ruby.git
+
+There are some other branches under development. Try the following command
+to see the list of branches:
+
+    $ svn ls https://svn.ruby-lang.org/repos/ruby/branches/
+
+Or if you are using git then use the following command:
+
+    $ git ls-remote https://github.com/ruby/ruby.git
+
+## Ruby home page
+
+The URL of the Ruby home page is:
+
+https://www.ruby-lang.org/
+
+## Mailing list
+
+There is a mailing list to talk about Ruby. To subscribe to this list, please
+send the following phrase:
+
+    subscribe
+
+in the mail body (not subject) to the address
+<ruby-talk-request@ruby-lang.org>.
+
+## How to compile and install
+
+This is what you need to do to compile and install Ruby:
+
+1.  If you want to use Microsoft Visual C++ to compile Ruby, read
+    [win32/README.win32](win32/README.win32) instead of this document.
+
+2.  If `./configure` does not exist or is older than `configure.ac`, run
+    `autoconf` to (re)generate configure.
+
+3.  Run `./configure`, which will generate `config.h` and `Makefile`.
+
+    Some C compiler flags may be added by default depending on your
+    environment. Specify `optflags=..` and `warnflags=..` as necessary to
+    override them.
+
+4.  Edit `defines.h` if you need. Usually this step will not be needed.
+
+5.  Remove comment mark(`#`) before the module names from `ext/Setup` (or add
+    module names if not present), if you want to link modules statically.
+
+    If you don't want to compile non static extension modules (probably on
+    architectures which do not allow dynamic loading), remove comment mark
+    from the line "`#option nodynamic`" in `ext/Setup`.
+
+    Usually this step will not be needed.
+
+6.  Run `make`.
+
+    * On Mac, set RUBY\_CODESIGN environment variable with a signing identity.
+      It uses the identity to sign `ruby` binary. See also codesign(1).
+
+7.  Optionally, run '`make check`' to check whether the compiled Ruby
+    interpreter works well. If you see the message "`check succeeded`", your
+    Ruby works as it should (hopefully).
+
+8.  Optionally, run `make update-gems` and `make extract-gems`.
+
+    If you want to install bundled gems, run `make update-gems` and
+    `make extract-gems` before running `make install`.
+
+9.  Run '`make install`'.
+
+    This command will create the following directories and install files into
+    them.
+
+    *   `${DESTDIR}${prefix}/bin`
+    *   `${DESTDIR}${prefix}/include/ruby-${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/include/ruby-${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib`
+    *   `${DESTDIR}${prefix}/lib/ruby`
+    *   `${DESTDIR}${prefix}/lib/ruby/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/lib/ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib/ruby/site_ruby`
+    *   `${DESTDIR}${prefix}/lib/ruby/site_ruby/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/lib/ruby/site_ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib/ruby/vendor_ruby`
+    *   `${DESTDIR}${prefix}/lib/ruby/vendor_ruby/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/lib/ruby/vendor_ruby/${MAJOR}.${MINOR}.${TEENY}/${PLATFORM}`
+    *   `${DESTDIR}${prefix}/lib/ruby/gems/${MAJOR}.${MINOR}.${TEENY}`
+    *   `${DESTDIR}${prefix}/share/man/man1`
+    *   `${DESTDIR}${prefix}/share/ri/${MAJOR}.${MINOR}.${TEENY}/system`
+
+
+    If Ruby's API version is '*x.y.z*', the `${MAJOR}` is '*x*', the
+    `${MINOR}` is '*y*', and the `${TEENY}` is '*z*'.
+
+    **NOTE**: teeny of the API version may be different from one of Ruby's
+    program version
+
+    You may have to be a super user to install Ruby.
+
+
+If you fail to compile Ruby, please send the detailed error report with the
+error log and machine/OS type, to help others.
+
+Some extension libraries may not get compiled because of lack of necessary
+external libraries and/or headers, then you will need to run '`make distclean-ext`'
+to remove old configuration after installing them in such case.
+
+## Copying
+
+See the file [COPYING](COPYING).
+
+## Feedback
+
+Questions about the Ruby language can be asked on the Ruby-Talk mailing list
+(https://www.ruby-lang.org/en/community/mailing-lists) or on websites like
+(https://stackoverflow.com).
+
+Bug reports should be filed at https://bugs.ruby-lang.org. Read [HowToReport] for more information.
+
+[HowToReport]: https://bugs.ruby-lang.org/projects/ruby/wiki/HowToReport
+
+## Contributing
+
+See the file [CONTRIBUTING.md](CONTRIBUTING.md)
+
+
+## The Author
+
+Ruby was originally designed and developed by Yukihiro Matsumoto (Matz) in
+1995.
+
+<matz@ruby-lang.org>
diff --git a/ToDo b/ToDo
deleted file mode 100644
index b55e399edf..0000000000
--- a/ToDo
+++ /dev/null
@@ -1,126 +0,0 @@
-Language Spec.
-
-- Class#allocate - basicNew
-- class Foo::Bar<Baz .. end, module Boo::Bar .. end
-* operator !! for rescue. ???
-* objectify characters
-* ../... outside condition invokes operator method too.
-* ... inside condition turns off just before right condition.???
-* package or access control for global variables??
-* named arguments like foo(nation:="german") or foo(nation: "german").
-* method to retrieve argument information (needs new C API)
-* multiple return values, yield values.  maybe incompatible ???
-* cascading method invocation ???
-* def Class#method .. end ??
-* def Foo::Bar::baz() .. end ??
-* I18N (or M17N) script/string/regexp
-* Fixnum 0 as false ????
-* discourage use of symbol variables (e.g. $/, etc.) in manual
-* discourage use of Perlish features by giving warnings.
-* non confusing in-block local variable (is it possible?)
-  + remove scope by block
-  + variables appears within block may have independent values.
-* Regexp: make /o thread safe.
-* decide whether begin with rescue or ensure make do..while loop.
-* a +1 to be a+1, not a(+1).
-* unify == and eql? again
-* to_i returns nil if str contains no digit.
-* raise exception by `` error
-* jar like combined library package. -> RubyGems?
-* resumable Exception via Exception#resume.
-* method combination, e.g. before, after, around, etc.
-* .. or something like defadvice in Emacs.
-* property - for methods, or for objects in general.
-* "in" modifier, to annotate, or to encourage assertion.
-* selector namespace - something like generic-flet in CLOS, to help RubyBehavior
-* private instance variable (as in Python?) @_foo in class Foo => @_Foo_foo
-* warn/error "bare word" method, like "foo",  you should type "foo()"
-* clarify evaluation order of operator argument (=~, .., ...)
-* :symbol => value hash in the form of {symbol: value, ...} ??
-
-Hacking Interpreter
-
-- generational GC
-* non-blocking open (e.g. for named pipe) for thread
-* avoid blocking with gethostbyname/gethostbyaddr (use fork ???)
-* objectify interpreters ???
-* remove rb_eval() recursions
-* syntax tree -> bytecode ???
-* scrambled script, or script filter
-* setuid ruby
-* performance tune for in-block (dynamic) local variables.
-* give warnings to assign magic variables.
-* export rb_io_{addstr,printf,puts,print}
-* autoload should work with threads [ruby-talk:4589]
-* remove stdio dependency from IOs.
-* warn for inconsistent local variable usage (lv m and method m at the same time). 
-* MicroRuby
-* Built-in Interactive Ruby.
-* Parser API
-* trap every method invocation, which can be enabled by e.g. trap_call :method.
-* unify Errno exceptions of same errno, or new exception comparison scheme.
-* 2.times{|i| if i==0 then a = 15 else puts eval("a") end} should print nil.
-* Thread#max_stack_size attribute (possible??)
-
-Standard Libraries
-
-- Module#define_method which takes a name and a body (block, proc or method).
-- Enume#inject
-- Array#fetch
-- IO::for_fd
-- Process::waitall [ruby-talk:4557]
-- Process::Status
-- File::lchown, File::lchmod; xxx - still need work for non existing platforms
-- move Time::times to Process.
-- Enumerable#sort_by for Schwartzian transformation
-- fork_and_kill_other_threads.
-- signal list (Signal::trap, Signal::list).
-- move NameError under StandardError.
-- Integer#to_s(base)
-- Hash::new{default}
-- hash etc. should handle self referenceing array/hash
-- Array#select(n1,n2...) works like Array#indexes(n1,n2...)
-- use Mersenne Twister RNG for random.
-- deprecate Array#indexes, and Array#indices.
-- remove dependency on MAXPATHLEN.
-* String#scanf(?)
-* Object#fmt(?)
-* Time::strptime
-* Integer[num], Float[num];  Fixnum[num]?
-* method to retrieve non-number trailer for to_i/to_f.
-* Stream or Port, abstract superclass of IO ?
-* String#{pred,prev}, String#downto
-* optional stepsize argument for succ()
-* Ruby module -- Ruby::Version, Ruby::Interpreter
-* introduce Boolean class; super of TrueClass, FalseClass
-* synchronized method - synchronized{...}, synchronized :foo, :bar
-* Array#&, Array#| to allow duplication. ???
-* way to specify immortal (fork endurance) thread;
-* or raise ForkException to every thread but fork caller.
-* new user-defined marshal scheme. _dump(dumper), _load(restorer)
-* library to load per-user profile seeking .ruby_profile or ruby.ini file.
-* warning framework (warn, warning for Ruby level)
-* marshal should not depend on sprintf (works bad with locale).
-* ternary arg pow: a.pow(b,c) == a**b%c
-* new caller(), e.g. call_stack; needs better name.
-* pointer share mechanism similar to one in String for Array.
-* require "1.6" etc. by /usr/lib/ruby/1.6/1.6.rb ;-)
-* save both "feature names" and "normalized path" in $"
-* implement Mutex_m (or MutexMixin) using Mutex.
-
-Extension Libraries
-
-* ptk.rb pTk wrapper that is compatible to tk.rb
-* Berkeley DB extension
-* BitVector
-* thread-safe fcgi
-
-Ruby Libraries
-
-* urllib.rb, nttplib.rb, etc.
-* format like perl's
-
-Tools
-
-* freeze or undump to bundle everything
-* bundle using zlib
diff --git a/aclocal.m4 b/aclocal.m4
new file mode 100644
index 0000000000..b0fe3eb959
--- /dev/null
+++ b/aclocal.m4
@@ -0,0 +1,46 @@
+# generated automatically by aclocal 1.16.1 -*- Autoconf -*-
+
+# Copyright (C) 1996-2018 Free Software Foundation, Inc.
+
+# This file is free software; the Free Software Foundation
+# gives unlimited permission to copy and/or distribute it,
+# with or without modifications, as long as this notice is preserved.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY, to the extent permitted by law; without
+# even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+# PARTICULAR PURPOSE.
+
+m4_ifndef([AC_CONFIG_MACRO_DIRS], [m4_defun([_AM_CONFIG_MACRO_DIRS], [])m4_defun([AC_CONFIG_MACRO_DIRS], [_AM_CONFIG_MACRO_DIRS($@)])])
+m4_include([tool/m4/_colorize_result_prepare.m4])
+m4_include([tool/m4/ac_msg_result.m4])
+m4_include([tool/m4/colorize_result.m4])
+m4_include([tool/m4/ruby_append_option.m4])
+m4_include([tool/m4/ruby_append_options.m4])
+m4_include([tool/m4/ruby_check_builtin_func.m4])
+m4_include([tool/m4/ruby_check_builtin_setjmp.m4])
+m4_include([tool/m4/ruby_check_printf_prefix.m4])
+m4_include([tool/m4/ruby_check_setjmp.m4])
+m4_include([tool/m4/ruby_check_signedness.m4])
+m4_include([tool/m4/ruby_check_sizeof.m4])
+m4_include([tool/m4/ruby_check_sysconf.m4])
+m4_include([tool/m4/ruby_cppoutfile.m4])
+m4_include([tool/m4/ruby_decl_attribute.m4])
+m4_include([tool/m4/ruby_default_arch.m4])
+m4_include([tool/m4/ruby_define_if.m4])
+m4_include([tool/m4/ruby_defint.m4])
+m4_include([tool/m4/ruby_dtrace_available.m4])
+m4_include([tool/m4/ruby_dtrace_postprocess.m4])
+m4_include([tool/m4/ruby_func_attribute.m4])
+m4_include([tool/m4/ruby_mingw32.m4])
+m4_include([tool/m4/ruby_prepend_option.m4])
+m4_include([tool/m4/ruby_prog_gnu_ld.m4])
+m4_include([tool/m4/ruby_replace_type.m4])
+m4_include([tool/m4/ruby_rm_recursive.m4])
+m4_include([tool/m4/ruby_setjmp_type.m4])
+m4_include([tool/m4/ruby_stack_grow_direction.m4])
+m4_include([tool/m4/ruby_try_cflags.m4])
+m4_include([tool/m4/ruby_try_ldflags.m4])
+m4_include([tool/m4/ruby_type_attribute.m4])
+m4_include([tool/m4/ruby_universal_arch.m4])
+m4_include([tool/m4/ruby_werror_flag.m4])
diff --git a/addr2line.c b/addr2line.c
new file mode 100644
index 0000000000..a11d32b73d
--- /dev/null
+++ b/addr2line.c
@@ -0,0 +1,2650 @@
+/**********************************************************************
+
+  addr2line.c -
+
+  $Author$
+
+  Copyright (C) 2010 Shinichiro Hamaji
+
+**********************************************************************/
+
+#if defined(__clang__)
+#pragma clang diagnostic ignored "-Wpedantic"
+#pragma clang diagnostic ignored "-Wgcc-compat"
+#elif defined(__GNUC__)
+#pragma GCC diagnostic ignored "-Wpedantic"
+#endif
+
+#include "ruby/config.h"
+#include "ruby/defines.h"
+#include "ruby/missing.h"
+#include "addr2line.h"
+
+#include <stdio.h>
+#include <errno.h>
+
+#ifdef HAVE_STDBOOL_H
+#include <stdbool.h>
+#else
+#include "missing/stdbool.h"
+#endif
+
+#if defined(USE_ELF) || defined(HAVE_MACH_O_LOADER_H)
+
+#include <fcntl.h>
+#include <limits.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+
+/* Make alloca work the best possible way.  */
+#ifdef __GNUC__
+# ifndef alloca
+#  define alloca __builtin_alloca
+# endif
+#else
+# ifdef HAVE_ALLOCA_H
+#  include <alloca.h>
+# else
+#  ifdef _AIX
+#pragma alloca
+#  else
+#   ifndef alloca		/* predefined by HP cc +Olibcalls */
+void *alloca();
+#   endif
+#  endif /* AIX */
+# endif	/* HAVE_ALLOCA_H */
+#endif /* __GNUC__ */
+
+#ifdef HAVE_DLADDR
+# include <dlfcn.h>
+#endif
+
+#ifdef HAVE_MACH_O_LOADER_H
+# include <mach-o/fat.h>
+# include <mach-o/ldsyms.h>
+# include <mach-o/loader.h>
+# include <mach-o/nlist.h>
+# include <mach-o/stab.h>
+#endif
+
+#ifdef USE_ELF
+# ifdef __OpenBSD__
+#  include <elf_abi.h>
+# else
+#  include <elf.h>
+# endif
+
+#ifndef ElfW
+# if SIZEOF_VOIDP == 8
+#  define ElfW(x) Elf64##_##x
+# else
+#  define ElfW(x) Elf32##_##x
+# endif
+#endif
+#ifndef ELF_ST_TYPE
+# if SIZEOF_VOIDP == 8
+#  define ELF_ST_TYPE ELF64_ST_TYPE
+# else
+#  define ELF_ST_TYPE ELF32_ST_TYPE
+# endif
+#endif
+#endif
+
+#ifdef SHF_COMPRESSED
+# if defined(ELFCOMPRESS_ZLIB) && defined(HAVE_LIBZ)
+   /* FreeBSD 11.0 lacks ELFCOMPRESS_ZLIB */
+#  include <zlib.h>
+#  define SUPPORT_COMPRESSED_DEBUG_LINE
+# endif
+#else /* compatibility with glibc < 2.22 */
+# define SHF_COMPRESSED 0
+#endif
+
+#ifndef PATH_MAX
+#define PATH_MAX 4096
+#endif
+
+#define DW_LNS_copy                     0x01
+#define DW_LNS_advance_pc               0x02
+#define DW_LNS_advance_line             0x03
+#define DW_LNS_set_file                 0x04
+#define DW_LNS_set_column               0x05
+#define DW_LNS_negate_stmt              0x06
+#define DW_LNS_set_basic_block          0x07
+#define DW_LNS_const_add_pc             0x08
+#define DW_LNS_fixed_advance_pc         0x09
+#define DW_LNS_set_prologue_end         0x0a /* DWARF3 */
+#define DW_LNS_set_epilogue_begin       0x0b /* DWARF3 */
+#define DW_LNS_set_isa                  0x0c /* DWARF3 */
+
+/* Line number extended opcode name. */
+#define DW_LNE_end_sequence             0x01
+#define DW_LNE_set_address              0x02
+#define DW_LNE_define_file              0x03
+#define DW_LNE_set_discriminator        0x04  /* DWARF4 */
+
+PRINTF_ARGS(static int kprintf(const char *fmt, ...), 1, 2);
+
+typedef struct line_info {
+    const char *dirname;
+    const char *filename;
+    const char *path; /* object path */
+    int line;
+
+    uintptr_t base_addr;
+    uintptr_t saddr;
+    const char *sname; /* function name */
+
+    struct line_info *next;
+} line_info_t;
+
+struct dwarf_section {
+    char *ptr;
+    size_t size;
+    uint64_t flags;
+};
+
+typedef struct obj_info {
+    const char *path; /* object path */
+    char *mapped;
+    size_t mapped_size;
+    void *uncompressed;
+    uintptr_t base_addr;
+    uintptr_t vmaddr;
+    struct dwarf_section debug_abbrev;
+    struct dwarf_section debug_info;
+    struct dwarf_section debug_line;
+    struct dwarf_section debug_ranges;
+    struct dwarf_section debug_str;
+    struct obj_info *next;
+} obj_info_t;
+
+#define DWARF_SECTION_COUNT 5
+
+static struct dwarf_section *
+obj_dwarf_section_at(obj_info_t *obj, int n)
+{
+    struct dwarf_section *ary[] = {
+        &obj->debug_abbrev,
+        &obj->debug_info,
+        &obj->debug_line,
+        &obj->debug_ranges,
+        &obj->debug_str
+    };
+    if (n < 0 || DWARF_SECTION_COUNT <= n) {
+        abort();
+    }
+    return ary[n];
+}
+
+struct debug_section_definition {
+    const char *name;
+    struct dwarf_section *dwarf;
+};
+
+/* Avoid consuming stack as this module may be used from signal handler */
+static char binary_filename[PATH_MAX];
+
+static unsigned long
+uleb128(char **p)
+{
+    unsigned long r = 0;
+    int s = 0;
+    for (;;) {
+	unsigned char b = *(unsigned char *)(*p)++;
+	if (b < 0x80) {
+	    r += (unsigned long)b << s;
+	    break;
+	}
+	r += (b & 0x7f) << s;
+	s += 7;
+    }
+    return r;
+}
+
+static long
+sleb128(char **p)
+{
+    long r = 0;
+    int s = 0;
+    for (;;) {
+	unsigned char b = *(unsigned char *)(*p)++;
+	if (b < 0x80) {
+	    if (b & 0x40) {
+		r -= (0x80 - b) << s;
+	    }
+	    else {
+		r += (b & 0x3f) << s;
+	    }
+	    break;
+	}
+	r += (b & 0x7f) << s;
+	s += 7;
+    }
+    return r;
+}
+
+static const char *
+get_nth_dirname(unsigned long dir, char *p)
+{
+    if (!dir--) {
+	return "";
+    }
+    while (dir--) {
+	while (*p) p++;
+	p++;
+	if (!*p) {
+	    kprintf("Unexpected directory number %lu in %s\n",
+		    dir, binary_filename);
+	    return "";
+	}
+    }
+    return p;
+}
+
+static void
+fill_filename(int file, char *include_directories, char *filenames, line_info_t *line, obj_info_t *obj)
+{
+    int i;
+    char *p = filenames;
+    char *filename;
+    unsigned long dir;
+    for (i = 1; i <= file; i++) {
+	filename = p;
+	if (!*p) {
+	    /* Need to output binary file name? */
+	    kprintf("Unexpected file number %d in %s at %tx\n",
+		    file, binary_filename, filenames - obj->mapped);
+	    return;
+	}
+	while (*p) p++;
+	p++;
+	dir = uleb128(&p);
+	/* last modified. */
+	uleb128(&p);
+	/* size of the file. */
+	uleb128(&p);
+
+	if (i == file) {
+	    line->filename = filename;
+	    line->dirname = get_nth_dirname(dir, include_directories);
+	}
+    }
+}
+
+static void
+fill_line(int num_traces, void **traces, uintptr_t addr, int file, int line,
+	  char *include_directories, char *filenames,
+	  obj_info_t *obj, line_info_t *lines, int offset)
+{
+    int i;
+    addr += obj->base_addr - obj->vmaddr;
+    for (i = offset; i < num_traces; i++) {
+	uintptr_t a = (uintptr_t)traces[i];
+	/* We assume one line code doesn't result >100 bytes of native code.
+       We may want more reliable way eventually... */
+	if (addr < a && a < addr + 100) {
+	    fill_filename(file, include_directories, filenames, &lines[i], obj);
+	    lines[i].line = line;
+	}
+    }
+}
+
+struct LineNumberProgramHeader {
+    uint64_t unit_length;
+    uint16_t version;
+    uint8_t format; /* 4 or 8 */
+    uint64_t header_length;
+    uint8_t minimum_instruction_length;
+    uint8_t maximum_operations_per_instruction;
+    uint8_t default_is_stmt;
+    int8_t line_base;
+    uint8_t line_range;
+    uint8_t opcode_base;
+    /* uint8_t standard_opcode_lengths[opcode_base-1]; */
+    const char *include_directories;
+    const char *filenames;
+    const char *cu_start;
+    const char *cu_end;
+};
+
+static int
+parse_debug_line_header(const char **pp, struct LineNumberProgramHeader *header)
+{
+    const char *p = *pp;
+    header->unit_length = *(uint32_t *)p;
+    p += sizeof(uint32_t);
+
+    header->format = 4;
+    if (header->unit_length == 0xffffffff) {
+	header->unit_length = *(uint64_t *)p;
+	p += sizeof(uint64_t);
+        header->format = 8;
+    }
+
+    header->cu_end = p + header->unit_length;
+
+    header->version = *(uint16_t *)p;
+    p += sizeof(uint16_t);
+    if (header->version > 4) return -1;
+
+    header->header_length = header->format == 4 ? *(uint32_t *)p : *(uint64_t *)p;
+    p += header->format;
+    header->cu_start = p + header->header_length;
+
+    header->minimum_instruction_length = *(uint8_t *)p++;
+
+    if (header->version >= 4) {
+        /* maximum_operations_per_instruction = *(uint8_t *)p; */
+        if (*p != 1) return -1; /* For non-VLIW architectures, this field is 1 */
+        p++;
+    }
+
+    header->default_is_stmt = *(uint8_t *)p++;
+    header->line_base = *(int8_t *)p++;
+    header->line_range = *(uint8_t *)p++;
+    header->opcode_base = *(uint8_t *)p++;
+    /* header->standard_opcode_lengths = (uint8_t *)p - 1; */
+    p += header->opcode_base - 1;
+
+    header->include_directories = p;
+
+    /* temporary measure for compress-debug-sections */
+    if (p >= header->cu_end) return -1;
+
+    /* skip include directories */
+    while (*p) {
+	p = memchr(p, '\0', header->cu_end - p);
+	if (!p) return -1;
+	p++;
+    }
+    p++;
+
+    header->filenames = p;
+
+    *pp = header->cu_start;
+
+    return 0;
+}
+
+static int
+parse_debug_line_cu(int num_traces, void **traces, char **debug_line,
+		obj_info_t *obj, line_info_t *lines, int offset)
+{
+    const char *p = (const char *)*debug_line;
+    struct LineNumberProgramHeader header;
+
+    /* The registers. */
+    unsigned long addr = 0;
+    unsigned int file = 1;
+    unsigned int line = 1;
+    /* unsigned int column = 0; */
+    int is_stmt;
+    /* int basic_block = 0; */
+    /* int end_sequence = 0; */
+    /* int prologue_end = 0; */
+    /* int epilogue_begin = 0; */
+    /* unsigned int isa = 0; */
+
+    if (parse_debug_line_header(&p, &header))
+        return -1;
+    is_stmt = header.default_is_stmt;
+
+#define FILL_LINE()						    \
+    do {							    \
+	fill_line(num_traces, traces, addr, file, line,		    \
+                  (char *)header.include_directories,               \
+                  (char *)header.filenames,                         \
+		  obj, lines, offset);				    \
+	/*basic_block = prologue_end = epilogue_begin = 0;*/	    \
+    } while (0)
+
+    while (p < header.cu_end) {
+	unsigned long a;
+	unsigned char op = *p++;
+	switch (op) {
+	case DW_LNS_copy:
+	    FILL_LINE();
+	    break;
+	case DW_LNS_advance_pc:
+	    a = uleb128((char **)&p);
+	    addr += a;
+	    break;
+	case DW_LNS_advance_line: {
+	    long a = sleb128((char **)&p);
+	    line += a;
+	    break;
+	}
+	case DW_LNS_set_file:
+	    file = (unsigned int)uleb128((char **)&p);
+	    break;
+	case DW_LNS_set_column:
+	    /*column = (unsigned int)*/(void)uleb128((char **)&p);
+	    break;
+	case DW_LNS_negate_stmt:
+	    is_stmt = !is_stmt;
+	    break;
+	case DW_LNS_set_basic_block:
+	    /*basic_block = 1; */
+	    break;
+	case DW_LNS_const_add_pc:
+	    a = ((255 - header.opcode_base) / header.line_range) *
+		header.minimum_instruction_length;
+	    addr += a;
+	    break;
+	case DW_LNS_fixed_advance_pc:
+	    a = *(unsigned char *)p++;
+	    addr += a;
+	    break;
+	case DW_LNS_set_prologue_end:
+	    /* prologue_end = 1; */
+	    break;
+	case DW_LNS_set_epilogue_begin:
+	    /* epilogue_begin = 1; */
+	    break;
+	case DW_LNS_set_isa:
+	    /* isa = (unsigned int)*/(void)uleb128((char **)&p);
+	    break;
+	case 0:
+	    a = *(unsigned char *)p++;
+	    op = *p++;
+	    switch (op) {
+	    case DW_LNE_end_sequence:
+		/* end_sequence = 1; */
+		FILL_LINE();
+		addr = 0;
+		file = 1;
+		line = 1;
+		/* column = 0; */
+		is_stmt = header.default_is_stmt;
+		/* end_sequence = 0; */
+		/* isa = 0; */
+		break;
+	    case DW_LNE_set_address:
+		addr = *(unsigned long *)p;
+		p += sizeof(unsigned long);
+		break;
+	    case DW_LNE_define_file:
+		kprintf("Unsupported operation in %s\n",
+			binary_filename);
+		break;
+	    case DW_LNE_set_discriminator:
+		/* TODO:currently ignore */
+		uleb128((char **)&p);
+		break;
+	    default:
+		kprintf("Unknown extended opcode: %d in %s\n",
+			op, binary_filename);
+	    }
+	    break;
+	default: {
+            uint8_t adjusted_opcode = op - header.opcode_base;
+            uint8_t operation_advance = adjusted_opcode / header.line_range;
+            /* NOTE: this code doesn't support VLIW */
+            addr += operation_advance * header.minimum_instruction_length;
+            line += header.line_base + (adjusted_opcode % header.line_range);
+	    FILL_LINE();
+	}
+	}
+    }
+    *debug_line = (char *)p;
+    return 0;
+}
+
+static int
+parse_debug_line(int num_traces, void **traces,
+		 char *debug_line, unsigned long size,
+		 obj_info_t *obj, line_info_t *lines, int offset)
+{
+    char *debug_line_end = debug_line + size;
+    while (debug_line < debug_line_end) {
+	if (parse_debug_line_cu(num_traces, traces, &debug_line, obj, lines, offset))
+	    return -1;
+    }
+    if (debug_line != debug_line_end) {
+	kprintf("Unexpected size of .debug_line in %s\n",
+		binary_filename);
+    }
+    return 0;
+}
+
+/* read file and fill lines */
+static uintptr_t
+fill_lines(int num_traces, void **traces, int check_debuglink,
+	   obj_info_t **objp, line_info_t *lines, int offset);
+
+static void
+append_obj(obj_info_t **objp)
+{
+    obj_info_t *newobj = calloc(1, sizeof(obj_info_t));
+    if (*objp) (*objp)->next = newobj;
+    *objp = newobj;
+}
+
+#ifdef USE_ELF
+static void
+follow_debuglink(const char *debuglink, int num_traces, void **traces,
+		 obj_info_t **objp, line_info_t *lines, int offset)
+{
+    /* Ideally we should check 4 paths to follow gnu_debuglink,
+       but we handle only one case for now as this format is used
+       by some linux distributions. See GDB's info for detail. */
+    static const char global_debug_dir[] = "/usr/lib/debug";
+    const size_t global_debug_dir_len = sizeof(global_debug_dir) - 1;
+    char *p;
+    obj_info_t *o1 = *objp, *o2;
+    size_t len;
+
+    p = strrchr(binary_filename, '/');
+    if (!p) {
+	return;
+    }
+    p[1] = '\0';
+
+    len = strlen(binary_filename);
+    if (len >= PATH_MAX - global_debug_dir_len)
+	len = PATH_MAX - global_debug_dir_len - 1;
+    memmove(binary_filename + global_debug_dir_len, binary_filename, len);
+    memcpy(binary_filename, global_debug_dir, global_debug_dir_len);
+    len += global_debug_dir_len;
+    strlcpy(binary_filename + len, debuglink, PATH_MAX - len);
+
+    append_obj(objp);
+    o2 = *objp;
+    o2->base_addr = o1->base_addr;
+    o2->path = o1->path;
+    fill_lines(num_traces, traces, 0, objp, lines, offset);
+}
+#endif
+
+enum
+{
+    DW_TAG_compile_unit = 0x11,
+    DW_TAG_inlined_subroutine = 0x1d,
+    DW_TAG_subprogram = 0x2e,
+};
+
+/* Attributes encodings */
+enum
+{
+    DW_AT_sibling = 0x01,
+    DW_AT_location = 0x02,
+    DW_AT_name = 0x03,
+    /* Reserved 0x04 */
+    /* Reserved 0x05 */
+    /* Reserved 0x06 */
+    /* Reserved 0x07 */
+    /* Reserved 0x08 */
+    DW_AT_ordering = 0x09,
+    /* Reserved 0x0a */
+    DW_AT_byte_size = 0x0b,
+    /* Reserved 0x0c */
+    DW_AT_bit_size = 0x0d,
+    /* Reserved 0x0e */
+    /* Reserved 0x0f */
+    DW_AT_stmt_list = 0x10,
+    DW_AT_low_pc = 0x11,
+    DW_AT_high_pc = 0x12,
+    DW_AT_language = 0x13,
+    /* Reserved 0x14 */
+    DW_AT_discr = 0x15,
+    DW_AT_discr_value = 0x16,
+    DW_AT_visibility = 0x17,
+    DW_AT_import = 0x18,
+    DW_AT_string_length = 0x19,
+    DW_AT_common_reference = 0x1a,
+    DW_AT_comp_dir = 0x1b,
+    DW_AT_const_value = 0x1c,
+    DW_AT_containing_type = 0x1d,
+    DW_AT_default_value = 0x1e,
+    /* Reserved 0x1f */
+    DW_AT_inline = 0x20,
+    DW_AT_is_optional = 0x21,
+    DW_AT_lower_bound = 0x22,
+    /* Reserved 0x23 */
+    /* Reserved 0x24 */
+    DW_AT_producer = 0x25,
+    /* Reserved 0x26 */
+    DW_AT_prototyped = 0x27,
+    /* Reserved 0x28 */
+    /* Reserved 0x29 */
+    DW_AT_return_addr = 0x2a,
+    /* Reserved 0x2b */
+    DW_AT_start_scope = 0x2c,
+    /* Reserved 0x2d */
+    DW_AT_bit_stride = 0x2e,
+    DW_AT_upper_bound = 0x2f,
+    /* Reserved 0x30 */
+    DW_AT_abstract_origin = 0x31,
+    DW_AT_accessibility = 0x32,
+    DW_AT_address_class = 0x33,
+    DW_AT_artificial = 0x34,
+    DW_AT_base_types = 0x35,
+    DW_AT_calling_convention = 0x36,
+    DW_AT_count = 0x37,
+    DW_AT_data_member_location = 0x38,
+    DW_AT_decl_column = 0x39,
+    DW_AT_decl_file = 0x3a,
+    DW_AT_decl_line = 0x3b,
+    DW_AT_declaration = 0x3c,
+    DW_AT_discr_list = 0x3d,
+    DW_AT_encoding = 0x3e,
+    DW_AT_external = 0x3f,
+    DW_AT_frame_base = 0x40,
+    DW_AT_friend = 0x41,
+    DW_AT_identifier_case = 0x42,
+    /* Reserved 0x43 */
+    DW_AT_namelist_item = 0x44,
+    DW_AT_priority = 0x45,
+    DW_AT_segment = 0x46,
+    DW_AT_specification = 0x47,
+    DW_AT_static_link = 0x48,
+    DW_AT_type = 0x49,
+    DW_AT_use_location = 0x4a,
+    DW_AT_variable_parameter = 0x4b,
+    DW_AT_virtuality = 0x4c,
+    DW_AT_vtable_elem_location = 0x4d,
+    DW_AT_allocated = 0x4e,
+    DW_AT_associated = 0x4f,
+    DW_AT_data_location = 0x50,
+    DW_AT_byte_stride = 0x51,
+    DW_AT_entry_pc = 0x52,
+    DW_AT_use_UTF8 = 0x53,
+    DW_AT_extension = 0x54,
+    DW_AT_ranges = 0x55,
+    DW_AT_trampoline = 0x56,
+    DW_AT_call_column = 0x57,
+    DW_AT_call_file = 0x58,
+    DW_AT_call_line = 0x59,
+    DW_AT_description = 0x5a,
+    DW_AT_binary_scale = 0x5b,
+    DW_AT_decimal_scale = 0x5c,
+    DW_AT_small = 0x5d,
+    DW_AT_decimal_sign = 0x5e,
+    DW_AT_digit_count = 0x5f,
+    DW_AT_picture_string = 0x60,
+    DW_AT_mutable = 0x61,
+    DW_AT_threads_scaled = 0x62,
+    DW_AT_explicit = 0x63,
+    DW_AT_object_pointer = 0x64,
+    DW_AT_endianity = 0x65,
+    DW_AT_elemental = 0x66,
+    DW_AT_pure = 0x67,
+    DW_AT_recursive = 0x68,
+    DW_AT_signature = 0x69,
+    DW_AT_main_subprogram = 0x6a,
+    DW_AT_data_bit_offset = 0x6b,
+    DW_AT_const_expr = 0x6c,
+    DW_AT_enum_class = 0x6d,
+    DW_AT_linkage_name = 0x6e,
+    DW_AT_string_length_bit_size = 0x6f,
+    DW_AT_string_length_byte_size = 0x70,
+    DW_AT_rank = 0x71,
+    DW_AT_str_offsets_base = 0x72,
+    DW_AT_addr_base = 0x73,
+    DW_AT_rnglists_base = 0x74,
+    /* Reserved 0x75 */
+    DW_AT_dwo_name = 0x76,
+    DW_AT_reference = 0x77,
+    DW_AT_rvalue_reference = 0x78,
+    DW_AT_macros = 0x79,
+    DW_AT_call_all_calls = 0x7a,
+    DW_AT_call_all_source_calls = 0x7b,
+    DW_AT_call_all_tail_calls = 0x7c,
+    DW_AT_call_return_pc = 0x7d,
+    DW_AT_call_value = 0x7e,
+    DW_AT_call_origin = 0x7f,
+    DW_AT_call_parameter = 0x80,
+    DW_AT_call_pc = 0x81,
+    DW_AT_call_tail_call = 0x82,
+    DW_AT_call_target = 0x83,
+    DW_AT_call_target_clobbered = 0x84,
+    DW_AT_call_data_location = 0x85,
+    DW_AT_call_data_value = 0x86,
+    DW_AT_noreturn = 0x87,
+    DW_AT_alignment = 0x88,
+    DW_AT_export_symbols = 0x89,
+    DW_AT_deleted = 0x8a,
+    DW_AT_defaulted = 0x8b,
+    DW_AT_loclists_base = 0x8c,
+    DW_AT_lo_user = 0x2000,
+    DW_AT_hi_user = 0x3fff
+};
+
+/* Attribute form encodings */
+enum
+{
+    DW_FORM_addr = 0x01,
+    /* Reserved 0x02 */
+    DW_FORM_block2 = 0x03,
+    DW_FORM_block4 = 0x04,
+    DW_FORM_data2 = 0x05,
+    DW_FORM_data4 = 0x06,
+    DW_FORM_data8 = 0x07,
+    DW_FORM_string = 0x08,
+    DW_FORM_block = 0x09,
+    DW_FORM_block1 = 0x0a,
+    DW_FORM_data1 = 0x0b,
+    DW_FORM_flag = 0x0c,
+    DW_FORM_sdata = 0x0d,
+    DW_FORM_strp = 0x0e,
+    DW_FORM_udata = 0x0f,
+    DW_FORM_ref_addr = 0x10,
+    DW_FORM_ref1 = 0x11,
+    DW_FORM_ref2 = 0x12,
+    DW_FORM_ref4 = 0x13,
+    DW_FORM_ref8 = 0x14,
+    DW_FORM_ref_udata = 0x15,
+    DW_FORM_indirect = 0x16,
+    DW_FORM_sec_offset = 0x17,
+    DW_FORM_exprloc = 0x18,
+    DW_FORM_flag_present = 0x19,
+    DW_FORM_strx = 0x1a,
+    DW_FORM_addrx = 0x1b,
+    DW_FORM_ref_sup4 = 0x1c,
+    DW_FORM_strp_sup = 0x1d,
+    DW_FORM_data16 = 0x1e,
+    DW_FORM_line_strp = 0x1f,
+    DW_FORM_ref_sig8 = 0x20,
+    DW_FORM_implicit_const = 0x21,
+    DW_FORM_loclistx = 0x22,
+    DW_FORM_rnglistx = 0x23,
+    DW_FORM_ref_sup8 = 0x24,
+    DW_FORM_strx1 = 0x25,
+    DW_FORM_strx2 = 0x26,
+    DW_FORM_strx3 = 0x27,
+    DW_FORM_strx4 = 0x28,
+    DW_FORM_addrx1 = 0x29,
+    DW_FORM_addrx2 = 0x2a,
+    DW_FORM_addrx3 = 0x2b,
+    DW_FORM_addrx4 = 0x2c
+};
+
+enum {
+    VAL_none = 0,
+    VAL_cstr = 1,
+    VAL_data = 2,
+    VAL_uint = 3,
+    VAL_int = 4
+};
+
+# define ABBREV_TABLE_SIZE 256
+typedef struct {
+    obj_info_t *obj;
+    char *file;
+    char *current_cu;
+    uint64_t current_low_pc;
+    char *debug_line_cu_end;
+    char *debug_line_files;
+    char *debug_line_directories;
+    char *p;
+    char *cu_end;
+    char *pend;
+    char *q0;
+    char *q;
+    int format; /* 4 or 8 */;
+    uint8_t address_size;
+    int level;
+    char *abbrev_table[ABBREV_TABLE_SIZE];
+} DebugInfoReader;
+
+typedef struct {
+    ptrdiff_t pos;
+    int tag;
+    int has_children;
+} DIE;
+
+typedef struct {
+    union {
+        char *ptr;
+        uint64_t uint64;
+        int64_t int64;
+    } as;
+    uint64_t off;
+    uint64_t at;
+    uint64_t form;
+    size_t size;
+    int type;
+} DebugInfoValue;
+
+/* TODO: Big Endian */
+#define MERGE_2INTS(a,b,sz) (((uint64_t)(b)<<sz)|(a))
+
+static uint16_t
+get_uint16(const uint8_t *p)
+{
+    return (uint16_t)MERGE_2INTS(p[0],p[1],8);
+}
+
+static uint32_t
+get_uint32(const uint8_t *p)
+{
+    return (uint32_t)MERGE_2INTS(get_uint16(p),get_uint16(p+2),16);
+}
+
+static uint64_t
+get_uint64(const uint8_t *p)
+{
+    return MERGE_2INTS(get_uint32(p),get_uint32(p+4),32);
+}
+
+static uint8_t
+read_uint8(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 1);
+    return *p;
+}
+
+static uint16_t
+read_uint16(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 2);
+    return get_uint16(p);
+}
+
+static uint32_t
+read_uint24(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 3);
+    return (*p << 16) | get_uint16(p+1);
+}
+
+static uint32_t
+read_uint32(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 4);
+    return get_uint32(p);
+}
+
+static uint64_t
+read_uint64(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + 8);
+    return get_uint64(p);
+}
+
+static uintptr_t
+read_uintptr(char **ptr)
+{
+    const unsigned char *p = (const unsigned char *)*ptr;
+    *ptr = (char *)(p + SIZEOF_VOIDP);
+#if SIZEOF_VOIDP == 8
+    return get_uint64(p);
+#else
+    return get_uint32(p);
+#endif
+}
+
+static uint64_t
+read_uint(DebugInfoReader *reader)
+{
+    if (reader->format == 4) {
+        return read_uint32(&reader->p);
+    } else { /* 64 bit */
+        return read_uint64(&reader->p);
+    }
+}
+
+static uint64_t
+read_uleb128(DebugInfoReader *reader)
+{
+    return uleb128(&reader->p);
+}
+
+static int64_t
+read_sleb128(DebugInfoReader *reader)
+{
+    return sleb128(&reader->p);
+}
+
+static void
+debug_info_reader_init(DebugInfoReader *reader, obj_info_t *obj)
+{
+    reader->file = obj->mapped;
+    reader->obj = obj;
+    reader->p = obj->debug_info.ptr;
+    reader->pend = obj->debug_info.ptr + obj->debug_info.size;
+    reader->debug_line_cu_end = obj->debug_line.ptr;
+}
+
+static void
+di_read_debug_abbrev_cu(DebugInfoReader *reader)
+{
+    uint64_t prev = 0;
+    char *p = reader->q0;
+    for (;;) {
+        uint64_t abbrev_number = uleb128(&p);
+        if (abbrev_number <= prev) break;
+        if (abbrev_number < ABBREV_TABLE_SIZE) {
+            reader->abbrev_table[abbrev_number] = p;
+        }
+        prev = abbrev_number;
+        uleb128(&p); /* tag */
+        p++; /* has_children */
+        /* skip content */
+        for (;;) {
+            uint64_t at = uleb128(&p);
+            uint64_t form = uleb128(&p);
+            if (!at && !form) break;
+        }
+    }
+}
+
+static int
+di_read_debug_line_cu(DebugInfoReader *reader)
+{
+    const char *p;
+    struct LineNumberProgramHeader header;
+
+    p = (const char *)reader->debug_line_cu_end;
+    if (parse_debug_line_header(&p, &header))
+        return -1;
+
+    reader->debug_line_cu_end = (char *)header.cu_end;
+    reader->debug_line_directories = (char *)header.include_directories;
+    reader->debug_line_files = (char *)header.filenames;
+
+    return 0;
+}
+
+static void
+set_uint_value(DebugInfoValue *v, uint64_t n)
+{
+    v->as.uint64 = n;
+    v->type = VAL_uint;
+}
+
+static void
+set_int_value(DebugInfoValue *v, int64_t n)
+{
+    v->as.int64 = n;
+    v->type = VAL_int;
+}
+
+static void
+set_cstr_value(DebugInfoValue *v, char *s)
+{
+    v->as.ptr = s;
+    v->off = 0;
+    v->type = VAL_cstr;
+}
+
+static void
+set_cstrp_value(DebugInfoValue *v, char *s, uint64_t off)
+{
+    v->as.ptr = s;
+    v->off = off;
+    v->type = VAL_cstr;
+}
+
+static void
+set_data_value(DebugInfoValue *v, char *s)
+{
+    v->as.ptr = s;
+    v->type = VAL_data;
+}
+
+static const char *
+get_cstr_value(DebugInfoValue *v)
+{
+    if (v->as.ptr) {
+        return v->as.ptr + v->off;
+    } else {
+        return NULL;
+    }
+}
+
+static void
+debug_info_reader_read_value(DebugInfoReader *reader, uint64_t form, DebugInfoValue *v)
+{
+    switch (form) {
+      case DW_FORM_addr:
+        if (reader->address_size == 4) {
+            set_uint_value(v, read_uint32(&reader->p));
+        } else if (reader->address_size == 8) {
+            set_uint_value(v, read_uint64(&reader->p));
+        } else {
+            fprintf(stderr,"unknown address_size:%d", reader->address_size);
+            abort();
+        }
+        break;
+      case DW_FORM_block2:
+        v->size = read_uint16(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_block4:
+        v->size = read_uint32(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_data2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_data4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_data8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_string:
+        v->size = strlen(reader->p);
+        set_cstr_value(v, reader->p);
+        reader->p += v->size + 1;
+        break;
+      case DW_FORM_block:
+        v->size = uleb128(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_block1:
+        v->size = read_uint8(&reader->p);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_data1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_flag:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_sdata:
+        set_int_value(v, read_sleb128(reader));
+        break;
+      case DW_FORM_strp:
+        set_cstrp_value(v, reader->obj->debug_str.ptr, read_uint(reader));
+        break;
+      case DW_FORM_udata:
+        set_uint_value(v, read_uleb128(reader));
+        break;
+      case DW_FORM_ref_addr:
+        if (reader->address_size == 4) {
+            set_uint_value(v, read_uint32(&reader->p));
+        } else if (reader->address_size == 8) {
+            set_uint_value(v, read_uint64(&reader->p));
+        } else {
+            fprintf(stderr,"unknown address_size:%d", reader->address_size);
+            abort();
+        }
+        break;
+      case DW_FORM_ref1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_ref2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_ref4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_ref8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_ref_udata:
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_indirect:
+        /* TODO: read the refered value */
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_sec_offset:
+        set_uint_value(v, read_uint(reader)); /* offset */
+        /* addrptr: debug_addr */
+        /* lineptr: debug_line */
+        /* loclist: debug_loclists */
+        /* loclistptr: debug_loclists */
+        /* macptr: debug_macro */
+        /* rnglist: debug_rnglists */
+        /* rnglistptr: debug_rnglists */
+        /* stroffsetsptr: debug_str_offsets */
+        break;
+      case DW_FORM_exprloc:
+        v->size = (size_t)read_uleb128(reader);
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_flag_present:
+        set_uint_value(v, 1);
+        break;
+      case DW_FORM_strx:
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_addrx:
+        /* TODO: read .debug_addr */
+        set_uint_value(v, uleb128(&reader->p));
+        break;
+      case DW_FORM_ref_sup4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_strp_sup:
+        set_uint_value(v, read_uint(reader));
+        /* *p = reader->sup_file + reader->sup_str->sh_offset + ret; */
+        break;
+      case DW_FORM_data16:
+        v->size = 16;
+        set_data_value(v, reader->p);
+        reader->p += v->size;
+        break;
+      case DW_FORM_line_strp:
+        set_uint_value(v, read_uint(reader));
+        /* *p = reader->file + reader->line->sh_offset + ret; */
+        break;
+      case DW_FORM_ref_sig8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_implicit_const:
+        set_int_value(v, sleb128(&reader->q));
+        break;
+      case DW_FORM_loclistx:
+        set_uint_value(v, read_uleb128(reader));
+        break;
+      case DW_FORM_rnglistx:
+        set_uint_value(v, read_uleb128(reader));
+        break;
+      case DW_FORM_ref_sup8:
+        set_uint_value(v, read_uint64(&reader->p));
+        break;
+      case DW_FORM_strx1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_strx2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_strx3:
+        set_uint_value(v, read_uint24(&reader->p));
+        break;
+      case DW_FORM_strx4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case DW_FORM_addrx1:
+        set_uint_value(v, read_uint8(&reader->p));
+        break;
+      case DW_FORM_addrx2:
+        set_uint_value(v, read_uint16(&reader->p));
+        break;
+      case DW_FORM_addrx3:
+        set_uint_value(v, read_uint24(&reader->p));
+        break;
+      case DW_FORM_addrx4:
+        set_uint_value(v, read_uint32(&reader->p));
+        break;
+      case 0:
+        goto fail;
+        break;
+    }
+    return;
+
+  fail:
+    fprintf(stderr, "%d: unsupported form: %#"PRIx64"\n", __LINE__, form);
+    exit(1);
+}
+
+/* find abbrev in current compilation unit */
+static char *
+di_find_abbrev(DebugInfoReader *reader, uint64_t abbrev_number)
+{
+    char *p;
+    if (abbrev_number < ABBREV_TABLE_SIZE) {
+        return reader->abbrev_table[abbrev_number];
+    }
+    p = reader->abbrev_table[ABBREV_TABLE_SIZE-1];
+    /* skip 255th record */
+    uleb128(&p); /* tag */
+    p++; /* has_children */
+    /* skip content */
+    for (;;) {
+        uint64_t at = uleb128(&p);
+        uint64_t form = uleb128(&p);
+        if (!at && !form) break;
+    }
+    for (uint64_t n = uleb128(&p); abbrev_number != n; n = uleb128(&p)) {
+        if (n == 0) {
+            fprintf(stderr,"%d: Abbrev Number %"PRId64" not found\n",__LINE__, abbrev_number);
+            exit(1);
+        }
+        uleb128(&p); /* tag */
+        p++; /* has_children */
+        /* skip content */
+        for (;;) {
+            uint64_t at = uleb128(&p);
+            uint64_t form = uleb128(&p);
+            if (!at && !form) break;
+        }
+    }
+    return p;
+}
+
+#if 0
+static void
+hexdump0(const unsigned char *p, size_t n)
+{
+    size_t i;
+    fprintf(stderr, "     0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F\n");
+    for (i=0; i < n; i++){
+        switch (i & 15) {
+          case 0:
+            fprintf(stderr, "%02zd: %02X ", i/16, p[i]);
+            break;
+          case 15:
+            fprintf(stderr, "%02X\n", p[i]);
+            break;
+          default:
+            fprintf(stderr, "%02X ", p[i]);
+            break;
+        }
+    }
+    if ((i & 15) != 15) {
+        fprintf(stderr, "\n");
+    }
+}
+#define hexdump(p,n) hexdump0((const unsigned char *)p, n)
+
+static void
+div_inspect(DebugInfoValue *v)
+{
+    switch (v->type) {
+      case VAL_uint:
+        fprintf(stderr,"%d: type:%d size:%zx v:%lx\n",__LINE__,v->type,v->size,v->as.uint64);
+        break;
+      case VAL_int:
+        fprintf(stderr,"%d: type:%d size:%zx v:%ld\n",__LINE__,v->type,v->size,(int64_t)v->as.uint64);
+        break;
+      case VAL_cstr:
+        fprintf(stderr,"%d: type:%d size:%zx v:'%s'\n",__LINE__,v->type,v->size,v->as.ptr);
+        break;
+      case VAL_data:
+        fprintf(stderr,"%d: type:%d size:%zx v:\n",__LINE__,v->type,v->size);
+        hexdump(v->as.ptr, 16);
+        break;
+    }
+}
+#endif
+
+static DIE *
+di_read_die(DebugInfoReader *reader, DIE *die)
+{
+    uint64_t abbrev_number = uleb128(&reader->p);
+    if (abbrev_number == 0) {
+        reader->level--;
+        return NULL;
+    }
+
+    reader->q = di_find_abbrev(reader, abbrev_number);
+
+    die->pos = reader->p - reader->obj->debug_info.ptr - 1;
+    die->tag = (int)uleb128(&reader->q); /* tag */
+    die->has_children = *reader->q++; /* has_children */
+    if (die->has_children) {
+        reader->level++;
+    }
+    return die;
+}
+
+static DebugInfoValue *
+di_read_record(DebugInfoReader *reader, DebugInfoValue *vp)
+{
+    uint64_t at = uleb128(&reader->q);
+    uint64_t form = uleb128(&reader->q);
+    if (!at || !form) return NULL;
+    vp->at = at;
+    vp->form = form;
+    debug_info_reader_read_value(reader, form, vp);
+    return vp;
+}
+
+static void
+di_skip_records(DebugInfoReader *reader)
+{
+    for (;;) {
+        DebugInfoValue v = {{}};
+        uint64_t at = uleb128(&reader->q);
+        uint64_t form = uleb128(&reader->q);
+        if (!at || !form) return;
+        debug_info_reader_read_value(reader, form, &v);
+    }
+}
+
+typedef struct {
+    uint64_t low_pc;
+    uint64_t high_pc;
+    uint64_t ranges;
+    bool low_pc_set;
+    bool high_pc_set;
+    bool ranges_set;
+} ranges_t;
+
+static void
+ranges_set(ranges_t *ptr, DebugInfoValue *v)
+{
+    switch (v->at) {
+      case DW_AT_low_pc:
+        ptr->low_pc = v->as.uint64;
+        ptr->low_pc_set = true;
+        break;
+      case DW_AT_high_pc:
+        if (v->form == DW_FORM_addr) {
+            ptr->high_pc = v->as.uint64;
+        }
+        else {
+            ptr->high_pc = ptr->low_pc + v->as.uint64;
+        }
+        ptr->high_pc_set = true;
+        break;
+      case DW_AT_ranges:
+        ptr->ranges = v->as.uint64;
+        ptr->ranges_set = true;
+        break;
+    }
+}
+
+static uintptr_t
+ranges_include(DebugInfoReader *reader, ranges_t *ptr, uint64_t addr)
+{
+    if (ptr->high_pc_set) {
+        if (ptr->ranges_set || !ptr->low_pc_set) {
+            exit(1);
+        }
+        if (ptr->low_pc <= addr && addr <= ptr->high_pc) {
+            return (uintptr_t)ptr->low_pc;
+        }
+    }
+    else if (ptr->ranges_set) {
+        /* TODO: support base address selection entry */
+        char *p = reader->obj->debug_ranges.ptr + ptr->ranges;
+        uint64_t base = ptr->low_pc_set ? ptr->low_pc : reader->current_low_pc;
+        for (;;) {
+            uintptr_t from = read_uintptr(&p);
+            uintptr_t to = read_uintptr(&p);
+            if (!from && !to) break;
+            if (from == UINTPTR_MAX) {
+                /* base address selection entry */
+                base = to;
+            }
+            else if (base + from <= addr && addr < base + to) {
+                return from;
+            }
+        }
+    }
+    else if (ptr->low_pc_set) {
+        if (ptr->low_pc == addr) {
+            return (uintptr_t)ptr->low_pc;
+        }
+    }
+    return false;
+}
+
+#if 0
+static void
+ranges_inspect(DebugInfoReader *reader, ranges_t *ptr)
+{
+    if (ptr->high_pc_set) {
+        if (ptr->ranges_set || !ptr->low_pc_set) {
+            fprintf(stderr,"low_pc_set:%d high_pc_set:%d ranges_set:%d\n",ptr->low_pc_set,ptr->high_pc_set,ptr->ranges_set);
+            exit(1);
+        }
+        fprintf(stderr,"low_pc:%"PRIx64" high_pc:%"PRIx64"\n",ptr->low_pc,ptr->high_pc);
+    }
+    else if (ptr->ranges_set) {
+        char *p = reader->obj->debug_ranges.ptr + ptr->ranges;
+        fprintf(stderr,"low_pc:%"PRIx64" ranges:%"PRIx64" %lx ",ptr->low_pc,ptr->ranges, p-reader->obj->mapped);
+        for (;;) {
+            uintptr_t from = read_uintptr(&p);
+            uintptr_t to = read_uintptr(&p);
+            if (!from && !to) break;
+            fprintf(stderr,"%"PRIx64"-%"PRIx64" ",ptr->low_pc+from,ptr->low_pc+to);
+        }
+        fprintf(stderr,"\n");
+    }
+    else if (ptr->low_pc_set) {
+        fprintf(stderr,"low_pc:%"PRIx64"\n",ptr->low_pc);
+    }
+    else {
+        fprintf(stderr,"empty\n");
+    }
+}
+#endif
+
+static int
+di_read_cu(DebugInfoReader *reader)
+{
+    uint64_t unit_length;
+    uint16_t version;
+    uint64_t debug_abbrev_offset;
+    reader->format = 4;
+    reader->current_cu = reader->p;
+    unit_length = read_uint32(&reader->p);
+    if (unit_length == 0xffffffff) {
+        unit_length = read_uint64(&reader->p);
+        reader->format = 8;
+    }
+    reader->cu_end = reader->p + unit_length;
+    version = read_uint16(&reader->p);
+    if (version > 5) {
+        return -1;
+    }
+    else if (version == 5) {
+        /* unit_type = */ read_uint8(&reader->p);
+        reader->address_size = read_uint8(&reader->p);
+        debug_abbrev_offset = read_uint(reader);
+    }
+    else {
+        debug_abbrev_offset = read_uint(reader);
+        reader->address_size = read_uint8(&reader->p);
+    }
+    reader->q0 = reader->obj->debug_abbrev.ptr + debug_abbrev_offset;
+
+    reader->level = 0;
+    di_read_debug_abbrev_cu(reader);
+    if (di_read_debug_line_cu(reader)) return -1;
+
+#if defined(__GNUC__) && !defined(__clang__) && !defined(__INTEL_COMPILER_BUILD_DATE)
+    /* Though DWARF specifies "the applicable base address defaults to the base
+       address of the compilation unit", but GCC seems to use zero as default */
+#else
+    do {
+        DIE die;
+
+        if (!di_read_die(reader, &die)) continue;
+
+        if (die.tag != DW_TAG_compile_unit) {
+            di_skip_records(reader);
+            break;
+        }
+
+        /* enumerate abbrev */
+        for (;;) {
+            DebugInfoValue v = {{}};
+            if (!di_read_record(reader, &v)) break;
+            switch (v.at) {
+              case DW_AT_low_pc:
+                reader->current_low_pc = v.as.uint64;
+                break;
+            }
+        }
+    } while (0);
+#endif
+    return 0;
+}
+
+static void
+read_abstract_origin(DebugInfoReader *reader, uint64_t abstract_origin, line_info_t *line)
+{
+    char *p = reader->p;
+    char *q = reader->q;
+    int level = reader->level;
+    DIE die;
+
+    reader->p = reader->current_cu + abstract_origin;
+    if (!di_read_die(reader, &die)) goto finish;
+
+    /* enumerate abbrev */
+    for (;;) {
+        DebugInfoValue v = {{}};
+        if (!di_read_record(reader, &v)) break;
+        switch (v.at) {
+          case DW_AT_name:
+            line->sname = get_cstr_value(&v);
+            break;
+        }
+    }
+
+  finish:
+    reader->p = p;
+    reader->q = q;
+    reader->level = level;
+}
+
+static void
+debug_info_read(DebugInfoReader *reader, int num_traces, void **traces,
+         line_info_t *lines, int offset) {
+    while (reader->p < reader->cu_end) {
+        DIE die;
+        ranges_t ranges = {};
+        line_info_t line = {};
+
+        if (!di_read_die(reader, &die)) continue;
+        /* fprintf(stderr,"%d:%tx: <%d>\n",__LINE__,die.pos,reader->level,die.tag); */
+
+        if (die.tag != DW_TAG_subprogram && die.tag != DW_TAG_inlined_subroutine) {
+          skip_die:
+            di_skip_records(reader);
+            continue;
+        }
+
+        /* enumerate abbrev */
+        for (;;) {
+            DebugInfoValue v = {{}};
+            /* ptrdiff_t pos = reader->p - reader->p0; */
+            if (!di_read_record(reader, &v)) break;
+            /* fprintf(stderr,"\n%d:%tx: AT:%lx FORM:%lx\n",__LINE__,pos,v.at,v.form); */
+            /* div_inspect(&v); */
+            switch (v.at) {
+              case DW_AT_name:
+                line.sname = get_cstr_value(&v);
+                break;
+              case DW_AT_call_file:
+                fill_filename((int)v.as.uint64, reader->debug_line_directories, reader->debug_line_files, &line, reader->obj);
+                break;
+              case DW_AT_call_line:
+                line.line = (int)v.as.uint64;
+                break;
+              case DW_AT_low_pc:
+              case DW_AT_high_pc:
+              case DW_AT_ranges:
+                ranges_set(&ranges, &v);
+                break;
+              case DW_AT_declaration:
+                goto skip_die;
+              case DW_AT_inline:
+                /* 1 or 3 */
+                break; /* goto skip_die; */
+              case DW_AT_abstract_origin:
+                read_abstract_origin(reader, v.as.uint64, &line);
+                break; /* goto skip_die; */
+            }
+        }
+        /* ranges_inspect(reader, &ranges); */
+        /* fprintf(stderr,"%d:%tx: %x ",__LINE__,diepos,die.tag); */
+        for (int i=offset; i < num_traces; i++) {
+            uintptr_t addr = (uintptr_t)traces[i];
+            uintptr_t offset = addr - reader->obj->base_addr + reader->obj->vmaddr;
+            uintptr_t saddr = ranges_include(reader, &ranges, offset);
+            if (saddr) {
+                /* fprintf(stderr, "%d:%tx: %d %lx->%lx %x %s: %s/%s %d %s %s %s\n",__LINE__,die.pos, i,addr,offset, die.tag,line.sname,line.dirname,line.filename,line.line,reader->obj->path,line.sname,lines[i].sname); */
+                if (lines[i].sname) {
+                    line_info_t *lp = malloc(sizeof(line_info_t));
+                    memcpy(lp, &lines[i], sizeof(line_info_t));
+                    lines[i].next = lp;
+                    lp->dirname = line.dirname;
+                    lp->filename = line.filename;
+                    lp->line = line.line;
+                    lp->saddr = 0;
+                }
+                lines[i].path = reader->obj->path;
+                lines[i].base_addr = line.base_addr;
+                lines[i].sname = line.sname;
+                lines[i].saddr = saddr + reader->obj->base_addr - reader->obj->vmaddr;
+            }
+        }
+    }
+}
+
+#ifdef USE_ELF
+static unsigned long
+uncompress_debug_section(ElfW(Shdr) *shdr, char *file, char **ptr)
+{
+#ifdef SUPPORT_COMPRESSED_DEBUG_LINE
+    ElfW(Chdr) *chdr = (ElfW(Chdr) *)(file + shdr->sh_offset);
+    unsigned long destsize = chdr->ch_size;
+    int ret = 0;
+
+    if (chdr->ch_type != ELFCOMPRESS_ZLIB) {
+	/* unsupported compression type */
+	return 0;
+    }
+
+    *ptr = malloc(destsize);
+    if (!*ptr) return 0;
+    ret = uncompress((Bytef *)*ptr, &destsize,
+	    (const Bytef*)chdr + sizeof(ElfW(Chdr)),
+            shdr->sh_size - sizeof(ElfW(Chdr)));
+    if (ret != Z_OK) goto fail;
+    return destsize;
+
+fail:
+    free(*ptr);
+#endif
+    return 0;
+}
+
+/* read file and fill lines */
+static uintptr_t
+fill_lines(int num_traces, void **traces, int check_debuglink,
+	   obj_info_t **objp, line_info_t *lines, int offset)
+{
+    int i, j;
+    char *shstr;
+    ElfW(Ehdr) *ehdr;
+    ElfW(Shdr) *shdr, *shstr_shdr;
+    ElfW(Shdr) *gnu_debuglink_shdr = NULL;
+    int fd;
+    off_t filesize;
+    char *file;
+    ElfW(Shdr) *symtab_shdr = NULL, *strtab_shdr = NULL;
+    ElfW(Shdr) *dynsym_shdr = NULL, *dynstr_shdr = NULL;
+    obj_info_t *obj = *objp;
+    uintptr_t dladdr_fbase = 0;
+
+    fd = open(binary_filename, O_RDONLY);
+    if (fd < 0) {
+	goto fail;
+    }
+    filesize = lseek(fd, 0, SEEK_END);
+    if (filesize < 0) {
+	int e = errno;
+	close(fd);
+	kprintf("lseek: %s\n", strerror(e));
+	goto fail;
+    }
+#if SIZEOF_OFF_T > SIZEOF_SIZE_T
+    if (filesize > (off_t)SIZE_MAX) {
+	close(fd);
+	kprintf("Too large file %s\n", binary_filename);
+	goto fail;
+    }
+#endif
+    lseek(fd, 0, SEEK_SET);
+    /* async-signal unsafe */
+    file = (char *)mmap(NULL, (size_t)filesize, PROT_READ, MAP_SHARED, fd, 0);
+    if (file == MAP_FAILED) {
+	int e = errno;
+	close(fd);
+	kprintf("mmap: %s\n", strerror(e));
+	goto fail;
+    }
+    close(fd);
+
+    ehdr = (ElfW(Ehdr) *)file;
+    if (memcmp(ehdr->e_ident, "\177ELF", 4) != 0) {
+	/*
+	 * Huh? Maybe filename was overridden by setproctitle() and
+	 * it match non-elf file.
+	 */
+	goto fail;
+    }
+    obj->mapped = file;
+    obj->mapped_size = (size_t)filesize;
+
+    shdr = (ElfW(Shdr) *)(file + ehdr->e_shoff);
+
+    shstr_shdr = shdr + ehdr->e_shstrndx;
+    shstr = file + shstr_shdr->sh_offset;
+
+    for (i = 0; i < ehdr->e_shnum; i++) {
+        char *section_name = shstr + shdr[i].sh_name;
+	switch (shdr[i].sh_type) {
+	  case SHT_STRTAB:
+	    if (!strcmp(section_name, ".strtab")) {
+		strtab_shdr = shdr + i;
+	    }
+	    else if (!strcmp(section_name, ".dynstr")) {
+		dynstr_shdr = shdr + i;
+	    }
+	    break;
+	  case SHT_SYMTAB:
+	    /* if (!strcmp(section_name, ".symtab")) */
+	    symtab_shdr = shdr + i;
+	    break;
+	  case SHT_DYNSYM:
+	    /* if (!strcmp(section_name, ".dynsym")) */
+	    dynsym_shdr = shdr + i;
+	    break;
+	  case SHT_PROGBITS:
+	    if (!strcmp(section_name, ".gnu_debuglink")) {
+		gnu_debuglink_shdr = shdr + i;
+	    }
+            else {
+                const char *debug_section_names[] = {
+                    ".debug_abbrev",
+                    ".debug_info",
+                    ".debug_line",
+                    ".debug_ranges",
+                    ".debug_str"
+                };
+
+                for (j=0; j < DWARF_SECTION_COUNT; j++) {
+                    struct dwarf_section *s = obj_dwarf_section_at(obj, j);
+
+                    if (strcmp(section_name, debug_section_names[j]) != 0)
+                        continue;
+
+                    s->ptr = file + shdr[i].sh_offset;
+                    s->size = shdr[i].sh_size;
+                    s->flags = shdr[i].sh_flags;
+                    if (s->flags & SHF_COMPRESSED) {
+                        s->size = uncompress_debug_section(&shdr[i], file, &s->ptr);
+                        if (!s->size) goto fail;
+                    }
+                    break;
+                }
+            }
+	    break;
+	}
+    }
+
+    if (offset == -1) {
+	/* main executable */
+	offset = 0;
+	if (dynsym_shdr && dynstr_shdr) {
+	    char *strtab = file + dynstr_shdr->sh_offset;
+	    ElfW(Sym) *symtab = (ElfW(Sym) *)(file + dynsym_shdr->sh_offset);
+	    int symtab_count = (int)(dynsym_shdr->sh_size / sizeof(ElfW(Sym)));
+            void *handle = dlopen(NULL, RTLD_NOW|RTLD_LOCAL);
+            if (handle) {
+                for (j = 0; j < symtab_count; j++) {
+                    ElfW(Sym) *sym = &symtab[j];
+                    Dl_info info;
+                    void *s;
+                    if (ELF_ST_TYPE(sym->st_info) != STT_FUNC || sym->st_size == 0) continue;
+                    s = dlsym(handle, strtab + sym->st_name);
+                    if (s && dladdr(s, &info)) {
+                        obj->base_addr = dladdr_fbase;
+                        dladdr_fbase = (uintptr_t)info.dli_fbase;
+                        break;
+                    }
+                }
+                dlclose(handle);
+            }
+	    if (ehdr->e_type == ET_EXEC) {
+		obj->base_addr = 0;
+	    }
+	    else {
+		/* PIE (position-independent executable) */
+		obj->base_addr = dladdr_fbase;
+	    }
+	}
+    }
+
+    if (obj->debug_info.ptr && obj->debug_abbrev.ptr) {
+        DebugInfoReader reader;
+        debug_info_reader_init(&reader, obj);
+        i = 0;
+        while (reader.p < reader.pend) {
+            /* fprintf(stderr, "%d:%tx: CU[%d]\n", __LINE__, reader.p - reader.obj->debug_info.ptr, i++); */
+            if (di_read_cu(&reader)) goto use_symtab;
+            debug_info_read(&reader, num_traces, traces, lines, offset);
+        }
+    }
+    else {
+        /* This file doesn't have dwarf, use symtab or dynsym */
+use_symtab:
+        if (!symtab_shdr) {
+            /* This file doesn't have symtab, use dynsym instead */
+            symtab_shdr = dynsym_shdr;
+            strtab_shdr = dynstr_shdr;
+        }
+
+        if (symtab_shdr && strtab_shdr) {
+            char *strtab = file + strtab_shdr->sh_offset;
+            ElfW(Sym) *symtab = (ElfW(Sym) *)(file + symtab_shdr->sh_offset);
+            int symtab_count = (int)(symtab_shdr->sh_size / sizeof(ElfW(Sym)));
+            for (j = 0; j < symtab_count; j++) {
+                ElfW(Sym) *sym = &symtab[j];
+                uintptr_t saddr = (uintptr_t)sym->st_value + obj->base_addr;
+                if (ELF_ST_TYPE(sym->st_info) != STT_FUNC) continue;
+                for (i = offset; i < num_traces; i++) {
+                    uintptr_t d = (uintptr_t)traces[i] - saddr;
+                    if (lines[i].line > 0 || d > (uintptr_t)sym->st_size)
+                        continue;
+                    /* fill symbol name and addr from .symtab */
+                    if (!lines[i].sname) lines[i].sname = strtab + sym->st_name;
+                    lines[i].saddr = saddr;
+                    lines[i].path  = obj->path;
+                    lines[i].base_addr = obj->base_addr;
+                }
+            }
+        }
+    }
+
+    if (!obj->debug_line.ptr) {
+	/* This file doesn't have .debug_line section,
+	   let's check .gnu_debuglink section instead. */
+	if (gnu_debuglink_shdr && check_debuglink) {
+	    follow_debuglink(file + gnu_debuglink_shdr->sh_offset,
+			     num_traces, traces,
+			     objp, lines, offset);
+	}
+	goto finish;
+    }
+
+    if (parse_debug_line(num_traces, traces,
+            obj->debug_line.ptr,
+            obj->debug_line.size,
+            obj, lines, offset) == -1)
+        goto fail;
+
+finish:
+    return dladdr_fbase;
+fail:
+    return (uintptr_t)-1;
+}
+#else /* Mach-O */
+/* read file and fill lines */
+static uintptr_t
+fill_lines(int num_traces, void **traces, int check_debuglink,
+        obj_info_t **objp, line_info_t *lines, int offset)
+{
+# ifdef __LP64__
+#  define LP(x) x##_64
+# else
+#  define LP(x) x
+# endif
+    int fd;
+    off_t filesize;
+    char *file, *p = NULL;
+    obj_info_t *obj = *objp;
+    struct LP(mach_header) *header;
+    uintptr_t dladdr_fbase = 0;
+
+    {
+        char *s = binary_filename;
+        char *base = strrchr(binary_filename, '/')+1;
+        size_t max = PATH_MAX;
+        size_t size = strlen(binary_filename);
+        size_t basesize = size - (base - binary_filename);
+        s += size;
+        max -= size;
+        p = s;
+        size = strlcpy(s, ".dSYM/Contents/Resources/DWARF/", max);
+        if (size == 0) goto fail;
+        s += size;
+        max -= size;
+        if (max <= basesize) goto fail;
+        memcpy(s, base, basesize);
+        s[basesize] = 0;
+
+        fd = open(binary_filename, O_RDONLY);
+        if (fd < 0) {
+            *p = 0; /* binary_filename becomes original file name */
+            fd = open(binary_filename, O_RDONLY);
+            if (fd < 0) {
+                goto fail;
+            }
+        }
+    }
+
+    filesize = lseek(fd, 0, SEEK_END);
+    if (filesize < 0) {
+        int e = errno;
+        close(fd);
+        kprintf("lseek: %s\n", strerror(e));
+        goto fail;
+    }
+#if SIZEOF_OFF_T > SIZEOF_SIZE_T
+    if (filesize > (off_t)SIZE_MAX) {
+        close(fd);
+        kprintf("Too large file %s\n", binary_filename);
+        goto fail;
+    }
+#endif
+    lseek(fd, 0, SEEK_SET);
+    /* async-signal unsafe */
+    file = (char *)mmap(NULL, (size_t)filesize, PROT_READ, MAP_SHARED, fd, 0);
+    if (file == MAP_FAILED) {
+        int e = errno;
+        close(fd);
+        kprintf("mmap: %s\n", strerror(e));
+        goto fail;
+    }
+    close(fd);
+
+    obj->mapped = file;
+    obj->mapped_size = (size_t)filesize;
+
+    header = (struct LP(mach_header) *)file;
+    if (header->magic == LP(MH_MAGIC)) {
+        /* non universal binary */
+        p = file;
+    }
+    else if (header->magic == FAT_CIGAM) {
+        struct fat_header *fat = (struct fat_header *)file;
+        char *q = file + sizeof(*fat);
+        uint32_t nfat_arch = __builtin_bswap32(fat->nfat_arch);
+        /* fprintf(stderr,"%d: fat:%s %d\n",__LINE__, binary_filename,nfat_arch); */
+        for (uint32_t i = 0; i < nfat_arch; i++) {
+            struct fat_arch *arch = (struct fat_arch *)q;
+            cpu_type_t cputype = __builtin_bswap32(arch->cputype);
+            cpu_subtype_t cpusubtype = __builtin_bswap32(arch->cpusubtype);
+            uint32_t offset = __builtin_bswap32(arch->offset);
+            /* fprintf(stderr,"%d: fat %d %x/%x %x/%x\n",__LINE__, i, _mh_execute_header.cputype,_mh_execute_header.cpusubtype, cputype,cpusubtype); */
+            if (_mh_execute_header.cputype == cputype &&
+                    (_mh_execute_header.cpusubtype & ~CPU_SUBTYPE_MASK) == cpusubtype) {
+                p = file + offset;
+                file = p;
+                header = (struct LP(mach_header) *)p;
+                if (header->magic == LP(MH_MAGIC)) {
+                    goto found_mach_header;
+                }
+                break;
+            }
+            q += sizeof(*arch);
+        }
+        kprintf("'%s' is not a Mach-O universal binary file!\n",binary_filename);
+        close(fd);
+        goto fail;
+    }
+    else {
+        kprintf("'%s' is not a "
+# ifdef __LP64__
+                "64"
+# else
+                "32"
+# endif
+                "-bit Mach-O file!\n",binary_filename);
+        close(fd);
+        goto fail;
+    }
+found_mach_header:
+    p += sizeof(*header);
+
+    for (uint32_t i = 0; i < (uint32_t)header->ncmds; i++) {
+        struct load_command *lcmd = (struct load_command *)p;
+        switch (lcmd->cmd) {
+          case LP(LC_SEGMENT):
+            {
+                static const char *debug_section_names[] = {
+                    "__debug_abbrev",
+                    "__debug_info",
+                    "__debug_line",
+                    "__debug_ranges",
+                    "__debug_str"
+                };
+                struct LP(segment_command) *scmd = (struct LP(segment_command) *)lcmd;
+                if (strcmp(scmd->segname, "__TEXT") == 0) {
+                    obj->vmaddr = scmd->vmaddr;
+                }
+                else if (strcmp(scmd->segname, "__DWARF") == 0) {
+                    p += sizeof(struct LP(segment_command));
+                    for (uint64_t i = 0; i < scmd->nsects; i++) {
+                        struct LP(section) *sect = (struct LP(section) *)p;
+                        p += sizeof(struct LP(section));
+                        for (int j=0; j < DWARF_SECTION_COUNT; j++) {
+                            struct dwarf_section *s = obj_dwarf_section_at(obj, j);
+
+                            if (strcmp(sect->sectname, debug_section_names[j]) != 0)
+                                continue;
+
+                            s->ptr = file + sect->offset;
+                            s->size = sect->size;
+                            s->flags = sect->flags;
+                            if (s->flags & SHF_COMPRESSED) {
+                                goto fail;
+                            }
+                            break;
+                        }
+                    }
+                }
+            }
+            break;
+
+          case LC_SYMTAB:
+            {
+                struct symtab_command *cmd = (struct symtab_command *)lcmd;
+                struct LP(nlist) *nl = (struct LP(nlist) *)(file + cmd->symoff);
+                char *strtab = file + cmd->stroff, *sname;
+                uint32_t j;
+                uintptr_t saddr;
+                /* kprintf("[%2d]: %x/symtab %p\n", i, cmd->cmd, p); */
+                for (j = 0; j < cmd->nsyms; j++) {
+                    uintptr_t symsize, d;
+                    struct LP(nlist) *e = &nl[j];
+                        /* kprintf("[%2d][%4d]: %02x/%x/%x: %s %llx\n", i, j, e->n_type,e->n_sect,e->n_desc,strtab+e->n_un.n_strx,e->n_value); */
+                    if (e->n_type != N_FUN) continue;
+                    if (e->n_sect) {
+                        saddr = (uintptr_t)e->n_value + obj->base_addr - obj->vmaddr;
+                        sname = strtab + e->n_un.n_strx;
+                        /* kprintf("[%2d][%4d]: %02x/%x/%x: %s %llx\n", i, j, e->n_type,e->n_sect,e->n_desc,strtab+e->n_un.n_strx,e->n_value); */
+                        continue;
+                    }
+                    for (int k = offset; k < num_traces; k++) {
+                        d = (uintptr_t)traces[k] - saddr;
+                        symsize = e->n_value;
+                        /* kprintf("%lx %lx %lx\n",saddr,symsize,traces[k]); */
+                        if (lines[k].line > 0 || d > (uintptr_t)symsize)
+                            continue;
+                        /* fill symbol name and addr from .symtab */
+                        if (!lines[k].sname) lines[k].sname = sname;
+                        lines[k].saddr = saddr;
+                        lines[k].path  = obj->path;
+                        lines[k].base_addr = obj->base_addr;
+                    }
+                }
+            }
+        }
+        p += lcmd->cmdsize;
+    }
+
+    if (obj->debug_info.ptr && obj->debug_abbrev.ptr) {
+        DebugInfoReader reader;
+        debug_info_reader_init(&reader, obj);
+        while (reader.p < reader.pend) {
+            if (di_read_cu(&reader)) goto fail;
+            debug_info_read(&reader, num_traces, traces, lines, offset);
+        }
+    }
+
+    if (parse_debug_line(num_traces, traces,
+            obj->debug_line.ptr,
+            obj->debug_line.size,
+            obj, lines, offset) == -1)
+        goto fail;
+
+    return dladdr_fbase;
+fail:
+    return (uintptr_t)-1;
+}
+#endif
+
+#define HAVE_MAIN_EXE_PATH
+#if defined(__FreeBSD__)
+# include <sys/sysctl.h>
+#endif
+/* ssize_t main_exe_path(void)
+ *
+ * store the path of the main executable to `binary_filename`,
+ * and returns strlen(binary_filename).
+ * it is NUL terminated.
+ */
+#if defined(__linux__)
+static ssize_t
+main_exe_path(void)
+{
+# define PROC_SELF_EXE "/proc/self/exe"
+    ssize_t len = readlink(PROC_SELF_EXE, binary_filename, PATH_MAX);
+    binary_filename[len] = 0;
+    return len;
+}
+#elif defined(__FreeBSD__)
+static ssize_t
+main_exe_path(void)
+{
+    int mib[4] = {CTL_KERN, KERN_PROC, KERN_PROC_PATHNAME, -1};
+    size_t len = PATH_MAX;
+    int err = sysctl(mib, 4, binary_filename, &len, NULL, 0);
+    if (err) {
+	kprintf("Can't get the path of ruby");
+	return -1;
+    }
+    len--; /* sysctl sets strlen+1 */
+    return len;
+}
+#else
+#undef HAVE_MAIN_EXE_PATH
+#endif
+
+static void
+print_line0(line_info_t *line, void *address)
+{
+    uintptr_t addr = (uintptr_t)address;
+    uintptr_t d = addr - line->saddr;
+    if (!address) {
+        /* inlined */
+        if (line->dirname && line->dirname[0]) {
+            kprintf("%s(%s) %s/%s:%d\n", line->path, line->sname, line->dirname, line->filename, line->line);
+        }
+        else {
+            kprintf("%s(%s) %s:%d\n", line->path, line->sname, line->filename, line->line);
+        }
+    }
+    else if (!line->path) {
+        kprintf("[0x%"PRIxPTR"]\n", addr);
+    }
+    else if (!line->saddr || !line->sname) {
+        kprintf("%s(0x%"PRIxPTR") [0x%"PRIxPTR"]\n", line->path, addr-line->base_addr, addr);
+    }
+    else if (line->line <= 0) {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"]\n", line->path, line->sname,
+                d, addr);
+    }
+    else if (!line->filename) {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] ???:%d\n", line->path, line->sname,
+                d, addr, line->line);
+    }
+    else if (line->dirname && line->dirname[0]) {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] %s/%s:%d\n", line->path, line->sname,
+                d, addr, line->dirname, line->filename, line->line);
+    }
+    else {
+        kprintf("%s(%s+0x%"PRIxPTR") [0x%"PRIxPTR"] %s:%d\n", line->path, line->sname,
+                d, addr, line->filename, line->line);
+    }
+}
+
+static void
+print_line(line_info_t *line, void *address)
+{
+    print_line0(line, address);
+    if (line->next) {
+        print_line(line->next, NULL);
+    }
+}
+
+void
+rb_dump_backtrace_with_lines(int num_traces, void **traces)
+{
+    int i;
+    /* async-signal unsafe */
+    line_info_t *lines = (line_info_t *)calloc(num_traces, sizeof(line_info_t));
+    obj_info_t *obj = NULL;
+    /* 2 is NULL + main executable */
+    void **dladdr_fbases = (void **)calloc(num_traces+2, sizeof(void *));
+#ifdef HAVE_MAIN_EXE_PATH
+    char *main_path = NULL; /* used on printing backtrace */
+    ssize_t len;
+    if ((len = main_exe_path()) > 0) {
+	main_path = (char *)alloca(len + 1);
+	if (main_path) {
+	    uintptr_t addr;
+	    memcpy(main_path, binary_filename, len+1);
+	    append_obj(&obj);
+	    obj->path = main_path;
+	    addr = fill_lines(num_traces, traces, 1, &obj, lines, -1);
+	    if (addr != (uintptr_t)-1) {
+		dladdr_fbases[0] = (void *)addr;
+	    }
+	}
+    }
+#endif
+
+    /* fill source lines by reading dwarf */
+    for (i = 0; i < num_traces; i++) {
+	Dl_info info;
+	if (lines[i].line) continue;
+	if (dladdr(traces[i], &info)) {
+	    const char *path;
+	    void **p;
+
+	    /* skip symbols which is in already checked objects */
+	    /* if the binary is strip-ed, this may effect */
+	    for (p=dladdr_fbases; *p; p++) {
+		if (*p == info.dli_fbase) {
+		    lines[i].path = info.dli_fname;
+		    lines[i].sname = info.dli_sname;
+		    goto next_line;
+		}
+	    }
+	    *p = info.dli_fbase;
+
+	    append_obj(&obj);
+	    obj->base_addr = (uintptr_t)info.dli_fbase;
+	    path = info.dli_fname;
+	    obj->path = path;
+	    lines[i].path = path;
+            lines[i].sname = info.dli_sname;
+            lines[i].saddr = (uintptr_t)info.dli_saddr;
+	    strlcpy(binary_filename, path, PATH_MAX);
+	    if (fill_lines(num_traces, traces, 1, &obj, lines, i) == (uintptr_t)-1)
+		break;
+	}
+next_line:
+	continue;
+    }
+
+    /* output */
+    for (i = 0; i < num_traces; i++) {
+        print_line(&lines[i], traces[i]);
+
+	/* FreeBSD's backtrace may show _start and so on */
+	if (lines[i].sname && strcmp("main", lines[i].sname) == 0)
+	    break;
+    }
+
+    /* free */
+    while (obj) {
+	obj_info_t *o = obj;
+        for (i=0; i < DWARF_SECTION_COUNT; i++) {
+            struct dwarf_section *s = obj_dwarf_section_at(obj, i);
+            if (s->flags & SHF_COMPRESSED) {
+                free(s->ptr);
+            }
+        }
+	if (obj->mapped_size) {
+	    munmap(obj->mapped, obj->mapped_size);
+	}
+	obj = o->next;
+	free(o);
+    }
+    for (i = 0; i < num_traces; i++) {
+        line_info_t *line = lines[i].next;
+        while (line) {
+            line_info_t *l = line;
+            line = line->next;
+            free(l);
+        }
+    }
+    free(lines);
+    free(dladdr_fbases);
+}
+
+/* From FreeBSD's lib/libstand/printf.c */
+/*-
+ * Copyright (c) 1986, 1988, 1991, 1993
+ *	The Regents of the University of California.  All rights reserved.
+ * (c) UNIX System Laboratories, Inc.
+ * All or some portions of this file are derived from material licensed
+ * to the University of California by American Telephone and Telegraph
+ * Co. or Unix System Laboratories, Inc. and are reproduced herein with
+ * the permission of UNIX System Laboratories, Inc.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 4. Neither the name of the University nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ *	@(#)subr_prf.c	8.3 (Berkeley) 1/21/94
+ */
+
+#include <stdarg.h>
+#define MAXNBUF (sizeof(intmax_t) * CHAR_BIT + 1)
+static inline int toupper(int c) { return ('A' <= c && c <= 'Z') ? (c&0x5f) : c; }
+#define    hex2ascii(hex)  (hex2ascii_data[hex])
+static const char hex2ascii_data[] = "0123456789abcdefghijklmnopqrstuvwxyz";
+static inline int imax(int a, int b) { return (a > b ? a : b); }
+static int kvprintf(char const *fmt, void (*func)(int), void *arg, int radix, va_list ap);
+
+static void putce(int c)
+{
+    char s[1];
+    ssize_t ret;
+
+    s[0] = (char)c;
+    ret = write(2, s, 1);
+    (void)ret;
+}
+
+static int
+kprintf(const char *fmt, ...)
+{
+	va_list ap;
+	int retval;
+
+	va_start(ap, fmt);
+	retval = kvprintf(fmt, putce, NULL, 10, ap);
+	va_end(ap);
+	return retval;
+}
+
+/*
+ * Put a NUL-terminated ASCII number (base <= 36) in a buffer in reverse
+ * order; return an optional length and a pointer to the last character
+ * written in the buffer (i.e., the first character of the string).
+ * The buffer pointed to by `nbuf' must have length >= MAXNBUF.
+ */
+static char *
+ksprintn(char *nbuf, uintmax_t num, int base, int *lenp, int upper)
+{
+	char *p, c;
+
+	p = nbuf;
+	*p = '\0';
+	do {
+		c = hex2ascii(num % base);
+		*++p = upper ? toupper(c) : c;
+	} while (num /= base);
+	if (lenp)
+		*lenp = (int)(p - nbuf);
+	return (p);
+}
+
+/*
+ * Scaled down version of printf(3).
+ *
+ * Two additional formats:
+ *
+ * The format %b is supported to decode error registers.
+ * Its usage is:
+ *
+ *	printf("reg=%b\n", regval, "<base><arg>*");
+ *
+ * where <base> is the output base expressed as a control character, e.g.
+ * \10 gives octal; \20 gives hex.  Each arg is a sequence of characters,
+ * the first of which gives the bit number to be inspected (origin 1), and
+ * the next characters (up to a control character, i.e. a character <= 32),
+ * give the name of the register.  Thus:
+ *
+ *	kvprintf("reg=%b\n", 3, "\10\2BITTWO\1BITONE\n");
+ *
+ * would produce output:
+ *
+ *	reg=3<BITTWO,BITONE>
+ *
+ * XXX:  %D  -- Hexdump, takes pointer and separator string:
+ *		("%6D", ptr, ":")   -> XX:XX:XX:XX:XX:XX
+ *		("%*D", len, ptr, " " -> XX XX XX XX ...
+ */
+static int
+kvprintf(char const *fmt, void (*func)(int), void *arg, int radix, va_list ap)
+{
+#define PCHAR(c) {int cc=(c); if (func) (*func)(cc); else *d++ = cc; retval++; }
+	char nbuf[MAXNBUF];
+	char *d;
+	const char *p, *percent, *q;
+	unsigned char *up;
+	int ch, n;
+	uintmax_t num;
+	int base, lflag, qflag, tmp, width, ladjust, sharpflag, neg, sign, dot;
+	int cflag, hflag, jflag, tflag, zflag;
+	int dwidth, upper;
+	char padc;
+	int stop = 0, retval = 0;
+
+	num = 0;
+	if (!func)
+		d = (char *) arg;
+	else
+		d = NULL;
+
+	if (fmt == NULL)
+		fmt = "(fmt null)\n";
+
+	if (radix < 2 || radix > 36)
+		radix = 10;
+
+	for (;;) {
+		padc = ' ';
+		width = 0;
+		while ((ch = (unsigned char)*fmt++) != '%' || stop) {
+			if (ch == '\0')
+				return (retval);
+			PCHAR(ch);
+		}
+		percent = fmt - 1;
+		qflag = 0; lflag = 0; ladjust = 0; sharpflag = 0; neg = 0;
+		sign = 0; dot = 0; dwidth = 0; upper = 0;
+		cflag = 0; hflag = 0; jflag = 0; tflag = 0; zflag = 0;
+reswitch:	switch (ch = (unsigned char)*fmt++) {
+		case '.':
+			dot = 1;
+			goto reswitch;
+		case '#':
+			sharpflag = 1;
+			goto reswitch;
+		case '+':
+			sign = 1;
+			goto reswitch;
+		case '-':
+			ladjust = 1;
+			goto reswitch;
+		case '%':
+			PCHAR(ch);
+			break;
+		case '*':
+			if (!dot) {
+				width = va_arg(ap, int);
+				if (width < 0) {
+					ladjust = !ladjust;
+					width = -width;
+				}
+			} else {
+				dwidth = va_arg(ap, int);
+			}
+			goto reswitch;
+		case '0':
+			if (!dot) {
+				padc = '0';
+				goto reswitch;
+			}
+		case '1': case '2': case '3': case '4':
+		case '5': case '6': case '7': case '8': case '9':
+				for (n = 0;; ++fmt) {
+					n = n * 10 + ch - '0';
+					ch = *fmt;
+					if (ch < '0' || ch > '9')
+						break;
+				}
+			if (dot)
+				dwidth = n;
+			else
+				width = n;
+			goto reswitch;
+		case 'b':
+			num = (unsigned int)va_arg(ap, int);
+			p = va_arg(ap, char *);
+			for (q = ksprintn(nbuf, num, *p++, NULL, 0); *q;)
+				PCHAR(*q--);
+
+			if (num == 0)
+				break;
+
+			for (tmp = 0; *p;) {
+				n = *p++;
+				if (num & (1 << (n - 1))) {
+					PCHAR(tmp ? ',' : '<');
+					for (; (n = *p) > ' '; ++p)
+						PCHAR(n);
+					tmp = 1;
+				} else
+					for (; *p > ' '; ++p)
+						continue;
+			}
+			if (tmp)
+				PCHAR('>');
+			break;
+		case 'c':
+			PCHAR(va_arg(ap, int));
+			break;
+		case 'D':
+			up = va_arg(ap, unsigned char *);
+			p = va_arg(ap, char *);
+			if (!width)
+				width = 16;
+			while(width--) {
+				PCHAR(hex2ascii(*up >> 4));
+				PCHAR(hex2ascii(*up & 0x0f));
+				up++;
+				if (width)
+					for (q=p;*q;q++)
+						PCHAR(*q);
+			}
+			break;
+		case 'd':
+		case 'i':
+			base = 10;
+			sign = 1;
+			goto handle_sign;
+		case 'h':
+			if (hflag) {
+				hflag = 0;
+				cflag = 1;
+			} else
+				hflag = 1;
+			goto reswitch;
+		case 'j':
+			jflag = 1;
+			goto reswitch;
+		case 'l':
+			if (lflag) {
+				lflag = 0;
+				qflag = 1;
+			} else
+				lflag = 1;
+			goto reswitch;
+		case 'n':
+			if (jflag)
+				*(va_arg(ap, intmax_t *)) = retval;
+			else if (qflag)
+				*(va_arg(ap, int64_t *)) = retval;
+			else if (lflag)
+				*(va_arg(ap, long *)) = retval;
+			else if (zflag)
+				*(va_arg(ap, size_t *)) = retval;
+			else if (hflag)
+				*(va_arg(ap, short *)) = retval;
+			else if (cflag)
+				*(va_arg(ap, char *)) = retval;
+			else
+				*(va_arg(ap, int *)) = retval;
+			break;
+		case 'o':
+			base = 8;
+			goto handle_nosign;
+		case 'p':
+			base = 16;
+			sharpflag = (width == 0);
+			sign = 0;
+			num = (uintptr_t)va_arg(ap, void *);
+			goto number;
+		case 'q':
+			qflag = 1;
+			goto reswitch;
+		case 'r':
+			base = radix;
+			if (sign)
+				goto handle_sign;
+			goto handle_nosign;
+		case 's':
+			p = va_arg(ap, char *);
+			if (p == NULL)
+				p = "(null)";
+			if (!dot)
+				n = (int)strlen (p);
+			else
+				for (n = 0; n < dwidth && p[n]; n++)
+					continue;
+
+			width -= n;
+
+			if (!ladjust && width > 0)
+				while (width--)
+					PCHAR(padc);
+			while (n--)
+				PCHAR(*p++);
+			if (ladjust && width > 0)
+				while (width--)
+					PCHAR(padc);
+			break;
+		case 't':
+			tflag = 1;
+			goto reswitch;
+		case 'u':
+			base = 10;
+			goto handle_nosign;
+		case 'X':
+			upper = 1;
+		case 'x':
+			base = 16;
+			goto handle_nosign;
+		case 'y':
+			base = 16;
+			sign = 1;
+			goto handle_sign;
+		case 'z':
+			zflag = 1;
+			goto reswitch;
+handle_nosign:
+			sign = 0;
+			if (jflag)
+				num = va_arg(ap, uintmax_t);
+			else if (qflag)
+				num = va_arg(ap, uint64_t);
+			else if (tflag)
+				num = va_arg(ap, ptrdiff_t);
+			else if (lflag)
+				num = va_arg(ap, unsigned long);
+			else if (zflag)
+				num = va_arg(ap, size_t);
+			else if (hflag)
+				num = (unsigned short)va_arg(ap, int);
+			else if (cflag)
+				num = (unsigned char)va_arg(ap, int);
+			else
+				num = va_arg(ap, unsigned int);
+			goto number;
+handle_sign:
+			if (jflag)
+				num = va_arg(ap, intmax_t);
+			else if (qflag)
+				num = va_arg(ap, int64_t);
+			else if (tflag)
+				num = va_arg(ap, ptrdiff_t);
+			else if (lflag)
+				num = va_arg(ap, long);
+			else if (zflag)
+				num = va_arg(ap, ssize_t);
+			else if (hflag)
+				num = (short)va_arg(ap, int);
+			else if (cflag)
+				num = (char)va_arg(ap, int);
+			else
+				num = va_arg(ap, int);
+number:
+			if (sign && (intmax_t)num < 0) {
+				neg = 1;
+				num = -(intmax_t)num;
+			}
+			p = ksprintn(nbuf, num, base, &n, upper);
+			tmp = 0;
+			if (sharpflag && num != 0) {
+				if (base == 8)
+					tmp++;
+				else if (base == 16)
+					tmp += 2;
+			}
+			if (neg)
+				tmp++;
+
+			if (!ladjust && padc == '0')
+				dwidth = width - tmp;
+			width -= tmp + imax(dwidth, n);
+			dwidth -= n;
+			if (!ladjust)
+				while (width-- > 0)
+					PCHAR(' ');
+			if (neg)
+				PCHAR('-');
+			if (sharpflag && num != 0) {
+				if (base == 8) {
+					PCHAR('0');
+				} else if (base == 16) {
+					PCHAR('0');
+					PCHAR('x');
+				}
+			}
+			while (dwidth-- > 0)
+				PCHAR('0');
+
+			while (*p)
+				PCHAR(*p--);
+
+			if (ladjust)
+				while (width-- > 0)
+					PCHAR(' ');
+
+			break;
+		default:
+			while (percent < fmt)
+				PCHAR(*percent++);
+			/*
+			 * Since we ignore an formatting argument it is no
+			 * longer safe to obey the remaining formatting
+			 * arguments as the arguments will no longer match
+			 * the format specs.
+			 */
+			stop = 1;
+			break;
+		}
+	}
+#undef PCHAR
+}
+#else /* defined(USE_ELF) */
+#error not supported
+#endif
diff --git a/addr2line.h b/addr2line.h
new file mode 100644
index 0000000000..4f6cf179ef
--- /dev/null
+++ b/addr2line.h
@@ -0,0 +1,21 @@
+/**********************************************************************
+
+  addr2line.h -
+
+  $Author$
+
+  Copyright (C) 2010 Shinichiro Hamaji
+
+**********************************************************************/
+
+#ifndef RUBY_ADDR2LINE_H
+#define RUBY_ADDR2LINE_H
+
+#if (defined(USE_ELF) || defined(HAVE_MACH_O_LOADER_H))
+
+void
+rb_dump_backtrace_with_lines(int num_traces, void **traces);
+
+#endif /* USE_ELF */
+
+#endif /* RUBY_ADDR2LINE_H */
diff --git a/appveyor.yml b/appveyor.yml
new file mode 100644
index 0000000000..3bf1568cd9
--- /dev/null
+++ b/appveyor.yml
@@ -0,0 +1,145 @@
+---
+version: '{build}'
+shallow_clone: true
+platform:
+  - x64
+environment:
+  ruby_version: "24-%Platform%"
+  zlib_version: "1.2.11"
+  matrix:
+    # to reduce time for finishing all jobs, run the slowest msys2 build first.
+    - build: msys2
+      APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2015
+    - build: vs
+      vs: 120
+      APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2013
+    - build: vs
+      vs: 140
+      APPVEYOR_BUILD_WORKER_IMAGE: Visual Studio 2015
+  GEMS_FOR_TEST: "timezone tzinfo"
+notifications:
+  - provider: Webhook
+    url:
+      secure: iMINHMS0nZabaDsxN9omRDsekxzVvAAzEq5ev7lN6vZ6r9zNhl3/F/7POIVyahAwVFpRDeQT/iUugpAGWmOt3eGL/lZIdqiJFZ9DjPSkP68= # #alerts
+    method: POST
+    # "icon_url" is the url used by `provider: Slack`
+    body: >-
+      {{^isPullRequest}}
+        {
+          "attachments": [
+            {
+              "title": "Build {{projectName}} {{buildVersion}} {{status}}",
+              "fallback": "AppVeyor Build {{projectName}} {{buildVersion}} {{status}}",
+              "title_link": "{{buildUrl}}",
+              "text": "Commit <{{commitUrl}}|{{commitId}}> by {{commitAuthor}} on {{commitDate}}: _{{commitMessage}}_",
+              {{#passed}}
+                "color": "#44ee44"
+              {{/passed}}
+              {{#failed}}
+                "color": "#ee4444"
+              {{/failed}}
+            }
+          ],
+          "icon_url": "https://slack-files2.s3-us-west-2.amazonaws.com/bot_icons/2018-02-10/314363543719_48.png",
+          "username": "AppVeyor CI"
+        }
+      {{/isPullRequest}}
+    on_build_success: false
+    on_build_failure: true
+    on_build_status_changed: true
+for:
+-
+  matrix:
+    only:
+      - build: vs
+  install:
+    - ver
+    - chcp
+    - SET BITS=%Platform:x86=32%
+    - SET BITS=%BITS:x=%
+    - SET OPENSSL_DIR=c:\OpenSSL-Win%BITS%
+    - CALL SET vcvars=%%^VS%VS%COMNTOOLS^%%..\..\VC\vcvarsall.bat
+    - SET vcvars
+    - '"%vcvars%" %Platform:x64=amd64%'
+    - SET ruby_path=C:\Ruby%ruby_version:-x86=%
+    - SET PATH=\usr\local\bin;%ruby_path%\bin;%PATH%;C:\msys64\mingw64\bin;C:\msys64\usr\bin
+    - ruby --version
+    - 'cl'
+    - echo> Makefile srcdir=.
+    - echo>> Makefile MSC_VER=0
+    - echo>> Makefile RT=none
+    - echo>> Makefile RT_VER=0
+    - echo>> Makefile BUILTIN_ENCOBJS=nul
+    - type win32\Makefile.sub >> Makefile
+    - nmake %mflags% touch-unicode-files
+    - nmake %mflags% up incs UNICODE_FILES=.
+    - del Makefile
+    - mkdir \usr\local\bin
+    - mkdir \usr\local\include
+    - mkdir \usr\local\lib
+    - appveyor DownloadFile https://zlib.net/zlib%zlib_version:.=%.zip
+    - 7z x -o%APPVEYOR_BUILD_FOLDER%\ext\zlib zlib%zlib_version:.=%.zip
+    - for %%I in (%OPENSSL_DIR%\*.dll) do mklink /h \usr\local\bin\%%~nxI %%I
+    - mkdir %Platform%-mswin_%vs%
+    - ps: Get-ChildItem "win32" -Recurse | foreach {$_.Attributes = 'Readonly'}
+    - ps: Get-Item $env:Platform"-mswin_"$env:vs | foreach {$_.Attributes = 'Normal'}
+  build_script:
+    - cd %APPVEYOR_BUILD_FOLDER%
+    - cd %Platform%-mswin_%vs%
+    - ..\win32\configure.bat --without-ext=+,dbm,gdbm,readline --with-opt-dir=/usr/local --with-openssl-dir=%OPENSSL_DIR:\=/%
+    - nmake -l
+    - nmake install-nodoc
+    - \usr\bin\ruby -v -e "p :locale => Encoding.find('locale'), :filesystem => Encoding.find('filesystem')"
+    - if not "%GEMS_FOR_TEST%" == "" \usr\bin\gem install --no-document %GEMS_FOR_TEST%
+  test_script:
+    - set /a JOBS=%NUMBER_OF_PROCESSORS%
+    - nmake -l "TESTOPTS=-v -q" btest
+    - nmake -l "TESTOPTS=-v -q" test-basic
+    - nmake -l "TESTOPTS=-q --subprocess-timeout-scale=3.0 --excludes=../test/excludes/_appveyor -j%JOBS% --exclude win32ole --exclude test_bignum --exclude test_syntax --exclude test_open-uri --exclude test_bundled_ca" test-all
+    # separately execute tests without -j which may crash worker with -j.
+    - nmake -l "TESTOPTS=-v --subprocess-timeout-scale=3.0 --excludes=../test/excludes/_appveyor" test-all TESTS="../test/win32ole ../test/ruby/test_bignum.rb ../test/ruby/test_syntax.rb ../test/open-uri/test_open-uri.rb ../test/rubygems/test_bundled_ca.rb"
+    - nmake -l test-spec MSPECOPT=-fs # not using `-j` because sometimes `mspec -j` silently dies on Windows
+-
+  matrix:
+    only:
+      - build: msys2
+  install:
+    - ver
+    - chcp
+    - set /a JOBS=%NUMBER_OF_PROCESSORS%
+    - set MSYS_NO_PATHCONV=1
+    - SET MSYSTEM=%Platform:x86=32%
+    - SET MSYSTEM=%MSYSTEM:x=MINGW%
+    - SET MSYS2_ARCH=%Platform:x86=i686%
+    - SET MSYS2_ARCH=%MSYS2_ARCH:x64=x86_64%
+    - set MSYSTEM_PREFIX=/mingw64
+    - set MINGW_CHOST=%MSYS2_ARCH%-w64-mingw32
+    - SET ruby_path=C:\Ruby%ruby_version:-x86=%
+    - cd ..
+    - mkdir build
+    - mkdir install
+    - SET PATH=%ruby_path%\bin;C:\msys64\%MSYSTEM%\bin;C:\msys64\usr\bin;%PATH%
+    - ruby --version
+  build_script:
+    # always update database
+    - pacman -Sy
+    - pacman -S --noconfirm --needed --noprogressbar mingw-w64-x86_64-toolchain
+    - pacman -S --noconfirm --needed --noprogressbar mingw-w64-x86_64-gdbm mingw-w64-x86_64-gmp mingw-w64-x86_64-libffi mingw-w64-x86_64-ncurses mingw-w64-x86_64-readline mingw-w64-x86_64-zlib
+    - cd %APPVEYOR_BUILD_FOLDER%
+    - set CFLAGS=-march=%MSYS2_ARCH:_=-% -mtune=generic -O3 -pipe
+    - set CXXFLAGS=%CFLAGS%
+    - set CPPFLAGS=-D_FORTIFY_SOURCE=2 -D__USE_MINGW_ANSI_STDIO=1 -DFD_SETSIZE=2048
+    - set LDFLAGS=-pipe
+    - sh -c "autoreconf -fi"
+    - cd ..\build
+    - sh ../ruby/configure --disable-install-doc --prefix=/. --build=%MINGW_CHOST% --host=%MINGW_CHOST% --target=%MINGW_CHOST%
+    - mingw32-make -j%JOBS% up
+    - mingw32-make -j%JOBS%
+    - mingw32-make DESTDIR=../install install-nodoc
+    - if not "%GEMS_FOR_TEST%" == "" ..\install\bin\gem install --no-document %GEMS_FOR_TEST%
+  test_script:
+    - mingw32-make test
+    - mingw32-make test-all TESTOPTS="--retry --job-status=normal --show-skip --subprocess-timeout-scale=1.5 --excludes=../ruby/test/excludes/_appveyor -j %JOBS% --exclude win32ole --exclude test_open-uri"
+    # separately execute tests without -j which may crash worker with -j.
+    - mingw32-make test-all TESTOPTS="--retry --job-status=normal --show-skip --subprocess-timeout-scale=1.5 --excludes=../ruby/test/excludes/_appveyor" TESTS="../ruby/test/win32ole ../ruby/test/open-uri/test_open-uri.rb"
+    - mingw32-make test-spec MSPECOPT=-fs # not using `-j` because sometimes `mspec -j` silently dies on Windows
diff --git a/array.c b/array.c
index baae2706e5..990bc076b1 100644
--- a/array.c
+++ b/array.c
@@ -2,164 +2,703 @@
 
   array.c -
 
-  $Author: akr $
-  $Date: 2006/06/24 14:53:36 $
+  $Author$
   created at: Fri Aug  6 09:46:12 JST 1993
 
-  Copyright (C) 1993-2003 Yukihiro Matsumoto
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
   Copyright (C) 2000  Network Applied Communication Laboratory, Inc.
   Copyright (C) 2000  Information-technology Promotion Agency, Japan
 
 **********************************************************************/
 
-#include "ruby.h"
-#include "util.h"
-#include "st.h"
+#include "ruby/encoding.h"
+#include "ruby/util.h"
+#include "ruby/st.h"
+#include "probes.h"
+#include "id.h"
+#include "debug_counter.h"
+#include "gc.h"
+#include "transient_heap.h"
+#include "internal.h"
+
+#if !ARRAY_DEBUG
+# define NDEBUG
+#endif
+#include "ruby_assert.h"
 
 VALUE rb_cArray;
-static ID id_cmp;
+
+/* for OPTIMIZED_CMP: */
+#define id_cmp idCmp
 
 #define ARY_DEFAULT_SIZE 16
+#define ARY_MAX_SIZE (LONG_MAX / (int)sizeof(VALUE))
+#define SMALL_ARRAY_LEN 16
+
+# define ARY_SHARED_P(ary) \
+    (assert(!FL_TEST((ary), ELTS_SHARED) || !FL_TEST((ary), RARRAY_EMBED_FLAG)), \
+     FL_TEST((ary),ELTS_SHARED)!=0)
+# define ARY_EMBED_P(ary) \
+    (assert(!FL_TEST((ary), ELTS_SHARED) || !FL_TEST((ary), RARRAY_EMBED_FLAG)), \
+     FL_TEST((ary), RARRAY_EMBED_FLAG)!=0)
+
+#define ARY_HEAP_PTR(a) (assert(!ARY_EMBED_P(a)), RARRAY(a)->as.heap.ptr)
+#define ARY_HEAP_LEN(a) (assert(!ARY_EMBED_P(a)), RARRAY(a)->as.heap.len)
+#define ARY_HEAP_CAPA(a) (assert(!ARY_EMBED_P(a)), RARRAY(a)->as.heap.aux.capa)
+
+#define ARY_EMBED_PTR(a) (assert(ARY_EMBED_P(a)), RARRAY(a)->as.ary)
+#define ARY_EMBED_LEN(a) \
+    (assert(ARY_EMBED_P(a)), \
+     (long)((RBASIC(a)->flags >> RARRAY_EMBED_LEN_SHIFT) & \
+	 (RARRAY_EMBED_LEN_MASK >> RARRAY_EMBED_LEN_SHIFT)))
+#define ARY_HEAP_SIZE(a) (assert(!ARY_EMBED_P(a)), assert(ARY_OWNS_HEAP_P(a)), ARY_HEAP_CAPA(a) * sizeof(VALUE))
+
+#define ARY_OWNS_HEAP_P(a) (!FL_TEST((a), ELTS_SHARED|RARRAY_EMBED_FLAG))
+#define FL_SET_EMBED(a) do { \
+    assert(!ARY_SHARED_P(a)); \
+    FL_SET((a), RARRAY_EMBED_FLAG); \
+    RARY_TRANSIENT_UNSET(a); \
+    ary_verify(a); \
+} while (0)
+#define FL_UNSET_EMBED(ary) FL_UNSET((ary), RARRAY_EMBED_FLAG|RARRAY_EMBED_LEN_MASK)
+#define FL_SET_SHARED(ary) do { \
+    assert(!ARY_EMBED_P(ary)); \
+    FL_SET((ary), ELTS_SHARED); \
+} while (0)
+#define FL_UNSET_SHARED(ary) FL_UNSET((ary), ELTS_SHARED)
+
+#define ARY_SET_PTR(ary, p) do { \
+    assert(!ARY_EMBED_P(ary)); \
+    assert(!OBJ_FROZEN(ary)); \
+    RARRAY(ary)->as.heap.ptr = (p); \
+} while (0)
+#define ARY_SET_EMBED_LEN(ary, n) do { \
+    long tmp_n = (n); \
+    assert(ARY_EMBED_P(ary)); \
+    assert(!OBJ_FROZEN(ary)); \
+    RBASIC(ary)->flags &= ~RARRAY_EMBED_LEN_MASK; \
+    RBASIC(ary)->flags |= (tmp_n) << RARRAY_EMBED_LEN_SHIFT; \
+} while (0)
+#define ARY_SET_HEAP_LEN(ary, n) do { \
+    assert(!ARY_EMBED_P(ary)); \
+    RARRAY(ary)->as.heap.len = (n); \
+} while (0)
+#define ARY_SET_LEN(ary, n) do { \
+    if (ARY_EMBED_P(ary)) { \
+        ARY_SET_EMBED_LEN((ary), (n)); \
+    } \
+    else { \
+        ARY_SET_HEAP_LEN((ary), (n)); \
+    } \
+    assert(RARRAY_LEN(ary) == (n)); \
+} while (0)
+#define ARY_INCREASE_PTR(ary, n) do  { \
+    assert(!ARY_EMBED_P(ary)); \
+    assert(!OBJ_FROZEN(ary)); \
+    RARRAY(ary)->as.heap.ptr += (n); \
+} while (0)
+#define ARY_INCREASE_LEN(ary, n) do  { \
+    assert(!OBJ_FROZEN(ary)); \
+    if (ARY_EMBED_P(ary)) { \
+        ARY_SET_EMBED_LEN((ary), RARRAY_LEN(ary)+(n)); \
+    } \
+    else { \
+        RARRAY(ary)->as.heap.len += (n); \
+    } \
+} while (0)
+
+#define ARY_CAPA(ary) (ARY_EMBED_P(ary) ? RARRAY_EMBED_LEN_MAX : \
+                       ARY_SHARED_ROOT_P(ary) ? RARRAY_LEN(ary) : ARY_HEAP_CAPA(ary))
+#define ARY_SET_CAPA(ary, n) do { \
+    assert(!ARY_EMBED_P(ary)); \
+    assert(!ARY_SHARED_P(ary)); \
+    assert(!OBJ_FROZEN(ary)); \
+    RARRAY(ary)->as.heap.aux.capa = (n); \
+} while (0)
+
+#define ARY_SHARED(ary) (assert(ARY_SHARED_P(ary)), RARRAY(ary)->as.heap.aux.shared)
+#define ARY_SET_SHARED(ary, value) do { \
+    const VALUE _ary_ = (ary); \
+    const VALUE _value_ = (value); \
+    assert(!ARY_EMBED_P(_ary_)); \
+    assert(ARY_SHARED_P(_ary_)); \
+    assert(ARY_SHARED_ROOT_P(_value_)); \
+    RB_OBJ_WRITE(_ary_, &RARRAY(_ary_)->as.heap.aux.shared, _value_); \
+} while (0)
+#define RARRAY_SHARED_ROOT_FLAG FL_USER5
+#define ARY_SHARED_ROOT_P(ary) (FL_TEST((ary), RARRAY_SHARED_ROOT_FLAG))
+#define ARY_SHARED_NUM(ary) \
+    (assert(ARY_SHARED_ROOT_P(ary)), RARRAY(ary)->as.heap.aux.capa)
+#define ARY_SHARED_OCCUPIED(ary) (ARY_SHARED_NUM(ary) == 1)
+#define ARY_SET_SHARED_NUM(ary, value) do { \
+    assert(ARY_SHARED_ROOT_P(ary)); \
+    RARRAY(ary)->as.heap.aux.capa = (value); \
+} while (0)
+#define FL_SET_SHARED_ROOT(ary) do { \
+    assert(!ARY_EMBED_P(ary)); \
+    assert(!RARRAY_TRANSIENT_P(ary)); \
+    FL_SET((ary), RARRAY_SHARED_ROOT_FLAG); \
+} while (0)
+
+#define ARY_SET(a, i, v) RARRAY_ASET((assert(!ARY_SHARED_P(a)), (a)), (i), (v))
+
+
+#if ARRAY_DEBUG
+#define ary_verify(ary) ary_verify_(ary, __FILE__, __LINE__)
+
+static VALUE
+ary_verify_(VALUE ary, const char *file, int line)
+{
+    assert(RB_TYPE_P(ary, T_ARRAY));
+
+    if (FL_TEST(ary, ELTS_SHARED)) {
+        VALUE root = RARRAY(ary)->as.heap.aux.shared;
+        const VALUE *ptr = ARY_HEAP_PTR(ary);
+        const VALUE *root_ptr = RARRAY_CONST_PTR_TRANSIENT(root);
+        long len = ARY_HEAP_LEN(ary), root_len = RARRAY_LEN(root);
+        assert(FL_TEST(root, RARRAY_SHARED_ROOT_FLAG));
+        assert(root_ptr <= ptr && ptr + len <= root_ptr + root_len);
+        ary_verify(root);
+    }
+    else if (ARY_EMBED_P(ary)) {
+        assert(!RARRAY_TRANSIENT_P(ary));
+        assert(!ARY_SHARED_P(ary));
+        assert(RARRAY_LEN(ary) <= RARRAY_EMBED_LEN_MAX);
+    }
+    else {
+#if 1
+        const VALUE *ptr = RARRAY_CONST_PTR_TRANSIENT(ary);
+        long i, len = RARRAY_LEN(ary);
+        volatile VALUE v;
+        if (len > 1) len = 1; /* check only HEAD */
+        for (i=0; i<len; i++) {
+            v = ptr[i]; /* access check */
+        }
+        v = v;
+#endif
+    }
+
+    if (RARRAY_TRANSIENT_P(ary)) {
+        assert(rb_transient_heap_managed_ptr_p(RARRAY_CONST_PTR_TRANSIENT(ary)));
+    }
+
+    rb_transient_heap_verify();
+
+    return ary;
+}
 
 void
-rb_mem_clear(mem, size)
-    register VALUE *mem;
-    register long size;
+rb_ary_verify(VALUE ary){
+    ary_verify(ary);
+}
+#else
+#define ary_verify(ary) ((void)0)
+#endif
+
+VALUE *
+rb_ary_ptr_use_start(VALUE ary)
+{
+#if ARRAY_DEBUG
+    FL_SET_RAW(ary, RARRAY_PTR_IN_USE_FLAG);
+#endif
+    return (VALUE *)RARRAY_CONST_PTR_TRANSIENT(ary);
+}
+
+void
+rb_ary_ptr_use_end(VALUE ary)
+{
+#if ARRAY_DEBUG
+    FL_UNSET_RAW(ary, RARRAY_PTR_IN_USE_FLAG);
+#endif
+}
+
+void
+rb_mem_clear(register VALUE *mem, register long size)
 {
     while (size--) {
 	*mem++ = Qnil;
     }
 }
 
+static void
+ary_mem_clear(VALUE ary, long beg, long size)
+{
+    RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+	rb_mem_clear(ptr + beg, size);
+    });
+}
+
 static inline void
-memfill(mem, size, val)
-    register VALUE *mem;
-    register long size;
-    register VALUE val;
+memfill(register VALUE *mem, register long size, register VALUE val)
 {
     while (size--) {
 	*mem++ = val;
     }
 }
 
-#define ARY_TMPLOCK  FL_USER1
+static void
+ary_memfill(VALUE ary, long beg, long size, VALUE val)
+{
+    RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+	memfill(ptr + beg, size, val);
+	RB_OBJ_WRITTEN(ary, Qundef, val);
+    });
+}
+
+static void
+ary_memcpy0(VALUE ary, long beg, long argc, const VALUE *argv, VALUE buff_owner_ary)
+{
+    assert(!ARY_SHARED_P(buff_owner_ary));
+
+    if (argc > (int)(128/sizeof(VALUE)) /* is magic number (cache line size) */) {
+        rb_gc_writebarrier_remember(buff_owner_ary);
+        RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+            MEMCPY(ptr+beg, argv, VALUE, argc);
+        });
+    }
+    else {
+        int i;
+        RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+            for (i=0; i<argc; i++) {
+                RB_OBJ_WRITE(buff_owner_ary, &ptr[i+beg], argv[i]);
+            }
+        });
+    }
+}
+
+static void
+ary_memcpy(VALUE ary, long beg, long argc, const VALUE *argv)
+{
+    ary_memcpy0(ary, beg, argc, argv, ary);
+}
+
+static VALUE *
+ary_heap_alloc(VALUE ary, size_t capa)
+{
+    VALUE *ptr = rb_transient_heap_alloc(ary, sizeof(VALUE) * capa);
+
+    if (ptr != NULL) {
+        RARY_TRANSIENT_SET(ary);
+    }
+    else {
+        RARY_TRANSIENT_UNSET(ary);
+        ptr = ALLOC_N(VALUE, capa);
+    }
+
+    return ptr;
+}
+
+static void
+ary_heap_free_ptr(VALUE ary, const VALUE *ptr, long size)
+{
+    if (RARRAY_TRANSIENT_P(ary)) {
+        /* ignore it */
+    }
+    else {
+        ruby_sized_xfree((void *)ptr, size);
+    }
+}
+
+static void
+ary_heap_free(VALUE ary)
+{
+    if (RARRAY_TRANSIENT_P(ary)) {
+        RARY_TRANSIENT_UNSET(ary);
+    }
+    else {
+        ary_heap_free_ptr(ary, ARY_HEAP_PTR(ary), ARY_HEAP_SIZE(ary));
+    }
+}
+
+static void
+ary_heap_realloc(VALUE ary, size_t new_capa)
+{
+    size_t old_capa = ARY_HEAP_CAPA(ary);
+
+    if (RARRAY_TRANSIENT_P(ary)) {
+        if (new_capa <= old_capa) {
+            /* do nothing */
+        }
+        else {
+            VALUE *new_ptr = rb_transient_heap_alloc(ary, sizeof(VALUE) * new_capa);
+
+            if (new_ptr == NULL) {
+                new_ptr = ALLOC_N(VALUE, new_capa);
+                RARY_TRANSIENT_UNSET(ary);
+            }
+
+            MEMCPY(new_ptr, ARY_HEAP_PTR(ary), VALUE, old_capa);
+            ARY_SET_PTR(ary, new_ptr);
+        }
+    }
+    else {
+        SIZED_REALLOC_N(RARRAY(ary)->as.heap.ptr, VALUE, new_capa, old_capa);
+    }
+    ary_verify(ary);
+}
 
+#if USE_TRANSIENT_HEAP
 static inline void
-rb_ary_modify_check(ary)
-    VALUE ary;
+rb_ary_transient_heap_evacuate_(VALUE ary, int transient, int promote)
+{
+    if (transient) {
+        VALUE *new_ptr;
+        const VALUE *old_ptr = ARY_HEAP_PTR(ary);
+        long capa = ARY_HEAP_CAPA(ary);
+        long len  = ARY_HEAP_LEN(ary);
+
+        if (ARY_SHARED_ROOT_P(ary)) {
+            capa = len;
+        }
+
+        assert(ARY_OWNS_HEAP_P(ary));
+        assert(RARRAY_TRANSIENT_P(ary));
+        assert(!ARY_PTR_USING_P(ary));
+
+        if (promote) {
+            new_ptr = ALLOC_N(VALUE, capa);
+            RARY_TRANSIENT_UNSET(ary);
+        }
+        else {
+            new_ptr = ary_heap_alloc(ary, capa);
+        }
+
+        MEMCPY(new_ptr, old_ptr, VALUE, capa);
+        /* do not use ARY_SET_PTR() because they assert !frozen */
+        RARRAY(ary)->as.heap.ptr = new_ptr;
+    }
+
+    ary_verify(ary);
+}
+
+void
+rb_ary_transient_heap_evacuate(VALUE ary, int promote)
+{
+    rb_ary_transient_heap_evacuate_(ary, RARRAY_TRANSIENT_P(ary), promote);
+}
+
+void
+rb_ary_detransient(VALUE ary)
+{
+    assert(RARRAY_TRANSIENT_P(ary));
+    rb_ary_transient_heap_evacuate_(ary, TRUE, TRUE);
+}
+#else
+void
+rb_ary_detransient(VALUE ary)
 {
-    if (OBJ_FROZEN(ary)) rb_error_frozen("array");
-    if (FL_TEST(ary, ARY_TMPLOCK))
-	rb_raise(rb_eRuntimeError, "can't modify array during iteration");
-    if (!OBJ_TAINTED(ary) && rb_safe_level() >= 4)
-	rb_raise(rb_eSecurityError, "Insecure: can't modify array");
+    /* do nothing */
 }
+#endif
 
 static void
-rb_ary_modify(ary)
-    VALUE ary;
+ary_resize_capa(VALUE ary, long capacity)
+{
+    assert(RARRAY_LEN(ary) <= capacity);
+    assert(!OBJ_FROZEN(ary));
+    assert(!ARY_SHARED_P(ary));
+
+    if (capacity > RARRAY_EMBED_LEN_MAX) {
+        if (ARY_EMBED_P(ary)) {
+            long len = ARY_EMBED_LEN(ary);
+            VALUE *ptr = ary_heap_alloc(ary, capacity);
+
+            MEMCPY(ptr, ARY_EMBED_PTR(ary), VALUE, len);
+            FL_UNSET_EMBED(ary);
+            ARY_SET_PTR(ary, ptr);
+            ARY_SET_HEAP_LEN(ary, len);
+        }
+        else {
+            ary_heap_realloc(ary, capacity);
+        }
+        ARY_SET_CAPA(ary, capacity);
+    }
+    else {
+        if (!ARY_EMBED_P(ary)) {
+            long len = ARY_HEAP_LEN(ary);
+            long old_capa = ARY_HEAP_CAPA(ary);
+            const VALUE *ptr = ARY_HEAP_PTR(ary);
+
+            if (len > capacity) len = capacity;
+            MEMCPY((VALUE *)RARRAY(ary)->as.ary, ptr, VALUE, len);
+            ary_heap_free_ptr(ary, ptr, old_capa);
+
+            FL_SET_EMBED(ary);
+            ARY_SET_LEN(ary, len);
+        }
+    }
+
+    ary_verify(ary);
+}
+
+static inline void
+ary_shrink_capa(VALUE ary)
+{
+    long capacity = ARY_HEAP_LEN(ary);
+    long old_capa = ARY_HEAP_CAPA(ary);
+    assert(!ARY_SHARED_P(ary));
+    assert(old_capa >= capacity);
+    if (old_capa > capacity) ary_heap_realloc(ary, capacity);
+
+    ary_verify(ary);
+}
+
+static void
+ary_double_capa(VALUE ary, long min)
+{
+    long new_capa = ARY_CAPA(ary) / 2;
+
+    if (new_capa < ARY_DEFAULT_SIZE) {
+	new_capa = ARY_DEFAULT_SIZE;
+    }
+    if (new_capa >= ARY_MAX_SIZE - min) {
+	new_capa = (ARY_MAX_SIZE - min) / 2;
+    }
+    new_capa += min;
+    ary_resize_capa(ary, new_capa);
+
+    ary_verify(ary);
+}
+
+static void
+rb_ary_decrement_share(VALUE shared)
+{
+    if (shared) {
+	long num = ARY_SHARED_NUM(shared) - 1;
+	if (num == 0) {
+	    rb_ary_free(shared);
+	    rb_gc_force_recycle(shared);
+	}
+	else if (num > 0) {
+	    ARY_SET_SHARED_NUM(shared, num);
+	}
+    }
+}
+
+static void
+rb_ary_unshare(VALUE ary)
+{
+    VALUE shared = RARRAY(ary)->as.heap.aux.shared;
+    rb_ary_decrement_share(shared);
+    FL_UNSET_SHARED(ary);
+}
+
+static inline void
+rb_ary_unshare_safe(VALUE ary)
 {
-    VALUE *ptr;
+    if (ARY_SHARED_P(ary) && !ARY_EMBED_P(ary)) {
+	rb_ary_unshare(ary);
+    }
+}
 
+static VALUE
+rb_ary_increment_share(VALUE shared)
+{
+    long num = ARY_SHARED_NUM(shared);
+    if (num >= 0) {
+	ARY_SET_SHARED_NUM(shared, num + 1);
+    }
+    return shared;
+}
+
+static void
+rb_ary_set_shared(VALUE ary, VALUE shared)
+{
+    rb_ary_increment_share(shared);
+    FL_SET_SHARED(ary);
+    ARY_SET_SHARED(ary, shared);
+}
+
+static inline void
+rb_ary_modify_check(VALUE ary)
+{
+    rb_check_frozen(ary);
+    ary_verify(ary);
+}
+
+void
+rb_ary_modify(VALUE ary)
+{
     rb_ary_modify_check(ary);
-    if (FL_TEST(ary, ELTS_SHARED)) {
-	ptr = ALLOC_N(VALUE, RARRAY(ary)->len);
-	FL_UNSET(ary, ELTS_SHARED);
-	RARRAY(ary)->aux.capa = RARRAY(ary)->len;
-	MEMCPY(ptr, RARRAY(ary)->ptr, VALUE, RARRAY(ary)->len);
-	RARRAY(ary)->ptr = ptr;
+    if (ARY_SHARED_P(ary)) {
+	long shared_len, len = RARRAY_LEN(ary);
+	VALUE shared = ARY_SHARED(ary);
+
+        ary_verify(shared);
+
+        if (len <= RARRAY_EMBED_LEN_MAX) {
+	    const VALUE *ptr = ARY_HEAP_PTR(ary);
+            FL_UNSET_SHARED(ary);
+            FL_SET_EMBED(ary);
+	    MEMCPY((VALUE *)ARY_EMBED_PTR(ary), ptr, VALUE, len);
+            rb_ary_decrement_share(shared);
+            ARY_SET_EMBED_LEN(ary, len);
+        }
+	else if (ARY_SHARED_OCCUPIED(shared) && len > ((shared_len = RARRAY_LEN(shared))>>1)) {
+            long shift = RARRAY_CONST_PTR_TRANSIENT(ary) - RARRAY_CONST_PTR_TRANSIENT(shared);
+	    FL_UNSET_SHARED(ary);
+            ARY_SET_PTR(ary, RARRAY_CONST_PTR_TRANSIENT(shared));
+	    ARY_SET_CAPA(ary, shared_len);
+            RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+		MEMMOVE(ptr, ptr+shift, VALUE, len);
+	    });
+	    FL_SET_EMBED(shared);
+	    rb_ary_decrement_share(shared);
+	}
+        else {
+            VALUE *ptr = ary_heap_alloc(ary, len);
+            MEMCPY(ptr, ARY_HEAP_PTR(ary), VALUE, len);
+            rb_ary_unshare(ary);
+            ARY_SET_CAPA(ary, len);
+            ARY_SET_PTR(ary, ptr);
+        }
+
+	rb_gc_writebarrier_remember(ary);
     }
+    ary_verify(ary);
 }
 
-VALUE
-rb_ary_freeze(ary)
-    VALUE ary;
+static VALUE
+ary_ensure_room_for_push(VALUE ary, long add_len)
 {
-    return rb_obj_freeze(ary);
+    long old_len = RARRAY_LEN(ary);
+    long new_len = old_len + add_len;
+    long capa;
+
+    if (old_len > ARY_MAX_SIZE - add_len) {
+	rb_raise(rb_eIndexError, "index %ld too big", new_len);
+    }
+    if (ARY_SHARED_P(ary)) {
+	if (new_len > RARRAY_EMBED_LEN_MAX) {
+	    VALUE shared = ARY_SHARED(ary);
+	    if (ARY_SHARED_OCCUPIED(shared)) {
+                if (ARY_HEAP_PTR(ary) - RARRAY_CONST_PTR_TRANSIENT(shared) + new_len <= RARRAY_LEN(shared)) {
+		    rb_ary_modify_check(ary);
+
+                    ary_verify(ary);
+                    ary_verify(shared);
+                    return shared;
+		}
+		else {
+		    /* if array is shared, then it is likely it participate in push/shift pattern */
+		    rb_ary_modify(ary);
+		    capa = ARY_CAPA(ary);
+		    if (new_len > capa - (capa >> 6)) {
+			ary_double_capa(ary, new_len);
+		    }
+                    ary_verify(ary);
+		    return ary;
+		}
+	    }
+	}
+        ary_verify(ary);
+        rb_ary_modify(ary);
+    }
+    else {
+	rb_ary_modify_check(ary);
+    }
+    capa = ARY_CAPA(ary);
+    if (new_len > capa) {
+	ary_double_capa(ary, new_len);
+    }
+
+    ary_verify(ary);
+    return ary;
 }
 
 /*
  *  call-seq:
- *     array.frozen?  -> true or false
+ *      ary.freeze -> ary
+ *
+ *  Calls Object#freeze on +ary+ to prevent any further
+ *  modification. A RuntimeError will be raised if a modification
+ *  attempt is made.
  *
- *  Return <code>true</code> if this array is frozen (or temporarily frozen
- *  while being sorted).
  */
 
-static VALUE
-rb_ary_frozen_p(ary)
-    VALUE ary;
+VALUE
+rb_ary_freeze(VALUE ary)
+{
+    return rb_obj_freeze(ary);
+}
+
+/* This can be used to take a snapshot of an array (with
+   e.g. rb_ary_replace) and check later whether the array has been
+   modified from the snapshot.  The snapshot is cheap, though if
+   something does modify the array it will pay the cost of copying
+   it.  If Array#pop or Array#shift has been called, the array will
+   be still shared with the snapshot, but the array length will
+   differ. */
+VALUE
+rb_ary_shared_with_p(VALUE ary1, VALUE ary2)
 {
-    if (OBJ_FROZEN(ary)) return Qtrue;
-    if (FL_TEST(ary, ARY_TMPLOCK)) return Qtrue;
+    if (!ARY_EMBED_P(ary1) && ARY_SHARED_P(ary1) &&
+	!ARY_EMBED_P(ary2) && ARY_SHARED_P(ary2) &&
+	RARRAY(ary1)->as.heap.aux.shared == RARRAY(ary2)->as.heap.aux.shared &&
+	RARRAY(ary1)->as.heap.len == RARRAY(ary2)->as.heap.len) {
+	return Qtrue;
+    }
     return Qfalse;
 }
 
-static VALUE ary_alloc _((VALUE));
 static VALUE
-ary_alloc(klass)
-    VALUE klass;
+ary_alloc(VALUE klass)
 {
-    NEWOBJ(ary, struct RArray);
-    OBJSETUP(ary, klass, T_ARRAY);
-
-    ary->len = 0;
-    ary->ptr = 0;
-    ary->aux.capa = 0;
-
+    NEWOBJ_OF(ary, struct RArray, klass, T_ARRAY | RARRAY_EMBED_FLAG | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0));
+    /* Created array is:
+     *   FL_SET_EMBED((VALUE)ary);
+     *   ARY_SET_EMBED_LEN((VALUE)ary, 0);
+     */
     return (VALUE)ary;
 }
 
 static VALUE
-ary_new(klass, len)
-    VALUE klass;
-    long len;
+empty_ary_alloc(VALUE klass)
 {
-    VALUE ary = ary_alloc(klass);
+    RUBY_DTRACE_CREATE_HOOK(ARRAY, 0);
+    return ary_alloc(klass);
+}
 
-    if (len < 0) {
+static VALUE
+ary_new(VALUE klass, long capa)
+{
+    VALUE ary,*ptr;
+
+    if (capa < 0) {
 	rb_raise(rb_eArgError, "negative array size (or size too big)");
     }
-    if (len > 0 && len * sizeof(VALUE) <= len) {
+    if (capa > ARY_MAX_SIZE) {
 	rb_raise(rb_eArgError, "array size too big");
     }
-    if (len == 0) len++;
-    RARRAY(ary)->ptr = ALLOC_N(VALUE, len);
-    RARRAY(ary)->aux.capa = len;
+
+    RUBY_DTRACE_CREATE_HOOK(ARRAY, capa);
+
+    ary = ary_alloc(klass);
+    if (capa > RARRAY_EMBED_LEN_MAX) {
+        ptr = ary_heap_alloc(ary, capa);
+        FL_UNSET_EMBED(ary);
+        ARY_SET_PTR(ary, ptr);
+        ARY_SET_CAPA(ary, capa);
+        ARY_SET_HEAP_LEN(ary, 0);
+    }
 
     return ary;
 }
 
 VALUE
-rb_ary_new2(len)
-    long len;
+rb_ary_new_capa(long capa)
 {
-    return ary_new(rb_cArray, len);
+    return ary_new(rb_cArray, capa);
 }
 
-
 VALUE
-rb_ary_new()
+rb_ary_new(void)
 {
-    return rb_ary_new2(ARY_DEFAULT_SIZE);
+    return rb_ary_new2(RARRAY_EMBED_LEN_MAX);
 }
 
-#ifdef HAVE_STDARG_PROTOTYPES
-#include <stdarg.h>
-#define va_init_list(a,b) va_start(a,b)
-#else
-#include <varargs.h>
-#define va_init_list(a,b) va_start(a)
-#endif
-
 VALUE
-#ifdef HAVE_STDARG_PROTOTYPES
-rb_ary_new3(long n, ...)
-#else
-rb_ary_new3(n, va_alist)
-    long n;
-    va_dcl
-#endif
+(rb_ary_new_from_args)(long n, ...)
 {
     va_list ar;
     VALUE ary;
@@ -167,120 +706,275 @@ rb_ary_new3(n, va_alist)
 
     ary = rb_ary_new2(n);
 
-    va_init_list(ar, n);
+    va_start(ar, n);
     for (i=0; i<n; i++) {
-	RARRAY(ary)->ptr[i] = va_arg(ar, VALUE);
+	ARY_SET(ary, i, va_arg(ar, VALUE));
     }
     va_end(ar);
 
-    RARRAY(ary)->len = n;
+    ARY_SET_LEN(ary, n);
     return ary;
 }
 
-VALUE
-rb_ary_new4(n, elts)
-    long n;
-    const VALUE *elts;
+MJIT_FUNC_EXPORTED VALUE
+rb_ary_tmp_new_from_values(VALUE klass, long n, const VALUE *elts)
 {
     VALUE ary;
 
-    ary = rb_ary_new2(n);
+    ary = ary_new(klass, n);
     if (n > 0 && elts) {
-	MEMCPY(RARRAY(ary)->ptr, elts, VALUE, n);
+	ary_memcpy(ary, 0, n, elts);
+	ARY_SET_LEN(ary, n);
     }
 
-    /* This assignment to len will be moved to the above "if" block in Ruby 1.9 */
-    RARRAY(ary)->len = n;
-
     return ary;
 }
 
 VALUE
-rb_assoc_new(car, cdr)
-    VALUE car, cdr;
+rb_ary_new_from_values(long n, const VALUE *elts)
 {
-    VALUE ary;
+    return rb_ary_tmp_new_from_values(rb_cArray, n, elts);
+}
 
-    ary = rb_ary_new2(2);
-    RARRAY(ary)->ptr[0] = car;
-    RARRAY(ary)->ptr[1] = cdr;
-    RARRAY(ary)->len = 2;
+VALUE
+rb_ary_tmp_new(long capa)
+{
+    VALUE ary = ary_new(0, capa);
+    rb_ary_transient_heap_evacuate(ary, TRUE);
+    return ary;
+}
 
+VALUE
+rb_ary_tmp_new_fill(long capa)
+{
+    VALUE ary = ary_new(0, capa);
+    ary_memfill(ary, 0, capa, Qnil);
+    ARY_SET_LEN(ary, capa);
+    rb_ary_transient_heap_evacuate(ary, TRUE);
     return ary;
 }
 
+void
+rb_ary_free(VALUE ary)
+{
+    if (ARY_OWNS_HEAP_P(ary)) {
+        if (RARRAY_TRANSIENT_P(ary)) {
+            RB_DEBUG_COUNTER_INC(obj_ary_transient);
+        }
+        else {
+            RB_DEBUG_COUNTER_INC(obj_ary_ptr);
+            ary_heap_free(ary);
+        }
+    }
+    else {
+	RB_DEBUG_COUNTER_INC(obj_ary_embed);
+    }
+}
+
+RUBY_FUNC_EXPORTED size_t
+rb_ary_memsize(VALUE ary)
+{
+    if (ARY_OWNS_HEAP_P(ary)) {
+	return ARY_CAPA(ary) * sizeof(VALUE);
+    }
+    else {
+	return 0;
+    }
+}
+
+static inline void
+ary_discard(VALUE ary)
+{
+    rb_ary_free(ary);
+    RBASIC(ary)->flags |= RARRAY_EMBED_FLAG;
+    RBASIC(ary)->flags &= ~(RARRAY_EMBED_LEN_MASK | RARRAY_TRANSIENT_FLAG);
+}
+
 static VALUE
-to_ary(ary)
-    VALUE ary;
+ary_make_shared(VALUE ary)
 {
-    return rb_convert_type(ary, T_ARRAY, "Array", "to_ary");
+    assert(!ARY_EMBED_P(ary));
+    ary_verify(ary);
+
+    if (ARY_SHARED_P(ary)) {
+	return ARY_SHARED(ary);
+    }
+    else if (ARY_SHARED_ROOT_P(ary)) {
+	return ary;
+    }
+    else if (OBJ_FROZEN(ary)) {
+        rb_ary_transient_heap_evacuate(ary, TRUE);
+	ary_shrink_capa(ary);
+	FL_SET_SHARED_ROOT(ary);
+	ARY_SET_SHARED_NUM(ary, 1);
+	return ary;
+    }
+    else {
+	long capa = ARY_CAPA(ary), len = RARRAY_LEN(ary);
+        const VALUE *ptr;
+	NEWOBJ_OF(shared, struct RArray, 0, T_ARRAY | (RGENGC_WB_PROTECTED_ARRAY ? FL_WB_PROTECTED : 0));
+
+        rb_ary_transient_heap_evacuate(ary, TRUE);
+        ptr = ARY_HEAP_PTR(ary);
+
+        FL_UNSET_EMBED(shared);
+	ARY_SET_LEN((VALUE)shared, capa);
+        ARY_SET_PTR((VALUE)shared, ptr);
+        ary_mem_clear((VALUE)shared, len, capa - len);
+	FL_SET_SHARED_ROOT(shared);
+	ARY_SET_SHARED_NUM((VALUE)shared, 1);
+	FL_SET_SHARED(ary);
+	ARY_SET_SHARED(ary, (VALUE)shared);
+	OBJ_FREEZE(shared);
+
+        ary_verify((VALUE)shared);
+        ary_verify(ary);
+        return (VALUE)shared;
+    }
+}
+
+static VALUE
+ary_make_substitution(VALUE ary)
+{
+    long len = RARRAY_LEN(ary);
+
+    if (len <= RARRAY_EMBED_LEN_MAX) {
+	VALUE subst = rb_ary_new2(len);
+        ary_memcpy(subst, 0, len, RARRAY_CONST_PTR_TRANSIENT(ary));
+        ARY_SET_EMBED_LEN(subst, len);
+        return subst;
+    }
+    else {
+        return rb_ary_increment_share(ary_make_shared(ary));
+    }
 }
 
 VALUE
-rb_check_array_type(ary)
-    VALUE ary;
+rb_assoc_new(VALUE car, VALUE cdr)
+{
+    return rb_ary_new3(2, car, cdr);
+}
+
+VALUE
+rb_to_array_type(VALUE ary)
+{
+    return rb_convert_type_with_id(ary, T_ARRAY, "Array", idTo_ary);
+}
+#define to_ary rb_to_array_type
+
+VALUE
+rb_check_array_type(VALUE ary)
+{
+    return rb_check_convert_type_with_id(ary, T_ARRAY, "Array", idTo_ary);
+}
+
+MJIT_FUNC_EXPORTED VALUE
+rb_check_to_array(VALUE ary)
 {
-    return rb_check_convert_type(ary, T_ARRAY, "Array", "to_ary");
+    return rb_check_convert_type_with_id(ary, T_ARRAY, "Array", idTo_a);
 }
 
-static VALUE rb_ary_replace _((VALUE, VALUE));
+/*
+ *  call-seq:
+ *     Array.try_convert(obj) -> array or nil
+ *
+ *  Tries to convert +obj+ into an array, using +to_ary+ method.  Returns the
+ *  converted array or +nil+ if +obj+ cannot be converted for any reason.
+ *  This method can be used to check if an argument is an array.
+ *
+ *     Array.try_convert([1])   #=> [1]
+ *     Array.try_convert("1")   #=> nil
+ *
+ *     if tmp = Array.try_convert(arg)
+ *       # the argument is an array
+ *     elsif tmp = String.try_convert(arg)
+ *       # the argument is a string
+ *     end
+ *
+ */
+
+static VALUE
+rb_ary_s_try_convert(VALUE dummy, VALUE ary)
+{
+    return rb_check_array_type(ary);
+}
 
 /*
  *  call-seq:
- *     Array.new(size=0, obj=nil)
+ *     Array.new(size=0, default=nil)
  *     Array.new(array)
  *     Array.new(size) {|index| block }
  *
- *  Returns a new array. In the first form, the new array is
- *  empty. In the second it is created with _size_ copies of _obj_
- *  (that is, _size_ references to the same
- *  _obj_). The third form creates a copy of the array
- *  passed as a parameter (the array is generated by calling
- *  to_ary  on the parameter). In the last form, an array
- *  of the given size is created. Each element in this array is
- *  calculated by passing the element's index to the given block and
- *  storing the return value.
- *
- *     Array.new
- *     Array.new(2)
- *     Array.new(5, "A")
- * 
- *     # only one copy of the object is created
+ *  Returns a new array.
+ *
+ *  In the first form, if no arguments are sent, the new array will be empty.
+ *  When a +size+ and an optional +default+ are sent, an array is created with
+ *  +size+ copies of +default+.  Take notice that all elements will reference the
+ *  same object +default+.
+ *
+ *  The second form creates a copy of the array passed as a parameter (the
+ *  array is generated by calling to_ary on the parameter).
+ *
+ *    first_array = ["Matz", "Guido"]
+ *
+ *    second_array = Array.new(first_array) #=> ["Matz", "Guido"]
+ *
+ *    first_array.equal? second_array       #=> false
+ *
+ *  In the last form, an array of the given size is created.  Each element in
+ *  this array is created by passing the element's index to the given block
+ *  and storing the return value.
+ *
+ *    Array.new(3) {|index| index ** 2}
+ *    # => [0, 1, 4]
+ *
+ *  == Common gotchas
+ *
+ *  When sending the second parameter, the same object will be used as the
+ *  value for all the array elements:
+ *
  *     a = Array.new(2, Hash.new)
+ *     # => [{}, {}]
+ *
  *     a[0]['cat'] = 'feline'
- *     a
+ *     a # => [{"cat"=>"feline"}, {"cat"=>"feline"}]
+ *
  *     a[1]['cat'] = 'Felix'
- *     a
- * 
- *     # here multiple copies are created
- *     a = Array.new(2) { Hash.new }
+ *     a # => [{"cat"=>"Felix"}, {"cat"=>"Felix"}]
+ *
+ *  Since all the Array elements store the same hash, changes to one of them
+ *  will affect them all.
+ *
+ *  If multiple copies are what you want, you should use the block
+ *  version which uses the result of that block each time an element
+ *  of the array needs to be initialized:
+ *
+ *     a = Array.new(2) {Hash.new}
  *     a[0]['cat'] = 'feline'
- *     a
- * 
- *     squares = Array.new(5) {|i| i*i}
- *     squares
- * 
- *     copy = Array.new(squares)
+ *     a # => [{"cat"=>"feline"}, {}]
+ *
  */
 
 static VALUE
-rb_ary_initialize(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_initialize(int argc, VALUE *argv, VALUE ary)
 {
     long len;
     VALUE size, val;
 
     rb_ary_modify(ary);
-    if (rb_scan_args(argc, argv, "02", &size, &val) == 0) {
-	RARRAY(ary)->len = 0;
+    if (argc == 0) {
+        if (ARY_OWNS_HEAP_P(ary) && ARY_HEAP_PTR(ary) != NULL) {
+            ary_heap_free(ary);
+	}
+        rb_ary_unshare_safe(ary);
+        FL_SET_EMBED(ary);
+	ARY_SET_EMBED_LEN(ary, 0);
 	if (rb_block_given_p()) {
 	    rb_warning("given block not used");
 	}
 	return ary;
     }
-
+    rb_scan_args(argc, argv, "02", &size, &val);
     if (argc == 1 && !FIXNUM_P(size)) {
 	val = rb_check_array_type(size);
 	if (!NIL_P(val)) {
@@ -290,16 +984,16 @@ rb_ary_initialize(argc, argv, ary)
     }
 
     len = NUM2LONG(size);
+    /* NUM2LONG() may call size.to_int, ary can be frozen, modified, etc */
     if (len < 0) {
 	rb_raise(rb_eArgError, "negative array size");
     }
-    if (len > 0 && len * (long)sizeof(VALUE) <= len) {
+    if (len > ARY_MAX_SIZE) {
 	rb_raise(rb_eArgError, "array size too big");
     }
-    if (len > RARRAY(ary)->aux.capa) {
-	REALLOC_N(RARRAY(ary)->ptr, VALUE, len);
-	RARRAY(ary)->aux.capa = len;
-    }
+    /* recheck after argument conversion */
+    rb_ary_modify(ary);
+    ary_resize_capa(ary, len);
     if (rb_block_given_p()) {
 	long i;
 
@@ -308,331 +1002,511 @@ rb_ary_initialize(argc, argv, ary)
 	}
 	for (i=0; i<len; i++) {
 	    rb_ary_store(ary, i, rb_yield(LONG2NUM(i)));
-	    RARRAY(ary)->len = i + 1;
+	    ARY_SET_LEN(ary, i + 1);
 	}
     }
     else {
-	memfill(RARRAY(ary)->ptr, len, val);
-	RARRAY(ary)->len = len;
+	ary_memfill(ary, 0, len, val);
+	ARY_SET_LEN(ary, len);
     }
-
     return ary;
 }
 
-
-/* 
-* Returns a new array populated with the given objects. 
-*
-*   Array.[]( 1, 'a', /^A/ )
-*   Array[ 1, 'a', /^A/ ]
-*   [ 1, 'a', /^A/ ]
-*/
+/*
+ * Returns a new array populated with the given objects.
+ *
+ *   Array.[]( 1, 'a', /^A/)  # => [1, "a", /^A/]
+ *   Array[ 1, 'a', /^A/ ]    # => [1, "a", /^A/]
+ *   [ 1, 'a', /^A/ ]         # => [1, "a", /^A/]
+ */
 
 static VALUE
-rb_ary_s_create(argc, argv, klass)
-    int argc;
-    VALUE *argv;
-    VALUE klass;
+rb_ary_s_create(int argc, VALUE *argv, VALUE klass)
 {
-    VALUE ary = ary_alloc(klass);
-
-    if (argc > 0) {
-	RARRAY(ary)->ptr = ALLOC_N(VALUE, argc);
-	MEMCPY(RARRAY(ary)->ptr, argv, VALUE, argc);
+    VALUE ary = ary_new(klass, argc);
+    if (argc > 0 && argv) {
+        ary_memcpy(ary, 0, argc, argv);
+        ARY_SET_LEN(ary, argc);
     }
-    RARRAY(ary)->len = RARRAY(ary)->aux.capa = argc;
 
     return ary;
 }
 
 void
-rb_ary_store(ary, idx, val)
-    VALUE ary;
-    long idx;
-    VALUE val;
+rb_ary_store(VALUE ary, long idx, VALUE val)
 {
+    long len = RARRAY_LEN(ary);
+
     if (idx < 0) {
-	idx += RARRAY(ary)->len;
+	idx += len;
 	if (idx < 0) {
-	    rb_raise(rb_eIndexError, "index %ld out of array",
-		    idx - RARRAY(ary)->len);
+	    rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld",
+		     idx - len, -len);
 	}
     }
+    else if (idx >= ARY_MAX_SIZE) {
+	rb_raise(rb_eIndexError, "index %ld too big", idx);
+    }
 
     rb_ary_modify(ary);
-    if (idx >= RARRAY(ary)->aux.capa) {
-	long new_capa = RARRAY(ary)->aux.capa / 2;
+    if (idx >= ARY_CAPA(ary)) {
+	ary_double_capa(ary, idx);
+    }
+    if (idx > len) {
+	ary_mem_clear(ary, len, idx - len + 1);
+    }
 
-	if (new_capa < ARY_DEFAULT_SIZE) {
-	    new_capa = ARY_DEFAULT_SIZE;
-	}
-	new_capa += idx;
-	if (new_capa * (long)sizeof(VALUE) <= new_capa) {
-	    rb_raise(rb_eArgError, "index too big");
-	}
-	REALLOC_N(RARRAY(ary)->ptr, VALUE, new_capa);
-	RARRAY(ary)->aux.capa = new_capa;
+    if (idx >= len) {
+	ARY_SET_LEN(ary, idx + 1);
     }
-    if (idx > RARRAY(ary)->len) {
-	rb_mem_clear(RARRAY(ary)->ptr + RARRAY(ary)->len,
-		     idx-RARRAY(ary)->len + 1);
+    ARY_SET(ary, idx, val);
+}
+
+static VALUE
+ary_make_partial(VALUE ary, VALUE klass, long offset, long len)
+{
+    assert(offset >= 0);
+    assert(len >= 0);
+    assert(offset+len <= RARRAY_LEN(ary));
+
+    if (len <= RARRAY_EMBED_LEN_MAX) {
+        VALUE result = ary_alloc(klass);
+        ary_memcpy(result, 0, len, RARRAY_CONST_PTR_TRANSIENT(ary) + offset);
+        ARY_SET_EMBED_LEN(result, len);
+        return result;
+    }
+    else {
+        VALUE shared, result = ary_alloc(klass);
+        FL_UNSET_EMBED(result);
+
+        shared = ary_make_shared(ary);
+        ARY_SET_PTR(result, RARRAY_CONST_PTR_TRANSIENT(ary));
+        ARY_SET_LEN(result, RARRAY_LEN(ary));
+        rb_ary_set_shared(result, shared);
+
+        ARY_INCREASE_PTR(result, offset);
+        ARY_SET_LEN(result, len);
+
+        ary_verify(shared);
+        ary_verify(result);
+        return result;
     }
+}
+
+static VALUE
+ary_make_shared_copy(VALUE ary)
+{
+    return ary_make_partial(ary, rb_obj_class(ary), 0, RARRAY_LEN(ary));
+}
+
+enum ary_take_pos_flags
+{
+    ARY_TAKE_FIRST = 0,
+    ARY_TAKE_LAST = 1
+};
 
-    if (idx >= RARRAY(ary)->len) {
-	RARRAY(ary)->len = idx + 1;
+static VALUE
+ary_take_first_or_last(int argc, const VALUE *argv, VALUE ary, enum ary_take_pos_flags last)
+{
+    long n;
+    long len;
+    long offset = 0;
+
+    argc = rb_check_arity(argc, 0, 1);
+    /* the case optional argument is ommited should be handled in
+     * callers of this function.  if another arity case is added,
+     * this arity check needs to rewrite. */
+    RUBY_ASSERT_WHEN(TRUE, argc == 1);
+
+    n = NUM2LONG(argv[0]);
+    len = RARRAY_LEN(ary);
+    if (n > len) {
+	n = len;
     }
-    RARRAY(ary)->ptr[idx] = val;
+    else if (n < 0) {
+	rb_raise(rb_eArgError, "negative array size");
+    }
+    if (last) {
+	offset = len - n;
+    }
+    return ary_make_partial(ary, rb_cArray, offset, n);
 }
 
 /*
  *  call-seq:
- *     array << obj            -> array
- *  
+ *     ary << obj            -> ary
+ *
  *  Append---Pushes the given object on to the end of this array. This
  *  expression returns the array itself, so several appends
  *  may be chained together.
  *
- *     [ 1, 2 ] << "c" << "d" << [ 3, 4 ]
+ *     a = [ 1, 2 ]
+ *     a << "c" << "d" << [ 3, 4 ]
+ *             #=>  [ 1, 2, "c", "d", [ 3, 4 ] ]
+ *     a
  *             #=>  [ 1, 2, "c", "d", [ 3, 4 ] ]
  *
  */
 
 VALUE
-rb_ary_push(ary, item)
-    VALUE ary;
-    VALUE item;
+rb_ary_push(VALUE ary, VALUE item)
 {
-    rb_ary_store(ary, RARRAY(ary)->len, item);
+    long idx = RARRAY_LEN((ary_verify(ary), ary));
+    VALUE target_ary = ary_ensure_room_for_push(ary, 1);
+    RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+	RB_OBJ_WRITE(target_ary, &ptr[idx], item);
+    });
+    ARY_SET_LEN(ary, idx + 1);
+    ary_verify(ary);
     return ary;
 }
 
-/* 
+VALUE
+rb_ary_cat(VALUE ary, const VALUE *argv, long len)
+{
+    long oldlen = RARRAY_LEN(ary);
+    VALUE target_ary = ary_ensure_room_for_push(ary, len);
+    ary_memcpy0(ary, oldlen, len, argv, target_ary);
+    ARY_SET_LEN(ary, oldlen + len);
+    return ary;
+}
+
+/*
  *  call-seq:
- *     array.push(obj, ... )   -> array
- *  
- *  Append---Pushes the given object(s) on to the end of this array. This
+ *     ary.push(obj, ...)    -> ary
+ *     ary.append(obj, ...)  -> ary
+ *
+ *  Append --- Pushes the given object(s) on to the end of this array. This
  *  expression returns the array itself, so several appends
- *  may be chained together.
+ *  may be chained together. See also Array#pop for the opposite
+ *  effect.
  *
  *     a = [ "a", "b", "c" ]
- *     a.push("d", "e", "f")  
+ *     a.push("d", "e", "f")
  *             #=> ["a", "b", "c", "d", "e", "f"]
+ *     [1, 2, 3].push(4).push(5)
+ *             #=> [1, 2, 3, 4, 5]
  */
 
 static VALUE
-rb_ary_push_m(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_push_m(int argc, VALUE *argv, VALUE ary)
 {
-    while (argc--) {
-	rb_ary_push(ary, *argv++);
-    }
-    return ary;
+    return rb_ary_cat(ary, argv, argc);
+}
+
+VALUE
+rb_ary_pop(VALUE ary)
+{
+    long n;
+    rb_ary_modify_check(ary);
+    n = RARRAY_LEN(ary);
+    if (n == 0) return Qnil;
+    if (ARY_OWNS_HEAP_P(ary) &&
+	n * 3 < ARY_CAPA(ary) &&
+	ARY_CAPA(ary) > ARY_DEFAULT_SIZE)
+    {
+	ary_resize_capa(ary, n * 2);
+    }
+    --n;
+    ARY_SET_LEN(ary, n);
+    ary_verify(ary);
+    return RARRAY_AREF(ary, n);
 }
 
 /*
  *  call-seq:
- *     array.pop  -> obj or nil
- *  
- *  Removes the last element from <i>self</i> and returns it, or
- *  <code>nil</code> if the array is empty.
- *     
- *     a = [ "a", "m", "z" ]
- *     a.pop   #=> "z"
- *     a       #=> ["a", "m"]
+ *     ary.pop    -> obj or nil
+ *     ary.pop(n) -> new_ary
+ *
+ *  Removes the last element from +self+ and returns it, or
+ *  +nil+ if the array is empty.
+ *
+ *  If a number +n+ is given, returns an array of the last +n+ elements
+ *  (or less) just like <code>array.slice!(-n, n)</code> does. See also
+ *  Array#push for the opposite effect.
+ *
+ *     a = [ "a", "b", "c", "d" ]
+ *     a.pop     #=> "d"
+ *     a.pop(2)  #=> ["b", "c"]
+ *     a         #=> ["a"]
  */
 
-VALUE
-rb_ary_pop(ary)
-    VALUE ary;
+static VALUE
+rb_ary_pop_m(int argc, VALUE *argv, VALUE ary)
 {
-    rb_ary_modify_check(ary);
-    if (RARRAY(ary)->len == 0) return Qnil;
-    if (!FL_TEST(ary, ELTS_SHARED) &&
-	    RARRAY(ary)->len * 2 < RARRAY(ary)->aux.capa &&
-	    RARRAY(ary)->aux.capa > ARY_DEFAULT_SIZE) {
-	RARRAY(ary)->aux.capa = RARRAY(ary)->len * 2;
-	REALLOC_N(RARRAY(ary)->ptr, VALUE, RARRAY(ary)->aux.capa);
+    VALUE result;
+
+    if (argc == 0) {
+	return rb_ary_pop(ary);
     }
-    return RARRAY(ary)->ptr[--RARRAY(ary)->len];
+
+    rb_ary_modify_check(ary);
+    result = ary_take_first_or_last(argc, argv, ary, ARY_TAKE_LAST);
+    ARY_INCREASE_LEN(ary, -RARRAY_LEN(result));
+    ary_verify(ary);
+    return result;
 }
 
-static VALUE
-ary_make_shared(ary)
-    VALUE ary;
+VALUE
+rb_ary_shift(VALUE ary)
 {
-    if (!FL_TEST(ary, ELTS_SHARED)) {
-	NEWOBJ(shared, struct RArray);
-	OBJSETUP(shared, rb_cArray, T_ARRAY);
+    VALUE top;
+    long len = RARRAY_LEN(ary);
 
-	shared->len = RARRAY(ary)->len;
-	shared->ptr = RARRAY(ary)->ptr;
-	shared->aux.capa = RARRAY(ary)->aux.capa;
-	RARRAY(ary)->aux.shared = (VALUE)shared;
-	FL_SET(ary, ELTS_SHARED);
-	OBJ_FREEZE(shared);
-	return (VALUE)shared;
+    rb_ary_modify_check(ary);
+    if (len == 0) return Qnil;
+    top = RARRAY_AREF(ary, 0);
+    if (!ARY_SHARED_P(ary)) {
+	if (len < ARY_DEFAULT_SIZE) {
+            RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+		MEMMOVE(ptr, ptr+1, VALUE, len-1);
+	    }); /* WB: no new reference */
+            ARY_INCREASE_LEN(ary, -1);
+            ary_verify(ary);
+	    return top;
+	}
+        assert(!ARY_EMBED_P(ary)); /* ARY_EMBED_LEN_MAX < ARY_DEFAULT_SIZE */
+
+	ARY_SET(ary, 0, Qnil);
+	ary_make_shared(ary);
     }
-    else {
-	return RARRAY(ary)->aux.shared;
+    else if (ARY_SHARED_OCCUPIED(ARY_SHARED(ary))) {
+        RARRAY_PTR_USE_TRANSIENT(ary, ptr, ptr[0] = Qnil);
     }
+    ARY_INCREASE_PTR(ary, 1);		/* shift ptr */
+    ARY_INCREASE_LEN(ary, -1);
+
+    ary_verify(ary);
+
+    return top;
 }
 
 /*
  *  call-seq:
- *     array.shift   ->   obj or nil
- *  
- *  Returns the first element of <i>self</i> and removes it (shifting all
- *  other elements down by one). Returns <code>nil</code> if the array
+ *     ary.shift    -> obj or nil
+ *     ary.shift(n) -> new_ary
+ *
+ *  Removes the first element of +self+ and returns it (shifting all
+ *  other elements down by one). Returns +nil+ if the array
  *  is empty.
- *     
+ *
+ *  If a number +n+ is given, returns an array of the first +n+ elements
+ *  (or less) just like <code>array.slice!(0, n)</code> does. With +ary+
+ *  containing only the remainder elements, not including what was shifted to
+ *  +new_ary+. See also Array#unshift for the opposite effect.
+ *
  *     args = [ "-m", "-q", "filename" ]
- *     args.shift   #=> "-m"
- *     args         #=> ["-q", "filename"]
+ *     args.shift     #=> "-m"
+ *     args           #=> ["-q", "filename"]
+ *
+ *     args = [ "-m", "-q", "filename" ]
+ *     args.shift(2)  #=> ["-m", "-q"]
+ *     args           #=> ["filename"]
  */
 
-VALUE
-rb_ary_shift(ary)
-    VALUE ary;
+static VALUE
+rb_ary_shift_m(int argc, VALUE *argv, VALUE ary)
 {
-    VALUE top;
+    VALUE result;
+    long n;
+
+    if (argc == 0) {
+	return rb_ary_shift(ary);
+    }
 
     rb_ary_modify_check(ary);
-    if (RARRAY(ary)->len == 0) return Qnil;
-    top = RARRAY(ary)->ptr[0];
-    ary_make_shared(ary);
-    RARRAY(ary)->ptr++;		/* shift ptr */
-    RARRAY(ary)->len--;
+    result = ary_take_first_or_last(argc, argv, ary, ARY_TAKE_FIRST);
+    n = RARRAY_LEN(result);
+    rb_ary_behead(ary,n);
 
-    return top;
+    return result;
 }
 
-VALUE
-rb_ary_unshift(ary, item)
-    VALUE ary, item;
+MJIT_FUNC_EXPORTED VALUE
+rb_ary_behead(VALUE ary, long n)
 {
-    rb_ary_modify(ary);
-    if (RARRAY(ary)->len == RARRAY(ary)->aux.capa) {
-	long capa_inc = RARRAY(ary)->aux.capa / 2;
-	if (capa_inc < ARY_DEFAULT_SIZE) {
-	    capa_inc = ARY_DEFAULT_SIZE;
+    if(n<=0) return ary;
+
+    rb_ary_modify_check(ary);
+    if (ARY_SHARED_P(ary)) {
+	if (ARY_SHARED_OCCUPIED(ARY_SHARED(ary))) {
+	  setup_occupied_shared:
+	    ary_mem_clear(ary, 0, n);
+	}
+        ARY_INCREASE_PTR(ary, n);
+    }
+    else {
+	if (RARRAY_LEN(ary) < ARY_DEFAULT_SIZE) {
+            RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+                MEMMOVE(ptr, ptr+n, VALUE, RARRAY_LEN(ary)-n);
+	    }); /* WB: no new reference */
+	}
+	else {
+	    ary_make_shared(ary);
+	    goto setup_occupied_shared;
 	}
-	RARRAY(ary)->aux.capa += capa_inc;
-	REALLOC_N(RARRAY(ary)->ptr, VALUE, RARRAY(ary)->aux.capa);
     }
+    ARY_INCREASE_LEN(ary, -n);
 
-    /* sliding items */
-    MEMMOVE(RARRAY(ary)->ptr + 1, RARRAY(ary)->ptr, VALUE, RARRAY(ary)->len);
+    ary_verify(ary);
+    return ary;
+}
 
-    RARRAY(ary)->len++;
-    RARRAY(ary)->ptr[0] = item;
+static VALUE
+ary_ensure_room_for_unshift(VALUE ary, int argc)
+{
+    long len = RARRAY_LEN(ary);
+    long new_len = len + argc;
+    long capa;
+    const VALUE *head, *sharedp;
+
+    if (len > ARY_MAX_SIZE - argc) {
+	rb_raise(rb_eIndexError, "index %ld too big", new_len);
+    }
+
+    if (ARY_SHARED_P(ary)) {
+	VALUE shared = ARY_SHARED(ary);
+	capa = RARRAY_LEN(shared);
+	if (ARY_SHARED_OCCUPIED(shared) && capa > new_len) {
+            head = RARRAY_CONST_PTR_TRANSIENT(ary);
+            sharedp = RARRAY_CONST_PTR_TRANSIENT(shared);
+	    goto makeroom_if_need;
+	}
+    }
 
-    return ary;
+    rb_ary_modify(ary);
+    capa = ARY_CAPA(ary);
+    if (capa - (capa >> 6) <= new_len) {
+	ary_double_capa(ary, new_len);
+    }
+
+    /* use shared array for big "queues" */
+    if (new_len > ARY_DEFAULT_SIZE * 4) {
+        ary_verify(ary);
+
+        /* make a room for unshifted items */
+	capa = ARY_CAPA(ary);
+	ary_make_shared(ary);
+
+        head = sharedp = RARRAY_CONST_PTR_TRANSIENT(ary);
+	goto makeroom;
+      makeroom_if_need:
+	if (head - sharedp < argc) {
+	    long room;
+	  makeroom:
+	    room = capa - new_len;
+	    room -= room >> 4;
+	    MEMMOVE((VALUE *)sharedp + argc + room, head, VALUE, len);
+	    head = sharedp + argc + room;
+	}
+	ARY_SET_PTR(ary, head - argc);
+	assert(ARY_SHARED_OCCUPIED(ARY_SHARED(ary)));
+
+        ary_verify(ary);
+	return ARY_SHARED(ary);
+    }
+    else {
+	/* sliding items */
+        RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+	    MEMMOVE(ptr + argc, ptr, VALUE, len);
+	});
+
+        ary_verify(ary);
+	return ary;
+    }
 }
 
 /*
  *  call-seq:
- *     array.unshift(obj, ...)  -> array
- *  
- *  Prepends objects to the front of <i>array</i>.
- *  other elements up one.
- *     
+ *     ary.unshift(obj, ...)  -> ary
+ *     ary.prepend(obj, ...)  -> ary
+ *
+ *  Prepends objects to the front of +self+, moving other elements upwards.
+ *  See also Array#shift for the opposite effect.
+ *
  *     a = [ "b", "c", "d" ]
  *     a.unshift("a")   #=> ["a", "b", "c", "d"]
  *     a.unshift(1, 2)  #=> [ 1, 2, "a", "b", "c", "d"]
  */
 
 static VALUE
-rb_ary_unshift_m(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_unshift_m(int argc, VALUE *argv, VALUE ary)
 {
-    long len = RARRAY(ary)->len;
+    long len = RARRAY_LEN(ary);
+    VALUE target_ary;
 
-    if (argc == 0) return ary;
-
-    /* make rooms by setting the last item */
-    rb_ary_store(ary, len + argc - 1, Qnil);
+    if (argc == 0) {
+	rb_ary_modify_check(ary);
+	return ary;
+    }
 
-    /* sliding items */
-    MEMMOVE(RARRAY(ary)->ptr + argc, RARRAY(ary)->ptr, VALUE, len);
-    MEMCPY(RARRAY(ary)->ptr, argv, VALUE, argc);
-    
+    target_ary = ary_ensure_room_for_unshift(ary, argc);
+    ary_memcpy0(ary, 0, argc, argv, target_ary);
+    ARY_SET_LEN(ary, len + argc);
     return ary;
 }
 
+VALUE
+rb_ary_unshift(VALUE ary, VALUE item)
+{
+    return rb_ary_unshift_m(1,&item,ary);
+}
+
 /* faster version - use this if you don't need to treat negative offset */
 static inline VALUE
-rb_ary_elt(ary, offset)
-    VALUE ary;
-    long offset;
+rb_ary_elt(VALUE ary, long offset)
 {
-    if (RARRAY(ary)->len == 0) return Qnil;
-    if (offset < 0 || RARRAY(ary)->len <= offset) {
+    long len = RARRAY_LEN(ary);
+    if (len == 0) return Qnil;
+    if (offset < 0 || len <= offset) {
 	return Qnil;
     }
-    return RARRAY(ary)->ptr[offset];
+    return RARRAY_AREF(ary, offset);
 }
 
 VALUE
-rb_ary_entry(ary, offset)
-    VALUE ary;
-    long offset;
+rb_ary_entry(VALUE ary, long offset)
 {
-    if (offset < 0) {
-	offset += RARRAY(ary)->len;
-    }
-    return rb_ary_elt(ary, offset);
+    return rb_ary_entry_internal(ary, offset);
 }
 
-static VALUE
-rb_ary_subseq(ary, beg, len)
-    VALUE ary;
-    long beg, len;
+VALUE
+rb_ary_subseq(VALUE ary, long beg, long len)
 {
-    VALUE klass, ary2, shared;
-    VALUE *ptr;
+    VALUE klass;
+    long alen = RARRAY_LEN(ary);
 
-    if (beg > RARRAY(ary)->len) return Qnil;
+    if (beg > alen) return Qnil;
     if (beg < 0 || len < 0) return Qnil;
 
-    if (RARRAY(ary)->len < len || RARRAY(ary)->len < beg + len) {
-	len = RARRAY(ary)->len - beg;
-	if (len < 0)
-	    len = 0;
+    if (alen < len || alen < beg + len) {
+	len = alen - beg;
     }
     klass = rb_obj_class(ary);
     if (len == 0) return ary_new(klass, 0);
 
-    shared = ary_make_shared(ary);
-    ptr = RARRAY(ary)->ptr;
-    ary2 = ary_alloc(klass);
-    RARRAY(ary2)->ptr = ptr + beg;
-    RARRAY(ary2)->len = len;
-    RARRAY(ary2)->aux.shared = shared;
-    FL_SET(ary2, ELTS_SHARED);
-
-    return ary2;
+    return ary_make_partial(ary, klass, beg, len);
 }
 
-/* 
+/*
  *  call-seq:
- *     array[index]                -> obj      or nil
- *     array[start, length]        -> an_array or nil
- *     array[range]                -> an_array or nil
- *     array.slice(index)          -> obj      or nil
- *     array.slice(start, length)  -> an_array or nil
- *     array.slice(range)          -> an_array or nil
+ *     ary[index]                -> obj     or nil
+ *     ary[start, length]        -> new_ary or nil
+ *     ary[range]                -> new_ary or nil
+ *     ary.slice(index)          -> obj     or nil
+ *     ary.slice(start, length)  -> new_ary or nil
+ *     ary.slice(range)          -> new_ary or nil
+ *
+ *  Element Reference --- Returns the element at +index+, or returns a
+ *  subarray starting at the +start+ index and continuing for +length+
+ *  elements, or returns a subarray specified by +range+ of indices.
  *
- *  Element Reference---Returns the element at _index_,
- *  or returns a subarray starting at _start_ and
- *  continuing for _length_ elements, or returns a subarray
- *  specified by _range_.
- *  Negative indices count backward from the end of the
- *  array (-1 is the last element). Returns nil if the index
- *  (or starting index) are out of range.
+ *  Negative indices count backward from the end of the array (-1 is the last
+ *  element).  For +start+ and +range+ cases the starting index is just before
+ *  an element.  Additionally, an empty array is returned when the starting
+ *  index for an element range is at the end of the array.
+ *
+ *  Returns +nil+ if the index (or starting index) are out of range.
  *
  *     a = [ "a", "b", "c", "d", "e" ]
  *     a[2] +  a[0] + a[1]    #=> "cab"
@@ -644,44 +1518,44 @@ rb_ary_subseq(ary, beg, len)
  *     a[-3, 3]               #=> [ "c", "d", "e" ]
  *     # special cases
  *     a[5]                   #=> nil
+ *     a[6, 1]                #=> nil
  *     a[5, 1]                #=> []
  *     a[5..10]               #=> []
  *
  */
 
 VALUE
-rb_ary_aref(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_aref(int argc, const VALUE *argv, VALUE ary)
 {
-    VALUE arg;
-    long beg, len;
-
+    rb_check_arity(argc, 1, 2);
     if (argc == 2) {
-	if (SYMBOL_P(argv[0])) {
-	    rb_raise(rb_eTypeError, "Symbol as array index");
-	}
-	beg = NUM2LONG(argv[0]);
-	len = NUM2LONG(argv[1]);
-	if (beg < 0) {
-	    beg += RARRAY(ary)->len;
-	}
-	return rb_ary_subseq(ary, beg, len);
+	return rb_ary_aref2(ary, argv[0], argv[1]);
     }
-    if (argc != 1) {
-	rb_scan_args(argc, argv, "11", 0, 0);
+    return rb_ary_aref1(ary, argv[0]);
+}
+
+VALUE
+rb_ary_aref2(VALUE ary, VALUE b, VALUE e)
+{
+    long beg = NUM2LONG(b);
+    long len = NUM2LONG(e);
+    if (beg < 0) {
+	beg += RARRAY_LEN(ary);
     }
-    arg = argv[0];
+    return rb_ary_subseq(ary, beg, len);
+}
+
+MJIT_FUNC_EXPORTED VALUE
+rb_ary_aref1(VALUE ary, VALUE arg)
+{
+    long beg, len;
+
     /* special case - speeding up */
     if (FIXNUM_P(arg)) {
 	return rb_ary_entry(ary, FIX2LONG(arg));
     }
-    if (SYMBOL_P(arg)) {
-	rb_raise(rb_eTypeError, "Symbol as array index");
-    }
     /* check if idx is Range */
-    switch (rb_range_beg_len(arg, &beg, &len, RARRAY(ary)->len, 0)) {
+    switch (rb_range_beg_len(arg, &beg, &len, RARRAY_LEN(ary), 0)) {
       case Qfalse:
 	break;
       case Qnil:
@@ -692,129 +1566,106 @@ rb_ary_aref(argc, argv, ary)
     return rb_ary_entry(ary, NUM2LONG(arg));
 }
 
-/* 
+/*
  *  call-seq:
- *     array.at(index)   ->   obj  or nil
+ *     ary.at(index)   ->   obj  or nil
  *
- *  Returns the element at _index_. A
- *  negative index counts from the end of _self_.  Returns +nil+
- *  if the index is out of range. See also <code>Array#[]</code>.
- *  (<code>Array#at</code> is slightly faster than <code>Array#[]</code>,
- *  as it does not accept ranges and so on.)
+ *  Returns the element at +index+. A negative index counts from the end of
+ *  +self+. Returns +nil+ if the index is out of range. See also
+ *  Array#[].
  *
  *     a = [ "a", "b", "c", "d", "e" ]
  *     a.at(0)     #=> "a"
  *     a.at(-1)    #=> "e"
  */
 
-static VALUE
-rb_ary_at(ary, pos)
-    VALUE ary, pos;
+VALUE
+rb_ary_at(VALUE ary, VALUE pos)
 {
     return rb_ary_entry(ary, NUM2LONG(pos));
 }
 
 /*
  *  call-seq:
- *     array.first   ->   obj or nil
- *     array.first(n) -> an_array
+ *     ary.first     ->   obj or nil
+ *     ary.first(n)  ->   new_ary
  *
  *  Returns the first element, or the first +n+ elements, of the array.
- *  If the array is empty, the first form returns <code>nil</code>, and the
- *  second form returns an empty array.
+ *  If the array is empty, the first form returns +nil+, and the
+ *  second form returns an empty array. See also Array#last for
+ *  the opposite effect.
  *
  *     a = [ "q", "r", "s", "t" ]
- *     a.first    #=> "q"
- *     a.first(1) #=> ["q"]
- *     a.first(3) #=> ["q", "r", "s"]
+ *     a.first     #=> "q"
+ *     a.first(2)  #=> ["q", "r"]
  */
 
 static VALUE
-rb_ary_first(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_first(int argc, VALUE *argv, VALUE ary)
 {
     if (argc == 0) {
-	if (RARRAY(ary)->len == 0) return Qnil;
-	return RARRAY(ary)->ptr[0];
+	if (RARRAY_LEN(ary) == 0) return Qnil;
+	return RARRAY_AREF(ary, 0);
     }
     else {
-	VALUE nv, result;
-	long n, i;
-
-	rb_scan_args(argc, argv, "01", &nv);
-	n = NUM2LONG(nv);
-	if (n > RARRAY(ary)->len) n = RARRAY(ary)->len;
-	result = rb_ary_new2(n);
-	for (i=0; i<n; i++) {
-	    rb_ary_push(result, RARRAY(ary)->ptr[i]);
-	}
-	return result;
+	return ary_take_first_or_last(argc, argv, ary, ARY_TAKE_FIRST);
     }
 }
 
 /*
  *  call-seq:
- *     array.last     ->  obj or nil
- *     array.last(n)  ->  an_array
- *  
- *  Returns the last element(s) of <i>self</i>. If the array is empty,
- *  the first form returns <code>nil</code>.
- *     
- *     [ "w", "x", "y", "z" ].last   #=> "z"
+ *     ary.last     ->  obj or nil
+ *     ary.last(n)  ->  new_ary
+ *
+ *  Returns the last element(s) of +self+. If the array is empty,
+ *  the first form returns +nil+.
+ *
+ *  See also Array#first for the opposite effect.
+ *
+ *     a = [ "w", "x", "y", "z" ]
+ *     a.last     #=> "z"
+ *     a.last(2)  #=> ["y", "z"]
  */
 
-static VALUE
-rb_ary_last(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+VALUE
+rb_ary_last(int argc, const VALUE *argv, VALUE ary)
 {
     if (argc == 0) {
-	if (RARRAY(ary)->len == 0) return Qnil;
-	return RARRAY(ary)->ptr[RARRAY(ary)->len-1];
+	long len = RARRAY_LEN(ary);
+	if (len == 0) return Qnil;
+	return RARRAY_AREF(ary, len-1);
     }
     else {
-	VALUE nv, result;
-	long n, i;
-
-	rb_scan_args(argc, argv, "01", &nv);
-	n = NUM2LONG(nv);
-	if (n > RARRAY(ary)->len) n = RARRAY(ary)->len;
-	result = rb_ary_new2(n);
-	for (i=RARRAY(ary)->len-n; n--; i++) {
-	    rb_ary_push(result, RARRAY(ary)->ptr[i]);
-	}
-	return result;
+	return ary_take_first_or_last(argc, argv, ary, ARY_TAKE_LAST);
     }
 }
 
 /*
  *  call-seq:
- *     array.fetch(index)                    -> obj
- *     array.fetch(index, default )          -> obj
- *     array.fetch(index) {|index| block }   -> obj
- *  
- *  Tries to return the element at position <i>index</i>. If the index
- *  lies outside the array, the first form throws an
- *  <code>IndexError</code> exception, the second form returns
- *  <i>default</i>, and the third form returns the value of invoking
- *  the block, passing in the index. Negative values of <i>index</i>
- *  count from the end of the array.
- *     
+ *     ary.fetch(index)                    -> obj
+ *     ary.fetch(index, default)           -> obj
+ *     ary.fetch(index) {|index| block}    -> obj
+ *
+ *  Tries to return the element at position +index+, but throws an IndexError
+ *  exception if the referenced +index+ lies outside of the array bounds.  This
+ *  error can be prevented by supplying a second argument, which will act as a
+ *  +default+ value.
+ *
+ *  Alternatively, if a block is given it will only be executed when an
+ *  invalid +index+ is referenced.
+ *
+ *  Negative values of +index+ count from the end of the array.
+ *
  *     a = [ 11, 22, 33, 44 ]
  *     a.fetch(1)               #=> 22
  *     a.fetch(-1)              #=> 44
  *     a.fetch(4, 'cat')        #=> "cat"
- *     a.fetch(4) { |i| i*i }   #=> 16
+ *     a.fetch(100) {|i| puts "#{i} is out of bounds"}
+ *                              #=> "100 is out of bounds"
  */
 
 static VALUE
-rb_ary_fetch(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_fetch(int argc, VALUE *argv, VALUE ary)
 {
     VALUE pos, ifnone;
     long block_given;
@@ -828,198 +1679,284 @@ rb_ary_fetch(argc, argv, ary)
     idx = NUM2LONG(pos);
 
     if (idx < 0) {
-	idx +=  RARRAY(ary)->len;
+	idx +=  RARRAY_LEN(ary);
     }
-    if (idx < 0 || RARRAY(ary)->len <= idx) {
+    if (idx < 0 || RARRAY_LEN(ary) <= idx) {
 	if (block_given) return rb_yield(pos);
 	if (argc == 1) {
-	    rb_raise(rb_eIndexError, "index %ld out of array", idx);
+	    rb_raise(rb_eIndexError, "index %ld outside of array bounds: %ld...%ld",
+			idx - (idx < 0 ? RARRAY_LEN(ary) : 0), -RARRAY_LEN(ary), RARRAY_LEN(ary));
 	}
 	return ifnone;
     }
-    return RARRAY(ary)->ptr[idx];
+    return RARRAY_AREF(ary, idx);
 }
 
 /*
  *  call-seq:
- *     array.index(obj)   ->  int or nil
- *  
- *  Returns the index of the first object in <i>self</i> such that is 
- *  <code>==</code> to <i>obj</i>. Returns <code>nil</code> if
- *  no match is found.
- *     
+ *     ary.find_index(obj)             ->  int or nil
+ *     ary.find_index {|item| block}  ->  int or nil
+ *     ary.find_index                  ->  Enumerator
+ *     ary.index(obj)             ->  int or nil
+ *     ary.index {|item| block}   ->  int or nil
+ *     ary.index                  ->  Enumerator
+ *
+ *  Returns the _index_ of the first object in +ary+ such that the object is
+ *  <code>==</code> to +obj+.
+ *
+ *  If a block is given instead of an argument, returns the _index_ of the
+ *  first object for which the block returns +true+.  Returns +nil+ if no
+ *  match is found.
+ *
+ *  See also Array#rindex.
+ *
+ *  An Enumerator is returned if neither a block nor argument is given.
+ *
  *     a = [ "a", "b", "c" ]
- *     a.index("b")   #=> 1
- *     a.index("z")   #=> nil
+ *     a.index("b")              #=> 1
+ *     a.index("z")              #=> nil
+ *     a.index {|x| x == "b"}    #=> 1
  */
 
 static VALUE
-rb_ary_index(ary, val)
-    VALUE ary;
-    VALUE val;
+rb_ary_index(int argc, VALUE *argv, VALUE ary)
 {
+    VALUE val;
     long i;
 
-    for (i=0; i<RARRAY(ary)->len; i++) {
-	if (rb_equal(RARRAY(ary)->ptr[i], val))
+    if (argc == 0) {
+	RETURN_ENUMERATOR(ary, 0, 0);
+	for (i=0; i<RARRAY_LEN(ary); i++) {
+	    if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) {
+		return LONG2NUM(i);
+	    }
+	}
+	return Qnil;
+    }
+    rb_check_arity(argc, 0, 1);
+    val = argv[0];
+    if (rb_block_given_p())
+	rb_warn("given block not used");
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	VALUE e = RARRAY_AREF(ary, i);
+	if (rb_equal(e, val)) {
 	    return LONG2NUM(i);
+	}
     }
     return Qnil;
 }
 
 /*
  *  call-seq:
- *     array.rindex(obj)    ->  int or nil
- *  
- *  Returns the index of the last object in <i>array</i> 
- *  <code>==</code> to <i>obj</i>. Returns <code>nil</code> if
- *  no match is found.
- *     
+ *     ary.rindex(obj)             ->  int or nil
+ *     ary.rindex {|item| block}   ->  int or nil
+ *     ary.rindex                  ->  Enumerator
+ *
+ *  Returns the _index_ of the last object in +self+ <code>==</code> to +obj+.
+ *
+ *  If a block is given instead of an argument, returns the _index_ of the
+ *  first object for which the block returns +true+, starting from the last
+ *  object.
+ *
+ *  Returns +nil+ if no match is found.
+ *
+ *  See also Array#index.
+ *
+ *  If neither block nor argument is given, an Enumerator is returned instead.
+ *
  *     a = [ "a", "b", "b", "b", "c" ]
- *     a.rindex("b")   #=> 3
- *     a.rindex("z")   #=> nil
+ *     a.rindex("b")             #=> 3
+ *     a.rindex("z")             #=> nil
+ *     a.rindex {|x| x == "b"}   #=> 3
  */
 
 static VALUE
-rb_ary_rindex(ary, val)
-    VALUE ary;
-    VALUE val;
+rb_ary_rindex(int argc, VALUE *argv, VALUE ary)
 {
-    long i = RARRAY(ary)->len;
+    VALUE val;
+    long i = RARRAY_LEN(ary), len;
 
-    while (i--) {
-	if (i > RARRAY(ary)->len) {
-	    i = RARRAY(ary)->len;
-	    continue;
+    if (argc == 0) {
+	RETURN_ENUMERATOR(ary, 0, 0);
+	while (i--) {
+	    if (RTEST(rb_yield(RARRAY_AREF(ary, i))))
+		return LONG2NUM(i);
+	    if (i > (len = RARRAY_LEN(ary))) {
+		i = len;
+	    }
 	}
-	if (rb_equal(RARRAY(ary)->ptr[i], val))
+	return Qnil;
+    }
+    rb_check_arity(argc, 0, 1);
+    val = argv[0];
+    if (rb_block_given_p())
+	rb_warn("given block not used");
+    while (i--) {
+	VALUE e = RARRAY_AREF(ary, i);
+	if (rb_equal(e, val)) {
 	    return LONG2NUM(i);
+	}
     }
     return Qnil;
 }
 
-/*
- *  call-seq:
- *     array.indexes( i1, i2, ... iN )   -> an_array
- *     array.indices( i1, i2, ... iN )   -> an_array
- *  
- *  Deprecated; use <code>Array#values_at</code>.
- */
-
-static VALUE
-rb_ary_indexes(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
-{
-    VALUE new_ary;
-    long i;
-
-    rb_warn("Array#%s is deprecated; use Array#values_at", rb_id2name(rb_frame_last_func()));
-    new_ary = rb_ary_new2(argc);
-    for (i=0; i<argc; i++) {
-	rb_ary_push(new_ary, rb_ary_aref(1, argv+i, ary));
-    }
-
-    return new_ary;
-}
-
 VALUE
-rb_ary_to_ary(obj)
-    VALUE obj;
+rb_ary_to_ary(VALUE obj)
 {
-    if (TYPE(obj) == T_ARRAY) {
-	return obj;
-    }
-    if (rb_respond_to(obj, rb_intern("to_ary"))) {
-	return rb_convert_type(obj, T_ARRAY, "Array", "to_ary");
-    }
+    VALUE tmp = rb_check_array_type(obj);
+
+    if (!NIL_P(tmp)) return tmp;
     return rb_ary_new3(1, obj);
 }
 
 static void
-rb_ary_splice(ary, beg, len, rpl)
-    VALUE ary;
-    long beg, len;
-    VALUE rpl;
+rb_ary_splice(VALUE ary, long beg, long len, const VALUE *rptr, long rlen)
 {
-    long rlen;
+    long olen;
+    long rofs;
 
     if (len < 0) rb_raise(rb_eIndexError, "negative length (%ld)", len);
+    olen = RARRAY_LEN(ary);
     if (beg < 0) {
-	beg += RARRAY(ary)->len;
+	beg += olen;
 	if (beg < 0) {
-	    beg -= RARRAY(ary)->len;
-	    rb_raise(rb_eIndexError, "index %ld out of array", beg);
+	    rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld",
+		     beg - olen, -olen);
 	}
     }
-    if (RARRAY(ary)->len < len || RARRAY(ary)->len < beg + len) {
-	len = RARRAY(ary)->len - beg;
+    if (olen < len || olen < beg + len) {
+	len = olen - beg;
     }
 
-    if (NIL_P(rpl)) {
-	rlen = 0;
-    }
-    else {
-	rpl = rb_ary_to_ary(rpl);
-	rlen = RARRAY(rpl)->len;
+    {
+        const VALUE *optr = RARRAY_CONST_PTR_TRANSIENT(ary);
+	rofs = (rptr >= optr && rptr < optr + olen) ? rptr - optr : -1;
     }
-    rb_ary_modify(ary);
 
-    if (beg >= RARRAY(ary)->len) {
-	len = beg + rlen;
-	if (len >= RARRAY(ary)->aux.capa) {
-	    REALLOC_N(RARRAY(ary)->ptr, VALUE, len);
-	    RARRAY(ary)->aux.capa = len;
+    if (beg >= olen) {
+	VALUE target_ary;
+	if (beg > ARY_MAX_SIZE - rlen) {
+	    rb_raise(rb_eIndexError, "index %ld too big", beg);
 	}
-	rb_mem_clear(RARRAY(ary)->ptr + RARRAY(ary)->len, beg - RARRAY(ary)->len);
+	target_ary = ary_ensure_room_for_push(ary, rlen-len); /* len is 0 or negative */
+	len = beg + rlen;
+	ary_mem_clear(ary, olen, beg - olen);
 	if (rlen > 0) {
-	    MEMCPY(RARRAY(ary)->ptr + beg, RARRAY(rpl)->ptr, VALUE, rlen);
+            if (rofs != -1) rptr = RARRAY_CONST_PTR_TRANSIENT(ary) + rofs;
+	    ary_memcpy0(ary, beg, rlen, rptr, target_ary);
 	}
-	RARRAY(ary)->len = len;
+	ARY_SET_LEN(ary, len);
     }
     else {
 	long alen;
 
-	if (beg + len > RARRAY(ary)->len) {
-	    len = RARRAY(ary)->len - beg;
+	if (olen - len > ARY_MAX_SIZE - rlen) {
+	    rb_raise(rb_eIndexError, "index %ld too big", olen + rlen - len);
 	}
-
-	alen = RARRAY(ary)->len + rlen - len;
-	if (alen >= RARRAY(ary)->aux.capa) {
-	    REALLOC_N(RARRAY(ary)->ptr, VALUE, alen);
-	    RARRAY(ary)->aux.capa = alen;
+	rb_ary_modify(ary);
+	alen = olen + rlen - len;
+	if (alen >= ARY_CAPA(ary)) {
+	    ary_double_capa(ary, alen);
 	}
 
 	if (len != rlen) {
-	    MEMMOVE(RARRAY(ary)->ptr + beg + rlen, RARRAY(ary)->ptr + beg + len,
-		    VALUE, RARRAY(ary)->len - (beg + len));
-	    RARRAY(ary)->len = alen;
+            RARRAY_PTR_USE_TRANSIENT(ary, ptr,
+                                     MEMMOVE(ptr + beg + rlen, ptr + beg + len,
+                                             VALUE, olen - (beg + len)));
+	    ARY_SET_LEN(ary, alen);
 	}
 	if (rlen > 0) {
-	    MEMMOVE(RARRAY(ary)->ptr + beg, RARRAY(rpl)->ptr, VALUE, rlen);
+            if (rofs != -1) rptr = RARRAY_CONST_PTR_TRANSIENT(ary) + rofs;
+            /* give up wb-protected ary */
+	    MEMMOVE(RARRAY_PTR(ary) + beg, rptr, VALUE, rlen);
 	}
     }
 }
 
-/* 
+void
+rb_ary_set_len(VALUE ary, long len)
+{
+    long capa;
+
+    rb_ary_modify_check(ary);
+    if (ARY_SHARED_P(ary)) {
+	rb_raise(rb_eRuntimeError, "can't set length of shared ");
+    }
+    if (len > (capa = (long)ARY_CAPA(ary))) {
+	rb_bug("probable buffer overflow: %ld for %ld", len, capa);
+    }
+    ARY_SET_LEN(ary, len);
+}
+
+/*!
+ * expands or shrinks \a ary to \a len elements.
+ * expanded region will be filled with Qnil.
+ * \param ary  an array
+ * \param len  new size
+ * \return     \a ary
+ * \post       the size of \a ary is \a len.
+ */
+VALUE
+rb_ary_resize(VALUE ary, long len)
+{
+    long olen;
+
+    rb_ary_modify(ary);
+    olen = RARRAY_LEN(ary);
+    if (len == olen) return ary;
+    if (len > ARY_MAX_SIZE) {
+	rb_raise(rb_eIndexError, "index %ld too big", len);
+    }
+    if (len > olen) {
+	if (len >= ARY_CAPA(ary)) {
+	    ary_double_capa(ary, len);
+	}
+	ary_mem_clear(ary, olen, len - olen);
+	ARY_SET_LEN(ary, len);
+    }
+    else if (ARY_EMBED_P(ary)) {
+        ARY_SET_EMBED_LEN(ary, len);
+    }
+    else if (len <= RARRAY_EMBED_LEN_MAX) {
+	VALUE tmp[RARRAY_EMBED_LEN_MAX];
+	MEMCPY(tmp, ARY_HEAP_PTR(ary), VALUE, len);
+	ary_discard(ary);
+	MEMCPY((VALUE *)ARY_EMBED_PTR(ary), tmp, VALUE, len); /* WB: no new reference */
+        ARY_SET_EMBED_LEN(ary, len);
+    }
+    else {
+	if (olen > len + ARY_DEFAULT_SIZE) {
+            ary_heap_realloc(ary, len);
+	    ARY_SET_CAPA(ary, len);
+	}
+	ARY_SET_HEAP_LEN(ary, len);
+    }
+    ary_verify(ary);
+    return ary;
+}
+
+/*
  *  call-seq:
- *     array[index]         = obj                     ->  obj
- *     array[start, length] = obj or an_array or nil  ->  obj or an_array or nil
- *     array[range]         = obj or an_array or nil  ->  obj or an_array or nil
+ *     ary[index]         = obj                      ->  obj
+ *     ary[start, length] = obj or other_ary or nil  ->  obj or other_ary or nil
+ *     ary[range]         = obj or other_ary or nil  ->  obj or other_ary or nil
+ *
+ *  Element Assignment --- Sets the element at +index+, or replaces a subarray
+ *  from the +start+ index for +length+ elements, or replaces a subarray
+ *  specified by the +range+ of indices.
+ *
+ *  If indices are greater than the current capacity of the array, the array
+ *  grows automatically.  Elements are inserted into the array at +start+ if
+ *  +length+ is zero.
+ *
+ *  Negative indices will count backward from the end of the array.  For
+ *  +start+ and +range+ cases the starting index is just before an element.
+ *
+ *  An IndexError is raised if a negative index points past the beginning of
+ *  the array.
+ *
+ *  See also Array#push, and Array#unshift.
  *
- *  Element Assignment---Sets the element at _index_,
- *  or replaces a subarray starting at _start_ and
- *  continuing for _length_ elements, or replaces a subarray
- *  specified by _range_.  If indices are greater than
- *  the current capacity of the array, the array grows
- *  automatically. A negative indices will count backward
- *  from the end of the array. Inserts elements if _length_ is
- *  zero. If +nil+ is used in the second and third form,
- *  deletes elements from _self_. An +IndexError+ is raised if a
- *  negative index points past the beginning of the array. See also
- *  <code>Array#push</code>, and <code>Array#unshift</code>.
- * 
  *     a = Array.new
  *     a[4] = "4";                 #=> [nil, nil, nil, nil, "4"]
  *     a[0, 3] = [ 'a', 'b', 'c' ] #=> ["a", "b", "c", nil, "4"]
@@ -1027,41 +1964,37 @@ rb_ary_splice(ary, beg, len, rpl)
  *     a[0, 2] = "?"               #=> ["?", 2, nil, "4"]
  *     a[0..2] = "A"               #=> ["A", "4"]
  *     a[-1]   = "Z"               #=> ["A", "Z"]
- *     a[1..-1] = nil              #=> ["A"]
+ *     a[1..-1] = nil              #=> ["A", nil]
+ *     a[1..-1] = []               #=> ["A"]
+ *     a[0, 0] = [ 1, 2 ]          #=> [1, 2, "A"]
+ *     a[3, 0] = "B"               #=> [1, 2, "A", "B"]
  */
 
 static VALUE
-rb_ary_aset(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_aset(int argc, VALUE *argv, VALUE ary)
 {
     long offset, beg, len;
+    VALUE rpl;
 
     if (argc == 3) {
-	if (SYMBOL_P(argv[0])) {
-	    rb_raise(rb_eTypeError, "Symbol as array index");
-	}
-	if (SYMBOL_P(argv[1])) {
-	    rb_raise(rb_eTypeError, "Symbol as subarray length");
-	}
-	rb_ary_splice(ary, NUM2LONG(argv[0]), NUM2LONG(argv[1]), argv[2]);
-	return argv[2];
-    }
-    if (argc != 2) {
-	rb_raise(rb_eArgError, "wrong number of arguments (%d for 2)", argc);
+	rb_ary_modify_check(ary);
+	beg = NUM2LONG(argv[0]);
+	len = NUM2LONG(argv[1]);
+	goto range;
     }
+    rb_check_arity(argc, 2, 2);
+    rb_ary_modify_check(ary);
     if (FIXNUM_P(argv[0])) {
 	offset = FIX2LONG(argv[0]);
 	goto fixnum;
     }
-    if (SYMBOL_P(argv[0])) {
-	rb_raise(rb_eTypeError, "Symbol as array index");
-    }
-    if (rb_range_beg_len(argv[0], &beg, &len, RARRAY(ary)->len, 1)) {
+    if (rb_range_beg_len(argv[0], &beg, &len, RARRAY_LEN(ary), 1)) {
 	/* check if idx is Range */
-	rb_ary_splice(ary, beg, len, argv[1]);
-	return argv[1];
+      range:
+	rpl = rb_ary_to_ary(argv[argc-1]);
+        rb_ary_splice(ary, beg, len, RARRAY_CONST_PTR_TRANSIENT(rpl), RARRAY_LEN(rpl));
+	RB_GC_GUARD(rpl);
+	return argv[argc-1];
     }
 
     offset = NUM2LONG(argv[0]);
@@ -1072,88 +2005,108 @@ fixnum:
 
 /*
  *  call-seq:
- *     array.insert(index, obj...)  -> array
- *  
- *  Inserts the given values before the element with the given index
- *  (which may be negative).
- *     
+ *     ary.insert(index, obj...)  -> ary
+ *
+ *  Inserts the given values before the element with the given +index+.
+ *
+ *  Negative indices count backwards from the end of the array, where +-1+ is
+ *  the last element. If a negative index is used, the given values will be
+ *  inserted after that element, so using an index of +-1+ will insert the
+ *  values at the end of the array.
+ *
  *     a = %w{ a b c d }
  *     a.insert(2, 99)         #=> ["a", "b", 99, "c", "d"]
  *     a.insert(-2, 1, 2, 3)   #=> ["a", "b", 99, "c", 1, 2, 3, "d"]
  */
 
 static VALUE
-rb_ary_insert(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_insert(int argc, VALUE *argv, VALUE ary)
 {
     long pos;
 
-    if (argc == 1) return ary;
-    if (argc < 1) {
-	rb_raise(rb_eArgError, "wrong number of arguments (at least 1)");
-    }
+    rb_check_arity(argc, 1, UNLIMITED_ARGUMENTS);
+    rb_ary_modify_check(ary);
     pos = NUM2LONG(argv[0]);
+    if (argc == 1) return ary;
     if (pos == -1) {
-	pos = RARRAY(ary)->len;
+	pos = RARRAY_LEN(ary);
     }
-    if (pos < 0) {
+    else if (pos < 0) {
+	long minpos = -RARRAY_LEN(ary) - 1;
+	if (pos < minpos) {
+	    rb_raise(rb_eIndexError, "index %ld too small for array; minimum: %ld",
+		     pos, minpos);
+	}
 	pos++;
     }
-    rb_ary_splice(ary, pos, 0, rb_ary_new4(argc - 1, argv + 1));
+    rb_ary_splice(ary, pos, 0, argv + 1, argc - 1);
     return ary;
 }
 
+static VALUE
+rb_ary_length(VALUE ary);
+
+static VALUE
+ary_enum_length(VALUE ary, VALUE args, VALUE eobj)
+{
+    return rb_ary_length(ary);
+}
+
 /*
  *  call-seq:
- *     array.each {|item| block }   ->   array
- *  
- *  Calls <i>block</i> once for each element in <i>self</i>, passing that
- *  element as a parameter.
- *     
+ *     ary.each {|item| block}    -> ary
+ *     ary.each                   -> Enumerator
+ *
+ *  Calls the given block once for each element in +self+, passing that element
+ *  as a parameter.  Returns the array itself.
+ *
+ *  If no block is given, an Enumerator is returned.
+ *
  *     a = [ "a", "b", "c" ]
  *     a.each {|x| print x, " -- " }
- *     
+ *
  *  produces:
- *     
+ *
  *     a -- b -- c --
  */
 
 VALUE
-rb_ary_each(ary)
-    VALUE ary;
+rb_ary_each(VALUE ary)
 {
     long i;
-
-    for (i=0; i<RARRAY(ary)->len; i++) {
-	rb_yield(RARRAY(ary)->ptr[i]);
+    ary_verify(ary);
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	rb_yield(RARRAY_AREF(ary, i));
     }
     return ary;
 }
 
 /*
  *  call-seq:
- *     array.each_index {|index| block }  ->  array
- *  
- *  Same as <code>Array#each</code>, but passes the index of the element
- *  instead of the element itself.
- *     
+ *     ary.each_index {|index| block}    -> ary
+ *     ary.each_index                    -> Enumerator
+ *
+ *  Same as Array#each, but passes the +index+ of the element instead of the
+ *  element itself.
+ *
+ *  An Enumerator is returned if no block is given.
+ *
  *     a = [ "a", "b", "c" ]
  *     a.each_index {|x| print x, " -- " }
- *     
+ *
  *  produces:
- *     
+ *
  *     0 -- 1 -- 2 --
  */
 
 static VALUE
-rb_ary_each_index(ary)
-    VALUE ary;
+rb_ary_each_index(VALUE ary)
 {
     long i;
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
 
-    for (i=0; i<RARRAY(ary)->len; i++) {
+    for (i=0; i<RARRAY_LEN(ary); i++) {
 	rb_yield(LONG2NUM(i));
     }
     return ary;
@@ -1161,29 +2114,32 @@ rb_ary_each_index(ary)
 
 /*
  *  call-seq:
- *     array.reverse_each {|item| block } 
- *  
- *  Same as <code>Array#each</code>, but traverses <i>self</i> in reverse
- *  order.
- *     
+ *     ary.reverse_each {|item| block}    -> ary
+ *     ary.reverse_each                   -> Enumerator
+ *
+ *  Same as Array#each, but traverses +self+ in reverse order.
+ *
  *     a = [ "a", "b", "c" ]
  *     a.reverse_each {|x| print x, " " }
- *     
+ *
  *  produces:
- *     
+ *
  *     c b a
  */
 
 static VALUE
-rb_ary_reverse_each(ary)
-    VALUE ary;
+rb_ary_reverse_each(VALUE ary)
 {
-    long len = RARRAY(ary)->len;
+    long len;
 
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    len = RARRAY_LEN(ary);
     while (len--) {
-	rb_yield(RARRAY(ary)->ptr[len]);
-	if (RARRAY(ary)->len < len) {
-	    len = RARRAY(ary)->len;
+	long nlen;
+	rb_yield(RARRAY_AREF(ary, len));
+	nlen = RARRAY_LEN(ary);
+	if (nlen < len) {
+	    len = nlen;
 	}
     }
     return ary;
@@ -1191,555 +2147,908 @@ rb_ary_reverse_each(ary)
 
 /*
  *  call-seq:
- *     array.length -> int
- *  
- *  Returns the number of elements in <i>self</i>. May be zero.
- *     
+ *     ary.length -> int
+ *
+ *  Returns the number of elements in +self+. May be zero.
+ *
  *     [ 1, 2, 3, 4, 5 ].length   #=> 5
+ *     [].length                  #=> 0
  */
 
 static VALUE
-rb_ary_length(ary)
-    VALUE ary;
+rb_ary_length(VALUE ary)
 {
-    return LONG2NUM(RARRAY(ary)->len);
+    long len = RARRAY_LEN(ary);
+    return LONG2NUM(len);
 }
 
 /*
  *  call-seq:
- *     array.empty?   -> true or false
- *  
- *  Returns <code>true</code> if <i>self</i> array contains no elements.
- *     
+ *     ary.empty?   -> true or false
+ *
+ *  Returns +true+ if +self+ contains no elements.
+ *
  *     [].empty?   #=> true
  */
 
 static VALUE
-rb_ary_empty_p(ary)
-    VALUE ary;
+rb_ary_empty_p(VALUE ary)
 {
-    if (RARRAY(ary)->len == 0)
+    if (RARRAY_LEN(ary) == 0)
 	return Qtrue;
     return Qfalse;
 }
 
 VALUE
-rb_ary_dup(ary)
-    VALUE ary;
+rb_ary_dup(VALUE ary)
 {
-    VALUE dup = rb_ary_new2(RARRAY(ary)->len);
+    long len = RARRAY_LEN(ary);
+    VALUE dup = rb_ary_new2(len);
+    ary_memcpy(dup, 0, len, RARRAY_CONST_PTR_TRANSIENT(ary));
+    ARY_SET_LEN(dup, len);
 
-    DUPSETUP(dup, ary);
-    MEMCPY(RARRAY(dup)->ptr, RARRAY(ary)->ptr, VALUE, RARRAY(ary)->len);
-    RARRAY(dup)->len = RARRAY(ary)->len;
+    ary_verify(ary);
+    ary_verify(dup);
     return dup;
 }
 
+VALUE
+rb_ary_resurrect(VALUE ary)
+{
+    return ary_make_partial(ary, rb_cArray, 0, RARRAY_LEN(ary));
+}
+
 extern VALUE rb_output_fs;
 
+static void ary_join_1(VALUE obj, VALUE ary, VALUE sep, long i, VALUE result, int *first);
+
 static VALUE
-inspect_join(ary, arg)
-    VALUE ary;
-    VALUE *arg;
+recursive_join(VALUE obj, VALUE argp, int recur)
 {
-    return rb_ary_join(arg[0], arg[1]);
+    VALUE *arg = (VALUE *)argp;
+    VALUE ary = arg[0];
+    VALUE sep = arg[1];
+    VALUE result = arg[2];
+    int *first = (int *)arg[3];
+
+    if (recur) {
+	rb_raise(rb_eArgError, "recursive array join");
+    }
+    else {
+	ary_join_1(obj, ary, sep, 0, result, first);
+    }
+    return Qnil;
 }
 
-VALUE
-rb_ary_join(ary, sep)
-    VALUE ary, sep;
+static void
+ary_join_0(VALUE ary, VALUE sep, long max, VALUE result)
 {
-    long len = 1, i;
-    int taint = Qfalse;
-    VALUE result, tmp;
-
-    if (RARRAY(ary)->len == 0) return rb_str_new(0, 0);
-    if (OBJ_TAINTED(ary) || OBJ_TAINTED(sep)) taint = Qtrue;
+    long i;
+    VALUE val;
 
-    for (i=0; i<RARRAY(ary)->len; i++) {
-	tmp = rb_check_string_type(RARRAY(ary)->ptr[i]);
-	len += NIL_P(tmp) ? 10 : RSTRING(tmp)->len;
-    }
-    if (!NIL_P(sep)) {
-	StringValue(sep);
-	len += RSTRING(sep)->len * (RARRAY(ary)->len - 1);
+    if (max > 0) rb_enc_copy(result, RARRAY_AREF(ary, 0));
+    for (i=0; i<max; i++) {
+	val = RARRAY_AREF(ary, i);
+	if (i > 0 && !NIL_P(sep))
+	    rb_str_buf_append(result, sep);
+	rb_str_buf_append(result, val);
+	if (OBJ_TAINTED(val)) OBJ_TAINT(result);
     }
-    result = rb_str_buf_new(len);
-    for (i=0; i<RARRAY(ary)->len; i++) {
-	tmp = RARRAY(ary)->ptr[i];
-	switch (TYPE(tmp)) {
-	  case T_STRING:
-	    break;
-	  case T_ARRAY:
-	    if (rb_inspecting_p(tmp)) {
-		tmp = rb_str_new2("[...]");
+}
+
+static void
+ary_join_1(VALUE obj, VALUE ary, VALUE sep, long i, VALUE result, int *first)
+{
+    VALUE val, tmp;
+
+    for (; i<RARRAY_LEN(ary); i++) {
+	if (i > 0 && !NIL_P(sep))
+	    rb_str_buf_append(result, sep);
+
+	val = RARRAY_AREF(ary, i);
+	if (RB_TYPE_P(val, T_STRING)) {
+	  str_join:
+	    rb_str_buf_append(result, val);
+	    if (*first) {
+		rb_enc_copy(result, val);
+		*first = FALSE;
+	    }
+	}
+	else if (RB_TYPE_P(val, T_ARRAY)) {
+	    obj = val;
+	  ary_join:
+	    if (val == ary) {
+		rb_raise(rb_eArgError, "recursive array join");
 	    }
 	    else {
-		VALUE args[2];
+		VALUE args[4];
 
-		args[0] = tmp;
+		*first = FALSE;
+		args[0] = val;
 		args[1] = sep;
-		tmp = rb_protect_inspect(inspect_join, ary, (VALUE)args);
+		args[2] = result;
+		args[3] = (VALUE)first;
+		rb_exec_recursive(recursive_join, obj, (VALUE)args);
 	    }
-	    break;
-	  default:
-	    tmp = rb_obj_as_string(tmp);
 	}
-	if (i > 0 && !NIL_P(sep))
-	    rb_str_buf_append(result, sep);
-	rb_str_buf_append(result, tmp);
-	if (OBJ_TAINTED(tmp)) taint = Qtrue;
+	else {
+	    tmp = rb_check_string_type(val);
+	    if (!NIL_P(tmp)) {
+		val = tmp;
+		goto str_join;
+	    }
+	    tmp = rb_check_array_type(val);
+	    if (!NIL_P(tmp)) {
+		obj = val;
+		val = tmp;
+		goto ary_join;
+	    }
+	    val = rb_obj_as_string(val);
+	    goto str_join;
+	}
+    }
+}
+
+VALUE
+rb_ary_join(VALUE ary, VALUE sep)
+{
+    long len = 1, i;
+    int taint = FALSE;
+    VALUE val, tmp, result;
+
+    if (RARRAY_LEN(ary) == 0) return rb_usascii_str_new(0, 0);
+    if (OBJ_TAINTED(ary)) taint = TRUE;
+
+    if (!NIL_P(sep)) {
+	StringValue(sep);
+	len += RSTRING_LEN(sep) * (RARRAY_LEN(ary) - 1);
+    }
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	val = RARRAY_AREF(ary, i);
+	tmp = rb_check_string_type(val);
+
+	if (NIL_P(tmp) || tmp != val) {
+	    int first;
+	    result = rb_str_buf_new(len + (RARRAY_LEN(ary)-i)*10);
+	    rb_enc_associate(result, rb_usascii_encoding());
+	    if (taint) OBJ_TAINT(result);
+	    ary_join_0(ary, sep, i, result);
+	    first = i == 0;
+	    ary_join_1(ary, ary, sep, i, result, &first);
+	    return result;
+	}
+
+	len += RSTRING_LEN(tmp);
     }
 
+    result = rb_str_buf_new(len);
     if (taint) OBJ_TAINT(result);
+    ary_join_0(ary, sep, RARRAY_LEN(ary), result);
+
     return result;
 }
 
 /*
  *  call-seq:
- *     array.join(sep=$,)    -> str
- *  
+ *     ary.join(separator=$,)    -> str
+ *
  *  Returns a string created by converting each element of the array to
- *  a string, separated by <i>sep</i>.
- *     
+ *  a string, separated by the given +separator+.
+ *  If the +separator+ is +nil+, it uses current <code>$,</code>.
+ *  If both the +separator+ and <code>$,</code> are +nil+,
+ *  it uses an empty string.
+ *
  *     [ "a", "b", "c" ].join        #=> "abc"
  *     [ "a", "b", "c" ].join("-")   #=> "a-b-c"
+ *
+ *  For nested arrays, join is applied recursively:
+ *
+ *     [ "a", [1, 2, [:x, :y]], "b" ].join("-")   #=> "a-1-2-x-y-b"
  */
 
 static VALUE
-rb_ary_join_m(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_join_m(int argc, VALUE *argv, VALUE ary)
 {
     VALUE sep;
 
-    rb_scan_args(argc, argv, "01", &sep);
-    if (NIL_P(sep)) sep = rb_output_fs;
-    
+    if (rb_check_arity(argc, 0, 1) == 0) {
+        sep = rb_output_fs;
+    }
+    else if (NIL_P(sep = argv[0])) {
+        sep = rb_output_fs;
+    }
+
     return rb_ary_join(ary, sep);
 }
 
+static VALUE
+inspect_ary(VALUE ary, VALUE dummy, int recur)
+{
+    int tainted = OBJ_TAINTED(ary);
+    long i;
+    VALUE s, str;
+
+    if (recur) return rb_usascii_str_new_cstr("[...]");
+    str = rb_str_buf_new2("[");
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	s = rb_inspect(RARRAY_AREF(ary, i));
+	if (OBJ_TAINTED(s)) tainted = TRUE;
+	if (i > 0) rb_str_buf_cat2(str, ", ");
+	else rb_enc_copy(str, s);
+	rb_str_buf_append(str, s);
+    }
+    rb_str_buf_cat2(str, "]");
+    if (tainted) OBJ_TAINT(str);
+    return str;
+}
+
 /*
  *  call-seq:
- *     array.to_s -> string
- *  
- *  Returns _self_<code>.join</code>.
- *     
- *     [ "a", "e", "i", "o" ].to_s   #=> "aeio"
+ *     ary.inspect  -> string
+ *     ary.to_s     -> string
+ *
+ *  Creates a string representation of +self+.
  *
+ *     [ "a", "b", "c" ].to_s     #=> "[\"a\", \"b\", \"c\"]"
  */
 
-VALUE
-rb_ary_to_s(ary)
-    VALUE ary;
+static VALUE
+rb_ary_inspect(VALUE ary)
 {
-    if (RARRAY(ary)->len == 0) return rb_str_new(0, 0);
-    
-    return rb_ary_join(ary, rb_output_fs);
+    if (RARRAY_LEN(ary) == 0) return rb_usascii_str_new2("[]");
+    return rb_exec_recursive(inspect_ary, ary, 0);
 }
 
-static ID inspect_key;
+VALUE
+rb_ary_to_s(VALUE ary)
+{
+    return rb_ary_inspect(ary);
+}
 
-struct inspect_arg {
-    VALUE (*func)();
-    VALUE arg1, arg2;
-};
+/*
+ *  call-seq:
+ *     ary.to_a     -> ary
+ *
+ *  Returns +self+.
+ *
+ *  If called on a subclass of Array, converts the receiver to an Array object.
+ */
 
 static VALUE
-inspect_call(arg)
-    struct inspect_arg *arg;
+rb_ary_to_a(VALUE ary)
 {
-    return (*arg->func)(arg->arg1, arg->arg2);
+    if (rb_obj_class(ary) != rb_cArray) {
+	VALUE dup = rb_ary_new2(RARRAY_LEN(ary));
+	rb_ary_replace(dup, ary);
+	return dup;
+    }
+    return ary;
 }
 
+/*
+ *  call-seq:
+ *     ary.to_h            -> hash
+ *     ary.to_h { block }  -> hash
+ *
+ *  Returns the result of interpreting <i>ary</i> as an array of
+ *  <tt>[key, value]</tt> pairs.
+ *
+ *     [[:foo, :bar], [1, 2]].to_h
+ *       # => {:foo => :bar, 1 => 2}
+ *
+ *  If a block is given, the results of the block on each element of
+ *  the array will be used as pairs.
+ *
+ *     ["foo", "bar"].to_h {|s| [s.ord, s]}
+ *       # => {102=>"foo", 98=>"bar"}
+ */
+
 static VALUE
-get_inspect_tbl(create)
-    int create;
+rb_ary_to_h(VALUE ary)
 {
-    VALUE inspect_tbl = rb_thread_local_aref(rb_thread_current(), inspect_key);
-
-    if (NIL_P(inspect_tbl)) {
-	if (create) {
-	  tbl_init:
-	    inspect_tbl = rb_ary_new();
-	    rb_thread_local_aset(rb_thread_current(), inspect_key, inspect_tbl);
+    long i;
+    VALUE hash = rb_hash_new_with_size(RARRAY_LEN(ary));
+    int block_given = rb_block_given_p();
+
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	const VALUE e = rb_ary_elt(ary, i);
+	const VALUE elt = block_given ? rb_yield_force_blockarg(e) : e;
+	const VALUE key_value_pair = rb_check_array_type(elt);
+	if (NIL_P(key_value_pair)) {
+	    rb_raise(rb_eTypeError, "wrong element type %"PRIsVALUE" at %ld (expected array)",
+		     rb_obj_class(elt), i);
 	}
+	if (RARRAY_LEN(key_value_pair) != 2) {
+	    rb_raise(rb_eArgError, "wrong array length at %ld (expected 2, was %ld)",
+		i, RARRAY_LEN(key_value_pair));
+	}
+	rb_hash_aset(hash, RARRAY_AREF(key_value_pair, 0), RARRAY_AREF(key_value_pair, 1));
     }
-    else if (TYPE(inspect_tbl) != T_ARRAY) {
-	rb_warn("invalid inspect_tbl value");
-	if (create) goto tbl_init;
-	rb_thread_local_aset(rb_thread_current(), inspect_key, Qnil);
-	return Qnil;
-    }
-    return inspect_tbl;
+    return hash;
 }
 
+/*
+ *  call-seq:
+ *     ary.to_ary -> ary
+ *
+ *  Returns +self+.
+ */
+
 static VALUE
-inspect_ensure(obj)
-    VALUE obj;
+rb_ary_to_ary_m(VALUE ary)
 {
-    VALUE inspect_tbl;
-
-    inspect_tbl = get_inspect_tbl(Qfalse);
-    if (!NIL_P(inspect_tbl)) {
-	rb_ary_pop(inspect_tbl);
-    }
-    return 0;
+    return ary;
 }
 
-VALUE
-rb_protect_inspect(func, obj, arg)
-    VALUE (*func)(ANYARGS);
-    VALUE obj, arg;
+static void
+ary_reverse(VALUE *p1, VALUE *p2)
 {
-    struct inspect_arg iarg;
-    VALUE inspect_tbl;
-    VALUE id;
-
-    inspect_tbl = get_inspect_tbl(Qtrue);
-    id = rb_obj_id(obj);
-    if (rb_ary_includes(inspect_tbl, id)) {
-	return (*func)(obj, arg);
+    while (p1 < p2) {
+	VALUE tmp = *p1;
+	*p1++ = *p2;
+	*p2-- = tmp;
     }
-    rb_ary_push(inspect_tbl, id);
-    iarg.func = func;
-    iarg.arg1 = obj;
-    iarg.arg2 = arg;
-
-    return rb_ensure(inspect_call, (VALUE)&iarg, inspect_ensure, obj);
 }
 
 VALUE
-rb_inspecting_p(obj)
-    VALUE obj;
-{
-    VALUE inspect_tbl;
-
-    inspect_tbl = get_inspect_tbl(Qfalse);
-    if (NIL_P(inspect_tbl)) return Qfalse;
-    return rb_ary_includes(inspect_tbl, rb_obj_id(obj));
-}
-
-static VALUE
-inspect_ary(ary)
-    VALUE ary;
+rb_ary_reverse(VALUE ary)
 {
-    int tainted = OBJ_TAINTED(ary);
-    long i;
-    VALUE s, str;
+    VALUE *p2;
+    long len = RARRAY_LEN(ary);
 
-    str = rb_str_buf_new2("[");
-    for (i=0; i<RARRAY(ary)->len; i++) {
-	s = rb_inspect(RARRAY(ary)->ptr[i]);
-	if (OBJ_TAINTED(s)) tainted = Qtrue;
-	if (i > 0) rb_str_buf_cat2(str, ", ");
-	rb_str_buf_append(str, s);
+    rb_ary_modify(ary);
+    if (len > 1) {
+        RARRAY_PTR_USE_TRANSIENT(ary, p1, {
+            p2 = p1 + len - 1;	/* points last item */
+            ary_reverse(p1, p2);
+	}); /* WB: no new reference */
     }
-    rb_str_buf_cat2(str, "]");
-    if (tainted) OBJ_TAINT(str);
-    return str;
+    return ary;
 }
 
 /*
  *  call-seq:
- *     array.inspect  -> string
+ *     ary.reverse!   -> ary
  *
- *  Create a printable version of <i>array</i>.
+ *  Reverses +self+ in place.
+ *
+ *     a = [ "a", "b", "c" ]
+ *     a.reverse!       #=> ["c", "b", "a"]
+ *     a                #=> ["c", "b", "a"]
  */
 
 static VALUE
-rb_ary_inspect(ary)
-    VALUE ary;
+rb_ary_reverse_bang(VALUE ary)
 {
-    if (RARRAY(ary)->len == 0) return rb_str_new2("[]");
-    if (rb_inspecting_p(ary)) return rb_str_new2("[...]");
-    return rb_protect_inspect(inspect_ary, ary, 0);
+    return rb_ary_reverse(ary);
 }
 
 /*
  *  call-seq:
- *     array.to_a     -> array
- *  
- *  Returns _self_. If called on a subclass of Array, converts
- *  the receiver to an Array object.
+ *     ary.reverse    -> new_ary
+ *
+ *  Returns a new array containing +self+'s elements in reverse order.
+ *
+ *     [ "a", "b", "c" ].reverse   #=> ["c", "b", "a"]
+ *     [ 1 ].reverse               #=> [1]
  */
 
 static VALUE
-rb_ary_to_a(ary)
-    VALUE ary;
+rb_ary_reverse_m(VALUE ary)
 {
-    if (rb_obj_class(ary) != rb_cArray) {
-	VALUE dup = rb_ary_new2(RARRAY(ary)->len);
-	rb_ary_replace(dup, ary);
-	return dup;
+    long len = RARRAY_LEN(ary);
+    VALUE dup = rb_ary_new2(len);
+
+    if (len > 0) {
+        const VALUE *p1 = RARRAY_CONST_PTR_TRANSIENT(ary);
+        VALUE *p2 = (VALUE *)RARRAY_CONST_PTR_TRANSIENT(dup) + len - 1;
+	do *p2-- = *p1++; while (--len > 0);
     }
-    return ary;
+    ARY_SET_LEN(dup, RARRAY_LEN(ary));
+    return dup;
 }
 
-/*
- *  call-seq:
- *     array.to_ary -> array
- *  
- *  Returns _self_.
- */
+static inline long
+rotate_count(long cnt, long len)
+{
+    return (cnt < 0) ? (len - (~cnt % len) - 1) : (cnt % len);
+}
 
-static VALUE
-rb_ary_to_ary_m(ary)
-    VALUE ary;
+static void
+ary_rotate_ptr(VALUE *ptr, long len, long cnt)
 {
-    return ary;
+    --len;
+    if (cnt < len) ary_reverse(ptr + cnt, ptr + len);
+    if (--cnt > 0) ary_reverse(ptr, ptr + cnt);
+    if (len > 0) ary_reverse(ptr, ptr + len);
 }
 
 VALUE
-rb_ary_reverse(ary)
-    VALUE ary;
+rb_ary_rotate(VALUE ary, long cnt)
 {
-    VALUE *p1, *p2;
-    VALUE tmp;
-
     rb_ary_modify(ary);
-    if (RARRAY(ary)->len > 1) {
-	p1 = RARRAY(ary)->ptr;
-	p2 = p1 + RARRAY(ary)->len - 1;	/* points last item */
 
-	while (p1 < p2) {
-	    tmp = *p1;
-	    *p1++ = *p2;
-	    *p2-- = tmp;
-	}
+    if (cnt != 0) {
+        long len = RARRAY_LEN(ary);
+        if (len > 0 && (cnt = rotate_count(cnt, len)) > 0) {
+            RARRAY_PTR_USE_TRANSIENT(ary, ptr, ary_rotate_ptr(ptr, len, cnt));
+            return ary;
+        }
     }
-    return ary;
+    return Qnil;
 }
 
 /*
  *  call-seq:
- *     array.reverse!   -> array 
- *  
- *  Reverses _self_ in place.
- *     
- *     a = [ "a", "b", "c" ]
- *     a.reverse!       #=> ["c", "b", "a"]
- *     a                #=> ["c", "b", "a"]
+ *     ary.rotate!(count=1)   -> ary
+ *
+ *  Rotates +self+ in place so that the element at +count+ comes first, and
+ *  returns +self+.
+ *
+ *  If +count+ is negative then it rotates in the opposite direction, starting
+ *  from the end of the array where +-1+ is the last element.
+ *
+ *     a = [ "a", "b", "c", "d" ]
+ *     a.rotate!        #=> ["b", "c", "d", "a"]
+ *     a                #=> ["b", "c", "d", "a"]
+ *     a.rotate!(2)     #=> ["d", "a", "b", "c"]
+ *     a.rotate!(-3)    #=> ["a", "b", "c", "d"]
  */
 
 static VALUE
-rb_ary_reverse_bang(ary)
-    VALUE ary;
+rb_ary_rotate_bang(int argc, VALUE *argv, VALUE ary)
 {
-    return rb_ary_reverse(ary);
+    long n = (rb_check_arity(argc, 0, 1) ? NUM2LONG(argv[0]) : 1);
+    rb_ary_rotate(ary, n);
+    return ary;
 }
 
 /*
  *  call-seq:
- *     array.reverse -> an_array
- *  
- *  Returns a new array containing <i>self</i>'s elements in reverse order.
- *     
- *     [ "a", "b", "c" ].reverse   #=> ["c", "b", "a"]
- *     [ 1 ].reverse               #=> [1]
+ *     ary.rotate(count=1)    -> new_ary
+ *
+ *  Returns a new array by rotating +self+ so that the element at +count+ is
+ *  the first element of the new array.
+ *
+ *  If +count+ is negative then it rotates in the opposite direction, starting
+ *  from the end of +self+ where +-1+ is the last element.
+ *
+ *     a = [ "a", "b", "c", "d" ]
+ *     a.rotate         #=> ["b", "c", "d", "a"]
+ *     a                #=> ["a", "b", "c", "d"]
+ *     a.rotate(2)      #=> ["c", "d", "a", "b"]
+ *     a.rotate(-3)     #=> ["b", "c", "d", "a"]
  */
 
 static VALUE
-rb_ary_reverse_m(ary)
-    VALUE ary;
+rb_ary_rotate_m(int argc, VALUE *argv, VALUE ary)
 {
-    return rb_ary_reverse(rb_ary_dup(ary));
+    VALUE rotated;
+    const VALUE *ptr;
+    long len;
+    long cnt = (rb_check_arity(argc, 0, 1) ? NUM2LONG(argv[0]) : 1);
+
+    len = RARRAY_LEN(ary);
+    rotated = rb_ary_new2(len);
+    if (len > 0) {
+	cnt = rotate_count(cnt, len);
+        ptr = RARRAY_CONST_PTR_TRANSIENT(ary);
+	len -= cnt;
+	ary_memcpy(rotated, 0, len, ptr + cnt);
+	ary_memcpy(rotated, len, cnt, ptr);
+    }
+    ARY_SET_LEN(rotated, RARRAY_LEN(ary));
+    return rotated;
 }
 
 struct ary_sort_data {
     VALUE ary;
-    VALUE *ptr;
-    long len;
+    struct cmp_opt_data cmp_opt;
 };
 
-static void
-ary_sort_check(data)
-    struct ary_sort_data *data;
+static VALUE
+sort_reentered(VALUE ary)
 {
-    if (RARRAY(data->ary)->ptr != data->ptr || RARRAY(data->ary)->len != data->len) {
-	rb_raise(rb_eArgError, "array modified during sort");
+    if (RBASIC(ary)->klass) {
+	rb_raise(rb_eRuntimeError, "sort reentered");
     }
+    return Qnil;
 }
 
 static int
-sort_1(a, b, data)
-    VALUE *a, *b;
-    struct ary_sort_data *data;
+sort_1(const void *ap, const void *bp, void *dummy)
 {
-    VALUE retval = rb_yield_values(2, *a, *b);
+    struct ary_sort_data *data = dummy;
+    VALUE retval = sort_reentered(data->ary);
+    VALUE a = *(const VALUE *)ap, b = *(const VALUE *)bp;
+    VALUE args[2];
     int n;
 
-    n = rb_cmpint(retval, *a, *b);
-    ary_sort_check(data);
+    args[0] = a;
+    args[1] = b;
+    retval = rb_yield_values2(2, args);
+    n = rb_cmpint(retval, a, b);
+    sort_reentered(data->ary);
     return n;
 }
 
 static int
-sort_2(ap, bp, data)
-    VALUE *ap, *bp;
-    struct ary_sort_data *data;
+sort_2(const void *ap, const void *bp, void *dummy)
 {
-    VALUE retval;
-    VALUE a = *ap, b = *bp;
+    struct ary_sort_data *data = dummy;
+    VALUE retval = sort_reentered(data->ary);
+    VALUE a = *(const VALUE *)ap, b = *(const VALUE *)bp;
     int n;
 
-    if (FIXNUM_P(a) && FIXNUM_P(b)) {
+    if (FIXNUM_P(a) && FIXNUM_P(b) && CMP_OPTIMIZABLE(data->cmp_opt, Fixnum)) {
 	if ((long)a > (long)b) return 1;
 	if ((long)a < (long)b) return -1;
 	return 0;
     }
-    if (TYPE(a) == T_STRING) {
-	if (TYPE(b) == T_STRING) return rb_str_cmp(a, b);
+    if (STRING_P(a) && STRING_P(b) && CMP_OPTIMIZABLE(data->cmp_opt, String)) {
+	return rb_str_cmp(a, b);
+    }
+    if (RB_FLOAT_TYPE_P(a) && CMP_OPTIMIZABLE(data->cmp_opt, Float)) {
+	return rb_float_cmp(a, b);
     }
 
-    retval = rb_funcall(a, id_cmp, 1, b);
+    retval = rb_funcallv(a, id_cmp, 1, &b);
     n = rb_cmpint(retval, a, b);
-    ary_sort_check(data);
+    sort_reentered(data->ary);
 
     return n;
 }
 
-static VALUE
-sort_internal(ary)
-    VALUE ary;
+/*
+ *  call-seq:
+ *     ary.sort!                   -> ary
+ *     ary.sort! {|a, b| block}    -> ary
+ *
+ *  Sorts +self+ in place.
+ *
+ *  Comparisons for the sort will be done using the <code><=></code> operator
+ *  or using an optional code block.
+ *
+ *  The block must implement a comparison between +a+ and +b+ and return
+ *  an integer less than 0 when +b+ follows +a+, +0+ when +a+ and +b+
+ *  are equivalent, or an integer greater than 0 when +a+ follows +b+.
+ *
+ *  The result is not guaranteed to be stable.  When the comparison of two
+ *  elements returns +0+, the order of the elements is unpredictable.
+ *
+ *     ary = [ "d", "a", "e", "c", "b" ]
+ *     ary.sort!                     #=> ["a", "b", "c", "d", "e"]
+ *     ary.sort! {|a, b| b <=> a}    #=> ["e", "d", "c", "b", "a"]
+ *
+ *  See also Enumerable#sort_by.
+ */
+
+VALUE
+rb_ary_sort_bang(VALUE ary)
 {
-    struct ary_sort_data data;
+    rb_ary_modify(ary);
+    assert(!ARY_SHARED_P(ary));
+    if (RARRAY_LEN(ary) > 1) {
+	VALUE tmp = ary_make_substitution(ary); /* only ary refers tmp */
+	struct ary_sort_data data;
+	long len = RARRAY_LEN(ary);
+	RBASIC_CLEAR_CLASS(tmp);
+	data.ary = tmp;
+	data.cmp_opt.opt_methods = 0;
+	data.cmp_opt.opt_inited = 0;
+	RARRAY_PTR_USE(tmp, ptr, {
+            ruby_qsort(ptr, len, sizeof(VALUE),
+                       rb_block_given_p()?sort_1:sort_2, &data);
+	}); /* WB: no new reference */
+	rb_ary_modify(ary);
+        if (ARY_EMBED_P(tmp)) {
+            if (ARY_SHARED_P(ary)) { /* ary might be destructively operated in the given block */
+                rb_ary_unshare(ary);
+		FL_SET_EMBED(ary);
+            }
+	    ary_memcpy(ary, 0, ARY_EMBED_LEN(tmp), ARY_EMBED_PTR(tmp));
+            ARY_SET_LEN(ary, ARY_EMBED_LEN(tmp));
+        }
+        else {
+            if (!ARY_EMBED_P(ary) && ARY_HEAP_PTR(ary) == ARY_HEAP_PTR(tmp)) {
+                FL_UNSET_SHARED(ary);
+                ARY_SET_CAPA(ary, RARRAY_LEN(tmp));
+            }
+            else {
+                assert(!ARY_SHARED_P(tmp));
+                if (ARY_EMBED_P(ary)) {
+                    FL_UNSET_EMBED(ary);
+                }
+                else if (ARY_SHARED_P(ary)) {
+                    /* ary might be destructively operated in the given block */
+                    rb_ary_unshare(ary);
+                }
+                else {
+                    ary_heap_free(ary);
+                }
+                ARY_SET_PTR(ary, ARY_HEAP_PTR(tmp));
+                ARY_SET_HEAP_LEN(ary, len);
+                ARY_SET_CAPA(ary, ARY_HEAP_LEN(tmp));
+            }
+            /* tmp was lost ownership for the ptr */
+            FL_UNSET(tmp, FL_FREEZE);
+            FL_SET_EMBED(tmp);
+            ARY_SET_EMBED_LEN(tmp, 0);
+            FL_SET(tmp, FL_FREEZE);
+        }
+        /* tmp will be GC'ed. */
+        RBASIC_SET_CLASS_RAW(tmp, rb_cArray); /* rb_cArray must be marked */
+    }
+    ary_verify(ary);
+    return ary;
+}
+
+/*
+ *  call-seq:
+ *     ary.sort                   -> new_ary
+ *     ary.sort {|a, b| block}    -> new_ary
+ *
+ *  Returns a new array created by sorting +self+.
+ *
+ *  Comparisons for the sort will be done using the <code><=></code> operator
+ *  or using an optional code block.
+ *
+ *  The block must implement a comparison between +a+ and +b+ and return
+ *  an integer less than 0 when +b+ follows +a+, +0+ when +a+ and +b+
+ *  are equivalent, or an integer greater than 0 when +a+ follows +b+.
+ *
+ *  The result is not guaranteed to be stable.  When the comparison of two
+ *  elements returns +0+, the order of the elements is unpredictable.
+ *
+ *     ary = [ "d", "a", "e", "c", "b" ]
+ *     ary.sort                     #=> ["a", "b", "c", "d", "e"]
+ *     ary.sort {|a, b| b <=> a}    #=> ["e", "d", "c", "b", "a"]
+ *
+ *  See also Enumerable#sort_by.
+ */
 
-    data.ary = ary;
-    data.ptr = RARRAY(ary)->ptr; data.len = RARRAY(ary)->len;
-    qsort(RARRAY(ary)->ptr, RARRAY(ary)->len, sizeof(VALUE),
-	  rb_block_given_p()?sort_1:sort_2, &data);
+VALUE
+rb_ary_sort(VALUE ary)
+{
+    ary = rb_ary_dup(ary);
+    rb_ary_sort_bang(ary);
     return ary;
 }
 
+static VALUE rb_ary_bsearch_index(VALUE ary);
+
+/*
+ *  call-seq:
+ *     ary.bsearch {|x| block }  -> elem
+ *
+ *  By using binary search, finds a value from this array which meets
+ *  the given condition in O(log n) where n is the size of the array.
+ *
+ *  You can use this method in two modes: a find-minimum mode and
+ *  a find-any mode.  In either case, the elements of the array must be
+ *  monotone (or sorted) with respect to the block.
+ *
+ *  In find-minimum mode (this is a good choice for typical use cases),
+ *  the block must always return true or false, and there must be an index i
+ *  (0 <= i <= ary.size) so that:
+ *
+ *  - the block returns false for any element whose index is less than
+ *    i, and
+ *  - the block returns true for any element whose index is greater
+ *    than or equal to i.
+ *
+ *  This method returns the i-th element.  If i is equal to ary.size,
+ *  it returns nil.
+ *
+ *     ary = [0, 4, 7, 10, 12]
+ *     ary.bsearch {|x| x >=   4 } #=> 4
+ *     ary.bsearch {|x| x >=   6 } #=> 7
+ *     ary.bsearch {|x| x >=  -1 } #=> 0
+ *     ary.bsearch {|x| x >= 100 } #=> nil
+ *
+ *  In find-any mode (this behaves like libc's bsearch(3)), the block
+ *  must always return a number, and there must be two indices i and j
+ *  (0 <= i <= j <= ary.size) so that:
+ *
+ *  - the block returns a positive number for ary[k] if 0 <= k < i,
+ *  - the block returns zero for ary[k] if i <= k < j, and
+ *  - the block returns a negative number for ary[k] if
+ *    j <= k < ary.size.
+ *
+ *  Under this condition, this method returns any element whose index
+ *  is within i...j.  If i is equal to j (i.e., there is no element
+ *  that satisfies the block), this method returns nil.
+ *
+ *     ary = [0, 4, 7, 10, 12]
+ *     # try to find v such that 4 <= v < 8
+ *     ary.bsearch {|x| 1 - x / 4 } #=> 4 or 7
+ *     # try to find v such that 8 <= v < 10
+ *     ary.bsearch {|x| 4 - x / 2 } #=> nil
+ *
+ *  You must not mix the two modes at a time; the block must always
+ *  return either true/false, or always return a number.  It is
+ *  undefined which value is actually picked up at each iteration.
+ */
+
 static VALUE
-sort_unlock(ary)
-    VALUE ary;
+rb_ary_bsearch(VALUE ary)
 {
-    FL_UNSET(ary, ARY_TMPLOCK);
-    return ary;
+    VALUE index_result = rb_ary_bsearch_index(ary);
+
+    if (FIXNUM_P(index_result)) {
+	return rb_ary_entry(ary, FIX2LONG(index_result));
+    }
+    return index_result;
 }
 
 /*
  *  call-seq:
- *     array.sort!                   -> array
- *     array.sort! {| a,b | block }  -> array 
- *  
- *  Sorts _self_. Comparisons for
- *  the sort will be done using the <code><=></code> operator or using
- *  an optional code block. The block implements a comparison between
- *  <i>a</i> and <i>b</i>, returning -1, 0, or +1. See also
- *  <code>Enumerable#sort_by</code>.
- *     
- *     a = [ "d", "a", "e", "c", "b" ]
- *     a.sort                    #=> ["a", "b", "c", "d", "e"]
- *     a.sort {|x,y| y <=> x }   #=> ["e", "d", "c", "b", "a"]
+ *     ary.bsearch_index {|x| block }  -> int or nil
+ *
+ *  By using binary search, finds an index of a value from this array which
+ *  meets the given condition in O(log n) where n is the size of the array.
+ *
+ *  It supports two modes, depending on the nature of the block. They are
+ *  exactly the same as in the case of the #bsearch method, with the only difference
+ *  being that this method returns the index of the element instead of the
+ *  element itself. For more details consult the documentation for #bsearch.
  */
 
-VALUE
-rb_ary_sort_bang(ary)
-    VALUE ary;
+static VALUE
+rb_ary_bsearch_index(VALUE ary)
 {
-    rb_ary_modify(ary);
-    if (RARRAY(ary)->len > 1) {
-	FL_SET(ary, ARY_TMPLOCK);	/* prohibit modification during sort */
-	rb_ensure(sort_internal, ary, sort_unlock, ary);
+    long low = 0, high = RARRAY_LEN(ary), mid;
+    int smaller = 0, satisfied = 0;
+    VALUE v, val;
+
+    RETURN_ENUMERATOR(ary, 0, 0);
+    while (low < high) {
+	mid = low + ((high - low) / 2);
+	val = rb_ary_entry(ary, mid);
+	v = rb_yield(val);
+	if (FIXNUM_P(v)) {
+	    if (v == INT2FIX(0)) return INT2FIX(mid);
+	    smaller = (SIGNED_VALUE)v < 0; /* Fixnum preserves its sign-bit */
+	}
+	else if (v == Qtrue) {
+	    satisfied = 1;
+	    smaller = 1;
+	}
+	else if (v == Qfalse || v == Qnil) {
+	    smaller = 0;
+	}
+	else if (rb_obj_is_kind_of(v, rb_cNumeric)) {
+	    const VALUE zero = INT2FIX(0);
+	    switch (rb_cmpint(rb_funcallv(v, id_cmp, 1, &zero), v, zero)) {
+	      case 0: return INT2FIX(mid);
+	      case 1: smaller = 1; break;
+	      case -1: smaller = 0;
+	    }
+	}
+	else {
+	    rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE
+		     " (must be numeric, true, false or nil)",
+		     rb_obj_class(v));
+	}
+	if (smaller) {
+	    high = mid;
+	}
+	else {
+	    low = mid + 1;
+	}
     }
-    return ary;
+    if (!satisfied) return Qnil;
+    return INT2FIX(low);
+}
+
+
+static VALUE
+sort_by_i(RB_BLOCK_CALL_FUNC_ARGLIST(i, dummy))
+{
+    return rb_yield(i);
 }
 
 /*
  *  call-seq:
- *     array.sort                   -> an_array 
- *     array.sort {| a,b | block }  -> an_array 
- *  
- *  Returns a new array created by sorting <i>self</i>. Comparisons for
- *  the sort will be done using the <code><=></code> operator or using
- *  an optional code block. The block implements a comparison between
- *  <i>a</i> and <i>b</i>, returning -1, 0, or +1. See also
- *  <code>Enumerable#sort_by</code>.
- *     
- *     a = [ "d", "a", "e", "c", "b" ]
- *     a.sort                    #=> ["a", "b", "c", "d", "e"]
- *     a.sort {|x,y| y <=> x }   #=> ["e", "d", "c", "b", "a"]
+ *     ary.sort_by! {|obj| block}      -> ary
+ *     ary.sort_by!                    -> Enumerator
+ *
+ *  Sorts +self+ in place using a set of keys generated by mapping the
+ *  values in +self+ through the given block.
+ *
+ *  The result is not guaranteed to be stable.  When two keys are equal,
+ *  the order of the corresponding elements is unpredictable.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *  See also Enumerable#sort_by.
  */
 
-VALUE
-rb_ary_sort(ary)
-    VALUE ary;
+static VALUE
+rb_ary_sort_by_bang(VALUE ary)
 {
-    ary = rb_ary_dup(ary);
-    rb_ary_sort_bang(ary);
+    VALUE sorted;
+
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    rb_ary_modify(ary);
+    sorted = rb_block_call(ary, rb_intern("sort_by"), 0, 0, sort_by_i, 0);
+    rb_ary_replace(ary, sorted);
     return ary;
 }
 
+
 /*
  *  call-seq:
- *     array.collect {|item| block }  -> an_array
- *     array.map     {|item| block }  -> an_array
- *  
- *  Invokes <i>block</i> once for each element of <i>self</i>. Creates a 
- *  new array containing the values returned by the block.
- *  See also <code>Enumerable#collect</code>.
- *     
+ *     ary.collect {|item| block}    -> new_ary
+ *     ary.map     {|item| block}    -> new_ary
+ *     ary.collect                   -> Enumerator
+ *     ary.map                       -> Enumerator
+ *
+ *  Invokes the given block once for each element of +self+.
+ *
+ *  Creates a new array containing the values returned by the block.
+ *
+ *  See also Enumerable#collect.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
  *     a = [ "a", "b", "c", "d" ]
- *     a.collect {|x| x + "!" }   #=> ["a!", "b!", "c!", "d!"]
- *     a                          #=> ["a", "b", "c", "d"]
+ *     a.collect {|x| x + "!"}           #=> ["a!", "b!", "c!", "d!"]
+ *     a.map.with_index {|x, i| x * i}   #=> ["", "b", "cc", "ddd"]
+ *     a                                 #=> ["a", "b", "c", "d"]
  */
 
 static VALUE
-rb_ary_collect(ary)
-    VALUE ary;
+rb_ary_collect(VALUE ary)
 {
     long i;
     VALUE collect;
 
-    if (!rb_block_given_p()) {
-	return rb_ary_new4(RARRAY(ary)->len, RARRAY(ary)->ptr);
-    }
-
-    collect = rb_ary_new2(RARRAY(ary)->len);
-    for (i = 0; i < RARRAY(ary)->len; i++) {
-	rb_ary_push(collect, rb_yield(RARRAY(ary)->ptr[i]));
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    collect = rb_ary_new2(RARRAY_LEN(ary));
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+        rb_ary_push(collect, rb_yield(RARRAY_AREF(ary, i)));
     }
     return collect;
 }
 
-/* 
+
+/*
  *  call-seq:
- *     array.collect! {|item| block }   ->   array
- *     array.map!     {|item| block }   ->   array
+ *     ary.collect! {|item| block }   -> ary
+ *     ary.map!     {|item| block }   -> ary
+ *     ary.collect!                   -> Enumerator
+ *     ary.map!                       -> Enumerator
+ *
+ *  Invokes the given block once for each element of +self+, replacing the
+ *  element with the value returned by the block.
+ *
+ *  See also Enumerable#collect.
+ *
+ *  If no block is given, an Enumerator is returned instead.
  *
- *  Invokes the block once for each element of _self_, replacing the
- *  element with the value returned by _block_.
- *  See also <code>Enumerable#collect</code>.
- *   
  *     a = [ "a", "b", "c", "d" ]
- *     a.collect! {|x| x + "!" }
- *     a             #=>  [ "a!", "b!", "c!", "d!" ]
+ *     a.map! {|x| x + "!" }
+ *     a #=>  [ "a!", "b!", "c!", "d!" ]
+ *     a.collect!.with_index {|x, i| x[0...i] }
+ *     a #=>  ["", "b", "c!", "d!"]
  */
 
 static VALUE
-rb_ary_collect_bang(ary)
-    VALUE ary;
+rb_ary_collect_bang(VALUE ary)
 {
     long i;
 
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
     rb_ary_modify(ary);
-    for (i = 0; i < RARRAY(ary)->len; i++) {
-	rb_ary_store(ary, i, rb_yield(RARRAY(ary)->ptr[i]));
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+	rb_ary_store(ary, i, rb_yield(RARRAY_AREF(ary, i)));
     }
     return ary;
 }
 
 VALUE
-rb_values_at(obj, olen, argc, argv, func)
-    VALUE obj;
-    long olen;
-    int argc;
-    VALUE *argv;
-    VALUE (*func) _((VALUE,long));
+rb_get_values_at(VALUE obj, long olen, int argc, const VALUE *argv, VALUE (*func) (VALUE, long))
 {
     VALUE result = rb_ary_new2(argc);
     long beg, len, i, j;
@@ -1750,15 +3059,13 @@ rb_values_at(obj, olen, argc, argv, func)
 	    continue;
 	}
 	/* check if idx is Range */
-	switch (rb_range_beg_len(argv[i], &beg, &len, olen, 0)) {
-	  case Qfalse:
-	    break;
-	  case Qnil:
-	    continue;
-	  default:
-	    for (j=0; j<len; j++) {
-		rb_ary_push(result, (*func)(obj, j+beg));
+	if (rb_range_beg_len(argv[i], &beg, &len, olen, 1)) {
+	    long end = olen < beg+len ? olen : beg+len;
+	    for (j = beg; j < end; j++) {
+		rb_ary_push(result, (*func)(obj, j));
 	    }
+	    if (beg + len > j)
+		rb_ary_resize(result, RARRAY_LEN(result) + (beg + len) - j);
 	    continue;
 	}
 	rb_ary_push(result, (*func)(obj, NUM2LONG(argv[i])));
@@ -1766,118 +3073,297 @@ rb_values_at(obj, olen, argc, argv, func)
     return result;
 }
 
-/* 
+static VALUE
+append_values_at_single(VALUE result, VALUE ary, long olen, VALUE idx)
+{
+    long beg, len;
+    if (FIXNUM_P(idx)) {
+	beg = FIX2LONG(idx);
+    }
+    /* check if idx is Range */
+    else if (rb_range_beg_len(idx, &beg, &len, olen, 1)) {
+	if (len > 0) {
+            const VALUE *const src = RARRAY_CONST_PTR_TRANSIENT(ary);
+	    const long end = beg + len;
+	    const long prevlen = RARRAY_LEN(result);
+	    if (beg < olen) {
+		rb_ary_cat(result, src + beg, end > olen ? olen-beg : len);
+	    }
+	    if (end > olen) {
+		rb_ary_store(result, prevlen + len - 1, Qnil);
+	    }
+	}
+	return result;
+    }
+    else {
+	beg = NUM2LONG(idx);
+    }
+    return rb_ary_push(result, rb_ary_entry(ary, beg));
+}
+
+/*
  *  call-seq:
- *     array.values_at(selector,... )  -> an_array
+ *     ary.values_at(selector, ...)  -> new_ary
+ *
+ *  Returns an array containing the elements in +self+ corresponding to the
+ *  given +selector+(s).
+ *
+ *  The selectors may be either integer indices or ranges.
+ *
+ *  See also Array#select.
  *
- *  Returns an array containing the elements in
- *  _self_ corresponding to the given selector(s). The selectors
- *  may be either integer indices or ranges. 
- *  See also <code>Array#select</code>.
- * 
  *     a = %w{ a b c d e f }
- *     a.values_at(1, 3, 5)
- *     a.values_at(1, 3, 5, 7)
- *     a.values_at(-1, -3, -5, -7)
- *     a.values_at(1..3, 2...5)
+ *     a.values_at(1, 3, 5)          # => ["b", "d", "f"]
+ *     a.values_at(1, 3, 5, 7)       # => ["b", "d", "f", nil]
+ *     a.values_at(-1, -2, -2, -7)   # => ["f", "e", "e", nil]
+ *     a.values_at(4..6, 3...6)      # => ["e", "f", nil, "d", "e", "f"]
  */
 
 static VALUE
-rb_ary_values_at(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_values_at(int argc, VALUE *argv, VALUE ary)
 {
-    return rb_values_at(ary, RARRAY(ary)->len, argc, argv, rb_ary_entry);
+    long i, olen = RARRAY_LEN(ary);
+    VALUE result = rb_ary_new_capa(argc);
+    for (i = 0; i < argc; ++i) {
+	append_values_at_single(result, ary, olen, argv[i]);
+    }
+    RB_GC_GUARD(ary);
+    return result;
 }
 
+
 /*
  *  call-seq:
- *     array.select {|item| block } -> an_array
- *  
- *  Invokes the block passing in successive elements from <i>array</i>,
- *  returning an array containing those elements for which the block
- *  returns a true value (equivalent to <code>Enumerable#select</code>).
- *     
- *     a = %w{ a b c d e f }
- *     a.select {|v| v =~ /[aeiou]/}   #=> ["a", "e"]
+ *     ary.select {|item| block}   -> new_ary
+ *     ary.select                  -> Enumerator
+ *     ary.filter {|item| block}   -> new_ary
+ *     ary.filter                  -> Enumerator
+ *
+ *  Returns a new array containing all elements of +ary+
+ *  for which the given +block+ returns a true value.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *     [1,2,3,4,5].select {|num| num.even? }     #=> [2, 4]
+ *
+ *     a = %w[ a b c d e f ]
+ *     a.select {|v| v =~ /[aeiou]/ }    #=> ["a", "e"]
+ *
+ *  See also Enumerable#select.
+ *
+ *  Array#filter is an alias for Array#select.
  */
 
 static VALUE
-rb_ary_select(ary)
-    VALUE ary;
+rb_ary_select(VALUE ary)
 {
     VALUE result;
     long i;
 
-    result = rb_ary_new2(RARRAY(ary)->len);
-    for (i = 0; i < RARRAY(ary)->len; i++) {
-	if (RTEST(rb_yield(RARRAY(ary)->ptr[i]))) {
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    result = rb_ary_new2(RARRAY_LEN(ary));
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+	if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) {
 	    rb_ary_push(result, rb_ary_elt(ary, i));
 	}
     }
     return result;
 }
 
+struct select_bang_arg {
+    VALUE ary;
+    long len[2];
+};
+
+static VALUE
+select_bang_i(VALUE a)
+{
+    volatile struct select_bang_arg *arg = (void *)a;
+    VALUE ary = arg->ary;
+    long i1, i2;
+
+    for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); arg->len[0] = ++i1) {
+	VALUE v = RARRAY_AREF(ary, i1);
+	if (!RTEST(rb_yield(v))) continue;
+	if (i1 != i2) {
+	    rb_ary_store(ary, i2, v);
+	}
+	arg->len[1] = ++i2;
+    }
+    return (i1 == i2) ? Qnil : ary;
+}
+
+static VALUE
+select_bang_ensure(VALUE a)
+{
+    volatile struct select_bang_arg *arg = (void *)a;
+    VALUE ary = arg->ary;
+    long len = RARRAY_LEN(ary);
+    long i1 = arg->len[0], i2 = arg->len[1];
+
+    if (i2 < len && i2 < i1) {
+	long tail = 0;
+	if (i1 < len) {
+	    tail = len - i1;
+            RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+		    MEMMOVE(ptr + i2, ptr + i1, VALUE, tail);
+		});
+	}
+	ARY_SET_LEN(ary, i2 + tail);
+    }
+    return ary;
+}
+
 /*
  *  call-seq:
- *     array.delete(obj)            -> obj or nil 
- *     array.delete(obj) { block }  -> obj or nil
- *  
- *  Deletes items from <i>self</i> that are equal to <i>obj</i>. If
- *  the item is not found, returns <code>nil</code>. If the optional
- *  code block is given, returns the result of <i>block</i> if the item
- *  is not found.
- *     
+ *     ary.select! {|item| block } -> ary or nil
+ *     ary.select!                 -> Enumerator
+ *     ary.filter! {|item| block } -> ary or nil
+ *     ary.filter!                 -> Enumerator
+ *
+ *  Invokes the given block passing in successive elements from +self+,
+ *  deleting elements for which the block returns a +false+ value.
+ *
+ *  The array may not be changed instantly every time the block is called.
+ *
+ *  If changes were made, it will return +self+, otherwise it returns +nil+.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *  See also Array#keep_if.
+ *
+ *  Array#filter! is an alias for Array#select!.
+ */
+
+static VALUE
+rb_ary_select_bang(VALUE ary)
+{
+    struct select_bang_arg args;
+
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    rb_ary_modify(ary);
+
+    args.ary = ary;
+    args.len[0] = args.len[1] = 0;
+    return rb_ensure(select_bang_i, (VALUE)&args, select_bang_ensure, (VALUE)&args);
+}
+
+/*
+ *  call-seq:
+ *     ary.keep_if {|item| block}   -> ary
+ *     ary.keep_if                  -> Enumerator
+ *
+ *  Deletes every element of +self+ for which the given block evaluates to
+ *  +false+, and returns +self+.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *     a = %w[ a b c d e f ]
+ *     a.keep_if {|v| v =~ /[aeiou]/ }    #=> ["a", "e"]
+ *     a                                  #=> ["a", "e"]
+ *
+ *  See also Array#select!.
+ */
+
+static VALUE
+rb_ary_keep_if(VALUE ary)
+{
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    rb_ary_select_bang(ary);
+    return ary;
+}
+
+static void
+ary_resize_smaller(VALUE ary, long len)
+{
+    rb_ary_modify(ary);
+    if (RARRAY_LEN(ary) > len) {
+	ARY_SET_LEN(ary, len);
+	if (len * 2 < ARY_CAPA(ary) &&
+	    ARY_CAPA(ary) > ARY_DEFAULT_SIZE) {
+	    ary_resize_capa(ary, len * 2);
+	}
+    }
+}
+
+/*
+ *  call-seq:
+ *     ary.delete(obj)            -> item or nil
+ *     ary.delete(obj) {block}    -> item or result of block
+ *
+ *  Deletes all items from +self+ that are equal to +obj+.
+ *
+ *  Returns the last deleted item, or +nil+ if no matching item is found.
+ *
+ *  If the optional code block is given, the result of the block is returned if
+ *  the item is not found.  (To remove +nil+ elements and get an informative
+ *  return value, use Array#compact!)
+ *
  *     a = [ "a", "b", "b", "b", "c" ]
  *     a.delete("b")                   #=> "b"
  *     a                               #=> ["a", "c"]
  *     a.delete("z")                   #=> nil
- *     a.delete("z") { "not found" }   #=> "not found"
+ *     a.delete("z") {"not found"}     #=> "not found"
  */
 
 VALUE
-rb_ary_delete(ary, item)
-    VALUE ary;
-    VALUE item;
+rb_ary_delete(VALUE ary, VALUE item)
 {
+    VALUE v = item;
     long i1, i2;
 
-    for (i1 = i2 = 0; i1 < RARRAY(ary)->len; i1++) {
-	VALUE e = RARRAY(ary)->ptr[i1];
+    for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); i1++) {
+	VALUE e = RARRAY_AREF(ary, i1);
 
-	if (rb_equal(e, item)) continue;
+	if (rb_equal(e, item)) {
+	    v = e;
+	    continue;
+	}
 	if (i1 != i2) {
 	    rb_ary_store(ary, i2, e);
 	}
 	i2++;
     }
-    if (RARRAY(ary)->len == i2) {
+    if (RARRAY_LEN(ary) == i2) {
 	if (rb_block_given_p()) {
 	    return rb_yield(item);
 	}
 	return Qnil;
     }
 
-    rb_ary_modify(ary);
-    if (RARRAY(ary)->len > i2) {
-	RARRAY(ary)->len = i2;
-	if (i2 * 2 < RARRAY(ary)->aux.capa &&
-	    RARRAY(ary)->aux.capa > ARY_DEFAULT_SIZE) {
-	    REALLOC_N(RARRAY(ary)->ptr, VALUE, i2 * 2);
-	    RARRAY(ary)->aux.capa = i2 * 2;
+    ary_resize_smaller(ary, i2);
+
+    ary_verify(ary);
+    return v;
+}
+
+void
+rb_ary_delete_same(VALUE ary, VALUE item)
+{
+    long i1, i2;
+
+    for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); i1++) {
+	VALUE e = RARRAY_AREF(ary, i1);
+
+	if (e == item) {
+	    continue;
+	}
+	if (i1 != i2) {
+	    rb_ary_store(ary, i2, e);
 	}
+	i2++;
+    }
+    if (RARRAY_LEN(ary) == i2) {
+	return;
     }
 
-    return item;
+    ary_resize_smaller(ary, i2);
 }
 
 VALUE
-rb_ary_delete_at(ary, pos)
-    VALUE ary;
-    long pos;
+rb_ary_delete_at(VALUE ary, long pos)
 {
-    long i, len = RARRAY(ary)->len;
+    long len = RARRAY_LEN(ary);
     VALUE del;
 
     if (pos >= len) return Qnil;
@@ -1887,52 +3373,48 @@ rb_ary_delete_at(ary, pos)
     }
 
     rb_ary_modify(ary);
-    del = RARRAY(ary)->ptr[pos];
-    for (i = pos + 1; i < len; i++, pos++) {
-	RARRAY(ary)->ptr[pos] = RARRAY(ary)->ptr[i];
-    }
-    RARRAY(ary)->len = pos;
-
+    del = RARRAY_AREF(ary, pos);
+    RARRAY_PTR_USE_TRANSIENT(ary, ptr, {
+        MEMMOVE(ptr+pos, ptr+pos+1, VALUE, len-pos-1);
+    });
+    ARY_INCREASE_LEN(ary, -1);
+    ary_verify(ary);
     return del;
 }
 
 /*
  *  call-seq:
- *     array.delete_at(index)  -> obj or nil
- *  
- *  Deletes the element at the specified index, returning that element,
- *  or <code>nil</code> if the index is out of range. See also
- *  <code>Array#slice!</code>.
- *     
- *     a = %w( ant bat cat dog )
+ *     ary.delete_at(index)  -> obj or nil
+ *
+ *  Deletes the element at the specified +index+, returning that element, or
+ *  +nil+ if the +index+ is out of range.
+ *
+ *  See also Array#slice!
+ *
+ *     a = ["ant", "bat", "cat", "dog"]
  *     a.delete_at(2)    #=> "cat"
  *     a                 #=> ["ant", "bat", "dog"]
  *     a.delete_at(99)   #=> nil
  */
 
 static VALUE
-rb_ary_delete_at_m(ary, pos)
-    VALUE ary, pos;
+rb_ary_delete_at_m(VALUE ary, VALUE pos)
 {
     return rb_ary_delete_at(ary, NUM2LONG(pos));
 }
 
 /*
  *  call-seq:
- *     array.slice!(index)         -> obj or nil
- *     array.slice!(start, length) -> sub_array or nil
- *     array.slice!(range)         -> sub_array or nil 
- *  
- *  Deletes the element(s) given by an index (optionally with a length)
- *  or by a range. Returns the deleted object, subarray, or
- *  <code>nil</code> if the index is out of range. Equivalent to:
- *     
- *     def slice!(*args)
- *       result = self[*args]
- *       self[*args] = nil
- *       result
- *     end
- *     
+ *     ary.slice!(index)         -> obj or nil
+ *     ary.slice!(start, length) -> new_ary or nil
+ *     ary.slice!(range)         -> new_ary or nil
+ *
+ *  Deletes the element(s) given by an +index+ (optionally up to +length+
+ *  elements) or by a +range+.
+ *
+ *  Returns the deleted object (or objects), or +nil+ if the +index+ is out of
+ *  range.
+ *
  *     a = [ "a", "b", "c" ]
  *     a.slice!(1)     #=> "b"
  *     a               #=> ["a", "c"]
@@ -1943,194 +3425,311 @@ rb_ary_delete_at_m(ary, pos)
  */
 
 static VALUE
-rb_ary_slice_bang(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_slice_bang(int argc, VALUE *argv, VALUE ary)
 {
     VALUE arg1, arg2;
-    long pos, len;
+    long pos, len, orig_len;
 
-    if (rb_scan_args(argc, argv, "11", &arg1, &arg2) == 2) {
-	pos = NUM2LONG(arg1);
-	len = NUM2LONG(arg2);
+    rb_ary_modify_check(ary);
+    if (argc == 2) {
+	pos = NUM2LONG(argv[0]);
+	len = NUM2LONG(argv[1]);
       delete_pos_len:
+	if (len < 0) return Qnil;
+	orig_len = RARRAY_LEN(ary);
 	if (pos < 0) {
-	    pos = RARRAY(ary)->len + pos;
+	    pos += orig_len;
+	    if (pos < 0) return Qnil;
 	}
-	arg2 = rb_ary_subseq(ary, pos, len);
-	rb_ary_splice(ary, pos, len, Qnil);	/* Qnil/rb_ary_new2(0) */
+	else if (orig_len < pos) return Qnil;
+	if (orig_len < pos + len) {
+	    len = orig_len - pos;
+	}
+	if (len == 0) return rb_ary_new2(0);
+        arg2 = rb_ary_new4(len, RARRAY_CONST_PTR_TRANSIENT(ary)+pos);
+	RBASIC_SET_CLASS(arg2, rb_obj_class(ary));
+	rb_ary_splice(ary, pos, len, 0, 0);
 	return arg2;
     }
 
-    if (!FIXNUM_P(arg1) && rb_range_beg_len(arg1, &pos, &len, RARRAY(ary)->len, 1)) {
-	goto delete_pos_len;
+    rb_check_arity(argc, 1, 2);
+    arg1 = argv[0];
+
+    if (!FIXNUM_P(arg1)) {
+	switch (rb_range_beg_len(arg1, &pos, &len, RARRAY_LEN(ary), 0)) {
+	  case Qtrue:
+	    /* valid range */
+	    goto delete_pos_len;
+	  case Qnil:
+	    /* invalid range */
+	    return Qnil;
+	  default:
+	    /* not a range */
+	    break;
+	}
     }
 
     return rb_ary_delete_at(ary, NUM2LONG(arg1));
 }
 
-/*
- *  call-seq:
- *     array.reject! {|item| block }  -> array or nil
- *  
- *  Equivalent to <code>Array#delete_if</code>, deleting elements from
- *  _self_ for which the block evaluates to true, but returns
- *  <code>nil</code> if no changes were made. Also see
- *  <code>Enumerable#reject</code>.
- */
+static VALUE
+ary_reject(VALUE orig, VALUE result)
+{
+    long i;
+
+    for (i = 0; i < RARRAY_LEN(orig); i++) {
+	VALUE v = RARRAY_AREF(orig, i);
+
+        if (!RTEST(rb_yield(v))) {
+	    rb_ary_push(result, v);
+	}
+    }
+    return result;
+}
 
 static VALUE
-rb_ary_reject_bang(ary)
-    VALUE ary;
+reject_bang_i(VALUE a)
 {
+    volatile struct select_bang_arg *arg = (void *)a;
+    VALUE ary = arg->ary;
     long i1, i2;
 
-    rb_ary_modify(ary);
-    for (i1 = i2 = 0; i1 < RARRAY(ary)->len; i1++) {
-	VALUE v = RARRAY(ary)->ptr[i1];
+    for (i1 = i2 = 0; i1 < RARRAY_LEN(ary); arg->len[0] = ++i1) {
+	VALUE v = RARRAY_AREF(ary, i1);
 	if (RTEST(rb_yield(v))) continue;
 	if (i1 != i2) {
 	    rb_ary_store(ary, i2, v);
 	}
-	i2++;
+	arg->len[1] = ++i2;
     }
-    if (RARRAY(ary)->len == i2) return Qnil;
-    if (i2 < RARRAY(ary)->len)
-	RARRAY(ary)->len = i2;
+    return (i1 == i2) ? Qnil : ary;
+}
 
-    return ary;
+static VALUE
+ary_reject_bang(VALUE ary)
+{
+    struct select_bang_arg args;
+    rb_ary_modify_check(ary);
+    args.ary = ary;
+    args.len[0] = args.len[1] = 0;
+    return rb_ensure(reject_bang_i, (VALUE)&args, select_bang_ensure, (VALUE)&args);
 }
 
 /*
  *  call-seq:
- *     array.reject {|item| block }  -> an_array
- *  
- *  Returns a new array containing the items in _self_
- *  for which the block is not true.
+ *     ary.reject! {|item| block}    -> ary or nil
+ *     ary.reject!                   -> Enumerator
+ *
+ *  Deletes every element of +self+ for which the block evaluates to +true+,
+ *  if no changes were made returns +nil+.
+ *
+ *  The array may not be changed instantly every time the block is called.
+ *
+ *  See also Enumerable#reject and Array#delete_if.
+ *
+ *  If no block is given, an Enumerator is returned instead.
  */
 
 static VALUE
-rb_ary_reject(ary)
-    VALUE ary;
+rb_ary_reject_bang(VALUE ary)
 {
-    ary = rb_ary_dup(ary);
-    rb_ary_reject_bang(ary);
-    return ary;
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    return ary_reject_bang(ary);
 }
 
 /*
  *  call-seq:
- *     array.delete_if {|item| block }  -> array
- *  
- *  Deletes every element of <i>self</i> for which <i>block</i> evaluates
- *  to <code>true</code>.
- *     
- *     a = [ "a", "b", "c" ]
- *     a.delete_if {|x| x >= "b" }   #=> ["a"]
+ *     ary.reject  {|item| block }  -> new_ary
+ *     ary.reject                   -> Enumerator
+ *
+ *  Returns a new array containing the items in +self+ for which the given
+ *  block is not +true+. The ordering of non-rejected elements is maintained.
+ *
+ *  See also Array#delete_if
+ *
+ *  If no block is given, an Enumerator is returned instead.
  */
 
 static VALUE
-rb_ary_delete_if(ary)
-    VALUE ary;
+rb_ary_reject(VALUE ary)
+{
+    VALUE rejected_ary;
+
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    rejected_ary = rb_ary_new();
+    ary_reject(ary, rejected_ary);
+    return rejected_ary;
+}
+
+/*
+ *  call-seq:
+ *     ary.delete_if {|item| block}    -> ary
+ *     ary.delete_if                   -> Enumerator
+ *
+ *  Deletes every element of +self+ for which block evaluates to +true+.
+ *
+ *  The array is changed instantly every time the block is called, not after
+ *  the iteration is over.
+ *
+ *  See also Array#reject!
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *     scores = [ 97, 42, 75 ]
+ *     scores.delete_if {|score| score < 80 }   #=> [97]
+ */
+
+static VALUE
+rb_ary_delete_if(VALUE ary)
 {
-    rb_ary_reject_bang(ary);
+    ary_verify(ary);
+    RETURN_SIZED_ENUMERATOR(ary, 0, 0, ary_enum_length);
+    ary_reject_bang(ary);
     return ary;
 }
 
+static VALUE
+take_i(RB_BLOCK_CALL_FUNC_ARGLIST(val, cbarg))
+{
+    VALUE *args = (VALUE *)cbarg;
+    if (args[1] == 0) rb_iter_break();
+    else args[1]--;
+    if (argc > 1) val = rb_ary_new4(argc, argv);
+    rb_ary_push(args[0], val);
+    return Qnil;
+}
+
+static VALUE
+take_items(VALUE obj, long n)
+{
+    VALUE result = rb_check_array_type(obj);
+    VALUE args[2];
+
+    if (!NIL_P(result)) return rb_ary_subseq(result, 0, n);
+    result = rb_ary_new2(n);
+    args[0] = result; args[1] = (VALUE)n;
+    if (rb_check_block_call(obj, idEach, 0, 0, take_i, (VALUE)args) == Qundef)
+	rb_raise(rb_eTypeError, "wrong argument type %"PRIsVALUE" (must respond to :each)",
+		 rb_obj_class(obj));
+    return result;
+}
+
+
 /*
  *  call-seq:
- *     array.zip(arg, ...)                   -> an_array
- *     array.zip(arg, ...) {| arr | block }  -> nil
- *  
- *  Converts any arguments to arrays, then merges elements of
- *  <i>self</i> with corresponding elements from each argument. This
- *  generates a sequence of <code>self.size</code> <em>n</em>-element
- *  arrays, where <em>n</em> is one more that the count of arguments. If
- *  the size of any argument is less than <code>enumObj.size</code>,
- *  <code>nil</code> values are supplied. If a block given, it is
- *  invoked for each output array, otherwise an array of arrays is
- *  returned.
- *     
+ *     ary.zip(arg, ...)                  -> new_ary
+ *     ary.zip(arg, ...) {|arr| block}    -> nil
+ *
+ *  Converts any arguments to arrays, then merges elements of +self+ with
+ *  corresponding elements from each argument.
+ *
+ *  This generates a sequence of <code>ary.size</code> _n_-element arrays,
+ *  where _n_ is one more than the count of arguments.
+ *
+ *  If the size of any argument is less than the size of the initial array,
+ *  +nil+ values are supplied.
+ *
+ *  If a block is given, it is invoked for each output +array+, otherwise an
+ *  array of arrays is returned.
+ *
  *     a = [ 4, 5, 6 ]
  *     b = [ 7, 8, 9 ]
- *     
- *     [1,2,3].zip(a, b)      #=> [[1, 4, 7], [2, 5, 8], [3, 6, 9]]
- *     [1,2].zip(a,b)         #=> [[1, 4, 7], [2, 5, 8]]
- *     a.zip([1,2],[8])       #=> [[4,1,8], [5,2,nil], [6,nil,nil]]
+ *     [1, 2, 3].zip(a, b)   #=> [[1, 4, 7], [2, 5, 8], [3, 6, 9]]
+ *     [1, 2].zip(a, b)      #=> [[1, 4, 7], [2, 5, 8]]
+ *     a.zip([1, 2], [8])    #=> [[4, 1, 8], [5, 2, nil], [6, nil, nil]]
  */
 
 static VALUE
-rb_ary_zip(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_zip(int argc, VALUE *argv, VALUE ary)
 {
     int i, j;
-    long len;
-    VALUE result;
+    long len = RARRAY_LEN(ary);
+    VALUE result = Qnil;
 
     for (i=0; i<argc; i++) {
-	argv[i] = to_ary(argv[i]);
+	argv[i] = take_items(argv[i], len);
     }
+
     if (rb_block_given_p()) {
-	for (i=0; i<RARRAY(ary)->len; i++) {
-	    VALUE tmp = rb_ary_new2(argc+1);
+	int arity = rb_block_arity();
 
-	    rb_ary_push(tmp, rb_ary_elt(ary, i));
-	    for (j=0; j<argc; j++) {
-		rb_ary_push(tmp, rb_ary_elt(argv[j], i));
+	if (arity > 1) {
+	    VALUE work, *tmp;
+
+	    tmp = ALLOCV_N(VALUE, work, argc+1);
+
+	    for (i=0; i<RARRAY_LEN(ary); i++) {
+		tmp[0] = RARRAY_AREF(ary, i);
+		for (j=0; j<argc; j++) {
+		    tmp[j+1] = rb_ary_elt(argv[j], i);
+		}
+		rb_yield_values2(argc+1, tmp);
+	    }
+
+	    if (work) ALLOCV_END(work);
+	}
+	else {
+	    for (i=0; i<RARRAY_LEN(ary); i++) {
+		VALUE tmp = rb_ary_new2(argc+1);
+
+		rb_ary_push(tmp, RARRAY_AREF(ary, i));
+		for (j=0; j<argc; j++) {
+		    rb_ary_push(tmp, rb_ary_elt(argv[j], i));
+		}
+		rb_yield(tmp);
 	    }
-	    rb_yield(tmp);
 	}
-	return Qnil;
     }
-    len = RARRAY(ary)->len;
-    result = rb_ary_new2(len);
-    for (i=0; i<len; i++) {
-	VALUE tmp = rb_ary_new2(argc+1);
+    else {
+	result = rb_ary_new_capa(len);
+
+	for (i=0; i<len; i++) {
+	    VALUE tmp = rb_ary_new_capa(argc+1);
 
-	rb_ary_push(tmp, rb_ary_elt(ary, i));
-	for (j=0; j<argc; j++) {
-	    rb_ary_push(tmp, rb_ary_elt(argv[j], i));
+	    rb_ary_push(tmp, RARRAY_AREF(ary, i));
+	    for (j=0; j<argc; j++) {
+		rb_ary_push(tmp, rb_ary_elt(argv[j], i));
+	    }
+	    rb_ary_push(result, tmp);
 	}
-	rb_ary_push(result, tmp);
     }
+
     return result;
 }
 
 /*
  *  call-seq:
- *     array.transpose -> an_array
- *  
- *  Assumes that <i>self</i> is an array of arrays and transposes the
- *  rows and columns.
- *     
+ *     ary.transpose -> new_ary
+ *
+ *  Assumes that +self+ is an array of arrays and transposes the rows and
+ *  columns.
+ *
  *     a = [[1,2], [3,4], [5,6]]
  *     a.transpose   #=> [[1, 3, 5], [2, 4, 6]]
+ *
+ *  If the length of the subarrays don't match, an IndexError is raised.
  */
 
 static VALUE
-rb_ary_transpose(ary)
-    VALUE ary;
+rb_ary_transpose(VALUE ary)
 {
     long elen = -1, alen, i, j;
     VALUE tmp, result = 0;
 
-    alen = RARRAY(ary)->len;
+    alen = RARRAY_LEN(ary);
     if (alen == 0) return rb_ary_dup(ary);
     for (i=0; i<alen; i++) {
 	tmp = to_ary(rb_ary_elt(ary, i));
 	if (elen < 0) {		/* first element */
-	    elen = RARRAY(tmp)->len;
+	    elen = RARRAY_LEN(tmp);
 	    result = rb_ary_new2(elen);
 	    for (j=0; j<elen; j++) {
 		rb_ary_store(result, j, rb_ary_new2(alen));
 	    }
 	}
-	else if (elen != RARRAY(tmp)->len) {
-	    rb_raise(rb_eIndexError, "element size differs (%d should be %d)",
-		     RARRAY(tmp)->len, elen);
+	else if (elen != RARRAY_LEN(tmp)) {
+	    rb_raise(rb_eIndexError, "element size differs (%ld should be %ld)",
+		     RARRAY_LEN(tmp), elen);
 	}
 	for (j=0; j<elen; j++) {
 	    rb_ary_store(rb_ary_elt(result, j), i, rb_ary_elt(tmp, j));
@@ -2141,75 +3740,111 @@ rb_ary_transpose(ary)
 
 /*
  *  call-seq:
- *     array.replace(other_array)  -> array
- *  
- *  Replaces the contents of <i>self</i> with the contents of
- *  <i>other_array</i>, truncating or expanding if necessary.
- *     
+ *     ary.replace(other_ary)  -> ary
+ *     ary.initialize_copy(other_ary)	-> ary
+ *
+ *  Replaces the contents of +self+ with the contents of +other_ary+,
+ *  truncating or expanding if necessary.
+ *
  *     a = [ "a", "b", "c", "d", "e" ]
  *     a.replace([ "x", "y", "z" ])   #=> ["x", "y", "z"]
  *     a                              #=> ["x", "y", "z"]
  */
 
-static VALUE
-rb_ary_replace(copy, orig)
-    VALUE copy, orig;
+VALUE
+rb_ary_replace(VALUE copy, VALUE orig)
 {
-    VALUE shared;
-
-    rb_ary_modify(copy);
+    rb_ary_modify_check(copy);
     orig = to_ary(orig);
     if (copy == orig) return copy;
-    shared = ary_make_shared(orig);
-    if (RARRAY(copy)->ptr && !FL_TEST(copy, ELTS_SHARED))
-	free(RARRAY(copy)->ptr);
-    RARRAY(copy)->ptr = RARRAY(orig)->ptr;
-    RARRAY(copy)->len = RARRAY(orig)->len;
-    RARRAY(copy)->aux.shared = shared;
-    FL_SET(copy, ELTS_SHARED);
 
+    if (RARRAY_LEN(orig) <= RARRAY_EMBED_LEN_MAX) {
+        VALUE shared = 0;
+
+        if (ARY_OWNS_HEAP_P(copy)) {
+            ary_heap_free(copy);
+	}
+        else if (ARY_SHARED_P(copy)) {
+            shared = ARY_SHARED(copy);
+            FL_UNSET_SHARED(copy);
+        }
+        FL_SET_EMBED(copy);
+        ary_memcpy(copy, 0, RARRAY_LEN(orig), RARRAY_CONST_PTR_TRANSIENT(orig));
+        if (shared) {
+            rb_ary_decrement_share(shared);
+        }
+        ARY_SET_LEN(copy, RARRAY_LEN(orig));
+    }
+    else {
+        VALUE shared = ary_make_shared(orig);
+        if (ARY_OWNS_HEAP_P(copy)) {
+            ary_heap_free(copy);
+        }
+        else {
+            rb_ary_unshare_safe(copy);
+        }
+        FL_UNSET_EMBED(copy);
+        ARY_SET_PTR(copy, ARY_HEAP_PTR(orig));
+        ARY_SET_LEN(copy, ARY_HEAP_LEN(orig));
+        rb_ary_set_shared(copy, shared);
+    }
+    ary_verify(copy);
     return copy;
 }
 
-/* 
+/*
  *  call-seq:
- *     array.clear    ->  array
+ *     ary.clear    -> ary
  *
- *  Removes all elements from _self_.
+ *  Removes all elements from +self+.
  *
  *     a = [ "a", "b", "c", "d", "e" ]
  *     a.clear    #=> [ ]
  */
 
 VALUE
-rb_ary_clear(ary)
-    VALUE ary;
+rb_ary_clear(VALUE ary)
 {
-    rb_ary_modify(ary);
-    RARRAY(ary)->len = 0;
-    if (ARY_DEFAULT_SIZE * 2 < RARRAY(ary)->aux.capa) {
-	REALLOC_N(RARRAY(ary)->ptr, VALUE, ARY_DEFAULT_SIZE * 2);
-	RARRAY(ary)->aux.capa = ARY_DEFAULT_SIZE * 2;
+    rb_ary_modify_check(ary);
+    if (ARY_SHARED_P(ary)) {
+	if (!ARY_EMBED_P(ary)) {
+	    rb_ary_unshare(ary);
+	    FL_SET_EMBED(ary);
+            ARY_SET_EMBED_LEN(ary, 0);
+	}
+    }
+    else {
+        ARY_SET_LEN(ary, 0);
+        if (ARY_DEFAULT_SIZE * 2 < ARY_CAPA(ary)) {
+            ary_resize_capa(ary, ARY_DEFAULT_SIZE * 2);
+        }
     }
+    ary_verify(ary);
     return ary;
 }
 
 /*
  *  call-seq:
- *     array.fill(obj)                                -> array
- *     array.fill(obj, start [, length])              -> array
- *     array.fill(obj, range )                        -> array
- *     array.fill {|index| block }                    -> array
- *     array.fill(start [, length] ) {|index| block } -> array
- *     array.fill(range) {|index| block }             -> array
- *  
- *  The first three forms set the selected elements of <i>self</i> (which
- *  may be the entire array) to <i>obj</i>. A <i>start</i> of
- *  <code>nil</code> is equivalent to zero. A <i>length</i> of
- *  <code>nil</code> is equivalent to <i>self.length</i>. The last three
- *  forms fill the array with the value of the block. The block is
- *  passed the absolute index of each element to be filled.
- *     
+ *     ary.fill(obj)                                 -> ary
+ *     ary.fill(obj, start [, length])               -> ary
+ *     ary.fill(obj, range)                          -> ary
+ *     ary.fill {|index| block}                      -> ary
+ *     ary.fill(start [, length]) {|index| block}    -> ary
+ *     ary.fill(range) {|index| block}               -> ary
+ *
+ *  The first three forms set the selected elements of +self+ (which
+ *  may be the entire array) to +obj+.
+ *
+ *  A +start+ of +nil+ is equivalent to zero.
+ *
+ *  A +length+ of +nil+ is equivalent to the length of the array.
+ *
+ *  The last three forms fill the array with the value of the given block,
+ *  which is passed the absolute index of each element to be filled.
+ *
+ *  Negative values of +start+ count from the end of the array, where +-1+ is
+ *  the last element.
+ *
  *     a = [ "a", "b", "c", "d" ]
  *     a.fill("x")              #=> ["x", "x", "x", "x"]
  *     a.fill("z", 2, 2)        #=> ["x", "x", "z", "z"]
@@ -2219,18 +3854,12 @@ rb_ary_clear(ary)
  */
 
 static VALUE
-rb_ary_fill(argc, argv, ary)
-    int argc;
-    VALUE *argv;
-    VALUE ary;
+rb_ary_fill(int argc, VALUE *argv, VALUE ary)
 {
-    VALUE item, arg1, arg2;
+    VALUE item = Qundef, arg1, arg2;
     long beg = 0, end = 0, len = 0;
-    VALUE *p, *pend;
-    int block_p = Qfalse;
 
     if (rb_block_given_p()) {
-	block_p = Qtrue;
 	rb_scan_args(argc, argv, "02", &arg1, &arg2);
 	argc += 1;		/* hackish */
     }
@@ -2240,112 +3869,162 @@ rb_ary_fill(argc, argv, ary)
     switch (argc) {
       case 1:
 	beg = 0;
-	len = RARRAY(ary)->len;
+	len = RARRAY_LEN(ary);
 	break;
       case 2:
-	if (rb_range_beg_len(arg1, &beg, &len, RARRAY(ary)->len, 1)) {
+	if (rb_range_beg_len(arg1, &beg, &len, RARRAY_LEN(ary), 1)) {
 	    break;
 	}
 	/* fall through */
       case 3:
 	beg = NIL_P(arg1) ? 0 : NUM2LONG(arg1);
 	if (beg < 0) {
-	    beg = RARRAY(ary)->len + beg;
+	    beg = RARRAY_LEN(ary) + beg;
 	    if (beg < 0) beg = 0;
 	}
-	len = NIL_P(arg2) ? RARRAY(ary)->len - beg : NUM2LONG(arg2);
+	len = NIL_P(arg2) ? RARRAY_LEN(ary) - beg : NUM2LONG(arg2);
 	break;
     }
     rb_ary_modify(ary);
-    end = beg + len;
-    if (end < 0) {
+    if (len < 0) {
+        return ary;
+    }
+    if (beg >= ARY_MAX_SIZE || len > ARY_MAX_SIZE - beg) {
 	rb_raise(rb_eArgError, "argument too big");
     }
-    if (end > RARRAY(ary)->len) {
-	if (end >= RARRAY(ary)->aux.capa) {
-	    REALLOC_N(RARRAY(ary)->ptr, VALUE, end);
-	    RARRAY(ary)->aux.capa = end;
+    end = beg + len;
+    if (RARRAY_LEN(ary) < end) {
+	if (end >= ARY_CAPA(ary)) {
+	    ary_resize_capa(ary, end);
 	}
-	rb_mem_clear(RARRAY(ary)->ptr + RARRAY(ary)->len, end - RARRAY(ary)->len);
-	RARRAY(ary)->len = end;
+	ary_mem_clear(ary, RARRAY_LEN(ary), end - RARRAY_LEN(ary));
+	ARY_SET_LEN(ary, end);
     }
 
-    if (block_p) {
+    if (item == Qundef) {
 	VALUE v;
 	long i;
 
 	for (i=beg; i<end; i++) {
 	    v = rb_yield(LONG2NUM(i));
-	    if (i>=RARRAY(ary)->len) break;
-	    RARRAY(ary)->ptr[i] = v;
+	    if (i>=RARRAY_LEN(ary)) break;
+	    ARY_SET(ary, i, v);
 	}
     }
     else {
-	p = RARRAY(ary)->ptr + beg;
-	pend = p + len;
-	while (p < pend) {
-	    *p++ = item;
-	}
+	ary_memfill(ary, beg, len, item);
     }
     return ary;
 }
 
-/* 
+/*
  *  call-seq:
- *     array + other_array   -> an_array
+ *     ary + other_ary   -> new_ary
  *
- *  Concatenation---Returns a new array built by concatenating the
+ *  Concatenation --- Returns a new array built by concatenating the
  *  two arrays together to produce a third array.
- * 
+ *
  *     [ 1, 2, 3 ] + [ 4, 5 ]    #=> [ 1, 2, 3, 4, 5 ]
+ *     a = [ "a", "b", "c" ]
+ *     c = a + [ "d", "e", "f" ]
+ *     c                         #=> [ "a", "b", "c", "d", "e", "f" ]
+ *     a                         #=> [ "a", "b", "c" ]
+ *
+ *  Note that
+ *     x += y
+ *  is the same as
+ *     x = x + y
+ *  This means that it produces a new array. As a consequence,
+ *  repeated use of <code>+=</code> on arrays can be quite inefficient.
+ *
+ *  See also Array#concat.
  */
 
 VALUE
-rb_ary_plus(x, y)
-    VALUE x, y;
+rb_ary_plus(VALUE x, VALUE y)
 {
     VALUE z;
-    long len;
+    long len, xlen, ylen;
 
     y = to_ary(y);
-    len = RARRAY(x)->len + RARRAY(y)->len;
+    xlen = RARRAY_LEN(x);
+    ylen = RARRAY_LEN(y);
+    len = xlen + ylen;
     z = rb_ary_new2(len);
-    MEMCPY(RARRAY(z)->ptr, RARRAY(x)->ptr, VALUE, RARRAY(x)->len);
-    MEMCPY(RARRAY(z)->ptr + RARRAY(x)->len, RARRAY(y)->ptr, VALUE, RARRAY(y)->len);
-    RARRAY(z)->len = len;
+
+    ary_memcpy(z, 0, xlen, RARRAY_CONST_PTR_TRANSIENT(x));
+    ary_memcpy(z, xlen, ylen, RARRAY_CONST_PTR_TRANSIENT(y));
+    ARY_SET_LEN(z, len);
     return z;
 }
 
-/* 
+static VALUE
+ary_append(VALUE x, VALUE y)
+{
+    long n = RARRAY_LEN(y);
+    if (n > 0) {
+        rb_ary_splice(x, RARRAY_LEN(x), 0, RARRAY_CONST_PTR_TRANSIENT(y), n);
+    }
+    return x;
+}
+
+/*
  *  call-seq:
- *     array.concat(other_array)   ->  array
+ *     ary.concat(other_ary1, other_ary2, ...)   -> ary
+ *
+ *  Appends the elements of <code>other_ary</code>s to +self+.
+ *
+ *     [ "a", "b" ].concat( ["c", "d"])   #=> [ "a", "b", "c", "d" ]
+ *     [ "a" ].concat( ["b"], ["c", "d"]) #=> [ "a", "b", "c", "d" ]
+ *     [ "a" ].concat #=> [ "a" ]
  *
- *  Appends the elements in other_array to _self_.
- *  
- *     [ "a", "b" ].concat( ["c", "d"] ) #=> [ "a", "b", "c", "d" ]
+ *     a = [ 1, 2, 3 ]
+ *     a.concat( [ 4, 5 ])
+ *     a                                 #=> [ 1, 2, 3, 4, 5 ]
+ *
+ *     a = [ 1, 2 ]
+ *     a.concat(a, a)                    #=> [1, 2, 1, 2, 1, 2]
+ *
+ *  See also Array#+.
  */
 
-
-VALUE
-rb_ary_concat(x, y)
-    VALUE x, y;
+static VALUE
+rb_ary_concat_multi(int argc, VALUE *argv, VALUE ary)
 {
-    y = to_ary(y);
-    if (RARRAY(y)->len > 0) {
-	rb_ary_splice(x, RARRAY(x)->len, 0, y);
+    rb_ary_modify_check(ary);
+
+    if (argc == 1) {
+	rb_ary_concat(ary, argv[0]);
     }
-    return x;
+    else if (argc > 1) {
+	int i;
+	VALUE args = rb_ary_tmp_new(argc);
+	for (i = 0; i < argc; i++) {
+	    rb_ary_concat(args, argv[i]);
+	}
+	ary_append(ary, args);
+    }
+
+    ary_verify(ary);
+    return ary;
 }
 
+VALUE
+rb_ary_concat(VALUE x, VALUE y)
+{
+    return ary_append(x, to_ary(y));
+}
 
-/* 
+/*
  *  call-seq:
- *     array * int     ->    an_array
- *     array * str     ->    a_string
+ *     ary * int     -> new_ary
+ *     ary * str     -> new_string
+ *
+ *  Repetition --- With a String argument, equivalent to
+ *  <code>ary.join(str)</code>.
  *
- *  Repetition---With a String argument, equivalent to
- *  self.join(str). Otherwise, returns a new array
- *  built by concatenating the _int_ copies of _self_.
+ *  Otherwise, returns a new array built by concatenating the +int+ copies of
+ *  +self+.
  *
  *
  *     [ 1, 2, 3 ] * 3    #=> [ 1, 2, 3, 1, 2, 3, 1, 2, 3 ]
@@ -2354,11 +4033,11 @@ rb_ary_concat(x, y)
  */
 
 static VALUE
-rb_ary_times(ary, times)
-    VALUE ary, times;
+rb_ary_times(VALUE ary, VALUE times)
 {
     VALUE ary2, tmp;
-    long i, len;
+    const VALUE *ptr;
+    long t, len;
 
     tmp = rb_check_string_type(times);
     if (!NIL_P(tmp)) {
@@ -2366,37 +4045,50 @@ rb_ary_times(ary, times)
     }
 
     len = NUM2LONG(times);
-    if (len == 0) return ary_new(rb_obj_class(ary), 0);
+    if (len == 0) {
+	ary2 = ary_new(rb_obj_class(ary), 0);
+	goto out;
+    }
     if (len < 0) {
 	rb_raise(rb_eArgError, "negative argument");
     }
-    if (LONG_MAX/len < RARRAY(ary)->len) {
+    if (ARY_MAX_SIZE/len < RARRAY_LEN(ary)) {
 	rb_raise(rb_eArgError, "argument too big");
     }
-    len *= RARRAY(ary)->len;
+    len *= RARRAY_LEN(ary);
 
     ary2 = ary_new(rb_obj_class(ary), len);
-    RARRAY(ary2)->len = len;
-
-    for (i=0; i<len; i+=RARRAY(ary)->len) {
-	MEMCPY(RARRAY(ary2)->ptr+i, RARRAY(ary)->ptr, VALUE, RARRAY(ary)->len);
-    }
+    ARY_SET_LEN(ary2, len);
+
+    ptr = RARRAY_CONST_PTR_TRANSIENT(ary);
+    t = RARRAY_LEN(ary);
+    if (0 < t) {
+	ary_memcpy(ary2, 0, t, ptr);
+	while (t <= len/2) {
+            ary_memcpy(ary2, t, t, RARRAY_CONST_PTR_TRANSIENT(ary2));
+            t *= 2;
+        }
+        if (t < len) {
+            ary_memcpy(ary2, t, len-t, RARRAY_CONST_PTR_TRANSIENT(ary2));
+        }
+    }
+  out:
     OBJ_INFECT(ary2, ary);
 
     return ary2;
 }
 
-/* 
+/*
  *  call-seq:
- *     array.assoc(obj)   ->  an_array  or  nil
+ *     ary.assoc(obj)   -> element_ary  or  nil
+ *
+ *  Searches through an array whose elements are also arrays comparing +obj+
+ *  with the first element of each contained array using <code>obj.==</code>.
  *
- *  Searches through an array whose elements are also arrays
- *  comparing _obj_ with the first element of each contained array
- *  using obj.==.
- *  Returns the first contained array that matches (that
- *  is, the first associated array),
- *  or +nil+ if no match is found.
- *  See also <code>Array#rassoc</code>.
+ *  Returns the first contained array that matches (that is, the first
+ *  associated array), or +nil+ if no match is found.
+ *
+ *  See also Array#rassoc
  *
  *     s1 = [ "colors", "red", "blue", "green" ]
  *     s2 = [ "letters", "a", "b", "c" ]
@@ -2407,17 +4099,15 @@ rb_ary_times(ary, times)
  */
 
 VALUE
-rb_ary_assoc(ary, key)
-    VALUE ary, key;
+rb_ary_assoc(VALUE ary, VALUE key)
 {
     long i;
     VALUE v;
 
-    for (i = 0; i < RARRAY(ary)->len; ++i) {
-	v = RARRAY(ary)->ptr[i];
-	if (TYPE(v) == T_ARRAY &&
-	    RARRAY(v)->len > 0 &&
-	    rb_equal(RARRAY(v)->ptr[0], key))
+    for (i = 0; i < RARRAY_LEN(ary); ++i) {
+	v = rb_check_array_type(RARRAY_AREF(ary, i));
+	if (!NIL_P(v) && RARRAY_LEN(v) > 0 &&
+	    rb_equal(RARRAY_AREF(v, 0), key))
 	    return v;
     }
     return Qnil;
@@ -2425,42 +4115,79 @@ rb_ary_assoc(ary, key)
 
 /*
  *  call-seq:
- *     array.rassoc(key) -> an_array or nil
- *  
- *  Searches through the array whose elements are also arrays. Compares
- *  <em>key</em> with the second element of each contained array using
- *  <code>==</code>. Returns the first contained array that matches. See
- *  also <code>Array#assoc</code>.
- *     
+ *     ary.rassoc(obj) -> element_ary or nil
+ *
+ *  Searches through the array whose elements are also arrays.
+ *
+ *  Compares +obj+ with the second element of each contained array using
+ *  <code>obj.==</code>.
+ *
+ *  Returns the first contained array that matches +obj+.
+ *
+ *  See also Array#assoc.
+ *
  *     a = [ [ 1, "one"], [2, "two"], [3, "three"], ["ii", "two"] ]
  *     a.rassoc("two")    #=> [2, "two"]
  *     a.rassoc("four")   #=> nil
  */
 
 VALUE
-rb_ary_rassoc(ary, value)
-    VALUE ary, value;
+rb_ary_rassoc(VALUE ary, VALUE value)
 {
     long i;
     VALUE v;
 
-    for (i = 0; i < RARRAY(ary)->len; ++i) {
-	v = RARRAY(ary)->ptr[i];
-	if (TYPE(v) == T_ARRAY &&
-	    RARRAY(v)->len > 1 &&
-	    rb_equal(RARRAY(v)->ptr[1], value))
+    for (i = 0; i < RARRAY_LEN(ary); ++i) {
+	v = RARRAY_AREF(ary, i);
+	if (RB_TYPE_P(v, T_ARRAY) &&
+	    RARRAY_LEN(v) > 1 &&
+	    rb_equal(RARRAY_AREF(v, 1), value))
 	    return v;
     }
     return Qnil;
 }
 
-/* 
+static VALUE
+recursive_equal(VALUE ary1, VALUE ary2, int recur)
+{
+    long i, len1;
+    const VALUE *p1, *p2;
+
+    if (recur) return Qtrue; /* Subtle! */
+
+    /* rb_equal() can evacuate ptrs */
+    p1 = RARRAY_CONST_PTR(ary1);
+    p2 = RARRAY_CONST_PTR(ary2);
+    len1 = RARRAY_LEN(ary1);
+
+    for (i = 0; i < len1; i++) {
+	if (*p1 != *p2) {
+	    if (rb_equal(*p1, *p2)) {
+		len1 = RARRAY_LEN(ary1);
+		if (len1 != RARRAY_LEN(ary2))
+		    return Qfalse;
+		if (len1 < i)
+		    return Qtrue;
+                p1 = RARRAY_CONST_PTR(ary1) + i;
+                p2 = RARRAY_CONST_PTR(ary2) + i;
+	    }
+	    else {
+		return Qfalse;
+	    }
+	}
+	p1++;
+	p2++;
+    }
+    return Qtrue;
+}
+
+/*
  *  call-seq:
- *     array == other_array   ->   bool
+ *     ary == other_ary   ->   bool
  *
- *  Equality---Two arrays are equal if they contain the same number
- *  of elements and if each element is equal to (according to
- *  Object.==) the corresponding element in the other array.
+ *  Equality --- Two arrays are equal if they contain the same number of
+ *  elements and if each element is equal to (according to Object#==) the
+ *  corresponding element in +other_ary+.
  *
  *     [ "a", "c" ]    == [ "a", "c", 7 ]     #=> false
  *     [ "a", "c", 7 ] == [ "a", "c", 7 ]     #=> true
@@ -2469,21 +4196,28 @@ rb_ary_rassoc(ary, value)
  */
 
 static VALUE
-rb_ary_equal(ary1, ary2)
-    VALUE ary1, ary2;
+rb_ary_equal(VALUE ary1, VALUE ary2)
 {
-    long i;
-
     if (ary1 == ary2) return Qtrue;
-    if (TYPE(ary2) != T_ARRAY) {
-	if (!rb_respond_to(ary2, rb_intern("to_ary"))) {
+    if (!RB_TYPE_P(ary2, T_ARRAY)) {
+	if (!rb_respond_to(ary2, idTo_ary)) {
 	    return Qfalse;
 	}
 	return rb_equal(ary2, ary1);
     }
-    if (RARRAY(ary1)->len != RARRAY(ary2)->len) return Qfalse;
-    for (i=0; i<RARRAY(ary1)->len; i++) {
-	if (!rb_equal(rb_ary_elt(ary1, i), rb_ary_elt(ary2, i)))
+    if (RARRAY_LEN(ary1) != RARRAY_LEN(ary2)) return Qfalse;
+    if (RARRAY_CONST_PTR_TRANSIENT(ary1) == RARRAY_CONST_PTR_TRANSIENT(ary2)) return Qtrue;
+    return rb_exec_recursive_paired(recursive_equal, ary1, ary2, ary2);
+}
+
+static VALUE
+recursive_eql(VALUE ary1, VALUE ary2, int recur)
+{
+    long i;
+
+    if (recur) return Qtrue; /* Subtle! */
+    for (i=0; i<RARRAY_LEN(ary1); i++) {
+	if (!rb_eql(rb_ary_elt(ary1, i), rb_ary_elt(ary2, i)))
 	    return Qfalse;
     }
     return Qtrue;
@@ -2491,346 +4225,747 @@ rb_ary_equal(ary1, ary2)
 
 /*
  *  call-seq:
- *     array.eql?(other)  -> true or false
+ *     ary.eql?(other)  -> true or false
  *
- *  Returns <code>true</code> if _array_ and _other_ are the same object,
- *  or are both arrays with the same content.
+ *  Returns +true+ if +self+ and +other+ are the same object,
+ *  or are both arrays with the same content (according to Object#eql?).
  */
 
 static VALUE
-rb_ary_eql(ary1, ary2)
-    VALUE ary1, ary2;
+rb_ary_eql(VALUE ary1, VALUE ary2)
 {
-    long i;
-
     if (ary1 == ary2) return Qtrue;
-    if (TYPE(ary2) != T_ARRAY) return Qfalse;
-    if (RARRAY(ary1)->len != RARRAY(ary2)->len) return Qfalse;
-    for (i=0; i<RARRAY(ary1)->len; i++) {
-	if (!rb_eql(rb_ary_elt(ary1, i), rb_ary_elt(ary2, i)))
-	    return Qfalse;
-    }
-    return Qtrue;
+    if (!RB_TYPE_P(ary2, T_ARRAY)) return Qfalse;
+    if (RARRAY_LEN(ary1) != RARRAY_LEN(ary2)) return Qfalse;
+    if (RARRAY_CONST_PTR_TRANSIENT(ary1) == RARRAY_CONST_PTR_TRANSIENT(ary2)) return Qtrue;
+    return rb_exec_recursive_paired(recursive_eql, ary1, ary2, ary2);
 }
 
 /*
  *  call-seq:
- *     array.hash   -> fixnum
+ *     ary.hash   -> integer
+ *
+ *  Compute a hash-code for this array.
  *
- *  Compute a hash-code for this array. Two arrays with the same content
- *  will have the same hash code (and will compare using <code>eql?</code>).
+ *  Two arrays with the same content will have the same hash code (and will
+ *  compare using #eql?).
+ *
+ *  See also Object#hash.
  */
 
 static VALUE
-rb_ary_hash(ary)
-    VALUE ary;
+rb_ary_hash(VALUE ary)
 {
-    long i, h;
+    long i;
+    st_index_t h;
     VALUE n;
 
-    h = RARRAY(ary)->len;
-    for (i=0; i<RARRAY(ary)->len; i++) {
-	h = (h << 1) | (h<0 ? 1 : 0);
-	n = rb_hash(RARRAY(ary)->ptr[i]);
-	h ^= NUM2LONG(n);
+    h = rb_hash_start(RARRAY_LEN(ary));
+    h = rb_hash_uint(h, (st_index_t)rb_ary_hash);
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	n = rb_hash(RARRAY_AREF(ary, i));
+	h = rb_hash_uint(h, NUM2LONG(n));
     }
-    return LONG2FIX(h);
+    h = rb_hash_end(h);
+    return ST2FIX(h);
 }
 
 /*
  *  call-seq:
- *     array.include?(obj)   -> true or false
- *  
- *  Returns <code>true</code> if the given object is present in
- *  <i>self</i> (that is, if any object <code>==</code> <i>anObject</i>),
- *  <code>false</code> otherwise.
- *     
+ *     ary.include?(object)   -> true or false
+ *
+ *  Returns +true+ if the given +object+ is present in +self+ (that is, if any
+ *  element <code>==</code> +object+), otherwise returns +false+.
+ *
  *     a = [ "a", "b", "c" ]
  *     a.include?("b")   #=> true
  *     a.include?("z")   #=> false
  */
 
 VALUE
-rb_ary_includes(ary, item)
-    VALUE ary;
-    VALUE item;
+rb_ary_includes(VALUE ary, VALUE item)
 {
     long i;
-    
-    for (i=0; i<RARRAY(ary)->len; i++) {
-	if (rb_equal(RARRAY(ary)->ptr[i], item)) {
+    VALUE e;
+
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	e = RARRAY_AREF(ary, i);
+	if (rb_equal(e, item)) {
 	    return Qtrue;
 	}
     }
     return Qfalse;
 }
 
+static VALUE
+rb_ary_includes_by_eql(VALUE ary, VALUE item)
+{
+    long i;
+    VALUE e;
 
-/* 
- *  call-seq:
- *     array <=> other_array   ->  -1, 0, +1
- *
- *  Comparison---Returns an integer (-1, 0,
- *  or +1) if this array is less than, equal to, or greater than
- *  other_array.  Each object in each array is compared
- *  (using <=>). If any value isn't
- *  equal, then that inequality is the return value. If all the
- *  values found are equal, then the return is based on a
- *  comparison of the array lengths.  Thus, two arrays are
- *  ``equal'' according to <code>Array#<=></code> if and only if they have
- *  the same length and the value of each element is equal to the
- *  value of the corresponding element in the other array.
- *  
- *     [ "a", "a", "c" ]    <=> [ "a", "b", "c" ]   #=> -1
- *     [ 1, 2, 3, 4, 5, 6 ] <=> [ 1, 2 ]            #=> +1
- *
- */
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	e = RARRAY_AREF(ary, i);
+	if (rb_eql(item, e)) {
+	    return Qtrue;
+	}
+    }
+    return Qfalse;
+}
 
-VALUE
-rb_ary_cmp(ary1, ary2)
-    VALUE ary1, ary2;
+static VALUE
+recursive_cmp(VALUE ary1, VALUE ary2, int recur)
 {
     long i, len;
 
-    ary2 = to_ary(ary2);
-    len = RARRAY(ary1)->len;
-    if (len > RARRAY(ary2)->len) {
-	len = RARRAY(ary2)->len;
+    if (recur) return Qundef;	/* Subtle! */
+    len = RARRAY_LEN(ary1);
+    if (len > RARRAY_LEN(ary2)) {
+	len = RARRAY_LEN(ary2);
     }
     for (i=0; i<len; i++) {
-	VALUE v = rb_funcall(rb_ary_elt(ary1, i), id_cmp, 1, rb_ary_elt(ary2, i));
+	VALUE e1 = rb_ary_elt(ary1, i), e2 = rb_ary_elt(ary2, i);
+	VALUE v = rb_funcallv(e1, id_cmp, 1, &e2);
 	if (v != INT2FIX(0)) {
 	    return v;
 	}
     }
-    len = RARRAY(ary1)->len - RARRAY(ary2)->len;
+    return Qundef;
+}
+
+/*
+ *  call-seq:
+ *     ary <=> other_ary   ->  -1, 0, +1 or nil
+ *
+ *  Comparison --- Returns an integer (+-1+, +0+, or <code>+1</code>) if this
+ *  array is less than, equal to, or greater than +other_ary+.
+ *
+ *  Each object in each array is compared (using the <=> operator).
+ *
+ *  Arrays are compared in an "element-wise" manner; the first element of +ary+
+ *  is compared with the first one of +other_ary+ using the <=> operator, then
+ *  each of the second elements, etc...
+ *  As soon as the result of any such comparison is non zero (i.e. the two
+ *  corresponding elements are not equal), that result is returned for the
+ *  whole array comparison.
+ *
+ *  If all the elements are equal, then the result is based on a comparison of
+ *  the array lengths. Thus, two arrays are "equal" according to Array#<=> if,
+ *  and only if, they have the same length and the value of each element is
+ *  equal to the value of the corresponding element in the other array.
+ *
+ *  +nil+ is returned if the +other_ary+ is not an array or if the comparison
+ *  of two elements returned +nil+.
+ *
+ *     [ "a", "a", "c" ]    <=> [ "a", "b", "c" ]   #=> -1
+ *     [ 1, 2, 3, 4, 5, 6 ] <=> [ 1, 2 ]            #=> +1
+ *     [ 1, 2 ]             <=> [ 1, :two ]         #=> nil
+ *
+ */
+
+VALUE
+rb_ary_cmp(VALUE ary1, VALUE ary2)
+{
+    long len;
+    VALUE v;
+
+    ary2 = rb_check_array_type(ary2);
+    if (NIL_P(ary2)) return Qnil;
+    if (ary1 == ary2) return INT2FIX(0);
+    v = rb_exec_recursive_paired(recursive_cmp, ary1, ary2, ary2);
+    if (v != Qundef) return v;
+    len = RARRAY_LEN(ary1) - RARRAY_LEN(ary2);
     if (len == 0) return INT2FIX(0);
     if (len > 0) return INT2FIX(1);
     return INT2FIX(-1);
 }
 
 static VALUE
-ary_make_hash(ary1, ary2)
-    VALUE ary1, ary2;
+ary_add_hash(VALUE hash, VALUE ary)
 {
-    VALUE hash = rb_hash_new();
     long i;
 
-    for (i=0; i<RARRAY(ary1)->len; i++) {
-	rb_hash_aset(hash, RARRAY(ary1)->ptr[i], Qtrue);
+    for (i=0; i<RARRAY_LEN(ary); i++) {
+	VALUE elt = RARRAY_AREF(ary, i);
+	rb_hash_add_new_element(hash, elt, elt);
     }
-    if (ary2) {
-	for (i=0; i<RARRAY(ary2)->len; i++) {
-	    rb_hash_aset(hash, RARRAY(ary2)->ptr[i], Qtrue);
-	}
+    return hash;
+}
+
+static inline VALUE
+ary_tmp_hash_new(VALUE ary)
+{
+    long size = RARRAY_LEN(ary);
+    VALUE hash = rb_hash_new_with_size(size);
+
+    RBASIC_CLEAR_CLASS(hash);
+    return hash;
+}
+
+static VALUE
+ary_make_hash(VALUE ary)
+{
+    VALUE hash = ary_tmp_hash_new(ary);
+    return ary_add_hash(hash, ary);
+}
+
+static VALUE
+ary_add_hash_by(VALUE hash, VALUE ary)
+{
+    long i;
+
+    for (i = 0; i < RARRAY_LEN(ary); ++i) {
+	VALUE v = rb_ary_elt(ary, i), k = rb_yield(v);
+	rb_hash_add_new_element(hash, k, v);
     }
     return hash;
 }
 
-/* 
+static VALUE
+ary_make_hash_by(VALUE ary)
+{
+    VALUE hash = ary_tmp_hash_new(ary);
+    return ary_add_hash_by(hash, ary);
+}
+
+static inline void
+ary_recycle_hash(VALUE hash)
+{
+    assert(RBASIC_CLASS(hash) == 0);
+    if (RHASH_ST_TABLE_P(hash)) {
+        st_table *tbl = RHASH_ST_TABLE(hash);
+	st_free_table(tbl);
+        RHASH_ST_CLEAR(hash);
+    }
+}
+
+/*
  *  call-seq:
- *     array - other_array    -> an_array
+ *     ary - other_ary    -> new_ary
+ *
+ *  Array Difference
  *
- *  Array Difference---Returns a new array that is a copy of
- *  the original array, removing any items that also appear in
- *  other_array. (If you need set-like behavior, see the
- *  library class Set.)
+ *  Returns a new array that is a copy of the original array, removing any
+ *  items that also appear in +other_ary+. The order is preserved from the
+ *  original array.
+ *
+ *  It compares elements using their #hash and #eql? methods for efficiency.
  *
  *     [ 1, 1, 2, 2, 3, 3, 4, 5 ] - [ 1, 2, 4 ]  #=>  [ 3, 3, 5 ]
+ *
+ *  If you need set-like behavior, see the library class Set.
+ *
+ *  See also Array#difference.
  */
 
 static VALUE
-rb_ary_diff(ary1, ary2)
-    VALUE ary1, ary2;
+rb_ary_diff(VALUE ary1, VALUE ary2)
 {
     VALUE ary3;
-    volatile VALUE hash;
+    VALUE hash;
     long i;
 
-    hash = ary_make_hash(to_ary(ary2), 0);
+    ary2 = to_ary(ary2);
     ary3 = rb_ary_new();
 
-    for (i=0; i<RARRAY(ary1)->len; i++) {
-	if (st_lookup(RHASH(hash)->tbl, RARRAY(ary1)->ptr[i], 0)) continue;
+    if (RARRAY_LEN(ary1) <= SMALL_ARRAY_LEN || RARRAY_LEN(ary2) <= SMALL_ARRAY_LEN) {
+	for (i=0; i<RARRAY_LEN(ary1); i++) {
+	    VALUE elt = rb_ary_elt(ary1, i);
+	    if (rb_ary_includes_by_eql(ary2, elt)) continue;
+	    rb_ary_push(ary3, elt);
+	}
+	return ary3;
+    }
+
+    hash = ary_make_hash(ary2);
+    for (i=0; i<RARRAY_LEN(ary1); i++) {
+        if (rb_hash_stlike_lookup(hash, RARRAY_AREF(ary1, i), NULL)) continue;
 	rb_ary_push(ary3, rb_ary_elt(ary1, i));
     }
+    ary_recycle_hash(hash);
     return ary3;
 }
 
-/* 
+/*
  *  call-seq:
- *     array & other_array
+ *     ary.difference(other_ary1, other_ary2, ...)   -> ary
+ *
+ *  Array Difference
+ *
+ *  Returns a new array that is a copy of the receiver, removing any items
+ *  that also appear in any of the arrays given as arguments.
+ *  The order is preserved from the original array.
+ *
+ *  It compares elements using their #hash and #eql? methods for efficiency.
+ *
+ *     [ 1, 1, 2, 2, 3, 3, 4, 5 ].difference([ 1, 2, 4 ])     #=> [ 3, 3, 5 ]
+ *     [ 1, 'c', :s, 'yep' ].difference([ 1 ], [ 'a', 'c' ])  #=> [ :s, "yep" ]
  *
- *  Set Intersection---Returns a new array
- *  containing elements common to the two arrays, with no duplicates.
+ *  If you need set-like behavior, see the library class Set.
  *
- *     [ 1, 1, 3, 5 ] & [ 1, 2, 3 ]   #=> [ 1, 3 ]
+ *  See also Array#-.
+ */
+
+static VALUE
+rb_ary_difference_multi(int argc, VALUE *argv, VALUE ary)
+{
+    VALUE ary_diff;
+    long i, length;
+    volatile VALUE t0;
+    bool *is_hash = ALLOCV_N(bool, t0, argc);
+    ary_diff = rb_ary_new();
+    length = RARRAY_LEN(ary);
+
+    for (i = 0; i < argc; i++) {
+        argv[i] = to_ary(argv[i]);
+        is_hash[i] = (length > SMALL_ARRAY_LEN && RARRAY_LEN(argv[i]) > SMALL_ARRAY_LEN);
+        if (is_hash[i]) argv[i] = ary_make_hash(argv[i]);
+    }
+
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+        int j;
+        VALUE elt = rb_ary_elt(ary, i);
+        for (j = 0; j < argc; j++){
+            if (is_hash[j]) {
+                if (rb_hash_stlike_lookup(argv[j], RARRAY_AREF(ary, i), NULL))
+                    break;
+            }
+            else {
+                if (rb_ary_includes_by_eql(argv[j], elt)) break;
+            }
+        }
+        if (j == argc) rb_ary_push(ary_diff, elt);
+    }
+
+    ALLOCV_END(t0);
+
+    return ary_diff;
+}
+
+
+/*
+ *  call-seq:
+ *     ary & other_ary      -> new_ary
+ *
+ *  Set Intersection --- Returns a new array containing unique elements common to the
+ *  two arrays. The order is preserved from the original array.
+ *
+ *  It compares elements using their #hash and #eql? methods for efficiency.
+ *
+ *     [ 1, 1, 3, 5 ] & [ 3, 2, 1 ]                 #=> [ 1, 3 ]
+ *     [ 'a', 'b', 'b', 'z' ] & [ 'a', 'b', 'c' ]   #=> [ 'a', 'b' ]
+ *
+ *  See also Array#uniq.
  */
 
 
 static VALUE
-rb_ary_and(ary1, ary2)
-    VALUE ary1, ary2;
+rb_ary_and(VALUE ary1, VALUE ary2)
 {
-    VALUE hash, ary3, v, vv;
+    VALUE hash, ary3, v;
+    st_data_t vv;
     long i;
 
     ary2 = to_ary(ary2);
-    ary3 = rb_ary_new2(RARRAY(ary1)->len < RARRAY(ary2)->len ?
-	    RARRAY(ary1)->len : RARRAY(ary2)->len);
-    hash = ary_make_hash(ary2, 0);
+    ary3 = rb_ary_new();
+    if (RARRAY_LEN(ary1) == 0 || RARRAY_LEN(ary2) == 0) return ary3;
+
+    if (RARRAY_LEN(ary1) <= SMALL_ARRAY_LEN && RARRAY_LEN(ary2) <= SMALL_ARRAY_LEN) {
+	for (i=0; i<RARRAY_LEN(ary1); i++) {
+	    v = RARRAY_AREF(ary1, i);
+	    if (!rb_ary_includes_by_eql(ary2, v)) continue;
+	    if (rb_ary_includes_by_eql(ary3, v)) continue;
+	    rb_ary_push(ary3, v);
+	}
+	return ary3;
+    }
 
-    for (i=0; i<RARRAY(ary1)->len; i++) {
-	v = vv = rb_ary_elt(ary1, i);
-	if (st_delete(RHASH(hash)->tbl, (st_data_t*)&vv, 0)) {
+    hash = ary_make_hash(ary2);
+
+    for (i=0; i<RARRAY_LEN(ary1); i++) {
+	v = RARRAY_AREF(ary1, i);
+	vv = (st_data_t)v;
+        if (rb_hash_stlike_delete(hash, &vv, 0)) {
 	    rb_ary_push(ary3, v);
 	}
     }
+    ary_recycle_hash(hash);
 
     return ary3;
 }
 
-/* 
+static int
+ary_hash_orset(st_data_t *key, st_data_t *value, st_data_t arg, int existing)
+{
+    if (existing) return ST_STOP;
+    *key = *value = (VALUE)arg;
+    return ST_CONTINUE;
+}
+
+static void
+rb_ary_union(VALUE ary_union, VALUE ary)
+{
+    long i;
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+        VALUE elt = rb_ary_elt(ary, i);
+        if (rb_ary_includes_by_eql(ary_union, elt)) continue;
+        rb_ary_push(ary_union, elt);
+    }
+}
+
+static void
+rb_ary_union_hash(VALUE hash, VALUE ary2)
+{
+    long i;
+    for (i = 0; i < RARRAY_LEN(ary2); i++) {
+        VALUE elt = RARRAY_AREF(ary2, i);
+        if (!rb_hash_stlike_update(hash, (st_data_t)elt, ary_hash_orset, (st_data_t)elt)) {
+            RB_OBJ_WRITTEN(hash, Qundef, elt);
+        }
+    }
+}
+
+/*
  *  call-seq:
- *     array | other_array     ->  an_array
+ *     ary | other_ary     -> new_ary
  *
- *  Set Union---Returns a new array by joining this array with
- *  other_array, removing duplicates.
+ *  Set Union --- Returns a new array by joining +ary+ with +other_ary+,
+ *  excluding any duplicates and preserving the order from the given arrays.
  *
- *     [ "a", "b", "c" ] | [ "c", "d", "a" ]
- *            #=> [ "a", "b", "c", "d" ]
+ *  It compares elements using their #hash and #eql? methods for efficiency.
+ *
+ *     [ "a", "b", "c" ] | [ "c", "d", "a" ]    #=> [ "a", "b", "c", "d" ]
+ *     [ "c", "d", "a" ] | [ "a", "b", "c" ]    #=> [ "c", "d", "a", "b" ]
+ *
+ *  See also Array#union.
  */
 
 static VALUE
-rb_ary_or(ary1, ary2)
-    VALUE ary1, ary2;
+rb_ary_or(VALUE ary1, VALUE ary2)
 {
     VALUE hash, ary3;
-    VALUE v, vv;
-    long i;
 
     ary2 = to_ary(ary2);
-    ary3 = rb_ary_new2(RARRAY(ary1)->len+RARRAY(ary2)->len);
-    hash = ary_make_hash(ary1, ary2);
+    if (RARRAY_LEN(ary1) + RARRAY_LEN(ary2) <= SMALL_ARRAY_LEN) {
+	ary3 = rb_ary_new();
+        rb_ary_union(ary3, ary1);
+        rb_ary_union(ary3, ary2);
+	return ary3;
+    }
 
-    for (i=0; i<RARRAY(ary1)->len; i++) {
-	v = vv = rb_ary_elt(ary1, i);
-	if (st_delete(RHASH(hash)->tbl, (st_data_t*)&vv, 0)) {
-	    rb_ary_push(ary3, v);
+    hash = ary_make_hash(ary1);
+    rb_ary_union_hash(hash, ary2);
+
+    ary3 = rb_hash_values(hash);
+    ary_recycle_hash(hash);
+    return ary3;
+}
+
+/*
+ *  call-seq:
+ *     ary.union(other_ary1, other_ary2, ...)   -> ary
+ *
+ *  Set Union --- Returns a new array by joining <code>other_ary</code>s with +self+,
+ *  excluding any duplicates and preserving the order from the given arrays.
+ *
+ *  It compares elements using their #hash and #eql? methods for efficiency.
+ *
+ *     [ "a", "b", "c" ].union( [ "c", "d", "a" ] )    #=> [ "a", "b", "c", "d" ]
+ *     [ "a" ].union( ["e", "b"], ["a", "c", "b"] )    #=> [ "a", "e", "b", "c" ]
+ *     [ "a" ].union #=> [ "a" ]
+ *
+ *  See also Array#|.
+ */
+
+static VALUE
+rb_ary_union_multi(int argc, VALUE *argv, VALUE ary)
+{
+    int i;
+    long sum;
+    VALUE hash, ary_union;
+
+    sum = RARRAY_LEN(ary);
+    for (i = 0; i < argc; i++){
+        argv[i] = to_ary(argv[i]);
+        sum += RARRAY_LEN(argv[i]);
+    }
+
+    if (sum <= SMALL_ARRAY_LEN) {
+        ary_union = rb_ary_new();
+
+        rb_ary_union(ary_union, ary);
+        for (i = 0; i < argc; i++) rb_ary_union(ary_union, argv[i]);
+
+        return ary_union;
+    }
+
+    hash = ary_make_hash(ary);
+    for (i = 0; i < argc; i++) rb_ary_union_hash(hash, argv[i]);
+
+    ary_union = rb_hash_values(hash);
+    ary_recycle_hash(hash);
+    return ary_union;
+}
+
+/*
+ *  call-seq:
+ *     ary.max                     -> obj
+ *     ary.max {|a, b| block}      -> obj
+ *     ary.max(n)                  -> array
+ *     ary.max(n) {|a, b| block}   -> array
+ *
+ *  Returns the object in _ary_ with the maximum value. The
+ *  first form assumes all objects implement <code>Comparable</code>;
+ *  the second uses the block to return <em>a <=> b</em>.
+ *
+ *     ary = %w(albatross dog horse)
+ *     ary.max                                   #=> "horse"
+ *     ary.max {|a, b| a.length <=> b.length}    #=> "albatross"
+ *
+ *  If the +n+ argument is given, maximum +n+ elements are returned
+ *  as an array.
+ *
+ *     ary = %w[albatross dog horse]
+ *     ary.max(2)                                  #=> ["horse", "dog"]
+ *     ary.max(2) {|a, b| a.length <=> b.length }  #=> ["albatross", "horse"]
+ */
+static VALUE
+rb_ary_max(int argc, VALUE *argv, VALUE ary)
+{
+    struct cmp_opt_data cmp_opt = { 0, 0 };
+    VALUE result = Qundef, v;
+    VALUE num;
+    long i;
+
+    if (rb_check_arity(argc, 0, 1) && !NIL_P(num = argv[0]))
+       return rb_nmin_run(ary, num, 0, 1, 1);
+
+    if (rb_block_given_p()) {
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	   v = RARRAY_AREF(ary, i);
+	   if (result == Qundef || rb_cmpint(rb_yield_values(2, v, result), v, result) > 0) {
+	       result = v;
+	   }
 	}
     }
-    for (i=0; i<RARRAY(ary2)->len; i++) {
-	v = vv = rb_ary_elt(ary2, i);
-	if (st_delete(RHASH(hash)->tbl, (st_data_t*)&vv, 0)) {
-	    rb_ary_push(ary3, v);
+    else {
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	   v = RARRAY_AREF(ary, i);
+	   if (result == Qundef || OPTIMIZED_CMP(v, result, cmp_opt) > 0) {
+	       result = v;
+	   }
 	}
     }
-    return ary3;
+    if (result == Qundef) return Qnil;
+    return result;
 }
 
 /*
  *  call-seq:
- *     array.uniq! -> array or nil
- *  
- *  Removes duplicate elements from _self_.
- *  Returns <code>nil</code> if no changes are made (that is, no
- *  duplicates are found).
- *     
+ *     ary.min                     -> obj
+ *     ary.min {| a,b | block }    -> obj
+ *     ary.min(n)                  -> array
+ *     ary.min(n) {| a,b | block } -> array
+ *
+ *  Returns the object in _ary_ with the minimum value. The
+ *  first form assumes all objects implement <code>Comparable</code>;
+ *  the second uses the block to return <em>a <=> b</em>.
+ *
+ *     ary = %w(albatross dog horse)
+ *     ary.min                                   #=> "albatross"
+ *     ary.min {|a, b| a.length <=> b.length}    #=> "dog"
+ *
+ *  If the +n+ argument is given, minimum +n+ elements are returned
+ *  as an array.
+ *
+ *     ary = %w[albatross dog horse]
+ *     ary.min(2)                                  #=> ["albatross", "dog"]
+ *     ary.min(2) {|a, b| a.length <=> b.length }  #=> ["dog", "horse"]
+ */
+static VALUE
+rb_ary_min(int argc, VALUE *argv, VALUE ary)
+{
+    struct cmp_opt_data cmp_opt = { 0, 0 };
+    VALUE result = Qundef, v;
+    VALUE num;
+    long i;
+
+    if (rb_check_arity(argc, 0, 1) && !NIL_P(num = argv[0]))
+       return rb_nmin_run(ary, num, 0, 0, 1);
+
+    if (rb_block_given_p()) {
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	   v = RARRAY_AREF(ary, i);
+	   if (result == Qundef || rb_cmpint(rb_yield_values(2, v, result), v, result) < 0) {
+	       result = v;
+	   }
+	}
+    }
+    else {
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	   v = RARRAY_AREF(ary, i);
+	   if (result == Qundef || OPTIMIZED_CMP(v, result, cmp_opt) < 0) {
+	       result = v;
+	   }
+	}
+    }
+    if (result == Qundef) return Qnil;
+    return result;
+}
+
+static int
+push_value(st_data_t key, st_data_t val, st_data_t ary)
+{
+    rb_ary_push((VALUE)ary, (VALUE)val);
+    return ST_CONTINUE;
+}
+
+/*
+ *  call-seq:
+ *     ary.uniq!                -> ary or nil
+ *     ary.uniq! {|item| ...}   -> ary or nil
+ *
+ *  Removes duplicate elements from +self+.
+ *
+ *  If a block is given, it will use the return value of the block for
+ *  comparison.
+ *
+ *  It compares values using their #hash and #eql? methods for efficiency.
+ *
+ *  +self+ is traversed in order, and the first occurrence is kept.
+ *
+ *  Returns +nil+ if no changes are made (that is, no duplicates are found).
+ *
  *     a = [ "a", "a", "b", "b", "c" ]
- *     a.uniq!   #=> ["a", "b", "c"]
+ *     a.uniq!   # => ["a", "b", "c"]
+ *
  *     b = [ "a", "b", "c" ]
- *     b.uniq!   #=> nil
+ *     b.uniq!   # => nil
+ *
+ *     c = [["student","sam"], ["student","george"], ["teacher","matz"]]
+ *     c.uniq! {|s| s.first}   # => [["student", "sam"], ["teacher", "matz"]]
+ *
  */
 
 static VALUE
-rb_ary_uniq_bang(ary)
-    VALUE ary;
+rb_ary_uniq_bang(VALUE ary)
 {
-    VALUE hash, v, vv;
-    long i, j;
-
-    hash = ary_make_hash(ary, 0);
+    VALUE hash;
+    long hash_size;
 
-    if (RARRAY(ary)->len == RHASH(hash)->tbl->num_entries) {
+    rb_ary_modify_check(ary);
+    if (RARRAY_LEN(ary) <= 1)
+        return Qnil;
+    if (rb_block_given_p())
+	hash = ary_make_hash_by(ary);
+    else
+	hash = ary_make_hash(ary);
+
+    hash_size = RHASH_SIZE(hash);
+    if (RARRAY_LEN(ary) == hash_size) {
 	return Qnil;
     }
-    for (i=j=0; i<RARRAY(ary)->len; i++) {
-	v = vv = rb_ary_elt(ary, i);
-	if (st_delete(RHASH(hash)->tbl, (st_data_t*)&vv, 0)) {
-	    rb_ary_store(ary, j++, v);
-	}
+    rb_ary_modify_check(ary);
+    ARY_SET_LEN(ary, 0);
+    if (ARY_SHARED_P(ary) && !ARY_EMBED_P(ary)) {
+	rb_ary_unshare(ary);
+	FL_SET_EMBED(ary);
     }
-    RARRAY(ary)->len = j;
+    ary_resize_capa(ary, hash_size);
+    rb_hash_foreach(hash, push_value, ary);
+    ary_recycle_hash(hash);
 
     return ary;
 }
 
 /*
  *  call-seq:
- *     array.uniq   -> an_array
- *  
- *  Returns a new array by removing duplicate values in <i>self</i>.
- *     
+ *     ary.uniq                -> new_ary
+ *     ary.uniq {|item| ...}   -> new_ary
+ *
+ *  Returns a new array by removing duplicate values in +self+.
+ *
+ *  If a block is given, it will use the return value of the block for comparison.
+ *
+ *  It compares values using their #hash and #eql? methods for efficiency.
+ *
+ *  +self+ is traversed in order, and the first occurrence is kept.
+ *
  *     a = [ "a", "a", "b", "b", "c" ]
- *     a.uniq   #=> ["a", "b", "c"]
+ *     a.uniq   # => ["a", "b", "c"]
+ *
+ *     b = [["student","sam"], ["student","george"], ["teacher","matz"]]
+ *     b.uniq {|s| s.first}   # => [["student", "sam"], ["teacher", "matz"]]
+ *
  */
 
 static VALUE
-rb_ary_uniq(ary)
-    VALUE ary;
+rb_ary_uniq(VALUE ary)
 {
-    ary = rb_ary_dup(ary);
-    rb_ary_uniq_bang(ary);
-    return ary;
+    VALUE hash, uniq;
+
+    if (RARRAY_LEN(ary) <= 1)
+        return rb_ary_dup(ary);
+    if (rb_block_given_p()) {
+	hash = ary_make_hash_by(ary);
+	uniq = rb_hash_values(hash);
+    }
+    else {
+	hash = ary_make_hash(ary);
+	uniq = rb_hash_values(hash);
+    }
+    RBASIC_SET_CLASS(uniq, rb_obj_class(ary));
+    ary_recycle_hash(hash);
+
+    return uniq;
 }
 
-/* 
+/*
  *  call-seq:
- *     array.compact!    ->   array  or  nil
+ *     ary.compact!    -> ary  or  nil
  *
- *  Removes +nil+ elements from array.
- *  Returns +nil+ if no changes were made.
+ *  Removes +nil+ elements from the array.
+ *
+ *  Returns +nil+ if no changes were made, otherwise returns the array.
  *
  *     [ "a", nil, "b", nil, "c" ].compact! #=> [ "a", "b", "c" ]
  *     [ "a", "b", "c" ].compact!           #=> nil
  */
 
 static VALUE
-rb_ary_compact_bang(ary)
-    VALUE ary;
+rb_ary_compact_bang(VALUE ary)
 {
     VALUE *p, *t, *end;
+    long n;
 
     rb_ary_modify(ary);
-    p = t = RARRAY(ary)->ptr;
-    end = p + RARRAY(ary)->len;
-    
+    p = t = (VALUE *)RARRAY_CONST_PTR_TRANSIENT(ary); /* WB: no new reference */
+    end = p + RARRAY_LEN(ary);
+
     while (t < end) {
 	if (NIL_P(*t)) t++;
 	else *p++ = *t++;
     }
-    if (RARRAY(ary)->len == (p - RARRAY(ary)->ptr)) {
+    n = p - RARRAY_CONST_PTR_TRANSIENT(ary);
+    if (RARRAY_LEN(ary) == n) {
 	return Qnil;
     }
-    RARRAY(ary)->len = RARRAY(ary)->aux.capa = (p - RARRAY(ary)->ptr);
-    REALLOC_N(RARRAY(ary)->ptr, VALUE, RARRAY(ary)->len);
+    ary_resize_smaller(ary, n);
 
     return ary;
 }
 
 /*
  *  call-seq:
- *     array.compact     ->  an_array
+ *     ary.compact     -> new_ary
  *
- *  Returns a copy of _self_ with all +nil+ elements removed.
+ *  Returns a copy of +self+ with all +nil+ elements removed.
  *
  *     [ "a", nil, "b", nil, "c", nil ].compact
  *                       #=> [ "a", "b", "c" ]
  */
 
 static VALUE
-rb_ary_compact(ary)
-    VALUE ary;
+rb_ary_compact(VALUE ary)
 {
     ary = rb_ary_dup(ary);
     rb_ary_compact_bang(ary);
@@ -2839,149 +4974,1813 @@ rb_ary_compact(ary)
 
 /*
  *  call-seq:
- *     array.nitems -> int
- *  
- *  Returns the number of non-<code>nil</code> elements in _self_.
- *  May be zero.
- *     
- *     [ 1, nil, 3, nil, 5 ].nitems   #=> 3
+ *     ary.count                   -> int
+ *     ary.count(obj)              -> int
+ *     ary.count {|item| block}    -> int
+ *
+ *  Returns the number of elements.
+ *
+ *  If an argument is given, counts the number of elements which equal +obj+
+ *  using <code>==</code>.
+ *
+ *  If a block is given, counts the number of elements for which the block
+ *  returns a true value.
+ *
+ *     ary = [1, 2, 4, 2]
+ *     ary.count                  #=> 4
+ *     ary.count(2)               #=> 2
+ *     ary.count {|x| x%2 == 0}   #=> 3
+ *
  */
 
 static VALUE
-rb_ary_nitems(ary)
-    VALUE ary;
+rb_ary_count(int argc, VALUE *argv, VALUE ary)
 {
-    long n = 0;
-    VALUE *p, *pend;
+    long i, n = 0;
+
+    if (rb_check_arity(argc, 0, 1) == 0) {
+	VALUE v;
+
+	if (!rb_block_given_p())
+	    return LONG2NUM(RARRAY_LEN(ary));
 
-    p = RARRAY(ary)->ptr;
-    pend = p + RARRAY(ary)->len;
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	    v = RARRAY_AREF(ary, i);
+	    if (RTEST(rb_yield(v))) n++;
+	}
+    }
+    else {
+        VALUE obj = argv[0];
 
-    while (p < pend) {
-	if (!NIL_P(*p)) n++;
-	p++;
+	if (rb_block_given_p()) {
+	    rb_warn("given block not used");
+	}
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	    if (rb_equal(RARRAY_AREF(ary, i), obj)) n++;
+	}
     }
+
     return LONG2NUM(n);
 }
 
-static long
-flatten(ary, idx, ary2, memo)
-    VALUE ary;
-    long idx;
-    VALUE ary2, memo;
+static VALUE
+flatten(VALUE ary, int level, int *modified)
 {
-    VALUE id;
-    long i = idx;
-    long n, lim = idx + RARRAY(ary2)->len;
-
-    id = rb_obj_id(ary2);
-    if (rb_ary_includes(memo, id)) {
-	rb_raise(rb_eArgError, "tried to flatten recursive array");
-    }
-    rb_ary_push(memo, id);
-    rb_ary_splice(ary, idx, 1, ary2);
-    while (i < lim) {
-	VALUE tmp;
-
-	tmp = rb_check_array_type(rb_ary_elt(ary, i));
-	if (!NIL_P(tmp)) {
-	    n = flatten(ary, i, tmp, memo);
-	    i += n; lim += n;
+    long i = 0;
+    VALUE stack, result, tmp, elt;
+    st_table *memo;
+    st_data_t id;
+
+    stack = ary_new(0, ARY_DEFAULT_SIZE);
+    result = ary_new(0, RARRAY_LEN(ary));
+    memo = st_init_numtable();
+    st_insert(memo, (st_data_t)ary, (st_data_t)Qtrue);
+    *modified = 0;
+
+    while (1) {
+	while (i < RARRAY_LEN(ary)) {
+	    elt = RARRAY_AREF(ary, i++);
+	    if (level >= 0 && RARRAY_LEN(stack) / 2 >= level) {
+		rb_ary_push(result, elt);
+		continue;
+	    }
+	    tmp = rb_check_array_type(elt);
+	    if (RBASIC(result)->klass) {
+		rb_raise(rb_eRuntimeError, "flatten reentered");
+	    }
+	    if (NIL_P(tmp)) {
+		rb_ary_push(result, elt);
+	    }
+	    else {
+		*modified = 1;
+		id = (st_data_t)tmp;
+		if (st_lookup(memo, id, 0)) {
+		    st_free_table(memo);
+		    rb_raise(rb_eArgError, "tried to flatten recursive array");
+		}
+		st_insert(memo, id, (st_data_t)Qtrue);
+		rb_ary_push(stack, ary);
+		rb_ary_push(stack, LONG2NUM(i));
+		ary = tmp;
+		i = 0;
+	    }
+	}
+	if (RARRAY_LEN(stack) == 0) {
+	    break;
 	}
-	i++;
+	id = (st_data_t)ary;
+	st_delete(memo, &id, 0);
+	tmp = rb_ary_pop(stack);
+	i = NUM2LONG(tmp);
+	ary = rb_ary_pop(stack);
     }
-    rb_ary_pop(memo);
 
-    return lim - idx - 1;	/* returns number of increased items */
+    st_free_table(memo);
+
+    RBASIC_SET_CLASS(result, rb_obj_class(ary));
+    return result;
 }
 
 /*
  *  call-seq:
- *     array.flatten! -> array or nil
- *  
- *  Flattens _self_ in place.
- *  Returns <code>nil</code> if no modifications were made (i.e.,
- *  <i>array</i> contains no subarrays.)
- *     
+ *     ary.flatten!        -> ary or nil
+ *     ary.flatten!(level) -> ary or nil
+ *
+ *  Flattens +self+ in place.
+ *
+ *  Returns +nil+ if no modifications were made (i.e., the array contains no
+ *  subarrays.)
+ *
+ *  The optional +level+ argument determines the level of recursion to flatten.
+ *
  *     a = [ 1, 2, [3, [4, 5] ] ]
  *     a.flatten!   #=> [1, 2, 3, 4, 5]
  *     a.flatten!   #=> nil
  *     a            #=> [1, 2, 3, 4, 5]
+ *     a = [ 1, 2, [3, [4, 5] ] ]
+ *     a.flatten!(1) #=> [1, 2, 3, [4, 5]]
  */
 
 static VALUE
-rb_ary_flatten_bang(ary)
-    VALUE ary;
+rb_ary_flatten_bang(int argc, VALUE *argv, VALUE ary)
 {
-    long i = 0;
-    int mod = 0;
-    VALUE memo = Qnil;
+    int mod = 0, level = -1;
+    VALUE result, lv;
 
-    while (i<RARRAY(ary)->len) {
-	VALUE ary2 = RARRAY(ary)->ptr[i];
-	VALUE tmp;
+    lv = (rb_check_arity(argc, 0, 1) ? argv[0] : Qnil);
+    rb_ary_modify_check(ary);
+    if (!NIL_P(lv)) level = NUM2INT(lv);
+    if (level == 0) return Qnil;
 
-	tmp = rb_check_array_type(ary2);
-	if (!NIL_P(tmp)) {
-	    if (NIL_P(memo)) {
-		memo = rb_ary_new();
-	    }
-	    i += flatten(ary, i, tmp, memo);
-	    mod = 1;
-	}
-	i++;
+    result = flatten(ary, level, &mod);
+    if (mod == 0) {
+	ary_discard(result);
+	return Qnil;
     }
-    if (mod == 0) return Qnil;
+    if (!(mod = ARY_EMBED_P(result))) rb_obj_freeze(result);
+    rb_ary_replace(ary, result);
+    if (mod) ARY_SET_EMBED_LEN(result, 0);
+
     return ary;
 }
 
 /*
  *  call-seq:
- *     array.flatten -> an_array
- *  
- *  Returns a new array that is a one-dimensional flattening of this
- *  array (recursively). That is, for every element that is an array,
- *  extract its elements into the new array.
- *     
+ *     ary.flatten -> new_ary
+ *     ary.flatten(level) -> new_ary
+ *
+ *  Returns a new array that is a one-dimensional flattening of +self+
+ *  (recursively).
+ *
+ *  That is, for every element that is an array, extract its elements into
+ *  the new array.
+ *
+ *  The optional +level+ argument determines the level of recursion to
+ *  flatten.
+ *
  *     s = [ 1, 2, 3 ]           #=> [1, 2, 3]
  *     t = [ 4, 5, 6, [7, 8] ]   #=> [4, 5, 6, [7, 8]]
  *     a = [ s, t, 9, 10 ]       #=> [[1, 2, 3], [4, 5, 6, [7, 8]], 9, 10]
- *     a.flatten                 #=> [1, 2, 3, 4, 5, 6, 7, 8, 9, 10
+ *     a.flatten                 #=> [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
+ *     a = [ 1, 2, [3, [4, 5] ] ]
+ *     a.flatten(1)              #=> [1, 2, 3, [4, 5]]
  */
 
 static VALUE
-rb_ary_flatten(ary)
-    VALUE ary;
+rb_ary_flatten(int argc, VALUE *argv, VALUE ary)
+{
+    int mod = 0, level = -1;
+    VALUE result;
+
+    if (rb_check_arity(argc, 0, 1) && !NIL_P(argv[0])) {
+        level = NUM2INT(argv[0]);
+        if (level == 0) return ary_make_shared_copy(ary);
+    }
+
+    result = flatten(ary, level, &mod);
+    OBJ_INFECT(result, ary);
+
+    return result;
+}
+
+#define OPTHASH_GIVEN_P(opts) \
+    (argc > 0 && !NIL_P((opts) = rb_check_hash_type(argv[argc-1])) && (--argc, 1))
+static ID id_random;
+
+#define RAND_UPTO(max) (long)rb_random_ulong_limited((randgen), (max)-1)
+
+/*
+ *  call-seq:
+ *     ary.shuffle!              -> ary
+ *     ary.shuffle!(random: rng) -> ary
+ *
+ *  Shuffles elements in +self+ in place.
+ *
+ *     a = [ 1, 2, 3 ]           #=> [1, 2, 3]
+ *     a.shuffle!                #=> [2, 3, 1]
+ *     a                         #=> [2, 3, 1]
+ *
+ *  The optional +rng+ argument will be used as the random number generator.
+ *
+ *     a.shuffle!(random: Random.new(1))  #=> [1, 3, 2]
+ */
+
+static VALUE
+rb_ary_shuffle_bang(int argc, VALUE *argv, VALUE ary)
+{
+    VALUE opts, randgen = rb_cRandom;
+    long i, len;
+
+    if (OPTHASH_GIVEN_P(opts)) {
+	VALUE rnd;
+	ID keyword_ids[1];
+
+	keyword_ids[0] = id_random;
+	rb_get_kwargs(opts, keyword_ids, 0, 1, &rnd);
+	if (rnd != Qundef) {
+	    randgen = rnd;
+	}
+    }
+    rb_check_arity(argc, 0, 0);
+    rb_ary_modify(ary);
+    i = len = RARRAY_LEN(ary);
+    RARRAY_PTR_USE(ary, ptr, {
+	while (i) {
+	    long j = RAND_UPTO(i);
+	    VALUE tmp;
+            if (len != RARRAY_LEN(ary) || ptr != RARRAY_CONST_PTR_TRANSIENT(ary)) {
+                rb_raise(rb_eRuntimeError, "modified during shuffle");
+	    }
+	    tmp = ptr[--i];
+	    ptr[i] = ptr[j];
+	    ptr[j] = tmp;
+	}
+    }); /* WB: no new reference */
+    return ary;
+}
+
+
+/*
+ *  call-seq:
+ *     ary.shuffle              -> new_ary
+ *     ary.shuffle(random: rng) -> new_ary
+ *
+ *  Returns a new array with elements of +self+ shuffled.
+ *
+ *     a = [ 1, 2, 3 ]           #=> [1, 2, 3]
+ *     a.shuffle                 #=> [2, 3, 1]
+ *     a                         #=> [1, 2, 3]
+ *
+ *  The optional +rng+ argument will be used as the random number generator.
+ *
+ *     a.shuffle(random: Random.new(1))  #=> [1, 3, 2]
+ */
+
+static VALUE
+rb_ary_shuffle(int argc, VALUE *argv, VALUE ary)
 {
     ary = rb_ary_dup(ary);
-    rb_ary_flatten_bang(ary);
+    rb_ary_shuffle_bang(argc, argv, ary);
+    return ary;
+}
+
+
+/*
+ *  call-seq:
+ *     ary.sample                  -> obj
+ *     ary.sample(random: rng)     -> obj
+ *     ary.sample(n)               -> new_ary
+ *     ary.sample(n, random: rng)  -> new_ary
+ *
+ *  Choose a random element or +n+ random elements from the array.
+ *
+ *  The elements are chosen by using random and unique indices into the array
+ *  in order to ensure that an element doesn't repeat itself unless the array
+ *  already contained duplicate elements.
+ *
+ *  If the array is empty the first form returns +nil+ and the second form
+ *  returns an empty array.
+ *
+ *     a = [ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 ]
+ *     a.sample         #=> 7
+ *     a.sample(4)      #=> [6, 4, 2, 5]
+ *
+ *  The optional +rng+ argument will be used as the random number generator.
+ *
+ *     a.sample(random: Random.new(1))     #=> 6
+ *     a.sample(4, random: Random.new(1))  #=> [6, 10, 9, 2]
+ */
+
+
+static VALUE
+rb_ary_sample(int argc, VALUE *argv, VALUE ary)
+{
+    VALUE nv, result;
+    VALUE opts, randgen = rb_cRandom;
+    long n, len, i, j, k, idx[10];
+    long rnds[numberof(idx)];
+    long memo_threshold;
+
+    if (OPTHASH_GIVEN_P(opts)) {
+	VALUE rnd;
+	ID keyword_ids[1];
+
+	keyword_ids[0] = id_random;
+	rb_get_kwargs(opts, keyword_ids, 0, 1, &rnd);
+	if (rnd != Qundef) {
+	    randgen = rnd;
+	}
+    }
+    len = RARRAY_LEN(ary);
+    if (rb_check_arity(argc, 0, 1) == 0) {
+	if (len < 2)
+	    i = 0;
+	else
+	    i = RAND_UPTO(len);
+
+	return rb_ary_elt(ary, i);
+    }
+    nv = argv[0];
+    n = NUM2LONG(nv);
+    if (n < 0) rb_raise(rb_eArgError, "negative sample number");
+    if (n > len) n = len;
+    if (n <= numberof(idx)) {
+	for (i = 0; i < n; ++i) {
+	    rnds[i] = RAND_UPTO(len - i);
+	}
+    }
+    k = len;
+    len = RARRAY_LEN(ary);
+    if (len < k && n <= numberof(idx)) {
+	for (i = 0; i < n; ++i) {
+	    if (rnds[i] >= len) return rb_ary_new_capa(0);
+	}
+    }
+    if (n > len) n = len;
+    switch (n) {
+      case 0:
+	return rb_ary_new_capa(0);
+      case 1:
+	i = rnds[0];
+	return rb_ary_new_from_values(1, &RARRAY_AREF(ary, i));
+      case 2:
+	i = rnds[0];
+	j = rnds[1];
+	if (j >= i) j++;
+	return rb_ary_new_from_args(2, RARRAY_AREF(ary, i), RARRAY_AREF(ary, j));
+      case 3:
+	i = rnds[0];
+	j = rnds[1];
+	k = rnds[2];
+	{
+	    long l = j, g = i;
+	    if (j >= i) l = i, g = ++j;
+	    if (k >= l && (++k >= g)) ++k;
+	}
+	return rb_ary_new_from_args(3, RARRAY_AREF(ary, i), RARRAY_AREF(ary, j), RARRAY_AREF(ary, k));
+    }
+    memo_threshold =
+	len < 2560 ? len / 128 :
+	len < 5120 ? len / 64 :
+	len < 10240 ? len / 32 :
+	len / 16;
+    if (n <= numberof(idx)) {
+	long sorted[numberof(idx)];
+	sorted[0] = idx[0] = rnds[0];
+	for (i=1; i<n; i++) {
+	    k = rnds[i];
+	    for (j = 0; j < i; ++j) {
+		if (k < sorted[j]) break;
+		++k;
+	    }
+	    memmove(&sorted[j+1], &sorted[j], sizeof(sorted[0])*(i-j));
+	    sorted[j] = idx[i] = k;
+	}
+	result = rb_ary_new_capa(n);
+        RARRAY_PTR_USE_TRANSIENT(result, ptr_result, {
+	    for (i=0; i<n; i++) {
+		ptr_result[i] = RARRAY_AREF(ary, idx[i]);
+	    }
+	});
+    }
+    else if (n <= memo_threshold / 2) {
+	long max_idx = 0;
+#undef RUBY_UNTYPED_DATA_WARNING
+#define RUBY_UNTYPED_DATA_WARNING 0
+	VALUE vmemo = Data_Wrap_Struct(0, 0, st_free_table, 0);
+	st_table *memo = st_init_numtable_with_size(n);
+	DATA_PTR(vmemo) = memo;
+	result = rb_ary_new_capa(n);
+	RARRAY_PTR_USE(result, ptr_result, {
+	    for (i=0; i<n; i++) {
+		long r = RAND_UPTO(len-i) + i;
+		ptr_result[i] = r;
+		if (r > max_idx) max_idx = r;
+	    }
+	    len = RARRAY_LEN(ary);
+	    if (len <= max_idx) n = 0;
+	    else if (n > len) n = len;
+            RARRAY_PTR_USE_TRANSIENT(ary, ptr_ary, {
+		for (i=0; i<n; i++) {
+		    long j2 = j = ptr_result[i];
+		    long i2 = i;
+		    st_data_t value;
+		    if (st_lookup(memo, (st_data_t)i, &value)) i2 = (long)value;
+		    if (st_lookup(memo, (st_data_t)j, &value)) j2 = (long)value;
+		    st_insert(memo, (st_data_t)j, (st_data_t)i2);
+		    ptr_result[i] = ptr_ary[j2];
+		}
+	    });
+	});
+	DATA_PTR(vmemo) = 0;
+	st_free_table(memo);
+    }
+    else {
+	result = rb_ary_dup(ary);
+	RBASIC_CLEAR_CLASS(result);
+	RB_GC_GUARD(ary);
+	RARRAY_PTR_USE(result, ptr_result, {
+	    for (i=0; i<n; i++) {
+		j = RAND_UPTO(len-i) + i;
+		nv = ptr_result[j];
+		ptr_result[j] = ptr_result[i];
+		ptr_result[i] = nv;
+	    }
+	});
+	RBASIC_SET_CLASS_RAW(result, rb_cArray);
+    }
+    ARY_SET_LEN(result, n);
+
+    return result;
+}
+
+static VALUE
+rb_ary_cycle_size(VALUE self, VALUE args, VALUE eobj)
+{
+    long mul;
+    VALUE n = Qnil;
+    if (args && (RARRAY_LEN(args) > 0)) {
+	n = RARRAY_AREF(args, 0);
+    }
+    if (RARRAY_LEN(self) == 0) return INT2FIX(0);
+    if (n == Qnil) return DBL2NUM(HUGE_VAL);
+    mul = NUM2LONG(n);
+    if (mul <= 0) return INT2FIX(0);
+    n = LONG2FIX(mul);
+    return rb_fix_mul_fix(rb_ary_length(self), n);
+}
+
+/*
+ *  call-seq:
+ *     ary.cycle(n=nil) {|obj| block}    -> nil
+ *     ary.cycle(n=nil)                  -> Enumerator
+ *
+ *  Calls the given block for each element +n+ times or forever if +nil+ is
+ *  given.
+ *
+ *  Does nothing if a non-positive number is given or the array is empty.
+ *
+ *  Returns +nil+ if the loop has finished without getting interrupted.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *     a = ["a", "b", "c"]
+ *     a.cycle {|x| puts x}       # print, a, b, c, a, b, c,.. forever.
+ *     a.cycle(2) {|x| puts x}    # print, a, b, c, a, b, c.
+ *
+ */
+
+static VALUE
+rb_ary_cycle(int argc, VALUE *argv, VALUE ary)
+{
+    long n, i;
+
+    rb_check_arity(argc, 0, 1);
+
+    RETURN_SIZED_ENUMERATOR(ary, argc, argv, rb_ary_cycle_size);
+    if (argc == 0 || NIL_P(argv[0])) {
+        n = -1;
+    }
+    else {
+        n = NUM2LONG(argv[0]);
+        if (n <= 0) return Qnil;
+    }
+
+    while (RARRAY_LEN(ary) > 0 && (n < 0 || 0 < n--)) {
+        for (i=0; i<RARRAY_LEN(ary); i++) {
+            rb_yield(RARRAY_AREF(ary, i));
+        }
+    }
+    return Qnil;
+}
+
+#define tmpary(n) rb_ary_tmp_new(n)
+#define tmpary_discard(a) (ary_discard(a), RBASIC_SET_CLASS_RAW(a, rb_cArray))
+
+/*
+ * Build a ruby array of the corresponding values and yield it to the
+ * associated block.
+ * Return the class of +values+ for reentry check.
+ */
+static int
+yield_indexed_values(const VALUE values, const long r, const long *const p)
+{
+    const VALUE result = rb_ary_new2(r);
+    long i;
+
+    for (i = 0; i < r; i++) RARRAY_ASET(result, i, RARRAY_AREF(values, p[i]));
+    ARY_SET_LEN(result, r);
+    rb_yield(result);
+    return !RBASIC(values)->klass;
+}
+
+/*
+ * Compute permutations of +r+ elements of the set <code>[0..n-1]</code>.
+ *
+ * When we have a complete permutation of array indices, copy the values
+ * at those indices into a new array and yield that array.
+ *
+ * n: the size of the set
+ * r: the number of elements in each permutation
+ * p: the array (of size r) that we're filling in
+ * used: an array of booleans: whether a given index is already used
+ * values: the Ruby array that holds the actual values to permute
+ */
+static void
+permute0(const long n, const long r, long *const p, char *const used, const VALUE values)
+{
+    long i = 0, index = 0;
+
+    for (;;) {
+	const char *const unused = memchr(&used[i], 0, n-i);
+	if (!unused) {
+	    if (!index) break;
+	    i = p[--index];                /* pop index */
+	    used[i++] = 0;                 /* index unused */
+	}
+	else {
+	    i = unused - used;
+	    p[index] = i;
+	    used[i] = 1;                   /* mark index used */
+	    ++index;
+	    if (index < r-1) {             /* if not done yet */
+		p[index] = i = 0;
+		continue;
+	    }
+	    for (i = 0; i < n; ++i) {
+		if (used[i]) continue;
+		p[index] = i;
+		if (!yield_indexed_values(values, r, p)) {
+		    rb_raise(rb_eRuntimeError, "permute reentered");
+		}
+	    }
+	    i = p[--index];                /* pop index */
+	    used[i] = 0;                   /* index unused */
+	    p[index] = ++i;
+	}
+    }
+}
+
+/*
+ * Returns the product of from, from-1, ..., from - how_many + 1.
+ * http://en.wikipedia.org/wiki/Pochhammer_symbol
+ */
+static VALUE
+descending_factorial(long from, long how_many)
+{
+    VALUE cnt;
+    if (how_many > 0) {
+	cnt = LONG2FIX(from);
+	while (--how_many > 0) {
+	    long v = --from;
+	    cnt = rb_int_mul(cnt, LONG2FIX(v));
+	}
+    }
+    else {
+	cnt = LONG2FIX(how_many == 0);
+    }
+    return cnt;
+}
+
+static VALUE
+binomial_coefficient(long comb, long size)
+{
+    VALUE r;
+    long i;
+    if (comb > size-comb) {
+	comb = size-comb;
+    }
+    if (comb < 0) {
+	return LONG2FIX(0);
+    }
+    else if (comb == 0) {
+	return LONG2FIX(1);
+    }
+    r = LONG2FIX(size);
+    for (i = 1; i < comb; ++i) {
+	r = rb_int_mul(r, LONG2FIX(size - i));
+	r = rb_int_idiv(r, LONG2FIX(i + 1));
+    }
+    return r;
+}
+
+static VALUE
+rb_ary_permutation_size(VALUE ary, VALUE args, VALUE eobj)
+{
+    long n = RARRAY_LEN(ary);
+    long k = (args && (RARRAY_LEN(args) > 0)) ? NUM2LONG(RARRAY_AREF(args, 0)) : n;
+
+    return descending_factorial(n, k);
+}
+
+/*
+ *  call-seq:
+ *     ary.permutation {|p| block}            -> ary
+ *     ary.permutation                        -> Enumerator
+ *     ary.permutation(n) {|p| block}         -> ary
+ *     ary.permutation(n)                     -> Enumerator
+ *
+ * When invoked with a block, yield all permutations of length +n+ of the
+ * elements of the array, then return the array itself.
+ *
+ * If +n+ is not specified, yield all permutations of all elements.
+ *
+ * The implementation makes no guarantees about the order in which the
+ * permutations are yielded.
+ *
+ * If no block is given, an Enumerator is returned instead.
+ *
+ * Examples:
+ *
+ *   a = [1, 2, 3]
+ *   a.permutation.to_a    #=> [[1,2,3],[1,3,2],[2,1,3],[2,3,1],[3,1,2],[3,2,1]]
+ *   a.permutation(1).to_a #=> [[1],[2],[3]]
+ *   a.permutation(2).to_a #=> [[1,2],[1,3],[2,1],[2,3],[3,1],[3,2]]
+ *   a.permutation(3).to_a #=> [[1,2,3],[1,3,2],[2,1,3],[2,3,1],[3,1,2],[3,2,1]]
+ *   a.permutation(0).to_a #=> [[]] # one permutation of length 0
+ *   a.permutation(4).to_a #=> []   # no permutations of length 4
+ */
+
+static VALUE
+rb_ary_permutation(int argc, VALUE *argv, VALUE ary)
+{
+    long r, n, i;
+
+    n = RARRAY_LEN(ary);                  /* Array length */
+    RETURN_SIZED_ENUMERATOR(ary, argc, argv, rb_ary_permutation_size);   /* Return enumerator if no block */
+    r = n;
+    if (rb_check_arity(argc, 0, 1) && !NIL_P(argv[0]))
+        r = NUM2LONG(argv[0]);            /* Permutation size from argument */
+
+    if (r < 0 || n < r) {
+	/* no permutations: yield nothing */
+    }
+    else if (r == 0) { /* exactly one permutation: the zero-length array */
+	rb_yield(rb_ary_new2(0));
+    }
+    else if (r == 1) { /* this is a special, easy case */
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	    rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i)));
+	}
+    }
+    else {             /* this is the general case */
+	volatile VALUE t0;
+	long *p = ALLOCV_N(long, t0, r+roomof(n, sizeof(long)));
+	char *used = (char*)(p + r);
+	VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */
+	RBASIC_CLEAR_CLASS(ary0);
+
+	MEMZERO(used, char, n); /* initialize array */
+
+	permute0(n, r, p, used, ary0); /* compute and yield permutations */
+	ALLOCV_END(t0);
+	RBASIC_SET_CLASS_RAW(ary0, rb_cArray);
+    }
+    return ary;
+}
+
+static void
+combinate0(const long len, const long n, long *const stack, const VALUE values)
+{
+    long lev = 0;
+
+    MEMZERO(stack+1, long, n);
+    stack[0] = -1;
+    for (;;) {
+	for (lev++; lev < n; lev++) {
+	    stack[lev+1] = stack[lev]+1;
+	}
+	if (!yield_indexed_values(values, n, stack+1)) {
+	    rb_raise(rb_eRuntimeError, "combination reentered");
+	}
+	do {
+	    if (lev == 0) return;
+	    stack[lev--]++;
+	} while (stack[lev+1]+n == len+lev+1);
+    }
+}
+
+static VALUE
+rb_ary_combination_size(VALUE ary, VALUE args, VALUE eobj)
+{
+    long n = RARRAY_LEN(ary);
+    long k = NUM2LONG(RARRAY_AREF(args, 0));
+
+    return binomial_coefficient(k, n);
+}
+
+/*
+ *  call-seq:
+ *     ary.combination(n) {|c| block}      -> ary
+ *     ary.combination(n)                  -> Enumerator
+ *
+ * When invoked with a block, yields all combinations of length +n+ of elements
+ * from the array and then returns the array itself.
+ *
+ * The implementation makes no guarantees about the order in which the
+ * combinations are yielded.
+ *
+ * If no block is given, an Enumerator is returned instead.
+ *
+ * Examples:
+ *
+ *     a = [1, 2, 3, 4]
+ *     a.combination(1).to_a  #=> [[1],[2],[3],[4]]
+ *     a.combination(2).to_a  #=> [[1,2],[1,3],[1,4],[2,3],[2,4],[3,4]]
+ *     a.combination(3).to_a  #=> [[1,2,3],[1,2,4],[1,3,4],[2,3,4]]
+ *     a.combination(4).to_a  #=> [[1,2,3,4]]
+ *     a.combination(0).to_a  #=> [[]] # one combination of length 0
+ *     a.combination(5).to_a  #=> []   # no combinations of length 5
+ *
+ */
+
+static VALUE
+rb_ary_combination(VALUE ary, VALUE num)
+{
+    long i, n, len;
+
+    n = NUM2LONG(num);
+    RETURN_SIZED_ENUMERATOR(ary, 1, &num, rb_ary_combination_size);
+    len = RARRAY_LEN(ary);
+    if (n < 0 || len < n) {
+	/* yield nothing */
+    }
+    else if (n == 0) {
+	rb_yield(rb_ary_new2(0));
+    }
+    else if (n == 1) {
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	    rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i)));
+	}
+    }
+    else {
+	VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */
+	volatile VALUE t0;
+	long *stack = ALLOCV_N(long, t0, n+1);
+
+	RBASIC_CLEAR_CLASS(ary0);
+	combinate0(len, n, stack, ary0);
+	ALLOCV_END(t0);
+	RBASIC_SET_CLASS_RAW(ary0, rb_cArray);
+    }
+    return ary;
+}
+
+/*
+ * Compute repeated permutations of +r+ elements of the set
+ * <code>[0..n-1]</code>.
+ *
+ * When we have a complete repeated permutation of array indices, copy the
+ * values at those indices into a new array and yield that array.
+ *
+ * n: the size of the set
+ * r: the number of elements in each permutation
+ * p: the array (of size r) that we're filling in
+ * values: the Ruby array that holds the actual values to permute
+ */
+static void
+rpermute0(const long n, const long r, long *const p, const VALUE values)
+{
+    long i = 0, index = 0;
+
+    p[index] = i;
+    for (;;) {
+	if (++index < r-1) {
+	    p[index] = i = 0;
+	    continue;
+	}
+	for (i = 0; i < n; ++i) {
+	    p[index] = i;
+	    if (!yield_indexed_values(values, r, p)) {
+		rb_raise(rb_eRuntimeError, "repeated permute reentered");
+	    }
+	}
+	do {
+	    if (index <= 0) return;
+	} while ((i = ++p[--index]) >= n);
+    }
+}
+
+static VALUE
+rb_ary_repeated_permutation_size(VALUE ary, VALUE args, VALUE eobj)
+{
+    long n = RARRAY_LEN(ary);
+    long k = NUM2LONG(RARRAY_AREF(args, 0));
+
+    if (k < 0) {
+	return LONG2FIX(0);
+    }
+    if (n <= 0) {
+	return LONG2FIX(!k);
+    }
+    return rb_int_positive_pow(n, (unsigned long)k);
+}
+
+/*
+ *  call-seq:
+ *     ary.repeated_permutation(n) {|p| block}   -> ary
+ *     ary.repeated_permutation(n)               -> Enumerator
+ *
+ * When invoked with a block, yield all repeated permutations of length +n+ of
+ * the elements of the array, then return the array itself.
+ *
+ * The implementation makes no guarantees about the order in which the repeated
+ * permutations are yielded.
+ *
+ * If no block is given, an Enumerator is returned instead.
+ *
+ * Examples:
+ *
+ *     a = [1, 2]
+ *     a.repeated_permutation(1).to_a  #=> [[1], [2]]
+ *     a.repeated_permutation(2).to_a  #=> [[1,1],[1,2],[2,1],[2,2]]
+ *     a.repeated_permutation(3).to_a  #=> [[1,1,1],[1,1,2],[1,2,1],[1,2,2],
+ *                                     #    [2,1,1],[2,1,2],[2,2,1],[2,2,2]]
+ *     a.repeated_permutation(0).to_a  #=> [[]] # one permutation of length 0
+ */
+
+static VALUE
+rb_ary_repeated_permutation(VALUE ary, VALUE num)
+{
+    long r, n, i;
+
+    n = RARRAY_LEN(ary);                  /* Array length */
+    RETURN_SIZED_ENUMERATOR(ary, 1, &num, rb_ary_repeated_permutation_size);      /* Return Enumerator if no block */
+    r = NUM2LONG(num);                    /* Permutation size from argument */
+
+    if (r < 0) {
+	/* no permutations: yield nothing */
+    }
+    else if (r == 0) { /* exactly one permutation: the zero-length array */
+	rb_yield(rb_ary_new2(0));
+    }
+    else if (r == 1) { /* this is a special, easy case */
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	    rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i)));
+	}
+    }
+    else {             /* this is the general case */
+	volatile VALUE t0;
+	long *p = ALLOCV_N(long, t0, r);
+	VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */
+	RBASIC_CLEAR_CLASS(ary0);
+
+	rpermute0(n, r, p, ary0); /* compute and yield repeated permutations */
+	ALLOCV_END(t0);
+	RBASIC_SET_CLASS_RAW(ary0, rb_cArray);
+    }
     return ary;
 }
 
+static void
+rcombinate0(const long n, const long r, long *const p, const long rest, const VALUE values)
+{
+    long i = 0, index = 0;
+
+    p[index] = i;
+    for (;;) {
+	if (++index < r-1) {
+	    p[index] = i;
+	    continue;
+	}
+	for (; i < n; ++i) {
+	    p[index] = i;
+	    if (!yield_indexed_values(values, r, p)) {
+		rb_raise(rb_eRuntimeError, "repeated combination reentered");
+	    }
+	}
+	do {
+	    if (index <= 0) return;
+	} while ((i = ++p[--index]) >= n);
+    }
+}
+
+static VALUE
+rb_ary_repeated_combination_size(VALUE ary, VALUE args, VALUE eobj)
+{
+    long n = RARRAY_LEN(ary);
+    long k = NUM2LONG(RARRAY_AREF(args, 0));
+    if (k == 0) {
+	return LONG2FIX(1);
+    }
+    return binomial_coefficient(k, n + k - 1);
+}
 
-/* Arrays are ordered, integer-indexed collections of any object. 
- * Array indexing starts at 0, as in C or Java.  A negative index is 
- * assumed to be relative to the end of the array---that is, an index of -1 
- * indicates the last element of the array, -2 is the next to last 
- * element in the array, and so on. 
+/*
+ *  call-seq:
+ *     ary.repeated_combination(n) {|c| block}   -> ary
+ *     ary.repeated_combination(n)               -> Enumerator
+ *
+ * When invoked with a block, yields all repeated combinations of length +n+ of
+ * elements from the array and then returns the array itself.
+ *
+ * The implementation makes no guarantees about the order in which the repeated
+ * combinations are yielded.
+ *
+ * If no block is given, an Enumerator is returned instead.
+ *
+ * Examples:
+ *
+ *   a = [1, 2, 3]
+ *   a.repeated_combination(1).to_a  #=> [[1], [2], [3]]
+ *   a.repeated_combination(2).to_a  #=> [[1,1],[1,2],[1,3],[2,2],[2,3],[3,3]]
+ *   a.repeated_combination(3).to_a  #=> [[1,1,1],[1,1,2],[1,1,3],[1,2,2],[1,2,3],
+ *                                   #    [1,3,3],[2,2,2],[2,2,3],[2,3,3],[3,3,3]]
+ *   a.repeated_combination(4).to_a  #=> [[1,1,1,1],[1,1,1,2],[1,1,1,3],[1,1,2,2],[1,1,2,3],
+ *                                   #    [1,1,3,3],[1,2,2,2],[1,2,2,3],[1,2,3,3],[1,3,3,3],
+ *                                   #    [2,2,2,2],[2,2,2,3],[2,2,3,3],[2,3,3,3],[3,3,3,3]]
+ *   a.repeated_combination(0).to_a  #=> [[]] # one combination of length 0
+ *
+ */
+
+static VALUE
+rb_ary_repeated_combination(VALUE ary, VALUE num)
+{
+    long n, i, len;
+
+    n = NUM2LONG(num);                 /* Combination size from argument */
+    RETURN_SIZED_ENUMERATOR(ary, 1, &num, rb_ary_repeated_combination_size);   /* Return enumerator if no block */
+    len = RARRAY_LEN(ary);
+    if (n < 0) {
+	/* yield nothing */
+    }
+    else if (n == 0) {
+	rb_yield(rb_ary_new2(0));
+    }
+    else if (n == 1) {
+	for (i = 0; i < RARRAY_LEN(ary); i++) {
+	    rb_yield(rb_ary_new3(1, RARRAY_AREF(ary, i)));
+	}
+    }
+    else if (len == 0) {
+	/* yield nothing */
+    }
+    else {
+	volatile VALUE t0;
+	long *p = ALLOCV_N(long, t0, n);
+	VALUE ary0 = ary_make_shared_copy(ary); /* private defensive copy of ary */
+	RBASIC_CLEAR_CLASS(ary0);
+
+	rcombinate0(len, n, p, n, ary0); /* compute and yield repeated combinations */
+	ALLOCV_END(t0);
+	RBASIC_SET_CLASS_RAW(ary0, rb_cArray);
+    }
+    return ary;
+}
+
+/*
+ *  call-seq:
+ *     ary.product(other_ary, ...)                -> new_ary
+ *     ary.product(other_ary, ...) {|p| block}    -> ary
+ *
+ *  Returns an array of all combinations of elements from all arrays.
+ *
+ *  The length of the returned array is the product of the length of +self+ and
+ *  the argument arrays.
+ *
+ *  If given a block, #product will yield all combinations and return +self+
+ *  instead.
+ *
+ *     [1,2,3].product([4,5])     #=> [[1,4],[1,5],[2,4],[2,5],[3,4],[3,5]]
+ *     [1,2].product([1,2])       #=> [[1,1],[1,2],[2,1],[2,2]]
+ *     [1,2].product([3,4],[5,6]) #=> [[1,3,5],[1,3,6],[1,4,5],[1,4,6],
+ *                                #     [2,3,5],[2,3,6],[2,4,5],[2,4,6]]
+ *     [1,2].product()            #=> [[1],[2]]
+ *     [1,2].product([])          #=> []
+ */
+
+static VALUE
+rb_ary_product(int argc, VALUE *argv, VALUE ary)
+{
+    int n = argc+1;    /* How many arrays we're operating on */
+    volatile VALUE t0 = tmpary(n);
+    volatile VALUE t1 = Qundef;
+    VALUE *arrays = RARRAY_PTR(t0); /* The arrays we're computing the product of */
+    int *counters = ALLOCV_N(int, t1, n); /* The current position in each one */
+    VALUE result = Qnil;      /* The array we'll be returning, when no block given */
+    long i,j;
+    long resultlen = 1;
+
+    RBASIC_CLEAR_CLASS(t0);
+
+    /* initialize the arrays of arrays */
+    ARY_SET_LEN(t0, n);
+    arrays[0] = ary;
+    for (i = 1; i < n; i++) arrays[i] = Qnil;
+    for (i = 1; i < n; i++) arrays[i] = to_ary(argv[i-1]);
+
+    /* initialize the counters for the arrays */
+    for (i = 0; i < n; i++) counters[i] = 0;
+
+    /* Otherwise, allocate and fill in an array of results */
+    if (rb_block_given_p()) {
+	/* Make defensive copies of arrays; exit if any is empty */
+	for (i = 0; i < n; i++) {
+	    if (RARRAY_LEN(arrays[i]) == 0) goto done;
+	    arrays[i] = ary_make_shared_copy(arrays[i]);
+	}
+    }
+    else {
+	/* Compute the length of the result array; return [] if any is empty */
+	for (i = 0; i < n; i++) {
+	    long k = RARRAY_LEN(arrays[i]);
+	    if (k == 0) {
+		result = rb_ary_new2(0);
+		goto done;
+	    }
+            if (MUL_OVERFLOW_LONG_P(resultlen, k))
+		rb_raise(rb_eRangeError, "too big to product");
+	    resultlen *= k;
+	}
+	result = rb_ary_new2(resultlen);
+    }
+    for (;;) {
+	int m;
+	/* fill in one subarray */
+	VALUE subarray = rb_ary_new2(n);
+	for (j = 0; j < n; j++) {
+	    rb_ary_push(subarray, rb_ary_entry(arrays[j], counters[j]));
+	}
+
+	/* put it on the result array */
+	if (NIL_P(result)) {
+	    FL_SET(t0, FL_USER5);
+	    rb_yield(subarray);
+	    if (! FL_TEST(t0, FL_USER5)) {
+		rb_raise(rb_eRuntimeError, "product reentered");
+	    }
+	    else {
+		FL_UNSET(t0, FL_USER5);
+	    }
+	}
+	else {
+	    rb_ary_push(result, subarray);
+	}
+
+	/*
+	 * Increment the last counter.  If it overflows, reset to 0
+	 * and increment the one before it.
+	 */
+	m = n-1;
+	counters[m]++;
+	while (counters[m] == RARRAY_LEN(arrays[m])) {
+	    counters[m] = 0;
+	    /* If the first counter overflows, we are done */
+	    if (--m < 0) goto done;
+	    counters[m]++;
+	}
+    }
+done:
+    tmpary_discard(t0);
+    ALLOCV_END(t1);
+
+    return NIL_P(result) ? ary : result;
+}
+
+/*
+ *  call-seq:
+ *     ary.take(n)               -> new_ary
+ *
+ *  Returns first +n+ elements from the array.
+ *
+ *  If a negative number is given, raises an ArgumentError.
+ *
+ *  See also Array#drop
+ *
+ *     a = [1, 2, 3, 4, 5, 0]
+ *     a.take(3)             #=> [1, 2, 3]
+ *
+ */
+
+static VALUE
+rb_ary_take(VALUE obj, VALUE n)
+{
+    long len = NUM2LONG(n);
+    if (len < 0) {
+	rb_raise(rb_eArgError, "attempt to take negative size");
+    }
+    return rb_ary_subseq(obj, 0, len);
+}
+
+/*
+ *  call-seq:
+ *     ary.take_while {|obj| block}    -> new_ary
+ *     ary.take_while                  -> Enumerator
+ *
+ *  Passes elements to the block until the block returns +nil+ or +false+, then
+ *  stops iterating and returns an array of all prior elements.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *  See also Array#drop_while
+ *
+ *     a = [1, 2, 3, 4, 5, 0]
+ *     a.take_while {|i| i < 3}    #=> [1, 2]
+ *
+ */
+
+static VALUE
+rb_ary_take_while(VALUE ary)
+{
+    long i;
+
+    RETURN_ENUMERATOR(ary, 0, 0);
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+	if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) break;
+    }
+    return rb_ary_take(ary, LONG2FIX(i));
+}
+
+/*
+ *  call-seq:
+ *     ary.drop(n)               -> new_ary
+ *
+ *  Drops first +n+ elements from +ary+ and returns the rest of the elements in
+ *  an array.
+ *
+ *  If a negative number is given, raises an ArgumentError.
+ *
+ *  See also Array#take
+ *
+ *     a = [1, 2, 3, 4, 5, 0]
+ *     a.drop(3)             #=> [4, 5, 0]
+ *
+ */
+
+static VALUE
+rb_ary_drop(VALUE ary, VALUE n)
+{
+    VALUE result;
+    long pos = NUM2LONG(n);
+    if (pos < 0) {
+	rb_raise(rb_eArgError, "attempt to drop negative size");
+    }
+
+    result = rb_ary_subseq(ary, pos, RARRAY_LEN(ary));
+    if (result == Qnil) result = rb_ary_new();
+    return result;
+}
+
+/*
+ *  call-seq:
+ *     ary.drop_while {|obj| block}     -> new_ary
+ *     ary.drop_while                  -> Enumerator
+ *
+ *  Drops elements up to, but not including, the first element for which the
+ *  block returns +nil+ or +false+ and returns an array containing the
+ *  remaining elements.
+ *
+ *  If no block is given, an Enumerator is returned instead.
+ *
+ *  See also Array#take_while
+ *
+ *     a = [1, 2, 3, 4, 5, 0]
+ *     a.drop_while {|i| i < 3 }   #=> [3, 4, 5, 0]
+ *
+ */
+
+static VALUE
+rb_ary_drop_while(VALUE ary)
+{
+    long i;
+
+    RETURN_ENUMERATOR(ary, 0, 0);
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+	if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) break;
+    }
+    return rb_ary_drop(ary, LONG2FIX(i));
+}
+
+/*
+ *  call-seq:
+ *     ary.any? [{|obj| block}  ]   -> true or false
+ *     ary.any?(pattern)            -> true or false
+ *
+ *  See also Enumerable#any?
+ */
+
+static VALUE
+rb_ary_any_p(int argc, VALUE *argv, VALUE ary)
+{
+    long i, len = RARRAY_LEN(ary);
+
+    rb_check_arity(argc, 0, 1);
+    if (!len) return Qfalse;
+    if (argc) {
+        if (rb_block_given_p()) {
+            rb_warn("given block not used");
+        }
+	for (i = 0; i < RARRAY_LEN(ary); ++i) {
+	    if (RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) return Qtrue;
+	}
+    }
+    else if (!rb_block_given_p()) {
+        for (i = 0; i < len; ++i) {
+            if (RTEST(RARRAY_AREF(ary, i))) return Qtrue;
+        }
+    }
+    else {
+	for (i = 0; i < RARRAY_LEN(ary); ++i) {
+	    if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) return Qtrue;
+	}
+    }
+    return Qfalse;
+}
+
+/*
+ *  call-seq:
+ *     ary.all? [{|obj| block}  ]   -> true or false
+ *     ary.all?(pattern)            -> true or false
+ *
+ *  See also Enumerable#all?
+ */
+
+static VALUE
+rb_ary_all_p(int argc, VALUE *argv, VALUE ary)
+{
+    long i, len = RARRAY_LEN(ary);
+
+    rb_check_arity(argc, 0, 1);
+    if (!len) return Qtrue;
+    if (argc) {
+        if (rb_block_given_p()) {
+            rb_warn("given block not used");
+        }
+        for (i = 0; i < RARRAY_LEN(ary); ++i) {
+            if (!RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) return Qfalse;
+        }
+    }
+    else if (!rb_block_given_p()) {
+        for (i = 0; i < len; ++i) {
+            if (!RTEST(RARRAY_AREF(ary, i))) return Qfalse;
+        }
+    }
+    else {
+        for (i = 0; i < RARRAY_LEN(ary); ++i) {
+            if (!RTEST(rb_yield(RARRAY_AREF(ary, i)))) return Qfalse;
+        }
+    }
+    return Qtrue;
+}
+
+/*
+ *  call-seq:
+ *     ary.none? [{|obj| block}  ]   -> true or false
+ *     ary.none?(pattern)            -> true or false
+ *
+ *  See also Enumerable#none?
+ */
+
+static VALUE
+rb_ary_none_p(int argc, VALUE *argv, VALUE ary)
+{
+    long i, len = RARRAY_LEN(ary);
+
+    rb_check_arity(argc, 0, 1);
+    if (!len) return Qtrue;
+    if (argc) {
+        if (rb_block_given_p()) {
+            rb_warn("given block not used");
+        }
+        for (i = 0; i < RARRAY_LEN(ary); ++i) {
+            if (RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) return Qfalse;
+        }
+    }
+    else if (!rb_block_given_p()) {
+        for (i = 0; i < len; ++i) {
+            if (RTEST(RARRAY_AREF(ary, i))) return Qfalse;
+        }
+    }
+    else {
+        for (i = 0; i < RARRAY_LEN(ary); ++i) {
+            if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) return Qfalse;
+        }
+    }
+    return Qtrue;
+}
+
+/*
+ *  call-seq:
+ *     ary.one? [{|obj| block}  ]   -> true or false
+ *     ary.one?(pattern)            -> true or false
+ *
+ *  See also Enumerable#one?
+ */
+
+static VALUE
+rb_ary_one_p(int argc, VALUE *argv, VALUE ary)
+{
+    long i, len = RARRAY_LEN(ary);
+    VALUE result = Qfalse;
+
+    rb_check_arity(argc, 0, 1);
+    if (!len) return Qfalse;
+    if (argc) {
+        if (rb_block_given_p()) {
+            rb_warn("given block not used");
+        }
+        for (i = 0; i < RARRAY_LEN(ary); ++i) {
+            if (RTEST(rb_funcall(argv[0], idEqq, 1, RARRAY_AREF(ary, i)))) {
+                if (result) return Qfalse;
+                result = Qtrue;
+            }
+        }
+    }
+    else if (!rb_block_given_p()) {
+        for (i = 0; i < len; ++i) {
+            if (RTEST(RARRAY_AREF(ary, i))) {
+                if (result) return Qfalse;
+                result = Qtrue;
+            }
+        }
+    }
+    else {
+        for (i = 0; i < RARRAY_LEN(ary); ++i) {
+            if (RTEST(rb_yield(RARRAY_AREF(ary, i)))) {
+                if (result) return Qfalse;
+                result = Qtrue;
+            }
+        }
+    }
+    return result;
+}
+
+/*
+ * call-seq:
+ *   ary.dig(idx, ...)                 -> object
+ *
+ * Extracts the nested value specified by the sequence of <i>idx</i>
+ * objects by calling +dig+ at each step, returning +nil+ if any
+ * intermediate step is +nil+.
+ *
+ *   a = [[1, [2, 3]]]
+ *
+ *   a.dig(0, 1, 1)                    #=> 3
+ *   a.dig(1, 2, 3)                    #=> nil
+ *   a.dig(0, 0, 0)                    #=> TypeError: Integer does not have #dig method
+ *   [42, {foo: :bar}].dig(1, :foo)    #=> :bar
+ */
+
+static VALUE
+rb_ary_dig(int argc, VALUE *argv, VALUE self)
+{
+    rb_check_arity(argc, 1, UNLIMITED_ARGUMENTS);
+    self = rb_ary_at(self, *argv);
+    if (!--argc) return self;
+    ++argv;
+    return rb_obj_dig(argc, argv, self, Qnil);
+}
+
+static inline VALUE
+finish_exact_sum(long n, VALUE r, VALUE v, int z)
+{
+    if (n != 0)
+        v = rb_fix_plus(LONG2FIX(n), v);
+    if (r != Qundef) {
+	/* r can be an Integer when mathn is loaded */
+	if (FIXNUM_P(r))
+	    v = rb_fix_plus(r, v);
+	else if (RB_TYPE_P(r, T_BIGNUM))
+	    v = rb_big_plus(r, v);
+	else
+	    v = rb_rational_plus(r, v);
+    }
+    else if (!n && z) {
+        v = rb_fix_plus(LONG2FIX(0), v);
+    }
+    return v;
+}
+
+/*
+ * call-seq:
+ *   ary.sum(init=0)                    -> number
+ *   ary.sum(init=0) {|e| expr }        -> number
+ *
+ * Returns the sum of elements.
+ * For example, [e1, e2, e3].sum returns init + e1 + e2 + e3.
+ *
+ * If a block is given, the block is applied to each element
+ * before addition.
+ *
+ * If <i>ary</i> is empty, it returns <i>init</i>.
+ *
+ *   [].sum                             #=> 0
+ *   [].sum(0.0)                        #=> 0.0
+ *   [1, 2, 3].sum                      #=> 6
+ *   [3, 5.5].sum                       #=> 8.5
+ *   [2.5, 3.0].sum(0.0) {|e| e * e }   #=> 15.25
+ *   [Object.new].sum                   #=> TypeError
+ *
+ * The (arithmetic) mean value of an array can be obtained as follows.
+ *
+ *   mean = ary.sum(0.0) / ary.length
+ *
+ * This method can be used for non-numeric objects by
+ * explicit <i>init</i> argument.
+ *
+ *   ["a", "b", "c"].sum("")            #=> "abc"
+ *   [[1], [[2]], [3]].sum([])          #=> [1, [2], 3]
+ *
+ * However, Array#join and Array#flatten is faster than Array#sum for
+ * array of strings and array of arrays.
+ *
+ *   ["a", "b", "c"].join               #=> "abc"
+ *   [[1], [[2]], [3]].flatten(1)       #=> [1, [2], 3]
+ *
+ *
+ * Array#sum method may not respect method redefinition of "+" methods
+ * such as Integer#+.
+ *
+ */
+
+static VALUE
+rb_ary_sum(int argc, VALUE *argv, VALUE ary)
+{
+    VALUE e, v, r;
+    long i, n;
+    int block_given;
+
+    v = (rb_check_arity(argc, 0, 1) ? argv[0] : LONG2FIX(0));
+
+    block_given = rb_block_given_p();
+
+    if (RARRAY_LEN(ary) == 0)
+        return v;
+
+    n = 0;
+    r = Qundef;
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+        e = RARRAY_AREF(ary, i);
+        if (block_given)
+            e = rb_yield(e);
+        if (FIXNUM_P(e)) {
+            n += FIX2LONG(e); /* should not overflow long type */
+            if (!FIXABLE(n)) {
+                v = rb_big_plus(LONG2NUM(n), v);
+                n = 0;
+            }
+        }
+        else if (RB_TYPE_P(e, T_BIGNUM))
+            v = rb_big_plus(e, v);
+        else if (RB_TYPE_P(e, T_RATIONAL)) {
+            if (r == Qundef)
+                r = e;
+            else
+                r = rb_rational_plus(r, e);
+        }
+        else
+            goto not_exact;
+    }
+    v = finish_exact_sum(n, r, v, argc!=0);
+    return v;
+
+  not_exact:
+    v = finish_exact_sum(n, r, v, i!=0);
+
+    if (RB_FLOAT_TYPE_P(e)) {
+        /*
+         * Kahan-Babuska balancing compensated summation algorithm
+         * See http://link.springer.com/article/10.1007/s00607-005-0139-x
+         */
+        double f, c;
+
+        f = NUM2DBL(v);
+        c = 0.0;
+        goto has_float_value;
+        for (; i < RARRAY_LEN(ary); i++) {
+            double x, t;
+            e = RARRAY_AREF(ary, i);
+            if (block_given)
+                e = rb_yield(e);
+            if (RB_FLOAT_TYPE_P(e))
+              has_float_value:
+                x = RFLOAT_VALUE(e);
+            else if (FIXNUM_P(e))
+                x = FIX2LONG(e);
+            else if (RB_TYPE_P(e, T_BIGNUM))
+                x = rb_big2dbl(e);
+            else if (RB_TYPE_P(e, T_RATIONAL))
+                x = rb_num2dbl(e);
+            else
+                goto not_float;
+
+            if (isnan(f)) continue;
+            if (isnan(x)) {
+                f = x;
+                continue;
+            }
+            if (isinf(x)) {
+                if (isinf(f) && signbit(x) != signbit(f))
+                    f = NAN;
+                else
+                    f = x;
+                continue;
+            }
+            if (isinf(f)) continue;
+
+            t = f + x;
+            if (fabs(f) >= fabs(x))
+                c += ((f - t) + x);
+            else
+                c += ((x - t) + f);
+            f = t;
+        }
+        f += c;
+        return DBL2NUM(f);
+
+      not_float:
+        v = DBL2NUM(f);
+    }
+
+    goto has_some_value;
+    for (; i < RARRAY_LEN(ary); i++) {
+        e = RARRAY_AREF(ary, i);
+        if (block_given)
+            e = rb_yield(e);
+      has_some_value:
+        v = rb_funcall(v, idPLUS, 1, e);
+    }
+    return v;
+}
+
+/*
+ *  Arrays are ordered, integer-indexed collections of any object.
+ *
+ *  Array indexing starts at 0, as in C or Java.  A negative index is assumed
+ *  to be relative to the end of the array---that is, an index of -1 indicates
+ *  the last element of the array, -2 is the next to last element in the
+ *  array, and so on.
+ *
+ *  == Creating Arrays
+ *
+ *  A new array can be created by using the literal constructor
+ *  <code>[]</code>.  Arrays can contain different types of objects.  For
+ *  example, the array below contains an Integer, a String and a Float:
+ *
+ *     ary = [1, "two", 3.0] #=> [1, "two", 3.0]
+ *
+ *  An array can also be created by explicitly calling Array.new with zero, one
+ *  (the initial size of the Array) or two arguments (the initial size and a
+ *  default object).
+ *
+ *     ary = Array.new    #=> []
+ *     Array.new(3)       #=> [nil, nil, nil]
+ *     Array.new(3, true) #=> [true, true, true]
+ *
+ *  Note that the second argument populates the array with references to the
+ *  same object.  Therefore, it is only recommended in cases when you need to
+ *  instantiate arrays with natively immutable objects such as Symbols,
+ *  numbers, true or false.
+ *
+ *  To create an array with separate objects a block can be passed instead.
+ *  This method is safe to use with mutable objects such as hashes, strings or
+ *  other arrays:
+ *
+ *     Array.new(4) {Hash.new}    #=> [{}, {}, {}, {}]
+ *     Array.new(4) {|i| i.to_s } #=> ["0", "1", "2", "3"]
+ *
+ *  This is also a quick way to build up multi-dimensional arrays:
+ *
+ *     empty_table = Array.new(3) {Array.new(3)}
+ *     #=> [[nil, nil, nil], [nil, nil, nil], [nil, nil, nil]]
+ *
+ *  An array can also be created by using the Array() method, provided by
+ *  Kernel, which tries to call #to_ary, then #to_a on its argument.
+ *
+ *	Array({:a => "a", :b => "b"}) #=> [[:a, "a"], [:b, "b"]]
+ *
+ *  == Example Usage
+ *
+ *  In addition to the methods it mixes in through the Enumerable module, the
+ *  Array class has proprietary methods for accessing, searching and otherwise
+ *  manipulating arrays.
+ *
+ *  Some of the more common ones are illustrated below.
+ *
+ *  == Accessing Elements
+ *
+ *  Elements in an array can be retrieved using the Array#[] method.  It can
+ *  take a single integer argument (a numeric index), a pair of arguments
+ *  (start and length) or a range. Negative indices start counting from the end,
+ *  with -1 being the last element.
+ *
+ *     arr = [1, 2, 3, 4, 5, 6]
+ *     arr[2]    #=> 3
+ *     arr[100]  #=> nil
+ *     arr[-3]   #=> 4
+ *     arr[2, 3] #=> [3, 4, 5]
+ *     arr[1..4] #=> [2, 3, 4, 5]
+ *     arr[1..-3] #=> [2, 3, 4]
+ *
+ *  Another way to access a particular array element is by using the #at method
+ *
+ *     arr.at(0) #=> 1
+ *
+ *  The #slice method works in an identical manner to Array#[].
+ *
+ *  To raise an error for indices outside of the array bounds or else to
+ *  provide a default value when that happens, you can use #fetch.
+ *
+ *     arr = ['a', 'b', 'c', 'd', 'e', 'f']
+ *     arr.fetch(100) #=> IndexError: index 100 outside of array bounds: -6...6
+ *     arr.fetch(100, "oops") #=> "oops"
+ *
+ *  The special methods #first and #last will return the first and last
+ *  elements of an array, respectively.
+ *
+ *     arr.first #=> 1
+ *     arr.last  #=> 6
+ *
+ *  To return the first +n+ elements of an array, use #take
+ *
+ *     arr.take(3) #=> [1, 2, 3]
+ *
+ *  #drop does the opposite of #take, by returning the elements after +n+
+ *  elements have been dropped:
+ *
+ *     arr.drop(3) #=> [4, 5, 6]
+ *
+ *  == Obtaining Information about an Array
+ *
+ *  Arrays keep track of their own length at all times.  To query an array
+ *  about the number of elements it contains, use #length, #count or #size.
+ *
+ *    browsers = ['Chrome', 'Firefox', 'Safari', 'Opera', 'IE']
+ *    browsers.length #=> 5
+ *    browsers.count #=> 5
+ *
+ *  To check whether an array contains any elements at all
+ *
+ *    browsers.empty? #=> false
+ *
+ *  To check whether a particular item is included in the array
+ *
+ *    browsers.include?('Konqueror') #=> false
+ *
+ *  == Adding Items to Arrays
+ *
+ *  Items can be added to the end of an array by using either #push or #<<
+ *
+ *    arr = [1, 2, 3, 4]
+ *    arr.push(5) #=> [1, 2, 3, 4, 5]
+ *    arr << 6    #=> [1, 2, 3, 4, 5, 6]
+ *
+ *  #unshift will add a new item to the beginning of an array.
+ *
+ *     arr.unshift(0) #=> [0, 1, 2, 3, 4, 5, 6]
+ *
+ *  With #insert you can add a new element to an array at any position.
+ *
+ *     arr.insert(3, 'apple')  #=> [0, 1, 2, 'apple', 3, 4, 5, 6]
+ *
+ *  Using the #insert method, you can also insert multiple values at once:
+ *
+ *     arr.insert(3, 'orange', 'pear', 'grapefruit')
+ *     #=> [0, 1, 2, "orange", "pear", "grapefruit", "apple", 3, 4, 5, 6]
+ *
+ *  == Removing Items from an Array
+ *
+ *  The method #pop removes the last element in an array and returns it:
+ *
+ *     arr =  [1, 2, 3, 4, 5, 6]
+ *     arr.pop #=> 6
+ *     arr #=> [1, 2, 3, 4, 5]
+ *
+ *  To retrieve and at the same time remove the first item, use #shift:
+ *
+ *     arr.shift #=> 1
+ *     arr #=> [2, 3, 4, 5]
+ *
+ *  To delete an element at a particular index:
+ *
+ *     arr.delete_at(2) #=> 4
+ *     arr #=> [2, 3, 5]
+ *
+ *  To delete a particular element anywhere in an array, use #delete:
+ *
+ *     arr = [1, 2, 2, 3]
+ *     arr.delete(2) #=> 2
+ *     arr #=> [1,3]
+ *
+ *  A useful method if you need to remove +nil+ values from an array is
+ *  #compact:
+ *
+ *     arr = ['foo', 0, nil, 'bar', 7, 'baz', nil]
+ *     arr.compact  #=> ['foo', 0, 'bar', 7, 'baz']
+ *     arr          #=> ['foo', 0, nil, 'bar', 7, 'baz', nil]
+ *     arr.compact! #=> ['foo', 0, 'bar', 7, 'baz']
+ *     arr          #=> ['foo', 0, 'bar', 7, 'baz']
+ *
+ *  Another common need is to remove duplicate elements from an array.
+ *
+ *  It has the non-destructive #uniq, and destructive method #uniq!
+ *
+ *     arr = [2, 5, 6, 556, 6, 6, 8, 9, 0, 123, 556]
+ *     arr.uniq #=> [2, 5, 6, 556, 8, 9, 0, 123]
+ *
+ *  == Iterating over Arrays
+ *
+ *  Like all classes that include the Enumerable module, Array has an each
+ *  method, which defines what elements should be iterated over and how.  In
+ *  case of Array's #each, all elements in the Array instance are yielded to
+ *  the supplied block in sequence.
+ *
+ *  Note that this operation leaves the array unchanged.
+ *
+ *     arr = [1, 2, 3, 4, 5]
+ *     arr.each {|a| print a -= 10, " "}
+ *     # prints: -9 -8 -7 -6 -5
+ *     #=> [1, 2, 3, 4, 5]
+ *
+ *  Another sometimes useful iterator is #reverse_each which will iterate over
+ *  the elements in the array in reverse order.
+ *
+ *     words = %w[first second third fourth fifth sixth]
+ *     str = ""
+ *     words.reverse_each {|word| str += "#{word} "}
+ *     p str #=> "sixth fifth fourth third second first "
+ *
+ *  The #map method can be used to create a new array based on the original
+ *  array, but with the values modified by the supplied block:
+ *
+ *     arr.map {|a| 2*a}     #=> [2, 4, 6, 8, 10]
+ *     arr                   #=> [1, 2, 3, 4, 5]
+ *     arr.map! {|a| a**2}   #=> [1, 4, 9, 16, 25]
+ *     arr                   #=> [1, 4, 9, 16, 25]
+ *
+ *  == Selecting Items from an Array
+ *
+ *  Elements can be selected from an array according to criteria defined in a
+ *  block.  The selection can happen in a destructive or a non-destructive
+ *  manner.  While the destructive operations will modify the array they were
+ *  called on, the non-destructive methods usually return a new array with the
+ *  selected elements, but leave the original array unchanged.
+ *
+ *  === Non-destructive Selection
+ *
+ *     arr = [1, 2, 3, 4, 5, 6]
+ *     arr.select {|a| a > 3}       #=> [4, 5, 6]
+ *     arr.reject {|a| a < 3}       #=> [3, 4, 5, 6]
+ *     arr.drop_while {|a| a < 4}   #=> [4, 5, 6]
+ *     arr                          #=> [1, 2, 3, 4, 5, 6]
+ *
+ *  === Destructive Selection
+ *
+ *  #select! and #reject! are the corresponding destructive methods to #select
+ *  and #reject
+ *
+ *  Similar to #select vs. #reject, #delete_if and #keep_if have the exact
+ *  opposite result when supplied with the same block:
+ *
+ *     arr.delete_if {|a| a < 4}   #=> [4, 5, 6]
+ *     arr                         #=> [4, 5, 6]
+ *
+ *     arr = [1, 2, 3, 4, 5, 6]
+ *     arr.keep_if {|a| a < 4}   #=> [1, 2, 3]
+ *     arr                       #=> [1, 2, 3]
+ *
  */
 
 void
-Init_Array()
+Init_Array(void)
 {
+#undef rb_intern
+#define rb_intern(str) rb_intern_const(str)
+
     rb_cArray  = rb_define_class("Array", rb_cObject);
     rb_include_module(rb_cArray, rb_mEnumerable);
 
-    rb_define_alloc_func(rb_cArray, ary_alloc);
+    rb_define_alloc_func(rb_cArray, empty_ary_alloc);
     rb_define_singleton_method(rb_cArray, "[]", rb_ary_s_create, -1);
+    rb_define_singleton_method(rb_cArray, "try_convert", rb_ary_s_try_convert, 1);
     rb_define_method(rb_cArray, "initialize", rb_ary_initialize, -1);
     rb_define_method(rb_cArray, "initialize_copy", rb_ary_replace, 1);
 
-    rb_define_method(rb_cArray, "to_s", rb_ary_to_s, 0);
     rb_define_method(rb_cArray, "inspect", rb_ary_inspect, 0);
+    rb_define_alias(rb_cArray,  "to_s", "inspect");
     rb_define_method(rb_cArray, "to_a", rb_ary_to_a, 0);
+    rb_define_method(rb_cArray, "to_h", rb_ary_to_h, 0);
     rb_define_method(rb_cArray, "to_ary", rb_ary_to_ary_m, 0);
-    rb_define_method(rb_cArray, "frozen?",  rb_ary_frozen_p, 0);
 
     rb_define_method(rb_cArray, "==", rb_ary_equal, 1);
     rb_define_method(rb_cArray, "eql?", rb_ary_eql, 1);
@@ -2993,12 +6792,16 @@ Init_Array()
     rb_define_method(rb_cArray, "fetch", rb_ary_fetch, -1);
     rb_define_method(rb_cArray, "first", rb_ary_first, -1);
     rb_define_method(rb_cArray, "last", rb_ary_last, -1);
-    rb_define_method(rb_cArray, "concat", rb_ary_concat, 1);
+    rb_define_method(rb_cArray, "concat", rb_ary_concat_multi, -1);
+    rb_define_method(rb_cArray, "union", rb_ary_union_multi, -1);
+    rb_define_method(rb_cArray, "difference", rb_ary_difference_multi, -1);
     rb_define_method(rb_cArray, "<<", rb_ary_push, 1);
     rb_define_method(rb_cArray, "push", rb_ary_push_m, -1);
-    rb_define_method(rb_cArray, "pop", rb_ary_pop, 0);
-    rb_define_method(rb_cArray, "shift", rb_ary_shift, 0);
+    rb_define_alias(rb_cArray,  "append", "push");
+    rb_define_method(rb_cArray, "pop", rb_ary_pop_m, -1);
+    rb_define_method(rb_cArray, "shift", rb_ary_shift_m, -1);
     rb_define_method(rb_cArray, "unshift", rb_ary_unshift_m, -1);
+    rb_define_alias(rb_cArray,  "prepend", "unshift");
     rb_define_method(rb_cArray, "insert", rb_ary_insert, -1);
     rb_define_method(rb_cArray, "each", rb_ary_each, 0);
     rb_define_method(rb_cArray, "each_index", rb_ary_each_index, 0);
@@ -3006,20 +6809,26 @@ Init_Array()
     rb_define_method(rb_cArray, "length", rb_ary_length, 0);
     rb_define_alias(rb_cArray,  "size", "length");
     rb_define_method(rb_cArray, "empty?", rb_ary_empty_p, 0);
-    rb_define_method(rb_cArray, "index", rb_ary_index, 1);
-    rb_define_method(rb_cArray, "rindex", rb_ary_rindex, 1);
-    rb_define_method(rb_cArray, "indexes", rb_ary_indexes, -1);
-    rb_define_method(rb_cArray, "indices", rb_ary_indexes, -1);
+    rb_define_method(rb_cArray, "find_index", rb_ary_index, -1);
+    rb_define_method(rb_cArray, "index", rb_ary_index, -1);
+    rb_define_method(rb_cArray, "rindex", rb_ary_rindex, -1);
     rb_define_method(rb_cArray, "join", rb_ary_join_m, -1);
     rb_define_method(rb_cArray, "reverse", rb_ary_reverse_m, 0);
     rb_define_method(rb_cArray, "reverse!", rb_ary_reverse_bang, 0);
+    rb_define_method(rb_cArray, "rotate", rb_ary_rotate_m, -1);
+    rb_define_method(rb_cArray, "rotate!", rb_ary_rotate_bang, -1);
     rb_define_method(rb_cArray, "sort", rb_ary_sort, 0);
     rb_define_method(rb_cArray, "sort!", rb_ary_sort_bang, 0);
+    rb_define_method(rb_cArray, "sort_by!", rb_ary_sort_by_bang, 0);
     rb_define_method(rb_cArray, "collect", rb_ary_collect, 0);
     rb_define_method(rb_cArray, "collect!", rb_ary_collect_bang, 0);
     rb_define_method(rb_cArray, "map", rb_ary_collect, 0);
     rb_define_method(rb_cArray, "map!", rb_ary_collect_bang, 0);
     rb_define_method(rb_cArray, "select", rb_ary_select, 0);
+    rb_define_method(rb_cArray, "select!", rb_ary_select_bang, 0);
+    rb_define_method(rb_cArray, "filter", rb_ary_select, 0);
+    rb_define_method(rb_cArray, "filter!", rb_ary_select_bang, 0);
+    rb_define_method(rb_cArray, "keep_if", rb_ary_keep_if, 0);
     rb_define_method(rb_cArray, "values_at", rb_ary_values_at, -1);
     rb_define_method(rb_cArray, "delete", rb_ary_delete, 1);
     rb_define_method(rb_cArray, "delete_at", rb_ary_delete_at_m, 1);
@@ -3047,14 +6856,38 @@ Init_Array()
     rb_define_method(rb_cArray, "&", rb_ary_and, 1);
     rb_define_method(rb_cArray, "|", rb_ary_or, 1);
 
+    rb_define_method(rb_cArray, "max", rb_ary_max, -1);
+    rb_define_method(rb_cArray, "min", rb_ary_min, -1);
+
     rb_define_method(rb_cArray, "uniq", rb_ary_uniq, 0);
     rb_define_method(rb_cArray, "uniq!", rb_ary_uniq_bang, 0);
     rb_define_method(rb_cArray, "compact", rb_ary_compact, 0);
     rb_define_method(rb_cArray, "compact!", rb_ary_compact_bang, 0);
-    rb_define_method(rb_cArray, "flatten", rb_ary_flatten, 0);
-    rb_define_method(rb_cArray, "flatten!", rb_ary_flatten_bang, 0);
-    rb_define_method(rb_cArray, "nitems", rb_ary_nitems, 0);
-
-    id_cmp = rb_intern("<=>");
-    inspect_key = rb_intern("__inspect_key__");
+    rb_define_method(rb_cArray, "flatten", rb_ary_flatten, -1);
+    rb_define_method(rb_cArray, "flatten!", rb_ary_flatten_bang, -1);
+    rb_define_method(rb_cArray, "count", rb_ary_count, -1);
+    rb_define_method(rb_cArray, "shuffle!", rb_ary_shuffle_bang, -1);
+    rb_define_method(rb_cArray, "shuffle", rb_ary_shuffle, -1);
+    rb_define_method(rb_cArray, "sample", rb_ary_sample, -1);
+    rb_define_method(rb_cArray, "cycle", rb_ary_cycle, -1);
+    rb_define_method(rb_cArray, "permutation", rb_ary_permutation, -1);
+    rb_define_method(rb_cArray, "combination", rb_ary_combination, 1);
+    rb_define_method(rb_cArray, "repeated_permutation", rb_ary_repeated_permutation, 1);
+    rb_define_method(rb_cArray, "repeated_combination", rb_ary_repeated_combination, 1);
+    rb_define_method(rb_cArray, "product", rb_ary_product, -1);
+
+    rb_define_method(rb_cArray, "take", rb_ary_take, 1);
+    rb_define_method(rb_cArray, "take_while", rb_ary_take_while, 0);
+    rb_define_method(rb_cArray, "drop", rb_ary_drop, 1);
+    rb_define_method(rb_cArray, "drop_while", rb_ary_drop_while, 0);
+    rb_define_method(rb_cArray, "bsearch", rb_ary_bsearch, 0);
+    rb_define_method(rb_cArray, "bsearch_index", rb_ary_bsearch_index, 0);
+    rb_define_method(rb_cArray, "any?", rb_ary_any_p, -1);
+    rb_define_method(rb_cArray, "all?", rb_ary_all_p, -1);
+    rb_define_method(rb_cArray, "none?", rb_ary_none_p, -1);
+    rb_define_method(rb_cArray, "one?", rb_ary_one_p, -1);
+    rb_define_method(rb_cArray, "dig", rb_ary_dig, -1);
+    rb_define_method(rb_cArray, "sum", rb_ary_sum, -1);
+
+    id_random = rb_intern("random");
 }
diff --git a/ast.c b/ast.c
new file mode 100644
index 0000000000..d6a3073f58
--- /dev/null
+++ b/ast.c
@@ -0,0 +1,757 @@
+/* indent-tabs-mode: nil */
+#include "ruby.h"
+#include "ruby/encoding.h"
+#include "ruby/util.h"
+#include "internal.h"
+#include "node.h"
+#include "vm_core.h"
+#include "iseq.h"
+
+static VALUE rb_mAST;
+static VALUE rb_cNode;
+
+struct ASTNodeData {
+    rb_ast_t *ast;
+    NODE *node;
+};
+
+static void
+node_gc_mark(void *ptr)
+{
+    struct ASTNodeData *data = (struct ASTNodeData *)ptr;
+    rb_gc_mark((VALUE)data->ast);
+}
+
+static const rb_data_type_t rb_node_type = {
+    "AST/node",
+    {node_gc_mark, RUBY_TYPED_DEFAULT_FREE, 0,},
+    0, 0,
+    RUBY_TYPED_FREE_IMMEDIATELY,
+};
+
+static VALUE rb_ast_node_alloc(VALUE klass);
+
+static void
+setup_node(VALUE obj, rb_ast_t *ast, NODE *node)
+{
+    struct ASTNodeData *data;
+
+    TypedData_Get_Struct(obj, struct ASTNodeData, &rb_node_type, data);
+    data->ast = ast;
+    data->node = node;
+}
+
+static VALUE
+ast_new_internal(rb_ast_t *ast, NODE *node)
+{
+    VALUE obj;
+
+    obj = rb_ast_node_alloc(rb_cNode);
+    setup_node(obj, ast, node);
+
+    return obj;
+}
+
+static VALUE rb_ast_parse_str(VALUE str);
+static VALUE rb_ast_parse_file(VALUE path);
+static VALUE rb_ast_parse_array(VALUE array);
+
+static VALUE
+ast_parse_new(void)
+{
+    return rb_parser_set_context(rb_parser_new(), NULL, 0);
+}
+
+static VALUE
+ast_parse_done(rb_ast_t *ast)
+{
+    if (!ast->body.root) {
+        rb_ast_dispose(ast);
+        rb_exc_raise(GET_EC()->errinfo);
+    }
+
+    return ast_new_internal(ast, (NODE *)ast->body.root);
+}
+
+/*
+ *  call-seq:
+ *     RubyVM::AbstractSyntaxTree.parse(string) -> RubyVM::AbstractSyntaxTree::Node
+ *
+ *  Parses the given string into an abstract syntax tree,
+ *  returning the root node of that tree.
+ *
+ *  SyntaxError is raised if the given string is invalid syntax.
+ *
+ *    RubyVM::AbstractSyntaxTree.parse("x = 1 + 2")
+ *    # => #<RubyVM::AbstractSyntaxTree::Node(NODE_SCOPE(0) 1:0, 1:9): >
+ */
+static VALUE
+rb_ast_s_parse(VALUE module, VALUE str)
+{
+    return rb_ast_parse_str(str);
+}
+
+static VALUE
+rb_ast_parse_str(VALUE str)
+{
+    rb_ast_t *ast = 0;
+
+    str = rb_check_string_type(str);
+    ast = rb_parser_compile_string_path(ast_parse_new(), Qnil, str, 1);
+    return ast_parse_done(ast);
+}
+
+/*
+ *  call-seq:
+ *     RubyVM::AbstractSyntaxTree.parse_file(pathname) -> RubyVM::AbstractSyntaxTree::Node
+ *
+ *   Reads the file from <code>pathname</code>, then parses it like ::parse,
+ *   returning the root node of the abstract syntax tree.
+ *
+ *   SyntaxError is raised if <code>pathname</code>'s contents are not
+ *   valid Ruby syntax.
+ *
+ *     RubyVM::AbstractSyntaxTree.parse_file("my-app/app.rb")
+ *     # => #<RubyVM::AbstractSyntaxTree::Node(NODE_SCOPE(0) 1:0, 31:3): >
+ */
+static VALUE
+rb_ast_s_parse_file(VALUE module, VALUE path)
+{
+    return rb_ast_parse_file(path);
+}
+
+static VALUE
+rb_ast_parse_file(VALUE path)
+{
+    VALUE f;
+    rb_ast_t *ast = 0;
+    rb_encoding *enc = rb_utf8_encoding();
+
+    FilePathValue(path);
+    f = rb_file_open_str(path, "r");
+    rb_funcall(f, rb_intern("set_encoding"), 2, rb_enc_from_encoding(enc), rb_str_new_cstr("-"));
+    ast = rb_parser_compile_file_path(ast_parse_new(), Qnil, f, 1);
+    rb_io_close(f);
+    return ast_parse_done(ast);
+}
+
+static VALUE
+lex_array(VALUE array, int index)
+{
+    VALUE str = rb_ary_entry(array, index);
+    if (!NIL_P(str)) {
+        StringValue(str);
+        if (!rb_enc_asciicompat(rb_enc_get(str))) {
+            rb_raise(rb_eArgError, "invalid source encoding");
+        }
+    }
+    return str;
+}
+
+static VALUE
+rb_ast_parse_array(VALUE array)
+{
+    rb_ast_t *ast = 0;
+
+    array = rb_check_array_type(array);
+    ast = rb_parser_compile_generic(ast_parse_new(), lex_array, Qnil, array, 1);
+    return ast_parse_done(ast);
+}
+
+static VALUE node_children(rb_ast_t*, NODE*);
+
+static VALUE
+node_find(VALUE self, const int node_id)
+{
+    VALUE ary;
+    long i;
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    if (nd_node_id(data->node) == node_id) return self;
+
+    ary = node_children(data->ast, data->node);
+
+    for (i = 0; i < RARRAY_LEN(ary); i++) {
+        VALUE child = RARRAY_AREF(ary, i);
+
+        if (CLASS_OF(child) == rb_cNode) {
+            VALUE result = node_find(child, node_id);
+            if (RTEST(result)) return result;
+        }
+    }
+
+    return Qnil;
+}
+
+extern VALUE rb_e_script;
+
+static VALUE
+script_lines(VALUE path)
+{
+    VALUE hash, lines;
+    ID script_lines;
+    CONST_ID(script_lines, "SCRIPT_LINES__");
+    if (!rb_const_defined_at(rb_cObject, script_lines)) return Qnil;
+    hash = rb_const_get_at(rb_cObject, script_lines);
+    if (!RB_TYPE_P(hash, T_HASH)) return Qnil;
+    lines = rb_hash_lookup(hash, path);
+    if (!RB_TYPE_P(lines, T_ARRAY)) return Qnil;
+    return lines;
+}
+
+/*
+ *  call-seq:
+ *     RubyVM::AbstractSyntaxTree.of(proc)   -> RubyVM::AbstractSyntaxTree::Node
+ *     RubyVM::AbstractSyntaxTree.of(method) -> RubyVM::AbstractSyntaxTree::Node
+ *
+ *   Returns AST nodes of the given proc or method.
+ *
+ *     RubyVM::AbstractSyntaxTree.of(proc {1 + 2})
+ *     # => #<RubyVM::AbstractSyntaxTree::Node(NODE_SCOPE(0) 1:35, 1:42): >
+ *
+ *     def hello
+ *       puts "hello, world"
+ *     end
+ *
+ *     RubyVM::AbstractSyntaxTree.of(method(:hello))
+ *     # => #<RubyVM::AbstractSyntaxTree::Node(NODE_SCOPE(0) 1:0, 3:3): >
+ */
+static VALUE
+rb_ast_s_of(VALUE module, VALUE body)
+{
+    VALUE path, node, lines;
+    int node_id;
+    const rb_iseq_t *iseq = NULL;
+
+    if (rb_obj_is_proc(body)) {
+        iseq = vm_proc_iseq(body);
+
+        if (!rb_obj_is_iseq((VALUE)iseq)) {
+            iseq = NULL;
+        }
+    }
+    else {
+        iseq = rb_method_iseq(body);
+    }
+
+    if (!iseq) return Qnil;
+
+    path = rb_iseq_path(iseq);
+    node_id = iseq->body->location.node_id;
+    if (!NIL_P(lines = script_lines(path))) {
+        node = rb_ast_parse_array(lines);
+    }
+    else if (RSTRING_LEN(path) == 2 && memcmp(RSTRING_PTR(path), "-e", 2) == 0) {
+        node = rb_ast_parse_str(rb_e_script);
+    }
+    else {
+        node = rb_ast_parse_file(path);
+    }
+
+    return node_find(node, node_id);
+}
+
+static VALUE
+rb_ast_node_alloc(VALUE klass)
+{
+    struct ASTNodeData *data;
+    VALUE obj = TypedData_Make_Struct(klass, struct ASTNodeData, &rb_node_type, data);
+
+    return obj;
+}
+
+static const char*
+node_type_to_str(const NODE *node)
+{
+    return (ruby_node_name(nd_type(node)) + rb_strlen_lit("NODE_"));
+}
+
+/*
+ *  call-seq:
+ *     node.type -> symbol
+ *
+ *  Returns the type of this node as a symbol.
+ *
+ *    root = RubyVM::AbstractSyntaxTree.parse("x = 1 + 2")
+ *    root.type # => :SCOPE
+ *    call = root.children[2]
+ *    call.type # => :OPCALL
+ */
+static VALUE
+rb_ast_node_type(VALUE self)
+{
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    return rb_sym_intern_ascii_cstr(node_type_to_str(data->node));
+}
+
+#define NEW_CHILD(ast, node) node ? ast_new_internal(ast, node) : Qnil
+
+static VALUE
+rb_ary_new_from_node_args(rb_ast_t *ast, long n, ...)
+{
+    va_list ar;
+    VALUE ary;
+    long i;
+
+    ary = rb_ary_new2(n);
+
+    va_start(ar, n);
+    for (i=0; i<n; i++) {
+        NODE *node;
+        node = va_arg(ar, NODE *);
+        rb_ary_push(ary, NEW_CHILD(ast, node));
+    }
+    va_end(ar);
+    return ary;
+}
+
+static VALUE
+dump_block(rb_ast_t *ast, NODE *node)
+{
+    VALUE ary = rb_ary_new();
+    do {
+        rb_ary_push(ary, NEW_CHILD(ast, node->nd_head));
+    } while (node->nd_next &&
+        nd_type(node->nd_next) == NODE_BLOCK &&
+        (node = node->nd_next, 1));
+    if (node->nd_next) {
+        rb_ary_push(ary, NEW_CHILD(ast, node->nd_next));
+    }
+
+    return ary;
+}
+
+static VALUE
+dump_array(rb_ast_t *ast, NODE *node)
+{
+    VALUE ary = rb_ary_new();
+    rb_ary_push(ary, NEW_CHILD(ast, node->nd_head));
+
+    while (node->nd_next && nd_type(node->nd_next) == NODE_ARRAY) {
+        node = node->nd_next;
+        rb_ary_push(ary, NEW_CHILD(ast, node->nd_head));
+    }
+    rb_ary_push(ary, NEW_CHILD(ast, node->nd_next));
+
+    return ary;
+}
+
+static VALUE
+var_name(ID id)
+{
+    if (!id) return Qnil;
+    if (!rb_id2str(id)) return Qnil;
+    return ID2SYM(id);
+}
+
+static VALUE
+node_children(rb_ast_t *ast, NODE *node)
+{
+    char name[DECIMAL_SIZE_OF_BITS(sizeof(long) * CHAR_BIT) + 2]; /* including '$' */
+
+    enum node_type type = nd_type(node);
+    switch (type) {
+      case NODE_BLOCK:
+        return dump_block(ast, node);
+      case NODE_IF:
+        return rb_ary_new_from_node_args(ast, 3, node->nd_cond, node->nd_body, node->nd_else);
+      case NODE_UNLESS:
+        return rb_ary_new_from_node_args(ast, 3, node->nd_cond, node->nd_body, node->nd_else);
+      case NODE_CASE:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body);
+      case NODE_CASE2:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body);
+      case NODE_WHEN:
+        return rb_ary_new_from_node_args(ast, 3, node->nd_head, node->nd_body, node->nd_next);
+      case NODE_WHILE:
+        goto loop;
+      case NODE_UNTIL:
+      loop:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_cond, node->nd_body);
+      case NODE_ITER:
+      case NODE_FOR:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_iter, node->nd_body);
+      case NODE_FOR_MASGN:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_var);
+      case NODE_BREAK:
+        goto jump;
+      case NODE_NEXT:
+        goto jump;
+      case NODE_RETURN:
+      jump:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_stts);
+      case NODE_REDO:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_RETRY:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_BEGIN:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_body);
+      case NODE_RESCUE:
+        return rb_ary_new_from_node_args(ast, 3, node->nd_head, node->nd_resq, node->nd_else);
+      case NODE_RESBODY:
+        return rb_ary_new_from_node_args(ast, 3, node->nd_args, node->nd_body, node->nd_head);
+      case NODE_ENSURE:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_ensr);
+      case NODE_AND:
+        goto andor;
+      case NODE_OR:
+      andor:
+        {
+            VALUE ary = rb_ary_new();
+
+            while (1) {
+                rb_ary_push(ary, NEW_CHILD(ast, node->nd_1st));
+                if (!node->nd_2nd || nd_type(node->nd_2nd) != (int)type)
+                    break;
+                node = node->nd_2nd;
+            }
+            rb_ary_push(ary, NEW_CHILD(ast, node->nd_2nd));
+            return ary;
+        }
+      case NODE_MASGN:
+        if (NODE_NAMED_REST_P(node->nd_args)) {
+            return rb_ary_new_from_node_args(ast, 3, node->nd_value, node->nd_head, node->nd_args);
+        }
+        return rb_ary_new_from_node_args(ast, 2, node->nd_value, node->nd_head);
+      case NODE_LASGN:
+        goto asgn;
+      case NODE_DASGN:
+        goto asgn;
+      case NODE_DASGN_CURR:
+        goto asgn;
+      case NODE_IASGN:
+        goto asgn;
+      case NODE_CVASGN:
+      asgn:
+        if (NODE_REQUIRED_KEYWORD_P(node)) {
+            return rb_ary_new_from_args(1, var_name(node->nd_vid));
+        }
+        return rb_ary_new_from_args(2, var_name(node->nd_vid), NEW_CHILD(ast, node->nd_value));
+      case NODE_GASGN:
+        goto asgn;
+      case NODE_CDECL:
+        if (node->nd_vid) {
+            return rb_ary_new_from_args(2, ID2SYM(node->nd_vid), NEW_CHILD(ast, node->nd_value));
+        }
+        return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_else), ID2SYM(node->nd_else->nd_mid), NEW_CHILD(ast, node->nd_value));
+      case NODE_OP_ASGN1:
+        return rb_ary_new_from_args(4, NEW_CHILD(ast, node->nd_recv),
+                                    ID2SYM(node->nd_mid),
+                                    NEW_CHILD(ast, node->nd_args->nd_head),
+                                    NEW_CHILD(ast, node->nd_args->nd_body));
+      case NODE_OP_ASGN2:
+        return rb_ary_new_from_args(4, NEW_CHILD(ast, node->nd_recv),
+                                    node->nd_next->nd_aid ? Qtrue : Qfalse,
+                                    ID2SYM(node->nd_next->nd_vid),
+                                    NEW_CHILD(ast, node->nd_value));
+      case NODE_OP_ASGN_AND:
+        return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_head), ID2SYM(idANDOP),
+                                    NEW_CHILD(ast, node->nd_value));
+      case NODE_OP_ASGN_OR:
+        return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_head), ID2SYM(idOROP),
+                                    NEW_CHILD(ast, node->nd_value));
+      case NODE_OP_CDECL:
+        return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_head),
+                                    ID2SYM(node->nd_aid),
+                                    NEW_CHILD(ast, node->nd_value));
+      case NODE_CALL:
+      case NODE_OPCALL:
+      case NODE_QCALL:
+        return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_recv),
+                                    ID2SYM(node->nd_mid),
+                                    NEW_CHILD(ast, node->nd_args));
+      case NODE_FCALL:
+        return rb_ary_new_from_args(2, ID2SYM(node->nd_mid),
+                                    NEW_CHILD(ast, node->nd_args));
+      case NODE_VCALL:
+        return rb_ary_new_from_args(1, ID2SYM(node->nd_mid));
+      case NODE_SUPER:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_args);
+      case NODE_ZSUPER:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_ARRAY:
+        goto ary;
+      case NODE_VALUES:
+      ary:
+        return dump_array(ast, node);
+      case NODE_ZARRAY:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_HASH:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_head);
+      case NODE_YIELD:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_head);
+      case NODE_LVAR:
+      case NODE_DVAR:
+        return rb_ary_new_from_args(1, var_name(node->nd_vid));
+      case NODE_IVAR:
+      case NODE_CONST:
+      case NODE_CVAR:
+      case NODE_GVAR:
+        return rb_ary_new_from_args(1, ID2SYM(node->nd_vid));
+      case NODE_NTH_REF:
+        snprintf(name, sizeof(name), "$%ld", node->nd_nth);
+        return rb_ary_new_from_args(1, ID2SYM(rb_intern(name)));
+      case NODE_BACK_REF:
+        name[0] = '$';
+        name[1] = (char)node->nd_nth;
+        name[2] = '\0';
+        return rb_ary_new_from_args(1, ID2SYM(rb_intern(name)));
+      case NODE_MATCH:
+        goto lit;
+      case NODE_MATCH2:
+        if (node->nd_args) {
+            return rb_ary_new_from_node_args(ast, 3, node->nd_recv, node->nd_value, node->nd_args);
+        }
+        return rb_ary_new_from_node_args(ast, 2, node->nd_recv, node->nd_value);
+      case NODE_MATCH3:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_recv, node->nd_value);
+      case NODE_LIT:
+        goto lit;
+      case NODE_STR:
+        goto lit;
+      case NODE_XSTR:
+      lit:
+        return rb_ary_new_from_args(1, node->nd_lit);
+      case NODE_ONCE:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_body);
+      case NODE_DSTR:
+        goto dlit;
+      case NODE_DXSTR:
+        goto dlit;
+      case NODE_DREGX:
+        goto dlit;
+      case NODE_DSYM:
+      dlit:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_next->nd_head, node->nd_next->nd_next);
+      case NODE_EVSTR:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_body);
+      case NODE_ARGSCAT:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body);
+      case NODE_ARGSPUSH:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body);
+      case NODE_SPLAT:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_head);
+      case NODE_BLOCK_PASS:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_head, node->nd_body);
+      case NODE_DEFN:
+        return rb_ary_new_from_args(2, ID2SYM(node->nd_mid), NEW_CHILD(ast, node->nd_defn));
+      case NODE_DEFS:
+        return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_recv), ID2SYM(node->nd_mid), NEW_CHILD(ast, node->nd_defn));
+      case NODE_ALIAS:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_1st, node->nd_2nd);
+      case NODE_VALIAS:
+        return rb_ary_new_from_args(2, ID2SYM(node->nd_alias), ID2SYM(node->nd_orig));
+      case NODE_UNDEF:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_undef);
+      case NODE_CLASS:
+        return rb_ary_new_from_node_args(ast, 3, node->nd_cpath, node->nd_super, node->nd_body);
+      case NODE_MODULE:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_cpath, node->nd_body);
+      case NODE_SCLASS:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_recv, node->nd_body);
+      case NODE_COLON2:
+        return rb_ary_new_from_args(2, NEW_CHILD(ast, node->nd_head), ID2SYM(node->nd_mid));
+      case NODE_COLON3:
+        return rb_ary_new_from_args(1, ID2SYM(node->nd_mid));
+      case NODE_DOT2:
+        goto dot;
+      case NODE_DOT3:
+        goto dot;
+      case NODE_FLIP2:
+        goto dot;
+      case NODE_FLIP3:
+      dot:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_beg, node->nd_end);
+      case NODE_SELF:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_NIL:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_TRUE:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_FALSE:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_ERRINFO:
+        return rb_ary_new_from_node_args(ast, 0);
+      case NODE_DEFINED:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_head);
+      case NODE_POSTEXE:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_body);
+      case NODE_ATTRASGN:
+        return rb_ary_new_from_args(3, NEW_CHILD(ast, node->nd_recv), ID2SYM(node->nd_mid), NEW_CHILD(ast, node->nd_args));
+      case NODE_LAMBDA:
+        return rb_ary_new_from_node_args(ast, 1, node->nd_body);
+      case NODE_OPT_ARG:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_body, node->nd_next);
+      case NODE_KW_ARG:
+        return rb_ary_new_from_node_args(ast, 2, node->nd_body, node->nd_next);
+      case NODE_POSTARG:
+        if (NODE_NAMED_REST_P(node->nd_1st)) {
+            return rb_ary_new_from_node_args(ast, 2, node->nd_1st, node->nd_2nd);
+        }
+        return rb_ary_new_from_node_args(ast, 1, node->nd_2nd);
+      case NODE_ARGS:
+        {
+            struct rb_args_info *ainfo = node->nd_ainfo;
+            return rb_ary_new_from_args(10,
+                                        INT2NUM(ainfo->pre_args_num),
+                                        NEW_CHILD(ast, ainfo->pre_init),
+                                        NEW_CHILD(ast, ainfo->opt_args),
+                                        var_name(ainfo->first_post_arg),
+                                        INT2NUM(ainfo->post_args_num),
+                                        NEW_CHILD(ast, ainfo->post_init),
+                                        var_name(ainfo->rest_arg),
+                                        NEW_CHILD(ast, ainfo->kw_args),
+                                        NEW_CHILD(ast, ainfo->kw_rest_arg),
+                                        var_name(ainfo->block_arg));
+        }
+      case NODE_SCOPE:
+        {
+            ID *tbl = node->nd_tbl;
+            int i, size = tbl ? (int)*tbl++ : 0;
+            VALUE locals = rb_ary_new_capa(size);
+            for (i = 0; i < size; i++) {
+                rb_ary_push(locals, var_name(tbl[i]));
+            }
+            return rb_ary_new_from_args(3, locals, NEW_CHILD(ast, node->nd_args), NEW_CHILD(ast, node->nd_body));
+        }
+      case NODE_ARGS_AUX:
+      case NODE_LAST:
+        break;
+    }
+
+    rb_bug("node_children: unknown node: %s", ruby_node_name(type));
+}
+
+/*
+ *  call-seq:
+ *     node.children -> array
+ *
+ *  Returns AST nodes under this one.  Each kind of node
+ *  has different children, depending on what kind of node it is.
+ *
+ *  The returned array may contain other nodes or <code>nil</code>.
+ */
+static VALUE
+rb_ast_node_children(VALUE self)
+{
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    return node_children(data->ast, data->node);
+}
+
+/*
+ *  call-seq:
+ *     node.first_lineno -> integer
+ *
+ *  The line number in the source code where this AST's text began.
+ */
+static VALUE
+rb_ast_node_first_lineno(VALUE self)
+{
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    return INT2NUM(nd_first_lineno(data->node));
+}
+
+/*
+ *  call-seq:
+ *     node.first_column -> integer
+ *
+ *  The column number in the source code where this AST's text began.
+ */
+static VALUE
+rb_ast_node_first_column(VALUE self)
+{
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    return INT2NUM(nd_first_column(data->node));
+}
+
+/*
+ *  call-seq:
+ *     node.last_lineno -> integer
+ *
+ *  The line number in the source code where this AST's text ended.
+ */
+static VALUE
+rb_ast_node_last_lineno(VALUE self)
+{
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    return INT2NUM(nd_last_lineno(data->node));
+}
+
+/*
+ *  call-seq:
+ *     node.last_column -> integer
+ *
+ *  The column number in the source code where this AST's text ended.
+ */
+static VALUE
+rb_ast_node_last_column(VALUE self)
+{
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    return INT2NUM(nd_last_column(data->node));
+}
+
+/*
+ *  call-seq:
+ *     node.inspect -> string
+ *
+ *  Returns debugging information about this node as a string.
+ */
+static VALUE
+rb_ast_node_inspect(VALUE self)
+{
+    VALUE str;
+    VALUE cname;
+    struct ASTNodeData *data;
+    TypedData_Get_Struct(self, struct ASTNodeData, &rb_node_type, data);
+
+    cname = rb_class_path(rb_obj_class(self));
+    str = rb_str_new2("#<");
+
+    rb_str_append(str, cname);
+    rb_str_cat2(str, "(");
+    rb_str_catf(str, "%s(%d) %d:%d, %d:%d", node_type_to_str(data->node), nd_type(data->node), nd_first_lineno(data->node), nd_first_column(data->node), nd_last_lineno(data->node), nd_last_column(data->node));
+    rb_str_cat2(str, "): >");
+
+    return str;
+}
+
+void
+Init_ast(void)
+{
+    /*
+     * AbstractSyntaxTree provides methods to parse Ruby code into
+     * abstract syntax trees. The nodes in the tree
+     * are instances of RubyVM::AbstractSyntaxTree::Node.
+     */
+    rb_mAST = rb_define_module_under(rb_cRubyVM, "AbstractSyntaxTree");
+    /*
+     * RubyVM::AbstractSyntaxTree::Node instances are created by parse methods in
+     * RubyVM::AbstractSyntaxTree.
+     */
+    rb_cNode = rb_define_class_under(rb_mAST, "Node", rb_cObject);
+
+    rb_undef_alloc_func(rb_cNode);
+    rb_define_singleton_method(rb_mAST, "parse", rb_ast_s_parse, 1);
+    rb_define_singleton_method(rb_mAST, "parse_file", rb_ast_s_parse_file, 1);
+    rb_define_singleton_method(rb_mAST, "of", rb_ast_s_of, 1);
+    rb_define_method(rb_cNode, "type", rb_ast_node_type, 0);
+    rb_define_method(rb_cNode, "first_lineno", rb_ast_node_first_lineno, 0);
+    rb_define_method(rb_cNode, "first_column", rb_ast_node_first_column, 0);
+    rb_define_method(rb_cNode, "last_lineno", rb_ast_node_last_lineno, 0);
+    rb_define_method(rb_cNode, "last_column", rb_ast_node_last_column, 0);
+    rb_define_method(rb_cNode, "children", rb_ast_node_children, 0);
+    rb_define_method(rb_cNode, "inspect", rb_ast_node_inspect, 0);
+}
diff --git a/basictest/runner.rb b/basictest/runner.rb
new file mode 100755
index 0000000000..0f398e7acc
--- /dev/null
+++ b/basictest/runner.rb
@@ -0,0 +1,33 @@
+#! ./miniruby
+
+exit if defined?(CROSS_COMPILING) and CROSS_COMPILING
+ruby = ENV["RUBY"]
+unless ruby
+  load './rbconfig.rb'
+  ruby = "./#{RbConfig::CONFIG['ruby_install_name']}#{RbConfig::CONFIG['EXEEXT']}"
+end
+unless File.exist? ruby
+  print "#{ruby} is not found.\n"
+  print "Try `make' first, then `make test', please.\n"
+  exit false
+end
+ARGV[0] and opt = ARGV[0][/\A--run-opt=(.*)/, 1] and ARGV.shift
+
+$stderr.reopen($stdout)
+error = ''
+
+srcdir = File.expand_path('..', File.dirname(__FILE__))
+if env = ENV["RUBYOPT"]
+  ENV["RUBYOPT"] = env + " -W1"
+end
+`#{ruby} #{opt} -W1 #{srcdir}/basictest/test.rb #{ARGV.join(' ')}`.each_line do |line|
+  if line =~ /^end of test/
+    print "\ntest succeeded\n"
+    exit true
+  end
+  error << line if %r:^(basictest/test.rb|not): =~ line
+end
+puts
+print error
+print "test failed\n"
+exit false
diff --git a/basictest/test.rb b/basictest/test.rb
new file mode 100755
index 0000000000..25d52ca1ef
--- /dev/null
+++ b/basictest/test.rb
@@ -0,0 +1,2367 @@
+#! /usr/bin/env ruby
+# -*- coding: us-ascii -*-
+
+$testnum=0
+$ntest=0
+$failed = 0
+class Progress
+  def initialize
+    @color = nil
+    @tty = nil
+    @quiet = nil
+    @verbose = nil
+    ARGV.each do |arg|
+      case arg
+      when /\A--color(?:=(?:always|(auto)|(never)|(.*)))?\z/
+        warn "unknown --color argument: #$3" if $3
+        @color = $1 ? nil : !$2
+      when /\A--tty(=(?:yes|(no)|(.*)))?\z/
+        warn "unknown --tty argument: #$3" if $3
+        @tty = !$1 || !$2
+        true
+      when /\A-(q|-quiet)\z/
+        @quiet = true
+      when /\A-(v|-verbose)\z/
+        @verbose = true
+      end
+    end
+    @tty = STDERR.tty? && !STDOUT.tty? && /dumb/ !~ ENV["TERM"] if @tty.nil?
+    @eol = @tty && !@verbose ? "\r\e[K\r" : "\n"
+    case @color
+    when nil
+      @color = @tty
+    end
+    if @color
+      # dircolors-like style
+      colors = (colors = ENV['TEST_COLORS']) ? Hash[colors.scan(/(\w+)=([^:\n]*)/)] : {}
+      begin
+        File.read(File.join(__dir__, "../test/colors")).scan(/(\w+)=([^:\n]*)/) do |n, c|
+          colors[n] ||= c
+        end
+      rescue
+      end
+      @passed = "\e[;#{colors["pass"] || "32"}m"
+      @failed = "\e[;#{colors["fail"] || "31"}m"
+      @reset = "\e[m"
+    else
+      @passed = @failed = @reset = ""
+    end
+    extend(Rotator) if @tty
+  end
+
+  def passed_string
+    "."
+  end
+  def failed_string
+    "#{@failed}F#{@reset}"
+  end
+  def init_string
+  end
+  def finish_string
+    if @quiet
+      @eol
+    else
+      "#{@passed}#{@ok ? 'OK' : ''} #{$testnum}#{@reset}#{@eol}"
+    end
+  end
+  def pass
+    STDERR.print passed_string
+  end
+  def fail
+    @ok = false
+    STDERR.print failed_string
+  end
+  def init
+    @ok = true
+    STDERR.print init_string
+  end
+  def finish
+    STDERR.print finish_string
+  end
+
+  module Rotator
+    ROTATOR = %w[- \\ | /]
+    BS = "\b" * ROTATOR[0].size
+    def passed_string
+      "#{BS}#{ROTATOR[(@count += 1) % ROTATOR.size]}"
+    end
+    def failed_string
+      "#{BS}#{super}#{ROTATOR[@count % ROTATOR.size]}"
+    end
+    def init_string
+      @count = 0
+      " "
+    end
+    def finish_string
+      s = "#{BS}#{' ' * BS.size}#{BS}#{super}"
+      s.gsub!(/\n/, "\r\e[2K\r") if @quiet
+      s
+    end
+  end
+end
+PROGRESS = Progress.new
+
+def test_check(what)
+  unless $ntest.zero?
+    PROGRESS.finish
+  end
+  STDERR.print "#{$0}:#{what} "
+  PROGRESS.init
+  $what = what
+  $testnum = 0
+end
+
+def test_ok(cond,n=1)
+  $testnum+=1
+  $ntest+=1
+  where = (st = caller(n)) ? st[0] : "caller error! (n=#{n}, trace=#{caller(0).join(', ')}"
+  if cond
+    PROGRESS.pass
+    printf "ok %d (%s)\n", $testnum, where
+  else
+    PROGRESS.fail
+    printf "not ok %s %d -- %s\n", $what, $testnum, where
+    $failed+=1
+  end
+  STDOUT.flush
+  STDERR.flush
+end
+
+# make sure conditional operators work
+
+test_check "assignment"
+
+a=[]; a[0] ||= "bar";
+test_ok(a[0] == "bar")
+h={}; h["foo"] ||= "bar";
+test_ok(h["foo"] == "bar")
+
+aa = 5
+aa ||= 25
+test_ok(aa == 5)
+bb ||= 25
+test_ok(bb == 25)
+cc &&=33
+test_ok(cc == nil)
+cc = 5
+cc &&=44
+test_ok(cc == 44)
+
+a = nil; test_ok(a == nil)
+a = 1; test_ok(a == 1)
+a = []; test_ok(a == [])
+a = [1]; test_ok(a == [1])
+a = [nil]; test_ok(a == [nil])
+a = [[]]; test_ok(a == [[]])
+a = [1,2]; test_ok(a == [1,2])
+a = [*[]]; test_ok(a == [])
+a = [*[1]]; test_ok(a == [1])
+a = [*[1,2]]; test_ok(a == [1,2])
+
+a = *[]; test_ok(a == [])
+a = *[1]; test_ok(a == [1])
+a = *[nil]; test_ok(a == [nil])
+a = *[[]]; test_ok(a == [[]])
+a = *[1,2]; test_ok(a == [1,2])
+a = *[*[]]; test_ok(a == [])
+a = *[*[1]]; test_ok(a == [1])
+a = *[*[1,2]]; test_ok(a == [1,2])
+
+a, = nil; test_ok(a == nil)
+a, = 1; test_ok(a == 1)
+a, = []; test_ok(a == nil)
+a, = [1]; test_ok(a == 1)
+a, = [nil]; test_ok(a == nil)
+a, = [[]]; test_ok(a == [])
+a, = 1,2; test_ok(a == 1)
+a, = [1,2]; test_ok(a == 1)
+a, = [*[]]; test_ok(a == nil)
+a, = [*[1]]; test_ok(a == 1)
+a, = *[1,2]; test_ok(a == 1)
+a, = [*[1,2]]; test_ok(a == 1)
+
+a, = *[]; test_ok(a == nil)
+a, = *[1]; test_ok(a == 1)
+a, = *[nil]; test_ok(a == nil)
+a, = *[[]]; test_ok(a == [])
+a, = *[1,2]; test_ok(a == 1)
+a, = *[*[]]; test_ok(a == nil)
+a, = *[*[1]]; test_ok(a == 1)
+a, = *[*[1,2]]; test_ok(a == 1)
+
+*a = nil; test_ok(a == [nil])
+*a = 1; test_ok(a == [1])
+*a = []; test_ok(a == [])
+*a = [1]; test_ok(a == [1])
+*a = [nil]; test_ok(a == [nil])
+*a = [[]]; test_ok(a == [[]])
+*a = [1,2]; test_ok(a == [1,2])
+*a = [*[]]; test_ok(a == [])
+*a = [*[1]]; test_ok(a == [1])
+*a = [*[1,2]]; test_ok(a == [1,2])
+
+*a = *[]; test_ok(a == [])
+*a = *[1]; test_ok(a == [1])
+*a = *[nil]; test_ok(a == [nil])
+*a = *[[]]; test_ok(a == [[]])
+*a = *[1,2]; test_ok(a == [1,2])
+*a = *[*[]]; test_ok(a == [])
+*a = *[*[1]]; test_ok(a == [1])
+*a = *[*[1,2]]; test_ok(a == [1,2])
+
+a,b,*c = nil; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = 1; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = []; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = [1]; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = [nil]; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = [[]]; test_ok([a,b,c] == [[],nil,[]])
+a,b,*c = [1,2]; test_ok([a,b,c] == [1,2,[]])
+a,b,*c = [*[]]; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = [*[1]]; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = [*[1,2]]; test_ok([a,b,c] == [1,2,[]])
+
+a,b,*c = *[]; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = *[1]; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = *[nil]; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = *[[]]; test_ok([a,b,c] == [[],nil,[]])
+a,b,*c = *[1,2]; test_ok([a,b,c] == [1,2,[]])
+a,b,*c = *[*[]]; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = *[*[1]]; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = *[*[1,2]]; test_ok([a,b,c] == [1,2,[]])
+
+def f; yield nil; end; f {|a| test_ok(a == nil)}
+def f; yield 1; end; f {|a| test_ok(a == 1)}
+def f; yield []; end; f {|a| test_ok(a == [])}
+def f; yield [1]; end; f {|a| test_ok(a == [1])}
+def f; yield [nil]; end; f {|a| test_ok(a == [nil])}
+def f; yield [[]]; end; f {|a| test_ok(a == [[]])}
+def f; yield [*[]]; end; f {|a| test_ok(a == [])}
+def f; yield [*[1]]; end; f {|a| test_ok(a == [1])}
+def f; yield [*[1,2]]; end; f {|a| test_ok(a == [1,2])}
+def f; yield *[]; end; f {|a| test_ok(a == nil)}
+def f; yield *[1]; end; f {|a| test_ok(a == 1)}
+def f; yield *[nil]; end; f {|a| test_ok(a == nil)}
+def f; yield *[[]]; end; f {|a| test_ok(a == [])}
+def f; yield *[*[]]; end; f {|a| test_ok(a == nil)}
+def f; yield *[*[1]]; end; f {|a| test_ok(a == 1)}
+def f; yield *[*[1,2]]; end; f {|a| test_ok(a == 1)}
+
+def f; yield; end; f {|a,| test_ok(a == nil)}
+def f; yield nil; end; f {|a,| test_ok(a == nil)}
+def f; yield 1; end; f {|a,| test_ok(a == 1)}
+def f; yield []; end; f {|a,| test_ok(a == nil)}
+def f; yield [1]; end; f {|a,| test_ok(a == 1)}
+def f; yield [nil]; end; f {|a,| test_ok(a == nil)}
+def f; yield [[]]; end; f {|a,| test_ok(a == [])}
+def f; yield [*[]]; end; f {|a,| test_ok(a == nil)}
+def f; yield [*[1]]; end; f {|a,| test_ok(a == 1)}
+def f; yield [*[1,2]]; end; f {|a,| test_ok(a == 1)}
+
+def f; yield *[]; end; f {|a,| test_ok(a == nil)}
+def f; yield *[1]; end; f {|a,| test_ok(a == 1)}
+def f; yield *[nil]; end; f {|a,| test_ok(a == nil)}
+def f; yield *[[]]; end; f {|a,| test_ok(a == nil)}
+def f; yield *[*[]]; end; f {|a,| test_ok(a == nil)}
+def f; yield *[*[1]]; end; f {|a,| test_ok(a == 1)}
+def f; yield *[*[1,2]]; end; f {|a,| test_ok(a == 1)}
+
+def f; yield; end; f {|*a| test_ok(a == [])}
+def f; yield nil; end; f {|*a| test_ok(a == [nil])}
+def f; yield 1; end; f {|*a| test_ok(a == [1])}
+def f; yield []; end; f {|*a| test_ok(a == [[]])}
+def f; yield [1]; end; f {|*a| test_ok(a == [[1]])}
+def f; yield [nil]; end; f {|*a| test_ok(a == [[nil]])}
+def f; yield [[]]; end; f {|*a| test_ok(a == [[[]]])}
+def f; yield [1,2]; end; f {|*a| test_ok(a == [[1,2]])}
+def f; yield [*[]]; end; f {|*a| test_ok(a == [[]])}
+def f; yield [*[1]]; end; f {|*a| test_ok(a == [[1]])}
+def f; yield [*[1,2]]; end; f {|*a| test_ok(a == [[1,2]])}
+
+def f; yield *[]; end; f {|*a| test_ok(a == [])}
+def f; yield *[1]; end; f {|*a| test_ok(a == [1])}
+def f; yield *[nil]; end; f {|*a| test_ok(a == [nil])}
+def f; yield *[[]]; end; f {|*a| test_ok(a == [[]])}
+def f; yield *[*[]]; end; f {|*a| test_ok(a == [])}
+def f; yield *[*[1]]; end; f {|*a| test_ok(a == [1])}
+def f; yield *[*[1,2]]; end; f {|*a| test_ok(a == [1,2])}
+
+def f; yield; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield nil; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield 1; end; f {|a,b,*c| test_ok([a,b,c] == [1,nil,[]])}
+def f; yield []; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield [1]; end; f {|a,b,*c| test_ok([a,b,c] == [1,nil,[]])}
+def f; yield [nil]; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield [[]]; end; f {|a,b,*c| test_ok([a,b,c] == [[],nil,[]])}
+def f; yield [*[]]; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield [*[1]]; end; f {|a,b,*c| test_ok([a,b,c] == [1,nil,[]])}
+def f; yield [*[1,2]]; end; f {|a,b,*c| test_ok([a,b,c] == [1,2,[]])}
+
+def f; yield *[]; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield *[1]; end; f {|a,b,*c| test_ok([a,b,c] == [1,nil,[]])}
+def f; yield *[nil]; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield *[[]]; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield *[*[]]; end; f {|a,b,*c| test_ok([a,b,c] == [nil,nil,[]])}
+def f; yield *[*[1]]; end; f {|a,b,*c| test_ok([a,b,c] == [1,nil,[]])}
+def f; yield *[*[1,2]]; end; f {|a,b,*c| test_ok([a,b,c] == [1,2,[]])}
+
+def r; return; end; a = r(); test_ok(a == nil)
+def r; return nil; end; a = r(); test_ok(a == nil)
+def r; return 1; end; a = r(); test_ok(a == 1)
+def r; return []; end; a = r(); test_ok(a == [])
+def r; return [1]; end; a = r(); test_ok(a == [1])
+def r; return [nil]; end; a = r(); test_ok(a == [nil])
+def r; return [[]]; end; a = r(); test_ok(a == [[]])
+def r; return [*[]]; end; a = r(); test_ok(a == [])
+def r; return [*[1]]; end; a = r(); test_ok(a == [1])
+def r; return [*[1,2]]; end; a = r(); test_ok(a == [1,2])
+
+def r; return *[]; end; a = r(); test_ok(a == [])
+def r; return *[1]; end; a = r(); test_ok(a == [1])
+def r; return *[nil]; end; a = r(); test_ok(a == [nil])
+def r; return *[[]]; end; a = r(); test_ok(a == [[]])
+def r; return *[*[]]; end; a = r(); test_ok(a == [])
+def r; return *[*[1]]; end; a = r(); test_ok(a == [1])
+def r; return *[*[1,2]]; end; a = r(); test_ok(a == [1,2])
+
+def r; return *[[]]; end; a = *r(); test_ok(a == [[]])
+def r; return *[*[1,2]]; end; a = *r(); test_ok(a == [1,2])
+
+def r; return; end; *a = r(); test_ok(a == [nil])
+def r; return nil; end; *a = r(); test_ok(a == [nil])
+def r; return 1; end; *a = r(); test_ok(a == [1])
+def r; return []; end; *a = r(); test_ok(a == [])
+def r; return [1]; end; *a = r(); test_ok(a == [1])
+def r; return [nil]; end; *a = r(); test_ok(a == [nil])
+def r; return [[]]; end; *a = r(); test_ok(a == [[]])
+def r; return [1,2]; end; *a = r(); test_ok(a == [1,2])
+def r; return [*[]]; end; *a = r(); test_ok(a == [])
+def r; return [*[1]]; end; *a = r(); test_ok(a == [1])
+def r; return [*[1,2]]; end; *a = r(); test_ok(a == [1,2])
+
+def r; return *[]; end; *a = r(); test_ok(a == [])
+def r; return *[1]; end; *a = r(); test_ok(a == [1])
+def r; return *[nil]; end; *a = r(); test_ok(a == [nil])
+def r; return *[[]]; end; *a = r(); test_ok(a == [[]])
+def r; return *[1,2]; end; *a = r(); test_ok(a == [1,2])
+def r; return *[*[]]; end; *a = r(); test_ok(a == [])
+def r; return *[*[1]]; end; *a = r(); test_ok(a == [1])
+def r; return *[*[1,2]]; end; *a = r(); test_ok(a == [1,2])
+
+def r; return *[[]]; end; *a = *r(); test_ok(a == [[]])
+def r; return *[1,2]; end; *a = *r(); test_ok(a == [1,2])
+def r; return *[*[1,2]]; end; *a = *r(); test_ok(a == [1,2])
+
+def r; return; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return nil; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return 1; end; a,b,*c = r(); test_ok([a,b,c] == [1,nil,[]])
+def r; return []; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return [1]; end; a,b,*c = r(); test_ok([a,b,c] == [1,nil,[]])
+def r; return [nil]; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return [[]]; end; a,b,*c = r(); test_ok([a,b,c] == [[],nil,[]])
+def r; return [1,2]; end; a,b,*c = r(); test_ok([a,b,c] == [1,2,[]])
+def r; return [*[]]; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return [*[1]]; end; a,b,*c = r(); test_ok([a,b,c] == [1,nil,[]])
+def r; return [*[1,2]]; end; a,b,*c = r(); test_ok([a,b,c] == [1,2,[]])
+
+def r; return *[]; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return *[1]; end; a,b,*c = r(); test_ok([a,b,c] == [1,nil,[]])
+def r; return *[nil]; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return *[[]]; end; a,b,*c = r(); test_ok([a,b,c] == [[],nil,[]])
+def r; return *[1,2]; end; a,b,*c = r(); test_ok([a,b,c] == [1,2,[]])
+def r; return *[*[]]; end; a,b,*c = r(); test_ok([a,b,c] == [nil,nil,[]])
+def r; return *[*[1]]; end; a,b,*c = r(); test_ok([a,b,c] == [1,nil,[]])
+def r; return *[*[1,2]]; end; a,b,*c = r(); test_ok([a,b,c] == [1,2,[]])
+
+f = lambda {|r,| test_ok([] == r)}
+f.call([], *[])
+
+f = lambda {|r,*l| test_ok([] == r); test_ok([1] == l)}
+f.call([], *[1])
+
+f = lambda{|x| x}
+test_ok(f.call(42) == 42)
+test_ok(f.call([42]) == [42])
+test_ok(f.call([[42]]) == [[42]])
+test_ok(f.call([42,55]) == [42,55])
+
+f = lambda{|x,| x}
+test_ok(f.call(42) == 42)
+test_ok(f.call([42]) == [42])
+test_ok(f.call([[42]]) == [[42]])
+test_ok(f.call([42,55]) == [42,55])
+
+f = lambda{|*x| x}
+test_ok(f.call(42) == [42])
+test_ok(f.call([42]) == [[42]])
+test_ok(f.call([[42]]) == [[[42]]])
+test_ok(f.call([42,55]) == [[42,55]])
+test_ok(f.call(42,55) == [42,55])
+
+f = lambda { |a, b=42, *c| [a,b,c] }
+test_ok(f.call(1      ) == [1,42,[  ]] )
+test_ok(f.call(1,43   ) == [1,43,[  ]] )
+test_ok(f.call(1,43,44) == [1,43,[44]] )
+
+f = lambda { |a, b=(a|16), *c, &block| [a,b,c,block&&block[]] }
+test_ok(f.call(8      )     == [8,24,[  ],nil] )
+test_ok(f.call(8,43   )     == [8,43,[  ],nil] )
+test_ok(f.call(8,43,44)     == [8,43,[44],nil] )
+test_ok(f.call(8      ){45} == [8,24,[  ],45 ] )
+test_ok(f.call(8,43   ){45} == [8,43,[  ],45 ] )
+test_ok(f.call(8,43,44){45} == [8,43,[44],45 ] )
+
+f = lambda { |a, b=42, *c, d| [a,b,c,d] }
+test_ok(f.call(1      ,99) == [1,42,[  ],99] )
+test_ok(f.call(1,43   ,99) == [1,43,[  ],99] )
+test_ok(f.call(1,43,44,99) == [1,43,[44],99] )
+
+f = lambda { |a, b=(a|16), &block| [a,b,block&&block[]] }
+test_ok(f.call(8   )     == [8,24,nil] )
+test_ok(f.call(8,43)     == [8,43,nil] )
+test_ok(f.call(8,43)     == [8,43,nil] )
+test_ok(f.call(8   ){45} == [8,24,45 ] )
+test_ok(f.call(8,43){45} == [8,43,45 ] )
+test_ok(f.call(8,43){45} == [8,43,45 ] )
+
+f = lambda { |a, b=42, d| [a,b,d] }
+test_ok(f.call(1   ,99) == [1,42,99] )
+test_ok(f.call(1,43,99) == [1,43,99] )
+test_ok(f.call(1,43,99) == [1,43,99] )
+
+f = lambda { |b=42, *c, &block| [b,c,block&&block[]] }
+test_ok(f.call(     )     == [42,[  ],nil] )
+test_ok(f.call(43   )     == [43,[  ],nil] )
+test_ok(f.call(43,44)     == [43,[44],nil] )
+test_ok(f.call(     ){45} == [42,[  ],45 ] )
+test_ok(f.call(43   ){45} == [43,[  ],45 ] )
+test_ok(f.call(43,44){45} == [43,[44],45 ] )
+
+f = lambda { |b=42, *c, d| [b,c,d] }
+test_ok(f.call(      99) == [42,[  ],99] )
+test_ok(f.call(43   ,99) == [43,[  ],99] )
+test_ok(f.call(43,44,99) == [43,[44],99] )
+
+f = lambda { |b=42, &block| [b,block&&block[]] }
+test_ok(f.call(  )     == [42,nil] )
+test_ok(f.call(43)     == [43,nil] )
+test_ok(f.call(43)     == [43,nil] )
+test_ok(f.call(  ){45} == [42,45 ] )
+test_ok(f.call(43){45} == [43,45 ] )
+test_ok(f.call(43){45} == [43,45 ] )
+
+f = lambda { |b=42, d| [b,d] }
+test_ok(f.call(   99) == [42,99] )
+test_ok(f.call(43,99) == [43,99] )
+test_ok(f.call(43,99) == [43,99] )
+
+
+a,=*[1]
+test_ok(a == 1)
+a,=*[[1]]
+test_ok(a == [1])
+a,=*[[[1]]]
+test_ok(a == [[1]])
+
+x, (y, z) = 1, 2, 3
+test_ok([1,2,nil] == [x,y,z])
+x, (y, z) = 1, [2,3]
+test_ok([1,2,3] == [x,y,z])
+x, (y, z) = 1, [2]
+test_ok([1,2,nil] == [x,y,z])
+
+a = loop do break; end; test_ok(a == nil)
+a = loop do break nil; end; test_ok(a == nil)
+a = loop do break 1; end; test_ok(a == 1)
+a = loop do break []; end; test_ok(a == [])
+a = loop do break [1]; end; test_ok(a == [1])
+a = loop do break [nil]; end; test_ok(a == [nil])
+a = loop do break [[]]; end; test_ok(a == [[]])
+a = loop do break [*[]]; end; test_ok(a == [])
+a = loop do break [*[1]]; end; test_ok(a == [1])
+a = loop do break [*[1,2]]; end; test_ok(a == [1,2])
+
+a = loop do break *[]; end; test_ok(a == [])
+a = loop do break *[1]; end; test_ok(a == [1])
+a = loop do break *[nil]; end; test_ok(a == [nil])
+a = loop do break *[[]]; end; test_ok(a == [[]])
+a = loop do break *[*[]]; end; test_ok(a == [])
+a = loop do break *[*[1]]; end; test_ok(a == [1])
+a = loop do break *[*[1,2]]; end; test_ok(a == [1,2])
+
+*a = loop do break; end; test_ok(a == [nil])
+*a = loop do break nil; end; test_ok(a == [nil])
+*a = loop do break 1; end; test_ok(a == [1])
+*a = loop do break []; end; test_ok(a == [])
+*a = loop do break [1]; end; test_ok(a == [1])
+*a = loop do break [nil]; end; test_ok(a == [nil])
+*a = loop do break [[]]; end; test_ok(a == [[]])
+*a = loop do break [1,2]; end; test_ok(a == [1,2])
+*a = loop do break [*[]]; end; test_ok(a == [])
+*a = loop do break [*[1]]; end; test_ok(a == [1])
+*a = loop do break [*[1,2]]; end; test_ok(a == [1,2])
+
+*a = loop do break *[]; end; test_ok(a == [])
+*a = loop do break *[1]; end; test_ok(a == [1])
+*a = loop do break *[nil]; end; test_ok(a == [nil])
+*a = loop do break *[[]]; end; test_ok(a == [[]])
+*a = loop do break *[1,2]; end; test_ok(a == [1,2])
+*a = loop do break *[*[]]; end; test_ok(a == [])
+*a = loop do break *[*[1]]; end; test_ok(a == [1])
+*a = loop do break *[*[1,2]]; end; test_ok(a == [1,2])
+
+*a = *loop do break *[[]]; end; test_ok(a == [[]])
+*a = *loop do break *[1,2]; end; test_ok(a == [1,2])
+*a = *loop do break *[*[1,2]]; end; test_ok(a == [1,2])
+
+a,b,*c = loop do break; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break nil; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break 1; end; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = loop do break []; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break [1]; end; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = loop do break [nil]; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break [[]]; end; test_ok([a,b,c] == [[],nil,[]])
+a,b,*c = loop do break [1,2]; end; test_ok([a,b,c] == [1,2,[]])
+a,b,*c = loop do break [*[]]; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break [*[1]]; end; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = loop do break [*[1,2]]; end; test_ok([a,b,c] == [1,2,[]])
+
+a,b,*c = loop do break *[]; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break *[1]; end; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = loop do break *[nil]; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break *[[]]; end; test_ok([a,b,c] == [[],nil,[]])
+a,b,*c = loop do break *[1,2]; end; test_ok([a,b,c] == [1,2,[]])
+a,b,*c = loop do break *[*[]]; end; test_ok([a,b,c] == [nil,nil,[]])
+a,b,*c = loop do break *[*[1]]; end; test_ok([a,b,c] == [1,nil,[]])
+a,b,*c = loop do break *[*[1,2]]; end; test_ok([a,b,c] == [1,2,[]])
+
+def r(val); a = yield(); test_ok(a == val, 2); end
+r(nil){next}
+r(nil){next nil}
+r(1){next 1}
+r([]){next []}
+r([1]){next [1]}
+r([nil]){next [nil]}
+r([[]]){next [[]]}
+r([]){next [*[]]}
+r([1]){next [*[1]]}
+r([1,2]){next [*[1,2]]}
+
+r([]){next *[]}
+r([1]){next *[1]}
+r([nil]){next *[nil]}
+r([[]]){next *[[]]}
+r([]){next *[*[]]}
+r([1]){next *[*[1]]}
+r([1,2]){next *[*[1,2]]}
+
+def r(val); *a = yield(); test_ok(a == val, 2); end
+r([nil]){next}
+r([nil]){next nil}
+r([1]){next 1}
+r([]){next []}
+r([1]){next [1]}
+r([nil]){next [nil]}
+r([[]]){next [[]]}
+r([1,2]){next [1,2]}
+r([]){next [*[]]}
+r([1]){next [*[1]]}
+r([1,2]){next [*[1,2]]}
+
+def r(val); *a = *yield(); test_ok(a == val, 2); end
+r([[]]){next *[[]]}
+r([1,2]){next *[1,2]}
+r([1,2]){next *[*[1,2]]}
+
+def r(val); a,b,*c = yield(); test_ok([a,b,c] == val, 2); end
+r([nil,nil,[]]){next}
+r([nil,nil,[]]){next nil}
+r([1,nil,[]]){next 1}
+r([nil,nil,[]]){next []}
+r([1,nil,[]]){next [1]}
+r([nil,nil,[]]){next [nil]}
+r([[],nil,[]]){next [[]]}
+r([1,2,[]]){next [1,2]}
+r([nil,nil,[]]){next [*[]]}
+r([1,nil,[]]){next [*[1]]}
+r([1,2,[]]){next [*[1,2]]}
+
+def r(val); a,b,*c = *yield(); test_ok([a,b,c] == val, 2); end
+r([[],nil,[]]){next *[[]]}
+r([1,2,[]]){next *[1,2]}
+r([1,2,[]]){next *[*[1,2]]}
+
+test_check "condition"
+
+$x = '0';
+
+$x == $x && test_ok(true)
+$x != $x && test_ok(false)
+$x == $x || test_ok(false)
+$x != $x || test_ok(true)
+
+# first test to see if we can run the tests.
+
+test_check "if/unless";
+
+$x = 'test';
+test_ok(if $x == $x then true else false end)
+$bad = false
+unless $x == $x
+  $bad = true
+end
+test_ok(!$bad)
+test_ok(unless $x != $x then true else false end)
+
+test_check "case"
+
+case 5
+when 1, 2, 3, 4, 6, 7, 8
+  test_ok(false)
+when 5
+  test_ok(true)
+end
+
+case 5
+when 5
+  test_ok(true)
+when 1..10
+  test_ok(false)
+end
+
+case 5
+when 1..10
+  test_ok(true)
+else
+  test_ok(false)
+end
+
+case 5
+when 5
+  test_ok(true)
+else
+  test_ok(false)
+end
+
+case "foobar"
+when /^f.*r$/
+  test_ok(true)
+else
+  test_ok(false)
+end
+
+test_check "while/until";
+
+while_tmp = "while_tmp.#{$$}"
+tmp = open(while_tmp, "w")
+tmp.print "tvi925\n";
+tmp.print "tvi920\n";
+tmp.print "vt100\n";
+tmp.print "Amiga\n";
+tmp.print "paper\n";
+tmp.close
+
+# test break
+
+tmp = open(while_tmp, "r")
+test_ok(tmp.kind_of?(File))
+
+while line = tmp.gets()
+  break if /vt100/ =~ line
+end
+
+test_ok(!tmp.eof? && /vt100/ =~ line)
+tmp.close
+
+# test next
+$bad = false
+tmp = open(while_tmp, "r")
+while line = tmp.gets()
+  next if /vt100/ =~ line
+  $bad = 1 if /vt100/ =~ line
+end
+test_ok(!(!tmp.eof? || /vt100/ =~ line || $bad))
+tmp.close
+
+# test redo
+$bad = false
+tmp = open(while_tmp, "r")
+while line = tmp.gets()
+  lastline = line
+  line = line.gsub(/vt100/, 'VT100')
+  if lastline != line
+    line.gsub!('VT100', 'Vt100')
+    redo
+  end
+  $bad = 1 if /vt100/ =~ line
+  $bad = 1 if /VT100/ =~ line
+end
+test_ok(tmp.eof? && !$bad)
+tmp.close
+
+sum=0
+for i in 1..10
+  sum += i
+  i -= 1
+  if i > 0
+    redo
+  end
+end
+test_ok(sum == 220)
+
+# test interval
+$bad = false
+tmp = open(while_tmp, "r")
+while line = tmp.gets()
+  break if 3
+  case line
+  when /vt100/, /Amiga/, /paper/
+    $bad = true
+  end
+end
+test_ok(!$bad)
+tmp.close
+
+File.unlink while_tmp or `/bin/rm -f "#{while_tmp}"`
+test_ok(!File.exist?(while_tmp))
+
+i = 0
+until i>4
+  i+=1
+end
+test_ok(i>4)
+
+
+# exception handling
+test_check "exception";
+
+begin
+  raise "this must be handled"
+  test_ok(false)
+rescue
+  test_ok(true)
+end
+
+$bad = true
+begin
+  raise "this must be handled no.2"
+rescue
+  if $bad
+    $bad = false
+    retry
+    test_ok(false)
+  end
+end
+test_ok(true)
+
+# exception in rescue clause
+$string = "this must be handled no.3"
+begin
+  begin
+    raise "exception in rescue clause"
+  rescue
+    raise $string
+  end
+  test_ok(false)
+rescue => e
+  test_ok($! == e)
+  test_ok(e.message == $string)
+  test_ok(e != $string)
+end
+
+# exception in ensure clause
+begin
+  begin
+    raise "this must be handled no.4"
+  ensure
+    raise "exception in ensure clause"
+  end
+  test_ok(false)
+rescue
+  test_ok(true)
+end
+
+$bad = true
+begin
+  begin
+    raise "this must be handled no.5"
+  ensure
+    $bad = false
+  end
+rescue
+end
+test_ok(!$bad)
+
+$bad = true
+begin
+  begin
+    raise "this must be handled no.6"
+  ensure
+    $bad = false
+  end
+rescue
+end
+test_ok(!$bad)
+
+$bad = true
+while true
+  begin
+    break
+  ensure
+    $bad = false
+  end
+end
+test_ok(!$bad)
+
+test_ok(catch(:foo) {
+     loop do
+       loop do
+	 throw :foo, true
+	 break
+       end
+       break
+       test_ok(false)			# should not reach here
+     end
+     false
+   })
+
+test_check "array"
+test_ok([1, 2] + [3, 4] == [1, 2, 3, 4])
+test_ok([1, 2] * 2 == [1, 2, 1, 2])
+test_ok([1, 2] * ":" == "1:2")
+
+test_ok([1, 2].hash == [1, 2].hash)
+
+test_ok([1,2,3] & [2,3,4] == [2,3])
+test_ok([1,2,3] | [2,3,4] == [1,2,3,4])
+test_ok([1,2,3] - [2,3] == [1])
+
+$x = [0, 1, 2, 3, 4, 5]
+test_ok($x[2] == 2)
+test_ok($x[1..3] == [1, 2, 3])
+test_ok($x[1,3] == [1, 2, 3])
+
+$x[0, 2] = 10
+test_ok($x[0] == 10 && $x[1] == 2)
+
+$x[0, 0] = -1
+test_ok($x[0] == -1 && $x[1] == 10)
+
+$x[-1, 1] = 20
+test_ok($x[-1] == 20 && $x.pop == 20)
+
+# array and/or
+test_ok(([1,2,3]&[2,4,6]) == [2])
+test_ok(([1,2,3]|[2,4,6]) == [1,2,3,4,6])
+
+# compact
+$x = [nil, 1, nil, nil, 5, nil, nil]
+$x.compact!
+test_ok($x == [1, 5])
+
+# uniq
+$x = [1, 1, 4, 2, 5, 4, 5, 1, 2]
+$x.uniq!
+test_ok($x == [1, 4, 2, 5])
+
+# empty?
+test_ok(!$x.empty?)
+$x = []
+test_ok($x.empty?)
+
+# sort
+$x = ["it", "came", "to", "pass", "that", "..."]
+$x = $x.sort.join(" ")
+test_ok($x == "... came it pass that to")
+$x = [2,5,3,1,7]
+$x.sort!{|a,b| a<=>b}		# sort with condition
+test_ok($x == [1,2,3,5,7])
+$x.sort!{|a,b| b-a}		# reverse sort
+test_ok($x == [7,5,3,2,1])
+
+# split test
+$x = "The Book of Mormon"
+test_ok($x.split(//).reverse!.join == $x.reverse)
+test_ok($x.reverse == $x.reverse!)
+test_ok("1 byte string".split(//).reverse.join(":") == "g:n:i:r:t:s: :e:t:y:b: :1")
+$x = "a b c  d"
+test_ok($x.split == ['a', 'b', 'c', 'd'])
+test_ok($x.split(' ') == ['a', 'b', 'c', 'd'])
+test_ok(defined? "a".chomp)
+test_ok("abc".scan(/./) == ["a", "b", "c"])
+test_ok("1a2b3c".scan(/(\d.)/) == [["1a"], ["2b"], ["3c"]])
+# non-greedy match
+test_ok("a=12;b=22".scan(/(.*?)=(\d*);?/) == [["a", "12"], ["b", "22"]])
+
+$x = [1]
+test_ok(($x * 5).join(":") == '1:1:1:1:1')
+test_ok(($x * 1).join(":") == '1')
+test_ok(($x * 0).join(":") == '')
+
+*$x = *(1..7).to_a
+test_ok($x.size == 7)
+test_ok($x == [1, 2, 3, 4, 5, 6, 7])
+
+$x = [1,2,3]
+$x[1,0] = $x
+test_ok($x == [1,1,2,3,2,3])
+
+$x = [1,2,3]
+$x[-1,0] = $x
+test_ok($x == [1,2,1,2,3,3])
+
+$x = [1,2,3]
+$x.concat($x)
+test_ok($x == [1,2,3,1,2,3])
+
+test_check "hash"
+$x = {1=>2, 2=>4, 3=>6}
+
+test_ok($x[1] == 2)
+
+test_ok(begin
+     for k,v in $x
+       raise if k*2 != v
+     end
+     true
+   rescue
+     false
+   end)
+
+test_ok($x.length == 3)
+test_ok($x.has_key?(1))
+test_ok($x.has_value?(4))
+test_ok($x.values_at(2,3) == [4,6])
+test_ok($x == {1=>2, 2=>4, 3=>6})
+
+$z = $x.keys.sort.join(":")
+test_ok($z == "1:2:3")
+
+$z = $x.values.sort.join(":")
+test_ok($z == "2:4:6")
+test_ok($x == $x)
+
+$x.shift
+test_ok($x.length == 2)
+
+$z = [1,2]
+$x[$z] = 256
+test_ok($x[$z] == 256)
+
+$x = Hash.new(0)
+$x[1] = 1
+test_ok($x[1] == 1)
+test_ok($x[2] == 0)
+
+$x = Hash.new([])
+test_ok($x[22] == [])
+test_ok($x[22].equal?($x[22]))
+
+$x = Hash.new{[]}
+test_ok($x[22] == [])
+test_ok(!$x[22].equal?($x[22]))
+
+$x = Hash.new{|h,k| $z = k; h[k] = k*2}
+$z = 0
+test_ok($x[22] == 44)
+test_ok($z == 22)
+$z = 0
+test_ok($x[22] == 44)
+test_ok($z == 0)
+$x.default = 5
+test_ok($x[23] == 5)
+
+$x = Hash.new
+def $x.default(k)
+  $z = k
+  self[k] = k*2
+end
+$z = 0
+test_ok($x[22] == 44)
+test_ok($z == 22)
+$z = 0
+test_ok($x[22] == 44)
+test_ok($z == 0)
+
+test_check "iterator"
+
+test_ok(!iterator?)
+
+def ttt
+  test_ok(iterator?)
+end
+ttt{}
+
+# yield at top level
+test_ok(!defined?(yield))
+
+$x = [1, 2, 3, 4]
+$y = []
+
+# iterator over array
+for i in $x
+  $y.push i
+end
+test_ok($x == $y)
+
+# nested iterator
+def tt
+  1.upto(10) {|i|
+    yield i
+  }
+end
+
+i=0
+tt{|i| break if i == 5}
+test_ok(i == 0)
+
+def tt2(dummy)
+  yield 1
+end
+
+def tt3(&block)
+  tt2(raise(ArgumentError,""),&block)
+end
+
+$x = false
+begin
+  tt3{}
+rescue ArgumentError
+  $x = true
+rescue Exception
+end
+test_ok($x)
+
+def tt4 &block
+  tt2(raise(ArgumentError,""),&block)
+end
+$x = false
+begin
+  tt4{}
+rescue ArgumentError
+  $x = true
+rescue Exception
+end
+test_ok($x)
+
+# iterator break/redo/next/retry
+done = true
+loop{
+  break
+  done = false			# should not reach here
+}
+test_ok(done)
+
+done = false
+$bad = false
+loop {
+  break if done
+  done = true
+  next
+  $bad = true			# should not reach here
+}
+test_ok(!$bad)
+
+done = false
+$bad = false
+loop {
+  break if done
+  done = true
+  redo
+  $bad = true			# should not reach here
+}
+test_ok(!$bad)
+
+$x = []
+for i in 1 .. 7
+  $x.push i
+end
+test_ok($x.size == 7)
+test_ok($x == [1, 2, 3, 4, 5, 6, 7])
+
+# append method to built-in class
+class Array
+  def iter_test1
+    collect{|e| [e, yield(e)]}.sort{|a,b|a[1]<=>b[1]}
+  end
+  def iter_test2
+    a = collect{|e| [e, yield(e)]}
+    a.sort{|a,b|a[1]<=>b[1]}
+  end
+end
+$x = [[1,2],[3,4],[5,6]]
+test_ok($x.iter_test1{|x|x} == $x.iter_test2{|x|x})
+
+class IterTest
+  def initialize(e); @body = e; end
+
+  def each0(&block); @body.each(&block); end
+  def each1(&block); @body.each {|*x| block.call(*x) } end
+  def each2(&block); @body.each {|*x| block.call(x) } end
+  def each3(&block); @body.each {|x| block.call(*x) } end
+  def each4(&block); @body.each {|x| block.call(x) } end
+  def each5; @body.each {|*x| yield(*x) } end
+  def each6; @body.each {|*x| yield(x) } end
+  def each7; @body.each {|x| yield(*x) } end
+  def each8; @body.each {|x| yield(x) } end
+
+  def f(a)
+    a
+  end
+end
+test_ok(IterTest.new(nil).method(:f).to_proc.call([1]) == [1])
+m = /\w+/.match("abc")
+test_ok(IterTest.new(nil).method(:f).to_proc.call([m]) == [m])
+
+IterTest.new([0]).each0 {|x| test_ok(x == 0)}
+IterTest.new([1]).each1 {|x| test_ok(x == 1)}
+IterTest.new([2]).each2 {|x| test_ok(x == [2])}
+#IterTest.new([3]).each3 {|x| test_ok(x == 3)}
+IterTest.new([4]).each4 {|x| test_ok(x == 4)}
+IterTest.new([5]).each5 {|x| test_ok(x == 5)}
+IterTest.new([6]).each6 {|x| test_ok(x == [6])}
+#IterTest.new([7]).each7 {|x| test_ok(x == 7)}
+IterTest.new([8]).each8 {|x| test_ok(x == 8)}
+
+IterTest.new([[0]]).each0 {|x| test_ok(x == [0])}
+IterTest.new([[1]]).each1 {|x| test_ok(x == [1])}
+IterTest.new([[2]]).each2 {|x| test_ok(x == [[2]])}
+IterTest.new([[3]]).each3 {|x| test_ok(x == 3)}
+IterTest.new([[4]]).each4 {|x| test_ok(x == [4])}
+IterTest.new([[5]]).each5 {|x| test_ok(x == [5])}
+IterTest.new([[6]]).each6 {|x| test_ok(x == [[6]])}
+IterTest.new([[7]]).each7 {|x| test_ok(x == 7)}
+IterTest.new([[8]]).each8 {|x| test_ok(x == [8])}
+
+IterTest.new([[0,0]]).each0 {|*x| test_ok(x == [[0,0]])}
+IterTest.new([[8,8]]).each8 {|*x| test_ok(x == [[8,8]])}
+
+def m0(v)
+  v
+end
+
+def m1
+  m0(block_given?)
+end
+test_ok(m1{p 'test'})
+test_ok(!m1)
+
+def m
+  m0(block_given?,&Proc.new{})
+end
+test_ok(m1{p 'test'})
+test_ok(!m1)
+
+class C
+  include Enumerable
+  def initialize
+    @a = [1,2,3]
+  end
+  def each(&block)
+    @a.each(&block)
+  end
+end
+
+test_ok(C.new.collect{|n| n} == [1,2,3])
+
+test_ok(Proc == lambda{}.class)
+test_ok(Proc == Proc.new{}.class)
+lambda{|a|test_ok(a==1)}.call(1)
+def block_test(klass, &block)
+  test_ok(klass === block)
+end
+
+block_test(NilClass)
+block_test(Proc){}
+
+def call_argument_test(state, proc, *args)
+  x = state
+  begin
+    proc.call(*args)
+  rescue ArgumentError
+    x = !x
+  end
+  test_ok(x,2)
+end
+
+call_argument_test(true, lambda{||})
+call_argument_test(false, lambda{||}, 1)
+call_argument_test(true, lambda{|a,|}, 1)
+call_argument_test(false, lambda{|a,|})
+call_argument_test(false, lambda{|a,|}, 1,2)
+
+call_argument_test(true, Proc.new{||})
+call_argument_test(true, Proc.new{||}, 1)
+call_argument_test(true, Proc.new{|a,|}, 1)
+call_argument_test(true, Proc.new{|a,|})
+call_argument_test(true, Proc.new{|a,|}, 1,2)
+
+def block_get(&block)
+  block
+end
+
+test_ok(Proc == block_get{}.class)
+call_argument_test(true, block_get{||})
+call_argument_test(true, block_get{||}, 1)
+call_argument_test(true, block_get{|a,|}, 1)
+call_argument_test(true, block_get{|a,|})
+call_argument_test(true, block_get{|a,|}, 1,2)
+
+call_argument_test(true, block_get(&lambda{||}))
+call_argument_test(false, block_get(&lambda{||}),1)
+call_argument_test(true, block_get(&lambda{|a,|}),1)
+call_argument_test(false, block_get(&lambda{|a,|}),1,2)
+
+blk = block_get{11}
+test_ok(blk.class == Proc)
+test_ok(blk.to_proc.class == Proc)
+test_ok(blk.clone.call == 11)
+test_ok(block_get(&blk).class == Proc)
+
+lmd = lambda{44}
+test_ok(lmd.class == Proc)
+test_ok(lmd.to_proc.class == Proc)
+test_ok(lmd.clone.call == 44)
+test_ok(block_get(&lmd).class == Proc)
+
+test_ok(Proc.new{|a,| a}.yield(1,2,3) == 1)
+call_argument_test(true, Proc.new{|a,|}, 1,2)
+
+test_ok(Proc.new{|&b| b.call(10)}.call {|x| x} == 10)
+test_ok(Proc.new{|a,&b| b.call(a)}.call(12) {|x| x} == 12)
+
+def test_return1
+  Proc.new {
+    return 55
+  }.yield + 5
+end
+test_ok(test_return1() == 55)
+def test_return2
+  lambda {
+    return 55
+  }.call + 5
+end
+test_ok(test_return2() == 60)
+
+def proc_call(&b)
+  b.call
+end
+def proc_yield()
+  yield
+end
+def proc_return1
+  lambda{return 42}.call+1
+end
+test_ok(proc_return1() == 43)
+def proc_return2
+  ->{return 42}.call+1
+end
+test_ok(proc_return2() == 43)
+def proc_return3
+  proc_call{return 42}+1
+end
+test_ok(proc_return3() == 42)
+def proc_return4
+  proc_yield{return 42}+1
+end
+test_ok(proc_return4() == 42)
+
+def ljump_test(state, proc, *args)
+  x = state
+  begin
+    proc.call(*args)
+  rescue LocalJumpError
+    x = !x
+  end
+  test_ok(x,2)
+end
+
+ljump_test(false, block_get{break})
+ljump_test(true, lambda{break})
+
+def exit_value_test(&block)
+  block.call
+rescue LocalJumpError
+  $!.exit_value
+end
+
+test_ok(45 == exit_value_test{break 45})
+
+test_ok(55 == begin
+              block_get{break 55}.call
+            rescue LocalJumpError
+              $!.exit_value
+            end)
+
+def block_call(&block)
+  block.call
+end
+
+def test_b1
+  block_call{break 11}
+end
+test_ok(test_b1() == 11)
+
+def ljump_rescue(r)
+  begin
+    yield
+  rescue LocalJumpError => e
+    r if /from proc-closure/ =~ e.message
+  end
+end
+
+def test_b2
+  ljump_rescue(22) do
+    block_get{break 21}.call
+  end
+end
+test_ok(test_b2() == 22)
+
+def test_b3
+  ljump_rescue(33) do
+    Proc.new{break 31}.yield
+  end
+end
+test_ok(test_b3() == 33)
+
+def test_b4
+  lambda{break 44}.call
+end
+test_ok(test_b4() == 44)
+
+def test_b5
+  ljump_rescue(55) do
+    b = block_get{break 54}
+    block_call(&b)
+  end
+end
+test_ok(test_b5() == 55)
+
+def test_b6
+  b = lambda{break 67}
+  block_call(&b)
+  66
+end
+test_ok(test_b6() == 66)
+
+def util_r7
+  block_get{break 78}
+end
+
+def test_b7
+  b = util_r7()
+  ljump_rescue(77) do
+    block_call(&b)
+  end
+end
+test_ok(test_b7() == 77)
+
+def util_b8(&block)
+  block_call(&block)
+end
+
+def test_b8
+  util_b8{break 88}
+end
+test_ok(test_b8() == 88)
+
+def util_b9(&block)
+  lambda{block.call; 98}.call
+end
+
+def test_b9
+  util_b9{break 99}
+end
+test_ok(test_b9() == 99)
+
+def util_b10
+  util_b9{break 100}
+end
+
+def test_b10
+  util_b10()
+end
+test_ok(test_b10() == 100)
+
+def test_b11
+  ljump_rescue(111) do
+    loop do
+      Proc.new{break 110}.yield
+      break 112
+    end
+  end
+end
+test_ok(test_b11() == 111)
+
+def test_b12
+  loop do
+    break lambda{break 122}.call
+    break 121
+  end
+end
+test_ok(test_b12() == 122)
+
+def test_b13
+  ljump_rescue(133) do
+    while true
+      Proc.new{break 130}.yield
+      break 131
+    end
+  end
+end
+test_ok(test_b13() == 133)
+
+def test_b14
+  while true
+    break lambda{break 144}.call
+    break 143
+  end
+end
+test_ok(test_b14() == 144)
+
+def test_b15
+  [0].each {|c| yield 1 }
+  156
+end
+test_ok(test_b15{|e| break 155 } == 155)
+
+def marity_test(m)
+  method = method(m)
+  test_ok(method.arity == method.to_proc.arity, 2)
+end
+marity_test(:test_ok)
+marity_test(:marity_test)
+marity_test(:p)
+
+lambda(&method(:test_ok)).call(true)
+lambda(&block_get{|a,n| test_ok(a,n)}).call(true, 2)
+
+class ITER_TEST1
+   def a
+     block_given?
+   end
+end
+
+class ITER_TEST2 < ITER_TEST1
+   def a
+     test_ok(super)
+     super
+   end
+end
+test_ok(ITER_TEST2.new.a {})
+
+class ITER_TEST3
+  def foo x
+    return yield if block_given?
+    x
+  end
+end
+
+class ITER_TEST4 < ITER_TEST3
+  def foo x
+    test_ok(super == yield)
+    test_ok(super(x, &nil) == x)
+  end
+end
+
+ITER_TEST4.new.foo(44){55}
+
+class ITER_TEST5
+   def tt(aa)
+     aa
+   end
+
+   def uu(a)
+      class << self
+         define_method(:tt) do |sym|
+            super(sym)
+         end
+      end
+   end
+
+   def xx(*x)
+     x.size
+   end
+end
+
+a = ITER_TEST5.new
+a.uu(12)
+test_ok(a.tt(1) == 1)
+
+class ITER_TEST6 < ITER_TEST5
+   def xx(*a)
+      a << 12
+      super
+   end
+end
+
+test_ok(ITER_TEST6.new.xx([24]) == 2)
+
+test_check "float"
+test_ok(2.6.floor == 2)
+test_ok((-2.6).floor == -3)
+test_ok(2.6.ceil == 3)
+test_ok((-2.6).ceil == -2)
+test_ok(2.6.truncate == 2)
+test_ok((-2.6).truncate == -2)
+test_ok(2.6.round == 3)
+test_ok((-2.4).truncate == -2)
+test_ok((13.4 % 1 - 0.4).abs < 0.0001)
+nan = 0.0/0
+def nan_test(x,y)
+  test_ok(x != y)
+  test_ok((x < y) == false)
+  test_ok((x > y) == false)
+  test_ok((x <= y) == false)
+  test_ok((x >= y) == false)
+end
+nan_test(nan, nan)
+nan_test(nan, 0)
+nan_test(nan, 1)
+nan_test(nan, -1)
+nan_test(nan, 1000)
+nan_test(nan, -1000)
+nan_test(nan, 1_000_000_000_000)
+nan_test(nan, -1_000_000_000_000)
+nan_test(nan, 100.0);
+nan_test(nan, -100.0);
+nan_test(nan, 0.001);
+nan_test(nan, -0.001);
+nan_test(nan, 1.0/0);
+nan_test(nan, -1.0/0);
+
+#s = "3.7517675036461267e+17"
+#test_ok(s == sprintf("%.16e", s.to_f))
+f = 3.7517675036461267e+17
+test_ok(f == sprintf("%.16e", f).to_f)
+
+
+test_check "bignum"
+def fact(n)
+  return 1 if n == 0
+  f = 1
+  while n>0
+    f *= n
+    n -= 1
+  end
+  return f
+end
+$x = fact(40)
+test_ok($x == $x)
+test_ok($x == fact(40))
+test_ok($x < $x+2)
+test_ok($x > $x-2)
+test_ok($x == 815915283247897734345611269596115894272000000000)
+test_ok($x != 815915283247897734345611269596115894272000000001)
+test_ok($x+1 == 815915283247897734345611269596115894272000000001)
+test_ok($x/fact(20) == 335367096786357081410764800000)
+$x = -$x
+test_ok($x == -815915283247897734345611269596115894272000000000)
+test_ok(2-(2**32) == -(2**32-2))
+test_ok(2**32 - 5 == (2**32-3)-2)
+
+$good = true;
+for i in 1000..1014
+  $good = false if ((1 << i) != (2**i))
+end
+test_ok($good)
+
+$good = true;
+n1= 1 << 1000
+for i in 1000..1014
+  $good = false if ((1 << i) != n1)
+  n1 *= 2
+end
+test_ok($good)
+
+$good = true;
+n2=n1
+for i in 1..10
+  n1 = n1 / 2
+  n2 = n2 >> 1
+  $good = false if (n1 != n2)
+end
+test_ok($good)
+
+$good = true;
+for i in 4000..4096
+  n1 = 1 << i;
+  if (n1**2-1) / (n1+1) != (n1-1)
+    $good = false
+  end
+end
+test_ok($good)
+
+b = 10**80
+a = b * 9 + 7
+test_ok(7 == a.modulo(b))
+test_ok(-b + 7 == a.modulo(-b))
+test_ok(b + -7 == (-a).modulo(b))
+test_ok(-7 == (-a).modulo(-b))
+test_ok(7 == a.remainder(b))
+test_ok(7 == a.remainder(-b))
+test_ok(-7 == (-a).remainder(b))
+test_ok(-7 == (-a).remainder(-b))
+
+test_ok(10**40+10**20 == 10000000000000000000100000000000000000000)
+test_ok(10**40/10**20 == 100000000000000000000)
+
+a = 677330545177305025495135714080
+b = 14269972710765292560
+test_ok(a % b == 0)
+test_ok(-a % b == 0)
+
+def shift_test(a)
+  b = a / (2 ** 32)
+  c = a >> 32
+  test_ok(b == c)
+
+  b = a * (2 ** 32)
+  c = a << 32
+  test_ok(b == c)
+end
+
+shift_test(-4518325415524767873)
+shift_test(-0xfffffffffffffffff)
+
+test_check "string & char"
+
+test_ok("abcd" == "abcd")
+test_ok("abcd" =~ /abcd/)
+test_ok("abcd" === "abcd")
+# compile time string concatenation
+test_ok("ab" "cd" == "abcd")
+test_ok("#{22}aa" "cd#{44}" == "22aacd44")
+test_ok("#{22}aa" "cd#{44}" "55" "#{66}" == "22aacd445566")
+test_ok("abc" !~ /^$/)
+test_ok("abc\n" !~ /^$/)
+test_ok("abc" !~ /^d*$/)
+test_ok(("abc" =~ /d*$/) == 3)
+test_ok("" =~ /^$/)
+test_ok("\n" =~ /^$/)
+test_ok("a\n\n" =~ /^$/)
+test_ok("abcabc" =~ /.*a/ && $& == "abca")
+test_ok("abcabc" =~ /.*c/ && $& == "abcabc")
+test_ok("abcabc" =~ /.*?a/ && $& == "a")
+test_ok("abcabc" =~ /.*?c/ && $& == "abc")
+test_ok(/(.|\n)*?\n(b|\n)/ =~ "a\nb\n\n" && $& == "a\nb")
+
+test_ok(/^(ab+)+b/ =~ "ababb" && $& == "ababb")
+test_ok(/^(?:ab+)+b/ =~ "ababb" && $& == "ababb")
+test_ok(/^(ab+)+/ =~ "ababb" && $& == "ababb")
+test_ok(/^(?:ab+)+/ =~ "ababb" && $& == "ababb")
+
+test_ok(/(\s+\d+){2}/ =~ " 1 2" && $& == " 1 2")
+test_ok(/(?:\s+\d+){2}/ =~ " 1 2" && $& == " 1 2")
+
+$x = <<END;
+ABCD
+ABCD
+END
+$x.gsub!(/((.|\n)*?)B((.|\n)*?)D/, '\1\3')
+test_ok($x == "AC\nAC\n")
+
+test_ok("foobar" =~ /foo(?=(bar)|(baz))/)
+test_ok("foobaz" =~ /foo(?=(bar)|(baz))/)
+
+$foo = "abc"
+test_ok("#$foo = abc" == "abc = abc")
+test_ok("#{$foo} = abc" == "abc = abc")
+
+foo = "abc"
+test_ok("#{foo} = abc" == "abc = abc")
+
+test_ok('-' * 5 == '-----')
+test_ok('-' * 1 == '-')
+test_ok('-' * 0 == '')
+
+foo = '-'
+test_ok(foo * 5 == '-----')
+test_ok(foo * 1 == '-')
+test_ok(foo * 0 == '')
+
+$x = "a.gif"
+test_ok($x.sub(/.*\.([^\.]+)$/, '\1') == "gif")
+test_ok($x.sub(/.*\.([^\.]+)$/, 'b.\1') == "b.gif")
+test_ok($x.sub(/.*\.([^\.]+)$/, '\2') == "")
+test_ok($x.sub(/.*\.([^\.]+)$/, 'a\2b') == "ab")
+test_ok($x.sub(/.*\.([^\.]+)$/, '<\&>') == "<a.gif>")
+
+# character constants(assumes ASCII)
+test_ok("a"[0] == ?a)
+test_ok(?a == ?a)
+test_ok(?\C-a == "\1")
+test_ok(?\M-a == "\341")
+test_ok(?\M-\C-a == "\201")
+test_ok("a".upcase![0] == ?A)
+test_ok("A".downcase![0] == ?a)
+test_ok("abc".tr!("a-z", "A-Z") == "ABC")
+test_ok("aabbcccc".tr_s!("a-z", "A-Z") == "ABC")
+test_ok("abcc".squeeze!("a-z") == "abc")
+test_ok("abcd".delete!("bc") == "ad")
+
+$x = "abcdef"
+$y = [ ?a, ?b, ?c, ?d, ?e, ?f ]
+$bad = false
+$x.each_byte {|i|
+  if i.chr != $y.shift
+    $bad = true
+    break
+  end
+}
+test_ok(!$bad)
+
+s = "a string"
+s[0..s.size]="another string"
+test_ok(s == "another string")
+
+s = <<EOS
+#{
+[1,2,3].join(",")
+}
+EOS
+test_ok(s == "1,2,3\n")
+test_ok("Just".to_i(36) == 926381)
+test_ok("-another".to_i(36) == -23200231779)
+test_ok(1299022.to_s(36) == "ruby")
+test_ok(-1045307475.to_s(36) == "-hacker")
+test_ok("Just_another_Ruby_hacker".to_i(36) == 265419172580680477752431643787347)
+test_ok(-265419172580680477752431643787347.to_s(36) == "-justanotherrubyhacker")
+
+a = []
+(0..255).each {|n|
+  ch = [n].pack("C")
+  a.push ch if /a#{Regexp.quote ch}b/x =~ "ab"
+}
+test_ok(a.size == 0)
+
+test_check "assignment"
+a = nil
+test_ok(defined?(a))
+test_ok(a == nil)
+
+# multiple asignment
+a, b = 1, 2
+test_ok(a == 1 && b == 2)
+
+a, b = b, a
+test_ok(a == 2 && b == 1)
+
+a, = 1,2
+test_ok(a == 1)
+
+a, *b = 1, 2, 3
+test_ok(a == 1 && b == [2, 3])
+
+a, (b, c), d = 1, [2, 3], 4
+test_ok(a == 1 && b == 2 && c == 3 && d == 4)
+
+*a = 1, 2, 3
+test_ok(a == [1, 2, 3])
+
+*a = 4
+test_ok(a == [4])
+
+*a = nil
+test_ok(a == [nil])
+
+test_check "call"
+def aaa(a, b=100, *rest)
+  res = [a, b]
+  res += rest if rest
+  return res
+end
+
+# not enough argument
+begin
+  aaa()				# need at least 1 arg
+  test_ok(false)
+rescue
+  test_ok(true)
+end
+
+begin
+  aaa				# no arg given (exception raised)
+  test_ok(false)
+rescue
+  test_ok(true)
+end
+
+test_ok(aaa(1) == [1, 100])
+test_ok(aaa(1, 2) == [1, 2])
+test_ok(aaa(1, 2, 3, 4) == [1, 2, 3, 4])
+test_ok(aaa(1, *[2, 3, 4]) == [1, 2, 3, 4])
+
+test_check "proc"
+$proc = Proc.new{|i| i}
+test_ok($proc.call(2) == 2)
+test_ok($proc.call(3) == 3)
+
+$proc = Proc.new{|i| i*2}
+test_ok($proc.call(2) == 4)
+test_ok($proc.call(3) == 6)
+
+Proc.new{
+  iii=5				# nested local variable
+  $proc = Proc.new{|i|
+    iii = i
+  }
+  $proc2 = Proc.new {
+    $x = iii			# nested variables shared by procs
+  }
+  # scope of nested variables
+  test_ok(defined?(iii))
+}.call
+test_ok(!defined?(iii))		# out of scope
+
+loop{iii=5; test_ok(eval("defined? iii")); break}
+loop {
+  iii = 10
+  def dyna_var_check
+    loop {
+      test_ok(!defined?(iii))
+      break
+    }
+  end
+  dyna_var_check
+  break
+}
+$x=0
+$proc.call(5)
+$proc2.call
+test_ok($x == 5)
+
+if defined? Process.kill
+  test_check "signal"
+
+  $x = 0
+  trap "SIGINT", Proc.new{|sig| $x = 2}
+  Process.kill "SIGINT", $$
+  100.times {
+    sleep 0.1
+    break if $x != 0
+  }
+  test_ok($x == 2)
+
+  trap "SIGINT", Proc.new{raise "Interrupt"}
+
+  x = false
+  begin
+    Process.kill "SIGINT", $$
+    sleep 0.1
+  rescue
+    x = $!
+  end
+  test_ok(x && /Interrupt/ =~ x.message)
+end
+
+test_check "eval"
+test_ok(eval("") == nil)
+$bad=false
+eval 'while false; $bad = true; print "foo\n" end'
+test_ok(!$bad)
+
+test_ok(eval('Object'))
+test_ok(eval('true'))
+test_ok(!eval('nil'))
+test_ok(!eval('false'))
+
+$foo = 'test_ok(true)'
+begin
+  eval $foo
+rescue
+  test_ok(false)
+end
+
+test_ok(eval("$foo") == 'test_ok(true)')
+test_ok(eval("true") == true)
+i = 5
+test_ok(eval("i == 5"))
+test_ok(eval("i") == 5)
+test_ok(eval("defined? i"))
+
+# eval with binding
+def test_ev
+  local1 = "local1"
+  lambda {
+    local2 = "local2"
+    return binding
+  }.call
+end
+
+$x = test_ev
+test_ok(eval("local1", $x) == "local1") # normal local var
+test_ok(eval("local2", $x) == "local2") # nested local var
+$bad = true
+begin
+  p eval("local1")
+rescue NameError		# must raise error
+  $bad = false
+end
+test_ok(!$bad)
+
+module EvTest
+  EVTEST1 = 25
+  evtest2 = 125
+  $x = binding
+end
+test_ok(eval("EVTEST1", $x) == 25)	# constant in module
+test_ok(eval("evtest2", $x) == 125)	# local var in module
+$bad = true
+begin
+  eval("EVTEST1")
+rescue NameError		# must raise error
+  $bad = false
+end
+test_ok(!$bad)
+
+x = binding #! YARV Limitation: Proc.new{}
+eval "i4 = 1", x
+test_ok(eval("i4", x) == 1)
+x = Proc.new{binding}.call #! YARV Limitation: Proc.new{Proc.new{}}.call
+eval "i4 = 22", x
+test_ok(eval("i4", x) == 22)
+$x = []
+x = Proc.new{binding}.call #! YARV Limitation: Proc.new{Proc.new{}}.call
+eval "(0..9).each{|i5| $x[i5] = Proc.new{i5*2}}", x
+test_ok($x[4].call == 8)
+
+x = binding
+eval "i = 1", x
+test_ok(eval("i", x) == 1)
+x = Proc.new{binding}.call
+eval "i = 22", x
+test_ok(eval("i", x) == 22)
+$x = []
+x = Proc.new{binding}.call
+eval "(0..9).each{|i5| $x[i5] = Proc.new{i5*2}}", x
+test_ok($x[4].call == 8)
+x = Proc.new{binding}.call
+eval "for i6 in 1..1; j6=i6; end", x
+test_ok(eval("defined? i6", x))
+test_ok(eval("defined? j6", x))
+
+Proc.new {
+  p = binding
+  eval "foo11 = 1", p
+  foo22 = 5
+  Proc.new{foo11=22}.call
+  Proc.new{foo22=55}.call
+  test_ok(eval("foo11", p) == eval("foo11"))
+  test_ok(eval("foo11") == 1)
+  test_ok(eval("foo22", p) == eval("foo22"))
+  test_ok(eval("foo22") == 55)
+}.call if false #! YARV Limitation
+
+#! YARV Limitation: p1 = Proc.new{i7 = 0; Proc.new{i7}}.call
+p1 = Proc.new{i7 = 0; binding}.call
+#! YARV Limitation: test_ok(p1.call == 0)
+eval "i7=5", p1
+#! YARV Limitation: test_ok(p1.call == 5)
+test_ok(!defined?(i7))
+
+if false #! YARV Limitation
+p1 = Proc.new{i7 = 0; Proc.new{i7}}.call
+i7 = nil
+test_ok(p1.call == 0)
+eval "i7=1", p1
+test_ok(p1.call == 1)
+eval "i7=5", p1
+test_ok(p1.call == 5)
+test_ok(i7 == nil)
+end
+
+test_check "system"
+test_ok(`echo foobar` == "foobar\n")
+test_ok(`./miniruby -e 'print "foobar"'` == 'foobar')
+
+script_tmp = "script_tmp.#{$$}"
+tmp = open(script_tmp, "w")
+tmp.print "print $zzz\n";
+tmp.close
+
+test_ok(`./miniruby -s #{script_tmp} -zzz` == 'true')
+test_ok(`./miniruby -s #{script_tmp} -zzz=555` == '555')
+
+tmp = open(script_tmp, "w")
+tmp.print "#! /usr/local/bin/ruby -s\n";
+tmp.print "print $zzz\n";
+tmp.close
+
+test_ok(`./miniruby #{script_tmp} -zzz=678` == '678')
+
+tmp = open(script_tmp, "w")
+tmp.print "this is a leading junk\n";
+tmp.print "#! /usr/local/bin/ruby -s\n";
+tmp.print "print $zzz\n";
+tmp.print "__END__\n";
+tmp.print "this is a trailing junk\n";
+tmp.close
+
+test_ok(`./miniruby -x #{script_tmp}` == '')
+test_ok(`./miniruby -x #{script_tmp} -zzz=555` == '555')
+
+tmp = open(script_tmp, "w")
+for i in 1..5
+  tmp.print i, "\n"
+end
+tmp.close
+
+`./miniruby -i.bak -pe '$_.sub!(/^[0-9]+$/){$&.to_i * 5}' #{script_tmp}`
+done = true
+tmp = open(script_tmp, "r")
+while tmp.gets
+  if $_.to_i % 5 != 0
+    done = false
+    break
+  end
+end
+tmp.close
+test_ok(done)
+
+File.unlink script_tmp or `/bin/rm -f "#{script_tmp}"`
+File.unlink "#{script_tmp}.bak" or `/bin/rm -f "#{script_tmp}.bak"`
+
+test_check "const"
+TEST1 = 1
+TEST2 = 2
+
+module Const
+  TEST3 = 3
+  TEST4 = 4
+end
+
+module Const2
+  TEST3 = 6
+  TEST4 = 8
+end
+
+include Const
+
+test_ok([TEST1,TEST2,TEST3,TEST4] == [1,2,3,4])
+
+include Const2
+STDERR.print "intentionally redefines TEST3, TEST4\n" if $VERBOSE
+test_ok([TEST1,TEST2,TEST3,TEST4] == [1,2,6,8])
+
+
+test_ok((String <=> Object) == -1)
+test_ok((Object <=> String) == 1)
+test_ok((Array <=> String) == nil)
+
+test_check "clone"
+foo = Object.new
+def foo.test
+  "test"
+end
+bar = foo.clone
+def bar.test2
+  "test2"
+end
+
+test_ok(bar.test2 == "test2")
+test_ok(bar.test == "test")
+test_ok(foo.test == "test")
+
+begin
+  foo.test2
+  test_ok false
+rescue NoMethodError
+  test_ok true
+end
+
+module M001; end
+module M002; end
+module M003; include M002; end
+module M002; include M001; end
+module M003; include M002; end
+
+test_ok(M003.ancestors == [M003, M002, M001])
+
+test_check "marshal"
+$x = [1,2,3,[4,5,"foo"],{1=>"bar"},2.5,fact(30)]
+$y = Marshal.dump($x)
+test_ok($x == Marshal.load($y))
+
+StrClone=String.clone;
+test_ok(Marshal.load(Marshal.dump(StrClone.new("abc"))).class == StrClone)
+
+[[1,2,3,4], [81, 2, 118, 3146]].each { |w,x,y,z|
+  a = (x.to_f + y.to_f / z.to_f) * Math.exp(w.to_f / (x.to_f + y.to_f / z.to_f))
+  ma = Marshal.dump(a)
+  b = Marshal.load(ma)
+  test_ok(a == b)
+}
+
+test_check "pack"
+
+$format = "c2x5CCxsdils_l_a6";
+# Need the expression in here to force ary[5] to be numeric.  This avoids
+# test2 failing because ary2 goes str->numeric->str and ary does not.
+ary = [1,-100,127,128,32767,987.654321098 / 100.0,12345,123456,-32767,-123456,"abcdef"]
+$x = ary.pack($format)
+ary2 = $x.unpack($format)
+
+test_ok(ary.length == ary2.length)
+test_ok(ary.join(':') == ary2.join(':'))
+test_ok($x =~ /def/)
+
+$x = [-1073741825]
+test_ok($x.pack("q").unpack("q") == $x)
+
+test_check "math"
+test_ok(Math.sqrt(4) == 2)
+
+include Math
+test_ok(sqrt(4) == 2)
+
+test_check "struct"
+struct_test = Struct.new("Test", :foo, :bar)
+test_ok(struct_test == Struct::Test)
+
+test = struct_test.new(1, 2)
+test_ok(test.foo == 1 && test.bar == 2)
+test_ok(test[0] == 1 && test[1] == 2)
+
+a, b = test.to_a
+test_ok(a == 1 && b == 2)
+
+test[0] = 22
+test_ok(test.foo == 22)
+
+test.bar = 47
+test_ok(test.bar == 47)
+
+test_check "variable"
+test_ok($$.instance_of?(Integer))
+
+# read-only variable
+begin
+  $$ = 5
+  test_ok false
+rescue NameError
+  test_ok true
+end
+
+foobar = "foobar"
+$_ = foobar
+test_ok($_ == foobar)
+
+class Gods
+  @@rule = "Uranus"		# private to Gods
+  def ruler0
+    @@rule
+  end
+
+  def self.ruler1		# <= per method definition style
+    @@rule
+  end
+  class << self			# <= multiple method definition style
+    def ruler2
+      @@rule
+    end
+  end
+end
+
+module Olympians
+  @@rule ="Zeus"
+  def ruler3
+    @@rule
+  end
+end
+
+class Titans < Gods
+  @@rule = "Cronus"		# do not affect @@rule in Gods
+  include Olympians
+  def ruler4
+    @@rule
+  end
+end
+
+test_ok(Gods.new.ruler0 == "Cronus")
+test_ok(Gods.ruler1 == "Cronus")
+test_ok(Gods.ruler2 == "Cronus")
+test_ok(Titans.ruler1 == "Cronus")
+test_ok(Titans.ruler2 == "Cronus")
+atlas = Titans.new
+test_ok(atlas.ruler0 == "Cronus")
+test_ok(atlas.ruler3 == "Zeus")
+test_ok(atlas.ruler4 == "Cronus")
+
+test_check "trace"
+$x = 1234
+$y = 0
+trace_var :$x, Proc.new{$y = $x}
+$x = 40414
+test_ok($y == $x)
+
+untrace_var :$x
+$x = 19660208
+test_ok($y != $x)
+
+trace_var :$x, Proc.new{$x *= 2}
+$x = 5
+test_ok($x == 10)
+
+untrace_var :$x
+
+test_check "defined?"
+
+test_ok(defined?($x))		# global variable
+test_ok(defined?($x) == 'global-variable')# returns description
+
+foo=5
+test_ok(defined?(foo))		# local variable
+
+test_ok(defined?(Array))	# constant
+test_ok(defined?(Object.new))	# method
+test_ok(!defined?(Object.print))# private method
+test_ok(defined?(1 == 2))	# operator expression
+
+class Foo
+  def foo
+    p :foo
+  end
+  protected :foo
+  def bar(f)
+    test_ok(defined?(self.foo))
+    test_ok(defined?(f.foo))
+  end
+end
+f = Foo.new
+test_ok(defined?(f.foo) == nil)
+f.bar(f)
+
+def defined_test
+  return !defined?(yield)
+end
+
+test_ok(defined_test)		# not iterator
+test_ok(!defined_test{})	# called as iterator
+
+test_check "alias"
+class Alias0
+  def foo; "foo" end
+end
+class Alias1 < Alias0
+  alias bar foo
+  def foo; "foo+" + super end
+end
+class Alias2 < Alias1
+  alias baz foo
+  undef foo
+end
+
+x = Alias2.new
+test_ok(x.bar == "foo")
+test_ok(x.baz == "foo+foo")
+
+# test_check for cache
+test_ok(x.baz == "foo+foo")
+
+class Alias3 < Alias2
+  def foo
+    defined? super
+  end
+  def bar
+    defined? super
+  end
+  def quux
+    defined? super
+  end
+end
+x = Alias3.new
+test_ok(!x.foo)
+test_ok(x.bar)
+test_ok(!x.quux)
+
+test_check "path"
+test_ok(File.basename("a") == "a")
+test_ok(File.basename("a/b") == "b")
+test_ok(File.basename("a/b/") == "b")
+test_ok(File.basename("/") == "/")
+test_ok(File.basename("//") == "/")
+test_ok(File.basename("///") == "/")
+test_ok(File.basename("a/b////") == "b")
+test_ok(File.basename("a.rb", ".rb") == "a")
+test_ok(File.basename("a.rb///", ".rb") == "a")
+test_ok(File.basename("a.rb///", ".*") == "a")
+test_ok(File.basename("a.rb///", ".c") == "a.rb")
+test_ok(File.dirname("a") == ".")
+test_ok(File.dirname("/") == "/")
+test_ok(File.dirname("/a") == "/")
+test_ok(File.dirname("a/b") == "a")
+test_ok(File.dirname("a/b/c") == "a/b")
+test_ok(File.dirname("/a/b/c") == "/a/b")
+test_ok(File.dirname("/a/b/") == "/a")
+test_ok(File.dirname("/a/b///") == "/a")
+case Dir.pwd
+when %r'\A\w:'
+  test_ok(/\A\w:\/\z/ =~ File.expand_path(".", "/"))
+  test_ok(/\A\w:\/a\z/ =~ File.expand_path("a", "/"))
+  dosish = true
+when %r'\A//'
+  test_ok(%r'\A//[^/]+/[^/]+\z' =~ File.expand_path(".", "/"))
+  test_ok(%r'\A//[^/]+/[^/]+/a\z' =~ File.expand_path(".", "/"))
+  dosish = true
+else
+  test_ok(File.expand_path(".", "/") == "/")
+  test_ok(File.expand_path("sub", "/") == "/sub")
+end
+if dosish
+  test_ok(File.expand_path("/", "//machine/share/sub") == "//machine/share")
+  test_ok(File.expand_path("/dir", "//machine/share/sub") == "//machine/share/dir")
+  test_ok(File.expand_path("/", "z:/sub") == "z:/")
+  test_ok(File.expand_path("/dir", "z:/sub") == "z:/dir")
+end
+test_ok(File.expand_path(".", "//") == "//")
+test_ok(File.expand_path("sub", "//") == "//sub")
+
+# test_check "Proc#binding"
+ObjectSpace.each_object(Proc){|o|
+  begin
+    b = o.binding
+    eval 'self', b
+  rescue ArgumentError
+  end
+}
+
+test_check "gc"
+begin
+  1.upto(10000) {
+    tmp = [0,1,2,3,4,5,6,7,8,9]
+  }
+  tmp = nil
+  test_ok true
+rescue
+  test_ok false
+end
+class S
+  def initialize(a)
+    @a = a
+  end
+end
+l=nil
+100000.times {
+  l = S.new(l)
+}
+GC.start
+test_ok true   # reach here or dumps core
+l = []
+100000.times {
+  l.push([l])
+}
+GC.start
+test_ok true   # reach here or dumps core
+
+ObjectSpace.each_object{|o|
+  o.class.name
+}
+
+test_ok true   # reach here or dumps core
+
+PROGRESS.finish
+if $failed > 0
+  printf "not ok/test: %d failed %d\n", $ntest, $failed
+else
+  printf "end of test(test: %d)\n", $ntest
+end
diff --git a/bcc32/Makefile.sub b/bcc32/Makefile.sub
deleted file mode 100644
index de7a2beeb8..0000000000
--- a/bcc32/Makefile.sub
+++ /dev/null
@@ -1,490 +0,0 @@
-# -*- makefile -*-
-
-SHELL = $(COMSPEC)
-MKFILES = Makefile
-
-#### Start of system configuration section. ####
-OS = bccwin32
-RT = $(OS)
-
-## variables may be overridden by $(compile_dir)/Makefile
-!ifndef srcdir
-srcdir = ..
-!endif
-!ifndef RUBY_INSTALL_NAME
-RUBY_INSTALL_NAME = ruby
-!endif
-!ifndef RUBYW_INSTALL_NAME
-RUBYW_INSTALL_NAME = $(RUBY_INSTALL_NAME:ruby=rubyw)
-!elif "$(RUBYW_INSTALL_NAME)" == "$(RUBY_INSTALL_NAME)"
-RUBYW_INSTALL_NAME = $(RUBY_INSTALL_NAME:ruby=rubyw)
-!endif
-!if "$(RUBYW_INSTALL_NAME)" == "$(RUBY_INSTALL_NAME)"
-RUBYW_INSTALL_NAME = $(RUBY_INSTALL_NAME)w
-!endif
-!ifndef RUBY_SO_NAME
-RUBY_SO_NAME = $(RT)-$(RUBY_INSTALL_NAME)$(MAJOR)$(MINOR)
-!endif
-!ifndef icondirs
-!ifdef ICONDIRS
-icondirs=$(ICONDIRS)
-!endif
-!endif
-!ifdef icondirs
-icondirs=$(icondirs:\=/)
-iconinc=-I$(icondirs: = -I)
-!endif
-###############
-
-VPATH = $(srcdir):$(srcdir)/missing
-.SUFFIXES: .y
-
-!ifndef CC
-CC = bcc32
-!endif
-!ifndef CPP
-CPP = cpp32
-!endif
-!ifndef RC
-RC = brcc32
-!endif
-!ifndef YACC
-YACC = byacc
-!endif
-!ifndef AR
-AR = tlib
-!endif
-
-PURIFY =
-AUTOCONF = autoconf
-RM = $(srcdir:/=\)\win32\rm.bat
-
-!if !defined(PROCESSOR_ARCHITECTURE)
-PROCESSOR_ARCHITECTURE = x86
-!endif
-MACHINE = $(PROCESSOR_ARCHITECTURE)
-!if "$(PROCESSOR_ARCHITECTURE)" == "x86"
-!ifndef PROCESSOR_LEVEL
-PROCESSOR_LEVEL = 5
-!endif
-!if 6 < $(PROCESSOR_LEVEL)
-PROCESSOR_LEVEL = 6
-!endif
-PROCESSOR_FLAG = -$(PROCESSOR_LEVEL)
-CPU = i$(PROCESSOR_LEVEL)86
-ARCH = i386
-!else
-CPU = $(PROCESSOR_ARCHITECTURE)
-ARCH = $(PROCESSOR_ARCHITECTURE)
-!endif
-!ifndef DEBUGFLAGS
-DEBUGFLAGS = 
-!endif
-!ifndef OPTFLAGS
-OPTFLAGS = -O
-!endif
-
-!ifndef prefix
-prefix = /usr
-!endif
-!ifndef exec_prefix
-exec_prefix = $(prefix)
-!endif
-!ifndef libdir
-libdir = $(exec_prefix)/lib
-!endif
-!if !defined(datadir)
-datadir = /share
-!endif
-!ifndef EXTOUT
-EXTOUT = .ext
-!endif
-!ifndef RIDATADIR
-RIDATADIR = $(DESTDIR)$(datadir)/ri/$(MAJOR).$(MINOR)/system
-!endif
-!ifndef TESTUI
-TESTUI = console
-!endif
-!ifndef TESTS
-TESTS =
-!endif
-!ifndef RDOCTARGET
-RDOCTARGET = install-nodoc
-!endif
-
-OUTFLAG       = -o
-!ifndef CFLAGS
-CFLAGS = -q -tWR -tWC $(DEBUGFLAGS) $(OPTFLAGS) $(PROCESSOR_FLAG) -w- -wsus -wcpt -wdup -wext -wrng -wrpt -wzdi
-!endif
-!ifndef LDFLAGS
-LDFLAGS = -S:$(STACK)
-!endif
-!ifndef RFLAGS
-RFLAGS = $(iconinc)
-!endif
-!ifndef EXTLIBS
-EXTLIBS =
-!endif
-!ifndef MEMLIB
-MEMLIB =
-!endif
-LIBS = $(MEMLIB) cw32i.lib import32.lib ws2_32.lib $(EXTLIBS)
-MISSING = acosh.obj crypt.obj erf.obj win32.obj
-
-!ifndef STACK
-STACK = 0x2000000
-!endif
-
-XCFLAGS = -DRUBY_EXPORT -I. -I$(srcdir) -I$(srcdir)/missing
-
-ARFLAGS = /a
-LD = ilink32 -q -Gn
-LDSHARED  = $(LD)
-XLDFLAGS  = -Tpe c0x32.obj
-WLDFLAGS  = -aa -Tpe c0w32.obj
-DLDFLAGS  = -Tpd c0d32.obj
-LIBRUBY_LDSHARED = $(LDSHARED)
-LIBRUBY_DLDFLAGS = -Gi $(DLDFLAGS) $(EXTLDFLAGS)
-LDOBJECTS = $(MAINOBJ)
-
-SOLIBS = 
-
-EXEEXT = .exe
-PROGRAM=$(RUBY_INSTALL_NAME)$(EXEEXT)
-WPROGRAM=$(RUBYW_INSTALL_NAME)$(EXEEXT)
-RUBYDEF = $(RUBY_SO_NAME).def
-MINIRUBY = .\miniruby$(EXEEXT)
-RUNRUBY = .\ruby$(EXEEXT) "$(srcdir)/runruby.rb" --extout="$(EXTOUT)" --
-
-ORGLIBPATH = $(LIB)
-
-#### End of system configuration section. ####
-
-LIBRUBY_A     = $(RUBY_SO_NAME)-static.lib
-LIBRUBY_SO    = $(RUBY_SO_NAME).dll
-LIBRUBY       = $(RUBY_SO_NAME).lib
-LIBRUBYARG    = $(LIBRUBY)
-
-PREP          = miniruby$(EXEEXT)
-
-OBJEXT        = obj
-
-WINMAINOBJ    = winmain.$(OBJEXT)
-MINIOBJS      = dmydln.$(OBJEXT)
-
-.path.c = .;$(srcdir);$(srcdir)/win32;$(srcdir)/missing
-.path.h = .;$(srcdir);$(srcdir)/win32;$(srcdir)/missing
-.path.y = $(srcdir)
-.path. = $(srcdir)
-
-.c.obj:
-	$(CC) $(CFLAGS) $(XCFLAGS) -I. $(CPPFLAGS) -c $(<:/=\)
-
-.rc.res:
-	$(RC) $(RFLAGS) -I. -I$(<D). $(iconinc) -I$(srcdir)/win32 $(RFLAGS) -fo$@ $(<:/=\)
-
-.y.c:
-	$(YACC) $(YFLAGS) $(<:\=/)
-	sed -e "s!^ *extern char \*getenv();!/* & */!;s/^\(#.*\)y\.tab/\1parse/" y.tab.c > $(@F)
-	@del y.tab.c
-
-all: $(srcdir)/bcc32/Makefile.sub $(srcdir)/common.mk
-
-ruby: $(PROGRAM)
-rubyw: $(WPROGRAM)
-
-!include $(srcdir)/common.mk
-
-PHONY: Makefile
-
-CONFIG_H = ./.config.h.time
-
-config: config.status
-
-config.status: $(CONFIG_H)
-
-$(CONFIG_H): $(MKFILES) $(srcdir)/bcc32/Makefile.sub
-	@$(srcdir:/=\)\win32\ifchange.bat config.h &&|
-\#define HAVE_SYS_TYPES_H 1
-\#define HAVE_SYS_STAT_H 1
-\#define HAVE_STDLIB_H 1
-\#define HAVE_STRING_H 1
-\#define HAVE_MEMORY_H 1
-\#define HAVE_OFF_T 1
-\#define SIZEOF_INT 4
-\#define SIZEOF_SHORT 2
-\#define SIZEOF_LONG 4
-\#define SIZEOF_LONG_LONG 0
-\#define SIZEOF___INT64 8
-\#define SIZEOF_OFF_T 4
-\#define SIZEOF_VOIDP 4
-\#define SIZEOF_FLOAT 4
-\#define SIZEOF_DOUBLE 8
-\#define SIZEOF_TIME_T 4
-\#define HAVE_PROTOTYPES 1
-\#define TOKEN_PASTE(x,y) x\#\#y
-\#define HAVE_STDARG_PROTOTYPES 1
-\#define NORETURN(x) x
-\#define RUBY_EXTERN extern __declspec(dllimport)
-\#define HAVE_DECL_SYS_NERR 1
-\#define HAVE_LIMITS_H 1
-\#define HAVE_FCNTL_H 1
-\#define HAVE_UTIME_H 1
-\#define HAVE_FLOAT_H 1
-\#define rb_uid_t uid_t
-\#define rb_gid_t gid_t
-\#define rb_pid_t int
-\#define HAVE_STRUCT_STAT_ST_RDEV 1
-\#define HAVE_ST_RDEV 1
-\#define GETGROUPS_T int
-\#define RETSIGTYPE void
-\#define HAVE_ALLOCA 1
-\#define HAVE_DUP2 1
-\#define HAVE_MEMMOVE 1
-\#define HAVE_MKDIR 1
-\#define HAVE_STRCASECMP 1
-\#define HAVE_STRNCASECMP 1
-\#define HAVE_STRERROR 1
-\#define HAVE_STRFTIME 1
-\#define HAVE_STRCHR 1
-\#define HAVE_STRSTR 1
-\#define HAVE_STRTOD 1
-\#define HAVE_STRTOL 1
-\#define HAVE_STRTOUL 1
-\#define HAVE_ISNAN 1
-\#define HAVE_FINITE 1
-\#define HAVE_HYPOT 1
-\#define HAVE_FMOD 1
-\#define HAVE_WAITPID 1
-\#define HAVE_FSYNC 1
-\#define HAVE_GETCWD 1
-\#define HAVE_CHSIZE 1
-\#define HAVE_TIMES 1
-\#define HAVE_FCNTL 1
-\#define HAVE_LINK 1
-\#define HAVE_TELLDIR 1
-\#define HAVE_SEEKDIR 1
-\#define HAVE_COSH 1
-\#define HAVE_SINH 1
-\#define HAVE_TANH 1
-\#define RSHIFT(x,y) ((x)>>(int)y)
-\#define FILE_COUNT level
-\#define FILE_READPTR curp
-\#define inline __inline
-\#define NEED_IO_SEEK_BETWEEN_RW 1
-\#define STACK_GROW_DIRECTION -1
-\#define DEFAULT_KCODE KCODE_NONE
-\#define DLEXT ".so"
-\#define RUBY_LIB "/lib/ruby/$(MAJOR).$(MINOR)"
-\#define RUBY_SITE_LIB "/lib/ruby/site_ruby"
-\#define RUBY_SITE_LIB2 "/lib/ruby/site_ruby/$(MAJOR).$(MINOR)"
-\#define RUBY_PLATFORM "$(ARCH)-$(OS)"
-\#define RUBY_ARCHLIB "/lib/ruby/$(MAJOR).$(MINOR)/$(ARCH)-$(OS)"
-\#define RUBY_SITE_ARCHLIB "/lib/ruby/site_ruby/$(MAJOR).$(MINOR)/$(ARCH)-$(OS)"
-|
-	@exit > $@
-
-config.status: $(MKFILES) $(srcdir)/bcc32/Makefile.sub $(srcdir)/common.mk
-	@echo Creating $@
-	@type > $@ &&|
-# Generated automatically by Makefile.sub.
-s,@SHELL@,$$(COMSPEC),;t t
-s,@BUILD_FILE_SEPARATOR@,\,;t t
-s,@PATH_SEPARATOR@,;,;t t
-s,@CFLAGS@,$(CFLAGS),;t t
-s,@CPPFLAGS@,$(CPPFLAGS),;t t
-s,@CXXFLAGS@,$(CXXFLAGS),;t t
-s,@FFLAGS@,$(FFLAGS),;t t
-s,@LDFLAGS@,,;t t
-s,@LIBS@,$(LIBS),;t t
-s,@exec_prefix@,$${prefix},;t t
-s,@prefix@,,;t t
-s,@program_transform_name@,s,,,,;t t
-s,@bindir@,$${exec_prefix}/bin,;t t
-s,@sbindir@,$${exec_prefix}/sbin,;t t
-s,@libexecdir@,$${exec_prefix}/libexec,;t t
-s,@datadir@,$${prefix}/share,;t t
-s,@sysconfdir@,$${prefix}/etc,;t t
-s,@sharedstatedir@,/etc,;t t
-s,@localstatedir@,/var,;t t
-s,@libdir@,$${exec_prefix}/lib,;t t
-s,@includedir@,$${prefix}/include,;t t
-s,@oldincludedir@,/usr/include,;t t
-s,@infodir@,$${prefix}/info,;t t
-s,@mandir@,$${prefix}/man,;t t
-s,@build@,$(CPU)-pc-$(OS),;t t
-s,@build_alias@,$(CPU)-$(OS),;t t
-s,@build_cpu@,$(CPU),;t t
-s,@build_vendor@,pc,;t t
-s,@build_os@,$(OS),;t t
-s,@host@,$(CPU)-pc-$(OS),;t t
-s,@host_alias@,$(CPU)-$(OS),;t t
-s,@host_cpu@,$(CPU),;t t
-s,@host_vendor@,pc,;t t
-s,@host_os@,$(OS),;t t
-s,@target@,$(ARCH)-pc-$(OS),;t t
-s,@target_alias@,$(ARCH)-$(OS),;t t
-s,@target_cpu@,$(ARCH),;t t
-s,@target_vendor@,pc,;t t
-s,@target_os@,$(OS),;t t
-s,@CC@,$(CC),;t t
-s,@CPP@,cpp32,;t t
-s,@YACC@,$(YACC),;t t
-s,@RANLIB@,,;t t
-s,@AR@,$(AR),;t t
-s,@ARFLAGS@,$(ARFLAGS) ,;t t
-s,@LN_S@,$(LN_S),;t t
-s,@SET_MAKE@,$(SET_MAKE),;t t
-s,@CP@,copy > nul,;t t
-s,@INSTALL@,copy > nul,;t t
-s,@INSTALL_PROG@,$$(INSTALL),;t t
-s,@INSTALL_DATA@,$$(INSTALL),;t t
-s,@LIBOBJS@, acosh.obj crypt.obj erf.obj win32.obj,;t t
-s,@ALLOCA@,$(ALLOCA),;t t
-s,@DEFAULT_KCODE@,$(DEFAULT_KCODE),;t t
-s,@EXEEXT@,.exe,;t t
-s,@OBJEXT@,obj,;t t
-s,@XCFLAGS@,$(XCFLAGS),;t t
-s,@XLDFLAGS@,$(XLDFLAGS),;t t
-s,@DLDFLAGS@,$(DLDFLAGS),;t t
-s,@ARCH_FLAG@,$(ARCH_FLAG),;t t
-s,@STATIC@,$(STATIC),;t t
-s,@CCDLFLAGS@,,;t t
-s,@LDSHARED@,$(LDSHARED),;t t
-s,@DLEXT@,so,;t t
-s,@LIBEXT@,lib,;t t
-s,@STRIP@,$(STRIP),;t t
-s,@EXTSTATIC@,$(EXTSTATIC),;t t
-s,@setup@,Setup,;t t
-s,@MINIRUBY@,$(MINIRUBY),;t t
-s,@PREP@,miniruby$(EXEEXT),;t t
-s,@RUNRUBY@,$(RUNRUBY),;t t
-s,@EXTOUT@,$(EXTOUT),;t t
-s,@ARCHFILE@,,;t t
-s,@RDOCTARGET@,,;t t
-s,@LIBRUBY_LDSHARED@,$$(LDSHARED),;t t
-s,@LIBRUBY_DLDFLAGS@,-Gi $$(DLDFLAGS),;t t
-s,@RUBY_INSTALL_NAME@,$(RUBY_INSTALL_NAME),;t t
-s,@rubyw_install_name@,$(RUBYW_INSTALL_NAME),;t t
-s,@RUBYW_INSTALL_NAME@,$(RUBYW_INSTALL_NAME),;t t
-s,@RUBY_SO_NAME@,$(RUBY_SO_NAME),;t t
-s,@LIBRUBY_A@,$$(RUBY_SO_NAME)-static.lib,;t t
-s,@LIBRUBY_SO@,$$(RUBY_SO_NAME).dll,;t t
-s,@LIBRUBY_ALIASES@,$(LIBRUBY_ALIASES),;t t
-s,@LIBRUBY@,$$(RUBY_SO_NAME).lib,;t t
-s,@LIBRUBYARG@,$$(LIBRUBYARG_SHARED),;t t
-s,@LIBRUBYARG_STATIC@,$$(LIBRUBY_A),;t t
-s,@LIBRUBYARG_SHARED@,$$(LIBRUBY),;t t
-s,@SOLIBS@,$(SOLIBS),;t t
-s,@DLDLIBS@,$(DLDLIBS),;t t
-s,@ENABLE_SHARED@,yes,;t t
-s,@OUTFLAG@,$(OUTFLAG),;t t
-s,@CPPOUTFILE@,,;t t
-s,@LIBPATHFLAG@, -L"%s",;t t
-s,@RPATHFLAG@,,;t t
-s,@LIBARG@,%s.lib,;t t
-s,@LINK_SO@,$$(LDSHARED) $$(DLDFLAGS) $$(LIBPATH) $$(OBJS), $$(@:/=\), nul, $$(LIBS) $$(LOCAL_LIBS), $$(DEFFILE), $$(RESFILE),;t t
-s,@COMPILE_C@,$$(CC) $$(INCFLAGS) $$(CFLAGS) $$(CPPFLAGS) -c $$(<:/=\),;t t
-s,@COMPILE_CXX@,$$(CXX) $$(INCFLAGS) $$(CXXFLAGS) $$(CPPFLAGS) -P -c $$(<:/=\),;t t
-s,@COMPILE_RULES@,{$$(srcdir)}.%s{}.%s: {$$(topdir)}.%s{}.%s: {$$(hdrdir)}.%s{}.%s: .%s.%s:,;t t
-s,@RULE_SUBST@,{.;$$(VPATH)}%s,;t t
-s,@COMMON_LIBS@,m advapi32 avicap32 avifil32 cap comctl32 comdlg32 dlcapi gdi32 glu32 imagehlp imm32 inetmib1 kernel32 loadperf lsapi32 lz32 mapi32 mgmtapi mpr msacm32 msvfw32 nddeapi netapi32 ole32 oleaut32 oledlg olepro32 opengl32 pdh pkpd32 rasapi32 rasdlg rassapi rpcrt4 setupapi shell32 shfolder snmpapi sporder tapi32 url user32 vdmdbg version win32spl winmm wintrust wsock32,;t t
-s,@COMMON_MACROS@,WIN32_LEAN_AND_MEAN WIN32,;t t
-s,@COMMON_HEADERS@,winsock2.h windows.h,;t t
-s,@TRY_LINK@,$$(CC) -oconftest $$(INCFLAGS) -I$$(hdrdir) $$(CPPFLAGS) $$(CFLAGS) $$(LIBPATH) $$(LDFLAGS) $$(src) $$(LOCAL_LIBS) $$(LIBS),;t t
-s,@EXPORT_PREFIX@,_,;t t
-s,@arch@,$(ARCH)-$(OS),;t t
-s,@sitearch@,$(ARCH)-$(OS),;t t
-s,@sitedir@,$${prefix}/lib/ruby/site_ruby,;t t
-s,@configure_args@,--enable-shared $(configure_args),;t t
-s,@configure_input@,$$configure_input,;t t
-s,@srcdir@,$(srcdir),;t t
-s,@top_srcdir@,$(srcdir),;t t
-|
-
-miniruby$(EXEEXT):
-		@echo $(LIBS)
-		$(LD) $(LDFLAGS) $(XLDFLAGS) $(MAINOBJ) $(MINIOBJS),$@,nul,$(LIBRUBY_A) $(LIBS)
-
-$(PROGRAM):	$(MAINOBJ) $(LIBRUBY_SO) $(RUBY_INSTALL_NAME).res
-		$(LD) $(LDFLAGS) $(XLDFLAGS) $(MAINOBJ),$@,nul,$(LIBRUBYARG) $(LIBS),,$(RUBY_INSTALL_NAME).res
-
-$(WPROGRAM):	$(MAINOBJ) $(WINMAINOBJ) $(LIBRUBY_SO) $(RUBYW_INSTALL_NAME).res
-		$(LD) $(LDFLAGS) $(WLDFLAGS) $(MAINOBJ) $(WINMAINOBJ),$@,nul,$(LIBRUBYARG) $(LIBS),,$(RUBYW_INSTALL_NAME).res
-
-$(LIBRUBY_A):	$(OBJS) $(DMYEXT)
-		@-if exist $@ del $@
-		$(AR) $(ARFLAGS) "$@" $(OBJS) $(DMYEXT)
-
-# $(LIBRUBY):	$(LIBRUBY_SO)
-#		implib $@ $(LIBRUBY_SO)
-
-$(LIBRUBY_SO): $(LIBRUBY_A) $(DLDOBJS) $(RUBYDEF) $(RUBY_SO_NAME).res
-		@echo $(DLDOBJS)
-		$(LIBRUBY_LDSHARED) $(LIBRUBY_DLDFLAGS) $(DLDOBJS:/=\),$(LIBRUBY_SO),nul,$(LIBRUBY_A) $(LIBS),$(RUBYDEF),$(RUBY_SO_NAME).res
-
-$(LIBRUBY): $(LIBRUBY_SO)
-
-$(RUBYDEF):	$(LIBRUBY_A) $(PREP)
-		$(MINIRUBY) $(srcdir)/bcc32/mkexports.rb -output=$@ -base=$(RUBY_SO_NAME) $(LIBRUBY_A)
-
-$(RUBY_INSTALL_NAME).rc $(RUBYW_INSTALL_NAME).rc $(RUBY_SO_NAME).rc: rbconfig.rb
-		@$(MINIRUBY) $(srcdir)/win32/resource.rb \
-			-ruby_name=$(RUBY_INSTALL_NAME) \
-			-rubyw_name=$(RUBYW_INSTALL_NAME) \
-			-so_name=$(RUBY_SO_NAME) \
-			. $(icondirs) $(srcdir)/win32
-
-post-install-ext::
-	$(MINIRUBY) -I$(srcdir)lib -rrbconfig -rfileutils \
-		-e "FileUtils.rm_f(Dir[ARGV[0]+Config::CONFIG['archdir']+'/**/*.tds'])" "$(DESTDIR:\=/)"
-
-clean-local::
-		@$(RM) ext\extinit.c ext\extinit.$(OBJEXT) *.tds *.il? $(RUBY_SO_NAME).lib
-		@$(RM) $(RUBY_INSTALL_NAME).res $(RUBYW_INSTALL_NAME).res $(RUBY_SO_NAME).res
-
-distclean-local::
-		@$(RM) ext\config.cache $(RBCONFIG:/=\)
-		@$(RM) *.map *.pdb *.ilk *.exp $(RUBYDEF)
-		@$(RM) $(RUBY_INSTALL_NAME).rc $(RUBYW_INSTALL_NAME).rc $(RUBY_SO_NAME).rc
-
-ext/extinit.obj: ext/extinit.c $(SETUP)
-	$(CC) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) -o$@ -c ext/extinit.c
-
-main.$(OBJEXT): win32.h
-array.$(OBJEXT): win32.h
-bignum.$(OBJEXT): win32.h
-class.$(OBJEXT): win32.h
-compar.$(OBJEXT): win32.h
-dir.$(OBJEXT): dir.h win32.h
-dln.$(OBJEXT): win32.h
-enum.$(OBJEXT): win32.h
-error.$(OBJEXT): win32.h
-eval.$(OBJEXT): win32.h
-file.$(OBJEXT): win32.h
-gc.$(OBJEXT): win32.h
-hash.$(OBJEXT): win32.h
-inits.$(OBJEXT): win32.h
-io.$(OBJEXT): win32.h
-marshal.$(OBJEXT): win32.h
-math.$(OBJEXT): win32.h
-numeric.$(OBJEXT): win32.h
-object.$(OBJEXT): win32.h
-pack.$(OBJEXT): win32.h
-parse.$(OBJEXT): win32.h
-process.$(OBJEXT): win32.h
-prec.$(OBJEXT): win32.h
-random.$(OBJEXT): win32.h
-range.$(OBJEXT): win32.h
-re.$(OBJEXT): win32.h
-regex.$(OBJEXT): win32.h
-ruby.$(OBJEXT): win32.h
-signal.$(OBJEXT): win32.h
-sprintf.$(OBJEXT): win32.h
-st.$(OBJEXT): win32.h
-string.$(OBJEXT): win32.h
-struct.$(OBJEXT): win32.h
-time.$(OBJEXT): win32.h
-util.$(OBJEXT): win32.h
-variable.$(OBJEXT): win32.h
-version.$(OBJEXT): win32.h
diff --git a/bcc32/README.bcc32 b/bcc32/README.bcc32
deleted file mode 100644
index c27a1261f1..0000000000
--- a/bcc32/README.bcc32
+++ /dev/null
@@ -1,137 +0,0 @@
-=begin
-
-= How to build ruby using Borland C++
-
-== Requirement
-
-(1) Borland C++ 5.0 or later.
-
-(2) Please set environment variable (({PATH}))
-    to run required commands properly from the command line.
-
-    Note: building ruby requires following commands.
-     * make
-     * bcc32
-     * tlib
-     * ilink32
-
-(3) If you want to build from CVS source, following commands are required.
-     * byacc ((<URL:http://gnuwin32.sourceforge.net/packages/byacc.htm>))
-     * sed   ((<URL:http://gnuwin32.sourceforge.net/packages/sed.htm>))
-
-(4) We strongly recommend to build ruby on C++Builder, to link following files.
-     * usebormm.lib
-     * memmgr.lib
-
-    RTL's internal memory manager cannot handle large memory block properly,
-    so we should use borlndmm.dll instead.
-     10000.times { "" << "." * 529671; GC.start } # crash
-
-== How to compile and install
-
-(1) Execute bcc32\configure.bat on your build directory.
-     ex. c:\ruby-1.6.7>bcc32\configure.bat
-     
-(2) Change ((|RUBY_INSTALL_NAME|)) and ((|RUBY_SO_NAME|)) in (({Makefile}))
-    if you want to change the name of the executable files. 
-    And add ((|RUBYW_INSTALL_NAME|)) to change the name of the
-    executable without console window if also you want.
-
-(3) Run `((%make%))'
-
-(4) Run `((%make test%))'
-
-(5) Run `((%make DESTDIR=<install_directory> install%))'
-
-    This command will create following directories and install files onto them.
-      * <install_directory>\bin
-      * <install_directory>\lib
-      * <install_directory>\lib\ruby
-      * <install_directory>\lib\ruby\<MAJOR>.<MINOR>
-      * <install_directory>\lib\ruby\<MAJOR>.<MINOR>\<PLATFORM>
-      * <install_directory>\lib\ruby\site_ruby
-      * <install_directory>\lib\ruby\site_ruby\<MAJOR>.<MINOR>
-      * <install_directory>\lib\ruby\site_ruby\<MAJOR>.<MINOR>\<PLATFORM>
-      * <install_directory>\man\man1
-    If Ruby's version is `x.y.z', the ((|<MAJOR>|)) is `x' and the ((|<MINOR>|)) is `y'.
-    The ((|<PLATFORM>|)) is usually `(({i586-bccwin32}))'.
-
-(6) Requires dynamic RTL (cc3250.dll on C++Builder5) and borlndmm.dll (If built with
-    usebormm.lib) to use installed binary. These files are ordinary in bcc32's bin
-    directory.
-
-== Icons
-
-Any icon files(*.ico) in the build directory, directories specified with
-((|icondirs|)) make variable and (({win32})) directory under the ruby
-source directory will be included in DLL or executable files, according
-to their base names.
-    $(RUBY_INSTALL_NAME).ico or ruby.ico   --> $(RUBY_INSTALL_NAME).exe
-    $(RUBYW_INSTALL_NAME).ico or rubyw.ico --> $(RUBYW_INSTALL_NAME).exe
-    the others                             --> $(RUBY_SO_NAME).dll
-
-Although no icons are distributed with the ruby source or in the official 
-site, you can use anything you like. For example, followings are written 
-in Japanese, but you can download at least.
-
-* ((<URL:http://member.nifty.ne.jp/ueivu/rubyico.html>)) or
-  ((<zipped icons|URL:http://member.nifty.ne.jp/ueivu/Ruby_ico.zip>))
-* ((<URL:http://homepage1.nifty.com/a_nakata/ruby/>)) or
-  ((<icon itself|URL:http://homepage1.nifty.com/a_nakata/ruby/RubyIcon.ico>))
-
-== Build examples
-
-* Build on the ruby source directory.
-
-  ex.)
-    ruby source directory:  C:\ruby
-    build directory:        C:\ruby
-    install directory:      C:\usr\local
-
-    C:
-    cd \ruby
-    bcc32\configure
-    make
-    make test
-    make DESTDIR=/usr/local install
-
-* Build on the relative directory from the ruby source directory and CPU type 
-  i386.
-
-  ex.)
-    ruby source directory:  C:\ruby
-    build directory:        C:\ruby\bccwin32
-    install directory:      C:\usr\local
-    CPU                     i386
-    
-    C:
-    cd \ruby
-    mkdir bccwin32
-    cd bccwin32
-    ..\bcc32\configure target i386-bccwin32
-    make
-    make test
-    make DESTDIR=/usr/local install
-
-* Build on the different drive.
-
-  ex.)
-    ruby source directory:  C:\src\ruby
-    build directory:        D:\build\ruby
-    install directory:      C:\usr\local
-
-    D:
-    cd D:\build\ruby
-    C:\src\ruby\bcc32\configure
-    make
-    make test
-    make DESTDIR=C:/usr/local install
-
-== Bugs
-
-You can ((*NOT*)) use a path name contains any white space characters as
-the ruby source directory, this restriction comes from the behavior of
-(({!INCLUDE})) directives of (({MAKE})).
-((- you may call it a bug. -))
-
-=end
diff --git a/bcc32/configure.bat b/bcc32/configure.bat
deleted file mode 100755
index 123a3f23c8..0000000000
--- a/bcc32/configure.bat
+++ /dev/null
@@ -1,92 +0,0 @@
-@echo off
-::: Don't set environment variable in batch file other than autoexec.bat
-::: to avoid "Out of environment space" problem on Windows 95/98.
-::: set TMPMAKE=~tmp~.mak
-
-echo> ~tmp~.mak ####
-echo>> ~tmp~.mak conf = %0
-echo>> ~tmp~.mak $(conf:\=/): nul
-echo>> ~tmp~.mak 	@del ~tmp~.mak
-echo>> ~tmp~.mak 	@-$(MAKE) -l$(MAKEFLAGS) -f $(@D)setup.mak \
-:loop
-if "%1" == "" goto :end
-if "%1" == "--prefix" goto :prefix
-if "%1" == "--srcdir" goto :srcdir
-if "%1" == "srcdir" goto :srcdir
-if "%1" == "--target" goto :target
-if "%1" == "target" goto :target
-if "%1" == "--with-static-linked-ext" goto :extstatic
-if "%1" == "--program-suffix" goto :suffix
-if "%1" == "--program-name" goto :progname
-if "%1" == "--enable-install-doc" goto :enable-rdoc
-if "%1" == "--disable-install-doc" goto :disable-rdoc
-if "%1" == "--extout" goto :extout
-if "%1" == "-h" goto :help
-if "%1" == "--help" goto :help
-  echo>> ~tmp~.mak 	"%1" \
-  shift
-goto :loop
-:srcdir
-  echo>> ~tmp~.mak 	-D"srcdir=%2" \
-  shift
-  shift
-goto :loop
-:prefix
-  echo>> ~tmp~.mak 	-D"prefix=%2" \
-  shift
-  shift
-goto :loop
-:suffix
-  echo>> ~tmp~.mak 	-D"RUBY_SUFFIX=%2" \
-  shift
-  shift
-goto :loop
-:installname
-  echo>> ~tmp~.mak 	-D"RUBY_INSTALL_NAME=%2" \
-  shift
-  shift
-goto :loop
-:soname
-  echo>> ~tmp~.mak 	-D"RUBY_SO_NAME=%2" \
-  shift
-  shift
-goto :loop
-:target
-  echo>> ~tmp~.mak 	"%2" \
-  shift
-  shift
-goto :loop
-:extstatic
-  echo>> ~tmp~.mak 	-D"EXTSTATIC=static" \
-  shift
-goto :loop
-:enable-rdoc
-  echo>> ~tmp~.mak 	-D"RDOCTARGET=install-doc" \
-  shift
-goto :loop
-:disable-rdoc
-  echo>> ~tmp~.mak 	-D"RDOCTARGET=install-nodoc" \
-  shift
-goto :loop
-:extout
-  echo>> ~tmp~.mak 	"EXTOUT=%2" \
-  shift
-  shift
-goto :loop
-:help
-  echo Configuration:
-  echo   --help                  display this help
-  echo   --srcdir=DIR            find the sources in DIR [configure dir or `..']
-  echo Installation directories:
-  echo   --prefix=PREFIX         install files in PREFIX (ignored currently)
-  echo System types:
-  echo   --target=TARGET         configure for TARGET [i386-bccwin32]
-  echo Optional Package:
-  echo   --with-static-linked-ext link external modules statically
-  echo   --enable-install-doc    install rdoc indexes during install
-  del ~tmp~.mak
-goto :exit
-:end
-echo>> ~tmp~.mak 	bcc32dir="$(@D)"
-make -s -f ~tmp~.mak
-:exit
diff --git a/bcc32/mkexports.rb b/bcc32/mkexports.rb
deleted file mode 100644
index e34b441e2f..0000000000
--- a/bcc32/mkexports.rb
+++ /dev/null
@@ -1,25 +0,0 @@
-#!./miniruby -s
-
-SYM = {}
-STDIN.reopen(open("nul"))
-ARGV.each do |obj|
-  IO.foreach("|tdump -q -oiPUBDEF -oiPUBD32 #{obj.tr('/', '\\')}") do |l|
-    next unless /(?:PUBDEF|PUBD32)/ =~ l
-    SYM[$1] = true if /'(.*?)'/ =~ l
-  end
-end
-
-exports = []
-if $name
-  exports << "Name " + $name
-elsif $library
-  exports << "Library " + $library
-end
-exports << "Description " + $description.dump if $description
-exports << "EXPORTS" << SYM.keys.sort
-
-if $output
-  open($output, 'w') {|f| f.puts exports.join("\n")}
-else
-  puts exports.join("\n")
-end
diff --git a/bcc32/setup.mak b/bcc32/setup.mak
deleted file mode 100644
index b7a2539d0a..0000000000
--- a/bcc32/setup.mak
+++ /dev/null
@@ -1,133 +0,0 @@
-# -*- makefile -*-
-
-!if "$(srcdir)" != ""
-bcc32dir = $(srcdir)/bcc32
-!elseif "$(bcc32dir)" == "bcc32/"
-srcdir = .
-!elseif "$(bcc32dir:/bcc32/=)/bcc32/" == "$(bcc32dir)"
-srcdir = $(bcc32dir:/bcc32/=)
-!else
-srcdir = $(bcc32dir)/..
-!endif
-!ifndef prefix
-prefix = /usr
-!endif
-OS = bccwin32
-RT = $(OS)
-BANG = !
-APPEND = echo>>$(MAKEFILE)
-!ifdef MAKEFILE
-MAKE = $(MAKE) -f $(MAKEFILE)
-!else
-MAKEFILE = Makefile
-!endif
-
-all: Makefile
-Makefile: -prologue- -generic- -epilogue-
-i386-$(OS): -prologue- -i386- -epilogue-
-i486-$(OS): -prologue- -i486- -epilogue-
-i586-$(OS): -prologue- -i586- -epilogue-
-i686-$(OS): -prologue- -i686- -epilogue-
-alpha-$(OS): -prologue- -alpha- -epilogue-
-
--prologue-: nul
-	@echo Creating $(MAKEFILE)
-	@type > $(MAKEFILE) &&|
-\#\#\# Makefile for ruby $(OS) \#\#\#
-$(BANG)ifndef srcdir
-srcdir = $(srcdir:\=/)
-$(BANG)endif
-$(BANG)ifndef prefix
-prefix = $(prefix:\=/)
-$(BANG)endif
-$(BANG)ifndef EXTSTATIC
-EXTSTATIC = $(EXTSTATIC)
-$(BANG)endif
-!if defined(RDOCTARGET)
-$(BANG)ifndef RDOCTARGET
-RDOCTARGET = $(RDOCTARGET)
-$(BANG)endif
-!endif
-!if defined(EXTOUT)
-$(BANG)ifndef EXTOUT
-EXTOUT = $(EXTOUT)
-$(BANG)endif
-!endif
-|
-	@type > usebormm.bat &&|
-@echo off
-ilink32 -Gn -x usebormm.lib > nul
-if exist usebormm.tds echo MEMLIB = usebormm.lib
-|
-	@usebormm.bat >> $(MAKEFILE)
-	@del usebormm.*
-
-	@cpp32 -I$(srcdir) -P- -o$(MAKEFILE) > nul &&|
-\#include "version.h"
-MAJOR = RUBY_VERSION_MAJOR
-MINOR = RUBY_VERSION_MINOR
-TEENY = RUBY_VERSION_TEENY
-|
-	@type $(MAKEFILE).i >> $(MAKEFILE)
-	@del $(MAKEFILE).i
-
--generic-: nul
-!if defined(PROCESSOR_ARCHITECTURE) ||  defined(PROCESSOR_LEVEL)
-	@type >> $(MAKEFILE) &&|
-!if defined(PROCESSOR_ARCHITECTURE)
-$(BANG)ifndef PROCESSOR_ARCHITECTURE
-PROCESSOR_ARCHITECTURE = $(PROCESSOR_ARCHITECTURE)
-$(BANG)endif
-!endif
-!if defined(PROCESSOR_LEVEL)
-$(BANG)ifndef PROCESSOR_LEVEL
-PROCESSOR_LEVEL = $(PROCESSOR_LEVEL)
-$(BANG)endif
-!endif
-
-|
-!endif
-
--alpha-: nul
-	@$(APPEND) !ifndef PROCESSOR_ARCHITECTURE
-	@$(APPEND) PROCESSOR_ARCHITECTURE = alpha
-	@$(APPEND) !endif
--ix86-: nul
-	@$(APPEND) !ifndef PROCESSOR_ARCHITECTURE
-	@$(APPEND) PROCESSOR_ARCHITECTURE = x86
-	@$(APPEND) !endif
-
--i386-: -ix86-
-	@$(APPEND) !ifndef PROCESSOR_LEVEL
-	@$(APPEND) PROCESSOR_LEVEL = 3
-	@$(APPEND) !endif
--i486-: -ix86-
-	@$(APPEND) !ifndef PROCESSOR_LEVEL
-	@$(APPEND) PROCESSOR_LEVEL = 4
-	@$(APPEND) !endif
--i586-: -ix86-
-	@$(APPEND) !ifndef PROCESSOR_LEVEL
-	@$(APPEND) PROCESSOR_LEVEL = 5
-	@$(APPEND) !endif
--i686-: -ix86-
-	@$(APPEND) !ifndef PROCESSOR_LEVEL
-	@$(APPEND) PROCESSOR_LEVEL = 6
-	@$(APPEND) !endif
-
--epilogue-: nul
-	@type >> $(MAKEFILE) &&|
-
-\# OS = $(OS)
-\# RT = $(RT)
-\# RUBY_INSTALL_NAME = ruby
-\# RUBY_SO_NAME = $$(RT)-$$(RUBY_INSTALL_NAME)$$(MAJOR)$$(MINOR)
-\# CFLAGS = -q $$(DEBUGFLAGS) $$(OPTFLAGS) $$(PROCESSOR_FLAG) -w- -wsus -wcpt -wdup -wext -wrng -wrpt -wzdi
-\# CPPFLAGS = -I. -I$$(srcdir) -I$$(srcdir)/missing -DLIBRUBY_SO=\"$$(LIBRUBY_SO)\"
-\# STACK = 0x2000000
-\# LDFLAGS = -S:$$(STACK)
-\# RFLAGS = $$(iconinc)
-\# EXTLIBS = cw32.lib import32.lib user32.lib kernel32.lib
-$(BANG)include $$(srcdir)/bcc32/Makefile.sub
-|
-	@$(srcdir:/=\)\win32\rm.bat config.h config.status
-	@echo type "`$(MAKE)'" to make ruby for $(OS).
diff --git a/benchmark/README.md b/benchmark/README.md
new file mode 100644
index 0000000000..24a2669143
--- /dev/null
+++ b/benchmark/README.md
@@ -0,0 +1,72 @@
+# ruby/benchmark
+
+This directory has benchmark definitions to be run with
+[benchmark\_driver.gem](https://github.com/benchmark-driver/benchmark-driver).
+
+## Normal usage
+
+Execute `gem install benchmark_driver` and run a command like:
+
+```bash
+# Run a benchmark script with the ruby in the $PATH
+benchmark-driver benchmark/app_fib.rb
+
+# Run benchmark scripts with multiple Ruby executables or options
+benchmark-driver benchmark/*.rb -e /path/to/ruby -e '/path/to/ruby --jit'
+
+# Or compare Ruby versions managed by rbenv
+benchmark-driver benchmark/*.rb --rbenv '2.5.1;2.6.0-preview2 --jit'
+
+# You can collect many metrics in many ways
+benchmark-driver benchmark/*.rb --runner memory --output markdown
+
+# Some are defined with YAML for complex setup or accurate measurement
+benchmark-driver benchmark/*.yml
+```
+
+See also:
+
+```console
+Usage: benchmark-driver [options] RUBY|YAML...
+    -r, --runner TYPE                Specify runner type: ips, time, memory, once (default: ips)
+    -o, --output TYPE                Specify output type: compare, simple, markdown, record (default: compare)
+    -e, --executables EXECS          Ruby executables (e1::path1 arg1; e2::path2 arg2;...)
+        --rbenv VERSIONS             Ruby executables in rbenv (x.x.x arg1;y.y.y arg2;...)
+        --repeat-count NUM           Try benchmark NUM times and use the fastest result or the worst memory usage
+        --repeat-result TYPE         Yield "best", "average" or "worst" result with --repeat-count (default: best)
+        --bundler                    Install and use gems specified in Gemfile
+        --filter REGEXP              Filter out benchmarks with given regexp
+        --run-duration SECONDS       Warmup estimates loop_count to run for this duration (default: 3)
+    -v, --verbose                    Verbose mode. Multiple -v options increase visilibity (max: 2)
+```
+
+## make benchmark
+
+Using `make benchmark`, `make update-benchmark-driver` automatically downloads
+the supported version of benchmark\_driver, and it runs benchmarks with the downloaded
+benchmark\_driver.
+
+```bash
+# Run all benchmarks with the ruby in the $PATH and the built ruby
+make benchmark
+
+# Or compare with specific ruby binary
+make benchmark COMPARE_RUBY="/path/to/ruby --jit"
+
+# Run vm1 benchmarks
+make benchmark ITEM=vm1
+
+# Run some limited benchmarks in ITEM-matched files
+make benchmark ITEM=vm1 OPTS=--filter=block
+
+# You can specify the benchmark by an exact filename instead of using the default argument:
+# ARGS = $$(find $(srcdir)/benchmark -maxdepth 1 -name '*$(ITEM)*.yml' -o -name '*$(ITEM)*.rb')
+make benchmark ARGS=../benchmark/erb_render.yml
+
+# You can specify any option via $OPTS
+make benchmark OPTS="--help"
+
+# With `make benchmark`, some special runner plugins are available:
+#   -r peak, -r size, -r total, -r utime, -r stime, -r cutime, -r cstime
+make benchmark ITEM=vm2_bigarray OPTS="-r peak"
+```
diff --git a/benchmark/app_answer.rb b/benchmark/app_answer.rb
new file mode 100644
index 0000000000..3cd8a8fd37
--- /dev/null
+++ b/benchmark/app_answer.rb
@@ -0,0 +1,15 @@
+def ack(m, n)
+  if m == 0 then
+    n + 1
+  elsif n == 0 then
+    ack(m - 1, 1)
+  else
+    ack(m - 1, ack(m, n - 1))
+  end
+end
+
+def the_answer_to_life_the_universe_and_everything
+  (ack(3,7).to_s.split(//).inject(0){|s,x| s+x.to_i}.to_s + "2" ).to_i
+end
+
+answer = the_answer_to_life_the_universe_and_everything
diff --git a/benchmark/app_aobench.rb b/benchmark/app_aobench.rb
new file mode 100644
index 0000000000..2bd6acfaf8
--- /dev/null
+++ b/benchmark/app_aobench.rb
@@ -0,0 +1,291 @@
+# AO render benchmark
+# Original program (C) Syoyo Fujita in Javascript (and other languages)
+#      https://code.google.com/p/aobench/
+# Ruby(yarv2llvm) version by Hideki Miura
+#
+
+IMAGE_WIDTH = 256
+IMAGE_HEIGHT = 256
+NSUBSAMPLES = 2
+NAO_SAMPLES = 8
+
+class Vec
+  def initialize(x, y, z)
+    @x = x
+    @y = y
+    @z = z
+  end
+
+  attr_accessor :x, :y, :z
+
+  def vadd(b)
+    Vec.new(@x + b.x, @y + b.y, @z + b.z)
+  end
+
+  def vsub(b)
+    Vec.new(@x - b.x, @y - b.y, @z - b.z)
+  end
+
+  def vcross(b)
+    Vec.new(@y * b.z - @z * b.y,
+            @z * b.x - @x * b.z,
+            @x * b.y - @y * b.x)
+  end
+
+  def vdot(b)
+    @x * b.x + @y * b.y + @z * b.z
+  end
+
+  def vlength
+    Math.sqrt(@x * @x + @y * @y + @z * @z)
+  end
+
+  def vnormalize
+    len = vlength
+    v = Vec.new(@x, @y, @z)
+    if len > 1.0e-17 then
+      v.x = v.x / len
+      v.y = v.y / len
+      v.z = v.z / len
+    end
+    v
+  end
+end
+
+
+class Sphere
+  def initialize(center, radius)
+    @center = center
+    @radius = radius
+  end
+
+  attr_reader :center, :radius
+
+  def intersect(ray, isect)
+    rs = ray.org.vsub(@center)
+    b = rs.vdot(ray.dir)
+    c = rs.vdot(rs) - (@radius * @radius)
+    d = b * b - c
+    if d > 0.0 then
+      t = - b - Math.sqrt(d)
+
+      if t > 0.0 and t < isect.t then
+        isect.t = t
+        isect.hit = true
+        isect.pl = Vec.new(ray.org.x + ray.dir.x * t,
+                          ray.org.y + ray.dir.y * t,
+                          ray.org.z + ray.dir.z * t)
+        n = isect.pl.vsub(@center)
+        isect.n = n.vnormalize
+      else
+        0.0
+      end
+    end
+    nil
+  end
+end
+
+class Plane
+  def initialize(p, n)
+    @p = p
+    @n = n
+  end
+
+  def intersect(ray, isect)
+    d = -@p.vdot(@n)
+    v = ray.dir.vdot(@n)
+    v0 = v
+    if v < 0.0 then
+      v0 = -v
+    end
+    if v0 < 1.0e-17 then
+      return
+    end
+
+    t = -(ray.org.vdot(@n) + d) / v
+
+    if t > 0.0 and t < isect.t then
+      isect.hit = true
+      isect.t = t
+      isect.n = @n
+      isect.pl = Vec.new(ray.org.x + t * ray.dir.x,
+                        ray.org.y + t * ray.dir.y,
+                        ray.org.z + t * ray.dir.z)
+    end
+    nil
+  end
+end
+
+class Ray
+  def initialize(org, dir)
+    @org = org
+    @dir = dir
+  end
+
+  attr_accessor :org, :dir
+end
+
+class Isect
+  def initialize
+    @t = 10000000.0
+    @hit = false
+    @pl = Vec.new(0.0, 0.0, 0.0)
+    @n = Vec.new(0.0, 0.0, 0.0)
+  end
+
+  attr_accessor :t, :hit, :pl, :n
+end
+
+def clamp(f)
+  i = f * 255.5
+  if i > 255.0 then
+    i = 255.0
+  end
+  if i < 0.0 then
+    i = 0.0
+  end
+  i.to_i
+end
+
+def otherBasis(basis, n)
+  basis[2] = Vec.new(n.x, n.y, n.z)
+  basis[1] = Vec.new(0.0, 0.0, 0.0)
+
+  if n.x < 0.6 and n.x > -0.6 then
+    basis[1].x = 1.0
+  elsif n.y < 0.6 and n.y > -0.6 then
+    basis[1].y = 1.0
+  elsif n.z < 0.6 and n.z > -0.6 then
+    basis[1].z = 1.0
+  else
+    basis[1].x = 1.0
+  end
+
+  basis[0] = basis[1].vcross(basis[2])
+  basis[0] = basis[0].vnormalize
+
+  basis[1] = basis[2].vcross(basis[0])
+  basis[1] = basis[1].vnormalize
+end
+
+class Scene
+  def initialize
+    @spheres = Array.new
+    @spheres[0] = Sphere.new(Vec.new(-2.0, 0.0, -3.5), 0.5)
+    @spheres[1] = Sphere.new(Vec.new(-0.5, 0.0, -3.0), 0.5)
+    @spheres[2] = Sphere.new(Vec.new(1.0, 0.0, -2.2), 0.5)
+    @plane = Plane.new(Vec.new(0.0, -0.5, 0.0), Vec.new(0.0, 1.0, 0.0))
+  end
+
+  def ambient_occlusion(isect)
+    basis = Array.new
+    otherBasis(basis, isect.n)
+
+    ntheta    = NAO_SAMPLES
+    nphi      = NAO_SAMPLES
+    eps       = 0.0001
+    occlusion = 0.0
+
+    p0 = Vec.new(isect.pl.x + eps * isect.n.x,
+                isect.pl.y + eps * isect.n.y,
+                isect.pl.z + eps * isect.n.z)
+    nphi.times do |j|
+      ntheta.times do |i|
+        r = rand
+        phi = 2.0 * 3.14159265 * rand
+        x = Math.cos(phi) * Math.sqrt(1.0 - r)
+        y = Math.sin(phi) * Math.sqrt(1.0 - r)
+        z = Math.sqrt(r)
+
+        rx = x * basis[0].x + y * basis[1].x + z * basis[2].x
+        ry = x * basis[0].y + y * basis[1].y + z * basis[2].y
+        rz = x * basis[0].z + y * basis[1].z + z * basis[2].z
+
+        raydir = Vec.new(rx, ry, rz)
+        ray = Ray.new(p0, raydir)
+
+        occisect = Isect.new
+        @spheres[0].intersect(ray, occisect)
+        @spheres[1].intersect(ray, occisect)
+        @spheres[2].intersect(ray, occisect)
+        @plane.intersect(ray, occisect)
+        if occisect.hit then
+          occlusion = occlusion + 1.0
+        else
+          0.0
+        end
+      end
+    end
+
+    occlusion = (ntheta.to_f * nphi.to_f - occlusion) / (ntheta.to_f * nphi.to_f)
+
+    Vec.new(occlusion, occlusion, occlusion)
+  end
+
+  def render(w, h, nsubsamples)
+    cnt = 0
+    nsf = nsubsamples.to_f
+    h.times do |y|
+      w.times do |x|
+        rad = Vec.new(0.0, 0.0, 0.0)
+
+        # Subsampling
+        nsubsamples.times do |v|
+          nsubsamples.times do |u|
+
+            cnt = cnt + 1
+            wf = w.to_f
+            hf = h.to_f
+            xf = x.to_f
+            yf = y.to_f
+            uf = u.to_f
+            vf = v.to_f
+
+            px = (xf + (uf / nsf) - (wf / 2.0)) / (wf / 2.0)
+            py = -(yf + (vf / nsf) - (hf / 2.0)) / (hf / 2.0)
+
+            eye = Vec.new(px, py, -1.0).vnormalize
+
+            ray = Ray.new(Vec.new(0.0, 0.0, 0.0), eye)
+
+            isect = Isect.new
+            @spheres[0].intersect(ray, isect)
+            @spheres[1].intersect(ray, isect)
+            @spheres[2].intersect(ray, isect)
+            @plane.intersect(ray, isect)
+            if isect.hit then
+              col = ambient_occlusion(isect)
+              rad.x = rad.x + col.x
+              rad.y = rad.y + col.y
+              rad.z = rad.z + col.z
+            end
+          end
+        end
+
+        r = rad.x / (nsf * nsf)
+        g = rad.y / (nsf * nsf)
+        b = rad.z / (nsf * nsf)
+        printf("%c", clamp(r))
+        printf("%c", clamp(g))
+        printf("%c", clamp(b))
+      end
+      nil
+    end
+
+    nil
+  end
+end
+
+alias printf_orig printf
+def printf *args
+end
+
+# File.open("ao.ppm", "w") do |fp|
+  printf("P6\n")
+  printf("%d %d\n", IMAGE_WIDTH, IMAGE_HEIGHT)
+  printf("255\n", IMAGE_WIDTH, IMAGE_HEIGHT)
+  Scene.new.render(IMAGE_WIDTH, IMAGE_HEIGHT, NSUBSAMPLES)
+# end
+
+undef printf
+alias printf printf_orig
diff --git a/benchmark/app_erb.yml b/benchmark/app_erb.yml
new file mode 100644
index 0000000000..8583858d20
--- /dev/null
+++ b/benchmark/app_erb.yml
@@ -0,0 +1,24 @@
+#
+# Create many HTML strings with ERB.
+#
+prelude: |
+  require 'erb'
+
+  data = <<erb
+  <html>
+    <head> <%= title %> </head>
+    <body>
+      <h1> <%= title %> </h1>
+      <p>
+        <%= content %>
+      </p>
+    </body>
+  </html>
+  erb
+
+  max = 15_000
+  title = "hello world!"
+  content = "hello world!\n" * 10
+benchmark:
+  app_erb: ERB.new(data).result(binding)
+loop_count: 15000
diff --git a/benchmark/app_factorial.rb b/benchmark/app_factorial.rb
new file mode 100644
index 0000000000..45f471dfdb
--- /dev/null
+++ b/benchmark/app_factorial.rb
@@ -0,0 +1,11 @@
+def fact(n)
+  if(n > 1)
+    n * fact(n-1)
+  else
+    1
+  end
+end
+
+100.times {
+  fact(5000)
+}
diff --git a/benchmark/app_fib.rb b/benchmark/app_fib.rb
new file mode 100644
index 0000000000..34a7b2e725
--- /dev/null
+++ b/benchmark/app_fib.rb
@@ -0,0 +1,10 @@
+def fib n
+  if n < 3
+    1
+  else
+    fib(n-1) + fib(n-2)
+  end
+end
+
+fib(34)
+
diff --git a/benchmark/app_lc_fizzbuzz.rb b/benchmark/app_lc_fizzbuzz.rb
new file mode 100644
index 0000000000..f09574bbeb
--- /dev/null
+++ b/benchmark/app_lc_fizzbuzz.rb
@@ -0,0 +1,52 @@
+#
+# FizzBuzz program using only lambda calculus
+#
+# This program is quoted from
+#   "Understanding Computation" by Tom Stuart
+#   http://computationbook.com/
+#
+# You can understand why this program works fine by reading this book.
+#
+
+solution = -> k { -> f { -> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> l { -> x { -> g { -> b { b }[-> p { p[-> x { -> y { x } }] }[l]][x][-> y { g[f[-> l { -> p { p[-> x { -> y { y } }] }[-> p { p[-> x { -> y { y } }] }[l]] }[l]][x][g]][-> l { -> p { p[-> x { -> y { x } }] }[-> p { p[-> x { -> y { y } }] }[l]] }[l]][y] }] } } } }][k][-> x { -> y { -> f { f[x][y] } } }[-> x { -> y { x } }][-> x { -> y { x } }]][-> l { -> x { -> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[l][f[x]] } }] } }[-> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> m { -> n { -> b { b }[-> m { -> n { -> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]] } }[m][n]][-> x { -> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[f[-> n { -> p { -> x { p[n[p][x]] } } }[m]][n]][m][x] }][-> x { -> y { -> f { f[x][y] } } }[-> x { -> y { x } }][-> x { -> y { x } }]] } } }][-> p { -> x { p[x] } }][-> p { -> x { p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[x]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]] } }]][-> n { -> b { b }[-> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> m { -> n { -> b { b }[-> m { -> n { -> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]] } }[n][m]][-> x { f[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]][n][x] }][m] } } }][n][-> p { -> x { p[p[p[p[p[p[p[p[p[p[p[p[p[p[p[x]]]]]]]]]]]]]]] } }]]][-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> x { -> y { -> f { f[x][y] } } }[-> x { -> y { x } }][-> x { -> y { x } }]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]][-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]][-> b { b }[-> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> m { -> n { -> b { b }[-> m { -> n { -> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]] } }[n][m]][-> x { f[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]][n][x] }][m] } } }][n][-> p { -> x { p[p[p[x]]] } }]]][-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> x { -> y { -> f { f[x][y] } } }[-> x { -> y { x } }][-> x { -> y { x } }]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]][-> b { b }[-> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> m { -> n { -> b { b }[-> m { -> n { -> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]] } }[n][m]][-> x { f[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]][n][x] }][m] } } }][n][-> p { -> x { p[p[p[p[p[x]]]]] } }]]][-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> x { -> y { -> f { f[x][y] } } }[-> x { -> y { x } }][-> x { -> y { x } }]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]]][-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> n { -> p { -> x { p[n[p][x]] } } }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]]]][-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]][-> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> n { -> l { -> x { -> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> l { -> x { -> g { -> b { b }[-> p { p[-> x { -> y { x } }] }[l]][x][-> y { g[f[-> l { -> p { p[-> x { -> y { y } }] }[-> p { p[-> x { -> y { y } }] }[l]] }[l]][x][g]][-> l { -> p { p[-> x { -> y { x } }] }[-> p { p[-> x { -> y { y } }] }[l]] }[l]][y] }] } } } }][l][-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }[-> x { -> y { -> f { f[x][y] } } }[-> x { -> y { x } }][-> x { -> y { x } }]][x]][-> l { -> x { -> x { -> y { -> f { f[x][y] } } }[-> x { -> y { y } }][-> x { -> y { -> f { f[x][y] } } }[x][l]] } }] } }[-> b { b }[-> m { -> n { -> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]] } }[n][-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }[-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]]][-> x { -> y { -> f { f[x][y] } } }[-> x { -> y { x } }][-> x { -> y { x } }]][-> x { f[-> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> m { -> n { -> b { b }[-> m { -> n { -> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]] } }[n][m]][-> x { -> n { -> p { -> x { p[n[p][x]] } } }[f[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]][n]][x] }][-> p { -> x { x } }] } } }][n][-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]][x] }]][-> f { -> x { f[-> y { x[x][y] }] }[-> x { f[-> y { x[x][y] }] }] }[-> f { -> m { -> n { -> b { b }[-> m { -> n { -> n { n[-> x { -> x { -> y { y } } }][-> x { -> y { x } }] }[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]] } }[n][m]][-> x { f[-> m { -> n { n[-> n { -> p { p[-> x { -> y { x } }] }[n[-> p { -> x { -> y { -> f { f[x][y] } } }[-> p { p[-> x { -> y { y } }] }[p]][-> n { -> p { -> x { p[n[p][x]] } } }[-> p { p[-> x { -> y { y } }] }[p]]] }][-> x { -> y { -> f { f[x][y] } } }[-> p { -> x { x } }][-> p { -> x { x } }]]] }][m] } }[m][n]][n][x] }][m] } } }][n][-> m { -> n { n[-> m { -> n { n[-> n { -> p { -> x { p[n[p][x]] } } }][m] } }[m]][-> p { -> x { x } }] } }[-> p { -> x { p[p[x]] } }][-> p { -> x { p[p[p[p[p[x]]]]] } }]]] } }][n]]]] }]
+
+FIRST     = -> l { LEFT[RIGHT[l]] }
+IF        = -> b { b }
+LEFT      = -> p { p[-> x { -> y { x } } ] }
+RIGHT     = -> p { p[-> x { -> y { y } } ] }
+IS_EMPTY  = LEFT
+REST      = -> l { RIGHT[RIGHT[l]] }
+
+def to_integer(proc)
+  proc[-> n { n + 1 }][0]
+end
+
+def to_boolean(proc)
+  IF[proc][true][false]
+end
+
+def to_array(proc)
+  array = []
+
+  until to_boolean(IS_EMPTY[proc])
+    array.push(FIRST[proc])
+    proc = REST[proc]
+  end
+
+  array
+end
+
+def to_char(c)
+  '0123456789BFiuz'.slice(to_integer(c))
+end
+
+def to_string(s)
+  to_array(s).map { |c| to_char(c) }.join
+end
+
+answer = to_array(solution).map do |p|
+  to_string(p)
+end
+
+answer_ary = answer.to_a
+# puts answer_ary
diff --git a/benchmark/app_mandelbrot.rb b/benchmark/app_mandelbrot.rb
new file mode 100644
index 0000000000..801b75e8e2
--- /dev/null
+++ b/benchmark/app_mandelbrot.rb
@@ -0,0 +1,23 @@
+require 'complex'
+
+def mandelbrot? z
+  i = 0
+  while i<100
+    i += 1
+    z = z * z
+    return false if z.abs > 2
+  end
+  true
+end
+
+ary = []
+
+(0..1000).each{|dx|
+  (0..1000).each{|dy|
+    x = dx / 50.0
+    y = dy / 50.0
+    c = Complex(x, y)
+    ary << c if mandelbrot?(c)
+  }
+}
+
diff --git a/benchmark/app_pentomino.rb b/benchmark/app_pentomino.rb
new file mode 100644
index 0000000000..47be7b203f
--- /dev/null
+++ b/benchmark/app_pentomino.rb
@@ -0,0 +1,130 @@
+#!/usr/local/bin/ruby
+# This program is contributed by Shin Nishiyama
+
+
+# modified by K.Sasada
+
+NP = 5
+ROW = 8 + NP
+COL = 8
+
+$p = []
+$b = []
+$no = 0
+
+def piece(n, a, nb)
+  nb.each{|x|
+    a[n] = x
+    if n == NP-1
+      $p << [a.sort]
+    else
+      nbc=nb.dup
+      [-ROW, -1, 1, ROW].each{|d|
+        if x+d > 0 and not a.include?(x+d) and not nbc.include?(x+d)
+          nbc << x+d
+        end
+      }
+      nbc.delete x
+      piece(n+1,a[0..n],nbc)
+    end
+  }
+end
+
+def kikaku(a)
+  a.collect {|x| x - a[0]}
+end
+def ud(a)
+  kikaku(a.collect {|x| ((x+NP)%ROW)-ROW*((x+NP)/ROW) }.sort)
+end
+def rl(a)
+  kikaku(a.collect {|x| ROW*((x+NP)/ROW)+ROW-((x+NP)%ROW)}.sort)
+end
+def xy(a)
+  kikaku(a.collect {|x| ROW*((x+NP)%ROW) + (x+NP)/ROW }.sort)
+end
+
+def mkpieces
+  piece(0,[],[0])
+  $p.each do |a|
+    a0 = a[0]
+    a[1] = ud(a0)
+    a[2] = rl(a0)
+    a[3] = ud(rl(a0))
+    a[4] = xy(a0)
+    a[5] = ud(xy(a0))
+    a[6] = rl(xy(a0))
+    a[7] = ud(rl(xy(a0)))
+    a.sort!
+    a.uniq!
+  end
+  $p.uniq!.sort! {|x,y| x[0] <=> y[0] }
+end
+
+def mkboard
+  (0...ROW*COL).each{|i|
+    if i % ROW >= ROW-NP
+      $b[i] = -2
+    else
+      $b[i] = -1
+    end
+    $b[3*ROW+3]=$b[3*ROW+4]=$b[4*ROW+3]=$b[4*ROW+4]=-2
+  }
+end
+
+def pboard
+  return # skip print
+  print "No. #$no\n"
+  (0...COL).each{|i|
+    print "|"
+    (0...ROW-NP).each{|j|
+      x = $b[i*ROW+j]
+      if x < 0
+        print "..|"
+      else
+        printf "%2d|",x+1
+      end
+    }
+    print "\n"
+  }
+  print "\n"
+end
+
+$pnum=[]
+def setpiece(a,pos)
+  if a.length == $p.length then
+    $no += 1
+    pboard
+    return
+  end
+  while $b[pos] != -1
+    pos += 1
+  end
+  ($pnum - a).each do |i|
+    $p[i].each do |x|
+      f = 0
+      x.each{|s|
+        if $b[pos+s] != -1
+          f=1
+          break
+        end
+      }
+      if f == 0 then
+        x.each{|s|
+          $b[pos+s] = i
+        }
+        a << i
+        setpiece(a.dup, pos)
+        a.pop
+        x.each{|s|
+          $b[pos+s] = -1
+        }
+      end
+    end
+  end
+end
+
+mkpieces
+mkboard
+$p[4] = [$p[4][0]]
+$pnum = (0...$p.length).to_a
+setpiece([],0)
diff --git a/benchmark/app_raise.rb b/benchmark/app_raise.rb
new file mode 100644
index 0000000000..5db8f95d50
--- /dev/null
+++ b/benchmark/app_raise.rb
@@ -0,0 +1,8 @@
+i = 0
+while i<300000
+  i += 1
+  begin
+    raise
+  rescue
+  end
+end
diff --git a/benchmark/app_strconcat.rb b/benchmark/app_strconcat.rb
new file mode 100644
index 0000000000..7eed7c1aed
--- /dev/null
+++ b/benchmark/app_strconcat.rb
@@ -0,0 +1,5 @@
+i = 0
+while i<2_000_000
+  "#{1+1} #{1+1} #{1+1}"
+  i += 1
+end
diff --git a/benchmark/app_tak.rb b/benchmark/app_tak.rb
new file mode 100644
index 0000000000..efe5380f4e
--- /dev/null
+++ b/benchmark/app_tak.rb
@@ -0,0 +1,13 @@
+
+def tak x, y, z
+  unless y < x
+    z
+  else
+    tak( tak(x-1, y, z),
+         tak(y-1, z, x),
+         tak(z-1, x, y))
+  end
+end
+
+tak(18, 9, 0)
+
diff --git a/benchmark/app_tarai.rb b/benchmark/app_tarai.rb
new file mode 100644
index 0000000000..4c146f5ccf
--- /dev/null
+++ b/benchmark/app_tarai.rb
@@ -0,0 +1,10 @@
+def tarai( x, y, z )
+  if x <= y
+  then y
+  else tarai(tarai(x-1, y, z),
+             tarai(y-1, z, x),
+             tarai(z-1, x, y))
+  end
+end
+
+tarai(12, 6, 0)
diff --git a/benchmark/app_uri.rb b/benchmark/app_uri.rb
new file mode 100644
index 0000000000..586edfd5dc
--- /dev/null
+++ b/benchmark/app_uri.rb
@@ -0,0 +1,8 @@
+require 'uri'
+
+100_000.times{
+  uri = URI.parse('http://www.ruby-lang.org')
+  uri.scheme
+  uri.host
+  uri.port
+}
diff --git a/benchmark/array_sample_100k_10.rb b/benchmark/array_sample_100k_10.rb
new file mode 100644
index 0000000000..5f41ecc32b
--- /dev/null
+++ b/benchmark/array_sample_100k_10.rb
@@ -0,0 +1,2 @@
+arr = [*0...100000]
+10_000.times {arr.sample 10}
diff --git a/benchmark/array_sample_100k_11.rb b/benchmark/array_sample_100k_11.rb
new file mode 100644
index 0000000000..18b1715319
--- /dev/null
+++ b/benchmark/array_sample_100k_11.rb
@@ -0,0 +1,2 @@
+arr = [*0...100000]
+10_000.times {arr.sample 11}
diff --git a/benchmark/array_sample_100k__100.rb b/benchmark/array_sample_100k__100.rb
new file mode 100644
index 0000000000..22863afe89
--- /dev/null
+++ b/benchmark/array_sample_100k__100.rb
@@ -0,0 +1,2 @@
+arr = [*0...100000]
+10_000.times {arr.sample 100}
diff --git a/benchmark/array_sample_100k__1k.rb b/benchmark/array_sample_100k__1k.rb
new file mode 100644
index 0000000000..4cd79e6c67
--- /dev/null
+++ b/benchmark/array_sample_100k__1k.rb
@@ -0,0 +1,2 @@
+arr = [*0...100000]
+10_000.times {arr.sample 1000}
diff --git a/benchmark/array_sample_100k__6k.rb b/benchmark/array_sample_100k__6k.rb
new file mode 100644
index 0000000000..b3d264249e
--- /dev/null
+++ b/benchmark/array_sample_100k__6k.rb
@@ -0,0 +1,2 @@
+arr = [*0...100000]
+10_000.times {arr.sample 6000}
diff --git a/benchmark/array_sample_100k___10k.rb b/benchmark/array_sample_100k___10k.rb
new file mode 100644
index 0000000000..5dd55ec058
--- /dev/null
+++ b/benchmark/array_sample_100k___10k.rb
@@ -0,0 +1,2 @@
+arr = [*0...100000]
+10_000.times {arr.sample 10_000}
diff --git a/benchmark/array_sample_100k___50k.rb b/benchmark/array_sample_100k___50k.rb
new file mode 100644
index 0000000000..1506732c3c
--- /dev/null
+++ b/benchmark/array_sample_100k___50k.rb
@@ -0,0 +1,2 @@
+arr = [*0...100000]
+10_000.times {arr.sample 50_000}
diff --git a/benchmark/array_shift.rb b/benchmark/array_shift.rb
new file mode 100644
index 0000000000..798bb9e3f4
--- /dev/null
+++ b/benchmark/array_shift.rb
@@ -0,0 +1,14 @@
+require 'benchmark'
+
+Benchmark.bm do |x|
+  [10_000,1_000_000,100_000_000].each do |n|
+    ary = Array.new(n,0)
+    GC.start
+    x.report("#{n}:shift"){ ary.shift }
+    (0..4).each do |i|
+      ary = Array.new(n,0)
+      GC.start
+      x.report("#{n}:shift(#{i})"){ ary.shift(i) }
+    end
+  end
+end
diff --git a/benchmark/array_small_and.rb b/benchmark/array_small_and.rb
new file mode 100644
index 0000000000..e53a6edae6
--- /dev/null
+++ b/benchmark/array_small_and.rb
@@ -0,0 +1,17 @@
+MIN_SIZE = ENV.fetch('SMALL_ARRAY_MIN', 0).to_i
+MAX_SIZE = ENV.fetch('SMALL_ARRAY_MAX', 16).to_i
+ITERATIONS = ENV.fetch('SMALL_ARRAY_ITERATIONS', 100).to_i
+
+ARRAYS = (MIN_SIZE..MAX_SIZE).map do |size1|
+  (MIN_SIZE..MAX_SIZE).map do |size2|
+    [Array.new(size1) { rand(MAX_SIZE) }, Array.new(size2) { rand(MAX_SIZE) }]
+  end
+end
+
+ITERATIONS.times do
+  ARRAYS.each do |group|
+    group.each do |arr1, arr2|
+      arr1 & arr2
+    end
+  end
+end
diff --git a/benchmark/array_small_diff.rb b/benchmark/array_small_diff.rb
new file mode 100644
index 0000000000..9661ee48db
--- /dev/null
+++ b/benchmark/array_small_diff.rb
@@ -0,0 +1,17 @@
+MIN_SIZE = ENV.fetch('SMALL_ARRAY_MIN', 0).to_i
+MAX_SIZE = ENV.fetch('SMALL_ARRAY_MAX', 16).to_i
+ITERATIONS = ENV.fetch('SMALL_ARRAY_ITERATIONS', 100).to_i
+
+ARRAYS = (MIN_SIZE..MAX_SIZE).map do |size1|
+  (MIN_SIZE..MAX_SIZE).map do |size2|
+    [Array.new(size1) { rand(MAX_SIZE) }, Array.new(size2) { rand(MAX_SIZE) }]
+  end
+end
+
+ITERATIONS.times do
+  ARRAYS.each do |group|
+    group.each do |arr1, arr2|
+      arr1 - arr2
+    end
+  end
+end
diff --git a/benchmark/array_small_or.rb b/benchmark/array_small_or.rb
new file mode 100644
index 0000000000..c58b5fd1ff
--- /dev/null
+++ b/benchmark/array_small_or.rb
@@ -0,0 +1,17 @@
+MIN_SIZE = ENV.fetch('SMALL_ARRAY_MIN', 0).to_i
+MAX_SIZE = ENV.fetch('SMALL_ARRAY_MAX', 16).to_i
+ITERATIONS = ENV.fetch('SMALL_ARRAY_ITERATIONS', 100).to_i
+
+ARRAYS = (MIN_SIZE..MAX_SIZE).map do |size1|
+  (MIN_SIZE..MAX_SIZE).map do |size2|
+    [Array.new(size1) { rand(MAX_SIZE) }, Array.new(size2) { rand(MAX_SIZE) }]
+  end
+end
+
+ITERATIONS.times do
+  ARRAYS.each do |group|
+    group.each do |arr1, arr2|
+      arr1 | arr2
+    end
+  end
+end
diff --git a/benchmark/array_sort_block.rb b/benchmark/array_sort_block.rb
new file mode 100644
index 0000000000..3579786056
--- /dev/null
+++ b/benchmark/array_sort_block.rb
@@ -0,0 +1,2 @@
+ary = Array.new(1000) { rand(1000) }
+10000.times { ary.sort { |a, b| a <=> b } }
diff --git a/benchmark/array_sort_float.rb b/benchmark/array_sort_float.rb
new file mode 100644
index 0000000000..9a6e2f8bd2
--- /dev/null
+++ b/benchmark/array_sort_float.rb
@@ -0,0 +1,2 @@
+arr = Array.new(1000) { rand }
+10000.times { arr.sort }
diff --git a/benchmark/array_values_at_int.rb b/benchmark/array_values_at_int.rb
new file mode 100644
index 0000000000..6cb394cb9f
--- /dev/null
+++ b/benchmark/array_values_at_int.rb
@@ -0,0 +1,2 @@
+ary = Array.new(10000) {|i| i}
+100000.times { ary.values_at(500) }
diff --git a/benchmark/array_values_at_range.rb b/benchmark/array_values_at_range.rb
new file mode 100644
index 0000000000..5b53806d1c
--- /dev/null
+++ b/benchmark/array_values_at_range.rb
@@ -0,0 +1,2 @@
+ary = Array.new(10000) {|i| i}
+100000.times { ary.values_at(1..2000) }
diff --git a/benchmark/bighash.rb b/benchmark/bighash.rb
new file mode 100644
index 0000000000..e2ad5a5c94
--- /dev/null
+++ b/benchmark/bighash.rb
@@ -0,0 +1 @@
+h = {}; 5000000.times {|n| h[n] = n }
diff --git a/benchmark/dir_empty_p.rb b/benchmark/dir_empty_p.rb
new file mode 100644
index 0000000000..8329c757cf
--- /dev/null
+++ b/benchmark/dir_empty_p.rb
@@ -0,0 +1,5 @@
+require 'tmpdir'
+max = 100_000
+Dir.mktmpdir('bm_dir_empty_p') do |dir|
+  max.times { Dir.empty?(dir) }
+end
diff --git a/benchmark/enum_lazy_grep_v_100.rb b/benchmark/enum_lazy_grep_v_100.rb
new file mode 100644
index 0000000000..8832392e65
--- /dev/null
+++ b/benchmark/enum_lazy_grep_v_100.rb
@@ -0,0 +1,4 @@
+grep_data = (1..10).to_a * 1000
+N = 100
+enum = grep_data.lazy.grep_v(->(i){i == 0}).grep_v(->(i){i == 0})
+N.times {enum.each {}}
diff --git a/benchmark/enum_lazy_grep_v_20.rb b/benchmark/enum_lazy_grep_v_20.rb
new file mode 100644
index 0000000000..329509fa8f
--- /dev/null
+++ b/benchmark/enum_lazy_grep_v_20.rb
@@ -0,0 +1,4 @@
+grep_data = (1..10).to_a * 1000
+N = 100
+enum = grep_data.lazy.grep_v(->(i){i > 2}).grep_v(->(i){i > 2})
+N.times {enum.each {}}
diff --git a/benchmark/enum_lazy_grep_v_50.rb b/benchmark/enum_lazy_grep_v_50.rb
new file mode 100644
index 0000000000..02ea4d4e71
--- /dev/null
+++ b/benchmark/enum_lazy_grep_v_50.rb
@@ -0,0 +1,4 @@
+grep_data = (1..10).to_a * 1000
+N = 100
+enum = grep_data.lazy.grep_v(->(i){i > 5}).grep_v(->(i){i > 5})
+N.times {enum.each {}}
diff --git a/benchmark/enum_lazy_uniq_100.rb b/benchmark/enum_lazy_uniq_100.rb
new file mode 100644
index 0000000000..2e6434d9c4
--- /dev/null
+++ b/benchmark/enum_lazy_uniq_100.rb
@@ -0,0 +1,4 @@
+uniq_data = (1..10_000).to_a
+N = 100
+enum = uniq_data.lazy.uniq {|i| i % 10000}.uniq {|i| i % 10000}
+N.times {enum.each {}}
diff --git a/benchmark/enum_lazy_uniq_20.rb b/benchmark/enum_lazy_uniq_20.rb
new file mode 100644
index 0000000000..75e6398fee
--- /dev/null
+++ b/benchmark/enum_lazy_uniq_20.rb
@@ -0,0 +1,4 @@
+uniq_data = (1..10_000).to_a
+N = 100
+enum = uniq_data.lazy.uniq {|i| i % 2000}.uniq {|i| i % 2000}
+N.times {enum.each {}}
diff --git a/benchmark/enum_lazy_uniq_50.rb b/benchmark/enum_lazy_uniq_50.rb
new file mode 100644
index 0000000000..59a39b78ff
--- /dev/null
+++ b/benchmark/enum_lazy_uniq_50.rb
@@ -0,0 +1,4 @@
+uniq_data = (1..10_000).to_a
+N = 100
+enum = uniq_data.lazy.uniq {|i| i % 5000}.uniq {|i| i % 5000}
+N.times {enum.each {}}
diff --git a/benchmark/erb_render.yml b/benchmark/erb_render.yml
new file mode 100644
index 0000000000..15f6c3880b
--- /dev/null
+++ b/benchmark/erb_render.yml
@@ -0,0 +1,24 @@
+prelude: |
+  require 'erb'
+
+  data = <<erb
+  <html>
+    <head> <%= title %> </head>
+    <body>
+      <h1> <%= title %> </h1>
+      <p>
+        <%= content %>
+      </p>
+    </body>
+  </html>
+  erb
+
+  title = "hello world!"
+  content = "hello world!\n" * 10
+
+  src = "def self.render(title, content); #{ERB.new(data).src}; end"
+  mod = Module.new
+  mod.instance_eval(src, "(ERB)")
+benchmark:
+  erb_render: mod.render(title, content)
+loop_count: 1500000
diff --git a/benchmark/fiber_chain.rb b/benchmark/fiber_chain.rb
new file mode 100755
index 0000000000..7e0a7f9d45
--- /dev/null
+++ b/benchmark/fiber_chain.rb
@@ -0,0 +1,40 @@
+# Check performance of fiber creation and transfer.
+
+def make_link(previous)
+  Fiber.new do
+    while message = previous.resume
+      Fiber.yield(message)
+    end
+  end
+end
+
+def make_chain(length, &block)
+  chain = Fiber.new(&block)
+
+  (length - 1).times do
+    chain = make_link(chain)
+  end
+
+  return chain
+end
+
+def run_benchmark(length, repeats, message = :hello)
+  chain = nil
+
+  chain = make_chain(length) do
+    while true
+      Fiber.yield(message)
+    end
+  end
+
+  repeats.times do
+    abort "invalid result" unless chain.resume == message
+  end
+end
+
+n = (ARGV[0] || 1000).to_i
+m = (ARGV[1] || 1000).to_i
+
+5.times do
+  run_benchmark(n, m)
+end
diff --git a/benchmark/file_chmod.rb b/benchmark/file_chmod.rb
new file mode 100644
index 0000000000..1cd4760c9d
--- /dev/null
+++ b/benchmark/file_chmod.rb
@@ -0,0 +1,9 @@
+# chmod file
+require 'tempfile'
+max = 200_000
+tmp = Tempfile.new('chmod')
+path = tmp.path
+max.times do
+  File.chmod(0777, path)
+end
+tmp.close!
diff --git a/benchmark/file_rename.rb b/benchmark/file_rename.rb
new file mode 100644
index 0000000000..bbb44aebac
--- /dev/null
+++ b/benchmark/file_rename.rb
@@ -0,0 +1,11 @@
+# rename file
+require 'tempfile'
+
+max = 100_000
+tmp = [ Tempfile.new('rename-a'), Tempfile.new('rename-b') ]
+a, b = tmp.map { |x| x.path }
+tmp.each { |t| t.close } # Windows can't rename files without closing them
+max.times do
+  File.rename(a, b)
+  File.rename(b, a)
+end
diff --git a/benchmark/gc/aobench.rb b/benchmark/gc/aobench.rb
new file mode 100644
index 0000000000..275f58b924
--- /dev/null
+++ b/benchmark/gc/aobench.rb
@@ -0,0 +1 @@
+require_relative '../app_aobench'
diff --git a/benchmark/gc/binary_trees.rb b/benchmark/gc/binary_trees.rb
new file mode 100644
index 0000000000..83347cdd20
--- /dev/null
+++ b/benchmark/gc/binary_trees.rb
@@ -0,0 +1 @@
+require_relative '../so_binary_trees'
diff --git a/benchmark/gc/gcbench.rb b/benchmark/gc/gcbench.rb
new file mode 100644
index 0000000000..23d0b91c6c
--- /dev/null
+++ b/benchmark/gc/gcbench.rb
@@ -0,0 +1,57 @@
+require 'benchmark'
+require 'pp'
+require 'optparse'
+
+$list = true
+$gcprof = false
+
+opt = OptionParser.new
+opt.on('-q'){$list = false}
+opt.on('-d'){$gcprof = false}
+opt.on('-p'){$gcprof = true}
+opt.parse!(ARGV)
+
+script = File.join(File.dirname(__FILE__), ARGV.shift)
+script += '.rb' unless FileTest.exist?(script)
+raise "#{script} not found" unless FileTest.exist?(script)
+
+puts "Script: #{script}"
+
+if $gcprof
+  GC::Profiler.enable
+end
+
+tms = Benchmark.measure{|x|
+  load script
+}
+
+gc_time = 0
+
+if $gcprof
+  gc_time = GC::Profiler.total_time
+  GC::Profiler.report if $list and RUBY_VERSION >= '2.0.0' # before 1.9.3, report() may run infinite loop
+  GC::Profiler.disable
+end
+
+pp GC.stat
+
+puts "#{RUBY_DESCRIPTION} #{GC::OPTS.inspect}" if defined?(GC::OPTS)
+
+desc = "#{RUBY_VERSION}#{RUBY_PATCHLEVEL >= 0 ? "p#{RUBY_PATCHLEVEL}" : "dev"}"
+name = File.basename(script, '.rb')
+
+puts
+puts script
+puts Benchmark::CAPTION
+puts tms
+puts "GC total time (sec): #{gc_time}"
+
+# show High-Water Mark on Linux
+if File.exist?('/proc/self/status') && /VmHWM:\s*(\d+.+)/ =~ File.read('/proc/self/status')
+  puts
+  puts "VmHWM: #{$1.chomp}"
+end
+
+puts
+puts "Summary of #{name} on #{desc}\t#{tms.real}\t#{gc_time}\t#{GC.count}"
+puts "         (real time in sec, GC time in sec, GC count)"
diff --git a/benchmark/gc/hash1.rb b/benchmark/gc/hash1.rb
new file mode 100644
index 0000000000..cb030d458d
--- /dev/null
+++ b/benchmark/gc/hash1.rb
@@ -0,0 +1,11 @@
+value = 0.01
+h = {}
+n = 50_000
+
+1.upto(n){|i|
+  h["%020d" % i] = "v-#{i}"
+}
+
+(n * 1_000).times{
+  ''
+}
diff --git a/benchmark/gc/hash2.rb b/benchmark/gc/hash2.rb
new file mode 100644
index 0000000000..e8c943fb21
--- /dev/null
+++ b/benchmark/gc/hash2.rb
@@ -0,0 +1,7 @@
+value = 0.01
+h = {}
+n = 4*(10**6)
+
+1.upto(n){|i|
+  h["%020d" % i] = value * i
+}
diff --git a/benchmark/gc/null.rb b/benchmark/gc/null.rb
new file mode 100644
index 0000000000..c05a79f561
--- /dev/null
+++ b/benchmark/gc/null.rb
@@ -0,0 +1 @@
+# null
diff --git a/benchmark/gc/pentomino.rb b/benchmark/gc/pentomino.rb
new file mode 100644
index 0000000000..8ebdff7d1d
--- /dev/null
+++ b/benchmark/gc/pentomino.rb
@@ -0,0 +1 @@
+require_relative '../app_pentomino'
diff --git a/benchmark/gc/rdoc.rb b/benchmark/gc/rdoc.rb
new file mode 100644
index 0000000000..14c89f5611
--- /dev/null
+++ b/benchmark/gc/rdoc.rb
@@ -0,0 +1,13 @@
+require 'rdoc/rdoc'
+require 'tmpdir'
+
+srcdir = File.expand_path('../..', __dir__)
+
+Dir.mktmpdir('rdocbench-'){|d|
+  dir = File.join(d, 'rdocbench')
+  args = %W(--root #{srcdir} --page-dir #{srcdir}/doc --encoding=UTF-8 --no-force-update --all --ri --debug --quiet #{srcdir})
+  args << '--op' << dir
+
+  r = RDoc::RDoc.new
+  r.document args
+}
diff --git a/benchmark/gc/redblack.rb b/benchmark/gc/redblack.rb
new file mode 100644
index 0000000000..c66290140a
--- /dev/null
+++ b/benchmark/gc/redblack.rb
@@ -0,0 +1,366 @@
+# This benchmark is imported from https://github.com/jruby/rubybench/blob/master/time/bench_red_black.rb
+# License is License is Apache-2
+
+require 'benchmark'
+
+# Algorithm based on "Introduction to Algorithms" by Cormen and others
+class RedBlackTree
+  class Node
+    attr_accessor :color
+    attr_accessor :key
+    attr_accessor :left
+    attr_accessor :right
+    attr_accessor :parent
+
+    RED = :red
+    BLACK = :black
+    COLORS = [RED, BLACK].freeze
+
+    def initialize(key, color = RED)
+      raise ArgumentError, "Bad value for color parameter" unless COLORS.include?(color)
+      @color = color
+      @key = key
+      @left = @right = @parent = NilNode.instance
+    end
+
+    def black?
+      return color == BLACK
+    end
+
+    def red?
+      return color == RED
+    end
+  end
+
+  class NilNode < Node
+    class << self
+      private :new
+
+      # it's not thread safe
+      def instance
+        @instance ||= begin
+          def instance
+            return @instance
+          end
+
+          new
+        end
+      end
+    end
+
+    def initialize
+      self.color = BLACK
+      self.key = 0
+      self.left = nil
+      self.right = nil
+      self.parent = nil
+    end
+
+    def nil?
+      return true
+    end
+  end
+
+  include Enumerable
+
+  attr_accessor :root
+  attr_accessor :size
+
+  def initialize
+    self.root = NilNode.instance
+    self.size = 0
+  end
+
+  def add(key)
+    insert(Node.new(key))
+  end
+
+  def insert(x)
+    insert_helper(x)
+
+    x.color = Node::RED
+    while x != root && x.parent.color == Node::RED
+      if x.parent == x.parent.parent.left
+        y = x.parent.parent.right
+        if !y.nil? && y.color == Node::RED
+          x.parent.color = Node::BLACK
+          y.color = Node::BLACK
+          x.parent.parent.color = Node::RED
+          x = x.parent.parent
+        else
+          if x == x.parent.right
+            x = x.parent
+            left_rotate(x)
+          end
+          x.parent.color = Node::BLACK
+          x.parent.parent.color = Node::RED
+          right_rotate(x.parent.parent)
+        end
+      else
+        y = x.parent.parent.left
+        if !y.nil? && y.color == Node::RED
+          x.parent.color = Node::BLACK
+          y.color = Node::BLACK
+          x.parent.parent.color = Node::RED
+          x = x.parent.parent
+        else
+          if x == x.parent.left
+            x = x.parent
+            right_rotate(x)
+          end
+          x.parent.color = Node::BLACK
+          x.parent.parent.color = Node::RED
+          left_rotate(x.parent.parent)
+        end
+      end
+    end
+    root.color = Node::BLACK
+  end
+
+  alias << insert
+
+  def delete(z)
+    y = (z.left.nil? || z.right.nil?) ? z : successor(z)
+    x = y.left.nil? ? y.right : y.left
+    x.parent = y.parent
+
+    if y.parent.nil?
+      self.root = x
+    else
+      if y == y.parent.left
+        y.parent.left = x
+      else
+        y.parent.right = x
+      end
+    end
+
+    z.key = y.key if y != z
+
+    if y.color == Node::BLACK
+      delete_fixup(x)
+    end
+
+    self.size -= 1
+    return y
+  end
+
+  def minimum(x = root)
+    while !x.left.nil?
+      x = x.left
+    end
+    return x
+  end
+
+  def maximum(x = root)
+    while !x.right.nil?
+      x = x.right
+    end
+    return x
+  end
+
+  def successor(x)
+    if !x.right.nil?
+      return minimum(x.right)
+    end
+    y = x.parent
+    while !y.nil? && x == y.right
+      x = y
+      y = y.parent
+    end
+    return y
+  end
+
+  def predecessor(x)
+    if !x.left.nil?
+      return maximum(x.left)
+    end
+    y = x.parent
+    while !y.nil? && x == y.left
+      x = y
+      y = y.parent
+    end
+    return y
+  end
+
+  def inorder_walk(x = root)
+    x = self.minimum
+    while !x.nil?
+      yield x.key
+      x = successor(x)
+    end
+  end
+
+  alias each inorder_walk
+
+  def reverse_inorder_walk(x = root)
+    x = self.maximum
+    while !x.nil?
+      yield x.key
+      x = predecessor(x)
+    end
+  end
+
+  alias reverse_each reverse_inorder_walk
+
+  def search(key, x = root)
+    while !x.nil? && x.key != key
+      key < x.key ? x = x.left : x = x.right
+    end
+    return x
+  end
+
+  def empty?
+    return self.root.nil?
+  end
+
+  def black_height(x = root)
+    height = 0
+    while !x.nil?
+      x = x.left
+      height +=1 if x.nil? || x.black?
+    end
+    return height
+  end
+
+private
+
+  def left_rotate(x)
+    raise "x.right is nil!" if x.right.nil?
+    y = x.right
+    x.right = y.left
+    y.left.parent = x if !y.left.nil?
+    y.parent = x.parent
+    if x.parent.nil?
+      self.root = y
+    else
+      if x == x.parent.left
+        x.parent.left = y
+      else
+        x.parent.right = y
+      end
+    end
+    y.left = x
+    x.parent = y
+  end
+
+  def right_rotate(x)
+    raise "x.left is nil!" if x.left.nil?
+    y = x.left
+    x.left = y.right
+    y.right.parent = x if !y.right.nil?
+    y.parent = x.parent
+    if x.parent.nil?
+      self.root = y
+    else
+      if x == x.parent.left
+        x.parent.left = y
+      else
+        x.parent.right = y
+      end
+    end
+    y.right = x
+    x.parent = y
+  end
+
+  def insert_helper(z)
+    y = NilNode.instance
+    x = root
+    while !x.nil?
+      y = x
+      z.key < x.key ? x = x.left : x = x.right
+    end
+    z.parent = y
+    if y.nil?
+      self.root = z
+    else
+      z.key < y.key ? y.left = z : y.right = z
+    end
+    self.size += 1
+  end
+
+  def delete_fixup(x)
+    while x != root && x.color == Node::BLACK
+      if x == x.parent.left
+        w = x.parent.right
+        if w.color == Node::RED
+          w.color = Node::BLACK
+          x.parent.color = Node::RED
+          left_rotate(x.parent)
+          w = x.parent.right
+        end
+        if w.left.color == Node::BLACK && w.right.color == Node::BLACK
+          w.color = Node::RED
+          x = x.parent
+        else
+          if w.right.color == Node::BLACK
+            w.left.color = Node::BLACK
+            w.color = Node::RED
+            right_rotate(w)
+            w = x.parent.right
+          end
+          w.color = x.parent.color
+          x.parent.color = Node::BLACK
+          w.right.color = Node::BLACK
+          left_rotate(x.parent)
+          x = root
+        end
+      else
+        w = x.parent.left
+        if w.color == Node::RED
+          w.color = Node::BLACK
+          x.parent.color = Node::RED
+          right_rotate(x.parent)
+          w = x.parent.left
+        end
+        if w.right.color == Node::BLACK && w.left.color == Node::BLACK
+          w.color = Node::RED
+          x = x.parent
+        else
+          if w.left.color == Node::BLACK
+            w.right.color = Node::BLACK
+            w.color = Node::RED
+            left_rotate(w)
+            w = x.parent.left
+          end
+          w.color = x.parent.color
+          x.parent.color = Node::BLACK
+          w.left.color = Node::BLACK
+          right_rotate(x.parent)
+          x = root
+        end
+      end
+    end
+    x.color = Node::BLACK
+  end
+end
+
+def rbt_bm
+  n = 100_000
+  a1 = []; n.times { a1 << rand(999_999) }
+  a2 = []; n.times { a2 << rand(999_999) }
+
+  start = Time.now
+
+  tree = RedBlackTree.new
+
+  n.times {|i| tree.add(i) }
+  n.times { tree.delete(tree.root) }
+
+  tree = RedBlackTree.new
+  a1.each {|e| tree.add(e) }
+  a2.each {|e| tree.search(e) }
+  tree.inorder_walk {|key| key + 1 }
+  tree.reverse_inorder_walk {|key| key + 1 }
+  n.times { tree.minimum }
+  n.times { tree.maximum }
+
+  return Time.now - start
+end
+
+N = (ARGV[0] || 10).to_i
+
+N.times do
+  # puts rbt_bm.to_f
+  rbt_bm.to_f
+  # puts "GC.count = #{GC.count}" if GC.respond_to?(:count)
+end
diff --git a/benchmark/gc/ring.rb b/benchmark/gc/ring.rb
new file mode 100644
index 0000000000..be2c7b7250
--- /dev/null
+++ b/benchmark/gc/ring.rb
@@ -0,0 +1,29 @@
+# create many old objects
+
+max = 30_000_000
+
+class Ring
+  attr_reader :next_ring
+  def initialize n = nil
+    @next_ring = n
+  end
+
+
+  def size
+    s = 1
+    ring = self
+    while ring.next_ring
+      s += 1
+      ring = ring.next_ring
+    end
+    s
+  end
+end
+
+ring = Ring.new
+
+max.times{
+  ring = Ring.new(ring)
+}
+
+# p ring.size
diff --git a/benchmark/hash_aref_dsym.rb b/benchmark/hash_aref_dsym.rb
new file mode 100644
index 0000000000..af4f8c36d4
--- /dev/null
+++ b/benchmark/hash_aref_dsym.rb
@@ -0,0 +1,4 @@
+h = {}
+syms = ('a'..'z').map { |s| s.to_sym }
+syms.each { |s| h[s] = 1 }
+200_000.times { syms.each { |s| h[s] } }
diff --git a/benchmark/hash_aref_dsym_long.rb b/benchmark/hash_aref_dsym_long.rb
new file mode 100644
index 0000000000..9d7759379e
--- /dev/null
+++ b/benchmark/hash_aref_dsym_long.rb
@@ -0,0 +1,21 @@
+# [ruby-core:70129] [Bug #11396]
+collection_size = 200000
+sample_size = 10000
+
+values = (1..collection_size).to_a.map do |x|
+  "THIS IS A LONGER STRING THAT IS ALSO UNIQUE #{x}"
+end
+
+symbol_hash = {}
+
+values.each do |x|
+  symbol_hash[x.to_sym] = 1
+end
+
+# use the same samples each time to minimize deviations
+rng = Random.new(0)
+symbol_sample_array = values.sample(sample_size, random: rng).map(&:to_sym)
+
+3000.times do
+  symbol_sample_array.each { |x| symbol_hash[x] }
+end
diff --git a/benchmark/hash_aref_fix.rb b/benchmark/hash_aref_fix.rb
new file mode 100644
index 0000000000..1346890582
--- /dev/null
+++ b/benchmark/hash_aref_fix.rb
@@ -0,0 +1,4 @@
+h = {}
+nums = (1..26).to_a
+nums.each { |i| h[i] = i }
+200_000.times { nums.each { |s| h[s] } }
diff --git a/benchmark/hash_aref_flo.rb b/benchmark/hash_aref_flo.rb
new file mode 100644
index 0000000000..2217274c82
--- /dev/null
+++ b/benchmark/hash_aref_flo.rb
@@ -0,0 +1,4 @@
+h = {}
+strs = [*1..10000].map! {|i| i.fdiv(10)}
+strs.each { |s| h[s] = s }
+50.times { strs.each { |s| h[s] } }
diff --git a/benchmark/hash_aref_miss.rb b/benchmark/hash_aref_miss.rb
new file mode 100644
index 0000000000..b0913dd4bb
--- /dev/null
+++ b/benchmark/hash_aref_miss.rb
@@ -0,0 +1,5 @@
+h = {}
+strs = ('a'..'z').to_a.map!(&:freeze)
+strs.each { |s| h[s] = s }
+strs = ('A'..'Z').to_a
+200_000.times { strs.each { |s| h[s] } }
diff --git a/benchmark/hash_aref_str.rb b/benchmark/hash_aref_str.rb
new file mode 100644
index 0000000000..19439b061b
--- /dev/null
+++ b/benchmark/hash_aref_str.rb
@@ -0,0 +1,4 @@
+h = {}
+strs = ('a'..'z').to_a.map!(&:freeze)
+strs.each { |s| h[s] = s }
+200_000.times { strs.each { |s| h[s] } }
diff --git a/benchmark/hash_aref_sym.rb b/benchmark/hash_aref_sym.rb
new file mode 100644
index 0000000000..f75d163fe6
--- /dev/null
+++ b/benchmark/hash_aref_sym.rb
@@ -0,0 +1,9 @@
+h = {}
+syms = ('a'..'z').to_a
+begin
+  syms = eval("%i[#{syms.join(' ')}]")
+rescue SyntaxError # <= 1.9.3
+  syms.map!(&:to_sym)
+end
+syms.each { |s| h[s] = s }
+200_000.times { syms.each { |s| h[s] } }
diff --git a/benchmark/hash_aref_sym_long.rb b/benchmark/hash_aref_sym_long.rb
new file mode 100644
index 0000000000..9dab8df7be
--- /dev/null
+++ b/benchmark/hash_aref_sym_long.rb
@@ -0,0 +1,13 @@
+h = {}
+syms = %w[puts warn syswrite write stat bacon lettuce tomato
+some symbols in this array may already be interned  others should not be
+hash browns make good breakfast but not cooked using prime numbers
+shift for division entries delete_if keys exist?
+]
+begin
+  syms = eval("%i[#{syms.join(' ')}]")
+rescue SyntaxError # <= 1.9.3
+  syms.map!(&:to_sym)
+end
+syms.each { |s| h[s] = s }
+200_000.times { syms.each { |s| h[s] } }
diff --git a/benchmark/hash_flatten.rb b/benchmark/hash_flatten.rb
new file mode 100644
index 0000000000..e944aae9f2
--- /dev/null
+++ b/benchmark/hash_flatten.rb
@@ -0,0 +1,9 @@
+h = {}
+
+10000.times do |i|
+  h[i] = nil
+end
+
+1000.times do
+  h.flatten
+end
diff --git a/benchmark/hash_ident_flo.rb b/benchmark/hash_ident_flo.rb
new file mode 100644
index 0000000000..0c7edfed3e
--- /dev/null
+++ b/benchmark/hash_ident_flo.rb
@@ -0,0 +1,4 @@
+h = {}.compare_by_identity
+strs = (1..10000).to_a.map!(&:to_f)
+strs.each { |s| h[s] = s }
+50.times { strs.each { |s| h[s] } }
diff --git a/benchmark/hash_ident_num.rb b/benchmark/hash_ident_num.rb
new file mode 100644
index 0000000000..b226736c6f
--- /dev/null
+++ b/benchmark/hash_ident_num.rb
@@ -0,0 +1,4 @@
+h = {}.compare_by_identity
+nums = (1..26).to_a
+nums.each { |n| h[n] = n }
+200_000.times { nums.each { |n| h[n] } }
diff --git a/benchmark/hash_ident_obj.rb b/benchmark/hash_ident_obj.rb
new file mode 100644
index 0000000000..4b3b58edec
--- /dev/null
+++ b/benchmark/hash_ident_obj.rb
@@ -0,0 +1,4 @@
+h = {}.compare_by_identity
+objs = 26.times.map { Object.new }
+objs.each { |o| h[o] = o }
+200_000.times { objs.each { |o| h[o] } }
diff --git a/benchmark/hash_ident_str.rb b/benchmark/hash_ident_str.rb
new file mode 100644
index 0000000000..8582b38e31
--- /dev/null
+++ b/benchmark/hash_ident_str.rb
@@ -0,0 +1,4 @@
+h = {}.compare_by_identity
+strs = ('a'..'z').to_a
+strs.each { |s| h[s] = s }
+200_000.times { strs.each { |s| h[s] } }
diff --git a/benchmark/hash_ident_sym.rb b/benchmark/hash_ident_sym.rb
new file mode 100644
index 0000000000..4c81e3d28e
--- /dev/null
+++ b/benchmark/hash_ident_sym.rb
@@ -0,0 +1,4 @@
+h = {}.compare_by_identity
+syms = ('a'..'z').to_a.map(&:to_sym)
+syms.each { |s| h[s] = s }
+200_000.times { syms.each { |s| h[s] } }
diff --git a/benchmark/hash_keys.rb b/benchmark/hash_keys.rb
new file mode 100644
index 0000000000..6863cd01f9
--- /dev/null
+++ b/benchmark/hash_keys.rb
@@ -0,0 +1,9 @@
+h = {}
+
+10000.times do |i|
+  h[i] = nil
+end
+
+5000.times do
+  h.keys
+end
diff --git a/benchmark/hash_literal_small2.rb b/benchmark/hash_literal_small2.rb
new file mode 100644
index 0000000000..c188529260
--- /dev/null
+++ b/benchmark/hash_literal_small2.rb
@@ -0,0 +1,3 @@
+# frozen_string_literal: true
+
+1_000_000.times.map { { "foo" => "bar", "bar" => "baz" }  }
diff --git a/benchmark/hash_literal_small4.rb b/benchmark/hash_literal_small4.rb
new file mode 100644
index 0000000000..739f71b5b0
--- /dev/null
+++ b/benchmark/hash_literal_small4.rb
@@ -0,0 +1,3 @@
+# frozen_string_literal: true
+
+1_000_000.times.map { { "foo" => "bar", "bar" => "baz", "baz" => "lol", "lol" => "lgtm" } }
diff --git a/benchmark/hash_literal_small8.rb b/benchmark/hash_literal_small8.rb
new file mode 100644
index 0000000000..53d80af535
--- /dev/null
+++ b/benchmark/hash_literal_small8.rb
@@ -0,0 +1,3 @@
+# frozen_string_literal: true
+
+1_000_000.times.map { { "foo" => "bar", "bar" => "baz", "baz" => "lol", "lol" => "lgtm", "lgtm" => "nope", "nope" => "ok", "ok" => "again", "again" => "wait" } }
diff --git a/benchmark/hash_long.rb b/benchmark/hash_long.rb
new file mode 100644
index 0000000000..03d9109602
--- /dev/null
+++ b/benchmark/hash_long.rb
@@ -0,0 +1,4 @@
+k1 = "Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong";
+k2 = "Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping Pong Ping";
+h = {k1 => 0, k2 => 0};
+3000000.times{|i| k = i % 2 ? k2 : k1; h [k] = h[k] + 1}
diff --git a/benchmark/hash_shift.rb b/benchmark/hash_shift.rb
new file mode 100644
index 0000000000..a645671a5b
--- /dev/null
+++ b/benchmark/hash_shift.rb
@@ -0,0 +1,10 @@
+h = {}
+
+10000.times do |i|
+  h[i] = nil
+end
+
+50000.times do
+  k, v = h.shift
+  h[k] = v
+end
diff --git a/benchmark/hash_shift_u16.rb b/benchmark/hash_shift_u16.rb
new file mode 100644
index 0000000000..ec800d0342
--- /dev/null
+++ b/benchmark/hash_shift_u16.rb
@@ -0,0 +1,10 @@
+h = {}
+
+(16384..65536).each do |i|
+  h[i] = nil
+end
+
+300000.times do
+  k, v = h.shift
+  h[k] = v
+end
diff --git a/benchmark/hash_shift_u24.rb b/benchmark/hash_shift_u24.rb
new file mode 100644
index 0000000000..de4e0fa696
--- /dev/null
+++ b/benchmark/hash_shift_u24.rb
@@ -0,0 +1,10 @@
+h = {}
+
+(0xff4000..0xffffff).each do |i|
+  h[i] = nil
+end
+
+300000.times do
+  k, v = h.shift
+  h[k] = v
+end
diff --git a/benchmark/hash_shift_u32.rb b/benchmark/hash_shift_u32.rb
new file mode 100644
index 0000000000..656aa55583
--- /dev/null
+++ b/benchmark/hash_shift_u32.rb
@@ -0,0 +1,10 @@
+h = {}
+
+(0xffff4000..0xffffffff).each do |i|
+  h[i] = nil
+end
+
+300000.times do
+  k, v = h.shift
+  h[k] = v
+end
diff --git a/benchmark/hash_small2.rb b/benchmark/hash_small2.rb
new file mode 100644
index 0000000000..45485d9c71
--- /dev/null
+++ b/benchmark/hash_small2.rb
@@ -0,0 +1 @@
+1000000.times.map{|i| a={}; 2.times{|j| a[j]=j}; a}
diff --git a/benchmark/hash_small4.rb b/benchmark/hash_small4.rb
new file mode 100644
index 0000000000..acd4084334
--- /dev/null
+++ b/benchmark/hash_small4.rb
@@ -0,0 +1 @@
+1000000.times.map{|i| a={}; 4.times{|j| a[j]=j}; a}
diff --git a/benchmark/hash_small8.rb b/benchmark/hash_small8.rb
new file mode 100644
index 0000000000..9cffcc91b6
--- /dev/null
+++ b/benchmark/hash_small8.rb
@@ -0,0 +1 @@
+1000000.times.map{|i| a={}; 8.times{|j| a[j]=j}; a}
diff --git a/benchmark/hash_to_proc.rb b/benchmark/hash_to_proc.rb
new file mode 100644
index 0000000000..2b675bf509
--- /dev/null
+++ b/benchmark/hash_to_proc.rb
@@ -0,0 +1,9 @@
+h = {}
+
+10000.times do |i|
+  h[i] = nil
+end
+
+5000.times do |i|
+  [i].map(&h)
+end
diff --git a/benchmark/hash_values.rb b/benchmark/hash_values.rb
new file mode 100644
index 0000000000..069441302f
--- /dev/null
+++ b/benchmark/hash_values.rb
@@ -0,0 +1,9 @@
+h = {}
+
+10000.times do |i|
+  h[i] = nil
+end
+
+5000.times do
+  h.values
+end
diff --git a/benchmark/int_quo.rb b/benchmark/int_quo.rb
new file mode 100644
index 0000000000..e22a3f8c30
--- /dev/null
+++ b/benchmark/int_quo.rb
@@ -0,0 +1 @@
+5000000.times { 42.quo(3) }
diff --git a/benchmark/io_copy_stream_write.rb b/benchmark/io_copy_stream_write.rb
new file mode 100644
index 0000000000..3fd87250a4
--- /dev/null
+++ b/benchmark/io_copy_stream_write.rb
@@ -0,0 +1,24 @@
+# The goal of this is to use a synthetic (non-IO) reader
+# to trigger the read/write loop of IO.copy_stream,
+# bypassing in-kernel mechanisms like sendfile for zero copy,
+# so we wrap the /dev/zero IO object:
+
+class Zero
+  def initialize
+    @n = 100000
+    @in = File.open('/dev/zero', 'rb')
+  end
+
+  def read(len, buf)
+    return if (@n -= 1) == 0
+    @in.read(len, buf)
+  end
+end
+
+begin
+  src = Zero.new
+  dst = File.open(IO::NULL, 'wb')
+  n = IO.copy_stream(src, dst)
+rescue Errno::ENOENT
+  # not *nix
+end if IO.respond_to?(:copy_stream) && IO.const_defined?(:NULL)
diff --git a/benchmark/io_copy_stream_write_socket.rb b/benchmark/io_copy_stream_write_socket.rb
new file mode 100644
index 0000000000..11f369bd0d
--- /dev/null
+++ b/benchmark/io_copy_stream_write_socket.rb
@@ -0,0 +1,35 @@
+# The goal of this is to use a synthetic (non-IO) reader
+# to trigger the read/write loop of IO.copy_stream,
+# bypassing in-kernel mechanisms like sendfile for zero copy,
+# so we wrap the /dev/zero IO object:
+class Zero
+  def initialize
+    @n = 100000
+    @in = File.open('/dev/zero', 'rb')
+  end
+
+  def read(len, buf)
+    return if (@n -= 1) == 0
+    @in.read(len, buf)
+  end
+end
+
+begin
+  require 'socket'
+  src = Zero.new
+  rd, wr = UNIXSocket.pair
+  pid = fork do
+    wr.close
+    buf = String.new
+    while rd.read(16384, buf)
+    end
+  end
+  rd.close
+  IO.copy_stream(src, wr)
+rescue Errno::ENOENT, NotImplementedError, NameError
+  # not *nix: missing /dev/zero, fork, or UNIXSocket
+rescue LoadError # no socket?
+ensure
+  wr.close if wr
+  Process.waitpid(pid) if pid
+end if IO.respond_to?(:copy_stream)
diff --git a/benchmark/io_file_create.rb b/benchmark/io_file_create.rb
new file mode 100644
index 0000000000..2f205c1333
--- /dev/null
+++ b/benchmark/io_file_create.rb
@@ -0,0 +1,13 @@
+#
+# Create files
+#
+
+max = 200_000
+file = './tmpfile_of_bm_io_file_create'
+
+max.times{
+  f = open(file, 'w')
+  f.close#(true)
+}
+File.unlink(file)
+
diff --git a/benchmark/io_file_read.rb b/benchmark/io_file_read.rb
new file mode 100644
index 0000000000..b9e796ed30
--- /dev/null
+++ b/benchmark/io_file_read.rb
@@ -0,0 +1,15 @@
+#
+# Seek and Read file.
+#
+
+require 'tempfile'
+
+max = 200_000
+str = "Hello world!  " * 1000
+f = Tempfile.new('yarv-benchmark')
+f.write str
+
+max.times{
+  f.seek 0
+  f.read
+}
diff --git a/benchmark/io_file_write.rb b/benchmark/io_file_write.rb
new file mode 100644
index 0000000000..aa1be0e5fe
--- /dev/null
+++ b/benchmark/io_file_write.rb
@@ -0,0 +1,14 @@
+#
+# Seek and Write file.
+#
+
+require 'tempfile'
+
+max = 200_000
+str = "Hello world!  " * 1000
+f = Tempfile.new('yarv-benchmark')
+
+max.times{
+  f.seek 0
+  f.write str
+}
diff --git a/benchmark/io_nonblock_noex.rb b/benchmark/io_nonblock_noex.rb
new file mode 100644
index 0000000000..da9357fdc6
--- /dev/null
+++ b/benchmark/io_nonblock_noex.rb
@@ -0,0 +1,22 @@
+nr = 1_000_000
+i = 0
+msg = '.'
+buf = '.'
+noex = { exception: false }
+begin
+  r, w = IO.pipe
+  while i < nr
+    i += 1
+    w.write_nonblock(msg, noex)
+    r.read_nonblock(1, buf, noex)
+  end
+rescue ArgumentError # old Rubies
+  while i < nr
+    i += 1
+    w.write_nonblock(msg)
+    r.read_nonblock(1, buf)
+  end
+ensure
+  r.close
+  w.close
+end
diff --git a/benchmark/io_nonblock_noex2.rb b/benchmark/io_nonblock_noex2.rb
new file mode 100644
index 0000000000..56819d049b
--- /dev/null
+++ b/benchmark/io_nonblock_noex2.rb
@@ -0,0 +1,21 @@
+nr = 1_000_000
+i = 0
+msg = '.'
+buf = '.'
+begin
+  r, w = IO.pipe
+  while i < nr
+    i += 1
+    w.write_nonblock(msg, exception: false)
+    r.read_nonblock(1, buf, exception: false)
+  end
+rescue ArgumentError # old Rubies
+  while i < nr
+    i += 1
+    w.write_nonblock(msg)
+    r.read_nonblock(1, buf)
+  end
+ensure
+  r.close
+  w.close
+end
diff --git a/benchmark/io_pipe_rw.rb b/benchmark/io_pipe_rw.rb
new file mode 100644
index 0000000000..6862a8ae61
--- /dev/null
+++ b/benchmark/io_pipe_rw.rb
@@ -0,0 +1,13 @@
+# Measure uncontended GVL performance via read/write with 1:1 threading
+# If we switch to M:N threading, this will benchmark something else...
+r, w = IO.pipe
+src = '0'.freeze
+dst = String.new
+i = 0
+while i < 1_000_000
+  i += 1
+  w.write(src)
+  r.read(1, dst)
+end
+w.close
+r.close
diff --git a/benchmark/io_select.rb b/benchmark/io_select.rb
new file mode 100644
index 0000000000..19248daeb1
--- /dev/null
+++ b/benchmark/io_select.rb
@@ -0,0 +1,9 @@
+# IO.select performance
+
+w = [ IO.pipe[1] ];
+
+nr = 1000000
+nr.times {
+  IO.select nil, w
+}
+
diff --git a/benchmark/io_select2.rb b/benchmark/io_select2.rb
new file mode 100644
index 0000000000..10e37d71b2
--- /dev/null
+++ b/benchmark/io_select2.rb
@@ -0,0 +1,22 @@
+# IO.select performance. worst case of single fd.
+
+ios = []
+nr = 1000000
+if defined?(Process::RLIMIT_NOFILE)
+  max = Process.getrlimit(Process::RLIMIT_NOFILE)[0]
+else
+  max = 64
+end
+puts "max fd: #{max} (results not apparent with <= 1024 max fd)"
+
+((max / 2) - 10).times do
+  ios.concat IO.pipe
+end
+
+last = [ ios[-1] ]
+puts "last IO: #{last[0].inspect}"
+
+nr.times do
+  IO.select nil, last
+end
+
diff --git a/benchmark/io_select3.rb b/benchmark/io_select3.rb
new file mode 100644
index 0000000000..7d0ba1f092
--- /dev/null
+++ b/benchmark/io_select3.rb
@@ -0,0 +1,21 @@
+# IO.select performance. a lot of fd
+
+ios = []
+nr = 100
+if defined?(Process::RLIMIT_NOFILE)
+  max = Process.getrlimit(Process::RLIMIT_NOFILE)[0]
+else
+  max = 64
+end
+puts "max fd: #{max} (results not apparent with <= 1024 max fd)"
+
+(max - 10).times do
+  r, w = IO.pipe
+  r.close
+  ios.push w
+end
+
+nr.times do
+  IO.select nil, ios
+end
+
diff --git a/benchmark/lib/benchmark_driver/output/driver.rb b/benchmark/lib/benchmark_driver/output/driver.rb
new file mode 100644
index 0000000000..d22236e9fb
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/output/driver.rb
@@ -0,0 +1,36 @@
+require 'benchmark_driver/output/simple'
+
+# This replicates the legacy benchmark/driver.rb behavior.
+class BenchmarkDriver::Output::Driver < BenchmarkDriver::Output::Simple
+  def initialize(*)
+    super
+    @stdout = $stdout
+    @strio  = StringIO.new
+    $stdout = IOMultiplexer.new(@stdout, @strio)
+  end
+
+  def with_benchmark(*)
+    super
+  ensure
+    logfile = "bmlog-#{Time.now.strftime('%Y%m%d-%H%M%S')}.#{$$}.log"
+    puts "\nLog file: #{logfile}"
+
+    $stdout = @stdout
+    File.write(logfile, @strio.tap(&:rewind).read)
+  end
+
+  class IOMultiplexer
+    def initialize(io1, io2)
+      @io1 = io1
+      @io2 = io2
+    end
+
+    [:write, :sync, :sync=, :puts, :print, :flush].each do |method|
+      define_method(method) do |*args|
+        @io1.send(method, *args)
+        @io2.send(method, *args)
+      end
+    end
+  end
+  private_constant :IOMultiplexer
+end
diff --git a/benchmark/lib/benchmark_driver/runner/cstime.rb b/benchmark/lib/benchmark_driver/runner/cstime.rb
new file mode 100644
index 0000000000..3c3453e527
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/runner/cstime.rb
@@ -0,0 +1,22 @@
+require 'benchmark_driver/runner/total'
+
+class BenchmarkDriver::Runner::Cstime < BenchmarkDriver::Runner::Total
+  METRIC = BenchmarkDriver::Metric.new(name: 'cstime', unit: 's', larger_better: false)
+
+  # JobParser returns this, `BenchmarkDriver::Runner.runner_for` searches "*::Job"
+  Job = Class.new(BenchmarkDriver::DefaultJob)
+  # Dynamically fetched and used by `BenchmarkDriver::JobParser.parse`
+  JobParser = BenchmarkDriver::DefaultJobParser.for(klass: Job, metrics: [METRIC])
+
+  private
+
+  # Overriding BenchmarkDriver::Runner::Total#metric
+  def metric
+    METRIC
+  end
+
+  # Overriding BenchmarkDriver::Runner::Total#target
+  def target
+    :cstime
+  end
+end
diff --git a/benchmark/lib/benchmark_driver/runner/cutime.rb b/benchmark/lib/benchmark_driver/runner/cutime.rb
new file mode 100644
index 0000000000..e139962ef2
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/runner/cutime.rb
@@ -0,0 +1,22 @@
+require 'benchmark_driver/runner/total'
+
+class BenchmarkDriver::Runner::Cutime < BenchmarkDriver::Runner::Total
+  METRIC = BenchmarkDriver::Metric.new(name: 'cutime', unit: 's', larger_better: false)
+
+  # JobParser returns this, `BenchmarkDriver::Runner.runner_for` searches "*::Job"
+  Job = Class.new(BenchmarkDriver::DefaultJob)
+  # Dynamically fetched and used by `BenchmarkDriver::JobParser.parse`
+  JobParser = BenchmarkDriver::DefaultJobParser.for(klass: Job, metrics: [METRIC])
+
+  private
+
+  # Overriding BenchmarkDriver::Runner::Total#metric
+  def metric
+    METRIC
+  end
+
+  # Overriding BenchmarkDriver::Runner::Total#target
+  def target
+    :cutime
+  end
+end
diff --git a/benchmark/lib/benchmark_driver/runner/peak.rb b/benchmark/lib/benchmark_driver/runner/peak.rb
new file mode 100644
index 0000000000..0ba8af42c5
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/runner/peak.rb
@@ -0,0 +1,151 @@
+require 'benchmark_driver/struct'
+require 'benchmark_driver/metric'
+require 'benchmark_driver/default_job'
+require 'benchmark_driver/default_job_parser'
+require 'tempfile'
+
+class BenchmarkDriver::Runner::Peak
+  METRIC = BenchmarkDriver::Metric.new(
+    name: 'Peak memory usage', unit: 'bytes', larger_better: false, worse_word: 'larger',
+  )
+
+  # JobParser returns this, `BenchmarkDriver::Runner.runner_for` searches "*::Job"
+  Job = Class.new(BenchmarkDriver::DefaultJob)
+  # Dynamically fetched and used by `BenchmarkDriver::JobParser.parse`
+  JobParser = BenchmarkDriver::DefaultJobParser.for(klass: Job, metrics: [METRIC])
+
+  # @param [BenchmarkDriver::Config::RunnerConfig] config
+  # @param [BenchmarkDriver::Output] output
+  # @param [BenchmarkDriver::Context] contexts
+  def initialize(config:, output:, contexts:)
+    @config = config
+    @output = output
+    @contexts = contexts
+  end
+
+  # This method is dynamically called by `BenchmarkDriver::JobRunner.run`
+  # @param [Array<BenchmarkDriver::Runner::Peak::Job>] jobs
+  def run(jobs)
+    if jobs.any? { |job| job.loop_count.nil? }
+      jobs = jobs.map do |job|
+        job.loop_count ? job : Job.new(job.to_h.merge(loop_count: 1))
+      end
+    end
+
+    @output.with_benchmark do
+      jobs.each do |job|
+        @output.with_job(name: job.name) do
+          job.runnable_contexts(@contexts).each do |context|
+            value = BenchmarkDriver::Repeater.with_repeat(config: @config, larger_better: false) do
+              run_benchmark(job, context: context)
+            end
+            @output.with_context(name: context.name, executable: context.executable, gems: context.gems, prelude: context.prelude) do
+              @output.report(values: { metric => value }, loop_count: job.loop_count)
+            end
+          end
+        end
+      end
+    end
+  end
+
+  private
+
+  # @param [BenchmarkDriver::Runner::Ips::Job] job - loop_count is not nil
+  # @param [BenchmarkDriver::Context] context
+  # @return [BenchmarkDriver::Metrics]
+  def run_benchmark(job, context:)
+    benchmark = BenchmarkScript.new(
+      preludes:   [context.prelude, job.prelude],
+      script:     job.script,
+      teardown:   job.teardown,
+      loop_count: job.loop_count,
+    )
+
+    memory_status = File.expand_path('../../../../test/lib/memory_status', __dir__)
+    Tempfile.open(['benchmark_driver-', '.rb']) do |f|
+      with_script(benchmark.render) do |path|
+        output = IO.popen([*context.executable.command, path, f.path, target, memory_status], &:read)
+        if $?.success?
+          Integer(f.read)
+        else
+          $stdout.print(output)
+          BenchmarkDriver::Result::ERROR
+        end
+      end
+    end
+  end
+
+  # Overridden by BenchmarkDriver::Runner::Size
+  def target
+    'peak'
+  end
+
+  # Overridden by BenchmarkDriver::Runner::Size
+  def metric
+    METRIC
+  end
+
+  def with_script(script)
+    if @config.verbose >= 2
+      sep = '-' * 30
+      $stdout.puts "\n\n#{sep}[Script begin]#{sep}\n#{script}#{sep}[Script end]#{sep}\n\n"
+    end
+
+    Tempfile.open(['benchmark_driver-', '.rb']) do |f|
+      f.puts script
+      f.close
+      return yield(f.path)
+    end
+  end
+
+  # @param [String] prelude
+  # @param [String] script
+  # @param [String] teardown
+  # @param [Integer] loop_count
+  BenchmarkScript = ::BenchmarkDriver::Struct.new(:preludes, :script, :teardown, :loop_count) do
+    def render
+      prelude = preludes.reject(&:nil?).reject(&:empty?).join("\n")
+      <<-RUBY
+#{prelude}
+#{while_loop(script, loop_count)}
+#{teardown}
+
+result_file, target, memory_status = ARGV
+require_relative memory_status
+
+ms = Memory::Status.new
+case target.to_sym
+when :peak
+  key = ms.respond_to?(:hwm) ? :hwm : :peak
+when :size
+  key = ms.respond_to?(:rss) ? :rss : :size
+else
+  raise('unexpected target: ' + target)
+end
+
+File.write(result_file, ms[key])
+      RUBY
+    end
+
+    private
+
+    def while_loop(content, times)
+      if !times.is_a?(Integer) || times <= 0
+        raise ArgumentError.new("Unexpected times: #{times.inspect}")
+      end
+
+      if times > 1
+        <<-RUBY
+__bmdv_i = 0
+while __bmdv_i < #{times}
+  #{content}
+  __bmdv_i += 1
+end
+        RUBY
+      else
+        content
+      end
+    end
+  end
+  private_constant :BenchmarkScript
+end
diff --git a/benchmark/lib/benchmark_driver/runner/size.rb b/benchmark/lib/benchmark_driver/runner/size.rb
new file mode 100644
index 0000000000..1b31f901c7
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/runner/size.rb
@@ -0,0 +1,25 @@
+require 'benchmark_driver/runner/peak'
+
+# Actually the same as BenchmarkDriver::Runner::Memory
+class BenchmarkDriver::Runner::Size < BenchmarkDriver::Runner::Peak
+  METRIC = BenchmarkDriver::Metric.new(
+    name: 'Max resident set size', unit: 'bytes', larger_better: false, worse_word: 'larger',
+  )
+
+  # JobParser returns this, `BenchmarkDriver::Runner.runner_for` searches "*::Job"
+  Job = Class.new(BenchmarkDriver::DefaultJob)
+  # Dynamically fetched and used by `BenchmarkDriver::JobParser.parse`
+  JobParser = BenchmarkDriver::DefaultJobParser.for(klass: Job, metrics: [METRIC])
+
+  private
+
+  # Overriding BenchmarkDriver::Runner::Peak#metric
+  def metric
+    METRIC
+  end
+
+  # Overriding BenchmarkDriver::Runner::Peak#target
+  def target
+    'size'
+  end
+end
diff --git a/benchmark/lib/benchmark_driver/runner/stime.rb b/benchmark/lib/benchmark_driver/runner/stime.rb
new file mode 100644
index 0000000000..4577fb0bf8
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/runner/stime.rb
@@ -0,0 +1,22 @@
+require 'benchmark_driver/runner/total'
+
+class BenchmarkDriver::Runner::Stime < BenchmarkDriver::Runner::Total
+  METRIC = BenchmarkDriver::Metric.new(name: 'stime', unit: 's', larger_better: false)
+
+  # JobParser returns this, `BenchmarkDriver::Runner.runner_for` searches "*::Job"
+  Job = Class.new(BenchmarkDriver::DefaultJob)
+  # Dynamically fetched and used by `BenchmarkDriver::JobParser.parse`
+  JobParser = BenchmarkDriver::DefaultJobParser.for(klass: Job, metrics: [METRIC])
+
+  private
+
+  # Overriding BenchmarkDriver::Runner::Total#metric
+  def metric
+    METRIC
+  end
+
+  # Overriding BenchmarkDriver::Runner::Total#target
+  def target
+    :stime
+  end
+end
diff --git a/benchmark/lib/benchmark_driver/runner/total.rb b/benchmark/lib/benchmark_driver/runner/total.rb
new file mode 100644
index 0000000000..64dc14f84e
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/runner/total.rb
@@ -0,0 +1,137 @@
+require 'benchmark_driver/struct'
+require 'benchmark_driver/metric'
+require 'benchmark_driver/default_job'
+require 'benchmark_driver/default_job_parser'
+require 'tempfile'
+
+class BenchmarkDriver::Runner::Total
+  METRIC = BenchmarkDriver::Metric.new(name: 'Total time', unit: 's', larger_better: false)
+
+  # JobParser returns this, `BenchmarkDriver::Runner.runner_for` searches "*::Job"
+  Job = Class.new(BenchmarkDriver::DefaultJob)
+  # Dynamically fetched and used by `BenchmarkDriver::JobParser.parse`
+  JobParser = BenchmarkDriver::DefaultJobParser.for(klass: Job, metrics: [METRIC])
+
+  # @param [BenchmarkDriver::Config::RunnerConfig] config
+  # @param [BenchmarkDriver::Output] output
+  # @param [BenchmarkDriver::Context] contexts
+  def initialize(config:, output:, contexts:)
+    @config = config
+    @output = output
+    @contexts = contexts
+  end
+
+  # This method is dynamically called by `BenchmarkDriver::JobRunner.run`
+  # @param [Array<BenchmarkDriver::Runner::Total::Job>] jobs
+  def run(jobs)
+    if jobs.any? { |job| job.loop_count.nil? }
+      raise 'missing loop_count is not supported in Ruby repository'
+    end
+
+    @output.with_benchmark do
+      jobs.each do |job|
+        @output.with_job(name: job.name) do
+          job.runnable_contexts(@contexts).each do |context|
+            duration = BenchmarkDriver::Repeater.with_repeat(config: @config, larger_better: false) do
+              run_benchmark(job, context: context)
+            end
+            @output.with_context(name: context.name, executable: context.executable, gems: context.gems, prelude: context.prelude) do
+              @output.report(values: { metric => duration }, duration: duration, loop_count: job.loop_count)
+            end
+          end
+        end
+      end
+    end
+  end
+
+  private
+
+  # @param [BenchmarkDriver::Runner::Ips::Job] job - loop_count is not nil
+  # @param [BenchmarkDriver::Context] context
+  # @return [BenchmarkDriver::Metrics]
+  def run_benchmark(job, context:)
+    benchmark = BenchmarkScript.new(
+      preludes:   [context.prelude, job.prelude],
+      script:     job.script,
+      teardown:   job.teardown,
+      loop_count: job.loop_count,
+    )
+
+    Tempfile.open(['benchmark_driver-', '.rb']) do |f|
+      with_script(benchmark.render(result: f.path, target: target)) do |path|
+        IO.popen([*context.executable.command, path], &:read) # TODO: print stdout if verbose=2
+        if $?.success?
+          Float(f.read)
+        else
+          BenchmarkDriver::Result::ERROR
+        end
+      end
+    end
+  end
+
+  # This method is overridden by some subclasses
+  def metric
+    METRIC
+  end
+
+  # This method is overridden by some subclasses
+  def target
+    :total
+  end
+
+  def with_script(script)
+    if @config.verbose >= 2
+      sep = '-' * 30
+      $stdout.puts "\n\n#{sep}[Script begin]#{sep}\n#{script}#{sep}[Script end]#{sep}\n\n"
+    end
+
+    Tempfile.open(['benchmark_driver-', '.rb']) do |f|
+      f.puts script
+      f.close
+      return yield(f.path)
+    end
+  end
+
+  # @param [String] prelude
+  # @param [String] script
+  # @param [String] teardown
+  # @param [Integer] loop_count
+  BenchmarkScript = ::BenchmarkDriver::Struct.new(:preludes, :script, :teardown, :loop_count) do
+    # @param [String] result - A file to write result
+    def render(result:, target:)
+      prelude = preludes.reject(&:nil?).reject(&:empty?).join("\n")
+      <<-RUBY
+#{prelude}
+
+require 'benchmark'
+__bmdv_result = Benchmark.measure {
+  #{while_loop(script, loop_count)}
+}
+
+#{teardown}
+
+File.write(#{result.dump}, __bmdv_result.#{target})
+      RUBY
+    end
+
+    private
+
+    def while_loop(content, times)
+      if !times.is_a?(Integer) || times <= 0
+        raise ArgumentError.new("Unexpected times: #{times.inspect}")
+      elsif times == 1
+        return content
+      end
+
+      # TODO: execute in batch
+      <<-RUBY
+__bmdv_i = 0
+while __bmdv_i < #{times}
+  #{content}
+  __bmdv_i += 1
+end
+      RUBY
+    end
+  end
+  private_constant :BenchmarkScript
+end
diff --git a/benchmark/lib/benchmark_driver/runner/utime.rb b/benchmark/lib/benchmark_driver/runner/utime.rb
new file mode 100644
index 0000000000..b61d83a188
--- /dev/null
+++ b/benchmark/lib/benchmark_driver/runner/utime.rb
@@ -0,0 +1,22 @@
+require 'benchmark_driver/runner/total'
+
+class BenchmarkDriver::Runner::Utime < BenchmarkDriver::Runner::Total
+  METRIC = BenchmarkDriver::Metric.new(name: 'utime', unit: 's', larger_better: false)
+
+  # JobParser returns this, `BenchmarkDriver::Runner.runner_for` searches "*::Job"
+  Job = Class.new(BenchmarkDriver::DefaultJob)
+  # Dynamically fetched and used by `BenchmarkDriver::JobParser.parse`
+  JobParser = BenchmarkDriver::DefaultJobParser.for(klass: Job, metrics: [METRIC])
+
+  private
+
+  # Overriding BenchmarkDriver::Runner::Total#metric
+  def metric
+    METRIC
+  end
+
+  # Overriding BenchmarkDriver::Runner::Total#target
+  def target
+    :utime
+  end
+end
diff --git a/benchmark/lib/load.rb b/benchmark/lib/load.rb
new file mode 100755
index 0000000000..4d73a63323
--- /dev/null
+++ b/benchmark/lib/load.rb
@@ -0,0 +1,2 @@
+$:.unshift(File.join(__dir__, '../benchmark-driver/lib'))
+require 'benchmark_driver'
diff --git a/benchmark/loop_for.rb b/benchmark/loop_for.rb
new file mode 100644
index 0000000000..0fc4cc1511
--- /dev/null
+++ b/benchmark/loop_for.rb
@@ -0,0 +1,3 @@
+for i in 1..30_000_000
+  #
+end
diff --git a/benchmark/loop_generator.rb b/benchmark/loop_generator.rb
new file mode 100644
index 0000000000..d3375c744c
--- /dev/null
+++ b/benchmark/loop_generator.rb
@@ -0,0 +1,14 @@
+max = 600000
+
+if defined? Fiber
+  gen = (1..max).each
+  loop do
+    gen.next
+  end
+else
+  require 'generator'
+  gen = Generator.new((0..max))
+  while gen.next?
+    gen.next
+  end
+end
diff --git a/benchmark/loop_times.rb b/benchmark/loop_times.rb
new file mode 100644
index 0000000000..521f72ad1a
--- /dev/null
+++ b/benchmark/loop_times.rb
@@ -0,0 +1 @@
+30_000_000.times{|e|}
diff --git a/benchmark/loop_whileloop.rb b/benchmark/loop_whileloop.rb
new file mode 100644
index 0000000000..0072822c06
--- /dev/null
+++ b/benchmark/loop_whileloop.rb
@@ -0,0 +1,4 @@
+i = 0
+while i<30_000_000 # benchmark loop 1
+  i += 1
+end
diff --git a/benchmark/loop_whileloop2.rb b/benchmark/loop_whileloop2.rb
new file mode 100644
index 0000000000..47d02dffc4
--- /dev/null
+++ b/benchmark/loop_whileloop2.rb
@@ -0,0 +1,4 @@
+i = 0
+while i< 6_000_000 # benchmark loop 2
+  i += 1
+end
diff --git a/benchmark/marshal_dump_flo.rb b/benchmark/marshal_dump_flo.rb
new file mode 100644
index 0000000000..9b8d0c6afb
--- /dev/null
+++ b/benchmark/marshal_dump_flo.rb
@@ -0,0 +1,2 @@
+bug10761 = 10000.times.map { |x| x.to_f }
+100.times { Marshal.dump(bug10761) }
diff --git a/benchmark/marshal_dump_load_geniv.rb b/benchmark/marshal_dump_load_geniv.rb
new file mode 100644
index 0000000000..8252ad90fa
--- /dev/null
+++ b/benchmark/marshal_dump_load_geniv.rb
@@ -0,0 +1,10 @@
+a = ''
+a.instance_eval do
+  @a = :a
+  @b = :b
+  @c = :c
+end
+100000.times do
+  a = Marshal.load(Marshal.dump(a))
+end
+#p(a.instance_eval { @a == :a && @b == :b && @c == :c })
diff --git a/benchmark/marshal_dump_load_time.rb b/benchmark/marshal_dump_load_time.rb
new file mode 100644
index 0000000000..e29743b791
--- /dev/null
+++ b/benchmark/marshal_dump_load_time.rb
@@ -0,0 +1 @@
+100000.times { Marshal.load(Marshal.dump(Time.now)) }
diff --git a/benchmark/other-lang/ack.pl b/benchmark/other-lang/ack.pl
new file mode 100644
index 0000000000..201e22ddfa
--- /dev/null
+++ b/benchmark/other-lang/ack.pl
@@ -0,0 +1,11 @@
+use integer;
+
+sub Ack {
+    return $_[0] ? ($_[1] ? Ack($_[0]-1, Ack($_[0], $_[1]-1))
+		    : Ack($_[0]-1, 1))
+	: $_[1]+1;
+}
+
+my $NUM = 9;
+$NUM = 1 if ($NUM < 1);
+my $ack = Ack(3, $NUM);
diff --git a/benchmark/other-lang/ack.py b/benchmark/other-lang/ack.py
new file mode 100644
index 0000000000..9968e7cfcf
--- /dev/null
+++ b/benchmark/other-lang/ack.py
@@ -0,0 +1,16 @@
+import sys
+sys.setrecursionlimit(5000000)
+
+def Ack(M, N):
+    if (not M):
+        return( N + 1 )
+    if (not N):
+        return( Ack(M-1, 1) )
+    return( Ack(M-1, Ack(M, N-1)) )
+
+def main():
+    NUM = 9
+    sys.setrecursionlimit(10000)
+    Ack(3, NUM)
+
+main()
diff --git a/benchmark/other-lang/ack.rb b/benchmark/other-lang/ack.rb
new file mode 100644
index 0000000000..7451bed6c4
--- /dev/null
+++ b/benchmark/other-lang/ack.rb
@@ -0,0 +1,12 @@
+def ack(m, n)
+    if m == 0 then
+        n + 1
+    elsif n == 0 then
+        ack(m - 1, 1)
+    else
+        ack(m - 1, ack(m, n - 1))
+    end
+end
+
+NUM = 9
+ack(3, NUM)
diff --git a/benchmark/other-lang/ack.scm b/benchmark/other-lang/ack.scm
new file mode 100644
index 0000000000..a80b73ba55
--- /dev/null
+++ b/benchmark/other-lang/ack.scm
@@ -0,0 +1,7 @@
+(define (ack m n)
+  (cond ((zero? m) (+ n 1))
+	((zero? n) (ack (- m 1) 1))
+	(else      (ack (- m 1) (ack m (- n 1))))))
+
+(ack 3 9)
+
diff --git a/benchmark/other-lang/eval.rb b/benchmark/other-lang/eval.rb
new file mode 100644
index 0000000000..48a2cea019
--- /dev/null
+++ b/benchmark/other-lang/eval.rb
@@ -0,0 +1,66 @@
+
+Bench = %w(
+  loop
+  ack
+  fib
+  tak
+  fact
+)
+
+Lang = <<EOP.map{|l| l.strip}
+  ruby-cyg
+  ../../../test6/miniruby
+  perl
+  python
+  gosh
+EOP
+
+Bench.replace ['loop2']
+Lang.replace ['ruby-cyg']
+
+Ext = %w(
+  .rb
+  .rb
+  .pl
+  .py
+  .scm
+)
+
+p Bench
+p Lang
+
+require 'benchmark'
+
+def bench cmd
+  m = Benchmark.measure{
+    #p cmd
+    system(cmd)
+  }
+  [m.utime, m.real]
+end
+
+Result = []
+Bench.each{|b|
+  r = []
+  Lang.each_with_index{|l, idx|
+    cmd = "#{l} #{b}#{Ext[idx]}"
+    r << bench(cmd)
+  }
+  Result << r
+}
+
+require 'pp'
+# utime
+puts Lang.join("\t")
+Bench.each_with_index{|b, bi|
+  print b, "\t"
+  puts Result[bi].map{|e| e[0]}.join("\t")
+}
+
+# rtime
+puts Lang.join("\t")
+Bench.each_with_index{|b, bi|
+  print b, "\t"
+  puts Result[bi].map{|e| e[1]}.join("\t")
+}
+
diff --git a/benchmark/other-lang/fact.pl b/benchmark/other-lang/fact.pl
new file mode 100644
index 0000000000..a9b0b69cdf
--- /dev/null
+++ b/benchmark/other-lang/fact.pl
@@ -0,0 +1,13 @@
+sub fact{
+  my $n = @_[0];
+  if($n < 2){
+    return 1;
+  }
+  else{
+    return $n * fact($n-1);
+  }
+}
+
+for($i=0; $i<10000; $i++){
+  &fact(100);
+}
diff --git a/benchmark/other-lang/fact.py b/benchmark/other-lang/fact.py
new file mode 100644
index 0000000000..01593965d9
--- /dev/null
+++ b/benchmark/other-lang/fact.py
@@ -0,0 +1,18 @@
+#import sys
+#sys.setrecursionlimit(1000)
+
+def factL(n):
+	r = 1
+	for x in range(2, n):
+		r *= x
+	return r
+
+def factR(n):
+	if n < 2:
+		return 1
+	else:
+		return n * factR(n-1)
+
+for i in range(10000):
+	factR(100)
+
diff --git a/benchmark/other-lang/fact.rb b/benchmark/other-lang/fact.rb
new file mode 100644
index 0000000000..6cedc752cd
--- /dev/null
+++ b/benchmark/other-lang/fact.rb
@@ -0,0 +1,13 @@
+def fact(n)
+  if n < 2
+    1
+  else
+    n * fact(n-1)
+  end
+end
+
+i = 0
+while i<10000
+  i += 1
+  fact(100)
+end
diff --git a/benchmark/other-lang/fact.scm b/benchmark/other-lang/fact.scm
new file mode 100644
index 0000000000..c98a7fedd3
--- /dev/null
+++ b/benchmark/other-lang/fact.scm
@@ -0,0 +1,8 @@
+(define (fact n)
+        (if (< n 2)
+            1
+          (* n (fact (- n 1)))))
+
+(dotimes (i 10000)
+  (fact 100))
+
diff --git a/benchmark/other-lang/fib.pl b/benchmark/other-lang/fib.pl
new file mode 100644
index 0000000000..a46f666d1e
--- /dev/null
+++ b/benchmark/other-lang/fib.pl
@@ -0,0 +1,11 @@
+sub fib{
+  my $n = $_[0];
+  if($n < 3){
+    return 1;
+  }
+  else{
+    return fib($n-1) + fib($n-2);
+  }
+};
+
+&fib(34);
diff --git a/benchmark/other-lang/fib.py b/benchmark/other-lang/fib.py
new file mode 100644
index 0000000000..45f2bceb8d
--- /dev/null
+++ b/benchmark/other-lang/fib.py
@@ -0,0 +1,7 @@
+def fib(n):
+  if n < 3:
+    return 1
+  else:
+    return fib(n-1) + fib(n-2)
+
+fib(34)
diff --git a/benchmark/other-lang/fib.rb b/benchmark/other-lang/fib.rb
new file mode 100644
index 0000000000..ec587eabe0
--- /dev/null
+++ b/benchmark/other-lang/fib.rb
@@ -0,0 +1,9 @@
+def fib n
+  if n < 3
+    1
+  else
+    fib(n-1) + fib(n-2)
+  end
+end
+
+fib(34)
diff --git a/benchmark/other-lang/fib.scm b/benchmark/other-lang/fib.scm
new file mode 100644
index 0000000000..2fc4e225bd
--- /dev/null
+++ b/benchmark/other-lang/fib.scm
@@ -0,0 +1,7 @@
+(define (fib n)
+        (if (< n 3)
+            1
+          (+ (fib (- n 1)) (fib (- n 2)))))
+
+(fib 34)
+
diff --git a/benchmark/other-lang/loop.pl b/benchmark/other-lang/loop.pl
new file mode 100644
index 0000000000..2777490aaa
--- /dev/null
+++ b/benchmark/other-lang/loop.pl
@@ -0,0 +1,3 @@
+for($i=0; $i<30000000; $i++){
+}
+
diff --git a/benchmark/other-lang/loop.py b/benchmark/other-lang/loop.py
new file mode 100644
index 0000000000..003749bf3a
--- /dev/null
+++ b/benchmark/other-lang/loop.py
@@ -0,0 +1,2 @@
+for i in xrange(30000000):
+	pass
diff --git a/benchmark/other-lang/loop.rb b/benchmark/other-lang/loop.rb
new file mode 100644
index 0000000000..b367b9dbf3
--- /dev/null
+++ b/benchmark/other-lang/loop.rb
@@ -0,0 +1,4 @@
+i = 0
+while i<30000000
+  i += 1
+end
diff --git a/benchmark/other-lang/loop.scm b/benchmark/other-lang/loop.scm
new file mode 100644
index 0000000000..3364f7e679
--- /dev/null
+++ b/benchmark/other-lang/loop.scm
@@ -0,0 +1 @@
+(dotimes (x 30000000))
diff --git a/benchmark/other-lang/loop2.rb b/benchmark/other-lang/loop2.rb
new file mode 100644
index 0000000000..df8fffc1ff
--- /dev/null
+++ b/benchmark/other-lang/loop2.rb
@@ -0,0 +1 @@
+30000000.times{}
diff --git a/benchmark/other-lang/tak.pl b/benchmark/other-lang/tak.pl
new file mode 100644
index 0000000000..7e748a67c6
--- /dev/null
+++ b/benchmark/other-lang/tak.pl
@@ -0,0 +1,11 @@
+sub tak {
+    local($x, $y, $z) = @_;
+    if (!($y < $x)) {
+	return $z;
+    } else {
+	return &tak(&tak($x - 1, $y, $z),
+		    &tak($y - 1, $z, $x),
+		    &tak($z - 1, $x, $y));
+    }
+}
+&tak(18, 9, 0);
diff --git a/benchmark/other-lang/tak.py b/benchmark/other-lang/tak.py
new file mode 100644
index 0000000000..04f3f6829c
--- /dev/null
+++ b/benchmark/other-lang/tak.py
@@ -0,0 +1,8 @@
+def tak(x, y, z):
+	if not(y<x):
+		return z
+	else:
+		return tak(tak(x-1, y, z),
+		           tak(y-1, z, x),
+		           tak(z-1, x, y))
+tak(18, 9, 0)
diff --git a/benchmark/other-lang/tak.rb b/benchmark/other-lang/tak.rb
new file mode 100644
index 0000000000..efe5380f4e
--- /dev/null
+++ b/benchmark/other-lang/tak.rb
@@ -0,0 +1,13 @@
+
+def tak x, y, z
+  unless y < x
+    z
+  else
+    tak( tak(x-1, y, z),
+         tak(y-1, z, x),
+         tak(z-1, x, y))
+  end
+end
+
+tak(18, 9, 0)
+
diff --git a/benchmark/other-lang/tak.scm b/benchmark/other-lang/tak.scm
new file mode 100644
index 0000000000..52a7629ee5
--- /dev/null
+++ b/benchmark/other-lang/tak.scm
@@ -0,0 +1,10 @@
+(define (tak x y z)
+        (if (not (< y x))
+            z
+          (tak (tak (- x 1) y z)
+               (tak (- y 1) z x)
+               (tak (- z 1) x y))))
+
+(tak 18 9 0)
+
+
diff --git a/benchmark/require.yml b/benchmark/require.yml
new file mode 100644
index 0000000000..711d8e11e9
--- /dev/null
+++ b/benchmark/require.yml
@@ -0,0 +1,36 @@
+prelude: |
+  require "fileutils"
+
+  def prepare
+    num_files = 10000
+
+    basename = File.dirname($0)
+    data_dir = File.join(basename, "bm_require.data")
+
+    # skip if all of files exists
+    if File.exist?(File.join(data_dir, "c#{num_files}.rb"))
+      return
+    end
+
+    FileUtils.mkdir_p(data_dir)
+
+    1.upto(num_files) do |i|
+      f = File.open("#{data_dir}/c#{i}.rb", "w")
+      f.puts <<-END
+        class C#{i}
+        end
+      END
+    end
+  end
+
+  prepare
+benchmark:
+  require: |
+    $:.push File.join(File.dirname(__FILE__), "bm_require.data")
+
+    1.upto(10000) do |i|
+      require "c#{i}"
+    end
+
+    $:.pop
+loop_count: 1
diff --git a/benchmark/require_thread.yml b/benchmark/require_thread.yml
new file mode 100644
index 0000000000..87e0ba888b
--- /dev/null
+++ b/benchmark/require_thread.yml
@@ -0,0 +1,44 @@
+prelude: |
+  require "fileutils"
+
+  def prepare
+    num_files = 10000
+
+    basename = File.dirname($0)
+    data_dir = File.join(basename, "bm_require.data")
+
+    # skip if all of files exists
+    if File.exist?(File.join(data_dir, "c#{num_files}.rb"))
+      return
+    end
+
+    FileUtils.mkdir_p(data_dir)
+
+    1.upto(num_files) do |i|
+      f = File.open("#{data_dir}/c#{i}.rb", "w")
+      f.puts <<-END
+        class C#{i}
+        end
+      END
+    end
+  end
+
+  prepare
+benchmark:
+  require_thread: |
+    $:.push File.join(File.dirname(__FILE__), "bm_require.data")
+
+    i=0
+    t = Thread.new do
+      while true
+        i = i+1 # dummy loop
+      end
+    end
+
+    1.upto(100) do |i|
+      require "c#{i}"
+    end
+
+    $:.pop
+    t.kill
+loop_count: 1
diff --git a/benchmark/securerandom.rb b/benchmark/securerandom.rb
new file mode 100644
index 0000000000..a082ea6d5b
--- /dev/null
+++ b/benchmark/securerandom.rb
@@ -0,0 +1,5 @@
+require "securerandom"
+
+20_0000.times do
+  SecureRandom.random_number(100)
+end
diff --git a/benchmark/so_ackermann.rb b/benchmark/so_ackermann.rb
new file mode 100644
index 0000000000..7db5be9050
--- /dev/null
+++ b/benchmark/so_ackermann.rb
@@ -0,0 +1,19 @@
+#!/usr/bin/ruby
+# -*- mode: ruby -*-
+# $Id: ackermann-ruby.code,v 1.4 2004/11/13 07:40:41 bfulgham Exp $
+# http://www.bagley.org/~doug/shootout/
+
+def ack(m, n)
+    if m == 0 then
+        n + 1
+    elsif n == 0 then
+        ack(m - 1, 1)
+    else
+        ack(m - 1, ack(m, n - 1))
+    end
+end
+
+NUM = 9
+ack(3, NUM)
+
+
diff --git a/benchmark/so_array.rb b/benchmark/so_array.rb
new file mode 100644
index 0000000000..2b8fce8f99
--- /dev/null
+++ b/benchmark/so_array.rb
@@ -0,0 +1,23 @@
+#!/usr/bin/ruby
+# -*- mode: ruby -*-
+# $Id: ary-ruby.code,v 1.4 2004/11/13 07:41:27 bfulgham Exp $
+# http://www.bagley.org/~doug/shootout/
+# with help from Paul Brannan and Mark Hubbart
+
+n = 9000 # Integer(ARGV.shift || 1)
+
+x = Array.new(n)
+y = Array.new(n, 0)
+
+n.times{|bi|
+  x[bi] = bi + 1
+}
+
+(0 .. 999).each do |e|
+  (n-1).step(0,-1) do |bi|
+    y[bi] += x.at(bi)
+  end
+end
+# puts "#{y.first} #{y.last}"
+
+
diff --git a/benchmark/so_binary_trees.rb b/benchmark/so_binary_trees.rb
new file mode 100644
index 0000000000..b1693e4109
--- /dev/null
+++ b/benchmark/so_binary_trees.rb
@@ -0,0 +1,62 @@
+# The Computer Language Shootout Benchmarks
+# http://shootout.alioth.debian.org
+#
+# contributed by Jesse Millikan
+
+# disable output
+alias puts_orig puts
+def puts str
+  # disable puts
+end
+
+def item_check(tree)
+ if tree[0] == nil
+  tree[1]
+ else
+  tree[1] + item_check(tree[0]) - item_check(tree[2])
+ end
+end
+
+def bottom_up_tree(item, depth)
+ if depth > 0
+  item_item = 2 * item
+  depth -= 1
+  [bottom_up_tree(item_item - 1, depth), item, bottom_up_tree(item_item, depth)]
+ else
+  [nil, item, nil]
+ end
+end
+
+max_depth = 16 # ARGV[0].to_i
+min_depth = 4
+
+max_depth = min_depth + 2 if min_depth + 2 > max_depth
+
+stretch_depth = max_depth + 1
+stretch_tree = bottom_up_tree(0, stretch_depth)
+
+puts "stretch tree of depth #{stretch_depth}\t check: #{item_check(stretch_tree)}"
+stretch_tree = nil
+
+long_lived_tree = bottom_up_tree(0, max_depth)
+
+min_depth.step(max_depth + 1, 2) do |depth|
+ iterations = 2**(max_depth - depth + min_depth)
+
+ check = 0
+
+ for i in 1..iterations
+  temp_tree = bottom_up_tree(i, depth)
+  check += item_check(temp_tree)
+
+  temp_tree = bottom_up_tree(-i, depth)
+  check += item_check(temp_tree)
+ end
+
+ puts "#{iterations * 2}\t trees of depth #{depth}\t check: #{check}"
+end
+
+puts "long lived tree of depth #{max_depth}\t check: #{item_check(long_lived_tree)}"
+
+undef puts
+alias puts puts_orig
diff --git a/benchmark/so_concatenate.rb b/benchmark/so_concatenate.rb
new file mode 100644
index 0000000000..873214de7c
--- /dev/null
+++ b/benchmark/so_concatenate.rb
@@ -0,0 +1,18 @@
+#!/usr/bin/ruby
+# -*- mode: ruby -*-
+# $Id: strcat-ruby.code,v 1.4 2004/11/13 07:43:28 bfulgham Exp $
+# http://www.bagley.org/~doug/shootout/
+# based on code from Aristarkh A Zagorodnikov and Dat Nguyen
+
+STUFF = "hello\n"
+i = 0
+while i<10
+  i += 1
+  hello = ''
+  4_000_000.times do |e|
+    hello << STUFF
+  end
+end
+# puts hello.length
+
+
diff --git a/benchmark/so_count_words.yml b/benchmark/so_count_words.yml
new file mode 100644
index 0000000000..d0a6c8dd3e
--- /dev/null
+++ b/benchmark/so_count_words.yml
@@ -0,0 +1,66 @@
+prelude: |
+  #!/usr/bin/ruby
+  # -*- mode: ruby -*-
+
+  wc_input_base = <<EOS
+  Subject:      Re: Who was Izchak Miller?
+  From:         "Jane D. Anonymous" <nobody@yale.edu>
+  Date:         1996/04/28
+  Message-Id:   <4lv7bc$oh@news.ycc.yale.edu>
+  References:   <317C405E.5DFA@panix.com> <4lk6vl$gde@ns.oar.net>
+  To:           75176.2330@compuserve.com
+  Content-Type: text/plain; charset=us-ascii
+  Organization: Yale University
+  X-Url:        news:4lk6vl$gde@ns.oar.net
+  Mime-Version: 1.0
+  Newsgroups:   rec.games.roguelike.nethack
+  X-Mailer:     Mozilla 1.1N (Macintosh; I; 68K)
+
+  Hello there, Izchak Miller was my father.  When I was younger I spent 
+  many a night, hunched over the keyboard with a cup of tea, playing 
+  nethack with him and my brother.  my dad was a philosopher with a strong 
+  weakness for fantasy/sci fi.  I remember when he started to get involved 
+  with the Nethack team- my brother's Dungeons and Dragons monster book 
+  found a regular place beside my dad's desk. it's nice to see him living 
+  on in the game he loved so much :-).  
+                                                                    Tamar Miller
+
+  The following is a really long word of 5000 characters:
+
+  wwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwwww
+  EOS
+
+  # prepare 'wc.input'
+
+  def prepare_wc_input(wcbase)
+    wcinput = File.join(File.dirname($0), 'wc.input')
+    unless FileTest.exist?(wcinput)
+      data = wcbase.dup
+      13.times{
+        data << data
+      }
+      open(wcinput, 'w'){|f| f.write data}
+    end
+  end
+
+  prepare_wc_input(wc_input_base)
+
+benchmark:
+  so_count_words: |
+    # $Id: wc-ruby.code,v 1.4 2004/11/13 07:43:32 bfulgham Exp $
+    # http://www.bagley.org/~doug/shootout/
+    # with help from Paul Brannan
+    input = open(File.join(File.dirname($0), 'wc.input'), 'rb')
+
+    nl = nw = nc = 0
+    while true
+      tmp = input.read(4096) or break
+      data = tmp << (input.gets || "")
+      nc += data.length
+      nl += data.count("\n")
+      ((data.strip! || data).tr!("\n", " ") || data).squeeze!
+      nw += data.count(" ") + 1
+    end
+    # STDERR.puts "#{nl} #{nw} #{nc}"
+
+loop_count: 1
diff --git a/benchmark/so_exception.rb b/benchmark/so_exception.rb
new file mode 100644
index 0000000000..deb003a594
--- /dev/null
+++ b/benchmark/so_exception.rb
@@ -0,0 +1,61 @@
+#!/usr/bin/ruby
+# -*- mode: ruby -*-
+# $Id: except-ruby.code,v 1.4 2004/11/13 07:41:33 bfulgham Exp $
+# http://www.bagley.org/~doug/shootout/
+
+$HI = 0
+$LO = 0
+NUM = 250000 # Integer(ARGV[0] || 1)
+
+
+class Lo_Exception < Exception
+  def initialize(num)
+    @value = num
+  end
+end
+
+class Hi_Exception < Exception
+  def initialize(num)
+    @value = num
+  end
+end
+
+def some_function(num)
+  begin
+    hi_function(num)
+  rescue
+    print "We shouldn't get here, exception is: #{$!.type}\n"
+  end
+end
+
+def hi_function(num)
+  begin
+    lo_function(num)
+  rescue Hi_Exception
+    $HI = $HI + 1
+  end
+end
+
+def lo_function(num)
+  begin
+    blowup(num)
+  rescue Lo_Exception
+    $LO = $LO + 1
+  end
+end
+
+def blowup(num)
+  if num % 2 == 0
+    raise Lo_Exception.new(num)
+  else
+    raise Hi_Exception.new(num)
+  end
+end
+
+
+i = 1
+max = NUM+1
+while i < max
+  i += 1
+  some_function(i+1)
+end
diff --git a/benchmark/so_fannkuch.rb b/benchmark/so_fannkuch.rb
new file mode 100644
index 0000000000..bac5ecd44c
--- /dev/null
+++ b/benchmark/so_fannkuch.rb
@@ -0,0 +1,45 @@
+# The Computer Language Shootout
+# http://shootout.alioth.debian.org/
+# Contributed by Sokolov Yura
+# Modified by Ryan Williams
+
+def fannkuch(n)
+   maxFlips, m, r, check = 0, n-1, n, 0
+   count = (1..n).to_a
+   perm = (1..n).to_a
+
+   while true
+      if check < 30
+         puts "#{perm}"
+         check += 1
+      end
+
+      while r != 1
+         count[r-1] = r
+         r -= 1
+      end
+
+      if perm[0] != 1 and perm[m] != n
+         perml = perm.clone #.dup
+         flips = 0
+         while (k = perml.first ) != 1
+            perml = perml.slice!(0, k).reverse + perml
+            flips += 1
+         end
+         maxFlips = flips if flips > maxFlips
+      end
+      while true
+         if r==n then return maxFlips end
+         perm.insert r,perm.shift
+         break if (count[r] -= 1) > 0
+         r += 1
+      end
+   end
+end
+
+def puts *args
+end
+
+N = 9 # (ARGV[0] || 1).to_i
+puts "Pfannkuchen(#{N}) = #{fannkuch(N)}"
+
diff --git a/benchmark/so_fasta.rb b/benchmark/so_fasta.rb
new file mode 100644
index 0000000000..dcc6b39507
--- /dev/null
+++ b/benchmark/so_fasta.rb
@@ -0,0 +1,81 @@
+# The Computer Language Shootout
+# http://shootout.alioth.debian.org/
+# Contributed by Sokolov Yura
+
+$last = 42.0
+def gen_random(max, im=139968, ia=3877, ic=29573)
+    (max * ($last = ($last * ia + ic) % im)) / im
+end
+
+alu =
+   "GGCCGGGCGCGGTGGCTCACGCCTGTAATCCCAGCACTTTGG"+
+   "GAGGCCGAGGCGGGCGGATCACCTGAGGTCAGGAGTTCGAGA"+
+   "CCAGCCTGGCCAACATGGTGAAACCCCGTCTCTACTAAAAAT"+
+   "ACAAAAATTAGCCGGGCGTGGTGGCGCGCGCCTGTAATCCCA"+
+   "GCTACTCGGGAGGCTGAGGCAGGAGAATCGCTTGAACCCGGG"+
+   "AGGCGGAGGTTGCAGTGAGCCGAGATCGCGCCACTGCACTCC"+
+   "AGCCTGGGCGACAGAGCGAGACTCCGTCTCAAAAA"
+
+iub = [
+    ["a", 0.27],
+    ["c", 0.12],
+    ["g", 0.12],
+    ["t", 0.27],
+
+    ["B", 0.02],
+    ["D", 0.02],
+    ["H", 0.02],
+    ["K", 0.02],
+    ["M", 0.02],
+    ["N", 0.02],
+    ["R", 0.02],
+    ["S", 0.02],
+    ["V", 0.02],
+    ["W", 0.02],
+    ["Y", 0.02],
+]
+homosapiens = [
+    ["a", 0.3029549426680],
+    ["c", 0.1979883004921],
+    ["g", 0.1975473066391],
+    ["t", 0.3015094502008],
+]
+
+def make_repeat_fasta(id, desc, src, n)
+    puts ">#{id} #{desc}"
+    v = nil
+    width = 60
+    l = src.length
+    s = src * ((n / l) + 1)
+    s.slice!(n, l)
+    puts(s.scan(/.{1,#{width}}/).join("\n"))
+end
+
+def make_random_fasta(id, desc, table, n)
+    puts ">#{id} #{desc}"
+    rand, v = nil,nil
+    width = 60
+    chunk = 1 * width
+    prob = 0.0
+    table.each{|v| v[1]= (prob += v[1])}
+    for i in 1..(n/width)
+        puts((1..width).collect{
+            rand = gen_random(1.0)
+            table.find{|v| v[1]>rand}[0]
+        }.join)
+    end
+    if n%width != 0
+        puts((1..(n%width)).collect{
+            rand = gen_random(1.0)
+            table.find{|v| v[1]>rand}[0]
+        }.join)
+    end
+end
+
+
+n = (ARGV[0] or 250_000).to_i
+
+make_repeat_fasta('ONE', 'Homo sapiens alu', alu, n*2)
+make_random_fasta('TWO', 'IUB ambiguity codes', iub, n*3)
+make_random_fasta('THREE', 'Homo sapiens frequency', homosapiens, n*5)
+
diff --git a/benchmark/so_k_nucleotide.yml b/benchmark/so_k_nucleotide.yml
new file mode 100644
index 0000000000..d7df086c39
--- /dev/null
+++ b/benchmark/so_k_nucleotide.yml
@@ -0,0 +1,155 @@
+prelude: |
+  bm_so_fasta = <<'EOS'
+  # The Computer Language Shootout
+  # http://shootout.alioth.debian.org/
+  # Contributed by Sokolov Yura
+
+  $last = 42.0
+  def gen_random(max, im=139968, ia=3877, ic=29573)
+      (max * ($last = ($last * ia + ic) % im)) / im
+  end
+
+  alu =
+     "GGCCGGGCGCGGTGGCTCACGCCTGTAATCCCAGCACTTTGG"+
+     "GAGGCCGAGGCGGGCGGATCACCTGAGGTCAGGAGTTCGAGA"+
+     "CCAGCCTGGCCAACATGGTGAAACCCCGTCTCTACTAAAAAT"+
+     "ACAAAAATTAGCCGGGCGTGGTGGCGCGCGCCTGTAATCCCA"+
+     "GCTACTCGGGAGGCTGAGGCAGGAGAATCGCTTGAACCCGGG"+
+     "AGGCGGAGGTTGCAGTGAGCCGAGATCGCGCCACTGCACTCC"+
+     "AGCCTGGGCGACAGAGCGAGACTCCGTCTCAAAAA"
+
+  iub = [
+      ["a", 0.27],
+      ["c", 0.12],
+      ["g", 0.12],
+      ["t", 0.27],
+
+      ["B", 0.02],
+      ["D", 0.02],
+      ["H", 0.02],
+      ["K", 0.02],
+      ["M", 0.02],
+      ["N", 0.02],
+      ["R", 0.02],
+      ["S", 0.02],
+      ["V", 0.02],
+      ["W", 0.02],
+      ["Y", 0.02],
+  ]
+  homosapiens = [
+      ["a", 0.3029549426680],
+      ["c", 0.1979883004921],
+      ["g", 0.1975473066391],
+      ["t", 0.3015094502008],
+  ]
+
+  def make_repeat_fasta(id, desc, src, n)
+      puts ">#{id} #{desc}"
+      v = nil
+      width = 60
+      l = src.length
+      s = src * ((n / l) + 1)
+      s.slice!(n, l)
+      puts(s.scan(/.{1,#{width}}/).join("\n"))
+  end
+
+  def make_random_fasta(id, desc, table, n)
+      puts ">#{id} #{desc}"
+      rand, v = nil,nil
+      width = 60
+      chunk = 1 * width
+      prob = 0.0
+      table.each{|v| v[1]= (prob += v[1])}
+      for i in 1..(n/width)
+          puts((1..width).collect{
+              rand = gen_random(1.0)
+              table.find{|v| v[1]>rand}[0]
+          }.join)
+      end
+      if n%width != 0
+          puts((1..(n%width)).collect{
+              rand = gen_random(1.0)
+              table.find{|v| v[1]>rand}[0]
+          }.join)
+      end
+  end
+
+
+  n = (ARGV[0] or 250_000).to_i
+
+  make_repeat_fasta('ONE', 'Homo sapiens alu', alu, n*2)
+  make_random_fasta('TWO', 'IUB ambiguity codes', iub, n*3)
+  make_random_fasta('THREE', 'Homo sapiens frequency', homosapiens, n*5)
+  EOS
+benchmark:
+  - name: so_k_nucleotide
+    prelude: |
+      script = File.join(File.dirname($0), 'bm_so_fasta.rb')
+      File.write(script, bm_so_fasta)
+
+      def prepare_fasta_output n
+        filebase = File.join(File.dirname($0), 'fasta.output')
+        script = File.join(File.dirname($0), 'bm_so_fasta.rb')
+        file = "#{filebase}.#{n}"
+
+        unless FileTest.exist?(file)
+          STDERR.puts "preparing #{file}"
+
+          open(file, 'w'){|f|
+            ARGV[0] = n
+            $stdout = f
+            load script
+            $stdout = STDOUT
+          }
+        end
+      end
+      prepare_fasta_output(100_000)
+    script: |
+      # The Computer Language Shootout
+      # http://shootout.alioth.debian.org
+      #
+      # contributed by jose fco. gonzalez
+      # modified by Sokolov Yura
+
+      seq = String.new
+
+      def frecuency( seq,length )
+          n, table = seq.length - length + 1, Hash.new(0)
+          f, i = nil, nil
+          (0 ... length).each do |f|
+              (f ... n).step(length) do |i|
+                  table[seq[i,length]] += 1
+              end
+          end
+          [n,table]
+
+      end
+
+      def sort_by_freq( seq,length )
+          n,table = frecuency( seq,length )
+          a, b, v = nil, nil, nil
+          table.sort{|a,b| b[1] <=> a[1]}.each do |v|
+              puts "%s %.3f" % [v[0].upcase,((v[1]*100).to_f/n)]
+          end
+          puts
+      end
+
+      def find_seq( seq,s )
+          n,table = frecuency( seq,s.length )
+          puts "#{table[s].to_s}\t#{s.upcase}"
+      end
+
+      input = open(File.join(File.dirname($0), 'fasta.output.100000'), 'rb')
+
+      line = input.gets while line !~ /^>THREE/
+      line = input.gets
+
+      while (line !~ /^>/) & line do
+          seq << line.chomp
+          line = input.gets
+      end
+
+      [1,2].each {|i| sort_by_freq( seq,i ) }
+
+      %w(ggt ggta ggtatt ggtattttaatt ggtattttaatttatagt).each{|s| find_seq( seq,s) }
+    loop_count: 1
diff --git a/benchmark/so_lists.rb b/benchmark/so_lists.rb
new file mode 100644
index 0000000000..e8f4a2a5f7
--- /dev/null
+++ b/benchmark/so_lists.rb
@@ -0,0 +1,47 @@
+#from http://www.bagley.org/~doug/shootout/bench/lists/lists.ruby
+
+NUM = 300
+SIZE = 10000
+
+def test_lists()
+  # create a list of integers (Li1) from 1 to SIZE
+  li1 = (1..SIZE).to_a
+  # copy the list to li2 (not by individual items)
+  li2 = li1.dup
+  # remove each individual item from left side of li2 and
+  # append to right side of li3 (preserving order)
+  li3 = Array.new
+  while (not li2.empty?)
+    li3.push(li2.shift)
+  end
+  # li2 must now be empty
+  # remove each individual item from right side of li3 and
+  # append to right side of li2 (reversing list)
+  while (not li3.empty?)
+    li2.push(li3.pop)
+  end
+  # li3 must now be empty
+  # reverse li1 in place
+  li1.reverse!
+  # check that first item is now SIZE
+  if li1[0] != SIZE then
+    p "not SIZE"
+    0
+  else
+    # compare li1 and li2 for equality
+    if li1 != li2 then
+      return(0)
+    else
+      # return the length of the list
+      li1.length
+    end
+  end
+end
+
+i = 0
+while i<NUM
+  i += 1
+  result = test_lists()
+end
+
+result
diff --git a/benchmark/so_mandelbrot.rb b/benchmark/so_mandelbrot.rb
new file mode 100644
index 0000000000..76331c64b8
--- /dev/null
+++ b/benchmark/so_mandelbrot.rb
@@ -0,0 +1,57 @@
+#  The Computer Language Benchmarks Game
+#  http://shootout.alioth.debian.org/
+#
+#  contributed by Karl von Laudermann
+#  modified by Jeremy Echols
+
+size = 600 # ARGV[0].to_i
+
+puts "P4\n#{size} #{size}"
+
+ITER = 49                           # Iterations - 1 for easy for..in looping
+LIMIT_SQUARED = 4.0                 # Presquared limit
+
+byte_acc = 0
+bit_num = 0
+
+count_size = size - 1               # Precomputed size for easy for..in looping
+
+# For..in loops are faster than .upto, .downto, .times, etc.
+for y in 0..count_size
+  for x in 0..count_size
+    zr = 0.0
+    zi = 0.0
+    cr = (2.0*x/size)-1.5
+    ci = (2.0*y/size)-1.0
+    escape = false
+
+    # To make use of the for..in code, we use a dummy variable,
+    # like one would in C
+    for dummy in 0..ITER
+      tr = zr*zr - zi*zi + cr
+      ti = 2*zr*zi + ci
+      zr, zi = tr, ti
+
+      if (zr*zr+zi*zi) > LIMIT_SQUARED
+        escape = true
+        break
+      end
+    end
+
+    byte_acc = (byte_acc << 1) | (escape ? 0b0 : 0b1)
+    bit_num += 1
+
+    # Code is very similar for these cases, but using separate blocks
+    # ensures we skip the shifting when it's unnecessary, which is most cases.
+    if (bit_num == 8)
+      print byte_acc.chr
+      byte_acc = 0
+      bit_num = 0
+    elsif (x == count_size)
+      byte_acc <<= (8 - bit_num)
+      print byte_acc.chr
+      byte_acc = 0
+      bit_num = 0
+    end
+  end
+end
diff --git a/benchmark/so_matrix.rb b/benchmark/so_matrix.rb
new file mode 100644
index 0000000000..e2c5c8e559
--- /dev/null
+++ b/benchmark/so_matrix.rb
@@ -0,0 +1,48 @@
+#!/usr/bin/ruby
+# -*- mode: ruby -*-
+# $Id: matrix-ruby.code,v 1.4 2004/11/13 07:42:14 bfulgham Exp $
+# http://www.bagley.org/~doug/shootout/
+
+n = 60 #Integer(ARGV.shift || 1)
+
+size = 40
+
+def mkmatrix(rows, cols)
+    count = 1
+    mx = Array.new(rows)
+    (0 .. (rows - 1)).each do |bi|
+        row = Array.new(cols, 0)
+        (0 .. (cols - 1)).each do |j|
+            row[j] = count
+            count += 1
+        end
+        mx[bi] = row
+    end
+    mx
+end
+
+def mmult(rows, cols, m1, m2)
+    m3 = Array.new(rows)
+    (0 .. (rows - 1)).each do |bi|
+        row = Array.new(cols, 0)
+        (0 .. (cols - 1)).each do |j|
+            val = 0
+            (0 .. (cols - 1)).each do |k|
+                val += m1.at(bi).at(k) * m2.at(k).at(j)
+            end
+            row[j] = val
+        end
+        m3[bi] = row
+    end
+    m3
+end
+
+m1 = mkmatrix(size, size)
+m2 = mkmatrix(size, size)
+mm = Array.new
+n.times do
+    mm = mmult(size, size, m1, m2)
+end
+# puts "#{mm[0][0]} #{mm[2][3]} #{mm[3][2]} #{mm[4][4]}"
+
+
diff --git a/benchmark/so_meteor_contest.rb b/benchmark/so_meteor_contest.rb
new file mode 100644
index 0000000000..8c136baa6c
--- /dev/null
+++ b/benchmark/so_meteor_contest.rb
@@ -0,0 +1,563 @@
+#!/usr/bin/env ruby
+#
+# The Computer Language Shootout
+#   http://shootout.alioth.debian.org
+#   contributed by Kevin Barnes (Ruby novice)
+
+# PROGRAM:  the main body is at the bottom.
+#   1) read about the problem here: http://www-128.ibm.com/developerworks/java/library/j-javaopt/
+#   2) see how I represent a board as a bitmask by reading the blank_board comments
+#   3) read as your mental paths take you
+
+def print *args
+end
+
+# class to represent all information about a particular rotation of a particular piece
+class Rotation
+  # an array (by location) containing a bit mask for how the piece maps at the given location.
+  # if the rotation is invalid at that location the mask will contain false
+  attr_reader :start_masks
+
+  # maps a direction to a relative location.  these differ depending on whether it is an even or
+  # odd row being mapped from
+  @@rotation_even_adder = { :west => -1, :east => 1, :nw => -7, :ne => -6, :sw => 5, :se => 6 }
+  @@rotation_odd_adder = { :west => -1, :east => 1, :nw => -6, :ne => -5, :sw => 6, :se => 7 }
+
+  def initialize( directions )
+    @even_offsets, @odd_offsets = normalize_offsets( get_values( directions ))
+
+    @even_mask = mask_for_offsets( @even_offsets)
+    @odd_mask = mask_for_offsets( @odd_offsets)
+
+    @start_masks = Array.new(60)
+
+    # create the rotational masks by placing the base mask at the location and seeing if
+    # 1) it overlaps the boundaries and 2) it produces a prunable board.  if either of these
+    # is true the piece cannot be placed
+    0.upto(59) do | offset |
+      mask = is_even(offset) ? (@even_mask << offset) : (@odd_mask << offset)
+      if (blank_board & mask == 0 && !prunable(blank_board | mask, 0, true)) then
+        imask = compute_required( mask, offset)
+        @start_masks[offset] = [ mask, imask, imask | mask ]
+      else
+        @start_masks[offset] = false
+      end
+    end
+  end
+
+  def compute_required( mask, offset )
+    board = blank_board
+    0.upto(offset) { | i | board |= 1 << i }
+    board |= mask
+    return 0 if (!prunable(board | mask, offset))
+    board = flood_fill(board,58)
+    count = 0
+    imask = 0
+    0.upto(59) do | i |
+      if (board[i] == 0) then
+        imask |= (1 << i)
+        count += 1
+      end
+    end
+    (count > 0 && count < 5) ? imask : 0
+  end
+
+  def flood_fill( board, location)
+    return board if (board[location] == 1)
+    board |= 1 << location
+    row, col = location.divmod(6)
+    board = flood_fill( board, location - 1) if (col > 0)
+    board = flood_fill( board, location + 1) if (col < 4)
+    if (row % 2 == 0) then
+      board = flood_fill( board, location - 7) if (col > 0 && row > 0)
+      board = flood_fill( board, location - 6) if (row > 0)
+      board = flood_fill( board, location + 6) if (row < 9)
+      board = flood_fill( board, location + 5) if (col > 0 && row < 9)
+    else
+      board = flood_fill( board, location - 5) if (col < 4 && row > 0)
+      board = flood_fill( board, location - 6) if (row > 0)
+      board = flood_fill( board, location + 6) if (row < 9)
+      board = flood_fill( board, location + 7) if (col < 4 && row < 9)
+    end
+    board
+  end
+
+  # given a location, produces a list of relative locations covered by the piece at this rotation
+  def offsets( location)
+    if is_even( location) then
+      @even_offsets.collect { | value | value + location }
+    else
+      @odd_offsets.collect { | value | value + location }
+    end
+  end
+
+  # returns a set of offsets relative to the top-left most piece of the rotation (by even or odd rows)
+  # this is hard to explain. imagine we have this partial board:
+  #   0 0 0 0 0 x        [positions 0-5]
+  #    0 0 1 1 0 x       [positions 6-11]
+  #   0 0 1 0 0 x        [positions 12-17]
+  #    0 1 0 0 0 x       [positions 18-23]
+  #   0 1 0 0 0 x        [positions 24-29]
+  #    0 0 0 0 0 x       [positions 30-35]
+  #       ...
+  # The top-left of the piece is at position 8, the
+  # board would be passed as a set of positions (values array) containing [8,9,14,19,25] not necessarily in that
+  # sorted order.  Since that array starts on an odd row, the offsets for an odd row are: [0,1,6,11,17] obtained
+  # by subtracting 8 from everything.  Now imagine the piece shifted up and to the right so it's on an even row:
+  #   0 0 0 1 1 x        [positions 0-5]
+  #    0 0 1 0 0 x       [positions 6-11]
+  #   0 0 1 0 0 x        [positions 12-17]
+  #    0 1 0 0 0 x       [positions 18-23]
+  #   0 0 0 0 0 x        [positions 24-29]
+  #    0 0 0 0 0 x       [positions 30-35]
+  #       ...
+  # Now the positions are [3,4,8,14,19] which after subtracting the lowest value (3) gives [0,1,5,11,16] thus, the
+  # offsets for this particular piece are (in even, odd order) [0,1,5,11,16],[0,1,6,11,17] which is what
+  # this function would return
+  def normalize_offsets( values)
+    min = values.min
+    even_min = is_even(min)
+    other_min = even_min ? min + 6 : min + 7
+    other_values = values.collect do | value |
+      if is_even(value) then
+        value + 6 - other_min
+      else
+        value + 7 - other_min
+      end
+    end
+    values.collect! { | value | value - min }
+
+    if even_min then
+      [values, other_values]
+    else
+      [other_values, values]
+    end
+  end
+
+  # produce a bitmask representation of an array of offset locations
+  def mask_for_offsets( offsets )
+    mask = 0
+    offsets.each { | value | mask = mask + ( 1 << value ) }
+    mask
+  end
+
+  # finds a "safe" position that a position as described by a list of directions can be placed
+  # without falling off any edge of the board.  the values returned a location to place the first piece
+  # at so it will fit after making the described moves
+  def start_adjust( directions )
+    south = east = 0;
+    directions.each do | direction |
+      east += 1 if ( direction == :sw || direction == :nw || direction == :west )
+      south += 1 if ( direction == :nw || direction == :ne )
+    end
+    south * 6 + east
+  end
+
+  # given a set of directions places the piece (as defined by a set of directions) on the board at
+  # a location that will not take it off the edge
+  def get_values( directions )
+    start = start_adjust(directions)
+    values = [ start ]
+    directions.each do | direction |
+      if (start % 12 >= 6) then
+        start += @@rotation_odd_adder[direction]
+      else
+        start += @@rotation_even_adder[direction]
+      end
+      values += [ start ]
+    end
+
+    # some moves take you back to an existing location, we'll strip duplicates
+    values.uniq
+  end
+end
+
+# describes a piece and caches information about its rotations to as to be efficient for iteration
+# ATTRIBUTES:
+#   rotations -- all the rotations of the piece
+#   type -- a numeic "name" of the piece
+#   masks -- an array by location of all legal rotational masks (a n inner array) for that location
+#   placed -- the mask that this piece was last placed at (not a location, but the actual mask used)
+class Piece
+  attr_reader :rotations, :type, :masks
+  attr_accessor :placed
+
+  # transform hashes that change one direction into another when you either flip or rotate a set of directions
+  @@flip_converter = { :west => :west, :east => :east, :nw => :sw, :ne => :se, :sw => :nw, :se => :ne }
+  @@rotate_converter = { :west => :nw, :east => :se, :nw => :ne, :ne => :east, :sw => :west, :se => :sw }
+
+  def initialize( directions, type )
+    @type = type
+    @rotations = Array.new();
+    @map = {}
+
+    generate_rotations( directions )
+    directions.collect! { | value | @@flip_converter[value] }
+    generate_rotations( directions )
+
+    # creates the masks AND a map that returns [location, rotation] for any given mask
+    # this is used when a board is found and we want to draw it, otherwise the map is unused
+    @masks = Array.new();
+    0.upto(59) do | i |
+      even = true
+      @masks[i] = @rotations.collect do | rotation |
+        mask = rotation.start_masks[i]
+        @map[mask[0]] = [ i, rotation ] if (mask)
+        mask || nil
+      end
+      @masks[i].compact!
+    end
+  end
+
+  # rotates a set of directions through all six angles and adds a Rotation to the list for each one
+  def generate_rotations( directions )
+    6.times do
+      rotations.push( Rotation.new(directions))
+      directions.collect! { | value | @@rotate_converter[value] }
+    end
+  end
+
+  # given a board string, adds this piece to the board at whatever location/rotation
+  # important: the outbound board string is 5 wide, the normal location notation is six wide (padded)
+  def fill_string( board_string)
+    location, rotation = @map[@placed]
+    rotation.offsets(location).each do | offset |
+      row, col = offset.divmod(6)
+      board_string[ row*5 + col, 1 ] = @type.to_s
+    end
+  end
+end
+
+# a blank bit board having this form:
+#
+#    0 0 0 0 0 1
+#     0 0 0 0 0 1
+#    0 0 0 0 0 1
+#     0 0 0 0 0 1
+#    0 0 0 0 0 1
+#     0 0 0 0 0 1
+#    0 0 0 0 0 1
+#     0 0 0 0 0 1
+#    0 0 0 0 0 1
+#     0 0 0 0 0 1
+#    1 1 1 1 1 1
+#
+# where left lest significant bit is the top left and the most significant is the lower right
+# the actual board only consists of the 0 places, the 1 places are blockers to keep things from running
+# off the edges or bottom
+def blank_board
+  0b111111100000100000100000100000100000100000100000100000100000100000
+end
+
+def full_board
+  0b111111111111111111111111111111111111111111111111111111111111111111
+end
+
+# determines if a location (bit position) is in an even row
+def is_even( location)
+  (location % 12) < 6
+end
+
+# support function that create three utility maps:
+#  $converter -- for each row an array that maps a five bit row (via array mapping)
+#                to the a five bit representation of the bits below it
+#  $bit_count -- maps a five bit row (via array mapping) to the number of 1s in the row
+#  @@new_regions -- maps a five bit row (via array mapping) to an array of "region" arrays
+#                   a region array has three values the first is a mask of bits in the region,
+#                   the second is the count of those bits and the third is identical to the first
+#                   examples:
+#                           0b10010 => [ 0b01100, 2, 0b01100 ], [ 0b00001, 1, 0b00001]
+#                           0b01010 => [ 0b10000, 1, 0b10000 ], [ 0b00100, 1, 0b00100 ], [ 0b00001, 1, 0b00001]
+#                           0b10001 => [ 0b01110, 3, 0b01110 ]
+def create_collector_support
+  odd_map = [0b11, 0b110, 0b1100, 0b11000, 0b10000]
+  even_map = [0b1, 0b11, 0b110, 0b1100, 0b11000]
+
+  all_odds = Array.new(0b100000)
+  all_evens = Array.new(0b100000)
+  bit_counts = Array.new(0b100000)
+  new_regions = Array.new(0b100000)
+  0.upto(0b11111) do | i |
+    bit_count = odd = even = 0
+    0.upto(4) do | bit |
+      if (i[bit] == 1) then
+        bit_count += 1
+        odd |= odd_map[bit]
+        even |= even_map[bit]
+      end
+    end
+    all_odds[i] = odd
+    all_evens[i] = even
+    bit_counts[i] = bit_count
+    new_regions[i] = create_regions( i)
+  end
+
+  $converter = []
+  10.times { | row | $converter.push((row % 2 == 0) ? all_evens : all_odds) }
+  $bit_counts = bit_counts
+  $regions = new_regions.collect { | set | set.collect { | value | [ value, bit_counts[value], value] } }
+end
+
+# determines if a board is punable, meaning that there is no possibility that it
+# can be filled up with pieces.  A board is prunable if there is a grouping of unfilled spaces
+# that are not a multiple of five.  The following board is an example of a prunable board:
+#    0 0 1 0 0
+#     0 1 0 0 0
+#    1 1 0 0 0
+#     0 1 0 0 0
+#    0 0 0 0 0
+#       ...
+#
+# This board is prunable because the top left corner is only 3 bits in area, no piece will ever fit it
+# parameters:
+#   board -- an initial bit board (6 bit padded rows, see blank_board for format)
+#   location -- starting location, everything above and to the left is already full
+#   slotting -- set to true only when testing initial pieces, when filling normally
+#               additional assumptions are possible
+#
+# Algorithm:
+#    The algorithm starts at the top row (as determined by location) and iterates a row at a time
+#    maintainng counts of active open areas (kept in the collector array) each collector contains
+#    three values at the start of an iteration:
+#          0: mask of bits that would be adjacent to the collector in this row
+#          1: the number of bits collected so far
+#          2: a scratch space starting as zero, but used during the computation to represent
+#             the empty bits in the new row that are adjacent (position 0)
+#  The exact procedure is described in-code
+def prunable( board, location, slotting = false)
+  collectors = []
+  # loop across the rows
+  (location / 6).to_i.upto(9) do | row_on |
+    # obtain a set of regions representing the bits of the current row.
+    regions = $regions[(board >> (row_on * 6)) & 0b11111]
+    converter = $converter[row_on]
+
+    # track the number of collectors at the start of the cycle so that
+    # we don't compute against newly created collectors, only existing collectors
+    initial_collector_count = collectors.length
+
+    # loop against the regions.  For each region of the row
+    # we will see if it connects to one or more existing collectors.
+    # if it connects to 1 collector, the bits from the region are added to the
+    # bits of the collector and the mask is placed in collector[2]
+    # If the region overlaps more than one collector then all the collectors
+    # it overlaps with are merged into the first one (the others are set to nil in the array)
+    # if NO collectors are found then the region is copied as a new collector
+    regions.each do | region |
+      collector_found = nil
+      region_mask = region[2]
+      initial_collector_count.times do | collector_num |
+        collector = collectors[collector_num]
+        if (collector) then
+          collector_mask = collector[0]
+          if (collector_mask & region_mask != 0) then
+            if (collector_found) then
+              collector_found[0] |= collector_mask
+              collector_found[1] += collector[1]
+              collector_found[2] |= collector[2]
+              collectors[collector_num] = nil
+            else
+              collector_found = collector
+              collector[1] += region[1]
+              collector[2] |= region_mask
+            end
+          end
+        end
+      end
+      if (collector_found == nil) then
+        collectors.push(Array.new(region))
+      end
+    end
+
+    # check the existing collectors, if any collector overlapped no bits in the region its [2] value will
+    # be zero.  The size of any such reaason is tested if it is not a multiple of five true is returned since
+    # the board is prunable.  if it is a multiple of five it is removed.
+    # Collector that are still active have a new adjacent value [0] set based n the matched bits
+    # and have [2] cleared out for the next cycle.
+    collectors.length.times do | collector_num |
+      collector = collectors[collector_num]
+      if (collector) then
+        if (collector[2] == 0) then
+          return true if (collector[1] % 5 != 0)
+          collectors[collector_num] = nil
+        else
+          # if a collector matches all bits in the row then we can return unprunable early for the
+          # following reasons:
+          #    1) there can be no more unavailable bits bince we fill from the top left downward
+          #    2) all previous regions have been closed or joined so only this region can fail
+          #    3) this region must be good since there can never be only 1 region that is nuot
+          #       a multiple of five
+          # this rule only applies when filling normally, so we ignore the rule if we are "slotting"
+          # in pieces to see what configurations work for them (the only other time this algorithm is used).
+          return false if (collector[2] == 0b11111 && !slotting)
+          collector[0] = converter[collector[2]]
+          collector[2] = 0
+        end
+      end
+    end
+
+    # get rid of all the empty converters for the next round
+    collectors.compact!
+  end
+  return false if (collectors.length <= 1) # 1 collector or less and the region is fine
+  collectors.any? { | collector | (collector[1] % 5) != 0 } # more than 1 and we test them all for bad size
+end
+
+# creates a region given a row mask.  see prunable for what a "region" is
+def create_regions( value )
+  regions = []
+  cur_region = 0
+  5.times do | bit |
+    if (value[bit] == 0) then
+      cur_region |= 1 << bit
+    else
+      if (cur_region != 0 ) then
+        regions.push( cur_region)
+        cur_region = 0;
+      end
+    end
+  end
+  regions.push(cur_region) if (cur_region != 0)
+  regions
+end
+
+# find up to the counted number of solutions (or all solutions) and prints the final result
+def find_all
+  find_top( 1)
+  find_top( 0)
+  print_results
+end
+
+# show the board
+def print_results
+  print "#{@boards_found} solutions found\n\n"
+  print_full_board( @min_board)
+  print "\n"
+  print_full_board( @max_board)
+  print "\n"
+end
+
+# finds solutions.  This special version of the main function is only used for the top level
+# the reason for it is basically to force a particular ordering on how the rotations are tested for
+# the first piece.  It is called twice, first looking for placements of the odd rotations and then
+# looking for placements of the even locations.
+#
+# WHY?
+#   Since any found solution has an inverse we want to maximize finding solutions that are not already found
+#   as an inverse.  The inverse will ALWAYS be 3 one of the piece configurations that is exactly 3 rotations away
+#   (an odd number).  Checking even vs odd then produces a higher probability of finding more pieces earlier
+#   in the cycle.  We still need to keep checking all the permutations, but our probability of finding one will
+#   diminsh over time.  Since we are TOLD how many to search for this lets us exit before checking all pieces
+#   this bennifit is very great when seeking small numbers of solutions and is 0 when looking for more than the
+#   maximum number
+def find_top( rotation_skip)
+  board = blank_board
+  (@pieces.length-1).times do
+    piece = @pieces.shift
+    piece.masks[0].each do | mask, imask, cmask |
+      if ((rotation_skip += 1) % 2 == 0) then
+        piece.placed = mask
+        find( 1, 1, board | mask)
+      end
+    end
+    @pieces.push(piece)
+  end
+  piece = @pieces.shift
+  @pieces.push(piece)
+end
+
+# the normail find routine, iterates through the available pieces, checks all rotations at the current location
+# and adds any boards found.  depth is achieved via recursion.  the overall approach is described
+# here: http://www-128.ibm.com/developerworks/java/library/j-javaopt/
+# parameters:
+#  start_location -- where to start looking for place for the next piece at
+#  placed -- number of pieces placed
+#  board -- current state of the board
+#
+# see in-code comments
+def find( start_location, placed, board)
+  # find the next location to place a piece by looking for an empty bit
+  while board[start_location] == 1
+    start_location += 1
+  end
+
+  @pieces.length.times do
+    piece = @pieces.shift
+    piece.masks[start_location].each do | mask, imask, cmask |
+      if ( board & cmask == imask) then
+        piece.placed = mask
+        if (placed == 9) then
+          add_board
+        else
+          find( start_location + 1, placed + 1, board | mask)
+        end
+      end
+    end
+    @pieces.push(piece)
+  end
+end
+
+# print the board
+def print_full_board( board_string)
+  10.times do | row |
+    print " " if (row % 2 == 1)
+    5.times do | col |
+      print "#{board_string[row*5 + col,1]} "
+    end
+    print "\n"
+  end
+end
+
+# when a board is found we "draw it" into a string and then flip that string, adding both to
+# the list (hash) of solutions if they are unique.
+def add_board
+  board_string = "99999999999999999999999999999999999999999999999999"
+  @all_pieces.each {  | piece | piece.fill_string( board_string ) }
+  save( board_string)
+  save( board_string.reverse)
+end
+
+# adds a board string to the list (if new) and updates the current best/worst board
+def save( board_string)
+  if (@all_boards[board_string] == nil) then
+    @min_board = board_string if (board_string < @min_board)
+    @max_board = board_string if (board_string > @max_board)
+    @all_boards.store(board_string,true)
+    @boards_found += 1
+
+    # the exit motif is a time saver.  Ideally the function should return, but those tests
+    # take noticeable time (performance).
+    if (@boards_found == @stop_count) then
+      print_results
+      exit(0)
+    end
+  end
+end
+
+
+##
+## MAIN BODY :)
+##
+create_collector_support
+@pieces = [
+  Piece.new( [ :nw, :ne, :east, :east ], 2),
+  Piece.new( [ :ne, :se, :east, :ne ], 7),
+  Piece.new( [ :ne, :east, :ne, :nw ], 1),
+  Piece.new( [ :east, :sw, :sw, :se ], 6),
+  Piece.new( [ :east, :ne, :se, :ne ], 5),
+  Piece.new( [ :east, :east, :east, :se ], 0),
+  Piece.new( [ :ne, :nw, :se, :east, :se ], 4),
+  Piece.new( [ :se, :se, :se, :west ], 9),
+  Piece.new( [ :se, :se, :east, :se ], 8),
+  Piece.new( [ :east, :east, :sw, :se ], 3)
+  ];
+
+@all_pieces = Array.new( @pieces)
+
+@min_board = "99999999999999999999999999999999999999999999999999"
+@max_board = "00000000000000000000000000000000000000000000000000"
+@stop_count = ARGV[0].to_i || 2089
+@all_boards = {}
+@boards_found = 0
+
+find_all ######## DO IT!!!
diff --git a/benchmark/so_nbody.rb b/benchmark/so_nbody.rb
new file mode 100644
index 0000000000..d6c5bb9e61
--- /dev/null
+++ b/benchmark/so_nbody.rb
@@ -0,0 +1,148 @@
+# The Computer Language Shootout
+# http://shootout.alioth.debian.org
+#
+# Optimized for Ruby by Jesse Millikan
+# From version ported by Michael Neumann from the C gcc version,
+# which was written by Christoph Bauer.
+
+SOLAR_MASS = 4 * Math::PI**2
+DAYS_PER_YEAR = 365.24
+
+def _puts *args
+end
+
+class Planet
+ attr_accessor :x, :y, :z, :vx, :vy, :vz, :mass
+
+ def initialize(x, y, z, vx, vy, vz, mass)
+  @x, @y, @z = x, y, z
+  @vx, @vy, @vz = vx * DAYS_PER_YEAR, vy * DAYS_PER_YEAR, vz * DAYS_PER_YEAR
+  @mass = mass * SOLAR_MASS
+ end
+
+ def move_from_i(bodies, nbodies, dt, i)
+  while i < nbodies
+   b2 = bodies[i]
+   dx = @x - b2.x
+   dy = @y - b2.y
+   dz = @z - b2.z
+
+   distance = Math.sqrt(dx * dx + dy * dy + dz * dz)
+   mag = dt / (distance * distance * distance)
+   b_mass_mag, b2_mass_mag = @mass * mag, b2.mass * mag
+
+   @vx -= dx * b2_mass_mag
+   @vy -= dy * b2_mass_mag
+   @vz -= dz * b2_mass_mag
+   b2.vx += dx * b_mass_mag
+   b2.vy += dy * b_mass_mag
+   b2.vz += dz * b_mass_mag
+   i += 1
+  end
+
+  @x += dt * @vx
+  @y += dt * @vy
+  @z += dt * @vz
+ end
+end
+
+def energy(bodies)
+  e = 0.0
+  nbodies = bodies.size
+
+  for i in 0 ... nbodies
+    b = bodies[i]
+    e += 0.5 * b.mass * (b.vx * b.vx + b.vy * b.vy + b.vz * b.vz)
+    for j in (i + 1) ... nbodies
+      b2 = bodies[j]
+      dx = b.x - b2.x
+      dy = b.y - b2.y
+      dz = b.z - b2.z
+      distance = Math.sqrt(dx * dx + dy * dy + dz * dz)
+      e -= (b.mass * b2.mass) / distance
+    end
+  end
+  e
+end
+
+def offset_momentum(bodies)
+  px, py, pz = 0.0, 0.0, 0.0
+
+  for b in bodies
+    m = b.mass
+    px += b.vx * m
+    py += b.vy * m
+    pz += b.vz * m
+  end
+
+  b = bodies[0]
+  b.vx = - px / SOLAR_MASS
+  b.vy = - py / SOLAR_MASS
+  b.vz = - pz / SOLAR_MASS
+end
+
+BODIES = [
+  # sun
+  Planet.new(0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0),
+
+  # jupiter
+  Planet.new(
+    4.84143144246472090e+00,
+    -1.16032004402742839e+00,
+    -1.03622044471123109e-01,
+    1.66007664274403694e-03,
+    7.69901118419740425e-03,
+    -6.90460016972063023e-05,
+    9.54791938424326609e-04),
+
+  # saturn
+  Planet.new(
+    8.34336671824457987e+00,
+    4.12479856412430479e+00,
+    -4.03523417114321381e-01,
+    -2.76742510726862411e-03,
+    4.99852801234917238e-03,
+    2.30417297573763929e-05,
+    2.85885980666130812e-04),
+
+  # uranus
+  Planet.new(
+    1.28943695621391310e+01,
+    -1.51111514016986312e+01,
+    -2.23307578892655734e-01,
+    2.96460137564761618e-03,
+    2.37847173959480950e-03,
+    -2.96589568540237556e-05,
+    4.36624404335156298e-05),
+
+  # neptune
+  Planet.new(
+    1.53796971148509165e+01,
+    -2.59193146099879641e+01,
+    1.79258772950371181e-01,
+    2.68067772490389322e-03,
+    1.62824170038242295e-03,
+    -9.51592254519715870e-05,
+    5.15138902046611451e-05)
+]
+
+init = 200_000 # ARGV[0]
+n = Integer(init)
+
+offset_momentum(BODIES)
+
+puts "%.9f" % energy(BODIES)
+
+nbodies = BODIES.size
+dt = 0.01
+
+n.times do
+  i = 0
+  while i < nbodies
+    b = BODIES[i]
+    b.move_from_i(BODIES, nbodies, dt, i + 1)
+    i += 1
+  end
+end
+
+puts "%.9f" % energy(BODIES)
diff --git a/benchmark/so_nested_loop.rb b/benchmark/so_nested_loop.rb
new file mode 100644
index 0000000000..a0513f8c47
--- /dev/null
+++ b/benchmark/so_nested_loop.rb
@@ -0,0 +1,24 @@
+#!/usr/bin/ruby
+# -*- mode: ruby -*-
+# $Id: nestedloop-ruby.code,v 1.4 2004/11/13 07:42:22 bfulgham Exp $
+# http://www.bagley.org/~doug/shootout/
+# from Avi Bryant
+
+n = 16 # Integer(ARGV.shift || 1)
+x = 0
+n.times do
+    n.times do
+        n.times do
+            n.times do
+                n.times do
+                    n.times do
+                        x += 1
+                    end
+                end
+            end
+        end
+    end
+end
+# puts x
+
+
diff --git a/benchmark/so_nsieve.rb b/benchmark/so_nsieve.rb
new file mode 100644
index 0000000000..a65cc78233
--- /dev/null
+++ b/benchmark/so_nsieve.rb
@@ -0,0 +1,35 @@
+# The Computer Language Shootout
+# http://shootout.alioth.debian.org/
+#
+# contributed by Glenn Parker, March 2005
+# modified by Evan Phoenix, Sept 2006
+
+def sieve(m)
+  flags = Flags.dup[0,m]
+  count = 0
+  pmax = m - 1
+  p = 2
+  while p <= pmax
+    unless flags[p].zero?
+      count += 1
+      mult = p
+      while mult <= pmax
+        flags[mult] = 0
+        mult += p
+      end
+    end
+    p += 1
+  end
+  count
+end
+
+n = 9 # (ARGV[0] || 2).to_i
+Flags = ("\x1" * ( 2 ** n * 10_000)).unpack("c*")
+
+n.downto(n-2) do |exponent|
+  break if exponent < 0
+  m = (1 << exponent) * 10_000
+  # m = (2 ** exponent) * 10_000
+  count = sieve(m)
+  printf "Primes up to %8d %8d\n", m, count
+end
diff --git a/benchmark/so_nsieve_bits.rb b/benchmark/so_nsieve_bits.rb
new file mode 100644
index 0000000000..6f958ee44e
--- /dev/null
+++ b/benchmark/so_nsieve_bits.rb
@@ -0,0 +1,43 @@
+#!/usr/bin/ruby
+#coding: us-ascii
+#
+# The Great Computer Language Shootout
+# http://shootout.alioth.debian.org/
+#
+# nsieve-bits in Ruby
+# Contributed by Glenn Parker, March 2005
+
+CharExponent = 3
+BitsPerChar = 1 << CharExponent
+LowMask = BitsPerChar - 1
+
+def sieve(m)
+  items = "\xFF" * ((m / BitsPerChar) + 1)
+  masks = ""
+  BitsPerChar.times do |b|
+    masks << (1 << b).chr
+  end
+
+  count = 0
+  pmax = m - 1
+  2.step(pmax, 1) do |p|
+    if items[p >> CharExponent][p & LowMask] == 1
+      count += 1
+      p.step(pmax, p) do |mult|
+	a = mult >> CharExponent
+	b = mult & LowMask
+	items[a] -= masks[b] if items[a][b] != 0
+      end
+    end
+  end
+  count
+end
+
+n = 9 # (ARGV[0] || 2).to_i
+n.step(n - 2, -1) do |exponent|
+  break if exponent < 0
+  m = 2 ** exponent * 10_000
+  count = sieve(m)
+  printf "Primes up to %8d %8d\n", m, count
+end
+
diff --git a/benchmark/so_object.rb b/benchmark/so_object.rb
new file mode 100644
index 0000000000..e8607c7199
--- /dev/null
+++ b/benchmark/so_object.rb
@@ -0,0 +1,56 @@
+#!/usr/bin/ruby
+# -*- mode: ruby -*-
+# $Id: objinst-ruby.code,v 1.4 2004/11/13 07:42:25 bfulgham Exp $
+# http://www.bagley.org/~doug/shootout/
+# with help from Aristarkh Zagorodnikov
+
+class Toggle
+    def initialize(start_state)
+        @bool = start_state
+    end
+
+    def value
+        @bool
+    end
+
+    def activate
+        @bool = !@bool
+        self
+    end
+end
+
+class NthToggle < Toggle
+    def initialize(start_state, max_counter)
+        super start_state
+        @count_max = max_counter
+        @counter = 0
+    end
+
+    def activate
+        @counter += 1
+        if @counter >= @count_max
+            @bool = !@bool
+            @counter = 0
+        end
+        self
+    end
+end
+
+n = 1500000 # (ARGV.shift || 1).to_i
+
+toggle = Toggle.new 1
+5.times do
+    toggle.activate.value ? 'true' : 'false'
+end
+n.times do
+    toggle = Toggle.new 1
+end
+
+ntoggle = NthToggle.new 1, 3
+8.times do
+    ntoggle.activate.value ? 'true' : 'false'
+end
+n.times do
+    ntoggle = NthToggle.new 1, 3
+end
+
diff --git a/benchmark/so_partial_sums.rb b/benchmark/so_partial_sums.rb
new file mode 100644
index 0000000000..630b45cb8d
--- /dev/null
+++ b/benchmark/so_partial_sums.rb
@@ -0,0 +1,31 @@
+n = 2_500_000 # (ARGV.shift || 1).to_i
+
+alt = 1.0 ; s0 = s1 = s2 = s3 = s4 = s5 = s6 = s7 = s8 = 0.0
+
+1.upto(n) do |d|
+  d = d.to_f ; d2 = d * d ; d3 = d2 * d ; ds = Math.sin(d) ; dc = Math.cos(d)
+
+  s0 += (2.0 / 3.0) ** (d - 1.0)
+  s1 += 1.0 / Math.sqrt(d)
+  s2 += 1.0 / (d * (d + 1.0))
+  s3 += 1.0 / (d3 * ds * ds)
+  s4 += 1.0 / (d3 * dc * dc)
+  s5 += 1.0 / d
+  s6 += 1.0 / d2
+  s7 += alt / d
+  s8 += alt / (2.0 * d - 1.0)
+
+  alt = -alt
+end
+
+if false
+  printf("%.9f\t(2/3)^k\n", s0)
+  printf("%.9f\tk^-0.5\n", s1)
+  printf("%.9f\t1/k(k+1)\n", s2)
+  printf("%.9f\tFlint Hills\n", s3)
+  printf("%.9f\tCookson Hills\n", s4)
+  printf("%.9f\tHarmonic\n", s5)
+  printf("%.9f\tRiemann Zeta\n", s6)
+  printf("%.9f\tAlternating Harmonic\n", s7)
+  printf("%.9f\tGregory\n", s8)
+end
diff --git a/benchmark/so_pidigits.rb b/benchmark/so_pidigits.rb
new file mode 100644
index 0000000000..9a537b2d1c
--- /dev/null
+++ b/benchmark/so_pidigits.rb
@@ -0,0 +1,92 @@
+# The Great Computer Language Shootout
+# http://shootout.alioth.debian.org/
+#
+# contributed by Gabriele Renzi
+
+class PiDigitSpigot
+
+    def initialize()
+        @z = Transformation.new 1,0,0,1
+        @x = Transformation.new 0,0,0,0
+        @inverse = Transformation.new 0,0,0,0
+    end
+
+    def next!
+        @y = @z.extract(3)
+        if safe? @y
+            @z = produce(@y)
+            @y
+        else
+            @z = consume @x.next!()
+            next!()
+        end
+    end
+
+    def safe?(digit)
+        digit == @z.extract(4)
+    end
+
+    def produce(i)
+        @inverse.qrst(10,-10*i,0,1).compose(@z)
+    end
+
+    def consume(a)
+        @z.compose(a)
+    end
+end
+
+
+class Transformation
+    attr_reader :q, :r, :s, :t
+    def initialize(q, r, s, t)
+        @q,@r,@s,@t,@k = q,r,s,t,0
+    end
+
+    def next!()
+        @q = @k = @k + 1
+        @r = 4 * @k + 2
+        @s = 0
+        @t = 2 * @k + 1
+        self
+    end
+
+    def extract(j)
+        (@q * j + @r) / (@s * j + @t)
+    end
+
+    def compose(a)
+        self.class.new( @q * a.q,
+                        @q * a.r + r * a.t,
+                        @s * a.q + t * a.s,
+                        @s * a.r + t * a.t
+                    )
+    end
+
+    def qrst *args
+        initialize *args
+        self
+    end
+
+
+end
+
+
+WIDTH = 10
+n = 2_500 # Integer(ARGV[0])
+j = 0
+
+digits = PiDigitSpigot.new
+
+while n > 0
+    if n >= WIDTH
+        WIDTH.times {print digits.next!}
+        j += WIDTH
+    else
+        n.times {print digits.next!}
+        (WIDTH-n).times {print " "}
+        j += n
+    end
+    puts "\t:"+j.to_s
+    n -= WIDTH
+end
+
diff --git a/benchmark/so_random.rb b/benchmark/so_random.rb
new file mode 100644
index 0000000000..a66b9e8e63
--- /dev/null
+++ b/benchmark/so_random.rb
@@ -0,0 +1,20 @@
+# from http://www.bagley.org/~doug/shootout/bench/random/random.ruby
+
+IM = 139968.0
+IA = 3877.0
+IC = 29573.0
+
+$last = 42.0
+
+def gen_random(max)
+  (max * ($last = ($last * IA + IC) % IM)) / IM
+end
+
+N = 3_000_000
+
+i = 0
+while i<N
+  i +=1
+  gen_random(100.0)
+end
+# "%.9f" % gen_random(100.0)
diff --git a/benchmark/so_reverse_complement.yml b/benchmark/so_reverse_complement.yml
new file mode 100644
index 0000000000..de05eedfc4
--- /dev/null
+++ b/benchmark/so_reverse_complement.yml
@@ -0,0 +1,137 @@
+prelude: |
+  bm_so_fasta = <<'EOS'
+  # The Computer Language Shootout
+  # http://shootout.alioth.debian.org/
+  # Contributed by Sokolov Yura
+
+  $last = 42.0
+  def gen_random(max, im=139968, ia=3877, ic=29573)
+      (max * ($last = ($last * ia + ic) % im)) / im
+  end
+
+  alu =
+     "GGCCGGGCGCGGTGGCTCACGCCTGTAATCCCAGCACTTTGG"+
+     "GAGGCCGAGGCGGGCGGATCACCTGAGGTCAGGAGTTCGAGA"+
+     "CCAGCCTGGCCAACATGGTGAAACCCCGTCTCTACTAAAAAT"+
+     "ACAAAAATTAGCCGGGCGTGGTGGCGCGCGCCTGTAATCCCA"+
+     "GCTACTCGGGAGGCTGAGGCAGGAGAATCGCTTGAACCCGGG"+
+     "AGGCGGAGGTTGCAGTGAGCCGAGATCGCGCCACTGCACTCC"+
+     "AGCCTGGGCGACAGAGCGAGACTCCGTCTCAAAAA"
+
+  iub = [
+      ["a", 0.27],
+      ["c", 0.12],
+      ["g", 0.12],
+      ["t", 0.27],
+
+      ["B", 0.02],
+      ["D", 0.02],
+      ["H", 0.02],
+      ["K", 0.02],
+      ["M", 0.02],
+      ["N", 0.02],
+      ["R", 0.02],
+      ["S", 0.02],
+      ["V", 0.02],
+      ["W", 0.02],
+      ["Y", 0.02],
+  ]
+  homosapiens = [
+      ["a", 0.3029549426680],
+      ["c", 0.1979883004921],
+      ["g", 0.1975473066391],
+      ["t", 0.3015094502008],
+  ]
+
+  def make_repeat_fasta(id, desc, src, n)
+      puts ">#{id} #{desc}"
+      v = nil
+      width = 60
+      l = src.length
+      s = src * ((n / l) + 1)
+      s.slice!(n, l)
+      puts(s.scan(/.{1,#{width}}/).join("\n"))
+  end
+
+  def make_random_fasta(id, desc, table, n)
+      puts ">#{id} #{desc}"
+      rand, v = nil,nil
+      width = 60
+      chunk = 1 * width
+      prob = 0.0
+      table.each{|v| v[1]= (prob += v[1])}
+      for i in 1..(n/width)
+          puts((1..width).collect{
+              rand = gen_random(1.0)
+              table.find{|v| v[1]>rand}[0]
+          }.join)
+      end
+      if n%width != 0
+          puts((1..(n%width)).collect{
+              rand = gen_random(1.0)
+              table.find{|v| v[1]>rand}[0]
+          }.join)
+      end
+  end
+
+
+  n = (ARGV[0] or 250_000).to_i
+
+  make_repeat_fasta('ONE', 'Homo sapiens alu', alu, n*2)
+  make_random_fasta('TWO', 'IUB ambiguity codes', iub, n*3)
+  make_random_fasta('THREE', 'Homo sapiens frequency', homosapiens, n*5)
+  EOS
+benchmark:
+  - name: so_reverse_complement
+    prelude: |
+      script = File.join(File.dirname($0), 'bm_so_fasta.rb')
+      File.write(script, bm_so_fasta)
+
+      def prepare_fasta_output n
+        filebase = File.join(File.dirname($0), 'fasta.output')
+        script = File.join(File.dirname($0), 'bm_so_fasta.rb')
+        file = "#{filebase}.#{n}"
+
+        unless FileTest.exist?(file)
+          STDERR.puts "preparing #{file}"
+
+          open(file, 'w'){|f|
+            ARGV[0] = n
+            $stdout = f
+            load script
+            $stdout = STDOUT
+          }
+        end
+      end
+      prepare_fasta_output(2_500_000)
+    script: |
+      # The Great Computer Language Shootout
+      # http://shootout.alioth.debian.org/
+      #
+      # Contributed by Peter Bjarke Olsen
+      # Modified by Doug King
+
+      seq=Array.new
+
+      def revcomp(seq)
+        seq.reverse!.tr!('wsatugcyrkmbdhvnATUGCYRKMBDHVN','WSTAACGRYMKVHDBNTAACGRYMKVHDBN')
+        stringlen=seq.length
+        0.step(stringlen-1,60) {|x| print seq.slice(x,60) , "\n"}
+      end
+
+      input = open(File.join(File.dirname($0), 'fasta.output.2500000'), 'rb')
+
+      while input.gets
+        if $_ =~ />/
+          if seq.length != 0
+            revcomp(seq.join)
+            seq=Array.new
+          end
+          puts $_
+        else
+          $_.sub(/\n/,'')
+          seq.push $_
+        end
+      end
+      revcomp(seq.join)
+    loop_count: 1
diff --git a/benchmark/so_sieve.rb b/benchmark/so_sieve.rb
new file mode 100644
index 0000000000..43dc302648
--- /dev/null
+++ b/benchmark/so_sieve.rb
@@ -0,0 +1,24 @@
+# from http://www.bagley.org/~doug/shootout/bench/sieve/sieve.ruby
+num = 500
+count = i = j = 0
+flags0 = Array.new(8192,1)
+k = 0
+while k < num
+  k += 1
+  count = 0
+  flags = flags0.dup
+  i = 2
+  while i<8192
+    i += 1
+    if flags[i]
+      # remove all multiples of prime: i
+      j = i*i
+      while j < 8192
+        j += i
+        flags[j] = nil
+      end
+      count += 1
+    end
+  end
+end
+count
diff --git a/benchmark/so_spectralnorm.rb b/benchmark/so_spectralnorm.rb
new file mode 100644
index 0000000000..6b97206689
--- /dev/null
+++ b/benchmark/so_spectralnorm.rb
@@ -0,0 +1,50 @@
+# The Computer Language Shootout
+# http://shootout.alioth.debian.org/
+# Contributed by Sokolov Yura
+
+def eval_A(i,j)
+	return 1.0/((i+j)*(i+j+1)/2+i+1)
+end
+
+def eval_A_times_u(u)
+        v, i = nil, nil
+	(0..u.length-1).collect { |i|
+                v = 0
+		for j in 0..u.length-1
+			v += eval_A(i,j)*u[j]
+                end
+                v
+        }
+end
+
+def eval_At_times_u(u)
+	v, i = nil, nil
+	(0..u.length-1).collect{|i|
+                v = 0
+		for j in 0..u.length-1
+			v += eval_A(j,i)*u[j]
+                end
+                v
+        }
+end
+
+def eval_AtA_times_u(u)
+	return eval_At_times_u(eval_A_times_u(u))
+end
+
+n = 500 # ARGV[0].to_i
+
+u=[1]*n
+for i in 1..10
+        v=eval_AtA_times_u(u)
+        u=eval_AtA_times_u(v)
+end
+vBv=0
+vv=0
+for i in 0..n-1
+        vBv += u[i]*v[i]
+        vv += v[i]*v[i]
+end
+
+str = "%0.9f" % (Math.sqrt(vBv/vv)), "\n"
+# print str
diff --git a/benchmark/string_index.rb b/benchmark/string_index.rb
new file mode 100644
index 0000000000..7783111082
--- /dev/null
+++ b/benchmark/string_index.rb
@@ -0,0 +1,3 @@
+str1 = "あ" * 1024 + "い" # not single byte optimizable
+str2 = "い"
+100_000.times { str1.index(str2) }
diff --git a/benchmark/string_scan_re.rb b/benchmark/string_scan_re.rb
new file mode 100644
index 0000000000..b0d60201a9
--- /dev/null
+++ b/benchmark/string_scan_re.rb
@@ -0,0 +1,2 @@
+str = Array.new(1_000, 'abc').join(',')
+1_000.times { str.scan(/abc/) }
diff --git a/benchmark/string_scan_str.rb b/benchmark/string_scan_str.rb
new file mode 100644
index 0000000000..42440bd948
--- /dev/null
+++ b/benchmark/string_scan_str.rb
@@ -0,0 +1,2 @@
+str = Array.new(1_000, 'abc').join(',')
+1_000.times { str.scan('abc') }
diff --git a/benchmark/time_subsec.rb b/benchmark/time_subsec.rb
new file mode 100644
index 0000000000..505021c701
--- /dev/null
+++ b/benchmark/time_subsec.rb
@@ -0,0 +1,2 @@
+t = Time.now
+4000000.times { t.subsec }
diff --git a/benchmark/vm1_attr_ivar.yml b/benchmark/vm1_attr_ivar.yml
new file mode 100644
index 0000000000..f714dd9bd9
--- /dev/null
+++ b/benchmark/vm1_attr_ivar.yml
@@ -0,0 +1,14 @@
+prelude: |
+  class C
+    attr_reader :a, :b
+    def initialize
+      @a = nil
+      @b = nil
+    end
+  end
+  obj = C.new
+benchmark:
+  vm1_attr_ivar: |
+    j = obj.a
+    k = obj.b
+loop_count: 30000000
diff --git a/benchmark/vm1_attr_ivar_set.yml b/benchmark/vm1_attr_ivar_set.yml
new file mode 100644
index 0000000000..f383e59ef4
--- /dev/null
+++ b/benchmark/vm1_attr_ivar_set.yml
@@ -0,0 +1,14 @@
+prelude: |
+  class C
+    attr_accessor :a, :b
+    def initialize
+      @a = nil
+      @b = nil
+    end
+  end
+  obj = C.new
+benchmark:
+  vm1_attr_ivar_set: |
+    obj.a = 1
+    obj.b = 2
+loop_count: 30000000
diff --git a/benchmark/vm1_block.yml b/benchmark/vm1_block.yml
new file mode 100644
index 0000000000..ac7c940f93
--- /dev/null
+++ b/benchmark/vm1_block.yml
@@ -0,0 +1,9 @@
+prelude: |
+  def m
+    yield
+  end
+benchmark:
+  vm1_block: |
+    m{
+    }
+loop_count: 30000000
diff --git a/benchmark/vm1_blockparam.yml b/benchmark/vm1_blockparam.yml
new file mode 100644
index 0000000000..947b8c53d5
--- /dev/null
+++ b/benchmark/vm1_blockparam.yml
@@ -0,0 +1,7 @@
+prelude: |
+  def m &b
+  end
+benchmark:
+  vm1_blockparam: |
+    m{}
+loop_count: 30000000
diff --git a/benchmark/vm1_blockparam_call.yml b/benchmark/vm1_blockparam_call.yml
new file mode 100644
index 0000000000..e2817a3ce2
--- /dev/null
+++ b/benchmark/vm1_blockparam_call.yml
@@ -0,0 +1,8 @@
+prelude: |
+  def m &b
+    b.call
+  end
+benchmark:
+  vm1_blockparam_call: |
+    m{}
+loop_count: 30000000
diff --git a/benchmark/vm1_blockparam_pass.yml b/benchmark/vm1_blockparam_pass.yml
new file mode 100644
index 0000000000..ca1bef3369
--- /dev/null
+++ b/benchmark/vm1_blockparam_pass.yml
@@ -0,0 +1,12 @@
+prelude: |
+  def bp_yield
+    yield
+  end
+
+  def bp_pass &b
+    bp_yield &b
+  end
+benchmark:
+  vm1_blockparam_pass: |
+    bp_pass{}
+loop_count: 30000000
diff --git a/benchmark/vm1_blockparam_yield.yml b/benchmark/vm1_blockparam_yield.yml
new file mode 100644
index 0000000000..56ae617798
--- /dev/null
+++ b/benchmark/vm1_blockparam_yield.yml
@@ -0,0 +1,8 @@
+prelude: |
+  def bp_yield &b
+    yield
+  end
+benchmark:
+  vm1_blockparam_yield: |
+    bp_yield{}
+loop_count: 30000000
diff --git a/benchmark/vm1_const.yml b/benchmark/vm1_const.yml
new file mode 100644
index 0000000000..b98db1545c
--- /dev/null
+++ b/benchmark/vm1_const.yml
@@ -0,0 +1,7 @@
+prelude: |
+  Const = 1
+benchmark:
+  vm1_const: |
+    j = Const
+    k = Const
+loop_count: 30000000
diff --git a/benchmark/vm1_ensure.yml b/benchmark/vm1_ensure.yml
new file mode 100644
index 0000000000..afbbe38bec
--- /dev/null
+++ b/benchmark/vm1_ensure.yml
@@ -0,0 +1,14 @@
+# Not utilizing loop_count since using it for this is too unstable for now
+benchmark:
+  vm1_ensure: |
+    i = 0
+    while i<30_000_000
+      i += 1
+      begin
+        begin
+        ensure
+        end
+      ensure
+      end
+    end
+loop_count: 1
diff --git a/benchmark/vm1_float_simple.yml b/benchmark/vm1_float_simple.yml
new file mode 100644
index 0000000000..4e9ad1852b
--- /dev/null
+++ b/benchmark/vm1_float_simple.yml
@@ -0,0 +1,8 @@
+prelude: |
+  f = 0.0
+benchmark:
+  vm1_float_simple: |
+    f += 0.1; f -= 0.1
+    f += 0.1; f -= 0.1
+    f += 0.1; f -= 0.1
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_short_lived.yml b/benchmark/vm1_gc_short_lived.yml
new file mode 100644
index 0000000000..8fdcb7371d
--- /dev/null
+++ b/benchmark/vm1_gc_short_lived.yml
@@ -0,0 +1,9 @@
+benchmark:
+  vm1_gc_short_lived: |
+    a = '' # short-lived String
+    b = ''
+    c = ''
+    d = ''
+    e = ''
+    f = ''
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_short_with_complex_long.yml b/benchmark/vm1_gc_short_with_complex_long.yml
new file mode 100644
index 0000000000..c22ea74a60
--- /dev/null
+++ b/benchmark/vm1_gc_short_with_complex_long.yml
@@ -0,0 +1,25 @@
+prelude: |
+  def nested_hash h, n
+    if n == 0
+      ''
+    else
+      10.times{
+        h[Object.new] = nested_hash(h, n-1)
+      }
+    end
+  end
+
+  long_lived = Hash.new
+  nested_hash long_lived, 6
+
+  GC.start
+  GC.start
+benchmark:
+  vm1_gc_short_with_complex_long: |
+    a = '' # short-lived String
+    b = ''
+    c = ''
+    d = ''
+    e = ''
+    f = ''
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_short_with_long.yml b/benchmark/vm1_gc_short_with_long.yml
new file mode 100644
index 0000000000..c731aae548
--- /dev/null
+++ b/benchmark/vm1_gc_short_with_long.yml
@@ -0,0 +1,13 @@
+prelude: |
+  long_lived = Array.new(1_000_000){|i| "#{i}"}
+  GC.start
+  GC.start
+benchmark:
+  vm1_gc_short_with_long: |
+    a = '' # short-lived String
+    b = ''
+    c = ''
+    d = ''
+    e = ''
+    f = ''
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_short_with_symbol.yml b/benchmark/vm1_gc_short_with_symbol.yml
new file mode 100644
index 0000000000..7fc1abedd8
--- /dev/null
+++ b/benchmark/vm1_gc_short_with_symbol.yml
@@ -0,0 +1,13 @@
+prelude: |
+  50_000.times{|i| sym = "sym#{i}".to_sym}
+  GC.start
+  GC.start
+benchmark:
+  vm1_gc_short_with_symbol: |
+    a = '' # short-lived String
+    b = ''
+    c = ''
+    d = ''
+    e = ''
+    f = ''
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_wb_ary.yml b/benchmark/vm1_gc_wb_ary.yml
new file mode 100644
index 0000000000..50fb4b6f84
--- /dev/null
+++ b/benchmark/vm1_gc_wb_ary.yml
@@ -0,0 +1,12 @@
+prelude: |
+  short_lived_ary = []
+
+  if RUBY_VERSION >= "2.2.0"
+    GC.start(full_mark: false, immediate_mark: true, immediate_sweep: true)
+  end
+
+  short_lived = ''
+benchmark:
+  vm1_gc_wb_ary: |
+    short_lived_ary[0] = short_lived # write barrier
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_wb_ary_promoted.yml b/benchmark/vm1_gc_wb_ary_promoted.yml
new file mode 100644
index 0000000000..cf9b5de005
--- /dev/null
+++ b/benchmark/vm1_gc_wb_ary_promoted.yml
@@ -0,0 +1,15 @@
+prelude: |
+  long_lived = []
+
+  if RUBY_VERSION > "2.2.0"
+    3.times{ GC.start(full_mark: false, immediate_mark: true, immediate_sweep: true) }
+  elsif
+    GC.start
+  end
+
+  short_lived = ''
+
+benchmark:
+  vm1_gc_wb_ary_promoted: |
+    long_lived[0] = short_lived # write barrier
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_wb_obj.yml b/benchmark/vm1_gc_wb_obj.yml
new file mode 100644
index 0000000000..9dc08e7e1a
--- /dev/null
+++ b/benchmark/vm1_gc_wb_obj.yml
@@ -0,0 +1,15 @@
+prelude: |
+  class C
+    attr_accessor :foo
+  end
+  short_lived_obj = C.new
+
+  if RUBY_VERSION >= "2.2.0"
+    GC.start(full_mark: false, immediate_mark: true, immediate_sweep: true)
+  end
+
+  short_lived = ''
+benchmark:
+  vm1_gc_wb_obj: |
+    short_lived_obj.foo = short_lived # write barrier
+loop_count: 30000000
diff --git a/benchmark/vm1_gc_wb_obj_promoted.yml b/benchmark/vm1_gc_wb_obj_promoted.yml
new file mode 100644
index 0000000000..26859d2a52
--- /dev/null
+++ b/benchmark/vm1_gc_wb_obj_promoted.yml
@@ -0,0 +1,17 @@
+prelude: |
+  class C
+    attr_accessor :foo
+  end
+  long_lived = C.new
+
+  if RUBY_VERSION >= "2.2.0"
+    3.times{ GC.start(full_mark: false, immediate_mark: true, immediate_sweep: true) }
+  elsif
+    GC.start
+  end
+
+  short_lived = ''
+benchmark:
+  vm1_gc_wb_obj_promoted: |
+    long_lived.foo = short_lived # write barrier
+loop_count: 30000000
diff --git a/benchmark/vm1_ivar.yml b/benchmark/vm1_ivar.yml
new file mode 100644
index 0000000000..7aa6fac729
--- /dev/null
+++ b/benchmark/vm1_ivar.yml
@@ -0,0 +1,6 @@
+prelude: "@a = 1\n"
+benchmark:
+  vm1_ivar: |
+    j = @a
+    k = @a
+loop_count: 30000000
diff --git a/benchmark/vm1_ivar_set.yml b/benchmark/vm1_ivar_set.yml
new file mode 100644
index 0000000000..6f19412d16
--- /dev/null
+++ b/benchmark/vm1_ivar_set.yml
@@ -0,0 +1,5 @@
+benchmark:
+  vm1_ivar_set: |
+    @a = 1
+    @b = 2
+loop_count: 30000000
diff --git a/benchmark/vm1_length.yml b/benchmark/vm1_length.yml
new file mode 100644
index 0000000000..a18e2ca2e6
--- /dev/null
+++ b/benchmark/vm1_length.yml
@@ -0,0 +1,8 @@
+prelude: |
+  a = 'abc'
+  b = [1, 2, 3]
+benchmark:
+  vm1_length: |
+    a.length
+    b.length
+loop_count: 30000000
diff --git a/benchmark/vm1_lvar_init.yml b/benchmark/vm1_lvar_init.yml
new file mode 100644
index 0000000000..10e2becef9
--- /dev/null
+++ b/benchmark/vm1_lvar_init.yml
@@ -0,0 +1,21 @@
+# while loop cost is not removed because `i` is used in the script
+benchmark:
+  vm1_lvar_init: |
+    def m v
+      unless v
+        # unreachable code
+        v1 = v2 = v3 = v4 = v5 = v6 = v7 = v8 = v9 = v10 =
+        v11 = v12 = v13 = v14 = v15 = v16 = v17 = v18 = v19 = v20 =
+        v21 = v22 = v23 = v24 = v25 = v26 = v27 = v28 = v29 = v30 =
+        v31 = v32 = v33 = v34 = v35 = v36 = v37 = v38 = v39 = v40 =
+        v41 = v42 = v43 = v44 = v45 = v46 = v47 = v48 = v49 = v50 = 1
+      end
+    end
+
+    i = 0
+
+    while i<30_000_000
+      i += 1
+      m i
+    end
+loop_count: 1
diff --git a/benchmark/vm1_lvar_set.yml b/benchmark/vm1_lvar_set.yml
new file mode 100644
index 0000000000..df8f6b6ea4
--- /dev/null
+++ b/benchmark/vm1_lvar_set.yml
@@ -0,0 +1,4 @@
+benchmark:
+  vm1_lvar_set: |
+    a = b = c = d = e = f = g = h = j = k = l = m = n = o = p = q = r = 1
+loop_count: 30000000
diff --git a/benchmark/vm1_neq.yml b/benchmark/vm1_neq.yml
new file mode 100644
index 0000000000..65a8128dda
--- /dev/null
+++ b/benchmark/vm1_neq.yml
@@ -0,0 +1,7 @@
+prelude: |
+  obj1 = Object.new
+  obj2 = Object.new
+benchmark:
+  vm1_neq: |
+    obj1 != obj2
+loop_count: 30000000
diff --git a/benchmark/vm1_not.yml b/benchmark/vm1_not.yml
new file mode 100644
index 0000000000..0fb7b282a9
--- /dev/null
+++ b/benchmark/vm1_not.yml
@@ -0,0 +1,6 @@
+prelude: |
+  obj = Object.new
+benchmark:
+  vm1_not: |
+    !obj
+loop_count: 30000000
diff --git a/benchmark/vm1_rescue.yml b/benchmark/vm1_rescue.yml
new file mode 100644
index 0000000000..a175b823af
--- /dev/null
+++ b/benchmark/vm1_rescue.yml
@@ -0,0 +1,6 @@
+benchmark:
+  vm1_rescue: |
+    begin
+    rescue
+    end
+loop_count: 30000000
diff --git a/benchmark/vm1_simplereturn.yml b/benchmark/vm1_simplereturn.yml
new file mode 100644
index 0000000000..3564aac7e2
--- /dev/null
+++ b/benchmark/vm1_simplereturn.yml
@@ -0,0 +1,7 @@
+prelude: |
+  def m
+    return 1
+  end
+benchmark:
+  vm1_simplereturn: m
+loop_count: 30000000
diff --git a/benchmark/vm1_swap.yml b/benchmark/vm1_swap.yml
new file mode 100644
index 0000000000..fed87ccd62
--- /dev/null
+++ b/benchmark/vm1_swap.yml
@@ -0,0 +1,7 @@
+prelude: |
+  a = 1
+  b = 2
+benchmark:
+  vm1_swap: |
+    a, b = b, a
+loop_count: 30000000
diff --git a/benchmark/vm1_yield.yml b/benchmark/vm1_yield.yml
new file mode 100644
index 0000000000..ae1f9316f9
--- /dev/null
+++ b/benchmark/vm1_yield.yml
@@ -0,0 +1,13 @@
+# while loop cost is not removed due to benchmark_driver.gem's limitation
+benchmark:
+  vm1_yield: |
+    def m
+      i = 0
+      while i<30_000_000
+        i += 1
+        yield
+      end
+    end
+
+    m{}
+loop_count: 1
diff --git a/benchmark/vm2_array.yml b/benchmark/vm2_array.yml
new file mode 100644
index 0000000000..7373098d5e
--- /dev/null
+++ b/benchmark/vm2_array.yml
@@ -0,0 +1,4 @@
+benchmark:
+  vm2_array: |
+    a = [1,2,3,4,5,6,7,8,9,10]
+loop_count: 6000000
diff --git a/benchmark/vm2_bigarray.yml b/benchmark/vm2_bigarray.yml
new file mode 100644
index 0000000000..2ad6da3905
--- /dev/null
+++ b/benchmark/vm2_bigarray.yml
@@ -0,0 +1,105 @@
+benchmark:
+  vm2_bigarray: |
+    a = [
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+      1,2,3,4,5,6,7,8,9,10,
+    ]
+loop_count: 6000000
diff --git a/benchmark/vm2_bighash.yml b/benchmark/vm2_bighash.yml
new file mode 100644
index 0000000000..e9154e4ba9
--- /dev/null
+++ b/benchmark/vm2_bighash.yml
@@ -0,0 +1,4 @@
+benchmark:
+  vm2_bighash: |
+    a = {0=>0, 1=>1, 2=>2, 3=>3, 4=>4, 5=>5, 6=>6, 7=>7, 8=>8, 9=>9, 10=>10, 11=>11, 12=>12, 13=>13, 14=>14, 15=>15, 16=>16, 17=>17, 18=>18, 19=>19, 20=>20, 21=>21, 22=>22, 23=>23, 24=>24, 25=>25, 26=>26, 27=>27, 28=>28, 29=>29, 30=>30, 31=>31, 32=>32, 33=>33, 34=>34, 35=>35, 36=>36, 37=>37, 38=>38, 39=>39, 40=>40, 41=>41, 42=>42, 43=>43, 44=>44, 45=>45, 46=>46, 47=>47, 48=>48, 49=>49, 50=>50, 51=>51, 52=>52, 53=>53, 54=>54, 55=>55, 56=>56, 57=>57, 58=>58, 59=>59, 60=>60, 61=>61, 62=>62, 63=>63, 64=>64, 65=>65, 66=>66, 67=>67, 68=>68, 69=>69, 70=>70, 71=>71, 72=>72, 73=>73, 74=>74, 75=>75, 76=>76, 77=>77, 78=>78, 79=>79, 80=>80, 81=>81, 82=>82, 83=>83, 84=>84, 85=>85, 86=>86, 87=>87, 88=>88, 89=>89, 90=>90, 91=>91, 92=>92, 93=>93, 94=>94, 95=>95, 96=>96, 97=>97, 98=>98, 99=>99, 100=>100, 101=>101, 102=>102, 103=>103, 104=>104, 105=>105, 106=>106, 107=>107, 108=>108, 109=>109, 110=>110, 111=>111, 112=>112, 113=>113, 114=>114, 115=>115, 116=>116, 117=>117, 118=>118, 119=>119, 120=>120, 121=>121, 122=>122, 123=>123, 124=>124, 125=>125, 126=>126, 127=>127, 128=>128, 129=>129, 130=>130, 131=>131, 132=>132, 133=>133, 134=>134, 135=>135, 136=>136, 137=>137, 138=>138, 139=>139, 140=>140, 141=>141, 142=>142, 143=>143, 144=>144, 145=>145, 146=>146, 147=>147, 148=>148, 149=>149, 150=>150, 151=>151, 152=>152, 153=>153, 154=>154, 155=>155, 156=>156, 157=>157, 158=>158, 159=>159, 160=>160, 161=>161, 162=>162, 163=>163, 164=>164, 165=>165, 166=>166, 167=>167, 168=>168, 169=>169, 170=>170, 171=>171, 172=>172, 173=>173, 174=>174, 175=>175, 176=>176, 177=>177, 178=>178, 179=>179, 180=>180, 181=>181, 182=>182, 183=>183, 184=>184, 185=>185, 186=>186, 187=>187, 188=>188, 189=>189, 190=>190, 191=>191, 192=>192, 193=>193, 194=>194, 195=>195, 196=>196, 197=>197, 198=>198, 199=>199, 200=>200, 201=>201, 202=>202, 203=>203, 204=>204, 205=>205, 206=>206, 207=>207, 208=>208, 209=>209, 210=>210, 211=>211, 212=>212, 213=>213, 214=>214, 215=>215, 216=>216, 217=>217, 218=>218, 219=>219, 220=>220, 221=>221, 222=>222, 223=>223, 224=>224, 225=>225, 226=>226, 227=>227, 228=>228, 229=>229, 230=>230, 231=>231, 232=>232, 233=>233, 234=>234, 235=>235, 236=>236, 237=>237, 238=>238, 239=>239, 240=>240, 241=>241, 242=>242, 243=>243, 244=>244, 245=>245, 246=>246, 247=>247, 248=>248, 249=>249, 250=>250, 251=>251, 252=>252, 253=>253, 254=>254, 255=>255, 256=>256, 257=>257, 258=>258, 259=>259, 260=>260, 261=>261, 262=>262, 263=>263, 264=>264, 265=>265, 266=>266, 267=>267, 268=>268, 269=>269, 270=>270, 271=>271, 272=>272, 273=>273, 274=>274, 275=>275, 276=>276, 277=>277, 278=>278, 279=>279, 280=>280, 281=>281, 282=>282, 283=>283, 284=>284, 285=>285, 286=>286, 287=>287, 288=>288, 289=>289, 290=>290, 291=>291, 292=>292, 293=>293, 294=>294, 295=>295, 296=>296, 297=>297, 298=>298, 299=>299, 300=>300, 301=>301, 302=>302, 303=>303, 304=>304, 305=>305, 306=>306, 307=>307, 308=>308, 309=>309, 310=>310, 311=>311, 312=>312, 313=>313, 314=>314, 315=>315, 316=>316, 317=>317, 318=>318, 319=>319, 320=>320, 321=>321, 322=>322, 323=>323, 324=>324, 325=>325, 326=>326, 327=>327, 328=>328, 329=>329, 330=>330, 331=>331, 332=>332, 333=>333, 334=>334, 335=>335, 336=>336, 337=>337, 338=>338, 339=>339, 340=>340, 341=>341, 342=>342, 343=>343, 344=>344, 345=>345, 346=>346, 347=>347, 348=>348, 349=>349, 350=>350, 351=>351, 352=>352, 353=>353, 354=>354, 355=>355, 356=>356, 357=>357, 358=>358, 359=>359, 360=>360, 361=>361, 362=>362, 363=>363, 364=>364, 365=>365, 366=>366, 367=>367, 368=>368, 369=>369, 370=>370, 371=>371, 372=>372, 373=>373, 374=>374, 375=>375, 376=>376, 377=>377, 378=>378, 379=>379, 380=>380, 381=>381, 382=>382, 383=>383, 384=>384, 385=>385, 386=>386, 387=>387, 388=>388, 389=>389, 390=>390, 391=>391, 392=>392, 393=>393, 394=>394, 395=>395, 396=>396, 397=>397, 398=>398, 399=>399, 400=>400, 401=>401, 402=>402, 403=>403, 404=>404, 405=>405, 406=>406, 407=>407, 408=>408, 409=>409, 410=>410, 411=>411, 412=>412, 413=>413, 414=>414, 415=>415, 416=>416, 417=>417, 418=>418, 419=>419, 420=>420, 421=>421, 422=>422, 423=>423, 424=>424, 425=>425, 426=>426, 427=>427, 428=>428, 429=>429, 430=>430, 431=>431, 432=>432, 433=>433, 434=>434, 435=>435, 436=>436, 437=>437, 438=>438, 439=>439, 440=>440, 441=>441, 442=>442, 443=>443, 444=>444, 445=>445, 446=>446, 447=>447, 448=>448, 449=>449, 450=>450, 451=>451, 452=>452, 453=>453, 454=>454, 455=>455, 456=>456, 457=>457, 458=>458, 459=>459, 460=>460, 461=>461, 462=>462, 463=>463, 464=>464, 465=>465, 466=>466, 467=>467, 468=>468, 469=>469, 470=>470, 471=>471, 472=>472, 473=>473, 474=>474, 475=>475, 476=>476, 477=>477, 478=>478, 479=>479, 480=>480, 481=>481, 482=>482, 483=>483, 484=>484, 485=>485, 486=>486, 487=>487, 488=>488, 489=>489, 490=>490, 491=>491, 492=>492, 493=>493, 494=>494, 495=>495, 496=>496, 497=>497, 498=>498, 499=>499, 500=>500,}
+loop_count: 60000
diff --git a/benchmark/vm2_case.yml b/benchmark/vm2_case.yml
new file mode 100644
index 0000000000..7716783c09
--- /dev/null
+++ b/benchmark/vm2_case.yml
@@ -0,0 +1,13 @@
+benchmark:
+  vm2_case: |
+    case :foo
+    when :bar
+      raise
+    when :baz
+      raise
+    when :boo
+      raise
+    when :foo
+      # noop
+    end
+loop_count: 6000000
diff --git a/benchmark/vm2_case_lit.yml b/benchmark/vm2_case_lit.yml
new file mode 100644
index 0000000000..c49b8dfe5e
--- /dev/null
+++ b/benchmark/vm2_case_lit.yml
@@ -0,0 +1,23 @@
+# loop_count is not utilized since `i` is involved in the script
+benchmark:
+  vm2_case_lit: |
+    i = 0
+    @ret = [ "foo", true, false, :sym, 6, nil, 0.1, 0xffffffffffffffff ]
+    def foo(i)
+      @ret[i % @ret.size]
+    end
+
+    while i<6_000_000
+      case foo(i)
+      when "foo" then :foo
+      when true then true
+      when false then false
+      when :sym then :sym
+      when 6 then :fix
+      when nil then nil
+      when 0.1 then :float
+      when 0xffffffffffffffff then :big
+      end
+      i += 1
+    end
+loop_count: 1
diff --git a/benchmark/vm2_defined_method.yml b/benchmark/vm2_defined_method.yml
new file mode 100644
index 0000000000..e1b0d55674
--- /dev/null
+++ b/benchmark/vm2_defined_method.yml
@@ -0,0 +1,8 @@
+prelude: |
+  class Object
+    define_method(:m){}
+  end
+benchmark:
+  vm2_defined_method: |
+    m; m; m; m; m; m; m; m;
+loop_count: 6000000
diff --git a/benchmark/vm2_dstr.yml b/benchmark/vm2_dstr.yml
new file mode 100644
index 0000000000..f8bd6e0133
--- /dev/null
+++ b/benchmark/vm2_dstr.yml
@@ -0,0 +1,6 @@
+prelude: |
+  x = y = 'z'
+benchmark:
+  vm2_dstr: |
+    str = "foo#{x}bar#{y}baz"
+loop_count: 6000000
diff --git a/benchmark/vm2_eval.yml b/benchmark/vm2_eval.yml
new file mode 100644
index 0000000000..d506a9c079
--- /dev/null
+++ b/benchmark/vm2_eval.yml
@@ -0,0 +1,4 @@
+benchmark:
+  vm2_eval: |
+    eval("1")
+loop_count: 6000000
diff --git a/benchmark/vm2_fiber_switch.yml b/benchmark/vm2_fiber_switch.yml
new file mode 100644
index 0000000000..f3e4c91283
--- /dev/null
+++ b/benchmark/vm2_fiber_switch.yml
@@ -0,0 +1,9 @@
+prelude: |
+  # based on benchmark for [ruby-core:65518] [Feature #10341] by Knut Franke
+  fib = Fiber.new do
+    loop { Fiber.yield }
+  end
+benchmark:
+  vm2_fiber_switch: |
+    fib.resume
+loop_count: 6000000
diff --git a/benchmark/vm2_freezestring.yml b/benchmark/vm2_freezestring.yml
new file mode 100644
index 0000000000..b78af91a20
--- /dev/null
+++ b/benchmark/vm2_freezestring.yml
@@ -0,0 +1,10 @@
+prelude: |
+  class String
+    def freeze
+      -self
+    end
+  end
+benchmark:
+  vm2_freezestring: |
+    "tXnL1BP5T1WPXMjuFNLQtallEtRcay1t2lHtJSrlVsDgvunlbtfpr/DGdH0NGYE9".freeze
+loop_count: 6000000
diff --git a/benchmark/vm2_method.yml b/benchmark/vm2_method.yml
new file mode 100644
index 0000000000..cc7b9b28ff
--- /dev/null
+++ b/benchmark/vm2_method.yml
@@ -0,0 +1,8 @@
+prelude: |
+  def m
+    nil
+  end
+benchmark:
+  vm2_method: |
+    m; m; m; m; m; m; m; m;
+loop_count: 6000000
diff --git a/benchmark/vm2_method_missing.yml b/benchmark/vm2_method_missing.yml
new file mode 100644
index 0000000000..cbfb794b25
--- /dev/null
+++ b/benchmark/vm2_method_missing.yml
@@ -0,0 +1,11 @@
+prelude: |
+  class C
+    def method_missing mid
+    end
+  end
+
+  obj = C.new
+benchmark:
+  vm2_method_missing: |
+    obj.m; obj.m; obj.m; obj.m; obj.m; obj.m; obj.m; obj.m;
+loop_count: 6000000
diff --git a/benchmark/vm2_method_with_block.yml b/benchmark/vm2_method_with_block.yml
new file mode 100644
index 0000000000..6e522adccc
--- /dev/null
+++ b/benchmark/vm2_method_with_block.yml
@@ -0,0 +1,8 @@
+prelude: |
+  def m
+    nil
+  end
+benchmark:
+  vm2_method_with_block: |
+    m{}; m{}; m{}; m{}; m{}; m{}; m{}; m{};
+loop_count: 6000000
diff --git a/benchmark/vm2_module_ann_const_set.yml b/benchmark/vm2_module_ann_const_set.yml
new file mode 100644
index 0000000000..b0becd9d3d
--- /dev/null
+++ b/benchmark/vm2_module_ann_const_set.yml
@@ -0,0 +1,4 @@
+benchmark:
+  vm2_module_ann_const_set: |
+    Module.new.const_set(:X, Module.new)
+loop_count: 6000000
diff --git a/benchmark/vm2_module_const_set.yml b/benchmark/vm2_module_const_set.yml
new file mode 100644
index 0000000000..05a640069c
--- /dev/null
+++ b/benchmark/vm2_module_const_set.yml
@@ -0,0 +1,8 @@
+prelude: |
+  module M
+  end
+  $VERBOSE = nil
+benchmark:
+  vm2_module_const_set: |
+    M.const_set(:X, Module.new)
+loop_count: 6000000
diff --git a/benchmark/vm2_mutex.yml b/benchmark/vm2_mutex.yml
new file mode 100644
index 0000000000..c40a90444a
--- /dev/null
+++ b/benchmark/vm2_mutex.yml
@@ -0,0 +1,8 @@
+prelude: |
+  require 'thread'
+
+  m = Thread::Mutex.new
+benchmark:
+  vm2_mutex: |
+    m.synchronize{}
+loop_count: 6000000
diff --git a/benchmark/vm2_newlambda.yml b/benchmark/vm2_newlambda.yml
new file mode 100644
index 0000000000..93133f9f30
--- /dev/null
+++ b/benchmark/vm2_newlambda.yml
@@ -0,0 +1,4 @@
+benchmark:
+  vm2_newlambda: |
+    lambda {}
+loop_count: 6000000
diff --git a/benchmark/vm2_poly_method.yml b/benchmark/vm2_poly_method.yml
new file mode 100644
index 0000000000..0104bdfb66
--- /dev/null
+++ b/benchmark/vm2_poly_method.yml
@@ -0,0 +1,24 @@
+# loop_count is not utilized since `i` is involved in the script
+benchmark:
+  vm2_poly_method: |
+    class C1
+      def m
+        1
+      end
+    end
+    class C2
+      def m
+        2
+      end
+    end
+
+    o1 = C1.new
+    o2 = C2.new
+
+    i = 0
+    while i<6_000_000
+      o = (i % 2 == 0) ? o1 : o2
+      o.m; o.m; o.m; o.m; o.m; o.m; o.m; o.m
+      i += 1
+    end
+loop_count: 1
diff --git a/benchmark/vm2_poly_method_ov.yml b/benchmark/vm2_poly_method_ov.yml
new file mode 100644
index 0000000000..3748073ba2
--- /dev/null
+++ b/benchmark/vm2_poly_method_ov.yml
@@ -0,0 +1,24 @@
+# loop_count is not utilized since `i` is involved in the script
+benchmark:
+  vm2_poly_method_ov: |
+    class C1
+      def m
+        1
+      end
+    end
+    class C2
+      def m
+        2
+      end
+    end
+
+    o1 = C1.new
+    o2 = C2.new
+
+    i = 0
+    while i<6_000_000
+      o = (i % 2 == 0) ? o1 : o2
+    #  o.m; o.m; o.m; o.m; o.m; o.m; o.m; o.m
+      i += 1
+    end
+loop_count: 1
diff --git a/benchmark/vm2_poly_singleton.yml b/benchmark/vm2_poly_singleton.yml
new file mode 100644
index 0000000000..e58d7bfb37
--- /dev/null
+++ b/benchmark/vm2_poly_singleton.yml
@@ -0,0 +1,18 @@
+# loop_count is not utilized since `i` is involved in the script
+benchmark:
+  vm2_poly_singleton: |
+    class C1
+      def m; 1; end
+    end
+
+    o1 = C1.new
+    o2 = C1.new
+    o2.singleton_class
+
+    i = 0
+    while i<6_000_000 # benchmark loop 2
+      o = (i % 2 == 0) ? o1 : o2
+      o.m; o.m; o.m; o.m; o.m; o.m; o.m; o.m
+      i += 1
+    end
+loop_count: 1
diff --git a/benchmark/vm2_proc.yml b/benchmark/vm2_proc.yml
new file mode 100644
index 0000000000..5c36e936d9
--- /dev/null
+++ b/benchmark/vm2_proc.yml
@@ -0,0 +1,12 @@
+prelude: |
+  def m &b
+    b
+  end
+
+  pr = m{
+    a = 1
+  }
+benchmark:
+  vm2_proc: |
+    pr.call
+loop_count: 6000000
diff --git a/benchmark/vm2_raise1.yml b/benchmark/vm2_raise1.yml
new file mode 100644
index 0000000000..f6eb308968
--- /dev/null
+++ b/benchmark/vm2_raise1.yml
@@ -0,0 +1,16 @@
+prelude: |
+  def rec n
+    if n > 0
+      rec n-1
+    else
+      raise
+    end
+  end
+benchmark:
+  vm2_raise1: |
+    begin
+      rec 1
+    rescue
+      # ignore
+    end
+loop_count: 6000000
diff --git a/benchmark/vm2_raise2.yml b/benchmark/vm2_raise2.yml
new file mode 100644
index 0000000000..7d51b1b314
--- /dev/null
+++ b/benchmark/vm2_raise2.yml
@@ -0,0 +1,16 @@
+prelude: |
+  def rec n
+    if n > 0
+      rec n-1
+    else
+      raise
+    end
+  end
+benchmark:
+  vm2_raise2: |
+    begin
+      rec 10
+    rescue
+      # ignore
+    end
+loop_count: 6000000
diff --git a/benchmark/vm2_regexp.yml b/benchmark/vm2_regexp.yml
new file mode 100644
index 0000000000..0f3968a99b
--- /dev/null
+++ b/benchmark/vm2_regexp.yml
@@ -0,0 +1,6 @@
+prelude: |
+  str = 'xxxhogexxx'
+benchmark:
+  vm2_regexp: |
+    /hoge/ =~ str
+loop_count: 6000000
diff --git a/benchmark/vm2_send.yml b/benchmark/vm2_send.yml
new file mode 100644
index 0000000000..44a12a27d9
--- /dev/null
+++ b/benchmark/vm2_send.yml
@@ -0,0 +1,11 @@
+prelude: |
+  class C
+    def m
+    end
+  end
+
+  o = C.new
+benchmark:
+  vm2_send: |
+    o.__send__ :m
+loop_count: 6000000
diff --git a/benchmark/vm2_string_literal.yml b/benchmark/vm2_string_literal.yml
new file mode 100644
index 0000000000..54b0aec1fe
--- /dev/null
+++ b/benchmark/vm2_string_literal.yml
@@ -0,0 +1,4 @@
+benchmark:
+  vm2_string_literal: |
+    x = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
+loop_count: 6000000
diff --git a/benchmark/vm2_struct_big_aref_hi.yml b/benchmark/vm2_struct_big_aref_hi.yml
new file mode 100644
index 0000000000..eed1846d28
--- /dev/null
+++ b/benchmark/vm2_struct_big_aref_hi.yml
@@ -0,0 +1,7 @@
+prelude: |
+  s = Struct.new(*('a'..'z').map { |x| x.to_sym })
+  x = s.new
+benchmark:
+  vm2_struct_big_aref_hi: |
+    x.z # x[25]
+loop_count: 6000000
diff --git a/benchmark/vm2_struct_big_aref_lo.yml b/benchmark/vm2_struct_big_aref_lo.yml
new file mode 100644
index 0000000000..0915435b76
--- /dev/null
+++ b/benchmark/vm2_struct_big_aref_lo.yml
@@ -0,0 +1,7 @@
+prelude: |
+  s = Struct.new(*('a'..'z').map { |x| x.to_sym })
+  x = s.new
+benchmark:
+  vm2_struct_big_aref_lo: |
+    x.k # x[10]
+loop_count: 6000000
diff --git a/benchmark/vm2_struct_big_aset.yml b/benchmark/vm2_struct_big_aset.yml
new file mode 100644
index 0000000000..6af50103d3
--- /dev/null
+++ b/benchmark/vm2_struct_big_aset.yml
@@ -0,0 +1,11 @@
+# loop_count is not utilized since `i` is involved in the script
+benchmark:
+  vm2_struct_big_aset: |
+    s = Struct.new(*('a'..'z').map { |x| x.to_sym })
+    x = s.new
+    i = 0
+    while i<6_000_000
+      i += 1
+      x.k = i # x[10] = i
+    end
+loop_count: 1
diff --git a/benchmark/vm2_struct_big_href_hi.yml b/benchmark/vm2_struct_big_href_hi.yml
new file mode 100644
index 0000000000..60aa7fddf3
--- /dev/null
+++ b/benchmark/vm2_struct_big_href_hi.yml
@@ -0,0 +1,7 @@
+prelude: |
+  s = Struct.new(*('a'..'z').map { |x| x.to_sym })
+  x = s.new
+benchmark:
+  vm2_struct_big_href_hi: |
+    x[:z]
+loop_count: 6000000
diff --git a/benchmark/vm2_struct_big_href_lo.yml b/benchmark/vm2_struct_big_href_lo.yml
new file mode 100644
index 0000000000..c55c0bd16c
--- /dev/null
+++ b/benchmark/vm2_struct_big_href_lo.yml
@@ -0,0 +1,7 @@
+prelude: |
+  s = Struct.new(*('a'..'z').map { |x| x.to_sym })
+  x = s.new
+benchmark:
+  vm2_struct_big_href_lo: |
+    x[:k]
+loop_count: 6000000
diff --git a/benchmark/vm2_struct_big_hset.yml b/benchmark/vm2_struct_big_hset.yml
new file mode 100644
index 0000000000..d199c5bd47
--- /dev/null
+++ b/benchmark/vm2_struct_big_hset.yml
@@ -0,0 +1,11 @@
+# loop_count is not utilized since `i` is involved in the script
+benchmark:
+  vm2_struct_big_hset: |
+    s = Struct.new(*('a'..'z').map { |x| x.to_sym })
+    x = s.new
+    i = 0
+    while i<6_000_000
+      i += 1
+      x[:k] = i
+    end
+loop_count: 1
diff --git a/benchmark/vm2_struct_small_aref.yml b/benchmark/vm2_struct_small_aref.yml
new file mode 100644
index 0000000000..83381bed3a
--- /dev/null
+++ b/benchmark/vm2_struct_small_aref.yml
@@ -0,0 +1,7 @@
+prelude: |
+  s = Struct.new(:a, :b, :c)
+  x = s.new
+benchmark:
+  vm2_struct_small_aref: |
+    x.a
+loop_count: 6000000
diff --git a/benchmark/vm2_struct_small_aset.yml b/benchmark/vm2_struct_small_aset.yml
new file mode 100644
index 0000000000..3e84a61dd0
--- /dev/null
+++ b/benchmark/vm2_struct_small_aset.yml
@@ -0,0 +1,11 @@
+# loop_count is not utilized since `i` is involved in the script
+benchmark:
+  vm2_struct_small_aset: |
+    s = Struct.new(:a, :b, :c)
+    x = s.new
+    i = 0
+    while i<6_000_000
+      i += 1
+      x.a = i
+    end
+loop_count: 1
diff --git a/benchmark/vm2_struct_small_href.yml b/benchmark/vm2_struct_small_href.yml
new file mode 100644
index 0000000000..b744f070d1
--- /dev/null
+++ b/benchmark/vm2_struct_small_href.yml
@@ -0,0 +1,7 @@
+prelude: |
+  s = Struct.new(:a, :b, :c)
+  x = s.new
+benchmark:
+  vm2_struct_small_href: |
+    x[:a]
+loop_count: 6000000
diff --git a/benchmark/vm2_struct_small_hset.yml b/benchmark/vm2_struct_small_hset.yml
new file mode 100644
index 0000000000..d43845d6e0
--- /dev/null
+++ b/benchmark/vm2_struct_small_hset.yml
@@ -0,0 +1,7 @@
+prelude: |
+  s = Struct.new(:a, :b, :c)
+  x = s.new
+benchmark:
+  vm2_struct_small_hset: |
+    x[:a] = 1
+loop_count: 6000000
diff --git a/benchmark/vm2_super.yml b/benchmark/vm2_super.yml
new file mode 100644
index 0000000000..674743762a
--- /dev/null
+++ b/benchmark/vm2_super.yml
@@ -0,0 +1,17 @@
+prelude: |
+  class C
+    def m
+      1
+    end
+  end
+
+  class CC < C
+    def m
+      super()
+    end
+  end
+
+  obj = CC.new
+benchmark:
+  vm2_super: obj.m
+loop_count: 6000000
diff --git a/benchmark/vm2_unif1.yml b/benchmark/vm2_unif1.yml
new file mode 100644
index 0000000000..caef13279f
--- /dev/null
+++ b/benchmark/vm2_unif1.yml
@@ -0,0 +1,7 @@
+prelude: |
+  def m a, b
+  end
+benchmark:
+  vm2_unif1: |
+    m 100, 200
+loop_count: 6000000
diff --git a/benchmark/vm2_zsuper.yml b/benchmark/vm2_zsuper.yml
new file mode 100644
index 0000000000..f760cfd48e
--- /dev/null
+++ b/benchmark/vm2_zsuper.yml
@@ -0,0 +1,18 @@
+prelude: |
+  class C
+    def m a
+      1
+    end
+  end
+
+  class CC < C
+    def m a
+      super
+    end
+  end
+
+  obj = CC.new
+benchmark:
+  vm2_zsuper: |
+    obj.m 10
+loop_count: 6000000
diff --git a/benchmark/vm3_backtrace.rb b/benchmark/vm3_backtrace.rb
new file mode 100644
index 0000000000..0fbf73e1ca
--- /dev/null
+++ b/benchmark/vm3_backtrace.rb
@@ -0,0 +1,22 @@
+# get last backtrace
+
+begin
+  caller(0, 0)
+rescue ArgumentError
+  alias caller_orig caller
+  def caller lev, n
+    caller_orig(lev)[0..n]
+  end
+end
+
+def rec n
+  if n < 0
+    100_000.times{
+      caller(0, 1)
+    }
+  else
+    rec(n-1)
+  end
+end
+
+rec 50
diff --git a/benchmark/vm3_clearmethodcache.rb b/benchmark/vm3_clearmethodcache.rb
new file mode 100644
index 0000000000..9661323cd2
--- /dev/null
+++ b/benchmark/vm3_clearmethodcache.rb
@@ -0,0 +1,8 @@
+i = 0
+while i<200_000
+  i += 1
+
+  Class.new{
+    def m; end
+  }
+end
diff --git a/benchmark/vm3_gc.rb b/benchmark/vm3_gc.rb
new file mode 100644
index 0000000000..e668026915
--- /dev/null
+++ b/benchmark/vm3_gc.rb
@@ -0,0 +1,6 @@
+5000.times do
+  100.times do
+    {"xxxx"=>"yyyy"}
+  end
+  GC.start
+end
diff --git a/benchmark/vm3_gc_old_full.rb b/benchmark/vm3_gc_old_full.rb
new file mode 100644
index 0000000000..cfdfc8c5a5
--- /dev/null
+++ b/benchmark/vm3_gc_old_full.rb
@@ -0,0 +1,4 @@
+old_object = Array.new(1_000_000){''}
+100.times do
+  GC.start
+end
diff --git a/benchmark/vm3_gc_old_immediate.rb b/benchmark/vm3_gc_old_immediate.rb
new file mode 100644
index 0000000000..ad22feb655
--- /dev/null
+++ b/benchmark/vm3_gc_old_immediate.rb
@@ -0,0 +1,4 @@
+old_object = Array.new(1_000_000){''}
+30_000.times do
+  GC.start(full_mark: false, immediate_sweep: true)
+end
diff --git a/benchmark/vm3_gc_old_lazy.rb b/benchmark/vm3_gc_old_lazy.rb
new file mode 100644
index 0000000000..b74d44baf1
--- /dev/null
+++ b/benchmark/vm3_gc_old_lazy.rb
@@ -0,0 +1,4 @@
+old_object = Array.new(1_000_000){''}
+30_000.times do
+  GC.start(full_mark: false, immediate_sweep: false)
+end
diff --git a/benchmark/vm_symbol_block_pass.rb b/benchmark/vm_symbol_block_pass.rb
new file mode 100644
index 0000000000..1d433353e1
--- /dev/null
+++ b/benchmark/vm_symbol_block_pass.rb
@@ -0,0 +1,13 @@
+class C
+  1000.times {|i|
+    eval("def i#{i};end")
+  }
+end
+
+c = C.new
+m = C.instance_methods(false)
+5_000.times do
+  m.each do |n|
+    c.tap(&n)
+  end
+end
diff --git a/benchmark/vm_thread_alive_check1.rb b/benchmark/vm_thread_alive_check1.rb
new file mode 100644
index 0000000000..c993accdda
--- /dev/null
+++ b/benchmark/vm_thread_alive_check1.rb
@@ -0,0 +1,6 @@
+5_000.times{
+  t = Thread.new{}
+  while t.alive?
+    Thread.pass
+  end
+}
diff --git a/benchmark/vm_thread_close.rb b/benchmark/vm_thread_close.rb
new file mode 100644
index 0000000000..3e9a265ce8
--- /dev/null
+++ b/benchmark/vm_thread_close.rb
@@ -0,0 +1,6 @@
+1000.times { Thread.new { sleep } }
+i = 0
+while i<100_000 # benchmark loop 3
+  i += 1
+  IO.pipe.each(&:close)
+end
diff --git a/benchmark/vm_thread_condvar1.rb b/benchmark/vm_thread_condvar1.rb
new file mode 100644
index 0000000000..cf5706b23e
--- /dev/null
+++ b/benchmark/vm_thread_condvar1.rb
@@ -0,0 +1,28 @@
+# two threads, two mutex, two condvar ping-pong
+require 'thread'
+m1 = Mutex.new
+m2 = Mutex.new
+cv1 = ConditionVariable.new
+cv2 = ConditionVariable.new
+max = 100000
+i = 0
+wait = nil
+m2.synchronize do
+  wait = Thread.new do
+    m1.synchronize do
+      m2.synchronize { cv2.signal }
+      while (i += 1) < max
+        cv1.wait(m1)
+        cv2.signal
+      end
+    end
+  end
+  cv2.wait(m2)
+end
+m1.synchronize do
+  while i < max
+    cv1.signal
+    cv2.wait(m1)
+  end
+end
+wait.join
diff --git a/benchmark/vm_thread_condvar2.rb b/benchmark/vm_thread_condvar2.rb
new file mode 100644
index 0000000000..7c8dc19481
--- /dev/null
+++ b/benchmark/vm_thread_condvar2.rb
@@ -0,0 +1,35 @@
+# many threads, one mutex, many condvars
+require 'thread'
+m = Mutex.new
+cv1 = ConditionVariable.new
+cv2 = ConditionVariable.new
+max = 1000
+n = 100
+waiting = 0
+scvs = []
+waiters = n.times.map do |i|
+  start_cv = ConditionVariable.new
+  scvs << start_cv
+  start_mtx = Mutex.new
+  start_mtx.synchronize do
+    th = Thread.new(start_mtx, start_cv) do |sm, scv|
+      m.synchronize do
+        sm.synchronize { scv.signal }
+        max.times do
+          cv2.signal if (waiting += 1) == n
+          cv1.wait(m)
+        end
+      end
+    end
+    start_cv.wait(start_mtx)
+    th
+  end
+end
+m.synchronize do
+  max.times do
+    cv2.wait(m) until waiting == n
+    waiting = 0
+    cv1.broadcast
+  end
+end
+waiters.each(&:join)
diff --git a/benchmark/vm_thread_create_join.rb b/benchmark/vm_thread_create_join.rb
new file mode 100644
index 0000000000..393cd45df9
--- /dev/null
+++ b/benchmark/vm_thread_create_join.rb
@@ -0,0 +1,6 @@
+i = 0
+while i<100_000 # benchmark loop 3
+  i += 1
+  Thread.new{
+  }.join
+end
diff --git a/benchmark/vm_thread_mutex1.rb b/benchmark/vm_thread_mutex1.rb
new file mode 100644
index 0000000000..66e42c85e1
--- /dev/null
+++ b/benchmark/vm_thread_mutex1.rb
@@ -0,0 +1,21 @@
+# one thread, one mutex (no contention)
+
+require 'thread'
+m = Thread::Mutex.new
+r = 0
+max = 2000
+lmax = max * max
+(1..1).map{
+  Thread.new{
+    i = 0
+    while i<lmax
+      i += 1
+      m.synchronize{
+        r += 1
+      }
+    end
+  }
+}.each{|e|
+  e.join
+}
+raise r.to_s if r != max * max
diff --git a/benchmark/vm_thread_mutex2.rb b/benchmark/vm_thread_mutex2.rb
new file mode 100644
index 0000000000..6e6c804c31
--- /dev/null
+++ b/benchmark/vm_thread_mutex2.rb
@@ -0,0 +1,21 @@
+# two threads, one mutex
+
+require 'thread'
+m = Thread::Mutex.new
+r = 0
+max = 2000
+lmax = (max * max)/2
+(1..2).map{
+  Thread.new{
+    i = 0
+    while i<lmax
+      i += 1
+      m.synchronize{
+        r += 1
+      }
+    end
+  }
+}.each{|e|
+  e.join
+}
+raise r.to_s if r != max * max
diff --git a/benchmark/vm_thread_mutex3.rb b/benchmark/vm_thread_mutex3.rb
new file mode 100644
index 0000000000..c750dc542a
--- /dev/null
+++ b/benchmark/vm_thread_mutex3.rb
@@ -0,0 +1,20 @@
+# 1000 threads, one mutex
+
+require 'thread'
+m = Thread::Mutex.new
+r = 0
+max = 2000
+(1..max).map{
+  Thread.new{
+    i = 0
+    while i<max
+      i += 1
+      m.synchronize{
+        r += 1
+      }
+    end
+  }
+}.each{|e|
+  e.join
+}
+raise r.to_s if r != max * max
diff --git a/benchmark/vm_thread_pass.rb b/benchmark/vm_thread_pass.rb
new file mode 100644
index 0000000000..b5b3c0bc85
--- /dev/null
+++ b/benchmark/vm_thread_pass.rb
@@ -0,0 +1,15 @@
+# Plenty Thtread.pass
+# A performance may depend on GVL implementation.
+
+tmax = (ARGV.shift || 2).to_i
+lmax = 200_000 / tmax
+
+(1..tmax).map{
+  Thread.new{
+    lmax.times{
+      Thread.pass
+    }
+  }
+}.each{|t| t.join}
+
+
diff --git a/benchmark/vm_thread_pass_flood.rb b/benchmark/vm_thread_pass_flood.rb
new file mode 100644
index 0000000000..a660aafc18
--- /dev/null
+++ b/benchmark/vm_thread_pass_flood.rb
@@ -0,0 +1,10 @@
+# n.b. this is a good test for GVL when pinned to a single CPU
+
+1000.times{
+  Thread.new{loop{Thread.pass}}
+}
+
+i = 0
+while i<10000
+  i += 1
+end
diff --git a/benchmark/vm_thread_pipe.rb b/benchmark/vm_thread_pipe.rb
new file mode 100644
index 0000000000..112a621905
--- /dev/null
+++ b/benchmark/vm_thread_pipe.rb
@@ -0,0 +1,17 @@
+# Measure small and plenty pipe read/write.
+# A performance may depend on GVL implementation.
+
+lmax = 100_000
+r, w = IO.pipe
+[Thread.new{
+  lmax.times{
+    w.write('a')
+  }
+  p "w:exit"
+}, Thread.new{
+  lmax.times{
+    r.read(1)
+  }
+  p "r:exit"
+}].each{|t| t.join}
+
diff --git a/benchmark/vm_thread_queue.rb b/benchmark/vm_thread_queue.rb
new file mode 100644
index 0000000000..274ceda366
--- /dev/null
+++ b/benchmark/vm_thread_queue.rb
@@ -0,0 +1,18 @@
+require 'thread'
+
+n = 1_000_000
+q = Thread::Queue.new
+consumer = Thread.new{
+  while q.pop
+    # consuming
+  end
+}
+
+producer = Thread.new{
+  n.times{
+    q.push true
+  }
+  q.push nil
+}
+
+consumer.join
diff --git a/benchmark/vm_thread_sized_queue.rb b/benchmark/vm_thread_sized_queue.rb
new file mode 100644
index 0000000000..7b9af5482b
--- /dev/null
+++ b/benchmark/vm_thread_sized_queue.rb
@@ -0,0 +1,20 @@
+require 'thread'
+# on producer, one consumer
+
+n = 1_000_000
+q = Thread::SizedQueue.new(100)
+consumer = Thread.new{
+  while q.pop
+    # consuming
+  end
+}
+
+producer = Thread.new{
+  while n > 0
+    q.push true
+    n -= 1
+  end
+  q.push nil
+}
+
+consumer.join
diff --git a/benchmark/vm_thread_sized_queue2.rb b/benchmark/vm_thread_sized_queue2.rb
new file mode 100644
index 0000000000..de9f55e978
--- /dev/null
+++ b/benchmark/vm_thread_sized_queue2.rb
@@ -0,0 +1,23 @@
+require 'thread'
+# one producer, many consumers
+n = 1_000_000
+m = 10
+q = Thread::SizedQueue.new(100)
+consumers = m.times.map do
+  Thread.new do
+    while q.pop
+      # consuming
+    end
+  end
+end
+
+producer = Thread.new do
+  while n > 0
+    q.push true
+    n -= 1
+  end
+  m.times { q.push nil }
+end
+
+producer.join
+consumers.each(&:join)
diff --git a/benchmark/vm_thread_sized_queue3.rb b/benchmark/vm_thread_sized_queue3.rb
new file mode 100644
index 0000000000..ce5f1796d8
--- /dev/null
+++ b/benchmark/vm_thread_sized_queue3.rb
@@ -0,0 +1,22 @@
+require 'thread'
+# many producers, one consumer
+n = 1_000_000
+m = 10
+q = Thread::SizedQueue.new(100)
+consumer = Thread.new do
+  while q.pop
+    # consuming
+  end
+end
+
+producers = m.times.map do
+  Thread.new do
+    while n > 0
+      q.push true
+      n -= 1
+    end
+  end
+end
+producers.each(&:join)
+q.push nil
+consumer.join
diff --git a/benchmark/vm_thread_sized_queue4.rb b/benchmark/vm_thread_sized_queue4.rb
new file mode 100644
index 0000000000..a9b7d80ec0
--- /dev/null
+++ b/benchmark/vm_thread_sized_queue4.rb
@@ -0,0 +1,26 @@
+require 'thread'
+# many producers, many consumers
+nr = 1_000_000
+n = 10
+m = 10
+q = Thread::SizedQueue.new(100)
+consumers = n.times.map do
+  Thread.new do
+    while q.pop
+      # consuming
+    end
+  end
+end
+
+producers = m.times.map do
+  Thread.new do
+    while nr > 0
+      q.push true
+      nr -= 1
+    end
+  end
+end
+
+producers.each(&:join)
+n.times { q.push nil }
+consumers.each(&:join)
diff --git a/bignum.c b/bignum.c
index cf7649c47e..ee3b49fd04 100644
--- a/bignum.c
+++ b/bignum.c
@@ -2,379 +2,4102 @@
 
   bignum.c -
 
-  $Author: shyouhei $
-  $Date: 2006/12/06 10:14:12 $
+  $Author$
   created at: Fri Jun 10 00:48:55 JST 1994
 
-  Copyright (C) 1993-2003 Yukihiro Matsumoto
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
 
 **********************************************************************/
 
-#include "ruby.h"
+#include "internal.h"
+#include "ruby/thread.h"
+#include "ruby/util.h"
+#include "id.h"
 
+#ifdef HAVE_STRINGS_H
+#include <strings.h>
+#endif
 #include <math.h>
+#include <float.h>
 #include <ctype.h>
 #ifdef HAVE_IEEEFP_H
 #include <ieeefp.h>
 #endif
+#include "ruby_assert.h"
+
+#if defined(HAVE_LIBGMP) && defined(HAVE_GMP_H)
+#define USE_GMP
+#include <gmp.h>
+#endif
+
+#define RB_BIGNUM_TYPE_P(x) RB_TYPE_P((x), T_BIGNUM)
 
+#ifndef RUBY_INTEGER_UNIFICATION
 VALUE rb_cBignum;
+#endif
+const char ruby_digitmap[] = "0123456789abcdefghijklmnopqrstuvwxyz";
+
+#ifndef SIZEOF_BDIGIT_DBL
+# if SIZEOF_INT*2 <= SIZEOF_LONG_LONG
+#  define SIZEOF_BDIGIT_DBL SIZEOF_LONG_LONG
+# else
+#  define SIZEOF_BDIGIT_DBL SIZEOF_LONG
+# endif
+#endif
 
-#if defined __MINGW32__
-#define USHORT _USHORT
+STATIC_ASSERT(sizeof_bdigit_dbl, sizeof(BDIGIT_DBL) == SIZEOF_BDIGIT_DBL);
+STATIC_ASSERT(sizeof_bdigit_dbl_signed, sizeof(BDIGIT_DBL_SIGNED) == SIZEOF_BDIGIT_DBL);
+STATIC_ASSERT(sizeof_bdigit, SIZEOF_BDIGIT <= sizeof(BDIGIT));
+STATIC_ASSERT(sizeof_bdigit_and_dbl, SIZEOF_BDIGIT*2 <= SIZEOF_BDIGIT_DBL);
+STATIC_ASSERT(bdigit_signedness, 0 < (BDIGIT)-1);
+STATIC_ASSERT(bdigit_dbl_signedness, 0 < (BDIGIT_DBL)-1);
+STATIC_ASSERT(bdigit_dbl_signed_signedness, 0 > (BDIGIT_DBL_SIGNED)-1);
+STATIC_ASSERT(rbignum_embed_len_max, BIGNUM_EMBED_LEN_MAX <= (BIGNUM_EMBED_LEN_MASK >> BIGNUM_EMBED_LEN_SHIFT));
+
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+STATIC_ASSERT(sizeof_long_and_sizeof_bdigit, SIZEOF_LONG % SIZEOF_BDIGIT == 0);
+#else
+STATIC_ASSERT(sizeof_long_and_sizeof_bdigit, SIZEOF_BDIGIT % SIZEOF_LONG == 0);
 #endif
 
-#define BDIGITS(x) ((BDIGIT*)RBIGNUM(x)->digits)
-#define BITSPERDIG (SIZEOF_BDIGITS*CHAR_BIT)
+#ifdef WORDS_BIGENDIAN
+#   define HOST_BIGENDIAN_P 1
+#else
+#   define HOST_BIGENDIAN_P 0
+#endif
+/* (!LSHIFTABLE(d, n) ? 0 : (n)) is same as n but suppress a warning, C4293, by Visual Studio.  */
+#define LSHIFTABLE(d, n) ((n) < sizeof(d) * CHAR_BIT)
+#define LSHIFTX(d, n) (!LSHIFTABLE(d, n) ? 0 : ((d) << (!LSHIFTABLE(d, n) ? 0 : (n))))
+#define CLEAR_LOWBITS(d, numbits) ((d) & LSHIFTX(~((d)*0), (numbits)))
+#define FILL_LOWBITS(d, numbits) ((d) | (LSHIFTX(((d)*0+1), (numbits))-1))
+#define POW2_P(x) (((x)&((x)-1))==0)
+
+#define BDIGITS(x) (BIGNUM_DIGITS(x))
+#define BITSPERDIG (SIZEOF_BDIGIT*CHAR_BIT)
 #define BIGRAD ((BDIGIT_DBL)1 << BITSPERDIG)
-#define DIGSPERLONG ((unsigned int)(SIZEOF_LONG/SIZEOF_BDIGITS))
-#if HAVE_LONG_LONG
-# define DIGSPERLL ((unsigned int)(SIZEOF_LONG_LONG/SIZEOF_BDIGITS))
+#define BIGRAD_HALF ((BDIGIT)(BIGRAD >> 1))
+#define BDIGIT_MSB(d) (((d) & BIGRAD_HALF) != 0)
+#define BIGUP(x) LSHIFTX(((x) + (BDIGIT_DBL)0), BITSPERDIG)
+#define BIGDN(x) RSHIFT((x),BITSPERDIG)
+#define BIGLO(x) ((BDIGIT)((x) & BDIGMAX))
+#define BDIGMAX ((BDIGIT)(BIGRAD-1))
+#define BDIGIT_DBL_MAX (~(BDIGIT_DBL)0)
+
+#if SIZEOF_BDIGIT == 2
+#   define swap_bdigit(x) swap16(x)
+#elif SIZEOF_BDIGIT == 4
+#   define swap_bdigit(x) swap32(x)
+#elif SIZEOF_BDIGIT == 8
+#   define swap_bdigit(x) swap64(x)
 #endif
-#define BIGUP(x) ((BDIGIT_DBL)(x) << BITSPERDIG)
-#define BIGDN(x) RSHIFT(x,BITSPERDIG)
-#define BIGLO(x) ((BDIGIT)((x) & (BIGRAD-1)))
-#define BDIGMAX ((BDIGIT)-1)
 
-#define BIGZEROP(x) (RBIGNUM(x)->len == 0 || (RBIGNUM(x)->len == 1 && BDIGITS(x)[0] == 0))
+#define BIGZEROP(x) (BIGNUM_LEN(x) == 0 || \
+		     (BDIGITS(x)[0] == 0 && \
+		      (BIGNUM_LEN(x) == 1 || bigzero_p(x))))
+#define BIGSIZE(x) (BIGNUM_LEN(x) == 0 ? (size_t)0 : \
+    BDIGITS(x)[BIGNUM_LEN(x)-1] ? \
+        (size_t)(BIGNUM_LEN(x)*SIZEOF_BDIGIT - nlz(BDIGITS(x)[BIGNUM_LEN(x)-1])/CHAR_BIT) : \
+    rb_absint_size(x, NULL))
+
+#define BIGDIVREM_EXTRA_WORDS 1
+#define bdigit_roomof(n) roomof(n, SIZEOF_BDIGIT)
+#define BARY_ARGS(ary) ary, numberof(ary)
+
+#define BARY_ADD(z, x, y) bary_add(BARY_ARGS(z), BARY_ARGS(x), BARY_ARGS(y))
+#define BARY_SUB(z, x, y) bary_sub(BARY_ARGS(z), BARY_ARGS(x), BARY_ARGS(y))
+#define BARY_SHORT_MUL(z, x, y) bary_short_mul(BARY_ARGS(z), BARY_ARGS(x), BARY_ARGS(y))
+#define BARY_DIVMOD(q, r, x, y) bary_divmod(BARY_ARGS(q), BARY_ARGS(r), BARY_ARGS(x), BARY_ARGS(y))
+#define BARY_ZERO_P(x) bary_zero_p(BARY_ARGS(x))
+
+#define BIGNUM_SET_NEGATIVE_SIGN(b) BIGNUM_SET_SIGN(b, 0)
+#define BIGNUM_SET_POSITIVE_SIGN(b) BIGNUM_SET_SIGN(b, 1)
+
+#define bignew(len,sign) bignew_1(rb_cInteger,(len),(sign))
+
+#define BDIGITS_ZERO(ptr, n) do { \
+  BDIGIT *bdigitz_zero_ptr = (ptr); \
+  size_t bdigitz_zero_n = (n); \
+  while (bdigitz_zero_n) { \
+    *bdigitz_zero_ptr++ = 0; \
+    bdigitz_zero_n--; \
+  } \
+} while (0)
+
+#define BARY_TRUNC(ds, n) do { \
+        while (0 < (n) && (ds)[(n)-1] == 0) \
+            (n)--; \
+    } while (0)
+
+#define KARATSUBA_BALANCED(xn, yn) ((yn)/2 < (xn))
+#define TOOM3_BALANCED(xn, yn) (((yn)+2)/3 * 2 < (xn))
+
+#define GMP_MUL_DIGITS 20
+#define KARATSUBA_MUL_DIGITS 70
+#define TOOM3_MUL_DIGITS 150
+
+#define GMP_DIV_DIGITS 20
+#define GMP_BIG2STR_DIGITS 20
+#define GMP_STR2BIG_DIGITS 20
+#ifdef USE_GMP
+# define NAIVE_MUL_DIGITS GMP_MUL_DIGITS
+#else
+# define NAIVE_MUL_DIGITS KARATSUBA_MUL_DIGITS
+#endif
 
-static VALUE
-bignew_1(klass, len, sign)
-    VALUE klass;
-    long len;
+typedef void (mulfunc_t)(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn);
+
+static mulfunc_t bary_mul_toom3_start;
+static mulfunc_t bary_mul_karatsuba_start;
+static BDIGIT bigdivrem_single(BDIGIT *qds, const BDIGIT *xds, size_t xn, BDIGIT y);
+static void bary_divmod(BDIGIT *qds, size_t qn, BDIGIT *rds, size_t rn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn);
+
+static VALUE bigmul0(VALUE x, VALUE y);
+static void bary_mul_toom3(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn);
+static VALUE bignew_1(VALUE klass, size_t len, int sign);
+static inline VALUE bigtrunc(VALUE x);
+
+static VALUE bigsq(VALUE x);
+static void bigdivmod(VALUE x, VALUE y, volatile VALUE *divp, volatile VALUE *modp);
+static inline VALUE power_cache_get_power(int base, int power_level, size_t *numdigits_ret);
+
+#if SIZEOF_BDIGIT <= SIZEOF_INT
+static int nlz(BDIGIT x) { return nlz_int((unsigned int)x) - (SIZEOF_INT-SIZEOF_BDIGIT) * CHAR_BIT; }
+#elif SIZEOF_BDIGIT <= SIZEOF_LONG
+static int nlz(BDIGIT x) { return nlz_long((unsigned long)x) - (SIZEOF_LONG-SIZEOF_BDIGIT) * CHAR_BIT; }
+#elif SIZEOF_BDIGIT <= SIZEOF_LONG_LONG
+static int nlz(BDIGIT x) { return nlz_long_long((unsigned LONG_LONG)x) - (SIZEOF_LONG_LONG-SIZEOF_BDIGIT) * CHAR_BIT; }
+#elif SIZEOF_BDIGIT <= SIZEOF_INT128_T
+static int nlz(BDIGIT x) { return nlz_int128((uint128_t)x) - (SIZEOF_INT128_T-SIZEOF_BDIGIT) * CHAR_BIT; }
+#endif
+
+#define U16(a) ((uint16_t)(a))
+#define U32(a) ((uint32_t)(a))
+#ifdef HAVE_UINT64_T
+#define U64(a,b) (((uint64_t)(a) << 32) | (b))
+#endif
+#ifdef HAVE_UINT128_T
+#define U128(a,b,c,d) (((uint128_t)U64(a,b) << 64) | U64(c,d))
+#endif
+
+/* The following script, maxpow.rb, generates the tables follows.
+
+def big(n, bits)
+  ns = []
+  ((bits+31)/32).times {
+    ns << sprintf("0x%08x", n & 0xffff_ffff)
+    n >>= 32
+  }
+  "U#{bits}(" + ns.reverse.join(",") + ")"
+end
+def values(ary, width, indent)
+  lines = [""]
+  ary.each {|e|
+    lines << "" if !ary.last.empty? && width < (lines.last + e + ", ").length
+    lines.last << e + ", "
+  }
+  lines.map {|line| " " * indent + line.chomp(" ") + "\n" }.join
+end
+[16,32,64,128].each {|bits|
+  max = 2**bits-1
+  exps = []
+  nums = []
+  2.upto(36) {|base|
+    exp = 0
+    n = 1
+    while n * base <= max
+      exp += 1
+      n *= base
+    end
+    exps << exp.to_s
+    nums << big(n, bits)
+  }
+  puts "#ifdef HAVE_UINT#{bits}_T"
+  puts "static const int maxpow#{bits}_exp[35] = {"
+  print values(exps, 70, 4)
+  puts "};"
+  puts "static const uint#{bits}_t maxpow#{bits}_num[35] = {"
+  print values(nums, 70, 4)
+  puts "};"
+  puts "#endif"
+}
+
+ */
+
+#if SIZEOF_BDIGIT_DBL == 2
+static const int maxpow16_exp[35] = {
+    15, 10, 7, 6, 6, 5, 5, 5, 4, 4, 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 3, 3,
+    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+};
+static const uint16_t maxpow16_num[35] = {
+    U16(0x00008000), U16(0x0000e6a9), U16(0x00004000), U16(0x00003d09),
+    U16(0x0000b640), U16(0x000041a7), U16(0x00008000), U16(0x0000e6a9),
+    U16(0x00002710), U16(0x00003931), U16(0x00005100), U16(0x00006f91),
+    U16(0x00009610), U16(0x0000c5c1), U16(0x00001000), U16(0x00001331),
+    U16(0x000016c8), U16(0x00001acb), U16(0x00001f40), U16(0x0000242d),
+    U16(0x00002998), U16(0x00002f87), U16(0x00003600), U16(0x00003d09),
+    U16(0x000044a8), U16(0x00004ce3), U16(0x000055c0), U16(0x00005f45),
+    U16(0x00006978), U16(0x0000745f), U16(0x00008000), U16(0x00008c61),
+    U16(0x00009988), U16(0x0000a77b), U16(0x0000b640),
+};
+#elif SIZEOF_BDIGIT_DBL == 4
+static const int maxpow32_exp[35] = {
+    31, 20, 15, 13, 12, 11, 10, 10, 9, 9, 8, 8, 8, 8, 7, 7, 7, 7, 7, 7,
+    7, 7, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+};
+static const uint32_t maxpow32_num[35] = {
+    U32(0x80000000), U32(0xcfd41b91), U32(0x40000000), U32(0x48c27395),
+    U32(0x81bf1000), U32(0x75db9c97), U32(0x40000000), U32(0xcfd41b91),
+    U32(0x3b9aca00), U32(0x8c8b6d2b), U32(0x19a10000), U32(0x309f1021),
+    U32(0x57f6c100), U32(0x98c29b81), U32(0x10000000), U32(0x18754571),
+    U32(0x247dbc80), U32(0x3547667b), U32(0x4c4b4000), U32(0x6b5a6e1d),
+    U32(0x94ace180), U32(0xcaf18367), U32(0x0b640000), U32(0x0e8d4a51),
+    U32(0x1269ae40), U32(0x17179149), U32(0x1cb91000), U32(0x23744899),
+    U32(0x2b73a840), U32(0x34e63b41), U32(0x40000000), U32(0x4cfa3cc1),
+    U32(0x5c13d840), U32(0x6d91b519), U32(0x81bf1000),
+};
+#elif SIZEOF_BDIGIT_DBL == 8 && defined HAVE_UINT64_T
+static const int maxpow64_exp[35] = {
+    63, 40, 31, 27, 24, 22, 21, 20, 19, 18, 17, 17, 16, 16, 15, 15, 15,
+    15, 14, 14, 14, 14, 13, 13, 13, 13, 13, 13, 13, 12, 12, 12, 12, 12,
+    12,
+};
+static const uint64_t maxpow64_num[35] = {
+    U64(0x80000000,0x00000000), U64(0xa8b8b452,0x291fe821),
+    U64(0x40000000,0x00000000), U64(0x6765c793,0xfa10079d),
+    U64(0x41c21cb8,0xe1000000), U64(0x36427987,0x50226111),
+    U64(0x80000000,0x00000000), U64(0xa8b8b452,0x291fe821),
+    U64(0x8ac72304,0x89e80000), U64(0x4d28cb56,0xc33fa539),
+    U64(0x1eca170c,0x00000000), U64(0x780c7372,0x621bd74d),
+    U64(0x1e39a505,0x7d810000), U64(0x5b27ac99,0x3df97701),
+    U64(0x10000000,0x00000000), U64(0x27b95e99,0x7e21d9f1),
+    U64(0x5da0e1e5,0x3c5c8000), U64(0xd2ae3299,0xc1c4aedb),
+    U64(0x16bcc41e,0x90000000), U64(0x2d04b7fd,0xd9c0ef49),
+    U64(0x5658597b,0xcaa24000), U64(0xa0e20737,0x37609371),
+    U64(0x0c29e980,0x00000000), U64(0x14adf4b7,0x320334b9),
+    U64(0x226ed364,0x78bfa000), U64(0x383d9170,0xb85ff80b),
+    U64(0x5a3c23e3,0x9c000000), U64(0x8e651373,0x88122bcd),
+    U64(0xdd41bb36,0xd259e000), U64(0x0aee5720,0xee830681),
+    U64(0x10000000,0x00000000), U64(0x172588ad,0x4f5f0981),
+    U64(0x211e44f7,0xd02c1000), U64(0x2ee56725,0xf06e5c71),
+    U64(0x41c21cb8,0xe1000000),
+};
+#elif SIZEOF_BDIGIT_DBL == 16 && defined HAVE_UINT128_T
+static const int maxpow128_exp[35] = {
+    127, 80, 63, 55, 49, 45, 42, 40, 38, 37, 35, 34, 33, 32, 31, 31, 30,
+    30, 29, 29, 28, 28, 27, 27, 27, 26, 26, 26, 26, 25, 25, 25, 25, 24,
+    24,
+};
+static const uint128_t maxpow128_num[35] = {
+    U128(0x80000000,0x00000000,0x00000000,0x00000000),
+    U128(0x6f32f1ef,0x8b18a2bc,0x3cea5978,0x9c79d441),
+    U128(0x40000000,0x00000000,0x00000000,0x00000000),
+    U128(0xd0cf4b50,0xcfe20765,0xfff4b4e3,0xf741cf6d),
+    U128(0x6558e2a0,0x921fe069,0x42860000,0x00000000),
+    U128(0x5080c7b7,0xd0e31ba7,0x5911a67d,0xdd3d35e7),
+    U128(0x40000000,0x00000000,0x00000000,0x00000000),
+    U128(0x6f32f1ef,0x8b18a2bc,0x3cea5978,0x9c79d441),
+    U128(0x4b3b4ca8,0x5a86c47a,0x098a2240,0x00000000),
+    U128(0xffd1390a,0x0adc2fb8,0xdabbb817,0x4d95c99b),
+    U128(0x2c6fdb36,0x4c25e6c0,0x00000000,0x00000000),
+    U128(0x384bacd6,0x42c343b4,0xe90c4272,0x13506d29),
+    U128(0x31f5db32,0xa34aced6,0x0bf13a0e,0x00000000),
+    U128(0x20753ada,0xfd1e839f,0x53686d01,0x3143ee01),
+    U128(0x10000000,0x00000000,0x00000000,0x00000000),
+    U128(0x68ca11d6,0xb4f6d1d1,0xfaa82667,0x8073c2f1),
+    U128(0x223e493b,0xb3bb69ff,0xa4b87d6c,0x40000000),
+    U128(0xad62418d,0x14ea8247,0x01c4b488,0x6cc66f59),
+    U128(0x2863c1f5,0xcdae42f9,0x54000000,0x00000000),
+    U128(0xa63fd833,0xb9386b07,0x36039e82,0xbe651b25),
+    U128(0x1d1f7a9c,0xd087a14d,0x28cdf3d5,0x10000000),
+    U128(0x651b5095,0xc2ea8fc1,0xb30e2c57,0x77aaf7e1),
+    U128(0x0ddef20e,0xff760000,0x00000000,0x00000000),
+    U128(0x29c30f10,0x29939b14,0x6664242d,0x97d9f649),
+    U128(0x786a435a,0xe9558b0e,0x6aaf6d63,0xa8000000),
+    U128(0x0c5afe6f,0xf302bcbf,0x94fd9829,0xd87f5079),
+    U128(0x1fce575c,0xe1692706,0x07100000,0x00000000),
+    U128(0x4f34497c,0x8597e144,0x36e91802,0x00528229),
+    U128(0xbf3a8e1d,0x41ef2170,0x7802130d,0x84000000),
+    U128(0x0e7819e1,0x7f1eb0fb,0x6ee4fb89,0x01d9531f),
+    U128(0x20000000,0x00000000,0x00000000,0x00000000),
+    U128(0x4510460d,0xd9e879c0,0x14a82375,0x2f22b321),
+    U128(0x91abce3c,0x4b4117ad,0xe76d35db,0x22000000),
+    U128(0x08973ea3,0x55d75bc2,0x2e42c391,0x727d69e1),
+    U128(0x10e425c5,0x6daffabc,0x35c10000,0x00000000),
+};
+#endif
+
+static BDIGIT_DBL
+maxpow_in_bdigit_dbl(int base, int *exp_ret)
+{
+    BDIGIT_DBL maxpow;
+    int exponent;
+
+    assert(2 <= base && base <= 36);
+
+    {
+#if SIZEOF_BDIGIT_DBL == 2
+        maxpow = maxpow16_num[base-2];
+        exponent = maxpow16_exp[base-2];
+#elif SIZEOF_BDIGIT_DBL == 4
+        maxpow = maxpow32_num[base-2];
+        exponent = maxpow32_exp[base-2];
+#elif SIZEOF_BDIGIT_DBL == 8 && defined HAVE_UINT64_T
+        maxpow = maxpow64_num[base-2];
+        exponent = maxpow64_exp[base-2];
+#elif SIZEOF_BDIGIT_DBL == 16 && defined HAVE_UINT128_T
+        maxpow = maxpow128_num[base-2];
+        exponent = maxpow128_exp[base-2];
+#else
+        maxpow = base;
+        exponent = 1;
+        while (maxpow <= BDIGIT_DBL_MAX / base) {
+            maxpow *= base;
+            exponent++;
+        }
+#endif
+    }
+
+    *exp_ret = exponent;
+    return maxpow;
+}
+
+static inline BDIGIT_DBL
+bary2bdigitdbl(const BDIGIT *ds, size_t n)
+{
+    assert(n <= 2);
+
+    if (n == 2)
+        return ds[0] | BIGUP(ds[1]);
+    if (n == 1)
+        return ds[0];
+    return 0;
+}
+
+static inline void
+bdigitdbl2bary(BDIGIT *ds, size_t n, BDIGIT_DBL num)
+{
+    assert(n == 2);
+
+    ds[0] = BIGLO(num);
+    ds[1] = (BDIGIT)BIGDN(num);
+}
+
+static int
+bary_cmp(const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    size_t i;
+    BARY_TRUNC(xds, xn);
+    BARY_TRUNC(yds, yn);
+
+    if (xn < yn)
+        return -1;
+    if (xn > yn)
+        return 1;
+
+    for (i = 0; i < xn; i++)
+        if (xds[xn - i - 1] != yds[yn - i - 1])
+            break;
+    if (i == xn)
+        return 0;
+    return xds[xn - i - 1] < yds[yn - i - 1] ? -1 : 1;
+}
+
+static BDIGIT
+bary_small_lshift(BDIGIT *zds, const BDIGIT *xds, size_t n, int shift)
+{
+    size_t i;
+    BDIGIT_DBL num = 0;
+    assert(0 <= shift && shift < BITSPERDIG);
+
+    for (i=0; i<n; i++) {
+	num = num | (BDIGIT_DBL)*xds++ << shift;
+	*zds++ = BIGLO(num);
+	num = BIGDN(num);
+    }
+    return BIGLO(num);
+}
+
+static void
+bary_small_rshift(BDIGIT *zds, const BDIGIT *xds, size_t n, int shift, BDIGIT higher_bdigit)
+{
+    size_t i;
+    BDIGIT_DBL num = 0;
+
+    assert(0 <= shift && shift < BITSPERDIG);
+
+    num = BIGUP(higher_bdigit);
+    for (i = 0; i < n; i++) {
+        BDIGIT x = xds[n - i - 1];
+	num = (num | x) >> shift;
+        zds[n - i - 1] = BIGLO(num);
+	num = BIGUP(x);
+    }
+}
+
+static int
+bary_zero_p(const BDIGIT *xds, size_t xn)
+{
+    if (xn == 0)
+        return 1;
+    do {
+	if (xds[--xn]) return 0;
+    } while (xn);
+    return 1;
+}
+
+static void
+bary_neg(BDIGIT *ds, size_t n)
+{
+    size_t i;
+    for (i = 0; i < n; i++)
+        ds[n - i - 1] = BIGLO(~ds[n - i - 1]);
+}
+
+static int
+bary_2comp(BDIGIT *ds, size_t n)
+{
+    size_t i;
+    i = 0;
+    for (i = 0; i < n; i++) {
+        if (ds[i] != 0) {
+            goto non_zero;
+        }
+    }
+    return 1;
+
+  non_zero:
+    ds[i] = BIGLO(~ds[i] + 1);
+    i++;
+    for (; i < n; i++) {
+        ds[i] = BIGLO(~ds[i]);
+    }
+    return 0;
+}
+
+static void
+bary_swap(BDIGIT *ds, size_t num_bdigits)
+{
+    BDIGIT *p1 = ds;
+    BDIGIT *p2 = ds + num_bdigits - 1;
+    for (; p1 < p2; p1++, p2--) {
+        BDIGIT tmp = *p1;
+        *p1 = *p2;
+        *p2 = tmp;
+    }
+}
+
+#define INTEGER_PACK_WORDORDER_MASK \
+    (INTEGER_PACK_MSWORD_FIRST | \
+     INTEGER_PACK_LSWORD_FIRST)
+#define INTEGER_PACK_BYTEORDER_MASK \
+    (INTEGER_PACK_MSBYTE_FIRST | \
+     INTEGER_PACK_LSBYTE_FIRST | \
+     INTEGER_PACK_NATIVE_BYTE_ORDER)
+
+static void
+validate_integer_pack_format(size_t numwords, size_t wordsize, size_t nails, int flags, int supported_flags)
+{
+    int wordorder_bits = flags & INTEGER_PACK_WORDORDER_MASK;
+    int byteorder_bits = flags & INTEGER_PACK_BYTEORDER_MASK;
+
+    if (flags & ~supported_flags) {
+        rb_raise(rb_eArgError, "unsupported flags specified");
+    }
+    if (wordorder_bits == 0) {
+        if (1 < numwords)
+            rb_raise(rb_eArgError, "word order not specified");
+    }
+    else if (wordorder_bits != INTEGER_PACK_MSWORD_FIRST &&
+        wordorder_bits != INTEGER_PACK_LSWORD_FIRST)
+        rb_raise(rb_eArgError, "unexpected word order");
+    if (byteorder_bits == 0) {
+        rb_raise(rb_eArgError, "byte order not specified");
+    }
+    else if (byteorder_bits != INTEGER_PACK_MSBYTE_FIRST &&
+        byteorder_bits != INTEGER_PACK_LSBYTE_FIRST &&
+        byteorder_bits != INTEGER_PACK_NATIVE_BYTE_ORDER)
+        rb_raise(rb_eArgError, "unexpected byte order");
+    if (wordsize == 0)
+        rb_raise(rb_eArgError, "invalid wordsize: %"PRI_SIZE_PREFIX"u", wordsize);
+    if (SSIZE_MAX < wordsize)
+        rb_raise(rb_eArgError, "too big wordsize: %"PRI_SIZE_PREFIX"u", wordsize);
+    if (wordsize <= nails / CHAR_BIT)
+        rb_raise(rb_eArgError, "too big nails: %"PRI_SIZE_PREFIX"u", nails);
+    if (SIZE_MAX / wordsize < numwords)
+        rb_raise(rb_eArgError, "too big numwords * wordsize: %"PRI_SIZE_PREFIX"u * %"PRI_SIZE_PREFIX"u", numwords, wordsize);
+}
+
+static void
+integer_pack_loop_setup(
+    size_t numwords, size_t wordsize, size_t nails, int flags,
+    size_t *word_num_fullbytes_ret,
+    int *word_num_partialbits_ret,
+    size_t *word_start_ret,
+    ssize_t *word_step_ret,
+    size_t *word_last_ret,
+    size_t *byte_start_ret,
+    int *byte_step_ret)
+{
+    int wordorder_bits = flags & INTEGER_PACK_WORDORDER_MASK;
+    int byteorder_bits = flags & INTEGER_PACK_BYTEORDER_MASK;
+    size_t word_num_fullbytes;
+    int word_num_partialbits;
+    size_t word_start;
+    ssize_t word_step;
+    size_t word_last;
+    size_t byte_start;
+    int byte_step;
+
+    word_num_partialbits = CHAR_BIT - (int)(nails % CHAR_BIT);
+    if (word_num_partialbits == CHAR_BIT)
+        word_num_partialbits = 0;
+    word_num_fullbytes = wordsize - (nails / CHAR_BIT);
+    if (word_num_partialbits != 0) {
+        word_num_fullbytes--;
+    }
+
+    if (wordorder_bits == INTEGER_PACK_MSWORD_FIRST) {
+        word_start = wordsize*(numwords-1);
+        word_step = -(ssize_t)wordsize;
+        word_last = 0;
+    }
+    else {
+        word_start = 0;
+        word_step = wordsize;
+        word_last = wordsize*(numwords-1);
+    }
+
+    if (byteorder_bits == INTEGER_PACK_NATIVE_BYTE_ORDER) {
+#ifdef WORDS_BIGENDIAN
+        byteorder_bits = INTEGER_PACK_MSBYTE_FIRST;
+#else
+        byteorder_bits = INTEGER_PACK_LSBYTE_FIRST;
+#endif
+    }
+    if (byteorder_bits == INTEGER_PACK_MSBYTE_FIRST) {
+        byte_start = wordsize-1;
+        byte_step = -1;
+    }
+    else {
+        byte_start = 0;
+        byte_step = 1;
+    }
+
+    *word_num_partialbits_ret = word_num_partialbits;
+    *word_num_fullbytes_ret = word_num_fullbytes;
+    *word_start_ret = word_start;
+    *word_step_ret = word_step;
+    *word_last_ret = word_last;
+    *byte_start_ret = byte_start;
+    *byte_step_ret = byte_step;
+}
+
+static inline void
+integer_pack_fill_dd(BDIGIT **dpp, BDIGIT **dep, BDIGIT_DBL *ddp, int *numbits_in_dd_p)
+{
+    if (*dpp < *dep && BITSPERDIG <= (int)sizeof(*ddp) * CHAR_BIT - *numbits_in_dd_p) {
+        *ddp |= (BDIGIT_DBL)(*(*dpp)++) << *numbits_in_dd_p;
+        *numbits_in_dd_p += BITSPERDIG;
+    }
+    else if (*dpp == *dep) {
+        /* higher bits are infinity zeros */
+        *numbits_in_dd_p = (int)sizeof(*ddp) * CHAR_BIT;
+    }
+}
+
+static inline BDIGIT_DBL
+integer_pack_take_lowbits(int n, BDIGIT_DBL *ddp, int *numbits_in_dd_p)
+{
+    BDIGIT_DBL ret;
+    ret = (*ddp) & (((BDIGIT_DBL)1 << n) - 1);
+    *ddp >>= n;
+    *numbits_in_dd_p -= n;
+    return ret;
+}
+
+#if !defined(WORDS_BIGENDIAN)
+static int
+bytes_2comp(unsigned char *buf, size_t len)
+{
+    size_t i;
+    for (i = 0; i < len; i++) {
+        signed char c = buf[i];
+        signed int d = ~c;
+        unsigned int e = d & 0xFF;
+        buf[i] = e;
+    }
+    for (i = 0; i < len; i++) {
+        buf[i]++;
+        if (buf[i] != 0)
+            return 0;
+    }
+    return 1;
+}
+#endif
+
+static int
+bary_pack(int sign, BDIGIT *ds, size_t num_bdigits, void *words, size_t numwords, size_t wordsize, size_t nails, int flags)
+{
+    BDIGIT *dp, *de;
+    unsigned char *buf, *bufend;
+
+    dp = ds;
+    de = ds + num_bdigits;
+
+    validate_integer_pack_format(numwords, wordsize, nails, flags,
+            INTEGER_PACK_MSWORD_FIRST|
+            INTEGER_PACK_LSWORD_FIRST|
+            INTEGER_PACK_MSBYTE_FIRST|
+            INTEGER_PACK_LSBYTE_FIRST|
+            INTEGER_PACK_NATIVE_BYTE_ORDER|
+            INTEGER_PACK_2COMP|
+            INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION);
+
+    while (dp < de && de[-1] == 0)
+        de--;
+    if (dp == de) {
+        sign = 0;
+    }
+
+    if (!(flags & INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION)) {
+        if (sign == 0) {
+            MEMZERO(words, unsigned char, numwords * wordsize);
+            return 0;
+        }
+        if (nails == 0 && numwords == 1) {
+            int need_swap = wordsize != 1 &&
+                (flags & INTEGER_PACK_BYTEORDER_MASK) != INTEGER_PACK_NATIVE_BYTE_ORDER &&
+                ((flags & INTEGER_PACK_MSBYTE_FIRST) ? !HOST_BIGENDIAN_P : HOST_BIGENDIAN_P);
+            if (0 < sign || !(flags & INTEGER_PACK_2COMP)) {
+                BDIGIT d;
+                if (wordsize == 1) {
+                    *((unsigned char *)words) = (unsigned char)(d = dp[0]);
+                    return ((1 < de - dp || CLEAR_LOWBITS(d, 8) != 0) ? 2 : 1) * sign;
+                }
+#if defined(HAVE_UINT16_T) && 2 <= SIZEOF_BDIGIT
+                if (wordsize == 2 && (uintptr_t)words % RUBY_ALIGNOF(uint16_t) == 0) {
+                    uint16_t u = (uint16_t)(d = dp[0]);
+                    if (need_swap) u = swap16(u);
+                    *((uint16_t *)words) = u;
+                    return ((1 < de - dp || CLEAR_LOWBITS(d, 16) != 0) ? 2 : 1) * sign;
+                }
+#endif
+#if defined(HAVE_UINT32_T) && 4 <= SIZEOF_BDIGIT
+                if (wordsize == 4 && (uintptr_t)words % RUBY_ALIGNOF(uint32_t) == 0) {
+                    uint32_t u = (uint32_t)(d = dp[0]);
+                    if (need_swap) u = swap32(u);
+                    *((uint32_t *)words) = u;
+                    return ((1 < de - dp || CLEAR_LOWBITS(d, 32) != 0) ? 2 : 1) * sign;
+                }
+#endif
+#if defined(HAVE_UINT64_T) && 8 <= SIZEOF_BDIGIT
+                if (wordsize == 8 && (uintptr_t)words % RUBY_ALIGNOF(uint64_t) == 0) {
+                    uint64_t u = (uint64_t)(d = dp[0]);
+                    if (need_swap) u = swap64(u);
+                    *((uint64_t *)words) = u;
+                    return ((1 < de - dp || CLEAR_LOWBITS(d, 64) != 0) ? 2 : 1) * sign;
+                }
+#endif
+            }
+            else { /* sign < 0 && (flags & INTEGER_PACK_2COMP) */
+                BDIGIT_DBL_SIGNED d;
+                if (wordsize == 1) {
+                    *((unsigned char *)words) = (unsigned char)(d = -(BDIGIT_DBL_SIGNED)dp[0]);
+                    return (1 < de - dp || FILL_LOWBITS(d, 8) != -1) ? -2 : -1;
+                }
+#if defined(HAVE_UINT16_T) && 2 <= SIZEOF_BDIGIT
+                if (wordsize == 2 && (uintptr_t)words % RUBY_ALIGNOF(uint16_t) == 0) {
+                    uint16_t u = (uint16_t)(d = -(BDIGIT_DBL_SIGNED)dp[0]);
+                    if (need_swap) u = swap16(u);
+                    *((uint16_t *)words) = u;
+                    return (wordsize == SIZEOF_BDIGIT && de - dp == 2 && dp[1] == 1 && dp[0] == 0) ? -1 :
+                        (1 < de - dp || FILL_LOWBITS(d, 16) != -1) ? -2 : -1;
+                }
+#endif
+#if defined(HAVE_UINT32_T) && 4 <= SIZEOF_BDIGIT
+                if (wordsize == 4 && (uintptr_t)words % RUBY_ALIGNOF(uint32_t) == 0) {
+                    uint32_t u = (uint32_t)(d = -(BDIGIT_DBL_SIGNED)dp[0]);
+                    if (need_swap) u = swap32(u);
+                    *((uint32_t *)words) = u;
+                    return (wordsize == SIZEOF_BDIGIT && de - dp == 2 && dp[1] == 1 && dp[0] == 0) ? -1 :
+                        (1 < de - dp || FILL_LOWBITS(d, 32) != -1) ? -2 : -1;
+                }
+#endif
+#if defined(HAVE_UINT64_T) && 8 <= SIZEOF_BDIGIT
+                if (wordsize == 8 && (uintptr_t)words % RUBY_ALIGNOF(uint64_t) == 0) {
+                    uint64_t u = (uint64_t)(d = -(BDIGIT_DBL_SIGNED)dp[0]);
+                    if (need_swap) u = swap64(u);
+                    *((uint64_t *)words) = u;
+                    return (wordsize == SIZEOF_BDIGIT && de - dp == 2 && dp[1] == 1 && dp[0] == 0) ? -1 :
+                        (1 < de - dp || FILL_LOWBITS(d, 64) != -1) ? -2 : -1;
+                }
+#endif
+            }
+        }
+#if !defined(WORDS_BIGENDIAN)
+        if (nails == 0 && SIZEOF_BDIGIT == sizeof(BDIGIT) &&
+            (flags & INTEGER_PACK_WORDORDER_MASK) == INTEGER_PACK_LSWORD_FIRST &&
+            (flags & INTEGER_PACK_BYTEORDER_MASK) != INTEGER_PACK_MSBYTE_FIRST) {
+            size_t src_size = (de - dp) * SIZEOF_BDIGIT;
+            size_t dst_size = numwords * wordsize;
+            int overflow = 0;
+            while (0 < src_size && ((unsigned char *)ds)[src_size-1] == 0)
+                src_size--;
+            if (src_size <= dst_size) {
+                MEMCPY(words, dp, char, src_size);
+                MEMZERO((char*)words + src_size, char, dst_size - src_size);
+            }
+            else {
+                MEMCPY(words, dp, char, dst_size);
+                overflow = 1;
+            }
+            if (sign < 0 && (flags & INTEGER_PACK_2COMP)) {
+                int zero_p = bytes_2comp(words, dst_size);
+                if (zero_p && overflow) {
+                    unsigned char *p = (unsigned char *)dp;
+                    if (dst_size == src_size-1 &&
+                        p[dst_size] == 1) {
+                        overflow = 0;
+                    }
+                }
+            }
+            if (overflow)
+                sign *= 2;
+            return sign;
+        }
+#endif
+        if (nails == 0 && SIZEOF_BDIGIT == sizeof(BDIGIT) &&
+            wordsize % SIZEOF_BDIGIT == 0 && (uintptr_t)words % RUBY_ALIGNOF(BDIGIT) == 0) {
+            size_t bdigits_per_word = wordsize / SIZEOF_BDIGIT;
+            size_t src_num_bdigits = de - dp;
+            size_t dst_num_bdigits = numwords * bdigits_per_word;
+            int overflow = 0;
+            int mswordfirst_p = (flags & INTEGER_PACK_MSWORD_FIRST) != 0;
+            int msbytefirst_p = (flags & INTEGER_PACK_NATIVE_BYTE_ORDER) ? HOST_BIGENDIAN_P :
+                (flags & INTEGER_PACK_MSBYTE_FIRST) != 0;
+            if (src_num_bdigits <= dst_num_bdigits) {
+                MEMCPY(words, dp, BDIGIT, src_num_bdigits);
+                BDIGITS_ZERO((BDIGIT*)words + src_num_bdigits, dst_num_bdigits - src_num_bdigits);
+            }
+            else {
+                MEMCPY(words, dp, BDIGIT, dst_num_bdigits);
+                overflow = 1;
+            }
+            if (sign < 0 && (flags & INTEGER_PACK_2COMP)) {
+                int zero_p = bary_2comp(words, dst_num_bdigits);
+                if (zero_p && overflow &&
+                    dst_num_bdigits == src_num_bdigits-1 &&
+                    dp[dst_num_bdigits] == 1)
+                    overflow = 0;
+            }
+            if (msbytefirst_p != HOST_BIGENDIAN_P) {
+                size_t i;
+                for (i = 0; i < dst_num_bdigits; i++) {
+                    BDIGIT d = ((BDIGIT*)words)[i];
+                    ((BDIGIT*)words)[i] = swap_bdigit(d);
+                }
+            }
+            if (mswordfirst_p ? !msbytefirst_p : msbytefirst_p) {
+                size_t i;
+                BDIGIT *p = words;
+                for (i = 0; i < numwords; i++) {
+                    bary_swap(p, bdigits_per_word);
+                    p += bdigits_per_word;
+                }
+            }
+            if (mswordfirst_p) {
+                bary_swap(words, dst_num_bdigits);
+            }
+            if (overflow)
+                sign *= 2;
+            return sign;
+        }
+    }
+
+    buf = words;
+    bufend = buf + numwords * wordsize;
+
+    if (buf == bufend) {
+        /* overflow if non-zero*/
+        if (!(flags & INTEGER_PACK_2COMP) || 0 <= sign)
+            sign *= 2;
+        else {
+            if (de - dp == 1 && dp[0] == 1)
+                sign = -1; /* val == -1 == -2**(numwords*(wordsize*CHAR_BIT-nails)) */
+            else
+                sign = -2; /* val < -1 == -2**(numwords*(wordsize*CHAR_BIT-nails)) */
+        }
+    }
+    else if (dp == de) {
+        memset(buf, '\0', bufend - buf);
+    }
+    else if (dp < de && buf < bufend) {
+        int word_num_partialbits;
+        size_t word_num_fullbytes;
+
+        ssize_t word_step;
+        size_t byte_start;
+        int byte_step;
+
+        size_t word_start, word_last;
+        unsigned char *wordp, *last_wordp;
+        BDIGIT_DBL dd;
+        int numbits_in_dd;
+
+        integer_pack_loop_setup(numwords, wordsize, nails, flags,
+            &word_num_fullbytes, &word_num_partialbits,
+            &word_start, &word_step, &word_last, &byte_start, &byte_step);
+
+        wordp = buf + word_start;
+        last_wordp = buf + word_last;
+
+        dd = 0;
+        numbits_in_dd = 0;
+
+#define FILL_DD \
+    integer_pack_fill_dd(&dp, &de, &dd, &numbits_in_dd)
+#define TAKE_LOWBITS(n) \
+    integer_pack_take_lowbits(n, &dd, &numbits_in_dd)
+
+        while (1) {
+            size_t index_in_word = 0;
+            unsigned char *bytep = wordp + byte_start;
+            while (index_in_word < word_num_fullbytes) {
+                FILL_DD;
+                *bytep = TAKE_LOWBITS(CHAR_BIT);
+                bytep += byte_step;
+                index_in_word++;
+            }
+            if (word_num_partialbits) {
+                FILL_DD;
+                *bytep = TAKE_LOWBITS(word_num_partialbits);
+                bytep += byte_step;
+                index_in_word++;
+            }
+            while (index_in_word < wordsize) {
+                *bytep = 0;
+                bytep += byte_step;
+                index_in_word++;
+            }
+
+            if (wordp == last_wordp)
+                break;
+
+            wordp += word_step;
+        }
+        FILL_DD;
+        /* overflow tests */
+        if (dp != de || 1 < dd) {
+            /* 2**(numwords*(wordsize*CHAR_BIT-nails)+1) <= abs(val) */
+            sign *= 2;
+        }
+        else if (dd == 1) {
+            /* 2**(numwords*(wordsize*CHAR_BIT-nails)) <= abs(val) < 2**(numwords*(wordsize*CHAR_BIT-nails)+1) */
+            if (!(flags & INTEGER_PACK_2COMP) || 0 <= sign)
+                sign *= 2;
+            else { /* overflow_2comp && sign == -1 */
+                /* test lower bits are all zero. */
+                dp = ds;
+                while (dp < de && *dp == 0)
+                    dp++;
+                if (de - dp == 1 && /* only one non-zero word. */
+                    POW2_P(*dp)) /* *dp contains only one bit set. */
+                    sign = -1; /* val == -2**(numwords*(wordsize*CHAR_BIT-nails)) */
+                else
+                    sign = -2; /* val < -2**(numwords*(wordsize*CHAR_BIT-nails)) */
+            }
+        }
+    }
+
+    if ((flags & INTEGER_PACK_2COMP) && (sign < 0 && numwords != 0)) {
+        int word_num_partialbits;
+        size_t word_num_fullbytes;
+
+        ssize_t word_step;
+        size_t byte_start;
+        int byte_step;
+
+        size_t word_start, word_last;
+        unsigned char *wordp, *last_wordp;
+
+        unsigned int partialbits_mask;
+        int carry;
+
+        integer_pack_loop_setup(numwords, wordsize, nails, flags,
+            &word_num_fullbytes, &word_num_partialbits,
+            &word_start, &word_step, &word_last, &byte_start, &byte_step);
+
+        partialbits_mask = (1 << word_num_partialbits) - 1;
+
+        buf = words;
+        wordp = buf + word_start;
+        last_wordp = buf + word_last;
+
+        carry = 1;
+        while (1) {
+            size_t index_in_word = 0;
+            unsigned char *bytep = wordp + byte_start;
+            while (index_in_word < word_num_fullbytes) {
+                carry += (unsigned char)~*bytep;
+                *bytep = (unsigned char)carry;
+                carry >>= CHAR_BIT;
+                bytep += byte_step;
+                index_in_word++;
+            }
+            if (word_num_partialbits) {
+                carry += (*bytep & partialbits_mask) ^ partialbits_mask;
+                *bytep = carry & partialbits_mask;
+                carry >>= word_num_partialbits;
+                bytep += byte_step;
+                index_in_word++;
+            }
+
+            if (wordp == last_wordp)
+                break;
+
+            wordp += word_step;
+        }
+    }
+
+    return sign;
+#undef FILL_DD
+#undef TAKE_LOWBITS
+}
+
+static size_t
+integer_unpack_num_bdigits_small(size_t numwords, size_t wordsize, size_t nails, int *nlp_bits_ret)
+{
+    /* nlp_bits stands for number of leading padding bits */
+    size_t num_bits = (wordsize * CHAR_BIT - nails) * numwords;
+    size_t num_bdigits = (num_bits + BITSPERDIG - 1) / BITSPERDIG;
+    *nlp_bits_ret = (int)(num_bdigits * BITSPERDIG - num_bits);
+    return num_bdigits;
+}
+
+static size_t
+integer_unpack_num_bdigits_generic(size_t numwords, size_t wordsize, size_t nails, int *nlp_bits_ret)
+{
+    /* BITSPERDIG = SIZEOF_BDIGIT * CHAR_BIT */
+    /* num_bits = (wordsize * CHAR_BIT - nails) * numwords */
+    /* num_bdigits = (num_bits + BITSPERDIG - 1) / BITSPERDIG */
+
+    /* num_bits = CHAR_BIT * (wordsize * numwords) - nails * numwords = CHAR_BIT * num_bytes1 - nails * numwords */
+    size_t num_bytes1 = wordsize * numwords;
+
+    /* q1 * CHAR_BIT + r1 = numwords */
+    size_t q1 = numwords / CHAR_BIT;
+    size_t r1 = numwords % CHAR_BIT;
+
+    /* num_bits = CHAR_BIT * num_bytes1 - nails * (q1 * CHAR_BIT + r1) = CHAR_BIT * num_bytes2 - nails * r1 */
+    size_t num_bytes2 = num_bytes1 - nails * q1;
+
+    /* q2 * CHAR_BIT + r2 = nails */
+    size_t q2 = nails / CHAR_BIT;
+    size_t r2 = nails % CHAR_BIT;
+
+    /* num_bits = CHAR_BIT * num_bytes2 - (q2 * CHAR_BIT + r2) * r1 = CHAR_BIT * num_bytes3 - r1 * r2 */
+    size_t num_bytes3 = num_bytes2 - q2 * r1;
+
+    /* q3 * BITSPERDIG + r3 = num_bytes3 */
+    size_t q3 = num_bytes3 / BITSPERDIG;
+    size_t r3 = num_bytes3 % BITSPERDIG;
+
+    /* num_bits = CHAR_BIT * (q3 * BITSPERDIG + r3) - r1 * r2 = BITSPERDIG * num_digits1 + CHAR_BIT * r3 - r1 * r2 */
+    size_t num_digits1 = CHAR_BIT * q3;
+
+    /*
+     * if CHAR_BIT * r3 >= r1 * r2
+     *   CHAR_BIT * r3 - r1 * r2 = CHAR_BIT * BITSPERDIG - (CHAR_BIT * BITSPERDIG - (CHAR_BIT * r3 - r1 * r2))
+     *   q4 * BITSPERDIG + r4 = CHAR_BIT * BITSPERDIG - (CHAR_BIT * r3 - r1 * r2)
+     *   num_bits = BITSPERDIG * num_digits1 + CHAR_BIT * BITSPERDIG - (q4 * BITSPERDIG + r4) = BITSPERDIG * num_digits2 - r4
+     * else
+     *   q4 * BITSPERDIG + r4 = -(CHAR_BIT * r3 - r1 * r2)
+     *   num_bits = BITSPERDIG * num_digits1 - (q4 * BITSPERDIG + r4) = BITSPERDIG * num_digits2 - r4
+     * end
+     */
+
+    if (CHAR_BIT * r3 >= r1 * r2) {
+        size_t tmp1 = CHAR_BIT * BITSPERDIG - (CHAR_BIT * r3 - r1 * r2);
+        size_t q4 = tmp1 / BITSPERDIG;
+        int r4 = (int)(tmp1 % BITSPERDIG);
+        size_t num_digits2 = num_digits1 + CHAR_BIT - q4;
+        *nlp_bits_ret = r4;
+        return num_digits2;
+    }
+    else {
+        size_t tmp1 = r1 * r2 - CHAR_BIT * r3;
+        size_t q4 = tmp1 / BITSPERDIG;
+        int r4 = (int)(tmp1 % BITSPERDIG);
+        size_t num_digits2 = num_digits1 - q4;
+        *nlp_bits_ret = r4;
+        return num_digits2;
+    }
+}
+
+static size_t
+integer_unpack_num_bdigits(size_t numwords, size_t wordsize, size_t nails, int *nlp_bits_ret)
+{
+    size_t num_bdigits;
+
+    if (numwords <= (SIZE_MAX - (BITSPERDIG-1)) / CHAR_BIT / wordsize) {
+        num_bdigits = integer_unpack_num_bdigits_small(numwords, wordsize, nails, nlp_bits_ret);
+#ifdef DEBUG_INTEGER_PACK
+        {
+            int nlp_bits1;
+            size_t num_bdigits1 = integer_unpack_num_bdigits_generic(numwords, wordsize, nails, &nlp_bits1);
+            assert(num_bdigits == num_bdigits1);
+            assert(*nlp_bits_ret == nlp_bits1);
+        }
+#endif
+    }
+    else {
+        num_bdigits = integer_unpack_num_bdigits_generic(numwords, wordsize, nails, nlp_bits_ret);
+    }
+    return num_bdigits;
+}
+
+static inline void
+integer_unpack_push_bits(int data, int numbits, BDIGIT_DBL *ddp, int *numbits_in_dd_p, BDIGIT **dpp)
+{
+    (*ddp) |= ((BDIGIT_DBL)data) << (*numbits_in_dd_p);
+    *numbits_in_dd_p += numbits;
+    while (BITSPERDIG <= *numbits_in_dd_p) {
+        *(*dpp)++ = BIGLO(*ddp);
+        *ddp = BIGDN(*ddp);
+        *numbits_in_dd_p -= BITSPERDIG;
+    }
+}
+
+static int
+integer_unpack_single_bdigit(BDIGIT u, size_t size, int flags, BDIGIT *dp)
+{
+    int sign;
+    if (flags & INTEGER_PACK_2COMP) {
+        sign = (flags & INTEGER_PACK_NEGATIVE) ?
+            ((size == SIZEOF_BDIGIT && u == 0) ? -2 : -1) :
+            ((u >> (size * CHAR_BIT - 1)) ? -1 : 1);
+        if (sign < 0) {
+            u |= LSHIFTX(BDIGMAX, size * CHAR_BIT);
+            u = BIGLO(1 + ~u);
+        }
+    }
+    else
+        sign = (flags & INTEGER_PACK_NEGATIVE) ? -1 : 1;
+    *dp = u;
+    return sign;
+}
+
+#ifdef HAVE_BUILTIN___BUILTIN_ASSUME_ALIGNED
+#define reinterpret_cast(type, value) (type) \
+    __builtin_assume_aligned((value), sizeof(*(type)NULL));
+#else
+#define reinterpret_cast(type, value) (type)value
+#endif
+
+static int
+bary_unpack_internal(BDIGIT *bdigits, size_t num_bdigits, const void *words, size_t numwords, size_t wordsize, size_t nails, int flags, int nlp_bits)
+{
     int sign;
+    const unsigned char *buf = words;
+    BDIGIT *dp;
+    BDIGIT *de;
+
+    dp = bdigits;
+    de = dp + num_bdigits;
+
+    if (!(flags & INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION)) {
+        if (nails == 0 && numwords == 1) {
+            int need_swap = wordsize != 1 &&
+                (flags & INTEGER_PACK_BYTEORDER_MASK) != INTEGER_PACK_NATIVE_BYTE_ORDER &&
+                ((flags & INTEGER_PACK_MSBYTE_FIRST) ? !HOST_BIGENDIAN_P : HOST_BIGENDIAN_P);
+            if (wordsize == 1) {
+                return integer_unpack_single_bdigit(*(uint8_t *)buf, sizeof(uint8_t), flags, dp);
+            }
+#if defined(HAVE_UINT16_T) && 2 <= SIZEOF_BDIGIT
+            if (wordsize == 2 && (uintptr_t)words % RUBY_ALIGNOF(uint16_t) == 0) {
+                uint16_t u = *reinterpret_cast(const uint16_t *, buf);
+                return integer_unpack_single_bdigit(need_swap ? swap16(u) : u, sizeof(uint16_t), flags, dp);
+            }
+#endif
+#if defined(HAVE_UINT32_T) && 4 <= SIZEOF_BDIGIT
+            if (wordsize == 4 && (uintptr_t)words % RUBY_ALIGNOF(uint32_t) == 0) {
+                uint32_t u = *reinterpret_cast(const uint32_t *, buf);
+                return integer_unpack_single_bdigit(need_swap ? swap32(u) : u, sizeof(uint32_t), flags, dp);
+            }
+#endif
+#if defined(HAVE_UINT64_T) && 8 <= SIZEOF_BDIGIT
+            if (wordsize == 8 && (uintptr_t)words % RUBY_ALIGNOF(uint64_t) == 0) {
+                uint64_t u = *reinterpret_cast(const uint64_t *, buf);
+                return integer_unpack_single_bdigit(need_swap ? swap64(u) : u, sizeof(uint64_t), flags, dp);
+            }
+#endif
+#undef reinterpret_cast
+        }
+#if !defined(WORDS_BIGENDIAN)
+        if (nails == 0 && SIZEOF_BDIGIT == sizeof(BDIGIT) &&
+            (flags & INTEGER_PACK_WORDORDER_MASK) == INTEGER_PACK_LSWORD_FIRST &&
+            (flags & INTEGER_PACK_BYTEORDER_MASK) != INTEGER_PACK_MSBYTE_FIRST) {
+            size_t src_size = numwords * wordsize;
+            size_t dst_size = num_bdigits * SIZEOF_BDIGIT;
+            MEMCPY(dp, words, char, src_size);
+            if (flags & INTEGER_PACK_2COMP) {
+                if (flags & INTEGER_PACK_NEGATIVE) {
+                    int zero_p;
+                    memset((char*)dp + src_size, 0xff, dst_size - src_size);
+                    zero_p = bary_2comp(dp, num_bdigits);
+                    sign = zero_p ? -2 : -1;
+                }
+                else if (buf[src_size-1] >> (CHAR_BIT-1)) {
+                    memset((char*)dp + src_size, 0xff, dst_size - src_size);
+                    bary_2comp(dp, num_bdigits);
+                    sign = -1;
+                }
+                else {
+                    MEMZERO((char*)dp + src_size, char, dst_size - src_size);
+                    sign = 1;
+                }
+            }
+            else {
+                MEMZERO((char*)dp + src_size, char, dst_size - src_size);
+                sign = (flags & INTEGER_PACK_NEGATIVE) ? -1 : 1;
+            }
+            return sign;
+        }
+#endif
+        if (nails == 0 && SIZEOF_BDIGIT == sizeof(BDIGIT) &&
+            wordsize % SIZEOF_BDIGIT == 0) {
+            size_t bdigits_per_word = wordsize / SIZEOF_BDIGIT;
+            int mswordfirst_p = (flags & INTEGER_PACK_MSWORD_FIRST) != 0;
+            int msbytefirst_p = (flags & INTEGER_PACK_NATIVE_BYTE_ORDER) ? HOST_BIGENDIAN_P :
+                (flags & INTEGER_PACK_MSBYTE_FIRST) != 0;
+            MEMCPY(dp, words, BDIGIT, numwords*bdigits_per_word);
+            if (mswordfirst_p) {
+                bary_swap(dp, num_bdigits);
+            }
+            if (mswordfirst_p ? !msbytefirst_p : msbytefirst_p) {
+                size_t i;
+                BDIGIT *p = dp;
+                for (i = 0; i < numwords; i++) {
+                    bary_swap(p, bdigits_per_word);
+                    p += bdigits_per_word;
+                }
+            }
+            if (msbytefirst_p != HOST_BIGENDIAN_P) {
+                BDIGIT *p;
+                for (p = dp; p < de; p++) {
+                    BDIGIT d = *p;
+                    *p = swap_bdigit(d);
+                }
+            }
+            if (flags & INTEGER_PACK_2COMP) {
+                if (flags & INTEGER_PACK_NEGATIVE) {
+                    int zero_p = bary_2comp(dp, num_bdigits);
+                    sign = zero_p ? -2 : -1;
+                }
+                else if (BDIGIT_MSB(de[-1])) {
+                    bary_2comp(dp, num_bdigits);
+                    sign = -1;
+                }
+                else {
+                    sign = 1;
+                }
+            }
+            else {
+                sign = (flags & INTEGER_PACK_NEGATIVE) ? -1 : 1;
+            }
+            return sign;
+        }
+    }
+
+    if (num_bdigits != 0) {
+        int word_num_partialbits;
+        size_t word_num_fullbytes;
+
+        ssize_t word_step;
+        size_t byte_start;
+        int byte_step;
+
+        size_t word_start, word_last;
+        const unsigned char *wordp, *last_wordp;
+        BDIGIT_DBL dd;
+        int numbits_in_dd;
+
+        integer_pack_loop_setup(numwords, wordsize, nails, flags,
+            &word_num_fullbytes, &word_num_partialbits,
+            &word_start, &word_step, &word_last, &byte_start, &byte_step);
+
+        wordp = buf + word_start;
+        last_wordp = buf + word_last;
+
+        dd = 0;
+        numbits_in_dd = 0;
+
+#define PUSH_BITS(data, numbits) \
+        integer_unpack_push_bits(data, numbits, &dd, &numbits_in_dd, &dp)
+
+        while (1) {
+            size_t index_in_word = 0;
+            const unsigned char *bytep = wordp + byte_start;
+            while (index_in_word < word_num_fullbytes) {
+                PUSH_BITS(*bytep, CHAR_BIT);
+                bytep += byte_step;
+                index_in_word++;
+            }
+            if (word_num_partialbits) {
+                PUSH_BITS(*bytep & ((1 << word_num_partialbits) - 1), word_num_partialbits);
+                bytep += byte_step;
+                index_in_word++;
+            }
+
+            if (wordp == last_wordp)
+                break;
+
+            wordp += word_step;
+        }
+        if (dd)
+            *dp++ = (BDIGIT)dd;
+        assert(dp <= de);
+        while (dp < de)
+            *dp++ = 0;
+#undef PUSH_BITS
+    }
+
+    if (!(flags & INTEGER_PACK_2COMP)) {
+        sign = (flags & INTEGER_PACK_NEGATIVE) ? -1 : 1;
+    }
+    else {
+        if (nlp_bits) {
+            if ((flags & INTEGER_PACK_NEGATIVE) ||
+                (bdigits[num_bdigits-1] >> (BITSPERDIG - nlp_bits - 1))) {
+                bdigits[num_bdigits-1] |= BIGLO(BDIGMAX << (BITSPERDIG - nlp_bits));
+                sign = -1;
+            }
+            else {
+                sign = 1;
+            }
+        }
+        else {
+            if (flags & INTEGER_PACK_NEGATIVE) {
+                sign = bary_zero_p(bdigits, num_bdigits) ? -2 : -1;
+            }
+            else {
+                if (num_bdigits != 0 && BDIGIT_MSB(bdigits[num_bdigits-1]))
+                    sign = -1;
+                else
+                    sign = 1;
+            }
+        }
+        if (sign == -1 && num_bdigits != 0) {
+            bary_2comp(bdigits, num_bdigits);
+        }
+    }
+
+    return sign;
+}
+
+static void
+bary_unpack(BDIGIT *bdigits, size_t num_bdigits, const void *words, size_t numwords, size_t wordsize, size_t nails, int flags)
 {
-    NEWOBJ(big, struct RBignum);
-    OBJSETUP(big, klass, T_BIGNUM);
-    big->sign = sign?1:0;
-    big->len = len;
-    big->digits = ALLOC_N(BDIGIT, len);
+    size_t num_bdigits0;
+    int nlp_bits;
+    int sign;
 
-    return (VALUE)big;
+    validate_integer_pack_format(numwords, wordsize, nails, flags,
+            INTEGER_PACK_MSWORD_FIRST|
+            INTEGER_PACK_LSWORD_FIRST|
+            INTEGER_PACK_MSBYTE_FIRST|
+            INTEGER_PACK_LSBYTE_FIRST|
+            INTEGER_PACK_NATIVE_BYTE_ORDER|
+            INTEGER_PACK_2COMP|
+            INTEGER_PACK_FORCE_BIGNUM|
+            INTEGER_PACK_NEGATIVE|
+            INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION);
+
+    num_bdigits0 = integer_unpack_num_bdigits(numwords, wordsize, nails, &nlp_bits);
+
+    assert(num_bdigits0 <= num_bdigits);
+
+    sign = bary_unpack_internal(bdigits, num_bdigits0, words, numwords, wordsize, nails, flags, nlp_bits);
+
+    if (num_bdigits0 < num_bdigits) {
+        BDIGITS_ZERO(bdigits + num_bdigits0, num_bdigits - num_bdigits0);
+        if (sign == -2) {
+            bdigits[num_bdigits0] = 1;
+        }
+    }
+}
+
+static int
+bary_subb(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, int borrow)
+{
+    BDIGIT_DBL_SIGNED num;
+    size_t i;
+    size_t sn;
+
+    assert(xn <= zn);
+    assert(yn <= zn);
+
+    sn = xn < yn ? xn : yn;
+
+    num = borrow ? -1 : 0;
+    for (i = 0; i < sn; i++) {
+	num += (BDIGIT_DBL_SIGNED)xds[i] - yds[i];
+	zds[i] = BIGLO(num);
+	num = BIGDN(num);
+    }
+    if (yn <= xn) {
+        for (; i < xn; i++) {
+            if (num == 0) goto num_is_zero;
+            num += xds[i];
+            zds[i] = BIGLO(num);
+            num = BIGDN(num);
+        }
+    }
+    else {
+        for (; i < yn; i++) {
+            num -= yds[i];
+            zds[i] = BIGLO(num);
+            num = BIGDN(num);
+        }
+    }
+    if (num == 0) goto num_is_zero;
+    for (; i < zn; i++) {
+	zds[i] = BDIGMAX;
+    }
+    return 1;
+
+  num_is_zero:
+    if (xds == zds && xn == zn)
+        return 0;
+    for (; i < xn; i++) {
+	zds[i] = xds[i];
+    }
+    for (; i < zn; i++) {
+	zds[i] = 0;
+    }
+    return 0;
+}
+
+static int
+bary_sub(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    return bary_subb(zds, zn, xds, xn, yds, yn, 0);
+}
+
+static int
+bary_sub_one(BDIGIT *zds, size_t zn)
+{
+    return bary_subb(zds, zn, zds, zn, NULL, 0, 1);
+}
+
+static int
+bary_addc(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, int carry)
+{
+    BDIGIT_DBL num;
+    size_t i;
+
+    assert(xn <= zn);
+    assert(yn <= zn);
+
+    if (xn > yn) {
+	const BDIGIT *tds;
+	tds = xds; xds = yds; yds = tds;
+	i = xn; xn = yn; yn = i;
+    }
+
+    num = carry ? 1 : 0;
+    for (i = 0; i < xn; i++) {
+	num += (BDIGIT_DBL)xds[i] + yds[i];
+	zds[i] = BIGLO(num);
+	num = BIGDN(num);
+    }
+    for (; i < yn; i++) {
+        if (num == 0) goto num_is_zero;
+	num += yds[i];
+	zds[i] = BIGLO(num);
+	num = BIGDN(num);
+    }
+    for (; i < zn; i++) {
+        if (num == 0) goto num_is_zero;
+	zds[i] = BIGLO(num);
+	num = BIGDN(num);
+    }
+    return num != 0;
+
+  num_is_zero:
+    if (yds == zds && yn == zn)
+        return 0;
+    for (; i < yn; i++) {
+	zds[i] = yds[i];
+    }
+    for (; i < zn; i++) {
+	zds[i] = 0;
+    }
+    return 0;
+}
+
+static int
+bary_add(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    return bary_addc(zds, zn, xds, xn, yds, yn, 0);
+}
+
+static int
+bary_add_one(BDIGIT *ds, size_t n)
+{
+    size_t i;
+    for (i = 0; i < n; i++) {
+        BDIGIT_DBL n = ds[i];
+        n += 1;
+        ds[i] = BIGLO(n);
+        if (ds[i] != 0)
+            return 0;
+    }
+    return 1;
+}
+
+static void
+bary_mul_single(BDIGIT *zds, size_t zn, BDIGIT x, BDIGIT y)
+{
+    BDIGIT_DBL n;
+
+    assert(2 <= zn);
+
+    n = (BDIGIT_DBL)x * y;
+    bdigitdbl2bary(zds, 2, n);
+    BDIGITS_ZERO(zds + 2, zn - 2);
+}
+
+static int
+bary_muladd_1xN(BDIGIT *zds, size_t zn, BDIGIT x, const BDIGIT *yds, size_t yn)
+{
+    BDIGIT_DBL n;
+    BDIGIT_DBL dd;
+    size_t j;
+
+    assert(zn > yn);
+
+    if (x == 0)
+        return 0;
+    dd = x;
+    n = 0;
+    for (j = 0; j < yn; j++) {
+        BDIGIT_DBL ee = n + dd * yds[j];
+        if (ee) {
+            n = zds[j] + ee;
+            zds[j] = BIGLO(n);
+            n = BIGDN(n);
+        }
+        else {
+            n = 0;
+        }
+
+    }
+    for (; j < zn; j++) {
+        if (n == 0)
+            break;
+        n += zds[j];
+        zds[j] = BIGLO(n);
+        n = BIGDN(n);
+    }
+    return n != 0;
+}
+
+static BDIGIT_DBL_SIGNED
+bigdivrem_mulsub(BDIGIT *zds, size_t zn, BDIGIT x, const BDIGIT *yds, size_t yn)
+{
+    size_t i;
+    BDIGIT_DBL t2;
+    BDIGIT_DBL_SIGNED num;
+
+    assert(zn == yn + 1);
+
+    num = 0;
+    t2 = 0;
+    i = 0;
+
+    do {
+        BDIGIT_DBL_SIGNED ee;
+        t2 += (BDIGIT_DBL)yds[i] * x;
+        ee = num - BIGLO(t2);
+        num = (BDIGIT_DBL_SIGNED)zds[i] + ee;
+        if (ee) zds[i] = BIGLO(num);
+        num = BIGDN(num);
+        t2 = BIGDN(t2);
+    } while (++i < yn);
+    num -= (BDIGIT_DBL_SIGNED)t2;
+    num += (BDIGIT_DBL_SIGNED)zds[yn]; /* borrow from high digit; don't update */
+    return num;
+}
+
+static int
+bary_mulsub_1xN(BDIGIT *zds, size_t zn, BDIGIT x, const BDIGIT *yds, size_t yn)
+{
+    BDIGIT_DBL_SIGNED num;
+
+    assert(zn == yn + 1);
+
+    num = bigdivrem_mulsub(zds, zn, x, yds, yn);
+    zds[yn] = BIGLO(num);
+    if (BIGDN(num))
+        return 1;
+    return 0;
 }
 
-#define bignew(len,sign) bignew_1(rb_cBignum,len,sign)
+static void
+bary_mul_normal(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    size_t i;
+
+    assert(xn + yn <= zn);
+
+    BDIGITS_ZERO(zds, zn);
+    for (i = 0; i < xn; i++) {
+        bary_muladd_1xN(zds+i, zn-i, xds[i], yds, yn);
+    }
+}
 
 VALUE
-rb_big_clone(x)
-    VALUE x;
+rb_big_mul_normal(VALUE x, VALUE y)
+{
+    size_t xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y), zn = xn + yn;
+    VALUE z = bignew(zn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+    bary_mul_normal(BDIGITS(z), zn, BDIGITS(x), xn, BDIGITS(y), yn);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+    return z;
+}
+
+/* efficient squaring (2 times faster than normal multiplication)
+ * ref: Handbook of Applied Cryptography, Algorithm 14.16
+ *      http://www.cacr.math.uwaterloo.ca/hac/about/chap14.pdf
+ */
+static void
+bary_sq_fast(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn)
 {
-    VALUE z = bignew_1(CLASS_OF(x), RBIGNUM(x)->len, RBIGNUM(x)->sign);
+    size_t i, j;
+    BDIGIT_DBL c, v, w;
+    BDIGIT vl;
+    int vh;
+
+    assert(xn * 2 <= zn);
+
+    BDIGITS_ZERO(zds, zn);
+
+    if (xn == 0)
+        return;
+
+    for (i = 0; i < xn-1; i++) {
+	v = (BDIGIT_DBL)xds[i];
+	if (!v)
+            continue;
+	c = (BDIGIT_DBL)zds[i + i] + v * v;
+	zds[i + i] = BIGLO(c);
+	c = BIGDN(c);
+	v *= 2;
+        vl = BIGLO(v);
+        vh = (int)BIGDN(v);
+	for (j = i + 1; j < xn; j++) {
+	    w = (BDIGIT_DBL)xds[j];
+	    c += (BDIGIT_DBL)zds[i + j] + vl * w;
+	    zds[i + j] = BIGLO(c);
+	    c = BIGDN(c);
+	    if (vh)
+                c += w;
+	}
+	if (c) {
+	    c += (BDIGIT_DBL)zds[i + xn];
+	    zds[i + xn] = BIGLO(c);
+	    c = BIGDN(c);
+            if (c)
+                zds[i + xn + 1] += (BDIGIT)c;
+	}
+    }
 
-    MEMCPY(BDIGITS(z), BDIGITS(x), BDIGIT, RBIGNUM(x)->len);
+    /* i == xn-1 */
+    v = (BDIGIT_DBL)xds[i];
+    if (!v)
+        return;
+    c = (BDIGIT_DBL)zds[i + i] + v * v;
+    zds[i + i] = BIGLO(c);
+    c = BIGDN(c);
+    if (c) {
+        zds[i + xn] += BIGLO(c);
+    }
+}
+
+VALUE
+rb_big_sq_fast(VALUE x)
+{
+    size_t xn = BIGNUM_LEN(x), zn = 2 * xn;
+    VALUE z = bignew(zn, 1);
+    bary_sq_fast(BDIGITS(z), zn, BDIGITS(x), xn);
+    RB_GC_GUARD(x);
     return z;
 }
 
-/* modify a bignum by 2's complement */
+/* balancing multiplication by slicing larger argument */
 static void
-get2comp(x)
-    VALUE x;
+bary_mul_balance_with_mulfunc(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn, mulfunc_t *mulfunc)
 {
-    long i = RBIGNUM(x)->len;
-    BDIGIT *ds = BDIGITS(x);
-    BDIGIT_DBL num;
+    VALUE work = 0;
+    size_t yn0 = yn;
+    size_t r, n;
+
+    assert(xn + yn <= zn);
+    assert(xn <= yn);
+    assert(!KARATSUBA_BALANCED(xn, yn) || !TOOM3_BALANCED(xn, yn));
+
+    BDIGITS_ZERO(zds, xn);
+
+    n = 0;
+    while (yn > 0) {
+        BDIGIT *tds;
+        size_t tn;
+	r = xn > yn ? yn : xn;
+        tn = xn + r;
+        if (2 * (xn + r) <= zn - n) {
+            tds = zds + n + xn + r;
+            mulfunc(tds, tn, xds, xn, yds + n, r, wds, wn);
+            BDIGITS_ZERO(zds + n + xn, r);
+            bary_add(zds + n, tn,
+                     zds + n, tn,
+                     tds, tn);
+        }
+        else {
+            if (wn < xn) {
+                wn = xn;
+                wds = ALLOCV_N(BDIGIT, work, wn);
+            }
+            tds = zds + n;
+            MEMCPY(wds, zds + n, BDIGIT, xn);
+            mulfunc(tds, tn, xds, xn, yds + n, r, wds+xn, wn-xn);
+            bary_add(zds + n, tn,
+                     zds + n, tn,
+                     wds, xn);
+        }
+	yn -= r;
+	n += r;
+    }
+    BDIGITS_ZERO(zds+xn+yn0, zn - (xn+yn0));
+
+    if (work)
+        ALLOCV_END(work);
+}
+
+VALUE
+rb_big_mul_balance(VALUE x, VALUE y)
+{
+    size_t xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y), zn = xn + yn;
+    VALUE z = bignew(zn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+    bary_mul_balance_with_mulfunc(BDIGITS(z), zn, BDIGITS(x), xn, BDIGITS(y), yn, NULL, 0, bary_mul_toom3_start);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+    return z;
+}
+
+/* multiplication by karatsuba method */
+static void
+bary_mul_karatsuba(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn)
+{
+    VALUE work = 0;
+
+    size_t n;
+    int sub_p, borrow, carry1, carry2, carry3;
+
+    int odd_y = 0;
+    int odd_xy = 0;
+    int sq;
+
+    const BDIGIT *xds0, *xds1, *yds0, *yds1;
+    BDIGIT *zds0, *zds1, *zds2, *zds3;
+
+    assert(xn + yn <= zn);
+    assert(xn <= yn);
+    assert(yn < 2 * xn);
+
+    sq = xds == yds && xn == yn;
+
+    if (yn & 1) {
+        odd_y = 1;
+        yn--;
+        if (yn < xn) {
+            odd_xy = 1;
+            xn--;
+        }
+    }
+
+    n = yn / 2;
+
+    assert(n < xn);
+
+    if (wn < n) {
+        /* This function itself needs only n BDIGITs for work area.
+         * However this function calls bary_mul_karatsuba and
+         * bary_mul_balance recursively.
+         * 2n BDIGITs are enough to avoid allocations in
+         * the recursively called functions.
+         */
+        wn = 2*n;
+        wds = ALLOCV_N(BDIGIT, work, wn);
+    }
+
+    /* Karatsuba algorithm:
+     *
+     * x = x0 + r*x1
+     * y = y0 + r*y1
+     * z = x*y
+     *   = (x0 + r*x1) * (y0 + r*y1)
+     *   = x0*y0 + r*(x1*y0 + x0*y1) + r*r*x1*y1
+     *   = x0*y0 + r*(x0*y0 + x1*y1 - (x1-x0)*(y1-y0)) + r*r*x1*y1
+     *   = x0*y0 + r*(x0*y0 + x1*y1 - (x0-x1)*(y0-y1)) + r*r*x1*y1
+     */
+
+    xds0 = xds;
+    xds1 = xds + n;
+    yds0 = yds;
+    yds1 = yds + n;
+    zds0 = zds;
+    zds1 = zds + n;
+    zds2 = zds + 2*n;
+    zds3 = zds + 3*n;
+
+    sub_p = 1;
+
+    /* zds0:? zds1:? zds2:? zds3:? wds:? */
+
+    if (bary_sub(zds0, n, xds, n, xds+n, xn-n)) {
+        bary_2comp(zds0, n);
+        sub_p = !sub_p;
+    }
+
+    /* zds0:|x1-x0| zds1:? zds2:? zds3:? wds:? */
+
+    if (sq) {
+        sub_p = 1;
+        bary_mul_karatsuba_start(zds1, 2*n, zds0, n, zds0, n, wds, wn);
+    }
+    else {
+        if (bary_sub(wds, n, yds, n, yds+n, n)) {
+            bary_2comp(wds, n);
+            sub_p = !sub_p;
+        }
+
+        /* zds0:|x1-x0| zds1:? zds2:? zds3:? wds:|y1-y0| */
+
+        bary_mul_karatsuba_start(zds1, 2*n, zds0, n, wds, n, wds+n, wn-n);
+    }
+
+    /* zds0:|x1-x0| zds1,zds2:|x1-x0|*|y1-y0| zds3:? wds:|y1-y0| */
+
+    borrow = 0;
+    if (sub_p) {
+        borrow = !bary_2comp(zds1, 2*n);
+    }
+    /* zds0:|x1-x0| zds1,zds2:-?|x1-x0|*|y1-y0| zds3:? wds:|y1-y0| */
+
+    MEMCPY(wds, zds1, BDIGIT, n);
+
+    /* zds0:|x1-x0| zds1,zds2:-?|x1-x0|*|y1-y0| zds3:? wds:lo(-?|x1-x0|*|y1-y0|) */
+
+    bary_mul_karatsuba_start(zds0, 2*n, xds0, n, yds0, n, wds+n, wn-n);
+
+    /* zds0,zds1:x0*y0 zds2:hi(-?|x1-x0|*|y1-y0|) zds3:? wds:lo(-?|x1-x0|*|y1-y0|) */
+
+    carry1 = bary_add(wds, n, wds, n, zds0, n);
+    carry1 = bary_addc(zds2, n, zds2, n, zds1, n, carry1);
+
+    /* zds0,zds1:x0*y0 zds2:hi(x0*y0-?|x1-x0|*|y1-y0|) zds3:? wds:lo(x0*y0-?|x1-x0|*|y1-y0|) */
+
+    carry2 = bary_add(zds1, n, zds1, n, wds, n);
+
+    /* zds0:lo(x0*y0) zds1:hi(x0*y0)+lo(x0*y0-?|x1-x0|*|y1-y0|) zds2:hi(x0*y0-?|x1-x0|*|y1-y0|) zds3:? wds:lo(x0*y0-?|x1-x0|*|y1-y0|) */
+
+    MEMCPY(wds, zds2, BDIGIT, n);
+
+    /* zds0:lo(x0*y0) zds1:hi(x0*y0)+lo(x0*y0-?|x1-x0|*|y1-y0|) zds2:_ zds3:? wds:hi(x0*y0-?|x1-x0|*|y1-y0|) */
+
+    bary_mul_karatsuba_start(zds2, zn-2*n, xds1, xn-n, yds1, n, wds+n, wn-n);
+
+    /* zds0:lo(x0*y0) zds1:hi(x0*y0)+lo(x0*y0-?|x1-x0|*|y1-y0|) zds2,zds3:x1*y1 wds:hi(x0*y0-?|x1-x0|*|y1-y0|) */
+
+    carry3 = bary_add(zds1, n, zds1, n, zds2, n);
+
+    /* zds0:lo(x0*y0) zds1:hi(x0*y0)+lo(x0*y0-?|x1-x0|*|y1-y0|)+lo(x1*y1) zds2,zds3:x1*y1 wds:hi(x0*y0-?|x1-x0|*|y1-y0|) */
+
+    carry3 = bary_addc(zds2, n, zds2, n, zds3, (4*n < zn ? n : zn-3*n), carry3);
+
+    /* zds0:lo(x0*y0) zds1:hi(x0*y0)+lo(x0*y0-?|x1-x0|*|y1-y0|)+lo(x1*y1) zds2,zds3:x1*y1+hi(x1*y1) wds:hi(x0*y0-?|x1-x0|*|y1-y0|) */
+
+    bary_add(zds2, zn-2*n, zds2, zn-2*n, wds, n);
+
+    /* zds0:lo(x0*y0) zds1:hi(x0*y0)+lo(x0*y0-?|x1-x0|*|y1-y0|)+lo(x1*y1) zds2,zds3:x1*y1+hi(x1*y1)+hi(x0*y0-?|x1-x0|*|y1-y0|) wds:_ */
+
+    if (carry2)
+        bary_add_one(zds2, zn-2*n);
+
+    if (carry1 + carry3 - borrow < 0)
+        bary_sub_one(zds3, zn-3*n);
+    else if (carry1 + carry3 - borrow > 0) {
+        BDIGIT c = carry1 + carry3 - borrow;
+        bary_add(zds3, zn-3*n, zds3, zn-3*n, &c, 1);
+    }
+
+    /*
+    if (SIZEOF_BDIGIT * zn <= 16) {
+        uint128_t z, x, y;
+        ssize_t i;
+        for (x = 0, i = xn-1; 0 <= i; i--) { x <<= SIZEOF_BDIGIT*CHAR_BIT; x |= xds[i]; }
+        for (y = 0, i = yn-1; 0 <= i; i--) { y <<= SIZEOF_BDIGIT*CHAR_BIT; y |= yds[i]; }
+        for (z = 0, i = zn-1; 0 <= i; i--) { z <<= SIZEOF_BDIGIT*CHAR_BIT; z |= zds[i]; }
+        assert(z == x * y);
+    }
+    */
+
+    if (odd_xy) {
+        bary_muladd_1xN(zds+yn, zn-yn, yds[yn], xds, xn);
+        bary_muladd_1xN(zds+xn, zn-xn, xds[xn], yds, yn+1);
+    }
+    else if (odd_y) {
+        bary_muladd_1xN(zds+yn, zn-yn, yds[yn], xds, xn);
+    }
+
+    if (work)
+        ALLOCV_END(work);
+}
+
+VALUE
+rb_big_mul_karatsuba(VALUE x, VALUE y)
+{
+    size_t xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y), zn = xn + yn;
+    VALUE z = bignew(zn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+    if (!((xn <= yn && yn < 2) || KARATSUBA_BALANCED(xn, yn)))
+        rb_raise(rb_eArgError, "unexpected bignum length for karatsuba");
+    bary_mul_karatsuba(BDIGITS(z), zn, BDIGITS(x), xn, BDIGITS(y), yn, NULL, 0);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+    return z;
+}
+
+static void
+bary_mul_toom3(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn)
+{
+    size_t n;
+    size_t wnc;
+    VALUE work = 0;
+
+    /* "p" stands for "positive".  Actually it means "non-negative", though. */
+    size_t x0n; const BDIGIT *x0ds;
+    size_t x1n; const BDIGIT *x1ds;
+    size_t x2n; const BDIGIT *x2ds;
+    size_t y0n; const BDIGIT *y0ds;
+    size_t y1n; const BDIGIT *y1ds;
+    size_t y2n; const BDIGIT *y2ds;
+
+    size_t u1n; BDIGIT *u1ds; int u1p;
+    size_t u2n; BDIGIT *u2ds; int u2p;
+    size_t u3n; BDIGIT *u3ds; int u3p;
+
+    size_t v1n; BDIGIT *v1ds; int v1p;
+    size_t v2n; BDIGIT *v2ds; int v2p;
+    size_t v3n; BDIGIT *v3ds; int v3p;
+
+    size_t t0n; BDIGIT *t0ds; int t0p;
+    size_t t1n; BDIGIT *t1ds; int t1p;
+    size_t t2n; BDIGIT *t2ds; int t2p;
+    size_t t3n; BDIGIT *t3ds; int t3p;
+    size_t t4n; BDIGIT *t4ds; int t4p;
+
+    size_t z0n; BDIGIT *z0ds;
+    size_t z1n; BDIGIT *z1ds; int z1p;
+    size_t z2n; BDIGIT *z2ds; int z2p;
+    size_t z3n; BDIGIT *z3ds; int z3p;
+    size_t z4n; BDIGIT *z4ds;
+
+    size_t zzn; BDIGIT *zzds;
+
+    int sq = xds == yds && xn == yn;
+
+    assert(xn <= yn);  /* assume y >= x */
+    assert(xn + yn <= zn);
+
+    n = (yn + 2) / 3;
+    assert(2*n < xn);
+
+    wnc = 0;
+
+    wnc += (u1n = n+1); /* BITSPERDIG*n+2 bits */
+    wnc += (u2n = n+1); /* BITSPERDIG*n+1 bits */
+    wnc += (u3n = n+1); /* BITSPERDIG*n+3 bits */
+    wnc += (v1n = n+1); /* BITSPERDIG*n+2 bits */
+    wnc += (v2n = n+1); /* BITSPERDIG*n+1 bits */
+    wnc += (v3n = n+1); /* BITSPERDIG*n+3 bits */
+
+    wnc += (t0n = 2*n); /* BITSPERDIG*2*n bits */
+    wnc += (t1n = 2*n+2); /* BITSPERDIG*2*n+4 bits but bary_mul needs u1n+v1n */
+    wnc += (t2n = 2*n+2); /* BITSPERDIG*2*n+2 bits but bary_mul needs u2n+v2n */
+    wnc += (t3n = 2*n+2); /* BITSPERDIG*2*n+6 bits but bary_mul needs u3n+v3n */
+    wnc += (t4n = 2*n); /* BITSPERDIG*2*n bits */
+
+    wnc += (z1n = 2*n+1); /* BITSPERDIG*2*n+5 bits */
+    wnc += (z2n = 2*n+1); /* BITSPERDIG*2*n+6 bits */
+    wnc += (z3n = 2*n+1); /* BITSPERDIG*2*n+8 bits */
+
+    if (wn < wnc) {
+        wn = wnc * 3 / 2; /* Allocate working memory for whole recursion at once. */
+        wds = ALLOCV_N(BDIGIT, work, wn);
+    }
+
+    u1ds = wds; wds += u1n;
+    u2ds = wds; wds += u2n;
+    u3ds = wds; wds += u3n;
+
+    v1ds = wds; wds += v1n;
+    v2ds = wds; wds += v2n;
+    v3ds = wds; wds += v3n;
+
+    t0ds = wds; wds += t0n;
+    t1ds = wds; wds += t1n;
+    t2ds = wds; wds += t2n;
+    t3ds = wds; wds += t3n;
+    t4ds = wds; wds += t4n;
+
+    z1ds = wds; wds += z1n;
+    z2ds = wds; wds += z2n;
+    z3ds = wds; wds += z3n;
+
+    wn -= wnc;
+
+    zzds = u1ds;
+    zzn = 6*n+1;
+
+    x0n = n;
+    x1n = n;
+    x2n = xn - 2*n;
+    x0ds = xds;
+    x1ds = xds + n;
+    x2ds = xds + 2*n;
+
+    if (sq) {
+        y0n = x0n;
+        y1n = x1n;
+        y2n = x2n;
+        y0ds = x0ds;
+        y1ds = x1ds;
+        y2ds = x2ds;
+    }
+    else {
+        y0n = n;
+        y1n = n;
+        y2n = yn - 2*n;
+        y0ds = yds;
+        y1ds = yds + n;
+        y2ds = yds + 2*n;
+    }
+
+    /*
+     * ref. http://en.wikipedia.org/wiki/Toom%E2%80%93Cook_multiplication
+     *
+     * x(b) = x0 * b^0 + x1 * b^1 + x2 * b^2
+     * y(b) = y0 * b^0 + y1 * b^1 + y2 * b^2
+     *
+     * z(b) = x(b) * y(b)
+     * z(b) = z0 * b^0 + z1 * b^1 + z2 * b^2 + z3 * b^3 + z4 * b^4
+     * where:
+     *   z0 = x0 * y0
+     *   z1 = x0 * y1 + x1 * y0
+     *   z2 = x0 * y2 + x1 * y1 + x2 * y0
+     *   z3 = x1 * y2 + x2 * y1
+     *   z4 = x2 * y2
+     *
+     * Toom3 method (a.k.a. Toom-Cook method):
+     * (Step1) calculating 5 points z(b0), z(b1), z(b2), z(b3), z(b4),
+     * where:
+     *   b0 = 0, b1 = 1, b2 = -1, b3 = -2, b4 = inf,
+     *   z(0)   = x(0)   * y(0)   = x0 * y0
+     *   z(1)   = x(1)   * y(1)   = (x0 + x1 + x2) * (y0 + y1 + y2)
+     *   z(-1)  = x(-1)  * y(-1)  = (x0 - x1 + x2) * (y0 - y1 + y2)
+     *   z(-2)  = x(-2)  * y(-2)  = (x0 - 2 * (x1 - 2 * x2)) * (y0 - 2 * (y1 - 2 * y2))
+     *   z(inf) = x(inf) * y(inf) = x2 * y2
+     *
+     * (Step2) interpolating z0, z1, z2, z3 and z4.
+     *
+     * (Step3) Substituting base value into b of the polynomial z(b),
+     */
+
+    /*
+     * [Step1] calculating 5 points z(b0), z(b1), z(b2), z(b3), z(b4)
+     */
+
+    /* u1 <- x0 + x2 */
+    bary_add(u1ds, u1n, x0ds, x0n, x2ds, x2n);
+    u1p = 1;
+
+    /* x(-1) : u2 <- u1 - x1 = x0 - x1 + x2 */
+    if (bary_sub(u2ds, u2n, u1ds, u1n, x1ds, x1n)) {
+        bary_2comp(u2ds, u2n);
+        u2p = 0;
+    }
+    else {
+        u2p = 1;
+    }
+
+    /* x(1) : u1 <- u1 + x1 = x0 + x1 + x2 */
+    bary_add(u1ds, u1n, u1ds, u1n, x1ds, x1n);
+
+    /* x(-2) : u3 <- 2 * (u2 + x2) - x0 = x0 - 2 * (x1 - 2 * x2) */
+    u3p = 1;
+    if (u2p) {
+        bary_add(u3ds, u3n, u2ds, u2n, x2ds, x2n);
+    }
+    else if (bary_sub(u3ds, u3n, x2ds, x2n, u2ds, u2n)) {
+        bary_2comp(u3ds, u3n);
+        u3p = 0;
+    }
+    bary_small_lshift(u3ds, u3ds, u3n, 1);
+    if (!u3p) {
+        bary_add(u3ds, u3n, u3ds, u3n, x0ds, x0n);
+    }
+    else if (bary_sub(u3ds, u3n, u3ds, u3n, x0ds, x0n)) {
+        bary_2comp(u3ds, u3n);
+        u3p = 0;
+    }
+
+    if (sq) {
+        v1n = u1n; v1ds = u1ds; v1p = u1p;
+        v2n = u2n; v2ds = u2ds; v2p = u2p;
+        v3n = u3n; v3ds = u3ds; v3p = u3p;
+    }
+    else {
+	/* v1 <- y0 + y2 */
+        bary_add(v1ds, v1n, y0ds, y0n, y2ds, y2n);
+        v1p = 1;
+
+	/* y(-1) : v2 <- v1 - y1 = y0 - y1 + y2 */
+        v2p = 1;
+        if (bary_sub(v2ds, v2n, v1ds, v1n, y1ds, y1n)) {
+            bary_2comp(v2ds, v2n);
+            v2p = 0;
+        }
+
+	/* y(1) : v1 <- v1 + y1 = y0 + y1 + y2 */
+        bary_add(v1ds, v1n, v1ds, v1n, y1ds, y1n);
+
+	/* y(-2) : v3 <- 2 * (v2 + y2) - y0 = y0 - 2 * (y1 - 2 * y2) */
+        v3p = 1;
+        if (v2p) {
+            bary_add(v3ds, v3n, v2ds, v2n, y2ds, y2n);
+        }
+        else if (bary_sub(v3ds, v3n, y2ds, y2n, v2ds, v2n)) {
+            bary_2comp(v3ds, v3n);
+            v3p = 0;
+        }
+        bary_small_lshift(v3ds, v3ds, v3n, 1);
+        if (!v3p) {
+            bary_add(v3ds, v3n, v3ds, v3n, y0ds, y0n);
+        }
+        else if (bary_sub(v3ds, v3n, v3ds, v3n, y0ds, y0n)) {
+            bary_2comp(v3ds, v3n);
+            v3p = 0;
+        }
+    }
+
+    /* z(0) : t0 <- x0 * y0 */
+    bary_mul_toom3_start(t0ds, t0n, x0ds, x0n, y0ds, y0n, wds, wn);
+    t0p = 1;
+
+    /* z(1) : t1 <- u1 * v1 */
+    bary_mul_toom3_start(t1ds, t1n, u1ds, u1n, v1ds, v1n, wds, wn);
+    t1p = u1p == v1p;
+    assert(t1ds[t1n-1] == 0);
+    t1n--;
+
+    /* z(-1) : t2 <- u2 * v2 */
+    bary_mul_toom3_start(t2ds, t2n, u2ds, u2n, v2ds, v2n, wds, wn);
+    t2p = u2p == v2p;
+    assert(t2ds[t2n-1] == 0);
+    t2n--;
+
+    /* z(-2) : t3 <- u3 * v3 */
+    bary_mul_toom3_start(t3ds, t3n, u3ds, u3n, v3ds, v3n, wds, wn);
+    t3p = u3p == v3p;
+    assert(t3ds[t3n-1] == 0);
+    t3n--;
+
+    /* z(inf) : t4 <- x2 * y2 */
+    bary_mul_toom3_start(t4ds, t4n, x2ds, x2n, y2ds, y2n, wds, wn);
+    t4p = 1;
+
+    /*
+     * [Step2] interpolating z0, z1, z2, z3 and z4.
+     */
+
+    /* z0 <- z(0) == t0 */
+    z0n = t0n; z0ds = t0ds;
+
+    /* z4 <- z(inf) == t4 */
+    z4n = t4n; z4ds = t4ds;
+
+    /* z3 <- (z(-2) - z(1)) / 3 == (t3 - t1) / 3 */
+    if (t3p == t1p) {
+        z3p = t3p;
+        if (bary_sub(z3ds, z3n, t3ds, t3n, t1ds, t1n)) {
+            bary_2comp(z3ds, z3n);
+            z3p = !z3p;
+        }
+    }
+    else {
+        z3p = t3p;
+        bary_add(z3ds, z3n, t3ds, t3n, t1ds, t1n);
+    }
+    bigdivrem_single(z3ds, z3ds, z3n, 3);
+
+    /* z1 <- (z(1) - z(-1)) / 2 == (t1 - t2) / 2 */
+    if (t1p == t2p) {
+        z1p = t1p;
+        if (bary_sub(z1ds, z1n, t1ds, t1n, t2ds, t2n)) {
+            bary_2comp(z1ds, z1n);
+            z1p = !z1p;
+        }
+    }
+    else {
+        z1p = t1p;
+        bary_add(z1ds, z1n, t1ds, t1n, t2ds, t2n);
+    }
+    bary_small_rshift(z1ds, z1ds, z1n, 1, 0);
+
+    /* z2 <- z(-1) - z(0) == t2 - t0 */
+    if (t2p == t0p) {
+        z2p = t2p;
+        if (bary_sub(z2ds, z2n, t2ds, t2n, t0ds, t0n)) {
+            bary_2comp(z2ds, z2n);
+            z2p = !z2p;
+        }
+    }
+    else {
+        z2p = t2p;
+        bary_add(z2ds, z2n, t2ds, t2n, t0ds, t0n);
+    }
+
+    /* z3 <- (z2 - z3) / 2 + 2 * z(inf) == (z2 - z3) / 2 + 2 * t4 */
+    if (z2p == z3p) {
+        z3p = z2p;
+        if (bary_sub(z3ds, z3n, z2ds, z2n, z3ds, z3n)) {
+            bary_2comp(z3ds, z3n);
+            z3p = !z3p;
+        }
+    }
+    else {
+        z3p = z2p;
+        bary_add(z3ds, z3n, z2ds, z2n, z3ds, z3n);
+    }
+    bary_small_rshift(z3ds, z3ds, z3n, 1, 0);
+    if (z3p == t4p) {
+        bary_muladd_1xN(z3ds, z3n, 2, t4ds, t4n);
+    }
+    else {
+        if (bary_mulsub_1xN(z3ds, z3n, 2, t4ds, t4n)) {
+            bary_2comp(z3ds, z3n);
+            z3p = !z3p;
+        }
+    }
+
+    /* z2 <- z2 + z1 - z(inf) == z2 + z1 - t4 */
+    if (z2p == z1p) {
+        bary_add(z2ds, z2n, z2ds, z2n, z1ds, z1n);
+    }
+    else {
+        if (bary_sub(z2ds, z2n, z2ds, z2n, z1ds, z1n)) {
+            bary_2comp(z2ds, z2n);
+            z2p = !z2p;
+        }
+    }
+
+    if (z2p == t4p) {
+        if (bary_sub(z2ds, z2n, z2ds, z2n, t4ds, t4n)) {
+            bary_2comp(z2ds, z2n);
+            z2p = !z2p;
+        }
+    }
+    else {
+        bary_add(z2ds, z2n, z2ds, z2n, t4ds, t4n);
+    }
+
+    /* z1 <- z1 - z3 */
+    if (z1p == z3p) {
+        if (bary_sub(z1ds, z1n, z1ds, z1n, z3ds, z3n)) {
+            bary_2comp(z1ds, z1n);
+            z1p = !z1p;
+        }
+    }
+    else {
+        bary_add(z1ds, z1n, z1ds, z1n, z3ds, z3n);
+    }
+
+    /*
+     * [Step3] Substituting base value into b of the polynomial z(b),
+     */
+
+    MEMCPY(zzds, z0ds, BDIGIT, z0n);
+    BDIGITS_ZERO(zzds + z0n, 4*n - z0n);
+    MEMCPY(zzds + 4*n, z4ds, BDIGIT, z4n);
+    BDIGITS_ZERO(zzds + 4*n + z4n, zzn - (4*n + z4n));
+    if (z1p)
+        bary_add(zzds +   n, zzn -   n, zzds +   n, zzn -   n, z1ds, z1n);
+    else
+        bary_sub(zzds +   n, zzn -   n, zzds +   n, zzn -   n, z1ds, z1n);
+    if (z2p)
+        bary_add(zzds + 2*n, zzn - 2*n, zzds + 2*n, zzn - 2*n, z2ds, z2n);
+    else
+        bary_sub(zzds + 2*n, zzn - 2*n, zzds + 2*n, zzn - 2*n, z2ds, z2n);
+    if (z3p)
+        bary_add(zzds + 3*n, zzn - 3*n, zzds + 3*n, zzn - 3*n, z3ds, z3n);
+    else
+        bary_sub(zzds + 3*n, zzn - 3*n, zzds + 3*n, zzn - 3*n, z3ds, z3n);
+
+    BARY_TRUNC(zzds, zzn);
+    MEMCPY(zds, zzds, BDIGIT, zzn);
+    BDIGITS_ZERO(zds + zzn, zn - zzn);
+
+    if (work)
+        ALLOCV_END(work);
+}
+
+VALUE
+rb_big_mul_toom3(VALUE x, VALUE y)
+{
+    size_t xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y), zn = xn + yn;
+    VALUE z = bignew(zn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+    if (xn > yn || yn < 3 || !TOOM3_BALANCED(xn,yn))
+        rb_raise(rb_eArgError, "unexpected bignum length for toom3");
+    bary_mul_toom3(BDIGITS(z), zn, BDIGITS(x), xn, BDIGITS(y), yn, NULL, 0);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+    return z;
+}
+
+#ifdef USE_GMP
+static void
+bary_mul_gmp(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    const size_t nails = (sizeof(BDIGIT)-SIZEOF_BDIGIT)*CHAR_BIT;
+    mpz_t x, y, z;
+    size_t count;
+
+    assert(xn + yn <= zn);
+
+    mpz_init(x);
+    mpz_init(y);
+    mpz_init(z);
+    mpz_import(x, xn, -1, sizeof(BDIGIT), 0, nails, xds);
+    if (xds == yds && xn == yn) {
+        mpz_mul(z, x, x);
+    }
+    else {
+        mpz_import(y, yn, -1, sizeof(BDIGIT), 0, nails, yds);
+        mpz_mul(z, x, y);
+    }
+    mpz_export(zds, &count, -1, sizeof(BDIGIT), 0, nails, z);
+    BDIGITS_ZERO(zds+count, zn-count);
+    mpz_clear(x);
+    mpz_clear(y);
+    mpz_clear(z);
+}
+
+VALUE
+rb_big_mul_gmp(VALUE x, VALUE y)
+{
+    size_t xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y), zn = xn + yn;
+    VALUE z = bignew(zn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+    bary_mul_gmp(BDIGITS(z), zn, BDIGITS(x), xn, BDIGITS(y), yn);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+    return z;
+}
+#endif
+
+static void
+bary_short_mul(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    assert(xn + yn <= zn);
+
+    if (xn == 1 && yn == 1) {
+        bary_mul_single(zds, zn, xds[0], yds[0]);
+    }
+    else {
+        bary_mul_normal(zds, zn, xds, xn, yds, yn);
+        rb_thread_check_ints();
+    }
+}
+
+/* determine whether a bignum is sparse or not by random sampling */
+static inline int
+bary_sparse_p(const BDIGIT *ds, size_t n)
+{
+    long c = 0;
+
+    if (          ds[rb_genrand_ulong_limited(n / 2) + n / 4]) c++;
+    if (c <= 1 && ds[rb_genrand_ulong_limited(n / 2) + n / 4]) c++;
+    if (c <= 1 && ds[rb_genrand_ulong_limited(n / 2) + n / 4]) c++;
+
+    return (c <= 1) ? 1 : 0;
+}
+
+static int
+bary_mul_precheck(BDIGIT **zdsp, size_t *znp, const BDIGIT **xdsp, size_t *xnp, const BDIGIT **ydsp, size_t *ynp)
+{
+    size_t nlsz; /* number of least significant zero BDIGITs */
+
+    BDIGIT *zds = *zdsp;
+    size_t zn = *znp;
+    const BDIGIT *xds = *xdsp;
+    size_t xn = *xnp;
+    const BDIGIT *yds = *ydsp;
+    size_t yn = *ynp;
+
+    assert(xn + yn <= zn);
+
+    nlsz = 0;
+
+    while (0 < xn) {
+        if (xds[xn-1] == 0) {
+            xn--;
+        }
+        else {
+            do {
+                if (xds[0] != 0)
+                    break;
+                xds++;
+                xn--;
+                nlsz++;
+            } while (0 < xn);
+            break;
+        }
+    }
+
+    while (0 < yn) {
+        if (yds[yn-1] == 0) {
+            yn--;
+        }
+        else {
+            do {
+                if (yds[0] != 0)
+                    break;
+                yds++;
+                yn--;
+                nlsz++;
+            } while (0 < yn);
+            break;
+        }
+    }
+
+    if (nlsz) {
+        BDIGITS_ZERO(zds, nlsz);
+        zds += nlsz;
+        zn -= nlsz;
+    }
+
+    /* make sure that y is longer than x */
+    if (xn > yn) {
+        const BDIGIT *tds;
+        size_t tn;
+	tds = xds; xds = yds; yds = tds;
+	tn = xn; xn = yn; yn = tn;
+    }
+    assert(xn <= yn);
+
+    if (xn <= 1) {
+        if (xn == 0) {
+            BDIGITS_ZERO(zds, zn);
+            return 1;
+        }
+
+        if (xds[0] == 1) {
+            MEMCPY(zds, yds, BDIGIT, yn);
+            BDIGITS_ZERO(zds+yn, zn-yn);
+            return 1;
+        }
+        if (POW2_P(xds[0])) {
+            zds[yn] = bary_small_lshift(zds, yds, yn, bit_length(xds[0])-1);
+            BDIGITS_ZERO(zds+yn+1, zn-yn-1);
+            return 1;
+        }
+        if (yn == 1 && yds[0] == 1) {
+            zds[0] = xds[0];
+            BDIGITS_ZERO(zds+1, zn-1);
+            return 1;
+        }
+        bary_mul_normal(zds, zn, xds, xn, yds, yn);
+        return 1;
+    }
+
+    *zdsp = zds;
+    *znp = zn;
+    *xdsp = xds;
+    *xnp = xn;
+    *ydsp = yds;
+    *ynp = yn;
+
+    return 0;
+}
+
+static void
+bary_mul_karatsuba_branch(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn)
+{
+    /* normal multiplication when x is small */
+    if (xn < KARATSUBA_MUL_DIGITS) {
+      normal:
+        if (xds == yds && xn == yn)
+            bary_sq_fast(zds, zn, xds, xn);
+        else
+            bary_short_mul(zds, zn, xds, xn, yds, yn);
+        return;
+    }
+
+    /* normal multiplication when x or y is a sparse bignum */
+    if (bary_sparse_p(xds, xn)) goto normal;
+    if (bary_sparse_p(yds, yn)) {
+        bary_short_mul(zds, zn, yds, yn, xds, xn);
+        return;
+    }
+
+    /* balance multiplication by slicing y when x is much smaller than y */
+    if (!KARATSUBA_BALANCED(xn, yn)) {
+        bary_mul_balance_with_mulfunc(zds, zn, xds, xn, yds, yn, wds, wn, bary_mul_karatsuba_start);
+        return;
+    }
+
+    /* multiplication by karatsuba method */
+    bary_mul_karatsuba(zds, zn, xds, xn, yds, yn, wds, wn);
+}
+
+static void
+bary_mul_karatsuba_start(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn)
+{
+    if (bary_mul_precheck(&zds, &zn, &xds, &xn, &yds, &yn))
+        return;
+
+    bary_mul_karatsuba_branch(zds, zn, xds, xn, yds, yn, wds, wn);
+}
+
+static void
+bary_mul_toom3_branch(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn)
+{
+    if (xn < TOOM3_MUL_DIGITS) {
+        bary_mul_karatsuba_branch(zds, zn, xds, xn, yds, yn, wds, wn);
+        return;
+    }
+
+    if (!TOOM3_BALANCED(xn, yn)) {
+        bary_mul_balance_with_mulfunc(zds, zn, xds, xn, yds, yn, wds, wn, bary_mul_toom3_start);
+        return;
+    }
+
+    bary_mul_toom3(zds, zn, xds, xn, yds, yn, wds, wn);
+}
+
+static void
+bary_mul_toom3_start(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, BDIGIT *wds, size_t wn)
+{
+    if (bary_mul_precheck(&zds, &zn, &xds, &xn, &yds, &yn))
+        return;
+
+    bary_mul_toom3_branch(zds, zn, xds, xn, yds, yn, wds, wn);
+}
+
+static void
+bary_mul(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    if (xn <= yn) {
+        if (xn < NAIVE_MUL_DIGITS) {
+            if (xds == yds && xn == yn)
+                bary_sq_fast(zds, zn, xds, xn);
+            else
+                bary_short_mul(zds, zn, xds, xn, yds, yn);
+            return;
+        }
+    }
+    else {
+        if (yn < NAIVE_MUL_DIGITS) {
+            bary_short_mul(zds, zn, yds, yn, xds, xn);
+            return;
+        }
+    }
+
+#ifdef USE_GMP
+    bary_mul_gmp(zds, zn, xds, xn, yds, yn);
+#else
+    bary_mul_toom3_start(zds, zn, xds, xn, yds, yn, NULL, 0);
+#endif
+}
+
+struct big_div_struct {
+    size_t yn, zn;
+    BDIGIT *yds, *zds;
+    volatile VALUE stop;
+};
+
+static void *
+bigdivrem1(void *ptr)
+{
+    struct big_div_struct *bds = (struct big_div_struct*)ptr;
+    size_t yn = bds->yn;
+    size_t zn = bds->zn;
+    BDIGIT *yds = bds->yds, *zds = bds->zds;
+    BDIGIT_DBL_SIGNED num;
+    BDIGIT q;
 
-    if (!i) return;
-    while (i--) ds[i] = ~ds[i];
-    i = 0; num = 1;
     do {
-	num += ds[i];
-	ds[i++] = BIGLO(num);
-	num = BIGDN(num);
-    } while (i < RBIGNUM(x)->len);
-    if (num != 0) {
-	REALLOC_N(RBIGNUM(x)->digits, BDIGIT, ++RBIGNUM(x)->len);
-	ds = BDIGITS(x);
-	ds[RBIGNUM(x)->len-1] = RBIGNUM(x)->sign ? ~0 : 1;
+	if (bds->stop) {
+	    bds->zn = zn;
+	    return 0;
+        }
+	if (zds[zn-1] == yds[yn-1]) q = BDIGMAX;
+	else q = (BDIGIT)((BIGUP(zds[zn-1]) + zds[zn-2])/yds[yn-1]);
+	if (q) {
+            num = bigdivrem_mulsub(zds+zn-(yn+1), yn+1,
+                                   q,
+                                   yds, yn);
+	    while (num) { /* "add back" required */
+		q--;
+                num = bary_add(zds+zn-(yn+1), yn,
+                               zds+zn-(yn+1), yn,
+                               yds, yn);
+                num--;
+	    }
+	}
+        zn--;
+	zds[zn] = q;
+    } while (zn > yn);
+    return 0;
+}
+
+static void
+rb_big_stop(void *ptr)
+{
+    struct big_div_struct *bds = ptr;
+    bds->stop = Qtrue;
+}
+
+static BDIGIT
+bigdivrem_single1(BDIGIT *qds, const BDIGIT *xds, size_t xn, BDIGIT x_higher_bdigit, BDIGIT y)
+{
+    assert(0 < xn);
+    assert(x_higher_bdigit < y);
+    if (POW2_P(y)) {
+        BDIGIT r;
+        r = xds[0] & (y-1);
+        bary_small_rshift(qds, xds, xn, bit_length(y)-1, x_higher_bdigit);
+        return r;
+    }
+    else {
+        size_t i;
+        BDIGIT_DBL t2;
+        t2 = x_higher_bdigit;
+        for (i = 0; i < xn; i++) {
+            t2 = BIGUP(t2) + xds[xn - i - 1];
+            qds[xn - i - 1] = (BDIGIT)(t2 / y);
+            t2 %= y;
+        }
+        return (BDIGIT)t2;
     }
 }
 
-void
-rb_big_2comp(x)			/* get 2's complement */
-    VALUE x;
+static BDIGIT
+bigdivrem_single(BDIGIT *qds, const BDIGIT *xds, size_t xn, BDIGIT y)
 {
-    get2comp(x);
+    return bigdivrem_single1(qds, xds, xn, 0, y);
+}
+
+static void
+bigdivrem_restoring(BDIGIT *zds, size_t zn, BDIGIT *yds, size_t yn)
+{
+    struct big_div_struct bds;
+    size_t ynzero;
+
+    assert(yn < zn);
+    assert(BDIGIT_MSB(yds[yn-1]));
+    assert(zds[zn-1] < yds[yn-1]);
+
+    for (ynzero = 0; !yds[ynzero]; ynzero++);
+
+    if (ynzero+1 == yn) {
+        BDIGIT r;
+        r = bigdivrem_single1(zds+yn, zds+ynzero, zn-yn, zds[zn-1], yds[ynzero]);
+        zds[ynzero] = r;
+        return;
+    }
+
+    bds.yn = yn - ynzero;
+    bds.zds = zds + ynzero;
+    bds.yds = yds + ynzero;
+    bds.stop = Qfalse;
+    bds.zn = zn - ynzero;
+    if (bds.zn > 10000 || bds.yn > 10000) {
+      retry:
+	bds.stop = Qfalse;
+	rb_thread_call_without_gvl(bigdivrem1, &bds, rb_big_stop, &bds);
+
+	if (bds.stop == Qtrue) {
+	    /* execute trap handler, but exception was not raised. */
+	    goto retry;
+	}
+    }
+    else {
+	bigdivrem1(&bds);
+    }
+}
+
+static void
+bary_divmod_normal(BDIGIT *qds, size_t qn, BDIGIT *rds, size_t rn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    int shift;
+    BDIGIT *zds, *yyds;
+    size_t zn;
+    VALUE tmpyz = 0;
+
+    assert(yn < xn || (xn == yn && yds[yn - 1] <= xds[xn - 1]));
+    assert(qds ? (xn - yn + 1) <= qn : 1);
+    assert(rds ? yn <= rn : 1);
+
+    zn = xn + BIGDIVREM_EXTRA_WORDS;
+
+    shift = nlz(yds[yn-1]);
+    if (shift) {
+        int alloc_y = !rds;
+        int alloc_z = !qds || qn < zn;
+        if (alloc_y && alloc_z) {
+            yyds = ALLOCV_N(BDIGIT, tmpyz, yn+zn);
+            zds = yyds + yn;
+        }
+        else {
+            yyds = alloc_y ? ALLOCV_N(BDIGIT, tmpyz, yn) : rds;
+            zds = alloc_z ? ALLOCV_N(BDIGIT, tmpyz, zn) : qds;
+        }
+        zds[xn] = bary_small_lshift(zds, xds, xn, shift);
+        bary_small_lshift(yyds, yds, yn, shift);
+    }
+    else {
+        if (qds && zn <= qn)
+            zds = qds;
+        else
+            zds = ALLOCV_N(BDIGIT, tmpyz, zn);
+        MEMCPY(zds, xds, BDIGIT, xn);
+        zds[xn] = 0;
+        /* bigdivrem_restoring will not modify y.
+         * So use yds directly.  */
+        yyds = (BDIGIT *)yds;
+    }
+
+    bigdivrem_restoring(zds, zn, yyds, yn);
+
+    if (rds) {
+        if (shift)
+            bary_small_rshift(rds, zds, yn, shift, 0);
+        else
+            MEMCPY(rds, zds, BDIGIT, yn);
+        BDIGITS_ZERO(rds+yn, rn-yn);
+    }
+
+    if (qds) {
+        size_t j = zn - yn;
+        MEMMOVE(qds, zds+yn, BDIGIT, j);
+        BDIGITS_ZERO(qds+j, qn-j);
+    }
+
+    if (tmpyz)
+        ALLOCV_END(tmpyz);
+}
+
+VALUE
+rb_big_divrem_normal(VALUE x, VALUE y)
+{
+    size_t xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y), qn, rn;
+    BDIGIT *xds = BDIGITS(x), *yds = BDIGITS(y), *qds, *rds;
+    VALUE q, r;
+
+    BARY_TRUNC(yds, yn);
+    if (yn == 0)
+        rb_num_zerodiv();
+    BARY_TRUNC(xds, xn);
+
+    if (xn < yn || (xn == yn && xds[xn - 1] < yds[yn - 1]))
+        return rb_assoc_new(LONG2FIX(0), x);
+
+    qn = xn + BIGDIVREM_EXTRA_WORDS;
+    q = bignew(qn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+    qds = BDIGITS(q);
+
+    rn = yn;
+    r = bignew(rn, BIGNUM_SIGN(x));
+    rds = BDIGITS(r);
+
+    bary_divmod_normal(qds, qn, rds, rn, xds, xn, yds, yn);
+
+    bigtrunc(q);
+    bigtrunc(r);
+
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+
+    return rb_assoc_new(q, r);
+}
+
+#ifdef USE_GMP
+static void
+bary_divmod_gmp(BDIGIT *qds, size_t qn, BDIGIT *rds, size_t rn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    const size_t nails = (sizeof(BDIGIT)-SIZEOF_BDIGIT)*CHAR_BIT;
+    mpz_t x, y, q, r;
+    size_t count;
+
+    assert(yn < xn || (xn == yn && yds[yn - 1] <= xds[xn - 1]));
+    assert(qds ? (xn - yn + 1) <= qn : 1);
+    assert(rds ? yn <= rn : 1);
+    assert(qds || rds);
+
+    mpz_init(x);
+    mpz_init(y);
+    if (qds) mpz_init(q);
+    if (rds) mpz_init(r);
+
+    mpz_import(x, xn, -1, sizeof(BDIGIT), 0, nails, xds);
+    mpz_import(y, yn, -1, sizeof(BDIGIT), 0, nails, yds);
+
+    if (!rds) {
+        mpz_fdiv_q(q, x, y);
+    }
+    else if (!qds) {
+        mpz_fdiv_r(r, x, y);
+    }
+    else {
+        mpz_fdiv_qr(q, r, x, y);
+    }
+
+    mpz_clear(x);
+    mpz_clear(y);
+
+    if (qds) {
+        mpz_export(qds, &count, -1, sizeof(BDIGIT), 0, nails, q);
+        BDIGITS_ZERO(qds+count, qn-count);
+        mpz_clear(q);
+    }
+
+    if (rds) {
+        mpz_export(rds, &count, -1, sizeof(BDIGIT), 0, nails, r);
+        BDIGITS_ZERO(rds+count, rn-count);
+        mpz_clear(r);
+    }
+}
+
+VALUE
+rb_big_divrem_gmp(VALUE x, VALUE y)
+{
+    size_t xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y), qn, rn;
+    BDIGIT *xds = BDIGITS(x), *yds = BDIGITS(y), *qds, *rds;
+    VALUE q, r;
+
+    BARY_TRUNC(yds, yn);
+    if (yn == 0)
+        rb_num_zerodiv();
+    BARY_TRUNC(xds, xn);
+
+    if (xn < yn || (xn == yn && xds[xn - 1] < yds[yn - 1]))
+        return rb_assoc_new(LONG2FIX(0), x);
+
+    qn = xn - yn + 1;
+    q = bignew(qn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+    qds = BDIGITS(q);
+
+    rn = yn;
+    r = bignew(rn, BIGNUM_SIGN(x));
+    rds = BDIGITS(r);
+
+    bary_divmod_gmp(qds, qn, rds, rn, xds, xn, yds, yn);
+
+    bigtrunc(q);
+    bigtrunc(r);
+
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+
+    return rb_assoc_new(q, r);
+}
+#endif
+
+static void
+bary_divmod_branch(BDIGIT *qds, size_t qn, BDIGIT *rds, size_t rn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+#ifdef USE_GMP
+    if (GMP_DIV_DIGITS < xn) {
+        bary_divmod_gmp(qds, qn, rds, rn, xds, xn, yds, yn);
+        return;
+    }
+#endif
+    bary_divmod_normal(qds, qn, rds, rn, xds, xn, yds, yn);
+}
+
+static void
+bary_divmod(BDIGIT *qds, size_t qn, BDIGIT *rds, size_t rn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn)
+{
+    assert(xn <= qn);
+    assert(yn <= rn);
+
+    BARY_TRUNC(yds, yn);
+    if (yn == 0)
+        rb_num_zerodiv();
+
+    BARY_TRUNC(xds, xn);
+    if (xn == 0) {
+        BDIGITS_ZERO(qds, qn);
+        BDIGITS_ZERO(rds, rn);
+        return;
+    }
+
+    if (xn < yn || (xn == yn && xds[xn - 1] < yds[yn - 1])) {
+        MEMCPY(rds, xds, BDIGIT, xn);
+        BDIGITS_ZERO(rds+xn, rn-xn);
+        BDIGITS_ZERO(qds, qn);
+    }
+    else if (yn == 1) {
+        MEMCPY(qds, xds, BDIGIT, xn);
+        BDIGITS_ZERO(qds+xn, qn-xn);
+        rds[0] = bigdivrem_single(qds, xds, xn, yds[0]);
+        BDIGITS_ZERO(rds+1, rn-1);
+    }
+    else if (xn == 2 && yn == 2) {
+        BDIGIT_DBL x = bary2bdigitdbl(xds, 2);
+        BDIGIT_DBL y = bary2bdigitdbl(yds, 2);
+        BDIGIT_DBL q = x / y;
+        BDIGIT_DBL r = x % y;
+        qds[0] = BIGLO(q);
+        qds[1] = BIGLO(BIGDN(q));
+        BDIGITS_ZERO(qds+2, qn-2);
+        rds[0] = BIGLO(r);
+        rds[1] = BIGLO(BIGDN(r));
+        BDIGITS_ZERO(rds+2, rn-2);
+    }
+    else {
+        bary_divmod_branch(qds, qn, rds, rn, xds, xn, yds, yn);
+    }
+}
+
+
+#define BIGNUM_DEBUG 0
+#if BIGNUM_DEBUG
+#define ON_DEBUG(x) do { x; } while (0)
+static void
+dump_bignum(VALUE x)
+{
+    long i;
+    printf("%c0x0", BIGNUM_SIGN(x) ? '+' : '-');
+    for (i = BIGNUM_LEN(x); i--; ) {
+        printf("_%0*"PRIxBDIGIT, SIZEOF_BDIGIT*2, BDIGITS(x)[i]);
+    }
+    printf(", len=%"PRIuSIZE, BIGNUM_LEN(x));
+    puts("");
 }
 
 static VALUE
-bignorm(x)
-    VALUE x;
+rb_big_dump(VALUE x)
 {
-    if (FIXNUM_P(x)) {
-      return x;
+    dump_bignum(x);
+    return x;
+}
+#else
+#define ON_DEBUG(x)
+#endif
+
+static int
+bigzero_p(VALUE x)
+{
+    return bary_zero_p(BDIGITS(x), BIGNUM_LEN(x));
+}
+
+int
+rb_bigzero_p(VALUE x)
+{
+    return BIGZEROP(x);
+}
+
+int
+rb_cmpint(VALUE val, VALUE a, VALUE b)
+{
+    if (NIL_P(val)) {
+	rb_cmperr(a, b);
     }
-    else if (TYPE(x) == T_BIGNUM) {
-	long len = RBIGNUM(x)->len;
-	BDIGIT *ds = BDIGITS(x);
+    if (FIXNUM_P(val)) {
+        long l = FIX2LONG(val);
+        if (l > 0) return 1;
+        if (l < 0) return -1;
+        return 0;
+    }
+    if (RB_BIGNUM_TYPE_P(val)) {
+	if (BIGZEROP(val)) return 0;
+	if (BIGNUM_SIGN(val)) return 1;
+	return -1;
+    }
+    if (RTEST(rb_funcall(val, '>', 1, INT2FIX(0)))) return 1;
+    if (RTEST(rb_funcall(val, '<', 1, INT2FIX(0)))) return -1;
+    return 0;
+}
 
-	while (--len && !ds[len]) ;
-	RBIGNUM(x)->len = ++len;
+#define BIGNUM_SET_LEN(b,l) \
+    ((RBASIC(b)->flags & BIGNUM_EMBED_FLAG) ? \
+     (void)(RBASIC(b)->flags = \
+	    (RBASIC(b)->flags & ~BIGNUM_EMBED_LEN_MASK) | \
+	    ((l) << BIGNUM_EMBED_LEN_SHIFT)) : \
+     (void)(RBIGNUM(b)->as.heap.len = (l)))
 
-	if (len*SIZEOF_BDIGITS <= sizeof(VALUE)) {
-	    long num = 0;
-	    while (len--) {
-		num = BIGUP(num) + ds[len];
+static void
+rb_big_realloc(VALUE big, size_t len)
+{
+    BDIGIT *ds;
+    if (RBASIC(big)->flags & BIGNUM_EMBED_FLAG) {
+	if (BIGNUM_EMBED_LEN_MAX < len) {
+	    ds = ALLOC_N(BDIGIT, len);
+	    MEMCPY(ds, RBIGNUM(big)->as.ary, BDIGIT, BIGNUM_EMBED_LEN_MAX);
+	    RBIGNUM(big)->as.heap.len = BIGNUM_LEN(big);
+	    RBIGNUM(big)->as.heap.digits = ds;
+	    RBASIC(big)->flags &= ~BIGNUM_EMBED_FLAG;
+	}
+    }
+    else {
+	if (len <= BIGNUM_EMBED_LEN_MAX) {
+	    ds = RBIGNUM(big)->as.heap.digits;
+	    RBASIC(big)->flags |= BIGNUM_EMBED_FLAG;
+	    BIGNUM_SET_LEN(big, len);
+            (void)VALGRIND_MAKE_MEM_UNDEFINED((void*)RBIGNUM(big)->as.ary, sizeof(RBIGNUM(big)->as.ary));
+	    if (ds) {
+		MEMCPY(RBIGNUM(big)->as.ary, ds, BDIGIT, len);
+		xfree(ds);
 	    }
-	    if (num >= 0) {
-		if (RBIGNUM(x)->sign) {
-		    if (POSFIXABLE(num)) return LONG2FIX(num);
-		}
-		else if (NEGFIXABLE(-(long)num)) return LONG2FIX(-(long)num);
+	}
+	else {
+	    if (BIGNUM_LEN(big) == 0) {
+		RBIGNUM(big)->as.heap.digits = ALLOC_N(BDIGIT, len);
+	    }
+	    else {
+		REALLOC_N(RBIGNUM(big)->as.heap.digits, BDIGIT, len);
 	    }
 	}
     }
+}
+
+void
+rb_big_resize(VALUE big, size_t len)
+{
+    rb_big_realloc(big, len);
+    BIGNUM_SET_LEN(big, len);
+}
+
+static VALUE
+bignew_1(VALUE klass, size_t len, int sign)
+{
+    NEWOBJ_OF(big, struct RBignum, klass, T_BIGNUM | (RGENGC_WB_PROTECTED_BIGNUM ? FL_WB_PROTECTED : 0));
+    BIGNUM_SET_SIGN(big, sign);
+    if (len <= BIGNUM_EMBED_LEN_MAX) {
+	RBASIC(big)->flags |= BIGNUM_EMBED_FLAG;
+	BIGNUM_SET_LEN(big, len);
+        (void)VALGRIND_MAKE_MEM_UNDEFINED((void*)RBIGNUM(big)->as.ary, sizeof(RBIGNUM(big)->as.ary));
+    }
+    else {
+	RBIGNUM(big)->as.heap.digits = ALLOC_N(BDIGIT, len);
+	RBIGNUM(big)->as.heap.len = len;
+    }
+    OBJ_FREEZE(big);
+    return (VALUE)big;
+}
+
+VALUE
+rb_big_new(size_t len, int sign)
+{
+    return bignew(len, sign != 0);
+}
+
+VALUE
+rb_big_clone(VALUE x)
+{
+    size_t len = BIGNUM_LEN(x);
+    VALUE z = bignew_1(CLASS_OF(x), len, BIGNUM_SIGN(x));
+
+    MEMCPY(BDIGITS(z), BDIGITS(x), BDIGIT, len);
+    return z;
+}
+
+static void
+big_extend_carry(VALUE x)
+{
+    rb_big_resize(x, BIGNUM_LEN(x)+1);
+    BDIGITS(x)[BIGNUM_LEN(x)-1] = 1;
+}
+
+/* modify a bignum by 2's complement */
+static void
+get2comp(VALUE x)
+{
+    long i = BIGNUM_LEN(x);
+    BDIGIT *ds = BDIGITS(x);
+
+    if (bary_2comp(ds, i)) {
+        big_extend_carry(x);
+    }
+}
+
+void
+rb_big_2comp(VALUE x)			/* get 2's complement */
+{
+    get2comp(x);
+}
+
+static BDIGIT
+abs2twocomp(VALUE *xp, long *n_ret)
+{
+    VALUE x = *xp;
+    long n = BIGNUM_LEN(x);
+    BDIGIT *ds = BDIGITS(x);
+    BDIGIT hibits = 0;
+
+    BARY_TRUNC(ds, n);
+
+    if (n != 0 && BIGNUM_NEGATIVE_P(x)) {
+        VALUE z = bignew_1(CLASS_OF(x), n, 0);
+        MEMCPY(BDIGITS(z), ds, BDIGIT, n);
+        bary_2comp(BDIGITS(z), n);
+        hibits = BDIGMAX;
+	*xp = z;
+    }
+    *n_ret = n;
+    return hibits;
+}
+
+static void
+twocomp2abs_bang(VALUE x, int hibits)
+{
+    BIGNUM_SET_SIGN(x, !hibits);
+    if (hibits) {
+        get2comp(x);
+    }
+}
+
+static inline VALUE
+bigtrunc(VALUE x)
+{
+    size_t len = BIGNUM_LEN(x);
+    BDIGIT *ds = BDIGITS(x);
+
+    if (len == 0) return x;
+    while (--len && !ds[len]);
+    if (BIGNUM_LEN(x) > len+1) {
+	rb_big_resize(x, len+1);
+    }
+    return x;
+}
+
+static inline VALUE
+bigfixize(VALUE x)
+{
+    size_t n = BIGNUM_LEN(x);
+    BDIGIT *ds = BDIGITS(x);
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    unsigned long u;
+#else
+    BDIGIT u;
+#endif
+
+    BARY_TRUNC(ds, n);
+
+    if (n == 0) return INT2FIX(0);
+
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    if (sizeof(long)/SIZEOF_BDIGIT < n)
+        goto return_big;
+    else {
+        int i = (int)n;
+        u = 0;
+        while (i--) {
+            u = (unsigned long)(BIGUP(u) + ds[i]);
+        }
+    }
+#else /* SIZEOF_BDIGIT >= SIZEOF_LONG */
+    if (1 < n)
+        goto return_big;
+    else
+        u = ds[0];
+#endif
+
+    if (BIGNUM_POSITIVE_P(x)) {
+        if (POSFIXABLE(u)) return LONG2FIX((long)u);
+    }
+    else {
+        if (u <= -FIXNUM_MIN) return LONG2FIX(-(long)u);
+    }
+
+  return_big:
+    rb_big_resize(x, n);
+    return x;
+}
+
+static VALUE
+bignorm(VALUE x)
+{
+    if (RB_BIGNUM_TYPE_P(x)) {
+	x = bigfixize(x);
+    }
     return x;
 }
 
 VALUE
-rb_big_norm(x)
-    VALUE x;
+rb_big_norm(VALUE x)
 {
     return bignorm(x);
 }
 
 VALUE
-rb_uint2big(n)
-    unsigned long n;
+rb_uint2big(uintptr_t n)
 {
-    BDIGIT_DBL num = n;
-    long i = 0;
-    BDIGIT *digits;
-    VALUE big;
+    long i;
+    VALUE big = bignew(bdigit_roomof(SIZEOF_VALUE), 1);
+    BDIGIT *digits = BDIGITS(big);
 
-    big = bignew(DIGSPERLONG, 1);
-    digits = BDIGITS(big);
-    while (i < DIGSPERLONG) {
-	digits[i++] = BIGLO(num);
-	num = BIGDN(num);
+#if SIZEOF_BDIGIT >= SIZEOF_VALUE
+    digits[0] = n;
+#else
+    for (i = 0; i < bdigit_roomof(SIZEOF_VALUE); i++) {
+	digits[i] = BIGLO(n);
+	n = BIGDN(n);
     }
+#endif
 
-    i = DIGSPERLONG;
+    i = bdigit_roomof(SIZEOF_VALUE);
     while (--i && !digits[i]) ;
-    RBIGNUM(big)->len = i+1;
+    BIGNUM_SET_LEN(big, i+1);
     return big;
 }
 
 VALUE
-rb_int2big(n)
-    long n;
+rb_int2big(intptr_t n)
 {
     long neg = 0;
+    VALUE u;
     VALUE big;
 
     if (n < 0) {
-	n = -n;
+        u = 1 + (VALUE)(-(n + 1)); /* u = -n avoiding overflow */
 	neg = 1;
     }
-    big = rb_uint2big(n);
+    else {
+        u = n;
+    }
+    big = rb_uint2big(u);
     if (neg) {
-	RBIGNUM(big)->sign = 0;
+	BIGNUM_SET_NEGATIVE_SIGN(big);
     }
     return big;
 }
 
 VALUE
-rb_uint2inum(n)
-    unsigned long n;
+rb_uint2inum(uintptr_t n)
 {
     if (POSFIXABLE(n)) return LONG2FIX(n);
     return rb_uint2big(n);
 }
 
 VALUE
-rb_int2inum(n)
-    long n;
+rb_int2inum(intptr_t n)
 {
     if (FIXABLE(n)) return LONG2FIX(n);
     return rb_int2big(n);
 }
 
-#ifdef HAVE_LONG_LONG
-
 void
-rb_quad_pack(buf, val)
-    char *buf;
-    VALUE val;
+rb_big_pack(VALUE val, unsigned long *buf, long num_longs)
+{
+    rb_integer_pack(val, buf, num_longs, sizeof(long), 0,
+            INTEGER_PACK_LSWORD_FIRST|INTEGER_PACK_NATIVE_BYTE_ORDER|
+            INTEGER_PACK_2COMP);
+}
+
+VALUE
+rb_big_unpack(unsigned long *buf, long num_longs)
 {
-    LONG_LONG q;
+    return rb_integer_unpack(buf, num_longs, sizeof(long), 0,
+            INTEGER_PACK_LSWORD_FIRST|INTEGER_PACK_NATIVE_BYTE_ORDER|
+            INTEGER_PACK_2COMP);
+}
+
+/*
+ * Calculate the number of bytes to be required to represent
+ * the absolute value of the integer given as _val_.
+ *
+ * [val] an integer.
+ * [nlz_bits_ret] number of leading zero bits in the most significant byte is returned if not NULL.
+ *
+ * This function returns ((val_numbits * CHAR_BIT + CHAR_BIT - 1) / CHAR_BIT)
+ * where val_numbits is the number of bits of abs(val).
+ * This function should not overflow.
+ *
+ * If nlz_bits_ret is not NULL,
+ * (return_value * CHAR_BIT - val_numbits) is stored in *nlz_bits_ret.
+ * In this case, 0 <= *nlz_bits_ret < CHAR_BIT.
+ *
+ */
+size_t
+rb_absint_size(VALUE val, int *nlz_bits_ret)
+{
+    BDIGIT *dp;
+    BDIGIT *de;
+    BDIGIT fixbuf[bdigit_roomof(sizeof(long))];
+
+    int num_leading_zeros;
 
     val = rb_to_int(val);
+
     if (FIXNUM_P(val)) {
-	q = FIX2LONG(val);
+        long v = FIX2LONG(val);
+        if (v < 0) {
+            v = -v;
+        }
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+        fixbuf[0] = v;
+#else
+        {
+            int i;
+            for (i = 0; i < numberof(fixbuf); i++) {
+                fixbuf[i] = BIGLO(v);
+                v = BIGDN(v);
+            }
+        }
+#endif
+        dp = fixbuf;
+        de = fixbuf + numberof(fixbuf);
     }
     else {
-	long len = RBIGNUM(val)->len;
-	BDIGIT *ds;
-
-	if (len > SIZEOF_LONG_LONG/SIZEOF_BDIGITS)
-	    rb_raise(rb_eRangeError, "bignum too big to convert into `quad int'");
-	ds = BDIGITS(val);
-	q = 0;
-	while (len--) {
-	    q = BIGUP(q);
-	    q += ds[len];
-	}
-	if (!RBIGNUM(val)->sign) q = -q;
+        dp = BDIGITS(val);
+        de = dp + BIGNUM_LEN(val);
+    }
+    while (dp < de && de[-1] == 0)
+        de--;
+    if (dp == de) {
+        if (nlz_bits_ret)
+            *nlz_bits_ret = 0;
+        return 0;
     }
-    memcpy(buf, (char*)&q, SIZEOF_LONG_LONG);
+    num_leading_zeros = nlz(de[-1]);
+    if (nlz_bits_ret)
+        *nlz_bits_ret = num_leading_zeros % CHAR_BIT;
+    return (de - dp) * SIZEOF_BDIGIT - num_leading_zeros / CHAR_BIT;
 }
 
-VALUE
-rb_quad_unpack(buf, sign)
-    const char *buf;
-    int sign;
+static size_t
+absint_numwords_small(size_t numbytes, int nlz_bits_in_msbyte, size_t word_numbits, size_t *nlz_bits_ret)
 {
-    unsigned LONG_LONG q;
-    long neg = 0;
-    long i;
-    BDIGIT *digits;
-    VALUE big;
+    size_t val_numbits = numbytes * CHAR_BIT - nlz_bits_in_msbyte;
+    size_t div = val_numbits / word_numbits;
+    size_t mod = val_numbits % word_numbits;
+    size_t numwords;
+    size_t nlz_bits;
+    numwords = mod == 0 ? div : div + 1;
+    nlz_bits = mod == 0 ? 0 : word_numbits - mod;
+    *nlz_bits_ret = nlz_bits;
+    return numwords;
+}
 
-    memcpy(&q, buf, SIZEOF_LONG_LONG);
-    if (sign) {
-	if (FIXABLE((LONG_LONG)q)) return LONG2FIX((LONG_LONG)q);
-	if ((LONG_LONG)q < 0) {
-	    q = -(LONG_LONG)q;
-	    neg = 1;
-	}
+static size_t
+absint_numwords_generic(size_t numbytes, int nlz_bits_in_msbyte, size_t word_numbits, size_t *nlz_bits_ret)
+{
+    static const BDIGIT char_bit[1] = { CHAR_BIT };
+    BDIGIT numbytes_bary[bdigit_roomof(sizeof(numbytes))];
+    BDIGIT val_numbits_bary[bdigit_roomof(sizeof(numbytes) + 1)];
+    BDIGIT nlz_bits_in_msbyte_bary[1];
+    BDIGIT word_numbits_bary[bdigit_roomof(sizeof(word_numbits))];
+    BDIGIT div_bary[numberof(val_numbits_bary) + BIGDIVREM_EXTRA_WORDS];
+    BDIGIT mod_bary[numberof(word_numbits_bary)];
+    BDIGIT one[1] = { 1 };
+    size_t nlz_bits;
+    size_t mod;
+    int sign;
+    size_t numwords;
+
+    nlz_bits_in_msbyte_bary[0] = nlz_bits_in_msbyte;
+
+    /*
+     * val_numbits = numbytes * CHAR_BIT - nlz_bits_in_msbyte
+     * div, mod = val_numbits.divmod(word_numbits)
+     * numwords = mod == 0 ? div : div + 1
+     * nlz_bits = mod == 0 ? 0 : word_numbits - mod
+     */
+
+    bary_unpack(BARY_ARGS(numbytes_bary), &numbytes, 1, sizeof(numbytes), 0,
+        INTEGER_PACK_NATIVE_BYTE_ORDER);
+    BARY_SHORT_MUL(val_numbits_bary, numbytes_bary, char_bit);
+    if (nlz_bits_in_msbyte)
+        BARY_SUB(val_numbits_bary, val_numbits_bary, nlz_bits_in_msbyte_bary);
+    bary_unpack(BARY_ARGS(word_numbits_bary), &word_numbits, 1, sizeof(word_numbits), 0,
+        INTEGER_PACK_NATIVE_BYTE_ORDER);
+    BARY_DIVMOD(div_bary, mod_bary, val_numbits_bary, word_numbits_bary);
+    if (BARY_ZERO_P(mod_bary)) {
+        nlz_bits = 0;
     }
     else {
-	if (POSFIXABLE(q)) return LONG2FIX(q);
+        BARY_ADD(div_bary, div_bary, one);
+        bary_pack(+1, BARY_ARGS(mod_bary), &mod, 1, sizeof(mod), 0,
+            INTEGER_PACK_NATIVE_BYTE_ORDER);
+        nlz_bits = word_numbits - mod;
     }
+    sign = bary_pack(+1, BARY_ARGS(div_bary), &numwords, 1, sizeof(numwords), 0,
+        INTEGER_PACK_NATIVE_BYTE_ORDER);
 
-    i = 0;
-    big = bignew(DIGSPERLL, 1);
-    digits = BDIGITS(big);
-    while (i < DIGSPERLL) {
-	digits[i++] = BIGLO(q);
-	q = BIGDN(q);
+    if (sign == 2) {
+#if defined __GNUC__ && (__GNUC__ == 4 && __GNUC_MINOR__ == 4)
+	*nlz_bits_ret = 0;
+#endif
+        return (size_t)-1;
     }
+    *nlz_bits_ret = nlz_bits;
+    return numwords;
+}
 
-    i = DIGSPERLL;
-    while (i-- && !digits[i]) ;
-    RBIGNUM(big)->len = i+1;
+/*
+ * Calculate the number of words to be required to represent
+ * the absolute value of the integer given as _val_.
+ *
+ * [val] an integer.
+ * [word_numbits] number of bits in a word.
+ * [nlz_bits_ret] number of leading zero bits in the most significant word is returned if not NULL.
+ *
+ * This function returns ((val_numbits * CHAR_BIT + word_numbits - 1) / word_numbits)
+ * where val_numbits is the number of bits of abs(val).
+ *
+ * This function can overflow.
+ * When overflow occur, (size_t)-1 is returned.
+ *
+ * If nlz_bits_ret is not NULL and overflow is not occur,
+ * (return_value * word_numbits - val_numbits) is stored in *nlz_bits_ret.
+ * In this case, 0 <= *nlz_bits_ret < word_numbits.
+ *
+ */
+size_t
+rb_absint_numwords(VALUE val, size_t word_numbits, size_t *nlz_bits_ret)
+{
+    size_t numbytes;
+    int nlz_bits_in_msbyte;
+    size_t numwords;
+    size_t nlz_bits;
 
-    if (neg) {
-	RBIGNUM(big)->sign = 0;
+    if (word_numbits == 0)
+        return (size_t)-1;
+
+    numbytes = rb_absint_size(val, &nlz_bits_in_msbyte);
+
+    if (numbytes <= SIZE_MAX / CHAR_BIT) {
+        numwords = absint_numwords_small(numbytes, nlz_bits_in_msbyte, word_numbits, &nlz_bits);
+#ifdef DEBUG_INTEGER_PACK
+        {
+            size_t numwords0, nlz_bits0;
+            numwords0 = absint_numwords_generic(numbytes, nlz_bits_in_msbyte, word_numbits, &nlz_bits0);
+            assert(numwords0 == numwords);
+            assert(nlz_bits0 == nlz_bits);
+        }
+#endif
     }
-    return bignorm(big);
+    else {
+        numwords = absint_numwords_generic(numbytes, nlz_bits_in_msbyte, word_numbits, &nlz_bits);
+    }
+    if (numwords == (size_t)-1)
+        return numwords;
+
+    if (nlz_bits_ret)
+        *nlz_bits_ret = nlz_bits;
+
+    return numwords;
 }
 
+/* Test abs(val) consists only a bit or not.
+ *
+ * Returns 1 if abs(val) == 1 << n for some n >= 0.
+ * Returns 0 otherwise.
+ *
+ * rb_absint_singlebit_p can be used to determine required buffer size
+ * for rb_integer_pack used with INTEGER_PACK_2COMP (two's complement).
+ *
+ * Following example calculates number of bits required to
+ * represent val in two's complement number, without sign bit.
+ *
+ *   size_t size;
+ *   int neg = FIXNUM_P(val) ? FIX2LONG(val) < 0 : BIGNUM_NEGATIVE_P(val);
+ *   size = rb_absint_numwords(val, 1, NULL)
+ *   if (size == (size_t)-1) ...overflow...
+ *   if (neg && rb_absint_singlebit_p(val))
+ *     size--;
+ *
+ * Following example calculates number of bytes required to
+ * represent val in two's complement number, with sign bit.
+ *
+ *   size_t size;
+ *   int neg = FIXNUM_P(val) ? FIX2LONG(val) < 0 : BIGNUM_NEGATIVE_P(val);
+ *   int nlz_bits;
+ *   size = rb_absint_size(val, &nlz_bits);
+ *   if (nlz_bits == 0 && !(neg && rb_absint_singlebit_p(val)))
+ *     size++;
+ */
+int
+rb_absint_singlebit_p(VALUE val)
+{
+    BDIGIT *dp;
+    BDIGIT *de;
+    BDIGIT fixbuf[bdigit_roomof(sizeof(long))];
+    BDIGIT d;
+
+    val = rb_to_int(val);
+
+    if (FIXNUM_P(val)) {
+        long v = FIX2LONG(val);
+        if (v < 0) {
+            v = -v;
+        }
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+        fixbuf[0] = v;
 #else
+        {
+            int i;
+            for (i = 0; i < numberof(fixbuf); i++) {
+                fixbuf[i] = BIGLO(v);
+                v = BIGDN(v);
+            }
+        }
+#endif
+        dp = fixbuf;
+        de = fixbuf + numberof(fixbuf);
+    }
+    else {
+        dp = BDIGITS(val);
+        de = dp + BIGNUM_LEN(val);
+    }
+    while (dp < de && de[-1] == 0)
+        de--;
+    while (dp < de && dp[0] == 0)
+        dp++;
+    if (dp == de) /* no bit set. */
+        return 0;
+    if (dp != de-1) /* two non-zero words. two bits set, at least. */
+        return 0;
+    d = *dp;
+    return POW2_P(d);
+}
 
-#define QUAD_SIZE 8
 
-void
-rb_quad_pack(buf, val)
-    char *buf;
-    VALUE val;
+/*
+ * Export an integer into a buffer.
+ *
+ * This function fills the buffer specified by _words_ and _numwords_ as
+ * val in the format specified by _wordsize_, _nails_ and _flags_.
+ *
+ * [val] Fixnum, Bignum or another integer like object which has to_int method.
+ * [words] buffer to export abs(val).
+ * [numwords] the size of given buffer as number of words.
+ * [wordsize] the size of word as number of bytes.
+ * [nails] number of padding bits in a word.
+ *   Most significant nails bits of each word are filled by zero.
+ * [flags] bitwise or of constants which name starts "INTEGER_PACK_".
+ *
+ * flags:
+ * [INTEGER_PACK_MSWORD_FIRST] Store the most significant word as the first word.
+ * [INTEGER_PACK_LSWORD_FIRST] Store the least significant word as the first word.
+ * [INTEGER_PACK_MSBYTE_FIRST] Store the most significant byte in a word as the first byte in the word.
+ * [INTEGER_PACK_LSBYTE_FIRST] Store the least significant byte in a word as the first byte in the word.
+ * [INTEGER_PACK_NATIVE_BYTE_ORDER] INTEGER_PACK_MSBYTE_FIRST or INTEGER_PACK_LSBYTE_FIRST corresponding to the host's endian.
+ * [INTEGER_PACK_2COMP] Use 2's complement representation.
+ * [INTEGER_PACK_LITTLE_ENDIAN] Same as INTEGER_PACK_LSWORD_FIRST|INTEGER_PACK_LSBYTE_FIRST
+ * [INTEGER_PACK_BIG_ENDIAN] Same as INTEGER_PACK_MSWORD_FIRST|INTEGER_PACK_MSBYTE_FIRST
+ * [INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION] Use generic implementation (for test and debug).
+ *
+ * This function fills the buffer specified by _words_
+ * as abs(val) if INTEGER_PACK_2COMP is not specified in _flags_.
+ * If INTEGER_PACK_2COMP is specified, 2's complement representation of val is
+ * filled in the buffer.
+ *
+ * This function returns the signedness and overflow condition.
+ * The overflow condition depends on INTEGER_PACK_2COMP.
+ *
+ * INTEGER_PACK_2COMP is not specified:
+ *   -2 : negative overflow.  val <= -2**(numwords*(wordsize*CHAR_BIT-nails))
+ *   -1 : negative without overflow.  -2**(numwords*(wordsize*CHAR_BIT-nails)) < val < 0
+ *   0 : zero.  val == 0
+ *   1 : positive without overflow.  0 < val < 2**(numwords*(wordsize*CHAR_BIT-nails))
+ *   2 : positive overflow.  2**(numwords*(wordsize*CHAR_BIT-nails)) <= val
+ *
+ * INTEGER_PACK_2COMP is specified:
+ *   -2 : negative overflow.  val < -2**(numwords*(wordsize*CHAR_BIT-nails))
+ *   -1 : negative without overflow.  -2**(numwords*(wordsize*CHAR_BIT-nails)) <= val < 0
+ *   0 : zero.  val == 0
+ *   1 : positive without overflow.  0 < val < 2**(numwords*(wordsize*CHAR_BIT-nails))
+ *   2 : positive overflow.  2**(numwords*(wordsize*CHAR_BIT-nails)) <= val
+ *
+ * The value, -2**(numwords*(wordsize*CHAR_BIT-nails)), is representable
+ * in 2's complement representation but not representable in absolute value.
+ * So -1 is returned for the value if INTEGER_PACK_2COMP is specified
+ * but returns -2 if INTEGER_PACK_2COMP is not specified.
+ *
+ * The least significant words are filled in the buffer when overflow occur.
+ */
+
+int
+rb_integer_pack(VALUE val, void *words, size_t numwords, size_t wordsize, size_t nails, int flags)
 {
-    long len;
+    int sign;
+    BDIGIT *ds;
+    size_t num_bdigits;
+    BDIGIT fixbuf[bdigit_roomof(sizeof(long))];
+
+    RB_GC_GUARD(val) = rb_to_int(val);
 
-    memset(buf, 0, QUAD_SIZE);
-    val = rb_to_int(val);
     if (FIXNUM_P(val)) {
-	val = rb_int2big(FIX2LONG(val));
-    }
-    len = RBIGNUM(val)->len * SIZEOF_BDIGITS;
-    if (len > QUAD_SIZE) {
-	rb_raise(rb_eRangeError, "bignum too big to convert into `quad int'");
-    }
-    memcpy(buf, (char*)BDIGITS(val), len);
-    if (!RBIGNUM(val)->sign) {
-	len = QUAD_SIZE;
-	while (len--) {
-	    *buf = ~*buf;
-	    buf++;
-	}
+        long v = FIX2LONG(val);
+        if (v < 0) {
+            sign = -1;
+            v = -v;
+        }
+        else {
+            sign = 1;
+        }
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+        fixbuf[0] = v;
+#else
+        {
+            int i;
+            for (i = 0; i < numberof(fixbuf); i++) {
+                fixbuf[i] = BIGLO(v);
+                v = BIGDN(v);
+            }
+        }
+#endif
+        ds = fixbuf;
+        num_bdigits = numberof(fixbuf);
+    }
+    else {
+        sign = BIGNUM_POSITIVE_P(val) ? 1 : -1;
+        ds = BDIGITS(val);
+        num_bdigits = BIGNUM_LEN(val);
     }
+
+    return bary_pack(sign, ds, num_bdigits, words, numwords, wordsize, nails, flags);
 }
 
-#define BNEG(b) (RSHIFT(((BDIGIT*)b)[QUAD_SIZE/SIZEOF_BDIGITS-1],BITSPERDIG-1) != 0)
+/*
+ * Import an integer into a buffer.
+ *
+ * [words] buffer to import.
+ * [numwords] the size of given buffer as number of words.
+ * [wordsize] the size of word as number of bytes.
+ * [nails] number of padding bits in a word.
+ *   Most significant nails bits of each word are ignored.
+ * [flags] bitwise or of constants which name starts "INTEGER_PACK_".
+ *
+ * flags:
+ * [INTEGER_PACK_MSWORD_FIRST] Interpret the first word as the most significant word.
+ * [INTEGER_PACK_LSWORD_FIRST] Interpret the first word as the least significant word.
+ * [INTEGER_PACK_MSBYTE_FIRST] Interpret the first byte in a word as the most significant byte in the word.
+ * [INTEGER_PACK_LSBYTE_FIRST] Interpret the first byte in a word as the least significant byte in the word.
+ * [INTEGER_PACK_NATIVE_BYTE_ORDER] INTEGER_PACK_MSBYTE_FIRST or INTEGER_PACK_LSBYTE_FIRST corresponding to the host's endian.
+ * [INTEGER_PACK_2COMP] Use 2's complement representation.
+ * [INTEGER_PACK_LITTLE_ENDIAN] Same as INTEGER_PACK_LSWORD_FIRST|INTEGER_PACK_LSBYTE_FIRST
+ * [INTEGER_PACK_BIG_ENDIAN] Same as INTEGER_PACK_MSWORD_FIRST|INTEGER_PACK_MSBYTE_FIRST
+ * [INTEGER_PACK_FORCE_BIGNUM] the result will be a Bignum
+ *   even if it is representable as a Fixnum.
+ * [INTEGER_PACK_NEGATIVE] Returns non-positive value.
+ *   (Returns non-negative value if not specified.)
+ * [INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION] Use generic implementation (for test and debug).
+ *
+ * This function returns the imported integer as Fixnum or Bignum.
+ *
+ * The range of the result value depends on INTEGER_PACK_2COMP and INTEGER_PACK_NEGATIVE.
+ *
+ * INTEGER_PACK_2COMP is not set:
+ *   0 <= val < 2**(numwords*(wordsize*CHAR_BIT-nails)) if !INTEGER_PACK_NEGATIVE
+ *   -2**(numwords*(wordsize*CHAR_BIT-nails)) < val <= 0 if INTEGER_PACK_NEGATIVE
+ *
+ * INTEGER_PACK_2COMP is set:
+ *   -2**(numwords*(wordsize*CHAR_BIT-nails)-1) <= val <= 2**(numwords*(wordsize*CHAR_BIT-nails)-1)-1 if !INTEGER_PACK_NEGATIVE
+ *   -2**(numwords*(wordsize*CHAR_BIT-nails)) <= val <= -1 if INTEGER_PACK_NEGATIVE
+ *
+ * INTEGER_PACK_2COMP without INTEGER_PACK_NEGATIVE means sign extension.
+ * INTEGER_PACK_2COMP with INTEGER_PACK_NEGATIVE mean assuming the higher bits are 1.
+ *
+ * Note that this function returns 0 when numwords is zero and
+ * INTEGER_PACK_2COMP is set but INTEGER_PACK_NEGATIVE is not set.
+ */
 
 VALUE
-rb_quad_unpack(buf, sign)
-    const char *buf;
+rb_integer_unpack(const void *words, size_t numwords, size_t wordsize, size_t nails, int flags)
+{
+    VALUE val;
+    size_t num_bdigits;
     int sign;
+    int nlp_bits;
+    BDIGIT *ds;
+    BDIGIT fixbuf[2] = { 0, 0 };
+
+    validate_integer_pack_format(numwords, wordsize, nails, flags,
+            INTEGER_PACK_MSWORD_FIRST|
+            INTEGER_PACK_LSWORD_FIRST|
+            INTEGER_PACK_MSBYTE_FIRST|
+            INTEGER_PACK_LSBYTE_FIRST|
+            INTEGER_PACK_NATIVE_BYTE_ORDER|
+            INTEGER_PACK_2COMP|
+            INTEGER_PACK_FORCE_BIGNUM|
+            INTEGER_PACK_NEGATIVE|
+            INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION);
+
+    num_bdigits = integer_unpack_num_bdigits(numwords, wordsize, nails, &nlp_bits);
+
+    if (LONG_MAX-1 < num_bdigits)
+        rb_raise(rb_eArgError, "too big to unpack as an integer");
+    if (num_bdigits <= numberof(fixbuf) && !(flags & INTEGER_PACK_FORCE_BIGNUM)) {
+        val = Qfalse;
+        ds = fixbuf;
+    }
+    else {
+        val = bignew((long)num_bdigits, 0);
+        ds = BDIGITS(val);
+    }
+    sign = bary_unpack_internal(ds, num_bdigits, words, numwords, wordsize, nails, flags, nlp_bits);
+
+    if (sign == -2) {
+        if (val) {
+            big_extend_carry(val);
+        }
+        else if (num_bdigits == numberof(fixbuf)) {
+            val = bignew((long)num_bdigits+1, 0);
+	    MEMCPY(BDIGITS(val), fixbuf, BDIGIT, num_bdigits);
+            BDIGITS(val)[num_bdigits++] = 1;
+        }
+        else {
+            ds[num_bdigits++] = 1;
+        }
+    }
+
+    if (!val) {
+        BDIGIT_DBL u = fixbuf[0] + BIGUP(fixbuf[1]);
+        if (u == 0)
+            return LONG2FIX(0);
+	if (0 < sign && POSFIXABLE(u))
+            return LONG2FIX(u);
+	if (sign < 0 && BDIGIT_MSB(fixbuf[1]) == 0 &&
+                NEGFIXABLE(-(BDIGIT_DBL_SIGNED)u))
+            return LONG2FIX(-(BDIGIT_DBL_SIGNED)u);
+        val = bignew((long)num_bdigits, 0 <= sign);
+        MEMCPY(BDIGITS(val), fixbuf, BDIGIT, num_bdigits);
+    }
+
+    if ((flags & INTEGER_PACK_FORCE_BIGNUM) && sign != 0 &&
+        bary_zero_p(BDIGITS(val), BIGNUM_LEN(val)))
+        sign = 0;
+    BIGNUM_SET_SIGN(val, 0 <= sign);
+
+    if (flags & INTEGER_PACK_FORCE_BIGNUM)
+        return bigtrunc(val);
+    return bignorm(val);
+}
+
+#define conv_digit(c) (ruby_digit36_to_number_table[(unsigned char)(c)])
+
+NORETURN(static inline void invalid_radix(int base));
+NORETURN(static inline void invalid_integer(VALUE s));
+
+static inline int
+valid_radix_p(int base)
+{
+    return (1 < base && base <= 36);
+}
+
+static inline void
+invalid_radix(int base)
+{
+    rb_raise(rb_eArgError, "invalid radix %d", base);
+}
+
+static inline void
+invalid_integer(VALUE s)
+{
+    rb_raise(rb_eArgError, "invalid value for Integer(): %+"PRIsVALUE, s);
+}
+
+static int
+str2big_scan_digits(const char *s, const char *str, int base, int badcheck, size_t *num_digits_p, ssize_t *len_p)
 {
-    VALUE big = bignew(QUAD_SIZE/SIZEOF_BDIGITS, 1);
+    char nondigit = 0;
+    size_t num_digits = 0;
+    const char *digits_start = str;
+    const char *digits_end = str;
+    ssize_t len = *len_p;
+
+    int c;
 
-    memcpy((char*)BDIGITS(big), buf, QUAD_SIZE);
-    if (sign && BNEG(buf)) {
-	long len = QUAD_SIZE;
-	char *tmp = (char*)BDIGITS(big);
+    if (!len) {
+	*num_digits_p = 0;
+	*len_p = 0;
+	return TRUE;
+    }
+
+    if (badcheck && *str == '_') goto bad;
 
-	RBIGNUM(big)->sign = 0;
-	while (len--) {
-	    *tmp = ~*tmp;
-	    tmp++;
+    while ((c = *str++) != 0) {
+	if (c == '_') {
+	    if (nondigit) {
+		if (badcheck) goto bad;
+		break;
+	    }
+	    nondigit = (char) c;
+	}
+	else if ((c = conv_digit(c)) < 0 || c >= base) {
+	    break;
+	}
+	else {
+	    nondigit = 0;
+	    num_digits++;
+	    digits_end = str;
 	}
+	if (len > 0 && !--len) break;
     }
+    if (badcheck && nondigit) goto bad;
+    if (badcheck && len) {
+	str--;
+	while (*str && ISSPACE(*str)) {
+	    str++;
+	    if (len > 0 && !--len) break;
+	}
+	if (len && *str) {
+	  bad:
+	    return FALSE;
+	}
+    }
+    *num_digits_p = num_digits;
+    *len_p = digits_end - digits_start;
+    return TRUE;
+}
 
-    return bignorm(big);
+static VALUE
+str2big_poweroftwo(
+    int sign,
+    const char *digits_start,
+    const char *digits_end,
+    size_t num_digits,
+    int bits_per_digit)
+{
+    BDIGIT *dp;
+    BDIGIT_DBL dd;
+    int numbits;
+
+    size_t num_bdigits;
+    const char *p;
+    int c;
+    VALUE z;
+
+    num_bdigits = (num_digits / BITSPERDIG) * bits_per_digit + roomof((num_digits % BITSPERDIG) * bits_per_digit, BITSPERDIG);
+    z = bignew(num_bdigits, sign);
+    dp = BDIGITS(z);
+    dd = 0;
+    numbits = 0;
+    for (p = digits_end; digits_start < p; p--) {
+        if ((c = conv_digit(p[-1])) < 0)
+            continue;
+        dd |= (BDIGIT_DBL)c << numbits;
+        numbits += bits_per_digit;
+        if (BITSPERDIG <= numbits) {
+            *dp++ = BIGLO(dd);
+            dd = BIGDN(dd);
+            numbits -= BITSPERDIG;
+        }
+    }
+    if (numbits) {
+        *dp++ = BIGLO(dd);
+    }
+    assert((size_t)(dp - BDIGITS(z)) == num_bdigits);
+
+    return z;
 }
 
-#endif
+static VALUE
+str2big_normal(
+    int sign,
+    const char *digits_start,
+    const char *digits_end,
+    size_t num_bdigits,
+    int base)
+{
+    size_t blen = 1;
+    BDIGIT *zds;
+    BDIGIT_DBL num;
 
-VALUE
-rb_cstr_to_inum(str, base, badcheck)
-    const char *str;
-    int base;
-    int badcheck;
+    size_t i;
+    const char *p;
+    int c;
+    VALUE z;
+
+    z = bignew(num_bdigits, sign);
+    zds = BDIGITS(z);
+    BDIGITS_ZERO(zds, num_bdigits);
+
+    for (p = digits_start; p < digits_end; p++) {
+        if ((c = conv_digit(*p)) < 0)
+            continue;
+        num = c;
+        i = 0;
+        for (;;) {
+            while (i<blen) {
+                num += (BDIGIT_DBL)zds[i]*base;
+                zds[i++] = BIGLO(num);
+                num = BIGDN(num);
+            }
+            if (num) {
+                blen++;
+                continue;
+            }
+            break;
+        }
+        assert(blen <= num_bdigits);
+    }
+
+    return z;
+}
+
+static VALUE
+str2big_karatsuba(
+    int sign,
+    const char *digits_start,
+    const char *digits_end,
+    size_t num_digits,
+    size_t num_bdigits,
+    int digits_per_bdigits_dbl,
+    int base)
 {
-    const char *s = str;
-    char *end;
-    char sign = 1, nondigit = 0;
+    VALUE powerv;
+    size_t unit;
+    VALUE tmpuv = 0;
+    BDIGIT *uds, *vds, *tds;
+    BDIGIT_DBL dd;
+    BDIGIT_DBL current_base;
+    int m;
+    int power_level = 0;
+
+    size_t i;
+    const char *p;
     int c;
-    BDIGIT_DBL num;
-    long len, blen = 1;
-    long i;
     VALUE z;
-    BDIGIT *zds;
 
-#define conv_digit(c) \
-    (!ISASCII(c) ? -1 : \
-     isdigit(c) ? ((c) - '0') : \
-     islower(c) ? ((c) - 'a' + 10) : \
-     isupper(c) ? ((c) - 'A' + 10) : \
-     -1)
+    uds = ALLOCV_N(BDIGIT, tmpuv, 2*num_bdigits);
+    vds = uds + num_bdigits;
 
-    if (!str) {
-	if (badcheck) goto bad;
-	return INT2FIX(0);
+    powerv = power_cache_get_power(base, power_level, NULL);
+
+    i = 0;
+    dd = 0;
+    current_base = 1;
+    m = digits_per_bdigits_dbl;
+    if (num_digits < (size_t)m)
+        m = (int)num_digits;
+    for (p = digits_end; digits_start < p; p--) {
+        if ((c = conv_digit(p[-1])) < 0)
+            continue;
+        dd = dd + c * current_base;
+        current_base *= base;
+        num_digits--;
+        m--;
+        if (m == 0) {
+            uds[i++] = BIGLO(dd);
+            uds[i++] = (BDIGIT)BIGDN(dd);
+            dd = 0;
+            m = digits_per_bdigits_dbl;
+            if (num_digits < (size_t)m)
+                m = (int)num_digits;
+            current_base = 1;
+        }
     }
-    if (badcheck) {
-	while (ISSPACE(*str)) str++;
+    assert(i == num_bdigits);
+    for (unit = 2; unit < num_bdigits; unit *= 2) {
+        for (i = 0; i < num_bdigits; i += unit*2) {
+            if (2*unit <= num_bdigits - i) {
+                bary_mul(vds+i, unit*2, BDIGITS(powerv), BIGNUM_LEN(powerv), uds+i+unit, unit);
+                bary_add(vds+i, unit*2, vds+i, unit*2, uds+i, unit);
+            }
+            else if (unit <= num_bdigits - i) {
+                bary_mul(vds+i, num_bdigits-i, BDIGITS(powerv), BIGNUM_LEN(powerv), uds+i+unit, num_bdigits-(i+unit));
+                bary_add(vds+i, num_bdigits-i, vds+i, num_bdigits-i, uds+i, unit);
+            }
+            else {
+                MEMCPY(vds+i, uds+i, BDIGIT, num_bdigits-i);
+            }
+        }
+        power_level++;
+        powerv = power_cache_get_power(base, power_level, NULL);
+        tds = vds;
+        vds = uds;
+        uds = tds;
     }
-    else {
-	while (ISSPACE(*str) || *str == '_') str++;
+    BARY_TRUNC(uds, num_bdigits);
+    z = bignew(num_bdigits, sign);
+    MEMCPY(BDIGITS(z), uds, BDIGIT, num_bdigits);
+
+    if (tmpuv)
+        ALLOCV_END(tmpuv);
+
+    return z;
+}
+
+#ifdef USE_GMP
+static VALUE
+str2big_gmp(
+    int sign,
+    const char *digits_start,
+    const char *digits_end,
+    size_t num_digits,
+    size_t num_bdigits,
+    int base)
+{
+    const size_t nails = (sizeof(BDIGIT)-SIZEOF_BDIGIT)*CHAR_BIT;
+    char *buf, *p;
+    const char *q;
+    VALUE tmps;
+    mpz_t mz;
+    VALUE z;
+    BDIGIT *zds;
+    size_t zn, count;
+
+    buf = ALLOCV_N(char, tmps, num_digits+1);
+    p = buf;
+    for (q = digits_start; q < digits_end; q++) {
+        if (conv_digit(*q) < 0)
+            continue;
+        *p++ = *q;
     }
+    *p = '\0';
 
-    if (str[0] == '+') {
-	str++;
+    mpz_init(mz);
+    mpz_set_str(mz, buf, base);
+    zn = num_bdigits;
+    z = bignew(zn, sign);
+    zds = BDIGITS(z);
+    mpz_export(BDIGITS(z), &count, -1, sizeof(BDIGIT), 0, nails, mz);
+    BDIGITS_ZERO(zds+count, zn-count);
+    mpz_clear(mz);
+
+    if (tmps)
+        ALLOCV_END(tmps);
+
+    return z;
+}
+#endif
+
+/*
+ * Parse +str+ as Ruby Integer, i.e., underscores, 0d and 0b prefixes.
+ *
+ * str:      pointer to the string to be parsed.
+ *	     should be NUL-terminated.
+ * base:     base of conversion, must be 2..36, or -36..0.
+ *           if +base+ > 0, the conversion is done according to the +base+
+ *           and unmatched prefix is parsed as a part of the result if
+ *           present.
+ *           if +base+ <= 0, the conversion is done according to the
+ *           prefix if present, in base <code>-base</code> if +base+ < -1,
+ *           or in base 10.
+ * badcheck: if non-zero, +ArgumentError+ is raised when +str+ is not
+ *           valid as an Integer.  if zero, Fixnum 0 is returned in
+ *           that case.
+ */
+VALUE
+rb_cstr_to_inum(const char *str, int base, int badcheck)
+{
+    char *end;
+    VALUE ret = rb_cstr_parse_inum(str, -1, (badcheck ? NULL : &end), base);
+    if (NIL_P(ret)) {
+	if (badcheck) rb_invalid_str(str, "Integer()");
+	ret = INT2FIX(0);
     }
-    else if (str[0] == '-') {
-	str++;
-	sign = 0;
+    return ret;
+}
+
+/*
+ * Parse +str+ as Ruby Integer, i.e., underscores, 0d and 0b prefixes.
+ *
+ * str:  pointer to the string to be parsed.
+ *       should be NUL-terminated if +len+ is negative.
+ * len:  length of +str+ if >= 0.  if +len+ is negative, +str+ should
+ *       be NUL-terminated.
+ * endp: if non-NULL, the address after parsed part is stored.  if
+ *       NULL, Qnil is returned when +str+ is not valid as an Integer.
+ * ndigits: if non-NULL, the number of parsed digits is stored.
+ * base: see +rb_cstr_to_inum+
+ * flags: bitwise OR of below flags:
+ *       RB_INT_PARSE_SIGN: allow preceding spaces and +/- sign
+ *       RB_INT_PARSE_UNDERSCORE: allow an underscore between digits
+ *       RB_INT_PARSE_PREFIX: allow preceding prefix
+ */
+
+VALUE
+rb_int_parse_cstr(const char *str, ssize_t len, char **endp, size_t *ndigits,
+		  int base, int flags)
+{
+    const char *const s = str;
+    char sign = 1;
+    int c;
+    VALUE z = Qnil;
+
+    unsigned long val;
+    int ov;
+
+    const char *digits_start, *digits_end;
+    size_t num_digits = 0;
+    size_t num_bdigits;
+    const ssize_t len0 = len;
+    const int badcheck = !endp;
+
+#define ADV(n) do {\
+	if (len > 0 && len <= (n)) goto bad; \
+	str += (n); \
+	len -= (n); \
+    } while (0)
+#define ASSERT_LEN() do {\
+	assert(len != 0); \
+	if (len0 >= 0) assert(s + len0 == str + len); \
+    } while (0)
+
+    if (!str) {
+      bad:
+	if (endp) *endp = (char *)str;
+	if (ndigits) *ndigits = num_digits;
+	return z;
     }
-    if (str[0] == '+' || str[0] == '-') {
-	if (badcheck) goto bad;
-	return INT2FIX(0);
+    if (len && (flags & RB_INT_PARSE_SIGN)) {
+	while (ISSPACE(*str)) ADV(1);
+
+	if (str[0] == '+') {
+	    ADV(1);
+	}
+	else if (str[0] == '-') {
+	    ADV(1);
+	    sign = 0;
+	}
+	ASSERT_LEN();
     }
     if (base <= 0) {
-	if (str[0] == '0') {
+	if (str[0] == '0' && len > 1) {
 	    switch (str[1]) {
 	      case 'x': case 'X':
 		base = 16;
+		ADV(2);
 		break;
 	      case 'b': case 'B':
 		base = 2;
+		ADV(2);
 		break;
 	      case 'o': case 'O':
 		base = 8;
+		ADV(2);
 		break;
 	      case 'd': case 'D':
 		base = 10;
+		ADV(2);
 		break;
 	      default:
 		base = 8;
@@ -387,68 +4110,75 @@ rb_cstr_to_inum(str, base, badcheck)
 	    base = 10;
 	}
     }
-    switch (base) {
-      case 2:
-	len = 1;
+    else if (len == 1 || !(flags & RB_INT_PARSE_PREFIX)) {
+	/* no prefix */
+    }
+    else if (base == 2) {
 	if (str[0] == '0' && (str[1] == 'b'||str[1] == 'B')) {
-	    str += 2;
+	    ADV(2);
 	}
-	break;
-      case 3:
-	len = 2;
-	break;
-      case 8:
+    }
+    else if (base == 8) {
 	if (str[0] == '0' && (str[1] == 'o'||str[1] == 'O')) {
-	    str += 2;
+	    ADV(2);
 	}
-      case 4: case 5: case 6: case 7:
-	len = 3;
-	break;
-      case 10:
+    }
+    else if (base == 10) {
 	if (str[0] == '0' && (str[1] == 'd'||str[1] == 'D')) {
-	    str += 2;
+	    ADV(2);
 	}
-      case 9: case 11: case 12: case 13: case 14: case 15:
-	len = 4;
-	break;
-      case 16:
-	len = 4;
+    }
+    else if (base == 16) {
 	if (str[0] == '0' && (str[1] == 'x'||str[1] == 'X')) {
-	    str += 2;
-	}
-	break;
-      default:
-	if (base < 2 || 36 < base) {
-	    rb_raise(rb_eArgError, "illegal radix %d", base);
-	}
-	if (base <= 32) {
-	    len = 5;
-	}
-	else {
-	    len = 6;
+	    ADV(2);
 	}
-	break;
     }
-    if (*str == '0') {		/* squeeze preceeding 0s */
-	while (*++str == '0');
-	if (!(c = *str) || ISSPACE(c)) --str;
+    if (!valid_radix_p(base)) {
+        invalid_radix(base);
+    }
+    if (!len) goto bad;
+    num_digits = str - s;
+    if (*str == '0' && len != 1) { /* squeeze preceding 0s */
+	int us = 0;
+	const char *end = len < 0 ? NULL : str + len;
+	++num_digits;
+	while ((c = *++str) == '0' ||
+	       ((flags & RB_INT_PARSE_UNDERSCORE) && c == '_')) {
+	    if (c == '_') {
+		if (++us >= 2)
+		    break;
+	    }
+	    else {
+		++num_digits;
+		us = 0;
+	    }
+	    if (str == end) break;
+	}
+	if (!c || ISSPACE(c)) --str;
+	if (end) len = end - str;
+	ASSERT_LEN();
     }
     c = *str;
     c = conv_digit(c);
     if (c < 0 || c >= base) {
-	if (badcheck) goto bad;
-	return INT2FIX(0);
+	if (!badcheck && num_digits) z = INT2FIX(0);
+	goto bad;
     }
-    len *= strlen(str)*sizeof(char);
-
-    if (len <= (sizeof(VALUE)*CHAR_BIT)) {
-	unsigned long val = strtoul((char*)str, &end, base);
 
-	if (*end == '_') goto bigparse;
+    if (ndigits) *ndigits = num_digits;
+    val = ruby_scan_digits(str, len, base, &num_digits, &ov);
+    if (!ov) {
+	const char *end = &str[num_digits];
+	if (num_digits > 0 && *end == '_' && (flags & RB_INT_PARSE_UNDERSCORE))
+	    goto bigparse;
+	if (endp) *endp = (char *)end;
+	if (ndigits) *ndigits += num_digits;
 	if (badcheck) {
-	    if (end == str) goto bad; /* no number */
-	    while (*end && ISSPACE(*end)) end++;
-	    if (*end) goto bad;	      /* trailing garbage */
+	    if (num_digits == 0) return Qnil; /* no number */
+	    while (len < 0 ? *end : end < str + len) {
+		if (!ISSPACE(*end)) return Qnil; /* trailing garbage */
+		end++;
+	    }
 	}
 
 	if (POSFIXABLE(val)) {
@@ -460,387 +4190,1028 @@ rb_cstr_to_inum(str, base, badcheck)
 	}
 	else {
 	    VALUE big = rb_uint2big(val);
-	    RBIGNUM(big)->sign = sign;
+	    BIGNUM_SET_SIGN(big, sign);
 	    return bignorm(big);
 	}
     }
-  bigparse:
-    len = (len/BITSPERDIG)+1;
-    if (badcheck && *str == '_') goto bad;
 
-    z = bignew(len, sign);
-    zds = BDIGITS(z);
-    for (i=len;i--;) zds[i]=0;
-    while ((c = *str++) != 0) {
-	if (c == '_') {
-	    if (badcheck) {
-		if (nondigit) goto bad;
-		nondigit = c;
-	    }
-	    continue;
-	}
-	else if ((c = conv_digit(c)) < 0) {
-	    break;
-	}
-	if (c >= base) break;
-	nondigit = 0;
-	i = 0;
-	num = c;
-	for (;;) {
-	    while (i<blen) {
-		num += (BDIGIT_DBL)zds[i]*base;
-		zds[i++] = BIGLO(num);
-		num = BIGDN(num);
-	    }
-	    if (num) {
-		blen++;
-		continue;
-	    }
-	    break;
-	}
+  bigparse:
+    digits_start = str;
+    if (!str2big_scan_digits(s, str, base, badcheck, &num_digits, &len))
+	goto bad;
+    if (endp) *endp = (char *)(str + len);
+    if (ndigits) *ndigits += num_digits;
+    digits_end = digits_start + len;
+
+    if (POW2_P(base)) {
+        z = str2big_poweroftwo(sign, digits_start, digits_end, num_digits,
+			       bit_length(base-1));
     }
-    if (badcheck) {
-	str--;
-	if (s+1 < str && str[-1] == '_') goto bad;
-	while (*str && ISSPACE(*str)) str++;
-	if (*str) {
-	  bad:
-	    rb_invalid_str(s, "Integer");
-	}
+    else {
+        int digits_per_bdigits_dbl;
+        maxpow_in_bdigit_dbl(base, &digits_per_bdigits_dbl);
+        num_bdigits = roomof(num_digits, digits_per_bdigits_dbl)*2;
+
+#ifdef USE_GMP
+        if (GMP_STR2BIG_DIGITS < num_bdigits) {
+            z = str2big_gmp(sign, digits_start, digits_end, num_digits,
+                    num_bdigits, base);
+        }
+        else
+#endif
+        if (num_bdigits < KARATSUBA_MUL_DIGITS) {
+            z = str2big_normal(sign, digits_start, digits_end,
+                    num_bdigits, base);
+        }
+        else {
+            z = str2big_karatsuba(sign, digits_start, digits_end, num_digits,
+                    num_bdigits, digits_per_bdigits_dbl, base);
+        }
     }
 
     return bignorm(z);
 }
 
 VALUE
-rb_str_to_inum(str, base, badcheck)
-    VALUE str;
-    int base;
-    int badcheck;
+rb_cstr_parse_inum(const char *str, ssize_t len, char **endp, int base)
 {
-    char *s;
+    return rb_int_parse_cstr(str, len, endp, NULL, base,
+			     RB_INT_PARSE_DEFAULT);
+}
+
+VALUE
+rb_str_convert_to_inum(VALUE str, int base, int badcheck, int raise_exception)
+{
+    VALUE ret;
+    const char *s;
     long len;
+    char *end;
 
     StringValue(str);
-    if (badcheck) {
-	s = StringValueCStr(str);
+    rb_must_asciicompat(str);
+    RSTRING_GETMEM(str, s, len);
+    ret = rb_cstr_parse_inum(s, len, (badcheck ? NULL : &end), base);
+    if (NIL_P(ret)) {
+        if (badcheck) {
+            if (!raise_exception) return Qnil;
+            invalid_integer(str);
+        }
+        ret = INT2FIX(0);
     }
-    else {
-	s = RSTRING(str)->ptr;
+    return ret;
+}
+
+VALUE
+rb_str_to_inum(VALUE str, int base, int badcheck)
+{
+    return rb_str_convert_to_inum(str, base, badcheck, TRUE);
+}
+
+VALUE
+rb_str2big_poweroftwo(VALUE arg, int base, int badcheck)
+{
+    int positive_p = 1;
+    const char *s, *str;
+    const char *digits_start, *digits_end;
+    size_t num_digits;
+    ssize_t len;
+    VALUE z;
+
+    if (!valid_radix_p(base) || !POW2_P(base)) {
+        invalid_radix(base);
     }
-    if (s) {
-	len = RSTRING(str)->len;
-	if (s[len]) {		/* no sentinel somehow */
-	    char *p = ALLOCA_N(char, len+1);
 
-	    MEMCPY(p, s, char, len);
-	    p[len] = '\0';
-	    s = p;
-	}
+    rb_must_asciicompat(arg);
+    s = str = StringValueCStr(arg);
+    len = RSTRING_LEN(arg);
+    if (*str == '-') {
+	len--;
+        str++;
+        positive_p = 0;
+    }
+
+    digits_start = str;
+    if (!str2big_scan_digits(s, str, base, badcheck, &num_digits, &len))
+	invalid_integer(arg);
+    digits_end = digits_start + len;
+
+    z = str2big_poweroftwo(positive_p, digits_start, digits_end, num_digits,
+            bit_length(base-1));
+
+    RB_GC_GUARD(arg);
+
+    return bignorm(z);
+}
+
+VALUE
+rb_str2big_normal(VALUE arg, int base, int badcheck)
+{
+    int positive_p = 1;
+    const char *s, *str;
+    const char *digits_start, *digits_end;
+    size_t num_digits;
+    ssize_t len;
+    VALUE z;
+
+    int digits_per_bdigits_dbl;
+    size_t num_bdigits;
+
+    if (!valid_radix_p(base)) {
+        invalid_radix(base);
     }
-    return rb_cstr_to_inum(s, base, badcheck); 
+
+    rb_must_asciicompat(arg);
+    s = str = StringValuePtr(arg);
+    len = RSTRING_LEN(arg);
+    if (len > 0 && *str == '-') {
+	len--;
+        str++;
+        positive_p = 0;
+    }
+
+    digits_start = str;
+    if (!str2big_scan_digits(s, str, base, badcheck, &num_digits, &len))
+	invalid_integer(arg);
+    digits_end = digits_start + len;
+
+    maxpow_in_bdigit_dbl(base, &digits_per_bdigits_dbl);
+    num_bdigits = roomof(num_digits, digits_per_bdigits_dbl)*2;
+
+    z = str2big_normal(positive_p, digits_start, digits_end,
+            num_bdigits, base);
+
+    RB_GC_GUARD(arg);
+
+    return bignorm(z);
 }
 
-#if HAVE_LONG_LONG
+VALUE
+rb_str2big_karatsuba(VALUE arg, int base, int badcheck)
+{
+    int positive_p = 1;
+    const char *s, *str;
+    const char *digits_start, *digits_end;
+    size_t num_digits;
+    ssize_t len;
+    VALUE z;
+
+    int digits_per_bdigits_dbl;
+    size_t num_bdigits;
+
+    if (!valid_radix_p(base)) {
+        invalid_radix(base);
+    }
+
+    rb_must_asciicompat(arg);
+    s = str = StringValuePtr(arg);
+    len = RSTRING_LEN(arg);
+    if (len > 0 && *str == '-') {
+	len--;
+        str++;
+        positive_p = 0;
+    }
+
+    digits_start = str;
+    if (!str2big_scan_digits(s, str, base, badcheck, &num_digits, &len))
+	invalid_integer(arg);
+    digits_end = digits_start + len;
+
+    maxpow_in_bdigit_dbl(base, &digits_per_bdigits_dbl);
+    num_bdigits = roomof(num_digits, digits_per_bdigits_dbl)*2;
+
+    z = str2big_karatsuba(positive_p, digits_start, digits_end, num_digits,
+            num_bdigits, digits_per_bdigits_dbl, base);
 
+    RB_GC_GUARD(arg);
+
+    return bignorm(z);
+}
+
+#ifdef USE_GMP
 VALUE
-rb_ull2big(n)
-    unsigned LONG_LONG n;
+rb_str2big_gmp(VALUE arg, int base, int badcheck)
 {
-    BDIGIT_DBL num = n;
-    long i = 0;
-    BDIGIT *digits;
-    VALUE big;
+    int positive_p = 1;
+    const char *s, *str;
+    const char *digits_start, *digits_end;
+    size_t num_digits;
+    ssize_t len;
+    VALUE z;
 
-    big = bignew(DIGSPERLL, 1);
-    digits = BDIGITS(big);
-    while (i < DIGSPERLL) {
-	digits[i++] = BIGLO(num);
-	num = BIGDN(num);
+    int digits_per_bdigits_dbl;
+    size_t num_bdigits;
+
+    if (!valid_radix_p(base)) {
+        invalid_radix(base);
+    }
+
+    rb_must_asciicompat(arg);
+    s = str = StringValuePtr(arg);
+    len = RSTRING_LEN(arg);
+    if (len > 0 && *str == '-') {
+	len--;
+        str++;
+        positive_p = 0;
+    }
+
+    digits_start = str;
+    if (!str2big_scan_digits(s, str, base, badcheck, &num_digits, &len))
+	invalid_integer(arg);
+    digits_end = digits_start + len;
+
+    maxpow_in_bdigit_dbl(base, &digits_per_bdigits_dbl);
+    num_bdigits = roomof(num_digits, digits_per_bdigits_dbl)*2;
+
+    z = str2big_gmp(positive_p, digits_start, digits_end, num_digits, num_bdigits, base);
+
+    RB_GC_GUARD(arg);
+
+    return bignorm(z);
+}
+#endif
+
+#if HAVE_LONG_LONG
+
+static VALUE
+rb_ull2big(unsigned LONG_LONG n)
+{
+    long i;
+    VALUE big = bignew(bdigit_roomof(SIZEOF_LONG_LONG), 1);
+    BDIGIT *digits = BDIGITS(big);
+
+#if SIZEOF_BDIGIT >= SIZEOF_LONG_LONG
+    digits[0] = n;
+#else
+    for (i = 0; i < bdigit_roomof(SIZEOF_LONG_LONG); i++) {
+	digits[i] = BIGLO(n);
+	n = BIGDN(n);
     }
+#endif
 
-    i = DIGSPERLL;
+    i = bdigit_roomof(SIZEOF_LONG_LONG);
     while (i-- && !digits[i]) ;
-    RBIGNUM(big)->len = i+1;
+    BIGNUM_SET_LEN(big, i+1);
     return big;
 }
 
-VALUE
-rb_ll2big(n)
-    LONG_LONG n;
+static VALUE
+rb_ll2big(LONG_LONG n)
 {
     long neg = 0;
+    unsigned LONG_LONG u;
     VALUE big;
 
     if (n < 0) {
-	n = -n;
+        u = 1 + (unsigned LONG_LONG)(-(n + 1)); /* u = -n avoiding overflow */
 	neg = 1;
     }
-    big = rb_ull2big(n);
+    else {
+        u = n;
+    }
+    big = rb_ull2big(u);
     if (neg) {
-	RBIGNUM(big)->sign = 0;
+	BIGNUM_SET_NEGATIVE_SIGN(big);
     }
     return big;
 }
 
 VALUE
-rb_ull2inum(n)
-    unsigned LONG_LONG n;
+rb_ull2inum(unsigned LONG_LONG n)
 {
     if (POSFIXABLE(n)) return LONG2FIX(n);
     return rb_ull2big(n);
 }
 
 VALUE
-rb_ll2inum(n)
-    LONG_LONG n;
+rb_ll2inum(LONG_LONG n)
 {
     if (FIXABLE(n)) return LONG2FIX(n);
     return rb_ll2big(n);
 }
 
 #endif  /* HAVE_LONG_LONG */
- 
+
+#ifdef HAVE_INT128_T
+static VALUE
+rb_uint128t2big(uint128_t n)
+{
+    long i;
+    VALUE big = bignew(bdigit_roomof(SIZEOF_INT128_T), 1);
+    BDIGIT *digits = BDIGITS(big);
+
+    for (i = 0; i < bdigit_roomof(SIZEOF_INT128_T); i++) {
+	digits[i] = BIGLO(RSHIFT(n ,BITSPERDIG*i));
+    }
+
+    i = bdigit_roomof(SIZEOF_INT128_T);
+    while (i-- && !digits[i]) ;
+    BIGNUM_SET_LEN(big, i+1);
+    return big;
+}
+
+MJIT_FUNC_EXPORTED VALUE
+rb_int128t2big(int128_t n)
+{
+    int neg = 0;
+    uint128_t u;
+    VALUE big;
+
+    if (n < 0) {
+        u = 1 + (uint128_t)(-(n + 1)); /* u = -n avoiding overflow */
+	neg = 1;
+    }
+    else {
+        u = n;
+    }
+    big = rb_uint128t2big(u);
+    if (neg) {
+	BIGNUM_SET_NEGATIVE_SIGN(big);
+    }
+    return big;
+}
+#endif
+
 VALUE
-rb_cstr2inum(str, base)
-    const char *str;
-    int base;
+rb_cstr2inum(const char *str, int base)
 {
     return rb_cstr_to_inum(str, base, base==0);
 }
 
 VALUE
-rb_str2inum(str, base)
-    VALUE str;
-    int base;
+rb_str2inum(VALUE str, int base)
 {
     return rb_str_to_inum(str, base, base==0);
 }
 
-const char ruby_digitmap[] = "0123456789abcdefghijklmnopqrstuvwxyz";
-VALUE
-rb_big2str(x, base)
-    VALUE x;
-    int base;
+static VALUE
+big_shift3(VALUE x, int lshift_p, size_t shift_numdigits, int shift_numbits)
 {
-    volatile VALUE t;
-    BDIGIT *ds;
-    long i, j, hbase;
-    VALUE ss;
-    char *s, c;
+    BDIGIT *xds, *zds;
+    long s1;
+    int s2;
+    VALUE z;
+    long xn;
+
+    if (lshift_p) {
+        if (LONG_MAX < shift_numdigits) {
+            rb_raise(rb_eArgError, "too big number");
+        }
+        s1 = shift_numdigits;
+        s2 = shift_numbits;
+        xn = BIGNUM_LEN(x);
+        z = bignew(xn+s1+1, BIGNUM_SIGN(x));
+        zds = BDIGITS(z);
+        BDIGITS_ZERO(zds, s1);
+        xds = BDIGITS(x);
+        zds[xn+s1] = bary_small_lshift(zds+s1, xds, xn, s2);
+    }
+    else {
+        long zn;
+        BDIGIT hibitsx;
+        if (LONG_MAX < shift_numdigits || (size_t)BIGNUM_LEN(x) <= shift_numdigits) {
+            if (BIGNUM_POSITIVE_P(x) ||
+                bary_zero_p(BDIGITS(x), BIGNUM_LEN(x)))
+                return INT2FIX(0);
+            else
+                return INT2FIX(-1);
+        }
+        s1 = shift_numdigits;
+        s2 = shift_numbits;
+        hibitsx = abs2twocomp(&x, &xn);
+        xds = BDIGITS(x);
+        if (xn <= s1) {
+            return hibitsx ? INT2FIX(-1) : INT2FIX(0);
+        }
+        zn = xn - s1;
+        z = bignew(zn, 0);
+        zds = BDIGITS(z);
+        bary_small_rshift(zds, xds+s1, zn, s2, hibitsx != 0 ? BDIGMAX : 0);
+        twocomp2abs_bang(z, hibitsx != 0);
+    }
+    RB_GC_GUARD(x);
+    return z;
+}
 
-    if (FIXNUM_P(x)) {
-	return rb_fix2str(x, base);
+static VALUE
+big_shift2(VALUE x, int lshift_p, VALUE y)
+{
+    int sign;
+    size_t lens[2];
+    size_t shift_numdigits;
+    int shift_numbits;
+
+    assert(POW2_P(CHAR_BIT));
+    assert(POW2_P(BITSPERDIG));
+
+    if (BIGZEROP(x))
+        return INT2FIX(0);
+    sign = rb_integer_pack(y, lens, numberof(lens), sizeof(size_t), 0,
+        INTEGER_PACK_LSWORD_FIRST|INTEGER_PACK_NATIVE_BYTE_ORDER);
+    if (sign < 0) {
+        lshift_p = !lshift_p;
+        sign = -sign;
+    }
+    if (lshift_p) {
+        if (1 < sign || CHAR_BIT <= lens[1])
+            rb_raise(rb_eRangeError, "shift width too big");
+    }
+    else {
+        if (1 < sign || CHAR_BIT <= lens[1])
+            return BIGNUM_POSITIVE_P(x) ? INT2FIX(0) : INT2FIX(-1);
     }
-    i = RBIGNUM(x)->len;
-    if (BIGZEROP(x)) {
-	return rb_str_new2("0");
-    }
-    j = SIZEOF_BDIGITS*CHAR_BIT*i;
-    switch (base) {
-      case 2: break;
-      case 3:
-	j = j * 647L / 1024;
-	break;
-      case 4: case 5: case 6: case 7:
-	j /= 2;
-	break;
-      case 8: case 9:
-	j /= 3;
-	break;
-      case 10: case 11: case 12: case 13: case 14: case 15:
-	j = j * 241L / 800;
-	break;
-      case 16: case 17: case 18: case 19: case 20: case 21:
-      case 22: case 23: case 24: case 25: case 26: case 27:
-      case 28: case 29: case 30: case 31:
-	j /= 4;
-	break;
-      case 32: case 33: case 34: case 35: case 36:
-	j /= 5;
-	break;
-      default:
-	rb_raise(rb_eArgError, "illegal radix %d", base);
-	break;
-    }
-    j += 2;
-
-    hbase = base * base;
-#if SIZEOF_BDIGITS > 2
-    hbase *= hbase;
-#endif
-
-    t = rb_big_clone(x);
-    ds = BDIGITS(t);
-    ss = rb_str_new(0, j);
-    s = RSTRING(ss)->ptr;
-
-    s[0] = RBIGNUM(x)->sign ? '+' : '-';
-    while (i && j) {
-	long k = i;
-	BDIGIT_DBL num = 0;
-
-	while (k--) {
-	    num = BIGUP(num) + ds[k];
-	    ds[k] = (BDIGIT)(num / hbase);
-	    num %= hbase;
+    shift_numbits = (int)(lens[0] & (BITSPERDIG-1));
+    shift_numdigits = (lens[0] >> bit_length(BITSPERDIG-1)) |
+      (lens[1] << (CHAR_BIT*SIZEOF_SIZE_T - bit_length(BITSPERDIG-1)));
+    return big_shift3(x, lshift_p, shift_numdigits, shift_numbits);
+}
+
+static VALUE
+big_lshift(VALUE x, unsigned long shift)
+{
+    long s1 = shift/BITSPERDIG;
+    int s2 = (int)(shift%BITSPERDIG);
+    return big_shift3(x, 1, s1, s2);
+}
+
+static VALUE
+big_rshift(VALUE x, unsigned long shift)
+{
+    long s1 = shift/BITSPERDIG;
+    int s2 = (int)(shift%BITSPERDIG);
+    return big_shift3(x, 0, s1, s2);
+}
+
+#define MAX_BASE36_POWER_TABLE_ENTRIES (SIZEOF_SIZE_T * CHAR_BIT + 1)
+
+static VALUE base36_power_cache[35][MAX_BASE36_POWER_TABLE_ENTRIES];
+static size_t base36_numdigits_cache[35][MAX_BASE36_POWER_TABLE_ENTRIES];
+
+static void
+power_cache_init(void)
+{
+    int i, j;
+    for (i = 0; i < 35; ++i) {
+	for (j = 0; j < MAX_BASE36_POWER_TABLE_ENTRIES; ++j) {
+	    base36_power_cache[i][j] = Qnil;
 	}
-	if (ds[i-1] == 0) i--;
-	k = SIZEOF_BDIGITS;
-	while (k--) {
-	    c = (char)(num % base);
-	    s[--j] = ruby_digitmap[(int)c];
-	    num /= base;
-	    if (i == 0 && num == 0) break;
+    }
+}
+
+static inline VALUE
+power_cache_get_power(int base, int power_level, size_t *numdigits_ret)
+{
+    /*
+     * MAX_BASE36_POWER_TABLE_ENTRIES is big enough to that
+     * base36_power_cache[base][MAX_BASE36_POWER_TABLE_ENTRIES-1] fills whole memory.
+     * So MAX_BASE36_POWER_TABLE_ENTRIES <= power_level is not possible to calculate.
+     *
+     * number-of-bytes =
+     * log256(base36_power_cache[base][MAX_BASE36_POWER_TABLE_ENTRIES-1]) =
+     * log256(maxpow_in_bdigit_dbl(base)**(2**(MAX_BASE36_POWER_TABLE_ENTRIES-1))) =
+     * log256(maxpow_in_bdigit_dbl(base)**(2**(SIZEOF_SIZE_T*CHAR_BIT))) =
+     * (2**(SIZEOF_SIZE_T*CHAR_BIT))*log256(maxpow_in_bdigit_dbl(base)) =
+     * (256**SIZEOF_SIZE_T)*log256(maxpow_in_bdigit_dbl(base)) >
+     * (256**SIZEOF_SIZE_T)*(sizeof(BDIGIT_DBL)-1) >
+     * 256**SIZEOF_SIZE_T
+     */
+    if (MAX_BASE36_POWER_TABLE_ENTRIES <= power_level)
+        rb_bug("too big power number requested: maxpow_in_bdigit_dbl(%d)**(2**%d)", base, power_level);
+
+    if (NIL_P(base36_power_cache[base - 2][power_level])) {
+        VALUE power;
+        size_t numdigits;
+        if (power_level == 0) {
+            int numdigits0;
+            BDIGIT_DBL dd = maxpow_in_bdigit_dbl(base, &numdigits0);
+            power = bignew(2, 1);
+            bdigitdbl2bary(BDIGITS(power), 2, dd);
+            numdigits = numdigits0;
+        }
+        else {
+            power = bigtrunc(bigsq(power_cache_get_power(base, power_level - 1, &numdigits)));
+            numdigits *= 2;
+        }
+        rb_obj_hide(power);
+        base36_power_cache[base - 2][power_level] = power;
+        base36_numdigits_cache[base - 2][power_level] = numdigits;
+	rb_gc_register_mark_object(power);
+    }
+    if (numdigits_ret)
+        *numdigits_ret = base36_numdigits_cache[base - 2][power_level];
+    return base36_power_cache[base - 2][power_level];
+}
+
+struct big2str_struct {
+    int negative;
+    int base;
+    BDIGIT_DBL hbase2;
+    int hbase2_numdigits;
+    VALUE result;
+    char *ptr;
+};
+
+static void
+big2str_alloc(struct big2str_struct *b2s, size_t len)
+{
+    if (LONG_MAX-1 < len)
+        rb_raise(rb_eArgError, "too big number");
+    b2s->result = rb_usascii_str_new(0, (long)(len + 1)); /* plus one for sign */
+    b2s->ptr = RSTRING_PTR(b2s->result);
+    if (b2s->negative)
+        *b2s->ptr++ = '-';
+}
+
+static void
+big2str_2bdigits(struct big2str_struct *b2s, BDIGIT *xds, size_t xn, size_t taillen)
+{
+    size_t j;
+    BDIGIT_DBL num;
+    char buf[SIZEOF_BDIGIT_DBL*CHAR_BIT], *p;
+    int beginning = !b2s->ptr;
+    size_t len = 0;
+
+    assert(xn <= 2);
+    num = bary2bdigitdbl(xds, xn);
+
+    if (beginning) {
+        if (num == 0)
+            return;
+        p = buf;
+        j = sizeof(buf);
+        do {
+            BDIGIT_DBL idx = num % b2s->base;
+            num /= b2s->base;
+            p[--j] = ruby_digitmap[idx];
+        } while (num);
+        len = sizeof(buf) - j;
+        big2str_alloc(b2s, len + taillen);
+	MEMCPY(b2s->ptr, buf + j, char, len);
+    }
+    else {
+        p = b2s->ptr;
+        j = b2s->hbase2_numdigits;
+        do {
+            BDIGIT_DBL idx = num % b2s->base;
+            num /= b2s->base;
+            p[--j] = ruby_digitmap[idx];
+        } while (j);
+        len = b2s->hbase2_numdigits;
+    }
+    b2s->ptr += len;
+}
+
+static void
+big2str_karatsuba(struct big2str_struct *b2s, BDIGIT *xds, size_t xn, size_t wn,
+		  int power_level, size_t taillen)
+{
+    VALUE b;
+    size_t half_numdigits, lower_numdigits;
+    int lower_power_level;
+    size_t bn;
+    const BDIGIT *bds;
+    size_t len;
+
+    /*
+     * Precondition:
+     * abs(x) < maxpow**(2**power_level)
+     * where
+     *   maxpow = maxpow_in_bdigit_dbl(base, &numdigits)
+     *
+     * This function generates sequence of zeros, and then stringized abs(x) into b2s->ptr.
+     *
+     * b2s->ptr can be NULL.
+     * It is allocated when the first character is generated via big2str_alloc.
+     *
+     * The prefix zeros should be generated if and only if b2s->ptr is not NULL.
+     * When the zeros are generated, the zeros and abs(x) consists
+     * numdigits*(2**power_level) characters at total.
+     *
+     * Note:
+     * power_cache_get_power(base, power_level, &len) may not be cached yet. It should not be called.
+     * power_cache_get_power(base, power_level-1, &len) should be cached already if 0 <= power_level-1.
+     */
+
+    if (xn == 0 || bary_zero_p(xds, xn)) {
+	if (b2s->ptr) {
+            /* When x is zero, power_cache_get_power(base, power_level) should be cached already. */
+            power_cache_get_power(b2s->base, power_level, &len);
+	    memset(b2s->ptr, '0', len);
+            b2s->ptr += len;
 	}
+        return;
+    }
+
+    if (power_level == 0) {
+	big2str_2bdigits(b2s, xds, xn, taillen);
+        return;
+    }
+
+    lower_power_level = power_level-1;
+    b = power_cache_get_power(b2s->base, lower_power_level, &lower_numdigits);
+    bn = BIGNUM_LEN(b);
+    bds = BDIGITS(b);
+
+    half_numdigits = lower_numdigits;
+
+    while (0 < lower_power_level &&
+            (xn < bn ||
+             (xn == bn && bary_cmp(xds, xn, bds, bn) < 0))) {
+        lower_power_level--;
+        b = power_cache_get_power(b2s->base, lower_power_level, &lower_numdigits);
+        bn = BIGNUM_LEN(b);
+        bds = BDIGITS(b);
     }
-    while (s[j] == '0') j++;
-    RSTRING(ss)->len -= RBIGNUM(x)->sign?j:j-1;
-    memmove(RBIGNUM(x)->sign?s:s+1, s+j, RSTRING(ss)->len);
-    s[RSTRING(ss)->len] = '\0';
 
-    return ss;
+    if (lower_power_level == 0 &&
+            (xn < bn ||
+             (xn == bn && bary_cmp(xds, xn, bds, bn) < 0))) {
+        if (b2s->ptr) {
+            len = half_numdigits * 2 - lower_numdigits;
+            memset(b2s->ptr, '0', len);
+            b2s->ptr += len;
+        }
+	big2str_2bdigits(b2s, xds, xn, taillen);
+    }
+    else {
+        BDIGIT *qds, *rds;
+        size_t qn, rn;
+        BDIGIT *tds;
+        int shift;
+
+        if (lower_power_level != power_level-1 && b2s->ptr) {
+            len = (half_numdigits - lower_numdigits) * 2;
+            memset(b2s->ptr, '0', len);
+            b2s->ptr += len;
+        }
+
+        shift = nlz(bds[bn-1]);
+
+        qn = xn + BIGDIVREM_EXTRA_WORDS;
+
+        if (shift == 0) {
+            /* bigdivrem_restoring will not modify y.
+             * So use bds directly.  */
+            tds = (BDIGIT *)bds;
+            xds[xn] = 0;
+        }
+        else {
+            /* bigdivrem_restoring will modify y.
+             * So use temporary buffer.  */
+            tds = xds + qn;
+            assert(qn + bn <= xn + wn);
+            bary_small_lshift(tds, bds, bn, shift);
+            xds[xn] = bary_small_lshift(xds, xds, xn, shift);
+        }
+
+        bigdivrem_restoring(xds, qn, tds, bn);
+
+        rds = xds;
+        rn = bn;
+
+        qds = xds + bn;
+        qn = qn - bn;
+
+        if (shift) {
+            bary_small_rshift(rds, rds, rn, shift, 0);
+        }
+
+        BARY_TRUNC(qds, qn);
+        assert(qn <= bn);
+        big2str_karatsuba(b2s, qds, qn, xn+wn - (rn+qn), lower_power_level, lower_numdigits+taillen);
+        BARY_TRUNC(rds, rn);
+        big2str_karatsuba(b2s, rds, rn, xn+wn - rn, lower_power_level, taillen);
+    }
 }
 
-/*
- *  call-seq:
- *     big.to_s(base=10)   =>  string
- *  
- *  Returns a string containing the representation of <i>big</i> radix
- *  <i>base</i> (2 through 36).
- *     
- *     12345654321.to_s         #=> "12345654321"
- *     12345654321.to_s(2)      #=> "1011011111110110111011110000110001"
- *     12345654321.to_s(8)      #=> "133766736061"
- *     12345654321.to_s(16)     #=> "2dfdbbc31"
- *     78546939656932.to_s(36)  #=> "rubyrules"
- */
+static VALUE
+big2str_base_poweroftwo(VALUE x, int base)
+{
+    int word_numbits = ffs(base) - 1;
+    size_t numwords;
+    VALUE result;
+    char *ptr;
+    numwords = rb_absint_numwords(x, word_numbits, NULL);
+    if (BIGNUM_NEGATIVE_P(x)) {
+        if (LONG_MAX-1 < numwords)
+            rb_raise(rb_eArgError, "too big number");
+        result = rb_usascii_str_new(0, 1+numwords);
+        ptr = RSTRING_PTR(result);
+        *ptr++ = BIGNUM_POSITIVE_P(x) ? '+' : '-';
+    }
+    else {
+        if (LONG_MAX < numwords)
+            rb_raise(rb_eArgError, "too big number");
+        result = rb_usascii_str_new(0, numwords);
+        ptr = RSTRING_PTR(result);
+    }
+    rb_integer_pack(x, ptr, numwords, 1, CHAR_BIT-word_numbits,
+                    INTEGER_PACK_BIG_ENDIAN);
+    while (0 < numwords) {
+        *ptr = ruby_digitmap[*(unsigned char *)ptr];
+        ptr++;
+        numwords--;
+    }
+    return result;
+}
+
+VALUE
+rb_big2str_poweroftwo(VALUE x, int base)
+{
+    return big2str_base_poweroftwo(x, base);
+}
 
 static VALUE
-rb_big_to_s(argc, argv, x)
-    int argc;
-    VALUE *argv;
-    VALUE x;
+big2str_generic(VALUE x, int base)
 {
-    VALUE b;
-    int base;
+    BDIGIT *xds;
+    size_t xn;
+    struct big2str_struct b2s_data;
+    int power_level;
+    VALUE power;
 
-    rb_scan_args(argc, argv, "01", &b);
-    if (argc == 0) base = 10;
-    else base = NUM2INT(b);
-    return rb_big2str(x, base);
+    xds = BDIGITS(x);
+    xn = BIGNUM_LEN(x);
+    BARY_TRUNC(xds, xn);
+
+    if (xn == 0) {
+	return rb_usascii_str_new2("0");
+    }
+
+    if (!valid_radix_p(base))
+	invalid_radix(base);
+
+    if (xn >= LONG_MAX/BITSPERDIG) {
+        rb_raise(rb_eRangeError, "bignum too big to convert into `string'");
+    }
+
+    power_level = 0;
+    power = power_cache_get_power(base, power_level, NULL);
+    while (power_level < MAX_BASE36_POWER_TABLE_ENTRIES &&
+           (size_t)BIGNUM_LEN(power) <= (xn+1)/2) {
+        power_level++;
+        power = power_cache_get_power(base, power_level, NULL);
+    }
+    assert(power_level != MAX_BASE36_POWER_TABLE_ENTRIES);
+
+    if ((size_t)BIGNUM_LEN(power) <= xn) {
+        /*
+         * This increment guarantees x < power_cache_get_power(base, power_level)
+         * without invoking it actually.
+         * (power_cache_get_power(base, power_level) can be slow and not used
+         * in big2str_karatsuba.)
+         *
+         * Although it is possible that x < power_cache_get_power(base, power_level-1),
+         * it is no problem because big2str_karatsuba checks it and
+         * doesn't affect the result when b2s_data.ptr is NULL.
+         */
+        power_level++;
+    }
+
+    b2s_data.negative = BIGNUM_NEGATIVE_P(x);
+    b2s_data.base = base;
+    b2s_data.hbase2 = maxpow_in_bdigit_dbl(base, &b2s_data.hbase2_numdigits);
+
+    b2s_data.result = Qnil;
+    b2s_data.ptr = NULL;
+
+    if (power_level == 0) {
+	big2str_2bdigits(&b2s_data, xds, xn, 0);
+    }
+    else {
+        VALUE tmpw = 0;
+        BDIGIT *wds;
+        size_t wn;
+        wn = power_level * BIGDIVREM_EXTRA_WORDS + BIGNUM_LEN(power);
+        wds = ALLOCV_N(BDIGIT, tmpw, xn + wn);
+        MEMCPY(wds, xds, BDIGIT, xn);
+	big2str_karatsuba(&b2s_data, wds, xn, wn, power_level, 0);
+        if (tmpw)
+            ALLOCV_END(tmpw);
+    }
+    RB_GC_GUARD(x);
+
+    *b2s_data.ptr = '\0';
+    rb_str_resize(b2s_data.result, (long)(b2s_data.ptr - RSTRING_PTR(b2s_data.result)));
+
+    RB_GC_GUARD(x);
+    return b2s_data.result;
+}
+
+VALUE
+rb_big2str_generic(VALUE x, int base)
+{
+    return big2str_generic(x, base);
+}
+
+#ifdef USE_GMP
+static VALUE
+big2str_gmp(VALUE x, int base)
+{
+    const size_t nails = (sizeof(BDIGIT)-SIZEOF_BDIGIT)*CHAR_BIT;
+    mpz_t mx;
+    size_t size;
+    VALUE str;
+    BDIGIT *xds = BDIGITS(x);
+    size_t xn = BIGNUM_LEN(x);
+
+    mpz_init(mx);
+    mpz_import(mx, xn, -1, sizeof(BDIGIT), 0, nails, xds);
+
+    size = mpz_sizeinbase(mx, base);
+
+    if (BIGNUM_NEGATIVE_P(x)) {
+        mpz_neg(mx, mx);
+        str = rb_usascii_str_new(0, size+1);
+    }
+    else {
+        str = rb_usascii_str_new(0, size);
+    }
+    mpz_get_str(RSTRING_PTR(str), base, mx);
+    mpz_clear(mx);
+
+    if (RSTRING_PTR(str)[RSTRING_LEN(str)-1] == '\0') {
+        rb_str_set_len(str, RSTRING_LEN(str)-1);
+    }
+
+    RB_GC_GUARD(x);
+    return str;
+}
+
+VALUE
+rb_big2str_gmp(VALUE x, int base)
+{
+    return big2str_gmp(x, base);
+}
+#endif
+
+static VALUE
+rb_big2str1(VALUE x, int base)
+{
+    BDIGIT *xds;
+    size_t xn;
+
+    if (FIXNUM_P(x)) {
+	return rb_fix2str(x, base);
+    }
+
+    bigtrunc(x);
+    xds = BDIGITS(x);
+    xn = BIGNUM_LEN(x);
+    BARY_TRUNC(xds, xn);
+
+    if (xn == 0) {
+	return rb_usascii_str_new2("0");
+    }
+
+    if (!valid_radix_p(base))
+	invalid_radix(base);
+
+    if (xn >= LONG_MAX/BITSPERDIG) {
+        rb_raise(rb_eRangeError, "bignum too big to convert into `string'");
+    }
+
+    if (POW2_P(base)) {
+        /* base == 2 || base == 4 || base == 8 || base == 16 || base == 32 */
+        return big2str_base_poweroftwo(x, base);
+    }
+
+#ifdef USE_GMP
+    if (GMP_BIG2STR_DIGITS < xn) {
+        return big2str_gmp(x, base);
+    }
+#endif
+
+    return big2str_generic(x, base);
+}
+
+VALUE
+rb_big2str(VALUE x, int base)
+{
+    return rb_big2str1(x, base);
 }
 
 static unsigned long
-big2ulong(x, type)
-    VALUE x;
-    char *type;
+big2ulong(VALUE x, const char *type)
 {
-    long len = RBIGNUM(x)->len;
-    BDIGIT_DBL num;
+#if SIZEOF_LONG > SIZEOF_BDIGIT
+    size_t i;
+#endif
+    size_t len = BIGNUM_LEN(x);
+    unsigned long num;
     BDIGIT *ds;
 
-    if (len > SIZEOF_LONG/SIZEOF_BDIGITS)
-	rb_raise(rb_eRangeError, "bignum too big to convert into `%s'", type);
+    if (len == 0)
+        return 0;
+    if (BIGSIZE(x) > sizeof(long)) {
+        rb_raise(rb_eRangeError, "bignum too big to convert into `%s'", type);
+    }
     ds = BDIGITS(x);
+#if SIZEOF_LONG <= SIZEOF_BDIGIT
+    num = (unsigned long)ds[0];
+#else
     num = 0;
-    while (len--) {
-	num = BIGUP(num);
-	num += ds[len];
+    for (i = 0; i < len; i++) {
+	num <<= BITSPERDIG;
+        num += (unsigned long)ds[len - i - 1]; /* overflow is already checked */
     }
+#endif
     return num;
 }
 
 unsigned long
-rb_big2ulong_pack(x)   
-    VALUE x;  
-{   
-    unsigned long num = big2ulong(x, "unsigned long");
-    if (!RBIGNUM(x)->sign) {
-	return -num;
-    }
-    return num;
-}  
-
-unsigned long
-rb_big2ulong(x)
-    VALUE x;
+rb_big2ulong(VALUE x)
 {
     unsigned long num = big2ulong(x, "unsigned long");
 
-    if (!RBIGNUM(x)->sign) {
-	if ((long)num < 0) {
-	    rb_raise(rb_eRangeError, "bignum out of range of unsigned long");
-	}
-	return -num;
+    if (BIGNUM_POSITIVE_P(x)) {
+        return num;
     }
-    return num;
+    else {
+        if (num <= 1+(unsigned long)(-(LONG_MIN+1)))
+            return -(long)(num-1)-1;
+    }
+    rb_raise(rb_eRangeError, "bignum out of range of unsigned long");
 }
 
 long
-rb_big2long(x)
-    VALUE x;
+rb_big2long(VALUE x)
 {
     unsigned long num = big2ulong(x, "long");
 
-    if ((long)num < 0 && (RBIGNUM(x)->sign || (long)num != LONG_MIN)) {
-	rb_raise(rb_eRangeError, "bignum too big to convert into `long'");
+    if (BIGNUM_POSITIVE_P(x)) {
+        if (num <= LONG_MAX)
+            return num;
     }
-    if (!RBIGNUM(x)->sign) return -(long)num;
-    return num;
+    else {
+        if (num <= 1+(unsigned long)(-(LONG_MIN+1)))
+            return -(long)(num-1)-1;
+    }
+    rb_raise(rb_eRangeError, "bignum too big to convert into `long'");
 }
 
 #if HAVE_LONG_LONG
 
 static unsigned LONG_LONG
-big2ull(x, type)
-    VALUE x;
-    char *type;
+big2ull(VALUE x, const char *type)
 {
-    long len = RBIGNUM(x)->len;
-    BDIGIT_DBL num;
-    BDIGIT *ds;
+#if SIZEOF_LONG_LONG > SIZEOF_BDIGIT
+    size_t i;
+#endif
+    size_t len = BIGNUM_LEN(x);
+    unsigned LONG_LONG num;
+    BDIGIT *ds = BDIGITS(x);
 
-    if (len > SIZEOF_LONG_LONG/SIZEOF_BDIGITS)
+    if (len == 0)
+        return 0;
+    if (BIGSIZE(x) > SIZEOF_LONG_LONG)
 	rb_raise(rb_eRangeError, "bignum too big to convert into `%s'", type);
-    ds = BDIGITS(x);
+#if SIZEOF_LONG_LONG <= SIZEOF_BDIGIT
+    num = (unsigned LONG_LONG)ds[0];
+#else
     num = 0;
-    while (len--) {
+    for (i = 0; i < len; i++) {
 	num = BIGUP(num);
-	num += ds[len];
+        num += ds[len - i - 1];
     }
+#endif
     return num;
 }
 
 unsigned LONG_LONG
-rb_big2ull(x)
-    VALUE x;
+rb_big2ull(VALUE x)
 {
     unsigned LONG_LONG num = big2ull(x, "unsigned long long");
 
-    if (!RBIGNUM(x)->sign) return -num;
-    return num;
+    if (BIGNUM_POSITIVE_P(x)) {
+        return num;
+    }
+    else {
+        if (num <= 1+(unsigned LONG_LONG)(-(LLONG_MIN+1)))
+            return -(LONG_LONG)(num-1)-1;
+    }
+    rb_raise(rb_eRangeError, "bignum out of range of unsigned long long");
 }
 
 LONG_LONG
-rb_big2ll(x)
-    VALUE x;
+rb_big2ll(VALUE x)
 {
     unsigned LONG_LONG num = big2ull(x, "long long");
 
-    if ((LONG_LONG)num < 0 && (RBIGNUM(x)->sign
-			       || (LONG_LONG)num != LLONG_MIN)) {
-	rb_raise(rb_eRangeError, "bignum too big to convert into `long long'");
+    if (BIGNUM_POSITIVE_P(x)) {
+        if (num <= LLONG_MAX)
+            return num;
     }
-    if (!RBIGNUM(x)->sign) return -(LONG_LONG)num;
-    return num;
+    else {
+        if (num <= 1+(unsigned LONG_LONG)(-(LLONG_MIN+1)))
+            return -(LONG_LONG)(num-1)-1;
+    }
+    rb_raise(rb_eRangeError, "bignum too big to convert into `long long'");
 }
 
 #endif  /* HAVE_LONG_LONG */
 
 static VALUE
-dbl2big(d)
-    double d;
+dbl2big(double d)
 {
     long i = 0;
     BDIGIT c;
@@ -855,7 +5226,7 @@ dbl2big(d)
 	rb_raise(rb_eFloatDomainError, "NaN");
     }
 
-    while (!POSFIXABLE(u) || 0 != (long)u) {
+    while (1.0 <= u) {
 	u /= (double)(BIGRAD);
 	i++;
     }
@@ -872,1271 +5243,1893 @@ dbl2big(d)
 }
 
 VALUE
-rb_dbl2big(d)
-    double d;
+rb_dbl2big(double d)
 {
     return bignorm(dbl2big(d));
 }
 
-double
-rb_big2dbl(x)
-    VALUE x;
+static double
+big2dbl(VALUE x)
 {
     double d = 0.0;
-    long i = RBIGNUM(x)->len;
-    BDIGIT *ds = BDIGITS(x);
+    long i = (bigtrunc(x), BIGNUM_LEN(x)), lo = 0, bits;
+    BDIGIT *ds = BDIGITS(x), dl;
 
-    while (i--) {
-	d = ds[i] + BIGRAD*d;
+    if (i) {
+	bits = i * BITSPERDIG - nlz(ds[i-1]);
+	if (bits > DBL_MANT_DIG+DBL_MAX_EXP) {
+	    d = HUGE_VAL;
+	}
+	else {
+	    if (bits > DBL_MANT_DIG+1)
+		lo = (bits -= DBL_MANT_DIG+1) / BITSPERDIG;
+	    else
+		bits = 0;
+	    while (--i > lo) {
+		d = ds[i] + BIGRAD*d;
+	    }
+	    dl = ds[i];
+	    if (bits && (dl & ((BDIGIT)1 << (bits %= BITSPERDIG)))) {
+		int carry = (dl & ~(BDIGMAX << bits)) != 0;
+		if (!carry) {
+		    while (i-- > 0) {
+			carry = ds[i] != 0;
+			if (carry) break;
+		    }
+		}
+		if (carry) {
+                    BDIGIT mask = BDIGMAX;
+                    BDIGIT bit = 1;
+                    mask <<= bits;
+                    bit <<= bits;
+                    dl &= mask;
+                    dl += bit;
+                    dl = BIGLO(dl);
+		    if (!dl) d += 1;
+		}
+	    }
+	    d = dl + BIGRAD*d;
+	    if (lo) {
+		if (lo > INT_MAX / BITSPERDIG)
+		    d = HUGE_VAL;
+		else if (lo < INT_MIN / BITSPERDIG)
+		    d = 0.0;
+		else
+		    d = ldexp(d, (int)(lo * BITSPERDIG));
+	    }
+	}
     }
+    if (BIGNUM_NEGATIVE_P(x)) d = -d;
+    return d;
+}
+
+double
+rb_big2dbl(VALUE x)
+{
+    double d = big2dbl(x);
+
     if (isinf(d)) {
-	rb_warn("Bignum out of Float range");
-	d = HUGE_VAL;
+	rb_warning("Bignum out of Float range");
+	if (d < 0.0)
+	    d = -HUGE_VAL;
+	else
+	    d = HUGE_VAL;
     }
-    if (!RBIGNUM(x)->sign) d = -d;
     return d;
 }
 
-/*
- *  call-seq:
- *     big.to_f -> float
- *  
- *  Converts <i>big</i> to a <code>Float</code>. If <i>big</i> doesn't
- *  fit in a <code>Float</code>, the result is infinity.
- *     
- */
+VALUE
+rb_integer_float_cmp(VALUE x, VALUE y)
+{
+    double yd = RFLOAT_VALUE(y);
+    double yi, yf;
+    VALUE rel;
+
+    if (isnan(yd))
+        return Qnil;
+    if (isinf(yd)) {
+        if (yd > 0.0) return INT2FIX(-1);
+        else return INT2FIX(1);
+    }
+    yf = modf(yd, &yi);
+    if (FIXNUM_P(x)) {
+#if SIZEOF_LONG * CHAR_BIT < DBL_MANT_DIG /* assume FLT_RADIX == 2 */
+        double xd = (double)FIX2LONG(x);
+        if (xd < yd)
+            return INT2FIX(-1);
+        if (xd > yd)
+            return INT2FIX(1);
+        return INT2FIX(0);
+#else
+        long xn, yn;
+        if (yi < FIXNUM_MIN)
+            return INT2FIX(1);
+        if (FIXNUM_MAX+1 <= yi)
+            return INT2FIX(-1);
+        xn = FIX2LONG(x);
+        yn = (long)yi;
+        if (xn < yn)
+            return INT2FIX(-1);
+        if (xn > yn)
+            return INT2FIX(1);
+        if (yf < 0.0)
+            return INT2FIX(1);
+        if (0.0 < yf)
+            return INT2FIX(-1);
+        return INT2FIX(0);
+#endif
+    }
+    y = rb_dbl2big(yi);
+    rel = rb_big_cmp(x, y);
+    if (yf == 0.0 || rel != INT2FIX(0))
+        return rel;
+    if (yf < 0.0)
+        return INT2FIX(1);
+    return INT2FIX(-1);
+}
 
-static VALUE
-rb_big_to_f(x)
-    VALUE x;
+VALUE
+rb_integer_float_eq(VALUE x, VALUE y)
 {
-    return rb_float_new(rb_big2dbl(x));
+    double yd = RFLOAT_VALUE(y);
+    double yi, yf;
+
+    if (isnan(yd) || isinf(yd))
+        return Qfalse;
+    yf = modf(yd, &yi);
+    if (yf != 0)
+        return Qfalse;
+    if (FIXNUM_P(x)) {
+#if SIZEOF_LONG * CHAR_BIT < DBL_MANT_DIG /* assume FLT_RADIX == 2 */
+        double xd = (double)FIX2LONG(x);
+        if (xd != yd)
+            return Qfalse;
+        return Qtrue;
+#else
+        long xn, yn;
+        if (yi < LONG_MIN || LONG_MAX < yi)
+            return Qfalse;
+        xn = FIX2LONG(x);
+        yn = (long)yi;
+        if (xn != yn)
+            return Qfalse;
+        return Qtrue;
+#endif
+    }
+    y = rb_dbl2big(yi);
+    return rb_big_eq(x, y);
 }
 
-/*
- *  call-seq:
- *     big <=> numeric   => -1, 0, +1
- *  
- *  Comparison---Returns -1, 0, or +1 depending on whether <i>big</i> is
- *  less than, equal to, or greater than <i>numeric</i>. This is the
- *  basis for the tests in <code>Comparable</code>.
- *     
- */
+VALUE
+rb_big_cmp(VALUE x, VALUE y)
+{
+    if (FIXNUM_P(y)) {
+	x = bigfixize(x);
+        if (FIXNUM_P(x)) {
+	    /* SIGNED_VALUE and Fixnum have same sign-bits, same
+	     * order */
+	    SIGNED_VALUE sx = (SIGNED_VALUE)x, sy = (SIGNED_VALUE)y;
+	    if (sx < sy) return INT2FIX(-1);
+	    return INT2FIX(sx > sy);
+        }
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+	if (BIGNUM_SIGN(x) == BIGNUM_SIGN(y)) {
+	    int cmp = bary_cmp(BDIGITS(x), BIGNUM_LEN(x), BDIGITS(y), BIGNUM_LEN(y));
+	    return INT2FIX(BIGNUM_SIGN(x) ? cmp : -cmp);
+	}
+    }
+    else if (RB_FLOAT_TYPE_P(y)) {
+        return rb_integer_float_cmp(x, y);
+    }
+    else {
+	return rb_num_coerce_cmp(x, y, idCmp);
+    }
+    return INT2FIX(BIGNUM_SIGN(x) ? 1 : -1);
+}
+
+enum big_op_t {
+    big_op_gt,
+    big_op_ge,
+    big_op_lt,
+    big_op_le
+};
 
 static VALUE
-rb_big_cmp(x, y)
-    VALUE x, y;
+big_op(VALUE x, VALUE y, enum big_op_t op)
 {
-    long xlen = RBIGNUM(x)->len;
-
-    switch (TYPE(y)) {
-      case T_FIXNUM:
-	y = rb_int2big(FIX2LONG(y));
-	break;
+    VALUE rel;
+    int n;
 
-      case T_BIGNUM:
-	break;
+    if (RB_INTEGER_TYPE_P(y)) {
+	rel = rb_big_cmp(x, y);
+    }
+    else if (RB_FLOAT_TYPE_P(y)) {
+        rel = rb_integer_float_cmp(x, y);
+    }
+    else {
+	ID id = 0;
+	switch (op) {
+	  case big_op_gt: id = '>'; break;
+	  case big_op_ge: id = idGE; break;
+	  case big_op_lt: id = '<'; break;
+	  case big_op_le: id = idLE; break;
+	}
+	return rb_num_coerce_relop(x, y, id);
+    }
 
-      case T_FLOAT:
-	return rb_dbl_cmp(rb_big2dbl(x), RFLOAT(y)->value);
+    if (NIL_P(rel)) return Qfalse;
+    n = FIX2INT(rel);
 
-      default:
-	return rb_num_coerce_cmp(x, y);
+    switch (op) {
+	case big_op_gt: return n >  0 ? Qtrue : Qfalse;
+	case big_op_ge: return n >= 0 ? Qtrue : Qfalse;
+	case big_op_lt: return n <  0 ? Qtrue : Qfalse;
+	case big_op_le: return n <= 0 ? Qtrue : Qfalse;
     }
+    return Qundef;
+}
 
-    if (RBIGNUM(x)->sign > RBIGNUM(y)->sign) return INT2FIX(1);
-    if (RBIGNUM(x)->sign < RBIGNUM(y)->sign) return INT2FIX(-1);
-    if (xlen < RBIGNUM(y)->len)
-	return (RBIGNUM(x)->sign) ? INT2FIX(-1) : INT2FIX(1);
-    if (xlen > RBIGNUM(y)->len)
-	return (RBIGNUM(x)->sign) ? INT2FIX(1) : INT2FIX(-1);
+VALUE
+rb_big_gt(VALUE x, VALUE y)
+{
+    return big_op(x, y, big_op_gt);
+}
+
+VALUE
+rb_big_ge(VALUE x, VALUE y)
+{
+    return big_op(x, y, big_op_ge);
+}
+
+VALUE
+rb_big_lt(VALUE x, VALUE y)
+{
+    return big_op(x, y, big_op_lt);
+}
 
-    while(xlen-- && (BDIGITS(x)[xlen]==BDIGITS(y)[xlen]));
-    if (-1 == xlen) return INT2FIX(0);
-    return (BDIGITS(x)[xlen] > BDIGITS(y)[xlen]) ?
-	(RBIGNUM(x)->sign ? INT2FIX(1) : INT2FIX(-1)) :
-	    (RBIGNUM(x)->sign ? INT2FIX(-1) : INT2FIX(1));
+VALUE
+rb_big_le(VALUE x, VALUE y)
+{
+    return big_op(x, y, big_op_le);
 }
 
 /*
  *  call-seq:
- *     big == obj  => true or false
- *  
+ *     big == obj  -> true or false
+ *
  *  Returns <code>true</code> only if <i>obj</i> has the same value
- *  as <i>big</i>. Contrast this with <code>Bignum#eql?</code>, which
- *  requires <i>obj</i> to be a <code>Bignum</code>.
- *     
+ *  as <i>big</i>. Contrast this with <code>Integer#eql?</code>, which
+ *  requires <i>obj</i> to be a <code>Integer</code>.
+ *
  *     68719476736 == 68719476736.0   #=> true
  */
 
-static VALUE
-rb_big_eq(x, y)
-    VALUE x, y;
+VALUE
+rb_big_eq(VALUE x, VALUE y)
 {
-    switch (TYPE(y)) {
-      case T_FIXNUM:
-	y = rb_int2big(FIX2LONG(y));
-	break;
-      case T_BIGNUM:
-	break;
-      case T_FLOAT:
-        {
-	    volatile double a, b;
-
-	    a = RFLOAT(y)->value;
-	    if (isnan(a)) return Qfalse;
-	    b = rb_big2dbl(x);
-	    return (a == b)?Qtrue:Qfalse;
-	}
-      default:
+    if (FIXNUM_P(y)) {
+	return bignorm(x) == y ? Qtrue : Qfalse;
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+    }
+    else if (RB_FLOAT_TYPE_P(y)) {
+        return rb_integer_float_eq(x, y);
+    }
+    else {
 	return rb_equal(y, x);
     }
-    if (RBIGNUM(x)->sign != RBIGNUM(y)->sign) return Qfalse;
-    if (RBIGNUM(x)->len != RBIGNUM(y)->len) return Qfalse;
-    if (MEMCMP(BDIGITS(x),BDIGITS(y),BDIGIT,RBIGNUM(y)->len) != 0) return Qfalse;
+    if (BIGNUM_SIGN(x) != BIGNUM_SIGN(y)) return Qfalse;
+    if (BIGNUM_LEN(x) != BIGNUM_LEN(y)) return Qfalse;
+    if (MEMCMP(BDIGITS(x),BDIGITS(y),BDIGIT,BIGNUM_LEN(y)) != 0) return Qfalse;
     return Qtrue;
 }
 
-/*
- *  call-seq:
- *     big.eql?(obj)   => true or false
- *  
- *  Returns <code>true</code> only if <i>obj</i> is a
- *  <code>Bignum</code> with the same value as <i>big</i>. Contrast this
- *  with <code>Bignum#==</code>, which performs type conversions.
- *     
- *     68719476736.eql?(68719476736.0)   #=> false
- */
-
-static VALUE
-rb_big_eql(x, y)
-    VALUE x, y;
+VALUE
+rb_big_eql(VALUE x, VALUE y)
 {
-    if (TYPE(y) != T_BIGNUM) return Qfalse;
-    if (RBIGNUM(x)->sign != RBIGNUM(y)->sign) return Qfalse;
-    if (RBIGNUM(x)->len != RBIGNUM(y)->len) return Qfalse;
-    if (MEMCMP(BDIGITS(x),BDIGITS(y),BDIGIT,RBIGNUM(y)->len) != 0) return Qfalse;
+    if (!RB_BIGNUM_TYPE_P(y)) return Qfalse;
+    if (BIGNUM_SIGN(x) != BIGNUM_SIGN(y)) return Qfalse;
+    if (BIGNUM_LEN(x) != BIGNUM_LEN(y)) return Qfalse;
+    if (MEMCMP(BDIGITS(x),BDIGITS(y),BDIGIT,BIGNUM_LEN(y)) != 0) return Qfalse;
     return Qtrue;
 }
 
-/*
- * call-seq:
- *    -big   =>  other_big
- *
- * Unary minus (returns a new Bignum whose value is 0-big)
- */
-
-static VALUE
-rb_big_uminus(x)
-    VALUE x;
+VALUE
+rb_big_uminus(VALUE x)
 {
     VALUE z = rb_big_clone(x);
 
-    RBIGNUM(z)->sign = !RBIGNUM(x)->sign;
+    BIGNUM_NEGATE(z);
 
     return bignorm(z);
 }
 
-/*
- * call-seq:
- *     ~big  =>  integer
- *
- * Inverts the bits in big. As Bignums are conceptually infinite
- * length, the result acts as if it had an infinite number of one
- * bits to the left. In hex representations, this is displayed
- * as two periods to the left of the digits.
- *  
- *   sprintf("%X", ~0x1122334455)    #=> "..FEEDDCCBBAA"
- */
-
-static VALUE
-rb_big_neg(x)
-    VALUE x;
+VALUE
+rb_big_comp(VALUE x)
 {
     VALUE z = rb_big_clone(x);
-    long i;
-    BDIGIT *ds;
+    BDIGIT *ds = BDIGITS(z);
+    long n = BIGNUM_LEN(z);
+
+    if (!n) return INT2FIX(-1);
 
-    if (!RBIGNUM(x)->sign) get2comp(z);
-    ds = BDIGITS(z);
-    i = RBIGNUM(x)->len;
-    if (!i) return INT2FIX(~0);
-    while (i--) ds[i] = ~ds[i];
-    RBIGNUM(z)->sign = !RBIGNUM(z)->sign;
-    if (RBIGNUM(x)->sign) get2comp(z);
+    if (BIGNUM_POSITIVE_P(z)) {
+        if (bary_add_one(ds, n)) {
+            big_extend_carry(z);
+        }
+        BIGNUM_SET_NEGATIVE_SIGN(z);
+    }
+    else {
+        bary_neg(ds, n);
+        if (bary_add_one(ds, n))
+            return INT2FIX(-1);
+        bary_neg(ds, n);
+        BIGNUM_SET_POSITIVE_SIGN(z);
+    }
 
     return bignorm(z);
 }
 
 static VALUE
-bigsub(x, y)
-    VALUE x, y;
+bigsub(VALUE x, VALUE y)
 {
-    VALUE z = 0;
-    BDIGIT *zds;
-    BDIGIT_DBL_SIGNED num;
-    long i = RBIGNUM(x)->len;
-    
-    /* if x is larger than y, swap */
-    if (RBIGNUM(x)->len < RBIGNUM(y)->len) {
-	z = x; x = y; y = z;	/* swap x y */
-    }
-    else if (RBIGNUM(x)->len == RBIGNUM(y)->len) {
-	while (i > 0) {
-	    i--;
-	    if (BDIGITS(x)[i] > BDIGITS(y)[i]) {
-		break;
-	    }
-	    if (BDIGITS(x)[i] < BDIGITS(y)[i]) {
-		z = x; x = y; y = z;	/* swap x y */
-		break;
-	    }
-	}
+    VALUE z;
+    BDIGIT *xds, *yds, *zds;
+    long xn, yn, zn;
+
+    xn = BIGNUM_LEN(x);
+    yn = BIGNUM_LEN(y);
+    zn = xn < yn ? yn : xn;
+
+    z = bignew(zn, 1);
+
+    xds = BDIGITS(x);
+    yds = BDIGITS(y);
+    zds = BDIGITS(z);
+
+    if (bary_sub(zds, zn, xds, xn, yds, yn)) {
+        bary_2comp(zds, zn);
+        BIGNUM_SET_NEGATIVE_SIGN(z);
     }
 
-    z = bignew(RBIGNUM(x)->len, z==0);
+    return z;
+}
+
+static VALUE bigadd_int(VALUE x, long y);
+
+static VALUE
+bigsub_int(VALUE x, long y0)
+{
+    VALUE z;
+    BDIGIT *xds, *zds;
+    long xn, zn;
+    BDIGIT_DBL_SIGNED num;
+    long i, y;
+
+    y = y0;
+    xds = BDIGITS(x);
+    xn = BIGNUM_LEN(x);
+
+    if (xn == 0)
+        return LONG2NUM(-y0);
+
+    zn = xn;
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    if (zn < bdigit_roomof(SIZEOF_LONG))
+        zn = bdigit_roomof(SIZEOF_LONG);
+#endif
+    z = bignew(zn, BIGNUM_SIGN(x));
     zds = BDIGITS(z);
 
-    for (i = 0, num = 0; i < RBIGNUM(y)->len; i++) { 
-	num += (BDIGIT_DBL_SIGNED)BDIGITS(x)[i] - BDIGITS(y)[i];
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+    assert(xn == zn);
+    num = (BDIGIT_DBL_SIGNED)xds[0] - y;
+    if (xn == 1 && num < 0) {
+	BIGNUM_NEGATE(z);
+	zds[0] = (BDIGIT)-num;
+	RB_GC_GUARD(x);
+	return bignorm(z);
+    }
+    zds[0] = BIGLO(num);
+    num = BIGDN(num);
+    i = 1;
+    if (i < xn)
+        goto y_is_zero_x;
+    goto finish;
+#else
+    num = 0;
+    for (i=0; i < xn; i++) {
+        if (y == 0) goto y_is_zero_x;
+	num += (BDIGIT_DBL_SIGNED)xds[i] - BIGLO(y);
 	zds[i] = BIGLO(num);
 	num = BIGDN(num);
-    } 
-    while (num && i < RBIGNUM(x)->len) {
-	num += BDIGITS(x)[i];
-	zds[i++] = BIGLO(num);
+	y = BIGDN(y);
+    }
+    for (; i < zn; i++) {
+        if (y == 0) goto y_is_zero_z;
+        num -= BIGLO(y);
+        zds[i] = BIGLO(num);
+        num = BIGDN(num);
+        y = BIGDN(y);
+    }
+    goto finish;
+#endif
+
+    for (; i < xn; i++) {
+      y_is_zero_x:
+        if (num == 0) goto num_is_zero_x;
+	num += xds[i];
+	zds[i] = BIGLO(num);
 	num = BIGDN(num);
     }
-    while (i < RBIGNUM(x)->len) {
-	zds[i] = BDIGITS(x)[i];
-	i++;
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    for (; i < zn; i++) {
+      y_is_zero_z:
+        if (num == 0) goto num_is_zero_z;
+        zds[i] = BIGLO(num);
+        num = BIGDN(num);
     }
-    
-    return z;
+#endif
+    goto finish;
+
+    for (; i < xn; i++) {
+      num_is_zero_x:
+	zds[i] = xds[i];
+    }
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    for (; i < zn; i++) {
+      num_is_zero_z:
+        zds[i] = 0;
+    }
+#endif
+    goto finish;
+
+  finish:
+    assert(num == 0 || num == -1);
+    if (num < 0) {
+        get2comp(z);
+	BIGNUM_NEGATE(z);
+    }
+    RB_GC_GUARD(x);
+    return bignorm(z);
 }
 
 static VALUE
-bigadd(x, y, sign)
-    VALUE x, y;
-    int sign;
+bigadd_int(VALUE x, long y)
 {
     VALUE z;
+    BDIGIT *xds, *zds;
+    long xn, zn;
     BDIGIT_DBL num;
-    long i, len;
+    long i;
 
-    sign = (sign == RBIGNUM(y)->sign);
-    if (RBIGNUM(x)->sign != sign) {
-	if (sign) return bigsub(y, x);
-	return bigsub(x, y);
-    }
+    xds = BDIGITS(x);
+    xn = BIGNUM_LEN(x);
+
+    if (xn == 0)
+        return LONG2NUM(y);
+
+    zn = xn;
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    if (zn < bdigit_roomof(SIZEOF_LONG))
+        zn = bdigit_roomof(SIZEOF_LONG);
+#endif
+    zn++;
 
-    if (RBIGNUM(x)->len > RBIGNUM(y)->len) {
-	len = RBIGNUM(x)->len + 1;
-        z = x; x = y; y = z;
+    z = bignew(zn, BIGNUM_SIGN(x));
+    zds = BDIGITS(z);
+
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+    num = (BDIGIT_DBL)xds[0] + y;
+    zds[0] = BIGLO(num);
+    num = BIGDN(num);
+    i = 1;
+    if (i < xn)
+        goto y_is_zero_x;
+    goto y_is_zero_z;
+#else
+    num = 0;
+    for (i=0; i < xn; i++) {
+        if (y == 0) goto y_is_zero_x;
+	num += (BDIGIT_DBL)xds[i] + BIGLO(y);
+	zds[i] = BIGLO(num);
+	num = BIGDN(num);
+	y = BIGDN(y);
     }
-    else {
-	len = RBIGNUM(y)->len + 1;
+    for (; i < zn; i++) {
+        if (y == 0) goto y_is_zero_z;
+	num += BIGLO(y);
+	zds[i] = BIGLO(num);
+	num = BIGDN(num);
+	y = BIGDN(y);
     }
-    z = bignew(len, sign);
+    goto finish;
+
+#endif
 
-    len = RBIGNUM(x)->len;
-    for (i = 0, num = 0; i < len; i++) {
-	num += (BDIGIT_DBL)BDIGITS(x)[i] + BDIGITS(y)[i];
-	BDIGITS(z)[i] = BIGLO(num);
+    for (;i < xn; i++) {
+      y_is_zero_x:
+        if (num == 0) goto num_is_zero_x;
+	num += (BDIGIT_DBL)xds[i];
+	zds[i] = BIGLO(num);
 	num = BIGDN(num);
     }
-    len = RBIGNUM(y)->len;
-    while (num && i < len) {
-	num += BDIGITS(y)[i];
-	BDIGITS(z)[i++] = BIGLO(num);
+    for (; i < zn; i++) {
+      y_is_zero_z:
+        if (num == 0) goto num_is_zero_z;
+	zds[i] = BIGLO(num);
 	num = BIGDN(num);
     }
-    while (i < len) {
-	BDIGITS(z)[i] = BDIGITS(y)[i];
-	i++;
+    goto finish;
+
+    for (;i < xn; i++) {
+      num_is_zero_x:
+	zds[i] = xds[i];
     }
-    BDIGITS(z)[i] = (BDIGIT)num;
+    for (; i < zn; i++) {
+      num_is_zero_z:
+	zds[i] = 0;
+    }
+    goto finish;
 
-    return z;
+  finish:
+    RB_GC_GUARD(x);
+    return bignorm(z);
 }
 
-/*
- *  call-seq:
- *     big + other  => Numeric
- *
- *  Adds big and other, returning the result.
- */
-
-VALUE
-rb_big_plus(x, y)
-    VALUE x, y;
+static VALUE
+bigadd(VALUE x, VALUE y, int sign)
 {
-    switch (TYPE(y)) {
-      case T_FIXNUM:
-	y = rb_int2big(FIX2LONG(y));
-	/* fall through */
-      case T_BIGNUM:
-	return bignorm(bigadd(x, y, 1));
+    VALUE z;
+    size_t len;
 
-      case T_FLOAT:
-	return rb_float_new(rb_big2dbl(x) + RFLOAT(y)->value);
+    sign = (sign == BIGNUM_SIGN(y));
+    if (BIGNUM_SIGN(x) != sign) {
+	if (sign) return bigsub(y, x);
+	return bigsub(x, y);
+    }
 
-      default:
-	return rb_num_coerce_bin(x, y);
+    if (BIGNUM_LEN(x) > BIGNUM_LEN(y)) {
+	len = BIGNUM_LEN(x) + 1;
     }
-}
+    else {
+	len = BIGNUM_LEN(y) + 1;
+    }
+    z = bignew(len, sign);
 
-/*
- *  call-seq:
- *     big - other  => Numeric
- *
- *  Subtracts other from big, returning the result.
- */
+    bary_add(BDIGITS(z), BIGNUM_LEN(z),
+             BDIGITS(x), BIGNUM_LEN(x),
+             BDIGITS(y), BIGNUM_LEN(y));
+
+    return z;
+}
 
 VALUE
-rb_big_minus(x, y)
-    VALUE x, y;
+rb_big_plus(VALUE x, VALUE y)
 {
-    switch (TYPE(y)) {
-      case T_FIXNUM:
-	y = rb_int2big(FIX2LONG(y));
-	/* fall through */
-      case T_BIGNUM:
-	return bignorm(bigadd(x, y, 0));
-
-      case T_FLOAT:
-	return rb_float_new(rb_big2dbl(x) - RFLOAT(y)->value);
+    long n;
 
-      default:
-	return rb_num_coerce_bin(x, y);
+    if (FIXNUM_P(y)) {
+	n = FIX2LONG(y);
+	if ((n > 0) != BIGNUM_SIGN(x)) {
+	    if (n < 0) {
+		n = -n;
+	    }
+	    return bigsub_int(x, n);
+	}
+	if (n < 0) {
+	    n = -n;
+	}
+	return bigadd_int(x, n);
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+	return bignorm(bigadd(x, y, 1));
+    }
+    else if (RB_FLOAT_TYPE_P(y)) {
+	return DBL2NUM(rb_big2dbl(x) + RFLOAT_VALUE(y));
+    }
+    else {
+	return rb_num_coerce_bin(x, y, '+');
     }
 }
 
-/*
- *  call-seq:
- *     big * other  => Numeric
- *
- *  Multiplies big and other, returning the result.
- */
-
 VALUE
-rb_big_mul(x, y)
-    VALUE x, y;
+rb_big_minus(VALUE x, VALUE y)
+{
+    long n;
+
+    if (FIXNUM_P(y)) {
+	n = FIX2LONG(y);
+	if ((n > 0) != BIGNUM_SIGN(x)) {
+	    if (n < 0) {
+		n = -n;
+	    }
+	    return bigadd_int(x, n);
+	}
+	if (n < 0) {
+	    n = -n;
+	}
+	return bigsub_int(x, n);
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+	return bignorm(bigadd(x, y, 0));
+    }
+    else if (RB_FLOAT_TYPE_P(y)) {
+	return DBL2NUM(rb_big2dbl(x) - RFLOAT_VALUE(y));
+    }
+    else {
+	return rb_num_coerce_bin(x, y, '-');
+    }
+}
+
+static VALUE
+bigsq(VALUE x)
 {
-    long i, j;
-    BDIGIT_DBL n = 0;
+    long xn, zn;
     VALUE z;
-    BDIGIT *zds;
+    BDIGIT *xds, *zds;
 
-    if (FIXNUM_P(x)) x = rb_int2big(FIX2LONG(x));
-    switch (TYPE(y)) {
-      case T_FIXNUM:
-	y = rb_int2big(FIX2LONG(y));
-	break;
+    xn = BIGNUM_LEN(x);
+    zn = 2 * xn;
 
-      case T_BIGNUM:
-	break;
+    z = bignew(zn, 1);
 
-      case T_FLOAT:
-	return rb_float_new(rb_big2dbl(x) * RFLOAT(y)->value);
+    xds = BDIGITS(x);
+    zds = BDIGITS(z);
 
-      default:
-	return rb_num_coerce_bin(x, y);
-    }
+    if (xn < NAIVE_MUL_DIGITS)
+        bary_sq_fast(zds, zn, xds, xn);
+    else
+        bary_mul(zds, zn, xds, xn, xds, xn);
+
+    RB_GC_GUARD(x);
+    return z;
+}
+
+static VALUE
+bigmul0(VALUE x, VALUE y)
+{
+    long xn, yn, zn;
+    VALUE z;
+    BDIGIT *xds, *yds, *zds;
+
+    if (x == y)
+        return bigsq(x);
 
-    j = RBIGNUM(x)->len + RBIGNUM(y)->len + 1;
-    z = bignew(j, RBIGNUM(x)->sign==RBIGNUM(y)->sign);
+    xn = BIGNUM_LEN(x);
+    yn = BIGNUM_LEN(y);
+    zn = xn + yn;
+
+    z = bignew(zn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+
+    xds = BDIGITS(x);
+    yds = BDIGITS(y);
     zds = BDIGITS(z);
-    while (j--) zds[j] = 0;
-    for (i = 0; i < RBIGNUM(x)->len; i++) {
-	BDIGIT_DBL dd = BDIGITS(x)[i]; 
-	if (dd == 0) continue;
-	n = 0;
-	for (j = 0; j < RBIGNUM(y)->len; j++) {
-	    BDIGIT_DBL ee = n + (BDIGIT_DBL)dd * BDIGITS(y)[j];
-	    n = zds[i + j] + ee;
-	    if (ee) zds[i + j] = BIGLO(n);
-	    n = BIGDN(n);
-	}
-	if (n) {
-	    zds[i + j] = n;
-	}
+
+    bary_mul(zds, zn, xds, xn, yds, yn);
+
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+    return z;
+}
+
+VALUE
+rb_big_mul(VALUE x, VALUE y)
+{
+    if (FIXNUM_P(y)) {
+	y = rb_int2big(FIX2LONG(y));
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+    }
+    else if (RB_FLOAT_TYPE_P(y)) {
+	return DBL2NUM(rb_big2dbl(x) * RFLOAT_VALUE(y));
+    }
+    else {
+	return rb_num_coerce_bin(x, y, '*');
     }
 
-    return bignorm(z);
+    return bignorm(bigmul0(x, y));
 }
 
-static void
-bigdivrem(x, y, divp, modp)
-    VALUE x, y;
-    VALUE *divp, *modp;
-{
-    long nx = RBIGNUM(x)->len, ny = RBIGNUM(y)->len;
-    long i, j;
-    VALUE yy, z;
-    BDIGIT *xds, *yds, *zds, *tds;
-    BDIGIT_DBL t2;
-    BDIGIT_DBL_SIGNED num;
-    BDIGIT dd, q;
+static VALUE
+bigdivrem(VALUE x, VALUE y, volatile VALUE *divp, volatile VALUE *modp)
+{
+    long xn = BIGNUM_LEN(x), yn = BIGNUM_LEN(y);
+    VALUE z;
+    BDIGIT *xds, *yds, *zds;
+    BDIGIT dd;
+
+    VALUE q = Qnil, r = Qnil;
+    BDIGIT *qds, *rds;
+    long qn, rn;
 
-    if (BIGZEROP(y)) rb_num_zerodiv();
     yds = BDIGITS(y);
-    if (nx < ny || (nx == ny && BDIGITS(x)[nx - 1] < BDIGITS(y)[ny - 1])) {
+    BARY_TRUNC(yds, yn);
+    if (yn == 0)
+        rb_num_zerodiv();
+
+    xds = BDIGITS(x);
+    BARY_TRUNC(xds, xn);
+
+    if (xn < yn || (xn == yn && xds[xn - 1] < yds[yn - 1])) {
 	if (divp) *divp = rb_int2big(0);
 	if (modp) *modp = x;
-	return;
+	return Qnil;
     }
-    xds = BDIGITS(x);
-    if (ny == 1) {
+    if (yn == 1) {
 	dd = yds[0];
-	z = rb_big_clone(x);
+	z = bignew(xn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
 	zds = BDIGITS(z);
-	t2 = 0; i = nx;
-	while (i--) {
-	    t2 = BIGUP(t2) + zds[i];
-	    zds[i] = (BDIGIT)(t2 / dd);
-	    t2 %= dd;
-	}
-	RBIGNUM(z)->sign = RBIGNUM(x)->sign==RBIGNUM(y)->sign;
+        dd = bigdivrem_single(zds, xds, xn, dd);
 	if (modp) {
-	    *modp = rb_uint2big((unsigned long)t2);
-	    RBIGNUM(*modp)->sign = RBIGNUM(x)->sign;
+	    *modp = rb_uint2big((uintptr_t)dd);
+	    BIGNUM_SET_SIGN(*modp, BIGNUM_SIGN(x));
 	}
 	if (divp) *divp = z;
-	return;
+	return Qnil;
+    }
+    if (xn == 2 && yn == 2) {
+        BDIGIT_DBL x0 = bary2bdigitdbl(xds, 2);
+        BDIGIT_DBL y0 = bary2bdigitdbl(yds, 2);
+        BDIGIT_DBL q0 = x0 / y0;
+        BDIGIT_DBL r0 = x0 % y0;
+        if (divp) {
+            z = bignew(bdigit_roomof(sizeof(BDIGIT_DBL)), BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+            zds = BDIGITS(z);
+            zds[0] = BIGLO(q0);
+            zds[1] = BIGLO(BIGDN(q0));
+            *divp = z;
+        }
+        if (modp) {
+            z = bignew(bdigit_roomof(sizeof(BDIGIT_DBL)), BIGNUM_SIGN(x));
+            zds = BDIGITS(z);
+            zds[0] = BIGLO(r0);
+            zds[1] = BIGLO(BIGDN(r0));
+            *modp = z;
+        }
+        return Qnil;
     }
-    z = bignew(nx==ny?nx+2:nx+1, RBIGNUM(x)->sign==RBIGNUM(y)->sign);
-    zds = BDIGITS(z);
-    if (nx==ny) zds[nx+1] = 0;
-    while (!yds[ny-1]) ny--;
 
-    dd = 0;
-    q = yds[ny-1];
-    while ((q & (1<<(BITSPERDIG-1))) == 0) {
-	q <<= 1;
-	dd++;
-    }
-    if (dd) {
-	yy = rb_big_clone(y);
-	tds = BDIGITS(yy);
-	j = 0;
-	t2 = 0;
-	while (j<ny) {
-	    t2 += (BDIGIT_DBL)yds[j]<<dd;
-	    tds[j++] = BIGLO(t2);
-	    t2 = BIGDN(t2);
-	}
-	yds = tds;
-	j = 0;
-	t2 = 0;
-	while (j<nx) {
-	    t2 += (BDIGIT_DBL)xds[j]<<dd;
-	    zds[j++] = BIGLO(t2);
-	    t2 = BIGDN(t2);
-	}
-	zds[j] = (BDIGIT)t2;
+    if (divp) {
+        qn = xn + BIGDIVREM_EXTRA_WORDS;
+        q = bignew(qn, BIGNUM_SIGN(x)==BIGNUM_SIGN(y));
+        qds = BDIGITS(q);
     }
     else {
-	zds[nx] = 0;
-	j = nx;
-	while (j--) zds[j] = xds[j];
+        qn = 0;
+        qds = NULL;
     }
 
-    j = nx==ny?nx+1:nx;
-    do {
-	if (zds[j] ==  yds[ny-1]) q = BIGRAD-1;
-	else q = (BDIGIT)((BIGUP(zds[j]) + zds[j-1])/yds[ny-1]);
-	if (q) {
-	    i = 0; num = 0; t2 = 0;
-	    do {			/* multiply and subtract */
-		BDIGIT_DBL ee;
-		t2 += (BDIGIT_DBL)yds[i] * q;
-		ee = num - BIGLO(t2);
-		num = (BDIGIT_DBL)zds[j - ny + i] + ee;
-		if (ee) zds[j - ny + i] = BIGLO(num);
-		num = BIGDN(num);
-		t2 = BIGDN(t2);
-	    } while (++i < ny);
-	    num += zds[j - ny + i] - t2;/* borrow from high digit; don't update */
-	    while (num) {		/* "add back" required */
-		i = 0; num = 0; q--;
-		do {
-		    BDIGIT_DBL ee = num + yds[i];
-		    num = (BDIGIT_DBL)zds[j - ny + i] + ee;
-		    if (ee) zds[j - ny + i] = BIGLO(num);
-		    num = BIGDN(num);
-		} while (++i < ny);
-		num--;
-	    }
-	}
-	zds[j] = q;
-    } while (--j >= ny);
-    if (divp) {			/* move quotient down in z */
-	*divp = rb_big_clone(z);
-	zds = BDIGITS(*divp);
-	j = (nx==ny ? nx+2 : nx+1) - ny;
-	for (i = 0;i < j;i++) zds[i] = zds[i+ny];
-	RBIGNUM(*divp)->len = i;
-    }
-    if (modp) {			/* normalize remainder */
-	*modp = rb_big_clone(z);
-	zds = BDIGITS(*modp);
-	while (--ny && !zds[ny]); ++ny;
-	if (dd) {
-	    t2 = 0; i = ny;
-	    while(i--) {
-		t2 = (t2 | zds[i]) >> dd;
-		q = zds[i];
-		zds[i] = BIGLO(t2);
-		t2 = BIGUP(q);
-	    }
-	}
-	RBIGNUM(*modp)->len = ny;
-	RBIGNUM(*modp)->sign = RBIGNUM(x)->sign;
+    if (modp) {
+        rn = yn;
+        r = bignew(rn, BIGNUM_SIGN(x));
+        rds = BDIGITS(r);
+    }
+    else {
+        rn = 0;
+        rds = NULL;
+    }
+
+    bary_divmod_branch(qds, qn, rds, rn, xds, xn, yds, yn);
+
+    if (divp) {
+        bigtrunc(q);
+        *divp = q;
     }
+    if (modp) {
+        bigtrunc(r);
+        *modp = r;
+    }
+
+    return Qnil;
 }
 
 static void
-bigdivmod(x, y, divp, modp)
-    VALUE x, y;
-    VALUE *divp, *modp;
+bigdivmod(VALUE x, VALUE y, volatile VALUE *divp, volatile VALUE *modp)
 {
     VALUE mod;
 
     bigdivrem(x, y, divp, &mod);
-    if (RBIGNUM(x)->sign != RBIGNUM(y)->sign && !BIGZEROP(mod)) {
+    if (BIGNUM_SIGN(x) != BIGNUM_SIGN(y) && !BIGZEROP(mod)) {
 	if (divp) *divp = bigadd(*divp, rb_int2big(1), 0);
 	if (modp) *modp = bigadd(mod, y, 1);
     }
-    else {
-	if (divp) *divp = *divp;
-	if (modp) *modp = mod;
+    else if (modp) {
+	*modp = mod;
     }
 }
 
-/*
- *  call-seq:
- *     big / other     => Numeric
- *     big.div(other)  => Numeric
- *
- *  Divides big by other, returning the result.
- */
 
 static VALUE
-rb_big_div(x, y)
-    VALUE x, y;
+rb_big_divide(VALUE x, VALUE y, ID op)
 {
     VALUE z;
 
-    switch (TYPE(y)) {
-      case T_FIXNUM:
+    if (FIXNUM_P(y)) {
 	y = rb_int2big(FIX2LONG(y));
-	break;
-
-      case T_BIGNUM:
-	break;
-
-      case T_FLOAT:
-	return rb_float_new(rb_big2dbl(x) / RFLOAT(y)->value);
-
-      default:
-	return rb_num_coerce_bin(x, y);
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+    }
+    else if (RB_FLOAT_TYPE_P(y)) {
+	if (op == '/') {
+            double dx = rb_big2dbl(x);
+            return rb_flo_div_flo(DBL2NUM(dx), y);
+	}
+	else {
+            VALUE v;
+	    double dy = RFLOAT_VALUE(y);
+	    if (dy == 0.0) rb_num_zerodiv();
+            v = rb_big_divide(x, y, '/');
+            return rb_dbl2big(RFLOAT_VALUE(v));
+	}
+    }
+    else {
+	return rb_num_coerce_bin(x, y, op);
     }
     bigdivmod(x, y, &z, 0);
 
     return bignorm(z);
 }
 
-/*
- *  call-seq:
- *     big % other         => Numeric
- *     big.modulo(other)   => Numeric
- *
- *  Returns big modulo other. See Numeric.divmod for more
- *  information.
- */
+VALUE
+rb_big_div(VALUE x, VALUE y)
+{
+    return rb_big_divide(x, y, '/');
+}
 
-static VALUE
-rb_big_modulo(x, y)
-    VALUE x, y;
+VALUE
+rb_big_idiv(VALUE x, VALUE y)
+{
+    return rb_big_divide(x, y, rb_intern("div"));
+}
+
+VALUE
+rb_big_modulo(VALUE x, VALUE y)
 {
     VALUE z;
 
-    switch (TYPE(y)) {
-      case T_FIXNUM:
+    if (FIXNUM_P(y)) {
 	y = rb_int2big(FIX2LONG(y));
-	break;
-
-      case T_BIGNUM:
-	break;
-
-      default:
-	return rb_num_coerce_bin(x, y);
+    }
+    else if (!RB_BIGNUM_TYPE_P(y)) {
+	return rb_num_coerce_bin(x, y, '%');
     }
     bigdivmod(x, y, 0, &z);
 
     return bignorm(z);
 }
 
-/*
- *  call-seq:
- *     big.remainder(numeric)    => number
- *  
- *  Returns the remainder after dividing <i>big</i> by <i>numeric</i>.
- *     
- *     -1234567890987654321.remainder(13731)      #=> -6966
- *     -1234567890987654321.remainder(13731.24)   #=> -9906.22531493148
- */
-static VALUE
-rb_big_remainder(x, y)
-    VALUE x, y;
+VALUE
+rb_big_remainder(VALUE x, VALUE y)
 {
     VALUE z;
 
-    switch (TYPE(y)) {
-      case T_FIXNUM:
+    if (FIXNUM_P(y)) {
 	y = rb_int2big(FIX2LONG(y));
-	break;
-
-      case T_BIGNUM:
-	break;
-
-      default:
-	return rb_num_coerce_bin(x, y);
+    }
+    else if (!RB_BIGNUM_TYPE_P(y)) {
+	return rb_num_coerce_bin(x, y, rb_intern("remainder"));
     }
     bigdivrem(x, y, 0, &z);
 
     return bignorm(z);
 }
 
-static VALUE big_lshift _((VALUE, unsigned long));
-static VALUE big_rshift _((VALUE, unsigned long));
-
-static VALUE big_shift(x, n)
-    VALUE x;
-    int n;
-{
-    if (n < 0)
-	return big_lshift(x, (unsigned int)n);
-    else if (n > 0)
-	return big_rshift(x, (unsigned int)n);
-    return x;
-}
-
-/*
- *  call-seq:
- *     big.divmod(numeric)   => array
- *  
- *  See <code>Numeric#divmod</code>.
- *     
- */
 VALUE
-rb_big_divmod(x, y)
-    VALUE x, y;
+rb_big_divmod(VALUE x, VALUE y)
 {
     VALUE div, mod;
 
-    switch (TYPE(y)) {
-      case T_FIXNUM:
+    if (FIXNUM_P(y)) {
 	y = rb_int2big(FIX2LONG(y));
-	break;
-
-      case T_BIGNUM:
-	break;
-
-      default:
-	return rb_num_coerce_bin(x, y);
+    }
+    else if (!RB_BIGNUM_TYPE_P(y)) {
+	return rb_num_coerce_bin(x, y, rb_intern("divmod"));
     }
     bigdivmod(x, y, &div, &mod);
 
     return rb_assoc_new(bignorm(div), bignorm(mod));
 }
 
-/*
- *  call-seq:
- *     big.quo(numeric) -> float
- *  
- *  Returns the floating point result of dividing <i>big</i> by
- *  <i>numeric</i>.
- *     
- *     -1234567890987654321.quo(13731)      #=> -89910996357705.5
- *     -1234567890987654321.quo(13731.24)   #=> -89909424858035.7
- *     
- */
-
 static VALUE
-rb_big_quo(x, y)
-    VALUE x, y;
+big_shift(VALUE x, long n)
+{
+    if (n < 0)
+	return big_lshift(x, 1+(unsigned long)(-(n+1)));
+    else if (n > 0)
+	return big_rshift(x, (unsigned long)n);
+    return x;
+}
+
+enum {DBL_BIGDIG = ((DBL_MANT_DIG + BITSPERDIG) / BITSPERDIG)};
+
+static double
+big_fdiv(VALUE x, VALUE y, long ey)
 {
-    double dx = rb_big2dbl(x);
-    double dy;
+    VALUE z;
+    long l, ex;
+
+    bigtrunc(x);
+    l = BIGNUM_LEN(x);
+    ex = l * BITSPERDIG - nlz(BDIGITS(x)[l-1]);
+    ex -= 2 * DBL_BIGDIG * BITSPERDIG;
+    if (ex > BITSPERDIG) ex -= BITSPERDIG;
+    else if (ex > 0) ex = 0;
+    if (ex) x = big_shift(x, ex);
+
+    bigdivrem(x, y, &z, 0);
+    l = ex - ey;
+#if SIZEOF_LONG > SIZEOF_INT
+    {
+	/* Visual C++ can't be here */
+	if (l > INT_MAX) return HUGE_VAL;
+	if (l < INT_MIN) return 0.0;
+    }
+#endif
+    return ldexp(big2dbl(z), (int)l);
+}
 
-    switch (TYPE(y)) {
-      case T_FIXNUM:
-	dy = (double)FIX2LONG(y);
-	break;
+static double
+big_fdiv_int(VALUE x, VALUE y)
+{
+    long l, ey;
+    bigtrunc(y);
+    l = BIGNUM_LEN(y);
+    ey = l * BITSPERDIG - nlz(BDIGITS(y)[l-1]);
+    ey -= DBL_BIGDIG * BITSPERDIG;
+    if (ey) y = big_shift(y, ey);
+    return big_fdiv(x, y, ey);
+}
 
-      case T_BIGNUM:
-	dy = rb_big2dbl(y);
-	break;
+static double
+big_fdiv_float(VALUE x, VALUE y)
+{
+    int i;
+    y = dbl2big(ldexp(frexp(RFLOAT_VALUE(y), &i), DBL_MANT_DIG));
+    return big_fdiv(x, y, i - DBL_MANT_DIG);
+}
 
-      case T_FLOAT:
-	dy = RFLOAT(y)->value;
-	break;
+double
+rb_big_fdiv_double(VALUE x, VALUE y)
+{
+    double dx, dy;
+    VALUE v;
 
-      default:
-	return rb_num_coerce_bin(x, y);
+    dx = big2dbl(x);
+    if (FIXNUM_P(y)) {
+	dy = (double)FIX2LONG(y);
+	if (isinf(dx))
+	    return big_fdiv_int(x, rb_int2big(FIX2LONG(y)));
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+	return big_fdiv_int(x, y);
     }
-    return rb_float_new(dx / dy);
+    else if (RB_FLOAT_TYPE_P(y)) {
+	dy = RFLOAT_VALUE(y);
+	if (isnan(dy))
+	    return dy;
+	if (isinf(dx))
+	    return big_fdiv_float(x, y);
+    }
+    else {
+	return NUM2DBL(rb_num_coerce_bin(x, y, rb_intern("fdiv")));
+    }
+    v = rb_flo_div_flo(DBL2NUM(dx), DBL2NUM(dy));
+    return NUM2DBL(v);
 }
 
-/*
- *  call-seq:
- *     big ** exponent   #=> numeric
- *
- *  Raises _big_ to the _exponent_ power (which may be an integer, float,
- *  or anything that will coerce to a number). The result may be
- *  a Fixnum, Bignum, or Float
- *
- *    123456789 ** 2      #=> 15241578750190521
- *    123456789 ** 1.2    #=> 5126464716.09932
- *    123456789 ** -2     #=> 6.5610001194102e-17
- */
+VALUE
+rb_big_fdiv(VALUE x, VALUE y)
+{
+    return DBL2NUM(rb_big_fdiv_double(x, y));
+}
 
 VALUE
-rb_big_pow(x, y)
-    VALUE x, y;
+rb_big_pow(VALUE x, VALUE y)
 {
     double d;
-    long yy;
-    
-    if (y == INT2FIX(0)) return INT2FIX(1);
-    switch (TYPE(y)) {
-      case T_FLOAT:
-	d = RFLOAT(y)->value;
-	break;
+    SIGNED_VALUE yy;
 
-      case T_BIGNUM:
+  again:
+    if (y == INT2FIX(0)) return INT2FIX(1);
+    if (y == INT2FIX(1)) return x;
+    if (RB_FLOAT_TYPE_P(y)) {
+	d = RFLOAT_VALUE(y);
+	if ((BIGNUM_NEGATIVE_P(x) && !BIGZEROP(x))) {
+            return rb_dbl_complex_new_polar_pi(pow(-rb_big2dbl(x), d), d);
+	}
+    }
+    else if (RB_BIGNUM_TYPE_P(y)) {
+	y = bignorm(y);
+	if (FIXNUM_P(y))
+	    goto again;
 	rb_warn("in a**b, b may be too big");
 	d = rb_big2dbl(y);
-	break;
-
-      case T_FIXNUM:
+    }
+    else if (FIXNUM_P(y)) {
 	yy = FIX2LONG(y);
-	if (yy > 0) {
-	    VALUE z = x;
-	    const long BIGLEN_LIMIT = 1024*1024 / SIZEOF_BDIGITS;
 
-	    if ((RBIGNUM(x)->len > BIGLEN_LIMIT) ||
-		(RBIGNUM(x)->len > BIGLEN_LIMIT / yy)) {
+        if (yy < 0) {
+            x = rb_big_pow(x, INT2NUM(-yy));
+            if (RB_INTEGER_TYPE_P(x))
+                return rb_rational_raw(INT2FIX(1), x);
+            else
+                return DBL2NUM(1.0 / NUM2DBL(x));
+        }
+	else {
+	    VALUE z = 0;
+	    SIGNED_VALUE mask;
+            const size_t xbits = rb_absint_numwords(x, 1, NULL);
+	    const size_t BIGLEN_LIMIT = 32*1024*1024;
+
+	    if (xbits == (size_t)-1 ||
+                (xbits > BIGLEN_LIMIT) ||
+                (xbits * yy > BIGLEN_LIMIT)) {
 		rb_warn("in a**b, b may be too big");
 		d = (double)yy;
-		break;
 	    }
-	    for (;;) {
-		yy -= 1;
-		if (yy == 0) break;
-		while (yy % 2 == 0) {
-		    yy /= 2;
-		    x = rb_big_mul(x, x);
+	    else {
+		for (mask = FIXNUM_MAX + 1; mask; mask >>= 1) {
+		    if (z) z = bigsq(z);
+		    if (yy & mask) {
+			z = z ? bigtrunc(bigmul0(z, x)) : x;
+		    }
 		}
-		z = rb_big_mul(z, x);
+		return bignorm(z);
 	    }
-	    return bignorm(z);
 	}
-	d = (double)yy;
-	break;
-
-      default:
-	return rb_num_coerce_bin(x, y);
     }
-    return rb_float_new(pow(rb_big2dbl(x), d));
+    else {
+	return rb_num_coerce_bin(x, y, idPow);
+    }
+    return DBL2NUM(pow(rb_big2dbl(x), d));
 }
 
-/*
- * call-seq:
- *     big & numeric   =>  integer
- *
- * Performs bitwise +and+ between _big_ and _numeric_.
- */
+static VALUE
+bigand_int(VALUE x, long xn, BDIGIT hibitsx, long y)
+{
+    VALUE z;
+    BDIGIT *xds, *zds;
+    long zn;
+    long i;
+    BDIGIT hibitsy;
+
+    if (y == 0) return INT2FIX(0);
+    if (xn == 0) return hibitsx ? LONG2NUM(y) : 0;
+    hibitsy = 0 <= y ? 0 : BDIGMAX;
+    xds = BDIGITS(x);
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+    if (!hibitsy) {
+	y &= xds[0];
+	return LONG2NUM(y);
+    }
+#endif
+
+    zn = xn;
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    if (hibitsx && zn < bdigit_roomof(SIZEOF_LONG))
+        zn = bdigit_roomof(SIZEOF_LONG);
+#endif
+
+    z = bignew(zn, 0);
+    zds = BDIGITS(z);
+
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+    i = 1;
+    zds[0] = xds[0] & BIGLO(y);
+#else
+    for (i=0; i < xn; i++) {
+        if (y == 0 || y == -1) break;
+        zds[i] = xds[i] & BIGLO(y);
+        y = BIGDN(y);
+    }
+    for (; i < zn; i++) {
+        if (y == 0 || y == -1) break;
+        zds[i] = hibitsx & BIGLO(y);
+        y = BIGDN(y);
+    }
+#endif
+    for (;i < xn; i++) {
+	zds[i] = xds[i] & hibitsy;
+    }
+    for (;i < zn; i++) {
+	zds[i] = hibitsx & hibitsy;
+    }
+    twocomp2abs_bang(z, hibitsx && hibitsy);
+    RB_GC_GUARD(x);
+    return bignorm(z);
+}
 
 VALUE
-rb_big_and(xx, yy)
-    VALUE xx, yy;
+rb_big_and(VALUE x, VALUE y)
 {
-    volatile VALUE x, y, z;
+    VALUE z;
     BDIGIT *ds1, *ds2, *zds;
-    long i, l1, l2;
-    char sign;
+    long i, xn, yn, n1, n2;
+    BDIGIT hibitsx, hibitsy;
+    BDIGIT hibits1, hibits2;
+    VALUE tmpv;
+    BDIGIT tmph;
+    long tmpn;
+
+    if (!RB_INTEGER_TYPE_P(y)) {
+	return rb_num_coerce_bit(x, y, '&');
+    }
 
-    x = xx;
-    y = rb_to_int(yy);
+    hibitsx = abs2twocomp(&x, &xn);
     if (FIXNUM_P(y)) {
-	y = rb_int2big(FIX2LONG(y));
-    }
-    if (!RBIGNUM(y)->sign) {
-	y = rb_big_clone(y);
-	get2comp(y);
-    }
-    if (!RBIGNUM(x)->sign) {
-	x = rb_big_clone(x);
-	get2comp(x);
+	return bigand_int(x, xn, hibitsx, FIX2LONG(y));
     }
-    if (RBIGNUM(x)->len > RBIGNUM(y)->len) {
-	l1 = RBIGNUM(y)->len;
-	l2 = RBIGNUM(x)->len;
-	ds1 = BDIGITS(y);
-	ds2 = BDIGITS(x);
-	sign = RBIGNUM(y)->sign;
+    hibitsy = abs2twocomp(&y, &yn);
+    if (xn > yn) {
+        tmpv = x; x = y; y = tmpv;
+        tmpn = xn; xn = yn; yn = tmpn;
+        tmph = hibitsx; hibitsx = hibitsy; hibitsy = tmph;
     }
-    else {
-	l1 = RBIGNUM(x)->len;
-	l2 = RBIGNUM(y)->len;
-	ds1 = BDIGITS(x);
-	ds2 = BDIGITS(y);
-	sign = RBIGNUM(x)->sign;
-    }
-    z = bignew(l2, RBIGNUM(x)->sign || RBIGNUM(y)->sign);
+    n1 = xn;
+    n2 = yn;
+    ds1 = BDIGITS(x);
+    ds2 = BDIGITS(y);
+    hibits1 = hibitsx;
+    hibits2 = hibitsy;
+
+    if (!hibits1)
+        n2 = n1;
+
+    z = bignew(n2, 0);
     zds = BDIGITS(z);
 
-    for (i=0; i<l1; i++) {
+    for (i=0; i<n1; i++) {
 	zds[i] = ds1[i] & ds2[i];
     }
-    for (; i<l2; i++) {
-	zds[i] = sign?0:ds2[i];
+    for (; i<n2; i++) {
+	zds[i] = hibits1 & ds2[i];
     }
-    if (!RBIGNUM(z)->sign) get2comp(z);
+    twocomp2abs_bang(z, hibits1 && hibits2);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
     return bignorm(z);
 }
 
-/*
- * call-seq:
- *     big | numeric   =>  integer
- *
- * Performs bitwise +or+ between _big_ and _numeric_.
- */
+static VALUE
+bigor_int(VALUE x, long xn, BDIGIT hibitsx, long y)
+{
+    VALUE z;
+    BDIGIT *xds, *zds;
+    long zn;
+    long i;
+    BDIGIT hibitsy;
+
+    if (y == -1) return INT2FIX(-1);
+    if (xn == 0) return hibitsx ? INT2FIX(-1) : LONG2FIX(y);
+    hibitsy = 0 <= y ? 0 : BDIGMAX;
+    xds = BDIGITS(x);
+
+    zn = BIGNUM_LEN(x);
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    if (zn < bdigit_roomof(SIZEOF_LONG))
+        zn = bdigit_roomof(SIZEOF_LONG);
+#endif
+    z = bignew(zn, 0);
+    zds = BDIGITS(z);
+
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+    i = 1;
+    zds[0] = xds[0] | BIGLO(y);
+    if (i < zn)
+        goto y_is_fixed_point;
+    goto finish;
+#else
+    for (i=0; i < xn; i++) {
+        if (y == 0 || y == -1) goto y_is_fixed_point;
+        zds[i] = xds[i] | BIGLO(y);
+        y = BIGDN(y);
+    }
+    if (hibitsx)
+        goto fill_hibits;
+    for (; i < zn; i++) {
+        if (y == 0 || y == -1) goto y_is_fixed_point;
+        zds[i] = BIGLO(y);
+        y = BIGDN(y);
+    }
+  goto finish;
+#endif
+
+  y_is_fixed_point:
+    if (hibitsy)
+        goto fill_hibits;
+    for (; i < xn; i++) {
+        zds[i] = xds[i];
+    }
+    if (hibitsx)
+        goto fill_hibits;
+    for (; i < zn; i++) {
+        zds[i] = 0;
+    }
+  goto finish;
+
+  fill_hibits:
+    for (; i < zn; i++) {
+        zds[i] = BDIGMAX;
+    }
+
+  finish:
+    twocomp2abs_bang(z, hibitsx || hibitsy);
+    RB_GC_GUARD(x);
+    return bignorm(z);
+}
 
 VALUE
-rb_big_or(xx, yy)
-    VALUE xx, yy;
+rb_big_or(VALUE x, VALUE y)
 {
-    volatile VALUE x, y, z;
+    VALUE z;
     BDIGIT *ds1, *ds2, *zds;
-    long i, l1, l2;
-    char sign;
+    long i, xn, yn, n1, n2;
+    BDIGIT hibitsx, hibitsy;
+    BDIGIT hibits1, hibits2;
+    VALUE tmpv;
+    BDIGIT tmph;
+    long tmpn;
+
+    if (!RB_INTEGER_TYPE_P(y)) {
+	return rb_num_coerce_bit(x, y, '|');
+    }
 
-    x = xx;
-    y = rb_to_int(yy);
+    hibitsx = abs2twocomp(&x, &xn);
     if (FIXNUM_P(y)) {
-	y = rb_int2big(FIX2LONG(y));
-    }
-    if (!RBIGNUM(y)->sign) {
-	y = rb_big_clone(y);
-	get2comp(y);
+	return bigor_int(x, xn, hibitsx, FIX2LONG(y));
     }
-    if (!RBIGNUM(x)->sign) {
-	x = rb_big_clone(x);
-	get2comp(x);
-    }
-    if (RBIGNUM(x)->len > RBIGNUM(y)->len) {
-	l1 = RBIGNUM(y)->len;
-	l2 = RBIGNUM(x)->len;
-	ds1 = BDIGITS(y);
-	ds2 = BDIGITS(x);
-	sign = RBIGNUM(y)->sign;
-    }
-    else {
-	l1 = RBIGNUM(x)->len;
-	l2 = RBIGNUM(y)->len;
-	ds1 = BDIGITS(x);
-	ds2 = BDIGITS(y);
-	sign = RBIGNUM(x)->sign;
+    hibitsy = abs2twocomp(&y, &yn);
+    if (xn > yn) {
+        tmpv = x; x = y; y = tmpv;
+        tmpn = xn; xn = yn; yn = tmpn;
+        tmph = hibitsx; hibitsx = hibitsy; hibitsy = tmph;
     }
-    z = bignew(l2, RBIGNUM(x)->sign && RBIGNUM(y)->sign);
+    n1 = xn;
+    n2 = yn;
+    ds1 = BDIGITS(x);
+    ds2 = BDIGITS(y);
+    hibits1 = hibitsx;
+    hibits2 = hibitsy;
+
+    if (hibits1)
+        n2 = n1;
+
+    z = bignew(n2, 0);
     zds = BDIGITS(z);
 
-    for (i=0; i<l1; i++) {
+    for (i=0; i<n1; i++) {
 	zds[i] = ds1[i] | ds2[i];
     }
-    for (; i<l2; i++) {
-	zds[i] = sign?ds2[i]:(BIGRAD-1);
+    for (; i<n2; i++) {
+	zds[i] = hibits1 | ds2[i];
     }
-    if (!RBIGNUM(z)->sign) get2comp(z);
-
+    twocomp2abs_bang(z, hibits1 || hibits2);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
     return bignorm(z);
 }
 
-/*
- * call-seq:
- *     big ^ numeric   =>  integer
- *
- * Performs bitwise +exclusive or+ between _big_ and _numeric_.
- */
+static VALUE
+bigxor_int(VALUE x, long xn, BDIGIT hibitsx, long y)
+{
+    VALUE z;
+    BDIGIT *xds, *zds;
+    long zn;
+    long i;
+    BDIGIT hibitsy;
+
+    hibitsy = 0 <= y ? 0 : BDIGMAX;
+    xds = BDIGITS(x);
+    zn = BIGNUM_LEN(x);
+#if SIZEOF_BDIGIT < SIZEOF_LONG
+    if (zn < bdigit_roomof(SIZEOF_LONG))
+        zn = bdigit_roomof(SIZEOF_LONG);
+#endif
+    z = bignew(zn, 0);
+    zds = BDIGITS(z);
+
+#if SIZEOF_BDIGIT >= SIZEOF_LONG
+    i = 1;
+    zds[0] = xds[0] ^ BIGLO(y);
+#else
+    for (i = 0; i < xn; i++) {
+        zds[i] = xds[i] ^ BIGLO(y);
+        y = BIGDN(y);
+    }
+    for (; i < zn; i++) {
+        zds[i] = hibitsx ^ BIGLO(y);
+        y = BIGDN(y);
+    }
+#endif
+    for (; i < xn; i++) {
+        zds[i] = xds[i] ^ hibitsy;
+    }
+    for (; i < zn; i++) {
+        zds[i] = hibitsx ^ hibitsy;
+    }
+    twocomp2abs_bang(z, (hibitsx ^ hibitsy) != 0);
+    RB_GC_GUARD(x);
+    return bignorm(z);
+}
 
 VALUE
-rb_big_xor(xx, yy)
-    VALUE xx, yy;
+rb_big_xor(VALUE x, VALUE y)
 {
-    volatile VALUE x, y;
     VALUE z;
     BDIGIT *ds1, *ds2, *zds;
-    long i, l1, l2;
-    char sign;
+    long i, xn, yn, n1, n2;
+    BDIGIT hibitsx, hibitsy;
+    BDIGIT hibits1, hibits2;
+    VALUE tmpv;
+    BDIGIT tmph;
+    long tmpn;
+
+    if (!RB_INTEGER_TYPE_P(y)) {
+	return rb_num_coerce_bit(x, y, '^');
+    }
 
-    x = xx;
-    y = rb_to_int(yy);
+    hibitsx = abs2twocomp(&x, &xn);
     if (FIXNUM_P(y)) {
-	y = rb_int2big(FIX2LONG(y));
-    }
-    if (!RBIGNUM(y)->sign) {
-	y = rb_big_clone(y);
-	get2comp(y);
+	return bigxor_int(x, xn, hibitsx, FIX2LONG(y));
     }
-    if (!RBIGNUM(x)->sign) {
-	x = rb_big_clone(x);
-	get2comp(x);
-    }
-    if (RBIGNUM(x)->len > RBIGNUM(y)->len) {
-	l1 = RBIGNUM(y)->len;
-	l2 = RBIGNUM(x)->len;
-	ds1 = BDIGITS(y);
-	ds2 = BDIGITS(x);
-	sign = RBIGNUM(y)->sign;
-    }
-    else {
-	l1 = RBIGNUM(x)->len;
-	l2 = RBIGNUM(y)->len;
-	ds1 = BDIGITS(x);
-	ds2 = BDIGITS(y);
-	sign = RBIGNUM(x)->sign;
+    hibitsy = abs2twocomp(&y, &yn);
+    if (xn > yn) {
+        tmpv = x; x = y; y = tmpv;
+        tmpn = xn; xn = yn; yn = tmpn;
+        tmph = hibitsx; hibitsx = hibitsy; hibitsy = tmph;
     }
-    RBIGNUM(x)->sign = RBIGNUM(x)->sign?1:0;
-    RBIGNUM(y)->sign = RBIGNUM(y)->sign?1:0;
-    z = bignew(l2, !(RBIGNUM(x)->sign ^ RBIGNUM(y)->sign));
+    n1 = xn;
+    n2 = yn;
+    ds1 = BDIGITS(x);
+    ds2 = BDIGITS(y);
+    hibits1 = hibitsx;
+    hibits2 = hibitsy;
+
+    z = bignew(n2, 0);
     zds = BDIGITS(z);
 
-    for (i=0; i<l1; i++) {
+    for (i=0; i<n1; i++) {
 	zds[i] = ds1[i] ^ ds2[i];
     }
-    for (; i<l2; i++) {
-	zds[i] = sign?ds2[i]:~ds2[i];
+    for (; i<n2; i++) {
+	zds[i] = hibitsx ^ ds2[i];
     }
-    if (!RBIGNUM(z)->sign) get2comp(z);
-
+    twocomp2abs_bang(z, (hibits1 ^ hibits2) != 0);
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
     return bignorm(z);
 }
 
-static VALUE
-check_shiftdown(VALUE y, VALUE x)
+VALUE
+rb_big_lshift(VALUE x, VALUE y)
 {
-    if (!RBIGNUM(x)->len) return INT2FIX(0);
-    if (RBIGNUM(y)->len > SIZEOF_LONG / SIZEOF_BDIGITS) {
-	return RBIGNUM(x)->sign ? INT2FIX(0) : INT2FIX(-1);
+    int lshift_p;
+    size_t shift_numdigits;
+    int shift_numbits;
+
+    for (;;) {
+	if (FIXNUM_P(y)) {
+	    long l = FIX2LONG(y);
+            unsigned long shift;
+	    if (0 <= l) {
+		lshift_p = 1;
+                shift = l;
+            }
+            else {
+		lshift_p = 0;
+		shift = 1+(unsigned long)(-(l+1));
+	    }
+            shift_numbits = (int)(shift & (BITSPERDIG-1));
+            shift_numdigits = shift >> bit_length(BITSPERDIG-1);
+            return bignorm(big_shift3(x, lshift_p, shift_numdigits, shift_numbits));
+	}
+	else if (RB_BIGNUM_TYPE_P(y)) {
+            return bignorm(big_shift2(x, 1, y));
+	}
+	y = rb_to_int(y);
     }
-    return Qnil;
 }
 
-/*
- * call-seq:
- *     big << numeric   =>  integer
- *
- * Shifts big left _numeric_ positions (right if _numeric_ is negative).
- */
-
 VALUE
-rb_big_lshift(x, y)
-    VALUE x, y;
+rb_big_rshift(VALUE x, VALUE y)
 {
-    long shift;
-    int neg = 0;
+    int lshift_p;
+    size_t shift_numdigits;
+    int shift_numbits;
 
     for (;;) {
 	if (FIXNUM_P(y)) {
-	    shift = FIX2LONG(y);
-	    if (shift < 0) {
-		neg = 1;
-		shift = -shift;
+	    long l = FIX2LONG(y);
+            unsigned long shift;
+            if (0 <= l) {
+                lshift_p = 0;
+                shift = l;
+            }
+            else {
+                lshift_p = 1;
+		shift = 1+(unsigned long)(-(l+1));
 	    }
-	    break;
+            shift_numbits = (int)(shift & (BITSPERDIG-1));
+            shift_numdigits = shift >> bit_length(BITSPERDIG-1);
+            return bignorm(big_shift3(x, lshift_p, shift_numdigits, shift_numbits));
 	}
-	else if (TYPE(y) == T_BIGNUM) {
-	    if (!RBIGNUM(y)->sign) {
-		VALUE t = check_shiftdown(y, x);
-		if (!NIL_P(t)) return t;
-		neg = 1;
-	    }
-	    shift = big2ulong(y, "long", Qtrue);
-	    break;
+	else if (RB_BIGNUM_TYPE_P(y)) {
+            return bignorm(big_shift2(x, 0, y));
 	}
 	y = rb_to_int(y);
     }
-
-    if (neg) return big_rshift(x, shift);
-    return big_lshift(x, shift);
 }
 
-static VALUE
-big_lshift(x, shift)
-    VALUE x;
-    unsigned long shift;
+VALUE
+rb_big_aref(VALUE x, VALUE y)
 {
-    BDIGIT *xds, *zds;
-    long s1 = shift/BITSPERDIG;
-    int s2 = shift%BITSPERDIG;
-    VALUE z;
-    BDIGIT_DBL num = 0;
-    long len, i;
+    BDIGIT *xds;
+    size_t shift;
+    size_t i, s1, s2;
+    long l;
+    BDIGIT bit;
 
-    len = RBIGNUM(x)->len;
-    z = bignew(len+s1+1, RBIGNUM(x)->sign);
-    zds = BDIGITS(z);
-    for (i=0; i<s1; i++) {
-	*zds++ = 0;
+    if (RB_BIGNUM_TYPE_P(y)) {
+	if (BIGNUM_NEGATIVE_P(y))
+	    return INT2FIX(0);
+	bigtrunc(y);
+	if (BIGSIZE(y) > sizeof(size_t)) {
+	  out_of_range:
+	    return BIGNUM_SIGN(x) ? INT2FIX(0) : INT2FIX(1);
+	}
+#if SIZEOF_SIZE_T <= SIZEOF_LONG
+	shift = big2ulong(y, "long");
+#else
+	shift = big2ull(y, "long long");
+#endif
     }
-    xds = BDIGITS(x);
-    for (i=0; i<len; i++) {
-	num = num | (BDIGIT_DBL)*xds++<<s2;
-	*zds++ = BIGLO(num);
-	num = BIGDN(num);
+    else {
+	l = NUM2LONG(y);
+	if (l < 0) return INT2FIX(0);
+	shift = (size_t)l;
     }
-    *zds = BIGLO(num);
-    return bignorm(z);
+    s1 = shift/BITSPERDIG;
+    s2 = shift%BITSPERDIG;
+    bit = (BDIGIT)1 << s2;
+
+    if (s1 >= BIGNUM_LEN(x)) goto out_of_range;
+
+    xds = BDIGITS(x);
+    if (BIGNUM_POSITIVE_P(x))
+        return (xds[s1] & bit) ? INT2FIX(1) : INT2FIX(0);
+    if (xds[s1] & (bit-1))
+        return (xds[s1] & bit) ? INT2FIX(0) : INT2FIX(1);
+    for (i = 0; i < s1; i++)
+        if (xds[i])
+            return (xds[s1] & bit) ? INT2FIX(0) : INT2FIX(1);
+    return (xds[s1] & bit) ? INT2FIX(1) : INT2FIX(0);
+}
+
+VALUE
+rb_big_hash(VALUE x)
+{
+    st_index_t hash;
+
+    hash = rb_memhash(BDIGITS(x), sizeof(BDIGIT)*BIGNUM_LEN(x)) ^ BIGNUM_SIGN(x);
+    return ST2FIX(hash);
 }
 
 /*
  * call-seq:
- *     big >> numeric   =>  integer
+ *   big.coerce(numeric)  ->  array
+ *
+ * Returns an array with both a +numeric+ and a +big+ represented as Bignum
+ * objects.
+ *
+ * This is achieved by converting +numeric+ to a Bignum.
+ *
+ * A TypeError is raised if the +numeric+ is not a Fixnum or Bignum type.
  *
- * Shifts big right _numeric_ positions (left if _numeric_ is negative).
+ *     (0x3FFFFFFFFFFFFFFF+1).coerce(42)   #=> [42, 4611686018427387904]
  */
 
-VALUE
-rb_big_rshift(x, y)
-    VALUE x, y;
+static VALUE
+rb_int_coerce(VALUE x, VALUE y)
 {
-    long shift;
-    int neg = 0;
+    if (RB_INTEGER_TYPE_P(y)) {
+        return rb_assoc_new(y, x);
+    }
+    else {
+        x = rb_Float(x);
+        y = rb_Float(y);
+        return rb_assoc_new(y, x);
+    }
+}
 
-    for (;;) {
-	if (FIXNUM_P(y)) {
-	    shift = FIX2LONG(y);
-	    if (shift < 0) {
-		neg = 1;
-		shift = -shift;
-	    }
-	    break;
-	}
-	else if (TYPE(y) == T_BIGNUM) {
-	    if (RBIGNUM(y)->sign) {
-		VALUE t = check_shiftdown(y, x);
-		if (!NIL_P(t)) return t;
-	    }
-	    else {
-		neg = 1;
-	    }
-	    shift = big2ulong(y, "long", Qtrue);
-	    break;
-	}
-	y = rb_to_int(y);
+VALUE
+rb_big_abs(VALUE x)
+{
+    if (BIGNUM_NEGATIVE_P(x)) {
+	x = rb_big_clone(x);
+	BIGNUM_SET_POSITIVE_SIGN(x);
     }
+    return x;
+}
 
-    if (neg) return big_lshift(x, shift);
-    return big_rshift(x, shift);
+int
+rb_big_sign(VALUE x)
+{
+    return BIGNUM_SIGN(x);
 }
 
-static VALUE
-big_rshift(x, shift)
-    VALUE x;
-    unsigned long shift;
+size_t
+rb_big_size(VALUE big)
 {
-    BDIGIT *xds, *zds;
-    long s1 = shift/BITSPERDIG;
-    int s2 = shift%BITSPERDIG;
-    VALUE z;
-    BDIGIT_DBL num = 0;
-    long i, j;
-    volatile VALUE save_x;
+    return BIGSIZE(big);
+}
 
-    if (s1 > RBIGNUM(x)->len) {
-	if (RBIGNUM(x)->sign)
-	    return INT2FIX(0);
-	else
-	    return INT2FIX(-1);
-    }
-    if (!RBIGNUM(x)->sign) {
-	save_x = x = rb_big_clone(x);
-	get2comp(x);
+VALUE
+rb_big_size_m(VALUE big)
+{
+    return SIZET2NUM(rb_big_size(big));
+}
+
+VALUE
+rb_big_bit_length(VALUE big)
+{
+    int nlz_bits;
+    size_t numbytes;
+
+    static const BDIGIT char_bit[1] = { CHAR_BIT };
+    BDIGIT numbytes_bary[bdigit_roomof(sizeof(size_t))];
+    BDIGIT nlz_bary[1];
+    BDIGIT result_bary[bdigit_roomof(sizeof(size_t)+1)];
+
+    numbytes = rb_absint_size(big, &nlz_bits);
+
+    if (numbytes == 0)
+        return LONG2FIX(0);
+
+    if (BIGNUM_NEGATIVE_P(big) && rb_absint_singlebit_p(big)) {
+        if (nlz_bits != CHAR_BIT-1) {
+            nlz_bits++;
+        }
+        else {
+            nlz_bits = 0;
+            numbytes--;
+        }
     }
-    xds = BDIGITS(x);
-    i = RBIGNUM(x)->len; j = i - s1;
-    z = bignew(j, RBIGNUM(x)->sign);
-    if (!RBIGNUM(x)->sign) {
-	num = ((BDIGIT_DBL)~0) << BITSPERDIG;
+
+    if (numbytes <= SIZE_MAX / CHAR_BIT) {
+        return SIZET2NUM(numbytes * CHAR_BIT - nlz_bits);
     }
-    zds = BDIGITS(z);
-    while (i--, j--) {
-	num = (num | xds[i]) >> s2;
-	zds[j] = BIGLO(num);
-	num = BIGUP(xds[i]);
+
+    nlz_bary[0] = nlz_bits;
+
+    bary_unpack(BARY_ARGS(numbytes_bary), &numbytes, 1, sizeof(numbytes), 0,
+            INTEGER_PACK_NATIVE_BYTE_ORDER);
+    BARY_SHORT_MUL(result_bary, numbytes_bary, char_bit);
+    BARY_SUB(result_bary, result_bary, nlz_bary);
+
+    return rb_integer_unpack(result_bary, numberof(result_bary), sizeof(BDIGIT), 0,
+            INTEGER_PACK_LSWORD_FIRST|INTEGER_PACK_NATIVE_BYTE_ORDER);
+}
+
+VALUE
+rb_big_odd_p(VALUE num)
+{
+    if (BIGNUM_LEN(num) != 0 && BDIGITS(num)[0] & 1) {
+	return Qtrue;
     }
-    if (!RBIGNUM(x)->sign) {
-	get2comp(z);
+    return Qfalse;
+}
+
+VALUE
+rb_big_even_p(VALUE num)
+{
+    if (BIGNUM_LEN(num) != 0 && BDIGITS(num)[0] & 1) {
+	return Qfalse;
     }
-    return bignorm(z);
+    return Qtrue;
 }
 
-/*
- *  call-seq:
- *     big[n] -> 0, 1
- *  
- *  Bit Reference---Returns the <em>n</em>th bit in the (assumed) binary
- *  representation of <i>big</i>, where <i>big</i>[0] is the least
- *  significant bit.
- *     
- *     a = 9**15
- *     50.downto(0) do |n|
- *       print a[n]
- *     end
- *     
- *  <em>produces:</em>
- *     
- *     000101110110100000111000011110010100111100010111001
- *     
- */
+unsigned long rb_ulong_isqrt(unsigned long);
+#if SIZEOF_BDIGIT*2 > SIZEOF_LONG
+BDIGIT rb_bdigit_dbl_isqrt(BDIGIT_DBL);
+# ifdef ULL_TO_DOUBLE
+#   define BDIGIT_DBL_TO_DOUBLE(n) ULL_TO_DOUBLE(n)
+# endif
+#else
+# define rb_bdigit_dbl_isqrt(x) (BDIGIT)rb_ulong_isqrt(x)
+#endif
+#ifndef BDIGIT_DBL_TO_DOUBLE
+# define BDIGIT_DBL_TO_DOUBLE(n) (double)(n)
+#endif
 
-static VALUE
-rb_big_aref(x, y)
-    VALUE x, y;
+static BDIGIT *
+estimate_initial_sqrt(VALUE *xp, const size_t xn, const BDIGIT *nds, size_t len)
 {
-    BDIGIT *xds;
-    int shift;
-    long s1, s2;
-
-    if (TYPE(y) == T_BIGNUM) {
-	if (!RBIGNUM(y)->sign || RBIGNUM(x)->sign)
-	    return INT2FIX(0);
-	return INT2FIX(1);
+    enum {dbl_per_bdig = roomof(DBL_MANT_DIG,BITSPERDIG)};
+    const int zbits = nlz(nds[len-1]);
+    VALUE x = *xp = bignew_1(0, xn, 1); /* division may release the GVL */
+    BDIGIT *xds = BDIGITS(x);
+    BDIGIT_DBL d = bary2bdigitdbl(nds+len-dbl_per_bdig, dbl_per_bdig);
+    BDIGIT lowbits = 1;
+    int rshift = (int)((BITSPERDIG*2-zbits+(len&BITSPERDIG&1) - DBL_MANT_DIG + 1) & ~1);
+    double f;
+
+    if (rshift > 0) {
+	lowbits = (BDIGIT)d & ~(~(BDIGIT)1U << rshift);
+	d >>= rshift;
     }
-    shift = NUM2INT(y);
-    if (shift < 0) return INT2FIX(0);
-    s1 = shift/BITSPERDIG;
-    s2 = shift%BITSPERDIG;
-
-    if (!RBIGNUM(x)->sign) {
-	if (s1 >= RBIGNUM(x)->len) return INT2FIX(1);
-	x = rb_big_clone(x);
-	get2comp(x);
+    else if (rshift < 0) {
+	d <<= -rshift;
+	d |= nds[len-dbl_per_bdig-1] >> (BITSPERDIG+rshift);
+    }
+    f = sqrt(BDIGIT_DBL_TO_DOUBLE(d));
+    d = (BDIGIT_DBL)ceil(f);
+    if (BDIGIT_DBL_TO_DOUBLE(d) == f) {
+	if (lowbits || (lowbits = !bary_zero_p(nds, len-dbl_per_bdig)))
+	    ++d;
     }
     else {
-	if (s1 >= RBIGNUM(x)->len) return INT2FIX(0);
+	lowbits = 1;
     }
-    xds = BDIGITS(x);
-    if (xds[s1] & (1<<s2))
-	return INT2FIX(1);
-    return INT2FIX(0);
-}
+    rshift /= 2;
+    rshift += (2-(len&1))*BITSPERDIG/2;
+    if (rshift >= 0) {
+	d <<= rshift;
+    }
+    BDIGITS_ZERO(xds, xn-2);
+    bdigitdbl2bary(&xds[xn-2], 2, d);
 
-/*
- * call-seq:
- *   big.hash   => fixnum
- *
- * Compute a hash based on the value of _big_.
- */
+    if (!lowbits) return NULL; /* special case, exact result */
+    return xds;
+}
 
-static VALUE
-rb_big_hash(x)
-    VALUE x;
+VALUE
+rb_big_isqrt(VALUE n)
 {
-    long i, len, key;
-    BDIGIT *digits;
+    BDIGIT *nds = BDIGITS(n);
+    size_t len = BIGNUM_LEN(n);
+    size_t xn = (len+1) / 2;
+    VALUE x;
+    BDIGIT *xds;
 
-    key = 0; digits = BDIGITS(x); len = RBIGNUM(x)->len;
-    for (i=0; i<len; i++) {
-	key ^= *digits++;
+    if (len <= 2) {
+	BDIGIT sq = rb_bdigit_dbl_isqrt(bary2bdigitdbl(nds, len));
+#if SIZEOF_BDIGIT > SIZEOF_LONG
+	return ULL2NUM(sq);
+#else
+	return ULONG2NUM(sq);
+#endif
+    }
+    else if ((xds = estimate_initial_sqrt(&x, xn, nds, len)) != 0) {
+	size_t tn = xn + BIGDIVREM_EXTRA_WORDS;
+	VALUE t = bignew_1(0, tn, 1);
+	BDIGIT *tds = BDIGITS(t);
+	tn = BIGNUM_LEN(t);
+
+	/* t = n/x */
+	while (bary_divmod_branch(tds, tn, NULL, 0, nds, len, xds, xn),
+	       bary_cmp(tds, tn, xds, xn) < 0) {
+	    int carry;
+	    BARY_TRUNC(tds, tn);
+	    /* x = (x+t)/2 */
+	    carry = bary_add(xds, xn, xds, xn, tds, tn);
+	    bary_small_rshift(xds, xds, xn, 1, carry);
+	    tn = BIGNUM_LEN(t);
+	}
+	rb_big_realloc(t, 0);
+	rb_gc_force_recycle(t);
     }
-    return LONG2FIX(key);
+    RBASIC_SET_CLASS_RAW(x, rb_cInteger);
+    return x;
 }
 
-/*
- * MISSING: documentation
- */
+#ifdef USE_GMP
+static void
+bary_powm_gmp(BDIGIT *zds, size_t zn, const BDIGIT *xds, size_t xn, const BDIGIT *yds, size_t yn, const BDIGIT *mds, size_t mn)
+{
+    const size_t nails = (sizeof(BDIGIT)-SIZEOF_BDIGIT)*CHAR_BIT;
+    mpz_t z, x, y, m;
+    size_t count;
+    mpz_init(x);
+    mpz_init(y);
+    mpz_init(m);
+    mpz_init(z);
+    mpz_import(x, xn, -1, sizeof(BDIGIT), 0, nails, xds);
+    mpz_import(y, yn, -1, sizeof(BDIGIT), 0, nails, yds);
+    mpz_import(m, mn, -1, sizeof(BDIGIT), 0, nails, mds);
+    mpz_powm(z, x, y, m);
+    mpz_export(zds, &count, -1, sizeof(BDIGIT), 0, nails, z);
+    BDIGITS_ZERO(zds+count, zn-count);
+    mpz_clear(x);
+    mpz_clear(y);
+    mpz_clear(m);
+    mpz_clear(z);
+}
+#endif
 
 static VALUE
-rb_big_coerce(x, y)
-    VALUE x, y;
+int_pow_tmp3(VALUE x, VALUE y, VALUE m, int nega_flg)
 {
+#ifdef USE_GMP
+    VALUE z;
+    size_t xn, yn, mn, zn;
+
+    if (FIXNUM_P(x)) {
+       x = rb_int2big(FIX2LONG(x));
+    }
     if (FIXNUM_P(y)) {
-	return rb_assoc_new(rb_int2big(FIX2LONG(y)), x);
+       y = rb_int2big(FIX2LONG(y));
     }
-    else if (TYPE(y) == T_BIGNUM) {
-       return rb_assoc_new(y, x);
+    assert(RB_BIGNUM_TYPE_P(m));
+    xn = BIGNUM_LEN(x);
+    yn = BIGNUM_LEN(y);
+    mn = BIGNUM_LEN(m);
+    zn = mn;
+    z = bignew(zn, 1);
+    bary_powm_gmp(BDIGITS(z), zn, BDIGITS(x), xn, BDIGITS(y), yn, BDIGITS(m), mn);
+    if (nega_flg & BIGNUM_POSITIVE_P(z)) {
+        z = rb_big_minus(z, m);
     }
-    else {
-	rb_raise(rb_eTypeError, "can't coerce %s to Bignum",
-		 rb_obj_classname(y));
+    RB_GC_GUARD(x);
+    RB_GC_GUARD(y);
+    RB_GC_GUARD(m);
+    return rb_big_norm(z);
+#else
+    VALUE tmp = LONG2FIX(1L);
+    long yy;
+
+    for (/*NOP*/; ! FIXNUM_P(y); y = rb_big_rshift(y, LONG2FIX(1L))) {
+        if (RTEST(rb_int_odd_p(y))) {
+            tmp = rb_int_mul(tmp, x);
+            tmp = rb_int_modulo(tmp, m);
+        }
+        x = rb_int_mul(x, x);
+        x = rb_int_modulo(x, m);
     }
-    /* not reached */
-    return Qnil;
+    for (yy = FIX2LONG(y); yy; yy >>= 1L) {
+        if (yy & 1L) {
+            tmp = rb_int_mul(tmp, x);
+            tmp = rb_int_modulo(tmp, m);
+        }
+        x = rb_int_mul(x, x);
+        x = rb_int_modulo(x, m);
+    }
+
+    if (nega_flg && rb_int_positive_p(tmp)) {
+        tmp = rb_int_minus(tmp, m);
+    }
+    return tmp;
+#endif
 }
 
 /*
- *  call-seq:
- *     big.abs -> aBignum
- *  
- *  Returns the absolute value of <i>big</i>.
- *     
- *     -1234567890987654321.abs   #=> 1234567890987654321
+ * Integer#pow
  */
 
 static VALUE
-rb_big_abs(x)
-    VALUE x;
+int_pow_tmp1(VALUE x, VALUE y, long mm, int nega_flg)
 {
-    if (!RBIGNUM(x)->sign) {
-	x = rb_big_clone(x);
-	RBIGNUM(x)->sign = 1;
+    long xx = FIX2LONG(x);
+    long tmp = 1L;
+    long yy;
+
+    for (/*NOP*/; ! FIXNUM_P(y); y = rb_big_rshift(y, LONG2FIX(1L))) {
+        if (RTEST(rb_int_odd_p(y))) {
+            tmp = (tmp * xx) % mm;
+        }
+        xx = (xx * xx) % mm;
     }
-    return x;
+    for (yy = FIX2LONG(y); yy; yy >>= 1L) {
+        if (yy & 1L) {
+            tmp = (tmp * xx) % mm;
+        }
+        xx = (xx * xx) % mm;
+    }
+
+    if (nega_flg && tmp) {
+        tmp -= mm;
+    }
+    return LONG2FIX(tmp);
 }
 
-VALUE
-rb_big_rand(max, rand_buf)
-    VALUE max;
-    double *rand_buf;
+static VALUE
+int_pow_tmp2(VALUE x, VALUE y, long mm, int nega_flg)
 {
-    VALUE v;
-    long len = RBIGNUM(max)->len;
+    long tmp = 1L;
+    long yy;
+#ifdef DLONG
+    const DLONG m = mm;
+    long tmp2 = tmp;
+    long xx = FIX2LONG(x);
+# define MUL_MODULO(a, b, c) (long)(((DLONG)(a) * (DLONG)(b)) % (c))
+#else
+    const VALUE m = LONG2FIX(mm);
+    VALUE tmp2 = LONG2FIX(tmp);
+    VALUE xx = x;
+# define MUL_MODULO(a, b, c) rb_int_modulo(rb_fix_mul_fix((a), (b)), (c))
+#endif
 
-    if (BIGZEROP(max)) {
-	return rb_float_new(rand_buf[0]);
+    for (/*NOP*/; ! FIXNUM_P(y); y = rb_big_rshift(y, LONG2FIX(1L))) {
+        if (RTEST(rb_int_odd_p(y))) {
+            tmp2 = MUL_MODULO(tmp2, xx, m);
+        }
+        xx = MUL_MODULO(xx, xx, m);
     }
-    v = bignew(len,1);
-    len--;
-    BDIGITS(v)[len] = BDIGITS(max)[len] * rand_buf[len];    
-    while (len--) {
-	BDIGITS(v)[len] = ((BDIGIT)~0) * rand_buf[len];
+    for (yy = FIX2LONG(y); yy; yy >>= 1L) {
+        if (yy & 1L) {
+            tmp2 = MUL_MODULO(tmp2, xx, m);
+        }
+        xx = MUL_MODULO(xx, xx, m);
     }
 
-    return v;
+#ifdef DLONG
+    tmp = tmp2;
+#else
+    tmp = FIX2LONG(tmp2);
+#endif
+    if (nega_flg && tmp) {
+        tmp -= mm;
+    }
+    return LONG2FIX(tmp);
 }
 
 /*
- *  call-seq:
- *     big.size -> integer
- *  
- *  Returns the number of bytes in the machine representation of
- *  <i>big</i>.
- *     
- *     (256**10 - 1).size   #=> 12
- *     (256**20 - 1).size   #=> 20
- *     (256**40 - 1).size   #=> 40
+ * Document-method: Integer#pow
+ * call-seq:
+ *    integer.pow(numeric)           ->  numeric
+ *    integer.pow(integer, integer)  ->  integer
+ *
+ * Returns (modular) exponentiation as:
+ *
+ *   a.pow(b)     #=> same as a**b
+ *   a.pow(b, m)  #=> same as (a**b) % m, but avoids huge temporary values
  */
-
-static VALUE
-rb_big_size(big)
-    VALUE big;
+VALUE
+rb_int_powm(int const argc, VALUE * const argv, VALUE const num)
 {
-    return LONG2FIX(RBIGNUM(big)->len*SIZEOF_BDIGITS);
+    rb_check_arity(argc, 1, 2);
+
+    if (argc == 1) {
+        return rb_int_pow(num, argv[0]);
+    }
+    else {
+        VALUE const a = num;
+        VALUE const b = argv[0];
+        VALUE m = argv[1];
+        int nega_flg = 0;
+        if ( ! RB_INTEGER_TYPE_P(b)) {
+            rb_raise(rb_eTypeError, "Integer#pow() 2nd argument not allowed unless a 1st argument is integer");
+        }
+        if (rb_int_negative_p(b)) {
+            rb_raise(rb_eRangeError, "Integer#pow() 1st argument cannot be negative when 2nd argument specified");
+        }
+        if (!RB_INTEGER_TYPE_P(m)) {
+            rb_raise(rb_eTypeError, "Integer#pow() 2nd argument not allowed unless all arguments are integers");
+        }
+
+        if (rb_int_negative_p(m)) {
+            m = rb_int_uminus(m);
+            nega_flg = 1;
+        }
+
+        if (FIXNUM_P(m)) {
+            long const half_val = (long)HALF_LONG_MSB;
+            long const mm = FIX2LONG(m);
+            if (!mm) rb_num_zerodiv();
+            if (mm <= half_val) {
+                return int_pow_tmp1(rb_int_modulo(a, m), b, mm, nega_flg);
+            }
+            else {
+                return int_pow_tmp2(rb_int_modulo(a, m), b, mm, nega_flg);
+            }
+        }
+        else {
+            if (rb_bigzero_p(m)) rb_num_zerodiv();
+            return int_pow_tmp3(rb_int_modulo(a, m), b, m, nega_flg);
+        }
+    }
+    UNREACHABLE_RETURN(Qnil);
 }
 
 /*
@@ -2146,49 +7139,33 @@ rb_big_size(big)
  *  Fixnum. When a calculation involving
  *  Bignum objects returns a result that will fit in a
  *  Fixnum, the result is automatically converted.
- *     
+ *
  *  For the purposes of the bitwise operations and <code>[]</code>, a
  *  Bignum is treated as if it were an infinite-length
  *  bitstring with 2's complement representation.
- *     
+ *
  *  While Fixnum values are immediate, Bignum
  *  objects are not---assignment and parameter passing work with
  *  references to objects, not the objects themselves.
- *     
+ *
  */
 
 void
-Init_Bignum()
-{
-    rb_cBignum = rb_define_class("Bignum", rb_cInteger);
-
-    rb_define_method(rb_cBignum, "to_s", rb_big_to_s, -1);
-    rb_define_method(rb_cBignum, "coerce", rb_big_coerce, 1);
-    rb_define_method(rb_cBignum, "-@", rb_big_uminus, 0);
-    rb_define_method(rb_cBignum, "+", rb_big_plus, 1);
-    rb_define_method(rb_cBignum, "-", rb_big_minus, 1);
-    rb_define_method(rb_cBignum, "*", rb_big_mul, 1);
-    rb_define_method(rb_cBignum, "/", rb_big_div, 1);
-    rb_define_method(rb_cBignum, "%", rb_big_modulo, 1);
-    rb_define_method(rb_cBignum, "div", rb_big_div, 1);
-    rb_define_method(rb_cBignum, "divmod", rb_big_divmod, 1);
-    rb_define_method(rb_cBignum, "modulo", rb_big_modulo, 1);
-    rb_define_method(rb_cBignum, "remainder", rb_big_remainder, 1);
-    rb_define_method(rb_cBignum, "quo", rb_big_quo, 1);
-    rb_define_method(rb_cBignum, "**", rb_big_pow, 1);
-    rb_define_method(rb_cBignum, "&", rb_big_and, 1);
-    rb_define_method(rb_cBignum, "|", rb_big_or, 1);
-    rb_define_method(rb_cBignum, "^", rb_big_xor, 1);
-    rb_define_method(rb_cBignum, "~", rb_big_neg, 0);
-    rb_define_method(rb_cBignum, "<<", rb_big_lshift, 1);
-    rb_define_method(rb_cBignum, ">>", rb_big_rshift, 1);
-    rb_define_method(rb_cBignum, "[]", rb_big_aref, 1);
-
-    rb_define_method(rb_cBignum, "<=>", rb_big_cmp, 1);
-    rb_define_method(rb_cBignum, "==", rb_big_eq, 1);
-    rb_define_method(rb_cBignum, "eql?", rb_big_eql, 1);
-    rb_define_method(rb_cBignum, "hash", rb_big_hash, 0);
-    rb_define_method(rb_cBignum, "to_f", rb_big_to_f, 0);
-    rb_define_method(rb_cBignum, "abs", rb_big_abs, 0);
-    rb_define_method(rb_cBignum, "size", rb_big_size, 0);
+Init_Bignum(void)
+{
+#ifndef RUBY_INTEGER_UNIFICATION
+    rb_cBignum = rb_cInteger;
+#endif
+    /* An obsolete class, use Integer */
+    rb_define_const(rb_cObject, "Bignum", rb_cInteger);
+    rb_deprecate_constant(rb_cObject, "Bignum");
+
+    rb_define_method(rb_cInteger, "coerce", rb_int_coerce, 1);
+
+#ifdef USE_GMP
+    /* The version of loaded GMP. */
+    rb_define_const(rb_cInteger, "GMP_VERSION", rb_sprintf("GMP %s", gmp_version));
+#endif
+
+    power_cache_init();
 }
diff --git a/bin/bundle b/bin/bundle
new file mode 100755
index 0000000000..1a0b06b005
--- /dev/null
+++ b/bin/bundle
@@ -0,0 +1,27 @@
+#!/usr/bin/env ruby
+#
+# This file was generated by RubyGems.
+#
+# The application 'bundler' is installed as part of a gem, and
+# this file is here to facilitate running it.
+#
+
+require 'rubygems'
+
+version = ">= 0.a"
+
+str = ARGV.first
+if str
+  str = str.b[/\A_(.*)_\z/, 1]
+  if str and Gem::Version.correct?(str)
+    version = str
+    ARGV.shift
+  end
+end
+
+if Gem.respond_to?(:activate_bin_path)
+load Gem.activate_bin_path('bundler', 'bundle', version)
+else
+gem "bundler", version
+load Gem.bin_path("bundler", "bundle", version)
+end
diff --git a/bin/bundler b/bin/bundler
new file mode 100755
index 0000000000..e15eb39ed7
--- /dev/null
+++ b/bin/bundler
@@ -0,0 +1,27 @@
+#!/usr/bin/env ruby
+#
+# This file was generated by RubyGems.
+#
+# The application 'bundler' is installed as part of a gem, and
+# this file is here to facilitate running it.
+#
+
+require 'rubygems'
+
+version = ">= 0.a"
+
+str = ARGV.first
+if str
+  str = str.b[/\A_(.*)_\z/, 1]
+  if str and Gem::Version.correct?(str)
+    version = str
+    ARGV.shift
+  end
+end
+
+if Gem.respond_to?(:activate_bin_path)
+load Gem.activate_bin_path('bundler', 'bundler', version)
+else
+gem "bundler", version
+load Gem.bin_path("bundler", "bundler", version)
+end
diff --git a/bin/erb b/bin/erb
index a6fcd5370c..d060caba90 100755
--- a/bin/erb
+++ b/bin/erb
@@ -1,6 +1,6 @@
 #!/usr/bin/env ruby
 # Tiny eRuby --- ERB2
-# Copyright (c) 1999-2000,2002 Masatoshi SEKI 
+# Copyright (c) 1999-2000,2002 Masatoshi SEKI
 # You can redistribute it and/or modify it under the same terms as Ruby.
 
 require 'erb'
@@ -11,19 +11,32 @@ class ERB
       return nil if self.empty?
       arg = self.shift
       return nil if arg == '--'
-      if arg =~ /^-(.)(.*)/
-        return arg if $1 == '-'
-        raise 'unknown switch "-"' if $2.index('-')
-        self.unshift "-#{$2}" if $2.size > 0
+      case arg
+      when /\A-(.)(.*)/
+        if $1 == '-'
+          arg, @maybe_arg = arg.split(/=/, 2)
+          return arg
+        end
+        raise 'unknown switch "-"' if $2[0] == ?- and $1 != 'T'
+        if $2.size > 0
+          self.unshift "-#{$2}"
+          @maybe_arg = $2
+        else
+          @maybe_arg = nil
+        end
         "-#{$1}"
+      when /\A(\w+)=/
+        arg
       else
         self.unshift arg
         nil
       end
     end
-    
+
     def ARGV.req_arg
-      self.shift || raise('missing argument')
+      (@maybe_arg || self.shift || raise('missing argument')).tap {
+        @maybe_arg = nil
+      }
     end
 
     def trim_mode_opt(trim_mode, disable_percent)
@@ -44,6 +57,7 @@ class ERB
     def run(factory=ERB)
       trim_mode = 0
       disable_percent = false
+      variables = {}
       begin
         while switch = ARGV.switch
           case switch
@@ -61,78 +75,99 @@ class ERB
           when '-r'                        # require
             require ARGV.req_arg
           when '-S'                        # security level
+            warn 'warning: -S option of erb command is deprecated. Please do not use this.'
             arg = ARGV.req_arg
-            raise "invalid safe_level #{arg.dump}" unless arg =~ /^[0-4]$/
+            raise "invalid safe_level #{arg.dump}" unless arg =~ /\A[0-1]\z/
             safe_level = arg.to_i
           when '-T'                        # trim mode
             arg = ARGV.req_arg
             if arg == '-'
-              trim_mode = arg 
+              trim_mode = arg
               next
             end
-            raise "invalid trim mode #{arg.dump}" unless arg =~ /^[0-2]$/
+            raise "invalid trim mode #{arg.dump}" unless arg =~ /\A[0-2]\z/
             trim_mode = arg.to_i
-          when '-K'                        # KCODE
+          when '-E', '--encoding'
             arg = ARGV.req_arg
-            case arg.downcase
-            when 'e', '-e', 'euc'
-              $KCODE = 'EUC'
-            when 's', '-s', 'sjis'
-              $KCODE = 'SJIS'
-            when 'u', '-u', 'utf8'
-              $KCODE = 'UTF8'
-            when 'n', '-n', 'none'
-              $KCODE = 'NONE'
-            else
-              raise "invalid KCODE #{arg.dump}"
-            end
+            set_encoding(*arg.split(/:/, 2))
+          when '-U'
+            set_encoding(Encoding::UTF_8, Encoding::UTF_8)
           when '-P'
             disable_percent = true
           when '--help'
             raise "print this help"
-          else
+          when /\A-/
             raise "unknown switch #{switch.dump}"
+          else
+            var, val = *switch.split('=', 2)
+            (variables ||= {})[var] = val
           end
         end
       rescue                               # usage
         STDERR.puts $!.to_s
-        STDERR.puts File.basename($0) + 
-          " [switches] [inputfile]"
+        STDERR.puts File.basename($0) +
+          " [switches] [var=value...] [inputfile]"
         STDERR.puts <<EOU
   -x               print ruby script
   -n               print ruby script with line number
   -v               enable verbose mode
   -d               set $DEBUG to true
-  -r [library]     load a library
-  -K [kcode]       specify KANJI code-set
-  -S [safe_level]  set $SAFE (0..4)
-  -T [trim_mode]   specify trim_mode (0..2, -)
+  -r library       load a library
+  -E ex[:in]       set default external/internal encodings
+  -U               set default encoding to UTF-8.
+  -T trim_mode     specify trim_mode (0..2, -)
   -P               ignore lines which start with "%"
+  var=value        set variable
 EOU
         exit 1
       end
 
+      $<.set_encoding(Encoding::UTF_8, nil)
       src = $<.read
       filename = $FILENAME
       exit 2 unless src
       trim = trim_mode_opt(trim_mode, disable_percent)
-      erb = factory.new(src.untaint, safe_level, trim)
+      if safe_level.nil?
+        erb = factory.new(src.untaint, trim_mode: trim)
+      else
+        # [deprecated] This will be removed at Ruby 2.7.
+        erb = factory.new(src.untaint, safe_level, trim_mode: trim)
+      end
       erb.filename = filename
       if output
         if number
-          l = 1
-          for line in erb.src
-            puts "%3d %s"%[l, line]
-            l += 1
+          erb.src.each_line.with_index do |line, l|
+            puts "%3d %s"%[l+1, line]
           end
         else
           puts erb.src
         end
       else
-        erb.run(TOPLEVEL_BINDING.taint)
+        bind = TOPLEVEL_BINDING.taint
+        if variables
+          enc = erb.encoding
+          for var, val in variables do
+            val = val.encode(enc) if val
+            bind.local_variable_set(var, val)
+          end
+        end
+        erb.run(bind)
       end
     end
     module_function :run
+
+    def set_encoding(extern, intern = nil)
+      verbose, $VERBOSE = $VERBOSE, nil
+      Encoding.default_external = extern unless extern.nil? || extern == ""
+      Encoding.default_internal = intern unless intern.nil? || intern == ""
+      [$stdin, $stdout, $stderr].each do |io|
+        io.set_encoding(extern, intern)
+      end
+    ensure
+      $VERBOSE = verbose
+    end
+    module_function :set_encoding
+    class << self; private :set_encoding; end
   end
 end
 
diff --git a/bin/gem b/bin/gem
new file mode 100755
index 0000000000..a4ec754abb
--- /dev/null
+++ b/bin/gem
@@ -0,0 +1,25 @@
+#!/usr/bin/env ruby
+#--
+# Copyright 2006 by Chad Fowler, Rich Kilmer, Jim Weirich and others.
+# All rights reserved.
+# See LICENSE.txt for permissions.
+#++
+
+require 'rubygems'
+require 'rubygems/gem_runner'
+require 'rubygems/exceptions'
+
+required_version = Gem::Requirement.new ">= 1.8.7"
+
+unless required_version.satisfied_by? Gem.ruby_version then
+  abort "Expected Ruby Version #{required_version}, is #{Gem.ruby_version}"
+end
+
+args = ARGV.clone
+
+begin
+  Gem::GemRunner.new.run args
+rescue Gem::SystemExitException => e
+  exit e.exit_code
+end
+
diff --git a/bin/irb b/bin/irb
index 13474a932b..ae6d358c9d 100644..100755
--- a/bin/irb
+++ b/bin/irb
@@ -1,21 +1,27 @@
 #!/usr/bin/env ruby
 #
-#   irb.rb - intaractive ruby
-#   	$Release Version: 0.9.5 $
-#   	$Revision: 1.2.2.1 $
-#   	$Date: 2005/04/19 19:24:56 $
-#   	by Keiju ISHITSUKA(keiju@ruby-lang.org)
+# This file was generated by RubyGems.
+#
+# The application 'irb' is installed as part of a gem, and
+# this file is here to facilitate running it.
 #
 
-require "irb"
+require 'rubygems'
 
-if __FILE__ == $0
-  IRB.start(__FILE__)
-else
-  # check -e option
-  if /^-e$/ =~ $0
-    IRB.start(__FILE__)
-  else
-    IRB.setup(__FILE__)
+version = ">= 0.a"
+
+str = ARGV.first
+if str
+  str = str.b[/\A_(.*)_\z/, 1]
+  if str and Gem::Version.correct?(str)
+    version = str
+    ARGV.shift
   end
 end
+
+if Gem.respond_to?(:activate_bin_path)
+load Gem.activate_bin_path('irb', 'irb', version)
+else
+gem "irb", version
+load Gem.bin_path("irb", "irb", version)
+end
diff --git a/bin/rdoc b/bin/rdoc
index d08ce838f6..8fa948cddb 100644..100755
--- a/bin/rdoc
+++ b/bin/rdoc
@@ -1,67 +1,27 @@
 #!/usr/bin/env ruby
 #
-#  RDoc: Documentation tool for source code
-#        (see lib/rdoc/rdoc.rb for more information)
+# This file was generated by RubyGems.
 #
-#  Copyright (c) 2003 Dave Thomas
-#  Released under the same terms as Ruby
+# The application 'rdoc' is installed as part of a gem, and
+# this file is here to facilitate running it.
 #
-#  $Revision: 1.1 $
 
-## Transitional Hack ####
-#
-#  RDoc was initially distributed independently, and installed
-#  itself into <prefix>/lib/ruby/site_ruby/<ver>/rdoc...
-#
-#  Now that RDoc is part of the distribution, it's installed into
-#  <prefix>/lib/ruby/<ver>, which unfortunately appears later in the
-#  search path. This means that if you have previously installed RDoc,
-#  and then install from ruby-lang, you'll pick up the old one by
-#  default. This hack checks for the condition, and readjusts the
-#  search path if necessary.
-
-def adjust_for_existing_rdoc(path)
-  
-  $stderr.puts %{
-  It seems as if you have a previously-installed RDoc in
-  the directory #{path}.
-
-  Because this is now out-of-date, you might want to consider
-  removing the directories:
-
-    #{File.join(path, "rdoc")}
-
-  and
+require 'rubygems'
 
-    #{File.join(path, "markup")}
+version = ">= 0.a"
 
-  }
-
-  # Move all the site_ruby directories to the end
-  p $:
-  $:.replace($:.partition {|path| /site_ruby/ !~ path}.flatten)
-  p $:
-end
-
-$:.each do |path|
-  if /site_ruby/ =~ path 
-    rdoc_path = File.join(path, 'rdoc', 'rdoc.rb')
-    if File.exists?(rdoc_path)
-      adjust_for_existing_rdoc(path)
-      break
-    end
+str = ARGV.first
+if str
+  str = str.b[/\A_(.*)_\z/, 1]
+  if str and Gem::Version.correct?(str)
+    version = str
+    ARGV.shift
   end
 end
 
-## End of Transitional Hack ##
-
-
-require 'rdoc/rdoc'
-
-begin
-  r = RDoc::RDoc.new
-  r.document(ARGV)
-rescue RDoc::RDocError => e
-  $stderr.puts e.message
-  exit(1)
+if Gem.respond_to?(:activate_bin_path)
+load Gem.activate_bin_path('rdoc', 'rdoc', version)
+else
+gem "rdoc", version
+load Gem.bin_path("rdoc", "rdoc", version)
 end
diff --git a/bin/ri b/bin/ri
index fb3e00eda3..0cc2f73bb6 100755
--- a/bin/ri
+++ b/bin/ri
@@ -1,49 +1,27 @@
 #!/usr/bin/env ruby
-# usage:
 #
-#   ri  name...
+# This file was generated by RubyGems.
 #
-# where name can be 
+# The application 'rdoc' is installed as part of a gem, and
+# this file is here to facilitate running it.
 #
-#   Class | Class::method | Class#method | Class.method | method
-#
-# All names may be abbreviated to their minimum unbiguous form. If a name
-# _is_ ambiguous, all valid options will be listed.
-#
-# The form '.' method matches either class or instance methods, while 
-# #method matches only instance and ::method matches only class methods.
-#
-#
-# == Installing Documentation
-#
-# 'ri' uses a database of documentation built by the RDoc utility.
-# 
-# So, how do you install this documentation on your system?
-# It depends on how you installed Ruby.
-#
-# <em>If you installed Ruby from source files</em> (that is, if it some point
-# you typed 'make' during the process :), you can install the RDoc
-# documentation yourself. Just go back to the place where you have 
-# your Ruby source and type
-#
-#    make install-doc
-#
-# You'll probably need to do this as a superuser, as the documentation
-# is installed in the Ruby target tree (normally somewhere under 
-# <tt>/usr/local</tt>.
-#
-# <em>If you installed Ruby from a binary distribution</em> (perhaps
-# using a one-click installer, or using some other packaging system),
-# then the team that produced the package probably forgot to package
-# the documentation as well. Contact them, and see if they can add
-# it to the next release.
-#
-
 
-require 'rdoc/ri/ri_driver'
+require 'rubygems'
 
-######################################################################
+version = ">= 0.a"
 
-ri = RiDriver.new
-ri.process_args
+str = ARGV.first
+if str
+  str = str.b[/\A_(.*)_\z/, 1]
+  if str and Gem::Version.correct?(str)
+    version = str
+    ARGV.shift
+  end
+end
 
+if Gem.respond_to?(:activate_bin_path)
+load Gem.activate_bin_path('rdoc', 'ri', version)
+else
+gem "rdoc", version
+load Gem.bin_path("rdoc", "ri", version)
+end
diff --git a/bin/testrb b/bin/testrb
deleted file mode 100755
index ff49cb5466..0000000000
--- a/bin/testrb
+++ /dev/null
@@ -1,5 +0,0 @@
-#!/usr/bin/env ruby
-require 'test/unit'
-(r = Test::Unit::AutoRunner.new(true)).process_args(ARGV) or
-  abort r.options.banner + " tests..."
-exit r.run
diff --git a/bootstraptest/pending.rb b/bootstraptest/pending.rb
new file mode 100644
index 0000000000..744c0adac1
--- /dev/null
+++ b/bootstraptest/pending.rb
@@ -0,0 +1,39 @@
+assert_equal 'A', %q{
+  class A
+    @@a = 'A'
+    def a=(x)
+      @@a = x
+    end
+    def a
+      @@a
+    end
+  end
+
+  B = A.dup
+  B.new.a = 'B'
+  A.new.a
+}, '[ruby-core:17019]'
+
+assert_equal 'ok', %q{
+  def m
+    lambda{
+      proc{
+        return :ng1
+      }
+    }.call.call
+    :ng2
+  end
+
+  begin
+    m()
+  rescue LocalJumpError
+    :ok
+  end
+}
+
+assert_normal_exit %q{
+  r = Range.allocate
+  def r.<=>(o) true end
+  r.instance_eval { initialize r, r }
+  r.inspect
+}
diff --git a/bootstraptest/runner.rb b/bootstraptest/runner.rb
new file mode 100755
index 0000000000..58619fae7e
--- /dev/null
+++ b/bootstraptest/runner.rb
@@ -0,0 +1,519 @@
+"exec" "${RUBY-ruby}" "-x" "$0" "$@" || true # -*- mode: ruby; coding: utf-8 -*-
+#!./ruby
+# $Id$
+
+# NOTE:
+# Never use optparse in this file.
+# Never use test/unit in this file.
+# Never use Ruby extensions in this file.
+# Maintain Ruby 1.8 compatibility for now
+
+begin
+  require 'fileutils'
+  require 'tmpdir'
+rescue LoadError
+  $:.unshift File.join(File.dirname(__FILE__), '../lib')
+  retry
+end
+
+if !Dir.respond_to?(:mktmpdir)
+  # copied from lib/tmpdir.rb
+  def Dir.mktmpdir(prefix_suffix=nil, tmpdir=nil)
+    case prefix_suffix
+    when nil
+      prefix = "d"
+      suffix = ""
+    when String
+      prefix = prefix_suffix
+      suffix = ""
+    when Array
+      prefix = prefix_suffix[0]
+      suffix = prefix_suffix[1]
+    else
+      raise ArgumentError, "unexpected prefix_suffix: #{prefix_suffix.inspect}"
+    end
+    tmpdir ||= Dir.tmpdir
+    t = Time.now.strftime("%Y%m%d")
+    n = nil
+    begin
+      path = "#{tmpdir}/#{prefix}#{t}-#{$$}-#{rand(0x100000000).to_s(36)}"
+      path << "-#{n}" if n
+      path << suffix
+      Dir.mkdir(path, 0700)
+    rescue Errno::EEXIST
+      n ||= 0
+      n += 1
+      retry
+    end
+
+    if block_given?
+      begin
+        yield path
+      ensure
+        FileUtils.remove_entry_secure path
+      end
+    else
+      path
+    end
+  end
+end
+
+def main
+  @ruby = File.expand_path('miniruby')
+  @verbose = false
+  $VERBOSE = false
+  $stress = false
+  @color = nil
+  @tty = nil
+  @quiet = false
+  dir = nil
+  quiet = false
+  tests = nil
+  ARGV.delete_if {|arg|
+    case arg
+    when /\A--ruby=(.*)/
+      @ruby = $1
+      @ruby.gsub!(/^([^ ]*)/){File.expand_path($1)}
+      @ruby.gsub!(/(\s+-I\s*)((?!(?:\.\/)*-(?:\s|\z))\S+)/){$1+File.expand_path($2)}
+      @ruby.gsub!(/(\s+-r\s*)(\.\.?\/\S+)/){$1+File.expand_path($2)}
+      true
+    when /\A--sets=(.*)/
+      tests = Dir.glob("#{File.dirname($0)}/test_{#{$1}}*.rb").sort
+      puts tests.map {|path| File.basename(path) }.inspect
+      true
+    when /\A--dir=(.*)/
+      dir = $1
+      true
+    when /\A(--stress|-s)/
+      $stress = true
+    when /\A--color(?:=(?:always|(auto)|(never)|(.*)))?\z/
+      warn "unknown --color argument: #$3" if $3
+      @color = $1 ? nil : !$2
+      true
+    when /\A--tty(=(?:yes|(no)|(.*)))?\z/
+      warn "unknown --tty argument: #$3" if $3
+      @tty = !$1 || !$2
+      true
+    when /\A(-q|--q(uiet))\z/
+      quiet = true
+      @quiet = true
+      true
+    when /\A(-v|--v(erbose))\z/
+      @verbose = true
+    when /\A(-h|--h(elp)?)\z/
+      puts(<<-End)
+Usage: #{File.basename($0, '.*')} --ruby=PATH [--sets=NAME,NAME,...]
+        --sets=NAME,NAME,...        Name of test sets.
+        --dir=DIRECTORY             Working directory.
+                                    default: /tmp/bootstraptestXXXXX.tmpwd
+        --color[=WHEN]              Colorize the output.  WHEN defaults to 'always'
+                                    or can be 'never' or 'auto'.
+    -s, --stress                    stress test.
+    -v, --verbose                   Output test name before exec.
+    -q, --quiet                     Don\'t print header message.
+    -h, --help                      Print this message and quit.
+End
+      exit true
+    when /\A-j/
+      true
+    else
+      false
+    end
+  }
+  if tests and not ARGV.empty?
+    $stderr.puts "--tests and arguments are exclusive"
+    exit false
+  end
+  tests ||= ARGV
+  tests = Dir.glob("#{File.dirname($0)}/test_*.rb").sort if tests.empty?
+  pathes = tests.map {|path| File.expand_path(path) }
+
+  @progress = %w[- \\ | /]
+  @progress_bs = "\b" * @progress[0].size
+  @tty = $stderr.tty? if @tty.nil?
+  case @color
+  when nil
+    @color = @tty && /dumb/ !~ ENV["TERM"]
+  end
+  @tty &&= !@verbose
+  if @color
+    # dircolors-like style
+    colors = (colors = ENV['TEST_COLORS']) ? Hash[colors.scan(/(\w+)=([^:\n]*)/)] : {}
+    begin
+      File.read(File.join(__dir__, "../test/colors")).scan(/(\w+)=([^:\n]*)/) do |n, c|
+        colors[n] ||= c
+      end
+    rescue
+    end
+    @passed = "\e[;#{colors["pass"] || "32"}m"
+    @failed = "\e[;#{colors["fail"] || "31"}m"
+    @reset = "\e[m"
+  else
+    @passed = @failed = @reset = ""
+  end
+  unless quiet
+    puts Time.now
+    if defined?(RUBY_DESCRIPTION)
+      puts "Driver is #{RUBY_DESCRIPTION}"
+    elsif defined?(RUBY_PATCHLEVEL)
+      puts "Driver is ruby #{RUBY_VERSION} (#{RUBY_RELEASE_DATE}#{RUBY_PLATFORM}) [#{RUBY_PLATFORM}]"
+    else
+      puts "Driver is ruby #{RUBY_VERSION} (#{RUBY_RELEASE_DATE}) [#{RUBY_PLATFORM}]"
+    end
+    puts "Target is #{`#{@ruby} -v`.chomp}"
+    puts
+    $stdout.flush
+  end
+
+  in_temporary_working_directory(dir) {
+    exec_test pathes
+  }
+end
+
+def erase(e = true)
+  if e and @columns > 0 and !@verbose
+    "\r#{" "*@columns}\r"
+  else
+    ""
+  end
+end
+
+def exec_test(pathes)
+  @count = 0
+  @error = 0
+  @errbuf = []
+  @location = nil
+  @columns = 0
+  @width = pathes.map {|path| File.basename(path).size}.max + 2
+  pathes.each do |path|
+    @basename = File.basename(path)
+    $stderr.printf("%s%-*s ", erase(@quiet), @width, @basename)
+    $stderr.flush
+    @columns = @width + 1
+    $stderr.puts if @verbose
+    count = @count
+    error = @error
+    load File.expand_path(path)
+    if @tty
+      if @error == error
+        msg = "PASS #{@count-count}"
+        @columns += msg.size - 1
+        $stderr.print "#{@progress_bs}#{@passed}#{msg}#{@reset}"
+      else
+        msg = "FAIL #{@error-error}/#{@count-count}"
+        $stderr.print "#{@progress_bs}#{@failed}#{msg}#{@reset}"
+        @columns = 0
+      end
+    end
+    $stderr.puts unless @quiet and @tty and @error == error
+  end
+  $stderr.print(erase) if @quiet
+  if @error == 0
+    if @count == 0
+      $stderr.puts "No tests, no problem"
+    else
+      $stderr.puts "#{@passed}PASS#{@reset} all #{@count} tests"
+    end
+    exit true
+  else
+    @errbuf.each do |msg|
+      $stderr.puts msg
+    end
+    $stderr.puts "#{@failed}FAIL#{@reset} #{@error}/#{@count} tests failed"
+    exit false
+  end
+end
+
+def show_progress(message = '')
+  if @verbose
+    $stderr.print "\##{@count} #{@location} "
+  elsif @tty
+    $stderr.print "#{@progress_bs}#{@progress[@count % @progress.size]}"
+  end
+  t = Time.now if @verbose
+  faildesc, errout = with_stderr {yield}
+  t = Time.now - t if @verbose
+  if !faildesc
+    if @tty
+      $stderr.print "#{@progress_bs}#{@progress[@count % @progress.size]}"
+    elsif @verbose
+      $stderr.printf(". %.3f\n", t)
+    else
+      $stderr.print '.'
+    end
+  else
+    $stderr.print "#{@failed}F"
+    $stderr.printf(" %.3f", t) if @verbose
+    $stderr.print "#{@reset}"
+    $stderr.puts if @verbose
+    error faildesc, message
+    unless errout.empty?
+      $stderr.print "#{@failed}stderr output is not empty#{@reset}\n", adjust_indent(errout)
+    end
+    if @tty and !@verbose
+      $stderr.printf("%-*s%s", @width, @basename, @progress[@count % @progress.size])
+    end
+  end
+rescue Interrupt
+  $stderr.puts "\##{@count} #{@location}"
+  raise
+rescue Exception => err
+  $stderr.print 'E'
+  $stderr.puts if @verbose
+  error err.message, message
+end
+
+def assert_check(testsrc, message = '', opt = '', **argh)
+  show_progress(message) {
+    result = get_result_string(testsrc, opt, **argh)
+    check_coredump
+    yield(result)
+  }
+end
+
+def assert_equal(expected, testsrc, message = '', opt = '', **argh)
+  newtest
+  assert_check(testsrc, message, opt, **argh) {|result|
+    if expected == result
+      nil
+    else
+      desc = "#{result.inspect} (expected #{expected.inspect})"
+      pretty(testsrc, desc, result)
+    end
+  }
+end
+
+def assert_match(expected_pattern, testsrc, message = '')
+  newtest
+  assert_check(testsrc, message) {|result|
+    if expected_pattern =~ result
+      nil
+    else
+      desc = "#{expected_pattern.inspect} expected to be =~\n#{result.inspect}"
+      pretty(testsrc, desc, result)
+    end
+  }
+end
+
+def assert_not_match(unexpected_pattern, testsrc, message = '')
+  newtest
+  assert_check(testsrc, message) {|result|
+    if unexpected_pattern !~ result
+      nil
+    else
+      desc = "#{unexpected_pattern.inspect} expected to be !~\n#{result.inspect}"
+      pretty(testsrc, desc, result)
+    end
+  }
+end
+
+def assert_valid_syntax(testsrc, message = '')
+  newtest
+  assert_check(testsrc, message, '-c') {|result|
+    result if /Syntax OK/ !~ result
+  }
+end
+
+def assert_normal_exit(testsrc, *rest, timeout: nil, **opt)
+  newtest
+  message, ignore_signals = rest
+  message ||= ''
+  show_progress(message) {
+    faildesc = nil
+    filename = make_srcfile(testsrc)
+    old_stderr = $stderr.dup
+    timeout_signaled = false
+    begin
+      $stderr.reopen("assert_normal_exit.log", "w")
+      io = IO.popen("#{@ruby} -W0 #{filename}")
+      pid = io.pid
+      th = Thread.new {
+        io.read
+        io.close
+        $?
+      }
+      if !th.join(timeout)
+        Process.kill :KILL, pid
+        timeout_signaled = true
+      end
+      status = th.value
+    ensure
+      $stderr.reopen(old_stderr)
+      old_stderr.close
+    end
+    if status && status.signaled?
+      signo = status.termsig
+      signame = Signal.list.invert[signo]
+      unless ignore_signals and ignore_signals.include?(signame)
+        sigdesc = "signal #{signo}"
+        if signame
+          sigdesc = "SIG#{signame} (#{sigdesc})"
+        end
+        if timeout_signaled
+          sigdesc << " (timeout)"
+        end
+        faildesc = pretty(testsrc, "killed by #{sigdesc}", nil)
+        stderr_log = File.read("assert_normal_exit.log")
+        if !stderr_log.empty?
+          faildesc << "\n" if /\n\z/ !~ faildesc
+          stderr_log << "\n" if /\n\z/ !~ stderr_log
+          stderr_log.gsub!(/^.*\n/) { '| ' + $& }
+          faildesc << stderr_log
+        end
+      end
+    end
+    faildesc
+  }
+end
+
+def assert_finish(timeout_seconds, testsrc, message = '')
+  timeout_seconds *= 3 if RubyVM::MJIT.enabled? # for --jit-wait
+  newtest
+  show_progress(message) {
+    faildesc = nil
+    filename = make_srcfile(testsrc)
+    io = IO.popen("#{@ruby} -W0 #{filename}")
+    pid = io.pid
+    waited = false
+    tlimit = Time.now + timeout_seconds
+    diff = timeout_seconds
+    while diff > 0
+      if Process.waitpid pid, Process::WNOHANG
+        waited = true
+        break
+      end
+      if io.respond_to?(:read_nonblock)
+        if IO.select([io], nil, nil, diff)
+          begin
+            io.read_nonblock(1024)
+          rescue Errno::EAGAIN, IO::WaitReadable, EOFError
+            break
+          end while true
+        end
+      else
+        sleep 0.1
+      end
+      diff = tlimit - Time.now
+    end
+    if !waited
+      Process.kill(:KILL, pid)
+      Process.waitpid pid
+      faildesc = pretty(testsrc, "not finished in #{timeout_seconds} seconds", nil)
+    end
+    io.close
+    faildesc
+  }
+end
+
+def flunk(message = '')
+  newtest
+  show_progress('') { message }
+end
+
+def pretty(src, desc, result)
+  src = src.sub(/\A\s*\n/, '')
+  (/\n/ =~ src ? "\n#{adjust_indent(src)}" : src) + "  #=> #{desc}"
+end
+
+INDENT = 27
+
+def adjust_indent(src)
+  untabify(src).gsub(/^ {#{INDENT}}/o, '').gsub(/^/, '   ').sub(/\s*\z/, "\n")
+end
+
+def untabify(str)
+  str.gsub(/^\t+/) {' ' * (8 * $&.size) }
+end
+
+def make_srcfile(src, frozen_string_literal: nil)
+  filename = 'bootstraptest.tmp.rb'
+  File.open(filename, 'w') {|f|
+    f.puts "#frozen_string_literal:true" if frozen_string_literal
+    f.puts "GC.stress = true" if $stress
+    f.puts "print(begin; #{src}; end)"
+  }
+  filename
+end
+
+def get_result_string(src, opt = '', **argh)
+  if @ruby
+    filename = make_srcfile(src, **argh)
+    begin
+      `#{@ruby} -W0 #{opt} #{filename}`
+    ensure
+      raise Interrupt if $? and $?.signaled? && $?.termsig == Signal.list["INT"]
+      raise CoreDumpError, "core dumped" if $? and $?.coredump?
+    end
+  else
+    eval(src).to_s
+  end
+end
+
+def with_stderr
+  out = err = nil
+  begin
+    r, w = IO.pipe
+    stderr = $stderr.dup
+    $stderr.reopen(w)
+    w.close
+    reader = Thread.start {r.read}
+    begin
+      out = yield
+    ensure
+      $stderr.reopen(stderr)
+      err = reader.value
+    end
+  ensure
+    w.close rescue nil
+    r.close rescue nil
+  end
+  return out, err
+end
+
+def newtest
+  @location = File.basename(caller(2).first)
+  @count += 1
+  cleanup_coredump
+end
+
+def error(msg, additional_message)
+  msg = "#{@failed}\##{@count} #{@location}#{@reset}: #{msg}  #{additional_message}"
+  if @tty
+    $stderr.puts "#{erase}#{msg}"
+  else
+    @errbuf.push msg
+  end
+  @error += 1
+end
+
+def in_temporary_working_directory(dir)
+  if dir
+    Dir.mkdir dir
+    Dir.chdir(dir) {
+      yield
+    }
+  else
+    Dir.mktmpdir(["bootstraptest", ".tmpwd"]) {|d|
+      Dir.chdir(d) {
+        yield
+      }
+    }
+  end
+end
+
+def cleanup_coredump
+  FileUtils.rm_f 'core'
+  FileUtils.rm_f Dir.glob('core.*')
+  FileUtils.rm_f @ruby+'.stackdump' if @ruby
+end
+
+class CoreDumpError < StandardError; end
+
+def check_coredump
+  if File.file?('core') or not Dir.glob('core.*').empty? or
+      (@ruby and File.exist?(@ruby+'.stackdump'))
+    raise CoreDumpError, "core dumped"
+  end
+end
+
+main
diff --git a/bootstraptest/test_attr.rb b/bootstraptest/test_attr.rb
new file mode 100644
index 0000000000..721a847145
--- /dev/null
+++ b/bootstraptest/test_attr.rb
@@ -0,0 +1,36 @@
+assert_equal 'ok', %q{
+  module M
+    class A
+      class << self
+        attr_accessor :at
+        def workflow_rule
+          yield self
+        end
+
+        def eval_str(str)
+          eval(str)
+        end
+      end
+    end
+  end
+  begin
+    M::A.eval_str(<<-END)
+    workflow_rule do |r|
+      r.at 1
+    end
+    END
+  rescue ArgumentError => e
+    print "ok"
+  end
+}, '[ruby-core:14641]'
+
+assert_equal %{ok}, %{
+  class A
+    attr :m
+  end
+  begin
+    A.new.m(3)
+  rescue ArgumentError => e
+    print "ok"
+  end
+}, '[ruby-core:15120]'
diff --git a/bootstraptest/test_autoload.rb b/bootstraptest/test_autoload.rb
new file mode 100644
index 0000000000..a9f8e6dacd
--- /dev/null
+++ b/bootstraptest/test_autoload.rb
@@ -0,0 +1,70 @@
+assert_equal 'ok', %q{
+  File.unlink('zzz.rb') if File.file?('zzz.rb')
+  instance_eval do
+    autoload :ZZZ, './zzz.rb'
+    begin
+      ZZZ
+    rescue LoadError
+      :ok
+    end
+  end
+}, '[ruby-dev:43816]'
+
+assert_equal 'ok', %q{
+  open('zzz.rb', 'w') {|f| f.puts '' }
+  instance_eval do
+    autoload :ZZZ, './zzz.rb'
+    begin
+      ZZZ
+    rescue NameError
+      :ok
+    end
+  end
+}, '[ruby-dev:43816]'
+
+assert_equal 'ok', %q{
+  open('zzz.rb', 'w') {|f| f.puts 'class ZZZ; def self.ok;:ok;end;end'}
+  instance_eval do
+    autoload :ZZZ, './zzz.rb'
+    ZZZ.ok
+  end
+}, '[ruby-dev:43816]'
+
+assert_equal 'ok', %q{
+  open("zzz.rb", "w") {|f| f.puts "class ZZZ; def self.ok;:ok;end;end"}
+  autoload :ZZZ, "./zzz.rb"
+  ZZZ.ok
+}
+
+assert_equal 'ok', %q{
+  open("zzz.rb", "w") {|f| f.puts "class ZZZ; def self.ok;:ok;end;end"}
+  autoload :ZZZ, "./zzz.rb"
+  require "./zzz.rb"
+  ZZZ.ok
+}
+
+assert_equal 'okok', %q{
+  open("zzz.rb", "w") {|f| f.puts "class ZZZ; def self.ok;:ok;end;end"}
+  autoload :ZZZ, "./zzz.rb"
+  t1 = Thread.new {ZZZ.ok}
+  t2 = Thread.new {ZZZ.ok}
+  [t1.value, t2.value].join
+}
+
+assert_finish 5, %q{
+  autoload :ZZZ, File.expand_path(__FILE__)
+  begin
+    ZZZ
+  rescue NameError
+  end
+}, '[ruby-core:21696]'
+
+assert_equal 'A::C', %q{
+  open("zzz.rb", "w") {}
+  class A
+    autoload :C, "./zzz"
+    class C
+    end
+    C
+  end
+}
diff --git a/bootstraptest/test_block.rb b/bootstraptest/test_block.rb
new file mode 100644
index 0000000000..cdc5960a59
--- /dev/null
+++ b/bootstraptest/test_block.rb
@@ -0,0 +1,613 @@
+assert_equal %q{1}, %q{
+  1.times{
+    begin
+      a = 1
+    ensure
+      foo = nil
+    end
+  }
+}
+assert_equal %q{2}, %q{
+  [1,2,3].find{|x| x == 2}
+}
+assert_equal %q{2}, %q{
+  class E
+    include Enumerable
+    def each(&block)
+      [1, 2, 3].each(&block)
+    end
+  end
+  E.new.find {|x| x == 2 }
+}
+assert_equal %q{6}, %q{
+  sum = 0
+  for x in [1, 2, 3]
+    sum += x
+  end
+  sum
+}
+assert_equal %q{15}, %q{
+  sum = 0
+  for x in (1..5)
+    sum += x
+  end
+  sum
+}
+assert_equal %q{0}, %q{
+  sum = 0
+  for x in []
+    sum += x
+  end
+  sum
+}
+assert_equal %q{1}, %q{
+  ans = []
+  1.times{
+    for n in 1..3
+      a = n
+      ans << a
+    end
+  }
+}
+assert_equal %q{1..3}, %q{
+  ans = []
+  for m in 1..3
+    for n in 1..3
+      a = [m, n]
+      ans << a
+    end
+  end
+}
+assert_equal %q{[1, 2, 3]}, %q{
+  (1..3).to_a
+}
+assert_equal %q{[4, 8, 12]}, %q{
+  (1..3).map{|e|
+    e * 4
+  }
+}
+assert_equal %q{[1, 2, 3]}, %q{
+  class C
+    include Enumerable
+    def each
+      [1,2,3].each{|e|
+        yield e
+      }
+    end
+  end
+
+  C.new.to_a
+}
+assert_equal %q{[4, 5, 6]}, %q{
+  class C
+    include Enumerable
+    def each
+      [1,2,3].each{|e|
+        yield e
+      }
+    end
+  end
+
+  C.new.map{|e|
+    e + 3
+  }
+}
+assert_equal %q{100}, %q{
+  def m
+    yield
+  end
+  def n
+    yield
+  end
+
+  m{
+    n{
+      100
+    }
+  }
+}
+assert_equal %q{20}, %q{
+  def m
+    yield 1
+  end
+
+  m{|ib|
+    m{|jb|
+      i = 20
+    }
+  }
+}
+assert_equal %q{2}, %q{
+  def m
+    yield 1
+  end
+
+  m{|ib|
+    m{|jb|
+      ib = 20
+      kb = 2
+    }
+  }
+}
+assert_equal %q{3}, %q{
+  def iter1
+    iter2{
+      yield
+    }
+  end
+
+  def iter2
+    yield
+  end
+
+  iter1{
+    jb = 2
+    iter1{
+      jb = 3
+    }
+    jb
+  }
+}
+assert_equal %q{2}, %q{
+  def iter1
+    iter2{
+      yield
+    }
+  end
+
+  def iter2
+    yield
+  end
+
+  iter1{
+    jb = 2
+    iter1{
+      jb
+    }
+    jb
+  }
+}
+assert_equal %q{2}, %q{
+  def m
+    yield 1
+  end
+  m{|ib|
+    ib*2
+  }
+}
+assert_equal %q{92580}, %q{
+  def m
+    yield 12345, 67890
+  end
+  m{|ib,jb|
+    ib*2+jb
+  }
+}
+assert_equal %q{[10, nil]}, %q{
+  def iter
+    yield 10
+  end
+
+  a = nil
+  [iter{|a|
+    a
+  }, a]
+}
+assert_equal %q{21}, %q{
+  def iter
+    yield 10
+  end
+
+  iter{|a|
+    iter{|a|
+      a + 1
+    } + a
+  }
+}
+assert_equal %q{[10, 20, 30, 40, nil, nil, nil, nil]}, %q{
+  def iter
+    yield 10, 20, 30, 40
+  end
+
+  a = b = c = d = nil
+  iter{|a, b, c, d|
+    [a, b, c, d]
+  } + [a, b, c, d]
+}
+assert_equal %q{[10, 20, 30, 40, nil, nil]}, %q{
+  def iter
+    yield 10, 20, 30, 40
+  end
+
+  a = b = nil
+  iter{|a, b, c, d|
+    [a, b, c, d]
+  } + [a, b]
+}
+assert_equal %q{[1]}, %q{
+  $a = []
+
+  def iter
+    yield 1
+  end
+
+  def m
+    x = iter{|x|
+      $a << x
+      y = 0
+    }
+  end
+  m
+  $a
+}
+assert_equal %q{[1, [2]]}, %q{
+  def iter
+    yield 1, 2
+  end
+
+  iter{|a, *b|
+    [a, b]
+  }
+}
+assert_equal %q{[[1, 2]]}, %q{
+  def iter
+    yield 1, 2
+  end
+
+  iter{|*a|
+    [a]
+  }
+}
+assert_equal %q{[1, 2, []]}, %q{
+  def iter
+    yield 1, 2
+  end
+
+  iter{|a, b, *c|
+    [a, b, c]
+  }
+}
+assert_equal %q{[1, 2, nil, []]}, %q{
+  def iter
+    yield 1, 2
+  end
+
+  iter{|a, b, c, *d|
+    [a, b, c, d]
+  }
+}
+assert_equal %q{1}, %q{
+  def m
+    yield
+  end
+  m{
+    1
+  }
+}
+assert_equal %q{15129}, %q{
+  def m
+    yield 123
+  end
+  m{|ib|
+    m{|jb|
+      ib*jb
+    }
+  }
+}
+assert_equal %q{2}, %q{
+  def m a
+    yield a
+  end
+  m(1){|ib|
+    m(2){|jb|
+      ib*jb
+    }
+  }
+}
+assert_equal %q{9}, %q{
+  sum = 0
+  3.times{|ib|
+    2.times{|jb|
+      sum += ib + jb
+    }}
+  sum
+}
+assert_equal %q{10}, %q{
+  3.times{|bl|
+    break 10
+  }
+}
+assert_equal %q{[1, 2]}, %q{
+  def iter
+    yield 1,2,3
+  end
+
+  iter{|i, j|
+    [i, j]
+  }
+}
+assert_equal %q{[1, nil]}, %q{
+  def iter
+    yield 1
+  end
+
+  iter{|i, j|
+    [i, j]
+  }
+}
+
+assert_equal '0', %q{
+def m()
+end
+m {|(v0,*,(*)),|}
+m {|(*v0,(*)),|}
+m {|(v0,*v1,(*)),|}
+m {|((v0,*v1,v2)),|}
+m {|(v0,*v1,v2),|}
+m {|(v0,*v1,(v2)),|}
+m {|((*),*v0,v1),|}
+m {|((v0),*v1,v2),|}
+m {|(v0,v1,*v2,v3),|}
+m {|v0,(v1,*v2,v3),|}
+m {|(v0,*v1,v2),v3,|}
+m {|(v0,*v1,v2)|}
+m {|(v0,*v1,v2),&v3|}
+m {|(v0,*v1,v2),*|}
+m {|(v0,*v1,v2),*,&v3|}
+m {|*,(v0,*v1,v2)|}
+m {|*,(v0,*v1,v2),&v3|}
+m {|v0,*,(v1,*v2,v3)|}
+m {|v0,*,(v1,*v2,v3),&v4|}
+m {|(v0,*v1,v2),*,v3|}
+m {|(v0,*v1,v2),*,v3,&v4|}
+m {|(v0, *v1, v2)|}
+m {|(*,v)|}
+0
+}, "block parameter (shouldn't SEGV: [ruby-dev:31143])"
+
+assert_equal 'nil', %q{
+  def m
+    yield
+  end
+  m{|&b| b}.inspect
+}, '[ruby-dev:31147]'
+
+assert_equal 'nil', %q{
+  def m()
+    yield
+  end
+  m {|(v,(*))|}.inspect
+}, '[ruby-dev:31160]'
+
+assert_equal 'nil', %q{
+  def m()
+    yield
+  end
+  m {|(*,a,b)|}.inspect
+}, '[ruby-dev:31153]'
+
+assert_equal 'nil', %q{
+  def m()
+    yield
+  end
+  m {|((*))|}.inspect
+}
+
+assert_equal %q{[1, 1, [1, nil], [1, nil], [1, nil], [1, nil], [1, 1], 1, [1, nil], [1, nil], [1, nil], [1, nil], [[1, 1], [1, 1]], [1, 1], [1, 1], [1, 1], [1, nil], [1, nil], [[[1, 1], [1, 1]], [[1, 1], [1, 1]]], [[1, 1], [1, 1]], [[1, 1], [1, 1]], [[1, 1], [1, 1]], [1, 1], [1, 1], [[[[1, 1], [1, 1]], [[1, 1], [1, 1]]], [[[1, 1], [1, 1]], [[1, 1], [1, 1]]]], [[[1, 1], [1, 1]], [[1, 1], [1, 1]]], [[[1, 1], [1, 1]], [[1, 1], [1, 1]]], [[[1, 1], [1, 1]], [[1, 1], [1, 1]]], [[1, 1], [1, 1]], [[1, 1], [1, 1]]]}, %q{
+def m(ary = [])
+  yield(ary)
+end
+
+$ans = []
+o = 1
+5.times{
+  v,(*) = o; $ans << o
+  m(o){|(v,(*))| $ans << v}
+  ((x, y)) = o; $ans << [x, y]
+  m(o){|((x, y))| $ans << [x, y]}
+  (((x, y))) = o; $ans << [x, y]
+  m(o){|(((x, y)))| $ans << [x, y]}
+  o = [o, o]
+}; $ans
+}
+
+assert_equal '0', %q{
+  def m()
+    yield [0]
+  end
+  m {|*,v| v}.inspect
+}, '[ruby-dev:31437]'
+assert_equal '[0]', %q{
+  def m
+    yield [0]
+  end
+  m{|v, &b| v}.inspect
+}, '[ruby-dev:31440]'
+assert_equal 'ok', %q{
+  begin
+    lambda{|a|}.call(1, 2)
+  rescue ArgumentError
+    :ok
+  else
+    :ng
+  end
+}, '[ruby-dev:31464]'
+assert_equal 'ok', %q{
+  begin
+    lambda{|&b|}.call(3)
+  rescue ArgumentError
+    :ok
+  else
+    :ng
+  end
+}, '[ruby-dev:31472]'
+assert_equal 'ok', %q{
+  class C
+    def each
+      yield [1,2]
+      yield 1,2
+    end
+  end
+  vs1 = []
+  C.new.each {|*v| vs1 << v }
+  vs2 = []
+  C.new.to_enum.each {|*v| vs2 << v }
+  vs1 == vs2 ? :ok : :ng
+}, '[ruby-dev:32329]'
+
+assert_normal_exit %q{
+  e = [1,2,3].each
+  10000.times {
+    e = [e].each
+  }
+  Thread.new { GC.start }.join
+}, '[ruby-dev:32604]'
+
+
+assert_equal '[nil, []]', %q{
+  def m() yield nil,[] end
+  l = lambda {|*v| v}
+  GC.stress=true
+  r = m(&l)
+  GC.stress=false
+  r.inspect
+}, '[ruby-dev:32567]'
+
+assert_equal NilClass.to_s, %q{
+  r = false; 1.times{|&b| r = b}; r.class
+}
+
+assert_equal 'ok', %q{
+  class C
+    define_method(:foo) do |arg, &block|
+      if block then block.call else arg end
+    end
+  end
+  C.new.foo("ng") {"ok"}
+}, '[ruby-talk:266422]'
+
+assert_equal 'ok', %q{
+  class C
+    define_method(:xyz) do |o, k, &block|
+      block.call(o, k)
+    end
+  end
+  C.new.xyz("o","k") {|o, k| o+k}
+}, '[ruby-core:20544]'
+
+assert_equal 'ok', %q{
+  class C
+    define_method(:xyz) do |*args, &block|
+      block.call(*args)
+    end
+  end
+  C.new.xyz("o","k") {|*args| args.join("")}
+}, '[ruby-core:20544]'
+
+assert_equal 'ok', %q{
+  STDERR.reopen(STDOUT)
+  class C
+    define_method(:foo) do |&block|
+      block.call if block
+    end
+    result = "ng"
+    new.foo() {result = "ok"}
+    result
+  end
+}
+
+assert_equal "ok", %q{
+  class Bar
+    def bar; :ok; end
+  end
+  def foo
+    yield(Bar.new) if block_given?
+  end
+  foo(&:bar)
+}, '[ruby-core:14279]'
+
+assert_normal_exit %q{
+  class Controller
+    def respond_to(&block)
+      responder = Responder.new
+      block.call(responder)
+      responder.respond
+    end
+    def test_for_bug
+      respond_to{|format|
+        format.js{
+          puts "in test"
+          render{|obj|
+            puts obj
+          }
+        }
+      }
+    end
+    def render(&block)
+      puts "in render"
+    end
+  end
+
+  class Responder
+    def method_missing(symbol, &block)
+      puts "enter method_missing"
+      @response = Proc.new{
+        puts 'in method missing'
+        block.call
+      }
+      puts "leave method_missing"
+    end
+    def respond
+      @response.call
+    end
+  end
+  t = Controller.new
+  t.test_for_bug
+}, '[ruby-core:14395]'
+
+assert_equal 'true', %q{
+  class C0
+    def foo
+      block_given?
+    end
+  end
+
+  class C1 < C0
+    def foo
+      super
+    end
+  end
+
+  C1.new.foo{}
+}
+
+assert_equal 'true', %q{
+  class C0
+    def foo
+      block_given?
+    end
+  end
+
+  class C1 < C0
+    def foo
+      super()
+    end
+  end
+
+  C1.new.foo{}
+}
+
+assert_equal 'ok', %q{
+  1.times do
+    begin
+      raise
+    rescue
+      begin
+        raise
+      rescue
+        break
+      end
+    end
+  end
+  'ok'
+}
diff --git a/bootstraptest/test_class.rb b/bootstraptest/test_class.rb
new file mode 100644
index 0000000000..b7fe0a1acd
--- /dev/null
+++ b/bootstraptest/test_class.rb
@@ -0,0 +1,169 @@
+# class
+assert_equal 'true',    %q( class C; end
+                            Object.const_defined?(:C) )
+assert_equal 'Class',   %q( class C; end
+                            C.class )
+assert_equal 'C',       %q( class C; end
+                            C.name )
+assert_equal 'C',       %q( class C; end
+                            C.new.class )
+assert_equal 'C',       %q( class C; end
+                            C.new.class.name )
+assert_equal 'Class',   %q( class C; end
+                            C.new.class.class )
+assert_equal 'true',    %q( Object.__send__(:remove_const, :TrueClass)
+                            GC.start
+                            true.inspect)
+assert_equal 'false',   %q( Object.__send__(:remove_const, :FalseClass)
+                            GC.start
+                            false.inspect)
+assert_equal 'nil',     %q( Object.__send__(:remove_const, :NilClass)
+                            GC.start
+                            nil.inspect)
+
+
+# inherited class
+assert_equal 'true',    %q( class A; end
+                            class C < A; end
+                            Object.const_defined?(:C) )
+assert_equal 'Class',   %q( class A; end
+                            class C < A; end
+                            C.class )
+assert_equal 'C',       %q( class A; end
+                            class C < A; end
+                            C.name )
+assert_equal 'C',       %q( class A; end
+                            class C < A; end
+                            C.new.class )
+assert_equal 'C',       %q( class A; end
+                            class C < A; end
+                            C.new.class.name )
+assert_equal 'Class',   %q( class A; end
+                            class C < A; end
+                            C.new.class.class )
+
+# module
+assert_equal 'true',    %q( module M; end
+                            Object.const_defined?(:M) )
+assert_equal 'Module',  %q( module M; end
+                            M.class )
+assert_equal 'M',       %q( module M; end
+                            M.name )
+assert_equal 'C',       %q( module M; end
+                            class C; include M; end
+                            C.new.class )
+
+# nested class
+assert_equal 'A::B',    %q( class A; end
+                            class A::B; end
+                            A::B )
+assert_equal 'A::B',    %q( class A; end
+                            class A::B; end
+                            A::B.name )
+assert_equal 'A::B',    %q( class A; end
+                            class A::B; end
+                            A::B.new.class )
+assert_equal 'Class',   %q( class A; end
+                            class A::B; end
+                            A::B.new.class.class )
+assert_equal 'A::B::C', %q( class A; end
+                            class A::B; end
+                            class A::B::C; end
+                            A::B::C )
+assert_equal 'A::B::C', %q( class A; end
+                            class A::B; end
+                            class A::B::C; end
+                            A::B::C.name )
+assert_equal 'Class',   %q( class A; end
+                            class A::B; end
+                            class A::B::C; end
+                            A::B::C.class )
+assert_equal 'A::B::C', %q( class A; end
+                            class A::B; end
+                            class A::B::C; end
+                            A::B::C.new.class )
+assert_equal 'Class',   %q( class A; end
+                            class A::B; end
+                            class A::B::C; end
+                            A::B::C.new.class.class )
+assert_equal 'A::B2',   %q( class A; end
+                            class A::B; end
+                            class A::B2 < A::B; end
+                            A::B2 )
+assert_equal 'Class',   %q( class A; end
+                            class A::B; end
+                            class A::B2 < A::B; end
+                            A::B2.class )
+
+# reopen
+assert_equal 'true',    %q( class C; end;  c1 = ::C
+                            class C; end;  c2 = ::C
+                            c1.equal?(c2) )
+assert_equal '1',       %q( class C; end
+                            class A; end
+                            begin class C < A; end; rescue TypeError; 1 end )
+assert_equal '1',       %q( class C; end
+                            begin module C; end; rescue TypeError; 1 end )
+assert_equal '1',       %q( C = 1   # [yarv-dev:782]
+                            begin class C; end; rescue TypeError; 1 end )
+assert_equal '1',       %q( C = 1   # [yarv-dev:800]
+                            begin module C; end; rescue TypeError; 1 end )
+
+# colon2, colon3
+assert_equal '1',       %q( class A; end;  A::C = 1;  A::C )
+assert_equal '1',       %q( A = 7;  begin A::C = 7; rescue TypeError; 1 end )
+assert_equal '1',       %q( begin 7::C = 7; rescue TypeError; 1 end )
+assert_equal 'C',       %q( class A; class ::C; end end;  C )
+assert_equal 'Class',   %q( class A; class ::C; end end;  C.class )
+assert_equal 'OK',      %q( class A; ::C = "OK"; end;  C )
+assert_equal 'String',  %q( class A; ::C = "OK"; end;  C.class )
+
+# class/module dup
+assert_equal 'Class',   %q( class C; end;  C.dup.class )
+assert_equal 'Module',  %q( module M; end;  M.dup.class )
+
+
+assert_equal "ok", %q{
+  module Foo
+  end
+
+  begin
+    def foo(&b)
+      Foo.module_eval &b
+    end
+    foo{
+      def bar
+      end
+    }
+    bar()
+  rescue NameError
+    :ok
+  end
+}, '[ruby-core:14378]'
+
+assert_equal '3', %q{
+  $i = 0
+  class C
+    def self.const_missing *args
+      $i+=1
+    end
+  end
+
+  3.times{
+    C::FOO
+  }
+  $i
+}
+
+assert_match /::C\z/, %q{
+  c = nil
+  Module.new{|m| c = class m::C; name; end}
+  c
+}, '[ruby-dev:38456]'
+
+assert_normal_exit %q{
+  s = Symbol.dup
+  class << s
+  end
+  s.allocate.to_s
+}, '[ruby-core:30843]'
diff --git a/bootstraptest/test_env.rb b/bootstraptest/test_env.rb
new file mode 100644
index 0000000000..7d1b45b75e
--- /dev/null
+++ b/bootstraptest/test_env.rb
@@ -0,0 +1,12 @@
+assert_equal "true", %q{
+  ENV["ENVTEST"] = "\u{e9 3042 d76c}"
+  env = ENV["ENVTEST"]
+  env.valid_encoding?
+}
+
+# different encoding is used for PATH
+assert_equal "true", %q{
+  ENV["PATH"] = "\u{e9 3042 d76c}"
+  env = ENV["PATH"]
+  env.valid_encoding?
+}
diff --git a/bootstraptest/test_eval.rb b/bootstraptest/test_eval.rb
new file mode 100644
index 0000000000..8e90ac2728
--- /dev/null
+++ b/bootstraptest/test_eval.rb
@@ -0,0 +1,324 @@
+assert_equal %q{ok}, %q{
+  def m
+    a = :ok
+    $b = binding
+  end
+  m
+  eval('a', $b)
+}
+assert_equal %q{[:ok, :ok2]}, %q{
+  def m
+    a = :ok
+    $b = binding
+  end
+  m
+  eval('b = :ok2', $b)
+  eval('[a, b]', $b)
+}
+assert_equal %q{[nil, 1]}, %q{
+  $ans = []
+  def m
+    $b = binding
+  end
+  m
+  $ans << eval(%q{
+    $ans << eval(%q{
+      a
+    }, $b)
+    a = 1
+  }, $b)
+  $ans
+}
+assert_equal %q{C}, %q{
+  Const = :top
+  class C
+    Const = :C
+    def m
+      binding
+    end
+  end
+  eval('Const', C.new.m)
+}
+assert_equal %q{top}, %q{
+  Const = :top
+  a = 1
+  class C
+    Const = :C
+    def m
+      eval('Const', TOPLEVEL_BINDING)
+    end
+  end
+  C.new.m
+}
+assert_equal %q{:ok
+ok}, %q{
+  class C
+    $b = binding
+  end
+  eval %q{
+    def m
+      :ok
+    end
+  }, $b
+  p C.new.m
+}
+assert_equal %q{ok}, %q{
+  b = proc{
+    a = :ok
+    binding
+  }.call
+  a = :ng
+  eval("a", b)
+}
+assert_equal %q{C}, %q{
+  class C
+    def foo
+      binding
+    end
+  end
+  C.new.foo.eval("self.class.to_s")
+}
+assert_equal %q{1}, %q{
+  eval('1')
+}
+assert_equal %q{1}, %q{
+  eval('a=1; a')
+}
+assert_equal %q{1}, %q{
+  a = 1
+  eval('a')
+}
+assert_equal %q{ok}, %q{
+  __send__ :eval, %{
+    :ok
+  }
+}
+assert_equal %q{ok}, %q{
+  1.__send__ :instance_eval, %{
+    :ok
+  }
+}
+assert_equal %q{1}, %q{
+  1.instance_eval{
+    self
+  }
+}
+assert_equal %q{foo}, %q{
+  'foo'.instance_eval{
+    self
+  }
+}
+assert_equal %q{1}, %q{
+  class Integer
+    Const = 1
+  end
+  1.instance_eval %{
+    Const
+  }
+}
+assert_equal %q{top}, %q{
+  Const = :top
+  class C
+    Const = :C
+  end
+  C.module_eval{
+    Const
+  }
+}
+assert_equal %q{C}, %q{
+  Const = :top
+  class C
+    Const = :C
+  end
+  C.class_eval %{
+    def m
+      Const
+    end
+  }
+  C.new.m
+}
+assert_equal %q{top}, %q{
+  Const = :top
+  class C
+    Const = :C
+  end
+  C.class_eval{
+    def m
+      Const
+    end
+  }
+  C.new.m
+}
+assert_equal %q{[:top, :C, :top, :C]}, %q{
+  Const = :top
+  class C
+    Const = :C
+  end
+  $nest = false
+  $ans = []
+  def m
+    $ans << Const
+    C.module_eval %{
+      $ans << Const
+      Boo = false unless defined? Boo
+      unless $nest
+        $nest = true
+        m
+      end
+    }
+  end
+  m
+  $ans
+}
+assert_equal %q{[10, main]}, %q{
+  $nested = false
+  $ans = []
+  $pr = proc{
+    $ans << self
+    unless $nested
+      $nested = true
+      $pr.call
+    end
+  }
+  class C
+    def initialize &b
+      10.instance_eval(&b)
+    end
+  end
+  C.new(&$pr)
+  $ans
+}
+
+%w[break next redo].each do |keyword|
+  assert_match %r"Can't escape from eval with #{keyword}\z", %{
+    begin
+      eval "0 rescue #{keyword}"
+    rescue SyntaxError => e
+      e.message
+    end
+  }, '[ruby-dev:31372]'
+end
+
+assert_normal_exit %q{
+  STDERR.reopen(STDOUT)
+  class Foo
+     def self.add_method
+       class_eval("def some-bad-name; puts 'hello' unless @some_variable.some_function(''); end")
+     end
+  end
+  Foo.add_method
+}, '[ruby-core:14556] reported by Frederick Cheung'
+
+assert_equal 'ok', %q{
+  class Module
+    def my_module_eval(&block)
+      module_eval(&block)
+    end
+  end
+  class String
+    Integer.my_module_eval do
+      def hoge; end
+    end
+  end
+  if Integer.instance_methods(false).map{|m|m.to_sym}.include?(:hoge) &&
+     !String.instance_methods(false).map{|m|m.to_sym}.include?(:hoge)
+    :ok
+  else
+    :ng
+  end
+}, "[ruby-dev:34236]"
+
+assert_equal 'ok', %q{
+  begin
+    eval("class nil::Foo; end")
+    :ng
+  rescue Exception
+    :ok
+  end
+}
+
+assert_equal 'ok', %q{
+  begin
+    0.instance_eval { def m() :m end }
+    1.m
+    :ng
+  rescue Exception
+    :ok
+  end
+}, '[ruby-dev:34579]'
+
+assert_equal 'ok', %q{
+  begin
+    12.instance_eval { @@a }
+  rescue NameError
+    :ok
+  end
+}, '[ruby-core:16794]'
+
+assert_equal 'ok', %q{
+  begin
+    12.instance_exec { @@a }
+  rescue NameError
+    :ok
+  end
+}, '[ruby-core:16794]'
+
+assert_equal 'ok', %q{
+  nil.instance_eval {
+    def defd_using_instance_eval() :ok end
+  }
+  nil.defd_using_instance_eval
+}, '[ruby-core:28324]'
+
+assert_equal 'ok', %q{
+  nil.instance_exec {
+    def defd_using_instance_exec() :ok end
+  }
+  nil.defd_using_instance_exec
+}, '[ruby-core:28324]'
+
+assert_normal_exit %q{
+  eval("", method(:proc).call {}.binding)
+}
+
+assert_equal "", %q{
+  b = binding
+  10.times{
+    eval('', b)
+  }
+  begin
+    eval('1.times{raise}', b)
+  rescue => e
+    e.message
+  end
+}, '[ruby-dev:35392]'
+
+assert_equal "[:x]", %q{
+  def kaboom!
+    yield.eval("local_variables")
+  end
+
+  for x in enum_for(:kaboom!)
+    binding
+  end
+}, '[ruby-core:25125]'
+
+assert_normal_exit %q{
+  hash = {}
+  ("aaaa".."matz").each_with_index do |s, i|
+    hash[s] = i
+  end
+  begin
+    eval "class C; @@h = #{hash.inspect}; end"
+  end
+}, '[ruby-core:25714]'
+
+assert_normal_exit %q{
+  begin
+    eval("# encoding:utf-16le\nfoo")
+  rescue Exception => e
+    p e
+    RubyVM::InstructionSequence.compile("p:hello")
+  end
+}, 'check escaping the internal value th->base_block'
+
diff --git a/bootstraptest/test_exception.rb b/bootstraptest/test_exception.rb
new file mode 100644
index 0000000000..0fb6f552b8
--- /dev/null
+++ b/bootstraptest/test_exception.rb
@@ -0,0 +1,432 @@
+assert_equal %q{2}, %q{
+  begin
+    1+1
+  ensure
+    2+2
+  end
+}
+assert_equal %q{4}, %q{
+  begin
+    1+1
+    begin
+      2+2
+    ensure
+      3+3
+    end
+  ensure
+    4+4
+  end
+}
+assert_equal %q{4}, %q{
+  begin
+    1+1
+    begin
+      2+2
+    ensure
+      3+3
+    end
+  ensure
+    4+4
+    begin
+      5+5
+    ensure
+      6+6
+    end
+  end
+}
+assert_equal %q{NilClass}, %q{
+  a = nil
+  1.times{|e|
+    begin
+    rescue => err
+    end
+    a = err.class
+  }
+  a
+}
+assert_equal %q{RuntimeError}, %q{
+  a = nil
+  1.times{|e|
+    begin
+      raise
+    rescue => err
+    end
+    a = err.class
+  }
+  a
+}
+assert_equal %q{}, %q{
+  $!
+}
+assert_equal %q{FOO}, %q{
+  begin
+    raise "FOO"
+  rescue
+    $!
+  end
+}
+assert_equal %q{FOO}, %q{
+  def m
+    $!
+  end
+  begin
+    raise "FOO"
+  rescue
+    m()
+  end
+}
+assert_equal %q{[#<RuntimeError: BAR>, #<RuntimeError: FOO>]}, %q{
+  $ans = []
+  def m
+    $!
+  end
+  begin
+    raise "FOO"
+  rescue
+    begin
+      raise "BAR"
+    rescue
+      $ans << m()
+    end
+    $ans << m()
+  end
+  $ans
+}
+assert_equal %q{[#<RuntimeError: FOO>, #<RuntimeError: FOO>]}, %q{
+  $ans = []
+  def m
+    $!
+  end
+
+  begin
+    begin
+      raise "FOO"
+    ensure
+      $ans << m()
+    end
+  rescue
+    $ans << m()
+  end
+}
+assert_equal %q{[nil]}, %q{
+  $ans = []
+  def m
+    $!
+  end
+  def m2
+    1.times{
+      begin
+        return
+      ensure
+        $ans << m
+      end
+    }
+  end
+  m2
+  $ans
+}
+assert_equal %q{ok}, %q{
+  begin
+    raise
+  rescue
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  begin
+    raise
+  rescue
+    :ok
+  ensure
+    :ng
+  end
+}
+assert_equal %q{RuntimeError}, %q{
+  begin
+    raise
+  rescue => e
+    e.class
+  end
+}
+assert_equal %q{ng}, %q{
+  begin
+    raise
+  rescue StandardError
+    :ng
+  rescue Exception
+    :ok
+  end
+}
+assert_equal %q{c}, %q{
+  begin
+    begin
+      raise "a"
+    rescue
+      raise "b"
+    ensure
+      raise "c"
+    end
+  rescue => e
+    e.message
+  end
+}
+assert_equal %q{33}, %q{
+  def m a, b
+    a + b
+  end
+  m(1, begin
+         raise
+       rescue
+         2
+       end) +
+  m(10, begin
+         raise
+       rescue
+         20
+       ensure
+         30
+       end)
+}
+assert_equal %q{3}, %q{
+  def m a, b
+    a + b
+  end
+  m(begin
+      raise
+    rescue
+      1
+    end,
+    begin
+      raise
+    rescue
+      2
+    end)
+}
+assert_equal %q{ok3}, %q{
+  class E1 < Exception
+  end
+
+  def m
+    yield
+  end
+
+  begin
+    begin
+      begin
+        m{
+          raise
+        }
+      rescue E1
+        :ok2
+      ensure
+      end
+    rescue
+      :ok3
+    ensure
+    end
+  rescue E1
+    :ok
+  ensure
+  end
+}
+assert_equal %q{7}, %q{
+  $i = 0
+  def m
+    iter{
+      begin
+        $i += 1
+        begin
+          $i += 2
+          break
+        ensure
+
+        end
+      ensure
+        $i += 4
+      end
+      $i = 0
+    }
+  end
+
+  def iter
+    yield
+  end
+  m
+  $i
+}
+assert_equal %q{10}, %q{
+  $i = 0
+  def m
+    begin
+      $i += 1
+      begin
+        $i += 2
+        return
+      ensure
+        $i += 3
+      end
+    ensure
+      $i += 4
+    end
+    p :end
+  end
+  m
+  $i
+}
+assert_equal %q{1}, %q{
+  begin
+    1
+  rescue
+    2
+  end
+}
+assert_equal %q{4}, %q{
+  begin
+    1
+    begin
+      2
+    rescue
+      3
+    end
+    4
+  rescue
+    5
+  end
+}
+assert_equal %q{3}, %q{
+  begin
+    1
+  rescue
+    2
+  else
+    3
+  end
+}
+assert_equal %q{2}, %q{
+  begin
+    1+1
+  rescue
+    2+2
+  ensure
+    3+3
+  end
+   }
+assert_equal %q{2}, %q{
+  begin
+    1+1
+  rescue
+    2+2
+  ensure
+    3+3
+  end
+   }
+assert_equal %q{6}, %q{
+  begin
+    1+1
+  rescue
+    2+2
+  else
+    3+3
+  ensure
+    4+4
+  end
+   }
+assert_equal %q{12}, %q{
+ begin
+   1+1
+   begin
+     2+2
+   rescue
+     3+3
+   else
+     4+4
+   end
+ rescue
+   5+5
+ else
+   6+6
+ ensure
+   7+7
+ end
+   }
+assert_equal %q{ok}, %q{ #
+  proc{
+    begin
+      raise
+      break
+    rescue
+      :ok
+    end
+  }.call
+}
+assert_equal %q{}, %q{
+  proc do
+    begin
+      raise StandardError
+      redo
+    rescue StandardError
+    end
+  end.call
+}
+
+##
+assert_match /undefined method `foo\'/, %q{#`
+  STDERR.reopen(STDOUT)
+  class C
+    def inspect
+      bar {}
+    end
+
+    def bar
+      raise
+    ensure
+    end
+  end
+  C.new.foo
+}, "[ruby-dev:31407]"
+
+assert_equal 'nil', %q{
+  doit = false
+  exc = nil
+  t = Thread.new {
+    begin
+      doit = true
+      sleep 10
+    ensure
+      exc = $!
+    end
+  }
+  Thread.pass until doit
+  t.kill
+  t.join
+  exc.inspect
+}, '[ruby-dev:32608]'
+
+assert_equal 'divided by 0', %q{
+  class ZeroDivisionError
+    def self.new(message)
+      42
+    end
+  end
+  begin
+    1/0
+  rescue Exception => e
+    e.message
+  end
+}, '[ruby-core:24767]'
+
+assert_equal 'ok', %q{
+  class C
+    def ===(o)
+      true
+    end
+  end
+  begin
+    begin
+      raise
+    rescue C.new
+    end
+  rescue TypeError
+    :ok
+  end
+}
diff --git a/bootstraptest/test_finalizer.rb b/bootstraptest/test_finalizer.rb
new file mode 100644
index 0000000000..22a16b1220
--- /dev/null
+++ b/bootstraptest/test_finalizer.rb
@@ -0,0 +1,8 @@
+assert_normal_exit %q{
+a1,a2,b1,b2=Array.new(4){""}
+ObjectSpace.define_finalizer(b2,proc{})
+ObjectSpace.define_finalizer(b1,proc{b1.inspect})
+
+ObjectSpace.define_finalizer(a2,proc{a1.inspect})
+ObjectSpace.define_finalizer(a1,proc{})
+}, '[ruby-dev:35778]'
diff --git a/bootstraptest/test_flip.rb b/bootstraptest/test_flip.rb
new file mode 100644
index 0000000000..ff194868b2
--- /dev/null
+++ b/bootstraptest/test_flip.rb
@@ -0,0 +1 @@
+assert_equal %q{E}, %q{$_ = "E"; eval("nil if true..~/^E/",nil,"-e"); $_}
diff --git a/bootstraptest/test_flow.rb b/bootstraptest/test_flow.rb
new file mode 100644
index 0000000000..9da6d45cbd
--- /dev/null
+++ b/bootstraptest/test_flow.rb
@@ -0,0 +1,601 @@
+assert_equal %q{[1, 2, 4, 5, 6, 7, 8]}, %q{$a = []; begin;  ; $a << 1
+  [1,2].each{; $a << 2
+    break; $a << 3
+  }; $a << 4
+  begin; $a << 5
+  ensure; $a << 6
+  end; $a << 7
+; $a << 8
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 5, 6, 7, 8]}, %q{$a = []; begin;  ; $a << 1
+  begin; $a << 2
+    [1,2].each do; $a << 3
+      break; $a << 4
+    end; $a << 5
+  ensure; $a << 6
+  end; $a << 7
+; $a << 8
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{ok}, %q{
+  ["a"].inject("ng"){|x,y|
+    break :ok
+  }
+}
+assert_equal %q{ok}, %q{
+  unless ''.respond_to? :lines
+    class String
+      def lines
+        self
+      end
+    end
+  end
+
+  ('a').lines.map{|e|
+    break :ok
+  }
+}
+assert_equal %q{[1, 2, 4, 5]}, %q{$a = []; begin; ; $a << 1
+  ["a"].inject("ng"){|x,y|; $a << 2
+    break :ok; $a << 3
+  }; $a << 4
+; $a << 5
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 4, 5]}, %q{$a = []; begin; ; $a << 1
+  ('a'..'b').map{|e|; $a << 2
+    break :ok; $a << 3
+  }; $a << 4
+; $a << 5
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 5, 7, 8]}, %q{$a = []; begin;  ; $a << 1
+  [1,2].each do; $a << 2
+    begin; $a << 3
+      break; $a << 4
+    ensure; $a << 5
+    end; $a << 6
+  end; $a << 7
+; $a << 8
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 6, 9, 10]}, %q{$a = []; begin; ; $a << 1
+  i=0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    begin; $a << 5
+    ensure; $a << 6
+      break; $a << 7
+    end; $a << 8
+  end; $a << 9
+; $a << 10
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 7, 10, 11]}, %q{$a = []; begin; ; $a << 1
+  i=0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    begin; $a << 5
+      raise; $a << 6
+    ensure; $a << 7
+      break; $a << 8
+    end; $a << 9
+  end; $a << 10
+; $a << 11
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 7, 10, 11]}, %q{$a = []; begin; ; $a << 1
+  i=0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    begin; $a << 5
+      raise; $a << 6
+    rescue; $a << 7
+      break; $a << 8
+    end; $a << 9
+  end; $a << 10
+; $a << 11
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 5, 8, 9]}, %q{$a = []; begin;  ; $a << 1
+  [1,2].each do; $a << 2
+    begin; $a << 3
+      raise StandardError; $a << 4
+    ensure; $a << 5
+      break; $a << 6
+    end; $a << 7
+  end; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 5, 8, 9]}, %q{$a = []; begin;  ; $a << 1
+  [1,2].each do; $a << 2
+    begin; $a << 3
+      raise StandardError; $a << 4
+    rescue; $a << 5
+      break; $a << 6
+    end; $a << 7
+  end; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 6, 8, 10, 11]}, %q{$a = []; begin;  ; $a << 1
+  [1,2].each do; $a << 2
+    begin; $a << 3
+      begin; $a << 4
+        break; $a << 5
+      ensure; $a << 6
+      end; $a << 7
+    ensure; $a << 8
+    end; $a << 9
+  end; $a << 10
+; $a << 11
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 6, 7, 8, 10, 13, 3, 4, 5, 6, 7, 8, 10, 13, 3, 4, 5, 6, 7, 8, 10, 13, 14, 15]}, %q{$a = []; begin; ; $a << 1
+  i = 0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    j = 0; $a << 5
+    while j<3; $a << 6
+      j+=1; $a << 7
+      begin; $a << 8
+        raise; $a << 9
+      rescue; $a << 10
+        break; $a << 11
+      end; $a << 12
+    end; $a << 13
+  end; $a << 14
+; $a << 15
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 15, 3, 4, 5, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 15, 3, 4, 5, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 15, 16, 17]}, %q{$a = []; begin; ; $a << 1
+  i = 0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    j = 0; $a << 5
+    while j<3; $a << 6
+      j+=1; $a << 7
+      1.times{; $a << 8
+        begin; $a << 9
+          raise; $a << 10
+        rescue; $a << 11
+          break; $a << 12
+        end; $a << 13
+      }; $a << 14
+    end; $a << 15
+  end; $a << 16
+; $a << 17
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 6, 7, 8, 10, 13, 3, 4, 5, 6, 7, 8, 10, 13, 3, 4, 5, 6, 7, 8, 10, 13, 14, 15]}, %q{$a = []; begin; ; $a << 1
+  i = 0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    j = 0; $a << 5
+    while j<3; $a << 6
+      j+=1; $a << 7
+      begin; $a << 8
+        raise; $a << 9
+      ensure; $a << 10
+        break; $a << 11
+      end; $a << 12
+    end; $a << 13
+  end; $a << 14
+; $a << 15
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 15, 3, 4, 5, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 15, 3, 4, 5, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 6, 7, 8, 9, 11, 14, 15, 16, 17]}, %q{$a = []; begin; ; $a << 1
+  i = 0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    j = 0; $a << 5
+    while j<3; $a << 6
+      j+=1; $a << 7
+      1.times{; $a << 8
+        begin; $a << 9
+          raise; $a << 10
+        ensure; $a << 11
+          break; $a << 12
+        end; $a << 13
+      }; $a << 14
+    end; $a << 15
+  end; $a << 16
+; $a << 17
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 5, 8, 9]}, %q{$a = []; begin; ; $a << 1
+  while true; $a << 2
+    begin; $a << 3
+      break; $a << 4
+    ensure; $a << 5
+      break; $a << 6
+    end; $a << 7
+  end; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 5, 99]}, %q{
+$a = [];
+begin; ; $a << 1
+  while true; $a << 2
+    begin; $a << 3
+      break; $a << 4
+    ensure; $a << 5
+      raise; $a << 6
+    end; $a << 7
+  end; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 6, 8, 9, 10, 11]}, %q{$a = []; begin;  ; $a << 1
+  begin; $a << 2
+    [1,2].each do; $a << 3
+      begin; $a << 4
+        break; $a << 5
+      ensure; $a << 6
+      end; $a << 7
+    end; $a << 8
+  ensure; $a << 9
+  end; $a << 10
+; $a << 11
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 4, 99]}, %q{$a = []; begin; ; $a << 1
+  begin; $a << 2
+    raise StandardError; $a << 3
+  ensure; $a << 4
+  end; $a << 5
+; $a << 6
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4]}, %q{$a = []; begin;  ; $a << 1
+  begin; $a << 2
+  ensure; $a << 3
+  end ; $a << 4
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 5, 99]}, %q{$a = []; begin;  ; $a << 1
+  [1,2].each do; $a << 2
+    begin; $a << 3
+      break; $a << 4
+    ensure; $a << 5
+      raise StandardError; $a << 6
+    end; $a << 7
+  end; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{3}, %q{
+  def m a, b
+    a + b
+  end
+  m(1,
+    while true
+      break 2
+    end
+    )
+}
+assert_equal %q{4}, %q{
+  def m a, b
+    a + b
+  end
+  m(1,
+    (i=0; while i<2
+       i+=1
+       class C
+         next 2
+       end
+     end; 3)
+    )
+}
+assert_equal %q{34}, %q{
+  def m a, b
+    a+b
+  end
+  m(1, 1.times{break 3}) +
+  m(10, (1.times{next 3}; 20))
+}
+assert_equal %q{[1, 2, 3, 6, 7]}, %q{$a = []; begin; ; $a << 1
+  3.times{; $a << 2
+    class C; $a << 3
+      break; $a << 4
+    end; $a << 5
+  }; $a << 6
+; $a << 7
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 8, 9]}, %q{$a = []; begin; ; $a << 1
+  3.times{; $a << 2
+    class A; $a << 3
+      class B; $a << 4
+        break; $a << 5
+      end; $a << 6
+    end; $a << 7
+  }; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 2, 3, 2, 3, 6, 7]}, %q{$a = []; begin; ; $a << 1
+  3.times{; $a << 2
+    class C; $a << 3
+      next; $a << 4
+    end; $a << 5
+  }; $a << 6
+; $a << 7
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 2, 3, 4, 2, 3, 4, 8, 9]}, %q{$a = []; begin; ; $a << 1
+  3.times{; $a << 2
+    class C; $a << 3
+      class D; $a << 4
+        next; $a << 5
+      end; $a << 6
+    end; $a << 7
+  }; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 6, 7]}, %q{$a = []; begin; ; $a << 1
+  while true; $a << 2
+    class C; $a << 3
+      break; $a << 4
+    end; $a << 5
+  end; $a << 6
+; $a << 7
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 8, 9]}, %q{$a = []; begin; ; $a << 1
+  while true; $a << 2
+    class C; $a << 3
+      class D; $a << 4
+        break; $a << 5
+      end; $a << 6
+    end; $a << 7
+  end; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 3, 4, 5, 3, 4, 5, 3, 4, 5, 8, 9]}, %q{$a = []; begin; ; $a << 1
+  i=0; $a << 2
+  while i<3; $a << 3
+    i+=1; $a << 4
+    class C; $a << 5
+      next 10; $a << 6
+    end; $a << 7
+  end; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{1}, %q{
+  1.times{
+    while true
+      class C
+        begin
+          break
+        ensure
+          break
+        end
+      end
+    end
+  }
+}
+assert_equal %q{[1, 2, 3, 5, 2, 3, 5, 7, 8]}, %q{$a = []; begin;  ; $a << 1
+  [1,2].each do; $a << 2
+    begin; $a << 3
+      next; $a << 4
+    ensure; $a << 5
+    end; $a << 6
+  end; $a << 7
+; $a << 8
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 2, 6, 3, 5, 7, 8]}, %q{$a = []; begin;  ; $a << 1
+  o = "test"; $a << 2
+  def o.test(a); $a << 3
+    return a; $a << 4
+  ensure; $a << 5
+  end; $a << 6
+  o.test(123); $a << 7
+; $a << 8
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 4, 7, 5, 8, 9]}, %q{$a = []; begin; ; $a << 1
+  def m1 *args; $a << 2
+    ; $a << 3
+  end; $a << 4
+  def m2; $a << 5
+    m1(:a, :b, (return 1; :c)); $a << 6
+  end; $a << 7
+  m2; $a << 8
+; $a << 9
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 8, 2, 3, 4, 5, 9, 10]}, %q{$a = []; begin; ; $a << 1
+  def m(); $a << 2
+    begin; $a << 3
+      2; $a << 4
+    ensure; $a << 5
+      return 3; $a << 6
+    end; $a << 7
+  end; $a << 8
+  m; $a << 9
+; $a << 10
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 3, 11, 4, 5, 6, 7, 12, 13]}, %q{$a = []; begin; ; $a << 1
+  def m2; $a << 2
+  end; $a << 3
+  def m(); $a << 4
+    m2(begin; $a << 5
+         2; $a << 6
+       ensure; $a << 7
+         return 3; $a << 8
+       end); $a << 9
+    4; $a << 10
+  end; $a << 11
+  m(); $a << 12
+; $a << 13
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[1, 16, 2, 3, 4, 5, 6, 7, 10, 11, 17, 18]}, %q{$a = []; begin; ; $a << 1
+  def m; $a << 2
+    1; $a << 3
+    1.times{; $a << 4
+      2; $a << 5
+      begin; $a << 6
+        3; $a << 7
+        return; $a << 8
+        4; $a << 9
+      ensure; $a << 10
+        5; $a << 11
+      end; $a << 12
+      6; $a << 13
+    }; $a << 14
+    7; $a << 15
+  end; $a << 16
+  m(); $a << 17
+; $a << 18
+; rescue Exception; $a << 99; end; $a}
+assert_equal %q{[:ok, :ok2, :last]}, %q{
+  a = []
+  i = 0
+  begin
+    while i < 1
+      i+=1
+      begin
+        begin
+          next
+        ensure
+          a << :ok
+        end
+      ensure
+        a << :ok2
+      end
+    end
+  ensure
+    a << :last
+  end
+  a
+}
+assert_equal %q{[:ok, :ok2, :last]}, %q{
+  a = []
+  i = 0
+  begin
+    while i < 1
+      i+=1
+      begin
+        begin
+          break
+        ensure
+          a << :ok
+        end
+      ensure
+        a << :ok2
+      end
+    end
+  ensure
+    a << :last
+  end
+  a
+}
+assert_equal %q{[:ok, :ok2, :last]}, %q{
+  a = []
+  i = 0
+  begin
+    while i < 1
+      if i>0
+        break
+      end
+      i+=1
+      begin
+        begin
+          redo
+        ensure
+          a << :ok
+        end
+      ensure
+        a << :ok2
+      end
+    end
+  ensure
+    a << :last
+  end
+  a
+}
+assert_equal %Q{ENSURE\n}, %q{
+  def test
+    while true
+      return
+    end
+  ensure
+    puts("ENSURE")
+  end
+  test
+}, '[ruby-dev:37967]'
+
+[['[ruby-core:28129]', %q{
+  class Bug2728
+    include Enumerable
+    define_method(:dynamic_method) do
+      "dynamically defined method"
+    end
+    def each
+      begin
+        yield :foo
+      ensure
+        dynamic_method
+      end
+    end
+  end
+  e = Bug2728.new
+}],
+ ['[ruby-core:28132]', %q{
+  class Bug2729
+    include Enumerable
+    def each
+      begin
+        yield :foo
+      ensure
+        proc {}.call
+      end
+    end
+  end
+  e = Bug2729.new
+}],
+ ['[ruby-core:39125]', %q{
+  class Bug5234
+    include Enumerable
+    def each
+      begin
+        yield :foo
+      ensure
+        proc
+      end
+    end
+  end
+  e = Bug5234.new
+}],
+ ['[ruby-dev:45656]', %q{
+  class Bug6460
+    include Enumerable
+    def each
+      begin
+        yield :foo
+      ensure
+        1.times { Proc.new }
+      end
+    end
+  end
+  e = Bug6460.new
+}]].each do |bug, src|
+  assert_equal "foo", src + %q{e.detect {true}}, bug
+  assert_equal "true", src + %q{e.any? {true}}, bug
+  assert_equal "false", src + %q{e.all? {false}}, bug
+  assert_equal "true", src + %q{e.include?(:foo)}, bug
+end
+
+assert_equal "foo", %q{
+  class Bug6460
+    def m1
+      m2 {|e|
+        return e
+      }
+    end
+
+    def m2
+      begin
+        yield :foo
+      ensure
+        begin
+          begin
+            yield :foo
+          ensure
+            Proc.new
+            raise ''
+          end
+        rescue
+        end
+      end
+    end
+  end
+  Bug6460.new.m1
+}, '[ruby-dev:46372]'
+
+assert_equal "foo", %q{
+  obj = "foo"
+  if obj || any1
+    any2 = any2
+  else
+    raise obj.inspect
+  end
+  obj
+}, '[ruby-core:87830]'
diff --git a/bootstraptest/test_fork.rb b/bootstraptest/test_fork.rb
new file mode 100644
index 0000000000..83923dad97
--- /dev/null
+++ b/bootstraptest/test_fork.rb
@@ -0,0 +1,77 @@
+assert_equal '0', %q{
+  begin
+    GC.stress = true
+    pid = fork {}
+    Process.wait pid
+    $?.to_i
+  rescue NotImplementedError
+    0
+  end
+}, '[ruby-dev:32404]'
+
+assert_finish 10, %q{
+  begin
+    children = (1..10).map{
+      Thread.start{fork{}}.value
+    }
+    while !children.empty? and pid = Process.wait
+      children.delete(pid)
+    end
+  rescue NotImplementedError
+  end
+}, '[ruby-core:22158]'
+
+# temporarily stop this test to enable explicit failure when
+# timer thread couldn't be created (r61706 and r61717).
+assert_normal_exit(<<'End', '[ruby-dev:37934]') if false
+  main = Thread.current
+  Thread.new { sleep 0.01 until main.stop?; Thread.kill main }
+  Process.setrlimit(:NPROC, 1) if defined?(Process::RLIMIT_NPROC)
+  fork {}
+End
+
+assert_equal 'ok', %q{
+  begin
+    r, w = IO.pipe
+    if pid1 = fork
+      w.close
+      r.read(1)
+      Process.kill("USR1", pid1)
+      _, s = Process.wait2(pid1)
+      s.success? ? :ok : :ng
+    else
+      r.close
+      if pid2 = fork
+        trap("USR1") { Time.now.to_s; Process.kill("USR2", pid2) }
+        w.close
+        Process.wait2(pid2)
+      else
+        w.close
+        sleep 0.2
+      end
+      exit true
+    end
+  rescue NotImplementedError
+    :ok
+  end
+}, '[ruby-core:28924]'
+
+assert_equal '[1, 2]', %q{
+  a = []
+  main = Thread.current
+  trap(:INT) { a.push(1).size == 2 and main.wakeup }
+  trap(:TERM) { a.push(2).size == 2 and main.wakeup }
+  pid = $$
+  begin
+    pid = fork do
+      Process.kill(:INT, pid)
+      Process.kill(:TERM, pid)
+    end
+    Process.wait(pid)
+    100.times {break if a.size > 1; sleep 0.001}
+    a.sort
+  rescue NotImplementedError
+    [1, 2]
+  end
+}, '[ruby-dev:44005] [Ruby 1.9 - Bug #4950]'
+
diff --git a/bootstraptest/test_gc.rb b/bootstraptest/test_gc.rb
new file mode 100644
index 0000000000..eb68c9845e
--- /dev/null
+++ b/bootstraptest/test_gc.rb
@@ -0,0 +1,34 @@
+assert_normal_exit %q{
+a = []
+ms = "a".."k"
+("A".."Z").each do |mod|
+  mod = eval("module #{mod}; self; end")
+  ms.each do |meth|
+    iseq = RubyVM::InstructionSequence.compile("module #{mod}; def #{meth}; end; end")
+    GC.stress = true
+    iseq.eval
+    GC.stress = false
+  end
+  o = Object.new.extend(mod)
+  ms.each do |meth|
+    o.send(meth)
+  end
+end
+}, '[ruby-dev:39453]'
+
+assert_normal_exit %q{
+a = []
+ms = "a".."k"
+("A".."Z").each do |mod|
+  mod = eval("module #{mod}; self; end")
+  ms.each do |meth|
+    GC.stress = true
+    mod.module_eval {define_method(meth) {}}
+    GC.stress = false
+  end
+  o = Object.new.extend(mod)
+  ms.each do |meth|
+    o.send(meth)
+  end
+end
+}, '[ruby-dev:39453]'
diff --git a/bootstraptest/test_insns.rb b/bootstraptest/test_insns.rb
new file mode 100644
index 0000000000..88f36bd0c2
--- /dev/null
+++ b/bootstraptest/test_insns.rb
@@ -0,0 +1,420 @@
+# C0 coverage of each instructions
+
+# :NOTE: This is for development purpose; never consider this file as
+# ISeq compilation specification.
+
+begin
+  # This library brings some additional coverage.
+  # Not mandatory.
+  require 'rbconfig/sizeof'
+rescue LoadError
+  # OK, just skip
+else
+  $FIXNUM_MAX = RbConfig::LIMITS["FIXNUM_MAX"]
+  $FIXNUM_MIN = RbConfig::LIMITS["FIXNUM_MIN"]
+end
+
+fsl   = { frozen_string_literal: true } # used later
+tests = [
+  # insn ,   expression to generate such insn
+  [ 'nop',   %q{ raise rescue true }, ],
+
+  [ 'setlocal *, 0', %q{ x = true }, ],
+  [ 'setlocal *, 1', %q{ x = nil; -> { x = true }.call }, ],
+  [ 'setlocal',      %q{ x = nil; -> { -> { x = true }.() }.() }, ],
+  [ 'getlocal *, 0', %q{ x = true; x }, ],
+  [ 'getlocal *, 1', %q{ x = true; -> { x }.call }, ],
+  [ 'getlocal',      %q{ x = true; -> { -> { x }.() }.() }, ],
+
+  [ 'setblockparam', <<~'},', ], # {
+    def m&b
+      b = # here
+        proc { true }
+    end
+    m { false }.call
+  },
+  [ 'getblockparam', <<~'},', ], # {
+    def m&b
+      b # here
+    end
+    m { true }.call
+  },
+  [ 'getblockparamproxy', <<~'},', ], # {
+    def m&b
+      b # here
+        .call
+    end
+    m { true }
+  },
+
+  [ 'setspecial', %q{ true if true..true }, ],
+  [ 'getspecial', %q{ $&.nil? }, ],
+  [ 'getspecial', %q{ $`.nil? }, ],
+  [ 'getspecial', %q{ $'.nil? }, ],
+  [ 'getspecial', %q{ $+.nil? }, ],
+  [ 'getspecial', %q{ $1.nil? }, ],
+  [ 'getspecial', %q{ $128.nil? }, ],
+
+  [ 'getglobal', %q{ String === $0 }, ],
+  [ 'getglobal', %q{ $_.nil? }, ],
+  [ 'setglobal', %q{ $0 = "true" }, ],
+
+  [ 'setinstancevariable', %q{ @x = true }, ],
+  [ 'getinstancevariable', %q{ @x = true; @x }, ],
+
+  [ 'setclassvariable', %q{ @@x = true }, ],
+  [ 'getclassvariable', %q{ @@x = true; @@x }, ],
+
+  [ 'setconstant', %q{ X = true }, ],
+  [ 'setconstant', %q{ Object::X = true }, ],
+  [ 'getconstant', %q{ X = true; X }, ],
+  [ 'getconstant', %q{ X = true; Object::X }, ],
+
+  [ 'getinlinecache / setinlinecache', %q{ def x; X; end; X = true; x; x; x }, ],
+
+  [ 'putnil',               %q{ $~ == nil }, ],
+  [ 'putself',              %q{ $~ != self }, ],
+  [ 'putobject INT2FIX(0)', %q{ $~ != 0 }, ],
+  [ 'putobject INT2FIX(1)', %q{ $~ != 1 }, ],
+  [ 'putobject',            %q{ $~ != -1 }, ],
+  [ 'putobject',            %q{ $~ != /x/ }, ],
+  [ 'putobject',            %q{ $~ != :x }, ],
+  [ 'putobject',            %q{ $~ != (1..2) }, ],
+  [ 'putobject',            %q{ $~ != true }, ],
+  [ 'putobject',            %q{ /(?<x>x)/ =~ "x"; x == "x" }, ],
+
+  [ 'putspecialobject',         %q{ {//=>true}[//] }, ],
+  [ 'putiseq',                  %q{ -> { true }.() }, ],
+  [ 'putstring',                %q{ "true" }, ],
+  [ 'tostring / concatstrings', %q{ "#{true}" }, ],
+  [ 'freezestring',             %q{ "#{true}" }, fsl, ],
+  [ 'freezestring',             %q{ "#{true}" }, '-d', fsl, ],
+  [ 'toregexp',                 %q{ /#{true}/ =~ "true" && $~ }, ],
+  [ 'intern',                   %q{ :"#{true}" }, ],
+
+  [ 'newarray',    %q{ ["true"][0] }, ],
+  [ 'duparray',    %q{ [ true ][0] }, ],
+  [ 'expandarray', %q{ y = [ true, false, nil ]; x, = y; x }, ],
+  [ 'expandarray', %q{ y = [ true, false, nil ]; x, *z = y; x }, ],
+  [ 'expandarray', %q{ y = [ true, false, nil ]; x, *z, w = y; x }, ],
+  [ 'splatarray',  %q{ x, = *(y = true), false; x }, ],
+  [ 'concatarray', %q{ ["t", "r", *x = "u", "e"].join }, ],
+  [ 'concatarray', <<~'},', ],  # {
+    class X; def to_a; ['u']; end; end
+    ['t', 'r', *X.new, 'e'].join
+  },
+  [ 'concatarray', <<~'},', ],  # {
+    r = false
+    t = [true, nil]
+    q, w, e = r, *t             # here
+    w
+  },
+
+  [ 'newhash',  %q{ x = {}; x[x] = true }, ],
+  [ 'newhash',  %q{ x = true; { x => x }[x] }, ],
+  [ 'newhashfromarray', %q{ { a: true }[:a] }, ],
+  [ 'newrange', %q{ x = 1; [*(0..x)][0] == 0 }, ],
+  [ 'newrange', %q{ x = 1; [*(0...x)][0] == 0 }, ],
+
+  [ 'pop',     %q{ def x; true; end; x }, ],
+  [ 'dup',     %q{ x = y = true; x }, ],
+  [ 'dupn',    %q{ Object::X ||= true }, ],
+  [ 'reverse', %q{ q, (w, e), r = 1, [2, 3], 4; e == 3 }, ],
+  [ 'swap',    <<~'},', ],      # {
+    x = [[false, true]]
+    for i, j in x               # here
+      ;
+    end
+    j
+  },
+
+  [ 'topn',        %q{ x, y = [], 0; x[*y], = [true, false]; x[0] }, ],
+  [ 'setn',        %q{ x, y = [], 0; x[*y]  =  true        ; x[0] }, ],
+  [ 'adjuststack', %q{ x = [true]; x[0] ||= nil; x[0] }, ],
+
+  [ 'defined',      %q{ !defined?(x) }, ],
+  [ 'checkkeyword', %q{ def x x:rand;x end; x x: true }, ],
+  [ 'checktype',    %q{ x = true; "#{x}" }, ],
+  [ 'checkmatch',   <<~'},', ], # {
+    x = y = true
+    case x
+    when false
+      y = false
+    when true                   # here
+      y = nil
+    end
+    y == nil
+  },
+  [ 'checkmatch',   <<~'},', ], # {
+    x, y = true, [false]
+    case x
+    when *y                     # here
+      z = false
+    else
+      z = true
+    end
+    z
+  },
+  [ 'checkmatch',   <<~'},', ], # {
+    x = false
+    begin
+      raise
+    rescue                      # here
+      x = true
+    end
+    x
+  },
+
+  [ 'defineclass', %q{                 module X;    true end }, ],
+  [ 'defineclass', %q{ X = Module.new; module X;    true end }, ],
+  [ 'defineclass', %q{                 class X;     true end }, ],
+  [ 'defineclass', %q{ X = Class.new;  class X;     true end }, ],
+  [ 'defineclass', %q{ X = Class.new;  class Y < X; true end }, ],
+  [ 'defineclass', %q{ X = Class.new;  class << X;  true end }, ],
+  [ 'defineclass', <<~'},', ], # {
+    X = Class.new
+    Y = Class.new(X)
+    class Y < X
+      true
+    end
+  },
+
+  [ 'opt_send_without_block', %q{ true.to_s }, ],
+  [ 'send',                   %q{ true.tap {|i| i.to_s } }, ],
+  [ 'leave',                  %q{ def x; true; end; x }, ],
+  [ 'invokesuper',            <<~'},', ], # {
+    class X < String
+      def empty?
+        super                   # here
+      end
+    end
+   X.new.empty?
+  },
+  [ 'invokeblock',            <<~'},', ], # {
+    def x
+      return yield self         # here
+    end
+    x do
+      true
+    end
+  },
+
+  [ 'opt_str_freeze', %q{ 'true'.freeze }, ],
+  [ 'opt_str_uminus', %q{ -'true' }, ],
+  [ 'opt_str_freeze', <<~'},', ], # {
+    class String
+      def freeze
+        true
+      end
+    end
+    'true'.freeze
+  },
+
+  [ 'opt_newarray_max', %q{ [ ].max.nil? }, ],
+  [ 'opt_newarray_max', %q{ [1, x = 2, 3].max == 3 }, ],
+  [ 'opt_newarray_max', <<~'},', ], # {
+    class Array
+      def max
+        true
+      end
+    end
+    [1, x = 2, 3].max
+  },
+  [ 'opt_newarray_min', %q{ [ ].min.nil? }, ],
+  [ 'opt_newarray_min', %q{ [3, x = 2, 1].min == 1 }, ],
+  [ 'opt_newarray_min', <<~'},', ], # {
+    class Array
+      def min
+        true
+      end
+    end
+    [3, x = 2, 1].min
+  },
+
+  [ 'throw',        %q{ false.tap { break true } }, ],
+  [ 'branchif',     %q{ x = nil;  x ||= true }, ],
+  [ 'branchif',     %q{ x = true; x ||= nil; x }, ],
+  [ 'branchunless', %q{ x = 1;    x &&= true }, ],
+  [ 'branchunless', %q{ x = nil;  x &&= true; x.nil? }, ],
+  [ 'branchnil',    %q{ x = true; x&.to_s }, ],
+  [ 'branchnil',    %q{ x = nil;  (x&.to_s).nil? }, ],
+  [ 'jump',         <<~'},', ], # {
+    y = 1
+    x = if y == 0 then nil elsif y == 1 then true else nil end
+    x
+  },
+  [ 'jump',         <<~'},', ], # {
+    # ultra complicated situation: this ||= assignment only generates
+    # 15 instructions, not including the class definition.
+    class X; attr_accessor :x; end
+    x = X.new
+    x&.x ||= true               # here
+  },
+
+  [ 'once', %q{ /#{true}/o =~ "true" && $~ }, ],
+  [ 'once', <<~'},', ],         # {
+    def once expr
+      return /#{expr}/o         # here
+    end
+    x = once(true); x = once(false); x = once(nil);
+    x =~ "true" && $~
+  },
+  [ 'once', <<~'},', ],         # {
+    # recursive once
+    def once n
+      return %r/#{
+        if n == 0
+          true
+        else
+          once(n-1)             # here
+        end
+      }/ox
+    end
+    x = once(128); x = once(7); x = once(16);
+    x =~ "true" && $~
+  },
+  [ 'once', <<~'},', ],         # {
+    # inter-thread lockup situation
+    def once n
+      return Thread.start n do |m|
+        Thread.pass
+        next %r/#{
+          sleep m               # here
+          true
+        }/ox
+      end
+    end
+    x = once(1); y = once(0.1); z = y.value
+    z =~ "true" && $~
+  },
+
+  [ 'opt_case_dispatch', %q{ case   0 when 1.1 then false else true end }, ],
+  [ 'opt_case_dispatch', %q{ case 1.0 when 1.1 then false else true end }, ],
+
+  [ 'opt_plus',    %q{ 1 + 1 == 2 }, ],
+  if defined? $FIXNUM_MAX then
+    [ 'opt_plus',  %Q{ #{ $FIXNUM_MAX } + 1 == #{ $FIXNUM_MAX + 1 } }, ]
+  end,
+  [ 'opt_plus',    %q{ 1.0 + 1.0 == 2.0 }, ],
+  [ 'opt_plus',    %q{ x = +0.0.next_float; x + x >= x }, ],
+  [ 'opt_plus',    %q{ 't' + 'rue' }, ],
+  [ 'opt_plus',    %q{ ( ['t'] + ['r', ['u', ['e'], ], ] ).join }, ],
+  [ 'opt_plus',    %q{ Time.at(1) + 1 == Time.at(2) }, ],
+  [ 'opt_minus',   %q{ 1 - 1 == 0 }, ],
+  if defined? $FIXNUM_MIN then
+    [ 'opt_minus', %Q{ #{ $FIXNUM_MIN } - 1 == #{ $FIXNUM_MIN - 1 } }, ]
+  end,
+  [ 'opt_minus',   %q{ 1.0 - 1.0 == 0.0 }, ],
+  [ 'opt_minus',   %q{ x = -0.0.prev_float; x - x == 0.0 }, ],
+  [ 'opt_minus',   %q{ ( [false, true] - [false] )[0] }, ],
+  [ 'opt_mult',    %q{ 1 * 1 == 1 }, ],
+  [ 'opt_mult',    %q{ 1.0 * 1.0 == 1.0 }, ],
+  [ 'opt_mult',    %q{ x = +0.0.next_float; x * x <= x }, ],
+  [ 'opt_mult',    %q{ ( "ruet" * 3 )[7,4] }, ],
+  [ 'opt_div',     %q{ 1 / 1 == 1 }, ],
+  [ 'opt_div',     %q{ 1.0 / 1.0 == 1.0 }, ],
+  [ 'opt_div',     %q{ x = +0.0.next_float; x / x >= x }, ],
+  [ 'opt_div',     %q{ x = 1/2r; x / x == 1 }, ],
+  [ 'opt_mod',     %q{ 1 % 1 == 0 }, ],
+  [ 'opt_mod',     %q{ 1.0 % 1.0 == 0.0 }, ],
+  [ 'opt_mod',     %q{ x = +0.0.next_float; x % x == 0.0 }, ],
+  [ 'opt_mod',     %q{ '%s' % [ true ] }, ],
+
+  [ 'opt_eq', %q{ 1 == 1 }, ],
+  [ 'opt_eq', <<~'},', ],       # {
+    class X; def == other; true; end; end
+    X.new == true
+  },
+  [ 'opt_neq', %q{ 1 != 0 }, ],
+  [ 'opt_neq', <<~'},', ],       # {
+    class X; def != other; true; end; end
+    X.new != true
+  },
+
+  [ 'opt_lt', %q{            -1   <  0 }, ],
+  [ 'opt_lt', %q{            -1.0 <  0.0 }, ],
+  [ 'opt_lt', %q{ -0.0.prev_float <  0.0 }, ],
+  [ 'opt_lt', %q{              ?a <  ?z }, ],
+  [ 'opt_le', %q{            -1   <= 0 }, ],
+  [ 'opt_le', %q{            -1.0 <= 0.0 }, ],
+  [ 'opt_le', %q{ -0.0.prev_float <= 0.0 }, ],
+  [ 'opt_le', %q{              ?a <= ?z }, ],
+  [ 'opt_gt', %q{             1   >  0 }, ],
+  [ 'opt_gt', %q{             1.0 >  0.0 }, ],
+  [ 'opt_gt', %q{ +0.0.next_float >  0.0 }, ],
+  [ 'opt_gt', %q{              ?z >  ?a }, ],
+  [ 'opt_ge', %q{             1   >= 0 }, ],
+  [ 'opt_ge', %q{             1.0 >= 0.0 }, ],
+  [ 'opt_ge', %q{ +0.0.next_float >= 0.0 }, ],
+  [ 'opt_ge', %q{              ?z >= ?a }, ],
+
+  [ 'opt_ltlt', %q{  '' << 'true' }, ],
+  [ 'opt_ltlt', %q{ ([] << 'true').join }, ],
+  [ 'opt_ltlt', %q{ (1 << 31) == 2147483648 }, ],
+
+  [ 'opt_aref', %q{ ['true'][0] }, ],
+  [ 'opt_aref', %q{ { 0 => 'true'}[0] }, ],
+  [ 'opt_aref', %q{ 'true'[0] == ?t }, ],
+  [ 'opt_aset', %q{ [][0] = true }, ],
+  [ 'opt_aset', %q{ {}[0] = true }, ],
+  [ 'opt_aset', %q{ x = 'frue'; x[0] = 't'; x }, ],
+  [ 'opt_aset', <<~'},', ], # {
+    # opt_aref / opt_aset mixup situation
+    class X; def x; {}; end; end
+    x = X.new
+    x&.x[true] ||= true         # here
+  },
+
+  [ 'opt_aref_with', %q{ { 'true' => true }['true'] }, ],
+  [ 'opt_aref_with', %q{ Struct.new(:nil).new['nil'].nil? }, ],
+  [ 'opt_aset_with', %q{ {}['true'] = true }, ],
+  [ 'opt_aset_with', %q{ Struct.new(:true).new['true'] = true }, ],
+
+  [ 'opt_length',  %q{   'true'       .length == 4 }, ],
+  [ 'opt_length',  %q{   :true        .length == 4 }, ],
+  [ 'opt_length',  %q{ [ 'true' ]     .length == 1 }, ],
+  [ 'opt_length',  %q{ { 'true' => 1 }.length == 1 }, ],
+  [ 'opt_size',    %q{   'true'       .size   == 4 }, ],
+  [ 'opt_size',    %q{               1.size   >= 4 }, ],
+  [ 'opt_size',    %q{ [ 'true' ]     .size   == 1 }, ],
+  [ 'opt_size',    %q{ { 'true' => 1 }.size   == 1 }, ],
+  [ 'opt_empty_p', %q{ ''.empty? }, ],
+  [ 'opt_empty_p', %q{ [].empty? }, ],
+  [ 'opt_empty_p', %q{ {}.empty? }, ],
+  [ 'opt_empty_p', %q{ Queue.new.empty? }, ],
+
+  [ 'opt_succ',  %q{ 1.succ == 2 }, ],
+  if defined? $FIXNUM_MAX then
+    [ 'opt_succ',%Q{ #{ $FIXNUM_MAX }.succ == #{ $FIXNUM_MAX + 1 } }, ]
+  end,
+  [ 'opt_succ',  %q{ '1'.succ == '2' }, ],
+  [ 'opt_succ',  %q{ x = Time.at(0); x.succ == Time.at(1) }, ],
+
+  [ 'opt_not',  %q{ ! false }, ],
+  [ 'opt_neq', <<~'},', ],       # {
+    class X; def !; true; end; end
+    ! X.new
+  },
+
+  [ 'opt_regexpmatch1',  %q{ /true/ =~ 'true' && $~ }, ],
+  [ 'opt_regexpmatch1', <<~'},', ],       # {
+    class Regexp; def =~ other; true; end; end
+    /true/ =~ 'true'
+  },
+  [ 'opt_regexpmatch2',  %q{ 'true' =~ /true/ && $~ }, ],
+  [ 'opt_regexpmatch2', <<~'},', ],       # {
+    class String; def =~ other; true; end; end
+    'true' =~ /true/
+  },
+
+  [ 'opt_call_c_function', 'Struct.new(:x).new.x = true', ],
+]
+
+# normal path
+tests.compact.each {|(insn, expr, *a)| assert_equal 'true', expr, insn, *a }
+
+# with trace
+tests.compact.each {|(insn, expr, *a)|
+  progn = "set_trace_func(proc{})\n" + expr
+  assert_equal 'true', progn, insn, *a
+}
diff --git a/bootstraptest/test_io.rb b/bootstraptest/test_io.rb
new file mode 100644
index 0000000000..89c00d0b88
--- /dev/null
+++ b/bootstraptest/test_io.rb
@@ -0,0 +1,112 @@
+assert_finish 5, %q{
+  r, w = IO.pipe
+  t1 = Thread.new { r.sysread(1) }
+  t2 = Thread.new { r.sysread(1) }
+  sleep 0.01 until t1.stop? and t2.stop?
+  w.write "a"
+  w.write "a"
+}, '[ruby-dev:31866]'
+
+assert_finish 10, %q{
+  begin
+    require "io/nonblock"
+    require "timeout"
+    timeout(3) do
+      r, w = IO.pipe
+      w.nonblock?
+      w.nonblock = true
+      w.write_nonblock("a" * 100000)
+      w.nonblock = false
+      t1 = Thread.new { w.write("b" * 4096) }
+      t2 = Thread.new { w.write("c" * 4096) }
+      sleep 0.5
+      r.sysread(4096).length
+      sleep 0.5
+      r.sysread(4096).length
+      t1.join
+      t2.join
+    end
+  rescue LoadError, Timeout::Error, NotImplementedError
+  end
+}, '[ruby-dev:32566]'
+
+assert_finish 1, %q{
+  r, w = IO.pipe
+  Thread.new {
+    w << "ab"
+    sleep 0.01
+    w << "ab"
+  }
+  r.gets("abab")
+}
+
+assert_equal 'ok', %q{
+  require 'tmpdir'
+  begin
+    tmpname = "#{Dir.tmpdir}/ruby-btest-#{$$}-#{rand(0x100000000).to_s(36)}"
+    rw = File.open(tmpname, File::RDWR|File::CREAT|File::EXCL)
+  rescue Errno::EEXIST
+    retry
+  end
+  save = STDIN.dup
+  STDIN.reopen(rw)
+  STDIN.reopen(save)
+  rw.close
+  File.unlink(tmpname)
+  :ok
+}
+
+assert_equal 'ok', %q{
+  require 'tmpdir'
+  begin
+    tmpname = "#{Dir.tmpdir}/ruby-btest-#{$$}-#{rand(0x100000000).to_s(36)}"
+    rw = File.open(tmpname, File::RDWR|File::CREAT|File::EXCL)
+  rescue Errno::EEXIST
+    retry
+  end
+  save = STDIN.dup
+  STDIN.reopen(rw)
+  STDIN.print "a"
+  STDIN.reopen(save)
+  rw.close
+  File.unlink(tmpname)
+  :ok
+}
+
+assert_equal 'ok', %q{
+  dup = STDIN.dup
+  dupfd = dup.fileno
+  dupfd == STDIN.dup.fileno ? :ng : :ok
+}, '[ruby-dev:46834]'
+
+assert_normal_exit %q{
+  ARGF.set_encoding "foo"
+}
+
+10.times do
+  assert_normal_exit %q{
+    at_exit { p :foo }
+
+    megacontent = "abc" * 12345678
+    #File.open("megasrc", "w") {|f| f << megacontent }
+
+    t0 = Thread.main
+    Thread.new { sleep 0.001 until t0.stop?; Process.kill(:INT, $$) }
+
+    r1, w1 = IO.pipe
+    r2, w2 = IO.pipe
+    t1 = Thread.new { w1 << megacontent; w1.close }
+    t2 = Thread.new { r2.read; r2.close }
+    IO.copy_stream(r1, w2) rescue nil
+    w2.close
+    r1.close
+    t1.join
+    t2.join
+  }, 'megacontent-copy_stream', ["INT"], :timeout => 10 or break
+end
+
+assert_normal_exit %q{
+  r, w = IO.pipe
+  STDOUT.reopen(w)
+  STDOUT.reopen(__FILE__, "r")
+}, '[ruby-dev:38131]'
diff --git a/bootstraptest/test_jump.rb b/bootstraptest/test_jump.rb
new file mode 100644
index 0000000000..595aaa7c4b
--- /dev/null
+++ b/bootstraptest/test_jump.rb
@@ -0,0 +1,308 @@
+assert_equal %q{ok}, %q{
+  def m
+    :ng1
+    mm{
+      yield
+    }
+    :ng2
+  end
+
+  def mm
+    :ng3
+    yield
+    :ng4
+  end
+
+  m{
+    break :ok
+  }
+}
+assert_equal %q{ok}, %q{
+  3.times{
+    break :ok
+  }
+}
+assert_equal %q{}, %q{
+  catch(:foo){
+    throw :foo
+  }
+}
+assert_equal %q{false}, %q{
+  catch(:foo){
+    throw :foo, false
+  }
+}
+assert_equal %q{}, %q{
+  catch(:foo){
+    throw :foo, nil
+  }
+}
+assert_equal %q{ok}, %q{
+  catch(:foo){
+    throw :foo, :ok
+  }
+}
+assert_equal %q{}, %q{
+  catch(:foo){
+    1.times{
+      throw :foo
+    }
+  }
+}
+assert_equal %q{ok}, %q{
+  catch(:foo){
+    1.times{
+      throw :foo, :ok
+    }
+  }
+}
+assert_equal %q{ok}, %q{
+  catch(:foo){
+    catch(:bar){
+      throw :foo, :ok
+    }
+    :ng
+  }
+}
+assert_equal %q{ok}, %q{
+  catch(:foo){
+    catch(:bar){
+      1.times{
+        throw :foo, :ok
+      }
+    }
+    :ng
+  }
+}
+assert_equal %q{2}, %q{
+  module Enumerable
+    def all_?
+      self.each{|e|
+        unless yield(e)
+          return false
+        end
+      }
+      true
+    end
+  end
+
+  xxx = 0
+  [1,2].each{|bi|
+    [3,4].each{|bj|
+      [true, nil, true].all_?{|be| be}
+      break
+    }
+    xxx += 1
+  }
+  xxx
+}
+assert_equal %q{ok}, %q{
+  def m
+    yield
+  end
+
+  m{
+    begin
+    ensure
+      break :ok
+    end
+  }
+}
+assert_equal %q{ok}, %q{
+  def m
+    yield
+    :ok
+  end
+  i=0
+  m{
+    if i>10
+      i*i
+    else
+      i+=1
+      next
+    end
+  }
+}
+assert_equal %q{ok}, %q{
+  def m
+    yield
+  end
+
+  m{
+    next :ok
+  }
+}
+assert_equal %q{131}, %q{
+  def m
+    yield + 10
+  end
+  i=0
+  m{
+    if i>10
+      i*i
+    else
+      i+=1
+      redo
+    end
+  }
+}
+assert_equal %q{ok}, %q{
+begin
+  eval %q{
+    1.times{
+      retry
+    }
+  }
+rescue SyntaxError
+  :ok
+end
+}
+assert_equal %q{3}, %q{
+  def m
+    return 3
+  end
+  m
+}
+assert_equal %q{ok}, %q{
+  def m
+    :ng1
+    mm{
+      return :ok
+    }
+    :ng2
+  end
+
+  def mm
+    :ng3
+    yield
+    :ng4
+  end
+  m
+}
+assert_equal %q{100}, %q{
+  $i = 0
+  def m
+    begin
+      iter{
+        return
+      }
+    ensure
+      $i = 100
+    end
+  end
+
+  def iter
+    yield
+  end
+  m
+  $i
+}
+assert_equal %q{ok}, %q{
+  def m
+    begin
+      raise
+    rescue
+      return :ok
+    end
+    :ng
+  end
+  m
+}
+assert_equal %q{1}, %q{
+  def m
+    begin
+      raise
+    rescue
+      return 1
+    end
+  end
+
+  m
+}
+assert_equal %q{1}, %q{
+  def m
+    begin
+      #
+    ensure
+      return 1
+    end
+  end
+
+  m
+}
+assert_equal 'ok', %q{
+  begin
+    catch {|t| throw t, :ok }
+  rescue ArgumentError
+    :ng
+  end
+}, '[ruby-dev:31609]'
+
+assert_equal "1", %q{
+  catch do |t|
+    begin
+      throw t, 1
+      2
+    ensure
+      3
+    end
+  end
+}, "[ruby-dev:31698]"
+
+assert_normal_exit %q{
+  f = 0
+  1.times do
+    begin
+      f += 1
+    ensure
+      redo unless f > 2
+    end
+  end
+}
+
+assert_normal_exit %q{
+  -> do
+    1.times do
+      begin
+        raise
+      rescue
+        return
+      end
+    end
+  end.call
+}
+
+assert_normal_exit %q{
+  while true
+    begin
+      raise
+      next
+    rescue
+    end
+    break
+  end
+}, '[ruby-core:28172]'
+
+assert_equal "true", %q{
+  class Object
+    def return_eigenclass
+      class << self
+        return self
+      end
+    end
+  end
+  s = "foo"
+  s.return_eigenclass == class << s; self; end
+}, '[ruby-core:21379]'
+
+assert_equal "true", %q{
+  class Object
+    def yield_eigenclass
+      class << self
+        yield self
+      end
+    end
+  end
+  s = "foo"
+  s.yield_eigenclass {|c| c == class << s; self; end }
+}, '[ruby-dev:40975]'
diff --git a/bootstraptest/test_literal.rb b/bootstraptest/test_literal.rb
new file mode 100644
index 0000000000..0c5102c46e
--- /dev/null
+++ b/bootstraptest/test_literal.rb
@@ -0,0 +1,229 @@
+# empty program
+assert_equal '',                ''
+assert_equal '',                ' '
+assert_equal '',                "\n"
+
+# special const
+assert_equal 'true',            'true'
+assert_equal 'TrueClass',       'true.class'
+assert_equal 'false',           'false'
+assert_equal 'FalseClass',      'false.class'
+assert_equal '',                'nil'
+assert_equal 'nil',             'nil.inspect'
+assert_equal 'NilClass',        'nil.class'
+assert_equal 'sym',             ':sym'
+assert_equal ':sym',            ':sym.inspect'
+assert_equal 'Symbol',          ':sym.class'
+assert_equal '1234',            '1234'
+assert_equal 'Integer',         '1234.class'
+assert_equal '1234',            '1_2_3_4'
+assert_equal 'Integer',         '1_2_3_4.class'
+assert_equal '18',              '0x12'
+assert_equal 'Integer',         '0x12.class'
+assert_equal '15',              '0o17'
+assert_equal 'Integer',         '0o17.class'
+assert_equal '5',               '0b101'
+assert_equal 'Integer',         '0b101.class'
+assert_equal '123456789012345678901234567890', '123456789012345678901234567890'
+assert_equal 'Integer',         '123456789012345678901234567890.class'
+assert_equal '2.0',             '2.0'
+assert_equal 'Float',           '1.3.class'
+
+# self
+assert_equal 'main',            'self'
+assert_equal 'Object',          'self.class'
+
+# string literal
+assert_equal 'a',               '?a'
+assert_equal 'String',          '?a.class'
+assert_equal 'A',               '?A'
+assert_equal 'String',          '?A.class'
+assert_equal "\n",              '?\n'
+assert_equal 'String',          '?\n.class'
+assert_equal ' ',               '?\ '
+assert_equal 'String',          '?\ .class'
+assert_equal 'string',          "'string'"
+assert_equal 'string',          '"string"'
+assert_equal 'string',          '%(string)'
+assert_equal 'string',          '%q(string)'
+assert_equal 'string',          '%Q(string)'
+assert_equal 'string string',   '"string string"'
+assert_equal ' ',               '" "'
+assert_equal "\0",              '"\0"'
+assert_equal "\1",              '"\1"'
+assert_equal "3",               '"\x33"'
+assert_equal "\n",              '"\n"'
+
+# dynamic string literal
+assert_equal '2',               '"#{1 + 1}"'
+assert_equal '16',              '"#{2 ** 4}"'
+assert_equal 'string',          's = "string"; "#{s}"'
+
+# dynamic symbol literal
+assert_equal 'a3c',             ':"a#{1+2}c"'
+assert_equal ':a3c',            ':"a#{1+2}c".inspect'
+assert_equal 'Symbol',          ':"a#{1+2}c".class'
+
+# xstring
+assert_equal "foo\n",           %q(`echo foo`)
+assert_equal "foo\n",           %q(s = "foo"; `echo #{s}`)
+
+# regexp
+assert_equal '',                '//.source'
+assert_equal 'Regexp',          '//.class'
+assert_equal '0',               '// =~ "a"'
+assert_equal '0',               '// =~ ""'
+assert_equal 'a',               '/a/.source'
+assert_equal 'Regexp',          '/a/.class'
+assert_equal '0',               '/a/ =~ "a"'
+assert_equal '0',               '/test/ =~ "test"'
+assert_equal '',                '/test/ =~ "tes"'
+assert_equal '0',               're = /test/; re =~ "test"'
+assert_equal '0',               'str = "test"; /test/ =~ str'
+assert_equal '0',               're = /test/; str = "test"; re =~ str'
+
+# dynamic regexp
+assert_equal 'regexp',          %q(/re#{'ge'}xp/.source)
+assert_equal 'Regexp',          %q(/re#{'ge'}xp/.class)
+
+# array
+assert_equal 'Array',           '[].class'
+assert_equal '0',               '[].size'
+assert_equal '0',               '[].length'
+assert_equal '[]',              '[].inspect'
+assert_equal 'Array',           '[0].class'
+assert_equal '1',               '[3].size'
+assert_equal '[3]',             '[3].inspect'
+assert_equal '3',               'a = [3]; a[0]'
+assert_equal 'Array',           '[1,2].class'
+assert_equal '2',               '[1,2].size'
+assert_equal '[1, 2]',          '[1,2].inspect'
+assert_equal 'Array',           '[1,2,3,4,5].class'
+assert_equal '5',               '[1,2,3,4,5].size'
+assert_equal '[1, 2, 3, 4, 5]', '[1,2,3,4,5].inspect'
+assert_equal '1',               'a = [1,2]; a[0]'
+assert_equal '2',               'a = [1,2]; a[1]'
+assert_equal 'Array',           'a = [1 + 2, 3 + 4, 5 + 6]; a.class'
+assert_equal '[3, 7, 11]',      'a = [1 + 2, 3 + 4, 5 + 6]; a.inspect'
+assert_equal '7',               'a = [1 + 2, 3 + 4, 5 + 6]; a[1]'
+assert_equal '1',               '([0][0] += 1)'
+assert_equal '1',               '([2][0] -= 1)'
+assert_equal 'Array',           'a = [obj = Object.new]; a.class'
+assert_equal '1',               'a = [obj = Object.new]; a.size'
+assert_equal 'true',            'a = [obj = Object.new]; a[0] == obj'
+assert_equal '5',               'a = [1,2,3]; a[1] = 5; a[1]'
+assert_equal 'bar',             '[*:foo];:bar'
+assert_equal '[1, 2]',          'def nil.to_a; [2]; end; [1, *nil]'
+assert_equal '[1, 2]',          'def nil.to_a; [1, 2]; end; [*nil]'
+assert_equal '[0, 1, {2=>3}]',  '[0, *[1], 2=>3]', "[ruby-dev:31592]"
+
+
+# hash
+assert_equal 'Hash',            '{}.class'
+assert_equal '{}',              '{}.inspect'
+assert_equal 'Hash',            '{1=>2}.class'
+assert_equal '{1=>2}',          '{1=>2}.inspect'
+assert_equal '2',               'h = {1 => 2}; h[1]'
+assert_equal '0',               'h = {1 => 2}; h.delete(1); h.size'
+assert_equal '',                'h = {1 => 2}; h.delete(1); h[1]'
+assert_equal '2',               'h = {"string" => "literal", "goto" => "hell"}; h.size'
+assert_equal 'literal', 'h = {"string"=>"literal", "goto"=>"hell"}; h["string"]'
+assert_equal 'hell', 'h = {"string"=>"literal", "goto"=>"hell"}; h["goto"]'
+
+# range
+assert_equal 'Range',           '(1..2).class'
+assert_equal '1..2',            '(1..2).inspect'
+assert_equal '1',               '(1..2).begin'
+assert_equal '2',               '(1..2).end'
+assert_equal 'false',           '(1..2).exclude_end?'
+assert_equal 'Range',           'r = 1..2; r.class'
+assert_equal '1..2',            'r = 1..2; r.inspect'
+assert_equal '1',               'r = 1..2; r.begin'
+assert_equal '2',               'r = 1..2; r.end'
+assert_equal 'false',           'r = 1..2; r.exclude_end?'
+assert_equal 'Range',           '(1...3).class'
+assert_equal '1...3',           '(1...3).inspect'
+assert_equal '1',               '(1...3).begin'
+assert_equal '3',               '(1...3).end'
+assert_equal 'true',            '(1...3).exclude_end?'
+assert_equal 'Range',           'r = (1...3); r.class'
+assert_equal '1...3',           'r = (1...3); r.inspect'
+assert_equal '1',               'r = (1...3); r.begin'
+assert_equal '3',               'r = (1...3); r.end'
+assert_equal 'true',            'r = (1...3); r.exclude_end?'
+assert_equal 'Range',           'r = (1+2 .. 3+4); r.class'
+assert_equal '3..7',            'r = (1+2 .. 3+4); r.inspect'
+assert_equal '3',               'r = (1+2 .. 3+4); r.begin'
+assert_equal '7',               'r = (1+2 .. 3+4); r.end'
+assert_equal 'false',           'r = (1+2 .. 3+4); r.exclude_end?'
+assert_equal 'Range',           'r = (1+2 ... 3+4); r.class'
+assert_equal '3...7',           'r = (1+2 ... 3+4); r.inspect'
+assert_equal '3',               'r = (1+2 ... 3+4); r.begin'
+assert_equal '7',               'r = (1+2 ... 3+4); r.end'
+assert_equal 'true',            'r = (1+2 ... 3+4); r.exclude_end?'
+assert_equal 'Range',           'r = ("a".."c"); r.class'
+assert_equal '"a".."c"',        'r = ("a".."c"); r.inspect'
+assert_equal 'a',               'r = ("a".."c"); r.begin'
+assert_equal 'c',               'r = ("a".."c"); r.end'
+
+assert_equal 'String',          '__FILE__.class'
+assert_equal 'Integer',         '__LINE__.class'
+
+###
+
+assert_equal 'ok', %q{
+  # this cause "called on terminated object".
+  ObjectSpace.each_object(Module) {|m| m.name.inspect }
+  :ok
+}
+
+assert_normal_exit %q{
+  begin
+    r = 0**-1
+    r + r
+  rescue
+  end
+}, '[ruby-dev:34524]'
+
+assert_normal_exit %q{
+  begin
+    r = Marshal.load("\x04\bU:\rRational[\ai\x06i\x05")
+    r + r
+  rescue
+  end
+}, '[ruby-dev:34536]'
+
+assert_equal 'ok', %q{
+  "#{}""#{}ok"
+}, '[ruby-dev:38968]'
+
+assert_equal 'ok', %q{
+  "#{}o""#{}k""#{}"
+}, '[ruby-core:25284]'
+
+assert_equal 'ok', %q{ #  long array literal
+  x = nil
+  eval "a = [#{(1..10_000).map{'x'}.join(", ")}]"
+  :ok
+}
+
+assert_equal 'ok', %q{ #  long array literal (optimized)
+  eval "a = [#{(1..10_000).to_a.join(", ")}]"
+  :ok
+}
+
+assert_equal 'ok', %q{ #  long hash literal
+  x = nil
+  eval "a = {#{(1..10_000).map{|n| "#{n} => x"}.join(', ')}}"
+  :ok
+}
+
+assert_equal 'ok', %q{ #  long hash literal (optimized)
+  eval "a = {#{(1..10_000).map{|n| "#{n} => #{n}"}.join(', ')}}"
+  :ok
+}
+
+assert_equal 'ok', %q{
+  [print(:ok), exit] # void literal with side-effect
+  :dummy
+}
diff --git a/bootstraptest/test_literal_suffix.rb b/bootstraptest/test_literal_suffix.rb
new file mode 100644
index 0000000000..4316c9e040
--- /dev/null
+++ b/bootstraptest/test_literal_suffix.rb
@@ -0,0 +1,54 @@
+# numbers with suffix
+assert_equal '0/1',             '0r'
+assert_equal 'Rational',        '0r.class'
+assert_equal '1/1',             '1r'
+assert_equal 'Rational',        '1r.class'
+assert_equal '-1/1',            '-1r'
+assert_equal 'Rational',        '(-1r).class'
+assert_equal '1/1',             '0x1r'
+assert_equal 'Rational',        '0x1r.class'
+assert_equal '1/1',             '0b1r'
+assert_equal 'Rational',        '0b1r.class'
+assert_equal '1/1',             '0d1r'
+assert_equal 'Rational',        '0d1r.class'
+assert_equal '1/1',             '0o1r'
+assert_equal 'Rational',        '0o1r.class'
+assert_equal '1/1',             '01r'
+assert_equal 'Rational',        '01r.class'
+assert_equal '6/5',             '1.2r'
+assert_equal 'Rational',        '1.2r.class'
+assert_equal '-6/5',            '-1.2r'
+assert_equal 'Rational',        '(-1.2r).class'
+assert_equal '0+0i',            '0i'
+assert_equal 'Complex',         '0i.class'
+assert_equal '0+1i',            '1i'
+assert_equal 'Complex',         '1i.class'
+assert_equal '0+1i',            '0x1i'
+assert_equal 'Complex',         '0x1i.class'
+assert_equal '0+1i',            '0b1i'
+assert_equal 'Complex',         '0b1i.class'
+assert_equal '0+1i',            '0d1i'
+assert_equal 'Complex',         '0d1i.class'
+assert_equal '0+1i',            '0o1i'
+assert_equal 'Complex',         '0o1i.class'
+assert_equal '0+1i',            '01i'
+assert_equal 'Complex',         '01i.class'
+assert_equal '0+1.2i',          '1.2i'
+assert_equal 'Complex',         '1.2i.class'
+assert_equal '0+1/1i',          '1ri'
+assert_equal 'Complex',         '1ri.class'
+assert_equal '0+6/5i',          '1.2ri'
+assert_equal 'Complex',         '1.2ri.class'
+assert_equal '0+10.0i',         '1e1i'
+assert_equal 'Complex',         '1e1i.class'
+assert_equal '1',               '1if true'
+assert_equal '1',               '1rescue nil'
+assert_equal '10000000000000000001/10000000000000000000',
+             '1.0000000000000000001r'
+
+assert_equal 'syntax error, unexpected tIDENTIFIER, expecting end-of-input',
+             %q{begin eval('1ir', nil, '', 0); rescue SyntaxError => e; e.message[/\A:(?:\d+:)? (.*)/, 1] end}
+assert_equal 'syntax error, unexpected tIDENTIFIER, expecting end-of-input',
+             %q{begin eval('1.2ir', nil, '', 0); rescue SyntaxError => e; e.message[/\A:(?:\d+:)? (.*)/, 1] end}
+assert_equal 'syntax error, unexpected tIDENTIFIER, expecting end-of-input',
+             %q{begin eval('1e1r', nil, '', 0); rescue SyntaxError => e; e.message[/\A:(?:\d+:)? (.*)/, 1] end}
diff --git a/bootstraptest/test_load.rb b/bootstraptest/test_load.rb
new file mode 100644
index 0000000000..e63c93a8f4
--- /dev/null
+++ b/bootstraptest/test_load.rb
@@ -0,0 +1,27 @@
+assert_equal 'ok', %q{
+  open("require-lock-test.rb", "w") {|f|
+    f.puts "sleep 0.1"
+    f.puts "module M"
+    f.puts "end"
+  }
+  $:.unshift Dir.pwd
+  vs = (1..2).map {|i|
+    Thread.start {
+      require "require-lock-test"
+      M
+    }
+  }.map {|t| t.value }
+  vs[0] == M && vs[1] == M ? :ok : :ng
+}, '[ruby-dev:32048]'
+
+assert_equal 'ok', %q{
+  %w[a a/foo b].each {|d| Dir.mkdir(d)}
+  open("b/foo", "w") {|f| f.puts "$ok = :ok"}
+  $:.replace(%w[a b])
+  begin
+    load "foo"
+    $ok
+  rescue => e
+    e.message
+  end
+}, '[ruby-dev:38097]'
diff --git a/bootstraptest/test_marshal.rb b/bootstraptest/test_marshal.rb
new file mode 100644
index 0000000000..7e34176169
--- /dev/null
+++ b/bootstraptest/test_marshal.rb
@@ -0,0 +1,5 @@
+
+assert_normal_exit %q{
+  Marshal.load(Marshal.dump({"k"=>"v"}), lambda {|v| v})
+}
+
diff --git a/bootstraptest/test_massign.rb b/bootstraptest/test_massign.rb
new file mode 100644
index 0000000000..0f63dd424a
--- /dev/null
+++ b/bootstraptest/test_massign.rb
@@ -0,0 +1,183 @@
+assert_equal '[[1], 2, 3]', '*v1, (a, b) = [1,[2, 3]]; [v1, a, b]'
+assert_equal '[[1], 2, 3]', '*v1,(*), (a, b) = [1,:x,[2, 3]]; [v1, a, b]'
+
+assert_equal '[]',          '*a = *nil; a'
+assert_equal '[nil]',       '*a = nil; a'
+assert_equal '2',           'a, a = 1, 2; a', "[ruby-dev:31522]"
+assert_equal '[1, 2]',      'a, b = 1, 2'
+assert_equal '[1, 2]', %q{
+  ans = []
+  trace_var(:$a){|v| ans << v}
+  trace_var(:$b){|v| ans << v}
+  $a, $b = 1, 2
+  ans
+}
+
+assert_equal 'ok', %q{
+  r = :ok
+  :ng.tap {|(r)|}
+  r
+}, '[ruby-dev:31507]'
+
+=begin
+# generated by this script:
+
+3.times{|i|
+  8.times{|e|
+    ary = (0...e).to_a
+    a,b,c,d,e,f = nil
+    vals = %w(a b c d e f)
+    vals[i] = '*' + vals[i]
+    program = "#{vals.join(", ")} = *ary"
+    eval(program)
+    ans = [a,b,c,d,e,f]
+    puts %Q{
+      assert_equal "#{ans.inspect}", %q{
+        ary = #{ary.inspect}
+        #{program}; [a, b, c, d, e, f]
+      }}
+  }
+}
+=end
+
+      assert_equal "[[], nil, nil, nil, nil, nil]", %q{
+        ary = []
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[[], 0, nil, nil, nil, nil]", %q{
+        ary = [0]
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[[], 0, 1, nil, nil, nil]", %q{
+        ary = [0, 1]
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[[], 0, 1, 2, nil, nil]", %q{
+        ary = [0, 1, 2]
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[[], 0, 1, 2, 3, nil]", %q{
+        ary = [0, 1, 2, 3]
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[[], 0, 1, 2, 3, 4]", %q{
+        ary = [0, 1, 2, 3, 4]
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[[0], 1, 2, 3, 4, 5]", %q{
+        ary = [0, 1, 2, 3, 4, 5]
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[[0, 1], 2, 3, 4, 5, 6]", %q{
+        ary = [0, 1, 2, 3, 4, 5, 6]
+        *a, b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[nil, [], nil, nil, nil, nil]", %q{
+        ary = []
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, [], nil, nil, nil, nil]", %q{
+        ary = [0]
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, [], 1, nil, nil, nil]", %q{
+        ary = [0, 1]
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, [], 1, 2, nil, nil]", %q{
+        ary = [0, 1, 2]
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, [], 1, 2, 3, nil]", %q{
+        ary = [0, 1, 2, 3]
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, [], 1, 2, 3, 4]", %q{
+        ary = [0, 1, 2, 3, 4]
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, [1], 2, 3, 4, 5]", %q{
+        ary = [0, 1, 2, 3, 4, 5]
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, [1, 2], 3, 4, 5, 6]", %q{
+        ary = [0, 1, 2, 3, 4, 5, 6]
+        a, *b, c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[nil, nil, [], nil, nil, nil]", %q{
+        ary = []
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, nil, [], nil, nil, nil]", %q{
+        ary = [0]
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, 1, [], nil, nil, nil]", %q{
+        ary = [0, 1]
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, 1, [], 2, nil, nil]", %q{
+        ary = [0, 1, 2]
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, 1, [], 2, 3, nil]", %q{
+        ary = [0, 1, 2, 3]
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, 1, [], 2, 3, 4]", %q{
+        ary = [0, 1, 2, 3, 4]
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, 1, [2], 3, 4, 5]", %q{
+        ary = [0, 1, 2, 3, 4, 5]
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+      assert_equal "[0, 1, [2, 3], 4, 5, 6]", %q{
+        ary = [0, 1, 2, 3, 4, 5, 6]
+        a, b, *c, d, e, f = *ary; [a, b, c, d, e, f]
+      }
+
+
+#
+assert_equal 'ok', %q{
+  a,s=[],"aaa"
+  300.times { a<<s; s=s.succ }
+  eval <<-END__
+  GC.stress=true
+  Fiber.new do
+    #{ a.join(",") },*zzz=1
+  end.resume
+  END__
+  :ok
+}, '[ruby-dev:32581]'
+
+assert_equal 'ok', %q{
+  while true
+    *, z = 1
+    break
+  end
+  :ok
+}, '[ruby-dev:32892]'
diff --git a/bootstraptest/test_method.rb b/bootstraptest/test_method.rb
new file mode 100644
index 0000000000..3462aa9434
--- /dev/null
+++ b/bootstraptest/test_method.rb
@@ -0,0 +1,1192 @@
+# regular argument
+assert_equal '1',       'def m() 1 end; m()'
+assert_equal '1',       'def m(a) a end; m(1)'
+assert_equal '[1, 2]',  'def m(a,b) [a, b] end; m(1,2)'
+assert_equal '[1, 2, 3]', 'def m(a,b,c) [a, b, c] end; m(1,2,3)'
+assert_match /\Awrong number of arguments \(.*\b1\b.* 0\)\z/, %q{
+  def m; end
+  begin
+    m(1)
+  rescue => e
+    e.message
+  end
+}
+
+assert_match /\Awrong number of arguments \(.*\b0\b.* 1\)\z/, %q{
+  def m a; end
+  begin
+    m
+  rescue => e
+    e.message
+  end
+}
+
+# default argument
+assert_equal '1',       'def m(x=1) x end; m()'
+assert_equal '1',       'def m(x=7) x end; m(1)'
+assert_equal '1',       'def m(a,x=1) x end; m(7)'
+assert_equal '1',       'def m(a,x=7) x end; m(7,1)'
+assert_equal '1',       'def m(a,b,x=1) x end; m(7,7)'
+assert_equal '1',       'def m(a,b,x=7) x end; m(7,7,1)'
+assert_equal '1',       'def m(a,x=1,y=1) x end; m(7)'
+assert_equal '1',       'def m(a,x=1,y=1) y end; m(7)'
+assert_equal '1',       'def m(a,x=7,y=1) x end; m(7,1)'
+assert_equal '1',       'def m(a,x=7,y=1) y end; m(7,1)'
+assert_equal '1',       'def m(a,x=7,y=7) x end; m(7,1,1)'
+assert_equal '1',       'def m(a,x=7,y=7) y end; m(7,1,1)'
+
+# rest argument
+assert_equal '[]',      'def m(*a) a end; m().inspect'
+assert_equal '[1]',     'def m(*a) a end; m(1).inspect'
+assert_equal '[1, 2]',  'def m(*a) a end; m(1,2).inspect'
+assert_equal '[]',      'def m(x,*a) a end; m(7).inspect'
+assert_equal '[1]',     'def m(x,*a) a end; m(7,1).inspect'
+assert_equal '[1, 2]',  'def m(x,*a) a end; m(7,1,2).inspect'
+assert_equal '[]',      'def m(x,y,*a) a end; m(7,7).inspect'
+assert_equal '[1]',     'def m(x,y,*a) a end; m(7,7,1).inspect'
+assert_equal '[1, 2]',  'def m(x,y,*a) a end; m(7,7,1,2).inspect'
+assert_equal '[]',      'def m(x,y=7,*a) a end; m(7).inspect'
+assert_equal '[]',      'def m(x,y,z=7,*a) a end; m(7,7).inspect'
+assert_equal '[]',      'def m(x,y,z=7,*a) a end; m(7,7,7).inspect'
+assert_equal '[]',      'def m(x,y,z=7,zz=7,*a) a end; m(7,7,7).inspect'
+assert_equal '[]',      'def m(x,y,z=7,zz=7,*a) a end; m(7,7,7,7).inspect'
+assert_equal '1',       'def m(x,y,z=7,zz=1,*a) zz end; m(7,7,7).inspect'
+assert_equal '1',       'def m(x,y,z=7,zz=1,*a) zz end; m(7,7,7).inspect'
+assert_equal '1',       'def m(x,y,z=7,zz=7,*a) zz end; m(7,7,7,1).inspect'
+
+# block argument
+assert_equal 'Proc',    'def m(&block) block end; m{}.class'
+assert_equal 'nil',     'def m(&block) block end; m().inspect'
+assert_equal 'Proc',    'def m(a,&block) block end; m(7){}.class'
+assert_equal 'nil',     'def m(a,&block) block end; m(7).inspect'
+assert_equal '1',       'def m(a,&block) a end; m(1){}'
+assert_equal 'Proc',    'def m(a,b=nil,&block) block end; m(7){}.class'
+assert_equal 'nil',     'def m(a,b=nil,&block) block end; m(7).inspect'
+assert_equal 'Proc',    'def m(a,b=nil,&block) block end; m(7,7){}.class'
+assert_equal '1',       'def m(a,b=nil,&block) b end; m(7,1){}'
+assert_equal 'Proc',    'def m(a,b=nil,*c,&block) block end; m(7){}.class'
+assert_equal 'nil',     'def m(a,b=nil,*c,&block) block end; m(7).inspect'
+assert_equal '1',       'def m(a,b=nil,*c,&block) a end; m(1).inspect'
+assert_equal '1',       'def m(a,b=1,*c,&block) b end; m(7).inspect'
+assert_equal '1',       'def m(a,b=7,*c,&block) b end; m(7,1).inspect'
+assert_equal '[1]',     'def m(a,b=7,*c,&block) c end; m(7,7,1).inspect'
+
+# splat
+assert_equal '1',       'def m(a) a end; m(*[1])'
+assert_equal '1',       'def m(x,a) a end; m(7,*[1])'
+assert_equal '1',       'def m(x,y,a) a end; m(7,7,*[1])'
+assert_equal '1',       'def m(a,b) a end; m(*[1,7])'
+assert_equal '1',       'def m(a,b) b end; m(*[7,1])'
+assert_equal '1',       'def m(x,a,b) b end; m(7,*[7,1])'
+assert_equal '1',       'def m(x,y,a,b) b end; m(7,7,*[7,1])'
+assert_equal '1',       'def m(a,b,c) a end; m(*[1,7,7])'
+assert_equal '1',       'def m(a,b,c) b end; m(*[7,1,7])'
+assert_equal '1',       'def m(a,b,c) c end; m(*[7,7,1])'
+assert_equal '1',       'def m(x,a,b,c) a end; m(7,*[1,7,7])'
+assert_equal '1',       'def m(x,y,a,b,c) a end; m(7,7,*[1,7,7])'
+
+# hash argument
+assert_equal '1',       'def m(h) h end; m(7=>1)[7]'
+assert_equal '1',       'def m(h) h end; m(7=>1).size'
+assert_equal '1',       'def m(h) h end; m(7=>1, 8=>7)[7]'
+assert_equal '2',       'def m(h) h end; m(7=>1, 8=>7).size'
+assert_equal '1',       'def m(h) h end; m(7=>1, 8=>7, 9=>7)[7]'
+assert_equal '3',       'def m(h) h end; m(7=>1, 8=>7, 9=>7).size'
+assert_equal '1',       'def m(x,h) h end; m(7, 7=>1)[7]'
+assert_equal '1',       'def m(x,h) h end; m(7, 7=>1, 8=>7)[7]'
+assert_equal '1',       'def m(x,h) h end; m(7, 7=>1, 8=>7, 9=>7)[7]'
+assert_equal '1',       'def m(x,y,h) h end; m(7,7, 7=>1)[7]'
+assert_equal '1',       'def m(x,y,h) h end; m(7,7, 7=>1, 8=>7)[7]'
+assert_equal '1',       'def m(x,y,h) h end; m(7,7, 7=>1, 8=>7, 9=>7)[7]'
+
+# block argument
+assert_equal '1',       %q(def m(&block) mm(&block) end
+                           def mm() yield 1 end
+                           m {|a| a })
+assert_equal '1',       %q(def m(x,&block) mm(x,&block) end
+                           def mm(x) yield 1 end
+                           m(7) {|a| a })
+assert_equal '1',       %q(def m(x,y,&block) mm(x,y,&block) end
+                           def mm(x,y) yield 1 end
+                           m(7,7) {|a| a })
+
+# recursive call
+assert_equal '1',       %q(def m(n) n == 0 ? 1 : m(n-1) end; m(5))
+
+# instance method
+assert_equal '1',       %q(class C; def m() 1 end end;  C.new.m)
+assert_equal '1',       %q(class C; def m(a) a end end;  C.new.m(1))
+assert_equal '1',       %q(class C; def m(a = 1) a end end;  C.new.m)
+assert_equal '[1]',     %q(class C; def m(*a) a end end;  C.new.m(1).inspect)
+assert_equal '1',       %q( class C
+                              def m() mm() end
+                              def mm() 1 end
+                            end
+                            C.new.m )
+
+# singleton method (const)
+assert_equal '1',       %q(class C; def C.m() 1 end end;  C.m)
+assert_equal '1',       %q(class C; def C.m(a) a end end;  C.m(1))
+assert_equal '1',       %q(class C; def C.m(a = 1) a end end;  C.m)
+assert_equal '[1]',     %q(class C; def C.m(*a) a end end;  C.m(1).inspect)
+assert_equal '1',       %q(class C; end; def C.m() 1 end;  C.m)
+assert_equal '1',       %q(class C; end; def C.m(a) a end;  C.m(1))
+assert_equal '1',       %q(class C; end; def C.m(a = 1) a end;  C.m)
+assert_equal '[1]',     %q(class C; end; def C.m(*a) a end;  C.m(1).inspect)
+assert_equal '1',       %q(class C; def m() 7 end end; def C.m() 1 end;  C.m)
+assert_equal '1',       %q( class C
+                              def C.m() mm() end
+                              def C.mm() 1 end
+                            end
+                            C.m )
+
+# singleton method (lvar)
+assert_equal '1',       %q(obj = Object.new; def obj.m() 1 end;  obj.m)
+assert_equal '1',       %q(obj = Object.new; def obj.m(a) a end;  obj.m(1))
+assert_equal '1',       %q(obj = Object.new; def obj.m(a=1) a end;  obj.m)
+assert_equal '[1]',     %q(obj = Object.new; def obj.m(*a) a end;  obj.m(1))
+assert_equal '1',       %q(class C; def m() 7 end; end
+                           obj = C.new
+                           def obj.m() 1 end
+                           obj.m)
+
+# inheritance
+assert_equal '1',       %q(class A; def m(a) a end end
+                           class B < A; end
+                           B.new.m(1))
+assert_equal '1',       %q(class A; end
+                           class B < A; def m(a) a end end
+                           B.new.m(1))
+assert_equal '1',       %q(class A; def m(a) a end end
+                           class B < A; end
+                           class C < B; end
+                           C.new.m(1))
+
+# include
+assert_equal '1',       %q(class A; def m(a) a end end
+                           module M; end
+                           class B < A; include M; end
+                           B.new.m(1))
+assert_equal '1',       %q(class A; end
+                           module M; def m(a) a end end
+                           class B < A; include M; end
+                           B.new.m(1))
+
+# alias
+assert_equal '1',       %q( def a() 1 end
+                            alias m a
+                            m() )
+assert_equal '1',       %q( class C
+                              def a() 1 end
+                              alias m a
+                            end
+                            C.new.m )
+assert_equal '1',       %q( class C
+                              def a() 1 end
+                              alias :m a
+                            end
+                            C.new.m )
+assert_equal '1',       %q( class C
+                              def a() 1 end
+                              alias m :a
+                            end
+                            C.new.m )
+assert_equal '1',       %q( class C
+                              def a() 1 end
+                              alias :m :a
+                            end
+                            C.new.m )
+assert_equal '1',       %q( class C
+                              def a() 1 end
+                              alias m a
+                              undef a
+                            end
+                            C.new.m )
+
+# undef
+assert_equal '1',       %q( class C
+                              def m() end
+                              undef m
+                            end
+                            begin C.new.m; rescue NoMethodError; 1 end )
+assert_equal '1',       %q( class A
+                              def m() end
+                            end
+                            class C < A
+                              def m() end
+                              undef m
+                            end
+                            begin C.new.m; rescue NoMethodError; 1 end )
+assert_equal '1',       %q( class A; def a() end end   # [yarv-dev:999]
+                            class B < A
+                              def b() end
+                              undef a, b
+                            end
+                            begin B.new.a; rescue NoMethodError; 1 end )
+assert_equal '1',       %q( class A; def a() end end   # [yarv-dev:999]
+                            class B < A
+                              def b() end
+                              undef a, b
+                            end
+                            begin B.new.b; rescue NoMethodError; 1 end )
+
+assert_equal '3',  %q{
+  def m1
+    1
+  end
+  alias m2 m1
+  alias :"#{'m3'}" m1
+  m1 + m2 + m3
+}, '[ruby-dev:32308]'
+assert_equal '1', %q{
+  def foobar
+  end
+  undef :"foo#{:bar}"
+  1
+}, '[ruby-dev:32308]'
+assert_equal '1', %q{
+  def foobar
+    1
+  end
+  alias :"bar#{:baz}" :"foo#{:bar}"
+  barbaz
+}, '[ruby-dev:32308]'
+
+# private
+assert_equal '1',       %q( class C
+                              def m() mm() end
+                              def mm() 1 end
+                              private :mm
+                            end
+                            C.new.m )
+assert_equal '1',       %q( class C
+                              def m() 7 end
+                              private :m
+                            end
+                            begin C.m; rescue NoMethodError; 1 end )
+assert_equal '1',       %q( class C
+                              def C.m() mm() end
+                              def C.mm() 1 end
+                              private_class_method :mm
+                            end
+                            C.m )
+assert_equal '1',       %q( class C
+                              def C.m() 7 end
+                              private_class_method :m
+                            end
+                            begin C.m; rescue NoMethodError; 1 end )
+assert_equal '1',       %q( class C; def m() 1 end end
+                            C.new.m   # cache
+                            class C
+                              alias mm m; private :mm
+                            end
+                            C.new.m
+                            begin C.new.mm; 7; rescue NoMethodError; 1 end )
+
+# nested method
+assert_equal '1',       %q( class C
+                              def m
+                                def mm() 1 end
+                              end
+                            end
+                            C.new.m
+                            C.new.mm )
+assert_equal '1',       %q( class C
+                              def m
+                                def mm() 1 end
+                              end
+                            end
+                            instance_eval "C.new.m; C.new.mm" )
+
+# method_missing
+assert_equal ':m',      %q( class C
+                              def method_missing(mid, *args) mid end
+                            end
+                            C.new.m.inspect )
+assert_equal ':mm',     %q( class C
+                              def method_missing(mid, *args) mid end
+                            end
+                            C.new.mm.inspect )
+assert_equal '[1, 2]',  %q( class C
+                              def method_missing(mid, *args) args end
+                            end
+                            C.new.m(1,2).inspect )
+assert_equal '1',       %q( class C
+                              def method_missing(mid, *args) yield 1 end
+                            end
+                            C.new.m {|a| a })
+assert_equal 'nil',     %q( class C
+                              def method_missing(mid, *args, &block) block end
+                            end
+                            C.new.m.inspect )
+
+# send
+assert_equal '1',       %q( class C; def m() 1 end end;
+                            C.new.__send__(:m) )
+assert_equal '1',       %q( class C; def m() 1 end end;
+                            C.new.send(:m) )
+assert_equal '1',       %q( class C; def m(a) a end end;
+                            C.new.send(:m,1) )
+assert_equal '1',       %q( class C; def m(a,b) a end end;
+                            C.new.send(:m,1,7) )
+assert_equal '1',       %q( class C; def m(x,a=1) a end end;
+                            C.new.send(:m,7) )
+assert_equal '1',       %q( class C; def m(x,a=7) a end end;
+                            C.new.send(:m,7,1) )
+assert_equal '[1, 2]',  %q( class C; def m(*a) a end end;
+                            C.new.send(:m,1,2).inspect )
+assert_equal '1',       %q( class C; def m() 7 end; private :m end
+                            begin C.new.public_send(:m); rescue NoMethodError; 1 end )
+assert_equal '1',       %q( class C; def m() 1 end; private :m end
+                            C.new.send(:m) )
+
+# with block
+assert_equal '[[:ok1, :foo], [:ok2, :foo, :bar]]',
+%q{
+  class C
+    def [](a)
+      $ary << [yield, a]
+    end
+    def []=(a, b)
+      $ary << [yield, a, b]
+    end
+  end
+
+  $ary = []
+  C.new[:foo, &lambda{:ok1}]
+  C.new[:foo, &lambda{:ok2}] = :bar
+  $ary
+}
+
+# with
+assert_equal '[:ok1, [:ok2, 11]]', %q{
+  class C
+    def []
+      $ary << :ok1
+      10
+    end
+    def []=(a)
+      $ary << [:ok2, a]
+    end
+  end
+  $ary = []
+  C.new[]+=1
+  $ary
+}
+
+# splat and block arguments
+assert_equal %q{[[[:x, :y, :z], NilClass], [[1, :x, :y, :z], NilClass], [[1, 2, :x, :y, :z], NilClass], [[:obj], NilClass], [[1, :obj], NilClass], [[1, 2, :obj], NilClass], [[], Proc], [[1], Proc], [[1, 2], Proc], [[], Proc], [[1], Proc], [[1, 2], Proc], [[:x, :y, :z], Proc], [[1, :x, :y, :z], Proc], [[1, 2, :x, :y, :z], Proc]]}, %q{
+def m(*args, &b)
+  $result << [args, b.class]
+end
+$result = []
+ary = [:x, :y, :z]
+obj = :obj
+b = Proc.new{}
+
+m(*ary)
+m(1,*ary)
+m(1,2,*ary)
+m(*obj)
+m(1,*obj)
+m(1,2,*obj)
+m(){}
+m(1){}
+m(1,2){}
+m(&b)
+m(1,&b)
+m(1,2,&b)
+m(*ary,&b)
+m(1,*ary,&b)
+m(1,2,*ary,&b)
+$result
+}
+
+# aset and splat
+assert_equal '4', %q{class Foo;def []=(a,b,c,d);end;end;Foo.new[1,*a=[2,3]]=4}
+assert_equal '4', %q{class Foo;def []=(a,b,c,d);end;end;def m(&blk)Foo.new[1,*a=[2,3],&blk]=4;end;m{}}
+
+# post test
+assert_equal %q{[1, 2, :o1, :o2, [], 3, 4, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4)}
+
+assert_equal %q{[1, 2, 3, :o2, [], 4, 5, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5)}
+
+assert_equal %q{[1, 2, 3, 4, [], 5, 6, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6)}
+
+assert_equal %q{[1, 2, 3, 4, [5], 6, 7, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7)}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6], 7, 8, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8)}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6, 7], 8, 9, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8, 9)}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6, 7, 8], 9, 10, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8, 9, 10)}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6, 7, 8, 9], 10, 11, NilClass, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11)}
+
+assert_equal %q{[1, 2, :o1, :o2, [], 3, 4, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4){}}
+
+assert_equal %q{[1, 2, 3, :o2, [], 4, 5, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5){}}
+
+assert_equal %q{[1, 2, 3, 4, [], 5, 6, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6){}}
+
+assert_equal %q{[1, 2, 3, 4, [5], 6, 7, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7){}}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6], 7, 8, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8){}}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6, 7], 8, 9, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8, 9){}}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6, 7, 8], 9, 10, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8, 9, 10){}}
+
+assert_equal %q{[1, 2, 3, 4, [5, 6, 7, 8, 9], 10, 11, Proc, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2, &b)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, b.class, x, y]
+end
+; m(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11){}}
+
+assert_equal %q{[1, 2, :o1, :o2, [], 3, 4, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, x, y]
+end
+; m(1, 2, 3, 4)}
+
+assert_equal %q{[1, 2, 3, :o2, [], 4, 5, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, x, y]
+end
+; m(1, 2, 3, 4, 5)}
+
+assert_equal %q{[1, 2, 3, 4, [], 5, 6, nil, nil]}, %q{
+def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2)
+  x, y = :x, :y if $foo
+  [m1, m2, o1, o2, r, p1, p2, x, y]
+end
+; m(1, 2, 3, 4, 5, 6)}
+
+
+#
+# super
+#
+=begin
+# below programs are generated by this program:
+
+BASE = <<EOS__
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; <TEST>; super; end; end
+EOS__
+
+tests = {
+%q{
+  def m
+} => %q{
+  C1.new.m
+},
+#
+%q{
+  def m a
+} => %q{
+  C1.new.m 1
+},
+%q{
+  def m a
+    a = :a
+} => %q{
+  C1.new.m 1
+},
+#
+%q{
+  def m a, o=:o
+} => %q{
+  C1.new.m 1
+  C1.new.m 1, 2
+},
+%q{
+  def m a, o=:o
+    a = :a
+} => %q{
+  C1.new.m 1
+  C1.new.m 1, 2
+},
+%q{
+  def m a, o=:o
+    o = :x
+} => %q{
+  C1.new.m 1
+  C1.new.m 1, 2
+},
+#
+%q{
+  def m a, *r
+} => %q{
+  C1.new.m 1
+  C1.new.m 1, 2
+  C1.new.m 1, 2, 3
+},
+%q{
+  def m a, *r
+    r = [:x, :y]
+} => %q{
+  C1.new.m 1
+  C1.new.m 1, 2
+  C1.new.m 1, 2, 3
+},
+#
+%q{
+  def m a, o=:o, *r
+} => %q{
+  C1.new.m 1
+  C1.new.m 1, 2
+  C1.new.m 1, 2, 3
+  C1.new.m 1, 2, 3, 4
+},
+#
+%q{
+  def m a, o=:o, *r, &b
+} => %q{
+  C1.new.m 1
+  C1.new.m 1, 2
+  C1.new.m 1, 2, 3
+  C1.new.m 1, 2, 3, 4
+  C1.new.m(1){}
+  C1.new.m(1, 2){}
+  C1.new.m(1, 2, 3){}
+  C1.new.m(1, 2, 3, 4){}
+},
+#
+"def m(m1, m2, o1=:o1, o2=:o2, p1, p2)" =>
+%q{
+C1.new.m(1,2,3,4)
+C1.new.m(1,2,3,4,5)
+C1.new.m(1,2,3,4,5,6)
+},
+#
+"def m(m1, m2, *r, p1, p2)" =>
+%q{
+C1.new.m(1,2,3,4)
+C1.new.m(1,2,3,4,5)
+C1.new.m(1,2,3,4,5,6)
+C1.new.m(1,2,3,4,5,6,7)
+C1.new.m(1,2,3,4,5,6,7,8)
+},
+#
+"def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2)" =>
+%q{
+C1.new.m(1,2,3,4)
+C1.new.m(1,2,3,4,5)
+C1.new.m(1,2,3,4,5,6)
+C1.new.m(1,2,3,4,5,6,7)
+C1.new.m(1,2,3,4,5,6,7,8)
+C1.new.m(1,2,3,4,5,6,7,8,9)
+},
+
+###
+}
+
+
+tests.each{|setup, methods| setup = setup.dup; setup.strip!
+  setup = BASE.gsub(/<TEST>/){setup}
+  methods.split(/\n/).each{|m| m = m.dup; m.strip!
+    next if m.empty?
+    expr = "#{setup}; #{m}"
+    result = eval(expr)
+    puts "assert_equal %q{#{result.inspect}}, %q{\n#{expr}}"
+    puts
+  }
+}
+
+=end
+
+assert_equal %q{[:C0_m, [1, 2, :o1, :o2, 3, 4]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, p1, p2); super; end; end
+; C1.new.m(1,2,3,4)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, :o2, 4, 5]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6)}
+
+assert_equal %q{[:C0_m, [1, :o]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, 2]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r; super; end; end
+; C1.new.m 1, 2}
+
+assert_equal %q{[:C0_m, [1, 2, 3]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r; super; end; end
+; C1.new.m 1, 2, 3}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r; super; end; end
+; C1.new.m 1, 2, 3, 4}
+
+assert_equal %q{[:C0_m, [:a]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a
+    a = :a; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6, 7]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6,7)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6, 7, 8]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6,7,8)}
+
+assert_equal %q{[:C0_m, [1, :o]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, 2]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m 1, 2}
+
+assert_equal %q{[:C0_m, [1, 2, 3]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m 1, 2, 3}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m 1, 2, 3, 4}
+
+assert_equal %q{[:C0_m, [1, :o]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m(1){}}
+
+assert_equal %q{[:C0_m, [1, 2]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m(1, 2){}}
+
+assert_equal %q{[:C0_m, [1, 2, 3]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m(1, 2, 3){}}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o, *r, &b; super; end; end
+; C1.new.m(1, 2, 3, 4){}}
+
+assert_equal %q{[:C0_m, [1, :x]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o
+    o = :x; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, :x]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o
+    o = :x; super; end; end
+; C1.new.m 1, 2}
+
+assert_equal %q{[:C0_m, [:a, :o]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o
+    a = :a; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [:a, 2]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o
+    a = :a; super; end; end
+; C1.new.m 1, 2}
+
+assert_equal %q{[:C0_m, [1]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, :x, :y]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, *r
+    r = [:x, :y]; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, :x, :y]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, *r
+    r = [:x, :y]; super; end; end
+; C1.new.m 1, 2}
+
+assert_equal %q{[:C0_m, [1, :x, :y]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, *r
+    r = [:x, :y]; super; end; end
+; C1.new.m 1, 2, 3}
+
+assert_equal %q{[:C0_m, [1, 2, :o1, :o2, 3, 4]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, :o2, 4, 5]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6, 7]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6,7)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6, 7, 8]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6,7,8)}
+
+assert_equal %q{[:C0_m, [1, 2, 3, 4, 5, 6, 7, 8, 9]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m(m1, m2, o1=:o1, o2=:o2, *r, p1, p2); super; end; end
+; C1.new.m(1,2,3,4,5,6,7,8,9)}
+
+assert_equal %q{[:C0_m, [1]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, *r; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, 2]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, *r; super; end; end
+; C1.new.m 1, 2}
+
+assert_equal %q{[:C0_m, [1, 2, 3]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, *r; super; end; end
+; C1.new.m 1, 2, 3}
+
+assert_equal %q{[:C0_m, []]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m; super; end; end
+; C1.new.m}
+
+assert_equal %q{[:C0_m, [1, :o]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o; super; end; end
+; C1.new.m 1}
+
+assert_equal %q{[:C0_m, [1, 2]]}, %q{
+class C0; def m *args; [:C0_m, args]; end; end
+class C1 < C0; def m a, o=:o; super; end; end
+; C1.new.m 1, 2}
+
+assert_equal 'ok', %q{
+  class C
+    def x=(n)
+    end
+    def m
+      self.x = :ok
+    end
+  end
+  C.new.m
+}
+
+assert_equal 'ok', %q{
+  proc{
+    $SAFE = 1
+    class C
+      def m
+        :ok
+      end
+    end
+  }.call
+  C.new.m
+}, '[ruby-core:11998]'
+
+assert_equal 'ok', %q{
+  class B
+    def m() :fail end
+  end
+  class C < B
+    undef m
+    begin
+      remove_method :m
+    rescue NameError
+    end
+  end
+  begin
+    C.new.m
+  rescue NameError
+    :ok
+  end
+}, '[ruby-dev:31816], [ruby-dev:31817]'
+
+assert_normal_exit %q{
+  begin
+    Process.setrlimit(Process::RLIMIT_STACK, 4_206_592)
+    # FreeBSD SEGVs this less than 4M + 12K bytes.
+  rescue Exception
+    exit
+  end
+  class C
+    attr "a" * (10*1024*1024)
+  end
+}, '[ruby-dev:31818]'
+
+assert_equal 'ok', %q{
+  class Module
+    def define_method2(name, &block)
+      define_method(name, &block)
+    end
+  end
+  class C
+    define_method2(:m) {|x, y| :fail }
+  end
+  begin
+    C.new.m([1,2])
+  rescue ArgumentError
+    :ok
+  end
+}
+
+assert_not_match /method_missing/, %q{
+  STDERR.reopen(STDOUT)
+  variable_or_mehtod_not_exist
+}
+
+assert_equal '[false, false, false, false, true, true]', %q{
+  class C
+    define_method(:foo) {
+      block_given?
+    }
+  end
+
+  C.new.foo {}
+
+  class D
+    def foo
+      D.module_eval{
+        define_method(:m1){
+          block_given?
+        }
+      }
+    end
+    def bar
+      D.module_eval{
+        define_method(:m2){
+          block_given?
+        }
+      }
+    end
+  end
+
+  D.new.foo
+  D.new.bar{}
+  [C.new.foo, C.new.foo{}, D.new.m1, D.new.m1{}, D.new.m2, D.new.m2{}]
+}, '[ruby-core:14813]'
+
+assert_equal 'ok', %q{
+  class Foo
+    define_method(:foo) do |&b|
+      b.call
+    end
+  end
+  Foo.new.foo do
+    break :ok
+  end
+}, '[ruby-dev:36028]'
+
+assert_equal '[1, 2, [3, 4]]', %q{
+  def regular(a, b, *c)
+    [a, b, c]
+  end
+  regular(*[], 1, *[], *[2, 3], *[], 4)
+}, '[ruby-core:19413]'
+
+assert_equal '[1, [:foo, 3, 4, :foo]]', %q{
+  def regular(a, *b)
+    [a, b]
+  end
+  a = b = [:foo]
+  regular(1, *a, *[3, 4], *b)
+}
+
+assert_equal '["B", "A"]', %q{
+  class A
+    def m
+      'A'
+    end
+  end
+
+  class B < A
+    define_method(:m) do
+      ['B', super()]
+    end
+  end
+
+  class C < B
+  end
+
+  C.new.m
+}
+
+assert_equal 'ok', %q{
+  module Foo
+    def foo
+      begin
+        super
+      rescue NoMethodError
+        :ok
+      end
+    end
+    module_function :foo
+  end
+  Foo.foo
+}, '[ruby-dev:37587]'
+
+assert_equal 'Object#foo', %q{
+  class Object
+    def self.foo
+      "Object.foo"
+    end
+    def foo
+      "Object#foo"
+    end
+  end
+
+  module Foo
+    def foo
+      begin
+        super
+      rescue NoMethodError
+        :ok
+      end
+    end
+    module_function :foo
+  end
+  Foo.foo
+}, '[ruby-dev:37587]'
+
+assert_normal_exit %q{
+  class BasicObject
+    remove_method :method_missing
+  end
+  begin
+    "a".lalala!
+  rescue NoMethodError => e
+    e.message == "undefined method `lalala!' for \"a\":String" ? :ok : :ng
+  end
+}, '[ruby-core:22298]'
+
+assert_equal 'ok', %q{
+  "hello"[0] ||= "H"
+  "ok"
+}
+
+assert_equal 'ok', %q{
+  "hello"[0, 1] ||= "H"
+  "ok"
+}
+
+assert_equal 'ok', %q{
+  class C
+    define_method(:foo) do
+      C.class_eval { remove_method(:foo) }
+      super()
+    end
+  end
+  begin
+    C.new.foo
+  rescue NoMethodError
+    'ok'
+  end
+}
+assert_equal 'ok', %q{
+  [0][0, &proc{}] += 21
+  'ok'
+}, '[ruby-core:30534]'
+
+# should not cache when splat
+assert_equal 'ok', %q{
+  class C
+    attr_reader :a
+    def initialize
+      @a = 1
+    end
+  end
+
+  def m *args
+    C.new.a(*args)
+  end
+
+  m()
+  begin
+    m(1)
+  rescue ArgumentError
+    'ok'
+  end
+}
+
+assert_equal 'DC', %q{
+  $result = []
+
+  class C
+    def foo *args
+      $result << 'C'
+    end
+  end
+  class D
+    def foo *args
+      $result << 'D'
+    end
+  end
+
+  o1 = $o1 = C.new
+  o2 = $o2 = D.new
+
+  args = Object.new
+  def args.to_a
+    test1 $o2, nil
+    []
+  end
+  def test1 o, args
+    o.foo(*args)
+  end
+  test1 o1, args
+  $result.join
+}
+
+assert_equal 'DC', %q{
+  $result = []
+
+  class C
+    def foo *args
+      $result << 'C'
+    end
+  end
+  class D
+    def foo *args
+      $result << 'D'
+    end
+  end
+
+  o1 = $o1 = C.new
+  o2 = $o2 = D.new
+
+  block = Object.new
+  def block.to_proc
+    test2 $o2, %w(a, b, c), nil
+    Proc.new{}
+  end
+  def test2 o, args, block
+    o.foo(*args, &block)
+  end
+  test2 o1, [], block
+  $result.join
+}
diff --git a/bootstraptest/test_objectspace.rb b/bootstraptest/test_objectspace.rb
new file mode 100644
index 0000000000..24a1a0ce2c
--- /dev/null
+++ b/bootstraptest/test_objectspace.rb
@@ -0,0 +1,46 @@
+assert_normal_exit %q{
+  eval("", TOPLEVEL_BINDING)
+  minobj = ObjectSpace.to_enum(:each_object).min_by {|a| a.object_id }
+  maxobj = ObjectSpace.to_enum(:each_object).max_by {|a| a.object_id }
+  (((minobj.object_id-100)..(minobj.object_id+100))+
+   ((maxobj.object_id-100)..(maxobj.object_id+100))).each {|id|
+    begin
+      o = ObjectSpace._id2ref(id)
+    rescue RangeError
+      next
+    end
+    o.inspect if defined?(o.inspect)
+  }
+}, '[ruby-dev:31911]'
+
+assert_normal_exit %q{
+  ary = (1..10).to_a
+  ary.permutation(2) {|x|
+    if x == [1,2]
+      ObjectSpace.each_object(String) {|s|
+        s.clear if !s.frozen? && (s.length == 40 || s.length == 80)
+      }
+    end
+  }
+}, '[ruby-dev:31982]'
+
+assert_normal_exit %q{
+  ary = (1..100).to_a
+  ary.permutation(2) {|x|
+    if x == [1,2]
+      ObjectSpace.each_object(Array) {|o| o.clear if o == ary && o.object_id != ary.object_id }
+    end
+  }
+}, '[ruby-dev:31985]'
+
+assert_normal_exit %q{
+  ObjectSpace.define_finalizer("") do
+    Thread::Mutex.new.lock
+  end
+}, '[ruby-dev:44049]'
+
+assert_normal_exit %q{
+  ObjectSpace.define_finalizer("") do
+    Thread.new {}
+  end
+}, '[ruby-core:37858]'
diff --git a/bootstraptest/test_proc.rb b/bootstraptest/test_proc.rb
new file mode 100644
index 0000000000..1e384411dc
--- /dev/null
+++ b/bootstraptest/test_proc.rb
@@ -0,0 +1,483 @@
+assert_equal %q{[1, 2, 3]}, %q{
+  def getproc &b
+    b
+  end
+
+  def m
+    yield
+  end
+
+  m{
+    i = 1
+    m{
+      j = 2
+      m{
+        k = 3
+        getproc{
+          [i, j, k]
+        }
+      }
+    }
+  }.call
+}
+assert_equal %q{7}, %q{
+  def make_proc(&b)
+    b
+  end
+
+  def make_closure
+    a = 0
+    make_proc{
+      a+=1
+    }
+  end
+
+  cl = make_closure
+  cl.call + cl.call * cl.call
+}
+assert_equal %q{ok}, %q{
+  class C
+    def foo
+      :ok
+    end
+  end
+
+  def block
+    C.method(:new).to_proc
+  end
+  b = block()
+  b.call.foo
+}
+assert_equal %q{[0, 1, :last, 0, 2, :last]}, %q{
+  def proc &b
+    b
+  end
+
+  pr = []
+  proc{|i_b|
+    p3 = proc{|j_b|
+      pr << proc{|k_b|
+        [i_b, j_b, k_b]
+      }
+    }
+    p3.call(1)
+    p3.call(2)
+  }.call(0)
+
+  pr[0].call(:last).concat pr[1].call(:last)
+}
+assert_equal %q{12}, %q{
+  def iter
+    yield
+  end
+
+  def getproc &b
+    b
+  end
+
+  iter{
+    bvar = 3
+    getproc{
+      bvar2 = 4
+      bvar * bvar2
+    }
+  }.call
+}
+assert_equal %q{200}, %q{
+  def iter
+    yield
+  end
+
+  def getproc &b
+    b
+  end
+
+  loc1 = 0
+  pr1 = iter{
+    bl1 = 1
+    getproc{
+      loc1 += 1
+      bl1  += 1
+      loc1 + bl1
+    }
+  }
+
+  pr2 = iter{
+    bl1 = 1
+    getproc{
+      loc1 += 1
+      bl1  += 1
+      loc1 + bl1
+    }
+  }
+
+  pr1.call; pr2.call
+  pr1.call; pr2.call
+  pr1.call; pr2.call
+  (pr1.call + pr2.call) * loc1
+}
+assert_equal %q{[1, 2]}, %q{
+  def proc(&pr)
+    pr
+  end
+
+  def m
+    a = 1
+    m2{
+      a
+    }
+  end
+
+  def m2
+    b = 2
+    proc{
+      [yield, b]
+    }
+  end
+
+  pr = m
+  x = ['a', 1,2,3,4,5,6,7,8,9,0,
+            1,2,3,4,5,6,7,8,9,0,
+            1,2,3,4,5,6,7,8,9,0,
+            1,2,3,4,5,6,7,8,9,0,
+            1,2,3,4,5,6,7,8,9,0,]
+  pr.call
+}
+assert_equal %q{1}, %q{
+  def proc(&pr)
+    pr
+  end
+
+  def m
+    a = 1
+    m2{
+      a
+    }
+  end
+
+  def m2
+    b = 2
+    proc{
+      [yield, b]
+    }
+    100000.times{|x|
+      "#{x}"
+    }
+    yield
+  end
+  m
+}
+assert_equal %q{[:C, :C]}, %q{
+  Const = :top
+  class C
+    Const = :C
+    $pr = proc{
+      (1..2).map{
+        Const
+      }
+    }
+  end
+  $pr.call
+}
+assert_equal %q{top}, %q{
+  Const = :top
+  class C
+    Const = :C
+  end
+  pr = proc{
+    Const
+  }
+  C.class_eval %q{
+    pr.call
+  }
+}
+assert_equal %q{1}, %q{
+  def m(&b)
+    b
+  end
+
+  m{|e_proctest| e_proctest}.call(1)
+}
+assert_equal %q{12}, %q{
+  def m(&b)
+    b
+  end
+
+  m{|e_proctest1, e_proctest2|
+    a = e_proctest1 * e_proctest2 * 2
+    a * 3
+  }.call(1, 2)
+}
+assert_equal %q{[[], [1], [1, 2], [1, 2, 3]]}, %q{
+  [
+  Proc.new{|*args| args}.call(),
+  Proc.new{|*args| args}.call(1),
+  Proc.new{|*args| args}.call(1, 2),
+  Proc.new{|*args| args}.call(1, 2, 3),
+  ]
+}
+assert_equal %q{[[nil, []], [1, []], [1, [2]], [1, [2, 3]]]}, %q{
+  [
+  Proc.new{|a, *b| [a, b]}.call(),
+  Proc.new{|a, *b| [a, b]}.call(1),
+  Proc.new{|a, *b| [a, b]}.call(1, 2),
+  Proc.new{|a, *b| [a, b]}.call(1, 2, 3),
+  ]
+}
+assert_equal %q{1}, %q{
+  pr = proc{
+    $SAFE
+  }
+  $SAFE = 1
+  pr.call
+}
+assert_equal %q{[1, 1]}, %q{
+  pr = proc{
+    $SAFE += 1
+  }
+  [pr.call, $SAFE]
+}
+assert_equal %q{1}, %q{
+  def m(&b)
+    b
+  end
+  m{1}.call
+}
+assert_equal %q{3}, %q{
+  def m(&b)
+    b
+  end
+
+  m{
+    a = 1
+    a + 2
+  }.call
+}
+assert_equal %Q{ok\n}, %q{
+  class A; def get_block; proc {puts "ok"} end end
+  block = A.new.get_block
+  GC.start
+  block.call
+}, '[ruby-core:14885]'
+
+assert_equal 'ok', %q{
+  a = lambda {|x, y, &b| b }
+  b = a.curry[1]
+  if b.call(2){} == nil
+    :ng
+  else
+    :ok
+  end
+}, '[ruby-core:15551]'
+
+assert_equal 'ok', %q{
+  lambda {
+    break :ok
+    :ng
+  }.call
+}, '[ruby-dev:34646]'
+
+assert_equal %q{[:bar, :foo]}, %q{
+  def foo
+    klass = Class.new do
+      define_method(:bar) do
+        return :bar
+      end
+    end
+    [klass.new.bar, :foo]
+  end
+  foo
+}, "[ ruby-Bugs-19304 ]"
+
+assert_equal 'ok', %q{
+   $x = :ok
+   def def7(x, y)
+      x[y]
+      $x = :ng
+   end
+   def test_def7
+      def7(lambda {|x| x.call}, Proc.new {return})
+      $x = :ng
+   end
+   test_def7
+   $x
+}, '[ruby-core:17164]'
+
+assert_equal 'ok', %q{
+  lambda { a = lambda { return }; $x = :ng; a[]; $x = :ok }.call
+  $x
+}, '[ruby-core:17164]'
+
+assert_equal 'ok', %q{
+  lambda { a = lambda { break }; $x = :ng; a[]; $x = :ok }.call
+  $x
+}, '[ruby-core:17164]'
+
+assert_equal 'ok', %q{
+  def def8
+    $x = :ng
+    lambda { a = Proc.new { return }; a[]}.call
+    $x = :ok
+  end
+  def8
+  $x
+}, '[ruby-core:17164]'
+
+
+assert_equal 'ok', %q{
+   def def9
+      lambda {|a| $x = :ok; a[]; $x = :ng }.call(Proc.new { return })
+      $x = :ng
+   end
+   def9
+   $x
+}, '[ruby-core:17164]'
+
+assert_equal 'ok', %q{
+   def def10
+     $x = :ng
+     lambda { 1.times { return } }.call
+     $x = :ok
+   end
+   $x = :ok
+   def10
+   $x
+}, '[ruby-core:17164]'
+
+assert_equal 'ok', %q{
+   def def11
+      yield
+   end
+   begin
+      lambda { def11 { return } }.call
+   rescue LocalJumpError
+      :ng
+   else
+      :ok
+   end
+}, '[ruby-core:17164]'
+
+assert_equal 'ok', %q{
+   def def12
+      b = Proc.new { $x = :ng; lambda { return }.call; $x = :ok }.call
+   end
+   def12
+   $x
+}, '[ruby-core:17164]'
+
+assert_equal 'ok', %q{
+  def m
+    pr = proc{
+      proc{
+        return :ok
+      }
+    }.call
+    pr.call
+    :ng
+  end
+  m()
+}
+
+assert_equal 'ok', %q{
+  class Foo
+    def call_it
+      p = Proc.new
+      p.call
+    end
+  end
+
+  def give_it
+    proc { :ok }
+  end
+
+  f = Foo.new
+  a_proc = give_it
+  f.call_it(&give_it())
+}, '[ruby-core:15711]'
+
+assert_equal 'foo!', %q{
+  class FooProc < Proc
+    def initialize
+      @foo = "foo!"
+    end
+
+    def bar
+      @foo
+    end
+  end
+
+  def bar
+    FooProc.new &lambda{
+      p 1
+    }
+  end
+
+  fp = bar(&lambda{
+    p 2
+  })
+
+  fp.bar
+}, 'Subclass of Proc'
+
+assert_equal 'ok', %q{
+  o = Object.new
+  def o.write(s); end
+  $stderr = o
+  at_exit{
+    print $!.message
+  }
+  raise "ok"
+}
+
+assert_equal 'ok', %q{
+  lambda do
+    class A
+      class B
+        proc{return :ng}.call
+      end
+    end
+  end.call
+  :ok
+}
+
+assert_equal 'ok', %q{
+  $proc = proc{return}
+  begin
+    lambda do
+      class A
+        class B
+          $proc.call
+        end
+      end
+    end.call
+    :ng
+  rescue LocalJumpError
+    :ok
+  end
+}
+
+assert_equal 'ok', %q{
+  def x
+    binding
+  end
+  b = x{|a| a }
+  b.eval('yield("ok")')
+}, '[Bug #5634]'
+
+assert_equal 'ok', %q{
+  def x
+    binding
+  end
+  eval("x { 'ok' }").eval "yield"
+}, '[Bug #5634]'
+
+assert_equal 'ok', %q{
+  def x
+    binding
+  end
+  def m
+    x{ 'ok' }
+  end
+  eval('yield', m)
+}, '[Bug #5634]'
+
diff --git a/bootstraptest/test_string.rb b/bootstraptest/test_string.rb
new file mode 100644
index 0000000000..849dcd45b0
--- /dev/null
+++ b/bootstraptest/test_string.rb
@@ -0,0 +1,3 @@
+assert_normal_exit %q{
+  inspect.clear
+}, '[ruby-core:68110]'
diff --git a/bootstraptest/test_struct.rb b/bootstraptest/test_struct.rb
new file mode 100644
index 0000000000..a65964d5f9
--- /dev/null
+++ b/bootstraptest/test_struct.rb
@@ -0,0 +1,5 @@
+assert_equal 'Struct::Foo', %q{
+  Struct.instance_eval { const_set(:Foo, nil) }
+  Struct.new("Foo")
+  Struct::Foo
+}
diff --git a/bootstraptest/test_syntax.rb b/bootstraptest/test_syntax.rb
new file mode 100644
index 0000000000..a111990a1f
--- /dev/null
+++ b/bootstraptest/test_syntax.rb
@@ -0,0 +1,904 @@
+assert_equal %q{4}, %q{1 && 2 && 3 && 4}
+assert_equal %q{}, %q{1 && nil && 3 && 4}
+assert_equal %q{}, %q{1 && 2 && 3 && nil}
+assert_equal %q{false}, %q{1 && 2 && 3 && false}
+assert_equal %q{4}, %q{1 and 2 and 3 and 4}
+assert_equal %q{}, %q{1 and nil and 3 and 4}
+assert_equal %q{}, %q{1 and 2 and 3 and nil}
+assert_equal %q{false}, %q{1 and 2 and 3 and false}
+assert_equal %q{}, %q{nil && true}
+assert_equal %q{false}, %q{false && true}
+assert_equal %q{}, %q{
+  case 1
+  when 2
+    :ng
+  end}
+assert_equal %q{ok}, %q{
+  case 1
+  when 10,20,30
+    :ng1
+  when 1,2,3
+    :ok
+  when 100,200,300
+    :ng2
+  else
+    :elseng
+  end}
+assert_equal %q{elseok}, %q{
+  case 123
+  when 10,20,30
+    :ng1
+  when 1,2,3
+    :ng2
+  when 100,200,300
+    :ng3
+  else
+    :elseok
+  end
+}
+assert_equal %q{ok}, %q{
+  case 'test'
+  when /testx/
+    :ng1
+  when /test/
+    :ok
+  when /tetxx/
+    :ng2
+  else
+    :ng_else
+  end
+}
+assert_equal %q{ok}, %q{
+  case Object.new
+  when Object
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  case Object
+  when Object.new
+    :ng
+  else
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  case 'test'
+  when 'tes'
+    :ng
+  when 'te'
+    :ng
+  else
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  case 'test'
+  when 'tes'
+    :ng
+  when 'te'
+    :ng
+  when 'test'
+    :ok
+  end
+}
+assert_equal %q{ng}, %q{
+  case 'test'
+  when 'tes'
+    :ng
+  when /te/
+    :ng
+  else
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  case 'test'
+  when 'tes'
+    :ng
+  when /test/
+    :ok
+  else
+    :ng
+  end
+}
+assert_equal %q{100}, %q{
+  def test(arg)
+    case 1
+    when 2
+      3
+    end
+    return arg
+  end
+
+  test(100)
+}
+assert_equal %q{ok}, %q{
+  ary = [1, 2]
+  case 1
+  when *ary
+    :ok
+  else
+    :ng
+  end
+}
+assert_equal %q{ok}, %q{
+  ary = [1, 2]
+  case 3
+  when *ary
+    :ng
+  else
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  ary = [1, 2]
+  case 1
+  when :x, *ary
+    :ok
+  when :z
+    :ng1
+  else
+    :ng2
+  end
+}
+assert_equal %q{ok}, %q{
+  ary = [1, 2]
+  case 3
+  when :x, *ary
+    :ng1
+  when :z
+    :ng2
+  else
+    :ok
+  end
+}
+assert_equal %q{[:false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :false, :false, :false, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :then, :false, :then, :then, :then, :false, :false, :false, :false, :false, :false, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :false, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :false, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :false, :then, :then, :then, :then, :then, :then, :then, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep, :then, :sep]}, %q{
+
+  def make_perm ary, num
+    if num == 1
+      ary.map{|e| [e]}
+    else
+      base = make_perm(ary, num-1)
+      res  = []
+      base.each{|b|
+        ary.each{|e|
+          res << [e] + b
+        }
+      }
+      res
+    end
+  end
+
+  def each_test
+    conds = make_perm(['fv', 'tv'], 3)
+    bangs = make_perm(['', '!'], 3)
+    exprs = make_perm(['and', 'or'], 3)
+    ['if', 'unless'].each{|syn|
+      conds.each{|cs|
+        bangs.each{|bs|
+          exprs.each{|es|
+            yield(syn, cs, bs, es)
+          }
+        }
+      }
+    }
+  end
+
+  fv = false
+  tv = true
+
+  $ans = []
+  each_test{|syn, conds, bangs, exprs|
+    c1, c2, c3 = conds
+    bang1, bang2, bang3 = bangs
+    e1, e2 = exprs
+    eval %Q{
+      #{syn} #{bang1}#{c1} #{e1} #{bang2}#{c2} #{e2} #{bang3}#{c3}
+        $ans << :then
+      else
+        $ans << :false
+      end
+    }
+  }
+
+  each_test{|syn, conds, bangs, exprs|
+    c1, c2, c3 = conds
+    bang1, bang2, bang3 = bangs
+    e1, e2 = exprs
+    eval %Q{
+      #{syn} #{bang1}#{c1} #{e1} #{bang2}#{c2} #{e2} #{bang3}#{c3}
+        $ans << :then
+      end
+      $ans << :sep
+    }
+  }
+  $ans
+}
+assert_equal %q{}, %q{
+  defined?(m)
+}
+assert_equal %q{method}, %q{
+  def m
+  end
+  defined?(m)
+}
+assert_equal %q{}, %q{
+  defined?(a.class)
+}
+assert_equal %q{method}, %q{
+  a = 1
+  defined?(a.class)
+}
+assert_equal %q{["method", "method", "method", "method", nil, nil, "method", "method", "method", nil]}, %q{
+  class C
+    def test
+      [defined?(m1()), defined?(self.m1), defined?(C.new.m1),
+       defined?(m2()), defined?(self.m2), defined?(C.new.m2),
+       defined?(m3()), defined?(self.m3), defined?(C.new.m3)]
+    end
+    def m1
+    end
+    private
+    def m2
+    end
+    protected
+    def m3
+    end
+  end
+  C.new.test + [defined?(C.new.m3)]
+}
+assert_equal %q{[nil, nil, nil, nil, "global-variable", "global-variable", nil, nil]}, %q{
+  $ans = [defined?($1), defined?($2), defined?($3), defined?($4)]
+  /(a)(b)/ =~ 'ab'
+  $ans + [defined?($1), defined?($2), defined?($3), defined?($4)]
+}
+assert_equal %q{nilselftruefalse}, %q{
+  defined?(nil) + defined?(self) +
+    defined?(true) + defined?(false)
+}
+assert_equal %q{}, %q{
+  defined?(@a)
+}
+assert_equal %q{instance-variable}, %q{
+  @a = 1
+  defined?(@a)
+}
+assert_equal %q{}, %q{
+  defined?(@@a)
+}
+assert_equal %q{class variable}, %q{
+  @@a = 1
+  defined?(@@a)
+}
+assert_equal %q{}, %q{
+  defined?($a)
+}
+assert_equal %q{global-variable}, %q{
+  $a = 1
+  defined?($a)
+}
+assert_equal %q{}, %q{
+  defined?(C_definedtest)
+}
+assert_equal %q{constant}, %q{
+  C_definedtest = 1
+  defined?(C_definedtest)
+}
+assert_equal %q{}, %q{
+  defined?(::C_definedtest)
+}
+assert_equal %q{constant}, %q{
+  C_definedtest = 1
+  defined?(::C_definedtest)
+}
+assert_equal %q{}, %q{
+  defined?(C_definedtestA::C_definedtestB::C_definedtestC)
+}
+assert_equal %q{constant}, %q{
+  class C_definedtestA
+    class C_definedtestB
+      C_definedtestC = 1
+    end
+  end
+  defined?(C_definedtestA::C_definedtestB::C_definedtestC)
+}
+assert_equal %q{30}, %q{
+  sum = 0
+  30.times{|ib|
+    if ib % 10 == 0 .. true
+      sum += ib
+    end
+  }
+  sum
+}
+assert_equal %q{63}, %q{
+  sum = 0
+  30.times{|ib|
+    if ib % 10 == 0 ... true
+      sum += ib
+    end
+  }
+  sum
+}
+assert_equal %q{[["NUM", "Type: NUM\n"], ["NUM", "123\n"], ["NUM", "456\n"], ["NUM", "Type: ARP\n"], ["NUM", "aaa\n"], ["NUM", "bbb\n"], ["NUM", "\f\n"], ["ARP", "Type: ARP\n"], ["ARP", "aaa\n"], ["ARP", "bbb\n"]]}, %q{
+  t = nil
+  unless ''.respond_to? :lines
+    class String
+      def lines
+        self
+      end
+    end
+  end
+  ary = []
+"this must not print
+Type: NUM
+123
+456
+Type: ARP
+aaa
+bbb
+\f
+this must not print
+hoge
+Type: ARP
+aaa
+bbb
+".lines.each{|l|
+    if (t = l[/^Type: (.*)/, 1])..(/^\f/ =~ l)
+      ary << [t, l]
+    end
+  }
+  ary
+}
+assert_equal %q{1}, %q{if true  then 1 ; end}
+assert_equal %q{}, %q{if false then 1 ; end}
+assert_equal %q{1}, %q{if true  then 1 ; else; 2; end}
+assert_equal %q{2}, %q{if false then 1 ; else; 2; end}
+assert_equal %q{}, %q{if true  then   ; elsif true then ; 1 ; end}
+assert_equal %q{1}, %q{if false then   ; elsif true then ; 1 ; end}
+assert_equal %q{}, %q{unless true  then 1 ; end}
+assert_equal %q{1}, %q{unless false then 1 ; end}
+assert_equal %q{2}, %q{unless true  then 1 ; else; 2; end}
+assert_equal %q{1}, %q{unless false then 1 ; else; 2; end}
+assert_equal %q{1}, %q{1 if true}
+assert_equal %q{}, %q{1 if false}
+assert_equal %q{}, %q{1 if nil}
+assert_equal %q{}, %q{1 unless true}
+assert_equal %q{1}, %q{1 unless false}
+assert_equal %q{1}, %q{1 unless nil}
+assert_equal %q{1}, %q{1 || 2 || 3 || 4}
+assert_equal %q{1}, %q{1 || false || 3 || 4}
+assert_equal %q{2}, %q{nil || 2 || 3 || 4}
+assert_equal %q{2}, %q{false || 2 || 3 || 4}
+assert_equal %q{false}, %q{nil || false || nil || false}
+assert_equal %q{1}, %q{1 or 2 or 3 or 4}
+assert_equal %q{1}, %q{1 or false or 3 or 4}
+assert_equal %q{2}, %q{nil or 2 or 3 or 4}
+assert_equal %q{2}, %q{false or 2 or 3 or 4}
+assert_equal %q{1}, %q{if true && ""; then 1; end}
+assert_equal %q{1}, %q{if nil || true; then 1; end}
+assert_equal %q{false}, %q{nil or false or nil or false}
+assert_equal %q{elseng}, %q{
+  case
+  when 1==2, 2==3
+    :ng1
+  when false, 4==5
+    :ok
+  when false
+    :ng2
+  else
+    :elseng
+  end
+}
+assert_equal %q{ok}, %q{
+  case
+  when nil, nil
+    :ng1
+  when 1,2,3
+    :ok
+  when false, false
+    :ng2
+  else
+    :elseng
+  end
+}
+assert_equal %q{elseok}, %q{
+  case
+  when nil
+    :ng1
+  when false
+    :ng2
+  else
+    :elseok
+  end}
+assert_equal %q{}, %q{
+  case
+  when 1
+  end
+}
+assert_equal %q{ok}, %q{
+  r = nil
+  ary = []
+  case
+  when false
+    r = :ng1
+  when false, false
+    r = :ng2
+  when *ary
+    r = :ng3
+  when false, *ary
+    r = :ng4
+  when true, *ary
+    r = :ok
+  end
+  r
+}
+assert_equal %q{ok}, %q{
+  ary = []
+  case
+  when false, *ary
+    :ng
+  else
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  ary = [false, nil]
+  case
+  when *ary
+    :ng
+  else
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  ary = [false, nil]
+  case
+  when *ary
+    :ng
+  when true
+    :ok
+  else
+    :ng2
+  end
+}
+assert_equal %q{ng}, %q{
+  ary = [false, nil]
+  case
+  when *ary
+    :ok
+  else
+    :ng
+  end
+}
+assert_equal %q{ok}, %q{
+  ary = [false, true]
+  case
+  when *ary
+    :ok
+  else
+    :ng
+  end
+}
+assert_equal %q{ok}, %q{
+  ary = [false, true]
+  case
+  when false, false
+  when false, *ary
+    :ok
+  else
+    :ng
+  end
+}
+assert_equal %q{}, %q{
+  i = 0
+  while i < 10
+    i+=1
+  end}
+assert_equal %q{10}, %q{
+  i = 0
+  while i < 10
+    i+=1
+  end; i}
+assert_equal %q{}, %q{
+  i = 0
+  until i > 10
+    i+=1
+  end}
+assert_equal %q{11}, %q{
+  i = 0
+  until i > 10
+    i+=1
+  end; i}
+assert_equal %q{1}, %q{
+  i = 0
+  begin
+    i+=1
+  end while false
+  i
+}
+assert_equal %q{1}, %q{
+  i = 0
+  begin
+    i+=1
+  end until true
+  i
+}
+def assert_syntax_error expected, code, message = ''
+  assert_equal "#{expected}",
+    "begin eval(%q{#{code}}, nil, '', 0)"'; rescue SyntaxError => e; e.message[/\A:(?:\d+:)? (.*)/, 1] end', message
+end
+assert_syntax_error "unterminated string meets end of file", '().."', '[ruby-dev:29732]'
+assert_equal %q{[]}, %q{$&;[]}, '[ruby-dev:31068]'
+assert_syntax_error "syntax error, unexpected *, expecting '}'", %q{{*0}}, '[ruby-dev:31072]'
+assert_syntax_error "`@0' is not allowed as an instance variable name", %q{@0..0}, '[ruby-dev:31095]'
+assert_syntax_error "identifier $00 is not valid to get", %q{$00..0}, '[ruby-dev:31100]'
+assert_syntax_error "identifier $00 is not valid to set", %q{0..$00=1}
+assert_equal %q{0}, %q{[*0];0}, '[ruby-dev:31102]'
+assert_syntax_error "syntax error, unexpected ')'", %q{v0,(*,v1,) = 0}, '[ruby-dev:31104]'
+assert_equal %q{1}, %q{
+  class << (ary=[]); def []; 0; end; def []=(x); super(0,x);end;end; ary[]+=1
+}, '[ruby-dev:31110]'
+assert_syntax_error "Can't set variable $1", %q{0..$1=1}, '[ruby-dev:31118]'
+assert_valid_syntax %q{1.times{1+(1&&next)}}, '[ruby-dev:31119]'
+assert_valid_syntax %q{x=-1;loop{x+=1&&redo if (x+=1).zero?}}, '[ruby-dev:31119]'
+assert_syntax_error %q{syntax error, unexpected end-of-input}, %q{!}, '[ruby-dev:31243]'
+assert_equal %q{[nil]}, %q{[()]}, '[ruby-dev:31252]'
+assert_equal %q{true}, %q{!_=()}, '[ruby-dev:31263]'
+assert_equal 'ok', %q{while true; redo; end if 1 == 2; :ok}, '[ruby-dev:31360]'
+assert_equal 'ok', %q{
+  1.times {
+    begin
+    ensure
+      next
+    end
+  }; :ok
+}, '[ruby-dev:31373]'
+assert_equal 'ok', %q{
+  flag = false
+  1.times {
+    next if flag
+    flag = true
+    begin
+    ensure
+      redo
+    end
+  }; :ok
+}, '[ruby-dev:31373]'
+
+assert_equal 'ok', %q{
+  1.times{
+    p(1, (next; 2))
+  }; :ok
+}
+assert_equal '3', %q{
+  i = 0
+  1 + (while true
+         break 2 if (i+=1) > 1
+         next
+       end)
+}
+assert_equal '3', %q{
+  i = 0
+  1 + (while true
+         break 2 if (i+=1) > 1
+         p(1, (next; 2))
+       end)
+}
+# redo
+assert_equal 'ok', %q{
+  i = 0
+  1.times{
+    break if i>1
+    i+=1
+    p(1, (redo; 2))
+  }; :ok
+}
+assert_equal '3', %q{
+  i = 0
+  1 + (while true
+         break 2 if (i+=1) > 1
+         redo
+       end)
+}
+assert_equal '3', %q{
+  i = 0
+  1 + (while true
+         break 2 if (i+=1) > 1
+         p(1, (redo; 2))
+       end)
+}
+assert_equal '1', %q{
+  a = [0]
+  a[*a]+=1
+}
+assert_equal '2', %q{
+  ary = [0]
+  case 1
+  when *ary, 1
+    1
+  end +
+  case
+  when *ary
+    1
+  end
+}
+
+assert_match /invalid multibyte char/, %q{
+  STDERR.reopen(STDOUT)
+  eval("\"\xf0".force_encoding("utf-8"))
+}, '[ruby-dev:32429]'
+
+# method ! and !=
+assert_equal 'true', %q{!false}
+assert_equal 'true', %q{1 == 1}
+assert_equal 'true', %q{1 != 2}
+assert_equal 'true', %q{
+  class C; def !=(obj); true; end; end
+  C.new != 1
+}
+assert_equal 'true', %q{
+  class C; def !@; true; end; end
+  !C.new
+}
+assert_normal_exit %q{
+  eval "while true; return; end rescue p $!"
+}, '[ruby-dev:31663]'
+assert_equal '1', %q{
+  def bar
+    raise
+  end
+
+  def foo
+    1.times{
+      begin
+        return bar
+      rescue
+        :ok
+      end
+    }
+  end
+
+  foo
+}
+
+assert_equal 'ok', %q{
+  counter = 2
+  while true
+    counter -= 1
+    next if counter != 0
+    break
+  end
+  :ok
+}, '[ruby-core:14385]'
+
+assert_equal 'ok', %q{
+  counter = 2
+  while true
+    counter -= 1
+    next if counter != 0
+    break :ok
+  end # direct
+}, '[ruby-core:14385]'
+
+assert_equal 'ok', %q{
+  counter = 2
+  while true
+    counter -= 1
+    break if counter == 0
+    "#{next}"
+  end
+  :ok
+}, 'reported by Yusuke ENDOH'
+
+assert_equal 'ok', %q{
+  counter = 2
+  while true
+    counter -= 1
+    break if counter == 0
+    next
+    redo
+  end
+  :ok
+}, 'reported by Yusuke ENDOH'
+
+assert_equal 'ok', %q{
+  counter = 2
+  while true
+    counter -= 1
+    break if counter == 0
+    next
+    "#{ redo }"
+  end
+  :ok
+}, 'reported by Yusuke ENDOH'
+
+assert_normal_exit %q{
+  begin
+    raise
+  rescue
+    counter = 2
+    while true
+      counter -= 1
+      break if counter == 0
+      next
+      retry
+    end
+  end
+}, 'reported by Yusuke ENDOH'
+
+assert_normal_exit %q{
+  counter = 2
+  while true
+    counter -= 1
+    break if counter == 0
+    next
+    "#{ break }"
+  end
+}, 'reported by Yusuke ENDOH'
+
+assert_normal_exit %q{
+  counter = 2
+  while true
+    counter -= 1
+    next if counter != 0
+    "#{ break }"
+  end
+}, 'reported by Yusuke ENDOH'
+
+assert_equal 'ok', %q{
+  1.times do
+    [
+      1, 2, 3, 4, 5, 6, 7, 8,
+      begin
+        false ? next : p
+        break while true
+      end
+    ]
+  end
+  :ok
+}, '[ruby-dev:32882]'
+
+assert_equal "1\n2\n", %q{
+  i = 0
+  while i<2
+    i += 1
+    next p(i)
+  end
+}
+
+assert_valid_syntax('1.times {|i|print (42),1;}', '[ruby-list:44479]')
+
+assert_equal 'ok', %q{
+  def a() end
+  begin
+    if defined?(a(1).a)
+      :ng
+    else
+      :ok
+    end
+  rescue
+    :ng
+  end
+}, '[ruby-core:16010]'
+
+assert_equal 'ok', %q{
+  def a() end
+  begin
+    if defined?(a::B)
+      :ng
+    else
+      :ok
+    end
+  rescue
+    :ng
+  end
+}, '[ruby-core:16010]'
+
+assert_normal_exit %q{
+  defined? C && 0
+}
+
+assert_normal_exit %q{
+  class C
+    def m
+      defined?(super())
+    end
+  end
+  C.new.m
+}
+
+assert_equal 'ok', %q{
+  class X < RuntimeError;end
+  x = [X]
+  begin
+   raise X
+  rescue *x
+   :ok
+  end
+}, '[ruby-core:14537]'
+
+assert_equal 'ok', %q{
+  a = [false]
+  (a[0] &&= true) == false ? :ok : :ng
+}, '[ruby-dev:34679]'
+
+assert_normal_exit %q{
+  a = []
+  100.times {|i| a << i << nil << nil }
+  p a.compact!
+}
+
+assert_equal 'ok', %q{
+  "#{}""#{}ok"
+}, '[ruby-dev:38968]'
+
+
+assert_equal 'ok', %q{
+  "o" "#{}k"
+}, '[ruby-dev:38980]'
+
+bug2415 = '[ruby-core:26961]'
+assert_normal_exit %q{
+  0.times do
+    0.times do
+      def x(a=1, b, *rest); nil end
+    end
+  end
+}, bug2415
+
+assert_normal_exit %q{
+  0.times do
+    0.times do
+      def x@; nil end
+    end
+  end
+}, bug2415
+
+assert_normal_exit %q{
+  0.times do
+    0.times do
+      def x(a = 0.times do
+              def y(a=1, b, *rest); nil; end
+            end)
+        nil
+      end
+    end
+  end
+}, bug2415
+
+assert_normal_exit %q{
+  0.times do
+    0.times do
+      def x(a = 0.times do
+              def x@; nil; end
+            end)
+        nil
+      end
+    end
+  end
+}, bug2415
+
+assert_normal_exit %q{
+  a {
+    b {|c.d| }
+    e
+  }
+}, '[ruby-dev:39861]'
+
+bug1240 = '[ruby-core:22637]'
+assert_valid_syntax('x y { "#{}".z { } }', bug1240)
+assert_valid_syntax('x y { "#{}".z do end }', bug1240)
+
+assert_valid_syntax('y "#{a 1}" do end', '[ruby-core:29579]')
+assert_normal_exit %q{
+  def foo(&block)
+    yield
+  end
+
+  foo do
+    s = defined?(raise + 1)
+    Class
+  end
+}, '[ruby-core:30293]'
diff --git a/bootstraptest/test_thread.rb b/bootstraptest/test_thread.rb
new file mode 100644
index 0000000000..d16295de8b
--- /dev/null
+++ b/bootstraptest/test_thread.rb
@@ -0,0 +1,484 @@
+# Thread and Fiber
+
+assert_equal %q{ok}, %q{
+  Thread.new{
+  }.join
+  :ok
+}
+assert_equal %q{ok}, %q{
+  Thread.new{
+    :ok
+  }.value
+}
+assert_equal %q{ok}, %q{
+begin
+  v = 0
+  (1..200).map{|i|
+    Thread.new{
+      i
+    }
+  }.each{|t|
+    v += t.value
+  }
+  v == 20100 ? :ok : v
+rescue ThreadError => e
+  :ok if /can't create Thread/ =~ e.message
+end
+}
+assert_equal %q{ok}, %q{
+begin
+  :ok if 5000 == 5000.times{|e|
+    (1..2).map{
+      Thread.new{
+      }
+    }.each{|e|
+      e.join()
+    }
+  }
+rescue ThreadError => e
+  :ok if /can't create Thread/ =~ e.message
+end
+}
+assert_equal %q{ok}, %q{
+begin
+  :ok if 5000 == 5000.times{|e|
+    (1..2).map{
+      Thread.new{
+      }
+    }.each{|e|
+      e.join(1000000000)
+    }
+  }
+rescue ThreadError => e
+  :ok if /can't create Thread/ =~ e.message
+end
+}
+assert_equal %q{ok}, %q{
+begin
+  :ok if 5000 == 5000.times{
+    t = Thread.new{}
+    while t.alive?
+      Thread.pass
+    end
+  }
+rescue NoMemoryError
+  :ok
+end
+}
+assert_equal %q{100}, %q{
+  100.times{
+    Thread.new{loop{Thread.pass}}
+  }
+}
+assert_equal %q{ok}, %q{
+  Thread.new{
+    :ok
+  }.join.value
+}
+assert_equal %q{ok}, %q{
+  begin
+    Thread.new{
+      raise "ok"
+    }.join
+  rescue => e
+    e
+  end
+}
+assert_equal %q{ok}, %q{
+  ans = nil
+  t = Thread.new{
+    begin
+      sleep 0.5
+    ensure
+      ans = :ok
+    end
+  }
+  Thread.pass until t.stop?
+  t.kill
+  t.join
+  ans
+}
+assert_equal %q{ok}, %q{
+  t = Thread.new{
+    sleep
+  }
+  sleep 0.1
+  t.raise
+  begin
+    t.join
+    :ng
+  rescue
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  t = Thread.new{
+    loop{}
+  }
+  Thread.pass
+  t.raise
+  begin
+    t.join
+    :ng
+  rescue
+    :ok
+  end
+}
+assert_equal %q{ok}, %q{
+  t = Thread.new{
+  }
+  Thread.pass
+  t.join
+  t.raise # raise to exited thread
+  begin
+    t.join
+    :ok
+  rescue
+    :ng
+  end
+}
+assert_equal %q{run}, %q{
+  t = Thread.new{
+    loop{}
+  }
+  st = t.status
+  t.kill
+  st
+}
+assert_equal %q{sleep}, %q{
+  t = Thread.new{
+    sleep
+  }
+  sleep 0.1
+  st = t.status
+  t.kill
+  st
+}
+assert_equal %q{false}, %q{
+  t = Thread.new{
+  }
+  t.kill
+  sleep 0.1
+  t.status
+}
+assert_equal %q{[ThreadGroup, true]}, %q{
+  ptg = Thread.current.group
+  Thread.new{
+    ctg = Thread.current.group
+    [ctg.class, ctg == ptg]
+  }.value
+}
+assert_equal %q{[1, 1]}, %q{
+  thg = ThreadGroup.new
+
+  t = Thread.new{
+    thg.add Thread.current
+    sleep
+  }
+  sleep 0.1
+  [thg.list.size, ThreadGroup::Default.list.size]
+}
+assert_equal %q{true}, %q{
+  thg = ThreadGroup.new
+
+  t = Thread.new{sleep 5}
+  thg.add t
+  thg.list.include?(t)
+}
+assert_equal %q{[true, nil, true]}, %q{
+  /a/ =~ 'a'
+  $a = $~
+  Thread.new{
+    $b = $~
+    /b/ =~ 'b'
+    $c = $~
+  }.join
+  $d = $~
+  [$a == $d, $b, $c != $d]
+}
+assert_equal %q{11}, %q{
+  Thread.current[:a] = 1
+  Thread.new{
+    Thread.current[:a] = 10
+    Thread.pass
+    Thread.current[:a]
+  }.value + Thread.current[:a]
+}
+assert_normal_exit %q{
+  begin
+    100.times do |i|
+      begin
+        th = Thread.start(Thread.current) {|u| u.raise }
+        raise
+      rescue
+      ensure
+        th.join
+      end
+    end
+  rescue
+  end
+}, '[ruby-dev:31371]'
+
+assert_equal 'true', %{
+  t = Thread.new { loop {} }
+  begin
+    pid = fork {
+      exit t.status != "run"
+    }
+    Process.wait pid
+    $?.success?
+  rescue NotImplementedError
+    true
+  end
+}
+
+assert_equal 'ok', %{
+  open("zzz.rb", "w") do |f|
+    f.puts <<-END
+      begin
+        Thread.new { fork { GC.start } }.join
+        pid, status = Process.wait2
+        $result = status.success? ? :ok : :ng
+      rescue NotImplementedError
+        $result = :ok
+      end
+    END
+  end
+  require "./zzz.rb"
+  $result
+}
+
+assert_finish 3, %{
+  th = Thread.new {sleep 0.2}
+  th.join(0.1)
+  th.join
+}
+
+assert_finish 3, %{
+  require 'timeout'
+  th = Thread.new {sleep 0.2}
+  begin
+    Timeout.timeout(0.1) {th.join}
+  rescue Timeout::Error
+  end
+  th.join
+}
+
+assert_normal_exit %q{
+  STDERR.reopen(STDOUT)
+  exec "/"
+}
+
+assert_normal_exit %q{
+  (0..10).map {
+    Thread.new {
+     10000.times {
+        Object.new.to_s
+      }
+    }
+  }.each {|t|
+    t.join
+  }
+}
+
+assert_equal 'ok', %q{
+  def m
+    t = Thread.new { while true; // =~ "" end }
+    sleep 0.01
+    10.times {
+      if /((ab)*(ab)*)*(b)/ =~ "ab"*7
+        return :ng if !$4
+        return :ng if $~.size != 5
+      end
+    }
+    :ok
+  ensure
+    Thread.kill t
+  end
+  m
+}, '[ruby-dev:34492]'
+
+assert_normal_exit %q{
+  at_exit { Fiber.new{}.resume }
+}
+
+assert_normal_exit %q{
+  g = enum_for(:local_variables)
+  loop { g.next }
+}, '[ruby-dev:34128]'
+
+assert_normal_exit %q{
+  g = enum_for(:block_given?)
+  loop { g.next }
+}, '[ruby-dev:34128]'
+
+assert_normal_exit %q{
+  g = enum_for(:binding)
+  loop { g.next }
+}, '[ruby-dev:34128]'
+
+assert_normal_exit %q{
+  g = "abc".enum_for(:scan, /./)
+  loop { g.next }
+}, '[ruby-dev:34128]'
+
+assert_normal_exit %q{
+  g = Module.enum_for(:new)
+  loop { g.next }
+}, '[ruby-dev:34128]'
+
+assert_normal_exit %q{
+  Fiber.new(&Object.method(:class_eval)).resume("foo")
+}, '[ruby-dev:34128]'
+
+assert_normal_exit %q{
+  Thread.new("foo", &Object.method(:class_eval)).join
+}, '[ruby-dev:34128]'
+
+assert_equal 'ok', %q{
+  begin
+    Thread.new { Thread.stop }
+    Thread.stop
+    :ng
+  rescue Exception
+    :ok
+  end
+}
+
+assert_equal 'ok', %q{
+  begin
+    m1, m2 = Thread::Mutex.new, Thread::Mutex.new
+    f1 = f2 = false
+    Thread.new { m1.lock; f2 = true; sleep 0.001 until f1; m2.lock }
+    m2.lock; f1 = true; sleep 0.001 until f2; m1.lock
+    :ng
+  rescue Exception
+    :ok
+  end
+}
+
+assert_equal 'ok', %q{
+  m = Thread::Mutex.new
+  Thread.new { m.lock }; sleep 0.1; m.lock
+  :ok
+}
+
+assert_equal 'ok', %q{
+  m = Thread::Mutex.new
+  Thread.new { m.lock }; m.lock
+  :ok
+}
+
+assert_equal 'ok', %q{
+  m = Thread::Mutex.new
+  Thread.new { m.lock }.join; m.lock
+  :ok
+}
+
+assert_equal 'ok', %q{
+  m = Thread::Mutex.new
+  Thread.new { m.lock; sleep 0.2 }
+  sleep 0.1; m.lock
+  :ok
+}
+
+assert_equal 'ok', %q{
+  m = Thread::Mutex.new
+  Thread.new { m.lock; sleep 0.2; m.unlock }
+  sleep 0.1; m.lock
+  :ok
+}
+
+assert_equal 'ok', %q{
+  t = Thread.new {`echo`}
+  t.join
+  $? ? :ng : :ok
+}, '[ruby-dev:35414]'
+
+assert_equal 'ok', %q{
+  begin
+    100.times{
+      (1..100).map{ Thread.new(true) {|x| x == false } }.each{|th| th.join}
+    }
+  rescue NoMemoryError, StandardError
+  end
+  :ok
+}
+
+assert_equal 'ok', %{
+  open("zzz.rb", "w") do |f|
+    f.puts <<-'end;' # do
+      begin
+        m = Thread::Mutex.new
+        parent = Thread.current
+        th1 = Thread.new { m.lock; sleep }
+        sleep 0.01 until th1.stop?
+        Thread.new do
+          sleep 0.01 until parent.stop?
+          begin
+            fork { GC.start }
+          rescue Exception
+            parent.raise $!
+          end
+          th1.run
+        end
+        m.lock
+        pid, status = Process.wait2
+        $result = status.success? ? :ok : :ng
+      rescue NotImplementedError
+        $result = :ok
+      end
+    end;
+  end
+  require "./zzz.rb"
+  $result
+}
+
+assert_finish 3, %q{
+  require 'thread'
+
+  lock = Thread::Mutex.new
+  cond = Thread::ConditionVariable.new
+  t = Thread.new do
+    lock.synchronize do
+      cond.wait(lock)
+    end
+  end
+
+  begin
+    pid = fork do
+      # Child
+      STDOUT.write "This is the child process.\n"
+      STDOUT.write "Child process exiting.\n"
+    end
+    Process.waitpid(pid)
+  rescue NotImplementedError
+  end
+}, '[ruby-core:23572]'
+
+assert_equal 'ok', %q{
+  begin
+    Process.waitpid2(fork {})[1].success? ? 'ok' : 'ng'
+  rescue NotImplementedError
+    'ok'
+  end
+}
+
+assert_equal 'foo', %q{
+  i = 0
+  Thread.start {sleep 1; exit!}
+  f = proc {|s, c| /#{c.call; s}/o }
+  th2 = Thread.new {
+    sleep 0.01 until i == 1
+    i = 2
+    f.call("bar", proc {sleep 2});
+    nil
+  }
+  th1 = Thread.new {
+    f.call("foo", proc {i = 1; sleep 0.01 until i == 2; sleep 0.01})
+    nil
+  }
+  [th1, th2].each {|t| t.join }
+  GC.start
+  f.call.source
+}
diff --git a/ccan/build_assert/build_assert.h b/ccan/build_assert/build_assert.h
new file mode 100644
index 0000000000..a04d1d4709
--- /dev/null
+++ b/ccan/build_assert/build_assert.h
@@ -0,0 +1,40 @@
+/* CC0 (Public domain) - see ccan/licenses/CC0 file for details */
+#ifndef CCAN_BUILD_ASSERT_H
+#define CCAN_BUILD_ASSERT_H
+
+/**
+ * BUILD_ASSERT - assert a build-time dependency.
+ * @cond: the compile-time condition which must be true.
+ *
+ * Your compile will fail if the condition isn't true, or can't be evaluated
+ * by the compiler.  This can only be used within a function.
+ *
+ * Example:
+ *	#include <stddef.h>
+ *	...
+ *	static char *foo_to_char(struct foo *foo)
+ *	{
+ *		// This code needs string to be at start of foo.
+ *		BUILD_ASSERT(offsetof(struct foo, string) == 0);
+ *		return (char *)foo;
+ *	}
+ */
+#define BUILD_ASSERT(cond) \
+	do { (void) sizeof(char [1 - 2*!(cond)]); } while(0)
+
+/**
+ * BUILD_ASSERT_OR_ZERO - assert a build-time dependency, as an expression.
+ * @cond: the compile-time condition which must be true.
+ *
+ * Your compile will fail if the condition isn't true, or can't be evaluated
+ * by the compiler.  This can be used in an expression: its value is "0".
+ *
+ * Example:
+ *	#define foo_to_char(foo)					\
+ *		 ((char *)(foo)						\
+ *		  + BUILD_ASSERT_OR_ZERO(offsetof(struct foo, string) == 0))
+ */
+#define BUILD_ASSERT_OR_ZERO(cond) \
+	(sizeof(char [1 - 2*!(cond)]) - 1)
+
+#endif /* CCAN_BUILD_ASSERT_H */
diff --git a/ccan/check_type/check_type.h b/ccan/check_type/check_type.h
new file mode 100644
index 0000000000..1f77a535e4
--- /dev/null
+++ b/ccan/check_type/check_type.h
@@ -0,0 +1,63 @@
+/* CC0 (Public domain) - see ccan/licenses/CC0 file for details */
+#ifndef CCAN_CHECK_TYPE_H
+#define CCAN_CHECK_TYPE_H
+
+/**
+ * check_type - issue a warning or build failure if type is not correct.
+ * @expr: the expression whose type we should check (not evaluated).
+ * @type: the exact type we expect the expression to be.
+ *
+ * This macro is usually used within other macros to try to ensure that a macro
+ * argument is of the expected type.  No type promotion of the expression is
+ * done: an unsigned int is not the same as an int!
+ *
+ * check_type() always evaluates to 0.
+ *
+ * If your compiler does not support typeof, then the best we can do is fail
+ * to compile if the sizes of the types are unequal (a less complete check).
+ *
+ * Example:
+ *	// They should always pass a 64-bit value to _set_some_value!
+ *	#define set_some_value(expr)			\
+ *		_set_some_value((check_type((expr), uint64_t), (expr)))
+ */
+
+/**
+ * check_types_match - issue a warning or build failure if types are not same.
+ * @expr1: the first expression (not evaluated).
+ * @expr2: the second expression (not evaluated).
+ *
+ * This macro is usually used within other macros to try to ensure that
+ * arguments are of identical types.  No type promotion of the expressions is
+ * done: an unsigned int is not the same as an int!
+ *
+ * check_types_match() always evaluates to 0.
+ *
+ * If your compiler does not support typeof, then the best we can do is fail
+ * to compile if the sizes of the types are unequal (a less complete check).
+ *
+ * Example:
+ *	// Do subtraction to get to enclosing type, but make sure that
+ *	// pointer is of correct type for that member.
+ *	#define container_of(mbr_ptr, encl_type, mbr)			\
+ *		(check_types_match((mbr_ptr), &((encl_type *)0)->mbr),	\
+ *		 ((encl_type *)						\
+ *		  ((char *)(mbr_ptr) - offsetof(enclosing_type, mbr))))
+ */
+#if HAVE_TYPEOF
+#define check_type(expr, type)			\
+	((typeof(expr) *)0 != (type *)0)
+
+#define check_types_match(expr1, expr2)		\
+	((typeof(expr1) *)0 != (typeof(expr2) *)0)
+#else
+#include "ccan/build_assert/build_assert.h"
+/* Without typeof, we can only test the sizes. */
+#define check_type(expr, type)					\
+	BUILD_ASSERT_OR_ZERO(sizeof(expr) == sizeof(type))
+
+#define check_types_match(expr1, expr2)				\
+	BUILD_ASSERT_OR_ZERO(sizeof(expr1) == sizeof(expr2))
+#endif /* HAVE_TYPEOF */
+
+#endif /* CCAN_CHECK_TYPE_H */
diff --git a/ccan/container_of/container_of.h b/ccan/container_of/container_of.h
new file mode 100644
index 0000000000..ae3e1fc81f
--- /dev/null
+++ b/ccan/container_of/container_of.h
@@ -0,0 +1,142 @@
+/* CC0 (Public domain) - see ccan/licenses/CC0 file for details */
+#ifndef CCAN_CONTAINER_OF_H
+#define CCAN_CONTAINER_OF_H
+#include "ccan/check_type/check_type.h"
+
+/**
+ * container_of - get pointer to enclosing structure
+ * @member_ptr: pointer to the structure member
+ * @containing_type: the type this member is within
+ * @member: the name of this member within the structure.
+ *
+ * Given a pointer to a member of a structure, this macro does pointer
+ * subtraction to return the pointer to the enclosing type.
+ *
+ * Example:
+ *	struct foo {
+ *		int fielda, fieldb;
+ *		// ...
+ *	};
+ *	struct info {
+ *		int some_other_field;
+ *		struct foo my_foo;
+ *	};
+ *
+ *	static struct info *foo_to_info(struct foo *foo)
+ *	{
+ *		return container_of(foo, struct info, my_foo);
+ *	}
+ */
+#define container_of(member_ptr, containing_type, member)		\
+	 ((containing_type *)						\
+	  ((char *)(member_ptr)						\
+	   - container_off(containing_type, member))			\
+	  + check_types_match(*(member_ptr), ((containing_type *)0)->member))
+
+
+/**
+ * container_of_or_null - get pointer to enclosing structure, or NULL
+ * @member_ptr: pointer to the structure member
+ * @containing_type: the type this member is within
+ * @member: the name of this member within the structure.
+ *
+ * Given a pointer to a member of a structure, this macro does pointer
+ * subtraction to return the pointer to the enclosing type, unless it
+ * is given NULL, in which case it also returns NULL.
+ *
+ * Example:
+ *	struct foo {
+ *		int fielda, fieldb;
+ *		// ...
+ *	};
+ *	struct info {
+ *		int some_other_field;
+ *		struct foo my_foo;
+ *	};
+ *
+ *	static struct info *foo_to_info_allowing_null(struct foo *foo)
+ *	{
+ *		return container_of_or_null(foo, struct info, my_foo);
+ *	}
+ */
+static inline char *container_of_or_null_(void *member_ptr, size_t offset)
+{
+	return member_ptr ? (char *)member_ptr - offset : NULL;
+}
+#define container_of_or_null(member_ptr, containing_type, member)	\
+	((containing_type *)						\
+	 container_of_or_null_(member_ptr,				\
+			       container_off(containing_type, member))	\
+	 + check_types_match(*(member_ptr), ((containing_type *)0)->member))
+
+/**
+ * container_off - get offset to enclosing structure
+ * @containing_type: the type this member is within
+ * @member: the name of this member within the structure.
+ *
+ * Given a pointer to a member of a structure, this macro does
+ * typechecking and figures out the offset to the enclosing type.
+ *
+ * Example:
+ *	struct foo {
+ *		int fielda, fieldb;
+ *		// ...
+ *	};
+ *	struct info {
+ *		int some_other_field;
+ *		struct foo my_foo;
+ *	};
+ *
+ *	static struct info *foo_to_info(struct foo *foo)
+ *	{
+ *		size_t off = container_off(struct info, my_foo);
+ *		return (void *)((char *)foo - off);
+ *	}
+ */
+#define container_off(containing_type, member)	\
+	offsetof(containing_type, member)
+
+/**
+ * container_of_var - get pointer to enclosing structure using a variable
+ * @member_ptr: pointer to the structure member
+ * @container_var: a pointer of same type as this member's container
+ * @member: the name of this member within the structure.
+ *
+ * Given a pointer to a member of a structure, this macro does pointer
+ * subtraction to return the pointer to the enclosing type.
+ *
+ * Example:
+ *	static struct info *foo_to_i(struct foo *foo)
+ *	{
+ *		struct info *i = container_of_var(foo, i, my_foo);
+ *		return i;
+ *	}
+ */
+#if HAVE_TYPEOF
+#define container_of_var(member_ptr, container_var, member) \
+	container_of(member_ptr, typeof(*container_var), member)
+#else
+#define container_of_var(member_ptr, container_var, member)	\
+	((void *)((char *)(member_ptr)	-			\
+		  container_off_var(container_var, member)))
+#endif
+
+/**
+ * container_off_var - get offset of a field in enclosing structure
+ * @container_var: a pointer to a container structure
+ * @member: the name of a member within the structure.
+ *
+ * Given (any) pointer to a structure and a its member name, this
+ * macro does pointer subtraction to return offset of member in a
+ * structure memory layout.
+ *
+ */
+#if HAVE_TYPEOF
+#define container_off_var(var, member)		\
+	container_off(typeof(*var), member)
+#else
+#define container_off_var(var, member)			\
+	((const char *)&(var)->member - (const char *)(var))
+#endif
+
+#endif /* CCAN_CONTAINER_OF_H */
diff --git a/ccan/licenses/BSD-MIT b/ccan/licenses/BSD-MIT
new file mode 100644
index 0000000000..89de354795
--- /dev/null
+++ b/ccan/licenses/BSD-MIT
@@ -0,0 +1,17 @@
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
diff --git a/ccan/licenses/CC0 b/ccan/licenses/CC0
new file mode 100644
index 0000000000..feb9b118e6
--- /dev/null
+++ b/ccan/licenses/CC0
@@ -0,0 +1,28 @@
+Statement of Purpose
+
+The laws of most jurisdictions throughout the world automatically confer exclusive Copyright and Related Rights (defined below) upon the creator and subsequent owner(s) (each and all, an "owner") of an original work of authorship and/or a database (each, a "Work").
+
+Certain owners wish to permanently relinquish those rights to a Work for the purpose of contributing to a commons of creative, cultural and scientific works ("Commons") that the public can reliably and without fear of later claims of infringement build upon, modify, incorporate in other works, reuse and redistribute as freely as possible in any form whatsoever and for any purposes, including without limitation commercial purposes. These owners may contribute to the Commons to promote the ideal of a free culture and the further production of creative, cultural and scientific works, or to gain reputation or greater distribution for their Work in part through the use and efforts of others.
+
+For these and/or other purposes and motivations, and without any expectation of additional consideration or compensation, the person associating CC0 with a Work (the "Affirmer"), to the extent that he or she is an owner of Copyright and Related Rights in the Work, voluntarily elects to apply CC0 to the Work and publicly distribute the Work under its terms, with knowledge of his or her Copyright and Related Rights in the Work and the meaning and intended legal effect of CC0 on those rights.
+
+1. Copyright and Related Rights. A Work made available under CC0 may be protected by copyright and related or neighboring rights ("Copyright and Related Rights"). Copyright and Related Rights include, but are not limited to, the following:
+
+    the right to reproduce, adapt, distribute, perform, display, communicate, and translate a Work;
+    moral rights retained by the original author(s) and/or performer(s);
+    publicity and privacy rights pertaining to a person's image or likeness depicted in a Work;
+    rights protecting against unfair competition in regards to a Work, subject to the limitations in paragraph 4(a), below;
+    rights protecting the extraction, dissemination, use and reuse of data in a Work;
+    database rights (such as those arising under Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the legal protection of databases, and under any national implementation thereof, including any amended or successor version of such directive); and
+    other similar, equivalent or corresponding rights throughout the world based on applicable law or treaty, and any national implementations thereof.
+
+2. Waiver. To the greatest extent permitted by, but not in contravention of, applicable law, Affirmer hereby overtly, fully, permanently, irrevocably and unconditionally waives, abandons, and surrenders all of Affirmer's Copyright and Related Rights and associated claims and causes of action, whether now known or unknown (including existing as well as future claims and causes of action), in the Work (i) in all territories worldwide, (ii) for the maximum duration provided by applicable law or treaty (including future time extensions), (iii) in any current or future medium and for any number of copies, and (iv) for any purpose whatsoever, including without limitation commercial, advertising or promotional purposes (the "Waiver"). Affirmer makes the Waiver for the benefit of each member of the public at large and to the detriment of Affirmer's heirs and successors, fully intending that such Waiver shall not be subject to revocation, rescission, cancellation, termination, or any other legal or equitable action to disrupt the quiet enjoyment of the Work by the public as contemplated by Affirmer's express Statement of Purpose.
+
+3. Public License Fallback. Should any part of the Waiver for any reason be judged legally invalid or ineffective under applicable law, then the Waiver shall be preserved to the maximum extent permitted taking into account Affirmer's express Statement of Purpose. In addition, to the extent the Waiver is so judged Affirmer hereby grants to each affected person a royalty-free, non transferable, non sublicensable, non exclusive, irrevocable and unconditional license to exercise Affirmer's Copyright and Related Rights in the Work (i) in all territories worldwide, (ii) for the maximum duration provided by applicable law or treaty (including future time extensions), (iii) in any current or future medium and for any number of copies, and (iv) for any purpose whatsoever, including without limitation commercial, advertising or promotional purposes (the "License"). The License shall be deemed effective as of the date CC0 was applied by Affirmer to the Work. Should any part of the License for any reason be judged legally invalid or ineffective under applicable law, such partial invalidity or ineffectiveness shall not invalidate the remainder of the License, and in such case Affirmer hereby affirms that he or she will not (i) exercise any of his or her remaining Copyright and Related Rights in the Work or (ii) assert any associated claims and causes of action with respect to the Work, in either case contrary to Affirmer's express Statement of Purpose.
+
+4. Limitations and Disclaimers.
+
+    No trademark or patent rights held by Affirmer are waived, abandoned, surrendered, licensed or otherwise affected by this document.
+    Affirmer offers the Work as-is and makes no representations or warranties of any kind concerning the Work, express, implied, statutory or otherwise, including without limitation warranties of title, merchantability, fitness for a particular purpose, non infringement, or the absence of latent or other defects, accuracy, or the present or absence of errors, whether or not discoverable, all to the greatest extent permissible under applicable law.
+    Affirmer disclaims responsibility for clearing rights of other persons that may apply to the Work or any use thereof, including without limitation any person's Copyright and Related Rights in the Work. Further, Affirmer disclaims responsibility for obtaining any necessary consents, permissions or other rights required for any use of the Work.
+    Affirmer understands and acknowledges that Creative Commons is not a party to this document and has no duty or obligation with respect to this CC0 or use of the Work.
diff --git a/ccan/list/list.h b/ccan/list/list.h
new file mode 100644
index 0000000000..59ab45ee53
--- /dev/null
+++ b/ccan/list/list.h
@@ -0,0 +1,788 @@
+/* Licensed under BSD-MIT - see ccan/licenses/BSD-MIT file for details */
+#ifndef CCAN_LIST_H
+#define CCAN_LIST_H
+#include <assert.h>
+#include "ccan/str/str.h"
+#include "ccan/container_of/container_of.h"
+#include "ccan/check_type/check_type.h"
+
+/**
+ * struct list_node - an entry in a doubly-linked list
+ * @next: next entry (self if empty)
+ * @prev: previous entry (self if empty)
+ *
+ * This is used as an entry in a linked list.
+ * Example:
+ *	struct child {
+ *		const char *name;
+ *		// Linked list of all us children.
+ *		struct list_node list;
+ *	};
+ */
+struct list_node
+{
+	struct list_node *next, *prev;
+};
+
+/**
+ * struct list_head - the head of a doubly-linked list
+ * @h: the list_head (containing next and prev pointers)
+ *
+ * This is used as the head of a linked list.
+ * Example:
+ *	struct parent {
+ *		const char *name;
+ *		struct list_head children;
+ *		unsigned int num_children;
+ *	};
+ */
+struct list_head
+{
+	struct list_node n;
+};
+
+#define LIST_LOC __FILE__  ":" stringify(__LINE__)
+#define list_debug(h, loc) ((void)loc, h)
+#define list_debug_node(n, loc) ((void)loc, n)
+
+/**
+ * LIST_HEAD_INIT - initializer for an empty list_head
+ * @name: the name of the list.
+ *
+ * Explicit initializer for an empty list.
+ *
+ * See also:
+ *	LIST_HEAD, list_head_init()
+ *
+ * Example:
+ *	static struct list_head my_list = LIST_HEAD_INIT(my_list);
+ */
+#define LIST_HEAD_INIT(name) { { &(name).n, &(name).n } }
+
+/**
+ * LIST_HEAD - define and initialize an empty list_head
+ * @name: the name of the list.
+ *
+ * The LIST_HEAD macro defines a list_head and initializes it to an empty
+ * list.  It can be prepended by "static" to define a static list_head.
+ *
+ * See also:
+ *	LIST_HEAD_INIT, list_head_init()
+ *
+ * Example:
+ *	static LIST_HEAD(my_global_list);
+ */
+#define LIST_HEAD(name) \
+	struct list_head name = LIST_HEAD_INIT(name)
+
+/**
+ * list_head_init - initialize a list_head
+ * @h: the list_head to set to the empty list
+ *
+ * Example:
+ *	...
+ *	struct parent *parent = malloc(sizeof(*parent));
+ *
+ *	list_head_init(&parent->children);
+ *	parent->num_children = 0;
+ */
+static inline void list_head_init(struct list_head *h)
+{
+	h->n.next = h->n.prev = &h->n;
+}
+
+/**
+ * list_node_init - initialize a list_node
+ * @n: the list_node to link to itself.
+ *
+ * You don't need to use this normally!  But it lets you list_del(@n)
+ * safely.
+ */
+static inline void list_node_init(struct list_node *n)
+{
+	n->next = n->prev = n;
+}
+
+/**
+ * list_add_after - add an entry after an existing node in a linked list
+ * @h: the list_head to add the node to (for debugging)
+ * @p: the existing list_node to add the node after
+ * @n: the new list_node to add to the list.
+ *
+ * The existing list_node must already be a member of the list.
+ * The new list_node does not need to be initialized; it will be overwritten.
+ *
+ * Example:
+ *	struct child c1, c2, c3;
+ *	LIST_HEAD(h);
+ *
+ *	list_add_tail(&h, &c1.list);
+ *	list_add_tail(&h, &c3.list);
+ *	list_add_after(&h, &c1.list, &c2.list);
+ */
+#define list_add_after(h, p, n) list_add_after_(h, p, n, LIST_LOC)
+static inline void list_add_after_(struct list_head *h,
+				   struct list_node *p,
+				   struct list_node *n,
+				   const char *abortstr)
+{
+	n->next = p->next;
+	n->prev = p;
+	p->next->prev = n;
+	p->next = n;
+	(void)list_debug(h, abortstr);
+}
+
+/**
+ * list_add - add an entry at the start of a linked list.
+ * @h: the list_head to add the node to
+ * @n: the list_node to add to the list.
+ *
+ * The list_node does not need to be initialized; it will be overwritten.
+ * Example:
+ *	struct child *child = malloc(sizeof(*child));
+ *
+ *	child->name = "marvin";
+ *	list_add(&parent->children, &child->list);
+ *	parent->num_children++;
+ */
+#define list_add(h, n) list_add_(h, n, LIST_LOC)
+static inline void list_add_(struct list_head *h,
+			     struct list_node *n,
+			     const char *abortstr)
+{
+	list_add_after_(h, &h->n, n, abortstr);
+}
+
+/**
+ * list_add_before - add an entry before an existing node in a linked list
+ * @h: the list_head to add the node to (for debugging)
+ * @p: the existing list_node to add the node before
+ * @n: the new list_node to add to the list.
+ *
+ * The existing list_node must already be a member of the list.
+ * The new list_node does not need to be initialized; it will be overwritten.
+ *
+ * Example:
+ *	list_head_init(&h);
+ *	list_add_tail(&h, &c1.list);
+ *	list_add_tail(&h, &c3.list);
+ *	list_add_before(&h, &c3.list, &c2.list);
+ */
+#define list_add_before(h, p, n) list_add_before_(h, p, n, LIST_LOC)
+static inline void list_add_before_(struct list_head *h,
+				    struct list_node *p,
+				    struct list_node *n,
+				    const char *abortstr)
+{
+	n->next = p;
+	n->prev = p->prev;
+	p->prev->next = n;
+	p->prev = n;
+	(void)list_debug(h, abortstr);
+}
+
+/**
+ * list_add_tail - add an entry at the end of a linked list.
+ * @h: the list_head to add the node to
+ * @n: the list_node to add to the list.
+ *
+ * The list_node does not need to be initialized; it will be overwritten.
+ * Example:
+ *	list_add_tail(&parent->children, &child->list);
+ *	parent->num_children++;
+ */
+#define list_add_tail(h, n) list_add_tail_(h, n, LIST_LOC)
+static inline void list_add_tail_(struct list_head *h,
+				  struct list_node *n,
+				  const char *abortstr)
+{
+	list_add_before_(h, &h->n, n, abortstr);
+}
+
+/**
+ * list_empty - is a list empty?
+ * @h: the list_head
+ *
+ * If the list is empty, returns true.
+ *
+ * Example:
+ *	assert(list_empty(&parent->children) == (parent->num_children == 0));
+ */
+#define list_empty(h) list_empty_(h, LIST_LOC)
+static inline int list_empty_(const struct list_head *h, const char* abortstr)
+{
+	(void)list_debug(h, abortstr);
+	return h->n.next == &h->n;
+}
+
+/**
+ * list_empty_nodebug - is a list empty (and don't perform debug checks)?
+ * @h: the list_head
+ *
+ * If the list is empty, returns true.
+ * This differs from list_empty() in that if CCAN_LIST_DEBUG is set it
+ * will NOT perform debug checks. Only use this function if you REALLY
+ * know what you're doing.
+ *
+ * Example:
+ *	assert(list_empty_nodebug(&parent->children) == (parent->num_children == 0));
+ */
+#ifndef CCAN_LIST_DEBUG
+#define list_empty_nodebug(h) list_empty(h)
+#else
+static inline int list_empty_nodebug(const struct list_head *h)
+{
+	return h->n.next == &h->n;
+}
+#endif
+
+/**
+ * list_empty_nocheck - is a list empty?
+ * @h: the list_head
+ *
+ * If the list is empty, returns true. This doesn't perform any
+ * debug check for list consistency, so it can be called without
+ * locks, racing with the list being modified. This is ok for
+ * checks where an incorrect result is not an issue (optimized
+ * bail out path for example).
+ */
+static inline bool list_empty_nocheck(const struct list_head *h)
+{
+	return h->n.next == &h->n;
+}
+
+/**
+ * list_del - delete an entry from an (unknown) linked list.
+ * @n: the list_node to delete from the list.
+ *
+ * Note that this leaves @n in an undefined state; it can be added to
+ * another list, but not deleted again.
+ *
+ * See also:
+ *	list_del_from(), list_del_init()
+ *
+ * Example:
+ *	list_del(&child->list);
+ *	parent->num_children--;
+ */
+#define list_del(n) list_del_(n, LIST_LOC)
+static inline void list_del_(struct list_node *n, const char* abortstr)
+{
+	(void)list_debug_node(n, abortstr);
+	n->next->prev = n->prev;
+	n->prev->next = n->next;
+#ifdef CCAN_LIST_DEBUG
+	/* Catch use-after-del. */
+	n->next = n->prev = NULL;
+#endif
+}
+
+/**
+ * list_del_init - delete a node, and reset it so it can be deleted again.
+ * @n: the list_node to be deleted.
+ *
+ * list_del(@n) or list_del_init() again after this will be safe,
+ * which can be useful in some cases.
+ *
+ * See also:
+ *	list_del_from(), list_del()
+ *
+ * Example:
+ *	list_del_init(&child->list);
+ *	parent->num_children--;
+ */
+#define list_del_init(n) list_del_init_(n, LIST_LOC)
+static inline void list_del_init_(struct list_node *n, const char *abortstr)
+{
+	list_del_(n, abortstr);
+	list_node_init(n);
+}
+
+/**
+ * list_del_from - delete an entry from a known linked list.
+ * @h: the list_head the node is in.
+ * @n: the list_node to delete from the list.
+ *
+ * This explicitly indicates which list a node is expected to be in,
+ * which is better documentation and can catch more bugs.
+ *
+ * See also: list_del()
+ *
+ * Example:
+ *	list_del_from(&parent->children, &child->list);
+ *	parent->num_children--;
+ */
+static inline void list_del_from(struct list_head *h, struct list_node *n)
+{
+#ifdef CCAN_LIST_DEBUG
+	{
+		/* Thorough check: make sure it was in list! */
+		struct list_node *i;
+		for (i = h->n.next; i != n; i = i->next)
+			assert(i != &h->n);
+	}
+#endif /* CCAN_LIST_DEBUG */
+
+	/* Quick test that catches a surprising number of bugs. */
+	assert(!list_empty(h));
+	list_del(n);
+}
+
+/**
+ * list_swap - swap out an entry from an (unknown) linked list for a new one.
+ * @o: the list_node to replace from the list.
+ * @n: the list_node to insert in place of the old one.
+ *
+ * Note that this leaves @o in an undefined state; it can be added to
+ * another list, but not deleted/swapped again.
+ *
+ * See also:
+ *	list_del()
+ *
+ * Example:
+ *	struct child x1, x2;
+ *	LIST_HEAD(xh);
+ *
+ *	list_add(&xh, &x1.list);
+ *	list_swap(&x1.list, &x2.list);
+ */
+#define list_swap(o, n) list_swap_(o, n, LIST_LOC)
+static inline void list_swap_(struct list_node *o,
+			      struct list_node *n,
+			      const char* abortstr)
+{
+	(void)list_debug_node(o, abortstr);
+	*n = *o;
+	n->next->prev = n;
+	n->prev->next = n;
+#ifdef CCAN_LIST_DEBUG
+	/* Catch use-after-del. */
+	o->next = o->prev = NULL;
+#endif
+}
+
+/**
+ * list_entry - convert a list_node back into the structure containing it.
+ * @n: the list_node
+ * @type: the type of the entry
+ * @member: the list_node member of the type
+ *
+ * Example:
+ *	// First list entry is children.next; convert back to child.
+ *	child = list_entry(parent->children.n.next, struct child, list);
+ *
+ * See Also:
+ *	list_top(), list_for_each()
+ */
+#define list_entry(n, type, member) container_of(n, type, member)
+
+/**
+ * list_top - get the first entry in a list
+ * @h: the list_head
+ * @type: the type of the entry
+ * @member: the list_node member of the type
+ *
+ * If the list is empty, returns NULL.
+ *
+ * Example:
+ *	struct child *first;
+ *	first = list_top(&parent->children, struct child, list);
+ *	if (!first)
+ *		printf("Empty list!\n");
+ */
+#define list_top(h, type, member)					\
+	((type *)list_top_((h), list_off_(type, member)))
+
+static inline const void *list_top_(const struct list_head *h, size_t off)
+{
+	if (list_empty(h))
+		return NULL;
+	return (const char *)h->n.next - off;
+}
+
+/**
+ * list_pop - remove the first entry in a list
+ * @h: the list_head
+ * @type: the type of the entry
+ * @member: the list_node member of the type
+ *
+ * If the list is empty, returns NULL.
+ *
+ * Example:
+ *	struct child *one;
+ *	one = list_pop(&parent->children, struct child, list);
+ *	if (!one)
+ *		printf("Empty list!\n");
+ */
+#define list_pop(h, type, member)					\
+	((type *)list_pop_((h), list_off_(type, member)))
+
+static inline const void *list_pop_(const struct list_head *h, size_t off)
+{
+	struct list_node *n;
+
+	if (list_empty(h))
+		return NULL;
+	n = h->n.next;
+	list_del(n);
+	return (const char *)n - off;
+}
+
+/**
+ * list_tail - get the last entry in a list
+ * @h: the list_head
+ * @type: the type of the entry
+ * @member: the list_node member of the type
+ *
+ * If the list is empty, returns NULL.
+ *
+ * Example:
+ *	struct child *last;
+ *	last = list_tail(&parent->children, struct child, list);
+ *	if (!last)
+ *		printf("Empty list!\n");
+ */
+#define list_tail(h, type, member) \
+	((type *)list_tail_((h), list_off_(type, member)))
+
+static inline const void *list_tail_(const struct list_head *h, size_t off)
+{
+	if (list_empty(h))
+		return NULL;
+	return (const char *)h->n.prev - off;
+}
+
+/**
+ * list_for_each - iterate through a list.
+ * @h: the list_head (warning: evaluated multiple times!)
+ * @i: the structure containing the list_node
+ * @member: the list_node member of the structure
+ *
+ * This is a convenient wrapper to iterate @i over the entire list.  It's
+ * a for loop, so you can break and continue as normal.
+ *
+ * Example:
+ *	list_for_each(&parent->children, child, list)
+ *		printf("Name: %s\n", child->name);
+ */
+#define list_for_each(h, i, member)					\
+	list_for_each_off(h, i, list_off_var_(i, member))
+
+/**
+ * list_for_each_rev - iterate through a list backwards.
+ * @h: the list_head
+ * @i: the structure containing the list_node
+ * @member: the list_node member of the structure
+ *
+ * This is a convenient wrapper to iterate @i over the entire list.  It's
+ * a for loop, so you can break and continue as normal.
+ *
+ * Example:
+ *	list_for_each_rev(&parent->children, child, list)
+ *		printf("Name: %s\n", child->name);
+ */
+#define list_for_each_rev(h, i, member)					\
+	list_for_each_rev_off(h, i, list_off_var_(i, member))
+
+/**
+ * list_for_each_rev_safe - iterate through a list backwards,
+ * maybe during deletion
+ * @h: the list_head
+ * @i: the structure containing the list_node
+ * @nxt: the structure containing the list_node
+ * @member: the list_node member of the structure
+ *
+ * This is a convenient wrapper to iterate @i over the entire list backwards.
+ * It's a for loop, so you can break and continue as normal.  The extra
+ * variable * @nxt is used to hold the next element, so you can delete @i
+ * from the list.
+ *
+ * Example:
+ *	struct child *next;
+ *	list_for_each_rev_safe(&parent->children, child, next, list) {
+ *		printf("Name: %s\n", child->name);
+ *	}
+ */
+#define list_for_each_rev_safe(h, i, nxt, member)			\
+	list_for_each_rev_safe_off(h, i, nxt, list_off_var_(i, member))
+
+/**
+ * list_for_each_safe - iterate through a list, maybe during deletion
+ * @h: the list_head
+ * @i: the structure containing the list_node
+ * @nxt: the structure containing the list_node
+ * @member: the list_node member of the structure
+ *
+ * This is a convenient wrapper to iterate @i over the entire list.  It's
+ * a for loop, so you can break and continue as normal.  The extra variable
+ * @nxt is used to hold the next element, so you can delete @i from the list.
+ *
+ * Example:
+ *	list_for_each_safe(&parent->children, child, next, list) {
+ *		list_del(&child->list);
+ *		parent->num_children--;
+ *	}
+ */
+#define list_for_each_safe(h, i, nxt, member)				\
+	list_for_each_safe_off(h, i, nxt, list_off_var_(i, member))
+
+/**
+ * list_next - get the next entry in a list
+ * @h: the list_head
+ * @i: a pointer to an entry in the list.
+ * @member: the list_node member of the structure
+ *
+ * If @i was the last entry in the list, returns NULL.
+ *
+ * Example:
+ *	struct child *second;
+ *	second = list_next(&parent->children, first, list);
+ *	if (!second)
+ *		printf("No second child!\n");
+ */
+#define list_next(h, i, member)						\
+	((list_typeof(i))list_entry_or_null(list_debug(h,		\
+					    __FILE__ ":" stringify(__LINE__)), \
+					    (i)->member.next,		\
+					    list_off_var_((i), member)))
+
+/**
+ * list_prev - get the previous entry in a list
+ * @h: the list_head
+ * @i: a pointer to an entry in the list.
+ * @member: the list_node member of the structure
+ *
+ * If @i was the first entry in the list, returns NULL.
+ *
+ * Example:
+ *	first = list_prev(&parent->children, second, list);
+ *	if (!first)
+ *		printf("Can't go back to first child?!\n");
+ */
+#define list_prev(h, i, member)						\
+	((list_typeof(i))list_entry_or_null(list_debug(h,		\
+					    __FILE__ ":" stringify(__LINE__)), \
+					    (i)->member.prev,		\
+					    list_off_var_((i), member)))
+
+/**
+ * list_append_list - empty one list onto the end of another.
+ * @to: the list to append into
+ * @from: the list to empty.
+ *
+ * This takes the entire contents of @from and moves it to the end of
+ * @to.  After this @from will be empty.
+ *
+ * Example:
+ *	struct list_head adopter;
+ *
+ *	list_append_list(&adopter, &parent->children);
+ *	assert(list_empty(&parent->children));
+ *	parent->num_children = 0;
+ */
+#define list_append_list(t, f) list_append_list_(t, f,			\
+				   __FILE__ ":" stringify(__LINE__))
+static inline void list_append_list_(struct list_head *to,
+				     struct list_head *from,
+				     const char *abortstr)
+{
+	struct list_node *from_tail = list_debug(from, abortstr)->n.prev;
+	struct list_node *to_tail = list_debug(to, abortstr)->n.prev;
+
+	/* Sew in head and entire list. */
+	to->n.prev = from_tail;
+	from_tail->next = &to->n;
+	to_tail->next = &from->n;
+	from->n.prev = to_tail;
+
+	/* Now remove head. */
+	list_del(&from->n);
+	list_head_init(from);
+}
+
+/**
+ * list_prepend_list - empty one list into the start of another.
+ * @to: the list to prepend into
+ * @from: the list to empty.
+ *
+ * This takes the entire contents of @from and moves it to the start
+ * of @to.  After this @from will be empty.
+ *
+ * Example:
+ *	list_prepend_list(&adopter, &parent->children);
+ *	assert(list_empty(&parent->children));
+ *	parent->num_children = 0;
+ */
+#define list_prepend_list(t, f) list_prepend_list_(t, f, LIST_LOC)
+static inline void list_prepend_list_(struct list_head *to,
+				      struct list_head *from,
+				      const char *abortstr)
+{
+	struct list_node *from_tail = list_debug(from, abortstr)->n.prev;
+	struct list_node *to_head = list_debug(to, abortstr)->n.next;
+
+	/* Sew in head and entire list. */
+	to->n.next = &from->n;
+	from->n.prev = &to->n;
+	to_head->prev = from_tail;
+	from_tail->next = to_head;
+
+	/* Now remove head. */
+	list_del(&from->n);
+	list_head_init(from);
+}
+
+/* internal macros, do not use directly */
+#define list_for_each_off_dir_(h, i, off, dir)				\
+	for (i = list_node_to_off_(list_debug(h, LIST_LOC)->n.dir,	\
+				   (off));				\
+	list_node_from_off_((void *)i, (off)) != &(h)->n;		\
+	i = list_node_to_off_(list_node_from_off_((void *)i, (off))->dir, \
+			      (off)))
+
+#define list_for_each_safe_off_dir_(h, i, nxt, off, dir)		\
+	for (i = list_node_to_off_(list_debug(h, LIST_LOC)->n.dir,	\
+				   (off)),				\
+	nxt = list_node_to_off_(list_node_from_off_(i, (off))->dir,	\
+				(off));					\
+	list_node_from_off_(i, (off)) != &(h)->n;			\
+	i = nxt,							\
+	nxt = list_node_to_off_(list_node_from_off_(i, (off))->dir,	\
+				(off)))
+
+/**
+ * list_for_each_off - iterate through a list of memory regions.
+ * @h: the list_head
+ * @i: the pointer to a memory region wich contains list node data.
+ * @off: offset(relative to @i) at which list node data resides.
+ *
+ * This is a low-level wrapper to iterate @i over the entire list, used to
+ * implement all oher, more high-level, for-each constructs. It's a for loop,
+ * so you can break and continue as normal.
+ *
+ * WARNING! Being the low-level macro that it is, this wrapper doesn't know
+ * nor care about the type of @i. The only assumption made is that @i points
+ * to a chunk of memory that at some @offset, relative to @i, contains a
+ * properly filled `struct list_node' which in turn contains pointers to
+ * memory chunks and it's turtles all the way down. With all that in mind
+ * remember that given the wrong pointer/offset couple this macro will
+ * happily churn all you memory until SEGFAULT stops it, in other words
+ * caveat emptor.
+ *
+ * It is worth mentioning that one of legitimate use-cases for that wrapper
+ * is operation on opaque types with known offset for `struct list_node'
+ * member(preferably 0), because it allows you not to disclose the type of
+ * @i.
+ *
+ * Example:
+ *	list_for_each_off(&parent->children, child,
+ *				offsetof(struct child, list))
+ *		printf("Name: %s\n", child->name);
+ */
+#define list_for_each_off(h, i, off)                                    \
+	list_for_each_off_dir_((h),(i),(off),next)
+
+/**
+ * list_for_each_rev_off - iterate through a list of memory regions backwards
+ * @h: the list_head
+ * @i: the pointer to a memory region wich contains list node data.
+ * @off: offset(relative to @i) at which list node data resides.
+ *
+ * See list_for_each_off for details
+ */
+#define list_for_each_rev_off(h, i, off)                                    \
+	list_for_each_off_dir_((h),(i),(off),prev)
+
+/**
+ * list_for_each_safe_off - iterate through a list of memory regions, maybe
+ * during deletion
+ * @h: the list_head
+ * @i: the pointer to a memory region wich contains list node data.
+ * @nxt: the structure containing the list_node
+ * @off: offset(relative to @i) at which list node data resides.
+ *
+ * For details see `list_for_each_off' and `list_for_each_safe'
+ * descriptions.
+ *
+ * Example:
+ *	list_for_each_safe_off(&parent->children, child,
+ *		next, offsetof(struct child, list))
+ *		printf("Name: %s\n", child->name);
+ */
+#define list_for_each_safe_off(h, i, nxt, off)                          \
+	list_for_each_safe_off_dir_((h),(i),(nxt),(off),next)
+
+/**
+ * list_for_each_rev_safe_off - iterate backwards through a list of
+ * memory regions, maybe during deletion
+ * @h: the list_head
+ * @i: the pointer to a memory region wich contains list node data.
+ * @nxt: the structure containing the list_node
+ * @off: offset(relative to @i) at which list node data resides.
+ *
+ * For details see `list_for_each_rev_off' and `list_for_each_rev_safe'
+ * descriptions.
+ *
+ * Example:
+ *	list_for_each_rev_safe_off(&parent->children, child,
+ *		next, offsetof(struct child, list))
+ *		printf("Name: %s\n", child->name);
+ */
+#define list_for_each_rev_safe_off(h, i, nxt, off)                      \
+	list_for_each_safe_off_dir_((h),(i),(nxt),(off),prev)
+
+/* Other -off variants. */
+#define list_entry_off(n, type, off)		\
+	((type *)list_node_from_off_((n), (off)))
+
+#define list_head_off(h, type, off)		\
+	((type *)list_head_off((h), (off)))
+
+#define list_tail_off(h, type, off)		\
+	((type *)list_tail_((h), (off)))
+
+#define list_add_off(h, n, off)                 \
+	list_add((h), list_node_from_off_((n), (off)))
+
+#define list_del_off(n, off)                    \
+	list_del(list_node_from_off_((n), (off)))
+
+#define list_del_from_off(h, n, off)			\
+	list_del_from(h, list_node_from_off_((n), (off)))
+
+/* Offset helper functions so we only single-evaluate. */
+static inline void *list_node_to_off_(struct list_node *node, size_t off)
+{
+	return (void *)((char *)node - off);
+}
+static inline struct list_node *list_node_from_off_(void *ptr, size_t off)
+{
+	return (struct list_node *)((char *)ptr + off);
+}
+
+/* Get the offset of the member, but make sure it's a list_node. */
+#define list_off_(type, member)					\
+	(container_off(type, member) +				\
+	 check_type(((type *)0)->member, struct list_node))
+
+#define list_off_var_(var, member)			\
+	(container_off_var(var, member) +		\
+	 check_type(var->member, struct list_node))
+
+#if HAVE_TYPEOF
+#define list_typeof(var) typeof(var)
+#else
+#define list_typeof(var) void *
+#endif
+
+/* Returns member, or NULL if at end of list. */
+static inline void *list_entry_or_null(const struct list_head *h,
+				       const struct list_node *n,
+				       size_t off)
+{
+	if (n == &h->n)
+		return NULL;
+	return (char *)n - off;
+}
+#endif /* CCAN_LIST_H */
diff --git a/ccan/str/str.h b/ccan/str/str.h
new file mode 100644
index 0000000000..9a9da9cd3f
--- /dev/null
+++ b/ccan/str/str.h
@@ -0,0 +1,16 @@
+/* CC0 (Public domain) - see ccan/licenses/CC0 file for details */
+#ifndef CCAN_STR_H
+#define CCAN_STR_H
+/**
+ * stringify - Turn expression into a string literal
+ * @expr: any C expression
+ *
+ * Example:
+ *	#define PRINT_COND_IF_FALSE(cond) \
+ *		((cond) || printf("%s is false!", stringify(cond)))
+ */
+#define stringify(expr)		stringify_1(expr)
+/* Double-indirection required to stringify expansions */
+#define stringify_1(expr)	#expr
+
+#endif /* CCAN_STR_H */
diff --git a/class.c b/class.c
index 9e41b1e260..c1cf6ab051 100644
--- a/class.c
+++ b/class.c
@@ -2,214 +2,648 @@
 
   class.c -
 
-  $Author: nobu $
-  $Date: 2005/09/28 14:42:46 $
+  $Author$
   created at: Tue Aug 10 15:05:44 JST 1993
 
-  Copyright (C) 1993-2003 Yukihiro Matsumoto
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
 
 **********************************************************************/
 
-#include "ruby.h"
-#include "rubysig.h"
-#include "node.h"
-#include "st.h"
+/*!
+ * \defgroup class Classes and their hierarchy.
+ * \par Terminology
+ * - class: same as in Ruby.
+ * - singleton class: class for a particular object
+ * - eigenclass: = singleton class
+ * - metaclass: class of a class. metaclass is a kind of singleton class.
+ * - metametaclass: class of a metaclass.
+ * - meta^(n)-class: class of a meta^(n-1)-class.
+ * - attached object: A singleton class knows its unique instance.
+ *   The instance is called the attached object for the singleton class.
+ * \{
+ */
+
+#include "internal.h"
+#include "ruby/st.h"
+#include "constant.h"
+#include "vm_core.h"
+#include "id_table.h"
 #include <ctype.h>
 
-extern st_table *rb_class_tbl;
+#define id_attached id__attached__
+
+void
+rb_class_subclass_add(VALUE super, VALUE klass)
+{
+    rb_subclass_entry_t *entry, *head;
+
+    if (super && super != Qundef) {
+	entry = ALLOC(rb_subclass_entry_t);
+	entry->klass = klass;
+	entry->next = NULL;
+
+	head = RCLASS_EXT(super)->subclasses;
+	if (head) {
+	    entry->next = head;
+	    RCLASS_EXT(head->klass)->parent_subclasses = &entry->next;
+	}
+
+	RCLASS_EXT(super)->subclasses = entry;
+	RCLASS_EXT(klass)->parent_subclasses = &RCLASS_EXT(super)->subclasses;
+    }
+}
+
+static void
+rb_module_add_to_subclasses_list(VALUE module, VALUE iclass)
+{
+    rb_subclass_entry_t *entry, *head;
+
+    entry = ALLOC(rb_subclass_entry_t);
+    entry->klass = iclass;
+    entry->next = NULL;
+
+    head = RCLASS_EXT(module)->subclasses;
+    if (head) {
+	entry->next = head;
+	RCLASS_EXT(head->klass)->module_subclasses = &entry->next;
+    }
+
+    RCLASS_EXT(module)->subclasses = entry;
+    RCLASS_EXT(iclass)->module_subclasses = &RCLASS_EXT(module)->subclasses;
+}
+
+void
+rb_class_remove_from_super_subclasses(VALUE klass)
+{
+    rb_subclass_entry_t *entry;
+
+    if (RCLASS_EXT(klass)->parent_subclasses) {
+	entry = *RCLASS_EXT(klass)->parent_subclasses;
+
+	*RCLASS_EXT(klass)->parent_subclasses = entry->next;
+	if (entry->next) {
+	    RCLASS_EXT(entry->next->klass)->parent_subclasses = RCLASS_EXT(klass)->parent_subclasses;
+	}
+	xfree(entry);
+    }
+
+    RCLASS_EXT(klass)->parent_subclasses = NULL;
+}
+
+void
+rb_class_remove_from_module_subclasses(VALUE klass)
+{
+    rb_subclass_entry_t *entry;
+
+    if (RCLASS_EXT(klass)->module_subclasses) {
+	entry = *RCLASS_EXT(klass)->module_subclasses;
+	*RCLASS_EXT(klass)->module_subclasses = entry->next;
+
+	if (entry->next) {
+	    RCLASS_EXT(entry->next->klass)->module_subclasses = RCLASS_EXT(klass)->module_subclasses;
+	}
+
+	xfree(entry);
+    }
+
+    RCLASS_EXT(klass)->module_subclasses = NULL;
+}
+
+void
+rb_class_foreach_subclass(VALUE klass, void (*f)(VALUE, VALUE), VALUE arg)
+{
+    rb_subclass_entry_t *cur = RCLASS_EXT(klass)->subclasses;
+
+    /* do not be tempted to simplify this loop into a for loop, the order of
+       operations is important here if `f` modifies the linked list */
+    while (cur) {
+	VALUE curklass = cur->klass;
+	cur = cur->next;
+	f(curklass, arg);
+    }
+}
+
+static void
+class_detach_subclasses(VALUE klass, VALUE arg)
+{
+    rb_class_remove_from_super_subclasses(klass);
+}
+
+void
+rb_class_detach_subclasses(VALUE klass)
+{
+    rb_class_foreach_subclass(klass, class_detach_subclasses, Qnil);
+}
+
+static void
+class_detach_module_subclasses(VALUE klass, VALUE arg)
+{
+    rb_class_remove_from_module_subclasses(klass);
+}
+
+void
+rb_class_detach_module_subclasses(VALUE klass)
+{
+    rb_class_foreach_subclass(klass, class_detach_module_subclasses, Qnil);
+}
+
+/**
+ * Allocates a struct RClass for a new class.
+ *
+ * \param flags     initial value for basic.flags of the returned class.
+ * \param klass     the class of the returned class.
+ * \return          an uninitialized Class object.
+ * \pre  \p klass must refer \c Class class or an ancestor of Class.
+ * \pre  \code (flags | T_CLASS) != 0  \endcode
+ * \post the returned class can safely be \c #initialize 'd.
+ *
+ * \note this function is not Class#allocate.
+ */
+static VALUE
+class_alloc(VALUE flags, VALUE klass)
+{
+    NEWOBJ_OF(obj, struct RClass, klass, (flags & T_MASK) | FL_PROMOTED1 /* start from age == 2 */ | (RGENGC_WB_PROTECTED_CLASS ? FL_WB_PROTECTED : 0));
+    obj->ptr = ZALLOC(rb_classext_t);
+    /* ZALLOC
+      RCLASS_IV_TBL(obj) = 0;
+      RCLASS_CONST_TBL(obj) = 0;
+      RCLASS_M_TBL(obj) = 0;
+      RCLASS_IV_INDEX_TBL(obj) = 0;
+      RCLASS_SET_SUPER((VALUE)obj, 0);
+      RCLASS_EXT(obj)->subclasses = NULL;
+      RCLASS_EXT(obj)->parent_subclasses = NULL;
+      RCLASS_EXT(obj)->module_subclasses = NULL;
+     */
+    RCLASS_SET_ORIGIN((VALUE)obj, (VALUE)obj);
+    RCLASS_SERIAL(obj) = rb_next_class_serial();
+    RCLASS_REFINED_CLASS(obj) = Qnil;
+    RCLASS_EXT(obj)->allocator = 0;
+
+    return (VALUE)obj;
+}
+
+static void
+RCLASS_M_TBL_INIT(VALUE c)
+{
+    RCLASS_M_TBL(c) = rb_id_table_create(0);
+}
 
+/*!
+ * A utility function that wraps class_alloc.
+ *
+ * allocates a class and initializes safely.
+ * \param super     a class from which the new class derives.
+ * \return          a class object.
+ * \pre  \a super must be a class.
+ * \post the metaclass of the new class is Class.
+ */
 VALUE
-rb_class_boot(super)
-    VALUE super;
+rb_class_boot(VALUE super)
 {
-    NEWOBJ(klass, struct RClass);
-    OBJSETUP(klass, rb_cClass, T_CLASS);
+    VALUE klass = class_alloc(T_CLASS, rb_cClass);
 
-    klass->super = super;
-    klass->iv_tbl = 0;
-    klass->m_tbl = 0;		/* safe GC */
-    klass->m_tbl = st_init_numtable();
+    RCLASS_SET_SUPER(klass, super);
+    RCLASS_M_TBL_INIT(klass);
 
     OBJ_INFECT(klass, super);
     return (VALUE)klass;
 }
 
-VALUE
-rb_class_new(super)
-    VALUE super;
+
+/*!
+ * Ensures a class can be derived from super.
+ *
+ * \param super a reference to an object.
+ * \exception TypeError if \a super is not a Class or \a super is a singleton class.
+ */
+void
+rb_check_inheritable(VALUE super)
 {
-    Check_Type(super, T_CLASS);
+    if (!RB_TYPE_P(super, T_CLASS)) {
+	rb_raise(rb_eTypeError, "superclass must be a Class (%"PRIsVALUE" given)",
+		 rb_obj_class(super));
+    }
+    if (RBASIC(super)->flags & FL_SINGLETON) {
+	rb_raise(rb_eTypeError, "can't make subclass of singleton class");
+    }
     if (super == rb_cClass) {
 	rb_raise(rb_eTypeError, "can't make subclass of Class");
     }
-    if (FL_TEST(super, FL_SINGLETON)) {
-	rb_raise(rb_eTypeError, "can't make subclass of virtual class");
-    }
+}
+
+
+/*!
+ * Creates a new class.
+ * \param super     a class from which the new class derives.
+ * \exception TypeError \a super is not inheritable.
+ * \exception TypeError \a super is the Class class.
+ */
+VALUE
+rb_class_new(VALUE super)
+{
+    Check_Type(super, T_CLASS);
+    rb_check_inheritable(super);
     return rb_class_boot(super);
 }
 
+static void
+clone_method(VALUE old_klass, VALUE new_klass, ID mid, const rb_method_entry_t *me)
+{
+    if (me->def->type == VM_METHOD_TYPE_ISEQ) {
+	rb_cref_t *new_cref;
+	rb_vm_rewrite_cref(me->def->body.iseq.cref, old_klass, new_klass, &new_cref);
+	rb_add_method_iseq(new_klass, mid, me->def->body.iseq.iseqptr, new_cref, METHOD_ENTRY_VISI(me));
+    }
+    else {
+	rb_method_entry_set(new_klass, mid, me, METHOD_ENTRY_VISI(me));
+    }
+}
+
+struct clone_method_arg {
+    VALUE new_klass;
+    VALUE old_klass;
+};
+
+static enum rb_id_table_iterator_result
+clone_method_i(ID key, VALUE value, void *data)
+{
+    const struct clone_method_arg *arg = (struct clone_method_arg *)data;
+    clone_method(arg->old_klass, arg->new_klass, key, (const rb_method_entry_t *)value);
+    return ID_TABLE_CONTINUE;
+}
+
+struct clone_const_arg {
+    VALUE klass;
+    struct rb_id_table *tbl;
+};
+
 static int
-clone_method(mid, body, tbl)
-    ID mid;
-    NODE *body;
-    st_table *tbl;
+clone_const(ID key, const rb_const_entry_t *ce, struct clone_const_arg *arg)
 {
-    st_insert(tbl, mid, (st_data_t)NEW_METHOD(body->nd_body, body->nd_noex));
-    return ST_CONTINUE;
+    rb_const_entry_t *nce = ALLOC(rb_const_entry_t);
+    MEMCPY(nce, ce, rb_const_entry_t, 1);
+    RB_OBJ_WRITTEN(arg->klass, Qundef, ce->value);
+    RB_OBJ_WRITTEN(arg->klass, Qundef, ce->file);
+
+    rb_id_table_insert(arg->tbl, key, (VALUE)nce);
+    return ID_TABLE_CONTINUE;
+}
+
+static enum rb_id_table_iterator_result
+clone_const_i(ID key, VALUE value, void *data)
+{
+    return clone_const(key, (const rb_const_entry_t *)value, data);
+}
+
+static void
+class_init_copy_check(VALUE clone, VALUE orig)
+{
+    if (orig == rb_cBasicObject) {
+	rb_raise(rb_eTypeError, "can't copy the root class");
+    }
+    if (RCLASS_SUPER(clone) != 0 || clone == rb_cBasicObject) {
+	rb_raise(rb_eTypeError, "already initialized class");
+    }
+    if (FL_TEST(orig, FL_SINGLETON)) {
+	rb_raise(rb_eTypeError, "can't copy singleton class");
+    }
 }
 
 /* :nodoc: */
 VALUE
-rb_mod_init_copy(clone, orig)
-    VALUE clone, orig;
+rb_mod_init_copy(VALUE clone, VALUE orig)
 {
-    rb_obj_init_copy(clone, orig);
+    if (RB_TYPE_P(clone, T_CLASS)) {
+	class_init_copy_check(clone, orig);
+    }
+    if (!OBJ_INIT_COPY(clone, orig)) return clone;
     if (!FL_TEST(CLASS_OF(clone), FL_SINGLETON)) {
-	RBASIC(clone)->klass = rb_singleton_class_clone(orig);
+	RBASIC_SET_CLASS(clone, rb_singleton_class_clone(orig));
+	rb_singleton_class_attached(RBASIC(clone)->klass, (VALUE)clone);
+    }
+    RCLASS_SET_SUPER(clone, RCLASS_SUPER(orig));
+    RCLASS_EXT(clone)->allocator = RCLASS_EXT(orig)->allocator;
+    if (RCLASS_IV_TBL(clone)) {
+	st_free_table(RCLASS_IV_TBL(clone));
+	RCLASS_IV_TBL(clone) = 0;
     }
-    RCLASS(clone)->super = RCLASS(orig)->super;
-    if (RCLASS(orig)->iv_tbl) {
-	ID id;
+    if (RCLASS_CONST_TBL(clone)) {
+	rb_free_const_table(RCLASS_CONST_TBL(clone));
+	RCLASS_CONST_TBL(clone) = 0;
+    }
+    RCLASS_M_TBL(clone) = 0;
+    if (RCLASS_IV_TBL(orig)) {
+	st_data_t id;
+
+	RCLASS_IV_TBL(clone) = rb_st_copy(clone, RCLASS_IV_TBL(orig));
+	CONST_ID(id, "__tmp_classpath__");
+	st_delete(RCLASS_IV_TBL(clone), &id, 0);
+	CONST_ID(id, "__classpath__");
+	st_delete(RCLASS_IV_TBL(clone), &id, 0);
+	CONST_ID(id, "__classid__");
+	st_delete(RCLASS_IV_TBL(clone), &id, 0);
+    }
+    if (RCLASS_CONST_TBL(orig)) {
+	struct clone_const_arg arg;
 
-	RCLASS(clone)->iv_tbl = st_copy(RCLASS(orig)->iv_tbl);
-	id = rb_intern("__classpath__");
-	st_delete(RCLASS(clone)->iv_tbl, (st_data_t*)&id, 0);
-	id = rb_intern("__classid__");
-	st_delete(RCLASS(clone)->iv_tbl, (st_data_t*)&id, 0);
+	arg.tbl = RCLASS_CONST_TBL(clone) = rb_id_table_create(0);
+	arg.klass = clone;
+	rb_id_table_foreach(RCLASS_CONST_TBL(orig), clone_const_i, &arg);
     }
-    if (RCLASS(orig)->m_tbl) {
-	RCLASS(clone)->m_tbl = st_init_numtable();
-	st_foreach(RCLASS(orig)->m_tbl, clone_method,
-	  (st_data_t)RCLASS(clone)->m_tbl);
+    if (RCLASS_M_TBL(orig)) {
+	struct clone_method_arg arg;
+	arg.old_klass = orig;
+	arg.new_klass = clone;
+	RCLASS_M_TBL_INIT(clone);
+	rb_id_table_foreach(RCLASS_M_TBL(orig), clone_method_i, &arg);
     }
 
     return clone;
 }
 
-/* :nodoc: */
 VALUE
-rb_class_init_copy(clone, orig)
-    VALUE clone, orig;
+rb_singleton_class_clone(VALUE obj)
 {
-    if (RCLASS(clone)->super != 0) {
-	rb_raise(rb_eTypeError, "already initialized class");
-    }
-    if (FL_TEST(orig, FL_SINGLETON)) {
-	rb_raise(rb_eTypeError, "can't copy singleton class");
-    }
-    return rb_mod_init_copy(clone, orig);
+    return rb_singleton_class_clone_and_attach(obj, Qundef);
 }
 
 VALUE
-rb_singleton_class_clone(obj)
-    VALUE obj;
+rb_singleton_class_clone_and_attach(VALUE obj, VALUE attach)
 {
-    VALUE klass = RBASIC(obj)->klass;
+    const VALUE klass = RBASIC(obj)->klass;
 
     if (!FL_TEST(klass, FL_SINGLETON))
 	return klass;
     else {
 	/* copy singleton(unnamed) class */
-	NEWOBJ(clone, struct RClass);
-	OBJSETUP(clone, 0, RBASIC(klass)->flags);
+	VALUE clone = class_alloc(RBASIC(klass)->flags, 0);
 
 	if (BUILTIN_TYPE(obj) == T_CLASS) {
-	    RBASIC(clone)->klass = (VALUE)clone;
+	    RBASIC_SET_CLASS(clone, clone);
 	}
 	else {
-	    RBASIC(clone)->klass = rb_singleton_class_clone(klass);
+	    RBASIC_SET_CLASS(clone, rb_singleton_class_clone(klass));
 	}
 
-	clone->super = RCLASS(klass)->super;
-	clone->iv_tbl = 0;
-	clone->m_tbl = 0;
-	if (RCLASS(klass)->iv_tbl) {
-	    clone->iv_tbl = st_copy(RCLASS(klass)->iv_tbl);
+	RCLASS_SET_SUPER(clone, RCLASS_SUPER(klass));
+	RCLASS_EXT(clone)->allocator = RCLASS_EXT(klass)->allocator;
+	if (RCLASS_IV_TBL(klass)) {
+	    RCLASS_IV_TBL(clone) = rb_st_copy(clone, RCLASS_IV_TBL(klass));
 	}
-	clone->m_tbl = st_init_numtable();
-	st_foreach(RCLASS(klass)->m_tbl, clone_method,
-	  (st_data_t)clone->m_tbl);
-	rb_singleton_class_attached(RBASIC(clone)->klass, (VALUE)clone);
+	if (RCLASS_CONST_TBL(klass)) {
+	    struct clone_const_arg arg;
+	    arg.tbl = RCLASS_CONST_TBL(clone) = rb_id_table_create(0);
+	    arg.klass = clone;
+	    rb_id_table_foreach(RCLASS_CONST_TBL(klass), clone_const_i, &arg);
+	}
+	if (attach != Qundef) {
+	    rb_singleton_class_attached(clone, attach);
+	}
+	RCLASS_M_TBL_INIT(clone);
+	{
+	    struct clone_method_arg arg;
+	    arg.old_klass = klass;
+	    arg.new_klass = clone;
+	    rb_id_table_foreach(RCLASS_M_TBL(klass), clone_method_i, &arg);
+	}
+	rb_singleton_class_attached(RBASIC(clone)->klass, clone);
 	FL_SET(clone, FL_SINGLETON);
-	return (VALUE)clone;
+
+	return clone;
     }
 }
 
+/*!
+ * Attach a object to a singleton class.
+ * @pre \a klass is the singleton class of \a obj.
+ */
 void
-rb_singleton_class_attached(klass, obj)
-    VALUE klass, obj;
+rb_singleton_class_attached(VALUE klass, VALUE obj)
 {
     if (FL_TEST(klass, FL_SINGLETON)) {
-	if (!RCLASS(klass)->iv_tbl) {
-	    RCLASS(klass)->iv_tbl = st_init_numtable();
+	if (!RCLASS_IV_TBL(klass)) {
+	    RCLASS_IV_TBL(klass) = st_init_numtable();
 	}
-	st_insert(RCLASS(klass)->iv_tbl, rb_intern("__attached__"), obj);
+	rb_class_ivar_set(klass, id_attached, obj);
     }
 }
 
-VALUE
-rb_make_metaclass(obj, super)
-    VALUE obj, super;
+
+
+#define METACLASS_OF(k) RBASIC(k)->klass
+#define SET_METACLASS_OF(k, cls) RBASIC_SET_CLASS(k, cls)
+
+/*!
+ * whether k is a meta^(n)-class of Class class
+ * @retval 1 if \a k is a meta^(n)-class of Class class (n >= 0)
+ * @retval 0 otherwise
+ */
+#define META_CLASS_OF_CLASS_CLASS_P(k)  (METACLASS_OF(k) == (k))
+
+static int
+rb_singleton_class_has_metaclass_p(VALUE sklass)
 {
-    VALUE klass = rb_class_boot(super);
-    FL_SET(klass, FL_SINGLETON);
-    RBASIC(obj)->klass = klass;
-    rb_singleton_class_attached(klass, obj);
-    if (BUILTIN_TYPE(obj) == T_CLASS && FL_TEST(obj, FL_SINGLETON)) {
-	RBASIC(klass)->klass = klass;
-	RCLASS(klass)->super = RBASIC(rb_class_real(RCLASS(obj)->super))->klass;
+    return rb_attr_get(METACLASS_OF(sklass), id_attached) == sklass;
+}
+
+int
+rb_singleton_class_internal_p(VALUE sklass)
+{
+    return (RB_TYPE_P(rb_attr_get(sklass, id_attached), T_CLASS) &&
+	    !rb_singleton_class_has_metaclass_p(sklass));
+}
+
+/*!
+ * whether k has a metaclass
+ * @retval 1 if \a k has a metaclass
+ * @retval 0 otherwise
+ */
+#define HAVE_METACLASS_P(k) \
+    (FL_TEST(METACLASS_OF(k), FL_SINGLETON) && \
+     rb_singleton_class_has_metaclass_p(k))
+
+/*!
+ * ensures \a klass belongs to its own eigenclass.
+ * @return the eigenclass of \a klass
+ * @post \a klass belongs to the returned eigenclass.
+ *       i.e. the attached object of the eigenclass is \a klass.
+ * @note this macro creates a new eigenclass if necessary.
+ */
+#define ENSURE_EIGENCLASS(klass) \
+    (HAVE_METACLASS_P(klass) ? METACLASS_OF(klass) : make_metaclass(klass))
+
+
+/*!
+ * Creates a metaclass of \a klass
+ * \param klass     a class
+ * \return          created metaclass for the class
+ * \pre \a klass is a Class object
+ * \pre \a klass has no singleton class.
+ * \post the class of \a klass is the returned class.
+ * \post the returned class is meta^(n+1)-class when \a klass is a meta^(n)-klass for n >= 0
+ */
+static inline VALUE
+make_metaclass(VALUE klass)
+{
+    VALUE super;
+    VALUE metaclass = rb_class_boot(Qundef);
+
+    FL_SET(metaclass, FL_SINGLETON);
+    rb_singleton_class_attached(metaclass, klass);
+
+    if (META_CLASS_OF_CLASS_CLASS_P(klass)) {
+	SET_METACLASS_OF(klass, metaclass);
+	SET_METACLASS_OF(metaclass, metaclass);
     }
     else {
-	VALUE metasuper = RBASIC(rb_class_real(super))->klass;
-
-	/* metaclass of a superclass may be NULL at boot time */
-	if (metasuper) {
-	    RBASIC(klass)->klass = metasuper;
-	}
+	VALUE tmp = METACLASS_OF(klass); /* for a meta^(n)-class klass, tmp is meta^(n)-class of Class class */
+	SET_METACLASS_OF(klass, metaclass);
+	SET_METACLASS_OF(metaclass, ENSURE_EIGENCLASS(tmp));
     }
 
-    return klass;
+    super = RCLASS_SUPER(klass);
+    while (RB_TYPE_P(super, T_ICLASS)) super = RCLASS_SUPER(super);
+    RCLASS_SET_SUPER(metaclass, super ? ENSURE_EIGENCLASS(super) : rb_cClass);
+
+    OBJ_INFECT(metaclass, RCLASS_SUPER(metaclass));
+
+    return metaclass;
 }
 
-VALUE
-rb_define_class_id(id, super)
-    ID id;
-    VALUE super;
+/*!
+ * Creates a singleton class for \a obj.
+ * \pre \a obj must not a immediate nor a special const.
+ * \pre \a obj must not a Class object.
+ * \pre \a obj has no singleton class.
+ */
+static inline VALUE
+make_singleton_class(VALUE obj)
 {
-    VALUE klass;
+    VALUE orig_class = RBASIC(obj)->klass;
+    VALUE klass = rb_class_boot(orig_class);
 
-    if (!super) super = rb_cObject;
-    klass = rb_class_new(super);
-    rb_make_metaclass(klass, RBASIC(super)->klass);
+    FL_SET(klass, FL_SINGLETON);
+    RBASIC_SET_CLASS(obj, klass);
+    rb_singleton_class_attached(klass, obj);
 
+    SET_METACLASS_OF(klass, METACLASS_OF(rb_class_real(orig_class)));
     return klass;
 }
 
+
+static VALUE
+boot_defclass(const char *name, VALUE super)
+{
+    VALUE obj = rb_class_boot(super);
+    ID id = rb_intern(name);
+
+    rb_name_class(obj, id);
+    rb_const_set((rb_cObject ? rb_cObject : obj), id, obj);
+    return obj;
+}
+
 void
-rb_check_inheritable(super)
-    VALUE super;
+Init_class_hierarchy(void)
+{
+    rb_cBasicObject = boot_defclass("BasicObject", 0);
+    rb_cObject = boot_defclass("Object", rb_cBasicObject);
+    rb_gc_register_mark_object(rb_cObject);
+
+    /* resolve class name ASAP for order-independence */
+    rb_class_name(rb_cObject);
+
+    rb_cModule = boot_defclass("Module", rb_cObject);
+    rb_cClass =  boot_defclass("Class",  rb_cModule);
+
+    rb_const_set(rb_cObject, rb_intern_const("BasicObject"), rb_cBasicObject);
+    RBASIC_SET_CLASS(rb_cClass, rb_cClass);
+    RBASIC_SET_CLASS(rb_cModule, rb_cClass);
+    RBASIC_SET_CLASS(rb_cObject, rb_cClass);
+    RBASIC_SET_CLASS(rb_cBasicObject, rb_cClass);
+}
+
+
+/*!
+ * \internal
+ * Creates a new *singleton class* for an object.
+ *
+ * \pre \a obj has no singleton class.
+ * \note DO NOT USE the function in an extension libraries. Use \ref rb_singleton_class.
+ * \param obj     An object.
+ * \param unused  ignored.
+ * \return        The singleton class of the object.
+ */
+VALUE
+rb_make_metaclass(VALUE obj, VALUE unused)
 {
-    if (TYPE(super) != T_CLASS) {
-	rb_raise(rb_eTypeError, "superclass must be a Class (%s given)",
-		 rb_obj_classname(super));
+    if (BUILTIN_TYPE(obj) == T_CLASS) {
+	return make_metaclass(obj);
     }
-    if (RBASIC(super)->flags & FL_SINGLETON) {
-	rb_raise(rb_eTypeError, "can't make subclass of virtual class");
+    else {
+	return make_singleton_class(obj);
     }
 }
 
+
+/*!
+ * Defines a new class.
+ * \param id     ignored
+ * \param super  A class from which the new class will derive. NULL means \c Object class.
+ * \return       the created class
+ * \throw TypeError if super is not a \c Class object.
+ *
+ * \note the returned class will not be associated with \a id.
+ *       You must explicitly set a class name if necessary.
+ */
 VALUE
-rb_class_inherited(super, klass)
-    VALUE super, klass;
+rb_define_class_id(ID id, VALUE super)
+{
+    VALUE klass;
+
+    if (!super) super = rb_cObject;
+    klass = rb_class_new(super);
+    rb_make_metaclass(klass, RBASIC(super)->klass);
+
+    return klass;
+}
+
+
+/*!
+ * Calls Class#inherited.
+ * \param super  A class which will be called #inherited.
+ *               NULL means Object class.
+ * \param klass  A Class object which derived from \a super
+ * \return the value \c Class#inherited's returns
+ * \pre Each of \a super and \a klass must be a \c Class object.
+ */
+MJIT_FUNC_EXPORTED VALUE
+rb_class_inherited(VALUE super, VALUE klass)
 {
+    ID inherited;
     if (!super) super = rb_cObject;
-    return rb_funcall(super, rb_intern("inherited"), 1, klass);
+    CONST_ID(inherited, "inherited");
+    return rb_funcall(super, inherited, 1, klass);
 }
 
+
+
+/*!
+ * Defines a top-level class.
+ * \param name   name of the class
+ * \param super  a class from which the new class will derive.
+ * \return the created class
+ * \throw TypeError if the constant name \a name is already taken but
+ *                  the constant is not a \c Class.
+ * \throw TypeError if the class is already defined but the class can not
+ *                  be reopened because its superclass is not \a super.
+ * \throw ArgumentError if the \a super is NULL.
+ * \post top-level constant named \a name refers the returned class.
+ *
+ * \note if a class named \a name is already defined and its superclass is
+ *       \a super, the function just returns the defined class.
+ */
 VALUE
-rb_define_class(name, super)
-    const char *name;
-    VALUE super;
+rb_define_class(const char *name, VALUE super)
 {
     VALUE klass;
     ID id;
@@ -217,19 +651,20 @@ rb_define_class(name, super)
     id = rb_intern(name);
     if (rb_const_defined(rb_cObject, id)) {
 	klass = rb_const_get(rb_cObject, id);
-	if (TYPE(klass) != T_CLASS) {
-	    rb_raise(rb_eTypeError, "%s is not a class", name);
+	if (!RB_TYPE_P(klass, T_CLASS)) {
+	    rb_raise(rb_eTypeError, "%s is not a class (%"PRIsVALUE")",
+		     name, rb_obj_class(klass));
 	}
-	if (rb_class_real(RCLASS(klass)->super) != super) {
-	    rb_name_error(id, "%s is already defined", name);
+	if (rb_class_real(RCLASS_SUPER(klass)) != super) {
+	    rb_raise(rb_eTypeError, "superclass mismatch for class %s", name);
 	}
 	return klass;
     }
     if (!super) {
-	rb_warn("no super class for `%s', Object assumed", name);
+	rb_raise(rb_eArgError, "no super class for `%s'", name);
     }
     klass = rb_define_class_id(id, super);
-    st_add_direct(rb_class_tbl, id, klass);
+    rb_vm_add_root_module(id, klass);
     rb_name_class(klass, id);
     rb_const_set(rb_cObject, id, klass);
     rb_class_inherited(super, klass);
@@ -237,55 +672,89 @@ rb_define_class(name, super)
     return klass;
 }
 
+
+/*!
+ * Defines a class under the namespace of \a outer.
+ * \param outer  a class which contains the new class.
+ * \param name   name of the new class
+ * \param super  a class from which the new class will derive.
+ *               NULL means \c Object class.
+ * \return the created class
+ * \throw TypeError if the constant name \a name is already taken but
+ *                  the constant is not a \c Class.
+ * \throw TypeError if the class is already defined but the class can not
+ *                  be reopened because its superclass is not \a super.
+ * \post top-level constant named \a name refers the returned class.
+ *
+ * \note if a class named \a name is already defined and its superclass is
+ *       \a super, the function just returns the defined class.
+ */
 VALUE
-rb_define_class_under(outer, name, super)
-    VALUE outer;
-    const char *name;
-    VALUE super;
+rb_define_class_under(VALUE outer, const char *name, VALUE super)
+{
+    return rb_define_class_id_under(outer, rb_intern(name), super);
+}
+
+
+/*!
+ * Defines a class under the namespace of \a outer.
+ * \param outer  a class which contains the new class.
+ * \param id     name of the new class
+ * \param super  a class from which the new class will derive.
+ *               NULL means \c Object class.
+ * \return the created class
+ * \throw TypeError if the constant name \a name is already taken but
+ *                  the constant is not a \c Class.
+ * \throw TypeError if the class is already defined but the class can not
+ *                  be reopened because its superclass is not \a super.
+ * \post top-level constant named \a name refers the returned class.
+ *
+ * \note if a class named \a name is already defined and its superclass is
+ *       \a super, the function just returns the defined class.
+ */
+VALUE
+rb_define_class_id_under(VALUE outer, ID id, VALUE super)
 {
     VALUE klass;
-    ID id;
 
-    id = rb_intern(name);
     if (rb_const_defined_at(outer, id)) {
 	klass = rb_const_get_at(outer, id);
-	if (TYPE(klass) != T_CLASS) {
-	    rb_raise(rb_eTypeError, "%s is not a class", name);
+	if (!RB_TYPE_P(klass, T_CLASS)) {
+	    rb_raise(rb_eTypeError, "%"PRIsVALUE"::%"PRIsVALUE" is not a class"
+		     " (%"PRIsVALUE")",
+		     outer, rb_id2str(id), rb_obj_class(klass));
 	}
-	if (rb_class_real(RCLASS(klass)->super) != super) {
-	    rb_name_error(id, "%s is already defined", name);
+	if (rb_class_real(RCLASS_SUPER(klass)) != super) {
+	    rb_raise(rb_eTypeError, "superclass mismatch for class "
+		     "%"PRIsVALUE"::%"PRIsVALUE""
+		     " (%"PRIsVALUE" is given but was %"PRIsVALUE")",
+		     outer, rb_id2str(id), RCLASS_SUPER(klass), super);
 	}
 	return klass;
     }
     if (!super) {
-	rb_warn("no super class for `%s::%s', Object assumed",
-		rb_class2name(outer), name);
+	rb_raise(rb_eArgError, "no super class for `%"PRIsVALUE"::%"PRIsVALUE"'",
+		 rb_class_path(outer), rb_id2str(id));
     }
     klass = rb_define_class_id(id, super);
-    rb_set_class_path(klass, outer, name);
+    rb_set_class_path_string(klass, outer, rb_id2str(id));
     rb_const_set(outer, id, klass);
     rb_class_inherited(super, klass);
+    rb_gc_register_mark_object(klass);
 
     return klass;
 }
 
 VALUE
-rb_module_new()
+rb_module_new(void)
 {
-    NEWOBJ(mdl, struct RClass);
-    OBJSETUP(mdl, rb_cModule, T_MODULE);
-
-    mdl->super = 0;
-    mdl->iv_tbl = 0;
-    mdl->m_tbl = 0;
-    mdl->m_tbl = st_init_numtable();
-
+    VALUE mdl = class_alloc(T_MODULE, rb_cModule);
+    RCLASS_M_TBL_INIT(mdl);
     return (VALUE)mdl;
 }
 
 VALUE
-rb_define_module_id(id)
-    ID id;
+rb_define_module_id(ID id)
 {
     VALUE mdl;
 
@@ -296,8 +765,7 @@ rb_define_module_id(id)
 }
 
 VALUE
-rb_define_module(name)
-    const char *name;
+rb_define_module(const char *name)
 {
     VALUE module;
     ID id;
@@ -305,61 +773,73 @@ rb_define_module(name)
     id = rb_intern(name);
     if (rb_const_defined(rb_cObject, id)) {
 	module = rb_const_get(rb_cObject, id);
-	if (TYPE(module) == T_MODULE)
-	    return module;
-	rb_raise(rb_eTypeError, "%s is not a module", rb_obj_classname(module));
+	if (!RB_TYPE_P(module, T_MODULE)) {
+	    rb_raise(rb_eTypeError, "%s is not a module (%"PRIsVALUE")",
+		     name, rb_obj_class(module));
+	}
+	return module;
     }
     module = rb_define_module_id(id);
-    st_add_direct(rb_class_tbl, id, module);
+    rb_vm_add_root_module(id, module);
     rb_const_set(rb_cObject, id, module);
 
     return module;
 }
 
 VALUE
-rb_define_module_under(outer, name)
-    VALUE outer;
-    const char *name;
+rb_define_module_under(VALUE outer, const char *name)
+{
+    return rb_define_module_id_under(outer, rb_intern(name));
+}
+
+VALUE
+rb_define_module_id_under(VALUE outer, ID id)
 {
     VALUE module;
-    ID id;
 
-    id = rb_intern(name);
     if (rb_const_defined_at(outer, id)) {
 	module = rb_const_get_at(outer, id);
-	if (TYPE(module) == T_MODULE)
-	    return module;
-	rb_raise(rb_eTypeError, "%s::%s is not a module",
-		 rb_class2name(outer), rb_obj_classname(module));
+	if (!RB_TYPE_P(module, T_MODULE)) {
+	    rb_raise(rb_eTypeError, "%"PRIsVALUE"::%"PRIsVALUE" is not a module"
+		     " (%"PRIsVALUE")",
+		     outer, rb_id2str(id), rb_obj_class(module));
+	}
+	return module;
     }
     module = rb_define_module_id(id);
     rb_const_set(outer, id, module);
-    rb_set_class_path(module, outer, name);
+    rb_set_class_path_string(module, outer, rb_id2str(id));
+    rb_gc_register_mark_object(module);
 
     return module;
 }
 
-static VALUE
-include_class_new(module, super)
-    VALUE module, super;
+VALUE
+rb_include_class_new(VALUE module, VALUE super)
 {
-    NEWOBJ(klass, struct RClass);
-    OBJSETUP(klass, rb_cClass, T_ICLASS);
+    VALUE klass = class_alloc(T_ICLASS, rb_cClass);
 
     if (BUILTIN_TYPE(module) == T_ICLASS) {
 	module = RBASIC(module)->klass;
     }
-    if (!RCLASS(module)->iv_tbl) {
-	RCLASS(module)->iv_tbl = st_init_numtable();
+    if (!RCLASS_IV_TBL(module)) {
+	RCLASS_IV_TBL(module) = st_init_numtable();
     }
-    klass->iv_tbl = RCLASS(module)->iv_tbl;
-    klass->m_tbl = RCLASS(module)->m_tbl;
-    klass->super = super;
-    if (TYPE(module) == T_ICLASS) {
-	RBASIC(klass)->klass = RBASIC(module)->klass;
+    if (!RCLASS_CONST_TBL(module)) {
+	RCLASS_CONST_TBL(module) = rb_id_table_create(0);
+    }
+    RCLASS_IV_TBL(klass) = RCLASS_IV_TBL(module);
+    RCLASS_CONST_TBL(klass) = RCLASS_CONST_TBL(module);
+
+    RCLASS_M_TBL(OBJ_WB_UNPROTECT(klass)) =
+      RCLASS_M_TBL(OBJ_WB_UNPROTECT(RCLASS_ORIGIN(module))); /* TODO: unprotected? */
+
+    RCLASS_SET_SUPER(klass, super);
+    if (RB_TYPE_P(module, T_ICLASS)) {
+	RBASIC_SET_CLASS(klass, RBASIC(module)->klass);
     }
     else {
-	RBASIC(klass)->klass = module;
+	RBASIC_SET_CLASS(klass, module);
     }
     OBJ_INFECT(klass, module);
     OBJ_INFECT(klass, super);
@@ -367,83 +847,184 @@ include_class_new(module, super)
     return (VALUE)klass;
 }
 
+static int include_modules_at(const VALUE klass, VALUE c, VALUE module, int search_super);
+
+static void
+ensure_includable(VALUE klass, VALUE module)
+{
+    rb_class_modify_check(klass);
+    Check_Type(module, T_MODULE);
+    if (!NIL_P(rb_refinement_module_get_refined_class(module))) {
+	rb_raise(rb_eArgError, "refinement module is not allowed");
+    }
+    OBJ_INFECT(klass, module);
+}
+
 void
-rb_include_module(klass, module)
-    VALUE klass, module;
+rb_include_module(VALUE klass, VALUE module)
 {
-    VALUE p, c;
     int changed = 0;
 
-    rb_frozen_class_p(klass);
-    if (!OBJ_TAINTED(klass)) {
-	rb_secure(4);
-    }
-    
-    if (NIL_P(module)) return;
-    if (klass == module) return;
+    ensure_includable(klass, module);
 
-    if (TYPE(module) != T_MODULE) {
-	Check_Type(module, T_MODULE);
-    }
+    changed = include_modules_at(klass, RCLASS_ORIGIN(klass), module, TRUE);
+    if (changed < 0)
+	rb_raise(rb_eArgError, "cyclic include detected");
+}
+
+static enum rb_id_table_iterator_result
+add_refined_method_entry_i(ID key, VALUE value, void *data)
+{
+    rb_add_refined_method_entry((VALUE)data, key);
+    return ID_TABLE_CONTINUE;
+}
+
+static int
+include_modules_at(const VALUE klass, VALUE c, VALUE module, int search_super)
+{
+    VALUE p, iclass;
+    int method_changed = 0, constant_changed = 0;
+    struct rb_id_table *const klass_m_tbl = RCLASS_M_TBL(RCLASS_ORIGIN(klass));
 
-    OBJ_INFECT(klass, module);
-    c = klass;
     while (module) {
-	int superclass_seen = Qfalse;
+	int superclass_seen = FALSE;
+	struct rb_id_table *tbl;
 
-	if (RCLASS(klass)->m_tbl == RCLASS(module)->m_tbl)
-	    rb_raise(rb_eArgError, "cyclic include detected");
+	if (RCLASS_ORIGIN(module) != module)
+	    goto skip;
+	if (klass_m_tbl && klass_m_tbl == RCLASS_M_TBL(module))
+	    return -1;
 	/* ignore if the module included already in superclasses */
-	for (p = RCLASS(klass)->super; p; p = RCLASS(p)->super) {
-	    switch (BUILTIN_TYPE(p)) {
-	      case T_ICLASS:
-		if (RCLASS(p)->m_tbl == RCLASS(module)->m_tbl) {
+	for (p = RCLASS_SUPER(klass); p; p = RCLASS_SUPER(p)) {
+	    int type = BUILTIN_TYPE(p);
+	    if (type == T_ICLASS) {
+		if (RCLASS_M_TBL(p) == RCLASS_M_TBL(module)) {
 		    if (!superclass_seen) {
-			c = p;	/* move insertion point */
+			c = p;  /* move insertion point */
 		    }
 		    goto skip;
 		}
-		break;
-	      case T_CLASS:
-		superclass_seen = Qtrue;
-		break;
 	    }
+	    else if (type == T_CLASS) {
+		if (!search_super) break;
+		superclass_seen = TRUE;
+	    }
+	}
+	iclass = rb_include_class_new(module, RCLASS_SUPER(c));
+	c = RCLASS_SET_SUPER(c, iclass);
+
+	{
+	    VALUE m = module;
+	    if (BUILTIN_TYPE(m) == T_ICLASS) m = RBASIC(m)->klass;
+	    rb_module_add_to_subclasses_list(m, iclass);
+	}
+
+	if (FL_TEST(klass, RMODULE_IS_REFINEMENT)) {
+	    VALUE refined_class =
+		rb_refinement_module_get_refined_class(klass);
+
+	    rb_id_table_foreach(RMODULE_M_TBL(module), add_refined_method_entry_i, (void *)refined_class);
+	    FL_SET(c, RMODULE_INCLUDED_INTO_REFINEMENT);
 	}
-	c = RCLASS(c)->super = include_class_new(module, RCLASS(c)->super);
-	changed = 1;
+
+	tbl = RMODULE_M_TBL(module);
+	if (tbl && rb_id_table_size(tbl)) method_changed = 1;
+
+	tbl = RMODULE_CONST_TBL(module);
+	if (tbl && rb_id_table_size(tbl)) constant_changed = 1;
       skip:
-	module = RCLASS(module)->super;
+	module = RCLASS_SUPER(module);
+    }
+
+    if (method_changed) rb_clear_method_cache_by_class(klass);
+    if (constant_changed) rb_clear_constant_cache();
+
+    return method_changed;
+}
+
+static enum rb_id_table_iterator_result
+move_refined_method(ID key, VALUE value, void *data)
+{
+    rb_method_entry_t *me = (rb_method_entry_t *) value;
+    VALUE klass = (VALUE)data;
+    struct rb_id_table *tbl = RCLASS_M_TBL(klass);
+
+    if (me->def->type == VM_METHOD_TYPE_REFINED) {
+	if (me->def->body.refined.orig_me) {
+	    const rb_method_entry_t *orig_me = me->def->body.refined.orig_me, *new_me;
+	    RB_OBJ_WRITE(me, &me->def->body.refined.orig_me, NULL);
+	    new_me = rb_method_entry_clone(me);
+	    rb_id_table_insert(tbl, key, (VALUE)new_me);
+	    RB_OBJ_WRITTEN(klass, Qundef, new_me);
+	    rb_method_entry_copy(me, orig_me);
+	    return ID_TABLE_CONTINUE;
+	}
+	else {
+	    rb_id_table_insert(tbl, key, (VALUE)me);
+	    return ID_TABLE_DELETE;
+	}
+    }
+    else {
+	return ID_TABLE_CONTINUE;
+    }
+}
+
+void
+rb_prepend_module(VALUE klass, VALUE module)
+{
+    VALUE origin;
+    int changed = 0;
+
+    ensure_includable(klass, module);
+
+    origin = RCLASS_ORIGIN(klass);
+    if (origin == klass) {
+	origin = class_alloc(T_ICLASS, klass);
+	OBJ_WB_UNPROTECT(origin); /* TODO: conservative shading. Need more survey. */
+	RCLASS_SET_SUPER(origin, RCLASS_SUPER(klass));
+	RCLASS_SET_SUPER(klass, origin);
+	RCLASS_SET_ORIGIN(klass, origin);
+	RCLASS_M_TBL(origin) = RCLASS_M_TBL(klass);
+	RCLASS_M_TBL_INIT(klass);
+	rb_id_table_foreach(RCLASS_M_TBL(origin), move_refined_method, (void *)klass);
+    }
+    changed = include_modules_at(klass, klass, module, FALSE);
+    if (changed < 0)
+	rb_raise(rb_eArgError, "cyclic prepend detected");
+    if (changed) {
+	rb_vm_check_redefinition_by_prepend(klass);
     }
-    if (changed) rb_clear_cache();
 }
 
 /*
  *  call-seq:
  *     mod.included_modules -> array
- *  
+ *
  *  Returns the list of modules included in <i>mod</i>.
- *     
+ *
  *     module Mixin
  *     end
- *     
+ *
  *     module Outer
  *       include Mixin
  *     end
- *     
+ *
  *     Mixin.included_modules   #=> []
  *     Outer.included_modules   #=> [Mixin]
  */
 
 VALUE
-rb_mod_included_modules(mod)
-    VALUE mod;
+rb_mod_included_modules(VALUE mod)
 {
     VALUE ary = rb_ary_new();
     VALUE p;
+    VALUE origin = RCLASS_ORIGIN(mod);
 
-    for (p = RCLASS(mod)->super; p; p = RCLASS(p)->super) {
-	if (BUILTIN_TYPE(p) == T_ICLASS) {
-	    rb_ary_push(ary, RBASIC(p)->klass);
+    for (p = RCLASS_SUPER(mod); p; p = RCLASS_SUPER(p)) {
+	if (p != origin && BUILTIN_TYPE(p) == T_ICLASS) {
+	    VALUE m = RBASIC(p)->klass;
+	    if (RB_TYPE_P(m, T_MODULE))
+		rb_ary_push(ary, m);
 	}
     }
     return ary;
@@ -451,11 +1032,11 @@ rb_mod_included_modules(mod)
 
 /*
  *  call-seq:
- *     mod.include?(module)    => true or false
- *  
+ *     mod.include?(module)    -> true or false
+ *
  *  Returns <code>true</code> if <i>module</i> is included in
  *  <i>mod</i> or one of <i>mod</i>'s ancestors.
- *     
+ *
  *     module A
  *     end
  *     class B
@@ -469,14 +1050,12 @@ rb_mod_included_modules(mod)
  */
 
 VALUE
-rb_mod_include_p(mod, mod2)
-    VALUE mod;
-    VALUE mod2;
+rb_mod_include_p(VALUE mod, VALUE mod2)
 {
     VALUE p;
 
     Check_Type(mod2, T_MODULE);
-    for (p = RCLASS(mod)->super; p; p = RCLASS(p)->super) {
+    for (p = RCLASS_SUPER(mod); p; p = RCLASS_SUPER(p)) {
 	if (BUILTIN_TYPE(p) == T_ICLASS) {
 	    if (RBASIC(p)->klass == mod2) return Qtrue;
 	}
@@ -487,547 +1066,993 @@ rb_mod_include_p(mod, mod2)
 /*
  *  call-seq:
  *     mod.ancestors -> array
- *  
- *  Returns a list of modules included in <i>mod</i> (including
- *  <i>mod</i> itself).
- *     
+ *
+ *  Returns a list of modules included/prepended in <i>mod</i>
+ *  (including <i>mod</i> itself).
+ *
  *     module Mod
  *       include Math
  *       include Comparable
+ *       prepend Enumerable
  *     end
- *     
- *     Mod.ancestors    #=> [Mod, Comparable, Math]
- *     Math.ancestors   #=> [Math]
+ *
+ *     Mod.ancestors        #=> [Enumerable, Mod, Comparable, Math]
+ *     Math.ancestors       #=> [Math]
+ *     Enumerable.ancestors #=> [Enumerable]
  */
 
 VALUE
-rb_mod_ancestors(mod)
-    VALUE mod;
+rb_mod_ancestors(VALUE mod)
 {
     VALUE p, ary = rb_ary_new();
 
-    for (p = mod; p; p = RCLASS(p)->super) {
-	if (FL_TEST(p, FL_SINGLETON))
-	    continue;
+    for (p = mod; p; p = RCLASS_SUPER(p)) {
 	if (BUILTIN_TYPE(p) == T_ICLASS) {
 	    rb_ary_push(ary, RBASIC(p)->klass);
 	}
-	else {
+	else if (p == RCLASS_ORIGIN(p)) {
 	    rb_ary_push(ary, p);
 	}
     }
     return ary;
 }
 
-#define VISI(x) ((x)&NOEX_MASK)
-#define VISI_CHECK(x,f) (VISI(x) == (f))
+static void
+ins_methods_push(st_data_t name, st_data_t ary)
+{
+    rb_ary_push((VALUE)ary, ID2SYM((ID)name));
+}
 
 static int
-ins_methods_push(name, type, ary, visi)
-    ID name;
-    long type;
-    VALUE ary;
-    long visi;
-{
-    if (type == -1) return ST_CONTINUE;
-    switch (visi) {
-      case NOEX_PRIVATE:
-      case NOEX_PROTECTED:
-      case NOEX_PUBLIC:
-	visi = (type == visi);
+ins_methods_i(st_data_t name, st_data_t type, st_data_t ary)
+{
+    switch ((rb_method_visibility_t)type) {
+      case METHOD_VISI_UNDEF:
+      case METHOD_VISI_PRIVATE:
 	break;
-      default:
-	visi = (type != NOEX_PRIVATE);
+      default: /* everything but private */
+	ins_methods_push(name, ary);
 	break;
     }
-    if (visi) {
-	rb_ary_push(ary, rb_str_new2(rb_id2name(name)));
-    }
     return ST_CONTINUE;
 }
 
 static int
-ins_methods_i(name, type, ary)
-    ID name;
-    long type;
-    VALUE ary;
-{
-    return ins_methods_push(name, type, ary, -1); /* everything but private */
-}
-
-static int
-ins_methods_prot_i(name, type, ary)
-    ID name;
-    long type;
-    VALUE ary;
+ins_methods_prot_i(st_data_t name, st_data_t type, st_data_t ary)
 {
-    return ins_methods_push(name, type, ary, NOEX_PROTECTED);
+    if ((rb_method_visibility_t)type == METHOD_VISI_PROTECTED) {
+	ins_methods_push(name, ary);
+    }
+    return ST_CONTINUE;
 }
 
 static int
-ins_methods_priv_i(name, type, ary)
-    ID name;
-    long type;
-    VALUE ary;
+ins_methods_priv_i(st_data_t name, st_data_t type, st_data_t ary)
 {
-    return ins_methods_push(name, type, ary, NOEX_PRIVATE);
+    if ((rb_method_visibility_t)type == METHOD_VISI_PRIVATE) {
+	ins_methods_push(name, ary);
+    }
+    return ST_CONTINUE;
 }
 
 static int
-ins_methods_pub_i(name, type, ary)
-    ID name;
-    long type;
-    VALUE ary;
+ins_methods_pub_i(st_data_t name, st_data_t type, st_data_t ary)
 {
-    return ins_methods_push(name, type, ary, NOEX_PUBLIC);
+    if ((rb_method_visibility_t)type == METHOD_VISI_PUBLIC) {
+	ins_methods_push(name, ary);
+    }
+    return ST_CONTINUE;
 }
 
-static int
-method_entry(key, body, list)
-    ID key;
-    NODE *body;
+struct method_entry_arg {
     st_table *list;
-{
-    long type;
+    int recur;
+};
 
-    if (key == ID_ALLOCATOR) return ST_CONTINUE;
-    if (!st_lookup(list, key, 0)) {
-	if (!body->nd_body) type = -1; /* none */
-	else type = VISI(body->nd_noex);
-	st_add_direct(list, key, type);
+static enum rb_id_table_iterator_result
+method_entry_i(ID key, VALUE value, void *data)
+{
+    const rb_method_entry_t *me = (const rb_method_entry_t *)value;
+    struct method_entry_arg *arg = (struct method_entry_arg *)data;
+    rb_method_visibility_t type;
+
+    if (me->def->type == VM_METHOD_TYPE_REFINED) {
+	VALUE owner = me->owner;
+	me = rb_resolve_refined_method(Qnil, me);
+	if (!me) return ID_TABLE_CONTINUE;
+	if (!arg->recur && me->owner != owner) return ID_TABLE_CONTINUE;
     }
-    return ST_CONTINUE;
+    if (!st_lookup(arg->list, key, 0)) {
+	if (UNDEFINED_METHOD_ENTRY_P(me)) {
+	    type = METHOD_VISI_UNDEF; /* none */
+	}
+	else {
+	    type = METHOD_ENTRY_VISI(me);
+	}
+	st_add_direct(arg->list, key, (st_data_t)type);
+    }
+    return ID_TABLE_CONTINUE;
 }
 
 static VALUE
-class_instance_method_list(argc, argv, mod, func)
-    int argc;
-    VALUE *argv;
-    VALUE mod;
-    int (*func) _((ID, long, VALUE));
+class_instance_method_list(int argc, const VALUE *argv, VALUE mod, int obj, int (*func) (st_data_t, st_data_t, st_data_t))
 {
     VALUE ary;
-    int recur;
-    st_table *list;
+    int recur = TRUE, prepended = 0;
+    struct method_entry_arg me_arg;
 
-    if (argc == 0) {
-	recur = Qtrue;
-    }
-    else {
-	VALUE r;
-	rb_scan_args(argc, argv, "01", &r);
-	recur = RTEST(r);
+    if (rb_check_arity(argc, 0, 1)) recur = RTEST(argv[0]);
+
+    if (!recur && RCLASS_ORIGIN(mod) != mod) {
+	mod = RCLASS_ORIGIN(mod);
+	prepended = 1;
     }
 
-    list = st_init_numtable();
-    for (; mod; mod = RCLASS(mod)->super) {
-	st_foreach(RCLASS(mod)->m_tbl, method_entry, (st_data_t)list);
-	if (BUILTIN_TYPE(mod) == T_ICLASS) continue;
-	if (FL_TEST(mod, FL_SINGLETON)) continue;
+    me_arg.list = st_init_numtable();
+    me_arg.recur = recur;
+    for (; mod; mod = RCLASS_SUPER(mod)) {
+	if (RCLASS_M_TBL(mod)) rb_id_table_foreach(RCLASS_M_TBL(mod), method_entry_i, &me_arg);
+	if (BUILTIN_TYPE(mod) == T_ICLASS && !prepended) continue;
+	if (obj && FL_TEST(mod, FL_SINGLETON)) continue;
 	if (!recur) break;
     }
     ary = rb_ary_new();
-    st_foreach(list, func, ary);
-    st_free_table(list);
+    st_foreach(me_arg.list, func, ary);
+    st_free_table(me_arg.list);
 
     return ary;
 }
 
 /*
  *  call-seq:
- *     mod.instance_methods(include_super=true)   => array
- *  
- *  Returns an array containing the names of public instance methods in
- *  the receiver. For a module, these are the public methods; for a
- *  class, they are the instance (not singleton) methods. With no
- *  argument, or with an argument that is <code>false</code>, the
- *  instance methods in <i>mod</i> are returned, otherwise the methods
- *  in <i>mod</i> and <i>mod</i>'s superclasses are returned.
- *     
+ *     mod.instance_methods(include_super=true)   -> array
+ *
+ *  Returns an array containing the names of the public and protected instance
+ *  methods in the receiver. For a module, these are the public and protected methods;
+ *  for a class, they are the instance (not singleton) methods. If the optional
+ *  parameter is <code>false</code>, the methods of any ancestors are not included.
+ *
  *     module A
  *       def method1()  end
  *     end
  *     class B
+ *       include A
  *       def method2()  end
  *     end
  *     class C < B
  *       def method3()  end
  *     end
- *     
- *     A.instance_methods                #=> ["method1"]
- *     B.instance_methods(false)         #=> ["method2"]
- *     C.instance_methods(false)         #=> ["method3"]
- *     C.instance_methods(true).length   #=> 43
+ *
+ *     A.instance_methods(false)                   #=> [:method1]
+ *     B.instance_methods(false)                   #=> [:method2]
+ *     B.instance_methods(true).include?(:method1) #=> true
+ *     C.instance_methods(false)                   #=> [:method3]
+ *     C.instance_methods.include?(:method2)       #=> true
  */
 
 VALUE
-rb_class_instance_methods(argc, argv, mod)
-    int argc;
-    VALUE *argv;
-    VALUE mod;
+rb_class_instance_methods(int argc, const VALUE *argv, VALUE mod)
 {
-    return class_instance_method_list(argc, argv, mod, ins_methods_i);
+    return class_instance_method_list(argc, argv, mod, 0, ins_methods_i);
 }
 
 /*
  *  call-seq:
- *     mod.protected_instance_methods(include_super=true)   => array
- *  
+ *     mod.protected_instance_methods(include_super=true)   -> array
+ *
  *  Returns a list of the protected instance methods defined in
- *  <i>mod</i>. If the optional parameter is not <code>false</code>, the
- *  methods of any ancestors are included.
+ *  <i>mod</i>. If the optional parameter is <code>false</code>, the
+ *  methods of any ancestors are not included.
  */
 
 VALUE
-rb_class_protected_instance_methods(argc, argv, mod)
-    int argc;
-    VALUE *argv;
-    VALUE mod;
+rb_class_protected_instance_methods(int argc, const VALUE *argv, VALUE mod)
 {
-    return class_instance_method_list(argc, argv, mod, ins_methods_prot_i);
+    return class_instance_method_list(argc, argv, mod, 0, ins_methods_prot_i);
 }
 
 /*
  *  call-seq:
- *     mod.private_instance_methods(include_super=true)    => array
- *  
+ *     mod.private_instance_methods(include_super=true)    -> array
+ *
  *  Returns a list of the private instance methods defined in
- *  <i>mod</i>. If the optional parameter is not <code>false</code>, the
- *  methods of any ancestors are included.
- *     
+ *  <i>mod</i>. If the optional parameter is <code>false</code>, the
+ *  methods of any ancestors are not included.
+ *
  *     module Mod
  *       def method1()  end
  *       private :method1
  *       def method2()  end
  *     end
- *     Mod.instance_methods           #=> ["method2"]
- *     Mod.private_instance_methods   #=> ["method1"]
+ *     Mod.instance_methods           #=> [:method2]
+ *     Mod.private_instance_methods   #=> [:method1]
  */
 
 VALUE
-rb_class_private_instance_methods(argc, argv, mod)
-    int argc;
-    VALUE *argv;
-    VALUE mod;
+rb_class_private_instance_methods(int argc, const VALUE *argv, VALUE mod)
 {
-    return class_instance_method_list(argc, argv, mod, ins_methods_priv_i);
+    return class_instance_method_list(argc, argv, mod, 0, ins_methods_priv_i);
 }
 
 /*
  *  call-seq:
- *     mod.public_instance_methods(include_super=true)   => array
- *  
+ *     mod.public_instance_methods(include_super=true)   -> array
+ *
  *  Returns a list of the public instance methods defined in <i>mod</i>.
- *  If the optional parameter is not <code>false</code>, the methods of
- *  any ancestors are included.
+ *  If the optional parameter is <code>false</code>, the methods of
+ *  any ancestors are not included.
+ */
+
+VALUE
+rb_class_public_instance_methods(int argc, const VALUE *argv, VALUE mod)
+{
+    return class_instance_method_list(argc, argv, mod, 0, ins_methods_pub_i);
+}
+
+/*
+ *  call-seq:
+ *     obj.methods(regular=true)    -> array
+ *
+ *  Returns a list of the names of public and protected methods of
+ *  <i>obj</i>. This will include all the methods accessible in
+ *  <i>obj</i>'s ancestors.
+ *  If the optional parameter is <code>false</code>, it
+ *  returns an array of <i>obj<i>'s public and protected singleton methods,
+ *  the array will not include methods in modules included in <i>obj</i>.
+ *
+ *     class Klass
+ *       def klass_method()
+ *       end
+ *     end
+ *     k = Klass.new
+ *     k.methods[0..9]    #=> [:klass_method, :nil?, :===,
+ *                        #    :==~, :!, :eql?
+ *                        #    :hash, :<=>, :class, :singleton_class]
+ *     k.methods.length   #=> 56
+ *
+ *     k.methods(false)   #=> []
+ *     def k.singleton_method; end
+ *     k.methods(false)   #=> [:singleton_method]
+ *
+ *     module M123; def m123; end end
+ *     k.extend M123
+ *     k.methods(false)   #=> [:singleton_method]
+ */
+
+VALUE
+rb_obj_methods(int argc, const VALUE *argv, VALUE obj)
+{
+    rb_check_arity(argc, 0, 1);
+    if (argc > 0 && !RTEST(argv[0])) {
+	return rb_obj_singleton_methods(argc, argv, obj);
+    }
+    return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_i);
+}
+
+/*
+ *  call-seq:
+ *     obj.protected_methods(all=true)   -> array
+ *
+ *  Returns the list of protected methods accessible to <i>obj</i>. If
+ *  the <i>all</i> parameter is set to <code>false</code>, only those methods
+ *  in the receiver will be listed.
+ */
+
+VALUE
+rb_obj_protected_methods(int argc, const VALUE *argv, VALUE obj)
+{
+    return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_prot_i);
+}
+
+/*
+ *  call-seq:
+ *     obj.private_methods(all=true)   -> array
+ *
+ *  Returns the list of private methods accessible to <i>obj</i>. If
+ *  the <i>all</i> parameter is set to <code>false</code>, only those methods
+ *  in the receiver will be listed.
+ */
+
+VALUE
+rb_obj_private_methods(int argc, const VALUE *argv, VALUE obj)
+{
+    return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_priv_i);
+}
+
+/*
+ *  call-seq:
+ *     obj.public_methods(all=true)   -> array
+ *
+ *  Returns the list of public methods accessible to <i>obj</i>. If
+ *  the <i>all</i> parameter is set to <code>false</code>, only those methods
+ *  in the receiver will be listed.
  */
 
 VALUE
-rb_class_public_instance_methods(argc, argv, mod)
-    int argc;
-    VALUE *argv;
-    VALUE mod;
+rb_obj_public_methods(int argc, const VALUE *argv, VALUE obj)
 {
-    return class_instance_method_list(argc, argv, mod, ins_methods_pub_i);
+    return class_instance_method_list(argc, argv, CLASS_OF(obj), 1, ins_methods_pub_i);
 }
 
 /*
  *  call-seq:
- *     obj.singleton_methods(all=true)    => array
- *  
+ *     obj.singleton_methods(all=true)    -> array
+ *
  *  Returns an array of the names of singleton methods for <i>obj</i>.
  *  If the optional <i>all</i> parameter is true, the list will include
  *  methods in modules included in <i>obj</i>.
- *     
+ *  Only public and protected singleton methods are returned.
+ *
  *     module Other
  *       def three() end
  *     end
- *     
+ *
  *     class Single
  *       def Single.four() end
  *     end
- *     
+ *
  *     a = Single.new
- *     
+ *
  *     def a.one()
  *     end
- *     
+ *
  *     class << a
  *       include Other
  *       def two()
  *       end
  *     end
- *     
- *     Single.singleton_methods    #=> ["four"]
- *     a.singleton_methods(false)  #=> ["two", "one"]
- *     a.singleton_methods         #=> ["two", "one", "three"]
+ *
+ *     Single.singleton_methods    #=> [:four]
+ *     a.singleton_methods(false)  #=> [:two, :one]
+ *     a.singleton_methods         #=> [:two, :one, :three]
  */
 
 VALUE
-rb_obj_singleton_methods(argc, argv, obj)
-    int argc;
-    VALUE *argv;
-    VALUE obj;
+rb_obj_singleton_methods(int argc, const VALUE *argv, VALUE obj)
 {
-    VALUE recur, ary, klass;
-    st_table *list;
+    VALUE ary, klass, origin;
+    struct method_entry_arg me_arg;
+    struct rb_id_table *mtbl;
+    int recur = TRUE;
 
-    rb_scan_args(argc, argv, "01", &recur);
-    if (argc == 0) {
-	recur = Qtrue;
-    }
+    if (rb_check_arity(argc, 0, 1)) recur = RTEST(argv[0]);
     klass = CLASS_OF(obj);
-    list = st_init_numtable();
+    origin = RCLASS_ORIGIN(klass);
+    me_arg.list = st_init_numtable();
+    me_arg.recur = recur;
     if (klass && FL_TEST(klass, FL_SINGLETON)) {
-	st_foreach(RCLASS(klass)->m_tbl, method_entry, (st_data_t)list);
-	klass = RCLASS(klass)->super;
+	if ((mtbl = RCLASS_M_TBL(origin)) != 0) rb_id_table_foreach(mtbl, method_entry_i, &me_arg);
+	klass = RCLASS_SUPER(klass);
     }
-    if (RTEST(recur)) {
-	while (klass && (FL_TEST(klass, FL_SINGLETON) || TYPE(klass) == T_ICLASS)) {
-	    st_foreach(RCLASS(klass)->m_tbl, method_entry, (st_data_t)list);
-	    klass = RCLASS(klass)->super;
+    if (recur) {
+	while (klass && (FL_TEST(klass, FL_SINGLETON) || RB_TYPE_P(klass, T_ICLASS))) {
+	    if (klass != origin && (mtbl = RCLASS_M_TBL(klass)) != 0) rb_id_table_foreach(mtbl, method_entry_i, &me_arg);
+	    klass = RCLASS_SUPER(klass);
 	}
     }
     ary = rb_ary_new();
-    st_foreach(list, ins_methods_i, ary);
-    st_free_table(list);
+    st_foreach(me_arg.list, ins_methods_i, ary);
+    st_free_table(me_arg.list);
 
     return ary;
 }
 
+/*!
+ * \}
+ */
+/*!
+ * \defgroup defmethod Defining methods
+ * There are some APIs to define a method from C.
+ * These API takes a C function as a method body.
+ *
+ * \par Method body functions
+ * Method body functions must return a VALUE and
+ * can be one of the following form:
+ * <dl>
+ * <dt>Fixed number of parameters</dt>
+ * <dd>
+ *     This form is a normal C function, excepting it takes
+ *     a receiver object as the first argument.
+ *
+ *     \code
+ *     static VALUE my_method(VALUE self, VALUE x, VALUE y);
+ *     \endcode
+ * </dd>
+ * <dt>argc and argv style</dt>
+ * <dd>
+ *     This form takes three parameters: \a argc, \a argv and \a self.
+ *     \a self is the receiver. \a argc is the number of arguments.
+ *     \a argv is a pointer to an array of the arguments.
+ *
+ *     \code
+ *     static VALUE my_method(int argc, VALUE *argv, VALUE self);
+ *     \endcode
+ * </dd>
+ * <dt>Ruby array style</dt>
+ * <dd>
+ *     This form takes two parameters: self and args.
+ *     \a self is the receiver. \a args is an Array object which
+ *     contains the arguments.
+ *
+ *     \code
+ *     static VALUE my_method(VALUE self, VALUE args);
+ *     \endcode
+ * </dd>
+ *
+ * \par Number of parameters
+ * Method defining APIs takes the number of parameters which the
+ * method will takes. This number is called \a argc.
+ * \a argc can be:
+ * <dl>
+ * <dt>zero or positive number</dt>
+ * <dd>This means the method body function takes a fixed number of parameters</dd>
+ * <dt>-1</dt>
+ * <dd>This means the method body function is "argc and argv" style.</dd>
+ * <dt>-2</dt>
+ * <dd>This means the method body function is "self and args" style.</dd>
+ * </dl>
+ * \{
+ */
+
 void
-rb_define_method_id(klass, name, func, argc)
-    VALUE klass;
-    ID name;
-    VALUE (*func)();
-    int argc;
+rb_define_method_id(VALUE klass, ID mid, VALUE (*func)(ANYARGS), int argc)
 {
-    rb_add_method(klass, name, NEW_CFUNC(func,argc), NOEX_PUBLIC);
+    rb_add_method_cfunc(klass, mid, func, argc, METHOD_VISI_PUBLIC);
 }
 
 void
-rb_define_method(klass, name, func, argc)
-    VALUE klass;
-    const char *name;
-    VALUE (*func)();
-    int argc;
+rb_define_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc)
 {
-    ID id = rb_intern(name);
-    int ex = NOEX_PUBLIC;
-
+    rb_add_method_cfunc(klass, rb_intern(name), func, argc, METHOD_VISI_PUBLIC);
+}
 
-    rb_add_method(klass, id, NEW_CFUNC(func, argc), ex);
+void
+rb_define_protected_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc)
+{
+    rb_add_method_cfunc(klass, rb_intern(name), func, argc, METHOD_VISI_PROTECTED);
 }
 
 void
-rb_define_protected_method(klass, name, func, argc)
-    VALUE klass;
-    const char *name;
-    VALUE (*func)();
-    int argc;
+rb_define_private_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc)
 {
-    rb_add_method(klass, rb_intern(name), NEW_CFUNC(func, argc), NOEX_PROTECTED);
+    rb_add_method_cfunc(klass, rb_intern(name), func, argc, METHOD_VISI_PRIVATE);
 }
 
 void
-rb_define_private_method(klass, name, func, argc)
-    VALUE klass;
-    const char *name;
-    VALUE (*func)();
-    int argc;
+rb_undef_method(VALUE klass, const char *name)
 {
-    rb_add_method(klass, rb_intern(name), NEW_CFUNC(func, argc), NOEX_PRIVATE);
+    rb_add_method(klass, rb_intern(name), VM_METHOD_TYPE_UNDEF, 0, METHOD_VISI_UNDEF);
+}
+
+static enum rb_id_table_iterator_result
+undef_method_i(ID name, VALUE value, void *data)
+{
+    VALUE klass = (VALUE)data;
+    rb_add_method(klass, name, VM_METHOD_TYPE_UNDEF, 0, METHOD_VISI_UNDEF);
+    return ID_TABLE_CONTINUE;
 }
 
 void
-rb_undef_method(klass, name)
-    VALUE klass;
-    const char *name;
+rb_undef_methods_from(VALUE klass, VALUE super)
 {
-    rb_add_method(klass, rb_intern(name), 0, NOEX_UNDEF);
+    struct rb_id_table *mtbl = RCLASS_M_TBL(super);
+    if (mtbl) {
+	rb_id_table_foreach(mtbl, undef_method_i, (void *)klass);
+    }
 }
 
+/*!
+ * \}
+ */
+/*!
+ * \addtogroup class
+ * \{
+ */
+
 #define SPECIAL_SINGLETON(x,c) do {\
     if (obj == (x)) {\
-	return c;\
+	return (c);\
     }\
 } while (0)
 
+static inline VALUE
+special_singleton_class_of(VALUE obj)
+{
+    SPECIAL_SINGLETON(Qnil, rb_cNilClass);
+    SPECIAL_SINGLETON(Qfalse, rb_cFalseClass);
+    SPECIAL_SINGLETON(Qtrue, rb_cTrueClass);
+    return Qnil;
+}
+
 VALUE
-rb_singleton_class(obj)
-    VALUE obj;
+rb_special_singleton_class(VALUE obj)
+{
+    return special_singleton_class_of(obj);
+}
+
+/*!
+ * \internal
+ * Returns the singleton class of \a obj. Creates it if necessary.
+ *
+ * \note DO NOT expose the returned singleton class to
+ *       outside of class.c.
+ *       Use \ref rb_singleton_class instead for
+ *       consistency of the metaclass hierarchy.
+ */
+static VALUE
+singleton_class_of(VALUE obj)
 {
     VALUE klass;
 
-    if (FIXNUM_P(obj) || SYMBOL_P(obj)) {
+    if (FIXNUM_P(obj) || FLONUM_P(obj) || STATIC_SYM_P(obj)) {
+      no_singleton:
 	rb_raise(rb_eTypeError, "can't define singleton");
     }
-    if (rb_special_const_p(obj)) {
-	SPECIAL_SINGLETON(Qnil, rb_cNilClass);
-	SPECIAL_SINGLETON(Qfalse, rb_cFalseClass);
-	SPECIAL_SINGLETON(Qtrue, rb_cTrueClass);
-	rb_bug("unknown immediate %ld", obj);
-    }
-
-    DEFER_INTS;
-    if (FL_TEST(RBASIC(obj)->klass, FL_SINGLETON) &&
-	rb_iv_get(RBASIC(obj)->klass, "__attached__") == obj) {
-	klass = RBASIC(obj)->klass;
+    if (SPECIAL_CONST_P(obj)) {
+	klass = special_singleton_class_of(obj);
+	if (NIL_P(klass))
+	    rb_bug("unknown immediate %p", (void *)obj);
+	return klass;
     }
     else {
-	klass = rb_make_metaclass(obj, RBASIC(obj)->klass);
+	switch (BUILTIN_TYPE(obj)) {
+	  case T_FLOAT: case T_BIGNUM: case T_SYMBOL:
+	    goto no_singleton;
+	  case T_STRING:
+	    if (FL_TEST_RAW(obj, RSTRING_FSTR)) goto no_singleton;
+	    break;
+	}
     }
+
+    klass = RBASIC(obj)->klass;
+    if (!(FL_TEST(klass, FL_SINGLETON) &&
+	  rb_ivar_get(klass, id_attached) == obj)) {
+	rb_serial_t serial = RCLASS_SERIAL(klass);
+	klass = rb_make_metaclass(obj, klass);
+	RCLASS_SERIAL(klass) = serial;
+    }
+
     if (OBJ_TAINTED(obj)) {
 	OBJ_TAINT(klass);
     }
     else {
 	FL_UNSET(klass, FL_TAINT);
     }
-    if (OBJ_FROZEN(obj)) OBJ_FREEZE(klass);
-    ALLOW_INTS;
+    RB_FL_SET_RAW(klass, RB_OBJ_FROZEN_RAW(obj));
 
     return klass;
 }
 
 void
-rb_define_singleton_method(obj, name, func, argc)
-    VALUE obj;
-    const char *name;
-    VALUE (*func)();
-    int argc;
+rb_freeze_singleton_class(VALUE x)
 {
-    rb_define_method(rb_singleton_class(obj), name, func, argc);
+    /* should not propagate to meta-meta-class, and so on */
+    if (!(RBASIC(x)->flags & FL_SINGLETON)) {
+	VALUE klass = RBASIC_CLASS(x);
+	if (klass && (klass = RCLASS_ORIGIN(klass)) != 0 &&
+	    FL_TEST(klass, (FL_SINGLETON|FL_FREEZE)) == FL_SINGLETON) {
+	    OBJ_FREEZE_RAW(klass);
+	}
+    }
 }
 
+/*!
+ * Returns the singleton class of \a obj, or nil if obj is not a
+ * singleton object.
+ *
+ * \param obj an arbitrary object.
+ * \return the singleton class or nil.
+ */
+VALUE
+rb_singleton_class_get(VALUE obj)
+{
+    VALUE klass;
+
+    if (SPECIAL_CONST_P(obj)) {
+	return rb_special_singleton_class(obj);
+    }
+    klass = RBASIC(obj)->klass;
+    if (!FL_TEST(klass, FL_SINGLETON)) return Qnil;
+    if (rb_ivar_get(klass, id_attached) != obj) return Qnil;
+    return klass;
+}
+
+/*!
+ * Returns the singleton class of \a obj. Creates it if necessary.
+ *
+ * \param obj an arbitrary object.
+ * \throw TypeError if \a obj is a Integer or a Symbol.
+ * \return the singleton class.
+ *
+ * \post \a obj has its own singleton class.
+ * \post if \a obj is a class,
+ *       the returned singleton class also has its own
+ *       singleton class in order to keep consistency of the
+ *       inheritance structure of metaclasses.
+ * \note a new singleton class will be created
+ *       if \a obj does not have it.
+ * \note the singleton classes for nil, true and false are:
+ *       NilClass, TrueClass and FalseClass.
+ */
+VALUE
+rb_singleton_class(VALUE obj)
+{
+    VALUE klass = singleton_class_of(obj);
+
+    /* ensures an exposed class belongs to its own eigenclass */
+    if (RB_TYPE_P(obj, T_CLASS)) (void)ENSURE_EIGENCLASS(klass);
+
+    return klass;
+}
+
+/*!
+ * \}
+ */
+
+/*!
+ * \addtogroup defmethod
+ * \{
+ */
+
+/*!
+ * Defines a singleton method for \a obj.
+ * \param obj    an arbitrary object
+ * \param name   name of the singleton method
+ * \param func   the method body
+ * \param argc   the number of parameters, or -1 or -2. see \ref defmethod.
+ */
 void
-rb_define_module_function(module, name, func, argc)
-    VALUE module;
-    const char *name;
-    VALUE (*func)();
-    int argc;
+rb_define_singleton_method(VALUE obj, const char *name, VALUE (*func)(ANYARGS), int argc)
+{
+    rb_define_method(singleton_class_of(obj), name, func, argc);
+}
+
+
+
+/*!
+ * Defines a module function for \a module.
+ * \param module  an module or a class.
+ * \param name    name of the function
+ * \param func    the method body
+ * \param argc    the number of parameters, or -1 or -2. see \ref defmethod.
+ */
+void
+rb_define_module_function(VALUE module, const char *name, VALUE (*func)(ANYARGS), int argc)
 {
     rb_define_private_method(module, name, func, argc);
     rb_define_singleton_method(module, name, func, argc);
 }
 
+
+/*!
+ * Defines a global function
+ * \param name    name of the function
+ * \param func    the method body
+ * \param argc    the number of parameters, or -1 or -2. see \ref defmethod.
+ */
 void
-rb_define_global_function(name, func, argc)
-    const char *name;
-    VALUE (*func)();
-    int argc;
+rb_define_global_function(const char *name, VALUE (*func)(ANYARGS), int argc)
 {
     rb_define_module_function(rb_mKernel, name, func, argc);
 }
 
+
+/*!
+ * Defines an alias of a method.
+ * \param klass  the class which the original method belongs to
+ * \param name1  a new name for the method
+ * \param name2  the original name of the method
+ */
 void
-rb_define_alias(klass, name1, name2)
-    VALUE klass;
-    const char *name1, *name2;
+rb_define_alias(VALUE klass, const char *name1, const char *name2)
 {
     rb_alias(klass, rb_intern(name1), rb_intern(name2));
 }
 
+/*!
+ * Defines (a) public accessor method(s) for an attribute.
+ * \param klass  the class which the attribute will belongs to
+ * \param name   name of the attribute
+ * \param read   a getter method for the attribute will be defined if \a read is non-zero.
+ * \param write  a setter method for the attribute will be defined if \a write is non-zero.
+ */
 void
-rb_define_attr(klass, name, read, write)
-    VALUE klass;
-    const char *name;
-    int read, write;
+rb_define_attr(VALUE klass, const char *name, int read, int write)
+{
+    rb_attr(klass, rb_intern(name), read, write, FALSE);
+}
+
+MJIT_FUNC_EXPORTED VALUE
+rb_keyword_error_new(const char *error, VALUE keys)
+{
+    long i = 0, len = RARRAY_LEN(keys);
+    VALUE error_message = rb_sprintf("%s keyword%.*s", error, len > 1, "s");
+
+    if (len > 0) {
+	rb_str_cat_cstr(error_message, ": ");
+	while (1) {
+            const VALUE k = RARRAY_AREF(keys, i);
+	    Check_Type(k, T_SYMBOL); /* wrong hash is given to rb_get_kwargs */
+	    rb_str_append(error_message, rb_sym2str(k));
+	    if (++i >= len) break;
+	    rb_str_cat_cstr(error_message, ", ");
+	}
+    }
+
+    return rb_exc_new_str(rb_eArgError, error_message);
+}
+
+NORETURN(static void rb_keyword_error(const char *error, VALUE keys));
+static void
+rb_keyword_error(const char *error, VALUE keys)
+{
+    rb_exc_raise(rb_keyword_error_new(error, keys));
+}
+
+NORETURN(static void unknown_keyword_error(VALUE hash, const ID *table, int keywords));
+static void
+unknown_keyword_error(VALUE hash, const ID *table, int keywords)
+{
+    int i;
+    for (i = 0; i < keywords; i++) {
+	st_data_t key = ID2SYM(table[i]);
+        rb_hash_stlike_delete(hash, &key, NULL);
+    }
+    rb_keyword_error("unknown", rb_hash_keys(hash));
+}
+
+struct extract_keywords {
+    VALUE kwdhash, nonsymkey;
+};
+
+static int
+separate_symbol(st_data_t key, st_data_t value, st_data_t arg)
 {
-    rb_attr(klass, rb_intern(name), read, write, Qfalse);
+    struct extract_keywords *argp = (struct extract_keywords *)arg;
+    VALUE k = (VALUE)key, v = (VALUE)value;
+
+    if (argp->kwdhash) {
+        if (UNLIKELY(!SYMBOL_P(k))) {
+            argp->nonsymkey = k;
+            return ST_STOP;
+        }
+    }
+    else if (SYMBOL_P(k)) {
+        if (UNLIKELY(argp->nonsymkey != Qundef)) {
+            argp->kwdhash = Qnil;
+            return ST_STOP;
+        }
+        argp->kwdhash = rb_hash_new();
+    }
+    else {
+        if (argp->nonsymkey == Qundef)
+            argp->nonsymkey = k;
+        return ST_CONTINUE;
+    }
+    rb_hash_aset(argp->kwdhash, k, v);
+    return ST_CONTINUE;
 }
 
-#ifdef HAVE_STDARG_PROTOTYPES
-#include <stdarg.h>
-#define va_init_list(a,b) va_start(a,b)
-#else
-#include <varargs.h>
-#define va_init_list(a,b) va_start(a)
-#endif
+VALUE
+rb_extract_keywords(VALUE *orighash)
+{
+    struct extract_keywords arg = {0, Qundef};
+    VALUE hash = *orighash;
+
+    if (RHASH_EMPTY_P(hash)) {
+	*orighash = 0;
+	return hash;
+    }
+    rb_hash_foreach(hash, separate_symbol, (st_data_t)&arg);
+    if (arg.kwdhash) {
+        if (arg.nonsymkey != Qundef) {
+            rb_raise(rb_eArgError, "non-symbol key in keyword arguments: %+"PRIsVALUE,
+                     arg.nonsymkey);
+        }
+        *orighash = 0;
+    }
+    return arg.kwdhash;
+}
 
 int
-#ifdef HAVE_STDARG_PROTOTYPES
+rb_get_kwargs(VALUE keyword_hash, const ID *table, int required, int optional, VALUE *values)
+{
+    int i = 0, j;
+    int rest = 0;
+    VALUE missing = Qnil;
+    st_data_t key;
+
+#define extract_kwarg(keyword, val) \
+    (key = (st_data_t)(keyword), values ? \
+     rb_hash_stlike_delete(keyword_hash, &key, (val)) : \
+     rb_hash_stlike_lookup(keyword_hash, key, (val)))
+
+    if (NIL_P(keyword_hash)) keyword_hash = 0;
+
+    if (optional < 0) {
+	rest = 1;
+	optional = -1-optional;
+    }
+    if (values) {
+	for (j = 0; j < required + optional; j++) {
+	    values[j] = Qundef;
+	}
+    }
+    if (required) {
+	for (; i < required; i++) {
+	    VALUE keyword = ID2SYM(table[i]);
+	    if (keyword_hash) {
+		st_data_t val;
+		if (extract_kwarg(keyword, &val)) {
+		    if (values) values[i] = (VALUE)val;
+		    continue;
+		}
+	    }
+	    if (NIL_P(missing)) missing = rb_ary_tmp_new(1);
+	    rb_ary_push(missing, keyword);
+	}
+	if (!NIL_P(missing)) {
+	    rb_keyword_error("missing", missing);
+	}
+    }
+    j = i;
+    if (optional && keyword_hash) {
+	for (i = 0; i < optional; i++) {
+	    st_data_t val;
+	    if (extract_kwarg(ID2SYM(table[required+i]), &val)) {
+		if (values) values[required+i] = (VALUE)val;
+		j++;
+	    }
+	}
+    }
+    if (!rest && keyword_hash) {
+	if (RHASH_SIZE(keyword_hash) > (unsigned int)(values ? 0 : j)) {
+	    unknown_keyword_error(keyword_hash, table, required+optional);
+	}
+    }
+    return j;
+#undef extract_kwarg
+}
+
+#undef rb_scan_args
+int
 rb_scan_args(int argc, const VALUE *argv, const char *fmt, ...)
-#else
-rb_scan_args(argc, argv, fmt, va_alist)
-    int argc;
-    const VALUE *argv;
-    const char *fmt;
-    va_dcl
-#endif
-{
-    int n, i = 0;
+{
+    int i;
     const char *p = fmt;
     VALUE *var;
     va_list vargs;
-
-    va_init_list(vargs, fmt);
-
-    if (*p == '*') goto rest_arg;
+    int f_var = 0, f_hash = 0, f_block = 0;
+    int n_lead = 0, n_opt = 0, n_trail = 0, n_mand;
+    int argi = 0, last_idx = -1;
+    VALUE hash = Qnil, last_hash = 0;
 
     if (ISDIGIT(*p)) {
-	n = *p - '0';
-	if (n > argc)
-	    rb_raise(rb_eArgError, "wrong number of arguments (%d for %d)", argc, n);
-	for (i=0; i<n; i++) {
-	    var = va_arg(vargs, VALUE*);
-	    if (var) *var = argv[i];
+	n_lead = *p - '0';
+	p++;
+	if (ISDIGIT(*p)) {
+	    n_opt = *p - '0';
+	    p++;
 	}
+    }
+    if (*p == '*') {
+	f_var = 1;
 	p++;
     }
-    else {
-	goto error;
+    if (ISDIGIT(*p)) {
+	n_trail = *p - '0';
+	p++;
+    }
+    if (*p == ':') {
+	f_hash = 1;
+	p++;
+    }
+    if (*p == '&') {
+	f_block = 1;
+	p++;
     }
+    if (*p != '\0') {
+	rb_fatal("bad scan arg format: %s", fmt);
+    }
+    n_mand = n_lead + n_trail;
 
-    if (ISDIGIT(*p)) {
-	n = i + *p - '0';
-	for (; i<n; i++) {
-	    var = va_arg(vargs, VALUE*);
-	    if (argc > i) {
-		if (var) *var = argv[i];
-	    }
-	    else {
-		if (var) *var = Qnil;
+    if (argc < n_mand)
+	goto argc_error;
+
+    va_start(vargs, fmt);
+
+    /* capture an option hash - phase 1: pop */
+    if (f_hash && n_mand < argc) {
+	VALUE last = argv[argc - 1];
+
+	if (NIL_P(last)) {
+	    /* nil is taken as an empty option hash only if it is not
+	       ambiguous; i.e. '*' is not specified and arguments are
+	       given more than sufficient */
+	    if (!f_var && n_mand + n_opt < argc)
+		argc--;
+	}
+	else {
+	    hash = rb_check_hash_type(last);
+	    if (!NIL_P(hash)) {
+		VALUE opts = rb_extract_keywords(&hash);
+		if (!(last_hash = hash)) argc--;
+		else last_idx = argc - 1;
+		hash = opts ? opts : Qnil;
 	    }
 	}
-	p++;
     }
-
-    if(*p == '*') {
-      rest_arg:
-	var = va_arg(vargs, VALUE*);
-	if (argc > i) {
-	    if (var) *var = rb_ary_new4(argc-i, argv+i);
-	    i = argc;
+    /* capture leading mandatory arguments */
+    for (i = n_lead; i-- > 0; ) {
+	var = va_arg(vargs, VALUE *);
+	if (var) *var = (argi == last_idx) ? last_hash : argv[argi];
+	argi++;
+    }
+    /* capture optional arguments */
+    for (i = n_opt; i-- > 0; ) {
+	var = va_arg(vargs, VALUE *);
+	if (argi < argc - n_trail) {
+	    if (var) *var = (argi == last_idx) ? last_hash : argv[argi];
+	    argi++;
+	}
+	else {
+	    if (var) *var = Qnil;
+	}
+    }
+    /* capture variable length arguments */
+    if (f_var) {
+	int n_var = argc - argi - n_trail;
+
+	var = va_arg(vargs, VALUE *);
+	if (0 < n_var) {
+	    if (var) {
+		int f_last = (last_idx + 1 == argc - n_trail);
+		*var = rb_ary_new4(n_var-f_last, &argv[argi]);
+		if (f_last) rb_ary_push(*var, last_hash);
+	    }
+	    argi += n_var;
 	}
 	else {
 	    if (var) *var = rb_ary_new();
 	}
-	p++;
     }
-
-    if (*p == '&') {
-	var = va_arg(vargs, VALUE*);
+    /* capture trailing mandatory arguments */
+    for (i = n_trail; i-- > 0; ) {
+	var = va_arg(vargs, VALUE *);
+	if (var) *var = (argi == last_idx) ? last_hash : argv[argi];
+	argi++;
+    }
+    /* capture an option hash - phase 2: assignment */
+    if (f_hash) {
+	var = va_arg(vargs, VALUE *);
+	if (var) *var = hash;
+    }
+    /* capture iterator block */
+    if (f_block) {
+	var = va_arg(vargs, VALUE *);
 	if (rb_block_given_p()) {
 	    *var = rb_block_proc();
 	}
 	else {
 	    *var = Qnil;
 	}
-	p++;
     }
     va_end(vargs);
 
-    if (*p != '\0') {
-	goto error;
-    }
-
-    if (argc > i) {
-	rb_raise(rb_eArgError, "wrong number of arguments (%d for %d)", argc, i);
+    if (argi < argc) {
+      argc_error:
+	rb_error_arity(argc, n_mand, f_var ? UNLIMITED_ARGUMENTS : n_mand + n_opt);
     }
 
     return argc;
+}
 
-  error:
-    rb_fatal("bad scan arg format: %s", fmt);
-    return 0;
+int
+rb_class_has_methods(VALUE c)
+{
+    return rb_id_table_size(RCLASS_M_TBL(c)) == 0 ? FALSE : TRUE;
 }
+
+/*!
+ * \}
+ */
diff --git a/common.mk b/common.mk
index 3584b83e35..992fb2325e 100644
--- a/common.mk
+++ b/common.mk
@@ -1,26 +1,93 @@
 bin: $(PROGRAM) $(WPROGRAM)
-lib: $(LIBRUBY);
-dll: $(LIBRUBY_SO);
+lib: $(LIBRUBY)
+dll: $(LIBRUBY_SO)
 
-RUBYOPT       =
+.SUFFIXES: .inc .h .c .y .i .$(ASMEXT) .$(DTRACE_EXT)
+
+# V=0 quiet, V=1 verbose.  other values don't work.
+V = 0
+Q1 = $(V:1=)
+Q = $(Q1:0=@)
+ECHO0 = $(ECHO1:0=echo)
+ECHO = @$(ECHO0)
+
+mflags = $(MFLAGS)
+gnumake_recursive =
+enable_shared = $(ENABLE_SHARED:no=)
+
+UNICODE_VERSION = 11.0.0
+UNICODE_EMOJI_VERSION = 11.0
+
+### set the following environment variable or uncomment the line if
+### the Unicode data files should be updated completely on every update ('make up',...).
+# ALWAYS_UPDATE_UNICODE = yes
+UNICODE_DATA_DIR = enc/unicode/data/$(UNICODE_VERSION)/ucd
+UNICODE_SRC_DATA_DIR = $(srcdir)/$(UNICODE_DATA_DIR)
+UNICODE_SRC_EMOJI_DATA_DIR = $(srcdir)/enc/unicode/data/emoji/$(UNICODE_EMOJI_VERSION)
+UNICODE_HDR_DIR = $(srcdir)/enc/unicode/$(UNICODE_VERSION)
+UNICODE_DATA_HEADERS = \
+	$(UNICODE_HDR_DIR)/casefold.h \
+	$(UNICODE_HDR_DIR)/name2ctype.h \
+	$(empty)
+
+RUBY_RELEASE_DATE = $(RUBY_RELEASE_YEAR)-$(RUBY_RELEASE_MONTH)-$(RUBY_RELEASE_DAY)
+RUBYLIB       = $(PATH_SEPARATOR)
+RUBYOPT       = -
+RUN_OPTS      = --disable-gems
+
+INCFLAGS = -I. -I$(arch_hdrdir) -I$(hdrdir) -I$(srcdir) -I$(UNICODE_HDR_DIR)
+
+GEM_HOME =
+GEM_PATH =
+GEM_VENDOR =
+
+BENCHMARK_DRIVER_GIT_URL = https://github.com/benchmark-driver/benchmark-driver
+BENCHMARK_DRIVER_GIT_REF = v0.14.11
+SIMPLECOV_GIT_URL = https://github.com/colszowka/simplecov.git
+SIMPLECOV_GIT_REF = v0.15.0
+SIMPLECOV_HTML_GIT_URL = https://github.com/colszowka/simplecov-html.git
+SIMPLECOV_HTML_GIT_REF = v0.10.2
+DOCLIE_GIT_URL = https://github.com/ms-ati/docile.git
+DOCLIE_GIT_REF = v1.1.5
 
 STATIC_RUBY   = static-ruby
 
+TIMESTAMPDIR  = $(EXTOUT)/.timestamp
 EXTCONF       = extconf.rb
-RBCONFIG      = ./.rbconfig.time
+LIBRUBY_EXTS  = ./.libruby-with-ext.time
+REVISION_H    = ./.revision.time
+PLATFORM_D    = $(TIMESTAMPDIR)/.$(PLATFORM_DIR).time
+ENC_TRANS_D   = $(TIMESTAMPDIR)/.enc-trans.time
+RDOCOUT       = $(EXTOUT)/rdoc
+HTMLOUT       = $(EXTOUT)/html
+CAPIOUT       = doc/capi
+INSTALL_DOC_OPTS = --rdoc-output="$(RDOCOUT)" --html-output="$(HTMLOUT)"
 
-DMYEXT	      = dmyext.$(OBJEXT)
-MAINOBJ	      = main.$(OBJEXT)
-EXTOBJS	      = 
-DLDOBJS	      = $(DMYEXT)
+INITOBJS      = dmyext.$(OBJEXT) dmyenc.$(OBJEXT)
+NORMALMAINOBJ = main.$(OBJEXT)
+MAINOBJ       = $(NORMALMAINOBJ)
+DLDOBJS	      = $(INITOBJS)
+EXTSOLIBS     =
+MINIOBJS      = $(ARCHMINIOBJS) miniinit.$(OBJEXT) dmyext.$(OBJEXT) miniprelude.$(OBJEXT)
+ENC_MK        = enc.mk
+MAKE_ENC      = -f $(ENC_MK) V="$(V)" UNICODE_HDR_DIR="$(UNICODE_HDR_DIR)" \
+		RUBY="$(MINIRUBY)" MINIRUBY="$(MINIRUBY)" $(mflags)
 
-OBJS	      = array.$(OBJEXT) \
+COMMONOBJS    = array.$(OBJEXT) \
+		ast.$(OBJEXT) \
 		bignum.$(OBJEXT) \
 		class.$(OBJEXT) \
 		compar.$(OBJEXT) \
+		compile.$(OBJEXT) \
+		complex.$(OBJEXT) \
+		cont.$(OBJEXT) \
+		debug.$(OBJEXT) \
+		debug_counter.$(OBJEXT) \
 		dir.$(OBJEXT) \
-		dln.$(OBJEXT) \
+		dln_find.$(OBJEXT) \
+		encoding.$(OBJEXT) \
 		enum.$(OBJEXT) \
+		enumerator.$(OBJEXT) \
 		error.$(OBJEXT) \
 		eval.$(OBJEXT) \
 		file.$(OBJEXT) \
@@ -28,278 +95,3129 @@ OBJS	      = array.$(OBJEXT) \
 		hash.$(OBJEXT) \
 		inits.$(OBJEXT) \
 		io.$(OBJEXT) \
+		iseq.$(OBJEXT) \
+		load.$(OBJEXT) \
 		marshal.$(OBJEXT) \
 		math.$(OBJEXT) \
+		mjit.$(OBJEXT) \
+		mjit_compile.$(OBJEXT) \
+		node.$(OBJEXT) \
 		numeric.$(OBJEXT) \
 		object.$(OBJEXT) \
 		pack.$(OBJEXT) \
 		parse.$(OBJEXT) \
+		proc.$(OBJEXT) \
 		process.$(OBJEXT) \
-		prec.$(OBJEXT) \
 		random.$(OBJEXT) \
 		range.$(OBJEXT) \
+		rational.$(OBJEXT) \
 		re.$(OBJEXT) \
-		regex.$(OBJEXT) \
+		regcomp.$(OBJEXT) \
+		regenc.$(OBJEXT) \
+		regerror.$(OBJEXT) \
+		regexec.$(OBJEXT) \
+		regparse.$(OBJEXT) \
+		regsyntax.$(OBJEXT) \
 		ruby.$(OBJEXT) \
+		safe.$(OBJEXT) \
 		signal.$(OBJEXT) \
 		sprintf.$(OBJEXT) \
 		st.$(OBJEXT) \
+		strftime.$(OBJEXT) \
 		string.$(OBJEXT) \
 		struct.$(OBJEXT) \
+		symbol.$(OBJEXT) \
+		thread.$(OBJEXT) \
 		time.$(OBJEXT) \
+		transcode.$(OBJEXT) \
+		transient_heap.$(OBJEXT) \
 		util.$(OBJEXT) \
 		variable.$(OBJEXT) \
 		version.$(OBJEXT) \
+		vm.$(OBJEXT) \
+		vm_backtrace.$(OBJEXT) \
+		vm_dump.$(OBJEXT) \
+		vm_trace.$(OBJEXT) \
+		$(COROUTINE_OBJ) \
+		$(DTRACE_OBJ) \
+		$(BUILTIN_ENCOBJS) \
+		$(BUILTIN_TRANSOBJS) \
 		$(MISSING)
 
+EXPORTOBJS    = $(DLNOBJ) \
+		localeinit.$(OBJEXT) \
+		loadpath.$(OBJEXT) \
+		$(COMMONOBJS)
+
+OBJS          = $(EXPORTOBJS) prelude.$(OBJEXT)
+ALLOBJS       = $(NORMALMAINOBJ) $(MINIOBJS) $(COMMONOBJS) $(INITOBJS)
+
+GOLFOBJS      = goruby.$(OBJEXT) golf_prelude.$(OBJEXT)
+
+DEFAULT_PRELUDES = $(GEM_PRELUDE)
+PRELUDE_SCRIPTS = $(srcdir)/prelude.rb $(DEFAULT_PRELUDES)
+GEM_PRELUDE   = $(srcdir)/gem_prelude.rb
+PRELUDES      = {$(srcdir)}prelude.c {$(srcdir)}miniprelude.c
+GOLFPRELUDES  = {$(srcdir)}golf_prelude.c
+
 SCRIPT_ARGS   =	--dest-dir="$(DESTDIR)" \
-		--make="$(MAKE)" \
+		--extout="$(EXTOUT)" \
 		--mflags="$(MFLAGS)" \
 		--make-flags="$(MAKEFLAGS)"
-EXTMK_ARGS    =	$(SCRIPT_ARGS) --extout="$(EXTOUT)" --extension $(EXTS) --extstatic $(EXTSTATIC) --
+EXTMK_ARGS    =	$(SCRIPT_ARGS) --extension $(EXTS) --extstatic $(EXTSTATIC) \
+		--make-flags="V=$(V) MINIRUBY='$(MINIRUBY)'" \
+		--gnumake=$(gnumake) --extflags="$(EXTLDFLAGS)" \
+		--
+INSTRUBY      =	$(SUDO) $(RUNRUBY) -r./$(arch)-fake $(srcdir)/tool/rbinstall.rb
+INSTRUBY_ARGS =	$(SCRIPT_ARGS) \
+		--data-mode=$(INSTALL_DATA_MODE) \
+		--prog-mode=$(INSTALL_PROG_MODE) \
+		--installed-list $(INSTALLED_LIST) \
+		--mantype="$(MANTYPE)"
+INSTALL_PROG_MODE = 0755
+INSTALL_DATA_MODE = 0644
+
+TESTSDIR      = $(srcdir)/test
+TEST_EXCLUDES = --excludes-dir=$(TESTSDIR)/excludes --name=!/memory_leak/
+EXCLUDE_TESTFRAMEWORK = --exclude=/testunit/ --exclude=/minitest/
+TESTWORKDIR   = testwork
+TESTOPTS      = $(RUBY_TESTOPTS)
+
+TESTRUN_SCRIPT = $(srcdir)/test.rb
+
+COMPILE_PRELUDE = $(srcdir)/tool/generic_erb.rb $(srcdir)/template/prelude.c.tmpl
+
+SHOWFLAGS = showflags
+
+MAKE_LINK = $(MINIRUBY) -rfileutils -e "include FileUtils::Verbose" \
+	  -e "src, dest = ARGV" \
+	  -e "exit if File.identical?(src, dest) or cmp(src, dest) rescue nil" \
+	  -e "def noraise; yield; rescue; rescue NotImplementedError; end" \
+	  -e "noraise {ln_sf('../'*dest.count('/')+src, dest)} or" \
+	  -e "noraise {ln(src, dest)} or" \
+	  -e "cp(src, dest)"
+
+
+all: $(SHOWFLAGS) main docs
+
+main: $(SHOWFLAGS) exts $(ENCSTATIC:static=lib)encs
+	@$(NULLCMD)
+
+mjit-headers: $(MJIT_SUPPORT)-mjit-headers
+no-mjit-headers: PHONY
+yes-mjit-headers: mjit_config.h PHONY
+
+mjit.$(OBJEXT): mjit_config.h
+mjit_config.h: Makefile
+
+
+# These rules using MJIT_HEADER_SUFFIX must be in common.mk, not
+# Makefile.in, in order to override the macro in defs/universal.mk.
+
+# Other `-Dxxx`s preceding `-DMJIT_HEADER` will be removed in transform_mjit_header.rb.
+# So `-DMJIT_HEADER` should be passed first when rb_mjit_header.h is generated.
+$(TIMESTAMPDIR)/$(MJIT_HEADER:.h=)$(MJIT_HEADER_SUFFIX).time: probes.h vm.$(OBJEXT) \
+		$(TIMESTAMPDIR)/$(arch)/.time
+	$(ECHO) building $(@F:.time=.h)
+	$(Q) $(CPP) -DMJIT_HEADER $(MJIT_HEADER_FLAGS) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(srcdir)/vm.c $(CPPOUTFLAG)$(@F:.time=.h).new
+	$(Q) $(IFCHANGE) "--timestamp=$@" $(@F:.time=.h) $(@F:.time=.h).new
+
+$(MJIT_HEADER:.h=)$(MJIT_HEADER_SUFFIX).h: $(TIMESTAMPDIR)/$(MJIT_HEADER:.h=)$(MJIT_HEADER_SUFFIX).time
+
+$(MJIT_MIN_HEADER:.h=)$(MJIT_HEADER_SUFFIX).h: \
+		$(TIMESTAMPDIR)/$(MJIT_HEADER:.h=)$(MJIT_HEADER_SUFFIX).time \
+		$(srcdir)/tool/transform_mjit_header.rb $(PREP) \
+		$(MJIT_HEADER:.h=)$(MJIT_HEADER_SUFFIX).h
+	$(ECHO) building $@
+	$(MINIRUBY) $(srcdir)/tool/transform_mjit_header.rb "$(CC) $(ARCH_FLAG)" $(MJIT_HEADER:.h=)$(MJIT_HEADER_ARCH).h $@
+	$(Q) $(MAKEDIRS) $(MJIT_HEADER_INSTALL_DIR)
+	$(Q) $(MAKE_LINK) $@ $(MJIT_HEADER_INSTALL_DIR)/$(@F)
+
+.PHONY: showflags
+exts enc trans: $(SHOWFLAGS)
+showflags:
+	$(MESSAGE_BEGIN) \
+	"	BASERUBY = $(BASERUBY)" \
+	"	CC = $(CC)" \
+	"	LD = $(LD)" \
+	"	LDSHARED = $(LDSHARED)" \
+	"	CFLAGS = $(CFLAGS)" \
+	"	XCFLAGS = $(XCFLAGS)" \
+	"	CPPFLAGS = $(CPPFLAGS)" \
+	"	DLDFLAGS = $(DLDFLAGS)" \
+	"	SOLIBS = $(SOLIBS)" \
+	"	LANG = $(LANG)" \
+	"	LC_ALL = $(LC_ALL)" \
+	"	LC_CTYPE = $(LC_CTYPE)" \
+	$(MESSAGE_END)
+	-@$(CC_VERSION)
+
+.PHONY: showconfig
+showconfig:
+	@$(ECHO_BEGIN) \
+	$(configure_args) \
+	$(ECHO_END)
+
+EXTS_NOTE = -f $(EXTS_MK) $(mflags) RUBY="$(MINIRUBY)" top_srcdir="$(srcdir)" note
+
+exts: build-ext
 
-all: $(MKFILES) $(PREP) $(RBCONFIG) $(LIBRUBY)
-	@$(MINIRUBY) $(srcdir)/ext/extmk.rb $(EXTMK_ARGS)
-prog: $(PROGRAM) $(WPROGRAM)
+EXTS_MK = exts.mk
+$(EXTS_MK): ext/configure-ext.mk $(TIMESTAMPDIR)/$(arch)/.time $(srcdir)/template/exts.mk.tmpl
+	$(Q)$(MAKE) -f ext/configure-ext.mk $(mflags) V=$(V) EXTSTATIC=$(EXTSTATIC) \
+		gnumake=$(gnumake) MINIRUBY="$(MINIRUBY)" \
+		EXTLDFLAGS="$(EXTLDFLAGS)" srcdir="$(srcdir)"
+	$(ECHO) generating makefile $@
+	$(Q)$(MINIRUBY) $(srcdir)/tool/generic_erb.rb -o $@ -c \
+	    $(srcdir)/template/exts.mk.tmpl --gnumake=$(gnumake)
 
-miniruby$(EXEEXT): config.status $(LIBRUBY_A) $(MAINOBJ) $(MINIOBJS) $(OBJS) $(DMYEXT)
+ext/configure-ext.mk: $(PREP) all-incs $(MKFILES) $(RBCONFIG) $(LIBRUBY) \
+		$(srcdir)/template/configure-ext.mk.tmpl
+	$(ECHO) generating makefiles $@
+	$(Q)$(MAKEDIRS) $(@D)
+	$(Q)$(MINIRUBY) $(srcdir)/tool/generic_erb.rb -o $@ -c \
+	    $(srcdir)/template/$(@F).tmpl --srcdir="$(srcdir)" \
+	    --miniruby="$(MINIRUBY)" --script-args='$(SCRIPT_ARGS)'
 
-$(PROGRAM): $(LIBRUBY) $(MAINOBJ) $(OBJS) $(EXTOBJS) $(SETUP) $(PREP)
+configure-ext: $(EXTS_MK)
 
-$(LIBRUBY_A):	$(OBJS) $(DMYEXT) $(ARCHFILE)
+build-ext: $(EXTS_MK)
+	$(Q)$(MAKE) -f $(EXTS_MK) $(mflags) libdir="$(libdir)" LIBRUBY_EXTS=$(LIBRUBY_EXTS) \
+	    EXTENCS="$(ENCOBJS)" UPDATE_LIBRARIES=no $(EXTSTATIC)
+	$(Q)$(MAKE) $(EXTS_NOTE)
 
-$(LIBRUBY_SO):	$(OBJS) $(DLDOBJS) $(LIBRUBY_A) $(PREP)
+exts-note: $(EXTS_MK)
+	$(Q)$(MAKE) $(EXTS_NOTE)
+
+ext/extinit.c: $(srcdir)/template/extinit.c.tmpl
+	$(Q)$(MINIRUBY) $(srcdir)/tool/generic_erb.rb -o $@ -c \
+	    $(srcdir)/template/extinit.c.tmpl $(EXTINITS)
+
+prog: program wprogram
+programs: $(PROGRAM) $(WPROGRAM)
+
+$(PREP): $(MKFILES)
+
+miniruby$(EXEEXT): config.status $(ALLOBJS) $(ARCHFILE)
+
+objs: $(ALLOBJS)
+
+GORUBY = go$(RUBY_INSTALL_NAME)
+GOLF = $(GORUBY)
+golf: $(GOLF)
+$(GOLF): $(LIBRUBY) $(GOLFOBJS) PHONY
+	$(Q) $(MAKE) $(mflags) \
+		GOLF=_dummy_golf_target_to_avoid_conflict_just_in_case_ \
+		MAINOBJ=goruby.$(OBJEXT) \
+		EXTOBJS="golf_prelude.$(OBJEXT) $(EXTOBJS)" \
+		PROGRAM=$(GORUBY)$(EXEEXT) \
+		V=$(V) \
+	program
+capi: $(CAPIOUT)/.timestamp PHONY
+
+$(CAPIOUT)/.timestamp: Doxyfile $(PREP)
+	$(Q) $(MAKEDIRS) "$(@D)"
+	$(ECHO) generating capi
+	-$(Q) $(DOXYGEN) -b
+	$(Q) $(MINIRUBY) -e 'File.open(ARGV[0], "w"){'"|f|"' f.puts(Time.now)}' "$@"
+
+Doxyfile: $(srcdir)/template/Doxyfile.tmpl $(PREP) $(srcdir)/tool/generic_erb.rb $(RBCONFIG)
+	$(ECHO) generating $@
+	$(Q) $(MINIRUBY) $(srcdir)/tool/generic_erb.rb -o $@ $(srcdir)/template/Doxyfile.tmpl \
+	--srcdir="$(srcdir)" --miniruby="$(MINIRUBY)"
+
+program: $(SHOWFLAGS) $(PROGRAM)
+wprogram: $(SHOWFLAGS) $(WPROGRAM)
+mini: PHONY miniruby$(EXEEXT)
+
+$(PROGRAM) $(WPROGRAM): $(LIBRUBY) $(MAINOBJ) $(OBJS) $(EXTOBJS) $(SETUP) $(PREP)
+
+$(LIBRUBY_A):	$(LIBRUBY_A_OBJS) $(MAINOBJ) $(INITOBJS) $(ARCHFILE)
+
+$(LIBRUBY_SO):	$(OBJS) $(DLDOBJS) $(LIBRUBY_A) $(PREP) $(BUILTIN_ENCOBJS)
+
+$(LIBRUBY_EXTS):
+	@exit > $@
 
 $(STATIC_RUBY)$(EXEEXT): $(MAINOBJ) $(DLDOBJS) $(EXTOBJS) $(LIBRUBY_A)
-	@$(RM) $@
-	$(PURIFY) $(CC) $(MAINOBJ) $(DLDOBJS) $(EXTOBJS) $(LIBRUBY_A) $(MAINLIBS) $(EXTLIBS) $(LIBS) $(OUTFLAG)$@ $(LDFLAGS) $(XLDFLAGS)
+	$(Q)$(RM) $@
+	$(PURIFY) $(CC) $(MAINOBJ) $(DLDOBJS) $(LIBRUBY_A) $(MAINLIBS) $(EXTLIBS) $(LIBS) $(OUTFLAG)$@ $(LDFLAGS) $(XLDFLAGS)
+
+ruby.imp: $(COMMONOBJS)
+	$(Q){ \
+	$(NM) -Pgp $(COMMONOBJS) | \
+	awk 'BEGIN{print "#!"}; $$2~/^[BDT]$$/&&$$1!~/^$(SYMBOL_PREFIX)(Init_|InitVM_|ruby_static_id_|.*_threadptr_|rb_ec_)|^\./{print $$1}'; \
+	($(CHDIR) $(srcdir) && \
+	 exec sed -n '/^MJIT_FUNC_EXPORTED/!d;N;s/.*\n\(rb_[a-zA-Z_0-9]*\).*/$(SYMBOL_PREFIX)\1/p' cont.c gc.c thread*c vm*.c) \
+	} | \
+	sort -u -o $@
+
+install: install-$(INSTALLDOC)
+docs: $(DOCTARGETS)
+pkgconfig-data: $(ruby_pc)
+$(ruby_pc): $(srcdir)/template/ruby.pc.in config.status
 
-ruby.imp: $(OBJS)
-	@$(NM) -Pgp $(OBJS) | awk 'BEGIN{print "#!"}; $$2~/^[BD]$$/{print $$1}' | sort -u -o $@
+install-all: docs pre-install-all do-install-all post-install-all
+pre-install-all:: all pre-install-local pre-install-ext pre-install-doc
+do-install-all: pre-install-all
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=all $(INSTALL_DOC_OPTS)
+post-install-all:: post-install-local post-install-ext post-install-doc
+	@$(NULLCMD)
 
-install: install-nodoc $(RDOCTARGET)
-install-all: install-nodoc install-doc
+install-nodoc: pre-install-nodoc do-install-nodoc post-install-nodoc
+pre-install-nodoc:: pre-install-local pre-install-ext
+do-install-nodoc: main pre-install-nodoc
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS)
+post-install-nodoc:: post-install-local post-install-ext
 
-install-nodoc: install-local install-ext
 install-local: pre-install-local do-install-local post-install-local
+pre-install-local:: pre-install-bin pre-install-lib pre-install-man
+do-install-local: $(PROGRAM) pre-install-local
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=local
+post-install-local:: post-install-bin post-install-lib post-install-man
+
 install-ext: pre-install-ext do-install-ext post-install-ext
+pre-install-ext:: pre-install-ext-arch pre-install-ext-comm
+do-install-ext: exts pre-install-ext
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=ext
+post-install-ext:: post-install-ext-arch post-install-ext-comm
+
+install-arch: pre-install-arch do-install-arch post-install-arch
+pre-install-arch:: pre-install-bin pre-install-ext-arch
+do-install-arch: main do-install-arch
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=arch
+post-install-arch:: post-install-bin post-install-ext-arch
+
+install-comm: pre-install-comm do-install-comm post-install-comm
+pre-install-comm:: pre-install-lib pre-install-ext-comm pre-install-man
+do-install-comm: $(PREP) pre-install-comm
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=lib --install=ext-comm --install=man
+post-install-comm:: post-install-lib post-install-ext-comm post-install-man
+
+install-bin: pre-install-bin do-install-bin post-install-bin
+pre-install-bin:: install-prereq
+do-install-bin: $(PROGRAM) pre-install-bin
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=bin
+post-install-bin::
+	@$(NULLCMD)
+
+install-lib: pre-install-lib do-install-lib post-install-lib
+pre-install-lib:: install-prereq
+do-install-lib: $(PREP) pre-install-lib
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=lib
+post-install-lib::
+	@$(NULLCMD)
+
+install-ext-comm: pre-install-ext-comm do-install-ext-comm post-install-ext-comm
+pre-install-ext-comm:: install-prereq
+do-install-ext-comm: exts pre-install-ext-comm
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=ext-comm
+post-install-ext-comm::
+	@$(NULLCMD)
+
+install-ext-arch: pre-install-ext-arch do-install-ext-arch post-install-ext-arch
+pre-install-ext-arch:: install-prereq
+do-install-ext-arch: exts pre-install-ext-arch
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=ext-arch
+post-install-ext-arch::
+	@$(NULLCMD)
+
+install-man: pre-install-man do-install-man post-install-man
+pre-install-man:: install-prereq
+do-install-man: $(PREP) pre-install-man
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=man
+post-install-man::
+	@$(NULLCMD)
 
-do-install-local: $(RBCONFIG)
-	$(MINIRUBY) $(srcdir)/instruby.rb $(SCRIPT_ARGS) --mantype="$(MANTYPE)"
-do-install-ext: $(RBCONFIG)
-	$(MINIRUBY) $(srcdir)/ext/extmk.rb $(EXTMK_ARGS) install
+install-capi: capi pre-install-capi do-install-capi post-install-capi
+pre-install-capi:: install-prereq
+do-install-capi: $(PREP) pre-install-capi
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=capi
+post-install-capi::
+	@$(NULLCMD)
 
-install-bin: $(RBCONFIG)
-	$(MINIRUBY) $(srcdir)/instruby.rb $(SCRIPT_ARGS) --install=bin
-install-lib: $(RBCONFIG)
-	$(MINIRUBY) $(srcdir)/instruby.rb $(SCRIPT_ARGS) --install=lib
-install-man: $(RBCONFIG)
-	$(MINIRUBY) $(srcdir)/instruby.rb $(SCRIPT_ARGS) --install=man --mantype="$(MANTYPE)"
+what-where: no-install
+no-install: no-install-$(INSTALLDOC)
+what-where-all: no-install-all
+no-install-all: pre-no-install-all dont-install-all post-no-install-all
+pre-no-install-all:: pre-no-install-local pre-no-install-ext pre-no-install-doc
+dont-install-all: $(PROGRAM)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=all $(INSTALL_DOC_OPTS)
+post-no-install-all:: post-no-install-local post-no-install-ext post-no-install-doc
+	@$(NULLCMD)
+
+uninstall: $(INSTALLED_LIST) sudo-precheck
+	$(Q)$(SUDO) $(MINIRUBY) $(srcdir)/tool/rbuninstall.rb --destdir=$(DESTDIR) $(INSTALLED_LIST)
+
+reinstall: all uninstall install
+
+what-where-nodoc: no-install-nodoc
+no-install-nodoc: pre-no-install-nodoc dont-install-nodoc post-no-install-nodoc
+pre-no-install-nodoc:: pre-no-install-local pre-no-install-ext
+dont-install-nodoc:  $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS)
+post-no-install-nodoc:: post-no-install-local post-no-install-ext
 
-what-where-all no-install-all: no-install no-install-doc
-what-where no-install: no-install-local no-install-ext
 what-where-local: no-install-local
-no-install-local: $(RBCONFIG)
-	$(MINIRUBY) $(srcdir)/instruby.rb -n $(SCRIPT_ARGS) --mantype="$(MANTYPE)"
+no-install-local: pre-no-install-local dont-install-local post-no-install-local
+pre-no-install-local:: pre-no-install-bin pre-no-install-lib pre-no-install-man
+dont-install-local: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=local
+post-no-install-local:: post-no-install-bin post-no-install-lib post-no-install-man
+
 what-where-ext: no-install-ext
-no-install-ext: $(RBCONFIG)
-	$(MINIRUBY) $(srcdir)/ext/extmk.rb -n $(EXTMK_ARGS) install
+no-install-ext: pre-no-install-ext dont-install-ext post-no-install-ext
+pre-no-install-ext:: pre-no-install-ext-arch pre-no-install-ext-comm
+dont-install-ext: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=ext
+post-no-install-ext:: post-no-install-ext-arch post-no-install-ext-comm
+
+what-where-arch: no-install-arch
+no-install-arch: pre-no-install-arch dont-install-arch post-no-install-arch
+pre-no-install-arch:: pre-no-install-bin pre-no-install-ext-arch
+dont-install-arch: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=bin --install=ext-arch
+post-no-install-arch:: post-no-install-lib post-no-install-man post-no-install-ext-arch
+
+what-where-comm: no-install-comm
+no-install-comm: pre-no-install-comm dont-install-comm post-no-install-comm
+pre-no-install-comm:: pre-no-install-lib pre-no-install-ext-comm pre-no-install-man
+dont-install-comm: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=lib --install=ext-comm --install=man
+post-no-install-comm:: post-no-install-lib post-no-install-ext-comm post-no-install-man
+
+what-where-bin: no-install-bin
+no-install-bin: pre-no-install-bin dont-install-bin post-no-install-bin
+pre-no-install-bin:: install-prereq
+dont-install-bin: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=bin
+post-no-install-bin::
+	@$(NULLCMD)
 
-install-doc: pre-install-doc do-install-doc post-install-doc
-do-install-doc: $(PROGRAM)
+what-where-lib: no-install-lib
+no-install-lib: pre-no-install-lib dont-install-lib post-no-install-lib
+pre-no-install-lib:: install-prereq
+dont-install-lib: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=lib
+post-no-install-lib::
+	@$(NULLCMD)
+
+what-where-ext-comm: no-install-ext-comm
+no-install-ext-comm: pre-no-install-ext-comm dont-install-ext-comm post-no-install-ext-comm
+pre-no-install-ext-comm:: install-prereq
+dont-install-ext-comm: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=ext-comm
+post-no-install-ext-comm::
+	@$(NULLCMD)
+
+what-where-ext-arch: no-install-ext-arch
+no-install-ext-arch: pre-no-install-ext-arch dont-install-ext-arch post-no-install-ext-arch
+pre-no-install-ext-arch:: install-prereq
+dont-install-ext-arch: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=ext-arch
+post-no-install-ext-arch::
+	@$(NULLCMD)
+
+what-where-man: no-install-man
+no-install-man: pre-no-install-man dont-install-man post-no-install-man
+pre-no-install-man:: install-prereq
+dont-install-man: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=man
+post-no-install-man::
+	@$(NULLCMD)
+
+install-doc: rdoc pre-install-doc do-install-doc post-install-doc
+pre-install-doc:: install-prereq
+do-install-doc: $(PROGRAM) pre-install-doc
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=rdoc $(INSTALL_DOC_OPTS)
+post-install-doc::
+	@$(NULLCMD)
+
+install-gem: pre-install-gem do-install-gem post-install-gem
+pre-install-gem:: pre-install-bin pre-install-lib pre-install-man
+do-install-gem: $(PROGRAM) pre-install-gem
+	$(INSTRUBY) --make="$(MAKE)" $(INSTRUBY_ARGS) --install=gem
+post-install-gem::
+	@$(NULLCMD)
+
+rdoc: PHONY main
 	@echo Generating RDoc documentation
-	$(RUNRUBY) "$(srcdir)/bin/rdoc" --all --ri --op "$(RIDATADIR)" "$(srcdir)"
-
-pre-install: pre-install-local pre-install-ext
-pre-install-local:: PHONY
-	$(PREINSTALL)
-pre-install-ext:: PHONY
-pre-install-doc:: PHONY
-
-post-install: post-install-local post-install-ext
-post-install-local:: PHONY
-post-install-ext:: PHONY
-post-install-doc:: PHONY
-
-clean: clean-ext clean-local
-clean-local::
-	@$(RM) $(OBJS) $(MAINOBJ) $(WINMAINOBJ) $(LIBRUBY_A) $(LIBRUBY_SO) $(LIBRUBY) $(LIBRUBY_ALIASES)
-	@$(RM) $(PROGRAM) $(WPROGRAM) miniruby$(EXEEXT) dmyext.$(OBJEXT) $(ARCHFILE)
-clean-ext:
-	@-$(MINIRUBY) $(srcdir)/ext/extmk.rb $(EXTMK_ARGS) clean
-
-distclean: distclean-ext distclean-local
+	$(Q) $(XRUBY) "$(srcdir)/libexec/rdoc" --root "$(srcdir)" --page-dir "$(srcdir)/doc" --encoding=UTF-8 --no-force-update --all --ri --op "$(RDOCOUT)" $(RDOCFLAGS) "$(srcdir)"
+
+html: PHONY main
+	@echo Generating RDoc HTML files
+	$(Q) $(XRUBY) "$(srcdir)/libexec/rdoc" --root "$(srcdir)" --page-dir "$(srcdir)/doc" --encoding=UTF-8 --no-force-update --all --op "$(HTMLOUT)" $(RDOCFLAGS) "$(srcdir)"
+
+rdoc-coverage: PHONY main
+	@echo Generating RDoc coverage report
+	$(Q) $(XRUBY) "$(srcdir)/libexec/rdoc" --root "$(srcdir)" --encoding=UTF-8 --all --quiet -C $(RDOCFLAGS) "$(srcdir)"
+
+RDOCBENCHOUT=/tmp/rdocbench
+
+GCBENCH_ITEM=null
+
+gcbench: PHONY
+	$(Q) $(XRUBY) "$(srcdir)/benchmark/gc/gcbench.rb" $(GCBENCH_ITEM)
+
+gcbench-rdoc: PHONY
+	$(Q) $(XRUBY) "$(srcdir)/benchmark/gc/gcbench.rb" rdoc
+
+nodoc: PHONY
+
+what-where-doc: no-install-doc
+no-install-doc: pre-no-install-doc dont-install-doc post-no-install-doc
+pre-no-install-doc:: install-prereq
+dont-install-doc:: $(PREP)
+	$(INSTRUBY) -n --make="$(MAKE)" $(INSTRUBY_ARGS) --install=rdoc $(INSTALL_DOC_OPTS)
+post-no-install-doc::
+	@$(NULLCMD)
+
+CLEAR_INSTALLED_LIST = clear-installed-list
+
+install-prereq: $(CLEAR_INSTALLED_LIST) yes-fake sudo-precheck PHONY
+
+clear-installed-list: PHONY
+	@> $(INSTALLED_LIST) set MAKE="$(MAKE)"
+
+clean: clean-ext clean-enc clean-golf clean-docs clean-extout clean-local clean-platform clean-spec
+clean-local:: clean-runnable
+	$(Q)$(RM) $(OBJS) $(MINIOBJS) $(MAINOBJ) $(LIBRUBY_A) $(LIBRUBY_SO) $(LIBRUBY) $(LIBRUBY_ALIASES)
+	$(Q)$(RM) $(PROGRAM) $(WPROGRAM) miniruby$(EXEEXT) dmyext.$(OBJEXT) dmyenc.$(OBJEXT) $(ARCHFILE) .*.time
+	$(Q)$(RM) y.tab.c y.output encdb.h transdb.h config.log rbconfig.rb $(ruby_pc)
+	$(Q)$(RM) probes.h probes.$(OBJEXT) probes.stamp ruby-glommed.$(OBJEXT) ruby.imp
+	$(Q)$(RM) GNUmakefile.old Makefile.old $(arch)-fake.rb bisect.sh $(ENC_TRANS_D)
+	-$(Q) $(RMDIR) enc/jis enc/trans enc 2> $(NULL) || exit 0
+clean-runnable:: PHONY
+	$(Q)$(CHDIR) bin 2>$(NULL) && $(RM) $(PROGRAM) $(WPROGRAM) $(GORUBY)$(EXEEXT) bin/*.$(DLEXT) 2>$(NULL) || exit 0
+	$(Q)$(CHDIR) lib 2>$(NULL) && $(RM) $(LIBRUBY_A) $(LIBRUBY) $(LIBRUBY_ALIASES) $(RUBY_BASE_NAME)/$(RUBY_PROGRAM_VERSION) $(RUBY_BASE_NAME)/vendor_ruby 2>$(NULL) || exit 0
+	$(Q)$(RMDIR) lib/$(RUBY_BASE_NAME) lib bin 2>$(NULL) || exit 0
+clean-ext:: PHONY
+clean-golf: PHONY
+	$(Q)$(RM) $(GORUBY)$(EXEEXT) $(GOLFOBJS)
+clean-rdoc: PHONY
+clean-html: PHONY
+clean-capi: PHONY
+clean-platform: PHONY
+clean-extout: PHONY
+	-$(Q)$(RMDIR) $(EXTOUT)/$(arch) $(EXTOUT) 2> $(NULL) || exit 0
+clean-docs: clean-rdoc clean-html clean-capi
+clean-spec: PHONY
+clean-rubyspec: clean-spec
+
+distclean: distclean-ext distclean-enc distclean-golf distclean-docs distclean-extout distclean-local distclean-platform distclean-spec
 distclean-local:: clean-local
-	@$(RM) $(MKFILES) config.h rbconfig.rb
-	@$(RM) config.cache config.log config.status
-	@$(RM) *~ *.bak *.stackdump core *.core gmon.out y.tab.c y.output $(PREP)
-distclean-ext:
-	@-$(MINIRUBY) $(srcdir)/ext/extmk.rb $(EXTMK_ARGS) distclean
+	$(Q)$(RM) $(MKFILES) yasmdata.rb *.inc $(PRELUDES)
+	$(Q)$(RM) config.cache config.status config.status.lineno
+	$(Q)$(RM) *~ *.bak *.stackdump core *.core gmon.out $(PREP)
+	-$(Q)$(RMALL) $(srcdir)/autom4te.cache
+distclean-ext:: PHONY
+distclean-golf: clean-golf
+distclean-rdoc: clean-rdoc
+distclean-html: clean-html
+distclean-capi: clean-capi
+distclean-docs: clean-docs
+distclean-extout: clean-extout
+distclean-platform: clean-platform
+distclean-spec: clean-spec
+distclean-rubyspec: distclean-spec
+
+realclean:: realclean-ext realclean-local realclean-enc realclean-golf realclean-extout
+realclean-local:: distclean-local realclean-srcs-local
+
+clean-srcs:: clean-srcs-local clean-srcs-ext
+realclean-srcs:: realclean-srcs-local realclean-srcs-ext
+
+clean-srcs-local::
+	$(Q)$(RM) parse.c parse.h lex.c enc/trans/newline.c revision.h
+	$(Q)$(RM) id.c id.h probes.dmyh probes.h
+	$(Q)$(RM) encdb.h transdb.h verconf.h ruby-runner.h
+	$(Q)$(RM) mjit_config.h rb_mjit_header.h
+	$(Q)$(RM) $(MJIT_MIN_HEADER) $(MJIT_MIN_HEADER:.h=)$(MJIT_HEADER_SUFFIX:%=*).h
+
+realclean-srcs-local:: clean-srcs-local
+	$(Q)$(CHDIR) $(srcdir) && $(exec) $(RM) parse.c parse.h lex.c enc/trans/newline.c $(PRELUDES) revision.h
+	$(Q)$(CHDIR) $(srcdir) && $(exec) $(RM) id.c id.h probes.dmyh
+	$(Q)$(CHDIR) $(srcdir) && $(exec) $(RM) configure aclocal.m4 tool/config.guess tool/config.sub gems/*.gem
+
+clean-srcs-ext::
+realclean-srcs-ext:: clean-srcs-ext
+
+realclean-ext:: PHONY
+realclean-golf: distclean-golf
+	$(Q)$(RM) $(GOLFPRELUDES)
+realclean-rdoc: distclean-rdoc
+realclean-html: distclean-html
+realclean-capi: distclean-capi
+realclean-docs: distclean-docs
+realclean-extout: distclean-extout
+realclean-platform: distclean-platform
+realclean-spec: distclean-spec
+realclean-rubyspec: realclean-spec
+
+clean-ext:: ext/clean gems/clean timestamp/clean
+distclean-ext:: ext/distclean gems/distclean timestamp/distclean
+realclean-ext:: ext/realclean gems/realclean timestamp/realclean
+
+ext/clean.mk ext/distclean.mk ext/realclean.mk::
+ext/clean gems/clean:: ext/clean.mk
+ext/distclean gems/distclean:: ext/distclean.mk
+ext/realclean gems/realclean:: ext/realclean.mk
+
+timestamp/clean:: ext/clean gems/clean
+timestamp/distclean:: ext/distclean gems/distclean
+timestamp/realclean:: ext/realclean gems/realclean
+
+timestamp/clean timestamp/distclean timestamp/realclean::
+	$(Q)$(RM) $(TIMESTAMPDIR)/.*.time $(TIMESTAMPDIR)/$(arch)/.time
+	$(Q)$(RMDIRS) $(TIMESTAMPDIR)/$(arch) 2> $(NULL) || exit 0
+
+clean-ext::
+	-$(Q)$(RM) ext/extinit.$(OBJEXT)
+
+distclean-ext realclean-ext::
+	-$(Q)$(RM) $(EXTS_MK) ext/extinit.* ext/configure-ext.mk
+	-$(Q)$(RMDIR) ext 2> $(NULL) || exit 0
 
-realclean:: distclean
-	@$(RM) parse.c lex.c
+clean-enc distclean-enc realclean-enc: PHONY
 
-check: test test-all
+clean-enc: clean-enc.d
 
-test: miniruby$(EXEEXT) $(RBCONFIG) $(PROGRAM) PHONY
-	@$(MINIRUBY) $(srcdir)/rubytest.rb
+clean-enc.d: PHONY
+	$(Q)$(RM) $(ENC_TRANS_D)
+	-$(Q) $(RMDIR) enc/jis enc/trans enc 2> $(NULL) || exit 0
 
-test-all:
-	$(RUNRUBY) -C "$(srcdir)/test" runner.rb --runner=$(TESTUI) $(TESTS)
+clean-rdoc distclean-rdoc realclean-rdoc:
+	@echo $(@:-rdoc=ing) rdoc
+	$(Q)$(RMALL) $(RDOCOUT)
 
-extconf:
-	$(MINIRUBY) -I$(srcdir)/lib -run -e mkdir -- -p "$(EXTCONFDIR)"
+clean-html distclean-html realclean-html:
+	@echo $(@:-html=ing) HTML
+	$(Q)$(RMALL) $(HTMLOUT)
+
+clean-capi distclean-capi realclean-capi:
+	@echo $(@:-capi=ing) capi
+	$(Q)$(RMALL) $(CAPIOUT)
+
+clean-platform:
+	$(Q) $(RM) $(PLATFORM_D)
+	-$(Q) $(RMDIR) $(PLATFORM_DIR) 2> $(NULL) || exit 0
+
+RUBYSPEC_CAPIEXT = spec/ruby/optional/capi/ext
+clean-spec: PHONY
+	-$(Q) $(RM) $(RUBYSPEC_CAPIEXT)/*.$(OBJEXT) $(RUBYSPEC_CAPIEXT)/*.$(DLEXT)
+	-$(Q) $(RMDIRS) $(RUBYSPEC_CAPIEXT) 2> $(NULL) || exit 0
+
+check: main test test-testframework test-almost test-spec
+	$(ECHO) check succeeded
+check-ruby: test test-ruby
+
+fake: $(CROSS_COMPILING)-fake
+yes-fake: $(arch)-fake.rb $(RBCONFIG) PHONY
+no-fake -fake: PHONY
+
+# really doesn't depend on .o, just ensure newer than headers which
+# version.o depends on.
+$(arch)-fake.rb: $(srcdir)/template/fake.rb.in $(srcdir)/tool/generic_erb.rb version.$(OBJEXT) miniruby$(EXEEXT)
+	$(ECHO) generating $@
+	$(Q) $(CPP) $(warnflags) $(XCFLAGS) $(CPPFLAGS) "$(srcdir)/version.c" | \
+	$(BOOTSTRAPRUBY) "$(srcdir)/tool/generic_erb.rb" -o $@ "$(srcdir)/template/fake.rb.in" \
+		i=- srcdir="$(srcdir)" BASERUBY="$(BASERUBY)"
+
+btest: $(TEST_RUNNABLE)-btest
+no-btest: PHONY
+yes-btest: fake miniruby$(EXEEXT) PHONY
+	$(Q)$(exec) $(BOOTSTRAPRUBY) "$(srcdir)/bootstraptest/runner.rb" --ruby="$(BTESTRUBY) $(RUN_OPTS)" $(OPTS) $(TESTOPTS) $(BTESTS)
+
+btest-ruby: $(TEST_RUNNABLE)-btest-ruby
+no-btest-ruby: PHONY
+yes-btest-ruby: prog PHONY
+	$(Q)$(exec) $(RUNRUBY) "$(srcdir)/bootstraptest/runner.rb" --ruby="$(PROGRAM) -I$(srcdir)/lib $(RUN_OPTS)" -q $(OPTS) $(TESTOPTS) $(BTESTS)
+
+test-basic: $(TEST_RUNNABLE)-test-basic
+no-test-basic: PHONY
+yes-test-basic: prog PHONY
+	$(Q)$(exec) $(RUNRUBY) "$(srcdir)/basictest/runner.rb" --run-opt=$(RUN_OPTS) $(OPTS) $(TESTOPTS)
+
+test-knownbugs: test-knownbug
+test-knownbug: $(TEST_RUNNABLE)-test-knownbug
+no-test-knownbug: PHONY
+yes-test-knownbug: prog PHONY
+	-$(exec) $(RUNRUBY) "$(srcdir)/bootstraptest/runner.rb" --ruby="$(PROGRAM) $(RUN_OPTS)" $(OPTS) $(TESTOPTS) $(srcdir)/KNOWNBUGS.rb
+
+test-testframework: $(TEST_RUNNABLE)-test-testframework
+yes-test-testframework: prog PHONY
+	$(gnumake_recursive)$(Q)$(exec) $(RUNRUBY) "$(srcdir)/test/runner.rb" --ruby="$(RUNRUBY)" $(TESTOPTS) testunit minitest
+no-test-testframework: PHONY
+
+test-sample: test-basic # backward compatibility for mswin-build
+test-short: btest-ruby test-knownbug test-basic
+test: test-short
+
+# $ make test-all TESTOPTS="--help" displays more detail
+# for example, make test-all TESTOPTS="-j2 -v -n test-name -- test-file-name"
+test-all: $(TEST_RUNNABLE)-test-all
+yes-test-all: programs PHONY
+	$(gnumake_recursive)$(Q)$(exec) $(RUNRUBY) "$(srcdir)/test/runner.rb" --ruby="$(RUNRUBY)" $(TEST_EXCLUDES) $(TESTOPTS) $(TESTS)
+TESTS_BUILD = mkmf
+no-test-all: PHONY
+	$(gnumake_recursive)$(MINIRUBY) -I"$(srcdir)/lib" "$(srcdir)/test/runner.rb" $(TESTOPTS) $(TESTS_BUILD)
+
+test-almost: $(TEST_RUNNABLE)-test-almost
+yes-test-almost: prog PHONY
+	$(gnumake_recursive)$(Q)$(exec) $(RUNRUBY) "$(srcdir)/test/runner.rb" --ruby="$(RUNRUBY)" $(TEST_EXCLUDES) $(TESTOPTS) $(EXCLUDE_TESTFRAMEWORK) $(TESTS)
+no-test-almost: PHONY
+
+test-ruby: $(TEST_RUNNABLE)-test-ruby
+no-test-ruby: PHONY
+yes-test-ruby: prog encs PHONY
+	$(gnumake_recursive)$(RUNRUBY) "$(srcdir)/test/runner.rb" $(TEST_EXCLUDES) $(TESTOPTS) -- ruby -ext-
+
+extconf: $(PREP)
+	$(Q) $(MAKEDIRS) "$(EXTCONFDIR)"
 	$(RUNRUBY) -C "$(EXTCONFDIR)" $(EXTCONF) $(EXTCONFARGS)
 
-$(RBCONFIG): $(srcdir)/mkconfig.rb config.status $(PREP)
-	@$(MINIRUBY) $(srcdir)/mkconfig.rb -timestamp=$@ \
+$(RBCONFIG): $(srcdir)/tool/mkconfig.rb config.status $(srcdir)/version.h
+	$(Q)$(BOOTSTRAPRUBY) -n \
+	-e 'BEGIN{version=ARGV.shift;mis=ARGV.dup}' \
+	-e 'END{abort "UNICODE version mismatch: #{mis}" unless mis.empty?}' \
+	-e '(mis.delete(ARGF.path); ARGF.close) if /ONIG_UNICODE_VERSION_STRING +"#{Regexp.quote(version)}"/o' \
+	$(UNICODE_VERSION) $(UNICODE_DATA_HEADERS)
+	$(Q)$(BOOTSTRAPRUBY) $(srcdir)/tool/mkconfig.rb \
+		-arch=$(arch) -version=$(RUBY_PROGRAM_VERSION) \
 		-install_name=$(RUBY_INSTALL_NAME) \
-		-so_name=$(RUBY_SO_NAME) rbconfig.rb
+		-so_name=$(RUBY_SO_NAME) \
+		-unicode_version=$(UNICODE_VERSION) \
+		-unicode_emoji_version=$(UNICODE_EMOJI_VERSION) \
+	> rbconfig.tmp
+	$(IFCHANGE) "--timestamp=$@" rbconfig.rb rbconfig.tmp
+
+test-rubyspec: test-spec
+yes-test-rubyspec: yes-test-spec
+
+test-spec-precheck: $(arch)-fake.rb programs
+
+test-spec: $(TEST_RUNNABLE)-test-spec
+yes-test-spec: test-spec-precheck
+	$(gnumake_recursive)$(Q) \
+	$(RUNRUBY) -r./$(arch)-fake $(srcdir)/spec/mspec/bin/mspec run -B $(srcdir)/spec/default.mspec $(MSPECOPT) $(SPECOPTS)
+no-test-spec:
+
+RUNNABLE = $(LIBRUBY_RELATIVE:no=un)-runnable
+runnable: $(RUNNABLE) prog $(srcdir)/tool/mkrunnable.rb PHONY
+	$(Q) $(MINIRUBY) $(srcdir)/tool/mkrunnable.rb -v $(EXTOUT)
+yes-runnable: PHONY
+
+encs: enc trans
+libencs: libenc libtrans
+encs enc trans libencs libenc libtrans: $(SHOWFLAGS) $(ENC_MK) $(LIBRUBY) $(PREP) PHONY
+	$(ECHO) making $@
+	$(Q) $(MAKE) $(MAKE_ENC) $@
+
+
+libenc enc: {$(VPATH)}encdb.h
+libtrans trans: {$(VPATH)}transdb.h
+
+# Use MINIRUBY which loads fake.rb for cross compiling
+$(ENC_MK): $(srcdir)/enc/make_encmake.rb $(srcdir)/enc/Makefile.in $(srcdir)/enc/depend \
+	$(srcdir)/enc/encinit.c.erb $(srcdir)/lib/mkmf.rb $(RBCONFIG) fake
+	$(ECHO) generating $@
+	$(Q) $(MINIRUBY) $(srcdir)/enc/make_encmake.rb --builtin-encs="$(BUILTIN_ENCOBJS)" --builtin-transes="$(BUILTIN_TRANSOBJS)" --module$(ENCSTATIC) $(ENCS) $@
 
 .PRECIOUS: $(MKFILES)
 
+.PHONY: PHONY all fake prereq incs srcs preludes help
 .PHONY: test install install-nodoc install-doc dist
+.PHONY: loadpath golf capi rdoc install-prereq clear-installed-list
+.PHONY: clean clean-ext clean-local clean-enc clean-golf clean-rdoc clean-html clean-extout
+.PHONY: distclean distclean-ext distclean-local distclean-enc distclean-golf distclean-extout
+.PHONY: realclean realclean-ext realclean-local realclean-enc realclean-golf realclean-extout
+.PHONY: exam check test test-short test-all btest btest-ruby test-basic test-knownbug
+.PHONY: run runruby parse benchmark gdb gdb-ruby
+.PHONY: update-mspec update-rubyspec test-rubyspec test-spec
+.PHONY: touch-unicode-files
 
 PHONY:
 
-{$(VPATH)}parse.c: parse.y
+{$(VPATH)}parse.c: {$(VPATH)}parse.y $(srcdir)/tool/ytab.sed {$(VPATH)}id.h
+{$(VPATH)}parse.h: {$(VPATH)}parse.c
+
+{$(srcdir)}.y.c:
+	$(ECHO) generating $@
+	$(Q)$(BASERUBY) $(srcdir)/tool/id2token.rb --path-separator=.$(PATH_SEPARATOR)./ --vpath=$(VPATH) id.h $(SRC_FILE) > parse.tmp.y
+	$(Q)$(YACC) -d $(YFLAGS) -o y.tab.c parse.tmp.y
+	$(Q)$(RM) parse.tmp.y
+	$(Q)sed -f $(srcdir)/tool/ytab.sed -e "/^#/s|parse\.tmp\.[iy]|$(SRC_FILE)|" -e "/^#/s!y\.tab\.c!$@!" y.tab.c > $@.new
+	$(Q)$(MV) $@.new $@
+	$(Q)sed -e "/^#line.*y\.tab\.h/d;/^#line.*parse.*\.y/d" y.tab.h > $(@:.c=.h)
+	$(Q)$(RM) y.tab.c y.tab.h
+
+$(PLATFORM_D):
+	$(Q) $(MAKEDIRS) $(PLATFORM_DIR) $(@D)
+	@exit > $@
+
+exe/$(PROGRAM): ruby-runner.c ruby-runner.h exe/.time miniruby$(EXEEXT) {$(VPATH)}config.h
+	$(Q) $(CC) $(CFLAGS) $(CPPFLAGS) -DRUBY_INSTALL_NAME=$(@F) $(COUTFLAG)ruby-runner.$(OBJEXT) -c $(CSRCFLAG)$(srcdir)/ruby-runner.c
+	$(Q) $(PURIFY) $(CC) $(CFLAGS) $(CPPFLAGS) $(LDFLAGS) $(OUTFLAG)$@ ruby-runner.$(OBJEXT) $(LIBS)
+	$(Q) $(POSTLINK)
+	$(Q) ./miniruby$(EXEEXT) \
+	    -e 'prog, dest = ARGV; dest += "/ruby"' \
+	    -e 'unless prog=="ruby"' \
+	    -e '  begin File.unlink(dest); rescue Errno::ENOENT; end' \
+	    -e '  File.symlink(prog, dest)' \
+	    -e 'end' \
+	$(@F) $(@D)
+
+exe/.time:
+	$(Q) $(MAKEDIRS) $(@D)
+	@exit > $@
+
+$(BUILTIN_ENCOBJS) $(BUILTIN_TRANSOBJS): $(ENC_TRANS_D)
+
+$(ENC_TRANS_D):
+	$(Q) $(MAKEDIRS) enc/trans $(@D)
+	@exit > $@
+
+$(TIMESTAMPDIR)/$(arch)/.time:
+	$(Q)$(MAKEDIRS) $(@D) $(EXTOUT)/$(arch)
+	@exit > $@
+
+###
+CCAN_DIR = {$(VPATH)}ccan
+
+RUBY_H_INCLUDES    = {$(VPATH)}ruby.h {$(VPATH)}config.h {$(VPATH)}defines.h \
+		     {$(VPATH)}intern.h {$(VPATH)}missing.h {$(VPATH)}st.h \
+		     {$(VPATH)}subst.h
+
+###
 
 acosh.$(OBJEXT): {$(VPATH)}acosh.c
-alloca.$(OBJEXT): {$(VPATH)}alloca.c
-crypt.$(OBJEXT): {$(VPATH)}crypt.c
+alloca.$(OBJEXT): {$(VPATH)}alloca.c {$(VPATH)}config.h
+crypt.$(OBJEXT): {$(VPATH)}crypt.c {$(VPATH)}crypt.h {$(VPATH)}missing/des_tables.c
 dup2.$(OBJEXT): {$(VPATH)}dup2.c
 erf.$(OBJEXT): {$(VPATH)}erf.c
+explicit_bzero.$(OBJEXT): {$(VPATH)}explicit_bzero.c
 finite.$(OBJEXT): {$(VPATH)}finite.c
 flock.$(OBJEXT): {$(VPATH)}flock.c
 memcmp.$(OBJEXT): {$(VPATH)}memcmp.c
 memmove.$(OBJEXT): {$(VPATH)}memmove.c
 mkdir.$(OBJEXT): {$(VPATH)}mkdir.c
-vsnprintf.$(OBJEXT): {$(VPATH)}vsnprintf.c
-strcasecmp.$(OBJEXT): {$(VPATH)}strcasecmp.c
-strncasecmp.$(OBJEXT): {$(VPATH)}strncasecmp.c
+setproctitle.$(OBJEXT): {$(VPATH)}setproctitle.c
 strchr.$(OBJEXT): {$(VPATH)}strchr.c
 strdup.$(OBJEXT): {$(VPATH)}strdup.c
 strerror.$(OBJEXT): {$(VPATH)}strerror.c
-strftime.$(OBJEXT): {$(VPATH)}strftime.c
+strlcat.$(OBJEXT): {$(VPATH)}strlcat.c
+strlcpy.$(OBJEXT): {$(VPATH)}strlcpy.c
 strstr.$(OBJEXT): {$(VPATH)}strstr.c
-strtod.$(OBJEXT): {$(VPATH)}strtod.c
-strtol.$(OBJEXT): {$(VPATH)}strtol.c
-strtoul.$(OBJEXT): {$(VPATH)}strtoul.c
 nt.$(OBJEXT): {$(VPATH)}nt.c
-x68.$(OBJEXT): {$(VPATH)}x68.c
-os2.$(OBJEXT): {$(VPATH)}os2.c
-dl_os2.$(OBJEXT): {$(VPATH)}dl_os2.c
-
-# when I use -I., there is confliction at "OpenFile" 
-# so, set . into environment varible "include"
-win32.$(OBJEXT): {$(VPATH)}win32.c
+ia64.$(OBJEXT): {$(VPATH)}ia64.s
+	$(CC) $(CFLAGS) -c $<
+.coroutine_obj $(COROUTINE_OBJ): \
+	{$(VPATH)}$(COROUTINE_H:.h=).$(ASMEXT) \
+	$(COROUTINE_H:/Context.h=/.time)
+$(COROUTINE_H:/Context.h=/.time):
+	$(Q) $(MAKEDIRS) $(@D)
+	@exit > $@
 
 ###
 
-array.$(OBJEXT): {$(VPATH)}array.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}util.h {$(VPATH)}st.h
-bignum.$(OBJEXT): {$(VPATH)}bignum.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-class.$(OBJEXT): {$(VPATH)}class.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}rubysig.h {$(VPATH)}node.h {$(VPATH)}st.h
-compar.$(OBJEXT): {$(VPATH)}compar.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-dir.$(OBJEXT): {$(VPATH)}dir.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}util.h
-dln.$(OBJEXT): {$(VPATH)}dln.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}dln.h
-dmydln.$(OBJEXT): {$(VPATH)}dmydln.c {$(VPATH)}dln.c {$(VPATH)}ruby.h \
-  config.h {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}dln.h
+# dependencies for generated C sources.
+parse.$(OBJEXT): {$(VPATH)}parse.c
+miniprelude.$(OBJEXT): {$(VPATH)}miniprelude.c
+prelude.$(OBJEXT): {$(VPATH)}prelude.c
+
+# dependencies for optional sources.
+compile.$(OBJEXT): {$(VPATH)}opt_sc.inc {$(VPATH)}optunifs.inc
+
+win32/win32.$(OBJEXT): {$(VPATH)}win32/win32.c {$(VPATH)}win32/file.h \
+  {$(VPATH)}dln.h {$(VPATH)}dln_find.c {$(VPATH)}encindex.h \
+  {$(VPATH)}internal.h {$(VPATH)}util.h $(RUBY_H_INCLUDES) \
+  {$(VPATH)}vm.h $(PLATFORM_D)
+win32/file.$(OBJEXT): {$(VPATH)}win32/file.c {$(VPATH)}win32/file.h \
+  $(RUBY_H_INCLUDES) $(PLATFORM_D)
+
+$(NEWLINE_C): $(srcdir)/enc/trans/newline.trans $(srcdir)/tool/transcode-tblgen.rb
+	$(Q) $(MAKEDIRS) $(@D)
+	$(Q) $(BASERUBY) "$(srcdir)/tool/transcode-tblgen.rb" -vo $@ $(srcdir)/enc/trans/newline.trans
+enc/trans/newline.$(OBJEXT): $(NEWLINE_C)
+
+verconf.h: $(srcdir)/template/verconf.h.tmpl $(srcdir)/tool/generic_erb.rb
+	$(ECHO) creating $@
+	$(Q) $(BOOTSTRAPRUBY) "$(srcdir)/tool/generic_erb.rb" -o $@ $(srcdir)/template/verconf.h.tmpl
+
+ruby-glommed.$(OBJEXT): $(OBJS)
+
+$(OBJS):  {$(VPATH)}config.h {$(VPATH)}missing.h
+
+INSNS2VMOPT = --srcdir="$(srcdir)"
+
+srcs_vpath = {$(VPATH)}
+
+# TODO: dependencies on tool/ruby_vm scripts.
+inc_common_headers = $(srcdir)/tool/ruby_vm/views/_copyright.erb $(srcdir)/tool/ruby_vm/views/_notice.erb
+$(srcs_vpath)opt_sc.inc: $(srcdir)/tool/ruby_vm/views/opt_sc.inc.erb $(inc_common_headers)
+$(srcs_vpath)optinsn.inc: $(srcdir)/tool/ruby_vm/views/optinsn.inc.erb $(inc_common_headers)
+$(srcs_vpath)optunifs.inc: $(srcdir)/tool/ruby_vm/views/optunifs.inc.erb $(inc_common_headers)
+$(srcs_vpath)insns.inc: $(srcdir)/tool/ruby_vm/views/insns.inc.erb $(inc_common_headers)
+$(srcs_vpath)insns_info.inc: $(srcdir)/tool/ruby_vm/views/insns_info.inc.erb $(inc_common_headers) \
+  $(srcdir)/tool/ruby_vm/views/_insn_type_chars.erb $(srcdir)/tool/ruby_vm/views/_insn_name_info.erb \
+  $(srcdir)/tool/ruby_vm/views/_insn_len_info.erb $(srcdir)/tool/ruby_vm/views/_insn_operand_info.erb \
+  $(srcdir)/tool/ruby_vm/views/_attributes.erb $(srcdir)/tool/ruby_vm/views/_insn_stack_increase.erb
+$(srcs_vpath)vmtc.inc: $(srcdir)/tool/ruby_vm/views/vmtc.inc.erb $(inc_common_headers)
+$(srcs_vpath)vm.inc: $(srcdir)/tool/ruby_vm/views/vm.inc.erb $(inc_common_headers) \
+  $(srcdir)/tool/ruby_vm/views/_insn_entry.erb $(srcdir)/tool/ruby_vm/views/_trace_instruction.erb
+$(srcs_vpath)mjit_compile.inc: $(srcdir)/tool/ruby_vm/views/mjit_compile.inc.erb $(inc_common_headers) \
+  $(srcdir)/tool/ruby_vm/views/_mjit_compile_insn.erb $(srcdir)/tool/ruby_vm/views/_mjit_compile_send.erb \
+  $(srcdir)/tool/ruby_vm/views/_mjit_compile_ivar.erb \
+  $(srcdir)/tool/ruby_vm/views/_mjit_compile_insn_body.erb $(srcdir)/tool/ruby_vm/views/_mjit_compile_pc_and_sp.erb
+
+common-srcs: $(srcs_vpath)parse.c $(srcs_vpath)lex.c $(srcs_vpath)enc/trans/newline.c $(srcs_vpath)id.c \
+	     srcs-lib srcs-ext incs
+
+missing-srcs: $(srcdir)/missing/des_tables.c
+
+srcs: common-srcs missing-srcs srcs-enc
+
+EXT_SRCS = $(srcdir)/ext/ripper/ripper.c \
+	   $(srcdir)/ext/rbconfig/sizeof/sizes.c \
+	   $(srcdir)/ext/rbconfig/sizeof/limits.c \
+	   $(srcdir)/ext/socket/constdefs.c \
+	   $(srcdir)/ext/etc/constdefs.h \
+	   # EXT_SRCS
+
+srcs-ext: $(EXT_SRCS)
+realclean-srcs-ext::
+	$(Q)$(RM) $(EXT_SRCS)
+
+EXTRA_SRCS = $(srcdir)/ext/json/parser/parser.c \
+	     $(srcdir)/ext/date/zonetab.h \
+	     $(empty)
+
+srcs-extra: $(EXTRA_SRCS)
+realclean-srcs-extra::
+	$(Q)$(RM) $(EXTRA_SRCS)
+
+LIB_SRCS = $(srcdir)/lib/unicode_normalize/tables.rb
+
+srcs-lib: $(LIB_SRCS)
+
+srcs-enc: $(ENC_MK)
+	$(ECHO) making srcs under enc
+	$(Q) $(MAKE) $(MAKE_ENC) srcs
+
+all-incs: incs {$(VPATH)}encdb.h {$(VPATH)}transdb.h
+incs: $(INSNS) {$(VPATH)}node_name.inc {$(VPATH)}known_errors.inc \
+      {$(VPATH)}vm_call_iseq_optimized.inc $(srcdir)/revision.h \
+      $(REVISION_H) \
+      $(UNICODE_DATA_HEADERS) $(srcdir)/enc/jis/props.h \
+      {$(VPATH)}id.h {$(VPATH)}probes.dmyh
+
+insns: $(INSNS)
+
+id.h: $(srcdir)/tool/generic_erb.rb $(srcdir)/template/id.h.tmpl $(srcdir)/defs/id.def
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) $(srcdir)/tool/generic_erb.rb --output=$@ \
+		$(srcdir)/template/id.h.tmpl
+
+id.c: $(srcdir)/tool/generic_erb.rb $(srcdir)/template/id.c.tmpl $(srcdir)/defs/id.def
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) $(srcdir)/tool/generic_erb.rb --output=$@ \
+		$(srcdir)/template/id.c.tmpl
+
+{$(VPATH)}node.h: $(srcdir)/tool/node_name.rb
+# This dependency is not really correct, but $? should not include
+# node_name.rb in the following recipe.
+node_name.inc: {$(VPATH)}node.h
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) -n $(srcdir)/tool/node_name.rb < $? > $@
+
+encdb.h: $(PREP) $(srcdir)/tool/generic_erb.rb $(srcdir)/template/encdb.h.tmpl
+	$(ECHO) generating $@
+	$(Q) $(MINIRUBY) $(srcdir)/tool/generic_erb.rb -c -o $@ $(srcdir)/template/encdb.h.tmpl $(srcdir)/enc enc
+
+transdb.h: $(PREP) srcs-enc $(srcdir)/tool/generic_erb.rb $(srcdir)/template/transdb.h.tmpl
+	$(ECHO) generating $@
+	$(Q) $(MINIRUBY) $(srcdir)/tool/generic_erb.rb -c -o $@ $(srcdir)/template/transdb.h.tmpl $(srcdir)/enc/trans enc/trans
+
+enc/encinit.c: $(ENC_MK) $(srcdir)/enc/encinit.c.erb
+
+known_errors.inc: $(srcdir)/template/known_errors.inc.tmpl $(srcdir)/defs/known_errors.def
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) $(srcdir)/tool/generic_erb.rb -c -o $@ $(srcdir)/template/known_errors.inc.tmpl $(srcdir)/defs/known_errors.def
+
+vm_call_iseq_optimized.inc: $(srcdir)/tool/mk_call_iseq_optimized.rb
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) $(srcdir)/tool/mk_call_iseq_optimized.rb > $@
+
+$(MINIPRELUDE_C): $(COMPILE_PRELUDE)
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) $(srcdir)/tool/generic_erb.rb -I$(srcdir) -o $@ \
+		$(srcdir)/template/prelude.c.tmpl
+
+$(PRELUDE_C): $(COMPILE_PRELUDE) \
+	   $(PRELUDE_SCRIPTS)
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) $(srcdir)/tool/generic_erb.rb -I$(srcdir) -c -o $@ \
+		$(srcdir)/template/prelude.c.tmpl $(PRELUDE_SCRIPTS)
+
+$(GOLF_PRELUDE_C): $(COMPILE_PRELUDE) {$(srcdir)}golf_prelude.rb
+	$(ECHO) generating $@
+	$(Q) $(BASERUBY) $(srcdir)/tool/generic_erb.rb -I$(srcdir) -c -o $@ \
+		$(srcdir)/template/prelude.c.tmpl golf_prelude.rb
+
+MAINCPPFLAGS = $(ENABLE_DEBUG_ENV:yes=-DRUBY_DEBUG_ENV=1)
+
+$(MAINOBJ): $(srcdir)/$(MAINSRC)
+	$(ECHO) compiling $(srcdir)/$(MAINSRC)
+	$(Q) $(CC) $(MAINCPPFLAGS) $(CFLAGS) $(XCFLAGS) $(CPPFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$(srcdir)/$(MAINSRC)
+
+{$(VPATH)}probes.dmyh: {$(srcdir)}probes.d $(srcdir)/tool/gen_dummy_probes.rb
+
+probes.dmyh:
+	$(BASERUBY) $(srcdir)/tool/gen_dummy_probes.rb $(srcdir)/probes.d > $@
+
+probes.h: {$(VPATH)}probes.$(DTRACE_EXT)
+
+prereq: incs srcs preludes PHONY
+
+preludes: {$(VPATH)}prelude.c
+preludes: {$(VPATH)}miniprelude.c
+preludes: {$(srcdir)}golf_prelude.c
+
+$(srcdir)/revision.h:
+	$(Q)$(gnumake:yes=#) $(RM) $(@F)
+	$(Q)$(gnumake:yes=#) exit > $@ || exit > $(@F)
+
+$(REVISION_H): $(srcdir)/version.h $(srcdir)/tool/file2lastrev.rb $(REVISION_FORCE)
+	-$(Q) $(BASERUBY) $(srcdir)/tool/file2lastrev.rb -q --revision.h "$(srcdir)" > revision.tmp
+	$(Q)$(IFCHANGE) "--timestamp=$@" "$(srcdir)/revision.h" revision.tmp
+
+$(srcdir)/ext/ripper/ripper.c: $(srcdir)/ext/ripper/tools/preproc.rb $(srcdir)/parse.y id.h
+	$(ECHO) generating $@
+	$(Q) VPATH=$${PWD-`pwd`} && $(CHDIR) $(@D) && \
+	sed -e 's/{\$$([^(){}]*)[^{}]*}//g' -e /AUTOGENERATED/q depend | \
+	$(exec) $(MAKE) -f - $(mflags) \
+		Q=$(Q) ECHO=$(ECHO) RM="$(RM)" top_srcdir=../.. srcdir=. VPATH="$${VPATH}" \
+		RUBY="$(BASERUBY)" PATH_SEPARATOR="$(PATH_SEPARATOR)"
+
+$(srcdir)/ext/json/parser/parser.c: $(srcdir)/ext/json/parser/parser.rl
+	$(ECHO) generating $@
+	$(Q) $(CHDIR) $(@D) && $(exec) $(MAKE) -f prereq.mk $(mflags) \
+		Q=$(Q) ECHO=$(ECHO) top_srcdir=../../.. srcdir=. VPATH=../../.. BASERUBY="$(BASERUBY)"
+
+$(srcdir)/ext/date/zonetab.h: $(srcdir)/ext/date/zonetab.list
+	$(ECHO) generating $@
+	$(Q) $(CHDIR) $(@D) && $(exec) $(MAKE) -f prereq.mk $(mflags) \
+		Q=$(Q) ECHO=$(ECHO) top_srcdir=../.. srcdir=. VPATH=../.. BASERUBY="$(BASERUBY)"
+
+$(srcdir)/ext/rbconfig/sizeof/sizes.c: $(srcdir)/ext/rbconfig/sizeof/depend \
+		$(srcdir)/tool/generic_erb.rb $(srcdir)/template/sizes.c.tmpl $(srcdir)/configure.ac
+	$(ECHO) generating $@
+	$(Q) $(CHDIR) $(@D) && \
+	sed '/AUTOGENERATED/q' depend | \
+	$(exec) $(MAKE) -f - $(mflags) \
+		Q=$(Q) ECHO=$(ECHO) top_srcdir=../../.. srcdir=. VPATH=../../.. RUBY="$(BASERUBY)" $(@F)
+
+$(srcdir)/ext/rbconfig/sizeof/limits.c: $(srcdir)/ext/rbconfig/sizeof/depend \
+		$(srcdir)/tool/generic_erb.rb $(srcdir)/template/limits.c.tmpl
+	$(ECHO) generating $@
+	$(Q) $(CHDIR) $(@D) && \
+	sed '/AUTOGENERATED/q' depend | \
+	$(exec) $(MAKE) -f - $(mflags) \
+		Q=$(Q) ECHO=$(ECHO) top_srcdir=../../.. srcdir=. VPATH=../../.. RUBY="$(BASERUBY)" $(@F)
+
+$(srcdir)/ext/socket/constdefs.c: $(srcdir)/ext/socket/depend
+	$(Q) $(CHDIR) $(@D) && \
+	sed '/AUTOGENERATED/q' depend | \
+	$(exec) $(MAKE) -f - $(mflags) \
+		Q=$(Q) ECHO=$(ECHO) top_srcdir=../.. srcdir=. VPATH=../.. RUBY="$(BASERUBY)"
+
+$(srcdir)/ext/etc/constdefs.h: $(srcdir)/ext/etc/depend
+	$(Q) $(CHDIR) $(@D) && \
+	sed '/AUTOGENERATED/q' depend | \
+	$(exec) $(MAKE) -f - $(mflags) \
+		Q=$(Q) ECHO=$(ECHO) top_srcdir=../.. srcdir=. VPATH=../.. RUBY="$(BASERUBY)"
+
+##
+
+run: fake miniruby$(EXEEXT) PHONY
+	$(BTESTRUBY) $(TESTRUN_SCRIPT) $(RUNOPT)
+
+runruby: $(PROGRAM) PHONY
+	$(RUNRUBY) $(TESTRUN_SCRIPT)
+
+parse: fake miniruby$(EXEEXT) PHONY
+	$(BTESTRUBY) --dump=parsetree_with_comment,insns $(TESTRUN_SCRIPT)
+
+bisect: PHONY
+	$(srcdir)/tool/bisect.sh miniruby $(srcdir)
+
+bisect-ruby: PHONY
+	$(srcdir)/tool/bisect.sh ruby $(srcdir)
+
+COMPARE_RUBY = $(BASERUBY)
+ITEM =
+ARGS = $$(find $(srcdir)/benchmark -maxdepth 1 -name '*$(ITEM)*.yml' -o -name '*$(ITEM)*.rb' | sort)
+OPTS =
+
+# You can pass several options through OPTS environment variable.
+# $ make benchmark ARGS="--help" displays more detail.
+# for example,
+#  $ make benchmark COMPARE_RUBY="ruby-trunk" OPTS="-e ruby-2.2.2"
+# This command compares trunk and built-ruby and 2.2.2
+benchmark: miniruby$(EXEEXT) update-benchmark-driver PHONY
+	$(BASERUBY) -rrubygems -I$(srcdir)/benchmark/lib $(srcdir)/benchmark/benchmark-driver/exe/benchmark-driver \
+	            --executables="compare-ruby::$(COMPARE_RUBY) -I$(EXTOUT)/common --disable-gem" \
+	            --executables="built-ruby::$(MINIRUBY) -r$(srcdir)/prelude --disable-gem" \
+	            $(ARGS) $(OPTS)
+
+run.gdb:
+	echo set breakpoint pending on         > run.gdb
+	echo b ruby_debug_breakpoint          >> run.gdb
+	echo '# handle SIGINT nostop'         >> run.gdb
+	echo '# handle SIGPIPE nostop'        >> run.gdb
+	echo '# b rb_longjmp'                 >> run.gdb
+	echo source $(srcdir)/breakpoints.gdb >> run.gdb
+	echo source $(srcdir)/.gdbinit        >> run.gdb
+	echo 'set $$_exitcode = -999'         >> run.gdb
+	echo run                              >> run.gdb
+	echo 'if $$_exitcode != -999'         >> run.gdb
+	echo '  quit'                         >> run.gdb
+	echo end                              >> run.gdb
+
+
+gdb: miniruby$(EXEEXT) run.gdb PHONY
+	gdb -x run.gdb --quiet --args $(MINIRUBY) $(TESTRUN_SCRIPT)
+
+gdb-ruby: $(PROGRAM) run.gdb PHONY
+	$(Q) $(RUNRUBY_COMMAND) $(RUNRUBY_DEBUGGER) -- $(TESTRUN_SCRIPT)
+
+LLDB_INIT = command script import -r $(srcdir)/misc/lldb_cruby.py
+
+lldb: miniruby$(EXEEXT) PHONY
+	lldb -o '$(LLDB_INIT)' miniruby$(EXEEXT) -- $(TESTRUN_SCRIPT)
+
+lldb-ruby: $(PROGRAM) PHONY
+	lldb $(enable_shared:yes=-o 'target modules add ${LIBRUBY_SO}') -o '$(LLDB_INIT)' $(PROGRAM) -- $(TESTRUN_SCRIPT)
+
+DISTPKGS = gzip,zip,all
+dist:
+	$(BASERUBY) $(srcdir)/tool/make-snapshot \
+	-srcdir=$(srcdir) -packages=$(DISTPKGS) \
+	-unicode-version=$(UNICODE_VERSION) \
+	tmp $(RELNAME)
+
+up:: update-remote
+
+up::
+	-$(Q)$(MAKE) $(mflags) Q=$(Q) REVISION_FORCE=PHONY "$(REVISION_H)"
+
+up::
+	-$(Q)$(MAKE) $(mflags) Q=$(Q) after-update
+
+after-update:: extract-extlibs
+
+update-remote:: update-src update-download
+update-download:: update-unicode update-gems download-extlibs
+
+update-mspec:
+update-rubyspec:
+
+update-config_files: PHONY
+	$(Q) $(BASERUBY) -C "$(srcdir)" tool/downloader.rb -d tool --cache-dir=$(CACHE_DIR) -e gnu \
+	    config.guess config.sub
+
+update-gems: PHONY
+	$(ECHO) Downloading bundled gem files...
+	$(Q) $(BASERUBY) -C "$(srcdir)" \
+	    -I./tool -rdownloader -answ \
+	    -e 'gem, ver = *$$F' \
+	    -e 'old = Dir.glob("gems/#{gem}-*.gem")' \
+	    -e 'gem = "#{gem}-#{ver}.gem"' \
+	    -e 'Downloader::RubyGems.download(gem, "gems", nil) and' \
+	    -e '(old.delete("gems/#{gem}"); !old.empty?) and' \
+	    -e 'File.unlink(*old) and' \
+	    -e 'FileUtils.rm_rf(old.map{'"|n|"'n.chomp(".gem")})' \
+	    gems/bundled_gems
+
+extract-gems: PHONY
+	$(ECHO) Extracting bundled gem files...
+	$(Q) $(RUNRUBY) -C "$(srcdir)/gems" \
+	    -I../tool -rgem-unpack -answ \
+	    -e 'gem, ver = *$$F' \
+	    -e 'Gem.unpack("#{gem}-#{ver}.gem")' \
+	    bundled_gems
+
+update-bundled_gems: PHONY
+	$(Q) $(RUNRUBY) -rrubygems \
+	    -pla \
+	    -e '(gem,src), = Gem::SpecFetcher.fetcher.detect(:latest) {'"|s|" \
+	    -e   's.platform=="ruby"&&s.name==$$F[0]' \
+	    -e '}' \
+	    -e 'gem = src.fetch_spec(gem)' \
+	    -e '$$_ = [gem.name, gem.version, gem.metadata["source_code_uri"]||gem.homepage].join(" ")' \
+	     "$(srcdir)/gems/bundled_gems" | \
+	"$(IFCHANGE)" "$(srcdir)/gems/bundled_gems" -
+
+test-bundled-gems-precheck: $(arch)-fake.rb programs
+
+test-bundled-gems-fetch: $(PREP)
+	$(Q) $(BASERUBY) -C $(srcdir)/gems ../tool/fetch-bundled_gems.rb src bundled_gems
+
+test-bundled-gems-prepare: test-bundled-gems-precheck test-bundled-gems-fetch
+	$(XRUBY) -C "$(srcdir)" bin/gem install --no-document \
+		--install-dir .bundle --conservative "bundler" "minitest:~> 5" 'test-unit' 'rake' 'hoe' 'yard' 'pry' 'packnga'
+
+PREPARE_BUNDLED_GEMS = test-bundled-gems-prepare
+test-bundled-gems: $(TEST_RUNNABLE)-test-bundled-gems
+yes-test-bundled-gems: test-bundled-gems-run
+no-test-bundled-gems:
+test-bundled-gems-run: $(PREPARE_BUNDLED_GEMS)
+
+test-bundler-precheck: $(arch)-fake.rb programs
+
+yes-test-bundler-prepare: test-bundler-precheck
+	$(XRUBY) -C "$(srcdir)" bin/gem install --no-document \
+		--install-dir .bundle --conservative "rspec:~> 3.5"
+
+RSPECOPTS = --format progress
+BUNDLER_SPECS =
+test-bundler: $(TEST_RUNNABLE)-test-bundler
+yes-test-bundler: yes-test-bundler-prepare
+	$(gnumake_recursive)$(Q) \
+	$(XRUBY) -C $(srcdir) -Ispec/bundler .bundle/bin/rspec \
+		--require spec_helper $(RSPECOPTS) spec/bundler/$(BUNDLER_SPECS)
+no-test-bundler:
+
+UNICODE_FILES = $(UNICODE_SRC_DATA_DIR)/UnicodeData.txt \
+		$(UNICODE_SRC_DATA_DIR)/CompositionExclusions.txt \
+		$(UNICODE_SRC_DATA_DIR)/NormalizationTest.txt \
+		$(UNICODE_SRC_DATA_DIR)/CaseFolding.txt \
+		$(UNICODE_SRC_DATA_DIR)/SpecialCasing.txt \
+		$(empty)
+
+UNICODE_PROPERTY_FILES =  \
+		$(UNICODE_SRC_DATA_DIR)/Blocks.txt \
+		$(UNICODE_SRC_DATA_DIR)/DerivedAge.txt \
+		$(UNICODE_SRC_DATA_DIR)/DerivedCoreProperties.txt \
+		$(UNICODE_SRC_DATA_DIR)/PropList.txt \
+		$(UNICODE_SRC_DATA_DIR)/PropertyAliases.txt \
+		$(UNICODE_SRC_DATA_DIR)/PropertyValueAliases.txt \
+		$(UNICODE_SRC_DATA_DIR)/Scripts.txt \
+		$(UNICODE_SRC_DATA_DIR)/auxiliary/GraphemeBreakProperty.txt \
+		$(UNICODE_SRC_DATA_DIR)/auxiliary/GraphemeBreakTest.txt \
+		$(empty)
+
+UNICODE_EMOJI_FILES = \
+		$(UNICODE_SRC_EMOJI_DATA_DIR)/emoji-data.txt \
+		$(UNICODE_SRC_EMOJI_DATA_DIR)/emoji-sequences.txt \
+		$(UNICODE_SRC_EMOJI_DATA_DIR)/emoji-test.txt \
+		$(UNICODE_SRC_EMOJI_DATA_DIR)/emoji-variation-sequences.txt \
+		$(UNICODE_SRC_EMOJI_DATA_DIR)/emoji-zwj-sequences.txt \
+		$(empty)
+
+update-unicode: $(UNICODE_FILES) $(UNICODE_PROPERTY_FILES)
+
+CACHE_DIR = $(srcdir)/.downloaded-cache
+UNICODE_DOWNLOAD = \
+	$(BASERUBY) $(srcdir)/tool/downloader.rb \
+	    --cache-dir=$(CACHE_DIR) \
+	    -d $(UNICODE_SRC_DATA_DIR) \
+	    -p $(UNICODE_VERSION)/ucd \
+	    -e $(ALWAYS_UPDATE_UNICODE:yes=-a) unicode
+UNICODE_EMOJI_DOWNLOAD = \
+	$(BASERUBY) $(srcdir)/tool/downloader.rb \
+	    --cache-dir=$(CACHE_DIR) \
+	    -d $(UNICODE_SRC_EMOJI_DATA_DIR) \
+	    -p emoji/$(UNICODE_EMOJI_VERSION) \
+	    -e $(ALWAYS_UPDATE_UNICODE:yes=-a) unicode
+
+$(UNICODE_PROPERTY_FILES): update-unicode-property-files
+update-unicode-property-files:
+	$(ECHO) Downloading Unicode $(UNICODE_VERSION) property files...
+	$(Q) $(MAKEDIRS) "$(UNICODE_SRC_DATA_DIR)/auxiliary"
+	$(Q) $(UNICODE_DOWNLOAD) $(UNICODE_PROPERTY_FILES)
+	$(ECHO) Downloading Unicode emoji $(UNICODE_EMOJI_VERSION) files...
+	$(Q) $(MAKEDIRS) "$(UNICODE_SRC_EMOJI_DATA_DIR)"
+	$(Q) $(UNICODE_EMOJI_DOWNLOAD) $(UNICODE_EMOJI_FILES)
+
+$(UNICODE_FILES): update-unicode-files
+update-unicode-files:
+	$(ECHO) Downloading Unicode $(UNICODE_VERSION) data files...
+	$(Q) $(MAKEDIRS) "$(UNICODE_SRC_DATA_DIR)"
+	$(Q) $(UNICODE_DOWNLOAD) $(UNICODE_FILES)
+
+$(srcdir)/$(HAVE_BASERUBY:yes=lib/unicode_normalize/tables.rb): \
+	$(UNICODE_SRC_DATA_DIR)/.unicode-tables.time
+
+$(UNICODE_SRC_DATA_DIR)/$(ALWAYS_UPDATE_UNICODE:yes=.unicode-tables.time): \
+	$(UNICODE_FILES) $(UNICODE_PROPERTY_FILES)
+
+touch-unicode-files:
+	$(MAKEDIRS) $(UNICODE_SRC_DATA_DIR)
+	touch $(UNICODE_SRC_DATA_DIR)/.unicode-tables.time $(UNICODE_DATA_HEADERS)
+
+$(UNICODE_SRC_DATA_DIR)/.unicode-tables.time: $(srcdir)/tool/generic_erb.rb \
+		$(srcdir)/template/unicode_norm_gen.tmpl \
+		$(ALWAYS_UPDATE_UNICODE:yes=update-unicode)
+	$(Q) $(MAKE) $(@D)
+	$(Q) $(BASERUBY) $(srcdir)/tool/generic_erb.rb \
+		-c -t$@ -o $(srcdir)/lib/unicode_normalize/tables.rb \
+		-I $(srcdir) \
+		$(srcdir)/template/unicode_norm_gen.tmpl \
+		$(UNICODE_DATA_DIR) lib/unicode_normalize
+
+$(UNICODE_SRC_DATA_DIR):
+	$(Q) $(exec) $(MAKEDIRS) $@ || exit && echo $(MAKE)
+
+$(UNICODE_HDR_DIR)/$(ALWAYS_UPDATE_UNICODE:yes=name2ctype.h): \
+		$(srcdir)/tool/enc-unicode.rb \
+		$(UNICODE_SRC_DATA_DIR)/UnicodeData.txt \
+		$(UNICODE_PROPERTY_FILES)
+
+$(UNICODE_HDR_DIR)/name2ctype.h:
+	$(MAKEDIRS) $(@D)
+	$(BOOTSTRAPRUBY) $(srcdir)/tool/enc-unicode.rb --header \
+		$(UNICODE_SRC_DATA_DIR) $(UNICODE_SRC_EMOJI_DATA_DIR) > $@.new
+	$(MV) $@.new $@
+
+# the next non-comment line was:
+# $(UNICODE_HDR_DIR)/casefold.h: $(srcdir)/enc/unicode/case-folding.rb \
+# but was changed to make sure CI works on systems that don't have gperf
+unicode-up: $(UNICODE_DATA_HEADERS)
+
+$(UNICODE_HDR_DIR)/$(ALWAYS_UPDATE_UNICODE:yes=casefold.h): \
+		$(srcdir)/enc/unicode/case-folding.rb \
+		$(UNICODE_SRC_DATA_DIR)/UnicodeData.txt \
+		$(UNICODE_SRC_DATA_DIR)/SpecialCasing.txt \
+		$(UNICODE_SRC_DATA_DIR)/CaseFolding.txt
+
+$(UNICODE_HDR_DIR)/casefold.h:
+	$(MAKEDIRS) $(@D)
+	$(Q) $(BASERUBY) $(srcdir)/enc/unicode/case-folding.rb \
+		--output-file=$@ \
+		--mapping-data-directory=$(UNICODE_SRC_DATA_DIR)
+
+download-extlibs:
+	$(Q) $(BASERUBY) -C $(srcdir) -w tool/extlibs.rb --download ext
+
+extract-extlibs:
+	$(Q) $(BASERUBY) -C $(srcdir) -w tool/extlibs.rb --all ext
+
+clean-extlibs:
+	$(Q) $(RMALL) $(srcdir)/.downloaded-cache
+
+clean-gems:
+	$(Q) $(RM) gems/*.gem
+
+CLEAN_CACHE = clean-extlibs
+
+info: info-program info-libruby_a info-libruby_so info-arch
+info-program: PHONY
+	@echo PROGRAM=$(PROGRAM)
+info-libruby_a: PHONY
+	@echo LIBRUBY_A=$(LIBRUBY_A)
+info-libruby_so: PHONY
+	@echo LIBRUBY_SO=$(LIBRUBY_SO)
+info-arch: PHONY
+	@echo arch=$(arch)
+
+change: PHONY
+	$(BASERUBY) -C "$(srcdir)" ./tool/change_maker.rb $(CHANGES) > change.log
+
+exam: check
+
+love: sudo-precheck up all test exam install
+	@echo love is all you need
+
+great: exam
+
+yes-test-all no-test-all: sudo-precheck
+
+sudo-precheck: PHONY
+	@$(SUDO) echo > $(NULL)
+
+update-man-date: PHONY
+	-$(Q) $(BASERUBY) -I"$(srcdir)/tool" -rvcs -i -p \
+	-e 'BEGIN{@vcs=VCS.detect(ARGV.shift)}' \
+	-e '$$_.sub!(/^(\.Dd ).*/){$$1+@vcs.modified(ARGF.path).strftime("%B %d, %Y")}' \
+	"$(srcdir)" "$(srcdir)"/man/*.1
+
+help: PHONY
+	$(MESSAGE_BEGIN) \
+	"                Makefile of Ruby" \
+	"" \
+	"targets:" \
+	"  all (default):       builds all of below" \
+	"  miniruby:            builds only miniruby" \
+	"  encs:                builds encodings" \
+	"  exts:                builds extensions" \
+	"  main:                builds encodings, extensions and ruby" \
+	"  docs:                builds documents" \
+	"  install-capi:        builds C API documents" \
+	"  run:                 runs test.rb by miniruby" \
+	"  runruby:             runs test.rb by ruby you just built" \
+	"  gdb:                 runs test.rb by miniruby under gdb" \
+	"  gdb-ruby:            runs test.rb by ruby under gdb" \
+	"  check:               equals make test test-all test-spec" \
+	"  exam:                equals make check" \
+	"  test:                ruby core tests" \
+	"  test-all:            all ruby tests [TESTOPTS=-j4 TESTS=<test files>]" \
+	"  test-spec:           run the Ruby spec suite" \
+	"  test-rubyspec:       same as test-spec" \
+	"  test-bundler:        run the Bundler spec" \
+	"  test-bundled-gems:   run the test suite of bundled gems" \
+	"  up:                  update local copy and autogenerated files" \
+	"  benchmark:           benchmark this ruby and COMPARE_RUBY." \
+	"  gcbench:             gc benchmark [GCBENCH_ITEM=<item_name>]" \
+	"  gcbench-rdoc:        gc benchmark with GCBENCH_ITEM=rdoc" \
+	"  install:             install all ruby distributions" \
+	"  install-nodoc:       install without rdoc" \
+	"  install-cross:       install cross compiling stuff" \
+	"  clean:               clean for tarball" \
+	"  distclean:           clean for repository" \
+	"  change:              make change log template" \
+	"  golf:                for golfers" \
+	"  goruby:              same as golf" \
+	"" \
+	"see DeveloperHowto for more detail: " \
+	"  https://bugs.ruby-lang.org/projects/ruby/wiki/DeveloperHowto" \
+	$(MESSAGE_END)
+
+# AUTOGENERATED DEPENDENCIES START
+addr2line.$(OBJEXT): {$(VPATH)}addr2line.c
+addr2line.$(OBJEXT): {$(VPATH)}addr2line.h
+addr2line.$(OBJEXT): {$(VPATH)}config.h
+addr2line.$(OBJEXT): {$(VPATH)}defines.h
+addr2line.$(OBJEXT): {$(VPATH)}missing.h
+array.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+array.$(OBJEXT): $(top_srcdir)/include/ruby.h
+array.$(OBJEXT): {$(VPATH)}array.c
+array.$(OBJEXT): {$(VPATH)}config.h
+array.$(OBJEXT): {$(VPATH)}debug_counter.h
+array.$(OBJEXT): {$(VPATH)}defines.h
+array.$(OBJEXT): {$(VPATH)}encoding.h
+array.$(OBJEXT): {$(VPATH)}gc.h
+array.$(OBJEXT): {$(VPATH)}id.h
+array.$(OBJEXT): {$(VPATH)}intern.h
+array.$(OBJEXT): {$(VPATH)}internal.h
+array.$(OBJEXT): {$(VPATH)}io.h
+array.$(OBJEXT): {$(VPATH)}missing.h
+array.$(OBJEXT): {$(VPATH)}onigmo.h
+array.$(OBJEXT): {$(VPATH)}oniguruma.h
+array.$(OBJEXT): {$(VPATH)}probes.dmyh
+array.$(OBJEXT): {$(VPATH)}probes.h
+array.$(OBJEXT): {$(VPATH)}ruby_assert.h
+array.$(OBJEXT): {$(VPATH)}st.h
+array.$(OBJEXT): {$(VPATH)}subst.h
+array.$(OBJEXT): {$(VPATH)}transient_heap.h
+array.$(OBJEXT): {$(VPATH)}util.h
+ast.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+ast.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+ast.$(OBJEXT): $(CCAN_DIR)/list/list.h
+ast.$(OBJEXT): $(CCAN_DIR)/str/str.h
+ast.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+ast.$(OBJEXT): $(top_srcdir)/include/ruby.h
+ast.$(OBJEXT): {$(VPATH)}ast.c
+ast.$(OBJEXT): {$(VPATH)}config.h
+ast.$(OBJEXT): {$(VPATH)}defines.h
+ast.$(OBJEXT): {$(VPATH)}encoding.h
+ast.$(OBJEXT): {$(VPATH)}id.h
+ast.$(OBJEXT): {$(VPATH)}intern.h
+ast.$(OBJEXT): {$(VPATH)}internal.h
+ast.$(OBJEXT): {$(VPATH)}iseq.h
+ast.$(OBJEXT): {$(VPATH)}method.h
+ast.$(OBJEXT): {$(VPATH)}missing.h
+ast.$(OBJEXT): {$(VPATH)}node.h
+ast.$(OBJEXT): {$(VPATH)}onigmo.h
+ast.$(OBJEXT): {$(VPATH)}oniguruma.h
+ast.$(OBJEXT): {$(VPATH)}ruby_assert.h
+ast.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+ast.$(OBJEXT): {$(VPATH)}st.h
+ast.$(OBJEXT): {$(VPATH)}subst.h
+ast.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+ast.$(OBJEXT): {$(VPATH)}thread_native.h
+ast.$(OBJEXT): {$(VPATH)}util.h
+ast.$(OBJEXT): {$(VPATH)}vm_core.h
+ast.$(OBJEXT): {$(VPATH)}vm_opts.h
+bignum.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+bignum.$(OBJEXT): $(top_srcdir)/include/ruby.h
+bignum.$(OBJEXT): {$(VPATH)}bignum.c
+bignum.$(OBJEXT): {$(VPATH)}config.h
+bignum.$(OBJEXT): {$(VPATH)}defines.h
+bignum.$(OBJEXT): {$(VPATH)}encoding.h
+bignum.$(OBJEXT): {$(VPATH)}id.h
+bignum.$(OBJEXT): {$(VPATH)}intern.h
+bignum.$(OBJEXT): {$(VPATH)}internal.h
+bignum.$(OBJEXT): {$(VPATH)}io.h
+bignum.$(OBJEXT): {$(VPATH)}missing.h
+bignum.$(OBJEXT): {$(VPATH)}onigmo.h
+bignum.$(OBJEXT): {$(VPATH)}oniguruma.h
+bignum.$(OBJEXT): {$(VPATH)}ruby_assert.h
+bignum.$(OBJEXT): {$(VPATH)}st.h
+bignum.$(OBJEXT): {$(VPATH)}subst.h
+bignum.$(OBJEXT): {$(VPATH)}thread.h
+bignum.$(OBJEXT): {$(VPATH)}util.h
+class.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+class.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+class.$(OBJEXT): $(CCAN_DIR)/list/list.h
+class.$(OBJEXT): $(CCAN_DIR)/str/str.h
+class.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+class.$(OBJEXT): $(top_srcdir)/include/ruby.h
+class.$(OBJEXT): {$(VPATH)}class.c
+class.$(OBJEXT): {$(VPATH)}config.h
+class.$(OBJEXT): {$(VPATH)}constant.h
+class.$(OBJEXT): {$(VPATH)}defines.h
+class.$(OBJEXT): {$(VPATH)}encoding.h
+class.$(OBJEXT): {$(VPATH)}id.h
+class.$(OBJEXT): {$(VPATH)}id_table.h
+class.$(OBJEXT): {$(VPATH)}intern.h
+class.$(OBJEXT): {$(VPATH)}internal.h
+class.$(OBJEXT): {$(VPATH)}io.h
+class.$(OBJEXT): {$(VPATH)}method.h
+class.$(OBJEXT): {$(VPATH)}missing.h
+class.$(OBJEXT): {$(VPATH)}node.h
+class.$(OBJEXT): {$(VPATH)}onigmo.h
+class.$(OBJEXT): {$(VPATH)}oniguruma.h
+class.$(OBJEXT): {$(VPATH)}ruby_assert.h
+class.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+class.$(OBJEXT): {$(VPATH)}st.h
+class.$(OBJEXT): {$(VPATH)}subst.h
+class.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+class.$(OBJEXT): {$(VPATH)}thread_native.h
+class.$(OBJEXT): {$(VPATH)}vm_core.h
+class.$(OBJEXT): {$(VPATH)}vm_debug.h
+class.$(OBJEXT): {$(VPATH)}vm_opts.h
+compar.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+compar.$(OBJEXT): {$(VPATH)}compar.c
+compar.$(OBJEXT): {$(VPATH)}config.h
+compar.$(OBJEXT): {$(VPATH)}defines.h
+compar.$(OBJEXT): {$(VPATH)}id.h
+compar.$(OBJEXT): {$(VPATH)}intern.h
+compar.$(OBJEXT): {$(VPATH)}missing.h
+compar.$(OBJEXT): {$(VPATH)}st.h
+compar.$(OBJEXT): {$(VPATH)}subst.h
+compile.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+compile.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+compile.$(OBJEXT): $(CCAN_DIR)/list/list.h
+compile.$(OBJEXT): $(CCAN_DIR)/str/str.h
+compile.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+compile.$(OBJEXT): $(top_srcdir)/include/ruby.h
+compile.$(OBJEXT): {$(VPATH)}compile.c
+compile.$(OBJEXT): {$(VPATH)}config.h
+compile.$(OBJEXT): {$(VPATH)}defines.h
+compile.$(OBJEXT): {$(VPATH)}encindex.h
+compile.$(OBJEXT): {$(VPATH)}encoding.h
+compile.$(OBJEXT): {$(VPATH)}gc.h
+compile.$(OBJEXT): {$(VPATH)}id.h
+compile.$(OBJEXT): {$(VPATH)}id_table.h
+compile.$(OBJEXT): {$(VPATH)}insns.def
+compile.$(OBJEXT): {$(VPATH)}insns.inc
+compile.$(OBJEXT): {$(VPATH)}insns_info.inc
+compile.$(OBJEXT): {$(VPATH)}intern.h
+compile.$(OBJEXT): {$(VPATH)}internal.h
+compile.$(OBJEXT): {$(VPATH)}io.h
+compile.$(OBJEXT): {$(VPATH)}iseq.h
+compile.$(OBJEXT): {$(VPATH)}method.h
+compile.$(OBJEXT): {$(VPATH)}missing.h
+compile.$(OBJEXT): {$(VPATH)}node.h
+compile.$(OBJEXT): {$(VPATH)}onigmo.h
+compile.$(OBJEXT): {$(VPATH)}oniguruma.h
+compile.$(OBJEXT): {$(VPATH)}opt_sc.inc
+compile.$(OBJEXT): {$(VPATH)}optinsn.inc
+compile.$(OBJEXT): {$(VPATH)}optunifs.inc
+compile.$(OBJEXT): {$(VPATH)}re.h
+compile.$(OBJEXT): {$(VPATH)}regex.h
+compile.$(OBJEXT): {$(VPATH)}ruby_assert.h
+compile.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+compile.$(OBJEXT): {$(VPATH)}st.h
+compile.$(OBJEXT): {$(VPATH)}subst.h
+compile.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+compile.$(OBJEXT): {$(VPATH)}thread_native.h
+compile.$(OBJEXT): {$(VPATH)}util.h
+compile.$(OBJEXT): {$(VPATH)}vm_core.h
+compile.$(OBJEXT): {$(VPATH)}vm_debug.h
+compile.$(OBJEXT): {$(VPATH)}vm_opts.h
+complex.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+complex.$(OBJEXT): $(top_srcdir)/include/ruby.h
+complex.$(OBJEXT): {$(VPATH)}complex.c
+complex.$(OBJEXT): {$(VPATH)}config.h
+complex.$(OBJEXT): {$(VPATH)}defines.h
+complex.$(OBJEXT): {$(VPATH)}encoding.h
+complex.$(OBJEXT): {$(VPATH)}id.h
+complex.$(OBJEXT): {$(VPATH)}intern.h
+complex.$(OBJEXT): {$(VPATH)}internal.h
+complex.$(OBJEXT): {$(VPATH)}io.h
+complex.$(OBJEXT): {$(VPATH)}missing.h
+complex.$(OBJEXT): {$(VPATH)}onigmo.h
+complex.$(OBJEXT): {$(VPATH)}oniguruma.h
+complex.$(OBJEXT): {$(VPATH)}ruby_assert.h
+complex.$(OBJEXT): {$(VPATH)}st.h
+complex.$(OBJEXT): {$(VPATH)}subst.h
+cont.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+cont.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+cont.$(OBJEXT): $(CCAN_DIR)/list/list.h
+cont.$(OBJEXT): $(CCAN_DIR)/str/str.h
+cont.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+cont.$(OBJEXT): $(top_srcdir)/include/ruby.h
+cont.$(OBJEXT): {$(VPATH)}$(COROUTINE_H)
+cont.$(OBJEXT): {$(VPATH)}config.h
+cont.$(OBJEXT): {$(VPATH)}cont.c
+cont.$(OBJEXT): {$(VPATH)}defines.h
+cont.$(OBJEXT): {$(VPATH)}encoding.h
+cont.$(OBJEXT): {$(VPATH)}eval_intern.h
+cont.$(OBJEXT): {$(VPATH)}gc.h
+cont.$(OBJEXT): {$(VPATH)}id.h
+cont.$(OBJEXT): {$(VPATH)}intern.h
+cont.$(OBJEXT): {$(VPATH)}internal.h
+cont.$(OBJEXT): {$(VPATH)}io.h
+cont.$(OBJEXT): {$(VPATH)}method.h
+cont.$(OBJEXT): {$(VPATH)}missing.h
+cont.$(OBJEXT): {$(VPATH)}mjit.h
+cont.$(OBJEXT): {$(VPATH)}node.h
+cont.$(OBJEXT): {$(VPATH)}onigmo.h
+cont.$(OBJEXT): {$(VPATH)}oniguruma.h
+cont.$(OBJEXT): {$(VPATH)}ruby_assert.h
+cont.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+cont.$(OBJEXT): {$(VPATH)}st.h
+cont.$(OBJEXT): {$(VPATH)}subst.h
+cont.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+cont.$(OBJEXT): {$(VPATH)}thread_native.h
+cont.$(OBJEXT): {$(VPATH)}vm_core.h
+cont.$(OBJEXT): {$(VPATH)}vm_debug.h
+cont.$(OBJEXT): {$(VPATH)}vm_opts.h
+debug.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+debug.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+debug.$(OBJEXT): $(CCAN_DIR)/list/list.h
+debug.$(OBJEXT): $(CCAN_DIR)/str/str.h
+debug.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+debug.$(OBJEXT): $(top_srcdir)/include/ruby.h
+debug.$(OBJEXT): {$(VPATH)}config.h
+debug.$(OBJEXT): {$(VPATH)}debug.c
+debug.$(OBJEXT): {$(VPATH)}defines.h
+debug.$(OBJEXT): {$(VPATH)}encoding.h
+debug.$(OBJEXT): {$(VPATH)}eval_intern.h
+debug.$(OBJEXT): {$(VPATH)}gc.h
+debug.$(OBJEXT): {$(VPATH)}id.h
+debug.$(OBJEXT): {$(VPATH)}intern.h
+debug.$(OBJEXT): {$(VPATH)}internal.h
+debug.$(OBJEXT): {$(VPATH)}io.h
+debug.$(OBJEXT): {$(VPATH)}method.h
+debug.$(OBJEXT): {$(VPATH)}missing.h
+debug.$(OBJEXT): {$(VPATH)}node.h
+debug.$(OBJEXT): {$(VPATH)}onigmo.h
+debug.$(OBJEXT): {$(VPATH)}oniguruma.h
+debug.$(OBJEXT): {$(VPATH)}ruby_assert.h
+debug.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+debug.$(OBJEXT): {$(VPATH)}st.h
+debug.$(OBJEXT): {$(VPATH)}subst.h
+debug.$(OBJEXT): {$(VPATH)}symbol.h
+debug.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+debug.$(OBJEXT): {$(VPATH)}thread_native.h
+debug.$(OBJEXT): {$(VPATH)}util.h
+debug.$(OBJEXT): {$(VPATH)}vm_core.h
+debug.$(OBJEXT): {$(VPATH)}vm_debug.h
+debug.$(OBJEXT): {$(VPATH)}vm_opts.h
+debug_counter.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+debug_counter.$(OBJEXT): $(top_srcdir)/include/ruby.h
+debug_counter.$(OBJEXT): {$(VPATH)}config.h
+debug_counter.$(OBJEXT): {$(VPATH)}debug_counter.c
+debug_counter.$(OBJEXT): {$(VPATH)}debug_counter.h
+debug_counter.$(OBJEXT): {$(VPATH)}defines.h
+debug_counter.$(OBJEXT): {$(VPATH)}encoding.h
+debug_counter.$(OBJEXT): {$(VPATH)}intern.h
+debug_counter.$(OBJEXT): {$(VPATH)}internal.h
+debug_counter.$(OBJEXT): {$(VPATH)}io.h
+debug_counter.$(OBJEXT): {$(VPATH)}missing.h
+debug_counter.$(OBJEXT): {$(VPATH)}onigmo.h
+debug_counter.$(OBJEXT): {$(VPATH)}oniguruma.h
+debug_counter.$(OBJEXT): {$(VPATH)}st.h
+debug_counter.$(OBJEXT): {$(VPATH)}subst.h
+dir.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+dir.$(OBJEXT): $(top_srcdir)/include/ruby.h
+dir.$(OBJEXT): {$(VPATH)}config.h
+dir.$(OBJEXT): {$(VPATH)}defines.h
+dir.$(OBJEXT): {$(VPATH)}dir.c
+dir.$(OBJEXT): {$(VPATH)}encindex.h
+dir.$(OBJEXT): {$(VPATH)}encoding.h
+dir.$(OBJEXT): {$(VPATH)}id.h
+dir.$(OBJEXT): {$(VPATH)}intern.h
+dir.$(OBJEXT): {$(VPATH)}internal.h
+dir.$(OBJEXT): {$(VPATH)}io.h
+dir.$(OBJEXT): {$(VPATH)}missing.h
+dir.$(OBJEXT): {$(VPATH)}onigmo.h
+dir.$(OBJEXT): {$(VPATH)}oniguruma.h
+dir.$(OBJEXT): {$(VPATH)}st.h
+dir.$(OBJEXT): {$(VPATH)}subst.h
+dir.$(OBJEXT): {$(VPATH)}thread.h
+dir.$(OBJEXT): {$(VPATH)}util.h
+dln.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+dln.$(OBJEXT): $(top_srcdir)/include/ruby.h
+dln.$(OBJEXT): {$(VPATH)}config.h
+dln.$(OBJEXT): {$(VPATH)}defines.h
+dln.$(OBJEXT): {$(VPATH)}dln.c
+dln.$(OBJEXT): {$(VPATH)}dln.h
+dln.$(OBJEXT): {$(VPATH)}intern.h
+dln.$(OBJEXT): {$(VPATH)}internal.h
+dln.$(OBJEXT): {$(VPATH)}missing.h
+dln.$(OBJEXT): {$(VPATH)}st.h
+dln.$(OBJEXT): {$(VPATH)}subst.h
+dln_find.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+dln_find.$(OBJEXT): {$(VPATH)}config.h
+dln_find.$(OBJEXT): {$(VPATH)}defines.h
+dln_find.$(OBJEXT): {$(VPATH)}dln.h
+dln_find.$(OBJEXT): {$(VPATH)}dln_find.c
+dln_find.$(OBJEXT): {$(VPATH)}intern.h
+dln_find.$(OBJEXT): {$(VPATH)}missing.h
+dln_find.$(OBJEXT): {$(VPATH)}st.h
+dln_find.$(OBJEXT): {$(VPATH)}subst.h
+dmydln.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+dmydln.$(OBJEXT): {$(VPATH)}config.h
+dmydln.$(OBJEXT): {$(VPATH)}defines.h
+dmydln.$(OBJEXT): {$(VPATH)}dmydln.c
+dmydln.$(OBJEXT): {$(VPATH)}intern.h
+dmydln.$(OBJEXT): {$(VPATH)}missing.h
+dmydln.$(OBJEXT): {$(VPATH)}st.h
+dmydln.$(OBJEXT): {$(VPATH)}subst.h
+dmyenc.$(OBJEXT): {$(VPATH)}dmyenc.c
 dmyext.$(OBJEXT): {$(VPATH)}dmyext.c
-enum.$(OBJEXT): {$(VPATH)}enum.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}node.h {$(VPATH)}util.h
-error.$(OBJEXT): {$(VPATH)}error.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}env.h {$(VPATH)}st.h
-eval.$(OBJEXT): {$(VPATH)}eval.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}node.h {$(VPATH)}env.h {$(VPATH)}util.h \
-  {$(VPATH)}rubysig.h {$(VPATH)}st.h {$(VPATH)}dln.h
-file.$(OBJEXT): {$(VPATH)}file.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}rubyio.h {$(VPATH)}rubysig.h {$(VPATH)}util.h \
-  {$(VPATH)}dln.h
-gc.$(OBJEXT): {$(VPATH)}gc.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}rubysig.h {$(VPATH)}st.h {$(VPATH)}node.h \
-  {$(VPATH)}env.h {$(VPATH)}re.h {$(VPATH)}regex.h
-hash.$(OBJEXT): {$(VPATH)}hash.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}st.h {$(VPATH)}util.h {$(VPATH)}rubysig.h
-inits.$(OBJEXT): {$(VPATH)}inits.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-io.$(OBJEXT): {$(VPATH)}io.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}rubyio.h {$(VPATH)}rubysig.h  {$(VPATH)}util.h \
-  {$(VPATH)}env.h
-main.$(OBJEXT): {$(VPATH)}main.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-marshal.$(OBJEXT): {$(VPATH)}marshal.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}rubyio.h {$(VPATH)}st.h {$(VPATH)}util.h
-math.$(OBJEXT): {$(VPATH)}math.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-numeric.$(OBJEXT): {$(VPATH)}numeric.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}env.h {$(VPATH)}defines.h {$(VPATH)}intern.h \
-  {$(VPATH)}missing.h
-object.$(OBJEXT): {$(VPATH)}object.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}st.h {$(VPATH)}util.h
-pack.$(OBJEXT): {$(VPATH)}pack.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-parse.$(OBJEXT): {$(VPATH)}parse.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}env.h {$(VPATH)}node.h {$(VPATH)}st.h \
-  {$(VPATH)}regex.h {$(VPATH)}util.h {$(VPATH)}lex.c
-prec.$(OBJEXT): {$(VPATH)}prec.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-process.$(OBJEXT): {$(VPATH)}process.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}rubysig.h {$(VPATH)}st.h
-random.$(OBJEXT): {$(VPATH)}random.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-range.$(OBJEXT): {$(VPATH)}range.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-re.$(OBJEXT): {$(VPATH)}re.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}re.h {$(VPATH)}regex.h
-regex.$(OBJEXT): {$(VPATH)}regex.c config.h {$(VPATH)}regex.h
-ruby.$(OBJEXT): {$(VPATH)}ruby.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}dln.h {$(VPATH)}node.h {$(VPATH)}util.h
-signal.$(OBJEXT): {$(VPATH)}signal.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}rubysig.h
-sprintf.$(OBJEXT): {$(VPATH)}sprintf.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-st.$(OBJEXT): {$(VPATH)}st.c config.h {$(VPATH)}st.h
-string.$(OBJEXT): {$(VPATH)}string.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}re.h {$(VPATH)}regex.h
-struct.$(OBJEXT): {$(VPATH)}struct.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-time.$(OBJEXT): {$(VPATH)}time.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h
-util.$(OBJEXT): {$(VPATH)}util.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}util.h
-variable.$(OBJEXT): {$(VPATH)}variable.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}env.h {$(VPATH)}node.h {$(VPATH)}st.h {$(VPATH)}util.h
-version.$(OBJEXT): {$(VPATH)}version.c {$(VPATH)}ruby.h config.h \
-  {$(VPATH)}defines.h {$(VPATH)}intern.h {$(VPATH)}missing.h \
-  {$(VPATH)}version.h
-
-dist: $(PROGRAM)
-	$(RUNRUBY) $(srcdir)/distruby.rb
+enc/ascii.$(OBJEXT): {$(VPATH)}config.h
+enc/ascii.$(OBJEXT): {$(VPATH)}defines.h
+enc/ascii.$(OBJEXT): {$(VPATH)}enc/ascii.c
+enc/ascii.$(OBJEXT): {$(VPATH)}encindex.h
+enc/ascii.$(OBJEXT): {$(VPATH)}missing.h
+enc/ascii.$(OBJEXT): {$(VPATH)}oniguruma.h
+enc/ascii.$(OBJEXT): {$(VPATH)}regenc.h
+enc/trans/newline.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+enc/trans/newline.$(OBJEXT): {$(VPATH)}config.h
+enc/trans/newline.$(OBJEXT): {$(VPATH)}defines.h
+enc/trans/newline.$(OBJEXT): {$(VPATH)}enc/trans/newline.c
+enc/trans/newline.$(OBJEXT): {$(VPATH)}intern.h
+enc/trans/newline.$(OBJEXT): {$(VPATH)}missing.h
+enc/trans/newline.$(OBJEXT): {$(VPATH)}st.h
+enc/trans/newline.$(OBJEXT): {$(VPATH)}subst.h
+enc/trans/newline.$(OBJEXT): {$(VPATH)}transcode_data.h
+enc/unicode.$(OBJEXT): $(UNICODE_HDR_DIR)/casefold.h
+enc/unicode.$(OBJEXT): $(UNICODE_HDR_DIR)/name2ctype.h
+enc/unicode.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+enc/unicode.$(OBJEXT): {$(VPATH)}config.h
+enc/unicode.$(OBJEXT): {$(VPATH)}defines.h
+enc/unicode.$(OBJEXT): {$(VPATH)}enc/unicode.c
+enc/unicode.$(OBJEXT): {$(VPATH)}intern.h
+enc/unicode.$(OBJEXT): {$(VPATH)}missing.h
+enc/unicode.$(OBJEXT): {$(VPATH)}onigmo.h
+enc/unicode.$(OBJEXT): {$(VPATH)}oniguruma.h
+enc/unicode.$(OBJEXT): {$(VPATH)}regenc.h
+enc/unicode.$(OBJEXT): {$(VPATH)}regint.h
+enc/unicode.$(OBJEXT): {$(VPATH)}st.h
+enc/unicode.$(OBJEXT): {$(VPATH)}subst.h
+enc/us_ascii.$(OBJEXT): {$(VPATH)}config.h
+enc/us_ascii.$(OBJEXT): {$(VPATH)}defines.h
+enc/us_ascii.$(OBJEXT): {$(VPATH)}enc/us_ascii.c
+enc/us_ascii.$(OBJEXT): {$(VPATH)}encindex.h
+enc/us_ascii.$(OBJEXT): {$(VPATH)}missing.h
+enc/us_ascii.$(OBJEXT): {$(VPATH)}oniguruma.h
+enc/us_ascii.$(OBJEXT): {$(VPATH)}regenc.h
+enc/utf_8.$(OBJEXT): {$(VPATH)}config.h
+enc/utf_8.$(OBJEXT): {$(VPATH)}defines.h
+enc/utf_8.$(OBJEXT): {$(VPATH)}enc/utf_8.c
+enc/utf_8.$(OBJEXT): {$(VPATH)}encindex.h
+enc/utf_8.$(OBJEXT): {$(VPATH)}missing.h
+enc/utf_8.$(OBJEXT): {$(VPATH)}oniguruma.h
+enc/utf_8.$(OBJEXT): {$(VPATH)}regenc.h
+encoding.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+encoding.$(OBJEXT): $(top_srcdir)/include/ruby.h
+encoding.$(OBJEXT): {$(VPATH)}config.h
+encoding.$(OBJEXT): {$(VPATH)}defines.h
+encoding.$(OBJEXT): {$(VPATH)}encindex.h
+encoding.$(OBJEXT): {$(VPATH)}encoding.c
+encoding.$(OBJEXT): {$(VPATH)}encoding.h
+encoding.$(OBJEXT): {$(VPATH)}intern.h
+encoding.$(OBJEXT): {$(VPATH)}internal.h
+encoding.$(OBJEXT): {$(VPATH)}io.h
+encoding.$(OBJEXT): {$(VPATH)}missing.h
+encoding.$(OBJEXT): {$(VPATH)}onigmo.h
+encoding.$(OBJEXT): {$(VPATH)}oniguruma.h
+encoding.$(OBJEXT): {$(VPATH)}regenc.h
+encoding.$(OBJEXT): {$(VPATH)}ruby_assert.h
+encoding.$(OBJEXT): {$(VPATH)}st.h
+encoding.$(OBJEXT): {$(VPATH)}subst.h
+encoding.$(OBJEXT): {$(VPATH)}util.h
+enum.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+enum.$(OBJEXT): $(top_srcdir)/include/ruby.h
+enum.$(OBJEXT): {$(VPATH)}config.h
+enum.$(OBJEXT): {$(VPATH)}defines.h
+enum.$(OBJEXT): {$(VPATH)}encoding.h
+enum.$(OBJEXT): {$(VPATH)}enum.c
+enum.$(OBJEXT): {$(VPATH)}id.h
+enum.$(OBJEXT): {$(VPATH)}intern.h
+enum.$(OBJEXT): {$(VPATH)}internal.h
+enum.$(OBJEXT): {$(VPATH)}io.h
+enum.$(OBJEXT): {$(VPATH)}missing.h
+enum.$(OBJEXT): {$(VPATH)}onigmo.h
+enum.$(OBJEXT): {$(VPATH)}oniguruma.h
+enum.$(OBJEXT): {$(VPATH)}st.h
+enum.$(OBJEXT): {$(VPATH)}subst.h
+enum.$(OBJEXT): {$(VPATH)}symbol.h
+enum.$(OBJEXT): {$(VPATH)}transient_heap.h
+enum.$(OBJEXT): {$(VPATH)}util.h
+enumerator.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+enumerator.$(OBJEXT): $(top_srcdir)/include/ruby.h
+enumerator.$(OBJEXT): {$(VPATH)}config.h
+enumerator.$(OBJEXT): {$(VPATH)}defines.h
+enumerator.$(OBJEXT): {$(VPATH)}encoding.h
+enumerator.$(OBJEXT): {$(VPATH)}enumerator.c
+enumerator.$(OBJEXT): {$(VPATH)}id.h
+enumerator.$(OBJEXT): {$(VPATH)}intern.h
+enumerator.$(OBJEXT): {$(VPATH)}internal.h
+enumerator.$(OBJEXT): {$(VPATH)}io.h
+enumerator.$(OBJEXT): {$(VPATH)}missing.h
+enumerator.$(OBJEXT): {$(VPATH)}onigmo.h
+enumerator.$(OBJEXT): {$(VPATH)}oniguruma.h
+enumerator.$(OBJEXT): {$(VPATH)}st.h
+enumerator.$(OBJEXT): {$(VPATH)}subst.h
+error.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+error.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+error.$(OBJEXT): $(CCAN_DIR)/list/list.h
+error.$(OBJEXT): $(CCAN_DIR)/str/str.h
+error.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+error.$(OBJEXT): $(top_srcdir)/include/ruby.h
+error.$(OBJEXT): {$(VPATH)}config.h
+error.$(OBJEXT): {$(VPATH)}defines.h
+error.$(OBJEXT): {$(VPATH)}encoding.h
+error.$(OBJEXT): {$(VPATH)}error.c
+error.$(OBJEXT): {$(VPATH)}id.h
+error.$(OBJEXT): {$(VPATH)}intern.h
+error.$(OBJEXT): {$(VPATH)}internal.h
+error.$(OBJEXT): {$(VPATH)}io.h
+error.$(OBJEXT): {$(VPATH)}known_errors.inc
+error.$(OBJEXT): {$(VPATH)}method.h
+error.$(OBJEXT): {$(VPATH)}missing.h
+error.$(OBJEXT): {$(VPATH)}node.h
+error.$(OBJEXT): {$(VPATH)}onigmo.h
+error.$(OBJEXT): {$(VPATH)}oniguruma.h
+error.$(OBJEXT): {$(VPATH)}ruby_assert.h
+error.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+error.$(OBJEXT): {$(VPATH)}st.h
+error.$(OBJEXT): {$(VPATH)}subst.h
+error.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+error.$(OBJEXT): {$(VPATH)}thread_native.h
+error.$(OBJEXT): {$(VPATH)}vm_core.h
+error.$(OBJEXT): {$(VPATH)}vm_debug.h
+error.$(OBJEXT): {$(VPATH)}vm_opts.h
+eval.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+eval.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+eval.$(OBJEXT): $(CCAN_DIR)/list/list.h
+eval.$(OBJEXT): $(CCAN_DIR)/str/str.h
+eval.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+eval.$(OBJEXT): $(top_srcdir)/include/ruby.h
+eval.$(OBJEXT): {$(VPATH)}config.h
+eval.$(OBJEXT): {$(VPATH)}defines.h
+eval.$(OBJEXT): {$(VPATH)}encoding.h
+eval.$(OBJEXT): {$(VPATH)}eval.c
+eval.$(OBJEXT): {$(VPATH)}eval_error.c
+eval.$(OBJEXT): {$(VPATH)}eval_intern.h
+eval.$(OBJEXT): {$(VPATH)}eval_jump.c
+eval.$(OBJEXT): {$(VPATH)}gc.h
+eval.$(OBJEXT): {$(VPATH)}id.h
+eval.$(OBJEXT): {$(VPATH)}intern.h
+eval.$(OBJEXT): {$(VPATH)}internal.h
+eval.$(OBJEXT): {$(VPATH)}io.h
+eval.$(OBJEXT): {$(VPATH)}iseq.h
+eval.$(OBJEXT): {$(VPATH)}method.h
+eval.$(OBJEXT): {$(VPATH)}missing.h
+eval.$(OBJEXT): {$(VPATH)}mjit.h
+eval.$(OBJEXT): {$(VPATH)}node.h
+eval.$(OBJEXT): {$(VPATH)}onigmo.h
+eval.$(OBJEXT): {$(VPATH)}oniguruma.h
+eval.$(OBJEXT): {$(VPATH)}probes.dmyh
+eval.$(OBJEXT): {$(VPATH)}probes.h
+eval.$(OBJEXT): {$(VPATH)}probes_helper.h
+eval.$(OBJEXT): {$(VPATH)}ruby_assert.h
+eval.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+eval.$(OBJEXT): {$(VPATH)}st.h
+eval.$(OBJEXT): {$(VPATH)}subst.h
+eval.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+eval.$(OBJEXT): {$(VPATH)}thread_native.h
+eval.$(OBJEXT): {$(VPATH)}vm.h
+eval.$(OBJEXT): {$(VPATH)}vm_core.h
+eval.$(OBJEXT): {$(VPATH)}vm_debug.h
+eval.$(OBJEXT): {$(VPATH)}vm_opts.h
+explicit_bzero.$(OBJEXT): {$(VPATH)}config.h
+explicit_bzero.$(OBJEXT): {$(VPATH)}explicit_bzero.c
+explicit_bzero.$(OBJEXT): {$(VPATH)}missing.h
+file.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+file.$(OBJEXT): $(top_srcdir)/include/ruby.h
+file.$(OBJEXT): {$(VPATH)}config.h
+file.$(OBJEXT): {$(VPATH)}defines.h
+file.$(OBJEXT): {$(VPATH)}dln.h
+file.$(OBJEXT): {$(VPATH)}encindex.h
+file.$(OBJEXT): {$(VPATH)}encoding.h
+file.$(OBJEXT): {$(VPATH)}file.c
+file.$(OBJEXT): {$(VPATH)}id.h
+file.$(OBJEXT): {$(VPATH)}intern.h
+file.$(OBJEXT): {$(VPATH)}internal.h
+file.$(OBJEXT): {$(VPATH)}io.h
+file.$(OBJEXT): {$(VPATH)}missing.h
+file.$(OBJEXT): {$(VPATH)}onigmo.h
+file.$(OBJEXT): {$(VPATH)}oniguruma.h
+file.$(OBJEXT): {$(VPATH)}st.h
+file.$(OBJEXT): {$(VPATH)}subst.h
+file.$(OBJEXT): {$(VPATH)}thread.h
+file.$(OBJEXT): {$(VPATH)}util.h
+gc.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+gc.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+gc.$(OBJEXT): $(CCAN_DIR)/list/list.h
+gc.$(OBJEXT): $(CCAN_DIR)/str/str.h
+gc.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+gc.$(OBJEXT): $(top_srcdir)/include/ruby.h
+gc.$(OBJEXT): {$(VPATH)}config.h
+gc.$(OBJEXT): {$(VPATH)}constant.h
+gc.$(OBJEXT): {$(VPATH)}debug.h
+gc.$(OBJEXT): {$(VPATH)}debug_counter.h
+gc.$(OBJEXT): {$(VPATH)}defines.h
+gc.$(OBJEXT): {$(VPATH)}encoding.h
+gc.$(OBJEXT): {$(VPATH)}eval_intern.h
+gc.$(OBJEXT): {$(VPATH)}gc.c
+gc.$(OBJEXT): {$(VPATH)}gc.h
+gc.$(OBJEXT): {$(VPATH)}id.h
+gc.$(OBJEXT): {$(VPATH)}id_table.h
+gc.$(OBJEXT): {$(VPATH)}intern.h
+gc.$(OBJEXT): {$(VPATH)}internal.h
+gc.$(OBJEXT): {$(VPATH)}io.h
+gc.$(OBJEXT): {$(VPATH)}method.h
+gc.$(OBJEXT): {$(VPATH)}missing.h
+gc.$(OBJEXT): {$(VPATH)}mjit.h
+gc.$(OBJEXT): {$(VPATH)}node.h
+gc.$(OBJEXT): {$(VPATH)}onigmo.h
+gc.$(OBJEXT): {$(VPATH)}oniguruma.h
+gc.$(OBJEXT): {$(VPATH)}probes.dmyh
+gc.$(OBJEXT): {$(VPATH)}probes.h
+gc.$(OBJEXT): {$(VPATH)}re.h
+gc.$(OBJEXT): {$(VPATH)}regenc.h
+gc.$(OBJEXT): {$(VPATH)}regex.h
+gc.$(OBJEXT): {$(VPATH)}regint.h
+gc.$(OBJEXT): {$(VPATH)}ruby_assert.h
+gc.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+gc.$(OBJEXT): {$(VPATH)}st.h
+gc.$(OBJEXT): {$(VPATH)}subst.h
+gc.$(OBJEXT): {$(VPATH)}thread.h
+gc.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+gc.$(OBJEXT): {$(VPATH)}thread_native.h
+gc.$(OBJEXT): {$(VPATH)}transient_heap.h
+gc.$(OBJEXT): {$(VPATH)}util.h
+gc.$(OBJEXT): {$(VPATH)}vm_core.h
+gc.$(OBJEXT): {$(VPATH)}vm_debug.h
+gc.$(OBJEXT): {$(VPATH)}vm_opts.h
+golf_prelude.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+golf_prelude.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+golf_prelude.$(OBJEXT): $(CCAN_DIR)/list/list.h
+golf_prelude.$(OBJEXT): $(CCAN_DIR)/str/str.h
+golf_prelude.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+golf_prelude.$(OBJEXT): $(top_srcdir)/include/ruby.h
+golf_prelude.$(OBJEXT): {$(VPATH)}config.h
+golf_prelude.$(OBJEXT): {$(VPATH)}defines.h
+golf_prelude.$(OBJEXT): {$(VPATH)}encoding.h
+golf_prelude.$(OBJEXT): {$(VPATH)}golf_prelude.c
+golf_prelude.$(OBJEXT): {$(VPATH)}golf_prelude.rb
+golf_prelude.$(OBJEXT): {$(VPATH)}id.h
+golf_prelude.$(OBJEXT): {$(VPATH)}intern.h
+golf_prelude.$(OBJEXT): {$(VPATH)}internal.h
+golf_prelude.$(OBJEXT): {$(VPATH)}io.h
+golf_prelude.$(OBJEXT): {$(VPATH)}iseq.h
+golf_prelude.$(OBJEXT): {$(VPATH)}method.h
+golf_prelude.$(OBJEXT): {$(VPATH)}missing.h
+golf_prelude.$(OBJEXT): {$(VPATH)}node.h
+golf_prelude.$(OBJEXT): {$(VPATH)}onigmo.h
+golf_prelude.$(OBJEXT): {$(VPATH)}oniguruma.h
+golf_prelude.$(OBJEXT): {$(VPATH)}ruby_assert.h
+golf_prelude.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+golf_prelude.$(OBJEXT): {$(VPATH)}st.h
+golf_prelude.$(OBJEXT): {$(VPATH)}subst.h
+golf_prelude.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+golf_prelude.$(OBJEXT): {$(VPATH)}thread_native.h
+golf_prelude.$(OBJEXT): {$(VPATH)}vm_core.h
+golf_prelude.$(OBJEXT): {$(VPATH)}vm_debug.h
+golf_prelude.$(OBJEXT): {$(VPATH)}vm_opts.h
+goruby.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+goruby.$(OBJEXT): $(top_srcdir)/include/ruby.h
+goruby.$(OBJEXT): {$(VPATH)}backward.h
+goruby.$(OBJEXT): {$(VPATH)}config.h
+goruby.$(OBJEXT): {$(VPATH)}defines.h
+goruby.$(OBJEXT): {$(VPATH)}goruby.c
+goruby.$(OBJEXT): {$(VPATH)}intern.h
+goruby.$(OBJEXT): {$(VPATH)}main.c
+goruby.$(OBJEXT): {$(VPATH)}missing.h
+goruby.$(OBJEXT): {$(VPATH)}node.h
+goruby.$(OBJEXT): {$(VPATH)}st.h
+goruby.$(OBJEXT): {$(VPATH)}subst.h
+goruby.$(OBJEXT): {$(VPATH)}vm_debug.h
+hash.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+hash.$(OBJEXT): $(top_srcdir)/include/ruby.h
+hash.$(OBJEXT): {$(VPATH)}config.h
+hash.$(OBJEXT): {$(VPATH)}debug_counter.h
+hash.$(OBJEXT): {$(VPATH)}defines.h
+hash.$(OBJEXT): {$(VPATH)}encoding.h
+hash.$(OBJEXT): {$(VPATH)}gc.h
+hash.$(OBJEXT): {$(VPATH)}hash.c
+hash.$(OBJEXT): {$(VPATH)}id.h
+hash.$(OBJEXT): {$(VPATH)}intern.h
+hash.$(OBJEXT): {$(VPATH)}internal.h
+hash.$(OBJEXT): {$(VPATH)}io.h
+hash.$(OBJEXT): {$(VPATH)}missing.h
+hash.$(OBJEXT): {$(VPATH)}onigmo.h
+hash.$(OBJEXT): {$(VPATH)}oniguruma.h
+hash.$(OBJEXT): {$(VPATH)}probes.dmyh
+hash.$(OBJEXT): {$(VPATH)}probes.h
+hash.$(OBJEXT): {$(VPATH)}ruby_assert.h
+hash.$(OBJEXT): {$(VPATH)}st.h
+hash.$(OBJEXT): {$(VPATH)}subst.h
+hash.$(OBJEXT): {$(VPATH)}symbol.h
+hash.$(OBJEXT): {$(VPATH)}transient_heap.h
+hash.$(OBJEXT): {$(VPATH)}util.h
+inits.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+inits.$(OBJEXT): $(top_srcdir)/include/ruby.h
+inits.$(OBJEXT): {$(VPATH)}config.h
+inits.$(OBJEXT): {$(VPATH)}defines.h
+inits.$(OBJEXT): {$(VPATH)}encoding.h
+inits.$(OBJEXT): {$(VPATH)}inits.c
+inits.$(OBJEXT): {$(VPATH)}intern.h
+inits.$(OBJEXT): {$(VPATH)}internal.h
+inits.$(OBJEXT): {$(VPATH)}io.h
+inits.$(OBJEXT): {$(VPATH)}missing.h
+inits.$(OBJEXT): {$(VPATH)}onigmo.h
+inits.$(OBJEXT): {$(VPATH)}oniguruma.h
+inits.$(OBJEXT): {$(VPATH)}st.h
+inits.$(OBJEXT): {$(VPATH)}subst.h
+io.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+io.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+io.$(OBJEXT): $(CCAN_DIR)/list/list.h
+io.$(OBJEXT): $(CCAN_DIR)/str/str.h
+io.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+io.$(OBJEXT): $(top_srcdir)/include/ruby.h
+io.$(OBJEXT): {$(VPATH)}config.h
+io.$(OBJEXT): {$(VPATH)}defines.h
+io.$(OBJEXT): {$(VPATH)}dln.h
+io.$(OBJEXT): {$(VPATH)}encindex.h
+io.$(OBJEXT): {$(VPATH)}encoding.h
+io.$(OBJEXT): {$(VPATH)}id.h
+io.$(OBJEXT): {$(VPATH)}intern.h
+io.$(OBJEXT): {$(VPATH)}internal.h
+io.$(OBJEXT): {$(VPATH)}io.c
+io.$(OBJEXT): {$(VPATH)}io.h
+io.$(OBJEXT): {$(VPATH)}method.h
+io.$(OBJEXT): {$(VPATH)}missing.h
+io.$(OBJEXT): {$(VPATH)}node.h
+io.$(OBJEXT): {$(VPATH)}onigmo.h
+io.$(OBJEXT): {$(VPATH)}oniguruma.h
+io.$(OBJEXT): {$(VPATH)}ruby_assert.h
+io.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+io.$(OBJEXT): {$(VPATH)}st.h
+io.$(OBJEXT): {$(VPATH)}subst.h
+io.$(OBJEXT): {$(VPATH)}thread.h
+io.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+io.$(OBJEXT): {$(VPATH)}thread_native.h
+io.$(OBJEXT): {$(VPATH)}util.h
+io.$(OBJEXT): {$(VPATH)}vm_core.h
+io.$(OBJEXT): {$(VPATH)}vm_opts.h
+iseq.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+iseq.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+iseq.$(OBJEXT): $(CCAN_DIR)/list/list.h
+iseq.$(OBJEXT): $(CCAN_DIR)/str/str.h
+iseq.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+iseq.$(OBJEXT): $(top_srcdir)/include/ruby.h
+iseq.$(OBJEXT): {$(VPATH)}config.h
+iseq.$(OBJEXT): {$(VPATH)}defines.h
+iseq.$(OBJEXT): {$(VPATH)}encoding.h
+iseq.$(OBJEXT): {$(VPATH)}eval_intern.h
+iseq.$(OBJEXT): {$(VPATH)}gc.h
+iseq.$(OBJEXT): {$(VPATH)}id.h
+iseq.$(OBJEXT): {$(VPATH)}id_table.h
+iseq.$(OBJEXT): {$(VPATH)}insns.def
+iseq.$(OBJEXT): {$(VPATH)}insns.inc
+iseq.$(OBJEXT): {$(VPATH)}insns_info.inc
+iseq.$(OBJEXT): {$(VPATH)}intern.h
+iseq.$(OBJEXT): {$(VPATH)}internal.h
+iseq.$(OBJEXT): {$(VPATH)}io.h
+iseq.$(OBJEXT): {$(VPATH)}iseq.c
+iseq.$(OBJEXT): {$(VPATH)}iseq.h
+iseq.$(OBJEXT): {$(VPATH)}method.h
+iseq.$(OBJEXT): {$(VPATH)}missing.h
+iseq.$(OBJEXT): {$(VPATH)}mjit.h
+iseq.$(OBJEXT): {$(VPATH)}node.h
+iseq.$(OBJEXT): {$(VPATH)}node_name.inc
+iseq.$(OBJEXT): {$(VPATH)}onigmo.h
+iseq.$(OBJEXT): {$(VPATH)}oniguruma.h
+iseq.$(OBJEXT): {$(VPATH)}ruby_assert.h
+iseq.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+iseq.$(OBJEXT): {$(VPATH)}st.h
+iseq.$(OBJEXT): {$(VPATH)}subst.h
+iseq.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+iseq.$(OBJEXT): {$(VPATH)}thread_native.h
+iseq.$(OBJEXT): {$(VPATH)}util.h
+iseq.$(OBJEXT): {$(VPATH)}vm_core.h
+iseq.$(OBJEXT): {$(VPATH)}vm_debug.h
+iseq.$(OBJEXT): {$(VPATH)}vm_opts.h
+load.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+load.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+load.$(OBJEXT): $(CCAN_DIR)/list/list.h
+load.$(OBJEXT): $(CCAN_DIR)/str/str.h
+load.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+load.$(OBJEXT): $(top_srcdir)/include/ruby.h
+load.$(OBJEXT): {$(VPATH)}config.h
+load.$(OBJEXT): {$(VPATH)}defines.h
+load.$(OBJEXT): {$(VPATH)}dln.h
+load.$(OBJEXT): {$(VPATH)}encoding.h
+load.$(OBJEXT): {$(VPATH)}eval_intern.h
+load.$(OBJEXT): {$(VPATH)}id.h
+load.$(OBJEXT): {$(VPATH)}intern.h
+load.$(OBJEXT): {$(VPATH)}internal.h
+load.$(OBJEXT): {$(VPATH)}io.h
+load.$(OBJEXT): {$(VPATH)}load.c
+load.$(OBJEXT): {$(VPATH)}method.h
+load.$(OBJEXT): {$(VPATH)}missing.h
+load.$(OBJEXT): {$(VPATH)}node.h
+load.$(OBJEXT): {$(VPATH)}onigmo.h
+load.$(OBJEXT): {$(VPATH)}oniguruma.h
+load.$(OBJEXT): {$(VPATH)}probes.dmyh
+load.$(OBJEXT): {$(VPATH)}probes.h
+load.$(OBJEXT): {$(VPATH)}ruby_assert.h
+load.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+load.$(OBJEXT): {$(VPATH)}st.h
+load.$(OBJEXT): {$(VPATH)}subst.h
+load.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+load.$(OBJEXT): {$(VPATH)}thread_native.h
+load.$(OBJEXT): {$(VPATH)}util.h
+load.$(OBJEXT): {$(VPATH)}vm_core.h
+load.$(OBJEXT): {$(VPATH)}vm_debug.h
+load.$(OBJEXT): {$(VPATH)}vm_opts.h
+loadpath.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+loadpath.$(OBJEXT): $(hdrdir)/ruby/version.h
+loadpath.$(OBJEXT): $(top_srcdir)/version.h
+loadpath.$(OBJEXT): {$(VPATH)}config.h
+loadpath.$(OBJEXT): {$(VPATH)}defines.h
+loadpath.$(OBJEXT): {$(VPATH)}intern.h
+loadpath.$(OBJEXT): {$(VPATH)}loadpath.c
+loadpath.$(OBJEXT): {$(VPATH)}missing.h
+loadpath.$(OBJEXT): {$(VPATH)}st.h
+loadpath.$(OBJEXT): {$(VPATH)}subst.h
+loadpath.$(OBJEXT): {$(VPATH)}verconf.h
+localeinit.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+localeinit.$(OBJEXT): $(top_srcdir)/include/ruby.h
+localeinit.$(OBJEXT): {$(VPATH)}config.h
+localeinit.$(OBJEXT): {$(VPATH)}defines.h
+localeinit.$(OBJEXT): {$(VPATH)}encindex.h
+localeinit.$(OBJEXT): {$(VPATH)}encoding.h
+localeinit.$(OBJEXT): {$(VPATH)}intern.h
+localeinit.$(OBJEXT): {$(VPATH)}internal.h
+localeinit.$(OBJEXT): {$(VPATH)}io.h
+localeinit.$(OBJEXT): {$(VPATH)}localeinit.c
+localeinit.$(OBJEXT): {$(VPATH)}missing.h
+localeinit.$(OBJEXT): {$(VPATH)}onigmo.h
+localeinit.$(OBJEXT): {$(VPATH)}oniguruma.h
+localeinit.$(OBJEXT): {$(VPATH)}st.h
+localeinit.$(OBJEXT): {$(VPATH)}subst.h
+main.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+main.$(OBJEXT): $(top_srcdir)/include/ruby.h
+main.$(OBJEXT): {$(VPATH)}backward.h
+main.$(OBJEXT): {$(VPATH)}config.h
+main.$(OBJEXT): {$(VPATH)}defines.h
+main.$(OBJEXT): {$(VPATH)}intern.h
+main.$(OBJEXT): {$(VPATH)}main.c
+main.$(OBJEXT): {$(VPATH)}missing.h
+main.$(OBJEXT): {$(VPATH)}node.h
+main.$(OBJEXT): {$(VPATH)}st.h
+main.$(OBJEXT): {$(VPATH)}subst.h
+main.$(OBJEXT): {$(VPATH)}vm_debug.h
+marshal.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+marshal.$(OBJEXT): $(top_srcdir)/include/ruby.h
+marshal.$(OBJEXT): {$(VPATH)}config.h
+marshal.$(OBJEXT): {$(VPATH)}defines.h
+marshal.$(OBJEXT): {$(VPATH)}encindex.h
+marshal.$(OBJEXT): {$(VPATH)}encoding.h
+marshal.$(OBJEXT): {$(VPATH)}id_table.h
+marshal.$(OBJEXT): {$(VPATH)}intern.h
+marshal.$(OBJEXT): {$(VPATH)}internal.h
+marshal.$(OBJEXT): {$(VPATH)}io.h
+marshal.$(OBJEXT): {$(VPATH)}marshal.c
+marshal.$(OBJEXT): {$(VPATH)}missing.h
+marshal.$(OBJEXT): {$(VPATH)}onigmo.h
+marshal.$(OBJEXT): {$(VPATH)}oniguruma.h
+marshal.$(OBJEXT): {$(VPATH)}st.h
+marshal.$(OBJEXT): {$(VPATH)}subst.h
+marshal.$(OBJEXT): {$(VPATH)}util.h
+math.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+math.$(OBJEXT): $(top_srcdir)/include/ruby.h
+math.$(OBJEXT): {$(VPATH)}config.h
+math.$(OBJEXT): {$(VPATH)}defines.h
+math.$(OBJEXT): {$(VPATH)}encoding.h
+math.$(OBJEXT): {$(VPATH)}intern.h
+math.$(OBJEXT): {$(VPATH)}internal.h
+math.$(OBJEXT): {$(VPATH)}io.h
+math.$(OBJEXT): {$(VPATH)}math.c
+math.$(OBJEXT): {$(VPATH)}missing.h
+math.$(OBJEXT): {$(VPATH)}onigmo.h
+math.$(OBJEXT): {$(VPATH)}oniguruma.h
+math.$(OBJEXT): {$(VPATH)}st.h
+math.$(OBJEXT): {$(VPATH)}subst.h
+miniinit.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+miniinit.$(OBJEXT): {$(VPATH)}config.h
+miniinit.$(OBJEXT): {$(VPATH)}defines.h
+miniinit.$(OBJEXT): {$(VPATH)}encoding.h
+miniinit.$(OBJEXT): {$(VPATH)}intern.h
+miniinit.$(OBJEXT): {$(VPATH)}miniinit.c
+miniinit.$(OBJEXT): {$(VPATH)}missing.h
+miniinit.$(OBJEXT): {$(VPATH)}onigmo.h
+miniinit.$(OBJEXT): {$(VPATH)}oniguruma.h
+miniinit.$(OBJEXT): {$(VPATH)}st.h
+miniinit.$(OBJEXT): {$(VPATH)}subst.h
+miniprelude.$(OBJEXT): {$(VPATH)}iseq.h
+miniprelude.$(OBJEXT): {$(VPATH)}miniprelude.c
+mjit.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+mjit.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+mjit.$(OBJEXT): $(CCAN_DIR)/list/list.h
+mjit.$(OBJEXT): $(CCAN_DIR)/str/str.h
+mjit.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+mjit.$(OBJEXT): $(top_srcdir)/include/ruby.h
+mjit.$(OBJEXT): {$(VPATH)}config.h
+mjit.$(OBJEXT): {$(VPATH)}constant.h
+mjit.$(OBJEXT): {$(VPATH)}debug.h
+mjit.$(OBJEXT): {$(VPATH)}defines.h
+mjit.$(OBJEXT): {$(VPATH)}dln.h
+mjit.$(OBJEXT): {$(VPATH)}gc.h
+mjit.$(OBJEXT): {$(VPATH)}id.h
+mjit.$(OBJEXT): {$(VPATH)}id_table.h
+mjit.$(OBJEXT): {$(VPATH)}intern.h
+mjit.$(OBJEXT): {$(VPATH)}internal.h
+mjit.$(OBJEXT): {$(VPATH)}method.h
+mjit.$(OBJEXT): {$(VPATH)}missing.h
+mjit.$(OBJEXT): {$(VPATH)}mjit.c
+mjit.$(OBJEXT): {$(VPATH)}mjit.h
+mjit.$(OBJEXT): {$(VPATH)}mjit_config.h
+mjit.$(OBJEXT): {$(VPATH)}mjit_worker.c
+mjit.$(OBJEXT): {$(VPATH)}node.h
+mjit.$(OBJEXT): {$(VPATH)}ruby_assert.h
+mjit.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+mjit.$(OBJEXT): {$(VPATH)}st.h
+mjit.$(OBJEXT): {$(VPATH)}subst.h
+mjit.$(OBJEXT): {$(VPATH)}thread.h
+mjit.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+mjit.$(OBJEXT): {$(VPATH)}thread_native.h
+mjit.$(OBJEXT): {$(VPATH)}util.h
+mjit.$(OBJEXT): {$(VPATH)}vm_core.h
+mjit.$(OBJEXT): {$(VPATH)}vm_opts.h
+mjit_compile.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+mjit_compile.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+mjit_compile.$(OBJEXT): $(CCAN_DIR)/list/list.h
+mjit_compile.$(OBJEXT): $(CCAN_DIR)/str/str.h
+mjit_compile.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+mjit_compile.$(OBJEXT): $(top_srcdir)/include/ruby.h
+mjit_compile.$(OBJEXT): {$(VPATH)}config.h
+mjit_compile.$(OBJEXT): {$(VPATH)}defines.h
+mjit_compile.$(OBJEXT): {$(VPATH)}id.h
+mjit_compile.$(OBJEXT): {$(VPATH)}insns.def
+mjit_compile.$(OBJEXT): {$(VPATH)}insns.inc
+mjit_compile.$(OBJEXT): {$(VPATH)}insns_info.inc
+mjit_compile.$(OBJEXT): {$(VPATH)}intern.h
+mjit_compile.$(OBJEXT): {$(VPATH)}internal.h
+mjit_compile.$(OBJEXT): {$(VPATH)}iseq.h
+mjit_compile.$(OBJEXT): {$(VPATH)}method.h
+mjit_compile.$(OBJEXT): {$(VPATH)}missing.h
+mjit_compile.$(OBJEXT): {$(VPATH)}mjit.h
+mjit_compile.$(OBJEXT): {$(VPATH)}mjit_compile.c
+mjit_compile.$(OBJEXT): {$(VPATH)}mjit_compile.inc
+mjit_compile.$(OBJEXT): {$(VPATH)}node.h
+mjit_compile.$(OBJEXT): {$(VPATH)}ruby_assert.h
+mjit_compile.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+mjit_compile.$(OBJEXT): {$(VPATH)}st.h
+mjit_compile.$(OBJEXT): {$(VPATH)}subst.h
+mjit_compile.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+mjit_compile.$(OBJEXT): {$(VPATH)}thread_native.h
+mjit_compile.$(OBJEXT): {$(VPATH)}vm_core.h
+mjit_compile.$(OBJEXT): {$(VPATH)}vm_exec.h
+mjit_compile.$(OBJEXT): {$(VPATH)}vm_insnhelper.h
+mjit_compile.$(OBJEXT): {$(VPATH)}vm_opts.h
+node.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+node.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+node.$(OBJEXT): $(CCAN_DIR)/list/list.h
+node.$(OBJEXT): $(CCAN_DIR)/str/str.h
+node.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+node.$(OBJEXT): $(top_srcdir)/include/ruby.h
+node.$(OBJEXT): {$(VPATH)}config.h
+node.$(OBJEXT): {$(VPATH)}defines.h
+node.$(OBJEXT): {$(VPATH)}encoding.h
+node.$(OBJEXT): {$(VPATH)}id.h
+node.$(OBJEXT): {$(VPATH)}intern.h
+node.$(OBJEXT): {$(VPATH)}internal.h
+node.$(OBJEXT): {$(VPATH)}io.h
+node.$(OBJEXT): {$(VPATH)}method.h
+node.$(OBJEXT): {$(VPATH)}missing.h
+node.$(OBJEXT): {$(VPATH)}node.c
+node.$(OBJEXT): {$(VPATH)}node.h
+node.$(OBJEXT): {$(VPATH)}onigmo.h
+node.$(OBJEXT): {$(VPATH)}oniguruma.h
+node.$(OBJEXT): {$(VPATH)}ruby_assert.h
+node.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+node.$(OBJEXT): {$(VPATH)}st.h
+node.$(OBJEXT): {$(VPATH)}subst.h
+node.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+node.$(OBJEXT): {$(VPATH)}thread_native.h
+node.$(OBJEXT): {$(VPATH)}vm_core.h
+node.$(OBJEXT): {$(VPATH)}vm_debug.h
+node.$(OBJEXT): {$(VPATH)}vm_opts.h
+numeric.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+numeric.$(OBJEXT): $(top_srcdir)/include/ruby.h
+numeric.$(OBJEXT): {$(VPATH)}config.h
+numeric.$(OBJEXT): {$(VPATH)}defines.h
+numeric.$(OBJEXT): {$(VPATH)}encoding.h
+numeric.$(OBJEXT): {$(VPATH)}id.h
+numeric.$(OBJEXT): {$(VPATH)}intern.h
+numeric.$(OBJEXT): {$(VPATH)}internal.h
+numeric.$(OBJEXT): {$(VPATH)}io.h
+numeric.$(OBJEXT): {$(VPATH)}missing.h
+numeric.$(OBJEXT): {$(VPATH)}numeric.c
+numeric.$(OBJEXT): {$(VPATH)}onigmo.h
+numeric.$(OBJEXT): {$(VPATH)}oniguruma.h
+numeric.$(OBJEXT): {$(VPATH)}st.h
+numeric.$(OBJEXT): {$(VPATH)}subst.h
+numeric.$(OBJEXT): {$(VPATH)}util.h
+object.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+object.$(OBJEXT): $(top_srcdir)/include/ruby.h
+object.$(OBJEXT): {$(VPATH)}config.h
+object.$(OBJEXT): {$(VPATH)}constant.h
+object.$(OBJEXT): {$(VPATH)}defines.h
+object.$(OBJEXT): {$(VPATH)}encoding.h
+object.$(OBJEXT): {$(VPATH)}id.h
+object.$(OBJEXT): {$(VPATH)}intern.h
+object.$(OBJEXT): {$(VPATH)}internal.h
+object.$(OBJEXT): {$(VPATH)}io.h
+object.$(OBJEXT): {$(VPATH)}missing.h
+object.$(OBJEXT): {$(VPATH)}object.c
+object.$(OBJEXT): {$(VPATH)}onigmo.h
+object.$(OBJEXT): {$(VPATH)}oniguruma.h
+object.$(OBJEXT): {$(VPATH)}probes.dmyh
+object.$(OBJEXT): {$(VPATH)}probes.h
+object.$(OBJEXT): {$(VPATH)}st.h
+object.$(OBJEXT): {$(VPATH)}subst.h
+object.$(OBJEXT): {$(VPATH)}util.h
+pack.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+pack.$(OBJEXT): $(top_srcdir)/include/ruby.h
+pack.$(OBJEXT): {$(VPATH)}config.h
+pack.$(OBJEXT): {$(VPATH)}defines.h
+pack.$(OBJEXT): {$(VPATH)}encoding.h
+pack.$(OBJEXT): {$(VPATH)}intern.h
+pack.$(OBJEXT): {$(VPATH)}internal.h
+pack.$(OBJEXT): {$(VPATH)}io.h
+pack.$(OBJEXT): {$(VPATH)}missing.h
+pack.$(OBJEXT): {$(VPATH)}onigmo.h
+pack.$(OBJEXT): {$(VPATH)}oniguruma.h
+pack.$(OBJEXT): {$(VPATH)}pack.c
+pack.$(OBJEXT): {$(VPATH)}st.h
+pack.$(OBJEXT): {$(VPATH)}subst.h
+parse.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+parse.$(OBJEXT): $(top_srcdir)/include/ruby.h
+parse.$(OBJEXT): {$(VPATH)}config.h
+parse.$(OBJEXT): {$(VPATH)}defines.h
+parse.$(OBJEXT): {$(VPATH)}defs/keywords
+parse.$(OBJEXT): {$(VPATH)}encoding.h
+parse.$(OBJEXT): {$(VPATH)}id.h
+parse.$(OBJEXT): {$(VPATH)}intern.h
+parse.$(OBJEXT): {$(VPATH)}internal.h
+parse.$(OBJEXT): {$(VPATH)}io.h
+parse.$(OBJEXT): {$(VPATH)}lex.c
+parse.$(OBJEXT): {$(VPATH)}missing.h
+parse.$(OBJEXT): {$(VPATH)}node.h
+parse.$(OBJEXT): {$(VPATH)}onigmo.h
+parse.$(OBJEXT): {$(VPATH)}oniguruma.h
+parse.$(OBJEXT): {$(VPATH)}parse.c
+parse.$(OBJEXT): {$(VPATH)}parse.h
+parse.$(OBJEXT): {$(VPATH)}parse.y
+parse.$(OBJEXT): {$(VPATH)}probes.dmyh
+parse.$(OBJEXT): {$(VPATH)}probes.h
+parse.$(OBJEXT): {$(VPATH)}regenc.h
+parse.$(OBJEXT): {$(VPATH)}regex.h
+parse.$(OBJEXT): {$(VPATH)}st.h
+parse.$(OBJEXT): {$(VPATH)}subst.h
+parse.$(OBJEXT): {$(VPATH)}symbol.h
+parse.$(OBJEXT): {$(VPATH)}util.h
+prelude.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+prelude.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+prelude.$(OBJEXT): $(CCAN_DIR)/list/list.h
+prelude.$(OBJEXT): $(CCAN_DIR)/str/str.h
+prelude.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+prelude.$(OBJEXT): $(top_srcdir)/include/ruby.h
+prelude.$(OBJEXT): {$(VPATH)}config.h
+prelude.$(OBJEXT): {$(VPATH)}defines.h
+prelude.$(OBJEXT): {$(VPATH)}encoding.h
+prelude.$(OBJEXT): {$(VPATH)}gem_prelude.rb
+prelude.$(OBJEXT): {$(VPATH)}id.h
+prelude.$(OBJEXT): {$(VPATH)}intern.h
+prelude.$(OBJEXT): {$(VPATH)}internal.h
+prelude.$(OBJEXT): {$(VPATH)}io.h
+prelude.$(OBJEXT): {$(VPATH)}iseq.h
+prelude.$(OBJEXT): {$(VPATH)}method.h
+prelude.$(OBJEXT): {$(VPATH)}missing.h
+prelude.$(OBJEXT): {$(VPATH)}node.h
+prelude.$(OBJEXT): {$(VPATH)}onigmo.h
+prelude.$(OBJEXT): {$(VPATH)}oniguruma.h
+prelude.$(OBJEXT): {$(VPATH)}prelude.c
+prelude.$(OBJEXT): {$(VPATH)}prelude.rb
+prelude.$(OBJEXT): {$(VPATH)}ruby_assert.h
+prelude.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+prelude.$(OBJEXT): {$(VPATH)}st.h
+prelude.$(OBJEXT): {$(VPATH)}subst.h
+prelude.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+prelude.$(OBJEXT): {$(VPATH)}thread_native.h
+prelude.$(OBJEXT): {$(VPATH)}vm_core.h
+prelude.$(OBJEXT): {$(VPATH)}vm_debug.h
+prelude.$(OBJEXT): {$(VPATH)}vm_opts.h
+proc.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+proc.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+proc.$(OBJEXT): $(CCAN_DIR)/list/list.h
+proc.$(OBJEXT): $(CCAN_DIR)/str/str.h
+proc.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+proc.$(OBJEXT): $(top_srcdir)/include/ruby.h
+proc.$(OBJEXT): {$(VPATH)}config.h
+proc.$(OBJEXT): {$(VPATH)}defines.h
+proc.$(OBJEXT): {$(VPATH)}encoding.h
+proc.$(OBJEXT): {$(VPATH)}eval_intern.h
+proc.$(OBJEXT): {$(VPATH)}gc.h
+proc.$(OBJEXT): {$(VPATH)}id.h
+proc.$(OBJEXT): {$(VPATH)}intern.h
+proc.$(OBJEXT): {$(VPATH)}internal.h
+proc.$(OBJEXT): {$(VPATH)}io.h
+proc.$(OBJEXT): {$(VPATH)}iseq.h
+proc.$(OBJEXT): {$(VPATH)}method.h
+proc.$(OBJEXT): {$(VPATH)}missing.h
+proc.$(OBJEXT): {$(VPATH)}node.h
+proc.$(OBJEXT): {$(VPATH)}onigmo.h
+proc.$(OBJEXT): {$(VPATH)}oniguruma.h
+proc.$(OBJEXT): {$(VPATH)}proc.c
+proc.$(OBJEXT): {$(VPATH)}ruby_assert.h
+proc.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+proc.$(OBJEXT): {$(VPATH)}st.h
+proc.$(OBJEXT): {$(VPATH)}subst.h
+proc.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+proc.$(OBJEXT): {$(VPATH)}thread_native.h
+proc.$(OBJEXT): {$(VPATH)}vm_core.h
+proc.$(OBJEXT): {$(VPATH)}vm_debug.h
+proc.$(OBJEXT): {$(VPATH)}vm_opts.h
+process.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+process.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+process.$(OBJEXT): $(CCAN_DIR)/list/list.h
+process.$(OBJEXT): $(CCAN_DIR)/str/str.h
+process.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+process.$(OBJEXT): $(top_srcdir)/include/ruby.h
+process.$(OBJEXT): {$(VPATH)}config.h
+process.$(OBJEXT): {$(VPATH)}defines.h
+process.$(OBJEXT): {$(VPATH)}dln.h
+process.$(OBJEXT): {$(VPATH)}encoding.h
+process.$(OBJEXT): {$(VPATH)}hrtime.h
+process.$(OBJEXT): {$(VPATH)}id.h
+process.$(OBJEXT): {$(VPATH)}intern.h
+process.$(OBJEXT): {$(VPATH)}internal.h
+process.$(OBJEXT): {$(VPATH)}io.h
+process.$(OBJEXT): {$(VPATH)}method.h
+process.$(OBJEXT): {$(VPATH)}missing.h
+process.$(OBJEXT): {$(VPATH)}node.h
+process.$(OBJEXT): {$(VPATH)}onigmo.h
+process.$(OBJEXT): {$(VPATH)}oniguruma.h
+process.$(OBJEXT): {$(VPATH)}process.c
+process.$(OBJEXT): {$(VPATH)}ruby_assert.h
+process.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+process.$(OBJEXT): {$(VPATH)}st.h
+process.$(OBJEXT): {$(VPATH)}subst.h
+process.$(OBJEXT): {$(VPATH)}thread.h
+process.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+process.$(OBJEXT): {$(VPATH)}thread_native.h
+process.$(OBJEXT): {$(VPATH)}util.h
+process.$(OBJEXT): {$(VPATH)}vm_core.h
+process.$(OBJEXT): {$(VPATH)}vm_debug.h
+process.$(OBJEXT): {$(VPATH)}vm_opts.h
+random.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+random.$(OBJEXT): $(top_srcdir)/include/ruby.h
+random.$(OBJEXT): {$(VPATH)}config.h
+random.$(OBJEXT): {$(VPATH)}defines.h
+random.$(OBJEXT): {$(VPATH)}encoding.h
+random.$(OBJEXT): {$(VPATH)}intern.h
+random.$(OBJEXT): {$(VPATH)}internal.h
+random.$(OBJEXT): {$(VPATH)}io.h
+random.$(OBJEXT): {$(VPATH)}missing.h
+random.$(OBJEXT): {$(VPATH)}onigmo.h
+random.$(OBJEXT): {$(VPATH)}oniguruma.h
+random.$(OBJEXT): {$(VPATH)}random.c
+random.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+random.$(OBJEXT): {$(VPATH)}siphash.c
+random.$(OBJEXT): {$(VPATH)}siphash.h
+random.$(OBJEXT): {$(VPATH)}st.h
+random.$(OBJEXT): {$(VPATH)}subst.h
+range.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+range.$(OBJEXT): $(top_srcdir)/include/ruby.h
+range.$(OBJEXT): {$(VPATH)}config.h
+range.$(OBJEXT): {$(VPATH)}defines.h
+range.$(OBJEXT): {$(VPATH)}encoding.h
+range.$(OBJEXT): {$(VPATH)}id.h
+range.$(OBJEXT): {$(VPATH)}intern.h
+range.$(OBJEXT): {$(VPATH)}internal.h
+range.$(OBJEXT): {$(VPATH)}io.h
+range.$(OBJEXT): {$(VPATH)}missing.h
+range.$(OBJEXT): {$(VPATH)}onigmo.h
+range.$(OBJEXT): {$(VPATH)}oniguruma.h
+range.$(OBJEXT): {$(VPATH)}range.c
+range.$(OBJEXT): {$(VPATH)}st.h
+range.$(OBJEXT): {$(VPATH)}subst.h
+rational.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+rational.$(OBJEXT): $(top_srcdir)/include/ruby.h
+rational.$(OBJEXT): {$(VPATH)}config.h
+rational.$(OBJEXT): {$(VPATH)}defines.h
+rational.$(OBJEXT): {$(VPATH)}encoding.h
+rational.$(OBJEXT): {$(VPATH)}id.h
+rational.$(OBJEXT): {$(VPATH)}intern.h
+rational.$(OBJEXT): {$(VPATH)}internal.h
+rational.$(OBJEXT): {$(VPATH)}io.h
+rational.$(OBJEXT): {$(VPATH)}missing.h
+rational.$(OBJEXT): {$(VPATH)}onigmo.h
+rational.$(OBJEXT): {$(VPATH)}oniguruma.h
+rational.$(OBJEXT): {$(VPATH)}rational.c
+rational.$(OBJEXT): {$(VPATH)}ruby_assert.h
+rational.$(OBJEXT): {$(VPATH)}st.h
+rational.$(OBJEXT): {$(VPATH)}subst.h
+re.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+re.$(OBJEXT): $(top_srcdir)/include/ruby.h
+re.$(OBJEXT): {$(VPATH)}config.h
+re.$(OBJEXT): {$(VPATH)}defines.h
+re.$(OBJEXT): {$(VPATH)}encindex.h
+re.$(OBJEXT): {$(VPATH)}encoding.h
+re.$(OBJEXT): {$(VPATH)}intern.h
+re.$(OBJEXT): {$(VPATH)}internal.h
+re.$(OBJEXT): {$(VPATH)}io.h
+re.$(OBJEXT): {$(VPATH)}missing.h
+re.$(OBJEXT): {$(VPATH)}onigmo.h
+re.$(OBJEXT): {$(VPATH)}oniguruma.h
+re.$(OBJEXT): {$(VPATH)}re.c
+re.$(OBJEXT): {$(VPATH)}re.h
+re.$(OBJEXT): {$(VPATH)}regenc.h
+re.$(OBJEXT): {$(VPATH)}regex.h
+re.$(OBJEXT): {$(VPATH)}regint.h
+re.$(OBJEXT): {$(VPATH)}st.h
+re.$(OBJEXT): {$(VPATH)}subst.h
+re.$(OBJEXT): {$(VPATH)}util.h
+regcomp.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+regcomp.$(OBJEXT): {$(VPATH)}config.h
+regcomp.$(OBJEXT): {$(VPATH)}defines.h
+regcomp.$(OBJEXT): {$(VPATH)}intern.h
+regcomp.$(OBJEXT): {$(VPATH)}missing.h
+regcomp.$(OBJEXT): {$(VPATH)}onigmo.h
+regcomp.$(OBJEXT): {$(VPATH)}oniguruma.h
+regcomp.$(OBJEXT): {$(VPATH)}regcomp.c
+regcomp.$(OBJEXT): {$(VPATH)}regenc.h
+regcomp.$(OBJEXT): {$(VPATH)}regint.h
+regcomp.$(OBJEXT): {$(VPATH)}regparse.h
+regcomp.$(OBJEXT): {$(VPATH)}st.h
+regcomp.$(OBJEXT): {$(VPATH)}subst.h
+regenc.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+regenc.$(OBJEXT): {$(VPATH)}config.h
+regenc.$(OBJEXT): {$(VPATH)}defines.h
+regenc.$(OBJEXT): {$(VPATH)}intern.h
+regenc.$(OBJEXT): {$(VPATH)}missing.h
+regenc.$(OBJEXT): {$(VPATH)}onigmo.h
+regenc.$(OBJEXT): {$(VPATH)}oniguruma.h
+regenc.$(OBJEXT): {$(VPATH)}regenc.c
+regenc.$(OBJEXT): {$(VPATH)}regenc.h
+regenc.$(OBJEXT): {$(VPATH)}regint.h
+regenc.$(OBJEXT): {$(VPATH)}st.h
+regenc.$(OBJEXT): {$(VPATH)}subst.h
+regerror.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+regerror.$(OBJEXT): {$(VPATH)}config.h
+regerror.$(OBJEXT): {$(VPATH)}defines.h
+regerror.$(OBJEXT): {$(VPATH)}intern.h
+regerror.$(OBJEXT): {$(VPATH)}missing.h
+regerror.$(OBJEXT): {$(VPATH)}onigmo.h
+regerror.$(OBJEXT): {$(VPATH)}oniguruma.h
+regerror.$(OBJEXT): {$(VPATH)}regenc.h
+regerror.$(OBJEXT): {$(VPATH)}regerror.c
+regerror.$(OBJEXT): {$(VPATH)}regint.h
+regerror.$(OBJEXT): {$(VPATH)}st.h
+regerror.$(OBJEXT): {$(VPATH)}subst.h
+regexec.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+regexec.$(OBJEXT): {$(VPATH)}config.h
+regexec.$(OBJEXT): {$(VPATH)}defines.h
+regexec.$(OBJEXT): {$(VPATH)}intern.h
+regexec.$(OBJEXT): {$(VPATH)}missing.h
+regexec.$(OBJEXT): {$(VPATH)}onigmo.h
+regexec.$(OBJEXT): {$(VPATH)}oniguruma.h
+regexec.$(OBJEXT): {$(VPATH)}regenc.h
+regexec.$(OBJEXT): {$(VPATH)}regexec.c
+regexec.$(OBJEXT): {$(VPATH)}regint.h
+regexec.$(OBJEXT): {$(VPATH)}st.h
+regexec.$(OBJEXT): {$(VPATH)}subst.h
+regparse.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+regparse.$(OBJEXT): {$(VPATH)}config.h
+regparse.$(OBJEXT): {$(VPATH)}defines.h
+regparse.$(OBJEXT): {$(VPATH)}intern.h
+regparse.$(OBJEXT): {$(VPATH)}missing.h
+regparse.$(OBJEXT): {$(VPATH)}onigmo.h
+regparse.$(OBJEXT): {$(VPATH)}oniguruma.h
+regparse.$(OBJEXT): {$(VPATH)}regenc.h
+regparse.$(OBJEXT): {$(VPATH)}regint.h
+regparse.$(OBJEXT): {$(VPATH)}regparse.c
+regparse.$(OBJEXT): {$(VPATH)}regparse.h
+regparse.$(OBJEXT): {$(VPATH)}st.h
+regparse.$(OBJEXT): {$(VPATH)}subst.h
+regsyntax.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+regsyntax.$(OBJEXT): {$(VPATH)}config.h
+regsyntax.$(OBJEXT): {$(VPATH)}defines.h
+regsyntax.$(OBJEXT): {$(VPATH)}intern.h
+regsyntax.$(OBJEXT): {$(VPATH)}missing.h
+regsyntax.$(OBJEXT): {$(VPATH)}onigmo.h
+regsyntax.$(OBJEXT): {$(VPATH)}oniguruma.h
+regsyntax.$(OBJEXT): {$(VPATH)}regenc.h
+regsyntax.$(OBJEXT): {$(VPATH)}regint.h
+regsyntax.$(OBJEXT): {$(VPATH)}regsyntax.c
+regsyntax.$(OBJEXT): {$(VPATH)}st.h
+regsyntax.$(OBJEXT): {$(VPATH)}subst.h
+ruby-runner.$(OBJEXT): {$(VPATH)}config.h
+ruby-runner.$(OBJEXT): {$(VPATH)}ruby-runner.c
+ruby-runner.$(OBJEXT): {$(VPATH)}ruby-runner.h
+ruby.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+ruby.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+ruby.$(OBJEXT): $(CCAN_DIR)/list/list.h
+ruby.$(OBJEXT): $(CCAN_DIR)/str/str.h
+ruby.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+ruby.$(OBJEXT): $(hdrdir)/ruby/version.h
+ruby.$(OBJEXT): $(top_srcdir)/include/ruby.h
+ruby.$(OBJEXT): {$(VPATH)}config.h
+ruby.$(OBJEXT): {$(VPATH)}defines.h
+ruby.$(OBJEXT): {$(VPATH)}dln.h
+ruby.$(OBJEXT): {$(VPATH)}encoding.h
+ruby.$(OBJEXT): {$(VPATH)}eval_intern.h
+ruby.$(OBJEXT): {$(VPATH)}id.h
+ruby.$(OBJEXT): {$(VPATH)}intern.h
+ruby.$(OBJEXT): {$(VPATH)}internal.h
+ruby.$(OBJEXT): {$(VPATH)}io.h
+ruby.$(OBJEXT): {$(VPATH)}method.h
+ruby.$(OBJEXT): {$(VPATH)}missing.h
+ruby.$(OBJEXT): {$(VPATH)}mjit.h
+ruby.$(OBJEXT): {$(VPATH)}node.h
+ruby.$(OBJEXT): {$(VPATH)}onigmo.h
+ruby.$(OBJEXT): {$(VPATH)}oniguruma.h
+ruby.$(OBJEXT): {$(VPATH)}ruby.c
+ruby.$(OBJEXT): {$(VPATH)}ruby_assert.h
+ruby.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+ruby.$(OBJEXT): {$(VPATH)}st.h
+ruby.$(OBJEXT): {$(VPATH)}subst.h
+ruby.$(OBJEXT): {$(VPATH)}thread.h
+ruby.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+ruby.$(OBJEXT): {$(VPATH)}thread_native.h
+ruby.$(OBJEXT): {$(VPATH)}util.h
+ruby.$(OBJEXT): {$(VPATH)}vm_core.h
+ruby.$(OBJEXT): {$(VPATH)}vm_debug.h
+ruby.$(OBJEXT): {$(VPATH)}vm_opts.h
+safe.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+safe.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+safe.$(OBJEXT): $(CCAN_DIR)/list/list.h
+safe.$(OBJEXT): $(CCAN_DIR)/str/str.h
+safe.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+safe.$(OBJEXT): $(top_srcdir)/include/ruby.h
+safe.$(OBJEXT): {$(VPATH)}config.h
+safe.$(OBJEXT): {$(VPATH)}defines.h
+safe.$(OBJEXT): {$(VPATH)}encoding.h
+safe.$(OBJEXT): {$(VPATH)}id.h
+safe.$(OBJEXT): {$(VPATH)}intern.h
+safe.$(OBJEXT): {$(VPATH)}internal.h
+safe.$(OBJEXT): {$(VPATH)}io.h
+safe.$(OBJEXT): {$(VPATH)}method.h
+safe.$(OBJEXT): {$(VPATH)}missing.h
+safe.$(OBJEXT): {$(VPATH)}node.h
+safe.$(OBJEXT): {$(VPATH)}onigmo.h
+safe.$(OBJEXT): {$(VPATH)}oniguruma.h
+safe.$(OBJEXT): {$(VPATH)}ruby_assert.h
+safe.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+safe.$(OBJEXT): {$(VPATH)}safe.c
+safe.$(OBJEXT): {$(VPATH)}st.h
+safe.$(OBJEXT): {$(VPATH)}subst.h
+safe.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+safe.$(OBJEXT): {$(VPATH)}thread_native.h
+safe.$(OBJEXT): {$(VPATH)}vm_core.h
+safe.$(OBJEXT): {$(VPATH)}vm_debug.h
+safe.$(OBJEXT): {$(VPATH)}vm_opts.h
+setproctitle.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+setproctitle.$(OBJEXT): $(top_srcdir)/include/ruby.h
+setproctitle.$(OBJEXT): {$(VPATH)}config.h
+setproctitle.$(OBJEXT): {$(VPATH)}defines.h
+setproctitle.$(OBJEXT): {$(VPATH)}intern.h
+setproctitle.$(OBJEXT): {$(VPATH)}missing.h
+setproctitle.$(OBJEXT): {$(VPATH)}setproctitle.c
+setproctitle.$(OBJEXT): {$(VPATH)}st.h
+setproctitle.$(OBJEXT): {$(VPATH)}subst.h
+setproctitle.$(OBJEXT): {$(VPATH)}util.h
+signal.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+signal.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+signal.$(OBJEXT): $(CCAN_DIR)/list/list.h
+signal.$(OBJEXT): $(CCAN_DIR)/str/str.h
+signal.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+signal.$(OBJEXT): $(top_srcdir)/include/ruby.h
+signal.$(OBJEXT): {$(VPATH)}config.h
+signal.$(OBJEXT): {$(VPATH)}debug_counter.h
+signal.$(OBJEXT): {$(VPATH)}defines.h
+signal.$(OBJEXT): {$(VPATH)}encoding.h
+signal.$(OBJEXT): {$(VPATH)}eval_intern.h
+signal.$(OBJEXT): {$(VPATH)}id.h
+signal.$(OBJEXT): {$(VPATH)}intern.h
+signal.$(OBJEXT): {$(VPATH)}internal.h
+signal.$(OBJEXT): {$(VPATH)}io.h
+signal.$(OBJEXT): {$(VPATH)}method.h
+signal.$(OBJEXT): {$(VPATH)}missing.h
+signal.$(OBJEXT): {$(VPATH)}node.h
+signal.$(OBJEXT): {$(VPATH)}onigmo.h
+signal.$(OBJEXT): {$(VPATH)}oniguruma.h
+signal.$(OBJEXT): {$(VPATH)}ruby_assert.h
+signal.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+signal.$(OBJEXT): {$(VPATH)}signal.c
+signal.$(OBJEXT): {$(VPATH)}st.h
+signal.$(OBJEXT): {$(VPATH)}subst.h
+signal.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+signal.$(OBJEXT): {$(VPATH)}thread_native.h
+signal.$(OBJEXT): {$(VPATH)}vm_core.h
+signal.$(OBJEXT): {$(VPATH)}vm_debug.h
+signal.$(OBJEXT): {$(VPATH)}vm_opts.h
+sprintf.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+sprintf.$(OBJEXT): $(top_srcdir)/include/ruby.h
+sprintf.$(OBJEXT): {$(VPATH)}config.h
+sprintf.$(OBJEXT): {$(VPATH)}defines.h
+sprintf.$(OBJEXT): {$(VPATH)}encoding.h
+sprintf.$(OBJEXT): {$(VPATH)}id.h
+sprintf.$(OBJEXT): {$(VPATH)}intern.h
+sprintf.$(OBJEXT): {$(VPATH)}internal.h
+sprintf.$(OBJEXT): {$(VPATH)}io.h
+sprintf.$(OBJEXT): {$(VPATH)}missing.h
+sprintf.$(OBJEXT): {$(VPATH)}onigmo.h
+sprintf.$(OBJEXT): {$(VPATH)}oniguruma.h
+sprintf.$(OBJEXT): {$(VPATH)}re.h
+sprintf.$(OBJEXT): {$(VPATH)}regex.h
+sprintf.$(OBJEXT): {$(VPATH)}sprintf.c
+sprintf.$(OBJEXT): {$(VPATH)}st.h
+sprintf.$(OBJEXT): {$(VPATH)}subst.h
+sprintf.$(OBJEXT): {$(VPATH)}vsnprintf.c
+st.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+st.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+st.$(OBJEXT): $(CCAN_DIR)/list/list.h
+st.$(OBJEXT): $(CCAN_DIR)/str/str.h
+st.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+st.$(OBJEXT): $(top_srcdir)/include/ruby.h
+st.$(OBJEXT): {$(VPATH)}config.h
+st.$(OBJEXT): {$(VPATH)}defines.h
+st.$(OBJEXT): {$(VPATH)}encoding.h
+st.$(OBJEXT): {$(VPATH)}intern.h
+st.$(OBJEXT): {$(VPATH)}internal.h
+st.$(OBJEXT): {$(VPATH)}io.h
+st.$(OBJEXT): {$(VPATH)}missing.h
+st.$(OBJEXT): {$(VPATH)}onigmo.h
+st.$(OBJEXT): {$(VPATH)}oniguruma.h
+st.$(OBJEXT): {$(VPATH)}st.c
+st.$(OBJEXT): {$(VPATH)}st.h
+st.$(OBJEXT): {$(VPATH)}subst.h
+strftime.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+strftime.$(OBJEXT): $(top_srcdir)/include/ruby.h
+strftime.$(OBJEXT): {$(VPATH)}config.h
+strftime.$(OBJEXT): {$(VPATH)}defines.h
+strftime.$(OBJEXT): {$(VPATH)}encoding.h
+strftime.$(OBJEXT): {$(VPATH)}intern.h
+strftime.$(OBJEXT): {$(VPATH)}internal.h
+strftime.$(OBJEXT): {$(VPATH)}io.h
+strftime.$(OBJEXT): {$(VPATH)}missing.h
+strftime.$(OBJEXT): {$(VPATH)}onigmo.h
+strftime.$(OBJEXT): {$(VPATH)}oniguruma.h
+strftime.$(OBJEXT): {$(VPATH)}st.h
+strftime.$(OBJEXT): {$(VPATH)}strftime.c
+strftime.$(OBJEXT): {$(VPATH)}subst.h
+strftime.$(OBJEXT): {$(VPATH)}timev.h
+string.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+string.$(OBJEXT): $(top_srcdir)/include/ruby.h
+string.$(OBJEXT): {$(VPATH)}config.h
+string.$(OBJEXT): {$(VPATH)}crypt.h
+string.$(OBJEXT): {$(VPATH)}debug_counter.h
+string.$(OBJEXT): {$(VPATH)}defines.h
+string.$(OBJEXT): {$(VPATH)}encindex.h
+string.$(OBJEXT): {$(VPATH)}encoding.h
+string.$(OBJEXT): {$(VPATH)}gc.h
+string.$(OBJEXT): {$(VPATH)}id.h
+string.$(OBJEXT): {$(VPATH)}intern.h
+string.$(OBJEXT): {$(VPATH)}internal.h
+string.$(OBJEXT): {$(VPATH)}io.h
+string.$(OBJEXT): {$(VPATH)}missing.h
+string.$(OBJEXT): {$(VPATH)}onigmo.h
+string.$(OBJEXT): {$(VPATH)}oniguruma.h
+string.$(OBJEXT): {$(VPATH)}probes.dmyh
+string.$(OBJEXT): {$(VPATH)}probes.h
+string.$(OBJEXT): {$(VPATH)}re.h
+string.$(OBJEXT): {$(VPATH)}regex.h
+string.$(OBJEXT): {$(VPATH)}ruby_assert.h
+string.$(OBJEXT): {$(VPATH)}st.h
+string.$(OBJEXT): {$(VPATH)}string.c
+string.$(OBJEXT): {$(VPATH)}subst.h
+string.$(OBJEXT): {$(VPATH)}util.h
+strlcat.$(OBJEXT): {$(VPATH)}config.h
+strlcat.$(OBJEXT): {$(VPATH)}missing.h
+strlcat.$(OBJEXT): {$(VPATH)}strlcat.c
+strlcpy.$(OBJEXT): {$(VPATH)}config.h
+strlcpy.$(OBJEXT): {$(VPATH)}missing.h
+strlcpy.$(OBJEXT): {$(VPATH)}strlcpy.c
+struct.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+struct.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+struct.$(OBJEXT): $(CCAN_DIR)/list/list.h
+struct.$(OBJEXT): $(CCAN_DIR)/str/str.h
+struct.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+struct.$(OBJEXT): $(top_srcdir)/include/ruby.h
+struct.$(OBJEXT): {$(VPATH)}config.h
+struct.$(OBJEXT): {$(VPATH)}defines.h
+struct.$(OBJEXT): {$(VPATH)}encoding.h
+struct.$(OBJEXT): {$(VPATH)}id.h
+struct.$(OBJEXT): {$(VPATH)}intern.h
+struct.$(OBJEXT): {$(VPATH)}internal.h
+struct.$(OBJEXT): {$(VPATH)}io.h
+struct.$(OBJEXT): {$(VPATH)}method.h
+struct.$(OBJEXT): {$(VPATH)}missing.h
+struct.$(OBJEXT): {$(VPATH)}node.h
+struct.$(OBJEXT): {$(VPATH)}onigmo.h
+struct.$(OBJEXT): {$(VPATH)}oniguruma.h
+struct.$(OBJEXT): {$(VPATH)}ruby_assert.h
+struct.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+struct.$(OBJEXT): {$(VPATH)}st.h
+struct.$(OBJEXT): {$(VPATH)}struct.c
+struct.$(OBJEXT): {$(VPATH)}subst.h
+struct.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+struct.$(OBJEXT): {$(VPATH)}thread_native.h
+struct.$(OBJEXT): {$(VPATH)}transient_heap.h
+struct.$(OBJEXT): {$(VPATH)}vm_core.h
+struct.$(OBJEXT): {$(VPATH)}vm_debug.h
+struct.$(OBJEXT): {$(VPATH)}vm_opts.h
+symbol.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+symbol.$(OBJEXT): $(top_srcdir)/include/ruby.h
+symbol.$(OBJEXT): {$(VPATH)}config.h
+symbol.$(OBJEXT): {$(VPATH)}defines.h
+symbol.$(OBJEXT): {$(VPATH)}encoding.h
+symbol.$(OBJEXT): {$(VPATH)}gc.h
+symbol.$(OBJEXT): {$(VPATH)}id.c
+symbol.$(OBJEXT): {$(VPATH)}id.h
+symbol.$(OBJEXT): {$(VPATH)}id_table.c
+symbol.$(OBJEXT): {$(VPATH)}id_table.h
+symbol.$(OBJEXT): {$(VPATH)}intern.h
+symbol.$(OBJEXT): {$(VPATH)}internal.h
+symbol.$(OBJEXT): {$(VPATH)}io.h
+symbol.$(OBJEXT): {$(VPATH)}missing.h
+symbol.$(OBJEXT): {$(VPATH)}onigmo.h
+symbol.$(OBJEXT): {$(VPATH)}oniguruma.h
+symbol.$(OBJEXT): {$(VPATH)}probes.dmyh
+symbol.$(OBJEXT): {$(VPATH)}probes.h
+symbol.$(OBJEXT): {$(VPATH)}ruby_assert.h
+symbol.$(OBJEXT): {$(VPATH)}st.h
+symbol.$(OBJEXT): {$(VPATH)}subst.h
+symbol.$(OBJEXT): {$(VPATH)}symbol.c
+symbol.$(OBJEXT): {$(VPATH)}symbol.h
+thread.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+thread.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+thread.$(OBJEXT): $(CCAN_DIR)/list/list.h
+thread.$(OBJEXT): $(CCAN_DIR)/str/str.h
+thread.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+thread.$(OBJEXT): $(top_srcdir)/include/ruby.h
+thread.$(OBJEXT): {$(VPATH)}config.h
+thread.$(OBJEXT): {$(VPATH)}debug.h
+thread.$(OBJEXT): {$(VPATH)}defines.h
+thread.$(OBJEXT): {$(VPATH)}encoding.h
+thread.$(OBJEXT): {$(VPATH)}eval_intern.h
+thread.$(OBJEXT): {$(VPATH)}gc.h
+thread.$(OBJEXT): {$(VPATH)}hrtime.h
+thread.$(OBJEXT): {$(VPATH)}id.h
+thread.$(OBJEXT): {$(VPATH)}intern.h
+thread.$(OBJEXT): {$(VPATH)}internal.h
+thread.$(OBJEXT): {$(VPATH)}io.h
+thread.$(OBJEXT): {$(VPATH)}iseq.h
+thread.$(OBJEXT): {$(VPATH)}method.h
+thread.$(OBJEXT): {$(VPATH)}missing.h
+thread.$(OBJEXT): {$(VPATH)}mjit.h
+thread.$(OBJEXT): {$(VPATH)}node.h
+thread.$(OBJEXT): {$(VPATH)}onigmo.h
+thread.$(OBJEXT): {$(VPATH)}oniguruma.h
+thread.$(OBJEXT): {$(VPATH)}ruby_assert.h
+thread.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+thread.$(OBJEXT): {$(VPATH)}st.h
+thread.$(OBJEXT): {$(VPATH)}subst.h
+thread.$(OBJEXT): {$(VPATH)}thread.c
+thread.$(OBJEXT): {$(VPATH)}thread.h
+thread.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).c
+thread.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+thread.$(OBJEXT): {$(VPATH)}thread_native.h
+thread.$(OBJEXT): {$(VPATH)}thread_sync.c
+thread.$(OBJEXT): {$(VPATH)}timev.h
+thread.$(OBJEXT): {$(VPATH)}vm_core.h
+thread.$(OBJEXT): {$(VPATH)}vm_debug.h
+thread.$(OBJEXT): {$(VPATH)}vm_opts.h
+time.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+time.$(OBJEXT): $(top_srcdir)/include/ruby.h
+time.$(OBJEXT): {$(VPATH)}config.h
+time.$(OBJEXT): {$(VPATH)}defines.h
+time.$(OBJEXT): {$(VPATH)}encoding.h
+time.$(OBJEXT): {$(VPATH)}id.h
+time.$(OBJEXT): {$(VPATH)}intern.h
+time.$(OBJEXT): {$(VPATH)}internal.h
+time.$(OBJEXT): {$(VPATH)}io.h
+time.$(OBJEXT): {$(VPATH)}missing.h
+time.$(OBJEXT): {$(VPATH)}onigmo.h
+time.$(OBJEXT): {$(VPATH)}oniguruma.h
+time.$(OBJEXT): {$(VPATH)}st.h
+time.$(OBJEXT): {$(VPATH)}subst.h
+time.$(OBJEXT): {$(VPATH)}time.c
+time.$(OBJEXT): {$(VPATH)}timev.h
+transcode.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+transcode.$(OBJEXT): $(top_srcdir)/include/ruby.h
+transcode.$(OBJEXT): {$(VPATH)}config.h
+transcode.$(OBJEXT): {$(VPATH)}defines.h
+transcode.$(OBJEXT): {$(VPATH)}encoding.h
+transcode.$(OBJEXT): {$(VPATH)}intern.h
+transcode.$(OBJEXT): {$(VPATH)}internal.h
+transcode.$(OBJEXT): {$(VPATH)}io.h
+transcode.$(OBJEXT): {$(VPATH)}missing.h
+transcode.$(OBJEXT): {$(VPATH)}onigmo.h
+transcode.$(OBJEXT): {$(VPATH)}oniguruma.h
+transcode.$(OBJEXT): {$(VPATH)}st.h
+transcode.$(OBJEXT): {$(VPATH)}subst.h
+transcode.$(OBJEXT): {$(VPATH)}transcode.c
+transcode.$(OBJEXT): {$(VPATH)}transcode_data.h
+transient_heap.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+transient_heap.$(OBJEXT): $(top_srcdir)/include/ruby.h
+transient_heap.$(OBJEXT): {$(VPATH)}config.h
+transient_heap.$(OBJEXT): {$(VPATH)}debug.h
+transient_heap.$(OBJEXT): {$(VPATH)}debug_counter.h
+transient_heap.$(OBJEXT): {$(VPATH)}defines.h
+transient_heap.$(OBJEXT): {$(VPATH)}gc.h
+transient_heap.$(OBJEXT): {$(VPATH)}intern.h
+transient_heap.$(OBJEXT): {$(VPATH)}internal.h
+transient_heap.$(OBJEXT): {$(VPATH)}missing.h
+transient_heap.$(OBJEXT): {$(VPATH)}node.h
+transient_heap.$(OBJEXT): {$(VPATH)}ruby_assert.h
+transient_heap.$(OBJEXT): {$(VPATH)}st.h
+transient_heap.$(OBJEXT): {$(VPATH)}subst.h
+transient_heap.$(OBJEXT): {$(VPATH)}transient_heap.c
+transient_heap.$(OBJEXT): {$(VPATH)}transient_heap.h
+transient_heap.$(OBJEXT): {$(VPATH)}vm_debug.h
+util.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+util.$(OBJEXT): $(top_srcdir)/include/ruby.h
+util.$(OBJEXT): {$(VPATH)}config.h
+util.$(OBJEXT): {$(VPATH)}defines.h
+util.$(OBJEXT): {$(VPATH)}encoding.h
+util.$(OBJEXT): {$(VPATH)}intern.h
+util.$(OBJEXT): {$(VPATH)}internal.h
+util.$(OBJEXT): {$(VPATH)}io.h
+util.$(OBJEXT): {$(VPATH)}missing.h
+util.$(OBJEXT): {$(VPATH)}onigmo.h
+util.$(OBJEXT): {$(VPATH)}oniguruma.h
+util.$(OBJEXT): {$(VPATH)}st.h
+util.$(OBJEXT): {$(VPATH)}subst.h
+util.$(OBJEXT): {$(VPATH)}util.c
+util.$(OBJEXT): {$(VPATH)}util.h
+variable.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+variable.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+variable.$(OBJEXT): $(CCAN_DIR)/list/list.h
+variable.$(OBJEXT): $(CCAN_DIR)/str/str.h
+variable.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+variable.$(OBJEXT): $(top_srcdir)/include/ruby.h
+variable.$(OBJEXT): {$(VPATH)}config.h
+variable.$(OBJEXT): {$(VPATH)}constant.h
+variable.$(OBJEXT): {$(VPATH)}debug_counter.h
+variable.$(OBJEXT): {$(VPATH)}defines.h
+variable.$(OBJEXT): {$(VPATH)}encoding.h
+variable.$(OBJEXT): {$(VPATH)}id.h
+variable.$(OBJEXT): {$(VPATH)}id_table.h
+variable.$(OBJEXT): {$(VPATH)}intern.h
+variable.$(OBJEXT): {$(VPATH)}internal.h
+variable.$(OBJEXT): {$(VPATH)}io.h
+variable.$(OBJEXT): {$(VPATH)}method.h
+variable.$(OBJEXT): {$(VPATH)}missing.h
+variable.$(OBJEXT): {$(VPATH)}node.h
+variable.$(OBJEXT): {$(VPATH)}onigmo.h
+variable.$(OBJEXT): {$(VPATH)}oniguruma.h
+variable.$(OBJEXT): {$(VPATH)}ruby_assert.h
+variable.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+variable.$(OBJEXT): {$(VPATH)}st.h
+variable.$(OBJEXT): {$(VPATH)}subst.h
+variable.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+variable.$(OBJEXT): {$(VPATH)}thread_native.h
+variable.$(OBJEXT): {$(VPATH)}transient_heap.h
+variable.$(OBJEXT): {$(VPATH)}util.h
+variable.$(OBJEXT): {$(VPATH)}variable.c
+variable.$(OBJEXT): {$(VPATH)}vm_core.h
+variable.$(OBJEXT): {$(VPATH)}vm_opts.h
+version.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+version.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+version.$(OBJEXT): $(CCAN_DIR)/list/list.h
+version.$(OBJEXT): $(CCAN_DIR)/str/str.h
+version.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+version.$(OBJEXT): $(hdrdir)/ruby/version.h
+version.$(OBJEXT): $(top_srcdir)/include/ruby.h
+version.$(OBJEXT): $(top_srcdir)/revision.h
+version.$(OBJEXT): $(top_srcdir)/version.h
+version.$(OBJEXT): {$(VPATH)}config.h
+version.$(OBJEXT): {$(VPATH)}defines.h
+version.$(OBJEXT): {$(VPATH)}id.h
+version.$(OBJEXT): {$(VPATH)}intern.h
+version.$(OBJEXT): {$(VPATH)}internal.h
+version.$(OBJEXT): {$(VPATH)}method.h
+version.$(OBJEXT): {$(VPATH)}missing.h
+version.$(OBJEXT): {$(VPATH)}mjit.h
+version.$(OBJEXT): {$(VPATH)}node.h
+version.$(OBJEXT): {$(VPATH)}ruby_assert.h
+version.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+version.$(OBJEXT): {$(VPATH)}st.h
+version.$(OBJEXT): {$(VPATH)}subst.h
+version.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+version.$(OBJEXT): {$(VPATH)}thread_native.h
+version.$(OBJEXT): {$(VPATH)}version.c
+version.$(OBJEXT): {$(VPATH)}vm_core.h
+version.$(OBJEXT): {$(VPATH)}vm_opts.h
+vm.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+vm.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+vm.$(OBJEXT): $(CCAN_DIR)/list/list.h
+vm.$(OBJEXT): $(CCAN_DIR)/str/str.h
+vm.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+vm.$(OBJEXT): $(top_srcdir)/include/ruby.h
+vm.$(OBJEXT): {$(VPATH)}config.h
+vm.$(OBJEXT): {$(VPATH)}constant.h
+vm.$(OBJEXT): {$(VPATH)}debug_counter.h
+vm.$(OBJEXT): {$(VPATH)}defines.h
+vm.$(OBJEXT): {$(VPATH)}defs/opt_operand.def
+vm.$(OBJEXT): {$(VPATH)}encoding.h
+vm.$(OBJEXT): {$(VPATH)}eval_intern.h
+vm.$(OBJEXT): {$(VPATH)}gc.h
+vm.$(OBJEXT): {$(VPATH)}id.h
+vm.$(OBJEXT): {$(VPATH)}id_table.h
+vm.$(OBJEXT): {$(VPATH)}insns.def
+vm.$(OBJEXT): {$(VPATH)}insns.inc
+vm.$(OBJEXT): {$(VPATH)}insns_info.inc
+vm.$(OBJEXT): {$(VPATH)}intern.h
+vm.$(OBJEXT): {$(VPATH)}internal.h
+vm.$(OBJEXT): {$(VPATH)}io.h
+vm.$(OBJEXT): {$(VPATH)}iseq.h
+vm.$(OBJEXT): {$(VPATH)}method.h
+vm.$(OBJEXT): {$(VPATH)}missing.h
+vm.$(OBJEXT): {$(VPATH)}mjit.h
+vm.$(OBJEXT): {$(VPATH)}node.h
+vm.$(OBJEXT): {$(VPATH)}onigmo.h
+vm.$(OBJEXT): {$(VPATH)}oniguruma.h
+vm.$(OBJEXT): {$(VPATH)}probes.dmyh
+vm.$(OBJEXT): {$(VPATH)}probes.h
+vm.$(OBJEXT): {$(VPATH)}probes_helper.h
+vm.$(OBJEXT): {$(VPATH)}ruby_assert.h
+vm.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+vm.$(OBJEXT): {$(VPATH)}st.h
+vm.$(OBJEXT): {$(VPATH)}subst.h
+vm.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+vm.$(OBJEXT): {$(VPATH)}thread_native.h
+vm.$(OBJEXT): {$(VPATH)}vm.c
+vm.$(OBJEXT): {$(VPATH)}vm.h
+vm.$(OBJEXT): {$(VPATH)}vm.inc
+vm.$(OBJEXT): {$(VPATH)}vm_args.c
+vm.$(OBJEXT): {$(VPATH)}vm_call_iseq_optimized.inc
+vm.$(OBJEXT): {$(VPATH)}vm_core.h
+vm.$(OBJEXT): {$(VPATH)}vm_debug.h
+vm.$(OBJEXT): {$(VPATH)}vm_eval.c
+vm.$(OBJEXT): {$(VPATH)}vm_exec.c
+vm.$(OBJEXT): {$(VPATH)}vm_exec.h
+vm.$(OBJEXT): {$(VPATH)}vm_insnhelper.c
+vm.$(OBJEXT): {$(VPATH)}vm_insnhelper.h
+vm.$(OBJEXT): {$(VPATH)}vm_method.c
+vm.$(OBJEXT): {$(VPATH)}vm_opts.h
+vm.$(OBJEXT): {$(VPATH)}vmtc.inc
+vm_backtrace.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+vm_backtrace.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+vm_backtrace.$(OBJEXT): $(CCAN_DIR)/list/list.h
+vm_backtrace.$(OBJEXT): $(CCAN_DIR)/str/str.h
+vm_backtrace.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+vm_backtrace.$(OBJEXT): $(top_srcdir)/include/ruby.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}config.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}debug.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}defines.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}encoding.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}eval_intern.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}id.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}intern.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}internal.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}io.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}iseq.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}method.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}missing.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}node.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}onigmo.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}oniguruma.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}ruby_assert.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}st.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}subst.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+vm_backtrace.$(OBJEXT): {$(VPATH)}thread_native.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}vm_backtrace.c
+vm_backtrace.$(OBJEXT): {$(VPATH)}vm_core.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}vm_debug.h
+vm_backtrace.$(OBJEXT): {$(VPATH)}vm_opts.h
+vm_dump.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+vm_dump.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+vm_dump.$(OBJEXT): $(CCAN_DIR)/list/list.h
+vm_dump.$(OBJEXT): $(CCAN_DIR)/str/str.h
+vm_dump.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+vm_dump.$(OBJEXT): $(top_srcdir)/include/ruby.h
+vm_dump.$(OBJEXT): {$(VPATH)}addr2line.h
+vm_dump.$(OBJEXT): {$(VPATH)}config.h
+vm_dump.$(OBJEXT): {$(VPATH)}defines.h
+vm_dump.$(OBJEXT): {$(VPATH)}encoding.h
+vm_dump.$(OBJEXT): {$(VPATH)}id.h
+vm_dump.$(OBJEXT): {$(VPATH)}intern.h
+vm_dump.$(OBJEXT): {$(VPATH)}internal.h
+vm_dump.$(OBJEXT): {$(VPATH)}io.h
+vm_dump.$(OBJEXT): {$(VPATH)}iseq.h
+vm_dump.$(OBJEXT): {$(VPATH)}method.h
+vm_dump.$(OBJEXT): {$(VPATH)}missing.h
+vm_dump.$(OBJEXT): {$(VPATH)}node.h
+vm_dump.$(OBJEXT): {$(VPATH)}onigmo.h
+vm_dump.$(OBJEXT): {$(VPATH)}oniguruma.h
+vm_dump.$(OBJEXT): {$(VPATH)}ruby_assert.h
+vm_dump.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+vm_dump.$(OBJEXT): {$(VPATH)}st.h
+vm_dump.$(OBJEXT): {$(VPATH)}subst.h
+vm_dump.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+vm_dump.$(OBJEXT): {$(VPATH)}thread_native.h
+vm_dump.$(OBJEXT): {$(VPATH)}vm_core.h
+vm_dump.$(OBJEXT): {$(VPATH)}vm_debug.h
+vm_dump.$(OBJEXT): {$(VPATH)}vm_dump.c
+vm_dump.$(OBJEXT): {$(VPATH)}vm_opts.h
+vm_trace.$(OBJEXT): $(CCAN_DIR)/check_type/check_type.h
+vm_trace.$(OBJEXT): $(CCAN_DIR)/container_of/container_of.h
+vm_trace.$(OBJEXT): $(CCAN_DIR)/list/list.h
+vm_trace.$(OBJEXT): $(CCAN_DIR)/str/str.h
+vm_trace.$(OBJEXT): $(hdrdir)/ruby/ruby.h
+vm_trace.$(OBJEXT): $(top_srcdir)/include/ruby.h
+vm_trace.$(OBJEXT): {$(VPATH)}config.h
+vm_trace.$(OBJEXT): {$(VPATH)}debug.h
+vm_trace.$(OBJEXT): {$(VPATH)}defines.h
+vm_trace.$(OBJEXT): {$(VPATH)}encoding.h
+vm_trace.$(OBJEXT): {$(VPATH)}eval_intern.h
+vm_trace.$(OBJEXT): {$(VPATH)}id.h
+vm_trace.$(OBJEXT): {$(VPATH)}intern.h
+vm_trace.$(OBJEXT): {$(VPATH)}internal.h
+vm_trace.$(OBJEXT): {$(VPATH)}io.h
+vm_trace.$(OBJEXT): {$(VPATH)}iseq.h
+vm_trace.$(OBJEXT): {$(VPATH)}method.h
+vm_trace.$(OBJEXT): {$(VPATH)}missing.h
+vm_trace.$(OBJEXT): {$(VPATH)}mjit.h
+vm_trace.$(OBJEXT): {$(VPATH)}node.h
+vm_trace.$(OBJEXT): {$(VPATH)}onigmo.h
+vm_trace.$(OBJEXT): {$(VPATH)}oniguruma.h
+vm_trace.$(OBJEXT): {$(VPATH)}ruby_assert.h
+vm_trace.$(OBJEXT): {$(VPATH)}ruby_atomic.h
+vm_trace.$(OBJEXT): {$(VPATH)}st.h
+vm_trace.$(OBJEXT): {$(VPATH)}subst.h
+vm_trace.$(OBJEXT): {$(VPATH)}thread_$(THREAD_MODEL).h
+vm_trace.$(OBJEXT): {$(VPATH)}thread_native.h
+vm_trace.$(OBJEXT): {$(VPATH)}vm_core.h
+vm_trace.$(OBJEXT): {$(VPATH)}vm_debug.h
+vm_trace.$(OBJEXT): {$(VPATH)}vm_opts.h
+vm_trace.$(OBJEXT): {$(VPATH)}vm_trace.c
+# AUTOGENERATED DEPENDENCIES END
diff --git a/compar.c b/compar.c
index 3cd3216b77..02529c9960 100644
--- a/compar.c
+++ b/compar.c
@@ -2,210 +2,227 @@
 
   compar.c -
 
-  $Author: dave $
-  $Date: 2003/12/19 00:01:18 $
+  $Author$
   created at: Thu Aug 26 14:39:48 JST 1993
 
-  Copyright (C) 1993-2003 Yukihiro Matsumoto
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
 
 **********************************************************************/
 
-#include "ruby.h"
+#include "ruby/ruby.h"
+#include "id.h"
 
 VALUE rb_mComparable;
 
-static ID cmp;
-
-int
-rb_cmpint(val, a, b)
-    VALUE val, a, b;
+static VALUE
+rb_cmp(VALUE x, VALUE y)
 {
-    if (NIL_P(val)) {
-	rb_cmperr(a, b);
-    }
-    if (FIXNUM_P(val)) return FIX2INT(val);
-    if (TYPE(val) == T_BIGNUM) {
-	if (RBIGNUM(val)->sign) return 1;
-	return -1;
-    }
-    if (RTEST(rb_funcall(val, '>', 1, INT2FIX(0)))) return 1;
-    if (RTEST(rb_funcall(val, '<', 1, INT2FIX(0)))) return -1;
-    return 0;
+    return rb_funcallv(x, idCmp, 1, &y);
 }
 
 void
-rb_cmperr(x, y)
-    VALUE x, y;
+rb_cmperr(VALUE x, VALUE y)
 {
-    const char *classname;
+    VALUE classname;
 
-    if (SPECIAL_CONST_P(y)) {
-	y = rb_inspect(y);
-	classname = StringValuePtr(y);
+    if (SPECIAL_CONST_P(y) || BUILTIN_TYPE(y) == T_FLOAT) {
+	classname = rb_inspect(y);
     }
     else {
-	classname = rb_obj_classname(y);
+	classname = rb_obj_class(y);
     }
-    rb_raise(rb_eArgError, "comparison of %s with %s failed",
-	     rb_obj_classname(x), classname);
+    rb_raise(rb_eArgError, "comparison of %"PRIsVALUE" with %"PRIsVALUE" failed",
+	     rb_obj_class(x), classname);
 }
 
-#define cmperr() (rb_cmperr(x, y), Qnil)
-
 static VALUE
-cmp_eq(a)
-    VALUE *a;
+invcmp_recursive(VALUE x, VALUE y, int recursive)
 {
-    VALUE c = rb_funcall(a[0], cmp, 1, a[1]);
+    if (recursive) return Qnil;
+    return rb_cmp(y, x);
+}
 
-    if (NIL_P(c)) return Qnil;
-    if (rb_cmpint(c, a[0], a[1]) == 0) return Qtrue;
-    return Qfalse;
+VALUE
+rb_invcmp(VALUE x, VALUE y)
+{
+    VALUE invcmp = rb_exec_recursive(invcmp_recursive, x, y);
+    if (invcmp == Qundef || NIL_P(invcmp)) {
+	return Qnil;
+    }
+    else {
+	int result = -rb_cmpint(invcmp, x, y);
+	return INT2FIX(result);
+    }
 }
 
 static VALUE
-cmp_failed()
+cmp_eq_recursive(VALUE arg1, VALUE arg2, int recursive)
 {
-    return Qnil;
+    if (recursive) return Qnil;
+    return rb_cmp(arg1, arg2);
 }
 
 /*
  *  call-seq:
- *     obj == other    => true or false
- *  
+ *     obj == other    -> true or false
+ *
  *  Compares two objects based on the receiver's <code><=></code>
  *  method, returning true if it returns 0. Also returns true if
  *  _obj_ and _other_ are the same object.
  */
 
 static VALUE
-cmp_equal(x, y)
-    VALUE x, y;
+cmp_equal(VALUE x, VALUE y)
 {
-    VALUE a[2];
-
+    VALUE c;
     if (x == y) return Qtrue;
 
-    a[0] = x; a[1] = y;
-    return rb_rescue(cmp_eq, (VALUE)a, cmp_failed, 0);
+    c = rb_exec_recursive_paired_outer(cmp_eq_recursive, x, y, y);
+
+    if (NIL_P(c)) return Qfalse;
+    if (rb_cmpint(c, x, y) == 0) return Qtrue;
+    return Qfalse;
+}
+
+static int
+cmpint(VALUE x, VALUE y)
+{
+    return rb_cmpint(rb_cmp(x, y), x, y);
 }
 
 /*
  *  call-seq:
- *     obj > other    => true or false
- *  
+ *     obj > other    -> true or false
+ *
  *  Compares two objects based on the receiver's <code><=></code>
  *  method, returning true if it returns 1.
  */
 
 static VALUE
-cmp_gt(x, y)
-    VALUE x, y;
+cmp_gt(VALUE x, VALUE y)
 {
-    VALUE c = rb_funcall(x, cmp, 1, y);
-
-    if (NIL_P(c)) return cmperr();
-    if (rb_cmpint(c, x, y) > 0) return Qtrue;
+    if (cmpint(x, y) > 0) return Qtrue;
     return Qfalse;
 }
 
 /*
  *  call-seq:
- *     obj >= other    => true or false
- *  
+ *     obj >= other    -> true or false
+ *
  *  Compares two objects based on the receiver's <code><=></code>
  *  method, returning true if it returns 0 or 1.
  */
 
 static VALUE
-cmp_ge(x, y)
-    VALUE x, y;
+cmp_ge(VALUE x, VALUE y)
 {
-    VALUE c = rb_funcall(x, cmp, 1, y);
-
-    if (NIL_P(c)) return cmperr();
-    if (rb_cmpint(c, x, y) >= 0) return Qtrue;
+    if (cmpint(x, y) >= 0) return Qtrue;
     return Qfalse;
 }
 
 /*
  *  call-seq:
- *     obj < other    => true or false
- *  
+ *     obj < other    -> true or false
+ *
  *  Compares two objects based on the receiver's <code><=></code>
  *  method, returning true if it returns -1.
  */
 
 static VALUE
-cmp_lt(x, y)
-    VALUE x, y;
+cmp_lt(VALUE x, VALUE y)
 {
-    VALUE c = rb_funcall(x, cmp, 1, y);
-
-    if (NIL_P(c)) return cmperr();
-    if (rb_cmpint(c, x, y) < 0) return Qtrue;
+    if (cmpint(x, y) < 0) return Qtrue;
     return Qfalse;
 }
 
-
 /*
  *  call-seq:
- *     obj <= other    => true or false
- *  
+ *     obj <= other    -> true or false
+ *
  *  Compares two objects based on the receiver's <code><=></code>
  *  method, returning true if it returns -1 or 0.
  */
 
 static VALUE
-cmp_le(x, y)
-    VALUE x, y;
+cmp_le(VALUE x, VALUE y)
 {
-    VALUE c = rb_funcall(x, cmp, 1, y);
-
-    if (NIL_P(c)) return cmperr();
-    if (rb_cmpint(c, x, y) <= 0) return Qtrue;
+    if (cmpint(x, y) <= 0) return Qtrue;
     return Qfalse;
 }
 
 /*
  *  call-seq:
- *     obj.between?(min, max)    => true or false
- *  
+ *     obj.between?(min, max)    -> true or false
+ *
  *  Returns <code>false</code> if <i>obj</i> <code><=></code>
  *  <i>min</i> is less than zero or if <i>anObject</i> <code><=></code>
  *  <i>max</i> is greater than zero, <code>true</code> otherwise.
- *     
+ *
  *     3.between?(1, 5)               #=> true
  *     6.between?(1, 5)               #=> false
  *     'cat'.between?('ant', 'dog')   #=> true
  *     'gnu'.between?('ant', 'dog')   #=> false
- *     
+ *
  */
 
 static VALUE
-cmp_between(x, min, max)
-    VALUE x, min, max;
+cmp_between(VALUE x, VALUE min, VALUE max)
 {
-    if (RTEST(cmp_lt(x, min))) return Qfalse;
-    if (RTEST(cmp_gt(x, max))) return Qfalse;
+    if (cmpint(x, min) < 0) return Qfalse;
+    if (cmpint(x, max) > 0) return Qfalse;
     return Qtrue;
 }
 
 /*
+ *  call-seq:
+ *     obj.clamp(min, max) ->  obj
+ *
+ * Returns <i>min</i> if <i>obj</i> <code><=></code> <i>min</i> is less
+ * than zero, <i>max</i> if <i>obj</i> <code><=></code> <i>max</i> is
+ * greater than zero and <i>obj</i> otherwise.
+ *
+ *     12.clamp(0, 100)         #=> 12
+ *     523.clamp(0, 100)        #=> 100
+ *     -3.123.clamp(0, 100)     #=> 0
+ *
+ *     'd'.clamp('a', 'f')      #=> 'd'
+ *     'z'.clamp('a', 'f')      #=> 'f'
+ */
+
+static VALUE
+cmp_clamp(VALUE x, VALUE min, VALUE max)
+{
+    int c;
+
+    if (cmpint(min, max) > 0) {
+	rb_raise(rb_eArgError, "min argument must be smaller than max argument");
+    }
+
+    c = cmpint(x, min);
+    if (c == 0) return x;
+    if (c < 0) return min;
+    c = cmpint(x, max);
+    if (c > 0) return max;
+    return x;
+}
+
+/*
  *  The <code>Comparable</code> mixin is used by classes whose objects
  *  may be ordered. The class must define the <code><=></code> operator,
  *  which compares the receiver against another object, returning -1, 0,
  *  or +1 depending on whether the receiver is less than, equal to, or
- *  greater than the other object. <code>Comparable</code> uses
+ *  greater than the other object. If the other object is not comparable
+ *  then the <code><=></code> operator should return nil.
+ *  <code>Comparable</code> uses
  *  <code><=></code> to implement the conventional comparison operators
  *  (<code><</code>, <code><=</code>, <code>==</code>, <code>>=</code>,
  *  and <code>></code>) and the method <code>between?</code>.
- *     
+ *
  *     class SizeMatters
  *       include Comparable
  *       attr :str
- *       def <=>(anOther)
- *         str.size <=> anOther.str.size
+ *       def <=>(other)
+ *         str.size <=> other.str.size
  *       end
  *       def initialize(str)
  *         @str = str
@@ -214,23 +231,26 @@ cmp_between(x, min, max)
  *         @str
  *       end
  *     end
- *     
+ *
  *     s1 = SizeMatters.new("Z")
  *     s2 = SizeMatters.new("YY")
  *     s3 = SizeMatters.new("XXX")
  *     s4 = SizeMatters.new("WWWW")
  *     s5 = SizeMatters.new("VVVVV")
- *     
+ *
  *     s1 < s2                       #=> true
  *     s4.between?(s1, s3)           #=> false
  *     s4.between?(s3, s5)           #=> true
  *     [ s3, s2, s5, s4, s1 ].sort   #=> [Z, YY, XXX, WWWW, VVVVV]
- *     
+ *
  */
 
 void
-Init_Comparable()
+Init_Comparable(void)
 {
+#undef rb_intern
+#define rb_intern(str) rb_intern_const(str)
+
     rb_mComparable = rb_define_module("Comparable");
     rb_define_method(rb_mComparable, "==", cmp_equal, 1);
     rb_define_method(rb_mComparable, ">", cmp_gt, 1);
@@ -238,6 +258,5 @@ Init_Comparable()
     rb_define_method(rb_mComparable, "<", cmp_lt, 1);
     rb_define_method(rb_mComparable, "<=", cmp_le, 1);
     rb_define_method(rb_mComparable, "between?", cmp_between, 2);
-
-    cmp = rb_intern("<=>");
+    rb_define_method(rb_mComparable, "clamp", cmp_clamp, 2);
 }
diff --git a/compile.c b/compile.c
new file mode 100644
index 0000000000..731b05323b
--- /dev/null
+++ b/compile.c
@@ -0,0 +1,10125 @@
+/**********************************************************************
+
+  compile.c - ruby node tree -> VM instruction sequence
+
+  $Author$
+  created at: 04/01/01 03:42:15 JST
+
+  Copyright (C) 2004-2007 Koichi Sasada
+
+**********************************************************************/
+
+#include "ruby/encoding.h"
+#include "ruby/re.h"
+#include "ruby/util.h"
+#include "internal.h"
+#include "encindex.h"
+#include <math.h>
+
+#include "vm_core.h"
+#include "vm_debug.h"
+#include "iseq.h"
+#include "insns.inc"
+#include "insns_info.inc"
+#include "id_table.h"
+#include "gc.h"
+
+#ifdef HAVE_DLADDR
+# include <dlfcn.h>
+#endif
+
+#undef RUBY_UNTYPED_DATA_WARNING
+#define RUBY_UNTYPED_DATA_WARNING 0
+
+#define FIXNUM_INC(n, i) ((n)+(INT2FIX(i)&~FIXNUM_FLAG))
+#define FIXNUM_OR(n, i) ((n)|INT2FIX(i))
+
+typedef struct iseq_link_element {
+    enum {
+	ISEQ_ELEMENT_ANCHOR,
+	ISEQ_ELEMENT_LABEL,
+	ISEQ_ELEMENT_INSN,
+	ISEQ_ELEMENT_ADJUST,
+	ISEQ_ELEMENT_TRACE
+    } type;
+    struct iseq_link_element *next;
+    struct iseq_link_element *prev;
+} LINK_ELEMENT;
+
+typedef struct iseq_link_anchor {
+    LINK_ELEMENT anchor;
+    LINK_ELEMENT *last;
+} LINK_ANCHOR;
+
+typedef enum {
+    LABEL_RESCUE_NONE,
+    LABEL_RESCUE_BEG,
+    LABEL_RESCUE_END,
+    LABEL_RESCUE_TYPE_MAX
+} LABEL_RESCUE_TYPE;
+
+typedef struct iseq_label_data {
+    LINK_ELEMENT link;
+    int label_no;
+    int position;
+    int sc_state;
+    int sp;
+    int refcnt;
+    unsigned int set: 1;
+    unsigned int rescued: 2;
+    unsigned int unremovable: 1;
+} LABEL;
+
+typedef struct iseq_insn_data {
+    LINK_ELEMENT link;
+    enum ruby_vminsn_type insn_id;
+    int operand_size;
+    int sc_state;
+    VALUE *operands;
+    struct {
+	int line_no;
+	rb_event_flag_t events;
+    } insn_info;
+} INSN;
+
+typedef struct iseq_adjust_data {
+    LINK_ELEMENT link;
+    LABEL *label;
+    int line_no;
+} ADJUST;
+
+typedef struct iseq_trace_data {
+    LINK_ELEMENT link;
+    rb_event_flag_t event;
+    long data;
+} TRACE;
+
+struct ensure_range {
+    LABEL *begin;
+    LABEL *end;
+    struct ensure_range *next;
+};
+
+struct iseq_compile_data_ensure_node_stack {
+    const NODE *ensure_node;
+    struct iseq_compile_data_ensure_node_stack *prev;
+    struct ensure_range *erange;
+};
+
+/**
+ * debug function(macro) interface depend on CPDEBUG
+ * if it is less than 0, runtime option is in effect.
+ *
+ * debug level:
+ *  0: no debug output
+ *  1: show node type
+ *  2: show node important parameters
+ *  ...
+ *  5: show other parameters
+ * 10: show every AST array
+ */
+
+#ifndef CPDEBUG
+#define CPDEBUG 0
+#endif
+
+#if CPDEBUG >= 0
+#define compile_debug CPDEBUG
+#else
+#define compile_debug ISEQ_COMPILE_DATA(iseq)->option->debug_level
+#endif
+
+#if CPDEBUG
+
+#define compile_debug_print_indent(level) \
+    ruby_debug_print_indent((level), compile_debug, gl_node_level * 2)
+
+#define debugp(header, value) (void) \
+  (compile_debug_print_indent(1) && \
+   ruby_debug_print_value(1, compile_debug, (header), (value)))
+
+#define debugi(header, id)  (void) \
+  (compile_debug_print_indent(1) && \
+   ruby_debug_print_id(1, compile_debug, (header), (id)))
+
+#define debugp_param(header, value)  (void) \
+  (compile_debug_print_indent(1) && \
+   ruby_debug_print_value(1, compile_debug, (header), (value)))
+
+#define debugp_verbose(header, value)  (void) \
+  (compile_debug_print_indent(2) && \
+   ruby_debug_print_value(2, compile_debug, (header), (value)))
+
+#define debugp_verbose_node(header, value)  (void) \
+  (compile_debug_print_indent(10) && \
+   ruby_debug_print_value(10, compile_debug, (header), (value)))
+
+#define debug_node_start(node)  ((void) \
+  (compile_debug_print_indent(1) && \
+   (ruby_debug_print_node(1, CPDEBUG, "", (const NODE *)(node)), gl_node_level)), \
+   gl_node_level++)
+
+#define debug_node_end()  gl_node_level --
+
+#else
+
+#define debugi(header, id)                 ((void)0)
+#define debugp(header, value)              ((void)0)
+#define debugp_verbose(header, value)      ((void)0)
+#define debugp_verbose_node(header, value) ((void)0)
+#define debugp_param(header, value)        ((void)0)
+#define debug_node_start(node)             ((void)0)
+#define debug_node_end()                   ((void)0)
+#endif
+
+#if CPDEBUG > 1 || CPDEBUG < 0
+#define printf ruby_debug_printf
+#define debugs if (compile_debug_print_indent(1)) ruby_debug_printf
+#define debug_compile(msg, v) ((void)(compile_debug_print_indent(1) && fputs((msg), stderr)), (v))
+#else
+#define debugs                             if(0)printf
+#define debug_compile(msg, v) (v)
+#endif
+
+#define LVAR_ERRINFO (1)
+
+/* create new label */
+#define NEW_LABEL(l) new_label_body(iseq, (l))
+#define LABEL_FORMAT "<L%03d>"
+
+#define NEW_ISEQ(node, name, type, line_no) \
+  new_child_iseq(iseq, (node), rb_fstring(name), 0, (type), (line_no))
+
+#define NEW_CHILD_ISEQ(node, name, type, line_no) \
+  new_child_iseq(iseq, (node), rb_fstring(name), iseq, (type), (line_no))
+
+/* add instructions */
+#define ADD_SEQ(seq1, seq2) \
+  APPEND_LIST((seq1), (seq2))
+
+/* add an instruction */
+#define ADD_INSN(seq, line, insn) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) new_insn_body(iseq, (line), BIN(insn), 0))
+
+/* insert an instruction before next */
+#define INSERT_BEFORE_INSN(next, line, insn) \
+  ELEM_INSERT_PREV(&(next)->link, (LINK_ELEMENT *) new_insn_body(iseq, (line), BIN(insn), 0))
+
+/* insert an instruction after prev */
+#define INSERT_AFTER_INSN(prev, line, insn) \
+  ELEM_INSERT_NEXT(&(prev)->link, (LINK_ELEMENT *) new_insn_body(iseq, (line), BIN(insn), 0))
+
+/* add an instruction with some operands (1, 2, 3, 5) */
+#define ADD_INSN1(seq, line, insn, op1) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) \
+           new_insn_body(iseq, (line), BIN(insn), 1, (VALUE)(op1)))
+
+/* insert an instruction with some operands (1, 2, 3, 5) before next */
+#define INSERT_BEFORE_INSN1(next, line, insn, op1) \
+  ELEM_INSERT_PREV(&(next)->link, (LINK_ELEMENT *) \
+           new_insn_body(iseq, (line), BIN(insn), 1, (VALUE)(op1)))
+
+/* insert an instruction with some operands (1, 2, 3, 5) after prev */
+#define INSERT_AFTER_INSN1(prev, line, insn, op1) \
+  ELEM_INSERT_NEXT(&(prev)->link, (LINK_ELEMENT *) \
+           new_insn_body(iseq, (line), BIN(insn), 1, (VALUE)(op1)))
+
+#define LABEL_REF(label) ((label)->refcnt++)
+
+/* add an instruction with label operand (alias of ADD_INSN1) */
+#define ADD_INSNL(seq, line, insn, label) (ADD_INSN1(seq, line, insn, label), LABEL_REF(label))
+
+#define ADD_INSN2(seq, line, insn, op1, op2) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) \
+           new_insn_body(iseq, (line), BIN(insn), 2, (VALUE)(op1), (VALUE)(op2)))
+
+#define ADD_INSN3(seq, line, insn, op1, op2, op3) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) \
+           new_insn_body(iseq, (line), BIN(insn), 3, (VALUE)(op1), (VALUE)(op2), (VALUE)(op3)))
+
+/* Specific Insn factory */
+#define ADD_SEND(seq, line, id, argc) \
+  ADD_SEND_R((seq), (line), (id), (argc), NULL, (VALUE)INT2FIX(0), NULL)
+
+#define ADD_SEND_WITH_FLAG(seq, line, id, argc, flag) \
+  ADD_SEND_R((seq), (line), (id), (argc), NULL, (VALUE)(flag), NULL)
+
+#define ADD_SEND_WITH_BLOCK(seq, line, id, argc, block) \
+  ADD_SEND_R((seq), (line), (id), (argc), (block), (VALUE)INT2FIX(0), NULL)
+
+#define ADD_CALL_RECEIVER(seq, line) \
+  ADD_INSN((seq), (line), putself)
+
+#define ADD_CALL(seq, line, id, argc) \
+  ADD_SEND_R((seq), (line), (id), (argc), NULL, (VALUE)INT2FIX(VM_CALL_FCALL), NULL)
+
+#define ADD_CALL_WITH_BLOCK(seq, line, id, argc, block) \
+  ADD_SEND_R((seq), (line), (id), (argc), (block), (VALUE)INT2FIX(VM_CALL_FCALL), NULL)
+
+#define ADD_SEND_R(seq, line, id, argc, block, flag, keywords) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) new_insn_send(iseq, (line), (id), (VALUE)(argc), (block), (VALUE)(flag), (keywords)))
+
+#define ADD_TRACE(seq, event) \
+  ADD_ELEM((seq), (LINK_ELEMENT *)new_trace_body(iseq, (event), 0))
+#define ADD_TRACE_WITH_DATA(seq, event, data) \
+  ADD_ELEM((seq), (LINK_ELEMENT *)new_trace_body(iseq, (event), (data)))
+
+
+#define DECL_BRANCH_BASE(branches, first_line, first_column, last_line, last_column, type) \
+  do { \
+      if (ISEQ_COVERAGE(iseq) && \
+	  ISEQ_BRANCH_COVERAGE(iseq) && \
+	  (first_line) > 0) { \
+	  VALUE structure = RARRAY_AREF(ISEQ_BRANCH_COVERAGE(iseq), 0); \
+	  branches = rb_ary_tmp_new(0); \
+	  rb_ary_push(structure, branches); \
+	  rb_ary_push(branches, ID2SYM(rb_intern(type))); \
+	  rb_ary_push(branches, INT2FIX(first_line)); \
+	  rb_ary_push(branches, INT2FIX(first_column)); \
+	  rb_ary_push(branches, INT2FIX(last_line)); \
+	  rb_ary_push(branches, INT2FIX(last_column)); \
+      } \
+  } while (0)
+#define ADD_TRACE_BRANCH_COVERAGE(seq, first_line, first_column, last_line, last_column, type, branches) \
+  do { \
+      if (ISEQ_COVERAGE(iseq) && \
+	  ISEQ_BRANCH_COVERAGE(iseq) && \
+	  (first_line) > 0) { \
+	  VALUE counters = RARRAY_AREF(ISEQ_BRANCH_COVERAGE(iseq), 1); \
+	  long counter_idx = RARRAY_LEN(counters); \
+	  rb_ary_push(counters, INT2FIX(0)); \
+	  rb_ary_push(branches, ID2SYM(rb_intern(type))); \
+	  rb_ary_push(branches, INT2FIX(first_line)); \
+	  rb_ary_push(branches, INT2FIX(first_column)); \
+	  rb_ary_push(branches, INT2FIX(last_line)); \
+	  rb_ary_push(branches, INT2FIX(last_column)); \
+	  rb_ary_push(branches, INT2FIX(counter_idx)); \
+          ADD_TRACE_WITH_DATA(seq, RUBY_EVENT_COVERAGE_BRANCH, counter_idx); \
+          ADD_INSN(seq, last_line, nop); \
+      } \
+  } while (0)
+
+static void iseq_add_getlocal(rb_iseq_t *iseq, LINK_ANCHOR *const seq, int line, int idx, int level);
+static void iseq_add_setlocal(rb_iseq_t *iseq, LINK_ANCHOR *const seq, int line, int idx, int level);
+
+#define ADD_GETLOCAL(seq, line, idx, level) iseq_add_getlocal(iseq, (seq), (line), (idx), (level))
+#define ADD_SETLOCAL(seq, line, idx, level) iseq_add_setlocal(iseq, (seq), (line), (idx), (level))
+
+/* add label */
+#define ADD_LABEL(seq, label) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) (label))
+
+#define APPEND_LABEL(seq, before, label) \
+  APPEND_ELEM((seq), (before), (LINK_ELEMENT *) (label))
+
+#define ADD_ADJUST(seq, line, label) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) new_adjust_body(iseq, (label), (line)))
+
+#define ADD_ADJUST_RESTORE(seq, label) \
+  ADD_ELEM((seq), (LINK_ELEMENT *) new_adjust_body(iseq, (label), -1))
+
+#define LABEL_UNREMOVABLE(label) \
+    ((label) ? (LABEL_REF(label), (label)->unremovable=1) : 0)
+#define ADD_CATCH_ENTRY(type, ls, le, iseqv, lc) do {				\
+    VALUE _e = rb_ary_new3(5, (type),						\
+			   (VALUE)(ls) | 1, (VALUE)(le) | 1,			\
+			   (VALUE)(iseqv), (VALUE)(lc) | 1);			\
+    LABEL_UNREMOVABLE(ls);							\
+    LABEL_REF(le);								\
+    LABEL_REF(lc);								\
+    rb_ary_push(ISEQ_COMPILE_DATA(iseq)->catch_table_ary, freeze_hide_obj(_e));	\
+} while (0)
+
+/* compile node */
+#define COMPILE(anchor, desc, node) \
+  (debug_compile("== " desc "\n", \
+                 iseq_compile_each(iseq, (anchor), (node), 0)))
+
+/* compile node, this node's value will be popped */
+#define COMPILE_POPPED(anchor, desc, node)    \
+  (debug_compile("== " desc "\n", \
+                 iseq_compile_each(iseq, (anchor), (node), 1)))
+
+/* compile node, which is popped when 'popped' is true */
+#define COMPILE_(anchor, desc, node, popped)  \
+  (debug_compile("== " desc "\n", \
+                 iseq_compile_each(iseq, (anchor), (node), (popped))))
+
+#define COMPILE_RECV(anchor, desc, node) \
+    (private_recv_p(node) ? \
+     (ADD_INSN(anchor, nd_line(node), putself), VM_CALL_FCALL) : \
+     (COMPILE(anchor, desc, node->nd_recv), 0))
+
+#define OPERAND_AT(insn, idx) \
+  (((INSN*)(insn))->operands[(idx)])
+
+#define INSN_OF(insn) \
+  (((INSN*)(insn))->insn_id)
+
+#define IS_INSN(link) ((link)->type == ISEQ_ELEMENT_INSN)
+#define IS_LABEL(link) ((link)->type == ISEQ_ELEMENT_LABEL)
+#define IS_ADJUST(link) ((link)->type == ISEQ_ELEMENT_ADJUST)
+#define IS_TRACE(link) ((link)->type == ISEQ_ELEMENT_TRACE)
+#define IS_INSN_ID(iobj, insn) (INSN_OF(iobj) == BIN(insn))
+#define IS_NEXT_INSN_ID(link, insn) \
+    ((link)->next && IS_INSN((link)->next) && IS_INSN_ID((link)->next, insn))
+
+/* error */
+#if CPDEBUG > 0
+NORETURN(static void append_compile_error(rb_iseq_t *iseq, int line, const char *fmt, ...));
+#endif
+
+static void
+append_compile_error(const rb_iseq_t *iseq, int line, const char *fmt, ...)
+{
+    VALUE err_info = ISEQ_COMPILE_DATA(iseq)->err_info;
+    VALUE file = rb_iseq_path(iseq);
+    VALUE err = err_info == Qtrue ? Qfalse : err_info;
+    va_list args;
+
+    va_start(args, fmt);
+    err = rb_syntax_error_append(err, file, line, -1, NULL, fmt, args);
+    va_end(args);
+    if (NIL_P(err_info)) {
+	RB_OBJ_WRITE(iseq, &ISEQ_COMPILE_DATA(iseq)->err_info, err);
+	rb_set_errinfo(err);
+    }
+    else if (!err_info) {
+	RB_OBJ_WRITE(iseq, &ISEQ_COMPILE_DATA(iseq)->err_info, Qtrue);
+    }
+    if (compile_debug) rb_exc_fatal(err);
+}
+
+#if 0
+static void
+compile_bug(rb_iseq_t *iseq, int line, const char *fmt, ...)
+{
+    va_list args;
+    va_start(args, fmt);
+    rb_report_bug_valist(rb_iseq_path(iseq), line, fmt, args);
+    va_end(args);
+    abort();
+}
+#endif
+
+#define COMPILE_ERROR append_compile_error
+
+#define ERROR_ARGS_AT(n) iseq, nd_line(n),
+#define ERROR_ARGS ERROR_ARGS_AT(node)
+
+#define EXPECT_NODE(prefix, node, ndtype, errval) \
+do { \
+    const NODE *error_node = (node); \
+    enum node_type error_type = nd_type(error_node); \
+    if (error_type != (ndtype)) { \
+	COMPILE_ERROR(ERROR_ARGS_AT(error_node) \
+		      prefix ": " #ndtype " is expected, but %s", \
+		      ruby_node_name(error_type)); \
+	return errval; \
+    } \
+} while (0)
+
+#define EXPECT_NODE_NONULL(prefix, parent, ndtype, errval) \
+do { \
+    COMPILE_ERROR(ERROR_ARGS_AT(parent) \
+		  prefix ": must be " #ndtype ", but 0"); \
+    return errval; \
+} while (0)
+
+#define UNKNOWN_NODE(prefix, node, errval) \
+do { \
+    const NODE *error_node = (node); \
+    COMPILE_ERROR(ERROR_ARGS_AT(error_node) prefix ": unknown node (%s)", \
+		  ruby_node_name(nd_type(error_node))); \
+    return errval; \
+} while (0)
+
+#define COMPILE_OK 1
+#define COMPILE_NG 0
+
+#define CHECK(sub) if (!(sub)) {BEFORE_RETURN;return COMPILE_NG;}
+#define BEFORE_RETURN
+
+/* leave name uninitialized so that compiler warn if INIT_ANCHOR is
+ * missing */
+#define DECL_ANCHOR(name) \
+    LINK_ANCHOR name[1] = {{{ISEQ_ELEMENT_ANCHOR,},}}
+#define INIT_ANCHOR(name) \
+    (name->last = &name->anchor)
+
+static inline VALUE
+freeze_hide_obj(VALUE obj)
+{
+    OBJ_FREEZE(obj);
+    RBASIC_CLEAR_CLASS(obj);
+    return obj;
+}
+
+#include "optinsn.inc"
+#if OPT_INSTRUCTIONS_UNIFICATION
+#include "optunifs.inc"
+#endif
+
+/* for debug */
+#if CPDEBUG < 0
+#define ISEQ_ARG iseq,
+#define ISEQ_ARG_DECLARE rb_iseq_t *iseq,
+#else
+#define ISEQ_ARG
+#define ISEQ_ARG_DECLARE
+#endif
+
+#if CPDEBUG
+#define gl_node_level ISEQ_COMPILE_DATA(iseq)->node_level
+#endif
+
+static void dump_disasm_list_with_cursor(const LINK_ELEMENT *link, const LINK_ELEMENT *curr, const LABEL *dest);
+static void dump_disasm_list(const LINK_ELEMENT *elem);
+
+static int insn_data_length(INSN *iobj);
+static int calc_sp_depth(int depth, INSN *iobj);
+
+static INSN *new_insn_body(rb_iseq_t *iseq, int line_no, enum ruby_vminsn_type insn_id, int argc, ...);
+static LABEL *new_label_body(rb_iseq_t *iseq, long line);
+static ADJUST *new_adjust_body(rb_iseq_t *iseq, LABEL *label, int line);
+static TRACE *new_trace_body(rb_iseq_t *iseq, rb_event_flag_t event, long data);
+
+
+static int iseq_compile_each(rb_iseq_t *iseq, LINK_ANCHOR *anchor, const NODE *n, int);
+static int iseq_setup(rb_iseq_t *iseq, LINK_ANCHOR *const anchor);
+static int iseq_setup_insn(rb_iseq_t *iseq, LINK_ANCHOR *const anchor);
+static int iseq_optimize(rb_iseq_t *iseq, LINK_ANCHOR *const anchor);
+static int iseq_insns_unification(rb_iseq_t *iseq, LINK_ANCHOR *const anchor);
+
+static int iseq_set_local_table(rb_iseq_t *iseq, const ID *tbl);
+static int iseq_set_exception_local_table(rb_iseq_t *iseq);
+static int iseq_set_arguments(rb_iseq_t *iseq, LINK_ANCHOR *const anchor, const NODE *const node);
+
+static int iseq_set_sequence_stackcaching(rb_iseq_t *iseq, LINK_ANCHOR *const anchor);
+static int iseq_set_sequence(rb_iseq_t *iseq, LINK_ANCHOR *const anchor);
+static int iseq_set_exception_table(rb_iseq_t *iseq);
+static int iseq_set_optargs_table(rb_iseq_t *iseq);
+
+static int compile_defined_expr(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, VALUE needstr);
+
+/*
+ * To make Array to LinkedList, use link_anchor
+ */
+
+static void
+verify_list(ISEQ_ARG_DECLARE const char *info, LINK_ANCHOR *const anchor)
+{
+#if CPDEBUG
+    int flag = 0;
+    LINK_ELEMENT *list, *plist;
+
+    if (!compile_debug) return;
+
+    list = anchor->anchor.next;
+    plist = &anchor->anchor;
+    while (list) {
+	if (plist != list->prev) {
+	    flag += 1;
+	}
+	plist = list;
+	list = list->next;
+    }
+
+    if (anchor->last != plist && anchor->last != 0) {
+	flag |= 0x70000;
+    }
+
+    if (flag != 0) {
+	rb_bug("list verify error: %08x (%s)", flag, info);
+    }
+#endif
+}
+#if CPDEBUG < 0
+#define verify_list(info, anchor) verify_list(iseq, (info), (anchor))
+#endif
+
+/*
+ * elem1, elem2 => elem1, elem2, elem
+ */
+static void
+ADD_ELEM(ISEQ_ARG_DECLARE LINK_ANCHOR *const anchor, LINK_ELEMENT *elem)
+{
+    elem->prev = anchor->last;
+    anchor->last->next = elem;
+    anchor->last = elem;
+    verify_list("add", anchor);
+}
+
+/*
+ * elem1, before, elem2 => elem1, before, elem, elem2
+ */
+static void
+APPEND_ELEM(ISEQ_ARG_DECLARE LINK_ANCHOR *const anchor, LINK_ELEMENT *before, LINK_ELEMENT *elem)
+{
+    elem->prev = before;
+    elem->next = before->next;
+    elem->next->prev = elem;
+    before->next = elem;
+    if (before == anchor->last) anchor->last = elem;
+    verify_list("add", anchor);
+}
+#if CPDEBUG < 0
+#define ADD_ELEM(anchor, elem) ADD_ELEM(iseq, (anchor), (elem))
+#define APPEND_ELEM(anchor, before, elem) APPEND_ELEM(iseq, (anchor), (before), (elem))
+#endif
+
+#define ISEQ_LAST_LINE(iseq) (ISEQ_COMPILE_DATA(iseq)->last_line)
+
+static int
+iseq_add_mark_object_compile_time(const rb_iseq_t *iseq, VALUE v)
+{
+    if (!SPECIAL_CONST_P(v)) {
+	rb_ary_push(ISEQ_COMPILE_DATA(iseq)->mark_ary, v);
+    }
+    return COMPILE_OK;
+}
+
+static inline VALUE
+freeze_literal(rb_iseq_t *iseq, VALUE lit)
+{
+    lit = rb_fstring(lit);
+    rb_ary_push(ISEQ_COMPILE_DATA(iseq)->mark_ary, lit);
+    return lit;
+}
+
+static int
+validate_label(st_data_t name, st_data_t label, st_data_t arg)
+{
+    rb_iseq_t *iseq = (rb_iseq_t *)arg;
+    LABEL *lobj = (LABEL *)label;
+    if (!lobj->link.next) {
+	do {
+	    COMPILE_ERROR(iseq, lobj->position,
+			  "%"PRIsVALUE": undefined label",
+			  rb_sym2str((VALUE)name));
+	} while (0);
+    }
+    return ST_CONTINUE;
+}
+
+static void
+validate_labels(rb_iseq_t *iseq, st_table *labels_table)
+{
+    st_foreach(labels_table, validate_label, (st_data_t)iseq);
+    st_free_table(labels_table);
+}
+
+VALUE
+rb_iseq_compile_ifunc(rb_iseq_t *iseq, const struct vm_ifunc *ifunc)
+{
+    DECL_ANCHOR(ret);
+    INIT_ANCHOR(ret);
+
+    (*ifunc->func)(iseq, ret, ifunc->data);
+
+    ADD_INSN(ret, ISEQ_COMPILE_DATA(iseq)->last_line, leave);
+
+    CHECK(iseq_setup_insn(iseq, ret));
+    return iseq_setup(iseq, ret);
+}
+
+VALUE
+rb_iseq_compile_node(rb_iseq_t *iseq, const NODE *node)
+{
+    DECL_ANCHOR(ret);
+    INIT_ANCHOR(ret);
+
+    if (imemo_type_p((VALUE)node, imemo_ifunc)) {
+        rb_raise(rb_eArgError, "unexpected imemo_ifunc");
+    }
+
+    if (node == 0) {
+	COMPILE(ret, "nil", node);
+	iseq_set_local_table(iseq, 0);
+    }
+    /* assume node is T_NODE */
+    else if (nd_type(node) == NODE_SCOPE) {
+	/* iseq type of top, method, class, block */
+	iseq_set_local_table(iseq, node->nd_tbl);
+	iseq_set_arguments(iseq, ret, node->nd_args);
+
+	switch (iseq->body->type) {
+	  case ISEQ_TYPE_BLOCK:
+	    {
+		LABEL *start = ISEQ_COMPILE_DATA(iseq)->start_label = NEW_LABEL(0);
+		LABEL *end = ISEQ_COMPILE_DATA(iseq)->end_label = NEW_LABEL(0);
+
+		start->rescued = LABEL_RESCUE_BEG;
+		end->rescued = LABEL_RESCUE_END;
+
+		ADD_TRACE(ret, RUBY_EVENT_B_CALL);
+		ADD_INSN (ret, FIX2INT(iseq->body->location.first_lineno), nop);
+		ADD_LABEL(ret, start);
+		CHECK(COMPILE(ret, "block body", node->nd_body));
+		ADD_LABEL(ret, end);
+		ADD_TRACE(ret, RUBY_EVENT_B_RETURN);
+		ISEQ_COMPILE_DATA(iseq)->last_line = iseq->body->location.code_location.end_pos.lineno;
+
+		/* wide range catch handler must put at last */
+		ADD_CATCH_ENTRY(CATCH_TYPE_REDO, start, end, NULL, start);
+		ADD_CATCH_ENTRY(CATCH_TYPE_NEXT, start, end, NULL, end);
+		break;
+	    }
+	  case ISEQ_TYPE_CLASS:
+	    {
+		ADD_TRACE(ret, RUBY_EVENT_CLASS);
+		CHECK(COMPILE(ret, "scoped node", node->nd_body));
+		ADD_TRACE(ret, RUBY_EVENT_END);
+		ISEQ_COMPILE_DATA(iseq)->last_line = nd_line(node);
+		break;
+	    }
+	  case ISEQ_TYPE_METHOD:
+	    {
+		ADD_TRACE(ret, RUBY_EVENT_CALL);
+		CHECK(COMPILE(ret, "scoped node", node->nd_body));
+		ADD_TRACE(ret, RUBY_EVENT_RETURN);
+		ISEQ_COMPILE_DATA(iseq)->last_line = nd_line(node);
+		break;
+	    }
+	  default: {
+	    CHECK(COMPILE(ret, "scoped node", node->nd_body));
+	    break;
+	  }
+	}
+    }
+    else {
+	const char *m;
+#define INVALID_ISEQ_TYPE(type) \
+	ISEQ_TYPE_##type: m = #type; goto invalid_iseq_type
+	switch (iseq->body->type) {
+	  case INVALID_ISEQ_TYPE(METHOD);
+	  case INVALID_ISEQ_TYPE(CLASS);
+	  case INVALID_ISEQ_TYPE(BLOCK);
+	  case INVALID_ISEQ_TYPE(EVAL);
+	  case INVALID_ISEQ_TYPE(MAIN);
+	  case INVALID_ISEQ_TYPE(TOP);
+#undef INVALID_ISEQ_TYPE /* invalid iseq types end */
+	  case ISEQ_TYPE_RESCUE:
+	    iseq_set_exception_local_table(iseq);
+	    CHECK(COMPILE(ret, "rescue", node));
+	    break;
+	  case ISEQ_TYPE_ENSURE:
+	    iseq_set_exception_local_table(iseq);
+	    CHECK(COMPILE_POPPED(ret, "ensure", node));
+	    break;
+	  case ISEQ_TYPE_PLAIN:
+	    CHECK(COMPILE(ret, "ensure", node));
+	    break;
+	  default:
+	    COMPILE_ERROR(ERROR_ARGS "unknown scope: %d", iseq->body->type);
+	    return COMPILE_NG;
+	  invalid_iseq_type:
+	    COMPILE_ERROR(ERROR_ARGS "compile/ISEQ_TYPE_%s should not be reached", m);
+	    return COMPILE_NG;
+	}
+    }
+
+    if (iseq->body->type == ISEQ_TYPE_RESCUE || iseq->body->type == ISEQ_TYPE_ENSURE) {
+	ADD_GETLOCAL(ret, 0, LVAR_ERRINFO, 0);
+	ADD_INSN1(ret, 0, throw, INT2FIX(0) /* continue throw */ );
+    }
+    else {
+	ADD_INSN(ret, ISEQ_COMPILE_DATA(iseq)->last_line, leave);
+    }
+
+#if SUPPORT_JOKE
+    if (ISEQ_COMPILE_DATA(iseq)->labels_table) {
+	st_table *labels_table = ISEQ_COMPILE_DATA(iseq)->labels_table;
+	ISEQ_COMPILE_DATA(iseq)->labels_table = 0;
+	validate_labels(iseq, labels_table);
+    }
+#endif
+    CHECK(iseq_setup_insn(iseq, ret));
+    return iseq_setup(iseq, ret);
+}
+
+int
+rb_iseq_translate_threaded_code(rb_iseq_t *iseq)
+{
+#if OPT_DIRECT_THREADED_CODE || OPT_CALL_THREADED_CODE
+    const void * const *table = rb_vm_get_insns_address_table();
+    unsigned int i;
+    VALUE *encoded = (VALUE *)iseq->body->iseq_encoded;
+
+    for (i = 0; i < iseq->body->iseq_size; /* */ ) {
+	int insn = (int)iseq->body->iseq_encoded[i];
+	int len = insn_len(insn);
+	encoded[i] = (VALUE)table[insn];
+	i += len;
+    }
+    FL_SET(iseq, ISEQ_TRANSLATED);
+#endif
+    return COMPILE_OK;
+}
+
+VALUE *
+rb_iseq_original_iseq(const rb_iseq_t *iseq) /* cold path */
+{
+    VALUE *original_code;
+
+    if (ISEQ_ORIGINAL_ISEQ(iseq)) return ISEQ_ORIGINAL_ISEQ(iseq);
+    original_code = ISEQ_ORIGINAL_ISEQ_ALLOC(iseq, iseq->body->iseq_size);
+    MEMCPY(original_code, iseq->body->iseq_encoded, VALUE, iseq->body->iseq_size);
+
+#if OPT_DIRECT_THREADED_CODE || OPT_CALL_THREADED_CODE
+    {
+	unsigned int i;
+
+	for (i = 0; i < iseq->body->iseq_size; /* */ ) {
+	    const void *addr = (const void *)original_code[i];
+	    const int insn = rb_vm_insn_addr2insn(addr);
+
+	    original_code[i] = insn;
+	    i += insn_len(insn);
+	}
+    }
+#endif
+    return original_code;
+}
+
+/*********************************************/
+/* definition of data structure for compiler */
+/*********************************************/
+
+/*
+ * On 32-bit SPARC, GCC by default generates SPARC V7 code that may require
+ * 8-byte word alignment. On the other hand, Oracle Solaris Studio seems to
+ * generate SPARCV8PLUS code with unaligned memory access instructions.
+ * That is why the STRICT_ALIGNMENT is defined only with GCC.
+ */
+#if defined(__sparc) && SIZEOF_VOIDP == 4 && defined(__GNUC__)
+  #define STRICT_ALIGNMENT
+#endif
+
+#ifdef STRICT_ALIGNMENT
+  #if defined(HAVE_TRUE_LONG_LONG) && SIZEOF_LONG_LONG > SIZEOF_VALUE
+    #define ALIGNMENT_SIZE SIZEOF_LONG_LONG
+  #else
+    #define ALIGNMENT_SIZE SIZEOF_VALUE
+  #endif
+  #define PADDING_SIZE_MAX    ((size_t)((ALIGNMENT_SIZE) - 1))
+  #define ALIGNMENT_SIZE_MASK PADDING_SIZE_MAX
+  /* Note: ALIGNMENT_SIZE == (2 ** N) is expected. */
+#else
+  #define PADDING_SIZE_MAX 0
+#endif /* STRICT_ALIGNMENT */
+
+#ifdef STRICT_ALIGNMENT
+/* calculate padding size for aligned memory access */
+static size_t
+calc_padding(void *ptr, size_t size)
+{
+    size_t mis;
+    size_t padding = 0;
+
+    mis = (size_t)ptr & ALIGNMENT_SIZE_MASK;
+    if (mis > 0) {
+        padding = ALIGNMENT_SIZE - mis;
+    }
+/*
+ * On 32-bit sparc or equivalents, when a single VALUE is requested
+ * and padding == sizeof(VALUE), it is clear that no padding is needed.
+ */
+#if ALIGNMENT_SIZE > SIZEOF_VALUE
+    if (size == sizeof(VALUE) && padding == sizeof(VALUE)) {
+        padding = 0;
+    }
+#endif
+
+    return padding;
+}
+#endif /* STRICT_ALIGNMENT */
+
+static void *
+compile_data_alloc(rb_iseq_t *iseq, size_t size)
+{
+    void *ptr = 0;
+    struct iseq_compile_data_storage *storage =
+	ISEQ_COMPILE_DATA(iseq)->storage_current;
+#ifdef STRICT_ALIGNMENT
+    size_t padding = calc_padding((void *)&storage->buff[storage->pos], size);
+#else
+    const size_t padding = 0; /* expected to be optimized by compiler */
+#endif /* STRICT_ALIGNMENT */
+
+    if (size >= INT_MAX - padding) rb_memerror();
+    if (storage->pos + size + padding > storage->size) {
+	unsigned int alloc_size = storage->size;
+
+	while (alloc_size < size + PADDING_SIZE_MAX) {
+	    if (alloc_size >= INT_MAX / 2) rb_memerror();
+	    alloc_size *= 2;
+	}
+	storage->next = (void *)ALLOC_N(char, alloc_size +
+					offsetof(struct iseq_compile_data_storage, buff));
+	storage = ISEQ_COMPILE_DATA(iseq)->storage_current = storage->next;
+	storage->next = 0;
+	storage->pos = 0;
+	storage->size = alloc_size;
+#ifdef STRICT_ALIGNMENT
+        padding = calc_padding((void *)&storage->buff[storage->pos], size);
+#endif /* STRICT_ALIGNMENT */
+    }
+
+#ifdef STRICT_ALIGNMENT
+    storage->pos += (int)padding;
+#endif /* STRICT_ALIGNMENT */
+
+    ptr = (void *)&storage->buff[storage->pos];
+    storage->pos += (int)size;
+    return ptr;
+}
+
+static INSN *
+compile_data_alloc_insn(rb_iseq_t *iseq)
+{
+    return (INSN *)compile_data_alloc(iseq, sizeof(INSN));
+}
+
+static LABEL *
+compile_data_alloc_label(rb_iseq_t *iseq)
+{
+    return (LABEL *)compile_data_alloc(iseq, sizeof(LABEL));
+}
+
+static ADJUST *
+compile_data_alloc_adjust(rb_iseq_t *iseq)
+{
+    return (ADJUST *)compile_data_alloc(iseq, sizeof(ADJUST));
+}
+
+static TRACE *
+compile_data_alloc_trace(rb_iseq_t *iseq)
+{
+    return (TRACE *)compile_data_alloc(iseq, sizeof(TRACE));
+}
+
+/*
+ * elem1, elemX => elem1, elem2, elemX
+ */
+static void
+ELEM_INSERT_NEXT(LINK_ELEMENT *elem1, LINK_ELEMENT *elem2)
+{
+    elem2->next = elem1->next;
+    elem2->prev = elem1;
+    elem1->next = elem2;
+    if (elem2->next) {
+	elem2->next->prev = elem2;
+    }
+}
+
+/*
+ * elem1, elemX => elemX, elem2, elem1
+ */
+static void
+ELEM_INSERT_PREV(LINK_ELEMENT *elem1, LINK_ELEMENT *elem2)
+{
+    elem2->prev = elem1->prev;
+    elem2->next = elem1;
+    elem1->prev = elem2;
+    if (elem2->prev) {
+	elem2->prev->next = elem2;
+    }
+}
+
+/*
+ * elemX, elem1, elemY => elemX, elem2, elemY
+ */
+static void
+ELEM_REPLACE(LINK_ELEMENT *elem1, LINK_ELEMENT *elem2)
+{
+    elem2->prev = elem1->prev;
+    elem2->next = elem1->next;
+    if (elem1->prev) {
+	elem1->prev->next = elem2;
+    }
+    if (elem1->next) {
+	elem1->next->prev = elem2;
+    }
+}
+
+static void
+ELEM_REMOVE(LINK_ELEMENT *elem)
+{
+    elem->prev->next = elem->next;
+    if (elem->next) {
+	elem->next->prev = elem->prev;
+    }
+}
+
+static LINK_ELEMENT *
+FIRST_ELEMENT(const LINK_ANCHOR *const anchor)
+{
+    return anchor->anchor.next;
+}
+
+static LINK_ELEMENT *
+LAST_ELEMENT(LINK_ANCHOR *const anchor)
+{
+    return anchor->last;
+}
+
+static LINK_ELEMENT *
+POP_ELEMENT(ISEQ_ARG_DECLARE LINK_ANCHOR *const anchor)
+{
+    LINK_ELEMENT *elem = anchor->last;
+    anchor->last = anchor->last->prev;
+    anchor->last->next = 0;
+    verify_list("pop", anchor);
+    return elem;
+}
+#if CPDEBUG < 0
+#define POP_ELEMENT(anchor) POP_ELEMENT(iseq, (anchor))
+#endif
+
+static LINK_ELEMENT *
+ELEM_FIRST_INSN(LINK_ELEMENT *elem)
+{
+    while (elem) {
+	switch (elem->type) {
+	  case ISEQ_ELEMENT_INSN:
+	  case ISEQ_ELEMENT_ADJUST:
+	    return elem;
+	  default:
+	    elem = elem->next;
+	}
+    }
+    return NULL;
+}
+
+static int
+LIST_INSN_SIZE_ONE(const LINK_ANCHOR *const anchor)
+{
+    LINK_ELEMENT *first_insn = ELEM_FIRST_INSN(FIRST_ELEMENT(anchor));
+    if (first_insn != NULL &&
+	ELEM_FIRST_INSN(first_insn->next) == NULL) {
+	return TRUE;
+    }
+    else {
+	return FALSE;
+    }
+}
+
+static int
+LIST_INSN_SIZE_ZERO(const LINK_ANCHOR *const anchor)
+{
+    if (ELEM_FIRST_INSN(FIRST_ELEMENT(anchor)) == NULL) {
+	return TRUE;
+    }
+    else {
+	return FALSE;
+    }
+}
+
+/*
+ * anc1: e1, e2, e3
+ * anc2: e4, e5
+ *#=>
+ * anc1: e1, e2, e3, e4, e5
+ * anc2: e4, e5 (broken)
+ */
+static void
+APPEND_LIST(ISEQ_ARG_DECLARE LINK_ANCHOR *const anc1, LINK_ANCHOR *const anc2)
+{
+    if (anc2->anchor.next) {
+	anc1->last->next = anc2->anchor.next;
+	anc2->anchor.next->prev = anc1->last;
+	anc1->last = anc2->last;
+    }
+    verify_list("append", anc1);
+}
+#if CPDEBUG < 0
+#define APPEND_LIST(anc1, anc2) APPEND_LIST(iseq, (anc1), (anc2))
+#endif
+
+/*
+ * anc1: e1, e2, e3
+ * anc2: e4, e5
+ *#=>
+ * anc1: e4, e5, e1, e2, e3
+ * anc2: e4, e5 (broken)
+ */
+static void
+INSERT_LIST(ISEQ_ARG_DECLARE LINK_ANCHOR *const anc1, LINK_ANCHOR *const anc2)
+{
+    if (anc2->anchor.next) {
+	LINK_ELEMENT *first = anc1->anchor.next;
+	anc1->anchor.next = anc2->anchor.next;
+	anc1->anchor.next->prev = &anc1->anchor;
+	anc2->last->next = first;
+	if (first) {
+	    first->prev = anc2->last;
+	}
+	else {
+	    anc1->last = anc2->last;
+	}
+    }
+
+    verify_list("append", anc1);
+}
+#if CPDEBUG < 0
+#define INSERT_LIST(anc1, anc2) INSERT_LIST(iseq, (anc1), (anc2))
+#endif
+
+#if CPDEBUG && 0
+static void
+debug_list(ISEQ_ARG_DECLARE LINK_ANCHOR *const anchor)
+{
+    LINK_ELEMENT *list = FIRST_ELEMENT(anchor);
+    printf("----\n");
+    printf("anch: %p, frst: %p, last: %p\n", &anchor->anchor,
+	   anchor->anchor.next, anchor->last);
+    while (list) {
+	printf("curr: %p, next: %p, prev: %p, type: %d\n", list, list->next,
+	       list->prev, FIX2INT(list->type));
+	list = list->next;
+    }
+    printf("----\n");
+
+    dump_disasm_list(anchor->anchor.next);
+    verify_list("debug list", anchor);
+}
+#if CPDEBUG < 0
+#define debug_list(anc) debug_list(iseq, (anc))
+#endif
+#else
+#define debug_list(anc) ((void)0)
+#endif
+
+static TRACE *
+new_trace_body(rb_iseq_t *iseq, rb_event_flag_t event, long data)
+{
+    TRACE *trace = compile_data_alloc_trace(iseq);
+
+    trace->link.type = ISEQ_ELEMENT_TRACE;
+    trace->link.next = NULL;
+    trace->event = event;
+    trace->data = data;
+
+    return trace;
+}
+
+static LABEL *
+new_label_body(rb_iseq_t *iseq, long line)
+{
+    LABEL *labelobj = compile_data_alloc_label(iseq);
+
+    labelobj->link.type = ISEQ_ELEMENT_LABEL;
+    labelobj->link.next = 0;
+
+    labelobj->label_no = ISEQ_COMPILE_DATA(iseq)->label_no++;
+    labelobj->sc_state = 0;
+    labelobj->sp = -1;
+    labelobj->refcnt = 0;
+    labelobj->set = 0;
+    labelobj->rescued = LABEL_RESCUE_NONE;
+    labelobj->unremovable = 0;
+    return labelobj;
+}
+
+static ADJUST *
+new_adjust_body(rb_iseq_t *iseq, LABEL *label, int line)
+{
+    ADJUST *adjust = compile_data_alloc_adjust(iseq);
+    adjust->link.type = ISEQ_ELEMENT_ADJUST;
+    adjust->link.next = 0;
+    adjust->label = label;
+    adjust->line_no = line;
+    LABEL_UNREMOVABLE(label);
+    return adjust;
+}
+
+static INSN *
+new_insn_core(rb_iseq_t *iseq, int line_no,
+	      int insn_id, int argc, VALUE *argv)
+{
+    INSN *iobj = compile_data_alloc_insn(iseq);
+
+    /* printf("insn_id: %d, line: %d\n", insn_id, line_no); */
+
+    iobj->link.type = ISEQ_ELEMENT_INSN;
+    iobj->link.next = 0;
+    iobj->insn_id = insn_id;
+    iobj->insn_info.line_no = line_no;
+    iobj->insn_info.events = 0;
+    iobj->operands = argv;
+    iobj->operand_size = argc;
+    iobj->sc_state = 0;
+    return iobj;
+}
+
+static INSN *
+new_insn_body(rb_iseq_t *iseq, int line_no, enum ruby_vminsn_type insn_id, int argc, ...)
+{
+    VALUE *operands = 0;
+    va_list argv;
+    if (argc > 0) {
+	int i;
+	va_init_list(argv, argc);
+	operands = (VALUE *)compile_data_alloc(iseq, sizeof(VALUE) * argc);
+	for (i = 0; i < argc; i++) {
+	    VALUE v = va_arg(argv, VALUE);
+	    operands[i] = v;
+	}
+	va_end(argv);
+    }
+    return new_insn_core(iseq, line_no, insn_id, argc, operands);
+}
+
+static struct rb_call_info *
+new_callinfo(rb_iseq_t *iseq, ID mid, int argc, unsigned int flag, struct rb_call_info_kw_arg *kw_arg, int has_blockiseq)
+{
+    size_t size = kw_arg != NULL ? sizeof(struct rb_call_info_with_kwarg) : sizeof(struct rb_call_info);
+    struct rb_call_info *ci = (struct rb_call_info *)compile_data_alloc(iseq, size);
+    struct rb_call_info_with_kwarg *ci_kw = (struct rb_call_info_with_kwarg *)ci;
+
+    ci->mid = mid;
+    ci->flag = flag;
+    ci->orig_argc = argc;
+
+    if (kw_arg) {
+	ci->flag |= VM_CALL_KWARG;
+	ci_kw->kw_arg = kw_arg;
+	ci->orig_argc += kw_arg->keyword_len;
+	iseq->body->ci_kw_size++;
+    }
+    else {
+	iseq->body->ci_size++;
+    }
+
+    if (!(ci->flag & (VM_CALL_ARGS_SPLAT | VM_CALL_ARGS_BLOCKARG | VM_CALL_KW_SPLAT)) &&
+	kw_arg == NULL && !has_blockiseq) {
+	ci->flag |= VM_CALL_ARGS_SIMPLE;
+    }
+    return ci;
+}
+
+static INSN *
+new_insn_send(rb_iseq_t *iseq, int line_no, ID id, VALUE argc, const rb_iseq_t *blockiseq, VALUE flag, struct rb_call_info_kw_arg *keywords)
+{
+    VALUE *operands = (VALUE *)compile_data_alloc(iseq, sizeof(VALUE) * 3);
+    operands[0] = (VALUE)new_callinfo(iseq, id, FIX2INT(argc), FIX2INT(flag), keywords, blockiseq != NULL);
+    operands[1] = Qfalse; /* cache */
+    operands[2] = (VALUE)blockiseq;
+    return new_insn_core(iseq, line_no, BIN(send), 3, operands);
+}
+
+static rb_iseq_t *
+new_child_iseq(rb_iseq_t *iseq, const NODE *const node,
+	       VALUE name, const rb_iseq_t *parent, enum iseq_type type, int line_no)
+{
+    rb_iseq_t *ret_iseq;
+    rb_ast_body_t ast;
+
+    ast.root = node;
+    ast.compile_option = 0;
+    ast.line_count = -1;
+
+    debugs("[new_child_iseq]> ---------------------------------------\n");
+    ret_iseq = rb_iseq_new_with_opt(&ast, name,
+				    rb_iseq_path(iseq), rb_iseq_realpath(iseq),
+				    INT2FIX(line_no), parent, type, ISEQ_COMPILE_DATA(iseq)->option);
+    debugs("[new_child_iseq]< ---------------------------------------\n");
+    iseq_add_mark_object_compile_time(iseq, (VALUE)ret_iseq);
+    return ret_iseq;
+}
+
+static rb_iseq_t *
+new_child_iseq_ifunc(rb_iseq_t *iseq, const struct vm_ifunc *ifunc,
+		     VALUE name, const rb_iseq_t *parent, enum iseq_type type, int line_no)
+{
+    rb_iseq_t *ret_iseq;
+
+    debugs("[new_child_iseq_ifunc]> ---------------------------------------\n");
+    ret_iseq = rb_iseq_new_ifunc(ifunc, name,
+				 rb_iseq_path(iseq), rb_iseq_realpath(iseq),
+				 INT2FIX(line_no), parent, type, ISEQ_COMPILE_DATA(iseq)->option);
+    debugs("[new_child_iseq_ifunc]< ---------------------------------------\n");
+    iseq_add_mark_object_compile_time(iseq, (VALUE)ret_iseq);
+    return ret_iseq;
+}
+
+static void
+set_catch_except_p(struct rb_iseq_constant_body *body)
+{
+    body->catch_except_p = TRUE;
+    if (body->parent_iseq != NULL) {
+        set_catch_except_p(body->parent_iseq->body);
+    }
+}
+
+/* Set body->catch_except_p to TRUE if the ISeq may catch an exception. If it is FALSE,
+   JIT-ed code may be optimized.  If we are extremely conservative, we should set TRUE
+   if catch table exists.  But we want to optimize while loop, which always has catch
+   table entries for break/next/redo.
+
+   So this function sets TRUE for limited ISeqs with break/next/redo catch table entries
+   whose child ISeq would really raise an exception. */
+static void
+update_catch_except_flags(struct rb_iseq_constant_body *body)
+{
+    unsigned int pos;
+    size_t i;
+    int insn;
+    const struct iseq_catch_table *ct = body->catch_table;
+
+    /* This assumes that a block has parent_iseq which may catch an exception from the block, and that
+       BREAK/NEXT/REDO catch table entries are used only when `throw` insn is used in the block. */
+    pos = 0;
+    while (pos < body->iseq_size) {
+#if OPT_DIRECT_THREADED_CODE || OPT_CALL_THREADED_CODE
+        insn = rb_vm_insn_addr2insn((void *)body->iseq_encoded[pos]);
+#else
+        insn = (int)body->iseq_encoded[pos];
+#endif
+        if (insn == BIN(throw)) {
+            set_catch_except_p(body);
+            break;
+        }
+        pos += insn_len(insn);
+    }
+
+    if (ct == NULL)
+        return;
+
+    for (i = 0; i < ct->size; i++) {
+        const struct iseq_catch_table_entry *entry = &ct->entries[i];
+        if (entry->type != CATCH_TYPE_BREAK
+            && entry->type != CATCH_TYPE_NEXT
+            && entry->type != CATCH_TYPE_REDO) {
+            body->catch_except_p = TRUE;
+            break;
+        }
+    }
+}
+
+static void
+iseq_insert_nop_between_end_and_cont(rb_iseq_t *iseq)
+{
+    VALUE catch_table_ary = ISEQ_COMPILE_DATA(iseq)->catch_table_ary;
+    unsigned int i, tlen = (unsigned int)RARRAY_LEN(catch_table_ary);
+    const VALUE *tptr = RARRAY_CONST_PTR_TRANSIENT(catch_table_ary);
+    for (i = 0; i < tlen; i++) {
+        const VALUE *ptr = RARRAY_CONST_PTR_TRANSIENT(tptr[i]);
+        LINK_ELEMENT *end = (LINK_ELEMENT *)(ptr[2] & ~1);
+        LINK_ELEMENT *cont = (LINK_ELEMENT *)(ptr[4] & ~1);
+        LINK_ELEMENT *e;
+        for (e = end; e && (IS_LABEL(e) || IS_TRACE(e)); e = e->next) {
+            if (e == cont) {
+                INSN *nop = new_insn_core(iseq, 0, BIN(nop), 0, 0);
+                ELEM_INSERT_NEXT(end, &nop->link);
+                break;
+            }
+        }
+    }
+}
+
+static int
+iseq_setup_insn(rb_iseq_t *iseq, LINK_ANCHOR *const anchor)
+{
+    if (RTEST(ISEQ_COMPILE_DATA(iseq)->err_info))
+	return COMPILE_NG;
+
+    /* debugs("[compile step 2] (iseq_array_to_linkedlist)\n"); */
+
+    if (compile_debug > 5)
+	dump_disasm_list(FIRST_ELEMENT(anchor));
+
+    debugs("[compile step 3.1 (iseq_optimize)]\n");
+    iseq_optimize(iseq, anchor);
+
+    if (compile_debug > 5)
+	dump_disasm_list(FIRST_ELEMENT(anchor));
+
+    if (ISEQ_COMPILE_DATA(iseq)->option->instructions_unification) {
+	debugs("[compile step 3.2 (iseq_insns_unification)]\n");
+	iseq_insns_unification(iseq, anchor);
+	if (compile_debug > 5)
+	    dump_disasm_list(FIRST_ELEMENT(anchor));
+    }
+
+    if (ISEQ_COMPILE_DATA(iseq)->option->stack_caching) {
+	debugs("[compile step 3.3 (iseq_set_sequence_stackcaching)]\n");
+	iseq_set_sequence_stackcaching(iseq, anchor);
+	if (compile_debug > 5)
+	    dump_disasm_list(FIRST_ELEMENT(anchor));
+    }
+
+    debugs("[compile step 3.4 (iseq_insert_nop_between_end_and_cont)]\n");
+    iseq_insert_nop_between_end_and_cont(iseq);
+
+    return COMPILE_OK;
+}
+
+static int
+iseq_setup(rb_iseq_t *iseq, LINK_ANCHOR *const anchor)
+{
+    if (RTEST(ISEQ_COMPILE_DATA(iseq)->err_info))
+        return COMPILE_NG;
+
+    debugs("[compile step 4.1 (iseq_set_sequence)]\n");
+    if (!iseq_set_sequence(iseq, anchor)) return COMPILE_NG;
+    if (compile_debug > 5)
+	dump_disasm_list(FIRST_ELEMENT(anchor));
+
+    debugs("[compile step 4.2 (iseq_set_exception_table)]\n");
+    if (!iseq_set_exception_table(iseq)) return COMPILE_NG;
+
+    debugs("[compile step 4.3 (set_optargs_table)] \n");
+    if (!iseq_set_optargs_table(iseq)) return COMPILE_NG;
+
+    debugs("[compile step 5 (iseq_translate_threaded_code)] \n");
+    if (!rb_iseq_translate_threaded_code(iseq)) return COMPILE_NG;
+
+    update_catch_except_flags(iseq->body);
+
+    if (compile_debug > 1) {
+	VALUE str = rb_iseq_disasm(iseq);
+	printf("%s\n", StringValueCStr(str));
+    }
+    debugs("[compile step: finish]\n");
+
+    return COMPILE_OK;
+}
+
+static int
+iseq_set_exception_local_table(rb_iseq_t *iseq)
+{
+    /* TODO: every id table is same -> share it.
+     * Current problem is iseq_free().
+     */
+    ID id_dollar_bang;
+    ID *ids = (ID *)ALLOC_N(ID, 1);
+
+    CONST_ID(id_dollar_bang, "#$!");
+    iseq->body->local_table_size = 1;
+    ids[0] = id_dollar_bang;
+    iseq->body->local_table = ids;
+    return COMPILE_OK;
+}
+
+static int
+get_lvar_level(const rb_iseq_t *iseq)
+{
+    int lev = 0;
+    while (iseq != iseq->body->local_iseq) {
+	lev++;
+	iseq = iseq->body->parent_iseq;
+    }
+    return lev;
+}
+
+static int
+get_dyna_var_idx_at_raw(const rb_iseq_t *iseq, ID id)
+{
+    unsigned int i;
+
+    for (i = 0; i < iseq->body->local_table_size; i++) {
+	if (iseq->body->local_table[i] == id) {
+	    return (int)i;
+	}
+    }
+    return -1;
+}
+
+static int
+get_local_var_idx(const rb_iseq_t *iseq, ID id)
+{
+    int idx = get_dyna_var_idx_at_raw(iseq->body->local_iseq, id);
+
+    if (idx < 0) {
+        COMPILE_ERROR(iseq, ISEQ_LAST_LINE(iseq),
+                      "get_local_var_idx: %d", idx);
+    }
+
+    return idx;
+}
+
+static int
+get_dyna_var_idx(const rb_iseq_t *iseq, ID id, int *level, int *ls)
+{
+    int lv = 0, idx = -1;
+    const rb_iseq_t *const topmost_iseq = iseq;
+
+    while (iseq) {
+	idx = get_dyna_var_idx_at_raw(iseq, id);
+	if (idx >= 0) {
+	    break;
+	}
+	iseq = iseq->body->parent_iseq;
+	lv++;
+    }
+
+    if (idx < 0) {
+        COMPILE_ERROR(topmost_iseq, ISEQ_LAST_LINE(topmost_iseq),
+                      "get_dyna_var_idx: -1");
+    }
+
+    *level = lv;
+    *ls = iseq->body->local_table_size;
+    return idx;
+}
+
+static int
+iseq_local_block_param_p(const rb_iseq_t *iseq, unsigned int idx, unsigned int level)
+{
+    const struct rb_iseq_constant_body *body;
+    while (level > 0) {
+	iseq = iseq->body->parent_iseq;
+	level--;
+    }
+    body = iseq->body;
+    if (body->local_iseq == iseq && /* local variables */
+	body->param.flags.has_block &&
+	body->local_table_size - body->param.block_start == idx) {
+	return TRUE;
+    }
+    else {
+	return FALSE;
+    }
+}
+
+static int
+iseq_block_param_id_p(const rb_iseq_t *iseq, ID id, int *pidx, int *plevel)
+{
+    int level, ls;
+    int idx = get_dyna_var_idx(iseq, id, &level, &ls);
+    if (iseq_local_block_param_p(iseq, ls - idx, level)) {
+	*pidx = ls - idx;
+	*plevel = level;
+	return TRUE;
+    }
+    else {
+	return FALSE;
+    }
+}
+
+static void
+iseq_add_getlocal(rb_iseq_t *iseq, LINK_ANCHOR *const seq, int line, int idx, int level)
+{
+    if (iseq_local_block_param_p(iseq, idx, level)) {
+	ADD_INSN2(seq, line, getblockparam, INT2FIX((idx) + VM_ENV_DATA_SIZE - 1), INT2FIX(level));
+    }
+    else {
+	ADD_INSN2(seq, line, getlocal, INT2FIX((idx) + VM_ENV_DATA_SIZE - 1), INT2FIX(level));
+    }
+}
+
+static void
+iseq_add_setlocal(rb_iseq_t *iseq, LINK_ANCHOR *const seq, int line, int idx, int level)
+{
+    if (iseq_local_block_param_p(iseq, idx, level)) {
+	ADD_INSN2(seq, line, setblockparam, INT2FIX((idx) + VM_ENV_DATA_SIZE - 1), INT2FIX(level));
+    }
+    else {
+	ADD_INSN2(seq, line, setlocal, INT2FIX((idx) + VM_ENV_DATA_SIZE - 1), INT2FIX(level));
+    }
+}
+
+
+
+static void
+iseq_calc_param_size(rb_iseq_t *iseq)
+{
+    struct rb_iseq_constant_body *const body = iseq->body;
+    if (body->param.flags.has_opt ||
+	body->param.flags.has_post ||
+	body->param.flags.has_rest ||
+	body->param.flags.has_block ||
+	body->param.flags.has_kw ||
+	body->param.flags.has_kwrest) {
+
+	if (body->param.flags.has_block) {
+	    body->param.size = body->param.block_start + 1;
+	}
+	else if (body->param.flags.has_kwrest) {
+	    body->param.size = body->param.keyword->rest_start + 1;
+	}
+	else if (body->param.flags.has_kw) {
+	    body->param.size = body->param.keyword->bits_start + 1;
+	}
+	else if (body->param.flags.has_post) {
+	    body->param.size = body->param.post_start + body->param.post_num;
+	}
+	else if (body->param.flags.has_rest) {
+	    body->param.size = body->param.rest_start + 1;
+	}
+	else if (body->param.flags.has_opt) {
+	    body->param.size = body->param.lead_num + body->param.opt_num;
+	}
+	else {
+            UNREACHABLE;
+	}
+    }
+    else {
+	body->param.size = body->param.lead_num;
+    }
+}
+
+static int
+iseq_set_arguments_keywords(rb_iseq_t *iseq, LINK_ANCHOR *const optargs,
+			    const struct rb_args_info *args, int arg_size)
+{
+    const NODE *node = args->kw_args;
+    struct rb_iseq_constant_body *const body = iseq->body;
+    struct rb_iseq_param_keyword *keyword;
+    const VALUE default_values = rb_ary_tmp_new(1);
+    const VALUE complex_mark = rb_str_tmp_new(0);
+    int kw = 0, rkw = 0, di = 0, i;
+
+    body->param.flags.has_kw = TRUE;
+    body->param.keyword = keyword = ZALLOC_N(struct rb_iseq_param_keyword, 1);
+
+    while (node) {
+	kw++;
+	node = node->nd_next;
+    }
+    arg_size += kw;
+    keyword->bits_start = arg_size++;
+
+    node = args->kw_args;
+    while (node) {
+	const NODE *val_node = node->nd_body->nd_value;
+	VALUE dv;
+
+	if (val_node == (const NODE *)-1) {
+	    ++rkw;
+	}
+	else {
+	    switch (nd_type(val_node)) {
+	      case NODE_LIT:
+		dv = val_node->nd_lit;
+		iseq_add_mark_object_compile_time(iseq, dv);
+		break;
+	      case NODE_NIL:
+		dv = Qnil;
+		break;
+	      case NODE_TRUE:
+		dv = Qtrue;
+		break;
+	      case NODE_FALSE:
+		dv = Qfalse;
+		break;
+	      default:
+		COMPILE_POPPED(optargs, "kwarg", node); /* nd_type(node) == NODE_KW_ARG */
+		dv = complex_mark;
+	    }
+
+	    keyword->num = ++di;
+	    rb_ary_push(default_values, dv);
+	}
+
+	node = node->nd_next;
+    }
+
+    keyword->num = kw;
+
+    if (args->kw_rest_arg->nd_vid != 0) {
+	keyword->rest_start = arg_size++;
+	body->param.flags.has_kwrest = TRUE;
+    }
+    keyword->required_num = rkw;
+    keyword->table = &body->local_table[keyword->bits_start - keyword->num];
+
+    {
+	VALUE *dvs = ALLOC_N(VALUE, RARRAY_LEN(default_values));
+
+	for (i = 0; i < RARRAY_LEN(default_values); i++) {
+	    VALUE dv = RARRAY_AREF(default_values, i);
+	    if (dv == complex_mark) dv = Qundef;
+	    if (!SPECIAL_CONST_P(dv)) {
+		RB_OBJ_WRITTEN(iseq, Qundef, dv);
+	    }
+	    dvs[i] = dv;
+	}
+
+	keyword->default_values = dvs;
+    }
+    return arg_size;
+}
+
+static int
+iseq_set_arguments(rb_iseq_t *iseq, LINK_ANCHOR *const optargs, const NODE *const node_args)
+{
+    debugs("iseq_set_arguments: %s\n", node_args ? "" : "0");
+
+    if (node_args) {
+	struct rb_iseq_constant_body *const body = iseq->body;
+	struct rb_args_info *args = node_args->nd_ainfo;
+	ID rest_id = 0;
+	int last_comma = 0;
+	ID block_id = 0;
+	int arg_size;
+
+	EXPECT_NODE("iseq_set_arguments", node_args, NODE_ARGS, COMPILE_NG);
+
+	body->param.lead_num = arg_size = (int)args->pre_args_num;
+	if (body->param.lead_num > 0) body->param.flags.has_lead = TRUE;
+	debugs("  - argc: %d\n", body->param.lead_num);
+
+	rest_id = args->rest_arg;
+	if (rest_id == 1) {
+	    last_comma = 1;
+	    rest_id = 0;
+	}
+	block_id = args->block_arg;
+
+	if (args->opt_args) {
+	    const NODE *node = args->opt_args;
+	    LABEL *label;
+	    VALUE labels = rb_ary_tmp_new(1);
+	    VALUE *opt_table;
+	    int i = 0, j;
+
+	    while (node) {
+		label = NEW_LABEL(nd_line(node));
+		rb_ary_push(labels, (VALUE)label | 1);
+		ADD_LABEL(optargs, label);
+		COMPILE_POPPED(optargs, "optarg", node->nd_body);
+		node = node->nd_next;
+		i += 1;
+	    }
+
+	    /* last label */
+	    label = NEW_LABEL(nd_line(node_args));
+	    rb_ary_push(labels, (VALUE)label | 1);
+	    ADD_LABEL(optargs, label);
+
+	    opt_table = ALLOC_N(VALUE, i+1);
+
+            MEMCPY(opt_table, RARRAY_CONST_PTR_TRANSIENT(labels), VALUE, i+1);
+	    for (j = 0; j < i+1; j++) {
+		opt_table[j] &= ~1;
+	    }
+	    rb_ary_clear(labels);
+
+	    body->param.flags.has_opt = TRUE;
+	    body->param.opt_num = i;
+	    body->param.opt_table = opt_table;
+	    arg_size += i;
+	}
+
+	if (rest_id) {
+	    body->param.rest_start = arg_size++;
+	    body->param.flags.has_rest = TRUE;
+	    assert(body->param.rest_start != -1);
+	}
+
+	if (args->first_post_arg) {
+	    body->param.post_start = arg_size;
+	    body->param.post_num = args->post_args_num;
+	    body->param.flags.has_post = TRUE;
+	    arg_size += args->post_args_num;
+
+	    if (body->param.flags.has_rest) { /* TODO: why that? */
+		body->param.post_start = body->param.rest_start + 1;
+	    }
+	}
+
+	if (args->kw_args) {
+	    arg_size = iseq_set_arguments_keywords(iseq, optargs, args, arg_size);
+	}
+	else if (args->kw_rest_arg) {
+	    struct rb_iseq_param_keyword *keyword = ZALLOC_N(struct rb_iseq_param_keyword, 1);
+	    keyword->rest_start = arg_size++;
+	    body->param.keyword = keyword;
+	    body->param.flags.has_kwrest = TRUE;
+	}
+
+	if (block_id) {
+	    body->param.block_start = arg_size++;
+	    body->param.flags.has_block = TRUE;
+	}
+
+	iseq_calc_param_size(iseq);
+	body->param.size = arg_size;
+
+	if (args->pre_init) { /* m_init */
+	    COMPILE_POPPED(optargs, "init arguments (m)", args->pre_init);
+	}
+	if (args->post_init) { /* p_init */
+	    COMPILE_POPPED(optargs, "init arguments (p)", args->post_init);
+	}
+
+	if (body->type == ISEQ_TYPE_BLOCK) {
+	    if (body->param.flags.has_opt    == FALSE &&
+		body->param.flags.has_post   == FALSE &&
+		body->param.flags.has_rest   == FALSE &&
+		body->param.flags.has_kw     == FALSE &&
+		body->param.flags.has_kwrest == FALSE) {
+
+		if (body->param.lead_num == 1 && last_comma == 0) {
+		    /* {|a|} */
+		    body->param.flags.ambiguous_param0 = TRUE;
+		}
+	    }
+	}
+    }
+
+    return COMPILE_OK;
+}
+
+static int
+iseq_set_local_table(rb_iseq_t *iseq, const ID *tbl)
+{
+    unsigned int size;
+
+    if (tbl) {
+	size = (unsigned int)*tbl;
+	tbl++;
+    }
+    else {
+	size = 0;
+    }
+
+    if (size > 0) {
+	ID *ids = (ID *)ALLOC_N(ID, size);
+	MEMCPY(ids, tbl, ID, size);
+	iseq->body->local_table = ids;
+    }
+    iseq->body->local_table_size = size;
+
+    debugs("iseq_set_local_table: %u\n", iseq->body->local_table_size);
+    return COMPILE_OK;
+}
+
+static int
+cdhash_cmp(VALUE val, VALUE lit)
+{
+    int tval, tlit;
+
+    if (val == lit) {
+        return 0;
+    }
+    else if ((tlit = OBJ_BUILTIN_TYPE(lit)) == -1) {
+        return val != lit;
+    }
+    else if ((tval = OBJ_BUILTIN_TYPE(val)) == -1) {
+        return -1;
+    }
+    else if (tlit != tval) {
+        return -1;
+    }
+    else if (tlit == T_SYMBOL) {
+        return val != lit;
+    }
+    else if (tlit == T_STRING) {
+        return rb_str_hash_cmp(lit, val);
+    }
+    else if (tlit == T_BIGNUM) {
+        long x = FIX2LONG(rb_big_cmp(lit, val));
+
+        /* Given lit and val are both Bignum, x must be -1, 0, 1.
+         * There is no need to call rb_fix2int here. */
+        RUBY_ASSERT((x == 1) || (x == 0) || (x == -1));
+        return (int)x;
+    }
+    else if (tlit == T_FLOAT) {
+        return rb_float_cmp(lit, val);
+    }
+    else {
+        UNREACHABLE_RETURN(-1);
+    }
+}
+
+static st_index_t
+cdhash_hash(VALUE a)
+{
+    switch (OBJ_BUILTIN_TYPE(a)) {
+      case -1:
+      case T_SYMBOL:
+        return (st_index_t)a;
+      case T_STRING:
+        return rb_str_hash(a);
+      case T_BIGNUM:
+        return FIX2LONG(rb_big_hash(a));
+      case T_FLOAT:
+        return rb_dbl_long_hash(RFLOAT_VALUE(a));
+      default:
+        UNREACHABLE_RETURN(0);
+    }
+}
+
+static const struct st_hash_type cdhash_type = {
+    cdhash_cmp,
+    cdhash_hash,
+};
+
+struct cdhash_set_label_struct {
+    VALUE hash;
+    int pos;
+    int len;
+};
+
+static int
+cdhash_set_label_i(VALUE key, VALUE val, void *ptr)
+{
+    struct cdhash_set_label_struct *data = (struct cdhash_set_label_struct *)ptr;
+    LABEL *lobj = (LABEL *)(val & ~1);
+    rb_hash_aset(data->hash, key, INT2FIX(lobj->position - (data->pos+data->len)));
+    return ST_CONTINUE;
+}
+
+
+static inline VALUE
+get_ivar_ic_value(rb_iseq_t *iseq,ID id)
+{
+    VALUE val;
+    struct rb_id_table *tbl = ISEQ_COMPILE_DATA(iseq)->ivar_cache_table;
+    if (tbl) {
+	if (rb_id_table_lookup(tbl,id,&val)) {
+	    return val;
+	}
+    }
+    else {
+	tbl = rb_id_table_create(1);
+	ISEQ_COMPILE_DATA(iseq)->ivar_cache_table = tbl;
+    }
+    val = INT2FIX(iseq->body->is_size++);
+    rb_id_table_insert(tbl,id,val);
+    return val;
+}
+
+#define BADINSN_DUMP(anchor, list, dest) \
+    dump_disasm_list_with_cursor(FIRST_ELEMENT(anchor), list, dest)
+
+#define BADINSN_ERROR \
+    (xfree(generated_iseq), \
+     xfree(insns_info), \
+     BADINSN_DUMP(anchor, list, NULL), \
+     COMPILE_ERROR)
+
+static int
+fix_sp_depth(rb_iseq_t *iseq, LINK_ANCHOR *const anchor)
+{
+    int stack_max = 0, sp = 0, line = 0;
+    LINK_ELEMENT *list;
+
+    for (list = FIRST_ELEMENT(anchor); list; list = list->next) {
+	if (list->type == ISEQ_ELEMENT_LABEL) {
+	    LABEL *lobj = (LABEL *)list;
+	    lobj->set = TRUE;
+	}
+    }
+
+    for (list = FIRST_ELEMENT(anchor); list; list = list->next) {
+	switch (list->type) {
+	  case ISEQ_ELEMENT_INSN:
+	    {
+		int j, len, insn;
+		const char *types;
+		VALUE *operands;
+		INSN *iobj = (INSN *)list;
+
+		/* update sp */
+		sp = calc_sp_depth(sp, iobj);
+		if (sp < 0) {
+		    BADINSN_DUMP(anchor, list, NULL);
+		    COMPILE_ERROR(iseq, iobj->insn_info.line_no,
+				  "argument stack underflow (%d)", sp);
+		    return -1;
+		}
+		if (sp > stack_max) {
+		    stack_max = sp;
+		}
+
+		line = iobj->insn_info.line_no;
+		/* fprintf(stderr, "insn: %-16s, sp: %d\n", insn_name(iobj->insn_id), sp); */
+		operands = iobj->operands;
+		insn = iobj->insn_id;
+		types = insn_op_types(insn);
+		len = insn_len(insn);
+
+		/* operand check */
+		if (iobj->operand_size != len - 1) {
+		    /* printf("operand size miss! (%d, %d)\n", iobj->operand_size, len); */
+		    BADINSN_DUMP(anchor, list, NULL);
+		    COMPILE_ERROR(iseq, iobj->insn_info.line_no,
+				  "operand size miss! (%d for %d)",
+				  iobj->operand_size, len - 1);
+		    return -1;
+		}
+
+		for (j = 0; types[j]; j++) {
+		    if (types[j] == TS_OFFSET) {
+			/* label(destination position) */
+			LABEL *lobj = (LABEL *)operands[j];
+			if (!lobj->set) {
+			    BADINSN_DUMP(anchor, list, NULL);
+			    COMPILE_ERROR(iseq, iobj->insn_info.line_no,
+					  "unknown label: "LABEL_FORMAT, lobj->label_no);
+			    return -1;
+			}
+			if (lobj->sp == -1) {
+			    lobj->sp = sp;
+			}
+		    }
+		}
+		break;
+	    }
+	  case ISEQ_ELEMENT_LABEL:
+	    {
+		LABEL *lobj = (LABEL *)list;
+		if (lobj->sp == -1) {
+		    lobj->sp = sp;
+		}
+		else {
+		    sp = lobj->sp;
+		}
+		break;
+	    }
+	  case ISEQ_ELEMENT_TRACE:
+	    {
+		/* ignore */
+		break;
+	    }
+	  case ISEQ_ELEMENT_ADJUST:
+	    {
+		ADJUST *adjust = (ADJUST *)list;
+		int orig_sp = sp;
+
+		sp = adjust->label ? adjust->label->sp : 0;
+		if (adjust->line_no != -1 && orig_sp - sp < 0) {
+		    BADINSN_DUMP(anchor, list, NULL);
+		    COMPILE_ERROR(iseq, adjust->line_no,
+				  "iseq_set_sequence: adjust bug %d < %d",
+				  orig_sp, sp);
+		    return -1;
+		}
+		break;
+	    }
+	  default:
+	    BADINSN_DUMP(anchor, list, NULL);
+	    COMPILE_ERROR(iseq, line, "unknown list type: %d", list->type);
+	    return -1;
+	}
+    }
+    return stack_max;
+}
+
+static int
+add_insn_info(struct iseq_insn_info_entry *insns_info, unsigned int *positions,
+              int insns_info_index, int code_index, const INSN *iobj)
+{
+    if (insns_info_index == 0 ||
+        insns_info[insns_info_index-1].line_no != iobj->insn_info.line_no ||
+        insns_info[insns_info_index-1].events  != iobj->insn_info.events) {
+        insns_info[insns_info_index].line_no    = iobj->insn_info.line_no;
+        insns_info[insns_info_index].events     = iobj->insn_info.events;
+        positions[insns_info_index]             = code_index;
+        return TRUE;
+    }
+    return FALSE;
+}
+
+static int
+add_adjust_info(struct iseq_insn_info_entry *insns_info, unsigned int *positions,
+                int insns_info_index, int code_index, const ADJUST *adjust)
+{
+    if (insns_info_index > 0 ||
+        insns_info[insns_info_index-1].line_no != adjust->line_no) {
+        insns_info[insns_info_index].line_no    = adjust->line_no;
+        insns_info[insns_info_index].events     = 0;
+        positions[insns_info_index]             = code_index;
+        return TRUE;
+    }
+    return FALSE;
+}
+
+/**
+  ruby insn object list -> raw instruction sequence
+ */
+static int
+iseq_set_sequence(rb_iseq_t *iseq, LINK_ANCHOR *const anchor)
+{
+    struct iseq_insn_info_entry *insns_info;
+    struct rb_iseq_constant_body *const body = iseq->body;
+    unsigned int *positions;
+    LINK_ELEMENT *list;
+    VALUE *generated_iseq;
+    rb_event_flag_t events = 0;
+    long data = 0;
+
+    int insn_num, code_index, insns_info_index, sp = 0;
+    int stack_max = fix_sp_depth(iseq, anchor);
+
+    if (stack_max < 0) return COMPILE_NG;
+
+    /* fix label position */
+    list = FIRST_ELEMENT(anchor);
+    insn_num = code_index = 0;
+    while (list) {
+	switch (list->type) {
+	  case ISEQ_ELEMENT_INSN:
+	    {
+		INSN *iobj = (INSN *)list;
+		/* update sp */
+		sp = calc_sp_depth(sp, iobj);
+		insn_num++;
+                if (ISEQ_COVERAGE(iseq)) {
+                    if (ISEQ_LINE_COVERAGE(iseq) && (events & RUBY_EVENT_COVERAGE_LINE) &&
+                        !(rb_get_coverage_mode() & COVERAGE_TARGET_ONESHOT_LINES)) {
+                        int line = iobj->insn_info.line_no;
+                        if (line >= 1) {
+                            RARRAY_ASET(ISEQ_LINE_COVERAGE(iseq), line - 1, INT2FIX(0));
+                        }
+                    }
+                    if (ISEQ_BRANCH_COVERAGE(iseq) && (events & RUBY_EVENT_COVERAGE_BRANCH)) {
+                        while (RARRAY_LEN(ISEQ_PC2BRANCHINDEX(iseq)) <= code_index) {
+                            rb_ary_push(ISEQ_PC2BRANCHINDEX(iseq), Qnil);
+                        }
+                        RARRAY_ASET(ISEQ_PC2BRANCHINDEX(iseq), code_index, INT2FIX(data));
+                    }
+		}
+                code_index += insn_data_length(iobj);
+		iobj->insn_info.events |= events;
+		events = 0;
+                data = 0;
+		break;
+	    }
+	  case ISEQ_ELEMENT_LABEL:
+	    {
+		LABEL *lobj = (LABEL *)list;
+		lobj->position = code_index;
+		sp = lobj->sp;
+		break;
+	    }
+	  case ISEQ_ELEMENT_TRACE:
+	    {
+		TRACE *trace = (TRACE *)list;
+		events |= trace->event;
+                if (trace->event & RUBY_EVENT_COVERAGE_BRANCH) data = trace->data;
+		break;
+	    }
+	  case ISEQ_ELEMENT_ADJUST:
+	    {
+		ADJUST *adjust = (ADJUST *)list;
+		if (adjust->line_no != -1) {
+		    int orig_sp = sp;
+		    sp = adjust->label ? adjust->label->sp : 0;
+		    if (orig_sp - sp > 0) {
+			if (orig_sp - sp > 1) code_index++; /* 1 operand */
+			code_index++; /* insn */
+			insn_num++;
+		    }
+		}
+		break;
+	    }
+	  default: break;
+	}
+	list = list->next;
+    }
+
+    /* make instruction sequence */
+    generated_iseq = ALLOC_N(VALUE, code_index);
+    insns_info = ALLOC_N(struct iseq_insn_info_entry, insn_num);
+    positions = ALLOC_N(unsigned int, insn_num);
+    body->is_entries = ZALLOC_N(union iseq_inline_storage_entry, body->is_size);
+    body->ci_entries = (struct rb_call_info *)ruby_xmalloc(sizeof(struct rb_call_info) * body->ci_size +
+								 sizeof(struct rb_call_info_with_kwarg) * body->ci_kw_size);
+    MEMZERO(body->ci_entries + body->ci_size, struct rb_call_info_with_kwarg,  body->ci_kw_size); /* need to clear ci_kw entries */
+    body->cc_entries = ZALLOC_N(struct rb_call_cache, body->ci_size + body->ci_kw_size);
+
+    ISEQ_COMPILE_DATA(iseq)->ci_index = ISEQ_COMPILE_DATA(iseq)->ci_kw_index = 0;
+
+    list = FIRST_ELEMENT(anchor);
+    insns_info_index = code_index = sp = 0;
+
+    while (list) {
+	switch (list->type) {
+	  case ISEQ_ELEMENT_INSN:
+	    {
+		int j, len, insn;
+		const char *types;
+		VALUE *operands;
+		INSN *iobj = (INSN *)list;
+
+		/* update sp */
+		sp = calc_sp_depth(sp, iobj);
+		/* fprintf(stderr, "insn: %-16s, sp: %d\n", insn_name(iobj->insn_id), sp); */
+		operands = iobj->operands;
+		insn = iobj->insn_id;
+		generated_iseq[code_index] = insn;
+		types = insn_op_types(insn);
+		len = insn_len(insn);
+
+		for (j = 0; types[j]; j++) {
+		    char type = types[j];
+		    /* printf("--> [%c - (%d-%d)]\n", type, k, j); */
+		    switch (type) {
+		      case TS_OFFSET:
+			{
+			    /* label(destination position) */
+			    LABEL *lobj = (LABEL *)operands[j];
+			    generated_iseq[code_index + 1 + j] = lobj->position - (code_index + len);
+			    break;
+			}
+		      case TS_CDHASH:
+			{
+			    VALUE map = operands[j];
+			    struct cdhash_set_label_struct data;
+                            data.hash = map;
+                            data.pos = code_index;
+                            data.len = len;
+			    rb_hash_foreach(map, cdhash_set_label_i, (VALUE)&data);
+
+			    rb_hash_rehash(map);
+			    freeze_hide_obj(map);
+			    generated_iseq[code_index + 1 + j] = map;
+			    RB_OBJ_WRITTEN(iseq, Qundef, map);
+			    FL_SET(iseq, ISEQ_MARKABLE_ISEQ);
+			    break;
+			}
+		      case TS_LINDEX:
+		      case TS_NUM:	/* ulong */
+			generated_iseq[code_index + 1 + j] = FIX2INT(operands[j]);
+			break;
+		      case TS_VALUE:	/* VALUE */
+		      case TS_ISEQ:	/* iseq */
+			{
+			    VALUE v = operands[j];
+			    generated_iseq[code_index + 1 + j] = v;
+			    /* to mark ruby object */
+			    if (!SPECIAL_CONST_P(v)) {
+				RB_OBJ_WRITTEN(iseq, Qundef, v);
+				FL_SET(iseq, ISEQ_MARKABLE_ISEQ);
+			    }
+			    break;
+			}
+		      case TS_ISE: /* inline storage entry */
+			/* Treated as an IC, but may contain a markable VALUE */
+			FL_SET(iseq, ISEQ_MARKABLE_ISEQ);
+		      case TS_IC: /* inline cache */
+			{
+			    unsigned int ic_index = FIX2UINT(operands[j]);
+			    IC ic = (IC)&body->is_entries[ic_index];
+			    if (UNLIKELY(ic_index >= body->is_size)) {
+                                BADINSN_DUMP(anchor, &iobj->link, 0);
+                                COMPILE_ERROR(iseq, iobj->insn_info.line_no,
+                                              "iseq_set_sequence: ic_index overflow: index: %d, size: %d",
+                                              ic_index, body->is_size);
+			    }
+			    generated_iseq[code_index + 1 + j] = (VALUE)ic;
+			    break;
+			}
+		      case TS_CALLINFO: /* call info */
+			{
+			    struct rb_call_info *base_ci = (struct rb_call_info *)operands[j];
+			    struct rb_call_info *ci;
+
+			    if (base_ci->flag & VM_CALL_KWARG) {
+				struct rb_call_info_with_kwarg *ci_kw_entries = (struct rb_call_info_with_kwarg *)&body->ci_entries[body->ci_size];
+				struct rb_call_info_with_kwarg *ci_kw = &ci_kw_entries[ISEQ_COMPILE_DATA(iseq)->ci_kw_index++];
+				*ci_kw = *((struct rb_call_info_with_kwarg *)base_ci);
+				ci = (struct rb_call_info *)ci_kw;
+				assert(ISEQ_COMPILE_DATA(iseq)->ci_kw_index <= body->ci_kw_size);
+			    }
+			    else {
+				ci = &body->ci_entries[ISEQ_COMPILE_DATA(iseq)->ci_index++];
+				*ci = *base_ci;
+				assert(ISEQ_COMPILE_DATA(iseq)->ci_index <= body->ci_size);
+			    }
+
+			    generated_iseq[code_index + 1 + j] = (VALUE)ci;
+			    break;
+			}
+		      case TS_CALLCACHE:
+			{
+			    struct rb_call_cache *cc = &body->cc_entries[ISEQ_COMPILE_DATA(iseq)->ci_index + ISEQ_COMPILE_DATA(iseq)->ci_kw_index - 1];
+			    generated_iseq[code_index + 1 + j] = (VALUE)cc;
+			    break;
+			}
+		      case TS_ID: /* ID */
+			generated_iseq[code_index + 1 + j] = SYM2ID(operands[j]);
+			break;
+		      case TS_GENTRY:
+			{
+			    struct rb_global_entry *entry =
+				(struct rb_global_entry *)(operands[j] & (~1));
+			    generated_iseq[code_index + 1 + j] = (VALUE)entry;
+			}
+			break;
+		      case TS_FUNCPTR:
+			generated_iseq[code_index + 1 + j] = operands[j];
+			break;
+		      default:
+			BADINSN_ERROR(iseq, iobj->insn_info.line_no,
+				      "unknown operand type: %c", type);
+			return COMPILE_NG;
+		    }
+		}
+		if (add_insn_info(insns_info, positions, insns_info_index, code_index, iobj)) insns_info_index++;
+		code_index += len;
+		break;
+	    }
+	  case ISEQ_ELEMENT_LABEL:
+	    {
+		LABEL *lobj = (LABEL *)list;
+		sp = lobj->sp;
+		break;
+	    }
+	  case ISEQ_ELEMENT_ADJUST:
+	    {
+		ADJUST *adjust = (ADJUST *)list;
+		int orig_sp = sp;
+
+		if (adjust->label) {
+		    sp = adjust->label->sp;
+		}
+		else {
+		    sp = 0;
+		}
+
+		if (adjust->line_no != -1) {
+		    const int diff = orig_sp - sp;
+		    if (diff > 0) {
+			if (add_adjust_info(insns_info, positions, insns_info_index, code_index, adjust)) insns_info_index++;
+		    }
+		    if (diff > 1) {
+			generated_iseq[code_index++] = BIN(adjuststack);
+			generated_iseq[code_index++] = orig_sp - sp;
+		    }
+		    else if (diff == 1) {
+			generated_iseq[code_index++] = BIN(pop);
+		    }
+		    else if (diff < 0) {
+			int label_no = adjust->label ? adjust->label->label_no : -1;
+			xfree(generated_iseq);
+			xfree(insns_info);
+			xfree(positions);
+			debug_list(anchor);
+			COMPILE_ERROR(iseq, adjust->line_no,
+				      "iseq_set_sequence: adjust bug to %d %d < %d",
+				      label_no, orig_sp, sp);
+			return COMPILE_NG;
+		    }
+		}
+		break;
+	    }
+	  default:
+	    /* ignore */
+	    break;
+	}
+	list = list->next;
+    }
+
+    body->iseq_encoded = (void *)generated_iseq;
+    body->iseq_size = code_index;
+    body->stack_max = stack_max;
+
+    /* get rid of memory leak when REALLOC failed */
+    body->insns_info.body = insns_info;
+    body->insns_info.positions = positions;
+
+    REALLOC_N(insns_info, struct iseq_insn_info_entry, insns_info_index);
+    body->insns_info.body = insns_info;
+    REALLOC_N(positions, unsigned int, insns_info_index);
+    body->insns_info.positions = positions;
+    body->insns_info.size = insns_info_index;
+
+    return COMPILE_OK;
+}
+
+static int
+label_get_position(LABEL *lobj)
+{
+    return lobj->position;
+}
+
+static int
+label_get_sp(LABEL *lobj)
+{
+    return lobj->sp;
+}
+
+static int
+iseq_set_exception_table(rb_iseq_t *iseq)
+{
+    const VALUE *tptr, *ptr;
+    unsigned int tlen, i;
+    struct iseq_catch_table_entry *entry;
+
+    tlen = (int)RARRAY_LEN(ISEQ_COMPILE_DATA(iseq)->catch_table_ary);
+    tptr = RARRAY_CONST_PTR_TRANSIENT(ISEQ_COMPILE_DATA(iseq)->catch_table_ary);
+
+    if (tlen > 0) {
+	struct iseq_catch_table *table = xmalloc(iseq_catch_table_bytes(tlen));
+	table->size = tlen;
+
+	for (i = 0; i < table->size; i++) {
+            ptr = RARRAY_CONST_PTR_TRANSIENT(tptr[i]);
+	    entry = &table->entries[i];
+	    entry->type = (enum catch_type)(ptr[0] & 0xffff);
+	    entry->start = label_get_position((LABEL *)(ptr[1] & ~1));
+	    entry->end = label_get_position((LABEL *)(ptr[2] & ~1));
+	    entry->iseq = (rb_iseq_t *)ptr[3];
+	    RB_OBJ_WRITTEN(iseq, Qundef, entry->iseq);
+
+	    /* stack depth */
+	    if (ptr[4]) {
+		LABEL *lobj = (LABEL *)(ptr[4] & ~1);
+		entry->cont = label_get_position(lobj);
+		entry->sp = label_get_sp(lobj);
+
+		/* TODO: Dirty Hack!  Fix me */
+		if (entry->type == CATCH_TYPE_RESCUE ||
+		    entry->type == CATCH_TYPE_BREAK ||
+		    entry->type == CATCH_TYPE_NEXT) {
+		    entry->sp--;
+		}
+	    }
+	    else {
+		entry->cont = 0;
+	    }
+	}
+	iseq->body->catch_table = table;
+	RB_OBJ_WRITE(iseq, &ISEQ_COMPILE_DATA(iseq)->catch_table_ary, 0); /* free */
+    }
+    else {
+	iseq->body->catch_table = NULL;
+    }
+
+    return COMPILE_OK;
+}
+
+/*
+ * set optional argument table
+ *   def foo(a, b=expr1, c=expr2)
+ *   =>
+ *    b:
+ *      expr1
+ *    c:
+ *      expr2
+ */
+static int
+iseq_set_optargs_table(rb_iseq_t *iseq)
+{
+    int i;
+    VALUE *opt_table = (VALUE *)iseq->body->param.opt_table;
+
+    if (iseq->body->param.flags.has_opt) {
+	for (i = 0; i < iseq->body->param.opt_num + 1; i++) {
+	    opt_table[i] = label_get_position((LABEL *)opt_table[i]);
+	}
+    }
+    return COMPILE_OK;
+}
+
+static LINK_ELEMENT *
+get_destination_insn(INSN *iobj)
+{
+    LABEL *lobj = (LABEL *)OPERAND_AT(iobj, 0);
+    LINK_ELEMENT *list;
+    rb_event_flag_t events = 0;
+
+    list = lobj->link.next;
+    while (list) {
+	switch (list->type) {
+	  case ISEQ_ELEMENT_INSN:
+	  case ISEQ_ELEMENT_ADJUST:
+	    goto found;
+	  case ISEQ_ELEMENT_LABEL:
+	    /* ignore */
+	    break;
+	  case ISEQ_ELEMENT_TRACE:
+	    {
+		TRACE *trace = (TRACE *)list;
+		events |= trace->event;
+	    }
+	    break;
+	  default: break;
+	}
+	list = list->next;
+    }
+  found:
+    if (list && IS_INSN(list)) {
+	INSN *iobj = (INSN *)list;
+	iobj->insn_info.events |= events;
+    }
+    return list;
+}
+
+static LINK_ELEMENT *
+get_next_insn(INSN *iobj)
+{
+    LINK_ELEMENT *list = iobj->link.next;
+
+    while (list) {
+	if (IS_INSN(list) || IS_ADJUST(list)) {
+	    return list;
+	}
+	list = list->next;
+    }
+    return 0;
+}
+
+static LINK_ELEMENT *
+get_prev_insn(INSN *iobj)
+{
+    LINK_ELEMENT *list = iobj->link.prev;
+
+    while (list) {
+	if (IS_INSN(list) || IS_ADJUST(list)) {
+	    return list;
+	}
+	list = list->prev;
+    }
+    return 0;
+}
+
+static void
+unref_destination(INSN *iobj, int pos)
+{
+    LABEL *lobj = (LABEL *)OPERAND_AT(iobj, pos);
+    --lobj->refcnt;
+    if (!lobj->refcnt) ELEM_REMOVE(&lobj->link);
+}
+
+static void
+replace_destination(INSN *dobj, INSN *nobj)
+{
+    VALUE n = OPERAND_AT(nobj, 0);
+    LABEL *dl = (LABEL *)OPERAND_AT(dobj, 0);
+    LABEL *nl = (LABEL *)n;
+    --dl->refcnt;
+    ++nl->refcnt;
+    OPERAND_AT(dobj, 0) = n;
+    if (!dl->refcnt) ELEM_REMOVE(&dl->link);
+}
+
+static LABEL*
+find_destination(INSN *i)
+{
+    int pos, len = insn_len(i->insn_id);
+    for (pos = 0; pos < len; ++pos) {
+	if (insn_op_types(i->insn_id)[pos] == TS_OFFSET) {
+	    return (LABEL *)OPERAND_AT(i, pos);
+	}
+    }
+    return 0;
+}
+
+static int
+remove_unreachable_chunk(rb_iseq_t *iseq, LINK_ELEMENT *i)
+{
+    LINK_ELEMENT *first = i, *end;
+    int *unref_counts = 0, nlabels = ISEQ_COMPILE_DATA(iseq)->label_no;
+
+    if (!i) return 0;
+    unref_counts = ALLOCA_N(int, nlabels);
+    MEMZERO(unref_counts, int, nlabels);
+    end = i;
+    do {
+	LABEL *lab;
+	if (IS_INSN(i)) {
+	    if (IS_INSN_ID(i, leave)) {
+		end = i;
+		break;
+	    }
+	    else if ((lab = find_destination((INSN *)i)) != 0) {
+		if (lab->unremovable) break;
+		unref_counts[lab->label_no]++;
+	    }
+	}
+	else if (IS_LABEL(i)) {
+	    lab = (LABEL *)i;
+	    if (lab->unremovable) return 0;
+	    if (lab->refcnt > unref_counts[lab->label_no]) {
+		if (i == first) return 0;
+		break;
+	    }
+	    continue;
+	}
+	else if (IS_TRACE(i)) {
+	    /* do nothing */
+	}
+	else if (IS_ADJUST(i)) {
+	    LABEL *dest = ((ADJUST *)i)->label;
+	    if (dest && dest->unremovable) return 0;
+	}
+	end = i;
+    } while ((i = i->next) != 0);
+    i = first;
+    do {
+	if (IS_INSN(i)) {
+	    struct rb_iseq_constant_body *body = iseq->body;
+	    VALUE insn = INSN_OF(i);
+	    int pos, len = insn_len(insn);
+	    for (pos = 0; pos < len; ++pos) {
+		switch (insn_op_types(insn)[pos]) {
+		  case TS_OFFSET:
+		    unref_destination((INSN *)i, pos);
+		    break;
+		  case TS_CALLINFO:
+		    if (((struct rb_call_info *)OPERAND_AT(i, pos))->flag & VM_CALL_KWARG)
+			--(body->ci_kw_size);
+		    else
+			--(body->ci_size);
+		    break;
+		}
+	    }
+	}
+	ELEM_REMOVE(i);
+    } while ((i != end) && (i = i->next) != 0);
+    return 1;
+}
+
+static int
+iseq_pop_newarray(rb_iseq_t *iseq, INSN *iobj)
+{
+    switch (OPERAND_AT(iobj, 0)) {
+      case INT2FIX(0): /* empty array */
+	ELEM_REMOVE(&iobj->link);
+	return TRUE;
+      case INT2FIX(1): /* single element array */
+	ELEM_REMOVE(&iobj->link);
+	return FALSE;
+      default:
+	iobj->insn_id = BIN(adjuststack);
+	return TRUE;
+    }
+}
+
+static int
+same_debug_pos_p(LINK_ELEMENT *iobj1, LINK_ELEMENT *iobj2)
+{
+    VALUE debug1 = OPERAND_AT(iobj1, 0);
+    VALUE debug2 = OPERAND_AT(iobj2, 0);
+    if (debug1 == debug2) return TRUE;
+    if (!RB_TYPE_P(debug1, T_ARRAY)) return FALSE;
+    if (!RB_TYPE_P(debug2, T_ARRAY)) return FALSE;
+    if (RARRAY_LEN(debug1) != 2) return FALSE;
+    if (RARRAY_LEN(debug2) != 2) return FALSE;
+    if (RARRAY_AREF(debug1, 0) != RARRAY_AREF(debug2, 0)) return FALSE;
+    if (RARRAY_AREF(debug1, 1) != RARRAY_AREF(debug2, 1)) return FALSE;
+    return TRUE;
+}
+
+static int
+is_frozen_putstring(INSN *insn, VALUE *op)
+{
+    if (IS_INSN_ID(insn, putstring)) {
+        *op = OPERAND_AT(insn, 0);
+        return 1;
+    }
+    else if (IS_INSN_ID(insn, putobject)) { /* frozen_string_literal */
+        *op = OPERAND_AT(insn, 0);
+        return RB_TYPE_P(*op, T_STRING);
+    }
+    return 0;
+}
+
+static int
+optimize_checktype(rb_iseq_t *iseq, INSN *iobj)
+{
+    /*
+     *   putobject obj
+     *   dup
+     *   checktype T_XXX
+     *   branchif l1
+     * l2:
+     *   ...
+     * l1:
+     *
+     * => obj is a T_XXX
+     *
+     *   putobject obj (T_XXX)
+     *   jump L1
+     * L1:
+     *
+     * => obj is not a T_XXX
+     *
+     *   putobject obj (T_XXX)
+     *   jump L2
+     * L2:
+     */
+    int line;
+    INSN *niobj, *ciobj, *dup = 0;
+    LABEL *dest = 0;
+    VALUE type;
+
+    switch (INSN_OF(iobj)) {
+      case BIN(putstring):
+	type = INT2FIX(T_STRING);
+	break;
+      case BIN(putnil):
+	type = INT2FIX(T_NIL);
+	break;
+      case BIN(putobject):
+	type = INT2FIX(TYPE(OPERAND_AT(iobj, 0)));
+	break;
+      default: return FALSE;
+    }
+
+    ciobj = (INSN *)get_next_insn(iobj);
+    if (IS_INSN_ID(ciobj, jump)) {
+	ciobj = (INSN *)get_next_insn((INSN*)OPERAND_AT(ciobj, 0));
+    }
+    if (IS_INSN_ID(ciobj, dup)) {
+	ciobj = (INSN *)get_next_insn(dup = ciobj);
+    }
+    if (!ciobj || !IS_INSN_ID(ciobj, checktype)) return FALSE;
+    niobj = (INSN *)get_next_insn(ciobj);
+    if (!niobj) {
+      no_branch:
+	/* TODO: putobject true/false */
+	return FALSE;
+    }
+    switch (INSN_OF(niobj)) {
+      case BIN(branchif):
+	if (OPERAND_AT(ciobj, 0) == type) {
+	    dest = (LABEL *)OPERAND_AT(niobj, 0);
+	}
+	break;
+      case BIN(branchunless):
+	if (OPERAND_AT(ciobj, 0) != type) {
+	    dest = (LABEL *)OPERAND_AT(niobj, 0);
+	}
+	break;
+      default:
+	goto no_branch;
+    }
+    line = ciobj->insn_info.line_no;
+    if (!dest) {
+	if (niobj->link.next && IS_LABEL(niobj->link.next)) {
+	    dest = (LABEL *)niobj->link.next; /* reuse label */
+	}
+	else {
+	    dest = NEW_LABEL(line);
+	    ELEM_INSERT_NEXT(&niobj->link, &dest->link);
+	}
+    }
+    INSERT_AFTER_INSN1(iobj, line, jump, dest);
+    LABEL_REF(dest);
+    if (!dup) INSERT_AFTER_INSN(iobj, line, pop);
+    return TRUE;
+}
+
+static int
+iseq_peephole_optimize(rb_iseq_t *iseq, LINK_ELEMENT *list, const int do_tailcallopt)
+{
+    INSN *const iobj = (INSN *)list;
+
+  again:
+    optimize_checktype(iseq, iobj);
+
+    if (IS_INSN_ID(iobj, jump)) {
+	INSN *niobj, *diobj, *piobj, *dniobj;
+	diobj = (INSN *)get_destination_insn(iobj);
+	niobj = (INSN *)get_next_insn(iobj);
+
+	if (diobj == niobj) {
+	    /*
+	     *   jump LABEL
+	     *  LABEL:
+	     * =>
+	     *   LABEL:
+	     */
+	    unref_destination(iobj, 0);
+	    ELEM_REMOVE(&iobj->link);
+	    return COMPILE_OK;
+	}
+	else if (iobj != diobj && IS_INSN_ID(diobj, jump) &&
+		 OPERAND_AT(iobj, 0) != OPERAND_AT(diobj, 0)) {
+	    /*
+	     *  useless jump elimination:
+	     *     jump LABEL1
+	     *     ...
+	     *   LABEL1:
+	     *     jump LABEL2
+	     *
+	     *   => in this case, first jump instruction should jump to
+	     *      LABEL2 directly
+	     */
+	    replace_destination(iobj, diobj);
+	    remove_unreachable_chunk(iseq, iobj->link.next);
+	    goto again;
+	}
+	else if (dniobj = 0,
+		 IS_INSN_ID(diobj, leave) ||
+		 (diobj->operand_size == 0 &&
+		  (dniobj = (INSN *)get_next_insn(diobj)) != 0 &&
+		  (IS_INSN_ID(dniobj, leave) || (dniobj = 0)))) {
+	    INSN *pop;
+	    /*
+	     *  jump LABEL
+	     *  ...
+	     * LABEL:
+	     *  leave
+	     * =>
+	     *  leave
+	     *  pop
+	     *  ...
+	     * LABEL:
+	     *  leave
+	     */
+	    /* replace */
+	    unref_destination(iobj, 0);
+	    iobj->insn_id = diobj->insn_id;
+	    iobj->operand_size = 0;
+	    iobj->insn_info = diobj->insn_info;
+	    if (dniobj) {
+		dniobj = new_insn_body(iseq, dniobj->insn_info.line_no, BIN(leave), 0);
+		ELEM_INSERT_NEXT(&iobj->link, &dniobj->link);
+	    }
+	    else {
+		dniobj = iobj;
+	    }
+	    /* adjust stack depth */
+	    pop = new_insn_body(iseq, diobj->insn_info.line_no, BIN(pop), 0);
+	    ELEM_INSERT_NEXT(&dniobj->link, &pop->link);
+	    goto again;
+	}
+	else if ((piobj = (INSN *)get_prev_insn(iobj)) != 0 &&
+		 (IS_INSN_ID(piobj, branchif) ||
+		  IS_INSN_ID(piobj, branchunless))) {
+	    INSN *pdiobj = (INSN *)get_destination_insn(piobj);
+	    if (niobj == pdiobj) {
+		int refcnt = IS_LABEL(piobj->link.next) ?
+		    ((LABEL *)piobj->link.next)->refcnt : 0;
+		/*
+		 * useless jump elimination (if/unless destination):
+		 *   if   L1
+		 *   jump L2
+		 * L1:
+		 *   ...
+		 * L2:
+		 *
+		 * ==>
+		 *   unless L2
+		 * L1:
+		 *   ...
+		 * L2:
+		 */
+		piobj->insn_id = (IS_INSN_ID(piobj, branchif))
+		  ? BIN(branchunless) : BIN(branchif);
+		replace_destination(piobj, iobj);
+		if (refcnt <= 1) {
+		    ELEM_REMOVE(&iobj->link);
+		}
+		else {
+		    /* TODO: replace other branch destinations too */
+		}
+		return COMPILE_OK;
+	    }
+	    else if (diobj == pdiobj) {
+		/*
+		 * useless jump elimination (if/unless before jump):
+		 * L1:
+		 *   ...
+		 *   if   L1
+		 *   jump L1
+		 *
+		 * ==>
+		 * L1:
+		 *   ...
+		 *   pop
+		 *   jump L1
+		 */
+		INSN *popiobj = new_insn_core(iseq, iobj->insn_info.line_no,
+					      BIN(pop), 0, 0);
+		ELEM_REPLACE(&piobj->link, &popiobj->link);
+	    }
+	}
+	if (remove_unreachable_chunk(iseq, iobj->link.next)) {
+	    goto again;
+	}
+    }
+
+    /*
+     * putstring "beg"
+     * putstring "end"
+     * newrange excl
+     *
+     * ==>
+     *
+     * putobject "beg".."end"
+     */
+    if (IS_INSN_ID(iobj, checkmatch)) {
+        INSN *range = (INSN *)get_prev_insn(iobj);
+        INSN *beg, *end;
+        VALUE str_beg, str_end;
+
+	if (range && IS_INSN_ID(range, newrange) &&
+                (end = (INSN *)get_prev_insn(range)) != 0 &&
+                is_frozen_putstring(end, &str_end) &&
+                (beg = (INSN *)get_prev_insn(end)) != 0 &&
+                is_frozen_putstring(beg, &str_beg)) {
+	    int excl = FIX2INT(OPERAND_AT(range, 0));
+	    VALUE lit_range = rb_range_new(str_beg, str_end, excl);
+
+	    iseq_add_mark_object_compile_time(iseq, lit_range);
+	    ELEM_REMOVE(&beg->link);
+	    ELEM_REMOVE(&end->link);
+	    range->insn_id = BIN(putobject);
+	    OPERAND_AT(range, 0) = lit_range;
+	}
+    }
+
+    if (IS_INSN_ID(iobj, leave)) {
+	remove_unreachable_chunk(iseq, iobj->link.next);
+    }
+
+    if (IS_INSN_ID(iobj, branchif) ||
+	IS_INSN_ID(iobj, branchnil) ||
+	IS_INSN_ID(iobj, branchunless)) {
+	/*
+	 *   if L1
+	 *   ...
+	 * L1:
+	 *   jump L2
+	 * =>
+	 *   if L2
+	 */
+	INSN *nobj = (INSN *)get_destination_insn(iobj);
+	INSN *pobj = (INSN *)iobj->link.prev;
+	int prev_dup = 0;
+	if (pobj) {
+	    if (!IS_INSN(&pobj->link))
+		pobj = 0;
+	    else if (IS_INSN_ID(pobj, dup))
+		prev_dup = 1;
+	}
+
+	for (;;) {
+	    if (IS_INSN_ID(nobj, jump)) {
+		replace_destination(iobj, nobj);
+	    }
+	    else if (prev_dup && IS_INSN_ID(nobj, dup) &&
+		     !!(nobj = (INSN *)nobj->link.next) &&
+		     /* basic blocks, with no labels in the middle */
+		     nobj->insn_id == iobj->insn_id) {
+		/*
+		 *   dup
+		 *   if L1
+		 *   ...
+		 * L1:
+		 *   dup
+		 *   if L2
+		 * =>
+		 *   dup
+		 *   if L2
+		 *   ...
+		 * L1:
+		 *   dup
+		 *   if L2
+		 */
+		replace_destination(iobj, nobj);
+	    }
+	    else if (pobj) {
+		/*
+		 *   putnil
+		 *   if L1
+		 * =>
+		 *   # nothing
+		 *
+		 *   putobject true
+		 *   if L1
+		 * =>
+		 *   jump L1
+		 *
+		 *   putstring ".."
+		 *   if L1
+		 * =>
+		 *   jump L1
+		 *
+		 *   putstring ".."
+		 *   dup
+		 *   if L1
+		 * =>
+		 *   putstring ".."
+		 *   jump L1
+		 *
+		 */
+		int cond;
+		if (prev_dup && IS_INSN(pobj->link.prev)) {
+		    pobj = (INSN *)pobj->link.prev;
+		}
+		if (IS_INSN_ID(pobj, putobject)) {
+		    cond = (IS_INSN_ID(iobj, branchif) ?
+			    OPERAND_AT(pobj, 0) != Qfalse :
+			    IS_INSN_ID(iobj, branchunless) ?
+			    OPERAND_AT(pobj, 0) == Qfalse :
+			    FALSE);
+		}
+		else if (IS_INSN_ID(pobj, putstring) ||
+			 IS_INSN_ID(pobj, duparray) ||
+			 IS_INSN_ID(pobj, newarray)) {
+		    cond = IS_INSN_ID(iobj, branchif);
+		}
+		else if (IS_INSN_ID(pobj, putnil)) {
+		    cond = !IS_INSN_ID(iobj, branchif);
+		}
+		else break;
+		if (prev_dup || !IS_INSN_ID(pobj, newarray)) {
+		    ELEM_REMOVE(iobj->link.prev);
+		}
+		else if (!iseq_pop_newarray(iseq, pobj)) {
+		    pobj = new_insn_core(iseq, pobj->insn_info.line_no, BIN(pop), 0, NULL);
+                    ELEM_INSERT_PREV(&iobj->link, &pobj->link);
+		}
+		if (cond) {
+		    if (prev_dup) {
+			pobj = new_insn_core(iseq, pobj->insn_info.line_no, BIN(putnil), 0, NULL);
+			ELEM_INSERT_NEXT(&iobj->link, &pobj->link);
+		    }
+		    iobj->insn_id = BIN(jump);
+		    goto again;
+		}
+		else {
+		    unref_destination(iobj, 0);
+		    ELEM_REMOVE(&iobj->link);
+		}
+		break;
+	    }
+	    else break;
+	    nobj = (INSN *)get_destination_insn(nobj);
+	}
+    }
+
+    if (IS_INSN_ID(iobj, pop)) {
+	/*
+	 *  putself / putnil / putobject obj / putstring "..."
+	 *  pop
+	 * =>
+	 *  # do nothing
+	 */
+	LINK_ELEMENT *prev = iobj->link.prev;
+	if (IS_INSN(prev)) {
+	    enum ruby_vminsn_type previ = ((INSN *)prev)->insn_id;
+	    if (previ == BIN(putobject) || previ == BIN(putnil) ||
+		previ == BIN(putself) || previ == BIN(putstring) ||
+		previ == BIN(dup) ||
+		previ == BIN(getlocal) ||
+		previ == BIN(getblockparam) ||
+		previ == BIN(getblockparamproxy) ||
+		/* getinstancevariable may issue a warning */
+		previ == BIN(duparray)) {
+		/* just push operand or static value and pop soon, no
+		 * side effects */
+		ELEM_REMOVE(prev);
+		ELEM_REMOVE(&iobj->link);
+	    }
+	    else if (previ == BIN(newarray) && iseq_pop_newarray(iseq, (INSN*)prev)) {
+		ELEM_REMOVE(&iobj->link);
+	    }
+	    else if (previ == BIN(concatarray)) {
+		INSN *piobj = (INSN *)prev;
+		INSERT_BEFORE_INSN1(piobj, piobj->insn_info.line_no, splatarray, Qfalse);
+		INSN_OF(piobj) = BIN(pop);
+	    }
+	    else if (previ == BIN(concatstrings)) {
+		if (OPERAND_AT(prev, 0) == INT2FIX(1)) {
+		    ELEM_REMOVE(prev);
+		}
+		else {
+		    ELEM_REMOVE(&iobj->link);
+		    INSN_OF(prev) = BIN(adjuststack);
+		}
+	    }
+	}
+    }
+
+    if (IS_INSN_ID(iobj, newarray) ||
+	IS_INSN_ID(iobj, duparray) ||
+	IS_INSN_ID(iobj, expandarray) ||
+	IS_INSN_ID(iobj, concatarray) ||
+	IS_INSN_ID(iobj, splatarray) ||
+	0) {
+	/*
+	 *  newarray N
+	 *  splatarray
+	 * =>
+	 *  newarray N
+	 * newarray always puts an array
+	 */
+	LINK_ELEMENT *next = iobj->link.next;
+	if (IS_INSN(next) && IS_INSN_ID(next, splatarray)) {
+	    /* remove splatarray following always-array insn */
+	    ELEM_REMOVE(next);
+	}
+    }
+
+    if (IS_INSN_ID(iobj, tostring)) {
+	LINK_ELEMENT *next = iobj->link.next;
+	/*
+	 *  tostring
+	 *  concatstrings 1
+	 * =>
+	 *  tostring
+	 */
+	if (IS_INSN(next) && IS_INSN_ID(next, concatstrings) &&
+	    OPERAND_AT(next, 0) == INT2FIX(1)) {
+	    ELEM_REMOVE(next);
+	}
+    }
+
+    if (IS_INSN_ID(iobj, putstring) ||
+	(IS_INSN_ID(iobj, putobject) && RB_TYPE_P(OPERAND_AT(iobj, 0), T_STRING))) {
+	/*
+	 *  putstring ""
+	 *  concatstrings N
+	 * =>
+	 *  concatstrings N-1
+	 */
+	if (IS_NEXT_INSN_ID(&iobj->link, concatstrings) &&
+	    RSTRING_LEN(OPERAND_AT(iobj, 0)) == 0) {
+	    INSN *next = (INSN *)iobj->link.next;
+	    if ((OPERAND_AT(next, 0) = FIXNUM_INC(OPERAND_AT(next, 0), -1)) == INT2FIX(1)) {
+		ELEM_REMOVE(&next->link);
+	    }
+	    ELEM_REMOVE(&iobj->link);
+	}
+    }
+
+    if (IS_INSN_ID(iobj, concatstrings)) {
+	/*
+	 *  concatstrings N
+	 *  concatstrings M
+	 * =>
+	 *  concatstrings N+M-1
+	 */
+	LINK_ELEMENT *next = iobj->link.next, *freeze = 0;
+	INSN *jump = 0;
+	if (IS_INSN(next) && IS_INSN_ID(next, freezestring))
+	    next = (freeze = next)->next;
+	if (IS_INSN(next) && IS_INSN_ID(next, jump))
+	    next = get_destination_insn(jump = (INSN *)next);
+	if (IS_INSN(next) && IS_INSN_ID(next, concatstrings)) {
+	    int n = FIX2INT(OPERAND_AT(iobj, 0)) + FIX2INT(OPERAND_AT(next, 0)) - 1;
+	    OPERAND_AT(iobj, 0) = INT2FIX(n);
+	    if (jump) {
+		LABEL *label = ((LABEL *)OPERAND_AT(jump, 0));
+		if (!--label->refcnt) {
+		    ELEM_REMOVE(&label->link);
+		}
+		else {
+		    label = NEW_LABEL(0);
+		    OPERAND_AT(jump, 0) = (VALUE)label;
+		}
+		label->refcnt++;
+		if (freeze && IS_NEXT_INSN_ID(next, freezestring)) {
+		    if (same_debug_pos_p(freeze, next->next)) {
+			ELEM_REMOVE(freeze);
+		    }
+		    else {
+			next = next->next;
+		    }
+		}
+		ELEM_INSERT_NEXT(next, &label->link);
+		CHECK(iseq_peephole_optimize(iseq, get_next_insn(jump), do_tailcallopt));
+	    }
+	    else {
+		if (freeze) ELEM_REMOVE(freeze);
+		ELEM_REMOVE(next);
+	    }
+	}
+    }
+
+    if (IS_INSN_ID(iobj, freezestring) &&
+	NIL_P(OPERAND_AT(iobj, 0)) &&
+	IS_NEXT_INSN_ID(&iobj->link, send)) {
+	INSN *niobj = (INSN *)iobj->link.next;
+	struct rb_call_info *ci = (struct rb_call_info *)OPERAND_AT(niobj, 0);
+	/*
+	 *  freezestring nil # no debug_info
+	 *  send <:+@, 0, ARG_SIMPLE>  # :-@, too
+	 * =>
+	 *  send <:+@, 0, ARG_SIMPLE>  # :-@, too
+	 */
+	if ((ci->mid == idUPlus || ci->mid == idUMinus) &&
+	    (ci->flag & VM_CALL_ARGS_SIMPLE) &&
+	    ci->orig_argc == 0) {
+	    ELEM_REMOVE(list);
+	    return COMPILE_OK;
+	}
+    }
+
+    if (do_tailcallopt &&
+	(IS_INSN_ID(iobj, send) ||
+	 IS_INSN_ID(iobj, opt_aref_with) ||
+	 IS_INSN_ID(iobj, opt_aset_with) ||
+	 IS_INSN_ID(iobj, invokesuper))) {
+	/*
+	 *  send ...
+	 *  leave
+	 * =>
+	 *  send ..., ... | VM_CALL_TAILCALL, ...
+	 *  leave # unreachable
+	 */
+	INSN *piobj = NULL;
+	if (iobj->link.next) {
+	    LINK_ELEMENT *next = iobj->link.next;
+	    do {
+		if (!IS_INSN(next)) {
+		    next = next->next;
+		    continue;
+		}
+		switch (INSN_OF(next)) {
+		  case BIN(nop):
+		    next = next->next;
+		    break;
+		  case BIN(jump):
+		    /* if cond
+		     *   return tailcall
+		     * end
+		     */
+		    next = get_destination_insn((INSN *)next);
+		    break;
+		  case BIN(leave):
+		    piobj = iobj;
+		  default:
+		    next = NULL;
+		    break;
+		}
+	    } while (next);
+	}
+
+	if (piobj) {
+	    struct rb_call_info *ci = (struct rb_call_info *)piobj->operands[0];
+	    if (IS_INSN_ID(piobj, send) || IS_INSN_ID(piobj, invokesuper)) {
+		if (piobj->operands[2] == 0) { /* no blockiseq */
+		    ci->flag |= VM_CALL_TAILCALL;
+		}
+	    }
+	    else {
+		ci->flag |= VM_CALL_TAILCALL;
+	    }
+	}
+    }
+
+    if (IS_INSN_ID(iobj, dup)) {
+	if (IS_NEXT_INSN_ID(&iobj->link, setlocal)) {
+	    LINK_ELEMENT *set1 = iobj->link.next, *set2 = NULL;
+	    if (IS_NEXT_INSN_ID(set1, setlocal)) {
+		set2 = set1->next;
+		if (OPERAND_AT(set1, 0) == OPERAND_AT(set2, 0) &&
+		    OPERAND_AT(set1, 1) == OPERAND_AT(set2, 1)) {
+		    ELEM_REMOVE(set1);
+		    ELEM_REMOVE(&iobj->link);
+		}
+	    }
+	    else if (IS_NEXT_INSN_ID(set1, dup) &&
+		     IS_NEXT_INSN_ID(set1->next, setlocal)) {
+		set2 = set1->next->next;
+		if (OPERAND_AT(set1, 0) == OPERAND_AT(set2, 0) &&
+		    OPERAND_AT(set1, 1) == OPERAND_AT(set2, 1)) {
+		    ELEM_REMOVE(set1->next);
+		    ELEM_REMOVE(set2);
+		}
+	    }
+	}
+    }
+
+    if (IS_INSN_ID(iobj, getlocal)) {
+	LINK_ELEMENT *niobj = &iobj->link;
+	if (IS_NEXT_INSN_ID(niobj, dup)) {
+	    niobj = niobj->next;
+	}
+	if (IS_NEXT_INSN_ID(niobj, setlocal)) {
+	    LINK_ELEMENT *set1 = niobj->next;
+	    if (OPERAND_AT(iobj, 0) == OPERAND_AT(set1, 0) &&
+		OPERAND_AT(iobj, 1) == OPERAND_AT(set1, 1)) {
+		ELEM_REMOVE(set1);
+		ELEM_REMOVE(niobj);
+	    }
+	}
+    }
+
+    return COMPILE_OK;
+}
+
+static int
+insn_set_specialized_instruction(rb_iseq_t *iseq, INSN *iobj, int insn_id)
+{
+    iobj->insn_id = insn_id;
+    iobj->operand_size = insn_len(insn_id) - 1;
+
+    if (insn_id == BIN(opt_neq)) {
+	VALUE *old_operands = iobj->operands;
+	iobj->operand_size = 4;
+	iobj->operands = (VALUE *)compile_data_alloc(iseq, iobj->operand_size * sizeof(VALUE));
+	iobj->operands[0] = (VALUE)new_callinfo(iseq, idEq, 1, 0, NULL, FALSE);
+	iobj->operands[1] = Qfalse; /* CALL_CACHE */
+	iobj->operands[2] = old_operands[0];
+	iobj->operands[3] = Qfalse; /* CALL_CACHE */
+    }
+
+    return COMPILE_OK;
+}
+
+static int
+iseq_specialized_instruction(rb_iseq_t *iseq, INSN *iobj)
+{
+    if (IS_INSN_ID(iobj, newarray) && iobj->link.next &&
+	IS_INSN(iobj->link.next)) {
+	/*
+	 *   [a, b, ...].max/min -> a, b, c, opt_newarray_max/min
+	 */
+	INSN *niobj = (INSN *)iobj->link.next;
+	if (IS_INSN_ID(niobj, send)) {
+	    struct rb_call_info *ci = (struct rb_call_info *)OPERAND_AT(niobj, 0);
+	    if ((ci->flag & VM_CALL_ARGS_SIMPLE) && ci->orig_argc == 0) {
+		switch (ci->mid) {
+		  case idMax:
+		    iobj->insn_id = BIN(opt_newarray_max);
+		    ELEM_REMOVE(&niobj->link);
+		    return COMPILE_OK;
+		  case idMin:
+		    iobj->insn_id = BIN(opt_newarray_min);
+		    ELEM_REMOVE(&niobj->link);
+		    return COMPILE_OK;
+		}
+	    }
+	}
+    }
+
+    if (IS_INSN_ID(iobj, send)) {
+	struct rb_call_info *ci = (struct rb_call_info *)OPERAND_AT(iobj, 0);
+	const rb_iseq_t *blockiseq = (rb_iseq_t *)OPERAND_AT(iobj, 2);
+
+#define SP_INSN(opt) insn_set_specialized_instruction(iseq, iobj, BIN(opt_##opt))
+	if (ci->flag & VM_CALL_ARGS_SIMPLE) {
+	    switch (ci->orig_argc) {
+	      case 0:
+		switch (ci->mid) {
+		  case idLength: SP_INSN(length); return COMPILE_OK;
+		  case idSize:	 SP_INSN(size);	  return COMPILE_OK;
+		  case idEmptyP: SP_INSN(empty_p);return COMPILE_OK;
+		  case idSucc:	 SP_INSN(succ);	  return COMPILE_OK;
+		  case idNot:	 SP_INSN(not);	  return COMPILE_OK;
+		}
+		break;
+	      case 1:
+		switch (ci->mid) {
+		  case idPLUS:	 SP_INSN(plus);	  return COMPILE_OK;
+		  case idMINUS:	 SP_INSN(minus);  return COMPILE_OK;
+		  case idMULT:	 SP_INSN(mult);	  return COMPILE_OK;
+		  case idDIV:	 SP_INSN(div);	  return COMPILE_OK;
+		  case idMOD:	 SP_INSN(mod);	  return COMPILE_OK;
+		  case idEq:	 SP_INSN(eq);	  return COMPILE_OK;
+		  case idNeq:	 SP_INSN(neq);	  return COMPILE_OK;
+		  case idLT:	 SP_INSN(lt);	  return COMPILE_OK;
+		  case idLE:	 SP_INSN(le);	  return COMPILE_OK;
+		  case idGT:	 SP_INSN(gt);	  return COMPILE_OK;
+		  case idGE:	 SP_INSN(ge);	  return COMPILE_OK;
+		  case idLTLT:	 SP_INSN(ltlt);	  return COMPILE_OK;
+		  case idAREF:	 SP_INSN(aref);	  return COMPILE_OK;
+                  case idAnd:    SP_INSN(and);    return COMPILE_OK;
+                  case idOr:     SP_INSN(or);    return COMPILE_OK;
+		}
+		break;
+	      case 2:
+		switch (ci->mid) {
+		  case idASET:	 SP_INSN(aset);	  return COMPILE_OK;
+		}
+		break;
+	    }
+	}
+
+	if ((ci->flag & VM_CALL_ARGS_BLOCKARG) == 0 && blockiseq == NULL) {
+	    iobj->insn_id = BIN(opt_send_without_block);
+	    iobj->operand_size = insn_len(iobj->insn_id) - 1;
+	}
+    }
+#undef SP_INSN
+
+    return COMPILE_OK;
+}
+
+static inline int
+tailcallable_p(rb_iseq_t *iseq)
+{
+    switch (iseq->body->type) {
+      case ISEQ_TYPE_TOP:
+      case ISEQ_TYPE_EVAL:
+      case ISEQ_TYPE_MAIN:
+	/* not tail callable because cfp will be over popped */
+      case ISEQ_TYPE_RESCUE:
+      case ISEQ_TYPE_ENSURE:
+	/* rescue block can't tail call because of errinfo */
+	return FALSE;
+      default:
+	return TRUE;
+    }
+}
+
+static int
+iseq_optimize(rb_iseq_t *iseq, LINK_ANCHOR *const anchor)
+{
+    LINK_ELEMENT *list;
+    const int do_peepholeopt = ISEQ_COMPILE_DATA(iseq)->option->peephole_optimization;
+    const int do_tailcallopt = tailcallable_p(iseq) &&
+	ISEQ_COMPILE_DATA(iseq)->option->tailcall_optimization;
+    const int do_si = ISEQ_COMPILE_DATA(iseq)->option->specialized_instruction;
+    const int do_ou = ISEQ_COMPILE_DATA(iseq)->option->operands_unification;
+    int rescue_level = 0;
+    int tailcallopt = do_tailcallopt;
+
+    list = FIRST_ELEMENT(anchor);
+
+    while (list) {
+	if (IS_INSN(list)) {
+	    if (do_peepholeopt) {
+		iseq_peephole_optimize(iseq, list, tailcallopt);
+	    }
+	    if (do_si) {
+		iseq_specialized_instruction(iseq, (INSN *)list);
+	    }
+	    if (do_ou) {
+		insn_operands_unification((INSN *)list);
+	    }
+	}
+	if (IS_LABEL(list)) {
+	    switch (((LABEL *)list)->rescued) {
+	      case LABEL_RESCUE_BEG:
+		rescue_level++;
+		tailcallopt = FALSE;
+		break;
+	      case LABEL_RESCUE_END:
+		if (!--rescue_level) tailcallopt = do_tailcallopt;
+		break;
+	    }
+	}
+	list = list->next;
+    }
+    return COMPILE_OK;
+}
+
+#if OPT_INSTRUCTIONS_UNIFICATION
+static INSN *
+new_unified_insn(rb_iseq_t *iseq,
+		 int insn_id, int size, LINK_ELEMENT *seq_list)
+{
+    INSN *iobj = 0;
+    LINK_ELEMENT *list = seq_list;
+    int i, argc = 0;
+    VALUE *operands = 0, *ptr = 0;
+
+
+    /* count argc */
+    for (i = 0; i < size; i++) {
+	iobj = (INSN *)list;
+	argc += iobj->operand_size;
+	list = list->next;
+    }
+
+    if (argc > 0) {
+	ptr = operands =
+	    (VALUE *)compile_data_alloc(iseq, sizeof(VALUE) * argc);
+    }
+
+    /* copy operands */
+    list = seq_list;
+    for (i = 0; i < size; i++) {
+	iobj = (INSN *)list;
+	MEMCPY(ptr, iobj->operands, VALUE, iobj->operand_size);
+	ptr += iobj->operand_size;
+	list = list->next;
+    }
+
+    return new_insn_core(iseq, iobj->insn_info.line_no, insn_id, argc, operands);
+}
+#endif
+
+/*
+ * This scheme can get more performance if do this optimize with
+ * label address resolving.
+ * It's future work (if compile time was bottle neck).
+ */
+static int
+iseq_insns_unification(rb_iseq_t *iseq, LINK_ANCHOR *const anchor)
+{
+#if OPT_INSTRUCTIONS_UNIFICATION
+    LINK_ELEMENT *list;
+    INSN *iobj, *niobj;
+    int id, k;
+    intptr_t j;
+
+    list = FIRST_ELEMENT(anchor);
+    while (list) {
+	if (IS_INSN(list)) {
+	    iobj = (INSN *)list;
+	    id = iobj->insn_id;
+	    if (unified_insns_data[id] != 0) {
+		const int *const *entry = unified_insns_data[id];
+		for (j = 1; j < (intptr_t)entry[0]; j++) {
+		    const int *unified = entry[j];
+		    LINK_ELEMENT *li = list->next;
+		    for (k = 2; k < unified[1]; k++) {
+			if (!IS_INSN(li) ||
+			    ((INSN *)li)->insn_id != unified[k]) {
+			    goto miss;
+			}
+			li = li->next;
+		    }
+		    /* matched */
+		    niobj =
+			new_unified_insn(iseq, unified[0], unified[1] - 1,
+					 list);
+
+		    /* insert to list */
+		    niobj->link.prev = (LINK_ELEMENT *)iobj->link.prev;
+		    niobj->link.next = li;
+		    if (li) {
+			li->prev = (LINK_ELEMENT *)niobj;
+		    }
+
+		    list->prev->next = (LINK_ELEMENT *)niobj;
+		    list = (LINK_ELEMENT *)niobj;
+		    break;
+		  miss:;
+		}
+	    }
+	}
+	list = list->next;
+    }
+#endif
+    return COMPILE_OK;
+}
+
+#if OPT_STACK_CACHING
+
+#define SC_INSN(insn, stat) sc_insn_info[(insn)][(stat)]
+#define SC_NEXT(insn)       sc_insn_next[(insn)]
+
+#include "opt_sc.inc"
+
+static int
+insn_set_sc_state(rb_iseq_t *iseq, const LINK_ELEMENT *anchor, INSN *iobj, int state)
+{
+    int nstate;
+    int insn_id;
+
+    insn_id = iobj->insn_id;
+    iobj->insn_id = SC_INSN(insn_id, state);
+    nstate = SC_NEXT(iobj->insn_id);
+
+    if (insn_id == BIN(jump) ||
+	insn_id == BIN(branchif) || insn_id == BIN(branchunless)) {
+	LABEL *lobj = (LABEL *)OPERAND_AT(iobj, 0);
+
+	if (lobj->sc_state != 0) {
+	    if (lobj->sc_state != nstate) {
+		BADINSN_DUMP(anchor, iobj, lobj);
+		COMPILE_ERROR(iseq, iobj->insn_info.line_no,
+			      "insn_set_sc_state error: %d at "LABEL_FORMAT
+			      ", %d expected\n",
+			      lobj->sc_state, lobj->label_no, nstate);
+		return COMPILE_NG;
+	    }
+	}
+	else {
+	    lobj->sc_state = nstate;
+	}
+	if (insn_id == BIN(jump)) {
+	    nstate = SCS_XX;
+	}
+    }
+    else if (insn_id == BIN(leave)) {
+	nstate = SCS_XX;
+    }
+
+    return nstate;
+}
+
+static int
+label_set_sc_state(LABEL *lobj, int state)
+{
+    if (lobj->sc_state != 0) {
+	if (lobj->sc_state != state) {
+	    state = lobj->sc_state;
+	}
+    }
+    else {
+	lobj->sc_state = state;
+    }
+
+    return state;
+}
+
+
+#endif
+
+static int
+iseq_set_sequence_stackcaching(rb_iseq_t *iseq, LINK_ANCHOR *const anchor)
+{
+#if OPT_STACK_CACHING
+    LINK_ELEMENT *list;
+    int state, insn_id;
+
+    /* initialize */
+    state = SCS_XX;
+    list = FIRST_ELEMENT(anchor);
+    /* dump_disasm_list(list); */
+
+    /* for each list element */
+    while (list) {
+      redo_point:
+	switch (list->type) {
+	  case ISEQ_ELEMENT_INSN:
+	    {
+		INSN *iobj = (INSN *)list;
+		insn_id = iobj->insn_id;
+
+		/* dump_disasm_list(list); */
+
+		switch (insn_id) {
+		  case BIN(nop):
+		    {
+			/* exception merge point */
+			if (state != SCS_AX) {
+			    INSN *rpobj =
+				new_insn_body(iseq, 0, BIN(reput), 0);
+
+			    /* replace this insn */
+			    ELEM_REPLACE(list, (LINK_ELEMENT *)rpobj);
+			    list = (LINK_ELEMENT *)rpobj;
+			    goto redo_point;
+			}
+			break;
+		    }
+		  case BIN(swap):
+		    {
+			if (state == SCS_AB || state == SCS_BA) {
+			    state = (state == SCS_AB ? SCS_BA : SCS_AB);
+
+			    ELEM_REMOVE(list);
+			    list = list->next;
+			    goto redo_point;
+			}
+			break;
+		    }
+		  case BIN(pop):
+		    {
+			switch (state) {
+			  case SCS_AX:
+			  case SCS_BX:
+			    state = SCS_XX;
+			    break;
+			  case SCS_AB:
+			    state = SCS_AX;
+			    break;
+			  case SCS_BA:
+			    state = SCS_BX;
+			    break;
+			  case SCS_XX:
+			    goto normal_insn;
+			  default:
+			    COMPILE_ERROR(iseq, iobj->insn_info.line_no,
+					  "unreachable");
+			    return COMPILE_NG;
+			}
+			/* remove useless pop */
+			ELEM_REMOVE(list);
+			list = list->next;
+			goto redo_point;
+		    }
+		  default:;
+		    /* none */
+		}		/* end of switch */
+	      normal_insn:
+		state = insn_set_sc_state(iseq, anchor, iobj, state);
+		break;
+	    }
+	  case ISEQ_ELEMENT_LABEL:
+	    {
+		LABEL *lobj;
+		lobj = (LABEL *)list;
+
+		state = label_set_sc_state(lobj, state);
+	    }
+	  default:
+	    break;
+	}
+	list = list->next;
+    }
+#endif
+    return COMPILE_OK;
+}
+
+static int
+all_string_result_p(const NODE *node)
+{
+    if (!node) return FALSE;
+    switch (nd_type(node)) {
+      case NODE_STR: case NODE_DSTR:
+	return TRUE;
+      case NODE_IF: case NODE_UNLESS:
+	if (!node->nd_body || !node->nd_else) return FALSE;
+	if (all_string_result_p(node->nd_body))
+	    return all_string_result_p(node->nd_else);
+	return FALSE;
+      case NODE_AND: case NODE_OR:
+	if (!node->nd_2nd)
+	    return all_string_result_p(node->nd_1st);
+	if (!all_string_result_p(node->nd_1st))
+	    return FALSE;
+	return all_string_result_p(node->nd_2nd);
+      default:
+	return FALSE;
+    }
+}
+
+static int
+compile_dstr_fragments(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int *cntp)
+{
+    const NODE *list = node->nd_next;
+    VALUE lit = node->nd_lit;
+    LINK_ELEMENT *first_lit = 0;
+    int cnt = 0;
+
+    debugp_param("nd_lit", lit);
+    if (!NIL_P(lit)) {
+	cnt++;
+	if (!RB_TYPE_P(lit, T_STRING)) {
+	    COMPILE_ERROR(ERROR_ARGS "dstr: must be string: %s",
+			  rb_builtin_type_name(TYPE(lit)));
+	    return COMPILE_NG;
+	}
+	lit = freeze_literal(iseq, lit);
+	ADD_INSN1(ret, nd_line(node), putobject, lit);
+	if (RSTRING_LEN(lit) == 0) first_lit = LAST_ELEMENT(ret);
+    }
+
+    while (list) {
+	const NODE *const head = list->nd_head;
+	if (nd_type(head) == NODE_STR) {
+	    lit = freeze_literal(iseq, head->nd_lit);
+	    ADD_INSN1(ret, nd_line(head), putobject, lit);
+	    lit = Qnil;
+	}
+	else {
+	    CHECK(COMPILE(ret, "each string", head));
+	}
+	cnt++;
+	list = list->nd_next;
+    }
+    if (NIL_P(lit) && first_lit) {
+	ELEM_REMOVE(first_lit);
+	--cnt;
+    }
+    *cntp = cnt;
+
+    return COMPILE_OK;
+}
+
+static int
+compile_dstr(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node)
+{
+    int cnt;
+    CHECK(compile_dstr_fragments(iseq, ret, node, &cnt));
+    ADD_INSN1(ret, nd_line(node), concatstrings, INT2FIX(cnt));
+    return COMPILE_OK;
+}
+
+static int
+compile_dregx(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node)
+{
+    int cnt;
+    CHECK(compile_dstr_fragments(iseq, ret, node, &cnt));
+    ADD_INSN2(ret, nd_line(node), toregexp, INT2FIX(node->nd_cflag), INT2FIX(cnt));
+    return COMPILE_OK;
+}
+
+static int
+compile_flip_flop(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int again,
+		  LABEL *then_label, LABEL *else_label)
+{
+    const int line = nd_line(node);
+    LABEL *lend = NEW_LABEL(line);
+    rb_num_t cnt = ISEQ_FLIP_CNT_INCREMENT(iseq->body->local_iseq)
+	+ VM_SVAR_FLIPFLOP_START;
+    VALUE key = INT2FIX(cnt);
+
+    ADD_INSN2(ret, line, getspecial, key, INT2FIX(0));
+    ADD_INSNL(ret, line, branchif, lend);
+
+    /* *flip == 0 */
+    CHECK(COMPILE(ret, "flip2 beg", node->nd_beg));
+    ADD_INSNL(ret, line, branchunless, else_label);
+    ADD_INSN1(ret, line, putobject, Qtrue);
+    ADD_INSN1(ret, line, setspecial, key);
+    if (!again) {
+	ADD_INSNL(ret, line, jump, then_label);
+    }
+
+    /* *flip == 1 */
+    ADD_LABEL(ret, lend);
+    CHECK(COMPILE(ret, "flip2 end", node->nd_end));
+    ADD_INSNL(ret, line, branchunless, then_label);
+    ADD_INSN1(ret, line, putobject, Qfalse);
+    ADD_INSN1(ret, line, setspecial, key);
+    ADD_INSNL(ret, line, jump, then_label);
+
+    return COMPILE_OK;
+}
+
+static int
+compile_branch_condition(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *cond,
+			 LABEL *then_label, LABEL *else_label)
+{
+  again:
+    switch (nd_type(cond)) {
+      case NODE_AND:
+	{
+	    LABEL *label = NEW_LABEL(nd_line(cond));
+	    CHECK(compile_branch_condition(iseq, ret, cond->nd_1st, label,
+					   else_label));
+	    if (!label->refcnt) break;
+	    ADD_LABEL(ret, label);
+	    cond = cond->nd_2nd;
+	    goto again;
+	}
+      case NODE_OR:
+	{
+	    LABEL *label = NEW_LABEL(nd_line(cond));
+	    CHECK(compile_branch_condition(iseq, ret, cond->nd_1st, then_label,
+					   label));
+	    if (!label->refcnt) break;
+	    ADD_LABEL(ret, label);
+	    cond = cond->nd_2nd;
+	    goto again;
+	}
+      case NODE_LIT:		/* NODE_LIT is always true */
+      case NODE_TRUE:
+      case NODE_STR:
+      case NODE_ZARRAY:
+      case NODE_LAMBDA:
+	/* printf("useless condition eliminate (%s)\n",  ruby_node_name(nd_type(cond))); */
+	ADD_INSNL(ret, nd_line(cond), jump, then_label);
+	break;
+      case NODE_FALSE:
+      case NODE_NIL:
+	/* printf("useless condition eliminate (%s)\n", ruby_node_name(nd_type(cond))); */
+	ADD_INSNL(ret, nd_line(cond), jump, else_label);
+	break;
+      case NODE_ARRAY:
+      case NODE_ARGSCAT:
+      case NODE_DREGX:
+      case NODE_DSTR:
+	CHECK(COMPILE_POPPED(ret, "branch condition", cond));
+	ADD_INSNL(ret, nd_line(cond), jump, then_label);
+	break;
+      case NODE_FLIP2:
+	CHECK(compile_flip_flop(iseq, ret, cond, TRUE, then_label, else_label));
+	break;
+      case NODE_FLIP3:
+	CHECK(compile_flip_flop(iseq, ret, cond, FALSE, then_label, else_label));
+	break;
+      case NODE_DEFINED:
+	CHECK(compile_defined_expr(iseq, ret, cond, Qfalse));
+	goto branch;
+      default:
+	CHECK(COMPILE(ret, "branch condition", cond));
+      branch:
+	ADD_INSNL(ret, nd_line(cond), branchunless, else_label);
+	ADD_INSNL(ret, nd_line(cond), jump, then_label);
+	break;
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_array_keyword_arg(rb_iseq_t *iseq, LINK_ANCHOR *const ret,
+			  const NODE *const root_node,
+			  struct rb_call_info_kw_arg **const kw_arg_ptr,
+			  unsigned int *flag)
+{
+    if (kw_arg_ptr == NULL) return FALSE;
+
+    if (nd_type(root_node) == NODE_HASH && root_node->nd_head && nd_type(root_node->nd_head) == NODE_ARRAY) {
+	const NODE *node = root_node->nd_head;
+
+	while (node) {
+	    const NODE *key_node = node->nd_head;
+
+	    assert(nd_type(node) == NODE_ARRAY);
+	    if (!key_node) {
+		if (flag && !root_node->nd_alen) *flag |= VM_CALL_KW_SPLAT;
+		return FALSE;
+	    }
+	    else if (nd_type(key_node) == NODE_LIT && RB_TYPE_P(key_node->nd_lit, T_SYMBOL)) {
+		/* can be keywords */
+	    }
+	    else {
+		return FALSE;
+	    }
+	    node = node->nd_next; /* skip value node */
+	    node = node->nd_next;
+	}
+
+	/* may be keywords */
+	node = root_node->nd_head;
+	{
+	    int len = (int)node->nd_alen / 2;
+	    struct rb_call_info_kw_arg *kw_arg  = (struct rb_call_info_kw_arg *)ruby_xmalloc(sizeof(struct rb_call_info_kw_arg) + sizeof(VALUE) * (len - 1));
+	    VALUE *keywords = kw_arg->keywords;
+	    int i = 0;
+	    kw_arg->keyword_len = len;
+
+	    *kw_arg_ptr = kw_arg;
+
+	    for (i=0; node != NULL; i++, node = node->nd_next->nd_next) {
+		const NODE *key_node = node->nd_head;
+		const NODE *val_node = node->nd_next->nd_head;
+		keywords[i] = key_node->nd_lit;
+		COMPILE(ret, "keyword values", val_node);
+	    }
+	    assert(i == len);
+	    return TRUE;
+	}
+    }
+    return FALSE;
+}
+
+enum compile_array_type_t {
+    COMPILE_ARRAY_TYPE_ARRAY,
+    COMPILE_ARRAY_TYPE_HASH,
+    COMPILE_ARRAY_TYPE_ARGS
+};
+
+static inline int
+static_literal_node_p(const NODE *node, const rb_iseq_t *iseq)
+{
+    node = node->nd_head;
+    switch (nd_type(node)) {
+      case NODE_LIT:
+      case NODE_NIL:
+      case NODE_TRUE:
+      case NODE_FALSE:
+	return TRUE;
+      case NODE_STR:
+        if (ISEQ_COMPILE_DATA(iseq)->option->frozen_string_literal) {
+            return TRUE;
+        } else {
+            return FALSE;
+        }
+      default:
+	return FALSE;
+    }
+}
+
+static inline VALUE
+static_literal_value(const NODE *node, rb_iseq_t *iseq)
+{
+    node = node->nd_head;
+    switch (nd_type(node)) {
+      case NODE_NIL:
+	return Qnil;
+      case NODE_TRUE:
+	return Qtrue;
+      case NODE_FALSE:
+	return Qfalse;
+      case NODE_STR:
+        if (ISEQ_COMPILE_DATA(iseq)->option->debug_frozen_string_literal || RTEST(ruby_debug)) {
+            VALUE lit;
+            VALUE debug_info = rb_ary_new_from_args(2, rb_iseq_path(iseq), INT2FIX((int)nd_line(node)));
+            lit = rb_str_dup(node->nd_lit);
+            rb_ivar_set(lit, id_debug_created_info, rb_obj_freeze(debug_info));
+            return rb_str_freeze(lit);
+        }
+        else {
+            return rb_fstring(node->nd_lit);
+        }
+      default:
+	return node->nd_lit;
+    }
+}
+
+static int
+compile_array(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node_root,
+	      enum compile_array_type_t type, struct rb_call_info_kw_arg **keywords_ptr,
+	      unsigned int *flag, int popped)
+{
+    const NODE *node = node_root;
+    int line = (int)nd_line(node);
+    int len = 0;
+
+    if (nd_type(node) == NODE_ZARRAY) {
+	if (!popped) {
+	    switch (type) {
+	      case COMPILE_ARRAY_TYPE_ARRAY: ADD_INSN1(ret, line, newarray, INT2FIX(0)); break;
+	      case COMPILE_ARRAY_TYPE_HASH: ADD_INSN1(ret, line, newhash, INT2FIX(0)); break;
+	      case COMPILE_ARRAY_TYPE_ARGS: /* do nothing */ break;
+	    }
+	}
+    }
+    else {
+	int opt_p = 1;
+	int first = 1, i;
+
+	while (node) {
+	    const NODE *start_node = node, *end_node;
+	    const NODE *kw = 0;
+	    const int max = 0x100;
+	    DECL_ANCHOR(anchor);
+	    INIT_ANCHOR(anchor);
+
+	    for (i=0; i<max && node; i++, len++, node = node->nd_next) {
+		if (CPDEBUG > 0) {
+		    EXPECT_NODE("compile_array", node, NODE_ARRAY, -1);
+		}
+
+		if (type != COMPILE_ARRAY_TYPE_ARRAY && !node->nd_head) {
+		    kw = node->nd_next;
+		    node = 0;
+		    if (kw) {
+			opt_p = 0;
+			node = kw->nd_next;
+			kw = kw->nd_head;
+		    }
+		    break;
+		}
+                if (opt_p && !static_literal_node_p(node, iseq)) {
+		    opt_p = 0;
+		}
+
+		if (type == COMPILE_ARRAY_TYPE_ARGS &&
+		    node->nd_next == NULL /* last node */ &&
+		    compile_array_keyword_arg(iseq, anchor, node->nd_head, keywords_ptr, flag)) {
+		    len--;
+		}
+		else {
+		    COMPILE_(anchor, "array element", node->nd_head, popped);
+		}
+	    }
+
+	    if (opt_p && type != COMPILE_ARRAY_TYPE_ARGS) {
+		if (!popped) {
+		    VALUE ary = rb_ary_tmp_new(i);
+
+		    end_node = node;
+		    node = start_node;
+
+		    while (node != end_node) {
+                        rb_ary_push(ary, static_literal_value(node, iseq));
+			node = node->nd_next;
+		    }
+		    while (node && node->nd_next &&
+                           static_literal_node_p(node, iseq) &&
+                           static_literal_node_p(node->nd_next, iseq)) {
+			VALUE elem[2];
+                        elem[0] = static_literal_value(node, iseq);
+                        elem[1] = static_literal_value(node->nd_next, iseq);
+			rb_ary_cat(ary, elem, 2);
+			node = node->nd_next->nd_next;
+			len++;
+		    }
+
+		    OBJ_FREEZE(ary);
+
+		    iseq_add_mark_object_compile_time(iseq, ary);
+
+		    if (first) {
+			first = 0;
+			if (type == COMPILE_ARRAY_TYPE_ARRAY) {
+			    ADD_INSN1(ret, line, duparray, ary);
+			}
+			else { /* COMPILE_ARRAY_TYPE_HASH */
+                            VALUE hash;
+
+                            hash = rb_hash_new_with_size(RARRAY_LEN(ary) / 2);
+                            rb_hash_bulk_insert(RARRAY_LEN(ary), RARRAY_CONST_PTR_TRANSIENT(ary), hash);
+                            iseq_add_mark_object_compile_time(iseq, hash);
+                            ADD_INSN1(ret, line, duphash, hash);
+			}
+		    }
+		    else {
+			if (type == COMPILE_ARRAY_TYPE_ARRAY) {
+			    ADD_INSN1(ret, line, putobject, ary);
+			    ADD_INSN(ret, line, concatarray);
+			}
+			else {
+			    COMPILE_ERROR(ERROR_ARGS "core#hash_merge_ary");
+			    return -1;
+			}
+		    }
+		}
+	    }
+	    else {
+		if (!popped || kw) {
+		    switch (type) {
+		      case COMPILE_ARRAY_TYPE_ARRAY:
+			ADD_INSN1(anchor, line, newarray, INT2FIX(i));
+
+			if (first) {
+			    first = 0;
+			}
+			else {
+			    ADD_INSN(anchor, line, concatarray);
+			}
+
+			APPEND_LIST(ret, anchor);
+			break;
+		      case COMPILE_ARRAY_TYPE_HASH:
+			if (i > 0) {
+			    if (first) {
+				if (!popped) {
+				    ADD_INSN1(anchor, line, newhash, INT2FIX(i));
+				}
+				APPEND_LIST(ret, anchor);
+			    }
+			    else {
+				if (!popped) {
+				    ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+				    ADD_INSN(ret, line, swap);
+				}
+				APPEND_LIST(ret, anchor);
+				if (!popped) {
+				    ADD_SEND(ret, line, id_core_hash_merge_ptr, INT2FIX(i + 1));
+				}
+			    }
+			}
+			if (kw) {
+			    if (!popped) {
+				ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+				if (i > 0 || !first) ADD_INSN(ret, line, swap);
+				else ADD_INSN1(ret, line, newhash, INT2FIX(0));
+			    }
+			    COMPILE(ret, "keyword splat", kw);
+			    if (popped) {
+				ADD_INSN(ret, line, pop);
+			    }
+			    else {
+				ADD_SEND(ret, line, id_core_hash_merge_kwd, INT2FIX(2));
+			    }
+			}
+			first = 0;
+			break;
+		      case COMPILE_ARRAY_TYPE_ARGS:
+			APPEND_LIST(ret, anchor);
+			break;
+		    }
+		}
+		else {
+		    /* popped */
+		    APPEND_LIST(ret, anchor);
+		}
+	    }
+	}
+    }
+    return len;
+}
+
+static VALUE
+case_when_optimizable_literal(const NODE *const node)
+{
+    switch (nd_type(node)) {
+      case NODE_LIT: {
+	VALUE v = node->nd_lit;
+	double ival;
+	if (RB_TYPE_P(v, T_FLOAT) &&
+	    modf(RFLOAT_VALUE(v), &ival) == 0.0) {
+	    return FIXABLE(ival) ? LONG2FIX((long)ival) : rb_dbl2big(ival);
+	}
+	if (SYMBOL_P(v) || rb_obj_is_kind_of(v, rb_cNumeric)) {
+	    return v;
+	}
+	break;
+      }
+      case NODE_NIL:
+	return Qnil;
+      case NODE_TRUE:
+	return Qtrue;
+      case NODE_FALSE:
+	return Qfalse;
+      case NODE_STR:
+	return rb_fstring(node->nd_lit);
+    }
+    return Qundef;
+}
+
+static int
+when_vals(rb_iseq_t *iseq, LINK_ANCHOR *const cond_seq, const NODE *vals,
+	  LABEL *l1, int only_special_literals, VALUE literals)
+{
+    while (vals) {
+	const NODE *val = vals->nd_head;
+	VALUE lit = case_when_optimizable_literal(val);
+
+	if (lit == Qundef) {
+	    only_special_literals = 0;
+	}
+	else {
+	    if (rb_hash_lookup(literals, lit) != Qnil) {
+		VALUE file = rb_iseq_path(iseq);
+		rb_compile_warning(RSTRING_PTR(file), nd_line(val),
+				   "duplicated when clause is ignored");
+	    }
+	    else {
+		rb_hash_aset(literals, lit, (VALUE)(l1) | 1);
+	    }
+	}
+
+	ADD_INSN(cond_seq, nd_line(val), dup); /* dup target */
+
+	if (nd_type(val) == NODE_STR) {
+	    debugp_param("nd_lit", val->nd_lit);
+	    lit = freeze_literal(iseq, val->nd_lit);
+	    ADD_INSN1(cond_seq, nd_line(val), putobject, lit);
+	}
+	else {
+	    if (!COMPILE(cond_seq, "when cond", val)) return -1;
+	}
+
+	ADD_INSN1(cond_seq, nd_line(vals), checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_CASE));
+	ADD_INSNL(cond_seq, nd_line(val), branchif, l1);
+	vals = vals->nd_next;
+    }
+    return only_special_literals;
+}
+
+static int
+when_splat_vals(rb_iseq_t *iseq, LINK_ANCHOR *const cond_seq, const NODE *vals,
+                LABEL *l1, int only_special_literals, VALUE literals)
+{
+    const int line = nd_line(vals);
+
+    switch (nd_type(vals)) {
+      case NODE_ARRAY:
+        if (when_vals(iseq, cond_seq, vals, l1, only_special_literals, literals) < 0)
+            return COMPILE_NG;
+        break;
+      case NODE_SPLAT:
+        ADD_INSN (cond_seq, line, dup);
+        CHECK(COMPILE(cond_seq, "when splat", vals->nd_head));
+        ADD_INSN1(cond_seq, line, splatarray, Qfalse);
+        ADD_INSN1(cond_seq, line, checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_CASE | VM_CHECKMATCH_ARRAY));
+        ADD_INSNL(cond_seq, line, branchif, l1);
+        break;
+      case NODE_ARGSCAT:
+        CHECK(when_splat_vals(iseq, cond_seq, vals->nd_head, l1, only_special_literals, literals));
+        CHECK(when_splat_vals(iseq, cond_seq, vals->nd_body, l1, only_special_literals, literals));
+        break;
+      case NODE_ARGSPUSH:
+        CHECK(when_splat_vals(iseq, cond_seq, vals->nd_head, l1, only_special_literals, literals));
+        ADD_INSN (cond_seq, line, dup);
+        CHECK(COMPILE(cond_seq, "when argspush body", vals->nd_body));
+        ADD_INSN1(cond_seq, line, checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_CASE));
+        ADD_INSNL(cond_seq, line, branchif, l1);
+        break;
+      default:
+        ADD_INSN (cond_seq, line, dup);
+        CHECK(COMPILE(cond_seq, "when val", vals));
+        ADD_INSN1(cond_seq, line, splatarray, Qfalse);
+        ADD_INSN1(cond_seq, line, checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_CASE | VM_CHECKMATCH_ARRAY));
+        ADD_INSNL(cond_seq, line, branchif, l1);
+        break;
+    }
+    return COMPILE_OK;
+}
+
+
+static int
+compile_massign_lhs(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node)
+{
+    switch (nd_type(node)) {
+      case NODE_ATTRASGN: {
+	INSN *iobj;
+	struct rb_call_info *ci;
+	VALUE dupidx;
+	int line = nd_line(node);
+
+	CHECK(COMPILE_POPPED(ret, "masgn lhs (NODE_ATTRASGN)", node));
+
+	iobj = (INSN *)get_prev_insn((INSN *)LAST_ELEMENT(ret)); /* send insn */
+	ci = (struct rb_call_info *)iobj->operands[0];
+	ci->orig_argc += 1;
+	dupidx = INT2FIX(ci->orig_argc);
+
+	INSERT_BEFORE_INSN1(iobj, line, topn, dupidx);
+	if (ci->flag & VM_CALL_ARGS_SPLAT) {
+	    --ci->orig_argc;
+	    INSERT_BEFORE_INSN1(iobj, line, newarray, INT2FIX(1));
+	    INSERT_BEFORE_INSN(iobj, line, concatarray);
+	}
+	ADD_INSN(ret, line, pop);	/* result */
+	break;
+      }
+      case NODE_MASGN: {
+	DECL_ANCHOR(anchor);
+	INIT_ANCHOR(anchor);
+	CHECK(COMPILE_POPPED(anchor, "nest masgn lhs", node));
+	ELEM_REMOVE(FIRST_ELEMENT(anchor));
+	ADD_SEQ(ret, anchor);
+	break;
+      }
+      default: {
+	DECL_ANCHOR(anchor);
+	INIT_ANCHOR(anchor);
+	CHECK(COMPILE_POPPED(anchor, "masgn lhs", node));
+	ELEM_REMOVE(FIRST_ELEMENT(anchor));
+	ADD_SEQ(ret, anchor);
+      }
+    }
+
+    return COMPILE_OK;
+}
+
+static int
+compile_massign_opt_lhs(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *lhsn)
+{
+    if (lhsn) {
+	CHECK(compile_massign_opt_lhs(iseq, ret, lhsn->nd_next));
+	CHECK(compile_massign_lhs(iseq, ret, lhsn->nd_head));
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_massign_opt(rb_iseq_t *iseq, LINK_ANCHOR *const ret,
+		    const NODE *rhsn, const NODE *orig_lhsn)
+{
+    VALUE mem[64];
+    const int memsize = numberof(mem);
+    int memindex = 0;
+    int llen = 0, rlen = 0;
+    int i;
+    const NODE *lhsn = orig_lhsn;
+
+#define MEMORY(v) { \
+    int i; \
+    if (memindex == memsize) return 0; \
+    for (i=0; i<memindex; i++) { \
+	if (mem[i] == (v)) return 0; \
+    } \
+    mem[memindex++] = (v); \
+}
+
+    if (rhsn == 0 || nd_type(rhsn) != NODE_ARRAY) {
+	return 0;
+    }
+
+    while (lhsn) {
+	const NODE *ln = lhsn->nd_head;
+	switch (nd_type(ln)) {
+	  case NODE_LASGN:
+	    MEMORY(ln->nd_vid);
+	    break;
+	  case NODE_DASGN:
+	  case NODE_DASGN_CURR:
+	  case NODE_IASGN:
+	  case NODE_CVASGN:
+	    MEMORY(ln->nd_vid);
+	    break;
+	  default:
+	    return 0;
+	}
+	lhsn = lhsn->nd_next;
+	llen++;
+    }
+
+    while (rhsn) {
+	if (llen <= rlen) {
+	    COMPILE_POPPED(ret, "masgn val (popped)", rhsn->nd_head);
+	}
+	else {
+	    COMPILE(ret, "masgn val", rhsn->nd_head);
+	}
+	rhsn = rhsn->nd_next;
+	rlen++;
+    }
+
+    if (llen > rlen) {
+	for (i=0; i<llen-rlen; i++) {
+	    ADD_INSN(ret, nd_line(orig_lhsn), putnil);
+	}
+    }
+
+    compile_massign_opt_lhs(iseq, ret, orig_lhsn);
+    return 1;
+}
+
+static void
+adjust_stack(rb_iseq_t *iseq, LINK_ANCHOR *const ret, int line, int rlen, int llen)
+{
+    if (rlen < llen) {
+	do {ADD_INSN(ret, line, putnil);} while (++rlen < llen);
+    }
+    else if (rlen > llen) {
+	do {ADD_INSN(ret, line, pop);} while (--rlen > llen);
+    }
+}
+
+static int
+compile_massign(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const NODE *rhsn = node->nd_value;
+    const NODE *splatn = node->nd_args;
+    const NODE *lhsn = node->nd_head;
+    int lhs_splat = (splatn && NODE_NAMED_REST_P(splatn)) ? 1 : 0;
+
+    if (!popped || splatn || !compile_massign_opt(iseq, ret, rhsn, lhsn)) {
+	int llen = 0;
+	int expand = 1;
+	DECL_ANCHOR(lhsseq);
+
+	INIT_ANCHOR(lhsseq);
+
+	while (lhsn) {
+	    CHECK(compile_massign_lhs(iseq, lhsseq, lhsn->nd_head));
+	    llen += 1;
+	    lhsn = lhsn->nd_next;
+	}
+
+	COMPILE(ret, "normal masgn rhs", rhsn);
+
+	if (!popped) {
+	    ADD_INSN(ret, nd_line(node), dup);
+	}
+	else if (!lhs_splat) {
+	    INSN *last = (INSN*)ret->last;
+	    if (IS_INSN(&last->link) &&
+		IS_INSN_ID(last, newarray) &&
+		last->operand_size == 1) {
+		int rlen = FIX2INT(OPERAND_AT(last, 0));
+		/* special case: assign to aset or attrset */
+		if (llen == 2) {
+		    POP_ELEMENT(ret);
+		    adjust_stack(iseq, ret, nd_line(node), rlen, llen);
+		    ADD_INSN(ret, nd_line(node), swap);
+		    expand = 0;
+		}
+		else if (llen > 2 && llen != rlen) {
+		    POP_ELEMENT(ret);
+		    adjust_stack(iseq, ret, nd_line(node), rlen, llen);
+		    ADD_INSN1(ret, nd_line(node), reverse, INT2FIX(llen));
+		    expand = 0;
+		}
+		else if (llen > 2) {
+		    last->insn_id = BIN(reverse);
+		    expand = 0;
+		}
+	    }
+	}
+	if (expand) {
+	    ADD_INSN2(ret, nd_line(node), expandarray,
+		      INT2FIX(llen), INT2FIX(lhs_splat));
+	}
+	ADD_SEQ(ret, lhsseq);
+
+	if (lhs_splat) {
+	    if (nd_type(splatn) == NODE_POSTARG) {
+		/*a, b, *r, p1, p2 */
+		const NODE *postn = splatn->nd_2nd;
+		const NODE *restn = splatn->nd_1st;
+		int num = (int)postn->nd_alen;
+		int flag = 0x02 | (NODE_NAMED_REST_P(restn) ? 0x01 : 0x00);
+
+		ADD_INSN2(ret, nd_line(splatn), expandarray,
+			  INT2FIX(num), INT2FIX(flag));
+
+		if (NODE_NAMED_REST_P(restn)) {
+		    CHECK(compile_massign_lhs(iseq, ret, restn));
+		}
+		while (postn) {
+		    CHECK(compile_massign_lhs(iseq, ret, postn->nd_head));
+		    postn = postn->nd_next;
+		}
+	    }
+	    else {
+		/* a, b, *r */
+		CHECK(compile_massign_lhs(iseq, ret, splatn));
+	    }
+	}
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_const_prefix(rb_iseq_t *iseq, const NODE *const node,
+		     LINK_ANCHOR *const pref, LINK_ANCHOR *const body)
+{
+    switch (nd_type(node)) {
+      case NODE_CONST:
+	debugi("compile_const_prefix - colon", node->nd_vid);
+	ADD_INSN1(body, nd_line(node), getconstant, ID2SYM(node->nd_vid));
+	break;
+      case NODE_COLON3:
+	debugi("compile_const_prefix - colon3", node->nd_mid);
+	ADD_INSN(body, nd_line(node), pop);
+	ADD_INSN1(body, nd_line(node), putobject, rb_cObject);
+	ADD_INSN1(body, nd_line(node), getconstant, ID2SYM(node->nd_mid));
+	break;
+      case NODE_COLON2:
+	CHECK(compile_const_prefix(iseq, node->nd_head, pref, body));
+	debugi("compile_const_prefix - colon2", node->nd_mid);
+	ADD_INSN1(body, nd_line(node), getconstant, ID2SYM(node->nd_mid));
+	break;
+      default:
+	CHECK(COMPILE(pref, "const colon2 prefix", node));
+	break;
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_cpath(LINK_ANCHOR *const ret, rb_iseq_t *iseq, const NODE *cpath)
+{
+    if (nd_type(cpath) == NODE_COLON3) {
+	/* toplevel class ::Foo */
+	ADD_INSN1(ret, nd_line(cpath), putobject, rb_cObject);
+	return VM_DEFINECLASS_FLAG_SCOPED;
+    }
+    else if (cpath->nd_head) {
+	/* Bar::Foo */
+	COMPILE(ret, "nd_else->nd_head", cpath->nd_head);
+	return VM_DEFINECLASS_FLAG_SCOPED;
+    }
+    else {
+	/* class at cbase Foo */
+	ADD_INSN1(ret, nd_line(cpath), putspecialobject,
+		  INT2FIX(VM_SPECIAL_OBJECT_CONST_BASE));
+	return 0;
+    }
+}
+
+#define private_recv_p(node) (nd_type((node)->nd_recv) == NODE_SELF)
+static int
+defined_expr(rb_iseq_t *iseq, LINK_ANCHOR *const ret,
+	     const NODE *const node, LABEL **lfinish, VALUE needstr);
+
+static int
+defined_expr0(rb_iseq_t *iseq, LINK_ANCHOR *const ret,
+	      const NODE *const node, LABEL **lfinish, VALUE needstr)
+{
+    enum defined_type expr_type = DEFINED_NOT_DEFINED;
+    enum node_type type;
+
+    switch (type = nd_type(node)) {
+
+	/* easy literals */
+      case NODE_NIL:
+	expr_type = DEFINED_NIL;
+	break;
+      case NODE_SELF:
+	expr_type = DEFINED_SELF;
+	break;
+      case NODE_TRUE:
+	expr_type = DEFINED_TRUE;
+	break;
+      case NODE_FALSE:
+	expr_type = DEFINED_FALSE;
+	break;
+
+      case NODE_ARRAY:{
+	const NODE *vals = node;
+
+	do {
+	    defined_expr0(iseq, ret, vals->nd_head, lfinish, Qfalse);
+
+	    if (!lfinish[1]) {
+		lfinish[1] = NEW_LABEL(nd_line(node));
+	    }
+	    ADD_INSNL(ret, nd_line(node), branchunless, lfinish[1]);
+	} while ((vals = vals->nd_next) != NULL);
+      }
+      case NODE_STR:
+      case NODE_LIT:
+      case NODE_ZARRAY:
+      case NODE_AND:
+      case NODE_OR:
+      default:
+	expr_type = DEFINED_EXPR;
+	break;
+
+	/* variables */
+      case NODE_LVAR:
+      case NODE_DVAR:
+	expr_type = DEFINED_LVAR;
+	break;
+
+      case NODE_IVAR:
+	ADD_INSN(ret, nd_line(node), putnil);
+	ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_IVAR),
+		  ID2SYM(node->nd_vid), needstr);
+	return 1;
+
+      case NODE_GVAR:
+	ADD_INSN(ret, nd_line(node), putnil);
+	ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_GVAR),
+		  ID2SYM(node->nd_entry->id), needstr);
+	return 1;
+
+      case NODE_CVAR:
+	ADD_INSN(ret, nd_line(node), putnil);
+	ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_CVAR),
+		  ID2SYM(node->nd_vid), needstr);
+	return 1;
+
+      case NODE_CONST:
+	ADD_INSN(ret, nd_line(node), putnil);
+	ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_CONST),
+		  ID2SYM(node->nd_vid), needstr);
+	return 1;
+      case NODE_COLON2:
+	if (!lfinish[1]) {
+	    lfinish[1] = NEW_LABEL(nd_line(node));
+	}
+	defined_expr0(iseq, ret, node->nd_head, lfinish, Qfalse);
+	ADD_INSNL(ret, nd_line(node), branchunless, lfinish[1]);
+	COMPILE(ret, "defined/colon2#nd_head", node->nd_head);
+
+	ADD_INSN3(ret, nd_line(node), defined,
+		  (rb_is_const_id(node->nd_mid) ?
+		   INT2FIX(DEFINED_CONST) : INT2FIX(DEFINED_METHOD)),
+		  ID2SYM(node->nd_mid), needstr);
+	return 1;
+      case NODE_COLON3:
+	ADD_INSN1(ret, nd_line(node), putobject, rb_cObject);
+	ADD_INSN3(ret, nd_line(node), defined,
+		  INT2FIX(DEFINED_CONST), ID2SYM(node->nd_mid), needstr);
+	return 1;
+
+	/* method dispatch */
+      case NODE_CALL:
+      case NODE_OPCALL:
+      case NODE_VCALL:
+      case NODE_FCALL:
+      case NODE_ATTRASGN:{
+	const int explicit_receiver =
+	    (type == NODE_CALL || type == NODE_OPCALL ||
+	     (type == NODE_ATTRASGN && !private_recv_p(node)));
+
+	if (!lfinish[1] && (node->nd_args || explicit_receiver)) {
+	    lfinish[1] = NEW_LABEL(nd_line(node));
+	}
+	if (node->nd_args) {
+	    defined_expr0(iseq, ret, node->nd_args, lfinish, Qfalse);
+	    ADD_INSNL(ret, nd_line(node), branchunless, lfinish[1]);
+	}
+	if (explicit_receiver) {
+	    defined_expr0(iseq, ret, node->nd_recv, lfinish, Qfalse);
+	    ADD_INSNL(ret, nd_line(node), branchunless, lfinish[1]);
+	    COMPILE(ret, "defined/recv", node->nd_recv);
+	    ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_METHOD),
+		      ID2SYM(node->nd_mid), needstr);
+	}
+	else {
+	    ADD_INSN(ret, nd_line(node), putself);
+	    ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_FUNC),
+		      ID2SYM(node->nd_mid), needstr);
+	}
+	return 1;
+      }
+
+      case NODE_YIELD:
+	ADD_INSN(ret, nd_line(node), putnil);
+	ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_YIELD), 0,
+		  needstr);
+	return 1;
+
+      case NODE_BACK_REF:
+      case NODE_NTH_REF:
+	ADD_INSN(ret, nd_line(node), putnil);
+	ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_REF),
+		  INT2FIX((node->nd_nth << 1) | (type == NODE_BACK_REF)),
+		  needstr);
+	return 1;
+
+      case NODE_SUPER:
+      case NODE_ZSUPER:
+	ADD_INSN(ret, nd_line(node), putnil);
+	ADD_INSN3(ret, nd_line(node), defined, INT2FIX(DEFINED_ZSUPER), 0,
+		  needstr);
+	return 1;
+
+      case NODE_OP_ASGN1:
+      case NODE_OP_ASGN2:
+      case NODE_OP_ASGN_OR:
+      case NODE_OP_ASGN_AND:
+      case NODE_MASGN:
+      case NODE_LASGN:
+      case NODE_DASGN:
+      case NODE_DASGN_CURR:
+      case NODE_GASGN:
+      case NODE_IASGN:
+      case NODE_CDECL:
+      case NODE_CVASGN:
+	expr_type = DEFINED_ASGN;
+	break;
+    }
+
+    if (expr_type) {
+	if (needstr != Qfalse) {
+	    VALUE str = rb_iseq_defined_string(expr_type);
+	    ADD_INSN1(ret, nd_line(node), putobject, str);
+	}
+	else {
+	    ADD_INSN1(ret, nd_line(node), putobject, Qtrue);
+	}
+	return 1;
+    }
+    return 0;
+}
+
+static VALUE
+build_defined_rescue_iseq(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *unused)
+{
+    ADD_INSN(ret, 0, putnil);
+    iseq_set_exception_local_table(iseq);
+    return Qnil;
+}
+
+static int
+defined_expr(rb_iseq_t *iseq, LINK_ANCHOR *const ret,
+	     const NODE *const node, LABEL **lfinish, VALUE needstr)
+{
+    LINK_ELEMENT *lcur = ret->last;
+    int done = defined_expr0(iseq, ret, node, lfinish, needstr);
+    if (lfinish[1]) {
+	int line = nd_line(node);
+	LABEL *lstart = NEW_LABEL(line);
+	LABEL *lend = NEW_LABEL(line);
+	const rb_iseq_t *rescue;
+	rescue = new_child_iseq_ifunc(iseq, IFUNC_NEW(build_defined_rescue_iseq, 0, 0),
+				      rb_str_concat(rb_str_new2("defined guard in "),
+						    iseq->body->location.label),
+				      iseq, ISEQ_TYPE_RESCUE, 0);
+	lstart->rescued = LABEL_RESCUE_BEG;
+	lend->rescued = LABEL_RESCUE_END;
+	APPEND_LABEL(ret, lcur, lstart);
+	ADD_LABEL(ret, lend);
+	ADD_CATCH_ENTRY(CATCH_TYPE_RESCUE, lstart, lend, rescue, lfinish[1]);
+    }
+    return done;
+}
+
+static int
+compile_defined_expr(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, VALUE needstr)
+{
+    const int line = nd_line(node);
+    if (!node->nd_head) {
+	VALUE str = rb_iseq_defined_string(DEFINED_NIL);
+	ADD_INSN1(ret, line, putobject, str);
+    }
+    else {
+	LABEL *lfinish[2];
+	LINK_ELEMENT *last = ret->last;
+	lfinish[0] = NEW_LABEL(line);
+	lfinish[1] = 0;
+	defined_expr(iseq, ret, node->nd_head, lfinish, needstr);
+	if (lfinish[1]) {
+	    ELEM_INSERT_NEXT(last, &new_insn_body(iseq, line, BIN(putnil), 0)->link);
+	    ADD_INSN(ret, line, swap);
+	    ADD_INSN(ret, line, pop);
+	    ADD_LABEL(ret, lfinish[1]);
+	}
+	ADD_LABEL(ret, lfinish[0]);
+    }
+    return COMPILE_OK;
+}
+
+static VALUE
+make_name_for_block(const rb_iseq_t *orig_iseq)
+{
+    int level = 1;
+    const rb_iseq_t *iseq = orig_iseq;
+
+    if (orig_iseq->body->parent_iseq != 0) {
+	while (orig_iseq->body->local_iseq != iseq) {
+	    if (iseq->body->type == ISEQ_TYPE_BLOCK) {
+		level++;
+	    }
+	    iseq = iseq->body->parent_iseq;
+	}
+    }
+
+    if (level == 1) {
+	return rb_sprintf("block in %"PRIsVALUE, iseq->body->location.label);
+    }
+    else {
+	return rb_sprintf("block (%d levels) in %"PRIsVALUE, level, iseq->body->location.label);
+    }
+}
+
+static void
+push_ensure_entry(rb_iseq_t *iseq,
+		  struct iseq_compile_data_ensure_node_stack *enl,
+		  struct ensure_range *er, const NODE *const node)
+{
+    enl->ensure_node = node;
+    enl->prev = ISEQ_COMPILE_DATA(iseq)->ensure_node_stack;	/* prev */
+    enl->erange = er;
+    ISEQ_COMPILE_DATA(iseq)->ensure_node_stack = enl;
+}
+
+static void
+add_ensure_range(rb_iseq_t *iseq, struct ensure_range *erange,
+		 LABEL *lstart, LABEL *lend)
+{
+    struct ensure_range *ne =
+	compile_data_alloc(iseq, sizeof(struct ensure_range));
+
+    while (erange->next != 0) {
+	erange = erange->next;
+    }
+    ne->next = 0;
+    ne->begin = lend;
+    ne->end = erange->end;
+    erange->end = lstart;
+
+    erange->next = ne;
+}
+
+static void
+add_ensure_iseq(LINK_ANCHOR *const ret, rb_iseq_t *iseq, int is_return)
+{
+    struct iseq_compile_data_ensure_node_stack *enlp =
+	ISEQ_COMPILE_DATA(iseq)->ensure_node_stack;
+    struct iseq_compile_data_ensure_node_stack *prev_enlp = enlp;
+    DECL_ANCHOR(ensure);
+
+    INIT_ANCHOR(ensure);
+    while (enlp) {
+	if (enlp->erange != NULL) {
+	    DECL_ANCHOR(ensure_part);
+	    LABEL *lstart = NEW_LABEL(0);
+	    LABEL *lend = NEW_LABEL(0);
+	    INIT_ANCHOR(ensure_part);
+
+	    add_ensure_range(iseq, enlp->erange, lstart, lend);
+
+	    ISEQ_COMPILE_DATA(iseq)->ensure_node_stack = enlp->prev;
+	    ADD_LABEL(ensure_part, lstart);
+	    COMPILE_POPPED(ensure_part, "ensure part", enlp->ensure_node);
+	    ADD_LABEL(ensure_part, lend);
+	    ADD_SEQ(ensure, ensure_part);
+	}
+	else {
+	    if (!is_return) {
+		break;
+	    }
+	}
+	enlp = enlp->prev;
+    }
+    ISEQ_COMPILE_DATA(iseq)->ensure_node_stack = prev_enlp;
+    ADD_SEQ(ret, ensure);
+}
+
+static VALUE
+setup_args(rb_iseq_t *iseq, LINK_ANCHOR *const args, const NODE *argn,
+	   unsigned int *flag, struct rb_call_info_kw_arg **keywords)
+{
+    VALUE argc = INT2FIX(0);
+    int nsplat = 0;
+    DECL_ANCHOR(arg_block);
+    DECL_ANCHOR(args_splat);
+
+    INIT_ANCHOR(arg_block);
+    INIT_ANCHOR(args_splat);
+    if (argn && nd_type(argn) == NODE_BLOCK_PASS) {
+	COMPILE(arg_block, "block", argn->nd_body);
+	*flag |= VM_CALL_ARGS_BLOCKARG;
+	argn = argn->nd_head;
+    }
+
+  setup_argn:
+    if (argn) {
+	switch (nd_type(argn)) {
+	  case NODE_SPLAT: {
+	    COMPILE(args, "args (splat)", argn->nd_head);
+	    ADD_INSN1(args, nd_line(argn), splatarray, nsplat ? Qtrue : Qfalse);
+	    argc = INT2FIX(1);
+	    nsplat++;
+	    *flag |= VM_CALL_ARGS_SPLAT;
+	    break;
+	  }
+	  case NODE_ARGSCAT:
+	  case NODE_ARGSPUSH: {
+	    int next_is_array = (nd_type(argn->nd_head) == NODE_ARRAY);
+	    DECL_ANCHOR(tmp);
+
+	    INIT_ANCHOR(tmp);
+	    COMPILE(tmp, "args (cat: splat)", argn->nd_body);
+	    if (nd_type(argn) == NODE_ARGSCAT) {
+		ADD_INSN1(tmp, nd_line(argn), splatarray, nsplat ? Qtrue : Qfalse);
+	    }
+	    else {
+		ADD_INSN1(tmp, nd_line(argn), newarray, INT2FIX(1));
+	    }
+	    INSERT_LIST(args_splat, tmp);
+	    nsplat++;
+	    *flag |= VM_CALL_ARGS_SPLAT;
+	    if (nd_type(argn->nd_body) == NODE_HASH)
+		*flag |= VM_CALL_KW_SPLAT;
+
+	    if (next_is_array) {
+		int len = compile_array(iseq, args, argn->nd_head, COMPILE_ARRAY_TYPE_ARGS, NULL, flag, FALSE);
+		if (len < 0) return Qnil;
+		argc = INT2FIX(len + 1);
+	    }
+	    else {
+		argn = argn->nd_head;
+		goto setup_argn;
+	    }
+	    break;
+	  }
+	  case NODE_ARRAY:
+	    {
+		int len = compile_array(iseq, args, argn, COMPILE_ARRAY_TYPE_ARGS, keywords, flag, FALSE);
+		if (len < 0) return Qnil;
+		argc = INT2FIX(len);
+		break;
+	    }
+	  default: {
+	    UNKNOWN_NODE("setup_arg", argn, Qnil);
+	  }
+	}
+    }
+
+    if (nsplat > 1) {
+	int i;
+	for (i=1; i<nsplat; i++) {
+	    ADD_INSN(args_splat, nd_line(argn), concatarray);
+	}
+    }
+
+    if (!LIST_INSN_SIZE_ZERO(args_splat)) {
+	ADD_SEQ(args, args_splat);
+    }
+
+    if (*flag & VM_CALL_ARGS_BLOCKARG) {
+	if (LIST_INSN_SIZE_ONE(arg_block)) {
+	    LINK_ELEMENT *elem = FIRST_ELEMENT(arg_block);
+	    if (elem->type == ISEQ_ELEMENT_INSN) {
+		INSN *iobj = (INSN *)elem;
+		if (iobj->insn_id == BIN(getblockparam)) {
+		    iobj->insn_id = BIN(getblockparamproxy);
+		}
+	    }
+	}
+	ADD_SEQ(args, arg_block);
+    }
+    return argc;
+}
+
+static VALUE
+build_postexe_iseq(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *body)
+{
+    int line = nd_line(body);
+    VALUE argc = INT2FIX(0);
+    const rb_iseq_t *block = NEW_CHILD_ISEQ(body, make_name_for_block(iseq->body->parent_iseq), ISEQ_TYPE_BLOCK, line);
+
+    ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+    ADD_CALL_WITH_BLOCK(ret, line, id_core_set_postexe, argc, block);
+    iseq_set_local_table(iseq, 0);
+    return Qnil;
+}
+
+static void
+compile_named_capture_assign(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node)
+{
+    const NODE *vars;
+    LINK_ELEMENT *last;
+    int line = nd_line(node);
+    LABEL *fail_label = NEW_LABEL(line), *end_label = NEW_LABEL(line);
+
+#if !(defined(NAMED_CAPTURE_BY_SVAR) && NAMED_CAPTURE_BY_SVAR-0)
+    ADD_INSN1(ret, line, getglobal, ((VALUE)rb_global_entry(idBACKREF) | 1));
+#else
+    ADD_INSN2(ret, line, getspecial, INT2FIX(1) /* '~' */, INT2FIX(0));
+#endif
+    ADD_INSN(ret, line, dup);
+    ADD_INSNL(ret, line, branchunless, fail_label);
+
+    for (vars = node; vars; vars = vars->nd_next) {
+	INSN *cap;
+	if (vars->nd_next) {
+	    ADD_INSN(ret, line, dup);
+	}
+	last = ret->last;
+	COMPILE_POPPED(ret, "capture", vars->nd_head);
+	last = last->next; /* putobject :var */
+	cap = new_insn_send(iseq, line, idAREF, INT2FIX(1),
+			    NULL, INT2FIX(0), NULL);
+	ELEM_INSERT_PREV(last->next, (LINK_ELEMENT *)cap);
+#if !defined(NAMED_CAPTURE_SINGLE_OPT) || NAMED_CAPTURE_SINGLE_OPT-0
+	if (!vars->nd_next && vars == node) {
+	    /* only one name */
+	    DECL_ANCHOR(nom);
+
+	    INIT_ANCHOR(nom);
+	    ADD_INSNL(nom, line, jump, end_label);
+	    ADD_LABEL(nom, fail_label);
+# if 0				/* $~ must be MatchData or nil */
+	    ADD_INSN(nom, line, pop);
+	    ADD_INSN(nom, line, putnil);
+# endif
+	    ADD_LABEL(nom, end_label);
+	    (nom->last->next = cap->link.next)->prev = nom->last;
+	    (cap->link.next = nom->anchor.next)->prev = &cap->link;
+	    return;
+	}
+#endif
+    }
+    ADD_INSNL(ret, line, jump, end_label);
+    ADD_LABEL(ret, fail_label);
+    ADD_INSN(ret, line, pop);
+    for (vars = node; vars; vars = vars->nd_next) {
+	last = ret->last;
+	COMPILE_POPPED(ret, "capture", vars->nd_head);
+	last = last->next; /* putobject :var */
+	((INSN*)last)->insn_id = BIN(putnil);
+	((INSN*)last)->operand_size = 0;
+    }
+    ADD_LABEL(ret, end_label);
+}
+
+static int
+number_literal_p(const NODE *n)
+{
+    return (n && nd_type(n) == NODE_LIT && RB_INTEGER_TYPE_P(n->nd_lit));
+}
+
+static int
+compile_if(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped, const enum node_type type)
+{
+    struct rb_iseq_constant_body *const body = iseq->body;
+    const NODE *const node_body = type == NODE_IF ? node->nd_body : node->nd_else;
+    const NODE *const node_else = type == NODE_IF ? node->nd_else : node->nd_body;
+
+    const int line = nd_line(node);
+    const int lineno = nd_first_lineno(node);
+    const int column = nd_first_column(node);
+    const int last_lineno = nd_last_lineno(node);
+    const int last_column = nd_last_column(node);
+    DECL_ANCHOR(cond_seq);
+    DECL_ANCHOR(then_seq);
+    DECL_ANCHOR(else_seq);
+    LABEL *then_label, *else_label, *end_label;
+    VALUE branches = 0;
+    int ci_size, ci_kw_size;
+
+    INIT_ANCHOR(cond_seq);
+    INIT_ANCHOR(then_seq);
+    INIT_ANCHOR(else_seq);
+    then_label = NEW_LABEL(line);
+    else_label = NEW_LABEL(line);
+    end_label = 0;
+
+    compile_branch_condition(iseq, cond_seq, node->nd_cond,
+			     then_label, else_label);
+
+    ci_size = body->ci_size;
+    ci_kw_size = body->ci_kw_size;
+    CHECK(COMPILE_(then_seq, "then", node_body, popped));
+    if (!then_label->refcnt) {
+        body->ci_size = ci_size;
+        body->ci_kw_size = ci_kw_size;
+    }
+
+    ci_size = body->ci_size;
+    ci_kw_size = body->ci_kw_size;
+    CHECK(COMPILE_(else_seq, "else", node_else, popped));
+    if (!else_label->refcnt) {
+        body->ci_size = ci_size;
+        body->ci_kw_size = ci_kw_size;
+    }
+
+    ADD_SEQ(ret, cond_seq);
+
+    if (then_label->refcnt && else_label->refcnt) {
+	DECL_BRANCH_BASE(branches, lineno, column, last_lineno, last_column, type == NODE_IF ? "if" : "unless");
+    }
+
+    if (then_label->refcnt) {
+	ADD_LABEL(ret, then_label);
+	if (else_label->refcnt) {
+	    ADD_TRACE_BRANCH_COVERAGE(
+		ret,
+		node_body ? nd_first_lineno(node_body) : lineno,
+		node_body ? nd_first_column(node_body) : column,
+		node_body ? nd_last_lineno(node_body) : last_lineno,
+		node_body ? nd_last_column(node_body) : last_column,
+		type == NODE_IF ? "then" : "else",
+		branches);
+	    end_label = NEW_LABEL(line);
+	    ADD_INSNL(then_seq, line, jump, end_label);
+	}
+	ADD_SEQ(ret, then_seq);
+    }
+
+    if (else_label->refcnt) {
+	ADD_LABEL(ret, else_label);
+	if (then_label->refcnt) {
+	    ADD_TRACE_BRANCH_COVERAGE(
+		ret,
+		node_else ? nd_first_lineno(node_else) : lineno,
+		node_else ? nd_first_column(node_else) : column,
+		node_else ? nd_last_lineno(node_else) : last_lineno,
+		node_else ? nd_last_column(node_else) : last_column,
+		type == NODE_IF ? "else" : "then",
+		branches);
+	}
+	ADD_SEQ(ret, else_seq);
+    }
+
+    if (end_label) {
+	ADD_LABEL(ret, end_label);
+    }
+
+    return COMPILE_OK;
+}
+
+static int
+compile_case(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const orig_node, int popped)
+{
+    const NODE *vals;
+    const NODE *node = orig_node;
+    LABEL *endlabel, *elselabel;
+    DECL_ANCHOR(head);
+    DECL_ANCHOR(body_seq);
+    DECL_ANCHOR(cond_seq);
+    int only_special_literals = 1;
+    VALUE literals = rb_hash_new();
+    int line, lineno, column, last_lineno, last_column;
+    enum node_type type;
+    VALUE branches = 0;
+
+    INIT_ANCHOR(head);
+    INIT_ANCHOR(body_seq);
+    INIT_ANCHOR(cond_seq);
+
+    RHASH_TBL_RAW(literals)->type = &cdhash_type;
+
+    CHECK(COMPILE(head, "case base", node->nd_head));
+
+    DECL_BRANCH_BASE(branches, nd_first_lineno(node), nd_first_column(node), nd_last_lineno(node), nd_last_column(node), "case");
+
+    node = node->nd_body;
+    EXPECT_NODE("NODE_CASE", node, NODE_WHEN, COMPILE_NG);
+    type = nd_type(node);
+    line = nd_line(node);
+    lineno = nd_first_lineno(node);
+    column = nd_first_column(node);
+    last_lineno = nd_last_lineno(node);
+    last_column = nd_last_column(node);
+
+    endlabel = NEW_LABEL(line);
+    elselabel = NEW_LABEL(line);
+
+    ADD_SEQ(ret, head);	/* case VAL */
+
+    while (type == NODE_WHEN) {
+	LABEL *l1;
+
+	l1 = NEW_LABEL(line);
+	ADD_LABEL(body_seq, l1);
+	ADD_INSN(body_seq, line, pop);
+	ADD_TRACE_BRANCH_COVERAGE(
+		body_seq,
+		node->nd_body ? nd_first_lineno(node->nd_body) : lineno,
+		node->nd_body ? nd_first_column(node->nd_body) : column,
+		node->nd_body ? nd_last_lineno(node->nd_body) : last_lineno,
+		node->nd_body ? nd_last_column(node->nd_body) : last_column,
+		"when",
+		branches);
+	CHECK(COMPILE_(body_seq, "when body", node->nd_body, popped));
+	ADD_INSNL(body_seq, line, jump, endlabel);
+
+	vals = node->nd_head;
+	if (vals) {
+	    switch (nd_type(vals)) {
+	      case NODE_ARRAY:
+		only_special_literals = when_vals(iseq, cond_seq, vals, l1, only_special_literals, literals);
+		if (only_special_literals < 0) return COMPILE_NG;
+		break;
+	      case NODE_SPLAT:
+	      case NODE_ARGSCAT:
+	      case NODE_ARGSPUSH:
+		only_special_literals = 0;
+		CHECK(when_splat_vals(iseq, cond_seq, vals, l1, only_special_literals, literals));
+		break;
+	      default:
+		UNKNOWN_NODE("NODE_CASE", vals, COMPILE_NG);
+	    }
+	}
+	else {
+	    EXPECT_NODE_NONULL("NODE_CASE", node, NODE_ARRAY, COMPILE_NG);
+	}
+
+	node = node->nd_next;
+	if (!node) {
+	    break;
+	}
+	type = nd_type(node);
+	line = nd_line(node);
+	lineno = nd_first_lineno(node);
+	column = nd_first_column(node);
+	last_lineno = nd_last_lineno(node);
+	last_column = nd_last_column(node);
+    }
+    /* else */
+    if (node) {
+	ADD_LABEL(cond_seq, elselabel);
+	ADD_INSN(cond_seq, line, pop);
+	ADD_TRACE_BRANCH_COVERAGE(cond_seq, nd_first_lineno(node), nd_first_column(node), nd_last_lineno(node), nd_last_column(node), "else", branches);
+	CHECK(COMPILE_(cond_seq, "else", node, popped));
+	ADD_INSNL(cond_seq, line, jump, endlabel);
+    }
+    else {
+	debugs("== else (implicit)\n");
+	ADD_LABEL(cond_seq, elselabel);
+	ADD_INSN(cond_seq, nd_line(orig_node), pop);
+	ADD_TRACE_BRANCH_COVERAGE(cond_seq, nd_first_lineno(orig_node), nd_first_column(orig_node), nd_last_lineno(orig_node), nd_last_column(orig_node), "else", branches);
+	if (!popped) {
+	    ADD_INSN(cond_seq, nd_line(orig_node), putnil);
+	}
+	ADD_INSNL(cond_seq, nd_line(orig_node), jump, endlabel);
+    }
+
+    if (only_special_literals && ISEQ_COMPILE_DATA(iseq)->option->specialized_instruction) {
+	iseq_add_mark_object_compile_time(iseq, literals);
+
+	ADD_INSN(ret, nd_line(orig_node), dup);
+	ADD_INSN2(ret, nd_line(orig_node), opt_case_dispatch, literals, elselabel);
+	LABEL_REF(elselabel);
+    }
+
+    ADD_SEQ(ret, cond_seq);
+    ADD_SEQ(ret, body_seq);
+    ADD_LABEL(ret, endlabel);
+    return COMPILE_OK;
+}
+
+static int
+compile_case2(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const orig_node, int popped)
+{
+    const NODE *vals;
+    const NODE *val;
+    const NODE *node = orig_node->nd_body;
+    LABEL *endlabel;
+    DECL_ANCHOR(body_seq);
+    VALUE branches = 0;
+
+    DECL_BRANCH_BASE(branches, nd_first_lineno(orig_node), nd_first_column(orig_node), nd_last_lineno(orig_node), nd_last_column(orig_node), "case");
+
+    INIT_ANCHOR(body_seq);
+    endlabel = NEW_LABEL(nd_line(node));
+
+    while (node && nd_type(node) == NODE_WHEN) {
+	const int line = nd_line(node);
+	const int lineno = nd_first_lineno(node);
+	const int column = nd_first_column(node);
+	const int last_lineno = nd_last_lineno(node);
+	const int last_column = nd_last_column(node);
+	LABEL *l1 = NEW_LABEL(line);
+	ADD_LABEL(body_seq, l1);
+	ADD_TRACE_BRANCH_COVERAGE(
+		body_seq,
+		node->nd_body ? nd_first_lineno(node->nd_body) : lineno,
+		node->nd_body ? nd_first_column(node->nd_body) : column,
+		node->nd_body ? nd_last_lineno(node->nd_body) : last_lineno,
+		node->nd_body ? nd_last_column(node->nd_body) : last_column,
+		"when",
+		branches);
+	CHECK(COMPILE_(body_seq, "when", node->nd_body, popped));
+	ADD_INSNL(body_seq, line, jump, endlabel);
+
+	vals = node->nd_head;
+	if (!vals) {
+	    COMPILE_ERROR(ERROR_ARGS "NODE_WHEN: must be NODE_ARRAY, but 0");
+	    return COMPILE_NG;
+	}
+	switch (nd_type(vals)) {
+	  case NODE_ARRAY:
+	    while (vals) {
+		LABEL *lnext;
+		val = vals->nd_head;
+		lnext = NEW_LABEL(nd_line(val));
+		debug_compile("== when2\n", (void)0);
+		CHECK(compile_branch_condition(iseq, ret, val, l1, lnext));
+		ADD_LABEL(ret, lnext);
+		vals = vals->nd_next;
+	    }
+	    break;
+	  case NODE_SPLAT:
+	  case NODE_ARGSCAT:
+	  case NODE_ARGSPUSH:
+	    ADD_INSN(ret, nd_line(vals), putnil);
+	    CHECK(COMPILE(ret, "when2/cond splat", vals));
+	    ADD_INSN1(ret, nd_line(vals), checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_WHEN | VM_CHECKMATCH_ARRAY));
+	    ADD_INSNL(ret, nd_line(vals), branchif, l1);
+	    break;
+	  default:
+	    UNKNOWN_NODE("NODE_WHEN", vals, COMPILE_NG);
+	}
+	node = node->nd_next;
+    }
+    /* else */
+    ADD_TRACE_BRANCH_COVERAGE(
+	ret,
+	node ? nd_first_lineno(node) : nd_first_lineno(orig_node),
+	node ? nd_first_column(node) : nd_first_column(orig_node),
+	node ? nd_last_lineno(node) : nd_last_lineno(orig_node),
+	node ? nd_last_column(node) : nd_last_column(orig_node),
+	"else",
+	branches);
+    CHECK(COMPILE_(ret, "else", node, popped));
+    ADD_INSNL(ret, nd_line(orig_node), jump, endlabel);
+
+    ADD_SEQ(ret, body_seq);
+    ADD_LABEL(ret, endlabel);
+    return COMPILE_OK;
+}
+
+static int
+compile_loop(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped, const enum node_type type)
+{
+    const int line = (int)nd_line(node);
+    const int lineno = nd_first_lineno(node);
+    const int column = nd_first_column(node);
+    const int last_lineno = nd_last_lineno(node);
+    const int last_column = nd_last_column(node);
+    LABEL *prev_start_label = ISEQ_COMPILE_DATA(iseq)->start_label;
+    LABEL *prev_end_label = ISEQ_COMPILE_DATA(iseq)->end_label;
+    LABEL *prev_redo_label = ISEQ_COMPILE_DATA(iseq)->redo_label;
+    int prev_loopval_popped = ISEQ_COMPILE_DATA(iseq)->loopval_popped;
+    VALUE branches = 0;
+
+    struct iseq_compile_data_ensure_node_stack enl;
+
+    LABEL *next_label = ISEQ_COMPILE_DATA(iseq)->start_label = NEW_LABEL(line);	/* next  */
+    LABEL *redo_label = ISEQ_COMPILE_DATA(iseq)->redo_label = NEW_LABEL(line);	/* redo  */
+    LABEL *break_label = ISEQ_COMPILE_DATA(iseq)->end_label = NEW_LABEL(line);	/* break */
+    LABEL *end_label = NEW_LABEL(line);
+    LABEL *adjust_label = NEW_LABEL(line);
+
+    LABEL *next_catch_label = NEW_LABEL(line);
+    LABEL *tmp_label = NULL;
+
+    ISEQ_COMPILE_DATA(iseq)->loopval_popped = 0;
+    push_ensure_entry(iseq, &enl, NULL, NULL);
+
+    if (node->nd_state == 1) {
+	ADD_INSNL(ret, line, jump, next_label);
+    }
+    else {
+	tmp_label = NEW_LABEL(line);
+	ADD_INSNL(ret, line, jump, tmp_label);
+    }
+    ADD_LABEL(ret, adjust_label);
+    ADD_INSN(ret, line, putnil);
+    ADD_LABEL(ret, next_catch_label);
+    ADD_INSN(ret, line, pop);
+    ADD_INSNL(ret, line, jump, next_label);
+    if (tmp_label) ADD_LABEL(ret, tmp_label);
+
+    ADD_LABEL(ret, redo_label);
+    DECL_BRANCH_BASE(branches, lineno, column, last_lineno, last_column, type == NODE_WHILE ? "while" : "until");
+    ADD_TRACE_BRANCH_COVERAGE(
+	ret,
+	node->nd_body ? nd_first_lineno(node->nd_body) : lineno,
+	node->nd_body ? nd_first_column(node->nd_body) : column,
+	node->nd_body ? nd_last_lineno(node->nd_body) : last_lineno,
+	node->nd_body ? nd_last_column(node->nd_body) : last_column,
+	"body",
+	branches);
+    CHECK(COMPILE_POPPED(ret, "while body", node->nd_body));
+    ADD_LABEL(ret, next_label);	/* next */
+
+    if (type == NODE_WHILE) {
+	compile_branch_condition(iseq, ret, node->nd_cond,
+				 redo_label, end_label);
+    }
+    else {
+	/* until */
+	compile_branch_condition(iseq, ret, node->nd_cond,
+				 end_label, redo_label);
+    }
+
+    ADD_LABEL(ret, end_label);
+    ADD_ADJUST_RESTORE(ret, adjust_label);
+
+    if (node->nd_state == Qundef) {
+	/* ADD_INSN(ret, line, putundef); */
+	COMPILE_ERROR(ERROR_ARGS "unsupported: putundef");
+	return COMPILE_NG;
+    }
+    else {
+	ADD_INSN(ret, line, putnil);
+    }
+
+    ADD_LABEL(ret, break_label);	/* break */
+
+    if (popped) {
+	ADD_INSN(ret, line, pop);
+    }
+
+    ADD_CATCH_ENTRY(CATCH_TYPE_BREAK, redo_label, break_label, NULL,
+		    break_label);
+    ADD_CATCH_ENTRY(CATCH_TYPE_NEXT, redo_label, break_label, NULL,
+		    next_catch_label);
+    ADD_CATCH_ENTRY(CATCH_TYPE_REDO, redo_label, break_label, NULL,
+		    ISEQ_COMPILE_DATA(iseq)->redo_label);
+
+    ISEQ_COMPILE_DATA(iseq)->start_label = prev_start_label;
+    ISEQ_COMPILE_DATA(iseq)->end_label = prev_end_label;
+    ISEQ_COMPILE_DATA(iseq)->redo_label = prev_redo_label;
+    ISEQ_COMPILE_DATA(iseq)->loopval_popped = prev_loopval_popped;
+    ISEQ_COMPILE_DATA(iseq)->ensure_node_stack = ISEQ_COMPILE_DATA(iseq)->ensure_node_stack->prev;
+    return COMPILE_OK;
+}
+
+static int
+compile_iter(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+    const rb_iseq_t *prevblock = ISEQ_COMPILE_DATA(iseq)->current_block;
+    LABEL *retry_label = NEW_LABEL(line);
+    LABEL *retry_end_l = NEW_LABEL(line);
+    const rb_iseq_t *child_iseq;
+
+    ADD_LABEL(ret, retry_label);
+    if (nd_type(node) == NODE_FOR) {
+	CHECK(COMPILE(ret, "iter caller (for)", node->nd_iter));
+
+	ISEQ_COMPILE_DATA(iseq)->current_block = child_iseq =
+	    NEW_CHILD_ISEQ(node->nd_body, make_name_for_block(iseq),
+			   ISEQ_TYPE_BLOCK, line);
+	ADD_SEND_WITH_BLOCK(ret, line, idEach, INT2FIX(0), child_iseq);
+    }
+    else {
+	ISEQ_COMPILE_DATA(iseq)->current_block = child_iseq =
+	    NEW_CHILD_ISEQ(node->nd_body, make_name_for_block(iseq),
+			   ISEQ_TYPE_BLOCK, line);
+	CHECK(COMPILE(ret, "iter caller", node->nd_iter));
+    }
+    ADD_LABEL(ret, retry_end_l);
+
+    if (popped) {
+	ADD_INSN(ret, line, pop);
+    }
+
+    ISEQ_COMPILE_DATA(iseq)->current_block = prevblock;
+
+    ADD_CATCH_ENTRY(CATCH_TYPE_BREAK, retry_label, retry_end_l, child_iseq, retry_end_l);
+    return COMPILE_OK;
+}
+
+static int
+compile_for_masgn(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    /* massign to var in "for"
+     * (args.length == 1 && Array.try_convert(args[0])) || args
+     */
+    const int line = nd_line(node);
+    const NODE *var = node->nd_var;
+    LABEL *not_single = NEW_LABEL(nd_line(var));
+    LABEL *not_ary = NEW_LABEL(nd_line(var));
+    CHECK(COMPILE(ret, "for var", var));
+    ADD_INSN(ret, line, dup);
+    ADD_CALL(ret, line, idLength, INT2FIX(0));
+    ADD_INSN1(ret, line, putobject, INT2FIX(1));
+    ADD_CALL(ret, line, idEq, INT2FIX(1));
+    ADD_INSNL(ret, line, branchunless, not_single);
+    ADD_INSN(ret, line, dup);
+    ADD_INSN1(ret, line, putobject, INT2FIX(0));
+    ADD_CALL(ret, line, idAREF, INT2FIX(1));
+    ADD_INSN1(ret, line, putobject, rb_cArray);
+    ADD_INSN(ret, line, swap);
+    ADD_CALL(ret, line, rb_intern("try_convert"), INT2FIX(1));
+    ADD_INSN(ret, line, dup);
+    ADD_INSNL(ret, line, branchunless, not_ary);
+    ADD_INSN(ret, line, swap);
+    ADD_LABEL(ret, not_ary);
+    ADD_INSN(ret, line, pop);
+    ADD_LABEL(ret, not_single);
+    return COMPILE_OK;
+}
+
+static int
+compile_break(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+    unsigned long throw_flag = 0;
+
+    if (ISEQ_COMPILE_DATA(iseq)->redo_label != 0) {
+	/* while/until */
+	LABEL *splabel = NEW_LABEL(0);
+	ADD_LABEL(ret, splabel);
+	ADD_ADJUST(ret, line, ISEQ_COMPILE_DATA(iseq)->redo_label);
+	CHECK(COMPILE_(ret, "break val (while/until)", node->nd_stts,
+		       ISEQ_COMPILE_DATA(iseq)->loopval_popped));
+	add_ensure_iseq(ret, iseq, 0);
+	ADD_INSNL(ret, line, jump, ISEQ_COMPILE_DATA(iseq)->end_label);
+	ADD_ADJUST_RESTORE(ret, splabel);
+
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	}
+    }
+    else if (iseq->body->type == ISEQ_TYPE_BLOCK) {
+      break_by_insn:
+	/* escape from block */
+	CHECK(COMPILE(ret, "break val (block)", node->nd_stts));
+        ADD_INSN1(ret, line, throw, INT2FIX(throw_flag | TAG_BREAK));
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+    }
+    else if (iseq->body->type == ISEQ_TYPE_EVAL) {
+      break_in_eval:
+	COMPILE_ERROR(ERROR_ARGS "Can't escape from eval with break");
+	return COMPILE_NG;
+    }
+    else {
+	const rb_iseq_t *ip = iseq->body->parent_iseq;
+
+	while (ip) {
+	    if (!ISEQ_COMPILE_DATA(ip)) {
+		ip = 0;
+		break;
+	    }
+
+	    if (ISEQ_COMPILE_DATA(ip)->redo_label != 0) {
+                throw_flag = VM_THROW_NO_ESCAPE_FLAG;
+		goto break_by_insn;
+	    }
+	    else if (ip->body->type == ISEQ_TYPE_BLOCK) {
+		goto break_by_insn;
+	    }
+	    else if (ip->body->type == ISEQ_TYPE_EVAL) {
+		goto break_in_eval;
+	    }
+
+	    ip = ip->body->parent_iseq;
+	}
+	COMPILE_ERROR(ERROR_ARGS "Invalid break");
+	return COMPILE_NG;
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_next(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+    unsigned long throw_flag = 0;
+
+    if (ISEQ_COMPILE_DATA(iseq)->redo_label != 0) {
+	LABEL *splabel = NEW_LABEL(0);
+	debugs("next in while loop\n");
+	ADD_LABEL(ret, splabel);
+	CHECK(COMPILE(ret, "next val/valid syntax?", node->nd_stts));
+	add_ensure_iseq(ret, iseq, 0);
+	ADD_ADJUST(ret, line, ISEQ_COMPILE_DATA(iseq)->redo_label);
+	ADD_INSNL(ret, line, jump, ISEQ_COMPILE_DATA(iseq)->start_label);
+	ADD_ADJUST_RESTORE(ret, splabel);
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	}
+    }
+    else if (ISEQ_COMPILE_DATA(iseq)->end_label) {
+	LABEL *splabel = NEW_LABEL(0);
+	debugs("next in block\n");
+	ADD_LABEL(ret, splabel);
+	ADD_ADJUST(ret, line, ISEQ_COMPILE_DATA(iseq)->start_label);
+	CHECK(COMPILE(ret, "next val", node->nd_stts));
+	add_ensure_iseq(ret, iseq, 0);
+	ADD_INSNL(ret, line, jump, ISEQ_COMPILE_DATA(iseq)->end_label);
+	ADD_ADJUST_RESTORE(ret, splabel);
+	splabel->unremovable = FALSE;
+
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	}
+    }
+    else if (iseq->body->type == ISEQ_TYPE_EVAL) {
+      next_in_eval:
+	COMPILE_ERROR(ERROR_ARGS "Can't escape from eval with next");
+	return COMPILE_NG;
+    }
+    else {
+	const rb_iseq_t *ip = iseq;
+
+	while (ip) {
+	    if (!ISEQ_COMPILE_DATA(ip)) {
+		ip = 0;
+		break;
+	    }
+
+            throw_flag = VM_THROW_NO_ESCAPE_FLAG;
+	    if (ISEQ_COMPILE_DATA(ip)->redo_label != 0) {
+		/* while loop */
+		break;
+	    }
+	    else if (ip->body->type == ISEQ_TYPE_BLOCK) {
+		break;
+	    }
+	    else if (ip->body->type == ISEQ_TYPE_EVAL) {
+		goto next_in_eval;
+	    }
+
+	    ip = ip->body->parent_iseq;
+	}
+	if (ip != 0) {
+	    CHECK(COMPILE(ret, "next val", node->nd_stts));
+            ADD_INSN1(ret, line, throw, INT2FIX(throw_flag | TAG_NEXT));
+
+	    if (popped) {
+		ADD_INSN(ret, line, pop);
+	    }
+	}
+	else {
+	    COMPILE_ERROR(ERROR_ARGS "Invalid next");
+	    return COMPILE_NG;
+	}
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_redo(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+
+    if (ISEQ_COMPILE_DATA(iseq)->redo_label) {
+	LABEL *splabel = NEW_LABEL(0);
+	debugs("redo in while");
+	ADD_LABEL(ret, splabel);
+	ADD_ADJUST(ret, line, ISEQ_COMPILE_DATA(iseq)->redo_label);
+	add_ensure_iseq(ret, iseq, 0);
+	ADD_INSNL(ret, line, jump, ISEQ_COMPILE_DATA(iseq)->redo_label);
+	ADD_ADJUST_RESTORE(ret, splabel);
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	}
+    }
+    else if (iseq->body->type == ISEQ_TYPE_EVAL) {
+      redo_in_eval:
+	COMPILE_ERROR(ERROR_ARGS "Can't escape from eval with redo");
+	return COMPILE_NG;
+    }
+    else if (ISEQ_COMPILE_DATA(iseq)->start_label) {
+	LABEL *splabel = NEW_LABEL(0);
+
+	debugs("redo in block");
+	ADD_LABEL(ret, splabel);
+	add_ensure_iseq(ret, iseq, 0);
+	ADD_ADJUST(ret, line, ISEQ_COMPILE_DATA(iseq)->start_label);
+	ADD_INSNL(ret, line, jump, ISEQ_COMPILE_DATA(iseq)->start_label);
+	ADD_ADJUST_RESTORE(ret, splabel);
+
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	}
+    }
+    else {
+	const rb_iseq_t *ip = iseq;
+
+	while (ip) {
+	    if (!ISEQ_COMPILE_DATA(ip)) {
+		ip = 0;
+		break;
+	    }
+
+	    if (ISEQ_COMPILE_DATA(ip)->redo_label != 0) {
+		break;
+	    }
+	    else if (ip->body->type == ISEQ_TYPE_BLOCK) {
+		break;
+	    }
+	    else if (ip->body->type == ISEQ_TYPE_EVAL) {
+		goto redo_in_eval;
+	    }
+
+	    ip = ip->body->parent_iseq;
+	}
+	if (ip != 0) {
+	    ADD_INSN(ret, line, putnil);
+            ADD_INSN1(ret, line, throw, INT2FIX(VM_THROW_NO_ESCAPE_FLAG | TAG_REDO));
+
+	    if (popped) {
+		ADD_INSN(ret, line, pop);
+	    }
+	}
+	else {
+	    COMPILE_ERROR(ERROR_ARGS "Invalid redo");
+	    return COMPILE_NG;
+	}
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_retry(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+
+    if (iseq->body->type == ISEQ_TYPE_RESCUE) {
+	ADD_INSN(ret, line, putnil);
+	ADD_INSN1(ret, line, throw, INT2FIX(TAG_RETRY));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+    }
+    else {
+	COMPILE_ERROR(ERROR_ARGS "Invalid retry");
+	return COMPILE_NG;
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_rescue(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+    LABEL *lstart = NEW_LABEL(line);
+    LABEL *lend = NEW_LABEL(line);
+    LABEL *lcont = NEW_LABEL(line);
+    const rb_iseq_t *rescue = NEW_CHILD_ISEQ(node->nd_resq,
+					     rb_str_concat(rb_str_new2("rescue in "), iseq->body->location.label),
+					     ISEQ_TYPE_RESCUE, line);
+
+    lstart->rescued = LABEL_RESCUE_BEG;
+    lend->rescued = LABEL_RESCUE_END;
+    ADD_LABEL(ret, lstart);
+    CHECK(COMPILE(ret, "rescue head", node->nd_head));
+    ADD_LABEL(ret, lend);
+    if (node->nd_else) {
+	ADD_INSN(ret, line, pop);
+	CHECK(COMPILE(ret, "rescue else", node->nd_else));
+    }
+    ADD_INSN(ret, line, nop);
+    ADD_LABEL(ret, lcont);
+
+    if (popped) {
+	ADD_INSN(ret, line, pop);
+    }
+
+    /* register catch entry */
+    ADD_CATCH_ENTRY(CATCH_TYPE_RESCUE, lstart, lend, rescue, lcont);
+    ADD_CATCH_ENTRY(CATCH_TYPE_RETRY, lend, lcont, NULL, lstart);
+    return COMPILE_OK;
+}
+
+static int
+compile_resbody(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+    const NODE *resq = node;
+    const NODE *narg;
+    LABEL *label_miss, *label_hit;
+
+    while (resq) {
+	label_miss = NEW_LABEL(line);
+	label_hit = NEW_LABEL(line);
+
+	narg = resq->nd_args;
+	if (narg) {
+	    switch (nd_type(narg)) {
+	      case NODE_ARRAY:
+		while (narg) {
+		    ADD_GETLOCAL(ret, line, LVAR_ERRINFO, 0);
+		    CHECK(COMPILE(ret, "rescue arg", narg->nd_head));
+		    ADD_INSN1(ret, line, checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_RESCUE));
+		    ADD_INSNL(ret, line, branchif, label_hit);
+		    narg = narg->nd_next;
+		}
+		break;
+	      case NODE_SPLAT:
+	      case NODE_ARGSCAT:
+	      case NODE_ARGSPUSH:
+		ADD_GETLOCAL(ret, line, LVAR_ERRINFO, 0);
+		CHECK(COMPILE(ret, "rescue/cond splat", narg));
+		ADD_INSN1(ret, line, checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_RESCUE | VM_CHECKMATCH_ARRAY));
+		ADD_INSNL(ret, line, branchif, label_hit);
+		break;
+	      default:
+		UNKNOWN_NODE("NODE_RESBODY", narg, COMPILE_NG);
+	    }
+	}
+	else {
+	    ADD_GETLOCAL(ret, line, LVAR_ERRINFO, 0);
+	    ADD_INSN1(ret, line, putobject, rb_eStandardError);
+	    ADD_INSN1(ret, line, checkmatch, INT2FIX(VM_CHECKMATCH_TYPE_RESCUE));
+	    ADD_INSNL(ret, line, branchif, label_hit);
+	}
+	ADD_INSNL(ret, line, jump, label_miss);
+	ADD_LABEL(ret, label_hit);
+	CHECK(COMPILE(ret, "resbody body", resq->nd_body));
+	if (ISEQ_COMPILE_DATA(iseq)->option->tailcall_optimization) {
+	    ADD_INSN(ret, line, nop);
+	}
+	ADD_INSN(ret, line, leave);
+	ADD_LABEL(ret, label_miss);
+	resq = resq->nd_head;
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_ensure(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+    DECL_ANCHOR(ensr);
+    const rb_iseq_t *ensure = NEW_CHILD_ISEQ(node->nd_ensr,
+					     rb_str_concat(rb_str_new2 ("ensure in "), iseq->body->location.label),
+					     ISEQ_TYPE_ENSURE, line);
+    LABEL *lstart = NEW_LABEL(line);
+    LABEL *lend = NEW_LABEL(line);
+    LABEL *lcont = NEW_LABEL(line);
+    LINK_ELEMENT *last;
+    int last_leave = 0;
+    struct ensure_range er;
+    struct iseq_compile_data_ensure_node_stack enl;
+    struct ensure_range *erange;
+
+    INIT_ANCHOR(ensr);
+    CHECK(COMPILE_POPPED(ensr, "ensure ensr", node->nd_ensr));
+    last = ensr->last;
+    last_leave = last && IS_INSN(last) && IS_INSN_ID(last, leave);
+
+    er.begin = lstart;
+    er.end = lend;
+    er.next = 0;
+    push_ensure_entry(iseq, &enl, &er, node->nd_ensr);
+
+    ADD_LABEL(ret, lstart);
+    CHECK(COMPILE_(ret, "ensure head", node->nd_head, (popped | last_leave)));
+    ADD_LABEL(ret, lend);
+    ADD_SEQ(ret, ensr);
+    if (!popped && last_leave) ADD_INSN(ret, line, putnil);
+    ADD_LABEL(ret, lcont);
+    if (last_leave) ADD_INSN(ret, line, pop);
+
+    erange = ISEQ_COMPILE_DATA(iseq)->ensure_node_stack->erange;
+    if (lstart->link.next != &lend->link) {
+	while (erange) {
+	    ADD_CATCH_ENTRY(CATCH_TYPE_ENSURE, erange->begin, erange->end,
+			    ensure, lcont);
+	    erange = erange->next;
+	}
+    }
+
+    ISEQ_COMPILE_DATA(iseq)->ensure_node_stack = enl.prev;
+    return COMPILE_OK;
+}
+
+static int
+compile_return(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    const int line = nd_line(node);
+
+    if (iseq) {
+	enum iseq_type type = iseq->body->type;
+	const rb_iseq_t *is = iseq;
+	enum iseq_type t = type;
+	const NODE *retval = node->nd_stts;
+	LABEL *splabel = 0;
+
+	while (t == ISEQ_TYPE_RESCUE || t == ISEQ_TYPE_ENSURE) {
+	    if (!(is = is->body->parent_iseq)) break;
+	    t = is->body->type;
+	}
+	switch (t) {
+	  case ISEQ_TYPE_TOP:
+	  case ISEQ_TYPE_MAIN:
+	    if (is == iseq) {
+		/* plain top-level, leave directly */
+		type = ISEQ_TYPE_METHOD;
+	    }
+	    break;
+	  default:
+	    break;
+	}
+
+	if (type == ISEQ_TYPE_METHOD) {
+	    splabel = NEW_LABEL(0);
+	    ADD_LABEL(ret, splabel);
+	    ADD_ADJUST(ret, line, 0);
+	}
+
+	CHECK(COMPILE(ret, "return nd_stts (return val)", retval));
+
+	if (type == ISEQ_TYPE_METHOD) {
+	    add_ensure_iseq(ret, iseq, 1);
+	    ADD_TRACE(ret, RUBY_EVENT_RETURN);
+	    ADD_INSN(ret, line, leave);
+	    ADD_ADJUST_RESTORE(ret, splabel);
+
+	    if (!popped) {
+		ADD_INSN(ret, line, putnil);
+	    }
+	}
+	else {
+	    ADD_INSN1(ret, line, throw, INT2FIX(TAG_RETURN));
+	    if (popped) {
+		ADD_INSN(ret, line, pop);
+	    }
+	}
+    }
+    return COMPILE_OK;
+}
+
+static int
+compile_evstr(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *const node, int popped)
+{
+    CHECK(COMPILE_(ret, "nd_body", node, popped));
+
+    if (!popped && !all_string_result_p(node)) {
+	const int line = nd_line(node);
+	const unsigned int flag = VM_CALL_FCALL;
+	LABEL *isstr = NEW_LABEL(line);
+	ADD_INSN(ret, line, dup);
+	ADD_INSN1(ret, line, checktype, INT2FIX(T_STRING));
+	ADD_INSNL(ret, line, branchif, isstr);
+	ADD_INSN(ret, line, dup);
+	ADD_SEND_R(ret, line, idTo_s, INT2FIX(0), NULL, INT2FIX(flag), NULL);
+	ADD_INSN(ret, line, tostring);
+	ADD_LABEL(ret, isstr);
+    }
+    return COMPILE_OK;
+}
+
+static int iseq_compile_each0(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *node, int popped);
+/**
+  compile each node
+
+  self:  InstructionSequence
+  node:  Ruby compiled node
+  popped: This node will be popped
+ */
+static int
+iseq_compile_each(rb_iseq_t *iseq, LINK_ANCHOR *ret, const NODE *node, int popped)
+{
+    if (node == 0) {
+	if (!popped) {
+	    int lineno = ISEQ_COMPILE_DATA(iseq)->last_line;
+	    if (lineno == 0) lineno = FIX2INT(rb_iseq_first_lineno(iseq));
+	    debugs("node: NODE_NIL(implicit)\n");
+	    ADD_INSN(ret, lineno, putnil);
+	}
+	return COMPILE_OK;
+    }
+    return iseq_compile_each0(iseq, ret, node, popped);
+}
+
+static int
+iseq_compile_each0(rb_iseq_t *iseq, LINK_ANCHOR *const ret, const NODE *node, int popped)
+{
+    const int line = (int)nd_line(node);
+    const enum node_type type = nd_type(node);
+    struct rb_iseq_constant_body *const body = iseq->body;
+
+    if (ISEQ_COMPILE_DATA(iseq)->last_line == line) {
+	/* ignore */
+    }
+    else {
+	if (node->flags & NODE_FL_NEWLINE) {
+	    int event = RUBY_EVENT_LINE;
+	    ISEQ_COMPILE_DATA(iseq)->last_line = line;
+	    if (ISEQ_COVERAGE(iseq) && ISEQ_LINE_COVERAGE(iseq)) {
+		event |= RUBY_EVENT_COVERAGE_LINE;
+	    }
+	    ADD_TRACE(ret, event);
+	}
+    }
+
+    debug_node_start(node);
+#undef BEFORE_RETURN
+#define BEFORE_RETURN debug_node_end()
+
+    switch (type) {
+      case NODE_BLOCK:{
+	while (node && nd_type(node) == NODE_BLOCK) {
+	    CHECK(COMPILE_(ret, "BLOCK body", node->nd_head,
+			   (node->nd_next ? 1 : popped)));
+	    node = node->nd_next;
+	}
+	if (node) {
+	    CHECK(COMPILE_(ret, "BLOCK next", node->nd_next, popped));
+	}
+	break;
+      }
+      case NODE_IF:
+      case NODE_UNLESS:
+	CHECK(compile_if(iseq, ret, node, popped, type));
+	break;
+      case NODE_CASE:
+	CHECK(compile_case(iseq, ret, node, popped));
+	break;
+      case NODE_CASE2:
+	CHECK(compile_case2(iseq, ret, node, popped));
+	break;
+      case NODE_WHILE:
+      case NODE_UNTIL:
+	CHECK(compile_loop(iseq, ret, node, popped, type));
+	break;
+      case NODE_FOR:
+      case NODE_ITER:
+	CHECK(compile_iter(iseq, ret, node, popped));
+	break;
+      case NODE_FOR_MASGN:
+	CHECK(compile_for_masgn(iseq, ret, node, popped));
+	break;
+      case NODE_BREAK:
+	CHECK(compile_break(iseq, ret, node, popped));
+	break;
+      case NODE_NEXT:
+	CHECK(compile_next(iseq, ret, node, popped));
+	break;
+      case NODE_REDO:
+	CHECK(compile_redo(iseq, ret, node, popped));
+	break;
+      case NODE_RETRY:
+	CHECK(compile_retry(iseq, ret, node, popped));
+	break;
+      case NODE_BEGIN:{
+	CHECK(COMPILE_(ret, "NODE_BEGIN", node->nd_body, popped));
+	break;
+      }
+      case NODE_RESCUE:
+	CHECK(compile_rescue(iseq, ret, node, popped));
+	break;
+      case NODE_RESBODY:
+	CHECK(compile_resbody(iseq, ret, node, popped));
+	break;
+      case NODE_ENSURE:
+	CHECK(compile_ensure(iseq, ret, node, popped));
+	break;
+
+      case NODE_AND:
+      case NODE_OR:{
+	LABEL *end_label = NEW_LABEL(line);
+	CHECK(COMPILE(ret, "nd_1st", node->nd_1st));
+	if (!popped) {
+	    ADD_INSN(ret, line, dup);
+	}
+	if (type == NODE_AND) {
+	    ADD_INSNL(ret, line, branchunless, end_label);
+	}
+	else {
+	    ADD_INSNL(ret, line, branchif, end_label);
+	}
+	if (!popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	CHECK(COMPILE_(ret, "nd_2nd", node->nd_2nd, popped));
+	ADD_LABEL(ret, end_label);
+	break;
+      }
+
+      case NODE_MASGN:{
+	compile_massign(iseq, ret, node, popped);
+	break;
+      }
+
+      case NODE_LASGN:{
+	ID id = node->nd_vid;
+	int idx = body->local_iseq->body->local_table_size - get_local_var_idx(iseq, id);
+
+	debugs("lvar: %s idx: %d\n", rb_id2name(id), idx);
+	CHECK(COMPILE(ret, "rvalue", node->nd_value));
+
+	if (!popped) {
+	    ADD_INSN(ret, line, dup);
+	}
+	ADD_SETLOCAL(ret, line, idx, get_lvar_level(iseq));
+	break;
+      }
+      case NODE_DASGN:
+      case NODE_DASGN_CURR:{
+	int idx, lv, ls;
+	ID id = node->nd_vid;
+	CHECK(COMPILE(ret, "dvalue", node->nd_value));
+	debugi("dassn id", rb_id2str(id) ? id : '*');
+
+	if (!popped) {
+	    ADD_INSN(ret, line, dup);
+	}
+
+	idx = get_dyna_var_idx(iseq, id, &lv, &ls);
+
+	if (idx < 0) {
+	    COMPILE_ERROR(ERROR_ARGS "NODE_DASGN(_CURR): unknown id (%"PRIsVALUE")",
+			  rb_id2str(id));
+	    goto ng;
+	}
+	ADD_SETLOCAL(ret, line, ls - idx, lv);
+	break;
+      }
+      case NODE_GASGN:{
+	CHECK(COMPILE(ret, "lvalue", node->nd_value));
+
+	if (!popped) {
+	    ADD_INSN(ret, line, dup);
+	}
+	ADD_INSN1(ret, line, setglobal,
+		  ((VALUE)node->nd_entry | 1));
+	break;
+      }
+      case NODE_IASGN:{
+	CHECK(COMPILE(ret, "lvalue", node->nd_value));
+	if (!popped) {
+	    ADD_INSN(ret, line, dup);
+	}
+	ADD_INSN2(ret, line, setinstancevariable,
+		  ID2SYM(node->nd_vid),
+		  get_ivar_ic_value(iseq,node->nd_vid));
+	break;
+      }
+      case NODE_CDECL:{
+	CHECK(COMPILE(ret, "lvalue", node->nd_value));
+
+	if (!popped) {
+	    ADD_INSN(ret, line, dup);
+	}
+
+	if (node->nd_vid) {
+	    ADD_INSN1(ret, line, putspecialobject,
+		      INT2FIX(VM_SPECIAL_OBJECT_CONST_BASE));
+	    ADD_INSN1(ret, line, setconstant, ID2SYM(node->nd_vid));
+	}
+	else {
+	    compile_cpath(ret, iseq, node->nd_else);
+	    ADD_INSN1(ret, line, setconstant, ID2SYM(node->nd_else->nd_mid));
+	}
+	break;
+      }
+      case NODE_CVASGN:{
+	CHECK(COMPILE(ret, "cvasgn val", node->nd_value));
+	if (!popped) {
+	    ADD_INSN(ret, line, dup);
+	}
+	ADD_INSN1(ret, line, setclassvariable,
+		  ID2SYM(node->nd_vid));
+	break;
+      }
+      case NODE_OP_ASGN1: {
+	DECL_ANCHOR(args);
+	VALUE argc;
+	unsigned int flag = 0;
+	unsigned int asgnflag = 0;
+	ID id = node->nd_mid;
+	int boff = 0;
+
+	/*
+	 * a[x] (op)= y
+	 *
+	 * nil       # nil
+	 * eval a    # nil a
+	 * eval x    # nil a x
+	 * dupn 2    # nil a x a x
+	 * send :[]  # nil a x a[x]
+	 * eval y    # nil a x a[x] y
+	 * send op   # nil a x ret
+	 * setn 3    # ret a x ret
+	 * send []=  # ret ?
+	 * pop       # ret
+	 */
+
+	/*
+	 * nd_recv[nd_args->nd_body] (nd_mid)= nd_args->nd_head;
+	 * NODE_OP_ASGN nd_recv
+	 *              nd_args->nd_head
+	 *              nd_args->nd_body
+	 *              nd_mid
+	 */
+
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	}
+	asgnflag = COMPILE_RECV(ret, "NODE_OP_ASGN1 recv", node);
+	switch (nd_type(node->nd_args->nd_head)) {
+	  case NODE_ZARRAY:
+	    argc = INT2FIX(0);
+	    break;
+	  case NODE_BLOCK_PASS:
+	    boff = 1;
+	  default:
+	    INIT_ANCHOR(args);
+	    argc = setup_args(iseq, args, node->nd_args->nd_head, &flag, NULL);
+	    CHECK(!NIL_P(argc));
+	    ADD_SEQ(ret, args);
+	}
+	ADD_INSN1(ret, line, dupn, FIXNUM_INC(argc, 1 + boff));
+	ADD_SEND_WITH_FLAG(ret, line, idAREF, argc, INT2FIX(flag));
+	flag |= asgnflag;
+
+	if (id == idOROP || id == idANDOP) {
+	    /* a[x] ||= y  or  a[x] &&= y
+
+	       unless/if a[x]
+	       a[x]= y
+	       else
+	       nil
+	       end
+	    */
+	    LABEL *label = NEW_LABEL(line);
+	    LABEL *lfin = NEW_LABEL(line);
+
+	    ADD_INSN(ret, line, dup);
+	    if (id == idOROP) {
+		ADD_INSNL(ret, line, branchif, label);
+	    }
+	    else { /* idANDOP */
+		ADD_INSNL(ret, line, branchunless, label);
+	    }
+	    ADD_INSN(ret, line, pop);
+
+	    CHECK(COMPILE(ret, "NODE_OP_ASGN1 args->body: ", node->nd_args->nd_body));
+	    if (!popped) {
+		ADD_INSN1(ret, line, setn, FIXNUM_INC(argc, 2+boff));
+	    }
+	    if (flag & VM_CALL_ARGS_SPLAT) {
+		ADD_INSN1(ret, line, newarray, INT2FIX(1));
+		if (boff > 0) {
+		    ADD_INSN1(ret, line, dupn, INT2FIX(3));
+		    ADD_INSN(ret, line, swap);
+		    ADD_INSN(ret, line, pop);
+		}
+		ADD_INSN(ret, line, concatarray);
+		if (boff > 0) {
+		    ADD_INSN1(ret, line, setn, INT2FIX(3));
+		    ADD_INSN(ret, line, pop);
+		    ADD_INSN(ret, line, pop);
+		}
+		ADD_SEND_WITH_FLAG(ret, line, idASET, argc, INT2FIX(flag));
+	    }
+	    else {
+		if (boff > 0)
+		    ADD_INSN(ret, line, swap);
+		ADD_SEND_WITH_FLAG(ret, line, idASET, FIXNUM_INC(argc, 1), INT2FIX(flag));
+	    }
+	    ADD_INSN(ret, line, pop);
+	    ADD_INSNL(ret, line, jump, lfin);
+	    ADD_LABEL(ret, label);
+	    if (!popped) {
+		ADD_INSN1(ret, line, setn, FIXNUM_INC(argc, 2+boff));
+	    }
+	    ADD_INSN1(ret, line, adjuststack, FIXNUM_INC(argc, 2+boff));
+	    ADD_LABEL(ret, lfin);
+	}
+	else {
+	    CHECK(COMPILE(ret, "NODE_OP_ASGN1 args->body: ", node->nd_args->nd_body));
+	    ADD_SEND(ret, line, id, INT2FIX(1));
+	    if (!popped) {
+		ADD_INSN1(ret, line, setn, FIXNUM_INC(argc, 2+boff));
+	    }
+	    if (flag & VM_CALL_ARGS_SPLAT) {
+		ADD_INSN1(ret, line, newarray, INT2FIX(1));
+		if (boff > 0) {
+		    ADD_INSN1(ret, line, dupn, INT2FIX(3));
+		    ADD_INSN(ret, line, swap);
+		    ADD_INSN(ret, line, pop);
+		}
+		ADD_INSN(ret, line, concatarray);
+		if (boff > 0) {
+		    ADD_INSN1(ret, line, setn, INT2FIX(3));
+		    ADD_INSN(ret, line, pop);
+		    ADD_INSN(ret, line, pop);
+		}
+		ADD_SEND_WITH_FLAG(ret, line, idASET, argc, INT2FIX(flag));
+	    }
+	    else {
+		if (boff > 0)
+		    ADD_INSN(ret, line, swap);
+		ADD_SEND_WITH_FLAG(ret, line, idASET, FIXNUM_INC(argc, 1), INT2FIX(flag));
+	    }
+	    ADD_INSN(ret, line, pop);
+	}
+
+	break;
+      }
+      case NODE_OP_ASGN2:{
+	ID atype = node->nd_next->nd_mid;
+	ID vid = node->nd_next->nd_vid, aid = rb_id_attrset(vid);
+	VALUE asgnflag;
+	LABEL *lfin = NEW_LABEL(line);
+	LABEL *lcfin = NEW_LABEL(line);
+	LABEL *lskip = 0;
+	/*
+	  class C; attr_accessor :c; end
+	  r = C.new
+	  r.a &&= v # asgn2
+
+	  eval r    # r
+	  dup       # r r
+	  eval r.a  # r o
+
+	  # or
+	  dup       # r o o
+	  if lcfin  # r o
+	  pop       # r
+	  eval v    # r v
+	  swap      # v r
+	  topn 1    # v r v
+	  send a=   # v ?
+	  jump lfin # v ?
+
+	  lcfin:      # r o
+	  swap      # o r
+
+	  lfin:       # o ?
+	  pop       # o
+
+	  # and
+	  dup       # r o o
+	  unless lcfin
+	  pop       # r
+	  eval v    # r v
+	  swap      # v r
+	  topn 1    # v r v
+	  send a=   # v ?
+	  jump lfin # v ?
+
+	  # others
+	  eval v    # r o v
+	  send ??   # r w
+	  send a=   # w
+
+	*/
+
+	asgnflag = COMPILE_RECV(ret, "NODE_OP_ASGN2#recv", node);
+	if (node->nd_next->nd_aid) {
+	    lskip = NEW_LABEL(line);
+	    ADD_INSN(ret, line, dup);
+	    ADD_INSNL(ret, line, branchnil, lskip);
+	}
+	ADD_INSN(ret, line, dup);
+	ADD_SEND(ret, line, vid, INT2FIX(0));
+
+	if (atype == idOROP || atype == idANDOP) {
+	    ADD_INSN(ret, line, dup);
+	    if (atype == idOROP) {
+		ADD_INSNL(ret, line, branchif, lcfin);
+	    }
+	    else { /* idANDOP */
+		ADD_INSNL(ret, line, branchunless, lcfin);
+	    }
+	    ADD_INSN(ret, line, pop);
+	    CHECK(COMPILE(ret, "NODE_OP_ASGN2 val", node->nd_value));
+	    ADD_INSN(ret, line, swap);
+	    ADD_INSN1(ret, line, topn, INT2FIX(1));
+	    ADD_SEND_WITH_FLAG(ret, line, aid, INT2FIX(1), INT2FIX(asgnflag));
+	    ADD_INSNL(ret, line, jump, lfin);
+
+	    ADD_LABEL(ret, lcfin);
+	    ADD_INSN(ret, line, swap);
+
+	    ADD_LABEL(ret, lfin);
+	    ADD_INSN(ret, line, pop);
+	    if (lskip) {
+		ADD_LABEL(ret, lskip);
+	    }
+	    if (popped) {
+		/* we can apply more optimize */
+		ADD_INSN(ret, line, pop);
+	    }
+	}
+	else {
+	    CHECK(COMPILE(ret, "NODE_OP_ASGN2 val", node->nd_value));
+	    ADD_SEND(ret, line, atype, INT2FIX(1));
+	    if (!popped) {
+		ADD_INSN(ret, line, swap);
+		ADD_INSN1(ret, line, topn, INT2FIX(1));
+	    }
+	    ADD_SEND_WITH_FLAG(ret, line, aid, INT2FIX(1), INT2FIX(asgnflag));
+	    if (lskip && popped) {
+		ADD_LABEL(ret, lskip);
+	    }
+	    ADD_INSN(ret, line, pop);
+	    if (lskip && !popped) {
+		ADD_LABEL(ret, lskip);
+	    }
+	}
+	break;
+      }
+      case NODE_OP_CDECL: {
+	LABEL *lfin = 0;
+	LABEL *lassign = 0;
+	ID mid;
+
+	switch (nd_type(node->nd_head)) {
+	  case NODE_COLON3:
+	    ADD_INSN1(ret, line, putobject, rb_cObject);
+	    break;
+	  case NODE_COLON2:
+	    CHECK(COMPILE(ret, "NODE_OP_CDECL/colon2#nd_head", node->nd_head->nd_head));
+	    break;
+	  default:
+	    COMPILE_ERROR(ERROR_ARGS "%s: invalid node in NODE_OP_CDECL",
+			  ruby_node_name(nd_type(node->nd_head)));
+	    goto ng;
+	}
+	mid = node->nd_head->nd_mid;
+	/* cref */
+	if (node->nd_aid == idOROP) {
+	    lassign = NEW_LABEL(line);
+	    ADD_INSN(ret, line, dup); /* cref cref */
+	    ADD_INSN3(ret, line, defined, INT2FIX(DEFINED_CONST),
+		      ID2SYM(mid), Qfalse); /* cref bool */
+	    ADD_INSNL(ret, line, branchunless, lassign); /* cref */
+	}
+	ADD_INSN(ret, line, dup); /* cref cref */
+	ADD_INSN1(ret, line, getconstant, ID2SYM(mid)); /* cref obj */
+
+	if (node->nd_aid == idOROP || node->nd_aid == idANDOP) {
+	    lfin = NEW_LABEL(line);
+	    if (!popped) ADD_INSN(ret, line, dup); /* cref [obj] obj */
+	    if (node->nd_aid == idOROP)
+		ADD_INSNL(ret, line, branchif, lfin);
+	    else /* idANDOP */
+		ADD_INSNL(ret, line, branchunless, lfin);
+	    /* cref [obj] */
+	    if (!popped) ADD_INSN(ret, line, pop); /* cref */
+	    if (lassign) ADD_LABEL(ret, lassign);
+	    CHECK(COMPILE(ret, "NODE_OP_CDECL#nd_value", node->nd_value));
+	    /* cref value */
+	    if (popped)
+		ADD_INSN1(ret, line, topn, INT2FIX(1)); /* cref value cref */
+	    else {
+		ADD_INSN1(ret, line, dupn, INT2FIX(2)); /* cref value cref value */
+		ADD_INSN(ret, line, swap); /* cref value value cref */
+	    }
+	    ADD_INSN1(ret, line, setconstant, ID2SYM(mid)); /* cref [value] */
+	    ADD_LABEL(ret, lfin);			    /* cref [value] */
+	    if (!popped) ADD_INSN(ret, line, swap); /* [value] cref */
+	    ADD_INSN(ret, line, pop); /* [value] */
+	}
+	else {
+	    CHECK(COMPILE(ret, "NODE_OP_CDECL#nd_value", node->nd_value));
+	    /* cref obj value */
+	    ADD_CALL(ret, line, node->nd_aid, INT2FIX(1));
+	    /* cref value */
+	    ADD_INSN(ret, line, swap); /* value cref */
+	    if (!popped) {
+		ADD_INSN1(ret, line, topn, INT2FIX(1)); /* value cref value */
+		ADD_INSN(ret, line, swap); /* value value cref */
+	    }
+	    ADD_INSN1(ret, line, setconstant, ID2SYM(mid));
+	}
+	break;
+      }
+      case NODE_OP_ASGN_AND:
+      case NODE_OP_ASGN_OR:{
+	LABEL *lfin = NEW_LABEL(line);
+	LABEL *lassign;
+
+	if (nd_type(node) == NODE_OP_ASGN_OR) {
+	    LABEL *lfinish[2];
+	    lfinish[0] = lfin;
+	    lfinish[1] = 0;
+	    defined_expr(iseq, ret, node->nd_head, lfinish, Qfalse);
+	    lassign = lfinish[1];
+	    if (!lassign) {
+		lassign = NEW_LABEL(line);
+	    }
+	    ADD_INSNL(ret, line, branchunless, lassign);
+	}
+	else {
+	    lassign = NEW_LABEL(line);
+	}
+
+	CHECK(COMPILE(ret, "NODE_OP_ASGN_AND/OR#nd_head", node->nd_head));
+	ADD_INSN(ret, line, dup);
+
+	if (nd_type(node) == NODE_OP_ASGN_AND) {
+	    ADD_INSNL(ret, line, branchunless, lfin);
+	}
+	else {
+	    ADD_INSNL(ret, line, branchif, lfin);
+	}
+
+	ADD_INSN(ret, line, pop);
+	ADD_LABEL(ret, lassign);
+	CHECK(COMPILE(ret, "NODE_OP_ASGN_AND/OR#nd_value", node->nd_value));
+	ADD_LABEL(ret, lfin);
+
+	if (popped) {
+	    /* we can apply more optimize */
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_CALL:
+      case NODE_OPCALL:
+	/* optimization shortcut
+	 *   "literal".freeze -> opt_str_freeze("literal")
+	 */
+	if (node->nd_recv && nd_type(node->nd_recv) == NODE_STR &&
+	    (node->nd_mid == idFreeze || node->nd_mid == idUMinus) &&
+	    node->nd_args == NULL &&
+	    ISEQ_COMPILE_DATA(iseq)->current_block == NULL &&
+	    ISEQ_COMPILE_DATA(iseq)->option->specialized_instruction) {
+	    VALUE str = freeze_literal(iseq, node->nd_recv->nd_lit);
+	    if (node->nd_mid == idUMinus) {
+		ADD_INSN3(ret, line, opt_str_uminus, str,
+                          new_callinfo(iseq, idUMinus, 0, 0, NULL, FALSE),
+                          Qundef /* CALL_CACHE */);
+	    }
+	    else {
+		ADD_INSN3(ret, line, opt_str_freeze, str,
+                          new_callinfo(iseq, idFreeze, 0, 0, NULL, FALSE),
+                          Qundef /* CALL_CACHE */);
+	    }
+	    if (popped) {
+		ADD_INSN(ret, line, pop);
+	    }
+	    break;
+	}
+	/* optimization shortcut
+	 *   obj["literal"] -> opt_aref_with(obj, "literal")
+	 */
+	if (node->nd_mid == idAREF && !private_recv_p(node) && node->nd_args &&
+	    nd_type(node->nd_args) == NODE_ARRAY && node->nd_args->nd_alen == 1 &&
+	    nd_type(node->nd_args->nd_head) == NODE_STR &&
+	    ISEQ_COMPILE_DATA(iseq)->current_block == NULL &&
+            !ISEQ_COMPILE_DATA(iseq)->option->frozen_string_literal &&
+	    ISEQ_COMPILE_DATA(iseq)->option->specialized_instruction) {
+	    VALUE str = freeze_literal(iseq, node->nd_args->nd_head->nd_lit);
+	    CHECK(COMPILE(ret, "recv", node->nd_recv));
+	    ADD_INSN3(ret, line, opt_aref_with, str,
+		      new_callinfo(iseq, idAREF, 1, 0, NULL, FALSE),
+		      NULL/* CALL_CACHE */);
+	    if (popped) {
+		ADD_INSN(ret, line, pop);
+	    }
+	    break;
+	}
+      case NODE_QCALL:
+      case NODE_FCALL:
+      case NODE_VCALL:{		/* VCALL: variable or call */
+	/*
+	  call:  obj.method(...)
+	  fcall: func(...)
+	  vcall: func
+	*/
+	DECL_ANCHOR(recv);
+	DECL_ANCHOR(args);
+	LABEL *else_label = 0;
+	LABEL *end_label = 0;
+	VALUE branches = 0;
+	ID mid = node->nd_mid;
+	VALUE argc;
+	unsigned int flag = 0;
+	struct rb_call_info_kw_arg *keywords = NULL;
+	const rb_iseq_t *parent_block = ISEQ_COMPILE_DATA(iseq)->current_block;
+	ISEQ_COMPILE_DATA(iseq)->current_block = NULL;
+
+	INIT_ANCHOR(recv);
+	INIT_ANCHOR(args);
+#if SUPPORT_JOKE
+	if (nd_type(node) == NODE_VCALL) {
+            ID id_bitblt;
+	    ID id_answer;
+
+            CONST_ID(id_bitblt, "bitblt");
+	    CONST_ID(id_answer, "the_answer_to_life_the_universe_and_everything");
+
+            if (mid == id_bitblt) {
+                ADD_INSN(ret, line, bitblt);
+                break;
+            }
+            else if (mid == id_answer) {
+		ADD_INSN(ret, line, answer);
+		break;
+	    }
+	}
+	/* only joke */
+	{
+	    ID goto_id;
+	    ID label_id;
+
+	    CONST_ID(goto_id, "__goto__");
+	    CONST_ID(label_id, "__label__");
+
+	    if (nd_type(node) == NODE_FCALL &&
+		(mid == goto_id || mid == label_id)) {
+		LABEL *label;
+		st_data_t data;
+		st_table *labels_table = ISEQ_COMPILE_DATA(iseq)->labels_table;
+		VALUE label_name;
+
+		if (!labels_table) {
+		    labels_table = st_init_numtable();
+		    ISEQ_COMPILE_DATA(iseq)->labels_table = labels_table;
+		}
+		if (nd_type(node->nd_args->nd_head) == NODE_LIT &&
+		    SYMBOL_P(node->nd_args->nd_head->nd_lit)) {
+
+		    label_name = node->nd_args->nd_head->nd_lit;
+		    if (!st_lookup(labels_table, (st_data_t)label_name, &data)) {
+			label = NEW_LABEL(line);
+			label->position = line;
+			st_insert(labels_table, (st_data_t)label_name, (st_data_t)label);
+		    }
+		    else {
+			label = (LABEL *)data;
+		    }
+		}
+		else {
+		    COMPILE_ERROR(ERROR_ARGS "invalid goto/label format");
+		    goto ng;
+		}
+
+
+		if (mid == goto_id) {
+		    ADD_INSNL(ret, line, jump, label);
+		}
+		else {
+		    ADD_LABEL(ret, label);
+		}
+		break;
+	    }
+	}
+#endif
+	/* receiver */
+	if (type == NODE_CALL || type == NODE_OPCALL || type == NODE_QCALL) {
+	    int idx, level;
+
+	    if (mid == idCall &&
+		nd_type(node->nd_recv) == NODE_LVAR &&
+		iseq_block_param_id_p(iseq, node->nd_recv->nd_vid, &idx, &level)) {
+		ADD_INSN2(recv, nd_line(node->nd_recv), getblockparamproxy, INT2FIX(idx + VM_ENV_DATA_SIZE - 1), INT2FIX(level));
+	    }
+	    else {
+		CHECK(COMPILE(recv, "recv", node->nd_recv));
+	    }
+
+	    if (type == NODE_QCALL) {
+		else_label = NEW_LABEL(line);
+		end_label = NEW_LABEL(line);
+
+		DECL_BRANCH_BASE(branches, nd_first_lineno(node), nd_first_column(node), nd_last_lineno(node), nd_last_column(node), "&.");
+		ADD_INSN(recv, line, dup);
+		ADD_INSNL(recv, line, branchnil, else_label);
+		ADD_TRACE_BRANCH_COVERAGE(recv, nd_first_lineno(node), nd_first_column(node), nd_last_lineno(node), nd_last_column(node), "then", branches);
+	    }
+	}
+	else if (type == NODE_FCALL || type == NODE_VCALL) {
+	    ADD_CALL_RECEIVER(recv, line);
+	}
+
+	/* args */
+	if (type != NODE_VCALL) {
+	    argc = setup_args(iseq, args, node->nd_args, &flag, &keywords);
+	    CHECK(!NIL_P(argc));
+	}
+	else {
+	    argc = INT2FIX(0);
+	}
+
+	ADD_SEQ(ret, recv);
+	ADD_SEQ(ret, args);
+
+	debugp_param("call args argc", argc);
+	debugp_param("call method", ID2SYM(mid));
+
+	switch ((int)type) {
+	  case NODE_VCALL:
+	    flag |= VM_CALL_VCALL;
+	    /* VCALL is funcall, so fall through */
+	  case NODE_FCALL:
+	    flag |= VM_CALL_FCALL;
+	}
+
+	ADD_SEND_R(ret, line, mid, argc, parent_block, INT2FIX(flag), keywords);
+
+	if (else_label && end_label) {
+	    ADD_INSNL(ret, line, jump, end_label);
+	    ADD_LABEL(ret, else_label);
+	    ADD_TRACE_BRANCH_COVERAGE(ret, nd_first_lineno(node), nd_first_column(node), nd_last_lineno(node), nd_last_column(node), "else", branches);
+	    ADD_LABEL(ret, end_label);
+	}
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_SUPER:
+      case NODE_ZSUPER:{
+	DECL_ANCHOR(args);
+	int argc;
+	unsigned int flag = 0;
+	struct rb_call_info_kw_arg *keywords = NULL;
+	const rb_iseq_t *parent_block = ISEQ_COMPILE_DATA(iseq)->current_block;
+
+	INIT_ANCHOR(args);
+	ISEQ_COMPILE_DATA(iseq)->current_block = NULL;
+	if (type == NODE_SUPER) {
+	    VALUE vargc = setup_args(iseq, args, node->nd_args, &flag, &keywords);
+	    CHECK(!NIL_P(vargc));
+	    argc = FIX2INT(vargc);
+	}
+	else {
+	    /* NODE_ZSUPER */
+	    int i;
+	    const rb_iseq_t *liseq = body->local_iseq;
+	    const struct rb_iseq_constant_body *const local_body = liseq->body;
+	    const struct rb_iseq_param_keyword *const local_kwd = local_body->param.keyword;
+	    int lvar_level = get_lvar_level(iseq);
+
+	    argc = local_body->param.lead_num;
+
+	    /* normal arguments */
+	    for (i = 0; i < local_body->param.lead_num; i++) {
+		int idx = local_body->local_table_size - i;
+		ADD_GETLOCAL(args, line, idx, lvar_level);
+	    }
+
+	    if (local_body->param.flags.has_opt) {
+		/* optional arguments */
+		int j;
+		for (j = 0; j < local_body->param.opt_num; j++) {
+		    int idx = local_body->local_table_size - (i + j);
+		    ADD_GETLOCAL(args, line, idx, lvar_level);
+		}
+		i += j;
+		argc = i;
+	    }
+	    if (local_body->param.flags.has_rest) {
+		/* rest argument */
+		int idx = local_body->local_table_size - local_body->param.rest_start;
+
+		ADD_GETLOCAL(args, line, idx, lvar_level);
+		ADD_INSN1(args, line, splatarray, Qfalse);
+
+		argc = local_body->param.rest_start + 1;
+		flag |= VM_CALL_ARGS_SPLAT;
+	    }
+	    if (local_body->param.flags.has_post) {
+		/* post arguments */
+		int post_len = local_body->param.post_num;
+		int post_start = local_body->param.post_start;
+
+		if (local_body->param.flags.has_rest) {
+		    int j;
+		    for (j=0; j<post_len; j++) {
+			int idx = local_body->local_table_size - (post_start + j);
+			ADD_GETLOCAL(args, line, idx, lvar_level);
+		    }
+		    ADD_INSN1(args, line, newarray, INT2FIX(j));
+		    ADD_INSN (args, line, concatarray);
+		    /* argc is settled at above */
+		}
+		else {
+		    int j;
+		    for (j=0; j<post_len; j++) {
+			int idx = local_body->local_table_size - (post_start + j);
+			ADD_GETLOCAL(args, line, idx, lvar_level);
+		    }
+		    argc = post_len + post_start;
+		}
+	    }
+
+	    if (local_body->param.flags.has_kw) { /* TODO: support keywords */
+		int local_size = local_body->local_table_size;
+		argc++;
+
+		ADD_INSN1(args, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+
+		if (local_body->param.flags.has_kwrest) {
+		    int idx = local_body->local_table_size - local_kwd->rest_start;
+		    ADD_GETLOCAL(args, line, idx, lvar_level);
+		    ADD_SEND (args, line, rb_intern("dup"), INT2FIX(0));
+		}
+		else {
+		    ADD_INSN1(args, line, newhash, INT2FIX(0));
+		}
+		for (i = 0; i < local_kwd->num; ++i) {
+		    ID id = local_kwd->table[i];
+		    int idx = local_size - get_local_var_idx(liseq, id);
+		    ADD_INSN1(args, line, putobject, ID2SYM(id));
+		    ADD_GETLOCAL(args, line, idx, lvar_level);
+		}
+		ADD_SEND(args, line, id_core_hash_merge_ptr, INT2FIX(i * 2 + 1));
+		if (local_body->param.flags.has_rest) {
+		    ADD_INSN1(args, line, newarray, INT2FIX(1));
+		    ADD_INSN (args, line, concatarray);
+		    --argc;
+		}
+	    }
+	    else if (local_body->param.flags.has_kwrest) {
+		int idx = local_body->local_table_size - local_kwd->rest_start;
+		ADD_GETLOCAL(args, line, idx, lvar_level);
+
+		ADD_SEND (args, line, rb_intern("dup"), INT2FIX(0));
+		if (local_body->param.flags.has_rest) {
+		    ADD_INSN1(args, line, newarray, INT2FIX(1));
+		    ADD_INSN (args, line, concatarray);
+		}
+		else {
+		    argc++;
+		}
+	    }
+	}
+
+	ADD_INSN(ret, line, putself);
+	ADD_SEQ(ret, args);
+	ADD_INSN3(ret, line, invokesuper,
+		  new_callinfo(iseq, 0, argc, flag | VM_CALL_SUPER | (type == NODE_ZSUPER ? VM_CALL_ZSUPER : 0) | VM_CALL_FCALL, keywords, parent_block != NULL),
+		  Qnil, /* CALL_CACHE */
+		  parent_block);
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_ARRAY:{
+	CHECK(compile_array(iseq, ret, node, COMPILE_ARRAY_TYPE_ARRAY, NULL, NULL, popped) >= 0);
+	break;
+      }
+      case NODE_ZARRAY:{
+	if (!popped) {
+	    ADD_INSN1(ret, line, newarray, INT2FIX(0));
+	}
+	break;
+      }
+      case NODE_VALUES:{
+	const NODE *n = node;
+	if (popped) {
+	    COMPILE_ERROR(ERROR_ARGS "NODE_VALUES: must not be popped");
+	}
+	while (n) {
+	    CHECK(COMPILE(ret, "values item", n->nd_head));
+	    n = n->nd_next;
+	}
+	ADD_INSN1(ret, line, newarray, INT2FIX(node->nd_alen));
+	break;
+      }
+      case NODE_HASH:{
+	DECL_ANCHOR(list);
+	enum node_type type = node->nd_head ? nd_type(node->nd_head) : NODE_ZARRAY;
+
+	INIT_ANCHOR(list);
+	switch (type) {
+	  case NODE_ARRAY:
+	    CHECK(compile_array(iseq, list, node->nd_head, COMPILE_ARRAY_TYPE_HASH, NULL, NULL, popped) >= 0);
+	    ADD_SEQ(ret, list);
+	    break;
+
+	  case NODE_ZARRAY:
+	    if (popped) break;
+	    ADD_INSN1(ret, line, newhash, INT2FIX(0));
+	    break;
+
+	  default:
+	    COMPILE_ERROR(ERROR_ARGS_AT(node->nd_head) "can't make hash with this node: %s",
+			  ruby_node_name(type));
+	    goto ng;
+	}
+	break;
+      }
+      case NODE_RETURN:
+	CHECK(compile_return(iseq, ret, node, popped));
+	break;
+      case NODE_YIELD:{
+	DECL_ANCHOR(args);
+	VALUE argc;
+	unsigned int flag = 0;
+	struct rb_call_info_kw_arg *keywords = NULL;
+
+	INIT_ANCHOR(args);
+	if (body->type == ISEQ_TYPE_TOP ||
+	    body->type == ISEQ_TYPE_MAIN) {
+	    COMPILE_ERROR(ERROR_ARGS "Invalid yield");
+	    goto ng;
+	}
+
+	if (node->nd_head) {
+	    argc = setup_args(iseq, args, node->nd_head, &flag, &keywords);
+	    CHECK(!NIL_P(argc));
+	}
+	else {
+	    argc = INT2FIX(0);
+	}
+
+	ADD_SEQ(ret, args);
+	ADD_INSN1(ret, line, invokeblock, new_callinfo(iseq, 0, FIX2INT(argc), flag, keywords, FALSE));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_LVAR:{
+	if (!popped) {
+	    ID id = node->nd_vid;
+	    int idx = body->local_iseq->body->local_table_size - get_local_var_idx(iseq, id);
+
+	    debugs("id: %s idx: %d\n", rb_id2name(id), idx);
+	    ADD_GETLOCAL(ret, line, idx, get_lvar_level(iseq));
+	}
+	break;
+      }
+      case NODE_DVAR:{
+	int lv, idx, ls;
+	debugi("nd_vid", node->nd_vid);
+	if (!popped) {
+	    idx = get_dyna_var_idx(iseq, node->nd_vid, &lv, &ls);
+	    if (idx < 0) {
+		COMPILE_ERROR(ERROR_ARGS "unknown dvar (%"PRIsVALUE")",
+			      rb_id2str(node->nd_vid));
+		goto ng;
+	    }
+	    ADD_GETLOCAL(ret, line, ls - idx, lv);
+	}
+	break;
+      }
+      case NODE_GVAR:{
+	ADD_INSN1(ret, line, getglobal,
+		  ((VALUE)node->nd_entry | 1));
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_IVAR:{
+	debugi("nd_vid", node->nd_vid);
+	if (!popped) {
+	    ADD_INSN2(ret, line, getinstancevariable,
+		      ID2SYM(node->nd_vid),
+		      get_ivar_ic_value(iseq,node->nd_vid));
+	}
+	break;
+      }
+      case NODE_CONST:{
+	debugi("nd_vid", node->nd_vid);
+
+	if (ISEQ_COMPILE_DATA(iseq)->option->inline_const_cache) {
+	    LABEL *lend = NEW_LABEL(line);
+	    int ic_index = body->is_size++;
+
+            ADD_INSN2(ret, line, opt_getinlinecache, lend, INT2FIX(ic_index));
+	    ADD_INSN1(ret, line, getconstant, ID2SYM(node->nd_vid));
+            ADD_INSN1(ret, line, opt_setinlinecache, INT2FIX(ic_index));
+	    ADD_LABEL(ret, lend);
+	}
+	else {
+	    ADD_INSN(ret, line, putnil);
+	    ADD_INSN1(ret, line, getconstant, ID2SYM(node->nd_vid));
+	}
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_CVAR:{
+	if (!popped) {
+	    ADD_INSN1(ret, line, getclassvariable,
+		      ID2SYM(node->nd_vid));
+	}
+	break;
+      }
+      case NODE_NTH_REF:{
+        if (!popped) {
+	    if (!node->nd_nth) {
+		ADD_INSN(ret, line, putnil);
+		break;
+	    }
+	    ADD_INSN2(ret, line, getspecial, INT2FIX(1) /* '~'  */,
+		      INT2FIX(node->nd_nth << 1));
+	}
+	break;
+      }
+      case NODE_BACK_REF:{
+	if (!popped) {
+	    ADD_INSN2(ret, line, getspecial, INT2FIX(1) /* '~' */,
+		      INT2FIX(0x01 | (node->nd_nth << 1)));
+	}
+	break;
+      }
+      case NODE_MATCH:
+      case NODE_MATCH2:
+      case NODE_MATCH3:{
+	DECL_ANCHOR(recv);
+	DECL_ANCHOR(val);
+
+	INIT_ANCHOR(recv);
+	INIT_ANCHOR(val);
+	switch (nd_type(node)) {
+	  case NODE_MATCH:
+	    ADD_INSN1(recv, line, putobject, node->nd_lit);
+	    ADD_INSN2(val, line, getspecial, INT2FIX(0),
+		      INT2FIX(0));
+	    break;
+	  case NODE_MATCH2:
+	    CHECK(COMPILE(recv, "receiver", node->nd_recv));
+	    CHECK(COMPILE(val, "value", node->nd_value));
+	    break;
+	  case NODE_MATCH3:
+	    CHECK(COMPILE(recv, "receiver", node->nd_value));
+	    CHECK(COMPILE(val, "value", node->nd_recv));
+	    break;
+	}
+
+	if (ISEQ_COMPILE_DATA(iseq)->option->specialized_instruction) {
+	    /* TODO: detect by node */
+	    if (recv->last == recv->anchor.next &&
+		INSN_OF(recv->last) == BIN(putobject) &&
+		nd_type(node) == NODE_MATCH2) {
+		ADD_SEQ(ret, val);
+		ADD_INSN1(ret, line, opt_regexpmatch1,
+			  OPERAND_AT(recv->last, 0));
+	    }
+	    else {
+		ADD_SEQ(ret, recv);
+		ADD_SEQ(ret, val);
+		ADD_INSN2(ret, line, opt_regexpmatch2, new_callinfo(iseq, idEqTilde, 1, 0, NULL, FALSE), Qnil);
+	    }
+	}
+	else {
+	    ADD_SEQ(ret, recv);
+	    ADD_SEQ(ret, val);
+	    ADD_SEND(ret, line, idEqTilde, INT2FIX(1));
+	}
+
+	if (node->nd_args) {
+	    compile_named_capture_assign(iseq, ret, node->nd_args);
+	}
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_LIT:{
+	debugp_param("lit", node->nd_lit);
+	if (!popped) {
+	    ADD_INSN1(ret, line, putobject, node->nd_lit);
+	}
+	break;
+      }
+      case NODE_STR:{
+	debugp_param("nd_lit", node->nd_lit);
+	if (!popped) {
+	    VALUE lit = node->nd_lit;
+	    if (!ISEQ_COMPILE_DATA(iseq)->option->frozen_string_literal) {
+		lit = freeze_literal(iseq, lit);
+		ADD_INSN1(ret, line, putstring, lit);
+	    }
+	    else {
+		if (ISEQ_COMPILE_DATA(iseq)->option->debug_frozen_string_literal || RTEST(ruby_debug)) {
+		    VALUE debug_info = rb_ary_new_from_args(2, rb_iseq_path(iseq), INT2FIX(line));
+		    lit = rb_str_dup(lit);
+		    rb_ivar_set(lit, id_debug_created_info, rb_obj_freeze(debug_info));
+		    lit = rb_str_freeze(lit);
+		}
+		else {
+		    lit = rb_fstring(lit);
+		}
+		ADD_INSN1(ret, line, putobject, lit);
+		iseq_add_mark_object_compile_time(iseq, lit);
+	    }
+	}
+	break;
+      }
+      case NODE_DSTR:{
+	compile_dstr(iseq, ret, node);
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	else {
+	    if (ISEQ_COMPILE_DATA(iseq)->option->frozen_string_literal) {
+		VALUE debug_info = Qnil;
+		if (ISEQ_COMPILE_DATA(iseq)->option->debug_frozen_string_literal || RTEST(ruby_debug)) {
+		    debug_info = rb_ary_new_from_args(2, rb_iseq_path(iseq), INT2FIX(line));
+		    iseq_add_mark_object_compile_time(iseq, rb_obj_freeze(debug_info));
+		}
+		ADD_INSN1(ret, line, freezestring, debug_info);
+	    }
+	}
+	break;
+      }
+      case NODE_XSTR:{
+	ADD_CALL_RECEIVER(ret, line);
+	ADD_INSN1(ret, line, putobject, freeze_literal(iseq, node->nd_lit));
+	ADD_CALL(ret, line, idBackquote, INT2FIX(1));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_DXSTR:{
+	ADD_CALL_RECEIVER(ret, line);
+	compile_dstr(iseq, ret, node);
+	ADD_CALL(ret, line, idBackquote, INT2FIX(1));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_EVSTR:
+	CHECK(compile_evstr(iseq, ret, node->nd_body, popped));
+	break;
+      case NODE_DREGX:{
+	compile_dregx(iseq, ret, node);
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_ONCE:{
+	int ic_index = body->is_size++;
+	const rb_iseq_t *block_iseq;
+	block_iseq = NEW_CHILD_ISEQ(node->nd_body, make_name_for_block(iseq), ISEQ_TYPE_PLAIN, line);
+
+	ADD_INSN2(ret, line, once, block_iseq, INT2FIX(ic_index));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_ARGSCAT:{
+	if (popped) {
+	    CHECK(COMPILE(ret, "argscat head", node->nd_head));
+	    ADD_INSN1(ret, line, splatarray, Qfalse);
+	    ADD_INSN(ret, line, pop);
+	    CHECK(COMPILE(ret, "argscat body", node->nd_body));
+	    ADD_INSN1(ret, line, splatarray, Qfalse);
+	    ADD_INSN(ret, line, pop);
+	}
+	else {
+	    CHECK(COMPILE(ret, "argscat head", node->nd_head));
+	    CHECK(COMPILE(ret, "argscat body", node->nd_body));
+	    ADD_INSN(ret, line, concatarray);
+	}
+	break;
+      }
+      case NODE_ARGSPUSH:{
+	if (popped) {
+	    CHECK(COMPILE(ret, "arsgpush head", node->nd_head));
+	    ADD_INSN1(ret, line, splatarray, Qfalse);
+	    ADD_INSN(ret, line, pop);
+	    CHECK(COMPILE_(ret, "argspush body", node->nd_body, popped));
+	}
+	else {
+	    CHECK(COMPILE(ret, "arsgpush head", node->nd_head));
+	    CHECK(COMPILE_(ret, "argspush body", node->nd_body, popped));
+	    ADD_INSN1(ret, line, newarray, INT2FIX(1));
+	    ADD_INSN(ret, line, concatarray);
+	}
+	break;
+      }
+      case NODE_SPLAT:{
+	CHECK(COMPILE(ret, "splat", node->nd_head));
+	ADD_INSN1(ret, line, splatarray, Qtrue);
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_DEFN:{
+	const rb_iseq_t *method_iseq = NEW_ISEQ(node->nd_defn,
+						rb_id2str(node->nd_mid),
+						ISEQ_TYPE_METHOD, line);
+
+	debugp_param("defn/iseq", rb_iseqw_new(method_iseq));
+
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+	ADD_INSN1(ret, line, putobject, ID2SYM(node->nd_mid));
+	ADD_INSN1(ret, line, putiseq, method_iseq);
+	ADD_SEND (ret, line, id_core_define_method, INT2FIX(2));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+
+	break;
+      }
+      case NODE_DEFS:{
+	const rb_iseq_t * singleton_method = NEW_ISEQ(node->nd_defn,
+						      rb_id2str(node->nd_mid),
+						      ISEQ_TYPE_METHOD, line);
+
+	debugp_param("defs/iseq", rb_iseqw_new(singleton_method));
+
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+	CHECK(COMPILE(ret, "defs: recv", node->nd_recv));
+	ADD_INSN1(ret, line, putobject, ID2SYM(node->nd_mid));
+	ADD_INSN1(ret, line, putiseq, singleton_method);
+	ADD_SEND (ret, line, id_core_define_singleton_method, INT2FIX(3));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_ALIAS:{
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_CBASE));
+	CHECK(COMPILE(ret, "alias arg1", node->nd_1st));
+	CHECK(COMPILE(ret, "alias arg2", node->nd_2nd));
+	ADD_SEND(ret, line, id_core_set_method_alias, INT2FIX(3));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_VALIAS:{
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+	ADD_INSN1(ret, line, putobject, ID2SYM(node->nd_alias));
+	ADD_INSN1(ret, line, putobject, ID2SYM(node->nd_orig));
+	ADD_SEND(ret, line, id_core_set_variable_alias, INT2FIX(2));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_UNDEF:{
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_CBASE));
+	CHECK(COMPILE(ret, "undef arg", node->nd_undef));
+	ADD_SEND(ret, line, id_core_undef_method, INT2FIX(2));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_CLASS:{
+	const rb_iseq_t *class_iseq = NEW_CHILD_ISEQ(node->nd_body,
+						     rb_sprintf("<class:%"PRIsVALUE">", rb_id2str(node->nd_cpath->nd_mid)),
+						     ISEQ_TYPE_CLASS, line);
+	const int flags = VM_DEFINECLASS_TYPE_CLASS |
+	    (node->nd_super ? VM_DEFINECLASS_FLAG_HAS_SUPERCLASS : 0) |
+	    compile_cpath(ret, iseq, node->nd_cpath);
+
+	CHECK(COMPILE(ret, "super", node->nd_super));
+	ADD_INSN3(ret, line, defineclass, ID2SYM(node->nd_cpath->nd_mid), class_iseq, INT2FIX(flags));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_MODULE:{
+        const rb_iseq_t *module_iseq = NEW_CHILD_ISEQ(node->nd_body,
+						      rb_sprintf("<module:%"PRIsVALUE">", rb_id2str(node->nd_cpath->nd_mid)),
+						      ISEQ_TYPE_CLASS, line);
+	const int flags = VM_DEFINECLASS_TYPE_MODULE |
+	    compile_cpath(ret, iseq, node->nd_cpath);
+
+	ADD_INSN (ret, line, putnil); /* dummy */
+	ADD_INSN3(ret, line, defineclass, ID2SYM(node->nd_cpath->nd_mid), module_iseq, INT2FIX(flags));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_SCLASS:{
+	ID singletonclass;
+	const rb_iseq_t *singleton_class = NEW_ISEQ(node->nd_body, rb_fstring_lit("singleton class"),
+						    ISEQ_TYPE_CLASS, line);
+
+	CHECK(COMPILE(ret, "sclass#recv", node->nd_recv));
+	ADD_INSN (ret, line, putnil);
+	CONST_ID(singletonclass, "singletonclass");
+	ADD_INSN3(ret, line, defineclass,
+		  ID2SYM(singletonclass), singleton_class,
+		  INT2FIX(VM_DEFINECLASS_TYPE_SINGLETON_CLASS));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_COLON2:{
+	if (rb_is_const_id(node->nd_mid)) {
+	    /* constant */
+	    LABEL *lend = NEW_LABEL(line);
+	    int ic_index = body->is_size++;
+
+	    DECL_ANCHOR(pref);
+	    DECL_ANCHOR(body);
+
+	    INIT_ANCHOR(pref);
+	    INIT_ANCHOR(body);
+	    CHECK(compile_const_prefix(iseq, node, pref, body));
+	    if (LIST_INSN_SIZE_ZERO(pref)) {
+		if (ISEQ_COMPILE_DATA(iseq)->option->inline_const_cache) {
+                    ADD_INSN2(ret, line, opt_getinlinecache, lend, INT2FIX(ic_index));
+		}
+		else {
+		    ADD_INSN(ret, line, putnil);
+		}
+
+		ADD_SEQ(ret, body);
+
+		if (ISEQ_COMPILE_DATA(iseq)->option->inline_const_cache) {
+                    ADD_INSN1(ret, line, opt_setinlinecache, INT2FIX(ic_index));
+		    ADD_LABEL(ret, lend);
+		}
+	    }
+	    else {
+		ADD_SEQ(ret, pref);
+		ADD_SEQ(ret, body);
+	    }
+	}
+	else {
+	    /* function call */
+	    ADD_CALL_RECEIVER(ret, line);
+	    CHECK(COMPILE(ret, "colon2#nd_head", node->nd_head));
+	    ADD_CALL(ret, line, node->nd_mid, INT2FIX(1));
+	}
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_COLON3:{
+	LABEL *lend = NEW_LABEL(line);
+	int ic_index = body->is_size++;
+
+	debugi("colon3#nd_mid", node->nd_mid);
+
+	/* add cache insn */
+	if (ISEQ_COMPILE_DATA(iseq)->option->inline_const_cache) {
+            ADD_INSN2(ret, line, opt_getinlinecache, lend, INT2FIX(ic_index));
+	    ADD_INSN(ret, line, pop);
+	}
+
+	ADD_INSN1(ret, line, putobject, rb_cObject);
+	ADD_INSN1(ret, line, getconstant, ID2SYM(node->nd_mid));
+
+	if (ISEQ_COMPILE_DATA(iseq)->option->inline_const_cache) {
+            ADD_INSN1(ret, line, opt_setinlinecache, INT2FIX(ic_index));
+	    ADD_LABEL(ret, lend);
+	}
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_DOT2:
+      case NODE_DOT3:{
+	int excl = type == NODE_DOT3;
+	VALUE flag = INT2FIX(excl);
+	const NODE *b = node->nd_beg;
+	const NODE *e = node->nd_end;
+	if (number_literal_p(b) && number_literal_p(e)) {
+	    if (!popped) {
+		VALUE val = rb_range_new(b->nd_lit, e->nd_lit, excl);
+		iseq_add_mark_object_compile_time(iseq, val);
+		ADD_INSN1(ret, line, putobject, val);
+	    }
+	}
+	else {
+	    CHECK(COMPILE_(ret, "min", b, popped));
+	    CHECK(COMPILE_(ret, "max", e, popped));
+	    if (!popped) {
+		ADD_INSN1(ret, line, newrange, flag);
+	    }
+	}
+	break;
+      }
+      case NODE_FLIP2:
+      case NODE_FLIP3:{
+	LABEL *lend = NEW_LABEL(line);
+	LABEL *ltrue = NEW_LABEL(line);
+	LABEL *lfalse = NEW_LABEL(line);
+	CHECK(compile_flip_flop(iseq, ret, node, type == NODE_FLIP2,
+				ltrue, lfalse));
+	ADD_LABEL(ret, ltrue);
+	ADD_INSN1(ret, line, putobject, Qtrue);
+	ADD_INSNL(ret, line, jump, lend);
+	ADD_LABEL(ret, lfalse);
+	ADD_INSN1(ret, line, putobject, Qfalse);
+	ADD_LABEL(ret, lend);
+	break;
+      }
+      case NODE_SELF:{
+	if (!popped) {
+	    ADD_INSN(ret, line, putself);
+	}
+	break;
+      }
+      case NODE_NIL:{
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	}
+	break;
+      }
+      case NODE_TRUE:{
+	if (!popped) {
+	    ADD_INSN1(ret, line, putobject, Qtrue);
+	}
+	break;
+      }
+      case NODE_FALSE:{
+	if (!popped) {
+	    ADD_INSN1(ret, line, putobject, Qfalse);
+	}
+	break;
+      }
+      case NODE_ERRINFO:{
+	if (!popped) {
+	    if (body->type == ISEQ_TYPE_RESCUE) {
+		ADD_GETLOCAL(ret, line, LVAR_ERRINFO, 0);
+	    }
+	    else {
+		const rb_iseq_t *ip = iseq;
+		int level = 0;
+		while (ip) {
+		    if (ip->body->type == ISEQ_TYPE_RESCUE) {
+			break;
+		    }
+		    ip = ip->body->parent_iseq;
+		    level++;
+		}
+		if (ip) {
+		    ADD_GETLOCAL(ret, line, LVAR_ERRINFO, level);
+		}
+		else {
+		    ADD_INSN(ret, line, putnil);
+		}
+	    }
+	}
+	break;
+      }
+      case NODE_DEFINED:
+	if (!popped) {
+	    CHECK(compile_defined_expr(iseq, ret, node, Qtrue));
+	}
+	break;
+      case NODE_POSTEXE:{
+	/* compiled to:
+	 *   ONCE{ rb_mRubyVMFrozenCore::core#set_postexe{ ... } }
+	 */
+	int is_index = body->is_size++;
+	const rb_iseq_t *once_iseq =
+	    new_child_iseq_ifunc(iseq, IFUNC_NEW(build_postexe_iseq, node->nd_body, 0),
+				 rb_fstring(make_name_for_block(iseq)), iseq, ISEQ_TYPE_BLOCK, line);
+
+	ADD_INSN2(ret, line, once, once_iseq, INT2FIX(is_index));
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_KW_ARG:
+	{
+	    LABEL *end_label = NEW_LABEL(nd_line(node));
+	    const NODE *default_value = node->nd_body->nd_value;
+
+	    if (default_value == (const NODE *)-1) {
+		/* required argument. do nothing */
+		COMPILE_ERROR(ERROR_ARGS "unreachable");
+		goto ng;
+	    }
+	    else if (nd_type(default_value) == NODE_LIT ||
+		     nd_type(default_value) == NODE_NIL ||
+		     nd_type(default_value) == NODE_TRUE ||
+		     nd_type(default_value) == NODE_FALSE) {
+		COMPILE_ERROR(ERROR_ARGS "unreachable");
+		goto ng;
+	    }
+	    else {
+		/* if keywordcheck(_kw_bits, nth_keyword)
+		 *   kw = default_value
+		 * end
+		 */
+		int kw_bits_idx = body->local_table_size - body->param.keyword->bits_start;
+		int keyword_idx = body->param.keyword->num;
+
+		ADD_INSN2(ret, line, checkkeyword, INT2FIX(kw_bits_idx + VM_ENV_DATA_SIZE - 1), INT2FIX(keyword_idx));
+		ADD_INSNL(ret, line, branchif, end_label);
+		CHECK(COMPILE_POPPED(ret, "keyword default argument", node->nd_body));
+		ADD_LABEL(ret, end_label);
+	    }
+
+	    break;
+	}
+      case NODE_DSYM:{
+	compile_dstr(iseq, ret, node);
+	if (!popped) {
+	    ADD_INSN(ret, line, intern);
+	}
+	else {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      case NODE_ATTRASGN:{
+	DECL_ANCHOR(recv);
+	DECL_ANCHOR(args);
+	unsigned int flag = 0;
+	ID mid = node->nd_mid;
+	LABEL *lskip = 0;
+	VALUE argc;
+
+	/* optimization shortcut
+	 *   obj["literal"] = value -> opt_aset_with(obj, "literal", value)
+	 */
+	if (mid == idASET && !private_recv_p(node) && node->nd_args &&
+	    nd_type(node->nd_args) == NODE_ARRAY && node->nd_args->nd_alen == 2 &&
+	    nd_type(node->nd_args->nd_head) == NODE_STR &&
+	    ISEQ_COMPILE_DATA(iseq)->current_block == NULL &&
+            !ISEQ_COMPILE_DATA(iseq)->option->frozen_string_literal &&
+	    ISEQ_COMPILE_DATA(iseq)->option->specialized_instruction)
+	{
+	    VALUE str = freeze_literal(iseq, node->nd_args->nd_head->nd_lit);
+	    CHECK(COMPILE(ret, "recv", node->nd_recv));
+	    CHECK(COMPILE(ret, "value", node->nd_args->nd_next->nd_head));
+	    if (!popped) {
+		ADD_INSN(ret, line, swap);
+		ADD_INSN1(ret, line, topn, INT2FIX(1));
+	    }
+	    ADD_INSN3(ret, line, opt_aset_with, str,
+		      new_callinfo(iseq, idASET, 2, 0, NULL, FALSE),
+		      NULL/* CALL_CACHE */);
+	    ADD_INSN(ret, line, pop);
+	    break;
+	}
+
+	INIT_ANCHOR(recv);
+	INIT_ANCHOR(args);
+	argc = setup_args(iseq, args, node->nd_args, &flag, NULL);
+	CHECK(!NIL_P(argc));
+
+	flag |= COMPILE_RECV(recv, "recv", node);
+
+	debugp_param("argc", argc);
+	debugp_param("nd_mid", ID2SYM(mid));
+
+	if (!rb_is_attrset_id(mid)) {
+	    /* safe nav attr */
+	    mid = rb_id_attrset(mid);
+	    ADD_INSN(recv, line, dup);
+	    lskip = NEW_LABEL(line);
+	    ADD_INSNL(recv, line, branchnil, lskip);
+	}
+	if (!popped) {
+	    ADD_INSN(ret, line, putnil);
+	    ADD_SEQ(ret, recv);
+	    ADD_SEQ(ret, args);
+
+	    if (flag & VM_CALL_ARGS_BLOCKARG) {
+		ADD_INSN1(ret, line, topn, INT2FIX(1));
+		if (flag & VM_CALL_ARGS_SPLAT) {
+		    ADD_INSN1(ret, line, putobject, INT2FIX(-1));
+		    ADD_SEND(ret, line, idAREF, INT2FIX(1));
+		}
+		ADD_INSN1(ret, line, setn, FIXNUM_INC(argc, 3));
+		ADD_INSN (ret, line, pop);
+	    }
+	    else if (flag & VM_CALL_ARGS_SPLAT) {
+		ADD_INSN(ret, line, dup);
+		ADD_INSN1(ret, line, putobject, INT2FIX(-1));
+		ADD_SEND(ret, line, idAREF, INT2FIX(1));
+		ADD_INSN1(ret, line, setn, FIXNUM_INC(argc, 2));
+		ADD_INSN (ret, line, pop);
+	    }
+	    else {
+		ADD_INSN1(ret, line, setn, FIXNUM_INC(argc, 1));
+	    }
+	}
+	else {
+	    ADD_SEQ(ret, recv);
+	    ADD_SEQ(ret, args);
+	}
+	ADD_SEND_WITH_FLAG(ret, line, mid, argc, INT2FIX(flag));
+	if (lskip) ADD_LABEL(ret, lskip);
+	ADD_INSN(ret, line, pop);
+
+	break;
+      }
+      case NODE_LAMBDA:{
+	/* compile same as lambda{...} */
+	const rb_iseq_t *block = NEW_CHILD_ISEQ(node->nd_body, make_name_for_block(iseq), ISEQ_TYPE_BLOCK, line);
+	VALUE argc = INT2FIX(0);
+
+	ADD_INSN1(ret, line, putspecialobject, INT2FIX(VM_SPECIAL_OBJECT_VMCORE));
+	ADD_CALL_WITH_BLOCK(ret, line, idLambda, argc, block);
+
+	if (popped) {
+	    ADD_INSN(ret, line, pop);
+	}
+	break;
+      }
+      default:
+	UNKNOWN_NODE("iseq_compile_each", node, COMPILE_NG);
+      ng:
+	debug_node_end();
+	return COMPILE_NG;
+    }
+
+    debug_node_end();
+    return COMPILE_OK;
+}
+
+/***************************/
+/* instruction information */
+/***************************/
+
+static int
+insn_data_length(INSN *iobj)
+{
+    return insn_len(iobj->insn_id);
+}
+
+static int
+calc_sp_depth(int depth, INSN *insn)
+{
+    return insn_stack_increase(depth, insn->insn_id, insn->operands);
+}
+
+static VALUE
+opobj_inspect(VALUE obj)
+{
+    struct RBasic *r = (struct RBasic *) obj;
+    if (!SPECIAL_CONST_P(r)  && r->klass == 0) {
+	switch (BUILTIN_TYPE(r)) {
+	  case T_STRING:
+	    obj = rb_str_new_cstr(RSTRING_PTR(obj));
+	    break;
+	  case T_ARRAY:
+	    obj = rb_ary_dup(obj);
+	    break;
+	}
+    }
+    return rb_inspect(obj);
+}
+
+
+
+static VALUE
+insn_data_to_s_detail(INSN *iobj)
+{
+    VALUE str = rb_sprintf("%-20s ", insn_name(iobj->insn_id));
+
+    if (iobj->operands) {
+	const char *types = insn_op_types(iobj->insn_id);
+	int j;
+
+	for (j = 0; types[j]; j++) {
+	    char type = types[j];
+
+	    switch (type) {
+	      case TS_OFFSET:	/* label(destination position) */
+		{
+		    LABEL *lobj = (LABEL *)OPERAND_AT(iobj, j);
+		    rb_str_catf(str, LABEL_FORMAT, lobj->label_no);
+		    break;
+		}
+		break;
+	      case TS_ISEQ:	/* iseq */
+		{
+		    rb_iseq_t *iseq = (rb_iseq_t *)OPERAND_AT(iobj, j);
+		    VALUE val = Qnil;
+		    if (0 && iseq) { /* TODO: invalidate now */
+			val = (VALUE)iseq;
+		    }
+		    rb_str_concat(str, opobj_inspect(val));
+		}
+		break;
+	      case TS_LINDEX:
+	      case TS_NUM:	/* ulong */
+	      case TS_VALUE:	/* VALUE */
+		{
+		    VALUE v = OPERAND_AT(iobj, j);
+		    rb_str_concat(str, opobj_inspect(v));
+		    break;
+		}
+	      case TS_ID:	/* ID */
+		rb_str_concat(str, opobj_inspect(OPERAND_AT(iobj, j)));
+		break;
+	      case TS_GENTRY:
+		{
+		    struct rb_global_entry *entry = (struct rb_global_entry *)
+		      (OPERAND_AT(iobj, j) & (~1));
+		    rb_str_append(str, rb_id2str(entry->id));
+		    break;
+		}
+	      case TS_IC:	/* inline cache */
+	      case TS_ISE:	/* inline storage entry */
+		rb_str_catf(str, "<ic:%d>", FIX2INT(OPERAND_AT(iobj, j)));
+		break;
+	      case TS_CALLINFO: /* call info */
+		{
+		    struct rb_call_info *ci = (struct rb_call_info *)OPERAND_AT(iobj, j);
+		    rb_str_cat2(str, "<callinfo:");
+		    if (ci->mid) rb_str_catf(str, "%"PRIsVALUE, rb_id2str(ci->mid));
+		    rb_str_catf(str, ", %d>", ci->orig_argc);
+		    break;
+		}
+	      case TS_CALLCACHE: /* call cache */
+		{
+		    rb_str_catf(str, "<call cache>");
+		    break;
+		}
+	      case TS_CDHASH:	/* case/when condition cache */
+		rb_str_cat2(str, "<ch>");
+		break;
+	      case TS_FUNCPTR:
+		{
+		    const void *func = (const void *)OPERAND_AT(iobj, j);
+#ifdef HAVE_DLADDR
+		    Dl_info info;
+		    if (dladdr(func, &info) && info.dli_sname) {
+			rb_str_cat2(str, info.dli_sname);
+			break;
+		    }
+#endif
+		    rb_str_catf(str, "<%p>", func);
+		}
+		break;
+	      default:{
+		rb_raise(rb_eSyntaxError, "unknown operand type: %c", type);
+	      }
+	    }
+	    if (types[j + 1]) {
+		rb_str_cat2(str, ", ");
+	    }
+	}
+    }
+    return str;
+}
+
+static void
+dump_disasm_list(const LINK_ELEMENT *link)
+{
+    dump_disasm_list_with_cursor(link, NULL, NULL);
+}
+
+static void
+dump_disasm_list_with_cursor(const LINK_ELEMENT *link, const LINK_ELEMENT *curr, const LABEL *dest)
+{
+    int pos = 0;
+    INSN *iobj;
+    LABEL *lobj;
+    VALUE str;
+
+    printf("-- raw disasm--------\n");
+
+    while (link) {
+	if (curr) printf(curr == link ? "*" : " ");
+	switch (link->type) {
+	  case ISEQ_ELEMENT_INSN:
+	    {
+		iobj = (INSN *)link;
+		str = insn_data_to_s_detail(iobj);
+		printf("%04d %-65s(%4u)\n", pos, StringValueCStr(str), iobj->insn_info.line_no);
+		pos += insn_data_length(iobj);
+		break;
+	    }
+	  case ISEQ_ELEMENT_LABEL:
+	    {
+		lobj = (LABEL *)link;
+		printf(LABEL_FORMAT" [sp: %d]%s\n", lobj->label_no, lobj->sp,
+		       dest == lobj ? " <---" : "");
+		break;
+	    }
+	  case ISEQ_ELEMENT_TRACE:
+	    {
+		TRACE *trace = (TRACE *)link;
+		printf("trace: %0x\n", trace->event);
+		break;
+	    }
+	  case ISEQ_ELEMENT_ADJUST:
+	    {
+		ADJUST *adjust = (ADJUST *)link;
+		printf("adjust: [label: %d]\n", adjust->label ? adjust->label->label_no : -1);
+		break;
+	    }
+	  default:
+	    /* ignore */
+	    rb_raise(rb_eSyntaxError, "dump_disasm_list error: %ld\n", FIX2LONG(link->type));
+	}
+	link = link->next;
+    }
+    printf("---------------------\n");
+    fflush(stdout);
+}
+
+const char *
+rb_insns_name(int i)
+{
+    return insn_name(i);
+}
+
+VALUE
+rb_insns_name_array(void)
+{
+    VALUE ary = rb_ary_new_capa(VM_INSTRUCTION_SIZE);
+    int i;
+    for (i = 0; i < VM_INSTRUCTION_SIZE; i++) {
+	rb_ary_push(ary, rb_fstring_cstr(insn_name(i)));
+    }
+    return rb_obj_freeze(ary);
+}
+
+static LABEL *
+register_label(rb_iseq_t *iseq, struct st_table *labels_table, VALUE obj)
+{
+    LABEL *label = 0;
+    st_data_t tmp;
+    obj = rb_to_symbol_type(obj);
+
+    if (st_lookup(labels_table, obj, &tmp) == 0) {
+	label = NEW_LABEL(0);
+	st_insert(labels_table, obj, (st_data_t)label);
+    }
+    else {
+	label = (LABEL *)tmp;
+    }
+    LABEL_REF(label);
+    return label;
+}
+
+static VALUE
+get_exception_sym2type(VALUE sym)
+{
+#undef rb_intern
+#define rb_intern(str) rb_intern_const(str)
+    static VALUE symRescue, symEnsure, symRetry;
+    static VALUE symBreak, symRedo, symNext;
+
+    if (symRescue == 0) {
+	symRescue = ID2SYM(rb_intern("rescue"));
+	symEnsure = ID2SYM(rb_intern("ensure"));
+	symRetry  = ID2SYM(rb_intern("retry"));
+	symBreak  = ID2SYM(rb_intern("break"));
+	symRedo   = ID2SYM(rb_intern("redo"));
+	symNext   = ID2SYM(rb_intern("next"));
+    }
+
+    if (sym == symRescue) return CATCH_TYPE_RESCUE;
+    if (sym == symEnsure) return CATCH_TYPE_ENSURE;
+    if (sym == symRetry)  return CATCH_TYPE_RETRY;
+    if (sym == symBreak)  return CATCH_TYPE_BREAK;
+    if (sym == symRedo)   return CATCH_TYPE_REDO;
+    if (sym == symNext)   return CATCH_TYPE_NEXT;
+    rb_raise(rb_eSyntaxError, "invalid exception symbol: %+"PRIsVALUE, sym);
+    return 0;
+}
+
+static int
+iseq_build_from_ary_exception(rb_iseq_t *iseq, struct st_table *labels_table,
+		     VALUE exception)
+{
+    int i;
+
+    for (i=0; i<RARRAY_LEN(exception); i++) {
+	const rb_iseq_t *eiseq;
+	VALUE v, type;
+	LABEL *lstart, *lend, *lcont;
+	unsigned int sp;
+
+	v = rb_to_array_type(RARRAY_AREF(exception, i));
+	if (RARRAY_LEN(v) != 6) {
+	    rb_raise(rb_eSyntaxError, "wrong exception entry");
+	}
+        type = get_exception_sym2type(RARRAY_AREF(v, 0));
+        if (RARRAY_AREF(v, 1) == Qnil) {
+	    eiseq = NULL;
+	}
+	else {
+            eiseq = rb_iseqw_to_iseq(rb_iseq_load(RARRAY_AREF(v, 1), (VALUE)iseq, Qnil));
+        }
+
+        lstart = register_label(iseq, labels_table, RARRAY_AREF(v, 2));
+        lend   = register_label(iseq, labels_table, RARRAY_AREF(v, 3));
+        lcont  = register_label(iseq, labels_table, RARRAY_AREF(v, 4));
+        sp     = NUM2UINT(RARRAY_AREF(v, 5));
+
+	/* TODO: Dirty Hack!  Fix me */
+	if (type == CATCH_TYPE_RESCUE ||
+	    type == CATCH_TYPE_BREAK ||
+	    type == CATCH_TYPE_NEXT) {
+	    ++sp;
+	}
+
+	lcont->sp = sp;
+
+	ADD_CATCH_ENTRY(type, lstart, lend, eiseq, lcont);
+
+	RB_GC_GUARD(v);
+    }
+    return COMPILE_OK;
+}
+
+static struct st_table *
+insn_make_insn_table(void)
+{
+    struct st_table *table;
+    int i;
+    table = st_init_numtable();
+
+    for (i=0; i<VM_INSTRUCTION_SIZE; i++) {
+	st_insert(table, ID2SYM(rb_intern(insn_name(i))), i);
+    }
+
+    return table;
+}
+
+static const rb_iseq_t *
+iseq_build_load_iseq(const rb_iseq_t *iseq, VALUE op)
+{
+    VALUE iseqw;
+    const rb_iseq_t *loaded_iseq;
+
+    if (RB_TYPE_P(op, T_ARRAY)) {
+	iseqw = rb_iseq_load(op, (VALUE)iseq, Qnil);
+    }
+    else if (CLASS_OF(op) == rb_cISeq) {
+	iseqw = op;
+    }
+    else {
+	rb_raise(rb_eSyntaxError, "ISEQ is required");
+    }
+
+    loaded_iseq = rb_iseqw_to_iseq(iseqw);
+    return loaded_iseq;
+}
+
+static VALUE
+iseq_build_callinfo_from_hash(rb_iseq_t *iseq, VALUE op)
+{
+    ID mid = 0;
+    int orig_argc = 0;
+    unsigned int flag = 0;
+    struct rb_call_info_kw_arg *kw_arg = 0;
+
+    if (!NIL_P(op)) {
+	VALUE vmid = rb_hash_aref(op, ID2SYM(rb_intern("mid")));
+	VALUE vflag = rb_hash_aref(op, ID2SYM(rb_intern("flag")));
+	VALUE vorig_argc = rb_hash_aref(op, ID2SYM(rb_intern("orig_argc")));
+	VALUE vkw_arg = rb_hash_aref(op, ID2SYM(rb_intern("kw_arg")));
+
+	if (!NIL_P(vmid)) mid = SYM2ID(vmid);
+	if (!NIL_P(vflag)) flag = NUM2UINT(vflag);
+	if (!NIL_P(vorig_argc)) orig_argc = FIX2INT(vorig_argc);
+
+	if (!NIL_P(vkw_arg)) {
+	    int i;
+	    int len = RARRAY_LENINT(vkw_arg);
+	    size_t n = rb_call_info_kw_arg_bytes(len);
+
+	    kw_arg = xmalloc(n);
+	    kw_arg->keyword_len = len;
+	    for (i = 0; i < len; i++) {
+		VALUE kw = RARRAY_AREF(vkw_arg, i);
+		SYM2ID(kw);	/* make immortal */
+		kw_arg->keywords[i] = kw;
+	    }
+	}
+    }
+
+    return (VALUE)new_callinfo(iseq, mid, orig_argc, flag, kw_arg, (flag & VM_CALL_ARGS_SIMPLE) == 0);
+}
+
+static rb_event_flag_t
+event_name_to_flag(VALUE sym)
+{
+#define CHECK_EVENT(ev) if (sym == ID2SYM(rb_intern(#ev))) return ev;
+		CHECK_EVENT(RUBY_EVENT_LINE);
+		CHECK_EVENT(RUBY_EVENT_CLASS);
+		CHECK_EVENT(RUBY_EVENT_END);
+		CHECK_EVENT(RUBY_EVENT_CALL);
+		CHECK_EVENT(RUBY_EVENT_RETURN);
+		CHECK_EVENT(RUBY_EVENT_B_CALL);
+		CHECK_EVENT(RUBY_EVENT_B_RETURN);
+#undef CHECK_EVENT
+    return RUBY_EVENT_NONE;
+}
+
+static int
+iseq_build_from_ary_body(rb_iseq_t *iseq, LINK_ANCHOR *const anchor,
+			 VALUE body, VALUE labels_wrapper)
+{
+    /* TODO: body should be frozen */
+    long i, len = RARRAY_LEN(body);
+    struct st_table *labels_table = DATA_PTR(labels_wrapper);
+    int j;
+    int line_no = 0;
+    int ret = COMPILE_OK;
+
+    /*
+     * index -> LABEL *label
+     */
+    static struct st_table *insn_table;
+
+    if (insn_table == 0) {
+	insn_table = insn_make_insn_table();
+    }
+
+    for (i=0; i<len; i++) {
+        VALUE obj = RARRAY_AREF(body, i);
+
+	if (SYMBOL_P(obj)) {
+	    rb_event_flag_t event;
+	    if ((event = event_name_to_flag(obj)) != RUBY_EVENT_NONE) {
+		ADD_TRACE(anchor, event);
+	    }
+	    else {
+		LABEL *label = register_label(iseq, labels_table, obj);
+		ADD_LABEL(anchor, label);
+	    }
+	}
+	else if (FIXNUM_P(obj)) {
+	    line_no = NUM2INT(obj);
+	}
+	else if (RB_TYPE_P(obj, T_ARRAY)) {
+	    VALUE *argv = 0;
+	    int argc = RARRAY_LENINT(obj) - 1;
+	    st_data_t insn_id;
+	    VALUE insn;
+
+	    insn = (argc < 0) ? Qnil : RARRAY_AREF(obj, 0);
+	    if (st_lookup(insn_table, (st_data_t)insn, &insn_id) == 0) {
+		/* TODO: exception */
+		COMPILE_ERROR(iseq, line_no,
+			      "unknown instruction: %+"PRIsVALUE, insn);
+		ret = COMPILE_NG;
+		break;
+	    }
+
+	    if (argc != insn_len((VALUE)insn_id)-1) {
+		COMPILE_ERROR(iseq, line_no,
+			      "operand size mismatch");
+		ret = COMPILE_NG;
+		break;
+	    }
+
+	    if (argc > 0) {
+		argv = compile_data_alloc(iseq, sizeof(VALUE) * argc);
+		for (j=0; j<argc; j++) {
+		    VALUE op = rb_ary_entry(obj, j+1);
+		    switch (insn_op_type((VALUE)insn_id, j)) {
+		      case TS_OFFSET: {
+			LABEL *label = register_label(iseq, labels_table, op);
+			argv[j] = (VALUE)label;
+			break;
+		      }
+		      case TS_LINDEX:
+		      case TS_NUM:
+			(void)NUM2INT(op);
+			argv[j] = op;
+			break;
+		      case TS_VALUE:
+			argv[j] = op;
+			iseq_add_mark_object_compile_time(iseq, op);
+			break;
+		      case TS_ISEQ:
+			{
+			    if (op != Qnil) {
+				VALUE v = (VALUE)iseq_build_load_iseq(iseq, op);
+				argv[j] = v;
+				iseq_add_mark_object_compile_time(iseq, v);
+			    }
+			    else {
+				argv[j] = 0;
+			    }
+			}
+			break;
+		      case TS_GENTRY:
+			op = rb_to_symbol_type(op);
+			argv[j] = (VALUE)rb_global_entry(SYM2ID(op));
+			break;
+		      case TS_ISE:
+			FL_SET(iseq, ISEQ_MARKABLE_ISEQ);
+		      case TS_IC:
+			argv[j] = op;
+			if (NUM2UINT(op) >= iseq->body->is_size) {
+			    iseq->body->is_size = NUM2INT(op) + 1;
+			}
+			break;
+		      case TS_CALLINFO:
+			argv[j] = iseq_build_callinfo_from_hash(iseq, op);
+			break;
+		      case TS_CALLCACHE:
+			argv[j] = Qfalse;
+			break;
+		      case TS_ID:
+			argv[j] = rb_to_symbol_type(op);
+			break;
+		      case TS_CDHASH:
+			{
+			    int i;
+			    VALUE map = rb_hash_new_with_size(RARRAY_LEN(op)/2);
+
+                            RHASH_TBL_RAW(map)->type = &cdhash_type;
+			    op = rb_to_array_type(op);
+			    for (i=0; i<RARRAY_LEN(op); i+=2) {
+				VALUE key = RARRAY_AREF(op, i);
+				VALUE sym = RARRAY_AREF(op, i+1);
+				LABEL *label =
+				  register_label(iseq, labels_table, sym);
+				rb_hash_aset(map, key, (VALUE)label | 1);
+			    }
+			    RB_GC_GUARD(op);
+			    argv[j] = map;
+			    iseq_add_mark_object_compile_time(iseq, map);
+			}
+			break;
+		      case TS_FUNCPTR:
+			{
+#if SIZEOF_VALUE <= SIZEOF_LONG
+			    long funcptr = NUM2LONG(op);
+#else
+			    LONG_LONG funcptr = NUM2LL(op);
+#endif
+			    argv[j] = (VALUE)funcptr;
+			}
+			break;
+		      default:
+			rb_raise(rb_eSyntaxError, "unknown operand: %c", insn_op_type((VALUE)insn_id, j));
+		    }
+		}
+	    }
+	    ADD_ELEM(anchor,
+		     (LINK_ELEMENT*)new_insn_core(iseq, line_no,
+						  (enum ruby_vminsn_type)insn_id, argc, argv));
+	}
+	else {
+	    rb_raise(rb_eTypeError, "unexpected object for instruction");
+	}
+    }
+    DATA_PTR(labels_wrapper) = 0;
+    validate_labels(iseq, labels_table);
+    if (!ret) return ret;
+    return iseq_setup(iseq, anchor);
+}
+
+#define CHECK_ARRAY(v)   rb_to_array_type(v)
+#define CHECK_SYMBOL(v)  rb_to_symbol_type(v)
+
+static int
+int_param(int *dst, VALUE param, VALUE sym)
+{
+    VALUE val = rb_hash_aref(param, sym);
+    if (FIXNUM_P(val)) {
+	*dst = FIX2INT(val);
+	return TRUE;
+    }
+    else if (!NIL_P(val)) {
+	rb_raise(rb_eTypeError, "invalid %+"PRIsVALUE" Fixnum: %+"PRIsVALUE,
+		 sym, val);
+    }
+    return FALSE;
+}
+
+static const struct rb_iseq_param_keyword *
+iseq_build_kw(rb_iseq_t *iseq, VALUE params, VALUE keywords)
+{
+    int i, j;
+    int len = RARRAY_LENINT(keywords);
+    int default_len;
+    VALUE key, sym, default_val;
+    VALUE *dvs;
+    ID *ids;
+    struct rb_iseq_param_keyword *keyword = ZALLOC(struct rb_iseq_param_keyword);
+
+    iseq->body->param.flags.has_kw = TRUE;
+
+    keyword->num = len;
+#define SYM(s) ID2SYM(rb_intern(#s))
+    (void)int_param(&keyword->bits_start, params, SYM(kwbits));
+    i = keyword->bits_start - keyword->num;
+    ids = (ID *)&iseq->body->local_table[i];
+#undef SYM
+
+    /* required args */
+    for (i = 0; i < len; i++) {
+	VALUE val = RARRAY_AREF(keywords, i);
+
+	if (!SYMBOL_P(val)) {
+	    goto default_values;
+	}
+	ids[i] = SYM2ID(val);
+	keyword->required_num++;
+    }
+
+  default_values: /* note: we intentionally preserve `i' from previous loop */
+    default_len = len - i;
+    if (default_len == 0) {
+	keyword->table = ids;
+	return keyword;
+    }
+    else if (default_len < 0) {
+        UNREACHABLE;
+    }
+
+    dvs = ALLOC_N(VALUE, (unsigned int)default_len);
+
+    for (j = 0; i < len; i++, j++) {
+	key = RARRAY_AREF(keywords, i);
+	CHECK_ARRAY(key);
+
+	switch (RARRAY_LEN(key)) {
+	  case 1:
+	    sym = RARRAY_AREF(key, 0);
+	    default_val = Qundef;
+	    break;
+	  case 2:
+	    sym = RARRAY_AREF(key, 0);
+	    default_val = RARRAY_AREF(key, 1);
+	    break;
+	  default:
+	    rb_raise(rb_eTypeError, "keyword default has unsupported len %+"PRIsVALUE, key);
+	}
+	ids[i] = SYM2ID(sym);
+	dvs[j] = default_val;
+    }
+
+    keyword->table = ids;
+    keyword->default_values = dvs;
+
+    return keyword;
+}
+
+void
+rb_iseq_build_from_ary(rb_iseq_t *iseq, VALUE misc, VALUE locals, VALUE params,
+			 VALUE exception, VALUE body)
+{
+#define SYM(s) ID2SYM(rb_intern(#s))
+    int i, len;
+    unsigned int arg_size, local_size, stack_max;
+    ID *tbl;
+    struct st_table *labels_table = st_init_numtable();
+    VALUE labels_wrapper = Data_Wrap_Struct(0, 0, st_free_table, labels_table);
+    VALUE arg_opt_labels = rb_hash_aref(params, SYM(opt));
+    VALUE keywords = rb_hash_aref(params, SYM(keyword));
+    VALUE sym_arg_rest = ID2SYM(rb_intern("#arg_rest"));
+    DECL_ANCHOR(anchor);
+    INIT_ANCHOR(anchor);
+
+    len = RARRAY_LENINT(locals);
+    iseq->body->local_table_size = len;
+    iseq->body->local_table = tbl = len > 0 ? (ID *)ALLOC_N(ID, iseq->body->local_table_size) : NULL;
+
+    for (i = 0; i < len; i++) {
+	VALUE lv = RARRAY_AREF(locals, i);
+
+	if (sym_arg_rest == lv) {
+	    tbl[i] = 0;
+	}
+	else {
+	    tbl[i] = FIXNUM_P(lv) ? (ID)FIX2LONG(lv) : SYM2ID(CHECK_SYMBOL(lv));
+	}
+    }
+
+#define INT_PARAM(F) int_param(&iseq->body->param.F, params, SYM(F))
+    if (INT_PARAM(lead_num)) {
+	iseq->body->param.flags.has_lead = TRUE;
+    }
+    if (INT_PARAM(post_num)) iseq->body->param.flags.has_post = TRUE;
+    if (INT_PARAM(post_start)) iseq->body->param.flags.has_post = TRUE;
+    if (INT_PARAM(rest_start)) iseq->body->param.flags.has_rest = TRUE;
+    if (INT_PARAM(block_start)) iseq->body->param.flags.has_block = TRUE;
+#undef INT_PARAM
+    {
+#define INT_PARAM(F) F = (int_param(&x, misc, SYM(F)) ? (unsigned int)x : 0)
+	int x;
+	INT_PARAM(arg_size);
+	INT_PARAM(local_size);
+	INT_PARAM(stack_max);
+#undef INT_PARAM
+    }
+
+    if (RB_TYPE_P(arg_opt_labels, T_ARRAY)) {
+	len = RARRAY_LENINT(arg_opt_labels);
+	iseq->body->param.flags.has_opt = !!(len - 1 >= 0);
+
+	if (iseq->body->param.flags.has_opt) {
+	    VALUE *opt_table = ALLOC_N(VALUE, len);
+
+	    for (i = 0; i < len; i++) {
+		VALUE ent = RARRAY_AREF(arg_opt_labels, i);
+		LABEL *label = register_label(iseq, labels_table, ent);
+		opt_table[i] = (VALUE)label;
+	    }
+
+	    iseq->body->param.opt_num = len - 1;
+	    iseq->body->param.opt_table = opt_table;
+	}
+    }
+    else if (!NIL_P(arg_opt_labels)) {
+	rb_raise(rb_eTypeError, ":opt param is not an array: %+"PRIsVALUE,
+		 arg_opt_labels);
+    }
+
+    if (RB_TYPE_P(keywords, T_ARRAY)) {
+	iseq->body->param.keyword = iseq_build_kw(iseq, params, keywords);
+    }
+    else if (!NIL_P(keywords)) {
+	rb_raise(rb_eTypeError, ":keywords param is not an array: %+"PRIsVALUE,
+		 keywords);
+    }
+
+    if (Qtrue == rb_hash_aref(params, SYM(ambiguous_param0))) {
+	iseq->body->param.flags.ambiguous_param0 = TRUE;
+    }
+
+    if (int_param(&i, params, SYM(kwrest))) {
+	struct rb_iseq_param_keyword *keyword = (struct rb_iseq_param_keyword *)iseq->body->param.keyword;
+	if (keyword == NULL) {
+	    iseq->body->param.keyword = keyword = ZALLOC(struct rb_iseq_param_keyword);
+	}
+	keyword->rest_start = i;
+	iseq->body->param.flags.has_kwrest = TRUE;
+    }
+#undef SYM
+    iseq_calc_param_size(iseq);
+
+    /* exception */
+    iseq_build_from_ary_exception(iseq, labels_table, exception);
+
+    /* body */
+    iseq_build_from_ary_body(iseq, anchor, body, labels_wrapper);
+
+    iseq->body->param.size = arg_size;
+    iseq->body->local_table_size = local_size;
+    iseq->body->stack_max = stack_max;
+}
+
+/* for parser */
+
+int
+rb_dvar_defined(ID id, const struct rb_block *base_block)
+{
+    const rb_iseq_t *iseq;
+
+    if (base_block && (iseq = vm_block_iseq(base_block)) != NULL) {
+	const struct rb_iseq_constant_body *body = iseq->body;
+	while (body->type == ISEQ_TYPE_BLOCK ||
+	       body->type == ISEQ_TYPE_RESCUE ||
+	       body->type == ISEQ_TYPE_ENSURE ||
+	       body->type == ISEQ_TYPE_EVAL ||
+	       body->type == ISEQ_TYPE_MAIN
+	       ) {
+	    unsigned int i;
+
+	    for (i = 0; i < body->local_table_size; i++) {
+		if (body->local_table[i] == id) {
+		    return 1;
+		}
+	    }
+	    iseq = body->parent_iseq;
+	    body = iseq->body;
+	}
+    }
+    return 0;
+}
+
+int
+rb_local_defined(ID id, const struct rb_block *base_block)
+{
+    const rb_iseq_t *iseq;
+
+    if (base_block && (iseq = vm_block_iseq(base_block)) != NULL) {
+	unsigned int i;
+	const struct rb_iseq_constant_body *const body = iseq->body->local_iseq->body;
+
+	for (i=0; i<body->local_table_size; i++) {
+	    if (body->local_table[i] == id) {
+		return 1;
+	    }
+	}
+    }
+    return 0;
+}
+
+static int
+caller_location(VALUE *path, VALUE *realpath)
+{
+    const rb_execution_context_t *ec = GET_EC();
+    const rb_control_frame_t *const cfp =
+        rb_vm_get_ruby_level_next_cfp(ec, ec->cfp);
+
+    if (cfp) {
+	int line = rb_vm_get_sourceline(cfp);
+	*path = rb_iseq_path(cfp->iseq);
+	*realpath = rb_iseq_realpath(cfp->iseq);
+	return line;
+    }
+    else {
+	*path = rb_fstring_lit("<compiled>");
+	*realpath = *path;
+	return 1;
+    }
+}
+
+typedef struct {
+    VALUE arg;
+    rb_insn_func_t func;
+    int line;
+} accessor_args;
+
+static const rb_iseq_t *
+method_for_self(VALUE name, VALUE arg, rb_insn_func_t func,
+		VALUE (*build)(rb_iseq_t *, LINK_ANCHOR *const, VALUE))
+{
+    VALUE path, realpath;
+    accessor_args acc;
+
+    acc.arg = arg;
+    acc.func = func;
+    acc.line = caller_location(&path, &realpath);
+    return rb_iseq_new_ifunc(IFUNC_NEW(build, (VALUE)&acc, 0),
+			     rb_sym2str(name), path, realpath,
+			     INT2FIX(acc.line), 0, ISEQ_TYPE_METHOD, 0);
+}
+
+static VALUE
+for_self_aref(rb_iseq_t *iseq, LINK_ANCHOR *const ret, VALUE a)
+{
+    const accessor_args *const args = (void *)a;
+    const int line = args->line;
+    struct rb_iseq_constant_body *const body = iseq->body;
+
+    iseq_set_local_table(iseq, 0);
+    body->param.lead_num = 0;
+    body->param.size = 0;
+
+    ADD_INSN1(ret, line, putobject, args->arg);
+    ADD_INSN1(ret, line, opt_call_c_function, (VALUE)args->func);
+    return Qnil;
+}
+
+static VALUE
+for_self_aset(rb_iseq_t *iseq, LINK_ANCHOR *const ret, VALUE a)
+{
+    const accessor_args *const args = (void *)a;
+    const int line = args->line;
+    struct rb_iseq_constant_body *const body = iseq->body;
+    static const ID vars[] = {1, idUScore};
+
+    iseq_set_local_table(iseq, vars);
+    body->param.lead_num = 1;
+    body->param.size = 1;
+
+    ADD_GETLOCAL(ret, line, numberof(vars)-1, 0);
+    ADD_INSN1(ret, line, putobject, args->arg);
+    ADD_INSN1(ret, line, opt_call_c_function, (VALUE)args->func);
+    ADD_INSN(ret, line, pop);
+    return Qnil;
+}
+
+/*
+ * func (index) -> (value)
+ */
+const rb_iseq_t *
+rb_method_for_self_aref(VALUE name, VALUE arg, rb_insn_func_t func)
+{
+    return method_for_self(name, arg, func, for_self_aref);
+}
+
+/*
+ * func (index, value) -> (index, value)
+ */
+const rb_iseq_t *
+rb_method_for_self_aset(VALUE name, VALUE arg, rb_insn_func_t func)
+{
+    return method_for_self(name, arg, func, for_self_aset);
+}
+
+/* ISeq binary format */
+
+#ifndef IBF_ISEQ_DEBUG
+#define IBF_ISEQ_DEBUG 0
+#endif
+
+typedef unsigned int ibf_offset_t;
+#define IBF_OFFSET(ptr) ((ibf_offset_t)(VALUE)(ptr))
+
+struct ibf_header {
+    char magic[4]; /* YARB */
+    unsigned int major_version;
+    unsigned int minor_version;
+    unsigned int size;
+    unsigned int extra_size;
+
+    unsigned int iseq_list_size;
+    unsigned int id_list_size;
+    unsigned int object_list_size;
+
+    ibf_offset_t iseq_list_offset;
+    ibf_offset_t id_list_offset;
+    ibf_offset_t object_list_offset;
+};
+
+struct ibf_dump {
+    VALUE str;
+    VALUE iseq_list;      /* [iseq0 offset, ...] */
+    VALUE obj_list;       /* [objs] */
+    st_table *iseq_table; /* iseq -> iseq number */
+    st_table *id_table;   /* id -> id number */
+};
+
+rb_iseq_t * iseq_alloc(void);
+
+struct ibf_load {
+    const RUBY_ALIGNAS(SIZEOF_VALUE) char *buff;
+    const struct ibf_header *header;
+    ID *id_list;     /* [id0, ...] */
+    VALUE iseq_list; /* [iseq0, ...] */
+    VALUE obj_list;  /* [obj0, ...] */
+    VALUE loader_obj;
+    VALUE str;
+    rb_iseq_t *iseq;
+};
+
+static ibf_offset_t
+ibf_dump_pos(struct ibf_dump *dump)
+{
+    long pos = RSTRING_LEN(dump->str);
+#if SIZEOF_LONG > SIZEOF_INT
+    if (pos >= UINT_MAX) {
+        rb_raise(rb_eRuntimeError, "dump size exceeds");
+    }
+#endif
+    return (unsigned int)pos;
+}
+
+static void
+ibf_dump_align(struct ibf_dump *dump, size_t align)
+{
+    ibf_offset_t pos = ibf_dump_pos(dump);
+    if (pos % align) {
+        static const char padding[sizeof(VALUE)];
+        size_t size = align - ((size_t)pos % align);
+#if SIZEOF_LONG > SIZEOF_INT
+        if (pos + size >= UINT_MAX) {
+            rb_raise(rb_eRuntimeError, "dump size exceeds");
+        }
+#endif
+        for (; size > sizeof(padding); size -= sizeof(padding)) {
+            rb_str_cat(dump->str, padding, sizeof(padding));
+        }
+        rb_str_cat(dump->str, padding, size);
+    }
+}
+
+static ibf_offset_t
+ibf_dump_write(struct ibf_dump *dump, const void *buff, unsigned long size)
+{
+    ibf_offset_t pos = ibf_dump_pos(dump);
+    rb_str_cat(dump->str, (const char *)buff, size);
+    /* TODO: overflow check */
+    return pos;
+}
+
+static void
+ibf_dump_overwrite(struct ibf_dump *dump, void *buff, unsigned int size, long offset)
+{
+    VALUE str = dump->str;
+    char *ptr = RSTRING_PTR(str);
+    if ((unsigned long)(size + offset) > (unsigned long)RSTRING_LEN(str))
+	rb_bug("ibf_dump_overwrite: overflow");
+    memcpy(ptr + offset, buff, size);
+}
+
+static void *
+ibf_load_alloc(const struct ibf_load *load, ibf_offset_t offset, int size)
+{
+    void *buff = ruby_xmalloc(size);
+    memcpy(buff, load->buff + offset, size);
+    return buff;
+}
+
+#define IBF_W_ALIGN(type) (RUBY_ALIGNOF(type) > 1 ? ibf_dump_align(dump, RUBY_ALIGNOF(type)) : (void)0)
+
+#define IBF_W(b, type, n) (IBF_W_ALIGN(type), (type *)(VALUE)IBF_WP(b, type, n))
+#define IBF_WV(variable)   ibf_dump_write(dump, &(variable), sizeof(variable))
+#define IBF_WP(b, type, n) ibf_dump_write(dump, (b), sizeof(type) * (n))
+#define IBF_R(val, type, n) (type *)ibf_load_alloc(load, IBF_OFFSET(val), sizeof(type) * (n))
+#define IBF_ZERO(variable) memset(&(variable), 0, sizeof(variable))
+
+static int
+ibf_table_lookup(struct st_table *table, st_data_t key)
+{
+    st_data_t val;
+
+    if (st_lookup(table, key, &val)) {
+	return (int)val;
+    }
+    else {
+	return -1;
+    }
+}
+
+static int
+ibf_table_index(struct st_table *table, st_data_t key)
+{
+    int index = ibf_table_lookup(table, key);
+
+    if (index < 0) { /* not found */
+	index = (int)table->num_entries;
+	st_insert(table, key, (st_data_t)index);
+    }
+
+    return index;
+}
+
+/* dump/load generic */
+
+static VALUE ibf_load_object(const struct ibf_load *load, VALUE object_index);
+static rb_iseq_t *ibf_load_iseq(const struct ibf_load *load, const rb_iseq_t *index_iseq);
+
+static VALUE
+ibf_dump_object(struct ibf_dump *dump, VALUE obj)
+{
+    long index = RARRAY_LEN(dump->obj_list);
+    long i;
+    for (i=0; i<index; i++) {
+	if (RARRAY_AREF(dump->obj_list, i) == obj) return (VALUE)i; /* dedup */
+    }
+    rb_ary_push(dump->obj_list, obj);
+    return (VALUE)index;
+}
+
+static VALUE
+ibf_dump_id(struct ibf_dump *dump, ID id)
+{
+    return (VALUE)ibf_table_index(dump->id_table, (st_data_t)id);
+}
+
+static ID
+ibf_load_id(const struct ibf_load *load, const ID id_index)
+{
+    ID id;
+
+    if (id_index == 0) {
+	id = 0;
+    }
+    else {
+	id = load->id_list[(long)id_index];
+
+	if (id == 0) {
+	    long *indices = (long *)(load->buff + load->header->id_list_offset);
+	    VALUE str = ibf_load_object(load, indices[id_index]);
+	    id = NIL_P(str) ? 0 : rb_intern_str(str); /* str == nil -> internal junk id */
+	    load->id_list[(long)id_index] = id;
+	}
+    }
+
+    return id;
+}
+
+/* dump/load: code */
+
+static VALUE
+ibf_dump_callinfo(struct ibf_dump *dump, const struct rb_call_info *ci)
+{
+    return (ci->flag & VM_CALL_KWARG) ? Qtrue : Qfalse;
+}
+
+static ibf_offset_t ibf_dump_iseq_each(struct ibf_dump *dump, const rb_iseq_t *iseq);
+
+static rb_iseq_t *
+ibf_dump_iseq(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    if (iseq == NULL) {
+	return (rb_iseq_t *)-1;
+    }
+    else {
+	int iseq_index = ibf_table_lookup(dump->iseq_table, (st_data_t)iseq);
+	if (iseq_index < 0) {
+	    iseq_index = ibf_table_index(dump->iseq_table, (st_data_t)iseq);
+	    rb_ary_store(dump->iseq_list, iseq_index, LONG2NUM(ibf_dump_iseq_each(dump, rb_iseq_check(iseq))));
+	}
+	return (rb_iseq_t *)(VALUE)iseq_index;
+    }
+}
+
+static VALUE
+ibf_dump_gentry(struct ibf_dump *dump, const struct rb_global_entry *entry)
+{
+    return (VALUE)ibf_dump_id(dump, entry->id);
+}
+
+static VALUE
+ibf_load_gentry(const struct ibf_load *load, const struct rb_global_entry *entry)
+{
+    ID gid = ibf_load_id(load, (ID)(VALUE)entry);
+    return (VALUE)rb_global_entry(gid);
+}
+
+static VALUE *
+ibf_dump_code(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    const struct rb_iseq_constant_body *const body = iseq->body;
+    const int iseq_size = body->iseq_size;
+    int code_index;
+    VALUE *code;
+    const VALUE *orig_code = rb_iseq_original_iseq(iseq);
+
+    code = ALLOCA_N(VALUE, iseq_size);
+
+    for (code_index=0; code_index<iseq_size;) {
+	const VALUE insn = orig_code[code_index];
+	const char *types = insn_op_types(insn);
+	int op_index;
+
+	code[code_index++] = (VALUE)insn;
+
+	for (op_index=0; types[op_index]; op_index++, code_index++) {
+	    VALUE op = orig_code[code_index];
+	    switch (types[op_index]) {
+	      case TS_CDHASH:
+	      case TS_VALUE:
+		code[code_index] = ibf_dump_object(dump, op);
+		break;
+	      case TS_ISEQ:
+		code[code_index] = (VALUE)ibf_dump_iseq(dump, (const rb_iseq_t *)op);
+		break;
+	      case TS_IC:
+	      case TS_ISE:
+		{
+		    unsigned int i;
+		    for (i=0; i<body->is_size; i++) {
+			if (op == (VALUE)&body->is_entries[i]) {
+			    break;
+			}
+		    }
+		    code[code_index] = i;
+		}
+		break;
+	      case TS_CALLINFO:
+		code[code_index] = ibf_dump_callinfo(dump, (const struct rb_call_info *)op);
+		break;
+	      case TS_CALLCACHE:
+		code[code_index] = 0;
+		break;
+	      case TS_ID:
+		code[code_index] = ibf_dump_id(dump, (ID)op);
+		break;
+	      case TS_GENTRY:
+		code[code_index] = ibf_dump_gentry(dump, (const struct rb_global_entry *)op);
+		break;
+	      case TS_FUNCPTR:
+		rb_raise(rb_eRuntimeError, "TS_FUNCPTR is not supported");
+		break;
+	      default:
+		code[code_index] = op;
+		break;
+	    }
+	}
+	assert(insn_len(insn) == op_index+1);
+    }
+
+    return IBF_W(code, VALUE, iseq_size);
+}
+
+static VALUE *
+ibf_load_code(const struct ibf_load *load, const rb_iseq_t *iseq, const struct rb_iseq_constant_body *body)
+{
+    const int iseq_size = body->iseq_size;
+    int code_index;
+    VALUE *code = IBF_R(body->iseq_encoded, VALUE, iseq_size);
+
+    struct rb_iseq_constant_body *load_body = iseq->body;
+    struct rb_call_info *ci_entries = load_body->ci_entries;
+    struct rb_call_info_with_kwarg *ci_kw_entries = (struct rb_call_info_with_kwarg *)&load_body->ci_entries[load_body->ci_size];
+    struct rb_call_cache *cc_entries = load_body->cc_entries;
+    union iseq_inline_storage_entry *is_entries = load_body->is_entries;
+
+    load_body->iseq_encoded = code;
+    load_body->iseq_size = 0;
+    for (code_index=0; code_index<iseq_size;) {
+	const VALUE insn = code[code_index++];
+	const char *types = insn_op_types(insn);
+	int op_index;
+
+	for (op_index=0; types[op_index]; op_index++, code_index++) {
+	    VALUE op = code[code_index];
+
+	    switch (types[op_index]) {
+	      case TS_CDHASH:
+	      case TS_VALUE:
+	        {
+		    VALUE v = ibf_load_object(load, op);
+		    code[code_index] = v;
+		    if (!SPECIAL_CONST_P(v)) {
+			RB_OBJ_WRITTEN(iseq, Qundef, v);
+			FL_SET(iseq, ISEQ_MARKABLE_ISEQ);
+		    }
+		    break;
+		}
+	      case TS_ISEQ:
+	        {
+		    VALUE v = (VALUE)ibf_load_iseq(load, (const rb_iseq_t *)op);
+		    code[code_index] = v;
+		    if (!SPECIAL_CONST_P(v)) {
+			RB_OBJ_WRITTEN(iseq, Qundef, v);
+			FL_SET(iseq, ISEQ_MARKABLE_ISEQ);
+		    }
+		    break;
+		}
+	      case TS_ISE:
+		FL_SET(iseq, ISEQ_MARKABLE_ISEQ);
+	      case TS_IC:
+		code[code_index] = (VALUE)&is_entries[(int)op];
+		break;
+	      case TS_CALLINFO:
+		code[code_index] = op ? (VALUE)ci_kw_entries++ : (VALUE)ci_entries++; /* op is Qtrue (kw) or Qfalse (!kw) */
+		break;
+	      case TS_CALLCACHE:
+		code[code_index] = (VALUE)cc_entries++;
+		break;
+	      case TS_ID:
+		code[code_index] = ibf_load_id(load, (ID)op);
+		break;
+	      case TS_GENTRY:
+		code[code_index] = ibf_load_gentry(load, (const struct rb_global_entry *)op);
+		break;
+	      case TS_FUNCPTR:
+		rb_raise(rb_eRuntimeError, "TS_FUNCPTR is not supported");
+		break;
+	      default:
+		/* code[code_index] = op; */
+		continue;
+	    }
+	}
+	if (insn_len(insn) != op_index+1) {
+	    rb_raise(rb_eRuntimeError, "operand size mismatch");
+	}
+    }
+    load_body->iseq_size = code_index;
+
+    return code;
+}
+
+static VALUE *
+ibf_dump_param_opt_table(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    int opt_num = iseq->body->param.opt_num;
+
+    if (opt_num > 0) {
+	return IBF_W(iseq->body->param.opt_table, VALUE, opt_num + 1);
+    }
+    else {
+	return NULL;
+    }
+}
+
+static VALUE *
+ibf_load_param_opt_table(const struct ibf_load *load, const struct rb_iseq_constant_body *body)
+{
+    int opt_num = body->param.opt_num;
+
+    if (opt_num > 0) {
+	ibf_offset_t offset = IBF_OFFSET(body->param.opt_table);
+	VALUE *table = ALLOC_N(VALUE, opt_num+1);
+	MEMCPY(table, load->buff + offset, VALUE, opt_num+1);
+	return table;
+    }
+    else {
+	return NULL;
+    }
+}
+
+static struct rb_iseq_param_keyword *
+ibf_dump_param_keyword(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    const struct rb_iseq_param_keyword *kw = iseq->body->param.keyword;
+
+    if (kw) {
+	struct rb_iseq_param_keyword dump_kw = *kw;
+	int dv_num = kw->num - kw->required_num;
+	ID *ids = kw->num > 0 ? ALLOCA_N(ID, kw->num) : NULL;
+	VALUE *dvs = dv_num > 0 ? ALLOCA_N(VALUE, dv_num) : NULL;
+	int i;
+
+	for (i=0; i<kw->num; i++) ids[i] = (ID)ibf_dump_id(dump, kw->table[i]);
+	for (i=0; i<dv_num; i++) dvs[i] = (VALUE)ibf_dump_object(dump, kw->default_values[i]);
+
+	dump_kw.table = IBF_W(ids, ID, kw->num);
+	dump_kw.default_values = IBF_W(dvs, VALUE, dv_num);
+	return IBF_W(&dump_kw, struct rb_iseq_param_keyword, 1);
+    }
+    else {
+	return NULL;
+    }
+}
+
+static const struct rb_iseq_param_keyword *
+ibf_load_param_keyword(const struct ibf_load *load, const struct rb_iseq_constant_body *body)
+{
+    if (body->param.keyword) {
+	struct rb_iseq_param_keyword *kw = IBF_R(body->param.keyword, struct rb_iseq_param_keyword, 1);
+	ID *ids = IBF_R(kw->table, ID, kw->num);
+	int dv_num = kw->num - kw->required_num;
+	VALUE *dvs = IBF_R(kw->default_values, VALUE, dv_num);
+	int i;
+
+	for (i=0; i<kw->num; i++) {
+	    ids[i] = ibf_load_id(load, ids[i]);
+	}
+	for (i=0; i<dv_num; i++) {
+	    dvs[i] = ibf_load_object(load, dvs[i]);
+	}
+
+	kw->table = ids;
+	kw->default_values = dvs;
+	return kw;
+    }
+    else {
+	return NULL;
+    }
+}
+
+static struct iseq_insn_info_entry *
+ibf_dump_insns_info_body(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    return IBF_W(iseq->body->insns_info.body, struct iseq_insn_info_entry, iseq->body->insns_info.size);
+}
+
+static struct iseq_insn_info_entry *
+ibf_load_insns_info_body(const struct ibf_load *load, const struct rb_iseq_constant_body *body)
+{
+    return IBF_R(body->insns_info.body, struct iseq_insn_info_entry, body->insns_info.size);
+}
+
+static unsigned int *
+ibf_dump_insns_info_positions(struct ibf_dump *dump, const struct rb_iseq_constant_body *body)
+{
+    return IBF_W(body->insns_info.positions, unsigned int, body->insns_info.size);
+}
+
+static unsigned int *
+ibf_load_insns_info_positions(const struct ibf_load *load, const struct rb_iseq_constant_body *body)
+{
+    return IBF_R(body->insns_info.positions, unsigned int, body->insns_info.size);
+}
+
+static ID *
+ibf_dump_local_table(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    const struct rb_iseq_constant_body *const body = iseq->body;
+    const int size = body->local_table_size;
+    ID *table = ALLOCA_N(ID, size);
+    int i;
+
+    for (i=0; i<size; i++) {
+	table[i] = ibf_dump_id(dump, body->local_table[i]);
+    }
+
+    return IBF_W(table, ID, size);
+}
+
+static ID *
+ibf_load_local_table(const struct ibf_load *load, const struct rb_iseq_constant_body *body)
+{
+    const int size = body->local_table_size;
+
+    if (size > 0) {
+	ID *table = IBF_R(body->local_table, ID, size);
+	int i;
+
+	for (i=0; i<size; i++) {
+	    table[i] = ibf_load_id(load, table[i]);
+	}
+	return table;
+    }
+    else {
+	return NULL;
+    }
+}
+
+static struct iseq_catch_table *
+ibf_dump_catch_table(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    const struct iseq_catch_table *table = iseq->body->catch_table;
+
+    if (table) {
+	int byte_size = iseq_catch_table_bytes(iseq->body->catch_table->size);
+	struct iseq_catch_table *dump_table = (struct iseq_catch_table *)ALLOCA_N(char, byte_size);
+	unsigned int i;
+	dump_table->size = table->size;
+	for (i=0; i<table->size; i++) {
+	    dump_table->entries[i] = table->entries[i];
+	    dump_table->entries[i].iseq = ibf_dump_iseq(dump, table->entries[i].iseq);
+	}
+	return (struct iseq_catch_table *)(VALUE)ibf_dump_write(dump, dump_table, byte_size);
+    }
+    else {
+	return NULL;
+    }
+}
+
+static struct iseq_catch_table *
+ibf_load_catch_table(const struct ibf_load *load, const struct rb_iseq_constant_body *body)
+{
+    if (body->catch_table) {
+	struct iseq_catch_table *table;
+	unsigned int i;
+	unsigned int size;
+	size = *(unsigned int *)(load->buff + IBF_OFFSET(body->catch_table));
+	table = ibf_load_alloc(load, IBF_OFFSET(body->catch_table), iseq_catch_table_bytes(size));
+	for (i=0; i<size; i++) {
+	    table->entries[i].iseq = ibf_load_iseq(load, table->entries[i].iseq);
+	}
+	return table;
+    }
+    else {
+	return NULL;
+    }
+}
+
+static struct rb_call_info *
+ibf_dump_ci_entries(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    const struct rb_iseq_constant_body *const body = iseq->body;
+    const unsigned int ci_size = body->ci_size;
+    const unsigned int ci_kw_size = body->ci_kw_size;
+    const struct rb_call_info *ci_entries = body->ci_entries;
+    struct rb_call_info *dump_ci_entries;
+    struct rb_call_info_with_kwarg *dump_ci_kw_entries;
+    int byte_size = ci_size * sizeof(struct rb_call_info) +
+                    ci_kw_size * sizeof(struct rb_call_info_with_kwarg);
+    unsigned int i;
+
+    dump_ci_entries = (struct rb_call_info *)ALLOCA_N(char, byte_size);
+    dump_ci_kw_entries = (struct rb_call_info_with_kwarg *)&dump_ci_entries[ci_size];
+    memcpy(dump_ci_entries, ci_entries, byte_size);
+
+    for (i=0; i<ci_size; i++) { /* conver ID for each ci */
+	dump_ci_entries[i].mid = ibf_dump_id(dump, dump_ci_entries[i].mid);
+    }
+    for (i=0; i<ci_kw_size; i++) {
+	const struct rb_call_info_kw_arg *kw_arg = dump_ci_kw_entries[i].kw_arg;
+	int j;
+	VALUE *keywords = ALLOCA_N(VALUE, kw_arg->keyword_len);
+	for (j=0; j<kw_arg->keyword_len; j++) {
+	    keywords[j] = (VALUE)ibf_dump_object(dump, kw_arg->keywords[j]); /* kw_arg->keywords[n] is Symbol */
+	}
+	dump_ci_kw_entries[i].kw_arg = (struct rb_call_info_kw_arg *)(VALUE)ibf_dump_write(dump, &kw_arg->keyword_len, sizeof(int));
+	ibf_dump_write(dump, keywords, sizeof(VALUE) * kw_arg->keyword_len);
+
+	dump_ci_kw_entries[i].ci.mid = ibf_dump_id(dump, dump_ci_kw_entries[i].ci.mid);
+    }
+    return (struct rb_call_info *)(VALUE)ibf_dump_write(dump, dump_ci_entries, byte_size);
+}
+
+static struct rb_call_info *
+ibf_load_ci_entries(const struct ibf_load *load, const struct rb_iseq_constant_body *body)
+{
+    unsigned int i;
+    const unsigned int ci_size = body->ci_size;
+    const unsigned int ci_kw_size = body->ci_kw_size;
+    struct rb_call_info *ci_entries = ibf_load_alloc(load, IBF_OFFSET(body->ci_entries),
+						     sizeof(struct rb_call_info) * body->ci_size +
+						     sizeof(struct rb_call_info_with_kwarg) * body->ci_kw_size);
+    struct rb_call_info_with_kwarg *ci_kw_entries = (struct rb_call_info_with_kwarg *)&ci_entries[ci_size];
+
+    for (i=0; i<ci_size; i++) {
+	ci_entries[i].mid = ibf_load_id(load, ci_entries[i].mid);
+    }
+    for (i=0; i<ci_kw_size; i++) {
+	int j;
+	ibf_offset_t kw_arg_offset = IBF_OFFSET(ci_kw_entries[i].kw_arg);
+	const int keyword_len = *(int *)(load->buff + kw_arg_offset);
+	const VALUE *keywords = (VALUE *)(load->buff + kw_arg_offset + sizeof(int));
+	struct rb_call_info_kw_arg *kw_arg = ruby_xmalloc(sizeof(struct rb_call_info_kw_arg) + sizeof(VALUE) * (keyword_len - 1));
+	kw_arg->keyword_len = keyword_len;
+	for (j=0; j<kw_arg->keyword_len; j++) {
+	    kw_arg->keywords[j] = (VALUE)ibf_load_object(load, keywords[j]);
+	}
+	ci_kw_entries[i].kw_arg = kw_arg;
+	ci_kw_entries[i].ci.mid = ibf_load_id(load, ci_kw_entries[i].ci.mid);
+    }
+
+    return ci_entries;
+}
+
+static ibf_offset_t
+ibf_dump_iseq_each(struct ibf_dump *dump, const rb_iseq_t *iseq)
+{
+    struct rb_iseq_constant_body dump_body;
+#if VM_INSN_INFO_TABLE_IMPL == 2
+    unsigned int *positions;
+#endif
+    dump_body = *iseq->body;
+
+    dump_body.location.pathobj = ibf_dump_object(dump, dump_body.location.pathobj); /* TODO: freeze */
+    dump_body.location.base_label = ibf_dump_object(dump, dump_body.location.base_label);
+    dump_body.location.label = ibf_dump_object(dump, dump_body.location.label);
+
+    dump_body.iseq_encoded =         ibf_dump_code(dump, iseq);
+    dump_body.param.opt_table =      ibf_dump_param_opt_table(dump, iseq);
+    dump_body.param.keyword =        ibf_dump_param_keyword(dump, iseq);
+    dump_body.insns_info.body =      ibf_dump_insns_info_body(dump, iseq);
+#if VM_INSN_INFO_TABLE_IMPL == 2
+    positions = rb_iseq_insns_info_decode_positions(&dump_body);
+    dump_body.insns_info.positions = positions;
+    dump_body.insns_info.succ_index_table = 0;
+#endif
+    dump_body.insns_info.positions = ibf_dump_insns_info_positions(dump, &dump_body);
+#if VM_INSN_INFO_TABLE_IMPL == 2
+    ruby_xfree(positions);
+#endif
+    dump_body.local_table =          ibf_dump_local_table(dump, iseq);
+    dump_body.catch_table =          ibf_dump_catch_table(dump, iseq);
+    dump_body.parent_iseq =          ibf_dump_iseq(dump, iseq->body->parent_iseq);
+    dump_body.local_iseq =           ibf_dump_iseq(dump, iseq->body->local_iseq);
+    dump_body.is_entries =           NULL;
+    dump_body.ci_entries =           ibf_dump_ci_entries(dump, iseq);
+    dump_body.cc_entries =           NULL;
+    dump_body.variable.coverage      = Qnil;
+    dump_body.variable.original_iseq = NULL;
+
+    IBF_W_ALIGN(struct rb_iseq_constant_body);
+    return IBF_WV(dump_body);
+}
+
+static VALUE
+ibf_load_location_str(const struct ibf_load *load, VALUE str_index)
+{
+    VALUE str = ibf_load_object(load, str_index);
+    if (str != Qnil) {
+	str = rb_fstring(str);
+    }
+    return str;
+}
+
+static void
+ibf_load_iseq_each(const struct ibf_load *load, rb_iseq_t *iseq, ibf_offset_t offset)
+{
+    struct rb_iseq_constant_body *load_body = iseq->body = ZALLOC(struct rb_iseq_constant_body);
+    const struct rb_iseq_constant_body *body = (struct rb_iseq_constant_body *)(load->buff + offset);
+
+    /* memcpy(load_body, load->buff + offset, sizeof(*load_body)); */
+    load_body->type = body->type;
+    load_body->stack_max = body->stack_max;
+    load_body->param = body->param;
+    load_body->param.flags.has_kw = FALSE;
+    load_body->local_table_size = body->local_table_size;
+    load_body->is_size = body->is_size;
+    load_body->ci_size = body->ci_size;
+    load_body->ci_kw_size = body->ci_kw_size;
+    load_body->insns_info.size = body->insns_info.size;
+
+    ISEQ_COVERAGE_SET(iseq, Qnil);
+    ISEQ_ORIGINAL_ISEQ_CLEAR(iseq);
+    iseq->body->variable.flip_count = body->variable.flip_count;
+
+    {
+	VALUE realpath = Qnil, path = ibf_load_object(load, body->location.pathobj);
+	if (RB_TYPE_P(path, T_STRING)) {
+	    realpath = path = rb_fstring(path);
+	}
+	else if (RB_TYPE_P(path, T_ARRAY)) {
+	    VALUE pathobj = path;
+	    if (RARRAY_LEN(pathobj) != 2) {
+		rb_raise(rb_eRuntimeError, "path object size mismatch");
+	    }
+	    path = rb_fstring(RARRAY_AREF(pathobj, 0));
+	    realpath = RARRAY_AREF(pathobj, 1);
+	    if (!NIL_P(realpath)) {
+		if (!RB_TYPE_P(realpath, T_STRING)) {
+		    rb_raise(rb_eArgError, "unexpected realpath %"PRIxVALUE
+			     "(%x), path=%+"PRIsVALUE,
+			     realpath, TYPE(realpath), path);
+		}
+		realpath = rb_fstring(realpath);
+	    }
+	}
+	else {
+	    rb_raise(rb_eRuntimeError, "unexpected path object");
+	}
+	rb_iseq_pathobj_set(iseq, path, realpath);
+    }
+
+    RB_OBJ_WRITE(iseq, &load_body->location.base_label,    ibf_load_location_str(load, body->location.base_label));
+    RB_OBJ_WRITE(iseq, &load_body->location.label,         ibf_load_location_str(load, body->location.label));
+    load_body->location.first_lineno = body->location.first_lineno;
+    load_body->location.node_id = body->location.node_id;
+    load_body->location.code_location = body->location.code_location;
+    load_body->catch_except_p = body->catch_except_p;
+
+    load_body->is_entries           = ZALLOC_N(union iseq_inline_storage_entry, body->is_size);
+    load_body->ci_entries           = ibf_load_ci_entries(load, body);
+    load_body->cc_entries           = ZALLOC_N(struct rb_call_cache, body->ci_size + body->ci_kw_size);
+    load_body->param.opt_table      = ibf_load_param_opt_table(load, body);
+    load_body->param.keyword        = ibf_load_param_keyword(load, body);
+    load_body->param.flags.has_kw   = body->param.flags.has_kw;
+    load_body->insns_info.body      = ibf_load_insns_info_body(load, body);
+    load_body->insns_info.positions = ibf_load_insns_info_positions(load, body);
+    load_body->local_table          = ibf_load_local_table(load, body);
+    load_body->catch_table          = ibf_load_catch_table(load, body);
+    load_body->parent_iseq          = ibf_load_iseq(load, body->parent_iseq);
+    load_body->local_iseq           = ibf_load_iseq(load, body->local_iseq);
+
+    ibf_load_code(load, iseq, body);
+#if VM_INSN_INFO_TABLE_IMPL == 2
+    rb_iseq_insns_info_encode_positions(iseq);
+#endif
+
+    rb_iseq_translate_threaded_code(iseq);
+}
+
+
+static void
+ibf_dump_iseq_list(struct ibf_dump *dump, struct ibf_header *header)
+{
+    const long size = RARRAY_LEN(dump->iseq_list);
+    ibf_offset_t *list = ALLOCA_N(ibf_offset_t, size);
+    long i;
+
+    for (i=0; i<size; i++) {
+	list[i] = (ibf_offset_t)NUM2LONG(rb_ary_entry(dump->iseq_list, i));
+    }
+
+    ibf_dump_align(dump, sizeof(ibf_offset_t));
+    header->iseq_list_offset = ibf_dump_write(dump, list, sizeof(ibf_offset_t) * size);
+    header->iseq_list_size = (unsigned int)size;
+}
+
+struct ibf_dump_id_list_i_arg {
+    struct ibf_dump *dump;
+    long *list;
+    int current_i;
+};
+
+static int
+ibf_dump_id_list_i(st_data_t key, st_data_t val, st_data_t ptr)
+{
+    struct ibf_dump_id_list_i_arg *arg = (struct ibf_dump_id_list_i_arg *)ptr;
+    int i = (int)val;
+    ID id = (ID)key;
+    assert(arg->current_i == i);
+    arg->current_i++;
+
+    if (rb_id2name(id)) {
+	arg->list[i] = (long)ibf_dump_object(arg->dump, rb_id2str(id));
+    }
+    else {
+	arg->list[i] = 0;
+    }
+
+    return ST_CONTINUE;
+}
+
+static void
+ibf_dump_id_list(struct ibf_dump *dump, struct ibf_header *header)
+{
+    const long size = dump->id_table->num_entries;
+    struct ibf_dump_id_list_i_arg arg;
+    arg.list = ALLOCA_N(long, size);
+    arg.dump = dump;
+    arg.current_i = 0;
+
+    st_foreach(dump->id_table, ibf_dump_id_list_i, (st_data_t)&arg);
+
+    ibf_dump_align(dump, sizeof(long));
+    header->id_list_offset = ibf_dump_write(dump, arg.list, sizeof(long) * size);
+    header->id_list_size = (unsigned int)size;
+}
+
+#define IBF_OBJECT_INTERNAL FL_PROMOTED0
+
+/*
+ * Binary format
+ * - ibf_object_header
+ * - ibf_object_xxx (xxx is type)
+ */
+
+struct ibf_object_header {
+    unsigned int type: 5;
+    unsigned int special_const: 1;
+    unsigned int frozen: 1;
+    unsigned int internal: 1;
+};
+static const size_t ibf_object_header_align =
+    RUBY_ALIGNOF(struct ibf_object_header);
+
+enum ibf_object_class_index {
+    IBF_OBJECT_CLASS_OBJECT,
+    IBF_OBJECT_CLASS_ARRAY,
+    IBF_OBJECT_CLASS_STANDARD_ERROR
+};
+
+struct ibf_object_string {
+    long encindex;
+    long len;
+    char ptr[FLEX_ARY_LEN];
+};
+
+struct ibf_object_regexp {
+    long srcstr;
+    char option;
+};
+
+struct ibf_object_array {
+    long len;
+    long ary[FLEX_ARY_LEN];
+};
+
+struct ibf_object_hash {
+    long len;
+    long keyval[FLEX_ARY_LEN];
+};
+
+struct ibf_object_struct_range {
+    long class_index;
+    long len;
+    long beg;
+    long end;
+    int excl;
+};
+
+struct ibf_object_bignum {
+    ssize_t slen;
+    BDIGIT digits[FLEX_ARY_LEN];
+};
+
+enum ibf_object_data_type {
+    IBF_OBJECT_DATA_ENCODING
+};
+
+struct ibf_object_complex_rational {
+    long a, b;
+};
+
+struct ibf_object_symbol {
+    long str;
+};
+
+#define IBF_ALIGNED_OFFSET(align, offset) /* offset > 0 */ \
+    ((((offset) - 1) / (align) + 1) * (align))
+#define IBF_OBJHEADER(offset)     (const struct ibf_object_header *)\
+    ibf_load_check_offset(load, IBF_ALIGNED_OFFSET(ibf_object_header_align, offset))
+#define IBF_OBJBODY(type, offset) (const type *)\
+    ibf_load_check_offset(load, IBF_ALIGNED_OFFSET(RUBY_ALIGNOF(type), offset))
+
+static const void *
+ibf_load_check_offset(const struct ibf_load *load, size_t offset)
+{
+    if (offset >= (size_t)RSTRING_LEN(load->str)) {
+	rb_raise(rb_eIndexError, "object offset out of range: %"PRIdSIZE, offset);
+    }
+    return load->buff + offset;
+}
+
+NORETURN(static void ibf_dump_object_unsupported(struct ibf_dump *dump, VALUE obj));
+
+static void
+ibf_dump_object_unsupported(struct ibf_dump *dump, VALUE obj)
+{
+    char buff[0x100];
+    rb_raw_obj_info(buff, sizeof(buff), obj);
+    rb_raise(rb_eNotImpError, "ibf_dump_object_unsupported: %s", buff);
+}
+
+static VALUE
+ibf_load_object_unsupported(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    rb_raise(rb_eArgError, "unsupported");
+    return Qnil;
+}
+
+static void
+ibf_dump_object_class(struct ibf_dump *dump, VALUE obj)
+{
+    enum ibf_object_class_index cindex;
+    if (obj == rb_cObject) {
+	cindex = IBF_OBJECT_CLASS_OBJECT;
+    }
+    else if (obj == rb_cArray) {
+	cindex = IBF_OBJECT_CLASS_ARRAY;
+    }
+    else if (obj == rb_eStandardError) {
+	cindex = IBF_OBJECT_CLASS_STANDARD_ERROR;
+    }
+    else {
+	rb_obj_info_dump(obj);
+	rb_p(obj);
+	rb_bug("unsupported class");
+    }
+    ibf_dump_write(dump, &cindex, sizeof(cindex));
+}
+
+static VALUE
+ibf_load_object_class(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const enum ibf_object_class_index *cindexp = IBF_OBJBODY(enum ibf_object_class_index, offset);
+    enum ibf_object_class_index cindex = *cindexp;
+
+    switch (cindex) {
+      case IBF_OBJECT_CLASS_OBJECT:
+	return rb_cObject;
+      case IBF_OBJECT_CLASS_ARRAY:
+	return rb_cArray;
+      case IBF_OBJECT_CLASS_STANDARD_ERROR:
+	return rb_eStandardError;
+    }
+
+    rb_raise(rb_eArgError, "ibf_load_object_class: unknown class (%d)", (int)cindex);
+}
+
+
+static void
+ibf_dump_object_float(struct ibf_dump *dump, VALUE obj)
+{
+    double dbl = RFLOAT_VALUE(obj);
+    (void)IBF_W(&dbl, double, 1);
+}
+
+static VALUE
+ibf_load_object_float(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const double *dblp = IBF_OBJBODY(double, offset);
+    return DBL2NUM(*dblp);
+}
+
+static void
+ibf_dump_object_string(struct ibf_dump *dump, VALUE obj)
+{
+    long encindex = (long)rb_enc_get_index(obj);
+    long len = RSTRING_LEN(obj);
+    const char *ptr = RSTRING_PTR(obj);
+    long buff[2];
+
+    if (encindex > RUBY_ENCINDEX_BUILTIN_MAX) {
+	rb_encoding *enc = rb_enc_from_index((int)encindex);
+	const char *enc_name = rb_enc_name(enc);
+	encindex = RUBY_ENCINDEX_BUILTIN_MAX + ibf_dump_object(dump, rb_str_new2(enc_name));
+    }
+
+    buff[0] = encindex;
+    buff[1] = len;
+    (void)IBF_W(buff, long, 2);
+    IBF_WP(ptr, char, len);
+}
+
+static VALUE
+ibf_load_object_string(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const struct ibf_object_string *string = IBF_OBJBODY(struct ibf_object_string, offset);
+    VALUE str = rb_str_new(string->ptr, string->len);
+    int encindex = (int)string->encindex;
+
+    if (encindex > RUBY_ENCINDEX_BUILTIN_MAX) {
+	VALUE enc_name_str = ibf_load_object(load, encindex - RUBY_ENCINDEX_BUILTIN_MAX);
+	encindex = rb_enc_find_index(RSTRING_PTR(enc_name_str));
+    }
+    rb_enc_associate_index(str, encindex);
+
+    if (header->internal) rb_obj_hide(str);
+    if (header->frozen)   str = rb_fstring(str);
+
+    return str;
+}
+
+static void
+ibf_dump_object_regexp(struct ibf_dump *dump, VALUE obj)
+{
+    struct ibf_object_regexp regexp;
+    VALUE srcstr = RREGEXP_SRC(obj);
+    IBF_ZERO(regexp);
+    regexp.option = (char)rb_reg_options(obj);
+    regexp.srcstr = (long)ibf_dump_object(dump, srcstr);
+    (void)IBF_W(&regexp, struct ibf_object_regexp, 1);
+}
+
+static VALUE
+ibf_load_object_regexp(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const struct ibf_object_regexp *regexp = IBF_OBJBODY(struct ibf_object_regexp, offset);
+    VALUE srcstr = ibf_load_object(load, regexp->srcstr);
+    VALUE reg = rb_reg_compile(srcstr, (int)regexp->option, NULL, 0);
+
+    if (header->internal) rb_obj_hide(reg);
+    if (header->frozen)   rb_obj_freeze(reg);
+
+    return reg;
+}
+
+static void
+ibf_dump_object_array(struct ibf_dump *dump, VALUE obj)
+{
+    long i, len = (int)RARRAY_LEN(obj);
+    (void)IBF_W(&len, long, 1);
+    for (i=0; i<len; i++) {
+	long index = (long)ibf_dump_object(dump, RARRAY_AREF(obj, i));
+	IBF_WV(index);
+    }
+}
+
+static VALUE
+ibf_load_object_array(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const struct ibf_object_array *array = IBF_OBJBODY(struct ibf_object_array, offset);
+    VALUE ary = rb_ary_new_capa(array->len);
+    int i;
+
+    for (i=0; i<array->len; i++) {
+	rb_ary_push(ary, ibf_load_object(load, array->ary[i]));
+    }
+
+    if (header->internal) rb_obj_hide(ary);
+    if (header->frozen)   rb_obj_freeze(ary);
+
+    return ary;
+}
+
+static int
+ibf_dump_object_hash_i(st_data_t key, st_data_t val, st_data_t ptr)
+{
+    struct ibf_dump *dump = (struct ibf_dump *)ptr;
+    long keyval[2];
+    keyval[0] = (long)ibf_dump_object(dump, (VALUE)key);
+    keyval[1] = (long)ibf_dump_object(dump, (VALUE)val);
+    (void)IBF_W(keyval, long, 2);
+    return ST_CONTINUE;
+}
+
+static void
+ibf_dump_object_hash(struct ibf_dump *dump, VALUE obj)
+{
+    long len = RHASH_SIZE(obj);
+    (void)IBF_W(&len, long, 1);
+    if (len > 0) rb_hash_foreach(obj, ibf_dump_object_hash_i, (VALUE)dump);
+}
+
+static VALUE
+ibf_load_object_hash(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const struct ibf_object_hash *hash = IBF_OBJBODY(struct ibf_object_hash, offset);
+    VALUE obj = rb_hash_new_with_size(hash->len);
+    int i;
+
+    for (i=0; i<hash->len; i++) {
+	VALUE key = ibf_load_object(load, hash->keyval[i*2  ]);
+	VALUE val = ibf_load_object(load, hash->keyval[i*2+1]);
+	rb_hash_aset(obj, key, val);
+    }
+    rb_hash_rehash(obj);
+
+    if (header->internal) rb_obj_hide(obj);
+    if (header->frozen)   rb_obj_freeze(obj);
+
+    return obj;
+}
+
+static void
+ibf_dump_object_struct(struct ibf_dump *dump, VALUE obj)
+{
+    if (rb_obj_is_kind_of(obj, rb_cRange)) {
+	struct ibf_object_struct_range range;
+	VALUE beg, end;
+	IBF_ZERO(range);
+	range.len = 3;
+	range.class_index = 0;
+
+	rb_range_values(obj, &beg, &end, &range.excl);
+	range.beg = (long)ibf_dump_object(dump, beg);
+	range.end = (long)ibf_dump_object(dump, end);
+
+	IBF_W_ALIGN(struct ibf_object_struct_range);
+	IBF_WV(range);
+    }
+    else {
+        rb_raise(rb_eNotImpError, "ibf_dump_object_struct: unsupported class %"PRIsVALUE,
+                 rb_class_name(CLASS_OF(obj)));
+    }
+}
+
+static VALUE
+ibf_load_object_struct(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const struct ibf_object_struct_range *range = IBF_OBJBODY(struct ibf_object_struct_range, offset);
+    VALUE beg = ibf_load_object(load, range->beg);
+    VALUE end = ibf_load_object(load, range->end);
+    VALUE obj = rb_range_new(beg, end, range->excl);
+    if (header->internal) rb_obj_hide(obj);
+    if (header->frozen)   rb_obj_freeze(obj);
+    return obj;
+}
+
+static void
+ibf_dump_object_bignum(struct ibf_dump *dump, VALUE obj)
+{
+    ssize_t len = BIGNUM_LEN(obj);
+    ssize_t slen = BIGNUM_SIGN(obj) > 0 ? len : len * -1;
+    BDIGIT *d = BIGNUM_DIGITS(obj);
+
+    (void)IBF_W(&slen, ssize_t, 1);
+    IBF_WP(d, BDIGIT, len);
+}
+
+static VALUE
+ibf_load_object_bignum(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const struct ibf_object_bignum *bignum = IBF_OBJBODY(struct ibf_object_bignum, offset);
+    int sign = bignum->slen > 0;
+    ssize_t len = sign > 0 ? bignum->slen : -1 * bignum->slen;
+    VALUE obj = rb_integer_unpack(bignum->digits, len * 2, 2, 0,
+				  INTEGER_PACK_LITTLE_ENDIAN | (sign == 0 ? INTEGER_PACK_NEGATIVE : 0));
+    if (header->internal) rb_obj_hide(obj);
+    if (header->frozen)   rb_obj_freeze(obj);
+    return obj;
+}
+
+static void
+ibf_dump_object_data(struct ibf_dump *dump, VALUE obj)
+{
+    if (rb_data_is_encoding(obj)) {
+	rb_encoding *enc = rb_to_encoding(obj);
+	const char *name = rb_enc_name(enc);
+	long len = strlen(name) + 1;
+	long data[2];
+	data[0] = IBF_OBJECT_DATA_ENCODING;
+	data[1] = len;
+	(void)IBF_W(data, long, 2);
+	IBF_WP(name, char, len);
+    }
+    else {
+	ibf_dump_object_unsupported(dump, obj);
+    }
+}
+
+static VALUE
+ibf_load_object_data(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const long *body = IBF_OBJBODY(long, offset);
+    const enum ibf_object_data_type type = (enum ibf_object_data_type)body[0];
+    /* const long len = body[1]; */
+    const char *data = (const char *)&body[2];
+
+    switch (type) {
+      case IBF_OBJECT_DATA_ENCODING:
+	{
+	    VALUE encobj = rb_enc_from_encoding(rb_enc_find(data));
+	    return encobj;
+	}
+    }
+
+    return ibf_load_object_unsupported(load, header, offset);
+}
+
+static void
+ibf_dump_object_complex_rational(struct ibf_dump *dump, VALUE obj)
+{
+    long data[2];
+    data[0] = (long)ibf_dump_object(dump, RCOMPLEX(obj)->real);
+    data[1] = (long)ibf_dump_object(dump, RCOMPLEX(obj)->imag);
+
+    (void)IBF_W(data, long, 2);
+}
+
+static VALUE
+ibf_load_object_complex_rational(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    const struct ibf_object_complex_rational *nums = IBF_OBJBODY(struct ibf_object_complex_rational, offset);
+    VALUE a = ibf_load_object(load, nums->a);
+    VALUE b = ibf_load_object(load, nums->b);
+    VALUE obj = header->type == T_COMPLEX ?
+      rb_complex_new(a, b) : rb_rational_new(a, b);
+
+    if (header->internal) rb_obj_hide(obj);
+    if (header->frozen)   rb_obj_freeze(obj);
+    return obj;
+}
+
+static void
+ibf_dump_object_symbol(struct ibf_dump *dump, VALUE obj)
+{
+    VALUE str = rb_sym2str(obj);
+    long str_index = (long)ibf_dump_object(dump, str);
+    (void)IBF_W(&str_index, long, 1);
+}
+
+static VALUE
+ibf_load_object_symbol(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t offset)
+{
+    /* const struct ibf_object_header *header = IBF_OBJHEADER(offset); */
+    const struct ibf_object_symbol *symbol = IBF_OBJBODY(struct ibf_object_symbol, offset);
+    VALUE str = ibf_load_object(load, symbol->str);
+    ID id = rb_intern_str(str);
+    return ID2SYM(id);
+}
+
+typedef void (*ibf_dump_object_function)(struct ibf_dump *dump, VALUE obj);
+static ibf_dump_object_function dump_object_functions[RUBY_T_MASK+1] = {
+    ibf_dump_object_unsupported, /* T_NONE */
+    ibf_dump_object_unsupported, /* T_OBJECT */
+    ibf_dump_object_class,       /* T_CLASS */
+    ibf_dump_object_unsupported, /* T_MODULE */
+    ibf_dump_object_float,       /* T_FLOAT */
+    ibf_dump_object_string,      /* T_STRING */
+    ibf_dump_object_regexp,      /* T_REGEXP */
+    ibf_dump_object_array,       /* T_ARRAY */
+    ibf_dump_object_hash,        /* T_HASH */
+    ibf_dump_object_struct,      /* T_STRUCT */
+    ibf_dump_object_bignum,      /* T_BIGNUM */
+    ibf_dump_object_unsupported, /* T_FILE */
+    ibf_dump_object_data,        /* T_DATA */
+    ibf_dump_object_unsupported, /* T_MATCH */
+    ibf_dump_object_complex_rational, /* T_COMPLEX */
+    ibf_dump_object_complex_rational, /* T_RATIONAL */
+    ibf_dump_object_unsupported, /* 0x10 */
+    ibf_dump_object_unsupported, /* 0x11 T_NIL */
+    ibf_dump_object_unsupported, /* 0x12 T_TRUE */
+    ibf_dump_object_unsupported, /* 0x13 T_FALSE */
+    ibf_dump_object_symbol,      /* 0x14 T_SYMBOL */
+    ibf_dump_object_unsupported, /* T_FIXNUM */
+    ibf_dump_object_unsupported, /* T_UNDEF */
+    ibf_dump_object_unsupported, /* 0x17 */
+    ibf_dump_object_unsupported, /* 0x18 */
+    ibf_dump_object_unsupported, /* 0x19 */
+    ibf_dump_object_unsupported, /* T_IMEMO 0x1a */
+    ibf_dump_object_unsupported, /* T_NODE 0x1b */
+    ibf_dump_object_unsupported, /* T_ICLASS 0x1c */
+    ibf_dump_object_unsupported, /* T_ZOMBIE 0x1d */
+    ibf_dump_object_unsupported, /* 0x1e */
+    ibf_dump_object_unsupported, /* 0x1f */
+};
+
+static ibf_offset_t
+ibf_dump_object_object(struct ibf_dump *dump, VALUE obj)
+{
+    struct ibf_object_header obj_header;
+    ibf_offset_t current_offset;
+    IBF_ZERO(obj_header);
+    obj_header.type = TYPE(obj);
+
+    IBF_W_ALIGN(ibf_offset_t);
+    current_offset = ibf_dump_pos(dump);
+
+    if (SPECIAL_CONST_P(obj)) {
+	if (RB_TYPE_P(obj, T_SYMBOL) ||
+	    RB_TYPE_P(obj, T_FLOAT)) {
+	    obj_header.internal = FALSE;
+	    goto dump_object;
+	}
+	obj_header.special_const = TRUE;
+	obj_header.frozen = TRUE;
+	obj_header.internal = TRUE;
+	IBF_WV(obj_header);
+	(void)IBF_W(&obj, VALUE, 1);
+    }
+    else {
+	obj_header.internal = (RBASIC_CLASS(obj) == 0) ? TRUE : FALSE;
+      dump_object:
+	obj_header.special_const = FALSE;
+	obj_header.frozen = FL_TEST(obj, FL_FREEZE) ? TRUE : FALSE;
+	IBF_WV(obj_header);
+	(*dump_object_functions[obj_header.type])(dump, obj);
+    }
+
+    return current_offset;
+}
+
+typedef VALUE (*ibf_load_object_function)(const struct ibf_load *load, const struct ibf_object_header *header, ibf_offset_t);
+static ibf_load_object_function load_object_functions[RUBY_T_MASK+1] = {
+    ibf_load_object_unsupported, /* T_NONE */
+    ibf_load_object_unsupported, /* T_OBJECT */
+    ibf_load_object_class,       /* T_CLASS */
+    ibf_load_object_unsupported, /* T_MODULE */
+    ibf_load_object_float,       /* T_FLOAT */
+    ibf_load_object_string,      /* T_STRING */
+    ibf_load_object_regexp,      /* T_REGEXP */
+    ibf_load_object_array,       /* T_ARRAY */
+    ibf_load_object_hash,        /* T_HASH */
+    ibf_load_object_struct,      /* T_STRUCT */
+    ibf_load_object_bignum,      /* T_BIGNUM */
+    ibf_load_object_unsupported, /* T_FILE */
+    ibf_load_object_data,        /* T_DATA */
+    ibf_load_object_unsupported, /* T_MATCH */
+    ibf_load_object_complex_rational, /* T_COMPLEX */
+    ibf_load_object_complex_rational, /* T_RATIONAL */
+    ibf_load_object_unsupported, /* 0x10 */
+    ibf_load_object_unsupported, /* T_NIL */
+    ibf_load_object_unsupported, /* T_TRUE */
+    ibf_load_object_unsupported, /* T_FALSE */
+    ibf_load_object_symbol,
+    ibf_load_object_unsupported, /* T_FIXNUM */
+    ibf_load_object_unsupported, /* T_UNDEF */
+    ibf_load_object_unsupported, /* 0x17 */
+    ibf_load_object_unsupported, /* 0x18 */
+    ibf_load_object_unsupported, /* 0x19 */
+    ibf_load_object_unsupported, /* T_IMEMO 0x1a */
+    ibf_load_object_unsupported, /* T_NODE 0x1b */
+    ibf_load_object_unsupported, /* T_ICLASS 0x1c */
+    ibf_load_object_unsupported, /* T_ZOMBIE 0x1d */
+    ibf_load_object_unsupported, /* 0x1e */
+    ibf_load_object_unsupported, /* 0x1f */
+};
+
+static VALUE
+ibf_load_object(const struct ibf_load *load, VALUE object_index)
+{
+    if (object_index == 0) {
+	return Qnil;
+    }
+    else if (object_index >= load->header->object_list_size) {
+	rb_raise(rb_eIndexError, "object index out of range: %"PRIdVALUE, object_index);
+    }
+    else {
+	VALUE obj = rb_ary_entry(load->obj_list, (long)object_index);
+	if (obj == Qnil) { /* TODO: avoid multiple Qnil load */
+	    ibf_offset_t *offsets = (ibf_offset_t *)(load->header->object_list_offset + load->buff);
+	    ibf_offset_t offset = offsets[object_index];
+	    const struct ibf_object_header *header = IBF_OBJHEADER(offset);
+	    size_t value_offset;
+
+#if IBF_ISEQ_DEBUG
+	    fprintf(stderr, "ibf_load_object: list=%#x offsets=%p offset=%#x\n",
+	            load->header->object_list_offset, offsets, offset);
+	    fprintf(stderr, "ibf_load_object: type=%#x special=%d frozen=%d internal=%d\n",
+	            header->type, header->special_const, header->frozen, header->internal);
+#endif
+	    value_offset = (const char *)(header + 1) - load->buff;
+	    if (value_offset >= (size_t)RSTRING_LEN(load->str)) {
+		rb_raise(rb_eIndexError, "object offset out of range: %"PRIdSIZE, value_offset);
+	    }
+	    offset = (ibf_offset_t)value_offset;
+
+	    if (header->special_const) {
+		const VALUE *vp = IBF_OBJBODY(VALUE, offset);
+#if IBF_ISEQ_DEBUG
+		fprintf(stderr, "ibf_load_object: vp=%p\n", vp);
+#endif
+		obj = *vp;
+	    }
+	    else {
+		obj = (*load_object_functions[header->type])(load, header, offset);
+	    }
+
+	    rb_ary_store(load->obj_list, (long)object_index, obj);
+	}
+#if IBF_ISEQ_DEBUG
+	fprintf(stderr, "ibf_load_object: index=%#"PRIxVALUE" obj=%#"PRIxVALUE"\n",
+	        object_index, obj);
+#endif
+	return obj;
+    }
+}
+
+static void
+ibf_dump_object_list(struct ibf_dump *dump, struct ibf_header *header)
+{
+    VALUE list = rb_ary_tmp_new(RARRAY_LEN(dump->obj_list));
+    int i, size;
+
+    for (i=0; i<RARRAY_LEN(dump->obj_list); i++) {
+	VALUE obj = RARRAY_AREF(dump->obj_list, i);
+	ibf_offset_t offset = ibf_dump_object_object(dump, obj);
+	rb_ary_push(list, UINT2NUM(offset));
+    }
+    size = i;
+    IBF_W_ALIGN(ibf_offset_t);
+    header->object_list_offset = ibf_dump_pos(dump);
+
+    for (i=0; i<size; i++) {
+	ibf_offset_t offset = NUM2UINT(RARRAY_AREF(list, i));
+	IBF_WV(offset);
+    }
+
+    header->object_list_size = size;
+}
+
+static void
+ibf_dump_mark(void *ptr)
+{
+    struct ibf_dump *dump = (struct ibf_dump *)ptr;
+    rb_gc_mark(dump->str);
+    rb_gc_mark(dump->iseq_list);
+    rb_gc_mark(dump->obj_list);
+}
+
+static void
+ibf_dump_free(void *ptr)
+{
+    struct ibf_dump *dump = (struct ibf_dump *)ptr;
+    if (dump->iseq_table) {
+	st_free_table(dump->iseq_table);
+	dump->iseq_table = 0;
+    }
+    if (dump->id_table) {
+	st_free_table(dump->id_table);
+	dump->id_table = 0;
+    }
+    ruby_xfree(dump);
+}
+
+static size_t
+ibf_dump_memsize(const void *ptr)
+{
+    struct ibf_dump *dump = (struct ibf_dump *)ptr;
+    size_t size = sizeof(*dump);
+    if (dump->iseq_table) size += st_memsize(dump->iseq_table);
+    if (dump->id_table) size += st_memsize(dump->id_table);
+    return size;
+}
+
+static const rb_data_type_t ibf_dump_type = {
+    "ibf_dump",
+    {ibf_dump_mark, ibf_dump_free, ibf_dump_memsize,},
+    0, 0, RUBY_TYPED_WB_PROTECTED | RUBY_TYPED_FREE_IMMEDIATELY
+};
+
+static void
+ibf_dump_setup(struct ibf_dump *dump, VALUE dumper_obj)
+{
+    RB_OBJ_WRITE(dumper_obj, &dump->str, rb_str_new(0, 0));
+    RB_OBJ_WRITE(dumper_obj, &dump->iseq_list, rb_ary_tmp_new(0));
+    RB_OBJ_WRITE(dumper_obj, &dump->obj_list, rb_ary_tmp_new(1));
+    rb_ary_push(dump->obj_list, Qnil); /* 0th is nil */
+    dump->iseq_table = st_init_numtable(); /* need free */
+    dump->id_table = st_init_numtable();   /* need free */
+
+    ibf_table_index(dump->id_table, 0); /* id_index:0 is 0 */
+}
+
+VALUE
+rb_iseq_ibf_dump(const rb_iseq_t *iseq, VALUE opt)
+{
+    struct ibf_dump *dump;
+    struct ibf_header header = {{0}};
+    VALUE dump_obj;
+    VALUE str;
+
+    if (iseq->body->parent_iseq != NULL ||
+	iseq->body->local_iseq != iseq) {
+	rb_raise(rb_eRuntimeError, "should be top of iseq");
+    }
+    if (RTEST(ISEQ_COVERAGE(iseq))) {
+	rb_raise(rb_eRuntimeError, "should not compile with coverage");
+    }
+
+    dump_obj = TypedData_Make_Struct(0, struct ibf_dump, &ibf_dump_type, dump);
+    ibf_dump_setup(dump, dump_obj);
+
+    ibf_dump_write(dump, &header, sizeof(header));
+    ibf_dump_write(dump, RUBY_PLATFORM, strlen(RUBY_PLATFORM) + 1);
+    ibf_dump_iseq(dump, iseq);
+
+    header.magic[0] = 'Y'; /* YARB */
+    header.magic[1] = 'A';
+    header.magic[2] = 'R';
+    header.magic[3] = 'B';
+    header.major_version = ISEQ_MAJOR_VERSION;
+    header.minor_version = ISEQ_MINOR_VERSION;
+    ibf_dump_iseq_list(dump, &header);
+    ibf_dump_id_list(dump, &header);
+    ibf_dump_object_list(dump, &header);
+    header.size = ibf_dump_pos(dump);
+
+    if (RTEST(opt)) {
+	VALUE opt_str = opt;
+	const char *ptr = StringValuePtr(opt_str);
+	header.extra_size = RSTRING_LENINT(opt_str);
+	ibf_dump_write(dump, ptr, header.extra_size);
+    }
+    else {
+	header.extra_size = 0;
+    }
+
+    ibf_dump_overwrite(dump, &header, sizeof(header), 0);
+
+    str = dump->str;
+    ibf_dump_free(dump);
+    DATA_PTR(dump_obj) = NULL;
+    RB_GC_GUARD(dump_obj);
+    return str;
+}
+
+static const ibf_offset_t *
+ibf_iseq_list(const struct ibf_load *load)
+{
+    return (ibf_offset_t *)(load->buff + load->header->iseq_list_offset);
+}
+
+void
+rb_ibf_load_iseq_complete(rb_iseq_t *iseq)
+{
+    struct ibf_load *load = RTYPEDDATA_DATA(iseq->aux.loader.obj);
+    rb_iseq_t *prev_src_iseq = load->iseq;
+    const ibf_offset_t offset = ibf_iseq_list(load)[iseq->aux.loader.index];
+    load->iseq = iseq;
+#if IBF_ISEQ_DEBUG
+    fprintf(stderr, "rb_ibf_load_iseq_complete: index=%#x offset=%#x size=%#x\n",
+	    iseq->aux.loader.index, offset,
+	    load->header->size);
+#endif
+    if (offset % sizeof(VALUE)) {
+        rb_raise(rb_eArgError, "unaligned iseq offset: %#x @ %u",
+                 offset, iseq->aux.loader.index);
+    }
+    ibf_load_iseq_each(load, iseq, offset);
+    ISEQ_COMPILE_DATA_CLEAR(iseq);
+    FL_UNSET(iseq, ISEQ_NOT_LOADED_YET);
+    rb_iseq_init_trace(iseq);
+    load->iseq = prev_src_iseq;
+}
+
+#if USE_LAZY_LOAD
+const rb_iseq_t *
+rb_iseq_complete(const rb_iseq_t *iseq)
+{
+    rb_ibf_load_iseq_complete((rb_iseq_t *)iseq);
+    return iseq;
+}
+#endif
+
+static rb_iseq_t *
+ibf_load_iseq(const struct ibf_load *load, const rb_iseq_t *index_iseq)
+{
+    int iseq_index = (int)(VALUE)index_iseq;
+
+#if IBF_ISEQ_DEBUG
+    fprintf(stderr, "ibf_load_iseq: index_iseq=%p iseq_list=%p\n",
+	    index_iseq, (void *)load->iseq_list);
+#endif
+    if (iseq_index == -1) {
+	return NULL;
+    }
+    else {
+	VALUE iseqv = rb_ary_entry(load->iseq_list, iseq_index);
+
+#if IBF_ISEQ_DEBUG
+	fprintf(stderr, "ibf_load_iseq: iseqv=%p\n", (void *)iseqv);
+#endif
+	if (iseqv != Qnil) {
+	    return (rb_iseq_t *)iseqv;
+	}
+	else {
+	    rb_iseq_t *iseq = iseq_imemo_alloc();
+#if IBF_ISEQ_DEBUG
+	    fprintf(stderr, "ibf_load_iseq: new iseq=%p\n", iseq);
+#endif
+	    FL_SET(iseq, ISEQ_NOT_LOADED_YET);
+	    iseq->aux.loader.obj = load->loader_obj;
+	    iseq->aux.loader.index = iseq_index;
+#if IBF_ISEQ_DEBUG
+	    fprintf(stderr, "ibf_load_iseq: iseq=%p loader_obj=%p index=%d\n",
+		    iseq, (void *)load->loader_obj, iseq_index);
+#endif
+	    rb_ary_store(load->iseq_list, iseq_index, (VALUE)iseq);
+
+#if !USE_LAZY_LOAD
+#if IBF_ISEQ_DEBUG
+	    fprintf(stderr, "ibf_load_iseq: loading iseq=%p\n", iseq);
+#endif
+	    rb_ibf_load_iseq_complete(iseq);
+#endif /* !USE_LAZY_LOAD */
+
+#if IBF_ISEQ_DEBUG
+	    fprintf(stderr, "ibf_load_iseq: iseq=%p loaded %p\n",
+		    iseq, load->iseq);
+#endif
+	    return iseq;
+	}
+    }
+}
+
+static void
+ibf_load_setup(struct ibf_load *load, VALUE loader_obj, VALUE str)
+{
+    rb_check_safe_obj(str);
+
+    if (RSTRING_LENINT(str) < (int)sizeof(struct ibf_header)) {
+	rb_raise(rb_eRuntimeError, "broken binary format");
+    }
+    RB_OBJ_WRITE(loader_obj, &load->str, str);
+    load->loader_obj = loader_obj;
+    load->buff = StringValuePtr(str);
+    load->header = (struct ibf_header *)load->buff;
+    RB_OBJ_WRITE(loader_obj, &load->iseq_list, rb_ary_tmp_new(0));
+    RB_OBJ_WRITE(loader_obj, &load->obj_list, rb_ary_tmp_new(0));
+    load->id_list = ZALLOC_N(ID, load->header->id_list_size);
+    load->iseq = NULL;
+
+    if (RSTRING_LENINT(str) < (int)load->header->size) {
+	rb_raise(rb_eRuntimeError, "broken binary format");
+    }
+    if (strncmp(load->header->magic, "YARB", 4) != 0) {
+	rb_raise(rb_eRuntimeError, "unknown binary format");
+    }
+    if (load->header->major_version != ISEQ_MAJOR_VERSION ||
+	load->header->minor_version != ISEQ_MINOR_VERSION) {
+	rb_raise(rb_eRuntimeError, "unmatched version file (%u.%u for %u.%u)",
+		 load->header->major_version, load->header->minor_version, ISEQ_MAJOR_VERSION, ISEQ_MINOR_VERSION);
+    }
+    if (strcmp(load->buff + sizeof(struct ibf_header), RUBY_PLATFORM) != 0) {
+	rb_raise(rb_eRuntimeError, "unmatched platform");
+    }
+    if (load->header->iseq_list_offset % RUBY_ALIGNOF(ibf_offset_t)) {
+        rb_raise(rb_eArgError, "unaligned iseq list offset: %u",
+                 load->header->iseq_list_offset);
+    }
+    if (load->header->id_list_offset % RUBY_ALIGNOF(long)) {
+        rb_raise(rb_eArgError, "unaligned ID list offset: %u",
+                 load->header->id_list_offset);
+    }
+    if (load->header->object_list_offset % RUBY_ALIGNOF(ibf_offset_t)) {
+        rb_raise(rb_eArgError, "unaligned object list offset: %u",
+                 load->header->object_list_offset);
+    }
+}
+
+static void
+ibf_loader_mark(void *ptr)
+{
+    struct ibf_load *load = (struct ibf_load *)ptr;
+    rb_gc_mark(load->str);
+    rb_gc_mark(load->iseq_list);
+    rb_gc_mark(load->obj_list);
+}
+
+static void
+ibf_loader_free(void *ptr)
+{
+    struct ibf_load *load = (struct ibf_load *)ptr;
+    ruby_xfree(load->id_list);
+    ruby_xfree(load);
+}
+
+static size_t
+ibf_loader_memsize(const void *ptr)
+{
+    struct ibf_load *load = (struct ibf_load *)ptr;
+    return sizeof(struct ibf_load) + load->header->id_list_size * sizeof(ID);
+}
+
+static const rb_data_type_t ibf_load_type = {
+    "ibf_loader",
+    {ibf_loader_mark, ibf_loader_free, ibf_loader_memsize,},
+    0, 0, RUBY_TYPED_WB_PROTECTED | RUBY_TYPED_FREE_IMMEDIATELY
+};
+
+const rb_iseq_t *
+rb_iseq_ibf_load(VALUE str)
+{
+    struct ibf_load *load;
+    rb_iseq_t *iseq;
+    VALUE loader_obj = TypedData_Make_Struct(0, struct ibf_load, &ibf_load_type, load);
+
+    ibf_load_setup(load, loader_obj, str);
+    iseq = ibf_load_iseq(load, 0);
+
+    RB_GC_GUARD(loader_obj);
+    return iseq;
+}
+
+VALUE
+rb_iseq_ibf_load_extra_data(VALUE str)
+{
+    struct ibf_load *load;
+    VALUE loader_obj = TypedData_Make_Struct(0, struct ibf_load, &ibf_load_type, load);
+    VALUE extra_str;
+
+    ibf_load_setup(load, loader_obj, str);
+    extra_str = rb_str_new(load->buff + load->header->size, load->header->extra_size);
+    RB_GC_GUARD(loader_obj);
+    return extra_str;
+}
diff --git a/complex.c b/complex.c
new file mode 100644
index 0000000000..685d4bc075
--- /dev/null
+++ b/complex.c
@@ -0,0 +1,2300 @@
+/*
+  complex.c: Coded by Tadayoshi Funaba 2008-2012
+
+  This implementation is based on Keiju Ishitsuka's Complex library
+  which is written in ruby.
+*/
+
+#include "ruby/config.h"
+#if defined _MSC_VER
+/* Microsoft Visual C does not define M_PI and others by default */
+# define _USE_MATH_DEFINES 1
+#endif
+#include <math.h>
+#include "internal.h"
+#include "id.h"
+
+#define NDEBUG
+#include "ruby_assert.h"
+
+#define ZERO INT2FIX(0)
+#define ONE INT2FIX(1)
+#define TWO INT2FIX(2)
+#if USE_FLONUM
+#define RFLOAT_0 DBL2NUM(0)
+#else
+static VALUE RFLOAT_0;
+#endif
+#if defined(HAVE_SIGNBIT) && defined(__GNUC__) && defined(__sun) && \
+    !defined(signbit)
+extern int signbit(double);
+#endif
+
+VALUE rb_cComplex;
+
+static ID id_abs, id_arg,
+    id_denominator, id_fdiv, id_numerator, id_quo,
+    id_real_p, id_i_real, id_i_imag,
+    id_finite_p, id_infinite_p, id_rationalize,
+    id_PI;
+#define id_to_i idTo_i
+#define id_to_r idTo_r
+#define id_negate idUMinus
+#define id_expt idPow
+#define id_to_f idTo_f
+
+#define f_boolcast(x) ((x) ? Qtrue : Qfalse)
+
+#define binop(n,op) \
+inline static VALUE \
+f_##n(VALUE x, VALUE y)\
+{\
+    return rb_funcall(x, (op), 1, y);\
+}
+
+#define fun1(n) \
+inline static VALUE \
+f_##n(VALUE x)\
+{\
+    return rb_funcall(x, id_##n, 0);\
+}
+
+#define fun2(n) \
+inline static VALUE \
+f_##n(VALUE x, VALUE y)\
+{\
+    return rb_funcall(x, id_##n, 1, y);\
+}
+
+#define PRESERVE_SIGNEDZERO
+
+inline static VALUE
+f_add(VALUE x, VALUE y)
+{
+    if (FIXNUM_ZERO_P(y))
+	return x;
+    if (FIXNUM_ZERO_P(x))
+	return y;
+    return rb_funcall(x, '+', 1, y);
+}
+
+inline static VALUE
+f_div(VALUE x, VALUE y)
+{
+    if (FIXNUM_P(y) && FIX2LONG(y) == 1)
+	return x;
+    return rb_funcall(x, '/', 1, y);
+}
+
+inline static int
+f_gt_p(VALUE x, VALUE y)
+{
+    if (RB_INTEGER_TYPE_P(x)) {
+        if (FIXNUM_P(x) && FIXNUM_P(y))
+            return (SIGNED_VALUE)x > (SIGNED_VALUE)y;
+        return RTEST(rb_int_gt(x, y));
+    }
+    else if (RB_FLOAT_TYPE_P(x))
+        return RTEST(rb_float_gt(x, y));
+    else if (RB_TYPE_P(x, T_RATIONAL)) {
+        int const cmp = rb_cmpint(rb_rational_cmp(x, y), x, y);
+        return cmp > 0;
+    }
+    return RTEST(rb_funcall(x, '>', 1, y));
+}
+
+inline static VALUE
+f_mul(VALUE x, VALUE y)
+{
+    if (FIXNUM_ZERO_P(y) && RB_INTEGER_TYPE_P(x))
+	return ZERO;
+    if (FIXNUM_ZERO_P(x) && RB_INTEGER_TYPE_P(y))
+	return ZERO;
+    if (y == ONE) return x;
+    if (x == ONE) return y;
+    return rb_funcall(x, '*', 1, y);
+}
+
+inline static VALUE
+f_sub(VALUE x, VALUE y)
+{
+    if (FIXNUM_ZERO_P(y))
+	return x;
+    return rb_funcall(x, '-', 1, y);
+}
+
+fun1(abs)
+fun1(arg)
+fun1(denominator)
+
+inline static VALUE
+f_negate(VALUE x)
+{
+    if (RB_INTEGER_TYPE_P(x)) {
+        return rb_int_uminus(x);
+    }
+    else if (RB_FLOAT_TYPE_P(x)) {
+        return rb_float_uminus(x);
+    }
+    else if (RB_TYPE_P(x, T_RATIONAL)) {
+        return rb_rational_uminus(x);
+    }
+    else if (RB_TYPE_P(x, T_COMPLEX)) {
+        return rb_complex_uminus(x);
+    }
+    return rb_funcall(x, id_negate, 0);
+}
+
+fun1(numerator)
+fun1(real_p)
+
+inline static VALUE
+f_to_i(VALUE x)
+{
+    if (RB_TYPE_P(x, T_STRING))
+	return rb_str_to_inum(x, 10, 0);
+    return rb_funcall(x, id_to_i, 0);
+}
+inline static VALUE
+f_to_f(VALUE x)
+{
+    if (RB_TYPE_P(x, T_STRING))
+	return DBL2NUM(rb_str_to_dbl(x, 0));
+    return rb_funcall(x, id_to_f, 0);
+}
+
+fun1(to_r)
+
+inline static int
+f_eqeq_p(VALUE x, VALUE y)
+{
+    if (FIXNUM_P(x) && FIXNUM_P(y))
+	return x == y;
+    else if (RB_FLOAT_TYPE_P(x) || RB_FLOAT_TYPE_P(y))
+	return NUM2DBL(x) == NUM2DBL(y);
+    return (int)rb_equal(x, y);
+}
+
+fun2(expt)
+fun2(fdiv)
+fun2(quo)
+
+inline static int
+f_negative_p(VALUE x)
+{
+    if (RB_INTEGER_TYPE_P(x))
+        return INT_NEGATIVE_P(x);
+    else if (RB_FLOAT_TYPE_P(x))
+        return RFLOAT_VALUE(x) < 0.0;
+    else if (RB_TYPE_P(x, T_RATIONAL))
+        return INT_NEGATIVE_P(RRATIONAL(x)->num);
+    return rb_num_negative_p(x);
+}
+
+#define f_positive_p(x) (!f_negative_p(x))
+
+inline static int
+f_zero_p(VALUE x)
+{
+    if (RB_INTEGER_TYPE_P(x)) {
+        return FIXNUM_ZERO_P(x);
+    }
+    else if (RB_TYPE_P(x, T_RATIONAL)) {
+        const VALUE num = RRATIONAL(x)->num;
+        return FIXNUM_ZERO_P(num);
+    }
+    return (int)rb_equal(x, ZERO);
+}
+
+#define f_nonzero_p(x) (!f_zero_p(x))
+
+VALUE rb_flo_is_finite_p(VALUE num);
+inline static int
+f_finite_p(VALUE x)
+{
+    if (RB_INTEGER_TYPE_P(x)) {
+        return TRUE;
+    }
+    else if (RB_FLOAT_TYPE_P(x)) {
+	return (int)rb_flo_is_finite_p(x);
+    }
+    else if (RB_TYPE_P(x, T_RATIONAL)) {
+	return TRUE;
+    }
+    return RTEST(rb_funcallv(x, id_finite_p, 0, 0));
+}
+
+VALUE rb_flo_is_infinite_p(VALUE num);
+inline static VALUE
+f_infinite_p(VALUE x)
+{
+    if (RB_INTEGER_TYPE_P(x)) {
+        return Qnil;
+    }
+    else if (RB_FLOAT_TYPE_P(x)) {
+	return rb_flo_is_infinite_p(x);
+    }
+    else if (RB_TYPE_P(x, T_RATIONAL)) {
+        return Qnil;
+    }
+    return rb_funcallv(x, id_infinite_p, 0, 0);
+}
+
+inline static int
+f_kind_of_p(VALUE x, VALUE c)
+{
+    return (int)rb_obj_is_kind_of(x, c);
+}
+
+inline static int
+k_numeric_p(VALUE x)
+{
+    return f_kind_of_p(x, rb_cNumeric);
+}
+
+#define k_exact_p(x) (!RB_FLOAT_TYPE_P(x))
+
+#define k_exact_zero_p(x) (k_exact_p(x) && f_zero_p(x))
+
+#define get_dat1(x) \
+    struct RComplex *dat = RCOMPLEX(x)
+
+#define get_dat2(x,y) \
+    struct RComplex *adat = RCOMPLEX(x), *bdat = RCOMPLEX(y)
+
+inline static VALUE
+nucomp_s_new_internal(VALUE klass, VALUE real, VALUE imag)
+{
+    NEWOBJ_OF(obj, struct RComplex, klass, T_COMPLEX | (RGENGC_WB_PROTECTED_COMPLEX ? FL_WB_PROTECTED : 0));
+
+    RCOMPLEX_SET_REAL(obj, real);
+    RCOMPLEX_SET_IMAG(obj, imag);
+    OBJ_FREEZE_RAW(obj);
+
+    return (VALUE)obj;
+}
+
+static VALUE
+nucomp_s_alloc(VALUE klass)
+{
+    return nucomp_s_new_internal(klass, ZERO, ZERO);
+}
+
+inline static VALUE
+f_complex_new_bang1(VALUE klass, VALUE x)
+{
+    assert(!RB_TYPE_P(x, T_COMPLEX));
+    return nucomp_s_new_internal(klass, x, ZERO);
+}
+
+inline static VALUE
+f_complex_new_bang2(VALUE klass, VALUE x, VALUE y)
+{
+    assert(!RB_TYPE_P(x, T_COMPLEX));
+    assert(!RB_TYPE_P(y, T_COMPLEX));
+    return nucomp_s_new_internal(klass, x, y);
+}
+
+#ifdef CANONICALIZATION_FOR_MATHN
+static int canonicalization = 0;
+
+RUBY_FUNC_EXPORTED void
+nucomp_canonicalization(int f)
+{
+    canonicalization = f;
+}
+#else
+#define canonicalization 0
+#endif
+
+inline static void
+nucomp_real_check(VALUE num)
+{
+    if (!RB_INTEGER_TYPE_P(num) &&
+	!RB_FLOAT_TYPE_P(num) &&
+	!RB_TYPE_P(num, T_RATIONAL)) {
+	if (!k_numeric_p(num) || !f_real_p(num))
+	    rb_raise(rb_eTypeError, "not a real");
+    }
+}
+
+inline static VALUE
+nucomp_s_canonicalize_internal(VALUE klass, VALUE real, VALUE imag)
+{
+    int complex_r, complex_i;
+#ifdef CANONICALIZATION_FOR_MATHN
+    if (k_exact_zero_p(imag) && canonicalization)
+	return real;
+#endif
+    complex_r = RB_TYPE_P(real, T_COMPLEX);
+    complex_i = RB_TYPE_P(imag, T_COMPLEX);
+    if (!complex_r && !complex_i) {
+	return nucomp_s_new_internal(klass, real, imag);
+    }
+    else if (!complex_r) {
+	get_dat1(imag);
+
+	return nucomp_s_new_internal(klass,
+				     f_sub(real, dat->imag),
+				     f_add(ZERO, dat->real));
+    }
+    else if (!complex_i) {
+	get_dat1(real);
+
+	return nucomp_s_new_internal(klass,
+				     dat->real,
+				     f_add(dat->imag, imag));
+    }
+    else {
+	get_dat2(real, imag);
+
+	return nucomp_s_new_internal(klass,
+				     f_sub(adat->real, bdat->imag),
+				     f_add(adat->imag, bdat->real));
+    }
+}
+
+/*
+ * call-seq:
+ *    Complex.rect(real[, imag])         ->  complex
+ *    Complex.rectangular(real[, imag])  ->  complex
+ *
+ * Returns a complex object which denotes the given rectangular form.
+ *
+ *    Complex.rectangular(1, 2)  #=> (1+2i)
+ */
+static VALUE
+nucomp_s_new(int argc, VALUE *argv, VALUE klass)
+{
+    VALUE real, imag;
+
+    switch (rb_scan_args(argc, argv, "11", &real, &imag)) {
+      case 1:
+	nucomp_real_check(real);
+	imag = ZERO;
+	break;
+      default:
+	nucomp_real_check(real);
+	nucomp_real_check(imag);
+	break;
+    }
+
+    return nucomp_s_canonicalize_internal(klass, real, imag);
+}
+
+inline static VALUE
+f_complex_new2(VALUE klass, VALUE x, VALUE y)
+{
+    assert(!RB_TYPE_P(x, T_COMPLEX));
+    return nucomp_s_canonicalize_internal(klass, x, y);
+}
+
+static VALUE nucomp_convert(VALUE klass, VALUE a1, VALUE a2, int raise);
+static VALUE nucomp_s_convert(int argc, VALUE *argv, VALUE klass);
+
+/*
+ * call-seq:
+ *    Complex(x[, y])  ->  numeric
+ *
+ * Returns x+i*y;
+ *
+ *    Complex(1, 2)    #=> (1+2i)
+ *    Complex('1+2i')  #=> (1+2i)
+ *    Complex(nil)     #=> TypeError
+ *    Complex(1, nil)  #=> TypeError
+ *
+ * Syntax of string form:
+ *
+ *   string form = extra spaces , complex , extra spaces ;
+ *   complex = real part | [ sign ] , imaginary part
+ *           | real part , sign , imaginary part
+ *           | rational , "@" , rational ;
+ *   real part = rational ;
+ *   imaginary part = imaginary unit | unsigned rational , imaginary unit ;
+ *   rational = [ sign ] , unsigned rational ;
+ *   unsigned rational = numerator | numerator , "/" , denominator ;
+ *   numerator = integer part | fractional part | integer part , fractional part ;
+ *   denominator = digits ;
+ *   integer part = digits ;
+ *   fractional part = "." , digits , [ ( "e" | "E" ) , [ sign ] , digits ] ;
+ *   imaginary unit = "i" | "I" | "j" | "J" ;
+ *   sign = "-" | "+" ;
+ *   digits = digit , { digit | "_" , digit };
+ *   digit = "0" | "1" | "2" | "3" | "4" | "5" | "6" | "7" | "8" | "9" ;
+ *   extra spaces = ? \s* ? ;
+ *
+ * See String#to_c.
+ */
+static VALUE
+nucomp_f_complex(int argc, VALUE *argv, VALUE klass)
+{
+    VALUE a1, a2, opts = Qnil;
+    int raise = TRUE;
+
+    if (rb_scan_args(argc, argv, "11:", &a1, &a2, &opts) == 1) {
+        a2 = Qundef;
+    }
+    if (!NIL_P(opts)) {
+        static ID kwds[1];
+        VALUE exception;
+        if (!kwds[0]) {
+            kwds[0] = rb_intern_const("exception");
+        }
+        rb_get_kwargs(opts, kwds, 0, 1, &exception);
+        raise = (exception != Qfalse);
+    }
+    return nucomp_convert(rb_cComplex, a1, a2, raise);
+}
+
+#define imp1(n) \
+inline static VALUE \
+m_##n##_bang(VALUE x)\
+{\
+    return rb_math_##n(x);\
+}
+
+imp1(cos)
+imp1(cosh)
+imp1(exp)
+
+static VALUE
+m_log_bang(VALUE x)
+{
+    return rb_math_log(1, &x);
+}
+
+imp1(sin)
+imp1(sinh)
+
+static VALUE
+m_cos(VALUE x)
+{
+    if (!RB_TYPE_P(x, T_COMPLEX))
+	return m_cos_bang(x);
+    {
+	get_dat1(x);
+	return f_complex_new2(rb_cComplex,
+			      f_mul(m_cos_bang(dat->real),
+				    m_cosh_bang(dat->imag)),
+			      f_mul(f_negate(m_sin_bang(dat->real)),
+				    m_sinh_bang(dat->imag)));
+    }
+}
+
+static VALUE
+m_sin(VALUE x)
+{
+    if (!RB_TYPE_P(x, T_COMPLEX))
+	return m_sin_bang(x);
+    {
+	get_dat1(x);
+	return f_complex_new2(rb_cComplex,
+			      f_mul(m_sin_bang(dat->real),
+				    m_cosh_bang(dat->imag)),
+			      f_mul(m_cos_bang(dat->real),
+				    m_sinh_bang(dat->imag)));
+    }
+}
+
+static VALUE
+f_complex_polar(VALUE klass, VALUE x, VALUE y)
+{
+    assert(!RB_TYPE_P(x, T_COMPLEX));
+    assert(!RB_TYPE_P(y, T_COMPLEX));
+    if (f_zero_p(x) || f_zero_p(y)) {
+	if (canonicalization) return x;
+	return nucomp_s_new_internal(klass, x, RFLOAT_0);
+    }
+    if (RB_FLOAT_TYPE_P(y)) {
+	const double arg = RFLOAT_VALUE(y);
+	if (arg == M_PI) {
+	    x = f_negate(x);
+	    if (canonicalization) return x;
+	    y = RFLOAT_0;
+	}
+	else if (arg == M_PI_2) {
+	    y = x;
+	    x = RFLOAT_0;
+	}
+	else if (arg == M_PI_2+M_PI) {
+	    y = f_negate(x);
+	    x = RFLOAT_0;
+	}
+	else if (RB_FLOAT_TYPE_P(x)) {
+	    const double abs = RFLOAT_VALUE(x);
+	    const double real = abs * cos(arg), imag = abs * sin(arg);
+	    x = DBL2NUM(real);
+	    if (canonicalization && imag == 0.0) return x;
+	    y = DBL2NUM(imag);
+	}
+	else {
+	    y = f_mul(x, DBL2NUM(sin(arg)));
+	    x = f_mul(x, DBL2NUM(cos(arg)));
+	    if (canonicalization && f_zero_p(y)) return x;
+	}
+	return nucomp_s_new_internal(klass, x, y);
+    }
+    return nucomp_s_canonicalize_internal(klass,
+					  f_mul(x, m_cos(y)),
+					  f_mul(x, m_sin(y)));
+}
+
+/* returns a Complex or Float of ang*PI-rotated abs */
+VALUE
+rb_dbl_complex_new_polar_pi(double abs, double ang)
+{
+    double fi;
+    const double fr = modf(ang, &fi);
+    int pos = fr == +0.5;
+
+    if (pos || fr == -0.5) {
+	if ((modf(fi / 2.0, &fi) != fr) ^ pos) abs = -abs;
+	return rb_complex_new(RFLOAT_0, DBL2NUM(abs));
+    }
+    else if (fr == 0.0) {
+	if (modf(fi / 2.0, &fi) != 0.0) abs = -abs;
+	return DBL2NUM(abs);
+    }
+    else {
+	ang *= M_PI;
+	return rb_complex_new(DBL2NUM(abs * cos(ang)), DBL2NUM(abs * sin(ang)));
+    }
+}
+
+/*
+ * call-seq:
+ *    Complex.polar(abs[, arg])  ->  complex
+ *
+ * Returns a complex object which denotes the given polar form.
+ *
+ *    Complex.polar(3, 0)            #=> (3.0+0.0i)
+ *    Complex.polar(3, Math::PI/2)   #=> (1.836909530733566e-16+3.0i)
+ *    Complex.polar(3, Math::PI)     #=> (-3.0+3.673819061467132e-16i)
+ *    Complex.polar(3, -Math::PI/2)  #=> (1.836909530733566e-16-3.0i)
+ */
+static VALUE
+nucomp_s_polar(int argc, VALUE *argv, VALUE klass)
+{
+    VALUE abs, arg;
+
+    switch (rb_scan_args(argc, argv, "11", &abs, &arg)) {
+      case 1:
+	nucomp_real_check(abs);
+	if (canonicalization) return abs;
+	return nucomp_s_new_internal(klass, abs, ZERO);
+      default:
+	nucomp_real_check(abs);
+	nucomp_real_check(arg);
+	break;
+    }
+    return f_complex_polar(klass, abs, arg);
+}
+
+/*
+ * call-seq:
+ *    cmp.real  ->  real
+ *
+ * Returns the real part.
+ *
+ *    Complex(7).real      #=> 7
+ *    Complex(9, -4).real  #=> 9
+ */
+VALUE
+rb_complex_real(VALUE self)
+{
+    get_dat1(self);
+    return dat->real;
+}
+
+/*
+ * call-seq:
+ *    cmp.imag       ->  real
+ *    cmp.imaginary  ->  real
+ *
+ * Returns the imaginary part.
+ *
+ *    Complex(7).imaginary      #=> 0
+ *    Complex(9, -4).imaginary  #=> -4
+ */
+VALUE
+rb_complex_imag(VALUE self)
+{
+    get_dat1(self);
+    return dat->imag;
+}
+
+/*
+ * call-seq:
+ *    -cmp  ->  complex
+ *
+ * Returns negation of the value.
+ *
+ *    -Complex(1, 2)  #=> (-1-2i)
+ */
+VALUE
+rb_complex_uminus(VALUE self)
+{
+    get_dat1(self);
+    return f_complex_new2(CLASS_OF(self),
+			  f_negate(dat->real), f_negate(dat->imag));
+}
+
+/*
+ * call-seq:
+ *    cmp + numeric  ->  complex
+ *
+ * Performs addition.
+ *
+ *    Complex(2, 3)  + Complex(2, 3)   #=> (4+6i)
+ *    Complex(900)   + Complex(1)      #=> (901+0i)
+ *    Complex(-2, 9) + Complex(-9, 2)  #=> (-11+11i)
+ *    Complex(9, 8)  + 4               #=> (13+8i)
+ *    Complex(20, 9) + 9.8             #=> (29.8+9i)
+ */
+VALUE
+rb_complex_plus(VALUE self, VALUE other)
+{
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+	VALUE real, imag;
+
+	get_dat2(self, other);
+
+	real = f_add(adat->real, bdat->real);
+	imag = f_add(adat->imag, bdat->imag);
+
+	return f_complex_new2(CLASS_OF(self), real, imag);
+    }
+    if (k_numeric_p(other) && f_real_p(other)) {
+	get_dat1(self);
+
+	return f_complex_new2(CLASS_OF(self),
+			      f_add(dat->real, other), dat->imag);
+    }
+    return rb_num_coerce_bin(self, other, '+');
+}
+
+/*
+ * call-seq:
+ *    cmp - numeric  ->  complex
+ *
+ * Performs subtraction.
+ *
+ *    Complex(2, 3)  - Complex(2, 3)   #=> (0+0i)
+ *    Complex(900)   - Complex(1)      #=> (899+0i)
+ *    Complex(-2, 9) - Complex(-9, 2)  #=> (7+7i)
+ *    Complex(9, 8)  - 4               #=> (5+8i)
+ *    Complex(20, 9) - 9.8             #=> (10.2+9i)
+ */
+VALUE
+rb_complex_minus(VALUE self, VALUE other)
+{
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+	VALUE real, imag;
+
+	get_dat2(self, other);
+
+	real = f_sub(adat->real, bdat->real);
+	imag = f_sub(adat->imag, bdat->imag);
+
+	return f_complex_new2(CLASS_OF(self), real, imag);
+    }
+    if (k_numeric_p(other) && f_real_p(other)) {
+	get_dat1(self);
+
+	return f_complex_new2(CLASS_OF(self),
+			      f_sub(dat->real, other), dat->imag);
+    }
+    return rb_num_coerce_bin(self, other, '-');
+}
+
+static VALUE
+safe_mul(VALUE a, VALUE b, int az, int bz)
+{
+    double v;
+    if (!az && bz && RB_FLOAT_TYPE_P(a) && (v = RFLOAT_VALUE(a), !isnan(v))) {
+	a = signbit(v) ? DBL2NUM(-1.0) : DBL2NUM(1.0);
+    }
+    if (!bz && az && RB_FLOAT_TYPE_P(b) && (v = RFLOAT_VALUE(b), !isnan(v))) {
+	b = signbit(v) ? DBL2NUM(-1.0) : DBL2NUM(1.0);
+    }
+    return f_mul(a, b);
+}
+
+static void
+comp_mul(VALUE areal, VALUE aimag, VALUE breal, VALUE bimag, VALUE *real, VALUE *imag)
+{
+    int arzero = f_zero_p(areal);
+    int aizero = f_zero_p(aimag);
+    int brzero = f_zero_p(breal);
+    int bizero = f_zero_p(bimag);
+    *real = f_sub(safe_mul(areal, breal, arzero, brzero),
+                  safe_mul(aimag, bimag, aizero, bizero));
+    *imag = f_add(safe_mul(areal, bimag, arzero, bizero),
+                  safe_mul(aimag, breal, aizero, brzero));
+}
+
+/*
+ * call-seq:
+ *    cmp * numeric  ->  complex
+ *
+ * Performs multiplication.
+ *
+ *    Complex(2, 3)  * Complex(2, 3)   #=> (-5+12i)
+ *    Complex(900)   * Complex(1)      #=> (900+0i)
+ *    Complex(-2, 9) * Complex(-9, 2)  #=> (0-85i)
+ *    Complex(9, 8)  * 4               #=> (36+32i)
+ *    Complex(20, 9) * 9.8             #=> (196.0+88.2i)
+ */
+VALUE
+rb_complex_mul(VALUE self, VALUE other)
+{
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+	VALUE real, imag;
+	get_dat2(self, other);
+
+        comp_mul(adat->real, adat->imag, bdat->real, bdat->imag, &real, &imag);
+
+	return f_complex_new2(CLASS_OF(self), real, imag);
+    }
+    if (k_numeric_p(other) && f_real_p(other)) {
+	get_dat1(self);
+
+	return f_complex_new2(CLASS_OF(self),
+			      f_mul(dat->real, other),
+			      f_mul(dat->imag, other));
+    }
+    return rb_num_coerce_bin(self, other, '*');
+}
+
+inline static VALUE
+f_divide(VALUE self, VALUE other,
+	 VALUE (*func)(VALUE, VALUE), ID id)
+{
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+        VALUE r, n, x, y;
+	int flo;
+	get_dat2(self, other);
+
+	flo = (RB_FLOAT_TYPE_P(adat->real) || RB_FLOAT_TYPE_P(adat->imag) ||
+	       RB_FLOAT_TYPE_P(bdat->real) || RB_FLOAT_TYPE_P(bdat->imag));
+
+	if (f_gt_p(f_abs(bdat->real), f_abs(bdat->imag))) {
+	    r = (*func)(bdat->imag, bdat->real);
+	    n = f_mul(bdat->real, f_add(ONE, f_mul(r, r)));
+	    if (flo)
+		return f_complex_new2(CLASS_OF(self),
+				      (*func)(self, n),
+				      (*func)(f_negate(f_mul(self, r)), n));
+            x = (*func)(f_add(adat->real, f_mul(adat->imag, r)), n);
+            y = (*func)(f_sub(adat->imag, f_mul(adat->real, r)), n);
+	}
+	else {
+	    r = (*func)(bdat->real, bdat->imag);
+	    n = f_mul(bdat->imag, f_add(ONE, f_mul(r, r)));
+	    if (flo)
+		return f_complex_new2(CLASS_OF(self),
+				      (*func)(f_mul(self, r), n),
+				      (*func)(f_negate(self), n));
+            x = (*func)(f_add(f_mul(adat->real, r), adat->imag), n);
+            y = (*func)(f_sub(f_mul(adat->imag, r), adat->real), n);
+	}
+        x = rb_rational_canonicalize(x);
+        y = rb_rational_canonicalize(y);
+        return f_complex_new2(CLASS_OF(self), x, y);
+    }
+    if (k_numeric_p(other) && f_real_p(other)) {
+	get_dat1(self);
+
+	return f_complex_new2(CLASS_OF(self),
+			      (*func)(dat->real, other),
+			      (*func)(dat->imag, other));
+    }
+    return rb_num_coerce_bin(self, other, id);
+}
+
+#define rb_raise_zerodiv() rb_raise(rb_eZeroDivError, "divided by 0")
+
+/*
+ * call-seq:
+ *    cmp / numeric     ->  complex
+ *    cmp.quo(numeric)  ->  complex
+ *
+ * Performs division.
+ *
+ *    Complex(2, 3)  / Complex(2, 3)   #=> ((1/1)+(0/1)*i)
+ *    Complex(900)   / Complex(1)      #=> ((900/1)+(0/1)*i)
+ *    Complex(-2, 9) / Complex(-9, 2)  #=> ((36/85)-(77/85)*i)
+ *    Complex(9, 8)  / 4               #=> ((9/4)+(2/1)*i)
+ *    Complex(20, 9) / 9.8             #=> (2.0408163265306123+0.9183673469387754i)
+ */
+VALUE
+rb_complex_div(VALUE self, VALUE other)
+{
+    return f_divide(self, other, f_quo, id_quo);
+}
+
+#define nucomp_quo rb_complex_div
+
+/*
+ * call-seq:
+ *    cmp.fdiv(numeric)  ->  complex
+ *
+ * Performs division as each part is a float, never returns a float.
+ *
+ *    Complex(11, 22).fdiv(3)  #=> (3.6666666666666665+7.333333333333333i)
+ */
+static VALUE
+nucomp_fdiv(VALUE self, VALUE other)
+{
+    return f_divide(self, other, f_fdiv, id_fdiv);
+}
+
+inline static VALUE
+f_reciprocal(VALUE x)
+{
+    return f_quo(ONE, x);
+}
+
+/*
+ * call-seq:
+ *    cmp ** numeric  ->  complex
+ *
+ * Performs exponentiation.
+ *
+ *    Complex('i') ** 2              #=> (-1+0i)
+ *    Complex(-8) ** Rational(1, 3)  #=> (1.0000000000000002+1.7320508075688772i)
+ */
+VALUE
+rb_complex_pow(VALUE self, VALUE other)
+{
+    if (k_numeric_p(other) && k_exact_zero_p(other))
+	return f_complex_new_bang1(CLASS_OF(self), ONE);
+
+    if (RB_TYPE_P(other, T_RATIONAL) && RRATIONAL(other)->den == LONG2FIX(1))
+	other = RRATIONAL(other)->num; /* c14n */
+
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+	get_dat1(other);
+
+	if (k_exact_zero_p(dat->imag))
+	    other = dat->real; /* c14n */
+    }
+
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+	VALUE r, theta, nr, ntheta;
+
+	get_dat1(other);
+
+	r = f_abs(self);
+	theta = f_arg(self);
+
+	nr = m_exp_bang(f_sub(f_mul(dat->real, m_log_bang(r)),
+			      f_mul(dat->imag, theta)));
+	ntheta = f_add(f_mul(theta, dat->real),
+		       f_mul(dat->imag, m_log_bang(r)));
+	return f_complex_polar(CLASS_OF(self), nr, ntheta);
+    }
+    if (FIXNUM_P(other)) {
+        long n = FIX2LONG(other);
+        if (n == 0) {
+            return nucomp_s_new_internal(CLASS_OF(self), ONE, ZERO);
+        }
+        if (n < 0) {
+            self = f_reciprocal(self);
+            other = rb_int_uminus(other);
+            n = -n;
+        }
+        {
+            get_dat1(self);
+            VALUE xr = dat->real, xi = dat->imag, zr = xr, zi = xi;
+
+            if (f_zero_p(xi)) {
+                zr = rb_num_pow(zr, other);
+            }
+            else if (f_zero_p(xr)) {
+                zi = rb_num_pow(zi, other);
+                if (n & 2) zi = f_negate(zi);
+                if (!(n & 1)) {
+                    VALUE tmp = zr;
+                    zr = zi;
+                    zi = tmp;
+                }
+            }
+            else {
+                while (--n) {
+                    long q, r;
+
+                    for (; q = n / 2, r = n % 2, r == 0; n = q) {
+                        VALUE tmp = f_sub(f_mul(xr, xr), f_mul(xi, xi));
+                        xi = f_mul(f_mul(TWO, xr), xi);
+                        xr = tmp;
+                    }
+                    comp_mul(zr, zi, xr, xi, &zr, &zi);
+                }
+            }
+            return nucomp_s_new_internal(CLASS_OF(self), zr, zi);
+	}
+    }
+    if (k_numeric_p(other) && f_real_p(other)) {
+	VALUE r, theta;
+
+	if (RB_TYPE_P(other, T_BIGNUM))
+	    rb_warn("in a**b, b may be too big");
+
+	r = f_abs(self);
+	theta = f_arg(self);
+
+	return f_complex_polar(CLASS_OF(self), f_expt(r, other),
+			       f_mul(theta, other));
+    }
+    return rb_num_coerce_bin(self, other, id_expt);
+}
+
+/*
+ * call-seq:
+ *    cmp == object  ->  true or false
+ *
+ * Returns true if cmp equals object numerically.
+ *
+ *    Complex(2, 3)  == Complex(2, 3)   #=> true
+ *    Complex(5)     == 5               #=> true
+ *    Complex(0)     == 0.0             #=> true
+ *    Complex('1/3') == 0.33            #=> false
+ *    Complex('1/2') == '1/2'           #=> false
+ */
+static VALUE
+nucomp_eqeq_p(VALUE self, VALUE other)
+{
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+	get_dat2(self, other);
+
+	return f_boolcast(f_eqeq_p(adat->real, bdat->real) &&
+			  f_eqeq_p(adat->imag, bdat->imag));
+    }
+    if (k_numeric_p(other) && f_real_p(other)) {
+	get_dat1(self);
+
+	return f_boolcast(f_eqeq_p(dat->real, other) && f_zero_p(dat->imag));
+    }
+    return f_boolcast(f_eqeq_p(other, self));
+}
+
+/* :nodoc: */
+static VALUE
+nucomp_coerce(VALUE self, VALUE other)
+{
+    if (k_numeric_p(other) && f_real_p(other))
+	return rb_assoc_new(f_complex_new_bang1(CLASS_OF(self), other), self);
+    if (RB_TYPE_P(other, T_COMPLEX))
+	return rb_assoc_new(other, self);
+
+    rb_raise(rb_eTypeError, "%"PRIsVALUE" can't be coerced into %"PRIsVALUE,
+	     rb_obj_class(other), rb_obj_class(self));
+    return Qnil;
+}
+
+/*
+ * call-seq:
+ *    cmp.abs        ->  real
+ *    cmp.magnitude  ->  real
+ *
+ * Returns the absolute part of its polar form.
+ *
+ *    Complex(-1).abs         #=> 1
+ *    Complex(3.0, -4.0).abs  #=> 5.0
+ */
+VALUE
+rb_complex_abs(VALUE self)
+{
+    get_dat1(self);
+
+    if (f_zero_p(dat->real)) {
+	VALUE a = f_abs(dat->imag);
+	if (RB_FLOAT_TYPE_P(dat->real) && !RB_FLOAT_TYPE_P(dat->imag))
+	    a = f_to_f(a);
+	return a;
+    }
+    if (f_zero_p(dat->imag)) {
+	VALUE a = f_abs(dat->real);
+	if (!RB_FLOAT_TYPE_P(dat->real) && RB_FLOAT_TYPE_P(dat->imag))
+	    a = f_to_f(a);
+	return a;
+    }
+    return rb_math_hypot(dat->real, dat->imag);
+}
+
+/*
+ * call-seq:
+ *    cmp.abs2  ->  real
+ *
+ * Returns square of the absolute value.
+ *
+ *    Complex(-1).abs2         #=> 1
+ *    Complex(3.0, -4.0).abs2  #=> 25.0
+ */
+static VALUE
+nucomp_abs2(VALUE self)
+{
+    get_dat1(self);
+    return f_add(f_mul(dat->real, dat->real),
+		 f_mul(dat->imag, dat->imag));
+}
+
+/*
+ * call-seq:
+ *    cmp.arg    ->  float
+ *    cmp.angle  ->  float
+ *    cmp.phase  ->  float
+ *
+ * Returns the angle part of its polar form.
+ *
+ *    Complex.polar(3, Math::PI/2).arg  #=> 1.5707963267948966
+ */
+VALUE
+rb_complex_arg(VALUE self)
+{
+    get_dat1(self);
+    return rb_math_atan2(dat->imag, dat->real);
+}
+
+/*
+ * call-seq:
+ *    cmp.rect         ->  array
+ *    cmp.rectangular  ->  array
+ *
+ * Returns an array; [cmp.real, cmp.imag].
+ *
+ *    Complex(1, 2).rectangular  #=> [1, 2]
+ */
+static VALUE
+nucomp_rect(VALUE self)
+{
+    get_dat1(self);
+    return rb_assoc_new(dat->real, dat->imag);
+}
+
+/*
+ * call-seq:
+ *    cmp.polar  ->  array
+ *
+ * Returns an array; [cmp.abs, cmp.arg].
+ *
+ *    Complex(1, 2).polar  #=> [2.23606797749979, 1.1071487177940904]
+ */
+static VALUE
+nucomp_polar(VALUE self)
+{
+    return rb_assoc_new(f_abs(self), f_arg(self));
+}
+
+/*
+ * call-seq:
+ *    cmp.conj       ->  complex
+ *    cmp.conjugate  ->  complex
+ *
+ * Returns the complex conjugate.
+ *
+ *    Complex(1, 2).conjugate  #=> (1-2i)
+ */
+VALUE
+rb_complex_conjugate(VALUE self)
+{
+    get_dat1(self);
+    return f_complex_new2(CLASS_OF(self), dat->real, f_negate(dat->imag));
+}
+
+/*
+ * call-seq:
+ *    cmp.real?  ->  false
+ *
+ * Returns false.
+ */
+static VALUE
+nucomp_false(VALUE self)
+{
+    return Qfalse;
+}
+
+/*
+ * call-seq:
+ *    cmp.denominator  ->  integer
+ *
+ * Returns the denominator (lcm of both denominator - real and imag).
+ *
+ * See numerator.
+ */
+static VALUE
+nucomp_denominator(VALUE self)
+{
+    get_dat1(self);
+    return rb_lcm(f_denominator(dat->real), f_denominator(dat->imag));
+}
+
+/*
+ * call-seq:
+ *    cmp.numerator  ->  numeric
+ *
+ * Returns the numerator.
+ *
+ *        1   2       3+4i  <-  numerator
+ *        - + -i  ->  ----
+ *        2   3        6    <-  denominator
+ *
+ *    c = Complex('1/2+2/3i')  #=> ((1/2)+(2/3)*i)
+ *    n = c.numerator          #=> (3+4i)
+ *    d = c.denominator        #=> 6
+ *    n / d                    #=> ((1/2)+(2/3)*i)
+ *    Complex(Rational(n.real, d), Rational(n.imag, d))
+ *                             #=> ((1/2)+(2/3)*i)
+ * See denominator.
+ */
+static VALUE
+nucomp_numerator(VALUE self)
+{
+    VALUE cd;
+
+    get_dat1(self);
+
+    cd = f_denominator(self);
+    return f_complex_new2(CLASS_OF(self),
+			  f_mul(f_numerator(dat->real),
+				f_div(cd, f_denominator(dat->real))),
+			  f_mul(f_numerator(dat->imag),
+				f_div(cd, f_denominator(dat->imag))));
+}
+
+/* :nodoc: */
+static VALUE
+nucomp_hash(VALUE self)
+{
+    st_index_t v, h[2];
+    VALUE n;
+
+    get_dat1(self);
+    n = rb_hash(dat->real);
+    h[0] = NUM2LONG(n);
+    n = rb_hash(dat->imag);
+    h[1] = NUM2LONG(n);
+    v = rb_memhash(h, sizeof(h));
+    return ST2FIX(v);
+}
+
+/* :nodoc: */
+static VALUE
+nucomp_eql_p(VALUE self, VALUE other)
+{
+    if (RB_TYPE_P(other, T_COMPLEX)) {
+	get_dat2(self, other);
+
+	return f_boolcast((CLASS_OF(adat->real) == CLASS_OF(bdat->real)) &&
+			  (CLASS_OF(adat->imag) == CLASS_OF(bdat->imag)) &&
+			  f_eqeq_p(self, other));
+
+    }
+    return Qfalse;
+}
+
+inline static int
+f_signbit(VALUE x)
+{
+    if (RB_FLOAT_TYPE_P(x)) {
+	double f = RFLOAT_VALUE(x);
+	return !isnan(f) && signbit(f);
+    }
+    return f_negative_p(x);
+}
+
+inline static int
+f_tpositive_p(VALUE x)
+{
+    return !f_signbit(x);
+}
+
+static VALUE
+f_format(VALUE self, VALUE (*func)(VALUE))
+{
+    VALUE s;
+    int impos;
+
+    get_dat1(self);
+
+    impos = f_tpositive_p(dat->imag);
+
+    s = (*func)(dat->real);
+    rb_str_cat2(s, !impos ? "-" : "+");
+
+    rb_str_concat(s, (*func)(f_abs(dat->imag)));
+    if (!rb_isdigit(RSTRING_PTR(s)[RSTRING_LEN(s) - 1]))
+	rb_str_cat2(s, "*");
+    rb_str_cat2(s, "i");
+
+    return s;
+}
+
+/*
+ * call-seq:
+ *    cmp.to_s  ->  string
+ *
+ * Returns the value as a string.
+ *
+ *    Complex(2).to_s                       #=> "2+0i"
+ *    Complex('-8/6').to_s                  #=> "-4/3+0i"
+ *    Complex('1/2i').to_s                  #=> "0+1/2i"
+ *    Complex(0, Float::INFINITY).to_s      #=> "0+Infinity*i"
+ *    Complex(Float::NAN, Float::NAN).to_s  #=> "NaN+NaN*i"
+ */
+static VALUE
+nucomp_to_s(VALUE self)
+{
+    return f_format(self, rb_String);
+}
+
+/*
+ * call-seq:
+ *    cmp.inspect  ->  string
+ *
+ * Returns the value as a string for inspection.
+ *
+ *    Complex(2).inspect                       #=> "(2+0i)"
+ *    Complex('-8/6').inspect                  #=> "((-4/3)+0i)"
+ *    Complex('1/2i').inspect                  #=> "(0+(1/2)*i)"
+ *    Complex(0, Float::INFINITY).inspect      #=> "(0+Infinity*i)"
+ *    Complex(Float::NAN, Float::NAN).inspect  #=> "(NaN+NaN*i)"
+ */
+static VALUE
+nucomp_inspect(VALUE self)
+{
+    VALUE s;
+
+    s = rb_usascii_str_new2("(");
+    rb_str_concat(s, f_format(self, rb_inspect));
+    rb_str_cat2(s, ")");
+
+    return s;
+}
+
+#define FINITE_TYPE_P(v) (RB_INTEGER_TYPE_P(v) || RB_TYPE_P(v, T_RATIONAL))
+
+/*
+ * call-seq:
+ *    cmp.finite?  ->  true or false
+ *
+ * Returns +true+ if +cmp+'s real and imaginary parts are both finite numbers,
+ * otherwise returns +false+.
+ */
+static VALUE
+rb_complex_finite_p(VALUE self)
+{
+    get_dat1(self);
+
+    if (f_finite_p(dat->real) && f_finite_p(dat->imag)) {
+	return Qtrue;
+    }
+    return Qfalse;
+}
+
+/*
+ * call-seq:
+ *    cmp.infinite?  ->  nil or 1
+ *
+ * Returns +1+ if +cmp+'s real or imaginary part is an infinite number,
+ * otherwise returns +nil+.
+ *
+ *  For example:
+ *
+ *     (1+1i).infinite?                   #=> nil
+ *     (Float::INFINITY + 1i).infinite?   #=> 1
+ */
+static VALUE
+rb_complex_infinite_p(VALUE self)
+{
+    get_dat1(self);
+
+    if (NIL_P(f_infinite_p(dat->real)) && NIL_P(f_infinite_p(dat->imag))) {
+	return Qnil;
+    }
+    return ONE;
+}
+
+/* :nodoc: */
+static VALUE
+nucomp_dumper(VALUE self)
+{
+    return self;
+}
+
+/* :nodoc: */
+static VALUE
+nucomp_loader(VALUE self, VALUE a)
+{
+    get_dat1(self);
+
+    RCOMPLEX_SET_REAL(dat, rb_ivar_get(a, id_i_real));
+    RCOMPLEX_SET_IMAG(dat, rb_ivar_get(a, id_i_imag));
+    OBJ_FREEZE_RAW(self);
+
+    return self;
+}
+
+/* :nodoc: */
+static VALUE
+nucomp_marshal_dump(VALUE self)
+{
+    VALUE a;
+    get_dat1(self);
+
+    a = rb_assoc_new(dat->real, dat->imag);
+    rb_copy_generic_ivar(a, self);
+    return a;
+}
+
+/* :nodoc: */
+static VALUE
+nucomp_marshal_load(VALUE self, VALUE a)
+{
+    Check_Type(a, T_ARRAY);
+    if (RARRAY_LEN(a) != 2)
+	rb_raise(rb_eArgError, "marshaled complex must have an array whose length is 2 but %ld", RARRAY_LEN(a));
+    rb_ivar_set(self, id_i_real, RARRAY_AREF(a, 0));
+    rb_ivar_set(self, id_i_imag, RARRAY_AREF(a, 1));
+    return self;
+}
+
+/* --- */
+
+VALUE
+rb_complex_raw(VALUE x, VALUE y)
+{
+    return nucomp_s_new_internal(rb_cComplex, x, y);
+}
+
+VALUE
+rb_complex_new(VALUE x, VALUE y)
+{
+    return nucomp_s_canonicalize_internal(rb_cComplex, x, y);
+}
+
+VALUE
+rb_complex_new_polar(VALUE x, VALUE y)
+{
+    return f_complex_polar(rb_cComplex, x, y);
+}
+
+VALUE
+rb_complex_polar(VALUE x, VALUE y)
+{
+    return rb_complex_new_polar(x, y);
+}
+
+VALUE
+rb_Complex(VALUE x, VALUE y)
+{
+    VALUE a[2];
+    a[0] = x;
+    a[1] = y;
+    return nucomp_s_convert(2, a, rb_cComplex);
+}
+
+/*!
+ * Creates a Complex object.
+ *
+ * \param real    real part value
+ * \param imag    imaginary part value
+ * \return        a new Complex object
+ */
+VALUE
+rb_dbl_complex_new(double real, double imag)
+{
+    return rb_complex_raw(DBL2NUM(real), DBL2NUM(imag));
+}
+
+/*
+ * call-seq:
+ *    cmp.to_i  ->  integer
+ *
+ * Returns the value as an integer if possible (the imaginary part
+ * should be exactly zero).
+ *
+ *    Complex(1, 0).to_i    #=> 1
+ *    Complex(1, 0.0).to_i  # RangeError
+ *    Complex(1, 2).to_i    # RangeError
+ */
+static VALUE
+nucomp_to_i(VALUE self)
+{
+    get_dat1(self);
+
+    if (!k_exact_zero_p(dat->imag)) {
+	rb_raise(rb_eRangeError, "can't convert %"PRIsVALUE" into Integer",
+		 self);
+    }
+    return f_to_i(dat->real);
+}
+
+/*
+ * call-seq:
+ *    cmp.to_f  ->  float
+ *
+ * Returns the value as a float if possible (the imaginary part should
+ * be exactly zero).
+ *
+ *    Complex(1, 0).to_f    #=> 1.0
+ *    Complex(1, 0.0).to_f  # RangeError
+ *    Complex(1, 2).to_f    # RangeError
+ */
+static VALUE
+nucomp_to_f(VALUE self)
+{
+    get_dat1(self);
+
+    if (!k_exact_zero_p(dat->imag)) {
+	rb_raise(rb_eRangeError, "can't convert %"PRIsVALUE" into Float",
+		 self);
+    }
+    return f_to_f(dat->real);
+}
+
+/*
+ * call-seq:
+ *    cmp.to_r  ->  rational
+ *
+ * Returns the value as a rational if possible (the imaginary part
+ * should be exactly zero).
+ *
+ *    Complex(1, 0).to_r    #=> (1/1)
+ *    Complex(1, 0.0).to_r  # RangeError
+ *    Complex(1, 2).to_r    # RangeError
+ *
+ * See rationalize.
+ */
+static VALUE
+nucomp_to_r(VALUE self)
+{
+    get_dat1(self);
+
+    if (!k_exact_zero_p(dat->imag)) {
+	rb_raise(rb_eRangeError, "can't convert %"PRIsVALUE" into Rational",
+		 self);
+    }
+    return f_to_r(dat->real);
+}
+
+/*
+ * call-seq:
+ *    cmp.rationalize([eps])  ->  rational
+ *
+ * Returns the value as a rational if possible (the imaginary part
+ * should be exactly zero).
+ *
+ *    Complex(1.0/3, 0).rationalize  #=> (1/3)
+ *    Complex(1, 0.0).rationalize    # RangeError
+ *    Complex(1, 2).rationalize      # RangeError
+ *
+ * See to_r.
+ */
+static VALUE
+nucomp_rationalize(int argc, VALUE *argv, VALUE self)
+{
+    get_dat1(self);
+
+    rb_check_arity(argc, 0, 1);
+
+    if (!k_exact_zero_p(dat->imag)) {
+       rb_raise(rb_eRangeError, "can't convert %"PRIsVALUE" into Rational",
+                self);
+    }
+    return rb_funcallv(dat->real, id_rationalize, argc, argv);
+}
+
+/*
+ * call-seq:
+ *    complex.to_c  ->  self
+ *
+ * Returns self.
+ *
+ *    Complex(2).to_c      #=> (2+0i)
+ *    Complex(-8, 6).to_c  #=> (-8+6i)
+ */
+static VALUE
+nucomp_to_c(VALUE self)
+{
+    return self;
+}
+
+/*
+ * call-seq:
+ *    nil.to_c  ->  (0+0i)
+ *
+ * Returns zero as a complex.
+ */
+static VALUE
+nilclass_to_c(VALUE self)
+{
+    return rb_complex_new1(INT2FIX(0));
+}
+
+/*
+ * call-seq:
+ *    num.to_c  ->  complex
+ *
+ * Returns the value as a complex.
+ */
+static VALUE
+numeric_to_c(VALUE self)
+{
+    return rb_complex_new1(self);
+}
+
+#include <ctype.h>
+
+inline static int
+issign(int c)
+{
+    return (c == '-' || c == '+');
+}
+
+static int
+read_sign(const char **s,
+	  char **b)
+{
+    int sign = '?';
+
+    if (issign(**s)) {
+	sign = **b = **s;
+	(*s)++;
+	(*b)++;
+    }
+    return sign;
+}
+
+inline static int
+isdecimal(int c)
+{
+    return isdigit((unsigned char)c);
+}
+
+static int
+read_digits(const char **s, int strict,
+	    char **b)
+{
+    int us = 1;
+
+    if (!isdecimal(**s))
+	return 0;
+
+    while (isdecimal(**s) || **s == '_') {
+	if (**s == '_') {
+	    if (strict) {
+		if (us)
+		    return 0;
+	    }
+	    us = 1;
+	}
+	else {
+	    **b = **s;
+	    (*b)++;
+	    us = 0;
+	}
+	(*s)++;
+    }
+    if (us)
+	do {
+	    (*s)--;
+	} while (**s == '_');
+    return 1;
+}
+
+inline static int
+islettere(int c)
+{
+    return (c == 'e' || c == 'E');
+}
+
+static int
+read_num(const char **s, int strict,
+	 char **b)
+{
+    if (**s != '.') {
+	if (!read_digits(s, strict, b))
+	    return 0;
+    }
+
+    if (**s == '.') {
+	**b = **s;
+	(*s)++;
+	(*b)++;
+	if (!read_digits(s, strict, b)) {
+	    (*b)--;
+	    return 0;
+	}
+    }
+
+    if (islettere(**s)) {
+	**b = **s;
+	(*s)++;
+	(*b)++;
+	read_sign(s, b);
+	if (!read_digits(s, strict, b)) {
+	    (*b)--;
+	    return 0;
+	}
+    }
+    return 1;
+}
+
+inline static int
+read_den(const char **s, int strict,
+	 char **b)
+{
+    if (!read_digits(s, strict, b))
+	return 0;
+    return 1;
+}
+
+static int
+read_rat_nos(const char **s, int strict,
+	     char **b)
+{
+    if (!read_num(s, strict, b))
+	return 0;
+    if (**s == '/') {
+	**b = **s;
+	(*s)++;
+	(*b)++;
+	if (!read_den(s, strict, b)) {
+	    (*b)--;
+	    return 0;
+	}
+    }
+    return 1;
+}
+
+static int
+read_rat(const char **s, int strict,
+	 char **b)
+{
+    read_sign(s, b);
+    if (!read_rat_nos(s, strict, b))
+	return 0;
+    return 1;
+}
+
+inline static int
+isimagunit(int c)
+{
+    return (c == 'i' || c == 'I' ||
+	    c == 'j' || c == 'J');
+}
+
+static VALUE
+str2num(char *s)
+{
+    if (strchr(s, '/'))
+	return rb_cstr_to_rat(s, 0);
+    if (strpbrk(s, ".eE"))
+	return DBL2NUM(rb_cstr_to_dbl(s, 0));
+    return rb_cstr_to_inum(s, 10, 0);
+}
+
+static int
+read_comp(const char **s, int strict,
+	  VALUE *ret, char **b)
+{
+    char *bb;
+    int sign;
+    VALUE num, num2;
+
+    bb = *b;
+
+    sign = read_sign(s, b);
+
+    if (isimagunit(**s)) {
+	(*s)++;
+	num = INT2FIX((sign == '-') ? -1 : + 1);
+	*ret = rb_complex_new2(ZERO, num);
+	return 1; /* e.g. "i" */
+    }
+
+    if (!read_rat_nos(s, strict, b)) {
+	**b = '\0';
+	num = str2num(bb);
+	*ret = rb_complex_new2(num, ZERO);
+	return 0; /* e.g. "-" */
+    }
+    **b = '\0';
+    num = str2num(bb);
+
+    if (isimagunit(**s)) {
+	(*s)++;
+	*ret = rb_complex_new2(ZERO, num);
+	return 1; /* e.g. "3i" */
+    }
+
+    if (**s == '@') {
+	int st;
+
+	(*s)++;
+	bb = *b;
+	st = read_rat(s, strict, b);
+	**b = '\0';
+	if (strlen(bb) < 1 ||
+	    !isdecimal(*(bb + strlen(bb) - 1))) {
+	    *ret = rb_complex_new2(num, ZERO);
+	    return 0; /* e.g. "1@-" */
+	}
+	num2 = str2num(bb);
+	*ret = rb_complex_new_polar(num, num2);
+	if (!st)
+	    return 0; /* e.g. "1@2." */
+	else
+	    return 1; /* e.g. "1@2" */
+    }
+
+    if (issign(**s)) {
+	bb = *b;
+	sign = read_sign(s, b);
+	if (isimagunit(**s))
+	    num2 = INT2FIX((sign == '-') ? -1 : + 1);
+	else {
+	    if (!read_rat_nos(s, strict, b)) {
+		*ret = rb_complex_new2(num, ZERO);
+		return 0; /* e.g. "1+xi" */
+	    }
+	    **b = '\0';
+	    num2 = str2num(bb);
+	}
+	if (!isimagunit(**s)) {
+	    *ret = rb_complex_new2(num, ZERO);
+	    return 0; /* e.g. "1+3x" */
+	}
+	(*s)++;
+	*ret = rb_complex_new2(num, num2);
+	return 1; /* e.g. "1+2i" */
+    }
+    /* !(@, - or +) */
+    {
+	*ret = rb_complex_new2(num, ZERO);
+	return 1; /* e.g. "3" */
+    }
+}
+
+inline static void
+skip_ws(const char **s)
+{
+    while (isspace((unsigned char)**s))
+	(*s)++;
+}
+
+static int
+parse_comp(const char *s, int strict, VALUE *num)
+{
+    char *buf, *b;
+    VALUE tmp;
+    int ret = 1;
+
+    buf = ALLOCV_N(char, tmp, strlen(s) + 1);
+    b = buf;
+
+    skip_ws(&s);
+    if (!read_comp(&s, strict, num, &b)) {
+        ret = 0;
+    }
+    else {
+        skip_ws(&s);
+
+        if (strict)
+            if (*s != '\0')
+                ret = 0;
+    }
+    ALLOCV_END(tmp);
+
+    return ret;
+}
+
+static VALUE
+string_to_c_strict(VALUE self, int raise)
+{
+    char *s;
+    VALUE num;
+
+    rb_must_asciicompat(self);
+
+    s = RSTRING_PTR(self);
+
+    if (!s || memchr(s, '\0', RSTRING_LEN(self))) {
+        if (!raise) return Qnil;
+	rb_raise(rb_eArgError, "string contains null byte");
+    }
+
+    if (s && s[RSTRING_LEN(self)]) {
+	rb_str_modify(self);
+	s = RSTRING_PTR(self);
+	s[RSTRING_LEN(self)] = '\0';
+    }
+
+    if (!s)
+	s = (char *)"";
+
+    if (!parse_comp(s, 1, &num)) {
+        if (!raise) return Qnil;
+	rb_raise(rb_eArgError, "invalid value for convert(): %+"PRIsVALUE,
+		 self);
+    }
+
+    return num;
+}
+
+/*
+ * call-seq:
+ *    str.to_c  ->  complex
+ *
+ * Returns a complex which denotes the string form.  The parser
+ * ignores leading whitespaces and trailing garbage.  Any digit
+ * sequences can be separated by an underscore.  Returns zero for null
+ * or garbage string.
+ *
+ *    '9'.to_c           #=> (9+0i)
+ *    '2.5'.to_c         #=> (2.5+0i)
+ *    '2.5/1'.to_c       #=> ((5/2)+0i)
+ *    '-3/2'.to_c        #=> ((-3/2)+0i)
+ *    '-i'.to_c          #=> (0-1i)
+ *    '45i'.to_c         #=> (0+45i)
+ *    '3-4i'.to_c        #=> (3-4i)
+ *    '-4e2-4e-2i'.to_c  #=> (-400.0-0.04i)
+ *    '-0.0-0.0i'.to_c   #=> (-0.0-0.0i)
+ *    '1/2+3/4i'.to_c    #=> ((1/2)+(3/4)*i)
+ *    'ruby'.to_c        #=> (0+0i)
+ *
+ * See Kernel.Complex.
+ */
+static VALUE
+string_to_c(VALUE self)
+{
+    char *s;
+    VALUE num;
+
+    rb_must_asciicompat(self);
+
+    s = RSTRING_PTR(self);
+
+    if (s && s[RSTRING_LEN(self)]) {
+	rb_str_modify(self);
+	s = RSTRING_PTR(self);
+	s[RSTRING_LEN(self)] = '\0';
+    }
+
+    if (!s)
+	s = (char *)"";
+
+    (void)parse_comp(s, 0, &num);
+
+    return num;
+}
+
+static VALUE
+to_complex(VALUE val)
+{
+    return rb_convert_type(val, T_COMPLEX, "Complex", "to_c");
+}
+
+static VALUE
+nucomp_convert(VALUE klass, VALUE a1, VALUE a2, int raise)
+{
+    if (NIL_P(a1) || NIL_P(a2))
+	rb_raise(rb_eTypeError, "can't convert nil into Complex");
+
+    if (RB_TYPE_P(a1, T_STRING)) {
+	a1 = string_to_c_strict(a1, raise);
+        if (NIL_P(a1)) return Qnil;
+    }
+
+    if (RB_TYPE_P(a2, T_STRING)) {
+	a2 = string_to_c_strict(a2, raise);
+        if (NIL_P(a2)) return Qnil;
+    }
+
+    if (RB_TYPE_P(a1, T_COMPLEX)) {
+	{
+	    get_dat1(a1);
+
+	    if (k_exact_zero_p(dat->imag))
+		a1 = dat->real;
+	}
+    }
+
+    if (RB_TYPE_P(a2, T_COMPLEX)) {
+	{
+	    get_dat1(a2);
+
+	    if (k_exact_zero_p(dat->imag))
+		a2 = dat->real;
+	}
+    }
+
+    if (RB_TYPE_P(a1, T_COMPLEX)) {
+	if (a2 == Qundef || (k_exact_zero_p(a2)))
+	    return a1;
+    }
+
+    if (a2 == Qundef) {
+	if (k_numeric_p(a1) && !f_real_p(a1))
+	    return a1;
+	/* should raise exception for consistency */
+	if (!k_numeric_p(a1)) {
+            if (!raise)
+                return rb_protect(to_complex, a1, NULL);
+	    return to_complex(a1);
+        }
+    }
+    else {
+	if ((k_numeric_p(a1) && k_numeric_p(a2)) &&
+	    (!f_real_p(a1) || !f_real_p(a2)))
+	    return f_add(a1,
+			 f_mul(a2,
+			       f_complex_new_bang2(rb_cComplex, ZERO, ONE)));
+    }
+
+    {
+        int argc;
+	VALUE argv2[2];
+	argv2[0] = a1;
+        if (a2 == Qundef) {
+            argv2[1] = Qnil;
+            argc = 1;
+        }
+        else {
+            if (!raise && !RB_INTEGER_TYPE_P(a2) && !RB_FLOAT_TYPE_P(a2) && !RB_TYPE_P(a2, T_RATIONAL))
+                return Qnil;
+            argv2[1] = a2;
+            argc = 2;
+        }
+	return nucomp_s_new(argc, argv2, klass);
+    }
+}
+
+static VALUE
+nucomp_s_convert(int argc, VALUE *argv, VALUE klass)
+{
+    VALUE a1, a2;
+
+    if (rb_scan_args(argc, argv, "11", &a1, &a2) == 1) {
+        a2 = Qundef;
+    }
+
+    return nucomp_convert(klass, a1, a2, TRUE);
+}
+
+/* --- */
+
+/*
+ * call-seq:
+ *    num.real  ->  self
+ *
+ * Returns self.
+ */
+static VALUE
+numeric_real(VALUE self)
+{
+    return self;
+}
+
+/*
+ * call-seq:
+ *    num.imag       ->  0
+ *    num.imaginary  ->  0
+ *
+ * Returns zero.
+ */
+static VALUE
+numeric_imag(VALUE self)
+{
+    return INT2FIX(0);
+}
+
+/*
+ * call-seq:
+ *    num.abs2  ->  real
+ *
+ * Returns square of self.
+ */
+static VALUE
+numeric_abs2(VALUE self)
+{
+    return f_mul(self, self);
+}
+
+/*
+ * call-seq:
+ *    num.arg    ->  0 or float
+ *    num.angle  ->  0 or float
+ *    num.phase  ->  0 or float
+ *
+ * Returns 0 if the value is positive, pi otherwise.
+ */
+static VALUE
+numeric_arg(VALUE self)
+{
+    if (f_positive_p(self))
+        return INT2FIX(0);
+    return DBL2NUM(M_PI);
+}
+
+/*
+ * call-seq:
+ *    num.rect  ->  array
+ *    num.rectangular  ->  array
+ *
+ * Returns an array; [num, 0].
+ */
+static VALUE
+numeric_rect(VALUE self)
+{
+    return rb_assoc_new(self, INT2FIX(0));
+}
+
+static VALUE float_arg(VALUE self);
+
+/*
+ * call-seq:
+ *    num.polar  ->  array
+ *
+ * Returns an array; [num.abs, num.arg].
+ */
+static VALUE
+numeric_polar(VALUE self)
+{
+    VALUE abs, arg;
+
+    if (RB_INTEGER_TYPE_P(self)) {
+        abs = rb_int_abs(self);
+        arg = numeric_arg(self);
+    }
+    else if (RB_FLOAT_TYPE_P(self)) {
+        abs = rb_float_abs(self);
+        arg = float_arg(self);
+    }
+    else if (RB_TYPE_P(self, T_RATIONAL)) {
+        abs = rb_rational_abs(self);
+        arg = numeric_arg(self);
+    }
+    else {
+        abs = f_abs(self);
+        arg = f_arg(self);
+    }
+    return rb_assoc_new(abs, arg);
+}
+
+/*
+ * call-seq:
+ *    num.conj       ->  self
+ *    num.conjugate  ->  self
+ *
+ * Returns self.
+ */
+static VALUE
+numeric_conj(VALUE self)
+{
+    return self;
+}
+
+/*
+ * call-seq:
+ *    flo.arg    ->  0 or float
+ *    flo.angle  ->  0 or float
+ *    flo.phase  ->  0 or float
+ *
+ * Returns 0 if the value is positive, pi otherwise.
+ */
+static VALUE
+float_arg(VALUE self)
+{
+    if (isnan(RFLOAT_VALUE(self)))
+	return self;
+    if (f_tpositive_p(self))
+	return INT2FIX(0);
+    return rb_const_get(rb_mMath, id_PI);
+}
+
+/*
+ * A complex number can be represented as a paired real number with
+ * imaginary unit; a+bi.  Where a is real part, b is imaginary part
+ * and i is imaginary unit.  Real a equals complex a+0i
+ * mathematically.
+ *
+ * Complex object can be created as literal, and also by using
+ * Kernel#Complex, Complex::rect, Complex::polar or to_c method.
+ *
+ *    2+1i                 #=> (2+1i)
+ *    Complex(1)           #=> (1+0i)
+ *    Complex(2, 3)        #=> (2+3i)
+ *    Complex.polar(2, 3)  #=> (-1.9799849932008908+0.2822400161197344i)
+ *    3.to_c               #=> (3+0i)
+ *
+ * You can also create complex object from floating-point numbers or
+ * strings.
+ *
+ *    Complex(0.3)         #=> (0.3+0i)
+ *    Complex('0.3-0.5i')  #=> (0.3-0.5i)
+ *    Complex('2/3+3/4i')  #=> ((2/3)+(3/4)*i)
+ *    Complex('1@2')       #=> (-0.4161468365471424+0.9092974268256817i)
+ *
+ *    0.3.to_c             #=> (0.3+0i)
+ *    '0.3-0.5i'.to_c      #=> (0.3-0.5i)
+ *    '2/3+3/4i'.to_c      #=> ((2/3)+(3/4)*i)
+ *    '1@2'.to_c           #=> (-0.4161468365471424+0.9092974268256817i)
+ *
+ * A complex object is either an exact or an inexact number.
+ *
+ *    Complex(1, 1) / 2    #=> ((1/2)+(1/2)*i)
+ *    Complex(1, 1) / 2.0  #=> (0.5+0.5i)
+ */
+void
+Init_Complex(void)
+{
+    VALUE compat;
+#undef rb_intern
+#define rb_intern(str) rb_intern_const(str)
+
+    id_abs = rb_intern("abs");
+    id_arg = rb_intern("arg");
+    id_denominator = rb_intern("denominator");
+    id_fdiv = rb_intern("fdiv");
+    id_numerator = rb_intern("numerator");
+    id_quo = rb_intern("quo");
+    id_real_p = rb_intern("real?");
+    id_i_real = rb_intern("@real");
+    id_i_imag = rb_intern("@image"); /* @image, not @imag */
+    id_finite_p = rb_intern("finite?");
+    id_infinite_p = rb_intern("infinite?");
+    id_rationalize = rb_intern("rationalize");
+    id_PI = rb_intern("PI");
+
+    rb_cComplex = rb_define_class("Complex", rb_cNumeric);
+
+    rb_define_alloc_func(rb_cComplex, nucomp_s_alloc);
+    rb_undef_method(CLASS_OF(rb_cComplex), "allocate");
+
+    rb_undef_method(CLASS_OF(rb_cComplex), "new");
+
+    rb_define_singleton_method(rb_cComplex, "rectangular", nucomp_s_new, -1);
+    rb_define_singleton_method(rb_cComplex, "rect", nucomp_s_new, -1);
+    rb_define_singleton_method(rb_cComplex, "polar", nucomp_s_polar, -1);
+
+    rb_define_global_function("Complex", nucomp_f_complex, -1);
+
+    rb_undef_methods_from(rb_cComplex, rb_mComparable);
+    rb_undef_method(rb_cComplex, "%");
+    rb_undef_method(rb_cComplex, "<=>");
+    rb_undef_method(rb_cComplex, "div");
+    rb_undef_method(rb_cComplex, "divmod");
+    rb_undef_method(rb_cComplex, "floor");
+    rb_undef_method(rb_cComplex, "ceil");
+    rb_undef_method(rb_cComplex, "modulo");
+    rb_undef_method(rb_cComplex, "remainder");
+    rb_undef_method(rb_cComplex, "round");
+    rb_undef_method(rb_cComplex, "step");
+    rb_undef_method(rb_cComplex, "truncate");
+    rb_undef_method(rb_cComplex, "i");
+
+    rb_define_method(rb_cComplex, "real", rb_complex_real, 0);
+    rb_define_method(rb_cComplex, "imaginary", rb_complex_imag, 0);
+    rb_define_method(rb_cComplex, "imag", rb_complex_imag, 0);
+
+    rb_define_method(rb_cComplex, "-@", rb_complex_uminus, 0);
+    rb_define_method(rb_cComplex, "+", rb_complex_plus, 1);
+    rb_define_method(rb_cComplex, "-", rb_complex_minus, 1);
+    rb_define_method(rb_cComplex, "*", rb_complex_mul, 1);
+    rb_define_method(rb_cComplex, "/", rb_complex_div, 1);
+    rb_define_method(rb_cComplex, "quo", nucomp_quo, 1);
+    rb_define_method(rb_cComplex, "fdiv", nucomp_fdiv, 1);
+    rb_define_method(rb_cComplex, "**", rb_complex_pow, 1);
+
+    rb_define_method(rb_cComplex, "==", nucomp_eqeq_p, 1);
+    rb_define_method(rb_cComplex, "coerce", nucomp_coerce, 1);
+
+    rb_define_method(rb_cComplex, "abs", rb_complex_abs, 0);
+    rb_define_method(rb_cComplex, "magnitude", rb_complex_abs, 0);
+    rb_define_method(rb_cComplex, "abs2", nucomp_abs2, 0);
+    rb_define_method(rb_cComplex, "arg", rb_complex_arg, 0);
+    rb_define_method(rb_cComplex, "angle", rb_complex_arg, 0);
+    rb_define_method(rb_cComplex, "phase", rb_complex_arg, 0);
+    rb_define_method(rb_cComplex, "rectangular", nucomp_rect, 0);
+    rb_define_method(rb_cComplex, "rect", nucomp_rect, 0);
+    rb_define_method(rb_cComplex, "polar", nucomp_polar, 0);
+    rb_define_method(rb_cComplex, "conjugate", rb_complex_conjugate, 0);
+    rb_define_method(rb_cComplex, "conj", rb_complex_conjugate, 0);
+
+    rb_define_method(rb_cComplex, "real?", nucomp_false, 0);
+
+    rb_define_method(rb_cComplex, "numerator", nucomp_numerator, 0);
+    rb_define_method(rb_cComplex, "denominator", nucomp_denominator, 0);
+
+    rb_define_method(rb_cComplex, "hash", nucomp_hash, 0);
+    rb_define_method(rb_cComplex, "eql?", nucomp_eql_p, 1);
+
+    rb_define_method(rb_cComplex, "to_s", nucomp_to_s, 0);
+    rb_define_method(rb_cComplex, "inspect", nucomp_inspect, 0);
+
+    rb_undef_method(rb_cComplex, "positive?");
+    rb_undef_method(rb_cComplex, "negative?");
+
+    rb_define_method(rb_cComplex, "finite?", rb_complex_finite_p, 0);
+    rb_define_method(rb_cComplex, "infinite?", rb_complex_infinite_p, 0);
+
+    rb_define_private_method(rb_cComplex, "marshal_dump", nucomp_marshal_dump, 0);
+    /* :nodoc: */
+    compat = rb_define_class_under(rb_cComplex, "compatible", rb_cObject);
+    rb_define_private_method(compat, "marshal_load", nucomp_marshal_load, 1);
+    rb_marshal_define_compat(rb_cComplex, compat, nucomp_dumper, nucomp_loader);
+
+    /* --- */
+
+    rb_define_method(rb_cComplex, "to_i", nucomp_to_i, 0);
+    rb_define_method(rb_cComplex, "to_f", nucomp_to_f, 0);
+    rb_define_method(rb_cComplex, "to_r", nucomp_to_r, 0);
+    rb_define_method(rb_cComplex, "rationalize", nucomp_rationalize, -1);
+    rb_define_method(rb_cComplex, "to_c", nucomp_to_c, 0);
+    rb_define_method(rb_cNilClass, "to_c", nilclass_to_c, 0);
+    rb_define_method(rb_cNumeric, "to_c", numeric_to_c, 0);
+
+    rb_define_method(rb_cString, "to_c", string_to_c, 0);
+
+    rb_define_private_method(CLASS_OF(rb_cComplex), "convert", nucomp_s_convert, -1);
+
+    /* --- */
+
+    rb_define_method(rb_cNumeric, "real", numeric_real, 0);
+    rb_define_method(rb_cNumeric, "imaginary", numeric_imag, 0);
+    rb_define_method(rb_cNumeric, "imag", numeric_imag, 0);
+    rb_define_method(rb_cNumeric, "abs2", numeric_abs2, 0);
+    rb_define_method(rb_cNumeric, "arg", numeric_arg, 0);
+    rb_define_method(rb_cNumeric, "angle", numeric_arg, 0);
+    rb_define_method(rb_cNumeric, "phase", numeric_arg, 0);
+    rb_define_method(rb_cNumeric, "rectangular", numeric_rect, 0);
+    rb_define_method(rb_cNumeric, "rect", numeric_rect, 0);
+    rb_define_method(rb_cNumeric, "polar", numeric_polar, 0);
+    rb_define_method(rb_cNumeric, "conjugate", numeric_conj, 0);
+    rb_define_method(rb_cNumeric, "conj", numeric_conj, 0);
+
+    rb_define_method(rb_cFloat, "arg", float_arg, 0);
+    rb_define_method(rb_cFloat, "angle", float_arg, 0);
+    rb_define_method(rb_cFloat, "phase", float_arg, 0);
+
+    /*
+     * The imaginary unit.
+     */
+    rb_define_const(rb_cComplex, "I",
+		    f_complex_new_bang2(rb_cComplex, ZERO, ONE));
+
+#if !USE_FLONUM
+    rb_gc_register_mark_object(RFLOAT_0 = DBL2NUM(0.0));
+#endif
+
+    rb_provide("complex.so");	/* for backward compatibility */
+}
+
+/*
+Local variables:
+c-file-style: "ruby"
+End:
+*/
diff --git a/config.guess b/config.guess
deleted file mode 100644
index dd1688b7b5..0000000000
--- a/config.guess
+++ /dev/null
@@ -1,1459 +0,0 @@
-#! /bin/sh
-# Attempt to guess a canonical system name.
-#   Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
-#   2000, 2001, 2002, 2003 Free Software Foundation, Inc.
-
-timestamp='2004-06-11'
-
-# This file is free software; you can redistribute it and/or modify it
-# under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful, but
-# WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
-# General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
-#
-# As a special exception to the GNU General Public License, if you
-# distribute this file as part of a program that contains a
-# configuration script generated by Autoconf, you may include it under
-# the same distribution terms that you use for the rest of that program.
-
-# Originally written by Per Bothner <per@bothner.com>.
-# Please send patches to <config-patches@gnu.org>.  Submit a context
-# diff and a properly formatted ChangeLog entry.
-#
-# This script attempts to guess a canonical system name similar to
-# config.sub.  If it succeeds, it prints the system name on stdout, and
-# exits with 0.  Otherwise, it exits with 1.
-#
-# The plan is that this can be called by configure scripts if you
-# don't specify an explicit build system type.
-
-me=`echo "$0" | sed -e 's,.*/,,'`
-
-usage="\
-Usage: $0 [OPTION]
-
-Output the configuration name of the system \`$me' is run on.
-
-Operation modes:
-  -h, --help         print this help, then exit
-  -t, --time-stamp   print date of last modification, then exit
-  -v, --version      print version number, then exit
-
-Report bugs and patches to <config-patches@gnu.org>."
-
-version="\
-GNU config.guess ($timestamp)
-
-Originally written by Per Bothner.
-Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001
-Free Software Foundation, Inc.
-
-This is free software; see the source for copying conditions.  There is NO
-warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
-
-help="
-Try \`$me --help' for more information."
-
-# Parse command line
-while test $# -gt 0 ; do
-  case $1 in
-    --time-stamp | --time* | -t )
-       echo "$timestamp" ; exit 0 ;;
-    --version | -v )
-       echo "$version" ; exit 0 ;;
-    --help | --h* | -h )
-       echo "$usage"; exit 0 ;;
-    -- )     # Stop option processing
-       shift; break ;;
-    - )	# Use stdin as input.
-       break ;;
-    -* )
-       echo "$me: invalid option $1$help" >&2
-       exit 1 ;;
-    * )
-       break ;;
-  esac
-done
-
-if test $# != 0; then
-  echo "$me: too many arguments$help" >&2
-  exit 1
-fi
-
-trap 'exit 1' 1 2 15
-
-# CC_FOR_BUILD -- compiler used by this script. Note that the use of a
-# compiler to aid in system detection is discouraged as it requires
-# temporary files to be created and, as you can see below, it is a
-# headache to deal with in a portable fashion.
-
-# Historically, `CC_FOR_BUILD' used to be named `HOST_CC'. We still
-# use `HOST_CC' if defined, but it is deprecated.
-
-# Portable tmp directory creation inspired by the Autoconf team.
-
-set_cc_for_build='
-trap "exitcode=\$?; (rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null) && exit \$exitcode" 0 ;
-trap "rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null; exit 1" 1 2 13 15 ;
-: ${TMPDIR=/tmp} ;
- { tmp=`(umask 077 && mktemp -d -q "$TMPDIR/cgXXXXXX") 2>/dev/null` && test -n "$tmp" && test -d "$tmp" ; } ||
- { test -n "$RANDOM" && tmp=$TMPDIR/cg$$-$RANDOM && (umask 077 && mkdir $tmp) ; } ||
- { tmp=$TMPDIR/cg-$$ && (umask 077 && mkdir $tmp) && echo "Warning: creating insecure temp directory" >&2 ; } ||
- { echo "$me: cannot create a temporary directory in $TMPDIR" >&2 ; exit 1 ; } ;
-dummy=$tmp/dummy ;
-tmpfiles="$dummy.c $dummy.o $dummy.rel $dummy" ;
-case $CC_FOR_BUILD,$HOST_CC,$CC in
- ,,)    echo "int x;" > $dummy.c ;
-	for c in cc gcc c89 c99 ; do
-	  if ($c -c -o $dummy.o $dummy.c) >/dev/null 2>&1 ; then
-	     CC_FOR_BUILD="$c"; break ;
-	  fi ;
-	done ;
-	if test x"$CC_FOR_BUILD" = x ; then
-	  CC_FOR_BUILD=no_compiler_found ;
-	fi
-	;;
- ,,*)   CC_FOR_BUILD=$CC ;;
- ,*,*)  CC_FOR_BUILD=$HOST_CC ;;
-esac ;'
-
-# This is needed to find uname on a Pyramid OSx when run in the BSD universe.
-# (ghazi@noc.rutgers.edu 1994-08-24)
-if (test -f /.attbin/uname) >/dev/null 2>&1 ; then
-	PATH=$PATH:/.attbin ; export PATH
-fi
-
-UNAME_MACHINE=`(uname -m) 2>/dev/null` || UNAME_MACHINE=unknown
-UNAME_RELEASE=`(uname -r) 2>/dev/null` || UNAME_RELEASE=unknown
-UNAME_SYSTEM=`(uname -s) 2>/dev/null`  || UNAME_SYSTEM=unknown
-UNAME_VERSION=`(uname -v) 2>/dev/null` || UNAME_VERSION=unknown
-
-# Note: order is significant - the case branches are not exclusive.
-
-case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in
-    *:NetBSD:*:*)
-	# NetBSD (nbsd) targets should (where applicable) match one or
-	# more of the tupples: *-*-netbsdelf*, *-*-netbsdaout*,
-	# *-*-netbsdecoff* and *-*-netbsd*.  For targets that recently
-	# switched to ELF, *-*-netbsd* would select the old
-	# object file format.  This provides both forward
-	# compatibility and a consistent mechanism for selecting the
-	# object file format.
-	#
-	# Note: NetBSD doesn't particularly care about the vendor
-	# portion of the name.  We always set it to "unknown".
-	sysctl="sysctl -n hw.machine_arch"
-	UNAME_MACHINE_ARCH=`(/sbin/$sysctl 2>/dev/null || \
-	    /usr/sbin/$sysctl 2>/dev/null || echo unknown)`
-	case "${UNAME_MACHINE_ARCH}" in
-	    armeb) machine=armeb-unknown ;;
-	    arm*) machine=arm-unknown ;;
-	    sh3el) machine=shl-unknown ;;
-	    sh3eb) machine=sh-unknown ;;
-	    *) machine=${UNAME_MACHINE_ARCH}-unknown ;;
-	esac
-	# The Operating System including object format, if it has switched
-	# to ELF recently, or will in the future.
-	case "${UNAME_MACHINE_ARCH}" in
-	    arm*|i386|m68k|ns32k|sh3*|sparc|vax)
-		eval $set_cc_for_build
-		if echo __ELF__ | $CC_FOR_BUILD -E - 2>/dev/null \
-			| grep __ELF__ >/dev/null
-		then
-		    # Once all utilities can be ECOFF (netbsdecoff) or a.out (netbsdaout).
-		    # Return netbsd for either.  FIX?
-		    os=netbsd
-		else
-		    os=netbsdelf
-		fi
-		;;
-	    *)
-	        os=netbsd
-		;;
-	esac
-	# The OS release
-	# Debian GNU/NetBSD machines have a different userland, and
-	# thus, need a distinct triplet. However, they do not need
-	# kernel version information, so it can be replaced with a
-	# suitable tag, in the style of linux-gnu.
-	case "${UNAME_VERSION}" in
-	    Debian*)
-		release='-gnu'
-		;;
-	    *)
-		release=`echo ${UNAME_RELEASE}|sed -e 's/[-_].*/\./'`
-		;;
-	esac
-	# Since CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM:
-	# contains redundant information, the shorter form:
-	# CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM is used.
-	echo "${machine}-${os}${release}"
-	exit 0 ;;
-    amd64:OpenBSD:*:*)
-	echo x86_64-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    amiga:OpenBSD:*:*)
-	echo m68k-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    arc:OpenBSD:*:*)
-	echo mipsel-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    cats:OpenBSD:*:*)
-	echo arm-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    hp300:OpenBSD:*:*)
-	echo m68k-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    luna88k:OpenBSD:*:*)
-    	echo m88k-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    mac68k:OpenBSD:*:*)
-	echo m68k-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    macppc:OpenBSD:*:*)
-	echo powerpc-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    mvme68k:OpenBSD:*:*)
-	echo m68k-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    mvme88k:OpenBSD:*:*)
-	echo m88k-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    mvmeppc:OpenBSD:*:*)
-	echo powerpc-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    pmax:OpenBSD:*:*)
-	echo mipsel-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    sgi:OpenBSD:*:*)
-	echo mipseb-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    sun3:OpenBSD:*:*)
-	echo m68k-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    wgrisc:OpenBSD:*:*)
-	echo mipsel-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    *:OpenBSD:*:*)
-	echo ${UNAME_MACHINE}-unknown-openbsd${UNAME_RELEASE}
-	exit 0 ;;
-    *:ekkoBSD:*:*)
-	echo ${UNAME_MACHINE}-unknown-ekkobsd${UNAME_RELEASE}
-	exit 0 ;;
-    macppc:MirBSD:*:*)
-	echo powerppc-unknown-mirbsd${UNAME_RELEASE}
-	exit 0 ;;
-    *:MirBSD:*:*)
-	echo ${UNAME_MACHINE}-unknown-mirbsd${UNAME_RELEASE}
-	exit 0 ;;
-    alpha:OSF1:*:*)
-	case $UNAME_RELEASE in
-	*4.0)
-		UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $3}'`
-		;;
-	*5.*)
-	        UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $4}'`
-		;;
-	esac
-	# According to Compaq, /usr/sbin/psrinfo has been available on
-	# OSF/1 and Tru64 systems produced since 1995.  I hope that
-	# covers most systems running today.  This code pipes the CPU
-	# types through head -n 1, so we only detect the type of CPU 0.
-	ALPHA_CPU_TYPE=`/usr/sbin/psrinfo -v | sed -n -e 's/^  The alpha \(.*\) processor.*$/\1/p' | head -n 1`
-	case "$ALPHA_CPU_TYPE" in
-	    "EV4 (21064)")
-		UNAME_MACHINE="alpha" ;;
-	    "EV4.5 (21064)")
-		UNAME_MACHINE="alpha" ;;
-	    "LCA4 (21066/21068)")
-		UNAME_MACHINE="alpha" ;;
-	    "EV5 (21164)")
-		UNAME_MACHINE="alphaev5" ;;
-	    "EV5.6 (21164A)")
-		UNAME_MACHINE="alphaev56" ;;
-	    "EV5.6 (21164PC)")
-		UNAME_MACHINE="alphapca56" ;;
-	    "EV5.7 (21164PC)")
-		UNAME_MACHINE="alphapca57" ;;
-	    "EV6 (21264)")
-		UNAME_MACHINE="alphaev6" ;;
-	    "EV6.7 (21264A)")
-		UNAME_MACHINE="alphaev67" ;;
-	    "EV6.8CB (21264C)")
-		UNAME_MACHINE="alphaev68" ;;
-	    "EV6.8AL (21264B)")
-		UNAME_MACHINE="alphaev68" ;;
-	    "EV6.8CX (21264D)")
-		UNAME_MACHINE="alphaev68" ;;
-	    "EV6.9A (21264/EV69A)")
-		UNAME_MACHINE="alphaev69" ;;
-	    "EV7 (21364)")
-		UNAME_MACHINE="alphaev7" ;;
-	    "EV7.9 (21364A)")
-		UNAME_MACHINE="alphaev79" ;;
-	esac
-	# A Pn.n version is a patched version.
-	# A Vn.n version is a released version.
-	# A Tn.n version is a released field test version.
-	# A Xn.n version is an unreleased experimental baselevel.
-	# 1.2 uses "1.2" for uname -r.
-	echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[PVTX]//' | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'`
-	exit 0 ;;
-    Alpha*:OpenVMS:*:*)
-	echo alpha-hp-vms
-	exit 0 ;;
-    Alpha\ *:Windows_NT*:*)
-	# How do we know it's Interix rather than the generic POSIX subsystem?
-	# Should we change UNAME_MACHINE based on the output of uname instead
-	# of the specific Alpha model?
-	echo alpha-pc-interix
-	exit 0 ;;
-    21064:Windows_NT:50:3)
-	echo alpha-dec-winnt3.5
-	exit 0 ;;
-    Amiga*:UNIX_System_V:4.0:*)
-	echo m68k-unknown-sysv4
-	exit 0;;
-    *:[Aa]miga[Oo][Ss]:*:*)
-	echo ${UNAME_MACHINE}-unknown-amigaos
-	exit 0 ;;
-    *:[Mm]orph[Oo][Ss]:*:*)
-	echo ${UNAME_MACHINE}-unknown-morphos
-	exit 0 ;;
-    *:OS/390:*:*)
-	echo i370-ibm-openedition
-	exit 0 ;;
-    *:OS400:*:*)
-        echo powerpc-ibm-os400
-	exit 0 ;;
-    arm:RISC*:1.[012]*:*|arm:riscix:1.[012]*:*)
-	echo arm-acorn-riscix${UNAME_RELEASE}
-	exit 0;;
-    SR2?01:HI-UX/MPP:*:* | SR8000:HI-UX/MPP:*:*)
-	echo hppa1.1-hitachi-hiuxmpp
-	exit 0;;
-    Pyramid*:OSx*:*:* | MIS*:OSx*:*:* | MIS*:SMP_DC-OSx*:*:*)
-	# akee@wpdis03.wpafb.af.mil (Earle F. Ake) contributed MIS and NILE.
-	if test "`(/bin/universe) 2>/dev/null`" = att ; then
-		echo pyramid-pyramid-sysv3
-	else
-		echo pyramid-pyramid-bsd
-	fi
-	exit 0 ;;
-    NILE*:*:*:dcosx)
-	echo pyramid-pyramid-svr4
-	exit 0 ;;
-    DRS?6000:unix:4.0:6*)
-	echo sparc-icl-nx6
-	exit 0 ;;
-    DRS?6000:UNIX_SV:4.2*:7*)
-	case `/usr/bin/uname -p` in
-	    sparc) echo sparc-icl-nx7 && exit 0 ;;
-	esac ;;
-    sun4H:SunOS:5.*:*)
-	echo sparc-hal-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
-	exit 0 ;;
-    sun4*:SunOS:5.*:* | tadpole*:SunOS:5.*:*)
-	echo sparc-sun-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
-	exit 0 ;;
-    i86pc:SunOS:5.*:*)
-	echo i386-pc-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
-	exit 0 ;;
-    sun4*:SunOS:6*:*)
-	# According to config.sub, this is the proper way to canonicalize
-	# SunOS6.  Hard to guess exactly what SunOS6 will be like, but
-	# it's likely to be more like Solaris than SunOS4.
-	echo sparc-sun-solaris3`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
-	exit 0 ;;
-    sun4*:SunOS:*:*)
-	case "`/usr/bin/arch -k`" in
-	    Series*|S4*)
-		UNAME_RELEASE=`uname -v`
-		;;
-	esac
-	# Japanese Language versions have a version number like `4.1.3-JL'.
-	echo sparc-sun-sunos`echo ${UNAME_RELEASE}|sed -e 's/-/_/'`
-	exit 0 ;;
-    sun3*:SunOS:*:*)
-	echo m68k-sun-sunos${UNAME_RELEASE}
-	exit 0 ;;
-    sun*:*:4.2BSD:*)
-	UNAME_RELEASE=`(sed 1q /etc/motd | awk '{print substr($5,1,3)}') 2>/dev/null`
-	test "x${UNAME_RELEASE}" = "x" && UNAME_RELEASE=3
-	case "`/bin/arch`" in
-	    sun3)
-		echo m68k-sun-sunos${UNAME_RELEASE}
-		;;
-	    sun4)
-		echo sparc-sun-sunos${UNAME_RELEASE}
-		;;
-	esac
-	exit 0 ;;
-    aushp:SunOS:*:*)
-	echo sparc-auspex-sunos${UNAME_RELEASE}
-	exit 0 ;;
-    # The situation for MiNT is a little confusing.  The machine name
-    # can be virtually everything (everything which is not
-    # "atarist" or "atariste" at least should have a processor
-    # > m68000).  The system name ranges from "MiNT" over "FreeMiNT"
-    # to the lowercase version "mint" (or "freemint").  Finally
-    # the system name "TOS" denotes a system which is actually not
-    # MiNT.  But MiNT is downward compatible to TOS, so this should
-    # be no problem.
-    atarist[e]:*MiNT:*:* | atarist[e]:*mint:*:* | atarist[e]:*TOS:*:*)
-        echo m68k-atari-mint${UNAME_RELEASE}
-	exit 0 ;;
-    atari*:*MiNT:*:* | atari*:*mint:*:* | atarist[e]:*TOS:*:*)
-	echo m68k-atari-mint${UNAME_RELEASE}
-        exit 0 ;;
-    *falcon*:*MiNT:*:* | *falcon*:*mint:*:* | *falcon*:*TOS:*:*)
-        echo m68k-atari-mint${UNAME_RELEASE}
-	exit 0 ;;
-    milan*:*MiNT:*:* | milan*:*mint:*:* | *milan*:*TOS:*:*)
-        echo m68k-milan-mint${UNAME_RELEASE}
-        exit 0 ;;
-    hades*:*MiNT:*:* | hades*:*mint:*:* | *hades*:*TOS:*:*)
-        echo m68k-hades-mint${UNAME_RELEASE}
-        exit 0 ;;
-    *:*MiNT:*:* | *:*mint:*:* | *:*TOS:*:*)
-        echo m68k-unknown-mint${UNAME_RELEASE}
-        exit 0 ;;
-    m68k:machten:*:*)
-	echo m68k-apple-machten${UNAME_RELEASE}
-	exit 0 ;;
-    powerpc:machten:*:*)
-	echo powerpc-apple-machten${UNAME_RELEASE}
-	exit 0 ;;
-    RISC*:Mach:*:*)
-	echo mips-dec-mach_bsd4.3
-	exit 0 ;;
-    RISC*:ULTRIX:*:*)
-	echo mips-dec-ultrix${UNAME_RELEASE}
-	exit 0 ;;
-    VAX*:ULTRIX*:*:*)
-	echo vax-dec-ultrix${UNAME_RELEASE}
-	exit 0 ;;
-    2020:CLIX:*:* | 2430:CLIX:*:*)
-	echo clipper-intergraph-clix${UNAME_RELEASE}
-	exit 0 ;;
-    mips:*:*:UMIPS | mips:*:*:RISCos)
-	eval $set_cc_for_build
-	sed 's/^	//' << EOF >$dummy.c
-#ifdef __cplusplus
-#include <stdio.h>  /* for printf() prototype */
-	int main (int argc, char *argv[]) {
-#else
-	int main (argc, argv) int argc; char *argv[]; {
-#endif
-	#if defined (host_mips) && defined (MIPSEB)
-	#if defined (SYSTYPE_SYSV)
-	  printf ("mips-mips-riscos%ssysv\n", argv[1]); exit (0);
-	#endif
-	#if defined (SYSTYPE_SVR4)
-	  printf ("mips-mips-riscos%ssvr4\n", argv[1]); exit (0);
-	#endif
-	#if defined (SYSTYPE_BSD43) || defined(SYSTYPE_BSD)
-	  printf ("mips-mips-riscos%sbsd\n", argv[1]); exit (0);
-	#endif
-	#endif
-	  exit (-1);
-	}
-EOF
-	$CC_FOR_BUILD -o $dummy $dummy.c \
-	  && $dummy `echo "${UNAME_RELEASE}" | sed -n 's/\([0-9]*\).*/\1/p'` \
-	  && exit 0
-	echo mips-mips-riscos${UNAME_RELEASE}
-	exit 0 ;;
-    Motorola:PowerMAX_OS:*:*)
-	echo powerpc-motorola-powermax
-	exit 0 ;;
-    Motorola:*:4.3:PL8-*)
-	echo powerpc-harris-powermax
-	exit 0 ;;
-    Night_Hawk:*:*:PowerMAX_OS | Synergy:PowerMAX_OS:*:*)
-	echo powerpc-harris-powermax
-	exit 0 ;;
-    Night_Hawk:Power_UNIX:*:*)
-	echo powerpc-harris-powerunix
-	exit 0 ;;
-    m88k:CX/UX:7*:*)
-	echo m88k-harris-cxux7
-	exit 0 ;;
-    m88k:*:4*:R4*)
-	echo m88k-motorola-sysv4
-	exit 0 ;;
-    m88k:*:3*:R3*)
-	echo m88k-motorola-sysv3
-	exit 0 ;;
-    AViiON:dgux:*:*)
-        # DG/UX returns AViiON for all architectures
-        UNAME_PROCESSOR=`/usr/bin/uname -p`
-	if [ $UNAME_PROCESSOR = mc88100 ] || [ $UNAME_PROCESSOR = mc88110 ]
-	then
-	    if [ ${TARGET_BINARY_INTERFACE}x = m88kdguxelfx ] || \
-	       [ ${TARGET_BINARY_INTERFACE}x = x ]
-	    then
-		echo m88k-dg-dgux${UNAME_RELEASE}
-	    else
-		echo m88k-dg-dguxbcs${UNAME_RELEASE}
-	    fi
-	else
-	    echo i586-dg-dgux${UNAME_RELEASE}
-	fi
- 	exit 0 ;;
-    M88*:DolphinOS:*:*)	# DolphinOS (SVR3)
-	echo m88k-dolphin-sysv3
-	exit 0 ;;
-    M88*:*:R3*:*)
-	# Delta 88k system running SVR3
-	echo m88k-motorola-sysv3
-	exit 0 ;;
-    XD88*:*:*:*) # Tektronix XD88 system running UTekV (SVR3)
-	echo m88k-tektronix-sysv3
-	exit 0 ;;
-    Tek43[0-9][0-9]:UTek:*:*) # Tektronix 4300 system running UTek (BSD)
-	echo m68k-tektronix-bsd
-	exit 0 ;;
-    *:IRIX*:*:*)
-	echo mips-sgi-irix`echo ${UNAME_RELEASE}|sed -e 's/-/_/g'`
-	exit 0 ;;
-    ????????:AIX?:[12].1:2)   # AIX 2.2.1 or AIX 2.1.1 is RT/PC AIX.
-	echo romp-ibm-aix      # uname -m gives an 8 hex-code CPU id
-	exit 0 ;;              # Note that: echo "'`uname -s`'" gives 'AIX '
-    i*86:AIX:*:*)
-	echo i386-ibm-aix
-	exit 0 ;;
-    ia64:AIX:*:*)
-	if [ -x /usr/bin/oslevel ] ; then
-		IBM_REV=`/usr/bin/oslevel`
-	else
-		IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
-	fi
-	echo ${UNAME_MACHINE}-ibm-aix${IBM_REV}
-	exit 0 ;;
-    *:AIX:2:3)
-	if grep bos325 /usr/include/stdio.h >/dev/null 2>&1; then
-		eval $set_cc_for_build
-		sed 's/^		//' << EOF >$dummy.c
-		#include <sys/systemcfg.h>
-
-		main()
-			{
-			if (!__power_pc())
-				exit(1);
-			puts("powerpc-ibm-aix3.2.5");
-			exit(0);
-			}
-EOF
-		$CC_FOR_BUILD -o $dummy $dummy.c && $dummy && exit 0
-		echo rs6000-ibm-aix3.2.5
-	elif grep bos324 /usr/include/stdio.h >/dev/null 2>&1; then
-		echo rs6000-ibm-aix3.2.4
-	else
-		echo rs6000-ibm-aix3.2
-	fi
-	exit 0 ;;
-    *:AIX:*:[45])
-	IBM_CPU_ID=`/usr/sbin/lsdev -C -c processor -S available | sed 1q | awk '{ print $1 }'`
-	if /usr/sbin/lsattr -El ${IBM_CPU_ID} | grep ' POWER' >/dev/null 2>&1; then
-		IBM_ARCH=rs6000
-	else
-		IBM_ARCH=powerpc
-	fi
-	if [ -x /usr/bin/oslevel ] ; then
-		IBM_REV=`/usr/bin/oslevel`
-	else
-		IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
-	fi
-	echo ${IBM_ARCH}-ibm-aix${IBM_REV}
-	exit 0 ;;
-    *:AIX:*:*)
-	echo rs6000-ibm-aix
-	exit 0 ;;
-    ibmrt:4.4BSD:*|romp-ibm:BSD:*)
-	echo romp-ibm-bsd4.4
-	exit 0 ;;
-    ibmrt:*BSD:*|romp-ibm:BSD:*)            # covers RT/PC BSD and
-	echo romp-ibm-bsd${UNAME_RELEASE}   # 4.3 with uname added to
-	exit 0 ;;                           # report: romp-ibm BSD 4.3
-    *:BOSX:*:*)
-	echo rs6000-bull-bosx
-	exit 0 ;;
-    DPX/2?00:B.O.S.:*:*)
-	echo m68k-bull-sysv3
-	exit 0 ;;
-    9000/[34]??:4.3bsd:1.*:*)
-	echo m68k-hp-bsd
-	exit 0 ;;
-    hp300:4.4BSD:*:* | 9000/[34]??:4.3bsd:2.*:*)
-	echo m68k-hp-bsd4.4
-	exit 0 ;;
-    9000/[34678]??:HP-UX:*:*)
-	HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'`
-	case "${UNAME_MACHINE}" in
-	    9000/31? )            HP_ARCH=m68000 ;;
-	    9000/[34]?? )         HP_ARCH=m68k ;;
-	    9000/[678][0-9][0-9])
-		if [ -x /usr/bin/getconf ]; then
-		    sc_cpu_version=`/usr/bin/getconf SC_CPU_VERSION 2>/dev/null`
-                    sc_kernel_bits=`/usr/bin/getconf SC_KERNEL_BITS 2>/dev/null`
-                    case "${sc_cpu_version}" in
-                      523) HP_ARCH="hppa1.0" ;; # CPU_PA_RISC1_0
-                      528) HP_ARCH="hppa1.1" ;; # CPU_PA_RISC1_1
-                      532)                      # CPU_PA_RISC2_0
-                        case "${sc_kernel_bits}" in
-                          32) HP_ARCH="hppa2.0n" ;;
-                          64) HP_ARCH="hppa2.0w" ;;
-			  '') HP_ARCH="hppa2.0" ;;   # HP-UX 10.20
-                        esac ;;
-                    esac
-		fi
-		if [ "${HP_ARCH}" = "" ]; then
-		    eval $set_cc_for_build
-		    sed 's/^              //' << EOF >$dummy.c
-
-              #define _HPUX_SOURCE
-              #include <stdlib.h>
-              #include <unistd.h>
-
-              int main ()
-              {
-              #if defined(_SC_KERNEL_BITS)
-                  long bits = sysconf(_SC_KERNEL_BITS);
-              #endif
-                  long cpu  = sysconf (_SC_CPU_VERSION);
-
-                  switch (cpu)
-              	{
-              	case CPU_PA_RISC1_0: puts ("hppa1.0"); break;
-              	case CPU_PA_RISC1_1: puts ("hppa1.1"); break;
-              	case CPU_PA_RISC2_0:
-              #if defined(_SC_KERNEL_BITS)
-              	    switch (bits)
-              		{
-              		case 64: puts ("hppa2.0w"); break;
-              		case 32: puts ("hppa2.0n"); break;
-              		default: puts ("hppa2.0"); break;
-              		} break;
-              #else  /* !defined(_SC_KERNEL_BITS) */
-              	    puts ("hppa2.0"); break;
-              #endif
-              	default: puts ("hppa1.0"); break;
-              	}
-                  exit (0);
-              }
-EOF
-		    (CCOPTS= $CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null) && HP_ARCH=`$dummy`
-		    test -z "$HP_ARCH" && HP_ARCH=hppa
-		fi ;;
-	esac
-	if [ ${HP_ARCH} = "hppa2.0w" ]
-	then
-	    # avoid double evaluation of $set_cc_for_build
-	    test -n "$CC_FOR_BUILD" || eval $set_cc_for_build
-	    if echo __LP64__ | (CCOPTS= $CC_FOR_BUILD -E -) | grep __LP64__ >/dev/null
-	    then
-		HP_ARCH="hppa2.0w"
-	    else
-		HP_ARCH="hppa64"
-	    fi
-	fi
-	echo ${HP_ARCH}-hp-hpux${HPUX_REV}
-	exit 0 ;;
-    ia64:HP-UX:*:*)
-	HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'`
-	echo ia64-hp-hpux${HPUX_REV}
-	exit 0 ;;
-    3050*:HI-UX:*:*)
-	eval $set_cc_for_build
-	sed 's/^	//' << EOF >$dummy.c
-	#include <unistd.h>
-	int
-	main ()
-	{
-	  long cpu = sysconf (_SC_CPU_VERSION);
-	  /* The order matters, because CPU_IS_HP_MC68K erroneously returns
-	     true for CPU_PA_RISC1_0.  CPU_IS_PA_RISC returns correct
-	     results, however.  */
-	  if (CPU_IS_PA_RISC (cpu))
-	    {
-	      switch (cpu)
-		{
-		  case CPU_PA_RISC1_0: puts ("hppa1.0-hitachi-hiuxwe2"); break;
-		  case CPU_PA_RISC1_1: puts ("hppa1.1-hitachi-hiuxwe2"); break;
-		  case CPU_PA_RISC2_0: puts ("hppa2.0-hitachi-hiuxwe2"); break;
-		  default: puts ("hppa-hitachi-hiuxwe2"); break;
-		}
-	    }
-	  else if (CPU_IS_HP_MC68K (cpu))
-	    puts ("m68k-hitachi-hiuxwe2");
-	  else puts ("unknown-hitachi-hiuxwe2");
-	  exit (0);
-	}
-EOF
-	$CC_FOR_BUILD -o $dummy $dummy.c && $dummy && exit 0
-	echo unknown-hitachi-hiuxwe2
-	exit 0 ;;
-    9000/7??:4.3bsd:*:* | 9000/8?[79]:4.3bsd:*:* )
-	echo hppa1.1-hp-bsd
-	exit 0 ;;
-    9000/8??:4.3bsd:*:*)
-	echo hppa1.0-hp-bsd
-	exit 0 ;;
-    *9??*:MPE/iX:*:* | *3000*:MPE/iX:*:*)
-	echo hppa1.0-hp-mpeix
-	exit 0 ;;
-    hp7??:OSF1:*:* | hp8?[79]:OSF1:*:* )
-	echo hppa1.1-hp-osf
-	exit 0 ;;
-    hp8??:OSF1:*:*)
-	echo hppa1.0-hp-osf
-	exit 0 ;;
-    i*86:OSF1:*:*)
-	if [ -x /usr/sbin/sysversion ] ; then
-	    echo ${UNAME_MACHINE}-unknown-osf1mk
-	else
-	    echo ${UNAME_MACHINE}-unknown-osf1
-	fi
-	exit 0 ;;
-    parisc*:Lites*:*:*)
-	echo hppa1.1-hp-lites
-	exit 0 ;;
-    C1*:ConvexOS:*:* | convex:ConvexOS:C1*:*)
-	echo c1-convex-bsd
-        exit 0 ;;
-    C2*:ConvexOS:*:* | convex:ConvexOS:C2*:*)
-	if getsysinfo -f scalar_acc
-	then echo c32-convex-bsd
-	else echo c2-convex-bsd
-	fi
-        exit 0 ;;
-    C34*:ConvexOS:*:* | convex:ConvexOS:C34*:*)
-	echo c34-convex-bsd
-        exit 0 ;;
-    C38*:ConvexOS:*:* | convex:ConvexOS:C38*:*)
-	echo c38-convex-bsd
-        exit 0 ;;
-    C4*:ConvexOS:*:* | convex:ConvexOS:C4*:*)
-	echo c4-convex-bsd
-        exit 0 ;;
-    CRAY*Y-MP:*:*:*)
-	echo ymp-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
-	exit 0 ;;
-    CRAY*[A-Z]90:*:*:*)
-	echo ${UNAME_MACHINE}-cray-unicos${UNAME_RELEASE} \
-	| sed -e 's/CRAY.*\([A-Z]90\)/\1/' \
-	      -e y/ABCDEFGHIJKLMNOPQRSTUVWXYZ/abcdefghijklmnopqrstuvwxyz/ \
-	      -e 's/\.[^.]*$/.X/'
-	exit 0 ;;
-    CRAY*TS:*:*:*)
-	echo t90-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
-	exit 0 ;;
-    CRAY*T3E:*:*:*)
-	echo alphaev5-cray-unicosmk${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
-	exit 0 ;;
-    CRAY*SV1:*:*:*)
-	echo sv1-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
-	exit 0 ;;
-    *:UNICOS/mp:*:*)
-	echo nv1-cray-unicosmp${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
-	exit 0 ;;
-    F30[01]:UNIX_System_V:*:* | F700:UNIX_System_V:*:*)
-	FUJITSU_PROC=`uname -m | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'`
-        FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'`
-        FUJITSU_REL=`echo ${UNAME_RELEASE} | sed -e 's/ /_/'`
-        echo "${FUJITSU_PROC}-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}"
-        exit 0 ;;
-    5000:UNIX_System_V:4.*:*)
-        FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'`
-        FUJITSU_REL=`echo ${UNAME_RELEASE} | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/ /_/'`
-        echo "sparc-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}"
-	exit 0 ;;
-    i*86:BSD/386:*:* | i*86:BSD/OS:*:* | *:Ascend\ Embedded/OS:*:*)
-	echo ${UNAME_MACHINE}-pc-bsdi${UNAME_RELEASE}
-	exit 0 ;;
-    sparc*:BSD/OS:*:*)
-	echo sparc-unknown-bsdi${UNAME_RELEASE}
-	exit 0 ;;
-    *:BSD/OS:*:*)
-	echo ${UNAME_MACHINE}-unknown-bsdi${UNAME_RELEASE}
-	exit 0 ;;
-    *:FreeBSD:*:*)
-	# Determine whether the default compiler uses glibc.
-	eval $set_cc_for_build
-	sed 's/^	//' << EOF >$dummy.c
-	#include <features.h>
-	#if __GLIBC__ >= 2
-	LIBC=gnu
-	#else
-	LIBC=
-	#endif
-EOF
-	eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^LIBC=`
-	# GNU/KFreeBSD systems have a "k" prefix to indicate we are using
-	# FreeBSD's kernel, but not the complete OS.
-	case ${LIBC} in gnu) kernel_only='k' ;; esac
-	echo ${UNAME_MACHINE}-unknown-${kernel_only}freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`${LIBC:+-$LIBC}
-	exit 0 ;;
-    i*:CYGWIN*:*)
-	echo ${UNAME_MACHINE}-pc-cygwin
-	exit 0 ;;
-    i*:MINGW*:*)
-	echo ${UNAME_MACHINE}-pc-mingw32
-	exit 0 ;;
-    i*:PW*:*)
-	echo ${UNAME_MACHINE}-pc-pw32
-	exit 0 ;;
-    x86:Interix*:[34]*)
-	echo i586-pc-interix${UNAME_RELEASE}|sed -e 's/\..*//'
-	exit 0 ;;
-    [345]86:Windows_95:* | [345]86:Windows_98:* | [345]86:Windows_NT:*)
-	echo i${UNAME_MACHINE}-pc-mks
-	exit 0 ;;
-    i*:Windows_NT*:* | Pentium*:Windows_NT*:*)
-	# How do we know it's Interix rather than the generic POSIX subsystem?
-	# It also conflicts with pre-2.0 versions of AT&T UWIN. Should we
-	# UNAME_MACHINE based on the output of uname instead of i386?
-	echo i586-pc-interix
-	exit 0 ;;
-    i*:UWIN*:*)
-	echo ${UNAME_MACHINE}-pc-uwin
-	exit 0 ;;
-    p*:CYGWIN*:*)
-	echo powerpcle-unknown-cygwin
-	exit 0 ;;
-    prep*:SunOS:5.*:*)
-	echo powerpcle-unknown-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
-	exit 0 ;;
-    *:GNU:*:*)
-	# the GNU system
-	echo `echo ${UNAME_MACHINE}|sed -e 's,[-/].*$,,'`-unknown-gnu`echo ${UNAME_RELEASE}|sed -e 's,/.*$,,'`
-	exit 0 ;;
-    *:GNU/*:*:*)
-	# other systems with GNU libc and userland
-	echo ${UNAME_MACHINE}-unknown-`echo ${UNAME_SYSTEM} | sed 's,^[^/]*/,,' | tr '[A-Z]' '[a-z]'``echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`-gnu
-	exit 0 ;;
-    i*86:Minix:*:*)
-	echo ${UNAME_MACHINE}-pc-minix
-	exit 0 ;;
-    arm*:Linux:*:*)
-	echo ${UNAME_MACHINE}-unknown-linux-gnu
-	exit 0 ;;
-    cris:Linux:*:*)
-	echo cris-axis-linux-gnu
-	exit 0 ;;
-    ia64:Linux:*:*)
-	echo ${UNAME_MACHINE}-unknown-linux-gnu
-	exit 0 ;;
-    m32r*:Linux:*:*)
-	echo ${UNAME_MACHINE}-unknown-linux-gnu
-	exit 0 ;;
-    m68*:Linux:*:*)
-	echo ${UNAME_MACHINE}-unknown-linux-gnu
-	exit 0 ;;
-    mips:Linux:*:*)
-	eval $set_cc_for_build
-	sed 's/^	//' << EOF >$dummy.c
-	#undef CPU
-	#undef mips
-	#undef mipsel
-	#if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL)
-	CPU=mipsel
-	#else
-	#if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB)
-	CPU=mips
-	#else
-	CPU=
-	#endif
-	#endif
-EOF
-	eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^CPU=`
-	test x"${CPU}" != x && echo "${CPU}-unknown-linux-gnu" && exit 0
-	;;
-    mips64:Linux:*:*)
-	eval $set_cc_for_build
-	sed 's/^	//' << EOF >$dummy.c
-	#undef CPU
-	#undef mips64
-	#undef mips64el
-	#if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL)
-	CPU=mips64el
-	#else
-	#if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB)
-	CPU=mips64
-	#else
-	CPU=
-	#endif
-	#endif
-EOF
-	eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^CPU=`
-	test x"${CPU}" != x && echo "${CPU}-unknown-linux-gnu" && exit 0
-	;;
-    ppc:Linux:*:*)
-	echo powerpc-unknown-linux-gnu
-	exit 0 ;;
-    ppc64:Linux:*:*)
-	echo powerpc64-unknown-linux-gnu
-	exit 0 ;;
-    alpha:Linux:*:*)
-	case `sed -n '/^cpu model/s/^.*: \(.*\)/\1/p' < /proc/cpuinfo` in
-	  EV5)   UNAME_MACHINE=alphaev5 ;;
-	  EV56)  UNAME_MACHINE=alphaev56 ;;
-	  PCA56) UNAME_MACHINE=alphapca56 ;;
-	  PCA57) UNAME_MACHINE=alphapca56 ;;
-	  EV6)   UNAME_MACHINE=alphaev6 ;;
-	  EV67)  UNAME_MACHINE=alphaev67 ;;
-	  EV68*) UNAME_MACHINE=alphaev68 ;;
-        esac
-	objdump --private-headers /bin/sh | grep ld.so.1 >/dev/null
-	if test "$?" = 0 ; then LIBC="libc1" ; else LIBC="" ; fi
-	echo ${UNAME_MACHINE}-unknown-linux-gnu${LIBC}
-	exit 0 ;;
-    parisc:Linux:*:* | hppa:Linux:*:*)
-	# Look for CPU level
-	case `grep '^cpu[^a-z]*:' /proc/cpuinfo 2>/dev/null | cut -d' ' -f2` in
-	  PA7*) echo hppa1.1-unknown-linux-gnu ;;
-	  PA8*) echo hppa2.0-unknown-linux-gnu ;;
-	  *)    echo hppa-unknown-linux-gnu ;;
-	esac
-	exit 0 ;;
-    parisc64:Linux:*:* | hppa64:Linux:*:*)
-	echo hppa64-unknown-linux-gnu
-	exit 0 ;;
-    s390:Linux:*:* | s390x:Linux:*:*)
-	echo ${UNAME_MACHINE}-ibm-linux
-	exit 0 ;;
-    sh64*:Linux:*:*)
-    	echo ${UNAME_MACHINE}-unknown-linux-gnu
-	exit 0 ;;
-    sh*:Linux:*:*)
-	echo ${UNAME_MACHINE}-unknown-linux-gnu
-	exit 0 ;;
-    sparc:Linux:*:* | sparc64:Linux:*:*)
-	echo ${UNAME_MACHINE}-unknown-linux-gnu
-	exit 0 ;;
-    x86_64:Linux:*:*)
-	echo x86_64-unknown-linux-gnu
-	exit 0 ;;
-    i*86:Linux:*:*)
-	# The BFD linker knows what the default object file format is, so
-	# first see if it will tell us. cd to the root directory to prevent
-	# problems with other programs or directories called `ld' in the path.
-	# Set LC_ALL=C to ensure ld outputs messages in English.
-	ld_supported_targets=`cd /; LC_ALL=C ld --help 2>&1 \
-			 | sed -ne '/supported targets:/!d
-				    s/[ 	][ 	]*/ /g
-				    s/.*supported targets: *//
-				    s/ .*//
-				    p'`
-        case "$ld_supported_targets" in
-	  elf32-i386)
-		TENTATIVE="${UNAME_MACHINE}-pc-linux-gnu"
-		;;
-	  a.out-i386-linux)
-		echo "${UNAME_MACHINE}-pc-linux-gnuaout"
-		exit 0 ;;
-	  coff-i386)
-		echo "${UNAME_MACHINE}-pc-linux-gnucoff"
-		exit 0 ;;
-	  "")
-		# Either a pre-BFD a.out linker (linux-gnuoldld) or
-		# one that does not give us useful --help.
-		echo "${UNAME_MACHINE}-pc-linux-gnuoldld"
-		exit 0 ;;
-	esac
-	# Determine whether the default compiler is a.out or elf
-	eval $set_cc_for_build
-	sed 's/^	//' << EOF >$dummy.c
-	#include <features.h>
-	#ifdef __ELF__
-	# ifdef __GLIBC__
-	#  if __GLIBC__ >= 2
-	LIBC=gnu
-	#  else
-	LIBC=gnulibc1
-	#  endif
-	# else
-	LIBC=gnulibc1
-	# endif
-	#else
-	#ifdef __INTEL_COMPILER
-	LIBC=gnu
-	#else
-	LIBC=gnuaout
-	#endif
-	#endif
-	#ifdef __dietlibc__
-	LIBC=dietlibc
-	#endif
-EOF
-	eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^LIBC=`
-	test x"${LIBC}" != x && echo "${UNAME_MACHINE}-pc-linux-${LIBC}" && exit 0
-	test x"${TENTATIVE}" != x && echo "${TENTATIVE}" && exit 0
-	;;
-    i*86:DYNIX/ptx:4*:*)
-	# ptx 4.0 does uname -s correctly, with DYNIX/ptx in there.
-	# earlier versions are messed up and put the nodename in both
-	# sysname and nodename.
-	echo i386-sequent-sysv4
-	exit 0 ;;
-    i*86:UNIX_SV:4.2MP:2.*)
-        # Unixware is an offshoot of SVR4, but it has its own version
-        # number series starting with 2...
-        # I am not positive that other SVR4 systems won't match this,
-	# I just have to hope.  -- rms.
-        # Use sysv4.2uw... so that sysv4* matches it.
-	echo ${UNAME_MACHINE}-pc-sysv4.2uw${UNAME_VERSION}
-	exit 0 ;;
-    i*86:OS/2:*:*)
-	# If we were able to find `uname', then EMX Unix compatibility
-	# is probably installed.
-	echo ${UNAME_MACHINE}-pc-os2-emx
-	exit 0 ;;
-    i*86:XTS-300:*:STOP)
-	echo ${UNAME_MACHINE}-unknown-stop
-	exit 0 ;;
-    i*86:atheos:*:*)
-	echo ${UNAME_MACHINE}-unknown-atheos
-	exit 0 ;;
-	i*86:syllable:*:*)
-	echo ${UNAME_MACHINE}-pc-syllable
-	exit 0 ;;
-    i*86:LynxOS:2.*:* | i*86:LynxOS:3.[01]*:* | i*86:LynxOS:4.0*:*)
-	echo i386-unknown-lynxos${UNAME_RELEASE}
-	exit 0 ;;
-    i*86:*DOS:*:*)
-	echo ${UNAME_MACHINE}-pc-msdosdjgpp
-	exit 0 ;;
-    i*86:*:4.*:* | i*86:SYSTEM_V:4.*:*)
-	UNAME_REL=`echo ${UNAME_RELEASE} | sed 's/\/MP$//'`
-	if grep Novell /usr/include/link.h >/dev/null 2>/dev/null; then
-		echo ${UNAME_MACHINE}-univel-sysv${UNAME_REL}
-	else
-		echo ${UNAME_MACHINE}-pc-sysv${UNAME_REL}
-	fi
-	exit 0 ;;
-    i*86:*:5:[78]*)
-	case `/bin/uname -X | grep "^Machine"` in
-	    *486*)	     UNAME_MACHINE=i486 ;;
-	    *Pentium)	     UNAME_MACHINE=i586 ;;
-	    *Pent*|*Celeron) UNAME_MACHINE=i686 ;;
-	esac
-	echo ${UNAME_MACHINE}-unknown-sysv${UNAME_RELEASE}${UNAME_SYSTEM}${UNAME_VERSION}
-	exit 0 ;;
-    i*86:*:3.2:*)
-	if test -f /usr/options/cb.name; then
-		UNAME_REL=`sed -n 's/.*Version //p' </usr/options/cb.name`
-		echo ${UNAME_MACHINE}-pc-isc$UNAME_REL
-	elif /bin/uname -X 2>/dev/null >/dev/null ; then
-		UNAME_REL=`(/bin/uname -X|grep Release|sed -e 's/.*= //')`
-		(/bin/uname -X|grep i80486 >/dev/null) && UNAME_MACHINE=i486
-		(/bin/uname -X|grep '^Machine.*Pentium' >/dev/null) \
-			&& UNAME_MACHINE=i586
-		(/bin/uname -X|grep '^Machine.*Pent *II' >/dev/null) \
-			&& UNAME_MACHINE=i686
-		(/bin/uname -X|grep '^Machine.*Pentium Pro' >/dev/null) \
-			&& UNAME_MACHINE=i686
-		echo ${UNAME_MACHINE}-pc-sco$UNAME_REL
-	else
-		echo ${UNAME_MACHINE}-pc-sysv32
-	fi
-	exit 0 ;;
-    pc:*:*:*)
-	# Left here for compatibility:
-        # uname -m prints for DJGPP always 'pc', but it prints nothing about
-        # the processor, so we play safe by assuming i386.
-	echo i386-pc-msdosdjgpp
-        exit 0 ;;
-    Intel:Mach:3*:*)
-	echo i386-pc-mach3
-	exit 0 ;;
-    paragon:*:*:*)
-	echo i860-intel-osf1
-	exit 0 ;;
-    i860:*:4.*:*) # i860-SVR4
-	if grep Stardent /usr/include/sys/uadmin.h >/dev/null 2>&1 ; then
-	  echo i860-stardent-sysv${UNAME_RELEASE} # Stardent Vistra i860-SVR4
-	else # Add other i860-SVR4 vendors below as they are discovered.
-	  echo i860-unknown-sysv${UNAME_RELEASE}  # Unknown i860-SVR4
-	fi
-	exit 0 ;;
-    mini*:CTIX:SYS*5:*)
-	# "miniframe"
-	echo m68010-convergent-sysv
-	exit 0 ;;
-    mc68k:UNIX:SYSTEM5:3.51m)
-	echo m68k-convergent-sysv
-	exit 0 ;;
-    M680?0:D-NIX:5.3:*)
-	echo m68k-diab-dnix
-	exit 0 ;;
-    M68*:*:R3V[5678]*:*)
-	test -r /sysV68 && echo 'm68k-motorola-sysv' && exit 0 ;;
-    3[345]??:*:4.0:3.0 | 3[34]??A:*:4.0:3.0 | 3[34]??,*:*:4.0:3.0 | 3[34]??/*:*:4.0:3.0 | 4400:*:4.0:3.0 | 4850:*:4.0:3.0 | SKA40:*:4.0:3.0 | SDS2:*:4.0:3.0 | SHG2:*:4.0:3.0)
-	OS_REL=''
-	test -r /etc/.relid \
-	&& OS_REL=.`sed -n 's/[^ ]* [^ ]* \([0-9][0-9]\).*/\1/p' < /etc/.relid`
-	/bin/uname -p 2>/dev/null | grep 86 >/dev/null \
-	  && echo i486-ncr-sysv4.3${OS_REL} && exit 0
-	/bin/uname -p 2>/dev/null | /bin/grep entium >/dev/null \
-	  && echo i586-ncr-sysv4.3${OS_REL} && exit 0 ;;
-    3[34]??:*:4.0:* | 3[34]??,*:*:4.0:*)
-        /bin/uname -p 2>/dev/null | grep 86 >/dev/null \
-          && echo i486-ncr-sysv4 && exit 0 ;;
-    m68*:LynxOS:2.*:* | m68*:LynxOS:3.0*:*)
-	echo m68k-unknown-lynxos${UNAME_RELEASE}
-	exit 0 ;;
-    mc68030:UNIX_System_V:4.*:*)
-	echo m68k-atari-sysv4
-	exit 0 ;;
-    TSUNAMI:LynxOS:2.*:*)
-	echo sparc-unknown-lynxos${UNAME_RELEASE}
-	exit 0 ;;
-    rs6000:LynxOS:2.*:*)
-	echo rs6000-unknown-lynxos${UNAME_RELEASE}
-	exit 0 ;;
-    PowerPC:LynxOS:2.*:* | PowerPC:LynxOS:3.[01]*:* | PowerPC:LynxOS:4.0*:*)
-	echo powerpc-unknown-lynxos${UNAME_RELEASE}
-	exit 0 ;;
-    SM[BE]S:UNIX_SV:*:*)
-	echo mips-dde-sysv${UNAME_RELEASE}
-	exit 0 ;;
-    RM*:ReliantUNIX-*:*:*)
-	echo mips-sni-sysv4
-	exit 0 ;;
-    RM*:SINIX-*:*:*)
-	echo mips-sni-sysv4
-	exit 0 ;;
-    *:SINIX-*:*:*)
-	if uname -p 2>/dev/null >/dev/null ; then
-		UNAME_MACHINE=`(uname -p) 2>/dev/null`
-		echo ${UNAME_MACHINE}-sni-sysv4
-	else
-		echo ns32k-sni-sysv
-	fi
-	exit 0 ;;
-    PENTIUM:*:4.0*:*) # Unisys `ClearPath HMP IX 4000' SVR4/MP effort
-                      # says <Richard.M.Bartel@ccMail.Census.GOV>
-        echo i586-unisys-sysv4
-        exit 0 ;;
-    *:UNIX_System_V:4*:FTX*)
-	# From Gerald Hewes <hewes@openmarket.com>.
-	# How about differentiating between stratus architectures? -djm
-	echo hppa1.1-stratus-sysv4
-	exit 0 ;;
-    *:*:*:FTX*)
-	# From seanf@swdc.stratus.com.
-	echo i860-stratus-sysv4
-	exit 0 ;;
-    *:VOS:*:*)
-	# From Paul.Green@stratus.com.
-	echo hppa1.1-stratus-vos
-	exit 0 ;;
-    mc68*:A/UX:*:*)
-	echo m68k-apple-aux${UNAME_RELEASE}
-	exit 0 ;;
-    news*:NEWS-OS:6*:*)
-	echo mips-sony-newsos6
-	exit 0 ;;
-    R[34]000:*System_V*:*:* | R4000:UNIX_SYSV:*:* | R*000:UNIX_SV:*:*)
-	if [ -d /usr/nec ]; then
-	        echo mips-nec-sysv${UNAME_RELEASE}
-	else
-	        echo mips-unknown-sysv${UNAME_RELEASE}
-	fi
-        exit 0 ;;
-    BeBox:BeOS:*:*)	# BeOS running on hardware made by Be, PPC only.
-	echo powerpc-be-beos
-	exit 0 ;;
-    BeMac:BeOS:*:*)	# BeOS running on Mac or Mac clone, PPC only.
-	echo powerpc-apple-beos
-	exit 0 ;;
-    BePC:BeOS:*:*)	# BeOS running on Intel PC compatible.
-	echo i586-pc-beos
-	exit 0 ;;
-    SX-4:SUPER-UX:*:*)
-	echo sx4-nec-superux${UNAME_RELEASE}
-	exit 0 ;;
-    SX-5:SUPER-UX:*:*)
-	echo sx5-nec-superux${UNAME_RELEASE}
-	exit 0 ;;
-    SX-6:SUPER-UX:*:*)
-	echo sx6-nec-superux${UNAME_RELEASE}
-	exit 0 ;;
-    Power*:Rhapsody:*:*)
-	echo powerpc-apple-rhapsody${UNAME_RELEASE}
-	exit 0 ;;
-    *:Rhapsody:*:*)
-	echo ${UNAME_MACHINE}-apple-rhapsody${UNAME_RELEASE}
-	exit 0 ;;
-    *:Darwin:*:*)
-	case `uname -p` in
-	    *86) UNAME_PROCESSOR=i686 ;;
-	    powerpc) UNAME_PROCESSOR=powerpc ;;
-	esac
-	echo ${UNAME_PROCESSOR}-apple-darwin${UNAME_RELEASE}
-	exit 0 ;;
-    *:procnto*:*:* | *:QNX:[0123456789]*:*)
-	UNAME_PROCESSOR=`uname -p`
-	if test "$UNAME_PROCESSOR" = "x86"; then
-		UNAME_PROCESSOR=i386
-		UNAME_MACHINE=pc
-	fi
-	echo ${UNAME_PROCESSOR}-${UNAME_MACHINE}-nto-qnx${UNAME_RELEASE}
-	exit 0 ;;
-    *:QNX:*:4*)
-	echo i386-pc-qnx
-	exit 0 ;;
-    NSR-?:NONSTOP_KERNEL:*:*)
-	echo nsr-tandem-nsk${UNAME_RELEASE}
-	exit 0 ;;
-    *:NonStop-UX:*:*)
-	echo mips-compaq-nonstopux
-	exit 0 ;;
-    BS2000:POSIX*:*:*)
-	echo bs2000-siemens-sysv
-	exit 0 ;;
-    DS/*:UNIX_System_V:*:*)
-	echo ${UNAME_MACHINE}-${UNAME_SYSTEM}-${UNAME_RELEASE}
-	exit 0 ;;
-    *:Plan9:*:*)
-	# "uname -m" is not consistent, so use $cputype instead. 386
-	# is converted to i386 for consistency with other x86
-	# operating systems.
-	if test "$cputype" = "386"; then
-	    UNAME_MACHINE=i386
-	else
-	    UNAME_MACHINE="$cputype"
-	fi
-	echo ${UNAME_MACHINE}-unknown-plan9
-	exit 0 ;;
-    *:TOPS-10:*:*)
-	echo pdp10-unknown-tops10
-	exit 0 ;;
-    *:TENEX:*:*)
-	echo pdp10-unknown-tenex
-	exit 0 ;;
-    KS10:TOPS-20:*:* | KL10:TOPS-20:*:* | TYPE4:TOPS-20:*:*)
-	echo pdp10-dec-tops20
-	exit 0 ;;
-    XKL-1:TOPS-20:*:* | TYPE5:TOPS-20:*:*)
-	echo pdp10-xkl-tops20
-	exit 0 ;;
-    *:TOPS-20:*:*)
-	echo pdp10-unknown-tops20
-	exit 0 ;;
-    *:ITS:*:*)
-	echo pdp10-unknown-its
-	exit 0 ;;
-    SEI:*:*:SEIUX)
-        echo mips-sei-seiux${UNAME_RELEASE}
-	exit 0 ;;
-    *:DragonFly:*:*)
-	echo ${UNAME_MACHINE}-unknown-dragonfly`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`
-	exit 0 ;;
-esac
-
-#echo '(No uname command or uname output not recognized.)' 1>&2
-#echo "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" 1>&2
-
-eval $set_cc_for_build
-cat >$dummy.c <<EOF
-#ifdef _SEQUENT_
-# include <sys/types.h>
-# include <sys/utsname.h>
-#endif
-main ()
-{
-#if defined (sony)
-#if defined (MIPSEB)
-  /* BFD wants "bsd" instead of "newsos".  Perhaps BFD should be changed,
-     I don't know....  */
-  printf ("mips-sony-bsd\n"); exit (0);
-#else
-#include <sys/param.h>
-  printf ("m68k-sony-newsos%s\n",
-#ifdef NEWSOS4
-          "4"
-#else
-	  ""
-#endif
-         ); exit (0);
-#endif
-#endif
-
-#if defined (__arm) && defined (__acorn) && defined (__unix)
-  printf ("arm-acorn-riscix"); exit (0);
-#endif
-
-#if defined (hp300) && !defined (hpux)
-  printf ("m68k-hp-bsd\n"); exit (0);
-#endif
-
-#if defined (NeXT)
-#if !defined (__ARCHITECTURE__)
-#define __ARCHITECTURE__ "m68k"
-#endif
-  int version;
-  version=`(hostinfo | sed -n 's/.*NeXT Mach \([0-9]*\).*/\1/p') 2>/dev/null`;
-  if (version < 4)
-    printf ("%s-next-nextstep%d\n", __ARCHITECTURE__, version);
-  else
-    printf ("%s-next-openstep%d\n", __ARCHITECTURE__, version);
-  exit (0);
-#endif
-
-#if defined (MULTIMAX) || defined (n16)
-#if defined (UMAXV)
-  printf ("ns32k-encore-sysv\n"); exit (0);
-#else
-#if defined (CMU)
-  printf ("ns32k-encore-mach\n"); exit (0);
-#else
-  printf ("ns32k-encore-bsd\n"); exit (0);
-#endif
-#endif
-#endif
-
-#if defined (__386BSD__)
-  printf ("i386-pc-bsd\n"); exit (0);
-#endif
-
-#if defined (sequent)
-#if defined (i386)
-  printf ("i386-sequent-dynix\n"); exit (0);
-#endif
-#if defined (ns32000)
-  printf ("ns32k-sequent-dynix\n"); exit (0);
-#endif
-#endif
-
-#if defined (_SEQUENT_)
-    struct utsname un;
-
-    uname(&un);
-
-    if (strncmp(un.version, "V2", 2) == 0) {
-	printf ("i386-sequent-ptx2\n"); exit (0);
-    }
-    if (strncmp(un.version, "V1", 2) == 0) { /* XXX is V1 correct? */
-	printf ("i386-sequent-ptx1\n"); exit (0);
-    }
-    printf ("i386-sequent-ptx\n"); exit (0);
-
-#endif
-
-#if defined (vax)
-# if !defined (ultrix)
-#  include <sys/param.h>
-#  if defined (BSD)
-#   if BSD == 43
-      printf ("vax-dec-bsd4.3\n"); exit (0);
-#   else
-#    if BSD == 199006
-      printf ("vax-dec-bsd4.3reno\n"); exit (0);
-#    else
-      printf ("vax-dec-bsd\n"); exit (0);
-#    endif
-#   endif
-#  else
-    printf ("vax-dec-bsd\n"); exit (0);
-#  endif
-# else
-    printf ("vax-dec-ultrix\n"); exit (0);
-# endif
-#endif
-
-#if defined (alliant) && defined (i860)
-  printf ("i860-alliant-bsd\n"); exit (0);
-#endif
-
-  exit (1);
-}
-EOF
-
-$CC_FOR_BUILD -o $dummy $dummy.c 2>/dev/null && $dummy && exit 0
-
-# Apollos put the system type in the environment.
-
-test -d /usr/apollo && { echo ${ISP}-apollo-${SYSTYPE}; exit 0; }
-
-# Convex versions that predate uname can use getsysinfo(1)
-
-if [ -x /usr/convex/getsysinfo ]
-then
-    case `getsysinfo -f cpu_type` in
-    c1*)
-	echo c1-convex-bsd
-	exit 0 ;;
-    c2*)
-	if getsysinfo -f scalar_acc
-	then echo c32-convex-bsd
-	else echo c2-convex-bsd
-	fi
-	exit 0 ;;
-    c34*)
-	echo c34-convex-bsd
-	exit 0 ;;
-    c38*)
-	echo c38-convex-bsd
-	exit 0 ;;
-    c4*)
-	echo c4-convex-bsd
-	exit 0 ;;
-    esac
-fi
-
-cat >&2 <<EOF
-$0: unable to guess system type
-
-This script, last modified $timestamp, has failed to recognize
-the operating system you are using. It is advised that you
-download the most up to date version of the config scripts from
-
-    ftp://ftp.gnu.org/pub/gnu/config/
-
-If the version you run ($0) is already up to date, please
-send the following data and any information you think might be
-pertinent to <config-patches@gnu.org> in order to provide the needed
-information to handle your system.
-
-config.guess timestamp = $timestamp
-
-uname -m = `(uname -m) 2>/dev/null || echo unknown`
-uname -r = `(uname -r) 2>/dev/null || echo unknown`
-uname -s = `(uname -s) 2>/dev/null || echo unknown`
-uname -v = `(uname -v) 2>/dev/null || echo unknown`
-
-/usr/bin/uname -p = `(/usr/bin/uname -p) 2>/dev/null`
-/bin/uname -X     = `(/bin/uname -X) 2>/dev/null`
-
-hostinfo               = `(hostinfo) 2>/dev/null`
-/bin/universe          = `(/bin/universe) 2>/dev/null`
-/usr/bin/arch -k       = `(/usr/bin/arch -k) 2>/dev/null`
-/bin/arch              = `(/bin/arch) 2>/dev/null`
-/usr/bin/oslevel       = `(/usr/bin/oslevel) 2>/dev/null`
-/usr/convex/getsysinfo = `(/usr/convex/getsysinfo) 2>/dev/null`
-
-UNAME_MACHINE = ${UNAME_MACHINE}
-UNAME_RELEASE = ${UNAME_RELEASE}
-UNAME_SYSTEM  = ${UNAME_SYSTEM}
-UNAME_VERSION = ${UNAME_VERSION}
-EOF
-
-exit 1
-
-# Local variables:
-# eval: (add-hook 'write-file-hooks 'time-stamp)
-# time-stamp-start: "timestamp='"
-# time-stamp-format: "%:y-%02m-%02d"
-# time-stamp-end: "'"
-# End:
diff --git a/config.sub b/config.sub
deleted file mode 100644
index 506d3ab77f..0000000000
--- a/config.sub
+++ /dev/null
@@ -1,1537 +0,0 @@
-#! /bin/sh
-# Configuration validation subroutine script.
-#   Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
-#   2000, 2001, 2002, 2003 Free Software Foundation, Inc.
-
-timestamp='2004-06-11'
-
-# This file is (in principle) common to ALL GNU software.
-# The presence of a machine in this file suggests that SOME GNU software
-# can handle that machine.  It does not imply ALL GNU software can.
-#
-# This file is free software; you can redistribute it and/or modify
-# it under the terms of the GNU General Public License as published by
-# the Free Software Foundation; either version 2 of the License, or
-# (at your option) any later version.
-#
-# This program is distributed in the hope that it will be useful,
-# but WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-# GNU General Public License for more details.
-#
-# You should have received a copy of the GNU General Public License
-# along with this program; if not, write to the Free Software
-# Foundation, Inc., 59 Temple Place - Suite 330,
-# Boston, MA 02111-1307, USA.
-
-# As a special exception to the GNU General Public License, if you
-# distribute this file as part of a program that contains a
-# configuration script generated by Autoconf, you may include it under
-# the same distribution terms that you use for the rest of that program.
-
-# Please send patches to <config-patches@gnu.org>.  Submit a context
-# diff and a properly formatted ChangeLog entry.
-#
-# Configuration subroutine to validate and canonicalize a configuration type.
-# Supply the specified configuration type as an argument.
-# If it is invalid, we print an error message on stderr and exit with code 1.
-# Otherwise, we print the canonical config type on stdout and succeed.
-
-# This file is supposed to be the same for all GNU packages
-# and recognize all the CPU types, system types and aliases
-# that are meaningful with *any* GNU software.
-# Each package is responsible for reporting which valid configurations
-# it does not support.  The user should be able to distinguish
-# a failure to support a valid configuration from a meaningless
-# configuration.
-
-# The goal of this file is to map all the various variations of a given
-# machine specification into a single specification in the form:
-#	CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM
-# or in some cases, the newer four-part form:
-#	CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM
-# It is wrong to echo any other type of specification.
-
-me=`echo "$0" | sed -e 's,.*/,,'`
-
-usage="\
-Usage: $0 [OPTION] CPU-MFR-OPSYS
-       $0 [OPTION] ALIAS
-
-Canonicalize a configuration name.
-
-Operation modes:
-  -h, --help         print this help, then exit
-  -t, --time-stamp   print date of last modification, then exit
-  -v, --version      print version number, then exit
-
-Report bugs and patches to <config-patches@gnu.org>."
-
-version="\
-GNU config.sub ($timestamp)
-
-Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001
-Free Software Foundation, Inc.
-
-This is free software; see the source for copying conditions.  There is NO
-warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
-
-help="
-Try \`$me --help' for more information."
-
-# Parse command line
-while test $# -gt 0 ; do
-  case $1 in
-    --time-stamp | --time* | -t )
-       echo "$timestamp" ; exit 0 ;;
-    --version | -v )
-       echo "$version" ; exit 0 ;;
-    --help | --h* | -h )
-       echo "$usage"; exit 0 ;;
-    -- )     # Stop option processing
-       shift; break ;;
-    - )	# Use stdin as input.
-       break ;;
-    -* )
-       echo "$me: invalid option $1$help"
-       exit 1 ;;
-
-    *local*)
-       # First pass through any local machine types.
-       echo $1
-       exit 0;;
-
-    * )
-       break ;;
-  esac
-done
-
-case $# in
- 0) echo "$me: missing argument$help" >&2
-    exit 1;;
- 1) ;;
- *) echo "$me: too many arguments$help" >&2
-    exit 1;;
-esac
-
-# Separate what the user gave into CPU-COMPANY and OS or KERNEL-OS (if any).
-# Here we must recognize all the valid KERNEL-OS combinations.
-maybe_os=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\2/'`
-case $maybe_os in
-  nto-qnx* | linux-gnu* | linux-dietlibc | linux-uclibc* | uclinux-uclibc* | uclinux-gnu* | \
-  kfreebsd*-gnu* | knetbsd*-gnu* | netbsd*-gnu* | storm-chaos* | os2-emx* | rtmk-nova*)
-    os=-$maybe_os
-    basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'`
-    ;;
-  *)
-    basic_machine=`echo $1 | sed 's/-[^-]*$//'`
-    if [ $basic_machine != $1 ]
-    then os=`echo $1 | sed 's/.*-/-/'`
-    else os=; fi
-    ;;
-esac
-
-### Let's recognize common machines as not being operating systems so
-### that things like config.sub decstation-3100 work.  We also
-### recognize some manufacturers as not being operating systems, so we
-### can provide default operating systems below.
-case $os in
-	-sun*os*)
-		# Prevent following clause from handling this invalid input.
-		;;
-	-dec* | -mips* | -sequent* | -encore* | -pc532* | -sgi* | -sony* | \
-	-att* | -7300* | -3300* | -delta* | -motorola* | -sun[234]* | \
-	-unicom* | -ibm* | -next | -hp | -isi* | -apollo | -altos* | \
-	-convergent* | -ncr* | -news | -32* | -3600* | -3100* | -hitachi* |\
-	-c[123]* | -convex* | -sun | -crds | -omron* | -dg | -ultra | -tti* | \
-	-harris | -dolphin | -highlevel | -gould | -cbm | -ns | -masscomp | \
-	-apple | -axis)
-		os=
-		basic_machine=$1
-		;;
-	-sim | -cisco | -oki | -wec | -winbond)
-		os=
-		basic_machine=$1
-		;;
-	-scout)
-		;;
-	-wrs)
-		os=-vxworks
-		basic_machine=$1
-		;;
-	-chorusos*)
-		os=-chorusos
-		basic_machine=$1
-		;;
- 	-chorusrdb)
- 		os=-chorusrdb
-		basic_machine=$1
- 		;;
-	-hiux*)
-		os=-hiuxwe2
-		;;
-	-sco5)
-		os=-sco3.2v5
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-sco4)
-		os=-sco3.2v4
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-sco3.2.[4-9]*)
-		os=`echo $os | sed -e 's/sco3.2./sco3.2v/'`
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-sco3.2v[4-9]*)
-		# Don't forget version if it is 3.2v4 or newer.
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-sco*)
-		os=-sco3.2v2
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-udk*)
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-isc)
-		os=-isc2.2
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-clix*)
-		basic_machine=clipper-intergraph
-		;;
-	-isc*)
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
-		;;
-	-lynx*)
-		os=-lynxos
-		;;
-	-ptx*)
-		basic_machine=`echo $1 | sed -e 's/86-.*/86-sequent/'`
-		;;
-	-windowsnt*)
-		os=`echo $os | sed -e 's/windowsnt/winnt/'`
-		;;
-	-psos*)
-		os=-psos
-		;;
-	-mint | -mint[0-9]*)
-		basic_machine=m68k-atari
-		os=-mint
-		;;
-esac
-
-# Decode aliases for certain CPU-COMPANY combinations.
-case $basic_machine in
-	# Recognize the basic CPU types without company name.
-	# Some are omitted here because they have special meanings below.
-	1750a | 580 \
-	| a29k \
-	| alpha | alphaev[4-8] | alphaev56 | alphaev6[78] | alphapca5[67] \
-	| alpha64 | alpha64ev[4-8] | alpha64ev56 | alpha64ev6[78] | alpha64pca5[67] \
-	| am33_2.0 \
-	| arc | arm | arm[bl]e | arme[lb] | armv[2345] | armv[345][lb] | avr \
-	| c4x | clipper \
-	| d10v | d30v | dlx | dsp16xx \
-	| fr30 | frv \
-	| h8300 | h8500 | hppa | hppa1.[01] | hppa2.0 | hppa2.0[nw] | hppa64 \
-	| i370 | i860 | i960 | ia64 \
-	| ip2k | iq2000 \
-	| m32r | m68000 | m68k | m88k | mcore \
-	| mips | mipsbe | mipseb | mipsel | mipsle \
-	| mips16 \
-	| mips64 | mips64el \
-	| mips64vr | mips64vrel \
-	| mips64orion | mips64orionel \
-	| mips64vr4100 | mips64vr4100el \
-	| mips64vr4300 | mips64vr4300el \
-	| mips64vr5000 | mips64vr5000el \
-	| mipsisa32 | mipsisa32el \
-	| mipsisa32r2 | mipsisa32r2el \
-	| mipsisa64 | mipsisa64el \
-	| mipsisa64r2 | mipsisa64r2el \
-	| mipsisa64sb1 | mipsisa64sb1el \
-	| mipsisa64sr71k | mipsisa64sr71kel \
-	| mipstx39 | mipstx39el \
-	| mn10200 | mn10300 \
-	| msp430 \
-	| ns16k | ns32k \
-	| openrisc | or32 \
-	| pdp10 | pdp11 | pj | pjl \
-	| powerpc | powerpc64 | powerpc64le | powerpcle | ppcbe \
-	| pyramid \
-	| sh | sh[1234] | sh[23]e | sh[34]eb | shbe | shle | sh[1234]le | sh3ele \
-	| sh64 | sh64le \
-	| sparc | sparc64 | sparc86x | sparclet | sparclite | sparcv9 | sparcv9b \
-	| strongarm \
-	| tahoe | thumb | tic4x | tic80 | tron \
-	| v850 | v850e \
-	| we32k \
-	| x86 | xscale | xstormy16 | xtensa \
-	| z8k)
-		basic_machine=$basic_machine-unknown
-		;;
-	m6811 | m68hc11 | m6812 | m68hc12)
-		# Motorola 68HC11/12.
-		basic_machine=$basic_machine-unknown
-		os=-none
-		;;
-	m88110 | m680[12346]0 | m683?2 | m68360 | m5200 | v70 | w65 | z8k)
-		;;
-
-	# We use `pc' rather than `unknown'
-	# because (1) that's what they normally are, and
-	# (2) the word "unknown" tends to confuse beginning users.
-	i*86 | x86_64)
-	  basic_machine=$basic_machine-pc
-	  ;;
-	# Object if more than one company name word.
-	*-*-*)
-		echo Invalid configuration \`$1\': machine \`$basic_machine\' not recognized 1>&2
-		exit 1
-		;;
-	# Recognize the basic CPU types with company name.
-	580-* \
-	| a29k-* \
-	| alpha-* | alphaev[4-8]-* | alphaev56-* | alphaev6[78]-* \
-	| alpha64-* | alpha64ev[4-8]-* | alpha64ev56-* | alpha64ev6[78]-* \
-	| alphapca5[67]-* | alpha64pca5[67]-* | arc-* \
-	| arm-*  | armbe-* | armle-* | armeb-* | armv*-* \
-	| avr-* \
-	| bs2000-* \
-	| c[123]* | c30-* | [cjt]90-* | c4x-* | c54x-* | c55x-* | c6x-* \
-	| clipper-* | cydra-* \
-	| d10v-* | d30v-* | dlx-* \
-	| elxsi-* \
-	| f30[01]-* | f700-* | fr30-* | frv-* | fx80-* \
-	| h8300-* | h8500-* \
-	| hppa-* | hppa1.[01]-* | hppa2.0-* | hppa2.0[nw]-* | hppa64-* \
-	| i*86-* | i860-* | i960-* | ia64-* \
-	| ip2k-* | iq2000-* \
-	| m32r-* \
-	| m68000-* | m680[012346]0-* | m68360-* | m683?2-* | m68k-* \
-	| m88110-* | m88k-* | mcore-* \
-	| mips-* | mipsbe-* | mipseb-* | mipsel-* | mipsle-* \
-	| mips16-* \
-	| mips64-* | mips64el-* \
-	| mips64vr-* | mips64vrel-* \
-	| mips64orion-* | mips64orionel-* \
-	| mips64vr4100-* | mips64vr4100el-* \
-	| mips64vr4300-* | mips64vr4300el-* \
-	| mips64vr5000-* | mips64vr5000el-* \
-	| mipsisa32-* | mipsisa32el-* \
-	| mipsisa32r2-* | mipsisa32r2el-* \
-	| mipsisa64-* | mipsisa64el-* \
-	| mipsisa64r2-* | mipsisa64r2el-* \
-	| mipsisa64sb1-* | mipsisa64sb1el-* \
-	| mipsisa64sr71k-* | mipsisa64sr71kel-* \
-	| mipstx39-* | mipstx39el-* \
-	| msp430-* \
-	| none-* | np1-* | nv1-* | ns16k-* | ns32k-* \
-	| orion-* \
-	| pdp10-* | pdp11-* | pj-* | pjl-* | pn-* | power-* \
-	| powerpc-* | powerpc64-* | powerpc64le-* | powerpcle-* | ppcbe-* \
-	| pyramid-* \
-	| romp-* | rs6000-* \
-	| sh-* | sh[1234]-* | sh[23]e-* | sh[34]eb-* | shbe-* \
-	| shle-* | sh[1234]le-* | sh3ele-* | sh64-* | sh64le-* \
-	| sparc-* | sparc64-* | sparc86x-* | sparclet-* | sparclite-* \
-	| sparcv9-* | sparcv9b-* | strongarm-* | sv1-* | sx?-* \
-	| tahoe-* | thumb-* \
-	| tic30-* | tic4x-* | tic54x-* | tic55x-* | tic6x-* | tic80-* \
-	| tron-* \
-	| v850-* | v850e-* | vax-* \
-	| we32k-* \
-	| x86-* | x86_64-* | xps100-* | xscale-* | xstormy16-* \
-	| xtensa-* \
-	| ymp-* \
-	| z8k-*)
-		;;
-	# Recognize the various machine names and aliases which stand
-	# for a CPU type and a company and sometimes even an OS.
-	386bsd)
-		basic_machine=i386-unknown
-		os=-bsd
-		;;
-	3b1 | 7300 | 7300-att | att-7300 | pc7300 | safari | unixpc)
-		basic_machine=m68000-att
-		;;
-	3b*)
-		basic_machine=we32k-att
-		;;
-	a29khif)
-		basic_machine=a29k-amd
-		os=-udi
-		;;
-	adobe68k)
-		basic_machine=m68010-adobe
-		os=-scout
-		;;
-	alliant | fx80)
-		basic_machine=fx80-alliant
-		;;
-	altos | altos3068)
-		basic_machine=m68k-altos
-		;;
-	am29k)
-		basic_machine=a29k-none
-		os=-bsd
-		;;
-	amd64)
-		basic_machine=x86_64-pc
-		;;
-	amd64-*)
-		basic_machine=x86_64-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	amdahl)
-		basic_machine=580-amdahl
-		os=-sysv
-		;;
-	amiga | amiga-*)
-		basic_machine=m68k-unknown
-		;;
-	amigaos | amigados)
-		basic_machine=m68k-unknown
-		os=-amigaos
-		;;
-	amigaunix | amix)
-		basic_machine=m68k-unknown
-		os=-sysv4
-		;;
-	apollo68)
-		basic_machine=m68k-apollo
-		os=-sysv
-		;;
-	apollo68bsd)
-		basic_machine=m68k-apollo
-		os=-bsd
-		;;
-	aux)
-		basic_machine=m68k-apple
-		os=-aux
-		;;
-	balance)
-		basic_machine=ns32k-sequent
-		os=-dynix
-		;;
-	c90)
-		basic_machine=c90-cray
-		os=-unicos
-		;;
-	convex-c1)
-		basic_machine=c1-convex
-		os=-bsd
-		;;
-	convex-c2)
-		basic_machine=c2-convex
-		os=-bsd
-		;;
-	convex-c32)
-		basic_machine=c32-convex
-		os=-bsd
-		;;
-	convex-c34)
-		basic_machine=c34-convex
-		os=-bsd
-		;;
-	convex-c38)
-		basic_machine=c38-convex
-		os=-bsd
-		;;
-	cray | j90)
-		basic_machine=j90-cray
-		os=-unicos
-		;;
-	crds | unos)
-		basic_machine=m68k-crds
-		;;
-	cris | cris-* | etrax*)
-		basic_machine=cris-axis
-		;;
-	da30 | da30-*)
-		basic_machine=m68k-da30
-		;;
-	decstation | decstation-3100 | pmax | pmax-* | pmin | dec3100 | decstatn)
-		basic_machine=mips-dec
-		;;
-	decsystem10* | dec10*)
-		basic_machine=pdp10-dec
-		os=-tops10
-		;;
-	decsystem20* | dec20*)
-		basic_machine=pdp10-dec
-		os=-tops20
-		;;
-	delta | 3300 | motorola-3300 | motorola-delta \
-	      | 3300-motorola | delta-motorola)
-		basic_machine=m68k-motorola
-		;;
-	delta88)
-		basic_machine=m88k-motorola
-		os=-sysv3
-		;;
-	dpx20 | dpx20-*)
-		basic_machine=rs6000-bull
-		os=-bosx
-		;;
-	dpx2* | dpx2*-bull)
-		basic_machine=m68k-bull
-		os=-sysv3
-		;;
-	ebmon29k)
-		basic_machine=a29k-amd
-		os=-ebmon
-		;;
-	elxsi)
-		basic_machine=elxsi-elxsi
-		os=-bsd
-		;;
-	encore | umax | mmax)
-		basic_machine=ns32k-encore
-		;;
-	es1800 | OSE68k | ose68k | ose | OSE)
-		basic_machine=m68k-ericsson
-		os=-ose
-		;;
-	fx2800)
-		basic_machine=i860-alliant
-		;;
-	genix)
-		basic_machine=ns32k-ns
-		;;
-	gmicro)
-		basic_machine=tron-gmicro
-		os=-sysv
-		;;
-	go32)
-		basic_machine=i386-pc
-		os=-go32
-		;;
-	h3050r* | hiux*)
-		basic_machine=hppa1.1-hitachi
-		os=-hiuxwe2
-		;;
-	h8300hms)
-		basic_machine=h8300-hitachi
-		os=-hms
-		;;
-	h8300xray)
-		basic_machine=h8300-hitachi
-		os=-xray
-		;;
-	h8500hms)
-		basic_machine=h8500-hitachi
-		os=-hms
-		;;
-	harris)
-		basic_machine=m88k-harris
-		os=-sysv3
-		;;
-	hp300-*)
-		basic_machine=m68k-hp
-		;;
-	hp300bsd)
-		basic_machine=m68k-hp
-		os=-bsd
-		;;
-	hp300hpux)
-		basic_machine=m68k-hp
-		os=-hpux
-		;;
-	hp3k9[0-9][0-9] | hp9[0-9][0-9])
-		basic_machine=hppa1.0-hp
-		;;
-	hp9k2[0-9][0-9] | hp9k31[0-9])
-		basic_machine=m68000-hp
-		;;
-	hp9k3[2-9][0-9])
-		basic_machine=m68k-hp
-		;;
-	hp9k6[0-9][0-9] | hp6[0-9][0-9])
-		basic_machine=hppa1.0-hp
-		;;
-	hp9k7[0-79][0-9] | hp7[0-79][0-9])
-		basic_machine=hppa1.1-hp
-		;;
-	hp9k78[0-9] | hp78[0-9])
-		# FIXME: really hppa2.0-hp
-		basic_machine=hppa1.1-hp
-		;;
-	hp9k8[67]1 | hp8[67]1 | hp9k80[24] | hp80[24] | hp9k8[78]9 | hp8[78]9 | hp9k893 | hp893)
-		# FIXME: really hppa2.0-hp
-		basic_machine=hppa1.1-hp
-		;;
-	hp9k8[0-9][13679] | hp8[0-9][13679])
-		basic_machine=hppa1.1-hp
-		;;
-	hp9k8[0-9][0-9] | hp8[0-9][0-9])
-		basic_machine=hppa1.0-hp
-		;;
-	hppa-next)
-		os=-nextstep3
-		;;
-	hppaosf)
-		basic_machine=hppa1.1-hp
-		os=-osf
-		;;
-	hppro)
-		basic_machine=hppa1.1-hp
-		os=-proelf
-		;;
-	i370-ibm* | ibm*)
-		basic_machine=i370-ibm
-		;;
-# I'm not sure what "Sysv32" means.  Should this be sysv3.2?
-	i*86v32)
-		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
-		os=-sysv32
-		;;
-	i*86v4*)
-		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
-		os=-sysv4
-		;;
-	i*86v)
-		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
-		os=-sysv
-		;;
-	i*86sol2)
-		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
-		os=-solaris2
-		;;
-	i386mach)
-		basic_machine=i386-mach
-		os=-mach
-		;;
-	i386-vsta | vsta)
-		basic_machine=i386-unknown
-		os=-vsta
-		;;
-	iris | iris4d)
-		basic_machine=mips-sgi
-		case $os in
-		    -irix*)
-			;;
-		    *)
-			os=-irix4
-			;;
-		esac
-		;;
-	isi68 | isi)
-		basic_machine=m68k-isi
-		os=-sysv
-		;;
-	m88k-omron*)
-		basic_machine=m88k-omron
-		;;
-	magnum | m3230)
-		basic_machine=mips-mips
-		os=-sysv
-		;;
-	merlin)
-		basic_machine=ns32k-utek
-		os=-sysv
-		;;
-	mingw32)
-		basic_machine=i386-pc
-		os=-mingw32
-		;;
-	miniframe)
-		basic_machine=m68000-convergent
-		;;
-	*mint | -mint[0-9]* | *MiNT | *MiNT[0-9]*)
-		basic_machine=m68k-atari
-		os=-mint
-		;;
-	mips3*-*)
-		basic_machine=`echo $basic_machine | sed -e 's/mips3/mips64/'`
-		;;
-	mips3*)
-		basic_machine=`echo $basic_machine | sed -e 's/mips3/mips64/'`-unknown
-		;;
-	mmix*)
-		basic_machine=mmix-knuth
-		os=-mmixware
-		;;
-	monitor)
-		basic_machine=m68k-rom68k
-		os=-coff
-		;;
-	morphos)
-		basic_machine=powerpc-unknown
-		os=-morphos
-		;;
-	msdos)
-		basic_machine=i386-pc
-		os=-msdos
-		;;
-	mvs)
-		basic_machine=i370-ibm
-		os=-mvs
-		;;
-	ncr3000)
-		basic_machine=i486-ncr
-		os=-sysv4
-		;;
-	netbsd386)
-		basic_machine=i386-unknown
-		os=-netbsd
-		;;
-	netwinder)
-		basic_machine=armv4l-rebel
-		os=-linux
-		;;
-	news | news700 | news800 | news900)
-		basic_machine=m68k-sony
-		os=-newsos
-		;;
-	news1000)
-		basic_machine=m68030-sony
-		os=-newsos
-		;;
-	news-3600 | risc-news)
-		basic_machine=mips-sony
-		os=-newsos
-		;;
-	necv70)
-		basic_machine=v70-nec
-		os=-sysv
-		;;
-	next | m*-next )
-		basic_machine=m68k-next
-		case $os in
-		    -nextstep* )
-			;;
-		    -ns2*)
-		      os=-nextstep2
-			;;
-		    *)
-		      os=-nextstep3
-			;;
-		esac
-		;;
-	nh3000)
-		basic_machine=m68k-harris
-		os=-cxux
-		;;
-	nh[45]000)
-		basic_machine=m88k-harris
-		os=-cxux
-		;;
-	nindy960)
-		basic_machine=i960-intel
-		os=-nindy
-		;;
-	mon960)
-		basic_machine=i960-intel
-		os=-mon960
-		;;
-	nonstopux)
-		basic_machine=mips-compaq
-		os=-nonstopux
-		;;
-	np1)
-		basic_machine=np1-gould
-		;;
-	nv1)
-		basic_machine=nv1-cray
-		os=-unicosmp
-		;;
-	nsr-tandem)
-		basic_machine=nsr-tandem
-		;;
-	op50n-* | op60c-*)
-		basic_machine=hppa1.1-oki
-		os=-proelf
-		;;
-	or32 | or32-*)
-		basic_machine=or32-unknown
-		os=-coff
-		;;
-	os400)
-		basic_machine=powerpc-ibm
-		os=-os400
-		;;
-	OSE68000 | ose68000)
-		basic_machine=m68000-ericsson
-		os=-ose
-		;;
-	os68k)
-		basic_machine=m68k-none
-		os=-os68k
-		;;
-	pa-hitachi)
-		basic_machine=hppa1.1-hitachi
-		os=-hiuxwe2
-		;;
-	paragon)
-		basic_machine=i860-intel
-		os=-osf
-		;;
-	pbd)
-		basic_machine=sparc-tti
-		;;
-	pbb)
-		basic_machine=m68k-tti
-		;;
-	pc532 | pc532-*)
-		basic_machine=ns32k-pc532
-		;;
-	pentium | p5 | k5 | k6 | nexgen | viac3)
-		basic_machine=i586-pc
-		;;
-	pentiumpro | p6 | 6x86 | athlon | athlon_*)
-		basic_machine=i686-pc
-		;;
-	pentiumii | pentium2 | pentiumiii | pentium3)
-		basic_machine=i686-pc
-		;;
-	pentium4)
-		basic_machine=i786-pc
-		;;
-	pentium-* | p5-* | k5-* | k6-* | nexgen-* | viac3-*)
-		basic_machine=i586-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	pentiumpro-* | p6-* | 6x86-* | athlon-*)
-		basic_machine=i686-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	pentiumii-* | pentium2-* | pentiumiii-* | pentium3-*)
-		basic_machine=i686-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	pentium4-*)
-		basic_machine=i786-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	pn)
-		basic_machine=pn-gould
-		;;
-	power)	basic_machine=power-ibm
-		;;
-	ppc)	basic_machine=powerpc-unknown
-		;;
-	ppc-*)	basic_machine=powerpc-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	ppcle | powerpclittle | ppc-le | powerpc-little)
-		basic_machine=powerpcle-unknown
-		;;
-	ppcle-* | powerpclittle-*)
-		basic_machine=powerpcle-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	ppc64)	basic_machine=powerpc64-unknown
-		;;
-	ppc64-*) basic_machine=powerpc64-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	ppc64le | powerpc64little | ppc64-le | powerpc64-little)
-		basic_machine=powerpc64le-unknown
-		;;
-	ppc64le-* | powerpc64little-*)
-		basic_machine=powerpc64le-`echo $basic_machine | sed 's/^[^-]*-//'`
-		;;
-	ps2)
-		basic_machine=i386-ibm
-		;;
-	pw32)
-		basic_machine=i586-unknown
-		os=-pw32
-		;;
-	rom68k)
-		basic_machine=m68k-rom68k
-		os=-coff
-		;;
-	rm[46]00)
-		basic_machine=mips-siemens
-		;;
-	rtpc | rtpc-*)
-		basic_machine=romp-ibm
-		;;
-	s390 | s390-*)
-		basic_machine=s390-ibm
-		;;
-	s390x | s390x-*)
-		basic_machine=s390x-ibm
-		;;
-	sa29200)
-		basic_machine=a29k-amd
-		os=-udi
-		;;
-	sb1)
-		basic_machine=mipsisa64sb1-unknown
-		;;
-	sb1el)
-		basic_machine=mipsisa64sb1el-unknown
-		;;
-	sei)
-		basic_machine=mips-sei
-		os=-seiux
-		;;
-	sequent)
-		basic_machine=i386-sequent
-		;;
-	sh)
-		basic_machine=sh-hitachi
-		os=-hms
-		;;
-	sh64)
-		basic_machine=sh64-unknown
-		;;
-	sparclite-wrs | simso-wrs)
-		basic_machine=sparclite-wrs
-		os=-vxworks
-		;;
-	sps7)
-		basic_machine=m68k-bull
-		os=-sysv2
-		;;
-	spur)
-		basic_machine=spur-unknown
-		;;
-	st2000)
-		basic_machine=m68k-tandem
-		;;
-	stratus)
-		basic_machine=i860-stratus
-		os=-sysv4
-		;;
-	sun2)
-		basic_machine=m68000-sun
-		;;
-	sun2os3)
-		basic_machine=m68000-sun
-		os=-sunos3
-		;;
-	sun2os4)
-		basic_machine=m68000-sun
-		os=-sunos4
-		;;
-	sun3os3)
-		basic_machine=m68k-sun
-		os=-sunos3
-		;;
-	sun3os4)
-		basic_machine=m68k-sun
-		os=-sunos4
-		;;
-	sun4os3)
-		basic_machine=sparc-sun
-		os=-sunos3
-		;;
-	sun4os4)
-		basic_machine=sparc-sun
-		os=-sunos4
-		;;
-	sun4sol2)
-		basic_machine=sparc-sun
-		os=-solaris2
-		;;
-	sun3 | sun3-*)
-		basic_machine=m68k-sun
-		;;
-	sun4)
-		basic_machine=sparc-sun
-		;;
-	sun386 | sun386i | roadrunner)
-		basic_machine=i386-sun
-		;;
-	sv1)
-		basic_machine=sv1-cray
-		os=-unicos
-		;;
-	symmetry)
-		basic_machine=i386-sequent
-		os=-dynix
-		;;
-	t3e)
-		basic_machine=alphaev5-cray
-		os=-unicos
-		;;
-	t90)
-		basic_machine=t90-cray
-		os=-unicos
-		;;
-	tic54x | c54x*)
-		basic_machine=tic54x-unknown
-		os=-coff
-		;;
-	tic55x | c55x*)
-		basic_machine=tic55x-unknown
-		os=-coff
-		;;
-	tic6x | c6x*)
-		basic_machine=tic6x-unknown
-		os=-coff
-		;;
-	tx39)
-		basic_machine=mipstx39-unknown
-		;;
-	tx39el)
-		basic_machine=mipstx39el-unknown
-		;;
-	toad1)
-		basic_machine=pdp10-xkl
-		os=-tops20
-		;;
-	tower | tower-32)
-		basic_machine=m68k-ncr
-		;;
-	tpf)
-		basic_machine=s390x-ibm
-		os=-tpf
-		;;
-	udi29k)
-		basic_machine=a29k-amd
-		os=-udi
-		;;
-	ultra3)
-		basic_machine=a29k-nyu
-		os=-sym1
-		;;
-	v810 | necv810)
-		basic_machine=v810-nec
-		os=-none
-		;;
-	vaxv)
-		basic_machine=vax-dec
-		os=-sysv
-		;;
-	vms)
-		basic_machine=vax-dec
-		os=-vms
-		;;
-	vpp*|vx|vx-*)
-		basic_machine=f301-fujitsu
-		;;
-	vxworks960)
-		basic_machine=i960-wrs
-		os=-vxworks
-		;;
-	vxworks68)
-		basic_machine=m68k-wrs
-		os=-vxworks
-		;;
-	vxworks29k)
-		basic_machine=a29k-wrs
-		os=-vxworks
-		;;
-	w65*)
-		basic_machine=w65-wdc
-		os=-none
-		;;
-	w89k-*)
-		basic_machine=hppa1.1-winbond
-		os=-proelf
-		;;
-	xps | xps100)
-		basic_machine=xps100-honeywell
-		;;
-	ymp)
-		basic_machine=ymp-cray
-		os=-unicos
-		;;
-	z8k-*-coff)
-		basic_machine=z8k-unknown
-		os=-sim
-		;;
-	none)
-		basic_machine=none-none
-		os=-none
-		;;
-
-# Here we handle the default manufacturer of certain CPU types.  It is in
-# some cases the only manufacturer, in others, it is the most popular.
-	w89k)
-		basic_machine=hppa1.1-winbond
-		;;
-	op50n)
-		basic_machine=hppa1.1-oki
-		;;
-	op60c)
-		basic_machine=hppa1.1-oki
-		;;
-	romp)
-		basic_machine=romp-ibm
-		;;
-	rs6000)
-		basic_machine=rs6000-ibm
-		;;
-	vax)
-		basic_machine=vax-dec
-		;;
-	pdp10)
-		# there are many clones, so DEC is not a safe bet
-		basic_machine=pdp10-unknown
-		;;
-	pdp11)
-		basic_machine=pdp11-dec
-		;;
-	we32k)
-		basic_machine=we32k-att
-		;;
-	sh3 | sh4 | sh[34]eb | sh[1234]le | sh[23]ele)
-		basic_machine=sh-unknown
-		;;
-	sh64)
-		basic_machine=sh64-unknown
-		;;
-	sparc | sparcv9 | sparcv9b)
-		basic_machine=sparc-sun
-		;;
-	cydra)
-		basic_machine=cydra-cydrome
-		;;
-	orion)
-		basic_machine=orion-highlevel
-		;;
-	orion105)
-		basic_machine=clipper-highlevel
-		;;
-	mac | mpw | mac-mpw)
-		basic_machine=m68k-apple
-		;;
-	pmac | pmac-mpw)
-		basic_machine=powerpc-apple
-		;;
-	*-unknown)
-		# Make sure to match an already-canonicalized machine name.
-		;;
-	*)
-		echo Invalid configuration \`$1\': machine \`$basic_machine\' not recognized 1>&2
-		exit 1
-		;;
-esac
-
-# Here we canonicalize certain aliases for manufacturers.
-case $basic_machine in
-	*-digital*)
-		basic_machine=`echo $basic_machine | sed 's/digital.*/dec/'`
-		;;
-	*-commodore*)
-		basic_machine=`echo $basic_machine | sed 's/commodore.*/cbm/'`
-		;;
-	*)
-		;;
-esac
-
-# Decode manufacturer-specific aliases for certain operating systems.
-
-if [ x"$os" != x"" ]
-then
-case $os in
-        # First match some system type aliases
-        # that might get confused with valid system types.
-	# -solaris* is a basic system type, with this one exception.
-	-solaris1 | -solaris1.*)
-		os=`echo $os | sed -e 's|solaris1|sunos4|'`
-		;;
-	-solaris)
-		os=-solaris2
-		;;
-	-svr4*)
-		os=-sysv4
-		;;
-	-unixware*)
-		os=-sysv4.2uw
-		;;
-	-gnu/linux*)
-		os=`echo $os | sed -e 's|gnu/linux|linux|'`
-		;;
-	# First accept the basic system types.
-	# The portable systems comes first.
-	# Each alternative MUST END IN A *, to match a version number.
-	# -sysv* is not here because it comes later, after sysvr4.
-	-gnu* | -bsd* | -mach* | -minix* | -genix* | -ultrix* | -irix* \
-	      | -*vms* | -sco* | -esix* | -isc* | -aix* | -sunos | -sunos[34]*\
-	      | -hpux* | -unos* | -osf* | -luna* | -dgux* | -solaris* | -sym* \
-	      | -amigaos* | -amigados* | -msdos* | -newsos* | -unicos* | -aof* \
-	      | -aos* \
-	      | -nindy* | -vxsim* | -vxworks* | -ebmon* | -hms* | -mvs* \
-	      | -clix* | -riscos* | -uniplus* | -iris* | -rtu* | -xenix* \
-	      | -hiux* | -386bsd* | -knetbsd* | -netbsd* | -openbsd* | -kfreebsd* | -freebsd* | -riscix* \
-	      | -lynxos* | -bosx* | -nextstep* | -cxux* | -aout* | -elf* | -oabi* \
-	      | -ptx* | -coff* | -ecoff* | -winnt* | -domain* | -vsta* \
-	      | -udi* | -eabi* | -lites* | -ieee* | -go32* | -aux* \
-	      | -chorusos* | -chorusrdb* \
-	      | -cygwin* | -pe* | -psos* | -moss* | -proelf* | -rtems* \
-	      | -mingw32* | -linux-gnu* | -linux-uclibc* | -uxpv* | -beos* | -mpeix* | -udk* \
-	      | -interix* | -uwin* | -mks* | -rhapsody* | -darwin* | -opened* \
-	      | -openstep* | -oskit* | -conix* | -pw32* | -nonstopux* \
-	      | -storm-chaos* | -tops10* | -tenex* | -tops20* | -its* \
-	      | -os2* | -vos* | -palmos* | -uclinux* | -nucleus* \
-	      | -morphos* | -superux* | -rtmk* | -rtmk-nova* | -windiss* \
-	      | -powermax* | -dnix* | -nx6 | -nx7 | -sei* | -dragonfly*)
-	# Remember, each alternative MUST END IN *, to match a version number.
-		;;
-	-qnx*)
-		case $basic_machine in
-		    x86-* | i*86-*)
-			;;
-		    *)
-			os=-nto$os
-			;;
-		esac
-		;;
-	-nto-qnx*)
-		;;
-	-nto*)
-		os=`echo $os | sed -e 's|nto|nto-qnx|'`
-		;;
-	-sim | -es1800* | -hms* | -xray | -os68k* | -none* | -v88r* \
-	      | -windows* | -osx | -abug | -netware* | -os9* | -beos* \
-	      | -macos* | -mpw* | -magic* | -mmixware* | -mon960* | -lnews*)
-		;;
-	-mac*)
-		os=`echo $os | sed -e 's|mac|macos|'`
-		;;
-	-linux-dietlibc)
-		os=-linux-dietlibc
-		;;
-	-linux*)
-		os=-linux
-		;;
-	-sunos5*)
-		os=`echo $os | sed -e 's|sunos5|solaris2|'`
-		;;
-	-sunos6*)
-		os=`echo $os | sed -e 's|sunos6|solaris3|'`
-		;;
-	-opened*)
-		os=-openedition
-		;;
-        -os400*)
-		os=-os400
-		;;
-	-wince*)
-		os=-wince
-		;;
-	-osfrose*)
-		os=-osfrose
-		;;
-	-osf*)
-		os=-osf
-		;;
-	-utek*)
-		os=-bsd
-		;;
-	-dynix*)
-		os=-bsd
-		;;
-	-acis*)
-		os=-aos
-		;;
-	-atheos*)
-		os=-atheos
-		;;
-	-syllable*)
-		os=-syllable
-		;;
-	-386bsd)
-		os=-bsd
-		;;
-	-ctix* | -uts*)
-		os=-sysv
-		;;
-	-nova*)
-		os=-rtmk-nova
-		;;
-	-ns2 )
-		os=-nextstep2
-		;;
-	-nsk*)
-		os=-nsk
-		;;
-	# Preserve the version number of sinix5.
-	-sinix5.*)
-		os=`echo $os | sed -e 's|sinix|sysv|'`
-		;;
-	-sinix*)
-		os=-sysv4
-		;;
-        -tpf*)
-		os=-tpf
-		;;
-	-triton*)
-		os=-sysv3
-		;;
-	-oss*)
-		os=-sysv3
-		;;
-	-svr4)
-		os=-sysv4
-		;;
-	-svr3)
-		os=-sysv3
-		;;
-	-sysvr4)
-		os=-sysv4
-		;;
-	# This must come after -sysvr4.
-	-sysv*)
-		;;
-	-ose*)
-		os=-ose
-		;;
-	-es1800*)
-		os=-ose
-		;;
-	-xenix)
-		os=-xenix
-		;;
-	-*mint | -mint[0-9]* | -*MiNT | -MiNT[0-9]*)
-		os=-mint
-		;;
-	-aros*)
-		os=-aros
-		;;
-	-kaos*)
-		os=-kaos
-		;;
-	-none)
-		;;
-	*)
-		# Get rid of the `-' at the beginning of $os.
-		os=`echo $os | sed 's/[^-]*-//'`
-		echo Invalid configuration \`$1\': system \`$os\' not recognized 1>&2
-		exit 1
-		;;
-esac
-else
-
-# Here we handle the default operating systems that come with various machines.
-# The value should be what the vendor currently ships out the door with their
-# machine or put another way, the most popular os provided with the machine.
-
-# Note that if you're going to try to match "-MANUFACTURER" here (say,
-# "-sun"), then you have to tell the case statement up towards the top
-# that MANUFACTURER isn't an operating system.  Otherwise, code above
-# will signal an error saying that MANUFACTURER isn't an operating
-# system, and we'll never get to this point.
-
-case $basic_machine in
-	*-acorn)
-		os=-riscix1.2
-		;;
-	arm*-rebel)
-		os=-linux
-		;;
-	arm*-semi)
-		os=-aout
-		;;
-    c4x-* | tic4x-*)
-        os=-coff
-        ;;
-	# This must come before the *-dec entry.
-	pdp10-*)
-		os=-tops20
-		;;
-	pdp11-*)
-		os=-none
-		;;
-	*-dec | vax-*)
-		os=-ultrix4.2
-		;;
-	m68*-apollo)
-		os=-domain
-		;;
-	i386-sun)
-		os=-sunos4.0.2
-		;;
-	m68000-sun)
-		os=-sunos3
-		# This also exists in the configure program, but was not the
-		# default.
-		# os=-sunos4
-		;;
-	m68*-cisco)
-		os=-aout
-		;;
-	mips*-cisco)
-		os=-elf
-		;;
-	mips*-*)
-		os=-elf
-		;;
-	or32-*)
-		os=-coff
-		;;
-	*-tti)	# must be before sparc entry or we get the wrong os.
-		os=-sysv3
-		;;
-	sparc-* | *-sun)
-		os=-sunos4.1.1
-		;;
-	*-be)
-		os=-beos
-		;;
-	*-ibm)
-		os=-aix
-		;;
-	*-wec)
-		os=-proelf
-		;;
-	*-winbond)
-		os=-proelf
-		;;
-	*-oki)
-		os=-proelf
-		;;
-	*-hp)
-		os=-hpux
-		;;
-	*-hitachi)
-		os=-hiux
-		;;
-	i860-* | *-att | *-ncr | *-altos | *-motorola | *-convergent)
-		os=-sysv
-		;;
-	*-cbm)
-		os=-amigaos
-		;;
-	*-dg)
-		os=-dgux
-		;;
-	*-dolphin)
-		os=-sysv3
-		;;
-	m68k-ccur)
-		os=-rtu
-		;;
-	m88k-omron*)
-		os=-luna
-		;;
-	*-next )
-		os=-nextstep
-		;;
-	*-sequent)
-		os=-ptx
-		;;
-	*-crds)
-		os=-unos
-		;;
-	*-ns)
-		os=-genix
-		;;
-	i370-*)
-		os=-mvs
-		;;
-	*-next)
-		os=-nextstep3
-		;;
-	*-gould)
-		os=-sysv
-		;;
-	*-highlevel)
-		os=-bsd
-		;;
-	*-encore)
-		os=-bsd
-		;;
-	*-sgi)
-		os=-irix
-		;;
-	*-siemens)
-		os=-sysv4
-		;;
-	*-masscomp)
-		os=-rtu
-		;;
-	f30[01]-fujitsu | f700-fujitsu)
-		os=-uxpv
-		;;
-	*-rom68k)
-		os=-coff
-		;;
-	*-*bug)
-		os=-coff
-		;;
-	*-apple)
-		os=-macos
-		;;
-	*-atari*)
-		os=-mint
-		;;
-	*)
-		os=-none
-		;;
-esac
-fi
-
-# Here we handle the case where we know the os, and the CPU type, but not the
-# manufacturer.  We pick the logical manufacturer.
-vendor=unknown
-case $basic_machine in
-	*-unknown)
-		case $os in
-			-riscix*)
-				vendor=acorn
-				;;
-			-sunos*)
-				vendor=sun
-				;;
-			-aix*)
-				vendor=ibm
-				;;
-			-beos*)
-				vendor=be
-				;;
-			-hpux*)
-				vendor=hp
-				;;
-			-mpeix*)
-				vendor=hp
-				;;
-			-hiux*)
-				vendor=hitachi
-				;;
-			-unos*)
-				vendor=crds
-				;;
-			-dgux*)
-				vendor=dg
-				;;
-			-luna*)
-				vendor=omron
-				;;
-			-genix*)
-				vendor=ns
-				;;
-			-mvs* | -opened*)
-				vendor=ibm
-				;;
-			-os400*)
-				vendor=ibm
-				;;
-			-ptx*)
-				vendor=sequent
-				;;
-			-tpf*)
-				vendor=ibm
-				;;
-			-vxsim* | -vxworks* | -windiss*)
-				vendor=wrs
-				;;
-			-aux*)
-				vendor=apple
-				;;
-			-hms*)
-				vendor=hitachi
-				;;
-			-mpw* | -macos*)
-				vendor=apple
-				;;
-			-*mint | -mint[0-9]* | -*MiNT | -MiNT[0-9]*)
-				vendor=atari
-				;;
-			-vos*)
-				vendor=stratus
-				;;
-		esac
-		basic_machine=`echo $basic_machine | sed "s/unknown/$vendor/"`
-		;;
-esac
-
-echo $basic_machine$os
-exit 0
-
-# Local variables:
-# eval: (add-hook 'write-file-hooks 'time-stamp)
-# time-stamp-start: "timestamp='"
-# time-stamp-format: "%:y-%02m-%02d"
-# time-stamp-end: "'"
-# End:
diff --git a/configure.ac b/configure.ac
new file mode 100644
index 0000000000..4e3f1ef0e5
--- /dev/null
+++ b/configure.ac
@@ -0,0 +1,4060 @@
+dnl Process this file with autoconf to produce a configure script.
+AC_INIT()
+{
+AC_CONFIG_AUX_DIR(tool)
+AC_CONFIG_MACRO_DIRS(tool/m4)
+
+AC_PREREQ(2.67)
+
+dnl   override AC_CHECKING
+dnl   placed here due to aclocal(1)'s
+dnl   ignoring this definition in separate files
+AC_DEFUN([AC_CHECKING],[dnl
+AC_REQUIRE([_COLORIZE_RESULT_PREPARE])dnl
+AS_MESSAGE([checking ${msg_checking}$1${msg_reset}...])])dnl
+
+AC_DISABLE_OPTION_CHECKING
+
+AC_ARG_VAR([cflags], [additional CFLAGS (ignored when CFLAGS is given)])
+AC_ARG_VAR([cppflags], [additional CPPFLAGS (ignored when CPPFLAGS is given)])
+AC_ARG_VAR([cxxflags], [additional CXXFLAGS (ignored when CXXFLAGS is given)])
+
+: "environment section" && {
+HAVE_BASERUBY=yes
+AC_ARG_WITH(baseruby,
+	AS_HELP_STRING([--with-baseruby=RUBY], [use RUBY as baseruby; RUBY is the pathname of ruby]),
+	[AS_CASE(["$withval"],
+	    [*ruby*],[BASERUBY=$withval],
+	    [no],[HAVE_BASERUBY=no],
+	    [AC_MSG_ERROR(need ruby)])
+	],
+	[
+		AC_PATH_PROG([BASERUBY], [ruby], [false])
+	])
+AS_IF([test "$HAVE_BASERUBY" = yes -a "`RUBYOPT=- $BASERUBY -e 'print 42' 2>/dev/null`" = 42], [
+    AS_IF([test "`RUBYOPT=- $BASERUBY --disable=gems -e 'print 42' 2>/dev/null`" = 42], [
+	BASERUBY="$BASERUBY --disable=gems"
+    ])
+    $BASERUBY -C "$srcdir" tool/downloader.rb -d tool -e gnu config.guess config.sub >&AS_MESSAGE_FD
+], [
+    BASERUBY="echo executable host ruby is required.  use --with-baseruby option.; false"
+    HAVE_BASERUBY=no
+])
+AC_SUBST(BASERUBY)
+AC_SUBST(HAVE_BASERUBY)
+
+: ${GIT=git}
+HAVE_GIT=yes
+AC_ARG_WITH(git,
+	AS_HELP_STRING([--without-git], [never use git]),
+	[AS_CASE([$withval],
+	    [no],  [GIT=never-use HAVE_GIT=no],
+	    [yes], [],
+	    [GIT=$withval])])
+AS_IF([test x"$HAVE_GIT" = xyes], [command -v "$GIT" > /dev/null || HAVE_GIT=no])
+AC_SUBST(GIT)
+AC_SUBST(HAVE_GIT)
+
+eval `sed -n ['s/^@%:@define RUBY_API_VERSION_\([A-Z][A-Z_0-9]*\) \([0-9][0-9]*\)/API_\1=\2/p'] $srcdir/include/ruby/version.h`
+RUBY_PROGRAM_VERSION=`sed -n 's/^@%:@define RUBY_VERSION "\(.*\)"/\1/p' $srcdir/version.h`
+MAJOR=`echo $RUBY_PROGRAM_VERSION | cut -d. -f1`
+MINOR=`echo $RUBY_PROGRAM_VERSION | cut -d. -f2`
+TEENY=`echo $RUBY_PROGRAM_VERSION | cut -d. -f3`
+for v in MAJOR MINOR TEENY; do
+    AS_IF([eval "test \"\$$v\" = ''"], [
+	AC_MSG_ERROR(could not determine $v number from version.h)
+    ])
+done
+AS_IF([test "$MAJOR.$MINOR" != "$API_MAJOR.$API_MINOR"], [
+    AC_MSG_ERROR([API version $API_MAJOR.$API_MINOR differs from program version $MAJOR.$MINOR])
+])
+AC_SUBST(MAJOR)
+AC_SUBST(MINOR)
+AC_SUBST(TEENY)
+AC_SUBST(RUBY_PROGRAM_VERSION)
+AC_SUBST(RUBY_API_VERSION, '$(MAJOR).$(MINOR)')
+RUBY_PATCHLEVEL=`sed -n 's/^#define RUBY_PATCHLEVEL //p' $srcdir/version.h`
+dnl checks for alternative programs
+AC_CANONICAL_BUILD
+RUBY_RM_RECURSIVE
+AC_ARG_WITH(gcc,
+	AS_HELP_STRING([--without-gcc], [never use gcc]),
+	[
+	AS_CASE([$withval],
+	    [no],  [: ${CC=cc}],
+	    [yes], [: ${CC=gcc}],
+	           [CC=$withval])])
+dnl If the user switches compilers, we can't believe the cache
+AS_IF([test ! -z "$ac_cv_prog_CC" -a ! -z "$CC" -a "$CC" != "$ac_cv_prog_CC"], [
+  AC_MSG_ERROR(cached CC is different -- throw away $cache_file
+(it is also a good idea to do 'make clean' before compiling))
+])
+test -z "$CC" || ac_cv_prog_CC="$CC"
+
+AS_IF([test "$program_prefix" = NONE], [
+  program_prefix=
+])
+AS_IF([test "$prefix" -ef .], [
+  AC_MSG_ERROR(--prefix cannot be the current working directory.)
+])
+RUBY_BASE_NAME=`echo ruby | sed "$program_transform_name"`
+RUBYW_BASE_NAME=`echo rubyw | sed "$program_transform_name"`
+AC_SUBST(RUBY_BASE_NAME)
+AC_SUBST(RUBYW_BASE_NAME)
+AC_SUBST(RUBY_VERSION_NAME, '${RUBY_BASE_NAME}-${ruby_version}')
+
+AC_CANONICAL_TARGET
+test x"$target_alias" = x &&
+target_os=`echo $target_os | sed 's/linux-gnu$/linux/;s/linux-gnu/linux-/'`
+ac_install_sh='' # unusable for extension libraries.
+
+AC_ARG_WITH(os-version-style,
+	AS_HELP_STRING([--with-os-version-style=TYPE],
+		       [OS version number for target and target_os [[full]]]
+		       [(full|teeny|minor+0|minor|major+0|major|none)]),
+	[os_version_style=$withval],
+	[os_version_style=full
+	    AS_CASE($target_os, [[*[0-9].*]],
+		[AS_CASE([`/usr/bin/ruby -e 'puts RUBY_PLATFORM' 2>/dev/null`],
+		    [[*-*[0-9].*.0]], [os_version_style=minor+0],
+		    [[*-*[0-9].*.*]], [os_version_style=full],
+		    [[*-*[0-9].0]  ], [os_version_style=major+0],
+		    [[*-*[0-9].*]  ], [os_version_style=minor],
+		    [[*-*[0-9]]    ], [os_version_style=major],
+		)])
+	])
+os_version_style_transform=
+AS_CASE("${os_version_style}",
+	[full|teeny], [],
+	[minor+0], [os_version_style_transform=['s/\([0-9]\.[0-9][0-9]*\)\.[0-9][.0-9]*$/\1.0/']],
+	[minor],   [os_version_style_transform=['s/\([0-9]\.[0-9][0-9]*\)\.[0-9][.0-9]*$/\1/']],
+	[major+0], [os_version_style_transform=['s/\([0-9]\)\.[0-9][.0-9]*$/\1.0/']],
+	[major],   [os_version_style_transform=['s/\([0-9]\)\.[0-9][.0-9]*$/\1/']],
+	[none],    [os_version_style_transform=['s/[0-9]\.[0-9][.0-9]*$//']],
+	[AC_MSG_ERROR(unknown --with-os-version-style: $withval)])
+AS_IF([test -z "$target_alias" -a -n "$os_version_style_transform"],
+	[
+	target=`echo ${target} | sed "$os_version_style_transform"`
+	target_os=`echo ${target_os} | sed "$os_version_style_transform"`
+	])
+
+AC_ARG_WITH(arch,
+	AS_HELP_STRING([--with-arch=ARCHS],
+		       [build an Apple/NeXT Multi Architecture Binary (MAB);
+                          ARCHS is a comma-delimited list of architectures for
+                          which to build; if this option is disabled or omitted
+			  entirely, then the package will be built only for the
+			  target platform]),
+       [target_archs="$withval"], [unset target_archs])
+
+AC_ARG_ENABLE(load-relative,
+       AS_HELP_STRING([--enable-load-relative], [resolve load paths at run time]),
+       [load_relative=$enableval])
+
+AC_ARG_PROGRAM
+
+dnl Checks for programs.
+
+cflagspat=
+test -z "$optflags" ||
+    cflagspat="$cflagspat;s|"`eval echo '"'"${optflags}"'"' | sed 's/[[][|.*]]/\\&/g;s/^ */ /;s/ *$/ /'`'| |g'
+test -z "$debugflags" ||
+    cflagspat="$cflagspat;s|"`eval echo '"'"${debugflags}"'"' | sed 's/[[][|.*]]/\\&/g;s/^ */ /;s/ *$/ /'`'| |g'
+test -z "warnflags" ||
+    cflagspat="$cflagspat;s|"`eval echo '"'"${warnflags}"'"' | sed 's/[[][|.*]]/\\&/g;s/^ */ /;s/ *$/ /'`'| |g'
+AS_IF([test -z "${CFLAGS+set}"], [
+    cflags=`echo " $cflags " | sed "$cflagspat;s/^ *//;s/ *$//"`
+    orig_cflags="$cflags"
+    cflags="$cflags "'${optflags} ${debugflags} ${warnflags}'
+])
+AS_IF([test -z "${CXXFLAGS+set}"], [
+    cxxflags=`echo " $cxxflags " | sed "$cflagspat;s/^ *//;s/ *$//"`
+    orig_cxxflags="$cxxflags"
+    cxxflags="$cxxflags "'${optflags} ${debugflags} ${warnflags}'
+])
+
+AS_CASE(["$host_os:$build_os"],
+[darwin*:darwin*], [
+    AC_CHECK_TOOLS(CC, [clang gcc cc])
+    # Following Apple deployed clang are broken
+    # clang version 1.0 (http://llvm.org/svn/llvm-project/cfe/tags/Apple/clang-23 exported)
+    # Apple clang version 2.0 (tags/Apple/clang-137) (based on LLVM 2.9svn)
+    # Apple clang version 2.1 (tags/Apple/clang-163.7.1) (based on LLVM 3.0svn)
+    AS_IF([! $CC -E -xc - <<SRC >/dev/null], [
+	@%:@if defined __APPLE_CC__ && defined __clang_major__ && __clang_major__ < 3
+	@%:@error premature clang
+	@%:@endif
+SRC
+	AC_MSG_ERROR([clang version 3.0 or later is required])
+    ])
+])
+AS_IF([test x"${build}" != x"${host}"], [
+  AC_CHECK_TOOL(CC, gcc)
+])
+
+AC_PROG_CC
+AS_CASE([$CC],
+[gcc-*], [
+    gcc_prefix=gcc- gcc_suffix=`echo "$CC" | sed 's/^gcc//'`
+    AC_PROG_CXX(g++${gcc_suffix})],
+[clang-*], [
+    gcc_prefix=clang- gcc_suffix=`echo "$CC" | sed 's/^clang//'`
+    AC_PROG_CXX(clang++${gcc_suffix})],
+[gcc_prefix= gcc_suffix=])
+
+dnl Select the appropriate C++ compiler in OS X
+AS_CASE(["$build_os"],
+  [darwin1*.*], [
+    AS_CASE(["x$CC"],
+      [xgcc-4.2|x/usr/bin/gcc-4.2], [: ${CXX=g++-4.2}],
+      [xgcc|x/usr/bin/gcc],         [: ${CXX=g++}],
+      [xcc|x/usr/bin/cc],           [: ${CXX=c++}],
+      [xicc],                       [: ${CXX=icpc}],
+      [xclang|x/usr/bin/clang],     [: ${CXX=clang++}])
+  ])
+test -z "$CXX" || ac_cv_prog_CXX="$CXX"
+
+AC_PROG_CXX
+RUBY_MINGW32
+AC_PROG_GCC_TRADITIONAL
+AC_SUBST(GCC)
+AS_CASE(["$target_os"],
+[solaris*], [AC_PATH_TOOL([LD], [ld], [/usr/ccs/bin/ld], [/usr/ccs/bin:$PATH])],
+[AC_CHECK_TOOL([LD], [ld], [ld])])
+AC_SUBST(LD)
+AS_IF([test "$GCC" = yes], [
+    linker_flag=-Wl,
+    : ${optflags=-O3}
+    gcc_major=`echo =__GNUC__ | $CC -E -xc - | sed '/^=/!d;s///'`
+    gcc_minor=`echo =__GNUC_MINOR__ | $CC -E -xc - | sed '/^=/!d;s///'`
+    test -n "$gcc_major" || gcc_major=0
+    test -n "$gcc_minor" || gcc_minor=0
+    AS_CASE(["x$CC"], [xicc], [
+      icc_version=`echo =__ICC | $CC -E -xc - | sed '/^=/!d;s///'`
+    ])
+    test -n "$icc_version" || icc_version=0
+    # RUBY_APPEND_OPTIONS(XCFLAGS, ["-include ruby/config.h" "-include ruby/missing.h"])
+], [
+    linker_flag=
+])
+
+AS_IF([test "$GCC" = yes -a "$gcc_major" -lt 3 ], [
+    AC_MSG_ERROR([too old GCC])
+])
+
+RUBY_PROG_GNU_LD
+RUBY_CPPOUTFILE
+
+: ${OUTFLAG='-o '}
+: ${COUTFLAG=${OUTFLAG}}
+: ${CSRCFLAG=''}
+AC_SUBST(OUTFLAG)
+AC_SUBST(COUTFLAG)
+AC_SUBST(CSRCFLAG)
+
+: ${MJIT_CC=$CC}
+AS_IF([test "x$cross_compiling" = xno], [
+    AC_PATH_PROG([MJIT_CC], ${MJIT_CC})
+    AS_CASE([$target_os],
+	[*mingw*], [command -v cygpath > /dev/null && MJIT_CC=`cygpath -ma $MJIT_CC`])
+    shift 2
+    MJIT_CC="$MJIT_CC${1+ }$*"
+])
+
+AS_CASE(["$build_os"],
+  [darwin1*.*], [
+    # Xcode linker warns for deprecated architecture and wrongly
+    # installed TBD files.
+    CC_WRAPPER=""
+    echo 'int main(void) {return 0;}' > conftest.c
+    AS_IF([$CC -framework Foundation -o conftest conftest.c 2>&1 |
+	   grep '^ld: warning: text-based stub file' >/dev/null], [
+	CC_WRAPPER=`cd -P "$srcdir/tool" && pwd`/darwin-cc
+	CC="$CC_WRAPPER $CC"
+    ])
+    rm -fr conftest*
+  ])
+
+cc_version=
+for option in --version -v -V -qversion; do
+    cc_version_message=`$CC $option 2>&1`
+    cc_version_status=$?
+    AS_CASE($cc_version_status, [0], [:], [continue])
+    AS_CASE($cc_version_message, [*Warning*], [continue])
+    cc_version='$(CC) '$option
+    break
+done
+AC_SUBST(CC_VERSION, $cc_version)
+AC_SUBST(CC_VERSION_MESSAGE, $cc_version_message)
+
+: ${DLDFLAGS="$LDFLAGS"}
+
+RUBY_UNIVERSAL_ARCH
+AS_IF([test "$target_cpu" != "$host_cpu" -a "$GCC" = yes -a "$cross_compiling" = no -a "$universal_binary" = no], [
+    RUBY_DEFAULT_ARCH("$target_cpu")
+])
+
+AS_CASE(["$target_os"], [darwin*], [
+if libtool 2>&1 | grep no_warning_for_no_symbols > /dev/null; then
+  ac_cv_prog_ac_ct_RANLIB=:
+  ac_cv_prog_ac_ct_AR='libtool -static'
+  rb_cv_arflags='-no_warning_for_no_symbols -o'
+fi
+])
+AC_CHECK_TOOLS(RANLIB, [${gcc_prefix}ranlib${gcc_suffix} ranlib], :)
+AC_CHECK_TOOLS(AR, [${gcc_prefix}ar${gcc_suffix} ar])
+AS_IF([test -z "$AR"], [
+  AC_CHECK_PROGS(AR, aal, ar)
+])
+AC_CACHE_CHECK([for $AR flags], [rb_cv_arflags], [
+    AS_IF([$AR rcD conftest.a > /dev/null 2>&1 && rm conftest.a],
+	[rb_cv_arflags=rcD], [rb_cv_arflags=rcu])
+])
+AC_SUBST(ARFLAGS, ["$rb_cv_arflags "])
+
+AC_CHECK_TOOL(AS, as)
+ASFLAGS=$ASFLAGS
+AC_SUBST(ASFLAGS)
+
+AS_CASE(["$target_os"],[cygwin*|mingw*], [ac_cv_prog_ac_ct_OBJCOPY=":"])
+
+# BSD's ports and MacPorts prefix GNU binutils with 'g'
+AC_CHECK_TOOLS(OBJDUMP, [objdump gobjdump])
+AC_CHECK_TOOLS(OBJCOPY, [objcopy gobjcopy])
+
+AS_CASE(["$target_os"],
+[cygwin*|mingw*], [
+    AC_CHECK_TOOL(WINDRES, windres)
+    AC_CHECK_TOOL(DLLWRAP, dllwrap)
+    target=`echo $target | sed "s/^$target_cpu-/-/"`
+    target_alias=`echo $target_alias | sed "s/^$target_cpu-/-/"`
+    target_cpu=`echo $target_cpu | sed s/i.86/i386/`
+    AS_CASE(["$target"], [-*], [ target="$target_cpu${target}"])
+    AS_CASE(["$target_alias"], [-*], [ target_alias="$target_cpu${target_alias}"])
+    AS_CASE(["$target_os"],
+    [mingw*], [
+	test "$rb_cv_msvcrt" = "" && unset rb_cv_msvcrt
+	AC_CACHE_CHECK(for mingw32 runtime DLL, rb_cv_msvcrt, [
+	AC_TRY_LINK([@%:@include <stdio.h>],
+		    [FILE* volatile f = stdin; return 0;],
+		    [rb_cv_msvcrt=`$OBJDUMP -p conftest$ac_exeext |
+				   tr A-Z a-z |
+				   sed -n '/^[[ 	]]*dll name: \(msvc.*\)\.dll$/{s//\1/p;q;}'`],
+		    [rb_cv_msvcrt=msvcrt])
+	test "$rb_cv_msvcrt" = "" && rb_cv_msvcrt=msvcrt])
+	RT_VER=`echo "$rb_cv_msvcrt" | tr -cd [0-9]`
+	test "$RT_VER" = "" && RT_VER=60
+	AC_DEFINE_UNQUOTED(RUBY_MSVCRT_VERSION, $RT_VER)
+	sysconfdir=
+    ])
+    : ${enable_shared=yes}
+    ],
+[aix*],     [AC_CHECK_TOOL(NM, nm, /usr/ccs/bin/nm, /usr/ccs/bin:$PATH)],
+[hiuxmpp*], [AC_DEFINE(__HIUX_MPP__)])    # by TOYODA Eizi <toyoda@npd.kishou.go.jp>
+AC_CHECK_TOOLS(NM, [${gcc_prefix}nm${gcc_suffix} nm])
+
+AC_PROG_LN_S
+AC_PROG_MAKE_SET
+AC_PROG_INSTALL
+AC_PROG_MKDIR_P
+AS_IF([test "x$MKDIR_P" = "x -d"], [
+  AS_IF([test x"$as_mkdir_p" != xfalse], [
+    MKDIR_P='mkdir -p'
+    echo "use 'mkdir -p' as MKDIR_P"
+  ], [
+    AC_MSG_ERROR([mkdir -p is required])
+  ])
+])
+MAKEDIRS="$MKDIR_P"
+AC_SUBST(MAKEDIRS)
+
+AC_CHECK_PROG([DTRACE], [${ac_tool_prefix}dtrace], [${ac_tool_prefix}dtrace])
+AS_IF([test "$cross_compiling:$ac_cv_prog_DTRACE" = no: -a -n "$ac_tool_prefix"], [
+    AC_CHECK_PROG([DTRACE], [dtrace], [dtrace])
+])
+
+AC_CHECK_PROGS(DOT, dot)
+AC_CHECK_PROGS(DOXYGEN, doxygen)
+
+AC_CHECK_PROG(PKG_CONFIG, pkg-config, [pkg-config], [], [],
+    [`"$as_dir/$ac_word$ac_exec_ext" --print-errors --version > /dev/null 2>&1 || echo "$as_dir/$ac_word$ac_exec_ext"`])
+
+# checks for UNIX variants that set C preprocessor variables
+AC_USE_SYSTEM_EXTENSIONS
+
+AC_SUBST(RM, ['rm -f'])
+AC_SUBST(CP, ['cp'])
+RMDIRS='$(top_srcdir)/tool/rmdirs'
+RMDIR=rmdir
+mkdir "rmdirs_$$_test" "rmdirs_$$_test/a"
+rmdir --ignore-fail-on-non-empty "rmdirs_$$_test" 2>/dev/null &&
+RMDIR='rmdir --ignore-fail-on-non-empty'
+$RMDIR -p "rmdirs_$$_test/a" 2>/dev/null &&
+{ test -d "rmdirs_$$_test" || RMDIRS="$RMDIR -p"; }
+rmdir "rmdirs_$$_test/a" "rmdirs_$$_test" 2>/dev/null
+AC_SUBST(RMDIR)
+AC_SUBST(RMDIRS)
+AC_SUBST(RMALL, ['rm -fr'])
+
+AC_MSG_CHECKING([for cd using physical directory])
+rm -fr conf$$.dir
+mkdir conf$$.dir &&
+(cd conf$$.dir && mkdir src build && cd src &&
+$as_ln_s ../build . > /dev/null 2>&1 && cd build &&
+for chdir in 'cd -P' 'PWD= cd'; do
+    /bin/sh -c "$chdir ../src && echo '$chdir' > cdcmd" 2> /dev/null && break
+done)
+AS_IF([test -f conf$$.dir/src/cdcmd], [
+    read CHDIR < conf$$.dir/src/cdcmd 2> /dev/null
+], [
+    CHDIR=cd
+])
+rm -fr conf$$.dir
+AC_MSG_RESULT([$CHDIR])
+AC_SUBST(CHDIR)
+}
+
+: "compiler section" && {
+RUBY_WERROR_FLAG([
+    AC_MSG_CHECKING([whether CFLAGS is valid])
+    AC_TRY_COMPILE([], [],
+	[AC_MSG_RESULT(yes)],
+	[
+	AC_MSG_RESULT(no)
+	AC_MSG_ERROR([something wrong with CFLAGS="$CFLAGS"])
+	]
+    )
+
+    AC_MSG_CHECKING([whether LDFLAGS is valid])
+    {
+	mkdir tmp.$$.try_link &&
+	cd tmp.$$.try_link &&
+	cp ../confdefs.h . &&
+	echo '<?xml?><plist><dict><key>CFBundleIdentifier</key><string></string></dict></plist>' > Info.plist &&
+	:
+    } || AC_MSG_ERROR([failed to make temporary directory])
+    AC_TRY_LINK([], [],
+	[AC_MSG_RESULT(yes)],
+	[
+	cd .. && rm -fr tmp.$$.try_link
+	AC_MSG_RESULT(no)
+	AC_MSG_ERROR([something wrong with LDFLAGS="$LDFLAGS"])
+	]
+    )
+    cd .. && rm -fr tmp.$$.try_link
+])
+
+: ${RPATHFLAG=''}
+rpathflag=''
+AS_IF([test x"${RPATHFLAG}" = x], [
+    AS_CASE(["$target_os"],
+	[hpux*], [AS_IF([test "$rb_cv_prog_gnu_ld" = no], [rpathflag='+b '])],
+	[aix*], [rpathflag='-blibpath:'],
+	[for rpathflag in -R "-rpath "; do
+	    AS_CASE("$rpathflag",
+		    [*" "], [AS_CASE(["${linker_flag}"],
+				     [*,], [rpathflag=`echo "$rpathflag" | tr ' ' ,`])])
+	    rpathflag="${linker_flag}${rpathflag}"
+	    RUBY_TRY_LDFLAGS([${rpathflag}.], [], [rpathflag=])
+	    AS_IF([test "x${rpathflag}" != x], [])
+        done])
+], [
+    rpathflag=`echo "$RPATHFLAG" | sed 's/%.*//'`
+])
+
+AS_CASE([$RUBY_PATCHLEVEL], [-*],
+	[RUBY_DEVEL=yes], [RUBY_DEVEL=no])
+particular_werror_flags=$RUBY_DEVEL
+AC_ARG_ENABLE(werror,
+	AS_HELP_STRING([--disable-werror],
+		       [don't make warnings into errors
+		       even if a compiler support -Werror feature
+		       [[disabled by default unless development version]]]),
+	[particular_werror_flags=$enableval])
+
+rb_cv_warnflags="$warnflags"
+AS_CASE(["$GCC:${warnflags+set}:${extra_warnflags:+set}:"],
+[yes::*|yes:*:set:], [# GCC && (!warnflags || extra_warnflags)
+    AS_IF([test $gcc_major -ge 4], [
+	extra_warnflags="$extra_warnflags -Werror=extra-tokens"
+    ])
+    AS_IF([test $gcc_major -ge 5 -a $gcc_major -le 6], [
+	extra_warnflags="$extra_warnflags -Wno-maybe-uninitialized"
+    ])
+    # ICC doesn't support -Werror=
+    AS_IF([test $icc_version -gt 0], [
+	particular_werror_flags=no
+    ])
+    for wflag in \
+		 -Werror=declaration-after-statement \
+		 -Werror=deprecated-declarations \
+		 -Werror=division-by-zero \
+		 -Werror=duplicated-cond \
+		 -Werror=implicit-function-declaration \
+		 -Werror=implicit-int \
+		 -Werror=misleading-indentation \
+		 -Werror=pointer-arith \
+		 -Werror=restrict \
+		 -Werror=shorten-64-to-32 \
+		 -Werror=write-strings \
+		 -Wimplicit-fallthrough=0 \
+		 -Wmissing-noreturn \
+		 -Wno-cast-function-type \
+		 -Wno-constant-logical-operand \
+		 -Wno-long-long \
+		 -Wno-missing-field-initializers \
+		 -Wno-overlength-strings \
+		 -Wno-packed-bitfield-compat \
+		 -Wno-parentheses-equality \
+		 -Wno-self-assign \
+		 -Wno-tautological-compare \
+		 -Wno-unused-parameter \
+		 -Wno-unused-value \
+		 -Wsuggest-attribute=format \
+		 -Wsuggest-attribute=noreturn \
+		 -Wunused-variable \
+		 -diag-disable=175,188,2259 \
+		 $extra_warnflags \
+		 ; do
+	AS_IF([test "$particular_werror_flags" != yes], [
+	    wflag=`echo x$wflag | sed 's/^x-Werror=/-W/;s/^x//'`
+	])
+	ok=no
+	RUBY_TRY_CFLAGS($wflag, [
+	    RUBY_APPEND_OPTIONS(warnflags, $wflag)
+	    ok=yes
+	])
+	AS_CASE([$ok:$wflag], [no:-Werror=*], [
+	    wflag=`echo x$wflag | sed 's/^x-Werror=/-W/'`
+	    RUBY_TRY_CFLAGS($wflag, [
+		RUBY_APPEND_OPTIONS(warnflags, $wflag)
+		particular_werror_flags=no
+	    ])
+	])
+    done
+    AS_CASE([" $warnflags "],[*" -Wno-missing-field-initializers "*], [wflag="-Wall -Wextra"],
+                             [wflag=-Wall])
+    RUBY_TRY_CFLAGS($wflag, [warnflags="$wflag${warnflags+ $warnflags}"])
+    # Disable warnflags while conftest. -Werror=* flags might make bad OS capability guess.
+    rb_cv_warnflags="$warnflags"
+    warnflags=
+])
+RUBY_TRY_CFLAGS(-Qunused-arguments, [RUBY_APPEND_OPTIONS(rb_cv_wsuppress_flags, -Qunused-arguments)])
+
+AC_ARG_WITH(compress-debug-sections,
+	AS_HELP_STRING([--with-compress-debug-sections=type],
+	    [enable debug section compression]),
+	[compress_debug_sections=$withval], [compress_debug_sections=])
+
+AS_IF([test "$GCC" = yes], [
+    # -D_FORTIFY_SOURCE
+    # When defined _FORTIFY_SOURCE, glibc enables some additional sanity
+    # argument check. The performance drop is very little and Ubuntu enables
+    # _FORTIFY_SOURCE=2 by default. So, let's support it for protecting us from
+    # a mistake of silly C extensions.
+    RUBY_TRY_CFLAGS(-D_FORTIFY_SOURCE=2, [RUBY_APPEND_OPTION(XCFLAGS, -D_FORTIFY_SOURCE=2)])
+
+    : ${MJIT_HEADER_FLAGS='-P -dD'}
+
+    # -fstack-protector
+    AS_CASE(["$target_os"],
+    [mingw*], [
+	stack_protector=no
+    ])
+    AS_IF([test -z "${stack_protector+set}"], [
+	AS_FOR(option, opt, [-fstack-protector-strong -fstack-protector], [
+	    RUBY_TRY_CFLAGS(option, [stack_protector=yes])
+	    AS_IF([test "x$stack_protector" = xyes], [
+		RUBY_TRY_LDFLAGS(option, [], [stack_protector=])
+	    ])
+	    AS_IF([test "x$stack_protector" = xyes], [stack_protector=option; break])
+	])
+    ])
+    AS_CASE(["$stack_protector"], [-*], [
+	RUBY_APPEND_OPTION(XCFLAGS, $stack_protector)
+	RUBY_APPEND_OPTION(XLDFLAGS, $stack_protector)
+	RUBY_APPEND_OPTION(LDFLAGS, $stack_protector)
+    ])
+
+    AS_CASE("${compress_debug_sections:-zlib}",
+    [none|no], [], [
+    RUBY_TRY_LDFLAGS(${linker_flag}--compress-debug-sections=${compress_debug_sections:-zlib},
+		     [compress_debug_sections=${compress_debug_sections:-zlib}],
+		     [compress_debug_sections=no])
+    ])
+    AS_IF([test "x$compress_debug_sections" != xno], [
+	RUBY_APPEND_OPTION(DLDFLAGS, ${linker_flag}--compress-debug-sections=$compress_debug_sections)
+    ])
+
+    AS_CASE(["$target_os"],[mingw*], [
+      # On  Windows  platforms,   system  provided  headers  are  VC++
+      # optimized.  That  is, C++  habits are often  contaminated into
+      # various  headers.  Most frequent  situation is  the use  of //
+      # comments.   We  bypass  ANSI   C  mode  for  them.   Otherwise
+      # extension libs cannot include those headers.
+
+      # Since math.h in some mingw64 wrongly declares frexp and modf
+      # to be pure, the variables pointed by the second arguments are
+      # considered uninitialized unexpectedly.
+      AC_CACHE_CHECK([whether frexp and modf are broken],
+	rb_cv_mingw64_broken_frexp_modf,
+	[
+	  save_CFLAGS="$CFLAGS"
+	  AS_IF([test "$particular_werror_flags" = "yes"], [
+	    CFLAGS="$CFLAGS -Werror=uninitialized"
+	  ], [
+	    CFLAGS="$CFLAGS -Werror -Wuninitialized"
+	  ])
+	  AC_TRY_COMPILE([@%:@include <math.h>
+	    int foo(double x)
+	    {
+	      int exp;
+	      frexp(x, &exp);
+	      return exp;
+	    }], [if (foo(0.0)) return 1;],
+	    [rb_cv_mingw64_broken_frexp_modf=no],
+	    [rb_cv_mingw64_broken_frexp_modf=yes])
+	  CFLAGS="$save_CFLAGS"
+	])
+      AS_IF([test "$rb_cv_mingw64_broken_frexp_modf" = yes], [
+	AC_DEFINE(RUBY_MINGW64_BROKEN_FREXP_MODF)
+      ])
+    ],
+    [cygwin*|darwin*|netbsd*], [
+      # need lgamma_r(), finite()
+    ],
+    [solaris*], [
+      # ANSI (no XCFLAGS because this is C only)
+      # Because "-std=gnu99" affects existance of functions on Solaris,
+      # "-std=gnu99" will be appended to CPPFLAGS.
+	for ansi_options in -std=gnu99; do
+	    RUBY_TRY_CFLAGS(${ansi_options}, [
+		RUBY_APPEND_OPTIONS(CPPFLAGS, ${ansi_options})
+	    ], [ansi_options=])
+	    test "x${ansi_options}" = x || break
+	done
+    ],
+    [
+      # ANSI (no XCFLAGS because this is C only)
+      rb_tmp_std_check=`echo $CC $CFLAGS $optflags $warnflags $debugflags | fgrep std= | tr -d  '\015'`
+      AS_IF([test "x$rb_tmp_std_check" = "x"],
+      [
+	for ansi_options in -std=gnu99; do
+	    RUBY_TRY_CFLAGS(${ansi_options}, [
+		RUBY_APPEND_OPTIONS(warnflags, ${ansi_options})
+		RUBY_APPEND_OPTIONS(strict_warnflags, ${ansi_options})
+	    ], [ansi_options=])
+	    test "x${ansi_options}" = x || break
+	done
+      ])
+    ])
+
+    # suppress annoying -Wstrict-overflow warnings
+    RUBY_TRY_CFLAGS(-fno-strict-overflow, [RUBY_APPEND_OPTION(XCFLAGS, -fno-strict-overflow)])
+
+    test "${debugflags+set}" || {RUBY_TRY_CFLAGS(-ggdb3, [debugflags=-ggdb3])}
+    test "${debugflags+set}" || {RUBY_TRY_CFLAGS(-ggdb, [debugflags=-ggdb])}
+    test "${debugflags+set}" || {RUBY_TRY_CFLAGS(-g3, [debugflags=-g3])}
+])
+test $ac_cv_prog_cc_g = yes && : ${debugflags=-g}
+AS_IF([test "x$RUBY_DEVEL" = xyes], [RUBY_APPEND_OPTION(XCFLAGS, -DRUBY_DEVEL=1)])
+
+AS_IF([test "$GCC" = ""], [
+    AS_CASE(["$target_os"],[aix*],[warnflags="$warnflags -qinfo=por" rb_cv_warnflags="$rb_cv_warnflags -qinfo=por"])
+])
+AS_IF([test "$GCC" = yes], [
+    AS_IF([test "$gcc_major" -ge 4], [
+	RUBY_TRY_CFLAGS(-fvisibility=hidden, [visibility_option=yes], [visibility_option=no])
+    ])
+    AC_SUBST(WERRORFLAG, "-Werror")
+    AS_IF([test "$visibility_option" = yes], [
+	RUBY_APPEND_OPTION(XCFLAGS, -fvisibility=hidden)
+	AC_DEFINE(RUBY_SYMBOL_EXPORT_BEGIN, [_Pragma("GCC visibility push(default)")])
+	AC_DEFINE(RUBY_SYMBOL_EXPORT_END,   [_Pragma("GCC visibility pop")])
+    ], [
+	RUBY_TRY_LDFLAGS([-Wl,-unexported_symbol,_Init_*], [visibility_option=ld], [visibility_option=no])
+    ])
+    test "$visibility_option" = no || OBJCOPY=:
+])
+
+AS_IF([test "$GCC" = yes], [
+    # optflags
+
+    AS_CASE(["$target_os"], [mingw*], [
+	RUBY_TRY_CFLAGS(-fno-omit-frame-pointer, [optflags="${optflags+$optflags }-fno-omit-frame-pointer"])
+	RUBY_TRY_CFLAGS(-static-libgcc, [static_libgcc=yes], [static_libgcc=no])
+	AS_IF([test "$static_libgcc" = yes], [
+	    RUBY_APPEND_OPTION(EXTLDFLAGS, -static-libgcc)
+	])
+    ])
+
+    # disable fast-math
+    for oflag in -fno-fast-math; do
+	RUBY_TRY_CFLAGS($oflag, [RUBY_APPEND_OPTION(CFLAGS, $oflag)])
+    done
+    for oflag in -fexcess-precision=standard -fp-model\ precise; do
+	RUBY_TRY_CFLAGS($oflag, [RUBY_APPEND_OPTION(XCFLAGS, $oflag)])
+    done
+])
+
+AS_CASE(["$target_cpu"], [[i[3-6]86*]], [
+    AC_CACHE_CHECK([for __sync_val_compare_and_swap], [rb_cv_gcc_compiler_cas], [
+	AC_TRY_LINK([unsigned long atomic_var;],
+	    [__sync_val_compare_and_swap(&atomic_var, 0, 1);],
+	    [rb_cv_gcc_compiler_cas=yes],
+	    [
+	    save_CFLAGS="$CFLAGS" CFLAGS="$CFLAGS -march=i486"
+	    AC_TRY_LINK([unsigned long atomic_var;],
+		[__sync_val_compare_and_swap(&atomic_var, 0, 1);],
+		[rb_cv_gcc_compiler_cas=i486],
+		[rb_cv_gcc_compiler_cas=no])
+	    CFLAGS="$save_CFLAGS"
+	    ])
+    ])
+    AS_IF([test "$rb_cv_gcc_compiler_cas" = i486], [ARCH_FLAG="-march=i486"])
+])
+
+AC_ARG_WITH(opt-dir,
+	AS_HELP_STRING([--with-opt-dir=DIR-LIST],
+		       [add optional headers and libraries directories separated by $PATH_SEPARATOR]),
+	[
+		val=`echo "$PATH_SEPARATOR$withval" | sed "s|$PATH_SEPARATOR\([[^$PATH_SEPARATOR]*]\)| -I\1/include|g;s/^ //"`
+		CPPFLAGS="$CPPFLAGS $val"
+		val=`echo "$PATH_SEPARATOR$withval" | sed "s|$PATH_SEPARATOR\([[^$PATH_SEPARATOR]*]\)| -L\1/lib${rpathflag:+ $rpathflag\\\\1/lib}|g;s/^ //"`
+		LDFLAGS="$LDFLAGS $val"
+		LDFLAGS_OPTDIR="$val"
+		OPT_DIR="$withval"
+	], [OPT_DIR=])
+
+test -z "${ac_env_CFLAGS_set}" -a -n "${cflags+set}" && eval CFLAGS="\"$cflags $ARCH_FLAG\""
+test -z "${ac_env_CXXFLAGS_set}" -a -n "${cxxflags+set}" && eval CXXFLAGS="\"$cxxflags $ARCH_FLAG\""
+}
+
+AC_CACHE_CHECK([whether compiler has statement and declarations in expressions],
+  rb_cv_have_stmt_and_decl_in_expr,
+  [AC_TRY_COMPILE([],[ __extension__ ({ int a = 0; a; }); ],
+  [rb_cv_have_stmt_and_decl_in_expr=yes],
+  [rb_cv_have_stmt_and_decl_in_expr=no])])
+AS_IF([test "$rb_cv_have_stmt_and_decl_in_expr" = yes], [
+  AC_DEFINE(HAVE_STMT_AND_DECL_IN_EXPR)
+])
+
+: "header and library section" && {
+AC_ARG_WITH(winnt-ver,
+  AS_HELP_STRING([--with-winnt-ver=0xXXXX], [target Windows NT version (default to 0x0600)]),
+  [with_winnt_ver="$withval"], [with_winnt_ver="0x0600"])
+AS_CASE(["$target_os"],
+[mingw*], [
+  RUBY_APPEND_OPTION(CPPFLAGS, -D_WIN32_WINNT=$with_winnt_ver)
+  RUBY_APPEND_OPTION(CPPFLAGS, -D__MINGW_USE_VC2005_COMPAT)
+])
+
+AS_CASE(["$target_os"],
+[freebsd*], [
+  AC_CACHE_CHECK([whether pthread should be enabled by default],
+    rb_cv_enable_pthread_default,
+    [AC_TRY_CPP([
+#include <osreldate.h>
+#if __FreeBSD_version < 502102
+#error pthread should be disabled on this platform
+#endif
+      ],
+      rb_cv_enable_pthread_default=yes,
+      rb_cv_enable_pthread_default=no)])
+  enable_pthread=$rb_cv_enable_pthread_default
+  ],
+[mingw*], [
+  enable_pthread=no
+  ],
+[
+  enable_pthread=yes
+  ])
+
+dnl Checks for libraries.
+AS_CASE(["$target_os"],[*bsd*|dragonfly*],[],[ac_cv_func_daemon=no])
+
+AS_UNSET(ORIG_LIBS)
+POSTLINK=:
+AC_SUBST(POSTLINK)
+AS_CASE(["$target_os"],
+[nextstep*], [	],
+[openstep*], [	],
+[rhapsody*], [	],
+[darwin*], [
+		ORIG_LIBS="$LIBS"
+		RUBY_PREPEND_OPTION(LIBS, -lobjc)
+		RUBY_APPEND_OPTIONS(CPPFLAGS, -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -D_DARWIN_UNLIMITED_SELECT -D_REENTRANT)
+		AC_MSG_CHECKING(whether Mac OS X 10.5 or later)
+		AC_TRY_CPP([#include <AvailabilityMacros.h>
+		    #if MAC_OS_X_VERSION_MAX_ALLOWED <= 1040
+		    #error pre OS X 10.4
+		    [!<===== pre OS X 10.4 =====>]
+		    #endif
+		    ],
+		    [macosx_10_5=yes], [macosx_10_5=no])
+		AC_MSG_RESULT($macosx_10_5)
+		AS_IF([test "${target_os@%:@darwin}" -ge 16], [
+		    ac_cv_func___syscall=no
+		    ac_cv_func_syscall=no
+		    ac_cv_header_sys_syscall_h=no
+		    ac_cv_header_syscall_h=no
+		])
+		AS_IF([test $macosx_10_5 = yes], [
+		    ac_cv_func_getcontext=no
+		    ac_cv_func_setcontext=no
+		], [
+		    AC_DEFINE(BROKEN_SETREUID, 1)
+		    AC_DEFINE(BROKEN_SETREGID, 1)
+		])
+		incs=`$CC -v -E -xc - < /dev/null 2>&1 | sed ['1,/^@%:@include </d;s/^ *//;s|[^./][^/]*/\.\./||g;/\/include$/!d;s||/lib|;/\/usr\/lib/d']`
+		for d in `$CC -print-search-dirs | sed -e '/^libraries: */!d;s///' | tr : '\012' | fgrep -v /../ | sed -n 's|^\(/.*/lib\)/$|\1|p'`; do
+		    incs=`echo "$incs" | fgrep -v "$d"`
+		done
+		for d in $incs; do
+		    test -d "$d" && RUBY_APPEND_OPTIONS(LDFLAGS, "-L$d")
+		done
+		ac_cv_type_getgroups=gid_t # getgroups() on Rosetta fills garbage
+		ac_cv_lib_crypt_crypt=no
+		ac_cv_func_fdatasync=no # Mac OS X wrongly reports it has fdatasync()
+		ac_cv_func_vfork=no
+		AS_IF([test $gcc_major -lt 4 -o \( $gcc_major -eq 4 -a $gcc_minor -lt 3 \)], [
+		    ac_cv_func___builtin_setjmp=no
+		])
+		with_setjmp_type=sigsetjmp # to hijack SIGCHLD handler
+		AC_CACHE_CHECK(for broken crypt with 8bit chars, rb_cv_broken_crypt,
+		    [AC_TRY_RUN([
+#include <stdio.h>
+#include <unistd.h>
+#include <string.h>
+
+void
+broken_crypt(const char *salt, const char *buf1, const char *buf2)
+{
+#if 0
+    printf("%.2x%.2x: %s -> %s\n", (unsigned char)salt[0], (unsigned char)salt[1],
+	   buf1+2, buf2+2);
+#endif
+}
+
+int
+main()
+{
+    int i;
+    char salt[2], buf[256], *s;
+    for (i = 0; i < 128*128; i++) {
+	salt[0] = 0x80 | (i & 0x7f);
+	salt[1] = 0x80 | (i >> 7);
+	strcpy(buf, crypt("", salt));
+	if (strcmp(buf, s = crypt("", salt))) {
+	    broken_crypt(salt, buf, s);
+	    return 1;
+	}
+    }
+    salt[0] = salt[1] = ' ';
+    strcpy(buf, crypt("", salt));
+    salt[0] = salt[1] = 0x80 | ' ';
+    if (strcmp(buf, s = crypt("", salt))) {
+	broken_crypt(salt, buf, s);
+	return 1;
+    }
+    return 0;
+}
+],
+		    rb_cv_broken_crypt=no,
+		    rb_cv_broken_crypt=yes,
+		    rb_cv_broken_crypt=yes)])
+		AS_IF([test "$rb_cv_broken_crypt" = yes], [
+		   AC_DEFINE(BROKEN_CRYPT, 1)
+		])
+		POSTLINK=""
+		AC_CHECK_PROGS(codesign, codesign)
+		AC_CHECK_PROGS(dsymutil, dsymutil)
+		AS_IF([test -n "$codesign"], [
+		    POSTLINK="{ test -z '\$(RUBY_CODESIGN)' || $codesign -s '\$(RUBY_CODESIGN)' -f \$@; }${POSTLINK:+; $POSTLINK}"
+		])
+		AS_IF([test -n "$dsymutil"], [
+		    POSTLINK="$dsymutil \$@${POSTLINK:+; $POSTLINK}"
+		])
+		AS_IF([test -n "${POSTLINK}"], [
+		    LINK_SO="$LINK_SO
+\$(POSTLINK)"
+		])
+		AC_CHECK_HEADERS(crt_externs.h, [], [], [
+		    #include <crt_externs.h>
+		])
+		],
+[hpux*], [	LIBS="-lm $LIBS"
+		ac_cv_c_inline=no],
+[solaris*], [	LIBS="-lm $LIBS"
+		ac_cv_func_vfork=no
+		AC_MSG_CHECKING(whether _XOPEN_SOURCE is already given)
+		AC_TRY_COMPILE([#include <unistd.h>
+			#ifndef _XOPEN_SOURCE
+			#error _XOPEN_SOURCE is not defined
+			#endif
+			], [],
+		        [given_xopen_source=yes], [given_xopen_source=no])
+		AC_MSG_RESULT($given_xopen_source)
+		AS_IF([test $given_xopen_source = no], [
+		  AC_MSG_CHECKING(appropriate _XOPEN_SOURCE value to define)
+		  define_xopen_source=""
+		  for tmp_xpg in 7 6 5; do
+		    AS_IF([test x"$define_xopen_source" != x], [
+		      break
+		    ])
+		    RUBY_WERROR_FLAG([AC_TRY_COMPILE([
+			#define _XOPEN_SOURCE ${tmp_xpg}00
+			#include <unistd.h>
+			#ifndef _XPG${tmp_xpg}
+			#error _XPG${tmp_xpg} should be defined by _XOPEN_SOURCE=${tmp_xpg}00
+			#endif
+			], [],
+			[define_xopen_source=${tmp_xpg}00], [])
+			])
+		  done
+		  AS_IF([test x"$define_xopen_source" = x], [
+		    define_xopen_source=no
+		  ])
+		  AC_MSG_RESULT($define_xopen_source)
+		  AS_IF([test x"$define_xopen_source" != xno], [
+		    RUBY_APPEND_OPTIONS(CPPFLAGS, -D_XOPEN_SOURCE=$define_xopen_source)
+		  ])
+		])
+		],
+[haiku*], [
+		LIBS="$LIBS" # m lib is include in root
+		],
+[cygwin*], [	ac_cv_header_langinfo_h=yes
+		RUBY_APPEND_OPTIONS(CPPFLAGS, -D_XOPEN_SOURCE -D_GNU_SOURCE)
+		AC_CHECK_FUNCS(cygwin_conv_path)
+		AC_LIBOBJ([langinfo])
+		],
+[mingw*], [	LIBS="-lshell32 -lws2_32 -liphlpapi -limagehlp -lshlwapi $LIBS"
+		ac_cv_header_a_out_h=no
+		ac_cv_header_pwd_h=no
+		ac_cv_header_utime_h=no
+		ac_cv_header_sys_ioctl_h=no
+		ac_cv_header_sys_param_h=no
+		ac_cv_header_sys_resource_h=no
+		ac_cv_header_sys_select_h=no
+		ac_cv_header_sys_time_h=no
+		ac_cv_header_sys_times_h=no
+		ac_cv_header_sys_socket_h=no
+		ac_cv_func_lstat=yes
+		ac_cv_func_times=yes
+		ac_cv_func_waitpid=yes
+		ac_cv_func_fsync=yes
+		ac_cv_func_seekdir=yes
+		ac_cv_func_telldir=yes
+		ac_cv_func_isinf=yes
+		ac_cv_func_isnan=yes
+		ac_cv_func_finite=yes
+		ac_cv_func_lchown=yes
+		ac_cv_func_link=yes
+		ac_cv_func_readlink=yes
+		ac_cv_func_symlink=yes
+		ac_cv_lib_crypt_crypt=no
+		ac_cv_func_getpgrp_void=no
+		ac_cv_func_memcmp_working=yes
+		ac_cv_lib_dl_dlopen=no
+		rb_cv_binary_elf=no
+		rb_cv_negative_time_t=no
+		ac_cv_func_fcntl=yes
+		ac_cv_func_flock=yes
+		ac_cv_func_gmtime_r=yes
+		rb_cv_large_fd_select=yes
+		ac_cv_type_struct_timeval=yes
+                ac_cv_func_clock_gettime=yes
+                ac_cv_func_clock_getres=yes
+		ac_cv_func_malloc_usable_size=no
+		ac_cv_type_off_t=yes
+		ac_cv_sizeof_off_t=8
+		AS_IF([test "$target_cpu" = x64], [
+		    ac_cv_func___builtin_setjmp=yes
+		    ac_cv_func_round=no
+		    rb_cv_fiber_coroutine=yes
+		])
+		ac_cv_func_tgamma=no
+		rb_cv_negative_time_t=yes
+		AC_CHECK_TYPE([NET_LUID], [], [],
+			      [@%:@include <winsock2.h>
+			      @%:@include <iphlpapi.h>])
+		AS_IF([test x"$ac_cv_type_NET_LUID" = xyes], [
+		    AC_DEFINE(HAVE_TYPE_NET_LUID, 1)
+		])
+		AC_CHECK_FUNCS(_gmtime64_s)
+		AC_CHECK_FUNCS(_wfreopen_s)
+		AC_LIBOBJ([langinfo])
+		],
+[bsdi*], [	LIBS="-lm $LIBS"
+		AC_DEFINE(BROKEN_SETREUID, 1)
+		AC_DEFINE(BROKEN_SETREGID, 1)
+                ac_cv_sizeof_rlim_t=8],
+[freebsd*], [	LIBS="-lm $LIBS"
+		ac_cv_func_getpeername=no
+		ac_cv_func_getsockname=no
+		ac_cv_func_shutdown=no
+		ac_cv_func_close=no
+		],
+[netbsd*], [	LIBS="-lm $LIBS"
+		],
+[dragonfly*], [	LIBS="-lm $LIBS"
+		# isinf() and isnan() are macros on DragonFly.
+		ac_cv_func_isinf=yes
+		ac_cv_func_isnan=yes
+		],
+[aix*],[	LIBS="-lm $LIBS"
+		ac_cv_func_round=no
+		ac_cv_func___builtin_setjmp=no
+		],
+[linux*],[	LIBS="-lm $LIBS"
+		# __builtin_longjmp in ppc64* Linux does not restore
+		# the TOC register (r2), which is problematic
+		# when a global exit happens from JITted .so code.
+		AS_CASE(["$target_cpu"], [powerpc64*], [
+			ac_cv_func___builtin_setjmp=no
+		])
+		# With gcc-8's -fcf-protection, MJIT's __builtin_longjmp fails.
+		AS_CASE(["$CC $CFLAGS "], [*" -fcf-protection "*], [cf_protection=yes], [cf_protection=no])
+		AS_IF([test "$cf_protection" = yes], [
+			ac_cv_func___builtin_setjmp=no
+		])
+		],
+[	LIBS="-lm $LIBS"])
+: ${ORIG_LIBS=$LIBS}
+
+AC_CHECK_LIB(crypt, crypt)      # glibc (GNU/Linux, GNU/Hurd, GNU/kFreeBSD)
+AC_CHECK_LIB(dl, dlopen)	# Dynamic linking for SunOS/Solaris and SYSV
+AC_CHECK_LIB(dld, shl_load)	# Dynamic linking for HP-UX
+AC_CHECK_LIB(socket, shutdown)  # SunOS/Solaris
+
+dnl Checks for header files.
+AC_HEADER_DIRENT
+dnl AC_HEADER_STDC has been checked in AC_USE_SYSTEM_EXTENSIONS
+AC_HEADER_STDBOOL
+AC_HEADER_SYS_WAIT
+
+AC_CHECK_HEADERS(a.out.h)
+AC_CHECK_HEADERS(atomic.h)
+AC_CHECK_HEADERS(direct.h)
+AC_CHECK_HEADERS(grp.h)
+AC_CHECK_HEADERS(fcntl.h)
+AC_CHECK_HEADERS(float.h)
+AC_CHECK_HEADERS(ieeefp.h)
+AC_CHECK_HEADERS(intrinsics.h)
+AC_CHECK_HEADERS(langinfo.h)
+AC_CHECK_HEADERS(limits.h)
+AC_CHECK_HEADERS(locale.h)
+AC_CHECK_HEADERS(malloc.h)
+AC_CHECK_HEADERS(malloc/malloc.h)
+AC_CHECK_HEADERS(malloc_np.h)
+AC_CHECK_HEADERS(net/socket.h)
+AC_CHECK_HEADERS(process.h)
+AC_CHECK_HEADERS(pwd.h)
+AC_CHECK_HEADERS(sanitizer/asan_interface.h)
+AC_CHECK_HEADERS(sanitizer/msan_interface.h)
+AC_CHECK_HEADERS(setjmpex.h)
+AC_CHECK_HEADERS(stdalign.h)
+AC_CHECK_HEADERS(sys/attr.h)
+AC_CHECK_HEADERS(sys/eventfd.h)
+AC_CHECK_HEADERS(sys/fcntl.h)
+AC_CHECK_HEADERS(sys/file.h)
+AC_CHECK_HEADERS(sys/id.h)
+AC_CHECK_HEADERS(sys/ioctl.h)
+AC_CHECK_HEADERS(sys/mkdev.h)
+AC_CHECK_HEADERS(sys/param.h)
+AC_CHECK_HEADERS(sys/prctl.h)
+AC_CHECK_HEADERS(sys/resource.h)
+AC_CHECK_HEADERS(sys/select.h)
+AC_CHECK_HEADERS(sys/sendfile.h)
+AC_CHECK_HEADERS(sys/socket.h)
+AC_CHECK_HEADERS(sys/syscall.h)
+AC_CHECK_HEADERS(sys/sysmacros.h)
+AC_CHECK_HEADERS(sys/time.h)
+AC_CHECK_HEADERS(sys/times.h)
+AC_CHECK_HEADERS(sys/uio.h)
+AC_CHECK_HEADERS(sys/utime.h)
+AC_CHECK_HEADERS(syscall.h)
+AC_CHECK_HEADERS(time.h)
+AC_CHECK_HEADERS(ucontext.h)
+AC_CHECK_HEADERS(utime.h)
+
+AC_ARG_WITH([gmp],
+  [AS_HELP_STRING([--without-gmp],
+    [disable GNU GMP to accelerate Bignum operations])],
+  [],
+  [with_gmp=yes])
+AS_IF([test "x$with_gmp" != xno],
+  [AC_CHECK_HEADERS(gmp.h)
+   AS_IF([test "x$ac_cv_header_gmp_h" != xno],
+     AC_SEARCH_LIBS([__gmpz_init], [gmp],
+       [AC_DEFINE(HAVE_LIBGMP, 1)]))])
+
+AC_ARG_WITH([jemalloc],
+  [AS_HELP_STRING([--with-jemalloc],[use jemalloc allocator])],
+  [with_jemalloc=$withval], [with_jemalloc=no])
+AS_IF([test "x$with_jemalloc" != xno],[
+  AC_SEARCH_LIBS([malloc_conf], [jemalloc],
+    [
+      AC_DEFINE(HAVE_LIBJEMALLOC, 1)
+      with_jemalloc=yes
+    ],
+    [test x$with_jemalloc = xyes && with_jemalloc=no])
+  AC_CHECK_HEADER(jemalloc/jemalloc.h, [
+    AC_DEFINE(RUBY_ALTERNATIVE_MALLOC_HEADER, [<jemalloc/jemalloc.h>])
+  ])
+  AS_IF([test "x$with_jemalloc" != xyes], [
+    AC_CACHE_CHECK([for jemalloc with JEMALLOC_MANGLE], rb_cv_jemalloc_demangle,
+      [AC_LINK_IFELSE([AC_LANG_PROGRAM([@%:@define JEMALLOC_MANGLE 1
+      @%:@ifdef RUBY_ALTERNATIVE_MALLOC_HEADER
+      @%:@include RUBY_ALTERNATIVE_MALLOC_HEADER
+      @%:@else
+      @%:@include <jemalloc.h>
+      @%:@endif], [return !&malloc_conf])],
+      [rb_cv_jemalloc_demangle=yes],
+      [rb_cv_jemalloc_demangle=no])
+    ])
+  ])
+  AS_IF([test "x$rb_cv_jemalloc_demangle" = xyes], [
+    AC_DEFINE(JEMALLOC_MANGLE)
+    with_jemalloc=yes
+  ])
+  AS_CASE(["$with_jemalloc"],
+    [yes],
+    [
+      AC_DEFINE(HAVE_MALLOC_CONF)
+      ac_cv_func_malloc_usable_size=yes
+    ],
+    [no],
+    [AC_MSG_ERROR([jemalloc requested but not found])
+  ])
+])
+
+dnl check for large file stuff
+mv confdefs.h confdefs1.h
+: > confdefs.h
+AC_SYS_LARGEFILE
+# On 32-bit Solaris, it is safe to define _LARGEFILE_SOURCE
+# which is not added by AC_SYS_LARGEFILE.
+AS_IF([test x"$enable_largefile" != xno], [
+  AS_CASE(["$target_os"], [solaris*], [
+    AC_MSG_CHECKING([wheather _LARGEFILE_SOURCE should be defined])
+    AS_CASE(["${ac_cv_sys_file_offset_bits}:${ac_cv_sys_large_files}"],
+      ["64:"|"64:no"|"64:unknown"], [
+	# insert _LARGEFILE_SOURCE before _FILE_OFFSET_BITS line
+	# that is the same order as "getconf LFS_CFLAGS" output
+	mv confdefs.h largefile0.h
+	: > confdefs.h
+	AC_DEFINE(_LARGEFILE_SOURCE)
+	cat largefile0.h >> confdefs.h
+	rm largefile0.h
+	AC_MSG_RESULT([yes])
+      ], [AC_MSG_RESULT([no])])
+  ])
+])
+mv confdefs.h largefile.h
+mv confdefs1.h confdefs.h
+cat largefile.h >> confdefs.h
+
+AS_CASE(["$target_os"],
+    [aix*], [
+    AS_CASE(["$target_cpu:$ac_cv_sys_large_files"],
+	[ppc64:*|powerpc64:*], [],
+	[*:no|*:unknown], [],
+	[
+	    # AIX currently does not support a 32-bit call to posix_fadvise()
+	    # if _LARGE_FILES is defined.
+	    ac_cv_func_posix_fadvise=no
+	])
+    ])
+
+AC_C_BIGENDIAN
+AC_C_CONST
+AC_C_CHAR_UNSIGNED
+AC_C_INLINE
+AC_C_VOLATILE
+AC_C_TYPEOF
+
+AS_CASE(":$ac_cv_c_const:$ac_cv_c_volatile:",
+    [*:no:*], [AC_MSG_ERROR(ANSI C-conforming const and volatile are mandatory)])
+
+AC_CHECK_TYPES([long long, off_t])
+
+AC_CACHE_CHECK([char bit], [rb_cv_char_bit],
+    [test "$universal_binary" = yes && cross_compiling=yes
+    AC_COMPUTE_INT([rb_cv_char_bit], [CHAR_BIT],
+	[AC_INCLUDES_DEFAULT([@%:@include <limits.h>])], [rb_cv_char_bit=8])
+    test "$universal_binary" = yes && cross_compiling=$real_cross_compiling])
+
+RUBY_CHECK_SIZEOF(int, [], [ILP])
+RUBY_CHECK_SIZEOF(short)
+RUBY_CHECK_SIZEOF(long, [int], [ILP LP])
+RUBY_CHECK_SIZEOF(long long)
+RUBY_CHECK_SIZEOF(__int64, [8], [ILP LP])
+RUBY_CHECK_SIZEOF(__int128, [16], [ILP LP])
+RUBY_CHECK_SIZEOF(off_t)
+RUBY_CHECK_SIZEOF(void*, [int long "long long"], [ILP LP LLP])
+RUBY_CHECK_SIZEOF(float)
+RUBY_CHECK_SIZEOF(double)
+RUBY_CHECK_SIZEOF(time_t, [long "long long"], [], [@%:@include <time.h>])
+RUBY_CHECK_SIZEOF(clock_t, [], [], [@%:@include <time.h>])
+
+AC_CACHE_CHECK(packed struct attribute, rb_cv_packed_struct,
+    [rb_cv_packed_struct=no
+    for mac in \
+	"__pragma(pack(push, 1)) x __pragma(pack(pop))" \
+	"x __attribute__((packed))" \
+	; do
+	AC_TRY_COMPILE([@%:@define PACKED_STRUCT(x) $mac
+			PACKED_STRUCT(struct { int a; });], [],
+		[rb_cv_packed_struct=$mac; break])
+    done])
+AS_IF([test "$rb_cv_packed_struct" != no], [
+    AC_DEFINE_UNQUOTED([PACKED_STRUCT(x)], [$rb_cv_packed_struct])
+    RUBY_TRY_CFLAGS(-Wno-address-of-packed-member, [AC_DEFINE(USE_UNALIGNED_MEMBER_ACCESS)])
+], [
+    AC_DEFINE_UNQUOTED([PACKED_STRUCT(x)], x)
+])
+
+AS_IF([test "x$ac_cv_type_long_long" = xyes], [
+    RUBY_CHECK_PRINTF_PREFIX(long long, ll I64, LL)
+], [test "x$ac_cv_type___int64" = xyes], [
+    RUBY_CHECK_PRINTF_PREFIX(__int64, ll I64, LL)
+])
+
+RUBY_REPLACE_TYPE(pid_t, int, PIDT)
+RUBY_REPLACE_TYPE(uid_t, int, UIDT)
+RUBY_REPLACE_TYPE(gid_t, int, GIDT)
+RUBY_REPLACE_TYPE(time_t, [], TIMET, [@%:@include <time.h>])
+RUBY_REPLACE_TYPE(dev_t, [int long "long long"], DEVT)
+RUBY_REPLACE_TYPE(mode_t, ["unsigned short" "unsigned int" long], MODET, [@%:@include <sys/stat.h>])
+RUBY_REPLACE_TYPE(rlim_t, [int long "long long"], RLIM, [
+@%:@ifdef HAVE_SYS_TYPES_H
+@%:@include <sys/types.h>
+@%:@endif
+@%:@ifdef HAVE_SYS_TYPES_H
+@%:@include <sys/time.h>
+@%:@endif
+@%:@include <sys/resource.h>
+])
+RUBY_REPLACE_TYPE(off_t, [], OFFT)
+RUBY_REPLACE_TYPE(clockid_t, [], CLOCKID, [@%:@ifdef HAVE_TIME_H
+@%:@ include <time.h>
+@%:@endif
+@%:@ifdef HAVE_SYS_TIME_H
+@%:@ include <sys/time.h>
+@%:@endif])
+
+AC_CACHE_CHECK(for prototypes, rb_cv_have_prototypes,
+  [AC_TRY_COMPILE([int foo(int x) { return 0; }], [return foo(10);],
+	rb_cv_have_prototypes=yes,
+	rb_cv_have_prototypes=no)])
+AS_IF([test "$rb_cv_have_prototypes" = yes], [
+  AC_DEFINE(HAVE_PROTOTYPES)
+])
+
+AC_CACHE_CHECK(token paste string, rb_cv_tokenpaste,
+  [AC_TRY_COMPILE([@%:@define paste(a,b) a@%:@@%:@b],
+		  [int xy = 1; return paste(x,y);],
+		  rb_cv_tokenpaste=ansi,
+		  rb_cv_tokenpaste=knr)])
+AS_IF([test "$rb_cv_tokenpaste" = ansi], [
+  AC_DEFINE(TOKEN_PASTE(x,y),[x@%:@@%:@y])
+], [
+  AC_DEFINE(TOKEN_PASTE(x,y),[x/**/y])
+])
+
+AC_CACHE_CHECK(stringization, rb_cv_stringization, [
+  rb_cv_stringization=no
+  for string in "#expr" '"expr"'; do
+    AC_COMPILE_IFELSE([
+      AC_LANG_BOOL_COMPILE_TRY([
+#define STRINGIZE0(expr) $string
+#define STRINGIZE(expr) STRINGIZE0(expr)
+#undef real_test_for_stringization
+#define test_for_stringization -.real_test_for_stringization.-
+const char stringized[[]] = STRINGIZE(test_for_stringization);
+], [sizeof(stringized) == 32])],
+      [rb_cv_stringization="$string"; break],
+      [rb_cv_stringization=no])
+  done]
+)
+AC_DEFINE(STRINGIZE(expr),STRINGIZE0(expr))
+AS_IF([test x"$rb_cv_stringization" != xno -a "$rb_cv_stringization" != "#expr"], [
+  AC_DEFINE_UNQUOTED(STRINGIZE0(expr),$rb_cv_stringization)
+  AC_DEFINE(OLD_FASHIONED_STRINGIZATION,1)
+])
+
+AC_CACHE_CHECK([string literal concatenation],
+  rb_cv_string_literal_concatenation, [
+  AC_COMPILE_IFELSE([
+    AC_LANG_BOOL_COMPILE_TRY([
+const char concatenated_literal[[]] = "literals" "to"
+  "be" "concatenated.";
+], [sizeof(concatenated_literal) == 26])],
+    [rb_cv_string_literal_concatenation=yes],
+    [rb_cv_string_literal_concatenation=no])]
+)
+AS_IF([test "$rb_cv_string_literal_concatenation" = no], [
+  AC_MSG_ERROR([No string literal concatenation])
+])
+
+AC_CACHE_CHECK(for variable length prototypes and stdarg.h, rb_cv_stdarg,
+  [AC_TRY_COMPILE([
+#include <stdarg.h>
+int foo(int x, ...) {
+	va_list va;
+	va_start(va, x);
+	va_arg(va, int);
+	va_arg(va, char *);
+	va_arg(va, double);
+	return 0;
+}
+], [return foo(10, "", 3.14);],
+	rb_cv_stdarg=yes,
+	rb_cv_stdarg=no)])
+AS_IF([test "$rb_cv_stdarg" = yes], [
+  AC_DEFINE(HAVE_STDARG_PROTOTYPES)
+])
+
+AC_CACHE_CHECK(for variable length macro, rb_cv_va_args_macro,
+  [AC_TRY_COMPILE([
+int foo(int x, ...);
+@%:@define FOO(a, ...) foo(a, @%:@@%:@__VA_ARGS__)
+], [FOO(1);FOO(1,2);FOO(1,2,3);],
+  rb_cv_va_args_macro=yes,
+  rb_cv_va_args_macro=no)])
+AS_IF([test "$rb_cv_va_args_macro" = yes], [
+  AC_DEFINE(HAVE_VA_ARGS_MACRO)
+])
+
+AC_CACHE_CHECK([for alignas() syntax], rb_cv_have_alignas, [
+rb_cv_have_alignas=no
+RUBY_WERROR_FLAG([
+for attr in \
+    "_Alignas(x)" \
+    "alignas(x)" \
+    "@<:@@<:@alignas(x)@:>@@:>@" \
+    "__declspec(aligned(x))" \
+    "__attribute__((__aligned__(x)))" \
+;
+do
+    # C11 _Alignas and GCC __attribute__((__aligned__)) behave
+    # slightly differently.  What we want is GCC's.  Check that
+    # here by something C11 does not allow (`struct ALIGNAS ...`)
+    AC_TRY_COMPILE(
+	[@%:@define ALIGNAS(x) $attr
+	struct ALIGNAS(128) conftest_tag { int foo; } foo; ], [],
+        [rb_cv_have_alignas="$attr"; break], [])
+done
+])])
+AS_IF([test "$rb_cv_have_alignas" != no], [
+    AC_DEFINE_UNQUOTED([RUBY_ALIGNAS(x)], $rb_cv_have_alignas)
+])
+
+AC_CACHE_CHECK([for alignof() syntax], rb_cv_have_alignof,[
+rb_cv_have_alignof=no
+RUBY_WERROR_FLAG([
+for expr in \
+    "alignof" \
+    "_Alignof" \
+    "__alignof" \
+    "__alignof__" \
+;
+do
+    AC_TRY_COMPILE([
+    @%:@ifdef HAVE_STDALIGN_H
+    @%:@include <stdalign.h>
+    @%:@endif],[return (int)$expr(int);],
+        [rb_cv_have_alignof="$expr"; break], [])
+done
+])])
+AS_IF([test "$rb_cv_have_alignof" != no], [
+    AC_DEFINE_UNQUOTED(RUBY_ALIGNOF, $rb_cv_have_alignof)
+])
+
+RUBY_FUNC_ATTRIBUTE(__const__, CONSTFUNC)
+RUBY_FUNC_ATTRIBUTE(__pure__, PUREFUNC)
+RUBY_FUNC_ATTRIBUTE(__noreturn__, NORETURN)
+RUBY_FUNC_ATTRIBUTE(__deprecated__, DEPRECATED)
+RUBY_FUNC_ATTRIBUTE(__deprecated__("by "@%:@n), DEPRECATED_BY(n,x), rb_cv_func_deprecated_by)
+RUBY_TYPE_ATTRIBUTE(__deprecated__ mesg, DEPRECATED_TYPE(mesg,x), rb_cv_type_deprecated)
+RUBY_FUNC_ATTRIBUTE(__noinline__, NOINLINE)
+RUBY_FUNC_ATTRIBUTE(__always_inline__, ALWAYS_INLINE)
+RUBY_FUNC_ATTRIBUTE(__no_sanitize__(san), NO_SANITIZE(san, x), rb_cv_func_no_sanitize)
+RUBY_FUNC_ATTRIBUTE(__no_sanitize_address__, NO_SANITIZE_ADDRESS)
+RUBY_FUNC_ATTRIBUTE(__no_address_safety_analysis__, NO_ADDRESS_SAFETY_ANALYSIS)
+RUBY_FUNC_ATTRIBUTE(__warn_unused_result__, WARN_UNUSED_RESULT)
+RUBY_FUNC_ATTRIBUTE(__unused__, MAYBE_UNUSED)
+RUBY_FUNC_ATTRIBUTE(__error__ mesg, ERRORFUNC(mesg,x), rb_cv_func___error__)
+RUBY_FUNC_ATTRIBUTE(__warning__ mesg, WARNINGFUNC(mesg,x), rb_cv_func___warning__)
+RUBY_FUNC_ATTRIBUTE(__weak__, WEAK, rb_cv_func_weak)
+AS_IF([test "$rb_cv_func_weak" != x], [
+   AC_DEFINE(HAVE_FUNC_WEAK)
+])
+
+if_i386=${universal_binary+[defined __i386__]}
+RUBY_FUNC_ATTRIBUTE(__stdcall__,  FUNC_STDCALL,  rb_cv_func_stdcall,  ${if_i386})
+RUBY_FUNC_ATTRIBUTE(__cdecl__,    FUNC_CDECL,    rb_cv_func_cdecl,    ${if_i386})
+RUBY_FUNC_ATTRIBUTE(__fastcall__, FUNC_FASTCALL, rb_cv_func_fastcall, ${if_i386})
+RUBY_FUNC_ATTRIBUTE(__optimize__("O0"), FUNC_UNOPTIMIZED, rb_cv_func_unoptimized)
+RUBY_FUNC_ATTRIBUTE(__optimize__("-Os","-fomit-frame-pointer"), FUNC_MINIMIZED, rb_cv_func_minimized)
+
+AS_IF([test "$GCC" = yes], [
+    AC_CACHE_CHECK([for function alias], [rb_cv_gcc_function_alias],
+	[rb_cv_gcc_function_alias=no
+	for a in alias weak,alias; do
+	    AC_TRY_LINK([void foo(void) {}
+		void bar(void) __attribute__(($a("foo")));], [bar()],
+		[rb_cv_gcc_function_alias=$a; break])
+	done])
+    AS_IF([test "$rb_cv_gcc_function_alias" != no], [
+	AC_DEFINE(HAVE_ATTRIBUTE_FUNCTION_ALIAS)
+	AC_DEFINE_UNQUOTED([RUBY_ALIAS_FUNCTION_TYPE(type, prot, name, args)],
+			   [type prot __attribute__(($rb_cv_gcc_function_alias(@%:@name)));])
+	AC_DEFINE_UNQUOTED([RUBY_ALIAS_FUNCTION_VOID(prot, name, args)],
+			   [RUBY_ALIAS_FUNCTION_TYPE(void, prot, name, args)])
+    ])
+
+    AC_CACHE_CHECK([for __atomic builtins], [rb_cv_gcc_atomic_builtins], [
+	AC_TRY_LINK([unsigned char atomic_var;],
+		    [
+			__atomic_exchange_n(&atomic_var, 0, __ATOMIC_SEQ_CST);
+			__atomic_exchange_n(&atomic_var, 1, __ATOMIC_SEQ_CST);
+			__atomic_fetch_add(&atomic_var, 1, __ATOMIC_SEQ_CST);
+			__atomic_fetch_sub(&atomic_var, 1, __ATOMIC_SEQ_CST);
+			__atomic_or_fetch(&atomic_var, 1, __ATOMIC_SEQ_CST);
+		    ],
+		    [rb_cv_gcc_atomic_builtins=yes],
+		    [rb_cv_gcc_atomic_builtins=no])])
+    AS_IF([test "$rb_cv_gcc_atomic_builtins" = yes], [
+	AC_DEFINE(HAVE_GCC_ATOMIC_BUILTINS)
+    ])
+
+    AC_CACHE_CHECK([for __sync builtins], [rb_cv_gcc_sync_builtins], [
+	AC_TRY_LINK([unsigned char atomic_var;],
+		    [
+			__sync_lock_test_and_set(&atomic_var, 0);
+			__sync_lock_test_and_set(&atomic_var, 1);
+			__sync_fetch_and_add(&atomic_var, 1);
+			__sync_fetch_and_sub(&atomic_var, 1);
+			__sync_or_and_fetch(&atomic_var, 1);
+			__sync_val_compare_and_swap(&atomic_var, 0, 1);
+		    ],
+		    [rb_cv_gcc_sync_builtins=yes],
+		    [rb_cv_gcc_sync_builtins=no])])
+    AS_IF([test "$rb_cv_gcc_sync_builtins" = yes], [
+	AC_DEFINE(HAVE_GCC_SYNC_BUILTINS)
+    ])
+
+    AC_CACHE_CHECK(for __builtin_unreachable, rb_cv_func___builtin_unreachable,
+    [RUBY_WERROR_FLAG(
+    [AC_TRY_LINK([volatile int zero;],
+	[if (zero) __builtin_unreachable();],
+	[rb_cv_func___builtin_unreachable=yes],
+	[rb_cv_func___builtin_unreachable=no])
+    ])
+    ])
+    AS_IF([test "$rb_cv_func___builtin_unreachable" = yes], [
+	AC_DEFINE_UNQUOTED(UNREACHABLE, [__builtin_unreachable()])
+    ])
+])
+
+AC_CACHE_CHECK(for exported function attribute, rb_cv_func_exported, [
+rb_cv_func_exported=no
+RUBY_WERROR_FLAG([
+for mac in '__attribute__ ((__visibility__("default")))' '__declspec(dllexport)'; do
+  AC_TRY_COMPILE([@%:@define RUBY_FUNC_EXPORTED $mac extern
+    RUBY_FUNC_EXPORTED void conftest_attribute_check(void);], [],
+    [rb_cv_func_exported="$mac"; break])
+done
+])])
+AS_IF([test "$rb_cv_func_exported" != no], [
+    AC_DEFINE_UNQUOTED(RUBY_FUNC_EXPORTED, [$rb_cv_func_exported extern])
+])
+
+RUBY_APPEND_OPTION(XCFLAGS, -DRUBY_EXPORT)
+
+AC_ARG_ENABLE(mathn,
+    AS_HELP_STRING([--disable-mathn], [disable canonicalization for mathn]),
+    [mathn=$enableval], [mathn=yes])
+test "x$mathn" = xyes || mathn=
+AC_SUBST(MATHN, $mathn)
+
+AC_CACHE_CHECK(for function name string predefined identifier,
+    rb_cv_function_name_string,
+    [rb_cv_function_name_string=no
+    RUBY_WERROR_FLAG([
+	for func in __func__ __FUNCTION__; do
+	    AC_TRY_LINK([@%:@include <stdio.h>],
+			[puts($func);],
+			[rb_cv_function_name_string=$func
+			break])
+	done
+    ])]
+)
+AS_IF([test "$rb_cv_function_name_string" != no], [
+    AC_DEFINE_UNQUOTED(RUBY_FUNCTION_NAME_STRING, [$rb_cv_function_name_string])
+])
+
+AC_CACHE_CHECK(if enum over int is allowed, rb_cv_enum_over_int, [
+    rb_cv_enum_over_int=no
+    AS_IF([test "x$ac_cv_type_long_long" = xyes], [
+	type="unsigned long long" max="ULLONG_MAX"
+    ], [
+	type="unsigned long" max="ULONG_MAX"
+    ])
+    RUBY_WERROR_FLAG([
+        AC_COMPILE_IFELSE([
+            AC_LANG_BOOL_COMPILE_TRY([
+                    @%:@include <limits.h>
+                    enum {conftest_max = $max};
+                ], [
+                    (conftest_max == $max) &&
+                    (sizeof(conftest_max) == sizeof($type))
+                ]
+	    )],
+	    [rb_cv_enum_over_int=yes],
+	    [rb_cv_enum_over_int=no]
+	)
+    ])
+])
+AS_IF([test $rb_cv_enum_over_int = yes], [
+    AC_DEFINE(ENUM_OVER_INT, 1)
+])
+
+dnl Check whether we need to define sys_nerr locally
+AC_CHECK_DECLS([sys_nerr], [], [], [$ac_includes_default
+@%:@include <errno.h>])
+
+AC_CHECK_DECLS([getenv])
+
+AS_CASE(["$target_cpu"],
+[alpha*|sh4|sh4el|sh4eb], [AS_CASE(["$target_os"::"$GCC"],
+		[*::yes],  # gcc
+			[CFLAGS="-mieee $CFLAGS"],
+		[osf*],    # ccc
+			[CFLAGS="-ieee $CFLAGS"],
+		)],
+[sparc*], [AC_LIBOBJ([sparc])])
+
+ac_cv_header_net_socket_h=${ac_cv_header_net_socket_h=no}
+AS_IF([test "$ac_cv_header_net_socket_h" = yes], [
+    ac_cv_header_sys_socket_h=${ac_cv_header_sys_socket_h=no}
+], [
+    ac_cv_header_sys_socket_h=${ac_cv_header_sys_socket_h=yes}
+])
+
+
+AC_TYPE_SIZE_T
+RUBY_CHECK_SIGNEDNESS(size_t, [AC_MSG_ERROR(size_t is signed)], [],
+		      [@%:@include <sys/types.h>])
+RUBY_CHECK_SIZEOF(size_t, [int long void*], [], [@%:@include <sys/types.h>])
+RUBY_CHECK_SIZEOF(ptrdiff_t, size_t, [], [@%:@include <stddef.h>])
+RUBY_CHECK_PRINTF_PREFIX(size_t, z)
+RUBY_CHECK_PRINTF_PREFIX(ptrdiff_t, t)
+AC_STRUCT_ST_BLKSIZE
+AC_STRUCT_ST_BLOCKS
+AC_STRUCT_ST_RDEV
+RUBY_CHECK_SIZEOF([struct stat.st_size], [off_t int long "long long"], [], [@%:@include <sys/stat.h>])
+AS_IF([test "$ac_cv_member_struct_stat_st_blocks" = yes], [
+    RUBY_CHECK_SIZEOF([struct stat.st_blocks], [off_t int long "long long"], [], [@%:@include <sys/stat.h>])
+])
+RUBY_CHECK_SIZEOF([struct stat.st_ino], [long "long long"], [], [@%:@include <sys/stat.h>])
+AC_CHECK_MEMBERS([struct stat.st_atim])
+AC_CHECK_MEMBERS([struct stat.st_atimespec])
+AC_CHECK_MEMBERS([struct stat.st_atimensec])
+AC_CHECK_MEMBERS([struct stat.st_mtim])
+AC_CHECK_MEMBERS([struct stat.st_mtimespec])
+AC_CHECK_MEMBERS([struct stat.st_mtimensec])
+AC_CHECK_MEMBERS([struct stat.st_ctim])
+AC_CHECK_MEMBERS([struct stat.st_ctimespec])
+AC_CHECK_MEMBERS([struct stat.st_ctimensec])
+AC_CHECK_MEMBERS([struct stat.st_birthtimespec])
+
+AC_CHECK_TYPES([struct timeval], [], [], [@%:@ifdef HAVE_TIME_H
+@%:@include <time.h>
+@%:@endif
+@%:@ifdef HAVE_SYS_TIME_H
+@%:@include <sys/time.h>
+@%:@endif])
+
+AS_IF([test "${ac_cv_type_struct_timeval}" = yes], [
+    RUBY_CHECK_SIZEOF([struct timeval.tv_sec], [time_t long "long long"], [],
+		      [@%:@ifdef HAVE_TIME_H
+@%:@include <time.h>
+@%:@endif
+@%:@ifdef HAVE_SYS_TIME_H
+@%:@include <sys/time.h>
+@%:@endif])
+    AS_CASE(${ac_cv_sizeof_struct_timeval_tv_sec},
+	    [SIZEOF_INT], [t=int],
+	    [SIZEOF_LONG], [t=long],
+	    [SIZEOF_LONG_LONG], [t=LONG_LONG],
+	    [t=])
+    AS_IF([test "${t}" != ""], [
+	AC_DEFINE_UNQUOTED(TYPEOF_TIMEVAL_TV_SEC, [$t])
+    ])
+])
+
+AC_CHECK_TYPES([struct timespec], [], [], [@%:@ifdef HAVE_TIME_H
+@%:@include <time.h>
+@%:@endif
+@%:@ifdef HAVE_SYS_TIME_H
+@%:@include <sys/time.h>
+@%:@endif])
+
+AC_CHECK_TYPES([struct timezone], [], [], [@%:@ifdef HAVE_TIME_H
+@%:@ include <time.h>
+@%:@endif
+@%:@ifdef HAVE_SYS_TIME_H
+@%:@ include <sys/time.h>
+@%:@endif])
+
+AC_CACHE_VAL([rb_cv_large_fd_select],
+    [AC_CHECK_TYPE(fd_mask, [rb_cv_large_fd_select=yes], [rb_cv_large_fd_select=no], [AC_INCLUDES_DEFAULT([])
+@%:@ifdef HAVE_SYS_SELECT_H
+@%:@ include <sys/select.h>
+@%:@endif])])
+AS_IF([test "$rb_cv_large_fd_select" = yes], [
+    AC_DEFINE(HAVE_RB_FD_INIT, 1)
+])
+
+RUBY_DEFINT(int8_t, 1)
+RUBY_DEFINT(uint8_t, 1, unsigned)
+RUBY_DEFINT(int16_t, 2)
+RUBY_DEFINT(uint16_t, 2, unsigned)
+RUBY_DEFINT(int32_t, 4)
+RUBY_DEFINT(uint32_t, 4, unsigned)
+RUBY_DEFINT(int64_t, 8)
+RUBY_DEFINT(uint64_t, 8, unsigned)
+RUBY_DEFINT(int128_t, 16)
+RUBY_DEFINT(uint128_t, 16, unsigned)
+RUBY_DEFINT(intptr_t, void*)
+RUBY_DEFINT(uintptr_t, void*, unsigned)
+AS_IF([test "x$rb_cv_type_intptr_t" != xno], [
+    RUBY_CHECK_PRINTF_PREFIX(intptr_t, '' ll I64 l, PTR)
+])
+RUBY_DEFINT(ssize_t, size_t, [], [@%:@include <sys/types.h>])	dnl may differ from int, so not use AC_TYPE_SSIZE_T.
+AS_IF([test "x$rb_cv_type_int64_t" != xno], [
+    RUBY_CHECK_PRINTF_PREFIX(int64_t, ll I64 l, 64)
+])
+
+AC_CACHE_CHECK(for stack end address, rb_cv_stack_end_address,
+[rb_cv_stack_end_address=no
+  AC_TRY_LINK(
+    [extern void *__libc_stack_end;],
+    [if (!__libc_stack_end) return 1;],
+    [rb_cv_stack_end_address="__libc_stack_end"])
+])
+AS_IF([test $rb_cv_stack_end_address != no], [
+  AC_DEFINE_UNQUOTED(STACK_END_ADDRESS, $rb_cv_stack_end_address)
+])
+
+dnl Checks for library functions.
+AC_TYPE_GETGROUPS
+AC_TYPE_SIGNAL
+AS_CASE(["${target_cpu}-${target_os}:${target_archs}"],
+[powerpc-darwin*], [
+  AC_LIBSOURCES(alloca.c)
+  AC_SUBST([ALLOCA], [\${LIBOBJDIR}alloca.${ac_objext}])
+  AC_DEFINE(C_ALLOCA)
+  AC_DEFINE_UNQUOTED(alloca, alloca)
+  ],
+[universal-darwin*:*ppc*], [
+  AC_LIBSOURCES(alloca.c)
+  AC_SUBST([ALLOCA], [\${LIBOBJDIR}alloca.${ac_objext}])
+  RUBY_DEFINE_IF([defined __powerpc__], C_ALLOCA, 1)
+  RUBY_DEFINE_IF([defined __powerpc__], alloca, alloca)
+  ],
+[
+  AC_FUNC_ALLOCA
+  ])
+AS_IF([test "x$ALLOCA" = "x"], [
+    AC_CACHE_CHECK([for dynamic size alloca], rb_cv_dynamic_alloca, [
+    for chk in ok __chkstk; do
+	AC_TRY_LINK([
+	    @%:@ifdef HAVE_ALLOCA_H
+	    @%:@include <alloca.h>
+	    @%:@endif
+	    void $chk() {}
+	    int dynamic_alloca_test;
+	    int dynamic_alloca_result;],
+	    [dynamic_alloca_result = alloca(dynamic_alloca_test) != 0;],
+	    [rb_cv_dynamic_alloca=$chk; break])
+    done])
+    AS_IF([test "x$rb_cv_dynamic_alloca" = "x__chkstk"], [
+	AC_DEFINE_UNQUOTED(RUBY_ALLOCA_CHKSTK, _$rb_cv_dynamic_alloca)
+	AS_CASE("$target_cpu",
+	[x64|x86_64], [
+	    AC_SUBST([ALLOCA], [\${LIBOBJDIR}x86_64-chkstk.${ac_objext}])
+	],)
+    ])
+])
+AC_FUNC_MEMCMP
+
+# http://sources.redhat.com/ml/libc-hacker/2005-08/msg00008.html
+# Debian GNU/Linux Etch's libc6.1 2.3.6.ds1-13etch5 has this problem.
+# Debian GNU/Linux Lenny's libc6.1 2.7-10 has no problem.
+AC_CACHE_CHECK(for broken erfc of glibc-2.3.6 on IA64, rb_cv_broken_glibc_ia64_erfc,
+  [AC_TRY_RUN([
+#include <math.h>
+int
+main()
+{
+   erfc(10000.0);
+   return 0;
+}
+],
+	rb_cv_broken_glibc_ia64_erfc=no,
+	rb_cv_broken_glibc_ia64_erfc=yes,
+	rb_cv_broken_glibc_ia64_erfc=no)])
+AS_CASE([$rb_cv_broken_glibc_ia64_erfc],[yes],[ac_cv_func_erf=no])
+
+AS_CASE(["$target_os"],[freebsd*],[
+	 AC_DEFINE(BROKEN_CLOSE)
+	 AC_REPLACE_FUNCS(close)
+	 ])
+
+AC_REPLACE_FUNCS(acosh)
+AC_REPLACE_FUNCS(cbrt)
+AC_REPLACE_FUNCS(crypt)
+AC_REPLACE_FUNCS(dup2)
+AC_REPLACE_FUNCS(erf)
+AC_REPLACE_FUNCS(explicit_bzero)
+AC_REPLACE_FUNCS(ffs)
+AC_REPLACE_FUNCS(finite)
+AC_REPLACE_FUNCS(flock)
+AC_REPLACE_FUNCS(hypot)
+AC_REPLACE_FUNCS(isinf)
+AC_REPLACE_FUNCS(isnan)
+AC_REPLACE_FUNCS(lgamma_r)
+AC_REPLACE_FUNCS(memmove)
+AC_REPLACE_FUNCS(nan)
+AC_REPLACE_FUNCS(nextafter)
+AC_REPLACE_FUNCS(setproctitle)
+AC_REPLACE_FUNCS(strchr)
+AC_REPLACE_FUNCS(strerror)
+AC_REPLACE_FUNCS(strlcat)
+AC_REPLACE_FUNCS(strlcpy)
+AC_REPLACE_FUNCS(strstr)
+AC_REPLACE_FUNCS(tgamma)
+
+# for missing/setproctitle.c
+AS_CASE(["$target_os"],
+[aix* | k*bsd*-gnu | kopensolaris*-gnu | linux* | darwin*], [AC_DEFINE(SPT_TYPE,SPT_REUSEARGV)],
+[hpux*], [AC_DEFINE(SPT_TYPE,SPT_PSTAT) ],
+[])
+AC_CHECK_HEADERS(sys/pstat.h)
+
+
+AC_CACHE_CHECK(for signbit, rb_cv_have_signbit,
+  [AC_TRY_LINK([
+#include <math.h>
+], [int v = signbit(-0.0);],
+	rb_cv_have_signbit=yes,
+	rb_cv_have_signbit=no)])
+AS_IF([test "$rb_cv_have_signbit" = yes], [
+  AC_DEFINE(HAVE_SIGNBIT)
+], [
+  AC_LIBOBJ([signbit])
+])
+
+AC_CACHE_CHECK(for broken memmem, rb_cv_broken_memmem, [
+    AC_TRY_RUN([
+@%:@include <string.h>
+
+int
+main(int argc, char **argv)
+{
+    const char *str = "hogefugafoobar";
+    const char *rs = "foo";
+    const char *empty = "";
+    char *p;
+
+    p = memmem(str, strlen(str), rs, strlen(rs));
+    if (p == str+8) {
+	p = memmem(str, strlen(str), empty, strlen(empty));
+	if (p == str)
+	    return 0;
+    }
+    return 1;
+}
+    ],
+    rb_cv_broken_memmem=no,
+    rb_cv_broken_memmem=yes,
+    rb_cv_broken_memmem=yes)
+])
+test x"$rb_cv_broken_memmem" = xyes && ac_cv_func_memmem=no
+
+AC_FUNC_FORK
+
+AC_CHECK_FUNCS(__syscall)
+AC_CHECK_FUNCS(_longjmp)		# used for AC_ARG_WITH(setjmp-type)
+# we don't use _setjmp if _longjmp doesn't exist.
+test x$ac_cv_func__longjmp = xno && ac_cv_func__setjmp=no
+AC_CHECK_FUNCS(arc4random_buf)
+AC_CHECK_FUNCS(atan2l atan2f)
+AC_CHECK_FUNCS(chroot)
+AC_CHECK_FUNCS(chsize)
+AC_CHECK_FUNCS(clock_gettime)
+AC_CHECK_FUNCS(cosh)
+AC_CHECK_FUNCS(crypt_r)
+AC_CHECK_FUNCS(daemon)
+AC_CHECK_FUNCS(dirfd)
+AC_CHECK_FUNCS(dl_iterate_phdr)
+AC_CHECK_FUNCS(dlopen)
+AC_CHECK_FUNCS(dladdr)
+AC_CHECK_FUNCS(dup)
+AC_CHECK_FUNCS(dup3)
+AC_CHECK_FUNCS(eaccess)
+AC_CHECK_FUNCS(endgrent)
+AC_CHECK_FUNCS(eventfd)
+AC_CHECK_FUNCS(fchmod)
+AC_CHECK_FUNCS(fchown)
+AC_CHECK_FUNCS(fcntl)
+AC_CHECK_FUNCS(fdatasync)
+AC_CHECK_FUNCS(fdopendir)
+AC_CHECK_FUNCS(fgetattrlist)
+AC_CHECK_FUNCS(fmod)
+AC_CHECK_FUNCS(fstatat)
+AC_CHECK_FUNCS(fsync)
+AC_CHECK_FUNCS(ftruncate)
+AC_CHECK_FUNCS(ftruncate64)		# used for Win32 platform
+AC_CHECK_FUNCS(getattrlist)
+AC_CHECK_FUNCS(getcwd)
+AC_CHECK_FUNCS(getgidx)
+AC_CHECK_FUNCS(getgrnam)
+AC_CHECK_FUNCS(getgrnam_r)
+AC_CHECK_FUNCS(getgroups)
+AC_CHECK_FUNCS(getpgid)
+AC_CHECK_FUNCS(getpgrp)
+AC_CHECK_FUNCS(getpriority)
+AC_CHECK_FUNCS(getpwnam_r)
+AC_CHECK_FUNCS(getresgid)
+AC_CHECK_FUNCS(getresuid)
+AC_CHECK_FUNCS(getrlimit)
+AC_CHECK_FUNCS(getsid)
+AC_CHECK_FUNCS(gettimeofday)		# for making ac_cv_func_gettimeofday
+AC_CHECK_FUNCS(getuidx)
+AC_CHECK_FUNCS(gmtime_r)
+AC_CHECK_FUNCS(grantpt)
+AC_CHECK_FUNCS(initgroups)
+AC_CHECK_FUNCS(ioctl)
+AC_CHECK_FUNCS(isfinite)
+AC_CHECK_FUNCS(issetugid)
+AC_CHECK_FUNCS(killpg)
+AC_CHECK_FUNCS(lchmod)
+AC_CHECK_FUNCS(lchown)
+AC_CHECK_FUNCS(link)
+AC_CHECK_FUNCS(llabs)
+AC_CHECK_FUNCS(lockf)
+AC_CHECK_FUNCS(log2)
+AC_CHECK_FUNCS(lstat)
+AC_CHECK_FUNCS(lutimes)
+AC_CHECK_FUNCS(malloc_usable_size)
+AC_CHECK_FUNCS(malloc_size)
+AC_CHECK_FUNCS(mblen)
+AC_CHECK_FUNCS(memalign)
+AC_CHECK_FUNCS(memset_s)
+AC_CHECK_FUNCS(writev)
+AC_CHECK_FUNCS(memrchr)
+AC_CHECK_FUNCS(memmem)
+AC_CHECK_FUNCS(mkfifo)
+AC_CHECK_FUNCS(mknod)
+AC_CHECK_FUNCS(mktime)
+AC_CHECK_FUNCS(openat)
+AC_CHECK_FUNCS(pipe2)
+AC_CHECK_FUNCS(poll)
+AC_CHECK_FUNCS(posix_fadvise)
+AC_CHECK_FUNCS(posix_memalign)
+AC_CHECK_FUNCS(ppoll)
+AC_CHECK_FUNCS(pread)
+AC_CHECK_FUNCS(pwrite)
+AC_CHECK_FUNCS(qsort_r)
+AC_CHECK_FUNCS(qsort_s)
+AC_CHECK_FUNCS(readlink)
+AC_CHECK_FUNCS(round)
+AC_CHECK_FUNCS(sched_getaffinity)
+AC_CHECK_FUNCS(seekdir)
+AC_CHECK_FUNCS(select_large_fdset)
+AC_CHECK_FUNCS(sendfile)
+AC_CHECK_FUNCS(setegid)
+AC_CHECK_FUNCS(setenv)
+AC_CHECK_FUNCS(seteuid)
+AC_CHECK_FUNCS(setgid)
+AC_CHECK_FUNCS(setgroups)
+AC_CHECK_FUNCS(setpgid)
+AC_CHECK_FUNCS(setpgrp)
+AC_CHECK_FUNCS(setregid)
+AC_CHECK_FUNCS(setresgid)
+AC_CHECK_FUNCS(setresuid)
+AC_CHECK_FUNCS(setreuid)
+AC_CHECK_FUNCS(setrgid)
+AC_CHECK_FUNCS(setrlimit)
+AC_CHECK_FUNCS(setruid)
+AC_CHECK_FUNCS(setsid)
+AC_CHECK_FUNCS(setuid)
+AC_CHECK_FUNCS(shutdown)
+AC_CHECK_FUNCS(sigaction)
+AC_CHECK_FUNCS(sigaltstack)
+AC_CHECK_FUNCS(sigprocmask)
+AC_CHECK_FUNCS(sinh)
+AC_CHECK_FUNCS(spawnv)
+AC_CHECK_FUNCS(symlink)
+AC_CHECK_FUNCS(syscall)
+AC_CHECK_FUNCS(sysconf)
+AC_CHECK_FUNCS(tanh)
+AC_CHECK_FUNCS(telldir)
+AC_CHECK_FUNCS(timegm)
+AC_CHECK_FUNCS(times)
+AC_CHECK_FUNCS(truncate)
+AC_CHECK_FUNCS(truncate64)		# used for Win32
+AC_CHECK_FUNCS(unsetenv)
+AC_CHECK_FUNCS(utimensat)
+AC_CHECK_FUNCS(utimes)
+AC_CHECK_FUNCS(wait4)
+AC_CHECK_FUNCS(waitpid)
+
+AS_CASE(["$ac_cv_func_memset_s:$ac_cv_func_qsort_s"], [*yes*],
+    [RUBY_DEFINE_IF([!defined __STDC_WANT_LIB_EXT1__], [__STDC_WANT_LIB_EXT1__], 1)])
+
+AS_IF([test "$ac_cv_func_getcwd" = yes], [
+    AC_CACHE_CHECK(if getcwd allocates buffer if NULL is given, [rb_cv_getcwd_malloc],
+	[AC_TRY_RUN([
+@%:@include <stddef.h>
+@%:@include <stdio.h>
+@%:@ifdef HAVE_UNISTD_H
+@%:@include <unistd.h>
+@%:@endif
+@%:@ifndef EXIT_SUCCESS
+@%:@define EXIT_SUCCESS 0
+@%:@endif
+@%:@ifndef EXIT_FAILURE
+@%:@define EXIT_FAILURE 1
+@%:@endif
+
+int
+main(int argc, char **argv)
+{
+    if (!getcwd(NULL, 0)) return EXIT_FAILURE;
+    return EXIT_SUCCESS;
+}
+],
+	    rb_cv_getcwd_malloc=yes,
+	    rb_cv_getcwd_malloc=no,
+	    AS_CASE($target_os,
+		[linux*|darwin*|*bsd|cygwin*|mingw*|mswin*],
+		[rb_cv_getcwd_malloc=yes],
+		[rb_cv_getcwd_malloc=no]))])
+    AS_IF([test "$rb_cv_getcwd_malloc" = no], [AC_DEFINE(NO_GETCWD_MALLOC, 1)])
+])
+
+AS_IF([test "$ac_cv_func_crypt_r" = yes],
+    [AC_CHECK_HEADERS(crypt.h)])
+AS_IF([test "$ac_cv_func_crypt_r:$ac_cv_header_crypt_h" = yes:yes],
+    [AC_CHECK_MEMBERS([struct crypt_data.initialized], [], [],
+		      [AC_INCLUDES_DEFAULT([@%:@include <crypt.h>])])])
+
+RUBY_CHECK_BUILTIN_FUNC(__builtin_alloca_with_align, [__builtin_alloca_with_align(1, 4096)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_assume_aligned, [__builtin_assume_aligned((void*)32, 32)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_bswap16, [__builtin_bswap16(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_bswap32, [__builtin_bswap32(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_bswap64, [__builtin_bswap64(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_popcount, [__builtin_popcount(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_popcountll, [__builtin_popcountll(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_clz, [__builtin_clz(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_clzl, [__builtin_clzl(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_clzll, [__builtin_clzll(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_ctz, [__builtin_ctz(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_ctzll, [__builtin_ctzll(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_add_overflow, [int x;__builtin_add_overflow(0,0,&x)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_sub_overflow, [int x;__builtin_sub_overflow(0,0,&x)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_mul_overflow, [int x;__builtin_mul_overflow(0,0,&x)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_mul_overflow_p, [__builtin_mul_overflow_p(0,0,(int)0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_constant_p, [__builtin_constant_p(0)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_choose_expr, [
+    [int x[__extension__(__builtin_choose_expr(1, 1, -1))]];
+    [int y[__extension__(__builtin_choose_expr(0, -1, 1))]];
+    ])
+AS_IF([test x$rb_cv_builtin___builtin_choose_expr = xyes], [
+    RUBY_CHECK_BUILTIN_FUNC(__builtin_choose_expr_constant_p, [
+    [int x[__extension__(__builtin_choose_expr(__builtin_constant_p(1), 1, -1))]];
+    [int y[__extension__(__builtin_choose_expr(__builtin_constant_p(foo), -1, 1))]];
+    ])
+])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_types_compatible_p, [__builtin_types_compatible_p(int, int)])
+RUBY_CHECK_BUILTIN_FUNC(__builtin_trap, [__builtin_trap()])
+
+AS_IF([test "$ac_cv_func_qsort_r" != no], [
+  AC_CACHE_CHECK(whether qsort_r is GNU version, rb_cv_gnu_qsort_r,
+    [AC_TRY_COMPILE([
+@%:@include <stdlib.h>
+void (qsort_r)(void *base, size_t nmemb, size_t size,
+	    int (*compar)(const void *, const void *, void *),
+	    void *arg);
+],[ ],
+      [rb_cv_gnu_qsort_r=yes],
+      [rb_cv_gnu_qsort_r=no])
+  ])
+  AC_CACHE_CHECK(whether qsort_r is BSD version, rb_cv_bsd_qsort_r,
+    [AC_TRY_COMPILE([
+@%:@include <stdlib.h>
+void (qsort_r)(void *base, size_t nmemb, size_t size,
+	     void *arg, int (*compar)(void *, const void *, const void *));
+],[ ],
+      [rb_cv_bsd_qsort_r=yes],
+      [rb_cv_bsd_qsort_r=no])
+  ])
+  AS_CASE("$rb_cv_gnu_qsort_r:$rb_cv_bsd_qsort_r",
+  [yes:no], [
+    AC_DEFINE(HAVE_GNU_QSORT_R, 1)
+  ],
+  [no:yes], [
+    AC_DEFINE(HAVE_BSD_QSORT_R, 1)
+  ])
+])
+
+AC_CACHE_CHECK(whether atan2 handles Inf as C99, rb_cv_atan2_inf_c99, [
+    AS_IF([test $ac_cv_func_atan2f:$ac_cv_func_atan2l = yes:yes], [
+	AC_TRY_RUN([
+@%:@include <math.h>
+@%:@ifdef HAVE_UNISTD_H
+@%:@include <unistd.h>
+@%:@endif
+@%:@ifndef EXIT_SUCCESS
+@%:@define EXIT_SUCCESS 0
+@%:@endif
+@%:@ifndef EXIT_FAILURE
+@%:@define EXIT_FAILURE 1
+@%:@endif
+
+int
+main(int argc, char **argv)
+{
+    if (fabs(atan2(INFINITY, INFINITY) - M_PI_4) <= 0.01) return EXIT_SUCCESS;
+    return EXIT_FAILURE;
+}
+],
+	[rb_cv_atan2_inf_c99=yes],
+	[rb_cv_atan2_inf_c99=no],
+	[AS_CASE($target_os, [mingw*|mswin*], [rb_cv_atan2_inf_c99=no], [rb_cv_atan2_inf_c99=yes])]
+	)
+    ], [rb_cv_atan2_inf_c99=no])
+])
+AS_IF([test "x$rb_cv_atan2_inf_c99" = xyes], [AC_DEFINE(ATAN2_INF_C99)])
+
+# Some platform need -lrt for clock_gettime, but the other don't.
+AS_IF([test x"$ac_cv_func_clock_gettime" != xyes], [
+    # glibc 2.17 moves clock_* functions from librt to the main C library.
+    # http://sourceware.org/ml/libc-announce/2012/msg00001.html
+    AC_CHECK_LIB(rt, clock_gettime)
+    AS_IF([test x"$ac_cv_lib_rt_clock_gettime" = xyes], [
+	AC_DEFINE(HAVE_CLOCK_GETTIME, 1)
+    ])
+])
+AC_CHECK_FUNCS(clock_getres) # clock_getres should be tested after clock_gettime test including librt test.
+AC_CHECK_LIB([rt], [timer_create])
+AC_CHECK_LIB([rt], [timer_settime])
+AS_IF([test x"$ac_cv_lib_rt_timer_create" = xyes], [
+    AC_DEFINE(HAVE_TIMER_CREATE, 1)
+])
+AS_IF([test x"$ac_cv_lib_rt_timer_settime" = xyes], [
+    AC_DEFINE(HAVE_TIMER_SETTIME, 1)
+])
+
+AC_CACHE_CHECK(for unsetenv returns a value, rb_cv_unsetenv_return_value,
+  [AC_TRY_COMPILE([
+#include <stdlib.h>
+], [int v = unsetenv("foo");],
+	rb_cv_unsetenv_return_value=yes,
+	rb_cv_unsetenv_return_value=no)])
+AS_IF([test "$rb_cv_unsetenv_return_value" = no], [
+  AC_DEFINE(VOID_UNSETENV)
+])
+
+# End of setjmp check.
+
+AC_ARG_ENABLE(setreuid,
+       AS_HELP_STRING([--enable-setreuid], [use setreuid()/setregid() according to need even if obsolete]),
+       [use_setreuid=$enableval])
+AS_IF([test "$use_setreuid" = yes], [
+    AC_DEFINE(USE_SETREUID)
+    AC_DEFINE(USE_SETREGID)
+])
+AC_STRUCT_TIMEZONE
+AC_CACHE_CHECK(for struct tm.tm_gmtoff, rb_cv_member_struct_tm_tm_gmtoff,
+  [AC_TRY_COMPILE([
+@%:@define _BSD_SOURCE
+@%:@define _DEFAULT_SOURCE
+@%:@include <time.h>
+   ],
+    [struct tm t; t.tm_gmtoff = 3600;],
+  [rb_cv_member_struct_tm_tm_gmtoff=yes],
+  [rb_cv_member_struct_tm_tm_gmtoff=no])])
+AS_IF([test "$rb_cv_member_struct_tm_tm_gmtoff" = yes], [
+  AC_DEFINE(HAVE_STRUCT_TM_TM_GMTOFF)
+])
+AC_CACHE_CHECK(for external int daylight, rb_cv_have_daylight,
+  [AC_TRY_LINK([#include <time.h>
+  int i;],
+	[i = daylight;],
+	rb_cv_have_daylight=yes,
+	rb_cv_have_daylight=no)])
+AS_IF([test "$rb_cv_have_daylight" = yes], [
+  AC_DEFINE(HAVE_DAYLIGHT)
+])
+
+AC_CACHE_CHECK(for negative time_t for gmtime(3), rb_cv_negative_time_t,
+  [AC_TRY_RUN([
+#include <stdlib.h>
+#include <time.h>
+
+void
+check(tm, y, m, d, h, s)
+    struct tm *tm;
+    int y, m, d, h, s;
+{
+    if (!tm ||
+	tm->tm_year != y ||
+	tm->tm_mon  != m-1 ||
+	tm->tm_mday != d ||
+	tm->tm_hour != h ||
+	tm->tm_sec  != s) {
+	exit(1);
+    }
+}
+
+int
+main()
+{
+   time_t t = -1;
+   struct tm *tm;
+
+   check(gmtime(&t), 69, 12, 31, 23, 59);
+   t = ~(time_t)0 << 31;
+   check(gmtime(&t), 1, 12, 13, 20, 52);
+   return 0;
+}
+],
+	rb_cv_negative_time_t=yes,
+	rb_cv_negative_time_t=no,
+	rb_cv_negative_time_t=yes)])
+AS_IF([test "$rb_cv_negative_time_t" = yes], [
+  AC_DEFINE(NEGATIVE_TIME_T)
+])
+
+# [ruby-dev:40910] overflow of time on FreeBSD
+# http://www.freebsd.org/cgi/query-pr.cgi?pr=145341
+AC_CACHE_CHECK(for localtime(3) overflow correctly, rb_cv_localtime_overflow,
+  [AC_TRY_RUN([
+#include <stdlib.h>
+#include <time.h>
+
+void
+check(time_t t1)
+{
+    struct tm *tm;
+    time_t t2;
+    tm = localtime(&t1);
+    if (!tm)
+	return; /* overflow detected.  ok. */
+    t2 = mktime(tm);
+    if (t1 == t2)
+        return; /* round-trip.  ok. */
+    exit(1);
+}
+
+int
+main()
+{
+    time_t t;
+    if (~(time_t)0 <= 0) {
+        t = (((time_t)1) << (sizeof(time_t) * 8 - 2));
+        t |= t - 1;
+    }
+    else {
+        t = ~(time_t)0;
+    }
+    check(t);
+    return 0;
+}
+],
+	rb_cv_localtime_overflow=yes,
+	rb_cv_localtime_overflow=no,
+	rb_cv_localtime_overflow=no)])
+AS_IF([test "$rb_cv_localtime_overflow" = no], [
+  AC_DEFINE(LOCALTIME_OVERFLOW_PROBLEM)
+])
+
+AS_IF([test "$ac_cv_func_sigprocmask" = yes && test "$ac_cv_func_sigaction" = yes], [
+   AC_DEFINE(POSIX_SIGNAL)
+], [
+  AC_CHECK_FUNCS(sigsetmask)
+  AC_CACHE_CHECK(for BSD signal semantics, rb_cv_bsd_signal,
+    [AC_TRY_RUN([
+#include <stdio.h>
+#include <signal.h>
+
+void
+sig_handler(dummy)
+     int dummy;
+{
+}
+
+int
+main()
+{
+  signal(SIGINT, sig_handler);
+  kill(getpid(), SIGINT);
+  kill(getpid(), SIGINT);
+  return 0;
+}
+],
+	rb_cv_bsd_signal=yes,
+	rb_cv_bsd_signal=no,
+	rb_cv_bsd_signal=$ac_cv_func_sigsetmask)])
+  AS_IF([test "$rb_cv_bsd_signal" = yes], [
+    AC_DEFINE(BSD_SIGNAL)
+  ])
+])
+
+AC_CHECK_TYPES([sig_t],[],[],[@%:@include <signal.h>])
+
+AS_IF([test "$ac_cv_func_getpgid" = no], [
+  # AC_FUNC_GETPGRP fails when cross-compiling with old autoconf.
+  # autoconf is changed between 2.52d and 2.52f?
+  # http://lists.gnu.org/archive/html/bug-gnu-utils/2001-09/msg00181.html
+  # "autoconf cleanup for AC_FUNC_GETPGRP and GETPGRP_VOID"
+AC_FUNC_GETPGRP
+])
+AS_IF([test "$ac_cv_func_setpgid:$ac_cv_func_setpgrp" = no:yes], [
+  # AC_FUNC_SETPGRP fails when cross-compiling.  (until autoconf 2.69?)
+  # https://lists.gnu.org/archive/html/bug-autoconf/2013-02/msg00002.html
+  # "AC_FUNC_SETPGRP fails to work properly when cross-compiling"
+AC_FUNC_SETPGRP
+])
+
+AS_IF([test x"$ac_cv_func_dirfd" = xno], [
+  AS_CASE(["$target_os"],[solaris*],
+          [AC_CHECK_MEMBERS([DIR.d_fd, DIR.dd_fd],,,[
+#include <sys/types.h>
+#include <dirent.h>
+])])
+])
+
+AS_IF([test x"$target_cpu" = xia64], [
+    AC_LIBOBJ([ia64])
+    AC_CACHE_CHECK(for __libc_ia64_register_backing_store_base,
+                   rb_cv___libc_ia64_register_backing_store_base,
+    [rb_cv___libc_ia64_register_backing_store_base=no
+    AC_TRY_LINK(
+      [extern unsigned long __libc_ia64_register_backing_store_base;],
+      [unsigned long p = __libc_ia64_register_backing_store_base;
+       printf("%ld\n", p);],
+      [rb_cv___libc_ia64_register_backing_store_base=yes])])
+    AS_IF([test $rb_cv___libc_ia64_register_backing_store_base = yes], [
+      AC_DEFINE(HAVE___LIBC_IA64_REGISTER_BACKING_STORE_BASE)
+    ])
+])
+
+AC_CACHE_CHECK(whether right shift preserve sign bit, rb_cv_rshift_sign,
+    [AC_COMPILE_IFELSE([AC_LANG_BOOL_COMPILE_TRY([], [(-1==(-1>>1))])],
+	rb_cv_rshift_sign=yes,
+	rb_cv_rshift_sign=no)])
+AS_IF([test "$rb_cv_rshift_sign" = yes], [
+  AC_DEFINE(RSHIFT(x,y), ((x)>>(int)(y)))
+], [
+  AC_DEFINE(RSHIFT(x,y), (((x)<0) ? ~((~(x))>>(int)(y)) : (x)>>(int)(y)))
+])
+
+AS_CASE(["$ac_cv_func_gettimeofday:$ac_cv_func_clock_gettime"],
+[*yes*], [],
+[
+    AC_MSG_ERROR(clock_gettime() or gettimeofday() must exist)
+])
+
+AS_IF([test "$ac_cv_func_sysconf" = yes], [
+  RUBY_CHECK_SYSCONF(CLK_TCK)
+])
+
+AS_IF([test "${universal_binary-no}" = yes ], [
+    archflagpat=`eval echo '"'"${ARCH_FLAG}"'"' | sed 's/[[][|.*]]/\\&/g'`
+    save_CFLAGS="$CFLAGS" new_cflags=`echo "$CFLAGS" | sed "s|$archflagpat"'||'`
+    save_LDFLAGS="$LDFLAGS" new_ldflags=`echo "$LDFLAGS" | sed "s|$archflagpat"'||'`
+    stack_dir=
+    for archs in ${universal_archnames}; do
+	archs=`echo $archs | sed 's/=.*//'`
+	CFLAGS="$new_cflags -arch $archs"
+	LDFLAGS="$new_ldflags -arch $archs"
+	RUBY_STACK_GROW_DIRECTION($archs, dir)
+	AS_IF([test x$stack_dir = x], [
+	    stack_dir=$dir
+	], [test x$stack_dir != x$dir], [
+	    stack_dir=no
+	])
+    done
+    CFLAGS="$save_CFLAGS" LDFLAGS="$save_LDFLAGS"
+    AS_IF([test x$stack_dir = xno], [
+	for archs in ${universal_archnames}; do
+	    archs=`echo $archs | sed 's/=.*//'`
+	    eval dir=\$[rb_cv_stack_grow_dir_]AS_TR_SH([$archs])
+	    RUBY_DEFINE_IF([defined __${archs}__], STACK_GROW_DIRECTION, $dir)
+	done
+    ], [
+	AC_DEFINE_UNQUOTED(STACK_GROW_DIRECTION, $stack_dir)
+    ])
+], [
+    RUBY_STACK_GROW_DIRECTION($target_cpu, dir)
+    AC_DEFINE_UNQUOTED(STACK_GROW_DIRECTION, $dir)
+])
+
+AC_ARG_ENABLE(fiber-coroutine,
+    AS_HELP_STRING([--disable-fiber-coroutine], [disable native coroutine implementation for fiber]),
+    [rb_cv_fiber_coroutine=$enableval])
+AS_CASE(["$rb_cv_fiber_coroutine"], [yes|''], [
+    AC_MSG_CHECKING(native coroutine implementation for ${target_cpu}-${target_os})
+    AS_CASE(["$target_cpu-$target_os"],
+	[x*64-darwin*], [
+	    rb_cv_fiber_coroutine=amd64
+	],
+	[x*64-linux], [
+	    AS_CASE(["$ac_cv_sizeof_voidp"],
+	        [8], [ rb_cv_fiber_coroutine=amd64 ],
+	        [4], [ rb_cv_fiber_coroutine=x86 ],
+		[*], [ rb_cv_fiber_coroutine= ]
+	    )
+	],
+	[*86-linux], [
+	    rb_cv_fiber_coroutine=x86
+	],
+	[x64-mingw32], [
+	    rb_cv_fiber_coroutine=win64
+	],
+	[powerpc64le-linux], [
+	    rb_cv_fiber_coroutine=ppc64le
+	],
+	[*], [
+	    rb_cv_fiber_coroutine=
+	]
+    )
+    AC_MSG_RESULT(${rb_cv_fiber_coroutine:-no})
+])
+AS_IF([test "${rb_cv_fiber_coroutine:-no}" != no], [
+    COROUTINE_H=coroutine/$rb_cv_fiber_coroutine/Context.h
+    AC_DEFINE_UNQUOTED(FIBER_USE_COROUTINE, ["$COROUTINE_H"])
+    AC_SUBST(X_FIBER_COROUTINE_H, [$COROUTINE_H])
+])
+
+AS_IF([test x"$enable_pthread" = xyes], [
+    for pthread_lib in thr pthread pthreads c c_r root; do
+	AC_CHECK_LIB($pthread_lib, pthread_create,
+		     rb_with_pthread=yes, rb_with_pthread=no)
+	AS_IF([test "$rb_with_pthread" = "yes"], [break])
+    done
+    AS_IF([test x"$rb_with_pthread" = xyes], [
+	AC_DEFINE(_REENTRANT)
+	AC_DEFINE(_THREAD_SAFE)
+	AC_DEFINE(HAVE_LIBPTHREAD)
+	AC_CHECK_HEADERS(pthread_np.h, [], [], [@%:@include <pthread.h>])
+	AS_CASE(["$pthread_lib:$target_os"],
+		[c:*], [],
+		[root:*], [],
+		[c_r:*|*:openbsd*|*:mirbsd*],  [LIBS="-pthread $LIBS"],
+		[LIBS="-l$pthread_lib $LIBS"])
+    ], [
+	AC_MSG_WARN("Don't know how to find pthread library on your system -- thread support disabled")
+    ])
+    AC_CACHE_CHECK([whether pthread_t is scalar type], [rb_cv_scalar_pthread_t], [
+	AC_TRY_COMPILE([
+	    @%:@include <pthread.h>
+	    ], [
+	    pthread_t thread_id;
+	    thread_id = 0;
+	    if (!thread_id) return 0;
+	    ], [rb_cv_scalar_pthread_t=yes], [rb_cv_scalar_pthread_t=no])
+    ])
+    AS_IF([test x"$rb_cv_scalar_pthread_t" = xyes], [
+	: # RUBY_CHECK_SIZEOF(pthread_t, [void* int long], [], [@%:@include <pthread.h>])
+    ], [
+	AC_DEFINE(NON_SCALAR_THREAD_ID)
+    ])
+    AC_CHECK_FUNCS(sched_yield pthread_attr_setinheritsched \
+	pthread_attr_get_np pthread_attr_getstack pthread_attr_getguardsize \
+	pthread_get_stackaddr_np pthread_get_stacksize_np \
+	thr_stksegment pthread_stackseg_np pthread_getthrds_np \
+	pthread_condattr_setclock \
+	pthread_sigmask pthread_setname_np pthread_set_name_np)
+    AS_CASE(["$target_os"],[aix*],[ac_cv_func_pthread_getattr_np=no],[AC_CHECK_FUNCS(pthread_getattr_np)])
+    set_current_thread_name=
+    AS_IF([test "$ac_cv_func_pthread_setname_np" = yes], [
+	AC_CACHE_CHECK([arguments of pthread_setname_np], [rb_cv_func_pthread_setname_np_arguments],
+	    [rb_cv_func_pthread_setname_np_arguments=
+	    # Linux,AIX,  (pthread_self(), name)
+	    # NetBSD (pthread_self(), name, \"%s\")
+	    # Darwin (name)
+	    for mac in \
+		"(pthread_self(), name)" \
+		"(pthread_self(), name, \"%s\")" \
+		"(name)" \
+		; do
+		AC_TRY_COMPILE([
+		    @%:@include <pthread.h>
+		    @%:@ifdef HAVE_PTHREAD_NP_H
+		    @%:@include <pthread_np.h>
+		    @%:@endif
+		    @%:@define SET_THREAD_NAME(name) pthread_setname_np${mac}
+		    ],
+		    [if (SET_THREAD_NAME("conftest")) return 1;],
+		    [rb_cv_func_pthread_setname_np_arguments="${mac}"
+		    break])
+	    done
+	    ]
+	)
+	AS_IF([test -n "${rb_cv_func_pthread_setname_np_arguments}"], [
+	    set_current_thread_name="pthread_setname_np${rb_cv_func_pthread_setname_np_arguments}"
+	])
+    ], [test "$ac_cv_func_pthread_set_name_np" = yes], [
+	set_current_thread_name="pthread_set_name_np(pthread_self(), name)"
+    ])
+    AS_IF([test -n "$set_current_thread_name"], [
+	AC_DEFINE_UNQUOTED(SET_CURRENT_THREAD_NAME(name), $set_current_thread_name)
+	AS_CASE([$set_current_thread_name],
+	    [*'pthread_self()'*], [
+		set_another_thread_name=`echo "$set_current_thread_name" | sed 's/pthread_self()/thid/'`
+		AC_DEFINE_UNQUOTED(SET_ANOTHER_THREAD_NAME(thid,name), $set_another_thread_name)
+	    ])
+    ])
+])
+
+AS_IF([test x"$ac_cv_header_ucontext_h" = xno], [
+    AC_CACHE_CHECK([if signal.h defines ucontext_t], [rb_cv_ucontext_in_signal_h],
+	[AC_TRY_COMPILE([@%:@include <signal.h>],
+	[size_t size = sizeof(ucontext_t);],
+	[rb_cv_ucontext_in_signal_h=yes], [rb_cv_ucontext_in_signal_h=no])])
+    AS_IF([test x"$rb_cv_ucontext_in_signal_h" = xyes], [
+	    AC_DEFINE_UNQUOTED(UCONTEXT_IN_SIGNAL_H, 1)
+    ])
+])
+AS_IF([test x"$ac_cv_header_ucontext_h" = xyes -o x"$rb_cv_ucontext_in_signal_h" = xyes], [
+    AC_CACHE_CHECK([if mcontext_t is a pointer], [rb_cv_mcontext_t_ptr],
+	[AC_TRY_COMPILE([
+	    @%:@include <signal.h>
+	    @%:@ifdef HAVE_UCONTEXT_H
+	    @%:@include <ucontext.h>
+	    @%:@endif
+        mcontext_t test(mcontext_t mc) {return mc+1;}
+	],
+	[test(0);],
+	[rb_cv_mcontext_t_ptr=yes], [rb_cv_mcontext_t_ptr=no])])
+    AS_IF([test x"$rb_cv_mcontext_t_ptr" = xyes], [
+	AC_DEFINE_UNQUOTED(DEFINE_MCONTEXT_PTR(mc, uc), mcontext_t mc = (uc)->uc_mcontext)
+    ], [
+	AC_DEFINE_UNQUOTED(DEFINE_MCONTEXT_PTR(mc, uc), mcontext_t *mc = &(uc)->uc_mcontext)
+    ])
+    AS_IF([test x"$rb_with_pthread" = xyes], [
+	AC_CHECK_FUNCS(getcontext setcontext)
+    ])
+])
+
+AS_IF([test "$ac_cv_func_fork_works" = "yes" -a "$rb_with_pthread" = "yes"], [
+    AC_CACHE_CHECK([if fork works with pthread], rb_cv_fork_with_pthread,
+	[AC_TRY_RUN([
+#include <stdlib.h>
+#include <unistd.h>
+#include <pthread.h>
+#include <stdio.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <signal.h>
+#ifndef EXIT_SUCCESS
+#define EXIT_SUCCESS 0
+#endif
+#ifndef EXIT_FAILURE
+#define EXIT_FAILURE 1
+#endif
+
+void *
+thread_func(void *dmy)
+{
+    return dmy;
+}
+
+int
+use_threads(void)
+{
+    pthread_t tid;
+    if (pthread_create(&tid, 0, thread_func, 0) != 0) {
+	return -1;
+    }
+    if (pthread_join(tid, 0) != 0) {
+	return -1;
+    }
+    return 0;
+}
+
+int
+main(int argc, char *argv[])
+{
+    pid_t pid;
+    if (use_threads()) return EXIT_FAILURE;
+    pid = fork();
+
+    if (pid) {
+	int loc;
+	sleep(1);
+	if (waitpid(pid, &loc, WNOHANG) == 0) {
+	    kill(pid, SIGKILL);
+	    return EXIT_FAILURE;
+	}
+        if (!WIFEXITED(loc) || WEXITSTATUS(loc) != EXIT_SUCCESS)
+           return EXIT_FAILURE;
+    }
+    else {
+	if (use_threads()) return EXIT_FAILURE;
+    }
+
+    return EXIT_SUCCESS;
+}],
+	rb_cv_fork_with_pthread=yes,
+	rb_cv_fork_with_pthread=no,
+	rb_cv_fork_with_pthread=yes)])
+    test x$rb_cv_fork_with_pthread = xyes || AC_DEFINE(CANNOT_FORK_WITH_PTHREAD)
+])
+}
+
+: "runtime section" && {
+dnl wheather use dln_a_out or not
+AC_ARG_WITH(dln-a-out,
+	AS_HELP_STRING([--with-dln-a-out], [use dln_a_out if possible]),
+	[
+	AS_CASE([$withval],
+	[yes], [
+	    AS_IF([test "$enable_shared" = yes], [
+		AC_MSG_ERROR(dln_a_out can not make shared library)
+	    ])
+	    with_dln_a_out=yes],
+	[
+	    with_dln_a_out=no])], [with_dln_a_out=no])
+
+AC_CACHE_CHECK(whether ELF binaries are produced, rb_cv_binary_elf,
+[AC_TRY_LINK([],[], [
+AS_CASE(["`head -1 conftest$EXEEXT | tr -dc '\177ELF' | tr '\177' .`"],
+[.ELF*], [rb_cv_binary_elf=yes], [rb_cv_binary_elf=no])],
+rb_cv_binary_elf=no)])
+
+AS_IF([test "$rb_cv_binary_elf" = yes], [
+  AC_DEFINE(USE_ELF)
+  AS_IF([test "$with_dln_a_out" = yes], [
+    AC_MSG_ERROR(dln_a_out does not work with ELF)
+  ])
+  AC_CHECK_HEADERS([elf.h elf_abi.h])
+  AS_IF([test $ac_cv_header_elf_h = yes -o $ac_cv_header_elf_abi_h = yes], [
+    AC_LIBOBJ([addr2line])
+    AS_IF([test "x$compress_debug_sections" = xzlib], [
+      AC_CHECK_LIB([z], [uncompress])
+    ])
+  ])
+])
+
+AC_CHECK_HEADERS([mach-o/loader.h])
+AS_IF([test "$ac_cv_header_mach_o_loader_h" = yes], [
+  AC_LIBOBJ([addr2line])
+])
+
+AS_CASE(["$target_os"],
+[linux* | gnu* | k*bsd*-gnu | bsdi* | kopensolaris*-gnu], [
+    AS_IF([test "$rb_cv_binary_elf" = no], [
+	with_dln_a_out=yes
+    ], [
+	LDFLAGS="$LDFLAGS -rdynamic"
+    ])])
+LIBEXT=a
+
+AC_SUBST(DLDFLAGS)dnl
+AC_SUBST(ARCH_FLAG)dnl
+AC_SUBST(MJIT_HEADER_FLAGS)dnl
+AC_SUBST(MJIT_HEADER_INSTALL_DIR)dnl
+AC_SUBST(MJIT_CC)dnl
+AS_IF([test "$GCC" = "yes"], [
+    AS_CASE(["$target_os"],[aix*],[mjit_std_cflag="-std=gnu99"])
+])
+AC_SUBST(MJIT_CFLAGS, [${MJIT_CFLAGS-"-w ${mjit_std_cflag} ${orig_cflags}"}])dnl
+AC_SUBST(MJIT_OPTFLAGS, [${MJIT_OPTFLAGS-'$(optflags)'}])dnl
+AC_SUBST(MJIT_DEBUGFLAGS, [${MJIT_DEBUGFLAGS-'$(debugflags)'}])dnl
+AC_SUBST(MJIT_LDSHARED)dnl
+
+AC_SUBST(STATIC)dnl
+AC_SUBST(CCDLFLAGS)dnl
+AC_SUBST(LDSHARED)dnl
+AC_SUBST(LDSHAREDXX)dnl
+AC_SUBST(DLEXT)dnl
+AC_SUBST(DLEXT2)dnl
+AC_SUBST(LIBEXT)dnl
+AC_SUBST(ASMEXT, S)dnl
+
+STATIC=
+
+AS_IF([test "$with_dln_a_out" != yes], [
+  rb_cv_dlopen=unknown
+  AC_MSG_CHECKING(whether OS depend dynamic link works)
+  AS_IF([test "$GCC" = yes], [
+    AS_CASE(["$target_os"],
+    [darwin*], [
+      # The -fno-common is needed if we wish to embed the Ruby interpreter
+      # into a plugin module of some project (as opposed to embedding it
+      # within the project's application).  The -I/usr/local/include is
+      # needed because CPP as discovered by configure (cc -E -traditional)
+      # fails to consult /usr/local/include by default.  This causes
+      # mkmf.rb's have_header() to fail if the desired resource happens to be
+      # installed in the /usr/local tree.
+      RUBY_APPEND_OPTION(CCDLFLAGS, -fno-common)],
+    [bsdi*|cygwin*|mingw*|aix*|interix*], [ ],
+    [
+      RUBY_APPEND_OPTION(CCDLFLAGS, -fPIC)])
+  ], [
+    AS_CASE(["$target_os"],
+	[hpux*],          [CCDLFLAGS="$CCDLFLAGS +Z"],
+	[solaris*|irix*], [CCDLFLAGS="$CCDLFLAGS -KPIC"],
+	[sunos*],         [CCDLFLAGS="$CCDLFLAGS -PIC"],
+	[esix*|uxpds*],   [CCDLFLAGS="$CCDLFLAGS -KPIC"],
+	                  [: ${CCDLFLAGS=""}])
+  ])
+
+
+  AC_ARG_ENABLE(rpath,
+       AS_HELP_STRING([--enable-rpath], [embed run path into extension libraries.
+       enabled by default on ELF platforms]),
+       [enable_rpath=$enableval], [enable_rpath="$rb_cv_binary_elf"])
+
+  AS_CASE(["$target_os"],
+	[hpux*], [	DLDFLAGS="$DLDFLAGS -E"
+			: ${LDSHARED='$(LD) -b'}
+			XLDFLAGS="$XLDFLAGS -Wl,-E"
+			: ${LIBPATHENV=SHLIB_PATH}
+			rb_cv_dlopen=yes],
+	[solaris*], [	AS_IF([test "$GCC" = yes], [
+			    : ${LDSHARED='$(CC) -shared'}
+			    AS_IF([test "$rb_cv_prog_gnu_ld" = yes], [
+				LDFLAGS="$LDFLAGS -Wl,-E"
+			    ])
+			], [
+			    : ${LDSHARED='$(CC) -G'}
+			])
+			AS_IF([test "$ac_cv_sizeof_voidp" = 8], [
+			    : ${LIBPATHENV=LD_LIBRARY_PATH_64}
+			    : ${PRELOADENV=LD_PRELOAD_64}
+			], [
+			    : ${LIBPATHENV=LD_LIBRARY_PATH_32}
+			    : ${PRELOADENV=LD_PRELOAD_32}
+			])
+			rb_cv_dlopen=yes],
+	[sunos*], [	: ${LDSHARED='$(LD) -assert nodefinitions'}
+			rb_cv_dlopen=yes],
+	[irix*], [	: ${LDSHARED='$(LD) -shared'}
+			rb_cv_dlopen=yes],
+	[sysv4*], [	: ${LDSHARED='$(LD) -G'}
+			rb_cv_dlopen=yes],
+	[nto-qnx*], [	: ${LDSHARED='$(CC) -shared'}
+			rb_cv_dlopen=yes],
+	[esix*|uxpds*], [ : ${LDSHARED='$(LD) -G'}
+			rb_cv_dlopen=yes],
+	[osf*], [	: ${LDSHARED='$(LD) -shared -expect_unresolved "*"'}
+			rb_cv_dlopen=yes],
+	[bsdi3*], [	AS_CASE(["$CC"],
+			[*shlicc*], [	: ${LDSHARED='$(CC) -r'}
+					rb_cv_dlopen=yes])],
+	[linux* | gnu* | k*bsd*-gnu | netbsd* | bsdi* | kopensolaris*-gnu | haiku*], [
+			: ${LDSHARED='$(CC) -shared'}
+			AS_IF([test "$rb_cv_binary_elf" = yes], [
+			    LDFLAGS="$LDFLAGS -Wl,-export-dynamic"
+			])
+			rb_cv_dlopen=yes],
+	[interix*], [	: ${LDSHARED='$(CC) -shared'}
+			XLDFLAGS="$XLDFLAGS -Wl,-E"
+			LIBPATHFLAG=" -L%1\$-s"
+			rb_cv_dlopen=yes],
+	[freebsd*|dragonfly*], [
+			: ${LDSHARED='$(CC) -shared'}
+			AS_IF([test "$rb_cv_binary_elf" = yes], [
+			    LDFLAGS="$LDFLAGS -rdynamic"
+			    DLDFLAGS="$DLDFLAGS "'-Wl,-soname,$@'
+			], [
+			  test "$GCC" = yes && test "$rb_cv_prog_gnu_ld" = yes || LDSHARED='$(LD) -Bshareable'
+			])
+			rb_cv_dlopen=yes],
+	[openbsd*|mirbsd*], [	: ${LDSHARED='$(CC) -shared ${CCDLFLAGS}'}
+			AS_IF([test "$rb_cv_binary_elf" = yes], [
+			    LDFLAGS="$LDFLAGS -Wl,-E"
+			])
+			rb_cv_dlopen=yes],
+	[darwin*], [	: ${LDSHARED='$(CC) -dynamic -bundle'}
+			: ${DLDSHARED='$(CC) -dynamiclib'}
+			: ${LDFLAGS=""}
+			: ${LIBPATHENV=DYLD_FALLBACK_LIBRARY_PATH}
+			: ${PRELOADENV=DYLD_INSERT_LIBRARIES}
+			rb_cv_dlopen=yes],
+        [aix*], [	: ${LDSHARED='$(CC)'}
+			AS_IF([test "$GCC" = yes], [
+			    LDSHARED="$LDSHARED ${linker_flag}-G -shared"
+			], [
+			    LDSHARED="$LDSHARED ${linker_flag}-G"
+			])
+			EXTDLDFLAGS='-e$(TARGET_ENTRY)'
+			XLDFLAGS="${linker_flag}"'-bE:$(ARCHFILE)'" ${linker_flag}-brtl"
+			XLDFLAGS="$XLDFLAGS ${linker_flag}-blibpath:${prefix}/lib:${LIBPATH:-/usr/lib:/lib}"
+			: ${ARCHFILE="ruby.imp"}
+                        TRY_LINK='$(CC) -oconftest $(INCFLAGS) -I$(hdrdir) $(CPPFLAGS)'
+                        TRY_LINK="$TRY_LINK"' $(CFLAGS) $(src) $(LIBPATH) $(LDFLAGS) $(LOCAL_LIBS) $(LIBS)'
+			: ${LIBPATHENV=LIBPATH}
+			: ${PRELOADENV=LDR_PRELOAD}
+			rb_cv_dlopen=yes],
+	[nto-qnx*], [	DLDFLAGS="$DLDFLAGS -L/lib -L/usr/lib -L/usr/local/lib"
+			: ${LDSHARED='$(LD) -Bshareable -x'}
+			LDFLAGS="$LDFLAGS -L/lib -L/usr/lib -L/usr/local/lib"
+			rb_cv_dlopen=yes],
+	[cygwin*|mingw*], [
+			: ${LDSHARED='$(CC) -shared'}
+			XLDFLAGS="$XLDFLAGS -Wl,--stack,0x00200000,--enable-auto-import"
+			DLDFLAGS="${DLDFLAGS} -Wl,--enable-auto-image-base,--enable-auto-import"
+			: ${LIBPATHENV=PATH}
+			: ${PRELOADENV=""}
+			rb_cv_dlopen=yes],
+	[hiuxmpp], [	: ${LDSHARED='$(LD) -r'}],
+	[atheos*], [	: ${LDSHARED='$(CC) -shared'}
+			rb_cv_dlopen=yes],
+	[	: ${LDSHARED='$(LD)'}])
+  AC_MSG_RESULT($rb_cv_dlopen)
+
+  AS_IF([test "$rb_cv_dlopen" = yes], [
+    AS_CASE(["$target_os"],
+      [darwin*], [
+	for flag in \
+	  "-undefined dynamic_lookup" \
+	  "-multiply_defined suppress" \
+	  ; do
+	  test "x${linker_flag}" = x || flag="${linker_flag}`echo ${flag} | tr ' ' ,`"
+	  RUBY_TRY_LDFLAGS([$flag], [], [flag=])
+	  AS_IF([test "x$flag" != x], [
+	    RUBY_APPEND_OPTIONS(DLDFLAGS, [$flag])
+	  ])
+	done
+      ])
+  ])
+
+  AS_IF([test "$enable_rpath:${RPATHFLAG}" = yes:], [
+      AS_IF([test "x$rpathflag" != x], [
+	  RPATHFLAG=" ${rpathflag}%1\$-s"
+      ])
+  ])
+])
+AS_IF([test "${LDSHAREDXX}" = ""], [
+    AS_CASE(["${LDSHARED}"],
+	[*'$(CC)'*], [
+	    LDSHAREDXX=`echo "${LDSHARED}" | sed 's/\$(CC)/$(CXX)/'`
+	    ],
+	[*'${CC}'*], [
+	    LDSHAREDXX=`echo "${LDSHARED}" | sed 's/\${CC}/${CXX}/'`
+	    ],
+	[*$CC*], [
+	    LDSHAREDXX=`echo "${LDSHARED}" | sed "s|$CC|$CXX|"`
+	    ],
+	[ld" "*], [
+	    ])
+])
+AS_CASE([${RPATHFLAG}],[*'%1$'*],[: ${LIBPATHFLAG=' -L%1$-s'}],[: ${LIBPATHFLAG=' -L%s'}])
+
+AC_SUBST(LINK_SO)
+AC_SUBST(LIBPATHFLAG)
+AC_SUBST(RPATHFLAG)
+AC_SUBST(LIBPATHENV, "${LIBPATHENV-LD_LIBRARY_PATH}")
+AC_SUBST(PRELOADENV, "${PRELOADENV-LD_PRELOAD}")
+AC_SUBST(TRY_LINK)
+
+AS_IF([test "x$OPT_DIR" != x], [
+    pat=`echo "${LDFLAGS_OPTDIR}" | sed ['s/[][\\.*|]/\\\\&/']`
+    LDFLAGS=`echo "${LDFLAGS}" | sed "s| ${pat}||"`
+    val=`IFS="$PATH_SEPARATOR"
+        for dir in $OPT_DIR; do
+            echo x ${LIBPATHFLAG} ${RPATHFLAG} |
+            sed "s/^x *//;s${IFS}"'%1\\$-s'"${IFS}${dir}/lib${IFS}g;s${IFS}%s${IFS}${dir}/lib${IFS}g"
+        done | tr '\012' ' ' | sed 's/ *$//'`
+    AS_IF([test x"$val" != x], [
+	test x"${LDFLAGS}" = x || LDFLAGS="$LDFLAGS "
+	LDFLAGS="$LDFLAGS$val"
+	test x"${DLDFLAGS}" = x || DLDFLAGS="$DLDFLAGS "
+	DLDFLAGS="$DLDFLAGS$val"
+    ])
+    LDFLAGS_OPTDIR="$val"
+])
+
+AS_CASE(["$target_os"],
+[freebsd*], [
+    AC_CHECK_LIB([procstat], [procstat_open_sysctl])
+    AS_IF([test "x$ac_cv_lib_procstat_procstat_open_sysctl" = xyes], [
+	AC_CHECK_FUNCS(procstat_getvmmap)
+    ])
+    ])
+AS_CASE(["$target_cpu-$target_os"],
+[*-darwin*], [
+    AC_CHECK_HEADERS([execinfo.h])
+    AS_IF([test "x$ac_cv_header_execinfo_h" = xyes], [
+	AC_CHECK_LIB([execinfo], [backtrace])
+	AC_CHECK_HEADERS([libunwind.h])
+    ])],
+[*-freebsd*|x86_64-netbsd*], [
+    AC_CHECK_HEADERS([execinfo.h])
+    AS_IF([test "x$ac_cv_header_execinfo_h" = xyes], [
+	AC_CHECK_LIB([execinfo], [backtrace])
+	AC_CHECK_LIB([unwind], [unw_backtrace])
+    ])])
+AC_CHECK_FUNCS(backtrace)
+
+AS_IF([test "x$ac_cv_func_backtrace" = xyes], [
+  AC_CACHE_CHECK(for broken backtrace, rb_cv_broken_backtrace,
+    [AC_TRY_RUN([
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <string.h>
+#include <signal.h>
+#include <execinfo.h>
+
+#define TRACE_SIZE 256
+
+void sigsegv(int signum, siginfo_t *info, void *ctx){
+    void *trace[TRACE_SIZE];
+    int n = backtrace(trace, TRACE_SIZE);
+    if (n > 0) {
+	/*fprintf(stdout, "backtrace:%d\n",n);*/
+    } else {
+	_exit(EXIT_FAILURE);
+    }
+    _exit(EXIT_SUCCESS);
+}
+int
+main(void)
+{
+    volatile int *a = NULL;
+    stack_t ss;
+    struct sigaction sa;
+
+    ss.ss_sp = malloc(SIGSTKSZ);
+    if (ss.ss_sp == NULL) {
+	fprintf(stderr, "cannot allocate memory for sigaltstack\n");
+	return EXIT_FAILURE;
+    }
+    ss.ss_size = SIGSTKSZ;
+    ss.ss_flags = 0;
+    if (sigaltstack(&ss, NULL) == -1) {
+	fprintf(stderr, "sigaltstack failed\n");
+	return EXIT_FAILURE;
+    }
+    memset(&sa, 0, sizeof(struct sigaction));
+    sigemptyset(&sa.sa_mask);
+    sa.sa_sigaction = sigsegv;
+    sa.sa_flags |= SA_SIGINFO;
+    sa.sa_flags |= SA_ONSTACK;
+    sigaction(SIGSEGV, &sa, NULL);
+    a[0] = 1;
+    return EXIT_SUCCESS;
+}
+],
+	rb_cv_broken_backtrace=no,
+	rb_cv_broken_backtrace=yes,
+	rb_cv_broken_backtrace=no)])
+  AS_IF([test "$rb_cv_broken_backtrace" = yes], [
+    AC_DEFINE(BROKEN_BACKTRACE, 1)
+  ])
+])
+
+AC_ARG_WITH(valgrind,
+        AS_HELP_STRING([--without-valgrind],[disable valgrind memcheck support]),
+        [], with_valgrind=yes)
+AS_IF([test x$with_valgrind != xno],
+        [AC_CHECK_HEADERS(valgrind/memcheck.h)])
+
+dln_a_out_works=no
+AS_IF([test "$ac_cv_header_a_out_h" = yes], [
+  AS_IF([test "$with_dln_a_out" = yes || test "$rb_cv_dlopen" = unknown], [
+    cat confdefs.h > config.h
+    AC_CACHE_CHECK(whether matz's dln works, rb_cv_dln_a_out,
+    [AC_TRY_COMPILE([
+#define USE_DLN_A_OUT
+#include "dln.c"
+],
+	[],
+	rb_cv_dln_a_out=yes,
+	rb_cv_dln_a_out=no)])
+    AS_IF([test "$rb_cv_dln_a_out" = yes], [
+      dln_a_out_works=yes
+      AC_DEFINE(USE_DLN_A_OUT)
+    ])
+  ])
+])
+
+AS_IF([test "$dln_a_out_works" = yes], [
+  AS_IF([test "$GCC" = yes], [
+    STATIC=-static
+  ], [
+    STATIC=-Bstatic
+  ])
+  DLEXT=so
+  CCDLFLAGS=
+], [
+  AS_CASE(["$target_os"],
+    [hpux*], [
+	DLEXT=sl],
+    [darwin*], [
+	SOEXT=dylib
+	DLEXT=bundle],
+    [cygwin*|mingw*|*djgpp*], [
+	LOAD_RELATIVE=1
+	SOEXT=dll
+	DLEXT=so],
+    [
+	DLEXT=so])
+])
+: ${SOEXT="${DLEXT}"}
+AC_SUBST(SOEXT)
+AS_IF([test "$rb_cv_dlopen:$load_relative" = yes:yes], [
+    AS_IF([test "$ac_cv_func_dladdr" = yes], [
+	LOAD_RELATIVE=1
+    ])
+])
+AS_IF([test x"$LOAD_RELATIVE" = x1], [
+    load_relative=yes
+], [
+    unset load_relative
+])
+
+len=2 # .rb
+n=`expr "$DLEXT"  : '.*'`; test "$n" -gt "$len" && len=$n
+n=`expr "$DLEXT2" : '.*'`; test "$n" -gt "$len" && len=$n
+AC_DEFINE_UNQUOTED(DLEXT_MAXLEN, `expr $len + 1`)
+test ".$DLEXT"  = "." || AC_DEFINE_UNQUOTED(DLEXT,  ".$DLEXT")
+test ".$DLEXT2" = "." || AC_DEFINE_UNQUOTED(DLEXT2, ".$DLEXT2")
+AC_SUBST(DLEXT)
+
+AS_IF([test "$with_dln_a_out" = yes], [
+  STRIP=true
+], [
+  AC_CHECK_TOOL(STRIP, strip, :)dnl
+])
+
+AS_CASE(["$target_os"],
+  [linux* | gnu* | k*bsd*-gnu | kopensolaris*-gnu], [
+	STRIP="$STRIP -S -x"],
+  [darwin*], [
+	STRIP="$STRIP -A -n"])
+
+AC_ARG_WITH(ext,
+            AC_HELP_STRING([--with-ext=EXTS],
+                           [pass to --with-ext option of extmk.rb]))
+AC_ARG_WITH(out-ext,
+            AC_HELP_STRING([--with-out-ext=EXTS],
+                           [pass to --without-ext option of extmk.rb]))
+EXTSTATIC=
+AC_SUBST(EXTSTATIC)dnl
+AC_ARG_WITH(static-linked-ext,
+	    AS_HELP_STRING([--with-static-linked-ext], [link external modules statically]),
+            [AS_CASE([$withval],[yes],[STATIC=;EXTSTATIC=static],[no],[],[EXTSTATIC="$withval"])])
+AS_CASE([",$EXTSTATIC,"], [,static,|*,enc,*], [
+  ENCOBJS='enc/encinit.$(OBJEXT) enc/libenc.$(LIBEXT) enc/libtrans.$(LIBEXT)'
+  EXTOBJS='ext/extinit.$(OBJEXT)'
+  AC_DEFINE_UNQUOTED(EXTSTATIC, 1)
+  AC_SUBST(ENCSTATIC, static)
+], [
+  ENCOBJS='dmyenc.$(OBJEXT)'
+  EXTOBJS='dmyext.$(OBJEXT)'
+])
+AC_SUBST(ENCOBJS)
+AC_SUBST(EXTOBJS)
+
+AC_ARG_WITH(setup,
+	    AS_HELP_STRING([--with-setup=SETUP], [use extension libraries setup]),
+	    [setup=$withval])
+AS_IF([test -n "$setup"], [
+    AS_IF([! test -f "ext/$setup" -o -f "$srcdir/ext/$setup"], [
+	AC_MSG_ERROR(Setup file $setup not found under ext or $srcdir/ext)
+    ])
+], [test -f "$srcdir/ext/Setup.$target_os"], [
+    setup="Setup.$target_os"
+], [
+    setup=
+    for file in "$srcdir"/ext/Setup.*; do
+	AS_CASE(["$file"], [*~|*.bak|*.orig|*.rej|*.tmp], [continue])
+	setup=`basename "$file"`
+	AS_CASE(["$target_os"], [`expr "$setup" : 'Setup.\(.*\)'`*], [break])
+	platform=`sed '/^option  *platform  */!d;s///;s/|/*|/g;q' "$file"`
+	AS_IF([test "x$platform" != x], [
+	    eval "AS_CASE([\"\$target_os\"], [$platform*], [break])"
+	])
+	setup=
+    done
+    : ${setup:=Setup}
+])
+AC_SUBST(setup)
+
+rubylibprefix='${libdir}/${RUBY_BASE_NAME}'
+AC_ARG_WITH(rubylibprefix,
+	    AS_HELP_STRING([--with-rubylibprefix=DIR], [prefix for ruby libraries [[LIBDIR/RUBY_BASE_NAME]]]),
+	    [AS_IF([test "x$withval" = xno], [
+		AC_MSG_ERROR([No ruby, No libprefix])
+	    ])
+	    rubylibprefix="$withval"])
+AC_SUBST(rubylibprefix)
+
+AS_IF([test x"${exec_prefix}" != xNONE], [
+    RUBY_EXEC_PREFIX="$exec_prefix"
+], [test x"$prefix" != xNONE], [
+    RUBY_EXEC_PREFIX="$prefix"
+], [
+    RUBY_EXEC_PREFIX=$ac_default_prefix
+])
+pat=`echo "${RUBY_EXEC_PREFIX}" | tr -c '\012' .`'\(.*\)'
+for var in bindir libdir rubylibprefix; do
+    eval val='"$'$var'"'
+    AS_CASE(["$val"], ["${RUBY_EXEC_PREFIX}"*], [val='${exec_prefix}'"`expr \"$val\" : \"$pat\"`"])
+    eval $var='"$val"'
+done
+
+BTESTRUBY='$(MINIRUBY)'
+AS_IF([test x"$cross_compiling" = xyes], [
+  test x"$MINIRUBY" = x && MINIRUBY="${RUBY-$BASERUBY} -I`$CHDIR .; pwd` "-r'$(arch)-fake'
+  XRUBY_LIBDIR=`${RUBY-$BASERUBY} -rrbconfig -e ['puts RbConfig::CONFIG["libdir"]']`
+  XRUBY_RUBYLIBDIR=`${RUBY-$BASERUBY} -rrbconfig -e ['puts RbConfig::CONFIG["rubylibdir"]']`
+  XRUBY_RUBYHDRDIR=`${RUBY-$BASERUBY} -rrbconfig -e ['puts RbConfig::CONFIG["rubyhdrdir"]']`
+  AC_SUBST(XRUBY_LIBDIR)
+  AC_SUBST(XRUBY_RUBYLIBDIR)
+  AC_SUBST(XRUBY_RUBYHDRDIR)
+  PREP='$(arch)-fake.rb'
+  RUNRUBY_COMMAND='$(MINIRUBY) -I`cd $(srcdir)/lib; pwd`'
+  RUNRUBY='$(RUNRUBY_COMMAND)'
+  XRUBY='$(MINIRUBY)'
+  BOOTSTRAPRUBY='$(BASERUBY)'
+  TEST_RUNNABLE=no
+  CROSS_COMPILING=yes
+], [
+  MINIRUBY='./miniruby$(EXEEXT) -I$(srcdir)/lib -I.'
+  MINIRUBY="$MINIRUBY"' -I$(EXTOUT)/common'
+  PREP='miniruby$(EXEEXT)'
+  RUNRUBY_COMMAND='$(MINIRUBY) $(srcdir)/tool/runruby.rb --extout=$(EXTOUT) $(RUNRUBYOPT)'
+  RUNRUBY='$(RUNRUBY_COMMAND) --'
+  XRUBY='$(RUNRUBY)'
+  BOOTSTRAPRUBY='$(MINIRUBY)'
+  TEST_RUNNABLE=yes
+  CROSS_COMPILING=no
+])
+AC_SUBST(TEST_RUNNABLE)
+AC_SUBST(CROSS_COMPILING)
+AC_SUBST(MINIRUBY)
+AC_SUBST(BTESTRUBY)
+AC_SUBST(PREP)
+AC_SUBST(RUNRUBY_COMMAND)
+AC_SUBST(RUNRUBY)
+AC_SUBST(XRUBY)
+AC_SUBST(BOOTSTRAPRUBY)
+AC_SUBST(EXTOUT, [${EXTOUT=.ext}])
+
+FIRSTMAKEFILE=""
+LIBRUBY_A='lib$(RUBY_SO_NAME)-static.a'
+LIBRUBY='$(LIBRUBY_A)'
+LIBRUBYARG_STATIC='-l$(RUBY_SO_NAME)-static'
+LIBRUBYARG='$(LIBRUBYARG_STATIC)'
+SOLIBS='$(MAINLIBS)'
+
+AS_CASE(["$target_os"],
+  [cygwin*|mingw*|haiku*|darwin*], [
+    : ${DLDLIBS=""}
+    ],
+  [
+    DLDLIBS="$DLDLIBS -lc"
+    ])
+
+AC_ARG_ENABLE(multiarch,
+	      AS_HELP_STRING([--enable-multiarch], [enable multiarch compatible directories]),
+	      [multiarch=], [unset multiarch])
+AS_IF([test ${multiarch+set}], [
+   AC_DEFINE(ENABLE_MULTIARCH)
+   MJIT_HEADER_INSTALL_DIR=include/'${arch}/${RUBY_VERSION_NAME}'
+], [
+   MJIT_HEADER_INSTALL_DIR=include/'${RUBY_VERSION_NAME}/${arch}'
+])
+
+archlibdir='${libdir}/${arch}'
+sitearchlibdir='${libdir}/${sitearch}'
+archincludedir='${includedir}/${arch}'
+sitearchincludedir='${includedir}/${sitearch}'
+
+AC_ARG_WITH(soname,
+	AS_HELP_STRING([--with-soname=SONAME], [base name of shared library]),
+	[RUBY_SO_NAME=$withval],
+	[
+	    AS_CASE(["$target_os"],
+	    [darwin*], [
+	        RUBY_SO_NAME='$(RUBY_BASE_NAME).$(RUBY_API_VERSION)'
+	    ],
+	    [cygwin*], [
+	        RUBY_SO_NAME='$(RUBY_BASE_NAME)$(MAJOR)$(MINOR)0'
+	    ],
+	    [mingw*], [
+		RUBY_SO_NAME="${rb_cv_msvcrt}"'-$(RUBY_BASE_NAME)$(MAJOR)$(MINOR)0'
+		AS_IF([test x"${target_cpu}" != xi386], [
+		    RUBY_SO_NAME="${target_cpu}-${RUBY_SO_NAME}"
+		])
+	    ],
+	    [RUBY_SO_NAME='$(RUBY_BASE_NAME)'])
+	])
+
+LIBRUBY_LDSHARED=${DLDSHARED=${LDSHARED}}
+LIBRUBY_DLDFLAGS=$DLDFLAGS
+LIBRUBY_SO='lib$(RUBY_SO_NAME).$(SOEXT).$(RUBY_PROGRAM_VERSION)'
+LIBRUBY_SONAME='lib$(RUBY_SO_NAME).$(SOEXT).$(RUBY_API_VERSION)'
+LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).$(SOEXT)'
+ENABLE_SHARED=no
+
+AC_ARG_ENABLE(shared,
+       AS_HELP_STRING([--enable-shared], [build a shared library for Ruby]),
+       [enable_shared=$enableval])
+libprefix=${multiarch+'$(archlibdir)'}${multiarch-'$(libdir)'}
+LIBRUBY_RELATIVE=${load_relative-no}
+AS_CASE("$enable_shared", [yes], [
+  LIBRUBY='$(LIBRUBY_SO)'
+  LIBRUBYARG_SHARED='-l$(RUBY_SO_NAME)'
+  LIBRUBYARG='$(LIBRUBYARG_SHARED)'
+  LIBRUBY_RELATIVE=no
+  test -z "$CCDLFLAGS" || CFLAGS="$CFLAGS $CCDLFLAGS"
+  ENABLE_SHARED=yes
+
+  # libdir can be overridden in config.site file (on OpenSUSE at least).
+  libdir_basename=lib
+  AS_IF([test "$bindir" = '${exec_prefix}/bin'], [
+    AS_CASE(["$libdir"], ['${exec_prefix}/'*], [libdir_basename=`basename "$libdir"`])
+  ])
+  AC_DEFINE_UNQUOTED(LIBDIR_BASENAME, ["${libdir_basename}"])
+  libdir_basename="${libdir_basename}"${multiarch+'/${arch}'}
+
+  AS_CASE(["$target_os"],
+    [freebsd*|dragonfly*], [],
+    [
+     AS_IF([test "$GCC" = yes], [
+       RUBY_TRY_LDFLAGS([${linker_flag}--no-undefined], [no_undefined=yes], [no_undefined=no])
+       AS_IF([test "no_undefined" = yes], [
+	  RUBY_APPEND_OPTION(EXTLDFLAGS, [${linker_flag}--no-undefined])
+       ])
+     ])
+    ])
+
+  AS_CASE(["$target_os"],
+    [sunos4*], [
+	LIBRUBY_ALIASES='$(LIBRUBY_SONAME) lib$(RUBY_SO_NAME).$(SOEXT)'
+	],
+    [linux* | gnu* | k*bsd*-gnu | atheos* | kopensolaris*-gnu | haiku*], [
+	RUBY_APPEND_OPTIONS(LIBRUBY_DLDFLAGS, ['-Wl,-soname,$(LIBRUBY_SONAME)' "$LDFLAGS_OPTDIR"])
+	LIBRUBY_ALIASES='$(LIBRUBY_SONAME) lib$(RUBY_SO_NAME).$(SOEXT)'
+	AS_IF([test "$load_relative" = yes], [
+	    libprefix="'\$\${ORIGIN}/../${libdir_basename}'"
+	    LIBRUBY_RPATHFLAGS="-Wl,-rpath,${libprefix}"
+	    LIBRUBY_RELATIVE=yes
+	])
+	],
+    [freebsd*|dragonfly*], [
+	LIBRUBY_SO='lib$(RUBY_SO_NAME).$(SOEXT).$(MAJOR)$(MINOR)'
+	LIBRUBY_SONAME='$(LIBRUBY_SO)'
+	AS_IF([test "$rb_cv_binary_elf" != "yes" ], [
+	    LIBRUBY_SO="$LIBRUBY_SO.\$(TEENY)"
+	    LIBRUBY_ALIASES=''
+	])
+	],
+    [netbsd*], [
+	LIBRUBY_SONAME='lib$(RUBY_SO_NAME).$(SOEXT).$(MAJOR)$(MINOR)'
+	LIBRUBY_SO="${LIBRUBY_SONAME}"'.$(TEENY)'
+	RUBY_APPEND_OPTIONS(LIBRUBY_DLDFLAGS, ['-Wl,-soname,$(LIBRUBY_SONAME)' "$LDFLAGS_OPTDIR"])
+	AS_IF([test "$rb_cv_binary_elf" = yes], [ # ELF platforms
+	   LIBRUBY_ALIASES='$(LIBRUBY_SONAME) lib$(RUBY_SO_NAME).$(SOEXT)'
+	], [	# a.out platforms
+	   LIBRUBY_ALIASES=""
+	])
+	],
+    [openbsd*|mirbsd*], [
+	LIBRUBY_SO='lib$(RUBY_SO_NAME).$(SOEXT).$(MAJOR).'`expr ${MINOR} \* 10 + ${TEENY}`
+	],
+    [solaris*], [
+	LIBRUBY_SO='lib$(RUBY_SO_NAME).$(SOEXT).$(MAJOR)'
+	LIBRUBY_SONAME='lib$(RUBY_SO_NAME).$(SOEXT).$(RUBY_PROGRAM_VERSION)'
+	LIBRUBY_ALIASES='$(LIBRUBY_SONAME) lib$(RUBY_SO_NAME).$(SOEXT)'
+	RUBY_APPEND_OPTIONS(LIBRUBY_DLDFLAGS, ["${linker_flag}-h${linker_flag:+,}"'$(@F)'])
+	XLDFLAGS="$XLDFLAGS "'-R${libdir}'
+	],
+    [hpux*], [
+	XLDFLAGS="$XLDFLAGS "'-Wl,+s,+b,$(libdir)'
+	LIBRUBY_ALIASES='$(LIBRUBY_SONAME) lib$(RUBY_SO_NAME).$(SOEXT)'
+	],
+    [aix*], [
+	RUBY_APPEND_OPTIONS(LIBRUBY_DLDFLAGS, ["${linker_flag}-bnoentry" "$XLDFLAGS" "$LDFLAGS_OPTDIR"])
+	LIBRUBYARG_SHARED='-L${libdir} -l${RUBY_SO_NAME}'
+	LIBS="$LIBS -lm -lc"
+	],
+    [darwin*], [
+	LIBRUBY_SO='lib$(RUBY_SO_NAME).$(SOEXT)'
+	LIBRUBY_SONAME='$(LIBRUBY_SO)'
+	LIBRUBY_ALIASES='lib$(RUBY_INSTALL_NAME).$(SOEXT)'
+	AS_IF([test "$load_relative" = yes], [
+	    libprefix="@executable_path/../${libdir_basename}"
+	    LIBRUBY_RELATIVE=yes
+	])
+	LIBRUBY_DLDFLAGS="$LIBRUBY_DLDFLAGS -install_name ${libprefix}"'/$(LIBRUBY_SONAME)'
+	LIBRUBY_DLDFLAGS="$LIBRUBY_DLDFLAGS "'-compatibility_version $(RUBY_API_VERSION)'
+	LIBRUBY_DLDFLAGS="$LIBRUBY_DLDFLAGS "'-current_version $(RUBY_PROGRAM_VERSION)'
+	AS_IF([test "$visibility_option" = ld], [
+	    LIBRUBY_DLDFLAGS="$LIBRUBY_DLDFLAGS "'-Wl,-unexported_symbol,_Init_*'
+	    LIBRUBY_DLDFLAGS="$LIBRUBY_DLDFLAGS "'-Wl,-unexported_symbol,_ruby_static_id_*'
+	    LIBRUBY_DLDFLAGS="$LIBRUBY_DLDFLAGS "'-Wl,-unexported_symbol,*_threadptr_*'
+	])
+	LIBRUBY_DLDFLAGS="$LIBRUBY_DLDFLAGS "' $(XLDFLAGS)'
+	],
+    [interix*], [
+	LIBRUBYARG_SHARED='-L. -L${libdir} -l$(RUBY_SO_NAME)'
+	],
+    [mingw*|cygwin*|mswin*], [
+	LIBRUBY_RELATIVE=yes
+	])
+], [
+  LIBRUBYARG_SHARED=
+
+  # enable PIE if possible
+  AC_ARG_ENABLE(pie,
+          AS_HELP_STRING([--disable-pie], [disable PIE feature]),
+          [pie=$enableval], [pie=])
+  AS_IF([test "$GCC" = yes -a -z "$EXTSTATIC" -a "x$pie" != xno], [
+    RUBY_TRY_CFLAGS(-fPIE, [pie=yes], [pie=no])
+    AS_IF([test "$pie" = yes], [
+      # Use -fPIE when testing -pie.  RUBY_TRY_LDFLAGS sets
+      # $save_CFLAGS internally, so set other name here.
+      save_CFLAGS_before_pie="$CFLAGS"
+      CFLAGS="$CFLAGS -fPIE"
+
+      # gcc need -pie but clang need -Wl,-pie.
+      for pie in -pie -Wl,-pie; do
+	RUBY_TRY_LDFLAGS([$pie], [], [pie=])
+	AS_IF([test "x$pie" != x], [
+	  RUBY_APPEND_OPTION(XCFLAGS, -fPIE)
+	  RUBY_APPEND_OPTION(XLDFLAGS, $pie)
+	  break
+	])
+      done
+      CFLAGS="$save_CFLAGS_before_pie"
+    ])
+  ])
+])
+AS_IF([test "$enable_rpath" = yes], [
+    test -z "$LIBRUBY_RPATHFLAGS" || LIBRUBY_RPATHFLAGS="$LIBRUBY_RPATHFLAGS "
+    rpathflag="${RPATHFLAG}"
+    AS_CASE(["${cross_compiling}${load_relative}"], [*yes*], [], [rpathflag="$RPATHFLAG$LIBPATHFLAG"])
+    rpathflag=`IFS="$PATH_SEPARATOR"
+        echo x "$rpathflag" |
+        sed "s/^x *//;s${IFS}"'%1\\$-s'"${IFS}${libprefix}${IFS}g;s${IFS}%s${IFS}${libprefix}${IFS}g"
+    `
+    LIBRUBY_RPATHFLAGS="$LIBRUBY_RPATHFLAGS${rpathflag}"
+    LIBRUBYARG_SHARED="$LIBRUBY_RPATHFLAGS $LIBRUBYARG_SHARED"
+    LIBRUBYARG_STATIC="$LIBRUBY_RPATHFLAGS $LIBRUBYARG_STATIC"
+])
+AC_SUBST(LIBRUBY_RELATIVE)
+
+LDFLAGS="-L. $LDFLAGS"
+AC_SUBST(ARCHFILE)
+
+AS_IF([test "$EXEEXT" = .exe], [
+    EXECUTABLE_EXTS='".exe",".com",".cmd",".bat"'
+    AC_DEFINE_UNQUOTED(EXECUTABLE_EXTS, $EXECUTABLE_EXTS)
+    EXECUTABLE_EXTS=`echo $EXECUTABLE_EXTS | tr -d '"' | tr , ' '`
+    AC_SUBST(EXECUTABLE_EXTS)
+])
+
+AS_CASE("$cross_compiling:${LIBPATHENV}", [yes:* | no:], [], [
+    AC_MSG_CHECKING(whether wrapper for $LIBPATHENV is needed)
+    AS_IF([env ${LIBPATHENV}=/lib /bin/sh -c ': ${'${LIBPATHENV}'?}' 2>/dev/null],
+	[AC_MSG_RESULT(no)],
+	[PREP="$PREP"' exe/$(PROGRAM)'
+	AC_MSG_RESULT(yes)]
+    )
+])
+
+AC_ARG_ENABLE(dtrace,
+        AS_HELP_STRING([--enable-dtrace],
+        [enable DTrace for tracing inside ruby. enabled by default on systems having dtrace]),
+        [enable_dtrace=$enableval], [enable_dtrace=auto])
+
+LIBRUBY_A_OBJS='$(OBJS)'
+DTRACE_REBUILD=
+AS_CASE(["${enable_dtrace}"],
+[yes|auto], [
+    RUBY_DTRACE_AVAILABLE()
+], [
+    rb_cv_dtrace_available=no
+])
+AS_IF([test "${enable_dtrace}" = yes], [dnl
+    AS_IF([test -z "$DTRACE"], [dnl
+	AC_MSG_ERROR([dtrace(1) is missing])
+    ], [test "$cross_compiling" = yes], [dnl
+	AC_MSG_ERROR([--enable-dtrace, however, cross compiling])
+    ], [test "${rb_cv_dtrace_available}" = "no"], [dnl
+       AC_MSG_ERROR([--enable-dtrace, however, USDT is not available])
+    ])
+])
+AS_CASE([$rb_cv_dtrace_available],
+[yes*], [dnl
+    RUBY_DTRACE_POSTPROCESS()
+    AS_IF([test "$rb_cv_prog_dtrace_g" != no], [dnl
+	DTRACE_OBJ='probes.$(OBJEXT)'
+    ])
+    AS_IF([test "$rb_cv_prog_dtrace_g" = rebuild], [dnl
+	DTRACE_REBUILD=yes
+	LIBRUBY_A_OBJS='$(DTRACE_GLOMMED_OBJ)'
+    ])
+    AS_CASE("${target_os}", [freebsd*], [dnl
+        # FreeBSD's dtrace requires libelf
+        LIBS="-lelf $LIBS"
+    ])
+    DTRACE_EXT=d
+], [dnl
+    enable_dtrace=no
+    DTRACE_EXT=dmyh
+])
+AC_SUBST(DTRACE_EXT)
+AC_SUBST(DTRACE_OBJ)
+AC_SUBST(DTRACE_REBUILD)
+AC_SUBST(DTRACE_OPT)
+AC_SUBST(LIBRUBY_A_OBJS)
+
+AC_ARG_ENABLE(gcov,
+       AS_HELP_STRING([--enable-gcov], [enable coverage measurement by gcov]),
+       [gcov=yes])
+AS_IF([test x"$gcov" = xyes], [
+    CFLAGS="$CFLAGS -coverage"
+    LDFLAGS="$LDFLAGS -coverage"
+])
+
+RUBY_SETJMP_TYPE
+}
+
+: "build section" && {
+dnl build rdoc index if requested
+RDOCTARGET=""
+CAPITARGET=""
+AC_ARG_ENABLE(install-doc,
+       AS_HELP_STRING([--disable-install-doc], [do not install either rdoc indexes or C API documents during install]),
+       [install_doc=$enableval], [install_doc=yes])
+AC_ARG_ENABLE(install-rdoc,
+      AS_HELP_STRING([--disable-install-rdoc], [do not install rdoc indexes during install]),
+      [install_rdoc=$enableval], [install_rdoc=yes])
+AC_ARG_ENABLE(install-capi,
+      AS_HELP_STRING([--disable-install-capi], [do not install C API documents during install]),
+      [install_capi=$enableval], [install_capi=no])
+
+AS_IF([test "$install_doc" != no], [
+    AS_IF([test "$install_rdoc" != no], [
+	RDOCTARGET="rdoc"
+    ], [
+	RDOCTARGET="nodoc"
+    ])
+    AS_IF([test "$install_capi" != no -a -n "$DOXYGEN"], [
+	CAPITARGET="capi"
+    ], [
+	CAPITARGET="nodoc"
+    ])
+], [
+    RDOCTARGET="nodoc"
+    CAPITARGET="nodoc"
+])
+
+AC_SUBST(RDOCTARGET)
+AC_SUBST(CAPITARGET)
+
+AS_CASE(["$RDOCTARGET:$CAPITARGET"],[nodoc:nodoc],[INSTALLDOC=nodoc],[INSTALLDOC=all])
+AC_SUBST(INSTALLDOC)
+
+AC_ARG_ENABLE(jit-support,
+        AS_HELP_STRING([--disable-jit-support], [disable JIT features]),
+        [MJIT_SUPPORT=$enableval
+         AS_IF([test x"$enable_jit_support" = "xyes"],
+                 [AC_DEFINE(USE_MJIT, 1)],
+                 [AC_DEFINE(USE_MJIT, 0)])],
+        [MJIT_SUPPORT=yes
+         AC_DEFINE(USE_MJIT, 1)])
+
+AC_SUBST(MJIT_SUPPORT)
+
+AC_ARG_ENABLE(install-static-library,
+	AS_HELP_STRING([--disable-install-static-library], [do not install static ruby library]),
+	[INSTALL_STATIC_LIBRARY=$enableval],
+	AS_IF([test x"$enable_shared" = xyes],
+	    [INSTALL_STATIC_LIBRARY=no],
+	    [INSTALL_STATIC_LIBRARY=yes]))
+AC_SUBST(INSTALL_STATIC_LIBRARY)
+
+AS_IF([test "$rb_with_pthread" = "yes"], [
+    THREAD_MODEL=pthread
+])
+AC_CACHE_CHECK([for prefix of external symbols], rb_cv_symbol_prefix, [
+    AC_TRY_COMPILE([extern void conftest_external(void) {}], [], [
+	rb_cv_symbol_prefix=`$NM conftest.$ac_objext |
+			     sed -n ['/.*T[ 	]\([^ 	]*\)conftest_external.*/!d;s//\1/p;q']`
+	],
+	[rb_cv_symbol_prefix=''])
+    test -n "$rb_cv_symbol_prefix" || rb_cv_symbol_prefix=NONE
+])
+SYMBOL_PREFIX="$rb_cv_symbol_prefix"
+test "x$SYMBOL_PREFIX" = xNONE && SYMBOL_PREFIX=''
+DLNOBJ=dln.o
+AC_ARG_ENABLE(dln,
+	      AC_HELP_STRING([--disable-dln], [disable dynamic link feature]),
+	      [test "$enableval" = yes || DLNOBJ=dmydln.o])
+AC_SUBST(DLNOBJ)
+MINIDLNOBJ=dmydln.o
+
+AS_CASE(["$target_os"],
+    [linux*], [
+	],
+    [netbsd*], [
+	RUBY_APPEND_OPTION(CFLAGS, -pipe)
+	],
+    [darwin*], [
+	RUBY_APPEND_OPTION(CFLAGS, -pipe)
+	AC_COMPILE_IFELSE([
+	    AC_LANG_BOOL_COMPILE_TRY([@%:@include <AvailabilityMacros.h>],
+		[MAC_OS_X_VERSION_MIN_REQUIRED >= MAC_OS_X_VERSION_10_7])],
+	    [dnl
+		RUBY_APPEND_OPTION(XLDFLAGS, [-framework Security])
+		RUBY_APPEND_OPTION(LIBRUBYARG_STATIC, [-framework Security])
+	    ]dnl
+	)
+	RUBY_APPEND_OPTION(XLDFLAGS, [-framework Foundation])
+	RUBY_APPEND_OPTION(LIBRUBYARG_STATIC, [-framework Foundation])
+	],
+    [osf*], [
+	AS_IF([test "$GCC" != "yes" ], [
+	  # compile something small: taint.c is fine for this.
+	  # the main point is the '-v' flag of 'cc'.
+	  AS_CASE(["`cc -v -I. -c main.c -o /tmp/main.o 2>&1`"],
+	  [*/gemc_cc*], [   # we have the new DEC GEM CC
+                        CFLAGS="$CFLAGS -oldc"
+                        ],
+          [            # we have the old MIPS CC
+                        ])
+	  # cleanup
+	  rm -f /tmp/main.o
+	  CFLAGS="$CFLAGS -std"
+	])
+	],
+    [cygwin*|mingw*], [
+	LIBRUBY_DLDFLAGS="${LIBRUBY_DLDFLAGS}"' -Wl,--out-implib=$(LIBRUBY)'
+	AS_CASE(["$target_os"],
+	[cygwin*], [
+	    AS_IF([test x"$enable_shared" = xyes], [
+		LIBRUBY_SO='cyg$(RUBY_SO_NAME)'.dll
+		LIBRUBY_DLDFLAGS="${LIBRUBY_DLDFLAGS}"' $(RUBYDEF)'
+	    ])
+	    ],
+	[mingw*], [
+	    AS_IF([test x"$enable_shared" = xyes], [
+		LIBRUBY_SO='$(RUBY_SO_NAME)'.dll
+		LIBRUBY_DLDFLAGS="${LIBRUBY_DLDFLAGS}"' $(RUBYDEF)'
+	    ])
+	    EXPORT_PREFIX=' '
+	    DLDFLAGS="${DLDFLAGS}"' $(DEFFILE)'
+	    AC_LIBOBJ([win32/win32])
+	    AC_LIBOBJ([win32/file])
+	    COMMON_LIBS=m
+#	    COMMON_MACROS="WIN32_LEAN_AND_MEAN="
+	    COMMON_HEADERS="winsock2.h windows.h"
+	    THREAD_MODEL=win32
+	    PLATFORM_DIR=win32
+	    ])
+	LIBRUBY_ALIASES=''
+	FIRSTMAKEFILE=GNUmakefile:cygwin/GNUmakefile.in
+	AS_IF([test x"$enable_shared" = xyes], [
+	    LIBRUBY='lib$(RUBY_SO_NAME).dll.a'
+	], [
+	    LIBRUBY_SO=dummy
+	    LIBRUBY='lib$(RUBY_SO_NAME).a'
+	    LIBRUBYARG='-l$(RUBY_SO_NAME)'
+	])
+	],
+    [hpux*], [
+	AS_CASE(["$YACC"],[*yacc*], [
+	    XCFLAGS="$XCFLAGS -DYYMAXDEPTH=300"
+	    YACC="$YACC -Nl40000 -Nm40000"
+	])
+])
+
+MINIOBJS="$MINIDLNOBJ"
+
+AS_CASE(["$THREAD_MODEL"],
+[pthread], [AC_CHECK_HEADERS(pthread.h)],
+[win32],   [],
+[""],      [AC_MSG_ERROR(thread model is missing)],
+           [AC_MSG_ERROR(unknown thread model $THREAD_MODEL)])
+
+AC_ARG_ENABLE(debug-env,
+       AS_HELP_STRING([--enable-debug-env], [enable RUBY_DEBUG environment variable]),
+       [AC_SUBST(ENABLE_DEBUG_ENV, yes)])
+
+AS_CASE(["$FIRSTMAKEFILE"], [*GNUmakefile:*], [gnumake=yes], [
+    AC_MSG_CHECKING([if ${MAKE-make} is GNU make])
+    mkdir conftest.dir
+    echo "all:; @echo yes" > conftest.dir/GNUmakefile
+    echo "all:; @echo no" > conftest.dir/Makefile
+    gnumake=`(cd conftest.dir; ${MAKE-make})`
+    rm -fr conftest.dir
+    AS_CASE(["$gnumake"],
+    [*yes*], [
+	FIRSTMAKEFILE=GNUmakefile:template/GNUmakefile.in
+	gnumake=yes],
+    [
+	gnumake=no])
+    AC_MSG_RESULT($gnumake)
+])
+AS_IF([test "$gnumake" = yes], [ NULLCMD=: ], [
+    AC_MSG_CHECKING([for safe null command for ${MAKE-make}])
+    mkdir conftest.dir
+    NULLCMD=
+    for cmd in : true; do
+	echo 'A=1' > conftest.dir/Makefile
+	echo 'B=$(A:1=@'$cmd')' >> conftest.dir/Makefile
+	echo 'all:; $B 1 2 3 4 5 6 7 8 9' >> conftest.dir/Makefile
+	AS_IF([(cd conftest.dir; ${MAKE-make} >/dev/null 2>/dev/null)], [
+	    NULLCMD=$cmd
+	    break
+	])
+    done
+    rm -fr conftest.dir
+    AS_IF([test -z "$NULLCMD"], [
+	AC_MSG_ERROR(no candidate for safe null command)
+    ])
+    AC_MSG_RESULT($NULLCMD)
+])
+AC_SUBST(NULLCMD)
+
+AS_IF([test "${universal_binary-no}" = yes ], [
+    AC_CACHE_CHECK([for architecture macros], rb_cv_architecture_macros, [
+    mv confdefs.h confdefs1.h
+    : > confdefs.h
+    AC_TRY_COMPILE([@%:@if defined __`echo ${universal_archnames} |
+    sed 's/=[^ ]*//g;s/ /__ || defined __/g'`__
+@%:@else
+@%:@error
+>>>>>><<<<<<
+@%:@endif], [],
+[
+    rb_cv_architecture_macros=yes
+    mv -f confdefs1.h confdefs.h
+], [
+    rb_cv_architecture_macros=no
+    archflagpat=`eval echo '"'"${ARCH_FLAG}"'"' | sed 's/[[][|.*]]/\\&/g'`
+    new_cflags=`echo "$CFLAGS" | sed "s|$archflagpat"'||'`
+    for archs in ${universal_archnames}; do
+	cpu=${archs@%:@*=}
+	archs=${archs%=*}
+	CFLAGS="$new_cflags -arch $archs"
+	archs="__${archs}__"
+	AC_MSG_CHECKING([for macro ${archs} on ${cpu}])
+	AC_TRY_COMPILE([@%:@ifndef ${archs}
+@%:@error
+@%:@endif], [], [AC_MSG_RESULT([yes])], [AC_MSG_RESULT([no])])
+    done
+    mv -f confdefs1.h confdefs.h
+    AC_MSG_ERROR([failed])
+    ])])
+    AC_CACHE_CHECK(whether __ARCHITECTURE__ is available, rb_cv_architecture_available,
+	AC_TRY_COMPILE([@%:@include <stdio.h>
+		const char arch[[]] = __ARCHITECTURE__;], [puts(arch);],
+		[rb_cv_architecture_available=yes], [rb_cv_architecture_available=no]))
+])
+
+: ${MJIT_LDSHARED=`echo "$LDSHARED" | sed ['s|\$(LD)|'"${LD}"'|g;s|\$(CC)|$(MJIT_CC)|g']`}
+
+MAINLIBS="$LIBS"
+LIBS=$ORIG_LIBS
+AS_IF([test -n "${LIBS}"], [
+    libspat=`echo "${LIBS}" | sed 's/[[][|.*$^]]/\\&/g;s/^  */ /;s/^  *$/ /'`
+    MAINFLAGS=`echo " $MAINLIBS " | sed "s|$libspat"'||;s/^ *//;s/ *$//'`
+])
+LIBRUBYARG_STATIC="${LIBRUBYARG_STATIC} \$(MAINLIBS)"
+CPPFLAGS="$CPPFLAGS "'$(DEFS)'
+test -z "$CPPFLAGS" || CPPFLAGS="$CPPFLAGS "; CPPFLAGS="$CPPFLAGS"'${cppflags}'
+AS_IF([test -n "${cflags+set}"], [
+    cflagspat=`eval echo '"'"${cflags}"'"' | sed 's/[[][|.*]]/\\&/g;s/^  */ /;s/^  *$/ /'`
+    CFLAGS=`echo " $CFLAGS " | sed "s|$cflagspat"'|${cflags}|;s/^ *//;s/ *$//'`
+])
+AS_IF([test -n "${cxxflags+set}"], [
+    cxxflagspat=`eval echo '"'"${cxxflags}"'"' | sed 's/[[][|.*]]/\\&/g;s/^  */ /;s/^  *$/ /'`
+    CXXFLAGS=`echo " $CXXFLAGS " | sed "s|$cxxflagspat"'|${cxxflags}|;s/^ *//;s/ *$//'`
+])
+AS_IF([test "${ARCH_FLAG}"], [
+    archflagpat=`eval echo '"'"${ARCH_FLAG}"'"' | sed 's/[[][|.*]]/\\&/g'`
+    CFLAGS=`echo "$CFLAGS" | sed "s| *$archflagpat"'||'`
+    CXXFLAGS=`echo "$CXXFLAGS" | sed "s| *$archflagpat"'||'`
+    LDFLAGS=`echo "$LDFLAGS" | sed "s| *$archflagpat"'||'`
+])
+warnflags="$rb_cv_warnflags"
+AC_SUBST(cppflags)dnl
+AC_SUBST(cflags, ["${orig_cflags:+$orig_cflags }"'${optflags} ${debugflags} ${warnflags}'])dnl
+AC_SUBST(cxxflags, ["${orig_cxxflags:+$orig_cxxflags }"'${optflags} ${debugflags} ${warnflags}'])dnl
+AC_SUBST(optflags)dnl
+AC_SUBST(debugflags)dnl
+AC_SUBST(warnflags)dnl
+AC_SUBST(strict_warnflags)dnl
+AC_SUBST(XCFLAGS)dnl
+AC_SUBST(XLDFLAGS)dnl
+AC_SUBST(EXTLDFLAGS)dnl
+AC_SUBST(EXTDLDFLAGS)dnl
+AC_SUBST(LIBRUBY_LDSHARED)
+AC_SUBST(LIBRUBY_DLDFLAGS)
+AC_SUBST(RUBY_INSTALL_NAME)
+AC_SUBST(rubyw_install_name)
+AC_SUBST(RUBYW_INSTALL_NAME)
+AC_SUBST(RUBY_SO_NAME)
+AC_SUBST(LIBRUBY_A)
+AC_SUBST(LIBRUBY_SO)
+AC_SUBST(LIBRUBY_SONAME)
+AC_SUBST(LIBRUBY_ALIASES)
+AC_SUBST(LIBRUBY)
+AC_SUBST(LIBRUBYARG)
+AC_SUBST(LIBRUBYARG_STATIC)
+AC_SUBST(LIBRUBYARG_SHARED)
+AC_SUBST(SOLIBS)
+AC_SUBST(DLDLIBS)
+AC_SUBST(DLDSHARED)
+AC_SUBST(ENABLE_SHARED)
+AC_SUBST(MAINLIBS)
+AC_SUBST(COMMON_LIBS)
+AC_SUBST(COMMON_MACROS)
+AC_SUBST(COMMON_HEADERS)
+AC_SUBST(EXPORT_PREFIX)
+AC_SUBST(SYMBOL_PREFIX)
+AC_SUBST(MINIOBJS)
+AC_SUBST(THREAD_MODEL)
+AC_SUBST(PLATFORM_DIR)
+
+firstmf=`echo $FIRSTMAKEFILE | sed 's/:.*//'`
+firsttmpl=`echo $FIRSTMAKEFILE | sed 's/.*://'`
+MAKEFILES="Makefile $firstmf"
+MAKEFILES="`echo $MAKEFILES`"
+AC_SUBST(MAKEFILES)
+
+ri_prefix=
+test "$program_prefix" != NONE &&
+  ri_prefix=$program_prefix
+
+ri_suffix=
+test "$program_suffix" != NONE &&
+  ri_suffix=$program_suffix
+
+RUBY_INSTALL_NAME="${ri_prefix}"'$(RUBY_BASE_NAME)'"${ri_suffix}"
+AS_CASE(["$target_os"],
+  [cygwin*|mingw*], [
+    RUBYW_INSTALL_NAME="${ri_prefix}"'$(RUBYW_BASE_NAME)'"${ri_suffix}"
+    rubyw_install_name='$(RUBYW_INSTALL_NAME)'
+    ])
+
+rubylibdir='${rubylibprefix}/${ruby_version}'
+rubyarchdir=${multiarch+'${rubyarchprefix}/${ruby_version}'}${multiarch-'${rubylibdir}/${arch}'}
+
+rubyarchprefix=${multiarch+'${archlibdir}/${RUBY_BASE_NAME}'}${multiarch-'${rubylibprefix}/${arch}'}
+AC_ARG_WITH(rubyarchprefix,
+	    AS_HELP_STRING([--with-rubyarchprefix=DIR],
+			   [prefix for architecture dependent ruby libraries [[RUBYLIBPREFIX/ARCH]]]),
+	    [rubyarchprefix="$withval"])
+AC_SUBST(rubyarchprefix)
+
+rubysitearchprefix=${multiarch+'${sitearchlibdir}/${RUBY_BASE_NAME}'}${multiarch-'${rubylibprefix}/${sitearch}'}
+AC_ARG_WITH(rubysitearchprefix,
+	    AS_HELP_STRING([--with-rubysitearchprefix=DIR],
+			   [prefix for architecture dependent site libraries [[RUBYLIBPREFIX/SITEARCH]]]),
+	    [rubysitearchprefix="$withval"])
+AC_SUBST(rubysitearchprefix)
+
+RI_BASE_NAME=`echo ${RUBY_BASE_NAME} | sed 's/ruby/ri/'`
+ridir='${datarootdir}/${RI_BASE_NAME}'
+AC_ARG_WITH(ridir,
+	    AS_HELP_STRING([--with-ridir=DIR], [ri documentation [[DATAROOTDIR/ri]]]),
+	    [ridir=$withval])
+AC_SUBST(ridir)
+AC_SUBST(RI_BASE_NAME)
+
+AC_ARG_WITH(ruby-version,
+	    AS_HELP_STRING([--with-ruby-version=STR], [ruby version string for version specific directories [[full]] (full|minor|STR)]),
+            [ruby_version=$withval],
+            [ruby_version=full])
+unset RUBY_LIB_VERSION
+unset RUBY_LIB_VERSION_STYLE
+AS_CASE(["$ruby_version"],
+  [full],  [RUBY_LIB_VERSION_STYLE='3	/* full */'],
+  [minor], [RUBY_LIB_VERSION_STYLE='2	/* minor */'])
+AS_IF([test ${RUBY_LIB_VERSION_STYLE+set}], [
+    {
+    echo "#define RUBY_LIB_VERSION_STYLE $RUBY_LIB_VERSION_STYLE"
+    echo '#define STRINGIZE(x) x'
+    test -f revision.h -o -f "${srcdir}/revision.h" || echo '#define RUBY_REVISION 0'
+    echo '#include "version.h"'
+    echo 'ruby_version=RUBY_LIB_VERSION'
+    } > conftest.c
+    ruby_version="`$CPP -I. -I"${srcdir}" -I"${srcdir}/include" conftest.c | sed '/^ruby_version=/!d;s/ //g'`"
+    eval $ruby_version
+], [test -z "${ruby_version}"], [
+    AC_MSG_ERROR([No ruby version, No place for bundled libraries])
+], [
+    RUBY_LIB_VERSION="${ruby_version}"
+])
+AC_SUBST(RUBY_LIB_VERSION_STYLE)
+AC_SUBST(RUBY_LIB_VERSION)
+
+AC_ARG_WITH(sitedir,
+	    AS_HELP_STRING([--with-sitedir=DIR], [site libraries in DIR [[RUBY_LIB_PREFIX/site_ruby]], "no" to disable site directory]),
+            [sitedir=$withval],
+            [sitedir='${rubylibprefix}/site_ruby'])
+sitelibdir='${sitedir}/${ruby_version}'
+
+AC_ARG_WITH(sitearchdir,
+	    AS_HELP_STRING([--with-sitearchdir=DIR],
+			   [architecture dependent site libraries in DIR [[SITEDIR/SITEARCH]], "no" to disable site directory]),
+            [sitearchdir=$withval],
+            [sitearchdir=${multiarch+'${rubysitearchprefix}/site_ruby/${ruby_version}'}${multiarch-'${sitelibdir}/${sitearch}'}])
+
+AC_ARG_WITH(vendordir,
+	    AS_HELP_STRING([--with-vendordir=DIR], [vendor libraries in DIR [[RUBY_LIB_PREFIX/vendor_ruby]], "no" to disable vendor directory]),
+            [vendordir=$withval],
+            [vendordir='${rubylibprefix}/vendor_ruby'])
+vendorlibdir='${vendordir}/${ruby_version}'
+
+AC_ARG_WITH(vendorarchdir,
+	    AS_HELP_STRING([--with-vendorarchdir=DIR],
+			   [architecture dependent vendor libraries in DIR [[VENDORDIR/SITEARCH]], "no" to disable vendor directory]),
+            [vendorarchdir=$withval],
+            [vendorarchdir=${multiarch+'${rubysitearchprefix}/vendor_ruby/${ruby_version}'}${multiarch-'${vendorlibdir}/${sitearch}'}])
+
+AS_IF([test "${LOAD_RELATIVE+set}"], [
+    AC_DEFINE_UNQUOTED(LOAD_RELATIVE, $LOAD_RELATIVE)
+    RUBY_EXEC_PREFIX=''
+])
+
+AC_SUBST(RUBY_EXEC_PREFIX)
+
+AC_SUBST(libdirname, ${multiarch+arch}libdir)
+AC_SUBST(archlibdir)dnl
+AC_SUBST(sitearchlibdir)dnl
+AC_SUBST(archincludedir)dnl
+AC_SUBST(sitearchincludedir)dnl
+AC_SUBST(arch)dnl
+AC_SUBST(sitearch)dnl
+AC_SUBST(ruby_version)dnl
+AC_SUBST(rubylibdir)dnl
+AC_SUBST(rubyarchdir)dnl
+AC_SUBST(sitedir)dnl
+AC_SUBST(sitelibdir)dnl
+AC_SUBST(sitearchdir)dnl
+AC_SUBST(vendordir)dnl
+AC_SUBST(vendorlibdir)dnl
+AC_SUBST(vendorarchdir)dnl
+
+AC_SUBST(CONFIGURE, "`echo $0 | sed 's|.*/||'`")dnl
+AC_SUBST(configure_args, "`echo "${ac_configure_args}" | sed 's/\\$/$$/g'`")dnl
+
+AS_IF([test "${universal_binary-no}" = yes ], [
+    arch="universal-${target_os}"
+    AS_IF([test "${rb_cv_architecture_available}" = yes], [
+	AC_DEFINE_UNQUOTED(RUBY_PLATFORM_CPU, __ARCHITECTURE__)
+    ], [
+	for archs in ${universal_archnames}; do
+	    cpu=`echo $archs | sed 's/.*=//'`
+	    archs=`echo $archs | sed 's/=.*//'`
+	    RUBY_DEFINE_IF([defined __${archs}__ &&! defined RUBY_PLATFORM_CPU], RUBY_PLATFORM_CPU, ["${cpu}"])
+	done
+    ])
+    ints='long int short'
+    test "$ac_cv_type_long_long" = yes && ints="'long long' $ints"
+    AC_SUBST(UNIVERSAL_ARCHNAMES, "${universal_archnames}")
+    AC_SUBST(UNIVERSAL_INTS, "${ints}")
+    AC_DEFINE_UNQUOTED(RUBY_PLATFORM_OS, "${target_os}")
+    AC_DEFINE_UNQUOTED(RUBY_ARCH, "universal-"RUBY_PLATFORM_OS)
+    AC_DEFINE_UNQUOTED(RUBY_PLATFORM, "universal."RUBY_PLATFORM_CPU"-"RUBY_PLATFORM_OS)
+], [
+    arch="${target_cpu}-${target_os}"
+    AC_DEFINE_UNQUOTED(RUBY_PLATFORM, "$arch")
+])
+
+unset sitearch
+AS_CASE(["$target_os"],[mingw*],[sitearch="$target_cpu-$rb_cv_msvcrt"])
+: ${sitearch='${arch}'}
+
+AC_ARG_WITH(search-path,
+		AS_HELP_STRING([--with-search-path=DIR], [specify the additional search path]),
+		[search_path=$withval])
+AS_IF([test "$search_path" != ""], [
+    AC_SUBST(RUBY_SEARCH_PATH, $search_path)
+])
+
+AC_ARG_WITH(rubyhdrdir,
+	    AS_HELP_STRING([--with-rubyhdrdir=DIR], [core headers in DIR [[INCLUDEDIR/RUBY_BASE_NAME-RUBY_VERSION]]]),
+	    [rubyhdrdir=$withval],
+	    [rubyhdrdir='${includedir}/${RUBY_VERSION_NAME}'])
+
+AC_ARG_WITH(rubyarchhdrdir,
+	    AS_HELP_STRING([--with-rubyarchhdrdir=DIR],
+			   [architecture dependent core headers in DIR [[$(rubyhdrdir)/$(arch)]]]),
+	    [rubyarchhdrdir=$withval],
+	    [rubyarchhdrdir=${multiarch+'${archincludedir}/${RUBY_VERSION_NAME}'}${multiarch-'${rubyhdrdir}/${arch}'}])
+
+AC_ARG_WITH(sitehdrdir,
+	    AS_HELP_STRING([--with-sitehdrdir=DIR], [core site headers in DIR [[RUBYHDRDIR/site_ruby]]]),
+	    [sitehdrdir=$withval],
+	    [sitehdrdir='${rubyhdrdir}/site_ruby'])
+
+AC_ARG_WITH(sitearchhdrdir,
+	    AS_HELP_STRING([--with-sitearchhdrdir=DIR],
+			   [architecture dependent core site headers in DIR [[RUBYHDRDIR/site_ruby]]]),
+	    [sitearchhdrdir=$withval],
+	    [sitearchhdrdir=${multiarch+'${sitearchincludedir}/${RUBY_VERSION_NAME}/site_ruby'}${multiarch-'${sitehdrdir}/${sitearch}'}])
+
+AC_ARG_WITH(vendorhdrdir,
+	    AS_HELP_STRING([--with-vendorhdrdir=DIR], [core vendor headers in DIR [[RUBYHDRDIR/vendor_ruby]]]),
+	    [vendorhdrdir=$withval],
+	    [vendorhdrdir='${rubyhdrdir}/vendor_ruby'])
+
+AC_ARG_WITH(vendorarchhdrdir,
+	    AS_HELP_STRING([--with-vendorarchhdrdir=DIR],
+			   [architecture dependent core vendor headers in DIR [[RUBYHDRDIR/vendor_ruby]]]),
+	    [vendorarchhdrdir=$withval],
+	    [vendorarchhdrdir=${multiarch+'${sitearchincludedir}/${RUBY_VERSION_NAME}/vendor_ruby'}${multiarch-'${vendorhdrdir}/${sitearch}'}])
+
+AC_SUBST(rubyhdrdir)dnl
+AC_SUBST(sitehdrdir)dnl
+AC_SUBST(vendorhdrdir)dnl
+AC_SUBST(rubyarchhdrdir)dnl
+AC_SUBST(sitearchhdrdir)dnl
+AC_SUBST(vendorarchhdrdir)dnl
+
+AC_ARG_WITH(mantype,
+	AS_HELP_STRING([--with-mantype=TYPE], [specify man page type; TYPE is one of man and doc]),
+		[
+			AS_CASE(["$withval"],
+			[man|man.gz|man.bz2|doc|doc.gz|doc.bz2], [MANTYPE=$withval],
+			[AC_MSG_ERROR(invalid man type: $withval)])
+		])
+AS_IF([test -z "$MANTYPE"], [
+	dnl Looks for nroff with -mdoc support.
+	AC_CACHE_VAL([ac_cv_path_NROFF], [
+		AC_PATH_PROGS_FEATURE_CHECK([NROFF],
+			[nroff awf mandoc],
+			[$ac_path_NROFF -mdoc ${srcdir}/man/ruby.1 \
+				>/dev/null 2>&1 &&
+				ac_cv_path_NROFF=$ac_path_NROFF \
+				ac_path_NROFF_found=:],
+			[], ["/usr/bin:/usr/ucb"]
+		)
+	])
+	AS_IF([test -n "$ac_cv_path_NROFF"], [
+		MANTYPE=doc
+	], [
+		MANTYPE=man
+	])
+])
+AC_SUBST(MANTYPE)
+
+AC_ARG_ENABLE(rubygems,
+	AS_HELP_STRING([--disable-rubygems], [disable rubygems by default]),
+	[enable_rubygems="$enableval"], [enable_rubygems=yes])
+AS_IF([test x"$enable_rubygems" = xno], [
+    USE_RUBYGEMS=no
+], [
+    USE_RUBYGEMS=yes
+])
+AC_SUBST(USE_RUBYGEMS)
+
+arch_hdrdir="${EXTOUT}/include/${arch}/ruby"
+AS_MKDIR_P("${arch_hdrdir}")
+config_h="${arch_hdrdir}/config.h"
+guard=INCLUDE_RUBY_CONFIG_H
+{
+  echo "#ifndef $guard"
+  echo "#define $guard 1"
+  grep -v "^#define PACKAGE_" confdefs.h
+  echo "#endif /* $guard */"
+} | tr -d '\015' |
+(
+  AS_IF([test "x$CONFIGURE_TTY" = xyes], [color=--color], [color=])
+  exec ${srcdir}/tool/ifchange $color "${config_h}" -
+) >&AS_MESSAGE_FD || AC_MSG_ERROR([failed to create ${config_h}])
+tr -d '\015' < largefile.h > confdefs.h
+rm largefile.h
+
+BUILTIN_ENCS=["`sed -n -e '/^BUILTIN_ENCS[ 	]*=/{' \
+	-e s/// -e :l -e '/\\\\$/N' -e 's/\\\\\\n/ /' -e 't l' -e p \
+	-e '}' "${srcdir}/enc/Makefile.in"`"]
+BUILTIN_ENCOBJS=
+for e in $BUILTIN_ENCS; do BUILTIN_ENCOBJS="$BUILTIN_ENCOBJS "`echo $e | sed 's/\.c$/.$(OBJEXT)/'`; done
+AC_SUBST(BUILTIN_ENCOBJS)
+
+BUILTIN_TRANSES=["`sed -n -e '/^BUILTIN_TRANSES[ 	]*=/{' \
+	-e s/// -e :l -e '/\\\\$/N' -e 's/\\\\\\n/ /' -e 't l' -e p \
+	-e '}' "${srcdir}/enc/Makefile.in"`"]
+BUILTIN_TRANSSRCS=
+BUILTIN_TRANSOBJS=
+for e in $BUILTIN_TRANSES; do
+  BUILTIN_TRANSSRCS="$BUILTIN_TRANSSRCS "`echo $e | sed 's/\.trans$/.c/'`
+  BUILTIN_TRANSOBJS="$BUILTIN_TRANSOBJS "`echo $e | sed 's/\.trans$/.$(OBJEXT)/'`
+done
+AC_SUBST(BUILTIN_TRANSSRCS)
+AC_SUBST(BUILTIN_TRANSOBJS)
+
+PACKAGE=$RUBY_BASE_NAME
+AC_SUBST(PACKAGE)
+AS_MESSAGE([$PACKAGE library version = $ruby_version])
+
+AS_IF([test x"$CC_WRAPPER" != x], [
+    CC='$(CC_WRAPPER) '"${CC@%:@$CC_WRAPPER }"
+    CPP='$(CC_WRAPPER) '"${CPP@%:@$CC_WRAPPER }"
+    CC_WRAPPER='$(rubyarchdir)/darwin-cc'
+    XCC_WRAPPER='$(top_srcdir)/tool/darwin-cc'
+])
+AC_SUBST(CC_WRAPPER, '')
+AC_SUBST(XCC_WRAPPER)
+
+AS_CASE([" $CPP "], [*" $CC "*], [CPP=`echo " $CPP " | sed "s| $CC |"' $(CC) |;s/^ *//;s/  *$//'`])
+
+AS_IF([test x"$firstmf" != x], [
+    AC_CONFIG_FILES($firstmf:$firsttmpl, [], [firstmf="$firstmf" firsttmpl="$firsttmpl"])
+])
+AC_CONFIG_FILES(Makefile, [
+    tmpmk=confmk$$.tmp
+    {
+	AS_IF([test ${VCS+set}], [
+	    :
+	], [svn info "$srcdir" > /dev/null 2>&1], [
+	    VCS='svn'
+	], [git_dir=`$GIT --work-tree="$srcdir" --git-dir="$srcdir/.git" rev-parse --git-dir 2>/dev/null`], [
+	    AS_IF([test -d "$git_dir/svn"], [
+		VCS='$(GIT) svn'
+	    ], [
+		VCS='$(GIT)'
+	    ])
+	], [
+	    VCS='echo cannot'
+	])
+	AS_CASE("$VCS",
+		[svn],          [VCSUP='$(VCS) up $(SVNUPOPTIONS)'],
+		['$(GIT) svn'], [VCSUP='$(VCS) rebase $(GITSVNREBASEOPTIONS)'],
+		['$(GIT)'|git], [VCSUP='$(VCS) pull $(GITPULLOPTIONS)'],
+		[VCSUP='$(VCS)'])
+	sed -n \
+	    -e '[/^@%:@define \(RUBY_RELEASE_[A-Z]*\) \([0-9][0-9]*\)/]{' \
+	    -e   's//\1 = \2/' \
+	    -e   '[s/ \([0-9]\)$/ 0\1/]' \
+	    -e   p \
+	    -e '}' "$srcdir/version.h"
+	sed '/^MISSING/s/\$U\././g;/^VCS *=/s#@VCS@#'"$VCS"'#;/^VCSUP *=/s#@VCSUP@#'"$VCSUP"'#' Makefile
+	echo; test x"$EXEEXT" = x || echo 'miniruby: miniruby$(EXEEXT)'
+	AS_IF([test "$gnumake" != yes], [
+	    echo ['$(MKFILES): $(srcdir)/common.mk']
+	    sed ['s/{\$([^(){}]*)[^{}]*}//g'] ${srcdir}/common.mk
+	], [
+	    echo 'distclean-local::; @$(RM) GNUmakefile uncommon.mk'
+	])
+    } > $tmpmk && AS_IF([! grep '^ruby:' $tmpmk > /dev/null], [
+	AS_IF([test "${gnumake}" = yes], [
+	    tmpgmk=confgmk$$.tmp
+	    {
+		echo "include $tmpmk"
+		echo "-include uncommon.mk"
+	    } > $tmpgmk
+	], [
+	    tmpgmk=$tmpmk
+	]) &&
+	test -z "`${MAKE-make} -f $tmpgmk info-program | grep '^PROGRAM=ruby$'`" &&
+	echo 'ruby: $(PROGRAM);' >> $tmpmk
+	test "$tmpmk" = "$tmpgmk" || rm -f "$tmpgmk"
+    ]) && mv -f $tmpmk Makefile],
+[EXEEXT='$EXEEXT' gnumake='$gnumake' GIT='$GIT'])
+
+AC_ARG_WITH([ruby-pc],
+	    AC_HELP_STRING([--with-ruby-pc=FILENAME], [pc file basename]),
+	    [ruby_pc="$withval"],
+	    [ruby_pc="${RUBY_BASE_NAME}-${MAJOR}.${MINOR}.pc"])
+AC_SUBST(ruby_pc)
+AC_SUBST(exec, [exec])
+
+AC_ARG_WITH(destdir,
+	    AS_HELP_STRING([--with-destdir=DESTDIR], [specify default directory to install]),
+	    [DESTDIR="$withval"])
+AC_SUBST(DESTDIR)
+
+AC_CONFIG_FILES($ruby_pc:template/ruby.pc.in,
+    [
+    AS_IF([sed ['s/\$(\([A-Za-z_][A-Za-z0-9_]*\))/${\1}/g;s/@[A-Za-z_][A-Za-z0-9_]*@//'] $ruby_pc > ruby.tmp.pc &&
+	{
+	    test -z "$PKG_CONFIG" ||
+	    PKG_CONFIG_PATH=. $PKG_CONFIG --print-errors ruby.tmp
+	}],
+    [
+	mv -f ruby.tmp.pc $ruby_pc
+    ], [
+	exit 1
+    ])
+    ],
+    [ruby_pc='$ruby_pc' PKG_CONFIG='$PKG_CONFIG'])
+
+AC_OUTPUT
+}
+}
+
+AS_IF([test "$silent" = yes], [], [
+AS_IF([${FOLD+:} false], [], [
+AS_IF([test "`echo abcdefg hijklmno | fold -s -w10 | sed 1d`" = hijklmno], [FOLD="fold"], [FOLD=])
+])
+AS_REQUIRE_SHELL_FN([config_summary],
+    [AS_FUNCTION_DESCRIBE([config_summary], [NAME, VAL], [configuration summary])],
+    [AS_IF([test -z "$2"], [], [
+	AS_ECHO_N(["   * $1:                     "]) | dd bs=1 count=26 2>/dev/null
+	AS_IF([test "$FOLD"], [
+	    echo "$2" | fold -s -w50 |
+	    sed '1!s/^/                          /;$!s/$/\\/'
+	], [echo "$2"])
+    ])]
+)
+
+echo "---"
+echo "Configuration summary for $RUBY_BASE_NAME version $RUBY_PROGRAM_VERSION"
+echo ""
+config_summary "Installation prefix" "$prefix"
+config_summary "exec prefix"         "$exec_prefix"
+config_summary "arch"                "$arch"
+config_summary "site arch"           "$sitearch"
+config_summary "RUBY_BASE_NAME"      "$RUBY_BASE_NAME"
+config_summary "enable shared"       "$enable_shared"
+config_summary "ruby lib prefix"     "$rubylibprefix"
+config_summary "site libraries path" "$rubysitearchprefix"
+config_summary "vendor path"         "$vendordir"
+config_summary "target OS"           "$target_os"
+config_summary "compiler"            "$CC"
+config_summary "with pthread"        "$enable_pthread"
+config_summary "enable shared libs"  "$ENABLE_SHARED"
+config_summary "dynamic library ext" "$DLEXT"
+config_summary "CFLAGS"              "$cflags"
+config_summary "CPPFLAGS"            "$cppflags"
+config_summary "LDFLAGS"             "$LDFLAGS"
+config_summary "DLDFLAGS"            "$DLDFLAGS"
+config_summary "optflags"            "$optflags"
+config_summary "debugflags"          "$debugflags"
+config_summary "warnflags"           "$warnflags"
+config_summary "strip command"       "$STRIP"
+config_summary "install doc"         "$install_doc"
+config_summary "JIT support"         "$MJIT_SUPPORT"
+config_summary "man page type"       "$MANTYPE"
+config_summary "search path"         "$search_path"
+config_summary "static-linked-ext"   ${EXTSTATIC:+"yes"}
+echo ""
+echo "---"
+])
diff --git a/configure.in b/configure.in
deleted file mode 100644
index e31a6fec49..0000000000
--- a/configure.in
+++ /dev/null
@@ -1,1645 +0,0 @@
-dnl Process this file with autoconf to produce a configure script.
-AC_INIT()
-
-AC_PREREQ(2.58)
-
-AC_DEFUN([RUBY_MINGW32],
-[case "$host_os" in
-cygwin*)
-AC_CACHE_CHECK(for mingw32 environment, rb_cv_mingw32,
-[AC_TRY_CPP([
-#ifndef __MINGW32__
-# error
-#endif
-], rb_cv_mingw32=yes,rb_cv_mingw32=no)
-rm -f conftest*])
-test "$rb_cv_mingw32" = yes && target_os="mingw32"
-  ;;
-esac])
-
-AC_DEFUN([RUBY_CPPOUTFILE],
-[AC_CACHE_CHECK(whether ${CPP} accepts -o, rb_cv_cppoutfile,
-[cppflags=$CPPFLAGS
-CPPFLAGS='-o conftest.i'
-AC_TRY_CPP([], rb_cv_cppoutfile=yes, rb_cv_cppoutfile=no)
-CPPFLAGS=$cppflags
-rm -f conftest*])
-if test "$rb_cv_cppoutfile" = yes; then
-  CPPOUTFILE='-o conftest.i'
-elif test "$rb_cv_cppoutfile" = no; then
-  CPPOUTFILE='> conftest.i'
-elif test -n "$rb_cv_cppoutfile"; then
-  CPPOUTFILE="$rb_cv_cppoutfile"
-fi
-AC_SUBST(CPPOUTFILE)])
-
-AC_DEFUN([RUBY_PROG_GNU_LD],
-[AC_CACHE_CHECK(whether the linker is GNU ld, rb_cv_prog_gnu_ld,
-[if `$CC $CFLAGS $CPPFLAGS $LDFLAGS --print-prog-name=ld 2>&1` -v 2>&1 | grep "GNU ld" > /dev/null; then
-  rb_cv_prog_gnu_ld=yes
-else
-  rb_cv_prog_gnu_ld=no
-fi
-])
-GNU_LD=$rb_cv_prog_gnu_ld
-AC_SUBST(GNU_LD)])
-
-rb_version=`grep RUBY_VERSION $srcdir/version.h`
-MAJOR=`expr "$rb_version" : '#define RUBY_VERSION "\([0-9][0-9]*\)\.[0-9][0-9]*\.[0-9][0-9]*"'`
-MINOR=`expr "$rb_version" : '#define RUBY_VERSION "[0-9][0-9]*\.\([0-9][0-9]*\)\.[0-9][0-9]*"'`
-TEENY=`expr "$rb_version" : '#define RUBY_VERSION "[0-9][0-9]*\.[0-9][0-9]*\.\([0-9][0-9]*\)"'`
-AC_SUBST(MAJOR)
-AC_SUBST(MINOR)
-AC_SUBST(TEENY)
-dnl checks for alternative programs
-AC_ARG_WITH(gcc, [  --without-gcc           never use gcc], [
-	case $withval in
-	no)	: ${CC=cc}
-		;;
-	yes)	: ${CC=gcc}
-		;;
-	*)	CC=$withval
-		;;
-	esac])
-dnl If the user switches compilers, we can't believe the cache
-if test ! -z "$ac_cv_prog_CC" -a ! -z "$CC" -a "$CC" != "$ac_cv_prog_CC"
-then
-  AC_MSG_ERROR(cached CC is different -- throw away $cache_file
-(it is also a good idea to do 'make clean' before compiling))
-fi
-
-if test "$program_prefix" = NONE; then
-  program_prefix=
-fi
-AC_CANONICAL_TARGET
-target_os=`echo $target_os | sed 's/linux-gnu$/linux/;s/linux-gnu/linux-/'`
-ac_install_sh='' # unusable for extension libraries.
-
-dnl checks for fat-binary
-AC_ARG_ENABLE(fat-binary,
-       [  --enable-fat-binary=ARCHS
-                          build an Apple/NeXT Multi Architecture Binary (MAB);
-                          ARCHS is a comma-delimited list of architectures for
-                          which to build; if ARCHS is omitted, then the package
-                          will be built for all architectures supported by the
-                          platform ("ppc" for MacOS/X and Darwin; "ppc,i386"
-                          for Rhapsody; "m68k,i386,sparc" for OpenStep;
-                          "m68k,i386,sparc,hppa" for NextStep); if this option
-                          is disabled or omitted entirely, then the package
-                          will be built only for the target platform],
-       [fat_binary=$enableval], [fat_binary=no])
-if test "$fat_binary" != no; then
-
-    AC_MSG_CHECKING([target architectures])
-
-    # Respect TARGET_ARCHS setting from environment if available.
-    if test -z "$TARGET_ARCHS"; then
-	# Respect ARCH given to --enable-fat-binary if present.
-	if test "$fat_binary" != yes; then
-	    TARGET_ARCHS=`echo "$fat_binary" | tr ',' ' '`
-	else
-	    # Choose a default set of architectures based upon platform.
-	    case "$target_os" in
-	    darwin*)
-		TARGET_ARCHS="ppc"
-		;;
-	    rhapsody*)
-		TARGET_ARCHS="ppc i386"
-		;;
-	    openstep*)
-		TARGET_ARCHS="m68k i386 sparc"
-		;;
-	    nextstep*)
-		TARGET_ARCHS="m68k i386 sparc hppa"
-		;;
-	    *)
-		TARGET_ARCHS=`arch`
-	    esac
-	fi
-    fi
-
-    AC_MSG_RESULT([$TARGET_ARCHS])
-
-    # /usr/lib/arch_tool -archify_list $TARGET_ARCHS
-    ARCH_FLAG=
-    for archs in $TARGET_ARCHS 
-    do
-        ARCH_FLAG="$ARCH_FLAG -arch $archs"
-    done
-    AC_DEFINE(NEXT_FAT_BINARY)
-fi
-
-case $target_cpu in
-  i?86) frame_address=yes;;
-  *)    frame_address=no;;
-esac
-AC_ARG_ENABLE(frame-address,
-       [  --enable-frame-address  use GCC __builtin_frame_address(). ],
-       [frame_address=$enableval])
-if test $frame_address = yes; then
-    AC_DEFINE(USE_BUILTIN_FRAME_ADDRESS)
-fi
-
-AC_ARG_PROGRAM
-
-dnl Checks for programs.
-
-if test x"${build}" != x"${host}"; then
-  AC_CHECK_TOOL(CC, gcc)
-fi
-AC_PROG_CC
-AC_PROG_GCC_TRADITIONAL
-
-RUBY_PROG_GNU_LD
-RUBY_CPPOUTFILE
-
-: ${OUTFLAG='-o '}
-AC_SUBST(OUTFLAG)
-
-RUBY_MINGW32
-
-AC_PROG_YACC
-if test "$YACC" = "yacc"; then
-  AC_DEFINE([OLD_YACC])
-fi
-
-AC_CHECK_TOOL(RANLIB, ranlib, :)
-AC_CHECK_TOOL(AR, ar)
-if test -z "$AR"; then
-  AC_CHECK_PROGS(AR, aal, ar)
-fi
-
-AC_CHECK_TOOL(AS, as)
-ASFLAGS=$ASFLAGS
-AC_SUBST(ASFLAGS)
-
-case "$target_os" in
-cygwin*|mingw*)
-    AC_CHECK_TOOL(NM, nm)
-    AC_CHECK_TOOL(WINDRES, windres)
-    AC_CHECK_TOOL(DLLWRAP, dllwrap)
-    target_cpu=`echo $target_cpu | sed s/i.86/i386/`
-    case "$target_os" in
-    mingw*)
-	test "$rb_cv_msvcrt" = "" && unset rb_cv_msvcrt
-	AC_CHECK_TOOL(OBJDUMP, objdump)
-	AC_CACHE_CHECK(for mingw32 runtime DLL, rb_cv_msvcrt, [
-	AC_TRY_LINK([#include <stdio.h>],
-		    [FILE* volatile f = stdin; return 0;],
-		    [rb_cv_msvcrt=`$OBJDUMP -p conftest$ac_exeext |
-				   tr A-Z a-z |
-				   sed -n '/^[[ 	]]*dll name: \(msvc.*\)\.dll$/{s//\1/p;q;}'`],
-		    [rb_cv_msvcrt=msvcrt])
-	test "$rb_cv_msvcrt" = "" && rb_cv_msvcrt=msvcrt])
-	AC_ARG_WITH(winsock2,
-		[  --with-winsock2         link winsock2 (MinGW only)], [
-		case $withval in
-		yes) with_winsock2=yes;;
-		*)   with_winsock2=no;;
-		esac], [with_winsock2=no])
-	if test "$with_winsock2" = yes; then
-	  AC_DEFINE(USE_WINSOCK2)
-	fi
-    esac
-    : ${enable_shared=yes}
-    ;;
-aix*)
-    AC_CHECK_TOOL(NM, nm, /usr/ccs/bin/nm, /usr/ccs/bin:$PATH)
-    ;;
-hiuxmpp*)
-    # by TOYODA Eizi <toyoda@npd.kishou.go.jp>
-    AC_DEFINE(__HIUX_MPP__)
-    ;;
-esac
-
-AC_PROG_LN_S
-AC_PROG_MAKE_SET
-AC_PROG_INSTALL
-
-# checks for UNIX variants that set C preprocessor variables
-AC_AIX
-AC_MINIX
-
-AC_SUBST(RM, ['rm -f'])
-AC_SUBST(CP, ['cp'])
-if $as_mkdir_p; then
-  AC_SUBST(MAKEDIRS, ['mkdir -p'])
-else
-  AC_SUBST(MAKEDIRS, ['install -d'])
-fi
-
-dnl check for large file stuff
-AC_SYS_LARGEFILE
-
-AC_CHECK_TYPES([long long, off_t])
-
-AC_CHECK_SIZEOF(int, 4)
-AC_CHECK_SIZEOF(short, 2)
-AC_CHECK_SIZEOF(long, 4)
-AC_CHECK_SIZEOF(long long, 0)
-AC_CHECK_SIZEOF(__int64, 0)
-AC_CHECK_SIZEOF(off_t, 0)
-AC_CHECK_SIZEOF(void*, 4)
-AC_CHECK_SIZEOF(float, 4)
-AC_CHECK_SIZEOF(double, 8)
-AC_CHECK_SIZEOF(time_t, 0)
-
-for id in pid_t gid_t uid_t; do
-    AC_CHECK_TYPE($id, [typ=$id], [typ=int])
-    AC_DEFINE_UNQUOTED(rb_$id, $typ)
-done
-
-AC_CACHE_CHECK(for prototypes, rb_cv_have_prototypes,
-  [AC_TRY_COMPILE([int foo(int x) { return 0; }], [return foo(10);],
-	rb_cv_have_prototypes=yes,
-	rb_cv_have_prototypes=no)])
-if test "$rb_cv_have_prototypes" = yes; then
-  AC_DEFINE(HAVE_PROTOTYPES)
-fi
-
-AC_CACHE_CHECK(token paste string, rb_cv_tokenpaste,
-  [AC_TRY_COMPILE([#define paste(a,b) a##b],
-		  [int xy = 1; return paste(x,y);],
-		  rb_cv_tokenpaste=ansi,
-		  rb_cv_tokenpaste=knr)])
-if test "$rb_cv_tokenpaste" = ansi; then
-  AC_DEFINE(TOKEN_PASTE(x,y),[x##y])
-else
-  AC_DEFINE(TOKEN_PASTE(x,y),[x/**/y])
-fi
-
-AC_CACHE_CHECK(for variable length prototypes and stdarg.h, rb_cv_stdarg,
-  [AC_TRY_COMPILE([
-#include <stdarg.h>
-int foo(int x, ...) {
-	va_list va;
-	va_start(va, x);
-	va_arg(va, int);
-	va_arg(va, char *);
-	va_arg(va, double);
-	return 0;
-}
-], [return foo(10, "", 3.14);],
-	rb_cv_stdarg=yes,
-	rb_cv_stdarg=no)])
-if test "$rb_cv_stdarg" = yes; then
-  AC_DEFINE(HAVE_STDARG_PROTOTYPES)
-fi
-
-AC_CACHE_CHECK([for noreturn], rb_cv_noreturn,
-[rb_cv_noreturn=x
-for mac in "x __attribute__ ((noreturn))" "__declspec(noreturn) x" x; do
-  AC_TRY_COMPILE(
-    [#define NORETURN(x) $mac
-NORETURN(void exit(int x));],
-    [],
-    [rb_cv_noreturn="$mac"; break])
-done])
-AC_DEFINE_UNQUOTED([NORETURN(x)], $rb_cv_noreturn)
-
-AC_CACHE_CHECK([for RUBY_EXTERN], rb_cv_ruby_extern,
-[rb_cv_ruby_extern=no
-for mac in "__attribute__((dllimport))" "__declspec(dllimport)"; do
-  AC_TRY_COMPILE(
-    [extern $mac void conftest(void);],
-    [rb_cv_ruby_extern="extern $mac"; break])
-done])
-test "x$rb_cv_ruby_extern" = xno || AC_DEFINE_UNQUOTED(RUBY_EXTERN, $rb_cv_ruby_extern)
-
-XCFLAGS="$XCFLAGS -DRUBY_EXPORT"
-
-dnl Check whether we need to define sys_nerr locally
-AC_CHECK_DECLS([sys_nerr], [], [], [$ac_includes_default
-#include <errno.h>])
-
-dnl whether link libc_r or not
-AC_ARG_WITH(libc_r, 
-	[  --with-libc_r           link libc_r if possible (FreeBSD only)], [
-	case $withval in
-	yes) with_libc_r=yes;;
-	*)   with_libc_r=no;;
-	esac], [with_libc_r=no])
-
-AC_ARG_ENABLE(pthread,
-       [  --enable-pthread        use pthread library.],
-       [enable_pthread=$enableval], [enable_pthread=no])
-
-dnl Checks for libraries.
-case "$target_os" in
-nextstep*)	;;
-openstep*)	;;
-rhapsody*)	;;
-darwin*)	LIBS="-lobjc $LIBS";;
-hpux*)		LIBS="-lm $LIBS"
-		ac_cv_c_inline=no;;
-human*)		ac_cv_func_getpgrp_void=yes
-		ac_cv_func_setitimer=no
-		;;
-beos*)		ac_cv_func_link=no;;
-cygwin*)	;;
-mingw*)		if test "$with_winsock2" = yes; then
-		  LIBS="-lws2_32 $LIBS"
-		else
-		  LIBS="-lwsock32 $LIBS"
-		fi
-		LIBS="-lshell32 $LIBS"
-		ac_cv_header_a_out_h=no
-		ac_cv_header_pwd_h=no
-		ac_cv_header_utime_h=no
-		ac_cv_header_sys_ioctl_h=no
-		ac_cv_header_sys_param_h=no
-		ac_cv_header_sys_resource_h=no
-		ac_cv_header_sys_select_h=no
-		ac_cv_header_sys_time_h=no
-		ac_cv_header_sys_times_h=no
-		ac_cv_func_times=yes
-		ac_cv_func_waitpid=yes
-		ac_cv_func_fsync=yes
-		ac_cv_func_vsnprintf=yes
-		ac_cv_func_seekdir=yes
-		ac_cv_func_telldir=yes
-		ac_cv_func_isinf=yes
-		ac_cv_func_isnan=yes
-		ac_cv_func_finite=yes
-		ac_cv_func_link=yes
-		ac_cv_lib_crypt_crypt=no
-		ac_cv_func_getpgrp_void=no
-		ac_cv_func_setpgrp_void=yes
-		ac_cv_func_memcmp_working=yes
-		ac_cv_lib_dl_dlopen=no
-		rb_cv_binary_elf=no
-		rb_cv_negative_time_t=no
-		enable_pthread=no
-		;;
-os2-emx*)	LIBS="-lm $LIBS"
-		ac_cv_lib_dir_opendir=no;;
-msdosdjgpp*)	LIBS="-lm $LIBS"
-		ac_cv_func_getpgrp_void=yes
-		ac_cv_func_setitimer=no
-                ac_cv_sizeof_rlim_t=4
-		;;
-bsdi*)          LIBS="-lm $LIBS"
-                ac_cv_sizeof_rlim_t=8;;
-freebsd*)	LIBS="-lm $LIBS"
-		AC_CACHE_CHECK([whether -lxpg4 has to be linked],
-		  rb_cv_lib_xpg4_needed,
-		  [AC_TRY_CPP([
-#include <osreldate.h>
-#if __FreeBSD_version < 400020 || \
-   (__FreeBSD_version >= 500000 && __FreeBSD_version < 500005)
-#error needs libxpg4
-#endif
-		   ],
-		   rb_cv_lib_xpg4_needed=no,
-		   rb_cv_lib_xpg4_needed=yes,
-		   rb_cv_lib_xpg4_needed=yes)])
-		if test "$rb_cv_lib_xpg4_needed" = yes; then
-			AC_CHECK_LIB(xpg4, setlocale)
-		fi
-		if test "$with_libc_r" = yes; then
-			AC_CACHE_CHECK([whether libc_r is supplementary to libc],
-			  rb_cv_supplementary_lib_c_r,
-			  [AC_TRY_CPP([
-#include <osreldate.h>
-#if 500016 <= __FreeBSD_version
-#error libc_r is supplementary to libc
-#endif
-			   ],
-			   rb_cv_supplementary_lib_c_r=no,
-			   rb_cv_supplementary_lib_c_r=yes,
-			   rb_cv_supplementary_lib_c_r=yes)])
-			if test "$rb_cv_supplementary_lib_c_r" = yes; then
-			   MAINLIBS="-lc_r $MAINLIBS"
-			fi
-		fi
-		;;
-dragonfly*)	LIBS="-lm $LIBS"
-		;;
-bow)		ac_cv_func_setitimer=no
-		;;
-superux*)     ac_cv_func_setitimer=no
-              ;;
-solaris*2.1*) if test -z "$GCC"; then
-	           ac_cv_func_isinf=yes
-              fi
-              LIBS="-lm $LIBS"
-              ;;
-*) LIBS="-lm $LIBS";;
-esac
-AC_CHECK_LIB(crypt, crypt)
-AC_CHECK_LIB(dl, dlopen)	# Dynamic linking for SunOS/Solaris and SYSV
-AC_CHECK_LIB(dld, shl_load)	# Dynamic linking for HP-UX
-
-case "$target_cpu" in
-alpha*)		case "$target_os"::"$GCC" in
-		*::yes)	CFLAGS="-mieee $CFLAGS" ;;   # gcc
-		osf*)   CFLAGS="-ieee $CFLAGS" ;;    # ccc
-		esac ;;
-esac
-
-dnl Checks for header files.
-AC_HEADER_DIRENT
-AC_HEADER_STDC
-AC_HEADER_SYS_WAIT
-AC_CHECK_HEADERS(stdlib.h string.h unistd.h limits.h sys/file.h sys/ioctl.h sys/syscall.h\
-		 fcntl.h sys/fcntl.h sys/select.h sys/time.h sys/times.h sys/param.h\
-		 syscall.h pwd.h grp.h a.out.h utime.h memory.h direct.h sys/resource.h \
-		 sys/mkdev.h sys/utime.h netinet/in_systm.h float.h ieeefp.h pthread.h \
-		 ucontext.h intrinsics.h)
-
-dnl Check additional types.
-AC_CHECK_SIZEOF(rlim_t, 0, [
-  #ifdef HAVE_SYS_TYPES_H
-  # include <sys/types.h>
-  #endif
-  #ifdef HAVE_SYS_TIME_H
-  # include <sys/time.h>
-  #endif
-  #ifdef HAVE_SYS_RESOURCE_H
-  # include <sys/resource.h>
-  #endif
-  #ifdef HAVE_UNISTD_H
-  # include <unistd.h>
-  #endif
-  #include <stdio.h>
-])
-
-dnl Checks for typedefs, structures, and compiler characteristics.
-AC_TYPE_SIZE_T
-AC_STRUCT_ST_BLKSIZE
-AC_STRUCT_ST_BLOCKS
-AC_STRUCT_ST_RDEV
-
-dnl Checks for library functions.
-AC_TYPE_GETGROUPS
-AC_TYPE_SIGNAL
-AC_FUNC_ALLOCA
-AC_FUNC_MEMCMP
-AC_FUNC_FSEEKO
-AC_CHECK_FUNCS(ftello)
-AC_REPLACE_FUNCS(dup2 memmove strcasecmp strncasecmp strerror strftime\
-		 strchr strstr strtoul crypt flock vsnprintf\
-		 isnan finite isinf hypot acosh erf)
-AC_CHECK_FUNCS(fmod killpg wait4 waitpid syscall chroot fsync getcwd eaccess\
-	      truncate ftruncate chsize times utimes fcntl lockf lstat symlink link\
-	      readlink setitimer setruid seteuid setreuid setresuid\
-	      setproctitle setrgid setegid setregid setresgid issetugid pause\
-	      lchown lchmod getpgrp setpgrp getpgid setpgid initgroups\
-	      getgroups setgroups getpriority getrlimit setrlimit sysconf\
-	      group_member dlopen sigprocmask\
-	      sigaction _setjmp setsid telldir seekdir fchmod\
-	      mktime timegm gettimeofday\
-	      cosh sinh tanh round setuid setgid setenv unsetenv)
-AC_ARG_ENABLE(setreuid,
-       [  --enable-setreuid       use setreuid()/setregid() according to need even if obsolete.],
-       [use_setreuid=$enableval])
-if test "$use_setreuid" = yes; then
-    AC_DEFINE(USE_SETREUID)
-    AC_DEFINE(USE_SETREGID)
-fi
-AC_STRUCT_TIMEZONE
-AC_CACHE_CHECK(for struct tm.tm_gmtoff, rb_cv_member_struct_tm_tm_gmtoff,
-  [AC_TRY_COMPILE([#include <time.h>],
-    [struct tm t; t.tm_gmtoff = 3600;],
-  [rb_cv_member_struct_tm_tm_gmtoff=yes],
-  [rb_cv_member_struct_tm_tm_gmtoff=no])])
-if test "$rb_cv_member_struct_tm_tm_gmtoff" = yes; then
-  AC_DEFINE(HAVE_STRUCT_TM_TM_GMTOFF)
-fi
-AC_CACHE_CHECK(for external int daylight, rb_cv_have_daylight,
-  [AC_TRY_LINK([#include <time.h>
-  int i;],
-  	[i = daylight;],
-	rb_cv_have_daylight=yes,
-	rb_cv_have_daylight=no)])
-if test "$rb_cv_have_daylight" = yes; then
-  AC_DEFINE(HAVE_DAYLIGHT)
-fi
-AC_DEFUN([RUBY_CHECK_VARTYPE], [dnl
-AC_CACHE_CHECK([for external $1], rb_cv_var_$1,
-  [rb_cv_var_$1=no
-  AC_TRY_COMPILE([#define _XOPEN_SOURCE 1
-    $2
-    const volatile void *volatile t;],
-    [t = &(&$1)[0];],
-    [for t in $3; do
-      AC_TRY_COMPILE([#define _XOPEN_SOURCE 1
-        $2
-        extern $t $1;
-        const volatile void *volatile t;],
-        [t = &(&$1)[0];],
-        [rb_cv_var_$1=$t; break])
-    done])])
-if test "[$rb_cv_var_]$1" != no; then
-  AC_DEFINE([HAVE_VAR_]m4_toupper($1))
-  AC_DEFINE_UNQUOTED([TYPEOF_VAR_]m4_toupper($1), $rb_cv_var_$1)
-fi])
-RUBY_CHECK_VARTYPE(timezone, [#include <time.h>], [long int])
-RUBY_CHECK_VARTYPE(altzone, [#include <time.h>], [long int])
-if test "$rb_cv_var_timezone" = no; then
-  AC_CHECK_FUNCS(timezone)
-  if test "$ ac_cv_func_timezone" = yes; then
-    AC_CACHE_CHECK([whether timezone requires zero arguments], rb_cv_func_timezone_void,
-      [AC_TRY_COMPILE([#include <time.h>],
-	[(void)timezone(0, 0);],
-	[rb_cv_func_timezone_void=no],
-	[rb_cv_func_timezone_void=yes])]
-    )
-    if test $rb_cv_func_timezone_void = yes; then
-      AC_DEFINE(TIMEZONE_VOID)
-    fi
-  fi
-fi
-
-AC_CACHE_CHECK(for negative time_t for gmtime(3), rb_cv_negative_time_t,
-  [AC_TRY_RUN([
-#include <time.h>
-
-void
-check(tm, y, m, d, h, s)
-    struct tm *tm;
-    int y, m, d, h, s;
-{
-    if (!tm ||
-	tm->tm_year != y ||
-	tm->tm_mon  != m-1 ||
-	tm->tm_mday != d ||
-	tm->tm_hour != h ||
-	tm->tm_sec  != s) {
-	exit(1);
-    }
-}
-
-int
-main()
-{
-   time_t t = -1;
-   struct tm *tm;
-
-   check(gmtime(&t), 69, 12, 31, 23, 59);
-   t = ~(time_t)0 << 31;
-   check(gmtime(&t), 1, 12, 13, 20, 52);
-   return 0;
-}
-],
-	rb_cv_negative_time_t=yes,
-	rb_cv_negative_time_t=no,
-	rb_cv_negative_time_t=yes)])
-if test "$rb_cv_negative_time_t" = yes; then
-  AC_DEFINE(NEGATIVE_TIME_T)
-fi
-
-if test "$ac_cv_func_sigprocmask" = yes && test "$ac_cv_func_sigaction" = yes; then
-   AC_DEFINE(POSIX_SIGNAL)
-else
-  AC_CHECK_FUNCS(sigsetmask)
-  AC_CACHE_CHECK(for BSD signal semantics, rb_cv_bsd_signal,
-    [AC_TRY_RUN([
-#include <stdio.h>
-#include <signal.h>
-
-void
-sig_handler(dummy)
-     int dummy;
-{
-}
-
-int
-main()
-{
-  signal(SIGINT, sig_handler);
-  kill(getpid(), SIGINT);
-  kill(getpid(), SIGINT);
-  return 0;
-}
-],
-	rb_cv_bsd_signal=yes,
-	rb_cv_bsd_signal=no,
-	rb_cv_bsd_signal=$ac_cv_func_sigsetmask)])
-  if test "$rb_cv_bsd_signal" = yes; then
-    AC_DEFINE(BSD_SIGNAL)
-  fi
-fi
-
-AC_FUNC_GETPGRP
-AC_FUNC_SETPGRP  
-
-AC_C_BIGENDIAN
-AC_C_CONST
-AC_C_CHAR_UNSIGNED
-AC_C_INLINE
-AC_C_VOLATILE
-
-AC_CACHE_CHECK(whether right shift preserve sign bit, rb_cv_rshift_sign,
-    [AC_TRY_RUN([
-int
-main()
-{
-  if (-1==(-1>>1))
-    return 0;
-  return 1;
-}
-],
-	rb_cv_rshift_sign=yes,
-	rb_cv_rshift_sign=no,
-	rb_cv_rshift_sign=yes)])
-if test "$rb_cv_rshift_sign" = yes; then
-  AC_DEFINE(RSHIFT(x,y), ((x)>>(int)y))
-else
-  AC_DEFINE(RSHIFT(x,y), (((x)<0) ? ~((~(x))>>y) : (x)>>y))
-fi
-
-AC_MSG_CHECKING(read count field in FILE structures)
-AC_CACHE_VAL(rb_cv_fcnt,
-[for fcnt in	dnl
-     _cnt	dnl
-     __cnt	dnl
-     _r		dnl
-     readCount	dnl
-     _rcount	dnl for emx0.9c
-; do
-    AC_TRY_COMPILE([#include <stdio.h>
-],
-	[FILE *f = stdin; f->$fcnt = 0;],
-	rb_cv_fcnt="$fcnt"; break,
-	rb_cv_fcnt="not found")
-done])
-if test "$rb_cv_fcnt" = "not found"; then
-  AC_MSG_RESULT([not found(OK if using GNU libc)])
-else
-  AC_MSG_RESULT($rb_cv_fcnt)
-  AC_DEFINE_UNQUOTED(FILE_COUNT, $rb_cv_fcnt)
-fi
-
-AC_MSG_CHECKING(read buffer ptr field in FILE structures)
-AC_CACHE_VAL(rb_cv_frptr,
-[for frptr in	dnl
-     _IO_read_ptr dnl
-     _ptr	dnl
-     __ptr	dnl
-     bufpos	dnl
-     _p	dnl
-; do
-    AC_TRY_COMPILE([#include <stdio.h>
-],
-	[FILE *f = stdin; char buf[256]; f->$frptr = buf;],
-	rb_cv_frptr="$frptr"; break,
-	rb_cv_frptr="not found")
-done])
-if test "$rb_cv_frptr" = "not found"; then
-  AC_MSG_RESULT([not found])
-else
-  AC_MSG_RESULT($rb_cv_frptr)
-  AC_DEFINE_UNQUOTED(FILE_READPTR, $rb_cv_frptr)
-
-  if test "$rb_cv_fcnt" = "not found"; then
-    AC_MSG_CHECKING(read buffer end field in FILE structures)
-    AC_CACHE_VAL(rb_cv_frend,
-    [for frend in	dnl
-         _IO_read_end	dnl
-         bufread 	dnl
-    ; do
-        AC_TRY_COMPILE([#include <stdio.h>
-    ],
-  	[FILE *f = stdin; char buf[256]; f->$frend = buf;],
-  	rb_cv_frend="$frend"; break,
-  	rb_cv_frend="not found")
-    done])
-    if test "$rb_cv_frend" = "not found"; then
-      AC_MSG_RESULT([not found])
-    else
-      AC_MSG_RESULT($rb_cv_frend)
-      AC_DEFINE_UNQUOTED(FILE_READEND, $rb_cv_frend)
-    fi
-  fi
-fi
-
-AC_DEFUN([RUBY_CHECK_IO_NEED],
-[AC_CACHE_CHECK(whether need to [$1], [$2],
-    [AC_TRY_RUN([
-#include <stdio.h>
-#ifndef SEEK_SET
-#define SEEK_SET 0
-#endif
-#ifndef SEEK_CUR
-#define SEEK_CUR 1
-#endif
-#define before_seek(f) ]ifelse(index($2,flush_before_seek),-1,[fflush(f)],[(f,0)])[
-#define reset_rw(f) ]ifelse(index($2,seek_between_rw),-1,[do_seek(f,SEEK_CUR)],[(f,0)])[
-#define do_seek(f, w) (before_seek(f), fseek(f,0,w))
-
-char *fn = "conftest.dat";
-char *wombat = "wombat\n";
-char *koara = "koara\n";
-char *kangaroo = "kangaroo\n";
-
-int main()
-{
-    char buf[BUFSIZ];
-    FILE *f;
-    int r = 1;
-
-    if (!(f = fopen(fn, "w+"))) return 1;
-    fputs(wombat, f);
-    do_seek(f, SEEK_SET);
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, wombat)) goto fail;
-    reset_rw(f);
-    fputs(koara, f);
-    fputs(kangaroo, f);
-    do_seek(f, SEEK_SET);
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, wombat)) goto fail;
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, koara)) goto fail;
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, kangaroo)) goto fail;
-    do_seek(f, SEEK_SET);
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, wombat)) goto fail;
-    reset_rw(f);
-    fputc('X', f);
-    reset_rw(f);
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, koara+1)) goto fail;
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, kangaroo)) goto fail;
-    do_seek(f, SEEK_SET);
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, wombat)) goto fail;
-    if (!fgets(buf, BUFSIZ, f) || buf[0] != 'X' || strcmp(buf+1, koara+1)) goto fail;
-    if (!fgets(buf, BUFSIZ, f) || strcmp(buf, kangaroo)) goto fail;
-    r = 0;
-  fail:
-    fclose(f);
-    unlink(fn);
-    return r;
-}
-], [$2]=no, [$2]=yes, [$2]=[$3])])])
-RUBY_CHECK_IO_NEED(seek between R/W, rb_cv_need_io_seek_between_rw, yes)
-if test "$rb_cv_need_io_seek_between_rw" = yes; then
-  AC_DEFINE(NEED_IO_SEEK_BETWEEN_RW, 1)
-fi
-dnl RUBY_CHECK_IO_NEED(flush before seek, rb_cv_need_io_flush_before_seek, no)
-dnl if test "$rb_cv_need_io_flush_before_seek" = yes; then
-dnl   AC_DEFINE(NEED_IO_FLUSH_BEFORE_SEEK, 1)
-dnl fi
-
-AC_CACHE_CHECK([whether st_ino is huge], rb_cv_huge_st_ino,
-[AC_COMPILE_IFELSE([AC_LANG_BOOL_COMPILE_TRY([
-#include <sys/stat.h>
-struct stat test_stat;
-], [sizeof(test_stat.st_ino)>sizeof(long)])],
-rb_cv_huge_st_ino=yes,
-rb_cv_huge_st_ino=no)
-])
-if test $rb_cv_huge_st_ino = yes; then
-  AC_DEFINE(HUGE_ST_INO)
-fi
-
-if test "$ac_cv_func_sysconf" = yes; then
-  AC_DEFUN([RUBY_CHECK_SYSCONF], [dnl
-  AC_CACHE_CHECK([whether _SC_$1 is supported], rb_cv_have_sc_[]m4_tolower($1),
-    [AC_TRY_COMPILE([#include <unistd.h>
-      ],
-      [_SC_$1 >= 0],
-      rb_cv_have_sc_[]m4_tolower($1)=yes,
-      rb_cv_have_sc_[]m4_tolower($1)=no)
-    ])
-  if test "$rb_cv_have_sc_[]m4_tolower($1)" = yes; then
-    AC_DEFINE(HAVE__SC_$1)
-  fi
-  ])
-  RUBY_CHECK_SYSCONF(CLK_TCK)
-fi
-
-case "$target_cpu" in
-m68*|i?86|ia64|sparc*|alpha*) rb_cv_stack_grow_dir=-1;;
-hppa*) rb_cv_stack_grow_dir=+1;;
-esac
-AC_CACHE_CHECK(stack growing direction, rb_cv_stack_grow_dir,
-  [AC_TRY_RUN([
-/* recurse to get rid of inlining */
-static int
-stack_growup_p(addr, n)
-    volatile int *addr, n;
-{
-    volatile int end;
-    if (n > 0)
-	return *addr = stack_growup_p(addr, n - 1);
-    else
-	return (&end > addr);
-}
-int main()
-{
-    int x;
-    return stack_growup_p(&x, 10);
-}
-], rb_cv_stack_grow_dir=-1, rb_cv_stack_grow_dir=+1, rb_cv_stack_grow_dir=0)])
-AC_DEFINE_UNQUOTED(STACK_GROW_DIRECTION, $rb_cv_stack_grow_dir)
-
-if test x"$enable_pthread" = xyes; then
-    for pthread_lib in pthread pthreads c c_r; do
- 	AC_CHECK_LIB($pthread_lib, pthread_kill,
-		     rb_with_pthread=yes, rb_with_pthread=no)
-	if test "$rb_with_pthread" = "yes"; then break; fi
-    done
-    if test x"$rb_with_pthread" = xyes; then
-	AC_DEFINE(_REENTRANT)
-	AC_DEFINE(_THREAD_SAFE)
-	AC_DEFINE(HAVE_LIBPTHREAD)
-	case $pthread_lib in
-	c)
-	    ;;
-	c_r)
-	    MAINLIBS="-pthread $MAINLIBS"
-	    ;;
-	*)
-      	    LIBS="-l$pthread_lib $LIBS"
-	    ;;
-	esac
-    else
-	AC_MSG_WARN("Don't know how to find pthread library on your system -- thread support disabled")
-    fi
-    AC_CHECK_FUNCS(nanosleep)
-    if test x"$ac_cv_func_nanosleep" = xno; then
-       AC_CHECK_LIB(rt, nanosleep)
-       if test x"$ac_cv_lib_rt_nanosleep" = xyes; then
-           AC_DEFINE(HAVE_NANOSLEEP)
-       fi
-    fi
-fi
-if test x"$ac_cv_header_ucontext_h" = xyes; then
-    if test x"$target_cpu" = xia64 -o x"$rb_with_pthread" = xyes; then
-	AC_CHECK_FUNCS(getcontext setcontext)
-    fi
-fi
-
-dnl default value for $KANJI
-DEFAULT_KCODE="KCODE_NONE"
-
-AC_ARG_WITH(default-kcode, 
-	[  --with-default-kcode=CODE specify default value for \$KCODE (utf8|euc|sjis|none)], 
-	[case $withval in
-	utf8) DEFAULT_KCODE="KCODE_UTF8";;
-	euc)  DEFAULT_KCODE="KCODE_EUC";;
-	sjis) DEFAULT_KCODE="KCODE_SJIS";;
-	none) DEFAULT_KCODE="KCODE_NONE";;
-        *)    AC_MSG_WARN($withval is not valid kcode; ignored);;
-	esac])
-AC_DEFINE_UNQUOTED(DEFAULT_KCODE, $DEFAULT_KCODE)
-
-dnl wheather use dln_a_out or not
-AC_ARG_WITH(dln-a-out, 
-	[  --with-dln-a-out        use dln_a_out if possible], [
-	case $withval in
-	yes) with_dln_a_out=yes;;
-	*)   with_dln_a_out=no;;
-	esac], [with_dln_a_out=no])
-
-AC_CACHE_CHECK(whether ELF binaries are produced, rb_cv_binary_elf,
-[AC_TRY_RUN([
-/* Test for whether ELF binaries are produced */
-#include <fcntl.h>
-#include <stdlib.h>
-main() {
-	char buffer[4];
-	int i=open("conftest",O_RDONLY);
-	if(i==-1)
-		exit(1); /* fail */
-	if(read(i,&buffer[0],4)<4)
-		exit(1); /* fail */
-	if(buffer[0] != 127 || buffer[1] != 'E' ||
-           buffer[2] != 'L' || buffer[3] != 'F')
-		exit(1); /* fail */
-	exit(0); /* succeed (yes, it's ELF) */
-}
-],
-rb_cv_binary_elf=yes,
-rb_cv_binary_elf=no,
-rb_cv_binary_elf=yes)])
-
-if test "$rb_cv_binary_elf" = yes; then
-  AC_DEFINE(USE_ELF)
-fi
-
-case "$target_os" in
-linux* | gnu* | k*bsd*-gnu | bsdi*)
-    if test "$rb_cv_binary_elf" = no; then
-	with_dln_a_out=yes
-    else
-	LDFLAGS="$LDFLAGS -rdynamic"
-    fi;;
-esac
-LIBEXT=a
-
-AC_SUBST(DLDFLAGS)dnl
-AC_SUBST(ARCH_FLAG)dnl
-
-AC_SUBST(STATIC)dnl
-AC_SUBST(CCDLFLAGS)dnl
-AC_SUBST(LDSHARED)dnl
-AC_SUBST(DLEXT)dnl
-AC_SUBST(DLEXT2)dnl
-AC_SUBST(LIBEXT)dnl
-
-STATIC=
-: ${LIBPATHFLAG=' -L"%s"'}
-: ${PATHFLAG=''}
-
-if test "$with_dln_a_out" != yes; then
-  rb_cv_dlopen=unknown
-  AC_MSG_CHECKING(whether OS depend dynamic link works)
-  if test "$GCC" = yes; then
-    case "$target_os" in
-    nextstep*)	CCDLFLAGS="$CCDLFLAGS -fno-common";;
-    openstep*)	CCDLFLAGS="$CCDLFLAGS -fno-common";;
-    rhapsody*)	CCDLFLAGS="$CCDLFLAGS -fno-common";;
-    darwin*)	CCDLFLAGS="$CCDLFLAGS -fno-common";;
-    human*|bsdi*|beos*|cygwin*|mingw*|aix*|interix*) ;;
-    *) CCDLFLAGS="$CCDLFLAGS -fPIC";;
-    esac
-  else
-    case "$target_os" in
-	hpux*)		CCDLFLAGS="$CCDLFLAGS +Z";;
-	solaris*|irix*)	CCDLFLAGS="$CCDLFLAGS -KPIC" ;;
-	sunos*)		CCDLFLAGS="$CCDLFLAGS -PIC" ;;
-	esix*|uxpds*)   CCDLFLAGS="$CCDLFLAGS -KPIC" ;;
-	*)		: ${CCDLFLAGS=""} ;;
-    esac
-  fi
-
-  case "$target_os" in
-	hpux*)		DLDFLAGS="$DLDFLAGS -E"
-			: ${LDSHARED='ld -b'}
-			XLDFLAGS="$XLDFLAGS -Wl,-E"
-			: ${LIBPATHENV=SHLIB_PATH}
-			rb_cv_dlopen=yes;;
-	solaris*) 	if test "$GCC" = yes; then
-                          : ${LDSHARED='$(CC) -shared'}
-			   if test "$rb_cv_prog_gnu_ld" = yes; then
-	 		       LDFLAGS="$LDFLAGS -Wl,-E"
-			   fi
-			else
-			   : ${LDSHARED='ld -G'}
-			fi
-			rb_cv_dlopen=yes;;
-	sunos*) 	: ${LDSHARED='ld -assert nodefinitions'}
-			rb_cv_dlopen=yes;;
-	irix*)		: ${LDSHARED='ld -shared'}
-			rb_cv_dlopen=yes;;
-	sysv4*) 	: ${LDSHARED='ld -G'}
-			rb_cv_dlopen=yes;;
-        nto-qnx*)       : ${LDSHARED="qcc -shared"}
-                        rb_cv_dlopen=yes ;;
-	esix*|uxpds*) 	: ${LDSHARED="ld -G"}
-			rb_cv_dlopen=yes ;;
-	osf*) 		: ${LDSHARED="ld -shared -expect_unresolved \"*\""}
-			rb_cv_dlopen=yes ;;
-	bsdi3*) 	case "$CC" in
-			*shlicc*)	: ${LDSHARED="$CC -r"}
-					rb_cv_dlopen=yes ;;
-			esac ;;
-	linux* | gnu* | k*bsd*-gnu | netbsd* | bsdi*)
-			: ${LDSHARED='${CC} -shared'}
-			if test "$rb_cv_binary_elf" = yes; then
-			    LDFLAGS="$LDFLAGS -Wl,-export-dynamic"
-			fi
-			rb_cv_dlopen=yes ;;
-	interix*) 	: ${LDSHARED="$CC -shared"}
-			XLDFLAGS="$XLDFLAGS -Wl,-E"
-			LIBPATHFLAG=" -L'%1\$-s'"
-			rb_cv_dlopen=yes ;;
-	freebsd*|dragonfly*)       : ${LDSHARED="$CC -shared"}
-			if test "$rb_cv_binary_elf" = yes; then
-			    LDFLAGS="$LDFLAGS -rdynamic"
-			    DLDFLAGS="$DLDFLAGS "'-Wl,-soname,$(.TARGET)'
-			else
-			  test "$GCC" = yes && test "$rb_cv_prog_gnu_ld" = yes || LDSHARED="ld -Bshareable"
-			fi
-			rb_cv_dlopen=yes ;;
-	openbsd*) 	: ${LDSHARED="\$(CC) -shared ${CCDLFLAGS}"}
-			if test "$rb_cv_binary_elf" = yes; then
-			    LDFLAGS="$LDFLAGS -Wl,-E"
-			fi
-			rb_cv_dlopen=yes ;;
-	nextstep*)	: ${LDSHARED='cc -r -nostdlib'}
-			LDFLAGS="$LDFLAGS -u libsys_s"
-			rb_cv_dlopen=yes ;;
-	openstep*)	: ${LDSHARED='cc -dynamic -bundle -undefined suppress'}
-			: ${LDFLAGS=""}
-			rb_cv_dlopen=yes ;;
-	rhapsody*)	: ${LDSHARED='cc -dynamic -bundle -undefined suppress'}
-			: ${LDFLAGS=""}
-			rb_cv_dlopen=yes ;;
-	darwin*)	: ${LDSHARED='cc -dynamic -bundle -undefined suppress -flat_namespace'}
-			: ${LDFLAGS=""}
-			: ${LIBPATHENV=DYLD_LIBRARY_PATH}
-			rb_cv_dlopen=yes ;;
-        aix*)           if test "$GCC" = yes; then
-                           : ${LDSHARED='gcc -shared'}
-                           DLDFLAGS='-Wl,-G -eInit_$(TARGET)'
-                           LDFLAGS='-Wl,-brtl -Wl,-bE:ruby.imp'
-                        else
-                          : ${LDSHARED='/usr/ccs/bin/ld'}
-                          DLDFLAGS='-G -eInit_$(TARGET)'
-                          LDFLAGS="-brtl -bE:ruby.imp"
-                        fi
-                        : ${ARCHFILE="ruby.imp"}
-                        TRY_LINK='$(CC) $(LDFLAGS) -oconftest $(INCFLAGS) -I$(hdrdir) $(CPPFLAGS)'
-                        TRY_LINK="$TRY_LINK"' $(CFLAGS) $(src) $(LIBPATH) $(LOCAL_LIBS) $(LIBS)'
-                        : ${LIBPATHENV=SHLIB_PATH}
-                        rb_cv_dlopen=yes ;;
-	human*)		: ${DLDFLAGS=''}
-			: ${LDSHARED=''}
-			: ${LDFLAGS=''}
-			: ${LINK_SO='ar cru $@ $(OBJS)'}
-			rb_cv_dlopen=yes ;;
-	beos*) 		case "$target_cpu" in
-			  powerpc*)
-			    : ${LDSHARED="ld -xms"}
-			    DLDFLAGS="$DLDFLAGS "'-export Init_$(TARGET) -lbe -lroot glue-noinit.a init_term_dyn.o start_dyn.o'
-                            ;;
-			  i586*)
-			    : ${LDSHARED="ld -shared"}
-			    DLDFLAGS="$DLDFLAGS -L/boot/develop/lib/x86 -lbe -lroot"
-			    ;;
-			esac
-			: ${LIBPATHENV=LIBRARY_PATH}
-			rb_cv_dlopen=yes ;;
-	nto-qnx*)	DLDFLAGS="$DLDFLAGS -L/lib -L/usr/lib -L/usr/local/lib"
-			: ${LDSHARED='ld -Bshareable -x'}
-			LDFLAGS="$LDFLAGS -L/lib -L/usr/lib -L/usr/local/lib"
-			rb_cv_dlopen=yes;;
-	cygwin*|mingw*)	: ${LDSHARED="${CC} -shared -s"}
-			XLDFLAGS="$XLDFLAGS -Wl,--stack,0x02000000"
-			DLDFLAGS="${DLDFLAGS} -Wl,--enable-auto-import,--export-all"
-			: ${LIBPATHENV=""}
-			rb_cv_dlopen=yes ;;
-	hiuxmpp)	: ${LDSHARED='ld -r'} ;;
-	atheos*) 	: ${LDSHARED="$CC -shared"}
-			rb_cv_dlopen=yes ;;
-	os2-emx*)	LDFLAGS="$LDFLAGS -Zbsd-signals"
-			;;
-	*) 		: ${LDSHARED='ld'} ;;
-  esac
-  AC_MSG_RESULT($rb_cv_dlopen)
-
-  AC_ARG_ENABLE(rpath,
-       [  --disable-rpath         embed run path into extension libraries.],
-       [enable_rpath=$enableval], [enable_rpath="$rb_cv_binary_elf"])
-  if test "$enable_rpath" = yes; then
-    LIBPATHFLAG=" -L'%1\$-s'"
-    if test "$GCC" = yes; then
-       RPATHFLAG=" -Wl,-R'%1\$-s'"
-    else
-       RPATHFLAG=" -R'%1\$-s'"
-    fi
-  fi
-fi
-AC_SUBST(LINK_SO)
-AC_SUBST(LIBPATHFLAG)
-AC_SUBST(RPATHFLAG)
-AC_SUBST(LIBPATHENV, "${LIBPATHENV-LD_LIBRARY_PATH}")
-AC_SUBST(TRY_LINK)
-
-dln_a_out_works=no
-if test "$ac_cv_header_a_out_h" = yes; then
-  if test "$with_dln_a_out" = yes || test "$rb_cv_dlopen" = unknown; then
-    cat confdefs.h > config.h
-    AC_CACHE_CHECK(whether matz's dln works, rb_cv_dln_a_out,
-    [AC_TRY_COMPILE([
-#define USE_DLN_A_OUT
-#include "dln.c"
-],
-	[], 
-	rb_cv_dln_a_out=yes,
-	rb_cv_dln_a_out=no)])
-    if test "$rb_cv_dln_a_out" = yes; then
-      dln_a_out_works=yes
-      AC_DEFINE(USE_DLN_A_OUT)
-    fi
-  fi
-fi
-
-if test "$dln_a_out_works" = yes; then
-  if test "$GCC" = yes; then
-    STATIC=-static
-  else
-    STATIC=-Bstatic
-  fi
-  DLEXT=so
-  AC_DEFINE(DLEXT, ".so")
-  CCDLFLAGS=
-else
-  case "$target_os" in
-    hpux*)	DLEXT=sl
-	 	AC_DEFINE(DLEXT, ".sl");;
-    nextstep*)	DLEXT=bundle
-	 	AC_DEFINE(DLEXT, ".bundle");;
-    openstep*)	DLEXT=bundle
-	 	AC_DEFINE(DLEXT, ".bundle");;
-    rhapsody*)	DLEXT=bundle
-	 	AC_DEFINE(DLEXT, ".bundle");;
-    darwin*)	DLEXT=bundle
-	 	AC_DEFINE(DLEXT, ".bundle");;
-    os2-emx*)	DLEXT=dll
-	 	AC_DEFINE(DLEXT, ".dll");;
-    cygwin*|mingw*)	DLEXT=so
-	 	AC_DEFINE(DLEXT, ".so")
-		DLEXT2=dll
-	 	AC_DEFINE(DLEXT2, ".dll");;
-    *)		DLEXT=so
-    		AC_DEFINE(DLEXT, ".so");;
-  esac
-fi
-
-AC_SUBST(STRIP)dnl
-if test "$with_dln_a_out" = yes; then
-  STRIP=true
-else
-  STRIP=strip
-fi
-
-case "$target_os" in
-  linux* | gnu* | k*bsd*-gnu)
-	STRIP='strip -S -x';;
-  nextstep*)
-	STRIP='strip -A -n';;
-  openstep*)
-	STRIP='strip -A -n';;
-  rhapsody*)
-	STRIP='strip -A -n';;
-  darwin*)
-	STRIP='strip -A -n';;
-esac
-
-EXTSTATIC=
-AC_SUBST(EXTSTATIC)dnl
-AC_ARG_WITH(static-linked-ext,
-	    [  --with-static-linked-ext link external modules statically],
-            [case $withval in
-	     yes) STATIC=
-		  EXTSTATIC=static;;
-	     *)	;;
-	     esac])
-
-case "$target_os" in
-  human*)
-    AC_CHECK_LIB(signal, _harderr)
-    AC_CHECK_LIB(hmem, hmemset)
-    AC_CHECK_FUNCS(select)
-    AC_CACHE_CHECK(whether PD libc _dtos18 fail to convert big number,
-		   rb_cv_missing__dtos18,
-    [AC_TRY_RUN(
-changequote(<<, >>)dnl
-<<
-#include <stdio.h>
-main ()
-{
-   char buf[256];
-   sprintf (buf, "%g", 1e+300);
-   exit (strcmp (buf, "1e+300") ? 0 : 1);
-}
->>,
-changequote([, ])dnl
-rb_cv_missing__dtos18=yes, rb_cv_missing__dtos18=no, rb_cv_missing__dtos18=no)])
-    if test "$rb_cv_missing__dtos18" = yes; then
-      AC_DEFINE(MISSING__DTOS18)
-    fi
-    AC_CACHE_CHECK(whether PD libc fconvert fail to round,
-		   rb_cv_missing_fconvert,
-    [AC_TRY_RUN(
-changequote(<<, >>)dnl
-<<
-#include <stdio.h>
-#include <math.h>
-main ()
-{
-  char buf[256];
-  sprintf (buf, "%f", log(exp(1.0)));
-  exit (strcmp (buf, "1.000000") ? 0 : 1);
-}
->>,
-changequote([, ])dnl
-rb_cv_missing_fconvert=yes, rb_cv_missing_fconvert=no, rb_cv_missing_fconvert=no)])
-    if test "$rb_cv_missing_fconvert" = yes; then
-      AC_DEFINE(MISSING_FCONVERT)
-    fi
-    AC_LIBOBJ([x68.o])
-    CFLAGS="$CFLAGS -fansi-only"
-    XCFLAGS="$XCFLAGS -cc1-stack=262144 -cpp-stack=2694144"
-    EXEEXT=.x
-    OBJEXT=o
-    setup=Setup.x68
-    ;;
-  dnl OS/2 environment w/ Autoconf 2.1x for EMX
-  os2-emx)
-    AC_LIBOBJ([os2])
-    setup=Setup.emx
-    ;;
-  *djgpp*)
-    setup=Setup.dj
-    ;;
-  *)
-    setup=Setup
-    ;;
-esac
-
-AC_SUBST(setup)
-
-if test "$prefix" = NONE; then
-  prefix=$ac_default_prefix
-fi
-
-#if test "$fat_binary" != no ; then
-#  CFLAGS="$CFLAGS $ARCH_FLAG"
-#fi
-
-if test x"$cross_compiling" = xyes; then
-  test x"$MINIRUBY" = x && MINIRUBY="${RUBY-ruby} -I`pwd` -rfake"
-  PREP=fake.rb
-  RUNRUBY='$(MINIRUBY) -I`cd $(srcdir)/lib; pwd`'
-else
-  MINIRUBY='./miniruby$(EXEEXT)'
-  PREP='miniruby$(EXEEXT)'
-  RUNRUBY='$(MINIRUBY) $(srcdir)/runruby.rb --extout=$(EXTOUT) --'
-fi
-AC_SUBST(MINIRUBY)
-AC_SUBST(PREP)
-AC_SUBST(RUNRUBY)
-AC_SUBST(EXTOUT, [${EXTOUT-.ext}])
-
-FIRSTMAKEFILE=""
-LIBRUBY_A='lib$(RUBY_SO_NAME)-static.a'
-LIBRUBY='$(LIBRUBY_A)'
-LIBRUBYARG_STATIC='-l$(RUBY_SO_NAME)-static'
-LIBRUBYARG='$(LIBRUBYARG_STATIC)'
-SOLIBS=
-
-case "$target_os" in
-  cygwin*|mingw*|beos*|openstep*|nextstep*|rhapsody*|darwin*|os2-emx*)
-    : ${DLDLIBS=""}
-    ;;
-  *)
-    DLDLIBS="$DLDLIBS -lc"
-    ;;
-esac
-
-RUBY_SO_NAME='$(RUBY_INSTALL_NAME)'
-LIBRUBY_LDSHARED=$LDSHARED
-LIBRUBY_DLDFLAGS=$DLDFLAGS
-LIBRUBY_SO='lib$(RUBY_SO_NAME).so.$(MAJOR).$(MINOR).$(TEENY)'
-LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).so'
-ENABLE_SHARED=no
-
-AC_ARG_ENABLE(shared,
-       [  --enable-shared         build a shared library for Ruby. ],
-       [enable_shared=$enableval])
-if test "$enable_shared" = 'yes'; then
-  LIBRUBY='$(LIBRUBY_SO)'
-  LIBRUBYARG_SHARED='-l$(RUBY_SO_NAME)'
-  LIBRUBYARG='$(LIBRUBYARG_SHARED)'
-  CFLAGS="$CFLAGS $CCDLFLAGS"
-  ENABLE_SHARED=yes
-  if test "$rb_cv_binary_elf" = yes; then
-    SOLIBS='$(LIBS)'
-  fi
-  case "$target_os" in
-    sunos4*)
-	LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).so.$(MAJOR).$(MINOR) lib$(RUBY_SO_NAME).so'
-	;;
-    linux* | gnu* | k*bsd*-gnu | atheos*)
-	LIBRUBY_DLDFLAGS='-Wl,-soname,lib$(RUBY_SO_NAME).so.$(MAJOR).$(MINOR)'
-	LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).so.$(MAJOR).$(MINOR) lib$(RUBY_SO_NAME).so'
-	;;
-    freebsd*|dragonfly*)
-	SOLIBS='$(LIBS)'
-	LIBRUBY_SO='lib$(RUBY_SO_NAME).so.$(MAJOR)$(MINOR)'
-	if test "$rb_cv_binary_elf" != "yes" ; then
-	    LIBRUBY_SO="$LIBRUBY_SO.\$(TEENY)"
-	    LIBRUBY_ALIASES=''
-	fi
-	;;
-    netbsd*)
-	SOLIBS='$(LIBS)'
-	LIBRUBY_SO='lib$(RUBY_SO_NAME).so.$(MAJOR)$(MINOR).$(TEENY)'
-	LIBRUBY_DLDFLAGS='-Wl,-soname,lib$(RUBY_SO_NAME).so.$(MAJOR)$(MINOR)'
-	if test "$rb_cv_binary_elf" = yes; then # ELF platforms
-	   LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).so.$(MAJOR)$(MINOR) lib$(RUBY_SO_NAME).so'
-	else	# a.out platforms
-	   LIBRUBY_ALIASES=""
-	fi
- 	;;
-    openbsd*)
-	SOLIBS='$(LIBS)'
-	LIBRUBY_SO='lib$(RUBY_INSTALL_NAME).so.$(MAJOR).'`expr ${MINOR} \* 10 + ${TEENY}`
-	;;
-    solaris*)
-	SOLIBS='$(LIBS)'
-	LIBRUBY_SO='lib$(RUBY_SO_NAME).so.$(MAJOR)'
-	LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).so.$(MAJOR).$(MINOR).$(TEENY) lib$(RUBY_SO_NAME).so'
-	if test "$GCC" = yes; then
-	    LIBRUBY_DLDFLAGS="$DLDFLAGS "'-Wl,-h,$(@F)'
-	fi
-	XLDFLAGS="$XLDFLAGS "'-R${libdir}'
- 	;;
-    hpux*)
-	XLDFLAGS="$XLDFLAGS "'-Wl,+s,+b,$(libdir)'
-	LIBRUBY_SO='lib$(RUBY_SO_NAME).sl.$(MAJOR).$(MINOR).$(TEENY)'
-	LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).sl.$(MAJOR).$(MINOR) lib$(RUBY_SO_NAME).sl'
-	;;
-    aix*)
-	if test "$GCC" = yes; then
-	    LIBRUBY_LDSHARED='$(CC) -shared'
-	    LIBRUBY_DLDFLAGS='-Wl,-G -Wl,-bE:ruby.imp -Wl,-bnoentry'
-	else
-	    LIBRUBY_LDSHARED='/usr/ccs/bin/ld'
-	    LIBRUBY_DLDFLAGS='-G -bE:ruby.imp -bnoentry'
-	fi
-	LIBRUBYARG_SHARED='-L${libdir} -lruby'
-	SOLIBS='-lm -lc'
-	;;
-    beos*)
-	case "$target_cpu" in
-	powerpc*)
-	    LIBRUBY_DLDFLAGS='-f ruby.exp -lnet -lbe -lroot glue-noinit.a init_term_dyn.o start_dyn.o'
-	    ;;
-	esac
-	;;
-    darwin*)
-	LIBRUBY_SO='lib$(RUBY_SO_NAME).$(MAJOR).$(MINOR).$(TEENY).dylib'
-	LIBRUBY_LDSHARED='cc -dynamiclib -undefined suppress -flat_namespace'
-	LIBRUBY_DLDFLAGS='-install_name $(libdir)/lib$(RUBY_SO_NAME).dylib -current_version $(MAJOR).$(MINOR).$(TEENY) -compatibility_version $(MAJOR).$(MINOR)'
-	LIBRUBY_ALIASES='lib$(RUBY_SO_NAME).$(MAJOR).$(MINOR).dylib lib$(RUBY_SO_NAME).dylib'
-	;;
-    interix*)
-	LIBRUBYARG_SHARED='-L${libdir} -L. -l$(RUBY_SO_NAME)'
-	;;
-    *)
-	;;
-  esac
-fi
-if test "$enable_rpath" = yes; then
-    if test "$GCC" = yes; then
-       LIBRUBYARG_SHARED='-Wl,-R -Wl,$(libdir) -L$(libdir) -L. '"$LIBRUBYARG_SHARED"
-    else
-       LIBRUBYARG_SHARED='-R $(libdir) -L$(libdir) -L. '"$LIBRUBYARG_SHARED"
-    fi
-fi
-
-XLDFLAGS="$XLDFLAGS -L."
-AC_SUBST(ARCHFILE)
-
-dnl build rdoc index if requested
-RDOCTARGET=""
-AC_ARG_ENABLE(install-doc,
-       [  --enable-install-doc    build and install rdoc indexes during install ],
-       [install_doc=$enableval], [install_doc=no])
-if test "$install_doc" != no; then
-   RDOCTARGET="install-doc"
-fi
-AC_SUBST(RDOCTARGET)
-
-case "$target_os" in
-    linux*)
-	XCFLAGS="$XCFLAGS -D_GNU_SOURCE=1"
-	;;
-    netbsd*)
-    	CFLAGS="$CFLAGS -pipe"
-	;;
-    nextstep*|openstep*)
-	# The -fno-common is needed if we wish to embed the Ruby interpreter
-	# into a plugin module of some project (as opposed to embedding it
-	# within the project's application).  The -I/usr/local/include is
-	# needed because CPP as discovered by configure (cc -E -traditional)
-	# fails to consult /usr/local/include by default.  This causes
-	# mkmf.rb's have_header() to fail if the desired resource happens to be
-	# installed in the /usr/local tree.
-    	CFLAGS="$CFLAGS -pipe -fno-common"
-	CPPFLAGS="$CPPFLAGS -I/usr/local/include"
-	;;
-    rhapsody*)
-    	CFLAGS="$CFLAGS -pipe -no-precomp -fno-common"
-	;;
-    darwin*)
-    	CFLAGS="$CFLAGS -pipe -fno-common"
-	;;
-    os2-emx)
-    	CFLAGS="$CFLAGS -DOS2 -Zmts"
-	LIBRUBY_A=`echo $LIBRUBY_A | sed 's/^lib//'`
-	LIBRUBY_SO=`echo $LIBRUBY_SO | sed 's/^lib//'`
-	LIBRUBY_ALIASES=`for i in $LIBRUBY_ALIASES; do echo "$i"; done | sed 's/^lib//'`
-	;;
-    osf*)
-	if test "$GCC" != "yes" ; then
-	  # compile something small: taint.c is fine for this.
-	  # the main point is the '-v' flag of 'cc'.
-	  case "`cc -v -I. -c main.c -o /tmp/main.o 2>&1`" in
-	  */gemc_cc*)   # we have the new DEC GEM CC
-                        CFLAGS="$CFLAGS -oldc"
-                        ;;
-          *)            # we have the old MIPS CC
-                        ;;
-          esac
-	  # cleanup
-	  rm -f /tmp/main.o
-	  CFLAGS="$CFLAGS -std"
-	fi
-		;;
-    beos*)
-	case "$target_cpu" in
-	powerpc*)
-	    CFLAGS="$CFLAGS -relax_pointers"
-	    ;;
-	esac
-	;;
-    cygwin*|mingw*)
-	case "$target_os" in
-	cygwin*)
-	    if test x"$enable_shared" = xyes; then
-		LIBRUBY_SO='cyg$(RUBY_SO_NAME)'${MAJOR}${MINOR}.dll
-		LIBRUBY='lib$(RUBY_SO_NAME).dll.a'
-	    fi
-	    ;;
-	mingw*)
-	    RUBY_SO_NAME=${rb_cv_msvcrt}-'$(RUBY_INSTALL_NAME)'${MAJOR}${MINOR}
-	    if test x"$enable_shared" = xyes; then
-		LIBRUBY_SO='$(RUBY_SO_NAME)'.dll
-		LIBRUBY='lib$(LIBRUBY_SO).a'
-	    fi
-	    AC_LIBOBJ([win32])
-	    COMMON_LIBS=m
-#	    COMMON_MACROS="WIN32_LEAN_AND_MEAN="
-	    COMMON_HEADERS="windows.h winsock.h"
-	    ;;
-	esac
-	LIBRUBY_DLDFLAGS="${DLDFLAGS}"' -Wl,--out-implib=$(LIBRUBY)'
-	LIBRUBY_ALIASES=''
-	FIRSTMAKEFILE=GNUmakefile:cygwin/GNUmakefile.in
-	SOLIBS='$(LIBS)'
-	if test x"$enable_shared" = xno; then
-	    LIBRUBY_SO=dummy
-	    LIBRUBY='lib$(RUBY_SO_NAME).a'
-	    LIBRUBYARG='-l$(RUBY_SO_NAME)'
-	fi
-	MINIOBJS=dmydln.o
-	;;
-    hpux*)
-	case "$YACC" in
-	  *yacc*)
-	    XCFLAGS="$XCFLAGS -DYYMAXDEPTH=300"
-	    YACC="$YACC -Nl40000 -Nm40000"
-	    ;;
-	esac
-	MINIOBJS=dmydln.o
-	;;
-    aix*)
-	PREINSTALL='@$(RM) -r $(prefix)/lib/$(LIBRUBY_A) $(prefix)/lib/$(LIBRUBY_SO) $(prefix)/lib/ruby/$(MAJOR).$(MINOR)/$(arch)'
-	;;
-    *)
-	;;
-esac
-
-case "$build_os" in
-  *msdosdjgpp*) FIRSTMAKEFILE=GNUmakefile:djgpp/GNUmakefile.in;;
-esac
-
-AC_SUBST(XCFLAGS)dnl
-AC_SUBST(XLDFLAGS)dnl
-AC_SUBST(LIBRUBY_LDSHARED)
-AC_SUBST(LIBRUBY_DLDFLAGS)
-AC_SUBST(RUBY_INSTALL_NAME)
-AC_SUBST(rubyw_install_name)
-AC_SUBST(RUBYW_INSTALL_NAME)
-AC_SUBST(RUBY_SO_NAME)
-AC_SUBST(LIBRUBY_A)
-AC_SUBST(LIBRUBY_SO)
-AC_SUBST(LIBRUBY_ALIASES)
-AC_SUBST(LIBRUBY)
-AC_SUBST(LIBRUBYARG)
-AC_SUBST(LIBRUBYARG_STATIC)
-AC_SUBST(LIBRUBYARG_SHARED)
-AC_SUBST(SOLIBS)
-AC_SUBST(DLDLIBS)
-AC_SUBST(ENABLE_SHARED)
-AC_SUBST(MAINLIBS)
-AC_SUBST(COMMON_LIBS)
-AC_SUBST(COMMON_MACROS)
-AC_SUBST(COMMON_HEADERS)
-AC_SUBST(EXPORT_PREFIX)
-AC_SUBST(MINIOBJS)
-AC_SUBST(PREINSTALL)
-
-MAKEFILES="Makefile `echo $FIRSTMAKEFILE | sed 's/:.*//'`"
-MAKEFILES="`echo $MAKEFILES`"
-AC_SUBST(MAKEFILES)
-
-ri_prefix=
-test "$program_prefix" != NONE &&
-  ri_prefix=$program_prefix
-
-ri_suffix=
-test "$program_suffix" != NONE &&
-  ri_suffix=$program_suffix
-
-RUBY_INSTALL_NAME="${ri_prefix}ruby${ri_suffix}"
-case "$target_os" in
-  cygwin*|mingw*)
-    RUBYW_INSTALL_NAME="${ri_prefix}rubyw${ri_suffix}"
-    rubyw_install_name="$RUBYW_INSTALL_NAME"
-    ;;
-esac
-case "$target_os" in
-  cygwin*|mingw*|*djgpp*|os2-emx*)
-    RUBY_LIB_PREFIX="/lib/ruby"
-    ;;
-  *)
-    RUBY_LIB_PREFIX="${prefix}/lib/ruby"
-    ;;
-esac
-RUBY_LIB_PATH="${RUBY_LIB_PREFIX}/${MAJOR}.${MINOR}"
-
-AC_ARG_WITH(sitedir,
-	    [  --with-sitedir=DIR      site libraries in DIR [PREFIX/lib/ruby/site_ruby]],
-            [sitedir=$withval],
-            [sitedir='${prefix}/lib/ruby/site_ruby'])
-SITE_DIR=`eval echo \\"${sitedir}\\"`
-case "$target_os" in
-  cygwin*|mingw*|*djgpp*|os2-emx*)
-    RUBY_SITE_LIB_PATH="`expr "$SITE_DIR" : "$prefix\(/.*\)"`" ||
-    RUBY_SITE_LIB_PATH="$SITE_DIR";;
-  *)
-    RUBY_SITE_LIB_PATH="$SITE_DIR";;
-esac
-RUBY_SITE_LIB_PATH2="${RUBY_SITE_LIB_PATH}/${MAJOR}.${MINOR}"
-
-AC_DEFINE_UNQUOTED(RUBY_LIB, "${RUBY_LIB_PATH}")
-AC_DEFINE_UNQUOTED(RUBY_SITE_LIB, "${RUBY_SITE_LIB_PATH}")
-AC_DEFINE_UNQUOTED(RUBY_SITE_LIB2, "${RUBY_SITE_LIB_PATH2}")
-
-AC_SUBST(arch)dnl
-AC_SUBST(sitearch)dnl
-AC_SUBST(sitedir)dnl
-
-configure_args=$ac_configure_args
-AC_SUBST(configure_args)dnl
-
-if test "$fat_binary" != no ; then
-    arch="fat-${target_os}"
-
-    AC_DEFINE_UNQUOTED(RUBY_THIN_ARCHLIB,
-                 "${RUBY_LIB_PATH}/" __ARCHITECTURE__ "-${target_os}")
-
-    AC_DEFINE_UNQUOTED(RUBY_SITE_THIN_ARCHLIB,
-                 "${RUBY_SITE_LIB_PATH}/" __ARCHITECTURE__ "-${target_os}")
-    AC_DEFINE_UNQUOTED(RUBY_PLATFORM, __ARCHITECTURE__ "-${target_os}")
-else
-    arch="${target_cpu}-${target_os}"
-    AC_DEFINE_UNQUOTED(RUBY_PLATFORM, "${arch}")
-fi
-
-case "$target_os" in
-  mingw*) sitearch="i386-$rb_cv_msvcrt" ;;
-  *) sitearch="${arch}" ;;
-esac
-
-AC_DEFINE_UNQUOTED(RUBY_ARCHLIB, "${RUBY_LIB_PATH}/${arch}")
-AC_DEFINE_UNQUOTED(RUBY_SITE_ARCHLIB, "${RUBY_SITE_LIB_PATH2}/${sitearch}")
-
-AC_ARG_WITH(search-path,
-		[  --with-search-path=DIR specify the additional search path],
-		[search_path=$withval])
-if test "$search_path" != ""; then
-    AC_DEFINE_UNQUOTED(RUBY_SEARCH_PATH,"$search_path")
-fi
-
-AC_ARG_WITH(mantype,
-		[  --with-mantype=TYPE specify man page type; TYPE is one of man and doc],
-		[
-			case "$withval" in
-			man|doc)
-				MANTYPE=$withval
-				;;
-			*)
-				AC_MSG_ERROR(invalid man type: $withval)
-				;;
-			esac
-		])
-if test -z "$MANTYPE"; then
-	AC_PATH_PROGS(NROFF, nroff awf, /bin/false, "/usr/bin:/usr/ucb")
-	if ${NROFF} -mdoc ${srcdir}/ruby.1 >/dev/null 2>&1; then
-		MANTYPE=doc
-	else
-		MANTYPE=man
-	fi
-fi
-AC_SUBST(MANTYPE)
-
-if test -f config.h && tr -d '\015' < confdefs.h | cmp -s config.h -; then
-  echo "config.h unchanged"
-else
-  echo "creating config.h"
-  tr -d '\015' < confdefs.h > config.h
-fi
-: > confdefs.h
-
-AC_CONFIG_FILES($FIRSTMAKEFILE)
-AC_CONFIG_FILES(Makefile, [{
-	echo; test x"$EXEEXT" = x || echo 'miniruby: miniruby$(EXEEXT)'
-	test "$RUBY_INSTALL_NAME$EXEEXT" = ruby || echo 'ruby: $(PROGRAM);'
-	sed ['s/{\$([^(){}]*)[^{}]*}//g'] ${srcdir}/common.mk
-} >> Makefile], [RUBY_INSTALL_NAME=$RUBY_INSTALL_NAME EXEEXT=$EXEEXT])
-AC_OUTPUT
diff --git a/constant.h b/constant.h
new file mode 100644
index 0000000000..fcccf07384
--- /dev/null
+++ b/constant.h
@@ -0,0 +1,51 @@
+/**********************************************************************
+
+  constant.h -
+
+  $Author$
+  created at: Sun Nov 15 00:09:33 2009
+
+  Copyright (C) 2009 Yusuke Endoh
+
+**********************************************************************/
+#ifndef CONSTANT_H
+#define CONSTANT_H
+
+typedef enum {
+    CONST_DEPRECATED = 0x100,
+
+    CONST_VISIBILITY_MASK = 0xff,
+    CONST_PUBLIC    = 0x00,
+    CONST_PRIVATE,
+    CONST_VISIBILITY_MAX
+} rb_const_flag_t;
+
+#define RB_CONST_PRIVATE_P(ce) \
+    (((ce)->flag & CONST_VISIBILITY_MASK) == CONST_PRIVATE)
+#define RB_CONST_PUBLIC_P(ce) \
+    (((ce)->flag & CONST_VISIBILITY_MASK) == CONST_PUBLIC)
+
+#define RB_CONST_DEPRECATED_P(ce) \
+    ((ce)->flag & CONST_DEPRECATED)
+
+typedef struct rb_const_entry_struct {
+    rb_const_flag_t flag;
+    int line;
+    const VALUE value;            /* should be mark */
+    const VALUE file;             /* should be mark */
+} rb_const_entry_t;
+
+VALUE rb_mod_private_constant(int argc, const VALUE *argv, VALUE obj);
+VALUE rb_mod_public_constant(int argc, const VALUE *argv, VALUE obj);
+VALUE rb_mod_deprecate_constant(int argc, const VALUE *argv, VALUE obj);
+void rb_free_const_table(struct rb_id_table *tbl);
+VALUE rb_public_const_get(VALUE klass, ID id);
+VALUE rb_public_const_get_at(VALUE klass, ID id);
+VALUE rb_public_const_get_from(VALUE klass, ID id);
+int rb_public_const_defined(VALUE klass, ID id);
+int rb_public_const_defined_at(VALUE klass, ID id);
+int rb_public_const_defined_from(VALUE klass, ID id);
+rb_const_entry_t *rb_const_lookup(VALUE klass, ID id);
+int rb_autoloading_value(VALUE mod, ID id, VALUE *value, rb_const_flag_t *flag);
+
+#endif /* CONSTANT_H */
diff --git a/cont.c b/cont.c
new file mode 100644
index 0000000000..da469b6cd5
--- /dev/null
+++ b/cont.c
@@ -0,0 +1,2140 @@
+/**********************************************************************
+
+  cont.c -
+
+  $Author$
+  created at: Thu May 23 09:03:43 2007
+
+  Copyright (C) 2007 Koichi Sasada
+
+**********************************************************************/
+
+#include "internal.h"
+#include "vm_core.h"
+#include "gc.h"
+#include "eval_intern.h"
+#include "mjit.h"
+
+/* FIBER_USE_NATIVE enables Fiber performance improvement using system
+ * dependent method such as make/setcontext on POSIX system or
+ * CreateFiber() API on Windows.
+ * This hack make Fiber context switch faster (x2 or more).
+ * However, it decrease maximum number of Fiber.  For example, on the
+ * 32bit POSIX OS, ten or twenty thousands Fiber can be created.
+ *
+ * Details is reported in the paper "A Fast Fiber Implementation for Ruby 1.9"
+ * in Proc. of 51th Programming Symposium, pp.21--28 (2010) (in Japanese).
+ */
+
+/*
+  Enable FIBER_USE_COROUTINE to make fiber yield/resume much faster by using native assembly implementations.
+
+  rvm install ruby-head-ioquatix-native-fiber --url https://github.com/ioquatix/ruby --branch native-fiber
+
+  # Without libcoro
+  koyoko% ./build/bin/ruby ./fiber_benchmark.rb 10000 1000
+  setup time for 10000 fibers:   0.099961
+  execution time for 1000 messages:  19.505909
+
+  # With libcoro
+  koyoko% ./build/bin/ruby ./fiber_benchmark.rb 10000 1000
+  setup time for 10000 fibers:   0.099268
+  execution time for 1000 messages:   8.491746
+*/
+
+#ifdef FIBER_USE_COROUTINE
+#include FIBER_USE_COROUTINE
+#define FIBER_USE_NATIVE 1
+#endif
+
+#if !defined(FIBER_USE_NATIVE)
+# if defined(HAVE_GETCONTEXT) && defined(HAVE_SETCONTEXT)
+#   if 0
+#   elif defined(__NetBSD__)
+/* On our experience, NetBSD doesn't support using setcontext() and pthread
+ * simultaneously.  This is because pthread_self(), TLS and other information
+ * are represented by stack pointer (higher bits of stack pointer).
+ * TODO: check such constraint on configure.
+ */
+#     define FIBER_USE_NATIVE 0
+#   elif defined(__sun)
+/* On Solaris because resuming any Fiber caused SEGV, for some reason.
+ */
+#     define FIBER_USE_NATIVE 0
+#   elif defined(__ia64)
+/* At least, Linux/ia64's getcontext(3) doesn't save register window.
+ */
+#     define FIBER_USE_NATIVE 0
+#   elif defined(__GNU__)
+/* GNU/Hurd doesn't fully support getcontext, setcontext, makecontext
+ * and swapcontext functions. Disabling their usage till support is
+ * implemented. More info at
+ * http://darnassus.sceen.net/~hurd-web/open_issues/glibc/#getcontext
+ */
+#     define FIBER_USE_NATIVE 0
+#   else
+#     define FIBER_USE_NATIVE 1
+#   endif
+# elif defined(_WIN32)
+#  define FIBER_USE_NATIVE 1
+# endif
+#endif
+#if !defined(FIBER_USE_NATIVE)
+#define FIBER_USE_NATIVE 0
+#endif
+
+#if FIBER_USE_NATIVE
+#ifndef _WIN32
+#include <unistd.h>
+#include <sys/mman.h>
+#include <ucontext.h>
+#endif
+#define RB_PAGE_SIZE (pagesize)
+#define RB_PAGE_MASK (~(RB_PAGE_SIZE - 1))
+static long pagesize;
+#endif /*FIBER_USE_NATIVE*/
+
+#define CAPTURE_JUST_VALID_VM_STACK 1
+
+enum context_type {
+    CONTINUATION_CONTEXT = 0,
+    FIBER_CONTEXT = 1
+};
+
+struct cont_saved_vm_stack {
+    VALUE *ptr;
+#ifdef CAPTURE_JUST_VALID_VM_STACK
+    size_t slen;  /* length of stack (head of ec->vm_stack) */
+    size_t clen;  /* length of control frames (tail of ec->vm_stack) */
+#endif
+};
+
+typedef struct rb_context_struct {
+    enum context_type type;
+    int argc;
+    VALUE self;
+    VALUE value;
+
+    struct cont_saved_vm_stack saved_vm_stack;
+
+    struct {
+	VALUE *stack;
+	VALUE *stack_src;
+	size_t stack_size;
+#ifdef __ia64
+	VALUE *register_stack;
+	VALUE *register_stack_src;
+	int register_stack_size;
+#endif
+    } machine;
+    rb_execution_context_t saved_ec;
+    rb_jmpbuf_t jmpbuf;
+    rb_ensure_entry_t *ensure_array;
+    /* Pointer to MJIT info about the continuation.  */
+    struct mjit_cont *mjit_cont;
+} rb_context_t;
+
+
+/*
+ * Fiber status:
+ *    [Fiber.new] ------> FIBER_CREATED
+ *                        | [Fiber#resume]
+ *                        v
+ *                   +--> FIBER_RESUMED ----+
+ *    [Fiber#resume] |    | [Fiber.yield]   |
+ *                   |    v                 |
+ *                   +-- FIBER_SUSPENDED    | [Terminate]
+ *                                          |
+ *                       FIBER_TERMINATED <-+
+ */
+enum fiber_status {
+    FIBER_CREATED,
+    FIBER_RESUMED,
+    FIBER_SUSPENDED,
+    FIBER_TERMINATED
+};
+
+#define FIBER_CREATED_P(fib)    ((fib)->status == FIBER_CREATED)
+#define FIBER_RESUMED_P(fib)    ((fib)->status == FIBER_RESUMED)
+#define FIBER_SUSPENDED_P(fib)  ((fib)->status == FIBER_SUSPENDED)
+#define FIBER_TERMINATED_P(fib) ((fib)->status == FIBER_TERMINATED)
+#define FIBER_RUNNABLE_P(fib)   (FIBER_CREATED_P(fib) || FIBER_SUSPENDED_P(fib))
+
+#if FIBER_USE_NATIVE && !defined(FIBER_USE_COROUTINE) && !defined(_WIN32)
+static inline int
+fiber_context_create(ucontext_t *context, void (*func)(), void *arg, void *ptr, size_t size)
+{
+    if (getcontext(context) < 0) return -1;
+    /*
+     * getcontext() may fail by some reasons:
+     *   1. SELinux policy banned one of "rt_sigprocmask",
+     *      "sigprocmask" or "swapcontext";
+     *   2. libseccomp (aka. syscall filter) banned one of them.
+     */
+    context->uc_link = NULL;
+    context->uc_stack.ss_sp = ptr;
+    context->uc_stack.ss_size = size;
+    makecontext(context, func, 0);
+    return 0;
+}
+#endif
+
+struct rb_fiber_struct {
+    rb_context_t cont;
+    VALUE first_proc;
+    struct rb_fiber_struct *prev;
+    BITFIELD(enum fiber_status, status, 2);
+    /* If a fiber invokes "transfer",
+     * then this fiber can't "resume" any more after that.
+     * You shouldn't mix "transfer" and "resume".
+     */
+    unsigned int transferred : 1;
+
+#if FIBER_USE_NATIVE
+#if defined(FIBER_USE_COROUTINE)
+#define FIBER_ALLOCATE_STACK
+    coroutine_context context;
+    void *ss_sp;
+    size_t ss_size;
+#elif defined(_WIN32)
+    void *fib_handle;
+#else
+#define FIBER_ALLOCATE_STACK
+    ucontext_t context;
+    /* Because context.uc_stack.ss_sp and context.uc_stack.ss_size
+     * are not necessarily valid after makecontext() or swapcontext(),
+     * they are saved in these variables for later use.
+     */
+    void *ss_sp;
+    size_t ss_size;
+#endif
+#endif
+};
+
+#ifdef FIBER_ALLOCATE_STACK
+#define MAX_MACHINE_STACK_CACHE  10
+static int machine_stack_cache_index = 0;
+typedef struct machine_stack_cache_struct {
+    void *ptr;
+    size_t size;
+} machine_stack_cache_t;
+static machine_stack_cache_t machine_stack_cache[MAX_MACHINE_STACK_CACHE];
+static machine_stack_cache_t terminated_machine_stack;
+#endif
+
+static const char *
+fiber_status_name(enum fiber_status s)
+{
+    switch (s) {
+      case FIBER_CREATED: return "created";
+      case FIBER_RESUMED: return "resumed";
+      case FIBER_SUSPENDED: return "suspended";
+      case FIBER_TERMINATED: return "terminated";
+    }
+    VM_UNREACHABLE(fiber_status_name);
+    return NULL;
+}
+
+static void
+fiber_verify(const rb_fiber_t *fib)
+{
+#if VM_CHECK_MODE > 0
+    VM_ASSERT(fib->cont.saved_ec.fiber_ptr == fib);
+
+    switch (fib->status) {
+      case FIBER_RESUMED:
+	VM_ASSERT(fib->cont.saved_ec.vm_stack != NULL);
+	break;
+      case FIBER_SUSPENDED:
+	VM_ASSERT(fib->cont.saved_ec.vm_stack != NULL);
+	break;
+      case FIBER_CREATED:
+      case FIBER_TERMINATED:
+	/* TODO */
+	break;
+      default:
+	VM_UNREACHABLE(fiber_verify);
+    }
+#endif
+}
+
+#if VM_CHECK_MODE > 0
+void
+rb_ec_verify(const rb_execution_context_t *ec)
+{
+    /* TODO */
+}
+#endif
+
+static void
+fiber_status_set(rb_fiber_t *fib, enum fiber_status s)
+{
+    if (0) fprintf(stderr, "fib: %p, status: %s -> %s\n", (void *)fib, fiber_status_name(fib->status), fiber_status_name(s));
+    VM_ASSERT(!FIBER_TERMINATED_P(fib));
+    VM_ASSERT(fib->status != s);
+    fiber_verify(fib);
+    fib->status = s;
+}
+
+void
+rb_ec_set_vm_stack(rb_execution_context_t *ec, VALUE *stack, size_t size)
+{
+    ec->vm_stack = stack;
+    ec->vm_stack_size = size;
+}
+
+static inline void
+ec_switch(rb_thread_t *th, rb_fiber_t *fib)
+{
+    rb_execution_context_t *ec = &fib->cont.saved_ec;
+
+    ruby_current_execution_context_ptr = th->ec = ec;
+
+    /*
+     * timer-thread may set trap interrupt on previous th->ec at any time;
+     * ensure we do not delay (or lose) the trap interrupt handling.
+     */
+    if (th->vm->main_thread == th && rb_signal_buff_size() > 0) {
+        RUBY_VM_SET_TRAP_INTERRUPT(ec);
+    }
+
+    VM_ASSERT(ec->fiber_ptr->cont.self == 0 || ec->vm_stack != NULL);
+}
+
+static const rb_data_type_t cont_data_type, fiber_data_type;
+static VALUE rb_cContinuation;
+static VALUE rb_cFiber;
+static VALUE rb_eFiberError;
+
+static rb_context_t *
+cont_ptr(VALUE obj)
+{
+    rb_context_t *cont;
+
+    TypedData_Get_Struct(obj, rb_context_t, &cont_data_type, cont);
+
+    return cont;
+}
+
+static rb_fiber_t *
+fiber_ptr(VALUE obj)
+{
+    rb_fiber_t *fib;
+
+    TypedData_Get_Struct(obj, rb_fiber_t, &fiber_data_type, fib);
+    if (!fib) rb_raise(rb_eFiberError, "uninitialized fiber");
+
+    return fib;
+}
+
+NOINLINE(static VALUE cont_capture(volatile int *volatile stat));
+
+#define THREAD_MUST_BE_RUNNING(th) do { \
+	if (!(th)->ec->tag) rb_raise(rb_eThreadError, "not running thread");	\
+    } while (0)
+
+static VALUE
+cont_thread_value(const rb_context_t *cont)
+{
+    return cont->saved_ec.thread_ptr->self;
+}
+
+static void
+cont_mark(void *ptr)
+{
+    rb_context_t *cont = ptr;
+
+    RUBY_MARK_ENTER("cont");
+    rb_gc_mark(cont->value);
+
+    rb_execution_context_mark(&cont->saved_ec);
+    rb_gc_mark(cont_thread_value(cont));
+
+    if (cont->saved_vm_stack.ptr) {
+#ifdef CAPTURE_JUST_VALID_VM_STACK
+	rb_gc_mark_locations(cont->saved_vm_stack.ptr,
+			     cont->saved_vm_stack.ptr + cont->saved_vm_stack.slen + cont->saved_vm_stack.clen);
+#else
+	rb_gc_mark_locations(cont->saved_vm_stack.ptr,
+			     cont->saved_vm_stack.ptr, cont->saved_ec.stack_size);
+#endif
+    }
+
+    if (cont->machine.stack) {
+	if (cont->type == CONTINUATION_CONTEXT) {
+	    /* cont */
+	    rb_gc_mark_locations(cont->machine.stack,
+				 cont->machine.stack + cont->machine.stack_size);
+	}
+	else {
+	    /* fiber */
+	    const rb_fiber_t *fib = (rb_fiber_t*)cont;
+
+	    if (!FIBER_TERMINATED_P(fib)) {
+		rb_gc_mark_locations(cont->machine.stack,
+				     cont->machine.stack + cont->machine.stack_size);
+	    }
+	}
+    }
+#ifdef __ia64
+    if (cont->machine.register_stack) {
+	rb_gc_mark_locations(cont->machine.register_stack,
+			     cont->machine.register_stack + cont->machine.register_stack_size);
+    }
+#endif
+
+    RUBY_MARK_LEAVE("cont");
+}
+
+static int
+fiber_is_root_p(const rb_fiber_t *fib)
+{
+    return fib == fib->cont.saved_ec.thread_ptr->root_fiber;
+}
+
+static void
+cont_free(void *ptr)
+{
+    rb_context_t *cont = ptr;
+
+    RUBY_FREE_ENTER("cont");
+    ruby_xfree(cont->saved_ec.vm_stack);
+
+#if FIBER_USE_NATIVE
+    if (cont->type == CONTINUATION_CONTEXT) {
+	/* cont */
+	ruby_xfree(cont->ensure_array);
+	RUBY_FREE_UNLESS_NULL(cont->machine.stack);
+    }
+    else {
+	/* fiber */
+	rb_fiber_t *fib = (rb_fiber_t*)cont;
+#if defined(FIBER_USE_COROUTINE)
+	coroutine_destroy(&fib->context);
+	if (fib->ss_sp != NULL) {
+	    if (fiber_is_root_p(fib)) {
+		rb_bug("Illegal root fiber parameter");
+	    }
+#ifdef _WIN32
+            VirtualFree((void*)fib->ss_sp, 0, MEM_RELEASE);
+#else
+	    munmap((void*)fib->ss_sp, fib->ss_size);
+#endif
+            fib->ss_sp = NULL;
+	}
+#elif defined(_WIN32)
+	if (!fiber_is_root_p(fib)) {
+	    /* don't delete root fiber handle */
+	    if (fib->fib_handle) {
+		DeleteFiber(fib->fib_handle);
+	    }
+	}
+#else /* not WIN32 */
+        /* fib->ss_sp == NULL is possible for root fiber */
+	if (fib->ss_sp != NULL) {
+	    munmap((void*)fib->ss_sp, fib->ss_size);
+	}
+#endif
+    }
+#else /* not FIBER_USE_NATIVE */
+    ruby_xfree(cont->ensure_array);
+    RUBY_FREE_UNLESS_NULL(cont->machine.stack);
+#endif
+#ifdef __ia64
+    RUBY_FREE_UNLESS_NULL(cont->machine.register_stack);
+#endif
+    RUBY_FREE_UNLESS_NULL(cont->saved_vm_stack.ptr);
+
+    if (mjit_enabled && cont->mjit_cont != NULL) {
+        mjit_cont_free(cont->mjit_cont);
+    }
+    /* free rb_cont_t or rb_fiber_t */
+    ruby_xfree(ptr);
+    RUBY_FREE_LEAVE("cont");
+}
+
+static size_t
+cont_memsize(const void *ptr)
+{
+    const rb_context_t *cont = ptr;
+    size_t size = 0;
+
+    size = sizeof(*cont);
+    if (cont->saved_vm_stack.ptr) {
+#ifdef CAPTURE_JUST_VALID_VM_STACK
+	size_t n = (cont->saved_vm_stack.slen + cont->saved_vm_stack.clen);
+#else
+	size_t n = cont->saved_ec.vm_stack_size;
+#endif
+	size += n * sizeof(*cont->saved_vm_stack.ptr);
+    }
+
+    if (cont->machine.stack) {
+	size += cont->machine.stack_size * sizeof(*cont->machine.stack);
+    }
+#ifdef __ia64
+    if (cont->machine.register_stack) {
+	size += cont->machine.register_stack_size * sizeof(*cont->machine.register_stack);
+    }
+#endif
+    return size;
+}
+
+void
+rb_fiber_mark_self(const rb_fiber_t *fib)
+{
+    if (fib->cont.self) {
+	rb_gc_mark(fib->cont.self);
+    }
+    else {
+	rb_execution_context_mark(&fib->cont.saved_ec);
+    }
+}
+
+static void
+fiber_mark(void *ptr)
+{
+    rb_fiber_t *fib = ptr;
+    RUBY_MARK_ENTER("cont");
+    fiber_verify(fib);
+    rb_gc_mark(fib->first_proc);
+    if (fib->prev) rb_fiber_mark_self(fib->prev);
+
+#if !FIBER_USE_NATIVE
+    if (fib->status == FIBER_TERMINATED) {
+	/* FIBER_TERMINATED fiber should not mark machine stack */
+	if (fib->cont.saved_ec.machine.stack_end != NULL) {
+	    fib->cont.saved_ec.machine.stack_end = NULL;
+	}
+    }
+#endif
+
+    cont_mark(&fib->cont);
+    RUBY_MARK_LEAVE("cont");
+}
+
+static void
+fiber_free(void *ptr)
+{
+    rb_fiber_t *fib = ptr;
+    RUBY_FREE_ENTER("fiber");
+
+    if (fib->cont.saved_ec.local_storage) {
+	st_free_table(fib->cont.saved_ec.local_storage);
+    }
+
+    cont_free(&fib->cont);
+    RUBY_FREE_LEAVE("fiber");
+}
+
+static size_t
+fiber_memsize(const void *ptr)
+{
+    const rb_fiber_t *fib = ptr;
+    size_t size = sizeof(*fib);
+    const rb_execution_context_t *saved_ec = &fib->cont.saved_ec;
+    const rb_thread_t *th = rb_ec_thread_ptr(saved_ec);
+
+    /*
+     * vm.c::thread_memsize already counts th->ec->local_storage
+     */
+    if (saved_ec->local_storage && fib != th->root_fiber) {
+	size += st_memsize(saved_ec->local_storage);
+    }
+    size += cont_memsize(&fib->cont);
+    return size;
+}
+
+VALUE
+rb_obj_is_fiber(VALUE obj)
+{
+    if (rb_typeddata_is_kind_of(obj, &fiber_data_type)) {
+	return Qtrue;
+    }
+    else {
+	return Qfalse;
+    }
+}
+
+static void
+cont_save_machine_stack(rb_thread_t *th, rb_context_t *cont)
+{
+    size_t size;
+
+    SET_MACHINE_STACK_END(&th->ec->machine.stack_end);
+#ifdef __ia64
+    th->ec->machine.register_stack_end = rb_ia64_bsp();
+#endif
+
+    if (th->ec->machine.stack_start > th->ec->machine.stack_end) {
+	size = cont->machine.stack_size = th->ec->machine.stack_start - th->ec->machine.stack_end;
+	cont->machine.stack_src = th->ec->machine.stack_end;
+    }
+    else {
+	size = cont->machine.stack_size = th->ec->machine.stack_end - th->ec->machine.stack_start;
+	cont->machine.stack_src = th->ec->machine.stack_start;
+    }
+
+    if (cont->machine.stack) {
+	REALLOC_N(cont->machine.stack, VALUE, size);
+    }
+    else {
+	cont->machine.stack = ALLOC_N(VALUE, size);
+    }
+
+    FLUSH_REGISTER_WINDOWS;
+    MEMCPY(cont->machine.stack, cont->machine.stack_src, VALUE, size);
+
+#ifdef __ia64
+    rb_ia64_flushrs();
+    size = cont->machine.register_stack_size = th->ec->machine.register_stack_end - th->ec->machine.register_stack_start;
+    cont->machine.register_stack_src = th->ec->machine.register_stack_start;
+    if (cont->machine.register_stack) {
+	REALLOC_N(cont->machine.register_stack, VALUE, size);
+    }
+    else {
+	cont->machine.register_stack = ALLOC_N(VALUE, size);
+    }
+
+    MEMCPY(cont->machine.register_stack, cont->machine.register_stack_src, VALUE, size);
+#endif
+}
+
+static const rb_data_type_t cont_data_type = {
+    "continuation",
+    {cont_mark, cont_free, cont_memsize,},
+    0, 0, RUBY_TYPED_FREE_IMMEDIATELY
+};
+
+static inline void
+cont_save_thread(rb_context_t *cont, rb_thread_t *th)
+{
+    rb_execution_context_t *sec = &cont->saved_ec;
+
+    VM_ASSERT(th->status == THREAD_RUNNABLE);
+
+    /* save thread context */
+    *sec = *th->ec;
+
+    /* saved_ec->machine.stack_end should be NULL */
+    /* because it may happen GC afterward */
+    sec->machine.stack_end = NULL;
+
+#ifdef __ia64
+    sec->machine.register_stack_start = NULL;
+    sec->machine.register_stack_end = NULL;
+#endif
+}
+
+static void
+cont_init(rb_context_t *cont, rb_thread_t *th)
+{
+    /* save thread context */
+    cont_save_thread(cont, th);
+    cont->saved_ec.thread_ptr = th;
+    cont->saved_ec.local_storage = NULL;
+    cont->saved_ec.local_storage_recursive_hash = Qnil;
+    cont->saved_ec.local_storage_recursive_hash_for_trace = Qnil;
+    if (mjit_enabled) {
+        cont->mjit_cont = mjit_cont_new(&cont->saved_ec);
+    }
+}
+
+static rb_context_t *
+cont_new(VALUE klass)
+{
+    rb_context_t *cont;
+    volatile VALUE contval;
+    rb_thread_t *th = GET_THREAD();
+
+    THREAD_MUST_BE_RUNNING(th);
+    contval = TypedData_Make_Struct(klass, rb_context_t, &cont_data_type, cont);
+    cont->self = contval;
+    cont_init(cont, th);
+    return cont;
+}
+
+#if 0
+void
+show_vm_stack(const rb_execution_context_t *ec)
+{
+    VALUE *p = ec->vm_stack;
+    while (p < ec->cfp->sp) {
+	fprintf(stderr, "%3d ", (int)(p - ec->vm_stack));
+	rb_obj_info_dump(*p);
+	p++;
+    }
+}
+
+void
+show_vm_pcs(const rb_control_frame_t *cfp,
+	    const rb_control_frame_t *end_of_cfp)
+{
+    int i=0;
+    while (cfp != end_of_cfp) {
+	int pc = 0;
+	if (cfp->iseq) {
+	    pc = cfp->pc - cfp->iseq->body->iseq_encoded;
+	}
+	fprintf(stderr, "%2d pc: %d\n", i++, pc);
+	cfp = RUBY_VM_PREVIOUS_CONTROL_FRAME(cfp);
+    }
+}
+#endif
+COMPILER_WARNING_PUSH
+#ifdef __clang__
+COMPILER_WARNING_IGNORED(-Wduplicate-decl-specifier)
+#endif
+static VALUE
+cont_capture(volatile int *volatile stat)
+{
+    rb_context_t *volatile cont;
+    rb_thread_t *th = GET_THREAD();
+    volatile VALUE contval;
+    const rb_execution_context_t *ec = th->ec;
+
+    THREAD_MUST_BE_RUNNING(th);
+    rb_vm_stack_to_heap(th->ec);
+    cont = cont_new(rb_cContinuation);
+    contval = cont->self;
+
+#ifdef CAPTURE_JUST_VALID_VM_STACK
+    cont->saved_vm_stack.slen = ec->cfp->sp - ec->vm_stack;
+    cont->saved_vm_stack.clen = ec->vm_stack + ec->vm_stack_size - (VALUE*)ec->cfp;
+    cont->saved_vm_stack.ptr = ALLOC_N(VALUE, cont->saved_vm_stack.slen + cont->saved_vm_stack.clen);
+    MEMCPY(cont->saved_vm_stack.ptr,
+	   ec->vm_stack,
+	   VALUE, cont->saved_vm_stack.slen);
+    MEMCPY(cont->saved_vm_stack.ptr + cont->saved_vm_stack.slen,
+	   (VALUE*)ec->cfp,
+	   VALUE,
+	   cont->saved_vm_stack.clen);
+#else
+    cont->saved_vm_stack.ptr = ALLOC_N(VALUE, ec->vm_stack_size);
+    MEMCPY(cont->saved_vm_stack.ptr, ec->vm_stack, VALUE, ec->vm_stack_size);
+#endif
+    rb_ec_set_vm_stack(&cont->saved_ec, NULL, 0);
+    cont_save_machine_stack(th, cont);
+
+    /* backup ensure_list to array for search in another context */
+    {
+	rb_ensure_list_t *p;
+	int size = 0;
+	rb_ensure_entry_t *entry;
+	for (p=th->ec->ensure_list; p; p=p->next)
+	    size++;
+	entry = cont->ensure_array = ALLOC_N(rb_ensure_entry_t,size+1);
+	for (p=th->ec->ensure_list; p; p=p->next) {
+	    if (!p->entry.marker)
+		p->entry.marker = rb_ary_tmp_new(0); /* dummy object */
+	    *entry++ = p->entry;
+	}
+	entry->marker = 0;
+    }
+
+    if (ruby_setjmp(cont->jmpbuf)) {
+	VALUE value;
+
+	VAR_INITIALIZED(cont);
+	value = cont->value;
+	if (cont->argc == -1) rb_exc_raise(value);
+	cont->value = Qnil;
+	*stat = 1;
+	return value;
+    }
+    else {
+	*stat = 0;
+	return contval;
+    }
+}
+COMPILER_WARNING_POP
+
+static inline void
+fiber_restore_thread(rb_thread_t *th, rb_fiber_t *fib)
+{
+    ec_switch(th, fib);
+    VM_ASSERT(th->ec->fiber_ptr == fib);
+}
+
+static inline void
+cont_restore_thread(rb_context_t *cont)
+{
+    rb_thread_t *th = GET_THREAD();
+
+    /* restore thread context */
+    if (cont->type == CONTINUATION_CONTEXT) {
+	/* continuation */
+	rb_execution_context_t *sec = &cont->saved_ec;
+	rb_fiber_t *fib = NULL;
+
+	if (sec->fiber_ptr != NULL) {
+	    fib = sec->fiber_ptr;
+	}
+	else if (th->root_fiber) {
+	    fib = th->root_fiber;
+	}
+
+	if (fib && th->ec != &fib->cont.saved_ec) {
+	    ec_switch(th, fib);
+	}
+
+        if (th->ec->trace_arg != sec->trace_arg) {
+            rb_raise(rb_eRuntimeError, "can't call across trace_func");
+        }
+
+	/* copy vm stack */
+#ifdef CAPTURE_JUST_VALID_VM_STACK
+	MEMCPY(th->ec->vm_stack,
+	       cont->saved_vm_stack.ptr,
+	       VALUE, cont->saved_vm_stack.slen);
+	MEMCPY(th->ec->vm_stack + th->ec->vm_stack_size - cont->saved_vm_stack.clen,
+	       cont->saved_vm_stack.ptr + cont->saved_vm_stack.slen,
+	       VALUE, cont->saved_vm_stack.clen);
+#else
+	MEMCPY(th->ec->vm_stack, cont->saved_vm_stack.ptr, VALUE, sec->vm_stack_size);
+#endif
+	/* other members of ec */
+
+	th->ec->cfp = sec->cfp;
+	th->ec->raised_flag = sec->raised_flag;
+	th->ec->tag = sec->tag;
+	th->ec->protect_tag = sec->protect_tag;
+	th->ec->root_lep = sec->root_lep;
+	th->ec->root_svar = sec->root_svar;
+	th->ec->ensure_list = sec->ensure_list;
+	th->ec->errinfo = sec->errinfo;
+
+	VM_ASSERT(th->ec->vm_stack != NULL);
+    }
+    else {
+	/* fiber */
+	fiber_restore_thread(th, (rb_fiber_t*)cont);
+    }
+}
+
+#if FIBER_USE_NATIVE
+#if defined(FIBER_USE_COROUTINE)
+static COROUTINE
+fiber_entry(coroutine_context * from, coroutine_context * to)
+{
+    rb_fiber_start();
+}
+#elif defined(_WIN32)
+static void
+fiber_set_stack_location(void)
+{
+    rb_thread_t *th = GET_THREAD();
+    VALUE *ptr;
+
+    SET_MACHINE_STACK_END(&ptr);
+    th->ec->machine.stack_start = (void*)(((VALUE)ptr & RB_PAGE_MASK) + STACK_UPPER((void *)&ptr, 0, RB_PAGE_SIZE));
+}
+
+NORETURN(static VOID CALLBACK fiber_entry(void *arg));
+static VOID CALLBACK
+fiber_entry(void *arg)
+{
+    fiber_set_stack_location();
+    rb_fiber_start();
+}
+#else
+NORETURN(static void fiber_entry(void *arg));
+static void
+fiber_entry(void *arg)
+{
+    rb_fiber_start();
+}
+#endif
+#endif
+
+#ifdef FIBER_ALLOCATE_STACK
+/*
+ * FreeBSD require a first (i.e. addr) argument of mmap(2) is not NULL
+ * if MAP_STACK is passed.
+ * http://www.FreeBSD.org/cgi/query-pr.cgi?pr=158755
+ */
+#if defined(MAP_STACK) && !defined(__FreeBSD__) && !defined(__FreeBSD_kernel__)
+#define FIBER_STACK_FLAGS (MAP_PRIVATE | MAP_ANON | MAP_STACK)
+#else
+#define FIBER_STACK_FLAGS (MAP_PRIVATE | MAP_ANON)
+#endif
+
+#define ERRNOMSG strerror(errno)
+
+static char*
+fiber_machine_stack_alloc(size_t size)
+{
+    char *ptr;
+#ifdef _WIN32
+    DWORD old_protect;
+#endif
+
+    if (machine_stack_cache_index > 0) {
+        if (machine_stack_cache[machine_stack_cache_index - 1].size == (size / sizeof(VALUE))) {
+            ptr = machine_stack_cache[machine_stack_cache_index - 1].ptr;
+            machine_stack_cache_index--;
+            machine_stack_cache[machine_stack_cache_index].ptr = NULL;
+            machine_stack_cache[machine_stack_cache_index].size = 0;
+        } else {
+            /* TODO handle multiple machine stack size */
+            rb_bug("machine_stack_cache size is not canonicalized");
+        }
+    } else {
+#ifdef _WIN32
+        ptr = VirtualAlloc(0, size, MEM_COMMIT, PAGE_READWRITE);
+
+        if (!ptr) {
+            rb_raise(rb_eFiberError, "can't allocate machine stack to fiber: %s", ERRNOMSG);
+        }
+
+        if (!VirtualProtect(ptr, RB_PAGE_SIZE, PAGE_READWRITE | PAGE_GUARD, &old_protect)) {
+            rb_raise(rb_eFiberError, "can't set a guard page: %s", ERRNOMSG);
+        }
+#else
+        void *page;
+        STACK_GROW_DIR_DETECTION;
+
+        errno = 0;
+        ptr = mmap(NULL, size, PROT_READ | PROT_WRITE, FIBER_STACK_FLAGS, -1, 0);
+        if (ptr == MAP_FAILED) {
+            rb_raise(rb_eFiberError, "can't alloc machine stack to fiber: %s", ERRNOMSG);
+        }
+
+        /* guard page setup */
+        page = ptr + STACK_DIR_UPPER(size - RB_PAGE_SIZE, 0);
+        if (mprotect(page, RB_PAGE_SIZE, PROT_NONE) < 0) {
+            rb_raise(rb_eFiberError, "can't set a guard page: %s", ERRNOMSG);
+        }
+#endif
+    }
+
+    return ptr;
+}
+#endif
+
+#if FIBER_USE_NATIVE
+static void
+fiber_initialize_machine_stack_context(rb_fiber_t *fib, size_t size)
+{
+    rb_execution_context_t *sec = &fib->cont.saved_ec;
+
+#if defined(FIBER_USE_COROUTINE)
+    char *ptr;
+    STACK_GROW_DIR_DETECTION;
+
+    ptr = fiber_machine_stack_alloc(size);
+    fib->ss_sp = ptr;
+    fib->ss_size = size;
+    coroutine_initialize(&fib->context, fiber_entry, ptr+size, size);
+    sec->machine.stack_start = (VALUE*)(ptr + STACK_DIR_UPPER(0, size));
+    sec->machine.stack_maxsize = size - RB_PAGE_SIZE;
+#elif defined(_WIN32)
+# if defined(_MSC_VER) && _MSC_VER <= 1200
+#   define CreateFiberEx(cs, stacksize, flags, entry, param) \
+    CreateFiber((stacksize), (entry), (param))
+# endif
+    fib->fib_handle = CreateFiberEx(size - 1, size, 0, fiber_entry, NULL);
+    if (!fib->fib_handle) {
+	/* try to release unnecessary fibers & retry to create */
+	rb_gc();
+	fib->fib_handle = CreateFiberEx(size - 1, size, 0, fiber_entry, NULL);
+	if (!fib->fib_handle) {
+	    rb_raise(rb_eFiberError, "can't create fiber");
+	}
+    }
+    sec->machine.stack_maxsize = size;
+#else /* not WIN32 */
+    char *ptr;
+    STACK_GROW_DIR_DETECTION;
+
+    ptr = fiber_machine_stack_alloc(size);
+    fib->ss_sp = ptr;
+    fib->ss_size = size;
+    if (fiber_context_create(&fib->context, fiber_entry, NULL, fib->ss_sp, fib->ss_size)) {
+	rb_raise(rb_eFiberError, "can't get context for creating fiber: %s", ERRNOMSG);
+    }
+    sec->machine.stack_start = (VALUE*)(ptr + STACK_DIR_UPPER(0, size));
+    sec->machine.stack_maxsize = size - RB_PAGE_SIZE;
+#endif
+#ifdef __ia64
+    sth->machine.register_stack_maxsize = sth->machine.stack_maxsize;
+#endif
+}
+
+NOINLINE(static void fiber_setcontext(rb_fiber_t *newfib, rb_fiber_t *oldfib));
+
+static void
+fiber_setcontext(rb_fiber_t *newfib, rb_fiber_t *oldfib)
+{
+    rb_thread_t *th = GET_THREAD();
+
+    /* save oldfib's machine stack / TODO: is it needed? */
+    if (!FIBER_TERMINATED_P(oldfib)) {
+	STACK_GROW_DIR_DETECTION;
+	SET_MACHINE_STACK_END(&th->ec->machine.stack_end);
+	if (STACK_DIR_UPPER(0, 1)) {
+	    oldfib->cont.machine.stack_size = th->ec->machine.stack_start - th->ec->machine.stack_end;
+	    oldfib->cont.machine.stack = th->ec->machine.stack_end;
+	}
+	else {
+	    oldfib->cont.machine.stack_size = th->ec->machine.stack_end - th->ec->machine.stack_start;
+	    oldfib->cont.machine.stack = th->ec->machine.stack_start;
+	}
+    }
+
+    /* exchange machine_stack_start between oldfib and newfib */
+    oldfib->cont.saved_ec.machine.stack_start = th->ec->machine.stack_start;
+
+    /* oldfib->machine.stack_end should be NULL */
+    oldfib->cont.saved_ec.machine.stack_end = NULL;
+
+    /* restore thread context */
+    fiber_restore_thread(th, newfib);
+
+    /* swap machine context */
+#if defined(FIBER_USE_COROUTINE)
+    coroutine_transfer(&oldfib->context, &newfib->context);
+#elif defined(_WIN32)
+    SwitchToFiber(newfib->fib_handle);
+#else
+    if (!newfib->context.uc_stack.ss_sp && th->root_fiber != newfib) {
+	rb_bug("non_root_fiber->context.uc_stac.ss_sp should not be NULL");
+    }
+    swapcontext(&oldfib->context, &newfib->context);
+#endif
+}
+#endif /* FIBER_USE_NATIVE */
+
+NOINLINE(NORETURN(static void cont_restore_1(rb_context_t *)));
+
+static void
+cont_restore_1(rb_context_t *cont)
+{
+    cont_restore_thread(cont);
+
+    /* restore machine stack */
+#ifdef _M_AMD64
+    {
+	/* workaround for x64 SEH */
+	jmp_buf buf;
+	setjmp(buf);
+	((_JUMP_BUFFER*)(&cont->jmpbuf))->Frame =
+	    ((_JUMP_BUFFER*)(&buf))->Frame;
+    }
+#endif
+    if (cont->machine.stack_src) {
+	FLUSH_REGISTER_WINDOWS;
+	MEMCPY(cont->machine.stack_src, cont->machine.stack,
+		VALUE, cont->machine.stack_size);
+    }
+
+#ifdef __ia64
+    if (cont->machine.register_stack_src) {
+	MEMCPY(cont->machine.register_stack_src, cont->machine.register_stack,
+	       VALUE, cont->machine.register_stack_size);
+    }
+#endif
+
+    ruby_longjmp(cont->jmpbuf, 1);
+}
+
+NORETURN(NOINLINE(static void cont_restore_0(rb_context_t *, VALUE *)));
+
+#ifdef __ia64
+#define C(a) rse_##a##0, rse_##a##1, rse_##a##2, rse_##a##3, rse_##a##4
+#define E(a) rse_##a##0= rse_##a##1= rse_##a##2= rse_##a##3= rse_##a##4
+static volatile int C(a), C(b), C(c), C(d), C(e);
+static volatile int C(f), C(g), C(h), C(i), C(j);
+static volatile int C(k), C(l), C(m), C(n), C(o);
+static volatile int C(p), C(q), C(r), C(s), C(t);
+#if 0
+{/* the above lines make cc-mode.el confused so much */}
+#endif
+int rb_dummy_false = 0;
+NORETURN(NOINLINE(static void register_stack_extend(rb_context_t *, VALUE *, VALUE *)));
+static void
+register_stack_extend(rb_context_t *cont, VALUE *vp, VALUE *curr_bsp)
+{
+    if (rb_dummy_false) {
+        /* use registers as much as possible */
+        E(a) = E(b) = E(c) = E(d) = E(e) =
+        E(f) = E(g) = E(h) = E(i) = E(j) =
+        E(k) = E(l) = E(m) = E(n) = E(o) =
+        E(p) = E(q) = E(r) = E(s) = E(t) = 0;
+        E(a) = E(b) = E(c) = E(d) = E(e) =
+        E(f) = E(g) = E(h) = E(i) = E(j) =
+        E(k) = E(l) = E(m) = E(n) = E(o) =
+        E(p) = E(q) = E(r) = E(s) = E(t) = 0;
+    }
+    if (curr_bsp < cont->machine.register_stack_src+cont->machine.register_stack_size) {
+        register_stack_extend(cont, vp, (VALUE*)rb_ia64_bsp());
+    }
+    cont_restore_0(cont, vp);
+}
+#undef C
+#undef E
+#endif
+
+static void
+cont_restore_0(rb_context_t *cont, VALUE *addr_in_prev_frame)
+{
+    if (cont->machine.stack_src) {
+#ifdef HAVE_ALLOCA
+#define STACK_PAD_SIZE 1
+#else
+#define STACK_PAD_SIZE 1024
+#endif
+	VALUE space[STACK_PAD_SIZE];
+
+#if !STACK_GROW_DIRECTION
+	if (addr_in_prev_frame > &space[0]) {
+	    /* Stack grows downward */
+#endif
+#if STACK_GROW_DIRECTION <= 0
+	    volatile VALUE *const end = cont->machine.stack_src;
+	    if (&space[0] > end) {
+# ifdef HAVE_ALLOCA
+		volatile VALUE *sp = ALLOCA_N(VALUE, &space[0] - end);
+		space[0] = *sp;
+# else
+		cont_restore_0(cont, &space[0]);
+# endif
+	    }
+#endif
+#if !STACK_GROW_DIRECTION
+	}
+	else {
+	    /* Stack grows upward */
+#endif
+#if STACK_GROW_DIRECTION >= 0
+	    volatile VALUE *const end = cont->machine.stack_src + cont->machine.stack_size;
+	    if (&space[STACK_PAD_SIZE] < end) {
+# ifdef HAVE_ALLOCA
+		volatile VALUE *sp = ALLOCA_N(VALUE, end - &space[STACK_PAD_SIZE]);
+		space[0] = *sp;
+# else
+		cont_restore_0(cont, &space[STACK_PAD_SIZE-1]);
+# endif
+	    }
+#endif
+#if !STACK_GROW_DIRECTION
+	}
+#endif
+    }
+    cont_restore_1(cont);
+}
+#ifdef __ia64
+#define cont_restore_0(cont, vp) register_stack_extend((cont), (vp), (VALUE*)rb_ia64_bsp())
+#endif
+
+/*
+ *  Document-class: Continuation
+ *
+ *  Continuation objects are generated by Kernel#callcc,
+ *  after having +require+d <i>continuation</i>. They hold
+ *  a return address and execution context, allowing a nonlocal return
+ *  to the end of the <code>callcc</code> block from anywhere within a
+ *  program. Continuations are somewhat analogous to a structured
+ *  version of C's <code>setjmp/longjmp</code> (although they contain
+ *  more state, so you might consider them closer to threads).
+ *
+ *  For instance:
+ *
+ *     require "continuation"
+ *     arr = [ "Freddie", "Herbie", "Ron", "Max", "Ringo" ]
+ *     callcc{|cc| $cc = cc}
+ *     puts(message = arr.shift)
+ *     $cc.call unless message =~ /Max/
+ *
+ *  <em>produces:</em>
+ *
+ *     Freddie
+ *     Herbie
+ *     Ron
+ *     Max
+ *
+ *  Also you can call callcc in other methods:
+ *
+ *     require "continuation"
+ *
+ *     def g
+ *       arr = [ "Freddie", "Herbie", "Ron", "Max", "Ringo" ]
+ *       cc = callcc { |cc| cc }
+ *       puts arr.shift
+ *       return cc, arr.size
+ *     end
+ *
+ *     def f
+ *       c, size = g
+ *       c.call(c) if size > 1
+ *     end
+ *
+ *     f
+ *
+ *  This (somewhat contrived) example allows the inner loop to abandon
+ *  processing early:
+ *
+ *     require "continuation"
+ *     callcc {|cont|
+ *       for i in 0..4
+ *         print "\n#{i}: "
+ *         for j in i*5...(i+1)*5
+ *           cont.call() if j == 17
+ *           printf "%3d", j
+ *         end
+ *       end
+ *     }
+ *     puts
+ *
+ *  <em>produces:</em>
+ *
+ *     0:   0  1  2  3  4
+ *     1:   5  6  7  8  9
+ *     2:  10 11 12 13 14
+ *     3:  15 16
+ */
+
+/*
+ *  call-seq:
+ *     callcc {|cont| block }   ->  obj
+ *
+ *  Generates a Continuation object, which it passes to
+ *  the associated block. You need to <code>require
+ *  'continuation'</code> before using this method. Performing a
+ *  <em>cont</em><code>.call</code> will cause the #callcc
+ *  to return (as will falling through the end of the block). The
+ *  value returned by the #callcc is the value of the
+ *  block, or the value passed to <em>cont</em><code>.call</code>. See
+ *  class Continuation for more details. Also see
+ *  Kernel#throw for an alternative mechanism for
+ *  unwinding a call stack.
+ */
+
+static VALUE
+rb_callcc(VALUE self)
+{
+    volatile int called;
+    volatile VALUE val = cont_capture(&called);
+
+    if (called) {
+	return val;
+    }
+    else {
+	return rb_yield(val);
+    }
+}
+
+static VALUE
+make_passing_arg(int argc, const VALUE *argv)
+{
+    switch (argc) {
+      case 0:
+	return Qnil;
+      case 1:
+	return argv[0];
+      default:
+	return rb_ary_new4(argc, argv);
+    }
+}
+
+/* CAUTION!! : Currently, error in rollback_func is not supported  */
+/* same as rb_protect if set rollback_func to NULL */
+void
+ruby_register_rollback_func_for_ensure(VALUE (*ensure_func)(ANYARGS), VALUE (*rollback_func)(ANYARGS))
+{
+    st_table **table_p = &GET_VM()->ensure_rollback_table;
+    if (UNLIKELY(*table_p == NULL)) {
+	*table_p = st_init_numtable();
+    }
+    st_insert(*table_p, (st_data_t)ensure_func, (st_data_t)rollback_func);
+}
+
+static inline VALUE
+lookup_rollback_func(VALUE (*ensure_func)(ANYARGS))
+{
+    st_table *table = GET_VM()->ensure_rollback_table;
+    st_data_t val;
+    if (table && st_lookup(table, (st_data_t)ensure_func, &val))
+	return (VALUE) val;
+    return Qundef;
+}
+
+
+static inline void
+rollback_ensure_stack(VALUE self,rb_ensure_list_t *current,rb_ensure_entry_t *target)
+{
+    rb_ensure_list_t *p;
+    rb_ensure_entry_t *entry;
+    size_t i, j;
+    size_t cur_size;
+    size_t target_size;
+    size_t base_point;
+    VALUE (*func)(ANYARGS);
+
+    cur_size = 0;
+    for (p=current; p; p=p->next)
+	cur_size++;
+    target_size = 0;
+    for (entry=target; entry->marker; entry++)
+	target_size++;
+
+    /* search common stack point */
+    p = current;
+    base_point = cur_size;
+    while (base_point) {
+	if (target_size >= base_point &&
+	    p->entry.marker == target[target_size - base_point].marker)
+	    break;
+	base_point --;
+	p = p->next;
+    }
+
+    /* rollback function check */
+    for (i=0; i < target_size - base_point; i++) {
+	if (!lookup_rollback_func(target[i].e_proc)) {
+	    rb_raise(rb_eRuntimeError, "continuation called from out of critical rb_ensure scope");
+	}
+    }
+    /* pop ensure stack */
+    while (cur_size > base_point) {
+	/* escape from ensure block */
+	(*current->entry.e_proc)(current->entry.data2);
+	current = current->next;
+	cur_size--;
+    }
+    /* push ensure stack */
+    for (j = 0; j < i; j++) {
+        func = (VALUE (*)(ANYARGS)) lookup_rollback_func(target[i - j - 1].e_proc);
+        if ((VALUE)func != Qundef) {
+            (*func)(target[i - j - 1].data2);
+        }
+    }
+}
+
+/*
+ *  call-seq:
+ *     cont.call(args, ...)
+ *     cont[args, ...]
+ *
+ *  Invokes the continuation. The program continues from the end of the
+ *  <code>callcc</code> block. If no arguments are given, the original
+ *  <code>callcc</code> returns <code>nil</code>. If one argument is
+ *  given, <code>callcc</code> returns it. Otherwise, an array
+ *  containing <i>args</i> is returned.
+ *
+ *     callcc {|cont|  cont.call }           #=> nil
+ *     callcc {|cont|  cont.call 1 }         #=> 1
+ *     callcc {|cont|  cont.call 1, 2, 3 }   #=> [1, 2, 3]
+ */
+
+static VALUE
+rb_cont_call(int argc, VALUE *argv, VALUE contval)
+{
+    rb_context_t *cont = cont_ptr(contval);
+    rb_thread_t *th = GET_THREAD();
+
+    if (cont_thread_value(cont) != th->self) {
+	rb_raise(rb_eRuntimeError, "continuation called across threads");
+    }
+    if (cont->saved_ec.protect_tag != th->ec->protect_tag) {
+	rb_raise(rb_eRuntimeError, "continuation called across stack rewinding barrier");
+    }
+    if (cont->saved_ec.fiber_ptr) {
+	if (th->ec->fiber_ptr != cont->saved_ec.fiber_ptr) {
+	    rb_raise(rb_eRuntimeError, "continuation called across fiber");
+	}
+    }
+    rollback_ensure_stack(contval, th->ec->ensure_list, cont->ensure_array);
+
+    cont->argc = argc;
+    cont->value = make_passing_arg(argc, argv);
+
+    cont_restore_0(cont, &contval);
+    return Qnil; /* unreachable */
+}
+
+/*********/
+/* fiber */
+/*********/
+
+/*
+ *  Document-class: Fiber
+ *
+ *  Fibers are primitives for implementing light weight cooperative
+ *  concurrency in Ruby. Basically they are a means of creating code blocks
+ *  that can be paused and resumed, much like threads. The main difference
+ *  is that they are never preempted and that the scheduling must be done by
+ *  the programmer and not the VM.
+ *
+ *  As opposed to other stackless light weight concurrency models, each fiber
+ *  comes with a stack.  This enables the fiber to be paused from deeply
+ *  nested function calls within the fiber block.  See the ruby(1)
+ *  manpage to configure the size of the fiber stack(s).
+ *
+ *  When a fiber is created it will not run automatically. Rather it must
+ *  be explicitly asked to run using the <code>Fiber#resume</code> method.
+ *  The code running inside the fiber can give up control by calling
+ *  <code>Fiber.yield</code> in which case it yields control back to caller
+ *  (the caller of the <code>Fiber#resume</code>).
+ *
+ *  Upon yielding or termination the Fiber returns the value of the last
+ *  executed expression
+ *
+ *  For instance:
+ *
+ *    fiber = Fiber.new do
+ *      Fiber.yield 1
+ *      2
+ *    end
+ *
+ *    puts fiber.resume
+ *    puts fiber.resume
+ *    puts fiber.resume
+ *
+ *  <em>produces</em>
+ *
+ *    1
+ *    2
+ *    FiberError: dead fiber called
+ *
+ *  The <code>Fiber#resume</code> method accepts an arbitrary number of
+ *  parameters, if it is the first call to <code>resume</code> then they
+ *  will be passed as block arguments. Otherwise they will be the return
+ *  value of the call to <code>Fiber.yield</code>
+ *
+ *  Example:
+ *
+ *    fiber = Fiber.new do |first|
+ *      second = Fiber.yield first + 2
+ *    end
+ *
+ *    puts fiber.resume 10
+ *    puts fiber.resume 14
+ *    puts fiber.resume 18
+ *
+ *  <em>produces</em>
+ *
+ *    12
+ *    14
+ *    FiberError: dead fiber called
+ *
+ */
+
+static const rb_data_type_t fiber_data_type = {
+    "fiber",
+    {fiber_mark, fiber_free, fiber_memsize,},
+    0, 0, RUBY_TYPED_FREE_IMMEDIATELY
+};
+
+static VALUE
+fiber_alloc(VALUE klass)
+{
+    return TypedData_Wrap_Struct(klass, &fiber_data_type, 0);
+}
+
+static rb_fiber_t*
+fiber_t_alloc(VALUE fibval)
+{
+    rb_fiber_t *fib;
+    rb_thread_t *th = GET_THREAD();
+
+    if (DATA_PTR(fibval) != 0) {
+	rb_raise(rb_eRuntimeError, "cannot initialize twice");
+    }
+
+    THREAD_MUST_BE_RUNNING(th);
+    fib = ZALLOC(rb_fiber_t);
+    fib->cont.self = fibval;
+    fib->cont.type = FIBER_CONTEXT;
+    cont_init(&fib->cont, th);
+    fib->cont.saved_ec.fiber_ptr = fib;
+    fib->prev = NULL;
+
+    /* fib->status == 0 == CREATED
+     * So that we don't need to set status: fiber_status_set(fib, FIBER_CREATED); */
+    VM_ASSERT(FIBER_CREATED_P(fib));
+
+    DATA_PTR(fibval) = fib;
+
+    return fib;
+}
+
+rb_control_frame_t *
+rb_vm_push_frame(rb_execution_context_t *sec,
+		 const rb_iseq_t *iseq,
+		 VALUE type,
+		 VALUE self,
+		 VALUE specval,
+		 VALUE cref_or_me,
+		 const VALUE *pc,
+		 VALUE *sp,
+		 int local_size,
+		 int stack_max);
+
+static VALUE
+fiber_init(VALUE fibval, VALUE proc)
+{
+    rb_fiber_t *fib = fiber_t_alloc(fibval);
+    rb_context_t *cont = &fib->cont;
+    rb_execution_context_t *sec = &cont->saved_ec;
+    rb_thread_t *cth = GET_THREAD();
+    rb_vm_t *vm = cth->vm;
+    size_t fib_stack_bytes = vm->default_params.fiber_vm_stack_size;
+    size_t thr_stack_bytes = vm->default_params.thread_vm_stack_size;
+    VALUE *vm_stack;
+
+    /* initialize cont */
+    cont->saved_vm_stack.ptr = NULL;
+    if (fib_stack_bytes == thr_stack_bytes) {
+        vm_stack = rb_thread_recycle_stack(fib_stack_bytes / sizeof(VALUE));
+    }
+    else {
+        vm_stack = ruby_xmalloc(fib_stack_bytes);
+    }
+    rb_ec_set_vm_stack(sec, vm_stack, fib_stack_bytes / sizeof(VALUE));
+    sec->cfp = (void *)(sec->vm_stack + sec->vm_stack_size);
+
+    rb_vm_push_frame(sec,
+		     NULL,
+		     VM_FRAME_MAGIC_DUMMY | VM_ENV_FLAG_LOCAL | VM_FRAME_FLAG_FINISH | VM_FRAME_FLAG_CFRAME,
+		     Qnil, /* self */
+		     VM_BLOCK_HANDLER_NONE,
+		     0, /* specval */
+		     NULL, /* pc */
+		     sec->vm_stack, /* sp */
+		     0, /* local_size */
+		     0);
+
+    sec->tag = NULL;
+    sec->local_storage = NULL;
+    sec->local_storage_recursive_hash = Qnil;
+    sec->local_storage_recursive_hash_for_trace = Qnil;
+
+    fib->first_proc = proc;
+
+#if !FIBER_USE_NATIVE
+    MEMCPY(&cont->jmpbuf, &cth->root_jmpbuf, rb_jmpbuf_t, 1);
+#endif
+
+    return fibval;
+}
+
+/* :nodoc: */
+static VALUE
+rb_fiber_init(VALUE fibval)
+{
+    return fiber_init(fibval, rb_block_proc());
+}
+
+VALUE
+rb_fiber_new(VALUE (*func)(ANYARGS), VALUE obj)
+{
+    return fiber_init(fiber_alloc(rb_cFiber), rb_proc_new(func, obj));
+}
+
+static void rb_fiber_terminate(rb_fiber_t *fib, int need_interrupt);
+
+void
+rb_fiber_start(void)
+{
+    rb_thread_t * volatile th = GET_THREAD();
+    rb_fiber_t *fib = th->ec->fiber_ptr;
+    rb_proc_t *proc;
+    enum ruby_tag_type state;
+    int need_interrupt = TRUE;
+
+    VM_ASSERT(th->ec == ruby_current_execution_context_ptr);
+    VM_ASSERT(FIBER_RESUMED_P(fib));
+
+    EC_PUSH_TAG(th->ec);
+    if ((state = EC_EXEC_TAG()) == TAG_NONE) {
+	rb_context_t *cont = &VAR_FROM_MEMORY(fib)->cont;
+	int argc;
+	const VALUE *argv, args = cont->value;
+	GetProcPtr(fib->first_proc, proc);
+	argv = (argc = cont->argc) > 1 ? RARRAY_CONST_PTR(args) : &args;
+	cont->value = Qnil;
+	th->ec->errinfo = Qnil;
+	th->ec->root_lep = rb_vm_proc_local_ep(fib->first_proc);
+	th->ec->root_svar = Qfalse;
+
+	EXEC_EVENT_HOOK(th->ec, RUBY_EVENT_FIBER_SWITCH, th->self, 0, 0, 0, Qnil);
+	cont->value = rb_vm_invoke_proc(th->ec, proc, argc, argv, VM_BLOCK_HANDLER_NONE);
+    }
+    EC_POP_TAG();
+
+    if (state) {
+	VALUE err = th->ec->errinfo;
+	VM_ASSERT(FIBER_RESUMED_P(fib));
+
+	if (state == TAG_RAISE || state == TAG_FATAL) {
+	    rb_threadptr_pending_interrupt_enque(th, err);
+	}
+	else {
+	    err = rb_vm_make_jump_tag_but_local_jump(state, err);
+	    if (!NIL_P(err)) {
+		rb_threadptr_pending_interrupt_enque(th, err);
+	    }
+	}
+	need_interrupt = TRUE;
+    }
+
+    rb_fiber_terminate(fib, need_interrupt);
+    VM_UNREACHABLE(rb_fiber_start);
+}
+
+static rb_fiber_t *
+root_fiber_alloc(rb_thread_t *th)
+{
+    VALUE fibval = fiber_alloc(rb_cFiber);
+    rb_fiber_t *fib = th->ec->fiber_ptr;
+
+    VM_ASSERT(DATA_PTR(fibval) == NULL);
+    VM_ASSERT(fib->cont.type == FIBER_CONTEXT);
+    VM_ASSERT(fib->status == FIBER_RESUMED);
+
+    th->root_fiber = fib;
+    DATA_PTR(fibval) = fib;
+    fib->cont.self = fibval;
+
+#if FIBER_USE_NATIVE
+#if defined(FIBER_USE_COROUTINE)
+    coroutine_initialize(&fib->context, NULL, NULL, 0);
+#elif defined(_WIN32)
+    /* setup fib_handle for root Fiber */
+    if (fib->fib_handle == 0) {
+        if ((fib->fib_handle = ConvertThreadToFiber(0)) == 0) {
+            rb_bug("root_fiber_alloc: ConvertThreadToFiber() failed - %s\n", rb_w32_strerror(-1));
+        }
+    }
+    else {
+        rb_bug("root_fiber_alloc: fib_handle is not NULL.");
+    }
+#endif
+#endif
+
+    return fib;
+}
+
+void
+rb_threadptr_root_fiber_setup(rb_thread_t *th)
+{
+    rb_fiber_t *fib = ruby_mimmalloc(sizeof(rb_fiber_t));
+    MEMZERO(fib, rb_fiber_t, 1);
+    fib->cont.type = FIBER_CONTEXT;
+    fib->cont.saved_ec.fiber_ptr = fib;
+    fib->cont.saved_ec.thread_ptr = th;
+    fiber_status_set(fib, FIBER_RESUMED); /* skip CREATED */
+    th->ec = &fib->cont.saved_ec;
+
+    /* NOTE: On WIN32, fib_handle is not allocated yet. */
+}
+
+void
+rb_threadptr_root_fiber_release(rb_thread_t *th)
+{
+    if (th->root_fiber) {
+	/* ignore. A root fiber object will free th->ec */
+    }
+    else {
+	VM_ASSERT(th->ec->fiber_ptr->cont.type == FIBER_CONTEXT);
+	VM_ASSERT(th->ec->fiber_ptr->cont.self == 0);
+	fiber_free(th->ec->fiber_ptr);
+
+	if (th->ec == ruby_current_execution_context_ptr) {
+	    ruby_current_execution_context_ptr = NULL;
+	}
+	th->ec = NULL;
+    }
+}
+
+static inline rb_fiber_t*
+fiber_current(void)
+{
+    rb_execution_context_t *ec = GET_EC();
+    if (ec->fiber_ptr->cont.self == 0) {
+	root_fiber_alloc(rb_ec_thread_ptr(ec));
+    }
+    return ec->fiber_ptr;
+}
+
+static inline rb_fiber_t*
+return_fiber(void)
+{
+    rb_fiber_t *fib = fiber_current();
+    rb_fiber_t *prev = fib->prev;
+
+    if (!prev) {
+	rb_thread_t *th = GET_THREAD();
+	rb_fiber_t *root_fiber = th->root_fiber;
+
+	VM_ASSERT(root_fiber != NULL);
+
+	if (root_fiber == fib) {
+	    rb_raise(rb_eFiberError, "can't yield from root fiber");
+	}
+	return root_fiber;
+    }
+    else {
+	fib->prev = NULL;
+	return prev;
+    }
+}
+
+VALUE
+rb_fiber_current(void)
+{
+    return fiber_current()->cont.self;
+}
+
+static inline VALUE
+fiber_store(rb_fiber_t *next_fib, rb_thread_t *th)
+{
+    rb_fiber_t *fib;
+
+    if (th->ec->fiber_ptr != NULL) {
+        fib = th->ec->fiber_ptr;
+    }
+    else {
+        /* create root fiber */
+        fib = root_fiber_alloc(th);
+    }
+
+    VM_ASSERT(FIBER_RESUMED_P(fib) || FIBER_TERMINATED_P(fib));
+    VM_ASSERT(FIBER_RUNNABLE_P(next_fib));
+
+#if FIBER_USE_NATIVE
+    if (FIBER_CREATED_P(next_fib)) {
+        fiber_initialize_machine_stack_context(next_fib, th->vm->default_params.fiber_machine_stack_size);
+    }
+#endif
+
+    if (FIBER_RESUMED_P(fib)) fiber_status_set(fib, FIBER_SUSPENDED);
+
+#if FIBER_USE_NATIVE == 0
+    /* should (re-)allocate stack are before fib->status change to pass fiber_verify() */
+    cont_save_machine_stack(th, &fib->cont);
+#endif
+
+    fiber_status_set(next_fib, FIBER_RESUMED);
+
+#if FIBER_USE_NATIVE
+    fiber_setcontext(next_fib, fib);
+    /* restored */
+#ifdef MAX_MACHINE_STACK_CACHE
+    if (terminated_machine_stack.ptr) {
+        if (machine_stack_cache_index < MAX_MACHINE_STACK_CACHE) {
+            machine_stack_cache[machine_stack_cache_index++] = terminated_machine_stack;
+        }
+        else {
+            if (terminated_machine_stack.ptr != fib->cont.machine.stack) {
+#ifdef _WIN32
+                VirtualFree(terminated_machine_stack.ptr, 0, MEM_RELEASE);
+#else
+                munmap((void*)terminated_machine_stack.ptr, terminated_machine_stack.size * sizeof(VALUE));
+#endif
+            }
+            else {
+                rb_bug("terminated fiber resumed");
+            }
+        }
+        terminated_machine_stack.ptr = NULL;
+        terminated_machine_stack.size = 0;
+    }
+#endif /* not _WIN32 */
+    fib = th->ec->fiber_ptr;
+    if (fib->cont.argc == -1) rb_exc_raise(fib->cont.value);
+    return fib->cont.value;
+
+#else /* FIBER_USE_NATIVE */
+    fib->cont.saved_ec.machine.stack_end = NULL;
+    if (ruby_setjmp(fib->cont.jmpbuf)) {
+        /* restored */
+        fib = th->ec->fiber_ptr;
+        if (fib->cont.argc == -1) rb_exc_raise(fib->cont.value);
+        if (next_fib->cont.value == Qundef) {
+            cont_restore_0(&next_fib->cont, &next_fib->cont.value);
+            VM_UNREACHABLE(fiber_store);
+        }
+        return fib->cont.value;
+    }
+    else {
+        VALUE undef = Qundef;
+        cont_restore_0(&next_fib->cont, &undef);
+        VM_UNREACHABLE(fiber_store);
+    }
+#endif /* FIBER_USE_NATIVE */
+}
+
+static inline VALUE
+fiber_switch(rb_fiber_t *fib, int argc, const VALUE *argv, int is_resume)
+{
+    VALUE value;
+    rb_context_t *cont = &fib->cont;
+    rb_thread_t *th = GET_THREAD();
+
+    /* make sure the root_fiber object is available */
+    if (th->root_fiber == NULL) root_fiber_alloc(th);
+
+    if (th->ec->fiber_ptr == fib) {
+	/* ignore fiber context switch
+         * because destination fiber is same as current fiber
+	 */
+	return make_passing_arg(argc, argv);
+    }
+
+    if (cont_thread_value(cont) != th->self) {
+	rb_raise(rb_eFiberError, "fiber called across threads");
+    }
+    else if (cont->saved_ec.protect_tag != th->ec->protect_tag) {
+	rb_raise(rb_eFiberError, "fiber called across stack rewinding barrier");
+    }
+    else if (FIBER_TERMINATED_P(fib)) {
+	value = rb_exc_new2(rb_eFiberError, "dead fiber called");
+
+	if (!FIBER_TERMINATED_P(th->ec->fiber_ptr)) {
+	    rb_exc_raise(value);
+	    VM_UNREACHABLE(fiber_switch);
+	}
+	else {
+	    /* th->ec->fiber_ptr is also dead => switch to root fiber */
+	    /* (this means we're being called from rb_fiber_terminate, */
+	    /* and the terminated fiber's return_fiber() is already dead) */
+	    VM_ASSERT(FIBER_SUSPENDED_P(th->root_fiber));
+
+	    cont = &th->root_fiber->cont;
+	    cont->argc = -1;
+	    cont->value = value;
+#if FIBER_USE_NATIVE
+	    fiber_setcontext(th->root_fiber, th->ec->fiber_ptr);
+#else
+	    cont_restore_0(cont, &value);
+#endif
+	    VM_UNREACHABLE(fiber_switch);
+	}
+    }
+
+    if (is_resume) {
+	fib->prev = fiber_current();
+    }
+
+    VM_ASSERT(FIBER_RUNNABLE_P(fib));
+
+    cont->argc = argc;
+    cont->value = make_passing_arg(argc, argv);
+    value = fiber_store(fib, th);
+    RUBY_VM_CHECK_INTS(th->ec);
+
+    EXEC_EVENT_HOOK(th->ec, RUBY_EVENT_FIBER_SWITCH, th->self, 0, 0, 0, Qnil);
+
+    return value;
+}
+
+VALUE
+rb_fiber_transfer(VALUE fibval, int argc, const VALUE *argv)
+{
+    return fiber_switch(fiber_ptr(fibval), argc, argv, 0);
+}
+
+void
+rb_fiber_close(rb_fiber_t *fib)
+{
+    rb_execution_context_t *ec = &fib->cont.saved_ec;
+    VALUE *vm_stack = ec->vm_stack;
+    size_t stack_bytes = ec->vm_stack_size * sizeof(VALUE);
+
+    fiber_status_set(fib, FIBER_TERMINATED);
+    if (stack_bytes == rb_ec_vm_ptr(ec)->default_params.thread_vm_stack_size) {
+        rb_thread_recycle_stack_release(vm_stack);
+    }
+    else {
+        ruby_xfree(vm_stack);
+    }
+    rb_ec_set_vm_stack(ec, NULL, 0);
+
+#if !FIBER_USE_NATIVE
+    /* should not mark machine stack any more */
+    ec->machine.stack_end = NULL;
+#endif
+}
+
+static void
+rb_fiber_terminate(rb_fiber_t *fib, int need_interrupt)
+{
+    VALUE value = fib->cont.value;
+    rb_fiber_t *ret_fib;
+
+    VM_ASSERT(FIBER_RESUMED_P(fib));
+    rb_fiber_close(fib);
+
+#if FIBER_USE_NATIVE
+#if defined(FIBER_USE_COROUTINE)
+    coroutine_destroy(&fib->context);
+#elif !defined(_WIN32)
+    fib->context.uc_stack.ss_sp = NULL;
+#endif
+
+#ifdef MAX_MACHINE_STACK_CACHE
+    /* Ruby must not switch to other thread until storing terminated_machine_stack */
+    terminated_machine_stack.ptr = fib->ss_sp;
+    terminated_machine_stack.size = fib->ss_size / sizeof(VALUE);
+    fib->ss_sp = NULL;
+    fib->cont.machine.stack = NULL;
+    fib->cont.machine.stack_size = 0;
+#endif
+#endif
+
+    ret_fib = return_fiber();
+    if (need_interrupt) RUBY_VM_SET_INTERRUPT(&ret_fib->cont.saved_ec);
+    fiber_switch(ret_fib, 1, &value, 0);
+}
+
+VALUE
+rb_fiber_resume(VALUE fibval, int argc, const VALUE *argv)
+{
+    rb_fiber_t *fib = fiber_ptr(fibval);
+
+    if (fib->prev != 0 || fiber_is_root_p(fib)) {
+	rb_raise(rb_eFiberError, "double resume");
+    }
+    if (fib->transferred != 0) {
+	rb_raise(rb_eFiberError, "cannot resume transferred Fiber");
+    }
+
+    return fiber_switch(fib, argc, argv, 1);
+}
+
+VALUE
+rb_fiber_yield(int argc, const VALUE *argv)
+{
+    return fiber_switch(return_fiber(), argc, argv, 0);
+}
+
+void
+rb_fiber_reset_root_local_storage(rb_thread_t *th)
+{
+    if (th->root_fiber && th->root_fiber != th->ec->fiber_ptr) {
+	th->ec->local_storage = th->root_fiber->cont.saved_ec.local_storage;
+    }
+}
+
+/*
+ *  call-seq:
+ *     fiber.alive? -> true or false
+ *
+ *  Returns true if the fiber can still be resumed (or transferred
+ *  to). After finishing execution of the fiber block this method will
+ *  always return false. You need to <code>require 'fiber'</code>
+ *  before using this method.
+ */
+VALUE
+rb_fiber_alive_p(VALUE fibval)
+{
+    return FIBER_TERMINATED_P(fiber_ptr(fibval)) ? Qfalse : Qtrue;
+}
+
+/*
+ *  call-seq:
+ *     fiber.resume(args, ...) -> obj
+ *
+ *  Resumes the fiber from the point at which the last <code>Fiber.yield</code>
+ *  was called, or starts running it if it is the first call to
+ *  <code>resume</code>. Arguments passed to resume will be the value of
+ *  the <code>Fiber.yield</code> expression or will be passed as block
+ *  parameters to the fiber's block if this is the first <code>resume</code>.
+ *
+ *  Alternatively, when resume is called it evaluates to the arguments passed
+ *  to the next <code>Fiber.yield</code> statement inside the fiber's block
+ *  or to the block value if it runs to completion without any
+ *  <code>Fiber.yield</code>
+ */
+static VALUE
+rb_fiber_m_resume(int argc, VALUE *argv, VALUE fib)
+{
+    return rb_fiber_resume(fib, argc, argv);
+}
+
+/*
+ *  call-seq:
+ *     fiber.transfer(args, ...) -> obj
+ *
+ *  Transfer control to another fiber, resuming it from where it last
+ *  stopped or starting it if it was not resumed before. The calling
+ *  fiber will be suspended much like in a call to
+ *  <code>Fiber.yield</code>. You need to <code>require 'fiber'</code>
+ *  before using this method.
+ *
+ *  The fiber which receives the transfer call is treats it much like
+ *  a resume call. Arguments passed to transfer are treated like those
+ *  passed to resume.
+ *
+ *  You cannot resume a fiber that transferred control to another one.
+ *  This will cause a double resume error. You need to transfer control
+ *  back to this fiber before it can yield and resume.
+ *
+ *  Example:
+ *
+ *    fiber1 = Fiber.new do
+ *      puts "In Fiber 1"
+ *      Fiber.yield
+ *    end
+ *
+ *    fiber2 = Fiber.new do
+ *      puts "In Fiber 2"
+ *      fiber1.transfer
+ *      puts "Never see this message"
+ *    end
+ *
+ *    fiber3 = Fiber.new do
+ *      puts "In Fiber 3"
+ *    end
+ *
+ *    fiber2.resume
+ *    fiber3.resume
+ *
+ *  <em>produces</em>
+ *
+ *    In fiber 2
+ *    In fiber 1
+ *    In fiber 3
+ *
+ */
+static VALUE
+rb_fiber_m_transfer(int argc, VALUE *argv, VALUE fibval)
+{
+    rb_fiber_t *fib = fiber_ptr(fibval);
+    fib->transferred = 1;
+    return fiber_switch(fib, argc, argv, 0);
+}
+
+/*
+ *  call-seq:
+ *     Fiber.yield(args, ...) -> obj
+ *
+ *  Yields control back to the context that resumed the fiber, passing
+ *  along any arguments that were passed to it. The fiber will resume
+ *  processing at this point when <code>resume</code> is called next.
+ *  Any arguments passed to the next <code>resume</code> will be the
+ *  value that this <code>Fiber.yield</code> expression evaluates to.
+ */
+static VALUE
+rb_fiber_s_yield(int argc, VALUE *argv, VALUE klass)
+{
+    return rb_fiber_yield(argc, argv);
+}
+
+/*
+ *  call-seq:
+ *     Fiber.current() -> fiber
+ *
+ *  Returns the current fiber. You need to <code>require 'fiber'</code>
+ *  before using this method. If you are not running in the context of
+ *  a fiber this method will return the root fiber.
+ */
+static VALUE
+rb_fiber_s_current(VALUE klass)
+{
+    return rb_fiber_current();
+}
+
+/*
+ * call-seq:
+ *   fiber.to_s   -> string
+ *
+ * Returns fiber information string.
+ *
+ */
+
+static VALUE
+fiber_to_s(VALUE fibval)
+{
+    const rb_fiber_t *fib = fiber_ptr(fibval);
+    const rb_proc_t *proc;
+    char status_info[0x10];
+
+    snprintf(status_info, 0x10, " (%s)", fiber_status_name(fib->status));
+    if (!rb_obj_is_proc(fib->first_proc)) {
+	VALUE str = rb_any_to_s(fibval);
+	strlcat(status_info, ">", sizeof(status_info));
+	rb_str_set_len(str, RSTRING_LEN(str)-1);
+	rb_str_cat_cstr(str, status_info);
+	return str;
+    }
+    GetProcPtr(fib->first_proc, proc);
+    return rb_block_to_s(fibval, &proc->block, status_info);
+}
+
+#ifdef HAVE_WORKING_FORK
+void
+rb_fiber_atfork(rb_thread_t *th)
+{
+    if (th->root_fiber) {
+        if (&th->root_fiber->cont.saved_ec != th->ec) {
+            th->root_fiber = th->ec->fiber_ptr;
+        }
+        th->root_fiber->prev = 0;
+    }
+}
+#endif
+
+/*
+ *  Document-class: FiberError
+ *
+ *  Raised when an invalid operation is attempted on a Fiber, in
+ *  particular when attempting to call/resume a dead fiber,
+ *  attempting to yield from the root fiber, or calling a fiber across
+ *  threads.
+ *
+ *     fiber = Fiber.new{}
+ *     fiber.resume #=> nil
+ *     fiber.resume #=> FiberError: dead fiber called
+ */
+
+void
+Init_Cont(void)
+{
+#if FIBER_USE_NATIVE
+    rb_thread_t *th = GET_THREAD();
+
+#ifdef _WIN32
+    SYSTEM_INFO info;
+    GetSystemInfo(&info);
+    pagesize = info.dwPageSize;
+#else /* not WIN32 */
+    pagesize = sysconf(_SC_PAGESIZE);
+#endif
+    SET_MACHINE_STACK_END(&th->ec->machine.stack_end);
+#endif
+
+    rb_cFiber = rb_define_class("Fiber", rb_cObject);
+    rb_define_alloc_func(rb_cFiber, fiber_alloc);
+    rb_eFiberError = rb_define_class("FiberError", rb_eStandardError);
+    rb_define_singleton_method(rb_cFiber, "yield", rb_fiber_s_yield, -1);
+    rb_define_method(rb_cFiber, "initialize", rb_fiber_init, 0);
+    rb_define_method(rb_cFiber, "resume", rb_fiber_m_resume, -1);
+    rb_define_method(rb_cFiber, "to_s", fiber_to_s, 0);
+    rb_define_alias(rb_cFiber, "inspect", "to_s");
+}
+
+RUBY_SYMBOL_EXPORT_BEGIN
+
+void
+ruby_Init_Continuation_body(void)
+{
+    rb_cContinuation = rb_define_class("Continuation", rb_cObject);
+    rb_undef_alloc_func(rb_cContinuation);
+    rb_undef_method(CLASS_OF(rb_cContinuation), "new");
+    rb_define_method(rb_cContinuation, "call", rb_cont_call, -1);
+    rb_define_method(rb_cContinuation, "[]", rb_cont_call, -1);
+    rb_define_global_function("callcc", rb_callcc, 0);
+}
+
+void
+ruby_Init_Fiber_as_Coroutine(void)
+{
+    rb_define_method(rb_cFiber, "transfer", rb_fiber_m_transfer, -1);
+    rb_define_method(rb_cFiber, "alive?", rb_fiber_alive_p, 0);
+    rb_define_singleton_method(rb_cFiber, "current", rb_fiber_s_current, 0);
+}
+
+RUBY_SYMBOL_EXPORT_END
diff --git a/coroutine/amd64/Context.S b/coroutine/amd64/Context.S
new file mode 100644
index 0000000000..6193993e03
--- /dev/null
+++ b/coroutine/amd64/Context.S
@@ -0,0 +1,46 @@
+##
+##  This file is part of the "Coroutine" project and released under the MIT License.
+##
+##  Created by Samuel Williams on 10/5/2018.
+##  Copyright, 2018, by Samuel Williams. All rights reserved.
+##
+
+.text
+
+# For older linkers
+.globl _coroutine_transfer
+_coroutine_transfer:
+
+.globl coroutine_transfer
+coroutine_transfer:
+	# Save caller state
+	pushq %rbp
+	pushq %rbx
+	pushq %r12
+	pushq %r13
+	pushq %r14
+	pushq %r15
+
+	# Save caller stack pointer
+	movq %rsp, (%rdi)
+
+	# Restore callee stack pointer
+	movq (%rsi), %rsp
+
+	# Restore callee stack
+	popq %r15
+	popq %r14
+	popq %r13
+	popq %r12
+	popq %rbx
+	popq %rbp
+
+	# Put the first argument into the return value
+	movq %rdi, %rax
+
+	# We pop the return address and jump to it
+	ret
+
+#if defined(__linux__) && defined(__ELF__)
+.section .note.GNU-stack,"",%progbits
+#endif
diff --git a/coroutine/amd64/Context.h b/coroutine/amd64/Context.h
new file mode 100644
index 0000000000..1801c1e2c5
--- /dev/null
+++ b/coroutine/amd64/Context.h
@@ -0,0 +1,59 @@
+/*
+ *  This file is part of the "Coroutine" project and released under the MIT License.
+ *
+ *  Created by Samuel Williams on 10/5/2018.
+ *  Copyright, 2018, by Samuel Williams. All rights reserved.
+*/
+
+#pragma once
+
+#include <assert.h>
+#include <string.h>
+
+#if __cplusplus
+extern "C" {
+#endif
+
+#define COROUTINE __attribute__((noreturn)) void
+
+const size_t COROUTINE_REGISTERS = 6;
+
+typedef struct
+{
+    void **stack_pointer;
+} coroutine_context;
+
+typedef COROUTINE(* coroutine_start)(coroutine_context *from, coroutine_context *self);
+
+static inline void coroutine_initialize(
+    coroutine_context *context,
+    coroutine_start start,
+    void *stack_pointer,
+    size_t stack_size
+) {
+    /* Force 16-byte alignment */
+    context->stack_pointer = (void**)((uintptr_t)stack_pointer & ~0xF);
+
+    if (!start) {
+        assert(!context->stack_pointer);
+        /* We are main coroutine for this thread */
+        return;
+    }
+
+    *--context->stack_pointer = NULL;
+    *--context->stack_pointer = (void*)start;
+
+    context->stack_pointer -= COROUTINE_REGISTERS;
+    memset(context->stack_pointer, 0, sizeof(void*) * COROUTINE_REGISTERS);
+}
+
+coroutine_context * coroutine_transfer(coroutine_context * current, coroutine_context * target);
+
+static inline void coroutine_destroy(coroutine_context * context)
+{
+    context->stack_pointer = NULL;
+}
+
+#if __cplusplus
+}
+#endif
diff --git a/coroutine/arm32/Context.S b/coroutine/arm32/Context.S
new file mode 100644
index 0000000000..c2b93d0a34
--- /dev/null
+++ b/coroutine/arm32/Context.S
@@ -0,0 +1,14 @@
+##
+##  This file is part of the "Coroutine" project and released under the MIT License.
+##
+##  Created by Samuel Williams on 10/5/2018.
+##  Copyright, 2018, by Samuel Williams. All rights reserved.
+##
+
+.text
+
+.globl coroutine_transfer
+coroutine_transfer:
+	stmia r1!, {r4-r11,sp,lr}
+	ldmia r0!, {r4-r11,sp,pc}
+	bx lr
diff --git a/coroutine/arm32/Context.h b/coroutine/arm32/Context.h
new file mode 100644
index 0000000000..60732df7a0
--- /dev/null
+++ b/coroutine/arm32/Context.h
@@ -0,0 +1,56 @@
+/*
+ *  This file is part of the "Coroutine" project and released under the MIT License.
+ *
+ *  Created by Samuel Williams on 10/5/2018.
+ *  Copyright, 2018, by Samuel Williams. All rights reserved.
+*/
+
+#pragma once
+
+#include <assert.h>
+#include <string.h>
+
+#if __cplusplus
+extern "C" {
+#endif
+
+#define COROUTINE __attribute__((noreturn)) void
+
+const size_t COROUTINE_REGISTERS = 9;
+
+typedef struct
+{
+    void **stack_pointer;
+} coroutine_context;
+
+typedef COROUTINE(* coroutine_start)(coroutine_context *from, coroutine_context *self);
+
+static inline void coroutine_initialize(
+    coroutine_context *context,
+    coroutine_start start,
+    void *stack_pointer,
+    size_t stack_size
+) {
+    context->stack_pointer = (void**)stack_pointer;
+
+    if (!start) {
+        assert(!context->stack_pointer);
+        /* We are main coroutine for this thread */
+        return;
+    }
+
+    *--context->stack_pointer = (void*)start;
+
+    context->stack_pointer -= COROUTINE_REGISTERS;
+    memset(context->stack_pointer, 0, sizeof(void*) * COROUTINE_REGISTERS);
+}
+
+coroutine_context * coroutine_transfer(coroutine_context * current, coroutine_context * target);
+
+static inline void coroutine_destroy(coroutine_context * context)
+{
+}
+
+#if __cplusplus
+}
+#endif
diff --git a/coroutine/arm64/Context.S b/coroutine/arm64/Context.S
new file mode 100644
index 0000000000..f6e5f0a6bc
--- /dev/null
+++ b/coroutine/arm64/Context.S
@@ -0,0 +1,59 @@
+##
+##  This file is part of the "Coroutine" project and released under the MIT License.
+##
+##  Created by Samuel Williams on 10/5/2018.
+##  Copyright, 2018, by Samuel Williams. All rights reserved.
+##
+
+.text
+.align 2
+
+.global coroutine_transfer
+coroutine_transfer:
+
+	# Make space on the stack for caller registers
+	sub sp, sp, 0xb0
+	
+	# Save caller registers
+	stp d8, d9, [sp, 0x00]
+	stp d10, d11, [sp, 0x10]
+	stp d12, d13, [sp, 0x20]
+	stp d14, d15, [sp, 0x30]
+	stp x19, x20, [sp, 0x40]
+	stp x21, x22, [sp, 0x50]
+	stp x23, x24, [sp, 0x60]
+	stp x25, x26, [sp, 0x70]
+	stp x27, x28, [sp, 0x80]
+	stp x29, x30, [sp, 0x90]
+
+	# Save return address
+	str x30, [sp, 0xa0]
+
+	# Save stack pointer to x0 (first argument)
+	mov x2, sp
+	str x2, [x0, 0]
+
+	# Load stack pointer from x1 (second argument)
+	ldr x3, [x1, 0]
+	mov sp, x3
+
+	# Restore caller registers
+	ldp d8, d9, [sp, 0x00]
+	ldp d10, d11, [sp, 0x10]
+	ldp d12, d13, [sp, 0x20]
+	ldp d14, d15, [sp, 0x30]
+	ldp x19, x20, [sp, 0x40]
+	ldp x21, x22, [sp, 0x50]
+	ldp x23, x24, [sp, 0x60]
+	ldp x25, x26, [sp, 0x70]
+	ldp x27, x28, [sp, 0x80]
+	ldp x29, x30, [sp, 0x90]
+
+	# Load return address into x4
+	ldr x4, [sp, 0xa0]
+
+	# Pop stack frame
+	add sp, sp, 0xb0
+
+	# Jump to return address (in x4)
+	ret x4
diff --git a/coroutine/arm64/Context.h b/coroutine/arm64/Context.h
new file mode 100644
index 0000000000..03b91fd937
--- /dev/null
+++ b/coroutine/arm64/Context.h
@@ -0,0 +1,57 @@
+/*
+ *  This file is part of the "Coroutine" project and released under the MIT License.
+ *
+ *  Created by Samuel Williams on 10/5/2018.
+ *  Copyright, 2018, by Samuel Williams. All rights reserved.
+*/
+
+#pragma once
+
+#include <assert.h>
+#include <string.h>
+
+#if __cplusplus
+extern "C" {
+#endif
+
+#define COROUTINE __attribute__((noreturn)) void
+
+const size_t COROUTINE_REGISTERS = 0xb0 / 8;
+
+typedef struct
+{
+    void **stack_pointer;
+} coroutine_context;
+
+typedef COROUTINE(* coroutine_start)(coroutine_context *from, coroutine_context *self);
+
+static inline void coroutine_initialize(
+    coroutine_context *context,
+    coroutine_start start,
+    void *stack_pointer,
+    size_t stack_size
+) {
+    /* Force 16-byte alignment */
+    context->stack_pointer = (void**)((uintptr_t)stack_pointer & ~0xF);
+
+    if (!start) {
+        assert(!context->stack_pointer);
+        /* We are main coroutine for this thread */
+        return;
+    }
+
+    context->stack_pointer -= COROUTINE_REGISTERS;
+    memset(context->stack_pointer, 0, sizeof(void*) * COROUTINE_REGISTERS);
+
+    context->stack_pointer[0xa0 / 8] = (void*)start;
+}
+
+coroutine_context * coroutine_transfer(coroutine_context * current, coroutine_context * target);
+
+static inline void coroutine_destroy(coroutine_context * context)
+{
+}
+
+#if __cplusplus
+}
+#endif
diff --git a/coroutine/ppc64le/Context.S b/coroutine/ppc64le/Context.S
new file mode 100644
index 0000000000..1b39086f8f
--- /dev/null
+++ b/coroutine/ppc64le/Context.S
@@ -0,0 +1,72 @@
+.text
+.align 2
+
+.globl coroutine_transfer
+.type   coroutine_transfer, @function
+coroutine_transfer:
+	# Make space on the stack for caller registers
+	addi 1,1,-152
+
+	# Save caller registers
+	std 14,0(1)
+	std 15,8(1)
+	std 16,16(1)
+	std 17,24(1)
+	std 18,32(1)
+	std 19,40(1)
+	std 20,48(1)
+	std 21,56(1)
+	std 22,64(1)
+	std 23,72(1)
+	std 24,80(1)
+	std 25,88(1)
+	std 26,96(1)
+	std 27,104(1)
+	std 28,112(1)
+	std 29,120(1)
+	std 30,128(1)
+	std 31,136(1)
+
+	# Save return address
+	mflr 0
+	std 0,144(1)
+
+	# Save stack pointer to first argument
+	std 1,0(3)
+
+	# Load stack pointer from second argument
+	ld 1,0(4)
+
+	# Restore caller registers
+	ld 14,0(1)
+	ld 15,8(1)
+	ld 16,16(1)
+	ld 17,24(1)
+	ld 18,32(1)
+	ld 19,40(1)
+	ld 20,48(1)
+	ld 21,56(1)
+	ld 22,64(1)
+	ld 23,72(1)
+	ld 24,80(1)
+	ld 25,88(1)
+	ld 26,96(1)
+	ld 27,104(1)
+	ld 28,112(1)
+	ld 29,120(1)
+	ld 30,128(1)
+	ld 31,136(1)
+
+	# Load return address
+	ld 0,144(1)
+	mtlr 0
+
+	# Pop stack frame
+	addi 1,1,152
+
+	# Jump to return address
+	blr
+
+#if defined(__linux__) && defined(__ELF__)
+.section .note.GNU-stack,"",%progbits
+#endif
diff --git a/coroutine/ppc64le/Context.h b/coroutine/ppc64le/Context.h
new file mode 100644
index 0000000000..de592f5a46
--- /dev/null
+++ b/coroutine/ppc64le/Context.h
@@ -0,0 +1,54 @@
+#pragma once
+
+#include <assert.h>
+#include <string.h>
+
+#if __cplusplus
+extern "C" {
+#endif
+
+#define COROUTINE __attribute__((noreturn)) void
+
+const size_t COROUTINE_REGISTERS =
+  19  /* 18 general purpose registers (r14-r31) and 1 return address */
+  + 4;  /* space for fiber_entry() to store the link register */
+
+typedef struct
+{
+    void **stack_pointer;
+} coroutine_context;
+
+typedef COROUTINE(* coroutine_start)(coroutine_context *from, coroutine_context *self);
+
+static inline void coroutine_initialize(
+    coroutine_context *context,
+    coroutine_start start,
+    void *stack_pointer,
+    size_t stack_size
+) {
+    /* Force 16-byte alignment */
+    context->stack_pointer = (void**)((uintptr_t)stack_pointer & ~0xF);
+
+    if (!start) {
+        assert(!context->stack_pointer);
+        /* We are main coroutine for this thread */
+        return;
+    }
+
+    context->stack_pointer -= COROUTINE_REGISTERS;
+    memset(context->stack_pointer, 0, sizeof(void*) * COROUTINE_REGISTERS);
+
+    /* Skip a global prologue that sets the TOC register */
+    context->stack_pointer[18] = ((char*)start) + 8;
+}
+
+coroutine_context * coroutine_transfer(coroutine_context * current, coroutine_context * target);
+
+static inline void coroutine_destroy(coroutine_context * context)
+{
+    context->stack_pointer = NULL;
+}
+
+#if __cplusplus
+}
+#endif
diff --git a/coroutine/win32/Context.asm b/coroutine/win32/Context.asm
new file mode 100644
index 0000000000..2647ea4bc4
--- /dev/null
+++ b/coroutine/win32/Context.asm
@@ -0,0 +1,55 @@
+;;
+;;  This file is part of the "Coroutine" project and released under the MIT License.
+;;
+;;  Created by Samuel Williams on 10/5/2018.
+;;  Copyright, 2018, by Samuel Williams. All rights reserved.
+;;
+
+.386
+.model flat
+
+.code
+
+assume fs:nothing
+
+; Using fastcall is a big win (and it's the same has how x64 works).
+; In coroutine transfer, the arguments are passed in ecx and edx. We don't need
+; to touch these in order to pass them to the destination coroutine.
+
+@coroutine_transfer@8 proc
+	; Save the thread information block:
+	push fs:[0]
+	push fs:[4]
+	push fs:[8]
+
+	; Save caller registers:
+	push ebp
+	push ebx
+	push edi
+	push esi
+
+	; Save caller stack pointer:
+	mov dword ptr [ecx], esp
+
+	; Restore callee stack pointer:
+	mov esp, dword ptr [edx]
+
+	; Restore callee stack:
+	pop esi
+	pop edi
+	pop ebx
+	pop ebp
+
+	; Restore the thread information block:
+	pop fs:[8]
+	pop fs:[4]
+	pop fs:[0]
+
+	; Save the first argument as the return value:
+	mov eax, dword ptr ecx
+
+	; Jump to the address on the stack:
+	ret
+@coroutine_transfer@8 endp
+
+end
diff --git a/coroutine/win32/Context.h b/coroutine/win32/Context.h
new file mode 100644
index 0000000000..aa9f17ddab
--- /dev/null
+++ b/coroutine/win32/Context.h
@@ -0,0 +1,62 @@
+/*
+ *  This file is part of the "Coroutine" project and released under the MIT License.
+ *
+ *  Created by Samuel Williams on 10/5/2018.
+ *  Copyright, 2018, by Samuel Williams. All rights reserved.
+*/
+
+#pragma once
+
+#include <assert.h>
+#include <string.h>
+
+#if __cplusplus
+extern "C" {
+#endif
+
+#define COROUTINE __declspec(noreturn) void __fastcall
+
+/* This doesn't include thread information block */
+const size_t COROUTINE_REGISTERS = 4;
+
+typedef struct
+{
+    void **stack_pointer;
+} coroutine_context;
+
+typedef void(__fastcall * coroutine_start)(coroutine_context *from, coroutine_context *self);
+
+static inline void coroutine_initialize(
+    coroutine_context *context,
+    coroutine_start start,
+    void *stack_pointer,
+    size_t stack_size
+) {
+    context->stack_pointer = (void**)stack_pointer;
+
+    if (!start) {
+        assert(!context->stack_pointer);
+        /* We are main coroutine for this thread */
+        return;
+    }
+
+    *--context->stack_pointer = (void*)start;
+
+    /* Windows Thread Information Block */
+    *--context->stack_pointer = 0; /* fs:[0] */
+    *--context->stack_pointer = (void*)stack_pointer; /* fs:[4] */
+    *--context->stack_pointer = (void*)((char *)stack_pointer - stack_size);  /* fs:[8] */
+
+    context->stack_pointer -= COROUTINE_REGISTERS;
+    memset(context->stack_pointer, 0, sizeof(void*) * COROUTINE_REGISTERS);
+}
+
+coroutine_context * __fastcall coroutine_transfer(coroutine_context * current, coroutine_context * target);
+
+static inline void coroutine_destroy(coroutine_context * context)
+{
+}
+
+#if __cplusplus
+}
+#endif
diff --git a/coroutine/win64/Context.S b/coroutine/win64/Context.S
new file mode 100644
index 0000000000..4b16e0ce8c
--- /dev/null
+++ b/coroutine/win64/Context.S
@@ -0,0 +1,77 @@
+##
+##  This file is part of the "Coroutine" project and released under the MIT License.
+##
+##  Created by Samuel Williams on 4/11/2018.
+##  Copyright, 2018, by Samuel Williams. All rights reserved.
+##
+
+.text
+
+.globl coroutine_transfer
+coroutine_transfer:
+	# Save the thread information block:
+	pushq %gs:8
+	pushq %gs:16
+
+	# Save caller registers:
+	pushq %rbp
+	pushq %rbx
+	pushq %rdi
+	pushq %rsi
+	pushq %r12
+	pushq %r13
+	pushq %r14
+	pushq %r15
+
+	movaps %xmm15, -168(%rsp)
+	movaps %xmm14, -152(%rsp)
+	movaps %xmm13, -136(%rsp)
+	movaps %xmm12, -120(%rsp)
+	movaps %xmm11, -104(%rsp)
+	movaps %xmm10, -88(%rsp)
+	movaps %xmm9, -72(%rsp)
+	movaps %xmm8, -56(%rsp)
+	movaps %xmm7, -40(%rsp)
+	movaps %xmm6, -24(%rsp)
+
+	# Save caller stack pointer:
+	mov %rsp, (%rcx)
+
+	# Restore callee stack pointer:
+	mov (%rdx), %rsp
+
+	movaps -24(%rsp), %xmm6
+	movaps -40(%rsp), %xmm7
+	movaps -56(%rsp), %xmm8
+	movaps -72(%rsp), %xmm9
+	movaps -88(%rsp), %xmm10
+	movaps -104(%rsp), %xmm11
+	movaps -120(%rsp), %xmm12
+	movaps -136(%rsp), %xmm13
+	movaps -152(%rsp), %xmm14
+	movaps -168(%rsp), %xmm15
+
+	# Restore callee stack:
+	popq %r15
+	popq %r14
+	popq %r13
+	popq %r12
+	popq %rsi
+	popq %rdi
+	popq %rbx
+	popq %rbp
+
+	# Restore the thread information block:
+	popq %gs:16
+	popq %gs:8
+
+	# Put the first argument into the return value:
+	mov %rcx, %rax
+
+	# We pop the return address and jump to it:
+	ret
+
+.globl coroutine_trampoline
+coroutine_trampoline:
+	# Do not remove this. This forces 16-byte alignment when entering the coroutine.
+	ret
diff --git a/coroutine/win64/Context.asm b/coroutine/win64/Context.asm
new file mode 100644
index 0000000000..59673ffa3e
--- /dev/null
+++ b/coroutine/win64/Context.asm
@@ -0,0 +1,79 @@
+;;
+;;  This file is part of the "Coroutine" project and released under the MIT License.
+;;
+;;  Created by Samuel Williams on 10/5/2018.
+;;  Copyright, 2018, by Samuel Williams. All rights reserved.
+;;
+
+.code
+
+coroutine_transfer proc
+	; Save the thread information block:
+	push qword ptr gs:[8]
+	push qword ptr gs:[16]
+
+	; Save caller registers:
+	push rbp
+	push rbx
+	push rdi
+	push rsi
+	push r12
+	push r13
+	push r14
+	push r15
+
+	movaps [rsp - 24], xmm6
+	movaps [rsp - 40], xmm7
+	movaps [rsp - 56], xmm8
+	movaps [rsp - 72], xmm9
+	movaps [rsp - 88], xmm10
+	movaps [rsp - 104], xmm11
+	movaps [rsp - 120], xmm12
+	movaps [rsp - 136], xmm13
+	movaps [rsp - 152], xmm14
+	movaps [rsp - 168], xmm15
+
+	; Save caller stack pointer:
+	mov [rcx], rsp
+
+	; Restore callee stack pointer:
+	mov rsp, [rdx]
+
+	movaps xmm15, [rsp - 168]
+	movaps xmm14, [rsp - 152]
+	movaps xmm13, [rsp - 136]
+	movaps xmm12, [rsp - 120]
+	movaps xmm11, [rsp - 104]
+	movaps xmm10, [rsp - 88]
+	movaps xmm9, [rsp - 72]
+	movaps xmm8, [rsp - 56]
+	movaps xmm7, [rsp - 40]
+	movaps xmm6, [rsp - 24]
+
+	; Restore callee stack:
+	pop r15
+	pop r14
+	pop r13
+	pop r12
+	pop rsi
+	pop rdi
+	pop rbx
+	pop rbp
+
+	; Restore the thread information block:
+	pop qword ptr gs:[16]
+	pop qword ptr gs:[8]
+
+	; Put the first argument into the return value:
+	mov rax, rcx
+
+	; We pop the return address and jump to it:
+	ret
+coroutine_transfer endp
+
+coroutine_trampoline proc
+	; Do not remove this. This forces 16-byte alignment when entering the coroutine.
+	ret
+coroutine_trampoline endp
+
+end
diff --git a/coroutine/win64/Context.h b/coroutine/win64/Context.h
new file mode 100644
index 0000000000..16a8f583ab
--- /dev/null
+++ b/coroutine/win64/Context.h
@@ -0,0 +1,72 @@
+/*
+ *  This file is part of the "Coroutine" project and released under the MIT License.
+ *
+ *  Created by Samuel Williams on 10/5/2018.
+ *  Copyright, 2018, by Samuel Williams. All rights reserved.
+*/
+
+#pragma once
+
+#include <assert.h>
+#include <string.h>
+
+#if __cplusplus
+extern "C" {
+#endif
+
+#define COROUTINE __declspec(noreturn) void
+
+const size_t COROUTINE_REGISTERS = 8;
+const size_t COROUTINE_XMM_REGISTERS = 1+10*2;
+
+typedef struct
+{
+    void **stack_pointer;
+} coroutine_context;
+
+typedef void(* coroutine_start)(coroutine_context *from, coroutine_context *self);
+
+void coroutine_trampoline();
+
+static inline void coroutine_initialize(
+    coroutine_context *context,
+    coroutine_start start,
+    void *stack_pointer,
+    size_t stack_size
+) {
+    /* Force 16-byte alignment */
+    context->stack_pointer = (void**)((uintptr_t)stack_pointer & ~0xF);
+
+    if (!start) {
+        assert(!context->stack_pointer);
+        /* We are main coroutine for this thread */
+        return;
+    }
+
+    /* Win64 ABI requires space for arguments */
+    context->stack_pointer -= 4;
+
+    /* Return address */
+    *--context->stack_pointer = 0;
+    *--context->stack_pointer = (void*)start;
+    *--context->stack_pointer = (void*)coroutine_trampoline;
+
+    /* Windows Thread Information Block */
+    /* *--context->stack_pointer = 0; */ /* gs:[0x00] is not used */
+    *--context->stack_pointer = (void*)stack_pointer; /* gs:[0x08] */
+    *--context->stack_pointer = (void*)((char *)stack_pointer - stack_size);  /* gs:[0x10] */
+
+    context->stack_pointer -= COROUTINE_REGISTERS;
+    memset(context->stack_pointer, 0, sizeof(void*) * COROUTINE_REGISTERS);
+    memset(context->stack_pointer - COROUTINE_XMM_REGISTERS, 0, sizeof(void*) * COROUTINE_XMM_REGISTERS);
+}
+
+coroutine_context * coroutine_transfer(coroutine_context * current, coroutine_context * target);
+
+static inline void coroutine_destroy(coroutine_context * context)
+{
+}
+
+#if __cplusplus
+}
+#endif
diff --git a/coroutine/x86/Context.S b/coroutine/x86/Context.S
new file mode 100644
index 0000000000..d6a0a2def7
--- /dev/null
+++ b/coroutine/x86/Context.S
@@ -0,0 +1,43 @@
+##
+##  This file is part of the "Coroutine" project and released under the MIT License.
+##
+##  Created by Samuel Williams on 3/11/2018.
+##  Copyright, 2018, by Samuel Williams. All rights reserved.
+##
+
+.text
+
+.globl coroutine_transfer
+coroutine_transfer:
+
+# For older linkers
+.globl _coroutine_transfer
+_coroutine_transfer:
+
+	# Save caller registers
+	pushl %ebp
+	pushl %ebx
+	pushl %edi
+	pushl %esi
+
+	# Save caller stack pointer
+	movl %esp, (%ecx)
+
+	# Restore callee stack pointer
+	movl (%edx), %esp
+
+	# Restore callee stack
+	popl %esi
+	popl %edi
+	popl %ebx
+	popl %ebp
+
+	# Save the first argument as the return value
+	movl %ecx, %eax
+
+	# Jump to the address on the stack
+	ret
+
+#if defined(__linux__) && defined(__ELF__)
+.section .note.GNU-stack,"",%progbits
+#endif
diff --git a/coroutine/x86/Context.h b/coroutine/x86/Context.h
new file mode 100644
index 0000000000..b077227a1d
--- /dev/null
+++ b/coroutine/x86/Context.h
@@ -0,0 +1,59 @@
+/*
+ *  This file is part of the "Coroutine" project and released under the MIT License.
+ *
+ *  Created by Samuel Williams on 3/11/2018.
+ *  Copyright, 2018, by Samuel Williams. All rights reserved.
+*/
+
+#pragma once
+
+#include <assert.h>
+#include <string.h>
+
+#if __cplusplus
+extern "C" {
+#endif
+
+#define COROUTINE __attribute__((noreturn, fastcall)) void
+
+const size_t COROUTINE_REGISTERS = 4;
+
+typedef struct
+{
+    void **stack_pointer;
+} coroutine_context;
+
+typedef COROUTINE(* coroutine_start)(coroutine_context *from, coroutine_context *self) __attribute__((fastcall));
+
+static inline void coroutine_initialize(
+    coroutine_context *context,
+    coroutine_start start,
+    void *stack_pointer,
+    size_t stack_size
+) {
+    /* Force 16-byte alignment */
+    context->stack_pointer = (void**)((uintptr_t)stack_pointer & ~0xF);
+
+    if (!start) {
+        assert(!context->stack_pointer);
+        /* We are main coroutine for this thread */
+        return;
+    }
+
+    *--context->stack_pointer = NULL;
+    *--context->stack_pointer = (void*)start;
+
+    context->stack_pointer -= COROUTINE_REGISTERS;
+    memset(context->stack_pointer, 0, sizeof(void*) * COROUTINE_REGISTERS);
+}
+
+coroutine_context * coroutine_transfer(coroutine_context * current, coroutine_context * target) __attribute__((fastcall));
+
+static inline void coroutine_destroy(coroutine_context * context)
+{
+    context->stack_pointer = NULL;
+}
+
+#if __cplusplus
+}
+#endif
diff --git a/coverage/README b/coverage/README
new file mode 100644
index 0000000000..7e4ff59e2a
--- /dev/null
+++ b/coverage/README
@@ -0,0 +1,17 @@
+Usage
+
+The make task `update-coverage' retrieves simplecov for coverage report.
+
+COVERAGE=1 make test-all TESTS=test/cgi
+
+it generate test coverage to coverage directory.
+
+Limitation
+
+ * test_coverage.rb and test_process.rb broke test suit with SimpleCov
+ * some tests failed randomly.
+
+TODO
+
+ * more reduce bundled simplecov(additional configuration, formatter, etc)
+ * measure rubyspec coverage
diff --git a/cygwin/GNUmakefile.in b/cygwin/GNUmakefile.in
index 03208df11c..192cf43b37 100644
--- a/cygwin/GNUmakefile.in
+++ b/cygwin/GNUmakefile.in
@@ -1,14 +1,16 @@
 include Makefile
 
 ENABLE_SHARED=@ENABLE_SHARED@
-DLLWRAP = @DLLWRAP@ --target=@target_os@
+DLLWRAP = @DLLWRAP@ --target=@target_os@ --driver-name="$(CC)"
+WINDRES = @WINDRES@ --preprocessor="$(CPP) -xc" -DRC_INVOKED
+STRIP = @STRIP@
 
 ifeq (@target_os@,cygwin)
-  DLL_BASE_NAME := $(subst .dll,,$(LIBRUBY_SO))
+  DLL_BASE_NAME := $(LIBRUBY_SO:.dll=)
 else
   DLL_BASE_NAME := $(RUBY_SO_NAME)
   DLLWRAP += -mno-cygwin
-  VPATH += $(srcdir)/win32
+  VPATH := $(VPATH):$(srcdir)/win32
 endif
 
 ifneq ($(ENABLE_SHARED),yes)
@@ -25,9 +27,15 @@ else
 endif
 
 WPROGRAM = $(RUBYW_INSTALL_NAME)$(EXEEXT)
+
+-include uncommon.mk
+
+include $(srcdir)/defs/gmake.mk
+
 SOLIBS := $(DLL_BASE_NAME).res.@OBJEXT@ $(SOLIBS)
-EXTOBJS += $(@:$(EXEEXT)=.res.@OBJEXT@)
+override EXTOBJS += $(if $(filter-out $(RUBYW_INSTALL_NAME),$(@:$(EXEEXT)=)),$(RUBY_INSTALL_NAME),$(@:$(EXEEXT)=)).res.$(OBJEXT)
 RCFILES = $(RUBY_INSTALL_NAME).rc $(RUBYW_INSTALL_NAME).rc $(DLL_BASE_NAME).rc
+RUBYDEF = $(DLL_BASE_NAME).def
 
 ruby: $(PROGRAM)
 rubyw: $(WPROGRAM)
@@ -36,43 +44,63 @@ $(LIBRUBY): $(RUBY_EXP) $(LIBRUBY_SO)
 $(RUBY_EXP) $(LIBRUBY_SO): $(DLL_BASE_NAME).res.@OBJEXT@
 
 %.res.@OBJEXT@: %.rc
-	@WINDRES@ --include-dir . --include-dir $(<D) --include-dir $(srcdir)/win32 $< $@
+	$(ECHO) compiling $@
+	$(Q) $(WINDRES) --include-dir . --include-dir $(<D) --include-dir $(srcdir)/win32 $< $@
 
-$(RCFILES): $(RBCONFIG)
-	@$(MINIRUBY) $(srcdir)/win32/resource.rb \
+%.rc: $(RBCONFIG) $(srcdir)/revision.h $(srcdir)/win32/resource.rb
+	$(ECHO) generating $@
+	$(Q) $(MINIRUBY) $(srcdir)/win32/resource.rb \
 	  -ruby_name=$(RUBY_INSTALL_NAME) -rubyw_name=$(RUBYW_INSTALL_NAME) \
-	  -so_name=$(DLL_BASE_NAME) \
+	  -so_name=$(DLL_BASE_NAME) -output=$(*F) \
 	  . $(icondirs) $(srcdir)/win32
 
 $(PROGRAM): $(RUBY_INSTALL_NAME).res.@OBJEXT@
 $(WPROGRAM): $(RUBYW_INSTALL_NAME).res.@OBJEXT@
 	@rm -f $@
-	$(PURIFY) $(CC) -mwindows -e _mainCRTStartup $(LDFLAGS) $(XLDFLAGS) \
+	$(ECHO) linking $@
+	$(Q) $(PURIFY) $(CC) -mwindows -e $(SYMBOL_PREFIX)mainCRTStartup $(LDFLAGS) $(XLDFLAGS) \
 	  $(MAINOBJ) $(EXTOBJS) $(LIBRUBYARG) $(LIBS) -o $@
+$(STUBPROGRAM): $(RUBY_INSTALL_NAME).res.@OBJEXT@
 
 $(RUBY_EXP): $(LIBRUBY_A)
-	$(DLLWRAP) \
+	$(ECHO) creating $@
+	$(Q) $(DLLWRAP) \
 	  --output-exp=$(RUBY_EXP) \
 	  --export-all $(LIBRUBY_A) $(LIBS) -o $(PROGRAM)
-	$(LDSHARED) $(DLDFLAGS) $(OBJS) dmyext.o $(SOLIBS) -o $(PROGRAM)
+	$(Q) $(LDSHARED) $(DLDFLAGS) $(OBJS) dmyext.o $(SOLIBS) -o $(PROGRAM)
 	@rm -f $(PROGRAM)
 
 GNUmakefile:	$(srcdir)/cygwin/GNUmakefile.in
 
 ifeq (@target_os@,mingw32)
-$(OBJS) $(MAINOBJ): win32/win32.h
+$(OBJS) $(MAINOBJ): win32.h
+
+dir.$(OBJEXT) win32/win32.$(OBJEXT): win32/dir.h
+file.$(OBJEXT) win32/win32.$(OBJEXT): win32/file.h
+
+MSYS2_ARG_CONV_EXCL_PARAM = --exclude=;--name=
+
+yes-test-ruby:	 export MSYS2_ARG_CONV_EXCL=$(MSYS2_ARG_CONV_EXCL_PARAM)
+yes-test-all:	 export MSYS2_ARG_CONV_EXCL=$(MSYS2_ARG_CONV_EXCL_PARAM)
+yes-test-almost: export MSYS2_ARG_CONV_EXCL=$(MSYS2_ARG_CONV_EXCL_PARAM)
+
 endif
 
+$(LIBRUBY_SO): $(RUBYDEF)
+
+$(RUBYDEF): $(LIBRUBY_A) $(PREP) $(RBCONFIG)
+	$(ECHO) generating $@
+	$(Q) $(MINIRUBY) $(srcdir)/win32/mkexports.rb -output=$@ $(LIBRUBY_A)
+
+clean-local::
+	@$(RM) $(RUBYDEF)
+
 ifeq (@target_os@,cygwin)
-cygwin-$(RUBY_INSTALL_NAME)$(MAJOR)$(MINOR).dll: $(LIBRUBY_A)
-	@NM@ --extern --defined $(LIBRUBY_A) | \
-	  $(MINIRUBY) -ne 'BEGIN{puts "EXPORTS"}; puts $$1+"=cyg$(RUBY_INSTALL_NAME)$(MAJOR)$(MINOR)."+$$1 if / [CDT] _(.*)$$/' >rubydll.def
-	@DLLWRAP@ -s --def=rubydll.def -o $@
-	@rm -f rubydll.def
+cygwin-$(RUBY_INSTALL_NAME)$(MAJOR)$(MINOR)0.dll: $(LIBRUBY_A) $(RUBYDEF)
+	$(ECHO) generating $@
+	$(Q) @DLLWRAP@ -s --def=$(RUBYDEF) -o $@
 endif
 
 clean-local::
 	@$(RM) $(RUBY_EXP) $(RCFILES:.rc=.res.@OBJEXT@)
-
-distclean-local::
 	@$(RM) $(RCFILES)
diff --git a/debug.c b/debug.c
new file mode 100644
index 0000000000..e4ad163ad1
--- /dev/null
+++ b/debug.c
@@ -0,0 +1,227 @@
+/**********************************************************************
+
+  debug.c -
+
+  $Author$
+  created at: 04/08/25 02:31:54 JST
+
+  Copyright (C) 2004-2007 Koichi Sasada
+
+**********************************************************************/
+
+#include "ruby/ruby.h"
+#include "ruby/encoding.h"
+#include "ruby/io.h"
+#include "ruby/util.h"
+#include "vm_debug.h"
+#include "eval_intern.h"
+#include "vm_core.h"
+#include "symbol.h"
+#include "id.h"
+
+/* for gdb */
+const union {
+    enum ruby_special_consts    special_consts;
+    enum ruby_value_type        value_type;
+    enum ruby_tag_type          tag_type;
+    enum node_type              node_type;
+    enum ruby_method_ids        method_ids;
+    enum ruby_id_types          id_types;
+    enum ruby_fl_type           fl_types;
+    enum ruby_encoding_consts   encoding_consts;
+    enum ruby_coderange_type    enc_coderange_types;
+    enum ruby_econv_flag_type   econv_flag_types;
+    enum ruby_robject_flags     robject_flags;
+    enum ruby_rmodule_flags     rmodule_flags;
+    enum ruby_rstring_flags     rstring_flags;
+    enum ruby_rarray_flags      rarray_flags;
+    enum {
+	RUBY_FMODE_READABLE		= FMODE_READABLE,
+	RUBY_FMODE_WRITABLE		= FMODE_WRITABLE,
+	RUBY_FMODE_READWRITE		= FMODE_READWRITE,
+	RUBY_FMODE_BINMODE		= FMODE_BINMODE,
+	RUBY_FMODE_SYNC 		= FMODE_SYNC,
+	RUBY_FMODE_TTY			= FMODE_TTY,
+	RUBY_FMODE_DUPLEX		= FMODE_DUPLEX,
+	RUBY_FMODE_APPEND		= FMODE_APPEND,
+	RUBY_FMODE_CREATE		= FMODE_CREATE,
+	RUBY_FMODE_NOREVLOOKUP		= 0x00000100,
+	RUBY_FMODE_TRUNC		= FMODE_TRUNC,
+	RUBY_FMODE_TEXTMODE		= FMODE_TEXTMODE,
+	RUBY_FMODE_PREP 		= 0x00010000,
+	RUBY_FMODE_SETENC_BY_BOM	= FMODE_SETENC_BY_BOM,
+	RUBY_FMODE_UNIX 		= 0x00200000,
+	RUBY_FMODE_INET 		= 0x00400000,
+	RUBY_FMODE_INET6		= 0x00800000,
+
+        RUBY_NODE_TYPESHIFT = NODE_TYPESHIFT,
+        RUBY_NODE_TYPEMASK  = NODE_TYPEMASK,
+        RUBY_NODE_LSHIFT    = NODE_LSHIFT,
+        RUBY_NODE_FL_NEWLINE   = NODE_FL_NEWLINE
+    } various;
+    union {
+	enum imemo_type                     types;
+	enum {RUBY_IMEMO_MASK = IMEMO_MASK} mask;
+	struct RIMemo                      *ptr;
+    } imemo;
+    struct RSymbol *symbol_ptr;
+    enum vm_call_flag_bits vm_call_flags;
+} ruby_dummy_gdb_enums;
+
+const SIGNED_VALUE RUBY_NODE_LMASK = NODE_LMASK;
+
+int
+ruby_debug_print_indent(int level, int debug_level, int indent_level)
+{
+    if (level < debug_level) {
+	fprintf(stderr, "%*s", indent_level, "");
+	fflush(stderr);
+	return TRUE;
+    }
+    return FALSE;
+}
+
+void
+ruby_debug_printf(const char *format, ...)
+{
+    va_list ap;
+    va_start(ap, format);
+    vfprintf(stderr, format, ap);
+    va_end(ap);
+}
+
+#include "gc.h"
+
+VALUE
+ruby_debug_print_value(int level, int debug_level, const char *header, VALUE obj)
+{
+    if (level < debug_level) {
+	char buff[0x100];
+	rb_raw_obj_info(buff, 0x100, obj);
+
+	fprintf(stderr, "DBG> %s: %s\n", header, buff);
+	fflush(stderr);
+    }
+    return obj;
+}
+
+void
+ruby_debug_print_v(VALUE v)
+{
+    ruby_debug_print_value(0, 1, "", v);
+}
+
+ID
+ruby_debug_print_id(int level, int debug_level, const char *header, ID id)
+{
+    if (level < debug_level) {
+	fprintf(stderr, "DBG> %s: %s\n", header, rb_id2name(id));
+	fflush(stderr);
+    }
+    return id;
+}
+
+NODE *
+ruby_debug_print_node(int level, int debug_level, const char *header, const NODE *node)
+{
+    if (level < debug_level) {
+	fprintf(stderr, "DBG> %s: %s (%u)\n", header,
+		ruby_node_name(nd_type(node)), nd_line(node));
+    }
+    return (NODE *)node;
+}
+
+void
+ruby_debug_breakpoint(void)
+{
+    /* */
+}
+
+#if defined _WIN32
+# if RUBY_MSVCRT_VERSION >= 80
+extern int ruby_w32_rtc_error;
+# endif
+#endif
+#if defined _WIN32 || defined __CYGWIN__
+#include <windows.h>
+UINT ruby_w32_codepage[2];
+#endif
+extern int ruby_rgengc_debug;
+
+int
+ruby_env_debug_option(const char *str, int len, void *arg)
+{
+    int ov;
+    size_t retlen;
+    unsigned long n;
+#define SET_WHEN(name, var, val) do {	    \
+	if (len == sizeof(name) - 1 &&	    \
+	    strncmp(str, (name), len) == 0) { \
+	    (var) = (val);		    \
+	    return 1;			    \
+	}				    \
+    } while (0)
+#define NAME_MATCH_VALUE(name)				\
+    ((size_t)len >= sizeof(name)-1 &&			\
+     strncmp(str, (name), sizeof(name)-1) == 0 &&	\
+     ((len == sizeof(name)-1 && !(len = 0)) ||		\
+      (str[sizeof(name)-1] == '=' &&			\
+       (str += sizeof(name), len -= sizeof(name), 1))))
+#define SET_UINT(val) do { \
+	n = ruby_scan_digits(str, len, 10, &retlen, &ov); \
+	if (!ov && retlen) { \
+	    val = (unsigned int)n; \
+	} \
+	str += retlen; \
+	len -= retlen; \
+    } while (0)
+#define SET_UINT_LIST(name, vals, num) do { \
+	int i; \
+	for (i = 0; i < (num); ++i) { \
+	    SET_UINT((vals)[i]); \
+	    if (!len || *str != ':') break; \
+	    ++str; \
+	    --len; \
+	} \
+	if (len > 0) { \
+	    fprintf(stderr, "ignored "name" option: `%.*s'\n", len, str); \
+	} \
+    } while (0)
+#define SET_WHEN_UINT(name, vals, num, req) \
+    if (NAME_MATCH_VALUE(name)) SET_UINT_LIST(name, vals, num);
+
+    SET_WHEN("gc_stress", *ruby_initial_gc_stress_ptr, Qtrue);
+    SET_WHEN("core", ruby_enable_coredump, 1);
+    if (NAME_MATCH_VALUE("rgengc")) {
+	if (!len) ruby_rgengc_debug = 1;
+	else SET_UINT_LIST("rgengc", &ruby_rgengc_debug, 1);
+	return 1;
+    }
+#if defined _WIN32
+# if RUBY_MSVCRT_VERSION >= 80
+    SET_WHEN("rtc_error", ruby_w32_rtc_error, 1);
+# endif
+#endif
+#if defined _WIN32 || defined __CYGWIN__
+    if (NAME_MATCH_VALUE("codepage")) {
+	if (!len) fprintf(stderr, "missing codepage argument");
+	else SET_UINT_LIST("codepage", ruby_w32_codepage, numberof(ruby_w32_codepage));
+	return 1;
+    }
+#endif
+    return 0;
+}
+
+static void
+set_debug_option(const char *str, int len, void *arg)
+{
+    if (!ruby_env_debug_option(str, len, arg)) {
+	fprintf(stderr, "unexpected debug option: %.*s\n", len, str);
+    }
+}
+
+void
+ruby_set_debug_option(const char *str)
+{
+    ruby_each_words(str, set_debug_option, 0);
+}
diff --git a/debug_counter.c b/debug_counter.c
new file mode 100644
index 0000000000..8c4004af41
--- /dev/null
+++ b/debug_counter.c
@@ -0,0 +1,55 @@
+/**********************************************************************
+
+  debug_counter.c -
+
+  created at: Tue Feb 21 16:51:18 2017
+
+  Copyright (C) 2017 Koichi Sasada
+
+**********************************************************************/
+
+#include "debug_counter.h"
+#if USE_DEBUG_COUNTER
+#include <stdio.h>
+#include <locale.h>
+#include "internal.h"
+
+static const char *const debug_counter_names[] = {
+    ""
+#define RB_DEBUG_COUNTER(name) #name,
+#include "debug_counter.h"
+#undef RB_DEBUG_COUNTER
+};
+
+size_t rb_debug_counter[numberof(debug_counter_names)];
+
+void
+rb_debug_counter_show_results(const char *msg)
+{
+    const char *env = getenv("RUBY_DEBUG_COUNTER_DISABLE");
+
+    setlocale(LC_NUMERIC, "");
+
+    if (env == NULL || strcmp("1", env) != 0) {
+	int i;
+        fprintf(stderr, "[RUBY_DEBUG_COUNTER]\t%d %s\n", getpid(), msg);
+	for (i=0; i<RB_DEBUG_COUNTER_MAX; i++) {
+            fprintf(stderr, "[RUBY_DEBUG_COUNTER]\t%-30s\t%'14"PRIuSIZE"\n",
+		    debug_counter_names[i],
+		    rb_debug_counter[i]);
+	}
+    }
+}
+
+__attribute__((destructor))
+static void
+debug_counter_show_results_at_exit(void)
+{
+    rb_debug_counter_show_results("normal exit.");
+}
+#else
+void
+rb_debug_counter_show_results(const char *msg)
+{
+}
+#endif /* USE_DEBUG_COUNTER */
diff --git a/debug_counter.h b/debug_counter.h
new file mode 100644
index 0000000000..bd03204af4
--- /dev/null
+++ b/debug_counter.h
@@ -0,0 +1,283 @@
+/**********************************************************************
+
+  debug_counter.h -
+
+  created at: Tue Feb 21 16:51:18 2017
+
+  Copyright (C) 2017 Koichi Sasada
+
+**********************************************************************/
+
+#ifndef USE_DEBUG_COUNTER
+#define USE_DEBUG_COUNTER 0
+#endif
+
+#ifdef RB_DEBUG_COUNTER
+
+/*
+ * method cache (mc) counts.
+ *
+ * * mc_inline_hit/miss: inline mc hit/miss counts (VM send insn)
+ * * mc_global_hit/miss: global method cache hit/miss counts
+ *                       two types: (1) inline cache miss (VM send insn)
+ *                                  (2) called from C (rb_funcall).
+ * * mc_global_state_miss: inline mc miss by global_state miss.
+ * * mc_class_serial_miss:            ... by mc_class_serial_miss
+ * * mc_cme_complement: cme complement counts.
+ * * mc_cme_complement_hit: cme cache hit counts.
+ * * mc_search_super: search_method() call counts.
+ */
+RB_DEBUG_COUNTER(mc_inline_hit)
+RB_DEBUG_COUNTER(mc_inline_miss)
+RB_DEBUG_COUNTER(mc_global_hit)
+RB_DEBUG_COUNTER(mc_global_miss)
+RB_DEBUG_COUNTER(mc_global_state_miss)
+RB_DEBUG_COUNTER(mc_class_serial_miss)
+RB_DEBUG_COUNTER(mc_cme_complement)
+RB_DEBUG_COUNTER(mc_cme_complement_hit)
+RB_DEBUG_COUNTER(mc_search_super)
+
+/*
+ * control frame push counts.
+ *
+ * * frame_push: frame push counts.
+ * * frame_push_*: frame push counts per each type.
+ * * frame_R2R: Ruby frame to Ruby frame
+ * * frame_R2C: Ruby frame to C frame
+ * * frame_C2C: C frame to C frame
+ * * frame_C2R: C frame to Ruby frame
+ */
+RB_DEBUG_COUNTER(frame_push)
+RB_DEBUG_COUNTER(frame_push_method)
+RB_DEBUG_COUNTER(frame_push_block)
+RB_DEBUG_COUNTER(frame_push_class)
+RB_DEBUG_COUNTER(frame_push_top)
+RB_DEBUG_COUNTER(frame_push_cfunc)
+RB_DEBUG_COUNTER(frame_push_ifunc)
+RB_DEBUG_COUNTER(frame_push_eval)
+RB_DEBUG_COUNTER(frame_push_rescue)
+RB_DEBUG_COUNTER(frame_push_dummy)
+
+RB_DEBUG_COUNTER(frame_R2R)
+RB_DEBUG_COUNTER(frame_R2C)
+RB_DEBUG_COUNTER(frame_C2C)
+RB_DEBUG_COUNTER(frame_C2R)
+
+/* instance variable counts
+ *
+ * * ivar_get_ic_hit/miss: ivar_get inline cache (ic) hit/miss counts (VM insn)
+ * * ivar_get_ic_miss_serial: ivar_get ic miss reason by serial (VM insn)
+ * * ivar_get_ic_miss_unset:                      ... by unset (VM insn)
+ * * ivar_get_ic_miss_noobject:                   ... by "not T_OBJECT" (VM insn)
+ * * ivar_set_...: same counts with ivar_set (VM insn)
+ * * ivar_get/set_base: call counts of "rb_ivar_get/set()".
+ *                      because of (1) ic miss.
+ *                                 (2) direct call by C extensions.
+ */
+RB_DEBUG_COUNTER(ivar_get_ic_hit)
+RB_DEBUG_COUNTER(ivar_get_ic_miss)
+RB_DEBUG_COUNTER(ivar_get_ic_miss_serial)
+RB_DEBUG_COUNTER(ivar_get_ic_miss_unset)
+RB_DEBUG_COUNTER(ivar_get_ic_miss_noobject)
+RB_DEBUG_COUNTER(ivar_set_ic_hit)
+RB_DEBUG_COUNTER(ivar_set_ic_miss)
+RB_DEBUG_COUNTER(ivar_set_ic_miss_serial)
+RB_DEBUG_COUNTER(ivar_set_ic_miss_unset)
+RB_DEBUG_COUNTER(ivar_set_ic_miss_oorange)
+RB_DEBUG_COUNTER(ivar_set_ic_miss_noobject)
+RB_DEBUG_COUNTER(ivar_get_base)
+RB_DEBUG_COUNTER(ivar_set_base)
+
+/* local variable counts
+ *
+ * * lvar_get: total lvar get counts (VM insn)
+ * * lvar_get_dynamic: lvar get counts if accessing upper env (VM insn)
+ * * lvar_set*: same as "get"
+ * * lvar_set_slowpath: counts using vm_env_write_slowpath()
+ */
+RB_DEBUG_COUNTER(lvar_get)
+RB_DEBUG_COUNTER(lvar_get_dynamic)
+RB_DEBUG_COUNTER(lvar_set)
+RB_DEBUG_COUNTER(lvar_set_dynamic)
+RB_DEBUG_COUNTER(lvar_set_slowpath)
+
+/* GC counts:
+ *
+ * * count: simple count
+ * * _minor: minor gc
+ * * _major: major gc
+ * * other suffix is corresponding to last_gc_info or
+ *   gc_profile_record_flag in gc.c.
+ */
+RB_DEBUG_COUNTER(gc_count)
+RB_DEBUG_COUNTER(gc_minor_newobj)
+RB_DEBUG_COUNTER(gc_minor_malloc)
+RB_DEBUG_COUNTER(gc_minor_method)
+RB_DEBUG_COUNTER(gc_minor_capi)
+RB_DEBUG_COUNTER(gc_minor_stress)
+RB_DEBUG_COUNTER(gc_major_nofree)
+RB_DEBUG_COUNTER(gc_major_oldgen)
+RB_DEBUG_COUNTER(gc_major_shady)
+RB_DEBUG_COUNTER(gc_major_force)
+RB_DEBUG_COUNTER(gc_major_oldmalloc)
+
+/* object allocation counts:
+ *
+ * * obj_newobj: newobj counts
+ * * obj_newobj_slowpath: newobj with slowpath counts
+ * * obj_newobj_wb_unprotected: newobj for wb_unprotecte.
+ * * obj_free: obj_free() counts
+ * * obj_promote: promoted counts (oldgen)
+ * * obj_wb_unprotect: wb unprotect counts
+ *
+ * * obj_[type]_[attr]: *free'ed counts* for each type.
+ *                      Note that it is not a allocated counts.
+ * * [type]
+ *   * _obj: T_OBJECT
+ *   * _str: T_STRING
+ *   * _ary: T_ARRAY
+ *   * _xxx: T_XXX (hash, struct, ...)
+ *
+ * * [attr]
+ *   * _ptr: R?? is not embed.
+ *   * _embed: R?? is embed.
+ *   * _transient: R?? uses transient heap.
+ * * type specific attr.
+ *   * str_shared: str is shared.
+ *   * str_nofree:        nofree
+ *   * str_fstr:          fstr
+ *   * hash_empty: hash is empty
+ *   * hash_under4:     has under 4 entries
+ *   * hash_ge4:        has n entries (4<=n<8)
+ *   * hash_ge8:        has n entries (8<=n)
+ *   * data_empty: T_DATA but no memory free.
+ *   * data_xfree:        free'ed by xfree().
+ *   * data_imm_free:     free'ed immediately.
+ *   * data_zombie:       free'ed with zombie.
+ *   * imemo_*: T_IMEMO with each type.
+ */
+RB_DEBUG_COUNTER(obj_newobj)
+RB_DEBUG_COUNTER(obj_newobj_slowpath)
+RB_DEBUG_COUNTER(obj_newobj_wb_unprotected)
+RB_DEBUG_COUNTER(obj_free)
+RB_DEBUG_COUNTER(obj_promote)
+RB_DEBUG_COUNTER(obj_wb_unprotect)
+
+RB_DEBUG_COUNTER(obj_obj_embed)
+RB_DEBUG_COUNTER(obj_obj_transient)
+RB_DEBUG_COUNTER(obj_obj_ptr)
+
+RB_DEBUG_COUNTER(obj_str_ptr)
+RB_DEBUG_COUNTER(obj_str_embed)
+RB_DEBUG_COUNTER(obj_str_shared)
+RB_DEBUG_COUNTER(obj_str_nofree)
+RB_DEBUG_COUNTER(obj_str_fstr)
+
+RB_DEBUG_COUNTER(obj_ary_embed)
+RB_DEBUG_COUNTER(obj_ary_transient)
+RB_DEBUG_COUNTER(obj_ary_ptr)
+
+RB_DEBUG_COUNTER(obj_hash_empty)
+RB_DEBUG_COUNTER(obj_hash_under4)
+RB_DEBUG_COUNTER(obj_hash_ge4)
+RB_DEBUG_COUNTER(obj_hash_ge8)
+RB_DEBUG_COUNTER(obj_hash_ar)
+RB_DEBUG_COUNTER(obj_hash_st)
+RB_DEBUG_COUNTER(obj_hash_transient)
+
+RB_DEBUG_COUNTER(obj_hash_force_convert)
+
+RB_DEBUG_COUNTER(obj_struct_embed)
+RB_DEBUG_COUNTER(obj_struct_transient)
+RB_DEBUG_COUNTER(obj_struct_ptr)
+
+RB_DEBUG_COUNTER(obj_regexp_ptr)
+
+RB_DEBUG_COUNTER(obj_data_empty)
+RB_DEBUG_COUNTER(obj_data_xfree)
+RB_DEBUG_COUNTER(obj_data_imm_free)
+RB_DEBUG_COUNTER(obj_data_zombie)
+
+RB_DEBUG_COUNTER(obj_match_ptr)
+RB_DEBUG_COUNTER(obj_file_ptr)
+RB_DEBUG_COUNTER(obj_bignum_ptr)
+
+RB_DEBUG_COUNTER(obj_symbol)
+
+RB_DEBUG_COUNTER(obj_imemo_ment)
+RB_DEBUG_COUNTER(obj_imemo_iseq)
+RB_DEBUG_COUNTER(obj_imemo_env)
+RB_DEBUG_COUNTER(obj_imemo_tmpbuf)
+RB_DEBUG_COUNTER(obj_imemo_ast)
+RB_DEBUG_COUNTER(obj_imemo_cref)
+RB_DEBUG_COUNTER(obj_imemo_svar)
+RB_DEBUG_COUNTER(obj_imemo_throw_data)
+RB_DEBUG_COUNTER(obj_imemo_ifunc)
+RB_DEBUG_COUNTER(obj_imemo_memo)
+RB_DEBUG_COUNTER(obj_imemo_parser_strterm)
+
+RB_DEBUG_COUNTER(obj_iclass_ptr)
+RB_DEBUG_COUNTER(obj_class_ptr)
+RB_DEBUG_COUNTER(obj_module_ptr)
+
+/* heap function counts
+ *
+ * * heap_xmalloc/realloc/xfree: call counts
+ */
+RB_DEBUG_COUNTER(heap_xmalloc)
+RB_DEBUG_COUNTER(heap_xrealloc)
+RB_DEBUG_COUNTER(heap_xfree)
+
+/* transient_heap */
+RB_DEBUG_COUNTER(theap_alloc)
+RB_DEBUG_COUNTER(theap_alloc_fail)
+RB_DEBUG_COUNTER(theap_evacuate)
+
+/* load (not implemented yet) */
+/*
+RB_DEBUG_COUNTER(load_files)
+RB_DEBUG_COUNTER(load_path_is_not_realpath)
+*/
+#endif
+
+#ifndef RUBY_DEBUG_COUNTER_H
+#define RUBY_DEBUG_COUNTER_H 1
+
+#if !defined(__GNUC__) && USE_DEBUG_COUNTER
+#error "USE_DEBUG_COUNTER is not supported by other than __GNUC__"
+#endif
+
+enum rb_debug_counter_type {
+#define RB_DEBUG_COUNTER(name) RB_DEBUG_COUNTER_##name,
+#include __FILE__
+    RB_DEBUG_COUNTER_MAX
+#undef RB_DEBUG_COUNTER
+};
+
+#if USE_DEBUG_COUNTER
+#include "ruby/ruby.h"
+
+extern size_t rb_debug_counter[];
+
+inline static int
+rb_debug_counter_add(enum rb_debug_counter_type type, int add, int cond)
+{
+    if (cond) {
+	rb_debug_counter[(int)type] += add;
+    }
+    return cond;
+}
+
+#define RB_DEBUG_COUNTER_INC(type)                rb_debug_counter_add(RB_DEBUG_COUNTER_##type, 1, 1)
+#define RB_DEBUG_COUNTER_INC_UNLESS(type, cond) (!rb_debug_counter_add(RB_DEBUG_COUNTER_##type, 1, !(cond)))
+#define RB_DEBUG_COUNTER_INC_IF(type, cond)       rb_debug_counter_add(RB_DEBUG_COUNTER_##type, 1, (cond))
+
+#else
+#define RB_DEBUG_COUNTER_INC(type)              ((void)0)
+#define RB_DEBUG_COUNTER_INC_UNLESS(type, cond) (cond)
+#define RB_DEBUG_COUNTER_INC_IF(type, cond)     (cond)
+#endif
+
+void rb_debug_counter_show_results(const char *msg);
+
+#endif /* RUBY_DEBUG_COUNTER_H */
diff --git a/defines.h b/defines.h
deleted file mode 100644
index ef80989925..0000000000
--- a/defines.h
+++ /dev/null
@@ -1,261 +0,0 @@
-/************************************************
-
-  defines.h -
-
-  $Author: nobu $
-  $Date: 2005/10/25 16:59:57 $
-  created at: Wed May 18 00:21:44 JST 1994
-
-************************************************/
-#ifndef DEFINES_H
-#define DEFINES_H
-
-#define RUBY
-
-#ifdef __cplusplus
-# ifndef  HAVE_PROTOTYPES
-#  define HAVE_PROTOTYPES 1
-# endif
-# ifndef  HAVE_STDARG_PROTOTYPES
-#  define HAVE_STDARG_PROTOTYPES 1
-# endif
-#endif
-
-#undef _
-#ifdef HAVE_PROTOTYPES
-# define _(args) args
-#else
-# define _(args) ()
-#endif
-
-#undef __
-#ifdef HAVE_STDARG_PROTOTYPES
-# define __(args) args
-#else
-# define __(args) ()
-#endif
-
-#ifdef __cplusplus
-#define ANYARGS ...
-#else
-#define ANYARGS
-#endif
-
-#define xmalloc ruby_xmalloc
-#define xcalloc ruby_xcalloc
-#define xrealloc ruby_xrealloc
-#define xfree ruby_xfree
-
-void *xmalloc _((long));
-void *xcalloc _((long,long));
-void *xrealloc _((void*,long));
-void xfree _((void*));
-
-#if SIZEOF_LONG_LONG > 0
-# define LONG_LONG long long
-#elif SIZEOF___INT64 > 0
-# define HAVE_LONG_LONG 1
-# define LONG_LONG __int64
-# undef SIZEOF_LONG_LONG
-# define SIZEOF_LONG_LONG SIZEOF___INT64
-#endif
-
-#if SIZEOF_INT*2 <= SIZEOF_LONG_LONG
-# define BDIGIT unsigned int
-# define SIZEOF_BDIGITS SIZEOF_INT
-# define BDIGIT_DBL unsigned LONG_LONG
-# define BDIGIT_DBL_SIGNED LONG_LONG
-#elif SIZEOF_INT*2 <= SIZEOF_LONG
-# define BDIGIT unsigned int
-# define SIZEOF_BDIGITS SIZEOF_INT
-# define BDIGIT_DBL unsigned long
-# define BDIGIT_DBL_SIGNED long
-#elif SIZEOF_SHORT*2 <= SIZEOF_LONG
-# define BDIGIT unsigned short
-# define SIZEOF_BDIGITS SIZEOF_SHORT
-# define BDIGIT_DBL unsigned long
-# define BDIGIT_DBL_SIGNED long
-#else
-# define BDIGIT unsigned short
-# define SIZEOF_BDIGITS (SIZEOF_LONG/2)
-# define BDIGIT_DBL unsigned long
-# define BDIGIT_DBL_SIGNED long
-#endif
-
-#ifdef __CYGWIN__
-#undef _WIN32
-#endif
-
-#if defined(MSDOS) || defined(_WIN32) || defined(__human68k__) || defined(__EMX__)
-#define DOSISH 1
-#ifndef _WIN32_WCE
-# define DOSISH_DRIVE_LETTER
-#endif
-#endif
-
-/* define RUBY_USE_EUC/SJIS for default kanji-code */
-#ifndef DEFAULT_KCODE
-#if defined(DOSISH) || defined(__CYGWIN__) || defined(__MACOS__) || defined(OS2)
-#define DEFAULT_KCODE KCODE_SJIS
-#else
-#define DEFAULT_KCODE KCODE_EUC
-#endif
-#endif
-
-#ifdef __NeXT__
-/* NextStep, OpenStep, Rhapsody */
-#ifndef S_IRUSR
-#define S_IRUSR 0000400        /* read permission, owner */
-#endif
-#ifndef S_IRGRP
-#define S_IRGRP 0000040        /* read permission, group */
-#endif
-#ifndef S_IROTH
-#define S_IROTH 0000004        /* read permission, other */
-#endif
-#ifndef S_IWUSR
-#define S_IWUSR 0000200        /* write permission, owner */
-#endif
-#ifndef S_IWGRP
-#define S_IWGRP 0000020        /* write permission, group */
-#endif
-#ifndef S_IWOTH
-#define S_IWOTH 0000002        /* write permission, other */
-#endif
-#ifndef S_IXUSR
-#define S_IXUSR 0000100        /* execute/search permission, owner */
-#endif
-#ifndef S_IXGRP
-#define S_IXGRP 0000010        /* execute/search permission, group */
-#endif
-#ifndef S_IXOTH
-#define S_IXOTH 0000001        /* execute/search permission, other */
-#endif
-#ifndef S_IRWXU
-#define S_IRWXU 0000700        /* read, write, execute permissions, owner */
-#endif
-#ifndef S_IRWXG
-#define S_IRWXG 0000070        /* read, write, execute permissions, group */
-#endif
-#ifndef S_IRWXO
-#define S_IRWXO 0000007        /* read, write, execute permissions, other */
-#endif
-#ifndef S_ISBLK
-#define S_ISBLK(mode)  (((mode) & (0170000)) == (0060000))
-#endif
-#ifndef S_ISCHR
-#define S_ISCHR(mode)  (((mode) & (0170000)) == (0020000))
-#endif
-#ifndef S_ISDIR
-#define S_ISDIR(mode)  (((mode) & (0170000)) == (0040000))
-#endif
-#ifndef S_ISFIFO
-#define S_ISFIFO(mode) (((mode) & (0170000)) == (0010000))
-#endif
-#ifndef S_ISREG
-#define S_ISREG(mode)  (((mode) & (0170000)) == (0100000))
-#endif
-/* Do not trust WORDS_BIGENDIAN from configure since -arch compiler flag may
-   result in a different endian.  Instead trust __BIG_ENDIAN__ and
-   __LITTLE_ENDIAN__ which are set correctly by -arch. */
-#undef WORDS_BIGENDIAN
-#ifdef __BIG_ENDIAN__
-#define WORDS_BIGENDIAN
-#endif
-#ifndef __APPLE__
-/* NextStep, OpenStep (but not Rhapsody) */
-#ifndef GETPGRP_VOID
-#define GETPGRP_VOID 1
-#endif
-#ifndef WNOHANG
-#define WNOHANG 01
-#endif
-#ifndef WUNTRACED
-#define WUNTRACED 02
-#endif
-#ifndef X_OK
-#define X_OK 1
-#endif
-#endif /* __APPLE__ */
-#endif /* NeXT */
-
-#ifdef _WIN32
-#include "win32/win32.h"
-#endif
-
-#if defined(__VMS)
-#include "vms.h"
-#endif
-
-#if defined(__BEOS__)
-#include <net/socket.h> /* intern.h needs fd_set definition */
-#endif
-
-#ifdef RUBY_EXPORT
-#undef RUBY_EXTERN
-#endif
-
-#ifndef RUBY_EXTERN
-#define RUBY_EXTERN extern
-#endif
-
-#ifndef EXTERN
-#define EXTERN RUBY_EXTERN	/* deprecated */
-#endif
-
-#ifndef RUBY_MBCHAR_MAXSIZE
-#define RUBY_MBCHAR_MAXSIZE INT_MAX
-        /* MB_CUR_MAX will not work well in C locale */
-#endif
-
-#if defined(sparc) || defined(__sparc__)
-static inline void
-flush_register_windows(void)
-{
-    asm
-#ifdef __GNUC__
-	volatile
-#endif
-# if defined(__sparc_v9__) || defined(__sparcv9) || defined(__arch64__)
-	("flushw")
-# else
-	("ta  0x03")
-# endif /* trap always to flush register windows if we are on a Sparc system */
-	;
-}
-#  define FLUSH_REGISTER_WINDOWS flush_register_windows()
-#elif defined(__ia64__)
-void flush_register_windows(void)
-#  if defined(__GNUC__) && (( __GNUC__ == 3 && __GNUC_MINOR__ > 0 ) || __GNUC__ > 3)
-__attribute__ ((noinline))
-#  endif
-  ;
-#  define FLUSH_REGISTER_WINDOWS flush_register_windows()
-#else
-#  define FLUSH_REGISTER_WINDOWS ((void)0)
-#endif
-
-#if defined(DOSISH)
-#define PATH_SEP ";"
-#elif defined(riscos)
-#define PATH_SEP ","
-#else
-#define PATH_SEP ":"
-#endif
-#define PATH_SEP_CHAR PATH_SEP[0]
-
-#if defined(__human68k__)
-#define PATH_ENV "path"
-#else
-#define PATH_ENV "PATH"
-#endif
-
-#if defined(DOSISH) && !defined(__human68k__) && !defined(__EMX__)
-#define ENV_IGNORECASE
-#endif
-
-#ifndef RUBY_PLATFORM
-#define RUBY_PLATFORM "unknown-unknown"
-#endif
-
-#endif
diff --git a/defs/gmake.mk b/defs/gmake.mk
new file mode 100644
index 0000000000..72955c5ce1
--- /dev/null
+++ b/defs/gmake.mk
@@ -0,0 +1,205 @@
+# -*- makefile-gmake -*-
+gnumake = yes
+override gnumake_recursive := $(if $(findstring n,$(firstword $(MFLAGS))),,+)
+override mflags := $(filter-out -j%,$(MFLAGS))
+MSPECOPT += $(if $(filter -j%,$(MFLAGS)),-j)
+
+CHECK_TARGETS := great exam love check test check% test% btest%
+# expand test targets, and those dependents
+TEST_TARGETS := $(filter $(CHECK_TARGETS),$(MAKECMDGOALS))
+TEST_DEPENDS := $(filter-out commit $(TEST_TARGETS),$(MAKECMDGOALS))
+TEST_TARGETS := $(patsubst great,exam,$(TEST_TARGETS))
+TEST_DEPENDS := $(filter-out great $(TEST_TARGETS),$(TEST_DEPENDS))
+TEST_TARGETS := $(patsubst exam,check,$(TEST_TARGETS))
+TEST_TARGETS := $(patsubst check,test-spec test-all,$(TEST_TARGETS))
+TEST_TARGETS := $(patsubst test-rubyspec,test-spec,$(TEST_TARGETS))
+TEST_DEPENDS := $(filter-out exam check test-spec $(TEST_TARGETS),$(TEST_DEPENDS))
+TEST_TARGETS := $(patsubst love,check,$(TEST_TARGETS))
+TEST_DEPENDS := $(filter-out love $(TEST_TARGETS),$(TEST_DEPENDS))
+TEST_TARGETS := $(patsubst test-all,test test-testframework test-almost,$(patsubst check-%,test test-%,$(TEST_TARGETS)))
+TEST_DEPENDS := $(filter-out test-all $(TEST_TARGETS),$(TEST_DEPENDS))
+TEST_TARGETS := $(patsubst test,test-short,$(TEST_TARGETS))
+TEST_DEPENDS := $(filter-out test $(TEST_TARGETS),$(TEST_DEPENDS))
+TEST_TARGETS := $(patsubst test-short,btest-ruby test-knownbug test-basic,$(TEST_TARGETS))
+TEST_DEPENDS := $(filter-out test-short $(TEST_TARGETS),$(TEST_DEPENDS))
+TEST_DEPENDS += $(if $(filter great exam love check,$(MAKECMDGOALS)),all exts)
+
+ifneq ($(filter -O0 -Od,$(optflags)),)
+override XCFLAGS := $(filter-out -D_FORTIFY_SOURCE=%,$(XCFLAGS))
+endif
+
+ifeq ($(if $(filter all main exts enc trans libencs libenc libtrans \
+		    prog program ruby ruby$(EXEEXT) \
+		    wprogram rubyw rubyw$(EXEEXT) \
+		    miniruby$(EXEEXT) mini,\
+	     $(MAKECMDGOALS)),,$(MAKECMDGOALS)),)
+-include $(SHOWFLAGS)
+endif
+
+ifneq ($(filter universal-%,$(arch)),)
+define archcmd
+%.$(1).S: %.c
+	@$$(ECHO) translating $$< with $(2)
+	$$(Q) $$(CC) $$(CFLAGS_NO_ARCH) $(2) $$(XCFLAGS) $$(CPPFLAGS) $$(COUTFLAG)$$@ -S $$<
+
+%.S: %.$(1).S
+
+%.$(1).i: %.c
+	@$$(ECHO) preprocessing $$< with $(2)
+	$$(Q) $$(CPP) $$(warnflags) $(2) $$(XCFLAGS) $$(CPPFLAGS) $$(COUTFLAG)$$@ -E $$< > $$@
+
+%.i: %.$(1).i
+endef
+
+$(foreach arch,$(arch_flags),\
+	$(eval $(call archcmd,$(patsubst -arch=%,%,$(value arch)),$(patsubst -arch=%,-arch %,$(value arch)))))
+endif
+
+.PHONY: $(addprefix yes-,$(TEST_TARGETS))
+
+ifneq ($(filter-out btest%,$(TEST_TARGETS)),)
+$(addprefix yes-,$(TEST_TARGETS)): $(TEST_DEPENDS)
+endif
+
+ORDERED_TEST_TARGETS := $(filter $(TEST_TARGETS), \
+	btest-ruby test-knownbug test-basic \
+	test-testframework test-ruby test-almost test-all \
+	test-spec test-bundler-prepare test-bundler \
+	)
+prev_test := $(if $(filter test-spec,$(ORDERED_TEST_TARGETS)),test-spec-precheck)
+$(foreach test,$(ORDERED_TEST_TARGETS), \
+	$(eval yes-$(value test) no-$(value test): $(value prev_test)); \
+	$(eval prev_test := $(value test)))
+
+ifneq ($(if $(filter install,$(MAKECMDGOALS)),$(filter uninstall,$(MAKECMDGOALS))),)
+install-targets := $(filter install uninstall,$(MAKECMDGOALS))
+$(word 1,$(install-targets)): $(word 0,$(install-targets))
+endif
+
+ifneq ($(filter reinstall,$(MAKECMDGOALS)),)
+install-prereq: uninstall
+uninstall sudo-precheck: all $(if $(filter all,$(INSTALLDOC)),docs)
+endif
+
+ifneq ($(filter love,$(MAKECMDGOALS)),)
+showflags: up
+sudo-precheck: test yes-test-testframework no-test-testframework
+install-prereq: sudo-precheck
+yes-test-all no-test-all: install
+yes-test-almost no-test-almost: install
+endif
+ifneq ($(filter great,$(MAKECMDGOALS)),)
+love: test-rubyspec
+endif
+
+$(srcdir)/missing/des_tables.c: $(srcdir)/missing/crypt.c
+ifeq ($(if $(filter yes,$(CROSS_COMPILING)),,$(CC)),)
+	touch $@
+else
+	@$(ECHO) building make_des_table
+	$(CC) $(CPPFLAGS) -DDUMP $(LDFLAGS) $(XLDFLAGS) $(LIBS) -omake_des_table $(srcdir)/missing/crypt.c
+	@[ -x ./make_des_table ]
+	@$(ECHO) generating $@
+	$(Q) $(MAKEDIRS) $(@D)
+	$(Q) ./make_des_table > $@.new
+	$(Q) mv $@.new $@
+	$(Q) $(RMALL) make_des_table*
+endif
+
+STUBPROGRAM = rubystub$(EXEEXT)
+IGNOREDPATTERNS = %~ .% %.orig %.rej \#%\#
+SCRIPTBINDIR := $(if $(EXEEXT),,exec/)
+SCRIPTPROGRAMS = $(addprefix $(SCRIPTBINDIR),$(addsuffix $(EXEEXT),$(filter-out $(IGNOREDPATTERNS),$(notdir $(wildcard $(srcdir)/libexec/*)))))
+
+stub: $(STUBPROGRAM)
+scriptbin: $(SCRIPTPROGRAMS)
+ifneq ($(STUBPROGRAM),rubystub)
+rubystub: $(STUBPROGRAM)
+endif
+
+$(SCRIPTPROGRAMS): $(STUBPROGRAM)
+
+$(STUBPROGRAM): rubystub.$(OBJEXT) $(LIBRUBY) $(MAINOBJ) $(OBJS) $(EXTOBJS) $(SETUP) $(PREP)
+
+rubystub$(EXEEXT):
+	@rm -f $@
+	$(ECHO) linking $@
+	$(Q) $(PURIFY) $(CC) $(LDFLAGS) $(XLDFLAGS) rubystub.$(OBJEXT) $(EXTOBJS) $(LIBRUBYARG) $(MAINLIBS) $(LIBS) $(EXTLIBS) $(OUTFLAG)$@
+	$(Q) $(POSTLINK)
+	$(if $(STRIP),$(Q) $(STRIP) $@)
+
+$(SCRIPTBINDIR)%$(EXEEXT): bin/% $(STUBPROGRAM) \
+			   $(if $(SCRIPTBINDIR),$(TIMESTAMPDIR)/.exec.time)
+	$(ECHO) generating $@
+	$(Q) { cat $(STUBPROGRAM); echo; sed -e '1{' -e '/^#!.*ruby/!i\' -e '#!/bin/ruby' -e '}' $<; } > $@
+	$(Q) chmod +x $@
+	$(Q) $(POSTLINK)
+
+$(TIMESTAMPDIR)/.exec.time:
+	$(Q) mkdir exec
+	$(Q) exit > $@
+
+.PHONY: commit
+commit: $(if $(filter commit,$(MAKECMDGOALS)),$(filter-out commit,$(MAKECMDGOALS)))
+	@$(BASERUBY) -C "$(srcdir)" -I./tool -rvcs -e 'VCS.detect(".").commit'
+	+$(Q) \
+	{ \
+	  cd "$(srcdir)"; \
+	  sed 's/^@.*@$$//;s/@[A-Za-z_][A-Za-z_0-9]*@//g;/^all-incs:/d' defs/gmake.mk Makefile.in; \
+	  sed 's/{[.;]*$$([a-zA-Z0-9_]*)}//g' common.mk; \
+	} | \
+	$(MAKE) $(mflags) Q=$(Q) srcdir="$(srcdir)" srcs_vpath="$(srcdir)/" CHDIR="$(CHDIR)" \
+		BOOTSTRAPRUBY="$(BOOTSTRAPRUBY)" MINIRUBY="$(BASERUBY)" BASERUBY="$(BASERUBY)" \
+		VCSUP="" ENC_MK=.top-enc.mk REVISION_FORCE=PHONY CONFIGURE="$(CONFIGURE)" -f - \
+		update-src srcs all-incs
+
+ifeq ($(words $(filter update-gems extract-gems,$(MAKECMDGOALS))),2)
+extract-gems: update-gems
+endif
+
+ifeq ($(filter 0 1,$(words $(arch_flags))),)
+$(foreach x,$(patsubst -arch=%,%,$(arch_flags)), \
+	  $(eval $$(MJIT_HEADER:.h=)-$(value x).h \
+		 $$(MJIT_MIN_HEADER:.h=)-$(value x).h \
+		 $$(TIMESTAMPDIR)/$$(MJIT_HEADER:.h=)-$(value x).time \
+		 : ARCH_FLAG := -arch $(value x)))
+
+$(foreach x,$(patsubst -arch=%,%,$(arch_flags)), \
+	$(eval $$(MJIT_HEADER:.h=)-$(value x).h: \
+		$$(TIMESTAMPDIR)/$$(MJIT_HEADER:.h=)-$(value x).time))
+
+mjit_min_headers := $(patsubst -arch=%,$(MJIT_MIN_HEADER:.h=-%.h),$(arch_flags))
+$(MJIT_MIN_HEADER): $(mjit_min_headers) $(PREP)
+	@ set -e; set $(patsubst -arch=%,%,$(arch_flags)); \
+	cd $(@D); h=$(@F:.h=); \
+	exec > $(@F).new; \
+	echo '#if 0'; \
+	for arch; do\
+	  echo "#elif defined __$${arch}__"; \
+	  echo "# include \"$$h-$$arch.h\""; \
+	done; \
+	echo "#else"; echo "# error unsupported platform"; echo "#endif"
+	$(IFCHANGE) $@ $@.new
+	$(Q) $(MAKEDIRS) $(MJIT_HEADER_INSTALL_DIR)
+	$(Q) $(MAKE_LINK) $@ $(MJIT_HEADER_INSTALL_DIR)/$(@F)
+
+endif
+
+# GNU make treat the target as unmodified when its dependents get
+# updated but it is not updated, while others may not.
+$(srcdir)/revision.h: $(REVISION_H)
+
+# Query on the generated rdoc
+#
+#   $ make rdoc:Integer#+
+rdoc\:%: PHONY
+	$(Q)$(RUNRUBY) $(srcdir)/libexec/ri --no-standard-docs --doc-dir=$(RDOCOUT) $(patsubst rdoc:%,%,$@)
+
+test_%.rb test/%: programs PHONY
+	+$(Q)$(exec) $(RUNRUBY) "$(srcdir)/test/runner.rb" --ruby="$(RUNRUBY)" $(TEST_EXCLUDES) $(TESTOPTS) -- $(patsubst test/%,%,$@)
+
+clean-srcs-ext::
+	$(Q)$(RM) $(patsubst $(srcdir)/%,%,$(EXT_SRCS))
+
+clean-srcs-extra::
+	$(Q)$(RM) $(patsubst $(srcdir)/%,%,$(EXTRA_SRCS))
diff --git a/defs/id.def b/defs/id.def
new file mode 100644
index 0000000000..bb7cc6f922
--- /dev/null
+++ b/defs/id.def
@@ -0,0 +1,182 @@
+# -*- mode: ruby; coding: us-ascii -*-
+firstline, predefined = __LINE__+1, %[\
+  max
+  min
+  freeze
+  inspect
+  intern
+  object_id
+  const_missing
+  method_missing                                        MethodMissing
+  method_added
+  singleton_method_added
+  method_removed
+  singleton_method_removed
+  method_undefined
+  singleton_method_undefined
+  length
+  size
+  gets
+  succ
+  each
+  proc
+  lambda
+  send
+  __send__
+  __attached__
+  initialize
+  initialize_copy
+  initialize_clone
+  initialize_dup
+  to_int
+  to_ary
+  to_str
+  to_sym
+  to_hash
+  to_proc
+  to_io
+  to_a
+  to_s
+  to_i
+  to_f
+  to_r
+  bt
+  bt_locations
+  call
+  mesg
+  exception
+  not                                                   NOT
+  and                                                   AND
+  or                                                    OR
+
+  _                                                     UScore
+  "/*NULL*/"                                            NULL
+  empty?
+  eql?
+  respond_to?                                           Respond_to
+  respond_to_missing?                                   Respond_to_missing
+  <IFUNC>
+  <CFUNC>
+  core#set_method_alias
+  core#set_variable_alias
+  core#undef_method
+  core#define_method
+  core#define_singleton_method
+  core#set_postexe
+  core#hash_merge_ptr
+  core#hash_merge_kwd
+
+  -                                                     debug#created_info
+
+  $_                                                    LASTLINE
+  $~                                                    BACKREF
+]
+
+# VM ID         OP      Parser Token
+token_ops = %[\
+  Dot2          ..      DOT2
+  Dot3          ...     DOT3
+  UPlus         +@      UPLUS
+  UMinus        -@      UMINUS
+  Pow           **      POW
+  Cmp           <=>     CMP
+  PLUS          +
+  MINUS         -
+  MULT          *
+  DIV           /
+  MOD           %
+  LTLT          <<      LSHFT
+  GTGT          >>      RSHFT
+  LT            <
+  LE            <=      LEQ
+  GT            >
+  GE            >=      GEQ
+  Eq            ==      EQ
+  Eqq           ===     EQQ
+  Neq           !=      NEQ
+  Not           !
+  And           &
+  Or            |
+  Backquote     `
+  EqTilde       =~      MATCH
+  NeqTilde      !~      NMATCH
+  AREF          []
+  ASET          []=
+  COLON2        ::
+  ANDOP         &&
+  OROP          ||
+  ANDDOT        &.
+]
+
+class KeywordError < RuntimeError
+  def self.raise(mesg, line)
+    super(self, mesg, ["#{__FILE__}:#{line}", *caller])
+  end
+end
+
+predefined_ids = {}
+preserved_ids = []
+local_ids = []
+instance_ids = []
+global_ids = []
+const_ids = []
+class_ids = []
+attrset_ids = []
+token_op_ids = []
+names = {}
+predefined.split(/^/).each_with_index do |line, num|
+  next if /^#/ =~ line
+  line.sub!(/\s+#.*/, '')
+  name, token = line.split
+  next unless name
+  token ||= name
+  if /#/ =~ token
+    token = "_#{token.gsub(/\W+/, '_')}"
+  else
+    token = token.sub(/\?/, 'P').sub(/\A[a-z]/) {$&.upcase}
+    token.sub!(/\A\$/, "_G_")
+    token.sub!(/\A@@/, "_C_")
+    token.sub!(/\A@/, "_I_")
+    token.gsub!(/\W+/, "")
+  end
+  if name == '-'
+    preserved_ids << token
+    next
+  end
+  if prev = names[name]
+    KeywordError.raise("#{name} is already registered at line #{prev+firstline}", firstline+num)
+  end
+  if prev = predefined_ids[token]
+    KeywordError.raise("#{token} is already used for #{prev} at line #{names[prev]+firstline}", firstline+num)
+  end
+  names[name] = num
+  case name
+  when /\A[A-Z]\w*\z/; const_ids
+  when /\A(?!\d)\w+\z/; local_ids
+  when /\A\$(?:\d+|(?!\d)\w+|\W)\z/; global_ids
+  when /\A@@(?!\d)\w+\z/; class_ids
+  when /\A@(?!\d)\w+\z/; instance_ids
+  when /\A((?!\d)\w+)=\z/; attrset_ids
+  else preserved_ids
+  end << token
+  predefined_ids[token] = name
+end
+token_ops.split(/^/).each do |line|
+  next if /^#/ =~ line
+  line.sub!(/\s+#.*/, '')
+  id, op, token = line.split
+  next unless id and op
+  token ||= (id unless /\A\W\z/ =~ op)
+  token_op_ids << [id, op, token]
+end
+{
+  "LOCAL" => local_ids,
+  "INSTANCE" => instance_ids,
+  "GLOBAL" => global_ids,
+  "CONST" => const_ids,
+  "CLASS" => class_ids,
+  "ATTRSET" => attrset_ids,
+  :preserved => preserved_ids,
+  :predefined => predefined_ids,
+  :token_op => token_op_ids,
+}
diff --git a/defs/keywords b/defs/keywords
new file mode 100644
index 0000000000..fc30ec2d15
--- /dev/null
+++ b/defs/keywords
@@ -0,0 +1,53 @@
+%{
+struct kwtable {short name, id[2], state;};
+const struct kwtable *rb_reserved_word(const char *, unsigned int);
+#ifndef RIPPER
+static const struct kwtable *reserved_word(/*!ANSI{*/const char *, unsigned int/*}!ANSI*/);
+#define rb_reserved_word(str, len) reserved_word(str, len)
+%}
+
+struct kwtable;
+%%
+__ENCODING__, {keyword__ENCODING__, keyword__ENCODING__}, EXPR_END
+__LINE__, {keyword__LINE__, keyword__LINE__}, EXPR_END
+__FILE__, {keyword__FILE__, keyword__FILE__}, EXPR_END
+BEGIN, {keyword_BEGIN, keyword_BEGIN}, EXPR_END
+END, {keyword_END, keyword_END}, EXPR_END
+alias, {keyword_alias, keyword_alias}, EXPR_FNAME|EXPR_FITEM
+and, {keyword_and, keyword_and}, EXPR_VALUE
+begin, {keyword_begin, keyword_begin}, EXPR_BEG
+break, {keyword_break, keyword_break}, EXPR_MID
+case, {keyword_case, keyword_case}, EXPR_VALUE
+class, {keyword_class, keyword_class}, EXPR_CLASS
+def, {keyword_def, keyword_def}, EXPR_FNAME
+defined?, {keyword_defined, keyword_defined}, EXPR_ARG
+do, {keyword_do, keyword_do}, EXPR_BEG
+else, {keyword_else, keyword_else}, EXPR_BEG
+elsif, {keyword_elsif, keyword_elsif}, EXPR_VALUE
+end, {keyword_end, keyword_end}, EXPR_END
+ensure, {keyword_ensure, keyword_ensure}, EXPR_BEG
+false, {keyword_false, keyword_false}, EXPR_END
+for, {keyword_for, keyword_for}, EXPR_VALUE
+if, {keyword_if, modifier_if}, EXPR_VALUE
+in, {keyword_in, keyword_in}, EXPR_VALUE
+module, {keyword_module, keyword_module}, EXPR_VALUE
+next, {keyword_next, keyword_next}, EXPR_MID
+nil, {keyword_nil, keyword_nil}, EXPR_END
+not, {keyword_not, keyword_not}, EXPR_ARG
+or, {keyword_or, keyword_or}, EXPR_VALUE
+redo, {keyword_redo, keyword_redo}, EXPR_END
+rescue, {keyword_rescue, modifier_rescue}, EXPR_MID
+retry, {keyword_retry, keyword_retry}, EXPR_END
+return, {keyword_return, keyword_return}, EXPR_MID
+self, {keyword_self, keyword_self}, EXPR_END
+super, {keyword_super, keyword_super}, EXPR_ARG
+then, {keyword_then, keyword_then}, EXPR_BEG
+true, {keyword_true, keyword_true}, EXPR_END
+undef, {keyword_undef, keyword_undef}, EXPR_FNAME|EXPR_FITEM
+unless, {keyword_unless, modifier_unless}, EXPR_VALUE
+until, {keyword_until, modifier_until}, EXPR_VALUE
+when, {keyword_when, keyword_when}, EXPR_VALUE
+while, {keyword_while, modifier_while}, EXPR_VALUE
+yield, {keyword_yield, keyword_yield}, EXPR_ARG
+%%
+#endif
diff --git a/defs/known_errors.def b/defs/known_errors.def
new file mode 100644
index 0000000000..e9694cfbda
--- /dev/null
+++ b/defs/known_errors.def
@@ -0,0 +1,157 @@
+E2BIG
+EACCES
+EADDRINUSE
+EADDRNOTAVAIL
+EADV
+EAFNOSUPPORT
+EAGAIN
+EALREADY
+EAUTH
+EBADARCH
+EBADE
+EBADEXEC
+EBADF
+EBADFD
+EBADMACHO
+EBADMSG
+EBADR
+EBADRPC
+EBADRQC
+EBADSLT
+EBFONT
+EBUSY
+ECANCELED
+ECAPMODE
+ECHILD
+ECHRNG
+ECOMM
+ECONNABORTED
+ECONNREFUSED
+ECONNRESET
+EDEADLK
+EDEADLOCK
+EDESTADDRREQ
+EDEVERR
+EDOM
+EDOOFUS
+EDOTDOT
+EDQUOT
+EEXIST
+EFAULT
+EFBIG
+EFTYPE
+EHOSTDOWN
+EHOSTUNREACH
+EHWPOISON
+EIDRM
+EILSEQ
+EINPROGRESS
+EINTR
+EINVAL
+EIO
+EIPSEC
+EISCONN
+EISDIR
+EISNAM
+EKEYEXPIRED
+EKEYREJECTED
+EKEYREVOKED
+EL2HLT
+EL2NSYNC
+EL3HLT
+EL3RST
+ELAST
+ELIBACC
+ELIBBAD
+ELIBEXEC
+ELIBMAX
+ELIBSCN
+ELNRNG
+ELOOP
+EMEDIUMTYPE
+EMFILE
+EMLINK
+EMSGSIZE
+EMULTIHOP
+ENAMETOOLONG
+ENAVAIL
+ENEEDAUTH
+ENETDOWN
+ENETRESET
+ENETUNREACH
+ENFILE
+ENOANO
+ENOATTR
+ENOBUFS
+ENOCSI
+ENODATA
+ENODEV
+ENOENT
+ENOEXEC
+ENOKEY
+ENOLCK
+ENOLINK
+ENOMEDIUM
+ENOMEM
+ENOMSG
+ENONET
+ENOPKG
+ENOPOLICY
+ENOPROTOOPT
+ENOSPC
+ENOSR
+ENOSTR
+ENOSYS
+ENOTBLK
+ENOTCAPABLE
+ENOTCONN
+ENOTDIR
+ENOTEMPTY
+ENOTNAM
+ENOTRECOVERABLE
+ENOTSOCK
+ENOTSUP
+ENOTTY
+ENOTUNIQ
+ENXIO
+EOPNOTSUPP
+EOVERFLOW
+EOWNERDEAD
+EPERM
+EPFNOSUPPORT
+EPIPE
+EPROCLIM
+EPROCUNAVAIL
+EPROGMISMATCH
+EPROGUNAVAIL
+EPROTO
+EPROTONOSUPPORT
+EPROTOTYPE
+EPWROFF
+EQFULL
+ERANGE
+EREMCHG
+EREMOTE
+EREMOTEIO
+ERESTART
+ERFKILL
+EROFS
+ERPCMISMATCH
+ESHLIBVERS
+ESHUTDOWN
+ESOCKTNOSUPPORT
+ESPIPE
+ESRCH
+ESRMNT
+ESTALE
+ESTRPIPE
+ETIME
+ETIMEDOUT
+ETOOMANYREFS
+ETXTBSY
+EUCLEAN
+EUNATCH
+EUSERS
+EWOULDBLOCK
+EXDEV
+EXFULL
diff --git a/defs/lex.c.src b/defs/lex.c.src
new file mode 100644
index 0000000000..fc30ec2d15
--- /dev/null
+++ b/defs/lex.c.src
@@ -0,0 +1,53 @@
+%{
+struct kwtable {short name, id[2], state;};
+const struct kwtable *rb_reserved_word(const char *, unsigned int);
+#ifndef RIPPER
+static const struct kwtable *reserved_word(/*!ANSI{*/const char *, unsigned int/*}!ANSI*/);
+#define rb_reserved_word(str, len) reserved_word(str, len)
+%}
+
+struct kwtable;
+%%
+__ENCODING__, {keyword__ENCODING__, keyword__ENCODING__}, EXPR_END
+__LINE__, {keyword__LINE__, keyword__LINE__}, EXPR_END
+__FILE__, {keyword__FILE__, keyword__FILE__}, EXPR_END
+BEGIN, {keyword_BEGIN, keyword_BEGIN}, EXPR_END
+END, {keyword_END, keyword_END}, EXPR_END
+alias, {keyword_alias, keyword_alias}, EXPR_FNAME|EXPR_FITEM
+and, {keyword_and, keyword_and}, EXPR_VALUE
+begin, {keyword_begin, keyword_begin}, EXPR_BEG
+break, {keyword_break, keyword_break}, EXPR_MID
+case, {keyword_case, keyword_case}, EXPR_VALUE
+class, {keyword_class, keyword_class}, EXPR_CLASS
+def, {keyword_def, keyword_def}, EXPR_FNAME
+defined?, {keyword_defined, keyword_defined}, EXPR_ARG
+do, {keyword_do, keyword_do}, EXPR_BEG
+else, {keyword_else, keyword_else}, EXPR_BEG
+elsif, {keyword_elsif, keyword_elsif}, EXPR_VALUE
+end, {keyword_end, keyword_end}, EXPR_END
+ensure, {keyword_ensure, keyword_ensure}, EXPR_BEG
+false, {keyword_false, keyword_false}, EXPR_END
+for, {keyword_for, keyword_for}, EXPR_VALUE
+if, {keyword_if, modifier_if}, EXPR_VALUE
+in, {keyword_in, keyword_in}, EXPR_VALUE
+module, {keyword_module, keyword_module}, EXPR_VALUE
+next, {keyword_next, keyword_next}, EXPR_MID
+nil, {keyword_nil, keyword_nil}, EXPR_END
+not, {keyword_not, keyword_not}, EXPR_ARG
+or, {keyword_or, keyword_or}, EXPR_VALUE
+redo, {keyword_redo, keyword_redo}, EXPR_END
+rescue, {keyword_rescue, modifier_rescue}, EXPR_MID
+retry, {keyword_retry, keyword_retry}, EXPR_END
+return, {keyword_return, keyword_return}, EXPR_MID
+self, {keyword_self, keyword_self}, EXPR_END
+super, {keyword_super, keyword_super}, EXPR_ARG
+then, {keyword_then, keyword_then}, EXPR_BEG
+true, {keyword_true, keyword_true}, EXPR_END
+undef, {keyword_undef, keyword_undef}, EXPR_FNAME|EXPR_FITEM
+unless, {keyword_unless, modifier_unless}, EXPR_VALUE
+until, {keyword_until, modifier_until}, EXPR_VALUE
+when, {keyword_when, keyword_when}, EXPR_VALUE
+while, {keyword_while, modifier_while}, EXPR_VALUE
+yield, {keyword_yield, keyword_yield}, EXPR_ARG
+%%
+#endif
diff --git a/defs/opt_insn_unif.def b/defs/opt_insn_unif.def
new file mode 100644
index 0000000000..31ae2eb6a1
--- /dev/null
+++ b/defs/opt_insn_unif.def
@@ -0,0 +1,27 @@
+#
+# a definition of instruction unification
+#
+#
+
+__END__
+
+putobject putobject
+putobject putstring
+putobject setlocal
+
+putstring putstring
+putstring putobject
+putstring setlocal
+
+# putnil end
+
+dup setlocal
+
+# from tarai
+getlocal getlocal
+# getlocal send
+
+# from tak, ackermann
+getlocal putobject
+
+
diff --git a/defs/opt_operand.def b/defs/opt_operand.def
new file mode 100644
index 0000000000..887e3da49a
--- /dev/null
+++ b/defs/opt_operand.def
@@ -0,0 +1,22 @@
+#
+# configuration file for operand union optimization
+#
+# format:
+#   [insn name] op1, op2 ...
+#
+#  wildcard: *
+#
+
+getlocal *, 0
+getlocal *, 1
+setlocal *, 0
+setlocal *, 1
+
+putobject INT2FIX(0)
+putobject INT2FIX(1)
+
+__END__
+
+putobject Qtrue
+putobject Qfalse
+
diff --git a/defs/separated_version.mk b/defs/separated_version.mk
new file mode 100644
index 0000000000..f086f4b24a
--- /dev/null
+++ b/defs/separated_version.mk
@@ -0,0 +1,38 @@
+# ******** FOR DEVELEPERS ONLY ********
+# Separate version.o into a shared library which varies every
+# revisions, in order to make the rest sharable.
+
+include $(firstword $(wildcard GNUmakefile Makefile))
+
+ifneq ($(filter @executable_path/%,$(DLDFLAGS)),)
+RUBY_VERSION_SO = $(subst ruby,ruby_version,$(LIBRUBY_SO))
+RUBY_VERSION_DLDFLAGS := $(patsubst @executable_path/%/$(LIBRUBY_SO),@loader_path/$(RUBY_VERSION_SO),$(DLDFLAGS)) -exported_symbol=Init_version
+else ifneq ($(findstring -soname,$(DLDFLAGS)),)
+RUBY_VERSION_SO = $(subst ruby,ruby_version,$(LIBRUBY_SO))
+RUBY_VERSION_DLDFLAGS := $(subst ruby,ruby_version,$(DLDFLAGS)) -Wl,-rpath-link,'$${ORIGIN}'
+else
+ERROR
+endif
+
+ifneq ($(RUBY_VERSION_SO),)
+
+version.$(OBJEXT): XCFLAGS := $(subst -fvisibility=hidden,,$(XCFLAGS))
+
+MINIOBJS += version.$(OBJEXT)
+DLDOBJS += $(RUBY_VERSION_SO)
+LIBRUBYARG_SHARED := -lruby_version $(LIBRUBYARG_SHARED)
+$(LIBRUBY_SO): COMMONOBJS := $(filter-out version.$(OBJEXT),$(COMMONOBJS))
+$(LIBRUBY_A): COMMONOBJS := $(sort version.$(OBJEXT) $(COMMONOBJS))
+
+$(LIBRUBY_SO): $(RUBY_VERSION_SO)
+
+$(RUBY_VERSION_SO): version.$(OBJEXT)
+	$(ECHO) linking shared-library $@
+	$(LDSHARED) $(RUBY_VERSION_DLDFLAGS) version.$(OBJEXT) $(OUTFLAG)$@
+	-$(Q) $(MINIRUBY) -e 'ARGV.each{|link|' \
+		-e   'File.delete link rescue nil' \
+		-e   'File.symlink "$(RUBY_VERSION_SO)", link' \
+		-e '}' \
+		$(subst ruby,ruby_version,$(LIBRUBY_ALIASES)) || true
+
+endif
diff --git a/defs/universal.mk b/defs/universal.mk
new file mode 100644
index 0000000000..c34a31b356
--- /dev/null
+++ b/defs/universal.mk
@@ -0,0 +1,5 @@
+arch_flags := $(filter -arch=%,$(subst -arch ,-arch=,$(ARCH_FLAG)))
+ifeq ($(filter 0 1,$(words $(arch_flags))),)
+override MJIT_HEADER_SUFFIX = -%
+override MJIT_HEADER_ARCH = -$(word 2,$(ARCH_FLAG))
+endif
diff --git a/dir.c b/dir.c
index 0798b3f76f..2f38e35976 100644
--- a/dir.c
+++ b/dir.c
@@ -2,17 +2,20 @@
 
   dir.c -
 
-  $Author: shyouhei $
-  $Date: 2006/12/14 14:50:13 $
+  $Author$
   created at: Wed Jan  5 09:51:01 JST 1994
 
-  Copyright (C) 1993-2003 Yukihiro Matsumoto
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
   Copyright (C) 2000  Network Applied Communication Laboratory, Inc.
   Copyright (C) 2000  Information-technology Promotion Agency, Japan
 
 **********************************************************************/
 
-#include "ruby.h"
+#include "ruby/encoding.h"
+#include "ruby/thread.h"
+#include "internal.h"
+#include "id.h"
+#include "encindex.h"
 
 #include <sys/types.h>
 #include <sys/stat.h>
@@ -21,6 +24,26 @@
 #include <unistd.h>
 #endif
 
+#ifndef O_CLOEXEC
+#  define O_CLOEXEC 0
+#endif
+
+#ifndef USE_OPENDIR_AT
+# if defined(HAVE_FDOPENDIR) && defined(HAVE_DIRFD) && \
+    defined(HAVE_OPENAT) && defined(HAVE_FSTATAT)
+#   define USE_OPENDIR_AT 1
+# else
+#   define USE_OPENDIR_AT 0
+# endif
+#endif
+#if USE_OPENDIR_AT
+# include <fcntl.h>
+#endif
+#ifndef AT_FDCWD
+# define AT_FDCWD -1
+#endif
+
+#undef HAVE_DIRENT_NAMLEN
 #if defined HAVE_DIRENT_H && !defined _WIN32
 # include <dirent.h>
 # define NAMLEN(dirent) strlen((dirent)->d_name)
@@ -29,12 +52,8 @@
 # define NAMLEN(dirent) strlen((dirent)->d_name)
 #else
 # define dirent direct
-# if !defined __NeXT__
-#  define NAMLEN(dirent) (dirent)->d_namlen
-# else
-#  /* On some versions of NextStep, d_namlen is always zero, so avoid it. */
-#  define NAMLEN(dirent) strlen((dirent)->d_name)
-# endif
+# define NAMLEN(dirent) (dirent)->d_namlen
+# define HAVE_DIRENT_NAMLEN 1
 # if HAVE_SYS_NDIR_H
 #  include <sys/ndir.h>
 # endif
@@ -56,259 +75,522 @@ char *getenv();
 #endif
 
 #ifndef HAVE_STRING_H
-char *strchr _((char*,char));
+char *strchr(char*,char);
 #endif
 
 #include <ctype.h>
 
-#include "util.h"
+#include "ruby/util.h"
+
+#define vm_initialized rb_cThread
+
+/* define system APIs */
+#ifdef _WIN32
+#undef chdir
+#define chdir(p) rb_w32_uchdir(p)
+#undef mkdir
+#define mkdir(p, m) rb_w32_umkdir((p), (m))
+#undef rmdir
+#define rmdir(p) rb_w32_urmdir(p)
+#undef opendir
+#define opendir(p) rb_w32_uopendir(p)
+#define ruby_getcwd() rb_w32_ugetcwd(NULL, 0)
+#define IS_WIN32 1
+#else
+#define IS_WIN32 0
+#endif
+
+#ifdef HAVE_SYS_ATTR_H
+#include <sys/attr.h>
+#endif
+
+#define USE_NAME_ON_FS_REAL_BASENAME 1	/* platform dependent APIs to
+					 * get real basenames */
+#define USE_NAME_ON_FS_BY_FNMATCH 2	/* select the matching
+					 * basename by fnmatch */
+
+#ifdef HAVE_GETATTRLIST
+# define USE_NAME_ON_FS USE_NAME_ON_FS_REAL_BASENAME
+# define RUP32(size) ((size)+3/4)
+# define SIZEUP32(type) RUP32(sizeof(type))
+#elif defined _WIN32
+# define USE_NAME_ON_FS USE_NAME_ON_FS_REAL_BASENAME
+#elif defined DOSISH
+# define USE_NAME_ON_FS USE_NAME_ON_FS_BY_FNMATCH
+#else
+# define USE_NAME_ON_FS 0
+#endif
 
-#if !defined HAVE_LSTAT && !defined lstat
-#define lstat stat
+#ifdef __APPLE__
+# define NORMALIZE_UTF8PATH 1
+#else
+# define NORMALIZE_UTF8PATH 0
 #endif
 
-#ifndef CASEFOLD_FILESYSTEM
-# if defined DOSISH || defined __VMS
-#   define CASEFOLD_FILESYSTEM 1
+#if NORMALIZE_UTF8PATH
+#include <sys/param.h>
+#include <sys/mount.h>
+#include <sys/vnode.h>
+
+# if defined HAVE_FGETATTRLIST || !defined HAVE_GETATTRLIST
+#   define need_normalization(dirp, path) need_normalization(dirp)
 # else
-#   define CASEFOLD_FILESYSTEM 0
+#   define need_normalization(dirp, path) need_normalization(path)
+# endif
+static inline int
+need_normalization(DIR *dirp, const char *path)
+{
+# if defined HAVE_FGETATTRLIST || defined HAVE_GETATTRLIST
+    u_int32_t attrbuf[SIZEUP32(fsobj_tag_t)];
+    struct attrlist al = {ATTR_BIT_MAP_COUNT, 0, ATTR_CMN_OBJTAG,};
+#   if defined HAVE_FGETATTRLIST
+    int ret = fgetattrlist(dirfd(dirp), &al, attrbuf, sizeof(attrbuf), 0);
+#   else
+    int ret = getattrlist(path, &al, attrbuf, sizeof(attrbuf), 0);
+#   endif
+    if (!ret) {
+	const fsobj_tag_t *tag = (void *)(attrbuf+1);
+	switch (*tag) {
+	  case VT_HFS:
+	  case VT_CIFS:
+	    return TRUE;
+	}
+    }
 # endif
+    return FALSE;
+}
+
+static inline int
+has_nonascii(const char *ptr, size_t len)
+{
+    while (len > 0) {
+	if (!ISASCII(*ptr)) return 1;
+	ptr++;
+	--len;
+    }
+    return 0;
+}
+
+# define IF_NORMALIZE_UTF8PATH(something) something
+#else
+# define IF_NORMALIZE_UTF8PATH(something) /* nothing */
 #endif
 
+#ifndef IFTODT
+# define IFTODT(m)	(((m) & S_IFMT) / ((~S_IFMT & (S_IFMT-1)) + 1))
+#endif
+
+typedef enum {
+#ifdef DT_UNKNOWN
+    path_exist     = DT_UNKNOWN,
+    path_directory = DT_DIR,
+    path_regular   = DT_REG,
+    path_symlink   = DT_LNK,
+#else
+    path_exist,
+    path_directory = IFTODT(S_IFDIR),
+    path_regular   = IFTODT(S_IFREG),
+    path_symlink   = IFTODT(S_IFLNK),
+#endif
+    path_noent = -1,
+    path_unknown = -2
+} rb_pathtype_t;
+
 #define FNM_NOESCAPE	0x01
 #define FNM_PATHNAME	0x02
 #define FNM_DOTMATCH	0x04
 #define FNM_CASEFOLD	0x08
+#define FNM_EXTGLOB	0x10
 #if CASEFOLD_FILESYSTEM
 #define FNM_SYSCASE	FNM_CASEFOLD
 #else
 #define FNM_SYSCASE	0
 #endif
+#if _WIN32
+#define FNM_SHORTNAME	0x20
+#else
+#define FNM_SHORTNAME	0
+#endif
 
 #define FNM_NOMATCH	1
 #define FNM_ERROR	2
 
-#define downcase(c) (nocase && ISUPPER(c) ? tolower(c) : (c))
-
-#ifndef CharNext		/* defined as CharNext[AW] on Windows. */
-# if defined(DJGPP)
-#   define CharNext(p) ((p) + mblen(p, MB_CUR_MAX))
-# else
-#   define CharNext(p) ((p) + 1)
-# endif
-#endif
-
-#if defined DOSISH
-#define isdirsep(c) ((c) == '/' || (c) == '\\')
-#else
-#define isdirsep(c) ((c) == '/')
-#endif
+# define Next(p, e, enc) ((p)+ rb_enc_mbclen((p), (e), (enc)))
+# define Inc(p, e, enc) ((p) = Next((p), (e), (enc)))
 
 static char *
-range(pat, test, flags)
-    const char *pat;
-    int test;
-    int flags;
+bracket(
+    const char *p, /* pattern (next to '[') */
+    const char *pend,
+    const char *s, /* string */
+    const char *send,
+    int flags,
+    rb_encoding *enc)
 {
-    int not, ok = 0;
-    int nocase = flags & FNM_CASEFOLD;
-    int escape = !(flags & FNM_NOESCAPE);
-
-    not = *pat == '!' || *pat == '^';
-    if (not)
-	pat++;
+    const int nocase = flags & FNM_CASEFOLD;
+    const int escape = !(flags & FNM_NOESCAPE);
+    unsigned int c1, c2;
+    int r;
+    int ok = 0, not = 0;
 
-    test = downcase(test);
+    if (p >= pend) return NULL;
+    if (*p == '!' || *p == '^') {
+	not = 1;
+	p++;
+    }
 
-    while (*pat != ']') {
-	int cstart, cend;
-        if (escape && *pat == '\\')
-	    pat++;
-	cstart = cend = *pat++;
-	if (!cstart)
+    while (*p != ']') {
+	const char *t1 = p;
+	if (escape && *t1 == '\\')
+	    t1++;
+	if (!*t1)
 	    return NULL;
-	if (*pat == '-' && pat[1] != ']') {
-	    pat++;
-	    if (escape && *pat == '\\')
-		pat++;
-	    cend = *pat++;
-	    if (!cend)
+	p = t1 + (r = rb_enc_mbclen(t1, pend, enc));
+	if (p >= pend) return NULL;
+	if (p[0] == '-' && p[1] != ']') {
+	    const char *t2 = p + 1;
+	    int r2;
+	    if (escape && *t2 == '\\')
+		t2++;
+	    if (!*t2)
 		return NULL;
+	    p = t2 + (r2 = rb_enc_mbclen(t2, pend, enc));
+	    if (ok) continue;
+	    if ((r <= (send-s) && memcmp(t1, s, r) == 0) ||
+		(r2 <= (send-s) && memcmp(t2, s, r2) == 0)) {
+		ok = 1;
+		continue;
+	    }
+	    c1 = rb_enc_codepoint(s, send, enc);
+	    if (nocase) c1 = rb_enc_toupper(c1, enc);
+	    c2 = rb_enc_codepoint(t1, pend, enc);
+	    if (nocase) c2 = rb_enc_toupper(c2, enc);
+	    if (c1 < c2) continue;
+	    c2 = rb_enc_codepoint(t2, pend, enc);
+	    if (nocase) c2 = rb_enc_toupper(c2, enc);
+	    if (c1 > c2) continue;
 	}
-	if (downcase(cstart) <= test && test <= downcase(cend))
-	    ok = 1;
+	else {
+	    if (ok) continue;
+	    if (r <= (send-s) && memcmp(t1, s, r) == 0) {
+		ok = 1;
+		continue;
+	    }
+	    if (!nocase) continue;
+	    c1 = rb_enc_toupper(rb_enc_codepoint(s, send, enc), enc);
+	    c2 = rb_enc_toupper(rb_enc_codepoint(p, pend, enc), enc);
+	    if (c1 != c2) continue;
+	}
+	ok = 1;
     }
-    return ok == not ? NULL : (char *)pat + 1;
+
+    return ok == not ? NULL : (char *)p + 1;
 }
 
-#define ISDIRSEP(c) (pathname && isdirsep(c))
-#define PERIOD(s) (period && *(s) == '.' && \
-		  ((s) == string || ISDIRSEP((s)[-1])))
+/* If FNM_PATHNAME is set, only path element will be matched. (up to '/' or '\0')
+   Otherwise, entire string will be matched.
+   End marker itself won't be compared.
+   And if function succeeds, *pcur reaches end marker.
+*/
+#define UNESCAPE(p) (escape && *(p) == '\\' ? (p) + 1 : (p))
+#define ISEND(p) (!*(p) || (pathname && *(p) == '/'))
+#define RETURN(val) return *pcur = p, *scur = s, (val);
+
 static int
-fnmatch(pat, string, flags)
-    const char *pat;
-    const char *string;
-    int flags;
+fnmatch_helper(
+    const char **pcur, /* pattern */
+    const char **scur, /* string */
+    int flags,
+    rb_encoding *enc)
 {
-    int c;
-    int test;
-    const char *s = string;
-    int escape = !(flags & FNM_NOESCAPE);
-    int pathname = flags & FNM_PATHNAME;
-    int period = !(flags & FNM_DOTMATCH);
-    int nocase = flags & FNM_CASEFOLD;
+    const int period = !(flags & FNM_DOTMATCH);
+    const int pathname = flags & FNM_PATHNAME;
+    const int escape = !(flags & FNM_NOESCAPE);
+    const int nocase = flags & FNM_CASEFOLD;
 
-    while ((c = *pat++) != '\0') {
-	switch (c) {
-	case '?':
-	    if (!*s || ISDIRSEP(*s) || PERIOD(s))
-		return FNM_NOMATCH;
-	    s++;
-	    break;
-	case '*':
-	    while ((c = *pat++) == '*')
-		;
+    const char *ptmp = 0;
+    const char *stmp = 0;
 
-	    if (PERIOD(s))
-		return FNM_NOMATCH;
+    const char *p = *pcur;
+    const char *pend = p + strlen(p);
+    const char *s = *scur;
+    const char *send = s + strlen(s);
 
-	    if (!c) {
-		if (pathname && *rb_path_next(s))
-		    return FNM_NOMATCH;
-		else
-		    return 0;
-	    }
-	    else if (ISDIRSEP(c)) {
-		s = rb_path_next(s);
-		if (*s) {
-                    s++;
-		    break;
-                }
-		return FNM_NOMATCH;
-	    }
+    int r;
 
-	    test = escape && c == '\\' ? *pat : c;
-	    test = downcase(test);
-	    pat--;
-	    while (*s) {
-		if ((c == '?' || c == '[' || downcase(*s) == test) &&
-		    !fnmatch(pat, s, flags | FNM_DOTMATCH))
-		    return 0;
-		else if (ISDIRSEP(*s))
-		    break;
-		s++;
+    if (period && *s == '.' && *UNESCAPE(p) != '.') /* leading period */
+	RETURN(FNM_NOMATCH);
+
+    while (1) {
+	switch (*p) {
+	  case '*':
+	    do { p++; } while (*p == '*');
+	    if (ISEND(UNESCAPE(p))) {
+		p = UNESCAPE(p);
+		RETURN(0);
 	    }
-	    return FNM_NOMATCH;
+	    if (ISEND(s))
+		RETURN(FNM_NOMATCH);
+	    ptmp = p;
+	    stmp = s;
+	    continue;
 
-	case '[':
-	    if (!*s || ISDIRSEP(*s) || PERIOD(s))
-		return FNM_NOMATCH;
-	    pat = range(pat, *s, flags);
-	    if (pat == NULL)
-		return FNM_NOMATCH;
-	    s++;
-	    break;
+	  case '?':
+	    if (ISEND(s))
+		RETURN(FNM_NOMATCH);
+	    p++;
+	    Inc(s, send, enc);
+	    continue;
 
-	case '\\':
-	    if (escape
-#if defined DOSISH
-		&& *pat && strchr("*?[]\\", *pat)
-#endif
-		) {
-		c = *pat;
-		if (!c)
-		    c = '\\';
-		else
-		    pat++;
+	  case '[': {
+	    const char *t;
+	    if (ISEND(s))
+		RETURN(FNM_NOMATCH);
+	    if ((t = bracket(p + 1, pend, s, send, flags, enc)) != 0) {
+		p = t;
+		Inc(s, send, enc);
+		continue;
 	    }
-	    /* FALLTHROUGH */
+	    goto failed;
+	  }
+	}
 
-	default:
-#if defined DOSISH
-	    if (ISDIRSEP(c) && isdirsep(*s))
-		;
-	    else
-#endif
-	    if(downcase(c) != downcase(*s))
-		return FNM_NOMATCH;
-	    s++;
-	    break;
+	/* ordinary */
+	p = UNESCAPE(p);
+	if (ISEND(s))
+	    RETURN(ISEND(p) ? 0 : FNM_NOMATCH);
+	if (ISEND(p))
+	    goto failed;
+	r = rb_enc_precise_mbclen(p, pend, enc);
+	if (!MBCLEN_CHARFOUND_P(r))
+	    goto failed;
+	if (r <= (send-s) && memcmp(p, s, r) == 0) {
+	    p += r;
+	    s += r;
+	    continue;
+	}
+	if (!nocase) goto failed;
+	if (rb_enc_toupper(rb_enc_codepoint(p, pend, enc), enc) !=
+	    rb_enc_toupper(rb_enc_codepoint(s, send, enc), enc))
+	    goto failed;
+	p += r;
+	Inc(s, send, enc);
+	continue;
+
+      failed: /* try next '*' position */
+	if (ptmp && stmp) {
+	    p = ptmp;
+	    Inc(stmp, send, enc); /* !ISEND(*stmp) */
+	    s = stmp;
+	    continue;
 	}
+	RETURN(FNM_NOMATCH);
     }
-    return !*s ? 0 : FNM_NOMATCH;
+}
+
+static int
+fnmatch(
+    const char *pattern,
+    rb_encoding *enc,
+    const char *string,
+    int flags)
+{
+    const char *p = pattern;
+    const char *s = string;
+    const char *send = s + strlen(string);
+    const int period = !(flags & FNM_DOTMATCH);
+    const int pathname = flags & FNM_PATHNAME;
+
+    const char *ptmp = 0;
+    const char *stmp = 0;
+
+    if (pathname) {
+	while (1) {
+	    if (p[0] == '*' && p[1] == '*' && p[2] == '/') {
+		do { p += 3; } while (p[0] == '*' && p[1] == '*' && p[2] == '/');
+		ptmp = p;
+		stmp = s;
+	    }
+	    if (fnmatch_helper(&p, &s, flags, enc) == 0) {
+		while (*s && *s != '/') Inc(s, send, enc);
+		if (*p && *s) {
+		    p++;
+		    s++;
+		    continue;
+		}
+		if (!*p && !*s)
+		    return 0;
+	    }
+	    /* failed : try next recursion */
+	    if (ptmp && stmp && !(period && *stmp == '.')) {
+		while (*stmp && *stmp != '/') Inc(stmp, send, enc);
+		if (*stmp) {
+		    p = ptmp;
+		    stmp++;
+		    s = stmp;
+		    continue;
+		}
+	    }
+	    return FNM_NOMATCH;
+	}
+    }
+    else
+	return fnmatch_helper(&p, &s, flags, enc);
 }
 
 VALUE rb_cDir;
 
 struct dir_data {
     DIR *dir;
-    char *path;
+    const VALUE path;
+    rb_encoding *enc;
 };
 
 static void
-free_dir(dir)
-    struct dir_data *dir;
+dir_mark(void *ptr)
 {
-    if (dir) {
-	if (dir->dir) closedir(dir->dir);
-	if (dir->path) free(dir->path);
-    }
-    free(dir);
+    struct dir_data *dir = ptr;
+    rb_gc_mark(dir->path);
+}
+
+static void
+dir_free(void *ptr)
+{
+    struct dir_data *dir = ptr;
+
+    if (dir->dir) closedir(dir->dir);
+    xfree(dir);
 }
 
-static VALUE dir_close _((VALUE));
+static size_t
+dir_memsize(const void *ptr)
+{
+    return sizeof(struct dir_data);
+}
+
+static const rb_data_type_t dir_data_type = {
+    "dir",
+    {dir_mark, dir_free, dir_memsize,},
+    0, 0, RUBY_TYPED_WB_PROTECTED | RUBY_TYPED_FREE_IMMEDIATELY
+};
+
+static VALUE dir_close(VALUE);
 
-static VALUE dir_s_alloc _((VALUE));
 static VALUE
-dir_s_alloc(klass)
-    VALUE klass;
+dir_s_alloc(VALUE klass)
 {
     struct dir_data *dirp;
-    VALUE obj = Data_Make_Struct(klass, struct dir_data, 0, free_dir, dirp);
+    VALUE obj = TypedData_Make_Struct(klass, struct dir_data, &dir_data_type, dirp);
 
     dirp->dir = NULL;
-    dirp->path = NULL;
+    RB_OBJ_WRITE(obj, &dirp->path, Qnil);
+    dirp->enc = NULL;
 
     return obj;
 }
 
+static void *
+nogvl_opendir(void *ptr)
+{
+    const char *path = ptr;
+
+    return (void *)opendir(path);
+}
+
+static DIR *
+opendir_without_gvl(const char *path)
+{
+    if (vm_initialized) {
+	union { const void *in; void *out; } u;
+
+	u.in = path;
+
+	return rb_thread_call_without_gvl(nogvl_opendir, u.out, RUBY_UBF_IO, 0);
+    }
+    else
+	return opendir(path);
+}
+
 /*
  *  call-seq:
  *     Dir.new( string ) -> aDir
+ *     Dir.new( string, encoding: enc ) -> aDir
  *
  *  Returns a new directory object for the named directory.
+ *
+ *  The optional <i>encoding</i> keyword argument specifies the encoding of the directory.
+ *  If not specified, the filesystem encoding is used.
  */
 static VALUE
-dir_initialize(dir, dirname)
-    VALUE dir, dirname;
+dir_initialize(int argc, VALUE *argv, VALUE dir)
 {
     struct dir_data *dp;
+    rb_encoding  *fsenc;
+    VALUE dirname, opt, orig;
+    static ID keyword_ids[1];
+    const char *path;
+
+    if (!keyword_ids[0]) {
+	keyword_ids[0] = rb_id_encoding();
+    }
+
+    fsenc = rb_filesystem_encoding();
 
-    SafeStringValue(dirname);
-    Data_Get_Struct(dir, struct dir_data, dp);
+    rb_scan_args(argc, argv, "1:", &dirname, &opt);
+
+    if (!NIL_P(opt)) {
+	VALUE enc;
+	rb_get_kwargs(opt, keyword_ids, 0, 1, &enc);
+	if (enc != Qundef && !NIL_P(enc)) {
+	    fsenc = rb_to_encoding(enc);
+	}
+    }
+
+    FilePathValue(dirname);
+    orig = rb_str_dup_frozen(dirname);
+    dirname = rb_str_encode_ospath(dirname);
+    dirname = rb_str_dup_frozen(dirname);
+
+    TypedData_Get_Struct(dir, struct dir_data, &dir_data_type, dp);
     if (dp->dir) closedir(dp->dir);
-    if (dp->path) free(dp->path);
     dp->dir = NULL;
-    dp->path = NULL;
-    dp->dir = opendir(RSTRING(dirname)->ptr);
+    RB_OBJ_WRITE(dir, &dp->path, Qnil);
+    dp->enc = fsenc;
+    path = RSTRING_PTR(dirname);
+    dp->dir = opendir_without_gvl(path);
     if (dp->dir == NULL) {
-	if (errno == EMFILE || errno == ENFILE) {
-	    rb_gc();
-	    dp->dir = opendir(RSTRING(dirname)->ptr);
+	int e = errno;
+	if (rb_gc_for_fd(e)) {
+	    dp->dir = opendir_without_gvl(path);
 	}
+#ifdef HAVE_GETATTRLIST
+	else if (e == EIO) {
+	    u_int32_t attrbuf[1];
+	    struct attrlist al = {ATTR_BIT_MAP_COUNT, 0};
+	    if (getattrlist(path, &al, attrbuf, sizeof(attrbuf), FSOPT_NOFOLLOW) == 0) {
+		dp->dir = opendir_without_gvl(path);
+	    }
+	}
+#endif
 	if (dp->dir == NULL) {
-	    rb_sys_fail(RSTRING(dirname)->ptr);
+	    RB_GC_GUARD(dirname);
+	    rb_syserr_fail_path(e, orig);
 	}
     }
-    dp->path = strdup(RSTRING(dirname)->ptr);
+    RB_OBJ_WRITE(dir, &dp->path, orig);
 
     return dir;
 }
 
 /*
  *  call-seq:
- *     Dir.open( string ) => aDir
- *     Dir.open( string ) {| aDir | block } => anObject
+ *     Dir.open( string ) -> aDir
+ *     Dir.open( string, encoding: enc ) -> aDir
+ *     Dir.open( string ) {| aDir | block } -> anObject
+ *     Dir.open( string, encoding: enc ) {| aDir | block } -> anObject
+ *
+ *  The optional <i>encoding</i> keyword argument specifies the encoding of the directory.
+ *  If not specified, the filesystem encoding is used.
  *
  *  With no block, <code>open</code> is a synonym for
  *  <code>Dir::new</code>. If a block is present, it is passed
@@ -316,15 +598,13 @@ dir_initialize(dir, dirname)
  *  the block, and <code>Dir::open</code> returns the value of the
  *  block.
  */
-
 static VALUE
-dir_s_open(klass, dirname)
-    VALUE klass, dirname;
+dir_s_open(int argc, VALUE *argv, VALUE klass)
 {
     struct dir_data *dp;
-    VALUE dir = Data_Make_Struct(klass, struct dir_data, 0, free_dir, dp);
+    VALUE dir = TypedData_Make_Struct(klass, struct dir_data, &dir_data_type, dp);
 
-    dir_initialize(dir, dirname);
+    dir_initialize(argc, argv, dir);
     if (rb_block_given_p()) {
 	return rb_ensure(rb_yield, dir, dir_close, dir);
     }
@@ -332,30 +612,103 @@ dir_s_open(klass, dirname)
     return dir;
 }
 
+NORETURN(static void dir_closed(void));
+
 static void
-dir_closed()
+dir_closed(void)
 {
     rb_raise(rb_eIOError, "closed directory");
 }
 
-static void
-dir_check(dir)
-    VALUE dir;
+static struct dir_data *
+dir_get(VALUE dir)
 {
-    if (!OBJ_TAINTED(dir) && rb_safe_level() >= 4)
-	rb_raise(rb_eSecurityError, "Insecure: operation on untainted Dir");
     rb_check_frozen(dir);
+    return rb_check_typeddata(dir, &dir_data_type);
+}
+
+static struct dir_data *
+dir_check(VALUE dir)
+{
+    struct dir_data *dirp = dir_get(dir);
+    if (!dirp->dir) dir_closed();
+    return dirp;
+}
+
+#define GetDIR(obj, dirp) ((dirp) = dir_check(obj))
+
+
+/*
+ *  call-seq:
+ *     dir.inspect -> string
+ *
+ *  Return a string describing this Dir object.
+ */
+static VALUE
+dir_inspect(VALUE dir)
+{
+    struct dir_data *dirp;
+
+    TypedData_Get_Struct(dir, struct dir_data, &dir_data_type, dirp);
+    if (!NIL_P(dirp->path)) {
+	VALUE str = rb_str_new_cstr("#<");
+	rb_str_append(str, rb_class_name(CLASS_OF(dir)));
+	rb_str_cat2(str, ":");
+	rb_str_append(str, dirp->path);
+	rb_str_cat2(str, ">");
+	return str;
+    }
+    return rb_funcallv(dir, idTo_s, 0, 0);
 }
 
-#define GetDIR(obj, dirp) do {\
-    dir_check(dir);\
-    Data_Get_Struct(obj, struct dir_data, dirp);\
-    if (dirp->dir == NULL) dir_closed();\
-} while (0)
+/* Workaround for Solaris 10 that does not have dirfd.
+   Note: Solaris 11 (POSIX.1-2008 compliant) has dirfd(3C).
+ */
+#if defined(__sun) && !defined(HAVE_DIRFD)
+# if defined(HAVE_DIR_D_FD)
+#  define dirfd(x) ((x)->d_fd)
+#  define HAVE_DIRFD 1
+# elif defined(HAVE_DIR_DD_FD)
+#  define dirfd(x) ((x)->dd_fd)
+#  define HAVE_DIRFD 1
+# endif
+#endif
 
+#ifdef HAVE_DIRFD
 /*
  *  call-seq:
- *     dir.path => string or nil
+ *     dir.fileno -> integer
+ *
+ *  Returns the file descriptor used in <em>dir</em>.
+ *
+ *     d = Dir.new("..")
+ *     d.fileno   #=> 8
+ *
+ *  This method uses dirfd() function defined by POSIX 2008.
+ *  NotImplementedError is raised on other platforms, such as Windows,
+ *  which doesn't provide the function.
+ *
+ */
+static VALUE
+dir_fileno(VALUE dir)
+{
+    struct dir_data *dirp;
+    int fd;
+
+    GetDIR(dir, dirp);
+    fd = dirfd(dirp->dir);
+    if (fd == -1)
+	rb_sys_fail("dirfd");
+    return INT2NUM(fd);
+}
+#else
+#define dir_fileno rb_f_notimplement
+#endif
+
+/*
+ *  call-seq:
+ *     dir.path -> string or nil
+ *     dir.to_path -> string or nil
  *
  *  Returns the path parameter passed to <em>dir</em>'s constructor.
  *
@@ -363,19 +716,59 @@ dir_check(dir)
  *     d.path   #=> ".."
  */
 static VALUE
-dir_path(dir)
-    VALUE dir;
+dir_path(VALUE dir)
 {
     struct dir_data *dirp;
 
-    GetDIR(dir, dirp);
-    if (!dirp->path) return Qnil;
-    return rb_str_new2(dirp->path);
+    TypedData_Get_Struct(dir, struct dir_data, &dir_data_type, dirp);
+    if (NIL_P(dirp->path)) return Qnil;
+    return rb_str_dup(dirp->path);
+}
+
+#if defined _WIN32
+static int
+fundamental_encoding_p(rb_encoding *enc)
+{
+    switch (rb_enc_to_index(enc)) {
+      case ENCINDEX_ASCII:
+      case ENCINDEX_US_ASCII:
+      case ENCINDEX_UTF_8:
+	return TRUE;
+      default:
+	return FALSE;
+    }
+}
+# define READDIR(dir, enc) rb_w32_readdir((dir), (enc))
+#else
+# define READDIR(dir, enc) readdir((dir))
+#endif
+
+/* safe to use without GVL */
+static int
+to_be_skipped(const struct dirent *dp)
+{
+    const char *name = dp->d_name;
+    if (name[0] != '.') return FALSE;
+#ifdef HAVE_DIRENT_NAMLEN
+    switch (NAMLEN(dp)) {
+      case 2:
+	if (name[1] != '.') return FALSE;
+      case 1:
+	return TRUE;
+      default:
+	break;
+    }
+#else
+    if (!name[1]) return TRUE;
+    if (name[1] != '.') return FALSE;
+    if (!name[2]) return TRUE;
+#endif
+    return FALSE;
 }
 
 /*
  *  call-seq:
- *     dir.read => string or nil
+ *     dir.read -> string or nil
  *
  *  Reads the next entry from <em>dir</em> and returns it as a string.
  *  Returns <code>nil</code> at the end of the stream.
@@ -386,34 +779,41 @@ dir_path(dir)
  *     d.read   #=> "config.h"
  */
 static VALUE
-dir_read(dir)
-    VALUE dir;
+dir_read(VALUE dir)
 {
     struct dir_data *dirp;
     struct dirent *dp;
 
     GetDIR(dir, dirp);
     errno = 0;
-    dp = readdir(dirp->dir);
-    if (dp) {
-	return rb_tainted_str_new(dp->d_name, NAMLEN(dp));
-    }
-    else if (errno == 0) {	/* end of stream */
-	return Qnil;
+    if ((dp = READDIR(dirp->dir, dirp->enc)) != NULL) {
+	return rb_external_str_new_with_enc(dp->d_name, NAMLEN(dp), dirp->enc);
     }
     else {
-	rb_sys_fail(0);
+	int e = errno;
+	if (e != 0) rb_syserr_fail(e, 0);
+	return Qnil;		/* end of stream */
     }
-    return Qnil;		/* not reached */
+}
+
+static VALUE dir_each_entry(VALUE, VALUE (*)(VALUE, VALUE), VALUE, int);
+
+static VALUE
+dir_yield(VALUE arg, VALUE path)
+{
+    return rb_yield(path);
 }
 
 /*
  *  call-seq:
- *     dir.each { |filename| block }  => dir
+ *     dir.each { |filename| block }  -> dir
+ *     dir.each                       -> an_enumerator
  *
  *  Calls the block once for each entry in this directory, passing the
  *  filename of each entry as a parameter to the block.
  *
+ *  If no block is given, an enumerator is returned instead.
+ *
  *     d = Dir.new("testdir")
  *     d.each  {|x| puts "Got #{x}" }
  *
@@ -425,25 +825,50 @@ dir_read(dir)
  *     Got main.rb
  */
 static VALUE
-dir_each(dir)
-    VALUE dir;
+dir_each(VALUE dir)
+{
+    RETURN_ENUMERATOR(dir, 0, 0);
+    return dir_each_entry(dir, dir_yield, Qnil, FALSE);
+}
+
+static VALUE
+dir_each_entry(VALUE dir, VALUE (*each)(VALUE, VALUE), VALUE arg, int children_only)
 {
     struct dir_data *dirp;
     struct dirent *dp;
+    IF_NORMALIZE_UTF8PATH(int norm_p);
 
     GetDIR(dir, dirp);
     rewinddir(dirp->dir);
-    for (dp = readdir(dirp->dir); dp != NULL; dp = readdir(dirp->dir)) {
-	rb_yield(rb_tainted_str_new(dp->d_name, NAMLEN(dp)));
+    IF_NORMALIZE_UTF8PATH(norm_p = need_normalization(dirp->dir, RSTRING_PTR(dirp->path)));
+    while ((dp = READDIR(dirp->dir, dirp->enc)) != NULL) {
+	const char *name = dp->d_name;
+	size_t namlen = NAMLEN(dp);
+	VALUE path;
+
+	if (children_only && name[0] == '.') {
+	    if (namlen == 1) continue; /* current directory */
+	    if (namlen == 2 && name[1] == '.') continue; /* parent directory */
+	}
+#if NORMALIZE_UTF8PATH
+	if (norm_p && has_nonascii(name, namlen) &&
+	    !NIL_P(path = rb_str_normalize_ospath(name, namlen))) {
+	    path = rb_external_str_with_enc(path, dirp->enc);
+	}
+	else
+#endif
+	path = rb_external_str_new_with_enc(name, namlen, dirp->enc);
+	(*each)(arg, path);
 	if (dirp->dir == NULL) dir_closed();
     }
     return dir;
 }
 
+#ifdef HAVE_TELLDIR
 /*
  *  call-seq:
- *     dir.pos => integer
- *     dir.tell => integer
+ *     dir.pos -> integer
+ *     dir.tell -> integer
  *
  *  Returns the current position in <em>dir</em>. See also
  *  <code>Dir#seek</code>.
@@ -454,24 +879,23 @@ dir_each(dir)
  *     d.tell   #=> 12
  */
 static VALUE
-dir_tell(dir)
-    VALUE dir;
+dir_tell(VALUE dir)
 {
-#ifdef HAVE_TELLDIR
     struct dir_data *dirp;
     long pos;
 
     GetDIR(dir, dirp);
     pos = telldir(dirp->dir);
     return rb_int2inum(pos);
+}
 #else
-    rb_notimplement();
+#define dir_tell rb_f_notimplement
 #endif
-}
 
+#ifdef HAVE_SEEKDIR
 /*
  *  call-seq:
- *     dir.seek( integer ) => dir
+ *     dir.seek( integer ) -> dir
  *
  *  Seeks to a particular location in <em>dir</em>. <i>integer</i>
  *  must be a value returned by <code>Dir#tell</code>.
@@ -484,24 +908,23 @@ dir_tell(dir)
  *     d.read                   #=> ".."
  */
 static VALUE
-dir_seek(dir, pos)
-    VALUE dir, pos;
+dir_seek(VALUE dir, VALUE pos)
 {
     struct dir_data *dirp;
-    off_t p = NUM2OFFT(pos);
+    long p = NUM2LONG(pos);
 
     GetDIR(dir, dirp);
-#ifdef HAVE_SEEKDIR
     seekdir(dirp->dir, p);
     return dir;
+}
 #else
-    rb_notimplement();
+#define dir_seek rb_f_notimplement
 #endif
-}
 
+#ifdef HAVE_SEEKDIR
 /*
  *  call-seq:
- *     dir.pos( integer ) => integer
+ *     dir.pos = integer  -> integer
  *
  *  Synonym for <code>Dir#seek</code>, but returns the position
  *  parameter.
@@ -514,16 +937,18 @@ dir_seek(dir, pos)
  *     d.read                   #=> ".."
  */
 static VALUE
-dir_set_pos(dir, pos)
-    VALUE dir, pos;
+dir_set_pos(VALUE dir, VALUE pos)
 {
     dir_seek(dir, pos);
     return pos;
 }
+#else
+#define dir_set_pos rb_f_notimplement
+#endif
 
 /*
  *  call-seq:
- *     dir.rewind => dir
+ *     dir.rewind -> dir
  *
  *  Repositions <em>dir</em> to the first entry.
  *
@@ -533,8 +958,7 @@ dir_set_pos(dir, pos)
  *     d.read     #=> "."
  */
 static VALUE
-dir_rewind(dir)
-    VALUE dir;
+dir_rewind(VALUE dir)
 {
     struct dir_data *dirp;
 
@@ -545,36 +969,40 @@ dir_rewind(dir)
 
 /*
  *  call-seq:
- *     dir.close => nil
+ *     dir.close -> nil
  *
- *  Closes the directory stream. Any further attempts to access
- *  <em>dir</em> will raise an <code>IOError</code>.
+ *  Closes the directory stream.
+ *  Calling this method on closed Dir object is ignored since Ruby 2.3.
  *
  *     d = Dir.new("testdir")
  *     d.close   #=> nil
  */
 static VALUE
-dir_close(dir)
-    VALUE dir;
+dir_close(VALUE dir)
 {
     struct dir_data *dirp;
 
-    if (rb_safe_level() >= 4 && !OBJ_TAINTED(dir)) {
-	rb_raise(rb_eSecurityError, "Insecure: can't close");
-    }
-    GetDIR(dir, dirp);
+    dirp = dir_get(dir);
+    if (!dirp->dir) return Qnil;
     closedir(dirp->dir);
     dirp->dir = NULL;
 
     return Qnil;
 }
 
+static void *
+nogvl_chdir(void *ptr)
+{
+    const char *path = ptr;
+
+    return (void *)(VALUE)chdir(path);
+}
+
 static void
-dir_chdir(path)
-    VALUE path;
+dir_chdir(VALUE path)
 {
-    if (chdir(RSTRING(path)->ptr) < 0)
-	rb_sys_fail(RSTRING(path)->ptr);
+    if (chdir(RSTRING_PTR(path)) < 0)
+	rb_sys_fail_path(path);
 }
 
 static int chdir_blocking = 0;
@@ -586,11 +1014,10 @@ struct chdir_data {
 };
 
 static VALUE
-chdir_yield(args)
-    struct chdir_data *args;
+chdir_yield(struct chdir_data *args)
 {
     dir_chdir(args->new_path);
-    args->done = Qtrue;
+    args->done = TRUE;
     chdir_blocking++;
     if (chdir_thread == Qnil)
 	chdir_thread = rb_thread_current();
@@ -598,8 +1025,7 @@ chdir_yield(args)
 }
 
 static VALUE
-chdir_restore(args)
-    struct chdir_data *args;
+chdir_restore(struct chdir_data *args)
 {
     if (args->done) {
 	chdir_blocking--;
@@ -612,8 +1038,8 @@ chdir_restore(args)
 
 /*
  *  call-seq:
- *     Dir.chdir( [ string] ) => 0
- *     Dir.chdir( [ string] ) {| path | block }  => anObject
+ *     Dir.chdir( [ string] ) -> 0
+ *     Dir.chdir( [ string] ) {| path | block }  -> anObject
  *
  *  Changes the current working directory of the process to the given
  *  string. When called without an argument, changes the directory to
@@ -650,16 +1076,12 @@ chdir_restore(args)
  *     /var/spool/mail
  */
 static VALUE
-dir_s_chdir(argc, argv, obj)
-    int argc;
-    VALUE *argv;
-    VALUE obj;
+dir_s_chdir(int argc, VALUE *argv, VALUE obj)
 {
     VALUE path = Qnil;
 
-    rb_secure(2);
-    if (rb_scan_args(argc, argv, "01", &path) == 1) {
-	SafeStringValue(path);
+    if (rb_check_arity(argc, 0, 1) == 1) {
+        path = rb_str_encode_ospath(rb_get_path(argv[0]));
     }
     else {
 	const char *dist = getenv("HOME");
@@ -677,62 +1099,111 @@ dir_s_chdir(argc, argv, obj)
 
     if (rb_block_given_p()) {
 	struct chdir_data args;
-	char *cwd = my_getcwd();
 
-	args.old_path = rb_tainted_str_new2(cwd); free(cwd);
+	args.old_path = rb_str_encode_ospath(rb_dir_getwd());
 	args.new_path = path;
-	args.done = Qfalse;
+	args.done = FALSE;
 	return rb_ensure(chdir_yield, (VALUE)&args, chdir_restore, (VALUE)&args);
     }
-    dir_chdir(path);
+    else {
+	char *p = RSTRING_PTR(path);
+	int r = (int)(VALUE)rb_thread_call_without_gvl(nogvl_chdir, p,
+							RUBY_UBF_IO, 0);
+	if (r < 0)
+	    rb_sys_fail_path(path);
+    }
 
     return INT2FIX(0);
 }
 
+#ifndef _WIN32
+VALUE
+rb_dir_getwd_ospath(void)
+{
+    char *path;
+    VALUE cwd;
+    VALUE path_guard;
+
+#undef RUBY_UNTYPED_DATA_WARNING
+#define RUBY_UNTYPED_DATA_WARNING 0
+    path_guard = Data_Wrap_Struct((VALUE)0, NULL, RUBY_DEFAULT_FREE, NULL);
+    path = ruby_getcwd();
+    DATA_PTR(path_guard) = path;
+#ifdef __APPLE__
+    cwd = rb_str_normalize_ospath(path, strlen(path));
+    OBJ_TAINT(cwd);
+#else
+    cwd = rb_tainted_str_new2(path);
+#endif
+    DATA_PTR(path_guard) = 0;
+
+    xfree(path);
+    return cwd;
+}
+#endif
+
+VALUE
+rb_dir_getwd(void)
+{
+    rb_encoding *fs = rb_filesystem_encoding();
+    int fsenc = rb_enc_to_index(fs);
+    VALUE cwd = rb_dir_getwd_ospath();
+
+    switch (fsenc) {
+      case ENCINDEX_US_ASCII:
+	fsenc = ENCINDEX_ASCII;
+      case ENCINDEX_ASCII:
+	break;
+#if defined _WIN32 || defined __APPLE__
+      default:
+	return rb_str_conv_enc(cwd, NULL, fs);
+#endif
+    }
+    return rb_enc_associate_index(cwd, fsenc);
+}
+
 /*
  *  call-seq:
- *     Dir.getwd => string
- *     Dir.pwd => string
+ *     Dir.getwd -> string
+ *     Dir.pwd -> string
  *
  *  Returns the path to the current working directory of this process as
  *  a string.
  *
  *     Dir.chdir("/tmp")   #=> 0
  *     Dir.getwd           #=> "/tmp"
+ *     Dir.pwd             #=> "/tmp"
  */
 static VALUE
-dir_s_getwd(dir)
-    VALUE dir;
+dir_s_getwd(VALUE dir)
 {
-    char *path;
-    VALUE cwd;
-
-    rb_secure(4);
-    path = my_getcwd();
-    cwd = rb_tainted_str_new2(path);
-
-    free(path);
-    return cwd;
+    return rb_dir_getwd();
 }
 
-static void check_dirname _((volatile VALUE *));
-static void
-check_dirname(dir)
-    volatile VALUE *dir;
+static VALUE
+check_dirname(VALUE dir)
 {
+    VALUE d = dir;
     char *path, *pend;
-
-    SafeStringValue(*dir);
-    rb_secure(2);
-    path = RSTRING(*dir)->ptr;
-    if (path && *(pend = rb_path_end(rb_path_skip_prefix(path)))) {
-	*dir = rb_str_new(path, pend - path);
+    long len;
+    rb_encoding *enc;
+
+    FilePathValue(d);
+    enc = rb_enc_get(d);
+    RSTRING_GETMEM(d, path, len);
+    pend = path + len;
+    pend = rb_enc_path_end(rb_enc_path_skip_prefix(path, pend, enc), pend, enc);
+    if (pend - path < len) {
+	d = rb_str_subseq(d, 0, pend - path);
+	StringValueCStr(d);
     }
+    return rb_str_encode_ospath(d);
 }
 
+#if defined(HAVE_CHROOT)
 /*
  *  call-seq:
- *     Dir.chroot( string ) => 0
+ *     Dir.chroot( string ) -> 0
  *
  *  Changes this process's idea of the file system root. Only a
  *  privileged process may make this call. Not available on all
@@ -740,25 +1211,34 @@ check_dirname(dir)
  *  information.
  */
 static VALUE
-dir_s_chroot(dir, path)
-    VALUE dir, path;
+dir_s_chroot(VALUE dir, VALUE path)
 {
-#if defined(HAVE_CHROOT) && !defined(__CHECKER__)
-    check_dirname(&path);
-
-    if (chroot(RSTRING(path)->ptr) == -1)
-	rb_sys_fail(RSTRING(path)->ptr);
+    path = check_dirname(path);
+    if (chroot(RSTRING_PTR(path)) == -1)
+	rb_sys_fail_path(path);
 
     return INT2FIX(0);
+}
 #else
-    rb_notimplement();
-    return Qnil;		/* not reached */
+#define dir_s_chroot rb_f_notimplement
 #endif
+
+struct mkdir_arg {
+    const char *path;
+    mode_t mode;
+};
+
+static void *
+nogvl_mkdir(void *ptr)
+{
+    struct mkdir_arg *m = ptr;
+
+    return (void *)(VALUE)mkdir(m->path, m->mode);
 }
 
 /*
  *  call-seq:
- *     Dir.mkdir( string [, integer] ) => 0
+ *     Dir.mkdir( string [, integer] ) -> 0
  *
  *  Makes a new directory named by <i>string</i>, with permissions
  *  specified by the optional parameter <i>anInteger</i>. The
@@ -768,648 +1248,1583 @@ dir_s_chroot(dir, path)
  *  also the discussion of permissions in the class documentation for
  *  <code>File</code>.
  *
+ *    Dir.mkdir(File.join(Dir.home, ".foo"), 0700) #=> 0
+ *
  */
 static VALUE
-dir_s_mkdir(argc, argv, obj)
-    int argc;
-    VALUE *argv;
-    VALUE obj;
+dir_s_mkdir(int argc, VALUE *argv, VALUE obj)
 {
+    struct mkdir_arg m;
     VALUE path, vmode;
-    int mode;
+    int r;
 
     if (rb_scan_args(argc, argv, "11", &path, &vmode) == 2) {
-	mode = NUM2INT(vmode);
+	m.mode = NUM2MODET(vmode);
     }
     else {
-	mode = 0777;
+	m.mode = 0777;
     }
 
-    check_dirname(&path);
-    if (mkdir(RSTRING(path)->ptr, mode) == -1)
-	rb_sys_fail(RSTRING(path)->ptr);
+    path = check_dirname(path);
+    m.path = RSTRING_PTR(path);
+    r = (int)(VALUE)rb_thread_call_without_gvl(nogvl_mkdir, &m, RUBY_UBF_IO, 0);
+    if (r < 0)
+	rb_sys_fail_path(path);
 
     return INT2FIX(0);
 }
 
+static void *
+nogvl_rmdir(void *ptr)
+{
+    const char *path = ptr;
+
+    return (void *)(VALUE)rmdir(path);
+}
+
 /*
  *  call-seq:
- *     Dir.delete( string ) => 0
- *     Dir.rmdir( string ) => 0
- *     Dir.unlink( string ) => 0
+ *     Dir.delete( string ) -> 0
+ *     Dir.rmdir( string ) -> 0
+ *     Dir.unlink( string ) -> 0
  *
  *  Deletes the named directory. Raises a subclass of
  *  <code>SystemCallError</code> if the directory isn't empty.
  */
 static VALUE
-dir_s_rmdir(obj, dir)
-    VALUE obj, dir;
+dir_s_rmdir(VALUE obj, VALUE dir)
 {
-    check_dirname(&dir);
-    if (rmdir(RSTRING(dir)->ptr) < 0)
-	rb_sys_fail(RSTRING(dir)->ptr);
+    const char *p;
+    int r;
+
+    dir = check_dirname(dir);
+    p = RSTRING_PTR(dir);
+    r = (int)(VALUE)rb_thread_call_without_gvl(nogvl_rmdir, (void *)p, RUBY_UBF_IO, 0);
+    if (r < 0)
+	rb_sys_fail_path(dir);
 
     return INT2FIX(0);
 }
 
+struct warning_args {
+#ifdef RUBY_FUNCTION_NAME_STRING
+    const char *func;
+#endif
+    const char *mesg;
+    rb_encoding *enc;
+};
+
+#ifndef RUBY_FUNCTION_NAME_STRING
+#define sys_enc_warning_in(func, mesg, enc) sys_enc_warning(mesg, enc)
+#endif
+
+static VALUE
+sys_warning_1(VALUE mesg)
+{
+    const struct warning_args *arg = (struct warning_args *)mesg;
+#ifdef RUBY_FUNCTION_NAME_STRING
+    rb_sys_enc_warning(arg->enc, "%s: %s", arg->func, arg->mesg);
+#else
+    rb_sys_enc_warning(arg->enc, "%s", arg->mesg);
+#endif
+    return Qnil;
+}
+
 static void
-sys_warning_1(mesg)
-    const char* mesg;
+sys_enc_warning_in(const char *func, const char *mesg, rb_encoding *enc)
 {
-    rb_sys_warning("%s", mesg);
+    struct warning_args arg;
+#ifdef RUBY_FUNCTION_NAME_STRING
+    arg.func = func;
+#endif
+    arg.mesg = mesg;
+    arg.enc = enc;
+    rb_protect(sys_warning_1, (VALUE)&arg, 0);
 }
 
 #define GLOB_VERBOSE	(1U << (sizeof(int) * CHAR_BIT - 1))
-#define sys_warning(val) \
-    (void)((flags & GLOB_VERBOSE) && rb_protect((VALUE (*)_((VALUE)))sys_warning_1, (VALUE)(val), 0))
+#define sys_warning(val, enc) \
+    ((flags & GLOB_VERBOSE) ? sys_enc_warning_in(RUBY_FUNCTION_NAME_STRING, (val), (enc)) :(void)0)
 
-#define GLOB_ALLOC(type) (type *)malloc(sizeof(type))
-#define GLOB_ALLOC_N(type, n) (type *)malloc(sizeof(type) * (n))
-#define GLOB_REALLOC_N(var, type, n) (type *)realloc((var), sizeof(type) * (n))
-#define GLOB_JUMP_TAG(status) ((status == -1) ? rb_memerror() : rb_jump_tag(status))
+#define GLOB_ALLOC(type) ((type *)malloc(sizeof(type)))
+#define GLOB_ALLOC_N(type, n) ((type *)malloc(sizeof(type) * (n)))
+#define GLOB_REALLOC(ptr, size) realloc((ptr), (size))
+#define GLOB_FREE(ptr) free(ptr)
+#define GLOB_JUMP_TAG(status) (((status) == -1) ? rb_memerror() : rb_jump_tag(status))
 
-/* Return nonzero if S has any special globbing chars in it.  */
+/*
+ * ENOTDIR can be returned by stat(2) if a non-leaf element of the path
+ * is not a directory.
+ */
+ALWAYS_INLINE(static int to_be_ignored(int e));
+static inline int
+to_be_ignored(int e)
+{
+    return e == ENOENT || e == ENOTDIR;
+}
+
+#ifdef _WIN32
+#define STAT(p, s)	rb_w32_ustati128((p), (s))
+#undef lstat
+#define lstat(p, s)	rb_w32_ulstati128((p), (s))
+#else
+#define STAT(p, s)	stat((p), (s))
+#endif
+
+typedef int ruby_glob_errfunc(const char*, VALUE, const void*, int);
+typedef struct {
+    ruby_glob_func *match;
+    ruby_glob_errfunc *error;
+} ruby_glob_funcs_t;
+
+static const char *
+at_subpath(int fd, size_t baselen, const char *path)
+{
+#if USE_OPENDIR_AT
+    if (fd != (int)AT_FDCWD && baselen > 0) {
+	path += baselen;
+	if (*path == '/') ++path;
+    }
+#endif
+    return *path ? path : ".";
+}
+
+/* System call with warning */
 static int
-has_magic(s, send, flags)
-    const char *s, *send;
-    int flags;
+do_stat(int fd, size_t baselen, const char *path, struct stat *pst, int flags, rb_encoding *enc)
+{
+#if USE_OPENDIR_AT
+    int ret = fstatat(fd, at_subpath(fd, baselen, path), pst, 0);
+#else
+    int ret = STAT(path, pst);
+#endif
+    if (ret < 0 && !to_be_ignored(errno))
+	sys_warning(path, enc);
+
+    return ret;
+}
+
+#if defined HAVE_LSTAT || defined lstat || USE_OPENDIR_AT
+static int
+do_lstat(int fd, size_t baselen, const char *path, struct stat *pst, int flags, rb_encoding *enc)
+{
+#if USE_OPENDIR_AT
+    int ret = fstatat(fd, at_subpath(fd, baselen, path), pst, AT_SYMLINK_NOFOLLOW);
+#else
+    int ret = lstat(path, pst);
+#endif
+    if (ret < 0 && !to_be_ignored(errno))
+	sys_warning(path, enc);
+
+    return ret;
+}
+#else
+#define do_lstat do_stat
+#endif
+
+struct opendir_at_arg {
+    int basefd;
+    const char *path;
+};
+
+static void *
+with_gvl_gc_for_fd(void *ptr)
+{
+    int *e = ptr;
+
+    return (void *)(rb_gc_for_fd(*e) ? Qtrue : Qfalse);
+}
+
+static int
+gc_for_fd_with_gvl(int e)
+{
+    if (vm_initialized)
+	return (int)(VALUE)rb_thread_call_with_gvl(with_gvl_gc_for_fd, &e);
+    else
+	return rb_gc_for_fd(e) ? Qtrue : Qfalse;
+}
+
+static void *
+nogvl_opendir_at(void *ptr)
+{
+    const struct opendir_at_arg *oaa = ptr;
+    DIR *dirp;
+
+#if USE_OPENDIR_AT
+    const int opendir_flags = (O_RDONLY|O_CLOEXEC|
+#  ifdef O_DIRECTORY
+			       O_DIRECTORY|
+#  endif /* O_DIRECTORY */
+			       0);
+    int fd = openat(oaa->basefd, oaa->path, opendir_flags);
+
+    dirp = fd >= 0 ? fdopendir(fd) : 0;
+    if (!dirp) {
+	int e = errno;
+
+	switch (gc_for_fd_with_gvl(e)) {
+	  default:
+	    if (fd < 0) fd = openat(oaa->basefd, oaa->path, opendir_flags);
+	    if (fd >= 0) dirp = fdopendir(fd);
+	    if (dirp) return dirp;
+
+	    e = errno;
+	    /* fallthrough*/
+	  case 0:
+	    if (fd >= 0) close(fd);
+	    errno = e;
+	}
+    }
+#else  /* !USE_OPENDIR_AT */
+    dirp = opendir(oaa->path);
+    if (!dirp && gc_for_fd_with_gvl(errno))
+	dirp = opendir(oaa->path);
+#endif /* !USE_OPENDIR_AT */
+
+    return dirp;
+}
+
+static DIR *
+opendir_at(int basefd, const char *path)
+{
+    struct opendir_at_arg oaa;
+
+    oaa.basefd = basefd;
+    oaa.path = path;
+
+    if (vm_initialized)
+	return rb_thread_call_without_gvl(nogvl_opendir_at, &oaa, RUBY_UBF_IO, 0);
+    else
+	return nogvl_opendir_at(&oaa);
+}
+
+static DIR *
+do_opendir(const int basefd, size_t baselen, const char *path, int flags, rb_encoding *enc,
+	   ruby_glob_errfunc *errfunc, VALUE arg, int *status)
+{
+    DIR *dirp;
+#ifdef _WIN32
+    VALUE tmp = 0;
+    if (!fundamental_encoding_p(enc)) {
+	tmp = rb_enc_str_new(path, strlen(path), enc);
+	tmp = rb_str_encode_ospath(tmp);
+	path = RSTRING_PTR(tmp);
+    }
+#endif
+    dirp = opendir_at(basefd, at_subpath(basefd, baselen, path));
+    if (!dirp) {
+	int e = errno;
+
+	*status = 0;
+	if (!to_be_ignored(e)) {
+	    if (errfunc) {
+		*status = (*errfunc)(path, arg, enc, e);
+	    }
+	    else {
+		sys_warning(path, enc);
+	    }
+	}
+    }
+#ifdef _WIN32
+    if (tmp) rb_str_resize(tmp, 0); /* GC guard */
+#endif
+
+    return dirp;
+}
+
+/* Globing pattern */
+enum glob_pattern_type { PLAIN, ALPHA, BRACE, MAGICAL, RECURSIVE, MATCH_ALL, MATCH_DIR };
+
+/* Return nonzero if S has any special globbing chars in it.  */
+static enum glob_pattern_type
+has_magic(const char *p, const char *pend, int flags, rb_encoding *enc)
 {
-    register const char *p = s;
-    register char c;
-    int open = 0;
     const int escape = !(flags & FNM_NOESCAPE);
-    const int nocase = flags & FNM_CASEFOLD;
+    int hasalpha = 0;
+    int hasmagical = 0;
+
+    register char c;
 
-    while ((c = *p++) != '\0') {
+    while (p < pend && (c = *p++) != 0) {
 	switch (c) {
-	  case '?':
-	  case '*':
-	    return Qtrue;
+	  case '{':
+	    return BRACE;
 
-	  case '[':	/* Only accept an open brace if there is a close */
-	    open++;	/* brace to match it.  Bracket expressions must be */
-	    continue;	/* complete, according to Posix.2 */
-	  case ']':
-	    if (open)
-		return Qtrue;
-	    continue;
+	  case '*':
+	  case '?':
+	  case '[':
+	    hasmagical = 1;
+	    break;
 
 	  case '\\':
-	    if (escape && *p++ == '\0')
-		return Qfalse;
+	    if (escape && p++ >= pend)
+		continue;
 	    break;
 
+#ifdef _WIN32
+	  case '.':
+	    break;
+
+	  case '~':
+	    hasalpha = 1;
+	    break;
+#endif
 	  default:
-	    if (!FNM_SYSCASE && ISALPHA(c) && nocase)
-		return Qtrue;
+	    if (IS_WIN32 || ISALPHA(c)) {
+		hasalpha = 1;
+	    }
+	    break;
 	}
 
-	if (send && p >= send) break;
+	p = Next(p-1, pend, enc);
     }
-    return Qfalse;
+
+    return hasmagical ? MAGICAL : hasalpha ? ALPHA : PLAIN;
 }
 
-static char*
-extract_path(p, pend)
-    const char *p, *pend;
-{
-    char *alloc;
-    int len;
-
-    len = pend - p;
-    alloc = GLOB_ALLOC_N(char, len+1);
-    if (!alloc) return NULL;
-    memcpy(alloc, p, len);
-    if (len > 1 && pend[-1] == '/'
-#if defined DOSISH_DRIVE_LETTER
-    && pend[-2] != ':'
-#endif
-    ) {
-	alloc[len-1] = 0;
+/* Find separator in globbing pattern. */
+static char *
+find_dirsep(const char *p, const char *pend, int flags, rb_encoding *enc)
+{
+    const int escape = !(flags & FNM_NOESCAPE);
+
+    register char c;
+    int open = 0;
+
+    while ((c = *p++) != 0) {
+	switch (c) {
+	  case '[':
+	    open = 1;
+	    continue;
+	  case ']':
+	    open = 0;
+	    continue;
+
+	  case '{':
+	    open = 1;
+	    continue;
+	  case '}':
+	    open = 0;
+	    continue;
+
+	  case '/':
+	    if (!open)
+		return (char *)p-1;
+	    continue;
+
+	  case '\\':
+	    if (escape && !(c = *p++))
+		return (char *)p-1;
+	    continue;
+	}
+
+	p = Next(p-1, pend, enc);
     }
-    else {
-	alloc[len] = 0;
+
+    return (char *)p-1;
+}
+
+/* Remove escaping backslashes */
+static char *
+remove_backslashes(char *p, register const char *pend, rb_encoding *enc)
+{
+    char *t = p;
+    char *s = p;
+
+    while (*p) {
+	if (*p == '\\') {
+	    if (t != s)
+		memmove(t, s, p - s);
+	    t += p - s;
+	    s = ++p;
+	    if (!*p) break;
+	}
+	Inc(p, pend, enc);
     }
 
-    return alloc;
+    while (*p++);
+
+    if (t != s)
+	memmove(t, s, p - s); /* move '\0' too */
+
+    return p;
 }
 
-static char*
-extract_elem(path)
-    const char *path;
+struct glob_pattern {
+    char *str;
+    enum glob_pattern_type type;
+    struct glob_pattern *next;
+};
+
+static void glob_free_pattern(struct glob_pattern *list);
+
+static struct glob_pattern *
+glob_make_pattern(const char *p, const char *e, int flags, rb_encoding *enc)
 {
-    const char *pend;
+    struct glob_pattern *list, *tmp, **tail = &list;
+    int dirsep = 0; /* pattern is terminated with '/' */
+    int recursive = 0;
+
+    while (p < e && *p) {
+	tmp = GLOB_ALLOC(struct glob_pattern);
+	if (!tmp) goto error;
+	if (p + 2 < e && p[0] == '*' && p[1] == '*' && p[2] == '/') {
+	    /* fold continuous RECURSIVEs (needed in glob_helper) */
+	    do { p += 3; while (*p == '/') p++; } while (p[0] == '*' && p[1] == '*' && p[2] == '/');
+	    tmp->type = RECURSIVE;
+	    tmp->str = 0;
+	    dirsep = 1;
+	    recursive = 1;
+	}
+	else {
+	    const char *m = find_dirsep(p, e, flags, enc);
+	    const enum glob_pattern_type magic = has_magic(p, m, flags, enc);
+	    const enum glob_pattern_type non_magic = (USE_NAME_ON_FS || FNM_SYSCASE) ? PLAIN : ALPHA;
+	    char *buf;
+
+	    if (!(FNM_SYSCASE || magic > non_magic) && !recursive && *m) {
+		const char *m2;
+		while (has_magic(m+1, m2 = find_dirsep(m+1, e, flags, enc), flags, enc) <= non_magic &&
+		       *m2) {
+		    m = m2;
+		}
+	    }
+	    buf = GLOB_ALLOC_N(char, m-p+1);
+	    if (!buf) {
+		GLOB_FREE(tmp);
+		goto error;
+	    }
+	    memcpy(buf, p, m-p);
+	    buf[m-p] = '\0';
+	    tmp->type = magic > MAGICAL ? MAGICAL : magic > non_magic ? magic : PLAIN;
+	    tmp->str = buf;
+	    if (*m) {
+		dirsep = 1;
+		p = m + 1;
+	    }
+	    else {
+		dirsep = 0;
+		p = m;
+	    }
+	}
+	*tail = tmp;
+	tail = &tmp->next;
+    }
 
-    pend = strchr(path, '/');
-    if (!pend) pend = path + strlen(path);
+    tmp = GLOB_ALLOC(struct glob_pattern);
+    if (!tmp) {
+      error:
+	*tail = 0;
+	glob_free_pattern(list);
+	return 0;
+    }
+    tmp->type = dirsep ? MATCH_DIR : MATCH_ALL;
+    tmp->str = 0;
+    *tail = tmp;
+    tmp->next = 0;
 
-    return extract_path(path, pend);
+    return list;
 }
 
 static void
-remove_backslashes(p)
-    char *p;
+glob_free_pattern(struct glob_pattern *list)
 {
-    char *pend = p + strlen(p);
-    char *t = p;
+    while (list) {
+	struct glob_pattern *tmp = list;
+	list = list->next;
+	if (tmp->str)
+	    GLOB_FREE(tmp->str);
+	GLOB_FREE(tmp);
+    }
+}
 
-    while (p < pend) {
-	if (*p == '\\') {
-	    if (++p == pend) break;
+static char *
+join_path(const char *path, size_t len, int dirsep, const char *name, size_t namlen)
+{
+    char *buf = GLOB_ALLOC_N(char, len+namlen+(dirsep?1:0)+1);
+
+    if (!buf) return 0;
+    memcpy(buf, path, len);
+    if (dirsep) {
+	buf[len++] = '/';
+    }
+    memcpy(buf+len, name, namlen);
+    buf[len+namlen] = '\0';
+    return buf;
+}
+
+#ifdef HAVE_GETATTRLIST
+# if defined HAVE_FGETATTRLIST
+#   define is_case_sensitive(dirp, path) is_case_sensitive(dirp)
+# else
+#   define is_case_sensitive(dirp, path) is_case_sensitive(path)
+# endif
+static int
+is_case_sensitive(DIR *dirp, const char *path)
+{
+    struct {
+	u_int32_t length;
+	vol_capabilities_attr_t cap[1];
+    } __attribute__((aligned(4), packed)) attrbuf[1];
+    struct attrlist al = {ATTR_BIT_MAP_COUNT, 0, 0, ATTR_VOL_INFO|ATTR_VOL_CAPABILITIES};
+    const vol_capabilities_attr_t *const cap = attrbuf[0].cap;
+    const int idx = VOL_CAPABILITIES_FORMAT;
+    const uint32_t mask = VOL_CAP_FMT_CASE_SENSITIVE;
+
+#   if defined HAVE_FGETATTRLIST
+    if (fgetattrlist(dirfd(dirp), &al, attrbuf, sizeof(attrbuf), FSOPT_NOFOLLOW))
+	return -1;
+#   else
+    if (getattrlist(path, &al, attrbuf, sizeof(attrbuf), FSOPT_NOFOLLOW))
+	return -1;
+#   endif
+    if (!(cap->valid[idx] & mask))
+	return -1;
+    return (cap->capabilities[idx] & mask) != 0;
+}
+
+static char *
+replace_real_basename(char *path, long base, rb_encoding *enc, int norm_p, int flags, rb_pathtype_t *type)
+{
+    struct {
+	u_int32_t length;
+	attrreference_t ref[1];
+	fsobj_type_t objtype;
+	char path[MAXPATHLEN * 3];
+    } __attribute__((aligned(4), packed)) attrbuf[1];
+    struct attrlist al = {ATTR_BIT_MAP_COUNT, 0, ATTR_CMN_NAME|ATTR_CMN_OBJTYPE};
+    const attrreference_t *const ar = attrbuf[0].ref;
+    const char *name;
+    long len;
+    char *tmp;
+    IF_NORMALIZE_UTF8PATH(VALUE utf8str = Qnil);
+
+    *type = path_noent;
+    if (getattrlist(path, &al, attrbuf, sizeof(attrbuf), FSOPT_NOFOLLOW)) {
+	if (!to_be_ignored(errno))
+	    sys_warning(path, enc);
+	return path;
+    }
+
+    switch (attrbuf[0].objtype) {
+      case VREG: *type = path_regular; break;
+      case VDIR: *type = path_directory; break;
+      case VLNK: *type = path_symlink; break;
+      default: *type = path_exist; break;
+    }
+    name = (char *)ar + ar->attr_dataoffset;
+    len = (long)ar->attr_length - 1;
+    if (name + len > (char *)attrbuf + sizeof(attrbuf))
+	return path;
+
+# if NORMALIZE_UTF8PATH
+    if (norm_p && has_nonascii(name, len)) {
+	if (!NIL_P(utf8str = rb_str_normalize_ospath(name, len))) {
+	    RSTRING_GETMEM(utf8str, name, len);
 	}
-	*t++ = *p++;
     }
-    *t = '\0';
+# endif
+
+    tmp = GLOB_REALLOC(path, base + len + 1);
+    if (tmp) {
+	path = tmp;
+	memcpy(path + base, name, len);
+	path[base + len] = '\0';
+    }
+    IF_NORMALIZE_UTF8PATH(if (!NIL_P(utf8str)) rb_str_resize(utf8str, 0));
+    return path;
 }
+#elif defined _WIN32
+VALUE rb_w32_conv_from_wchar(const WCHAR *wstr, rb_encoding *enc);
+int rb_w32_reparse_symlink_p(const WCHAR *path);
+
+static char *
+replace_real_basename(char *path, long base, rb_encoding *enc, int norm_p, int flags, rb_pathtype_t *type)
+{
+    char *plainname = path;
+    volatile VALUE tmp = 0;
+    WIN32_FIND_DATAW fd;
+    WIN32_FILE_ATTRIBUTE_DATA fa;
+    WCHAR *wplain;
+    HANDLE h = INVALID_HANDLE_VALUE;
+    long wlen;
+    int e = 0;
+    if (!fundamental_encoding_p(enc)) {
+	tmp = rb_enc_str_new_cstr(plainname, enc);
+	tmp = rb_str_encode_ospath(tmp);
+	plainname = RSTRING_PTR(tmp);
+    }
+    wplain = rb_w32_mbstr_to_wstr(CP_UTF8, plainname, -1, &wlen);
+    if (tmp) rb_str_resize(tmp, 0);
+    if (!wplain) return path;
+    if (GetFileAttributesExW(wplain, GetFileExInfoStandard, &fa)) {
+	h = FindFirstFileW(wplain, &fd);
+	e = rb_w32_map_errno(GetLastError());
+    }
+    if (fa.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) {
+	if (!rb_w32_reparse_symlink_p(wplain))
+	    fa.dwFileAttributes &= ~FILE_ATTRIBUTE_REPARSE_POINT;
+    }
+    free(wplain);
+    if (h == INVALID_HANDLE_VALUE) {
+	*type = path_noent;
+	if (e && !to_be_ignored(e)) {
+	    errno = e;
+	    sys_warning(path, enc);
+	}
+	return path;
+    }
+    FindClose(h);
+    *type =
+	(fa.dwFileAttributes & FILE_ATTRIBUTE_REPARSE_POINT) ? path_symlink :
+	(fa.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) ? path_directory :
+	path_regular;
+    if (tmp) {
+	char *buf;
+	tmp = rb_w32_conv_from_wchar(fd.cFileName, enc);
+	wlen = RSTRING_LEN(tmp);
+	buf = GLOB_REALLOC(path, base + wlen + 1);
+	if (buf) {
+	    path = buf;
+	    memcpy(path + base, RSTRING_PTR(tmp), wlen);
+	    path[base + wlen] = 0;
+	}
+	rb_str_resize(tmp, 0);
+    }
+    else {
+	char *utf8filename;
+	wlen = WideCharToMultiByte(CP_UTF8, 0, fd.cFileName, -1, NULL, 0, NULL, NULL);
+	utf8filename = GLOB_REALLOC(0, wlen);
+	if (utf8filename) {
+	    char *buf;
+	    WideCharToMultiByte(CP_UTF8, 0, fd.cFileName, -1, utf8filename, wlen, NULL, NULL);
+	    buf = GLOB_REALLOC(path, base + wlen + 1);
+	    if (buf) {
+		path = buf;
+		memcpy(path + base, utf8filename, wlen);
+		path[base + wlen] = 0;
+	    }
+	    GLOB_FREE(utf8filename);
+	}
+    }
+    return path;
+}
+#elif USE_NAME_ON_FS == USE_NAME_ON_FS_REAL_BASENAME
+# error not implemented
+#endif
 
 #ifndef S_ISDIR
-#   define S_ISDIR(m) ((m & S_IFMT) == S_IFDIR)
+#   define S_ISDIR(m) (((m) & S_IFMT) == S_IFDIR)
+#endif
+
+#ifndef S_ISLNK
+#  ifndef S_IFLNK
+#    define S_ISLNK(m) (0)
+#  else
+#    define S_ISLNK(m) (((m) & S_IFMT) == S_IFLNK)
+#  endif
 #endif
 
 struct glob_args {
-    void (*func) _((const char*, VALUE));
-    const char *c;
-    VALUE v;
+    void (*func)(const char *, VALUE, void *);
+    const char *path;
+    const char *base;
+    size_t baselen;
+    VALUE value;
+    rb_encoding *enc;
 };
 
-static VALUE glob_func_caller _((VALUE));
+#define glob_call_func(func, path, arg, enc) (*(func))((path), (arg), (void *)(enc))
 
 static VALUE
-glob_func_caller(val)
-    VALUE val;
+glob_func_caller(VALUE val)
 {
     struct glob_args *args = (struct glob_args *)val;
 
-    (*args->func)(args->c, args->v);
+    glob_call_func(args->func, args->path, args->value, args->enc);
+    return Qnil;
+}
+
+struct glob_error_args {
+    const char *path;
+    rb_encoding *enc;
+    int error;
+};
+
+static VALUE
+glob_func_warning(VALUE val)
+{
+    struct glob_error_args *arg = (struct glob_error_args *)val;
+    rb_syserr_enc_warning(arg->error, arg->enc, "%s", arg->path);
     return Qnil;
 }
 
-#define glob_call_func(func, path, arg) (*func)(path, arg)
+#if 0
+static int
+rb_glob_warning(const char *path, VALUE a, const void *enc, int error)
+{
+    int status;
+    struct glob_error_args args;
 
-static int glob_helper _((const char *path, const char *sub, int flags, int (*func)(const char *,VALUE), VALUE arg));
+    args.path = path;
+    args.enc = enc;
+    args.error = error;
+    rb_protect(glob_func_warning, (VALUE)&args, &status);
+    return status;
+}
+#endif
+
+static VALUE
+glob_func_error(VALUE val)
+{
+    struct glob_error_args *arg = (struct glob_error_args *)val;
+    VALUE path = rb_enc_str_new_cstr(arg->path, arg->enc);
+    rb_syserr_fail_str(arg->error, path);
+    return Qnil;
+}
 
 static int
-glob_helper(path, sub, flags, func, arg)
+rb_glob_error(const char *path, VALUE a, const void *enc, int error)
+{
+    int status;
+    struct glob_error_args args;
+    VALUE (*errfunc)(VALUE) = glob_func_error;
+
+    if (error == EACCES) {
+	errfunc = glob_func_warning;
+    }
+    args.path = path;
+    args.enc = enc;
+    args.error = error;
+    rb_protect(errfunc, (VALUE)&args, &status);
+    return status;
+}
+
+static inline int
+dirent_match(const char *pat, rb_encoding *enc, const char *name, const struct dirent *dp, int flags)
+{
+    if (fnmatch(pat, enc, name, flags) == 0) return 1;
+#ifdef _WIN32
+    if (dp->d_altname && (flags & FNM_SHORTNAME)) {
+	if (fnmatch(pat, enc, dp->d_altname, flags) == 0) return 1;
+    }
+#endif
+    return 0;
+}
+
+struct push_glob_args {
+    int fd;
     const char *path;
-    const char *sub;
+    size_t baselen;
+    size_t namelen;
+    int dirsep; /* '/' should be placed before appending child entry's name to 'path'. */
+    rb_pathtype_t pathtype; /* type of 'path' */
     int flags;
-    int (*func) _((const char *, VALUE));
+    const ruby_glob_funcs_t *funcs;
     VALUE arg;
+};
+
+struct dirent_brace_args {
+    const char *name;
+    const struct dirent *dp;
+    int flags;
+};
+
+static int
+dirent_match_brace(const char *pattern, VALUE val, void *enc)
+{
+    struct dirent_brace_args *arg = (struct dirent_brace_args *)val;
+
+    return dirent_match(pattern, enc, arg->name, arg->dp, arg->flags);
+}
+
+/* join paths from pattern list of glob_make_pattern() */
+static char*
+join_path_from_pattern(struct glob_pattern **beg)
+{
+    struct glob_pattern *p;
+    char *path = NULL;
+    size_t path_len = 0;
+
+    for (p = *beg; p; p = p->next) {
+	const char *str;
+	switch (p->type) {
+	  case RECURSIVE:
+	    str = "**";
+	    break;
+	  default:
+	    str = p->str;
+	    if (!str) continue;
+	}
+	if (!path) {
+	    path_len = strlen(str);
+	    path = GLOB_ALLOC_N(char, path_len + 1);
+	    memcpy(path, str, path_len);
+	    path[path_len] = '\0';
+	} else {
+	    size_t len = strlen(str);
+	    char *tmp;
+	    tmp = GLOB_REALLOC(path, path_len + len + 2);
+	    if (tmp) {
+		path = tmp;
+		path[path_len++] = '/';
+		memcpy(path + path_len, str, len);
+		path_len += len;
+		path[path_len] = '\0';
+	    }
+	}
+    }
+    return path;
+}
+
+static int push_caller(const char *path, VALUE val, void *enc);
+
+static int ruby_brace_expand(const char *str, int flags, ruby_glob_func *func, VALUE arg,
+			     rb_encoding *enc, VALUE var);
+
+static int
+glob_helper(
+    int fd,
+    const char *path,
+    size_t baselen,
+    size_t namelen,
+    int dirsep, /* '/' should be placed before appending child entry's name to 'path'. */
+    rb_pathtype_t pathtype, /* type of 'path' */
+    struct glob_pattern **beg,
+    struct glob_pattern **end,
+    int flags,
+    const ruby_glob_funcs_t *funcs,
+    VALUE arg,
+    rb_encoding *enc)
 {
     struct stat st;
-    const char *p, *m;
     int status = 0;
-    char *buf = 0;
-    char *newpath = 0;
-    char *newbuf;
-
-    p = sub ? sub : path;
-    if (!has_magic(p, 0, flags)) {
-#if !defined DOSISH
-	if (!(flags & FNM_NOESCAPE))
+    struct glob_pattern **cur, **new_beg, **new_end;
+    int plain = 0, brace = 0, magical = 0, recursive = 0, match_all = 0, match_dir = 0;
+    int escape = !(flags & FNM_NOESCAPE);
+    size_t pathlen = baselen + namelen;
+
+    for (cur = beg; cur < end; ++cur) {
+	struct glob_pattern *p = *cur;
+	if (p->type == RECURSIVE) {
+	    recursive = 1;
+	    p = p->next;
+	}
+	switch (p->type) {
+	  case PLAIN:
+	    plain = 1;
+	    break;
+	  case ALPHA:
+#if USE_NAME_ON_FS == USE_NAME_ON_FS_REAL_BASENAME
+	    plain = 1;
+#else
+	    magical = 1;
 #endif
-	{
-	    newpath = strdup(path);
-	    if (!newpath) return -1;
-	    if (sub) {
-		p = newpath + (sub - path);
-		remove_backslashes(newpath + (sub - path));
-		sub = p;
+	    break;
+	  case BRACE:
+	    if (!recursive) {
+		brace = 1;
+	    }
+	    break;
+	  case MAGICAL:
+	    magical = 2;
+	    break;
+	  case MATCH_ALL:
+	    match_all = 1;
+	    break;
+	  case MATCH_DIR:
+	    match_dir = 1;
+	    break;
+	  case RECURSIVE:
+	    rb_bug("continuous RECURSIVEs");
+	}
+    }
+
+    if (brace) {
+	struct push_glob_args args;
+	char* brace_path = join_path_from_pattern(beg);
+	if (!brace_path) return -1;
+	args.fd = fd;
+	args.path = path;
+	args.baselen = baselen;
+	args.namelen = namelen;
+	args.dirsep = dirsep;
+	args.pathtype = pathtype;
+	args.flags = flags;
+	args.funcs = funcs;
+	args.arg = arg;
+	status = ruby_brace_expand(brace_path, flags, push_caller, (VALUE)&args, enc, Qfalse);
+	GLOB_FREE(brace_path);
+	return status;
+    }
+
+    if (*path) {
+	if (match_all && pathtype == path_unknown) {
+	    if (do_lstat(fd, baselen, path, &st, flags, enc) == 0) {
+		pathtype = IFTODT(st.st_mode);
+	    }
+	    else {
+		pathtype = path_noent;
+	    }
+	}
+	if (match_dir && (pathtype == path_unknown || pathtype == path_symlink)) {
+	    if (do_stat(fd, baselen, path, &st, flags, enc) == 0) {
+		pathtype = IFTODT(st.st_mode);
 	    }
 	    else {
-		remove_backslashes(newpath);
-		p = path = newpath;
+		pathtype = path_noent;
 	    }
 	}
-	if (lstat(path, &st) == 0) {
-	    status = glob_call_func(func, path, arg);
+	if (match_all && pathtype > path_noent) {
+	    const char *subpath = path + baselen + (baselen && path[baselen] == '/');
+	    status = glob_call_func(funcs->match, subpath, arg, enc);
+	    if (status) return status;
 	}
-	else if (errno != ENOENT) {
-	    /* In case stat error is other than ENOENT and
-	       we may want to know what is wrong. */
-	    sys_warning(path);
+	if (match_dir && pathtype == path_directory) {
+	    int seplen = (baselen && path[baselen] == '/');
+	    const char *subpath = path + baselen + seplen;
+	    char *tmp = join_path(subpath, namelen - seplen, dirsep, "", 0);
+	    if (!tmp) return -1;
+	    status = glob_call_func(funcs->match, tmp, arg, enc);
+	    GLOB_FREE(tmp);
+	    if (status) return status;
 	}
-	if (newpath) free(newpath);
-	return status;
     }
 
-    while (p && !status) {
-	if (*p == '/') p++;
-	m = strchr(p, '/');
-	if (has_magic(p, m, flags)) {
-	    char *dir, *base, *magic;
-	    DIR *dirp;
-	    struct dirent *dp;
-	    int recursive = 0;
+    if (pathtype == path_noent) return 0;
 
-	    struct d_link {
-		char *path;
-		struct d_link *next;
-	    } *tmp, *link, **tail = &link;
+    if (magical || recursive) {
+	struct dirent *dp;
+	DIR *dirp;
+# if USE_NAME_ON_FS == USE_NAME_ON_FS_BY_FNMATCH
+	char *plainname = 0;
+# endif
+	IF_NORMALIZE_UTF8PATH(int norm_p);
+# if USE_NAME_ON_FS == USE_NAME_ON_FS_BY_FNMATCH
+	if (cur + 1 == end && (*cur)->type <= ALPHA) {
+	    plainname = join_path(path, pathlen, dirsep, (*cur)->str, strlen((*cur)->str));
+	    if (!plainname) return -1;
+	    dirp = do_opendir(fd, basename, plainname, flags, enc, funcs->error, arg, &status);
+	    GLOB_FREE(plainname);
+	}
+	else
+# else
+	    ;
+# endif
+	dirp = do_opendir(fd, baselen, path, flags, enc, funcs->error, arg, &status);
+	if (dirp == NULL) {
+# if FNM_SYSCASE || NORMALIZE_UTF8PATH
+	    if ((magical < 2) && !recursive && (errno == EACCES)) {
+		/* no read permission, fallback */
+		goto literally;
+	    }
+# endif
+	    return status;
+	}
+	IF_NORMALIZE_UTF8PATH(norm_p = need_normalization(dirp, *path ? path : "."));
 
-	    base = extract_path(path, p);
-	    if (!base) {
-		status = -1;
-		break;
+# if NORMALIZE_UTF8PATH
+	if (!(norm_p || magical || recursive)) {
+	    closedir(dirp);
+	    goto literally;
+	}
+# endif
+# ifdef HAVE_GETATTRLIST
+	if (is_case_sensitive(dirp, path) == 0)
+	    flags |= FNM_CASEFOLD;
+# endif
+	while ((dp = READDIR(dirp, enc)) != NULL) {
+	    char *buf;
+	    rb_pathtype_t new_pathtype = path_unknown;
+	    const char *name;
+	    size_t namlen;
+	    int dotfile = 0;
+	    IF_NORMALIZE_UTF8PATH(VALUE utf8str = Qnil);
+
+	    name = dp->d_name;
+	    namlen = NAMLEN(dp);
+	    if (recursive && name[0] == '.') {
+		++dotfile;
+		if (namlen == 1) {
+		    /* unless DOTMATCH, skip current directories not to recurse infinitely */
+		    if (!(flags & FNM_DOTMATCH)) continue;
+		    ++dotfile;
+		    new_pathtype = path_directory; /* force to skip stat/lstat */
+		}
+		else if (namlen == 2 && name[1] == '.') {
+		    /* always skip parent directories not to recurse infinitely */
+		    continue;
+		}
 	    }
-	    if (path == p) dir = ".";
-	    else dir = base;
 
-	    magic = extract_elem(p);
-	    if (!magic) {
+# if NORMALIZE_UTF8PATH
+	    if (norm_p && has_nonascii(name, namlen)) {
+		if (!NIL_P(utf8str = rb_str_normalize_ospath(name, namlen))) {
+		    RSTRING_GETMEM(utf8str, name, namlen);
+		}
+	    }
+# endif
+	    buf = join_path(path, pathlen, dirsep, name, namlen);
+	    IF_NORMALIZE_UTF8PATH(if (!NIL_P(utf8str)) rb_str_resize(utf8str, 0));
+	    if (!buf) {
 		status = -1;
 		break;
 	    }
-	    if (stat(dir, &st) < 0) {
-	        if (errno != ENOENT)
-		    sys_warning(dir);
-	        free(base);
-	        free(magic);
-	        break;
+	    name = buf + pathlen + (dirsep != 0);
+#ifdef DT_UNKNOWN
+	    if (dp->d_type != DT_UNKNOWN) {
+		/* Got it. We need no more lstat. */
+		new_pathtype = dp->d_type;
 	    }
-	    if (S_ISDIR(st.st_mode)) {
-		if (m && strcmp(magic, "**") == 0) {
-		    int n = strlen(base);
-		    recursive = 1;
-		    newbuf = GLOB_REALLOC_N(buf, char, n+strlen(m)+3);
-		    if (!newbuf) {
-			status = -1;
-			goto finalize;
-		    }
-		    buf = newbuf;
-		    sprintf(buf, "%s%s", base, *base ? m : m+1);
-		    status = glob_helper(buf, buf+n, flags, func, arg);
-		    if (status) goto finalize;
-		}
-		dirp = opendir(dir);
-		if (dirp == NULL) {
-		    sys_warning(dir);
-		    free(base);
-		    free(magic);
-		    break;
-		}
+#endif
+	    if (recursive && dotfile < ((flags & FNM_DOTMATCH) ? 2 : 1) &&
+		new_pathtype == path_unknown) {
+		/* RECURSIVE never match dot files unless FNM_DOTMATCH is set */
+		if (do_lstat(fd, baselen, buf, &st, flags, enc) == 0)
+		    new_pathtype = IFTODT(st.st_mode);
+		else
+		    new_pathtype = path_noent;
 	    }
-	    else {
-		free(base);
-		free(magic);
+
+	    new_beg = new_end = GLOB_ALLOC_N(struct glob_pattern *, (end - beg) * 2);
+	    if (!new_beg) {
+		GLOB_FREE(buf);
+		status = -1;
 		break;
 	    }
 
-#if defined DOSISH_DRIVE_LETTER
-#define BASE (*base && !((isdirsep(*base) && !base[1]) || (base[1] == ':' && isdirsep(base[2]) && !base[3])))
-#else
-#define BASE (*base && !(isdirsep(*base) && !base[1]))
-#endif
-
-	    for (dp = readdir(dirp); dp != NULL; dp = readdir(dirp)) {
-		if (recursive) {
-		    if (strcmp(".", dp->d_name) == 0 || strcmp("..", dp->d_name) == 0)
-			continue;
-		    if (fnmatch("*", dp->d_name, flags) != 0)
-			continue;
-		    newbuf = GLOB_REALLOC_N(buf, char, strlen(base)+NAMLEN(dp)+strlen(m)+6);
-		    if (!newbuf) {
-			status = -1;
-			break;
-		    }
-		    buf = newbuf;
-		    sprintf(buf, "%s%s%s", base, (BASE) ? "/" : "", dp->d_name);
-		    if (lstat(buf, &st) < 0) {
-			if (errno != ENOENT)
-			    sys_warning(buf);
-			continue;
+	    for (cur = beg; cur < end; ++cur) {
+		struct glob_pattern *p = *cur;
+		struct dirent_brace_args args;
+		if (p->type == RECURSIVE) {
+		    if (new_pathtype == path_directory || /* not symlink but real directory */
+			new_pathtype == path_exist) {
+			if (dotfile < ((flags & FNM_DOTMATCH) ? 2 : 1))
+			    *new_end++ = p; /* append recursive pattern */
 		    }
-		    if (S_ISDIR(st.st_mode)) {
-			char *t = buf+strlen(buf);
-		        strcpy(t, "/**");
-			strcpy(t+3, m);
-			status = glob_helper(buf, t, flags, func, arg);
-			if (status) break;
-			continue;
-		    }
-		    continue;
+		    p = p->next; /* 0 times recursion */
 		}
-		if (fnmatch(magic, dp->d_name, flags) == 0) {
-		    newbuf = GLOB_REALLOC_N(buf, char, strlen(base)+NAMLEN(dp)+2);
-		    if (!newbuf) {
-			status = -1;
-			break;
-		    }
-		    buf = newbuf;
-		    sprintf(buf, "%s%s%s", base, (BASE) ? "/" : "", dp->d_name);
-		    if (!m) {
-			status = glob_call_func(func, buf, arg);
-			if (status) break;
-			continue;
-		    }
-		    tmp = GLOB_ALLOC(struct d_link);
-		    if (!tmp) {
-			status = -1;
+		switch (p->type) {
+		  case BRACE:
+		    args.name = name;
+		    args.dp = dp;
+		    args.flags = flags;
+		    if (ruby_brace_expand(p->str, flags, dirent_match_brace,
+					  (VALUE)&args, enc, Qfalse) > 0)
+			*new_end++ = p->next;
+		    break;
+		  case ALPHA:
+# if USE_NAME_ON_FS == USE_NAME_ON_FS_BY_FNMATCH
+		    if (plainname) {
+			*new_end++ = p->next;
 			break;
 		    }
-		    tmp->path = buf;
-		    buf = 0;
-		    *tail = tmp;
-		    tail = &tmp->next;
+# endif
+		  case PLAIN:
+		  case MAGICAL:
+		    if (dirent_match(p->str, enc, name, dp, flags))
+			*new_end++ = p->next;
+		  default:
+		    break;
 		}
 	    }
-	    closedir(dirp);
-	  finalize:
-	    *tail = 0;
-	    free(base);
-	    free(magic);
-	    if (link) {
-		while (link) {
-		    if (status == 0) {
-			if (stat(link->path, &st) == 0) {
-			    if (S_ISDIR(st.st_mode)) {
-				int len = strlen(link->path);
-				int mlen = strlen(m);
-
-				newbuf = GLOB_REALLOC_N(buf, char, len+mlen+1);
-				if (!newbuf) {
-				    status = -1;
-				    goto next_elem;
-				}
-				buf = newbuf;
-				sprintf(buf, "%s%s", link->path, m);
-				status = glob_helper(buf, buf+len, flags, func, arg);
-			    }
-			}
-			else {
-			    sys_warning(link->path);
-			}
+
+	    status = glob_helper(fd, buf, baselen, name - buf - baselen + namlen, 1,
+				 new_pathtype, new_beg, new_end,
+				 flags, funcs, arg, enc);
+	    GLOB_FREE(buf);
+	    GLOB_FREE(new_beg);
+	    if (status) break;
+	}
+
+	closedir(dirp);
+    }
+    else if (plain) {
+	struct glob_pattern **copy_beg, **copy_end, **cur2;
+
+# if FNM_SYSCASE || NORMALIZE_UTF8PATH
+      literally:
+# endif
+	copy_beg = copy_end = GLOB_ALLOC_N(struct glob_pattern *, end - beg);
+	if (!copy_beg) return -1;
+	for (cur = beg; cur < end; ++cur)
+	    *copy_end++ = (*cur)->type <= ALPHA ? *cur : 0;
+
+	for (cur = copy_beg; cur < copy_end; ++cur) {
+	    if (*cur) {
+		rb_pathtype_t new_pathtype = path_unknown;
+		char *buf;
+		char *name;
+		size_t len = strlen((*cur)->str) + 1;
+		name = GLOB_ALLOC_N(char, len);
+		if (!name) {
+		    status = -1;
+		    break;
+		}
+		memcpy(name, (*cur)->str, len);
+		if (escape)
+		    len = remove_backslashes(name, name+len-1, enc) - name;
+
+		new_beg = new_end = GLOB_ALLOC_N(struct glob_pattern *, end - beg);
+		if (!new_beg) {
+		    GLOB_FREE(name);
+		    status = -1;
+		    break;
+		}
+		*new_end++ = (*cur)->next;
+		for (cur2 = cur + 1; cur2 < copy_end; ++cur2) {
+		    if (*cur2 && fnmatch((*cur2)->str, enc, name, flags) == 0) {
+			*new_end++ = (*cur2)->next;
+			*cur2 = 0;
 		    }
-		  next_elem:
-		    tmp = link;
-		    link = link->next;
-		    free(tmp->path);
-		    free(tmp);
 		}
-		break;
+
+		buf = join_path(path, pathlen, dirsep, name, len);
+		GLOB_FREE(name);
+		if (!buf) {
+		    GLOB_FREE(new_beg);
+		    status = -1;
+		    break;
+		}
+#if USE_NAME_ON_FS == USE_NAME_ON_FS_REAL_BASENAME
+		if ((*cur)->type == ALPHA) {
+		    buf = replace_real_basename(buf, pathlen + (dirsep != 0), enc,
+						IF_NORMALIZE_UTF8PATH(1)+0,
+						flags, &new_pathtype);
+		    if (!buf) break;
+		}
+#endif
+		status = glob_helper(fd, buf, baselen,
+				     namelen + strlen(buf + pathlen), 1,
+				     new_pathtype, new_beg, new_end,
+				     flags, funcs, arg, enc);
+		GLOB_FREE(buf);
+		GLOB_FREE(new_beg);
+		if (status) break;
 	    }
 	}
-	p = m;
+
+	GLOB_FREE(copy_beg);
     }
-    if (buf) free(buf);
-    if (newpath) free(newpath);
+
     return status;
 }
 
-int
-ruby_glob(path, flags, func, arg)
-    const char *path;
-    int flags;
-    int (*func) _((const char *, VALUE));
-    VALUE arg;
+static int
+push_caller(const char *path, VALUE val, void *enc)
 {
+    struct push_glob_args *arg = (struct push_glob_args *)val;
+    struct glob_pattern *list;
+    int status;
+
+    list = glob_make_pattern(path, path + strlen(path), arg->flags, enc);
+    if (!list) {
+	return -1;
+    }
+    status = glob_helper(arg->fd, arg->path, arg->baselen, arg->namelen, arg->dirsep,
+			 arg->pathtype, &list, &list + 1, arg->flags, arg->funcs,
+			 arg->arg, enc);
+    glob_free_pattern(list);
+    return status;
+}
+
+static int
+ruby_glob0(const char *path, int fd, const char *base, int flags,
+	   const ruby_glob_funcs_t *funcs, VALUE arg,
+	   rb_encoding *enc)
+{
+    struct glob_pattern *list;
+    const char *root, *start;
+    char *buf;
+    size_t n, baselen = 0;
+    int status, dirsep = FALSE;
+
+    start = root = path;
     flags |= FNM_SYSCASE;
-    return glob_helper(path, 0, flags & ~GLOB_VERBOSE, func, arg);
+#if defined DOSISH
+    root = rb_enc_path_skip_prefix(root, root + strlen(root), enc);
+#endif
+
+    if (*root == '/') root++;
+
+    n = root - start;
+    if (!n && base) {
+	n = strlen(base);
+	baselen = n;
+	start = base;
+	dirsep = TRUE;
+    }
+    buf = GLOB_ALLOC_N(char, n + 1);
+    if (!buf) return -1;
+    MEMCPY(buf, start, char, n);
+    buf[n] = '\0';
+
+    list = glob_make_pattern(root, root + strlen(root), flags, enc);
+    if (!list) {
+	GLOB_FREE(buf);
+	return -1;
+    }
+    status = glob_helper(fd, buf, baselen, n-baselen, dirsep,
+			 path_unknown, &list, &list + 1,
+			 flags, funcs, arg, enc);
+    glob_free_pattern(list);
+    GLOB_FREE(buf);
+
+    return status;
 }
 
 int
-ruby_globi(path, flags, func, arg)
-    const char *path;
-    int flags;
-    int (*func) _((const char *, VALUE));
-    VALUE arg;
+ruby_glob(const char *path, int flags, ruby_glob_func *func, VALUE arg)
 {
-    return glob_helper(path, 0, flags | FNM_CASEFOLD, func, arg);
+    ruby_glob_funcs_t funcs;
+    funcs.match = func;
+    funcs.error = NULL;
+    return ruby_glob0(path, AT_FDCWD, 0, flags & ~GLOB_VERBOSE,
+		      &funcs, arg, rb_ascii8bit_encoding());
 }
 
-static int rb_glob_caller _((const char *, VALUE));
-
 static int
-rb_glob_caller(path, a)
-    const char *path;
-    VALUE a;
+rb_glob_caller(const char *path, VALUE a, void *enc)
 {
     int status;
     struct glob_args *args = (struct glob_args *)a;
 
-    args->c = path;
+    args->path = path;
     rb_protect(glob_func_caller, a, &status);
     return status;
 }
 
-static int
-rb_glob2(path, flags, func, arg)
-    const char *path;
-    int flags;
-    void (*func) _((const char *, VALUE));
-    VALUE arg;
+static const ruby_glob_funcs_t rb_glob_funcs = {
+    rb_glob_caller, rb_glob_error,
+};
+
+void
+rb_glob(const char *path, void (*func)(const char *, VALUE, void *), VALUE arg)
 {
     struct glob_args args;
+    int status;
 
     args.func = func;
-    args.v = arg;
-
-    flags |= FNM_SYSCASE;
-    return glob_helper(path, 0, flags | GLOB_VERBOSE, rb_glob_caller, (VALUE)&args);
-}
+    args.value = arg;
+    args.enc = rb_ascii8bit_encoding();
 
-void
-rb_glob(path, func, arg)
-    const char *path;
-    void (*func) _((const char*, VALUE));
-    VALUE arg;
-{
-    int status = rb_glob2(path, 0, func, arg);
-    if (status) rb_jump_tag(status);
-}
-
-void
-rb_globi(path, func, arg)
-    const char *path;
-    void (*func) _((const char*, VALUE));
-    VALUE arg;
-{
-    int status = rb_glob2(path, FNM_CASEFOLD, func, arg);
-    if (status) rb_jump_tag(status);
+    status = ruby_glob0(path, AT_FDCWD, 0, GLOB_VERBOSE, &rb_glob_funcs,
+			(VALUE)&args, args.enc);
+    if (status) GLOB_JUMP_TAG(status);
 }
 
 static void
-push_pattern(path, ary)
-    const char *path;
-    VALUE ary;
-{
-    rb_ary_push(ary, rb_tainted_str_new2(path));
-}
-
-static int
-push_globs(ary, s, flags)
-    VALUE ary;
-    const char *s;
-    int flags;
+push_pattern(const char *path, VALUE ary, void *enc)
 {
-    return rb_glob2(s, flags, push_pattern, ary);
+#if defined _WIN32 || defined __APPLE__
+    VALUE name = rb_utf8_str_new_cstr(path);
+    rb_encoding *eenc = rb_default_internal_encoding();
+    OBJ_TAINT(name);
+    name = rb_str_conv_enc(name, NULL, eenc ? eenc : enc);
+#else
+    VALUE name = rb_external_str_new_with_enc(path, strlen(path), enc);
+#endif
+    rb_ary_push(ary, name);
 }
 
 static int
-push_braces(ary, str, flags)
-    VALUE ary;
-    const char *str;
-    int flags;
+ruby_brace_expand(const char *str, int flags, ruby_glob_func *func, VALUE arg,
+		  rb_encoding *enc, VALUE var)
 {
-    char *buf = 0;
-    char *b, *newbuf;
-    const char *s, *p, *t;
-    const char *lbrace, *rbrace;
-    int nest = 0;
-    int status = 0;
+    const int escape = !(flags & FNM_NOESCAPE);
+    const char *p = str;
+    const char *pend = p + strlen(p);
+    const char *s = p;
+    const char *lbrace = 0, *rbrace = 0;
+    int nest = 0, status = 0;
 
-    s = p = str;
-    lbrace = rbrace = 0;
     while (*p) {
-	if (*p == '{') {
+	if (*p == '{' && nest++ == 0) {
 	    lbrace = p;
-	    break;
 	}
-	p++;
-    }
-    while (*p) {
-	if (*p == '{') nest++;
-	if (*p == '}' && --nest == 0) {
+	if (*p == '}' && lbrace && --nest == 0) {
 	    rbrace = p;
 	    break;
 	}
-	p++;
+	if (*p == '\\' && escape) {
+	    if (!*++p) break;
+	}
+	Inc(p, pend, enc);
     }
 
     if (lbrace && rbrace) {
-	int len = strlen(s);
+	size_t len = strlen(s) + 1;
+	char *buf = GLOB_ALLOC_N(char, len);
+	long shift;
+
+	if (!buf) return -1;
+	memcpy(buf, s, lbrace-s);
+	shift = (lbrace-s);
 	p = lbrace;
-	while (*p != '}') {
-	    t = p + 1;
-	    for (p = t; *p!='}' && *p!=','; p++) {
-		/* skip inner braces */
-		if (*p == '{') {
-		    nest = 1;
-		    while (*++p != '}' || --nest) {
-			if (*p == '{') nest++;
-		    }
+	while (p < rbrace) {
+	    const char *t = ++p;
+	    nest = 0;
+	    while (p < rbrace && !(*p == ',' && nest == 0)) {
+		if (*p == '{') nest++;
+		if (*p == '}') nest--;
+		if (*p == '\\' && escape) {
+		    if (++p == rbrace) break;
 		}
+		Inc(p, pend, enc);
 	    }
-	    newbuf = GLOB_REALLOC_N(buf, char, len+1);
-	    if (!newbuf) {
-		status = -1;
-		break;
-	    }
-	    buf = newbuf;
-	    memcpy(buf, s, lbrace-s);
-	    b = buf + (lbrace-s);
-	    memcpy(b, t, p-t);
-	    strcpy(b+(p-t), rbrace+1);
-	    status = push_braces(ary, buf, flags);
+	    memcpy(buf+shift, t, p-t);
+	    strlcpy(buf+shift+(p-t), rbrace+1, len-(shift+(p-t)));
+	    status = ruby_brace_expand(buf, flags, func, arg, enc, var);
 	    if (status) break;
 	}
+	GLOB_FREE(buf);
     }
-    else {
-	status = push_globs(ary, str, flags);
+    else if (!lbrace && !rbrace) {
+	status = glob_call_func(func, s, arg, enc);
     }
-    if (buf) free(buf);
 
+    RB_GC_GUARD(var);
     return status;
 }
 
-#define isdelim(c) ((c)=='\0')
+struct brace_args {
+    ruby_glob_funcs_t funcs;
+    VALUE value;
+    int flags;
+};
+
+static int
+glob_brace(const char *path, VALUE val, void *enc)
+{
+    struct brace_args *arg = (struct brace_args *)val;
+
+    return ruby_glob0(path, AT_FDCWD, 0, arg->flags, &arg->funcs, arg->value, enc);
+}
+
+int
+ruby_brace_glob_with_enc(const char *str, int flags, ruby_glob_func *func, VALUE arg, rb_encoding *enc)
+{
+    struct brace_args args;
+
+    flags &= ~GLOB_VERBOSE;
+    args.funcs.match = func;
+    args.funcs.error = NULL;
+    args.value = arg;
+    args.flags = flags;
+    return ruby_brace_expand(str, flags, glob_brace, (VALUE)&args, enc, Qfalse);
+}
+
+int
+ruby_brace_glob(const char *str, int flags, ruby_glob_func *func, VALUE arg)
+{
+    return ruby_brace_glob_with_enc(str, flags, func, arg, rb_ascii8bit_encoding());
+}
+
+static int
+push_glob(VALUE ary, VALUE str, VALUE base, int flags)
+{
+    struct glob_args args;
+    int fd;
+    rb_encoding *enc = rb_enc_get(str);
+
+#if defined _WIN32 || defined __APPLE__
+    str = rb_str_encode_ospath(str);
+#endif
+    if (rb_enc_to_index(enc) == ENCINDEX_US_ASCII)
+	enc = rb_filesystem_encoding();
+    if (rb_enc_to_index(enc) == ENCINDEX_US_ASCII)
+	enc = rb_ascii8bit_encoding();
+    flags |= GLOB_VERBOSE;
+    args.func = push_pattern;
+    args.value = ary;
+    args.enc = enc;
+    args.base = 0;
+    fd = AT_FDCWD;
+    if (!NIL_P(base)) {
+	if (!RB_TYPE_P(base, T_STRING) || !rb_enc_check(str, base)) {
+	    struct dir_data *dirp = DATA_PTR(base);
+	    if (!dirp->dir) dir_closed();
+#ifdef HAVE_DIRFD
+	    if ((fd = dirfd(dirp->dir)) == -1)
+		rb_sys_fail_path(dir_inspect(base));
+#endif
+	    base = dirp->path;
+	}
+	args.base = RSTRING_PTR(base);
+    }
+#if defined _WIN32 || defined __APPLE__
+    enc = rb_utf8_encoding();
+#endif
+
+    return ruby_glob0(RSTRING_PTR(str), fd, args.base, flags, &rb_glob_funcs,
+		      (VALUE)&args, enc);
+}
 
 static VALUE
-rb_push_glob(str, flags)
-    VALUE str;
-    int flags;
+rb_push_glob(VALUE str, VALUE base, int flags) /* '\0' is delimiter */
 {
-    const char *p, *pend, *buf;
-    int nest, maxnest;
-    int status = 0;
-    int noescape = flags & FNM_NOESCAPE;
+    long offset = 0;
+    long len;
     VALUE ary;
+    int warned = FALSE;
 
+    /* can contain null bytes as separators */
+    if (!RB_TYPE_P((str), T_STRING)) {
+	FilePathValue(str);
+    }
+    else {
+	rb_check_safe_obj(str);
+	rb_enc_check(str, rb_enc_from_encoding(rb_usascii_encoding()));
+    }
     ary = rb_ary_new();
-    SafeStringValue(str);
-    p = RSTRING(str)->ptr;
-    pend = p + RSTRING(str)->len;
-
-    while (p < pend) {
-	nest = maxnest = 0;
-	while (p < pend && isdelim(*p)) p++;
-	buf = p;
-	while (p < pend && !isdelim(*p)) {
-	    if (*p == '{') nest++, maxnest++;
-	    if (*p == '}') nest--;
-	    if (!noescape && *p == '\\') {
-		if (++p == pend) break;
-	    }
-	    p++;
-	}
-	if (maxnest == 0) {
-	    status = push_globs(ary, buf, flags);
-	    if (status) break;
-	}
-	else if (nest == 0) {
-	    status = push_braces(ary, buf, flags);
-	    if (status) break;
-	}
-	/* else unmatched braces */
+
+    while (offset < (len = RSTRING_LEN(str))) {
+	int status;
+        long rest = len - offset;
+        const char *pbeg = RSTRING_PTR(str), *p = pbeg + offset;
+        const char *pend = memchr(p, '\0', rest);
+        if (pend) {
+            if (!warned) {
+                rb_warn("use glob patterns list instead of nul-separated patterns");
+                warned = TRUE;
+            }
+            rest = ++pend - p;
+            offset = pend - pbeg;
+        }
+        else {
+            offset = len;
+        }
+	status = push_glob(ary, rb_str_subseq(str, p-pbeg, rest),
+			   base, flags);
+	if (status) GLOB_JUMP_TAG(status);
     }
-    if (status) GLOB_JUMP_TAG(status);
-    if (rb_block_given_p()) {
-	rb_ary_each(ary);
-	return Qnil;
+
+    return ary;
+}
+
+static VALUE
+dir_globs(long argc, const VALUE *argv, VALUE base, int flags)
+{
+    VALUE ary = rb_ary_new();
+    long i;
+
+    for (i = 0; i < argc; ++i) {
+	int status;
+	VALUE str = argv[i];
+	FilePathValue(str);
+	status = push_glob(ary, str, base, flags);
+	if (status) GLOB_JUMP_TAG(status);
     }
+
     return ary;
 }
 
+static void
+dir_glob_options(VALUE opt, VALUE *base, int *flags)
+{
+    ID kw[2];
+    VALUE args[2];
+    kw[0] = rb_intern("base");
+    if (flags) kw[1] = rb_intern("flags");
+    rb_get_kwargs(opt, kw, 0, flags ? 2 : 1, args);
+    if (args[0] == Qundef || NIL_P(args[0])) {
+	*base = Qnil;
+    }
+#if USE_OPENDIR_AT
+    else if (rb_typeddata_is_kind_of(args[0], &dir_data_type)) {
+	*base = args[0];
+    }
+#endif
+    else {
+	FilePathValue(args[0]);
+	if (!RSTRING_LEN(args[0])) args[0] = Qnil;
+	*base = args[0];
+    }
+    if (flags && args[1] != Qundef) {
+	*flags = NUM2INT(args[1]);
+    }
+}
+
 /*
  *  call-seq:
- *     Dir[ string ] => array
+ *     Dir[ string [, string ...] [, base: path] ] -> array
  *
  *  Equivalent to calling
- *  <em>dir</em>.<code>glob(</code><i>string,</i><code>0)</code>.
+ *  <code>Dir.glob([</code><i>string,...</i><code>], 0)</code>.
  *
  */
 static VALUE
-dir_s_aref(obj, str)
-    VALUE obj, str;
+dir_s_aref(int argc, VALUE *argv, VALUE obj)
 {
-    return rb_push_glob(str, 0);
+    VALUE opts, base;
+    argc = rb_scan_args(argc, argv, "*:", NULL, &opts);
+    dir_glob_options(opts, &base, NULL);
+    if (argc == 1) {
+	return rb_push_glob(argv[0], base, 0);
+    }
+    return dir_globs(argc, argv, base, 0);
 }
 
 /*
  *  call-seq:
- *     Dir.glob( string, [flags] ) => array
- *     Dir.glob( string, [flags] ) {| filename | block }  => nil
- *
- *  Returns the filenames found by expanding the pattern given in
- *  <i>string</i>, either as an <i>array</i> or as parameters to the
- *  block. Note that this pattern is not a regexp (it's closer to a
- *  shell glob). See <code>File::fnmatch</code> for the meaning of
- *  the <i>flags</i> parameter.
- *
- *  <code>*</code>::        Matches any file. Can be restricted by
- *                          other values in the glob. <code>*</code>
- *                          will match all files; <code>c*</code> will
- *                          match all files beginning with
- *                          <code>c</code>; <code>*c</code> will match
- *                          all files ending with <code>c</code>; and
- *                          <code>*c*</code> will match all files that
- *                          have <code>c</code> in them (including at
- *                          the beginning or end). Equivalent to
- *                          <code>/ .* /x</code> in regexp.
- *  <code>**</code>::       Matches directories recursively.
- *  <code>?</code>::        Matches any one character. Equivalent to
- *                          <code>/.{1}/</code> in regexp.
- *  <code>[set]</code>::    Matches any one character in +set+.
- *                          Behaves exactly like character sets in
- *                          Regexp, including set negation
- *                          (<code>[^a-z]</code>).
- *  <code>{p,q}</code>::    Matches either literal <code>p</code> or
- *                          literal <code>q</code>. Matching literals
- *                          may be more than one character in length.
- *                          More than two literals may be specified.
- *                          Equivalent to pattern alternation in
- *                          regexp.
- *  <code>\</code>::        Escapes the next metacharacter.
+ *     Dir.glob( pattern, [flags], [base: path] )                       -> array
+ *     Dir.glob( pattern, [flags], [base: path] ) { |filename| block }  -> nil
+ *
+ *  Expands +pattern+, which is a pattern string or an Array of pattern
+ *  strings, and returns an array containing the matching filenames.
+ *  If a block is given, calls the block once for each matching filename,
+ *  passing the filename as a parameter to the block.
+ *
+ *  The optional +base+ keyword argument specifies the base directory for
+ *  interpreting relative pathnames instead of the current working directory.
+ *  As the results are not prefixed with the base directory name in this
+ *  case, you will need to prepend the base directory name if you want real
+ *  paths.
+ *
+ *  Note that the pattern is not a regexp, it's closer to a shell glob.
+ *  See File::fnmatch for the meaning of the +flags+ parameter.
+ *  Case sensitivity depends on your system (File::FNM_CASEFOLD is ignored),
+ *  as does the order in which the results are returned.
+ *
+ *  <code>*</code>::
+ *    Matches any file. Can be restricted by other values in the glob.
+ *    Equivalent to <code>/ .* /mx</code> in regexp.
+ *
+ *    <code>*</code>::     Matches all files
+ *    <code>c*</code>::    Matches all files beginning with <code>c</code>
+ *    <code>*c</code>::    Matches all files ending with <code>c</code>
+ *    <code>\*c\*</code>:: Match all files that have <code>c</code> in them
+ *                         (including at the beginning or end).
+ *
+ *    Note, this will not match Unix-like hidden files (dotfiles).  In order
+ *    to include those in the match results, you must use the
+ *    File::FNM_DOTMATCH flag or something like <code>"{*,.*}"</code>.
+ *
+ *  <code>**</code>::
+ *    Matches directories recursively.
+ *
+ *  <code>?</code>::
+ *    Matches any one character. Equivalent to <code>/.{1}/</code> in regexp.
+ *
+ *  <code>[set]</code>::
+ *    Matches any one character in +set+.  Behaves exactly like character sets
+ *    in Regexp, including set negation (<code>[^a-z]</code>).
+ *
+ *  <code>{p,q}</code>::
+ *    Matches either literal <code>p</code> or literal <code>q</code>.
+ *    Equivalent to pattern alternation in regexp.
+ *
+ *    Matching literals may be more than one character in length.  More than
+ *    two literals may be specified.
+ *
+ *  <code> \\ </code>::
+ *    Escapes the next metacharacter.
+ *
+ *    Note that this means you cannot use backslash on windows as part of a
+ *    glob, i.e.  <code>Dir["c:\\foo*"]</code> will not work, use
+ *    <code>Dir["c:/foo*"]</code> instead.
+ *
+ *  Examples:
  *
  *     Dir["config.?"]                     #=> ["config.h"]
  *     Dir.glob("config.?")                #=> ["config.h"]
@@ -1418,60 +2833,78 @@ dir_s_aref(obj, str)
  *     Dir.glob("*.{rb,h}")                #=> ["main.rb", "config.h"]
  *     Dir.glob("*")                       #=> ["config.h", "main.rb"]
  *     Dir.glob("*", File::FNM_DOTMATCH)   #=> [".", "..", "config.h", "main.rb"]
+ *     Dir.glob(["*.rb", "*.h"])           #=> ["main.rb", "config.h"]
  *
  *     rbfiles = File.join("**", "*.rb")
  *     Dir.glob(rbfiles)                   #=> ["main.rb",
- *                                              "lib/song.rb",
- *                                              "lib/song/karaoke.rb"]
+ *                                         #    "lib/song.rb",
+ *                                         #    "lib/song/karaoke.rb"]
+ *
+ *     Dir.glob(rbfiles, base: "lib")      #=> ["song.rb",
+ *                                         #    "song/karaoke.rb"]
+ *
  *     libdirs = File.join("**", "lib")
  *     Dir.glob(libdirs)                   #=> ["lib"]
  *
  *     librbfiles = File.join("**", "lib", "**", "*.rb")
  *     Dir.glob(librbfiles)                #=> ["lib/song.rb",
- *                                              "lib/song/karaoke.rb"]
+ *                                         #    "lib/song/karaoke.rb"]
  *
  *     librbfiles = File.join("**", "lib", "*.rb")
  *     Dir.glob(librbfiles)                #=> ["lib/song.rb"]
  */
 static VALUE
-dir_s_glob(argc, argv, obj)
-    int argc;
-    VALUE *argv;
-    VALUE obj;
+dir_s_glob(int argc, VALUE *argv, VALUE obj)
 {
-    VALUE str, rflags;
+    VALUE str, rflags, ary, opts, base;
     int flags;
 
-    if (rb_scan_args(argc, argv, "11", &str, &rflags) == 2)
+    argc = rb_scan_args(argc, argv, "11:", &str, &rflags, &opts);
+    if (argc == 2)
 	flags = NUM2INT(rflags);
     else
 	flags = 0;
+    dir_glob_options(opts, &base, &flags);
 
-    return rb_push_glob(str, flags);
+    ary = rb_check_array_type(str);
+    if (NIL_P(ary)) {
+	ary = rb_push_glob(str, base, flags);
+    }
+    else {
+	VALUE v = ary;
+	ary = dir_globs(RARRAY_LEN(v), RARRAY_CONST_PTR(v), base, flags);
+	RB_GC_GUARD(v);
+    }
+
+    if (rb_block_given_p()) {
+	rb_ary_each(ary);
+	return Qnil;
+    }
+    return ary;
 }
 
 static VALUE
-dir_open_dir(path)
-    VALUE path;
+dir_open_dir(int argc, VALUE *argv)
 {
-    VALUE dir = rb_funcall(rb_cDir, rb_intern("open"), 1, path);
+    VALUE dir = rb_funcallv(rb_cDir, rb_intern("open"), argc, argv);
 
-    if (TYPE(dir) != T_DATA ||
-	RDATA(dir)->dfree != (RUBY_DATA_FUNC)free_dir) {
-	rb_raise(rb_eTypeError, "wrong argument type %s (expected Dir)",
-		 rb_obj_classname(dir));
-    }
+    rb_check_typeddata(dir, &dir_data_type);
     return dir;
 }
 
 
 /*
  *  call-seq:
- *     Dir.foreach( dirname ) {| filename | block }  => nil
+ *     Dir.foreach( dirname ) {| filename | block }                 -> nil
+ *     Dir.foreach( dirname, encoding: enc ) {| filename | block }  -> nil
+ *     Dir.foreach( dirname )                                       -> an_enumerator
+ *     Dir.foreach( dirname, encoding: enc )                        -> an_enumerator
  *
  *  Calls the block once for each entry in the named directory, passing
  *  the filename of each entry as a parameter to the block.
  *
+ *  If no block is given, an enumerator is returned instead.
+ *
  *     Dir.foreach("testdir") {|x| puts "Got #{x}" }
  *
  *  <em>produces:</em>
@@ -1483,100 +2916,277 @@ dir_open_dir(path)
  *
  */
 static VALUE
-dir_foreach(io, dirname)
-    VALUE io, dirname;
+dir_foreach(int argc, VALUE *argv, VALUE io)
 {
     VALUE dir;
 
-    dir = dir_open_dir(dirname);
+    RETURN_ENUMERATOR(io, argc, argv);
+    dir = dir_open_dir(argc, argv);
     rb_ensure(dir_each, dir, dir_close, dir);
     return Qnil;
 }
 
+static VALUE
+dir_collect(VALUE dir)
+{
+    VALUE ary = rb_ary_new();
+    dir_each_entry(dir, rb_ary_push, ary, FALSE);
+    return ary;
+}
+
 /*
  *  call-seq:
- *     Dir.entries( dirname ) => array
+ *     Dir.entries( dirname )                -> array
+ *     Dir.entries( dirname, encoding: enc ) -> array
  *
  *  Returns an array containing all of the filenames in the given
  *  directory. Will raise a <code>SystemCallError</code> if the named
  *  directory doesn't exist.
  *
+ *  The optional <i>encoding</i> keyword argument specifies the encoding of the
+ *  directory. If not specified, the filesystem encoding is used.
+ *
  *     Dir.entries("testdir")   #=> [".", "..", "config.h", "main.rb"]
  *
  */
 static VALUE
-dir_entries(io, dirname)
-    VALUE io, dirname;
+dir_entries(int argc, VALUE *argv, VALUE io)
 {
     VALUE dir;
 
-    dir = dir_open_dir(dirname);
-    return rb_ensure(rb_Array, dir, dir_close, dir);
+    dir = dir_open_dir(argc, argv);
+    return rb_ensure(dir_collect, dir, dir_close, dir);
+}
+
+static VALUE
+dir_each_child(VALUE dir)
+{
+    return dir_each_entry(dir, dir_yield, Qnil, TRUE);
 }
 
 /*
  *  call-seq:
- *     File.fnmatch( pattern, path, [flags] ) => (true or false)
- *     File.fnmatch?( pattern, path, [flags] ) => (true or false)
- *
- *  Returns true if <i>path</i> matches against <i>pattern</i> The
- *  pattern is not a regular expression; instead it follows rules
- *  similar to shell filename globbing. It may contain the following
- *  metacharacters:
- *
- *  <code>*</code>::        Matches any file. Can be restricted by
- *                          other values in the glob. <code>*</code>
- *                          will match all files; <code>c*</code> will
- *                          match all files beginning with
- *                          <code>c</code>; <code>*c</code> will match
- *                          all files ending with <code>c</code>; and
- *                          <code>*c*</code> will match all files that
- *                          have <code>c</code> in them (including at
- *                          the beginning or end). Equivalent to
- *                          <code>/ .* /x</code> in regexp.
- *  <code>?</code>::        Matches any one character. Equivalent to
- *                          <code>/.{1}/</code> in regexp.
- *  <code>[set]</code>::    Matches any one character in +set+.
- *                          Behaves exactly like character sets in
- *                          Regexp, including set negation
- *                          (<code>[^a-z]</code>).
- *  <code>\</code>::        Escapes the next metacharacter.
- *
- *  <i>flags</i> is a bitwise OR of the <code>FNM_xxx</code>
- *  parameters. The same glob pattern and flags are used by
- *  <code>Dir::glob</code>.
- *
- *     File.fnmatch('cat',       'cat')        #=> true
- *     File.fnmatch('cat',       'category')   #=> false
- *     File.fnmatch('c{at,ub}s', 'cats')       #=> false
- *     File.fnmatch('c{at,ub}s', 'cubs')       #=> false
- *     File.fnmatch('c{at,ub}s', 'cat')        #=> false
- *
- *     File.fnmatch('c?t',    'cat')                       #=> true
- *     File.fnmatch('c\?t',   'cat')                       #=> false
- *     File.fnmatch('c??t',   'cat')                       #=> false
- *     File.fnmatch('c*',     'cats')                      #=> true
- *     File.fnmatch('c/ * FIXME * /t', 'c/a/b/c/t')                 #=> true
- *     File.fnmatch('c*t',    'cat')                       #=> true
- *     File.fnmatch('c\at',   'cat')                       #=> true
- *     File.fnmatch('c\at',   'cat', File::FNM_NOESCAPE)   #=> false
- *     File.fnmatch('a?b',    'a/b')                       #=> true
- *     File.fnmatch('a?b',    'a/b', File::FNM_PATHNAME)   #=> false
- *
- *     File.fnmatch('*',   '.profile')                            #=> false
- *     File.fnmatch('*',   '.profile', File::FNM_DOTMATCH)        #=> true
- *     File.fnmatch('*',   'dave/.profile')                       #=> true
- *     File.fnmatch('*',   'dave/.profile', File::FNM_DOTMATCH)   #=> true
- *     File.fnmatch('*',   'dave/.profile', File::FNM_PATHNAME)   #=> false
- *     File.fnmatch('* / FIXME *', 'dave/.profile', File::FNM_PATHNAME)   #=> false
- *     STRICT = File::FNM_PATHNAME | File::FNM_DOTMATCH
- *     File.fnmatch('* / FIXME *', 'dave/.profile', STRICT)               #=> true
+ *     Dir.each_child( dirname ) {| filename | block }                 -> nil
+ *     Dir.each_child( dirname, encoding: enc ) {| filename | block }  -> nil
+ *     Dir.each_child( dirname )                                       -> an_enumerator
+ *     Dir.each_child( dirname, encoding: enc )                        -> an_enumerator
+ *
+ *  Calls the block once for each entry except for "." and ".." in the
+ *  named directory, passing the filename of each entry as a parameter
+ *  to the block.
+ *
+ *  If no block is given, an enumerator is returned instead.
+ *
+ *     Dir.each_child("testdir") {|x| puts "Got #{x}" }
+ *
+ *  <em>produces:</em>
+ *
+ *     Got config.h
+ *     Got main.rb
+ *
  */
 static VALUE
-file_s_fnmatch(argc, argv, obj)
-    int argc;
-    VALUE *argv;
-    VALUE obj;
+dir_s_each_child(int argc, VALUE *argv, VALUE io)
+{
+    VALUE dir;
+
+    RETURN_ENUMERATOR(io, argc, argv);
+    dir = dir_open_dir(argc, argv);
+    rb_ensure(dir_each_child, dir, dir_close, dir);
+    return Qnil;
+}
+
+/*
+ *  call-seq:
+ *     dir.each_child {| filename | block }  -> nil
+ *     dir.each_child                        -> an_enumerator
+ *
+ *  Calls the block once for each entry except for "." and ".." in
+ *  this directory, passing the filename of each entry as a parameter
+ *  to the block.
+ *
+ *  If no block is given, an enumerator is returned instead.
+ *
+ *     d = Dir.new("testdir")
+ *     d.each_child  {|x| puts "Got #{x}" }
+ *
+ *  <em>produces:</em>
+ *
+ *     Got config.h
+ *     Got main.rb
+ *
+ */
+static VALUE
+dir_each_child_m(VALUE dir)
+{
+    RETURN_ENUMERATOR(dir, 0, 0);
+    return dir_each_entry(dir, dir_yield, Qnil, TRUE);
+}
+
+/*
+ *  call-seq:
+ *     dir.children  -> array
+ *
+ *  Returns an array containing all of the filenames except for "."
+ *  and ".." in this directory.
+ *
+ *     d = Dir.new("testdir")
+ *     d.children   #=> ["config.h", "main.rb"]
+ *
+ */
+static VALUE
+dir_collect_children(VALUE dir)
+{
+    VALUE ary = rb_ary_new();
+    dir_each_entry(dir, rb_ary_push, ary, TRUE);
+    return ary;
+}
+
+/*
+ *  call-seq:
+ *     Dir.children( dirname )                -> array
+ *     Dir.children( dirname, encoding: enc ) -> array
+ *
+ *  Returns an array containing all of the filenames except for "."
+ *  and ".." in the given directory. Will raise a
+ *  <code>SystemCallError</code> if the named directory doesn't exist.
+ *
+ *  The optional <i>encoding</i> keyword argument specifies the encoding of the
+ *  directory. If not specified, the filesystem encoding is used.
+ *
+ *     Dir.children("testdir")   #=> ["config.h", "main.rb"]
+ *
+ */
+static VALUE
+dir_s_children(int argc, VALUE *argv, VALUE io)
+{
+    VALUE dir;
+
+    dir = dir_open_dir(argc, argv);
+    return rb_ensure(dir_collect_children, dir, dir_close, dir);
+}
+
+static int
+fnmatch_brace(const char *pattern, VALUE val, void *enc)
+{
+    struct brace_args *arg = (struct brace_args *)val;
+    VALUE path = arg->value;
+    rb_encoding *enc_pattern = enc;
+    rb_encoding *enc_path = rb_enc_get(path);
+
+    if (enc_pattern != enc_path) {
+	if (!rb_enc_asciicompat(enc_pattern))
+	    return FNM_NOMATCH;
+	if (!rb_enc_asciicompat(enc_path))
+	    return FNM_NOMATCH;
+	if (!rb_enc_str_asciionly_p(path)) {
+	    int cr = ENC_CODERANGE_7BIT;
+	    long len = strlen(pattern);
+	    if (rb_str_coderange_scan_restartable(pattern, pattern + len,
+						  enc_pattern, &cr) != len)
+		return FNM_NOMATCH;
+	    if (cr != ENC_CODERANGE_7BIT)
+		return FNM_NOMATCH;
+	}
+    }
+    return (fnmatch(pattern, enc, RSTRING_PTR(path), arg->flags) == 0);
+}
+
+/*
+ *  call-seq:
+ *     File.fnmatch( pattern, path, [flags] ) -> (true or false)
+ *     File.fnmatch?( pattern, path, [flags] ) -> (true or false)
+ *
+ *  Returns true if +path+ matches against +pattern+.  The pattern is not a
+ *  regular expression; instead it follows rules similar to shell filename
+ *  globbing.  It may contain the following metacharacters:
+ *
+ *  <code>*</code>::
+ *    Matches any file. Can be restricted by other values in the glob.
+ *    Equivalent to <code>/ .* /x</code> in regexp.
+ *
+ *    <code>*</code>::    Matches all files regular files
+ *    <code>c*</code>::   Matches all files beginning with <code>c</code>
+ *    <code>*c</code>::   Matches all files ending with <code>c</code>
+ *    <code>\*c*</code>:: Matches all files that have <code>c</code> in them
+ *                        (including at the beginning or end).
+ *
+ *    To match hidden files (that start with a <code>.</code> set the
+ *    File::FNM_DOTMATCH flag.
+ *
+ *  <code>**</code>::
+ *    Matches directories recursively or files expansively.
+ *
+ *  <code>?</code>::
+ *    Matches any one character. Equivalent to <code>/.{1}/</code> in regexp.
+ *
+ *  <code>[set]</code>::
+ *    Matches any one character in +set+.  Behaves exactly like character sets
+ *    in Regexp, including set negation (<code>[^a-z]</code>).
+ *
+ *  <code> \ </code>::
+ *    Escapes the next metacharacter.
+ *
+ *  <code>{a,b}</code>::
+ *    Matches pattern a and pattern b if File::FNM_EXTGLOB flag is enabled.
+ *    Behaves like a Regexp union (<code>(?:a|b)</code>).
+ *
+ *  +flags+ is a bitwise OR of the <code>FNM_XXX</code> constants. The same
+ *  glob pattern and flags are used by Dir::glob.
+ *
+ *  Examples:
+ *
+ *     File.fnmatch('cat',       'cat')        #=> true  # match entire string
+ *     File.fnmatch('cat',       'category')   #=> false # only match partial string
+ *
+ *     File.fnmatch('c{at,ub}s', 'cats')                    #=> false # { } isn't supported by default
+ *     File.fnmatch('c{at,ub}s', 'cats', File::FNM_EXTGLOB) #=> true  # { } is supported on FNM_EXTGLOB
+ *
+ *     File.fnmatch('c?t',     'cat')          #=> true  # '?' match only 1 character
+ *     File.fnmatch('c??t',    'cat')          #=> false # ditto
+ *     File.fnmatch('c*',      'cats')         #=> true  # '*' match 0 or more characters
+ *     File.fnmatch('c*t',     'c/a/b/t')      #=> true  # ditto
+ *     File.fnmatch('ca[a-z]', 'cat')          #=> true  # inclusive bracket expression
+ *     File.fnmatch('ca[^t]',  'cat')          #=> false # exclusive bracket expression ('^' or '!')
+ *
+ *     File.fnmatch('cat', 'CAT')                     #=> false # case sensitive
+ *     File.fnmatch('cat', 'CAT', File::FNM_CASEFOLD) #=> true  # case insensitive
+ *
+ *     File.fnmatch('?',   '/', File::FNM_PATHNAME)  #=> false # wildcard doesn't match '/' on FNM_PATHNAME
+ *     File.fnmatch('*',   '/', File::FNM_PATHNAME)  #=> false # ditto
+ *     File.fnmatch('[/]', '/', File::FNM_PATHNAME)  #=> false # ditto
+ *
+ *     File.fnmatch('\?',   '?')                       #=> true  # escaped wildcard becomes ordinary
+ *     File.fnmatch('\a',   'a')                       #=> true  # escaped ordinary remains ordinary
+ *     File.fnmatch('\a',   '\a', File::FNM_NOESCAPE)  #=> true  # FNM_NOESCAPE makes '\' ordinary
+ *     File.fnmatch('[\?]', '?')                       #=> true  # can escape inside bracket expression
+ *
+ *     File.fnmatch('*',   '.profile')                      #=> false # wildcard doesn't match leading
+ *     File.fnmatch('*',   '.profile', File::FNM_DOTMATCH)  #=> true  # period by default.
+ *     File.fnmatch('.*',  '.profile')                      #=> true
+ *
+ *     rbfiles = '**' '/' '*.rb' # you don't have to do like this. just write in single string.
+ *     File.fnmatch(rbfiles, 'main.rb')                    #=> false
+ *     File.fnmatch(rbfiles, './main.rb')                  #=> false
+ *     File.fnmatch(rbfiles, 'lib/song.rb')                #=> true
+ *     File.fnmatch('**.rb', 'main.rb')                    #=> true
+ *     File.fnmatch('**.rb', './main.rb')                  #=> false
+ *     File.fnmatch('**.rb', 'lib/song.rb')                #=> true
+ *     File.fnmatch('*',           'dave/.profile')                      #=> true
+ *
+ *     pattern = '*' '/' '*'
+ *     File.fnmatch(pattern, 'dave/.profile', File::FNM_PATHNAME)  #=> false
+ *     File.fnmatch(pattern, 'dave/.profile', File::FNM_PATHNAME | File::FNM_DOTMATCH) #=> true
+ *
+ *     pattern = '**' '/' 'foo'
+ *     File.fnmatch(pattern, 'a/b/c/foo', File::FNM_PATHNAME)     #=> true
+ *     File.fnmatch(pattern, '/a/b/c/foo', File::FNM_PATHNAME)    #=> true
+ *     File.fnmatch(pattern, 'c:/a/b/c/foo', File::FNM_PATHNAME)  #=> true
+ *     File.fnmatch(pattern, 'a/.b/c/foo', File::FNM_PATHNAME)    #=> false
+ *     File.fnmatch(pattern, 'a/.b/c/foo', File::FNM_PATHNAME | File::FNM_DOTMATCH) #=> true
+ */
+static VALUE
+file_s_fnmatch(int argc, VALUE *argv, VALUE obj)
 {
     VALUE pattern, path;
     VALUE rflags;
@@ -1588,15 +3198,164 @@ file_s_fnmatch(argc, argv, obj)
 	flags = 0;
 
     StringValue(pattern);
-    StringValue(path);
+    FilePathStringValue(path);
+
+    if (flags & FNM_EXTGLOB) {
+	struct brace_args args;
 
-    if (fnmatch(RSTRING(pattern)->ptr, RSTRING(path)->ptr, flags) == 0)
-	return Qtrue;
+	args.value = path;
+	args.flags = flags;
+	if (ruby_brace_expand(RSTRING_PTR(pattern), flags, fnmatch_brace,
+			      (VALUE)&args, rb_enc_get(pattern), pattern) > 0)
+	    return Qtrue;
+    }
+    else {
+	rb_encoding *enc = rb_enc_compatible(pattern, path);
+	if (!enc) return Qfalse;
+	if (fnmatch(RSTRING_PTR(pattern), enc, RSTRING_PTR(path), flags) == 0)
+	    return Qtrue;
+    }
+    RB_GC_GUARD(pattern);
 
     return Qfalse;
 }
 
 /*
+ *  call-seq:
+ *    Dir.home()       -> "/home/me"
+ *    Dir.home("root") -> "/root"
+ *
+ *  Returns the home directory of the current user or the named user
+ *  if given.
+ */
+static VALUE
+dir_s_home(int argc, VALUE *argv, VALUE obj)
+{
+    VALUE user;
+    const char *u = 0;
+
+    rb_check_arity(argc, 0, 1);
+    user = (argc > 0) ? argv[0] : Qnil;
+    if (!NIL_P(user)) {
+	SafeStringValue(user);
+	rb_must_asciicompat(user);
+	u = StringValueCStr(user);
+	if (*u) {
+	    return rb_home_dir_of(user, rb_str_new(0, 0));
+	}
+    }
+    return rb_default_home_dir(rb_str_new(0, 0));
+
+}
+
+#if 0
+/*
+ * call-seq:
+ *   Dir.exist?(file_name)   ->  true or false
+ *
+ * Returns <code>true</code> if the named file is a directory,
+ * <code>false</code> otherwise.
+ *
+ */
+VALUE
+rb_file_directory_p(void)
+{
+}
+#endif
+
+/*
+ * call-seq:
+ *   Dir.exists?(file_name)  ->  true or false
+ *
+ * Deprecated method. Don't use.
+ */
+static VALUE
+rb_dir_exists_p(VALUE obj, VALUE fname)
+{
+    rb_warning("Dir.exists? is a deprecated name, use Dir.exist? instead");
+    return rb_file_directory_p(obj, fname);
+}
+
+static void *
+nogvl_dir_empty_p(void *ptr)
+{
+    const char *path = ptr;
+    DIR *dir = opendir(path);
+    struct dirent *dp;
+    VALUE result = Qtrue;
+
+    if (!dir) {
+	int e = errno;
+	switch (gc_for_fd_with_gvl(e)) {
+	  default:
+	    dir = opendir(path);
+	    if (dir) break;
+	    e = errno;
+	    /* fall through */
+	  case 0:
+	    if (e == ENOTDIR) return (void *)Qfalse;
+	    errno = e; /* for rb_sys_fail_path */
+	    return (void *)Qundef;
+	}
+    }
+    while ((dp = READDIR(dir, NULL)) != NULL) {
+	if (!to_be_skipped(dp)) {
+	    result = Qfalse;
+	    break;
+	}
+    }
+    closedir(dir);
+    return (void *)result;
+}
+
+/*
+ * call-seq:
+ *   Dir.empty?(path_name)  ->  true or false
+ *
+ * Returns <code>true</code> if the named file is an empty directory,
+ * <code>false</code> if it is not a directory or non-empty.
+ */
+static VALUE
+rb_dir_s_empty_p(VALUE obj, VALUE dirname)
+{
+    VALUE result, orig;
+    const char *path;
+    enum {false_on_notdir = 1};
+
+    FilePathValue(dirname);
+    orig = rb_str_dup_frozen(dirname);
+    dirname = rb_str_encode_ospath(dirname);
+    dirname = rb_str_dup_frozen(dirname);
+    path = RSTRING_PTR(dirname);
+
+#if defined HAVE_GETATTRLIST && defined ATTR_DIR_ENTRYCOUNT
+    {
+	u_int32_t attrbuf[SIZEUP32(fsobj_tag_t)];
+	struct attrlist al = {ATTR_BIT_MAP_COUNT, 0, ATTR_CMN_OBJTAG,};
+	if (getattrlist(path, &al, attrbuf, sizeof(attrbuf), 0) != 0)
+	    rb_sys_fail_path(orig);
+	if (*(const fsobj_tag_t *)(attrbuf+1) == VT_HFS) {
+	    al.commonattr = 0;
+	    al.dirattr = ATTR_DIR_ENTRYCOUNT;
+	    if (getattrlist(path, &al, attrbuf, sizeof(attrbuf), 0) == 0) {
+		if (attrbuf[0] >= 2 * sizeof(u_int32_t))
+		    return attrbuf[1] ? Qfalse : Qtrue;
+		if (false_on_notdir) return Qfalse;
+	    }
+	    rb_sys_fail_path(orig);
+	}
+    }
+#endif
+
+    result = (VALUE)rb_thread_call_without_gvl(nogvl_dir_empty_p, (void *)path,
+					    RUBY_UBF_IO, 0);
+    if (result == Qundef) {
+	rb_sys_fail_path(orig);
+    }
+    return result;
+}
+
+/*
  *  Objects of class <code>Dir</code> are directory streams representing
  *  directories in the underlying file system. They provide a variety of
  *  ways to list directories and their contents. See also
@@ -1608,21 +3367,28 @@ file_s_fnmatch(argc, argv, obj)
  *  (<code>.</code>).
  */
 void
-Init_Dir()
+Init_Dir(void)
 {
     rb_cDir = rb_define_class("Dir", rb_cObject);
 
     rb_include_module(rb_cDir, rb_mEnumerable);
 
     rb_define_alloc_func(rb_cDir, dir_s_alloc);
-    rb_define_singleton_method(rb_cDir, "open", dir_s_open, 1);
-    rb_define_singleton_method(rb_cDir, "foreach", dir_foreach, 1);
-    rb_define_singleton_method(rb_cDir, "entries", dir_entries, 1);
-
-    rb_define_method(rb_cDir,"initialize", dir_initialize, 1);
+    rb_define_singleton_method(rb_cDir, "open", dir_s_open, -1);
+    rb_define_singleton_method(rb_cDir, "foreach", dir_foreach, -1);
+    rb_define_singleton_method(rb_cDir, "entries", dir_entries, -1);
+    rb_define_singleton_method(rb_cDir, "each_child", dir_s_each_child, -1);
+    rb_define_singleton_method(rb_cDir, "children", dir_s_children, -1);
+
+    rb_define_method(rb_cDir,"initialize", dir_initialize, -1);
+    rb_define_method(rb_cDir,"fileno", dir_fileno, 0);
     rb_define_method(rb_cDir,"path", dir_path, 0);
+    rb_define_method(rb_cDir,"to_path", dir_path, 0);
+    rb_define_method(rb_cDir,"inspect", dir_inspect, 0);
     rb_define_method(rb_cDir,"read", dir_read, 0);
     rb_define_method(rb_cDir,"each", dir_each, 0);
+    rb_define_method(rb_cDir,"each_child", dir_each_child_m, 0);
+    rb_define_method(rb_cDir,"children", dir_collect_children, 0);
     rb_define_method(rb_cDir,"rewind", dir_rewind, 0);
     rb_define_method(rb_cDir,"tell", dir_tell, 0);
     rb_define_method(rb_cDir,"seek", dir_seek, 1);
@@ -1638,16 +3404,61 @@ Init_Dir()
     rb_define_singleton_method(rb_cDir,"rmdir", dir_s_rmdir, 1);
     rb_define_singleton_method(rb_cDir,"delete", dir_s_rmdir, 1);
     rb_define_singleton_method(rb_cDir,"unlink", dir_s_rmdir, 1);
+    rb_define_singleton_method(rb_cDir,"home", dir_s_home, -1);
 
     rb_define_singleton_method(rb_cDir,"glob", dir_s_glob, -1);
-    rb_define_singleton_method(rb_cDir,"[]", dir_s_aref, 1);
+    rb_define_singleton_method(rb_cDir,"[]", dir_s_aref, -1);
+    rb_define_singleton_method(rb_cDir,"exist?", rb_file_directory_p, 1);
+    rb_define_singleton_method(rb_cDir,"exists?", rb_dir_exists_p, 1);
+    rb_define_singleton_method(rb_cDir,"empty?", rb_dir_s_empty_p, 1);
 
     rb_define_singleton_method(rb_cFile,"fnmatch", file_s_fnmatch, -1);
     rb_define_singleton_method(rb_cFile,"fnmatch?", file_s_fnmatch, -1);
 
+    /*  Document-const: File::Constants::FNM_NOESCAPE
+     *
+     *  Disables escapes in File.fnmatch and Dir.glob patterns
+     */
     rb_file_const("FNM_NOESCAPE", INT2FIX(FNM_NOESCAPE));
+
+    /*  Document-const: File::Constants::FNM_PATHNAME
+     *
+     *  Wildcards in File.fnmatch and Dir.glob patterns do not match directory
+     *  separators
+     */
     rb_file_const("FNM_PATHNAME", INT2FIX(FNM_PATHNAME));
+
+    /*  Document-const: File::Constants::FNM_DOTMATCH
+     *
+     *  The '*' wildcard matches filenames starting with "." in File.fnmatch
+     *  and Dir.glob patterns
+     */
     rb_file_const("FNM_DOTMATCH", INT2FIX(FNM_DOTMATCH));
+
+    /*  Document-const: File::Constants::FNM_CASEFOLD
+     *
+     *  Makes File.fnmatch patterns case insensitive (but not Dir.glob
+     *  patterns).
+     */
     rb_file_const("FNM_CASEFOLD", INT2FIX(FNM_CASEFOLD));
+
+    /*  Document-const: File::Constants::FNM_EXTGLOB
+     *
+     *  Allows file globbing through "{a,b}" in File.fnmatch patterns.
+     */
+    rb_file_const("FNM_EXTGLOB", INT2FIX(FNM_EXTGLOB));
+
+    /*  Document-const: File::Constants::FNM_SYSCASE
+     *
+     *  System default case insensitiveness, equals to FNM_CASEFOLD or
+     *  0.
+     */
     rb_file_const("FNM_SYSCASE", INT2FIX(FNM_SYSCASE));
+
+    /*  Document-const: File::Constants::FNM_SHORTNAME
+     *
+     *  Makes patterns to match short names if existing.  Valid only
+     *  on Microsoft Windows.
+     */
+    rb_file_const("FNM_SHORTNAME", INT2FIX(FNM_SHORTNAME));
 }
diff --git a/distruby.rb b/distruby.rb
deleted file mode 100644
index 613da0eb9c..0000000000
--- a/distruby.rb
+++ /dev/null
@@ -1,59 +0,0 @@
-#!./miniruby
-
-if RUBY_PATCHLEVEL.zero?
-	dirname = sprintf 'ruby-%s', RUBY_VERSION
-	tagname = dirname.gsub /ruby-(\d)\.(\d)\.(\d)/, 'v\1_\2_\3'
-else
-	dirname = sprintf 'ruby-%s-p%u', RUBY_VERSION, RUBY_PATCHLEVEL
-	tagname = dirname.gsub /ruby-(\d)\.(\d)\.(\d)-p/, 'v\1_\2_\3_'
-end
-tgzname = dirname + '.tar.gz'
-tbzname = dirname + '.tar.bz2'
-zipname = dirname + '.zip'
-repos   = 'http://svn.ruby-lang.org/repos/ruby/tags/' + tagname
-
-STDERR.puts 'exporting sources...'
-system 'svn',  'export',  '-q', repos, dirname
-Dir.chdir dirname do
-	STDERR.puts 'generating configure...'
-	system 'autoconf'
-	system 'rm', '-rf', 'autom4te.cache'
-
-	STDERR.puts 'generating parse.c...'
-	system 'bison', '-y', '-o', 'parse.c', 'parse.y'
-end
-
-STDERR.puts 'generating tarballs...'
-ENV['GZIP'] = '-9'
-system 'tar', 'chofzp', tgzname, dirname
-system 'tar', 'chojfp', tbzname, dirname
-system 'zip', '-q9r', zipname, dirname
-
-require 'digest/md5'
-require 'digest/sha2'
-for name in [tgzname, tbzname, zipname] do
-	open name, 'rb' do |fp|
-		str = fp.read
-		md5 = Digest::MD5.hexdigest str
-		sha = Digest::SHA256.hexdigest str
-		printf "MD5(%s)= %s\nSHA256(%s)= %s\nSIZE(%s)= %s\n\n",
-				 name, md5,
-				 name, sha,
-				 name, str.size
-	end
-end
-
-
-
-# 
-# Local Variables:
-# mode: ruby
-# code: utf-8
-# indent-tabs-mode: t
-# tab-width: 3
-# ruby-indent-level: 3
-# fill-column: 79
-# default-justification: full
-# End:
-# vi: ts=3 sw=3
-
diff --git a/djgpp/GNUmakefile.in b/djgpp/GNUmakefile.in
deleted file mode 100644
index 0a7e1fb131..0000000000
--- a/djgpp/GNUmakefile.in
+++ /dev/null
@@ -1,2 +0,0 @@
-include Makefile
-VPATH = $(srcdir) $(srcdir)/missing
diff --git a/djgpp/README.djgpp b/djgpp/README.djgpp
deleted file mode 100644
index f1f413a478..0000000000
--- a/djgpp/README.djgpp
+++ /dev/null
@@ -1,21 +0,0 @@
-* How to compile and install on djgpp
-
-This is what you need to do to compile and install Ruby:
-
-  1. Run configure.bat, which will generate config.h and Makefile
-     (GNU sed required).
-     Message like this is normal:
-       sed.exe: can't read 123456789: No such file or directory (ENOENT)
-
-  2. Edit defines.h if you need.  Probably this step will not need.
-
-  3. Remove comment mark(#) before the module names from ext/Setup.dj (or
-     add module names if not present).
-
-  4. Run make.
-
-  5. Optionally, run 'make test' to check whether the compiled Ruby
-     interpreter works well.  If you see the message "test succeeded",
-     your ruby works as it should (hopefully).
-
-  6. Run 'make install'
diff --git a/djgpp/config.hin b/djgpp/config.hin
deleted file mode 100644
index 8ee427c92f..0000000000
--- a/djgpp/config.hin
+++ /dev/null
@@ -1,114 +0,0 @@
-
-#define PACKAGE_NAME ""
-#define PACKAGE_TARNAME ""
-#define PACKAGE_VERSION ""
-#define PACKAGE_STRING ""
-#define PACKAGE_BUGREPORT ""
-#define USE_BUILTIN_FRAME_ADDRESS 1
-#define STDC_HEADERS 1
-#define HAVE_SYS_TYPES_H 1
-#define HAVE_SYS_STAT_H 1
-#define HAVE_STDLIB_H 1
-#define HAVE_STRING_H 1
-#define HAVE_MEMORY_H 1
-#define HAVE_STRINGS_H 1
-#define HAVE_UNISTD_H 1
-#define HAVE_LONG_LONG 1
-#define HAVE_OFF_T 1
-#define SIZEOF_INT 4
-#define SIZEOF_SHORT 2
-#define SIZEOF_LONG 4
-#define SIZEOF_LONG_LONG 8
-#define SIZEOF___INT64 0
-#define SIZEOF_OFF_T 4
-#define SIZEOF_VOIDP 4
-#define SIZEOF_FLOAT 4
-#define SIZEOF_DOUBLE 8
-#define HAVE_PROTOTYPES 1
-#define TOKEN_PASTE(x,y) x##y
-#define HAVE_STDARG_PROTOTYPES 1
-#define NORETURN(x) x __attribute__ ((noreturn))
-#define HAVE_DECL_SYS_NERR 1
-#define HAVE_DIRENT_H 1
-#define STDC_HEADERS 1
-#define HAVE_SYS_WAIT_H 1
-#define HAVE_STDLIB_H 1
-#define HAVE_STRING_H 1
-#define HAVE_UNISTD_H 1
-#define HAVE_LIMITS_H 1
-#define HAVE_SYS_FILE_H 1
-#define HAVE_SYS_IOCTL_H 1
-#define HAVE_FCNTL_H 1
-#define HAVE_SYS_FCNTL_H 1
-#define HAVE_SYS_TIME_H 1
-#define HAVE_SYS_TIMES_H 1
-#define HAVE_SYS_PARAM_H 1
-#define HAVE_PWD_H 1
-#define HAVE_UTIME_H 1
-#define HAVE_MEMORY_H 1
-#define HAVE_DIRECT_H 1
-#define HAVE_SYS_RESOURCE_H 1
-#define HAVE_STRUCT_STAT_ST_BLKSIZE 1
-#define HAVE_ST_BLKSIZE 1
-#define HAVE_STRUCT_STAT_ST_RDEV 1
-#define HAVE_ST_RDEV 1
-#define GETGROUPS_T gid_t
-#define RETSIGTYPE void
-#define HAVE_ALLOCA 1
-#define HAVE_DUP2 1
-#define HAVE_MEMMOVE 1
-#define HAVE_MKDIR 1
-#define HAVE_STRCASECMP 1
-#define HAVE_STRNCASECMP 1
-#define HAVE_STRERROR 1
-#define HAVE_STRFTIME 1
-#define HAVE_STRCHR 1
-#define HAVE_STRSTR 1
-#define HAVE_STRTOUL 1
-#define HAVE_ISINF 1
-#define HAVE_ISNAN 1
-#define HAVE_FINITE 1
-#define HAVE_HYPOT 1
-#define HAVE_ACOSH 1
-#define HAVE_FMOD 1
-#define HAVE_WAITPID 1
-#define HAVE_FSYNC 1
-#define HAVE_TRUNCATE 1
-#define HAVE_CHSIZE 1
-#define HAVE_TIMES 1
-#define HAVE_UTIMES 1
-#define HAVE_FCNTL 1
-#define HAVE_SYMLINK 1
-#define HAVE_SETITIMER 1
-#define HAVE_PAUSE 1
-#define HAVE_GETPGRP 1
-#define HAVE_SETPGID 1
-#define HAVE_GETGROUPS 1
-#define HAVE_GETRLIMIT 1
-#define HAVE_SIGPROCMASK 1
-#define HAVE_SIGACTION 1
-#define HAVE_SETSID 1
-#define HAVE_TELLDIR 1
-#define HAVE_SEEKDIR 1
-#define HAVE_MKTIME 1
-#define HAVE_COSH 1
-#define HAVE_SINH 1
-#define HAVE_TANH 1
-#define HAVE_STRUCT_TM_TM_ZONE 1
-#define HAVE_TM_ZONE 1
-#define HAVE_STRUCT_TM_TM_GMTOFF 1
-#define POSIX_SIGNAL 1
-#define GETPGRP_VOID 1
-#define SETPGRP_VOID 1
-#define RSHIFT(x,y) ((x)>>(int)y)
-#define FILE_COUNT _cnt
-#define FILE_READPTR _ptr
-#define NEED_IO_FLUSH_BETWEEN_RW 1
-#define DEFAULT_KCODE KCODE_NONE
-#define DLEXT ".so"
-#define RUBY_LIB "/lib/ruby/@MAJOR@.@MINOR@"
-#define RUBY_SITE_LIB "/lib/ruby/site_ruby"
-#define RUBY_SITE_LIB2 "/lib/ruby/site_ruby/@MAJOR@.@MINOR@"
-#define RUBY_PLATFORM "i386-msdosdjgpp"
-#define RUBY_ARCHLIB "/lib/ruby/@MAJOR@.@MINOR@/i386-msdosdjgpp"
-#define RUBY_SITE_ARCHLIB "/lib/ruby/site_ruby/@MAJOR@.@MINOR@/i386-msdosdjgpp"
diff --git a/djgpp/config.sed b/djgpp/config.sed
deleted file mode 100644
index 1805789520..0000000000
--- a/djgpp/config.sed
+++ /dev/null
@@ -1,128 +0,0 @@
-/^SHELL/s,/bin/sh,$(COMSPEC),
-;s%/bin/rm%rm%
-;s%|| true%%
-;/\/dev\/null/ {
-;s,/dev/null 2>&1, nul,
-;s,2> /dev/null,,
-;}
-;/^config.status/ {
-;    N;N;N;N;N;d
-;}
-:t
-  /@[a-zA-Z_][a-zA-Z_0-9]*@/!b
-s,@srcdir@,.,g;t t
-s,@top_srcdir@,..,;t t
-s,@PATH_SEPARATOR@,:,;t t
-s,@PACKAGE_NAME@,,;t t
-s,@PACKAGE_TARNAME@,,;t t
-s,@PACKAGE_VERSION@,,;t t
-s,@PACKAGE_STRING@,,;t t
-s,@PACKAGE_BUGREPORT@,,;t t
-s,@exec_prefix@,${prefix},;t t
-s,@prefix@,/dev/env/DJDIR,;t t
-s%@program_transform_name@%s,^,,%;t t
-s,@bindir@,${exec_prefix}/bin,;t t
-s,@sbindir@,${exec_prefix}/sbin,;t t
-s,@libexecdir@,${exec_prefix}/libexec,;t t
-s,@datadir@,${prefix}/share,;t t
-s,@sysconfdir@,${prefix}/etc,;t t
-s,@sharedstatedir@,${prefix}/com,;t t
-s,@localstatedir@,${prefix}/var,;t t
-s,@libdir@,${exec_prefix}/lib,;t t
-s,@includedir@,${prefix}/include,;t t
-s,@oldincludedir@,/usr/include,;t t
-s,@infodir@,${prefix}/info,;t t
-s,@mandir@,${prefix}/man,;t t
-s,@build_alias@,i586-pc-msdosdjgpp,;t t
-s,@host_alias@,i586-pc-msdosdjgpp,;t t
-s,@target_alias@,i386-msdosdjgpp,;t t
-s,@DEFS@,,;t t
-s,@ECHO_C@,,;t t
-s,@ECHO_N@,-n,;t t
-s,@ECHO_T@,,;t t
-s,@LIBS@,-lm ,;t t
-s,@MAJOR@,1,;t t
-s,@MINOR@,7,;t t
-s,@TEENY@,3,;t t
-s,@build@,i586-pc-msdosdjgpp,;t t
-s,@build_cpu@,i586,;t t
-s,@build_vendor@,pc,;t t
-s,@build_os@,msdosdjgpp,;t t
-s,@host@,i586-pc-msdosdjgpp,;t t
-s,@host_cpu@,i586,;t t
-s,@host_vendor@,pc,;t t
-s,@host_os@,msdosdjgpp,;t t
-s,@target@,i386-pc-msdosdjgpp,;t t
-s,@target_cpu@,i386,;t t
-s,@target_vendor@,pc,;t t
-s,@target_os@,msdosdjgpp,;t t
-s,@CC@,gcc,;t t
-s,@ac_ct_CC@,,;t t
-s,@CFLAGS@,-Os,;t t
-s,@LDFLAGS@,,;t t
-s,@CPPFLAGS@,,;t t
-s,@EXEEXT@,.exe,;t t
-s,@OBJEXT@,o,;t t
-s,@CPP@,gcc -E,;t t
-s,@EGREP@,grep -E,;t t
-s,@GNU_LD@,yes,;t t
-s,@CPPOUTFILE@,-o conftest.i,;t t
-s,@OUTFLAG@,-o ,;t t
-s,@YACC@,bison -y,;t t
-s,@RANLIB@,ranlib,;t t
-s,@ac_ct_RANLIB@,,;t t
-s,@AR@,ar,;t t
-s,@ac_ct_AR@,,;t t
-s,@NM@,,;t t
-s,@ac_ct_NM@,,;t t
-s,@WINDRES@,,;t t
-s,@ac_ct_WINDRES@,,;t t
-s,@DLLWRAP@,,;t t
-s,@ac_ct_DLLWRAP@,,;t t
-s,@LN_S@,ln -s,;t t
-s,@SET_MAKE@,,;t t
-s,@LIBOBJS@,crypt.o flock.o vsnprintf.o,;t t
-s,@ALLOCA@,,;t t
-s,@XCFLAGS@,,;t t
-s,@XLDFLAGS@, -L.,;t t
-s,@DLDFLAGS@,,;t t
-s,@STATIC@,,;t t
-s,@CCDLFLAGS@,,;t t
-s,@LDSHARED@,ld,;t t
-s,@DLEXT@,so,;t t
-s,@DLEXT2@,,;t t
-s,@LIBEXT@,a,;t t
-s,@LINK_SO@,,;t t
-s,@LIBPATHFLAG@, -L%s,;t t
-s,@STRIP@,strip,;t t
-s,@EXTSTATIC@,,;t t
-s,@setup@,Setup.dj,;t t
-s,@MINIRUBY@,./miniruby,;t t
-s,@PREP@,,;t t
-s,@ARCHFILE@,,;t t
-s,@LIBRUBY_LDSHARED@,ld,;t t
-s,@LIBRUBY_DLDFLAGS@,,;t t
-s,@RUBY_INSTALL_NAME@,ruby,;t t
-s,@rubyw_install_name@,,;t t
-s,@RUBYW_INSTALL_NAME@,,;t t
-s,@RUBY_SO_NAME@,$(RUBY_INSTALL_NAME),;t t
-s,@LIBRUBY_A@,lib$(RUBY_INSTALL_NAME).a,;t t
-s,@LIBRUBY_SO@,lib$(RUBY_SO_NAME).so.$(MAJOR).$(MINOR).$(TEENY),;t t
-s,@LIBRUBY_ALIASES@,lib$(RUBY_SO_NAME).so,;t t
-s,@LIBRUBY@,$(LIBRUBY_A),;t t
-s,@LIBRUBYARG@,-l$(RUBY_INSTALL_NAME),;t t
-s,@SOLIBS@,,;t t
-s,@DLDLIBS@,-lc,;t t
-s,@ENABLE_SHARED@,no,;t t
-s,@MAINLIBS@,,;t t
-s,@COMMON_LIBS@,,;t t
-s,@COMMON_MACROS@,,;t t
-s,@COMMON_HEADERS@,,;t t
-s,@EXPORT_PREFIX@,,;t t
-s,@MAKEFILES@,Makefile,;t t
-s,@arch@,i386-msdosdjgpp,;t t
-s,@sitearch@,i386-msdosdjgpp,;t t
-s,@sitedir@,${prefix}/lib/ruby/site_ruby,;t t
-s,@configure_args@,,;t t
-/^,THIS_IS_DUMMY_PATTERN_/i\
-ac_given_srcdir=.
diff --git a/djgpp/configure.bat b/djgpp/configure.bat
deleted file mode 100755
index e6a5d79d4a..0000000000
--- a/djgpp/configure.bat
+++ /dev/null
@@ -1,20 +0,0 @@
-@echo off
-if exist configure.bat cd ..
-if exist djgpp\version.sed goto exist
-  sed -n -f djgpp\mkver.sed < version.h > djgpp\version.sed
-:exist
-set _conv_=-f djgpp\config.sed -f djgpp\version.sed
-sed %_conv_% < Makefile.in > Makefile
-sed %_conv_% < djgpp\config.hin > config.h
-echo LFN check > 12345678
-sed -n /LFN/d 123456789 > nul
-if errorlevel 2 goto LFN
-    copy missing\vsnprintf.c missing\vsnprint.c > nul
-    copy djgpp\config.sed config.sta > nul
-goto end
-:LFN
-    copy djgpp\config.sed config.status > nul
-:end
-set _conv_=
-del 12345678
-echo Now you must run a make.
diff --git a/djgpp/mkver.sed b/djgpp/mkver.sed
deleted file mode 100644
index f29b9ddd3e..0000000000
--- a/djgpp/mkver.sed
+++ /dev/null
@@ -1 +0,0 @@
-/RUBY_VERSION /s/^.*\([0-9][0-9]*\)\.\([0-9][0-9]*\)\.\([0-9][0-9]*\).*$/s,@MAJOR@,\1,;s,@MINOR@,\2,;s,@TEENY@,\3,/p
diff --git a/dln.c b/dln.c
index fd59bdab8e..7acea5f888 100644
--- a/dln.c
+++ b/dln.c
@@ -2,27 +2,32 @@
 
   dln.c -
 
-  $Author: matz $
-  $Date: 2006/08/07 03:43:39 $
+  $Author$
   created at: Tue Jan 18 17:05:06 JST 1994
 
-  Copyright (C) 1993-2003 Yukihiro Matsumoto
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
 
 **********************************************************************/
 
-#include "ruby.h"
+#ifdef RUBY_EXPORT
+#include "ruby/ruby.h"
+#define dln_notimplement rb_notimplement
+#define dln_memerror rb_memerror
+#define dln_exit rb_exit
+#define dln_loaderror rb_loaderror
+#else
+#define dln_notimplement --->>> dln not implemented <<<---
+#define dln_memerror abort
+#define dln_exit exit
+static void dln_loaderror(const char *format, ...);
+#endif
 #include "dln.h"
+#include "internal.h"
 
 #ifdef HAVE_STDLIB_H
 # include <stdlib.h>
 #endif
 
-#ifdef __CHECKER__
-#undef HAVE_DLOPEN
-#undef USE_DLN_A_OUT
-#undef USE_DLN_DLOPEN
-#endif
-
 #ifdef USE_DLN_A_OUT
 char *dln_argv0;
 #endif
@@ -43,15 +48,18 @@ void *xcalloc();
 void *xrealloc();
 #endif
 
+#undef free
+#define free(x) xfree(x)
+
 #include <stdio.h>
-#if defined(_WIN32) || defined(__VMS)
+#if defined(_WIN32)
 #include "missing/file.h"
 #endif
 #include <sys/types.h>
 #include <sys/stat.h>
 
 #ifndef S_ISDIR
-#   define S_ISDIR(m) ((m & S_IFMT) == S_IFDIR)
+#   define S_ISDIR(m) (((m) & S_IFMT) == S_IFDIR)
 #endif
 
 #ifdef HAVE_SYS_PARAM_H
@@ -69,75 +77,74 @@ void *xrealloc();
 char *getenv();
 #endif
 
-#if defined(__VMS)
-#pragma builtins
-#include <dlfcn.h>
-#endif
-
-#ifdef __MACOS__
-# include <TextUtils.h>
-# include <CodeFragments.h>
-# include <Aliases.h>
-# include "macruby_private.h"
+#ifdef __APPLE__
+# if defined(HAVE_DLOPEN)
+   /* Mac OS X with dlopen (10.3 or later) */
+#  define MACOSX_DLOPEN
+# else
+#  define MACOSX_DYLD
+# endif
 #endif
 
-#ifdef __BEOS__
-# include <image.h>
+#ifndef dln_loaderror
+static void
+dln_loaderror(const char *format, ...)
+{
+    va_list ap;
+    va_start(ap, format);
+    vfprintf(stderr, format, ap);
+    va_end(ap);
+    abort();
+}
 #endif
 
-#ifndef NO_DLN_LOAD
-
-#if defined(HAVE_DLOPEN) && !defined(USE_DLN_A_OUT) && !defined(_AIX) && !defined(__APPLE__) && !defined(_UNICOSMP)
+#if defined(HAVE_DLOPEN) && !defined(USE_DLN_A_OUT) && !defined(_AIX) && !defined(MACOSX_DYLD) && !defined(_UNICOSMP)
 /* dynamic load with dlopen() */
 # define USE_DLN_DLOPEN
 #endif
 
-#ifndef FUNCNAME_PATTERN
-# if defined(__hp9000s300) ||  (defined(__NetBSD__) && !defined(__ELF__)) || defined(__BORLANDC__) || (defined(__FreeBSD__) && !defined(__ELF__)) || (defined(__OpenBSD__) && !defined(__ELF__)) || defined(NeXT) || defined(__WATCOMC__) || defined(__APPLE__)
-#  define FUNCNAME_PATTERN "_Init_%s"
-# else
-#  define FUNCNAME_PATTERN "Init_%s"
-# endif
+#if defined(__hp9000s300) || ((defined(__NetBSD__) || defined(__FreeBSD__) || defined(__OpenBSD__)) && !defined(__ELF__)) || defined(NeXT) || defined(MACOSX_DYLD)
+# define EXTERNAL_PREFIX "_"
+#else
+# define EXTERNAL_PREFIX ""
 #endif
+#define FUNCNAME_PREFIX EXTERNAL_PREFIX"Init_"
 
-static int
-init_funcname_len(buf, file)
-    char **buf;
-    const char *file;
-{
-    char *p;
-    const char *slash;
-    int len;
-
-    /* Load the file as an object one */
-    for (slash = file-1; *file; file++) /* Find position of last '/' */
-#ifdef __MACOS__
-	if (*file == ':') slash = file;
+#if defined __CYGWIN__ || defined DOSISH
+#define isdirsep(x) ((x) == '/' || (x) == '\\')
 #else
-	if (*file == '/') slash = file;
+#define isdirsep(x) ((x) == '/')
 #endif
 
-    len = strlen(FUNCNAME_PATTERN) + strlen(slash + 1);
-    *buf = xmalloc(len);
-    snprintf(*buf, len, FUNCNAME_PATTERN, slash + 1);
-    for (p = *buf; *p; p++) {         /* Delete suffix if it exists */
-	if (*p == '.') {
-	    *p = '\0'; break;
-	}
+static size_t
+init_funcname_len(const char **file)
+{
+    const char *p = *file, *base, *dot = NULL;
+
+    /* Load the file as an object one */
+    for (base = p; *p; p++) { /* Find position of last '/' */
+	if (*p == '.' && !dot) dot = p;
+	if (isdirsep(*p)) base = p+1, dot = NULL;
     }
-    return p - *buf;
+    *file = base;
+    /* Delete suffix if it exists */
+    return (dot ? dot : p) - base;
 }
 
+static const char funcname_prefix[sizeof(FUNCNAME_PREFIX) - 1] = FUNCNAME_PREFIX;
+
 #define init_funcname(buf, file) do {\
-    int len = init_funcname_len(buf, file);\
-    char *tmp = ALLOCA_N(char, len+1);\
+    const char *base = (file);\
+    const size_t flen = init_funcname_len(&base);\
+    const size_t plen = sizeof(funcname_prefix);\
+    char *const tmp = ALLOCA_N(char, plen+flen+1);\
     if (!tmp) {\
-	free(*buf);\
-	rb_memerror();\
+	dln_memerror();\
     }\
-    strcpy(tmp, *buf);\
-    free(*buf);\
-    *buf = tmp;\
+    memcpy(tmp, funcname_prefix, plen);\
+    memcpy(tmp+plen, base, flen);\
+    tmp[plen+flen] = '\0';\
+    *(buf) = tmp;\
 } while (0)
 
 #ifdef USE_DLN_A_OUT
@@ -175,8 +182,8 @@ static int dln_init_p = 0;
 
 #define INVALID_OBJECT(h) (N_MAGIC(h) != OMAGIC)
 
-#include "util.h"
-#include "st.h"
+#include "ruby/util.h"
+#include "ruby/st.h"
 
 static st_table *sym_tbl;
 static st_table *undef_tbl;
@@ -184,10 +191,7 @@ static st_table *undef_tbl;
 static int load_lib();
 
 static int
-load_header(fd, hdrp, disp)
-    int fd;
-    struct exec *hdrp;
-    long disp;
+load_header(int fd, struct exec *hdrp, long disp)
 {
     int size;
 
@@ -221,20 +225,20 @@ load_header(fd, hdrp, disp)
 #define RELOC_TARGET_SIZE(r)		((r)->r_length)
 #endif
 
-#if defined(sun) && defined(sparc)
+#if defined(__sun) && defined(__sparc)
 /* Sparc (Sun 4) macros */
 #  undef relocation_info
 #  define relocation_info reloc_info_sparc
 #  define R_RIGHTSHIFT(r)	(reloc_r_rightshift[(r)->r_type])
 #  define R_BITSIZE(r) 		(reloc_r_bitsize[(r)->r_type])
 #  define R_LENGTH(r)		(reloc_r_length[(r)->r_type])
-static int reloc_r_rightshift[] = {
+static const int reloc_r_rightshift[] = {
   0, 0, 0, 0, 0, 0, 2, 2, 10, 0, 0, 0, 0, 0, 0,
 };
-static int reloc_r_bitsize[] = {
+static const int reloc_r_bitsize[] = {
   8, 16, 32, 8, 16, 32, 30, 22, 22, 22, 13, 10, 32, 32, 16,
 };
-static int reloc_r_length[] = {
+static const int reloc_r_length[] = {
   0, 1, 2, 0, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
 };
 #  define R_PCREL(r) \
@@ -257,10 +261,7 @@ static int reloc_r_length[] = {
 #endif
 
 static struct relocation_info *
-load_reloc(fd, hdrp, disp)
-     int fd;
-     struct exec *hdrp;
-     long disp;
+load_reloc(int fd, struct exec *hdrp, long disp)
 {
     struct relocation_info *reloc;
     int size;
@@ -283,10 +284,7 @@ load_reloc(fd, hdrp, disp)
 }
 
 static struct nlist *
-load_sym(fd, hdrp, disp)
-    int fd;
-    struct exec *hdrp;
-    long disp;
+load_sym(int fd, struct exec *hdrp, long disp)
 {
     struct nlist * buffer;
     struct nlist * sym;
@@ -327,9 +325,7 @@ load_sym(fd, hdrp, disp)
 }
 
 static st_table *
-sym_hash(hdrp, syms)
-    struct exec *hdrp;
-    struct nlist *syms;
+sym_hash(struct exec *hdrp, struct nlist *syms)
 {
     st_table *tbl;
     struct nlist *sym = syms;
@@ -349,17 +345,16 @@ sym_hash(hdrp, syms)
 }
 
 static int
-dln_init(prog)
-    const char *prog;
+dln_init(const char *prog)
 {
-    char *file;
+    char *file, fbuf[MAXPATHLEN];
     int fd;
     struct exec hdr;
     struct nlist *syms;
 
     if (dln_init_p == 1) return 0;
 
-    file = dln_find_exe(prog, NULL);
+    file = dln_find_exe_r(prog, NULL, fbuf, sizeof(fbuf));
     if (file == NULL || (fd = open(file, O_RDONLY)) < 0) {
 	dln_errno = errno;
 	return -1;
@@ -424,11 +419,7 @@ dln_init(prog)
 }
 
 static long
-load_text_data(fd, hdrp, bss, disp)
-    int fd;
-    struct exec *hdrp;
-    int bss;
-    long disp;
+load_text_data(int fd, struct exec *hdrp, int bss, long disp)
 {
     int size;
     unsigned char* addr;
@@ -462,27 +453,26 @@ load_text_data(fd, hdrp, bss, disp)
 }
 
 static int
-undef_print(key, value)
-    char *key, *value;
+undef_print(char *key, char *value)
 {
     fprintf(stderr, "  %s\n", key);
     return ST_CONTINUE;
 }
 
 static void
-dln_print_undef()
+dln_print_undef(void)
 {
     fprintf(stderr, " Undefined symbols:\n");
     st_foreach(undef_tbl, undef_print, NULL);
 }
 
 static void
-dln_undefined()
+dln_undefined(void)
 {
     if (undef_tbl->num_entries > 0) {
 	fprintf(stderr, "dln: Calling undefined function\n");
 	dln_print_undef();
-	rb_exit(1);
+	dln_exit(1);
     }
 }
 
@@ -500,10 +490,7 @@ struct undef {
 
 static st_table *reloc_tbl = NULL;
 static void
-link_undef(name, base, reloc)
-    const char *name;
-    long base;
-    struct relocation_info *reloc;
+link_undef(const char *name, long base, struct relocation_info *reloc)
 {
     static int u_no = 0;
     struct undef *obj;
@@ -536,14 +523,11 @@ struct reloc_arg {
 };
 
 static int
-reloc_undef(no, undef, arg)
-    int no;
-    struct undef *undef;
-    struct reloc_arg *arg;
+reloc_undef(int no, struct undef *undef, struct reloc_arg *arg)
 {
     int datum;
     char *address;
-#if defined(sun) && defined(sparc)
+#if defined(__sun) && defined(__sparc)
     unsigned int mask = 0;
 #endif
 
@@ -552,7 +536,7 @@ reloc_undef(no, undef, arg)
     datum = arg->value;
 
     if (R_PCREL(&(undef->reloc))) datum -= undef->base;
-#if defined(sun) && defined(sparc)
+#if defined(__sun) && defined(__sparc)
     datum += undef->reloc.r_addend;
     datum >>= R_RIGHTSHIFT(&(undef->reloc));
     mask = (1 << R_BITSIZE(&(undef->reloc))) - 1;
@@ -600,9 +584,7 @@ reloc_undef(no, undef, arg)
 }
 
 static void
-unlink_undef(name, value)
-    const char *name;
-    long value;
+unlink_undef(const char *name, long value)
 {
     struct reloc_arg arg;
 
@@ -617,10 +599,7 @@ struct indr_data {
 };
 
 static int
-reloc_repl(no, undef, data)
-    int no;
-    struct undef *undef;
-    struct indr_data *data;
+reloc_repl(int no, struct undef *undef, struct indr_data *data)
 {
     if (strcmp(data->name0, undef->name) == 0) {
 	free(undef->name);
@@ -631,12 +610,9 @@ reloc_repl(no, undef, data)
 #endif
 
 static int
-load_1(fd, disp, need_init)
-    int fd;
-    long disp;
-    const char *need_init;
+load_1(int fd, long disp, const char *need_init)
 {
-    static char *libc = LIBC_NAME;
+    static const char *libc = LIBC_NAME;
     struct exec hdr;
     struct relocation_info *reloc = NULL;
     long block = 0;
@@ -784,11 +760,11 @@ load_1(fd, disp, need_init)
 	while (rel < rel_end) {
 	    char *address = (char*)(rel->r_address + block);
 	    long datum = 0;
-#if defined(sun) && defined(sparc)
+#if defined(__sun) && defined(__sparc)
 	    unsigned int mask = 0;
 #endif
 
-	    if(rel >= rel_beg)
+	    if (rel >= rel_beg)
 		address += hdr.a_text;
 
 	    if (rel->r_extern) { /* Look it up in symbol-table */
@@ -805,7 +781,7 @@ load_1(fd, disp, need_init)
 		}
 	    } /* end.. look it up */
 	    else { /* is static */
-		switch (R_SYMBOL(rel)) { 
+		switch (R_SYMBOL(rel)) {
 		  case N_TEXT:
 		  case N_DATA:
 		    datum = block;
@@ -819,7 +795,7 @@ load_1(fd, disp, need_init)
 	    } /* end .. is static */
 	    if (R_PCREL(rel)) datum -= block;
 
-#if defined(sun) && defined(sparc)
+#if defined(__sun) && defined(__sparc)
 	    datum += rel->r_addend;
 	    datum >>= R_RIGHTSHIFT(rel);
 	    mask = (1 << R_BITSIZE(rel)) - 1;
@@ -915,10 +891,7 @@ load_1(fd, disp, need_init)
 
 static int target_offset;
 static int
-search_undef(key, value, lib_tbl)
-    const char *key;
-    int value;
-    st_table *lib_tbl;
+search_undef(const char *key, int value, st_table *lib_tbl)
 {
     long offset;
 
@@ -932,13 +905,13 @@ struct symdef {
     int lib_offset;
 };
 
-char *dln_librrb_ary_path = DLN_DEFAULT_LIB_PATH;
+const char *dln_librrb_ary_path = DLN_DEFAULT_LIB_PATH;
 
 static int
-load_lib(lib)
-    const char *lib;
+load_lib(const char *lib)
 {
-    char *path, *file;
+    char *path, *file, fbuf[MAXPATHLEN];
+    char *envpath = 0;
     char armagic[SARMAG];
     int fd, size;
     struct ar_hdr ahdr;
@@ -968,8 +941,10 @@ load_lib(lib)
     /* if path is still NULL, use "." for path. */
     path = getenv("DLN_LIBRARY_PATH");
     if (path == NULL) path = dln_librrb_ary_path;
+    else path = envpath = strdup(path);
 
-    file = dln_find_file(lib, path);
+    file = dln_find_file_r(lib, path, fbuf, sizeof(fbuf));
+    if (envpath) free(envpath);
     fd = open(file, O_RDONLY);
     if (fd == -1) goto syserr;
     size = read(fd, armagic, SARMAG);
@@ -1073,8 +1048,7 @@ load_lib(lib)
 }
 
 static int
-load(file)
-    const char *file;
+load(const char *file)
 {
     int fd;
     int result;
@@ -1099,8 +1073,7 @@ load(file)
 }
 
 void*
-dln_sym(name)
-    const char *name;
+dln_sym(const char *name)
 {
     struct nlist *sym;
 
@@ -1112,12 +1085,7 @@ dln_sym(name)
 #endif /* USE_DLN_A_OUT */
 
 #ifdef USE_DLN_DLOPEN
-# if defined(__NetBSD__) && defined(__NetBSD_Version__) && __NetBSD_Version__ < 105000000
-#  include <nlist.h>
-#  include <link.h>
-# else
-#  include <dlfcn.h>
-# endif
+# include <dlfcn.h>
 #endif
 
 #ifdef __hpux
@@ -1141,26 +1109,40 @@ dln_sym(name)
 #endif
 #endif
 #else
-#ifdef __APPLE__
+#ifdef MACOSX_DYLD
 #include <mach-o/dyld.h>
 #endif
 #endif
 
-#if defined _WIN32 && !defined __CYGWIN__
+#ifdef _WIN32
 #include <windows.h>
+#include <imagehlp.h>
 #endif
 
-#ifdef _WIN32_WCE
-#undef FormatMessage
-#define FormatMessage FormatMessageA
-#undef LoadLibrary
-#define LoadLibrary LoadLibraryA
-#undef GetProcAddress
-#define GetProcAddress GetProcAddressA
-#endif
-
+#ifdef _WIN32
 static const char *
-dln_strerror()
+dln_strerror(char *message, size_t size)
+{
+    int error = GetLastError();
+    char *p = message;
+    size_t len = snprintf(message, size, "%d: ", error);
+
+#define format_message(sublang) FormatMessage(\
+	FORMAT_MESSAGE_FROM_SYSTEM | FORMAT_MESSAGE_IGNORE_INSERTS,	\
+	NULL, error, MAKELANGID(LANG_NEUTRAL, (sublang)),		\
+	message + len, size - len, NULL)
+    if (format_message(SUBLANG_ENGLISH_US) == 0)
+	format_message(SUBLANG_DEFAULT);
+    for (p = message + len; *p; p++) {
+	if (*p == '\n' || *p == '\r')
+	    *p = ' ';
+    }
+    return message;
+}
+#define dln_strerror() dln_strerror(message, sizeof message)
+#elif ! defined _AIX
+static const char *
+dln_strerror(void)
 {
 #ifdef USE_DLN_A_OUT
     char *strerror();
@@ -1186,127 +1168,140 @@ dln_strerror()
 #ifdef USE_DLN_DLOPEN
     return (char*)dlerror();
 #endif
-
-#if defined _WIN32 && !defined __CYGWIN__
-    static char message[1024];
-    int error = GetLastError();
-    char *p = message;
-    p += sprintf(message, "%d: ", error);
-    FormatMessage(
-	FORMAT_MESSAGE_FROM_SYSTEM	 | FORMAT_MESSAGE_IGNORE_INSERTS,
-	NULL,
-	error,
-	MAKELANGID(LANG_NEUTRAL, SUBLANG_DEFAULT),
-	p,
-	sizeof message - strlen(message),
-	NULL);
-
-    for (p = message; *p; p++) {
-	if (*p == '\n' || *p == '\r')
-	    *p = ' ';
-    }
-    return message;
-#endif
 }
-
+#endif
 
 #if defined(_AIX) && ! defined(_IA64)
 static void
 aix_loaderror(const char *pathname)
 {
-    char *message[8], errbuf[1024];
-    int i,j;
-
-    struct errtab { 
-	int errnum;
-	char *errstr;
-    } load_errtab[] = {
-	{L_ERROR_TOOMANY,	"too many errors, rest skipped."},
-	{L_ERROR_NOLIB,		"can't load library:"},
-	{L_ERROR_UNDEF,		"can't find symbol in library:"},
-	{L_ERROR_RLDBAD,
-	     "RLD index out of range or bad relocation type:"},
-	{L_ERROR_FORMAT,	"not a valid, executable xcoff file:"},
-	{L_ERROR_MEMBER,
-	     "file not an archive or does not contain requested member:"},
-	{L_ERROR_TYPE,		"symbol table mismatch:"},
-	{L_ERROR_ALIGN,		"text alignment in file is wrong."},
-	{L_ERROR_SYSTEM,	"System error:"},
-	{L_ERROR_ERRNO,		NULL}
-    };
-
-#define LOAD_ERRTAB_LEN	(sizeof(load_errtab)/sizeof(load_errtab[0]))
-#define ERRBUF_APPEND(s) strncat(errbuf, s, sizeof(errbuf)-strlen(errbuf)-1)
-
-    snprintf(errbuf, 1024, "load failed - %s ", pathname);
-
-    if (!loadquery(1, &message[0], sizeof(message))) 
-	ERRBUF_APPEND(strerror(errno));
-    for(i = 0; message[i] && *message[i]; i++) {
-	int nerr = atoi(message[i]);
-	for (j=0; j<LOAD_ERRTAB_LEN; j++) {
-           if (nerr == load_errtab[i].errnum && load_errtab[i].errstr)
-		ERRBUF_APPEND(load_errtab[i].errstr);
+    char *message[1024], errbuf[1024];
+    int i;
+#define ERRBUF_APPEND(s) strlcat(errbuf, (s), sizeof(errbuf))
+    snprintf(errbuf, sizeof(errbuf), "load failed - %s. ", pathname);
+
+    if (loadquery(L_GETMESSAGES, &message[0], sizeof(message)) != -1) {
+	ERRBUF_APPEND("Please issue below command for detailed reasons:\n\t");
+	ERRBUF_APPEND("/usr/sbin/execerror ruby ");
+	for (i=0; message[i]; i++) {
+	    ERRBUF_APPEND("\"");
+	    ERRBUF_APPEND(message[i]);
+	    ERRBUF_APPEND("\" ");
 	}
-	while (isdigit(*message[i])) message[i]++; 
-	ERRBUF_APPEND(message[i]);
 	ERRBUF_APPEND("\n");
     }
-    errbuf[strlen(errbuf)-1] = '\0';	/* trim off last newline */
-    rb_loaderror(errbuf);
-    return;
+    else {
+	ERRBUF_APPEND(strerror(errno));
+	ERRBUF_APPEND("[loadquery failed]");
+    }
+    dln_loaderror("%s", errbuf);
+}
+#endif
+
+#if defined _WIN32 && defined RUBY_EXPORT
+HANDLE rb_libruby_handle(void);
+
+static int
+rb_w32_check_imported(HMODULE ext, HMODULE mine)
+{
+    ULONG size;
+    const IMAGE_IMPORT_DESCRIPTOR *desc;
+
+    desc = ImageDirectoryEntryToData(ext, TRUE, IMAGE_DIRECTORY_ENTRY_IMPORT, &size);
+    if (!desc) return 0;
+    while (desc->Name) {
+	PIMAGE_THUNK_DATA pint = (PIMAGE_THUNK_DATA)((char *)ext + desc->Characteristics);
+	PIMAGE_THUNK_DATA piat = (PIMAGE_THUNK_DATA)((char *)ext + desc->FirstThunk);
+	for (; piat->u1.Function; piat++, pint++) {
+	    static const char prefix[] = "rb_";
+	    PIMAGE_IMPORT_BY_NAME pii;
+	    const char *name;
+
+	    if (IMAGE_SNAP_BY_ORDINAL(pint->u1.Ordinal)) continue;
+	    pii = (PIMAGE_IMPORT_BY_NAME)((char *)ext + (size_t)pint->u1.AddressOfData);
+	    name = (const char *)pii->Name;
+	    if (strncmp(name, prefix, sizeof(prefix) - 1) == 0) {
+		FARPROC addr = GetProcAddress(mine, name);
+		if (addr) return (FARPROC)piat->u1.Function == addr;
+	    }
+	}
+	desc++;
+    }
+    return 1;
 }
 #endif
 
-#if defined(__VMS)
-#include <starlet.h>
-#include <rms.h>
-#include <stsdef.h>
-#include <unixlib.h>
-#include <descrip.h>
-#include <lib$routines.h>
-
-static char *vms_filespec;
-static int vms_fileact(char *filespec, int type);
-static long vms_fisexh(long *sigarr, long *mecarr);
+#if defined(DLN_NEEDS_ALT_SEPARATOR) && DLN_NEEDS_ALT_SEPARATOR
+#define translit_separator(src) do { \
+	char *tmp = ALLOCA_N(char, strlen(src) + 1), *p = tmp, c; \
+	do { \
+	    *p++ = ((c = *file++) == '/') ? DLN_NEEDS_ALT_SEPARATOR : c; \
+	} while (c); \
+	(src) = tmp; \
+    } while (0)
+#else
+#define translit_separator(str) (void)(str)
 #endif
 
-#endif /* NO_DLN_LOAD */
+#ifdef USE_DLN_DLOPEN
+COMPILER_WARNING_PUSH
+#if defined(__clang__) || GCC_VERSION_SINCE(4, 2, 0)
+COMPILER_WARNING_IGNORED(-Wpedantic)
+#endif
+static bool
+dln_incompatible_library_p(void *handle)
+{
+    void *ex = dlsym(handle, EXTERNAL_PREFIX"ruby_xmalloc");
+    return ex && ex != ruby_xmalloc;
+}
+COMPILER_WARNING_POP
+#endif
 
 void*
-dln_load(file)
-    const char *file;
+dln_load(const char *file)
 {
-#ifdef NO_DLN_LOAD
-    rb_raise(rb_eLoadError, "this executable file can't load extension libraries");
-#else
-
+#if (defined _WIN32 || defined USE_DLN_DLOPEN) && defined RUBY_EXPORT
+    static const char incompatible[] = "incompatible library version";
+#endif
 #if !defined(_AIX) && !defined(NeXT)
     const char *error = 0;
-#define DLN_ERROR() (error = dln_strerror(), strcpy(ALLOCA_N(char, strlen(error) + 1), error))
 #endif
 
-#if defined _WIN32 && !defined __CYGWIN__
+#if defined _WIN32
     HINSTANCE handle;
-    char winfile[MAXPATHLEN];
+    WCHAR *winfile;
+    char message[1024];
     void (*init_fct)();
     char *buf;
 
-    if (strlen(file) >= MAXPATHLEN) rb_loaderror("filename too long");
-
     /* Load the file as an object one */
     init_funcname(&buf, file);
 
-    strcpy(winfile, file);
+    /* Convert the file path to wide char */
+    winfile = rb_w32_mbstr_to_wstr(CP_UTF8, file, -1, NULL);
+    if (!winfile) {
+	dln_memerror();
+    }
 
     /* Load file */
-    if ((handle = LoadLibrary(winfile)) == NULL) {
+    handle = LoadLibraryW(winfile);
+    free(winfile);
+
+    if (!handle) {
 	error = dln_strerror();
 	goto failed;
     }
 
+#if defined _WIN32 && defined RUBY_EXPORT
+    if (!rb_w32_check_imported(handle, rb_libruby_handle())) {
+	FreeLibrary(handle);
+	error = incompatible;
+	goto failed;
+    }
+#endif
+
     if ((init_fct = (void(*)())GetProcAddress(handle, buf)) == NULL) {
-	rb_loaderror("%s - %s\n%s", dln_strerror(), buf, file);
+	dln_loaderror("%s - %s\n%s", dln_strerror(), buf, file);
     }
 
     /* Call the init code */
@@ -1324,6 +1319,7 @@ dln_load(file)
     char *buf;
     /* Load the file as an object one */
     init_funcname(&buf, file);
+    translit_separator(file);
 
 #ifdef USE_DLN_DLOPEN
 #define DLN_DEFINED
@@ -1346,10 +1342,28 @@ dln_load(file)
 	    error = dln_strerror();
 	    goto failed;
 	}
+# if defined RUBY_EXPORT
+	{
+	    if (dln_incompatible_library_p(handle)) {
+
+#   if defined __APPLE__ && \
+    defined(MAC_OS_X_VERSION_MIN_REQUIRED) && \
+    (MAC_OS_X_VERSION_MIN_REQUIRED < MAC_OS_X_VERSION_10_11)
+		/* dlclose() segfaults */
+		rb_fatal("%s - %s", incompatible, file);
+#   else
+		dlclose(handle);
+		error = incompatible;
+		goto failed;
+#   endif
+	    }
+	}
+# endif
 
-	init_fct = (void(*)())dlsym(handle, buf);
+	init_fct = (void(*)())(VALUE)dlsym(handle, buf);
 	if (init_fct == NULL) {
-	    error = DLN_ERROR();
+	    const size_t errlen = strlen(error = dln_strerror()) + 1;
+	    error = memcpy(ALLOCA_N(char, errlen), error, errlen);
 	    dlclose(handle);
 	    goto failed;
 	}
@@ -1371,14 +1385,14 @@ dln_load(file)
 	lib = shl_load(file, flags, 0);
 	if (lib == NULL) {
 	    extern int errno;
-	    rb_loaderror("%s - %s", strerror(errno), file);
+	    dln_loaderror("%s - %s", strerror(errno), file);
 	}
 	shl_findsym(&lib, buf, TYPE_PROCEDURE, (void*)&init_fct);
 	if (init_fct == NULL) {
 	    shl_findsym(&lib, buf, TYPE_UNDEFINED, (void*)&init_fct);
 	    if (init_fct == NULL) {
 		errno = ENOSYM;
-		rb_loaderror("%s - %s", strerror(ENOSYM), file);
+		dln_loaderror("%s - %s", strerror(ENOSYM), file);
 	    }
 	}
 	(*init_fct)();
@@ -1403,54 +1417,17 @@ dln_load(file)
     }
 #endif /* _AIX */
 
-#if defined(NeXT) || defined(__APPLE__)
+#if defined(MACOSX_DYLD)
 #define DLN_DEFINED
 /*----------------------------------------------------
    By SHIROYAMA Takayuki Psi@fortune.nest.or.jp
- 
+
    Special Thanks...
     Yu tomoak-i@is.aist-nara.ac.jp,
     Mi hisho@tasihara.nest.or.jp,
     sunshine@sunshineco.com,
     and... Miss ARAI Akino(^^;)
  ----------------------------------------------------*/
-#if defined(NeXT) && (NS_TARGET_MAJOR < 4)/* NeXTSTEP rld functions */
-
-    {
-        NXStream* s;
-	unsigned long init_address;
-	char *object_files[2] = {NULL, NULL};
-
-	void (*init_fct)();
-	
-	object_files[0] = (char*)file;
-	
-	s = NXOpenFile(2,NX_WRITEONLY);
-
-	/* Load object file, if return value ==0 ,  load failed*/
-	if(rld_load(s, NULL, object_files, NULL) == 0) {
-	    NXFlush(s);
-	    NXClose(s);
-	    rb_loaderror("Failed to load %.200s", file);
-	}
-
-	/* lookup the initial function */
-	if(rld_lookup(s, buf, &init_address) == 0) {
-	    NXFlush(s);
-	    NXClose(s);
-	    rb_loaderror("Failed to lookup Init function %.200s", file);
-	}
-
-	NXFlush(s);
-	NXClose(s);
-
-	/* Cannot call *init_address directory, so copy this value to
-	   funtion pointer */
-	init_fct = (void(*)())init_address;
-	(*init_fct)();
-	return (void*)init_address;
-    }
-#else/* OPENSTEP dyld functions */
     {
 	int dyld_result;
 	NSObjectFileImage obj_file; /* handle, but not use it */
@@ -1463,398 +1440,32 @@ dln_load(file)
 	dyld_result = NSCreateObjectFileImageFromFile(file, &obj_file);
 
 	if (dyld_result != NSObjectFileImageSuccess) {
-	    rb_loaderror("Failed to load %.200s", file);
+	    dln_loaderror("Failed to load %.200s", file);
 	}
 
 	NSLinkModule(obj_file, file, NSLINKMODULE_OPTION_BINDNOW);
 
 	/* lookup the initial function */
-	if(!NSIsSymbolNameDefined(buf)) {
-	    rb_loaderror("Failed to lookup Init function %.200s",file);
-	}	
+	if (!NSIsSymbolNameDefined(buf)) {
+	    dln_loaderror("Failed to lookup Init function %.200s",file);
+	}
 	init_fct = NSAddressOfSymbol(NSLookupAndBindSymbol(buf));
 	(*init_fct)();
 
 	return (void*)init_fct;
     }
-#endif /* rld or dyld */
 #endif
 
-#ifdef __BEOS__
-# define DLN_DEFINED
-    {
-      status_t err_stat;  /* BeOS error status code */
-      image_id img_id;    /* extention module unique id */
-      void (*init_fct)(); /* initialize function for extention module */
-
-      /* load extention module */
-      img_id = load_add_on(file);
-      if (img_id <= 0) {
-	rb_loaderror("Failed to load %.200s", file);
-      }
-      
-      /* find symbol for module initialize function. */
-      /* The Be Book KernelKit Images section described to use
-	 B_SYMBOL_TYPE_TEXT for symbol of function, not
-	 B_SYMBOL_TYPE_CODE. Why ? */
-      /* strcat(init_fct_symname, "__Fv"); */  /* parameter nothing. */
-      /* "__Fv" dont need! The Be Book Bug ? */
-      err_stat = get_image_symbol(img_id, buf,
-				  B_SYMBOL_TYPE_TEXT, (void **)&init_fct);
-
-      if (err_stat != B_NO_ERROR) {
-	char real_name[MAXPATHLEN];
-
-	strcpy(real_name, buf);
-	strcat(real_name, "__Fv");
-        err_stat = get_image_symbol(img_id, real_name,
-				    B_SYMBOL_TYPE_TEXT, (void **)&init_fct);
-      }
-
-      if ((B_BAD_IMAGE_ID == err_stat) || (B_BAD_INDEX == err_stat)) {
-	unload_add_on(img_id);
-	rb_loaderror("Failed to lookup Init function %.200s", file);
-      }
-      else if (B_NO_ERROR != err_stat) {
-	char errmsg[] = "Internal of BeOS version. %.200s (symbol_name = %s)";
-	unload_add_on(img_id);
-	rb_loaderror(errmsg, strerror(err_stat), buf);
-      }
-
-      /* call module initialize function. */
-      (*init_fct)();
-      return (void*)img_id;
-    }
-#endif /* __BEOS__*/
-
-#ifdef __MACOS__
-# define DLN_DEFINED
-    {
-      OSErr err;
-      FSSpec libspec;
-      CFragConnectionID connID;
-      Ptr mainAddr;
-      char errMessage[1024];
-      Boolean isfolder, didsomething;
-      Str63 fragname;
-      Ptr symAddr;
-      CFragSymbolClass class;
-      void (*init_fct)();
-      char fullpath[MAXPATHLEN];
-
-      strcpy(fullpath, file);
-
-      /* resolve any aliases to find the real file */
-      c2pstr(fullpath);
-      (void)FSMakeFSSpec(0, 0, fullpath, &libspec);
-      err = ResolveAliasFile(&libspec, 1, &isfolder, &didsomething);
-      if (err) {
-	  rb_loaderror("Unresolved Alias - %s", file);
-      }
-
-      /* Load the fragment (or return the connID if it is already loaded */
-      fragname[0] = 0;
-      err = GetDiskFragment(&libspec, 0, 0, fragname, 
-			    kLoadCFrag, &connID, &mainAddr,
-			    errMessage);
-      if (err) {
-	  p2cstr(errMessage);
-	  rb_loaderror("%s - %s",errMessage , file);
-      }
-
-      /* Locate the address of the correct init function */
-      c2pstr(buf);
-      err = FindSymbol(connID, buf, &symAddr, &class);
-      if (err) {
-	  rb_loaderror("Unresolved symbols - %s" , file);
-      }
-      init_fct = (void (*)())symAddr;
-      (*init_fct)();
-      return (void*)init_fct;
-    }
-#endif /* __MACOS__ */
-
-#if defined(__VMS)
-#define DLN_DEFINED
-    {
-	long status;
-	void (*init_fct)();
-	char *fname, *p1, *p2;
-
-	$DESCRIPTOR(fname_d, "");
-	$DESCRIPTOR(image_d, "");
-	$DESCRIPTOR(buf_d, "");
-
-	decc$to_vms(file, vms_fileact, 0, 0);
-
-	fname = (char *)__alloca(strlen(file)+1);
-	strcpy(fname,file);
-	if (p1 = strrchr(fname,'/'))
-	    fname = p1 + 1;
-	if (p2 = strrchr(fname,'.'))
-	    *p2 = '\0';
-
-	fname_d.dsc$w_length  = strlen(fname);
-	fname_d.dsc$a_pointer = fname;
-	image_d.dsc$w_length  = strlen(vms_filespec);
-	image_d.dsc$a_pointer = vms_filespec;
-	buf_d.dsc$w_length    = strlen(buf);
-	buf_d.dsc$a_pointer   = buf;
-
-	lib$establish(vms_fisexh);
-
-	status = lib$find_image_symbol (
-		     &fname_d,
-		     &buf_d, 
-		     &init_fct, 
-		     &image_d);
-
-	lib$establish(0);
-
-	if (status == RMS$_FNF) {
-	    error = dln_strerror();
-	    goto failed;
-	} else if (!$VMS_STATUS_SUCCESS(status)) {
-	    error = DLN_ERROR();
-	    goto failed;
-	}
-
-	/* Call the init code */
-	(*init_fct)();
-
-	return 1;
-    }
-#endif /* __VMS */
-
 #ifndef DLN_DEFINED
-    rb_notimplement();
+    dln_notimplement();
 #endif
 
 #endif /* USE_DLN_A_OUT */
 #endif
 #if !defined(_AIX) && !defined(NeXT)
   failed:
-    rb_loaderror("%s - %s", error, file);
+    dln_loaderror("%s - %s", error, file);
 #endif
 
-#endif /* NO_DLN_LOAD */
     return 0;			/* dummy return */
 }
-
-static char *dln_find_1();
-
-char *
-dln_find_exe(fname, path)
-    const char *fname;
-    const char *path;
-{
-    if (!path) {
-	path = getenv(PATH_ENV);
-    }
-
-    if (!path) {
-#if defined(MSDOS) || defined(_WIN32) || defined(__human68k__) || defined(__MACOS__)
-	path = "/usr/local/bin;/usr/ucb;/usr/bin;/bin;.";
-#else
-	path = "/usr/local/bin:/usr/ucb:/usr/bin:/bin:.";
-#endif
-    }
-    return dln_find_1(fname, path, 1);
-}
-
-char *
-dln_find_file(fname, path)
-    const char *fname;
-    const char *path;
-{
-#ifndef __MACOS__
-    if (!path) path = ".";
-    return dln_find_1(fname, path, 0);
-#else
-    if (!path) path = ".";
-    return _macruby_path_conv_posix_to_macos(dln_find_1(fname, path, 0));
-#endif
-}
-
-static char fbuf[MAXPATHLEN];
-
-static char *
-dln_find_1(fname, path, exe_flag)
-    char *fname;
-    char *path;
-    int exe_flag;		/* non 0 if looking for executable. */
-{
-    register char *dp;
-    register char *ep;
-    register char *bp;
-    struct stat st;
-#ifdef __MACOS__
-    const char* mac_fullpath;
-#endif
-
-    if (!fname) return fname;
-    if (fname[0] == '/') return fname;
-    if (strncmp("./", fname, 2) == 0 || strncmp("../", fname, 3) == 0)
-      return fname;
-    if (exe_flag && strchr(fname, '/')) return fname;
-#ifdef DOSISH
-    if (fname[0] == '\\') return fname;
-# ifdef DOSISH_DRIVE_LETTER
-    if (strlen(fname) > 2 && fname[1] == ':') return fname;
-# endif
-    if (strncmp(".\\", fname, 2) == 0 || strncmp("..\\", fname, 3) == 0)
-      return fname;
-    if (exe_flag && strchr(fname, '\\')) return fname;
-#endif
-
-    for (dp = path;; dp = ++ep) {
-	register int l;
-	int i;
-	int fspace;
-
-	/* extract a component */
-	ep = strchr(dp, PATH_SEP[0]);
-	if (ep == NULL)
-	    ep = dp+strlen(dp);
-
-	/* find the length of that component */
-	l = ep - dp;
-	bp = fbuf;
-	fspace = sizeof fbuf - 2;
-	if (l > 0) {
-	    /*
-	    **	If the length of the component is zero length,
-	    **	start from the current directory.  If the
-	    **	component begins with "~", start from the
-	    **	user's $HOME environment variable.  Otherwise
-	    **	take the path literally.
-	    */
-
-	    if (*dp == '~' && (l == 1 ||
-#if defined(DOSISH)
-			       dp[1] == '\\' || 
-#endif
-			       dp[1] == '/')) {
-		char *home;
-
-		home = getenv("HOME");
-		if (home != NULL) {
-		    i = strlen(home);
-		    if ((fspace -= i) < 0)
-			goto toolong;
-		    memcpy(bp, home, i);
-		    bp += i;
-		}
-		dp++;
-		l--;
-	    }
-	    if (l > 0) {
-		if ((fspace -= l) < 0)
-		    goto toolong;
-		memcpy(bp, dp, l);
-		bp += l;
-	    }
-
-	    /* add a "/" between directory and filename */
-	    if (ep[-1] != '/')
-		*bp++ = '/';
-	}
-
-	/* now append the file name */
-	i = strlen(fname);
-	if ((fspace -= i) < 0) {
-	  toolong:
-	    fprintf(stderr, "openpath: pathname too long (ignored)\n");
-	    *bp = '\0';
-	    fprintf(stderr, "\tDirectory \"%s\"\n", fbuf);
-	    fprintf(stderr, "\tFile \"%s\"\n", fname);
-	    goto next;
-	}
-	memcpy(bp, fname, i + 1);
-
-#ifndef __MACOS__
-	if (stat(fbuf, &st) == 0) {
-	    if (exe_flag == 0) return fbuf;
-	    /* looking for executable */
-	    if (!S_ISDIR(st.st_mode) && eaccess(fbuf, X_OK) == 0)
-		return fbuf;
-	}
-#else
-	if (mac_fullpath = _macruby_exist_file_in_libdir_as_posix_name(fbuf)) {
-	    if (exe_flag == 0) return mac_fullpath;
-	    /* looking for executable */
-	    if (stat(mac_fullpath, &st) == 0) {
-		if (!S_ISDIR(st.st_mode) && eaccess(mac_fullpath, X_OK) == 0)
-		    return mac_fullpath;
-	    }
-	}
-#endif
-#if defined(DOSISH)
-	if (exe_flag) {
-	    static const char *extension[] = {
-#if defined(MSDOS)
-		".com", ".exe", ".bat",
-#if defined(DJGPP)
-		".btm", ".sh", ".ksh", ".pl", ".sed",
-#endif
-#elif defined(__EMX__) || defined(_WIN32)
-		".exe", ".com", ".cmd", ".bat",
-/* end of __EMX__ or _WIN32 */
-#else
-		".r", ".R", ".x", ".X", ".bat", ".BAT",
-/* __human68k__ */
-#endif
-		(char *) NULL
-	    };
-	    int j;
-
-	    for (j = 0; extension[j]; j++) {
-		if (fspace < strlen(extension[j])) {
-		    fprintf(stderr, "openpath: pathname too long (ignored)\n");
-		    fprintf(stderr, "\tDirectory \"%.*s\"\n", (int) (bp - fbuf), fbuf);
-		    fprintf(stderr, "\tFile \"%s%s\"\n", fname, extension[j]);
-		    continue;
-		}
-		strcpy(bp + i, extension[j]);
-#ifndef __MACOS__
-		if (stat(fbuf, &st) == 0)
-		    return fbuf;
-#else
-		if (mac_fullpath = _macruby_exist_file_in_libdir_as_posix_name(fbuf))
-		    return mac_fullpath;
-
-#endif
-	    }
-	}
-#endif /* MSDOS or _WIN32 or __human68k__ or __EMX__ */
-
-      next:
-	/* if not, and no other alternatives, life is bleak */
-	if (*ep == '\0') {
-	    return NULL;
-	}
-
-	/* otherwise try the next component in the search path */
-    }
-}
-
-#if defined(__VMS)
-
-/* action routine for decc$to_vms */
-static int vms_fileact(char *filespec, int type)
-{
-    if (vms_filespec)
-	free(vms_filespec);
-    vms_filespec = malloc(strlen(filespec)+1);
-    strcpy(vms_filespec, filespec);
-    return 1;
-}
-
-/* exception handler for LIB$FIND_IMAGE_SYMBOL */
-static long vms_fisexh(long *sigarr, long *mecarr)
-{
-    sys$unwind(1, 0);
-    return 1;
-}
-
-#endif /* __VMS */
diff --git a/dln.h b/dln.h
index 3d52ea2827..d98b2607e2 100644
--- a/dln.h
+++ b/dln.h
@@ -2,11 +2,10 @@
 
   dln.h -
 
-  $Author: michal $
-  $Date: 2003/01/16 07:34:01 $
+  $Author$
   created at: Wed Jan 19 16:53:09 JST 1994
 
-  Copyright (C) 1993-2003 Yukihiro Matsumoto
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
 
 **********************************************************************/
 
@@ -29,12 +28,24 @@
 # define _(args) ()
 #endif
 
-char *dln_find_exe _((const char*,const char*));
-char *dln_find_file _((const char*,const char*));
+RUBY_SYMBOL_EXPORT_BEGIN
+
+#ifndef DLN_FIND_EXTRA_ARG
+#define DLN_FIND_EXTRA_ARG
+#endif
+#ifndef DLN_FIND_EXTRA_ARG_DECL
+#define DLN_FIND_EXTRA_ARG_DECL
+#endif
+
+char *dln_find_exe_r(const char*,const char*,char*,size_t DLN_FIND_EXTRA_ARG_DECL);
+char *dln_find_file_r(const char*,const char*,char*,size_t DLN_FIND_EXTRA_ARG_DECL);
 
 #ifdef USE_DLN_A_OUT
 extern char *dln_argv0;
 #endif
 
-void *dln_load _((const char*));
+void *dln_load(const char*);
+
+RUBY_SYMBOL_EXPORT_END
+
 #endif
diff --git a/dln_find.c b/dln_find.c
new file mode 100644
index 0000000000..b08612764e
--- /dev/null
+++ b/dln_find.c
@@ -0,0 +1,290 @@
+/**********************************************************************
+
+  dln_find.c -
+
+  $Author$
+  created at: Tue Jan 18 17:05:06 JST 1994
+
+  Copyright (C) 1993-2007 Yukihiro Matsumoto
+
+**********************************************************************/
+
+#ifdef RUBY_EXPORT
+#include "ruby/ruby.h"
+#define dln_warning rb_warning
+#define dln_warning_arg
+#else
+#define dln_warning fprintf
+#define dln_warning_arg stderr,
+#endif
+#include "dln.h"
+
+#ifdef HAVE_STDLIB_H
+# include <stdlib.h>
+#endif
+
+#ifdef USE_DLN_A_OUT
+char *dln_argv0;
+#endif
+
+#if defined(HAVE_ALLOCA_H)
+#include <alloca.h>
+#endif
+
+#ifdef HAVE_STRING_H
+# include <string.h>
+#else
+# include <strings.h>
+#endif
+
+#include <stdio.h>
+#if defined(_WIN32)
+#include "missing/file.h"
+#endif
+#include <sys/types.h>
+#include <sys/stat.h>
+
+#ifndef S_ISDIR
+#   define S_ISDIR(m) (((m) & S_IFMT) == S_IFDIR)
+#endif
+
+#ifdef HAVE_UNISTD_H
+# include <unistd.h>
+#endif
+
+#if !defined(_WIN32) && !HAVE_DECL_GETENV
+char *getenv();
+#endif
+
+static char *dln_find_1(const char *fname, const char *path, char *buf, size_t size, int exe_flag
+			DLN_FIND_EXTRA_ARG_DECL);
+
+char *
+dln_find_exe_r(const char *fname, const char *path, char *buf, size_t size
+	       DLN_FIND_EXTRA_ARG_DECL)
+{
+    char *envpath = 0;
+
+    if (!path) {
+	path = getenv(PATH_ENV);
+	if (path) path = envpath = strdup(path);
+    }
+
+    if (!path) {
+	path =
+	    "/usr/local/bin" PATH_SEP
+	    "/usr/ucb" PATH_SEP
+	    "/usr/bin" PATH_SEP
+	    "/bin" PATH_SEP
+	    ".";
+    }
+    buf = dln_find_1(fname, path, buf, size, 1 DLN_FIND_EXTRA_ARG);
+    if (envpath) free(envpath);
+    return buf;
+}
+
+char *
+dln_find_file_r(const char *fname, const char *path, char *buf, size_t size
+		DLN_FIND_EXTRA_ARG_DECL)
+{
+    if (!path) path = ".";
+    return dln_find_1(fname, path, buf, size, 0 DLN_FIND_EXTRA_ARG);
+}
+
+static char *
+dln_find_1(const char *fname, const char *path, char *fbuf, size_t size,
+	   int exe_flag /* non 0 if looking for executable. */
+	   DLN_FIND_EXTRA_ARG_DECL)
+{
+    register const char *dp;
+    register const char *ep;
+    register char *bp;
+    struct stat st;
+    size_t i, fnlen, fspace;
+#ifdef DOSISH
+    static const char extension[][5] = {
+	EXECUTABLE_EXTS,
+    };
+    size_t j;
+    int is_abs = 0, has_path = 0;
+    const char *ext = 0;
+#endif
+    const char *p = fname;
+
+    static const char pathname_too_long[] = "openpath: pathname too long (ignored)\n\
+\tDirectory \"%.*s\"%s\n\tFile \"%.*s\"%s\n";
+#define PATHNAME_TOO_LONG() dln_warning(dln_warning_arg pathname_too_long, \
+					((bp - fbuf) > 100 ? 100 : (int)(bp - fbuf)), fbuf, \
+					((bp - fbuf) > 100 ? "..." : ""), \
+					(fnlen > 100 ? 100 : (int)fnlen), fname, \
+					(fnlen > 100 ? "..." : ""))
+
+#define RETURN_IF(expr) if (expr) return (char *)fname;
+
+    RETURN_IF(!fname);
+    fnlen = strlen(fname);
+    if (fnlen >= size) {
+	dln_warning(dln_warning_arg
+		    "openpath: pathname too long (ignored)\n\tFile \"%.*s\"%s\n",
+		    (fnlen > 100 ? 100 : (int)fnlen), fname,
+		    (fnlen > 100 ? "..." : ""));
+	return NULL;
+    }
+#ifdef DOSISH
+# ifndef CharNext
+# define CharNext(p) ((p)+1)
+# endif
+# ifdef DOSISH_DRIVE_LETTER
+    if (((p[0] | 0x20) - 'a') < 26  && p[1] == ':') {
+	p += 2;
+	is_abs = 1;
+    }
+# endif
+    switch (*p) {
+      case '/': case '\\':
+	is_abs = 1;
+	p++;
+    }
+    has_path = is_abs;
+    while (*p) {
+	switch (*p) {
+	  case '/': case '\\':
+	    has_path = 1;
+	    ext = 0;
+	    p++;
+	    break;
+	  case '.':
+	    ext = p;
+	    p++;
+	    break;
+	  default:
+	    p = CharNext(p);
+	}
+    }
+    if (ext) {
+	for (j = 0; STRCASECMP(ext, extension[j]); ) {
+	    if (++j == sizeof(extension) / sizeof(extension[0])) {
+		ext = 0;
+		break;
+	    }
+	}
+    }
+    ep = bp = 0;
+    if (!exe_flag) {
+	RETURN_IF(is_abs);
+    }
+    else if (has_path) {
+	RETURN_IF(ext);
+	i = p - fname;
+	if (i + 1 > size) goto toolong;
+	fspace = size - i - 1;
+	bp = fbuf;
+	ep = p;
+	memcpy(fbuf, fname, i + 1);
+	goto needs_extension;
+    }
+    p = fname;
+#endif
+
+    if (*p == '.' && *++p == '.') ++p;
+    RETURN_IF(*p == '/');
+    RETURN_IF(exe_flag && strchr(fname, '/'));
+
+#undef RETURN_IF
+
+    for (dp = path;; dp = ++ep) {
+	register size_t l;
+
+	/* extract a component */
+	ep = strchr(dp, PATH_SEP[0]);
+	if (ep == NULL)
+	    ep = dp+strlen(dp);
+
+	/* find the length of that component */
+	l = ep - dp;
+	bp = fbuf;
+	fspace = size - 2;
+	if (l > 0) {
+	    /*
+	    **	If the length of the component is zero length,
+	    **	start from the current directory.  If the
+	    **	component begins with "~", start from the
+	    **	user's $HOME environment variable.  Otherwise
+	    **	take the path literally.
+	    */
+
+	    if (*dp == '~' && (l == 1 ||
+#if defined(DOSISH)
+			       dp[1] == '\\' ||
+#endif
+			       dp[1] == '/')) {
+		char *home;
+
+		home = getenv("HOME");
+		if (home != NULL) {
+		    i = strlen(home);
+		    if (fspace < i)
+			goto toolong;
+		    fspace -= i;
+		    memcpy(bp, home, i);
+		    bp += i;
+		}
+		dp++;
+		l--;
+	    }
+	    if (l > 0) {
+		if (fspace < l)
+		    goto toolong;
+		fspace -= l;
+		memcpy(bp, dp, l);
+		bp += l;
+	    }
+
+	    /* add a "/" between directory and filename */
+	    if (ep[-1] != '/')
+		*bp++ = '/';
+	}
+
+	/* now append the file name */
+	i = fnlen;
+	if (fspace < i) {
+	  toolong:
+	    PATHNAME_TOO_LONG();
+	    goto next;
+	}
+	fspace -= i;
+	memcpy(bp, fname, i + 1);
+
+#if defined(DOSISH)
+	if (exe_flag && !ext) {
+	  needs_extension:
+	    for (j = 0; j < sizeof(extension) / sizeof(extension[0]); j++) {
+		if (fspace < strlen(extension[j])) {
+		    PATHNAME_TOO_LONG();
+		    continue;
+		}
+		strlcpy(bp + i, extension[j], fspace);
+		if (stat(fbuf, &st) == 0)
+		    return fbuf;
+	    }
+	    goto next;
+	}
+#endif
+
+#ifndef S_ISREG
+# define S_ISREG(m) (((m) & S_IFMT) == S_IFREG)
+#endif
+	if (stat(fbuf, &st) == 0 && S_ISREG(st.st_mode)) {
+	    if (exe_flag == 0) return fbuf;
+	    /* looking for executable */
+	    if (eaccess(fbuf, X_OK) == 0) return fbuf;
+	}
+      next:
+	/* if not, and no other alternatives, life is bleak */
+	if (*ep == '\0') {
+	    return NULL;
+	}
+
+	/* otherwise try the next component in the search path */
+    }
+}
diff --git a/dmydln.c b/dmydln.c
index 09ea06806c..d05cda0b8e 100644
--- a/dmydln.c
+++ b/dmydln.c
@@ -1,2 +1,10 @@
-#define NO_DLN_LOAD 1
-#include "dln.c"
+#include "ruby/ruby.h"
+
+NORETURN(void *dln_load(const char *));
+void*
+dln_load(const char *file)
+{
+    rb_loaderror("this executable file can't load extension libraries");
+
+    UNREACHABLE_RETURN(NULL);
+}
diff --git a/dmyenc.c b/dmyenc.c
new file mode 100644
index 0000000000..7e006e826c
--- /dev/null
+++ b/dmyenc.c
@@ -0,0 +1,10 @@
+#define require(name) ruby_require_internal(name, (unsigned int)sizeof(name)-1)
+int ruby_require_internal(const char *, int);
+
+void
+Init_enc(void)
+{
+    if (require("enc/encdb.so") == 1) {
+	require("enc/trans/transdb.so");
+    }
+}
diff --git a/dmyext.c b/dmyext.c
index 4120d493c3..4d273f7faf 100644
--- a/dmyext.c
+++ b/dmyext.c
@@ -1,4 +1,4 @@
 void
-Init_ext()
+Init_ext(void)
 {
 }
diff --git a/doc/.document b/doc/.document
new file mode 100644
index 0000000000..d739c9f6bc
--- /dev/null
+++ b/doc/.document
@@ -0,0 +1,3 @@
+*.rdoc
+NEWS-*
+syntax
diff --git a/doc/ChangeLog-0.06_to_0.52 b/doc/ChangeLog-0.06_to_0.52
new file mode 100644
index 0000000000..63826081b3
--- /dev/null
+++ b/doc/ChangeLog-0.06_to_0.52
@@ -0,0 +1,1147 @@
+Fri Oct 14 13:22:18 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.52: ……なんてこったい.
+
+	* eval.c(rb_call): returnの処理が間違っていたので, マシンによって
+	  はreturnで関数を終了するだけでなくtoplevelまでつき抜けていた.
+
+	* object.c: Builtinクラスを新設. 組み込み関数をKernelから移した.
+	  nilが組み込み関数を理解するとトラブルの元である.
+
+	* dbm.c: Dictと同様にeachが[key,value]を返すように.
+
+	* version 0.51
+
+Thu Oct 13 12:13:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(SETUP_ARGS): 付加演算子が配列でない時には配列に変換する.
+
+	* parse.y: 括弧なしのメソッド呼び出しでも`*'による付加引数が使える
+	  ようにした. ただし, 通常引数が一つもない場合は乗算演算子と区別が
+	  つかないので, 必ず括弧が必要.
+
+Wed Oct 12 10:09:07 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(rb_call): キャッシュの計算をinline化. キャッシュミスがあ
+	  れば関数呼び出しでメソッドを検索する. methods.cはなくなった.
+
+	* eval.c(rb_eval): ローカル変数用の領域をalloca()するように変更.
+	  サイズの変更が必要になれば改めてmalloc()するように.
+
+	* parse.y: error recoveryの際にlex_stateを更新しておくように.
+
+Tue Oct 11 17:10:46 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* socket.c(for_fd): ファイル記述子(Fixnum)からソケットインスタンス
+	  を得るメソッド. たとえばinetdから起動されたサーバで標準入出力に
+	  ソケット操作を行なうために使う. つまりSocket.for_fd($stdin)で標
+	  準入力に対応するソケットオブジェクトが得られる.
+
+	* io.c(to_i): IOクラスのインスタンスを整数に変換するとそのファイル
+	  記述子を返すように.
+
+	* numeric.c(num2int): to_iメソッドを使ってできる限り整数に変換する.
+	  以前はnum2fixだけが全てのオブジェクトに対してto_iメソッドを適用
+	  していた.
+
+	* sprintf.c(Fsprintf): 整数表示の際, オブジェクトをできる限り整数
+	  に変換するように(to_iメソッドを使う).
+
+Fri Oct  7 14:06:32 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(Fcaller): 必要性がよく分からないのでドキュメントから削除.
+	  将来デバッガを作る時に復活させよう.
+
+	* eval.c(rb_call): Cで記述されたメソッド呼び出しでは環境をスタック
+	  にセーブしないことによって高速化.
+
+Wed Oct  5 15:00:58 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby.h: 一時env.hに移動してたQselfの定義を復活. ただし今回は関数
+	  として実現(env.hでは変数として再定義してある).
+
+	* ruby.h: TRUEでsyntax errorにならないよう#undefを追加.
+
+	* eval.c(rb_eval): thread化に挑戦したが, 失敗(速くならなかった).
+	  が, Scopingなどの無駄なコードの削除とメソッド呼び出しの引数セッ
+	  トのinline化で若干の高速化を実現した. 副作用として, argc, argv形
+	  式の関数呼び出しの仕様が変化した(argvにselfを含まなくなった).
+
+	* eval.c(rb_call): メソッド呼び出しの高速化.
+
+Tue Oct  4 11:40:53 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby-mode.el: 修飾子に対応した.
+
+	* parse.y: 多重代入にrestをつけた. この機能を使えばoptional引数の
+	  解析が簡単にできる(はず).
+
+	* pack.c(unpack): uuencode形式のdecodeの際に文字列の長さが間違って
+	  いた.
+
+Mon Oct  3 15:58:41 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* file.c(type): ファイルタイプを文字列で返すメソッド.
+
+Fri Sep 30 11:36:07 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* object.c: デフォルトの真の値である%TRUEの値を1(Fixnum)からtに変
+	  更した. to_sで文字列に変換した時にも`t'と表示される. 更に踏み込
+	  んで`t'を予約語にしてlispのようにしようかとも思ったが, そこまで
+	  は決心できなかった. 一文字のローカル変数はかなり使いそうな気がす
+	  るので….
+
+	* array.c,dict.c: equalを再定義しているクラスで, hashを正しく定義
+	  した.
+
+Wed Sep 28 23:30:28 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c(Ffail): 今までfailはカーネルクラスのメソッドであったが,
+	  構文に組み込んだ. この変更によって, 1)`fail'は予約語となり, ロー
+	  カル変数に用いることができなくなった. 2)`fail'単体で例外を発生す
+	  るようになった. 3)failはメソッドではなくなったので再定義される可
+	  能性がなくなった.
+
+	* dic.c, dbm.c(indexes): Arrayのindexesと同様の機能を持つメソッド
+	  を追加.
+
+	* array.c(indexes): 引数をインデックスとする要素の配列を返す. 整数
+	  の配列を引数とする時には引数の要素をインデックスとする要素の配列
+	  を返す.
+
+Mon Sep 19 13:42:31 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* array.c(aset): 部分配列に対する代入で配列以外のオブジェクトが指
+	  定された場合に多重代入と同じルールで配列化するようにした.
+
+	* io.c(print): 引数として与えられた各オブジェクトにprint_onメッセー
+	  ジを与えるように. 実行速度は落ちるが柔軟性は増す.
+
+Fri Sep 16 14:59:18 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* glob.c: ワイルドカードの導入. bashに使われているGNUのglobルーチ
+	  ンを流用した.
+
+Mon Sep 12 18:36:58 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(value_expr): 式がnilの時に対応.
+
+	* class.c: ICLASSのclassが必ずClass/Moduleを指すように.
+
+Tue Sep  6 16:23:28 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* re.c:	正規表現内で「\数字」形式が指定できるように.
+
+	* parse.y:「do expr using var ... end」形式はなくなった. 寂しい気
+	  もする. *BACKWARD INCOMPATIBILITY*
+
+Mon Sep  5 10:59:01 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* numeric.c(next): Numericクラスにもnextを提供.
+
+	* string.c(upto): uptoを提供.
+
+	* range.c(each): nextを使ったインタフェースからuptoを使うように変
+	  更した. この方が一つのメソッドで処理をまとめで行なうことができる.
+
+Fri Sep  2 15:25:39 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* dict.c(each): 戻り値を[key, value]のペアに変更. 今までのeachは
+	  each_valueとして残る. *BACKWARD INCOMPATIBILITY*
+
+Thu Sep  1 10:49:04 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* 成功した(特別な戻り値を持たない)システムコールは`0'を返すように.
+
+Wed Aug 31 00:26:51 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* string.c: チェックサムを得るメソッド`sum'を作った.
+
+	* class.c(include_class_new): ICLASSのclassをもとのクラスにした.
+	  gcの際に元クラスをマークする必要があるのが, フィールドを増やす余
+	  地が無いので, classフィールドを流用した. 私の見積りが間違ってい
+	  て, ICLASSのインスタンスにメッセージを送る事があれば, おかしな動
+	  作をするだろう.
+
+	* eval.c(masign): 式(a,b = nil)の値を[nil]からnilに変更した.
+
+Mon Aug 29 11:56:09 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* class.c: rb_define_mfuncを無くして, メタクラスにモジュールをイン
+	  クルードするようにした.
+
+	* error.c(yyerror): 同じ行で複数のsyntax errorをリポートしないよう
+	  にした.
+
+	* file.c: FileTestモジュールにファイルテストメソッドを分離した.
+
+	* parse.y: 演算子を指定する時のlex_stateを正しく設定した.
+
+Sat Aug 27 01:23:34 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y: if/whileなどの複合式をprimaryに移動した. これによって例
+	  えば「if cond then a else b end.message()」のような式が書けるよ
+	  うになった.
+
+Fri Aug 26 10:46:30 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* spec: 整理された文法にしたがって書き直した.
+
+	* parse.y: ここ数日で混乱していた文法を整理した. 括弧を省略したメ
+	  ソッド呼び出しができるようになったこと, modifierが付けられるよう
+	  になったこと, returnにリストが渡せるようになったことが主な変更点
+	  である.
+
+	* process周りが怪しいがとにかくSolaris 2.3で動くように.
+
+	* parse.y: 曖昧性がない場合にはメソッド呼び出しの引数の括弧を省略
+	  できるように. 省略できるメソッド呼び出しの条件は, 1)かならず1個
+	  以上の引数を必要とすること, 2)第1引数が`+', `-', `(', `[', `{',
+	  `/'など, 式の始まりに置かれた時と途中に現れた時とで解釈が違う記
+	  号で始まらないこと, である.
+
+Thu Aug 25 13:54:58 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(cond): 条件式の展開部にbugがあった.
+
+Wed Aug 24 00:01:15 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y: returnはコンマで区切ったリストも受け取るように. つまり,
+	  return a, b, cはreturn [a, b, c]と同じ意味になる.
+
+	* parse.y: yield以外の大域脱出制御式をexprからexpr0に移した. よっ
+	  てメソッドの引数に制御式を使えなくなる(これで困る人はいないはず).
+
+	* parse.y: `+'の定数展開の際に演算子の優先順位を忘れていた.
+
+	* eval.c: untilの戻り値はnilになった.
+
+	* parse.y: modifierとしてのif/unless/while/untilを追加.
+
+	* parse.y: 文法からendの後ろにつけるキーワードを削除. ほとんど使わ
+	  なかった上に, emacsではruby-modeがあれば対応のチェックが機械的に
+	  出来るため.
+
+Tue Aug 23 18:08:33 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: スクリプト実行開始前に例外が発生した時にcore dumpした.
+	  組み込み用にコードを変更した時にenbugしてしまった.
+
+Tue Aug 23 00:07:17 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c: doの戻り値がいつもnilになっていた.
+
+	* parse.y: loop制御変数の多重代入化にbugがあった.
+
+	* parse.y(expand_op): 文字列も畳み込みの対象に.
+
+Mon Aug 22 10:50:01 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(expand_op): `+'に関しては結合則を使って, より多く定数畳
+	  み込みを行なうように.
+
+	* ruby.c(proc_options): argcが0の時にも対応.
+
+	* parse.y: forなどの制御変数に多重代入も使えるように.
+
+Sat Aug 20 00:59:40 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y(call_op): 演算子`~'の取り扱いをルール部へ移動.
+
+Fri Aug 19 11:44:13 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* main.c: rubyをほかのプログラムに組み込めるようにmain()を分割した.
+	  それにともない, プログラムの呼び出し構造を修正した.
+
+	* parse.y: 条件式の定義を変更. ifなどの条件式の中でだけ`&&'や`||'
+	  および`!'の引数が条件式になるように. この変更により条件式以外の
+	  場所での `&&', `||', `!'演算子の動作が直観に一致する.
+
+	* parse.y: 実引数の`*'の後に続く引数はexprに制限した. 今までは全て
+	  の文が有効であったが, ここで定義文があってもしょうがない.
+
+Thu Aug 18 10:21:45 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* re.c: 正規表現ルーチンの初期化部分を削除してしまっていた. これで
+	  はemacsの正規表現になってしまう.
+
+	* version.c: copyright表示を追加.
+
+	* version.c: バージョン表示をstderrに.
+
+	* configure.in: gccがない場合testに失敗していた.
+
+Fri Aug 12 14:12:23 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* array.c(astore): 配列の拡大する時にある程度の大きさをまとめて拡
+	  大するように.
+
+	* io.c(Fprint): 配列に対しては一度文字列に変換することなく, 直接内
+	  容を出力するように.
+
+	* string.c(str_new): memmoveからmemcpyへ置き換えた. これでもかなり
+	  速度が違う.
+
+	* ruby.h: データメンバの取り出しで名前を文字列からIDで指定するよう
+	  にした. かなりの高速化になる.
+
+	* io.c: $ARGFという変数で引数列からなる仮想ファイルをオブジェクト
+	  として扱えるようにした. 今まではトップレベルのgets()などを使って
+	  アクセスしていたが, どうもオブジェクト指向的ではなかった.
+
+Thu Aug 11 11:43:15 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* gc.c: mark_location()の間違った呼び出し方の行が残っていた.
+
+	* method.c: プロトタイプ宣言が足りなかった.
+
+Wed Aug 10 15:54:46 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* variable.c: -vオプションが指定されている時は初期化されていない,
+	  大域変数, インスタンス変数, ローカル変数を参照した時点でwarning
+	  を出すようにした.
+
+Tue Aug  9 11:50:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* bignum.c: 冪乗に関しても多倍長演算を行なうように. 特に浮動小数点
+	  数の範囲を越えた時の処理を的確に行なうように.
+
+	* eval.c: メソッド定義後は構文木から, メソッド定義部分を外す. 無駄
+	  な再定義が起こらないようにするためと2重にfree()されないため.
+
+	* array.c(Fary_aref): 引数が1つでFixnumの時, Range checkを行なわな
+	  いように修正.
+
+	* eval.c: メソッドの引数の数をコンパイル時に計算して若干の高速化.
+
+Mon Aug  8 13:06:24 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* object.c: nilによる比較連鎖をなくした.
+
+	* parse.y: bit演算子の優先順位を比較演算子よりも強くした.  Cとは異
+	  なることになるが, 直観には合致する.
+
+	* gc.c: クラスを解放する時, 個々のメソッド毎にキャッシュをクリアす
+	  るのではなく, クラス単位でクリアするように.
+
+Thu Aug  4 18:45:09 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* methods.c(method_free): 解放されたメソッドに関してキャッシュをク
+	  リアしておく必要があった.
+
+	* gc.c: Dataクラスのデータ部分をfree()し忘れていた.
+
+Wed Aug  3 09:58:14 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: def func .. end形式による関数メソッドの定義はなくなった.
+
+	* methods.c: func形式のメソッドをなくした. あっても, あまり意味が
+	  ないので.
+
+	* eval.c: $0への代入でps(1)の出力が変化するように.
+
+	* io.c(Fsyscall): syscall()を実現.
+
+Mon Aug  1 13:41:11 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: ダブルクォートで囲まれた文字列や正規表現中で"#{変数名}"
+	  または"#変数名"という形式で変数の内容を埋め込むことができるよう
+	  になった.
+
+	* io.c: 関数メソッドsystem2()はなくなった. 今はバッククォートがあ
+	  るからね.
+
+	* parse.y: `cmd`によってコマンドを文字列に展開することができるよう
+	  になった.
+
+	* parse.y: __FILE__, __LINE__を追加. それぞれファイル名(文字列),
+	  行番号(整数)を値とする疑似変数.
+
+Fri Jul 29 13:16:07 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* methods.h: メソッドをオブジェクトとして扱うのをやめる. メソッド
+	  のメモリ管理にはリファレンスカウントを使うことにした. これでオブ
+	  ジェクトの数が減ってほんの少しだけGCが速くなる(かな).
+
+	* purifyによってメモリ関係のバグを検査した(見つかる,見つかる…).
+
+	* gc.c: GCをプログラマが変数をマークする形式から, スタックとレジス
+	  タからマークする方法に変更. 移植性が下がるような気もするが, siod
+	  やscmでも採用されているから多分大丈夫だろう. Linux on i486でも動
+	  作を確認した.
+
+Wed Jul 27 16:13:13 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(Eval): トップレベルでは構造木をfreeしないように. どうせ解
+	  放されるから時間の無駄である.
+
+	* array.c, dict.c: "=="を構造一致に変更.
+
+Fri Jul 22 10:14:09 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* error.c: 組み込みタイプの名前を登録し忘れていた.
+
+Thu Jul 21 14:06:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(freenode),eval.c(Eval): 解析木を解放し忘れていた.
+
+Mon Jul 18 10:19:15 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: 多重代入を処理するルールにバグがあって, 3要素以上の多重
+	  代入に失敗していた.
+
+	* eval.c(rb_eval): 多重代入で, 右辺が配列でない時には`to_a'メソッ
+	  ドで配列に変換して代入するようにした. 今までの仕様だと右辺値が第
+	  1要素にそのまま代入されていたが, structなど配列に変換できるもの
+	  は変換した方が嬉しい気がする.
+
+	* dbm.c,dict.c(delete_if): メソッド追加.
+
+	* process.c(wait,waitpid): システムコールwaitpidまたはwait4がある
+	  時はそちらを使うように. configureもそれらをチェックするように変更.
+
+	* dbm.c, dict.c(clear): メソッド追加.
+
+Fri Jul 15 10:54:45 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* array.c(Fary_fill,Fary_clear): メソッドを追加.
+
+	* string.c(Fstr_split): $;の値が長さ1の文字列である時, これを正規
+	  表現化しないで, 単なる文字として分割する.
+
+	* string.c(Fstr_aset/Fstr_aref): インデックスが文字列の範囲外だっ
+	  た時の動作をArrayを参考に修正した.
+
+	* array.c(astore,Fary_aset): 領域をreallocした後, ゼロでクリアする
+	  ように. 今まで配列にゴミが入っていた.
+
+	* array.c: []/[]=でのインデックス関係を整理. 基本的に負のインデッ
+	  クスに代入しない限り例外は起きないように変更した. 必要に応じて適
+	  当に解釈して, 必要ならば領域を拡張するように.
+
+Thu Jul 14 11:18:07 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* autoexec.c: 削除. autoload関係の機能は今後検討しよう.
+
+	* dict.c: 辞書クラスの正式名称をDictに変更した. 別名としてHashを用
+	  意した. 今までDictionaryなどと長い名前にしていたが誰も使っていな
+	  かったしね. *BACKWARD INCOMPATIBILITY*
+
+	* parse.y: Dictを生成する構文を追加. こちらを{..}にした.
+
+	* parse.y: 配列を生成する構文を[..]に変更した. 過去のRubyスクリプ
+	  トとの互換性が保てないが, Dictを生成する構文を導入するに当たり,
+	  perl5に合わせて(意識して), 変更する時期は今しかないと考えた.
+	  *BACKWARD INCOMPATIBILITY*
+
+	* eval.c(Feval): eval()でメソッドを定義する時, 定義されるクラスを
+	  メソッドの所属するクラスにした. 今まではObjectクラスに定義されて
+	  いた.
+
+	* parse.y: ローカル引数がない時のeval()で落ちていた.
+
+Tue Jul 12 09:41:28 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* pack.c: uuencode形式のサポート.
+
+	* `-0'を`-R'に. 出力レコードセパレータをコマンドラインから指定する
+	  方法はなくなった. どうも, 仕様がゆれるなあ.
+
+Mon Jul 11 09:51:24 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby.c: `-r'オプションは`-0'オプションになった. 当面は`-r'も有効
+	  だが変更される可能性がある. `-R'は当面はそのまま.
+
+	* version.c: version表示に日付を含めた.
+
+	* parse.y: private methodの復活. RubyのprivateメソッドはC++におけ
+	  るprotected methodに該当するもので, `@'で始まる名前を持つ.
+
+	* env.h: struct ENVIRONの定義を分離.
+
+	* parse.y: `\$var', `\@var', `%var'も許すように.
+
+	* variable.c(Fdefined): idも引数として受け付けるように.
+
+	* parse.y: if文/unless文にキーワードthenを追加. thenなしというのは,
+	  意外と間違いが多いので. もちろん省略できる.
+
+Sat Jul  9 02:16:04 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c(rb_eval): class/moduleの評価で新しいスコープを割り当てて
+	  いなかった. スコープ割り当て部分をrb_call()からrb_eval()に移した.
+
+	* eval.c(rb_call): realloc()に渡される事のある, ローカル変数用の領
+	  域をalloca()していた. たまに落ちるわけだ.
+
+	* string.c(Fstr_times): 割り当てた領域を越えた部分を変更していた.
+
+Wed Jul  6 15:52:42 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* socket.c: Socket -> BasicSocket, RawSocket -> Socket に改名.
+
+	* string.c(ucfirst,lcfirst): 最初の1文字だけの大文字/小文字変換.
+
+	* numeric.c(chr): 整数の文字列化メソッド.
+
+	* inits.c, dbm.c: DBMが使えない時はクラスそのものを定義しないよう
+	  にした.  利用できないクラスはnilとすることを今後のポリシーとしよ
+	  う(いままではアクセスした時点でエラーが発生していた). autoexec()
+	  のあり方も検討が必要になりそうだ.
+
+	* bignum.c(bigadd): バグ修正.
+
+Thu Jul  7 11:12:18 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(Fload,Feval): eval_treeをクリアし忘れていた.
+
+	* _inspect: オブジェクトを可読形式の文字列に変換する(主にデバッグ
+	  出力用).
+
+Wed Jul  6 00:57:18 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* numeric.c, bignum.c: 整数に対する`[]'演算子. nビット目がセットさ
+	  れているかどうかを返す.
+
+Tue Jul  5 12:48:39 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* io.c(Feof): 追加. コマンドラインからなる仮想ファイルについても
+	  EOFが検出できるように.
+
+	* ruby.c: -l/-r/-R/-Xオプションの追加.
+
+	* ruby.c: -n/-pオプションのloopの付加などをメインルーチンに移動し
+	  た. これで, オプションの解析途中で(`-c'オプションのせいで)終了な
+	  どといったことはない.
+
+	* io.c(Fgets): 高速化. 凝ったことをしない方が速かった. 虚しい.
+
+Mon Jul  4 15:55:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* Socket:getsockname/getpeername - ようやく仕様が安定した.
+
+	* io.c(Fgets): eachでgetsを記述するのではなく, getsでeachを記述す
+	  るようにした.
+
+Fri Jul  1 10:35:49 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* $ENV[env] = nil/$ENV.delete(env)で環境変数を削除できるようになっ
+	  た. $ENV.deleteは環境変数の以前の値を返す.
+
+	* !~の定義が間違っていた.
+
+	* Dict,DBM:[]= - nilの代入によって要素を削除できるようになった. こ
+	  れにともないnilはDictの要素になれなくなった.
+
+	* ソースの整理. 盲腸のような使われていないコードをなくしたり, 変数
+	  名を付け変えたりした.
+
+Fri Jul  1 00:21:29 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* Array:join() - 要素数0の配列に対して空文字列を返す.
+
+	* RawSocket:open(),socketpair() - 文字列で指定できるドメインとタイ
+	  プをいくつか追加した.
+
+Thu Jun 30 13:51:29 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* -fオプションをなくした. 昔(loadのなかった頃)の名残なので, 今となっ
+	  ては必要ないだろう.
+
+	* -sオプションを追加. perlの-sオプションと同じ動きをする.
+
+	* RawSocketクラスを提供する. Socketに対するシステムコールレベルの
+	  アクセスが可能になった.
+
+Thu Jun 30 00:27:19 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* Socket - bug fixes.
+
+	* linuxではsyscall(SYS_select)が正常に動作しない.
+
+	* Socket:addr,peeraddr - 配列としてsockaddrの情報を返す.
+
+Wed Jun 29 00:14:20 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* Socket:setopt,getopt - setsockopt(2), getsockopt(2)へのアクセス
+	  を実現.
+
+	* sprintf() - rubyにはunsignedは無いので, %uを取り除いた.
+
+	* sprintf() - %b, %x, %oでは2の補数表現, %B, %X, %Oでは符号付き表
+	  現で出力するように. ここ数日でsprintf()の仕様がゆらいでいたが,
+	  これで落ち着きそうだ.
+
+Tue Jun 28 14:42:03 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* Bignum:<<,>> - 2の補数をとる処理を除いた. シフト演算には関係ない
+	  処理だった.
+
+	* Bignum:^ - bug fix. 符合が反対だった.
+
+	* sprintf() - 2進出力子"%b"を追加.
+
+	* sprintf() - %x, %oでFixnumを出力する時, 2の補数表示を行なわない.
+
+	* sprintf() - %x, %oはやはり負の数の時は`-'を出力するように.
+
+Mon Jun 27 14:56:13 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* sprintf() - Bignumについても%d, %oは(2の補数表現に変換して)正の
+	  整数を表示するようにした.
+
+	* Bignumに対する論理演算の定義を修正した. 負の数は2の補数表現であ
+	  るとみなし, かつ仮想的に左側に無限に1が連続しているような演算結
+	  果を得る.
+
+	* Fixnum:<<,>> - 符合付シフトに変更.
+
+	* Bignum:>> - 負の整数のシフトに対応した.
+
+	* __END__, ^D, ^Zでスクリプトを終了できる.
+
+	* -xオプションを追加. #! ..rubyなる行まで読み飛ばす.
+
+	* -cオプションを追加. コンパイルのみを行う.
+
+Sat Jun 25 01:37:21 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* Fixnum:<< - 必要に応じてBignumに拡張して左シフトするように. よっ
+	  て, シフト幅が32を越えるとCやPerlとは違った値を返す.
+
+Fri Jun 24 10:01:28 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ioctl()/fcntl() - システムコールを呼び出す前にバッファの大きさを
+	  調節するようにした.
+
+	* String:toupper/tolower - 文字列を置き換えたコピーを作るのではな
+	  く, 元の文字列の内容を変更するようにした.
+
+	* inplace editを実現した. perlと同じように`-i'オプションで指定する.
+	  もっとも, こちらはMS-DOSのこととか考えてないけど.
+
+	* デフォルトの出力先を追加した. 今までは$stdoutに代入するしか方法
+	  はなかった.
+
+Fri Jun 17 10:55:08 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* 環境変数にアクセスする方法としてgetenv()/setenv()以外に$ENVを用
+	  意した. $ENVは文字列-文字列の辞書であるかのように動作するEnvDict
+	  オブジェクトが代入されている(eachはassocを与える).
+
+	* nilに代入するとcore dumpした. コンパイル時のチェックを強化.
+
+	* Struct: struct_new()の引数をGCプロテクトする必要がある. せめてス
+	  タック領域だけでもスキャンできるようにしなければいけないんだろう
+	  か? でも, 移植性がなあ.
+
+Fri Jun 17 01:01:46 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* Time::asctime() - 日付のフォーマットで日が落ちていた.
+
+	* Stat: StatはEtcなどと同様にStructで実現したので, Statクラスは無
+	  くなった.
+
+Thu Jun 16 10:32:23 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* bignum.c: いくつかのバグを修正した. Fixnumを渡すべきところで普通
+	  のintを渡していた. 失敗.
+
+	* big2str() - 1桁ずつbaseで割る代わりに, 4桁ずつ割算を行なうように
+	  した. これで多倍長割算の回数が1/4になる. さらに整数->数(文字)へ
+	  の変換をテーブルを用いるようにした.
+
+	* rb_ivar_get_1() - すでに何らかのインスタンス変数を持つオブジェク
+	  トでは, 未定義のインスタンス変数の値が不定値になっていた.
+
+	* yylex() - インスタンス変数の認識に失敗していた. attr()は正しく動
+	  作していたので, 混用すると動作しなかった. 全部違っていたから動い
+	  ていたのね.
+
+	* Object:attr() - すでにアクセスメソッドが定義されている時にはデフォ
+	  ルトのアクセスメソッドを定義しないようにした. もっともアクセスメ
+	  ソッドと同名のメソッドの区別はRubyには存在しないけど, それは仕方
+	  がないよね.
+
+	* pack.c: エンディアンをautoconfで判定するようにしたので, v/Vが使
+	  えるようになった. またntoh?()/hton?()も自前で用意した.
+
+	* Stat: st_rdevをアクセスするメソッドを追加. さらにシステムがstat
+	  構造体にst_blksize, st_blockを持っているかをautoconfでチェックす
+	  るようにした.
+
+	* ドキュメントを少し整備した.
+
+	* INT2FIX()のうち, 31bit幅が保証できないものは, int2inum()に置き換
+	  えた.
+
+Wed Jun 15 10:18:27 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* sprintf() - bignumの出力の時, 出力幅を正しく計算するようにした.
+
+	* str2inum() - baseが0の時, baseを自動判定するように(0xで始まる時
+	  16 進, 0で始まる時8進).
+
+Tue Jun 14 16:08:42 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* gc.c: Bignum型を追加するのを忘れていた. 組み込み型を追加した時に
+	  は必ずmark()とsweep()にその型に関する処理を追加する必要がある.
+
+	* bignum: 割算も動いたような気がする. アルゴリズムを理解していない
+	  ので, 自信がない.
+
+Mon Jun 13 14:36:55 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* まだサポートしていないメソッドなどがあるが, 曲がりなりにもBignum
+	  が使えるようになる. これでioctlも使える.
+
+Fri Jun 10 17:26:42 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* Comparable: 基礎となるメソッドを`=='と`>'から`<=>'に変更した. 今
+	  後Comparableのサブクラスは`<=>'だけを再定義する必要がある.
+
+Wed Jun  8 13:12:18 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* Need_Fixnum()をほとんどなくして, NUM2INT()で直接intに変換するこ
+	  とにした. これで31bitに丸めて桁落ちをおこす問題がなくなる.
+
+Tue Jun  7 09:45:31 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby.h: マクロFIXABLE(n)を追加. ついでにFIXNUM周りの定義を変更し
+	  て, 移植性を高めた(つもり).
+
+	* C++の予約語であるnewを削除した. しかし, もうひとつの予約語である
+	  classに関しては, 置き換える単語が思いつかないこともあってそのま
+	  まになっている.
+
+	* 31bitを越えそうなINT2FIX()を関数呼び出しに変えた. 将来bignumが導
+	  入された時には自動的にbignumを返すようにする.
+
+	* readline() - 引数の`-'は標準入力を意味するようになった.
+
+	* ruby.h: 右シフトが論理シフトか算術シフトかは処理系依存のようなの
+	  で, ruby.hでcppを使ってチェックするようにした. これでうまくいく
+	  と思うのだが, 手元に符合付intを論理シフトする処理系がないので確
+	  認できない. NEWS-OSのCCは確か右シフトはいつも論理シフトだったよ
+	  うな気がするんだけど….
+
+Mon Jun  6 10:10:22 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* FIX2INT()の定義を変更した. どうして昔はうまく動かなかったんだろ
+	  うか? もしかして, 右シフトの符号拡張は処理系依存?
+
+	* FIX2INT()とFIX2UINT()を使い分けるようにした. もっともfixnumは31
+	  ビットしかないので, 本質的な解決にはならないのだが(ioctlが組み込
+	  みたかった).
+
+	* printを関数的メソッドから通常メソッドに変更. 引数が与えられない
+	  時にはレシーバをプリントするようにした. これでprintをメッセージ
+	  形式でも実行できるようになった. 例:
+
+		ruby -e 'readlines().sort.print'
+
+	  上のスクリプトは, 引数として与えられた(あるいは標準入力から読み
+	  込まれた)文字列を各行毎にソートして表示する.
+
+	* eval.c: argc,argvパターンで引数を受けるメソッドに引数が一つも与
+	  えられない時, argvがnilになっていた(argv[0]にアクセスすると落ち
+	  てしまう).
+
+	* _exit()を追加. こちらは例外処理など行なわない.
+
+	* dbmクラス: クラス名称をDBM(大文字)に統一した.
+
+Sat Jun  4 00:51:04 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* ループ変数にも属性や配列要素を指定できるようにした.
+
+Fri Jun  3 09:49:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* 多重代入において, 属性代入, 配列要素への代入も行なえるようにした.
+
+	* Need_Fixnum(): nilを0に変換するように.
+
+	* Enumerable:min, max, index, includes - 追加. min, maxは要素が
+	  `<=>'メソッドを持つことを仮定している.
+
+	* Dict/Dbm:length - 要素数を返すメソッド.
+
+	* Dbmクラスにto_aメソッドを追加.
+
+	* Sunにおけるsortの誤動作の件, 昨日の修正でfixされた. しかし, それ
+	  でなぜ動かなかったのかは明らかではないが…. 比較関数がどんな値を
+	  返しても指定した領域外をアクセスするのはバグではないか.
+
+	* ファイルの全内容を読んで, 各行を配列として返すメソッドはpythonを
+	  参考にして`readlines'という名前にした. それにともないgetsに対し
+	  てreadlineという別名を用意した.
+
+Fri Jun  3 00:08:38 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* Array:sort - 判別関数の戻り値はFixnumではなく, Intであるべきだっ
+	  た. 間違い. Sunで動作がおかしかったのはこのせいかも知れない.
+
+Thu Jun  2 11:48:37 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* IO:read_all() - ストリームの最後まで入力して, 各行を要素とする配
+	  列を返すメソッドを追加. また関数メソッド read_all()も追加した.
+  	  これは引数のファイルから読み込んで各行を要素とする配列を返す. 意
+	  味的には
+
+		def read_all()
+	  	  ary = {}
+	  	  while gets()
+		    ary.push($_)
+		  end
+		end
+
+	  とほぼ等価である.
+
+	* String:atoiメソッドを削除. to_aメソッドからaが配列であるとの連想
+	  を呼んで, 混乱を招かないため. 代わりにto_iメソッドを使うこと.
+
+	* 配列への変換メソッドto_aを導入した. 通常のオブジェクトは自分自身
+	  を唯一の要素とする長さ1の配列を返す. 配列は自分自身を, 辞書はキー
+	  と値のペアの配列を返す. Enumeratedをincludeしたクラスは, eachが
+	  返す各要素を含む配列を返す.
+
+	* file.c: 不定個の引数を受けとるメソッド(chmod,chown,utimes)を書き
+	  換えて, 整理した. それに伴い, 最初に全ての引数の型チェックを行な
+	  うようにした. 型チェックに失敗すると処理を行なわずに例外を発生さ
+	  せる.
+
+	* configure.in: 不必要なテストを行なわないように修正した.
+
+Tue May 31 10:41:08 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* String:pack(): 2進数の文字列変換(B,b)で0と1が逆だった.
+
+	* Math.c: 実数系のメソッドに引数として整数が渡された時に自動的に変
+	  換するようにした.
+
+	* toupper(), tolower(): 文字列の判定ミスで変換されていなかった.
+
+	* getopt_long()の仕様によって, スクリプトへの引数がインタプリタの
+	  引数だと解釈されていた. 引数パターン文字列の先頭に`+'を追加.
+
+	* config.hを削除した. DEFINEはMakefileで与えられる.
+
+	* sprintf(): "%d"に文字列が与えられた時にはアドレスではなく内容を
+	  整数に変換するようにした. ついでに浮動小数点数も変換するように変
+	  更した.
+
+	* regexp.c: rubyの拡張正規表現(\d, \D, \s, \S)の処理で割り当てた領
+	  域を越えてバッファに書き込んでいた. 処理前にバッファをきちんと拡
+	  張するようにした. これで昨日問題にしていたメモリの問題は解決でき
+	  たと思う.
+
+	* yylex(): ダブルクォート文字列中でダブルクォートを表現するため
+	  のバックスラッシュ表現ができなかった.
+
+Mon May 30 10:07:42 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* 演算子`!'の右辺も条件式であるとした. これによって, この演算子を
+	  再定義する人は混乱するかも知れないが, 大多数のこの演算子を使う人
+	  は混乱を避けることができると思う.
+
+	* autoconfを使って, 自動的にMakefile, config.hを生成するようにした.
+	  これで, 多くのマシンでは`configure'を実行した後, `make'一発でコ
+	  ンパイルできると思う.
+
+	* clone: サブクラスに対して用いられた場合, 元のオブジェクトと同じ
+	  クラスのインスタンスを返すように(以前はビルトインクラスの場合を
+	  考えてなかった).
+
+	* ビルトインクラスのサブクラスも作れるように, リテラルのあるクラス
+	  にもnewメソッドを追加した.
+
+	* malloc()で落ちる. purifyが必要かも知れない.
+
+	* re.c: rb_global_variable()の呼びだし形式の間違い. 変数へのポイン
+	  タを渡さなければいけない.
+
+	* parse.y: ローカル変数の扱いに引数の評価順に依存する移植性のない
+	  部分があった.
+
+	* attr(): 属性設定のバグを直した. いつ内部仕様が変わったんだろう…?
+
+Sat May 28 23:08:18 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* 正規表現キャッシュの文字列一致判定をポインタ一致から内容一致に変
+	  更した. そういえば文字列リテラルは一回毎に新しくオブジェクトが生
+	  成されるのだった.
+
+Fri May 27 11:42:00 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* trから文字削除(delete), 文字圧縮(squeeze)を分離した. それにとも
+	  ないtrのオプション引数はなくなった.
+
+Thu May 26 10:32:55 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* スクリプト読み込みルーチンを書き直して, 通常ファイル以外のファイ
+	  ル名や空文字列がスクリプトとして与えられた場合に対応した. また,
+	  標準入力からスクリプトを読み込む時に, 一時ファイルが/tmpに残らな
+	  いようにした.
+
+	* Fixnum:id2name - IDから文字列に戻す関数. String:internの逆.
+
+	* Array: 配列の範囲外の要素をアクセスした時に例外を発生させずに,
+	  nilを返すようにした. 配列は自動的に拡張される.
+
+	* string:stripを追加.
+
+	* -nオプションが-eオプションを複数指定した時も動作するように.
+
+	* parse.yで<sys/types.h>もインクルードするようにした.
+
+	* fname周りの細かいbugを修正.
+
+Wed May 26 11:45:10 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* 定数をキャッシュするようにした. 繰り返しが多い場合には有効のはず
+	  だが, 一度しかアクセスしない場合は遅くなるなあ.
+
+Wed May 25 00:42:24 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* 多重代入文(foo, bar = 1, 2)の採用.
+
+	* 条件式部に文字列あるいは正規表現リテラルをおくと`=~'演算子によっ
+	  て`$_'と比較される. 更に`...'の両辺では整数リテラルが`$.'と比較
+	  される.
+
+Mon May 23 23:27:03 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* &式 形式はなくなった. 代わりにkernel:apply(id, args..)を導入.
+
+	* def op () ..形式の導入. opは再定義可能な演算子.
+
+	* constantの代入時チェック. 既に初期化されている定数に代入した場合
+	  は例外が発生する.
+
+	* 多重代入文.
+
+Thu May 19 22:57:07 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* 複合文でもvoid valueのチェックを行うようにした.
+
+	* untilの動作の修正(do..until型だった).
+
+Wed May 18 01:06:25 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* 移植に関する若干の問題を修正.
+
+	* 別名の構文を「def a b」にした.
+
+	* until/unless: 演算子から制御文へ. 例外を捕捉する機能はそのまま.
+
+	* 選択可能な機能をconfig.hからdefines.hに移動.
+
+Fri May 13 23:20:21 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* -yオプションを新設. -dオプションからコンパイラのデバッグ部分を分
+	  離した.
+
+Tue Apr 25 20:17:33 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* マルチバイト文字列を識別子に使えるように. 個人的には使いたくは無
+	  いけどなあ.
+
+	* `-v'フラグの状態を$verboseでアクセスできるように.
+
+	* CVSの導入に伴い, バージョン管理の方法を変更.
+
+	* 真面目にChangeLogをつける事にした.
+
+Tue Mar  8 10:09:25 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* %変数名 によるクラス定数を導入.
+
+	* undef メソッド によるメソッド定義の取り消しを導入.
+
+	* rb_get_method_bodyではthe_envを変更せず, rb_call()で明示的に変更
+	  するように. これでresponds_toなどで環境が破壊されない.
+
+Mon Mar  7 17:46:15 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* 「&文字列」形式. 「式.文字列」型のメッセージセンドはなくなった.
+
+	* 自己代入形式(+=. -=, ...)
+
+	* obj.attr = expr形式の採用.
+
+Thu Feb 24 16:23:28 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* toint, tofloat, print_stringをそれぞれto_i, to_f, to_sに変更.
+
+	* String:clone - Copy on Writeの実現.
+
+Tue Feb 22 11:11:44 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* re.c: マッチした文字列の保存に失敗していた.
+
+	* trap: 可能ならば処理に時間のかかるシステムコール(read, wait,
+	  sigpause, select)をフックして割り込み処理の即答性を高める(DOSな
+	  どでは無理だなあ).
+
+	* trap: 割り込みをその場で処理するか(迅速だが危険), 安全なタイミン
+	  グで処理するかを選択できるように.
+
+Tue Feb 17 11:11:12 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* trap: 割り込みハンドラ.
+
+Wed Feb 16 12:29:12 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* String:crypt: 暗号化ルーチン
+
+	* "::"演算子の追加. a::b は {a, b}と同義. a::b::c は {a, {b, c}}と
+	  同義(右結合). 同義とはいうものの, "::"演算子を使った方が少しだけ
+	  メモリ効率が良い.
+
+	* Dir.rmdir(), File.unlink(), File.utime() -- 各システムコールへの
+	  インタフェース.
+
+	* kill -- kill(2) I/F
+
+	* select(): readのチェックではstdioにバッファリングされているかど
+	  うかをチェックするように.
+
+Tue Feb 15 15:08:31 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* file.c: statをキャッシュするように.
+
+	* File:utime()を追加.
+
+	* unliteralize(): フラグを破壊していた.
+
+	* Bug(): coreを吐くように.
+
+	* String:tr -- tr(1)互換. 引数パターンがちょっと違うけど….
+
+Mon Feb 14 18:24:13 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* unless, untilが例外も偽と見なすように.
+
+	* select() -- select(2) I/F
+
+	* Array:pack, String:unpack: perlのpack/unpackの同等品
+
+Tue Feb  8 17:11:10 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* setenv()のないシステムのためにputenv()を使ったコードも用意した.
+
+Mon Feb  7 09:52:44 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* 引数の一番最後に`*'を置けるようにした. これでrest引数のリストを
+	  操作する必要が少なくなる.
+
+Fri Feb  4 18:23:26 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* ruby-mode.elを書き直す. ずいぶんましになったと思う.
+
+	* 文字列リテラルのCopy on Writeを実現. これで文字列がリテラルであ
+	  るからといっていちいちcloneしなくても済む.
+
+Tue Feb  1 09:21:09 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* sub(), gsub()で, マッチした文字列を$&, $1..$9でアクセスできるよ
+	  うにした. 同時にマッチした部分文字列をコピーしておくように(元の
+	  文字列が変更されても状態を保存するため).
+
+Mon Jan 31 15:16:58 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* プライベートメソッドの仕様を変更. 今までは同じクラスのメソッドか
+	  らしかアクセスできなかったが, サブクラスのメソッドからもアクセス
+	  できるようにした(C++におけるprotected メンバ関数).
+
+	* メソッドサーチのアルゴリズムを改善し, 10%程度の高速化を行なった.
+
+	* 高速化. Cで記述されたメソッドを呼び出す時にはsetjmpを呼ばないよ
+	  うにした. これでCメソッドを多用する場合には3倍程度高速になった.
+
+Fri Jan 28 15:44:04 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* sh-modeを元にruby-mode.elを作る. 演算子で終る, 2行に渡る文には対
+	  応していないけど….
+
+Thu Jan 27 11:35:19 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* freenode(): NODE_NILの解放忘れ.
+
+	* 字句解析部のバグ修正(コメントの後の状態を戻し忘れ).
+
+	* protect .. endのバグ修正. GC_LINKのネストが不正だった.
+
+	* joinのバグ修正(使っているオブジェクトをfreeしていた).
+
+	* splitのバグ修正(アルゴリズムがおかしかった).
+
+	* fork()を追加.
+
+Wed Jan 26 17:09:56 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* ファイルテストメソッドの追加.
+
+	* rb_autoexec(): クラスを初めてアクセスした時の挙動を制御できるよ
+	  うにした. これでautoloadも実現できる. これにともないメソッド
+	  unknownはなくなった.
+
+Tue Jan 25 15:51:36 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* Dbmクラス, Mathモジュールを作成.
+
+	* -Iオプションでサーチパスに追加できるように.
+
+	* サーチパスを変数$load_pathに設定できるように.
+
+	* load(): ダイナミックロードを使えるようにした.
+
+Tue Jan 18 14:14:01 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* Comparable:"<=>"
+
+	* Float,Fixnum:"**"
+
+	* Array:sort
+
+Fri Jan 14 16:53:37 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* version 0.07
+
+	* メソッドに関するドキュメントを充実させた.
+
+	* String:index(): 引数positionを増やした.
+
+Thu Jan 13 15:13:52 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* 未初期化の変数アクセスをなくした.
+
+	* 無駄なhash tableのアロケーションを削除.
+
+	* Purify'd(on Sun)
+
+	* ~RE と ~STRのコンパイル時展開の抑制.
+
+	* Sunへ移植. signal()の戻り値. RDataのbug修正.
+
+	* parse.y: nlsルールを削除.
+
+	* yylex(): 改行と符合の解析部分を変更.
+
+	* missing/strftime.c: 移植用.
+
+	* Time:strftime: その他のメソッドもstrftimeを利用するように.
+
+	* メソッド再定義時にメソッドキャッシュをクリアする.
+
+Fri Jan  7 15:23:20 1994  Yukihiro Matsumoto  (matz at nws119)
+
+	* Float:coerce(): FixnumとFloat以外の引数を与えられるた時には例外
+	  を発生するように.
+
+	* Stat: stat構造体の全てのメンバに対するアクセスメソッドを用意.
+
+	* 未定義のクラス/モジュールへの参照がunknownメソッドを呼び出すよう
+	  にした.
+
+	* baseline - version 0.06.
diff --git a/doc/ChangeLog-0.50_to_0.60 b/doc/ChangeLog-0.50_to_0.60
new file mode 100644
index 0000000000..5f5b03ff40
--- /dev/null
+++ b/doc/ChangeLog-0.50_to_0.60
@@ -0,0 +1,462 @@
+Thu Dec  8 00:32:21 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* io.c($<.file,$<.filename): きちんと初期化．
+
+	* parse.y(rb_class2name): includeしているクラス名を正常に表示でき
+	  なかった．
+
+Wed Dec  7 15:40:36 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* dln.c: config.hをincludeしていなかった．
+
+	* missing/strdup.c: 忘れていた．
+
+Fri Dec  2 15:21:44 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* 関数の命名ルールを変えた(クラスメソッド: Sxxx_xxx，ユーティリティ
+	  メソッドxxx_xxxなど)．
+
+	* re.c(casefold): 個々の正規表現オブジェクトが大文字小文字を無視す
+	  るかどうかを設定できるようにした．
+
+Thu Dec  1 01:44:51 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* missing.c: システムで提供されないファイルの扱いを変更した．
+
+	* io.c($<): 変数`$<'が指しているものを現在読み込み中のファイル名か
+	  ら仮想ファイル($ARGF)に変更した．また，現在読み込み中のファイル
+	  名とファイルオブジェクトにアクセスするメソッドも用意した．
+
+	* ruby.h(data_new): Dataオブジェクトの割り当て方法を改善した．
+
+Wed Nov 30 15:36:13 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* Makefile: ルールを整理した．
+
+Tue Nov 29 00:09:26 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* array.c(`|'): 引数が配列でなく，かつそのオブジェクトが要素として
+	  含まれていない時，そのオブジェクトを追加する．あればなにもしない．
+
+	* process.c(gid,egid): GIDをアクセスするメソッド．
+
+	* io.c(print): 文字列と配列に対する処理を組み込んで若干の高速化を
+	  図った．出力が多い場合に効いてくるようだ．
+
+	* parse.y(parse_string): ""で囲まれた文字列内では対応するエスケー
+	  プがない`\'を残さないように．つまり，今後は"\k" -> "k"である(以
+	  前は"\k"だった)．
+
+Mon Nov 28 18:02:31 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* re.c(re_regsub): 置換文字列の置換えが不完全だった．
+
+	* 一時オブジェクトの解放をすべてGCに任せた．今のままではまずいタイ
+	  ミングでオブジェクトを解放する可能性があった．昔のGCの時にはこれ
+	  でも良かったのだが．
+
+Tue Nov 22 00:15:24 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c($"): 既にロードしたファイル名の配列．
+
+Mon Nov 21 01:09:50 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* array.c(&,|): 集合としての積演算と和演算．
+
+	* io.c($<): 文字列を代入した時には自動的にopenするように．IOのサブ
+	  クラスかどうかのチェックも行う．
+
+Sun Nov 19 23:02:27 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c($:): $LOAD_PATHの別名．
+
+	* io.c($>): デフォルト出力先．IO.defaultは無くなった．
+
+	* io.c($<): $FILENAMEの別名
+
+	* glob.c(each): ワイルドカードにマッチするものがない場合，パターン
+	  そのものを与えるように．
+
+	* file.c(chmod,chown,unlink,utime): `\'でワイルドカードをエスケー
+	  プできるように．
+
+Fri Nov 18 00:20:42 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.55 released
+
+	* file.c(chmod,chown,unlink,utime): 引数として与えられた文字列にワ
+	  イルドカードが含まれている場合にはGlobオブジェクトに変換する.
+
+Wed Nov 16 17:33:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* file.c(chmod,chown,unlink,utime): 引数としてワイルドカードも受け
+	  付けるように. その場合はマッチするファイル全体に適用される.
+
+Fri Nov 11 00:07:28 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* string.c(strip): 文字列そのものを変更するように.
+
+	* eval.c(rb_cal): 環境をスタックに積むタイミングが間違っていたので，
+	  不適切なエラーメッセージが出る時があった．
+
+	* eval.c(rb_undefined): メソッドが無い時とメソッドにアクセスできな
+	  い時とでメッセージを替えた．
+
+	* string.c: toupper/tolowerはupcase/downcaseを使って実現．
+
+	* string.c: lcfirst/ucfirstは削除．
+
+Thu Nov 10 16:15:16 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* string.c: sub/gsubは置換後の文字列を返すようにした.
+
+	* string.c: toupper/tolower/lcfirstなどを文字列を変更するものとし
+	  ないものの2 種類を提供するようにした.
+
+Tue Nov  1 17:52:09 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: 変数展開にバグがあって元の文字列を破壊していた.
+
+Thu Oct 27 09:56:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: rest引数の処理が間違っていた.
+
+	* env.c(rb_yield): ローカル変数の設定が間違っていた.
+
+Wed Oct 26 19:01:43 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: 引数セットの高速化.
+
+	* process.c: sleepのバグ.
+
+	* parse.y, gnuglob.c: Sunのccでもコンパイルできるように.
+
+Tue Oct 25 00:36:16 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: 若干の高速化. スタックの処理によってrb_call()が少し遅く
+	  なった分の穴埋めくらいか.
+
+	* eval.c: blockをenvから外すなどしてスタックを軽くした.
+
+Mon Oct 24 11:47:54 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* re.c: 正規表現アクセス用の変数($`, $', $+)を追加.
+
+	* re.c($KANJI): 処理する漢字コードをrubyプログラム内から設定できる
+	  ようにした.
+
+Sat Oct 22 00:42:41 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c: イテレータ内のredoとretryの処理にバグがあった．
+
+Fri Oct 21 00:53:21 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* string.c(swapcase): 文字列の大文字小文字を置き換える.
+
+	* string.c(ljust,rjust,center): 文字列のレイアウト用メソッド.
+	  sprintf()でも同じようなことはできるが.
+
+	* socket.c(recv,recvfrom): flags引数を省略可能にした. 省略時の値は
+	  0である.
+
+	* socket.c(recvfrom): recvと同様だが, データと相手アドレスのペアを
+	  返す.
+
+	* socket.c(accept): 戻り値をつながったソケットから, ソケットとアド
+	  レスのペアに変更した.
+
+	* eval.c(eval): the_classをセットする時に, last_moduleがincludeさ
+	  れているモジュール(T_ICLASS)であれば, 定義元のモジュールの方をセッ
+	  トするように. この修正がないとICLASSに対してメソッドが呼ばれる可
+	  能性があり, 不審な動作をする(はず).
+
+	* class.c: オブジェクトのcloneの際に特異クラス(特異メソッド用のク
+	  ラス)をきちんとコピーするようにした. 今までは個々のメソッドのコ
+	  ピーを忘れていた.
+
+	* numeric.c: Numericのnewとcloneをundefした.
+
+Thu Oct 20 11:30:00 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.53 released.
+
+	* parse.y: TopLevelのdef文は関数メソッドを定義するようにした.
+
+	* parse.y: def文のprivate指定はなくなった. export/noexportメソッド
+	  を使ってもらうことにした.
+
+	* parse.y: case文にthenを含むことができるようになった.
+
+Wed Oct 19 13:09:58 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* object.c(export,unexport): 関数メソッドの設定用メソッド.
+
+	* eval.c, class.c: 関数的メソッドを復活させ(またか…), `@'による
+	  privateメソッドをなくした. やはり, 関数的なものは関数的に呼びた
+	  い気がしてきた. こう仕様が変動しててはいけないような気もするなあ.
+	   *BACKWARD INCOMPATIBILITY*
+
+	* eval.c: メソッドの構成を変更し, 別名管理と関数メソッドの管理を分
+	  離した.
+
+	* eval.c: the_env->last_funcをCメソッドの時にも更新する. これがな
+	  いとStructのアクセスが動作しない.
+
+Fri Oct 14 13:22:18 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.52 released: ……なんてこったい.
+
+	* eval.c(rb_call): returnの処理が間違っていたので, マシンによって
+	  はreturnで関数を終了するだけでなくtoplevelまでつき抜けていた.
+
+	* dbm.c: Dictと同様にeachが[key,value]を返すように.
+
+	* version 0.51 released
+
+Thu Oct 13 12:13:48 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(SETUP_ARGS): 付加引数が配列でない時には配列に変換する.
+
+	* parse.y: 括弧なしのメソッド呼び出しでも`*'による付加引数が使える
+	  ようにした. ただし, 通常引数が一つもない場合は乗算演算子と区別が
+	  つかないので, 必ず括弧が必要.
+
+Wed Oct 12 10:09:07 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(rb_call): キャッシュの計算をinline化. キャッシュミスがあ
+	  れば関数呼び出しでメソッドを検索する. methods.cはなくなった.
+
+	* eval.c(rb_eval): ローカル変数用の領域をalloca()するように変更.
+	  サイズの変更が必要になれば改めてmalloc()するように.
+
+	* parse.y: error recoveryの際にlex_stateを更新しておくように.
+
+Tue Oct 11 17:10:46 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* socket.c(for_fd): ファイル記述子(Fixnum)からソケットインスタンス
+	  を得るメソッド. たとえばinetdから起動されたサーバで標準入出力に
+	  ソケット操作を行なうために使う. つまりSocket.for_fd($stdin)で標
+	  準入力に対応するソケットオブジェクトが得られる.
+
+	* io.c(to_i): IOクラスのインスタンスを整数に変換するとそのファイル
+	  記述子を返すように.
+
+	* numeric.c(num2int): to_iメソッドを使ってできる限り整数に変換する.
+	  以前はnum2fixだけが全てのオブジェクトに対してto_iメソッドを適用
+	  していた.
+
+	* sprintf.c(Fsprintf): 整数表示の際, オブジェクトをできる限り整数
+	  に変換するように(to_iメソッドを使う).
+
+Fri Oct  7 14:06:32 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(Fcaller): 必要性がよく分からないのでドキュメントから削除.
+	  将来デバッガを作る時に復活させよう.
+
+	* eval.c(rb_call): Cで記述されたメソッド呼び出しでは環境をスタック
+	  にセーブしないことによって高速化.
+
+Wed Oct  5 15:00:58 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby.h: 一時env.hに移動してたQselfの定義を復活. ただし今回は関数
+	  として実現(env.hでは変数として再定義してある).
+
+	* ruby.h: TRUEでsyntax errorにならないよう#undefを追加.
+
+	* eval.c(rb_eval): thread化に挑戦したが, 失敗(速くならなかった).
+	  が, Scopingなどの無駄なコードの削除とメソッド呼び出しの引数セッ
+	  トのinline化で若干の高速化を実現した. 副作用として, argc, argv形
+	  式の関数呼び出しの仕様が変化した(argvにselfを含まなくなった).
+
+	* eval.c(rb_call): メソッド呼び出しの高速化.
+
+Tue Oct  4 11:40:53 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby-mode.el: 修飾子に対応した.
+
+	* parse.y: 多重代入にrestをつけた. この機能を使えばoptional引数の
+	  解析が簡単にできる(はず).
+
+	* pack.c(unpack): uuencode形式のdecodeの際に文字列の長さが間違って
+	  いた.
+
+Mon Oct  3 15:58:41 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* file.c(type): ファイルタイプを文字列で返すメソッド.
+
+Fri Sep 30 11:36:07 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* object.c: デフォルトの真の値である%TRUEの値を1(Fixnum)からtに変
+	  更した. to_sで文字列に変換した時にも`t'と表示される. 更に踏み込
+	  んで`t'を予約語にしてlispのようにしようかとも思ったが, そこまで
+	  は決心できなかった. 一文字のローカル変数はかなり使いそうな気がす
+	  るので….
+
+	* array.c,dict.c: equalを再定義しているクラスで, hashを正しく定義
+	  した.
+
+Wed Sep 28 23:30:28 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c(Ffail): 今までfailはカーネルクラスのメソッドであったが,
+	  構文に組み込んだ. この変更によって, 1)`fail'は予約語となり, ロー
+	  カル変数に用いることができなくなった. 2)`fail'単体で例外を発生す
+	  るようになった. 3)failはメソッドではなくなったので再定義される可
+	  能性がなくなった.
+
+	* dic.c, dbm.c(indexes): Arrayのindexesと同様の機能を持つメソッド
+	  を追加.
+
+	* array.c(indexes): 引数をインデックスとする要素の配列を返す. 整数
+	  の配列を引数とする時には引数の要素をインデックスとする要素の配列
+	  を返す.
+
+Mon Sep 19 13:42:31 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* array.c(aset): 部分配列に対する代入で配列以外のオブジェクトが指
+	  定された場合に多重代入と同じルールで配列化するようにした.
+
+	* io.c(print): 引数として与えられた各オブジェクトにprint_onメッセー
+	  ジを与えるように. 実行速度は落ちるが柔軟性は増す.
+
+Fri Sep 16 14:59:18 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* glob.c: ワイルドカードの導入. bashに使われているGNUのglobルーチ
+	  ンを流用した.
+
+Mon Sep 12 18:36:58 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(value_expr): 式がnilの時に対応.
+
+	* class.c: ICLASSのclassが必ずClass/Moduleを指すように.
+
+Tue Sep  6 16:23:28 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* re.c:	正規表現内で「\数字」形式が指定できるように.
+
+	* parse.y:「do expr using var ... end」形式はなくなった. 寂しい気
+	  もする. *BACKWARD INCOMPATIBILITY*
+
+Mon Sep  5 10:59:01 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* numeric.c(next): Numericクラスにもnextを提供.
+
+	* string.c(upto): uptoを提供.
+
+	* range.c(each): nextを使ったインタフェースからuptoを使うように変
+	  更した. この方が一つのメソッドで処理をまとめで行なうことができる.
+
+Fri Sep  2 15:25:39 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* dict.c(each): 戻り値を[key, value]のペアに変更. 今までのeachは
+	  each_valueとして残る. *BACKWARD INCOMPATIBILITY*
+
+Thu Sep  1 10:49:04 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* 成功した(特別な戻り値を持たない)システムコールは`0'を返すように.
+
+Wed Aug 31 00:26:51 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* string.c: チェックサムを得るメソッド`sum'を作った.
+
+	* class.c(include_class_new): ICLASSのclassをもとのクラスにした.
+	  gcの際に元クラスをマークする必要があるのが, フィールドを増やす余
+	  地が無いので, classフィールドを流用した. 私の見積りが間違ってい
+	  て, ICLASSのインスタンスにメッセージを送る事があれば, おかしな動
+	  作をするだろう.
+
+	* eval.c(masign): 式(a,b = nil)の値を[nil]からnilに変更した.
+
+Mon Aug 29 11:56:09 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* class.c: rb_define_mfuncを無くして, メタクラスにモジュールをイン
+	  クルードするようにした.
+
+	* error.c(yyerror): 同じ行で複数のsyntax errorをリポートしないよう
+	  にした.
+
+	* file.c: FileTestモジュールにファイルテストメソッドを分離した.
+
+	* parse.y: 演算子を指定する時のlex_stateを正しく設定した.
+
+Sat Aug 27 01:23:34 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y: if/whileなどの複合式をprimaryに移動した. これによって例
+	  えば「if cond then a else b end.message()」のような式が書けるよ
+	  うになった.
+
+Fri Aug 26 10:46:30 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* spec: 整理された文法にしたがって書き直した.
+
+	* parse.y: ここ数日で混乱していた文法を整理した. 括弧を省略したメ
+	  ソッド呼び出しができるようになったこと, modifierが付けられるよう
+	  になったこと, returnにリストが渡せるようになったことが主な変更点
+	  である.
+
+	* process周りが怪しいがとにかくSolaris 2.3で動くように.
+
+	* parse.y: 曖昧性がない場合にはメソッド呼び出しの引数の括弧を省略
+	  できるように. 省略できるメソッド呼び出しの条件は, 1)かならず1個
+	  以上の引数を必要とすること, 2)第1引数が`+', `-', `(', `[', `{',
+	  `/'など, 式の始まりに置かれた時と途中に現れた時とで解釈が違う記
+	  号で始まらないこと, である.
+
+Thu Aug 25 13:54:58 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(cond): 条件式の展開部にbugがあった.
+
+Wed Aug 24 00:01:15 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y: returnはコンマで区切ったリストも受け取るように. つまり,
+	  return a, b, cはreturn [a, b, c]と同じ意味になる.
+
+	* parse.y: yield以外の大域脱出制御式をexprからexpr0に移した. よっ
+	  てメソッドの引数に制御式を使えなくなる(これで困る人はいないはず).
+
+	* parse.y: `+'の定数展開の際に演算子の優先順位を忘れていた.
+
+	* eval.c: untilの戻り値はnilになった.
+
+	* parse.y: modifierとしてのif/unless/while/untilを追加.
+
+	* parse.y: 文法からendの後ろにつけるキーワードを削除. ほとんど使わ
+	  なかった上に, emacsではruby-modeがあれば対応のチェックが機械的に
+	  出来るため.
+
+Tue Aug 23 18:08:33 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: スクリプト実行開始前に例外が発生した時にcore dumpした.
+	  組み込み用にコードを変更した時にenbugしてしまった.
+
+Tue Aug 23 00:07:17 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c: doの戻り値がいつもnilになっていた.
+
+	* parse.y: loop制御変数の多重代入化にbugがあった.
+
+	* parse.y(expand_op): 文字列も畳み込みの対象に.
+
+Mon Aug 22 10:50:01 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(expand_op): `+'に関しては結合則を使って, より多く定数畳
+	  み込みを行なうように.
+
+	* ruby.c(proc_options): argcが0の時にも対応.
+
+	* parse.y: forなどの制御変数に多重代入も使えるように.
+
+Sat Aug 20 00:59:40 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y(call_op): 演算子`~'の取り扱いをルール部へ移動.
+
+Fri Aug 19 11:44:13 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* main.c: rubyをほかのプログラムに組み込めるようにmain()を分割した.
+	  それにともない, プログラムの呼び出し構造を修正した.
+
+	* parse.y: 条件式の定義を変更. ifなどの条件式の中でだけ`&&'や`||'
+	  および`!'の引数が条件式になるように. この変更により条件式以外の
+	  場所での `&&', `||', `!'演算子の動作が直観に一致する.
+
+	* parse.y: 実引数の`*'の後に続く引数はexprに制限した. 今までは全て
+	  の文が有効
diff --git a/doc/ChangeLog-0.60_to_1.1 b/doc/ChangeLog-0.60_to_1.1
new file mode 100644
index 0000000000..33b0326892
--- /dev/null
+++ b/doc/ChangeLog-0.60_to_1.1
@@ -0,0 +1,3955 @@
+Tue Aug 12 16:02:18 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* variable.c: option variables: $-0, $-p(readonly), $-v,
+	  $-I(load_path), $-a(readonly), $-K, $-d, $-F, $-i, $-l.
+
+	* parse.y (yylex): ignore rd (ruby document) in the code.
+
+Mon Aug 11 12:37:58 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* re.c (Init_Regexp): $-K as alias to the $KCODE.
+
+	* io.c (Init_IO): new virtual variable $-i for the value of -i
+ 	  option.
+
+	* enum.c (Init_Enumerable): include? as alias of member?
+
+Fri Aug  8 11:16:50 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* io.c (io_foreach): now the record separator can be specified.
+
+	* io.c (io_s_readlines): new method to read in whole file (or
+ 	  command output) from path.
+
+	* ext/socket/socket.c (Init_socket): recvfrom did not work.
+
+	* ext/socket/socket.c (sock_send): forgot to check nil for false
+	  value.
+
+Thu Aug  7 11:40:01 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* variable.c (mod_constants): lists constants defiend in the
+	  modules/classes.
+
+	* variable.c (rb_const_set): no longer warns about constant
+	  overriding by subclasses.
+
+	* eval.c (mod_eval): does eval() on module's context. local
+	  variables are shared with outer scope.
+
+	* object.c (Init_Object): remove private_attr/public_attr.
+
+Wed Aug  6 14:21:36 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* object.c (mod_attr): forgot to check nil for false value.
+
+Mon Aug  4 11:50:28 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* object.c (class_s_new): need not to specify names to create
+	  classes (or modules) dynamically.
+
+	* variable.c (rb_class_path): scan class constants for anonymous
+	  classes/modules to make up pathes.
+
+Wed Jul 30 08:45:12 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* eval.c (rb_eval): stop to cache const value in nodes.
+
+Sat Jul 26 03:17:22 1997  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* numeric.c (flo_to_s): wrong .0 at end.
+
+Sat Jul 26 00:36:36 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* eval.c (error_print): always print exception type in the
+ 	  toplevel exception handler.
+
+	* string.c (str_hash): wrong hash value.
+
+Thu Jul 24 11:05:51 1997  Yukihiro Matsumoto  <matz@netlab.co.jp>
+
+	* string.c (uscore_get): proper error message for unset $_.
+
+Wed Jul 23 09:56:55 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (obj_methods): returns list of method names of the
+	  specified object.
+
+	* class.c (mod_instance_methods): returns list of method names of
+	  the class instnace.
+
+Fri Jul 11 22:38:55 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (class_superclass): returns class's superclass
+ 	  itself. (1.1)
+
+	* object.c (obj_type): returns object's class itself. (1.1)
+
+	* class.c (mod_included_modules): list included modules.
+
+	* object.c (class_superclass): raises error for Object.
+
+Thu Jul  3 09:54:02 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (SETUP_ARGS): save source position, remove nd_line().
+
+	* eval.c (rb_call): replace modulo by bit-masking.
+
+	* eval.c (POP_SCOPE): force recycle scope object to reduce gc rate.
+
+	* gc.c (obj_free): aboid calling run_final() when no finalizer is set.
+
+	* eval.c (PUSH_VARS): do not allocate the dynamic scope's end-mark
+	  object.
+
+Wed Jul  2 14:25:07 1997  KIMURA Koichi  <kkimura@pure.cpdc.canon.co.jp>
+
+	* Native mswin32 support.
+
+Tue Jul  1 09:59:00 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970701
+
+	* parse.y (mrhs): allow rest-star(*) in right hand side.
+
+Tue Jun 24 19:04:31 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970624
+
+Sat Jun 20 22:22:51 1997  Michio "Karl" Jinbo  <karl@marcer.nagaokaut.ac.jp>
+
+	* eval.c: freebsd 3.0 <sys/select.h> support.
+
+Fri Jun 20 01:24:45 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970620
+
+	* gc.c: eliminate uninitilalized field of Hash, Array etc., to
+	  avoid dumping core.
+
+Thu Jun 19 01:29:44 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970619
+
+	* string.c (str_split_method): wrong limit.
+
+Sat Jun 14 01:54:16 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* class.c (rb_singleton_class): no singleton for special
+	  constants (now raises exception).
+
+	* eval.c (ruby_init): cbase in TOPLEVEL_BINDING need to be
+	  initialized.
+
+Sat Jun 14 01:01:16 1997  maeda shugo  <shugo@po.aianet.ne.jp>
+
+	* array.c (sort_2): wrong comparison.
+
+Sat Jun 14 00:53:44 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* hash.c (hash_foreach): safe iteration.
+
+Fri Jun 13 14:04:56 1997  Michio "Karl" Jinbo  <karl@marcer.nagaokaut.ac.jp>
+
+	* configure.in: -Bshareable option for netbsd.
+
+Fri Jun 13 01:16:22 1997  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* io.c (pipe_open): call io_unbuffered() only for writable pipes.
+
+Thu Jun 12 01:14:15 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970612
+
+	* ext/socket/socket.c (sock_new): use io_unbuffered().
+
+	* ext/marshal/marshal.c (w_long): compact long format, which
+	  supports 64 bit architectures (unless longs are >32 bit size).
+
+	* ext/marshal/marshal.c: allows recursive data for marshaling.
+
+	* parse.y (rb_intern): raise exception for non-internable string.
+
+	* ext/marshal/marshal.c (marshal_load): allows direct loading from
+ 	  strings.
+
+	* ext/marshal/marshal.c (marshal_dump): allows direct dump to strings.
+
+	* ext/marshal/marshal.c (marshal_dump): interface changed.
+
+Wed Jun 11 18:26:00 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* gc.c (rb_newobj): remove needless memset().
+
+Mon Jun  9 13:03:43 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): reduce condition checks from while/until loop.
+
+	* eval.c (rb_eval): wrong jump point for `next'.
+
+Fri Jun  6 11:47:39 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.c (ruby_set_argv): initialize dln_argv0 for dln_a_out.
+
+	* ext/socket/socket.c (open_unix): display path name for exceptions.
+
+	* ruby.c (proc_options): option -S did not work well.
+
+Fri May 30 02:14:44 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970530
+
+	* eval.c (eval): set $! properly if exception raised in eval().
+
+	* io.c (io_write): now handles non T_FILE object.
+
+	* io.c (io_defset): $< can be anything which has `write' method.
+
+Thu May 29 15:40:22 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (eval): $@ is always an array (not string).
+
+	* pack.c (pack_unpack): avoid corrupting memory for unexpected
+	  input strings.
+
+Wed May 28 12:46:13 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970528
+
+	* process.c (rb_waitpid): do not block other threads.
+
+Tue May 27 12:02:31 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (ruby_init): split initialize and processing command line
+	  options.
+
+	* ruby.c (ruby_options): ruby_init(0, 0, envp) dumps core.
+
+Tue May 20 18:59:45 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* variable.c (rb_ivar_set): invalid instance variable access for
+	  built-in object raises TypeError.
+
+Fri May 16 17:32:21 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970516
+
+	* dir.c (push_globs): was freeing non heap pointer.
+
+	* gc.c: remove some duplicated prototypes.
+
+	* ext/kconv/kconv.c: fix prototypes.
+
+Fri May  9 11:38:59 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970509
+
+	* gc.c (obj_free): avoid free(NULL).
+
+	* eval.c (rb_check_safe_str): argument missing for TypeError().
+
+Thu May  8 01:14:28 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (file_s_dirname): need to return "." for path without
+	  slashes.
+
+Wed May  7 19:18:48 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* process.c (f_fork): child processe does not inherit parent's
+ 	  itimer setting on linux.  call setitimer() again in the child
+	  process.
+
+Sat May  3 02:49:43 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/curses/curses.c: modified for portability and add to the
+ 	  standard distribution.
+
+Wed Apr 30 00:34:00 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (file_s_size): returns 0 for empty files (not FALSE).
+
+Fri Apr 25 02:17:50 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970425
+
+	* eval.c (f_load): free unused name-table.
+
+	* eval.c (f_load): copy local variable name-table.
+
+	* gc.c (obj_free): avoid free(NULL).
+
+	* eval.c (rb_eval): forgot to make link from the scope object to
+	  NODE_SCOPE.  It may crash the interpreter.
+
+Thu Apr 24 00:35:09 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* random.c (f_srand): save old seed anyway.  srand() returns no
+ 	  value on some systems.
+
+	* gc.c (obj_free): avoid double free of the local variable name
+	  table.
+
+	* parse.y (top_local_setup): modify realloc to handle offset.
+
+Tue Apr 22 12:58:26 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970422
+
+Thu Apr 17 00:40:51 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* configure.in (rb_cv_bsdpgrp): proper check for BSD
+	  setpgrp/setpgrp.
+
+Wed Apr 16 16:14:02 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (proc_call): proc called in other thread must be orphan.
+
+Tue Apr 15 10:46:31 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970415
+
+	* gc.c (obj_free): NODE_SCOPE marked from SCOPE object.
+
+	* gc.c (gc_mark): some nodes marked wrong.
+
+	* process.c (proc_getpgrp): wrong argument
+
+Fri Apr 14 18:32:42 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970414
+
+Fri Apr 12 01:20:12 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.h: String pointer changed to unsigned char.
+
+Fri Apr 11 10:27:29 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970411
+
+	* Makefile.in: create libruby.a before linking ruby.
+
+	* string.c (str_strip_bang): >0x80 characters for isspace().
+
+	* eval.c (proc_call): set safe-level temporally
+
+	* eval.c (proc_s_new): save safe-level in the proc context.
+
+	* eval.c (rb_eval): no class/module extension in safe mode.
+
+Thu Apr 10 02:10:41 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* gc.c (gc_mark): remove some pointer checks for speeding up.
+
+	* ruby.c (ruby_options): set $0 temporally for -r option.
+
+	* eval.c: built-in security feature.
+
+	* gc.c (gc_sweep): do not free nodes during compile.
+
+	* parse.y (yycompile): set flag when compiling.
+
+Wed Apr  9 10:19:02 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.c: forgot to include <ctype.h> for isspace().
+
+	* file.c: provide S_ISREG for some platforms.
+
+	* io.c (Init_IO): added some $< operations.
+
+	* lib/ping.rb: check host upness using TCP echo.
+
+Tue Apr  8 00:10:15 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (arg_read): bug with 0 length input.
+
+Mon Apr  7 11:36:16 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/fcntl/fcntl.c: module for fcntl constants.
+
+	* eval.c (rb_alias): bug when original was an alias.
+
+	* parse.y (primary): syntax to access singleton class.
+
+	* eval.c (mod_public_method): method's to specify visibitily of
+	  the class methods.  make_method_{public,private} removed.
+
+Fri Apr  4 21:43:57 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970404
+
+	* gc.c (obj_free): finalizer added for experiment.
+
+Thu Apr  3 02:12:31 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_schedule): make Fatal rise on main_thread on
+	  deadlocks.
+
+	* eval.c (thread_join): raise ThreadError instead of Fatal, in
+	  case of deadlock.
+
+	* regex.c (re_compile_fastmap): uninitialized local variable.
+
+	* parse.y (parse_regx): new option //[nes] to specify character
+	  code for regexp literals.  Last specified code option is valid.
+
+	* re.c (reg_s_new): additional 3rd argument to specify compiled
+ 	  regexp's character code.
+
+	* re.c (reg_new_1): regexp character code can be specified for
+	  each regexp object.
+
+Wed Apr  2 14:51:06 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_create): handle uncaught throw.
+
+	* eval.c (thread_create): halt on some deadlock conditions.
+
+	* regex.c (is_in_list): wrong result for non-mbc higher-byte
+	  characters.
+
+	* regex.c (re_match): wrong skip for multi-byte characters.
+
+	* regex.c (re_compile_fastmap): wrong fastmap in non-mbc mode.
+
+	* hash.c (Init_Hash): hash compatible features added to ENV.
+
+Tue Apr  1 15:24:06 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (obj_extend): remove Object#extend as an iterator which
+ 	  is in experimental state, since it unveils internal singleton
+ 	  classes.
+
+Mon Mar 31 14:29:39 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970331
+
+Sun Mar 30 19:40:57 1997  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* parse.y (terms): avoided win32 gcc's optimization bug.
+
+Sat Mar 29 11:21:58 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* struct.c (make_struct): St[val,..] creates new structure.
+
+Fri Mar 28 11:24:51 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (obj_make_private): new method make_method_{public,private}
+	  to change visibility of singleton methods.
+
+	* regex.c (re_compile_pattern): enables numeric literal >= 0x80 in
+	  the character class.
+
+	* regex.c (re_compile_pattern): enabled numeric literal >= 0x80,
+	  in multibyte mode.
+
+	* regex.c (re_compile_fastmap): modified exantn and charset(_not)
+	  to set fastmap for higher bytes properly.
+
+	* regex.c (is_in_list): now matches numeric literals.
+
+Thu Mar 27 13:34:20 1997  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* pack.c (pack_unpack): extra null byte after unpacked string.
+
+Wed Mar 26 15:20:34 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* regex.c (re_compile_pattern): register numbers must be fit in a
+ 	  byte (0 <= regnum <= 0xff).
+
+	* regex.c (re_compile_fastmap): forgot to set mbchar map for
+ 	  charset_not if RE_MBCTYPE is on.
+
+	* regex.c (re_compile_pattern): set list bits for multi-byte
+	  characters for \W, \S, \D in range expression.
+
+	* object.c (obj_is_kind_of): defined that nil itself is kind of
+ 	  nil.  TRUE is kind of TRUE, FALSE is kind of FALSE likewise.
+  	  This change makes `obj.kind_of?(eval(obj.type))' always true.
+
+Tue Mar 25 14:08:43 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* lib/English.rb: provides nicer English alias for the variables.
+
+	* parse.y (expr): alias $var1 $var2 makes alias of the global
+	  variable.
+
+Mon Mar 24 18:23:20 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970324
+
+Thu Mar 20 22:04:59 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (mod_modfunc): forget to clear method cache.
+
+Wed Mar 19 17:06:55 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (program): set methods' default private/public status
+	  correctly under eval().
+
+	* eval.c (eval): set the_class correctly while evaluating string.
+
+Tue Mar 18 12:23:53 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (eval): yield can be called from eval().
+
+	* version 1.0-970318
+
+	* parse.y (program): regexp in condition expression should do
+	  matching operation with $_.
+
+	* re.c (reg_regsub): wrong substitution.
+
+Fri Mar 14 14:36:28 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* hash.c (hash_invert): returns value to key mapping of the
+	  associative array.
+
+	* ext/socket/extconf.rb: set environment variable SOCKS_SERVER to
+	  compile with libsocks.a.
+
+	* ext/socket/socket.c (socks_s_open): SOCKSsocket class to access
+	  internet via SOCKS library.
+
+	* sprintf.c (f_sprintf): unsigned formats display leading double
+ 	  dots for imaginary sequence of signed bit to the left.
+
+	* sprintf.c (f_sprintf): correct width and precision formatting
+	  for big integers.
+
+	* parse.y (yylex): enables negative hex/octal numbers and `_' in
+	  non-decimal numbers.
+
+	* sprintf.c (f_sprintf): %u added for unsigned decimal format.
+
+Thu Mar 13 10:24:27 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sprintf.c (f_sprintf): wrong output for bignums.
+
+	* array.c (ary_reverse_each): iterates in reverse order.
+
+	* pack.c (pack_unpack): L unpacked signed long.
+
+	* io.c (f_backquote): now returns an empty string for no output.
+
+Wed Mar 12 10:20:30 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/socks/socks.c: socket module with socks library.
+
+Mon Mar 10 20:44:22 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* re.c (reg_regsub): \& for substitution.  \`, \', and \+ are
+	  avaiable also.
+
+Thu Mar  6 01:47:03 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970306
+
+	* sample/rubydb.el (gud): ruby debugger emacs interface
+
+	* lib/debug.rb: ruby debugger
+
+	* parse.y (exprs): more accurate line number display.
+
+Wed Mar  5 21:31:46 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970305
+
+Tue Mar  4 12:28:32 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.c (proc_options): search through RUBYPATH and PATH for
+	  option -S.
+
+Mon Mar  3 22:44:55 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_status): returns nil for exception terminated
+	  threads.
+
+	* eval.c (thread_value): re-raise exceptions.
+
+Sat Mar  1 00:59:47 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): restore $! value after rescue clause, to
+	  re-raise exceptions correctly.
+
+Fri Feb 28 16:43:38 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970228
+
+Thu Feb 27 11:23:41 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_yield_0): redo raises exception
+
+	* eval.c (thread_schedule): bug in interrupt handling by rescue.
+
+Wed Feb 26 00:55:36 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (eval): forgot to restore dynamic local variable
+	  bindings.
+
+Tue Feb 25 11:22:08 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/aix_ld.rb: AIX dynamic load support (not tested).
+
+	* eval.c (rb_eval): wrong return value for defined? super.
+
+	* error.c (exception): more error check.
+
+	* re.c (reg_regsub): wrong substitution when sub expanded to null
+ 	  string.
+
+Fri Feb 21 13:01:47 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970221
+
+	* eval.c (f_require): volatile added.  register variable was
+ 	  recycled, so that GC did not mark that variable.
+
+	* object.c (Init_Object): forget to mark main object (was mostly
+	  ok, but made trouble with early GC.)
+
+Thu Feb 20 11:50:50 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970220
+
+Thu Feb 20 11:25:50 1997  Yasuo OHBA  <jammy@shljapan.co.jp>
+
+	* lib/date.rb: update
+
+Thu Feb 20 08:25:57 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (yylex): forgot tokfix() before rb_intern().
+
+	* lib/tk.rb (TkVariable): give up using trace_var.
+
+Wed Feb 19 00:24:35 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970219
+
+	* pack.c (pack_pack): packed by null for A specifier. must be
+	  space filled.
+
+	* pack.c (pack_unpack): bug in skipping spaces
+
+	* gc.c (xmalloc): garbage collect for every 4 Meg. allocation.
+
+	* string.c (str_split_method): limit worked wrong way.
+
+	* io.c (io_gets_method): misunderstand 0xff in binary files when
+	  $/ == nil.
+
+	* re.c (reg_regsub): re-implement.
+
+	* ext/socket/socket.c (thread_connect): remove O_NONBLOCK, which
+	  is not defined on some platform like NeXT.
+
+Mon Feb 17 13:08:30 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970217
+
+	* object.c (mod_eqq): === extended for subclass check (to use case
+	  as typecase).
+
+Sat Feb 15 02:07:22 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* regex.c (re_compile_pattern): wrong match backref at end of pattern.
+
+	* io.c (arg_read): now works beyond end of file.
+
+Thu Feb 13 16:21:24 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (expr): return/yield now accept normal argument format.
+
+	* parse.y (yylex): a star in `yield *x' must not be multiplication
+	  operator.
+
+Wed Feb 12 15:06:44 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* time.c (time_plus): bug in simple addition.
+
+	* eval.c (thread_raise): raise exceptions from outside.
+
+	* eval.c (Init_Thread): Thread#alive? -- alias for Thread#status.
+
+Mon Feb 10 00:38:55 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.h (Data_Make_Struct): rename macros.
+
+Sun Feb  8 11:48:13 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (f_syscall): argument offset was wrong.
+
+Fri Feb  7 18:01:17 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970207
+
+	* eval.c: add volatiles to avoid variable crobbering by longjmp().
+
+	* eval.c (f_raise): 1st argument can be the GlobalExit object now.
+
+	* array.c (ary_unshift): no longer accept more than 2 args.
+
+	* eval.c (f_raise): bug if 2nd argument is the exception.
+
+Tue Feb  4 00:37:29 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970204
+
+	* eval.c (eval): check compile errors by nerrs.
+
+	* eval.c (rb_eval): check syntax error by nerrs, not by the return
+ 	  value, which may be NULL.
+
+	* eval.c (compile): Do not clear errinfo.
+
+Mon Feb  3 10:13:06 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (obj_extend): move real inclusion to Module#extend_object
+	  to allow redfinition.
+
+	* object.c (Init_Object): Kernel class is now Module. Object class
+	  became the true root class.
+
+	* object.c (obj_inspect): remove useless buffer.
+
+	* hash.c (any_cmp): disable interrupts and context switching.
+
+	* st.c: remove ALLOW_INTS to disable interrupt during operations.
+
+Fri Jan 31 22:10:08 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* hash.c (hash_rehash): re-register all key-value.
+
+Thu Jan 30 02:14:49 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (io_reopen): re-implement according to clone() way.
+
+	* io.c (io_clone): copy IO object.
+
+	* struct.c (struct_eql): compare elements by eql?.
+
+	* io.c (io_mode_flags): detect "rb", "wb" etc.
+
+	* io.h (FMODE_BINMODE): added.
+
+	* ext/socket/socket.c (Init_socket): undef BasicSocket.new
+
+	* file.c (Init_File): File.new(path[,mode])
+
+	* io.c (Init_IO): IO.new(fd[,mode])
+
+	* eval.c (rb_method_boundp): forgot to enable priv argument.
+
+	* object.c (Init_Object): remove `=~' from Kernel class.
+
+	* ext/socket/socket.c (open_inet): initialize sockaddr before
+	  calling bind(2).
+
+	* sample/ruby-mode.el (ruby-calculate-indent): skip comment lines
+
+Wed Jan 29 18:43:22 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (Init_Thread): DEFER_INTS during initializing threads.
+
+	* hash.c (Init_Hash): Hash#eql? checks for object identity.
+
+	* eval.c (thread_set_critical): wrong value assigned.
+
+Mon Jan 27 16:10:51 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (io_print): remove print_on().
+
+	* eval.c (f_missing): proper error message for undefined method
+	  without argument
+
+Sat Jan 25 23:32:32 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_sub_s): false alert - sub() does not modify string.
+
+	* array.c (ary_times): negative multiplication detected
+
+	* string.c (str_times): negative multiplication detected
+
+Fri Jan 24 10:51:39 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* time.c (time_arg): month -> 0 == "jan" == "1" == "01", little bit
+	  confusing but wanted to conform japanese style.
+
+	* version 1.0-970124
+
+Fri Jan 24 09:52:49 1997  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* util.c (_fixpath): supports SJIS filenames on DJGPP.
+
+Thu Jan 23 16:52:06 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* README.EXT: update. partially translated into English.
+
+	* ext/extmk.rb.in: inherit $LDFLAGS to the final link.
+
+	* ext/socket/socket.c (Init_socket): add various constants.
+
+Mon Jan 23 11:40:59 1997  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* eval.c (Init_Thread): allocate main_thread first to avoid crash.
+
+Thu Jan 23 02:09:26 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* gc.c (ObjectSpace): API modified. each_object method will do all
+ 	  the iteration.
+
+	* eval.c (proc_call): wrong return from nested lambda.
+
+	* ext/GD/GD.c: debugged.
+
+Wed Jan 22 16:12:25 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970122
+
+	* gc.c (gc_mark): forgot to mark match->str.
+
+	* ext/GD/GD.c: GD interface module.
+
+	* eval.c (PUSH_BLOCK): wrong value pushed as the block level.
+
+Mon Jan 20 14:01:31 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_run): no context switch in the critical section.
+
+Mon Jan 20 09:40:59 1997  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* utils.c: supports 8+3 filenames
+
+Sat Jan 18 01:23:03 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970118
+
+	* regex.c (PATFETCH): need cast to unsigned char.
+
+	* io.c (io_ctl): bug in case when arg is not a string.
+
+	* lib/tk.rb: forgot that Kernel#type returns the class name now.
+
+	* regex.c (re_search): "abc\n" =~ "^$" should not match.
+
+Fri Jan 17 12:31:37 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970117
+
+	* ruby.c (ruby_options): constant PLATFORM, which is in the {cpu}-{os}
+	  form, defined.
+
+	* configure.in: platform information embedded in the interpreter.
+
+	* regex.c (re_search): /^$/ did not match to "" by wrong exit condition.
+
+	* lib/thread.rb: re-write Mutex/Queue based on Thread.critical.
+
+	* eval.c (thread_set_critical): remove Thread.exclusive, add
+	  Thread.critical = TRUE/FALSE instead.
+
+	* re.c (reg_search): re-compile pattern if needed
+
+	* regex.c (PATFETCH): do translate at compile time
+
+Thu Jan 16 00:49:10 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* gc.c (gc_mark_frame): forgot to mark frame->cbase.
+
+	* regex.c (re_compile_pattern): /a$|b)/ causes error.
+
+	* regex.c (re_compile_pattern): /(^|b)/ causes error.
+
+	* version 1.0-970116
+
+	* re.c (Init_Regexp): set RE_CONTEXTUAL_INVALID_OPS flag.
+
+Tue Jan 14 02:09:06 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (proc_call): Proc#callをイテレータとして呼んだ時に対応
+
+	* configure.in: nextstep対応？
+
+	* eval.c (rb_eval): a[b]=cで無駄な配列を割り当てない
+
+	* eval.c (f_send): イテレータとして呼ばれたらイテレータとしてメソッ
+	  ドを呼ぶ．
+
+	* string.c (str_new4): match共有用の生成関数
+
+	* re.c (reg_search): matchの実体(文字列)をマッチを行った文字列と
+	  copy-on-writeで共有
+
+	* string.c (str_hash): toupperをかける条件が違っていた
+
+	* array.c (sort_2): FixnumとStringを特別扱いして高速化
+
+Mon Jan 13 11:03:53 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_create): threadが生成されるまで割込みを設定しない
+
+	* eval.c (Init_Thread): 割込みタイミングを100msecに
+
+Sat Jan 11 00:17:05 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* regex.c (re_search): マッチに失敗する場合があった(本当に直ったか？)
+
+	* io.c (io_ioctl,io_fcntl): 第2引数を省略可能に
+
+	* io.c (io_ioctl,io_fcntl): 戻り値がIOだった．整数(システムコール
+	  の戻り値)を返すようにした．
+
+	* io.c (io_ctl): 引数が整数の時に対応
+
+	* io.c (io_fcntl): file.cから移動
+
+Fri Jan 10 17:01:47 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-970110
+
+	* ext/socket/socket.c (thread_connect): open(connect(2))で他の
+	  threadをブロックしないように
+
+	* eval.c (thread_create): exitでないときにexitだと思い込む
+
+Mon Jan  6 17:42:22 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_sub_s): 文字列長より長いoffsetの検出
+
+	* regex.c (re_search): 空にマッチするパターン後の$で失敗
+
+Thu Jan  2 16:36:23 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (file_reopen): Fileのreopen(pathまたはIOで指定)．
+
+	* io.c (io_reopen): IOのreopen(IOで指定) -- change classつき
+
+Wed Jan  1 11:09:01 1997  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (f_select): timeoutでnilを返す
+
+Fri Dec 27 13:06:44 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (file_s_open): サブクラスではそのクラスのインスタンスを返
+	  すように．
+
+Fri Dec 27 08:58:27 1996  <ono@isl.nara.sharp.co.jp>
+
+	* numeric.c (flo_to_s): index()を使わない．strstr()に．
+
+Thu Dec 26 01:34:17 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* lib/tk.rb: placeが使えるように
+
+	* pack.c (endian): マクロDYNAMIC_ENDIANを指定すると実行時にendian
+	  を判定するように．
+
+	* eval.c (thread_alloc): 初期化忘れのメンバがあった．
+
+Wed Dec 25 00:33:19 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 1.0-961225
+
+	* io.c (Init_IO): newを無効化
+
+	* lib/tkthcore.rb: tk_call "global $foo; set foo 5"などもできるように
+
+	* eval.c (thread_restore_context): $~, $_でスタックを壊していた
+
+	* process.c (rb_waitpid): threadに一応対応
+
+Tue Dec 24 15:20:58 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.4-961224
+
+	* configure.in: charがunsignedかどうかもチェック
+
+	* regex.c (SIGN_EXTEND_CHAR): __CHAR_UNSIGNED__にも対応
+
+	* pack.c (pack_unpack): 明示的にsigned charを指定．
+
+Mon Dec 23 14:41:23 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.c (load_file): 標準入力からのスクリプトで一時ファイルを使わ
+	  ないように
+
+	* object.c (f_integer): `0x', `0'などでbaseを解釈するように．
+
+Fri Dec 20 01:44:39 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* Makefile.in (flock.o): flockに対応
+
+Thu Dec 19 20:13:32 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.4-961219
+
+Wed Dec 18 00:06:48 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* glob.c (glob_filename): strrchrがマクロの場合に対応
+
+	* configure.in: <sys/select.h>をチェック
+
+	* ext/kconv/kconv.c: 1.62ベースに
+
+	* ext/kconv/kconv.c: Kconvモジュール
+
+	* string.c (str_substr): lenが元の文字列より長い時に対応
+
+	* parse.y (iterator): 「$bar do .. end」などは許さないように
+
+	* parse.y (iterator): FID(foo!,foo?)をdo形式のイテレータにできる．
+
+	* missing/flock.c (flock): lockf()を使って代替
+
+	* file.c (file_flock): flockを実装
+
+Tue Dec 17 12:13:38 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.4-961217
+
+Fri Dec 13 02:05:03 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* configure.in: RUBYLIBのカレントを後回し(@mix/awk offline)
+
+	* dln.c: AIXに対応した？(@mix/awk offline)
+
+	* eval.c (thread_schedule): critical sectionでも明示的なコンテキス
+	  トスイッチは起きないとまずい
+
+	* re.c (reg_search): matchに失敗した時に$~をnilに．
+
+	* re.c (reg_search): 毎回matchを生成するように
+
+Thu Dec 12 17:03:30 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* numeric.c (flo_to_s): 2.0.to_s -> 2.0に
+
+	* eval.c (thread_save_context): $_, $~をthread毎に保存
+
+	* eval.c (thread_kill): main threadではexit(0)
+
+	* string.c (str_split_method): 間違った結果を返していた
+
+Thu Dec 12 15:32:48 1996  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* dir.c: CYGWIN32対応
+
+	* ext/socket/socket.c: CYGWIN32対応
+
+	* io.c: CYGWIN32対応
+
+Thu Dec 12 14:43:51 1996  Jun Kuroda  <j_kuro@pluto.ai.kutech.ac.jp>
+
+	* lib/tk.rb: wish4.2も探索候補に含める
+
+	* config.guess: JCC対応
+
+Thu Dec 12 00:41:17 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.4-961212
+
+	* parse.y (parse_string): """..."""はやはり無くすことにした
+
+	* parse.y (parse_regx): %r|...|でterminatorを \ でエスケープできる
+	  ように
+
+	* signal.c (posix_signal): sigactionを使うsignal
+
+	* configure.in: posix signal/bsd signalの検出
+
+Wed Dec 11 17:47:35 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_schedule): critical sectionではコンテキストスイッ
+	  チが起きないように
+
+	* lib/thread.rb: SharedMutexクラス
+
+	* lib/jcode.rb: String#scanを使うように
+
+Tue Dec 10 12:21:28 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961210
+
+	* string.c (str_split_method): 正規表現に()を含む時にバグ
+
+	* lib/jcode.rb: ちょっとましになった
+
+	* string.c (tr_setup_table): 置換文字が短すぎる(2文字)のときのバグ
+
+Mon Dec  9 11:38:04 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_scan): 文字列のマッチを行う．イテレータとしても動
+	  作する
+
+	* regex.c (re_copy_registers): allocatedが初期化されていなかった
+
+	* re.c (match_to_s): $~の文字列化
+
+	* re.c (match_to_a): $~を配列化できるように
+
+	* re.c (match_getter): レジスタが初期化されていなかった
+
+Thu Dec  5 11:06:10 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_split_method): マッチしなかった括弧は空文字列を
+	  pushするべきではない
+
+	* string.c (str_succ): アルファベットを含まない文字に対応
+
+Wed Dec  4 10:48:09 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961204
+
+	* io.c (io_binmode): DJGPPでのbinmode対応
+
+	* sprintf.c (f_sprintf): intの範囲の数値は直接sprintfで変換する
+
+	* sprintf.c (f_sprintf): "%02s"に頼らない
+
+	* re.c (reg_search): indexでSEGV
+
+Tue Dec  3 10:09:36 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961203
+
+	* ext/extmk.rb.in (install): INSTALL_DATAからINSTALLに変更
+
+	* dln.c: hpux対応
+
+	* string.c (str_aset_method): 負の値を含む範囲でも例外を起こさない
+
+	* array.c (ary_replace): 負の値を含む範囲でも例外を起こさない
+
+	* array.c (beg_len): beg==endの時，長さ0に
+
+Mon Dec  2 14:07:12 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* configure.in: HP shl対応
+
+	* string.c (str_upto): beg > endの時無限ループに落ちるのを止めた
+
+	* range.c (range_each): String#uptoが再定義された場合に対応
+
+	* string.c (str_split_method): "ABC".split(/(B)/)が誤動作
+
+Sat Nov 30 01:43:52 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): undefでSEGV
+
+Fri Nov 29 12:17:59 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-parse-region): %Q#..#などに対応．しか
+	  し，区切り文字が演算子で行末にある場合には対応できなかった．
+
+	* re.c (reg_raise): 例外でもスラッシュをエスケープ
+
+	* re.c (reg_inspect): スラッシュをエスケープ
+
+	* parse.y (parse_string): `%[QqXxRr](.)..\1'なる文字列形式(テスト
+	  採用)
+
+	* parse.y (parse_qstring): '''...'''の形式
+
+	* ext/dbm/dbm.c (Init_dbm): 述語key?,value?の追加
+
+	* ext/dbm/dbm.c (Init_dbm): includes->include?
+
+	* hash.c (Init_Hash): 述語key?,value?,include?の追加
+
+	* eval.c (rb_eval): unlessでelse節が実行されない(うーん)
+
+	* string.c (str_sub_iter_s): イテレータブロック内でマッチが行われ
+	  ると位置がずれる(時に無限ループに落ちる)
+
+	* string.c (str_resize): lenが0の時sizeの調整が行われなかった
+
+Thu Nov 28 00:59:54 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961128
+
+	* parse.y (parse_string): 3-quote styleの文字列(例:"""abc"d"e""")
+
+	* configure.in (EXTSTATIC): extを静的にリンクする時にはrubyはdllを
+	  使うように
+
+	* io.c (Init_IO): getsの引数が間違っていた
+
+	* string.c (str_each_line): RSを明示的に指定できるように
+
+Wed Nov 27 12:37:46 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961127
+
+	* eval.c (rb_eval): iver defined? でselfを指定するのを忘れた
+
+	* io.c: gets等でRSを明示的に指定できるように
+
+	* ext/extmk.rb.in (install): static linkに失敗
+
+Tue Nov 26 10:33:04 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961126
+
+	* string.c (str_sub_s): 置換後の文字列長さが間違っていた
+
+Mon Nov 25 09:11:22 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* numeric.c (fix_rshift): 32以上の右シフトで0を返すように(Cの
+	  rshiftは(x>>(y%32))を返していた)．
+
+	* string.c (str_gsub): 置換が行われない場合があった
+
+	* string.c (str_resize): 本当に必要な時だけrealloc
+
+Thu Nov 21 04:13:21 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* configure.in (EXTSTATIC): --with-static-linked-extで全てのモジュー
+	  ルを静的リンクするように
+
+	* pack.c (pack_unpack): 行末の改行がない時にもチェックサムをスキッ
+	  プするように
+
+Wed Nov 20 21:42:51 1996  Yasuo OHBA  <jammy@shljapan.co.jp>
+
+	* configure.in: freebsd対応
+
+Wed Nov 20 10:24:24 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/extmk.rb.in (install): 通常リンク用のLDFLAGSとダイナミックリ
+	  ンク用のDLDFALGSを分離
+
+	* ext/extmk.rb.in (install): コンパイルの成功したものを静的リンク
+	  のリストに追加する
+
+	* eval.c (f_missing): オブジェクトの文字列表現が長すぎる時バッファ
+	  を書き潰していた
+
+	* process.c (proc_exec_v): forkした後例外を発生させてはいけない
+
+Tue Nov 19 13:28:15 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961119
+
+	* eval.c (mod_method_defined): Module#method_defined? の追加
+
+	* parse.y (call_args): 引数が唯一のコマンドコールである時のバグ(戻
+	  り値が展開されてしまう)
+
+Mon Nov 18 13:28:18 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_sub): 失敗した時にnilを返していた
+
+	* string.c (str_split_method): 検索開始位置が移動してなかった
+
+	* ext/socket/socket.c (sock_s_getservbyaname): まだ間違っていた
+
+	* version 0.99.3-961118
+
+	* string.c (str_sub_s): 元の文字列を置換するのを止めた
+
+	* pack.c (encodes): 領域外をアクセスしていた
+
+Fri Nov 15 17:10:35 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* bignum.c (big_divmod): Bignumが引数の場合の対応忘れ
+
+	* sample/ruby-mode.el (ruby-expr-beg): word?形式への対応が不完全
+
+Wed Nov 13 15:42:40 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_tr_s_bang): tr_sでtrが行われていなかった
+
+	* eval.c (rb_eval): autoloadクラスのチェック
+
+	* string.c (f_sub): subがsub!と同じ動作になっていた
+
+	* eval.c (thread_sleep): stopとsleepの分離
+
+Mon Nov 11 13:53:19 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961111
+
+	* numeric.c (fix_step): to, stepが整数以外の場合に対応
+
+	* eval.c (rb_call): dynamic varがdynamic scopingになっていた(これ
+	  はまずい)
+
+	* string.c (str_chop_bang): 長さ0の文字列のchopで，領域外のアクセ
+	  スが発生していた．
+
+	* parse.y (yyerror): 割り当てた領域外をアクセスしていた
+
+Fri Nov  8 11:54:46 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_yield): scopeをheapにコピー
+
+Thu Nov  7 09:56:53 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* numeric.c (num_coerce): とりあえず両辺をFloatに変換することに
+
+Wed Nov  6 10:45:13 1996  Yasuo OHBA  <jammy@shljapan.co.jp>
+
+	* lib/parsearg.rb: 第2引数を変更．
+
+Tue Nov  5 14:21:09 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961105
+
+Sat Nov  2 01:11:40 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* bignum.c (big_pow): typo (dy -> dx)
+
+	* bignum.c (big_divmod): 知らない型はfloatに変換してみる
+
+	* numeric.c (fix_lshift): 境界条件のバグ(負になっていた)
+
+	* bignum.c (big_pow): 無駄なfloatへの変換をなくした
+
+	* math.c (math_atan2): typo(x -> y)
+
+Fri Nov  1 15:30:59 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/socket/socket.c (sock_gethostname): gethostnameがない時には
+	  unameを使ってホスト名を得る
+
+	* ext/etc/etc.c (etc_getlogin): getloginがNULLを返しても環境変数を
+	  調べるように
+
+	* object.c (krn_clone): オブジェクトのフラグもコピー
+
+	* hash.c (rb_cmp): ハッシュの比較を`=='でなく`eql?'に変更
+
+	* math.c (Need_Float): Float()を使って変換する
+
+	* compar.c (cmp_gt): 以前の右辺を返す仕様の名残が残っていた
+
+Thu Oct 31 12:55:51 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961031
+
+	* numeric.c (Init_Numeric): typo
+
+	* eval.c (error_print): 長すぎるtrace backを途中省略する
+
+	* regex.c (re_compile_pattern): 全角のrangeに対応
+
+Wed Oct 30 03:03:18 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.3-961030
+
+	* io.c (f_ungetc): 関数を追加
+
+	* eval.c (dyna_var_asgn): return値忘れ
+
+Tue Oct 29 10:05:28 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (f_split): 関数splitを追加
+
+	* eval.c (rb_call): ネストした外側のクラス/モジュールの定数を参照
+	  できるように
+
+Mon Oct 28 09:51:03 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_sub): offsetが文字の末尾にある時のチェック
+
+	* regex.c (re_match): 割り当てるレジスタの数が1多かった
+
+	* io.c (io_gets): $/ = ""の動作をperlに合わせる(awkとはちょっと違
+	  うらしい)
+
+	* io.c (io_gets): $/ = nilの時少し高速化
+
+	* string.c (str_split_method): 括弧がnullにマッチした時にも無視し
+	  ないように
+
+	* string.c (str_split_method): 括弧にマッチした分はlimitの数に含め
+	  ないように．
+
+	* numeric.c (num_coerce_bin): coerceの定義を変更，2要素の配列
+	  [x,y]を返すように
+
+	* sample/ruby-mode.el (ruby-calculate-indent): "do |aa|"の対応を改
+	  善した．
+
+Sat Oct 26 01:43:51 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/marshal/marshal.c (w_object): ビルトインクラスのサブクラスを
+	  正しく復旧できるように
+
+	* ext/marshal/marshal.c (w_object): ユーザ定義dumpの優先
+
+	* numeric.c (flo_coerce): Float()を使って定義
+
+	* numeric.c (Init_Numeric): Numericのnewのundefはまずい
+
+	* ext/marshal/marshal.c (w_symbol): シンボルの内容(文字列)は一度し
+	  かファイルに書き出さない．
+
+	* sample/ruby-mode.el (ruby-parse-region): if/while修飾子に対応し
+	  なくなっていた
+
+	* bignum.c (Init_Bignum): Bignum.newを除く
+
+	* eval.c (rb_eval): 引数評価後にファイル名と行番号を再設定
+
+	* numeric.c (flo_div): typo
+
+	* sample/ruby-mode.el (ruby-parse-region): def /, def `に対応
+
+Fri Oct 25 09:26:29 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-calculate-indent): "do |aa|"に対応
+
+	* array.c (ary_aset): indexがfixnumの場合ちょっと高速化
+
+	* eval.c (thread_fd_writable): 書き込み前のselectチェック
+
+	* array.c (ary_assoc): 無限ループに落ちた
+
+	* eval.c (thread_wait_for): selectがエラー終了した時，linux以外で
+	  の動作が正しくなかった．
+
+Thu Oct 24 08:26:48 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (backtrace): `$@'を文字列から配列に変更した．
+
+	* eval.c (eval): eval中の例外発生位置を保存する
+
+	* bignum.c (bigsub): オペランドの大小比較の失敗
+
+	* re.c (reg_search): 直接参照がない時にも`$~'がセットされるように
+
+Wed Oct 23 10:40:10 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-961023
+
+	* ext/marshal/marshal.c (r_bytes): mallocをやめ，allocaを使う
+
+	* sample/ruby-mode.el (ruby-calculate-indent): 括弧の対応を変更．
+	  ()内ではインデントをレベルを合わせるように
+
+Tue Oct 22 12:59:11 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* hash.c (hash_s_new): sizeを指定できるように
+
+	* ext/marshal/marshal.c (w_object): dumpする深さ制限を指定できるよ
+	  うに
+
+	* array.c (ary_s_new): sizeを指定した時の初期化忘れ
+
+	* object.c (f_float): big2dblの宣言忘れ．
+
+	* bignum.c (bigsub): 大きさの近いBignum同士の演算で結果が負になる
+	  場合に間違いがあった．
+
+	* array.c (ary_aset): 置換先と置換元が同じ長さの時内容を
+	  shift(memmove)しないように．
+
+	* ext/marshal/marshal.c (marshal_dump): ファイルフォーマットにバー
+	  ジョンを埋め込むように
+
+	* ext/marshal/marshal.c (tmpnam): linux-aout-dln用に定義
+
+Mon Oct 21 08:40:20 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/socket/socket.c (sock_s_gethostbyname): hostent構造体の情報
+	  を返す
+	  (sock_s_gethostbyaddr): IPアドレスからhostent構造体を得る
+	  (sock_s_getservbyaname): getservbyname(3)
+
+Fri Oct 18 10:37:36 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-indent-to): 移動先カラムが負になるバグ
+
+	* eval.c (compile): evalで元ソースの行番号でエラーを表示する
+
+Thu Oct 17 09:52:28 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (eval): evalで文法エラーがあった時にSEGV
+
+	* lib/safe.rb: Restricted.evalの中だけ制限を加える．
+
+	* eval.c (error_print): バックトレースの出力．callerで例外発生位置
+	  を調整した時に問題が出る(そんなことをしなければ良いのだが…)
+
+	* eval.c (make_backtrace): バックトレースの生成
+
+Wed Oct 16 12:56:22 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby-man-0.99.2-jp/index.html: 日本語版ドキュメントの完成(長かった…)
+
+	* re.c (reg_regcomp): $=がnilの時の処理
+
+	* string.c (f_chop): $_に対するchop
+
+Tue Oct 15 11:04:23 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-961015
+
+Mon Oct 14 18:22:38 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_schedule): BOW対応．selectが-1を返した時にバグ(実
+	  はdo .. whileがcontinueで先頭にジャンプすると思い込んでいた．条
+	  件の直前だったのね ^^);;;;;
+
+	* sample/ruby-mode.el (ruby-mode-syntax-table): ?のsyntaxが"/"では
+	  まずいらしい
+
+	* hash.c (rb_hash): name conflict
+
+Fri Oct 11 00:23:05 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-961011
+
+	* ext/marshal/marshal.c (w_object): 結局動いていなかった循環オブジェ
+	  クト対応を外した．
+
+	* hash.c (rb_hash): Fixnumと文字列の高速化
+
+	* ext/marshal/marshal.c (w_object): 無駄なデータの削除(フォーマッ
+	  トの非互換性)
+
+	* io.c (io_readline): 戻り値の不備
+
+	* ext/marshal/marshal.c (marshal_dumps): MSDOS対応
+
+	* ruby.c (load_file): MSDOS対応
+
+Wed Oct  9 17:46:27 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/extmk.rb.in (install): 無駄なコピーを避ける
+
+	* string.c (str_sub_method): マッチがなかった時のString#subの値が
+	  違っていた．
+
+	* eval.c (obj_extend): extendした時にobject_extendedを呼ぶように
+
+Tue Oct  8 00:55:38 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_alloc): 割当の平均化
+
+	* eval.c (thread_schedule): joinのバグを修正
+
+	* eval.c (thread_wait_for): selectへの割込みなどに対応
+
+	* eval.c (thread_select): linuxのselectの挙動に対応(timeoutが変化
+	  する)
+
+Mon Oct  7 09:47:19 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-961007
+
+	* eval.c (PUSH_BLOCK): the_classの保存を忘れていた．
+
+	* ext/dbm/dbm.c (fdbm_store): sizeの保存する場所が間違っていた
+
+	* ext/socket/socket.c (s_accept): thread対応していなかった
+
+Sat Oct  5 01:32:27 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (io_readchar): EOFで例外を発生させる
+
+Fri Oct  4 11:59:54 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/marshal/marshal.c (w_object): HashとObjectの復旧に必要なハッ
+	  シュテーブルが渡されていなかった．
+
+	* variable.c (rb_path2class): ユーザ定義クラスの復旧に失敗していた
+
+	* variable.c (rb_path2class): クラスが存在しない時のエラーをFatal
+	  からNameErrorへ．
+
+	* range.c (range_s_new): first,lastが両方Numericの時エラーになって
+	  いた．
+
+	* range.c: start->first, end->last
+
+Wed Oct  2 02:02:46 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c: DJGPPでchmod,chownを使えるように(ってDOSにchownがあるのか?)
+
+	* class.c (rb_singleton_class): ビルトインクラスもextendしたり特異
+	  メソッドを追加したりできるように
+
+	* variable.c (rb_set_class_path): ユーザ定義のトップレベルクラスに
+	  pathを設定しない
+
+	* eval.c (eval): 例外がRuntimeErrorに化けていた
+
+	* eval.c (eval): eval中の例外の表現の改善
+
+	* eval.c (eval): eval_with_bindingとの一本化
+
+	* eval.c (rb_eval): クラス/モジュール定義の中から定義中のクラス/モ
+	  ジュールが参照できるように
+
+Tue Oct  1 01:40:09 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-961001
+
+	* parse.y: cur_crefが2度宣言されていた
+
+	* signal.c (trap): SIGSEGV，SIGBUSのない機種に対応
+
+	* io.c (Init_IO): 引数タイプの指定間違い
+
+Mon Sep 30 15:28:00 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-960930
+
+	* config.guess,config.sub: $host_osが正しく設定されない
+
+	* eval.c (rb_eval): yieldで正しくないselfが設定されていた
+
+	* eval.c (ruby_run): toplevelの例外処理のバグ
+
+Mon Sep 30 09:13:26 1996  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* djgpp対応
+
+Sat Sep 28 02:45:10 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-960928
+
+	* sample/ruby-mode.el (ruby-beginning-of-block): ブロックの先頭に
+	  移動(正しくインデントしていないと動作しない)
+	  (ruby-end-of-block): 同上
+
+	* eval.c (class_s_new): Class#newがイテレータとして呼ばれた時は
+	  initializeもイテレータとして呼ばれるように
+
+	* signal.c (sigsegv): SEGVでbacktraceを表示するように
+
+Fri Sep 27 09:51:07 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-960927
+
+	* eval.c (error_print): 引数のないraiseでメッセージが正しく表示さ
+	  れるように．
+
+	* eval.c (rb_longjmp): mesgがnilの時RuntimeErrorを生成する．
+
+	* eval.c (f_raise): 引数がない時に対応
+
+	* eval.c (thread_mark): stack上にないデータのアドレス変換を行って
+	  いた．
+
+	* eval.c (Init_Thread): 割込みの間隔が1秒と長すぎた．
+
+Thu Sep 26 16:02:45 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_schedule): 一度ペンディングになるとフラグがクリア
+	  されていなかった．
+
+	* process.c (rb_proc_exec): system/execの引数が空文字列であった場
+	  合，例外を発生すべきだった．
+
+	* config.sub/config.guess: 新しいものに置き換え
+
+Thu Sep 26 15:41:35 1996  WATANABE Hirofumi  <watanabe@ase.ptg.sony.co.jp>
+
+	* io.c (next_argv): -i.bakをBOWとDOSに対応．
+
+Thu Sep 26 01:31:43 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (io_sysread): EOFで例外
+
+	* io.c (f_readline): EOFで例外を発生するように．getsは互換性のため
+	  nilを返すままにする
+
+	* eval.c (proc_call): lambdaからのreturnでIN_BLOCKフラグが立ったま
+	  まだった
+
+	* eval.c (PUSH_BLOCK2): threadに対応するためBlockを一度stackにコピー
+
+Wed Sep 25 11:54:11 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (method_call): Const::method()形式を使えるようにしてみた．
+	  引数括弧は省略できない．
+
+	* sample/test.rb: Process.killの存在を確かめてからテストを行う
+
+	* eval.c (eval_with_binding): 第2引数としてbinding(またはlambda)を
+	  与えるとその環境でevalを実行するようにした
+
+	* eval.c (f_binding): 現在のbindingを返す関数
+
+	* eval.c: block構造体にthe_classを保存するメンバを追加
+
+	* process.c (Init_process): kill,wait,waitpidをProcessに移動
+
+Tue Sep 24 02:44:43 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el: いろいろ問題が多いので以前の高速化は破棄．
+	  別のアプローチを使った．
+
+	* lib/tk.rb (Tk.pack): 複数のウィンドウを受け付けるpack
+
+Sat Sep 21 11:08:09 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (exprs): 空文も受け付けるように文法を変更．今までは改行
+	  の連続だけが許されていた．
+
+Fri Sep 20 11:39:18 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* Failの大半を名前つき例外に変更．
+
+	* re.c (Init_Regexp): 名前つき例外を導入．
+
+	* eval.c (f_missing): Objectはinspectしない．
+
+	* object.c (inspect_i): Object#inspectでloopに対応．
+
+	* regex.c (re_search): /^$/が""にマッチしなかった．
+
+Thu Sep 19 19:25:12 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* regex.c (re_search): /^$/が非空行にマッチしていた．
+
+Tue Sep 17 10:28:11 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-960917
+
+Mon Sep 16 10:47:56 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-calculate-indent): 演算子継続の場合の
+	  文字列の判定のバグ
+
+	* sample/ruby-mode.el (ruby-calculate-indent): elseなどの次の行の
+	  インデント計算を正しく．
+
+Sat Sep 14 08:37:19 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.2-960914
+
+Fri Sep 13 08:06:03 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/socket/socket.c (tcpaddr): port番号にntohsをつけ忘れ
+
+	* dln.c (link_undef): テーブルの種類が間違っていた．
+
+	* bignum.c (bigadd): 引き算が発生する時に計算違いが起きていた．
+
+	* parse.y (iter_do_block): do..endでもdynamic variableを．
+
+	* bignum.c (big_pow): より正確な計算を(整数同士ではfloatに変換しな
+	  い)．
+
+Thu Sep 12 13:11:55 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* variable.c (rb_set_class_path): Stringクラスが初期化される前に
+	  Stringを作っていた．組込みクラスにはpathはいらない
+
+	* parse.y (yylex): 0.1が0になっていた
+
+	* parse.y (yylex): 行番号の不整合
+
+	* gc.c (oblist_live_obj): 今「生きている」全部のオブジェクトを返す
+	  イテレータ．そのクラス(またはサブクラス)の全部のインスタンスを返
+	  すeach_object_ofも定義した．
+
+	* class.c (rb_define_class_id): 無駄なクラスを割り当てていた．結果
+	  として未初期化のクラスオブジェクトが存在していた．
+
+Wed Sep 11 00:56:23 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (yylex): octalの定数の検出をより正確に(090はエラーとか)．
+
+	* bignum.c (big_minus): yがxより大きい場合にエラー．
+
+	* parse.y (yylex): エラー行番号の表示をより正確に
+
+	* sample/ruby-mode.el (ruby-expr-beg): 変数名が1文字の時誤動作して
+	  いた．
+
+	* sample/ruby-mode.el (ruby-calculate-indent): ?/でループに落ちい
+	  たバグを修正．
+
+	* enum.c (enum_min,enum_max): sortのようにイテレータとしても動作す
+	  るように．
+
+	* enum.c (enum_find_all): typo
+
+Tue Sep 10 12:07:12 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* node.h (nd_line): NODEのlineをflagsに押し込めてオブジェクトサイ
+	  ズを小さくした．制限:32bit intのマシンの場合，ファイルの行数が
+	  32767を越えると正常に表示されない．
+
+	* st.c: hashとcompareの関数メンバを構造体にパック，クラス的な使い
+	  方を行う．1 tableあたり4 byteの節約．
+
+Mon Sep  9 16:35:54 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (file_truncate): 提供されない時には特別な例外を発生するよ
+	  うに．
+
+	* eval.c (Init_Proc): 不適切な位置のlocal-jumpを例外に．
+
+Sat Sep  7 17:06:15 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (proc_call): まだスコープがスタック上にある時には局所脱出
+	  を有効にする．これで，procを生成してcallすることは，スコープを脱
+	  出しない限り，yieldと同じ意味を持つことになる．
+
+Fri Sep  6 13:30:59 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-indent-to): インデントが変わらない時に
+	  はバッファを変更しない．
+	  (ruby-calculate-indent): まず文字列の内部か判断してから，前の行
+	  からパーズを行う．defunが大きくなった時の高速化．
+	  (ruby-in-string-p): 文字列の内部かどうかを判断する関数(以前の
+	  parseから分離)
+	  (ruby-parse-region): 文字列に対する処理をはずす．
+	  (ruby-beginning-of-block): ブロックの先頭に
+	  (ruby-end-of-block): ブロックの末尾に(遅い…)
+
+Thu Sep  5 14:23:07 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (file_s_split): [dirname,basename]にsplitする．
+
+	* eval.c (rb_eval): evalの中でも定数の値が正しくなるように．これで
+	  定数に関しては静的なスコープが保証されるようになった．
+
+	* st.c (rehash): ハッシュ拡大の系数を2から1.79に．割算がより良い値
+	  を返すように．
+
+Thu Sep  5 00:32:07 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (class_superclass) クラスのスーパークラスを返すメソッド．
+
+Wed Sep  4 16:54:56 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* random.c (f_rand): Bignumやlongの範囲を越えるFloatに対する乱数も
+	  発生できるように．
+
+	* struct.c (struct_alloc): Fatalではなく例外を発生させるように(通
+	  常の使用で発生しうる)．
+
+	* struct.c (struct_s_members): Structの特異メソッドではなく，生成
+	  されたStructクラスの特異メソッドにした．
+
+	* st.c (st_init_table): ruby専用にパラメタを固定にした(サイ
+	  ズが減った)
+
+Mon Sep  2 11:37:59 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* array.c (ary_shift): capaがあまりにも大きい時には領域をREALLOC
+	(ary_pop): 同上
+
+	* string.c (str_inspect): multibyte character 対応にミス．
+	(str_inspect): unsigned charにしないと符号展開されてしまう
+
+	* parse.y (primary): `::'をprimaryに移動 Foo::Bar.Bazがエラーにな
+	  らないように．
+
+	* parse.y (primary): オペレータ形式の特異メソッドが定義できない
+
+	* random.c (f_rand): maxが0の時に対応
+
+	* io.c (io_printf): 関数を定義していたがインタプリタに登録していな
+	  かった．
+
+	* file.c (file_s_basename): 第2引数が無い時にエラー．
+
+Thu Aug 29 10:49:40 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (expr): イテレータの新形式に「method do .. end」形式を採
+	  用した．もちろん昔の形式も有効．
+
+	* sample/ruby-mode.el (ruby-calculate-indent): endの数の方が多い場
+	  合にもエラーを起こさないように．
+
+Wed Aug 28 09:41:36 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* numeric.c (upto,downto,step,times): 対象がfixnumの範囲を越えても
+	  動作するように．
+
+Mon Aug 26 10:04:37 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* missing/setenv.c (envix): typo(missing `== 0' for memcmp)
+
+	* dir.c (dir_foreach): foreach(dir open -> read loop -> closeまで)
+
+	* io.c (io_foreach): foreach(file open -> read loop -> closeまで)
+
+	* Fatalのうち捕捉可能ないくつかを例外に．
+
+Sat Aug 24 23:56:37 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* bignum.c (bigdivmod): FIX2INT -> INT2FIX 大間違い
+
+Fri Aug 23 18:13:03 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* regex.c (re_free_registers): allocateしていない時には当然 free
+	  してはいけない．
+
+Thu Aug 22 01:20:35 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_create): 外側から強制終了させられたthreadは
+	  cleanupする必要が無い．
+
+Wed Aug 21 09:57:28 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_create): threadを終了させた大域脱出の情報を
+	  main_threadに渡すように．
+
+	* parse.y (call_args): 最終引数に括弧を省略したメソッド呼出しを置
+	  けるように(例: print foo bar, baz == print(foo(bar,baz)))
+
+Tue Aug 20 13:37:16 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (masign): 多重代入とrest引数の動作を合わせて空の配列を代
+	  入するように．
+
+	* parse.y (arg): defined?の強度をもうちょっと強く
+
+	* eval.c (error_print): -wで例外名も表示するように
+
+	* eval.c (rb_eval): 新構文に対応
+	(handle_rescue): 捕捉する例外を kind_of? で同定
+
+	* parse.y (primary): rescueの構文を変更(同定引数の追加，複数rescue)
+
+	* Fail()のかなりを適当な例外を使うように
+
+	* eval.c (thread_interrupt): Interrupt(今はnon-local jump)は
+	  main-threadに送られるように．
+
+	* eval.c (rb_longjmp): $! の内容を文字列から例外クラスに変更
+	(rb_raise): rb_fail から名称変更
+	(rb_interrupt): 例外化
+	(rb_exit): 例外化
+
+	* error.c (Init_Exception): 例外クラスの新設(文字列のサブクラス)
+
+Mon Aug 19 19:40:52 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* signal.c (trap): 古いハンドラを返すように．
+
+Wed Aug 14 00:07:18 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_trap_eval): ハンドラのためにthreadをforkすることを止
+	  めた．
+
+	* eval.c (thread_mark): thread毎の $!, $@ をマークし忘れ
+
+	* ext/dbm/dbm.c (fdbm_delete): イテレータとして呼ばれた場合，要素
+	  が無ければブロックを評価する．
+
+	* hash.c (hash_delete): イテレータとして呼ばれた場合，要素が無けれ
+	  ばブロックを評価する．
+
+	* array.c (ary_delete): イテレータとして呼ばれた場合，要素が無けれ
+	  ばブロックを評価する．
+
+	* eval.c (rb_interrupt): SIGINTのデフォルトをexitから特別な大域脱
+	  出に．やはり割り込まれた位置の表示が無いのは寂しいので．
+
+Tue Aug 13 01:34:00 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_exit): sub-thread内でのexitもstatusを保存するように
+	(thread_create): 自thread内のexitに対応
+
+	* signal.c (sighandle): SIGINTのデフォルトハンドラはexitするように
+	  (以前は例外を発生していた)．
+
+	* 例外の一部をFatalに．
+
+	* string.c (str_aset): 文字列の置換の対象が部分文字列でなかった時，
+	  例外を発生させないように
+
+	* eval.c (proc_call): Procの中からbreak/nextは通し，他のものは通さ
+	  ないように
+
+Mon Aug 12 14:15:09 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (krn_type): 文字列を返す
+
+	* eval.c (thread_create): sub-thread内でのexitに対応
+
+	* numeric.c (fix_type): 文字列を返す
+
+	* io.c (f_p): デバッグ用データ表示メソッド
+
+	* eval.c (f_missing): nil/TRUE/FALSEを特別扱い
+
+	* string.c (str_inspect): 長い文字列を短縮表示．inspectの働きを
+	  human readable stringの生成に統一(re-generatable string は正式に
+	  無くなった)．
+
+Sat Aug 10 16:54:21 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (Init_Object): kernel/nil/false/trueのクラス名を変更(小
+	  文字に)，rubyスクリプトからアクセスできないように．
+
+	* eval.c (rb_eval): CONSTANTのアクセス先を単純化．crefを使わない．
+
+	* eval.c (f_eval): 特異メソッド内でも定数の値が正しくなるように
+
+Fri Aug  9 12:23:17 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* array.c (ary_concat): append -> concat Stringに合わせた
+
+	* parse.y (yylex): `$;'が使えなかった．
+
+	* array.c (ary_push_method): 複数引数を受け付けるように．
+	(ary_unshift): 複数引数を受け付けるように．
+
+	* io.c (io_popen): IO.popenでcommand pipeが開けるように．
+
+	* object.c (Init_Object): KernelとNilをruby scriptからアクセスでき
+	  ないように．
+
+Thu Aug  8 01:21:47 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (f_integer): 整数への変換関数
+	(f_float): 実数への変換関数
+	(f_string): 文字列への変換関数
+	(f_array): 配列への変換関数
+
+	* bignum.c (big_to_i): FIXNUMの範囲でない時はBignumのまま返すよう
+	  に変更．
+
+Wed Aug  7 09:28:38 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99.1-960807
+
+	* parse.y (mlhs): 「*foo = 1,2,3」タイプの多重代入も可能に．
+
+	* object.c (Init_Object): クラスTrue/Falseをruby scriptからアクセ
+	  スできないように．
+
+	* object.c (nil_inspect): inspect表現は"nil"に
+
+	* io.c (io_print): nilのprintをnilに．
+
+	* object.c (nil_to_s): nilの文字列表現を""に．
+
+Tue Aug  6 01:12:32 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* dir.c (dir_s_open): file descripterが足りない時にはgcしてからも
+	  う一度openしてみる．
+
+	* io.c (rb_fopen): すべてのfopen()についてfile descripterが足りな
+	  い時にはgcしてからもう一度openしてみる．
+
+	* ext/socket/socket.c (Init_socket): 定数の追加．
+
+	* sample/ruby-mode.el (ruby-indent-to): インデント後のカーソル位置
+	  の調整を正しく．
+
+	* gc.c (gc): 割込みチェックを行わない(Cコードの中で安心して
+	  malloc()が使えなくなるので)．
+
+	* st.c (call_hash_func): signalとthreadによる割込みに対応．
+
+	* sig.h (DEFER_INTS): 割込み禁止区間の指定
+
+	* eval.c (f_require): threadによるrequireの競合に対応(最初の
+	  requireが終了するまで他のthreadは待つ)．
+
+	* bignum.c (str2inum): 0x80000000の値が負になっていた
+
+	* sprintf.c (f_sprintf): 文字列末尾，行末の単独の`%'に対応
+
+	* bignum.c (big_cmp): 比較の結果が逆になる時があった．
+
+Mon Aug  5 10:58:13 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* process.c (proc_exec_v): 例外のメッセージを分かりやすく．
+
+	* ext/dbm/dbm.c (fdbm_store): nilを格納すると要素の削除になる
+
+	* ext/dbm/dbm.c: サイズをキャッシュ．
+
+Sat Aug  3 01:52:52 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_fail): `fail'が引数無しで呼ばれた時だけ以前の`$@'を保
+	  存するように．
+
+	* eval.c (f_fail): frameの調整
+
+Fri Aug  2 11:26:21 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/socket/socket.c (bsock_setopt): valとしてTRUE/FALSE/Fixnumも
+	  受け付けるように．
+
+	* ext/socket/socket.c (Init_socket): SO_REUSEADDR等の定数の追加
+
+	* ext/md5/md5init.c: md5モジュール(初の複数ファイルからなるモジュー
+	  ルでもある)
+
+	* ruby.h (Make_Data_Struct): Data: objectのinstance変数に格納 ->
+	  Data型のObjectに(Dir,Time,Proc,Thread,DBM)
+
+Thu Aug  1 11:38:44 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/dbm/dbm.c (fdbm_store): valueが文字で無い時に対応
+
+Wed Jul 31 10:53:42 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/socket/socket.c (open_inet): htonsが必要であった
+	(tcpaddr): ntohlで変換した
+
+	* process.c (rb_proc_exec): execvp -> execv
+
+Tue Jul 30 17:48:33 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c: `$?'をthread localに
+
+	* Makefile.in (install): install時にstripを行う
+
+	* configure.in: install時のstripの検出
+
+	* configure.in: NEXTSTEP対応
+
+	* version 0.99.1-960730
+
+Tue Jul 30 16:40:35 1996  SHIROYAMA Takayuki  <psi@fortune.nest.or.jp>
+
+	* dln.c (dln_load): NeXT dln(mach-o)対応．configureは未対応
+
+Tue Jul 30 09:46:51 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* process.c (f_system): 複数引数もとれるように
+
+	* process.c (f_exec): 複数引数もとれるように
+
+	* array.c (ary_append): 配列(またはEnum)の要素を破壊的に追加
+
+	* array.c (ary_plus): Enumはその要素を追加
+
+	* file.c (file_s_open): File.openを追加
+
+	* struct.c (struct_new): FIX2INTを忘れていた
+
+	* file.c (Init_File): exists? -> exist?
+
+	* object.c (obj_is_kind_of): is_kind_of? -> kind_of?, is_a?
+
+	* object.c (obj_is_instance_of): is_instance_of? -> instance_of?
+
+Mon Jul 29 16:40:02 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (parse_regx): 式展開を行った場合，casefoldの設定ができて
+	  いなかった．
+
+	* object.c (true_type): TRUE/FALSEにtypeを実装．
+
+	* parse.y (read_escape): 3文字以内のoctalに対応(\0とか)
+
+Fri Jul 26 00:31:45 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* array.c (ary_reverse_bang): in-placeで配列を反転させる
+	(ary_sort_bang): in-placeでsortする
+	(ary_sort): sortした配列を返すように
+	(ary_delete_at): 指定した位置の要素を削除する
+
+	* eval.c (rb_call): stack深さチェックを毎回は行わないように
+
+	* error.c (Warning): 実行中のwarningが表示されていなかった
+
+	* eval.c (compile): 例外発生を分離．
+
+	* eval.c (f_eval): 変数rb_in_evalを正しく管理するように
+
+	* ext/dbm/dbm.c (fdbm_store): 格納するkeyを文字列に変換
+
+	* eval.c (rb_call): 無限再帰のチェックを大域脱出を行うC methodにも
+	  対応させた．threadのstack深さチェックルーチンを流用．
+
+	* parse.y (yylex): 第1引数のunary -/+の判定が間違っていた．
+
+	* parse.y (yylex): unary +で数字を余計に読んでいた(ex. +5 -> 55)
+
+Thu Jul 25 12:15:04 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (yylex): 曖昧でない引数に対して警告を出していた．
+
+	* eval.c (iterator_p): 引数で呼んでも正しい結果を返すように．
+
+	* parse.y: break/next/redo/retryのメソッド化．
+
+	* sample/ruby-mode.el (ruby-calculate-indent): nestのチェックミス
+
+	* sample/ruby-mode.el (ruby-parse-region): 予約語のチェックを強化
+
+	* parse.y (primary): unless/untilの復活
+
+Tue Jul 23 18:50:10 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* array.c (Array#empty?), Hash.c (Hash#empty?), ext/dbm/dbm.c (DBM#empty?):
+	  空の判定述語
+
+	* eval.c (f_unless): ifの逆をするイテレータ
+
+	* eval.c (f_until): whileの逆をするイテレータ
+
+	* parse.y: notの優先順位をand/orより高く
+
+	* parse.y (expr): `!'を引数括弧を省略したcallでも有効に
+
+Mon Jul 22 10:15:38 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99-960722
+
+	* array.c (ary_print_on): OFSのNILチェックが不完全
+
+	* ruby.c  (load_file): 標準入力からのスクリプトが空の時に対応．
+
+	* ruby.c (proc_options): -wでは引数無しの時には標準入力からスクリ
+	  プトをとる(-vではたんに終了する)．
+
+	* array.c (ary_compact): nilの要素を取り除くメソッド
+
+	* array.c (ary_nitems): nilでない要素を数えるメソッド
+
+Sun Jul 20 00:51:53 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.c (proc_options): -w optionを追加
+
+	* parse.y: {}が閉じていない時には展開しない文字列を
+
+Fri Jul 19 16:16:05 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99-960719
+
+	* lib/find.rb: 石塚版(pruneの拡張付き)
+
+	* file.c (test_l): lstatで調べないとね．
+
+	* eval.c (f_throw): 第2引数を省略可能に．
+
+	* parse.y (str_extend): {}のネストに対応
+
+Thu Jul 18 18:25:46 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99-960718
+
+	* parse.y (str_extend): 文字列中の式展開に \" ' ` / を含む事ができ
+	  るように．
+
+Tue Jul 16 15:55:31 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-parse-region): 正規表現内のエスケープ
+	  に対応
+
+	* version 0.99-960716
+
+Fri Jul 12 10:06:19 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (f_select): 引数のclose check.
+
+	* ruby.c (load_file): #!行の引数チェックを第1引数に限定(実をいうと
+	  DOS改行対策)
+
+Wed Jul 10 17:18:35 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99-960710
+
+	* time.c (time_s_timegm/time_s_timelocal): 時間を生成するメソッド
+
+Mon Jun 17 15:59:20 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99-960617
+
+	* parse.y (yyerror): エラー表示の簡略化．
+
+Wed Jun 12 14:11:01 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* signal.c (rb_trap_exit): trap 0はthreadを生成せずに処理する．
+
+Fri Jun  7 10:17:01 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* array.c/hash.c (indexes): 配列1引数のパターンを無くした．配列の
+	  場合は`*ary'を使ってもらおう．
+
+	* eval.c (thread_wait_threads): main_threadが終了する前に他の
+	  threadを待つ(強制的には終了させない)．
+	  (ruby_run): 他のthreadを待っている間にシグナルが来たら，全thread
+	  を強制終了させる．
+
+	* eval.c (rb_fail): メソッド名を`$!'に埋め込む．
+
+	* eval.c (thread_create): main_threadのコンテクストがセーブされな
+	  い場合があった．
+
+	* process.c (f_sleep): 時間を指定せず，threadがひとつしかない状況
+	  にも対応．
+
+	* eval.c (thread_create): create後，fnを呼び出す前にcontext switch
+	  が起きると違うcontextでfnが実行されてしまうバグ．
+
+Mon Jun  3 08:03:17 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* struct.c (struct_s_def): メンバの指定を文字列，シンボル(FIXNUM)
+	  双方で可能にした．
+
+	* ext/etc/etc.c (Init_etc): 構造体オブジェクトをGCから保護した．
+
+	* error.c (rb_sys_fail): nil/FALSEを引数として受け付けるように．
+
+Thu May 30 16:19:08 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_select): EINTRに対応．
+
+Wed May 29 11:04:51 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (f_catch): catch/throwを実装した．
+
+Tue May 28 13:30:52 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99-960528
+
+	* eval.c (thread_cleanup): main threadが終了すると他のthreadも終了
+	  することの明確化．
+
+	* signal.c (trap): SIGINTのデフォルトの設定ミス(本当にSIG_DFLでは
+	  まずかった)．rubyではちゃんとハンドルしないと．
+
+	* eval.c (thread_interrupt): SIGINTはmain_threadに例外を発生させる
+	  ように．
+
+Mon May 27 15:13:31 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_status): threadの状態を返すメソッド．threadの終了
+	  を待たない．
+
+	* eval.c (thread_value): 一種のpromiseを実装するためのメソッド．
+
+	* eval.c (thread_join): 待っているthreadが例外を起こした時には，
+	  joinがその例外を発生するように．
+
+	* eval.c (thread_create): threadでの例外をpropagateしないように．
+
+Fri May 24 10:47:53 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* enum.c (Init_Enumerable): `size' as alias to the `length'
+
+	* eval.c (thread_save_context): `$@', `$!'をスレッド毎にセーブ．
+
+	* eval.c (superclass): エラー表示をより親切に．
+
+Thu May 23 10:38:41 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.99-960523
+
+	* eval.c (superclass): エラー時にスーパークラス名を(分かれば)表示
+	  するように．
+
+Wed May 22 19:48:42 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (superclass): スーパークラスの指定子を`:'から`<'に変更．
+
+Tue May 21 09:27:59 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* lib/thread.rb: threadをサポートするクラス(Mutex, Queue)．
+
+Mon May 20 09:39:49 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* time.c (time_cmp): 浮動小数点数も扱えるように．
+	(time_minus): Time - Timeが浮動小数点数を返すように．
+
+Fri May 17 15:40:10 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* process.c (rb_proc_exec): Thread対応時にexecの直前に
+	  ITIMER_VIRTUALをリセットする．
+
+Tue May 14 02:12:44 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* signal.c (sighandle): SIGINTに対してデフォルトで例外を発生させる
+	  のをやめ，status 130でexitするようにした．
+
+	* eval.c (thread_schedule): Threadのバグはほとんどとれたようだ．
+
+Fri May 10 11:21:08 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (thread_schedule): ユーザレベルThread機能．効率はともかく
+	  移植性はある．今後，thread間の通信機能を実装する予定．
+
+Thu May  2 21:22:31 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* time.c (time_timeval): struct timevalを直接返すように(static変数
+	  を使わない)．
+
+Wed May  1 17:27:32 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* process.c (f_sleep): 整数以外のtimeを指定できるように．
+
+Thu Apr 25 08:19:15 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (file_s_dirname): ファイル名が"/"を含まない時，"."を返す
+	  ように(GNU dirnameの仕様)．
+
+	* file.c (file_s_basename): まだnilと0を混同しているソースが残って
+	  いた．
+
+	* parse.y (exprs): エラーリカバリを追加．
+
+Wed Apr 24 15:51:05 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_chop_bang): CRLFの場合2 bytesをchop!するように．
+
+	* ext/socket/socket.c (tcp_svr_s_open): まだnilと0を混同しているソー
+	  スが残っていた．
+
+Tue Apr 23 18:14:25 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* pack.c (pack_pack): "A/a"のバグ．余計なpaddingが入っていた．
+
+Thu Apr 18 13:02:11 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* configure.in: アーキテクチャ依存部を別ディレクトリにインストール
+	  するように．
+
+	* parse.y (yyerror): エラー発生時にエラー行とその位置を表示するよ
+	  うに．
+
+Wed Apr 17 14:22:42 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* defines.h: SAFE_SIGHANDLEを無くし，危険な選択はできないように．
+
+	* io.c (io_ungetc): 新機能．
+
+	* ruby.c (load_file): ファイルからの読み込み方式が変わったのに対応．
+
+	* parse.y (compile_file): ファイルからの入力を一度全部読み込むのを
+	  止めて，getsを使うことにした．
+
+Wed Apr 10 17:40:11 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.98
+
+Tue Apr  9 09:54:30 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (iter_block): イテレータブロックの指定をメソッド呼び出し
+	  に限定．文法の明確化．
+
+	* eval.c (rb_eval): 条件式の正規表現の比較をinline化．
+
+	* eval.c (rb_eval): defined? の 定義情報(種別)を文字列で返す．
+
+	* node.h: NODE_BEGIN -> NODE_RESCUE, NODE_ENSUREに分離．
+
+	* eval.c (rb_eval): option -n/-pのトップレベルループのinline展開．
+
+	* parse.y (cond0): 条件式中の文字列は比較の対象としない
+
+Wed Mar 27 12:33:54 1996  Tairo Nomura  <tairo@hucom.tp.titech.ac.jp>
+
+	* defines.h: NeXT対応
+
+Wed Mar 27 10:02:44 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y: 予約語の変更 continue -> next
+
+Mon Mar 25 07:34:37 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (parse_regx): o(once)オプションを追加．
+
+Fri Mar 22 14:25:35 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.97d
+
+	* eval.c (dyna_var_defined): 動的ローカル変数の定義チェック用ルー
+	  チン．
+
+	* parse.y (gettable): eval()の中での動的ローカル変数(既に値を持っ
+	  ているもの)の検出に失敗していた．
+
+Tue Mar 19 10:46:47 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.97c
+
+	* re.c (reg_s_new): compile時にsegmentation fault．
+
+	* parse.y (str_extend): いつもevalするように．
+
+Wed Mar 13 11:00:42 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (str_extend): 文字列中の式展開の不備を無くした．
+
+	* parse.y: 下手なエラーリカバリを外した．
+
+Tue Mar 12 12:30:20 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rescue): 間違ってensureでも例外を捕捉していた．
+
+Wed Mar  6 12:11:03 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (var_extend): 変数展開"#{}"で，任意の式を書けるようにし
+	  た，これで「変数」展開では無くなっちゃったなあ．
+
+	* regex.c (init_syntax_once): `_'をwordに追加．
+
+	* regex.c (re_compile_pattern): `\w',`\W'の判定をsyntax tableを使
+	  うように．
+
+Tue Feb 27 10:15:32 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (obj_inspect): 表示するインスタンス変数が無い時には，
+	  to_sを使う．
+
+	* configure.in: dlnの検出を自動的に．
+
+Mon Feb 26 19:55:33 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.c (readin): read(2)で一度にファイルが読み込めない場合に対応．
+
+Sat Feb 24 14:47:18 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.97b
+
+Fri Feb 23 11:26:02 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* class.c (rb_define_module): C言語で定義されたモジュールのPATHの
+	  設定忘れ．文字列化でcore dump．
+
+	* eval.c (mod_include): 戻り値をnilに．
+
+	* version 0.97a
+
+Thu Feb 22 21:03:42 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* array.c (ary_times): 「配列*文字列」がjoinと同じ働きをするように．
+
+Wed Feb 21 11:18:09 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* configure.in : fileCountをcache．
+
+	* configure.in : LinuxでELF環境を自動的に検出できるよう．
+
+Tue Feb 20 11:18:09 1996  Mitsuhide Satou  <mit-sato@aries.bekkoame.or.jp>
+
+	* FreeBSD dynamic link対応．
+
+Fri Feb 16 08:50:01 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (obj_inspect): インスタンス変数を持たないオブジェクトも
+	  正しく表示されるように．
+
+Wed Feb 14 16:56:44 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): 条件式の`2..2'など左辺成立直後に右辺が成立する
+	  パターンにバグ．
+
+Tue Feb 13 18:22:22 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.97
+
+Fri Feb  9 21:32:55 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* lib/tkscrollbox.rb: スクロールでtclの設定を行い，ruby<->wishの不
+	  要な通信を無くした．
+
+Wed Feb  7 10:26:52 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_aref): indexをunsigned intでとっていた．
+
+	* string.c (str_aref): 範囲外のindexに対してnilを返す．
+
+	* parse.y (special_local_set): `$_'が宣言無しに使われた場合に対応．
+	  関数をvariable.cから移動．
+
+	* string.c (str_sub): 置換開始位置が間違っていた．
+
+Tue Feb  6 16:17:31 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-parse-region): コメントの読み飛ばしの
+	  バグ．
+
+Fri Feb  2 18:35:28 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* variable.c (lastline_get): `$_'を`$~'と同じようにSCOPEローカルな
+	  変数にした．
+
+Thu Feb  1 14:14:07 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c: statのcacheをやめた．
+
+Wed Jan 31 07:13:08 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (proc_s_new): procの中でyieldを呼ばれた時にcore dumpして
+	  いた．とりあえず例外を発生させる．
+
+	* variable.c (rb_class2path): singleton classに対応．
+
+	* ext/etc/etc.c (Init_etc): struct_defineのターミネータがnilだった
+	  (0でなければならない)．
+
+	* ext/marshal/marshal.c: TRUE/FALSEを吐き出せるように．
+
+	* eval.c (rb_get_method_body): キャッシュのalias対応，いままでは
+	  aliasはキャッシュに入っていなかった．
+
+Tue Jan 30 09:55:13 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): NODE_BLOCK - tail recursive(というほどでもない
+	  が)．
+
+	* io.c (io_pipe): pipe(2)を実装した．
+
+	* eval.c (rb_eval): Qselfをなくした．thread対応への第一歩．先は遠
+	  いが…．
+
+	* eval.c (proc_call): procの中でのreturnはprocの終了を意味するよう
+	  に．ただし，procからのyieldの中でのreturnは例外を発生する．
+
+Wed Jan 24 11:33:48 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.96a
+
+	* dir.c (dir_each): `$_'の値を変更するのをやめた．
+
+	* io.c (f_readlines): nilとFALSEの分離のあおりで無限ループに落ちて
+	  いた．
+
+	* ruby.c (ruby_options): $0の設定ミス．
+
+Tue Jan 23 15:28:21 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): ``は文字列を引数とするメソッド(`)呼び出しのシ
+	  ンタックスシュガーであるとした．
+
+	* ruby.c (addpath): `-I'オプションでディレクトリが「前に」追加され
+	  るように変更．
+
+Fri Jan 19 11:23:12 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* dln.c (load_1): N_INDR対応(出来たような気がする)．
+
+Thu Jan 18 18:14:20 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.texi: FALSEとnilの分離を反映した．
+
+Tue Jan 16 17:39:23 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.96 - とりあえずnilとFALSEを区別する版
+
+Wed Jan 10 15:31:48 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* re.c (reg_match): マッチしなかった時の戻り値はFALSE．
+
+	* object.c (rb_equal): `0 == nil'がTRUEになるバグ．
+
+Tue Jan  9 00:44:58 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* nilとFALSEが分離可能に変更．
+
+	* nilとFALSEと0の区別を厳密に．
+
+	* struct.c (struct_new): 引数を0で終る必要が無くなった．
+
+	* object.c (inspect_i): オブジェクトのチェックのバグ(Fixnumでcore
+	  dumpしていた)．
+
+	* range.c (range_to_s): Rangeの表示を改善．
+
+	* object.c (true_inspect): TRUEの表示を`TRUE'に．
+
+Mon Jan  8 15:02:33 1996  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* numeric.c (fix_mul): divide by zero errorが発生した(オーバーフロー
+	  検出のバグ)
+
+	* texinfo.texをパッケージに含めた．
+
+Sun Dec 31 00:08:49 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): `::'では，そのクラスで定義された定数を参照する
+	  ように変更．
+
+	* string.c (Init_String): eachをeach_lineに戻した．
+
+Thu Dec 28 12:31:55 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): caseの演算子を`=~'から`==='に．
+
+	* variable.c (rb_const_set): クラス定数の再定義を許す(同じクラスで
+	  は不可)．警告は出す．
+
+Wed Dec 27 13:27:52 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.95c
+
+	* ext/tkutil/tkutil.c: wishがあってもなくても一応コンパイルだけは
+	  するように．
+
+	* lib/tk.rb: 環境変数PATHから{wish|wish4.0}を探すように．
+
+Tue Dec 26 01:03:42 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-parse-region): 正規表現の検出強化．
+
+	* numeric.c (fix_mul): 乗算のオーバーフロー検出アルゴリズムのバグ．
+
+	* ext/extmk.rb.in: ./install-shを使う場合のPATHを調整．
+
+	* Makefile.in (install): lib/*.rbを一つずつインストール．
+
+	* io.c (io_each_line): イテレータの戻り値をnilで統一．
+
+Fri Dec 22 10:34:32 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.95b
+
+	* variable.c (f_untrace_var): 第2引数を指定すると特定のtraceを削除
+	  できるように．
+
+	* variable.c (f_trace_var): 第2引数がnilの時，traceを削除する．
+
+	* lib/tk.rb (file_readable/file_writable): 第2引数をnilにすること
+	  によるevent handlerの削除．
+
+	* parse.y (variable): ドキュメントに`__FILE__'と`__LINE__'が残って
+	  いた．`caller(0)'で代用したはずだったのに．
+
+	* eval.c (f_eval): $!のリセット．
+
+	* error.c (err_sprintf): 勝手に"\n"を付加するのを止めた．
+
+	* parse.y (f_arglist): 引数リスト直後のif/whileの読み間違い．
+	  lex_stateの値が設定されていなかった．
+
+Thu Dec 21 00:56:57 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.95a - ^^;;;
+
+	* lib/tkscrollbox.rb: パッケージに入ってなかった．
+
+	* configure.in: FILE structureのチェックにバグ．
+
+	* Makefile.in (clean): ext以下をinstallしていた．
+
+	* ext/socket/extconf.rb: Solarisにおける-lnlsのチェック．
+
+	* array.c (beg_len): バグがあった…．悲しい．
+
+	* version 0.95 - fj.sourcesに
+
+	* eval.c (rb_eval): rescueのロジックをrb_rescue()に一元化．
+
+Wed Dec 20 19:30:58 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* Makefile.in: 不要なコンパイルの回避(より完全に)．
+
+	* class.c (singleton_class_new): `single'->`singleton'
+
+Tue Dec 19 07:14:33 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* io.c (closed?): IOがcloseされているかどうかを知る述語．
+
+	* parse.y (primary): 特異メソッドの引数のlex_stateが不適切．
+
+	* lib/tk.rb: tcl->rubyの変換関数の用意．
+
+	* ext/extmk.rb.in (install): installの2重コンパイルの回避．
+
+	* array.c (range_beg_len): range指定の不適切なエラーを訂正．
+
+	* string.c (str_aref): range指定のバグを削除．
+
+	* lib/tk.rb (tk_split_list): Tclのリストに対応．
+
+Mon Dec 18 09:58:12 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.94
+
+	* dln.c (dln_load): HP対応(未確認)
+
+	* eval.c (Init_Proc): BlockをProcに改名．
+
+Sat Dec 16 13:46:14 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): retryでイテレータの再実行ができるように．
+
+Fri Dec 15 17:14:30 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c: proc:lambdaの親しみやすい別名
+
+Thu Dec 14 17:21:55 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (dyna_var_asgn): イテレータブロック内で最初に初期化された
+	  ローカル変数の有効範囲をそのブロック内に限定．これでlambdaと呼べ
+	  ないことはない．
+
+Wed Dec 13 02:30:58 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* gc.c: autoloadのマークミス．
+
+	* lib/tk.rb: wishからの複数行の戻り値に対応
+
+	* lib/tkcomposite.rb: 複合widget
+
+	* variable.c (rb_class2path): ICLASSに対応してなかった．
+
+	* eval.c (ruby_run): exit(0)のバグ
+
+Sat Dec  9 01:21:24 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/marshal/marshal.c (dumps|load): 文字列に対する入出力を可能に
+	  した(ただし実はファイル経由なのだ)．
+
+Fri Dec  8 18:29:11 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/marshal/marshal.c: シンボルを一度だけ初期化する．
+
+Thu Dec  7 07:58:50 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (yylex): 第1引数の正規表現の認識にエラーがあった．同時に
+	  状態数を減らした．
+
+	* string.c (str_sub): 置換でスキップ幅が大きすぎた．
+
+Wed Dec  6 15:14:23 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_sub_method): sub/gsub(!なし)は置換が行なわれなかっ
+	  た時，置換前の文字列を返す．
+
+Tue Dec  5 00:55:15 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (yylex): 括弧を省略した時の引数展開の`*'に対応．
+
+	* eval.c (ruby_run): EXITハンドラ内での例外に対応．
+
+	* bignum.c (big_cmp): BignumとFixnumの比較で落ちる．
+
+Mon Dec  4 14:21:18 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (call_op): コンパイル時の定数式の展開をやめた．労多くし
+	  て益少ないと判断したので．
+
+Thu Nov 30 01:35:15 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* lib/tk.rb: {Radio,Check}Buttonのvariableの実装．
+
+	* eval.c (rb_yield_0): Block.callがネストした時のバグ．
+
+	* io.c (f_select): 常に配列3つをふくむ配列を返すように
+
+	* lib/tk.rb: fileeventをruby側で実装．
+
+Wed Nov 29 17:53:23 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* variable.c (rb_ivar_get): selfを常に指定するように．
+
+Tue Nov 14 00:07:29 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* lib/tk.rb: Tk4.0対応
+
+Mon Nov 13 16:23:32 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.93
+
+Thu Nov  9 23:26:01 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* gc.c (gc_mark): モジュールのMixinのマーク忘れ．
+
+	* parse.y (f_arglist): メソッド定義の引数を括弧で括らなくても良い
+	  ようにした．
+
+Wed Nov  8 00:17:51 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_yield_0): 未初期化のローカル変数があった．
+
+	* eval.c (rb_eval): pendig signalのチェックをeval実行後に行うよう
+	  にした．でないとシグナルの発生と検出が遠く離れてしまう事がある．
+
+	* parse.y: class文のsuperclass部を定数から式に拡張した．
+
+	* lib/tk.rb: Tkのほぼ全ウィンドウクラスに対応．キャンバスとテキス
+	  ト上のオブジェクトが残っている．
+
+Tue Nov  7 08:18:37 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* signal.c (trap): ブロックを指定できるように．
+
+Mon Nov  6 16:44:00 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (f_caller): 呼出元の情報を得る．
+
+	* ext/tkutil/tkutil.c: wishのstderr出力を監視することで，エラー処
+	  理を行う．
+
+	* ext/tkutil/tkutil.c: wishとの通信部をCで記述．
+
+Sat Nov  4 01:12:59 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el (ruby-calculate-indent): インデントの計算を
+	  もう少しスマートにした(正規表現のチェック，継続行のチェック)．
+
+	* eval.c (rb_call): 無限再帰を避けるため，関数のネストレベルの制限
+	  を行なう．
+
+	* lib/tk.rb: Tkインターフェース．まだ不完全だが．
+
+	* eval.c (rb_yield_0): 空のBlockのバグ．
+
+	* sample/ruby-mode.el (ruby-calculate-indent): 行末の演算子による
+	  行継続に対応．
+
+Fri Nov  3 12:56:21 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_call): 本体が空の関数の実行にバグ．
+
+	* parse.y (var_extend): 文字列の末尾の変数展開のバグ．
+
+	* variable.c (rb_gvar_set): traceの評価時にに変数値を与えるように．
+
+	* eval.c (f_require): ruby scriptのrequireにbug．
+
+	* variable.c (rb_const_get): モジュールのinclude対策．
+
+Thu Oct 19 13:56:06 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* dln.c (dln_load): HP対応でのtypo.
+
+Wed Oct 18 17:39:39 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.92
+
+	* object.c (krn_type): オブジェクトの動的な型を返すメソッド．
+
+Tue Oct 17 00:48:18 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.c (proc_options): -X オプション．chdirだけを行う．
+
+	* re.c (reg_search): 漢字コードを途中で変更できるように．コンパイ
+	  ル時のコードが変更された時にはマッチの直前に正規表現の再コンパイ
+	  ルを行う．定数KCODEから変数$KCODEへ．
+
+	* parse.y: ()のなかにcompexprを許す．
+
+	* re.c (reg_search): メモリリークを直した．
+
+Fri Oct 13 13:19:19 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (str_sub): 文字列置換にバグ．
+
+	* string.c (str_strip_bang): 文字列の後ろの長さの調整が行われてい
+	  なかった．
+
+	* re.c (reg_search): $&, $1...はローカルに束縛するようになった．呼
+	  び出したメソッドでのマッチは現スコープの$&などの値に影響しない．
+	  マッチの情報をスコープ外で得たいときには$~を使って束縛情報を持ち
+	  出す必要がある．
+
+Thu Oct 12 00:33:33 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* re.c (reg_search): String:split, String:indexでは$&, $1...が変化
+	  しないようにした．
+
+	* io.c (rb_str_setter): setterの仕様が変更になっていた．
+
+	* variable.c (f_trace_var): 第2引数を省略してイテレータとして呼べ
+	  るように．
+
+Wed Oct 11 11:50:59 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.91
+
+	* variable.c (var_setter): 引数が間違っていた．致命的バグ．
+
+	* io.c (pipe_open): $stderrの値が変更されている時にはそちらを
+	  子プロセスのstderrに設定する．
+
+Mon Oct  9 13:06:33 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (mod_to_s): モジュール内のモジュールは`::'を使った表現
+	  で表示されるように．
+
+	* variable.c (rb_gvar_set): 代入によるループが発生しないように，
+	  trace内での代入ではtraceを評価しない．
+
+	* struct.c (struct_equal): structのequal判定にクラスの一致を含めた．
+
+Sat Oct  7 00:18:32 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_eval): defined?の機能を拡張(yieldのチェック，superの
+	  存在など)．
+
+Fri Oct  6 12:06:47 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.90
+
+	* st.c (st_foreach): 要素を削除した時に要素数が変化していなかった．
+
+	* hash.c (hash_values): バグ修正．keysを返していた…．
+
+	* parse.y (call_op): defined? の引数では定数の畳み込みを行わない
+	  (チェックする前にコンパイルエラーになっては困る)．
+
+	* スコープ生成の一部見直し．
+
+Thu Oct  5 00:29:43 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* 関数とクラスの命名規則を変更した．関数名，変数名の全面書き換え．
+
+	* gc.c (looks_pointerp): ヒープチェックの高速化．
+
+	* struct.c (Fstruct_aset): 構造体に対する`[]='．
+	(struct_set): 構造体メンバに対する代入．
+
+Wed Oct  4 09:54:07 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.89
+
+	* eval.c (Frequire): ダイナミックロードのエラーチェックを厳しく．
+
+	* struct.c: structの構造を完全に書き換えた．以前は順序付きの
+	  id->valueの連想配列であったが，今度は構造体毎に新しいクラスを生
+	  成するようにした．
+
+	* parse.y: `::'の意味をAssocの生成からクラス(モジュール)内の定数ア
+	  クセスへ変更．
+
+	* assoc.c: なくす．
+
+Tue Oct  3 13:31:08 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* variable.c (Ftrace_var): trace_var, 大域変数への書き込みhookを設
+	  定する．
+
+	* variable.c: global_entryの構成を書き換えた．これでtrace_varを実
+	  装できる．
+
+	* file.c (Ffile_stat): "&"で直前のfstatの結果も参照できるように．
+
+Fri Sep 29 14:15:13 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.88
+
+	* dln.c (dln_load): AIXとHPに対応したコードを入れた(動作は未確認)．
+
+	* ext/extmk.rb.in: 必要に応じて，定数EXTLIBを定義するように．
+
+	* dln.c (dln_load): dln独立に書き換える．将来の拡張用．
+	(load_1): dln_a_outにおいてソースコードでライブラリを明示的にロー
+	ドする必要がないように変更した．
+
+Thu Sep 28 13:31:37 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/ruby-mode.el: もっとましなhilit19対応(正規表現)．
+
+Wed Sep 27 04:12:44 1995  Takahasi Mamoru  <taka@soum.co.jp>
+
+	* sample/test.rb: echoで-nを使わないように(SysV対策)．
+
+	* ext/extmk.rb.in: sub -> sub!
+
+Tue Sep 26 19:12:42 1995  Yasuo OHBA  <jammy@csg.mes.co.jp>
+
+	* dln.c (dln_find_1): `.', `..'から始まるパスに対応した.
+
+Mon Sep 25 12:33:03 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.87
+
+Sat Sep 23 10:00:18 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (Fmod_modfunc): メソッドをprivateにし，同時に特異メソッド
+	  も定義するメソッド．パッケージ的使い方のモジュール用．
+
+Fri Sep 22 11:02:44 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* lib/find.rb: findを提供するライブラリ
+
+	* variable.c (rb_define_variable): hookの設定を分離．
+	(add_hook): 1変数に対して複数のhookを設定できるように．
+
+Thu Sep 21 00:22:11 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (Fstr_frozen): 文字列が更新不可かどうかをチェックする述
+	  語メソッド．
+
+	* hash.c (Fhash_aset): keyが文字列の時，キーの内容が変化しないよう
+	  に，dupしてfreezeする．
+
+Wed Sep 20 16:12:44 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.86
+
+	* ext/extmk.rb.in (have_header): キャッシュにバグ．
+
+	* ext/extmk.rb.in (have_library): 引数の順序が変わった．
+
+Thu Sep 14 18:00:59 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* object.c (obj_is_instance_of): is_member_ofから名称変更．
+
+	Wed Sep 13 15:44:35 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (Fstr_tr_bang): 範囲外の文字に対する変換バグ．
+
+Tue Sep 12 14:27:58 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (Sfile_expand_path): expand_file_name -> expand_pathに名
+	  称変更．
+
+	* enum.c (Fenum_member): includes? -> member? に名称変更．
+
+	* string.c (Fstr_each_byte): StringはByteArrayであるという基本に戻っ
+	  て，eachの定義をeach_byteに変更した．今までのeachはeach_lineでア
+	  クセスできる．
+
+Mon Sep 11 18:31:17 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (cache_stat): ファイル名として"&"を指定すると直前の
+	  stat(2)の結果を再利用するように．
+
+Fri Sep  8 14:18:51 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ruby.texi: `!', `?'に対応してアップデート．
+
+	* parse.y: defined -> defined?
+
+	* file.c: FileOpの一文字メソッドをなくす．一文字テストはtestメソッ
+	  ドにまかせる．
+
+	* parse.y (yylex): 変数名の後ろに`?'も許す．述語メソッドの後ろに
+	  `?'を追加する．
+
+Thu Sep  7 20:01:33 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c: 文字列の中身を更新するメソッドの名前の終りに`!'を付加．
+	  `!'の無いバージョンも用意した．
+
+	* parse.y: 変数名の後ろに`!'を許す．
+
+Wed Sep  6 14:12:19 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.85
+
+	* string.c (Fstr_dup): 文字列の複製を作る
+	(Fstr_freeze): 文字列の更新不可属性を設定できるように．
+	(Fsub/Fgsub): $_の内容をdupしてから置換を行うように．
+
+	* ruby.h (CLONESETUP): flagsの状態もコピー
+
+Tue Sep  5 01:27:50 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* sample/test.rb: 失敗の検出を厳しく．
+
+Fri Aug 25 14:31:02 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* process.c (Ffork): イテレータとしても動作するように．
+
+	* version 0.84
+
+	* signal.c (sig_beg): ハンドラが設定されている時には再設定しない．
+
+	* ext/extmk.rb.in (create_makefile): shared objectのリンクの際に
+	  `-l'オプションを指定するように．
+
+	* signal.c (trap): `EXIT'で終了処理を行う設定が出来る．
+
+Wed Aug 16 00:13:22 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* signal.c (sig_beg): デフォルトではbegin節の中でだけSIGINTを捕捉
+	  するように変更．
+
+	* io.c (io_ctl): fcntlを持たないシステムにも対応．
+
+	* 各ディレクトリに分散していたMANIFESTをまとめた．拡張モジュール毎
+	  には必要．
+
+	* string.c (Sstr_new,str_sub,Fstr_crypt): 引数を自動的に文字列に変
+	  換するように．
+
+Sat Aug 12 00:44:02 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* string.c (Fstr_crypt): PD cryptを用意した．
+
+Fri Aug 11 14:37:03 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* assoc.c (Fassoc_clone): assocもcloneできるように．
+
+	* io.c: マクロREAD_DATA_PENDINGの定義を変更(Linux対応)
+
+	* io.c (io_fptr_finalize): fptrの開放時の処理を指定できるように．
+
+Wed Aug  9 16:52:41 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* eval.c (rb_provided): 複数のfeatureをロードすると無限ループに落
+	  ちるという単純な(しかし凶悪な)ミス．
+
+	* ext/extmk.rb.in (install): dlopen対応を行った．今までdlnにしか十
+	  分に対応していなかった．
+
+Tue Aug  8 14:17:06 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.83
+
+Mon Aug  7 12:47:41 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y: resque -> rescue．恥ずかしいがtypoを残しておくわけには
+	  いかないよなあ．なんで今まで気がつかなかったのか…．
+
+Thu Aug  3 18:18:05 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* missing/nt.c: NT移植用の関数群をまとめた．
+
+	* variable.c (rb_const_get): また例外を発生するようにした．defined
+	  がある以上例外を発生させない理由がないので(例外が発生した方がタ
+	  イプミスの検出などの点で有利)．
+
+	* variable.c (Fautoload): autoloadを実装．今度は使えるか．
+
+Mon Jul 31 15:44:21 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (arg_ambiguous): 第1引数のあいまいさを警告(-vオプション
+	  で有効)．
+
+	* eval.c (rb_eval): `-v'オプションをつけて`def'が呼ばれると不必要
+	  なエラーメッセージが出た．
+
+	* parse.y (yylex): メソッドの第1引数の判定をもうちょっと賢くした．
+
+Fri Jul 28 19:04:43 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (yylex): `+/-/['の直前に空白が来るかどうかで動作を変更し
+	  た(混乱のもとか?)
+
+Wed Jul 26 09:21:23 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.82a
+
+	* sprintf.c (Fsprintf): `%s'で'\0'を含む文字列に対応．
+
+	* pack.c (Fpck_pack): packの要素確保のバグ．
+
+	* eval.c (Floop): 無限ループのイテレータ．
+
+	* io.c (next_argv): 存在しないファイル名が指定された時のエラー処理
+	  が行われていなかった．
+
+Mon Jul 24 17:37:34 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.82
+
+	* ext/extmk.rb.in (install): 拡張モジュールをstatic linkする場合は
+	  そのモジュールが既にrequireされたのと同じようにfeatureを設定する．
+	  これで拡張モジュールの機能が必要な時には(static linkされているか
+	  どうかにかかわらず)requireすればよくなる．
+
+	* eval.c (Frequire): `$"'に格納する文字列をフルパスでなくフィーチャ
+	  名とする．rubyスクリプトをロードした時には`.rb'，オブジェクトを
+	  ロードした時には`.o'をフィーチャ名に付加する．lispのrequireと
+	  provideの働きに(少し)近い．
+
+Thu Jul 20 12:50:05 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* Makefile.in (test): make testができるように．
+
+	* struct.c (struct_new): typo.
+
+	* eval.c (rb_eval): `defined'を追加．メソッド/変数/定数の定義状態
+	  を知る事が出来る．
+
+Wed Jul 19 18:04:01 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.81
+
+Mon Jul 17 14:53:51 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* variable.c (rb_const_get): 未初期化のCONSTANTの値をnilにした．し
+	  かし，今後また例外に戻す可能性はある．要はoptionalなクラス/モジュー
+	  ルが存在するかチェックしたいだけなんだな．
+
+	* st.c (int): grow_factorを固定にした(大嶋さんのマシンに対応)．
+
+Fri Jul 14 00:48:40 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* ext/extmk.rb.in: キャッシュのバグを修正．
+
+	* parse.y (var_extend): #{$数字}に対応した．
+
+	* dln.c (dln_load_1): `Init_FILENAME'だけを有効に．`init_*'は今後
+	  実行しない．
+
+	* ext/etc/etc.c : Etcモジュールを拡張モジュールとして分離．実はNT
+	  対応への布石だったりするかもしれない．
+
+Tue Jul 11 17:12:48 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* gcc -Wallで出たwarningを元にソースを変更．
+
+	* signal.c (trap): typo.
+
+Fri Jul  7 10:08:51 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.80
+
+	* ruby.texi: texinfo documentを提供．specとruby.1は無くなった．
+
+	* signal.c (Ftrap): 割込み禁止中の例外発生に対応．
+
+	* eval.c (Flambda): Blockオブジェクトを返す．Block.newと同義．
+
+Thu Jul  6 00:35:03 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* signal.c (Ftrap): SIG_DFLの処理を変更．SIGINTへのデフォルトハン
+	ドラを用意(例外を発生する)．
+
+	* file.c (Sfile_expand_fname): パス名を絶対パスに展開するメソッド．
+	(Sfile_basename): basenameを得るメソッド．拡張子も外せる．
+	(Sfile_dirname): basenameの反対．
+
+	* eval.c (rb_call): argument評価中の例外発生に対応．
+
+	* file.c (Ftest): `M', `A', `C'を追加.
+
+Tue Jul  4 12:36:33 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* file.c (Ftest): ファイルテスト用メソッド．
+
+	* ruby.c (proc_options): `-r'オプションを追加．
+
+	* parse.y (f_args): デフォルト引数を追加．
+
+	* eval.c (rb_call): 該当する引数が無い時，rest引数の値をnilに．
+
+	* numeric.c (num_equal): 数値以外との比較で例外が発生していた．
+	  FALSEを返すように．
+
+	* eval.c (masign): 多重代入のrest部の動作がおかしかった．
+
+Sat Jun 17 01:03:16 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* parse.y (gettable): 未初期化のローカル変数の参照(独立した識別子)
+	  は正式にメソッド呼び出しとした．
+
+	* parse.y (read_escape): tokenbufを使わないように修正．それにとも
+	  ない，`\C-x',`\M-x'などのエスケープ表現を復活．これでドキュメン
+	  トと実際の処理系が一致した．
+
+Thu Jun 15 15:42:00 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* re.c (re_regcomp): cacheのチェックを改善．
+
+Mon Jun 12 18:50:51 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* version 0.79
+
+Sat Jun 10 00:25:01 1995  Yukihiro Matsumoto  <matz@caelum.co.jp>
+
+	* re.c (re_regcomp): cache判定に`$='の値も反映させた．
+
+	* sample/test.rb: test suite作成．
+
+Fri Jun  9 15:58:34 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* re.c (re_regcomp): cacheの判定が間違っていた．
+
+Fri Jun  9 00:01:35 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c (rb_yield): block構造体に初期化していないメンバ(iter)があっ
+	  たのでイテレータのネストが正しく動作しなかった．
+
+Thu Jun  8 00:59:03 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* re.c (=~): String以外との比較がFALSEを返すように(例外を発生して
+	  いた)．
+
+	* extmk.rb.in: 判定した値をファイルにキャッシュするようにした．
+
+	* assoc.c (to_a): to_aメソッドが再定義されていなかった．
+
+	* eval.c (rb_eval): 初期化されていないローカル変数へのアクセスを引
+	  数の無いメソッド呼び出しと解釈する．ただし，(現状では)メソッドが
+	  定義されていない場合，エラーにせず変数未初期化のwaringを出して
+	  nilを返している．「ruby -pe print」などが実行できるという意味で
+	  はありがたいこの仕様は，しかし今後の検討が必要である．-- メソッ
+	  ド呼び出しとするのを止めるか(以前の仕様)，いつもメソッド呼び出し
+	  とする(未定義ならばエラー)か，今の仕様で行くか．
+
+	* eval.c (rb_eval): 初期化されていないローカル変数へのアクセスで
+	  (evalなどで)初期化された事が分かった時には以後初期化されたローカ
+	  ル変数とみなす．
+
+Wed Jun  7 11:58:12 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* eval.c (rb_fail): 例外処理後も`$!'をクリアしないように．
+	  (rb_fail): `$!'変数に最後に改行を追加しない．
+
+	* io.c (Fprint): privateメソッドに変更．引数を取らない時の動作を変
+	  更(`$_'を出力する)．
+	  (Fio_print): 出力先指定のprintメソッド．
+	  (Fio_printf): 出力先指定のprintfメソッド．
+
+	* parse.y: not演算子の追加．優先順位の低い`!'演算子．
+
+Mon Jun  5 19:00:55 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* version 0.78
+
+Fri Jun  2 17:52:03 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* ruby.c (proc_options): -Iオプションで`$:'への追加される順番を修
+	  正した．
+
+Fri Jun  2 00:36:34 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y: while修飾子の動作を通常のwhileと同じにした．ただし，
+	  begin式へのwhile修飾子だけはdo..while型のループとなる．
+
+Wed May 31 18:36:30 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* version 0.77
+
+Mon May 29 18:39:37 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* ext/extmk.rb.in (install): 拡張モジュールもインストールできるよ
+	  うに．
+
+Fri May 26 14:43:01 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* process.c (Fsystem): 戻り値をサブプロセスの失敗/成功を表す真偽値
+	  にした．終了ステータスは`$?'で得る．
+
+Tue May 23 10:58:11 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* string.c (Fstr_upto): 無限ループに陥らないように．
+
+	* parse.y (cond): `||'などの右辺に制御式が書けるように，条件式がか
+	  ならずしも値を持たなくても良いようにした．
+
+	* ext/marshal/marshal.c: オブジェクトの読み書きをメソッドの再定義
+	  でコントロールできるように．インスタンスが`_dump_to'というメソッ
+	  ドを定義している時はそちらを使うように．
+
+	* ext/extmk.rb.in: static linkも設定できるような仕様にした．
+	  ext/Setupというファイルにディレクトリ名を記述するとそのディレク
+	  トリに存在するモジュールはstatic linkされる(はず)．
+
+	* eval.c (rb_eval): `..'を文法に組み込み，`..'と`...'の動作をperl
+	  に合わせた．
+
+Sat May 20 01:22:48 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* io.c (select): timeout時と割込み時の動作の明確化．
+
+Fri May 19 15:33:23 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* version 0.76
+
+Fri May 19 00:48:08 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* string.c (Fstr_each): イテレータブロック中で文字列の変更が行われ
+	  たかどうかをチェック．ポインタの値が変わっていれば例外を発生する．
+
+	* ruby-mode.el: ruby-electric-braceの新設．
+
+Thu May 18 12:27:23 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* string.c (Fstr_tr): trの置換対象に`\0'を含む時に正しく置換を行わ
+	  ないバグがあった．更に置換文字列をASCII順に指定しないと動作しな
+	  い問題もあった．結果としてtrを書き換えたので，copyrightの問題は
+	  無くなった(と思う)．
+
+	* gc.c (gc): the_scopeをマークしていなかったので，ローカル変数の指
+	  しているオブジェクトが間違って開放される場合があった．
+
+	* gc.c (mark_locations_array): 若干の高速化．
+
+Mon May 15 11:43:49 1995  Yukihiro Matsumoto  <matz@ix-02>
+
+	* ext/extmk.rb.in: Dynamic Loadモジュールのコンパイル用チェックを
+	  行うruby script．autoconfに近い感覚で使える．新しいモジュールを
+	  提供したい人はextの下にディレクトリを作るだけで良い．必須のファ
+	  イルはファイル名の一覧を記録した`MANIFEST'というファイルのみ．必
+	  要に応じて`depend'(ファイルの依存関係を記述するファイル gcc -MM
+	  の出力)，`extconf.rb'(コンパイル用にライブラリと関数の存在チェッ
+	  クするファイル)を用意できる．
+
+	* eval.c (rb_call): rubyメソッドの引数チェック時に未初期化の
+	  jmp_bufを使用していた．
+
+	* parse.y: `or'と`and'の優先順位を同じにした．
+
+Wed May  3 18:21:36 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* dln.c: Linuxでは`__.SYMDEF/'であった．
+
+	* dln.c: system callのエラーチェックを忘れていた．
+
+Wed Apr 26 09:50:56 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: イテレータブロックの変数宣言を`|'で括るようにした．これ
+	  でイテレータ変数がない時は宣言そのものを省略できる．文法の変更は
+	  久しぶりだ．
+
+Tue Apr 25 12:04:17 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(require): loadからダイナミックロードの機能を移してきた．
+	  さらに拡張子の補完機能を追加してユーザがdln/dlopenの差を意識する
+	  必要のないようにした．
+
+	* string.c(sub,sub): イテレータとしても動作するように．
+
+	* object.c: init_object -> initialize.
+
+Mon Apr 24 14:22:39 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* NEWS-OS 3.4対応
+
+	* io.c: Solarisのstdioの動作が違うようだ．signalでEOFを返してしま
+	  う…．perlでも同様の問題がある．
+
+Fri Apr 21 20:04:39 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.75
+
+	* signal.c: trapがなくなっていた．うーむ．
+
+	* configure: Solaris 2.3対応．
+
+	* io.c: #elifのないcppもある．
+
+	* dir.c: autoconf 2.xへの対応が不十分
+
+Thu Apr 20 12:31:24 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.74
+
+	* env.h, gc.c, regex.c: IRIXへの移植対応
+
+	* configure: dlopen用にpicを生成するoptionの検出のため，システムタ
+	  イプをチェックするように．
+
+Tue Apr 18 19:08:17 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* gc.c(xrealloc): ptr=nilの時，malloc()と同じ働きを
+
+	* array.c(astore): 空の配列の0番目の要素に代入するとsize=0で
+	  realloc()を呼んでいた．
+
+	* configure, glob.c: Solaris 2.xでコンパイルできるように
+
+Mon Apr 10 18:36:06 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.73
+
+Fri Apr  7 13:51:08 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* cons.c->assoc.c: consの余計な機能は外してpairとしての機能だけを
+	  残した．Enumerableをincludeするのもやめた．
+
+	* string.c(esub): 文字列置換イテレータ．perlのs///eの相当する．
+
+Wed Apr  5 11:35:21 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.72
+
+	* EWS4800対応
+
+	* file.c: utimesがない時はutimeを使うように．
+
+Mon Apr  3 15:19:41 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.71
+
+	* regexp.c(re_match): バグがあった．match_2を削除した時にenbugして
+	  いたのだった．
+
+Mon Mar 27 15:41:43 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* dict.c: Dict->Hashに全面的に移行．
+
+Thu Mar 23 20:30:00 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* dbm.c,socket.c: extディレクトリに分離．
+
+	* configure: dln周りのチェックの強化
+
+	* dln.c: initの呼び出しをdlopen()版に合わせた．
+
+Mon Mar 20 17:45:08 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* configure: autoconf 2.2に対応(一部)．
+
+Fri Mar 17 15:56:44 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* dln.c: dlopenのあるマシンではそちらを使うように．ただし，ちゃん
+	  と動いているかどうかは自信がない．
+
+	* regex.c: virtual concatinationをやめた．
+
+Thu Mar 16 11:32:57 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.70
+
+	* eval.c,regex.c: gccでのコンパイルエラー．
+
+	* io.c: inplace-editで拡張子が指定されない場合，もとのファイルを削
+	  除する．
+
+Wed Mar 15 14:59:18 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.69
+
+	* eval.c(method_missing): unknownから名称変更．
+
+	* eval.c(single_method_added): 特異メソッドが定義された時に呼ばれ
+	  るメソッド．hookとして使える．実際に定義される直前に呼ばれる．
+
+Tue Mar 14 14:46:44 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby.c(proc_options): 引数の解析を自分でやることにより引数指定の
+	  方法がperlに近付いた．getopt_longはもう使わない．
+
+	* dir.c(glob): `{}'のネストを許すようにした．
+
+Mon Mar 13 17:56:25 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* glob.c: Glob(ワイルドカードオブジェクト)はなくなった．ワイルドカー
+	  ドの展開はDir.glob(文字列)を使う．ワイルドカードのマッチは正規表
+	  現で代用．
+
+Fri Mar 10 18:35:46 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: Mathのようなモジュールは自分自身でextendする．
+
+	* eval.c: クラスやモジュールを定義する時，既に同名のものがあれば追
+	  加定義となるように．ただし．superクラスの違いなどはチェックする．
+
+	* regex.c: debug.
+
+	* math.c: 定数PIとEを定義．
+
+Thu Mar  9 21:35:12 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* regex.c: EUC,SJISモードでは0x80以上の8進，16進リテラルを禁止．
+
+	* regex.c: クラス内でも数値リテラル・文字クラスが使えるようした．
+
+Wed Mar  8 17:39:05 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* regex.c: \200など括弧の数以上の表現は8進リテラルと解釈する．ただ
+	  し，\1から\9までは例外．
+
+	* regex.c: \9以上のリファレンスも有効にした．
+
+Tue Mar  7 14:26:01 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(public/private): スコープ制御メソッドの名称変更．静的なア
+	  クセスも出来るようにしてみたが，不採用．
+
+Mon Mar  6 19:34:32 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c(inlcude): メソッド化．動的にモジュールをインクルードでき
+	  るように．さらに任意のオブジェクトにもモジュールをインクルードで
+	  きるメソッド `extend'も用意した．
+
+	* parse.y: 文法からincludeを削除．メソッド化．
+
+Tue Feb 28 15:35:10 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: 配列，連想配列の最後に`,'をおけるように．
+
+Fri Feb 24 13:15:43 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.68
+
+Thu Feb 23 11:19:19 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: resque節のselfの値が間違っていた．
+
+	* eval.c(rb_clear_cache): キャッシュのクリアし忘れがあった．
+
+	* eval.c: 定数のスコープをクラス内の静的スコープに変更した．これに
+	  よって，特異メソッドから参照される定数は，レシーバのクラスではな
+	  く，定義されたスコープのクラスの定数となる．
+
+Wed Feb 22 00:51:38 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* regex.c: ignorecaseを正規表現のコンパイル前に指定しないと正しく
+	  動作しない．修正．
+
+	* string.c(toupper,tolower): bug fix.
+
+	* ENV,VERSION: readonly変数から定数へ．
+
+Tue Feb 21 18:56:56 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* io.c(STDIN, STDOUT, STDERR): 定数として定義．
+
+	* io.c(select): bug fix.
+
+	* version 0.67
+
+Mon Feb 20 16:10:14 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(yylex): 定数を`%識別子'から，第1文字が大文字の識別子に変
+	  更．それにともないクラスは定数となった．
+
+	* eval.c: クラス定義内のselfがクラス定義外部のthe_classだった．
+
+	* variable.c(rb_name_class): クラス名をインスタンス変数に格納する．
+
+Thu Feb 16 15:36:17 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: BLOCKをbraceで表現する文法に変更したものを作ってみる．
+	  MLに提示してみるが反応がない．
+
+	* object.c(do,forever): なくした．
+
+Wed Feb 15 13:20:49 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* re.c(new): 第2引数が与えられて，かつnilでないときだけ設定するよ
+	  うに(以前はnilの時にも設定を行なっていた)．
+
+	* parse.y(parse_regexp): 正規表現リテラルで大文字小文字を無視する
+	  かどうか指定できるように．
+
+Tue Feb 14 00:55:33 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y: (compexpr) -> (expr).
+
+Fri Feb 10 16:30:00 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* ruby.c(load_file): scriptを読み込む時だけ"#!"の解析を行うように．
+
+	* ruby.c(readin): ファイル読み込み時に先頭に"#!"があり，その行が
+	  "ruby"という文字列を含む時，rubyに引数が与えられていれば，その引
+	  数も有効になる．
+
+	* parse.y(yylex): コメント行の終りが`\'であった時，次の行に継続し
+	  ているとみなすようにした．
+
+Thu Feb  9 16:18:37 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.66
+
+	* parse.y: protectをbeginに変更．begin..endは例外処理だけでなく，
+	  文括弧としても働くことになった．
+
+Wed Feb  1 19:48:24 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* version 0.65
+
+	* string.c(str_replace): 置き換える文字列の長さが等しい時メモリコ
+	  ピーをしない．
+
+	* string.c(rindex): バグ修正．
+
+Mon Jan 30 11:23:05 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(value_expr): ifのチェックを追加．
+
+	* gc.c(gc_mark): free cellの扱いにバグ．
+
+	* parse.y: 文法の変更(よりシンプルに)．例外を減らした．
+
+Thu Jan 26 00:52:55 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* parse.y: 引数として連想配列を置くことができるように．この場合，
+	  連想配列リテラルが最終引数となる．
+
+	* parse.y: 配列参照の`[]'内が空でもよいことにした．
+
+Tue Jan 24 14:45:15 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* class.c(rb_include_module): `-v'を指定した時にはincludeしたモジュー
+	  ルとクラス定数が衝突していないかチェックする．
+
+Mon Jan 23 10:42:09 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(rb_class2name): メタクラスに関するbug fix.
+
+	* dict.c: Dict[..]で辞書の生成が出来るように．
+
+	* array.c: Array[..]で配列の生成が出来るように．
+
+	* parse.y: 辞書の表現として{a,b,..}という形式も許すように．
+
+Fri Jan 20 10:28:38 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* re.c(Regexp.quote): 正規表現をエスケープするメソッド．
+
+	* 無駄なrb_intern()を減らした．
+
+	* parse.y: `!', `!=', `!~'を特殊演算子にする．よってこれらは再定義
+	  できなくなった．
+
+Wed Jan 18 13:20:41 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: 文法の整理(unless,untilをなくした)．
+
+Tue Jan 17 11:11:27 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: defでメソッド再定義時にはスーパークラスのメソッドの可視
+	  性を継承する．最初の定義の時は今までと同じデフォルト(トップレベ
+	  ルで関数的，クラス定義内で通常メソッド)．
+
+	* object.c(Class#new): オブジェクトの生成時に関数的メソッド
+	  init_objectが必ず呼ばれるように変更．
+
+	* eval.c: 未定義のメソッドに対してunknownメソッドが呼ばれるように
+	  なった．エラー表示が今までと同じになるようにenvを調節している．
+
+Fri Jan 13 14:40:30 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* gc.c: gcを若干書き換えて整理した．が，あまり変化はなかったようだ．
+
+	* parse.y(yylex): symbolを\symから:symに変更した．
+
+Thu Jan 12 01:39:28 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c: 新規関数 rb_eval_string()．
+
+	* gc.c: gc_mark()を一部非再帰化．
+
+	* variable.c(rb_ivar_{get,set}): インスタンス変数のアクセス周りで
+	  チェックが足りなかった．
+
+	* variable.c: クラス定数とインスタンス変数でハッシュテーブルを共有
+	  するようにした．
+
+	* ruby.h: iv_tblをRBasicからRObjectとRClassへ移動した．これにより，
+	  ObjectとClass,Moduleしかインスタンス変数を持てなくなる．が，メモ
+	  リ効率は若干向上する．
+
+Tue Jan 10 00:58:20 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* 0.64 released
+
+	* eval.c: レシーバと引数は常にiterではない．
+
+	* cons.c(aref,aset): negative offset対応．
+
+Mon Jan  9 14:40:39 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: foo{..}の形式において，fooをローカル変数やクラス名では
+	  なく，引数なしの関数型メソッド呼び出しとみなすようにした．
+
+	* list.c -> cons.c: 名称変更(クラス名も)．
+
+	* list.c: a::b::c::nilをリスト(a b c)とみなすlisp形式から，a::b::c
+	  をリスト(a b c)とみなすruby形式に変更．[], []=, eachもそれに会わ
+	  せた仕様とする．
+
+	* list.c: consペアとしての機能を強調．仕様変更．
+
+Sat Jan  7 01:26:26 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c: 自己代入の不具合修正．
+
+	* eval.c(masign): 多重代入が配列もリストもとれるようにした．
+
+	* list.c: assocを2要素の配列からList(CONSペア)に変更した．
+
+Fri Jan  6 13:42:12 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: a[b]+=cやa.b+=cなどの自己代入形式で，aやbを2度評価しな
+	  くなった．
+
+	* eval.c: iterator設定のバグフィックス．
+
+	* list.c: Listクラスを新設．
+
+Thu Jan  5 13:55:00 1995  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: SCOPEのメモリリークをなくした．
+
+	* eval.c: built-inメソッドへの引数の引き渡し方を変更して，配列の生
+	  成数を減らした．
+
+	* re.c: match-dataを毎回生成することをやめた．`$~'をアクセスした時
+	  にon-demandで生成する．
+
+	* string.c etc: 不必要なmemmoveをmemcpyに置換．
+
+	* parse.y: =~, !~は副作用があるのでコンパイル時に展開できない．
+
+Tue Jan  3 02:04:36 1995  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c: rest引数のbug fix.
+
+	* eval.c,gc.c: scopeをオブジェクトにした．
+
+	* eval.c: envとscopeの扱いを変更した．
+
+Wed Dec 28 09:46:57 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: evalでローカル変数が追加された場合に対応した．
+
+	* parse.y: 演算子を含むaliasのbug fix.
+
+Tue Dec 27 16:45:20 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: def A Bをalias A Bに変更．
+
+	* eval.c: alias関係のbug修正．nodeをオブジェクト化した時にenbugし
+	  たようだ．
+
+	* signal.c: システムコールの再定義を止めた．
+
+	* io.c(select): write/exceptのフラグ設定にバグ．
+
+	* Makefile.in: static link用オプションをMake変数として独立させた．
+
+Tue Dec 20 00:46:19 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* 0.63 released
+
+	* eval.c(rb_call): superの呼び出しで落ちる．argc, argvの設定を忘れ
+	  ていた．
+
+	* parse.y(read_escape): 展開エラー．
+
+	* variable.c: 定義済みの変数のhookを変更しないように．
+
+Mon Dec 19 12:01:10 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y(cond): 条件式に代入式が置かれた場合，`-v'オプションで警
+	  告が出るように．
+
+	* parse.y(**): 冪乗演算子`**'の優先順位を単項演算子より高くした．
+
+	* parse.y(and,or): 優先順位の低い演算子`and', `or'.
+
+	* 0.62 released.
+
+	* eval.c: 不必要になったPUSH_ENV, POP_ENVを減らした．
+
+	* env.h: ENVIONからselfをはずした．PUSH_ENVはsuperの準備のためだけ
+	  に用いることにした．
+
+	* eval.c: 下記のオブジェクト化で遅くなった実行速度をもとに戻した．
+
+Mon Dec 17 23:01:10 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* eval.c: env.{argv,argc}とscope.local_varsのオブジェクト化．
+
+	* eval.c: 1スコープ内で複数Blockを生成したときのバグを修正．
+
+Fri Dec 16 15:52:06 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* parse.y: `&&'と`||'の両辺はいつでも条件式とした．
+
+Thu Dec 15 00:16:04 1994  Yukihiro Matsumoto  (matz@dyna)
+
+	* eval.c(Block): Blockオブジェクトを実現．
+
+	* node.h: NODE_QLISTはなくなった．
+
+	* eval.c(rb_call): 引数への代入を名前で一つずつ代入するのをやめて，
+	  一度にコピーするようにした．
+
+	* eval.c(rb_call): rubyで記述されたメソッドへの引数渡しをinline化．
+
+	* eval.c: イテレータ判定処理の全面書き換え．不適切なイテレータ呼び
+	  出しをなくした．例えば「[foo(),bar()]{i|baz(i)}」でfooもbarもイ
+	  テレータとして呼び出され*ない*．
+
+	* eval.c(rb_call): SCOPE処理をinline化．メソッド呼び出しの若干の高
+	  速化．
+
+Wed Dec 14 18:09:33 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* node.h: nodeもオブジェクトにする．よってGCで回収される．
+
+Thu Dec  8 14:17:29 1994  Yukihiro Matsumoto  (matz@ix-02)
+
+	* 0.60 released - alpha test baseline.
diff --git a/doc/ChangeLog-1.8.0 b/doc/ChangeLog-1.8.0
index d168a50f80..3f7d6bfb3c 100644
--- a/doc/ChangeLog-1.8.0
+++ b/doc/ChangeLog-1.8.0
@@ -42,7 +42,7 @@ Sun Aug  3 23:56:50 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
 
 Sun Aug  3 22:07:47 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tkentry.rb: support 'validatecommand' option of 
+	* ext/tk/lib/tkentry.rb: support 'validatecommand' option of
 	  TkEntry/TkSpinbox widget
 
 	* ext/tk/sample/{demos-en,demos-jp}/spin.rb: add
@@ -60,7 +60,7 @@ Sun Aug  3 18:03:44 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
 
 Sun Aug  3 08:53:06 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/sample/{demos-en,demos-jp}/image3.rb: add 
+	* ext/tk/sample/{demos-en,demos-jp}/image3.rb: add
 
 	* ext/tk/lib/tkcanvas.rb: bug fix on Tk object ID management
 
@@ -84,7 +84,7 @@ Sat Aug  2 23:51:52 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 Sat Aug  2 09:58:13 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: bug fix --- TkGrid failed to treat 
+	* ext/tk/lib/tk.rb: bug fix --- TkGrid failed to treat
 	  RELATIVE PLACEMENT
 
 	* ext/tk/sample/demos-en/, demos-jp/: add or modify some
@@ -122,7 +122,7 @@ Sat Aug  2 14:02:39 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
 
 Sat Aug  2 09:58:13 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: bug fix --- forgot to entry a widget class 
+	* ext/tk/lib/tk.rb: bug fix --- forgot to entry a widget class
 	  name of 'labelframe' widget
 
 	* ext/tk/sample/{demos-en,demos-jp}/{labelframe.rb,paned1.rb,
@@ -238,7 +238,7 @@ Fri Aug  1 09:54:38 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 Fri Aug  1 04:58:55 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: bug fix --- forget to eval given block to 
+	* ext/tk/lib/tk.rb: bug fix --- forget to eval given block to
 	  TkRoot.new method
 
 	* ext/tk/sample/tkoptdb-safeTk.rb: new sample script
@@ -255,13 +255,13 @@ Thu Jul 31 23:44:00 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
 
 Thu Jul 31 23:04:45 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/sample/resource.en, ext/tk/sample/resource.jp: 
+	* ext/tk/sample/resource.en, ext/tk/sample/resource.jp:
 	  wrong resource file format
 
-	* ext/tk/lib/tk.rb: add Tk::Encoding.{encoding_convertfrom, 
+	* ext/tk/lib/tk.rb: add Tk::Encoding.{encoding_convertfrom,
 	  encoding_convertto}
 
-	* ext/tk/lib/tk.rb: add TkOptionDB.read_with_encoding to read 
+	* ext/tk/lib/tk.rb: add TkOptionDB.read_with_encoding to read
 	  non-utf8 resource file
 
 Thu Jul 31 23:02:47 2003  NAKAMURA Usaku <usa@ruby-lang.org>
@@ -274,15 +274,15 @@ Thu Jul 31 20:52:40 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 	* ext/tk/lib/tk.rb: (IMPORTANT BUG FIX) scan of event keywords
 	  doesn't work on recent versions of Tck/Tk
 
-	* ext/tk/lib/tk.rb: initialize error of instance variable on 
+	* ext/tk/lib/tk.rb: initialize error of instance variable on
 	  TkComposite
 
-	* ext/tk/lib/multi-tk.rb: initialize error on encoding-system on 
+	* ext/tk/lib/multi-tk.rb: initialize error on encoding-system on
 	  MultiTkIp
 
 	* ext/tk/lib/tk.rb: trouble on destroying widgets
 
-	* ext/tk/sample/demos-en/, demos-jp/: add JP and EN version of 
+	* ext/tk/sample/demos-en/, demos-jp/: add JP and EN version of
 	  Ruby/Tk widget demos
 
 Thu Jul 31 15:25:12 2003  NAKAMURA Usaku <usa@ruby-lang.org>
@@ -311,13 +311,13 @@ Thu Jul 31 08:18:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
 
 Thu Jul 31 07:59:18 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: wrap the command-proc of TkScale --- pass 
+	* ext/tk/lib/tk.rb: wrap the command-proc of TkScale --- pass
 	  the numeric object to the proc
 
-	* ext/tk/lib/tk.rb: better support for widgets created on 
+	* ext/tk/lib/tk.rb: better support for widgets created on
 	  Tk interpreter (without Ruby)
 
-	* ext/tk/lib/multi-tk.rb: a little more stable on Multiple Tk 
+	* ext/tk/lib/multi-tk.rb: a little more stable on Multiple Tk
 	  interpreters running
 
 Thu Jul 31 00:17:19 2003  Shugo Maeda  <shugo@ruby-lang.org>
@@ -411,17 +411,17 @@ Wed Jul 30 07:23:14 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tkentry.rb: fix lack of methods for TkEntry
 
-	* ext/tk/lib/multi-tk.rb, ext/tk/lib/tk.rb, 
-	  ext/tk/lib/tkdialog.rb, ext/tk/lib/tkentry.rb, 
+	* ext/tk/lib/multi-tk.rb, ext/tk/lib/tk.rb,
+	  ext/tk/lib/tkdialog.rb, ext/tk/lib/tkentry.rb,
 	  ext/tk/sample/safe-tk.rb, ext/tk/sample/tktimer2.rb: bug fix
 
-	* ext/tk/lib/multi-tk.rb: MultiTkIp.new_* accept a block to 
+	* ext/tk/lib/multi-tk.rb: MultiTkIp.new_* accept a block to
 	  eval under the new interpreter
 
 Wed Jul 30 04:36:30 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tcltklib/tcltklib.c, 
-	  ext/tk/lib/tk.rb, ext/tk/lib/tkafter.rb: additional check of 
+	* ext/tcltklib/tcltklib.c,
+	  ext/tk/lib/tk.rb, ext/tk/lib/tkafter.rb: additional check of
 	  Tk interpreters' status for a little more safety
 
 Wed Jul 30 02:37:12 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
@@ -514,14 +514,14 @@ Mon Jul 28 22:57:52 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 Tue Jul 29 16:20:36 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tcltklib/tcltklib.c: bug fix and 
+	* ext/tcltklib/tcltklib.c: bug fix and
 	  change mainloop_abort_on_no_widget_cmd => mainloop_abort_on_exception
 	  ( to avoid thread timing trouble on accessing destroyed widgets )
 
-	* ext/tk/lib/multi-tk.rb: change default mode of 
+	* ext/tk/lib/multi-tk.rb: change default mode of
 	  mainloop_abort_on_exception on multi-tk.rb
 
-	* ext/tk/lib/multi-tk.rb: fix a bug of the procedure for 
+	* ext/tk/lib/multi-tk.rb: fix a bug of the procedure for
 	  'Delete' button on the safe-Tk frmae
 
 Tue Jul 29 12:22:28 2003  why the lucky stiff  <ruby-cvs@whytheluckystiff.net>
@@ -541,8 +541,8 @@ Tue Jul 29 12:15:37 2003  NAKAMURA Usaku <usa@ruby-lang.org>
 
 Tue Jul 29 08:05:30 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb, ext/tk/lib/tkdialog.rb, ext/tk/lib/tktext.rb, 
-	  ext/tk/sample/tkbiff.rb, ext/tk/sample/tkdialog.rb, 
+	* ext/tk/lib/tk.rb, ext/tk/lib/tkdialog.rb, ext/tk/lib/tktext.rb,
+	  ext/tk/sample/tkbiff.rb, ext/tk/sample/tkdialog.rb,
 	  ext/tk/sample/tkform.rb: bug fix ( tested with Ruby/Tk widget demo )
 
 Tue Jul 29 04:22:08 2003  why the lucky stiff  <ruby-cvs@whytheluckystiff.net>
@@ -571,7 +571,7 @@ Tue Jul 29 01:24:32 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 	* ext/tk/lib/multi-tk.rb: bug fix and pack options are pssed
 	  to the safeTk container
 
-	* ext/tk/sample/safe-tk.rb: add example for pack options of 
+	* ext/tk/sample/safe-tk.rb: add example for pack options of
 	  safeTk container
 
 Mon Jul 28 23:23:08 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
@@ -643,7 +643,7 @@ Sun Jul 27 19:35:06 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 	* ext/tcltklib/tcltklib.c: add some methods to support
 	  multiple interpreters (low level)
 
-	* ext/tk/lib/multi-tk.rb: new library to support multiple Tk 
+	* ext/tk/lib/multi-tk.rb: new library to support multiple Tk
 	  interpreters (high level)
 
 	* ext/tcltklib/demo/safeTk.rb: new sample of safeTk interpreter
@@ -732,12 +732,12 @@ Fri Jul 26 00:04:25 2003  NAKAMURA, Hiroshi  <nakahiro@sarion.co.jp>
 
 Fri Jul 25 16:43:03 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tcltklib/tcltklib.c: add TclTkIp#create_slave, 
+	* ext/tcltklib/tcltklib.c: add TclTkIp#create_slave,
 	  TclTkIp#_make_safe and TclTkIp#safe?
 
 	* ext/tcltklib/MANUAL.euc: modify descriptions
 
-	* ext/tk/lib/tk.rb: bug fix [ruby-talk:76980] and modify to 
+	* ext/tk/lib/tk.rb: bug fix [ruby-talk:76980] and modify to
 	  support multi Tk IPs
 
 	* ext/tk/lib/tkafter.rb: modify to support multi Tk IPs
@@ -747,10 +747,10 @@ Fri Jul 25 15:47:39 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
 	* ext/openssl/extconf.rb: add check for BN_rand_range() and
 	  BN_pseudo_rand_range().
 
-	* ext/openssl/ossl_bn.c (ossl_bn_s_rand_range): should raise 
+	* ext/openssl/ossl_bn.c (ossl_bn_s_rand_range): should raise
 	  NotImplementedError if BN_rand_range() wan not defined.
 
-	* ext/openssl/ossl_bn.c (ossl_bn_s_pseudo_rand_range): should raise 
+	* ext/openssl/ossl_bn.c (ossl_bn_s_pseudo_rand_range): should raise
 	  NotImplementedError if BN_pseudo_rand_range() wan not defined.
 
 	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_s_encrypt): avoid compiler
@@ -1110,7 +1110,7 @@ Thu Jul 17 13:42:53 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
 
 Thu Jul 17 06:40:28 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: recover and fix typo : Tk.chooseDirectory 
+	* ext/tk/lib/tk.rb: recover and fix typo : Tk.chooseDirectory
 	  (Tk8.4 feature)
 
 Wed Jul 16 16:23:58 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
@@ -1139,7 +1139,7 @@ Tue Jul 15 14:38:21 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 Tue Jul 15 03:30:41 2003  why the lucky stiff  <ruby-cvs@whytheluckystiff.net>
 
-	* ext/syck/rubyext.c (syck_mark_emitter): forgot to rb_gc_mark the 
+	* ext/syck/rubyext.c (syck_mark_emitter): forgot to rb_gc_mark the
 	  outgoing IO object.
 
 Sun Jul 13 14:55:36 2003  Koji Arai  <jca02266@nifty.ne.jp>
@@ -1196,7 +1196,7 @@ Fri Jul 11 16:09:09 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 Fri Jul 11 07:17:47 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: not create a Tcl/Tk interpreter if already 
+	* ext/tk/lib/tk.rb: not create a Tcl/Tk interpreter if already
 	  defined TkCore::INTERP
 
 	* ext/tk/lib/tk.rb: bugfix on TkWindow#configure
@@ -1306,7 +1306,7 @@ Thu Jul  3 14:22:46 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 Thu Jul  3 12:13:05 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
 
-	* lib/mkmf.rb (VPATH): convert from Windows form to Unix form on 
+	* lib/mkmf.rb (VPATH): convert from Windows form to Unix form on
 	  MinGW.  This fixes the build with GNU make 3.80-1 for Cygwin.
 
 Wed Jul  2 23:27:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
@@ -1454,16 +1454,16 @@ Fri Jun 27 03:24:54 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
 Thu Jun 26 21:34:49 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
 
 	* class.c (class_instance_method_list): get rid of warning about
-	  arguement type mismatch, and inline method_list().
+	  argument type mismatch, and inline method_list().
 	  [ruby-core:01198]
 
 Wed Jun 25 14:40:33 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: add and modify methods --- 
-	  TkWidget.database_class, TkWidget.database_classname, 
-          TkWidget#database_class, TkWidget#database_classname 
+	* ext/tk/lib/tk.rb: add and modify methods ---
+	  TkWidget.database_class, TkWidget.database_classname,
+          TkWidget#database_class, TkWidget#database_classname
 
-	* ext/tk/lib/tk.rb: instances of a subclass of TkToplevel or 
+	* ext/tk/lib/tk.rb: instances of a subclass of TkToplevel or
 	  TkFrame are created with ":class=>subclass" option as default.
 
 	* ext/tk/sample/tkoptdb.rb: add a new part
@@ -1478,7 +1478,7 @@ Wed Jun 25 05:49:10 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tk.rb: add widget destroy hook binding to TkBindTag::ALL
 
-	* ext/tk/lib/tkcanvas.rb: Although requiring manual control of GC, 
+	* ext/tk/lib/tkcanvas.rb: Although requiring manual control of GC,
 	  memory eating problem of TkCanvas Items is fixed.
 
 	* ext/tk/lib/tktext.rb: add some methods and bug fix
@@ -1493,10 +1493,10 @@ Tue Jun 24 16:46:07 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 	* ext/tk/lib/tk.rb: bug fix on TkToplevel, TkFrame,
 	  TkPanedwindow, TkOptionDB
 
-	* ext/tk/lib/tk.rb: TkOptionDB --- make it more secure to use procs 
+	* ext/tk/lib/tk.rb: TkOptionDB --- make it more secure to use procs
 	  defined on resourceDB
 
-	* ext/tk/sample/tkoptdb.rb, resource.ja, resource.en: 
+	* ext/tk/sample/tkoptdb.rb, resource.ja, resource.en:
 	  sample script how to use TkOptionDB.
 
 Tue Jun 24 14:22:41 2003  why the lucky stiff  <ruby-cvs@whytheluckystiff.net>
@@ -1610,14 +1610,14 @@ Sun Jun 22 23:42:20 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
 
 Sun Jun 22 16:17:02 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: TkRoot.new and TkToplevel.new accept Wm 
+	* ext/tk/lib/tk.rb: TkRoot.new and TkToplevel.new accept Wm
 	  commands as elements
 
 	* ext/tk/lib/tk.rb: TkMenu --- add some methods
 
 	* ext/tk/lib/tk.rb: TkOptionMenubutton --- bug fix
 
-	* ext/tk/sample/tkmenubutton.rb: sample of TkMenubutton and 
+	* ext/tk/sample/tkmenubutton.rb: sample of TkMenubutton and
 	  TkOptionMenubutton
 
 Sat Jun 21 23:15:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
@@ -1625,7 +1625,7 @@ Sat Jun 21 23:15:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 	* eval.c (proc_invoke): should not propagate distination tag if
 	  tag is already handled in this level. (ruby-bugs-ja PR#501)
 
-	* object.c (str_to_id): check for empty string before intern. 
+	* object.c (str_to_id): check for empty string before intern.
 	  [ruby-talk:74006]
 
 Sat Jun 21 13:56:09 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
@@ -1636,10 +1636,10 @@ Sat Jun 21 13:56:09 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
 
 Sat Jun 21 12:55:17 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
-	* ext/tk/lib/tk.rb: TkRoot.new and TkToplevel.new accept Wm commands 
-	  as elements of a hash argument. 
+	* ext/tk/lib/tk.rb: TkRoot.new and TkToplevel.new accept Wm commands
+	  as elements of a hash argument.
 
-	* ext/tk/sample/tktimer2.rb: add comments about the usage of a 
+	* ext/tk/sample/tktimer2.rb: add comments about the usage of a
 	  TkTimer object.
 
 Sat Jun 21 08:47:22 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
@@ -1647,10 +1647,10 @@ Sat Jun 21 08:47:22 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 	* ext/tk/lib/tk*.rb: remove direct-accesses to TkComm::INTERP and
 	  TkComm::INITIALIZE_TARGETS
 
-	* ext/tk/lib/tk*.rb: use TkINTERP_SETUP_SCRIPTS constant for setting 
+	* ext/tk/lib/tk*.rb: use TkINTERP_SETUP_SCRIPTS constant for setting
 	  up the interpreter
 
-	* ext/tcltklib/tcltklib.c: support to create a safe interpreter 
+	* ext/tcltklib/tcltklib.c: support to create a safe interpreter
 	  with safe-Tk
 
 Fri Jun 20 23:28:27 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
@@ -1660,7 +1660,7 @@ Fri Jun 20 23:28:27 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 Fri Jun 20 15:04:28 2003  NAKAMURA Usaku <usa@ruby-lang.org>
 
-	* defines.h (PATH_ENV): name of PATH environment. [new]. 
+	* defines.h (PATH_ENV): name of PATH environment. [new].
 
 	* defines.h (ENV_IGNORECASE): define for case insensitive platforms
 	  to access environment variables.
@@ -1678,7 +1678,7 @@ Fri Jun 20 14:52:46 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 Fri Jun 20 03:09:21 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
-	* parse.y (new_yield): distinguish "yield 1,2" and "yield [1,2]". 
+	* parse.y (new_yield): distinguish "yield 1,2" and "yield [1,2]".
 	  [ruby-dev:20360]
 
 	* eval.c (rb_eval): support new_yield() change.
@@ -1688,7 +1688,7 @@ Fri Jun 20 03:09:21 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
 	  [ruby-list:36935]
 
 	* parse.y (no_blockarg): separate no block argument check and
-	  ret_args argument processing. 
+	  ret_args argument processing.
 
 Fri Jun 20 00:45:19 2003  NAKAMURA, Hiroshi <nahi@ruby-lang.org>
 
@@ -1696,9 +1696,9 @@ Fri Jun 20 00:45:19 2003  NAKAMURA, Hiroshi <nahi@ruby-lang.org>
 
 Thu Jun 19 22:51:41 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
 
-	* lib/drb.rb, lib/drb/drb.rb, lib/drb/eq.rb, 
-	  lib/drb/extserv.rb, lib/drb/extservm.rb, lib/drb/gw.rb, 
-	  lib/drb/invokemethod.rb, lib/drb/observer.rb, 
+	* lib/drb.rb, lib/drb/drb.rb, lib/drb/eq.rb,
+	  lib/drb/extserv.rb, lib/drb/extservm.rb, lib/drb/gw.rb,
+	  lib/drb/invokemethod.rb, lib/drb/observer.rb,
 	  lib/drb/timeridconv.rb, lib/drb/unix.rb: import drb-2.0.4b3
 
 Thu Jun 19 16:14:43 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
@@ -1710,17 +1710,17 @@ Thu Jun 19 16:14:43 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tcltklib/tcltklib.c: add TclTkLib::EventFlag::NONE ( == 0 )
 
-	* ext/tcltklib/tcltklib.c: add set_no_event_wait() and 
+	* ext/tcltklib/tcltklib.c: add set_no_event_wait() and
 	  get_no_event_wait()
 
 	* ext/tcltklib/MANUAL.euc: modify
 
 	* ext/tcltklib/README.euc: ditto
 
-	* ext/tk/lib/tk.rb: change default value of TkCore.do_one_event 
+	* ext/tk/lib/tk.rb: change default value of TkCore.do_one_event
 	  argument
 
-	* ext/tk/lib/tk.rb: add TkCore.set_no_event_wait(wait) and 
+	* ext/tk/lib/tk.rb: add TkCore.set_no_event_wait(wait) and
 	  TkCore.get_no_event_wait
 
 	* ext/tk/lib/tk.rb: add Tk.exit ( == destroy root widget )
@@ -1730,7 +1730,7 @@ Thu Jun 19 16:14:43 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tkafter.rb: set_callback returns self
 
-	* ext/tk/lib/tkafter.rb: continue() raises an exception, if already 
+	* ext/tk/lib/tkafter.rb: continue() raises an exception, if already
 	  running or no procedure.
 
 	* ext/tk/lib/tkafter.rb: skip() raises an exception, if not running.
@@ -1768,15 +1768,15 @@ Wed Jun 18 19:46:21 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tk.rb: rename 'no_create' option to 'without_creating'
 
-	* ext/tk/lib/tk.rb: add TkWindow#pack_in, TkWindow#grid_in, 
+	* ext/tk/lib/tk.rb: add TkWindow#pack_in, TkWindow#grid_in,
 	  TkWindow#place_in
 
 	* ext/tk/lib/tk.rb: add TkWindow#bind_class and TkWindow#database_class
 
-	* ext/tk/lib/tk.rb: add TkBindTag.new_by_name and TkDatabaseClass 
+	* ext/tk/lib/tk.rb: add TkBindTag.new_by_name and TkDatabaseClass
 	  for binding to database class
 
-	* ext/tk/lib/tk.rb: check varname whether already exsist or not. 
+	* ext/tk/lib/tk.rb: check varname whether already exsist or not.
 	  (TkVarAccess.new)
 
 	* ext/tk/lib/tk.rb: TkTextWin#bbox returns an array of four numbers
@@ -1788,21 +1788,21 @@ Wed Jun 18 19:46:21 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tk.rb: TkBindTag.new accepts a block
 
-	* ext/tk/lib/tk.rb: If given taglist, TkWindow#bindtags(taglist) 
+	* ext/tk/lib/tk.rb: If given taglist, TkWindow#bindtags(taglist)
 	  returns taglist
 
 	* ext/tk/lib/tk.rb: add TkWindow#bindtags=(taglist)
 
-	* ext/tk/lib/tk.rb: Tk.focue and Tk.focus_lastfor return nil 
+	* ext/tk/lib/tk.rb: Tk.focue and Tk.focus_lastfor return nil
 	  if there is no target widget.
 
-	* ext/tk/lib/tk.rb: Tk::Wm.client returns the argument string 
+	* ext/tk/lib/tk.rb: Tk::Wm.client returns the argument string
 	  when setting name
 
-	* ext/tk/lib/tk.rb: TkGrid.columnconfiginfo and rowconfiginfo 
+	* ext/tk/lib/tk.rb: TkGrid.columnconfiginfo and rowconfiginfo
 	  given a slot return a number.
 
-	* ext/tk/lib/tk.rb: TkWindow.grid_columnconfiginfo and 
+	* ext/tk/lib/tk.rb: TkWindow.grid_columnconfiginfo and
 	  grid_rowconfiginfo --- ditto
 
 	* ext/tk/lib/tk.rb: rename and define alias :: TkOption ==> TkOptionDB
@@ -1813,10 +1813,10 @@ Wed Jun 18 19:46:21 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tk.rb: some TkComm methods change to module functions
 
-	* ext/tk/lib/tk.rb: add support for -displayof option to some 
+	* ext/tk/lib/tk.rb: add support for -displayof option to some
 	  TkWinfo methods
 
-	* ext/tk/lib/tk.rb: bind, bind_append and bind_remove --- 
+	* ext/tk/lib/tk.rb: bind, bind_append and bind_remove ---
 	  returns the target of event-binding
 
 	* ext/tk/lib/tk.rb: add Tk8.4 features
@@ -1837,7 +1837,7 @@ Wed Jun 18 19:46:21 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tkentry.rb: TkEntry#bbox returns an array of four numbers
 
-	* ext/tk/lib/tkentry.rb: scan validatecommand arguments and 
+	* ext/tk/lib/tkentry.rb: scan validatecommand arguments and
 	  convert to proper type
 
 	* ext/tk/lib/tkbgerror.rb: support to define a error handler by user
@@ -1932,7 +1932,7 @@ Fri Jun 13 09:24:39 2003  Shugo Maeda  <shugo@ruby-lang.org>
 Thu Jun 12 22:13:13 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tk.rb : add 'no_create' option to widget
-	  initialize method. 
+	  initialize method.
 
 	* ext/tk/MANIFEST : forgot to commit when added tkmacpkg.rb
 	  and tkwinpkg.rb
@@ -1947,7 +1947,7 @@ Thu Jun 12 21:14:11 2003  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 	* ext/tk/lib/tkmacpkg.rb : Mac resource (not new but not
 	  included until now)
 
-	* ext/tk/lib/tkwinpkg.rb : Win DDE and registry (not new but not 
+	* ext/tk/lib/tkwinpkg.rb : Win DDE and registry (not new but not
 	  included until now)
 
 Tue Jun 10 14:26:30 2003  why the lucky stiff  <ruby-cvs@whytheluckystiff.net>
@@ -3361,7 +3361,7 @@ Fri Apr 11 02:41:35 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
 
 Thu Apr 10 21:12:19 2003  Minero Aoki  <aamine@loveruby.net>
 
-	* lib/net/pop.rb: Exception line was accidentaly removed.
+	* lib/net/pop.rb: Exception line was accidentally removed.
 	  [ruby-dev:19989]
 
 Thu Apr 10 18:42:13 2003  Tadayoshi Funaba  <tadf@dotrb.org>
@@ -5986,7 +5986,7 @@ Thu Nov  7 09:51:37 2002  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
 Wed Nov  6 16:57:06 2002  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 	* class.c (rb_define_method): do not set NOEX_CFUNC if klass is
-	  really a module, whose methods must be safe for reciever's type.
+	  really a module, whose methods must be safe for receiver's type.
 
 	* eval.c (rb_eval): nosuper should not be inherited unless the
 	  overwritten method is an undef placeholder.
@@ -8346,7 +8346,7 @@ Tue Jun  4 07:03:33 2002  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tkfont.rb:  Fix bugs on TkFont.init_widget_font for Tk8.x.
 
-	* ext/tk/lib/tkafter.rb: Add self to 1st argument of interval- 
+	* ext/tk/lib/tkafter.rb: Add self to 1st argument of interval-
 	  and loop-proc
 	  TkAfter#current_interval returns an interval (sleep) time value
 	  TkAfter#current_args returns an array of arguments
@@ -8354,7 +8354,7 @@ Tue Jun  4 07:03:33 2002  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 
 	* ext/tk/lib/tk*.rb: Allow to use Symbols for parameters.
 
-	* ext/tk/lib/tkcanvas.rb: (TkcItem) Add 'coords' parameter to the 
+	* ext/tk/lib/tkcanvas.rb: (TkcItem) Add 'coords' parameter to the
 	  canvas item constructor (for new notation of constructor).
 
 	* ext/tcltklib/tcltklib.c: New 'mainloop' and 'mainloop_watchdog'.
@@ -8363,7 +8363,7 @@ Tue Jun  4 07:03:33 2002  Hidetoshi NAGAI <nagai@ai.kyutech.ac.jp>
 	  'use' parameter.
 
 	* ext/tk/lib/tk.rb: Add new parameter 'widgetname' to the widget
-	  constructor to support effective use of Resource Database. 
+	  constructor to support effective use of Resource Database.
 
 	* ext/tk/lib/tk.rb: TkOption::get always returns a tainted string.
 
@@ -10516,6 +10516,11 @@ Thu Jan 17 16:21:42 2002  Yukihiro Matsumoto  <matz@ruby-lang.org>
 	* eval.c (block_pass): should allow block argument relay even in
 	  the tainted mode.
 
+Thu Jan 17 09:43:05 2002  TAKAHASHI Kaoru  <kaoru@kaisei.org>
+
+	* misc/ruby-mode.el (ruby-add-log-current-method): New function
+	for `add-log-current-defun-function'.
+
 Thu Jan 17 04:51:48 2002  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 	* ext/socket/socket.c: support subclassing by proper "initialize"
@@ -13758,7 +13763,7 @@ Tue Apr 17 17:33:55 2001  Yukihiro Matsumoto  <matz@ruby-lang.org>
 
 	* eval.c (handle_rescue): use === to compare exception match.
 
-	* error.c (syserr_eqq): comparison between SytemCallErrors should
+	* error.c (syserr_eqq): comparison between SystemCallErrors should
 	  based on their error numbers.
 
 Tue Apr 17 16:54:39 2001  K.Kosako  <kosako@sofnec.co.jp>
@@ -21231,7 +21236,7 @@ Sun Jan  3 01:37:58 1999  Takao KAWAMURA  <kawamura@ike.tottori-u.ac.jp>
 Sat Jan  2 17:09:06 1999  Yukihiro Matsumoto  <matz@netlab.co.jp>
 
 	* eval.c (rb_jump_tag): new api to invoke JUMP_TAG.  tag values
-	  can obtained from rb_eval_string_protect()/rb_load_protect().
+	  can be obtained from rb_eval_string_protect()/rb_load_protect().
 
 	* eval.c (rb_rescue): now catches all exceptions but SystemExit.
 
diff --git a/doc/ChangeLog-1.9.3 b/doc/ChangeLog-1.9.3
new file mode 100644
index 0000000000..eecfc44325
--- /dev/null
+++ b/doc/ChangeLog-1.9.3
@@ -0,0 +1,92772 @@
+Sun Jul 10 22:50:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (rmpd_power_by_big_decimal): fix
+	  precision treatment errors.
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests for the above change.
+	  fix precision treatment errors.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_power): precision argument
+	  should be optional for its compatibility.
+
+Sun Jul 10 22:38:09 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (var_ref): distinguish vcall from local variable
+	  references.  based on a patch by Michael Edgar michael.j.edgar
+	  AT dartmouth.edu.  Bug #5002
+
+Sun Jul 10 21:51:29 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: add comments (cautions).
+
+Sun Jul 10 20:59:38 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Add new class variable `@@testfile_prefix`.
+	  This is for changing test name prefix. (For testing)
+
+	* test/testunit/tests_for_parallel/ptest_first.rb: Renamed from
+	  test_first.rb
+
+	* test/testunit/tests_for_parallel/ptest_second.rb: Renamed from
+	  test_second.rb
+
+	* test/testunit/tests_for_parallel/ptest_third.rb: Renamed from
+	  test_third.rb
+
+	* test/testunit/tests_for_parallel/ptest_forth.rb: Renamed from
+	  test_forth.rb
+
+	* test/testunit/tests_for_parallel/runner.rb: Remove misc.rb
+
+	* test/testunit/tests_for_parallel/ptest_first.rb: ditto.
+
+	* test/testunit/tests_for_parallel/ptest_second.rb: ditto.
+
+	* test/testunit/tests_for_parallel/ptest_third.rb: ditto.
+
+	* test/testunit/tests_for_parallel/ptest_forth.rb: ditto.
+
+	* test/testunit/tests_for_parallel/misc.rb: Removed because no longer
+	  needed.
+
+	* test/testunit/test_parallel.rb: Fix assertions for above.
+
+Sun Jul 10 16:57:08 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_throw): check a class frame.
+	  Fixes Bug #4648.
+	  The patch is contributed by Kazuki Tsujimoto.
+
+	* bootstraptest/test_proc.rb: add tests for above.
+
+Sun Jul 10 17:28:01 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (mutex_debug): use exit(EXIT_FAILURE) instead of
+	  exit(1).
+	* thread_pthread.c (add_signal_thread_list): ditto.
+	* thread.c (rb_thread_call_with_gvl): ditto.
+	* util.c (Bug): ditto.
+
+Sun Jul 10 15:58:12 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Merge json gem 1.5.4+ (f7f78896607b6f6226cd).
+	  [Bug #4700]
+
+Sun Jul 10 16:41:32 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (typedef struct rb_vm_struct): create a new
+	  'inhibit_thread_creation' field.
+	* thread.c (rb_thread_terminate_all): set inhibit_thread_creation.
+	* thread.c (thread_s_new): don't permit to create new thread
+	  if the VM is under destruction. Otherwise evil finalizer code
+	  can make SEGV. [Bug #4992][ruby-core:37858]
+
+	* bootstraptest/test_objectspace.rb: new test for this fix.
+
+Sun Jul 10 16:06:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (sigsegv): use abort() instead of exit() when nested
+	  SEGV was happen. Because unnested SEGV use abort().
+	  [Bug #5013][ruby-dev:44078]
+
+Sun Jul 10 15:30:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* load.c (rb_f_autoload): prevent to autoload for singleton
+	  classes.  fixes [Bug #4886] [ruby-dev:43816]
+
+	* bootstraptest/test_autoload.rb: add tests for the above change.
+
+Sun Jul 10 15:09:17 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/assertions.rb: Import documentation patch by Justin
+	  Collins. [ruby-core:37225] [Feature #4903]
+
+Sun Jul 10 14:57:36 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: canonicalizes nth and sf.
+
+Sun Jul 10 14:13:50 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h (rb_thread_call_with_gvl, rb_thread_call_without_gvl):
+	  make them visible as experimental C APIs.  fixes Feature #4328.
+
+Sun Jul 10 12:18:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_power): support non-integral
+	  exponent.  fixes [Bug #3271]
+
+	* ext/bigdecimal/bigdecimal.c (rmpd_power_by_big_decimal): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_power_op): add a function to
+	  only use for "**" operator.
+
+	* test/bigdecimal/test_bigdecimal.rb: add a bunch of tests for the
+	  above changes.
+
+	* ext/bigdecimal/bigdecimal.c (is_integer): add an utility function.
+
+	* ext/bigdecimal/bigdecimal.c (is_negative): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (is_positive): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (is_zero): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (is_one): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (is_even): ditto.
+
+Sun Jul 10 12:08:39 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (when_vals): when a string literal is written on when
+	  clause, skip string creation to make it faster.  [ruby-dev:44068]
+	  [Feature #5000]
+
+Sun Jul 10 11:35:29 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (reduce_nodes_gen): NODE_RETURN in rescue body must not be
+	  reduced when there is an else clause.  This caused bizarre behavior
+	  in [Bug #4473] [ruby-core:35629] [ruby-core:37884].
+
+Sun Jul 10 09:46:14 2011  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* range.c (range_max): fix behavior with excluded end value.
+	  [Bug #4591]
+
+Sun Jul 10 09:13:18 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS:  Fix RubyGems version.  [Ruby 1.9 - Bug #5004]
+
+Sat Jul  9 20:01:59 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* internal.h: rb_rational_reciprocal is defined in rational.c.
+
+Sat Jul  9 19:48:31 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* internal.h: added declarations.
+	* complex.c: followed the above change.
+
+Sat Jul  9 17:24:41 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* NEWS: bigdecimal is not a builtin.
+
+Sat Jul  9 17:17:53 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: some improvements for performance.
+
+Sat Jul  9 16:56:01 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* atomic.h (ATOMIC_OR): _InterlockedOr is unavailable in VC6.
+
+	* numeric.c (ULLONG_MAX): fallback definition.
+
+Sat Jul  9 15:59:19 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_{read,write}): should be signed.
+	  Bug #5001
+
+Sat Jul  9 14:02:20 2011  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb: fix multipart form parsing bug. [Bug #3866]
+
+Sat Jul  9 11:41:03 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Vector#normalize [ruby-dev:43829]
+
+Sat Jul  9 09:25:06 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* enumerator.c:  Remove "enumeration sequenced by".
+	  [Ruby 1.9 - Bug #4757]
+
+Sat Jul  9 09:14:56 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* io.c:  Note that methods other than IO#gets may increase IO#lineno.
+	  [Ruby 1.9 - Bug #4902]
+
+Sat Jul  9 08:39:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RUN_OPT): disable gems.
+
+Sat Jul  9 08:37:05 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_close): close(2) on a fd which is being read by
+	  another thread causes deadlock on Mac OS X 10.5
+
+Fri Jul  8 21:20:39 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: use USE_ELF instead of __ELF__ because Solaris
+	  doesn't define it. USE_ELF is already provided by configure.
+	  patched by Naohisa Goto. [ruby-dev:44066] [Bug #4998]
+
+	* addr2line.h: ditto.
+
+	* vm_dump.c: ditto.
+
+Fri Jul  8 16:40:38 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/parser/parser.rl (convert_encoding): should not modify
+	  the argument.
+
+	* ext/json/parser/parser.rl (convert_encoding): no needs to use
+	  force_encoding.
+
+Fri Jul  8 15:53:31 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_bug): get rid of segfault after all threads
+	  disposed.
+
+Fri Jul  8 15:01:06 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl.h: include openssl/e_os2.h before checking the
+	  definition of OPENSSL_SYS_WIN32.
+
+Fri Jul  8 14:40:39 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (wunlink): reverted a part of r32426. it was mistakenly
+	  mixed.
+
+Fri Jul  8 14:29:47 2011  Narihiro Nakamura  <authornari@gmail.com>
+
+	* configure.in: can't subtract void *.
+
+Fri Jul  8 14:33:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): define GC_MARK_STACKFRAME_WORD.
+	  fixed build problem of r32438.  the value (30) is temporary value.
+	  maybe it's enough by 20~24 according to my observation.
+
+Fri Jul  8 13:47:39 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_unlock_all): folded into
+	  rb_threadptr_unlock_all_locking_mutexes.
+	* thread.c (rb_threadptr_unlock_all_locking_mutexes) ditto.
+
+Fri Jul  8 13:36:02 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_unlock_all_locking_mutexes): rename to
+	  rb_threadptr_unlock_all_locking_mutexes and remove static.
+	* vm_core.h: add rb_threadptr_unlock_all_locking_mutexes declaration.
+	* thread.c (thread_start_func_2): adjust the above rename.
+
+	* eval.c (ruby_cleanup): call rb_threadptr_unlock_all_locking_mutexes
+	  again after finalizer. [Bug #4988] [ruby-dev:44049]
+
+Fri Jul  8 13:06:09 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c (FIBER_MACHINE_STACK_ALLOCATION_SIZE): Fiber stack size
+	  don't need to keep multiple number of sizeof(VALUE).
+
+Fri Jul  8 11:39:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (sudo-precheck): true command is not standard on
+	  Windows.
+
+Fri Jul  8 10:39:52 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (gvl_destroy): fix cond_t leak.
+
+Fri Jul  8 09:17:59 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* gc.c:  Improve documentation
+
+Thu Jul  7 23:35:31 2011  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: change water_mark value that may call
+	  gc_mark(lev <= GC_LEVEL_MAX) in gc_mark().
+	  In ruby_stack_check(), water_mark is a value that may call some
+	  C function. Fixes Bug #3781
+
+	* configure.in: define GC_MARK_STACKFRAME_WORD that approximate
+	  size of gc_mark() and gc_mark_children() stackframes.
+
+Thu Jul  7 17:55:05 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/testunit/test_parallel.rb (TestParallelWorker#teardown): wait
+	  the child process even if the communication pipe is broken.
+
+Thu Jul  7 15:44:42 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_set_index, rb_enc_associate_index): should
+	  check if frozen.
+
+	* parse.y (rb_intern3), ruby.c (process_options, ruby_script):
+	  defer freezing after associating encodings.
+
+Thu Jul  7 15:16:51 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (rb_num2ull): use own switch sentence.
+	  Current implementation can't convert 18446744073709551615.
+
+Thu Jul  7 06:56:15 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c (FIBER_STACK_FLAGS): workaround fix for r32420 on FreeBSD.
+
+Thu Jul  7 06:46:12 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* benchmark/driver.rb:  Add difference column to report that averages
+	  across all runs of a benchmark.  [Ruby 1.9 - Feature #4982]
+
+Thu Jul  7 06:19:38 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Reduce requires to improve `make benchmark`.
+	  [#4962]
+	* lib/rubygems/specification.rb:  Delay initialization of rubygems
+	  until require is called.
+
+Thu Jul  7 04:31:26 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (arg): rescue_mod is in inverse order from other
+	  modifiers.  patched by michael.j.edgar AT dartmouth.edu at
+	  [ruby-core:36248].  fixed #4716.
+
+Thu Jul  7 00:40:16 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (kill): check that the process exited or not before
+	  terminating it. [Bug #4943]
+
+Wed Jul  6 23:13:19 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (opt_call_args): allow trailing comma after assoc
+	  argument e.g. 'foo(bar:1,)'.  fixed #3456
+
+Wed Jul  6 22:11:12 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/cgi/test_cgi_header.rb(test_cgi_header_nph): Validate date in
+	  "Date:" header
+
+Wed Jul  6 21:29:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c (fiber_machine_stack_alloc): cleanup pointer arithmetic.
+	  "size/sizeof(VALUE)" is ugly and easy confusing.
+	* cont.c (fiber_initialize_machine_stack_context): ditto.
+
+Wed Jul  6 21:24:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c (fiber_machine_stack_alloc): fix mprotect misuse. A stack
+	  guard page should have PROT_NONE.
+	* cont.c (fiber_initialize_machine_stack_context):
+	  th->machine_stack_maxsize shouldn't be included guard pages size.
+	  [Bug #4983][ruby-dev:44043]
+
+Wed Jul  6 21:23:38 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c (fiber_machine_stack_alloc): use MAP_STACK if it's provided.
+
+Wed Jul  6 21:22:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c (fiber_machine_stack_alloc): use MAP_FAILED instead of -1.
+
+Wed Jul  6 21:21:05 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c (fiber_machine_stack_alloc): remove unnecessary cast.
+
+Wed Jul  6 18:10:13 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/cgi/test_cgi_header.rb(test_cgi_header_nph): Adding space after
+	  comma.
+
+	* test/cgi/test_cgi_header.rb(test_cgi_header_nph): Remove variable
+	  `now`. Suppress warning.
+
+Wed Jul  6 12:18:09 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/cgi/test_cgi_header.rb(test_cgi_header_nph): Fix bug depends to
+	  time. The test fails if time past 1 second in line 136-145
+
+Tue Jul  5 15:28:04 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_here_document): should dispatch heredoc_end
+	  scanner event on an empty here document.  fixed Bug#4543.
+
+Tue Jul  5 13:49:26 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* addr2line.c: fix r32407 to check HAVE_ALLOCA_H.
+
+Tue Jul  5 14:05:43 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httpauth/digestauth.rb (_authenticate):
+	  Literal texts in HTTP ABNF is case-insensitive (RFC2616 2.1),
+	  and a sample implementation in RFC2617 also ignores the case
+	  of algorithms. So now this ignores those cases.
+	  [ruby-dev:43965] [Feature #4936]
+
+	* lib/webrick/httpauth/digestauth.rb (initialize):
+	  Because of above, opera_hack is useless and removed.
+
+Tue Jul  5 01:30:01 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread_pthread.c (native_sleep): cut the waiting time up to
+	  100,000,000 because Solaris cond_timedwait() return EINVAL if an
+	  argument is greater than current_time + 100,000,000.  This is
+	  considered as a kind of spurious wakeup.  The caller to native_sleep
+	  should care about spurious wakeup.
+
+Tue Jul  5 01:24:26 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c: disable FIBER_USE_NATIVE on Solaris because resuming any
+	  Fiber caused SEGV.  I haven't follow up the issue deeply, but it
+	  works when disabling the feature.
+
+Tue Jul  5 01:22:46 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* addr2line.c: include <alloca.h> to fix a build issue on Solaris.
+
+Tue Jul  5 00:49:05 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/coverage/coverage.c: resurrect r32071 + add GC guard for
+	  rb_coverages. [ruby-core:37352] [Bug #4927]
+	  [ruby-core:36539] [Feature #4796]
+
+	* test/coverage/test_coverage.rb resurrect r32071.
+
+Mon Jul  4 22:24:46 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (get_stack): For NetBSD/FreeBSD, use
+	  pthread_attr_getstack() if possible. and, remove an assumption
+	  of stack growing direction.
+
+Mon Jul  4 20:42:31 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/coverage/coverage.c: revert r32071.  The commit caused SEGV on
+	  some minor nonfree OS.  I have no means of debugging the bug.  My
+	  personal opinion is that such OS should be unsupported unless there
+	  is an active maintainer.  [ruby-core:37352]
+
+	* test/coverage/test_coverage.rb: ditto.
+
+Mon Jul  4 07:14:12 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (get_stack): the return address of get_stack
+	  must be the highest address of the current thread's stack.
+
+Mon Jul  4 06:37:22 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/intern.h, thread_pthread.c (rb_reserved_fd_p,
+	  RB_RESERVED_FD_P): added.  This C API is to limit to access
+	  fds which are used by RubyVM internal.  In this version of
+	  CRuby, return 1 if fd is communication pipe.
+	  If your application needs to close all file descriptors to
+	  present resource leak, skip internal fds using this C API.
+	  We also define a macro RB_RESERVED_FD_P(fd).  So you can write
+	    #ifndef RB_RESERVED_FD_P
+	    #define RB_RESERVED_FD_P(fd) 0
+	    #endif
+	  for Ruby 1.9.2 or previous version to write compatible extensions.
+	  See [ruby-core:37727]
+
+	* thread_win32.c (rb_reserved_fd_p): added (return 0 for any fds).
+
+	* io.c (rb_io_initialize): raise ArgumentError if given fd is reserved by Ruby.
+
+Sun Jul  3 23:43:56 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ext/extmk.rb (extmake): suppresses outputs from extconf.rb.
+	  (extmake) warns a failure in extconf.rb.
+
+Sun Jul  3 13:44:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_reject_bang): should not remove elements which are
+	  not yielded.  [Bug #2545]
+
+Sun Jul  3 06:10:26 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (get_stack): pthread_attr_getstack() doesn't
+	  return stack start address, but stack base address. Thus,
+	  we need to add stack size for getting stack start address.
+	  And, we don't have to decrease guard size twice.
+	* thread_pthread.c (thread_start_func_1): don't use inaccurate
+	  stack start guess if native_thread_init_stack() can be used.
+	  [Bug #1813] [ruby-core:24540]
+
+Sun Jul  3 04:50:08 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (get_stack): add to a care of guard page on Mac
+	  OS X. [Bug #1813] [ruby-core:24540]
+	* signal.c (ruby_signal): SIGBUS use alternative stack too.
+	* signal.c (sigbus): On Mac, thread stack overflow makes SIGBUS
+	  instead of SIGSEGV. thus, added stackoverflow check.
+	* signal.c (default_handler): get rid of compilation warning.
+	* signal.c (Init_signal): ditto.
+
+Sat Jul 02 08:59:20 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_ocsp.rb
+	* test/openssl/test_x509_cert.rb: Perform SHA-256 tests only if
+	  supported by the available OpenSSL version. Versions < 0.9.8 don't
+	  support it. [ruby-core:37724]
+
+Sat Jul  2 07:17:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_reject_bang, rb_ary_delete_if): rejected
+	  elements should be removed.  fixed [Bug #2545]
+
+Sat Jul  2 01:57:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: remove a description of Kernel#respond_to? because it has
+	  been reverted at revision 28564.
+
+Sat Jul  2 00:58:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: describe a change of multiplication of Bignum.
+
+Fri Jul  1 18:52:31 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_app_erb.rb: increase loop count. too short
+	  measurement time makes less accuracy.
+	* benchmark/bm_app_factorial.rb: ditto.
+	* benchmark/bm_app_mandelbrot.rb: ditto.
+	* benchmark/bm_app_strconcat.rb: ditto.
+
+	* benchmark/bm_io_file_create.rb: ditto.
+	* benchmark/bm_io_file_read.rb: ditto.
+	* benchmark/bm_io_file_write.rb: ditto.
+
+	* benchmark/bm_so_concatenate.rb: ditto.
+	* benchmark/bm_so_lists.rb: ditto.
+	* benchmark/bm_so_matrix.rb: ditto.
+	* benchmark/bm_so_random.rb: ditto.
+	* benchmark/bm_so_sieve.rb: ditto.
+
+	* benchmark/bm_vm_thread_mutex1.rb: ditto.
+	* benchmark/bm_vm_thread_mutex2.rb: ditto.
+	* benchmark/bm_vm_thread_mutex3.rb: ditto.
+
+	* benchmark/bm_vm1_block.rb: cleanup.
+	* benchmark/bm_vm1_const.rb: cleanup.
+	* benchmark/bm_vm1_ensure.rb: cleanup.
+	* benchmark/bm_vm1_ivar.rb: cleanup.
+	* benchmark/bm_vm1_length.rb: cleanup.
+	* benchmark/bm_vm1_neq.rb: cleanup.
+	* benchmark/bm_vm1_not.rb: cleanup.
+	* benchmark/bm_vm1_rescue.rb: cleanup.
+	* benchmark/bm_vm1_simplereturn.rb: cleanup.
+	* benchmark/bm_vm1_swap.rb: cleanup.
+
+	* benchmark/bm_vm2_array.rb: cleanup.
+	* benchmark/bm_vm2_case.rb: cleanup.
+	* benchmark/bm_vm2_defined_method.rb: cleanup.
+	* benchmark/bm_vm2_eval.rb: cleanup.
+	* benchmark/bm_vm2_method.rb: cleanup.
+	* benchmark/bm_vm2_mutex.rb: cleanup.
+	* benchmark/bm_vm2_poly_method.rb: cleanup.
+	* benchmark/bm_vm2_poly_method_ov.rb: cleanup.
+	* benchmark/bm_vm2_proc.rb: cleanup.
+	* benchmark/bm_vm2_regexp.rb: cleanup.
+	* benchmark/bm_vm2_send.rb: cleanup.
+	* benchmark/bm_vm2_super.rb: cleanup.
+	* benchmark/bm_vm2_unif1.rb: cleanup.
+	* benchmark/bm_vm2_zsuper.rb: cleanup.
+
+	* benchmark/bm_vm_thread_alive_check1.rb: cleanup.
+
+Fri Jul  1 15:23:00 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix: Add LUP decomposition
+
+Fri Jul  1 15:21:14 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Allow non integer exponents for Matrix#**
+
+Fri Jul  1 15:13:25 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix: Add Eigenvalue Decomposition
+
+Fri Jul  1 15:10:22 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix: Add Matrix#round
+
+Fri Jul  1 11:41:12 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_trans): free heap ptr when the str is not embedded.
+	  patched by Eric Wong. [Bug #4956] [ruby-core:37708]
+
+Fri Jul  1 11:07:43 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (do_select): fix memory leak.
+	  Patch by Eric Wong. Thank you! [Bug #4953] [ruby-core:37702]
+
+Fri Jul  1 08:21:28 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_getivar): check vm state version
+	  to invalidate inline cache (ivar index).
+	  fixes Bug #4926.
+
+	* vm_insnhelper.c (vm_setivar): ditto.
+
+Fri Jul  1 08:03:15 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c, thread_pthread.c (WRITE_CONST): suppress warnings
+	  `ignoring return value'.
+
+Fri Jul  1 06:41:36 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_threadptr_check_signal): only wake up main thread.
+
+	* thread.c (rb_threadptr_execute_interrupts_common): check signal
+	  delivery if it is main thread.
+	  fixes [ruby-dev:44005] [Ruby 1.9 - Bug #4950]
+
+	* bootstraptest/test_fork.rb: add a test for above.
+
+	* signal.c (rb_get_next_signal): skip if signal_buff is empty.
+	  (check signal_buff.size first)
+
+	* vm_core.h: remove unused variable rb_thread_t::exec_signal.
+
+	* thread.c (rb_thread_check_trap_pending): check
+	  rb_signal_buff_size() because rb_thread_t::exec_signal
+	  is no longer available.
+
+Fri Jul  1 03:28:25 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c (Init_class_hierarchy): should name BasicObject
+	  explicitly.
+
+	* variable.c (rb_const_defined_0): should not check for
+	  superclasses as const_get.
+
+Fri Jul  1 03:24:03 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: mathn is still alive (should die soon).
+
+Thu Jun 30 23:50:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* misc/ruby-mode.el (ruby-indent-beg-re): Fix broken regular
+	  expression.  Fixes #4546
+
+Thu Jun 30 23:43:30 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl.c/.h: Added ossl_x509_name_sk2ary.
+	* ext/openssl/ossl.c: Replaced ossl_x509_ary2k by generic macro to
+	  simplify future conversions.
+	* ext/openssl/ossl_ssl.c: Implement SSLSocket#client_ca.
+	* test/openssl/test_ssl.rb: Add test for SSLSocket#client_ca.
+	  Thanks to Ippei Obayashi for providing the patch!
+	  [ Ruby 1.9 - Feature #4481 ] [ruby-core:35461]
+
+Thu Jun 30 22:38:58 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm2_defined_method.rb: added to measure performance of
+	  bmethod (method defined by define_method()).
+
+Thu Jun 30 22:17:04 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_bmethod): fix to hook call/return event
+	  for methods defined by define_method().  fixes Bug #4613.
+
+	* thread.c (call_trace_proc): Fix to skip if class is not given (0).
+	  Note that ID and Class object are passed for call/return event
+	  if the called method was defined by define_method().
+	  If you are author of tracer/profiler/debugger, this may be an
+	  important change.  You should check passed class as zero or
+	  non-zero instead of checking the event type.
+
+	* test/ruby/test_settracefunc.rb: add a test for above.
+
+Thu Jun 30 21:18:35 2011  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* configure.in: Add warnflags for XL/C on AIX during configure
+	  to avoid [Bug #3971]. See [ruby-core:32859]
+
+Thu Jun 30 21:16:04 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (m_ajd): refers a constant.
+
+Thu Jun 30 20:54:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	  * ext/io/console/io-console.gemspec: spin-off gem for 1.9.2.
+
+Thu Jun 30 20:36:23 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: trivial changes.
+
+Thu Jun 30 20:12:36 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_module.rb: tests for [Bug #3422] and [Bug #3423].
+
+Thu Jun 30 20:03:39 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: modified doc.
+
+Thu Jun 30 19:09:19 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (thread_timer): ignore unknown errno.
+	  (we observed that select(2) was canceled by errno=514 on
+	  boron == Linux/Xen environment)
+
+Thu Jun 30 17:33:25 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (ObjectSpace.count_tdata_objects):
+	  Fix rdoc.  Fixes Bug #3892.
+
+	* ext/objspace/objspace.c (ObjectSpace.count_tdata_objects):
+	  Change key type if the klass of a object is zero (internal object).
+	  Read rdoc for details.
+
+	* internal.h: export rb_objspace_data_type_name().
+
+Thu Jun 30 17:25:08 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (ping_signal_thread_list, thread_timer):
+	  fix to keep polling state if there are any ping-tasks.
+
+Thu Jun 30 12:25:34 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): allocate
+	  machine stack for the timer thread at least 12KB.  FreeBSD 8.2
+	  AMD64 causes machine stack overflow (SIGSEGV) only with
+	  PTHREAD_STACK_MIN (maybe defined as 2KB).
+
+Thu Jun 30 09:36:37 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/weakref.rb:  Attach documentation to WeakRef and add missing
+	  documentation
+
+Thu Jun 30 09:30:14 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/yaml.rb:  Document toplevel YAML and YAML::ENGINE to describe
+	  Psych vs Syck engines.
+
+Thu Jun 30 09:21:52 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/cmath.rb:  Hide handle_no_method_error from RDoc.
+	* error.c:  Document or hide undocumented error classes.
+
+Thu Jun 30 07:49:04 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* hash.c:  Document ENV
+
+Thu Jun 30 06:37:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): ensure RUBYLIB_PREFIX stored
+	  before RUBYLIB, even if MANGLED_PATH is enabled.  fixed #1679.
+	  MANGLED_PATH is disabled by the default and will be removed
+	  completely in the future.
+
+Thu Jun 30 06:32:21 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/drb/drb.rb:  Hide deprecated toplevel DRb constants.
+
+Thu Jun 30 06:17:02 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cmath.rb (CMath.log): second argument: b can be nil.
+
+Thu Jun 30 06:23:28 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* thread.c (ruby_thread_s_pass):  Fix typo.
+
+Thu Jun 30 06:16:53 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 3.8 which contains fixes for documentation
+	  in trunk.
+
+Thu Jun 30 02:53:26 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_threadptr_execute_interrupts_common): remove
+	  meaningless native_thread_yield(). It never close a race.
+
+Thu Jun 30 02:41:47 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_schedule_limits): minor optimization.
+	  eliminate machine context saving when running time is enough small.
+
+Thu Jun 30 02:28:11 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_schedule_rec): move interrupt_flag check to
+	  rb_thread_schedule().
+	  And also rename to rb_thread_schedule_limits() and remove
+	  sched_depth argument. It's no longer called recursive.
+	* thread.c (rb_thread_schedule): add to check interrupt_flag as
+	  above explained.
+
+	* thread.c (rb_threadptr_execute_interrupts_rec): rename to
+	  rb_threadptr_execute_interrupts_common() and remove sched_depth
+	  argument. It's no longer called recursive.
+
+	* thread.c (rb_thread_sleep): adapt the renaming.
+	* thread.c (rb_threadptr_execute_interrupts): ditto.
+	* thread.c (rb_thread_execute_interrupts): ditto.
+
+Thu Jun 30 01:31:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_s_pass): change RDoc description and remove
+	  a sample code. The actual implementation never behave as explained by
+	  an example. It's a documentation bug.
+
+Thu Jun 30 00:54:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_stop): change RDoc sample code. The old
+	  example is buggy and may cause deadlock. The patch is
+	  suggested by Heesob Park <phasis@gmail.com>. Thank you!
+	  [Bug #3606][ruby-core:31454]
+
+Thu Jun 30 00:49:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_wakeup): change RDoc sample code. The old
+	  example is buggy and may not display anything by a race.
+	  The patch is suggested by Heesob Park <phasis@gmail.com>.
+	  Thank you! [Bug #3606][ruby-core:31454]
+
+Thu Jun 30 00:43:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_run): change RDoc. The old example is buggy
+	  and may cause deadlock. The patch is suggested by Heesob Park
+	  <phasis@gmail.com>. Thank you! [Bug #3606][ruby-core:31454]
+
+Thu Jun 30 00:30:15 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/cmath.rb: make same exception for Math. fix [Bug #3137].
+
+Thu Jun 30 00:03:20 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/completion.rb: complement correctly string literal. fix
+	  [Bug #1145].
+
+Wed Jun 29 23:42:51 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: avoided using timev.
+	* ext/date/date_strftime.c: ditto.
+	* ext/date/date_tmx.h: ditto.
+
+Wed Jun 29 23:17:57 2011  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/openssl/ossl.h (OPENSSL_SYS_WIN32): support for mingw(msys).
+
+Wed Jun 29 23:09:14 2011  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/tk/extconf.rb (intptr_t, uintptr_t): support for the latest ActiveTcl with mingw.
+
+Wed Jun 29 22:49:10 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/cmd/help.rb: support RDoc 3.7. fix [Bug #3760].
+
+Wed Jun 29 22:04:14 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* lib/tracer.rb: Tracer.on only if required by -r command-line option.
+	  and consider --disable-gems option.
+	* test/test_tracer.rb: add tests for it.
+
+Wed Jun 29 13:55:36 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_const_get_0): should not look for superclasses if
+	  the second optional argument is given for #const_get().
+	  fix [Bug #3422] [Bug #3423]
+
+Wed Jun 29 12:07:27 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* math.c:  Attach documentation for Math.
+	* object.c:  Document NIL, TRUE, FALSE.
+	* io.c:  Improve grammar in ARGF comment.  Document STDIN/OUT/ERR.
+	  Document ARGF global constant.
+	* lib/rake:  Hide deprecated toplevel constants from RDoc (import from
+	  rake trunk).
+	* lib/thwait.rb:  Document ThWait.
+	* lib/mathn.rb:  Hide Math redefinition from RDoc
+	* lib/sync.rb:  Add a basic comment for Sync_m, Synchronizer_m, Sync,
+	  Synchronizer.
+	* parse.y:  Document SCRIPT_LINES__.
+	* hash.c:  Document ENV class and global constant.
+	* vm.c:  Document TOPLEVEL_BINDING.
+	* version.c:  Document RUBY_* constants.
+	* ruby.c:  Document DATA and ARGV.
+
+Wed Jun 29 10:13:12 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Matrix.zero can build rectangular matrices.
+	  Vector#r should be called #magnitude
+
+Wed Jun 29 10:11:08 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Matrix#diagonal?, hermitian?, normal?,
+	  orthogonal?
+	  permutation?, symmetric?, {lower|upper}triangular?, unitary?, zero?
+
+Wed Jun 29 10:09:35 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Specialize Matrix#find_index to return [row, col]
+	  and accept the same optional argument as #each
+
+Wed Jun 29 10:07:32 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Matrix#each{_with_index} can iterate over a subset
+	  of the Matrix
+
+Wed Jun 29 06:21:02 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (native_stop_timer_thread): skip to close
+	  communication pipe to avoid timing bug (process termination timing).
+	  The communication pipe will closed by OS.
+
+Wed Jun 29 06:09:54 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c (rb_async_bug_errno): async-safe bug report function.
+	  In timer thread, signal handler should use it.
+	  The patch is contributed by Eric Wong <normalperson@yhbt.net>.
+	  Refs: [ruby-core:37644] and [ruby-core:37647]
+
+	* thread_pthread.c: use rb_async_bug_errno().
+	  And replace all fprintf() to write().
+
+	* internal.h (rb_async_bug_errno): add decl. of above func.
+
+Tue Jun 28 23:46:08 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/tracer.rb: count only non-internal libraries in stack trace,
+	  ignoring custom_require.
+
+Tue Jun 28 21:44:58 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ruby-lex.rb: recognize '\char' in ruby statement.
+
+Tue Jun 28 20:39:29 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* lib/debug.rb (var_list): Command 'var *' did not work on 1.9(!).
+	  global_variables, local_variables, and instance_variables returns
+	  Symbols from 1.9 and need to stringify before evaling it.
+	  See #4931.
+
+Tue Jun 28 19:23:01 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c: Now, USE_TRAP_SIGMASK depend on HAVE_PTHREAD_SIGMASK.
+	  The code have already depended on pthread_sigmask since r27464.
+
+Tue Jun 28 15:09:21 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/benchmark.rb: merge eregon/benchmark.
+	  https://github.com/eregon/ruby/tree/benchmark
+	  patched by Benoit Daloze. [ruby-core:37593] [Bug #4940]
+
+	* lib/benchmark (Benchmark#bmbm): bmbm should be consistent with bm
+	  for the return value.
+
+	* test/benchmark: remove preemptive test instead of skipping
+	  I removed the preemptive test I wrote for Feature #4197.
+	  I'll add it back when the implementation will be able to satisfy it.
+
+	* lib/benchmark (Benchmark#bmbm): remove useless explicit call,
+	  #format is an alias of #to_s test/benchmark: add a test for
+	  format of long time.
+
+	* lib/benchmark: fix label width: always add 1 to ensure there is a
+	  space delimiter even with times over 100s
+	  When I asked for Feature #4197, I wanted to make delimiting spaces
+	  consistent for #bm and #bmbm.
+	  But with times over 100s, the output contains no space between the
+	  label and the first time (user).
+	  Now both ensure there is always a space, even if that means 3 spaces
+	  with times under 10s (because it is formatted as %10.6f)
+
+	* test/benchmark: let labels be a constant
+	  lib/benchmark (Benchmark#realtime): avoid creating an unused Proc
+	  lib/benchmark (Benchmark#benchmark): use ensure clause to restore
+	  STDOUT.sync, as in #bmbm
+
+Tue Jun 28 13:41:51 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (native_stop_timer_thread): fixed commit miss of
+	  r32244.  grep sources before changing the signature of a function.
+
+Tue Jun 28 11:49:14 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (consume_communication_pipe):
+	  Make "buff" as static.  (Maybe) "buff" can be shared between
+	  any caller (any threads) because no one use the read values.
+	  "buff" (1024 byte) on stack may cause stack overflow on
+	  several environment (we found a crash on FreeBSD).
+	  And remove const value "buff_size", and define CCP_READ_BUFF_SIZE
+	  macro.
+
+Tue Jun 28 11:45:30 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake:  Update rake to fix some bugs and hide deprecated features
+	  from RDoc.
+	* lib/rake/version.rb:  Bump version to 0.9.2.1 to distinguish it from
+	  the released version.
+	* NEWS:  ditto
+
+Tue Jun 28 11:17:28 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 3.7 (final)
+	* NEWS:  ditto
+
+Tue Jun 28 10:18:42 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c (rb_daemon): fix wrong #endif position.
+
+Tue Jun 28 07:50:32 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (Init_Object):  Teach RDoc what Init_class_hierarchy does to
+	  hook up ri for BasicObject, Object, Module and Class.
+
+Tue Jun 28 05:03:32 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* thread.c (rb_thread_local_aref): RDoc fix. Thread#[] example
+	  had a race. See #4480.
+
+Tue Jun 28 01:22:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_log): move BigMath.log from
+	  bigdecimal/math.rb.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb: ditto.
+
+	* test/bigdecimal/test_bigdecimal.rb: move test for BigMath.log from
+	  test/bigdecimal/test_bigmath.rb.
+
+	* test/bigdecimal/test_bigmath.rb: ditto.
+
+Tue Jun 28 01:19:52 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ruby-lex.rb: fix [Bug #4232].
+
+Tue Jun 28 00:14:13 2011  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: fix [Bug #4409]. add DRbServer#here?.
+
+	* test/drb/test_drb.rb: ditto.
+
+	* test/drb/drbtest.rb: ditto.
+
+	* test/drb/ut_eq.rb: ditto.
+
+Tue Jun 28 00:08:43 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/workspace.rb: fix BUG#4793.
+
+Mon Jun 27 22:06:12 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (consume_communication_pipe): don't use C99
+	  style variable length array.
+
+Mon Jun 27 22:04:27 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (consume_communication_pipe): change return
+	  type to void. caller doesn't use it.
+
+Mon Jun 27 21:29:50 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.h (rb_global_vm_lock_struct): add volatile to
+	  gvl->waiting. now thread_timer() access it w/o lock.
+
+Mon Jun 27 21:16:11 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c: s/__gvl_acquire/gvl_acquire_common/ and
+	  s/__gvl_release/gvl_release_common/.
+
+Mon Jun 27 11:41:47 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (rb_thread_create_timer_thread):
+	  the type of return value of write(2) is ssize_t.
+
+Mon Jun 27 09:57:02 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (rb_thread_create_timer_thread):
+	  Fixes missing initialization of oflags.
+
+Mon Jun 27 09:07:42 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c: Stop polling in the timer thread when there are
+	  no waiting thread.  If there are 2 or more runnable threads,
+	  the timer thread does polling.  Avoid polling makes power save
+	  for several computers (0.2W per a Ruby process, when I measured).
+	  If outside-event such as signal or Thread#kill was occurred
+	  when the timer thread does not do polling, then wake-up
+	  the timer thread using communication-pipe (the timer thread
+	  waits this communication-pipe with select(2)).
+	  The discussion about this modification can be found from the post
+	  [ruby-core:33456] and other related posts.
+	  Note that Eric Wong and KOSAKI Motohiro give us the huge
+	  contributions for this modification.  Thanks.
+
+	* thread_pthread.c (rb_thread_wakeup_timer_thread): add a function.
+	  This function wakes up the timer thread using communication-pipe.
+
+	* thread.c (rb_thread_stop_timer_thread): add a parameter which
+	  specify closing communication-pipe or not.
+
+	* thread.c (rb_thread_terminate_all): do not stop timer thread here
+	  (ruby_cleanup() terminate timer thread).
+
+	* signal.c: wake up timer thread using
+	  rb_thread_wakeup_timer_thread() from signal handler.
+
+	* eval.c (ruby_cleanup): use rb_thread_stop_timer_thread(1).
+
+	* process.c: use rb_thread_stop_timer_thread(0)
+	  (reuse communication-pipe).
+
+	* thread_win32.c (rb_thread_wakeup_timer_thread): add a dummy
+	  function.
+
+	* vm_core.h: add and fix decl. of functions.
+
+Mon Jun 27 08:01:19 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c: should use ALLOCA_N.
+
+Mon Jun 27 01:34:18 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/etc/test_etc.rb (TestEtc#test_get{pw,gr}nam): skip entries
+	  start with + sign, which means NIS.  these are returned in the
+	  case that passwd and group entries in /etc/nsswitch.conf are set
+	  to use "nis" explicitly on Debian.  fixed #3683
+
+Mon Jun 27 00:44:53 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (rb_parser_end_seen_p): fix documentation about return
+	  value. patched by Sho Hashimoto. [Bug #4511]
+
+Mon Jun 27 00:40:47 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* hash.c (rb_hash_reject): add documentation that Hash#reject
+	  without block returns enumerator.
+	  patched by Michael Edgar [Bug #4847] [ruby-core:36800]
+
+Sun Jun 26 23:49:21 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/test_switch_hitter.rb: added a test.
+
+ Sun Jun 26 22:21:23 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: refactoring.
+
+Sun Jun 26 18:03:30 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* parse.y: comma at the end of line is no longer allowed.
+	  A patch from Yukihiro Matsumoto  <matz AT ruby-lang.org>.
+	  (fixed #3456).
+
+Sun Jun 26 13:35:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_dump.c (rb_vm_bugreport): change CrashReporter suggestion messages
+	  on Mac. It should be placed after "-- C level backtrace" line.
+	  Suggested by Endoh-san.
+
+	  <before>
+	  -- See Crash Report log file under ~/Library/Logs/CrashReporter or ---------
+	  -- /Library/Logs/CrashReporter, for the more detail of ---------------------
+	  -- C level backtrace information -------------------------------------------
+
+	<after>
+	  -- C level backtrace information -------------------------------------------
+
+	     See Crash Report log file under ~/Library/Logs/CrashReporter or
+	     /Library/Logs/CrashReporter, for the more detail of.
+
+Sun Jun 26 10:08:28 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/extconf.rb
+	* ext/openssl/ossl_missing.h/.c: add ASN1_put_eoc if missing.
+
+	* ext/openssl/ossl_asn1.c: introduce ossl_asn1_object_size and
+	  ossl_asn1_put_object to wrap functionality depending on OpenSSL
+	  version in use.
+	  Fixes	[ Ruby 1.9 - Bug #4916 ] reported by Hiroshi Nakamura.
+	  [ruby-core:37286]
+
+Sun Jun 26 01:00:15 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (date_strftime_internal): removed meaningless braces.
+	* ext/date/date_core.c (gengo): the value should be int.
+
+Sat Jun 25 23:45:30 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* vm_insnhelper.c (vm_search_superclass): avoid control frame
+	  stack overrun. currently super() in Proc created in a method
+	  defined by Module#define_method raise NoMethodError. [Bug #4881]
+	* test/ruby/test_method.rb t_super_in_proc_from_define_method):
+	  add test for it.
+
+Sat Jun 25 23:23:14 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread.c (sleep_forever): now Kernel#sleep don't wakeup by
+	  signal handler execution. [Bug #4072]
+
+Sat Jun 25 23:14:47 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread.c (rb_threadptr_check_signal): remove unnecessary th->status
+	  backup. fix race condition which may results unexpected main thread's
+	  status transition. see #4072
+
+Fri Jun 24 19:57:30 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* lib/webrick/httprequest.rb (setup_forwarded_info): Parsing request
+	  header failed when the request is from 2 or more Apache reverse
+	  proxies. It's said that all X-Forwarded-* headers will contain more
+	  than one (comma-separated) value if the original request already
+	  contained one of these headers.  Since we could use these values as
+	  Host header, we choose the initial(first) value. See #4922.
+
+	* test/webrick/test_httprequest.rb (test_forwarded): Test it.
+
+Fri Jun 24 17:06:33 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_daemon): should not start timer thread
+	  twice.  fixed Bug#4920.
+
+Fri Jun 24 15:54:14 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_shutdown): Try to shutdown SSL
+	  connection more gracefully. Call SSL_shutdown() max 4 times until it
+	  returns 1 (success). Bi-directional SSL close has several states but
+	  SSL_shutdown() kicks only 1 transition per call. Max 4 is from
+	  mod_ssl.c of Apache httpd that says 'max 2x pending + 2x data = 4'.
+	  See #4237.
+
+Fri Jun 24 07:24:37 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake/version.rb:  Fixed VERSION to work with tool/rbinstall.rb
+	* bin/rake:  Import bin/rake from 0.9.2
+	* tool/rbinstall.rb (install):  Rake::VERSION is now in
+	  lib/rake/version.rb.  Fixes `make install`
+
+Fri Jun 24 07:11:37 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake: Import Rake 0.9.2
+
+Fri Jun 24 00:44:15 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (c_valid_{julian,gregorian}_p): fixed the range of month.
+
+Fri Jun 24 00:14:23 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: trivial changes on text.
+
+Thu Jun 23 22:46:57 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_x509name.c: Add X509::Name#hash_old as a wrapper
+	  for X509_NAME_hash_old in OpenSSL 1.0.0. See #4805
+
+	* test/openssl/test_x509name.rb (test_hash): Make test pass with
+	  OpenSSL 1.0.0.
+
+	* NEWS: Add it.
+
+Thu Jun 23 19:30:53 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_set_time): Check
+	  argument type with NUM2LONG if the arg is not a Time object.
+	  See #4919.
+
+	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_set_timeout): Check
+	  type with NUM2LONG. Time as an arg is not allowed. See #4919.
+
+	* test/openssl/test_ssl_session.rb (test_session_time,
+	  test_session_timeout): Test it.
+
+Wed Jun 23 13:30:30 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* signal.c(ruby_atomic_exchange): Fix definition style.
+
+Wed Jun 22 22:34:05 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_session_new_cb): Return 0 to
+	  OpenSSL from the callback for SSL_CTX_sess_set_get_cb().
+	  Returning 0 means to OpenSSL that the session is still valid
+	  (since we created Ruby Session object) and was not freed by us with
+	  SSL_SESSION_free(). Call SSLContext#remove_session(sess) in
+	  session_get_cb block if you don't want OpenSSL to cache the session
+	  internally.
+	  This potential issue was pointed by Ippei Obayashi. See #4416.
+
+	* test/openssl/test_ssl_session.rb (test_ctx_server_session_cb): Test
+	  it.
+
+Wed Jun 22 22:21:17 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl_session.c: Respect T_BIGNUM time values. Patch by
+	  Tomoyuki Chikanaga.
+	  [ Ruby 1.9 - Bug #4919 ] [ruby-dev:43869]
+
+Wed Jun 22 21:29:25 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/socket/depend (SOCK_HEADERS): use $(top_srcdir) instead of
+	  $(topdir). sorry!
+
+Wed Jun 22 19:47:03 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c (cont_capture): add volatile.
+	  On clang -O, it is needed to avoid the optimization.
+	  With this and llvm/clang's recent fix, clang 3.0 can
+	  build ruby-trunk with -O option.
+
+	* cont.c (cont_capture): use for-loop.
+
+	* array.c (rb_ary_each): add volatile and use it.
+
+	* vm_insnhelper.c (vm_call_cfunc): ditto.
+
+Wed Jun 22 18:20:46 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_session_remove_cb):
+	  OpenSSL::SSL::SSLContext#session_remove_cb was broken. It wrongly
+	  tried to call the session_*new*_cb callback.
+
+	* test/openssl/test_ssl_session.rb (class OpenSSL): Test it.
+
+Wed Jun 22 17:37:49 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl.h: Introduced OSSL_BIO_reset macro for PEM/DER
+	  fallback scenarios.
+
+	* ext/openssl/ossl_pkey_dsa.c
+	* ext/openssl/ossl_x509req.c
+	* ext/openssl/ossl_pkey_rsa.c
+	* ext/openssl/ossl_pkey_ec.c
+	* ext/openssl/ossl_ssl_session.c
+	* ext/openssl/ossl_x509crl.c
+	* ext/openssl/ossl_pkey.c
+	* ext/openssl/ossl_pkey_dh.c
+	* ext/openssl/ossl_x509cert.c
+	* ext/openssl/ossl_pkcs7.c: Use OSSL_BIO_reset.
+
+	* ext/openssl/ossl_ssl.c
+	* ext/openssl/ossl_cipher.c
+	* ext/openssl/ossl_pkey_ec.c
+	* ext/openssl/ossl_pkcs12.c
+	* ext/openssl/ossl_ssl_session.c: Replace rb_raise occurrences by
+	  ossl_raise. This automatically flushes OpenSSL's error queue.
+
+	* ext/openssl/ossl_pkcs7.c: Raise error if DER fallback for parsing
+	  fails.
+
+	* test/openssl/test_pkey_ec.rb
+	* test/openssl/test_pkey_dsa.rb
+	* test/openssl/test_pkey_rsa.rb: Add assertions that OpenSSL.errors is
+	  empty.
+
+	* test/openssl/test_pkey_rsa.rb: Remove initial OpenSSL.errors call in
+	  test_new.
+	  [ Ruby 1.9 - Bug #4885 ] [ruby-core:37134]
+
+Wed Jun 22 15:01:24 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: Use SSL_MODE_RELEASE_BUFFERS if available.
+	  Thanks, Eric Wong, for providing the patch.
+	  [ Ruby 1.9 - Feature #4672 ] [ruby-core:36127]
+
+Wed Jun 22 14:47:53 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_buffering.rb
+	* test/openssl/test_pkcs12.rb: Inherit from Test::Unit::TestCase
+	  instead of MiniTest::Unit::TestCase. [ruby-core:37275]
+
+Wed Jun 22 12:41:03 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_to_der):
+	  OpenSSL::SSL::Session#to_der was broken. Fix buffer handling.
+
+	* test/openssl/test_ssl_session.rb (test_session): Test it.
+
+Wed Jun 22 12:38:52 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* test/openssl/test_ssl_session.rb: Split out SSL::Session related
+	  tests from test_ssl.rb
+
+Wed Jun 22 03:20:52 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Fix cyclic references of
+	  objects.  Thanks to CvX for reporting the bug and a test case.
+	* test/psych/test_object.rb: test for cyclic object references.
+
+Wed Jun 22 02:39:54 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP.post_form): Do not ignore query part of
+	  the given URI to post. See #655.
+
+	* test/net/http/test_http.rb, test/net/http/utils.rb: Test it.
+
+Wed Jun 22 01:28:13 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* test/openssl/test_x509store.rb (test_set_errors): Redhat is
+	  distributing a patched version of OpenSSL that allows multiple CRL
+	  for a key (multi-crl.patch.) Make test pass on such env. See #4122,
+	  #4554.
+
+Tue Jun 21 21:50:37 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* lib/webrick/httpresponse.rb (HTTPResponse#setup_header): Close
+	  HTTP/1.1 connection when returning an IO object as response body
+	  without setting HTTPResponse#chunked to true. See #855 no.1.
+
+	* test/webrick/test_httpserver.rb: Test it.
+
+Tue Jun 21 21:27:34 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* internal.h: move rb_thread_io_blocking_region() declaration
+	  from intern.h to internal.h. It's still experimental API and
+	  need more discussion. [ruby-dev:43698]
+	* include/ruby/intern.h: ditto.
+
+	* ext/socket/rubysocket.h: include internal.h.
+	* ext/socket/depend: add internal.h dependency.
+	* ext/socket/extconf.rb: add $INCFLAGS to topdir.
+
+Tue Jun 21 20:38:47 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (datetime_s_*): canonicalize 24 o'clock.
+
+Tue Jun 21 19:46:23 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_thread.rb (TestThread#test_priority): enable
+	  this test again. Current GVL respect thread priority rather
+	  than past.
+
+Tue Jun 21 13:25:35 2011  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c (readline_getc): applied a patch in
+	  #3827 by by Akio Tajima <artonx AT yahoo.co.jp>. (see #3827)
+
+Tue Jun 21 13:16:31 2011  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/extconf.rb: fixed bug, specify --disable-libedit
+	  then disable libedit, does not specify then check readline and
+	  libedit if failed checking readline. (fixes #3375)
+
+Mon Jun 20 22:52:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (before_exec): use sig_do_nothing instead of SIG_DFL
+	  for avoiding a race.
+	* process.c (sig_do_nothing): new function.
+
+Mon Jun 20 21:31:03 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (thread_timer): rename timeout_10ms to
+	  time_quantum. it's no longer 10ms.
+
+Mon Jun 20 18:46:02 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_cipher.c, ext/openssl/lib/openssl/cipher.rb:
+	  Documentation fix by Ippei Obayashi. See #4419.
+
+Mon Jun 20 15:41:33 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* lib/webrick/cookie.rb (WEBrick::Cookie.parse): Revert r31228.
+	  r31228 was for allowing the 'Cookie:' header which did not have no
+	  SP after ';' for separating cookie-pairs but RFC6265 requires single
+	  SP after ';' there.  We allow multiple SPs here for compatibility
+	  with older WEBrick version.
+
+	* test/webrick/test_cookie.rb: Test it.
+
+Sun Jun 19 13:31:26 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* NEWS: Introduce --hide-skip on test/unit.
+
+Sun Jun 19 10:05:16 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/parallel.rb: Override Test::Unit::TestCase#on_parallel_worker?
+	  only when $0 == __FILE__.
+
+	* lib/test/unit/parallel.rb: Run Test::Unit::Worker.run only when
+	  $0 == __FILE__.
+
+Sat Jun 18 23:59:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (fill_cbuf): finish reading at EOF, and the readconv has
+	  been cleared by another thread while io_fillbuf() is waiting at
+	  select().  a patch in [ruby-core:37197] by Hiroshi Shirosaki
+	  <h.shirosaki AT gmail.com>.  fixed #3840
+
+Sat Jun 18 21:36:29 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c: remove GVL_DEBUG
+
+Sat Jun 18 21:32:02 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm.c, vm_core.h (rb_vm_stack_to_heap): remove const.
+	  It makes compilations warnings.
+
+Sat Jun 18 18:54:15 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, vm_core.h (rb_vm_stack_to_heap): fix "const" place.
+
+Sat Jun 18 17:23:38 2011  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c, hash.c, load.c, proc.c, range.c, thread.c, time.c: don't
+	  declare internal functions.
+
+	* internal.h, vm_core.h: declare internal functions.
+
+	* array.c: include internal.h.
+
+	* common.mk: update dependency for array.o.
+
+Sat Jun 18 13:39:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: declarations declared in include/ruby/*.h removed.
+
+Sat Jun 18 12:42:17 2011  Tanaka Akira  <akr@fsij.org>
+
+	* method.h, internal.h iseq.h: declare internal functions.
+
+	* compile.c, eval.c, iseq.c, object.c, parse.y, proc.c, process.c,
+	  thread.c, vm.c, vm_eval.c, vm_insnhelper.c, vm_method.c: don't
+	  declare internal functions.
+
+	  Note that rb_method_entry_eq() is defined in vm_method.c but
+	  there was a declaration in proc.c with different const-ness.
+	  Now it is declared in method.h with same const-ness to the
+	  definition.
+
+	* object.c (rb_mod_module_exec): don't declare functions declared in
+	  include/ruby/intern.h.
+
+Sat Jun 18 12:05:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/backward/classext.h: for evil gems.  fixed #4803
+
+Sat Jun 18 11:12:13 2011  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: update dependencies.
+
+Sat Jun 18 11:09:03 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: suppress warnings.
+
+Sat Jun 18 10:22:39 2011  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: declare more internal functions.
+
+	* iseq.h (rb_method_get_iseq): declared.
+
+	* compile.c, eval.c, eval_error.c, iseq.c, parse.y, proc.c, range.c,
+	  ruby.c, time.c, util.c, vm.c: don't declare internal functions.
+
+	* eval.c, parse.y, thread_pthread.c: non-existing function declarations
+	  removed.
+
+Sat Jun 18 08:12:54 2011  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: dependencies updated.
+
+	* tool/update-deps: new file to assist update dependencies in
+	  common.mk.
+
+Sat Jun 18 07:27:27 2011  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: declare internal functions here.
+
+	* node.h: declare NODE dependent internal functions here.
+
+	* iseq.h: declare rb_iseq_t dependent internal functions here.
+
+	* vm_core.h: declare rb_thread_t dependent internal functions here.
+
+	* bignum.c, class.c, compile.c, complex.c, cont.c, dir.c, encoding.c,
+	  enumerator.c, error.c, eval.c, file.c, gc.c, hash.c, inits.c, io.c,
+	  iseq.c, load.c, marshal.c, math.c, numeric.c, object.c, parse.y,
+	  proc.c, process.c, range.c, rational.c, re.c, ruby.c, string.c,
+	  thread.c, time.c, transcode.c, variable.c, vm.c,
+	  tool/compile_prelude.rb: don't declare internal functions declared
+	  in above headers.  include above headers if required.
+
+	  Note that rb_thread_mark() was declared as
+	  void rb_thread_mark(rb_thread_t *th) in cont.c but defined as
+	  void rb_thread_mark(void *ptr) in vm.c.  Now it is declared as
+	  the later in internal.h.
+
+Sat Jun 18 02:36:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpNewRbClass): fix type of the 2nd
+	  argument.
+
+	* ext/bigdecimal/bigdecimal.h: ditto.
+
+Sat Jun 18 02:30:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): move BigMath.exp from
+	  bigdecimal/math.rb.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb: ditto.
+
+	* test/bigdecimal/test_bigdecimal.rb: move test for BigMath.exp from
+	  test/bigdecimal/test_bigmath.rb.
+
+	* test/bigdecimal/test_bigmath.rb: ditto.
+
+Sat Jun 18 00:20:54 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: do not define wnum[01].
+
+Fri Jun 17 18:57:36 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): fix return value of obj[a,*b]=c.
+
+Fri Jun 17 13:09:45 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/curses/curses.c:  Clean up documentation.
+
+Fri Jun 17 09:25:14 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/curses/curses.c:  Document curses constants.  Patch by Vincent
+	  Batts.  [Ruby 1.9 - Bug #4880]
+
+Fri Jun 17 09:11:05 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c:  Document Module#method_added and #method_removed.
+	  Patch by Bryce Kerley.  [Ruby 1.9 - Feature #4867]
+
+Fri Jun 17 08:50:16 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* io.c:  Improve documentation of IO and File open and new.
+	  Patch by Roger Pack.  [Ruby 1.9 - Bug #4790]
+
+Fri Jun 17 07:53:50 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/csv.rb:  Document #raw_encoding.  Patch by David Czarnecki.
+	  [Ruby 1.9 - Bug #4874]
+
+Fri Jun 17 07:46:50 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/benchmark.rb:  Document Benchmark::Tms#memberwise.  Patch by
+	  David Czarnecki.  [Ruby 1.9 - Bug #4873]
+
+Fri Jun 17 07:38:31 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/prettyprint.rb:  Improve documentation.  Patch by Ysiad
+	  Ferreiras.  [#4834]
+
+Fri Jun 17 07:23:03 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c (rb_ary_drop):  Improve documentation.  Patch by Caley Woods.
+	  [Ruby 1.9 - Bug #4858]
+
+Fri Jun 17 06:11:31 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/log.rb:  Improve documentation of BasicLog and Log.
+	  Patch by Olivier Brisse.  [Ruby 1.9 - Bug #4833]
+	* lib/webrick/httpstatus.rb:  Improve documentation of
+	  WEBrick::HTTPStatus.  Patch by Olivier Brisse.
+	  [Ruby 1.9 - Bug #4833]
+
+Fri Jun 17 04:48:22 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c, thread_pthread.h: remove unused variables.
+	  (native_thread_data_t::gvl_cond, native_thread_data_t::gvl_next)
+
+Thu Jun 16 14:32:31 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (rb_time_new): prevent overflow by "* 1000".
+
+Fri Jun 17 03:07:36 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm4_thread_create_join.rb,
+	  benchmark/bm_vm4_thread_mutex[1-3].rb: renamed to
+	  bm_thread_* (fix last rename).
+
+Fri Jun 17 02:26:47 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (native_thread_create): fix debug message.
+	  (add last newline)
+
+Thu Jun 16 23:40:49 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_schedule_rec): fix {UN,}LIKELY macro misuse.
+	* gc.c (rb_newobj): ditto.
+	* vm_insnhelper.c (vm_method_search): ditto.
+
+Thu Jun 16 20:06:15 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/test_parallel.rb: Fix Regexp for test.
+
+	* test/testunit/tests_for_parallel/test_third.rb: Use
+	  Test::Unit::TestCase#on_parallel_worker? for detecting worker.
+
+	* lib/test/unit/testcase.rb(Test::Unit::TestCase#on_parallel_worker?):
+	  New Method Test::Unit::TestCase#on_parallel_worker? returns true
+	  when a testcase is running on parallel worker.
+	* lib/test/unit/parallel.rb(Test::Unit::TestCase#on_parallel_worker?):
+	  ditto.
+
+Thu Jun 16 19:27:12 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* test/test_securerandom.rb: Add testcase.  This testcase does NOT aim
+	  to test cryptographically strongness and randomness.  It includes
+	  the test for PID recycle issue of OpenSSL described in #4579 but
+	  it's disabled by default.
+
+Thu Jun 16 17:55:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (TestIO#test_copy_stream_socket): fix
+	  test hanging up issue. Patch by CHIKANAGA Tomoyuki.
+
+Thu Jun 16 15:17:39 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* variable.c (const_missing):  Add simple example of const_missing.
+	  Patch by Anuj Dutta.  [Ruby 1.9 - Bug #4794]
+
+Thu Jun 16 15:09:29 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/monitor.rb:  Improve documentation.  Patch by Sandor Szucs.
+	  [Ruby 1.9 - Bug #4823]
+
+Thu Jun 16 14:54:09 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/utils.rb: Document WEBrick::Utils.  Patch by Olivier
+	  Brisse.  [Ruby 1.9 - Bug #4819]
+
+Thu Jun 16 14:26:46 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/httpservlet/erbhandler.rb:  Allow the ERB document to
+	  alter the content-type of the response.  [Ruby 1.9 - Bug #4685]
+
+Thu Jun 16 14:15:47 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/timeout.rb: Clarify timeout duration types.  Patch by Alf Mikula.
+	  [Ruby 1.9 - Bug #4791]
+	* lib/net/http.rb: ditto
+
+Thu Jun 16 13:25:25 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc*:  Import RDoc 3.7 release candidate
+
+Thu Jun 16 11:35:09 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (search_response): parses SEARCH responses from
+	  the Yahoo IMAP server correctly.  patched by Mark Nadig.  [Bug #4509]
+
+Thu Jun 16 09:12:38 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* fix for build on solaris 10.
+
+Thu Jun 16 09:08:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/io/console/test_io_console.rb (TestIO_Console#test_sync):
+	  fix for daemon process.
+
+Thu Jun 16 07:58:01 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/test_parallel.rb(test_ignore_tzero): Test for r32109.
+
+	* test/testunit/tests_for_parallel/test_third.rb: Use another way to
+	  detect that test is running on worker. This fixes sometimes
+	  TestParallel failing.
+
+Thu Jun 16 07:20:06 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb(Test::Unit::Runner#_run_parallel): Ignore -j0
+	  because it makes blocking forever by IO.select.
+
+Thu Jun 16 03:08:11 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/testunit/test_parallel.rb (TestParallel::TestParallelWorker#
+	  setup): now can run on Windows, probably.
+
+	* test/testunit/test_parallel.rb (TestParallel::TestParallel#setup):
+	  ditto.
+
+Thu Jun 16 03:00:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (do_select): Windows: no need to poll if select(2) is
+	  cancelable.
+
+	* thread_win32.c (native_fd_select): new function to make select(2)
+	  cancelable.
+
+	* thread_win32.c (rb_w32_check_interrupt): new function for checking
+	  interrupt.
+
+	* win32/win32.c (rb_w32_select_with_thread): new function. cancelable
+	  select(2).
+
+	* win32/win32.c (rb_w32_select): use above function internally.
+
+Wed Jun 15 23:30:45 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* gc.c: fix a regression by r31690 on AIX because AIX malloc
+	  return NULL if it's passed 0. But some caller don't expect it.
+	  patch by Yutaka Kanemoto. [ruby-dev:43779]
+	(vm_malloc_prepare): return calculated size.
+	(vm_xmalloc): use above result.
+	(vm_xcalloc): ditto.
+
+Wed Jun 15 23:11:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c: remove BLOCKING_REGION_CORE() macro. It's no longer used
+	  since r32022.
+
+Wed Jun 15 21:00:47 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_config.rb: execute based on the existence of the
+	  OpenSSL module.
+
+Wed Jun 15 12:35:11 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb (test_copy_stream_socket): wait a child process
+	  before SIGUSR1 handler is removed.
+
+	* test/pathname/test_pathname.rb (define_assertion): use line number
+	  for test method names.
+
+Wed Jun 15 10:37:43 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_stat_rdev): use DEVT2NUM.
+
+	* file.c (rb_stat_rdev_major): ditto.
+
+	* file.c (rb_stat_rdev_minor): ditto.
+
+Wed Jun 15 05:12:59 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (lib_eventloop_core): revert the last change (it's
+	  the part for ruby_1_8), and use rb_thread_check_ints() when RUBY_VM
+	  is defined.
+
+Wed Jun 15 04:42:47 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm3_thread_*.rb: renamed bm_vm3_thread_*.rb to
+	  benchmark/bm_vm_thread_*.rb.
+
+Wed Jun 15 04:28:39 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_vm4_thread_pass.rb: rename bm_vm4* to
+	  bm_vm_thread_*. suggested by ko1.
+	* benchmark/bm_vm4_pipe.rb: ditto.
+	* benchmark/bm_vm4_alive_check1.rb: ditto.
+	* benchmark/bm_vm4_pass_flood.rb: ditto.
+
+Wed Jun 15 03:52:50 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_vm4_pass_flood.rb: new benchmark for GVL fairness.
+	* benchmark/bm_vm4_alive_check1.rb: ditto.
+
+Wed Jun 15 01:27:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (gvl_yield): fix live lock issue on 1-2 cpus
+	  system. It's additional fix for r32021.
+	* thread_pthread.c (gvl_init): add switch_wait_cond.
+	* thread_pthread.h (typedef struct rb_global_vm_lock_struct): ditto.
+
+Tue Jun 14 23:16:22 2011  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (show_progress): refine verbose mode.
+	  (exec_test): ditto.
+
+Tue Jun 14 23:02:36 2011  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (show_progress): extracted from assert_check.
+	  (assert_check): use show_progress.
+	  (assert_normal_exit): ditto.
+	  (assert_finish): ditto.
+	  (flunk): ditto.
+
+Tue Jun 14 22:51:42 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/test_*.rb: added tests.
+
+Tue Jun 14 22:09:58 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: renamed some functions.
+	* ext/date/date_core.c: modified doc.
+
+Tue Jun 14 21:26:01 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* cont.c (cont_save_thread): add new utility function.
+	  rb_context_t::saved_thread.machine_stack_start and
+	  machine_stack_end should be cleared immediately after a snapshot of
+	  current thread is stored to saved_thread. [ruby-dev:43680] [Bug #4855]
+	  this change aims to get rid of unnecessary GC mark at machine stack.
+
+Tue Jun 14 19:50:49 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_autoload.rb: remove temporary directory.
+
+Tue Jun 14 11:05:03 2011  Narihiro Nakamura  <narihiro@netlab.jp>
+
+	* gc.c (rb_gc_set_params): call initial_expand_heap if
+	  initial_heap_min_slots is set.
+
+Tue Jun 14 11:02:08 2011  Narihiro Nakamura  <narihiro@netlab.jp>
+
+	* gc.c: use size_t.
+
+Tue Jun 14 01:10:38 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/coverage/test_coverage.rb: add a test for restart.  a patch
+	  from Xavier Shay.  [ruby-core:36745]
+
+Tue Jun 14 01:05:10 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/coverage/coverage.c: make it restartable.  [ruby-core:36539]
+
+Mon Jun 13 23:55:40 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_schedule_rec): call gvl_yield() unconditionally.
+	* thread_pthread.c: remove HAVE_GVL_YIELD macro.
+	* thread_win32.c (gvl_yield): new. this fallback logic was moved from
+	  rb_thread_schedule_rec().
+
+Mon Jun 13 23:50:25 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/io/console/console.c (console_dev): typo.
+
+Mon Jun 13 23:38:23 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_parse_string): flush delayed token.  based on a
+	  patch by Masaya Tarui in [ruby-dev:43762].  Bug #4544
+
+	* parse.y (yylex): revert r24557.  delayed token at the end of
+	  string should be flushed already by the above change.
+
+Mon Jun 13 23:33:31 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_dev): console should be
+	  unbuffered.
+
+	* ext/io/console/console.c (console_dev): take care of no-ctty
+	  case.
+
+Mon Jun 13 23:06:12 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c: rewrite GVL completely. This fix improve some
+	  benchmark dramatically (e.g. vm3_thread_mutex result changed
+	  form 120.601sec to 3.426sec).
+	* thread_win32.c: ditto.
+	* thread_pthread.h: ditto.
+	* vm_core.h: ditto.
+	* thread.c: ditto.
+
+Mon Jun 13 23:11:52 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_unix.rb: don't use Thread.abort_on_exception.
+
+Mon Jun 13 23:05:01 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/unixsocket.c (unix_send_io): race condition fixed.
+	  (unix_recv_io): ditto.
+	  fixed by Eric Wong.  [ruby-core:35574]
+
+	* test/socket/test_unix.rb: test added for above problem.
+
+Mon Jun 13 21:41:05 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_win32.c (native_cond_signal): remove unnecessary rb_bug().
+	  It's additional fix for r32021. [Bug #4696]
+
+Mon Jun 13 20:50:49 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_ec.rb
+	  test/openssl/test_pkey_ec.rb: merge both files into test_pkey_ec.rb.
+	  Removed redundant group instantiation from PKey tests.
+	* test/openssl/utils.rb: only create TEST_PKEY_EC_P256V1 if EC is
+	  defined.
+
+Mon Jun 13 20:28:24 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/digest.rb: remove MDC2 from test, it is not available
+	  by default in an OpenSSL installation.
+
+Mon Jun 13 20:18:55 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, vm_insnhelper.h: move decl. of
+	  ruby_vm_global_state_version and related macros
+	  from vm_core.h to vm_insnhelper.h.
+
+	* vm.c (vm_clear_all_cache): added.  This function is called
+	  when ruby_vm_global_state_version overflows.
+	  TODO: vm_clear_all_inline_method_cache() is only place holder.
+	        We need to implement it ASAP.
+
+	* vm_method.c (vm_clear_global_method_cache): added.
+
+Mon Jun 13 19:46:21 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/cmath.rb: add new method Object#real?. fix #3137
+
+Mon Jun 13 18:52:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (is_kind_of_BigDecimal): new function to
+	  examine the whether the object is kind of BigDecimal.
+
+Mon Jun 13 18:49:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimalCmp): use GetVpValueWithPrec
+	  for Float and Rational arguments.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_new, test_cmp, test_power):
+	  add and modify tests for the above change.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimalCmp): modify coding style to
+	  match ruby's standard.
+
+Mon Jun 13 18:33:04 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (SecureRandom.random_bytes): modify PRNG state
+	  to prevent random number sequence repetition at forked child
+	  process which has same pid.
+	  reported by Eric Wong.  [ruby-core:35765]
+
+Mon Jun 13 17:02:34 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#use_ssl?): require 'openssl' only when
+	  https is needed. fixes r31933.
+
+Mon Jun 13 14:35:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* lib/cmath.rb (CMath.cbrt): returns the principal value of the cube
+	  root of the argument.  fix #3676
+
+	* test/test_cmath.rb (test_cbrt_returns_principal_value_of_cube_root):
+	  test for the above change.
+
+Mon Jun 13 14:17:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* lib/test/unit.rb (Test::Unit::GlobOption#non_options): fix typo.
+
+Mon Jun 13 13:04:20 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_digest.c: fix error for digests that have no oid
+	(e.g. DSS1).
+	* test/openssl/test_digest.c: add tests for this.
+
+Mon Jun 13 12:51:51 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/yaml.rb: load psych only when syck is not loaded.
+
+Mon Jun 13 12:23:39 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+Mon Jun 13 12:23:39 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/psych/lib/psych/deprecated.rb (Object#to_yaml_properties):
+	  undef to_yaml_properties before redefine it.
+
+	* ext/syck/lib/syck/rubytypes.rb: ditto.
+
+Mon Jun 13 11:30:10 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_digest.c: allow Digests to be created by sn, ln or
+	  oid.
+	* test/openssl/test_digest.rb: add tests for this.
+	  [Ruby 1.9 - Feature #4412] [ruby-core:35319]
+
+Mon Jun 13 10:54:03 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/pkey_dh.c: corrected documentation.
+	* test/openssl/utils.rb: add test key for DH.
+	* test/openssl/test_pkey_dh.rb: add tests.
+
+Mon Jun 13 10:13:08 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/pkey_dh.c: clarify difference between DH#public_key and
+	DH#pub_key in documentation.
+
+Mon Jun 13 05:50:43 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* NEWS: introduce PKey.read.
+
+Mon Jun 13 05:17:29 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey.c: added PKey.read module function that allows
+	  reading arbitrary public/private keys from DER-/PEM-encoded File or
+	  string instances.
+	* ext/openssl/ossl_pkey_dh.c: improved documentation.
+	* test/openssl/utils.rb: added EC test key.
+	* test/openssl/test_pkey_rsa.rb
+	  test/openssl/test_pkey_dsa.rb: Test PKey.read. Reuse keys from
+	  OpenSSL::TestUtils.
+	* test/openssl/test_pkey_ec.rb: Created test file for EC tests.
+	  Test PKey.read.
+	  [Ruby 1.9 - Feature #4424] [ruby-core:35330]
+
+Mon Jun 13 04:42:24 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (total_i): fix to skip no ruby objects.
+
+Mon Jun 13 03:07:38 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/benchmark/test_benchmark.rb (capture_output):
+	  replace '-' as space. On NetBSD, subtract between two Process.times
+	  after and before the short process may return negative value like:
+	  t0=Process.times; yield; t1=Process.times; p t1.utime-t0.utime
+
+Mon Jun 13 02:40:23 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkey_dsa.rb: Test for DSA#syssign/sysverify.
+
+Mon Jun 13 01:59:19 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey_dh.c: completed documentation.
+	* ext/openssl/ossl_pkey_dsa.c: corrected examples. Improved parameter
+	  sections.
+
+Mon Jun 13 00:25:10 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey_dsa.c: completed documentation.
+
+Sun Jun 12 23:36:46 2011  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (kill_sub_thread): remove the method. [ruby-core:34185]
+
+Sun Jun 12 21:01:56 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (d_lite_marshal_load): should give converted value.
+
+Sun Jun 12 20:36:30 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: edited doc.
+
+Sun Jun 12 18:12:07 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm3_clearmethodcache.rb: added.
+
+Sun Jun 12 17:40:29 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_clear_cache*): update only vm state version.
+
+	* vm_method.c (rb_method_entry_get_without_cache, rb_method_entry):
+	  Fill method cache entry with vm state version, and
+	  check current vm state version for method (cache) look up.
+	  This modification speed-up invalidating of global method cache table.
+	  [Ruby 1.9 - Feature #3905] [ruby-core:36908]
+
+Sun Jun 12 16:19:48 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: fail on Mac OS X. [Bug #4853][ruby-dev:43655]
+
+Sun Jun 12 15:56:08 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c: remove th->transition_for_lock. It's thread unsafe.
+	  [Bug #4723][ruby-dev:43563]
+
+Sun Jun 12 15:47:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c: introduce spurious wakeup safe deadlock check.
+	  [Bug #4696][ruby-dev:43554]
+
+Sun Jun 12 13:33:52 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm3_thread_mutex.rb: remove it.
+
+	* benchmark/bm_vm3_thread_mutex[1-3].rb: added 3 benchmarks.
+	    1: one thread with one mutex (no contention).
+	    2: two threads with one mutex (contention).
+	    3: 1000 threads with one mutex (huge number of contention)
+	  Above removed benchmark was type 3.
+	  Therefore, this commit adds type 1 and 2 benchmark.
+
+Sun Jun 12 11:16:59 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: use select() appropriately for sendfile().
+	  Fixed by Eric Wong.  [ruby-core:36150]
+	  (maygvl_copy_stream_wait_readwrite): removed.
+	  (nogvl_copy_stream_sendfile): use nogvl_copy_stream_wait_write and
+	  maygvl_copy_stream_wait_read instead of
+	  maygvl_copy_stream_wait_readwrite.
+
+Sun Jun 12 09:32:13 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* atomic.h (ATOMIC_OR): _InterlockedOr is not available on mingw.h
+	* gc.c (rb_gc_set_params): VM_OBJSPACE is disabled on mingw.
+
+Sun Jun 12 01:07:09 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: edited doc.
+
+Sat Jun 11 23:18:00 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (RUBY_VM_SET_TIMER_INTERRUPT, RUBY_VM_SET_INTERRUPT,
+	  RUBY_VM_SET_FINALIZER_INTERRUPT): use atomic ops for preventing
+	  interrupt_flag bit lost. [Bug #4770][ruby-dev:43467]
+	* thread.c (rb_threadptr_execute_interrupts_rec): ditto.
+	* vm_core.h (typedef struct rb_thread_struct): change type of
+	  interrupt_flag to rb_atomic_t.
+	* atomic.h: move atomic ops definition from signal.c.
+	* signal.c: remove atomic ops definition.
+	* common.mk (gc, signal, thread, cont): add to dependency to atomic.h.
+
+Sat Jun 11 23:23:52 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: edited doc.
+
+Sat Jun 11 23:02:36 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/buffering.rb (module OpenSSL):
+	  Buffering#each_byte should return String in accordance with IO in
+	  1.9.
+
+	* test/openssl/test_buffering.rb (class OpenSSL): add tests for getc
+	  and each_byte.
+
+Sat Jun 11 22:41:37 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* time.c: a correction of doc for strftime (%v).
+
+Sat Jun 11 22:30:53 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: replaced doc for strftime based on Time's one.
+
+Sat Jun 11 22:07:56 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (datetime_s_{iso8601,rfc3339,xmlschema,rfc2822,httpdate}):
+	  do not take argument comp.
+
+Sat Jun 11 21:58:31 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: added examples.
+
+Sat Jun 11 19:40:45 2011  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: expand heap if initial_heap_min_slots is bigger than
+	  HEAP_MIN_SLOTS.
+
+Sat Jun 11 19:42:50 2011  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ChangeLog (vim): set shiftwidth to 2.
+
+Sat Jun 11 19:27:06 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_x509req.c: raise RequestError instead of
+	  CertificateError when Request#to_der gets an error from OpenSSL.
+	  Patch from Ippei Obayashi, see #4420. I cannot write a test for
+	  this... Request does not allow to create broken bytes...
+
+Sat Jun 11 19:34:51 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (Date::(ABBR_)?(MONTH|DAY)NAMES): should be usascii.
+
+Sat Jun 11 19:24:33 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: rewrote doc.
+
+Sat Jun 11 19:04:15 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::GlobOption#non_options): should run
+	  with 1.8.
+
+Sat Jun 11 18:05:57 2011  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* bootstraptest/runner.rb: should initialize $stress to avoid warnings.
+
+Sat Jun 11 18:02:09 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_getc): should be 7bit if ascii.  fixes #4557
+
+Sat Jun 11 16:52:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (rb_enc_uint_chr): fix message format. Bug#4869
+
+Sat Jun 11 16:28:25 2011  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/formatters/pretty.rb
+	  (REXML::Formatters::Pretty#write_text),
+	  test/rexml/test_core.rb
+	  (Tester#test_pretty_format_long_text_finite): don't ignore
+	  'width' parameter in pretty formatter. fixes #4498
+	  Reported by Michael Frasca. Thanks!!!
+
+Sat Jun 11 16:11:36 2011  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_core.rb
+	  (Tester#test_pretty_format_long_text_finite): remove needless
+	  assert_nothing_raised.
+
+Sat Jun 11 16:04:03 2011  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/xpathparser.rb
+	  (REXML::Parsers::XPathParser#parse),
+	  test/rexml/test_elements.rb
+	  (ElementsTester#test_each_with_frozen_condition):
+	  don't modify original XPath. fixes #4161
+	  Reported by Pavel Shved. Thanks!!!
+
+Sat Jun 11 15:53:27 2011  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_elements.rb (ElementsTester): remove needless
+	  prefix from test name.
+
+Sat Jun 11 15:36:36 2011  Martin Duerst  <duerst@it.aoyama.ac.jp>
+	* common.mk: fixed a grammatical error
+
+Sat Jun 11 14:20:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm.c (thread_memsize): don't ignore size of th->local_storage.
+
+Sat Jun 11 10:32:46 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: should quote arch_hdrdir and libpath for the case
+	  installed prefix contains spaces.
+
+Sat Jun 11 10:20:52 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (native_cond_timeout): wrap conditionally used
+	  label.
+
+	* thread_pthread.c (native_sleep): remove unused variable.
+
+Sat Jun 11 10:15:50 2011  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (rb_thread_execute_interrupts): use GetThreadPtr to extract
+	  rb_thread_t from VALUE.
+	  reported by Motohiro KOSAKI.  [ruby-dev:43700]
+
+Sat Jun 11 10:00:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_process_options): add missing return type.
+
+Fri Jun 10 23:18:09 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/tk/tcltklib.c (lib_eventloop_core): replace CHECK_INTS with
+	  rb_thread_check_ints(). Because current code can't be compiled.
+
+Fri Jun 10 16:38:13 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_locale_charmap): When ruby process is run as Windows
+	  Service the console codepage is not set, GetConsoleCP returns 0.
+	  So on such environment, use GetACP().
+	  http://blogs.msdn.com/b/michkap/archive/2005/02/08/369197.aspx
+	  patched by Rafal Bigaj [ruby-core:36832] [Bug #4854]
+
+Fri Jun 10 14:34:24 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: restore TESTRUN_SCRIPT to "$(srcdir)/test.rb".
+	  TESTRUN_SCRIPT is used by "make run", "make gdb" and so on.
+
+Fri Jun 10 13:01:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_module.rb (TestModule#remove_rake_mixins): remove all
+	  module related to Rake.
+
+Fri Jun 10 09:52:38 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* encoding.c:  Mention that Encoding.compatible? can work with more
+	  than just Strings.
+
+Fri Jun 10 02:25:53 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: updating version to match released gem.
+
+Fri Jun 10 01:06:29 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/bigdecimal (BigDecimal_to_i): Integer#** may return flonum.
+
+Fri Jun 10 00:35:12 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (string_to_c_internal): uses rb_reg_nth_match;
+	* rational.c (string_to_r_internal): ditto.
+
+Fri Jun 10 00:25:03 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* gc.c: remove an unused declaration.
+
+Fri Jun 10 00:24:04 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (string_to_r): Rational#** may return flonum.
+
+Thu Jun  9 23:57:53 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: fix IO.copy_stream interrupt handling.
+	  based on the patch by Eric Wong.  [ruby-core:36156]
+
+	* vm_core.h (rb_thread_call_with_gvl): don't declare here.
+
+	* thread.c: include internal.h.
+	  (rb_thread_execute_interrupts): new function.
+
+	* internal.h (rb_thread_execute_interrupts): declared.
+	  (rb_thread_call_with_gvl): declared.
+
+Thu Jun  9 23:34:01 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* gc.c (rb_objspace_call_finalizer): use rb_typeddata_is_kind_of() for
+	  type check to get rid of a double free when main Thread has singleton
+	  class. [ruby-core:36741] [Bug #4828]
+	* thread.c (rb_obj_is_mutex): add a new utility function.
+	* vm.c (rb_obj_is_thread): ditto.
+
+Thu Jun  9 22:53:49 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/test_thread.rb (TestThread#test_kill_thread_subclass):
+	  add test for Thread.kill with Thread subclass instance.
+
+Thu Jun  9 22:31:47 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/test_thread.rb (TestThread#test_kill_wrong_argument):
+	  test for [ruby-core:35086].
+	  partially forward porting r31402 from ruby_1_9_2 branch.
+
+Thu Jun  9 18:36:00 2011  Kenta Murata <mrkn@mrkn.jp>
+
+	* string.c: Fix the ambiguous description of the behavior of
+	  rb_str_aref_m with a range.  It returns nil when the beginning of
+	  the range is greater than the end of the string rather than the range.
+
+Thu Jun  9 10:57:03 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Hash subclasses can be read
+	  from YAML files.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: Hash subclasses can be
+	  dumped to YAML files.
+	* test/psych/test_hash.rb: corresponding test.
+
+Thu Jun  9 09:18:51 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Ruby modules can be loaded
+	  from YAML files.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: Ruby modules can be
+	  dumped to YAML files.
+	* test/psych/test_class.rb: corresponding test.
+
+Thu Jun  9 09:05:04 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Ruby classes can be loaded
+	  from YAML files.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: Ruby classes can be
+	  dumped to YAML files.
+	* test/psych/test_class.rb: corresponding test.
+
+Wed Jun  8 21:38:57 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* cont.c (root_fiber_alloc): set root fiber's status RUNNING.
+	  in cont_mark() only RUNNING fiber's machine stack is marked.
+	  root fiber's status should be RUNNING at the beginning regardless of
+	  FIBER_USE_NATIVE. [ruby-core:36735] fixes #4827
+
+Tue Jun  7 20:50:11 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* doc/irb/irb.rd: fix typo. patch by Nobuhiro IMAI.
+	  [Bug #4843] [ruby-dev:43639]
+	* doc/irb/irb.rd.ja: ditto.
+	* doc/ChangeLog-YARV: ditto.
+
+Tue Jun  7 18:52:55 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rubygems/user_interaction.rb (Gem::StreamUI#tty?): IO#tty? of
+	  Windows has been fixed at r29969.
+
+	* test/rubygems/test_gem_stream_ui.rb: now can run tests.
+
+Tue Jun  7 18:36:41 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem.rb (TestGem#{test_self_user_home_userprofile,
+	  test_self_user_home_user_drive_and_path}): should simply ignore
+	  meaningless tests instead of skipping them.
+
+Tue Jun  7 18:15:00 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_set_program_name):
+	  should show some messages when skipping tests.
+
+Tue Jun  7 13:59:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_core.c (date_s_today, datetime_s_now): check the
+	  result of localtime_r().
+
+Tue Jun  7 13:36:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/extconf.rb: use $defs not $CPPFLAGS to get rid of
+	  command line escape issues on Windows.  fixed #4835.
+
+Tue Jun  7 03:18:45 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_io.rb (TestIO#test_s_{,bin}write): do not create a
+	  file under /tmp. [Bug #4846]
+
+Mon Jun  6 22:51:43 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* cont.c: use #if FIBER_USE_NATIVE instead of #ifdef.
+	  you can suppress use of setcontext for Fiber with compile option
+	  -DFIBER_USE_NATIVE=0
+
+Mon Jun  6 21:59:23 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/test_switch_hitter.rb: added a test.
+
+Mon Jun  6 21:37:45 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: added notes.
+
+Mon Jun  6 21:02:12 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: flattened format to strftimev.
+	* ext/date/date_core.c (date_strftime_internal): taints run.
+
+Mon Jun  6 15:10:17 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/{defines,missing}.h (rb_infinity, rb_nan): move from
+	  defines.h to missing.h.  (couldn't use RUBY_EXTERN there.)
+
+Mon Jun  6 14:35:48 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_markup_pre_process.rb (TestRDocMarkupPreProcess#
+	  {test_include_file,test_include_file_encoding_incompatible}): no
+	  need to write such workaround.  don't hide the bug of ruby. (and the
+	  bug is already fixed.)
+
+Mon Jun  6 14:11:11 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/date/date_core.c (valid_jd_sub): need to convert from VALUE to
+	  double.
+
+	* ext/date/date_core.c (offset_to_sec): get rid of a compiler warning.
+
+Mon Jun  6 14:09:08 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/defines.h (rb_infinity, rb_nan): export for Windows.
+
+Mon Jun  6 10:54:45 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb(Test::Unit::Runner#puke):
+	  Add overriding from MiniTest::Unit#puke. This reverts minitest's fix
+	  that skip messages are hidden when not verbose mode (-v option).
+	  To hide skip messages, use --hide-skip option instead.
+
+Mon Jun  6 10:52:13 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: don't use autoload.
+
+Mon Jun  6 09:39:43 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (parse): release event objects to plug memory
+	leak.  Thanks Mark J. Titorenko!
+
+Sun Jun  5 23:26:15 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* eval.c: remove rb_thread_stop_timer_thread function declaration.
+	  Instead, include vm_core.h.
+	* process.c: ditto.
+
+Sun Jun  5 21:38:51 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (thread_timer): add to care a spurious wakeup.
+	  When native_cond_timedwait() return 0 by spurious wakeup, we
+	  don't have to neither 1) call timer_thread_function and 2)
+	  exit the timer thread.
+
+Sun Jun  5 17:50:01 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (m_real_cwyear): new.  derived from m_cwyear.
+	* ext/date/date_strftime.c: trivial changes.
+
+Sun Jun  5 17:22:01 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/config_list.in: add new options for tcltklib.
+
+Sun Jun  5 10:06:50 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_tmx.h: now does not place decoded data.  allows to
+	  access indirectly via functions on demand.
+	* ext/date/date_strftime.c: ditto.
+	* ext/date/date_core.c: ditto.
+	* ext/date/date_core.c ({d|dt}_lite_to_s): use strftime.
+
+Sun Jun  5 06:22:02 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* NEWS: wrote about changes of date.
+
+Sat Jun  4 16:59:26 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (d_lite_inspect): changed the format.
+	* ext/date/date_core.c: refactoring and fixing some bugs.
+
+Sat Jun  4 11:30:57 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (check_mix_method_i, do_mix_method_i): not mix methods
+	  renamed as nil.
+
+Sat Jun  4 04:04:41 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rubygems/test_gem_commands_which_command.rb:
+	  "missing" exists on ruby's top source directory. [Bug #4815]
+
+Fri Jun  3 21:48:12 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/test_case.rb: Refix for test-all in separate directory.
+	  r31147 + r31151.
+
+Fri Jun  3 20:58:47 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/date/date_core.c (d_lite_plus): get rid of compiler warnings.
+
+Fri Jun  3 20:56:40 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/missing.h, numeric.c (round): moved prototype of round()
+	  from numeric.c to missing.h.  (note: round() is C99 feature, so ruby
+	  provides it if not exist in C runtime.)
+
+Fri Jun  3 20:42:04 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/parallel.rb: Keep $stdin, $stdout before run testcase
+	  and restore after run. Because some test break $stdin, $stdout.
+	  Fixes [Bug #4433] [ruby-core:35353]
+
+Fri Jun  3 19:58:14 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_fflush): windows -- call fsync() only when the FD is tied to
+	  file, because if the FD is pipe, it blocks.
+
+Fri Jun  3 09:27:31 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (URI::HTTP#request_uri): return nil when the uri
+	  is path-rootless form. Bug #4759
+
+Thu Jun  2 23:51:03 2011  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Improve the line ending detection algorithm
+	  patch by Alexey).
+
+Thu Jun  2 20:05:57 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_s_write, rb_io_s_binwrite): return!!!
+
+Thu Jun  2 16:29:34 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* io.c: Add File.write, File.binwrite. [Feature #1081] [ruby-core:21701]
+
+	* test/ruby/test_io.rb: Test for File.write, File.binwrite.
+
+	* NEWS: News for above.
+
+Thu Jun  2 12:33:09 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_fflush, rb_io_flush): need to fsync() when ruby calls
+	  internal flush. [ruby-core:36670] [Bug #4813]
+
+Thu Jun  2 07:56:24 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: reduce CPU power required by an eventloop.
+
+Tue May 31 21:28:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_signal.rb (TestSignal#test_signal_requiring):
+	  redirect stderr to null device.
+
+Thu Jun  2 00:45:26 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: fix for uninitialized global variables.
+	  [Ruby 1.9 - Bug #4811]
+
+Wed Jun  1 21:57:08 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (native_sleep): fix 1000times calculation error.
+	  this is a regression since r31457. [Bug #4808] [ruby-dev:43606]
+
+Wed Jun  1 17:19:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c: remove unused macro.
+
+Wed Jun  1 15:42:18 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (peek_n): new macro to see next nth char.
+
+Wed Jun  1 15:40:46 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (gem): fix for rubygems change.
+
+Wed Jun  1 14:07:57 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 2.2.2 (r6281)
+	* test/minitest/*: ditto
+
+Wed Jun  1 12:35:50 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/rubygems*: Import rubygems 1.8.5 (released @ 137c80f)
+	* test/rubygems: Ditto
+
+Wed Jun  1 12:34:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: add new features of bigdecimal.
+
+Wed Jun  1 09:41:14 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/cgi/util.rb:  Improve documentation.  Patch by Kuba Fietkiewicz.
+	  [Ruby 1.9 - Bug #4775]
+	* lib/cgi/core.rb:  ditto
+
+Wed Jun  1 09:26:05 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mathn.rb:  Improve documentation and attach it to Numeric.
+	  Modified from patch by Anil V.  [Ruby 1.9 - Bug #4762]
+
+Wed Jun  1 09:21:30 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/prime.rb:  Indent examples enough to appear as code sections.
+	  Note that Prime is Enumerable.  [#4762]
+
+Wed Jun  1 07:34:57 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* hash.c (key_i):  Change rdoc from "the first occurrence" to "an
+	  occurrence" since first occurrence is not a specification of
+	  Hash#key.  [Ruby 1.9 - Bug #4760]
+
+Wed Jun  1 07:26:19 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/pty/pty.c (pty_check):  Restore "not reached" comment.
+	  [Ruby 1.9 - Bug #4756]
+
+Wed Jun  1 07:21:40 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c:  Fix document-method declarations for set_sync and
+	  set_comment.  [Ruby 1.9 - Bug #4695]
+
+Wed Jun  1 06:43:13 2011  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* load.c (loaded_feature_path): cut nonsense loop execution to fix
+	  performance bug.
+
+Wed Jun  1 01:16:02 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mix_module): implement Module#mix.
+
+Wed Jun  1 01:15:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_encoding_set): should honor already set ecflags since it
+	  might be set by mode option.  fixed #4804
+
+Wed Jun  1 00:34:04 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValueWithPrec): remove unused
+	  variable.
+
+Wed Jun  1 00:32:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValueWithPrec): support instantiation from
+	  a Float through Rational.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_new): ditto.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_global_new_float): add a test for
+	  the above changes.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_new_with_float): ditto.
+
+Wed Jun  1 00:07:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_coerce): support coerce with a
+	  Rational.  The precision used for instantiate a BigDecimal from the
+	  given Rational is obtained from the receiver BigDecimal.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_coerce): add a test for the
+	  above change.
+
+Tue May 31 23:49:08 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (offset_to_sec): fixed invalid validation.
+
+Tue May 31 23:43:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValueWithPrec): replace the algorithm for
+	  coercing from a Rational to stop requiring "bigdecimal/util.rb".
+	  [ruby-core:34318]
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValue): refactoring.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_new): support instantiation from a
+	  Rational.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_global_new_with_rational): add a
+	  test for the above change.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_new_with_rational): ditto.
+
+Tue May 31 22:44:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_new): support instantiation a
+	  BigDecimal object from an Integer.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_new_with_integer):
+	  add for testing the above change.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_global_new): replace its body
+	  with a BigDecimal_new call.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_global_new_with_integer):
+	  add for testing the above change.
+
+Tue May 31 22:24:39 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: use simple/complex mode instead of light/right mode.
+	* test/date/*.rb: followed the above changes.
+
+Tue May 31 21:28:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_signal.rb (TestSignal#test_signal_requiring):
+	  initialize SIGINT handler.
+
+Tue May 31 17:03:24 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* lib/net/http.rb, lib/net/protocol.rb: Allow to configure to wait
+	  server returning '100 continue' response before sending HTTP request
+	  body. See NEWS for more detail. See #3622.
+	  Original patch is made by Eric Hodel <drbrain@segment7.net>.
+
+	* test/net/http/test_http.rb: test it.
+
+	* NEWS: Add new feature.
+
+Tue May 31 14:17:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_s_pipe): potential bug. the mode of read IO is set as
+	  DEFAULT_TEXTMODE in call of io_set_encoding(), and of write IO is
+	  also set as it in call of io_new_instance() via rb_protect().
+	  so, if DEFAULT_TEXTMODE is not 0, we should check the result of
+	  extract_binmode() and avoid crush of default IO mode and the result.
+
+Tue May 31 13:00:17 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* strftime.c (rb_strftime_with_timespec): improved style consistency.
+	  constified some variables.
+
+	* test/test_time.rb (TestTime#test_huge_precision): test for #4456.
+
+Tue May 31 12:53:10 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/io/wait/test_io_wait.rb (TestIOWait#{test_nread,test_ready?,
+	  test_wait}: give system some time to process the written data.
+
+Tue May 31 12:40:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#test_open_mode): MUST release resources
+	  explicitly. fix problem of r31671
+
+Tue May 31 10:49:55 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_exec.c: remove conditions for clang
+	  because clang version 3.0 (trunk 132165) doesn't need them.
+
+Mon May 30 22:19:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_signal.rb (TestSignal#test_signal_requiring): don't
+	  close stderr.
+
+Mon May 30 20:22:00 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_signal.rb (TestSignal#test_signal_requiring): small
+	  but critical typo of r31642.  sorry...
+	  [Bug #4798] [ruby-core:36550]
+
+Mon May 30 15:44:16 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (opt_mult): as r31805, volatile it.
+	  Without this, clang -O fails calculation.
+
+	* numeric.c (fix_mul): ditto.
+
+	* rational.c (f_imul): ditto.
+
+Mon May 30 10:26:51 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (int_pow): make sure to assign the result of x * z.
+	  If xz is optimized out, the value won't overflow.
+
+Sun May 29 23:17:29 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_match): fix rdoc of Regexp#=~.
+	  patched by Tsuyoshi Sawada. [Bug #4781]
+
+Sun May 29 23:10:42 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/https.rb (WEBrick::HTTPRequest#parse_uri):
+	  keep parse_uri as private. patched by okkez. [Bug #4773]
+
+Sun May 29 17:53:03 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* numeric.c: add #include "internal.h" for rb_big_uminus() prototype.
+
+Sun May 29 15:09:05 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_round): fix for negative value.
+
+Sun May 29 02:16:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/net/http/utils.rb (TestNetHTTPUtils#teardown): add nil check.
+
+Sun May 29 00:22:40 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (before_exec, after_exec): change from macro to function.
+
+Sat May 28 19:30:17 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (before_exec, after_exec): change SIGPIPE handler to SIG_DFL
+	  before calling execve(). Because r31760 reintroduced an issue that
+	  system() may hang up (i.e. [ruby-dev:12261]).
+	* process.c (save_sigpipe, restore_sigpipe): new.
+
+Sat May 28 16:08:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (Init_signal, default_handler): change default SIGPIPE handler
+	  from empty function to SIG_IGN. [ruby-dev:43215]
+	* signal.c (sigpipe): removed.
+
+Sat May 28 03:04:27 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (fill_cbuf): return MORE_CHAR_SUSPENDED when cbuf is not empty.
+
+Sat May 28 02:22:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* string.c (rb_str_bytesize): rb_str_bytesize() should use LONG2NUM().
+	  Patch by Nikolai Weibull. [Bug #4789] [ruby-core:36511]
+
+Sat May 28 02:06:26 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (fill_cbuf): Fix test-all crash.
+
+Sat May 28 00:58:40 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb (method_missing): Handle [] and []= correctly.
+	  Based on a patch by Caius Durling, bug #4179 [ruby-core:33792]
+
+Fri May 27 23:56:54 2011  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_core.rb (Tester::test_text_frozen): split frozen
+	  string test. refs #4783
+
+Fri May 27 22:46:23 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (gem): install gemspec of json.  fixed #4784
+
+Fri May 27 22:45:10 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (validate_enc_binmode): do not clear textmode flag if
+	  default.  fixed #4732
+
+	* io.c (fill_cbuf): finish reading at EOF.
+
+Fri May 27 11:31:51 2011  misfo <tedwardo2@gmail.com>
+
+	* lib/rexml/text.rb (REXML::Text#initialize): prevent an error
+	  when passing a frozen string to REXML::Text.new
+
+	  dup the string passed in instead of cloning so that it's frozen
+	  state is ignored
+
+Fri May 27 08:47:26 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* thread.c (ppoll): typo bug fix.
+
+Fri May 27 08:35:04 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: hang-up at exit before calling Tk.mainloop.
+
+	* ext/tk/lib/tk/extconf.rb: cannot make on MinGW [Ruby 1.9 - Bug #4141].
+
+Thu May 27 00:34:07 2011  James Edward Gray II  <jeg2@ruby-lang.org>
+
+    * lib/csv.rb: Enhance each() to support Enumerator.
+
+Thu May 26 10:32:11 2011  James Edward Gray II  <jeg2@ruby-lang.org>
+
+    * lib/csv.rb: Documentation improvements from Ysiad Ferreiras.
+      [Ruby 1.9 - Bug #4785]
+
+Thu May 26 15:42:02 2011  Cezary Baginski <cezary.baginski@gmail.com>
+
+	* lib/xmlrpc/parser.rb (FaultException): fix to_s and inspect
+
+	* test/xmlrpc/parser.rb: test for the above
+
+Wed May 25 11:54:31 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/curses/curses.c:  Remove color constants block.
+	  [Ruby 1.9 - Bug #4748]
+
+Wed May 25 09:56:45 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/curses/curses.c:  Add missing documentation for button state, MIN
+	  and MAX comments.  Add Curses. to TABSIZE= and ESCDELAY= methods.
+	  [Ruby 1.9 - Bug #4747]
+
+Wed May 25 09:35:31 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/benchmark.rb:  Restore nodoc for Benchmark::Job and
+	  Benchmark::Report.  [Ruby 1.9 - Bug #4726]
+
+Wed May 25 09:29:38 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/pop.rb:  Hide implementation details from RDoc.
+	  [Ruby 1.9 - Bug #4711]
+
+Wed May 25 09:26:29 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/ftp.rb:  Add :nodoc: for private methods.
+	  [Ruby 1.9 - Bug #4710]
+
+Wed May 25 09:19:17 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c:  Fix Document-method declarations.  Improve
+	  Zlib::GzipFile's method catalog.  [Ruby 1.9 - Bug #4695]
+
+Wed May 25 08:22:12 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/erb.rb:  Hide documentation for implementation details of ERB.
+	  [Ruby 1.9 - Bug #4694]
+
+Wed May 25 07:58:14 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile.{mkdir,rmdir}): revert for backward
+	  compatibility.
+
+Wed May 25 07:13:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* spec/README: update the description.
+
+Wed May 25 07:12:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile.{lock,unlock}_tempfile): refactor.
+
+Tue May 24 17:30:36 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* spec/README: fix typo.
+	  patched by bowsersenior. https://github.com/ruby/ruby/pull/24
+
+Tue May 24 07:06:34 2011  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fail to start Tk.mainloop (exit immediately) on
+	  some environment (reported on [ruby-talk:381444]).
+
+	* ext/tk/lib/tk/canvas.rb: support creating a canvas item object from
+	  an item ID number.
+
+	* ext/tk/lib/tk/image.rb: import documents which are pull-requested.
+	  [Ruby 1.9 - Feature #4595]
+
+	* ext/tk/lib/tk/extconf.rb: search directories for 64bit library (e.g.
+	  /usr/lib64), add some new configure options (see README.tcltklib),
+	  and bug fix.
+
+	* ext/tk/lib/tk/README.tcltklib: modify docs for some new configure
+	  options.
+
+Tue May 24 04:01:46 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/yaml.rb: switch default YAML engine to Psych, old syck engine
+	  may be enabled via YAML::ENGINE.yamler = "syck". [ruby-core:36374]
+
+Mon May 23 09:45:26 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* include/ruby/defines.h (CASEFOLD_FILESYSTEM): Revert r30508. Forgot to
+	  include this file in the commit r31692. __APPLE__ is not
+	  CASEFOLD_FILESYSTEM again, from this time.
+
+Mon May 23 10:01:02 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Do not parse zero-tagged values as EOC. Do
+	  not let current length become negative for infinite length constructed
+	  values. Support constructed values of length zero. Added tests.
+
+Mon May 23 09:19:53 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/smtp.rb:  Document Net::SMTP::Response.  Patch by J.R. Garcia.
+	  [Ruby 1.9 - Bug #4768]
+
+Mon May 23 09:03:52 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/parallel.rb: Never Ignore SIGINT. When received
+	Interrupt, immediately puts result and exit. [ruby-dev:43571]
+
+	* lib/test/unit.rb: When received Interrupt, wait results from workers
+	and collect them. [ruby-dev:43571]
+
+Mon May 23 09:08:07 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mathn.rb:  Improve documentation.  Patch by Sandor Szucs.
+	  [Ruby 1.9 - Bug #4767]
+
+Mon May 23 08:45:55 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/ostruct.rb:  Improve documentation.  Patch by Franklin Webber.
+	  [Ruby 1.9 - Bug #4761]
+
+Mon May 23 08:35:24 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* hash.c:  Improve documentation of Hash#key.  Patch by Utkarsh
+	  Kukreti.  [Ruby 1.9 - Bug #4760]
+
+Mon May 23 08:32:59 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* enumerator.c:  Improve documentation.  Patch by Dave Copeland.
+	  [Ruby 1.9 - Bug #4757]
+
+Mon May 23 07:19:45 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* NEWS (openssl): Infinite length support. Different behavior of
+	  Constructive and Primitive constructors.
+
+Mon May 23 06:58:33 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Forbid Constructives whose value is not an
+	  Array to prevent segfault. Added test.
+
+Mon May 23 06:33:17 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Forbid Constructive without infinite
+	  length. This also prevents a segfault. Added test and improved
+	  documentation.
+
+Mon May 23 05:58:14 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Fix decoding of infinite length values.
+	  Simplified ossl_asn1_decode0 by splitting it into three separate
+	  functions. Add tests.
+	  [Ruby 1.9 - Bug #4374][ruby-core:35123]
+
+Mon May 23 04:03:46 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_initialize): Allow creation of
+	  Constructives with an explicit tag_class parameter without
+	  automatically setting tagging to :EXPLICIT. Fixes a bug when encoding
+	  infinite length primitive values.
+
+Mon May 23 04:03:46 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_cons_to_der): Add an additional
+	  EOC for infinite length Constructives that are supposed to be encoded
+	  with explicit tagging. Also tabify method correctly.
+
+Mon May 23 03:44:39 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1data_to_der): Remove redundant
+	  flag tmp_cons.
+
+Mon May 23 00:35:00 2001  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (dump_bignum, bigmul1_balance, big_split, biglsh_bang),
+	  (bigrsh_bang, big_split3, bigmul1_toom3, bigmul0): implement Toom3 (Toom-Cook)
+	  multiplication.
+
+	* include/ruby/defines.h: add format prefixes for BDIGIT and BDIGIT_DBL.
+
+Sun May 22 23:24:02 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Instead of rb_intern use static symbols to
+	  improve performance.
+
+Sun May 22 21:56:51 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Use OpenSSL constants V_ASN1_xxx instead of
+	  hardcoded numbers for initializing class_tag_map.
+
+Sun May 22 21:29:29 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* include/ruby/defines.h (CASEFOLD_FILESYSTEM): Revert r30508. See #4255.
+	  Now __APPLE__ is not CASEFOLD_FILESYSTEM again.
+
+	* load.c (loaded_feature_path, rb_feature_p, load_lock): Revert r30508.
+	  See #4255. Make $LOADED_FEATURES scanning case-sensitive again.
+
+Sun May 22 18:59:27 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_asn1.c(ossl_asn1_default_tag): avoid using RCLASS_SUPER
+	  to make it compilable. Plus, tabify and change variable definition style.
+
+Sun May 22 18:26:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (vm_xcalloc): use calloc provided by platforms.
+	  fixes #4754
+
+Sun May 22 11:44:53 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/pty/pty.c:  Improve documentation.  Patch by David Copeland.
+	  [Ruby 1.9 - Bug #4756]
+
+Sun May 22 11:26:39 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/timeout.rb:  Improve documentation.  Patch by David Copeland.
+	  [Ruby 1.9 - Bug #4755]
+
+Sun May 22 11:21:41 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/ipaddr.rb:  Improve documentation.  Patch by Sandor Szucs.
+	  [Ruby 1.9 - Bug #4753]
+
+Sun May 22 11:14:40 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/forwardable.rb:  Document def_delegator.  Patch by Sandor Szucs.
+	  [Ruby 1.9 - Bug #4752]
+
+Sun May 22 11:11:41 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/fileutils.rb:  Document block behavior of FileUtils.cd.  Patch by
+	  Bil Kleb.  [Ruby 1.9 - Bug #4751]
+
+Sun May 22 11:07:47 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/curses/curses.c:  Complete documentation.  Patch by Vincent
+	  Batts.  [Ruby 1.9 - Bug #4748]
+
+Sun May 22 09:29:43 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_set_program_name):
+	  use spawn. it prevent that other tests inherit renamed $0.
+
+Sun May 22 08:57:13 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Default tag lookup in constant time via hash
+	  instead of previous linear algorithm.
+	  [Ruby 1.9 - Feature #4309][ruby-core:34813]
+
+Sun May 22 07:54:16 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_digest.c: Explain DSS and DSS1 in documentation.
+
+Sun May 22 07:10:25 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkey_dsa.rb: Add tests for sign/verify.
+
+Sun May 22 06:07:17 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_x509cert.rb: Merge DSA-related tests from ruby_1_8
+	  branch.
+
+Sun May 22 04:11:12 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (Init_Thread): add a code comment why the meaningless
+	  line is necessary.
+
+Sun May 22 01:35:12 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: modified documentation.
+
+Sat May 21 22:46:26 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime(date_strftime_with_tmx): "%v" means "%e-%b-%Y".
+
+Sat May 21 22:14:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_extract_modeenc): accept combination hash and
+	  File::Constants. (eg. File.open('yo', :mode => File::WRONLY))
+	  [Feature #4742][ruby-core:36338]
+	* test/ruby/test_io.rb (TestIO#test_open_mode): new test.
+
+Sat May 21 21:44:14 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/test_switch_hitter.rb: new.
+
+Sat May 21 21:18:29 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_{core,parse}.c: moved nearly all core code from ext/date/lib.
+	* ext/date/lib/{date,date/format}.rb: removed nearly all code.
+
+Sat May 21 02:58:46 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/.document:  Add curses to documented extensions.
+	* ext/curses/curses.c:  Improve documentation.  Patch by Vincent Batts.
+	  [Ruby 1.9 - Bug #4747]
+
+Sat May 21 02:51:01 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb:  Improve documentation.  Patch
+	  by Pete Higgins.  [Ruby 1.9 - Bug #4746]
+
+Sat May 21 02:44:10 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/bigdecimal/lib/bigdecimal/jacobian.rb:  Document isEqual.  Patch
+	  by Kuba Fietkiewicz.  [Ruby 1.9 - Bug #4744]
+
+Sat May 21 02:22:34 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/date/lib/date/format.rb:  Document date formats.  Patch by
+	  Clinton Nixon.  [Ruby 1.9 - Bug #4743]
+
+Fri May 20 05:15:19 2011  Nobuhiro Iwamatsu  <iwamatsu@nigauri.org>
+
+	* gc.c: Fix build on m68k by 'error: too few arguments to
+	  function 'mark_locations_array''.
+
+Fri May 20 04:23:42 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/scanf.rb:  Improve documentation.  Patch by Gabe McArthur.
+	  [Ruby 1.9 - Bug #4735]
+
+Fri May 20 00:58:01 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/ibm737-tbl.rb: greek code page.  fixes #4738
+
+Thu May 19 14:44:05 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_signal.rb (test_signal_requiring): skip on Windows.
+	  we can send SIGINT only to pid 0 and the process itself.
+
+Thu May 19 09:07:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib: revert r31635-r31638 and untabify with expand(1).
+
+Thu May 19 07:47:26 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkey_rsa.rb: Add tests for sign/verify.
+
+Thu May 19 07:19:16 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey.c: Add documentation.
+
+Thu May 19 07:06:56 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/benchmark.rb:  Fix indentation.
+	* lib/net/imap.rb:  Fix indentation of regular expressions.
+
+Thu May 19 06:36:11 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/imap.rb:  Fix indentation of EOF for heredoc.
+	* lib/debug.rb (Commands):  Fix indentation of EOHELP for heredoc.
+
+Thu May 19 06:30:38 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mkmf.rb:  Fix indentation of EOM for heredoc.
+
+Thu May 19 06:16:41 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib:  Convert tabs to spaces for ruby files per
+	  https://bugs.ruby-lang.org/projects/ruby/wiki/DeveloperHowto#coding-style
+	  Patch by Steve Klabnik [Ruby 1.9 - Bug #4730]
+	  Patch by Jason Dew [Ruby 1.9 - Feature #4718]
+
+Thu May 19 06:06:07 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/cgi/util.rb:  Improve documentation.  Patch by Clinton Nixon.
+	  [Ruby 1.9 - Bug #4733]
+	* lib/cgi/core.rb:  ditto
+	* lib/cgi/cookie.rb:  ditto
+
+Thu May 19 06:02:21 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/tempfile.rb:  Document Dir.mkdir and Dir.rmdir.  Patch by Clinton
+	  Nixon.  [Ruby 1.9 - Bug #4728]
+
+Thu May 19 05:57:52 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* encoding.c:  Improve documentation for Encoding#default_external and
+	  Encoding#default_internal.
+
+Wed May 18 22:45:26 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/lib/console/size.rb (IO#console_size): new
+	  method.  (EXPERIMENTAL)
+
+Wed May 18 22:41:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h: add for internal use only.
+
+Wed May 18 22:36:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (setup_exception): internal exception should be hidden
+
+Wed May 18 20:25:04 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/timeout.rb (Timeout#timeout): don't leak "execution expired"
+	  exception. [Bug #4283] [ruby-core:34534].
+
+Wed May 18 06:09:24 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/cmath.rb:  Add some examples and improve documentation.  Patch by
+	  Sandor Szucs.  [Ruby 1.9 - Bug #4727]
+
+Wed May 18 05:40:31 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/benchmark.rb:  Remove nodoc from Benchmark::Job and
+	  Benchmark::Report.  Patch by Sandor Szucs.  [Ruby 1.9 - Bug #4726]
+
+Wed May 18 05:29:26 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/compat.rb:  Improve documentation.  Patch by Sandor
+	  Szucs.  [Ruby 1.9 - Bug #4725]
+
+Wed May 18 05:10:35 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/tracer.rb:  Improve documentation.  Patch by Richard Ramsden.
+	  [Ruby 1.9 - Feature #4720]
+
+Wed May 18 04:53:41 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/cmath.rb:  Improve documentation.  Patch by Jason Dew.
+	  [Ruby 1.9 - Feature #4717]
+
+Wed May 18 04:50:24 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/ftp.rb:  Improve documentation.  Patch by Vincent Batts.
+	  [Ruby 1.9 - Bug #4710]
+
+Wed May 18 03:14:49 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* test/test_singleton.rb:  Add tests from lib/singleton.rb.  Patch by
+	  Pete Higgins.  [Ruby 1.9 - Bug #4715]
+
+Wed May 18 03:03:07 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/singleton.rb:  Improve documentation.  Patch by Pete Higgins.
+	  [Ruby 1.9 - Bug #4709]
+
+Tue May 17 21:24:04 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_lock): remove remove_signal_thread_list() call.
+	  It's meaningless because lock_interrupt doesn't call
+	  add_signal_thread_list().
+
+Tue May 17 20:20:49 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (rb_thread_struct): add volatile to
+	  transition_for_lock because it is not protected by lock.
+
+Tue May 17 20:08:53 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* LEGAL (missing/{elf,tgamma,lgamma_r}.c): they've been replaced by
+	  public domain implementations.
+
+	* LEGAL (vsnprintf.c): it has moved to srcdir from missing/.
+
+	* LEGAL (missing/crypt.c): list its original license.
+
+Tue May 17 19:54:26 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* LEGAL (configure): add missing/setproctitle.c
+
+Tue May 17 19:35:01 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+	Fix FreeBSD test failure.
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_set_program_name):
+	  use ps -o command instead of ps -o cmd. FreeBSD doesn't support
+	  -o cmd option.
+
+Tue May 17 08:04:26 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_digest.c: Add documentation.
+
+Tue May 17 07:14:58 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Improve documentation of proxy configuration
+	  methods.  Patch by Alf Mikula.  [Ruby 1.9 - Bug #4714]
+
+Tue May 17 07:09:01 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/pop.rb:  Improve documentation.  Patch by Vincent Batts.
+	  [Ruby 1.9 - Bug #4711]
+	* lib/net/telnet.rb:  ditto
+
+Tue May 17 07:00:41 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Fix nodoc for Net::HTTP::version_1_1?.  Patch by
+	  Alf Mikula.  [Ruby 1.9 - Bug #4713]
+
+Tue May 17 06:56:26 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/optparse.rb:  Add link to make_switch to improve documentation.
+	  Patch by David Copeland.  [Ruby 1.9 - Bug #4708]
+
+Tue May 17 06:50:40 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/observer.rb:  Improve documentation.  Patch by David Copeland.
+	  [Ruby 1.9 - Bug #4707]
+
+Tue May 17 06:42:53 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/logger.rb:  Improve documentation.  Patch by David Copeland.
+	  [Ruby 1.9 - Bug #4706]
+
+Tue May 17 06:28:14 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/gserver.rb:  Improve documentation.  Patch by David Copeland.
+	  [Ruby 1.9 - Bug #4705]
+
+Tue May 17 06:21:15 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/cgi.rb: Add toplevel documentation to class CGI
+	* lib/cgi/session.rb: Add overview documentation to CGI::Cookie
+	* lib/cgi/html.rb:  Don't add CGI::TagMaker documentation to CGI.
+	  Patch by David Copeland.  [Ruby 1.9 - Bug #4704]
+	* lib/cgi/core.rb:  Clean up CGI documentation.  Patch by David
+	  Copeland.  [Ruby 1.9 - Bug #4704]
+	* lib/cgi/cookie.rb:  Clean up CGI::Cookie documentation.  Patch by
+	  David Copeland.  [Ruby 1.9 - Bug #4704]
+
+Tue May 17 05:52:30 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/digest: Improve documentation of Digest, Digest::HMAC and
+	  Digest::SHA2.  Patch by Pete Higgins.  [Ruby 1.9 - Bug #4702]
+
+Tue May 17 03:51:42 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/abbrev.rb:  Hide copyright and revision information from RDoc.
+	  Inspired by patch from David Copeland, bug #4703.
+
+Tue May 17 03:33:21 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/timeout.rb (module Timeout):  Hide internal constants.  Patch by
+	  Pete Higgins.  [Ruby 1.9 - Bug #4701]
+
+Mon May 16 11:21:09 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in, win32/Makefile.sub (RUBY_SO_NAME): add CPU as prefix
+	  of RUBY_SO_NAME on x64/ia64 mswin/mingw.
+	  [Feature #4602]
+
+Mon May 16 08:00:05 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc.rb:  Update to RDoc 3.6.1, allows OpenSSL::Digest to be
+	  found.
+
+Mon May 16 05:49:54 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/drb/acl.rb:  Add documentation.
+
+Mon May 16 05:13:20 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Add documentation.
+
+Mon May 16 00:32:05 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_signal.rb (TestSignal#test_signal_process_group):
+	  skip if the platform doesn't have :pgroup capability. (i.e. skip
+	  if mswin32)
+
+Sun May 15 23:53:31 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/intern.h: resurrect old rb_fd_copy().
+	* thread.c (rb_fd_copy): ditto.
+
+Sun May 15 23:45:11 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/intern.h: remove rb_fd_copy() to rb_fd_dup() and
+	  rb_w32_fdcopy() to rb_w32_fd_dup().
+	* win32/win32.c: ditto.
+	* thread.c: ditto.
+
+Sun May 15 22:26:39 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* signal.c (rb_f_kill): accept '-SIGXXX' style signal with Symbol or
+	  implicit conversion with #to_str. [ruby-dev:43169] fixes #4362
+	* test/ruby/test_signal.rb (test_signal_process_group): add a test
+	  for send signal to process group.
+
+Sun May 15 21:22:35 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* cont.c (cont_init): clear macihne_stack_start/end of saved thread to
+	  prevent mark machine stack of GC'ed Thread. root Fiber is not
+	  initialized by fiber_init().
+	  based on a patch by Serge Balyuk [ruby-core:35891] fixes #4612
+	* test/ruby/test_fiber.rb (test_gc_root_fiber): add test for it.
+
+Sun May 15 21:04:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (econv_init): revert r31353.  [ruby-dev:43512]
+
+Sun May 15 03:39:35 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c:  Improve documentation.  Patch by Vincent Batts.
+	  [Ruby 1.9 - Bug #4695]
+
+Sun May 15 03:23:46 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/erb.rb:  Document ERB::Compiler.  Patch by Simon Chiang.
+	  [Ruby 1.9 - Bug #4694]
+
+Sun May 15 00:58:47 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+	fix mswin32 build error.
+
+	* missing/setproctitle.c: add #ifdef HAVE_UNISTD_H.
+	* win32/Makefile.sub (MISSING): add setproctitle.obj
+
+Sat May 14 22:45:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* missing/setproctitle.c: add to include "ruby/util.h".
+
+Sat May 14 19:52:22 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_set_program_name):
+	  add for $0 test.
+
+Sat May 14 19:50:46 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* missing/setproctitle.c (compat_init_setproctitle): use
+	  ruby_strdup() instead of strdup().
+
+Sat May 14 19:37:31 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/missing.h: add setproctitle() declaration.
+	* missing/setproctitle.c: added.
+	* configure.in: add check for missing/setproctitle.c.
+
+	* ruby.c (ruby_process_options): add to call compat_init_setproctitle().
+	* ruby.c (set_arg0): remove all platform specific code. it's
+	  moved to missing/setproctitle.c.
+	* ruby.c (origarg): remove len field. It's no longer used.
+	* ruby.c (get_arglen): removed.
+
+	This patch makes a lot of cleanup set_arg0 related code and fixes
+	[Feature #4689].
+
+Sat May 14 17:42:21 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* process.c (rb_proc_times): improve documentation.
+	  [ruby-core:35785] fixes #4581, reported by Andrew Grimm.
+
+Sat May 14 12:12:54 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkey_dsa.rb: Add basic tests and tests that
+	  ensure new public key PEM encoding behavior and ensure backward
+	  compatibility.
+	  [Ruby 1.9 - Bug #4422] [ruby-core:35328]
+	* test/openssl/test_pkey_rsa.rb: Remove line with 'puts'.
+
+Sat May 14 12:06:49 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/context.rb (class RDoc):  Fix infinite loop caused by
+	  re-encountering BasicObject.
+
+Sat May 14 10:32:36 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkey_rsa.rb: Add tests that ensure new public key
+	  encoding behavior and also ensure backward compatibility.
+	  [Ruby 1.9 - Bug #4421] [ruby-core:35327]
+
+Sat May 14 09:50:10 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/yaml/dbm.rb:  Add documentation.  Patch by Justin Collins.
+	  [Ruby 1.9 - Bug #4693]
+	* lib/yaml/store.rb:  ditto
+
+Sat May 14 09:31:43 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc.rb: Updated to RDoc 3.6
+
+Sat May 14 07:30:29 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: released a new gem, so increasing version.
+
+Sat May 14 05:08:32 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_digest.c
+	* ext/openssl/ossl_pkey.c
+	* ext/openssl/ossl_pkey.h
+	* test/openssl/pkey/test_pkey_rsa.rb
+	  Reverted premature commit. Sorry for the noise!
+
+Sat May 14 05:02:58 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/uri.rb:  Add toplevel documentation.  Patch by Vincent Batts.
+	  [Ruby 1.9 - Bug #4690]
+
+Sat May 14 04:19:06 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* NEWS: Describe altered behaviour for RSA and DSA public key
+	  encoding. [Ruby 1.9 - Bug #4421, Bug #4422]
+	  [ruby-core:35327,35328]
+
+Sat May 14 02:57:52 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/ipaddr.rb (unless Socket):  Document valid*? methods.  Patch by
+	  Sebastian Martinez.  [Ruby 1.9 - Feature #4687]
+
+Sat May 14 02:54:04 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rexml/functions.rb:  Add some documentation for REXML::Functions.
+	  Patch by Sebastian Martinez.  [Ruby 1.9 - Feature #4688]
+
+Sat May 14 02:51:42 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/resolv.rb:  Hide private method and state-tracking constants from
+	  RDoc.  Patch by Mark Turner.  [Ruby 1.9 - Feature #4691]
+
+Fri May 13 19:23:21 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* numeric.c (flo_coerce): Add #flo_coerce documentation.
+	  Patch by Sebastian Martinez.
+	  https://github.com/ruby/ruby/pull/21
+
+Fri May 13 18:42:22 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* README.EXT: fix typo.  Patch by William Blackerby.
+	  https://github.com/ruby/ruby/pull/19
+
+	* README.EXT.ja: ditto.
+
+Fri May 13 15:22:34 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_select): check invalid handle before doing
+	  select operations.  see [ruby-dev:43513], [ruby-dev:43535]
+
+Fri May 13 08:34:00 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/rdoc.rb:  Output summary after documentation report.
+	* lib/rdoc/stats/normal.rb:  Don't output information for users when
+	  we're not on a TTY
+
+Fri May 13 07:49:02 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/fileutils.rb:  Hide internal methods from RDoc.  Patch by Darragh
+	  Curran.  [Ruby 1.9 - Bug #4684]
+
+Fri May 13 07:36:23 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/httpservlet/erbhandler.rb:  Add documentation.
+
+Fri May 13 07:04:33 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mathn.rb:  Fix indentation.  Patch by Jason Dew.
+	  [Ruby 1.9 - Feature #4682]
+
+Fri May 13 06:50:43 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mathn.rb:  Add documentation.  Patch by Jason Dew.  [Ruby 1.9 -
+	  Feature #4667]
+
+Fri May 13 05:44:19 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/logger.rb (class Logger):  Document Logger#datetime_format.
+	  Patch by Sergio Gil Perez de la Manga.  [Ruby 1.9 - Bug #4678]
+
+Fri May 13 05:39:11 2011  Eric Hodel <drbrain@segment7.net>
+
+	* re.c (Init_Regexp): Document option constants.  Patch by Vincent
+	  Batts.  [Ruby 1.9 - Bug #4677]
+	* lib/uri/common.rb (module URI):  Documentation for URI.  Patch by
+	  Vincent Batts.  [Ruby 1.9- Bug #4677]
+	* lib/uri/ftp.rb (module URI):  ditto
+	* lib/uri/generic.rb (module URI):  ditto
+	* lib/uri/http.rb (module URI):  ditto
+	* lib/uri/https.rb (module URI):  ditto
+	* lib/uri/ldap.rb (module URI):  ditto
+	* lib/uri/ldaps.rb (module URI):  ditto
+	* lib/uri/mailto.rb (module URI):  ditto
+	* process.c (Init_process):  Document Process constants.  Patch by
+	  Vincent Batts.  [Ruby 1.9- Bug #4677]
+
+Fri May 13 05:16:38 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rss/atom.rb (module RSS):  Document URIs.  Patch by Mark Turner.
+	  [Ruby 1.9 - #4671]
+	* lib/rss/rss.rb (module RSS):  Document exception classes.  Patch by
+	  Mark Turner.  [Ruby 1.9 - #4671]
+
+Fri May 13 02:15:18 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (select_internal): remove unused variable (interrupt_flag).
+
+Thu May 12 18:24:34 2011  Kouhei Sutou  <kou@clear-code.com>
+
+	* configure.in: limit to "T" type for prefix of external symbols
+	  because x86_64-w64-mingw32-gcc on Debian GNU/Linux generates the
+	  following symbol:
+	    0000000068483390 D _GLOBAL__F__conftest_external
+
+	  Approved by nobu.
+
+Thu May 12 14:50:52 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/dl/test_base.rb (Fiddle::LIBC_SO): its always msvc*.dll on
+	  mswin/mingw.
+
+Thu May 12 14:47:53 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (Logging.postpone): copy only when temporary logfile
+	  exists.
+
+Thu May 12 12:24:22 2011  Nobuhiro Iwamatsu  <iwamatsu@nigauri.org>
+
+	* ext/openssl/ossl_ssl.c: By trunk@31346, function check of SSLv2 is
+	  executed.
+	  However, the problem is not revised in this.
+	  This adds the control of using function of SSLv2 in made macro by
+	  function check.
+
+Thu May 12 08:10:46 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/set.rb (class Set):  Add nodoc to internal-use methods.  Patch
+	  by Pete Higgins.  [Ruby 1.9 - Bug #4665]
+
+Thu May 12 08:01:14 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey_ec.c: Allow encryption when PEM-encoding
+	  Elliptic Curve private keys.
+	  [ruby-core:35329] [Bug #4423]
+
+Thu May 12 07:54:59 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (rb_obj_equal):  Add documentation.  Patch by Vincent Batts.
+	  [Ruby 1.9 - Bug #4664]
+	* lib/rexml:  ditto
+	* lib/mkmf.rb:  ditto
+	* ext/socket/lib/socket.rb:  ditto
+
+Thu May 12 07:30:08 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* Various .document files: Update .document files to match files which
+	  have documentation.
+
+Thu May 12 07:18:45 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey_dsa.c: Use generic X.509 SubjectPublicKeyInfo
+	  format for PEM-encoding DSA public keys.
+	  [ruby-core:35328] [Bug #4422]
+
+Thu May 12 06:27:31 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey_rsa.c: Use generic X.509 SubjectPublicKeyInfo
+	  format for encoding RSA public keys.
+	  [ruby-core:35327] [Bug #4421]
+
+Wed May 11 19:45:27 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/forwardable.rb: support 'delegate :foo => :bar' for to meet
+	  by specification of RDOC.
+
+Wed May 11 08:36:38 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick: Add documentation for WEBrick::HTTPAuth
+
+Wed May 11 03:06:35 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rss.rb: Add documentation for RSS.  Patch by Steve Klabnik.
+	  [Ruby 1.9 - Bug #4663]
+
+Tue May 10 14:50:32 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Add option for hiding skip messages when test
+	  ends. #4657
+
+	* test/testunit/test_hideskip.rb, test/testunit/test4test_hideskip.rb:
+	  test for above.
+
+Tue May 10 10:53:04 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* common.mk (rdoc): Add rdoc-coverage rule
+
+Tue May 10 09:13:21 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick: Add Documentation
+
+Tue May 10 04:22:09  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/log.rb: Hide copyright info from ri
+	* lib/webrick/httpstatus.rb: ditto
+	* lib/webrick/htmlutils.rb: ditto
+	* lib/webrick/httpversion.rb: ditto
+	* lib/webrick/version.rb: ditto
+	* lib/webrick/httpauth/userdb.rb: ditto
+	* lib/webrick/httpauth/authenticator.rb: ditto
+	* lib/webrick/accesslog.rb: ditto
+
+Mon May  9 20:57:13 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_{complex,rational}.rb: added tests.
+
+Mon May  9 20:29:44 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (string_to_c_internal): a refactoring.
+
+Mon May  9 18:33:05 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c: Improve documentation for String#start_with? and
+	  String#end_with?. fixes #4652
+	  patched by Andrew Grimm <andrew.j.grimm at gmail.com>
+
+Mon May  9 13:49:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* complex.c (string_to_c_internal): support scientific notation.
+	  patched by Tinco Andringa. https://github.com/ruby/ruby/pull/16
+	  [ruby-core:36046][Bug #4655]
+
+Mon May  9 11:52:48 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (int_ord): remove K&R style.
+	  patched by Daehyub Kim. https://github.com/ruby/ruby/pull/17
+
+Sun May  8 22:17:24 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_{complex2,complexrational}.rb: use skip.
+	* test/date/*.rb: ditto.
+
+Sun May  8 21:02:31 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_{complex2,complexrational}.rb: NEVER SKIP.
+
+Sun May  8 21:01:21 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/test_date_base.rb: fixed.
+
+Sun May  8 20:54:11 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/*.rb: NEVER SKIP.
+
+Sun May  8 20:37:33 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/*.rb: reverted 31432.
+
+Sun May  8 20:32:43 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/*.rb: reverted 31483.
+
+Sun May  8 19:39:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (native_cond_timedwait): add to care EINTR.
+	* thread_pthread.c (thread_timer): remove EINTR check.
+
+Sun May  8 19:04:15 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/time.rb (xmlschema): avoid passing any negative numbers.
+
+Sun May  8 18:40:03 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_{parse,strptime}.c: introduced some macros.
+
+Sun May  8 17:17:13 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/*.rb: use skip /w messages.
+
+Sun May  8 17:04:55 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib/date/format.rb (_httpdate): omitted to call zone_to_diff.
+
+Sun May  8 16:56:19 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (date_s_test_*): use macros.
+
+Sun May  8 10:24:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c: cleanup signal_thread_list related ifdef.
+	  1) we don't have to use #ifdef FOO-PLATFORM directly 2) About
+	  half #ifdef didn't care symbian properly.
+
+Sun May  8 05:19:37 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/io/wait/test_io_wait.rb: Linux socketpair(2) only support
+	  AF_UNIX, but windows socketpair doesn't support it. we can't
+	  avoid platform check. sigh!
+
+Sun May  8 00:13:05 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/io/wait/test_io_wait.rb: use Socket.pair instead of pipe.
+	  Windows can only treat a socket.
+
+Sat May  7 22:43:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_fd_zero): remove redundant zero fill.
+
+Sat May  7 22:38:04 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_fd_init): remove volatile qualifier.
+
+Sat May  7 22:34:29 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_fd_init_copy): new internal api. It provide efficient
+	  copy constructor semantics.
+	* thread.c (do_select): use rb_fd_init_copy().
+
+Sat May  7 15:18:06 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+	fix incorrect native_cond_signal call when deadlock was detected.
+
+	* thread.c (lock_func): decrement cond_waiting if timeout was happen.
+
+Sat May  7 18:28:37 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (USE_MONOTONIC_COND): check the availability
+	  more strictly.
+
+	* thread_pthread.h (rb_thread_cond_t): ditto.
+
+Sat May  7 15:15:10 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	fix win32 native_cond_timedwait() makes SIGSEGV.
+
+	* thread_win32.h (rb_thread_cond_struct): add prev field instead of
+	  last. (ie cond_event_entry is now using double linked list instead of
+	  single)
+	* thread_win32.c (cond_event_entry): add prev field.
+
+	* thread_win32.c (__cond_timedwait): remove entry properly if timeout
+	  was happen.
+
+	* thread_win32.c (native_cond_signal): change for double linked list.
+	* thread_win32.c (native_cond_broadcast): ditto.
+	* thread_win32.c (native_cond_initialize): ditto.
+
+Sat May  7 12:41:04 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+	fix mutex deadlock test hang-up.
+
+	* thread_win32.c (abs_timespec_to_timeout_ms): fix 1000x calculation
+	  mistake. (ie fix hang-up native_cond_timedwait())
+
+Sat May  7 03:14:13 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	sleep_cond use monotonic time if possible.
+
+	* thread_pthread.c (native_thread_init): change sleep_cond
+	  attribute to monotonic.
+	* thread_pthread.c (native_sleep): use native_cond_timeout().
+
+	* thread_pthread.c (native_cond_timeout): add overflow care.
+	* thread_win32.c (native_cond_timeout): ditto.
+
+Sat May  7 02:49:12 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+	  fix win32 compile error.
+
+	* thread_win32.c (RB_CONDATTR_CLOCK_MONOTONIC): define
+	  RB_CONDATTR_CLOCK_MONOTONIC always.
+	* thread_pthread.c (RB_CONDATTR_CLOCK_MONOTONIC): ditto.
+
+Sat May  7 02:29:41 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	  mutex: deadlock check timeout use monotonic time.
+
+	* thread_pthread.c (native_cond_timeout): new internal api.
+	  it calculate a proper time for argument of native_cond_timedwait().
+	* thread_win32.c (native_cond_timeout): ditto.
+
+	* thread_pthread.c (thread_timer): use native_cond_timeout()
+	  instead of get_ts.
+	* thread.c (lock_func): ditto.
+
+	* thread_pthread.c (get_ts): removed. use native_cond_timeout().
+	* thread.c (init_lock_timeout): ditto.
+
+Sat May  7 01:54:21 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (get_ts): add monotonic clock capability.
+	* thread_pthread.c (rb_thread_create_timer_thread): use monotonic
+	  clock if possible.
+
+Sat May  7 01:43:37 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.h (rb_thread_cond_t): add clockid field. it's
+	  no longer an alias of pthread_cond_t.
+	* thread_pthread.c: adapt new rb_thread_cond_t type.
+	* thread.c (mutex_alloc): ditto.
+	* thread_win32.c (native_cond_initialize): ditto.
+	* configure.in: add check for pthread_cond_attr_setclock() and
+	  clockid_t type.
+
+Fri May  6 23:29:47 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_wait_for_single_fd): use ppoll() instead of poll()
+	  if possible. based on a patch from Eric Wong. [ruby-core:36003].
+
+Fri May  6 23:13:43 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: remove nanosleep check. we no longer use it.
+	  r20124 removed last usage.
+
+Fri May  6 22:35:56 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c (mktime_do): extra digits are not used.
+
+Fri May  6 17:43:07 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/syck/rubyext.c (mktime_do): remove unused variable offset.
+
+	* ext/syck/syck.h: use #ifdef instead of #if DEBUG.
+
+Fri May  6 16:27:33 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/date_core.c (DAY_IN_NANOSECONDS): refix: 31438.
+	  check with LONG_MAX and cast as long; without this the calculation
+	  will be done as int and overflow.
+
+Fri May  6 15:01:11 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* ext/syck/rubyext.c (mktime_do): avoid buffer overrun, by
+	  silently ignoring lesser significant digits.  Required buffer
+	  length can be computable so you might at first think of
+	  allocating enough memory space on the fly using alloca().  That
+	  is a wrong idea because when using alloca there is always risk
+	  of integer overflow.  A function that accepts outer-process
+	  resources like this should not blindly trust its inputs.  In
+	  this particular case we just want to generate milliseconds
+	  resolution by strtod() so the string in question needs no more
+	  length than what we originally have.  Ignoring lesser
+	  significant digits should suffice I believe.
+
+Fri May  6 14:25:53 2011  Tinco Andringa <mail@tinco.nl>
+
+	* ext/syck/rubyext.c (mktime_do): YAML.load time correctly parse
+	  usecs smaller than 1 fixes #4571
+
+Thu May  5 22:23:34 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (native_mutex_reinitialize_atfork): removed
+	  unused macro.
+	* thread_win32.c (native_mutex_reinitialize_atfork): ditto.
+
+Thu May  5 22:09:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_core.c (DAY_IN_NANOSECONDS): long long int is not
+	  available on all platforms.
+
+Thu May  5 17:36:31 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* eval.c (frame_func_id): store result of method_entry_of_iseq() to
+	  cfp->me because method_entry_of_iseq() might become expensive.
+
+Thu May  5 15:03:51 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* eval.c (frame_func_id): __method__ return different name from
+	  methods defined by Module#define_method with a same block.
+	  [ruby-core:35386] fixes #4606
+	* eval.c (method_entry_of_iseq): new helper function. search control
+	  frame stack for a method entry which has given iseq.
+	* test/ruby/test_method.rb: add tests for #4606
+
+Wed May  4 22:13:09 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_vm4_pipe.rb: Reduced iterations. Too slow benchmark
+	  is bad.
+	* benchmark/bm_vm4_thread_pass.rb: ditto.
+
+Wed May  4 22:08:22 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/date/test_date_base.rb: don't use no message skip().
+
+Wed May  4 21:11:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_io_select2.rb: reduce number of using file
+	  descriptors. because gdb need some fds.
+
+Wed May  4 19:00:59 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_wait_for_single_fd): Fix wrong return value.
+	* test/-ext-/wait_for_single_fd/test_wait_for_single_fd.rb
+	  (TestWaitForSingleFD#test_wait_for_closed_pipe): test for it.
+
+Wed May  4 18:46:39 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/-test-/wait_for_single_fd: New. for testing
+	  rb_wait_for_single_fd() internal function.
+	  The patch was written by Eric Wong. [ruby-core:35991]
+
+	* test/-ext-/wait_for_single_fd/test_wait_for_single_fd.rb: ditto.
+
+Wed May  4 12:46:25 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_wait_for_single_fd): Added POLLNVAL check.
+	  based on a patch from Eric Wong at [ruby-core:35991].
+
+Wed May  4 11:51:01 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_f_select): remove useless ifdef.
+
+Wed May  4 11:42:47 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/socket/init.c (wait_connectable): fix error handling code.
+	  RB_WAITFD_OUT is turned on even though an error occur.
+
+Wed May  4 10:12:39 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/readline/readline.c (readline_event): use rb_wait_for_single_fd().
+	The patch was written by Eric Wong. [Ruby 1.9 - Feature #4531]
+
+Wed May  4 10:10:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/socket/init.c (wait_connectable): use rb_wait_for_single_fd().
+	  The patch was written by Eric Wong. [Ruby 1.9 - Feature #4531]
+
+	* ext/socket/init.c (try_wait_connectable, wait_connectable_ensure):
+	  removed.
+
+Wed May  4 10:07:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/io/wait/wait.c (io_wait): use rb_wait_for_single_fd().
+	  The patch was written by Eric Wong. [Ruby 1.9 - Feature #4531]
+
+Wed May  4 10:01:27 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_wait_for_single_fd): new. poll(2) based backend for
+	  rb_wait_for_single_fd().
+	  Now only Linux uses it.
+
+	The patch was written by Eric Wong. [Ruby 1.9 - Feature #4531]
+
+Wed May  4 09:56:57 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_wait_for_single_fd): new.
+	* thread.c (select_single): select(2) based backend for
+	  rb_wait_for_single_fd().
+
+	* io.c (make_writeconv): use rb_wait_for_single_fd() instead of
+	  rb_thread_fd_select().
+	* io.c (rb_io_wait_readable): ditto.
+	* thread.c (rb_thread_wait_fd_rw): ditto.
+
+	* io.c (wait_readable): removed.
+	* thread.c (init_set_fd): new helper function.
+	* include/ruby/io.h (RB_WAITFD_IN, RB_WAITFD_PRI, RB_WAITFD_OUT):
+	  new constant for rb_single_wait_fd().
+
+	The patch was written by Eric Wong. [Ruby 1.9 - Feature #4531]
+
+Wed May  4 08:04:59 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: fix time dumping so that
+	  Syck can load UTC times that Psych dumps.
+
+Wed May  4 07:33:00 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_fd_copy): fix wrong argument.This issue was pointed
+	  out by Eric Wong. [ruby-core:35982]
+
+Tue May  3 20:29:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/fileutils/test_fileutils.rb (TestFileUtils#test_chmod_symbol_mode):
+	  Skip sticky bit test if the platform is FreeBSD. It doesn't allow to
+	  change sticky bit if a target is regular file.
+
+Tue May  3 18:23:57 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/date/test_date.rb (TestDate#test_coerce):
+	  test for [ruby-core:35127].
+
+Tue May  3 04:27:53 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_select): preserve errno if no error
+	  occurred.
+
+Tue May  3 03:57:04 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_w32_fdcopy): add prototype.  fixes
+	  #4640
+
+Mon May  2 01:02:04 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/fileutils.rb (FileUtils#chmod): accept symbolic mode argument.
+	  The patch was written by takkanm. [ruby-core:26029][Feature #2190]
+
+	* lib/fileutils.rb (FileUtils#fu_mode): new helper function.
+	* lib/fileutils.rb (FileUtils#symbolic_modes_to_i): ditto.
+	* lib/fileutils.rb (FileUtils#mode_mask): ditto.
+	* lib/fileutils.rb (FileUtils#user_mask): ditto.
+
+	* test/fileutils/test_fileutils.rb (TestFileUtils#test_chmod_symbol_mode):
+	  new test for the above symbolic mode.
+	* test/fileutils/test_fileutils.rb (TestFileUtils#test_chmod_R): ditto.
+
+Mon May  2 00:36:12 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/socket/init.c (rsock_connect): add to care EINTR. based
+	  on a patch from Eric Wong at [ruby-core:35621][Bug #4555]
+
+Sun May  1 01:06:24 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_select): release GVL while waiting select().
+
+Sat Apr 30 23:10:15 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/win32.c (rb_w32_fdcopy): New. This can copy even though
+	  fdset size exceed FD_SETSIZE.
+	* include/ruby/intern.h (rb_fd_copy): use rb_w32_fdcopy()
+
+Sat Apr 30 20:18:43 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (do_select): Change argument type to rb_fdset_t.
+	  Now do_select() is free from unexpected hangup if
+	  HAVE_RB_FD_INIT=1 [Bug #4636]
+
+	* thread.c (rb_thread_fd_select, rb_thread_wait_fd_rw):
+	  adapt new argument type.
+
+	* thread.c (rb_thread_select): make dummy implementation.
+
+Sat Apr 30 20:16:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_fd_copy): Change function argument. Now
+	  rb_fd_copy() has fully copy semantics.
+	* include/ruby/intern.h: ditto.
+
+Sat Apr 30 20:11:47 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/intern.h (rb_thread_select): mark as deprecated.
+
+	* ext/io/wait/wait.c (wait_readable): use rb_thread_fd_select
+	  instead of rb_thread_select.
+	* ext/socket/init.c (wait_connectable0): ditto.
+	* ext/readline/readline.c (readline_event): ditto.
+	* io.c (rb_io_wait_readable, wait_readable, rb_io_wait_writable,
+	  wait_writable): ditto.
+
+Sat Apr 30 20:06:36 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (do_select): remove useless ifdef. time calculation
+	  is not heavy weight.
+
+Sat Apr 30 16:48:36 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_io_select3.rb: New.
+
+Sat Apr 30 16:27:09 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (copy_stream_body, rb_io_s_copy_stream): move rb_fd_init()
+	  from copy_stream_body to rb_io_s_copy_stream. fds of passing
+	  rb_fd_term() have to be guaranteed initialized.
+
+Sat Apr 30 16:13:17 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_io_select.rb, benchmark/bm_io_select2.rb: New.
+	  based on a patch from Eric Wong at [Feature #4531]
+
+Sat Apr 30 03:25:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/io/wait/test_io_wait.rb: New. for testing ext/io/wait.
+	  the patch was written by Eric Wong. [Feature #4531]
+
+Sat Apr 30 00:34:56 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/win32.h: remove redundant declaration of
+	  rb_w32_time_subtract().
+
+Sat Apr 30 00:16:40 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (gvl_init): fix hangup if GVL_SIMPLE_LOCK=1.
+	  We don't have to call mutex_unlock() before initialize it!
+
+Fri Apr 29 13:15:15 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_win32.c (native_cond_timedwait): New. r31373 caused
+	  win32 build failure.
+
+	* thread_win32.c (__cond_timedwait, abs_timespec_to_timeout_ms):
+	  New helper functions.
+
+	* win32/win32.c (rb_w32_time_subtract): rename from subtract and
+	  remove static.
+
+Fri Apr 29 10:43:09 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/bm_vm4_pipe.rb: Add two new benchmark for GVL
+	  performance. They was written by Koichi Sasada.
+	* benchmark/bm_vm4_thread_pass.rb: ditto.
+
+Fri Apr 29 10:25:31 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_method.c (rb_clear_cache_by_class): Revert r29673. It made
+	  a segmentation fault regression. [Bug #4289][ruby-core:34554].
+
+Fri Apr 29 10:24:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (make_writeconv): do not add textmode newline decorator if any
+	  newline decorator is set already.  fixes #4618, fixes #4619
+
+Fri Apr 29 10:17:42 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (lock_func): small cleanup.
+
+Fri Apr 29 10:07:13 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_lock, lock_func): Avoid busy loop and
+	  performance regression. bm_vm3_thread_mutex.rb performance
+	  change from 109.064sec to 16.331sec. [Feature #4607]
+
+	* thread.c (init_lock_timeout): New helper function.
+
+Thu Apr 28 16:15:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{win32.c,dir.h} (rb_w32_uopendir): new API to pass UTF-8 path.
+
+	* win32/win32.c (opendir_internal, rb_w32_opendir): extract and merge
+	  common part of rb_w32_opendir() and rb_w32_uopendir().
+
+	* dir.c (do_opendir, glob_helper): encoding.
+
+	* dir.c (dir_initialize, do_opendir): convert path to UTF-8 and call
+	  rb_w32_uopendir() instead of rb_w32_opendir() on Windows.
+	  fixes #4491, reported by Joey Zhou.
+
+Thu Apr 28 15:32:53 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/dl/test_base.rb (DL::LIBC_SO): its always msvc*.dll on
+	  mswin/mingw.
+
+Thu Apr 28 06:07:06 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/csv.rb (CSV::open): suppress universal newline decorator.
+	  fixes #4603
+
+	* lib/csv.rb (CSV.read): no mode is needed.
+
+Thu Apr 28 06:06:56 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_extract_modeenc, rb_f_backquote): set default text
+	  mode.  fixes #4619
+
+	* io.c (pipe_open): set universal newline decorator if needed.
+
+Wed Apr 27 11:33:08 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/emoji_iso2022_kddi.trans: ISO-2022-JP-KDDI doesn't have
+	  CP932 UDA. Another reason is emacs-mule: the implementation of
+	  stateless-iso-2022-jp doesn't support beyond 94x94 (0x7fxx);
+	  but CP932 UDA is in 7Fxx-92xx.
+
+Wed Apr 27 07:42:44 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (STRIP): use proper toolchain.  based on a patch
+	  from Jon Forums at [ruby-core:35909].  fixes #4617
+
+Wed Apr 27 01:20:59 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (date_zone_to_diff): renamed.
+	* ext/date/date_parse.c: ditto.
+	* ext/date/date_strptime.c: ditto.
+
+Wed Apr 27 01:16:59 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_find): accept Encoding objects.
+
+Wed Apr 27 00:55:56 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (econv_opts): add newline option.
+
+	* io.c (validate_enc_binmode, rb_io_extract_modeenc): set newline
+	  decorator according to open mode.
+
+	* transcode.c (rb_econv_prepare_options): new function, to prepare
+	  econv options with newline flags.
+
+	* include/ruby/encoding.h (ECONV_NEWLINE_DECORATOR_MASK): add.
+
+Wed Apr 27 00:51:01 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_truncate): fix function.
+
+	* include/ruby/win32.h (ftruncate, truncate, ftello, fseeko): non-64
+	  versions on mingw are useless because they use int32_t.  fixes #4564
+
+Wed Apr 27 00:50:33 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: modified validation methods.
+	* ext/date/lib/date.rb: ditto.
+
+Wed Apr 27 00:00:37 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (dt_lite_set_tmx): should get df value.
+
+Tue Apr 26 22:34:04 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib/date/format.rb (_iso8601): allowed day only civil
+	  date.  disallowed separatorless day only ordinal date.
+
+Mon Apr 25 21:31:36 2011  Nobuhiro Iwamatsu  <iwamatsu@nigauri.org>
+
+	* ext/openssl/extconf.rb: Should check SSLv2_*method.
+	  openssl compiled with "no-ssl2" the extconf don't fail
+	  when running `make' having this compilation errors.
+	  Patched by Laurent Arnoud. fixes #4562, #4556
+
+Mon Apr 25 20:53:32 2011  Tajima, Akio <artonx@yahoo.co.jp>
+
+	* win32/win32.c (kill): accept 0 only sig is SIGINT #4596
+
+Mon Apr 25 19:59:47 2011  Tajima, Akio <artonx@yahoo.co.jp>
+
+	* win32/win32.c (kill): accept 0 as pid, fixes #4596
+
+Mon Apr 25 16:43:45 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c (random_rand): remove unused variables.
+
+	* struct.c (rb_struct_define_without_accessor): ditto.
+
+	* strftime.c (rb_strftime_with_timespec): ditto.
+
+	* sprintf.c: ditto.
+
+	* time.c (time_asctime): remove useless GetTimeval().
+
+	* thread_pthread.c: cast to (void *) for %p.
+
+Mon Apr 25 11:02:11 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/ripper/lib/ripper/sexp.rb: fix rdoc around sexp.
+	  patched by Sho Hashimoto. fixes #4599
+
+Mon Apr 25 08:24:04 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* random.c (rb_f_rand, random_s_rand): RDocs for them.
+
+Mon Apr 25 07:18:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* random.c (random_s_rand, Init_Random): Random.rand should behave as
+	  Random::DEFAULT.rand rather than Kernel#rand.
+
+	* random.c (rand_range, random_rand): rand_range function extracted
+	  from random_rand function.
+
+	* random.c (rb_f_rand): accept a Range argument as Random#rand
+	  [ruby-dev:43427] #4605
+
+Mon Apr 25 03:31:06 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/time.rb: require 'date'.
+	* ext/date/lib/date/format.rb: removed require line.
+
+Mon Apr 25 03:08:39 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib/date/format.rb: require 'date'.
+
+Mon Apr 25 03:04:16 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib/date/format.rb (_iso8601): added a pattern.
+
+Mon Apr 25 02:51:22 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/lib/date/format.rb: require 'date_core.so'.
+	  date/format needs methods which are now in date_core.so.
+	  This breaks make rdoc which uses Date._parse from time.rb.
+
+Mon Apr 25 02:47:46 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib/date/format.rb (_iso8601): fixed a bug of regex.
+
+Mon Apr 25 02:12:26 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib/date/format.rb: an adjustment of regex.
+
+Mon Apr 25 01:58:50 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib/date/format.rb: omitted to call _parse.
+
+Mon Apr 25 01:03:03 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* string.c (rb_to_id): remove unused variable.
+
+Sun Apr 24 22:19:05 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c, rational.c: omitted some method calls.
+
+Sun Apr 24 02:57:27 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (n2i): takes long.
+
+Sun Apr 24 02:51:06 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c: reverted.
+
+Sun Apr 24 02:25:23 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/intern.h: pcc can't use __builtin_constant_p.
+
+	* vm_exec.c: change condition.
+
+Sun Apr 24 01:58:01 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/date_core.c (leap_p): suppress warning: parentheses.
+
+	* ext/date/date_core.c (date_s__parse_internal): remove unused
+	  variable "str".
+
+	* ext/date/date_parse.c (parse_ddd_cb): use RSTRING_LENINT.
+
+	* ext/date/date_strftime.c (date_strftime_with_tmx): remove unused
+	  variable.
+
+Sun Apr 24 00:34:23 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c: removed some unused macros.  use strchr()
+	  instead of index().
+
+Sat Apr 23 21:29:42 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: replacement of implementation of
+	  _parse.  [experimental]
+	* ext/date/date_parse.c: new.
+	* ext/date/lib/date/format.rb: removed ruby version of _parse.
+
+Fri Apr 22 12:04:15 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): fix rdoc.
+	  patched by burningTyger. https://github.com/ruby/ruby/pull/11
+
+Fri Apr 22 11:49:49 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/xmlrpc/create.rb (XMLRPC::Create#conv2value):
+	  XML-RPC's int is 32bit int, and Fixnum also may be beyond 32bit.
+
+	* lib/xmlrpc/create.rb (XMLRPC::Create#conv2value):
+	  XML-RPC doesn't allow Infinity and NaN.
+	  http://www.xmlrpc.com/spec
+
+Fri Apr 22 04:16:14 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (parse): strings from psych have proper taint
+	  markings.
+
+	* test/psych/test_tainted.rb: test for string taint
+
+Thu Apr 21 01:30:02 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c (rb_f_srand): fix rdoc: srand(0)'s 0 is a seed.
+	  [ruby-core:35833] fixes #4590
+
+Thu Apr 21 01:01:28 2011  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): maximum length of lpCommandLine is
+	  32,768 characters, including the Unicode terminating null character.
+
+Wed Apr 20 21:32:11 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strptime.c (date__strptime_internal): do not
+	  overwrite local variables.
+
+Wed Apr 20 14:41:28 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_each_line): check string's length when compare
+	  separator and string. [ruby-core:35815] fixes #4586
+
+Wed Apr 20 00:02:13 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-parse-partial): use position of open paren.
+
+Tue Apr 19 01:00:21 2011  Tajima Akio <artonx@yahoo.co.jp>
+
+	* test/ruby/test_io.rb (TestIO#test_cross_thread_close_fd):
+	  skip cross thread pipe close if windows
+
+Mon Apr 18 12:15:46 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_range.rb (TestRange#test_step_ruby_core_35753):
+	  avoid float error. [ruby-core:35804]
+
+Sun Apr 17 00:20:14 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_{core,strftime}.c: use struct tmx instead of vtm.
+	* ext/date/date_tmx.h: new.
+
+Sat Apr 16 22:23:52 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c (date_strftime_wo_timespec): changed
+	  the way of validation of locale modifiers.
+
+Sat Apr 16 21:55:12 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: replacement of implementation of
+	  _strptime.  [experimental]
+	* ext/date/date_strptime.c: new.
+	* ext/date/lib/date/format.rb: removed ruby version of _strptime.
+
+Sat Apr 16 10:18:30 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* vm.c (Init_VM): suppress warning: "OPT_BASIC_OPERATIONS" is not
+	  defined.
+
+Fri Apr 15 23:41:18 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ruby.c (proc_options): suppress warning:
+	  "ALLOW_DEFAULT_SOURCE_ENCODING" is not defined.
+
+Fri Apr 15 15:10:29 2011  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/uri/generic.rb (#route_from_path): Fix a bug where
+	  URI('http://h/b/').route_to('http://h/b') wrongly returned './'
+	  (should be '../b'). [Bug #4476]
+
+Fri Apr 15 14:58:06 2011  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/fileutils.rb (FileUtils#touch): Fix corrupted output when
+	  mtime is specified in addition to nocreate (and verbose).
+	  ref [ruby-dev:43401]
+
+Thu Apr 14 23:43:43 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (ruby_float_step): wrong loop condition.
+	  fixes [ruby-core:35753], reported by Joey Zhou.
+
+	* test/ruby/test_range.rb (TestRange#test_step_ruby_core_35753):
+	  test above change.
+
+Thu Apr 14 22:48:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Options#setup_options): set possible
+	  values for completion. no conversion is needed.
+
+	* lib/test/unit.rb (Test::Unit::Runner::Worker#initialize): use
+	  positional arguments instead of keyword arguments.
+
+	* lib/test/unit.rb (Test::Unit::Runner#jobs_status): io/console may
+	  not be available. use 80 as the last resort if IO#winsize and
+	  COLUMNS are unavailable.
+
+	* lib/test/unit.rb (Test::Unit::Runner::Worker#died): rename using a
+	  verb.
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): check if worker
+	  is signaled and use its exit status.
+
+	* lib/test/unit.rb (Test::Unit::Runner::Worker#dead): no longer @in
+	  and @out are separated.
+
+Thu Apr 14 21:23:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_autoload_p): search superclasses as same as actual
+	  loading.  fixes [ruby-core:35679]
+
+Thu Apr 14 21:21:06 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (frexp, modf): wrongly declared as pure in
+	  mingw math.h.
+
+	* include/ruby/win32.h (ftruncate, truncate): mingw64 misses
+	  prototypes.
+
+	* win32/win32.c (rb_w32_read): suppress warning.
+
+Thu Apr 14 19:55:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/fileutils.rb (FileUtils#touch): fix corrupted output when
+	  FileUtils.touch(:nocreate => true, :verbose => true) case.
+	  The patch was written by Hiroyuki Iwatsuki. [ruby-dev:43401]
+
+Thu Apr 14 16:01:45 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* io.c (rb_f_syscall): suppress warning: "HAVE___SYSCALL" is not
+	  defined.
+
+Thu Apr 14 00:41:09 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread.c (thread_fd_close_i): IOError exception should be assigned
+	  to rb_thread_t::thrown_errinfo.
+
+Wed Apr 13 20:12:26 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* io.c (rb_io_fdatasync): remove unused variable.
+
+Tue Apr 12 20:54:12 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/st.h: parenthesize macro arguments.
+
+Tue Apr 12 19:19:50 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb: avoid race condition. fixes #4572
+
+Tue Apr 12 18:07:13 2011  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/extconf.rb: --disable-libedit to disable
+	  libedit. fixes #4550
+
+Tue Apr 12 10:37:39 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h: VC doesn't have ftruncate() and others, but
+	  ruby needs HAVE_ macros to use our emulation functions.
+	  (fix the problem of 31262)
+
+Tue Apr 12 01:33:00 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* configure.in: properly evaluate existence of truncate, ftruncate
+	  and ftello for MinGW. [ruby-core:35678]
+	* win32/win32.c: rename truncate, ftruncate and ftello to avoid
+	  redefinitions.
+	* win32/win32.h: ditto.
+
+Mon Apr 11 21:51:52 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c: revert r31230. Because it made a regression.
+	  [ruby-core:35631]
+
+Mon Apr 11 21:49:18 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb: Added TestIO#test_cross_thread_close_stdio
+	  and TestIO#test_cross_thread_close_fd.
+	  The patch was written by Eric Wong. [ruby-core:35669]
+
+Mon Apr 11 21:15:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (rb_group_member): kill 256K of stack usage.
+	  the patch was written by Eric Wong. [ruby-core:35699]
+
+Mon Apr 11 07:24:13 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl.c:  Fix typo, document version constants.
+
+Sun Apr 10 22:23:45 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h: parenthesize macro arguments.
+
+Sat Apr  9 23:31:47 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* ext/stringio/stringio.c (strio_each, strio_readlines):
+	  Use `NUM2LONG` instead of `FIX2INT`. Fixes [ruby-dev:43395].
+
+Sat Apr  9 23:22:27 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* ext/stringio/stringio.c (strio_each):
+	  Fix exception message and don't raise immediately if block is not
+	  given.
+	  Fixes [ruby-dev:43394].
+
+	* test/stringio/test_stringio.rb (test_each_line_limit_0):
+	  Fix test for above.
+
+Sat Apr  9 21:54:15 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* ext/stringio/stringio.c (strio_each, strio_readlines):
+	  limit must not be zero. Fixes [ruby-dev:43392].
+
+	* test/stringio/test_stringio.rb: Add tests for above.
+
+Sat Apr  9 18:01:36 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/util.h: parenthesize macro arguments.
+
+Fri Apr  8 16:01:56 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_getline): check whether str is
+	  a string when str and lim are given.
+	  https://twitter.com/watson1978/status/56225052152168449
+
+Thu Apr  7 20:03:52 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h: parenthesize macro arguments.
+
+Wed Apr  6 21:08:31 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h: parenthesize macro arguments.
+
+Wed Apr  6 15:12:40 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey_dh.c (ossl_dh_initialize):
+	  pop pushed error after each try of reading. fixes #4550
+
+	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_initialize): ditto.
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_initialize): ditto.
+
+Wed Apr  6 11:36:44 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_initialize):
+	  pop pushed error after each try of reading. fixes #4550
+
+Tue Apr  5 20:33:43 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h: parenthesize macro arguments.
+
+Mon Apr  4 22:02:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/io/nonblock/nonblock.c (io_nonblock_set): Avoid F_SETFL if
+	  we're not changing the O_NONBLOCK bit. F_SETFL is an expensive
+	  operation since it needs to affect all processes with the same
+	  file object.
+	  The patch is written by Eric Wong. [ruby-core:35556]
+
+Mon Apr  4 21:41:26 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_syswrite): While local FS writes are usually
+	  buffered, the buffers can be full or the file opened with
+	  O_SYNC. IO#syswrite can also be used on blocking IOs
+	  (pipe/socket) just like IO#write.
+	  The patch is written by Eric Wong. [ruby-core:35554]
+
+Mon Apr  4 11:50:40 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/test_tempfile.rb: simply ignore platform dependent testcases
+	  instead of skipping.
+
+Sun Apr  3 22:52:22 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/syslog/syslog.c: improve rdoc.
+	  a patch by Jonas Pfenniger. [ruby-core:35592] fixes #4545
+
+Sun Apr  3 22:10:09 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/zlib/zlib.c: parenthesize macro arguments.
+
+Sun Apr  3 21:33:58 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: disable fdatasync again on Mac OS X.
+	  [ruby-core:35493][Bug #4500]
+
+Sun Apr  3 21:16:20 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_reopen): IO#close releases GVL if possible.
+	  close() may block for certain file types (NFS, SO_LINGER
+	  sockets, inotify), so let other threads run. The patch was
+	  created by Eric Wong [ruby-core:35555][Bug #4527]
+
+	* io.c (fptr_finalize): ditto.
+
+	* io.c (maygvl_fclose): new.
+	* io.c (nogvl_fclose): ditto.
+	* io.c (maygvl_close): ditto.
+	* io.c (nogvl_close): ditto.
+
+Fri Apr  1 22:25:50 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/syslog/syslog.c: parenthesize macro arguments.
+
+Fri Apr 1 18:53:06 2011  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/webrick/cookie.rb (WEBrick::Cookie.parse): 'white space is
+	  permitted between tokens' according to RFC2965. Though 'Netscape
+	  spec' does not define the syntax clearly, make it tolerant as a
+	  server. As a real-world example, rest-client gem sends
+	  'Cookie: foo=1;bar=2'
+
+	* test/webrick/test_cookie.rb (test_parse_non_whitespace): test it.
+
+Fri Apr  1 13:19:20 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (RUBY_VM_CHECK_INTS_TH): merge a patch by ko1
+	  in [ruby-dev:43373].
+
+Thu Mar 31 23:15:46 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-brace-to-do-end, ruby-do-end-to-brace):
+	  adjust space between block beginning and block arguments
+
+Thu Mar 31 20:42:05 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/strscan/strscan.c:  parenthesize macro arguments.
+
+Thu Mar 31 18:06:12 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_get_ev_const): should ignore crefs with
+	  the NODE_FL_CREF_PUSHED_BY_EVAL flag.
+
+Thu Mar 31 16:49:56 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_get_ev_const): search root cref properly.
+	  [ruby-dev:43365]
+
+Thu Mar 31 14:50:25 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_s_constants): should ignore crefs with
+	  the NODE_FL_CREF_PUSHED_BY_EVAL flag.
+
+Wed Mar 30 22:55:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-toggle-block): toggle do/end and {}.
+
+	* misc/ruby-mode.el (ruby-move-to-block): move to opening of
+	  block.
+
+Wed Mar 30 14:35:15 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.h (COPY_CREF): should copy
+	  the NODE_FL_CREF_PUSHED_BY_EVAL flag to hide constants from
+	  methods defined by class_eval.  [ruby-dev:43365]
+
+Wed Mar 30 00:24:53 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/stringio/stringio.c: parenthesize macro arguments.
+
+Tue Mar 29 21:51:31 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* object.c (rb_String): Kernel#String should call to_str before to_s.
+
+Tue Mar 29 10:28:08 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/webrick/test_filehandler.rb
+	  (WEBrick::TestFileHandler#test_short_filename): the cgi doesn't exist
+	  on current directory.
+
+Tue Mar 29 05:19:57 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c: parenthesize macro arguments.
+
+Tue Mar 29 00:03:51 2011 Tajima Akio <artonx@yahoo.co.jp>
+
+	* test/webrick/test_filehandler.rb (test_short_filename):
+	  read real short filename by cmd because smb mounted files
+	  have different naming convention.
+
+Mon Mar 28 11:38:08 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/date_core.c (date_s_today): use int for year.
+
+	* ext/date/date_core.c (datetime_s_now): ditto.
+
+Mon Mar 28 11:07:41 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/extmk.rb: set MFLAGS from MAKEFLAGS when using nmake.
+
+Mon Mar 28 11:07:00 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (love): all you need is love.
+
+Sun Mar 27 23:16:31 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c: parenthesize macro arguments.
+
+Sun Mar 27 16:55:34 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode-map): remove unnecessary
+	  binding. fixes
+	  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=468952
+
+	* misc/ruby-mode.el: suppress warnings at byte compile.  fixes
+	  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=502926
+
+Sun Mar 27 11:18:35 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: removed unused variables.
+
+Sat Mar 26 15:16:09 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/getaddrinfo.c: parenthesize macro arguments.
+
+Sat Mar 26 05:27:34 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/lib/date/format.rb (DateTime#strftime): removed because
+	  date_core defines it.
+
+Fri Mar 25 21:59:45 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: should not force cast with macros.
+
+Fri Mar 25 21:56:10 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/sdbm/init.c: parenthesize macro arguments.
+
+Fri Mar 25 19:39:40 2011  Ben Walton <bwalton@artsci.utoronto.ca>
+
+	* test/test_syslog.rb:
+	  Skip syslog tests that rely on LOG_PERROR unless it's defined
+
+	  Instead of checking looking at the platform to determine if the tests
+	  relying on LOG_PERROR should be run, look for the definition of the
+	  constant as this will be robust against all platforms as long as the
+	  underlying syslog.c code sets it up correctly.
+
+	  This specifically addresses failures on Solaris 9.
+
+	  Use LOG_PID instead of LOG_PERROR in Syslog.open test
+
+	  LOG_PERROR isn't a POSIX option for syslog, so it fails on platforms
+	  that don't define it.  Solaris 9 and 10 are examples of this.
+
+	  Use LOG_PID instead.
+
+Fri Mar 25 15:42:17 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/sdbm/_sdbm.c (sdbm_open): use size_t.
+
+	* ext/syck/bytecode.c: ditto.
+
+	* ext/sdbm/_sdbm.c (delpair): use ptrdiff_t.
+
+	* ext/sdbm/init.c: use RSTRING_LENINT.
+
+	* ext/dl/handle.c: suppress warning: shorten-64-to-32.
+
+	* ext/strscan/strscan.c: ditto.
+
+	* ext/syck/emitter.c: ditto.
+
+	* ext/syck/implicit.c: ditto.
+
+	* ext/syck/syck.c: ditto.
+
+	* ext/syck/token.c: ditto.
+
+Fri Mar 25 12:14:58 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: import nkf 7f18e30.
+
+Fri Mar 25 11:49:29 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_process.rb (TestProcess#test_no_curdir): skip silently
+	  on Windows, because this tests a platform specific feature and it'll
+	  never be supported on ruby on Windows.
+
+	* test/ruby/test_dir_m17n.rb
+	  (TestDir_M17N#test_filename_extutf8_invalid,
+	  TestDir_M17N#test_filename_as_bytes_extutf8): ditto.
+
+	* test/open-uri/test_open-uri.rb
+	  (TestOpenURI#test_find_proxy_case_sensitive_env): ditto.
+
+	* test/dl/test_handle.rb (DL::TestHandle#test_NEXT,
+	  DL::TestHandle#test_DEFAULT): ditto.
+
+Thu Mar 24 23:06:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_get_ev_const): should not autoload in
+	  defined? mode.
+
+	* variable.c (rb_const_defined_0): fix autoloading base.
+	  [ruby-core:35509]
+
+Thu Mar 24 22:48:43 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/sdbm/_sdbm.c: parenthesize macro arguments.
+
+Thu Mar 24 14:45:57 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl.c: suppress warning: shorten-64-to-32.
+
+	* ext/openssl/ossl.h: ditto.
+
+	* ext/openssl/ossl_asn1.c: ditto.
+
+	* ext/openssl/ossl_bio.c: ditto.
+
+	* ext/openssl/ossl_bn.c: ditto.
+
+	* ext/openssl/ossl_cipher.c: ditto.
+
+	* ext/openssl/ossl_hmac.c: ditto.
+
+	* ext/openssl/ossl_ns_spki.c: ditto.
+
+	* ext/openssl/ossl_ocsp.c: ditto.
+
+	* ext/openssl/ossl_pkcs5.c: ditto.
+
+	* ext/openssl/ossl_pkey.c: ditto.
+
+	* ext/openssl/ossl_pkey_dh.c: ditto.
+
+	* ext/openssl/ossl_pkey_dsa.c: ditto.
+
+	* ext/openssl/ossl_pkey_ec.c: ditto.
+
+	* ext/openssl/ossl_pkey_rsa.c: ditto.
+
+	* ext/openssl/ossl_rand.c: ditto.
+
+	* ext/openssl/ossl_ssl.c: ditto.
+
+	* ext/openssl/ossl_x509ext.c: ditto.
+
+	* ext/openssl/ossl_x509name.c: ditto.
+
+Thu Mar 24 11:48:19 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_rand.c (ossl_rand_egd_bytes): use NUM2INT because
+	  the result is used with functions whose argument is int.
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): ditto.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_set_purpose): ditto.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_set_trust): ditto.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509stctx_set_purpose): ditto.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509stctx_set_trust): ditto.
+
+Thu Mar 24 11:36:55 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_x509name.c: id_aref's type is ID.
+
+Thu Mar 24 10:04:35 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/io/console/console.c (console_set_winsize):
+	  suppress warning: shorten-64-to-32.
+
+Thu Mar 24 09:56:19 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_verify): flags is VALUE,
+	  so it should use NUM2INT.
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspbres_verify): ditto.
+
+Wed Mar 23 21:09:29 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/readline/readline.c: parenthesize macro arguments.
+
+Wed Mar 23 08:07:33 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_round): fix inaccurate results.
+
+Wed Mar 23 00:12:16 2011  Tajima Akio <artonx@yahoo.co.jp>
+
+	* win32/win32.c: wait process real termination after reading
+	  exit code. fixes #4518
+
+Tue Mar 22 21:20:10 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/test_case.rb: save current dir to @current_dir
+	  before Dir.chdir.
+
+Tue Mar 22 20:10:04 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/psych/parser.c: parenthesize macro arguments.
+
+Tue Mar 22 20:10:01 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ruby_missing.h: parenthesize macro arguments.
+
+Tue Mar 22 13:33:22 2011  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/buffering.rb: removed circular require of
+	  openssl.rb.
+
+	* ext/openssl/lib/openssl/*: removed following comment for transition
+	  measures of avoiding circular require. No one claimed about this as
+	  far as I know.
+	  ##
+	  # Should we care what if somebody require this file directly?
+	  # require "openssl"
+
+Tue Mar 22 10:57:36 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/runner.rb: set Gem::TestCase's @@project_dir.
+
+	* lib/rubygems/test_case.rb: set Gem::TestCase's @@project_dir only
+	  when it is not defined.
+
+Tue Mar 22 09:38:19 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (flo_round): use pow instead of while-loop. fixes #4510
+	  patched by Alex Young [ruby-core:35526]
+
+Tue Mar 22 06:47:46 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/date_strftime.c (date_strftime_wo_timespec):
+	  suppress warning: shorten-64-to-32.
+
+Tue Mar 22 06:42:42 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/date_core.c: suppress warning: shorten-64-to-32.
+
+Tue Mar 22 06:41:37 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/test/unit/parallel.rb: remove unused variable.
+
+Tue Mar 22 06:19:42 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/utf_16le.c: suppress warning: shorten-64-to-32.
+
+	* ext/dbm/dbm.c: ditto.
+
+	* ext/gdbm/gdbm.c: ditto.
+
+	* parse.y (Init_ripper): suppress warning: unused value.
+
+Mon Mar 21 11:21:32 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Refactoring. Unified if and elsif.
+
+Sun Mar 20 23:09:34 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: checks duplicated modifiers.
+
+Sun Mar 20 22:32:30 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: removed unused code and arguments.
+
+Sun Mar 20 21:34:49 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: replacement of implementation of
+	  strftime. It has some limitations that is same as Time's
+	  one.  [experimental]
+	* ext/date/date_strftime.c: new.
+	* ext/date/lib/date/format.rb: removed ruby version of strftime.
+
+Sun Mar 20 12:43:12 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509store.c: parenthesize macro arguments.
+
+Sun Mar 20 01:39:48 2011  Tajima Akio <artonx@yahoo.co.jp>
+
+	* hash.c (ruby_setenv): check env process block size with OS ver.
+	* win32/win32.c: export rb_w32_osver for above patch.
+	* include/ruby/win32.h: declare rb_w32_osver for Win32 Libs.
+
+Sat Mar 19 18:35:05 2011  Tajima Akio <artonx@yahoo.co.jp>
+
+	* hash.c (ruby_setenv): calculate total env block size for win32.
+	* test/ruby/test_env.rb: add test for above patch.
+
+Sat Mar 19 17:14:46 2011  Tajima Akio <artonx@yahoo.co.jp>
+
+	* hash.c (ruby_setenv): checking with max process environment
+	  block size for Win32. 32767 for 2000/XP, 2003. if failed to
+	  read the block, then checking with 5120 for earlier Windows.
+
+Sat Mar 19 12:30:25 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509revoked.c: parenthesize macro arguments.
+
+Fri Mar 18 20:44:36 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509req.c: parenthesize macro arguments.
+
+Fri Mar 18 08:48:06 2011  Oleg Shaldybin  <oleg.shaldybin@gmail.com>
+
+	* lib/fileutils.rb (FileUtils::Entry_#copy_file): updated FileUtils.cp
+	  to still copy file permissions when :preserve is false (as cp does
+	  this even when -p isn't set).
+
+Fri Mar 18 00:59:38 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/extconf.rb (create_docfile): removed.  should not
+	  modify source directory unnecessarily, platform dependent
+	  documentation should be dealt with by rdoc.  [ruby-core:35495]
+
+Fri Mar 18 00:54:20 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_funcall_passing_block): add prototype.
+	  a patch by James M. Lawrence at [ruby-core:35501]
+
+Wed Mar 17 06:23:31 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509name.c: parenthesize macro arguments.
+
+Wed Mar 16 20:36:56 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): raise an error on
+	  listen(2) failure.
+	  reported by Xavier Shay.  [ruby-core:35505]
+
+Wed Mar 16 15:06:21 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/lib/openssl/buffering.rb (module OpenSSL): #flush should
+	  not change sync mode on exception.
+	* test/openssl/test_buffering.rb: added
+
+Wed Mar 16 13:45:28 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/lib/openssl/buffering.rb: de-nest Buffering module
+
+	* ext/openssl/lib/openssl/buffering.rb: add RDoc
+
+Wed Mar 16 08:40:39 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509ext.c: parenthesize macro arguments.
+
+Tue Mar 15 18:34:27 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509crl.c: parenthesize macro arguments.
+
+Tue Mar 15 09:49:03 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/misc/test_ruby_mode.rb (test_singleton_class): Skip for Pending.
+
+Mon Mar 14 21:20:44 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/test_require.rb (test_require_too_long_filename):
+	  increase path length, because MAXPATHLEN is defined as 4096 on linux.
+
+	* test/ruby/test_require.rb (test_require_path_home_1): ditto.
+
+	* test/ruby/test_require.rb (test_require_path_home_2): ditto.
+
+Mon Mar 14 19:54:37 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509cert.c: parenthesize macro arguments.
+
+Sun Mar 13 18:11:28 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509attr.c: parenthesize macro arguments.
+
+Sun Mar 13 16:07:58 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/pstore.rb: Fix don't raise "nested transaction" when thread_safe
+	  is true. Patch by Masaki Matsushita (Glass_saga). [ruby-dev:43337]
+
+	* test/test_pstore.rb: Test for above.
+	  Patch by Masaki Matsushita (Glass_saga) [ruby-dev:43337]
+
+Sat Mar 12 04:12:41 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ssl_session.c: parenthesize macro arguments.
+
+Sat Mar 12 02:27:07 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c ({d,dt}_lite_marshal_load): checks the given
+	  argument.
+
+Sat Mar 12 01:26:24 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: changed some directives.
+
+Sat Mar 12 01:16:02 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c, ext/date/lib/*: moved rdoc descriptions.
+
+Sat Mar 12 00:06:24 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/lib: moved from lib.
+
+Fri Mar 11 23:32:38 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/delta*: removed undocumented delta.
+
+Fri Mar 11 18:42:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (find_executable0): should exclude directories.
+
+Fri Mar 11 01:40:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (proc_getmaxgroups, proc_setmaxgroups): Process#maxgroups
+	  and Process#maxgroups= now raise NotImplementedError if the
+	  platform don't support supplementary groups concept.
+
+Fri Mar 11 01:25:03 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (get_sc_ngroups_max): return -1 if platform don't
+	  support NGROUPS_MAX.
+
+Thu Mar 10 22:28:15 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ssl.h: parenthesize macro arguments.
+
+Thu Mar 10 21:59:37 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (parser_encode_length): add exception as UTF8-MAC for
+	  magic comment's emacs newline specifier
+	  patched by James M. Lawrence [ruby-core:35476] fixes #4489
+
+Thu Mar 10 16:00:22 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (parser_encode_length): fix typo: the length of
+	  "-dos" and "-mac" is not 5 but 4.
+	  patched by James M. Lawrence [ruby-core:35476] fixes #4489
+
+Thu Mar 10 10:52:01 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_require.rb: setting too long string to ENV causes
+	  Errno::EINVAL on Windows.  long path name errors may causes over
+	  about 1024 bytes, then limit it about 4000 bytes.
+
+Thu Mar 10 10:09:35 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner::Worker#read): fix for the case
+	  when IO#read or IO#gets returns nil.
+
+Thu Mar 10 07:12:03 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/rubygems*: Import rubygems 1.6.2 (release candidate @ 2026fbb5)
+	* test/rubygems: Ditto
+	* test/runner.rb: Added test to load path to fix test requires.
+
+Thu Mar 10 03:00:43 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ssl.c: parenthesize macro arguments.
+
+Wed Mar  9 23:51:26 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/test_io_m17n.rb (test_io_new_enc): "sjis" is now an alias
+	  of Windows-31J.
+
+Wed Mar  9 23:06:13 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-parse-partial): fix indent after aref.
+
+Wed Mar  9 12:50:24 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: Rescue exceptions when
+	  people implement the method method. Thanks Lin Jen-Shin.
+	  [ruby-core:35255]
+
+	* test/psych/visitors/test_yaml_tree.rb: test for implementation of
+	  method method.
+
+Wed Mar  9 11:53:31 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/shift_jis.c: Change SJIS as an alias of Windows-31J.
+	  [ruby-dev:43027] fixes #4280
+
+	* enc/shift_jis.c: Add PCK as an alias of Windows-31J.
+
+Wed Mar  9 00:45:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: nmake substitutes all occurrences in macro.
+
+	* ext/extmk.rb: workaround for nmake.
+
+Tue Mar  8 23:49:45 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (proc_setgroups): cleanup.
+
+Tue Mar  8 23:40:30 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/misc/test_ruby_mode.rb: test for ruby-mode.el.
+
+Tue Mar  8 23:27:38 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (get_sc_ngroups_max): try to use NGROUPS_MAX at first if
+	  _SC_NGROUP_MAX is not defined.
+
+Tue Mar  8 23:10:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-parse-partial): fix for array in block.
+
+Tue Mar  8 21:44:49 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_rand.c: parenthesize macro arguments.
+
+Tue Mar  8 16:45:31 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (ruby_setenv): MSDN says that Windows XP or earlier limits
+	  the total size of environment block to 5,120 chars.  and on such
+	  OS, putenv() causes SEGV.  So, ruby should limit the size of an
+	  environment variable to 5,120 bytes for workaround.
+
+Tue Mar  8 15:57:20 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_spec_fetcher.rb
+	  (TestGemSpecFetcher#test_cache_dir_escapes_windows_paths): cache_dir
+	  may have driveletter and `:' for base of cache_dir itself, so need
+	  to skip it for checking.
+
+Tue Mar  8 12:30:06 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-deep-indent-paren-p, ruby-calculate-indent):
+	  do not apply deep-indent inside parens at the beginning of
+	  expressions.
+
+Tue Mar  8 09:32:48 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (configure-ext, build-ext), ext/extmk.rb (extmake):
+	  support parallel-make under ext.
+
+Tue Mar  8 09:25:23 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_setgroups): use getgrnam() if getgrnam_r() is
+	  not available.
+
+	* process.c: RARRAY_LEN() returns long int.
+
+Tue Mar  8 09:07:03 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_REPLACE_TYPE): enclose in quotes for multiple
+	  type names.
+
+Tue Mar  8 01:43:11 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c (get_sc_ngroups_max): define to wrap sysconf(3).
+	  this also supports Windows which doesn't have sysconf(3).
+
+	* process.c (maxgroups): use get_sc_ngroups_max.
+
+	* process.c (proc_setmaxgroups): ditto.
+
+Tue Mar  8 01:16:49 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (rb_objspace): an initializer must be a constant.
+
+Tue Mar  8 01:11:44 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c (maxgroups): cast because sysconf(3)'s return value is long.
+
+	* process.c (proc_setmaxgroups): ditto.
+
+	* process.c (proc_setgroups): cast because RARRAY_LEN() is long.
+
+Tue Mar  8 00:02:47 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkey_rsa.c: parenthesize macro arguments.
+
+Mon Mar  7 22:59:39 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/pstore.rb: Delete variable @transaction and fix #4474. Patch by
+	  Masaki Matsushita (Glass_saga).
+
+	* test/test_pstore.rb(test_thread_safe): Add test for #4474.
+
+Mon Mar  7 21:31:38 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (proc_setgroups): replace getgrnam() with getgrnam_r()
+	  because getgrnam() isn't thread safe.
+
+Mon Mar  7 20:49:12 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (proc_getmaxgroups, proc_setmaxgroups): reflect
+	  platform maxgroups limitation by default instead hardcoded 65536.
+
+Mon Mar  7 17:13:00 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (rb_gc_set_params): allow GC parameter configuration by
+	  environment variables.  based on a patch from funny-falcon at
+	  https://gist.github.com/856296, but honors safe level.
+
+Mon Mar  7 09:05:18 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c: NUM2RLIM is defined but no getrlimit and setrlimit on
+	  mingw.
+
+Mon Mar  7 08:38:14 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_core.c (DateTimeData): should not use bare 'long long'
+	  and 'long double', which are not defined by C89.
+
+	* ext/date/date_core.c (dt_lite_plus): get rid of overflow at casting
+	  down double to integer.
+
+Mon Mar  7 00:21:11 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (proc_getgroups): get rid of maxgroups dependency.
+	  ngroups can be calculated dynamically.
+
+Sun Mar  6 23:45:40 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: rlim_t use standard RUBY_REPLACE_TYPE mechanism.
+
+Sun Mar  6 23:26:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (proc_setmaxgroups): added negative value check.
+	  This was suggested by Daniel Berger. Thanks Daniel!
+	  [ruby-core:35426][Bug#4467]
+
+Sun Mar  6 23:18:23 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (maxgroups, proc_setmaxgroups): increase max groups
+	  limitation up to 65536.
+
+Sun Mar  6 22:20:59 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkey_ec.c: parenthesize macro arguments.
+
+Sun Mar  6 21:49:04 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* sample/list.rb (MyElem#initialize): initialize @head
+	  explicitly. Otherwise -W2 option makes following warning.
+	  "warning: instance variable @head not initialized".
+	  This issue was founded by Andrew Grimm. Thanks Andrew!
+	  [ruby-core:35435][Bug#4471]
+
+Sun Mar  6 05:21:41 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* class.c: fix camelCase to snake_case in documentation code examples.
+	  patched by Andrew Grimm. fixes Bug #4469
+
+	* marshal.c: ditto.
+
+	* proc.c: ditto.
+
+	* sample/biorhythm.rb: ditto.
+
+	* vm_eval.c: ditto.
+
+	* vm_method.c: ditto.
+
+
+Sun Mar  6 03:22:27 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_cntl): use rb_thread_io_blocking_region() instead
+	  rb_thread_blocking_region().
+
+Sat Mar  5 22:54:36 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* include/ruby/intern.h: fix a typo of prototype declaration.
+	  rb_mutex_try_lock -> rb_mutex_trylock [ruby-dev:43213]
+
+Sat Mar  5 19:44:03 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (TestIO#test_fcntl_lock): small clean up.
+
+Sat Mar  5 01:33:46 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_cntl, nogvl_io_cntl): IO.fcntl() and IO.ioctl()
+	  release GVL during calling kernel interface.
+	  Suggested by Eric Wong. [ruby-core:35417][Bug #4463]
+
+	* test/ruby/test_io.rb (TestIO#test_fcntl_lock): add new test for
+	  IO.fcntl().
+
+Fri Mar  4 23:09:12 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/testunit/test_parallel.rb
+	  (test_should_run_all_without_any_leaks): consider that the order of
+	  testcase could change. [ruby-dev:43300] [Bug #4466]
+
+Fri Mar  4 22:01:14 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_cntl): change 'cmd' type to int. ioctl and fcntl need to
+	  be passed int.
+	* io.c (rb_io_ctl): ditto.
+
+Fri Mar  4 21:10:40 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: save warnflags. the patch is created by Eric Wong.
+	  [Bug #4465]
+
+Wed Mar  2 21:15:00 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkey_dsa.c: parenthesize macro arguments.
+
+Thu Mar  3 22:10:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (check_exec_redirect_fd, check_exec_redirect): raise
+	  ArgumentError if fd >= 3 on Windows because the feature is not
+	  supported.
+
+	* test/ruby/test_process.rb (test_execopts_redirect): remove meaningless
+	  argument.
+
+Thu Mar  3 21:21:42 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_process.rb (test_execopts_redirect): redirecting fd
+	  >= 3 is not supported on Windows, so should not specify such options
+	  when calling spawn or others.
+
+Thu Mar  3 18:59:04 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_slice_bang): raise error when the string is frozen.
+
+Thu Mar  3 14:25:19 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* strftime.c (STRFTIME): return 0 and ERANGE when precision is too
+	  large. [ruby-dev:43284] fixes #4456
+
+Thu Mar  3 00:46:51 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (uleb128): cast the value to unsigned long.
+
+	* addr2line.c (fill_lines): print error when lseek fails.
+
+Thu Mar  3 00:36:29 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rexml/encoding.rb (REXML::Encoding#encoding=): store @encoding
+	  a String which means the name of the encoding.
+	  this partially revert r29646.
+
+	* lib/rexml/document.rb: follow above.
+
+	* lib/rexml/output.rb: ditto.
+
+	* lib/rexml/parsers/baseparser.rb: ditto.
+
+	* lib/rexml/source.rb: ditto.
+
+	* lib/rexml/xmldecl.rb: ditto.
+
+Wed Mar  2 23:19:56 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_byte_substr): return nil for negative length.
+
+Wed Mar  2 21:15:00 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkey_dh.c: parenthesize macro arguments.
+
+Wed Mar  2 14:24:04 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/parallel.rb: Fix name from `inclement_io` to
+	  `increment_io`.
+
+Wed Mar  2 14:06:01 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_slice_bang): move treatments which is only needed
+	  when the result is not nil.
+
+Wed Mar  2 14:02:29 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/test_parallel.rb(TestParallel#spawn_runner):
+	  Fix outputing empty line in running test.
+
+	* test/testunit/tests_for_parallel/test_third.rb: Remove `sleep`
+
+Tue Mar  1 22:29:10 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkey.h: parenthesize macro arguments.
+
+Tue Mar  1 22:02:35 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/parallel.rb: Fix number.
+
+Tue Mar  1 21:48:22 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/parallel.rb: For Windows.
+
+	* test/testunit/test_parallel.rb(TestParallelWorker#test_quit_in_test):
+	  Fix for above specification change.
+	* test/testunit/test_parallel.rb(TestParallel#spawn_runner):
+	  Fix outputing empty line in running test.
+
+Tue Mar  1 20:51:57 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_system.rb (TestSystem#test_system_at):
+	  remove tests for [bug#4396]. because we decided to reject this
+	  ticket.
+
+Tue Mar  1 19:46:19 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/{test_date.rb,test_date_attr.rb}: [ruby-dev:43280]
+
+Tue Mar  1 18:40:38 2011  Ryan Davis  <ryan@YPCMC09457>
+
+	* lib/rubygems*: Import rubygems 1.6.0 (released version @ 58d8a0b9)
+	* test/rubygems: Ditto
+
+Tue Mar  1 16:22:22 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c: revert r30987 because it causes some failures in
+	  test-all, especially webrick.
+
+Tue Mar  1 15:59:53 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_byteslice): the resulted encoding should keep
+	  original encoding. this also fixes the encoding when the result
+	  shares internal string. [ruby-core:35376]
+
+Tue Mar  1 13:25:00 2011  Kenta Murata <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpMemAlloc): CVE-2011-0188.
+	  Fixes a bug reported by Drew Yao <ayao at apple.com>
+
+Tue Mar  1 10:34:39 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_byteslice): Add String#byteslice. [ruby-core:35376]
+
+Tue Mar  1 00:12:49 2011  Tajima Akio <artonx@yahoo.co.jp>
+
+	* include/ruby/win32.h: define WIN32 if neither _WIN64 nor WIN32
+	  defined. it forces to use push/pop for pack(4) pragma.
+
+Mon Feb 28 23:52:13 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/testunit/test_rake_integration.rb (test_with_rake_runner):
+	  use assert_in_out_err for suppress messages.
+
+Mon Feb 28 22:48:56 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/win32.c (rb_w32_spawn): use shell if a commandline contain
+	  double-quote character.
+	* win32/win32.c (is_internal_cmd): similar, use shell if a commandline
+	  contain caret character.
+
+	* test/ruby/test_system.rb (TestSystem#test_system_at): fix
+	  wrong test case. if system() invoke a command by using shell,
+	  system() never return nil. Also, "" quotation must not appear
+	  twice in a command line.
+
+Mon Feb 28 17:36:57 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkcs7.c: parenthesize macro arguments.
+
+Mon Feb 28 16:48:42 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkcs12.c: parenthesize macro arguments.
+
+Mon Feb 28 16:28:15 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_trans): when the hash for multibyte repl is empty,
+	  tr is inverse mode, and a character doesn't much the table, the
+	  character should be replaced by last replacement. Bug #4449
+
+Mon Feb 28 16:38:56 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ocsp.c: parenthesize macro arguments.
+
+Mon Feb 28 13:02:15 2011  Danial Pearce <github@tigris.id.au>
+
+	* lib/tempfile.rb: Fix example file paths in docs for tempfile.
+	  https://github.com/ruby/ruby/pull/5
+
+Mon Feb 28 12:56:18 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_init): typo fix.
+	  https://github.com/ruby/ruby/pull/8
+
+Mon Feb 28 12:28:13 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/date/date_core.c (datetime_s_now): localtime() and localtime_r()
+	  required time_t pointer as 1st parameter, and tv_sec member of struct
+	  timeval is long.
+
+Mon Feb 28 11:57:40 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/test_parallel.rb: Temporally disable test on Windows.
+
+Mon Feb 28 07:28:35 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb(Test::Unit::Runner#after_worker_quit):
+	  method name more be natural English.
+
+	* lib/test/unit.rb(Test::Unit::Runner::Worker.launch):
+	  IO.sync doesn't need. Should use "b" for mode.
+
+Sun Feb 27 21:59:37 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_system.rb (TestSystem#test_system_redirect_win):
+	  add test for system().
+
+Sun Feb 27 18:00:09 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Refactoring; Worker never use Hash for internal
+	  storage.
+
+	* lib/test/unit.rb: Never use Kernel#spawn. Use IO.popen instead.
+
+Sun Feb 27 13:16:48 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ns_spki.c: parenthesize macro arguments.
+
+Sat Feb 26 17:07:53 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: [Feature #4257]
+
+	* ext/date/extconf.rb: new
+
+	* ext/date/date_core.c: new
+
+Sat Feb 26 16:10:23 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: --jobs-status won't puts over 2 lines.
+
+	* test/testunit/test_parallel.rb: Fix test for above.
+
+	* lib/test/*: refactoring.
+
+Sat Feb 26 07:10:05 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: fix parsing timezone's whose
+	  whose format is (+/-)hhmm.  Thanks Goncalo Silva!
+
+	* test/psych/test_scalar_scanner.rb: test for bug.
+
+Thu Feb 24 23:02:55 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_hmac.c: parenthesize macro arguments.
+
+Thu Feb 24 22:53:10 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (love): for the birthday.
+
+Thu Feb 24 22:51:54 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (ruby_vm_destruct): run vm exit hooks after all objects are
+	  destructed.
+
+Thu Feb 24 14:40:33 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* ChangeLog (vim): Modeline for vim
+
+Thu Feb 24 13:39:25 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* common.mk: Use $RUNRUBY for worker process.
+
+	* lib/test/unit.rb: Fix bug.
+
+	* lib/test/unit.rb: @options[:ruby](@opts[:ruby]) is now Array.
+
+	* test/testunit/parallel.rb: Fix for above.
+
+Thu Feb 24 10:05:55 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/tests_for_parallel/misc.rb: Fix bug in r30947.
+
+	* lib/test/unit.rb, lib/test/unit/assertions.rb: For this test.
+
+Wed Feb 23 23:07:38 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/test_parallel.rb, test/testunit/parallel/*:
+	  Test for r30939.
+
+	* lib/test/unit.rb: For test.
+
+	* lib/test/parallel.rb: For test.
+
+	* lib/test/unit/testcase.rb: For test.
+
+Wed Feb 23 22:05:13 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_engine.c: parenthesize macro arguments.
+
+Tue Feb 22 23:15:17 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Fix --ruby option doesn't effect.
+
+	* lib/test/unit.rb: Fix typo.
+
+Tue Feb 22 21:39:28 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_digest.c: parenthesize macro arguments.
+
+Tue Feb 22 14:34:26 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Fix merging miss.
+
+Tue Feb 22 12:27:26 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Add new options; --jobs,-j,--ruby,--jobs-status,
+	  --no-retry.
+	  [Feature #4415] [ruby-dev:43226],[ruby-dev:43222],[ruby-core:35294]
+
+	* lib/test/unit/parallel.rb: Used at test/unit --jobs(-j) option.
+
+	* test/csv/test_serialization.rb: test/unit parallel running ready.
+
+	* test/rake/test_file_task.rb: test/unit parallel running ready.
+
+Tue Feb 22 06:09:10 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/syslog/syslog.c: Apply documentation patch from mathew murphy.
+	  [Bug #4149]
+
+Tue Feb 22 03:09:10 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: increase Psych to 1.1.0 for help with
+	  debugging.
+
+Tue Feb 22 03:04:46 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/streaming.rb: refactor streaming methods to a
+	  module.
+	* ext/psych/lib/psych/stream.rb: extracted streaming specific methods
+	  to a module.
+	* ext/psych/lib/psych/json/stream.rb: JSON stream inherits from
+	  JSONTree and includes streaming methods.
+	* ext/psych/lib/psych/visitors/json_tree.rb: JSON does not support
+	  object references, so remove object reference testing when building
+	  JSON trees.
+
+Tue Feb 22 02:41:51 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb (accept): use Hash#key?
+	  when looking up object references to err on the side of cache
+	  misses.
+
+Mon Feb 21 10:58:39 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/json/yaml_events.rb: refactoring JSON event
+	  handling methods to a module for reuse.
+	* ext/psych/lib/psych/json/tree_builder.rb: AST builder uses JSON
+	  event methods.
+	* ext/psych/lib/psych/json/stream.rb: stream emitter uses JSON event
+	  methods.
+
+Mon Feb 21 10:54:29 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/json/stream.rb: do not emit custom tags in maps
+	  or sequences when emitting JSON.
+	* ext/psych/lib/psych/json/tree_builder.rb: do not emit custom tags in
+	  sequences when emitting JSON.
+	* test/psych/json/test_stream.rb: tests for custom stream emits.
+	* test/psych/test_json_tree.rb: tests for JSON emits.
+
+Mon Feb 21 10:05:10 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/json/ruby_events.rb: DRY up ruby event handling
+	  for JSON.
+	* ext/psych/lib/psych/visitors/json_tree.rb: use ruby events module
+	* ext/psych/lib/psych/json/stream.rb: ditto
+
+Mon Feb 21 10:01:01 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/json/stream.rb: fix JSON stream emits to use
+	  double quotes during stream.
+	* test/psych/json/test_stream.rb: tests to reflect changes.
+
+Mon Feb 21 00:38:56 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_system.rb (TestSystem#test_system_at):
+	  add testcase for bug4396.
+
+Sun Feb 20 19:59:32 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_cipher.c: parenthesize macro arguments.
+
+Sun Feb 20 16:26:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (exec_recursive): prevent temporary objects from GC.
+
+	* prevent temporary objects from GC, and should not use
+	  RSTRING_PTR() for function calls since it evaluates the argument
+	  a couple of times.
+
+Sun Feb 20 16:22:53 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_flock): use rb_thread_io_blocking_region for the
+	  time being.
+
+Sun Feb 20 05:33:17 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 2.0.2 r6207.
+	* test/minitest/*: ditto
+
+Sun Feb 20 02:14:09 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (sig_trap): avoid pthread_sigmask(xx, &mask, &mask) usage
+	  because FreeBSD don't permit it. If it's used, it behave as
+	  pthread_sigmask(xx, NULL, &mask).
+
+	* signal.c (init_sigchld): ditto.
+
+Sun Feb 20 00:46:51 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_bn.c: parenthesize macro arguments.
+
+Sat Feb 19 22:37:42 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* vm_insnhelper.c (vm_check_if_namespace): guard temporary object
+	  from GC.
+
+Sat Feb 19 06:36:27 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/test/unit.rb: partial revert of r30849. [ruby-core:32864]
+
+	* test/testunit/test_rake_integration.rb: adding an integration test
+	  with the rake loader to prevent regressions.
+
+Fri Feb 18 19:31:31 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::remove_entry_secure): there is a
+	  race condition in the case where the given path is a directory,
+	  and some other user can move that directory, and create a
+	  symlink while this method is executing.
+	  Reported by: Nicholas Jefferson <nicholas at pythonic.com.au>
+
+Fri Feb 18 00:28:39 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* compile.c (get_exception_sym2type): guard temporary object from GC.
+
+Thu Feb 17 23:54:29 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* iseq.c (prepare_iseq_build): initialize iseq_compile_data::err_info
+	  with nil. this fix exception in rb_iseq_load().
+
+Thu Feb 17 22:32:35 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/test_marshal.rb (test_marshal_dump_extra_iv):
+	  fix a typo of local variable. [Bug #3720] [ruby-dev:42083]
+
+Thu Feb 17 21:32:53 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl.h: parenthesize macro arguments.
+
+Wed Feb 16 20:37:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* eval_jump.c (rb_exec_end_proc): changed at_exit and END proc
+	  evaluation order. [Bug #4400] [ruby-core:35237]
+	* eval_jump.c (rb_mark_end_proc): ditto.
+
+	* test/ruby/test_beginendblock.rb (TestBeginEndBlock#test_nested_at_exit):
+	  added a test for nested at_exit.
+	* test/ruby/test_beginendblock.rb (TestBeginEndBlock#test_beginendblock):
+	  changed the test to adopt new spec.
+
+Wed Feb 16 20:17:06 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/openssl_missing.h: parenthesize macro arguments.
+
+Tue Feb 15 21:37:45 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/gdbm/gdbm.c: parenthesize macro arguments.
+
+Tue Feb 15 20:34:53 2011  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (ary_join_1): fix array size.
+
+Tue Feb 15 19:43:23 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: fix and resubmit r30621. [ruby-dev:43203]
+
+Tue Feb 15 15:41:30 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (array_join): copy the encoding of the first element as
+	  an initial encoding.
+
+	* array.c (array_join_0): ditto.
+
+	* array.c (array_join_1): ditto.
+
+	* array.c (inspect_ary): ditto.
+
+	* array.c (array_join_1): add an argument to check the appending is
+	  first one or not.
+
+Tue Feb 15 15:40:53 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* hash.c (inspect_i): copy the encoding of the first key as
+	  an initial encoding.
+
+Mon Feb 14 15:00:16 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (inspect_ary): don't taint the inspected result of a
+	  recursive array.
+
+Tue Feb 15 15:43:29 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): change the rule for empty strings:
+	  remove the special treatment of the US-ASCII encoded empty string.
+	  Now Encoding.compatible? usually respect the encoding of the
+	  receiver.
+
+Tue Feb 15 15:39:37 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_buf_cat): remove special treatment of
+	  ASCII-8BIT receivers.
+
+	* string.c (str_gsub): set initial encoding of the buffer as the
+	  same of the receiver. [ruby-core:35141]
+
+Tue Feb 15 09:49:33 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_system.rb (TestSystem#test_system_at): use findstr
+	  command instead of find command, because the latter is confusing
+	  another famous Unix command.
+
+Mon Feb 14 23:01:19 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread.c (rb_thread_io_blocking_region): reset th->waiting_fd
+	  after blocking region, because remaining waiting_fd might
+	  cause unnecessary IOError.
+
+Mon Feb 14 21:06:50 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* configure.in: revert r30621.  That revision introduced mkmf test
+	  failures and it turned out to be OK to revert. [ruby-dev:43203]
+
+Mon Feb 14 21:04:01 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/fiddle/conversions.h: parenthesize macro arguments.
+
+Mon Feb 14 18:41:47 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/setup.mak (USE_RUBYGEMS): fixed r30835. It didn't work on
+	  mswin32 port. If you changed win32/configure.bat, you should change
+	  setup.mak too.
+
+Mon Feb 14 17:28:34 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_system.rb (TestSystem#test_system_at):
+	  added test. [ruby-core:35218] (#4393)
+
+Mon Feb 14 13:15:35 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (is_internal_cmd): if the first char of prog is '@',
+	  execute it via shell.  [ruby-core:35218] (#4393)
+
+Mon Feb 14 10:33:45 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb: revert r30863, because it causes too many noise.
+
+Mon Feb 14 07:34:55 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c: parenthesize macro arguments.
+
+Sun Feb 13 19:41:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::RequireFiles#non_options): skip
+	  test suites failed to load instead of mere messages.
+
+Sun Feb 13 09:56:44 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/openssl/test_config.rb (OpenSSL#test_freeze): fix error
+	  message assertion.
+
+	* test/io/nonblock/test_flush.rb (TestIONonblock#flush_test):
+	  return true to finish the test.
+
+	* test/syck/test_string.rb (Syck::TestString#test_non_binary_string):
+	  use assert_not instead of refute, unless required minitest
+	  explicitly.
+
+	* test/test_prime.rb (TestPrime::sieve.Integer): ditto.
+
+	* test/xmlrpc/webrick_testing.rb (WEBrick_Testing#start_server):
+	  catch IOError when server socket was closed.
+
+Sun Feb 13 07:39:51 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_inject): typo fixed. a patch from Gaku Ueda in
+	  [ruby-core:35216].
+
+Sun Feb 13 00:48:47 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (Date#===): [ruby-core:35127]
+
+Sun Feb 13 00:29:18 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Options#process_args): always
+	  return options.
+
+	* lib/test/unit.rb (Test::Unit::RequireFiles#non_options): return
+	  if any test case get loaded.
+
+	* lib/test/unit.rb (Test::Unit::AutoRunner#initialize): do not add
+	  default directory if it is nil.
+
+	* lib/test/unit.rb (Test::Unit::AutoRunner#process_args): return
+	  true if any test cases to run.
+
+Sat Feb 12 23:17:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (assert_include): add alias.
+
+Sat Feb 12 14:44:20 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_io_blocking_region): new function to run
+	  blocking region with GIL released, for fd.
+
+	* thread.c (rb_thread_fd_close): implement.  [ruby-core:35203]
+
+	* vm.c (th_init): rename from th_init2.
+
+Sat Feb 12 14:41:36 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::AutoRunner#initialize): use
+	  default_dir if no test case given.
+
+	* lib/test/unit.rb (Test::Unit::Runner): rename from Test::Unit::Mini.
+
+	* lib/test/unit.rb (Test::Unit::GlobOption#non_options): run tests
+	  under base directory if no argument given.
+
+Sat Feb 12 08:03:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_settracefunc.rb (TestSetTraceFunc): ensure to use
+	  method_added hook defined in Module.
+
+Sat Feb 12 01:04:02 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): enable rubygems if --gem option is given.
+
+	* ruby.c (process_options): load rubygems if it is disabled but
+	  --gem option is given.
+
+Fri Feb 11 23:27:50 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ruby.c (proc_options): add --gem=enabled as an alias of
+	  --enable=gems and --gem=disabled as an alias of --disable=gems.
+	  Gem named "enabled" or "disabled" has already been reserved
+	  legitimately for this purpose.
+
+Fri Feb 11 23:17:04 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/cfunc.c: parenthesize macro arguments.
+
+Fri Feb 11 21:41:53 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bin/testrb, test/runner.rb, lib/test/unit.rb: improve backward
+	  compatibility.
+
+Fri Feb 11 19:45:26 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): use rb_ary_free to free internal object.
+
+	* gc.h (RUBY_FREE_UNLESS_NULL): get rid of double free.
+	  [ruby-core:35192]
+
+Fri Feb 11 16:57:03 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_transcode.rb (test_from_cp50221): fix wrong
+	  assertion and move back.
+
+Fri Feb 11 14:33:18 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (assert_no_match): alias for
+	  backward compatibility.
+
+Fri Feb 11 12:06:48 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (add_gems, require_libraries, proc_options): add
+	  --require and --gem options.
+
+Fri Feb 11 12:03:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rubygems): add --disable-rubygems option.
+
+Fri Feb 11 11:39:03 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in (CROSS_COMPILING): get rid of NameError.
+
+Thu Feb 10 23:12:34 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/dl.h: parenthesize macro arguments.
+
+Wed Feb  9 23:11:27 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c: parenthesize macro arguments.
+
+Tue Feb  8 11:47:11 2011  Loren Sands-Ramshaw  <lorensr@gmail.com>
+
+	* array.c: documentation clarification in rotate, rotate!,
+	  index, and rindex. [ruby-core:35144]
+
+Wed Feb  9 09:45:43 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_encoding.rb: remove unnecessary (and wrong)
+	  platform-dependent hacks.
+
+Wed Feb  9 00:47:18 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/etc/etc.c: parenthesize macro arguments.
+
+Tue Feb  8 19:38:00 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-expr-beg): fix for invalid nest errors.
+
+Tue Feb  8 19:22:59 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in (AC_MSG_CHECKING): fixed typo. the patch is
+	  created by Benoit Daloze. Thanks a lot. [Bug #4384][ruby-core:35148]
+
+Tue Feb  8 16:04:03 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_s_sysopen): use NUM2MODET() instead NUM2UINT().
+
+Tue Feb  8 15:59:23 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (rb_run_exec_options_err): use MODET2NUM() instead
+	  LONG2NUM().
+
+Tue Feb  8 13:59:56 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: revert r30725. Now we have proper runtime fallback.
+	  Therefore, no need compile time disabling. (see r30762).
+
+Tue Feb  8 01:00:21 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* process.c (proc_setgroups): add GC guard to prevent intermediate
+	  variable from GC.
+
+Tue Feb  8 00:56:33 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-expr-beg, ruby-in-here-doc-p): tell
+	  singleton class definitions from here documents.
+
+	* misc/ruby-mode.el (ruby-expr-beg, ruby-parse-partial): keyword
+	  followed by colon is label.
+
+Mon Feb  7 22:56:16 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* lib/benchmark.rb (Benchmark#bmbm): use ensure clause instead of
+	  Object#tap to restore STDOUT.sync.
+
+Mon Feb  7 22:34:20 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/net/http.rb (Net::HTTP#connect): support SNI (Server Name
+	  Indication) for HTTPS.  [ruby-dev:43164]
+	  http://stackoverflow.com/questions/4685736/openssl-server-name-indication-support-in-ruby
+
+Mon Feb  7 16:05:32 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Upgrade to RDoc 3.5.3  Fixes [Bug #4376]
+
+Mon Feb  7 11:46:59 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (rdoc): add --encoding=UTF-8; ruby's rdoc must be UTF-8.
+
+Mon Feb  7 10:21:50 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_options.rb (TestRDocOptions#test_check_files):
+	  there is no easy way to create owner unreadable file on Windows.
+	  So, skip the test.
+
+Sun Feb  6 13:48:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/lib/json/common.rb (JSON::MissingUnicodeSupport.iconv):
+	  should not drop rest of the result.  use Iconv.conv instead.
+
+Sun Feb  6 12:46:02 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* string.c (gsub): Ensure result encoding is the same as input
+	  encoding.  [Bug #4340].
+
+Sun Feb  6 12:18:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (words, qwords): dispatch array events.  based on a
+	  patch from Michael Edgar.  [Bug #4365].
+
+Sun Feb  6 12:12:59 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fileutils/fileasserts.rb (FileAssertions): separate module.
+
+Sun Feb  6 11:29:23 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/dbm.c: parenthesize macro arguments.
+
+Sat Feb  5 22:01:59 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  revert r30796. r30797 and r30798 are an alternative fix.
+	  [ruby-dev:43174]
+
+Sat Feb  5 21:47:09 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (mlhs_basic): include mlhs_post for ripper.  a patch
+	  from Michael Edgar at [ruby-core:35078].
+
+Sat Feb  5 21:22:21 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (assert_block): move from
+	  test/fileutils/fileasserts.rb.
+
+	* test/fileutils/fileasserts.rb (assert_block): pass arguments
+	  as-is.  [ruby-dev:43174]
+
+Sat Feb  5 16:47:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  msg can be passed nil. [Bug #4371] [ruby-dev:43174]
+
+Sat Feb  5 15:18:25 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Upgrade to RDoc 3.5.2
+
+Sat Feb  5 12:05:27 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_node_init_copy): SyckNode is not
+	  copiable.  [ruby-core:35094]
+
+Sat Feb  5 11:48:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_alloc): leave data ptr
+	  NULL.
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_new, ossl_cipher_initialize):
+	  allocate internal structure.  [ruby-core:35094]
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_copy): ditto.
+
+Sat Feb  5 11:29:10 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/parser/parser.h (GET_PARSER): raise TypeError.
+
+	* ext/json/parser/parser.rl (cParser_initialize): ditto.
+
+	* ext/json/parser/parser.h (GET_PARSER): check if initialized.
+	  [ruby-core:35079]
+
+	* ext/json/parser/parser.rl (cParser_initialize): ditto.
+
+Sat Feb  5 10:09:31 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_get_expanded_load_path): always expand load paths.
+
+Sat Feb  5 09:38:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (encoded_dup): extract.
+
+Sat Feb  5 03:37:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::LowMethods): make low level methods
+	  in NoWrite and DryRun to do nothing.  [ruby-dev:43129]
+
+	* test/fileutils/fileasserts.rb: add message arguments.
+
+	* test/fileutils/fileasserts.rb (Test::Unit::Assertions#assert_block):
+	  show the given message.
+
+Sat Feb  5 02:09:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lex_getline, parser_set_encode): set encoding of lines
+	  in SCRIPT_LINES__ as source encoding.  [ruby-dev:43168]
+
+Sat Feb  5 02:08:37 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (ruby_thread_data_type): add prefix.
+
+Sat Feb  5 00:59:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (GetThreadPtr): use TypedData_Get_Struct() instead
+	  CoreDataFromValue() because we need type check. Otherwise,
+	  type mismatch can cause segmentation fault crash.
+	  [ruby-core:35086] [Ruby 1.9-Bug#4367]
+
+	* vm.c (thread_data_type): remove static.
+
+Fri Feb  4 19:14:27 2011  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/utf8_mac.trans: parenthesize macro arguments.
+
+Fri Feb  4 12:11:51 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* string.c (str_utf8_nth): fixed a condition of optimized lead
+	  byte counting. [Bug #4366][ruby-dev:43170]
+
+Fri Feb  4 01:50:13 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* string.c (count_utf8_lead_bytes_with_word): wrote function
+	  comments.
+
+Fri Feb  4 00:14:55 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_reader_get_unused): no need to dup
+	  before rb_str_resurrect.
+
+Thu Feb  3 20:04:44 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c (CHECK): unused macro removed.
+
+Thu Feb  3 18:33:26 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_reader_get_unused): use rb_str_resurrect
+	  because gz->z.input is hidden string. [ruby-core:35057]
+
+Thu Feb  3 16:34:10 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/shift_jis.c (code_to_mbc): cast as int from the subtraction of
+	  pointers.
+
+	* enc/utf_16le.c (utf16le_mbc_enc_len): use ptrdiff_t.
+
+	* enc/utf_32be.c (utf32be_left_adjust_char_head): ditto.
+
+	* enc/utf_32le.c (utf32le_left_adjust_char_head): ditto.
+
+Thu Feb  3 16:31:43 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/missing.h: don't use HAVE_STDDEF_H because it never
+	  defined by configure though configure.bat defines it.
+
+	* include/ruby/ruby.h: move include stddef.h to defines.h
+
+	* include/ruby/defines.h: ditto.
+
+Wed Feb  2 20:25:04 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_enc_step_back): cast 4th argument 'n'
+	  as int because Ruby usually treats length value as long but
+	  onigenc_step_back's 4th argument is int.
+
+Thu Feb  3 07:20:46 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: use Regexp::NOENCODING
+	  rather than magic number.
+
+	* ext/syck/lib/syck/rubytypes.rb: ditto
+
+Thu Feb  3 07:16:11 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* re.c (Init_Regexp): added a constant for ARG_ENCODING_NONE
+	  [ruby-core:35054]
+
+	* test/ruby/test_regexp.rb: corresponding test.
+
+Thu Feb  3 07:02:16 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: ARG_ENCODING_NONE regular
+	  expressions can round trip. [ruby-core:34969]
+
+	* test/psych/test_yaml.rb: test for ARG_ENCODING_NONE regex
+
+	* ext/sych/lib/syck/rubytypes.rb: ARG_ENCODING_NONE regular
+	  expressions can round trip.
+
+	* test/syck/test_yaml.rb: test for ARG_ENCODING_NONE regex
+
+Wed Feb  2 17:09:22 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_fdatasync): Use fsync(2) if the underlying
+	  operating system does not support fdatasync(2).
+
+Wed Feb  2 14:51:08 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup/to_tt_only.rb: commit miss
+	* test/rdoc/test_rdoc_markup_to_tt_only.rb: ditto
+	* test/rdoc/test_rdoc_single_class.rb: ditto
+
+Wed Feb  2 09:27:53 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Upgrade to RDoc 3.5.1
+
+Wed Feb  2 00:30:43 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/st.h (st_table): Added comment why we need __extension__.
+
+Tue Feb  1 20:45:44 2011  Tanaka Akira  <akr@fsij.org>
+
+	* enc/encdb.c: parenthesize macro arguments.
+
+Tue Feb  1 15:12:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_require.rb (TestRequire#test_require_with_unc):
+	  use ``127.0.0.1'' instead of ``localhost'' as host name, because
+	  XP or earlier cannot resolv it as NBT hostname.
+
+Tue Feb  1 13:20:39 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/benchmark/test_benchmark.rb (#capture_bench_output):
+	  Added explicit sleep. Windows have imprecise time support.
+	  Thus Tms.new.Add!{} may be or may be not equal 0. The
+	  test failure started since r30747.
+
+Tue Feb  1 11:03:47 2011  Ryan Davis  <ryan@lust.local>
+
+	* lib/rubygems*: Import rubygems 1.5.0 (released version @ 1fb59d0)
+	* test/rubygems: Ditto
+
+Tue Feb  1 08:01:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_set_winsize): new method to set
+	  console size.  [EXPERIMENTAL]
+
+	* ext/io/console/console.c (console_winsize): use GetWriteFD.
+
+Tue Feb  1 02:28:14 2011  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* include/ruby/win32.h, win32/win32.c: add rb_w32_inet_ntop.
+	  inet_ntop's minimum supported client is Vista.
+
+Tue Feb  1 00:10:30 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/benchmark.rb: fix benchmark to work with current ruby.
+	  patched by Benoit Daloze [ruby-core:33846] [ruby-dev:43143]
+	  merged from https://github.com/eregon/ruby/commits/benchmark
+
+	*  lib/benchmark (Report#width): update documentation
+	*  lib/benchmark: document the return value of #benchmark and the
+	   :list attribute in Report
+	*  lib/benchmark (Tms#format): rename variables, use String#%
+	   instead of Kernel.format
+	*  lib/benchmark: remove undocumented Benchmark::times (an alias
+	   of Process::times used twice)
+	*  lib/benchmark (#benchmark): use label_width for the caption
+	*  lib/benchmark (Tms#initialize): rename variables
+	*  lib/benchmark: allow title to not be a String and call #to_s
+	*  lib/benchmark (Benchmark#bm): return an Array of the times with
+	   the labels
+	*  lib/benchmark: correct output for Benchmark#bmbm
+	   (remove the extra space)
+	*  lib/benchmark: add a few tests for Benchmark::Tms output
+	*  lib/benchmark: improve style (enumerators, ljust, unused vars)
+	*  lib/benchmark: add spec about output and return value
+	*  lib/benchmark: improve basic style and consistency
+	   no parenthesis for print and use interpolation instead of printf
+	*  lib/benchmark: remove unnecessary conversions and variables
+	*  lib/benchmark: correct indentation
+	*  lib/benchmark: rename the FMTSTR constant and variable to FORMAT
+	*  lib/benchmark: remove useless exception
+
+	*  test/benchmark: remove unused variable warnings
+
+Mon Jan 31 23:27:23 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* node.c (add_id): remove duplicated rb_id2str() call.
+
+Sun Jan 30 17:19:46 2011  Tanaka Akira  <akr@fsij.org>
+
+	* missing/langinfo.c: parenthesize macro arguments.
+
+Mon Jan 31 21:57:39 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: revert r30698.
+
+Mon Jan 31 21:32:44 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread.c (thread_start_func_2): check deadlock condition before
+	  release thread stack. fix memory violation when deadlock detected.
+	  reported by Max Aller. [Bug #4009] [ruby-core:32982]
+
+Mon Jan 31 14:45:47 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/locale.rb (IRB::Locale::#search_file):
+	  Gem might be undefined if --disable-gems. [ruby-core:34990]
+
+Mon Jan 31 12:26:14 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: suppressed shorten-64-to-32 warnings.
+	* regcomp.c: ditto.
+	* regexec.c: ditto.
+	* regint.h: ditto.
+	* regparse.c: ditto.
+	* regparse.h: ditto.
+	* time.c: ditto.
+	* variable.c: ditto.
+
+Mon Jan 31 04:45:12 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_uniq_bang): call ARY_SET_LEN(ary, 0) before
+	  ary_resize_capa because ary_resize_capa expects resized length is
+	  smaller than current array length. call rb_ary_unshare before
+	  ary_resize_capa because ary_resize_capa lost the reference to
+	  original shared array. [ruby-core:34997]
+
+Sun Jan 30 17:19:46 2011  Tanaka Akira  <akr@fsij.org>
+
+	* missing/crypt.c: parenthesize macro arguments.
+
+Sun Jan 30 16:40:27 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_security.rb (TestGemSecurity): valid only
+	  if OpenSSL is available.
+
+	* test/dl/test_dl2.rb (TestDL#test_sin): math functions do not
+	  work on x86_64 due to the design of DL2.
+
+	* test/dl/test_func.rb (DL::TestFunc#test_{sinf,sin): ditto.
+
+Sun Jan 30 16:09:22 2011  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c (rb_strftime_with_timespec): %G produces 4 digits.
+
+Sun Jan 30 15:13:19 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/emacs_mule.c (emacsmule_islead): 7bit range is also leading
+	  byte.
+
+Sun Jan 30 13:03:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_fetch_m): use useful message for longer key, not a
+	  nonsense id value.
+
+	* string.c (rb_str_ellipsize): new function to ellipsize a string.
+
+	* include/ruby/encoding.h (rb_enc_step_back): new function to step
+	  back n characters.
+
+Sun Jan 30 12:53:38 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/emacs_mule.c (emacsmule_islead): fix inverse condition.
+
+Sun Jan 30 09:37:25 2011  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* io.c (struct argf): char behaves like an unsigned char
+	  by default on AIX.
+
+Sun Jan 30 08:02:55 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: Mac OS X wrongly reports it has fdatasync(3).
+
+Sun Jan 30 03:29:47 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_bn.c (GetBNPtr): add missing nil case.
+	  patched by Martin Bosslet. [ruby-core:34987]
+
+Sun Jan 30 01:02:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/ruby.h: Added NUM2MODET() and MODET2NUM() default
+	  definition.
+	  Because r30686 introduced win32 build failure.
+
+Sat Jan 29 22:16:26 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_join): [].join.encoding must be US-ASCII.
+	  [ruby-list:47790]
+
+Sat Jan 29 20:22:39 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/driver.rb (BenchmarkDriver#measure): Show command line
+	  when abnormal exiting occur.
+
+Sat Jan 29 10:53:16 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_insnhelper.c (vm_get_ev_const): no-scope reference to toplevel
+	  private constant has been prohibited incorrectly.
+
+	* test/ruby/test_module.rb (test_toplevel_private_constant): add a
+	  test for above.
+
+Sat Jan 29 08:43:23 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/rubygems*: Import rubygems 1.5.0 (release candidate @ 09893d9)
+	* test/rubygems: Ditto
+
+Sat Jan 29 02:02:37 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* variable.c (rb_mod_const_of, sv_i): Module#constant should exclude
+	  private constants.  see [ruby-core:32912].
+
+	* test/ruby/test_module.rb (test_constants_with_private_constant): add
+	  a test for above.
+
+Sat Jan 29 01:36:41 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* variable.c (rb_const_set): const_set should preserve constant
+	  visibility.  see [ruby-core:32912].
+
+	* test/ruby/test_module.rb: add a test for above.
+
+Sat Jan 29 01:24:57 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (NODE_CLASS, NODE_MODULE), insns.def (defineclass): raise
+	  an exception when "class Foo::Bar" is evaluated and Foo::Bar is
+	  private.  To implement this, define_type of "defineclass" is added
+	  so that the instruction can distinguish whether the class definition
+	  is scoped (class Foo::Bar) or not (class Bar).
+
+	* test/ruby/test_class.rb (test_redefine_private_class),
+	  test/ruby/test_module.rb
+	  (test_define_module_under_private_constant): add tests for above.
+
+Sat Jan 29 01:19:17 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* constant.h, variable.c: to ensure compatibility, rb_const_get_* must
+	  not raise an exception even when the constant is private.  Instead,
+	  rb_public_const_get_* and rb_public_const_defined_* are introduced,
+	  which raise an exception when the referring constant is private.
+	  see [ruby-core:32912].
+
+	* vm_insnhelper.c (vm_get_ev_const): use rb_public_const_get_* instead
+	  of rb_const_get_* to follow the constant visibility when user code
+	  refers a constant.
+
+	* test/ruby/test_marshal.rb (test_marshal_private_class): add a test.
+	  This test had failed because of incompatibility of rb_const_get.
+
+Sat Jan 29 00:30:44 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* variable.c (set_const_visibility): fix typo.  a patch from Tomoyuki
+	  Chikanaga in [ruby-core:32919].
+
+Fri Jan 28 23:20:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/gdbm/test_gdbm.rb (TestGDBM#test_s_open_no_create,
+	  TestGDBM2#test_writer_open_notexist): We only need to skip libgdbm
+	  1.8.0, not all 1.8.x. 1.8.1 or later don't have GDBM_WRITER sickness.
+
+Fri Jan 28 21:56:57 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/dbm/extconf.rb: Added new header places for Fedora13.
+
+Fri Jan 28 21:49:30 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/zlib/zlib.c: parenthesize macro arguments.
+
+Fri Jan 28 17:47:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/gdbm/test_gdbm.rb (TestGDBM2#test_writer_open_notexist):
+	  gdbm 1.8.x changed GDBM::WRITER behavior. Thus our testcase need
+	  to be changed too.
+
+Fri Jan 28 17:33:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/gdbm/test_gdbm.rb (TestGDBM#test_s_open_no_create): skip
+	  the test if gdbm version is 1.8.x.
+
+Fri Jan 28 16:30:51 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_require.rb (TestRequire#test_require_too_long_filename):
+	  Added -w option because too long path error don't output a message
+	  by default since r30660. [Bug #4336] [ruby-dev:43134]
+
+Fri Jan 28 16:19:38 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_require.rb (TestRequire#test_require_path_home_{1,2}):
+	  Added -w option because too long path error don't output a message
+	  by default since r30660. [Bug #4336] [ruby-dev:43134]
+
+Fri Jan 28 16:04:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_require.rb (TestRequire#test_require_path_home_{1,2,3}):
+	  split from test_require_path_home.
+
+Fri Jan 28 13:04:38 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in (--with-valgrind): Fixed r29683. Now this option
+	  is really default on.
+
+Fri Jan 28 12:05:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: Add #include<sys/stat.h> when struct stat is
+	  tested. Otherwise, incomplete type dereference error will occur.
+
+Fri Jan 28 11:53:19 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: redundant variable names made strange conftest
+	  error. Fixed it.
+
+Fri Jan 28 11:47:00 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_process.rb (TestProcess#test_too_long_path{,2}):
+	  should handle Errno::E2BIG, because this test checks crash of ruby,
+	  not the error type system.
+
+Fri Jan 28 11:23:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_open): Use NUM2MODET() instead NUM2UINT().
+	* io.c (rb_scan_open_args): ditto.
+
+Fri Jan 28 10:58:20 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: Added mode_t type checking.
+	* process.c (rb_exec_arg_addopt): Use NUM2MODET() instead
+	  NUM2LONG because clang makes compile error by this narrowing
+	  conversion.
+	* process.c (rb_run_exec_options_err): ditto.
+
+Fri Jan 28 02:37:18 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* re.c (rb_reg_raise): add GC guard to prevent intermediate
+	  variable from GC.
+
+Fri Jan 28 02:35:41 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* hash.c (rb_hash_fetch_m): add GC guard to prevent intermediate
+	  variable from GC.
+
+Fri Jan 28 01:33:51 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_process.rb (TestProcess#test_too_long_path)
+	  TestProcess#test_too_long_path): Reduced string size from 100MB
+	  to 10MB. 100MB may cause no memory error. It isn't intended.
+
+Fri Jan 28 01:27:42 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_process.rb (TestProcess#test_too_long_path2):
+	  Factored out from test_too_long_path. A test should only do
+	  one test.
+
+Thu Jan 27 23:29:36 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_foreach): check if unpacked.
+
+Thu Jan 27 23:14:19 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode-map): remove deprecated binding.
+	  use M-; instead.
+
+Thu Jan 27 21:58:32 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bignum.c (rb_str_to_inum): get rid of too huge alloca().
+
+Thu Jan 27 21:43:29 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* object.c (rb_str_to_dbl): rewrite again. use ALLOCV instead
+	  rb_str_tmp_new().
+
+Thu Jan 27 21:41:47 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/win32.c: get rid of STRNDUPA(). It's dangerous API.
+
+Thu Jan 27 21:31:57 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/win32.c (rb_w32_aspawn): get rid of too huge alloca().
+	  [Bug #4330] [ruby-core:34898]
+
+Thu Jan 27 20:30:18 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/win32.c (rb_w32_spawn): get rid of too huge alloca().
+
+Thu Jan 27 18:49:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/win32.c (open_dir_handle): get rid of too huge alloca().
+
+Thu Jan 27 18:34:58 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (w32_io_info): get rid of too huge alloca().
+	  [Bug #4313] [ruby-core:34830]
+
+Thu Jan 27 18:19:51 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/win32.c (wstati64): get rid of too huge alloca().
+	  [Bug #4316] [ruby-core:34834]
+
+Thu Jan 27 15:11:52 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn): get rid of too huge
+	  alloca(). this is the real fix of [ruby-core:34833].
+
+Thu Jan 27 12:46:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (ALLOC_ARGV_WITH_STR): fix void pointer arithmetic.
+
+Thu Jan 27 08:41:40 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_exec_v, rb_proc_exec_n, rb_proc_exec)
+	  (proc_spawn_n, proc_spawn): get rid of too huge alloca().
+	  [ruby-core:34827], [ruby-core:34833]
+
+Thu Jan 27 08:32:17 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ALLOCV): new API for exception-safe
+	  temporary buffer.  [ruby-core:34844]
+
+	* string.c (rb_alloc_tmp_buffer, rb_free_tmp_buffer):
+	  implementation of the API.
+
+Thu Jan 27 08:22:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln_find.c (dln_find_1): use rb_warning and return immediately
+	  if fname is longer than buffer.
+
+Wed Jan 26 22:57:30 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* class.c (clone_method): add GC guard to prevent intermediate
+	  variable from GC. [Bug #4321] [ruby-dev:43107]
+
+Wed Jan 26 22:45:16 2011  Tanaka Akira  <akr@fsij.org>
+
+	* template/id.h.tmpl: parenthesize macro arguments.
+
+Wed Jan 26 22:28:49 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* vm_eval.c (rb_throw_obj): add GC guard to prevent intermediate
+	  variable from GC. [Bug #4322] [ruby-dev:43108]
+
+Wed Jan 26 17:08:59 2011  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_decode0): OpenSSL::ASN1.decode
+	  should reject indefinite length primitive encodings as that is
+	  illegal. Patch by Martin Bosslet. See #4324.
+
+Wed Jan 26 10:36:28 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (=~): documentation fix; the return value is nil when
+	  it doesn't match. patched by Andrei Kulakov [ruby-core:34562]
+
+Tue Jan 25 08:41:58 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln_find.c (dln_find_1): omit too long pathnames.
+
+Tue Jan 25 08:28:19 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_resize): get rid of out-of-bound access.
+
+Tue Jan 25 07:48:22 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_thread.rb: remove unused variables.
+
+Tue Jan 25 07:45:44 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_thread.rb (TestThread#test_condvar_nolock_2): get
+	  rid of method redefined.
+
+Tue Jan 25 07:00:52 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_string_value_cstr): rb_str_modify can change
+	  RSTRING_PTR.
+
+Tue Jan 25 03:24:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_thread.rb: Added various ConditionVariable tests.
+
+Mon Jan 24 22:26:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* object.c (rb_str_to_dbl): Fix again. use rb_str_tmp_new()
+	  instead ALLOC_N.
+
+Mon Jan 24 21:50:48 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vm_insnhelper.h: parenthesize macro arguments.
+
+Mon Jan 24 21:28:34 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* object.c (rb_str_to_dbl): use ALLOC_N instead ALLOCA_N because
+	  ALLOC_N may cause stack overflow.
+
+Mon Jan 24 21:04:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_invalid_str): prevent intermediate variable from GC.
+	  [ruby-core:34820]
+
+Sun Jan 23 23:01:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/io/console/test_io_console.rb: Don't run test if the system
+	  don't support io/console.
+
+Sun Jan 23 22:17:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/fiddle/test_fiddle.rb: Don't run test if the system don't support
+	  fiddle.
+
+	* test/fiddle/test_function.rb: ditto.
+	* test/fiddle/test_closure.rb: ditto.
+
+Sun Jan 23 11:39:18 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vm_exec.h: parenthesize macro arguments.
+
+Sun Jan 23 10:33:02 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in (ruby): suppress warnings.
+
+Sun Jan 23 08:00:09 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_nth_len, str_utf8_nth): return the rest length together.
+
+	* string.c (rb_str_substr): get rid of measure the length always
+	  to improve performance for huge string.  [ruby-core:34648]
+
+Sun Jan 23 00:40:10 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/test_syslog.rb: Fix to make a lot of test failure if
+	  the platform doesn't support syslog.
+
+Sat Jan 22 11:49:55 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: fixing merge key support
+	  when multiple merge keys are specified.
+
+	* test/psych/test_merge_keys.rb: tests for multi-merge key support
+
+Sat Jan 22 11:33:04 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: merge keys are actually
+	  part of YAML 1.1, so they should be supported.  Remove warning and
+	  merge keys to parent.  [ruby-core:34679]
+
+	* test/psych/test_merge_keys.rb: test for merge keys
+
+Sat Jan 22 10:25:19 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (parse): add the file name to the exception when
+	  parse errors occur.
+
+	* test/psych/test_parser.rb: test for parse error file name
+
+Sat Jan 22 10:12:30 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (parse): fix assertion error when reusing a
+	  parser after an exception has been raised
+
+	* test/psych/test_parser.rb: test for assertion error
+
+Sat Jan 22 04:09:22 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/nodes/node.rb: Make Psych::Nodes::Node
+	  enumerable.
+
+	* ext/psych/lib/psych/visitors/depth_first.rb: Add a depth-first
+	  visitor to enumerate over a YAML AST in a depth-first fashion
+
+	* test/psych/nodes/test_enumerable.rb: test for enumerating nodes
+
+	* test/psych/visitors/test_depth_first.rb: test for depth-first
+	  visitor
+
+Sat Jan 22 00:53:42 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vm_core.h: parenthesize macro arguments.
+
+Fri Jan 21 18:15:09 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: should not use -Werror=* flags while conftests.
+
+Fri Jan 21 09:17:00 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* configure.in: Fix incorrectly detected x86_64-w64-mingw32 due
+	  canonalization of target_os. Bug #3889 [ruby-core:32634]
+
+Thu Jan 20 23:44:00 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: Fix rb_cv_va_args_macro was broken. We are using
+	  -Werror=implicit-function-declaration compile option. therefore
+	  we need a function declaration explicitly.
+
+Thu Jan 20 23:58:02 2011  Tanaka Akira  <akr@fsij.org>
+
+	* node.h: parenthesize macro arguments.
+
+Thu Jan 20 23:25:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: Add '#include <stdlib.h>' to
+	  rb_cv_localtime_overflow test too. It's reported by Tomoyuki
+	  Chikanaga. Thanks.
+
+Thu Jan 20 16:11:00 2011  Kenta Murata <mrkn@mrkn.jp>
+
+	* README.EXT, README.EXT.ja: You shouldn't choose ``conftest.c'' as a
+	  name of a source file.
+
+Thu Jan 20 12:15:44 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: Add stdlib.h inclusion into rb_cv_negative_time_t
+	  test because it's required for exit(3). The patch is
+	  created by Tomoyuki Chikanaga. [Bug #4287] [ruby-dev:43060]
+
+Thu Jan 20 11:39:41 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/webrick/utils.rb (TestWEBrick::RubyBin): test CGI does not need
+	  to load rubygems.  if it activated, ruby raises LoadError about
+	  rbconfig.rb.
+
+Thu Jan 20 09:19:42 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/json_tree.rb: Fix JSON emit for
+	  DateTime and Time classes.
+
+	* test/psych/test_json_tree.rb: test for JSON emit
+
+Thu Jan 20 08:02:46 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/coder.rb (represent_object): arbitrary objects
+	  may be passed to the Psych::Coder object.
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: support for visiting
+	  arbitrary objects set on the coder.
+
+	* test/psych/test_coder.rb: supporting test case.
+
+Thu Jan 20 06:03:17 2011  Tanaka Akira  <akr@fsij.org>
+
+	* method.h: parenthesize macro arguments.
+
+Wed Jan 19 13:16:05 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/sources_command.rb: Finish removing code,
+	  (fixes sources command test).
+
+Wed Jan 19 13:04:16 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* proc.c (proc_call): Add gc guard to avoid segfault. The fix
+	  is created by Tomoyuki Chikanaga. [Bug #4238][ruby-dev:42963]
+
+Wed Jan 19 12:31:28 2011  Eric Hodel <drbrain@segment7.net>
+
+	* lib/rubygems.rb: Since gem_prelude requires rubygems, enable
+	  custom_require always.
+
+Wed Jan 19 12:08:08 2011  Eric Hodel <drbrain@segment7.net>
+
+	* lib/rubygems/commands/dependency_command.rb: Remove require of
+	  deleted file.
+	* lib/rubygems/commands/fetch_command.rb: ditto
+	* lib/rubygems/commands/setup_command.rb: ditto
+	* lib/rubygems/commands/sources_command.rb: ditto
+	* lib/rubygems/commands/specification_command.rb: ditto
+
+Wed Jan 19 08:13:59 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/rubygems*: Import rubygems 1.5.0 (release candidate)
+	* test/rubygems: Ditto
+
+Tue Jan 18 23:31:01 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y: avoid NULL reference.  [ruby-dev:43067]
+
+Wed Jan 19 02:54:04 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vsnprintf.c (cvt): set first byte of buf to NUL for the case when
+	  no bytes are written to the buf. [ruby-dev:43062]
+
+Tue Jan 18 23:04:51 2011  Tanaka Akira  <akr@fsij.org>
+
+	* gc.h: parenthesize macro arguments.
+
+Tue Jan 18 18:31:14 2011  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/completion.rb: Irb tab completion support for XX::method
+	  forms.
+
+Tue Jan 18 15:05:55 2011  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: added RDoc document for logging message escape
+	  by Hal Brodigan. See #3869
+
+Tue Jan 18 07:53:52 2011  Tanaka Akira  <akr@fsij.org>
+
+	* eval_intern.h: parenthesize macro arguments.
+
+Tue Jan 18 04:42:44 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/parser.rb (Mark): Adding a class to wrap
+	  marker information
+
+	* ext/psych/parser.c (mark): Add a method to return the mark object
+	  for the parser
+
+	* test/psych/test_parser.rb: tests for the Mark class.
+
+Tue Jan 18 02:46:55 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/json_tree.rb (visit_String): JSON
+	  strings should be dumped with double quotes. [ruby-core:34186]
+
+	* test/psych/test_json_tree.rb: test for double quotes
+
+Mon Jan 17 23:36:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_times): less MEMCPY calls.
+
+Mon Jan 17 22:54:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* debug.h: parenthesize macro arguments.
+
+Mon Jan 17 21:40:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): revert r30549.
+
+Sun Jan 16 20:55:45 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vsnprintf.c: parenthesize macro arguments.
+
+Sat Jan 15 11:57:30 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (warnflags): add -Werror=implicit-function-declaration
+	  if available.
+
+	* lib/mkmf.rb (init_mkmf): ignore warnings in mkmf tests.
+
+	* test/mkmf/base.rb (setup, teardown): restore config values.
+
+	* test/mkmf/test_flags.rb: split from test_find_executable.rb.
+
+Sat Jan 15 10:04:14 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): autoload rubygems.
+
+	* tool/compile_prelude.rb (Prelude#initialize): ignore empty
+	  preludes.
+
+	* ruby.c (ruby_init_prelude): get rid of global namespace
+	  pollution.
+
+Sat Jan 15 09:42:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h: missing prototypes.
+
+Fri Jan 14 23:25:55 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vm_method.c: parenthesize macro arguments.
+
+Fri Jan 14 15:32:29 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/net/imap/test_imap.rb: call neither logout nor disconnect
+	  unless connected.  patch by Kazuhiro NISHIYAMA.  [ruby-dev:42860]
+
+Fri Jan 14 14:56:57 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/imap.rb: use bytesize for binary strings.
+	  patched by Yoshimasa Niwa. [ruby-core:34222]
+
+Fri Jan 14 14:01:12 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_unpack): the resulted string of unpack('M') must have
+	  ASCII-8BIT encoding (and ENC_CODERANGE_VALID). [ruby-core:34482]
+
+Fri Jan 14 13:38:58 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_check_footer): ISIZE (Input SIZE) in
+	  gzip's header is the size of uncompressed input data modulo 2^32.
+	  [ruby-core:34481] http://www.ietf.org/rfc/rfc1952.txt
+
+Fri Jan 14 11:36:25 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in, win32/Makefile.sub (RUNRUBY): require path should
+	  include "." because rbconfig.rb is there.
+
+Fri Jan 14 10:40:11 2011  Ryan Davis  <ryan@lust.local>
+
+	* gem_prelude.rb: Just require rubygems. Fixes rubygems 1.4.
+	* lib/rubygems.rb: removed all Gem::Quickloader code.
+	* ruby.c: renamed ruby_init_gems to ruby_init_prelude. Set
+	  $disable_rubygems since there is no fine grained mechanism to
+	  skip parts of the prelude. Open to suggestions on how to do this
+	  better.
+	* test/*.rb: Load path isn't set up correctly, so add
+	  --disable-gems as needed to failing tests that are explicitly
+	  testing stderr w/ ==.
+
+Fri Jan 14 07:30:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_next_argv): go advance when the next file cannot be
+	  read.  [ruby-core:34446]
+
+Thu Jan 13 20:49:19 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vm_insnhelper.c: parenthesize macro arguments.
+
+Thu Jan 13 13:21:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* vm_dump.c: delete dashes to make lines 80 chars, Patched by
+	  Shota Fukumori (sora_h).  [Bug #4275] [ruby-dev:43021]
+
+Thu Jan 13 13:21:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* vm_dump.c: fix misspelling of CrashReporter, Patched by Shota
+	  Fukumori (sora_h).  [Bug #4275] [ruby-dev:43021]
+
+Thu Jan 13 06:27:29 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* error.c: Exception#to_s should actually call to_s.
+
+Thu Jan 13 00:32:54 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (get_nth_dirname): decrement the directory index
+	  because the index specifies the index of given included_directories
+	  which is separated by NUL and its index is begun from 1.
+	  Note that 0 specifies the current directory of the compilation.
+	  see also http://www.dwarfstd.org/doc/dwarf-2.0.0.pdf
+
+Thu Jan 13 00:06:38 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_f_syscall): Add warning messages. [ruby-core:34062]
+
+Thu Jan 13 00:00:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_f_syscall): Some syscall return unsigned or pointer value.
+	  Therefore we should only check the result is -1 or not.
+	  [ruby-core:34062]
+
+Wed Jan 12 23:55:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_f_syscall): Add 64bit Linux support. Some syscall takes
+	  long type arguments.
+
+Wed Jan 12 19:37:10 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vm_dump.c: parenthesize macro arguments.
+
+Wed Jan 12 19:28:23 2011  Tanaka Akira  <akr@fsij.org>
+
+	* vm.c (thread_free): reset ruby_current_thread if it points the
+	  thread to free.
+	* gc.c (slot_sweep): don't call RUBY_VM_SET_FINALIZER_INTERRUPT if
+	  there is no current thread.
+	  [ruby-dev:43000]
+
+Wed Jan 12 19:09:29 2011  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (sort_by_i): reenter check more strictly.
+	  (sort_by_cmp): ditto.
+	  [ruby-dev:43003] reported by Usaku NAKAMURA.
+
+Wed Jan 12 16:25:12 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/net/http.rb (Net::HTTP#connect): makes it timeout during
+	  SSL handshake too. [ruby-core:34203]
+	  Patch by Marc Slemko.
+
+	* test/net/http/test_http.rb (TestNetHTTP_v1_2#test_timeout_during_HTTP_session):
+	  test for [ruby-core:34203]
+
+	* test/net/http/test_https.rb (TestNetHTTPS#test_timeout_during_SSL_handshake):
+	  ditto.
+
+Wed Jan 12 16:24:53 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ext/readline/extconf.rb: new checks for RL_PROMPT_START_IGNORE
+	  and RL_PROMPT_END_IGNORE. [ruby-core:34331]
+
+	* ext/readline/readline.c: enables USE_INSERT_IGNORE_ESCAPE only if
+	  RL_PROMPT_{START,END}_IGNORE are available to get rid of compilation
+	  error with libedit.
+
+Wed Jan 12 15:53:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* README.EXT.ja (rb_ensure): typo.
+
+Wed Jan 12 11:33:46 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: OpenBSD uses the elf_abi.h header file instead of the
+	  elf.h header file. patched by Jeremy Evans [ruby-core:34384]
+
+Wed Jan 12 03:59:36 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/webrick/test_cgi.rb: Removes usage of deprecated
+	  :RequestHandler option.
+	  patched by Peter Weldon [ruby-core:34010]
+
+	* test/webrick/test_httpproxy.rb: ditto.
+
+	* test/webrick/test_httpserver.rb: Add a test of the deprecation
+	  behaviour.
+
+Wed Jan 12 08:37:07 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (hash_i): return different values for inverse hash.
+	  [ruby-core:34334]
+
+Tue Jan 11 20:32:59 2011  Tanaka Akira  <akr@fsij.org>
+
+	* variable.c: parenthesize macro arguments.
+
+Tue Jan 11 13:06:38 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* array.c (rb_ary_resize): should care of embeded array when extending
+	  the array.
+
+	* array.c (rb_ary_resize): need to set capa when changing the real
+	  size of the array.
+	  these are latent bugs.
+
+Mon Jan 10 22:46:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/defines.h (CASEFOLD_FILESYSTEM): HFS+ is case
+	  insensitive.
+
+	* load.c (loaded_feature_path, rb_feature_p, load_lock): on a
+	  case-insensitive filesystem, loaded features search should
+	  ignore case.  [ruby-core:34297]
+
+Mon Jan 10 21:34:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (showflags): show LD commands.
+
+Mon Jan 10 14:32:55 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_method.rb (TestMethod#test_define_method): method
+	  transplanting between class and module is impossible.
+
+Mon Jan 10 13:51:17 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/rdoc-mode.el (rdoc-mode): show trailing whitespace.
+
+Mon Jan 10 11:22:02 2011  Tanaka Akira  <akr@fsij.org>
+
+	* util.c: parenthesize macro arguments.
+
+Mon Jan 10 07:41:31 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* misc/README: mention rdoc-mode.el and ruby-style.el.
+
+Sun Jan  9 20:37:21 2011  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c: parenthesize macro arguments.
+
+Sun Jan  9 16:31:53 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* io.c (Kernel.#syscall): implemented on LP64/LLP64 environments too.
+	  also uses __syscall if available for *BSD on 64bit architecture.
+	  [ruby-core:34062]
+
+Sun Jan  9 16:31:34 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/locale.rb (IRB::Locale::LOCALE_NAME_RE):
+	  some platform has a locale without territory but with
+	  encoding.
+	  (#each_sub_locale): ditto.
+
+Sun Jan  9 14:47:50 2011  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c: apply a patch from Nobuyoshi Nakada.
+	  fixed #3616 [ruby-core:31484] IRB + readline incorrectly counts
+	  non-printing characters in prompt
+
+Sat Jan  8 21:47:26 2011  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_sort_by): use rb_ary_resize.
+	  (ary_cutoff): removed.
+
+Sat Jan  8 21:24:17 2011  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (swapf): compilation condition simplified.
+	  (swapd): ditto.
+
+Sat Jan  8 20:51:25 2011  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (swapd): remove duplicated code.
+
+Sat Jan  8 19:28:55 2011  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c: parenthesize macro arguments.
+
+Fri Jan  7 23:07:40 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): backref needs to capture.
+
+Fri Jan  7 21:57:22 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode-variables), misc/ruby-style.el:
+	  show trailing whitespace.
+
+	* misc/ruby-mode.el (ruby-font-lock-syntactic-keywords): highlight
+	  regexp after open bracket.  [ruby-core:34183]
+
+Fri Jan  7 00:37:35 2011  Tanaka Akira  <akr@fsij.org>
+
+	* string.c: parenthesize macro arguments.
+
+Thu Jan  6 22:42:02 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigmul1_karatsuba): avoid overflow that make assertion
+	  fail in certain case.  this patch is contributed from Ray Chason
+	  <chasonr at gmail.com> in personal communication.
+
+Thu Jan  6 20:55:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): ignore rest from first dot from
+	  TARGET to generate init function name.
+	  this is followup of r30464.
+
+Thu Jan  6 11:27:01 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/json/tree_builder.rb (start_mapping): tags
+	  should not be included in JSON mapping
+
+Thu Jan  6 09:23:33 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/net/protocol.rb (eof?): BufferedIO should proxy eof? to the
+	  underlying IO object.
+
+Thu Jan  6 09:12:31 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/mkmf.rb (configuration): fixing gsub when multiple error flags
+	  are passed to GCC.
+
+Thu Jan  6 05:25:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_modify): export.
+
+Thu Jan  6 05:14:41 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (get_strio, strio_set_string)
+	  (strio_reopen): check if frozen.  [ruby-core:33648]
+
+Thu Jan  6 05:10:58 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_resize): new utility function.  [ruby-dev:42912]
+
+Thu Jan  6 05:03:26 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (init_funcname_len): ignore rest from first dot.
+	  [ruby-dev:41774]
+
+Thu Jan  6 02:55:48 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: use YAML 1.0 output
+	  format for serializing nil values.  Thanks Eric Hodel!
+
+	* test/psych/test_nil.rb: test for nil values
+
+Wed Jan  5 14:21:34 2011  Mark Dodwell  <hi@mkdynamic.co.uk>
+
+	* string.c: fix rdoc typo.
+	  https://github.com/shyouhei/ruby/pull/3
+
+Wed Jan  5 14:06:01 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_options.rb (TestRDocOptions#test_check_files):
+	  skip on Windows because chmod 0 doesn't mean unreadable by owner.
+
+Wed Jan  5 13:56:54 2011  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/net/http.rb (Net::HTTP#get): A header hash given should not
+	  be modified.
+
+Wed Jan  5 12:10:08 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/dl/{cfunc.c,dl.h,handle.c}, ext/fiddle/fiddle.{h,c}: Use _WIN32
+	  rather than checking for windows.h.  Thanks Jon Forums!
+	  [ruby-core:33977]
+
+Sat Jan  1 17:02:50 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/locale.rb (IRB::Locale#search_file): make it possible
+	  to load a localization from a gem.
+	  (IRB::Locale#lc_path): obsoleted because of the change of #search_file
+	  (IRB::Locale#each_localized_path): new private method, based on
+	  lc_path
+	  (IRB::Locale#find): follows the change of #search_file.
+	  (IRB::Locale#load): removed duplicate with #find.
+
+Sat Jan  1 11:44:42 2011  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c: parenthesize macro arguments.
+
+Sat Jan  1 11:10:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c: take care of platforms where long is bigger
+	  than int.
+
+Sat Jan  1 11:03:58 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* NEWS (optparse): shell completion support.
+
+	* misc/README (rb_optparse.{bash,zsh}): for shell completion.
+
+	* include/ruby/intern.h (VALUE rb_ary_print_on): I have never seen
+	  this function anywhere.
+
+Sat Jan  1 04:20:11 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write_console): don't raise exception when
+	  the conversion is for writing to console.
+	  Patched by Heesob Park [ruby-core:33999]
+
+Fri Dec 31 12:02:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_sort_by): use less temporary objects.
+
+Fri Dec 31 11:46:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (warnflags), lib/mkmf.rb (configuration): turn
+	  warnings into errors only for bundled extensions.
+	  [ruby-core:33815]
+
+Fri Dec 31 11:15:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (sizeof): zlib.h mistakenly assumes the result
+	  of sizeof to be int, not size_t.
+
+Fri Dec 31 10:27:34 2010  Tanaka Akira  <akr@fsij.org>
+
+	* st.c: parenthesize macro arguments.
+
+Fri Dec 31 03:23:26 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vsnprintf.c (BSD__uqtoa): Fix overflow when long != quad_t.
+	  patched by Peter Weldon <peter.weldon AT null.net>
+	  [ruby-core:33985]
+
+Fri Dec 31 03:00:34 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* Makefile.in: remove unnecessary semicolons.
+
+Thu Dec 30 23:09:47 2010  wanabe  <s.wanabe@gmail.com>
+
+	* vm.c (vm_define_method): guard iseq from GC while method definition.
+	  [ruby-dev:42832]
+
+Thu Dec 30 20:18:32 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/Makefile.sub: ditto.
+
+Thu Dec 30 20:57:09 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* Makefile.in: Check V=1 argument if run "make clean" or similar.
+
+Thu Dec 30 20:41:50 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* Makefile.in: Kill ugly line continuation.
+
+Thu Dec 30 11:49:40 2010  Tanaka Akira  <akr@fsij.org>
+
+	* sprintf.c: parenthesize macro arguments.
+
+Wed Dec 29 21:20:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (maygvl_copy_stream_wait_readwrite): define if USE_SENDFILE
+
+Wed Dec 29 20:37:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: strip current directory prefix.
+
+	* enc/depend (clean): remove name2ctype.h when out-of-place build.
+
+	* win32/Makefile.sub (clean-enc): pass V to inferior make.
+
+Wed Dec 29 18:23:46 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* re.c (rb_reg_expr_str): need to escape if the coderange is invalid.
+
+Wed Dec 29 10:06:51 2010  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c: parenthesize macro arguments.
+
+Wed Dec 29 07:22:15 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake/rdoctask.rb: Deprecate in favor of rdoc/task.
+
+Wed Dec 29 07:07:06 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Import RDoc 3.1
+
+Tue Dec 28 18:36:38 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* error.c, include/ruby/intern.h (rb_compile_error_with_enc): new
+	  function to raise syntax error, with source encoding'ed message.
+
+	* parse.y (compile_error): use above function.
+	  [ruby-core:33951] (#4217)
+
+Tue Dec 28 07:37:38 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c: parenthesize macro arguments.
+
+Tue Dec 28 07:17:11 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add ARGF.write and so on.
+
+Tue Dec 28 07:12:38 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add new magic-comment. (warn-indent) [ruby-core:25442]
+
+Tue Dec 28 04:32:37 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/extconf.rb: check for windows.h while building fiddle.
+	  Thanks Jon Forums! [ruby-core:33923]
+
+Tue Dec 28 01:45:12 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: Add Zlib.deflate and Zlib.inflate.
+	  [ruby-dev:42833]
+
+Mon Dec 27 21:22:33 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/configure.bat: Remove obsoleted coding rule. Now, we
+	  don't support to build on Windows 95/98 and Me.
+
+Mon Dec 27 18:27:13 2010  Tanaka Akira  <akr@fsij.org>
+
+	* re.c: parenthesize macro arguments.
+
+Mon Dec 27 15:22:23 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/README.win32: note to need NT based OS to build ruby.
+
+Mon Dec 27 12:14:46 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (EXTMK_ARGS): specify to pass macro V, because nmake
+	  doesn't pass it via MAKEFLAGS.
+
+Mon Dec 27 10:33:43 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/zlib/zlib.c (Init_zlib): Add Zlib.deflate and Zlib.inflate.
+	  [ruby-dev:42833]
+
+Mon Dec 27 07:38:07 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* misc/rb_optparse.zsh: add compdef for generator.
+
+Mon Dec 27 07:32:07 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/optparse.rb (OptionParser#compsys): escape brackets too.
+	  [ruby-dev:42754]
+
+Mon Dec 27 01:30:08 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: add IF_NAMESIZE.
+	  add a default for INET6_ADDRSTRLEN.
+
+Sun Dec 26 23:49:47 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* win32/Makefile.sub: suppress a strange error message when RMALL
+	  found no such file.
+	* win32/rmall.bat: new.
+
+Sun Dec 26 21:23:23 2010    <kosaki.motohiro@gmail.com>
+
+	* win32/Makefile.sub: fix 'nmake clean-enc' breakage since r28322.
+
+Sun Dec 26 22:25:07 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/ripper/depend (ripper.y): fix messages with nmake.
+	  [ruby-dev:42896]
+
+Sun Dec 26 22:24:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): get rid of warnings caused by
+	  -Wdeclaration-after-statement on cygwin.
+
+Sun Dec 26 20:28:34 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (before_exec): add small comment.
+
+Sun Dec 26 20:52:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: define INET_ADDRSTRLEN as 16 if not
+	  available.  fix compilation error on mswin32-60.  reported by nobu.
+
+Sun Dec 26 19:37:37 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c: define IFNAMSIZ if not available.
+	  fix compilation error on mingw32.  reported by nobu.
+
+Sun Dec 26 12:16:29 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/paths.rb (RDoc::RI::Paths::HOMEDIR): no exception if
+	  HOME is not set.  [ruby-core:33867]
+
+Sun Dec 26 11:39:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt): missing ripper rule.  i.e., `a::B ||= c 1'.
+	  http://twitter.com/#!/wannabe53/status/18797576396472321
+	  http://twitter.com/#!/wannabe53/status/18798416150663168
+
+Sun Dec 26 11:15:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/with_different_ofs.rb (DifferentOFS): should not affect
+	  original classes.
+
+Sun Dec 26 09:35:07 2010  Tanaka Akira  <akr@fsij.org>
+
+	* rational.c: parenthesize macro arguments.
+
+Sun Dec 26 09:22:19 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (rb_if_indextoname): new function to abstract
+	  environments without if_indextoname.
+	  (inspect_ipv6_multicast_if): new function to inspect
+	  IPV6_MULTICAST_IF.
+	  Socket::Option.new(:INET6, :IPV6, :MULTICAST_IF,
+	  [2].pack("I!")).inspect is
+	  "#<Socket::Option: INET6 IPV6 MULTICAST_IF eth0>".
+
+Sun Dec 26 04:31:15 2010  Luis Lavena  <luislavena@gmail.com>
+
+	* ext/dl/win32/registry.rb: Corrected RegCreateKeyExA signature.
+	  Patch by Rafal Michalski [ruby-core:33874] [Ruby 1.9-Bug#4203]
+
+Sun Dec 26 02:31:58 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (advice_arg_check): Change argument check.
+	  Now, an unsupported advice makes NotImplementedError.
+	  [ruby-dev:42887] [Ruby 1.9-Feature#4204]
+
+Sun Dec 26 03:00:53 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/socket/extconf.rb: Fix build error which was introduced r30372.
+
+Sun Dec 26 01:37:10 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: check the existence of if_indextoname().
+
+	* ext/socket/option.c: yesterday's akr's commits destroyed the build of
+	  some unrelated platforms (such as Windows).
+
+Sat Dec 25 23:29:11 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_ipv4_add_drop_membership): new function
+	  to inspect struct ip_mreq and struct ip_mreqn for
+	  IP_ADD_MEMBERSHIP and IP_DROP_MEMBERSHIP.
+	  Socket::Option.new(:INET, :IP, :ADD_MEMBERSHIP,
+	  [239,255,99,81, 0,0,0,0].pack("CCCCCCCC")).inspect is now
+	  "#<Socket::Option: INET IP ADD_MEMBERSHIP 239.255.99.81 0.0.0.0>".
+	  (inspect_ipv4_multicast_if): new function to inspect struct in_addr
+	  and struct ip_mreqn for IP_MULTICAST_IF.
+	  Socket::Option.new(:INET, :IP, :MULTICAST_IF,
+	  [192,168,0,7].pack("CCCC")).inspect is now
+	  "#<Socket::Option: INET IP MULTICAST_IF 192.168.0.7>".
+
+	* ext/socket/extconf.rb: check struct ip_mreq and struct ip_mreqn.
+
+Sat Dec 25 22:49:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/csv: DifferentOFS needs to be include in each classes.
+
+	* test/digest/test_digest_extend.rb (TestDigestExtend#setup):
+	  should not depend on the result of previous tests
+
+	* test/with_different_ofs.rb (DifferentOFS::WithDifferentOFS): give
+	  name.
+
+	* test/with_different_ofs.rb (DifferentOFS): test suite for test
+	  suites affected by $,.
+
+	* test/digest/test_digest_extend.rb (TestDigestExtend): should not
+	  assume $, invariant.
+
+	* test/csv/test_data_converters.rb, test/csv/test_table.rb: don't
+	  call setup within tests.
+
+Sat Dec 25 20:01:40 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (pipe_open): Added rb_thread_atfork(). We must reinitialize
+	  GVL at new process creation.
+
+Sat Dec 25 18:26:55 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_ipv6_mreq): new function to inspect
+	  struct ipv6_mreq for IPV6_JOIN_GROUP and IPV6_LEAVE_GROUP.
+	  Socket::Option.new(:INET6, :IPV6, :JOIN_GROUP,
+	  [0xff12,0,0,0,0,0,0,1, 2].pack("nnnnnnnnI!")).inspect is now
+	  "#<Socket::Option: INET6 IPV6 JOIN_GROUP ff12::1 eth0>".
+
+	* ext/socket/extconf.rb: check struct ipv6_mreq.
+
+Sat Dec 25 18:04:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/csv.rb (CSV.foreach): 'rb' mode is defaulted in open.
+
+	* lib/csv.rb (CSV#init_separators): cannonicalize encoding options
+	  as Encoding objects.
+
+Sat Dec 25 18:30:34 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_atfork): Add small comment why we need
+	  reset random seed.
+
+Sat Dec 25 17:33:55 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/csv/base.rb (TestCSV.with_different_ofs): give name to
+	  anonymous classes.
+
+	* lib/csv.rb (CSV#init_separators): use IO#gets with length
+	  parameter to get rid of wrong conversion.
+
+	* lib/csv.rb (CSV::foreach, CSV#initialize): directly use encoding
+
+	* lib/csv.rb, test/csv: should not assume $, invariant.
+
+Sat Dec 25 16:08:06 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c: change rb_atomic_t definition from uchar to uint.
+
+Sat Dec 25 15:04:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/csv/test_encodings.rb (TestEncodings#setup): fix evil test
+	  suite writing to the source directory.
+
+Sat Dec 25 15:08:08 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/pty/pty.c (chfunc): Added rb_thread_atfork_before_exec().
+	  We must reinitialize GVL at new process creation. Otherwise
+	  we may meet an insane deadlock. [Bug#4121][ruby-dev:42686]
+
+Sat Dec 25 14:27:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_extract_encoding_option): accept Encoding object as
+	  encoding: optional argument.  [ruby-dev:42884]
+
+Sat Dec 25 13:37:55 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 2.0.2 r6093.
+
+Sat Dec 25 13:05:59 2010  Tanaka Akira  <akr@fsij.org>
+
+	* random.c: parenthesize macro arguments.
+
+Sat Dec 25 12:48:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_f_require_relative): don't omit return type.
+
+Sat Dec 25 11:06:00 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* load.c (rb_f_require_relative): Add documentation.
+
+Sat Dec 25 11:02:52 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzreader_gets): support optional length
+	  parameter.
+
+	* ext/zlib/zlib.c (gzfile_read, gzfile_readpartial): length should
+	  be long.
+
+Sat Dec 25 10:51:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/generator/generator.{c,h} (fbuffer_free_only_buffer):
+	  unused.
+
+	* ext/openssl/ossl_pkcs5.c (ossl_pkcs5_pbkdf2_hmac): add casts.
+
+Fri Dec 24 08:46:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: parenthesize macro arguments.
+
+Thu Dec 23 19:17:14 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/net/imap/cacert.pem: updated because it has been expired.
+
+	* test/net/imap/server.crt: signed again because CA cert was expired.
+
+Thu Dec 23 11:16:52 2010  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y: parenthesize macro arguments.
+
+Thu Dec 23 11:00:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_check_type): check for type from extensions for ruby
+	  1.8.  see [ruby-core:33797].
+
+Thu Dec 23 08:12:59 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/net/smtp.rb: refactoring Net::SMTP#esmtp= to use an
+	  attr_accessor
+
+Thu Dec 23 06:35:41 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/net/smtp.rb: Net::SMTP should close the SSL connection if the
+	  connection verification fails.
+
+Thu Dec 23 01:47:58 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: remove #object_id. [ruby-dev:42840]
+
+Wed Dec 22 08:56:39 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add Module#private_constant and Module#public_constant.
+	  [ruby-dev:39685][ruby-core:32698]
+
+Wed Dec 22 07:59:23 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add IO#advise. [ruby-core:33110] [Ruby 1.9-Feature#4038]
+
+Tue Dec 21 23:45:31 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (Init_GC): move back object_id to Kernel. [ruby-dev:42840]
+
+Tue Dec 21 12:45:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (target_archs): remove temporary objects.
+
+	* enc/Makefile.in, enc/depend (clean): remove work directories.
+
+Tue Dec 21 07:39:12 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c: parenthesize macro arguments.
+
+Tue Dec 21 06:25:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rexml/test_contrib.rb (ContribTester#test_pos): should not
+	  use fixed path name for tests.  [ruby-dev:42827]
+
+	* test/rexml/test_sax.rb (SAX2Tester#test_socket): should not use
+	  fixed port for tests.  [ruby-dev:42828]
+
+Tue Dec 21 06:10:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (setup_args), vm.c (invoke_block_from_c),
+	  vm_insnhelper.c (caller_setup_args): reverted r30241 and r30243
+	  except for the test.
+
+Tue Dec 21 01:41:42 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* io.c : add an extra byte to buffer for the specification of read
+	  in Windows. see [ruby-core:33460] and r29980. and, we have to
+	  discuss how to do this one byte.
+
+Tue Dec 21 01:18:06 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* error.c: Fix build error for win32. This regression was
+	  introduced by r30271.
+
+Tue Dec 21 00:59:40 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_cleanup_func): Moved interrupted_lock
+	  destroying code from native_thread_destroy() to
+	  thread_cleanup_func() because it's platform independent logic.
+
+	* thread_win32.c (native_thread_destroy): ditto.
+	* thread_pthread.c (native_thread_destroy): ditto.
+
+Tue Dec 21 00:46:20 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_cleanup_func): Don't touch native threading
+	  resource at fork. Sadly this is purely bandaid. We need to
+	  implement proper fix later. [Bug #4169] [ruby-core:33767]
+
+Tue Dec 21 00:22:44 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* error.c (exit_success_p): Check status code more carefully.
+	  status code may have garbage in upper bit.
+
+Mon Dec 20 23:12:37 2010  Tanaka Akira  <akr@fsij.org>
+
+	* node.c: parenthesize macro arguments.
+
+Mon Dec 20 20:04:41 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add #__id__ and #object_id. [ruby-dev:42778]
+
+Mon Dec 20 20:03:21 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (native_thread_destroy): Fixed gvl_cond leak.
+
+Mon Dec 20 13:49:05 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS: Add item for RDoc 3.0.1
+
+	* lib/rdoc: Import RDoc 3.0.1, remove require for perl parser.
+
+Mon Dec 20 12:15:32 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Import RDoc 3.0.
+
+Mon Dec 20 01:55:03 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (Init_IO): Added O_DIRECT. This feature was proposed by
+	  Run Paint Run Run.
+	  [Feature #4015] [ruby-core:33018]
+
+Sun Dec 19 19:15:23 2010  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c: parenthesize macro arguments.
+
+Sat Dec 18 21:52:37 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* vsnprintf.c (BSD_vfprintf): suppress warning: "_WIN32" is not
+	  defined.
+
+Sat Dec 18 16:02:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (setup_args), vm.c (invoke_block_from_c),
+	  vm_insnhelper.c (caller_setup_args): fix of r30241. lambda block
+	  should check argument number.
+
+Sat Dec 18 14:42:29 2010  Tanaka Akira  <akr@fsij.org>
+
+	* load.c: parenthesize macro arguments.
+
+Sat Dec 18 10:07:04 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (setup_args, iseq_compile_each): optimize AMPER LAMBDA
+	  combination as block.
+
+Fri Dec 17 22:07:16 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (Init_GC): move #__id__ and #object_id to BasicObject.
+	  [ruby-dev:42778]
+
+Fri Dec 17 19:35:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/mkmf/base.rb (TestMkmf::FakeLog): capture output from mkmf.
+
+	* test/mkmf/test_find_executable.rb (test_find_executable):
+	  suppress meaningless differences for chkbuild.
+
+Fri Dec 17 13:26:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.mak (BASERUBY): quit with an error when BASERUBY was not
+	  able to set, just like configure.in does. [ruby-dev:42782]
+
+Fri Dec 17 07:04:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* iseq.c: parenthesize macro arguments.
+
+Fri Dec 17 04:18:37 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* transcode.c (str_encode): Alter comment for better wording and ri
+	  output.
+
+Fri Dec 17 00:05:40 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_advise): New API. IO#advise() allows to tell the
+	  ruby runtime how it expects to use a file handle. This feature
+	  can be improved a performance some situations.
+	  Note: This feature is mainly developed by Run Paint Run Run.
+	  Thank you! [ruby-core:33110] [Ruby 1.9-Feature#4038]
+
+	* io.c (do_io_advise): Helper function.
+	* io.c (io_advise_sym_to_const): ditto.
+
+Thu Dec 16 23:29:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (bin-comm): use transformed name.
+	  [ruby-dev:42777]
+
+Thu Dec 16 21:52:07 2010  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: parenthesize macro arguments.
+
+Thu Dec 16 21:46:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb (RbConfig): honor ARCHFLAGS and RC_ARCHS to
+	  override embedded ARCH_FLAG value on universal-darwin.
+
+Thu Dec 16 19:50:12 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{configure.bat,setup.mak,Makefile.sub} (PROGRAM_PREFIX,
+	  PROGRAM_SUFFIX): unite the differences of the names of macros of
+	  prefix and suffix.
+	  reported by HANEDA Norikatsu. [ruby-dev:42775]
+
+Thu Dec 16 08:04:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.h (RNode): match the type of flags to RBasic, and renamed
+	  nd_file as nd_reserved.
+
+	* iseq.c (set_relation), vm_insnhelper.c (vm_cref_push): nd_file
+	  is always zero-cleared.
+
+Thu Dec 16 07:22:30 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/unit.rb: Imported minitest 2.0.1 r6079.
+
+Wed Dec 15 20:45:02 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb (process_args): need to setup @help to print options.
+
+Wed Dec 15 11:19:33 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/zlib/test_zlib.rb (test_to_io): forgotten to fix with r30201.
+
+Wed Dec 15 11:07:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (simple_sendfile): enable on Mac OS X.
+
+	* io.c (nogvl_copy_stream_sendfile): moved precheck of copy length.
+
+	* io.c (nogvl_copy_stream_sendfile): should wait for both of
+	  read/write fds.
+
+Wed Dec 15 07:11:55 2010  Tanaka Akira  <akr@fsij.org>
+
+	* hash.c: parenthesize macro arguments.
+
+Wed Dec 15 04:02:00 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_set_config):
+	  fix compile error when !HAVE_X509V3_SET_NCONF. Thanks
+	  Chikanaga-san. [ruby-dev:42761] [Ruby 1.9-Bug#4158]
+
+Wed Dec 15 03:41:31 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ripper/test_parser_events.rb (TestRipper#test_block_variables):
+	  Limit address space 100MB instead 100KB. Quite frankly, This
+	  margin is too narrow to contain ruby. [ruby-dev:42763] [Bug#4159]
+
+Tue Dec 14 23:53:52 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (simple_sendfile): improve linux compatibility on FreeBSD,
+	  and now it works. But without cpuset -l 0, it still gets stuck.
+
+Tue Dec 14 20:31:33 2010  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c: parenthesize macro arguments.
+
+Tue Dec 14 18:31:48 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb: help messages.
+
+Tue Dec 14 18:19:03 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (help): there is no reason to use the abbreviation for here.
+
+Tue Dec 14 15:03:46 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (test_reopen, test_reinitialize): should close
+	  the temporary files.
+
+Tue Dec 14 14:24:15 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (make_tempfile): change the prefix from 'foo'
+	  to 'test_io' because the old one is meaningless and inconvenient.
+
+	* test/ruby/test_io.rb (test_binmode_after_closed): the temporary file
+	  made by make_temfile is already closed.
+
+Tue Dec 14 13:52:19 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (test_flush_in_finalizer[12]): should close
+	  temporary file because it's only used for taking pathname and
+	  unlinking the file after the end of the test (in GC phase).
+
+Tue Dec 14 13:34:33 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_s_open): should close the IO if some error
+	  occurs in initializing.
+
+Tue Dec 14 13:04:16 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTPRequest#send_request_body_data):
+	  set binmode to tempfile.
+
+Tue Dec 14 12:55:46 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/zlib/test_zlib.rb (*): should close files associated with zlib.
+
+Tue Dec 14 11:30:17 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_argf.rb (test_inplace_rename_impossible): unlink
+	  the renamed temporary file on no_safe_rename platforms.
+
+	* test/ruby/test_argf.rb (test_readlines_limit_0,
+	  test_each_line_limit_0): should close argf because the associated
+	  Tempfile object cannot unlink the temporary file when it's gc'ed
+	  on some platforms (Windows, etc.)
+
+Tue Dec 14 11:27:07 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/minitest/unit.rb (Minitest::Unit#_run_suite): split test
+	  name and its time. Thiw allows to know test's name when you are
+	  running tests and meet a test which spends long time at realtime.
+
+Tue Dec 14 11:25:20 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: Add -Werror=declaration-after-statement to default
+	  warning flag. If you are using GCC, this flag is useful to
+	  prevent breaking VC build.
+
+Tue Dec 14 10:25:57 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_decode0): how many gcc-c99isms
+	  must a man mend; before he can build with VC? r30178
+
+Mon Dec 13 21:26:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (simple_sendfile): disable the use of sendfile(2) on
+	  FreeBSD. It blocks on TestIO#test_copy_stream_socket.
+
+Mon Dec 13 18:35:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c: define USE_SENDFILE on FreeBSD or DragonFly BSD.
+	  Remove Mac OS X because its argument is different from them.
+
+Mon Dec 13 12:00:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* file.c: parenthesize macro arguments.
+
+Mon Dec 13 11:21:14 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (simple_sendfile): added for BSD version of sendfile(2).
+
+Mon Dec 13 09:50:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTPRequest#set_form): Added to support
+	  both application/x-www-form-urlencoded and multipart/form-data.
+	  There is a similar API, Net::HTTPRequest#set_form_data, but
+	  to keep its compatibility this is newly added. [ruby-dev:42729]
+
+Sun Dec 12 23:45:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): fix for __goto__ and __label__
+	  where were totally broken.
+
+Sun Dec 12 22:45:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ID_H_INCLUDES): now id.h depends on vm_opts.h.
+
+Sun Dec 12 20:42:47 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* template/id.h.tmpl: suppress all warning: "SUPPORT_JOKE" is not
+	  defined.  [ruby-dev:42730]
+
+Sun Dec 12 20:35:07 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* misc/rb_optparse.zsh: update how to install.
+
+	* misc/rb_optparse.zsh: avoid error when setopt noclobber.
+
+	* lib/optparse.rb: fix typo. pointed out at
+	  <http://d.hatena.ne.jp/nagachika/20101207>.
+
+Sun Dec 12 13:27:35 2010  Tanaka Akira  <akr@fsij.org>
+
+	* eval_error.c: parenthesize macro arguments.
+
+Sun Dec 12 11:53:24 2010  Tanaka Akira  <akr@fsij.org>
+
+	* error.c: parenthesize macro arguments.
+
+Sun Dec 12 04:01:58 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): fix: extra back slash is added when
+	  the string is dummy encoding and includes \x22 or \x5C.
+
+Sun Dec 12 02:42:24 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/openssl/ossl_asn1.c: indefinite length BER to DER encoding is
+	  properly supported.  Thanks Martin Bosslet! [ruby-core:33082]
+
+Sat Dec 11 17:43:34 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/bigdecimal/bigdecimal.h: suppress "warning: 'VPrint' declared
+	  'static' but never defined".
+
+Sat Dec 11 09:24:57 2010  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c: parenthesize macro arguments.
+
+Sat Dec 11 08:12:48 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl.c, ext/openssl/ossl_pkey_rsa.c: Document RSA, RSA
+	  encryption/decryption and PKCS #5 encryption/decryption.
+
+Sat Dec 11 06:23:41 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_x509name.c: include Comparable to provide #==.
+	  Document OpenSSL::X509::Name#<=>.  [Ruby 1.9-Feature#4116]
+
+Sat Dec 11 05:48:28 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: infinite loop on method_missing at loading.
+	  [ruby-dev:42716] [Ruby 1.9-Bug#4129]
+
+	* ext/tk/lib/multi-tk.rb: when no eventloop is running, ruby freezes at
+	  exit.
+
+Sat Dec 11 02:23:15 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/openssl/extconf.rb: try pkgconfig first, then fall back to
+	  normal have_library, etc.  Thanks Erik Hollensbe. [ruby-core:32406]
+
+Fri Dec 10 22:33:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* dln_find.c: parenthesize macro arguments.
+
+Fri Dec 10 20:05:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/id.h.tmpl (ruby_method_ids): suppress warnings.
+	  [ruby-dev:42730]
+
+Fri Dec 10 18:29:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): relatively called non-shared
+	  binary cannot be found in PATH, so use given pathname.
+
+Fri Dec 10 18:28:40 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (SCRIPTPROGRAMS): ignore backup files and etc.
+
+	* cygwin/GNUmakefile.in (scriptbin): set executable bit.
+
+	* tool/rbinstall.rb (install_recursive): always skip default ignored
+	  files.  if block is given, call it instead of calling install.
+
+	* tool/rbinstall.rb (bin-comm): use install_recursive.
+
+Fri Dec 10 18:12:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/mkmf/base.rb (TestMkmf#config_value): extract macro value from
+	  config.h.
+
+	* test/mkmf/test_sizeof.rb (TestMkmf::TestSizeof#test_sizeof_builtin),
+	  (TestMkmf::TestSizeof#test_sizeof_struct): more tests.
+
+	* lib/mkmf.rb (check_signedness): should use the prelude code.
+	  [ruby-dev:42731]
+
+	* lib/mkmf.rb (Logging.log_close): separate from Logging.logfile.
+
+	* test/mkmf/base.rb (TestMkmf::MKMFLOG): show mkmf.log at failures.
+
+	* test/mkmf/base.rb (TestMkmf#teardown): close log file for each tests.
+
+Fri Dec 10 11:36:43 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* compile.c (enum): remove a comma at end of enumerator list.
+
+	* constant.h (rb_const_flag_t): ditto.
+
+	* iseq.h (enum catch_type): ditto.
+
+	* iseq.h (enum defined_type): ditto.
+
+	* vm_core.h (enum iseq_type): ditto.
+
+	* vm_core.h (enum vm_special_object_type): ditto.
+
+Fri Dec 10 10:47:53 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* sprintf.c (_HAVE_SANE_QUAD_): Don't forget LP64, r30156.
+
+Fri Dec 10 10:37:17 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* sprintf.c (_HAVE_SANE_QUAD_): if a certain platform has LONG_LONG in
+	  8 byte, it might be sane quad. [ruby-core:33634]
+
+Fri Dec 10 10:07:59 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: remove version 1.1 features.
+
+Fri Dec 10 02:18:02 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/openssl/ossl_x509store.c (ossl_x509stctx_cleanup): removing C
+	  implementation of `cleanup`.
+
+	* ext/openssl/lib/openssl/x509.rb: adding ruby implementation of
+	  `cleanup`.  OpenSSL::X509::StoreContext#cleanup is deprecated since
+	  reusing the underlying struct doesn't make sense. [ruby-dev:42546]
+
+Thu Dec  9 20:14:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lvar_defined_gen, shadowing_lvar_gen, dvar_defined): no
+	  warnings for unused method and block arguments.
+	  [ruby-dev:42718] [ruby-dev:42724]
+
+Thu Dec  9 19:25:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* dln.c: parenthesize macro arguments.
+
+Thu Dec  9 18:51:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/accesslog.rb (WEBrick::AccessLog#format): support
+	  %{remote}p for logging remote (client) port number.
+	  [ruby-dev:42670]
+
+Thu Dec  9 11:00:30 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_dup): should copy contents only.  no instance
+	  variable, no class would be copied.  it would affect methods
+	  #sort, #reject, #transpose, #uniq, #compact, and #shuffle.
+	  [ruby-core:33640]
+
+	* array.c (rb_ary_reverse_m): ditto.
+
+	* array.c (rb_ary_rotate_m): ditto.
+
+Wed Dec  8 21:38:40 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/dl/lib/dl/struct.rb: clean a warning: assigned but unused
+	  variable. patched by Kouhei Yanagita. [ruby-dev:42722]
+
+	* ext/dl/lib/dl/import.rb: ditto.
+
+Wed Dec  8 21:36:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (shadowing_lvar_gen): fix line number.  [ruby-dev:42718]
+
+Wed Dec  8 20:37:11 2010  Tanaka Akira  <akr@fsij.org>
+
+	* dir.c: parenthesize macro arguments.
+
+Tue Dec  7 22:37:15 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* io.c (io_read): duplicate string if shared. [ruby-dev:42719]
+
+Tue Dec  7 22:31:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Officious): separate completion
+	  options from --help.  [ruby-dev:42690]
+
+	* lib/optparse.rb (OptionParser::Completion#candidate),
+	  (OptionParser::Switch#compsys): remove unused variables.
+
+Tue Dec  7 22:05:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (transcode_loop): call default handler of the given
+	  hash, method, proc or [] method as fallback.  [ruby-dev:42692]
+
+Tue Dec  7 21:59:37 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/light/node.rb: remove circular require.
+
+Tue Dec  7 21:56:01 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_light.rb: really suppress a warning.
+
+Tue Dec  7 21:51:57 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_light.rb: suppress a warning.
+
+Tue Dec  7 21:14:03 2010  Tanaka Akira  <akr@fsij.org>
+
+	* debug.c: parenthesize macro arguments.
+
+Tue Dec  7 21:06:38 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/doctype.rb, test/rexml/test_doctype.rb: suppress warnings.
+	  [ruby-core:33305]
+	  Reported by Aaron Patterson. Thanks!!!
+
+Tue Dec  7 18:56:52 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb (String#kconv): fix typo and update rdoc.
+	  patched by Kouhei Yanagita [ruby-dev:42696]
+
+Tue Dec  7 20:32:11 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_doctype.rb: add Accessor to test case name.
+
+Tue Dec  7 20:31:02 2010  Kouhei Sutou  <kou@clear-code.com>
+
+	* test/rexml/test_doctype.rb: Doctype -> DocType.
+
+Tue Dec  7 20:29:23 2010  Kouhei Sutou  <kou@clear-code.com>
+
+	* test/rexml/test_doctype_mixin.rb: rename to ...
+	* test/rexml/test_doctype.rb: ... this to remove needless name.
+
+Tue Dec  7 17:03:16 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (xlist): supported the XLIST command, which is an
+	  extension by Apple and Google.  patch by Geoff Youngs.
+	  [ruby-core:33521]
+
+Tue Dec  7 08:00:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, win32/Makefile.sub (WERRORFLAG): flag to treat
+	  warnings as errors.
+
+	* lib/mkmf.rb (Logging.postpone): yield log file object.
+
+	* lib/mkmf.rb (xsystem): add options, :werror only right now.
+
+	* lib/mkmf.rb (with_werror): check as if warnings are errors.
+
+	* lib/mkmf.rb (convertible_int): make declaration conflict
+	  warnings errors not to pass wrong type.  [ruby-dev:42684]
+
+	* lib/mkmf.rb (COMMON_MACROS): get rid of conflicts.
+
+	* win32/Makefile.sub (WARNFLAGS): make declaration conflict
+	  warnings errors if possible.
+
+Sun Dec  7 21:16:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* cont.c: parenthesize macro arguments.
+
+Tue Dec  7 00:27:14 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* win32/win32.c (rb_w32_read): fixed more for readline,
+	  and so on. [ruby-core:33511]
+
+Mon Dec  6 23:18:22 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/mkmf/base.rb (TestMkmf#setup): run quietly.
+
+	* test/mkmf/test_find_executable.rb (test_find_executable): use
+	  configured results.
+
+	* common.mk (test-build): test for build process.
+
+Mon Dec  6 22:47:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#candidate): skip separators.
+
+	* sample/optparse/opttest.rb: should not override --help.
+	  [ruby-dev:42690]
+
+Mon Dec  6 19:00:48 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* misc/rb_optparse.zsh: fix typos.
+
+Mon Dec  6 18:59:04 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add new encodings.
+
+Mon Dec  6 18:56:42 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_string.rb (TestString#test_scan): add a test for
+	  [ruby-core:33338] #4087.
+
+Mon Dec  6 18:55:36 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/uri/test_common.rb (TestCommon#test_encode_www_form): add
+	  tests for r30015.
+
+Mon Dec  6 10:39:54 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI::Parser#initialize_pattern):
+	  refix for restrict the pattern.
+
+Mon Dec  6 09:45:11 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl (OpenSSL): add toplevel documentation
+	* ext/openssl/ossl_ssl.c (SSLContext, SSLSocket: add additional
+	  documentation
+	* ext/openssl: move "let rdoc know about mOSSL" comments so they don't
+	  show up in output
+
+Mon Dec  6 09:16:46 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI::Parser#initialize_pattern):
+	  workaround fix pattern of hostname for RFC 3986. [ruby-dev:42672]
+
+Mon Dec  6 09:14:38 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/mkmf.rb (check_signedness): rename unused variable prelude.
+
+Sun Dec  5 17:56:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (make_metaclass): fix probable typo.  builtin type flag
+	  cannot be used with FL_TEST.
+
+Sun Dec  5 12:09:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb/init.rb (IRB.parse_opts): fix typo. [ruby-core:33574]
+
+Sun Dec  5 11:27:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): decrement for ungotten line.
+	  [ruby-dev:42680]
+
+Sun Dec  5 10:32:11 2010  Tanaka Akira  <akr@fsij.org>
+
+	* complex.c: parenthesize macro arguments.
+
+Sat Dec  4 11:39:17 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_x509ext.c (initialize): add documentation.
+
+Sat Dec  4 11:21:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_update_by): new API for Hash#update.
+
+Sat Dec  4 11:18:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* class.c: parenthesize macro arguments.
+
+Sat Dec  4 11:07:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_vm_inc_const_missing_count): missing prototype.
+
+Sat Dec  4 08:50:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (Init_iconv): no warnings if $VERBOSE is nil.
+
+Sat Dec  4 08:25:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): revert r30064 and r30071,
+	  because of [ruby-core:26761]. Bug#4106 rejected.
+
+Sat Dec  4 07:46:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (String#tr_cpp): substitute * with P like as
+	  autoconf.
+
+Fri Dec  3 22:36:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): protected methods should be
+	  checked against the real class.
+
+Fri Dec  3 20:23:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (convertible_int): define printf format prefix too.
+
+	* lib/mkmf.rb (convertible_int): detect convertible integer type.
+	  port RUBY_REPLACE_INT from configure.in.
+
+	* lib/mkmf.rb (check_sizeof): should return integer always.
+
+Fri Dec  3 12:54:48 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (RCFLAGS): VC10 and after only. fixed the problem
+	  of r30015. [ruby-core:33530]
+
+Fri Dec  3 12:41:52 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* gc.c (rb_objspace_free): With our "lazy-sweep" GC engine, it is
+	  possible for an object to survive until its surrounding object
+	  space is about to be freed.  Those objects, if any, remains
+	  leaked for the rest of a process life.  This is problematic
+	  because for instance a T_DATA object may have its own destructor
+	  to terminate something.
+
+	* vm.c (ruby_vm_destruct): ruby_current_vm termination should be
+	  somewhere after rb_objspace_free for above reason.
+
+Fri Dec  3 12:17:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): protected singleton methods should
+	  be visible from same real class methods.  [ruby-core:33506]
+
+Fri Dec  3 07:08:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_getline): round upto next char
+	  boundary.  [ruby-dev:42674]
+
+Fri Dec  3 06:52:46 2010  Tanaka Akira  <akr@fsij.org>
+
+	* compile.c: parenthesize macro arguments.
+
+Fri Dec  3 04:08:59 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_alias_internal): use st_insert2 and change return
+	  value to int.
+
+	* encoding.c (enc_alias): follow enc_alias_internal.
+
+Fri Dec  3 01:52:43 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_alias_internal): use xfree instead of free.
+
+Thu Dec  2 23:52:26 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* NEWS: entry for ruby_vm_at_exit().
+
+	* eval.c (ruby_cleanup): bug fix around at_exit (1) timing was
+	  wrong. (2) execution order was opposite.
+
+Thu Dec  2 23:05:14 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (RCFLAGS): -nologo switch is only available in
+	  newer versions of rc.exe. fixed the problem of r30012.
+
+Thu Dec  2 21:28:07 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/add/rails.rb: removed.
+
+Thu Dec  2 21:22:05 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_alias_internal): free the copied key and
+	  return NULL when given key is already registered.
+
+	* encoding.c (enc_alias): call set_encoding_const only when the
+	  alias is not registered yet.
+
+Thu Dec  2 19:58:24 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* vm.c (ruby_vm_at_exit): new API.  This enables extension libs to
+	  hook a VM termination.  Right now, because the VM we have is
+	  process global, most extensions do not deallocate resources and
+	  leave them to Operating System's reaping userland processes.  But
+	  in a future we plan to have multiple VMs to run simultaneously in
+	  a single process (MVM project).  At that stage we can no longer
+	  rely on OSes and have to manage every resources to be reclaimed
+	  properly.  So it is.  For a forward-compatibility reason this API
+	  is introduced now, encouraging you to be as gentle as you can for
+	  your resources; that is, tidy up your room.
+
+	* include/ruby/vm.h: ditto.
+
+	* vm_core.h (rb_vm_struct): new field.
+
+	* vm.c (vm_init2): initialize above new field.
+
+	* eval.c (ruby_cleanup): trigger those hooks.
+
+Thu Dec  2 17:00:44 2010  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: parenthesize macro arguments.
+
+Thu Dec  2 15:31:14 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): more fix. [ruby-core:33513]
+
+Thu Dec  2 13:41:43 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): workaround for console reading troubles.
+	  fixed [ruby-core:33511]
+
+Thu Dec  2 13:10:42 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.encode_www_form):
+	  split key-value when the value is Array like object.
+
+Thu Dec  2 10:39:39 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#set_form_data):
+	  use URI.encode_www_form for application/x-www-form-urlencoded.
+
+Thu Dec  2 10:38:40 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/extmk.rb: remove $makeflags.defined?, it should be $mflags.
+
+Thu Dec  2 10:19:47 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (rc): suppress meaningless message.
+
+Thu Dec  2 10:09:40 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/generator/extconf.rb: remove the lines which set -O3
+	  when -O option is not set.
+	  Note that -O3 doesn't always exist.
+
+	* ext/json/parser/extconf.rb: ditto.
+
+Thu Dec  2 10:01:59 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/extmk.rb: define $makeflags.defined? like $mflags.
+
+Thu Dec  2 07:20:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::GCStressOption): --gc-stress
+	  option.
+
+	* lib/test/unit.rb (Test::Unit::Mini#_run_suites): show the result
+	  even when interrupted on the way.
+
+Thu Dec  2 07:08:38 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (setattr): should retry on EINTR.
+	  [ruby-dev:42666]
+
+Thu Dec  2 02:30:50 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb: fixed positional wording to match revised order.
+
+Thu Dec  2 01:24:39 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/common.rb: don't use iconv on 1.9.
+	  patched by Shota Fukumori [ruby-core:33164]
+
+Thu Dec  2 01:02:03 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Update github/flori/json from 1.4.2+ to
+	  e22b2f2bdfe6a9b0. this fixes some bugs.
+
+Thu Dec  2 00:05:44 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: improve rdoc.
+	  This change the order of chapter because such overview should
+	  begin with simple examples.
+	  patched by Eric Hodel [ruby-core:33469]
+
+Wed Dec  1 22:01:49 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (Init_Numeric): fixed a potential bug when using bccwin32
+	  ruby with Microsoft's dll, though we already gave up of supporting
+	  bccwin32. [ruby-core:33503]
+
+Wed Dec  1 21:43:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* array.c: parenthesize macro arguments.
+
+Wed Dec  1 21:41:57 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_addrinfo.rb: extract Errno::EADDRINUSE as a method.
+
+	* test/socket/test_socket.rb: ditto.
+
+Wed Dec  1 15:08:32 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/openssl/test_ssl.rb (test_not_started_session): non socket
+	  argument of SSLSocket.new is not supported on Windows.
+
+Wed Dec  1 14:36:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_memhash): zero-filled strings should return
+	  different values.  [ruby-core:33500]
+
+Wed Dec  1 14:27:49 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 2.0.0 r5952.
+	* test/minitest/*.rb: ditto.
+	* lib/test/unit.rb: Compatibility fix for minitest changes.
+
+Wed Dec  1 10:16:41 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): fix typo (not 0xFD but 0xFE).
+
+Wed Dec  1 09:28:27 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: Follow .gnu_debuglink section.
+	  A user of distribution provided ruby will see line
+	  info if s/he has a debug package for ruby.
+	  patched by Shinichiro Hamaji [ruby-dev:42655]
+
+Wed Dec  1 01:29:15 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): inspect as a dummy encoding string
+	  when a UTF-16/32 (not BE/LE) string does not have a BOM.
+	  Unicode and some RFCs say that a string labeled as UTF-16/32
+	  doesn't have a BOM, it should be considered big endian.
+	  But many Windows programs generates little endian UTF-16
+	  strings without a BOM. So String#inspect treats a string
+	  labeled UTF-16/32 without a BOM as a dummy encoding string.
+	  patched by Martin Duerst. [ruby-core:33461]
+
+Tue Nov 30 17:04:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (parse_debug_line_cu): ignore DW_LNE_set_discriminator.
+	  To ignore, it needs to read a single unsigned LEB128 integer.
+
+Tue Nov 30 16:29:19 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c: undef HAVE_BACKTRACE when the OS is FreeBSD (in other
+	  words backtrace() is libexecinfo) and it is optimized.
+	  This temporary hack may be also applied to other libexecinfo
+	  environments.
+
+Tue Nov 30 16:23:23 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: improve rdoc.
+	  patched by Eric Hodel ref #4100
+
+Tue Nov 30 12:23:52 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): read only 1 byte at once on console.
+	  workaround of Windows bug. see [ruby-core:33460].
+	  this is not the final solution.
+
+Tue Nov 30 11:39:13 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: improve rdoc.
+	  patched by mathew murphy [ruby-core:33472] ref #4100
+
+Tue Nov 30 05:03:44 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/uri/common.rb (encode_www_form, encode_www_form_component):
+	  Improve English in documentation.
+
+	* ext/openssl/ossl_ssl.c (ssl_version=, ciphers=): Document
+	  #ssl_version=, add documentation for #ciphers=.
+
+Mon Nov 29 22:55:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/uri/common.rb (URI::WFKV_): get rid of backtrack explosion
+	  by nested repeat operators.  [ruby-core:33464]
+
+Mon Nov 29 22:53:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (scriptbin.mk): fix generated rules.
+
+	* win32/win32.c (rb_w32_write_console): fix argument type.
+
+Mon Nov 29 21:12:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-forward-sexp): stop after literal hash
+	  key labels.
+
+	* misc/ruby-mode.el (ruby-font-lock-keywords): highlight literal
+	  hash key labels as symbols.
+
+Mon Nov 29 18:31:31 2010  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb (test_unicode_public_review_issue_121):
+	  - Removed commented-out options that are no longer under discussion.
+	  - Added two more tests for forthcomming clarifications.
+
+Mon Nov 29 14:31:17 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_isatty): use GetConsoleMode() to determine the
+	  fd is console or not, just like rb_w32_write_console(). [experimental]
+
+Mon Nov 29 14:19:40 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (rb_w32_write_console): wrong prototype.
+
+Mon Nov 29 14:10:55 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write_console): fixed indentation.
+
+Sun Nov 28 22:13:39 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (NATIVE_MUTEX_LOCK_DEBUG): move and use it.
+
+	* ChangeLog: fix my timezone.
+
+Mon Nov 28 21:58:58 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c: remove pthread_atfork().
+
+Mon Nov 28 21:54:22 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (native_cond_*): Check return code.
+	  (Some OSes except Linux return error code).
+
+Sun Nov 28 21:46:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (thread_start_func_1): initialize native thread
+	  data immediately before starting.
+
+Sun Nov 28 14:56:32 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (struct argf): make lineno long, and reorder members.
+
+Sun Nov 28 14:55:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_win32.c (gvl_release, gvl_init): suppress warnings.
+
+Sun Nov 28 14:48:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (gvl_release, gvl_init): suppress warnings.
+
+	* vm_core.h (rb_vm_gvl_destroy): add prototype.
+
+Sun Nov 28 14:46:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (gvl_reinit): register atfork handler only in
+	  the parent process, to get rid of dead lock.
+
+Sun Nov 28 12:23:57 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, vm_core.h: make gvl_acquire/release/init/destruct
+	  APIs to modularize GVL implementation.
+
+	* thread_pthread.c, thread_pthread.h: Two GVL implementations.
+	  (1) Simple locking GVL which is same as existing GVL.
+	  (2) Wake-up queued threads.  The wake-up order is simple FIFO.
+	     (We can make several queues to support exact priorities, however
+	      this causes some issues such as priority inversion and so on.)
+	      This impl. prevents spin-loop (*1) caused on SMP environments.
+	  *1: Only one Ruby thread acquires GVL again and again.
+	      Bug #2359 [ruby-core:26694]
+
+	* thread_win32.c, thread_win32.h: Using simple lock
+	  not by CRITICAL_SECTION but by Mutex.
+	  Bug #3890 [ruby-dev:42315]
+
+	* vm.c (ruby_vm_destruct): ditto.
+
+Sun Nov 28 04:40:00 2010  Luis Lavena  <luislavena@gmail.com>
+
+	* io.c (io_fwrite): use rb_w32_write_console under Windows.
+
+	* win32/win32.c (rb_w32_write_console): added to write to write
+	  Unicode using WriteConsoleW for stdout/stderr. [ruby-core:33166]
+
+Sun Nov 28 03:58:47 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: improve rdoc.
+	  patched by Mike Perham [ruby-core:33433]
+
+Sat Nov 27 19:12:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: parenthesize macro arguments.
+
+Sat Nov 27 18:08:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (leap_year_v_p): fixed typo.  [ruby-dev:42631]
+
+Sat Nov 27 17:57:08 2010  Tanaka Akira  <akr@fsij.org>
+
+	* resolv.rb (Resolv::DNS): use the same DNS server when retry using
+	  TCP.  reported by Julian Mehnle.  [ruby-core:32970]
+
+Sat Nov 27 15:45:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): see CrashReport log on Mac OS X.
+
+	* configure.in: link addr2line only for ELF.
+
+Sat Nov 27 13:58:55 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#candidate): : was missing.  Thanks,
+	  Shota Fukumori.  [ruby-dev:42634]
+
+Sat Nov 27 12:07:05 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* man/ruby.1: Ruby man page from Arthur Gunn in [ruby-core:33412]
+
+Sat Nov 27 11:29:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#candidate): get rid of 1.9 syntax
+	  so that BASERUBY can be 1.8.
+
+Sat Nov 27 08:16:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* addr2line.c (rb_dump_backtrace_with_lines): should close fd on
+	  edge case.
+
+Fri Nov 26 13:33:24 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: apply a patch from shinichiro.h.
+
+Fri Nov 26 12:21:20 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: added to show source filename and line number of
+	  functions in backtrace. [ruby-dev:42625]
+	  a patch from shinichiro.h <shinichiro.hamaji AT gmail.com>
+
+	* addr2line.h: ditto.
+
+	* common.mk: add addr2line.$(OBJEXT).
+
+	* configure.in: check dl_iterate_phdr.
+
+	* vm_dump.c (rb_vm_bugreport): use rb_dump_backtrace_with_lines in
+	  addr2line.c when the binary is ELF.
+
+Fri Nov 26 12:12:50 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (setup_tree): restart setup_tree() for a node whose
+	  AnchorNode's type is ANCHOR_PREC_BEHIND or ANCHOR_PREC_BEHIND_NOT
+	  and divide_look_behind_alternatives() divided it to NT_ALT or
+	  NT_LIST. [ruby-core:33370]
+
+Fri Nov 26 11:40:11 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (dump_thread): get only required rights of the target
+	  thread because THREAD_ALL_ACCESS causes an access error on XP.
+	  reported by Masaya TARUI via IRC.
+
+Fri Nov 26 11:09:07 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (dump_thread): show the displacement from the beginning
+	  of the symbol.
+
+Fri Nov 26 10:48:23 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (dump_thread): follow the output of glibc.
+	  see [ruby-dev:42627]
+
+Fri Nov 26 09:48:45 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* re.c (rb_reg_initialize_str): should succeed the taint status from
+	  the origin. [ruby-core:33338]
+
+Fri Nov 26 09:32:37 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (dump_thread): seems to be necessary the 3rd argument of
+	  SymGetLineFromAddr64(), even though MSDN says it can be zero.
+
+Fri Nov 26 09:03:38 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (onig_is_prelude): added to check whether ruby is still
+	  in prelude (or other boot processes) or not.
+
+	* regcomp.c (optimize_node_left): use onig_is_prelude for printing.
+
+	* regcomp.c (set_optimize_info_from_tree): ditto.
+
+	* regcomp.c (onig_compile): ditto.
+
+	* regcomp.c (print_compiled_byte_code_list): print its address.
+
+	* regcomp.c (print_indent_tree): print its contents tree of
+	    ANCHOR_PREC_READ(_NOT) and ANCHOR_PREC_BEHIND(_NOT).
+
+Thu Nov 25 23:10:49 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (print_distance_range): use PRIuSIZE.
+
+	* regcomp.c (print_optimize_info): use %ld because the type of
+	  calculated value of integers is long.
+
+	* regexec.c (onig_print_compiled_byte_code): add prototype.
+
+	* regexec.c (match_at): add 2nd argument.
+
+Thu Nov 25 10:29:55 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/callback/mkcallback.rb (gencallback): shouldn't assume that
+	  VALUE is the same size with long.
+
+Thu Nov 25 10:03:14 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/win32ole/test_err_in_callback.rb (teardown): remove tmp file
+	  only when it exists.
+
+Thu Nov 25 01:38:25 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/big5-hkscs-tbl.rb: Update table as HKSCS-2008.
+	  patched by oCameLo oTnTh [ruby-core:33256]
+
+	* enc/big5.c: add alias Big5-HKSCS:2008 to Big5-HKSCS.
+
+Wed Nov 24 15:18:07 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vsnprintf (BSD_vfprintf): use QUADINT macro only when _HAVE_SANE_QUAD_
+	  macro is defined.
+
+Wed Nov 24 12:47:16 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vsnprintf (BSD_vfprintf): added VC++ compatible size specifications
+	  (I, I32, I64).
+
+Wed Nov 24 11:19:13 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): treat UTF-16 and UTF-32 as BE or LE.
+
+Wed Nov 24 06:35:32 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf_16_32.trans: add the UTF-32 converter.
+
+Wed Nov 24 05:40:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+Wed Nov 24 06:13:32 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (filecp, wstr_to_mbstr, mbstr_to_wstr):
+	  refactored.
+
+Wed Nov 24 05:40:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf_16_32.trans: add a converter from UTF-8 to UTF-16.
+
+Wed Nov 24 03:21:35 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf_16_32.trans: raise error on unpaired upper
+	  surrogates.
+
+Wed Nov 24 01:40:23 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/utf_16_32.h: add UTF-16 and UTF-32 as a dummy encoding.
+
+	* enc/trans/utf_16_32.trans: add a converter from UTF-16 to UTF-8.
+
+Tue Nov 23 21:59:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (wlink, rb_w32_getppid): use typedef instead of
+	  repeating complicated function prototypes.
+
+Tue Nov 23 18:54:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_thread_mark): should mark self in control
+	  frames.  [ruby-core:33289]
+
+Tue Nov 23 07:57:31 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/delta/parser.{ry,rb}: fixed a bug of token scanner.
+
+Tue Nov 23 07:29:24 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c, rational.c ({nucomp,nurat}_expt): added a check.
+
+Tue Nov 23 07:27:27 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (daynum): should be private.
+
+Tue Nov 23 07:22:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ChangeLog (change-log-indent-text): hanging indent.
+
+Tue Nov 23 06:30:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (SITE_DIR, VENDOR_DIR),
+	  version.c (ruby_initial_load_paths): exclude directories that
+	  are configured without them from $LOAD_PATH. [ruby-core:33267]
+
+	* configure.in (rubylibprefix): No ruby, No libprefix.
+
+Tue Nov 23 01:05:27 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): don't output floating point
+	  when the precision is 0. [ruby-dev:42615]
+
+Mon Nov 22 21:30:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_inspect): fix for ascii-compatible external
+	  encoding and different encoding string.  [ruby-core:33283]
+
+Mon Nov 22 18:45:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): should not duplicate rules.
+	  bug fix of r29842.
+
+Mon Nov 22 18:04:40 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/big5.c: split CP950 from Big5.
+
+	* enc/big5.c: split CP951 from Big5-HKSCS.
+
+	* enc/trans/big5.trans: import conversion table of Big5, Big5-HKSCS,
+	  CP950, and CP951 from ICU. they need fallback conversions.
+	  ref [ruby-core:33256]
+	  http://source.icu-project.org/repos/icu/data/trunk/charset/data/ucm/
+
+	* tool/transcode-tblgen.rb (import_ucm): add to import ucm files.
+
+Mon Nov 22 18:33:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_inspect): append for each chars instead of bulk
+	  copy if encoding conversion is needed.  [ruby-core:33283]
+
+Mon Nov 22 14:22:45 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (time_zone): use rb_locale_str_new_cstr to set encoding
+	  as locale and convert its content to internal encoding.
+	  [ruby-core:33278]
+
+Mon Nov 22 11:58:11 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_concat): set ENC_CODERANGE_VALID when the
+	  receiver is 7BIT and the argument is non ASCII.
+
+Mon Nov 22 01:48:58 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: some improvements for performance.
+
+Sat Nov 20 07:45:50 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/mkmf.rb: adding compilation support for ObjC/ObjC++ extensions.
+	  Thanks Scott Gonyea! [ruby-core:33260]
+
+Sat Nov 20 01:57:55 2010  Akio Tajima  <artonx@yahoo.co.jp>
+
+	* common.mk: add dependency(insns.inc) to compile.obj
+
+Fri Nov 19 23:05:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (insns_rules.mk): remove extra backslash.
+
+	* cygwin/GNUmakefile.in, win32/Makefile.sub (clean): rc files are
+	  made at compile time, so should be removed by clean.
+
+Fri Nov 19 22:09:46 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_core.rb
+	  (Tester#test_pretty_format_long_text_finite): skip a test that
+	  uses long string on small memory system. [ruby-dev:42599]
+
+Fri Nov 19 21:07:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb: shell completion support for zsh.  based on
+	  <http://d.hatena.ne.jp/rubikitch/20071002/zshcomplete>
+
+	* lib/optparse.rb: shell completion support for bash.
+
+Fri Nov 19 00:00:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (SCRIPTPROGRAMS): no needs on cygwin.
+
+	* win32/Makefile.sub (scriptbin): create script binaries.
+
+Thu Nov 18 23:21:23 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/formatters/pretty.rb (REXML::Formatters::Pretty#wrap):
+	  REXML::Formatters::Pretty#wrap used a recursive method call to
+	  format text.  This switches it to use an iterative approach.
+	  [ruby-core:33245]
+	  Patch by Jeremy Evans. Thanks!!!
+
+	* test/rexml/test_core.rb: add a test for it.
+
+Thu Nov 18 22:58:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h (rb_io_buffer_t): extract from rb_io_t.
+
+Thu Nov 18 07:37:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (reconfig): force reconfigure with previous options.
+
+	* common.mk (showconfig): show configure flags, like as
+	  `config.status --config' generated by recent autoconf.
+
+Thu Nov 18 07:16:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/langinfo.c (strncasecmp): get rid of redefinition.
+
+Thu Nov 18 00:02:17 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Upgrading output encoding with ASCII content
+	  as needed.  [ruby-core:33229]
+
+Wed Nov 17 23:19:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/configure.bat: remove quotes from arguments to be quoted.
+
+	* lib/mkmf.rb (create_makefile): use forward slashes in messages.
+
+	* lib/mkmf.rb (create_makefile): make extension libraries messages
+	  brief.
+
+	* win32/Makefile.sub (MAKEDIRS): should not include silent flag.
+
+	* common.mk (ext/ripper/ripper.c, ext/json/parser/parser.c): pass
+	  Q and ECHO.  [ruby-core:33226]
+
+Wed Nov 17 16:09:52 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/test_tracer.rb: new test case.
+	  minimal regression test for r29280.
+
+Wed Nov 17 16:04:23 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/envutil.rb (Test::Unit::Assersions#assert_warn):
+	  new assertion to assert that a particular warning message is
+	  displayed.
+	  forward port from branches/ruby_1_9_2@29795.
+
+Wed Nov 17 15:16:48 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regint.h (OnigOpInfoType): constify name.
+
+	* regcomp.c (op2name): constify return value.
+
+	* regcomp.c (onig_print_compiled_byte_code): use PRIuPTR and
+	  uintptr_t to clean warnings.
+
+	* regcomp.c (print_indent_tree): use PRIxPTR and intptr_t.
+
+	* regexec.c (match_at): use PRIdPTR and intptr_t.
+
+Wed Nov 17 09:49:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/shift_jis.c (property_name_to_ctype): fix memory leak.
+
+	* enc/euc_jp.c (property_name_to_ctype): ditto.
+
+Wed Nov 17 08:54:04 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Upgrading output encoding as needed.  [ruby-core:33135]
+
+Tue Nov 16 22:30:39 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_insnhelper.c (vm_throw): remove fear of undefined behavior :-)
+	  Coverity Scan found this bug.
+
+Tue Nov 16 09:33:00 2010  Kenta Murata <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb (to_digits): avoid unused
+	  variables warning, reported by Aaron Patterson.
+
+Tue Nov 16 06:39:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (PACK_ITEM_ADJUST): return nil not result array and yield
+	  values if block is given.  [ruby-core:33193]
+
+Tue Nov 16 00:21:20 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* regparse.c (and_cclass, or_cclass): fix memory leak.  Coverity Scan
+	  found this bug.  [ruby-dev:42579]
+
+Tue Nov 16 00:07:32 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gc.c (assign_heap_slot): fix fear of memory leak and memory
+	  violation.  Coverity Scan found this bug.
+
+Mon Nov 15 23:54:45 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval_intern.h (CHECK_STACK_OVERFLOW): it was not intended to add
+	  size_t to a pointer typed VALUE*.  Coverity Scan found this defect.
+
+Mon Nov 15 23:41:21 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_set_exception_local_table, iseq_set_local_table,
+	  rb_iseq_build_from_ary): fix type inconsistency (which is benign
+	  because sizeof(ID) == sizeof(ID*), though).  Coverity Scan found
+	  these bugs.
+
+Mon Nov 15 22:47:27 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_funcall): ensure va_end after va_init_list.  Coverity
+	  Scan found this bug.
+
+Mon Nov 15 08:36:12 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/racc/parser.rb (do_parse, yyparse): using class eval to define
+	  method and avoid __send__.
+
+Mon Nov 15 06:43:48 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* etc/openssl/ossl_ssl.c (ossl_ssl_get_cert): raise exception if
+	  pointer is invalid.  Thanks Ippei Obayashi! [ruby-dev:42573]
+
+Sun Nov 14 17:57:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/Makefile.in (distclean): should not remove sources which are
+	  distributed in tarball.
+
+Sun Nov 14 16:48:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_set_token_info): turn on/off with directives.
+	  [ruby-core:25442]
+
+Sun Nov 14 12:05:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_readlines): forward to current_file for arguments
+	  check.  http://twitter.com/nagachika/status/3634254856589312
+
+Sun Nov 14 08:48:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/setup.mak (-basic-vars-, -runtime-): suppress trailing
+	  space and compiler command line.
+
+Sun Nov 14 04:22:32 2010  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup (config.h): Added HAVE_LABS and HAVE_LLABS to config.h.
+
+	* symbian/configure.bat: Changed packaging version in line with API
+	  style 3 versioning.
+
+Sat Nov 13 16:37:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (showflags, help): emit messages at once.
+
+	* win32/Makefile.sub (MSG, EOM): remove surrounding quotes by %~I.
+
+Sat Nov 13 01:31:30 2010  Akio Tajima <artonx@yahoo.co.jp>
+
+	* win32/Makefile.sub: reorder variable End Of Message (don't display it)
+
+Fri Nov 12 20:52:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (showflags, help): use caret to quote leading spaces on
+	  Windows.
+
+	* Makefile.in, common.mk, cygwin/GNUmakefile.in, enc/depend,
+	  ext/ripper/depend, lib/mkmf.rb, win32/Makefile.sub: caddle up.
+
+Fri Nov 12 16:35:31 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: support C level backtrace information on FreeBSD.
+	  When devel/libexecinfo is installed on FreeBSD, now ruby
+	  can show C level backtrace information.
+	  http://www.freebsd.org/cgi/cvsweb.cgi/ports/devel/libexecinfo/
+
+Fri Nov 12 09:58:30 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.mak: use findstr.exe instead of find.exe, because all
+	  target build platforms should have findstr.exe, and, find.exe often
+	  means another command such as cygwin's.
+
+Fri Nov 12 00:30:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): need PRI_LL_PREFIX.
+
+Thu Nov 11 23:38:32 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: ANSI C-conforming const and volatile are mandatory
+
+	* configure.in (AC_C_CONST, AC_C_INLINE, AC_C_VOLATILE): check
+	  before used in other checks.
+
+	* configure.in (RUBY_CHECK_PRINTF_PREFIX): should not break from
+	  RUBY_WERROR_FLAG, so that ac_c_werror_flag gets restored.
+
+Thu Nov 11 23:04:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (warn_deprecated): show caller position.
+
+Thu Nov 11 23:03:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_close): untie tied io before closing.
+
+	* io.c (argf_write): add ARGF.write and so on.
+
+	* io.c (argf_read_nonblock): add ARGF.read_nonblock.
+
+Thu Nov 11 21:49:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/stats.rb (RDoc#print): get rid of NaN.
+
+Thu Nov 11 21:47:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (SHOWFLAGS): show compile flags.
+
+	* common.mk: hide long command lines by default.  verbose-mode is
+	  turned on by V=1 as before.
+	  http://jarp.does.notwork.org/diary/200605b.html#200605121
+
+Thu Nov 11 21:32:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_func): accept variable address.
+
+	* ext/win32ole/extconf.rb: libuuid is needed on cygwin.
+
+Thu Nov 11 21:24:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): use cygwin_conv_path on cygwin 1.7 or
+	  later.
+
+	* ruby.c (push_include_cygwin): ditto.
+
+Thu Nov 11 20:49:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRI_LL_PREFIX): format type specifier for
+	  LONG_LONG may vary on platforms.
+
+Thu Nov 11 20:45:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (SYMBOL_PREFIX): separate from EXPORT_PREFIX.
+
+	* win32/mkexports.rb (Exports#each_export): use SYMBOL_PREFIX.
+
+Wed Nov 10 07:20:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (scriptbin): make executable file from
+	  scripts with stub.
+
+	* ruby.c (load_file_internal): assume xflag for exe file as well
+	  as no-shebang file.
+
+	* tool/rbinstall.rb: install script programs.
+
+	* win32/mkexports.rb (Exports#initialize): alias ruby_sysinit for
+	  stub.
+
+	* win32/stub.c: stub for scripts.  [EXPERIMENTAL]
+
+Tue Nov  9 21:57:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (init_funcname): allocate and build initialization
+	  funciton name at once.
+
+Tue Nov  9 21:14:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (AC_FUNC_GETPGRP, AC_FUNC_SETPGRP): no need when
+	  not used.
+
+	* configure.in (EXPORT_PREFIX): check generic prefix.
+
+Tue Nov  9 13:24:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regenc.c (onigenc_minimum_property_name_to_ctype):
+	  \p{...} should be case insensitive. [ruby-core:33000]
+
+	* regenc.c (onigenc_property_list_add_property):
+	  ditto.
+
+	* enc/euc_jp.c (init_property_list, property_name_to_ctype):
+	  to lowercase property names.
+
+	* enc/shift_jis.c (init_property_list, property_name_to_ctype):
+	  ditto.
+
+Tue Nov  9 13:29:36 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (overlapped_socket_io): get rid of a warning of 64bit
+	  mingw.
+
+Tue Nov  9 10:44:19 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_strtod): this code uses FPU's rounding system.
+	  But x86's FPU calculates double precision floating-point
+	  numbers in 80bit precision, so it fails to round the value.
+	  So ensure the value is assigned a variable. [ruby-dev:42551]
+	  see also [ruby-math:00802]
+	  http://www.shudo.net/java-grandprix99/strictfp/
+
+Tue Nov  9 07:30:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_syserr_new): new function to make SystemCallError
+	  instance without errno.  [EXPERIMENTAL]
+
+	* error.c (rb_syserr_fail, rb_mod_syserr_fail): ditto.
+
+Tue Nov  9 05:54:57 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/*.rb: Remove unused variable warnings.
+	  Patch by Run Paint [ruby-core:30991]
+
+	* lib/rubygems/*, lib/rdoc/*.rb, lib/rake/*.rb: ditto
+
+Mon Nov  8 18:26:03 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_hdtoa): fix type cast and bufsize.
+
+Mon Nov  8 15:40:56 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): fix precision specifier doesn't
+	  work well on %f. [ruby-dev:42552]
+
+Mon Nov  8 14:41:40 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (get_wsa_extension_function): typos.
+
+Mon Nov  8 13:41:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb,
+	  enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  Add Age property to regexp. [ruby-core:33019]
+	  patched by Ammar Ali, tested by Run Paint Run Run
+
+Mon Nov  8 12:16:39 2010  Ben Walton  <bwalton@artsci.utoronto.ca>
+
+	* configure.in: support -h for solaris linker when gcc not used
+
+Mon Nov  8 11:47:39 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (finish_overlapped_socket): refactoring.
+
+Mon Nov  8 11:02:21 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (get_proc_address): refactoring.
+
+	* win32/win32.c (get_wsa_exetinsion_function): refactoring.
+
+Mon Nov  8 09:45:35 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/gbk-tbl.rb: Add euro sign. [ruby-core:33094]
+	  CP936, which is de facto definition of GBK, has it.
+	  http://msdn.microsoft.com/en-us/goglobal/cc305153.aspx
+
+Mon Nov  8 07:26:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check only the first symbol to get rid of
+	  duplication.  [ruby-core:33084] #4031
+
+Sun Nov  7 10:13:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (NM): check on all platforms.  #4031
+
+Sun Nov  7 06:16:33 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* test/openssl/test_ocsp.rb: adding test for r29699.  Thanks Elise
+	  Huard! [ruby-core:32460]
+
+Sat Nov  6 07:33:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_export_prefix): check for prefixed
+	  underscore of exported symbols
+
+	* tool/rbinstall.rb (bin-comm): prepend prolog shell script if
+	  necessary.
+
+	* configure.in (LIBRUBY_RELATIVE): use rpath token expansion.
+
+Sat Nov  6 07:24:01 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/ruby.pc.in (arch, sitearch): reordered.
+
+	* configure.in: keep failed file.
+
+Sat Nov  6 07:03:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_fork_err): save errinfo before fdopen.
+
+Sat Nov  6 00:43:58 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspcid_initialize): an optional
+	  parameter may be used to specify the OpenSSL::OCSP::CertificateId on
+	  initialization.  Thanks Elise Huard! [ruby-core:32460]
+
+Fri Nov  5 12:23:01 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_{process,system}.rb (test_fallback_to_sh):
+	  meaningless and wrong tests where /bin/sh does not exist.
+
+	* process.c (proc_spawn_v): should spawn, not exec.
+
+Fri Nov  5 01:21:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_exec_v, proc_spawn_v): try to execute with sh if
+	  no shebang.  [ruby-core:32745] [EXPERIMENTAL]
+
+Fri Nov  5 00:39:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_readlines, rb_io_each_line): limit must not be zero.
+	  a patch from Tomoyuki Chikanaga at [ruby-dev:42538].  #4024
+
+Fri Nov  5 00:14:15 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/extconf.rb: fixing ffi library location on windows.
+	  Thanks Usa! [ruby-core:32930]
+
+Thu Nov  4 20:04:44 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_newobj): force garbage_collect() if GC.stress == true.
+
+Thu Nov  4 19:48:22 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* ChangeLog: missed to write a last ChangeLog.
+
+	* gc.c (gc_finalize_deferred): removed.
+
+	* gc.c (rb_gc_finalize_deferred): Do not invoke a free_unused_heaps().
+
+Thu Nov  4 19:45:27 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (run_final): do not need argument obj.
+
+Thu Nov  4 19:26:10 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (before_gc_sweep): fix commit miss.
+
+Thu Nov  4 19:20:46 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (after_gc_sweep, before_gc_sweep):
+	  invoke rb_sweep_method_entry() as soon as possible.
+
+Thu Nov  4 19:13:58 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (after_gc_sweep, slot_sweep): finalizers should be invoked
+	  as soon as possible.
+
+Thu Nov  4 10:30:40 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* configure.in (--with-valgrind): Now this option is default on.
+	  You can still explicitly disable this feature by specifying
+	  --without-valgrind.
+
+Thu Nov  4 02:06:16 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c (fiber_t_alloc): raise an error when fiber is going to be
+	  initialized twice.  [ruby-dev:42524]
+
+Thu Nov  4 02:04:25 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c (rb_fiber_resume): raise an "double resume" error when root
+	  fiber is going to be resumed.  [ruby-dev:42523]
+
+Wed Nov  3 14:17:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (OpenStruct#delete_field): also undefine
+	  accessor methods.  [ruby-core:33010]
+
+Wed Nov  3 14:13:46 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_buf_cat): concatenation of valid
+	  encoding string and invalid encoding string should result
+	  invalid encoding.  [ruby-core:33027]
+
+Wed Nov  3 08:58:59 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, vm.c, vm_core.h: remove USE_VALUE_CACHE option.
+
+Wed Nov  3 07:47:25 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (RubyLex#identify_string): parse multiple
+	  regex options.  a patch from Heesob Park in [ruby-core:32988].
+
+Wed Nov  3 07:33:57 2010  Tanaka Akira  <akr@fsij.org>
+
+	* vm_method.c (rb_clear_cache_by_class): just return if the class has
+	  no method.  reported by Eric Wong.  [ruby-core:32689]
+
+Tue Nov  2 22:50:25 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/visitor.rb (initialize): push accessor
+	  methods to subclass that actually uses them.
+
+Tue Nov  2 22:47:08 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/visitor.rb (accept): switch to
+	  a dispatch cache rather than case / when statement.
+
+Tue Nov  2 21:46:52 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS: fix a typo.
+
+Tue Nov  2 20:10:32 2010  Tajima Akio <artonx@yahoo.co.jp>
+
+	* test/rake/test_tasks.rb: clear env var which is used by the test.
+	  [ruby-dev:42508]
+
+Tue Nov  2 00:25:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: win64 is just same with win32 about socket.
+	  notice: but wince is not same.
+
+Mon Nov  1 21:25:57 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* main.c: <stdlib.h> is needed, to introduce the getenv(3)
+	  prototype declaration.  Without it a C compiler shall infer
+	  the getenv type as "int getenv(...);", but this is totally
+	  wrong, especially when your machine's sizeof(int) and
+	  sizeof(char*) differs.  On such environment a return value
+	  of getenv(3), which is in fact a char*, might first casted
+	  into a int (loses data here), and then casted back to char*
+	  by automatic integral promotion to fit to the prototype of
+	  ruby_set_debug_option().
+
+Sun Oct 31 23:27:09 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (finalizer_table, objspace->final.table):
+	  Create finalizer_table at Init_heap().
+	  Remove all null checks of finalizer_table.
+
+	* gc.c (mark_tbl): skip if no table entries.
+
+	* gc.c (slot_swee): remove useless need_call_final check.
+
+Sun Oct 31 22:32:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_objspace_free): finalizers should be called separately
+	  from freeing objspace.  [ruby-dev:42479]
+
+Sun Oct 31 22:24:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): free current VM and its objspace even
+	  when exiting by SystemExit.
+
+Sun Oct 31 22:10:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (new_child_iseq): adjust argument types.
+
+	* iseq.c (prepare_iseq_build, rb_iseq_new),
+	  (rb_iseq_new_with_bopt_and_opt, rb_iseq_new_with_opt),
+	  (rb_iseq_new_with_bopt): ditto.
+
+	* compile.c (iseq_set_exception_table): suppress warnings.
+
+	* insns.def (putspecialobject, defined): ditto.
+
+	* iseq.c (iseq_load): ditto.
+
+Sun Oct 31 09:30:51 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: some refactoring.
+	  - move decl. of rb_compile_option_struct to iseq.h.
+	  - define enum iseq_type.
+	  - define enum vm_special_object_type.
+
+	* compile.c: some refactoring.
+	  - apply above changes.
+	  - (struct iseq_link_element): change value of type.
+	  - remove unused decl.
+	  - fix comment.
+	  - rename iseq_build_body and iseq_build_exception to
+	    iseq_build_from_ary_body and iseq_build_from_ary_exception.
+
+	* iseq.h: define enum catch_type and enum defined_type.
+
+	* insns.def: apply above changes.
+
+	* iseq.c: define ISEQ_MAJOR_VERSION and ISEQ_MINOR_VERSION.
+
+Sat Oct 30 23:38:59 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/encoding.rb: untabify.
+
+Sat Oct 30 21:06:37 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/encoding.rb: use Ruby native encoding mechanism.
+	  [ruby-dev:42464]
+	* lib/rexml/encodings/: remove.
+
+	* lib/rexml/document.rb, lib/rexml/formatters/default.rb,
+	  lib/rexml/output.rb, lib/rexml/parseexception.rb,
+	  lib/rexml/parsers/baseparser.rb, lib/rexml/source.rb,
+	  lib/rexml/xmldecl.rb: use Ruby's native Encoding object.
+
+	* test/rexml/, test/rss/: follow the above encoding changes.
+
+	* NEWS: add REXML's incompatible change about encoding.
+
+Sat Oct 30 17:23:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strtod): get rid of overflow/underflow as possible.
+
+Sat Oct 30 14:37:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ruby_pc): erase runtime-defined variables and
+	  check if generated pc file is valid.
+
+	* template/ruby.pc.in (DEFFILE): need for mingw.
+
+	* template/ruby.pc.in (LIBRUBY): fix the order.
+
+Sat Oct 30 11:33:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (ruby_pc): ignore missing variables.
+
+	* template/ruby.pc.in: add missing variables for mswin.
+
+Sat Oct 30 10:24:35 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* object.c: Make BasicObject.new accept no parameter.
+	  Revert of r26135 [ruby-core:27080], as per [ruby-core:32952].
+
+Sat Oct 30 09:40:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c: use constants in id.h.
+
+Sat Oct 30 09:08:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/closure.c (fiddle_closure): embed cif not reference
+	  so that the content surely get initialized.  [ruby-dev:42480]
+
+Sat Oct 30 07:01:53 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv-replace.rb: suppress warning.
+
+	* lib/open-uri.rb: ditto.
+
+Sat Oct 30 06:32:52 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/pathname/test_pathname.rb (TestPathname#test_grpowned?): the
+	  group of the created file is inherited from the parent
+	  directory on BSDs and MacOS X.  Linux also inherit the group if
+	  the setgid bit of the directory is set.  It causes the test fail.
+	  fixed by Shota Fukumori.  [ruby-dev:42458]
+
+Sat Oct 30 05:58:54 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: retry via TCP if UDP reply is truncated.
+	  fixed by Julian Mehnle.  [ruby-core:32407]
+
+Sat Oct 30 00:35:13 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_s_compile): fix optional argument.
+	  a patch from Yutaka HARA [ruby-core:32953] [Ruby 1.9-Bug#4001]
+
+Sat Oct 30 00:24:42 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (memsize_of): take care of
+	  T_CLASS/const_tbl.
+	  a patch from nagachika <nagachika00@gmail.com> [ruby-dev:42490]
+
+Fri Oct 29 23:32:36 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* test/profile_test_all.rb: added.
+	  You can use test-all profiler with the following command:
+	    RUBY_TEST_ALL_PROFILE=true make test-all
+	  This command generates ./test_all_profile and you can analyze
+	  which tests consume memories.
+
+	* test/runner.rb: ditto.
+
+Fri Oct 29 10:02:03 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb,
+	  enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  Add 'Unknown' Script.
+	  patched by Run Paint Run Run. [ruby-core:32937] #3998
+
+Fri Oct 29 05:13:34 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (ObjectSpace.memsize_of_all): rename
+	  ObjectSpace.total_memsize_of_all_objects() to
+	  ObjectSpace.memsize_of_all([klass]).
+	  Accept Class object to filter the objects.
+
+	* test/objspace/test_objspace.rb: fix test for above change.
+
+Fri Oct 29 03:04:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_dump): fix expected length. [ruby-core:32935]
+
+Thu Oct 28 23:31:39 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (before_gc_sweep, run_final): fix decrement timing of final_num.
+
+Thu Oct 28 20:11:30 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb,
+	  enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  Update Oniguruma for Unicode 6.
+	  patched by Run Paint Run Run. [ruby-core:32923] #3989
+
+Thu Oct 28 20:06:25 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/oniguruma.h (ONIGENC_CTYPE_SPECIAL_MASK):
+	  change mask from 128 to 256. [ruby-core:32931]
+
+Thu Oct 28 12:06:54 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httprequest.rb (read_request_line): extend max
+	  length to 2083. This is from Internet Explorer's max uri
+	  length. http://support.microsoft.com/kb/208427 [ruby-core:32924]
+
+Thu Oct 28 04:00:08 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (GC.stat): added.  [ruby-dev:38607]
+
+	* test/ruby/test_gc.rb: add a test for above.
+
+Thu Oct 28 03:13:06 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (memsize_of): fix rdoc.
+
+	* ext/objspace/objspace.c (total_memsize_of_all_objects): added.
+
+	* test/objspace/test_objspace.rb:
+	  - add a test for ObjectSpace.total_memsize_of_all_objects.
+	  - add two tests for ObjectSpace.memsize_of (for nil and Fixnum).
+
+Wed Oct 27 23:55:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (Init_iconv): warn deprecated use.
+
+Wed Oct 27 18:50:17 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (rb_big2long, rb_big2ulong): rb2ulong() returns VALUE, but
+	  its real range is ulong. So, if the size of VALUE is bigger than
+	  ulong, upper bits are always zero even if the actual value is
+	  negative.
+	  fixed #3490
+
+Wed Oct 27 18:27:17 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#pipe): should close write end of pipe
+	  before closing read end, to get rid of timing problem.
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#pipe): ditto.
+
+Wed Oct 27 18:14:27 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_getppid): support Win64.
+
+Wed Oct 27 15:07:19 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (w32_error): should get error no only once, because
+	  the result of the second getting will indicate the error of the
+	  first FormatMessage() call.
+
+Wed Oct 27 13:51:25 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#pipe): need to propagate exceptions
+	  in read/write thread. fix r29541.
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#pipe): ditto.
+
+Wed Oct 27 12:05:40 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* class.c (clone_const): need to return value. fix r29602.
+
+Wed Oct 27 11:58:58 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (NUM2LONG_internal): add cast to get rid of a
+	  non GCC compiler warning. this is intentional type conversion.
+
+Wed Oct 27 09:25:46 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c: apply documentation patch by Run Paint Run Run.
+	  [ruby-core:32915]
+
+Wed Oct 27 02:12:10 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* object.c (Init_Object), constant.h, variable.c
+	  (rb_mod_private_constant, rb_mod_public_constant,
+	  set_const_visibility, rb_const_get_0): add Module#public_constant
+	  and private_constant.  [ruby-dev:39685][ruby-core:32698]
+
+	* test/ruby/test_module.rb: add tests for above.
+
+Wed Oct 27 02:02:54 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* class.c, constant.h, gc.c, method.h, object.c, variable.c,
+	  vm_insnhelper.c: use struct rb_constant_entry_t as entry of
+	  RCLASS_CONST_TBL.  RCLASS_CONST_TBL has contained VALUE of constant
+	  directly.  Now instead rb_const_entry_t is contained in
+	  RCLASS_CONST_TBL,  rb_const_entry_t is managed by malloc, and
+	  have not only the value itself but also visibility flag.
+	  This is another preparation for private constant (see
+	  [ruby-dev:39685][ruby-core:32698]).
+
+Wed Oct 27 01:56:34 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* class.c, gc.c, object.c, variable.c, vm_insnhelper.c,
+	  include/ruby/ruby.h: separate RCLASS_CONST_TBL from RCLASS_IV_TBL.
+	  RCLASS_IV_TBL has contained not only instance variable table but
+	  also constant table.  Now the two table are separated to
+	  RCLASS_CONST_TBL and RCLASS_IV_TBL.  This is a preparation for
+	  private constant (see [ruby-dev:39685][ruby-core:32698]).
+
+Tue Oct 26 18:51:00 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/scanf.rb (extract_float): allow 2.e+2 style.
+	  [ruby-dev:42452] #3978
+
+Tue Oct 26 18:09:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (ASFLAGS): needs INCFLAGS.
+
+	* configure.in (rb_cv_dynamic_alloca): check if extra source for
+	  dynamic size alloca.
+
+	* missing/x86_64-chkstk.s (___chkstk): necessary for alloca of
+	  amd64-mingw32msvc-gcc on Ubuntu.
+
+	* thread_win32.c (ruby_alloca_chkstk): check stack overflow
+
+Tue Oct 26 18:04:53 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/ruby.pc.in (Libs): needs DLDFLAGS.
+
+Tue Oct 26 12:47:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (pkgconfig-data): moved from Makefile.in.
+
+	* tool/rbinstall.rb: install pc file only if non-empty.
+	  [ruby-core:32901] #3983
+
+	* win32/Makefile.sub (ruby_pc): create pc file.
+
+Tue Oct 26 09:13:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_gcc_atomic_builtins): check for atomic
+	  builtins, all are not available in Apple derivative gcc.
+
+Tue Oct 26 00:29:26 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (pkgconfig-data): create pkg-config metadata file.
+
+	* tool/rbinstall.rb: install pkg-config metadata file.
+
+	* template/ruby.pc.in: template of pkg-config metadata file.
+
+Mon Oct 25 16:38:07 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* signal.c  (rb_atomic_t): GCC (of  at least  recent versions)
+	  has  ubiquitous  support  for  atomic operations.   On  that
+	  compiler a C program can  issue a memory barrier using these
+	  dedicated  instructions.  According to  the GCC  manual they
+	  cargo culted  this feature form  the Itanium ABI  so chances
+	  are that  other compilers  could also support  this feature.
+	  But so far GCC is the  only compiler that I know to have it.
+	  Also note that this works on non-Itanium machines.
+
+Mon Oct 25 06:21:35 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): prec digits fractal part should be
+	  appended to 0 if prec is given.  [ruby-dev:42453] #3979
+
+Mon Oct 25 02:57:21 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk (run.gdb): Quit gdb on 'make gdb' when
+	  no signals are received.
+
+Mon Oct 25 00:25:23 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: some corrections of documentation.
+
+Sun Oct 24 17:14:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c, gc.c, hash.c, object.c, string.c, struct.c,
+	  transcode.c, variable.c, vm.c, vm_insnhelper.c, vm_method.c:
+	  replace calls to rb_error_frozen() with rb_check_frozen().  a
+	  patch from Run Paint Run Run at [ruby-core:32014]
+
+	* include/ruby/intern.h (rb_check_frozen): optimize.
+	  [ruby-core:32878]
+
+Sun Oct 24 15:16:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Mini#run): abort if interrupted.
+
+	* lib/test/unit.rb (Test::Unit::Mini#run_test_suites): show the
+	  result even when interrupted on the way.
+
+	* lib/test/unit.rb (Test::Unit::Mini#run_test_suites): ensure
+	  output sync mode to be restored.
+
+Sun Oct 24 14:11:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_define_method): defined method is run with the default
+	  public visibility regardless the visibility context of definition.
+	  [ruby-core:30638]
+
+Sun Oct 24 12:08:54 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/test/unit.rb: make test/unit play nicely with the rake test
+	  loader. [ruby-core:32864]
+
+Sun Oct 24 00:25:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (test_segv_test): follow up the
+	  change at r29556.
+
+Sat Oct 23 14:39:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: $extmk should be true for test/runner.
+
+Sat Oct 23 10:55:37 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c (rb_vm_bugreport): fix to add bug outputs.
+	  - loaded script ($0)
+	  - loaded features ($")
+	  - process memory map on Linux (/proc/self/maps)
+
+	* vm_dump.c (rb_vmdebug_stack_dump_raw): fix header message.
+
+Fri Oct 22 14:50:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  assertion message must not be nil.
+
+Fri Oct 22 13:59:50 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  treat nil case. Please run test-all before commit such change.
+
+Thu Oct 21 23:58:14 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_lazy_sweep): Variable declarations should be at
+	  the head of block.
+
+Thu Oct 21 23:56:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (objspace_each_objects, rb_objspace_each_objects): use
+	  struct.
+
+	* gc.c (objspace_each_objects): fix return with no value.
+
+Thu Oct 21 23:47:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_initialize): remove useless intermediate variable.
+
+Thu Oct 21 16:07:20 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_f_select): change rdoc.
+	  patched by Eito Katagiri [ruby-core:31805]
+
+Thu Oct 21 15:55:21 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httpauth/digestauth.rb
+	  (WEBrick::HTTPAuth::ProxyDigestAuth#check_uri): privated.
+	  [ruby-dev:42344]
+
+Thu Oct 21 15:50:25 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_f_select): add correct rdoc.
+	  patched by Dave Thomas [ruby-core:32467]
+
+Thu Oct 21 15:42:01 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/telnet.rb (Net::Telnet#close): added.
+	  patched by Erik Hollensbe [ruby-dev:42260] #3830
+
+Thu Oct 21 13:08:00 2010  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (rb_objspace_each_objects): don't lazy sweep in
+	  rb_objspace_each_objects. [Bug #3940] [ruby-dev:42369]
+
+Thu Oct 21 00:05:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#pipe): get rid of deadlock on pipe.
+	  a patch from Tomoyuki Chikanaga at [ruby-dev:42435].  #3970
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#pipe): ditto.
+
+Wed Oct 20 23:54:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dbm/dbm.c: rdoc based on a patch by mathew meta AT
+	  pobox.com, at [ruby-core:32853].
+
+Wed Oct 20 10:47:21 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_strtod): reject 0x1.p+0. [ruby-dev:42432] #3966
+
+Wed Oct 20 10:00:57 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): print floating point on "%#a".
+	  [ruby-dev:42431] Bug#3965
+
+Tue Oct 19 19:30:11 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): clear ALT flag for %a.
+	  [ruby-core:32841] [ruby-core:32848]
+
+Tue Oct 19 12:19:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): fix over-count of field size.
+
+Tue Oct 19 03:08:52 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): use HEXPREFIX flag for prefix of %a.
+	  [ruby-core:32841]
+
+Mon Oct 18 13:18:01 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_class_init): Define
+	  Digest::Class.new(). [Feature #3954]
+
+Mon Oct 18 12:58:40 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_pack): refine the document.  [ruby-dev:42397]
+	  (pack_unpack): ditto.
+
+Mon Oct 18 10:19:00 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (transport_request): @socket may be nil.
+	  patched by Egbert Eich [ruby-core:32829]
+
+Mon Oct 18 09:57:28 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* sprintf.c (BSD_vfprintf): wrong padding around prefix and
+	  floating point with %a. [ruby-dev:42403] Bug #3956
+
+Sun Oct 17 22:36:33 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/delta.rb: added an rdoc tag.
+
+Sun Oct 17 10:47:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_mod_remove_const): update rdoc.
+	  [ruby-core:31957]
+
+Sun Oct 17 10:40:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_define_{class,module}_id_under): register to be
+	  marked, which probably are defined and used internally.
+
+Sat Oct 16 11:10:55 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_encoding2cp): set codepage 20936
+	  according to GB2312. [Bug #3937] [ruby-core:32758]
+
+Sat Oct 16 10:54:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (CPP): already used in .c.i rule.
+
+	* cygwin/GNUmakefile.in (DLLWRAP, WINDRES): add --driver-name and
+	  --preprocessor options explicitly.  [ruby-core:32776]
+
+Sat Oct 16 10:06:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/sdbm/_sdbm.c (SEEDUPS, BADMESS): make settable using command
+	  line options.
+
+	* ext/sdbm/_sdbm.c (makroom): suppress unused result warning.
+
+	* ext/sdbm/extconf.rb: disable BADMESS, a library should not emit
+	  messages directly.
+
+Sat Oct 16 08:39:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_strerror): get English message first, instead of
+	  system default. see [ruby-dev:42358].
+
+Sat Oct 16 00:08:00 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c (rb_hash_aref): skip calling "default" method
+	  if it is not needed, for speed-up.
+
+Fri Oct 15 23:36:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (NUM2DEVT, DEVT2NUM, PRI_DEVT_PREFIX): fallback to
+	  unsigned int.
+
+Fri Oct 15 22:54:46 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_hresult2msg): get English message first,
+	  instead of system default. [ruby-core:32765]
+
+Fri Oct 15 22:47:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (VALUE): prefer long over uintptr_t,
+	  FIX2LONG expects VALUE to be long at least.
+
+	* include/ruby/ruby.h (FIX2LONG): parenthesize the argument.
+
+Fri Oct 15 20:30:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (dev_t): use RUBY_REPLACE_TYPE.
+
+	* file.c (rb_stat_inspect): use PRI_DEVT_PREFIX.
+
+Fri Oct 15 17:26:57 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_pack): simplify comparison of explicit_endian
+	  as pointed by nobu.
+
+	* pack.c (pack_unpack): ditto.
+
+Fri Oct 15 16:40:37 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_pack): fix more than one modifiers appear in the
+	  format string. [ruby-core:32793]
+
+	* pack.c (pack_unpack): ditto.
+
+Thu Oct 14 23:20:42 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_folderitem2_invokeverb.rb: refactoring.
+
+Thu Oct 14 22:18:29 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, iseq.c, vm_insnhelper.c: rename variable name
+	  (ip -> iseq).
+
+Thu Oct 14 20:41:27 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_pack): support endian modifiers: < and >.
+	  [ruby-dev:42376] Feature #3491
+
+	* pack.c (pack_unpack): ditto.
+
+Thu Oct 14 20:50:51 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (reg_get_val): expand environment in
+	  the pathname. [Bug #3907]
+
+Thu Oct 14 07:35:07 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (DEVT2NUM): added. Size of dev_t is depend on the
+	  environment even if POSIX defines dev_t as unsigned integer.
+	  For example, OpenVMS, 64bit Solaris 9, and NetBSD 6 defines
+	  dev_t as 64bit unsigned integer.
+
+	* file.c (rb_stat_dev): use DEVT2NUM.
+
+	* file.c (rb_stat_dev_major): dev_t is not long. major(3)'s return
+	  value is int.
+
+	* file.c (rb_stat_dev_minor): dev_t is not long. minor(3)'s return
+	  value is int.
+
+	* configure.in: check size of dev_t.
+
+Thu Oct 14 07:22:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_and, rb_ary_or), class.c (rb_mod_init_copy),
+	  gc.c (undefine_final), time.c (time_mload): get rid of
+	  type-punning casts.
+
+Thu Oct 14 04:16:41 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (ruby_float_step): fix Numeric#step with infinity unit
+	  doesn't works well. [ruby-core:32779]
+
+Wed Oct 13 23:16:46 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/enc-unicode.rb: get rid of lots of warnings.
+
+	* iseq.c (insn_operand_intern, rb_iseq_disasm): fix format specifiers.
+
+	* vm.c (thread_free): ditto.
+
+	* numeric.c (check_uint): get rid of overflow on LLP64 platforms.
+
+	* insns.def (opt_case_dispatch): use st_data_t.
+
+Wed Oct 13 22:32:34 2010  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/util.rb (CGI::unescape): bugfix to unescape the multibyte
+	  string. Thanks nobu and tDiary dev members. [Bug #3909]
+
+Wed Oct 13 21:13:00 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (int_chr): raise error when the value is negative.
+
+Wed Oct 13 19:24:08 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* vm.c (ruby_vm_destruct): This function type was wrong; correct to the prototype.
+
+Wed Oct 13 14:58:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (rb_num_to_uint): fix 32bit logic.
+
+Wed Oct 13 12:53:43 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (rb_num_to_uint): added to check the range of arguments.
+	  Mainly for negative value with NUM2UINT on 32bit environment.
+
+	* string.c (rb_str_concat): use rb_num_to_uint.
+
+Wed Oct 13 12:10:02 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (w32_error): get English message first, instead
+	  of system default. see [ruby-core:32765].
+	  [experimental]
+
+Wed Oct 13 11:04:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* debug.c (ruby_set_debug_option): define always for binary
+	  compatibility with debug env enabled binary.
+
+	* signal.c (ruby_enable_coredump): ditto.
+
+Wed Oct 13 10:52:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ruby_executable_node): missing prototype.
+
+Wed Oct 13 05:23:04 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* missing/strchr.c: add strlen's prototype.
+
+	* missing/strstr.c: ditto.
+
+Wed Oct 13 00:21:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c (struct mktime_arg): constified.
+
+	* ext/syck/rubyext.c (mktime_do, mktime_r, rb_syck_mktime): fix
+	  function signatures.
+
+Wed Oct 13 00:18:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (MurmurMagic): get rid of literal overflow.
+
+	* configure.in (RUBY_CHECK_PRINTF_PREFIX): check for printf format
+	  specifier if possible.
+
+Tue Oct 12 23:58:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_open_osfhandle, rb_w32_wopen, rb_w32_pipe):
+	  use uintptr_t instead of long for win64.
+
+	* win32/win32.c (socketpair_internal): suppress warnings.
+
+	* win32/win32.c (ftruncate): use HANDLE instead of long for win64.
+
+	* vsnprintf.c (BSD_vfprintf): fix cast.
+
+	* numeric.c (rb_num2fix): result of rb_num2long is SIGNED_VALUE.
+
+	* compile.c (iseq_build_body), error.c (set_syserr, get_syserr),
+	  (syserr_initialize), gc.c (define_final, rb_gc_copy_finalizer),
+	  (run_final), hash.c (rb_hash_aref, rb_hash_lookup2),
+	  (rb_hash_fetch_m, rb_hash_clear, rb_hash_aset, eql_i),
+	  iseq.c (iseq_load, iseq_data_to_ary), marshal.c (r_symlink),
+	  thread.c (rb_thread_local_aref),
+	  variable.c (generic_ivar_remove, ivar_get, rb_const_get_0),
+	  (rb_cvar_get), vm.c (rb_vm_check_redefinition_opt_method),
+	  vm_insnhelper.c (vm_get_ev_const), vm_method.c (remove_method),
+	  ext/iconv/iconv.c (map_charset): use st_data_t.
+
+	* compile.c (iseq_build_body), insns.def (getglobal, setglobal),
+	  iseq.c (iseq_load, iseq_data_to_ary), util.c (valid_filename):
+	  use VALUE.
+
+	* gc.c (obj_free, rb_objspace_call_finalizer): fix truncating
+	  cast.
+
+	* gc.c (mark_current_machine_context): suppress warnings.
+
+	* compile.c (iseq_compile_each): fix truncating cast.
+
+	* cont.c (fiber_setcontext): missing variable definition.
+
+Tue Oct 12 19:25:13 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* error.c (exc_to_s): use OBJ_INFECT.
+
+	* error.c (name_err_to_s): ditto.
+
+	* error.c (name_err_mesg_to_str): ditto.
+
+	* error.c (syserr_initialize): ditto.
+
+Tue Oct 12 19:07:55 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* error.c (syserr_initialize): taint message if mesg is given
+	  and it is tainted.
+
+Tue Oct 12 18:25:43 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_ungetc): always see Bignum. On 32bit valid value
+	  may be a Bignum. On 64bit for errors. [ruby-dev:42366]
+
+Tue Oct 12 18:25:04 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_concat): use unsigned int for GB18030.
+
+Tue Oct 12 17:53:49 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric (check_uint): the mask must refer to VALUE.
+
+Tue Oct 12 17:47:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric (check_uint): set MSB for negative value.
+
+	* numeric (rb_num2uint): return value's type of rb_num2ulong
+	  is VALUE.
+
+	* numeric (int_chr): variable i can't be negative.
+
+Tue Oct 12 16:04:37 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_strerror): get English message first, instead
+	  of system default. see [ruby-dev:42358].
+	  [experimental]
+
+Tue Oct 12 15:52:35 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_strerror): unmap some range of errno for
+	  workaround of VC10's strerror() and sys_nerr problem.
+	  based on a patch from Akio Tajima, [ruby-dev:42355].
+
+Tue Oct 12 15:36:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_ungetc): use unsigned int for GB18030.
+
+Tue Oct 12 15:14:21 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_putc): support multibyte characters.
+	  [ruby-core:30697]
+
+Tue Oct 12 15:10:31 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (rb_enc_uint_chr): split from int_chr.
+
+	* numeric.c (int_chr): use rb_enc_uint_chr.
+
+	* include/ruby/encoding.h (rb_enc_uint_chr): added.
+
+Tue Oct 12 14:04:41 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (int_chr): a codepoint of Ruby M17N must be 32bit
+	  unsigned int; GB18030 uses MSB. Also note that OnigCodePoint
+	  is defined as unsigned int.
+
+Tue Oct 12 12:20:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (dump_thread): foolish mistake.
+
+Tue Oct 12 10:39:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_MINGW32): canonicalize only on mingw.
+
+Mon Oct 11 20:20:23 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (HTTP.get): specify ASCII-8BIT as the result
+	  encoding of Zlib::GzipReader.
+	  http://hibari.2ch.net/test/read.cgi/tech/1281473294/271
+
+Mon Oct 11 17:42:54 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* error.c (syserr_initialize): use mesg's encoding when locale
+	  encoding is US-ASCII. If locale encoding is not US-ASCII,
+	  assume err has non ASCII characters. [ruby-dev:42358]
+
+Mon Oct 11 14:03:54 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* error.c (syserr_initialize): set the encoding of Errno::*#message
+	  as locale. [ruby-dev:42358]
+
+Mon Oct 11 06:38:27 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_set_encoding):
+	  StringIO#set_encoding can get 2nd argument and optional hash
+	  for API compatibility to IO. [ruby-dev:42356]
+
+Mon Oct 11 06:11:30 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_set_encoding): use rb_funcall2 when the io is not
+	  a T_FILE. [ruby-dev:42356]
+
+Sun Oct 10 18:42:23 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#replace): Apply a bit of optimization.
+
+Sun Oct 10 10:20:07 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_MINGW32): canonicalize as like mswin version.
+
+Sun Oct 10 05:33:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_signal_buff_size, rb_signal_exec): moved
+	  declarations from thread.c.
+
+Sat Oct  9 16:54:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RSHIFT): quote to get rid of argument expansion
+	  for autoconf 2.68.
+
+Sat Oct  9 11:00:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_reset_event_flags, exec_event_hooks): ignore
+	  hooks marked as removed.
+
+	* thread.c (thread_exec_event_hooks): remove hooks to be removed.
+
+	* thread.c (rb_threadptr_remove_event_hook, rb_remove_event_hook):
+	  defer removing hooks if running the hooks.  [ruby-dev:42350]
+
+Sat Oct  9 10:51:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_threadptr_exec_event_hooks): suppress each event
+	  hooks separately.
+
+	* thread.c (thread_suppress_tracing): split from
+	  ruby_suppress_tracing, accepting thread pointer and event mask.
+
+Sat Oct  9 08:16:01 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_threadptr_remove_event_hook): fix typo.
+
+Fri Oct  8 10:52:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RBCONFIG): depends on version.h due to
+	  RUBY_PATCHLEVEL.  [ruby-core:32709]
+
+Fri Oct  8 00:24:54 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Fixing documentation typos.  [ruby-core:32712]
+
+Thu Oct  7 09:14:28 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_exec.c (vm_exec_core):  Treat clang as non gcc on this
+	  context: It has __asm__ but doesn't works well.
+
+Wed Oct  6 12:28:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/uri/generic.rb (URI::Generic#hostname): new method.
+	  (URI::Generic#hostname=): ditto.
+
+	* lib/open-uri.rb: use URI#hostname
+
+	* lib/net/http.rb: ditto.
+
+	  reported by Adam Majer.  [ruby-core:32056]
+
+Wed Oct  6 11:52:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (fptr_finalize): write_mutex might have been destroyed
+	  already in finalization phase, as the order of finalizers is not
+	  guaranteed.  rb_mutex_t should be used in place of Mutex object
+	  in the future.
+
+Tue Oct  5 22:17:02 2010  wanabe  <s.wanabe@gmail.com>
+
+	* win32/mkexports.rb: revert r29320 and r29402.
+
+Mon Oct  4 12:43:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (regexp): dregexp has literal string only at the head
+	  and successors are array.  [ruby-core:32682]
+
+Mon Oct  4 10:22:21 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c (rand_init): This checks the value is in 32bit or not,
+	  so use int32_t, not int.
+
+Mon Oct  4 09:47:39 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c (rand_init): remove useless assignment.
+
+	* re.c (update_char_offset): remove unused variable.
+
+	* re.c (read_escaped_byte): ditto.
+
+Mon Oct  4 09:30:42 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/bn.rb (Integer#to_bn): OpenSSL::BN.new
+	  accepts only Strings, so call Integer#to_s(16).
+	  16 is for an optimization. [ruby-dev:42336]
+
+Mon Oct  4 07:57:51 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c (fiber_memsize): Return size.
+	  Before this change, fiber_memsize always returns 0.
+
+Mon Oct  4 07:16:55 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/unicode.c (onigenc_unicode_property_name_to_ctype):
+	  remove useless assignment.
+
+	* vm.c (vm_make_proc_from_block): ditto.
+
+	* variable.c (rb_ivar_count): ditto.
+
+Mon Oct  4 06:40:24 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (clean-rdoc): Don't use \ in variable expansion.
+	  BSD make treats it as an escape character.
+
+Mon Oct  4 00:01:53 2010  wanabe  <s.wanabe@gmail.com>
+
+	* tool/config.sub: revert r29320, r29324, r29347, r29354, r29365
+	  to automake-1.11.1. [ruby-core:32634]
+
+	* win32/mkexports.rb: no longer use 'mingw64'. a patch from Luis Lavena
+	  at [ruby-core:32678].
+
+Sun Oct  3 20:36:37 2010  Akio Tajima (arton)  <artonx@yahoo.co.jp>
+
+	* test/win32ole/test_folderitem2_invokeverb.rb: Change creating
+	  shortcut verb to 'Link' [Bug #3339]
+
+Sun Oct  3 19:44:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (Makefile): get rid of duplicated ruby target when
+	  already there it was.
+
+Sat Oct  2 22:59:32 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_thread.rb: add for win32ole with Thread.
+
+Fri Oct  1 17:03:00 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/webrick/test_httpproxy.rb (TestWEBrickHTTPProxy::test_upstream_proxy):
+	  My machine fails this test at this line, saying 503 service unavailable.
+
+Thu Sep 30 16:11:08 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_getenv): should return NULL if specified name
+	  is empty. a patch from Heesob Park at [ruby-core:32650]
+
+Thu Sep 30 15:18:23 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (command_asgn): allow command_call to be right hand side
+	  expression of chained assignment.  [ruby-dev:42313]
+
+Thu Sep 30 10:55:38 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (ruby_setenv): workaround for old Windows. a patch from
+	  Heesob Park. [ruby-core:32353]
+
+Thu Sep 30 09:29:06 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.encode_www_form): change treatment of
+	  undefined value in given array as latest internet draft for
+	  application/www-form-urlencoded.
+	  http://tools.ietf.org/html/draft-hoehrmann-urlencoded-01
+
+Thu Sep 30 09:34:03 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (dump_thread): fixed wrong type of return value of
+	  SymGetModuleBase64(). [ruby-dev:42306]
+
+Wed Sep 29 21:04:05 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions::test_script_from_stdin):
+	  As usual, PTY is not always available.
+
+Wed Sep 29 18:38:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/config.sub (x64): regularize only for pc vendor, and strip
+	  useless 64 suffix.
+
+Wed Sep 29 17:53:02 2010  Kenta Murata <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BIGDECIMAL_ENABLE_VPRINT):
+	  VPrint is usually disabled.  It's only used in debugging.
+
+Wed Sep 29 17:41:34 2010  Kenta Murata <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_save_limit):
+	  return the result of a block.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_save_limit):
+	  add a test for the above change.
+
+Wed Sep 29 16:18:03 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (dump_thread): remove unused optional arguments.
+
+Wed Sep 29 13:26:30 2010  Kenta Murata <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_save_rounding_mode):
+	  return the result of a block.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_save_rounding_mode):
+	  add a test for the above change.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_save_exception_mode):
+	  add a test for the return value of BigDecimal.save_exception_mode.
+
+Wed Sep 29 12:45:30 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_div2, BigDecimal_add2,
+	  BigDecimal_sub2, BigDecimal_mult2, VpLimitRound): remove meaningless
+	  casts to get rid of compiler warnings.
+
+Wed Sep 29 12:35:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VPrint, VpToString): fix format.
+
+Wed Sep 29 12:31:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/known_classes.rb (RDoc::KNOWN_CLASSES): add Encoding.
+
+Tue Sep 28 20:50:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/config.sub (x64): regularize same as mswin.
+
+Tue Sep 28 20:06:14 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): add windows support.
+	  based on patches from Peter Weldon at [ruby-core:32551]
+
+Mon Sep 27 23:30:34 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (opt_case_dispatch), vm_insnhelper.c:
+	  execute the procedures of "when" clauses by bytecode
+	  instead of st_foreach() when the object does not hit
+	  prepared hash.  [ruby-dev:42304]
+
+Mon Sep 27 15:54:03 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/net/http/test_https.rb: As always, localhost is not
+	  guaranteed to be resolved as 127.0.0.1.  But a SSL
+	  certificate needs a socket to listen on a specific address
+	  where a CN resolves to.  On situations where localhost is
+	  not 127.0.0.1, these tests are not possible.
+
+Mon Sep 27 15:25:05 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/net/imap/test_imap.rb: resurrection of r29259.
+	  this change depends on minitest 1.7.1.
+
+	* lib/test/unit/assertions.rb: ditto.
+
+Sun Sep 26 22:59:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/config.sub (x86_64-pc-mingw64): regularize.
+
+Sun Sep 26 22:21:07 2010  wanabe  <s.wanabe@gmail.com>
+
+	* ext/openssl/ossl_hmac.c (ossl_hmac_hexdigest, ossl_hmac_s_hexdigest),
+	  ext/openssl/ossl_pkey_ec.c (ossl_ec_group_set_seed),
+	  ext/openssl/ossl_ssl_session.c (ossl_ssl_session_to_der),
+	  ext/openssl/ossl_pkcs7.c (numberof): suppress warnings.
+	  [ruby-core:31932]
+
+Sun Sep 26 10:25:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/config.{guess,sub}: updated to automake-1.11.1.
+
+Sat Sep 25 22:48:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBY_DLDFLAGS): fix quoting.
+
+Sat Sep 25 10:30:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBY_DLDFLAGS): use -unexported_symbol only
+	  when available.  http://trac.macports.org/ticket/26341
+
+Sat Sep 25 10:05:49 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: Always add -mieee for Renesas SH4.
+	  Thanks, Nobuhiro Iwamatsu. [Feature #3874] [ruby-core:32548]
+
+Sat Sep 25 01:34:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (install-cross): target to install cross-compiling
+	  stuff.
+
+Fri Sep 24 23:44:59 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_call), vm.c (invoke_block_from_c),
+	  vm_insnhelper.c (vm_yield_with_cfunc): pass given block.
+	  [ruby-core:32075]
+
+	* vm_eval.c (rb_funcall_passing_block): new function to call
+	  method with passing given block.
+
+Fri Sep 24 15:50:43 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_to_i): fix rdoc: String#to_i raises an
+	  exception when base is invalid. [ruby-core:31685]
+
+Fri Sep 24 15:28:35 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_rindex): use rb_enc_prev_char instead of repeated
+	  str_nth.
+	  patched by Michael Selig [ruby-core:32498]
+
+Fri Sep 24 14:19:12 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/test_pty.rb: Same as 229281; existence of PTY class do not
+	  guarantee a successful pty operation.
+
+Fri Sep 24 06:25:55 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.7.2 r5879.
+	* test/minitest/*.rb: ditto.
+
+Thu Sep 23 23:09:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_get_cref0): cref is stacked only in normal
+	  iseqs, so check if it is the case first.
+
+Thu Sep 23 23:08:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/config.sub: mingw64 should use x86_64.  [ruby-core:32514]
+
+Thu Sep 23 21:40:40 2010  wanabe  <s.wanabe@gmail.com>
+
+	* ext/socket/raddrinfo.c (init_addrinfo, inspect_sockaddr): suppress
+	  warnings.  see [ruby-core:31932].
+
+Thu Sep 23 19:27:57 2010  wanabe  <s.wanabe@gmail.com>
+
+	* thread_win32.c (w32_wait_events, w32_close_handle): suppress warnings.
+	  see [ruby-core:31932].
+
+Thu Sep 23 18:54:39 2010  wanabe  <s.wanabe@gmail.com>
+
+	* tool/config.sub: add mingw64.
+
+	* win32/mkexports.rb (Exports::Mingw64): added.
+
+	* win32/mkexports.rb (Exports::Mingw32): renamed from Exports::Mingw.
+
+Thu Sep 23 09:01:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_cref_push): no outer cref is needed for proc
+	  from method.  Bug #3786, Bug #3860, [ruby-core:32501]
+
+Wed Sep 22 17:12:01 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/openssl/utils.rb (OpenSSL#silent): always restore $VERBOSE.
+	  [ruby-dev:42285]
+
+Wed Sep 22 16:59:40 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/test_prime.rb (TestPrime#test_new): the warning expected have
+	  not been displayed when $VERBOSE == nil. Patch by Shota Fukumori
+	  a.k.a. sora_h. [ruby-dev:42272]
+	  Recovers $stderr even if StringIO.new fails. Reported by unak.
+
+Wed Sep 22 01:55:48 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_method.rb: fix last commit.
+
+Wed Sep 22 01:49:52 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_method.rb: add a test for [ruby-core:30534].
+
+Wed Sep 22 00:52:44 2010  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/rdoc/ri/store.rb (save_cache): remove duplicate entries.
+
+Wed Sep 22 00:00:05 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_f_pathname): Pathname() translated
+	  from pathname.rb.
+
+Tue Sep 21 22:18:30 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/mkconfig.rb: fixed build problem on mswin64 introduced in r29278.
+
+Tue Sep 21 02:42:35 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/pathname/test_pathname.rb (TestPathname#test_mkdir): fix typo.
+
+Mon Sep 20 23:23:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (bracket): get rid of scanning at the end of the pattern
+	  string, not to raise an exception while globbing command line.
+	  [ruby-core:32478]
+
+Mon Sep 20 11:25:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (Init_pathname): Pathname#=~ undefinition
+	  translated from pathname.rb.
+
+Mon Sep 20 02:34:11 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (check_rounding_mode, BigDecimal_mode):
+	  raise ArgumentError instead of TypeError passing invalid modes.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_mode, test_round):
+	  change against the above modifications.
+
+Sun Sep 19 22:08:39 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/mkmf.rb (try_link): rdoc
+	  (try_compile): ditto
+	  (try_cpp): ditto
+	  (try_func): ditto
+	  (try_var): ditto
+	  (try_run): ditto
+	  (egrep_cpp): ditto
+
+Sun Sep 19 20:43:33 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* configure.in (--disable-install-doc): disables capi too, in addition
+	  to rdoc.
+	  (--disable-install-rdoc): a new option for disabling only rdoc.
+	  (--disable-install-capi): a new option for disabling only capi.
+
+Sun Sep 19 20:37:45 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (clean): removes all documents on cleaning.o
+	  (CAPIOUT): new variable.
+	  (clean-capi, distclean-capi, realclean-capi): new targets
+
+	* Makefile.in (clean-capi, distclean-capi, realclean-capi): ditto.
+
+	* win32/Makefile.sub (clean-capi, distclean-capi, realclean-capi):
+	  ditto.
+
+Sun Sep 19 13:44:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBY_SO): fix an oversight of replace
+	  RUBY_INSTALL_NAME with RUBY_SO_NAME.  a patch from Jeremy Evans
+	  at [ruby-core:32474].
+
+Sun Sep 19 07:48:20 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_unlink): Pathname#unlink and
+	  Pathname#delete translated from pathname.rb.
+
+Sun Sep 19 06:06:07 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (check_rounding_mode): added for
+	  converting symbol to rounding mode number.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_mode, BigDecimal_round):
+	  support to specify rounding mode by symbol.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_mode, test_round):
+	  add tests for above changes.
+
+Sun Sep 19 05:14:35 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c: fix rounding algorithms for half-down
+	  and half-even.  This change is based on the patch created by Matthew
+	  Willson, the reporter of this bug.  [Bug #3803] [ruby-core:32136]
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests for above changes.
+
+Sat Sep 18 20:09:51 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_each_entry): Pathname#each_entry
+	  translated from pathname.rb.
+
+Fri Sep 17 23:44:07 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/xpath_parser.rb, test/rexml/test_xpath.rb:
+	  add missing method availability check. [ruby-core:32447]
+	  Reported by Wiebe Cazemier. Thanks!!!
+
+Fri Sep 17 23:23:26 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_sax.rb: don't use thread and sleep to avoid slow test.
+
+Fri Sep 17 23:10:44 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_core.rb: enable.
+
+Fri Sep 17 22:46:02 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/: untabify.
+
+Fri Sep 17 22:29:56 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/: fix fixture data path. All REXML tests are worked.
+
+Fri Sep 17 22:15:15 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_listener.rb: remove needless codes.
+
+Fri Sep 17 22:12:23 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/: import REXML tests from
+	  http://www.germane-software.com/repos/rexml/trunk/test/.
+	  Many tests are failed temporary. I'll fix them quickly. Sorry.
+
+Fri Sep 17 16:48:49 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/io/console/test_io_console.rb (TestIO_Console::helper):
+	  PTY.open is not guaranteed to work.  On my machine opening a
+	  pty is prohibited via process control group.  On those cases
+	  exceptions shall occur, and that doesn't mean our fault.
+	  Skip those tests on such situations.
+
+Fri Sep 17 08:30:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tracer.rb: count only non-internal libraries in stack trace,
+	  ignoring custom_require.  [ruby-core:31858]
+
+Fri Sep 17 02:18:41 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* tool/mkconfig.rb: Fix build with m4 1.4.15 generating duplicate
+	  lines in config.status.  According to nobu, the mswin32 port may
+	  depend on the piece of code in question, so the behavior is left
+	  unchanged on mswin32.
+
+Thu Sep 16 23:47:59 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_opendir): Pathname#opendir translated
+	  from pathname.rb.
+
+Thu Sep 16 21:40:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::GlobOption): merged RejectOption.
+
+	* test/runner.rb: utilize GlobOption.
+
+Thu Sep 16 21:31:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/driver.rb (RDoc::RI::Driver.setup_options)
+	  (RDoc::RI::Driver.fixup_options): split from process_args.
+	  libraries should not parse ARGV inside, since it's a task of
+	  applications, not libraries.
+
+Thu Sep 16 21:02:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/paths.rb (RDoc::RI::Paths.each): HOMEDIR can be nil
+	  if $HOME is unset.
+
+Thu Sep 16 14:50:42 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/ruby/test_file_exhaustive.rb (TestFileExhaustive::test_expand_path):
+	  ENV["HOME"] might not be set.  On those cases without it an
+	  exception raises here, which effectively disables later
+	  tests on this method.
+
+Thu Sep 16 08:30:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): fix rdoc.  pointed out by Tomoyuki
+	  Chikanaga at [ruby-core:32395], and a patch from Daniel
+	  Bovensiepen at [ruby-core:32403].
+
+Thu Sep 16 08:27:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/extconf.rb: use expanded sysconfdir with empty DESTDIR.
+	  [ruby-core:32394]
+
+Thu Sep 16 06:07:24 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_rmdir): Pathname#rmdir translated
+	  from pathname.rb.
+
+Thu Sep 16 00:36:25 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/extconf.rb: Remove Readline 6 check because
+	  Ruby's license is now GPLv3 compatible. [ruby-core:28736]
+
+Thu Sep 16 00:26:00 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* COPYING: change Ruby's License from a dual license with GPLv2
+	  to a dual license with 2-clause BSDL.
+	  [ruby-dev:42166] [ruby-core:31971]
+	  [ruby-dev:39167] [ruby-core:25272]
+
+	* COPYING.ja: ditto.
+
+	* BSDL: added. this is from The FreeBSD License.
+
+Wed Sep 15 21:07:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_mkdir): Pathname#mkdir translated
+	  from pathname.rb.
+
+Wed Sep 15 13:37:00 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/net/imap/test_imap.rb: "localhost" not guaranteed to
+	  resolve to "127.0.0.1".  On my machine it is "::1" instead.
+	  The problem is, you have to connect to the imaps server via
+	  the canonical name written in a server certificate, and that
+	  of the server.cert is "localhost".  So you have to listen to
+	  the address of what "localhost" resolves to.  I think this
+	  situation cannot be resolved in a handy manner because the
+	  test "test_imaps_post_connection_check" is actually
+	  expecting to connect to a server via an address other than
+	  the CN.  On my machine several assertions won't pass because
+	  the test cannot connect to the server.
+
+Wed Sep 15 09:12:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_puts): fix for wide char encoding strings.
+	  [ruby-dev:42212]
+
+Wed Sep 15 07:27:52 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_format_m): mentioned about Hash argument.  a patch
+	  from Daniel Bovensiepen at [ruby-core:32386].
+
+	* sprintf.c (get_hash): ditto, and fix typo.
+
+Wed Sep 15 07:22:20 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_entries): Pathname#entries translated
+	  from pathname.rb.
+
+Wed Sep 15 02:13:44 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/closure.c : Don't use FFI closure alloc on OpenBSD.
+	  Thanks Jeremy Evans! [ruby-core:32384]
+
+Tue Sep 14 20:17:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_s_getwd): Pathname.getwd and
+	  Pathname.pwd translated from pathname.rb.
+
+Tue Sep 14 05:13:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_s_glob): Pathname.glob translated
+	  from pathname.rb.
+
+Tue Sep 14 01:24:51 2010  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (ruby_getaddrinfo__aix): suppress a
+	  warning.
+
+Mon Sep 13 20:48:30 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_zero_p): Pathname#zero? translated
+	  from pathname.rb.
+
+Mon Sep 13 19:56:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/rubysocket.h (__DARWIN_ALIGNBYTES): workaround of a
+	  bug in system header of darwin 9.  [ruby-core:32341]
+
+Mon Sep 13 18:11:55 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_do): fix typo.  a patch from Peter Weldon
+	  at [ruby-core:32327].
+
+Mon Sep 13 10:12:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_strtod): reject Float('0x0.').
+	  [ruby-dev:42239] Bug #3820
+
+Mon Sep 13 09:23:58 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_is_prime): fix comparison
+	  with rb_scan_args. Before this fix, OpenSSL::BN#prime?
+	  is fully broken. [ruby-dev:42225]
+
+Mon Sep 13 06:45:24 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_writable_real_p):
+	  Pathname#writable_real? translated from pathname.rb.
+
+Sun Sep 12 21:21:50 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: [ruby-core:32096]  Thanks Colin Bartlett.
+
+Sun Sep 12 19:30:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_world_writable_p):
+	  Pathname#world_writable? translated from pathname.rb.
+
+Sun Sep 12 09:16:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_writable_p): Pathname#writable?
+	  translated from pathname.rb.
+
+Sun Sep 12 08:36:15 2010  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rlimit_resource_name2int): support more limits:
+	  RLIMIT_MSGQUEUE, RLIMIT_NICE, RLIMIT_RTPRIO, RLIMIT_RTTIME and
+	  RLIMIT_SIGPENDING.
+	  (Init_process): ditto.
+	  patch by Run Paint Run Run.  [ruby-core:32262]
+
+Sun Sep 12 04:27:13 2010  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rlimit_resource_name2int): use STRCASECMP to avoid
+	  ALLOCA_N.
+
+Sat Sep 11 16:47:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (ruby_setenv): raise if putenv and SetEnvironmentVariable
+	  failed, because of the restriction of the size on Windows.
+	  based on a patch from Peter Weldon at [ruby-core:32304].  fix:
+	  Bug#3812, [ruby-core:32250]
+
+Sat Sep 11 15:19:57 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/httpauth/digestauth.rb (WEBrick::Config::DigestAuth):
+	  Add documentation
+
+	* lib/webrick/config.rb (WEBrick::Config::DigestAuth): Add
+	  documentation
+
+Sat Sep 11 12:32:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_set_kcode, rb_get_kcode): removed
+	  zombie prototype declarations.  a patch from Eric Hodel
+	  at [ruby-core:32305].
+
+Sat Sep 11 06:53:12 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_symlink_p): Pathname#symlink?
+	  translated from pathname.rb.
+
+Fri Sep 10 23:03:43 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_sticky_p): Pathname#sticky? translated
+	  from pathname.rb.
+
+Fri Sep 10 19:11:13 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#continue): add
+	  method for generating HTTP/1.1 100 continue response if the client
+	  expects it, otherwise does nothing.  Patch by Brian Candler.
+	  ref #855.
+
+	* test/webrick/test_httprequest.rb: test added.
+
+Fri Sep 10 17:49:34 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/x509-internal.rb: removed unused local
+	  variable.
+
+	* test/openssl/*: less warnings while test running with -w.
+
+Fri Sep 10 16:49:20 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* class.c (rb_scan_args): Add support for optional keyword
+	  argument hash. [ruby-dev:42221] [ruby-dev:38048]
+
+	* README.EXT, README.EXT.ja: Update documentation accordingly.
+
+	* dir.c (dir_initialize): Make use of the new rb_scan_args()
+	  feature.
+
+	* io.c (rb_io_s_popen, rb_scan_open_args, rb_io_initialize)
+	  (rb_io_s_pipe, open_key_args, io_s_foreach, io_s_readlines)
+	  (rb_io_s_read, rb_io_set_encoding): Ditto.
+
+	* transcode.c (str_transcode, econv_args)
+	  (econv_primitive_convert): Ditto.
+
+	* ext/zlib/zlib.c (rb_gzreader_initialize): Ditto.
+
+Fri Sep 10 10:33:18 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c (rb_genrand_ulong_limited): renamed from
+	  rb_rand_internal and now this is public API.
+
+	* include/ruby/ruby.h (rb_genrand_ulong_limited): added.
+
+	* bignum.c (big_sparse_p): use rb_genrand_ulong_limited.
+
+Fri Sep 10 13:07:22 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/digest/lib/digest.rb: removed unused exception variable
+	  assignment to avoid a warning.
+
+Fri Sep 10 07:29:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_systmpdir): assume system default tmpdir
+	  safe.  [ruby-dev:42089]
+
+Fri Sep 10 07:03:23 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_size_p): Pathname#size? translated from
+	  pathname.rb.
+
+Fri Sep 10 02:15:29 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_peercred): support OpenBSD-current.
+	  patch by Jeremy Evans.  [ruby-core:32240]
+
+Thu Sep  9 23:25:53 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_backtrace_each): skip allocator frames which have no
+	  name.  [ruby-core:32231]
+
+Thu Sep  9 22:39:08 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_size): Pathname#size translated from
+	  pathname.rb.
+
+Thu Sep  9 22:34:48 2010  wanabe  <s.wanabe@gmail.com>
+
+	* compile.c (case_when_optimizable_literal): When float value can be
+	  treated as integer, add to table hash of case that way.
+	  based on a patch from Ikuo KOBORI. [ruby-dev:42038]
+
+	* insns.def (opt_case_dispatch): ditto.
+
+	* test/ruby/test_case.rb: add tests.
+
+Thu Sep  9 17:15:15 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/net/http/test_https.rb (test_identity_verify_failure): follows
+	  the SSL hostname check error message of openssl.
+
+Thu Sep  9 10:44:46 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_env.rb (test_aset): OpenBSD acts like NetBSD in
+	  that it ignores characters after = in ENV.
+	  patched by Jeremy Evans [ruby-core:32184]
+
+Thu Sep  9 09:02:01 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/rbinstall.rb (install?): gemspec filename should include
+	  its version. patched by Luis Lavena [ruby-core:32165]
+
+Wed Sep  8 22:46:31 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ssl_get_error): Thread context switch was
+	  blocked on Windows while blocking call for SSLSocket.  Need to
+	  convert errno for letting rb_io_wait_readable detect EWOULDBLOCK.
+	  Patch by arton. ref #3794.
+
+Wed Sep  8 20:56:57 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_setgid_p): Pathname#setgid? translated
+	  from pathname.rb.
+
+Wed Sep  8 06:25:41 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_setuid_p): Pathname#setuid? translated
+	  from pathname.rb.
+
+Tue Sep  7 21:03:35 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_readable_real_p):
+	  Pathname#readable_real?  translated from pathname.rb.
+
+Mon Sep  6 23:07:25 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_world_readable_p):
+	  Pathname#world_readable?  translated from pathname.rb.
+
+Mon Sep  6 11:03:13 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* Fixed wrong check of missing functions. Patch by Adrian Quark.
+	  ref #3400
+	  The patch contains following comment:
+	    This patch should avoid unnecessary incompatibility with future
+	    versions of Openssl.  Changes suggested by bmaher_at_amazon.com.
+
+Mon Sep  6 10:46:55 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* Fixed exception message for SSL post connection check failure. Patch
+	  by Paul Betteridge. ref [Bug #3704]
+
+Mon Sep  6 10:31:59 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_s_get_line_buffer):
+	  Readline.line_buffer should return locale string.
+	  [ruby-dev:42184] #3791
+
+Mon Sep  6 09:47:24 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_strtod): check there is at least 1 digit after
+	  "0x" before ".". [ruby-dev:42183] #3790
+
+Mon Sep  6 09:44:50 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_strtod): check integer overflow.
+	  [ruby-dev:42180] #3789
+
+Mon Sep  6 06:17:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_readable_p): Pathname#readable?
+	  translated from pathname.rb.
+
+Sun Sep  5 23:02:34 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_owned_p): Pathname#owned?
+	  translated from pathname.rb.
+
+Sat Sep  4 23:48:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_readlink): symlink target should be in
+	  filesystem encoding.
+
+Sat Sep  4 10:40:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (ruby_init_ext): export for golfers.
+
+	* vm_core.h (rb_iseq_eval, rb_iseq_compile_with_option): ditto.
+
+Sun May 23 17:29:41 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (capi): uses a timestamp file to get rid of
+	  generating twice.
+
+Fri Jun 18 01:33:21 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* enc/Makefile.in (realclean): has been missing. necessary
+	  for make realclean-enc.
+
+Fri Sep  3 23:51:26 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_socket_p): Pathname#socket?
+	  translated from pathname.rb.
+
+Fri Sep  3 06:40:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (chfunc): pass through exceptions.
+
+	* io.c (rb_io_bufwrite, rb_io_bufread): added.
+
+	* process.c (rb_fork_err): protect from exceptions.
+
+Fri Sep  3 06:16:07 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_pipe_p): Pathname#pipe?
+	  translated from pathname.rb.
+
+Fri Sep  3 06:14:40 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (chfunc): restore errno from SystemCallError and
+	  propagate proper exception to the parent.  [ruby-dev:41965]
+
+Thu Sep  2 22:10:38 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_file_p): Pathname#file?
+	  translated from pathname.rb.
+
+Thu Sep  2 09:12:02 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn): don't forget to free
+	  memory.
+
+Thu Sep  2 09:01:13 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): unicodize.
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn): convert arguments of
+	  CreateChild() from ACP to WideChar.
+
+Thu Sep  2 06:53:43 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_directory_p): Pathname#directory?
+	  translated from pathname.rb.
+
+Wed Sep  1 22:03:41 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_grpowned_p): Pathname#grpowned?
+	  translated from pathname.rb.
+
+Wed Sep  1 17:39:02 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.7.1 r5835.
+	* test/minitest/*.rb: ditto.
+
+Wed Sep  1 16:50:42 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_setup_table): optimized. don't create hash objects
+	  when given pattern is ASCII only.
+
+	* string.c (tr_find): ditto.
+
+Wed Sep  1 14:35:29 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_rotate_m): fix typo of rdoc.
+	  patched by Andrei Kulakov [ruby-core:31975]
+
+Wed Sep  1 14:33:36 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enum.c (enum_zip): fix typo of rdoc.
+	  patched by Andrei Kulakov [ruby-core:31974]
+
+Wed Sep  1 12:56:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (ruby_suppress_tracing): restore the state and invoke
+	  the func with normal state.  a patch from Satoshi Shiba <shiba
+	  AT rvm.jp> at [ruby-dev:42162].  [ruby-core:31783]
+
+Tue Aug 31 21:10:23 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_exist_p): Pathname#exist? translated
+	  from pathname.rb.
+
+Tue Aug 31 17:32:34 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/stubs.c: fix [Bug #3771] "VC++ can't make ext/tk with enabling
+	  stubs". Thanks, Akio Tajima [ruby-dev:42159].
+
+Tue Aug 31 03:42:14 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_setup_table): fix bug in r29146.
+	  Initialize table even if cflag is 0; tr_find see whether
+	  del is empty or not.
+
+	* string.c (tr_find): nodel can't be NULL; if NULL, it means
+	  it is not specified.
+
+Mon Aug 30 21:29:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_executable_real_p):
+	  Pathname#executable_real? translated from pathname.rb.
+
+Mon Aug 30 15:00:13 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_setup_table): initialize negating table when
+	  negating string is given. [ruby-core:31851]
+
+	* string.c (tr_find): add a sentence for the time when
+	  target characters include negating one.
+
+	* string.c (rb_str_count): move definition.
+
+Mon Aug 30 07:32:41 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_executable_p): Pathname#executable?
+	  translated from pathname.rb.
+
+Sun Aug 29 23:54:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parser/ruby.rb (RDoc#parse_call_parameters): don't
+	  include assignment.  [Bug #3759], [ruby-dev:42154]
+
+	* lib/rdoc/parser/ruby.rb (RDoc#parse_class): ignore non-constant
+	  name singleton class.  [Bug #3759], [ruby-dev:42154]
+
+Sun Aug 29 23:25:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_get_path_check): clarify error message for
+	  ASCII-incompatible path name.
+
+Sun Aug 29 16:02:54 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (node_name.inc): remove command option -n and give
+	  file as stdin, because IronRuby 1.1 still doesn't support it.
+	  So now we can use ir.exe as BASERUBY.
+
+	* tool/node_name.rb: read stdin with while gets.
+
+Sun Aug 29 13:22:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_thread_method_id_and_class): curried proc has no
+	  method.  [ruby-core:31871]
+
+Sun Aug 29 12:51:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_provide_feature): clarify error message for frozen
+	  $LOADED_FEATURES.  based on a patch from Run Paint Run Run at
+	  [ruby-core:31913].
+
+Sun Aug 29 12:19:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_failed): should honor encoding.  [ruby-core:31915]
+
+Sun Aug 29 09:35:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (clean): exclude *.inc. [ruby-dev:41931]
+
+	* common.mk (distclean): include *.inc.
+
+	* common.mk (help): change description about clean and distclean.
+
+Sat Aug 29 06:34:52 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_chardev_p): Pathname#chardev?
+	  translated from pathname.rb.
+
+Sat Aug 28 17:39:33 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_save_exception_mode,
+	  BigDecimal_save_rounding_mode, BigDecimal_save_limit): added.
+
+	* test/bigdecimal/test_bigdecimal.rb: added tests for the above
+	  features.
+
+Sat Aug 28 08:11:05 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_blockdev_p): Pathname#blockdev?
+	  translated from pathname.rb.
+
+Fri Aug 27 16:20:01 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* string.c (rb_str_prepend): new method by Shota Fukumori (sora_h)
+	  [Feature #3765]
+
+Fri Aug 27 15:24:20 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* math.c (math_atan2): you should know that M_PI is not the feature
+	  of C90.
+	  fixed build failure caused by r29115.
+
+Fri Aug 27 15:26:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (null_device): move from io.c.
+
+Fri Aug 27 12:47:44 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* math.c (math_atan2): change the behavior when x and y are zero.
+	  [ruby-dev:42090] [Bug #3736] [ruby-dev:42116]
+
+	* test/ruby/test_math.rb (test_atan2): add tests for the above
+	  changes.
+
+Fri Aug 27 12:26:23 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* object.c (rb_obj_class): remove mention of obsolete method.
+	  a patch from Run Paint Run Run at [ruby-core:31842].
+
+Fri Aug 27 12:25:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (null_device): the name of null device.  [ruby-dev:41791]
+
+Fri Aug 27 07:57:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_shuffle_bang): bail out from modification during
+	  shuffle.
+
+	* array.c (rb_ary_sample): ditto.
+
+Fri Aug 27 05:11:51 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_sysopen): Pathname#sysopen translated
+	  from pathname.rb.
+
+Thu Aug 26 22:53:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_shuffle): rdoc fix.  argument name was missing.
+	  a patch from Run Paint Run Run at [ruby-core:31848].
+
+Thu Aug 26 21:49:46 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_readlines): Pathname#readlines
+	  translated from pathname.rb.
+
+Thu Aug 26 10:37:00 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regint.h (OnigStackIndex): the type should be intptr_t.
+	  Original Oniguruma assumes the size of long and that of void *
+	  are equal, but it's not true on LLP64 platform: mswin64.
+	  originally patched by shintaro kuwamoto [ruby-dev:42133]
+
+Thu Aug 26 10:38:11 2010  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* test/dl/test_base.rb: AIX does not have dynamically loadable lib[cm].
+
+	* test/fiddle/helper.rb: AIX does not have dynamically loadable lib[cm].
+
+Thu Aug 26 09:49:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_shuffle_bang): check number of argument.
+
+Tue Aug 26 09:11:40 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (Init_bigdecimal,
+	  rmpd_set_thread_local_exception_mode, VpGetException,
+	  VpSetException): thread-local exception mode.
+
+	* ext/bigdecimal/bigdecimal.c (Init_bigdecimal,
+	  rmpd_set_thread_local_precision_limit, VpGetPrecLimit,
+	  VpSetPrecLimit): thread-local precision limit.
+
+	* ext/bigdecimal/bigdecimal.c (Init_bigdecimal,
+	  rmpd_set_thread_local_rounding_mode, VpGetRoundMode,
+	  VpSetRoundMode, VpException, VpInternalRound):
+	  thread-local rounding mode.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_mode, BigDecimal_round,
+	  VpIsRoundMode, VpGetRoundMode, VpSetRoundMode, VpActiveRound,
+	  VpMidRound, VpLeftRound), ext/bigdecimal/bigdecimal.h:
+	  use unsigned short for rounding mode.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_mode): add test for
+	  setting rounding mode.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_thread_local_mode):
+	  add test for setting mode thread-locally.
+
+Thu Aug 26 07:29:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_{shuffle_bang,sample}): use Random class object.
+
+	* random.c (try_get_rnd): use default_rand for Random as same as
+	  singleton methods.
+
+	* random.c (rb_random_real): check the range of result.
+
+Wed Aug 25 22:11:11 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_binread): Pathname#binread translated
+	  from pathname.rb.
+
+Wed Aug 25 03:42:43 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): workaround for VC9 for x64.
+	  reported by kuwamoto shintaro in [ruby-dev:42125].
+
+Tue Aug 24 23:28:50 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* .gitignore: updated.
+
+Tue Aug 24 22:07:28 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_read): Pathname#read translated from
+	  pathname.rb.
+
+Tue Aug 24 10:11:04 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: read API version from include/ruby/version.h.
+
+	* {bcc,win}32/setup.mak (-version-): ditto.
+
+	* version.h (RUBY_LIB_VERSION): use API version numbers.
+
+Tue Aug 24 07:07:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_shuffle_bang, rb_ary_sample): add optional
+	  argument random.  [ruby-dev:41923] [EXPERIMENTAL]
+
+	* random.c (rb_random_{int32,real,bytes}): fallback to normal
+	  method invocation.
+
+Tue Aug 24 06:08:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/version.h (RUBY_API_VERSION_*): renamed and moved
+	  from version.h.  [ruby-dev:42103]
+
+Tue Aug 24 05:58:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ChangeLog: flushed.  [ruby-dev:42050]
+
+Tue Aug 24 01:14:58 2010  Kenta Murata <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): to_f must underflow
+	  when the exponent is less than DBL_MIN_10_EXP - BASE_FIG.
+
+	* test/bigdecimal/test_bigdecimal.rb (test_to_f): added test for the above changes.
+
+Mon Aug 23 23:14:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_each_line): Pathname#each_line
+	  translated from pathname.rb.
+
+Mon Aug 23 22:30:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (version.o): depends on both of version.h and
+	  include/ruby/version.h.   [ruby-dev:42063]
+
+Mon Aug 23 13:09:27 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* backport r29071 from ruby_1_8;
+
+	  * ext/openssl/ossl_asn1.c (obj_to_asn1bool): fixed ASN1::Boolean
+	    encoding issue for OpenSSL 1.0.0 compatibility.
+	    ASN1::Boolean.new(false).to_der wrongly generated "\1\1\377" which
+	    means 'true'.
+
+	    ASN1_TYPE_set of OpenSSL <= 0.9.8 treats value 0x100 as 'false'
+	    but OpenSSL >= 1.0.0 treats it as 'true'.  ruby-ossl was using
+	    0x100 for 'false' for backward compatibility.  Just use 0x0 for
+	    the case OpenSSL >= OpenSSL 0.9.7.
+
+	  * test/openssl/test_asn1.rb: test added.
+
+Mon Aug 23 12:28:58 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_search): fix: 4th argument should be regexp
+	  object. patched by shintaro kuwamoto [ruby-dev:41667] #3459
+
+Mon Aug 23 02:23:05 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_strtod): make sure to have digit-sequence after 'p'
+	  for hexadecimal-floating-constant. [ruby-dev:42105]
+
+Mon Aug 23 00:23:07 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: [ruby-core:31695]
+
+Sun Aug 22 09:08:02 2010  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (UINT2NUM): fix ifdef condition for LLP64.
+	  reported by Daniel Gutmanas.  [ruby-core:31778]
+
+Sun Aug 22 06:46:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (RAND_UPTO): macro for random number.
+
+Sun Aug 22 05:55:01 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: fix typo.  a patch from Eric Wong at
+	  [ruby-core:31810].
+
+Sat Aug 21 20:09:43 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm.c: add missing prototype of rb_iseq_clone for r29063.
+
+Sat Aug 21 18:54:56 2010  wanabe  <s.wanabe@gmail.com>
+
+	* vm.c (vm_define_method): copy iseq to avoid overwriting iseq->klass.
+	  #2502, #3136. see #2420.
+
+Sat Aug 21 17:36:42 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_split): Pathname#split translated from
+	  pathname.rb.
+
+Fri Aug 20 10:40:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_expand_path): Pathname#expand_path
+	  translated from pathname.rb.
+
+Thu Aug 19 22:44:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (usage): add usage.
+
+	* tool/make-snapshot (package): accept revision as BRANCH@REV.
+
+Thu Aug 19 19:53:26 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_timestamp): fix a constant type.
+
+	* ext/socket/raddrinfo.c (get_afamily): suppress a warning.
+
+Thu Aug 19 19:28:47 2010  Tanaka Akira  <akr@fsij.org>
+
+	* thread_pthread.c (native_sleep): suppress a warning for platforms
+	  which time_t is int.
+
+Thu Aug 19 18:15:50 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: Add a tool to merge commits between branches.
+	  developed by URABE Shyouhei
+
+Thu Aug 19 17:19:09 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* backport r28621 and r28632 from ruby_1_8;
+
+	  * ext/openssl/ossl_config.c, ext/openssl/lib/openssl/config.rb,
+	    ext/openssl/lib/openssl.rb: reimplement OpenSSL::Config in Ruby.
+	    Now it should work on windows.
+
+	  * test/openssl/test_config.rb: added tests for OpenSSL::Config#dup.
+
+	  * test/openssl/test_config.rb: added tests for Config#clone.
+
+Thu Aug 19 12:04:39 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* array.c (rb_ary_permutation, rb_ary_repeated_permutation,
+	  rb_ary_repeated_combination, rb_ary_product):
+	  use ary_make_shared_copy instead of ary_make_substitution.
+	  [ruby-dev:42067] [Bug #3708]
+
+	* test/ruby/test_array.rb (test_product, test_repeated_permutation,
+	  test_repeated_combination): append assertions against [Bug #3708].
+
+Thu Aug 19 11:11:24 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enum.c (enum_inject): fix typo of rdoc.
+	  patched by Andrei Kulakov [ruby-core:31768]
+
+Thu Aug 19 10:06:08 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_extname): Pathname#extname translated
+	  from pathname.rb.
+
+Thu Aug 19 06:03:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_push_frame): initialize block_iseq.  a patch
+	  from Peter Weldon at [ruby-core:31757].
+
+Wed Aug 18 19:08:17 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* array.c (ary_make_shared): revert r29037, it occurs SEGV some
+	  situations.
+
+Wed Aug 18 18:36:32 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* array.c (ary_make_shared),
+	  test/ruby/test_array.rb (test_permutation):
+	  always return the original array when it is shared.
+	  [ruby-dev:42067] [Bug #3708]
+
+Wed Aug 18 01:37:49 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c: revert r26701; it introduces Bug #3681.
+	  [ruby-core:31677]
+
+Tue Aug 17 20:05:29 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb (test_threaded_flush): test "hi" is not output
+	  twice.
+
+Tue Aug 17 15:13:28 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/date.rb: Re-revert the part of r28950.
+	  It reverted the part of r28619.
+
+Tue Aug 17 13:00:07 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_dirname): Pathname#dirname translated
+	  from pathname.rb.
+
+Tue Aug 17 07:50:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/extconf.rb: check functions more.
+	  [ruby-core:31722]
+
+Tue Aug 17 07:42:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_make_independent_expand): set capacity properly. a
+	  patch from Peter Weldon at [ruby-core:31734].  [ruby-core:31653]
+
+Tue Aug 17 07:38:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gem_prelude.rb, lib/rubygems.rb (Gem.suffixes): include empty
+	  suffix.  [ruby-core:31730]
+
+Mon Aug 16 22:16:12 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_basename): unused variable removed.
+
+Mon Aug 16 21:39:33 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_basename): Pathname#basename translated
+	  from pathname.rb.
+
+Mon Aug 16 20:26:13 2010  Narihiro Nakamura  <narihiro@netlab.jp>
+
+	* gc.c (gc_profile_result): Index begins with 1.
+
+Mon Aug 16 20:21:17 2010  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_profile_result): use size_t. based on patches from
+	Tomoaki NISHIYAMA <tomoakin@kenroku.kanazawa-u.ac.jp> at
+	[ruby-dev:42042].
+
+Mon Aug 16 19:23:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_dtoa, ruby_hdtoa): use same representations for
+	  Infinity and NaN.  a part of a patch from Peter Weldon at
+	  [ruby-core:31725].
+
+	* util.c (ruby_hdtoa): fixed buffer overrun.  based on a patch
+	  from Peter Weldon at [ruby-core:31725].
+
+Mon Aug 16 17:44:55 2010  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (capi): use -b option for doxygen.  It disables stdout
+	  buffering and prevents wrong reordering between stdout and stderr
+	  even when the output of "make" is redirected.
+
+Sun Aug 15 19:59:58 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/webrick/httpresponse.rb (WEBrick::HTTPResponse#set_error):
+	  Fix for possible cross-site scripting (CVE-2010-0541).
+	  Found by Apple, reported by Hideki Yamane.
+	  Patch by Hirokazu Nishio <nishio.hirokazu AT gmail.com>.
+
+Mon Aug 16 12:29:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub (test-rubyspec-precheck): split
+	  from test-rubyspec.
+
+	* common.mk (test-rubyspec): moved from Makefile.in.
+
+Mon Aug 16 10:00:45 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.h (llabs): never never never never never
+	  use "long long".
+
+	* ext/bigdecimal.bigdecimal.c (BigDecimal_to_i): get rid of a warning.
+
+Sun Aug 15 18:14:20 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_dump, BigDecimal_inspect,
+	  VPrint, VpToString, VpVarCheck): use PRIuSIZE, PRIdSIZE, PRIdVALUE,
+	  and PRIxVALUE.
+
+Sun Aug 15 16:14:31 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (Logging.postpone): close @log only when it's available.
+
+Sun Aug 15 13:30:35 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.h (VpVtoD): fix a type of e.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_dump, BigDecimal_inspect):
+	  fix format for size_t.
+
+Sun Aug 15 00:22:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/extconf.rb, ext/bigdecimal/bigdecimal.h
+	  (labs, llabs): support environments missing labs and llabs.
+
+	* ext/bigdecimal/bigdecimal.h (vabs): added.
+
+	* ext/bigdecimal/extconf.rb, ext/bigdecimal/bigdecimal.h,
+	  ext/bigdecimal/bigdecimal.c, test/bigdecimal/test_bigdecimal.rb
+	  (TestBigDecimal#test_new): replace U_LONG, S_LONG, S_INT, and U_INT
+	  with appropriate standard or ruby-provided types.
+
+Sun Aug 15 00:15:45 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_utime): Pathname#utime translated
+	  from pathname.rb.
+
+Sat Aug 14 21:04:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_gc_mark_threads): deprecated.
+
+Sat Aug 14 19:12:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/{coverage,objspace}/extconf.rb ($INCFLAGS): explicitly add
+	  topdir and top_srcdir.  [ruby-dev:42031]
+
+Sat Aug 14 17:48:41 2010  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* test/objspace/test_objspace.rb: added test for objspace.
+
+	* ext/objspace/objspace.c: considers T_ZOMBIE by lazy sweep GC.
+
+	* gc.c: considers that dsize was 0. [ruby-dev:42022]
+
+Sat Aug 14 15:33:02 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, include/ruby/defines.h (RUBY_FUNC_EXPORTED): macro
+	  to declare exported function.
+
+	* array.c (rb_ary_memsize), string.c (rb_str_memsize),
+	  variable.c (rb_objspace_data_type_memsize): used in
+	  objspace.  [ruby-dev:42022]
+
+	* io.c (rb_io_memsize): constified.
+
+	* regcomp.c (onig_memsize): constified.
+
+	* gc.h (rb_objspace_each_objects): used in objspace.
+
+	* vm_core.h (rb_{get,set,reset}_coverages): used in coverage.
+
+	* include/ruby/intern.h (rb_time_interval): used in io/wait.
+
+	* include/ruby/intern.h (rb_make_backtrace, rb_make_exception):
+	  used in ripper.
+
+	* node.h (rb_parser_{malloc,realloc,calloc,free}): ditto.
+
+Fri Aug 13 22:48:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_truncate): Pathname#truncate translated
+	  from pathname.rb.
+
+Fri Aug 13 16:11:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (MiniTest::Unit.new): extend before initialize.
+	  [ruby-dev:41970]
+
+	* lib/test/unit.rb (MiniTest::Unit.autorun): use Test::Unit::Mini.
+
+Thu Aug 12 23:39:02 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_make_symlink): Pathname#make_symlink
+	  translated from pathname.rb.
+
+Thu Aug 12 21:18:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_lstat): Pathname#lstat translated
+	  from pathname.rb.
+
+Thu Aug 12 05:58:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_char_to_option_kcode): used in
+	  ripper.
+
+	* node.h (rb_reserved_word): ditto.
+
+Thu Aug 12 00:19:53 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_rubyoptions.rb (test_script_from_stdin): disable
+	  echo.  [ruby-dev:41966]
+
+Wed Aug 11 23:28:22 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* lib/cmath.rb (CMath.sqrt): use floating-point value.
+	  [ruby-core:31672] [Bug #3678]
+
+	* test/test_cmath.rb: added for testing lib/cmath.rb.
+
+Wed Aug 11 20:57:25 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: merge from branches/ruby_1_9_2, and move io/console.
+	  [ruby-dev:41924]
+
+Wed Aug 11 11:01:15 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c: don't use // comment.
+
+Wed Aug 11 06:18:20 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_stat): Pathname#stat translated
+	  from pathname.rb.
+
+Tue Aug 10 21:18:04 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_to_[ifr]): don't allow complex with in-exact
+	  imaginary zero to be converted.
+
+Tue Aug 10 20:56:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/optparse.rb: suppress a warning.
+
+	* test/test_mutex_m.rb: ditto.
+
+Tue Aug 10 20:56:13 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (Date::Format::Bag): use Struct.
+	  as an experimental.
+
+Tue Aug 10 20:45:35 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_rename): Pathname#rename translated
+	  from pathname.rb.
+
+Tue Aug 10 20:16:19 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/resolv.rb (get_info): get rid of a warning.
+
+Tue Aug 10 19:56:13 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: [ruby-core:31658]
+
+Tue Aug 10 17:24:56 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* README.EXT.ja: rb_tainted_str_new_cstr() is an alias of
+	  rb_tainted_str_new2().
+
+Tue Aug 10 15:24:30 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c, ext/bigdecimal/bigdecimal.h,
+	  ext/bigdecimal/extconf.rb: BASE and BASE_FIG are defined
+	  based on the size of BDIGIT.
+
+	* ext/bigdecimal/bigdecimal.c, ext/bigdecimal/bigdecimal.h:
+	  use BDIGIT for Real#frac.
+
+Tue Aug 10 15:19:45 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.h, ext/bigdecimal/bigdecimal.c:
+	  add extern declaration of rb_cBigDecimal.
+
+Tue Aug 10 14:09:31 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest::parse_uri):
+	  rollup leading slashes. [ruby-core:31657]
+	  patched by Jamison Wilde
+	  NOTE: //authority/path is valid relative URI both RFC2396 and
+	  RFC3986. So when give a relative URI-like string to URI lib,
+	  users must care leading slashes.
+
+Tue Aug 10 11:26:33 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/un.rb (httpd): SIGQUIT and SIGHUP are not guaranteed to exist.
+
+Tue Aug 10 06:26:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_readlink): Pathname#readlink translated
+	  from pathname.rb.
+
+Mon Aug  9 22:15:19 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_open): Pathname#open translated from
+	  pathname.rb.
+
+Mon Aug  9 22:11:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_signal.rb (TestSignal#test_exit_action): use
+	  Process.spawn instead of Process.fork.
+
+Mon Aug  9 15:59:02 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/source_index.rb: rename unused variable.
+
+	* lib/rubygems/specification.rb: rename unused variable.
+
+	* lib/rubygems/specification.rb: remove unused variable.
+
+Mon Aug  9 14:10:06 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Fix type of mimeout_state.buf.
+	  [nkf-bug:20079]
+
+Mon Aug  9 06:52:59 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (sys_warning_1): show error message.
+
+Mon Aug  9 06:50:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_require.rb (test_require_too_long_filename):
+	  clear -w option.
+
+Mon Aug  9 00:18:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb ($ignore_error): initialize after $nmake.
+
+Sun Aug  8 19:16:26 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_make_link): Pathname#make_link
+	  translated from pathname.rb.
+
+Sun Aug  8 16:42:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems.rb (Gem.find_files): reverted to use globbing.
+
+	* lib/rubygems.rb (Gem.find_files): fixed search order same as
+	  default behavior.
+
+	* gem_prelude.rb, lib/rubygems.rb (Gem.suffixes): return truly
+	  require-able suffixes only.
+
+Sun Aug  8 13:20:57 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_ftype): Pathname#ftype translated
+	  from pathname.rb.
+
+Sun Aug  8 12:41:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_escape): no similar messages twice.
+	  [ruby-core:31048]
+
+Sun Aug  8 10:07:45 2010  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c (rb_strftime_with_timespec): suppress warnings.
+
+Sun Aug  8 08:58:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RUBY_H_INCLUDES): add subst.h.
+
+Sun Aug  8 08:55:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ruby_snprintf): printf attribute.
+
+Sun Aug  8 07:47:29 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rbconfig/obsolete.rb: show the location which use Config.
+
+Sun Aug  8 07:29:55 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_fnmatch): Pathname#fnmatch and
+	  Pathname#fnmatch? translated from pathname.rb.
+
+Sun Aug  8 07:18:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/subst.h (snprintf): redefinition moved from ruby.h.
+	  (vsnprintf): ditto.
+
+Sun Aug  8 00:41:42 2010  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/subst.h: new file for substitute standard functions.
+
+	* include/ruby/missing.h: don't substitute "close", etc. here.
+
+	* include/ruby/ruby.h: include ruby/subst.h at last.
+	  This prevents substituting "close" in unistd.h.
+
+Sat Aug  7 19:51:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (EnvUtil.invoke_ruby): merge stdout and stderr
+	  if capture_stderr is :merge_to_stdout.
+	  (assert_normal_exit): print abnormal output properly.
+
+Sat Aug  7 19:04:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* missing/close.c: undef the macros "getpeername", "getsockname" and
+	  "shutdown" to prevent infinite recursion.
+
+Sat Aug  7 18:37:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* missing/close.c: undef the macro "close" to prevent infinite
+	  recursion.
+
+Sat Aug  7 18:20:41 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_lchown): Pathname#lchown translated
+	  from pathname.rb.
+
+Sat Aug  7 17:49:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/missing.h: moved BROKEN_CLOSE replacements from
+	  include/ruby/ruby.h.
+
+Sat Aug  7 17:40:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb, lib/rbconfig/obsolete.rb: warn obsolete and
+	  deprecated Config.
+
+Sat Aug  7 17:18:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): follow the change of
+	  install-rb-default line format at r28850, which causes second
+	  run to create dummy makefiles.
+
+Sat Aug  7 14:08:44 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/mkmf.rb: change instance variable to global.
+
+Sat Aug  7 13:59:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pathname/pathname.c (path_sub_ext): don't clobber shared string.
+	  [ruby-core:31640]
+
+Sat Aug  7 13:33:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call0): fix for VM_METHOD_TYPE_NOTIMPLEMENTED.
+	  [ruby-dev:41953]
+
+Sat Aug  7 09:47:06 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: fix checking version of GCC.
+
+Sat Aug  7 09:31:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (Init_pathname): fix number of arguments for
+	  Pathname#chown.
+
+Sat Aug  7 09:28:12 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_normal_exit): use assert.  fix
+	  faildesc.
+
+	* test/pathname/test_pathname.rb (test_lchmod): test Pathname#lchmod
+	  using assert_normal_exit.
+
+Sat Aug  7 08:26:26 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_chown): Pathname#chown translated
+	  from pathname.rb.
+
+Sat Aug  7 02:09:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_lchmod): Pathname#lchmod translated
+	  from pathname.rb.
+
+Fri Aug  6 21:30:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_chmod): Pathname#chmod translated from
+	  pathname.rb.
+
+Fri Aug  6 20:13:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* rational.c (nurat_div): divided by infinity should be zero.
+
+	* rational.c (nurat_div): divided by float zero should be
+	  infinity.  [ruby-core:31626]
+
+Fri Aug  6 18:59:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (realpath_rec): rb_str_modify depends on the length, so
+	  resize instead.
+
+Fri Aug  6 18:56:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (clean-ext): fixed directories to be removed.
+
+Fri Aug  6 17:42:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/depend: add space at the beginning of @ignore_error.
+
+	* lib/mkmf.rb: ditto.
+
+Fri Aug  6 17:38:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (clean): add  prelude.c, config.log, and rbconfig.rb.
+
+	* common.mk (distclean): remove encdb.h, config.log, and rbconfig.rb.
+
+Fri Aug  6 17:35:24 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in: ignore error from RMDIRS.
+
+	* enc/depend: ditto.
+
+	* lib/mkmf.rb: ditto.
+
+Fri Aug  6 05:53:32 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_mtime): Pathname#mtime translated from
+	  pathname.rb.
+
+Thu Aug  5 22:09:30 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_ctime): Pathname#ctime translated from
+	  pathname.rb.
+
+Thu Aug  5 21:20:31 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (NODE_ARGSCAT, NODE_ARGSPUSH): drop unused ARGSCAT
+	  results when popped is true.  [ruby-dev:41933], [Bug #3658]
+	  This is retry of r28870 and r28873 which were reverted.
+
+Thu Aug  5 20:13:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja (String functions): mention
+	  rb_str_resize and rb_str_set_len.
+
+Thu Aug  5 19:59:55 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_set_len): bail out when buffer overflowed
+	  probably.
+
+Thu Aug  5 18:36:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (nucomp_marshal_load): should check the argument.
+	  [ruby-core:31622]
+
+	* rational.c (nurat_marshal_load): ditto
+
+	* marshal.c (w_float): should not append a dot if no fractal part
+	  exists.  [ruby-dev:41936]
+
+Thu Aug  5 17:11:43 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (void_expr_gen): add 'possibly' to warning message.
+	  [ruby-core:31611]
+
+Thu Aug  5 16:57:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_make_pattern): fold continuous PLAINs to get rid of
+	  snail at too deep path.  [ruby-dev:41871]
+
+Thu Aug  5 16:42:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_set_len): should fail to modify shared string.
+
+Thu Aug  5 14:12:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): ensure sopath to be modifiable
+
+Thu Aug  5 12:39:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_make_independent_expand): fix buffer overflow
+	  while shrinking.
+
+Thu Aug  5 06:42:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (realpath_rec): call rb_str_modify before rb_str_set_len.
+
+Wed Aug  4 19:37:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_set_len): rb_str_modify cannot work before the
+	  length is set, which is a precondition for rb_str_modify.
+
+Wed Aug  4 15:47:21 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): no need to create the directory
+	  at each library file. this change makes making ext faster.
+
+Wed Aug  4 06:25:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_atime): Pathname#atime translated from
+	  pathname.rb.
+
+Tue Aug  3 21:16:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (Init_Random): add Random::DEFAULT.
+
+Tue Aug  3 21:01:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_check_hash_type): added.
+
+Tue Aug  3 20:30:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_eq): never equal to infinity.
+	  [ruby-core:31603]
+
+	* rational.c (nurat_div): hack for integral float divisor.
+
+Tue Aug  3 14:42:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/extmk.rb: remove purelib, fixes a bug in r28440, r28441.
+
+	* spec/default.mspec: ditto.
+
+	* template/Doxyfile.tmpl: ditto.
+
+Tue Aug  3 14:19:42 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/ri/driver.rb (setup_pager): no need to check the existence
+	  of pagers. the following code checks whether they are executable or
+	  not.
+
+Tue Aug  3 12:03:57 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* tool/runruby.rb: no purelib.
+
+Tue Aug  3 07:52:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/purelib.rb: no longer used since the default load path of
+	  miniruby is empty.
+
+Tue Aug  3 00:53:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def, vm_core.h: fixed typo.  reported by Rocky Bernstein
+	  <rockyb AT rubyforge.org> at [ruby-core:31596].
+
+Mon Aug  2 21:51:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): rb_str_set_len modifies ptr
+	  since r26303.
+
+Mon Aug  2 21:16:03 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: NONAMELESSUNION defined only if gcc
+	  version is older than 3.4.4. [ruby-core:31567] [Bug #3637]
+	* ext/win32ole/extconf.rb: ditto.
+
+Mon Aug  2 16:34:32 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (native_thread_join): need to wait thread, of course.
+	  [ruby-dev:41911]
+
+Mon Aug  2 12:22:22 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (file_expand_path): wrong condition. [ruby-core:31591]
+
+Mon Aug  2 05:34:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_realdirpath): Pathname#realdirpath
+	  translated from pathname.rb.
+
+Sun Aug  1 23:04:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb: describe superclasses explicitly.
+
+Sun Aug  1 23:04:35 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_modify_expand, rb_str_resize): get rid of
+	  repeating malloc and realloc.
+
+Sun Aug  1 22:16:30 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_hash): use rb_memhash and
+	  take care of negative finite numbers properly.
+	  [ruby-core:31582][Bug #3640]
+
+Sun Aug  1 20:57:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_realpath): Pathname#realpath translated
+	  from pathname.rb.
+
+Sun Aug  1 17:27:32 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub: needs stddef.h for size_t.  a patch from
+	  Akio Tajima at [ruby-dev:41905].
+
+Sun Aug  1 14:59:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_sub_ext): Pathname#sub_ext translated
+	  from pathname.rb.
+
+Sun Aug  1 10:23:48 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/irb/init.rb (IRB.parse_opts): set VERBOSE to true when debug
+	  switch called in irb.  a patch from Andrew Grimm in
+	  [ruby-core:31558].
+
+Sun Aug  1 09:35:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (big_op): comparison of bignum and infinity has returned 1
+	  or -1, but it must return boolean.
+
+Sun Aug  1 09:44:25 2010  Tanaka Akira  <akr@fsij.org>
+
+	* class.c (rb_include_module): don't clear the method cache if the
+	  included module has no method.
+	  reported by Eric Wong.  [ruby-core:31559]
+
+Sun Aug  1 09:21:32 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_sub): Pathname#sub translated
+	  from pathname.rb.
+
+Sun Aug  1 09:12:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h: latest x86_64 mingw defines stati64.
+	  [ruby-core:27516]
+
+Sun Aug  1 08:51:52 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: fix typo. [ruby-core:31564][Bug #3636]
+
+Sat Jul 31 23:15:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_inspect): Pathname#inspect translated
+	  from pathname.rb.
+
+Sat Jul 31 15:50:03 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (Init_pathname): The alias,
+	  Pathname#to_path, is translated from pathname.rb.
+
+Sat Jul 31 09:12:36 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_to_s): Pathname#to_s translated
+	  from pathname.rb.
+
+Sat Jul 31 08:20:07 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/webrick/ssl.rb (WEBrick::Utils.create_self_signed_cert):
+	  merged r28784 from ruby_1_8: wrongly created dummy SSL certificate
+	  with version == 3 (no such version) and serial == 0 (must be >0).
+
+Fri Jul 30 21:43:55 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_hash): Pathname#hash translated
+	  from pathname.rb.
+
+Fri Jul 30 12:42:02 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/imap.rb (example): use IO#noecho to read password if
+	  possible, and defer until needed.  [ruby-dev:41889]
+
+	* lib/net/imap.rb (example): support starttls option.
+	  [ruby-dev:41888]
+
+Fri Jul 30 08:51:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): home directory must be absolute.
+	  [ruby-core:31537]
+
+Fri Jul 30 08:33:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): should check if could find user.
+	  [ruby-core:31538]
+
+Fri Jul 30 07:59:53 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_add_suffix): fixed a bug returning uninitialized
+	  value.
+
+Fri Jul 30 07:48:04 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: use TK_XINCLUDES on tkConfig.sh when not empty,
+	  even if MacOS X Aqua version [ruby-dev:41883].
+
+Thu Jul 29 22:28:35 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_cmp): Pathname#<=> translated
+	  from pathname.rb.
+
+Thu Jul 29 06:51:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (EXT_SRCS): add ext/json/parser/parser.c.
+
+Thu Jul 29 05:52:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_basename): pass baselen to rmext.
+
+Thu Jul 29 02:38:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: not [freebsd] but [freebsd*] for AS_CASE.
+
+Wed Jul 28 22:23:59 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_eq): Pathname#{==,===,eql?} translated
+	  from pathname.rb.
+
+Wed Jul 28 19:37:33 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): VC6 or later have stddef.h.
+
+	* include/ruby/missing.h: need to include stddef.h for size_t.
+
+	* include/ruby/missing.h: shouldn't declare as dllimport when building
+	  ruby itself (for Windows).
+
+Wed Jul 28 17:11:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/*.c: include missing.h
+
+Wed Jul 28 09:38:03 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* cont.c: typos.
+
+Wed Jul 28 09:24:43 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c: (RB_PAGE_SIZE): renamed from PAGE_SIZE. [ruby-dev:41870]
+
+	* cont.c: (RB_PAGE_MASK): renamed from PAGE_MASK.
+
+Tue Jul 27 23:09:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_taint): use rb_obj_taint.
+	  (path_untaint): Pathname#untaint translated from
+	  pathname.rb.
+
+Tue Jul 27 18:59:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_framework): added.
+
+Tue Jul 27 18:33:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defines.h, intern.h, missing.h, ruby.h, st.h, util.h: include
+	  config.h and defines.h for autoconf macros.
+
+Tue Jul 27 16:27:38 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (flo_cmp): typo.
+
+Tue Jul 27 16:09:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* class.c, compile.c, dir.c, file.c, iseq.c, parse.y, random.c:
+	  clean unused-value warnings.
+
+	* cont.c, process.c, vm_exec.h: clean cast warnings.
+
+Mon Jul 26 22:34:37 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_taint): Pathname#taint translated from
+	  pathname.rb.
+
+Mon Jul 26 18:18:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (config.h): add include guard.
+
+	* win32/Makefile.sub, bcc32/Makefile.sub (config.h): ditto.
+
+Mon Jul 26 14:38:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (XCFLAGS): missing.h needs config.h.
+
+Mon Jul 26 14:08:47 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/missing.h: revert a part of r28727.  config.h is expected
+	  to include only once, before including defines.h.
+	  including it here breaks some macro definitions.
+
+Mon Jul 26 13:52:25 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: define BROKEN_CLOSE only on FreeBSD.
+	  This needs to merge to 1.9.2.
+
+Mon Jul 26 13:01:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_thread_mark): should mark iseq itself other than normal
+	  iseq.  [ruby-dev:41880]
+
+Mon Jul 26 11:51:01 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: define BROKEN_CLOSE on FreeBSD.
+	  This fixes build failure on MSVC. [ruby-core:31481]
+
+	* include/ruby/ruby.h, include/ruby/missing.h:
+	  use BROKEN_CLOSE for replacing close(2).
+
+Mon Jul 26 09:51:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_thread_mark): mark only self of normal iseqs, not
+	  nodes.  [ruby-dev:41874]
+
+Sun Jul 25 05:37:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_cmp): honor the result of infinite? method of the
+	  other.  [ruby-core:31470]
+
+	* test/ruby/envutil.rb (EnvUtil#.suppress_warning): added.
+
+	* test/ruby/test_float.rb (TestFloat#test_Float): suppress
+	  warnings under --verbose.
+
+Fri Jul 23 16:07:32 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/mutex_m.rb (sleep): added Mutex_m#sleep to support
+	  ConditionVariable.
+
+Fri Jul 23 15:09:22 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_MINGW32): ignore msvc suffix.
+
+	* configure.in (RUBY_TRY_CFLAGS): try with werror turned on.
+
+Fri Jul 23 09:02:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_expr_str): fixed out-of-boundary access at invalid
+	  multibyte characters.
+
+Fri Jul 23 09:00:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (XCFLAGS): reverted mistakenly removed option.
+	  [ruby-dev:41872]
+
+	* include/ruby/missing.h: needs ruby/config.h.
+
+Thu Jul 22 20:58:55 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_EXTERN): unnecessary after all.
+
+Thu Jul 22 17:33:47 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* complex.c (nucomp_to_i): allow complex with imaginary zero to be
+	  converted.
+
+	* complex.c (nucomp_to_f, nucomp_to_r): ditto.
+
+Thu Jul 22 20:12:56 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread_pthread.c (get_stack): fix memory leak; pthread_attr_destroy
+	  must be called even when pthread_getattr_np is used.
+	  [ruby-core:31269]
+
+Thu Jul 22 16:27:41 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_expr_str): fix broken Regexp#inspect when it
+	  is ASCII-8BIT and non-ASCII character.
+	  The length of character should be from original byte string.
+	  [ruby-core:31431]
+
+Thu Jul 22 14:30:17 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/missing.h: add prototype for ruby_close().
+
+Thu Jul 22 12:58:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_EXTERN): ignore invalid attribute.
+
+	* configure.in: fix for older autoconf.
+
+Thu Jul 22 12:02:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/{intern,io}.h: add missing prototypes.
+
+Thu Jul 22 06:38:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (XCFLAGS): use -fvisibility=hidden if possible.
+
+	* configure.in (RMDIR): use --ignore-fail-on-non-empty if possible.
+
+	* configure.in (RUBY_EXTERN): FUNCTION-BODY was missing.
+
+Thu Jul 22 03:02:55 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* README.EXT: fix signature of rb_ensure and rb_rescue to match
+	  implementation.
+
+Wed Jul 21 23:07:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in: get rid of embedding an absolute path.
+
+Wed Jul 21 15:22:17 2010  Evan Phoenix <evan@fallingsnow.net>
+
+	* lib/rubygems/custom_require.rb, gem_prelude.rb: Load code from
+	  from gems properly.
+
+Wed Jul 21 15:15:02 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* sprintf.c: add short documentation about named reference.
+	  [ruby-core:31294]
+
+Wed Jul 21 15:00:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (local_push_gen): disable unused variable warnings in
+	  eval.  [ruby-dev:41869]
+
+Wed Jul 21 13:57:37 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_method.c (rb_method_boundp): revert r28543, r28564.
+	  They may be merged in Ruby 2.0. [ruby-core:31217]
+
+Wed Jul 21 13:37:35 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/uri/common.rb: Have URI#route_to, URI#route_from accept
+	  string-like arguments [ruby-core:30961]
+
+	* lib/uri/generic.rb: ditto for URI.join, URI#merge
+
+Wed Jul 21 12:39:15 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/cmath.rb (CMath#cbrt): cbrt should accept a negative real
+	  numbers.  [ruby-core:31234]
+
+Wed Jul 21 12:31:30 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/uri/common.rb: Have URI() and URI.join accept URI objects in
+	  addition to strings. [ruby-core:30960]
+
+Wed Jul 21 11:55:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_hdtoa): renamed from BSD__hdtoa.
+
+Tue Jul 20 11:35:11 2010  Evan Phoenix  <evan@ruby-lang.org>
+
+	* lib/rubygems.rb: Load the custom require if --disable-rubygems
+	  was used.
+
+Tue Jul 20 11:27:18 2010  Evan Phoenix  <evan@ruby-lang.org>
+
+	* gem_prelude.rb: Pull in rubygem's custom require
+
+	* lib/rubygems.rb: Handle always having custom require available
+
+Tue Jul 20 18:39:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RSTRING_GETMEM): new macro to get ptr and
+	  len at once.
+
+	* string.c (rb_str_cmp, str_eql, rb_str_eql): trivial improvements.
+
+Tue Jul 20 18:23:04 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/closure.c (closure_data_type),
+	  ext/fiddle/function.c (function_data_type): rb_data_type_t has
+	  changed.
+
+Tue Jul 20 14:21:38 2010  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c: don't call after_gc_sweep() after when garbage_collect()
+	  was called, because rb_sweep_method_entry() free live unlinked
+	  method entries. [ruby-core:31169]
+
+Tue Jul 20 12:42:06 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (rb_getaddrinfo): according to my
+	  valgrind output this variable should be initialized beforehand.
+
+Tue Jul 20 12:50:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_flush_buffer): write and buffer operations should be
+	  monolithic.  [ruby-core:31348]
+
+Tue Jul 20 12:27:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::Entry_#copy): check file name
+	  separator boundary.  [ruby-core:31360]
+
+Mon Jul 19 18:34:12 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (get_strpath): check the type.
+	  (path_initialize): bypass to_path call for T_STRING.
+	  (path_freeze): implemented.
+
+	* ext/pathname/lib/pathname.rb (Pathname#freeze): removed.
+
+Mon Jul 19 12:33:29 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_clear): should not unshare embedded array, and
+	  should make unshared array embedded.
+
+Mon Jul 19 09:00:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* NEWS (Kernel#instance_eval): add an incompatible change since
+	  the 1.9.1 release.  [ruby-core:31336]
+
+Mon Jul 19 05:46:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_data_type_t): add new feature macros.
+
+Sun Jul 18 21:20:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_clear): no need to duplicate buffer just before
+	  clearing.
+
+Sun Jul 18 16:31:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_data_type_t): restructured.  [ruby-dev:41862]
+	  add parent member.
+
+	* error.c (rb_typeddata_inherited_p): new function.
+
+Sun Jul 18 01:09:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_initialize): return a value.
+
+Sun Jul 18 00:02:19 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/lib/pathname.rb (Pathname#initialize): removed.
+
+	* ext/pathname/pathname.c (path_initialize): implemented.
+	  (get_strpath): new function.
+	  (set_strpath): ditto.
+
+Sat Jul 17 19:01:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (MiniTest::Unit#process_args): refactored.
+
+Sat Jul 17 18:30:05 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/file2lastrev.rb: don't depend on pathname.rb if File.realpath
+	  is available.
+
+Sat Jul 17 14:10:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_add_suffix): fixed type warnings.
+
+Sat Jul 17 13:48:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/lib/pathname.rb: moved from lib/pathname.rb.
+	  require pathname.so.
+
+	* ext/pathname/pathname.c: new file to define empty Pathname class.
+
+	* ext/pathname/extconf.rb: new file.
+
+Sat Jul 17 13:04:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_wakeup_alive): split from rb_thread_wakeup.
+	  merged from r13476.  c.f. [ruby-core:31320]
+
+Sat Jul 17 10:07:52 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb: MiniTest::Unit is different class from
+	  Test::Unit, and install runner before loading test/minitest.
+
+Fri Jul 16 14:58:38 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit.setup_argv): run tests only when
+	  any tests have not run.
+
+	* bin/testrb, test/runner.rb: use Test::Unit.start.
+
+Fri Jul 16 12:06:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (ext-arch): prune directories start with '-'.
+
+Thu Jul 15 21:43:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* numeric.c (fix_divide): must not use rb_rational_new1 for coercion
+	  because it returns an argument itself when canonical mode is set.
+	  [ruby-core:31279]
+
+Thu Jul 15 21:38:31 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (bm_free): fix memory leak.  [ruby-core:30869] [Bug #3466]
+
+Thu Jul 15 15:47:32 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regexec.c (onig_search): don't skip non-ANYCHARs when
+	  .* fails to match. This causes to fail matching
+	  ANYCHAR_STAR with LOOK_BEHIND. This fix is workaround
+	  and disable the optimization. [ruby-dev:41851]
+
+Thu Jul 15 06:11:29 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/rdoc/markup/attribute_manager.rb: fixing ri output when special
+	  characters are inside html tags.  Thanks Tomo Kazahaya! [Bug #3512]
+
+Thu Jul 15 06:01:42 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (guess_local_offset): use the UTC offset of an older date on
+	  64bit time_t environment.
+
+Thu Jul 15 02:42:51 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/test/unit.rb (setup_argv): convert to using optparse, adding
+	  --help and -I. [ruby-core:30154]
+
+Thu Jul 15 00:35:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (localtime_with_gmtoff_zone): renamed from
+	  localtime_with_gmtoff and return the timezone abbreviation name.
+	  (guess_local_offset): return the isdst and timezone abbreviation name.
+	  (localtimew): use the returned isdst and timezone abbreviation name.
+	  [ruby-core:31275]
+
+Wed Jul 14 20:23:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (frame_func_id), vm_eval.c (rb_iterate),
+	  vm_insnhelper.c (vm_yield_with_cfunc): as the name of a C-level
+	  block, use the current method ID at the creation point.
+	  [ruby-dev:41852]
+
+Wed Jul 14 18:18:05 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regexec.c (match_at): add end point to enclen's argument.
+	  This only effect on compiling with -DONIG_DEBUG_MATCH.
+
+Tue Jul 13 21:34:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c: libedit is missing declaration of
+	  rl_getc.
+
+Tue Jul 13 21:31:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_daemon): split from proc_daemon.
+
+	* process.c (rb_fork_err): suppress gcc 4.4 warnings.
+
+	* random.c (fill_random_seed): ditto.
+
+Tue Jul 13 21:01:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (establishShell): chfunc must not raise any
+	  exceptions.
+
+Tue Jul 13 20:58:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_DEFAULT_ARCH): adjust for target
+	  cpu. [ruby-core:31243]
+
+Tue Jul 13 20:46:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/handle.c (rb_dlhandle_close): should not pass a dynamic
+	          string to rb_raise directly.
+
+Tue Jul 13 12:04:57 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* missing/close.c: keep original errno.
+
+Mon Jul 12 01:58:56 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/csv.rb: Fix unused variable warnings.
+	  Patch by Run Paint [ruby-core:30991]
+
+	* lib/date.rb: ditto
+
+	* lib/debug.rb: ditto
+
+	* lib/drb/drb.rb: ditto
+
+	* lib/drb/invokemethod.rb: ditto
+
+	* lib/irb/ruby-lex.rb: ditto
+
+	* lib/irb/slex.rb: ditto
+
+	* lib/logger.rb: ditto
+
+	* lib/pathname.rb: ditto
+
+Sun Jul 11 21:20:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT{,.ja} (rb_block_call): fixed about third/fourth
+	  arguments to the block.  based on [ruby-core:31192] by Asher
+	  Haig.  [Bug #3558]
+
+Sun Jul 11 17:01:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in ({,dist-,real-}clean-ext): should also omit
+	  semicolon when omitting "in WORDS" of FOR-statement.
+
+	* tool/rmdirs: ditto.
+
+Sun Jul 11 11:07:42 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/rdoc/test_rdoc_ri_driver.rb (test_formatter): add a test.  a
+	  patch from caleb clausen in [ruby-core:31182].
+
+Sat Jul 10 21:43:11 2010  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c (rb_strftime_with_timespec): fix precision handling for
+	  %:z and %::z.
+
+Fri Jul  9 22:32:54 2010  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c: unused declarations removed.
+
+Fri Jul  9 21:20:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c (rb_strftime_with_timespec): clear colons at unknown
+	  directive.
+
+Fri Jul  9 21:04:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c: don't call tzset.  strftime.c doesn't depend on
+	  the global timezone.
+
+Fri Jul  9 20:30:26 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): 24:00 should be the beginning of the next
+	  day even if the leap second, 23:59:60, exists.
+
+Fri Jul  9 01:08:46 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/emitter.rb (initialize): line_width is
+	  a valid option passed to the emitter.
+
+	* test/psych/test_psych.rb: corresponding tests.
+
+Fri Jul  9 00:49:46 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/emitter.c (line_width, set_line_width): preferred line may
+	  be set on the emitter.
+
+	* test/psych/test_emitter.rb: corresponding tests.
+
+Thu Jul  8 15:47:34 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_conv_enc_opts): fix infinite loop because
+	  of ISO-2022-JP conversion with empty string.
+	  patched by Brian Buchanan [ruby-core:31107]
+
+Thu Jul  8 08:16:57 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb (push): adding version
+	  and header emit options.
+
+	* test/psych/test_psych.rb: corresponding test.
+
+Thu Jul  8 08:01:03 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/emitter.c: updating documentation about emit options
+
+	* ext/psych/lib/psych/core_ext.rb: ditto
+
+	* ext/psych/lib/psych.rb (dump): passing emit options to emitter.
+
+	* ext/psych/lib/psych/nodes/node.rb: ditto
+
+	* ext/psych/lib/psych/visitors/emitter.rb: sending emit options to
+	  YAML emitter. [ruby-core:28318]
+
+Thu Jul  8 06:05:58 2010  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c (rb_strftime_with_timespec): support %:z and %::z.
+	  [ruby-dev:41841]
+
+Thu Jul  8 00:15:50 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gem_prelude.rb: provide workaround for gem activation.  Currently,
+	  gem activation does not work by default.  Now it can be worked
+	  around by requiring "rubygems" first.  [ruby-core:29486]
+	  a patch from Evan Phoenix in [ruby-core:31096].
+
+	* lib/rubygems.rb: ditto.
+
+Wed Jul  7 10:01:34 2010  Adrian Bloomer  <adrian.bloomer@gmail.com>
+
+	* numeric.c (fix_rev): Replaced fix_rev with '~num | FIXNUM_FLAG'.
+
+Wed Jul  7 13:22:20 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (ruby_find_basename): set correct baselen.
+
+Wed Jul  7 13:02:59 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* vm_method.c (rb_method_boundp): respond_to?(:protected_method,
+	  true) should return true.  Pointed out by Marc-Andre Lafortune.
+	  [ruby-dev:41837]
+
+Wed Jul  7 12:00:24 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (ruby_find_basename): should initialize f.
+
+Wed Jul  7 11:44:01 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_raise): add invalid header to
+	  exceptions.
+
+Wed Jul  7 10:26:20 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* missing/close.c: ignore ECONNRESET.
+	  FreeBSD wrongly sets ECONNRESET on close(2) and
+	  it causes false-negative exceptions. [ruby-dev:41778]
+
+	* configure.in: ditto.
+Tue Jul  6 22:57:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (nogvl_copy_stream_sendfile): jump to retry_sendfile directly
+	  to avoid select() on a socket which TCP state is CLOSED.
+	  patch by Eric Wong.  [ruby-core:31053]
+
+Tue Jul  6 21:45:34 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io/console/console.c (winsize_row): should return actual screen
+	  size, not buffer size.
+
+Tue Jul  6 08:35:58 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/dl/lib/dl/import.rb (handler): add a more helpful error message
+	  when calling import_symbol or import_function without calling
+	  dlload.  Thanks nobu! [ruby-core:30996]
+
+Tue Jul  6 00:34:50 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (thread_free): free altstack to prevent memory leak.  a patch
+	  from Tomoyuki Chikanaga in [ruby-dev:41815]. [Bug #3537]
+
+Tue Jul  6 00:29:27 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (NONASCII_MASK): NONASCII_MASK must be unsigned.
+	  [ruby-dev:41782]
+
+Mon Jul  5 16:05:39 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* vm_method.c (rb_method_boundp): Return false for protected
+	  methods when called from Kernel#respond_to?. [ruby-dev:40461]
+
+Mon Jul  5 12:32:01 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb (parse_string): support
+	  timezones that are not one hour off. [ruby-core:31023]
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ditto
+
+Sun Jul  4 22:49:54 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_syntax.rb: split test_syntax from test_system.rb.
+
+Sun Jul  4 22:02:02 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_system.rb (test_syntax): unified with test_syntax2.
+
+Sun Jul  4 21:00:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_system.rb (test_syntax2): moved from sample/test.rb
+	  [ruby-dev:41721]
+
+Sun Jul  4 17:13:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (swallow, prepare_getline_args, rb_io_getline_1): fix for
+	  paragraph mode reading in non-ascii-compatible encoding.
+	  [ruby-dev:41803]
+
+Sat Jul  3 16:14:10 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (Init_bigdecimal): add two new constants
+	  BigDecimal::INFINITY and BigDecimal::NAN.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (BigMath.exp): modify the
+	  behaviors for infinity arguments as same as Math.exp.
+
+Sat Jul  3 09:47:26 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb(visit_Psych_Nodes_Scalar):
+	  teaching Psych to deserialize DateTime objects. [Bug #1390]
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb(visit_DateTime): added a
+	  method for serializing DateTime objects.
+
+	* ext/psych/lib/psych/scalar_scanner.rb(parse_time): add method for
+	  parsing times objects from a string.
+
+	* test/psych/test_date_time.rb: tests for dumping DateTime objects.
+
+Sat Jul  3 09:13:55 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb (visit_Time): use
+	  Time#nsec to accurately serialize time objects. [ruby-core:29233]
+
+Fri Jul  2 23:30:23 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (argf_inplace_mode_set): prohibits an assignment of a tainted
+	  value.
+
+	* file.c (ruby_find_basename, ruby_find_extname): split from
+	  rb_file_s_basename() and rb_file_s_extname().
+
+	* util.c (ruby_add_suffix): support arbitrary length of the suffix
+	  to get rid of the potential buffer overflow.
+	  reported by tarui.
+
+Fri Jul  2 05:31:51 2010  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num2long): accept LONG_MAX < x < LONG_MAX+1 and
+	  LONG_MIN-1 < x < LONG_MIN as well because they are converted
+	  into the valid range of long by truncation.
+	  (rb_num2ulong): accept ULONG_MAX < x < ULONG_MAX+1 and
+	  LONG_MIN-1 < x < LONG_MIN as well.
+	  (rb_num2ll): accept LLONG_MAX < x < LLONG_MAX+1 and
+	  LLONG_MIN-1 < x < LLONG_MIN.
+
+Thu Jul  1 23:10:25 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Improving documentation.
+
+Thu Jul  1 22:15:01 2010  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num2ulong): fix the lower limit for float.
+	  [ruby-dev:41361]
+
+Thu Jul  1 21:37:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (EnvUtil#invoke_ruby): integrated with
+	  runexec.
+
+Thu Jul  1 16:19:53 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (thread_start_func_1): don't call
+	  native_thread_init_stack(th) on cygwin to avoid the segv
+	  introduced by r27789. Cygwin's signal implementation is half
+	  baked so USE_SIGNALSTACK is not defined and it needs another
+	  treatment.
+
+Thu Jul  1 13:00:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (test-knownbug): ignore known bugs.
+
+Thu Jul  1 08:40:26 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): suppress empty instance variable entry on
+	  compatible dump objects.
+
+Wed Jun 30 07:29:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (test_script_from_stdin): by using
+	  a pipe, get rid of not-well-defined behavior after the child
+	  process terminated in pty.
+
+Wed Jun 30 02:30:26 2010  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* thread_pthread.c (get_stack): use pthread_getthrds_np() for AIX.
+
+	* configure.in: ditto.
+
+Tue Jun 29 21:11:15 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* ext/stringio/stringio.c (strio_write): add RB_GC_GUARD.
+
+Tue Jun 29 19:39:59 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_method.rb (is_ruby64): check
+	  ruby binary is mswin64 or mingw64. [ruby-dev:41756]
+
+Tue Jun 29 14:18:21 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_realpath_internal, realpath_rec): skip UNC share root
+	  on DOSISH platforms.
+	  http://pc12.2ch.net/test/read.cgi/tech/1272248179/600
+
+Tue Jun 29 11:52:33 2010  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (unlink_heap_slot, slot_sweep): unlink heaps_slot of
+	  heaps_slot linked list if heaps_slot is empty at slot_sweep.
+	  fixed [ruby-dev:41543], [ruby-core:24894].
+
+Tue Jun 29 01:22:08 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/fiddle.h: added FFI type detection to avoid bug in ffi
+	  header files. Thanks Yugui! [ruby-core:30917]
+
+Mon Jun 28 22:14:22 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/rdoc/test_rdoc_ri_driver.rb (TestRDocRIDriver#test_formatter):
+	  fix a test accordingly to r28455.
+
+Mon Jun 28 21:56:14 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_fd_resize, rb_fd_copy): avoid NULL dereference upon
+	  failed realloc by using xrealloc instead of not realloc.  a patch
+	  from Jim Meyering <meyering at redhat.com> in [ruby-core:30920]
+	  [Bug #3489]
+
+Mon Jun 28 20:32:33 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_method.rb (test_offset_vtbl): check
+	  that Ruby is 32bit or 64bit binary in order to get correct offset
+	  value. [ruby-dev:41741]
+
+Mon Jun 28 05:32:51 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rdoc/ri/driver.rb (RDoc::RI::Driver#formatter): should use bs
+	  format when stdout is piped.  [ruby-core:30734]
+
+Mon Jun 28 03:12:03 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bootstraptest/test_class.rb: add a test for [ruby-core:30843].
+
+Mon Jun 28 02:43:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* class.c (rb_mod_init_copy): when class is dup'ed, a metaclass of the
+	  class should be attached to the dup'ed class, not the original
+	  class.  [ruby-core:30843] [Bug #3461]
+
+Sun Jun 27 23:31:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h, io.c: reverted r21709.
+
+	* ruby.c (load_file_internal): nothing to read if EOF reached
+	  while reading shebang.  [ruby-core:30910]
+
+Sun Jun 27 13:25:07 2010  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (simple_sendfile): don't try to send data more than SSIZE_MAX
+	  with single sendfile call..
+	  based on the patch by Eric Wong.  [ruby-core:30908]
+
+Sun Jun 27 10:41:38 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/require_paths_builder.rb (write_require_paths_file_if_needed):
+	  no reason that bin directory should be included in $LOAD_PATH.
+	  it is for executable files, but not libraries.  [ruby-core:25936]
+
+Sat Jun 26 13:07:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (enable-debug-env): option for RUBY_DEBUG env.
+
+Sat Jun 26 11:56:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (clear-installed-list): put redirection before set
+	  command, since it seems to be handled by nmake in special way.
+	  [ruby-dev:41711]
+
+Sat Jun 26 10:08:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (EnvUtil#invoke_ruby): no needs to copy the
+	  original ENV, since it's done in spawn automatically.
+	  [ruby-dev:41733]
+
+Sat Jun 26 08:44:22 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* NEWS (zlib): new methods.  [ruby-dev:41706] [Bug #3472]
+
+Sat Jun 26 07:59:18 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_method.rb (test_offset_vtbl): check
+	  that OS is Windows 32bit or Windows 64bit in order to get
+	  correct offset value.
+
+Sat Jun 26 04:39:12 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* ext/dl/lib/dl/func.rb (call): don't overwrite original arguments
+	  to defend from GC.
+	* test/dl/test_func.rb (test_string): add test for above.
+
+Fri Jun 25 11:45:36 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Fixing a bug that prevented CSV from parsing
+	  all multi-line fields correctly.  Patch by Rob Biedenharn.
+
+Fri Jun 25 10:07:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_in_out_err):
+	  return the exit status.
+
+	* test/ruby/envutil.rb (EnvUtil#invoke_ruby): change the
+	  environment of spawned process only.
+
+	* test/ruby/envutil.rb (EnvUtil#invoke_ruby): get rid of possible
+	  deadlock.
+
+Fri Jun 25 06:24:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/un.rb (httpd): MaxClients also should be integer.
+	  [ruby-dev:41724] [Bug #3477]
+
+	* lib/un.rb (setup): fix of word splitting.  [ruby-dev:41723]
+	  [Bug #3476]
+
+Fri Jun 25 01:07:17 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* include/ruby/ruby.h (static inline rb_gc_guarded_ptr): prevent
+	  RB_GC_GUARD_PTR being removed by optimization.
+
+Thu Jun 24 06:22:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (ruby_getaddrinfo__darwin): the workaround
+	  for Mac OS X moved from rsock_getaddrinfo.
+
+Thu Jun 24 05:44:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (struct dump_arg, struct load_arg): merge taint and
+	  untrust flags into infection as bit flags.
+
+	* marshal.c (w_nbyte, clear_dump_arg): infect the buffer as soon
+	  as appending, because it might have been finalized already at
+	  exit.  based on a patch by Tomoyuki Chikanaga
+	  at [ruby-dev:41672].  [Bug #3463]
+
+Wed Jun 23 23:49:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (str_is_number): renamed from str_isnumber to
+	  avoid confusion to str_isnumber in ext/socket/getaddrinfo.c.
+
+Wed Jun 23 23:45:31 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* NEWS (stringio): new methods.  [ruby-dev:41687] [Bug #3469]
+
+Wed Jun 23 22:52:00 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c: don't use __P.
+
+Wed Jun 23 21:32:08 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (reg_get_typelib_file_path): try win64
+	  registry entry at first. [ruby-dev:41674] [Bug #3464]
+
+Wed Jun 23 21:17:32 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (ruby_getaddrinfo__darwin): new workaround for
+	  getaddrinfo problem on Mac OS X Snow Leopard.  [ruby-core:29427]
+	  patch by Wataru Kimura.  [ruby-core:30842]
+
+Wed Jun 23 17:12:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* NEWS (ptr): new method and deprecated methods.  [ruby-dev:41681]
+
+	* ext/pty/{README,README.ja}: ditto.
+
+	* ext/pty/pty.c (pty_check): add rdoc.
+
+Wed Jun 23 12:44:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_regexp.rb (test_dup_warn): read in UTF-8
+	  encoding regardless environment.
+
+	* test/ruby/envutil.rb (invoke_ruby): add encoding option.
+
+Wed Jun 23 06:44:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (fchown): need to check.  a patch by Eric Wong
+	  at [ruby-core:30818].
+
+	* io.c (argf_next_argv): check for setting owner/group.
+
+Tue Jun 22 23:10:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (num_exact): fix for mathn.  [ruby-dev:41599]
+
+Tue Jun 22 22:00:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb: suppress warnings.
+
+Tue Jun 22 21:33:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/optparse/test_summary.rb: fixed superclass so that it run
+	  solely.
+
+Tue Jun 22 19:57:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (shadowing_lvar_gen): should add lvar to vars and
+	  used.  [ruby-dev:41666]
+
+Tue Jun 22 13:23:13 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/delegate.rb: Forward #trust, #untrust, #taint and #untaint
+	  to both the delegator and __getobj__ [ruby-core:26138]
+
+Mon Jun 21 23:41:08 2010  wanabe  <s.wanabe@gmail.com>
+
+	* eval_error.c (error_print): clear raised_flag while error-printing
+	  to avoid hang. [ruby-core:27608]
+
+	* test/ruby/test_beginendblock.rb (test_endblock_raise): add test for
+	  above.
+
+Sun Jun 20 16:17:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* id.c (Init_id): add underscore name.
+
+	* parse.y (warn_unused_var): ignore underscore name.
+
+	* parse.y (warn_unused_var): use same format as shadowing local
+	  variable.
+
+	* parse.y (shadowing_lvar_gen): get rid of adding extra name for
+	  shadowing local variable.  [ruby-dev:41628]
+
+Sat Jun 19 11:11:37 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb.rb: remove double exclamation marks.  a patch from Diego
+	  Viola.  [ruby-core:30589]
+
+Sat Jun 19 03:35:58 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb (resolve_klass): fix the
+	  exception message when attempting to load an unknown class.  Thanks
+	  nobu! [ruby-dev:41399]
+
+	* test/psych/test_psych.rb: test for the exception message
+
+Fri Jun 18 10:37:46 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (gc_lazy_sweep): clean a warning.
+	  "suggest parentheses around assignment used as truth value"
+
+	* transcode_data.h (getGB4bt1): clean a warning.
+	  "suggest parentheses around arithmetic in operand of |"
+
+	* transcode_data.h (getGB4bt3): ditto.
+
+	* vm.c (thread_free): clean a warning. "format 'p' expects type
+	  'void *', but argument 3 has type 'struct rb_mutex_struct *'"
+
+Fri Jun 18 10:15:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c: add prototype of rb_realpath_internal.
+
+	* load.c: ditto.
+
+Fri Jun 18 01:50:21 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* object.c: Object#public_methods, private_methods, etc. returns
+	  method ids that belong to the class or the singleton class(es) of
+	  the object.  [ruby-dev:41613]
+
+	* class.c: on the other hand, Module#public_instance_methods, etc.
+	  returns method ids that belong to the module itself (even if the
+	  module is singleton, it does not return method ids of super
+	  class(es); see [ruby-core:28837]).
+
+Fri Jun 18 01:22:55 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/webrick/httpservlet/filehandler.rb
+	  (prevent_directory_traversal): apply filesystem encoding to path
+	  only during calling File.expand_path.  [ruby-dev:41423]
+
+Thu Jun 17 23:20:14 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* load.c (rb_load_internal): remove call to rb_realpath_internal
+	  within rb_load_internal which caused big performance degradation.
+	  Instead, call rb_realpath_internal in the caller of
+	  rb_load_internal.  [ruby-dev:41502] [ruby-dev:41610]
+
+	* vm.c (rb_vm_call_cfunc): ditto.
+
+	* eval_intern.h (rb_vm_call_cfunc): ditto.
+
+	* ruby.c (process_options): ditto.
+
+Thu Jun 17 18:37:47 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_str_encode_ospath): when the encoding of the parameter
+	  is ASCII-8BIT, should recognize as filesystem encoding, and convert
+	  to UTF-8 on Windows.
+
+	* file.c (realpath_rec): should convert to ospath encoding before
+	  calling lstat().
+
+	* file.c (rb_realpath_internal): resolved string should take over
+	  the encoding of base string.
+
+	* transcode.c (rb_str_encode): should return new string always.
+	  fixed #3444.
+
+Wed Jun 16 18:34:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y: ripper needs warn_unused_var(), too.
+
+Wed Jun 16 18:20:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (warn_unused_var): warn unused variables.
+	  [ruby-dev:41620]
+
+Wed Jun 16 15:40:53 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* error.c (rb_bug): existence of _set_abort_behavior() depends on
+	  runtime version, not compiler version.
+
+Wed Jun 16 01:38:40 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/fiddle/extconf.rb: check ffi.h even when pkg-config succeeded.
+	  On Debian/lenny, which is a "supported" platform, ffi.h is installed
+	  in /usr/include/i486-linux-gnu/.  This causes build error when using
+	  gcc whose target is not i486-linux-gnu.
+
+Wed Jun 16 00:04:38 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_io.rb (safe_4): does not use Timeout because
+	  Timeout.timeout uses Thread#kill which raises SecurityError when
+	  $SAFE == 4.  based on a patch from Tomoyuki Chikanaga.
+	  [ruby-dev:41484]
+
+	* test/ruby/test_io.rb (test_print_separators): use pipe (test helper
+	  method) instead of IO.pipe.  [ruby-dev:41484]
+
+Tue Jun 15 17:14:58 2010  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/fiddle/extconf.rb: De Morgan's laws.
+
+Tue Jun 15 12:09:00 2010  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/fiddle/extconf.rb: use pkg_config to find ffi.h.
+
+Tue Jun 15 02:31:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.mk (clean-ext, clean-enc): clean up cleaning
+	  targets.
+
+Mon Jun 14 22:36:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_socket.rb (test_udp_server): rescue Errno::ENOSYS
+	  for Windows.
+
+	* test/socket/test_nonblock.rb (test_sendmsg_nonblock_error): ditto.
+
+	  [ruby-dev:41597] reported by Masaya Tarui.
+
+Mon Jun 14 17:44:39 2010  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: raise DRbConnError instead of ArgumentError if too
+	  many arguments. [ruby-dev:41481]
+
+	* test/drb/test_drb.rb: ditto.
+
+	* test/drb/drbtest.rb: ditto.
+
+Mon Jun 14 04:03:55 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/fiddle/closure.c (dealloc): refix workaround r28300.
+	  don't use ffi_closure_alloc, ffi_prep_closure_loc and
+	  ffi_closure_free on MACOSX and __linux__.
+	  [ruby-dev:41483] [ruby-dev:41214]
+
+Sun Jun 13 15:46:07 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rubyoptions.rb: add a simple test for __END__ and
+	  DATA.
+
+Sun Jun 13 04:24:18 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/delegate.rb: Delegate !=, eql? and hash [ruby-core:26139]
+
+Sun Jun 13 02:12:46 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf8_mac.trans (buf_apply): fix for patterns
+	  whose result is 2 bytes. [ruby-core:30751]
+
+Sun Jun 13 01:38:17 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (rb_localtime_r2): fix mixed declarations and code.
+
+Sun Jun 13 00:27:24 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/dl/lib/dl.rb: don't require when already loaded.
+
+Sun Jun 13 00:02:56 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/fiddle/closure.c (dealloc): workaround fix for libffi's
+	  ffi_closure_free. [ruby-dev:41483] [ruby-dev:41214]
+
+Sat Jun 12 10:02:26 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_syscall): should check argument string taint before
+	  invoking system calls.
+
+Sat Jun 12 09:18:31 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (uninitialized_constant): process through
+	  rb_class_real() to remove "Object::" from error message.
+	  [ruby-dev:40951]
+
+Sat Jun 12 00:38:37 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/dbm/extconf.rb: accept Berkley DB version 3, 4 and 5.
+	  a patch from Takahiro Kambe.  [ruby-dev:41531]
+
+Fri Jun 11 23:21:35 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (init_leap_second_info): check the result of
+	  gmtime_with_leapsecond.
+
+Fri Jun 11 23:04:59 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_get_ciphers):
+	  use sk_SSL_CIPHER_num and sk_SSL_CIPHER_value instead of cast.
+	  patched by Takahiro Kambe [ruby-dev:41530]
+
+Fri Jun 11 22:59:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/missing.h (isnan): fix compilation error on OpenBSD.
+
+Fri Jun 11 22:39:50 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rubygems/install_update_options.rb
+	  (Gem::InstallUpdateOptions#add_install_update_options): deprecate
+	  --test option which has not worked.  [ruby-core:21714]
+
+	* test/rubygems/test_gem_command_manager.rb: ditto.
+
+Fri Jun 11 07:34:25 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): test the result of LOCALTIME.
+
+Fri Jun 11 00:42:45 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (rb_localtime_r2): fix localtime overflow check.
+
+Thu Jun 10 23:05:44 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (rb_localtime_r2): refine localtime overflow check for
+	  FreeBSD 6.4.
+
+Thu Jun 10 09:10:08 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_putc): documentation updated to mention putc would
+	  not work well with multi-byte characters.  [ruby-core:30697]
+
+	* io.c (rb_f_putc): ditto.
+
+Wed Jun  9 22:51:50 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): always extrapolate from past.
+	  [ruby-core:30672] reported by Benoit Daloze.
+
+Wed Jun  9 22:13:08 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (calc_tm_yday): extracted from timegmw_noleapsecond.
+
+Tue Jun  8 06:27:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gem_prelude.rb: load full rubygems at LoadError for activation
+	  check.  [ruby-core:29486]
+
+Tue Jun  8 06:04:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (parse_mode_enc): set set_by_bom bit.  [ruby-core:30641]
+
+Sun Jun  6 22:37:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_buf_cat_escaped_char): get rid of buffer
+	  overflow on platforms int is bigger than 32bit, and warnings.
+
+Sun Jun  6 19:55:21 2010  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_product): clear uninitialized elements in temporary
+	  array.
+
+Sun Jun  6 12:31:57 2010  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c: set_heaps_increment is not
+	  called before gc_marks of lazy sweeping because live_num not
+	  contain finalize objects. So call set_heaps_increment after lazy
+	  sweeping if free_num are not enough. And move free_min to struct
+	  rb_objspace for above.  [ruby-dev:41499]
+
+Sun Jun  6 10:44:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_product): need to set the length in order to get
+	  the entries marked.  [ruby-dev:41540]
+
+Sun Jun  6 08:26:01 2010  Tanaka Akira  <akr@fsij.org>
+
+	* vm.c (Init_BareVM): call Init_native_thread here.
+
+	* thread.c (Init_Thread): don't call Init_native_thread.
+
+	* thread_pthread.c (Init_native_thread): exported.
+
+	* thread_win32.c (Init_native_thread): ditto.
+
+	[ruby-dev:41536]
+
+Sun Jun  6 08:21:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* NEWS: Enumerable#join has been reverted.  [ruby-core:30604]
+
+Sat Jun  5 20:30:49 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_expr_str): ASCII incompatible strings
+	  must always escape or converted.
+
+	* re.c (rb_reg_expr_str): use rb_str_buf_cat_escaped_char
+	  when resenc is given: for Regexp#inspect or error message.
+
+	* re.c (rb_reg_desc): add 'n' for ENCODING_NONE.
+
+Sat Jun  5 01:20:14 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (sym_inspect): Escape when the symbol is not
+	  resulted encoding and not ascii_only. It had escaped
+	  ascii-incompatible string, but it is wrong.
+
+Sat Jun  5 01:10:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_buf_cat_escaped_char): defined.
+	  Splited from rb_str_inspect.
+
+Sat Jun  5 16:39:13 2010  Tanaka Akira  <akr@fsij.org>
+
+	* .gdbinit (rp): show type name for RTypedData.
+
+Sat Jun  5 15:59:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options, load_file_internal): $0 seen from
+	  required libraries by -r option should be the main script.
+	  [ruby-core:23717]
+
+Sat Jun  5 15:15:52 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (EnvUtil#invoke_ruby): support for child
+	  process env.
+
+Sat Jun  5 14:11:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/net/imap/test_imap.rb (test_exception_during_shutdown):
+	  need to raise always.
+
+Sat Jun  5 13:13:30 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (eql?, ==, minor): Fix bugs when comparing/returning
+	  some empty matrices.
+
+Sat Jun  5 11:00:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* error.c (rb_name_err_mesg_new): guard mesg, recv and method.
+	  The problem in [ruby-dev:41464] is caused because mesg is collected.
+
+Sat Jun  5 10:03:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* .gdbinit (rp): detect and show RTypedData.
+
+Sat Jun  5 08:30:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_f_test): 'W' should test writable by real uid/git,
+	  not world writable.   [ruby-core:30587]
+
+Sat Jun  5 06:20:57 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* dir.c: Clarification of what '*' matches.  Patch by John Wells
+	  <john.wells at greatworx.com>
+
+Fri Jun  4 10:46:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_inspect): should taint the result.
+	  [ruby-dev:41533]
+
+Fri Jun  4 09:37:15 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (gc_sweep): suppress a warning on VC, again.
+
+Thu Jun  3 23:34:55 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb: Emit a deprecation warning in
+	  verbose mode and add a caution to the overview section of the
+	  document. [ruby-dev:41525]
+
+Thu Jun  3 19:33:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_inspect): suppress warnings by
+	  rb_sprintf.
+
+	* ext/dl/cptr.c (rb_dlptr_s_malloc): suppress warnings.
+
+Thu Jun  3 18:58:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (ruby_init_ext): statically linked extensions have no
+	  real path.  [ruby-dev:41526]
+
+	* vm.c (rb_vm_call_cfunc): add filepath argument.
+
+Thu Jun  3 18:17:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rake/test_win32.rb (Rake::TestWin32): update tests.
+	  [ruby-core:30309]
+
+	* test/io/nonblock/test_flush.rb (test_flush): try pipe and
+	  socketpair.  [ruby-dev:41517]
+
+Thu Jun  3 09:36:43 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/dl.h (DLSTACK_TYPE): type of stack is same as VALUE.
+	  reported by sakiyama shin in [ruby-dev:41514]
+
+Thu Jun  3 06:30:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_path.rb (test_path): workaround for drive
+	  letter.
+
+	* test/ruby/test_path.rb (test_path): get rid of blocking on
+	  access to non-existent host.
+
+Thu Jun  3 05:37:46 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* tool/rbinstall.rb (install-bin): Allow bin/* install from dot-dirs.
+	  Fixes rvm and multiruby installations.
+
+Thu Jun  3 01:22:45 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fix typo and race condition.
+
+Thu Jun  3 00:58:45 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/dl/lib/dl/cparser.rb (parse_ctype): add backwards compatibility
+	  by supporting "uint" types in the c parser. [ruby-core:29750]
+	* test/dl/test_cparser.rb: adding a test for "uint" changes.
+
+Wed Jun  2 11:40:02 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): should consider block on stack,
+	  if block argument is passed.  [ruby-core:30534]
+
+	* parse.c (arg_concat_gen): should append to nd_head, not to
+	  nd_iter for NODE_BLOCK_PASS.
+
+Tue Jun  1 23:12:06 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (unescape_nonascii): \P{FOO} is also Unicode property in
+	  regexp. [ruby-core:30540]
+
+Tue Jun  1 21:29:39 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: adding support to rfc5789 patch verb.
+	  Added a Net::HTTP::Patch class which expects a message body
+	  and response body. It recycles the post method into a patch one,
+	  that will send the encoded representation to the server.
+	  Summarizing, a new class has been created, the post method
+	  extracted into send_entity, including a new argument,
+	  which defines which class to use (Post or Patch) and
+	  finally a patch method was created. [ruby-core:30426]
+	  Patched by Guilherme Silveira
+	  <guilherme.silveira AT caelum.com.br>
+
+Tue Jun  1 03:46:08 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/extconf.rb: reject GPLv3 readline. [ruby-dev:39172]
+	  [ruby-core:25272] [ruby-dev:39167] [ruby-core:28736]
+
+Tue Jun  1 01:14:31 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* io.c (pipe_open): add RB_GC_GUARD.
+	  This caused failure when test/ruby/test_argf.rb is executed with
+	  GC.stress = true in mswin32_90
+
+Mon May 31 23:44:22 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: use tclConfig.sh/tkConfig.sh when frameworks
+	  are enabled on MacOS X.
+
+	* ext/tk/stubs.c: dirty hack for frameworks and stubs on MacOS X.
+
+	* ext/tk/lib/tk.rb: stop creating a dummy Tcl/Tk interpreter.
+	  And hide a root window before starting eventloop. (for ruby 1.9)
+
+	* ext/tk/tcltklib.c: add codes to support Ruby/Tk-Kit (Rubykit).
+
+Mon May 31 21:49:42 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Requester#request): rescue ECONNRESET
+	  for Windows.  reported by U.Nakamura.  [ruby-dev:41477]
+
+Mon May 31 19:25:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regparse.c (onig_syntax_warn): do not use external strings as
+	  printf format.
+
+Mon May 31 18:22:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode-set-encoding): skip shebang line
+	  always regardless of ruby.
+
+Mon May 31 15:07:18 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_expand_path): Refix r28102: this breaks
+	  r28039. test for [ruby-dev:41429] is added. [ruby-core:30516]
+
+Mon May 31 14:47:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): inspect as ASCII when the codepoint
+	  of a character in Unicode string is ASCII printable one.
+
+Mon May 31 13:44:40 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_unicode_p): check the encoding is Unicode
+	  or not by the name; not function's pointer. [ruby-dev:41479]
+
+Mon May 31 04:03:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): check if expanded dname encoding is
+	  compatible with fname, not just copying.  [ruby-core:30516]
+
+	* test/ruby/test_beginendblock.rb (test_endblockwarn): needs
+	  encoding comment.
+
+Mon May 31 02:17:54 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (defined): respond_to_missing? may not be available
+	  (for BasicObject instances).
+
+Mon May 31 01:43:42 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/delegate: Delegator: combine (public|protected) methods with
+	  those of the delegated object. [ruby-core:27224]
+	  DelegateClass: combine (public|protected) instance methods
+	  with those of the delegated superclass.
+
+Sun May 30 22:18:49 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/set.rb (keep_if, select!): New methods [ruby-core:29749]
+
+Sun May 30 21:51:59 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/rake/test_application.rb: update a test because of r28089.
+
+Sun May 30 21:48:07 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/tempfile.rb (Tempfile#unlink): leave @data.  Assigning nil to
+	  @data caused double closing error of the same IO in finalizer.  a
+	  patch from Simon Nicholls.  [ruby-core:29395]
+
+Sun May 30 18:47:15 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: updated to b856dd07.
+	  this cleans warnings.
+
+Sun May 30 18:25:55 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* load.c (load_failed): use more accurate error message.
+	  [ruby-core:23851]
+
+Sun May 30 16:54:34 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* file.c (rb_file_directory_p): update rdoc.  a patch from Ilkka
+	  Laukkanen.  [ruby-core:30016]
+
+Sun May 30 14:59:13 2010  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (gc_mark_all_clear): don't call obj_free() in
+	  gc_mark_all_clear(), because obj_free() may make T_ZOMBIE
+	  objects.
+
+	* gc.c (gc_clear_mark_on_sweep_slots): renamed.
+
+	* gc.c (rb_objspace_call_finalizer): call
+	  gc_clear_mark_on_sweep_slots() before calling mark_tbl().
+	  [ruby-dev:41459]
+
+Sun May 30 03:40:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_utf8, parser_tokadd_string): allow NUL
+	  containing symbol literals, as well as String#to_sym.
+	  [ruby-dev:41447]
+
+Sun May 30 03:03:20 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/generator/template/darkfish/classpage.rhtml:
+	  fix encoding value of XML declaration. [ruby-dev:41452]
+
+Sun May 30 02:20:26 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* thread.c (RB_GC_SAVE_MACHINE_CONTEXT), gc.c (ruby_gc_stress_start):
+	  revert r28078.
+
+Sun May 30 02:21:34 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/zlib/zlib.c (zstream_append_input2): add RB_GC_GUARD.
+	  This caused failure when test/csv is executed with GC.stress = true.
+
+Sun May 30 01:25:48 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* thread.c (RB_GC_SAVE_MACHINE_CONTEXT): start GC
+	  by switching the thread if gc_stress == true
+
+	* gc.c (ruby_gc_stress_start): ditto.
+
+Sun May 30 00:02:39 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gc.c (force_chain_object, rb_objspace_call_finalizer): delete
+	  finalizer entry after corresponding finalizer is executed.
+	  This caused SEGV when test/cgi is executed with GC.stress = true.
+
+Sat May 29 23:30:33 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (str_replace_shared): change embedded state atomically.
+	  [ruby-core:29953] [ruby-dev:41456]
+
+Sat May 29 12:56:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encindex): avoid infinite require
+	  loop. [ruby-core:30467]
+
+Sat May 29 12:05:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (RDoc#read_file_contents): take care of BOM.
+	  [ruby-dev:41452]
+
+Sat May 29 10:12:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (GC_PROF_TIMER_STOP): regularize `marked' to suppress
+	  warnings.
+
+Sat May 29 09:34:32 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb: fix for --disable-gems.
+
+Sat May 29 09:04:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Requester#request): rescue ECONNREFUSED.
+
+Sat May 29 08:46:29 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (allocate_sorted_heaps, before_gc_sweep, gc_sweep): removed
+	  unused variables.
+
+Fri May 28 20:35:21 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (blocking_region_begin): avoid RUBY_VM_CHECK_INTS()
+	  call during GVL state transition. [Bug#3354] [ruby-dev:41435]
+
+Fri May 28 19:37:47 2010  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c: use simple lazy sweep algorithm for response performance
+	  gain. See [ruby-dev:41067].
+
+	* object.c: FL_MARK of some objects by lazy sweep is copied when
+	  RVALUE is cloned. These objects are not marked in the mark phase.
+	  So delete FL_MARK.
+
+	* class.c: ditto.
+
+Fri May 28 18:39:38 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_inspect): escape ASCII-compatible strings.
+
+	* string.c (rb_str_inspect): escape ASCII-compatible strings.
+
+Fri May 28 17:34:48 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* array.c (rb_ary_product): Use tmpary instead, to ensure marking
+	  arrays by GC.
+
+Fri May 28 11:40:07 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* array.c (rb_ary_product): Do not rely on GC, t0 should be
+	  checked explicitly.
+
+Fri May 28 10:40:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (checking_for): ignore toplevel.
+
+Fri May 28 00:47:16 2010  Masaya Tarui  <tarui@ruby-lnag.org>
+
+	* error.c (rb_bug): suppress the error report dialog if report_bug()
+	  is finished successfully.
+
+Fri May 28 00:32:25 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* insns.def (invokesuper): check consistency between class of self and
+	  class of method being invoked by super.  This is temporary measure
+	  for YARV.  See [ruby-core:30313] in detail.  See [ruby-dev:40959]
+	  [ruby-dev:39772] [ruby-core:27000] [ruby-core:27230]
+
+	* vm_insnhelper.c (vm_search_superclass): ditto.
+
+Thu May 27 23:38:31 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_home_dir): set filesystem encoding.
+
+Thu May 27 23:29:18 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_expand_path): set encoding as the same of fname
+	  when _result_ is not filesystem encoding. [ruby-dev:41429]
+
+Thu May 27 23:07:45 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_expand_path): use rb_enc_associate_index and
+	  rb_filesystem_encindex. Strings related FileSystem should
+	  have filesystem_encoding.
+
+	* file.c (SET_EXTERNAL_ENCODING): removed.
+
+Thu May 27 23:03:19 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_filesystem_encindex): defined.
+
+	* include/ruby/encoding.h (rb_locale_encindex): ditto.
+
+	* encoding.c (rb_filesystem_encindex): remove static.
+
+	* encoding.c (rb_locale_encindex): ditto.
+
+Thu May 27 22:34:27 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): call ReadFile() with len = 0 before
+	  reading really on console, because the first ReadFile() call after
+	  PeekConsoleInput() always returns broken data. (Windows's bug).
+	  [ruby-core:29018]
+
+Thu May 27 12:42:23 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* LGPL: Removed.  We no longer have LGPL'd source files in our
+	  tree.
+
+Wed May 26 20:19:22 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_rand): add the result of random to the
+	  beginning of range, not the opposite.  [ruby-dev:41415]
+
+Wed May 26 19:55:33 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (file_expand_path): revert a part of r22392. it's commit miss.
+
+Wed May 26 18:40:23 2010  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* test/rinda/test_rinda.rb (TupleSpaceTest, TupleSpaceProxyTest):
+	  kill a used thread at teardown.  [ruby-dev:41397]
+
+Wed May 26 12:08:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_rand): subtraction method of non-numeric can
+	  return Float.  [ruby-dev:41410]
+
+Wed May 26 11:50:09 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* marshal.c (Init_marshal): document marshal_dump and marshal_load.
+
+Wed May 26 10:35:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_large_fd_select): needed on mingw, even
+	  though fd_mask is not available.  [ruby-core:30401]
+
+Tue May 25 14:00:51 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (add_code_range_to_buf0): fix false negative
+	  warning when given range is just before previous range.
+	  [ruby-dev:41406]
+
+Tue May 25 16:37:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el (ruby-style-version): take revision if
+	  keyword is expanded. [ruby-dev:41408]
+
+Tue May 25 13:26:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/vm.h, include/ruby/encoding.h: add external
+	  linkage.
+
+Tue May 25 01:46:49 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_f_caller): update rdoc.  a patch from Nobuhiro IMAI
+	  <nov at yo.rim.or.jp> in [ruby-dev:41387].
+
+Mon May 24 23:04:41 2010  Tanaka Akira  <akr@fsij.org>
+
+	* error.c: fix wrong exception class name in rdoc.
+
+Mon May 24 19:00:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (warn_balanced, parser_yylex): no warnings after closing
+	  parens.
+
+Mon May 24 12:52:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (change), tool/change_maker.rb: make a brief template
+	  for ChangeLog.
+
+Mon May 24 09:19:59 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (gc_sweep): suppress a warning on VC.
+
+Mon May 24 08:16:02 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_f_spawn): use correct command name for the error
+	  message.  [ruby-dev:41395]
+
+Sun May 23 17:48:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/x509-internal.rb, lib/forwardable.rb,
+	  lib/irb/cmd/fork.rb, lib/mutex_m.rb,
+	  lib/shell/process-controller.rb, lib/sync.rb, object.c:
+	  suppress warnings patched by Benoit Daloze at [ruby-core:30366].
+
+	* parse.y (warn_balanced): no warning for singleton class.
+	  [ruby-core:30366]
+
+Sun May 23 16:56:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (CONST_ID_CACHE, rb_intern_const): suppress
+	  warnings with -Wconversion.
+
+Sun May 23 07:08:34 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/json/stream.rb: adding a JSON streaming API
+
+	* ext/psych/lib/psych/stream.rb: ditto
+
+	* ext/psych/lib/psych.rb: using autoload
+
+	* ext/psych/lib/psych/json.rb: ditto
+
+	* ext/psych/lib/psych/json/tree_builder.rb: refactor
+
+	* ext/psych/lib/psych/visitors/json_tree.rb: refactor
+
+Sat May 22 03:53:05 2010  Satoshi Shiba  <shiba@rvm.jp>
+
+	* cont.c (fiber_setcontext): Use swapcontext() instead longjmp().
+	  [ruby-dev:41316] [Bug #3295]
+
+Fri May 21 19:11:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_convert, rb_nkf_guess): check too huge
+	  string.
+
+Fri May 21 18:12:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_intern): C90 needs nonempty macro
+	  arguments.
+
+Fri May 21 13:55:46 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBYARG_SHARED): shared library will not be
+	  created unless enable-shared.
+
+	* ruby.c (ruby_init_loadpath_safe): use real path for non-shared
+	  build.
+
+Fri May 21 12:25:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_float): use dtoa directly instead of stripping
+	  needless trailing .0.
+
+	* numeric.c (flo_to_s): reverted.  [ruby-dev:41341]
+
+Fri May 21 01:06:05 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (subtract_tv): if the rest is zero, should finish waiting
+	  immediately.
+
+	* win32/win32.c (subtract): ditto.
+	  based on a patch from Roger Pack in [ruby-core:27957].
+
+Thu May 20 22:49:04 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* re.c (rb_reg_s_union_m): update rdoc.  [ruby-dev:41354]
+
+Thu May 20 22:08:28 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_f_caller): update rdoc.  a patch from Nobuhiro IMAI
+	  <nov at yo.rim.or.jp> in [ruby-dev:41348].
+
+Thu May 20 22:04:05 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (vm_backtrace_each): now takes an init function to distinguish
+	  an empty stack from out of stack.  [ruby-dev:41366]
+
+	* vm_eval.c (print_backtrace, rb_thread_backtrace): ditto.
+
+Thu May 20 20:47:46 2010  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): raise NoMethodError
+	  when COM method is not found. [ruby-core:30160] [Bug #3277]
+
+	* test/win32ole/test_win32ole.rb (test_no_method_error): ditto.
+
+Thu May 20 16:17:14 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/psych/emitter: C99(gcc)-ism.
+
+Thu May 20 12:59:49 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/stream.rb: adding YAML streaming API for
+	  infinite length streams.
+
+	* ext/psych/lib/psych.rb: refactoring for streaming API
+
+	* ext/psych/lib/psych/{handler, stream, tree_builder}.rb: ditto
+
+Thu May 20 02:12:20 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/emitter.c: output strings are automatically transcoded
+
+Wed May 19 23:36:57 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: [ruby-dev:41334] [Bug #3307] invalid result
+	  on searching tcl.h/tk.h. Thanks, Masaya Tarui.
+
+Wed May 19 23:19:30 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (disconnect): closes the socket of a Net::IMAP
+	  object only when it is not closed.  [ruby-dev:41350]
+
+Wed May 19 20:09:38 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_f_caller): return [] instead of nil when the function
+	  is called on toplevel.  [ruby-dev:41348]
+
+Wed May 19 19:58:01 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: mswin/mingw ruby has socketpair(), but it's
+	  not exist as such name in ruby static library, so mkmf.rb cannot
+	  find it.
+
+Wed May 19 19:45:10 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* iseq_compile_each (NODE_DEFINED): put nil first to fix stack
+	  consistency.  [ruby-core:30293]
+	  Now, lfinish[0] of defined_expr seems not to be used.  Refactoring
+	  may be needed.
+
+Wed May 19 16:55:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (rb_num2ulong): use rb_big2ulong for data from
+	  Bignum. Without this 32bit integer on 32bit environment
+	  can't converted into long.
+	  This fixes 1) and 2) of [ruby-dev:41289]
+
+Mon May 17 22:19:16 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* process.c: suppress warning for signed and unsigned type
+	  inconsistency.
+
+	* ext/psych/parser.c: ditto.
+
+	* ext/sdbm/_sdbm.c: ditto.
+
+	* ext/syck/rubyext.c: ditto.
+
+Mon May 17 21:30:50 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* .gitignore: updated.
+
+Mon May 17 21:08:53 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (argv_size): merged into join_argv() to maintain the
+	  agreement with it.  removed code has a calculation bug.
+	  fixed [Bug#2388]
+
+	* win32/win32.c (join_argv): calc and return the length of joined
+	  argv.
+	  the cause of the original bug was clarified by Masaya TARUI
+	  <tarui AT prx.jp> and the solution was suggested by him, too.
+
+Mon Apr  5 09:20:08 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/iso_2022_jp.h: add CP50220.
+
+	* enc/trans/iso2022.trans: add converter for CP50220.
+
+Mon May 17 09:37:25 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::Entry_#entries): returns pathname in
+	  UTF-8 on Windows to allow FileUtils accessing all pathnames
+	  internally.
+
+Mon May 17 01:07:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (rb_w32_check_imported): workaround for VC6.
+
+Sun May 16 22:21:32 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (mnew): initialize a field.  a patch from Takahiro Kambe.
+	  [ruby-dev:41312]
+
+Sun May 16 22:17:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval.c (setup_exception): let SystemStackError#backtrace return an
+	  single element array instead of string itself.  [ruby-core:30196]
+
+Sun May 16 21:51:04 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (parser_read_escape, parser_tokadd_escape): allow a hex or
+	  octal encoded character after \c.  This seemed to be prohibited at
+	  r13836, but its ChangeLog mentions nothing about this prohibition.
+	  So I assume this prohibition is not intended.  [ruby-core:27229]
+
+Sun May 16 21:14:04 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ruby/test_rubyoptions.rb (TestRubyOptions#test_segv_test):
+	  fixes "NoMethodError: undefined method `[]=' for nil:NilClass"
+
+Sun May 16 17:16:09 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (proc_binding): don't propagate filename and line_no of
+	  binding that is created from C level.  [ruby-dev:41322]
+
+	* vm_eval.c (eval_string_with_cref): ditto.
+
+Sun May 16 15:09:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (rb_w32_check_imported): check if extension library to be
+	  loaded imports from different ruby dll.
+
+Sun May 16 14:55:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (dln_find.o): fix dependency.
+
+Sun May 16 13:55:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_loaderror): use locale string, not ascii-8bit.
+
+Sun May 16 11:39:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/missing.h (signbit): add missing prototype.
+
+Sun May 16 10:49:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (rb_iconv_sys_fail): fix number of arguments.
+	  a patch by Masaya TARUI <tarui AT prx.jp>.
+
+Sun May 16 02:24:27 2010  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* configure.in: can not load libruby.so if gcc is used with
+	  --enable-shared on AIX.
+
+Sat May 15 17:16:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (rb_iconv_sys_fail): raise BrokenLibrary if
+	  errno is not set.  [ruby-dev:41317]
+
+Fri May 14 07:27:07 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_float): strip ".0" from end for rubyspec.
+
+Fri May 14 01:17:10 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_core.c (rb_thread_struct): add a field for sigaltstack.
+
+	* thread_pthread.c (thread_start_func_1): initialize machine stack
+	  information.
+
+	* thread.c (thread_start_func_2): set sigaltstack for each sub thread.
+	  [ruby-core:24540] [ruby-core:30207]
+
+Thu May 13 21:40:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* missing/ffs.c (ffs): fixed for non-zero values.
+
+Thu May 13 18:45:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (get_write_fd): return primary fd if no
+	  io is tied for writing.
+
+Thu May 13 18:15:33 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (swallow): should use more_char() instead of fill_cbuf().
+	  suggested by akr.
+
+Thu May 13 17:56:07 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_set_echo, console_echo_p): use
+	  primary fd.  [ruby-dev:41309]
+
+Thu May 13 13:30:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_float): use minimal representation.
+
+	* numeric.c (ruby_dbl2cstr): split from rb_float_new.
+
+Thu May 13 13:09:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_vm_get_sourceline): moved from eval_intern.h for
+	  vm_dump.c.
+
+Thu May 13 12:53:13 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (swallow): support text mode and UTF-16/32 as internal encoding.
+	  [Bug #1576]
+
+	* io.c (io_shift_cbuf): read and throw it away when str is NULL.
+
+Thu May 13 09:45:27 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub: link with ffs.obj.
+
+Thu May 13 07:37:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/sdbm/_sdbm.c: include unistd.h before sdbm.h for off_t.
+	  fix compilation problem on FreeBSD 6.4.
+
+Wed May 12 23:48:37 2010  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_quad_pack): deprecated.
+	  (rb_quad_unpack): ditto.
+
+Wed May 12 22:22:05 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (rb_big_abs_find_minbit): use ffs().
+
+	* configure.in: check ffs().
+
+	* missing/ffs.c: new file.
+
+	* include/ruby/missing.h (ffs): declared.
+
+Wed May 12 16:43:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_to_s): fixed broken output including nuls.
+
+Wed May 12 16:25:46 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_to_s): exponent needs 2 digits.
+
+Wed May 12 16:02:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_to_s): fill lower zeros.
+
+Wed May 12 15:45:36 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/generator/ri.rb (generate): should load existing cache
+	  before generating it.
+
+Wed May 12 15:04:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (BigMath#log): improved
+	  precision and performance. [ruby-dev:41295]
+
+Wed May 12 11:39:10 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: test all IPPROTO_* constants for recent Win32
+	  SDK.
+
+Wed May 12 10:57:04 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_to_s): make minimum string representation.
+	  [ruby-core:30145]
+
+Wed May 12 09:21:05 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_initialize_m): fix wrong index for the lang
+	  option's value 'N'. reported by Masaya TARUI via IRC.
+
+Tue May 11 23:07:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test IPPROTO_IP and IPPROTO_IPV6 constants.
+
+	* ext/socket/mkconstants.rb: define macros for enum.
+
+	  [ruby-dev:38849]
+
+Tue May 11 21:53:18 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/rake/test_task_arguments.rb: makes ENV empty during tests
+	  because ENV may change the behavior of Rake::TaskArguments.
+	  [ruby-core:29984]
+
+Tue May 11 15:14:30 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/utf_8.c: Add new alias UTF-8-HFS for UTF8-MAC.
+	  http://www.gnu.org/software/emacs/NEWS.23.2
+
+Tue May 11 13:46:00 2010  Kenta Murata <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (atan),
+	  test/bigdecimal/test_bigmath.rb (test_atan): explicitly specify
+	  the precision for calculating a reciprocal number of an argument.
+	  [Bug #3267]
+
+Tue May 11 11:49:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/sdbm/sdbm.h (DBM): large file support on win32.
+	  [ruby-core:23039]
+
+	* ext/sdbm/depend: objects depend on sdbm.h.
+
+Tue May 11 09:57:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (rb_vm_get_sourceline): add prototype.
+
+Tue May 11 09:53:07 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/psych/parser.c (PSYCH_TRANSCODE): get rid of bare use of gcc
+	  extension.
+
+Tue May 11 01:20:43 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (parse): Return strings encoded as
+	  Encoding.default_internal if set.
+
+Mon May 10 23:50:19 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_unpack): add a missing break.
+
+Mon May 10 14:13:04 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (rb_big_abs_find_minbit): get rid of a warning of VC.
+
+Mon May 10 13:59:42 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_has_cancel_io): new
+	  function.
+
+	* io.c (WAIT_FD_IN_WIN32): check only when it's not cancelable.
+
+Mon May 10 06:59:19 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (rdoc): rdoc should depend main. Moreover when XRUBY
+	  is miniruby, it can run after miniruby is built.
+
+Mon May 10 03:36:56 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (eval_string_with_cref): propagate filename and line_no
+	  of binding.  [ruby-dev:38767] [ruby-core:28307]
+
+	* vm_core.h (rb_binding_t), proc.c: add filename and line_no fields to
+	  preserve them.
+
+Mon May 10 02:58:33 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_compile_each), vm_insnhelper.c (vm_invoke_block,
+	  vm_throw): allow "return" and "yield" even in singleton class
+	  definition.  based on a patch from wanabe <s.wanabe AT gmail.com>
+	  for "return".  [ruby-core:21379] [ruby-dev:40975]
+
+	* insns.def (defineclass): ditto (straightforwardly push block ptr,
+	  instead of dfp ptr with special flag).
+
+	* vm_core.h (RUBY_VM_CLASS_SPECIAL_P): ditto (no longer needed).
+
+	* proc.c (proc_new): ditto (remove handling for special flag).
+
+	* bootstraptest/test_jump.rb: add tests for above.
+
+Mon May 10 02:29:51 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c (fiber_switch): raise FiberError when returning to dead
+	  fiber.  [ruby-dev:40833]
+
+Mon May 10 02:07:20 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/thread.rb (ConditionVariable#wait): ensure to remove the current
+	  thread from waiters.  [ruby-core:29835]
+
+Mon May 10 00:54:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_set_raw): new method.
+
+	* ext/io/console/console.c (ttymode): reverted previous commit.
+
+Sun May  9 23:53:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_GC_GUARD_PTR): get rid of removal by
+	  optimization.
+
+Sun May  9 23:07:53 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (proc_dup): copy blockprocval.  proc_dup is used by
+	  define_method, which made blockprocval be GC'ed mistakenly.
+	  [ruby-core:30023]
+
+Sun May  9 16:28:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (fiber_initialize_machine_stack_context): try to release
+	  unnecessary fibers and retry to create.  based on a patch from
+	  masaya tarui at [ruby-dev:41230].
+
+Sun May  9 08:32:56 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/generator/generator.c (fbuffer_inc_capa):
+	  fix the use of REALLOC_N. [ruby-dev:41227]
+
+Sun May  9 09:30:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/{generator/generator.c,parser/parser.rl}: fixed indent.
+
+	* ext/json/{generator,parser}/depend: added.
+
+Sun May  9 09:15:03 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/net/imap/test_imap.rb: use IPv4 for
+	  test_imaps_post_connection_check.  [ruby-dev:41189]
+
+Sun May  9 08:24:24 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (disconnect): terminates @receiver_thread even if
+	  @sock.shutdown raises an exception.  [ruby-dev:34881]
+
+Sun May  9 06:15:21 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* io.c (nogvl_copy_stream_sendfile): ISO C90 forbids mixed
+	  declarations and code.
+
+Sun May  9 02:57:02 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/lib/fiddle.rb: only require DL if it hasn't been required
+	  yet. [ruby-core:30095]
+
+Sun May  9 01:15:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (stackgrowdirection): removed duplicated code, use
+	  STACK_UPPER macro instead.
+
+	* gc.h (STACK_DIR_UPPER): moved from thread_pthread.c.
+
+Sun May  9 00:35:56 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/dl/test_base.rb (libc_so, libm_so): supports solaris.
+
+Sat May  8 19:03:31 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/default.mspec (config): . is no longer in $:.
+
+Sun May  9 00:27:46 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_aset): fixed argument type.
+
+Sat May  8 23:09:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* NEWS (io/console): IO#noecho and IO#raw without block are
+	  obsolete already.  [ruby-dev:41226]
+
+	* ext/io/console/console.c (ttymode): save duped file descriptors
+	  and restore tty modes using them, so that original modes can be
+	  restored even if original fds are closed.  [ruby-dev:41225]
+
+Sat May  8 13:48:31 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c (rb_ary_fetch, rb_ary_splice, rb_ary_store): Improve
+	  IndexError
+	  messages [ruby-core:28394]
+
+	* hash.c (rb_hash_fetch_m): Improve KeyError message
+
+Sat May  8 13:11:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/user_interaction.rb (Gem::StreamUI#ask_for_password):
+	  use io/console.
+
+	* ext/io/console: imported.  [ruby-dev:40897]
+
+Sat May  8 12:25:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_systmpdir): moved from ext/tmpdir.
+
+	* ext/etc/etc.c (etc_sysconfdir): added.
+
+	* lib/rubygems/config_file.rb, lib/tmpdir.rb: use etc.
+
+Sat May  8 11:07:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpAlloc): ensure buf does not get
+	  collected.  based on a patch masaya tarui at [ruby-dev:41213].
+
+Sat May  8 10:03:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (BasicSocket#connect_address): MacOS X 10.6
+	  returns "::ffff:a.b.c.d" for IPv4-mapped IPv6 address.
+	  [ruby-dev:41215] patch by Tomoyuki Chikanaga.
+
+Fri May  7 23:51:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* version.h: bump up to 1.9.3.
+
+Fri May  7 11:30:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (install-man): install mdocs directly without
+	  temporary files.  [ruby-dev:41204]
+
+Fri May  7 09:16:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/extconf.rb (get_tclConfig_dirs): glob with EXEEXT.
+
+	* ext/tk/extconf.rb (search_tclConfig): fixed typo.
+
+Fri May  7 06:45:28 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: search directories on PATH, only if containing
+	  tclsh or wish command (probably right fix for [ruby-core:30010]).
+
+Fri May  7 03:40:58 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpCtoV): fix to check overflow.
+
+Thu May  6 22:19:38 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* cont.c: define FIBER_USE_NATIVE only when _WIN32_WINNT >= 0x0400
+	  on Windows because Fiber APIs are supported only such building
+	  (and running) environments.
+	  [ruby-dev:41192]
+
+Thu May  6 19:13:43 2010  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/openssl/test_ec.rb: added test_dsa_sign_asn1_FIPS186_3. dgst is
+	  truncated with ec_key.group.order.size after openssl 0.9.8m for
+	  FIPS 186-3 compliance.
+
+	  WARNING: ruby-openssl aims to wrap an OpenSSL so when you're using
+	  openssl 0.9.8l or earlier version, EC.dsa_sign_asn1 raises
+	  OpenSSL::PKey::ECError as before and EC.dsa_verify_asn1 just returns
+	  false when you pass dgst longer than expected (no truncation
+	  performed).
+
+	* ext/openssl/ossl_pkey_ec.c: rdoc typo fixed.
+
+Thu May  6 18:12:43 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (fiber_setcontext): Fix last commit.
+
+Thu May  6 17:16:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (rb_localtime_r2): get rid of infinite loop bug of
+	  localtime() on x86_64-darwin.  [ruby-core:30031]
+
+Thu May  6 15:56:12 2010  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/*: Adding fiddle library to wrap libffi
+	* test/fiddle/*: testing fiddle extension
+	* ext/dl/lib/dl.rb: Requiring fiddle if it is available
+	* ext/dl/lib/dl/callback.rb: using Fiddle if it is available
+	* ext/dl/lib/dl/func.rb: ditto
+
+Thu May  6 15:04:37 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_match_m): add description about optional
+	  position parameter. [ruby-list:47064]
+	  patched by KISHIMOTO, Makoto <ksmakoto AT dd.iij4u.or.jp>
+
+Thu May  6 14:12:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb: test RUBY_PLATFORM for loading tmpdir.so to avoid
+	  LoadError.
+
+Thu May  6 12:41:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: add rationalize methods of Complex, Float, Integer, NilClass,
+	  and Rational classes.
+
+Thu May  6 12:12:26 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README: update.
+
+Thu May  6 03:34:29 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: apply FIBER_USE_NATIVE patch.  This patch improve
+	  Fiber context switching cost using system APIs.  Detail comments
+	  are written in cont.c.
+
+Thu May  6 02:16:48 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_unlink_method_entry, rb_sweep_method_entry):
+	  added.  Unlinked method entries are collected to
+	  vm->unlinked_method_entry_list.  On the GC timing, mark all method
+	  entries which are on all living threads.  Only non-marked method
+	  entries are collected.  This hack prevents releasing living method
+	  entry.
+	  [Performance Consideration] Since this Method Entry GC (MEGC)
+	  doesn't occur frequently, MEGC will not be a performance bottleneck.
+	  However, to traverse living method entries, every control frame push
+	  needs to clear cfp->me field.  This will be a performance issue
+	  (because pushing control frame is occurred frequently).
+	  Bug #2777 [ruby-dev:40457]
+
+	* cont.c (fiber_init): init cfp->me.
+
+	* gc.c (garbage_collect): kick rb_sweep_method_entry().
+
+	* method.h (rb_method_entry_t): add a mark field.
+
+	* vm.c (invoke_block_from_c): set passed me.
+
+	* vm.c (rb_thread_mark): mark cfp->me.
+
+	* vm_core.h (rb_thread_t): add a field passed_me.
+
+	* vm_core.h (rb_vm_t): add a field unlinked_method_entry_list.
+
+	* vm_insnhelper.c (vm_push_frame): clear cfp->me at all times.
+
+	* vm_insnhelper.c (vm_call_bmethod): pass me.
+
+	* bootstraptest/test_method.rb: add a test.
+
+Wed May  5 22:22:51 2010  wanabe  <s.wanabe@gmail.com>
+
+	* compile.c (iseq_set_sequence): fix check range of ic_index.
+	  a patch from Tomoyuki Chikanaga.  #3236
+
+Wed May  5 21:49:31 2010  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* dln.c (aix_loaderror): use execerror for load error on AIX.
+	  This can avoid SEGV reported by Perry Smith.  [Bug #2063]
+
+Wed May  5 20:37:54 2010  wanabe  <s.wanabe@gmail.com>
+
+	* thread_win32.c (w32_wait_events): get GVL before handle interrupt
+	  event.   [ruby-core:27199], [ruby-core:29698]
+
+Wed May  5 19:00:01 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: revert. enbugged by last commit.
+	  ([ruby-dev:41133], [ruby-dev:41134], [ruby-core:30010])
+
+	* ext/tk/README.tcltklib: ditto.
+
+Wed May  5 15:54:35 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_stat): use STAT macro instead of calling stat() directly.
+	  reported by Bill Kelly.  [ruby-core:30012]
+
+Wed May  5 11:43:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/test_io.rb (megacontent-copy_stream): get rid of
+	  deadlock.
+
+Wed May  5 07:27:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_mark_method_entry): renamed.
+
+Wed May  5 05:20:27 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h, vm_method.c: rename some internal functions related to
+	  rb_method_entry_t.
+	  rb_add_method_me()        -> rb_method_entry_set().
+	  rb_get_method_entry()     -> rb_method_entry_without_cache().
+	  rb_gc_mark_method_entry() -> rb_mark_method_entry().
+
+	* class.c, proc.c: ditto.
+
+Tue May  4 22:59:48 2010  wanabe  <s.wanabe@gmail.com>
+
+	* compile.c (iseq_build_body): update iseq->ic_size.
+	  a patch from Tomoyuki Chikanaga. see #3236.
+
+Tue May  4 18:51:22 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (name_err_mesg_to_str): preserve encoding of inspection.
+	  [ruby-core:29948]
+
+Tue May  4 18:38:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb (test_copy_stream):
+	  IO.copy_stream support binmode only currently.
+	  [ruby-core:23724]
+
+Tue May  4 12:46:09 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (argument_error): push correct backtrace.
+	  Bug #2281 [ruby-core:26333]
+
+Tue May  4 12:38:40 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (sockopt_inspect): use rb_str_cat2 and
+	  rb_str_append.
+
+	* ext/socket/ancdata.c (ancillary_inspect): refined.
+
+Tue May  4 11:48:06 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_inspect): use rb_str_cat2 and
+	  rb_str_append.
+
+Tue May  4 07:52:33 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/net/http.rb (Net::HTTPResponse#read_chunked): ensure to skip the
+	  last newline of chunk.  [ruby-core:29229]
+
+	* test/net/http/utils.rb: add an option for chunked response test.
+
+	* test/net/http/test_http.rb: add tests for chunked response.
+
+Tue May  4 03:37:54 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Update nkf 2010-04-28.
+
+Mon May  3 21:08:16 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* sample/README: update file list. [ruby-core:28981]
+
+Mon May  3 08:15:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/emitter.c (syck_scan_scalar): set SCAN_WHITEEDGE flag
+	  when scalar begins with newline.  patches from Dave B <daz AT
+	  d10.karoo.co.uk> at [ruby-core:23019] and caleb clausen at
+	  [ruby-core:25851].  [ruby-core:23006][ruby-core:29925]
+
+Sun May  2 17:52:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/task.rb (RDoc::Task): should not override newer code.
+
+Sun May  2 17:25:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): ignore signedness.
+
+Sun May  2 12:04:30 2010  wanabe  <s.wanabe@gmail.com>
+
+	* iseq.c (iseq_load): allow filepath to be nil.
+	  a patch from Tomoyuki Chikanaga in [Bug #3229].
+
+Sun May  2 08:54:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::NEEDS_BINMODE): check if O_BINARY value instead
+	  of fragile check by platform name.
+
+Sun May  2 07:38:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (dir_config): should not modify argument arrays
+	  themselves.
+
+Sat May  1 22:53:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#size): stat by path name when it is
+	  closed.
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  keep the first trace.
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  fixed filtering.  [ruby-core:29908]
+
+Sat May  1 16:46:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_IsInfinite): don't use
+	  non-ascii character.
+
+Sat May  1 14:08:29 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/server.rb:  Fix bug when serving gems.
+
+Sat May  1 14:05:36 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Upgrade to RDoc 2.5.8.
+
+Sat May  1 09:43:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  filter out traces in minitest directory.  [ruby-core:29908]
+
+Sat May  1 06:42:57 2010  wanabe  <s.wanabe@gmail.com>
+
+	* win32/win32.c: add declaration of wstati64 for gcc.
+
+Sat May  1 02:41:33 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* merge some patches from win32-unicode-test branch.
+	  see #1685.
+
+	* file.c, include/ruby/intern.h (rb_str_encode_ospath): new function
+	  to convert encoding for pathname.
+
+	* win32.c, include/ruby/win32.h (rb_w32_ulink, rb_w32_urename,
+	  rb_w32_ustati64, rb_w32_uopen, rb_w32_uutime, rb_w32_uchdir,
+	  rb_w32_umkdir, rb_w32_urmdir, rb_w32_uunlink): new functions to
+	  accept UTF-8 path.
+
+	* win32/win32.c (rb_w32_opendir, link, rb_w32_stati64, rb_w32_utime,
+	  rb_w32_unlink): use WCHAR path internally.
+
+	* file.c (rb_stat, eaccess, access_internal, rb_file_s_ftype,
+	  chmod_internal, rb_file_chmod, rb_file_chown, utime_internal,
+	  rb_file_s_link, unlink_internal, rb_file_s_rename): use UTF-8 version
+	  functions on Win32.
+
+	* file.c (apply2files, rb_stat, rb_file_s_lstat, rb_file_symlink_p,
+	  rb_file_readable_p, rb_file_writable_p, rb_file_executable_p,
+	  check3rdbyte, rb_file_identical_p, rb_file_chmod, rb_file_chown,
+	  rb_file_s_link, rb_file_s_symlink, rb_file_s_rename): call
+	  rb_str_encode_ospath() before passing the path to system.
+
+	* io.c (rb_sysopen): ditto.
+
+	* dir.c (dir_chdir, dir_s_mkdir, dir_s_rmdir): ditto.
+
+Sat May  1 00:26:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  assertion message must be String or Proc.  suggested by caleb
+	  clausen at [ruby-core:29884].
+
+Sat May  1 00:14:47 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_io.rb: skip false tests on Windows.
+	  [ruby-core:29886]
+
+Fri Apr 30 22:46:27 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* math.c: Math::DomainError < StandardError [ruby-core:29855]
+
+Fri Apr 30 21:40:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/test_open3.rb (TestOpen3#test_commandline): use simple
+	  command via shell.  [ruby-dev:41100]
+
+Fri Apr 30 15:38:45 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: better support for MinGW environment.
+
+Fri Apr 30 12:05:20 2010  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (file_expand_path): call rb_str_set_len before BUFCHECK to
+	  prevent rb_str_resize in BUFCHECK discard the content.
+
+Fri Apr 30 11:48:31 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* string.c (rb_str_sub_bang): String#sub! now raises an error when
+	  called on a frozen string, even if no change is made.
+	  See [ruby-core:23657]
+
+Fri Apr 30 11:40:44 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* eval.c (make_exception, rb_obj_extend): Fix error messages in case
+	  of wrong
+	  number of arguments
+
+	* file.c (rb_f_test, rb_file_s_umask): ditto
+
+	* numeric.c (int_chr, num_step): ditto
+
+	* process.c (rb_f_sleep): ditto
+
+	* re.c (rb_reg_initialize_m): ditto
+
+	* signal.c (rb_f_kill, sig_trap): ditto
+
+	* string.c (rb_str_aref_m, rb_str_aset_m, rb_str_count,
+	  rb_str_delete_bang,
+	  rb_str_slice_bang, rb_str_sub_bang, str_gsub): ditto
+
+	* proc.c (curry): rdoc fix
+
+Fri Apr 30 04:09:30 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (TIMET_MAX): should be time_t, not unsigned_time_t.
+
+Fri Apr 30 03:59:08 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (rb_time_unmagnify_to_float): should cast.
+
+Fri Apr 30 03:38:14 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (fix_mul): the width of fixnum is same as long's on all
+	  platforms.
+
+Fri Apr 30 03:17:20 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Improve algorithm for Matrix#determinant and
+	  Matrix#rank
+	  {determinant,det,rank}_e are now deprecated. [ruby-core:28273]
+	  Also fixes a bug in Determinant#rank (e.g. [[0,1][0,1][0,1]])
+	  Matrix#singular?, Matrix#regular? now raise on rectangular matrices
+	  and use determinant instead of rank.
+
+Fri Apr 30 00:52:56 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): define some constants to select
+	  collect code for win64.
+
+Thu Apr 29 20:10:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/test_open3.rb (test_commandline): use dump instead of
+	  shellwords.  [ruby-core:23797]
+
+Thu Apr 29 18:39:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/thwait.rb (ThreadsWait#join): refined rdoc again.
+	  [ruby-core:29863] [ruby-dev:41092]
+
+Thu Apr 29 14:46:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby: use more descriptive assertions.
+
+	* test/ruby: fixed nonsense assertions.
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  reverted.  [ruby-core:29872]
+
+Thu Apr 29 12:33:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/thwait.rb (ThreadsWait#join): refined rdoc.  [ruby-core:29863]
+
+Thu Apr 29 12:16:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert): reject
+	  non-boolean values.  [ruby-core:29868]
+
+Thu Apr 29 11:20:53 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (seek_before_access): renamed.  see [ruby-core:29861].
+
+	* io.c (rb_io_s_binread): fixed rdoc.  parts of a patch from Roger
+	  Pack in [ruby-core:29861].
+
+Thu Apr 29 03:50:49 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/xmlrpc/utils.rb: Fixing an incompatibility with
+	  reflection methods returning Symbols.  Patch by
+	  HD Moore.  [ruby-core:29821]
+
+Wed Apr 28 23:53:15 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/thwait.rb: revise rdoc.  a patch from Roger Pack in
+	  [ruby-core:27362].
+
+Wed Apr 28 23:21:42 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Update to JSON 1.4.2.
+
+Wed Apr 28 22:53:57 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread. (rb_thread_terminate_all): revert r24591 which caused SEGV
+	  sometimes.  [ruby-dev:40936] [ruby-core:27245] [ruby-core:27416]
+
+Wed Apr 28 22:26:55 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/psych/visitors/test_yaml_tree.rb (test_struct_const): remove
+	  Struct::Foo which affects test/ruby/test_struct to warn redefining
+	  constant.
+
+Wed Apr 28 18:04:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_timespec): use rb_check_funcall.
+
+Wed Apr 28 17:14:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/extconf.rb: CMSG_ macros are broken on 64bit darwin,
+	  because of use of __DARWIN_ALIGN.
+
+Wed Apr 28 16:44:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h ({RSTRING,RSTRUCT,RBIGNUM}_LENINT): check long
+	  to cast to int.
+
+Wed Apr 28 16:28:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/unixsocket.c (sendmsg_blocking, recvmsg_blocking):
+	  define only when used.
+
+	* ext/socket/rubysocket.h: non-implemented method definitions need
+	  to be shared.
+
+Wed Apr 28 10:27:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (ruby_vsnprintf, ruby_snprintf): suppress warnings.
+	  why return int in spite of the arguments of size_t?
+
+Wed Apr 28 09:49:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/compile_prelude.rb: enable optimization.
+
+Wed Apr 28 09:43:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (warn_balanced): do not warn for EXPR_ENDFN.
+
+Tue Apr 27 23:57:04 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval.c (ruby_cleanup): add exception guard around code running
+	  signal handler (of r27513).  Absence of the guard made
+	  bootstraptest/test_thread.rb fail sometimes.
+
+Tue Apr 27 22:55:29 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* insns.def (onceinlinecache): add exclusion control for a region
+	  between onceinlinecache and setinlinecache.  [ruby-dev:39768]
+
+Tue Apr 27 22:40:54 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* iseq.c (set_relation): do not use top_wrapper as bottom of cref,
+	  which caused constant lookup error when "wrapped load" is used;
+	  instead, use Object as bottom, and push top_wrapper on Object.
+	  [ruby-core:25039]
+
+Tue Apr 27 21:24:40 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval.c (ruby_cleanup): before cleanup, check signal buffer and run
+	  handler if any.  [ruby-core:20970]
+
+	* thread.c (rb_threadptr_check_signal): separated from
+	  timer_thread_function.
+
+Tue Apr 27 18:00:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/smtp.rb (Net::SMTP#rcptto_list): fixed typo.
+	  [ruby-core:29809]
+
+Tue Apr 27 12:44:23 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Import RDoc 2.5.7.  Fixes #1318 and ruby-core:29780.
+
+Tue Apr 27 10:54:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_read_escape): deny extra character escapes.
+	  [ruby-core:27228]
+
+Tue Apr 27 06:20:13 2010  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (select_internal): IO which cbuf is not empty is readable.
+
+Tue Apr 27 00:07:32 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (program): check void_expr when rb_parse_in_main().
+	  [ruby-dev:38717]
+
+Mon Apr 26 20:11:05 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c, rational.c, lib/cmath.rb, lib/date.rb lib/date/delta*:
+	  reverted r27484-27486.  now official spec.
+
+Mon Apr 26 15:42:59 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/json/generator/generator.c (convert_UTF8_to_JSON_ASCII): get rid
+	  of a warning.
+
+Mon Apr 26 13:11:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ripper_get_value): escape Qundef.
+
+Mon Apr 26 12:42:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (warn_balanced): warn according to last state.
+
+Mon Apr 26 09:05:30 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Update to JSON 1.4.1.
+
+Sun Apr 25 23:50:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (ruby_dup): update max fd.
+
+Sun Apr 25 21:56:49 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (fetch_internal): do not quote message data item
+	  names.  Thanks, Eric Hodel.  [ruby-core:23508]
+
+Sun Apr 25 15:59:02 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb: reverted experimental r23900.
+
+Sun Apr 25 15:51:00 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c, rational.c: reverted experimental r24565.
+
+Sun Apr 25 15:34:48 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/delta*: reverted experimental r24567 and
+	  r25393.
+
+Sun Apr 25 11:02:20 2010  wanabe  <s.wanabe@gmail.com>
+
+	* hash.c (ruby_setenv): putenv on msvcrt.dll can't remove empty value.
+
+Sat Apr 24 23:40:50 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb: use [""].pack("p").size to detect the pointer size.
+
+Sat Apr 24 21:07:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* test/psych/test_psych.rb: don't leave temporary files.
+
+Sat Apr 24 16:27:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (RUBY_PROGRAM_VERSION): added
+
+	* tool/rbinstall.rb: expand target file name.
+
+	* template/fake.rb.in: no need for hook if extmk.
+
+Sat Apr 24 14:40:20 2010  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: raise IOError when byte oriented operations occur with
+	  non-empty character buffer.
+	  [ruby-dev:40493] [ruby-dev:40506]
+
+Sat Apr 24 13:06:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (get_arglen): skip the last terminator of argv before
+	  checking environ.
+
+	* ruby.c (get_arglen): duplicate environ area if setenv and unsetenv
+	  are provided.
+
+	* ruby.c (set_arg0): keep empty strings.
+
+Sat Apr 24 09:44:40 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/yaml2byte.c (bytestring_append, bytestring_extend):
+	  removed wrong extra assignments.  a patch from Marcus Rueckert
+	  <darix AT opensu.se> at [ruby-core:29759].
+
+Sat Apr 24 00:41:52 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* process.c (rb_f_fork): remove after_exec() which sometimes caused
+	  two timer threads started.  [ruby-core:25217]
+
+	* signal.c: use pthread_sigmask() instead of sigprocmask().
+	  sigprocmask() is unspecified behavior on multi-thread programs.
+	  [ruby-core:25217]
+
+Sat Apr 24 00:36:05 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/dl/test_base.rb: add kfreebsd support.  based on a patch from
+	  Petr Salinger in [ruby-core:29769].
+	  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=560293
+
+Sat Apr 24 00:14:41 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl_session.c (SSL_SESSION_cmp): remove static for
+	  loopy version of openssl (0.9.8l?).
+
+Fri Apr 23 23:27:17 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_config.c: fix compilation failure with
+	  OpenSSL 0.9.8.
+
+Fri Apr 23 17:29:35 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_config.c: OpenSSL 1.0.0 support.
+	  [ruby-core:29256]
+
+	* ext/openssl/extconf.rb: check some functions added/removed at
+	  OpenSSL 1.0.0.
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_load): use engines which
+	  exists.
+
+	* ext/openssl/ossl_ssl_session (SSL_SESSION_cmp): removed at 1.0.0,
+	  so implement compatible function here.
+
+Fri Apr 23 14:37:22 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gem_prelude.rb (Gem::QuickLoader.load_full_rubygems_library):
+	  get rid of creating same regexps many times.
+
+	* lib/rubygems/custom_require.rb (Kernel#require): ditto.
+
+Fri Apr 23 14:07:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/etc.rd, doc/etc.rd.ja: moved from ext/etc.
+
+	* ext/etc.c (Init_etc): fixed rdoc.
+
+Fri Apr 23 11:31:25 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Update to RDoc 2.5.6.
+
+Fri Apr 23 04:11:26 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ripper/test_parser_events.rb (test_unterminated_regexp):
+	  separate test.
+
+Fri Apr 23 01:47:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_wopen): shouldn't use FILE_FLAG_OVERLAPPED with
+	  normal file.  #1807, #2510, #2549
+
+Fri Apr 23 01:28:03 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (ubf_handle): remove unused typedef.
+
+Thu Apr 22 23:10:59 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (mnew): fix wrong error message when Kernel#public_method
+	  receives name of private method.  [Bug #2425]
+
+Thu Apr 22 22:56:42 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bootstraptest/test_thread.rb: fix test.  [Bug #2342]
+
+Thu Apr 22 21:50:17 2010  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/ext/multi-irb.rb: fix multi-irb running
+	  parallelly. [ruby-dev:41031] [Bug #3182]
+
+Thu Apr 22 17:13:34 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems: Import RubyGems 1.3.7.pre.1 (as 1.3.6.1).
+
+Thu Apr 22 16:43:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (warn_balanced): removed false warning.
+
+	* parse.y (parser_yylex): skip whitespaces after method name.
+
+Wed Apr 22 02:16:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.[ch]: replace _DEBUG with BIGDECIMAL_DEBUG
+	  to prevent activating debugging code when compiled with -DEBUG flag
+	  on Windows. [ruby-core:22989] #1308
+
+Wed Apr 21 21:24:20 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/digest/sha2/extconf.rb: check SHA256_CTX and SHA512_CTX to fix
+	  compilation failure on OpenBSD 4.4.
+
+Wed Apr 21 15:13:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_getc): fix incomplete character immediately before EOF
+	  with newline converter. [ruby-dev:41024]
+
+Wed Apr 21 13:44:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (regexp): check each fragments.   [ruby-core:27374]
+
+Wed Apr 21 13:36:01 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): reverted r27388 due to backward
+	  compatibility.
+
+	* parse.y (parser_yylex): warn confusing binary operators after
+	  local variable.
+
+Wed Apr 21 11:53:47 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 2.5.5.  Fixes bugs in ri, Darkfish and
+	  rdoc option handling.
+
+Wed Apr 21 11:31:35 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): abandoned EXPR_VCALL.
+
+Wed Apr 21 03:17:17 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_getc): set read_encoding to resulted one character
+	  string. [ruby-dev:41023]
+
+Wed Apr 21 00:29:39 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (bigmul1_karatsuba): fix calculation order to prevent
+	  underflow.  [ruby-core:29088]
+
+Wed Apr 21 00:26:17 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (NODE_NEXT, NODE_REDO): add dummy putnil instruction to
+	  fix stack consistency error.  [ruby-core:28172]
+
+	* bootstraptest/test_jump.rb: add a test for above.
+
+Wed Apr 21 00:16:44 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/zlib/zlib.c (rb_deflate_params): update buf_filled count because
+	  deflateParams writes to out buffer.  And, revert r18029 because the
+	  flush was not needed now and emits garbage.  [ruby-dev:40802]
+
+Wed Apr 21 00:01:05 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_method.c (rb_add_method_def): decrement alias count of
+	  rb_method_definition_t to prevent memory leak.  see
+	  [ruby-dev:41021].
+
+Tue Apr 20 21:33:06 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/registry.rb (PredefinedKey#create): root key
+	  name should be a string.  fixed [ruby-core:28192]
+
+Tue Apr 20 19:25:50 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): fix for tLPAREN_ARG.
+
+Tue Apr 20 12:34:23 2010  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/{core, util}.rb (RFC822_DAYS, RFC822_MONTHS):
+	  move the constant because used only util.rb.
+
+Tue Apr 20 12:24:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lex_state_e, IS_END, IS_SPCARG, parser_yylex): separate
+	  the state after vcall.  [ruby-core:29578]
+
+Tue Apr 20 01:03:00 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bootstraptest/test_fork.rb: add a test for [ruby-core:28924].
+
+Tue Apr 20 00:43:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_extname): skip last directory separators.
+	  [ruby-core:29627]
+
+Tue Apr 20 00:41:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (internal_{read,write}_func, rb_{read,write}_internal):
+	  reverted r27265, since now rb_thread_blocking_region() preserves
+	  errno.
+
+Mon Apr 19 23:14:45 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_thread_blocking_region, rb_thread_blocking_region_end):
+	  preserve errno.  [Bug #2606] [ruby-core:28924]
+
+Mon Apr 19 19:41:10 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::cp_r): backport r11156 from ruby_1_8.
+
+Mon Apr 19 19:18:53 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): reduced duplicated conditions.
+
+Mon Apr 19 13:58:04 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 2.5.4.  Fixes #3169, #3160, #3023.
+
+Mon Apr 19 12:46:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout#timeout): propagate errors to the
+	  caller.  [ruby-dev:41010]
+
+Mon Apr 19 00:27:03 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_method.c (rb_add_method_def): workaround fix for redefinition
+	  of methods. This is because cfp->me uses the rb_method_entry
+	  which is freed by redefinition of remove_method. Note that
+	  reusing may cause another problem when the usage is changed.
+	  [ruby-core:27320] [ruby-core:29464]
+
+Sun Apr 18 22:13:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (string_content): cond_stack and cmdarg_stack are VALUE.
+
+Sun Apr 18 18:07:47 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::IPv4::Regex): make it only accept 0 to 255.
+	  [ruby-core:29501]
+
+Sun Apr 18 12:48:51 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): seems like a symbol-literal when spaces
+	  are unbalanced.  [ruby-core:29578]
+
+Sun Apr 18 12:32:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (string_content): preserve cond_stack and cmdarg_stack.
+	  [ruby-core:29579]
+
+Sun Apr 18 05:50:58 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: defaulting binary string
+	  format to "literal" format.
+
+Sun Apr 18 02:13:59 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (identify_identifier): add '_' to the regexp.
+
+Sat Apr 17 23:21:15 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: use AC_TRY_LINK instead of AC_TRY_COMPILE.
+	  On Solaris it can compile with signbit but can't link.
+
+Sat Apr 17 20:34:09 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/envutil.rb: update Gem::ConfigMap to fake environment for
+	  tests.  Some test of rubygems had failed because envutil.rb changed
+	  only RbConfig::CONFIG but not Gem::ConfigMap.
+
+Sat Apr 17 20:31:08 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/irb/ruby-lex.rb (identify_identifier): allow multibyte character
+	  as identifier.  [ruby-core:27275]
+
+Sat Apr 17 09:19:27 2010  wanabe  <s.wanabe@gmail.com>
+
+	* compile.c (iseq_compile_each): fix splat condition in NODE_WHEN.
+	  [Bug #2226]
+
+Sat Apr 17 08:57:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/smtp.rb (Net::SMTP#rcptto_list): continue when at least
+	  one RCPT is accepted.  based on a patch from Kero van Gelder at
+	  [ruby-core:26190].
+
+Sat Apr 17 07:43:55 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/thread/test_queue.rb (TestQueue#grind): fix typos.
+
+Sat Apr 17 07:27:53 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add Array#repeated_{combination,permutation} [Feature #2981]
+
+Sat Apr 17 05:30:22 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/core_ext.rb (yaml_as): supporting deprecated
+	  "yaml_as" method
+	* ext/syck/lib/syck/tag.rb: ditto
+
+Sat Apr 17 05:25:15 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/coder.rb (scalar): supporting deprecated methods
+	* ext/psych/lib/psych/deprecated.rb: supporting deprecated
+	  to_yaml_properties method
+
+Sat Apr 17 01:32:50 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (rb_io_rewind, rb_io_eof): add rdoc.  based on a patch from
+	  Roger Pack in [ruby-core:26771] [Bug #2377].
+
+Fri Apr 16 23:42:56 2010  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_folderitem2_invokeverb.rb (test_invokeverb):
+	  run test only when "Create Shortcut (&S)" menu is found.
+	  [ruby-core:29550] [Bug #1602]
+
+Fri Apr 16 21:52:16 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (string_content, etc): counts of CMDARG_PUSH and POP were
+	  inconsistent.  [ruby-core:22637]
+
+Fri Apr 16 21:27:57 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_coerce): revert r23389,
+	  which not only is unnecessary but also causes the inconsistency of
+	  return type.  [ruby-core:25706] [Bug #2129]
+
+Fri Apr 16 20:05:24 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/thread.rb (Queue#push, #pop, SizedQueue#push, #pop): remove
+	  code that kicks waiting thread twice, which caused race and
+	  deadlock.  a patch from James M. Lawrence.  [ruby-core:25537]
+
+	* test/thread/test_queue.rb: added.
+
+Fri Apr 16 20:01:47 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* .gitignore: updated.
+
+Fri Apr 16 16:51:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/file2lastrev.rb (VCS::SVN::get_revisions.): remind about
+	  DOSISH, sometimes.
+
+Fri Apr 16 16:15:40 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_repeated_permutation): new method added. a patch
+	  from Makoto Kishimoto in [ruby-core:29267]   [ruby-core:28724]
+
+	* array.c (rb_ary_repeated_combination): ditto.
+
+Thu Apr 15 22:41:47 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_thread_priority, rb_thread_priority_set): fix rdoc.
+	  Lower-priority thread may run even if there are higher-priority
+	  threads.  See [ruby-dev:40977].
+
+Thu Apr 15 22:33:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/uri/ftp.rb (URI::FTP#set_path): added to correct handling of
+	  special case where path of ftp is relative.  This converts relative
+	  path to absolute one, because external representation of ftp path is
+	  relative and internal representation is absolute.  [ruby-core:24077]
+
+	* lib/uri/ftp.rb (URI::FTP#initialize): converts absolute to relative.
+
+	* lib/uri/generic.rb (URI::Generic#check_path): allow relative path
+	  when scheme is ftp.
+
+Thu Apr 15 21:54:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/file2lastrev.rb: use backtick for ruby 1.8.
+
+Thu Apr 15 21:13:29 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/file2lastrev.rb: this should run with ruby 1.8.
+
+Thu Apr 15 20:41:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/file2lastrev.rb: make -q effective for files not version
+	  controlled.
+
+Thu Apr 15 14:50:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_backtrace_each): get rid of use of malloc from signal
+	  handler by using ruby_engine_name.  [ruby-core:29497]
+
+	* vm_eval.c (print_backtrace): file may be nil when segfaulted in
+	  very early stage.
+
+	* vm_dump.c (bugreport_backtrace): ditto.
+
+Thu Apr 15 11:51:49 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (help): small fix.
+
+Wed Apr 14 22:09:28 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rexml/source.rb: force_encoding("UTF-8") when the input
+	  is already UTF-8. patched by Kouhei Sutou [ruby-core:23404]
+
+Wed Apr 14 18:23:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* configure.in (signbit): signbit is a macro in C99.
+
+Wed Apr 14 17:56:06 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_shift_cbuf): should terminate.
+	  http://golf.shinh.org/p.rb?DECimal+number+to+BINary#Ruby
+
+Wed Apr 14 16:29:02 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (GET_STACK_BOUNDS): fixed macro argument names.  a patch from
+	  Satoshi Shiba <shiba AT rvm.jp> at [ruby-dev:40973].
+
+	* gc.h (ruby_get_stack_grow_direction): fixed prototype.
+
+	* thread_pthread.c (get_stack, ruby_stack_overflowed_p): both side
+	  should be same type.  [Bug #3145]
+
+Wed Apr 14 15:58:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_string_value_cstr): make NUL terminated if it is
+	  not done.
+
+Wed Apr 14 12:56:21 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap (encode_utf7): encode & properly.  Thanks, Kengo
+	  Matsuyama.  [ruby-dev:38063]
+
+Wed Apr 14 11:39:26 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/lib/syck/basenode.rb (Syck::BaseNode#match_path): fixed
+	  library name.
+
+	* ext/syck/lib/yaml/syck.rb: split from ext/syck/lib/syck.rb for
+	  backward compatibility.
+
+Wed Apr 14 09:24:07 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Refix [nkf-bug:21393]. (explicit -X)
+
+Tue Apr 13 20:36:50 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Fix: Windows Codepages wrongly convert
+	  Halfwidth Katakana. [nkf-bug:21393]
+
+Tue Apr 13 15:00:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (make_seed_value): fix leading-zero-guard condition on
+	  bdigit is smaller than 32bit.
+
+Tue Apr 13 13:57:36 2010  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/completion.rb (CompletionProc): fix [ruby-dev:40953]
+	  [Bug #3126]. Thanks Kazuhiro NISHIYAMA.
+
+Tue Apr 13 09:32:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (transcode_loop): insert output the value when
+	  fallback hash has a related key. [ruby-dev:40540]
+	  [ruby-dev:40829] #3036
+
+	* transcode.c (rb_econv_prepare_opts): pass to newhash
+	  a value with the key :fallback.
+
+Tue Apr 13 00:12:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (rand_init): use the absolute value of seed to
+	  make srand(-2**40) portable with 32bit and 64bit.
+	  [ruby-core:29292](2)
+
+Mon Apr 12 22:48:47 2010  Tanaka Akira  <akr@fsij.org>
+
+	* enumerator.c (inspect_enumerator): don't raise for uninitialized
+	  enumerator.
+
+Mon Apr 12 21:47:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* LEGAL: separated the section for parse.c.  contributed by Paul
+	  Betteridge in [ruby-core:29472].
+
+Mon Apr 12 09:19:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (fu_each_src_dest): ensure src is accessible.
+
+	* lib/fileutils.rb (fu_same): use File.identical? to get rid of
+	  exceptions.  [ruby-core:28141]
+
+	* lib/fileutils.rb (fu_have_st_ino): no longer used.
+
+	* lib/fileutils.rb (fu_have_st_ino): check if required method is
+	  defined, instead of platform name.
+
+Mon Apr 12 05:10:20 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: New method Matrix.build [ruby-core:28272]
+
+Mon Apr 12 03:45:25 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Deprecate elements_to_{f/i/r}
+
+Mon Apr 12 03:30:29 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Handle coercion errors by raising TypeErrors
+	  [ruby-core:26736]
+
+Mon Apr 12 00:51:21 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* io.c (rb_io_getline_fast, rb_io_getline_1): fix ARGF.lineno
+	  behavior. [ruby-core:25205]
+
+	* test/ruby/test_argf.rb (TestArgf#test_lineno3): add a test for
+	  above.
+
+Sun Apr 11 23:25:17 2010  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* configure.in: can not load libraries if --with-opt-dir is used
+	  on AIX. [Bug #3107]
+
+Sun Apr 11 14:44:45 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/irb/ruby-lex.rb (RubyLex#getc): revert r10817, which caused
+	  regression for handling case when there is no newline at the end of
+	  file.  [ruby-core:28248]
+
+Sun Apr 11 12:08:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_replicate): new encoding name must be valid
+	  c-string.  [ruby-dev:40954]
+
+Sun Apr 11 11:58:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (ftruncate): rename to get rid of conflict
+	  with same name but non-LFS function on MinGW.  [ruby-core:24757]
+
+Sun Apr 11 10:33:34 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 2.5.3.  Includes r27288 and r27290.
+
+Sun Apr 11 09:31:39 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* test/syck/*: Moved test/yaml to test/syck since it's actually
+	  testing the syck YAML engine.
+
+Sun Apr 11 08:56:44 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (setup_output_dir): compare by Time#to_i.
+	  Cached created time doesn't have fractional times,
+	  but FileStat#mtime has; so rdoc almost always judged files
+	  are updated.
+
+Sun Apr 11 07:40:48 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (setup_output_dir): fix typo. [ruby-core:29415]
+
+Sun Apr 11 07:01:41 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (decode_www_form_component): validate.
+	  [ruby-dev:40938]
+
+	* lib/uri/common.rb (decode_www_form): allow empty string.
+
+	* lib/uri/common.rb: fix nodoc for constant. [ruby-dev:40949]
+
+Sat Apr 10 21:26:22 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/markup/preprocess.rb (RDoc::Markup::PreProcess#handle):
+	  use File.binread to avoid locale dependency.
+
+Sat Apr 10 15:18:26 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc: Import RDoc 2.5.2
+	* lib/rdoc/parser/ruby.rb (RDoc::Parser::Ruby): Don't parse rdoc
+	  files, reverts r24976 in favor of include directive support in C
+	  parser.
+
+Sat Apr 10 13:14:22 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): add SSLContext#ssl_timeout=,
+	  which allows net/https to specify timeout.  [ruby-core:24812]
+
+Sat Apr 10 07:36:13 2010  wanabe  <s.wanabe@gmail.com>
+
+	* vm_method.c (rb_add_method): clear attr.location before rb_ary_new3.
+	  a patch from Masaya TARUI in [Bug #3108]
+
+Sat Apr 10 05:27:26 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/deprecated.rb: implementing Psych.quick_emit and
+	  adding deprecation warnings.
+	* ext/psych/lib/psych/visitors/to_ruby.rb: supporting deprecated
+	  yaml_initialize api.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: supporting deprecated
+	  to_yaml api.
+
+Sat Apr 10 01:08:53 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c: parse raises a TypeError when nil is passed in.
+
+Sat Apr 10 00:01:51 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/pty/lib/expect.rb: preserve buffer read, instead of discard.
+	  based on a patch from Luiz Angelo Daros de Luca in
+	  [ruby-core:23464].
+
+Fri Apr  9 23:58:58 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/pty/lib/expect.rb: raise an error when argument is unexpected
+	  type.  based on a patch from Luiz Angelo Daros de Luca in
+	  [ruby-core:23464].
+
+Fri Apr  9 23:57:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/pty/lib/expect.rb: add rdoc.  based on a patch from Luiz Angelo
+	  Daros de Luca in [ruby-core:23464].
+
+Fri Apr  9 23:54:16 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/irb/init.rb, lib/irb/lc/help-message,
+	  lib/irb/lc/ja/help-message: add -w flag.  [ruby-core:24594]
+
+Fri Apr  9 23:51:45 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal),
+	  ext/socket/unixsocket.c (rsock_init_unixsock): check the result of
+	  listen(2).  based on a patch from Mike Pomraning.  [ruby-core:23698]
+
+Fri Apr  9 21:22:10 2010  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/completion.rb (CompletionProc): irb will be stuck with
+	  long variable name at completion. [Bug#1969]. refix [ruby-core:28366].
+
+Fri Apr  9 20:54:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (decode_www_form): don't ignore leading '?'.
+	  [ruby-dev:40938]
+
+	* lib/uri/common.rb (decode_www_form): check whether argument is
+	  valid application/x-www-form-urlencoded data.
+
+Fri Apr  9 20:29:13 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* dir.c (push_glob): clear up the previous commit (RB_GC_GUARD can
+	  precede the last reference).  [ruby-dev:40942]
+
+Fri Apr  9 20:05:05 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* dir.c (push_glob): add RB_GC_GUARD to prevent str being free'ed.
+	  a patch from Masaya TARUI in [ruby-dev:40939].  [ruby-dev:40514]
+
+Fri Apr  9 10:53:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* compile.c (iseq_compile_each): stop hiding, and freeze unpopped
+	  string nodes to allow method redefinition.  [ruby-dev:40641]
+
+Fri Apr  9 01:26:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (internal_{read,write}_func, rb_{read,write}_internal):
+	  preserve errno.  a patch from Takehiro Kubo in [ruby-core:29340].
+	  [ruby-core:28924]
+
+Fri Apr  9 01:12:07 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/irb/completion.rb (CompletionProc): calling the method "methods"
+	  may raise an exception.  [ruby-core:28366]
+
+Thu Apr  8 14:33:24 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* configure.in: --disable-install-doc should prohibit doxygen.
+
+Thu Apr  8 14:23:51 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h: include errno.h before defining errnos.
+	  see [ruby-core:29314]
+
+Thu Apr  8 11:45:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/yaml.rb: use Psych if only it is loaded already.
+
+	* lib/yaml.rb (YAML::EngineManager#yamler): set the result after
+	  successfully switched.
+
+Thu Apr  8 07:22:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rand_init): ignore higher bits if all they are same as
+	  the lower sign bit.  [ruby-core:29292](2)
+
+Thu Apr  8 07:16:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb/cmd/help.rb (IRB::ExtendCommand::Help#execute): use RI
+	  interactive mode if no argument.  [ruby-dev:39839]
+
+Wed Apr  8 02:33:55 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c (rb_ary_permutation): Remove limitation for lengthy permutations
+	  [ruby-core:29240]
+
+	* test/ruby/test_array.rb: ditto
+
+Wed Apr  7 23:33:55 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* misc/ruby-mode.el (ruby-mode-map): binded C-c C-c and C-c C-c C-u
+	  to comment-region and uncomment-region. Thanks Michael Klishin!
+	  [Feature #872] [ruby-core:20552]
+
+Wed Apr  7 21:44:14 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (ASCTIME): unused macro removed.
+
+Wed Apr  7 16:45:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (Makefile): reverted r27244.
+
+Wed Apr  7 11:49:34 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rexml/quickpath.rb (REXML::QuickPath::predicate):
+	  fix regexp.
+
+Wed Apr  7 09:44:48 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (Makefile): quote init-commands.
+
+	* configure.in (Makefile): `make ruby` builds ruby command
+	  even if RUBY_INSTALL_NAME is customized or it has extension.
+
+Wed Apr  7 03:28:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (RDoc#{initialize,setup_output_dir}):
+	  initialize @last_created and use it.
+
+Wed Apr  7 02:05:41 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in: not define $extout to make target in cwd.
+
+Tue Apr  6 23:01:35 2010  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h: Introduce UNINITIALIZED_VAR() macro.
+	* thread.c (do_select): Use UNINITIALIZED_VAR() instead FAKE_FD_ZERO().
+	  Also, remove FAKE_FD_ZERO completely. [Feature #3018]
+
+Tue Apr  6 14:53:17 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h: check definition existence before defining
+	  errno macros.
+
+	* win32/win32.c (errmap): define winsock errors mappings.
+	  these are VC++10 support. see [ruby-core:29278]
+
+Tue Apr  6 21:55:25 2010  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: test localtime(3) overflow.  [ruby-dev:40910]
+
+	* time.c (rb_gmtime_r): renamed from rb_gmtime.
+	  (rb_localtime_r): renamed from rb_localtime.
+	  (rb_localtime_r2): call rb_localtime_r and validate the result if
+	  there is overflow problem.
+	  (rb_gmtime_r2): call rb_gmtime_r and validate the result if there
+	  is overflow problem.
+
+Tue Apr  6 11:21:23 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in: hooks for extconf.rb.
+
+Tue Apr  6 06:19:36 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rexml/text.rb (REXML::Text.check): comment out
+	  broken logic.
+
+Tue Apr  6 05:59:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rexml/parsers/baseparser.rb (LETTER, DIGIT):
+	  always use POSIX charclass.
+
+	* lib/rexml/parsers/baseparser.rb (NAMECHAR):
+	  remove duplicated range.
+
+	* lib/rexml/xmltokens.rb (NCNAME_STR, NAMECHAR): ditto.
+
+	* lib/rexml/parsers/xpathparser.rb (PathExpr): ditto.
+
+	* lib/rexml/text.rb (REXML::Text#initialize):
+	  initialize @parent.
+
+Mon Apr  5 19:54:58 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (make_regexp): use onig_new_with_source to keep
+	  sourcefile and sourceline.
+
+	* re.c (onig_new_with_source): copied from onig_new in
+	  regcomp.c for keep sourcefile and sourceline.
+
+Mon Apr  5 13:20:45 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h, win32/win32.c (EWOULDBLOCK): VC10 already
+	  defined EWOULDBLOCK.  based on a patch from Charlie Savage,
+	  see [ruby-core:29255]
+
+Mon Apr  5 13:10:35 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/psych/lib/psych/core_ext.rb: remove before alias.
+
+	* ext/syck/lib/syck.rb: don't warn called by itself.
+
+Mon Apr  5 13:08:24 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/mkexports.rb (each_export): VC10 support.
+	  a patch from Charlie Savage in [ruby-core:29254]
+
+Sun Apr  4 22:18:32 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (cmp): optimize for fixnums.
+	  (lt): use cmp.
+	  (gt): ditto.
+	  (le): ditto.
+	  (ge): ditto.
+	  (wlt): use wcmp.
+	  (wgt): ditto.
+	  (wle): ditto.
+	  (wge): ditto.
+	  (time_subsec): use wmod.
+
+Sun Apr  4 10:04:28 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h: replace snprintf and vsnprintf by
+	  ruby_snprintf and ruby_vsnprintf. [ruby-dev:40909]
+
+	* configure.in: don't check snprintf(3) and vsnprintf(3).
+
+	* include/ruby/missing.h: don't declare snprintf and vsnprintf.
+
+	* include/ruby/ruby.h: include stdarg.h.
+
+	* sprintf.c: include vsnprintf.c
+
+	* vsnprintf.c: renamed from missing/vsnprintf.c.
+
+	* vsnprintf.c: remove useless ifdefs.
+
+	* win32/mkexports.rb: use ruby_snprintf and ruby_vsnprintf.
+
+Sun Apr  4 09:44:01 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (lt): call <=> instead of <.
+
+Sun Apr  4 06:46:16 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* lib/yaml: Moved to ext/syck/lib, Syck only uses Syck constant.
+	* lib/yaml.rb: Added an engine manager for choosing YAML engine.
+	* ext/syck/lib/syck/rubytypes.rb: squashed warnings when using Psych
+
+Sat Apr  3 22:47:51 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: don't define STRUCT_WIDEVAL.
+
+Sat Apr  3 18:15:02 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (RDoc#{setup,update}_output_dir): store mtimes
+	  per files in the flag file.
+
+	* lib/rdoc/rdoc.rb (RDoc#normalized_file_list): skip already
+	  processed and unmodified files.
+
+Sat Apr  3 15:09:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (rdoc): no force-update, and add RDOCFLAGS.
+
+	* lib/rdoc/options.rb (RDoc#parse): no-force-update option.
+
+Sat Apr  3 10:51:10 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_rand): raise ArgumentError on nil, as the
+	  documentation implies.  [ruby-core:29075]
+
+	* random.c (rb_f_rand): mentioned the case of when max is nil.
+
+Sat Apr  3 06:56:11 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c (rb_ary_product): Accept a block [ruby-core:29045]
+
+Sat Apr  3 04:46:00 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/file2lastrev.rb (VCS#relative_to): path and @srcdir may have
+	  different relative-ness.
+
+Sat Apr  3 03:19:01 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/benchmark.rb (Benchmark::Tms#add): fix NameError.
+	  [ruby-dev:40906]
+
+	* test/benchmark/test_benchmark.rb: add a test for above.
+
+Sat Apr  3 01:50:02 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parser/ruby.rb (RDoc#make_message): no empty lines.
+
+	* lib/rdoc/stats.rb (RDoc::Normal#print_file): send to stdout and
+	  flush always
+
+Sat Apr  3 00:03:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/class_module.rb (RDoc#merge): get rid of stack overflow.
+
+	* lib/rdoc/parser/c.rb (RDoc#handle_method): see source files in
+	  source directory.
+
+Fri Apr  2 22:13:38 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb: reverted r27186 to add generators.
+
+Fri Apr  2 21:17:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* NEWS: added reverse_lookup option.
+
+Fri Apr  2 15:26:14 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_ri_driver.rb (with_dummy_pager): don't grasp
+	  the terminal.
+
+Fri Apr  2 14:43:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (RDoc::Generator): defer by autoload.
+
+	* lib/rdoc/generator/darkfish.rb: use normal library instead of
+	  gem.
+
+Fri Apr  2 14:26:26 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc#parse_files): don't branch by
+	  RUBY_VERSION.  reapplied r24990.
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc#parse_files): emacs local variables
+	  are delimited by a semicolon.  supported Vim style.  reapplied
+	  r24988 with fixing a typo for shebang.
+
+Fri Apr  2 14:16:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parser/ruby.rb (RDoc::Parser::Ruby): parse also rdoc
+	  files.  reapplied r24976.
+
+	* lib/rdoc/parser/c.rb: fixed a small error in the documentation.
+	  [ruby-core:24744]  reapplied r24395.
+
+	* lib/rdoc/parser.rb (RDoc::Parser.binary?): blksize may be nil
+	  and is irrelevant to whether a file is binary.  TAB and newlines
+	  would be usually considered to be included in text data.
+	  reapplied r23071 and r24297.
+
+Fri Apr  2 13:59:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/paths.rb (RDoc::RI::Paths): Gem::Enable has been
+	  obsolete.  reapplied r24599.
+
+	* lib/rdoc/ri/paths.rb: follows $ridir.  reapplied r23554.
+
+	* lib/rdoc/ri/paths.rb (RDoc::HOMEDIR): expand_path deals with
+	  platform dependent environments.  reapplied r21312.
+
+Fri Apr  2 12:52:25 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/parser.rb (binary?): regression.
+
+Fri Apr  2 11:54:49 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (init_env): wrong calculation of the head of setenv
+	  string.  [ruby-core:29216]
+	  a patch from Heesob Park in [ruby-core:29218]
+
+Fri Apr  2 08:24:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (ruby_setsid): split from proc_setsid.
+
+	* process.c (proc_daemon): use EXIT_SUCCESS instead of magic number.
+
+Fri Apr  2 07:00:28 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cptr.c (rb_dlptr_inspect): fix format specifiers.
+
+Fri Apr  2 05:38:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (InitVM): calls per-VM initialization in
+	  single VM, but does nothing in MVM.
+
+Fri Apr  2 02:56:56 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: New instance methods:
+	  empty? [ruby-core:26284], each, each_with_index [ruby-core:28400],
+	  conjugate, conj, imaginary, imag, real, real?, rectangular, rect
+	  [ruby-core:26285]
+
+	  Removed compare_by*, inspect_org, cf [ruby-core:26268]
+
+	  Matrix.empty: raise on negative sizes
+
+	  Matrix.determinant: raise on rectangular matrices [ruby-core:28271]
+
+Thu Apr  1 17:17:00 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/iso2022.trans: CP50221 supports 8bit JIS.
+
+Thu Apr  1 16:44:00 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc, test/rdoc: Imported RDoc 2.5
+
+Thu Apr  1 14:30:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/utf_16{be,le}.c (utf16{be,le}_mbc_to_code): simplified.
+
+Thu Apr  1 14:07:51 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (BSD__hdtoa): don't use C99 macros. (FP_NORMAL etc)
+
+Thu Apr  1 13:30:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): support %a format. [ruby-dev:40650]
+
+	* missing/vsnprintf.c (BSD_vfprintf): ditto.
+
+	* missing/vsnprintf.c (cvt): ditto.
+
+	* util.c (BSD__hdtoa): added.  This is 2-clause BSDL licensed
+	  by David Schultz and from FreeBSD.
+
+	* LEGAL: add about hdtoa() in util.c.
+
+Thu Apr  1 13:24:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): return 0.0 if hexadecimal and
+	  badcheck is FALSE: Float("0x1p+0") works, but "0x1p+0".to_f
+	  doesn't. [ruby-dev:40650]
+
+	* util.c (ruby_strtod): allow hexadecimal integers.
+
+Thu Apr  1 13:20:50 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/scanf.rb: support %a format. [ruby-dev:40650]
+
+Thu Apr  1 12:04:10 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRE_TIMET_PREFIX): define if configure doesn't
+	  defines it.  [ruby-core:29190]
+
+Thu Apr  1 11:36:04 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_initialize): fixed a rdoc miss that textmode and
+	  binmode flag explanation was reversed each other.  a patch from
+	  Heesob Park in [ruby-core:29166].
+
+Thu Apr  1 09:54:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RUBY_{DEFAULT,NEVER}_FREE): added.
+	  [ruby-dev:40889]
+
+Thu Apr  1 04:55:08 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (rb_big_abs_find_minbit): fix handling the last word.
+
+Thu Apr  1 04:30:11 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (long_mul): overflow condition changed.
+	  (wi_mul): ditto.
+
+Wed Mar 31 23:33:29 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (long_mul): extracted from mul and avoid integer overflow.
+	  (wi_mul): extracted from wmul and avoid integer overflow.
+
+Wed Mar 31 21:30:38 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: less bignum allocations.
+
+	* strftime.c (rb_strftime_timespec): defined to avoid rational for
+	  nano second resolution time.
+
+Wed Mar 31 16:05:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rake/test_{package_task,rules,task_manager}.rb: use
+	  temporary directories.
+
+Wed Mar 31 13:53:13 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rubygems/gemutilities.rb (setup, teardown):
+	  Object#to_yaml must use syck on RubyGem.
+
+Wed Mar 31 06:57:14 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (timew2timespec): fix previous change.
+
+Wed Mar 31 00:32:19 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (wdivmod): renamed from wdivmodv.  fix neg/pos and pos/neg.
+	  (split_second): return number of seconds as wideval_t.
+	  (timet2wv): new function.
+	  (wv2timet): new function.
+	  (timet2wideval): removed.
+
+Tue Mar 30 22:24:19 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: use more polymorphic operations.
+
+Tue Mar 30 22:12:38 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add URI.decode and URI.encode to Obsoleted methods,
+	  and fix typos. [ruby-dev:40852]
+
+	* NEWS: add Time#subsec.
+
+Tue Mar 30 22:07:12 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: reorder wide value implementation.
+
+Tue Mar 30 22:00:44 2010  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb.rb: support [Feature#988], [ruby-core:21177].
+	* lib/irb/context.rb: fix Context#verbose?.
+
+Tue Mar 30 21:56:33 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: make wide value more polymorphic.
+
+Tue Mar 30 21:18:32 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/psych/extconf.rb: install psych .so to not psych/ but direct.
+	  This prepends to install *.rb files to psych/psych.
+
+	* ext/psych/lib/psych.rb: require 'psych.so'.
+
+Tue Mar 30 20:55:14 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (w2v): renamed from w2xv.
+	  (v2w): renamed from xv2w.
+	  (rb_time_magnify): takes wideval_t.
+	  (rb_time_unmagnify): returns wideval_t.
+
+Tue Mar 30 19:46:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/psych/lib: moved external library dependent files from lib.
+	  [ruby-core:29129]
+
+Tue Mar 30 18:25:52 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_dirname): split from rb_file_s_dirname.
+
+	* load.c (rb_f_require_relative): use absolute path instead of
+	  expanded path.
+
+Tue Mar 30 13:57:08 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* process.c (proc_daemon): Process.daemon should raise an error on
+	  failure regardless of whether the implementation uses daemon(3)
+	  or not. [ruby-dev:40832]
+
+Tue Mar 30 13:11:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/file2lastrev.rb (VCS::GIT_SVN): removed because git-log can
+	  deal with git-svn repository faster than git-svn-info.
+
+	* tool/file2lastrev.rb (VCS#get_revisions): particular commands do
+	  not depend on instance.
+
+Tue Mar 30 08:55:50 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/extconf.rb: Making library detection more agnostic.
+
+Tue Mar 30 08:10:59 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* lib/psych.rb: Fix problem with empty and white-space only strings.
+	  Thanks Peter McLain!
+	* test/psych/test_psych.rb: tests for change.
+
+Tue Mar 30 05:31:39 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* lib/psych.rb: documentation updates.  Thanks Peter McLain!
+	* lib/psych/{coder,tree_builder}.rb: ditto
+
+Tue Mar 30 03:56:13 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/psych/{emitter,parser,psych}.c: move variable
+	  declaration to the first of the block.
+
+Mon Mar 29 21:47:44 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* prelude.rb, .document: Stuff in prelude.rb should be documented
+	  as well.
+
+Mon Mar 29 20:23:05 2010  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/matrix.rb(Vector#each2, Vector#collect2): add type check for
+	  Integer[Bug #2495].
+
+Mon Mar 29 19:45:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: wide value condition changed.
+
+Mon Mar 29 18:10:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (Numeric): accept rationals.
+
+Mon Mar 29 15:10:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#open): re-open with same mode and
+	  options as initialize.
+
+Mon Mar 29 09:16:45 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c: change include order; ruby.h should be at first.
+
+Mon Mar 29 06:47:25 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/psych/*: importing Psych to trunk
+	* test/psych/*: ditto
+	* lib/psych/*: ditto
+
+Sun Mar 28 10:03:51 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/scanf.rb: fix max width modifier for %f.
+
+Sun Mar 28 10:35:45 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.6.0 r5717.
+	* test/minitest/*.rb: ditto.
+
+Sun Mar 28 10:12:28 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (rb_time_magnify): fix for LP64.
+
+Sun Mar 28 09:28:33 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/scanf.rb: fix %[egEFG] isn't accept.
+
+Sun Mar 28 09:53:02 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: rename small integer specialization related identifiers.
+
+Sun Mar 28 08:20:37 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (weq): specialize for small integer.
+	  (wne): ditto.
+	  (wlt): ditto.
+	  (wgt): ditto.
+	  (wle): ditto.
+	  (wge): ditto.
+	  (rb_time_magnify): ditto.
+	  (rb_time_unmagnify): ditto.
+	  (rb_time_unmagnify_to_float): new function to avoid rational for
+	  Time#to_f and Time#-.
+
+Sun Mar 28 07:12:41 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (mul): condition refined.
+
+Sun Mar 28 02:14:13 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: fix previous commit.
+
+Sat Mar 27 23:17:52 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: use 64bit arithmetic even on platforms with 32bit VALUE.
+
+Fri Mar 26 23:52:07 2010  wanabe  <s.wanabe@gmail.com>
+
+	* NEWS: add Thread#add_trace_func and Thread#set_trace_func.
+
+Fri Mar 26 22:58:10 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_initialize): initialize
+	  store->ex_data.sk.  a patch from Christian Holtje in
+	  [ruby-core:28907].  [ruby-core:23971] [ruby-core:18121]
+
+Fri Mar 26 19:55:41 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* Makefile.in (DOXYGEN): Define a missing variable DOXYGEN.  Build
+	  has been failing when doxygen(1) is found by configure but the
+	  variable is not defined by the system and make(1) does not allow
+	  an empty command. ("@$(DOXYGEN)" was the cause)
+
+Fri Mar 26 19:28:03 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_hash.rb (test_recursive_key): recursive keys are
+	  permitted now.  [ruby-dev:40735]
+
+Fri Mar 26 12:36:10 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* lib/yaml/basenode.rb: deprecating YPath methods
+
+	* lib/yaml/stream.rb: deprecating YAML::Stream#edit
+
+Fri Mar 26 12:29:28 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* lib/yaml/encoding.rb: YAML.encode, YAML.decode are deprecated.
+
+	* lib/yaml/stringio.rb: yaml/stringio.rb is deprecated.
+
+	* lib/yaml/ypath.rb: YAML::YPath is deprecated.
+
+Fri Mar 26 04:52:19 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: fix [ruby-core:28901] [Bug #2997].
+	  Thanks, Michael Graff.
+
+	* ext/tk/tcltklib.c: use xfree() for memories allocated by ALLOC().
+
+	* ext/tk/tkutil/tkutil.c: ditto.
+
+	* ext/tk/lib/tkextlib/version.rb: forgot updating.
+
+Thu Mar 25 23:53:22 2010  wanabe  <s.wanabe@gmail.com>
+
+	* thread.c (thread_reset_event_flags): set flags to thread.  [Bug #693]
+
+Thu Mar 25 23:17:16 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* file.c (rb_get_path_check): prefer #to_path over #to_str.  a patch
+	  originally written by me, and modified by Nobuyoshi Nakada.
+	  [ruby-core:24257]
+
+Thu Mar 25 18:10:08 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: Array#{uniq,uniq!} can take a block. see [ruby-dev:37998]
+
+Thu Mar 25 17:21:49 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add IO#autoclose=, IO#autoclose?, IO#codepoints,
+	  IO#each_codepoint.
+
+Thu Mar 25 14:50:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_{read,write}): fix handle leaks.  based on
+	  a patch from Heesob Park in [ruby-core:28919].  [ruby-core:28833]
+
+Thu Mar 25 12:37:37 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add Kernel#singleton_class.
+
+Thu Mar 25 11:34:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c, node.h, strftime.c, enc/trans/utf8_mac.trans:
+	  added explicit casts for suppressing warnings.
+
+Thu Mar 25 11:34:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* test/ruby/test_dir_m17n.rb: HFS+ escapes invalid byte sequences of
+	  filenames.
+
+Thu Mar 25 05:44:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ruby_version): needs runnable CPP.
+
+Thu Mar 25 04:42:19 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): take a breath for HP-UX make
+	  [ruby-core:28938].
+
+	* tool/make-snapshot (Dir.mktmpdir): for 1.8.6 backward compatibility.
+
+	* tool/make-snapshot (package): $: no longer contains ".", in trunk.
+
+Thu Mar 25 03:52:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/charset_alias.rb: update of URL.  [ruby-dev:38360]
+
+Wed Mar 24 23:43:40 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (thread_create_core): let new thread inherit RUBY_EVENT_VM
+	  of event_flags.  [ruby-core:25191]
+
+	* thread.c (rb_threadptr_exec_event_hooks): delete RUBY_EVENT_VM if
+	  all event_hooks are removed.
+
+Wed Mar 24 22:58:02 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* configure.in: revert the previous commit, which seemed to break make
+	  install.  The files of standard libraries are directly installed to
+	  $(PREFIX)/lib/ruby instead of $(PREFIX)/lib/ruby/1.9.1
+
+Wed Mar 24 18:16:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LDSHARED, CPP): keep $(CC).  [ruby-core:28912]
+
+Wed Mar 24 14:33:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): limit read size to 16KB if the file
+	  seems to be console.  [ruby-core:28902]
+
+Wed Mar 24 10:18:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_expand_path): set length of string before calling
+	  rb_enc_check because rb_enc_check scans its content.
+	  This prevents warnings by valgrind.
+
+Tue Mar 23 23:58:51 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Incorporating the fixes from the recent
+	  FasterCSV releases:  1.5.2 and 1.5.3.  New parser
+	  by Tim Elliott.  [ruby-core:25038]
+
+Tue Mar 23 18:35:46 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* object.c (rb_obj_singleton_class): new method
+	  Kernel#singleton_class.  [ruby-core:21702]
+
+Tue Mar 23 01:13:59 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: use rsock_ prefix for internal initialization functions.
+
+Mon Mar 22 20:32:22 2010  wanabe  <s.wanabe@gmail.com>
+
+	* proc.c (rb_method_location): return attr's location if it is setup.
+	  [Feature #2084]
+
+	* NEWS: follow above.
+
+	* vm_method.c (rb_add_method): save attr's location.
+
+	* gc.c (mark_method_entry): mark attr's location.
+
+	* method.h (rb_method_definition_t): add member to save attr's location.
+
+	* vm_eval.c (vm_call0): follow above.
+
+	* vm_insnhelper.c (vm_call_method): ditto.
+
+	* vm_method.c (rb_method_definition_eq): ditto.
+
+Mon Mar 22 19:31:06 2010  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (Date#>>): fixed.  [ruby-core:28011]
+
+Mon Mar 22 18:01:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RMDIRS): use rmdir -p if possible.
+
+	* Makefile.in ({dist,real}clean-ext): fix for removing ext
+	  directories.
+
+Mon Mar 22 09:06:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/.document: extracted from ext/.document.
+
+Mon Mar 22 00:50:33 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (rsock_revlookup_flag): declared.
+
+Mon Mar 22 00:01:24 2010  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup (config.h): RUBY_EXEC_PREFIX defined.
+
+Sun Mar 21 00:46:29 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.encode_www_form_component):
+	  convert strings of HTML5 ASCII incompatible encoding
+	  to UTF-8.
+
+	* lib/uri/common.rb (URI.encode_www_form_component):
+	  "\x83\x41" of Shift_JIS should be encoded as "%83A".
+	  This follows real implementations.
+
+	* lib/uri/common.rb (URI.decode_www_form_component):
+	  use given encoding for force_encoding. [ruby-dev:40721]
+
+Sun Mar 21 21:09:17 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv-replace.rb: specify super class for rdoc.
+
+	* lib/ipaddr.rb: ditto.
+
+Sun Mar 21 19:52:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_initialize): add autoclose argument to control close
+	  at finalization.  [ruby-core:26222]
+
+	* io.c (rb_io_autoclose_p, rb_io_set_autoclose): new methods.
+
+Sun Mar 21 19:50:04 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: make sources rdoc friendly.
+
+Sun Mar 21 17:57:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (IO_RBUF_CAPA_FOR): use large buffer as cbuf if readconv is
+	  needed for performance improvement.  based on a patch by Roger Pack
+	  in [ruby-core:28163].  [ruby-core:28162]
+
+Sun Mar 21 17:14:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: reuse hash object.
+
+Sun Mar 21 12:32:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (transcode_compile_tree): make
+	  valid_encoding mandatory unless from_encoding is registered in
+	  ValidEncoding.
+	  (transcode_tbl_only): ditto.
+	  (transcode_tblgen): ditto.
+	  (ValidEncoding): new function.
+
+	* enc/trans/escape.trans: specify valid_encoding.
+
+	* enc/trans/emoji_sjis_docomo.trans: ditto.
+
+	* enc/trans/emoji.trans: ditto.
+
+	* enc/trans/emoji_iso2022_kddi.trans: ditto.
+
+	* enc/trans/big5.trans: ditto.
+
+	* enc/trans/emoji_sjis_softbank.trans: ditto.
+
+	* enc/trans/emoji_sjis_kddi.trans: ditto.
+
+	* enc/trans/chinese.trans: use ValidEncoding() instead of
+	  ValidEncoding[].
+
+Sun Mar 21 09:43:01 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (IO_[CRW]BUF_CAPA_MIN): replaced magic numbers.
+
+Sun Mar 21 09:22:10 2010  Tanaka Akira  <akr@fsij.org>
+
+	* object.c (rb_obj_hash): use LONG2FIX to avoid Bignum allocation.
+
+Sat Mar 20 22:51:46 2010  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_sum): use UCHAR_MAX.
+
+Sat Mar 20 19:16:42 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (ActionMap.unambiguous_action): use
+	  Array#delete.
+	  (ActionMap.merge2): defined.
+	  (ActionMap::PostMemo): removed.
+	  (transcode_compile_tree): use ActionMap.merge2.
+
+Sat Mar 20 13:26:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_uniq_bang): the array is already unique if the
+	  length is zero or one.
+	  (rb_ary_uniq): ditto.
+
+Sat Mar 20 12:30:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib: fixed typo.  a patch by Sho Hashimoto in [ruby-dev:40716].
+
+Sat Mar 20 11:32:18 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (each_firstbyte_range): tuned for less
+	  method calls.
+
+Sat Mar 20 00:59:50 2010  wanabe  <s.wanabe@gmail.com>
+
+	* ext/dl/lib/dl/pack.rb (DL#pack): allow LLP64.
+
+Fri Mar 19 05:26:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_sum): don't call method for each byte.
+
+Thu Mar 18 21:24:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (miniruby): link $(NORMALMAINOBJ).
+
+Thu Mar 18 19:52:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: scan singleton mappings sequentially.
+
+Thu Mar 18 06:28:32 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_EXEC_PREFIX): added to config.h.
+
+	* ruby.c (ruby_init_loadpath_safe): TMP_RUBY_PREFIX should be
+	  exec_prefix, not rubylibprefix.  [ruby-core:28718]
+
+	* version.c (RUBY_LIB_PREFIX): fallback to RUBY_EXEC_PREFIX.
+
+Thu Mar 18 05:58:27 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_{to_i,mod,divmod,div2}):
+	  removed unused variables.
+
+Thu Mar 18 04:40:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_intern3): get rid of type-punning cast.
+
+Thu Mar 18 02:29:42 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_build_body): add mark to jump table hash of case.
+
+Thu Mar 18 00:58:27 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_strtod): Add support for Hexadecimal
+	  floating-point expression [ruby-dev:40650] #2969
+
+Thu Mar 18 00:00:58 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.{en,de}code_www_form_component):
+	  renamed from URI.{en,de}code_www_component. [ruby-dev:40672]
+
+	* lib/uri/common.rb (URI.encode_www_form_component): %-encoded
+	  element should have always two hex.
+
+	* lib/uri/common.rb (URI.encode_www_form_component):
+	  better treatment for ASCII incompatible encodings and
+	  encodings whose lead byte may use 7bit.
+
+	* lib/uri/common.rb (URI.decode_www_form_component): add %20.
+
+	* lib/uri/common.rb (URI.decode_www_form_component): add
+	  result's encoding as 2nd argument.
+
+	* lib/uri/common.rb (URI.decode_www_form): added.
+
+Wed Mar 17 16:25:53 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_aset): allow recursive key.  [ruby-core:24648]
+
+Wed Mar 17 06:39:59 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.h (rb_warning, rb_sys_warning): fixed typo in rdoc.
+	  [ruby-core:28696]
+
+Wed Mar 17 02:29:46 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c, iseq.c, ruby.c, vm.c, vm_core.h, vm_eval.c: add absolute
+	  path field into rb_iseq_t.  The field contains a string representing
+	  a path to corresponding source file. or nil when the iseq is created
+	  from -e, stdin, eval, etc.  This field is used for require_relative.
+	  [ruby-dev:40004]
+
+	* load.c (rb_f_require_relative): add C implementation of
+	  require_relative.
+
+	* prelude.rb (require_relative): get rid of Ruby implementation of
+	  require_relative.
+
+Wed Mar 17 01:24:01 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (rb_intern3): prohibit Symbol with an invalid encoding.
+	  [ruby-core:24621]
+
+	* test/ruby/test_m17n_comb.rb: modify a test for above.
+
+Tue Mar 16 22:51:11 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: specialize for singleton mappings.
+
+Tue Mar 16 20:13:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enc/trans/emoji.trans: added codepoints leading 0xf4 into
+	  nomap_table.
+
+Tue Mar 16 17:33:30 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpCtoV): keep e's sign.
+	  When e doesn't over flow e * BASE_FIG but overflow e * 10,
+	  e will overflow.
+
+Tue Mar 16 17:18:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_str_conv_enc_opts): missing prototype.
+
+Tue Mar 16 15:20:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tmpdir: split from lib/tmpdir.rb.
+
+	* win32/win32.c (rb_w32_system_tmpdir): extracted from init_env.
+
+	* win32/win32.c (rb_w32_getcwd): removed duplicated code.
+
+Tue Mar 16 14:06:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (ENC_TO_ENCINDEX): suppress warnings.
+
+Tue Mar 16 14:03:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RSHIFT): parenthesize the argument of cast instead
+	  of the result.
+
+Tue Mar 16 11:23:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): ignore dname if it has different
+	  drive letter or UNC.
+
+Tue Mar 16 07:16:56 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: redundant loop removed.
+
+Tue Mar 16 07:01:43 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: refactored to use tree as memo key.
+
+Tue Mar 16 04:05:13 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: more info in generating macro names.
+
+Mon Mar 15 21:58:03 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: ActionMap#each_firstbyte inlined.
+
+Mon Mar 15 21:22:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (transcode_tblgen): add valid_encoding
+	  optional argument.
+
+	* enc/trans/single_byte.trans use valid_encoding argument for
+	  transcode_tblgen.
+
+	* enc/trans/chinese.trans: ditto.
+
+Mon Mar 15 18:33:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (default_rand): removed initial buffer.
+
+	* random.c (Init_RandomSeed): initialize seed of default random.
+
+	* random.c (Init_RandomSeed2): turn the seed to Bignum object.
+
+Mon Mar 15 17:28:30 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_print): RDoc update.  a patch from Daniel Kelley
+	  in [ruby-core:28643].
+
+Mon Mar 15 14:06:07 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (next_state): no initialization here.
+
+	* random.c (default_mt): always return initialized MT.
+
+Mon Mar 15 11:49:48 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c (rb_reset_random_seed): set seed in this.
+	  [ruby-core:28655]
+
+Mon Mar 15 10:26:02 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: import latest nkf. [master 9306cb0]
+	  this also fixes [ruby-dev:40607]
+
+Mon Mar 15 09:34:17 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.encode_www_component):
+	  call str.to_s at first.
+
+Mon Mar 15 09:36:22 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbian/README.SYMBIAN: fixed broken patch and converted to
+	  unified diff.
+
+Mon Mar 15 07:51:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (Init_File): do not define File::ALT_SEPARATOR on cygwin.
+
+Mon Mar 15 07:41:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (realpath_rec): use same cache.
+
+	* file.c (realpath_internal): regulate separators in prefix.
+	  [ruby-core:28653]
+
+	* file.c (FILE_ALT_SEPARATOR): separated condition.
+
+Mon Mar 15 04:41:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_each_codepoint): read directly when readconv is
+	  needed but internal encoding is not set.  [ruby-core:28650]
+
+Mon Mar 15 04:18:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/file2lastrev.rb (VCS::{SVN,GIT}#get_revisions):
+	  use block argument 'path' to get the path given by super.
+
+Mon Mar 15 02:43:59 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (Action#hash): defined.
+	  (Action#eql?): ditto.
+	  (Action#==): ditto.
+
+Mon Mar 15 01:52:46 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: refactored.
+
+Mon Mar 15 01:18:31 2010  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup (*.pkg): Ruby Core installation separated from standard extensions.
+
+	* symbian/configure.bat: ditto.
+
+	* symbian/README.SYMBIAN: ditto.
+
+	* symbian/setup (config.h): support for 2nd internal drive added.
+
+Mon Mar 15 00:11:23 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/file2lastrev.rb (VCS::GIT_SVN#get_revisions) :
+	  use block argument 'path' to get the path given by super.
+
+	* tool/file2lastrev.rb (VCS::GIT#get_revisions):
+	  use double quotes for Windows.
+	  patched by Vladimir Sizikov [ruby-core:28651]
+
+Sun Mar 14 22:38:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: consider valid_encoding for max_input.
+
+Sun Mar 14 15:46:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/emoji.trans: fix nomap_table.
+
+Sun Mar 14 09:50:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (insn_data_to_s_detail), iseq.c (insn_operand_intern):
+	  fixed format specifiers.
+
+Sun Mar 14 07:20:17 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (EXPAND_PATH_BUFFER): make it back to usascii, to prevent
+	  infinite loop on some platform. [ruby-dev:40629]
+
+Sun Mar 14 02:40:38 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: reject ambiguous mapping.
+
+	* enc/trans/single_byte.trans: remove ambiguous mapping such as
+	  \xD6 -> U+05F2 and \xD6\xC7 -> U+FB1F in Windows-1255
+
+Sat Mar 13 23:48:27 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (file_expand_path): should not just copy the encoding
+	  from fname.  [ruby-core:28635]
+
+	* file.c (EXPAND_PATH_BUFFER): set filesystem_encoding, not
+	  usascii for path buffer.
+
+Sat Mar 13 17:48:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/file2lastrev.rb: refactored.  fixed changed revision of git.
+
+Sat Mar 13 15:44:20 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_print): should not print field separator at the end
+	  of arguments.  [ruby-talk:358633]
+
+Sat Mar 13 14:49:55 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_join): remove Enumerable#join. [ruby-core:24786]
+
+	* array.c (ary_join_1): use #to_ary to detect recursive array.
+
+Sat Mar 13 12:26:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h (MakeOpenFile): finalize fptr to get rid of
+	  memory leak.
+
+Sat Mar 13 11:14:26 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* load.c (rb_get_expanded_load_path): expand paths if any item in $:
+	  is not a string.
+
+Sat Mar 13 10:16:32 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* load.c (rb_get_expanded_load_path): does not expand paths if all
+	  the items in $: are absolute paths.  [ruby-core:28113]
+
+Sat Mar 13 10:03:52 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: fix [Bug #2840] Tk doesn't built in mingw.
+
+Sat Mar 13 03:24:15 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: show consumed time at last.
+
+Sat Mar 13 00:44:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_basename): check encoding of suffix.
+
+Sat Mar 13 00:11:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): mark initial load paths.
+
+	* gem_prelude.rb (push_all_highest_version_gems_on_load_path):
+	  search insertion position by initial load path mark.
+
+	* lib/rubygems.rb (Gem.load_path_insert_index): ditto.
+
+Fri Mar 12 21:34:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: emoji encodings.
+
+Fri Mar 12 17:14:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.encode_www_form): new method to
+	  generate URL-encoded form data. [ruby-dev:39246]
+
+	* lib/uri/common.rb (URI.encode_www_component,
+	  URI.decode_www_component): new method for encode/decode
+	  a name/value of HTML form.
+
+Fri Mar 12 17:36:35 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb
+	  (WEBrick::HTTPServlet::CGIHandler#do_GET):
+	  set binary mode for tempfile.
+	  http://pc12.2ch.net/test/read.cgi/tech/1265467681/286
+
+Fri Mar 12 13:52:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* tool/compile_prelude.rb: TMP_RUBY_PREFIX should replace
+	  rubylibprefix but not prefix.
+
+Fri Mar 12 09:43:11 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* complex.c (f_signbit): remove condition for signbit because
+	  all platforms have signbit from r26871.
+
+Fri Mar 12 07:25:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (init_env): get rid of alloca() for outer string.
+
+Fri Mar 12 07:17:15 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/racc/cparse/cparse.c: suppressed warnings for shortening on
+	  platforms where pointer is bigger than int.
+
+Fri Mar 12 07:15:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe, ruby_init_gems): set and remove
+	  TMP_RUBY_PREFIX.
+
+	* variable.c (rb_const_remove): new function.
+
+	* tool/compile_prelude.rb: split each preludes.
+
+Fri Mar 12 07:09:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (config.status): setup MINIRUBY environment for
+	  cross-compiling.
+
+Fri Mar 12 06:48:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/extconf.rb: log() is built-in in some compilers and needs
+	  proper declaration.
+
+Fri Mar 12 06:37:13 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/timeout.rb: split from lib/timeout.rb.
+
+Thu Mar 11 22:35:27 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_round): new method Time#round.  [ruby-dev:40595]
+
+Thu Mar 11 17:31:59 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (str_transcode0): confirm the code range.
+
+Thu Mar 11 12:14:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/signbit.c: added.
+
+Thu Mar 11 11:16:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if target_archs has changed.
+
+Thu Mar 11 01:33:30 2010  wanabe  <s.wanabe@gmail.com>
+
+	* win32/win32.c (signbit): allow x64.
+
+Thu Mar 11 01:04:48 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* iseq.c (rb_iseq_clone): sets local_iseq and klass properly.
+
+Wed Mar 10 21:25:41 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enum.c (min_ii, max_ii, minmax_ii): remove wrong optimization that
+	  reuses array for yield parameter, which caused unexpected behavior.
+	  [ruby-core:25989]
+
+Wed Mar 10 12:10:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enc/x_emoji.h: renamed from enc/x-emoji.c.
+
+Wed Mar 10 04:17:08 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (defined_expr), insns.def (defined): fix to pass a Symbol
+	  object such as :$foo for the "defined?($foo)" expression
+	  on defined instruction.  This patch fixes ISeq#to_a/load.
+
+Tue Mar  9 20:50:09 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/dl/test_method.rb: delete a residual test for dl on libffi.
+
+Tue Mar  9 18:00:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enc/trans/EMOJI/*.src, enc/trans/emoji*,
+	  enc/x-emoji.c, test/ruby/enc/test_emoji.rb,
+	  tool/enc-emoji-citrus-gen.rb, tool/enc-emoji4unicode.rb,
+	  tool/jisx0208.rb, tool/test/test_jisx0208.rb:
+	  new encodings to support emoji charsets, which are used by
+	  Japanese mobile phones [ruby-dev:40528].
+	  Thanks Yoji Shidara for a lot of contribution.
+
+	* tool/transcode-tblgen.rb: modified for enc-emoji4unicode.rb.
+
+Tue Mar  9 14:57:40 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/transcode-tblgen.rb (ActionMap#generate_node):
+	  remove useless dup.
+
+	* tool/transcode-tblgen.rb (transcode_tbl_only): remove duplicated
+	  method call for encode_utf8.
+
+Tue Mar  9 14:55:00 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_to_s): remove unused variable.
+
+Tue Mar  9 12:17:34 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* tool/transcode-tblgen.rb (StrSet#parse, ActionMap#generate_node):
+	  optimization (reusing object and improving memoize).
+	  [ruby-dev:40582]
+
+Sun Mar  7 02:27:03 2010  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* dln.c [DLN_NEEDS_ALT_SEPARATOR] (translit_separator): small typo
+	  fixed.
+
+	* symbian/setup (*.pkg): Shared library installation path fixed to
+	  match changes in version.c.
+
+	* symbian/setup (ruby.mmp): SOURCE dln_find.c added.
+
+Sun Mar  7 23:14:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_pack): use DIGSPERLONG and BITSPERDIG.
+	  (rb_big_unpack): use DIGSPERLONG.
+
+Sun Mar  7 19:21:10 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* io.c: Fix documentation for each/each_line/lines, bytes/each_byte,
+	  codepoints/each_code_point [ruby-core:23948]
+
+	* string.c: ditto
+
+	* ext/stringio/stringio.c: ditto
+
+Sun Mar  7 13:49:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* file.c: add optional basedir argument for realpath/realdirpath.
+	  (realpath_internal): handle basedir.
+	  (rb_file_s_realpath): extract basedir from argument list.
+	  (rb_file_s_realdirpath): extract basedir from argument list.
+
+	* lib/pathname.rb (realpath): pass basedir.
+	  (realdirpath): ditto.
+
+Sun Mar  7 02:05:38 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_set_filesystem_encoding):
+	  filesystem encoding on Mac OS X is now
+	  default external encoding. so Mac OS X is now
+	  treated as one of Unix. [ruby-dev:40439]
+
+	* file.c (file_path_convert): ditto.
+
+Sat Mar  6 22:07:09 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c: fix the prototype of rb_big2ulong_pack.
+
+Sat Mar  6 13:45:37 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (rb_io_s_write, rb_io_s_binwrite): delete File#write and
+	  #binwrite.  It may take a long time to make them stable, so 1.9.2
+	  should not include them.  We need refactoring to implement them
+	  again.
+
+	* test/ruby/test_io.rb: delete tests for them.
+
+Sat Mar  6 10:03:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (rb_io_s_binwrite): use mode "r+b" instead of "ab" to handle
+	  offset correctly.  [ruby-core:28517]
+
+	* test/ruby/test_io.rb: use assert_equal instead of assert.
+
+Sat Mar  6 01:08:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): VC6 has __int64.
+
+Fri Mar  5 19:36:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (econv_opts): set :undef => :replace when
+	  :replace is given and :invalid is not given. [ruby-dev:40554]
+
+Fri Mar  5 17:51:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): suppressed a warning.
+
+	* win32/win32.c (rb_w32_map_errno, winnt_stat, rb_chsize): ditto.
+
+	* win32/win32.c (rb_w32_aspawn): get rid of overflow.
+
+Fri Mar  5 11:59:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (signbit): defined on mingw.
+
+Thu Mar  4 22:49:42 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval.c (setup_exception): avoid line number 0 when setting
+	  $DEBUG=true.  [ruby-dev:39116]
+
+Thu Mar  4 22:45:02 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (io_s_write, io_s_binwrite): add File#write and #binwrite.
+	  [ruby-core:21701]
+
+	* test/ruby/test_io.rb: add tests for above.
+
+Mon Mar  4 13:14:34 2010  wanabe  <s.wanabe@gmail.com>
+
+	* gc.c (gc_profile_total_time): add GC::Profiler.total_time.
+	  [ruby-core:27169]
+
+	* NEWS: ditto.
+
+Thu Mar  4 10:15:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* complex.c (m_log, m_exp): remove unused functions.
+
+Thu Mar  4 02:34:59 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_env.rb (TestEnv#test_select_bang): add tests.
+
+Thu Mar  4 02:29:52 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_hash.rb (TestHash#test_keep_if): fix typo.
+
+Thu Mar  4 00:10:54 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (HEX2): extracted from regexps.
+
+Wed Mar  3 23:16:32 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* complex.c (nucomp_real_check): raise TypeError instead of
+	  ArgumentError when argument is not a real as expected
+	  [ruby-core:28395]
+
+	* rational.c (nurat_int_check): ditto (for integers)
+
+Wed Mar  3 23:07:08 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (Vector#each): Return self and optimization
+	  [ruby-core:28405]
+
+Wed Mar  3 18:35:55 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_select_bang): add #select! and keep_if to Hash.
+
+	* hash.c (env_select_bang): ..and to ENV.
+
+Wed Mar  3 15:54:20 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/matrix.rb (Vector#each2, collect2): small refactoring.
+
+Wed Mar  3 16:14:24 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/matrix.rb (Vector#each): make Vector enumerable.
+	  [ruby-core:28405]
+
+Wed Mar  3 14:28:23 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_select_bang): select! removes all elements for
+	  which block returns false.  [ruby-core:27286]
+
+	* array.c (rb_ary_keep_if): #keep_if, new method.
+
+Wed Mar  3 06:19:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (signbit): defined.
+
+	* win32/Makefile.sub, symbian/setup (config.h): pack.c requires
+	  SIZEOF_INT*_T now.
+
+Tue Mar  2 21:16:48 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (rb_io_lines, rb_io_bytes, rb_io_chars, rb_io_codepoints):
+	  change to alias to each_*, in similar way to ARGF and String.
+	  [ruby-core:23948]
+
+Tue Mar  2 15:54:40 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (noname_disable_map): add NT_ANCHOR case.
+	  Without this change, captured groups in anchors (look-ahead,
+	  look-behind, and so on) are not removed and
+	  unintended invalid backref error occur. [ruby-core:28235]
+
+	* regcomp.c (renumber_by_map): ditto.
+
+Tue Mar  2 14:42:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* test/dl/test_closure.rb: forget to revert at r26764.
+
+Tue Mar  2 14:29:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext_safe, rb_find_file_safe): skip argument
+	  checks in file_expand_path().
+
+Tue Mar  2 13:54:44 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* test/ruby/test_math.rb: Fix lgamma test
+
+Mon Mar  1 23:26:56 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (pty_open): refine the path for master IO.
+
+Mon Mar  1 20:07:06 2010  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* ext/zlib/zlib.c (zstream_expand_buffer_into): remove compare
+	  different type values warning.
+
+Mon Mar  1 17:42:45 2010  wanabe  <s.wanabe@gmail.com>
+
+	* configure.in (mingw): do not detect snprintf/vsnprintf.
+
+Mon Mar  1 16:54:21 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/oniguruma.h: updated to follow Oniguruma 5.9.2.
+
+	* re.c (make_regexp): use onig_new() instead of onig_alloc_init().
+
+	* re.c (rb_reg_to_s): ditto.
+Sun Feb 28 21:32:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (get_device_once): raise on error when fail is
+	  non-zero.
+
+Sun Feb 28 18:58:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* math.c (rb_eMathDomainError): new exception class
+	  for representing mathematical domain error instead
+	  of Errno::EDOM.
+
+	* math.c (domain_check, infinity_check): removed,
+	  no longer needed.
+
+	* math.c (math_atan2, math_acos, math_asin, math_acosh,
+	  math_atanh, math_log, math_log2, math_log10, math_sqrt,
+	  math_gamma, math_lgamma): mathematical domain errors
+	  are checked and raised before calling libm's functions.
+
+	* test/ruby/test_math.rb: updated for changes of math.c.
+
+Sun Feb 28 15:07:28 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_pack): use union to avoid pointer cast.
+	  (pack_unpack): ditto.
+
+Sun Feb 28 11:49:35 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* openssl/ossl.c (OSSL_IMPL_SK2ARY): for OpenSSL 1.0.
+	  patched by Jeroen van Meeuwen at [ruby-core:25210]
+	  fixed by Nobuyoshi Nakada [ruby-core:25238],
+	  Hongli Lai [ruby-core:27417],
+	  and Motohiro KOSAKI [ruby-core:28063]
+
+	* ext/openssl/ossl_pkcs7.c (pkcs7_get_certs, pkcs7_get_crls):
+	  split pkcs7_get_certs_or_crls.
+
+Sun Feb 28 11:42:55 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/openssl/{test_x509cert.rb,openssl,test_x509crl.rb,
+	  test_x509req.rb}:
+	  fix false positive tests because of OpenSSL spec change.
+	  patched by originally Hongli Lai [ruby-core:27417],
+	  and fixed by Motohiro KOSAKI [ruby-core:28063]
+
+Sun Feb 28 11:21:03 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/openssl/{test_x509cert.rb,openssl,test_x509req.rb}:
+	  remove false positive tests because of OpenSSL spec change.
+	  patched by Motohiro KOSAKI [ruby-core:28063]
+	  see also [ruby-dev:40077]
+
+Sun Feb 28 11:25:16 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_pack): generalized integer packer implemented.
+	  (pack_unpack): generalized integer unpacker implemented.
+
+Sun Feb 28 06:58:53 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (swap32): use __builtin_bswap32 on gcc 4.3.0 or later.
+	  (swap64): ditto.
+
+Sun Feb 28 00:38:18 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c: use integer types with explicit size.
+
+Sat Feb 27 15:54:55 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c: check assumption on QUAD_SIZE and SIZEOF_LONG.
+
+	* bignum.c: check assumption on SIZEOF_LONG and SIZEOF_BDIGITS.
+
+Sat Feb 27 03:48:18 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c: fix q and Q for big endian environments which have no
+	  8 bytes integer type.
+	  (pack_pack): use rb_big_pack.
+	  (pack_unpack): use rb_big_unpack.
+
+	* include/ruby/intern.h (rb_big_pack): declared.
+	  (rb_big_unpack): ditto.
+
+	* bignum.c (rb_big_pack): new function.
+	  (rb_big_unpack): ditto.
+
+Fri Feb 26 21:36:51 2010  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: fix rb_quad_pack and rb_quad_unpack for environments
+	  which don't have 8bytes integer type.  This still depends on little
+	  endian.
+	  (rb_quad_pack): use quad_buf_complement.  don't raise for large
+	  values.
+	  (rb_quad_unpack): use quad_buf_complement.
+	  (quad_buf_complement): new function extracted from rb_quad_pack.
+	  add one after bitwise negation.
+
+Fri Feb 26 21:29:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (RSHIFT): add parenthesis to suppress warning.
+
+Fri Feb 26 20:51:47 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/socket/raddrinfo.c, ext/socket/udpsocket.c: avoid illegal cast
+	  from void-returning function to VALUE-returning one.  It causes SEGV
+	  on RubySpec with mingw32.
+
+Fri Feb 26 20:07:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (QUAD_SIZE): it should be always 8.
+
+	* bignum.c (rb_quad_pack): use LONG_LONG version only if
+	  SIZEOF_LONG_LONG == QUAD_SIZE.
+
+Fri Feb 26 16:49:29 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (swap64): fix for VC6.
+
+Fri Feb 26 14:17:09 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack, pack_unpack): suppressed warnings.
+
+Fri Feb 26 07:44:51 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/dl: revert dl with libffi because it can't run on mswin now.
+
+Thu Feb 25 22:59:46 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c: consider DYNAMIC_ENDIAN.  refactored.
+
+Thu Feb 25 11:17:01 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* rational.c (nurat_expt): use Float#** when Rational ** Float.
+	  This fixes Rational(3,1)**3.0=>26.99999999999999 on FreeBSD.
+
+	* complex.c (rb_fexpt): removed.
+	  Note that this function is not static but is private.
+
+Thu Feb 25 00:43:57 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: Change the operand type of setinlinecache
+	  (OFFSET to IC).  This IC must be same as corresponding
+	  getinlinecache instruction's IC operand.
+	  This change is for a little performance improvement
+	  (getting IC directly) and is for the AOT compilation development.
+
+	* compile.c, iseq.c, insns.def: Change the approach to handling inline
+	  cache (IC) type operand to enable the above change.
+	  This change also affects ISeq#to_a method.  The inline cache operand
+	  will be dumped by fixnum, the index of inline cache, in other words,
+	  inline cache identity.
+
+	* template/insns_info.inc.tmpl, tool/instruction.rb: No need to count
+	  inline cache size (insn_iclen()).
+
+Thu Feb 25 00:57:59 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (swap64): don't redefine.
+
+Thu Feb 25 00:29:55 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_unpack): don't use OFF32 for gcc 4.5.
+
+Wed Feb 24 22:39:15 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: fix [ruby-core:28320] reported by Paul Clegg.
+	  (Resolv::DNS::Requester#request): raise ResolvTimeout
+	  consistently for timeout.
+
+Wed Feb 24 14:38:16 2010  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/indexer: Removed obsolete files.
+
+Wed Feb 24 11:52:05 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (translit_separator): moved back from load.c again.
+
+	* dln_find.c: split from dln.c.
+
+Wed Feb 24 09:31:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/extconf.rb: use OpenSSL only when all transform
+	  functions are available.
+
+Wed Feb 24 00:39:17 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (str_new_empty): String#split, partition, rpartition
+	  taints the resulting strings if self is tainted.
+
+Mon Feb 22 21:35:33 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/digest/sha2/sha2init.c: test OpenSSL more strictly.
+
+Mon Feb 22 11:52:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems: update to 1.3.6.
+
+Mon Feb 22 11:21:18 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/sha2: Use OpenSSL's SHA1 engine if available.
+
+Sun Feb 21 21:20:17 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile, install_files): honor srcprefix
+	  argument if given.  [ruby-dev:40449]
+
+Sun Feb 21 13:29:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/rubygems_version.rb: outdated.  [ruby-core:28275]
+
+Sun Feb 21 06:24:12 2010  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup (config.h): HAVE_STRUCT_TIMEZONE, VOID_UNSETENV, and RUBY_LIB_VERSION_STYLE defined.
+	* symbian/setup (ruby.mmp): SOURCE node.c added.
+
+Mon Feb 22 09:15:45 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_mark_children): lost comment added.
+
+Sat Feb 20 14:42:19 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_object0): should prepare placeholder before
+	  processing instance variables.  [ruby-dev:40414]
+
+	* marshal.c (id2encidx): no longer need arg.
+
+Sat Feb 20 14:36:16 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm.c (vm_backtrace_each): use called_id when method definition
+	  structure is already freed.  [ruby-dev:40234] [ruby-core:27959]
+
+Fri Feb 19 00:04:19 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/drb/drb.rb (DRbServer#stop_service): join killed thread to
+	  ensure service stops.  [ruby-dev:40441]
+
+Thu Feb 18 22:31:15 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* math.c (math_atanh): raise ERANGE without calling atanh if absolute
+	  value is 1 to achieve platform-independent math.  [ruby-core:28219]
+
+	* math.c (math_lgamma): return [Infinity, 1] without calling lgamma_r
+	  if argument is infinity or -infinity.  [ruby-core:28219]
+
+Thu Feb 18 22:28:00 2010  Kenta Murata  <mrkn@mrkn.jp>
+
+	* configure.in: new --with-ext and --with-out-ext options for extmk.
+	* tool/mkconfig.rb: normalizing --with-out-ext to --without-ext.
+
+Thu Feb 18 21:50:00 2010  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_unpack): call PACK_ITEM_ADJUST for 'Q'.
+
+Thu Feb 18 02:14:26 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (io_fread, io_getpartial, io_read, io_sysread): by using lock,
+	  prohibit modification of buffer string during read (which had caused
+	  EFAULT or SEGV).  [ruby-dev:40437]
+
+	* test/ruby/test_io.rb: rewrite tests for the old behavior.
+
+Wed Feb 17 21:34:01 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* regcomp.c (setup_tree, onig_compile): optimize .* at last by
+	  converting into (?>.*), which does not backtrack.  [ruby-core:27791]
+
+	* test/ruby/test_regexp.rb: add a test for above.
+
+Wed Feb 17 21:26:53 2010  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_normal_exit): add :timeout option.
+
+Wed Feb 17 17:05:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_cfunc): removed unused variable.
+
+	* vm.c (vm_frametype_name): define only when VMDEBUG.
+
+Wed Feb 17 15:34:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/dl_conversions.c (rb_dl_type_to_ffi_type): support signed
+	  long long.
+
+	* ext/dl/cfunc.c (rb_dlcfunc_inspect): get rid of overflow.
+
+	* ext/dl/closure.c (dlc_callback, rb_dlclosure_init): ditto.
+
+	* ext/dl/cptr.c (rb_dlptr_s_malloc): ditto.
+
+	* ext/dl/method.c (rb_dlfunction_initialize): ditto.
+
+Wed Feb 17 01:16:12 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* hash.c (hash_update): always raise an exception when adding a new
+	  key during iteration.  Traditionally, an exception was raised only
+	  when rehash occurs, but it may lead to difficult bug to reproduce.
+	  [ruby-core:23614]
+
+Tue Feb 16 22:09:27 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gc.c (chain_finalized_object): fix precedence.
+
+Tue Feb 16 21:32:01 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gc.c (rb_objspace_call_finalizer): always run finalizer at exit,
+	  before object is free'd.  [ruby-dev:40382]
+
+Tue Feb 16 20:56:39 2010  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (rb_group_member): renamed from group_member.
+	  don't use group_member() in glibc because it's not valgrind clean.
+	  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=570047
+
+Tue Feb 16 19:19:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): convert signed value to
+	  unsigned.
+
+Tue Feb 16 19:02:59 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/dl/test_{base,handle}.rb: use more verbose assertions.
+
+	* test/dl/test_import.rb (DL::LIBC::BoundQsortCallback): renamed
+	  to get rid of overwriting warning.
+
+Tue Feb 16 11:03:19 2010  Aaron Patterson  <aaron@tenderlovemaking.com>
+
+	* ext/dl/method.c: Adding DL::Method as a superclass for DL::Function
+
+Mon Feb 15 23:37:30 2010  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: check lseek error by errno.  NetBSD 4.0.1 may return -1 as
+	  a file position of tty.
+
+Mon Feb 15 23:08:56 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pstore.rb (PStore#initialize): initialize @thread_safe.
+	  [ruby-core:27853]
+
+Mon Feb 15 22:45:26 2010  Tanaka Akira  <akr@fsij.org>
+
+	* st.c (st_foreach): don't access ptr->hash after func call.
+	  It may access freed area.
+
+Mon Feb 15 22:25:16 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/zlib/zlib.c (zlib_mem_alloc): suppress valgrind warnings.
+	  http://www.zlib.net/zlib_faq.html#faq36
+
+Mon Feb 15 22:18:49 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_add): propagate fixed time offset.
+
+Mon Feb 15 17:42:20 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* signal.c (USE_SIGALTSTACK): NetBSD can't use sigaltstack(2)
+	  with pthread.
+	  http://netbsd.gw.com/cgi-bin/man-cgi?sigaltstack++NetBSD-current
+
+Mon Feb 15 13:11:47 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_push_m): use rb_ary_modify instead of
+	  rb_ary_modify_check. This fixes regression due to r26632.
+
+Sun Feb 14 12:54:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): removed duplicated entry for regexp.
+	  [ruby-dev:40416]
+
+Sun Feb 14 04:45:31 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): register regexp object before encoding
+	  name.  [ruby-dev:40414]
+
+	* re.c (rb_reg_alloc, rb_reg_init_str): split from rb_reg_new_str.
+
+Sat Feb 13 17:07:20 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_delete): RDoc update.  a patch from Hugh Sasse.
+	  [ruby-core:28128]
+
+	* array.c (rb_ary_compact_bang): ditto.
+
+Sat Feb 13 15:01:24 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (id2encidx): duplicated entry for encoding name.
+	  [ruby-dev:40388]
+
+Sat Feb 13 12:17:52 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tempfile.rb (Tempfile::Remover): new class to replace
+	  Tempfile.callback.  port r24902 from Ruby 1.8.
+
+Fri Feb 12 17:55:21 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (thread_free): fixed typo.
+
+Fri Feb 12 02:27:39 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#initialize, Set#replace, Set#merge)
+	  (Set#subtract, Set#&): Fix duck type tests. [ruby-core:28078]
+
+	* lib/set.rb (Set#initialize, Set#replace, Set#merge)
+	  (Set#subtract, Set#&): Try #each if #each_entry fails.
+
+Thu Feb 11 20:43:00 2010  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_oflags_modestr): return "r" for O_RDONLY|O_APPEND.
+	  [ruby-dev:40379]
+
+Thu Feb 11 19:19:21 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* missing/alloca.c: s/RUBY_LIB/RUBY_LIB_PREFIX/ [ruby-dev:40395]
+
+Thu Feb 11 17:52:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_exec): reset thread state before restarting vm loop
+	  from catch scope.  [ruby-core:28129], [ruby-core:28143]
+
+Thu Feb 11 14:43:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): read sequentially since marshal source
+	  may not be possible to rewind.  [ruby-dev:40386]
+
+	* marshal.c (r_object0): replace non-1.8 escapes directly.
+
+Thu Feb 11 09:49:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: fix [ruby-core:28144] reported by Hans de Graaff.
+	  (Resolv::DNS#make_requester): pass nameserver_port to
+	  UnconnectedUDP.new.
+	  (Resolv::DNS.bind_random_port): change the is_ipv6 argument to
+	  bind_host.
+	  (Resolv::DNS::Requester#initialize): change instance variable to
+	  store multiple sockets.
+	  (Resolv::DNS::Requester#request): pass readable sockets to
+	  recv_reply.
+	  (Resolv::DNS::Requester#close): close all sockets.
+	  (Resolv::DNS::Requester::UnconnectedUDP#initialize): allocate
+	  a socket for each address family of name servers.
+	  (Resolv::DNS::Requester::UnconnectedUDP#recv_reply): read from the
+	  passwd readable socket.
+	  (Resolv::DNS::Requester::UnconnectedUDP#sender): use appropriate
+	  socket for the target nameserver.
+	  (Resolv::DNS::Requester::ConnectedUDP): follow the instance variable
+	  change.
+	  (Resolv::DNS::Requester::TCP#sender): ditto.
+	  (Resolv::DNS::Config#nameserver_port): new method.
+
+Thu Feb 11 01:45:04 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (vm_exec): temporarily revert r26628, which causes SEGV when
+	  executing rubyspec.
+
+Wed Feb 10 16:31:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_push_m, rb_ary_unshift_m, rb_ary_aset),
+	  (rb_ary_insert, rb_ary_replace, rb_ary_concat),
+	  (rb_ary_uniq_bang, rb_ary_flatten_bang): check if frozen after
+	  wrong number of arguments but before TypeError.
+	  [ruby-core:28140]
+
+	* hash.c (rb_hash_replace): ditto.
+
+	* string.c (rb_str_replace): ditto.
+
+Wed Feb 10 04:06:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_exec): reset thread state before starting vm loop.
+	  [ruby-core:28129]
+
+Tue Feb  9 23:48:25 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_divide): fix precision too.
+
+Tue Feb  9 23:26:07 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_DoDivmod): fix precision.
+	  [ruby-core:17472][ruby-dev:35372][ruby-dev:40105][ruby-dev:40358]
+
+Tue Feb  9 22:23:27 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (READ_CHECK): do not select fd before reading, that had made
+	  TCPServer#gets stuck.  [ruby-dev:40317]
+
+Tue Feb  9 21:27:38 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rexml/text.rb (REXML::Text#initialize): REXML::Text.new checks
+	  raw text for illegal characters without entity check, for the sake
+	  of 1.8 compatibility.  This had caused rubyspec error.
+
+Mon Feb  8 23:49:24 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_getline_fast): wrong calculation of new position
+	  from rb_str_coderange_scan_restartable().  [ruby-core:28103]
+
+	* io.c (read_all): ditto.
+
+	* sprintf.c (rb_str_format): ditto.
+
+Mon Feb  8 21:03:53 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (socket_s_ip_address_list): obtain the scope_id
+	  of IPv6 link local address on OpenSolaris.
+
+Mon Feb  8 16:27:57 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dmyversion.c: empty load path in miniruby.
+
+	* common.mk (COMPILE_PRELUDE): rbconfig is loaded on demand.
+
+	* tool/eval.rb, win32/{mkexports,resource}.rb: miniruby no longer
+	  contains "." in $:.
+
+Mon Feb  8 15:15:07 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_clone): call initialize_clone hook method to
+	  call initialize_copy.
+
+	* object.c (rb_obj_dup): call initialize_dup hook.
+
+	* lib/delegate.rb (Delegator#initialize_clone): use new hook to
+	  implement deep copy.  [ruby-dev:40242]
+
+	* lib/delegate.rb (Delegator#initialize_dup): ditto.
+
+	* test/test_delegate.rb (TestDelegateClass#test_copy_frozen): add
+	  a test to ensure #clone copies frozen status.
+
+Mon Feb  8 10:28:58 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: check pthread_attr_getstack to prevent obsolete
+	  warning for pthread_attr_getstackaddr.
+
+Sun Feb  7 23:12:34 2010  Tanaka Akira  <akr@fsij.org>
+
+	* sample/test.rb: sort files for syntax validation.
+
+Sun Feb  7 23:08:53 2010  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb: sort test files.
+
+Sun Feb  7 12:53:12 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rake.rb (Rake::FileList::ARRAY_METHODS): fix test failure by <=>
+	  definition at Kernel.
+
+Sun Feb  7 03:01:46 2010  Tanaka Akira  <akr@fsij.org>
+
+	* math.c (math_lgamma): initialize sign because
+	  lgamma(NaN) doesn't set the sign in OpenSolaris.
+
+Sun Feb  7 00:23:21 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* class.c (rb_class_init_copy): raise a TypeError if the argument is
+	  BasicObject.  [ruby-core:27060]
+
+Sat Feb  6 23:37:11 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (initialize): set @sock to a NullSocket instance to
+	  raise FTPConnectionError when not connected.   [ruby-dev:40258]
+
+Sat Feb  6 23:25:57 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/curses/view2.rb: replaced with Hugh Sasse's version.
+	  [ruby-core:27894]
+
+Sat Feb  6 22:57:00 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (receive_responses): does not hang when an
+	  unexpected BYE response received.  fixed [ruby-core:27944].
+	  Thanks, Bob Potter.
+
+Sat Feb  6 21:31:23 2010  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb (wait): supported timeout.
+
+	* test/net/imap/test_imap.rb (test_exception_during_idle): use timeout.
+
+Sat Feb  6 19:35:16 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* tool/compile_prelude.rb: fix require path.
+
+Sat Feb  6 12:02:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#method_missing),
+	  (Delegator.delegating_block): don't hide backtrace from
+	  __getobj__ and reduced exception messages when $DEBUG.
+
+Sat Feb  6 11:35:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/ipsocket.c (ip_addr, ip_peeraddr),
+	  ext/socket/socket.c (sock_s_getaddrinfo): added optional
+	  reverse_lookup flag.  [ruby-core:28007]
+
+Sat Feb  6 01:55:02 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/stringio/stringio.c (strio_ungetc): pads with \000 when the
+	  current position is after the end.  [ruby-dev:40271]
+
+Sat Feb  6 01:14:54 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/purelib.rb, common.mk: to simulate ruby command more precisely,
+	  remove "." from $: of virtual environment for build and test.
+
+Sat Feb  6 00:02:31 2010  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (fill_random_seed): don't use O_NOFOLLOW because
+	  /dev/urandom is a symlink in OpenSolaris.
+
+	* lib/securerandom.rb (SecureRandom.random_bytes): ditto.
+
+Fri Feb  5 16:38:08 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator): include copy of Kernel.
+	  [ruby-dev:40314]
+
+	* lib/delegate.rb (Delegator#{dup,clone}): class of copy should be
+	  Delegator.  [ruby-dev:40313]
+
+Fri Feb  5 09:26:54 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bignum.c (big_op): remove unused variables.
+
+Fri Feb  5 02:06:57 2010  Aaron Patterson  <tenderlove@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb: Struct members are emitted without a leading
+	  colon.  Thanks Yusuke Endoh! [ruby-core:28052]
+
+	* test/yaml/test_struct.rb: fixed tests to go with Struct changes
+
+	* test/yaml/test_yaml.rb: fixed tests to go with Struct changes
+
+Fri Feb  5 00:34:24 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (big_gt, big_ge, big_lt, big_ge): added Bignum#>, >=, < and
+	  <= to allow to compare with BigDecimal.  [ruby-dev:40167]
+
+Thu Feb  4 15:47:27 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (native_thread_init_stack): use get_stack.
+	  patched by KOSAKI Motohiro [ruby-dev:40309]
+
+	* thread_pthread.c (ruby_init_stack): use get_stack
+	  on platforms which have pthread_attr_get_np.
+	  (FreeBSD, DragonFlyBSD and NetBSD)
+	  This is because FreeBSD and DragonFly BSD must use
+	  pthread_attr_get_np to get stack size of main thread,
+	  but Mac OS X and Linux with LinuxThreads must use getrlimit.
+	  <http://www.nminoru.jp/~nminoru/programming/stackoverflow_handling.html>
+	  <http://d.hatena.ne.jp/nurse/20100204>
+
+Thu Feb  4 09:55:38 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: FreeBSD, DragonFly BSD and OpenBSD needs
+	  pthread_np.h to use pthread_*_np functions.
+	  OpenBSD's pthread_*_np also depend sys/signal.h,
+	  but it is included at signal.h via vm_core.h via thread.c.
+
+Thu Feb  4 08:15:53 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator): now inherits BasicObject.
+	  [ruby-dev:39154], [Bug #2679], [ruby-dev:40242]
+
+Thu Feb  4 03:00:59 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_DoDivmod): get red of
+	  floating point exception.
+
+Thu Feb  4 01:24:01 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_DoDivmod): fix infinity
+	  handling.  1 / Infinity was evaluated to NaN.
+
+Thu Feb  4 00:37:43 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_mdump, time_mload): dump/load utc_offset.
+	  [ruby-dev:40063]
+
+Wed Feb  3 22:22:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check for non-portable stack attribute functions.
+
+Wed Feb  3 20:10:52 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/iconv/charset_alias.rb: pass block argument to outer local
+	  variable.
+
+Wed Feb  3 20:08:05 2010  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (realpath_rec): rb_path_last_separator may return NULL.
+
+Wed Feb  3 13:15:24 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c: DragonFlyBSD is also the same as FreeBSD
+	  on getting the stack size of the main thread.
+
+Wed Feb  3 12:30:10 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack): use pthread_get_attr_np
+	  to get the stack size of the main thread on FreeBSD.
+
+	* thread_pthread.c: include pthread_np.h on FreeBSD.
+
+Wed Feb  3 11:38:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/{closure,function}.c: removed C99 features and warnings.
+
+Wed Feb  3 10:12:09 2010  Aaron Patterson  <tenderlove@ruby-lang.org>
+
+	* ext/dl/function.c: DL::Function now uses libffi
+
+	* ext/dl/cfunc.c (rb_dl_set_last_error): set to non static so errors
+	  can be exposed.
+
+	* ext/dl/closure.c: DL::Closure will now be used in place of
+	  ext/dl/callback/*.
+
+	* ext/dl/dl.c: legacy callbacks removed in favor of libffi
+
+	* ext/dl/dl_conversions.(c,h): used for converting ruby types to FFI
+	  types.
+
+	* ext/dl/callback/*: replaced by libffi callbacks.
+
+	* ext/dl/lib/dl/callback.rb: Converting internal callbacks to use
+	  DL::Closure
+
+	* ext/dl/lib/dl/closure.rb: Ruby parts of the new DL::Closure object
+
+	* ext/dl/lib/dl/import.rb: More conversion to use DL::Closure object
+
+	* ext/dl/lib/dl/value.rb (ruby2ffi): adding private method for
+	  DL::CPtr to ffi value conversion.
+
+Tue Feb  2 18:15:12 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c: turn on do_not_reverse_lookup by default,
+	  which has been reverted in r9880 probably unintentionally,
+	  according to matz.  [ruby-core:24530]
+
+Tue Feb  2 14:46:06 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c: move implementation of each_slice, each_cons,
+	  each_with_object to enum.c.
+
+	* enum.c (each_slice_i): convert multiple values from yield into
+	  an array.
+
+	* enum.c (each_cons_i): ditto.
+
+	* enum.c (each_with_object_i): ditto.
+
+Tue Feb  2 14:30:27 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_each_entry): new method #each_entry to pack values
+	  from yield into an array.
+
+	* lib/set.rb (Set#merge): use Enumerable#each_entry to implement
+	  Set compatible to 1.8 behavior.  [ruby-core:27985]
+
+	* lib/set.rb: replace is_a?(Enumerable) with respond_to?(:each)
+	  for duck typing.
+
+	* lib/set.rb (SortedSet#add): typo fixed.
+
+Tue Feb  2 11:13:56 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#marshal_dump): exclude
+	  delegator-specific instance variables.
+
+Mon Feb  1 21:26:41 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/matrix.rb (Vector#each2): returns a self.  [ruby-dev:40241]
+
+Mon Feb  1 17:08:42 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_minus): RDoc update.  a patch from red stun
+	  in [ruby-core:27951]
+
+Mon Feb  1 07:36:33 2010  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tkutil/tkutil.c: fix SEGV on TkUtil::CallbackSubst._setup_subst_table.
+
+	* ext/tk/lib/tk.rb: [ruby1.9] fix freeze at exit.
+
+	* ext/tk/lib/tk.rb: [POTENTIAL INCOMPATIBLE] return NoMethodError
+	  for TkWindow#to_ary and to_str.
+
+	* ext/tk/lib/tkextlib/tcllib/plotchart.rb: wrong arguments.
+
+	* ext/tk/sample/tkballoonhelp.rb: fail to support TkEntry widgets.
+
+Sun Jan 31 23:20:43 2010  wanabe  <s.wanabe@gmail.com>
+
+	* io.c (rb_io_each_codepoint): use cbuf when needs readconv.
+
+Sun Jan 31 23:07:23 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_dir.rb: use string instead of symbol as file name.
+
+Sun Jan 31 22:41:33 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (rb_string_value): fix the previous commit.
+
+Sun Jan 31 21:43:35 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (rb_string_value): make no exception for Symbol.
+	  [ruby-dev:40274]
+
+Sun Jan 31 21:10:15 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_insnhelper.c (vm_throw): fixed infinite loop.  [ruby-core:27969]
+
+Sun Jan 31 21:29:58 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rexml/text.rb (REXML::Text#initialize): do Text.check only when
+	  parent is specified, since Text.check may need doctype.  partially
+	  revert r26518.
+
+Sun Jan 31 15:50:34 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rexml/text.rb (REXML::Text#initialize): fix typo and a bug that
+	  seems to be caused by refactoring.
+
+Sun Jan 31 15:46:37 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rexml/parent.rb (REXML::Parent#delete): return the deleted node
+	  because the rdoc of REXML::Element#delete_element says it returns
+	  "the element that was removed."  [REXMLTracker#161]
+
+Sun Jan 31 14:33:00 2010  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: A bug fix for deleting blank Table rows from Andy Hartford.
+
+Sun Jan 31 13:31:43 2010  wanabe  <s.wanabe@gmail.com>
+
+	* gc.c (obj_free): free rb_classext_t of eigenclass.  [Bug #1392]
+
+Sun Jan 31 13:00:14 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rexml/document.rb (REXML::Document#add): fix duplicate XMLDecls
+	  and bad DocTypes in REXML::Document.    (Bug #19058) [ruby-core:27979]
+	  based on the patch by Federico Builes.
+
+Fri Jan 29 22:49:21 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/getoptlong.rb (set_options): ensure that the type of argument is
+	  Array, restoring this check that was deleted at r10239.  This caused
+	  rubyspec error.
+
+Fri Jan 29 12:59:33 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (mingw): needs $(DEFFILE) for extension libraries.
+	  [ruby-core:27946]
+
+Fri Jan 29 11:09:49 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_do): log no source when no development env.
+
+	* lib/mkmf.rb (create_makefile): srcprefix always needs $(srcdir).
+
+	* lib/mkmf.rb (create_makefile): yield configuration if a block is
+	  given.
+
+Fri Jan 29 09:43:30 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf8_mac.trans (buf_shift_char): don't see uninitialized
+	  value. [ruby-dev:40233]
+
+Fri Jan 29 01:42:24 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#initialize_copy): use initialize_copy
+	  instead of overriding clone/dup.  [ruby-dev:40221]
+	  it now always clones the target, it might cause incompatibility.
+
+Fri Jan 29 01:26:53 2010  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_i): update RDoc to
+	  denote that #to_i raises FloatDomainError for Inf and NaN.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_i): fast #to_i using
+	  BigDecimal_split().
+
+	* bignum.c (conv_digit): use faster ISDIGIT() assuming ASCII.
+
+Fri Jan 29 00:18:54 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/cgi.rb: set autoload to CGI::HtmlExtension.  [ruby-dev:40194]
+
+Thu Jan 28 09:44:19 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (onig_compile): initialize ScanEnv.
+	  mainly to initialize env->warnings_flag [ruby-dev:40196]
+
+	* regparse.c (scan_env_clear): clear warnings_flag.
+
+Wed Jan 27 23:33:21 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/matrix/test_matrix.rb, test/matrix/test_vector.rb: add some
+	  tests.
+
+Wed Jan 27 23:29:36 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/thread.rb (ConditionVariable#wait, signal, broadcast): return
+	  self (for 1.8 compatibility).
+
+Wed Jan 27 23:27:54 2010  Keiju Ishitsuka  <keiju@emperor2.pendome>
+
+	* lib/matrix.rb: add exception Matrix::ErrOperationNotImplemented
+	  [ruby-dev:40149].
+	* lib/matrix.rb: change message of exception
+	  Matrix::ErrOperationNotDefined [ruby-dev:40150], [ruby-dev:40176].
+	* lib/matrix.rb: add method Vector#/ [ruby-dev:40151].
+	* lib/matrix.rb(Matrix::Scalar#+,-,/): delete meaningless when
+	  switch. [ruby-dev:40149]
+
+Wed Jan 27 23:22:54 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* vm_dump.c (bugreport_backtrace): trivial change.
+
+	* vm_dump.c (rb_vm_bugreport): uninitialized local variable i.
+	  [ruby-dev:40169]
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_segv_test):
+	  follow above change.
+
+Wed Jan 27 23:20:52 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_exception.rb
+	  (TestException#test_thread_signal_location):
+	  change test method name.
+
+Wed Jan 27 22:48:40 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/matrix.rb (determinant): fix name error.
+
+Wed Jan 27 22:26:25 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_backtrace): "circular require" warning was output to
+	  stdout except the first line.  All line is output to stderr now.
+	  [ruby-dev:40147]
+
+Wed Jan 27 00:22:20 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_DoDivmod, BigDecimal_mod,
+	  BigDecimal_divmod, BigDecimal_div2): BigDecimal#% and #div returned
+	  an array itself that was returned by #divmod.
+
+Wed Jan 27 00:19:30 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpCtoV): do not raise infinity
+	  exception for BigDecimal("0E200000000000").
+
+Tue Jan 26 21:50:31 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpCtoV): BigDecimal("0E200000000000")
+	  was Infinity, not 0.
+
+	* test/bigdecimal/test_bigdecimal.rb: add a test for above.
+
+Tue Jan 26 21:36:22 2010  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: test unsetenv returns a value.
+	  unsetenv is void in older BSDs (FreeBSD 6 and OpenBSD 4.5 at least).
+
+	* hash.c (ruby_setenv): don't use the result of unsetenv if unsetenv
+	  doesn't return a value.
+
+Tue Jan 26 21:32:03 2010  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: suppress a warning.
+
+	* ext/extmk.rb: ditto.
+
+Tue Jan 26 20:23:22 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (rb_vm_invoke_proc): this function must not catch TAG_RETURN
+	  because vm_exec does.  This caused rubyspec error.  [ruby-dev:40158]
+
+Tue Jan 26 20:21:28 2010  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/eq.rb: fix circular require in drb.
+	  reported by akr. see [ruby-dev:40156] [ruby-core:27661]
+
+Tue Jan 26 19:59:17 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/zlib/zlib.c (rb_gzfile_s_wrap): add rdoc. [Bug #2656]
+	  patched by Hugh Sasse [ruby-core:27692] [ruby-core:27852]
+
+	* ext/zlib/doc/zlib.rd: removed.
+
+Tue Jan 26 16:43:34 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/strscan/strscan.c: fix rdoc. (length -> bytesize)
+	  reported by Kornelius Kalnbach. see [ruby-core:27792]
+
+Tue Jan 26 07:06:05 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/matrix.rb: suppress warnings.
+
+Tue Jan 26 03:16:45 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c, vm_core.h, eval.c: because rb_protect must not be jumped by
+	  callcc, revert r26407.  And rename trap_tag to protect_tag and
+	  change exception message (across trap -> across stack rewinding
+	  barrier).
+
+Mon Jan 25 23:08:10 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (do_select): wrong conditions.  [ruby-core:27753]
+
+Mon Jan 25 22:31:53 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/envutil.rb: use method_defined? instead of
+	  instance_methods.include?.
+
+Mon Jan 25 22:08:20 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c (rb_cont_call, cont_restore_1): remove trap_tag check because
+	  it seems not to make sense.  [ruby-dev:40121]
+
+	* vm_core.h, eval.c (rb_protect): ditto.
+
+Mon Jan 25 21:43:05 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/envutil.rb: fix the check if instance method `ruby' is
+	  defined or not.
+
+Mon Jan 25 21:17:32 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rubyoptions.rb (test_segv_test): add a test for
+	  bugreport trace dumper.
+
+Mon Jan 25 17:47:02 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (pty_check): needs WNOHANG to poll, return $?, and
+	  call raise_from_check() with pid_t.  [ruby-dev:40141]
+
+Mon Jan 25 17:36:11 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* version.c (RUBY_LIB, RUBY_*_LIB): moved from configures.
+
+Mon Jan 25 12:11:45 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (USE_SIGALTSTACK): only when SA_SIGINFO also is
+	  available.  see [ruby-core:27768].
+
+Mon Jan 25 12:02:34 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dln.c, file.c, io.c, signal.c: add __HAIKU__.
+	  patched by Alexander von Gluck [ruby-core:27767]
+
+Mon Jan 25 11:45:47 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* math.c (domain_check): ignore errno if y is inf.
+	  r26335 is because NetBSD 5.0's asin and acos returns
+	  0.0 with errno EDOM. But it breaks Linux whose gamma returns inf
+	  with errno ERANGE on.
+
+Sun Jan 24 22:48:05 2010  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, vm.c, vm_eval.c, vm_insnhelper.c: fix issues about
+	  return and c-return trace.  This issue skips (c-)return event
+	  with global jump such as break or return.  This fix make vm invoke
+	  hooks at stack rewind timing.  fix [ruby-core:27606] [Bug #2610].
+
+	* test/ruby/test_settracefunc.rb: add a test for above.
+
+Sun Jan 24 14:21:48 2010  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_enc_strlen_cr): increment by rb_enc_mbminlen(enc) for
+	  broken byte sequence.  [ruby-core:27748]
+	  (rb_str_inspect): ditto.
+
+Sun Jan 24 05:18:34 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): unlock all locking mutexes
+	  before clean up.  [ruby-core:26877]
+
+	* thread.c (rb_thread_atfork): no other threads to be joined.
+
+	* vm_core.h (rb_thread_lock_unlock, rb_thread_lock_destroy):
+	  new functions.
+
+	* vm.c (ruby_vm_destruct): unlock and destroy global VM lock.
+
+Sun Jan 24 00:31:39 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/parser/ruby.rb: fix typo.
+	  patched by Hal Brodigan [ruby-core:21536]
+
+Sun Jan 24 00:02:18 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/markup/to_html_crossref.rb: fix failure of the test.
+	  patched by Tomoyuki Chikanaga and nobu [ruby-core:20564]
+
+Sat Jan 23 23:27:12 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (dir_s_glob): add rdoc by Roger Pack. [ruby-core:27669]
+
+Sat Jan 23 23:12:56 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/generator/html.rb (RDoc::Generator::HTML#gen_into):
+	  make the rdoc(generating html) run faster and use less memory.
+	  patch by Tetsu Soh [ruby-core:27656]
+
+Sat Jan 23 19:54:48 2010  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE.ole_initialize,
+	  WIN32OLE.ole_uninitialize to use in win32ole.rb.
+	  You must not use these methods.
+
+	* ext/win32ole/lib/win32ole.rb: add win32ole.rb
+	  re-define Thread#initialize (fix ruby-core:27634)
+
+Sat Jan 23 00:25:19 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* sample/coverage.rb: preserve exit status.
+
+Sat Jan 23 00:21:18 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_dir.rb: get rid of debug print.
+
+	* test/ruby/test_module.rb: fixed to make test-all work.
+
+Fri Jan 22 23:54:04 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb: add a test for Array#rotate, rotate!.
+
+	* test/ruby/test_dir.rb, test/ruby/test_fnmatch.rb: add some tests
+	  (for coverage of dir.c).
+
+	* test/ruby/test_enum.rb: add a test for Enumerable#minmax.
+
+	* test/ruby/test_enumerator.rb: add some tests for Enumerator#inspect,
+	  Enumerator::Generator and Yielder.
+
+	* test/ruby/test_env.rb: add a test for ENV#index.
+
+	* test/ruby/test_exception.rb: add some tests (for coverage of
+	  error.c).
+
+	* test/ruby/test_hash.rb: add a test for recursive check.
+
+	* test/ruby/test_integer.rb: add a test for number of argument of
+	  Integer.
+
+	* test/ruby/test_method.rb: add a test for define_method.
+
+	* test/ruby/test_module.rb: add a test for constant of included
+	  module.
+
+	* test/ruby/test_proc.rb: add a test for parameters with cfunc.
+
+Fri Jan 22 23:50:03 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_regexp.rb, test/ruby/test_symbol.rb,
+	  test/ruby/test_variable.rb: add some tests (for coverage of
+	  compile.c).
+
+Fri Jan 22 21:05:34 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_mload): add submicro into vtm.subsecx.  [ruby-dev:40133]
+
+Fri Jan 22 14:26:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (rdoc): needs encodings and exts.
+
+Fri Jan 22 14:16:18 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: add missing comma.
+
+Fri Jan 22 01:07:16 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_bignum.rb, test/ruby/test_class.rb,
+	  test/ruby/test_defined.rb, test/ruby/test_hash.rb,
+	  test/ruby/test_primitive.rb, test/ruby/test_variable.rb: add some
+	  tests (for coverage).
+
+Fri Jan 22 01:03:28 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_require.rb (test_define_class): expect TypeError
+	  instead of NameError.  [ruby-core:27504]
+
+Thu Jan 21 15:09:35 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* configure.in: some tidy.
+
+Thu Jan 21 11:15:46 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_rotate): new methods, Array#rotate! and
+	  Array#rotate.  [ruby-dev:17194]
+
+	* array.c (rb_ary_reverse_m): copy directly.
+
+Thu Jan 21 09:38:00 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: use AS_CASE instead of when, to get rid of
+	  an unintentional substitution.
+
+Thu Jan 21 08:45:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_s_disasm): check for proc first.  based on the
+	  patch by Roger Pack in [ruby-core:27626].  [ruby-core:27227]
+
+Wed Jan 20 16:09:59 2010  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* common.mk (compile.$(OBJEXT)): dependencies lacking.
+
+	* vm_method.c (rb_add_method_cfunc): invalid initializer for C89
+
+	* compile.c (iseq_insns_unification): int might be smaller than int*
+
+Tue Jan 19 20:00:30 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::Config.default_config_hash): return an
+	  empty hash when resolv.conf is not available.  [ruby-core:27620]
+
+	* lib/resolv.rb (Resolv::DNS::Config#lazy_initialize): fixed the
+	  defaults of nameserver and port.
+
+Tue Jan 19 14:29:16 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (clean-ext): allow glob patterns.
+
+	* ext/extmk.rb: ditto.
+
+Tue Jan 19 14:19:26 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/zlib/{extconf.rb, zlib.c): crc32_combine and adler32_combine is
+	  supported on Zlib 1.2.2.1, so check them for old zlib.
+
+Tue Jan 19 09:03:37 2010  Aaron Patterson  <tenderlove@ruby-lang.org>
+
+	* ext/zlib/zlib.c: added Zlib.crc32_combine and Zlib.adler32_combine
+
+	* test/zlib/test_zlib.rb: corresponding tests [ruby-core:27551]
+
+Tue Jan 19 02:02:32 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb: add some tests (for coverage).
+
+	* test/ruby/test_bignum.rb: ditto.
+
+Tue Jan 19 01:57:12 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_bignum.rb: some coerce definitions (for test) was
+	  wrong.
+
+Tue Jan 19 01:53:11 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (bigsub_int): remove nonsense loop.
+
+Tue Jan 19 01:42:36 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (top_compstmt, top_stmts, top_stmt): prohibit BEGIN {} in
+	  non-toplevel scope.  [ruby-core:21657]
+
+	* test/ruby/test_beginendblock.rb (test_begininclass): add a test for
+	  above.
+
+Mon Jan 18 17:16:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/httpservlet/filehandler.rb (make_partial_content):
+	  add bytes-unit.  [ruby-dev:40030]
+
+Mon Jan 18 15:49:42 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTPHeader#{content_range,range_length}):
+	  use inclusive range same as the header representation.
+
+Mon Jan 18 03:59:57 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_instance_hexdigest_bang): Fix
+	  rdoc.
+
+	* ext/digest/lib/digest.rb (Digest::Class.base64digest)
+	  (Digest::Instance#base64digest{,!}): New methods.
+
+Sun Jan 17 22:48:44 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_instance_digest)
+	  (rb_digest_instance_hexdigest): Save a method call of reset()
+	  for a disposable clone.
+
+Sun Jan 17 19:24:25 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (domain_check): check errno first.
+	  NetBSD 5.0's asin and acos returns 0.0 with errno EDOM.
+
+Sun Jan 17 14:24:35 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (rb_str_derive): use long.
+
+	* ext/iconv/iconv.c (iconv_convert): suppress a warning.
+
+	* lib/mkmf.rb (check_signedness): new method.
+
+	* lib/mkmf.rb (have_header, create_header): use String#tr_cpp.
+
+Thu Jan 14 13:06:58 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_hash, int_chr): fixed type.
+
+Thu Jan 14 12:50:37 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_concat): fixed range check for Fixnum, and
+	  added checks for integer overflow and invalid char code.
+
+Thu Jan 14 09:34:31 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_concat): raise RangeError when the argument is
+	  negative value. [ruby-core:27583]
+
+Thu Jan 14 08:49:59 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_to_r): convert to rational if internal representation
+	  is not rational.
+
+Thu Jan 14 04:01:50 2010  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_mdump): use nano_num and nano_den instead of subnano to
+	  avoid Rational class in marshaled data which prevent unmarshal by
+	  Ruby 1.8.
+	  (time_mload): use nano_num and nano_den.
+
+Wed Jan 13 11:57:38 2010  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* object.c (rb_class_initialize): Make sure BasicObject doesn't get
+	  initialized twice [ruby-core:27577]
+
+	* class.c (rb_class_init_copy): ditto
+
+Wed Jan 13 06:54:44 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check for if struct timezone is defined.
+
+	* missing.h (struct timezone): define if not defined.
+
+	* win32/win32.h (struct timezone): defined in the newer w32api.
+	  [ruby-core:27515]
+
+Wed Jan 13 00:33:09 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/shell/command-processor.rb: fix typo by Sho Hashimoto.
+	  reported and patched at [ruby-dev:40058] [Bug #2599]
+
+Tue Jan 12 23:48:29 2010  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb, test/rss/test_maker_0.9.rb:
+	  accept any time format in maker. [ruby-core:26923]
+
+Tue Jan 12 21:56:00 2010  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_set_len): call rb_str_modify.
+
+	* file.c (realpath_rec): don't call rb_str_modify before
+	  rb_str_set_len.
+
+Tue Jan 12 20:44:14 2010  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (realpath_internal): call rb_secure.
+
+Tue Jan 12 16:48:46 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP.start): options may not be given.
+
+Tue Jan 12 16:48:03 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_readline): check if instream
+	  is closed instead of fd 0.
+
+	* ext/readline/readline.c (Init_readline): use STDIN for input.
+	  Reported by Sora Harakami. See
+	  http://d.hatena.ne.jp/codnote/20100111/1263174134
+
+Tue Jan 12 16:09:02 2010  wanabe  <s.wanabe@gmail.com>
+
+	* test/ruby/test_process.rb (test_execopts_env): MANDATORY_ENVS might
+	  not be a part of ENV.  e.g. TMPDIR.
+
+Tue Jan 12 14:07:31 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP.start): add hash argument to
+	  set ssl related options. when use_ssl is set default value
+	  of verify_mode is OpenSSL::SSL::VERIFY_PEER. [ruby-dev:40003]
+
+Tue Jan 12 14:53:07 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (init_env): use _wputenv() instead of
+	  SetEnvironmentVariableW() because latter doesn't set msvcrt's environ
+	  work area, of course.
+	  [Bug #2552]
+
+Tue Jan 12 13:33:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (realpath_rec): trace symbolic link only when supporting
+	  readlink().
+
+Tue Jan 12 12:49:39 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-here-doc-beg-match): fix for here-doc
+	  which ends with an underscore.
+
+Tue Jan 12 09:58:03 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c: need to include errno.h for EINVAL.
+
+	* hash.c (ruby_setenv): fixed typo.  see [ruby-dev:40026]
+
+Tue Jan 12 09:22:43 2010  Tanaka Akira  <akr@fsij.org>
+
+	* prelude.rb (require_relative): use File.realpath.  [ruby-dev:40040]
+
+	* include/ruby/intern.h: declare rb_dir_getwd.
+
+	* dir.c (rb_dir_getwd): copied from dir_s_getwd to export.
+	  (dir_s_getwd): use rb_dir_getwd.
+
+	* file.c (rb_file_s_realpath): new method File.realpath.
+	  (rb_file_s_realdirpath): new method File.realdirpath.
+
+	* lib/pathname.rb (Pathname#realpath): use File.realpath.
+	  (Pathname#realdirpath): use File.realdirpath.
+
+Mon Jan 11 22:45:08 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* hash.c (ruby_setenv): Improve the emulation of setenv(3) on
+	  environments where putenv(3) is used.  Raise EINVAL If a
+	  variable name contains an '='.
+
+Mon Jan 11 18:16:38 2010  wanabe  <s.wanabe@gmail.com>
+
+	* vm_insnhelper.h (GET_BLOCK_PTR): return 0 when in class frame.
+	  [Bug #2583]
+
+Mon Jan 11 16:52:05 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_strlen): added. [ruby-dev:40028]
+
+	* include/ruby/intern.h (rb_str_strlen): declared.
+
+Mon Jan 11 13:30:35 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/accesslog.rb (WEBrick::AccessLog#format): fixed typo.
+
+	* lib/webrick/httpstatus.rb (WEBrick::HTTPStatus::Status#initialize):
+	  accept 0 or more arguments.  [ruby-dev:40021]
+
+Mon Jan 11 12:47:58 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* hash.c (ruby_setenv): ENV.[]= should raise an error if setenv(3)
+	  or putenv(3) fails.  [ruby-dev:40023]
+
+Sun Jan 10 17:25:24 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/accesslog.rb : Escape needed.
+
+	* lib/webrick/httpstatus.rb : ditto.
+
+	* lib/webrick/httprequest.rb : ditto.
+
+	* lib/webrick/httputils.rb : ditto.
+
+	* test/webrick/test_cgi.rb (TestWEBrickCGI::test_bad_): Test for it.
+
+Sun Jan 10 04:54:36 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_define_class): raise TypeError same as class
+	  statement.  [ruby-core:27504]
+
+Sun Jan 10 04:41:20 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): needs ARCH_FLAG.
+
+Sat Jan  9 14:20:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (install-all): maybe typo.
+
+Sat Jan  9 14:01:14 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* missing/vsnprintf.c (BSD_vfprintf): get rid of a warning of VC++ x64.
+
+Sat Jan  9 08:40:54 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/configure.bat (WIN32DIR): regularise file separators.
+
+Fri Jan  8 23:35:18 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (mnew): don't check visibility of method body if public
+	  ZSUPER method is found.  [ruby-dev:39767]
+
+	* test/ruby/test_method.rb: add a test for above.
+
+Fri Jan  8 22:59:40 2010  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_method.c (rb_alias): skip ZSUPER method when searching body of
+	  source method.  [ruby-dev:39760]
+
+	* test/ruby/test_alias.rb: add a test for above.
+
+Fri Jan  8 21:15:21 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http, lib/net/https: move content from net/https to
+	  net/http. [ruby-dev:39986]
+
+Fri Jan  8 14:06:01 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_s_read): close the IO if an exception is raised on
+	  seeking. [ruby-core:27429]
+
+Fri Jan  8 13:12:26 2010  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_symbol): dump no encoding for 7bit only coderange
+	  symbol.  [ruby-core:27375]
+
+Thu Jan  7 07:56:09 2010  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.5.0 r5596.
+	* test/minitest/*.rb: ditto.
+
+Tue Jan  5 19:30:53 2010  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_exception.rb: add a test. cf [ruby-dev:39116]
+
+Tue Jan  5 02:03:34 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (DTRACE): clear this when `dtrace -l` fails.
+	  Note that current FreeBSD needs privilege to call it;
+	  so to use DTrace on FreeBSD, you should run configure as root.
+
+Mon Jan  4 21:24:18 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* gc.c: added UNLIKELY to probes for optimization.
+
+	* vm.c: ditto.
+
+	* thread.c: ditto.
+
+Mon Jan  4 09:30:54 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (TRACING_MODEL): follow yugui's previous changes.
+
+Mon Jan  4 09:30:25 2010  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (InterlockedExchangePointer): old SDK support.
+
+Sun Jan  3 23:54:51 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* trace.h: new file. wraps tracing mechanisms.
+
+	* defs/dtrace.d: new file. defined a dtrace provider "ruby".
+
+	* include/ruby/ruby.h (LIKELY): moved from vm.c.
+	  (UNLIKELY): ditto.
+	  (OBJSETUP): probe "object-create".
+	  (RUBY_EVENT_RESCUE): new event.
+
+	* vm_exec.c (DEBUG_ENTER_INSN): embedded a probe insn-entry into it.
+	  (DEBUG_END_INSN): insn-return.
+
+	* vm.c (LIKELY): moved into ruby.h.
+	  (UNLIKELY): ditto.
+	  (Init_BareVM): embedded a probe "raise" into it.
+
+	* variable.c (rb_class2name_without_alloc): new utility function.
+
+	* tool/rbinstall.rb (install?(:ext, :arch, :'ext-arch')): installs
+	  dtrace.d if necessary.
+
+	* thread_pthread.c (add_signal_thread_list): probe "raise".
+	  (rb_thread_create_timer_thread): ditto.
+
+	* thread.c (rb_thread_schedule_rec): probes "thread-enter" and
+	  "thread-leave",
+	  (thread_start_func_2): ditto.
+	  (thread_cleanup_func): probe "thread-term"
+
+	* lib/mkmf.rb: supports dtrace postprocessor on making an extension.
+
+	* iseq.c (rb_vm_insn_name): new utility function.
+	  (rb_vm_insn_len): ditto.
+
+	* insns.def (hook): probes "method-entry", "method-return", "line",
+	  and "rescue".
+
+	* compile.c (iseq_compile_each): adds a trace op for "rescue" probe.
+
+	* gc.c (garbage_collect): probes "gc-begin" and "gc-end".
+	  (obj_free): probe "object-free"
+	  (garbage_collect_with_gvl): probe "raise"
+	  (negative_size_allocation_error): ditto.
+	  (rb_memerror): ditto.
+
+	* eval.c (rb_rescue2): probe "rescue"
+	  (rb_longjmp): probe "raise"
+
+	* ext/probe/probe.c: new extension for application defined probes.
+
+	* ext/probe/extconf.rb: ditto.
+
+	* configure.in (--with-tracing-model): new option to choose a tracing
+	  mechanism.
+	  (DTRACE): new substitution. name of dtrace(1).
+	  (RUBY_TRACING_MODEL): new substitution.
+	  (DTRACE_OBJ): ditto.
+	  (MINIDTRACE_OBJ): ditto.
+	  (GOLFDTRACE_OBJ): ditto.
+	  (LIBRUBY_DTRACE_OBJ): ditto.
+	  (RUBY_DTRACE_POSTPROCESS): new macro. checks whether the dtrace on
+	  the system needs postprocessing.
+	  (RUBY_DTRACE_BSD_BROKEN): new macro. checks whether the dtrace
+	  supports USDT.
+
+	* Makefile.in:
+	  (DTRACE): new variable. name of dtrace(1).
+	  (TRACING_MODEL): new variable. name of the chosen tracing mechanism.
+	  (DTRACE_OBJ): same as the one in configure.in.
+	  (MINIDTRACE_OBJ): ditto.
+	  (GOLFDTRACE_OBJ): ditto.
+	  (LIBRUBY_DTRACE_OBJ): ditto.
+	  (CPPOUTFILE): new substitution. necessary for generating dtrace.d
+	  (trace_none.h): new target for TRACING_MODEL=none
+	  (RUBY_H_INCLUDES): appended a header for tracing.
+	  (distclean-local): also removes preprocessed version of dtrace.d
+	  ($(LIBRUBY_A)): needs $(LIBRUBY_DTRACE_OBJ) if dtrace needs
+	  postprocessing.
+	  ($(PROGRAM)): ditto.
+	  (golf): ditto.
+	  (miniruby): ditto.
+	  ($(arch_hdrdir)/ruby/dtrace.d): new target. preprocessed version
+	  of defs/dtrace.d. generated if necessary.
+	  ($(arch_hdrdir)/ruby/trace_dtrace.h): new target.
+	  definition of probes.
+	  ($(LIBRUBY_DTRACE_OBJ)): new target. generated if dtrace needs
+	  postprocessing.
+	  ($(DTRACE_OBJ)): ditto.
+	  ($(MINIDTRACE_OBJ)): ditto.
+	  ($(GOLFDTRACE_OBJ)): ditto.
+
+Sun Jan  3 15:34:19 2010  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/rexml/text.rb: String no longer has #each.
+	  Patch by Mitsutaka Mimura (takkanm). [ruby-dev:39949].
+
+Sun Jan  3 01:29:18 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Message::MessageDecoder#inspect):
+	  implemented.
+
+Sat Jan  2 15:57:54 2010  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS#initialize): new option :nameserver_port.
+	  (Resolv::DNS#each_resource): pass port number.
+	  (Resolv::DNS#make_requester): ditto.
+	  (Resolv::DNS::Config#lazy_initialize): initialize @nameserver_port
+	  instead of @nameserver.
+	  (Resolv::DNS::Config#single?): return port number addition to the
+	  nameserver.
+	  (Resolv::DNS::Config#generate_timeouts): use @nameserver_port.
+	  (Resolv::DNS::Config#resolv): yield port number.
+
+Sat Jan  2 00:43:22 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (initialize_regexp): allow leading
+	  and trailing white space, and forbid extra characters
+	  on another lines. [ruby-core:26223]
+	  RFC 3986  Appendix C.  Delimiting a URI in Context
+	  draft-duerst-iri-bis-07  7.2.  Web Address processing
+
+Fri Jan  1 23:17:49 2010  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI#normalize!): normalize case of
+	  scheme. [ruby-core:27309]
+
+Fri Jan  1 00:47:57 2010  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_instance_method_unimpl): Suppress
+	  compiler warnings.
+
+Fri Jan  1 00:00:00 2010  Tanaka Akira  <akr@fsij.org>
+
+	* tool/mkconfig.rb: generate RbConfig.ruby in rbconfig.rb.
+
+	* lib/webrick/httpservlet/cgihandler.rb: use RbConfig.ruby.
+
+	* test/ruby/envutil.rb: ditto.
+
+	* benchmark/report.rb: ditto.
+
+	* benchmark/runc.rb: ditto.
+
+	* tool/eval.rb: ditto.
+
+Thu Dec 31 18:18:55 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/rubysocket.h: include addrinfo.h only when using our own
+	  getaddrinfo.c.
+
+Thu Dec 31 14:20:11 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (save_redirect_fd): consider EBADF that the fd is not used.
+	  [ruby-dev:39938]
+
+Thu Dec 31 06:03:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_type): typed.
+
+	* ext/socket/extconf.rb: fix for wide-getaddrinfo option.
+
+	* ext/socket/addrinfo.c: rename {addr,name}info functions to ensure
+	  those are used on darwin.
+
+Thu Dec 31 03:27:53 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (native_thread_destroy): decreased the probability of
+	  using the interrupt event in the thread termination.
+	  see [ruby-core:27199].
+
+Thu Dec 31 02:35:57 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (w32_error): should report the function.
+
+Thu Dec 31 01:58:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: fix for extstatic.
+
+Wed Dec 30 19:43:00 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (RREGEXP_SRC_END): added.
+
+Wed Dec 30 19:40:00 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/defines.h (INFINITY): this is float.
+
+	* include/ruby/defines.h (NAN): ditto.
+
+	* numeric.c (rb_infinity): change content as float.
+
+	* numeric.c (rb_nan): ditto.
+
+Wed Dec 30 17:59:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gem_prelude.rb (push_all_highest_version_gems_on_load_path):
+	  simplified.
+
+	* lib/rubygems/command_manager.rb (Gem#load_and_instantiate):
+	  rescue only NameError from const_get.
+
+	* lib/rubygems/source_index.rb (Gem#load_specification): don't use
+	  RUBY_VERSION to branch.
+
+	* lib/rubygems/validator.rb (Gem::TestRunner, Gem#alien): ditto.
+
+	* lib/rubygems.rb: Kernel#gem is already defined, and workaround
+	  for home directory and custom_require are no longer needed.
+
+Tue Dec 29 16:03:33 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (Init_Numeric): Add Float::INFINITY and Float::NAN.
+	  [ruby-dev:1657] [ruby-dev:4760] [ruby-list:7023]
+	  [ruby-list:46690]
+	  [ruby-core:26632] [ruby-talk:41352] [ruby-talk:203333]
+
+	* include/ruby/defines.h (INFINITY): defined.
+
+	* include/ruby/defines.h (NAN): defined.
+
+	* include/ruby/util.h (ruby_div0): removed.
+
+	* numeric.c (fix_pow): use INFINITY and NAN
+	  instead of ruby_div0(1.0).
+
+	* marshal.c (r_object0): ditto.
+
+	* bignum.c (big_fdiv): ditto.
+
+Tue Dec 29 10:36:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser::STANDALONE):
+	  any number spaces can be placed between equal-sign and the value.
+	  patch from Ed Howland in [ruby-core:27345].
+
+Mon Dec 28 22:33:58 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): force to inherit standard I/O handles.
+	  this change fixes [ruby-core:27273], but other side effects might
+	  exist.
+
+Mon Dec 28 22:00:10 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_marshal.rb: added tests for taintness/untrustness
+	  propagation.
+
+Mon Dec 28 18:13:26 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* sample/occur2.rb: reimplemented in modern style. [ruby-dev:39927].
+
+Mon Dec 28 17:19:37 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/install-sh: correction by Hiro Asari.
+	  https://gist.github.com/264558/58ad1cae45cde49600bbb39286af2aae23e639d5
+
+Mon Dec 28 13:18:52 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/install-sh: wrote the intention. [ruby-dev:39928]
+
+Sun Dec 27 10:45:00 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (foleparam_initialize): add foleparam_initialize
+	  to check argument of WIN32OLE_PARAM.new
+
+	* test/win32ole/test_win32ole_param.rb (test_s_new): add some assertion
+	  to test WIN32OLE_PARAM.new
+
+Sun Dec 27 09:41:54 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/rbinstall.rb (install?(:local, :comm, :bin, :'bin-comm')):
+	  Makes it vim friendly.  __END__ in a heredoc is confusing with
+	  the script end for vim.
+
+Sun Dec 27 09:13:55 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* sample/occur2.rb: have been broken. fixed for Ruby 1.9 feature.
+
+Thu Dec 24 16:32:30 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/uri/generic.rb (eql?): Check the class of the compared object.
+	  Based on a patch by Peter McLain [ruby-core:27019]
+
+Thu Dec 24 15:20:03 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regexec.c (match_at): follow enclen's change.
+
+Thu Dec 24 12:08:00 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/delegate.rb (marshal_dump/load): dump & load instance variables
+	  by default [ruby-core:24211]
+
+Thu Dec 24 10:31:50 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/object.c (rb_obj_cmp): Default <=> operator returns 0 if
+	  objects are == [ruby-core:24063]
+
+Wed Dec 23 09:12:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/cgi/test_cgi_multipart.rb (CGIMultipartTest#_prepare):
+	  always must be binary mode.  c.f. Bug#2341
+
+Thu Dec 10 09:20:58 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (Init_Encoding): undef Encoding.new because
+	  a class which is rb_undef_alloc-func-ed can't call new method.
+	  [ruby-dev:39862]
+
+	* vm.c (Init_VM): undef RubyVM.new and RubyVM::Env.new.
+
+Mon Dec 21 17:51:44 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (optimize_node_left): include equal on the condition of for-loop.
+	  This bug also affects original Oniguruma. [ruby-core:27247]
+
+Mon Dec 21 10:03:33 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (print_enc_string): follow enclen's change.
+
+	* regcomp.c (onig_print_compiled_byte_code): ditto.
+
+	* regcomp.c (onig_print_compiled_byte_code): change prototype.
+
+	* regint.c (onig_print_compiled_byte_code): comment out.
+
+Mon Dec 21 08:04:34 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* object.c: BasicObject#initialize accepts any number of arguments
+	  [ruby-core:27080]
+
+Mon Dec 21 02:45:46 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_iterate): remove SEGV (retry).
+
+Sun Dec 20 23:18:23 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_iterate): remove SEGV (use the original patch).
+	  [ruby-dev:39874]
+
+Sun Dec 20 21:26:05 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (rb_iterate): pass current block when the argument bl_proc
+	  is NULL.  This behavior can be used to make enumerator faster
+	  [ruby-dev:39874]
+
+	* enumerator.c (enumerator_each): pass current block directly instead
+	  of trampoline block (enumerator_each_i).
+
+	* io.c (argf_each_line, argf_each_byte, argf_each_char): ditto.
+
+Sat Dec 19 14:59:30 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gem_prelude.rb (Kernel#gem): should make gem private.  a patch
+	  from Sho Hashimoto in [ruby-dev:39838].
+
+Sat Dec 19 14:57:17 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/webrick/config.rb (WEBrick::Config): typo fixed.  a patch
+	  from Sho Hashimoto in [ruby-dev:39835].
+
+Sat Dec 19 11:06:48 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (each2,collect2,map2): Fix enumerator
+	  [ruby-core:27225]
+
+Sat Dec 19 09:58:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (check_funcall): reset method_missing_reason before
+	  trying the call.  based on a patch from Yehuda Katz in
+	  [ruby-core:27219].
+
+Sat Dec 19 09:29:22 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/set.rb: Add checks that passed argument is Enumerable.
+	  [ruby-core:23844]
+
+Wed Dec 16 20:28:46 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/envutil.rb: fix a typo in assert message.
+
+Wed Dec 16 16:57:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/un.rb (httpd): try to convert port number to integer.
+
+Wed Dec 16 11:18:30 2009  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in(MINIRUBY): use "$BASERUBY" as a default ruby executable
+	  name.
+
+Wed Dec 16 00:53:14 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/find.rb (Find.find): rescue more exceptions which is possible to
+	  occur by other process change the directory tree.
+
+Tue Dec 15 09:06:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/find.rb (Find.find): get rid of race condition.
+
+Mon Dec 14 22:33:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/find.rb (Find.find): rescue only ENOENT and EACCES for lstat.
+
+Mon Dec 14 21:49:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/find.rb (Find.find): narrow rescue region.
+
+Mon Dec 14 09:20:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/find.rb (Find.find): removed already unnecessary code.
+
+Sun Dec 13 23:48:25 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/find.rb (Find.find): sort directory entries.  [ruby-dev:39847]
+
+Sun Dec 13 20:55:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (invoke_ruby): call to_str for stdin_data to
+	  reject non-string.
+
+Sun Dec 13 20:26:57 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (rb_parser_dump_tree): add prototype.
+
+Thu Dec 10 20:55:55 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode_data.h, transcode.c, tool/transcode-tblgen.rb: Added
+	  support for new transcoding instruction FUNsio (with Tatsuya Mizuno)
+
+	* enc/trans/gb18030.trans: Significantly reduced GB18030 conversion
+	  table footprint using FUNsio and differences (with Tatsuya Mizuno)
+
+	* test/ruby/test_transcode.rb: Minor name fix (from Tatsuya Mizuno)
+
+Thu Dec 10 17:22:36 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_eval.c (yield_under): yields self the same as 1.8.
+
+Thu Dec 10 15:49:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (native_mutex_reinitialize_atfork): release and
+	  re-acquire the lock at re-initialization.
+
+Thu Dec 10 12:56:02 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_replicate): add Encoding#replicate(name).
+
+	* encoding.c (enc_replicate_with_index): renamed from old
+	  enc_replicate.
+
+	* encoding.c (rb_enc_from_encoding_index): split from
+	  rb_enc_from_encoding.
+
+Thu Dec 10 09:15:00 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): CHAR_ESC_LEN should be 13.
+
+Thu Dec 10 01:12:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (dyna_pop_gen): pop dvars.  [ruby-dev:39861]
+
+Thu Dec 10 00:46:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.c (dump_node): fixed for long members.
+
+Wed Dec  9 22:57:04 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* node.c: node management added.  Currently, only pretty-dumper is
+	  implemented.  [ruby-dev:39853]
+
+	* ruby.c: --dump=parsetree and --dump=parsetree_with_comment options
+	  added.  This is just for debug or research purpose.  Note that the
+	  compatibility of these options are not supported at all.
+
+Wed Dec  9 09:50:35 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_justify): fixed the case a fill size is a
+	  multiple of the length of the padding.  [ruby-dev:39856]
+
+Tue Dec  8 23:41:34 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/configure.bat: now recognize --with-*-{dir,include,lib} options
+	  and pass them to mkmf.
+
+	* win32/configure.bat: general conversion from ``/'' to ``\'' of
+	  configure_args is not necessary any longer.
+
+	* win32/setup.mak: BASERUBY is used as command, so always need to
+	  convert ``/'' to ``\'' within it.
+
+Tue Dec  8 23:39:52 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (flo_eq): suppress a warning on VC++ for x64.
+
+Tue Dec  8 16:19:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems: update to 1.3.5.
+
+	* lib/rubygems/defaults.rb (Gem::default_dir): removed a clause
+	  doing nothing.
+
+Tue Dec  8 03:50:24 2009  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (rb_f_test): use string form in unknown command error
+	  message.
+
+Mon Dec  7 14:11:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): reverted r26007.  [ruby-dev:39845]
+
+	* test/test_delegate.rb (test_marshal): moved from test_marshal.rb.
+
+Mon Dec  7 13:05:59 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* string.c (rb_str_justify): CVE-2009-4124.
+	  Fixes a bug reported by
+	  Emmanouel Kellinis <Emmanouel.Kellinis AT kpmg.co.uk>, KPMG London;
+	  Patch by nobu.
+
+Sun Dec  6 23:50:46 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* strftime.c: %l should be 1..12 instead of 0..12 [ruby-core:27072]
+
+Sun Dec  6 23:16:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/find.rb (Find.find): reduce stat system call.
+
+Sun Dec  6 16:02:15 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httpservlet/filehandler.rb: escape filename of index.
+	  [ruby-dev:37768]
+
+Sun Dec  6 00:35:16 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup (config.h): EXECUTABLE_EXTS moved from
+	  dln.c:dln_find_1().
+
+Sat Dec  5 15:35:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): dump instance variables when using
+	  marshal_dump.   [ruby-core:24211]
+
+	* variable.c (rb_ivar_count): added.
+
+Sat Dec  5 13:19:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: default ac_cv_prog_CC to CC.
+
+Sat Dec  5 10:18:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb/extend-command.rb (def_extend_command): fixed argument
+	  number for negative arity.
+
+Fri Dec  4 16:50:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (k_def): adjust the location of method definition to the
+	  line of def.  [Bug #2427]
+
+Fri Dec  4 19:05:28 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (defined): should respect #respond_to_missing? as
+	  #respond_to? does.
+
+Fri Dec  4 15:50:18 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_eval.c (yield_under): does not yield self, and passes blockptr
+	  instead of &block to vm_cref_push().  [ruby-dev:39833]
+
+Fri Dec  4 15:15:43 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (num_exact): should not accept strings as operands, even
+	  though they respond to #to_r.  ideally, strict rational
+	  conversion should be done by a method like #to_rational, not #to_r.
+	  [ruby-core:23729]
+
+Fri Dec  4 13:10:23 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/stringio/stringio.c (ungetc): RDoc updated.  trunk allows
+	  pushing back behind the beginning of the pseudo stream.
+
+Fri Dec  4 03:10:38 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* compile.c (compile_cpath, iseq_compile_each): reverted
+	  constant/class variable lookup in instance_eval etc. to the
+	  behavior of 1.8.
+
+	* eval.c (rb_mod_nesting): ditto.
+
+	* insns.def (putspecialobject, defineclass): ditto.
+
+	* node.h (NODE_FL_CREF_PUSHED_BY_EVAL): ditto.
+
+	* vm_core.h (VM_SPECIAL_OBJECT_CONST_BASE): ditto.
+
+	* vm_eval.c (yield_under, eval_under): ditto.
+
+	* vm_insnhelper.c (vm_cref_push, vm_get_const_base,
+	  vm_get_ev_const, vm_get_cvar_base): ditto.
+
+Thu Dec 3 20:27:27 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/gb18030-tbl.rb: Fix omission of C1 region in code table
+	  (from Tatsuya Mizuno)
+
+	* test/ruby/test_transcode.rb: Added test for converting full range of
+	  Unicode codepoints from/to GB18030 (from Tatsuya Mizuno)
+
+Wed Dec  2 23:51:28 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c: initialize @hostname of SSLSocket to avoid
+	  warning at SSLSocket#connect.
+
+Tue Dec  1 18:01:43 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/digest/test_digest_extend.rb: added tests for digest framework.
+
+Tue Dec  1 12:01:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (dyna_push_gen, dyna_pop_gen): no needs to save both of
+	  args and vars.  [ruby-core:26961]
+
+	* parse.y (dyna_push_gen): use rb_node_newnode() for ripper.
+	  [ruby-core:26961]
+
+	* parse.y (dyna_push_gen, dyna_pop_gen): adjust local vtable level
+	  for the case of syntax errors in method name or argument inside
+	  do block.   [ruby-core:26961]
+
+Mon Nov 30 16:57:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (command_output): $makeflags are already quoted.
+
+Mon Nov 30 16:54:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, win32/Makefile.sub (EXECUTABLE_EXTS): moved from
+	  dln.c:dln_find_1().
+
+	* lib/mkmf.rb (def find_executable0): use EXECUTABLE_EXTS, not
+	  only EXEEXT.  [ruby-core:26821]
+
+Mon Nov 30 11:00:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): suppress an extra error message after
+	  numeric literal without digits.  based on a patch from ujihisa .
+	  in [ruby-dev:39811].  [ruby-dev:39798]
+
+Sun Nov 29 16:56:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (check_funcall_failed): pass ID.  [ruby-core:26934]
+
+Sun Nov 29 06:37:53 2009  Aaron Patterson  <tenderlove@ruby-lang.org>
+
+	* lib/rexml/formatters/default.rb (write_attribute): fix an
+	  exception when printing a document when duplicate namespaced
+	  attributes exist. Thanks, Alexey Froloff [ruby-core:26837]
+
+Sat Nov 28 09:05:53 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_eval.c (check_funcall_failed): should rescue user raised
+	  NoMethodError.  rescue all NoMethodError if receiver does not
+	  respond to the method name.  [ruby-dev:39796]
+
+Thu Nov 26 21:14:30 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_complex.rb (@unify): fix the detection if math
+	  loaded. This makes test_complex.rb work fine with the previous
+	  commit.
+
+	* test/ruby/test_rational.rb (@unify): ditto.
+
+Thu Nov 26 21:13:36 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/test_mathn.rb (TestMathn): new test case.
+	  test for r25067.
+
+Thu Nov 26 21:11:23 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/openssl/test_config.rb (OpenSSL::TestConfig): new test case.
+	  test for r25017.
+
+Thu Nov 26 21:08:54 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_range.rb (TestRange#test_comparison_when_recursive):
+	  test for r25010.
+
+	* test/ruby/test_struct.rb (TestStruct#test_comparison_when_recursive):
+	  ditto.
+
+Thu Nov 26 20:18:02 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gem_prelude.rb (Gem.set_home): must dup before force_encoding
+	  and must force_encoding before gsub.
+	  cf. Yen Sign problem of SJIS [ruby-core:26910]
+
+Thu Nov 26 17:54:37 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{configure.bat, setup.mak, Makefile.sub}: add new configure
+	  option ``--with-ntver''.
+
+Thu Nov 26 11:42:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/mkmf/base.rb: use $INCFLAGS to add -I option.  [Bug#2387]
+
+Thu Nov 26 07:17:58 2009  wanabe  <s.wanabe@gmail.com>
+
+	* marshal.c (mark_dump_arg): mark str.  see also [ruby-dev:39735]
+
+Thu Nov 26 00:05:58 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/digest/test_digest_extend.rb: Added tests for current digest
+	  framework.
+
+Wed Nov 25 20:46:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* vm_eval.c (rb_search_method_entry): refine error message.
+
+Wed Nov 25 19:29:05 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_instance_method_unimpl): Do not
+	  call rb_inspect() on an object that does not implement necessary
+	  methods; reported by NaHi.
+
+Wed Nov 25 19:30:30 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c: Added a check for an internal error
+	  (with Tatsuya Mizuno)
+
+Tue Nov 24 22:57:46 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (EnvUtil.invoke_ruby): raise Timeout::Error
+	  instead of flunk.  reported by Yusuke Endoh.
+
+Tue Nov 24 22:31:44 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (ruby_vm_destruct, thread_memsize): fix argument type to make
+	  RUBY_MARK_FREE_DEBUG available.
+
+Tue Nov 24 21:25:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* error.c: include errno.h at beginning.
+
+Tue Nov 24 20:11:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c: %Y format a year with 4 digits at least.
+
+	* lib/time.rb: format a year with 4 digits at least.
+
+Tue Nov 24 20:05:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* defs/known_errors.def: more errors.
+
+Tue Nov 24 20:01:49 2009  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (rb_bug_errno): declared.
+
+	* include/ruby/intern.h (rb_strerrno): declaration removed.
+
+	* error.c (rb_strerrno): make it static.  return NULL for unknown
+	  errors.
+	  (rb_bug_errno): defined.
+
+	* thread_pthread.c: use rb_bug_errno.
+
+	* signal.c (ruby_signal): use rb_bug_errno.
+
+Tue Nov 24 10:17:38 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_path_convert): fix fs_encoding is not assign.
+
+Tue Nov 24 10:00:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_strerrno): constified.
+
+Tue Nov 24 09:49:31 2009  Tanaka Akira  <akr@fsij.org>
+
+	* error.c (rb_strerrno): return "UNKNOWNERROR" for non-zero unknown
+	  error.
+
+Tue Nov 24 09:18:33 2009  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_strerrno): declared.
+
+	* template/known_errors.inc.tmpl: generate defined_error() and
+	  undefined_error() instead of set_syserr.
+
+	* error.c (Init_syserr): define defined_error() and undefined_error()
+	  to follow the above change.
+	  (rb_strerrno): defined.
+
+	* thread_pthread.c: show error message and errno macro name with
+	  rb_bug.
+
+Mon Nov 23 16:06:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (RUBY_STACK_MIN, RUBY_STACK_SPACE): delay for
+	  platforms where PTHREAD_STACK_MIN is not compile time constant.
+	  [ruby-dev:39751]
+
+Mon Nov 23 11:26:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_find_1): removed duplication.
+
+Mon Nov 23 04:12:00 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/newline.trans (fun_so_universal_newline): generate \n
+	  after \r\n detection instead of just after \r.
+	  [ruby-list:45988] [ruby-core:25881] [ruby-core:26788]
+
+Sat Nov 21 18:48:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* vm_eval.c (rb_search_method_entry): show flags and klass value in
+	  not implemented error message.
+
+Sat Nov 21 16:38:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (marshal_dump): use normal object as the buffer so
+	  that no hidden object is exposed to ruby-level.  [ruby-dev:39744]
+
+Sat Nov 21 15:58:43 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (read_all): fix: false negative invalid byte sequence
+	  on reading from pipes. [ruby-dev:39743]
+	  fix: assign the variable 'pos' as relative value from recent pos.
+
+Sat Nov 21 14:44:16 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_path_convert): delay getting UTF8-MAC encoding
+	  while really needed. [ruby-core:26807]
+
+Fri Nov 20 21:16:54 2009  Tanaka Akira  <akr@fsij.org>
+
+	* vm_eval.c (rb_search_method_entry): avoid trigraph.
+
+Thu Nov 19 23:17:06 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (putbinaryfile): use APPE for resume.
+	  Thanks, Tomoyuki Chikanaga.
+
+Thu Nov 19 22:50:05 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (flag_list): untaint strings to intern in the safe
+	  level 1.
+
+	* lib/net/imap.rb (max_flag_count=): new methods to set the max
+	  number of flags interned to symbols.
+
+Thu Nov 19 20:43:39 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/getnameinfo.c: need to include extconf.h for HAVE_* macros.
+	  reported by Kenta Murata <mrkn AT mrkn.jp> via IRC.
+
+Thu Nov 19 17:00:59 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub: nmake execute the file named echo if it exists
+	  in the PATH.  reported by Kenta Murata <mrkn AT mrkn.jp> via IRC.
+
+Thu Nov 19 02:50:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* vm_eval.c (rb_search_method_entry): show the type of the hidden
+	  object.
+	  (rb_type_str): new function for above.
+
+Thu Nov 19 00:47:09 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (MORE_CHAR_SUSPENDED): renamed from MORE_CHAR_CBUF_FULL.
+
+Wed Nov 18 22:00:13 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (fill_cbuf): extracted from more_char.
+	  (io_shift_cbuf): fix memmove condition.
+	  (read_all): use fill_cbuf directly to avoid ECONV_AFTER_OUTPUT.
+	  [ruby-dev:39708]
+
+Wed Nov 18 18:25:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_atfork_internal): reinitialize global lock
+	  at fork to get rid of deadlock.  based on the patch from Hongli
+	  Lai in [ruby-core:26783].  [ruby-core:23572]
+
+Wed Nov 18 17:48:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (terminate_atfork_i): all mutex locks by other threads
+	  have been abandoned at fork.
+
+Wed Nov 18 15:27:20 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (file_path_convert): delay getting filesystem encoding
+	  while really needed.
+
+Wed Nov 18 12:33:42 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_set_default_encoding): reset filesystem
+	  encoding because on resetting default_external because
+	  Unix's filesystem encoding depends on default_external.
+
+	* encoding.c (enc_set_filesystem_encoding): added.
+
+	* ruby.c (process_options): don't call rb_filesystem_encoding
+	  because filesystem encoding is reset when default_external
+	  is reset.
+
+Wed Nov 18 11:57:32 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* math.c (math_gamma): fix incorrect comparison expression.
+	  see also [ruby-dev:39709] [Bug #2381]
+
+Wed Nov 18 11:37:05 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_scan_open_args): move path encoding conversion
+	  for filesystem encoding of Mac OS X.
+
+	* file.c (file_path_convert): added for convert encoding
+	  of file path.
+
+	* file.c (rb_get_path_check): add file_path_convert.
+
+Wed Nov 18 10:17:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_utf8_nth): no count_utf8_lead_bytes_with_word
+	  optimization for short string.  [ruby-core:26787]
+
+	* string.c (str_utf8_offset): str_utf8_nth never return NULL.
+
+Wed Nov 18 10:12:34 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_undef): should raise TypeError if klass is nil.
+	  1.instance_eval { undef to_s } causes SEGV before this fix.
+
+	* test/ruby/test_undef.rb: new tests for undef.
+
+Wed Nov 18 08:41:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_utf8_nth): fixed overrun.  [ruby-core:26787]
+
+Wed Nov 18 07:51:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (parse_mode_enc): fix invalid access.
+
+Tue Nov 17 23:50:06 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_alias): should raise TypeError if klass is nil.
+	  1.instance_eval { alias to_string to_s } causes SEGV before this
+	  fix.
+
+	* test/ruby/test_alias.rb (test_special_const_alias): ditto.
+
+Tue Nov 17 17:53:53 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/big5.c, enc/trans/big5.trans, enc/trans/big5-uao-tbl.rb,
+	  test/ruby/test-transcode.rb: Added Encoding 'Big5-UAO' and transcoding
+	  for it (from Tatsuya Mizuno) (see Bug #1784)
+
+Tue Nov 17 16:26:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (opt_case_dispatch): runtime value cannot be used as
+	  an element initializer.
+
+	* vm_insnhelper.c (opt_case_dispatch_i): gets rid of type-punning
+	  calls.
+
+Mon Nov 16 15:51:53 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): protected singleton methods of
+	  an object should not be able to called from other instances of the
+	  class of the object.  [ruby-core:26761]
+
+	* vm_eval.c (rb_method_call_status): ditto.
+
+	* test/ruby/test_module.rb (test_protected_singleton_method): ditto.
+
+Mon Nov 16 14:03:53 2009  wanabe  <s.wanabe@gmail.com>
+
+	* io.c (read_all): shift read buffer if exception occurred.
+	  pointed out in [ruby-dev:39702].
+
+Mon Nov 16 07:59:38 2009  wanabe  <s.wanabe@gmail.com>
+
+	* io.c (read_all): don't call io_shift_cbuf until buffering enough or
+	  econv_finished.   [ruby-dev:39696]
+
+	* io.c (more_char): don't call clear_readconv to read buffer after
+	  econv_finished.
+
+	* io.c (appendline, rb_io_each_codepoint): clear readconv when done.
+
+Mon Nov 16 01:58:08 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* include/ruby/ruby.h (rb_classext_t): annotate @internal.
+
+Mon Nov 16 01:35:34 2009  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_filesystem_encindex): use default external encoding
+	  instead of locale encoding in Unix.
+
+	* ruby.c (process_options): delay filesystem encoding
+	  initialization until default external encoding initialization.
+
+Mon Nov 16 00:06:26 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (thread_cleanup_func): delete locking_mutex when thread
+	  object become dummy because of fork.  [ruby-core:26744]
+	  [ruby-core:26745]
+
+	* bootstraptest/test_thread.rb: add a test for above.
+
+Sat Nov 14 21:54:46 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* class.c (rb_mod_init_copy): fix memory leak of Class#dup.
+	  [ruby-dev:39687]
+
+Sat Nov 14 17:09:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--with-opt-dir): ignore and suppress a warning.
+	  [ruby-dev:39684]
+
+Sat Nov 14 16:43:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (ruby_setenv): get rid of crash in Solaris 8 and 10.
+	  [ruby-core:26668]
+
+Sat Nov 14 09:16:54 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimalCmp): Fix comparisons
+	  [ruby-core:26646]
+
+	* test/bigdecimal/test_bigdecimal.rb (class): Fix and improve tests.
+
+Sat Nov 14 04:07:06 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb (TkVariable::coerce): fix bug on a
+	  numeric value.
+
+Sat Nov 14 03:35:29 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb: TkVariable#*(other) and /(other) have a
+	  bug on handling of the "other" value.
+
+Fri Nov 13 21:18:15 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/zlib/zlib.c (rb_zlib_adler32): fix typo.
+
+Fri Nov 13 16:22:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): see first_func, not first_proc,
+	  to decide which to use.
+
+	* gc.c (vm_xrealloc): use the given object space.
+
+Fri Nov 13 00:46:24 2009  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/util.rb (CGI::pretty): fix the overflow bug
+	  if stripped string.[ruby-dev:37975]
+
+Thu Nov 12 23:08:11 2009  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (read_from_cmdline): this code is better.
+	  [ruby-core:25991]
+
+Thu Nov 12 22:55:42 2009  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/net/telnet.rb: don't use simple delegate.
+	  because SimpleDelegate behavior changed.
+
+Thu Nov 12 14:33:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_win32.c (thread_errno): CreateThread does not set errno.
+
+	* thread.c (thread_create_core): moved failure handling from
+	  native_thread_core().
+
+	* thread_pthread.c (native_thread_create): constified.
+
+Thu Nov 12 10:08:56 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .document: remove documents not in rdoc format until
+	  rdoc supports non rdoc files. [ruby-core:26459]
+
+Thu Nov 12 06:42:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#initialize): option hash may not be
+	  given.  [ruby-core:26681]
+
+Thu Nov 12 01:29:15 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enumerator.c (yielder_yield_push): Yielder#<< should return self.
+	  [ruby-dev:39660]
+
+Wed Nov 11 19:17:35 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#initialize): merge mode option.
+
+	* lib/tmpdir.rb (Dir::Tmpname#create): splat options.
+
+Wed Nov 11 12:54:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (ruby_setenv): use ruby_strdup().
+
+Wed Nov 11 12:19:27 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (ruby_setenv): also set CRT workarea.  ref [ruby-core:25010]
+
+Wed Nov 11 09:36:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object, r_object0): use RHASH_IFNONE but not ifnone
+	  directly.
+
+Wed Nov 11 08:32:45 2009  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (blocking_region_begin): define before BLOCKING_REGION.
+	  reported by Luis Lavena.  [ruby-core:26670]
+
+Wed Nov 11 08:22:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* util.c (ruby_strtod): use dval() consistently.
+
+Wed Nov 11 02:14:48 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/matrix/test_matrix.rb (TestMatrix#test_rank):
+	  added a test method for r24969.
+
+Tue Nov 10 08:23:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_f_exec, rb_f_system, rb_f_spawn): mentioned about
+	  the shell to be used when the command line is single string.
+	  See [ruby-core:26652]
+
+Mon Nov  9 20:15:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (rb_gc_save_machine_context): don't save the stack pointer
+	  in this function.
+	  (RB_GC_SAVE_MACHINE_CONTEXT): call rb_gc_save_machine_context and
+	  save the stack pointer at caller side.
+	  (rb_thread_schedule_rec): use RB_GC_SAVE_MACHINE_CONTEXT instead of
+	  rb_gc_save_machine_context.
+	  (blocking_region_begin): changed to a macro.  use
+	  RB_GC_SAVE_MACHINE_CONTEXT instead of rb_gc_save_machine_context.
+	  [ruby-dev:39659]
+
+Mon Nov  9 11:28:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (warnflags): add -Wno-long-long.
+
+Sun Nov  8 18:13:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c: use RHASH_IFNONE but not ifnone directly.
+
+Sat Nov  7 21:02:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb (PP::ObjectMixin#pretty_print): use to_s regardless of
+	  instance variables.
+
+Sat Nov  7 20:59:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_push_frame): get rid of out-of-bounds
+	  access.
+
+Sat Nov  7 12:41:19 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_insnhelper.c (VM_CALLEE_SETUP_ARG): revert r25521.
+	   [ruby-core:26427] [ruby-core:26447]
+
+Fri Nov  6 18:33:47 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_push_frame): add CHECK_STACK_OVERFLOW.
+	  [ruby-dev:39592]
+
+	* eval.c (rb_longjmp): add 1 level backtrace for sysstack_error
+	  without calling any method to prevent further stack overflow.
+
+	* eval.c (make_exception): don't call #exception for
+	  sysstack_error to prevent stack overflow.
+
+	* proc.c (Init_Proc): don't freeze sysstack_error.
+
+	* eval.c (rb_longjmp): move reentrant check after exception
+	  preparation.
+
+Fri Nov  6 17:13:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_header): split the line by tabs.
+
+Fri Nov  6 12:02:32 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (MINIDLNOBJ): set default as dmydln.o.
+	  Now LLVM clang seems to be able to build Ruby 1.9.
+
+Fri Nov  6 09:29:32 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encindex): add fallback to
+	  ASCII-8BIT on Windows.
+
+Fri Nov  6 07:29:07 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_case.rb: merged r25658 from ruby_1_8.
+
+Thu Nov  5 12:27:07 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gem_prelude.rb (Gem.user_home): force_encoding(
+	  Encoding.find('filesystem')). [ruby-core:26525]
+
+Thu Nov  5 07:47:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (BEGIN): now local_push and local_pop are necessary in
+	  ripper.
+
+Wed Nov  4 21:39:10 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .document: add some README files to default rdoc's of trunk.
+	  patched by Roger Pack [ruby-core:26459]
+
+Wed Nov  4 16:06:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/st.h (SIZEOF_ST_INDEX_T): moved from st.c for
+	  Init_RandomSeed().
+
+Wed Nov  4 15:50:16 2009  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* include/ruby/ruby.h (NUM2CHR): prefix __extension__ for
+	  braced-groups within expressions.
+
+Wed Nov  4 15:46:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (set_{internal,external}_encoding_once): fixed typos.
+
+Wed Nov  4 12:49:18 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c (mnew): Fix scope issue [ruby-core:26069]
+
+Wed Nov  4 08:50:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_hash_start): moved from string.c.
+
+	* random.c (Init_RandomSeed2): register global address before set.
+
+	* random.c (Init_RandomSeed): initialize hashseed.
+
+Wed Nov  4 08:26:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_check): moved other checks from GetDIR.
+
+	* dir.c (GetDIR): fixed a variable name.
+
+Wed Nov  4 08:19:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (rb_fiber_start): fixed unconditional test.
+
+Wed Nov  4 02:08:14 2009  Aaron Patterson  <tenderlove@ruby-lang.org>
+
+	* ext/dl/cptr.c (rb_dlptr_eql, rb_dlptr_cmp): DL::CPtr#== and DL::CPtr#<=>
+	  should not raise an exception when compared to a different object.
+
+Wed Nov  4 00:05:36 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_upto): make next object before yield its block.
+	  fix: can modify original begin string of String#upto.
+	  [ruby-dev:26384] [ruby-dev:39626]
+
+Mon Nov  2 18:33:21 2009  wanabe  <s.wanabe@gmail.com>
+
+	* cont.c (fiber_free): don't free unallocated local_storage. see #1325.
+
+	* cont.c (cont_init): clear local_storage not to use current thread's.
+
+	* cont.c (fiber_t_alloc, root_fiber_alloc): link itself always for
+	  a case that fiber_link_remove() is called before fiber_link_join().
+
+	* cont.c (fiber_init): clear cont->vm_stack and th->stack before
+	  root_fiber_alloc() in rb_fiber_current().
+
+Mon Nov  2 14:52:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_exc_raise, rb_exc_fatal, rb_make_exception):
+	  suppressed shorten-64-to-32 warnings.
+
+Mon Nov  2 14:43:48 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_pthread.c (native_stop_timer_thread): delay joining timer
+	  thread after unlocking mutex.
+
+Mon Nov  2 13:31:14 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_pthread.c (native_stop_timer_thread): need to join timer thread
+	  only when really stopping it.
+
+Mon Nov  2 12:55:50 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_{pthread,win32}.c (native_stop_timer_thread): join the thread
+	  here.
+
+	* thread_{pthread,win32}.c (native_reset_timer_thread): new function.
+
+	* thread.c (rb_thread_stop_timer_thread, rb_thread_reset_timer_thread):
+	  call above function instead of simply setting 0.
+
+Mon Nov  2 11:22:19 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/registry.rb: update rdoc. [ruby-core:26022]
+
+Sun Nov  1 20:16:03 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: fixed rdoc. [ruby-core:26457]
+
+Sun Nov  1 16:24:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_stack_grow_dir): fix for universal binary.
+
+	* configure.in (sitehdrdir, vendorhdrdir): fixed default values.
+
+Sun Nov  1 13:31:16 2009  wanabe  <s.wanabe@gmail.com>
+
+	* win32/win32.c (overlapped_socket_io, recvmsg, sendmsg): pass handle
+	  instead of pointer to CloseHandle().
+
+Sun Nov  1 13:11:27 2009  wanabe  <s.wanabe@gmail.com>
+
+	* ext/socket/socket.c (socket_s_ip_address_list): use FreeLibrary() to
+	  free HMODULE.
+
+Sun Nov  1 08:17:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (GET_STACK_BOUNDS): refactored common code.  based on a
+	  patch from Suraj N. Kurapati <sunaku AT gmail.com> in
+	  [ruby-core:26443].
+
+Sat Oct 31 23:44:35 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_count): remove optimization using #size.
+	  revert r25560.
+
+	* vm_eval.c (rb_funcall_no_recursive): remove method.
+
+Sat Oct 31 23:28:49 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_eval.c (check_funcall): logic updated according to
+	  [ruby-dev:39594].  search method entry, call if it exists;
+	  otherwise check method_missing, call if it was overridden,
+	  protecting exceptions; if NoMethodError happens, check method
+	  name.
+
+	* vm_eval.c (vm_call0): use idMethodMissing.
+
+	* vm_eval.c (rb_search_method_entry): typo fixed.
+
+Sat Oct 31 17:19:28 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTPResponse#each_response_header):
+	  cosmetic: '?\ ' -> '?\s'
+
+Fri Oct 30 22:09:47 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTPResponse#each_response_header):
+	  accept multiline message header of HTTP response.  see #1796.
+	  cf. RFC 2616 '4.2 Message Header'.
+
+	* test/net/http/test_httpresponse.rb: added.
+
+Fri Oct 30 18:54:04 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (trnext): detect empty range and raise exception.
+	   [ruby-dev:39108]
+
+Fri Oct 30 17:01:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (enum call_type): get rid of last comma.
+
+	* vm_eval.c (vm_call0, vm_call_super, rb_f_send, rb_f_public_send):
+	  fixed call_type.  [ruby-dev:39581]
+
+	* vm_eval.c (rb_search_method_entry, rb_method_call_status): split
+	  from rb_call0().
+
+	* vm_eval.c (rb_check_funcall): get rid of raising exceptions and
+	  hiding further exceptions.  [ruby-dev:39584]
+
+	* vm_eval.c (rb_funcall_no_recursive): ditto.
+
+Fri Oct 30 13:36:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_blocking_region): reverted r25566, and added
+	  description that no exception is allowed inside `func', instead.
+	  see [ruby-dev:39582]
+
+Fri Oct 30 13:13:16 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32.c (recvmsg, sendmsg, link): shouldn't raise ruby's exceptions
+	  in the functions expected as system API.  see [ruby-dev:39579] and
+	  [ruby-dev:39582]
+
+Fri Oct 30 12:59:20 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (rb_thread_blocking_region): standard C doesn't accept
+	  preprocessing directive within macro expansion.
+
+Fri Oct 30 10:55:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_blocking_region): must ensure to unlock GVL.
+	  [ruby-dev:39579]
+
+Fri Oct 30 04:47:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RSTRING_END): trivial optimization.
+
+	* string.c (rb_str_sub_bang): trivial optimization.
+
+Fri Oct 30 02:11:36 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_count): call #size using rb_funcall_no_recursive()
+	  to prevent infinite recursive calls.  [ruby-core:24794]
+
+	* vm_eval.c (rb_funcall_no_recursive): utility function to check
+	  direct recursive call.
+
+	* vm_eval.c (rb_check_funcall): move from eval.c. [refactoring]
+
+Thu Oct 29 18:37:02 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (Init_Encoding): revert previous commit.
+
+	* ruby.c (process_options): ditto.
+
+Thu Oct 29 15:35:39 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (Init_Encoding): set locale and filesystem encindex.
+
+	* ruby.c (process_options): move setting func of filesystem
+	  encoding to Init_Encoding.
+
+Thu Oct 29 15:43:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_name_err_mesg_new): added prototype.
+
+Thu Oct 29 13:53:18 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_to_ary): do not use #respond_to? to detect
+	  to_ary.  Just call.  [ruby-core:23738]
+
+	* eval.c (rb_check_funcall): new function with method existence
+	  check.  returns Qundef when the method does not exist.
+
+	* enumerator.c (enumerator_rewind): just call method, using
+	  rb_check_funcall().  [ruby-core:23738]
+
+	* error.c (exc_equal): ditto.
+
+	* object.c (convert_type): ditto.
+
+	* error.c (rb_name_err_mesg_new): export function.
+
+	* eval.c (make_exception): ditto.
+
+	* io.c (pop_last_hash): return early when the last argument is nil.
+
+	* io.c (rb_io_puts): treat T_STRING specially for small
+	  optimization.
+
+	* vm_eval.c (raise_method_missing): skip method call if possible
+	  using rb_method_basic_definition_p().
+
+	* vm_eval.c (method_missing): ditto.
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_debug): test
+	  suites changed to ignore exceptions caused by just-call policy.
+
+Thu Oct 29 04:41:44 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (process_options): call rb_filesystem_encoding().
+
+Thu Oct 29 04:40:36 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (Init_IO): rb_default_rs should be US-ASCII.
+
+Thu Oct 29 01:22:01 2009  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* ext/socket/extconf.rb : Compilation failure on AIX.
+	  ss_len (a member of struct sockaddr_storage) has preceding __,
+	  but ss_family does not have it from AIX 5.2.
+
+Wed Oct 28 16:32:49 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (get_filesystem_encoding): removed.
+
+	* encoding.c (rb_locale_encindex): added.
+
+	* encoding.c (rb_filesystem_encindex): added.
+
+	* encoding.c (rb_filesystem_encindex): add an alias 'filesystem'.
+	  [ruby-dev:39574]
+
+	* encoding.c (enc_find): add rdoc about special aliases.
+
+	* gem_prelude.rb (Gem.set_home): use Encoding.find('filesystem').
+
+	* gem_prelude.rb (Gem.set_paths): ditto.
+
+Wed Oct 28 15:02:31 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gem_prelude.rb (Gem.set_home):
+	  force_encoding(Encoding.filesystem_encoding)
+	  [ruby-core:25959]
+
+	* gem_prelude.rb (Gem.set_paths): ditto.
+
+Wed Oct 28 14:24:45 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (get_filesystem_encoding):
+	  add Encoding.filesystem_encoding [ruby-dev:39546]
+	  also see [ruby-core:25959]
+
+Wed Oct 28 14:51:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_each_group): fixed typo.
+
+Wed Oct 28 13:02:10 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::FTP#initialize): sets @binary to true.
+
+	* lib/net/ftp.rb (Net::FTP#binary=): sends a TYPE command only when
+	  logged in. [ruby-dev:39548]
+
+	* lib/net/ftp.rb (Net::FTP#send_type_command): new private method
+	  which sends an appropriate TYPE command according to the value of
+	  @binary.
+
+	* lib/net/ftp.rb (Net::FTP#login): calls send_type_command instead
+	  of binary=.
+
+Wed Oct 28 12:26:51 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_setup_method): should push call frame before
+	  raising exception, to put the Ruby-defined method name in the
+	  error message.  [ruby-core:26333]
+
+	* vm_insnhelper.c (VM_CALLEE_SETUP_ARG): macro modified.
+
+	* vm_insnhelper.c (vm_yield_setup_args): modified for new
+	  VM_CALLEE_SETUP_ARG macro.
+
+Tue Oct 27 22:46:44 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::FTP#initialize): @sock = nil.
+
+	* lib/net/ftp.rb (Net::FTP#binary=): send command only when
+	  socket is open. [ruby-dev:39548]
+
+Tue Oct 27 22:22:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (Init_stringio): added read_nonblock and
+	  write_nonblock aliases.  [ruby-dev:39551]
+
+	* ext/stringio/stringio.c (strio_data_type): typed.
+
+Tue Oct 27 21:20:35 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb: add TkVariable#+@ and -@ method.
+
+Tue Oct 27 16:36:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (chopped_length): get rid of unexpected exception.
+	  see [ruby-core:26336].
+
+Tue Oct 27 15:53:10 2009  Tanaka Akira  <akr@fsij.org>
+
+	* gc.h (SET_MACHINE_STACK_END): use __i386.
+	  Some compiler may not define __i386__.
+	  Solaris 64-bit Developer's Guide:
+	  http://docs.sun.com/app/docs/doc/816-5138/dev-env?l=en&a=view
+
+Tue Oct 27 15:44:48 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (getbinaryfile, list): call to_s to convert
+	  a Pathname instance into a string.  [ruby-core:26237]
+
+Tue Oct 27 12:30:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.h (SET_MACHINE_STACK_END): use __i386__ instead of __i386,
+	  and explicit size qualifiers.
+
+Tue Oct 27 09:40:55 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.h (SET_MACHINE_STACK_END): add x86_64 version by nobu.
+
+Tue Oct 27 09:27:59 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_exec.c (DECL_SC_REG): use __clang__.
+
+Tue Oct 27 08:56:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: reverted r25443 because build_os is used in mkmf.rb
+	  on some platforms.  [ruby-core:26332]
+
+Tue Oct 27 08:01:57 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* parse.y (ripper_filename): add Ripper#filename. [ruby-dev:37856]
+
+	* test/ripper/test_filter.rb: add more tests.
+
+Tue Oct 27 07:53:25 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (garbage_collect_with_gvl): do not garbage_collect when
+	  dont_gc flag turned on.  [ruby-core:26327]
+
+Tue Oct 27 07:38:39 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c,stubs.c: remove errors or warnings when compiled
+	  with old ruby 1.8.x.
+
+	* ext/tk/tkutil/tkutil.c: ditto.
+
+Tue Oct 27 05:56:39 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm.c (invoke_block_from_c): return Qnil when its iseq is
+	  SPECIAL CONST. [ruby-core:26335]
+
+Tue Oct 27 05:11:49 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_exec.c (DECL_SC_REG): check defined(__asm__) before use it.
+	  LLVM/clang defines __GNUC__ but doesn't have __asm__.
+
+Tue Oct 27 03:45:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): added RUBY_SITEARCH to use
+	  different name from RUBY_ARCH.  [ruby-core:26324]
+
+Mon Oct 26 20:04:13 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c (rb_ary_cmp): Array#<=> returns nil when comparison fails
+	  [ruby-core:26316]
+
+Mon Oct 26 18:37:57 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (default_proc_arity_check): new support function.
+
+	* hash.c (rb_hash_initialize): should do arity check as #default_proc=.
+	  [ruby-core:26281]
+
+Mon Oct 26 13:24:17 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/remote-tk.rb: typo fixed.
+
+Mon Oct 26 12:34:08 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ruby.c (usage): -T and -W descriptions updated.  see
+	  [ruby-dev:39539]
+
+	* man/ruby.1: ditto.
+
+Mon Oct 26 12:06:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_fwrite): adjust stdio file position after direct write on
+	  BSDish platforms.   [ruby-core:26300]
+
+Sun Oct 25 15:44:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/handle.c (dlhandle_sym): fixed an invalid local variable
+	  declaration.
+
+Sun Oct 25 13:33:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_cntl): F_DUPFD is platform dependent.
+
+Sun Oct 25 10:19:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/handle.c (rb_dlhandle_close): fixed an invalid local
+	  variable declaration.
+
+Sun Oct 25 05:44:34 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (**): Optimization (up to 45% faster)
+
+Sat Oct 24 14:28:40 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_set_default_proc): checks arity of default_proc
+	  of a Hash.  [ruby-core:26087]
+
+Sat Oct 24 13:38:45 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_cmp): defines Object#<=>.  [ruby-core:24063]
+
+Sat Oct 24 09:51:28 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_flat_map): new method that concatenates the values
+	  from given block.  also provides alias #collect_concat.
+
+Sat Oct 24 00:36:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_cntl): update max file descriptor by the result of
+	  fcntl(F_DUPFD).
+
+Fri Oct 23 16:31:14 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c (rb_class_new): move class check to rb_check_inheritable().
+
+	* class.c (rb_check_inheritable): should not allow subclass of
+	  class Class.  [ruby-core:26225]
+
+Fri Oct 23 14:25:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (target, target_alias): replace with real cpu.
+
+	* mkconfig.rb: build* are not needed in rbconfig.rb.
+
+	* configure.in (warnflags): use -Wextra only when
+	  -Wno-missing-field-initializers is available.
+
+	* configure.in (optflags): removed -fomit-frame-pointer by default.
+
+Fri Oct 23 09:12:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pathname.rb (Pathname::SAME_PATHS): FNM_SYSCASE is always
+	  non-nil.
+
+Fri Oct 23 07:32:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, Makefile.in, win32/Makefile.sub (XRUBY): runnable
+	  ruby without current libraries.
+
+	* common.mk (rdoc): use XRUBY.
+
+Fri Oct 23 07:28:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (warnflags): use -Wextra instead of -Wall.
+
+	* gc.c (mark_current_machine_context): get rid of warning.
+
+Thu Oct 22 21:10:39 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* class.c (rb_obj_basic_to_s_p): typo.  Please become familiar with
+	  the ANSI style.
+
+Thu Oct 22 20:20:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_in_out_err): test_stdout and
+	  test_stderr should be an array.
+
+	* test/ruby/test_rubyoptions.rb (test_notfound): test_stdin of
+	  assert_in_out_err should be a string.
+
+Thu Oct 22 17:49:05 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/fileutils.rb (FileUtils#fu_get_uid, fu_get_gid): Do not
+	  convert an integer back and forth.
+
+Thu Oct 22 17:29:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (arg_prepend): removed.  a patch from Mikhail T. in
+	  [ruby-core:26217].
+
+Thu Oct 22 04:54:41 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_inspect): print instance variables only when
+	  Object#to_s is not overridden.  [ruby-core:24425]
+
+	* class.c (rb_obj_basic_to_s_p): new function.
+
+Wed Oct 21 19:32:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_inspect): fixed rdoc about the case that to_s
+	  is called.  [ruby-core:24425]
+
+Wed Oct 21 08:17:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/logger/test_logger.rb (TestLogDevice#test_write): check
+	  also error message.
+
+Wed Oct 21 03:54:41 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb:
+	  Creator functions now strict with their arguments
+	  Support for empty matrices (see new method Matrix.empty)
+	  Matrix#trace raises an ErrDimensionMismatch if the matrix is not square
+	  Enumerators are returned when no block given
+	  Consistent results when accessing elements with out of bounds indices
+	  Details in [ruby-core:23598].
+
+Wed Oct 21 00:27:15 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/webrick/httpauth/digestauth.rb: typo in exception message fixed.
+
+Wed Oct 21 00:17:28 2009  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: imported upstream version (logger/1.2.7) see #2238.
+	  * do not raise an exception even if log writing failed.
+	  * do not raise ShiftingError if an aged file already exists.
+	    (no ShiftingError will be raised from 1.2.7, just warn() instead)
+	* test/logger/test_logger.rb: ditto.
+
+Tue Oct 20 22:29:06 2009  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/matrix.rb:  Bug fix. See detail [ruby-core:23598].
+
+Tue Oct 20 17:57:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_symbol, r_symreal): fixed the order of symbol and
+	  its encoding modifier, in order to make the dump readable from
+	  1.8.  [ruby-dev:39515]
+
+Tue Oct 20 16:41:18 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (finite, scalb): inline'ed non-standard
+	  identifier macros.  [ruby-core:26166]
+
+Tue Oct 20 15:38:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ripper_intern): enable literal optimization.
+
+	* parse.y (method_call): dispatch symbols.  a patch from Andy Keep in
+	  [ruby-core:26169].  [ruby-core:26165]
+
+	* parse.y (mlhs_basic): fixed handling splat in middle of mlhs.  a
+	  patch from Andy Keep in [ruby-core:26163]
+
+	* parse.y (parser_here_document): dispatch delayed heredoc
+	  contents.  based on a patch from Andy Keep in [ruby-core:24855].
+
+Mon Oct 19 15:17:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .gdbinit (rb_method_entry): search method entry by class and id.
+
+Mon Oct 19 15:03:31 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (enc/unicode/name2ctype.h): no need to create
+	  directory if it already exists.
+
+Mon Oct 19 11:34:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .gdbinit (rb_ps): dump all threads and their callstacks.  based
+	  on [ruby-core:26155] by Joshua ben Jore <twists AT gmail.com>.
+
+Mon Oct 19 10:59:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (prepare_iseq_build, rb_iseq_build_for_ruby2cext):
+	  untrust mark array.  [ruby-core:26137]
+
+Mon Oct 19 05:49:53 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/rexml/element.rb (text=): false should be converted to string.
+	  A patch by Teruo Oshida [ruby-dev:38351]
+
+Sun Oct 18 22:33:25 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: do not require lib/delta.rb.
+
+	* lib/date/delta.rb: follows the above change.
+
+Sun Oct 18 19:14:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (is_special_global_name): add boundary check.
+
+Sun Oct 18 18:31:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): should not dup tmp string.  a
+	  patch from neomjp neomjp in [ruby-core:24251].
+
+Sun Oct 18 09:49:14 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/test_prime.rb
+	  (TestPrime#test_eratosthenes_works_fine_after_timeout):
+	  test for [ruby-dev:39465].
+
+	* lib/prime.rb (Prime::EratosthenesSieve):
+	  fixed [ruby-dev:39465].
+	  suppressed memory reallocation.
+	  constantified some magic numbers.
+
+Sat Oct 17 22:11:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (id2encidx): register encoding name.
+	  (r_object0): register object before encoding name.
+	  [ruby-core:24882]
+
+Sat Oct 17 17:56:58 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/rake/test_fileutils.rb (Rake::TestFileUtils#test_sh): uses
+	  FileUtils::RUBY instead of fixed "ruby" so that the ruby command
+	  works fine in Ruby's "make test-all".
+	  (test_sh_with_a_single_string_argument): ditto.
+	  (test_sh_with_multiple_arguments): ditto.
+	  (test_sh_failure): ditto
+	  (test_sh_special_handling): ditto.
+
+Sat Oct 17 17:30:06 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (big_split): fix off-by-one error.  [ruby-dev:39501]
+
+Sat Oct 17 16:34:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (parser_yylex): fix token even after trailing under score.
+
+Sat Oct 17 11:27:44 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_method.c (basic_obj_respond_to): call #respond_to_missing?
+	  always with two arguments.  [ruby-core:26090]
+
+Sat Oct 17 08:51:44 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#respond_to_missing): warn only when
+	  specified method is a private.  [ruby-dev:39498]
+
+Fri Oct 17 00:05:53 2009  wanabe  <s.wanabe@gmail.com>
+
+	* st.c (unpack_entries): save table->bins and never change the table
+	  during unpacking. Because st_insert() may cause GC and refer the
+	  table, i.e. st_foreach().  [Bug #2196]
+
+Fri Oct 16 22:20:25 2009  Tanaka Akira  <akr@fsij.org>
+
+	* prelude.rb (require_relative): defined as a module function of
+	  Kernel.
+
+Fri Oct 16 20:18:28 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#method_missing): remove backtrace
+	  lines _until_ `method_missing'.
+
+Fri Oct 16 20:09:55 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#freeze): #freeze should freeze self
+	  and the target at once.   [ruby-core:26118]
+
+Fri Oct 16 19:39:28 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#respond_to_missing): warn if optional
+	  include_private argument is not false.  Delegator does (and
+	  should) not forward private methods.  [ruby-core:26080]
+
+	* lib/delegate.rb (Delegator#respond_to_missing): instead of
+	  redefining #respond_to?, use #respond_to_missing?.  [ruby-core:26081]
+
+Fri Oct 16 18:42:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/test_gc.rb: added tests based on [ruby-dev:39484]
+	  from wanabe <s.wanabe AT gmail.com>.
+
+Fri Oct 16 16:09:01 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn): `\'' is also quote character.
+
+Fri Oct 16 13:40:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.h (rb_gc_debug_body): constified.
+
+Fri Oct 16 13:20:39 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpMidRound): remove warnings.
+	  patch from Charlie Savage. [ruby-core:22869]
+
+	* ext/digest/bubblebabble/bubblebabble.c (bubblebabble_str_new): ditto.
+
+	* ext/digest/digest.c (hexencode_str_new): ditto.
+
+	* ext/iconv/iconv.c (iconv_convert): ditto.
+
+	* ext/socket/socket.c (inspect_sockaddr): ditto.
+
+	* ext/socket/raddrinfo.c (sockaddr_obj): ditto.
+
+	* ext/syck/emitter.c (syck_emitter_write): ditto.
+
+	* ext/syck/emitter.c (syck_emitter_flush): ditto.
+
+	* ext/syck/emitter.c (syck_emit_tag): ditto.
+
+Fri Oct 16 12:03:31 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/csv.rb (CSV#raw_encoding): returns ASCII-8BIT when the io
+	  doesn't have encoding.
+
+Fri Oct 16 03:15:52 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/csv.rb (CSV#read_to_char): set encoding and verify data
+	  which read from io before encode it to @encoding.
+
+	* lib/csv.rb (CSV#raw_encoding): add to get @io's encoding.
+
+	* lib/csv.rb (CSV#read_io): add to read string and set @io's
+	  encoding.
+
+Thu Oct 15 18:26:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_intern3): check symbol table overflow before generate
+	  next id.  [ruby-core:26092]
+
+Thu Oct 15 15:14:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_encoding_set): get rid of parsing non-ascii string, and
+	  refine messages for invalid name encoding.
+
+	* io.c (io_reopen): unread current buffer before telling the
+	  position, for the case of reopening same file.  [ruby-dev:39479]
+
+Thu Oct 15 14:20:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_create): cannot retry with given block.
+	  [ruby-dev:39487]
+
+Thu Oct 15 09:25:07 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/init.c (rsock_init_sock): mswin doesn't have S_IFSOCK
+	  flag in st_mode of struct stat. so, use rb_w32_issocket() function
+	  instead of S_ISSOCK macro.
+
+Thu Oct 15 00:47:42 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb,
+	  enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  use UTS#18 for POSIX character class.
+	  http://rubyspec.org/issues/show/161
+
+Thu Oct 15 00:26:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (rsock_init_sock): validate file descriptor.
+
+Wed Oct 14 13:24:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): script name should not be shown in an
+	  error message before loaded.
+
+Wed Oct 14 09:06:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_run_node): need to call ruby_cleanup() always even
+	  if any error occurred so far.
+
+Wed Oct 14 08:08:12 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c (mnew): Method#new checks for respond_to_missing? with
+	  private set to true
+	  [ruby-core:26069]
+
+Tue Oct 13 21:05:01 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb,
+	  enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  Add DerivedCoreProperties, PropList (Binary Property),
+	  PropertyAlias and PropertyValueAlias.
+	  Now users of tool/enc-unicode.rb should specify
+	  the directory of UCD files.
+
+Tue Oct 13 18:54:25 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/variable.rb: bug fix. additional trace definition changes the
+	  option of first trace definition.
+
+Tue Oct 13 18:23:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (token_info_push, token_info_pop): reduced ifdefs.
+
+	* parse.y (parser_magic_comment): fixed normalization.
+
+Tue Oct 13 09:04:14 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* thread.c: Revert changes to Thread#raise made in r25278
+	  [ruby-core:25367]
+
+	* eval_intern.h: ditto
+
+Mon Oct 12 23:27:57 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (login): use "anonymous@" as a default password.
+	  [ruby-dev:39451]
+
+Mon Oct 12 22:48:25 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (retrlines): added a new block parameter.
+
+	* lib/net/ftp.rb (gettextfile): preserve missing end-of-line at end
+	  of files.  [ruby-core:24590]
+
+Mon Oct 12 19:48:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_run_node): if an exception occurred in ruby_option,
+	  the result is not executable.
+
+Mon Oct 12 05:51:11 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (fetch_token): warn invalid back reference
+	  and subexp call. (\k and \g).
+
+Mon Oct 12 03:47:42 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* marshal.c (has_encoding): added for check the regexp
+	  is dumped by 1.8 or 1.9.
+
+	* marshal.c (r_object0): use has_encoding.
+
+Sun Oct 11 15:54:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/gdbm/gdbm.c (fgdbm_select): fixed rdoc.  a patch from Justin
+	  Collins in [ruby-core:26050].
+
+Sun Oct 11 10:27:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb/context.rb (IRB::Context#irb_name): removed duplicated
+	  attr_reader.  [ruby-core:26047]
+
+	* lib/irb/ruby-lex.rb (RubyLex#lex_int2): removed duplicated
+	  character class range.
+
+Sun Oct 11 10:04:35 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (fetch_token_in_cc): warn when \p is not
+	  followed by property name.
+
+	* regparse.c (fetch_token): ditto.
+
+Sun Oct 11 09:44:46 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regerror.c (onig_vsnprintf_with_pattern): added.
+
+	* regparse.c (onig_syntax_warn): use above.
+
+Sun Oct 11 09:04:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_GC_GUARD_PTR): workaround for gcc
+	  optimization.
+
+	* include/ruby/ruby.h (ruby_exec_node): declared.
+
+Sun Oct 11 03:10:50 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (onig_syntax_warn): added.
+
+	* regparse.c (CC_ESC_WARN, CLOSE_BRACKET_WITHOUT_ESC_WARN,
+	  CC_DUP_WARN, UNKNOWN_ESC_WARN): use onig_syntax_warn.
+
+Sun Oct 11 00:14:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (marshal_dump, marshal_load): prevent from GC.
+
+Sat Oct 10 23:57:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (path_check_0): prevent from GC.
+
+Sat Oct 10 23:51:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_transcoding, str_transcoding_resize): fixed
+	  types.
+
+Sat Oct 10 20:35:27 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* math.c (math_atanh): reverted r25279.
+
+Sat Oct 10 19:03:29 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c: use rb_thread_blocking_region to avoid
+	  rb_read_check.  This makes other threads runnable in getstr and
+	  wgetstr.
+	  (getch_func): extracted from curses_getch.
+	  (curses_getch): use rb_thread_blocking_region with getch_func.
+	  (getstr_func): extracted from curses_getstr.
+	  (curses_getstr): use rb_thread_blocking_region with getstr_func.
+	  (wgetch_func): extracted from window_getch.
+	  (window_getch): use rb_thread_blocking_region with wgetch_func.
+	  (wgetstr_func): extracted from window_getstr.
+	  (window_getstr): use rb_thread_blocking_region with wgetstr_func.
+
+	* include/ruby/io.h (rb_read_check): deprecated because it access
+	  internal of stdio.
+
+Sat Oct 10 18:59:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (cflags, cxxflags): remove duplicating options.
+
+Sat Oct 10 18:19:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, Makefile.in (LIBRUBY_SO), common.mk (ruby.imp),
+	  win32/mkexports.rb (each_export): exclude _threadptr_ functions.
+
+Sat Oct 10 17:55:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_threadptr_errinfo): renamed.
+
+Sat Oct 10 17:03:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_exec_node): removed unused argument.
+
+Sat Oct 10 14:55:55 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* math.c (math_atanh): Fix bug for Math.atanh(+/-1). It now returns
+	  +-Infinity. [ruby-core:26028]
+
+Sat Oct 10 14:09:40 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* thread.c (rb_threadptr_execute_interrupts_rec, rb_threadptr_raise):
+	  Thread#raise with no argument will now re-raise the current exception
+	  if there is one [ruby-core:25367]
+
+	* eval.c (get_errinfo, rb_rubylevel_thread_errinfo): Getter for
+	  current exception for a given thread
+
+Fri Oct  9 23:10:04 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_method.c (rb_method_boundp): should exclude NOEX_RESPONDS.
+	  based on the patch from Nikolai Lugovoi.  [ruby-core:25949]
+
+Fri Oct  9 21:14:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Requester::ConnectedUDP): disable
+	  reverse lookup.
+	  (Resolv::DNS::Requester::UnconnectedUDP): ditto.
+
+Fri Oct  9 10:12:13 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/irb/context.rb (IRB::Context#initialize):
+	  remove warnings when $VERBOSE is set as true in .irbrc.
+
+Fri Oct  9 02:58:18 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb: optimized.
+
+	* enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  U+100000-U+10FFFD is assigned, not Cn.
+
+Fri Oct  9 02:12:02 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* ext/curses/curses.c: Many functions of module Curses could cause a
+	  crash if the ncurses library was not properly initialized.
+	  Fix pointed out by Alexander Beisig [ruby-core:22592]
+	  Functions fixed: attroff, attron, attrset, bkgd, bkgdset,
+	  can_change_color, close_screen, closed, color_content, curs_set,
+	  def_prog_mode, delch, deleteln, getmouse, getstr, has_colors,
+	  init_color, init_pair, insertln, keyname, mouseinterval, mousemask,
+	  pair_content, pair_number, reset_prog_mode, resizeterm, scrl,
+	  setscrreg, standend, standout, start_color, timeout, ungetmouse,
+	  use_default_colors
+
+Fri Oct  9 01:07:34 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (ADD_TRACE): fire coverage event in ensure clause.
+	  [ruby-dev:39303]
+
+	* iseq.h, iseq.c: ditto.
+
+Fri Oct  9 00:33:29 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/net/telnet.rb (cmd): Pass FailEOF options: patch by Brian
+	  Candler [ruby-core:22723]
+
+Fri Oct  9 00:01:17 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.udp_server_recv): extracted from
+	  Socket.udp_server_loop_on.
+
+Thu Oct  8 05:45:14 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb: parse range notation of UnicodeData.txt.
+
+	* enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  follow above change. [ruby-dev:39444]
+
+Thu Oct  8 02:46:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_inspect): copy by chunks.
+
+Thu Oct  8 01:23:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_s_disasm): accept proc objects.  [ruby-core:18762]
+
+Wed Oct  7 16:42:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (mark_dump_arg, mark_load_arg): ignore already cleaned
+	  data.  [ruby-core:25969]
+
+	* marshal.c (clear_dump_arg, clear_load_arg): clear freed fields.
+
+Wed Oct  7 16:06:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_long2int): evaluates the argument only
+	  once.
+
+	* struct.c (rb_struct_alloc): check array length overflow.
+
+Wed Oct  7 09:23:49 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): don't assign -1 to unsigned int.
+
+Wed Oct  7 00:27:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS.bind_random_port): bind to "::" for IPv6.
+	  (Resolv::DNS::ConnectedUDP#initialize): specify is_ipv6 argument of
+	  bind_random_port.
+	  [ruby-core:25970]
+
+Tue Oct  6 23:32:38 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_upto): RDoc updated.  a patch from Nobuhiro
+	  IMAI <nov at yo.rim.or.jp> in [ruby-dev:39440]. [ruby-dev:39439]
+
+Tue Oct  6 21:30:58 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_reopen): avoid close if possible.
+
+Tue Oct  6 18:56:09 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_select): Struct#select should return
+	  enumerator when no block given.
+
+Tue Oct  6 06:26:00 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* * ext/tk/lib/tk/canvas.rb: *** POTENTIALLY INCOMPATIBLE ***
+	  'tags' option of a TkcItem object should give a list of TkcTag objs.
+
+	* ext/tk/lib/tkextlib/vu/dial.rb: fix logical bug.
+
+	* ext/tk/lib/tk/canvas.rb, ext/tk/lib/tkextlib/blt/component.rb:
+	  lack of support for methodcall_optkeys.
+
+Mon Oct  5 17:19:33 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator::public_api): take snapshot of
+	  public method at the beginning time.
+
+	* lib/delegate.rb (SimpleDelegator#initialize): use
+	  Delegator.public_api since public_method might be added after
+	  initialization.  [ruby-dev:39383]
+
+	* lib/delegate.rb (DelegateClass): ditto.
+
+Mon Oct  5 12:22:12 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_{times, shuffle_bang, sample}): reducing macro
+	  calls inside of the loop by keeping pointers in local
+	  variables. a patch from Masahiro Kanai (CanI) in [ruby-dev:39406].
+	  It was found and fixed at Security and Programming camp 2009.
+
+	* string.c (rb_str_{times, split_m}): ditto.
+
+	* struct.c (rb_struct_{getmember, set, aref_id, aset_id}, {make,
+	  inspect}_struct, recursive_{equal, hash, eql}): ditto.
+
+Mon Oct  5 00:09:57 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_method.c (basic_obj_respond_to): should not call
+	  #respond_to_missing? for not implemented methods.
+	  [ruby-core:25909]
+
+	* vm_method.c (rb_method_boundp): returns exceptional value 2 for
+	  not-implemented methods when called from #respond_to? (specifies
+	  by new constant NOEX_RESPONDS).
+
+	* method.h (enum): new constant NOEX_RESPONDS added.
+
+Sun Oct  4 22:16:29 2009  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/cookie.rb: add default value to @@accept_charset
+	  if have not defined. [ruby-dev:38987]
+
+	* lib/cgi/util.rb: ditto.
+
+Sun Oct  4 19:30:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (struct {dump,load}_arg): manage with dfree, instead
+	  of using local variable which may be moved by context switch.
+	  [ruby-dev:39425]
+
+Sun Oct  4 15:00:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (NATINT_LEN, pack_pack): suppressed warnings.
+
+Sun Oct  4 14:01:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems.rb (Gem::binary_mode): binary mode uses binary
+	  encoding.
+
+Sun Oct  4 08:27:10 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (slicebefore_ii): use id_eqq.
+
+Sun Oct  4 06:40:09 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup (ruby.mmp): added macro RUBY_EXPORT to match the
+	  change in dln.c
+
+Sun Oct  4 05:34:34 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb: add TkVariable#to_hash,to_proc,to_int,
+	  to_str,to_ary
+
+Sun Oct  4 00:59:52 2009  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb: fix command-line option of
+	  non-interactive terminal. [ruby-core:23016]
+
+Sun Oct  4 00:40:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake/contrib: added.  [ruby-core:25918]
+
+Sat Oct  3 22:14:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (bv_decls, bvar): fix for block variables.
+	  [ruby-dev:39423]
+
+Sat Oct  3 21:19:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_add_method_def): no redefinition warning on
+	  undef.
+
+Sat Oct  3 18:51:11 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_f_integer): now Integer() takes optional base
+	  argument.  base will be ignored for non string values.
+	  suggested by Sam Carr at RubyFoo Lounge at London.
+
+	* test/ruby/test_integer.rb (TestInteger#test_Integer): test
+	  updated.
+
+Sat Oct  3 04:34:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assignable_gen): parser_yyerror takes two arguments.
+
+Sat Oct  3 04:07:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake: updated to rake code to rake-0.8.7 source code base.
+
+	* lib/rake/loaders/makefile.rb (Rake::MakefileLoader#process_line):
+	  respace dependencies too.
+
+Sat Oct  3 02:59:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assignable_gen): get rid of macro collision.
+
+Sat Oct  3 02:49:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_make_shared): should count frozen array itself.
+
+Sat Oct  3 01:01:20 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/unicode/name2ctype.h: Updated to Unicode 5.2.0.
+
+Sat Oct  3 01:01:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: workaround for regexp metacharacters of expr in
+	  prefix path.
+
+Sat Oct  3 00:47:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ripper/dummyparser.rb (DummyParser): improvement by Magnus
+	  Holm in [ruby-core:25884].
+	  * remove scanner events which simply returned the first argument.
+	  * all parser events are now automatically generated.
+	  * simplify blocks.
+
+Sat Oct  3 00:43:52 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* method.h (rb_method_type_t): remove a comma at end of
+	  enumerator list.
+
+Sat Oct  3 00:31:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (discrete_object_p): needs the argument type to get rid
+	  of truncation on platforms where VALUE is larger than int.
+
+Fri Oct  2 22:30:15 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/unicode/name2ctype.h.blt, enc/unicode/name2ctype.kwd,
+	  enc/unicode/name2ctype.src: Updated to Unicode 5.2.0.
+	  NOTE: when you update these data, download UnicodeData.txt
+	  and Scripts.txt from http://www.unicode.org/Public/UNIDATA/
+	  and run
+	  ruby1.9 tool/enc-unicode.rb UnicodeData.txt Scripts.txt \
+	  > enc/unicode/name2ctype.kwd
+
+	* enc/unicode/Scripts.txt: removed.
+
+	* enc/unicode/UnicodeData.txt: removed.
+
+Fri Oct  2 20:49:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_slice_before): take a pattern if no block given.
+
+Fri Oct  2 20:37:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ripper_yylval_id, ripper_get_{id,value}): wrap ID by
+	  NODE to track local variable assignment.
+
+	* parse.y (lvar_defined_gen, assignable_gen): enable local
+	  variable check.  [ruby-core:24923]
+
+	* parse.y (validate): use value only.
+
+	* test/ripper/test_parser_events.rb (test_local_variables): tests
+	  based on a patch from Magnus Holm in [ruby-core:25885].
+
+Fri Oct  2 15:34:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (ruby_suppress_tracing): get rid of clobbering by
+	  longjmp.
+
+Fri Oct  2 09:20:35 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (rb_origenviron): remove unused old variable.
+	  [ruby-dev:39412]
+
+Thu Oct  1 14:16:39 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{setup.mak,Makefile.sub,win32.c},include/ruby/win32.h (RT_VER):
+	  split compiler specification and runtime library specification.
+
+	* win32/Makefile.sub (LD_SHARED*, config.status): no need to embed
+	  manifest if not exist.
+
+Thu Oct  1 13:23:14 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_getpid): simply call GetCurrentProcessId()
+	  instead of calling MSVCRT's getpid().
+
+Wed Sep 30 13:15:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_add_method_def): show the location where
+	  overwritten method was defined.  [ruby-dev:39400]
+
+Wed Sep 30 00:37:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_block_call): extracted.
+
+Wed Sep 30 00:00:25 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): escape as \x{XXXX} when the encoding is
+	  other than Unicode. [ruby-dev:39388]
+
+Wed Sep 30 00:00:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (THREAD_MODEL): modified message when no thread
+	  model is available.
+
+Tue Sep 29 23:17:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_scan_open_args): rb_utf8mac_encoding is undefined.
+
+Tue Sep 29 22:25:41 2009  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_struct_iv_get): deprecated because it is
+	  not used and access internal structure.
+
+Tue Sep 29 22:19:36 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb (assert_equal): use Time#subsec if nsec
+	  is not enough to distinguish arguments.
+
+Tue Sep 29 21:16:35 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_scan_open_args): add UTF8-MAC to no-conversion encoding.
+
+Tue Sep 29 21:21:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--enable-pthread): deprecated.
+
+Tue Sep 29 21:03:59 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/pp.rb (Kernel#pp): returns its arguments, like Kernel#p.
+	  [ruby-dev:34830]
+
+Tue Sep 29 17:02:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/lib/json/common.rb (recurse_proc): removed needless
+	  module_function, since visibility is already module_function.
+
+Tue Sep 29 13:48:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_method_definition_eq): non-null definition is
+	  not equal to null definition.
+
+	* vm_method.c (rb_add_method_def): nothing to do if old method had
+	  same definition.  [ruby-dev:39397]
+
+Tue Sep 29 06:50:32 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): dump as \uXXXX when the
+	  string is in Unicode. [ruby-dev:39388]
+
+Tue Sep 29 06:49:16 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_unicode_p): defined.
+	  Returns 1 when the encoding is Unicode series
+	  other than UTF-7 else 0.
+
+Tue Sep 29 04:14:08 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encoding): On Unix systems,
+	  filesystem encoding should be locale encoding.
+	  [ruby-dev:39393]
+
+Tue Sep 29 04:07:58 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* hash.c (rb_f_getenv): use rb_filesystem_str_new_cstr
+	  instead of rb_str_new2.
+	  ENV['PATH'].encoding should be Filesystem Encoding
+	  because its content is related to filesystem.
+	  see [ruby-dev:39393]
+
+	* hash.c (env_fetch): ditto.
+
+	* string.c (rb_filesystem_str_new): defined.
+
+	* string.c (rb_filesystem_str_new_cstr): ditto.
+
+	* include/ruby/intern.h (rb_filesystem_str_new): added.
+
+	* include/ruby/intern.h (rb_filesystem_str_new_cstr): ditto.
+
+Tue Sep 29 04:06:18 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/st.h: include inttypes.h and stdint.h.
+
+Tue Sep 29 00:07:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_f_getenv, env_fetch): env string may be overwritten.
+
+Mon Sep 28 23:30:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (load_lib, dln_find_exe_r): env string may be overwritten.
+
+	* dln.c (dln_{exit,loaderror,memerror,notimplement}): renamed as
+	  independent names.
+
+	* dln.c (aix_loaderror): needs format string.
+
+Mon Sep 28 19:36:20 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (LK_ERR): with overlapped I/O, LockFileEx() returns
+	  ERROR_IO_PENDING if the file is locked.
+
+Mon Sep 28 19:05:05 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* include/ruby/st.h: aligned prototype of st_hash_uint32 with function
+	  definition (fixing compiling problem on cygwin)
+
+Mon Sep 28 12:13:15 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* method.h (enum): new method type VM_METHOD_TYPE_MISSING.
+
+	* vm_eval.c (vm_call0): invoking VM_METHOD_TYPE_MISSING method
+	  objects.
+
+	* vm_insnhelper.c (vm_call_method): invoking method defined from
+	  VM_METHOD_TYPE_MISSING.
+
+	* proc.c (rb_method_entry_arity): ditto.
+
+	* vm_method.c (rb_method_entry_eq): two method object wraps
+	  method_missing with same symbol should be equal.
+	  [ruby-core:25755]
+
+	* proc.c (mnew): should always return method object.
+
+Mon Sep 28 11:38:07 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (parser_tokadd_string): the byte after ``\'' may be a part of
+	  multibyte character, so pushback it. [ruby-list:46416]
+
+Mon Sep 28 10:06:38 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* stringio/stringio.c (strio_read): set ASCII-8BIT encoding
+	  when length argument is given.
+
+Mon Sep 28 01:28:17 2009  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* Makefile.in (miniruby): suppress duplication warning on AIX.
+
+Mon Sep 28 01:13:25 2009  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* common.mk (ruby.imp): add text section [Bug #2064].
+
+	* common.mk (ruby.imp): do not export Init_*.
+
+Sun Sep 27 13:06:43 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb (PP:ObjectMixin#pretty_print): delegates has no inspect
+	  method.  [ruby-core:25804]
+
+Sun Sep 27 12:01:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_buf_cat2): optimize since all second arguments are
+	  constant literals.
+
+	* string.c (str_cat_char): unused now.
+
+	* string.c (rb_hash_{uint{32,},end}): removed.
+
+Sun Sep 27 11:58:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (optflags): add -fomit-frame-pointer by default.
+
+Sun Sep 27 11:28:15 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_minmax): use struct for memo.
+	  (enum_minmax_by): ditto.
+
+Sun Sep 27 10:21:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (warnflags): check all flags if each are available.
+
+Sun Sep 27 05:35:17 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/ext/generator/generator.c: Documentation patch by okkez.
+	  [Bug #2075]
+
+Sun Sep 27 04:20:55 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): result's encoding should be fixed.
+	  If default_internal is not nil, the encoding is default_internal.
+	  Else if default_external is not nil, the encoding is default_external.
+	  But the encoding is not ASCII-compatible, the encoding is replaced by
+	  US-ASCII.
+	  Characters in ASCII-incompatible encoding or non ASCII characters
+	  in other than the encoding will be \xXX escaped.
+	  [ruby-dev:39343]
+
+	* string.c (str_buf_cat2): defined.
+
+	* string.c (prefix_escape): removed.
+
+Sun Sep 27 05:37:45 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/missing-pips.c: Updated to work with the latest PIPS 1.6.
+
+	* symbian/setup: ditto.
+
+	* symbian/README.SYMBIAN: ditto.
+
+Sun Sep 27 02:00:46 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: use rename-macro instead of RUBY_ALIAS_FUNCTION_TYPE.
+	  Because build causes failure.
+
+Sat Sep 26 23:29:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c: moved murmur hash from string.c.  [ruby-dev:39376]
+
+Sun Sep 26 00:24:14 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup: Updated .mmp file generation due to blockinlining.c removal.
+
+Sat Sep 26 22:39:24 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* Makefile.in(test-rubyspec): explicitly executes run subcommand of
+	  mspec.
+
+	* spec/README: typo fix
+
+Sat Sep 26 17:53:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (COLLISION): improved collision log feature.
+
+	* string.c (hash): updated to MurmurHash 2.0 2009-09-19.
+
+	* string.c (rb_hash_start): fixed shift width on 128bit platform.
+
+	* include/ruby/intern.h (rb_hash_{start,uint32,uint,end}): fixed
+	  prototypes.
+
+Sat Sep 26 13:26:55 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/net/http.rb (transport_request): Handle timeout error by
+	  closing socket if exception raised. [ruby-core:20976]
+
+Sat Sep 26 12:08:17 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_eq): method defined from same
+	  block/proc should be equal.  [ruby-core:25755] [ruby-core:24791]
+
+Sat Sep 26 08:35:12 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (compile_string): rename to parse_string(), because
+	  this function only parse String to NODE.
+
+Fri Sep 25 16:01:45 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_access): new function to
+	  replace MSVCRT's access().
+	  [ruby-core:25761]
+
+	* file.c (eaccess): workaround for recent MSVCRT is no longer needed.
+
+Fri Sep 25 13:04:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew): fix for instance method of Module, BasicObject
+	  and subclass of a class which overrides respond_to_missing?.
+	  based on a patch from Nikolai Lugovoi <nlugovoi AT gmail.com> in
+	  [ruby-core:25748].
+
+Fri Sep 25 11:56:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_mod_method_defined): should return true or false.
+
+Thu Sep 24 13:32:53 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (mnew): generate method object that wraps method_missing,
+	  when #respond_to_missing? is defined.
+
+	* test/ruby/test_object.rb (test_respond_to_missing): add test
+	  suites for #respond_to_missing? changes.
+
+Thu Sep 24 09:41:42 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/mathn.rb (Bignum#**): Fixed bignum**fixnum that was broken when
+	  requiring lib/mathn
+	  [ruby-core:25740]
+
+Thu Sep 24 02:21:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (atan): atan(Infinity) is
+	  PI/2.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (atan): reduce loop with
+	  the double-angle formula.  based on a patch from Masahiro
+	  Kanai (CanI) in [ruby-dev:39367].
+
+Thu Sep 24 01:14:18 2009  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* dln.c (aix_loaderror): fixed typo. suppress warnings.
+
+Thu Sep 24 00:17:06 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_minmax): reduce comparison.
+	  (enum_minmax_by): ditto.
+
+Wed Sep 23 22:58:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/thread.rb (ConditionVariable#wait): add timeout argument.
+	  [ruby-talk:346154]
+
+Wed Sep 23 21:25:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (atan): refined.
+
+Wed Sep 23 17:08:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assign_in_cond): also should warn assignment to dvar in
+	  conditional.  [ruby-dev:39363]
+
+Wed Sep 23 13:14:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_upto): keep first width.  [ruby-dev:39361]
+
+Wed Sep 23 11:28:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/instruction.rb (make_header_prepare_stack): check stack
+	  overflow.  [ruby-core:25714]
+
+	* tool/instruction.rb (make_footer_stack_val): ditto.
+
+Wed Sep 23 05:03:36 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c (umethod_bind, rb_mod_define_method): Fix bug that
+	  disallowed methods from singleton classes to be used for
+	  UnboundMethod#bind, Kernel#define_singleton_method and
+	  Module#define_method, even when that singleton class was of the right
+	  kind_of. A patch by Shane O'Brien [ruby-core:25632]
+
+Tue Sep 22 22:56:48 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_method.c (basic_obj_respond_to): new function to fundamental
+	  behavior for #respond_to?
+
+	* vm_method.c (basic_obj_respond_to): calls #respond_to_missing
+	  method if overridden, to check responsiveness of methods
+	  implemented by #method_missing.
+
+Tue Sep 22 16:34:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_table_entry, st_get_key): use st_index_t.
+
+Tue Sep 22 16:28:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regenc.h (PosixBracketEntryType): constified.
+
+	* regenc.h (PosixBracketEntryInit): suppress warnings.
+
+	* regerror.c (onig_error_code_to_str, onig_snprintf_with_pattern):
+	  fixed type.
+
+	* regparse.c (st_str_end_key, str_end_cmp, str_end_hash):
+	  constified.
+
+	* tool/transcode-tblgen.rb (transcode_generated_code): fixed type.
+
+Tue Sep 22 10:29:06 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_chunk): new method Enumerable#chunk.
+	* enum.c (enum_slice_before): new method Enumerable#slice_before.
+	  [ruby-dev:38392] [ruby-dev:39240]
+
+Tue Sep 22 05:58:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c, cont.c, gc.c, insns.def, iseq.c, iseq.h, process.c,
+	  thread.c, vm.c, vm_core.h, vm_dump.c, vm_eval.c,
+	  vm_insnhelper.c, vm_method.c, template/insns_info.inc.tmpl,
+	  tool/instruction.rb: fixed types.
+
+Tue Sep 22 05:04:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/{ludcmp,math}.rb: depend on
+	  bigdecimal.
+
+	* ext/bigdecimal/lib/bigdecimal/*.rb: made module functions.
+
+Tue Sep 22 04:47:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValue): support conversion from
+	  Rational.  [ruby-core:25697]
+
+Tue Sep 22 04:43:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub (INSNS): depend on tools.
+
+Tue Sep 22 01:10:22 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* ossl_ocsp.c (ossl_ocspres_to_der): Bug fix in Response#to_def.
+	  Patch by Chris Chandler [ruby-core:18411]
+
+Tue Sep 22 01:10:02 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* ossl_config.c (ossl_config_add_value_m, ossl_config_set_section):
+	  Check if frozen (or untrusted for $SAFE >= 4) [ruby-core:18377]
+
+Mon Sep 21 17:12:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_binding): allow proc from method.  [ruby-core:25589]
+
+	* vm.c (collect_local_variables_in_env): block iseq can be NULL.
+
+Mon Sep 21 10:50:37 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (rb_time_succ): make Time#succ obsolete since time is not
+	  a discrete value.
+
+	* range.c (discrete_object_p): treat time objects specially to
+	  determine discrete values, since time objects have #succ yet are
+	  discrete (for now at least).
+
+Mon Sep 21 10:13:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_new, cont_capture, fiber_t_alloc): needs already
+	  running thread.  cf. [ruby-core:25681]
+
+Mon Sep 21 00:07:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (sin, cos, atan, exp, log):
+	  improved precision and performance.  based on a patch from Makoto
+	  Yamashita in [ruby-core:25600] and [ruby-core:25602].
+
+Sun Sep 20 11:11:34 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* struct.c (rb_struct_equal, rb_struct_eql): Handle comparison of
+	  recursive structures [ruby-core:24759]
+
+	* range.c (range_eq, range_eql):  ditto for ranges
+
+Sat Sep 19 17:46:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (ENABLE_VM_OBJSPACE): socklist needs st_table in
+	  rb_w32_sysinit(), before object space initialization.
+
+Sat Sep 19 17:32:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (GlobPathValue), file.c (rb_get_path_check): path names
+	  must be ASCII compatible.
+
+Sat Sep 19 00:02:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_type): forward declaration to suppress a
+	  warning.  a patch from Naohisa Goto at [ruby-dev:39350]
+
+Fri Sep 18 23:59:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (GlobPathValue): adjust return types.  based on a patch
+	  from Naohisa Goto at [ruby-dev:39350].
+
+Fri Sep 18 23:51:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): entry regexp object before its encoding
+	  name.  [ruby-core:25625]
+
+Fri Sep 18 16:29:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (eval.o): needs vm.h.
+
+	* eval.c (ruby_cleanup): destruct current VM before exit.
+
+	* gc.c (rb_objspace_free): free object space.
+
+	* vm.c (ruby_vm_destruct): destruct and free VM struct.
+
+Fri Sep 18 16:15:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each), parse.y (stmt, arg): arg_concat()
+	  on op_asgn was inversed.  [ruby-core:25629] [Bug #2050]
+
+Fri Sep 18 16:06:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (GlobPathValue): glob allows null bytes as separators.
+
+Fri Sep 18 10:11:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/re.rdoc: use rdoc mode.
+
+	* misc/rdoc-mode.el: added.
+
+Fri Sep 18 09:02:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc#parse_files): don't branch by
+	  RUBY_VERSION.
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc#parse_files): emacs local variables
+	  are delimited by a semicolon.  supported Vim style.
+
+Fri Sep 18 07:06:41 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/fileutils/test_fileutils.rb: add a test for [ruby-dev:39345]
+
+Fri Sep 18 06:47:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (push_glob): str should be a string always.
+
+	* dir.c (rb_push_glob, dir_globs): use #to_path to convert non-
+	  string values.  cf. [ruby-dev:39345]
+
+Fri Sep 18 06:36:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (rm_f, rm_rf): FileUtils.rm can take an array.
+	  [ruby-dev:39345]
+
+	* lib/mkmf.rb (create_header): open in binmode.
+
+Thu Sep 17 18:34:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (vm_xrealloc): free as like standard free if size is zero.
+
+Thu Sep 17 15:41:02 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h: use rb_node_newnode() directly.
+
+Thu Sep 17 15:01:32 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c: removed.
+
+	* README.EXT: ditto.
+
+	* README.EXT.ja: ditto.
+
+Thu Sep 17 13:50:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_s_debug_set): set level, not only boolean.
+
+Thu Sep 17 13:12:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parser/ruby.rb (RDoc::Parser::Ruby): parse also rdoc
+	  files.
+
+	* doc/re.rdoc: renamed from re.rb.
+
+Thu Sep 17 09:37:28 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* doc/re.rb: New document for Ruby's fork of Oniguruma.
+	  written by Run Paint Run Run [ruby-core:25420]
+
+	* re.c: import document in doc/re.rb.
+
+	* .document: add doc/re.rb.
+
+Thu Sep 17 06:03:40 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (Matrix#rank): Two bug fixes. One made
+	  Matrix[[0,0],[0,0],[1,0]].rank raise a NoMethodError while the other
+	  one had Matrix[[0,1],[0,0],[1,0]].rank raise a TypeError.
+
+Thu Sep 17 06:02:04 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Optimizations
+
+Thu Sep 17 00:36:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_mdump): error message refined.
+
+Wed Sep 16 19:27:43 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_thread.rb (TestThread#test_recursive): remove
+	  implementation dependent test.
+
+Wed Sep 16 17:42:52 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): install-capi needs docdir.  And,
+	  mandir, infodir and ridir are also depend on datadir like docdir.
+
+Wed Sep 16 17:20:49 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (Matrix#/): Fix obvious bug
+
+Wed Sep 16 16:59:34 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (DOCTARGETS): rdoc is default.
+
+Wed Sep 16 16:27:40 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub: typo.
+
+Wed Sep 16 16:20:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (CAPITARGET): enable iff doxygen is available.
+
+	* configure.in (INSTALLDOC): enable if rdoc or doxygen are enabled.
+
+	* common.mk (docs): target to make documents.
+
+	* Makefile.in, win32/Makefile.sub (install-{all,nodoc}): trigger
+	  by $(INSTALLDOC) from install.
+
+Wed Sep 16 13:39:10 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (determinant): Bug fix where determinant failed on
+	  some matrices [ruby-core:23597]
+
+Wed Sep 16 13:30:20 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb:  trivial optimizations
+
+Wed Sep 16 13:15:17 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* template/Doxyfile.tmpl: RbConfig::CONFIG['DOT'] is sometimes nil.
+
+Wed Sep 16 10:14:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja (rb_protect, rb_jump_tag): added.
+
+Wed Sep 16 10:12:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (main): makes also encs.
+
+Wed Sep 16 06:30:07 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* thread.c (rb_exec_recursive_outer, rb_exec_recursive): Added method
+	  to short-circuit to the outermost level in case of recursion
+
+	* test/ruby/test_thread.rb (test_recursive_outer): Test for above
+
+	* hash.c (rb_hash_hash): Return a sensible hash for in case of
+	  recursion [ruby-core:24648]
+
+	* range.c (rb_range_hash): ditto
+
+	* struct.c (rb_struct_hash): ditto
+
+	* array.c (rb_array_hash): ditto
+
+	* test/ruby/test_array.rb (test_hash2): test for above
+
+Wed Sep 16 06:17:33 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* vm_eval.c (rb_catch_obj, rb_catch, rb_f_catch): No longer use the
+	  obsolete function rb_iterate.
+
+Tue Sep 15 21:48:12 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (--enable-frame-address): removed.
+
+Tue Sep 15 15:21:01 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{configure.bat,Makefile.sub} (RDOCTARGET): the meaning of this
+	  macro was changed at r24923.
+
+Tue Sep 15 15:16:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (install-rdoc), configure.in (RDOCTARGET): removed
+	  circular dependency.  [ruby-dev:39339]
+
+Tue Sep 15 15:09:13 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/configure.bat: need a space before a tab to output the tab by
+	  echo.
+
+Tue Sep 15 14:24:52 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_split_m): use rb_isspace when the
+	  string may be ASCII-incompatible.
+
+	* string.c (rb_str_lstrip_bang): ditto.
+
+	* string.c (rb_str_rstrip_bang): ditto.
+
+Tue Sep 15 12:12:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_USE_BUILTIN_FRAME_ADDRESS): check after real
+	  target CPU is set.
+
+	* configure.in (RUBY_UNIVERSAL_ARCH): check real target CPU after
+	  AC_PROG_CC.
+
+Tue Sep 15 06:42:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (install-all): target to install all.
+
+	* tool/rbinstall.rb (parse_args): accept all install targets.
+
+	* tool/rbinstall.rb (parse_args): show help message and exit if
+	  rbconfig could not load.
+
+Tue Sep 15 04:25:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RDOCTARGET): use install-all.  [ruby-dev:39334]
+
+Tue Sep 15 03:00:35 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_lstrip_bang): use ascii_isspace().  [ruby-dev:39322]
+
+	* string.c (rb_str_rstrip_bang): ditto.
+
+	* string.c (rb_str_split_m): ditto.
+
+Mon Sep 14 16:39:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* debug.c, parse.y: fixed types.
+
+	* node.h (nd_line): limit to int.
+
+Mon Sep 14 11:23:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, common.mk, */configure.bat (rdoc): make before
+	  install if rdoc is enabled.  [ruby-dev:39325]
+
+Mon Sep 14 10:56:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (check-ruby): run all test of ruby itself.
+
+Mon Sep 14 10:44:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_TYPE_P): should not use BUILTIN_TYPE for
+	  special constants.
+
+Mon Sep 14 10:08:19 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): use rb_enc_mbc_to_codepoint
+	  because we already knew char is found and got length.
+
+Mon Sep 14 09:59:03 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (NUM2INT, NUM2LL, INT2NUM, UINT2NUM),
+	  (LONG2NUM, ULONG2NUM, NUM2CHR): get rid of backward references of
+	  macros.  the code couldn't compile by VC.
+
+Mon Sep 14 08:33:11 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/net/http.rb (each_*): return enumerator if no block present.
+	  Patch by Arthur Schreiber [ruby-core:18310]
+
+Mon Sep 14 06:42:21 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/cgi/cookie.rb (value): Keep CGI::Cookie#value in sync with the
+	  cookie itself. Based on a patch by Arthur Schreiber [ruby-core:17634]
+
+Mon Sep 14 05:21:12 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/net/http.rb (fetch): Handle properly default values; a patch by
+	  Arthur Schreiber [ruby-core:18308]
+
+Mon Sep 14 04:07:09 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/set.rb (==): Optimization; patch by Arthur Schreiber [ruby-core:17203]
+
+Mon Sep 14 03:30:23 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/pre-build: added rule to generate id.h for Symbian build.
+
+	* symbian/configure.bat: fixed harmless error message.
+
+Sun Sep 13 22:18:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_mod_remove_const): do not change VM state when an
+	  exception will occur.
+
+Sun Sep 13 21:25:01 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (oletypelib_get_libattr): some refactoring
+	  by adding oletypelib_get_libattr.
+
+Sun Sep 13 20:55:19 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (oletypelib_get_typelib): some refactoring
+	  by adding oletypelib_get_typelib.
+
+Sun Sep 13 20:18:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_sflag, proc_options, load_file_internal): fixed
+	  types.
+
+Sun Sep 13 19:39:59 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: some refactoring.
+
+Sun Sep 13 19:38:34 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): no need to define DATA when error.
+
+Sun Sep 13 18:48:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (XLDFLAGS): link startup code with ObjC support.
+
+Sun Sep 13 13:38:00 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* thread.c (recursive_*): refactored the access to the inspect
+	  table used by rb_exec_recursive_*. The functions recursive_push,
+	  pop and check now assume a valid hash table as their first
+	  argument. Added documentation.
+
+Sun Sep 13 12:07:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (SYM2ID): needs parens.
+
+	* include/ruby/ruby.h (NUM2INT, NUM2LL, INT2NUM, UINT2NUM),
+	  (LONG2NUM, ULONG2NUM, NUM2CHR, rb_type_p, rb_special_const_p):
+	  GCC specific optimization.
+
+Sun Sep 13 11:06:12 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI::Meta#content_type_parse): strip quotes.
+
+Sun Sep 13 09:38:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/prime.rb (EratosthenesGenerator#initialize): call super.
+	  (TrialDivisionGenerator, Generator23): ditto.  [ruby-core:25539]
+
+Sun Sep 13 09:34:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_add_method_def): no warning for inherited
+	  method.
+
+Sun Sep 13 08:30:30 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: fix WIN32OLE_TYPELIB.new when the 1st
+	  argument is the non-registered file.
+
+Sun Sep 13 02:08:43 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: change members of iseq_inline_cache_entry.
+	  make cache value members to one union member "ic_value".
+
+	* insns.def: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+Sun Sep 13 01:15:49 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (URI::FTP#buffer_open): fix the %2F handling.
+
+Sun Sep 13 00:46:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (CALL_METHOD_HOOK): refined with macro.
+
+	* vm_method.c (rb_remove_method_id): no definition body is
+	  undefined.
+
+	* vm_method.c (rb_add_method, rb_add_method_me): call method added
+	  hook after definition.  [ruby-core:25536]
+
+	* vm_method.c (rb_alias): hooks are called from rb_add_method_def.
+
+Sat Sep 12 22:47:24 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (URI::FTP#buffer_open): use the port specified in
+	  the URI.
+
+Sat Sep 12 17:31:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (GCC): subst for GCC depending extension libraries.
+
+Sat Sep 12 07:52:59 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (EVENTSINK_Invoke): initialize result
+	  variant value.
+
+Fri Sep 11 21:52:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* NEWS: update for 1.9.2.  based on a patch from Run Paint Run Run
+	  in [ruby-core:25534].
+
+Fri Sep 11 21:38:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check for real target cpu on darwin 10.
+
+Fri Sep 11 18:51:57 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/fileutils: use require_relative to require fileasserts.
+
+Fri Sep 11 14:22:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (debugflags): use gdb by default on linux and
+	  darwin.
+
+	* configure.in (XLDFLAGS): keep ARCHFILE macro on AIX.
+
+	* configure.in (RUBY_REPLACE_TYPE): define typename to default
+	  type if the latter is found.
+
+Fri Sep 11 13:57:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (fstat): override if large file support is
+	  enabled.
+
+Fri Sep 11 11:33:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/config.{guess,sub}: updated to automake-1.11.
+
+Fri Sep 11 10:38:33 2009  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTPHeader::encode_kvpair): also call to_s
+	  to k.  A patch from swdyh <youhei@gmail.com>
+	  http://github.com/swdyh/ruby/tree/c847f43c2ccb679b9ff728f8b1b16c6ceeb57f39
+
+Fri Sep 11 09:45:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (aix_loaderror): get rid of using uninitialized value in the
+	  case loadquery fails.  fixed wrong index variable usage.  see
+	  [ruby-core:25479].
+
+Fri Sep 11 07:52:43 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* unicode.c (onigenc_unicode_property_name_to_ctype):
+	  ignore case of properties.
+
+	* tool/enc-unicode.rb: downcase properties list.
+
+	* enc/unicode/name2ctype.h, enc/unicode/name2ctype.h.blt,
+	  enc/unicode/name2ctype.kwd, enc/unicode/name2ctype.src:
+	  follow above.
+
+Fri Sep 11 05:00:19 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h (rb_data_type_t): Add comments.
+	  And add a member variable "data", a multi-purpose storage
+	  area for rb_data_type.
+
+Fri Sep 11 02:14:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/dl: use require_relative to require test_base.rb.
+
+Fri Sep 11 02:06:44 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (rb_char_to_option_kcode): ASCII-8BIT should
+	  also delay.
+
+	* re.c (parser_regx_options): return rb_ascii8bit_encindex on
+	  ASCII-8BIT. [ruby-dev:39300]
+
+Fri Sep 11 01:14:00 2009  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/net/http.rb (fetch): rdoc fix, patch by Arthur Schreiber.
+	  [ruby-core:18309].
+
+Thu Sep 10 23:00:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): fix for parallel execution.
+	  [ruby-core:25509]
+
+Thu Sep 10 21:22:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/dl/test_cptr.rb (test_free=): test SEGV at first.
+	  [ruby-dev:39269]
+
+Thu Sep 10 21:20:59 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_normal_exit): Don't use
+	  AssertionMessage.
+
+Thu Sep 10 15:59:05 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (UNKNOWN_ESC_WARN): added.
+
+	* regparse.c (conv_backslash_value): Warn unknown
+	  escaped chars in regexp. [ruby-dev:39104]
+
+Wed Sep  9 22:02:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::Entry_#copy_file): open the source
+	  file first to ensure it can be copied.  [ruby-core:25498]
+
+Wed Sep  9 21:20:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_data_type): typed.
+
+	* ext/dl/cfunc.c (dlcfunc_data_type): typed.
+
+	* ext/dl/cptr.c (dlptr_data_type): ditto.
+
+	* ext/dl/handle.c (dlhandle_data_type): ditto.
+
+Wed Sep  9 17:17:31 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (parser_regx_options): only one kcode should effect
+	  options [ruby-core:25411]
+
+Wed Sep  9 15:46:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_feature_provided): fixed for autoloading extension
+	  library without suffix.
+
+Wed Sep  9 15:24:32 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* include/ruby/st.h : revert previous commit.
+
+	* ext/objspace/objspace.c : remove st_memsize declare.
+
+Wed Sep  9 14:07:19 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* include/ruby/st.h : fix duplicate st_memsize declare.
+
+Wed Sep  9 13:33:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_autoload): initialize typed data.
+
+Wed Sep  9 13:10:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_free_table): constified.
+
+Wed Sep  9 13:09:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_data_type): typed.
+
+	* enumerator.c (enumerator_data_type): typed.
+	  (yielder_data_type, generator_data_type): ditto.
+
+	* error.c (name_err_mesg_data_type): typed.
+
+	* file.c (stat_data_type): typed.
+
+	* thread.c (thgroup_data_type, mutex_data_type, barrier_data_type):
+	  typed.
+
+	* time.c (time_data_type): typed.
+
+	* transcode.c (econv_data_type): typed.
+
+	* variable.c (autoload_data_type): typed.
+
+Wed Sep  9 11:11:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_data_type_struct): constified dsize.
+
+Wed Sep  9 11:07:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_memsize): fixed wrong expression on IA64.
+
+Wed Sep  9 10:51:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_restore_1, rb_cont_call): should be Fiber.
+
+Wed Sep  9 00:27:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_data_type, fiber_data_type): typed.
+
+Tue Sep  8 22:37:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb, lib/tmpdir.rb (Tmpname): extracted new module.
+	  [ruby-dev:39197]
+
+Tue Sep  8 22:18:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_init_*table_with_size): use st_index_t.
+
+	* include/ruby/st.h (st_hash_func): use st_index_t.
+
+Tue Sep  8 21:48:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_thread_mark): mark callers iseqs.  [ruby-core:25474]
+	  [Bug #2062]
+
+Tue Sep  8 11:53:58 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_mark): no need to mark inline cache entries.
+
+	* insns.def (onceinlinecache, setinlinecache): save a value
+	  to mark cached value.
+
+Tue Sep  8 08:32:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/dl/test_{cfunc,ptr}.rb: added tests from Aaron Patterson.
+	  see [ruby-dev:39249].
+
+Mon Sep  7 17:22:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_tell): adjustment for ungotten data.
+
+Mon Sep  7 17:13:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_encname_bom_p): removed magic number.
+
+Mon Sep  7 12:26:04 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/inspector.rb (IRB::INSPECTORS.def_inspector): support
+	  object without #inspect defined.  a patch from Daniel
+	  Bovensiepen.  [ruby-core:25200]
+
+Mon Sep  7 05:38:34 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (opt_*): add IC operands.
+
+	* vm_insnhelper.h (CALL_SIMPLE_METHOD): add a version which
+	  use an inline cache.  USE_IC_FOR_SPECIALIZED_METHOD macro
+	  switches the behaviour.  This change also removes
+	  CALL_SIMPLE_METHOD_IC() macro.
+
+	* tool/instruction.rb: fix elimination process to ignore
+	  variable "ic".
+
+Mon Sep  7 05:21:09 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* Makefile.in, common.mk: move a id.h generation rule.
+
+Mon Sep  7 05:07:59 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: remove RUBY_VERSION output.
+
+Mon Sep  7 05:06:16 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: rename macro name ENABLE_IC_FOR_IVAR
+	  to USE_IC_FOR_IVAR.
+
+Mon Sep  7 03:21:40 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.escape): obsoleted.
+
+	* lib/uri/common.rb (URI.unescape): ditto.
+
+Sun Sep  6 18:13:54 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h (CALL_SIMPLE_METHOD_IC): make a macro
+	  invoke simple method with inline cache entry.
+
+	* insns.def (opt_length, opt_size): fix to use inline method cache.
+
+Sun Sep  6 17:47:21 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* template/id.h.tmpl: fix this.
+
+	* id.h: removed.  Because this file is generated automatically.
+
+Sun Sep  6 17:31:28 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_specialized_instruction), insns.def (opt_size):
+	  optimize #size methods (by specialized instruction).
+
+	* id.c, id.h, vm.c, vm_insnhelper.h: ditto.
+
+Sun Sep  6 16:13:06 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (setinstancevariable), vm_insnhelper.c (vm_setivar):
+	  fix to use inline cache (trivial optimization).
+
+Sun Sep  6 10:34:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c: fixed rdoc, a patch from Nobuhiro IMAI at [ruby-core:25433].
+
+Sun Sep  6 05:19:09 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c: Add rdoc for ARGF.
+	  contributed by Run Paint Run Run. [ruby-core:23854]
+
+Sat Sep  5 15:21:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): op_asgn to aref should return rhs.
+	  [ruby-core:25387]
+
+Sat Sep  5 10:38:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): &&= and ||= should return rhs.
+	  [ruby-dev:39163] (#1996), [ruby-core:25143]
+
+Sat Sep  5 08:51:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (update_char_offset): position should be long.
+
+	* re.c (match_hash, match_equal): new methods.  [ruby-core:24748]
+
+	* re.c (reg_match_pos, rb_reg_eqq, rb_reg_s_quote): get rid of use
+	  VALUE as int.
+
+Fri Sep  4 20:40:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (round): added declaration.  [ruby-dev:39222]
+
+Fri Sep  4 06:15:39 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (Init_Dir): alias Dir#path to Dir#to_path.  [ruby-core:25326]
+
+Fri Sep  4 04:49:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_rand): fixed rdoc.  [ruby-core:25332]
+
+Fri Sep  4 04:46:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/httpservlet/abstract.rb (do_OPTIONS): method names
+	  are symbols now.  [ruby-core:24580]
+
+Thu Sep  3 17:56:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (literal_concat_gen): concat body from dstr instead of
+	  nd_next.  [ruby-core:25284]
+
+Wed Sep  2 16:49:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (math_gamma): get rid of direct comparison between too
+	  big double and integer, with gcc on x86_64.  [ruby-core:25257]
+
+Wed Sep  2 13:47:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (domain_check): simplified.
+
+Wed Sep  2 11:32:24 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_free): fix to free method table (fix memory leak).
+
+Wed Sep  2 07:42:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/instruction.rb (RubyVM::InstructionsLoader#make_stackcaching_insns):
+	  simplified.
+
+Wed Sep  2 02:32:46 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/common.rb (NaN): Change definition
+	  of NaN to 0.0/0 for 1.8/1.9 compatibility.
+
+Wed Sep  2 01:16:32 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Update to JSON 1.1.9.
+
+Tue Sep  1 19:56:28 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (eval_string_with_cref): fix to check local_table_size.
+	  [ruby-dev:39205] [Bug #2024]
+
+Mon Aug 31 16:20:41 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* class.c (make_singleton_class): variable name changed.
+	  removed an unnecessary conditional.
+
+Mon Aug 31 14:17:09 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* class.c: refactored singleton class related matters.
+	  Handles eigenclasses and plain classes transparently.
+
+	  (make_metaclass): renamed from make_metametaclass.
+	  (METACLASS_OF): new utility macro
+	  (META_CLASS_OF_CLASS_CLASS): ditto.
+	  (ENSURE_EIGENCLASS): ditto.
+	  (make_singleton_class): extracted from rb_singleton_class.
+	  (boot_defclass): moved from object.c
+	  (Init_class_hierarchy): extracted from Init_Object.
+	  (rb_make_metaclass): refactored.
+	  (singleton_class_of): extracted from rb_singleton_class.
+	  (rb_singleton_class): refactored.
+	  (rb_define_singleton_method): it needs a metaclass only
+	  but not its metametaclass.
+
+	* object.c: booting class hierarchy was moved to class.c
+	  for keeping dependency between compilation units least.
+	  (Init_Object): extracting the booting into
+	  Init_class_hierarchy.
+	  (boot_defclass): moved to class.c.
+
+Sun Aug 30 23:44:09 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): use mktime for the first guess.
+
+Sun Aug 30 16:38:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_enc_symname2_p): not depend on nul terminator.
+
+Sun Aug 30 14:11:45 2009  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: dependencies updated.
+
+Sun Aug 30 13:00:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (add): shortcut implemented for fixnums.
+	  (sub): ditto.
+	  (mul): ditto.
+
+Sun Aug 30 10:24:43 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (eq): apply RTEST.
+	  (ne): ditto.
+	  (add): avoid method dispatch for bignums.
+	  (sub): ditto.
+	  (mul): ditto.
+	  (mod): ditto.
+
+Sun Aug 30 09:45:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigmul1_single): new function specialized respect to
+	  multiply two single digit bignums.
+	  (bigmul0): use bigmul1_single.
+
+Sun Aug 30 03:59:43 2009  Tanaka Akira  <akr@fsij.org>
+
+	* timev.h (TIME_SCALE): defined as 1000000000.
+	  (struct vtm): subsec is replaced by subsecx.
+	  subsec * TIME_SCALE == subsecx.
+
+	* time.c: avoid rational in most cases.
+	  (struct time_object): timev is replaced by timexv.
+	  timev * TIME_SCALE == timexv.
+
+Sun Aug 30 03:17:25 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (init_leap_second_info): use TIMET_MAX.
+
+Sun Aug 30 01:15:31 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_read_all): use gzfile_newstr;
+	  set and convert its encoding. [ruby-dev:38304]
+
+Sat Aug 29 20:40:02 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* vm_eval.c (rb_call0): gets rid of checking method cache twice.
+
+	* method.h (rb_get_method_entry): added a prototype of the function.
+	  (rb_method_entry_without_cache): more friendly name.
+
+Sat Aug 29 12:16:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tmpdir.rb (Dir.mktmpdir): rolled back r24699.  [ruby-dev:39193]
+
+Sat Aug 29 03:27:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#make_tmpname): removed thread race
+	  condition.
+
+	* lib/tmpdir.rb (Dir.mktmpdir): ditto.
+
+Fri Aug 28 20:29:34 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/tempfile.rb (Tempfile#callback): Debug information should be
+	  output to stderr, not stdout; pointed out by akira yamada.
+	  cf. [ruby-dev:39072]
+
+Fri Aug 28 20:34:24 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: use SafeStringValue instead of
+	  Check_SafeStr.
+
+Fri Aug 28 13:30:43 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (do_select): rollback r24680. void struct initializer is
+	  invalid.
+
+Fri Aug 28 11:45:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* method.h (rb_method_definition_t): split from rb_method_entry_
+	  to deal aliases.  [ruby-dev:39165]
+
+	* proc.c (struct METHOD): contains rb_method_entry_t copy.
+
+Fri Aug 28 10:21:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_mark): skip outdated cache entries.
+
+	* vm_core.h ({GET,INC}_VM_STATE_VERSION): moved from
+	  vm_insnhelper.h.
+
+Fri Aug 28 07:25:25 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (next_i): typo fixed (reached at end -> reached an
+	  end).  pointed out by James Edward Gray II at LoneStar RubyConf.
+
+Thu Aug 27 18:31:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_remove_method_id): exported.
+
+	* numeric.c (num_sadded): fix for non-ascii method name.
+
+Thu Aug 27 14:32:31 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_preprocess_dregexp): set encoding as ASCII-8BIT
+	  when /n is specified and the embedded string is escaped text.
+
+Thu Aug 27 13:51:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_rand): random integer can be a fixnum for
+	  bignum range.  [ruby-dev:39173]
+
+Thu Aug 27 08:16:34 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/strscan/strscan.c (strscan_set_string): set string should not be
+	  duped or frozen, because freezing it causes #concat method failure,
+	  and unnecessary to dup without freezing.  a patch from Aaron
+	  Patterson at [ruby-core:25145].
+
+Thu Aug 27 02:06:11 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (enc/unicode/name2ctype.kwd):
+	  remove rules to generate name2ctype.kwd from
+	  UnicodeData.txt and Scripts.txt.
+	  if you want to generate, use tool/enc-unicode.rb.
+
+Thu Aug 27 02:00:09 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* unicode.c (PROPERTY_NAME_MAX_SIZE): use MAX_WORD_LENGTH.
+
+Wed Aug 26 23:59:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_rand): refined error message.
+
+	* random.c (random_rand): fixed for edge cases of ranges.
+	  [ruby-dev:39166]
+
+Wed Aug 26 21:49:23 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/tempfile.rb: add documents from Hongli Lai's fork.
+	  cf [ruby-core:25131].
+
+Wed Aug 26 19:51:13 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/mkconfig.rb (program_transform_name): fix for multiple trans
+	  rules for autoconf 2.61 or earlier.
+
+	* tool/rbinstall.rb (program_transform_name): ditto.
+
+Wed Aug 26 19:20:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_rand): unified random_int and random_float.
+	  [ruby-dev:39158].  and fixes [ruby-core:24655], [ruby-core:24677],
+	  [ruby-core:24679].
+
+Wed Aug 26 18:59:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/test_tempfile.rb: merged from Hongli Lai's fork.
+	  cf [ruby-core:25131].
+
+Wed Aug 26 18:49:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#close!): should not undefine finalizer
+	  by just unlink.
+
+Wed Aug 26 17:00:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb (program_transform_name): fix for autoconf 2.61
+	  or earlier.
+
+Wed Aug 26 14:34:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#close!, Tempfile#path): added side
+	  notes from Hongli Lai's fork.
+
+	* lib/tempfile.rb (Tempfile#unlink, Tempfile.callback): do nothing
+	  any more once unlinked.
+
+Wed Aug 26 13:48:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#unlink): reverted r23494, since the
+	  usage in RubyInline is considered wrong.
+
+Wed Aug 26 12:36:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (collect_local_variables_in_env): skips internal variables.
+	  [ruby-core:25125]
+
+Tue Aug 25 23:51:07 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb: added for generate name2ctype.kwd.
+	  contributed by Run Paint Run Run [ruby-core:24775]
+
+	* enc/unicode.c (CodeRanges): move definitions to name2ctype.h.
+
+	* enc/unicode/name2ctype.h.blt, enc/unicode/name2ctype.kwd,
+	  enc/unicode/name2ctype.src: updated to v5.1.
+
+	* enc/unicode/UnicodeData.txt, enc/unicode/Scripts.txt: added v5.1.
+
+	* Makefile.in: add rule to generate name2ctype.kwd from
+	  UnicodeData.txt and Scripts.txt.
+
+Tue Aug 25 22:31:51 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (MKDIR_P): Set 'mkdir -p' to MKDIR_P
+	  when AC_PROG_MKDIR_P doesn't set MKDIR_P.
+
+Tue Aug 25 17:38:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_clone, bigmul1_normal, bigdivrem): trivial
+	  optimization.
+
+	* bignum.c (big2dbl): truncates zero digits to get rid of possible
+	  underflow.
+
+Tue Aug 25 12:22:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (enc/unicode/name2ctype.h): explicitly ignores the
+	  result of diff and turns -e option off, because *BSD make passes
+	  it by default.
+
+Tue Aug 25 02:16:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DLDFLAGS): use linker_flag and changed undefined
+	  and multiply_defined behaviors. cf [ruby-core:25086].
+
+Mon Aug 24 21:31:37 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb, test/rss/test_maker_2.0.rb: fix a bug
+	  that RSS Maker doesn't accept 'false' as guid's isPermaLink.
+	  Reported by Joe Holt. Thanks!!!
+
+Mon Aug 24 18:58:56 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/missing.h (vsnprintf): rollback a part of r24179, because
+	  it's meaningless.
+
+Mon Aug 24 16:35:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_delete_safe): deals with packed entries.
+	  [ruby-core:25080]
+
+	* st.c (st_cleanup_safe): ditto.  [ruby-core:25081]
+
+Mon Aug 24 13:24:07 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (MAKEDIRS): define.
+
+	* common.mk (capi): using $(MAKEDIRS), so depends on $(PREP).
+
+Mon Aug 24 13:14:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): set cross_compiling only when
+	  universal binary.
+
+Mon Aug 24 12:55:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (enc/unicode/name2ctype.h): use md instead of
+	  $(MAKEDIRS).
+
+Sun Aug 23 15:22:45 2009  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (main): "usage" description updated.
+
+Sun Aug 23 15:12:22 2009  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (Dir.mktmpdir): updated to latest.
+	  (in_temporary_working_directory): temporary directory name changed.
+
+Sun Aug 23 00:56:13 2009  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (rb_thread_schedule): don't recur infinitely.
+	  (rb_threadptr_execute_interrupts): ditto.
+	  [ruby-dev:38060]
+
+Sat Aug 22 15:07:23 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/syck/rubyext.c (id_hash_new): new function to create a hash
+	  which key is compared by object id.
+	  (syck_emitter_reset): use id_hash_new for bonus->data.
+
+	* lib/yaml.rb (YAML.quick_emit): give the object itself to emitter.
+	  don't use object_id and hash.
+
+Sat Aug 22 13:05:22 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in: use CP and MV macros.
+
+Sat Aug 22 01:29:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (rm_f, rm_rf): pass the last hash through if exists.
+	  [ruby-dev:39153]
+
+Sat Aug 22 00:48:08 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enumerator.c (ary2sv): add dup argument.
+	  (enumerator_next): call ary2sv with dup=0.
+	  (enumerator_peek): call ary2sv with dup=1 to return duplicated array.
+	  (enumerator_peek_values_m): new function to return duplicated array.
+	  (Init_Enumerator): use enumerator_peek_values_m as
+	  Enumerator#peek_value.
+
+Sat Aug 22 00:03:19 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_check_deadlock): decrease number of sleepers before
+	  deadlock detection because the deadlock exception makes main thread
+	  run.  [ruby-dev:39142]
+
+Fri Aug 21 22:34:58 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enumerator.c (get_next_values): extracted from
+	  enumerator_next_values.
+	  (enumerator_next_values): use get_next_values.
+	  (enumerator_peek_values): ditto.
+
+Fri Aug 21 17:01:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/unicode/name2ctype.h: split from enc/unicode.c and made a
+	  perfect hash.
+
+Fri Aug 21 15:13:08 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/io.h, io.c (FMODE_SETENC_BY_BOM):
+	  renamed from FMODE_STRIP_BOM.
+
+Thu Aug 20 01:24:55 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_fmode_modestr): change modestr syntax for BOM
+	  to "BOM|UTF-*". [ruby-dev:39106]
+
+	* io.c (parse_mode_enc): ditto.
+
+Fri Aug 21 15:01:35 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_readline): use rb_prep_terminal
+	  only on Windows.
+
+Fri Aug 21 07:25:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/gemdirs.rb: split from lib/rdoc/ri/paths.rb to ge
+	  rid of loading rubygems and searching all gems always.
+
+Fri Aug 21 07:14:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_PROGRAM_VERSION, RUBY_RELEASE_DATE): extracts
+	  from version.h for cross-compiling.
+
+	* template/fake.rb.in (RUBY_VERSION, RUBY_DESCRIPTION): use above.
+
+Fri Aug 21 00:08:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/paths.rb (RDoc::RI::Paths): Gem::Enable has been obsolete.
+
+Thu Aug 20 23:56:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_sysopen): moved sysopen_struct from rb_sysopen_internal.
+
+Thu Aug 20 23:39:51 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (reduce_nodes_gen): preserve NODE_FL_NEWLINE flag during
+	  node reducing.  [ruby-core:24463]
+
+Thu Aug 20 14:39:47 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_get): add rl_prep_terminal(1).
+	  incited by jitte [ruby-list:43546]
+
+Thu Aug 20 12:09:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ivar2_hash_type): disabled for now.
+
+Thu Aug 20 08:39:50 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (rb_thread_terminate_all): do not ignore interrupt when
+	  reaping threads on termination.  [ruby-dev:39107]
+
+Thu Aug 20 02:32:08 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enumerator.c (next_init): don't clear feedvalue.
+
+Thu Aug 20 01:28:42 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enumerator.c: implement Enumerator#{next_values,peek_values,feed}
+	  and StopIteration#result.  [ruby-dev:39109]
+	  (struct enumerator): replace no_next by stop_exc.
+	  new field feedvalue.
+	  (enumerator_mark): mark feedvalue and stop_exc.
+	  (enumerator_init): initialize feedvalue and stop_exc.
+	  (enumerator_init_copy): initialize feedvalue.
+	  (next_ii): send yield arguments as an array.  return feedvalue.
+	  (next_i): generate StopIteration exception here.  set result.
+	  (next_init): initialize feedvalue.
+	  (enumerator_next_values): new method Enumerator#next_values.
+	  (ary2sv): new function.
+	  (enumerator_peek_values): new method Enumerator#peek_values.
+	  (enumerator_feed): new method Enumerator#feed.
+	  (yielder_yield): return the yield value.
+	  (generator_each): return the iterator value.
+	  (stop_result): new method StopIteration#result.
+
+Thu Aug 20 01:06:48 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (DEFINE_STRUCT_DIRENT): use union to allocate sufficient
+	  memory space for Solaris.  a patch from Naohisa GOTO
+	  <ngoto at gen-info.osaka-u.ac.jp> in [ruby-dev:39132].
+	  [ruby-dev:39062]
+
+	* configure.in (SIZEOF_STRUCT_DIRENT_TOO_SMALL): Solaris dirent
+	  check.
+
+Wed Aug 19 11:32:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/unicode.c (CodeRanges): initialized statically.
+
+Wed Aug 19 02:54:01 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_settracefunc.rb (test_return, test_return2): add two
+	  tests for [ruby-dev:38701] and [ruby-core:24463].
+
+Wed Aug 19 01:08:34 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (NODE_RETURN): fire return event at explicit return.
+	  [ruby-dev:38701]
+
+Tue Aug 18 21:00:26 2009  Tanaka Akira  <akr@fsij.org>
+
+	* enumerator.c (enumerator_peek): new method Enumerator#peek.
+	  (enumerator_next): don't rewind at end.
+	  [ruby-dev:38932]
+
+Tue Aug 18 13:46:14 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* touch test/rdoc/empty.dat to run test_rdoc_parser.rb
+
+Tue Aug 18 11:37:24 2009  wanabe  <s.wanabe@gmail.com>
+
+	* vm_insnhelper.c (vm_call_cfunc): ensure hook c-return.
+	  [Bug #1588]
+
+	* test/ruby/test_settracefunc.rb (TestSetTraceFunc#test_raise):
+	  follow above.
+
+Tue Aug 18 01:57:00 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_step): treat symbols specially so that iterating
+	  over symbols should work like strings.  [ruby-core:24780]
+
+	* range.c (range_each): ditto.
+
+Tue Aug 18 01:21:31 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_each): should honor to_str conversion.
+
+Mon Aug 17 23:45:40 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/delta.rb: removed require 'date'.  added to_c.
+
+Mon Aug 17 14:35:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lex_get_str, lex_io_gets, rb_parser_compile_string):
+	  must be ascii compatible.
+
+Mon Aug 17 10:37:41 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (add_code_range_to_buf0): added with checkdup argument.
+
+	* regparse.c (add_code_range_to_buf): use above.
+
+	* regparse.c (add_code_range0): added with checkdup argument.
+
+	* regparse.c (add_code_range): use above.
+
+	* regparse.c (i_apply_case_fold): don't warn if the duplicate is
+	  caused by case folding.
+
+Mon Aug 17 08:31:56 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/delta.rb: merged from date4.  [experimental]
+
+	* lib/date/delta/parser.*: ditto.
+
+	* lib/date.rb: followed the above changes.
+
+Mon Aug 17 08:19:03 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (strptime): removed \v; since \s includes \v.
+
+Mon Aug 17 08:14:26 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_rationalize) added.  [experimental]
+
+	* rational.c ({nurat,nilclass,integer,float}_rationalize) ditto.
+
+Mon Aug 17 08:11:53 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb: use num#i.
+
+Mon Aug 17 07:59:00 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c (flo_pow,fix_pow): may return complex number.
+
+	* bignum.c (rb_big_pow): ditto.
+
+Mon Aug 17 07:16:10 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (num_imaginary): num#i to return imaginary counterpart
+	  of the given numeric.
+
+	* complex.c (Init_Complex): undef #i for complex numbers.
+
+Mon Aug 17 00:17:33 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/complex.rb, lib/rational.rb: added warning messages.
+
+Sun Aug 16 23:58:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (yylex): should dispatch scan-event even when follows
+	  just after delayed-token.  [ruby-dev:37855] [Bug #1071]
+
+Sun Aug 16 22:20:16 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: reverted.
+
+Sun Aug 16 21:31:21 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/date/format.rb: suppressed a warning.
+
+	* lib/irb/ruby-lex.rb: ditto.
+
+Sun Aug 16 15:25:26 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/csv.rb: Change magic comment to US-ASCII in order to
+	  make literals as US-ASCII.
+
+Sun Aug 16 10:45:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regparse.c (parse_char_class, parse_exp, parse_branch),
+	  (parse_subexp): fixed memory leak.  a patch from Ralf Junker
+	  <ralfjunker AT gmx.de> at [ruby-core:24921].
+
+Sun Aug 16 10:38:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_backtrace_each, vm_backtrace_push),
+	  vm_eval.c (print_backtrace), vm_dump.c (bugreport_backtrace):
+	  rb_backtrace_iter_func now takes VALUE as file and method names.
+
+Sun Aug 16 03:06:59 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (opt_eq_func): fix optimization bug.  This issue
+	  was found out and debugged with Takuto Hayashi at Security and
+	  Programming camp 2009.
+
+Sun Aug 16 01:10:00 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (add_ctype_to_cc_by_range): fix the first
+	  character bigger than sb_out was dropped.
+
+	* test/ruby/test_regexp.rb (TestRegexp#test_posix_bracket):
+	  add tests for above.
+
+Sun Aug 16 00:30:33 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/oniguruma.h
+	  (ONIGENC_CTYPE_SPECIAL_MASK): added.
+	  (ONIGENC_CTYPE_D): ditto.
+	  (ONIGENC_CTYPE_S): ditto.
+	  (ONIGENC_CTYPE_W): ditto.
+
+	* regparse.c: \d, \s and \w are now non Unicode class.
+	  [ruby-dev:39026]
+	  (fetch_token_in_cc): use ONIGENC_CTYPE_[DSW] for \d/\s/\w.
+	  (fetch_token): ditto.
+	  (add_ctype_to_cc): add routines for ONIGENC_CTYPE_[DSW].
+	  (parse_exp): ditto.
+
+	* test/ruby/test_regexp.rb (TestRegexp#test_char_class):
+	  add tests for above.
+
+Sat Aug 15 10:39:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (fname, string_dvar, sym, dsym, f_arglist): removed
+	  duplications.
+
+Fri Aug 14 20:03:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_check_safe_str): deprecated.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_add_{file,path}):
+	  replaced deprecated function.
+
+Fri Aug 14 17:59:12 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_preprocess_dregexp): change Exception class to
+	  RegexpError.
+
+	* test/ruby/test_m17n.rb (test_regexp_usascii): follow above.
+
+	* test/ruby/test_m17n.rb (test_regexp_embed): ditto.
+
+Fri Aug 14 17:17:42 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/Makefile.in (MKDIRS): revert r24525.
+
+Fri Aug 14 16:28:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/lib/digest/hmac.rb (Digest::HMAC#initialize): faster
+	  code.
+
+Fri Aug 14 14:31:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rubyhdrdir): fixed typo.  [ruby-dev:39079]
+
+Fri Aug 14 00:29:22 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* doc/ChangeLog-1.8.0: add forgotten entry contributed by
+	  TAKAHASHI Kaoru.  [ruby-dev:39065]
+
+Fri Aug 14 00:19:49 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#method_missing): __FILE__ may contain
+	  multi-byte characters.  a patch from Kenta Murata in [ruby-dev:39066].
+
+Thu Aug 13 21:01:03 2009  wanabe  <s.wanabe@gmail.com>
+
+	* vm.c (vm_exec): returning from lambda runs ensure section.
+	  [Bug #1729]
+
+Thu Aug 13 18:40:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): set cross_compiling to yes
+	  only when AC_COMPUTE_INT, and inverted the arguments.
+
+Thu Aug 13 18:22:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): use AC_COMPUTE_INT instead of
+	  _AC_COMPUTE_INT_COMPILE with cross compiling, since its arguments
+	  have been changed at autoconf 2.64.
+
+Thu Aug 13 16:31:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (opt_case_dispatch): suppressed a warning.
+
+	* vm_insnhelper.c (opt_case_dispatch_i): ditto.
+
+Thu Aug 13 16:20:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, Makefile.in (MAKEDIRS): used MKDIR_P instead of
+	  as_mkdir_p.  [ruby-dev:39063]
+
+Thu Aug 13 15:37:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regenc.c (onigenc_strlen_null, onigenc_str_bytelen_null): fixed
+	  infinite loop for wide encodings.  reported by Ralf Junker a
+	  [ruby-core:24892].  [ruby-core:24904]
+
+Wed Aug 12 21:07:46 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: if ipv6 is enabled, the version of Windows
+	  must be XP or later.
+	  [ruby-core:24601]
+
+Wed Aug 12 15:59:29 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (yes-test-knownbug): use RUNRUBY instead of MINIRUBY.
+
+Wed Aug 12 15:52:04 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* class.c (rb_define_module_id_under): fix the name.
+
+	* class.c (rb_define_module_under): fix for previous changes.
+
+Wed Aug 12 15:32:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_define_class_id_under, rb_define_module_id_under):
+	  new functions to define a nested class/module with non-ascii
+	  name.
+
+	* struct.c (make_struct): use name with encoding.
+
+	* struct.c (inspect_struct): ditto.  [ruby-core:24849]
+
+Wed Aug 12 Wed Aug 12 14:54:34 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm.c, vm_insnhelper.c, vm_insnhelper.h: check
+	  definition of (classes)#=== for case/when optimization.
+	  Fix Bug #1376 [ruby-core:23190].
+
+	* string.c (Init_String), bignum.c (Init_Bignum),
+	  numeric.c (Init_Numeric): define String#===, Symbol#===,
+	  Bignum#===, Fixnum#===, Float#=== as same as (classes)#==.
+
+Wed Aug 12 14:14:42 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (readdir_internal): free old temporary filename.
+	  [ruby-core:24820]
+
+Wed Aug 12 12:59:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_new_frozen): must not change encoding of frozen
+	  shared string.  [ruby-dev:39068]
+
+Wed Aug 12 11:51:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_broken_crypt): needs more checks.
+
+Wed Aug 12 07:41:31 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): If a string is empty and
+	  other's encoding is US-ASCII, returns the empty string's encoding.
+	  [ruby-list:46274]
+
+Wed Aug 12 07:38:12 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (is_data_encoding): fix condition.
+
+	* encoding.c (enc_capable): ditto.
+
+Tue Aug 11 23:12:31 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (complete_frags): Monday must be suitable for %W's
+	  default day.
+
+Tue Aug 11 21:42:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_divmod): fixed rdoc.  [ruby-core:24862]
+
+Mon Aug 10 21:45:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_path2class): no deprecation.
+
+Mon Aug 10 10:57:59 2009  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c: reject unused longlife gc. longlife gc target is longlife
+	  NODE by method table and vm inline cache. but, fixed it at
+	  r24085, r24128. so I rejected longlife gc.
+
+	* debug.c: ditto.
+
+	* include/ruby/intern.h: ditto.
+
+	* include/ruby/ruby.h: ditto.
+
+	* iseq.c: ditto.
+
+	* node.h: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_insnhelper.h: ditto.
+
+Mon Aug 10 06:55:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (class2path, w_unique, w_extended, w_class, w_uclass):
+	  deal with non-ascii class path.  [ruby-core:24790]
+
+	* marshal.c (r_unique, path2class, path2module, obj_alloc_by_path),
+	  (r_object0): ditto.
+
+	* variable.c (rb_path_to_class): new encoding-aware function to
+	  get a class from its name.
+
+Sun Aug  9 21:14:03 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb (search_tclConfig): last change isn't enough.
+	  fixed it.
+
+Sun Aug  9 16:36:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ruby.imp): excluded prelude.o to get rid of circular
+	  dependency.  [ruby-dev:39052]
+
+Sun Aug  9 14:49:24 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (rb_big_cmp, bigsub, big_real_len, bigmul1_normal,
+	  bigmul1_balance, big_split): remove BDIGITS() inside of the loops.
+	  same as r24444.
+
+Sat Aug  8 17:03:21 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_ascii_compatible_p): added. [ruby-core:24793]
+	  (Init_Encoding): New API Encoding#ascii_compatible?.
+
+Sun Aug  9 07:25:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/ripper/eventids2.c (token_to_eventid): added
+	  keyword_do_LAMBDA.  [ruby-dev:39049]
+
+Sun Aug  9 02:07:41 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/compile_prelude.rb: fixes a regexp pattern for require.
+	  It had matched 'require("foo"('.
+
+Sat Aug  8 11:42:44 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigzero_p): removing BDIGITS() inside of the
+	  loop. inspired by Masahiro Kanai's blog entry
+	  <http://d.hatena.ne.jp/CanI/20090807/1249657492>.
+
+Sat Aug  8 06:18:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_symbol r_symlink, r_symbol, r_object0): fix for
+	  non-ascii symbols.  loading such symbols can cause segfaults in
+	  older versions.  [ruby-core:24788]
+
+Fri Aug  7 03:25:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.c (error_print): removed an extra argument.
+
+Fri Aug  7 03:22:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_exc_raise, rb_exc_fatal): nil is used to reraise.
+
+Fri Aug  7 01:49:41 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/sha2/sha2.c: The ULL suffix is not supported by
+	  pre-C99 compilers, so resurrect the ULL() macro to regain
+	  portability. [ruby-dev:39032]
+
+Fri Aug  7 01:35:26 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/*/extconf.rb: inttypes.h and unistd.h need not be
+	  checked here. [ruby-dev:39032]
+
+Fri Aug  7 01:04:17 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb (search_tclConfig): fix logic bug.
+
+Thu Aug  6 21:18:15 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/digest/sha2/lib/sha2.rb: should require sha2.so.
+
+Thu Aug  6 21:11:40 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/digest/sha2/sha2.c (*_Final): typos.
+
+Thu Aug  6 19:46:56 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/sha2/sha2.h (BYTE_ORDER): Define BYTE_ORDER as
+	  necessary. [ruby-dev:39029]
+
+Thu Aug  6 16:36:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (valid_encoding_name_p): rejects too long encoding
+	  names.
+
+	* encoding.c (encoding_data_type): typed data.
+
+	* encoding.c (enc_capable, rb_enc_get_index): Symbol is encoding
+	  capable.
+
+	* encoding.c (rb_enc_associate_index): cannot set encoding on
+	  special constants.
+
+Thu Aug  6 15:44:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_eq): deals with optimized method
+	  properly.  [ruby-core:24789]
+
+Thu Aug  6 13:30:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/mkmf/base.rb (TestMkmf): was turned into a class, because
+	  MiniUnit doesn't complain even if a testcase has no tests.
+
+Thu Aug  6 13:00:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (check_sizeof): added optional compiler option
+	  argument.  [ruby-core:24785]
+
+	* lib/mkmf.rb (create_makefile): suppressed shadowing outer local
+	  variable warnings.
+
+Thu Aug  6 12:05:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/testcase.rb (Test::Unit): removes silly TestCase
+	  class.
+
+Thu Aug  6 01:37:20 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/sha2/sha2.[ch]: Update to 1.0 RELEASE which fixes an
+	  off-by-one bug in SHA-256 hashing.  Reduce differences from
+	  the original while at it. [Bug #1799]
+
+Thu Aug  6 00:09:56 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (IPAddr#hash): Take account of netmask; submitted
+	  by Nobuhiro IMAI in [ruby-dev:39011]
+
+Wed Aug  5 19:19:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): assumes -x flag if no "ruby" is in
+	  the shebang line.  [ruby-dev:39015]
+
+Wed Aug  5 19:11:01 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (rb_stdio_set_default_encoding): declared.
+
+Wed Aug  5 18:46:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* debug.c (ruby_dummy_gdb_enums): made public.  [ruby-dev:39001]
+
+Wed Aug  5 13:49:09 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_upto): generate numerical sequence when
+	  characters in both edges are all digits.  [ruby-talk:343186]
+
+Wed Aug  5 12:54:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_exc_raise, rb_exc_fatal): require exception object.
+	  [ruby-core:24767]
+
+Wed Aug  5 12:39:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_longjmp): reset raised flag before fatal error.
+
+Wed Aug  5 10:20:39 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_preprocess_dregexp): add options to arguments.
+
+	* re.c (rb_reg_new_ary): follow above.
+
+	* re.c (rb_reg_preprocess_dregexp): change error message when
+	  /.../n has a non escaped non ASCII character in non ASCII-8BIT
+	  script. [ruby-dev:38524]
+
+Wed Aug  5 03:28:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit.setup_argv): expands paths before
+	  requiring.  [ruby-dev:39012]
+
+Wed Aug  5 01:38:27 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/pp.rb (guard_inspect_key): untrust internal hash to prevent
+	  unexpected SecurityError.
+
+	* test/ruby/test_object.rb: add a test for [ruby-dev:38982].
+
+Wed Aug  5 00:33:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parser/c.rb: fixed a small error in the documentation.
+	  [ruby-core:24744]
+
+Tue Aug  4 22:10:34 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (has_redirection): need to execute shell if commandline
+	  includes newline.  cf. [ruby-core:24560]
+
+Tue Aug  4 15:06:58 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (IPAddr#{eql?,hash}): Add IPAddr#{eql?,hash} so
+	  that an IPAddr object can be used as a hash key, a set element,
+	  etc.; suggested by Nick Brown <nick@nick-brown.com>.
+
+Tue Aug  4 13:07:10 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/oniguruma.h (ONIG_SYN_WARN_CC_DUP): defined.
+	* regparse.h (ScanEnv): add warnings_flag.
+	* regparse.c (CC_DUP_WARN): defined for warn duplicated characters in
+	  character class of regexp. [ruby-core:24593]
+	  (add_code_range_to_buf): add CC_DUP_WARN.
+	  (next_state_val): add CC_DUP_WARN.
+	  (OnigSyntaxRuby): add ONIG_SYN_WARN_CC_DUP.
+	  (SET_ALL_MULTI_BYTE_RANGE): add env to arguments.
+	  (add_code_range): ditto.
+	  (add_code_range_to_buf): ditto.
+	  (not_code_range_buf): ditto.
+	  (or_code_range_buf): ditto.
+	  (and_code_range1): ditto.
+	  (and_code_range_buf): ditto.
+	  (and_cclass): ditto.
+	  (or_cclass): ditto.
+	  (add_ctype_to_cc_by_range): ditto.
+	  (add_ctype_to_cc): ditto.
+	  (parse_char_class): ditto.
+
+Tue Aug  4 12:40:45 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/encdb.c (ENC_SET_BASE): fix typo.
+	  patch by ujihisa [ruby-dev:39004]
+
+Tue Aug  4 11:57:39 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_eql, range_eq): fixed equality to work for
+	  subclasses of Range.  a patch from Marc-Andre Lafortune.
+	   [ruby-core:22190]
+
+	* test/ruby/test_range.rb: add assertions for above.
+
+Tue Aug  4 09:41:11 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/big5.c (EncLen_Big5): back to original Big5 table.
+	  (EncLen_Big5_HKSCS): for Big5-HKSCS.
+	  (trans): add the lead byte table for Big5-HKSCS.
+	  (big5_mbc_enc_len): abstract function for Big5 series.
+	  (big5_mbc_enc_len): for Big5.
+	  (big5_hkscs_mbc_enc_len): for Big5-HKSCS.
+	  (BIG5_HKSCS_P): added.
+	  (BIG5_ISMB_FIRST): add routine for Big5-HKSCS.
+	  (big5_hkscs): add for Big5-HKSCS.
+
+Tue Aug  4 09:33:54 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_set_base): Add for setting base encoding
+	  with their names. this is internal function.
+
+	* template/encdb.h.tmpl: specify ENC_SET_BASE for second encodings in
+	  each encoding files.
+
+	* enc/encdb.c (rb_enc_set_base): add a declaration.
+	  (ENC_SET_BASE): ditto.
+
+Tue Aug  4 06:30:01 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_replace): should copy compare_by_identity status as well.
+	  [ruby-core:24728]
+
+Tue Aug  4 05:43:03 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (recursive_push): need to set UNTRUST.  [ruby-dev:38997]
+
+Tue Aug  4 03:56:51 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tcltklib.c: fix trouble on old-style C function
+	  declarations [ruby-core:22871].
+
+	* ext/tk/lib/tcltklib.c: (ruby_1_8) fix warning about RUBY_RELEASE_DATE
+
+	* ext/tk/lib/tk/multi-tk.rb: kill zombie threads.
+
+	* ext/tk/lib/tk/fontchooser.rb: fix typo and support OptionObj.
+
+	* ext/tk/lib/tk/{canvas.rb,virtevent.rb,image.rb,timer.rb}:
+	  don't create unnecessary array.
+
+Mon Aug  3 22:19:24 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval.c (rb_mod_include): fix document.  [ruby-core:24675]
+
+Mon Aug  3 18:25:08 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/svr.rb: obsolete TCPserver renamed.  [ruby-core:24712]
+
+	* sample/tsvr.rb: ditto.
+
+	* sample/dualstack-httpd.rb: ditto.
+
+Mon Aug  3 18:12:54 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (recursive_push): untrust internal hash to prevent
+	  unexpected SecurityError.  a patch from Kazuhiro NISHIYAMA.
+	  Fix: #1864  [ruby-dev:38982]
+
+Mon Aug  3 17:06:05 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/README: updated.  a patch from Daniel Bovensiepen.
+	  [ruby-core:24693]
+
+Mon Aug  3 16:28:09 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_connect): return value was broken when some
+	  error occurred.
+	  [ruby-core:24234]
+
+Mon Aug  3 15:56:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (permute0): use chars for boolean array.
+
+	* array.c (rb_ary_{permutation,combination}): disallow reentrance
+	  with continuation since work-buffers cannot restore.
+
+	* array.c (rb_ary_{permutation,combination,product}): must not use
+	  ary_discard on strings.
+
+Mon Aug  3 06:43:25 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_hash): normalize -0.0 to 0.0.  [ruby-core:24577]
+
+Mon Aug  3 00:32:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_random_int): arguments have to be converted to
+	  integer.  [ruby-core:24679]
+
+Sun Aug  2 21:04:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (literal_concat0): tail can be nil.  [ruby-dev:38980]
+
+Sun Aug  2 20:09:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_combination, rb_ary_product): prevent from GC.
+
+Sun Aug  2 16:53:19 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rand.rb: add tests for Random#float's rejection
+	  against Infinity and NaN.
+
+Sun Aug  2 14:20:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rand_int): prevent from GC.
+
+Sat Aug  1 19:23:27 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_trans): change condition of singlebyte
+	  optimization.
+
+Sat Aug  1 18:50:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_float): rejects Infinity and NaN.
+	  [ruby-core:24651]
+
+Sat Aug  1 18:34:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (gem): suppressed warnings.
+	  cf: [ruby-dev:38975]
+
+Sat Aug  1 05:18:36 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_trans): can't use singlebyte optimization when
+	  the replacement is multibyte. [ruby-core:24612]
+
+Fri Jul 31 18:01:34 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/securerandom.rb (SecureRandom.random_bytes): return string should
+	  be ASCII-8BIT.  [ruby-core:24640]
+
+Fri Jul 31 16:28:33 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_stdio_set_default_encoding): added.
+
+	* ruby.c (process_options): call rb_stdio_set_default_encoding
+	  after setting default internal and external.
+
+Fri Jul 31 15:06:33 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/compile_prelude.rb: too long string literal causes compile error
+	  on some platforms.
+
+Fri Jul 31 13:15:27 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* transcode.c (str_encode_bang): C99ism.
+
+Fri Jul 31 11:48:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): used more appropriate construct.
+
+Fri Jul 31 10:54:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (literal_concat_gen): reduced unnecessary node at string
+	  literal concatenation with empty head dstr.  [ruby-dev:38968]
+
+Fri Jul 31 02:57:39 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (literal_concat_gen): NODE_DSTR was incorrectly handled as
+	  NODE_STR.  [ruby-dev:38968]
+
+	* bootstraptest/test_syntax.rb: add a test for above.
+
+Fri Jul 31 00:55:48 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_module.rb (test_ancestors, test_included_modules):
+	  ignore rake mixins.
+
+Fri Jul 31 00:30:54 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_insnhelper.c (vm_call_cfunc): let set_trace_func use called_id
+	  instead of original_id.
+
+Thu Jul 30 23:04:32 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* gem_prelude.rb (Gem.path): uses Gem.default_path as a default value
+	  so that ruby finds gems in ~/.gem/.
+	  (Gem.user_home): reduced version of lib/rubygems.rb's.
+	  Gem.default_path needs it.
+
+Thu Jul 30 22:28:04 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/compile_prelude.rb: replaces "require" with in-place evaluation
+	  so that copy & paste for lib/rubygems/default.rb is not necessary.
+
+	* gem_prelude.rb: removes copied codes from lib/rubygems/defaults.rb.
+	  uses require instead.
+
+	* common.mk (prelude.c): adds dependency for lib/rubygems/defaults.rb.
+
+Thu Jul 30 21:56:18 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rand.rb: add tests for Random class.
+
+Thu Jul 30 21:48:56 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* random.c (init_genrand): ensure invariant of mt->next and mt->left.
+	  mt->next should always equal mt->state + N + 1 - mt->left.
+	  In fact, 'r = Random.new(0); r == r.dup' has returned false.
+
+Thu Jul 30 21:43:41 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* random.c (random_bytes): use NUM2LONG instead of FIX2LONG because
+	  Random#bytes may receive bignum.
+
+Thu Jul 30 21:39:42 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_compile_each): fix stack consistency error.
+	  [ruby-core:24611]
+
+	* bootstraptest/test_method.rb: add tests for above.
+
+Thu Jul 30 18:39:39 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c: added check for frozen string for encode! (see Bug #1836)
+
+	* test/ruby/test_transcode.rb: added tests for the above
+
+Thu Jul 30 16:45:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defineclass): preserve encoding of class/module
+	  names.  [ruby-core:24600]
+
+	* variable.c (rb_set_class_path_string): set class path with a
+	  string value.
+
+Thu Jul 30 16:12:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (Init_var_tables): initializes __classid__ ID.
+
+	* variable.c: use st_data_t for st functions.
+
+Thu Jul 29 14:25:14 2009  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/util.rb (CGI::unescape): support encoding option.
+
+	* lib/cgi/cookie.rb (CGI::Cookie.parse): fix for the encoded value.
+
+Wed Jul 29 08:08:07 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (regexp): regexp literal at the top of dstr is still needed
+	  even if it is empty.
+
+Wed Jul 29 03:36:24 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/json/lib/json/common.rb (JSON#recurse_proc): remove
+	  unnecessary private specifier.  [ruby-dev:38929]
+
+Wed Jul 29 03:34:46 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, vm_insnhelper.c (vm_call_method): revive
+	  VM_CALL_OPT_SEND_BIT and use it to recognize "send" method.
+
+Wed Jul 29 03:11:59 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_method_entry_arity): support not_implemented method.
+	  (I have no idea to test it)
+
+Tue Jul 28 19:36:26 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_method_entry_arity): support optimized method (send).
+
+	* test/ruby/test_method.rb: add a test for above.
+
+Tue Jul 28 04:34:05 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/extconf.rb: bug fix and ignore invalid Tcl/Tk libraries.
+
+	* ext/tk/lib/config_list.in: bug fix and add a new option.
+
+	* ext/tk/lib/README.tcltklib: update for a new option.
+
+Mon Jul 27 19:03:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_*str_new_cstr, rb_str_buf_new_cstr),
+	  (rb_str_buf_cat2, rb_str_cat2, rb_exc_new2): suppress warnings.
+
+Mon Jul 27 10:24:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parser.rb (RDoc::Parser.binary?): fix for empty files.
+	  [ruby-dev:38848]
+
+Mon Jul 27 06:51:41 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_eof): should call next_argv() before testing.
+	  [ruby-core:24561]
+
+Sun Jul 26 19:17:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_eof): go to the next file if called after ARGF.close
+	  or ARGF.skip.  a patch from Mike Kasick at [ruby-core:24561].
+
+Sun Jul 26 18:30:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): __send__ can call protected
+	  methods.  [ruby-core:24500]
+
+Sun Jul 26 01:09:14 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* ext/bigdecimal.c: moved BASE_FIG definition before it is used
+	  first time.
+
+	* include/ruby/defines.h [__SYMBIAN32__]: defined TRUE and FALSE to match
+	  changes in bignum.c and array.c.
+
+Sat Jul 25 17:49:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_eof): should not have reached EOF before trying to
+	  read.  based on a patch by Heesob Park <phasis AT gmail.com> in
+	  [ruby-core:24559].  [ruby-core:24557]
+
+Sat Jul 25 13:44:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (nucomp_hash), rational.c (nurat_hash): not to use
+	  hash value of class so that equality against subclasses can
+	  work.  [ruby-dev:38850]
+
+Sat Jul 25 01:05:59 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/big5.c: Fix EncLen_BIG5 for Big5-HKSCS. see [ruby-core:24390]
+
+Fri Jul 24 19:19:19 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/big5.trans, big5-hkscs-tbl.rb:
+	  new Chinese BIG5-HKSCS transcoding (with Tatsuya Mizuno)
+	  see [ruby-core:24390]
+
+	* test/ruby/test_transcode.rb: added tests for the above
+	  (with Tatsuya Mizuno)
+
+	* enc/big5.c: Added BIG5-HKSCS as a replicate encoding of BIG5
+	  (short term solution, needs more work; with Tatsuya Mizuno)
+
+	* tool/transcode-tblgen.rb: made 'pat' directly accessible in
+	  class StrSet
+
+Fri Jul 24 18:03:01 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (idle): leaves IDLE in a ensure clause.  a patch
+	  from Eric Hodel.
+
+	* test/net/imap/test_imap.rb (setup, teardown): turn on
+	  Socket.do_not_reverse_lookup in tests.
+
+Fri Jul 24 00:13:41 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (resp_text_code): accepts response codes without
+	  text.  [ruby-core:24194]
+
+	* lib/net/imap.rb (idle, idle_done): new methods for the IMAP4 IDLE
+	  command (RFC 2177).  Thanks, Eric Hodel.
+
+	* lib/net/imap.rb (format_date, format_datetime): new method to
+	  format IMAP-style date/datetime. Thanks, Eric Hodel.
+
+Thu Jul 23 17:31:02 2009  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb: make irb be able to parse
+	  string_dvar. [ruby-core: 24051]
+
+Thu Jul 23 17:26:51 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: should not create "config_list" in a $srcdir.
+
+Thu Jul 23 14:35:02 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_{open,wopen}): fixed typos. these conditions
+	  mean to call runtime's open() if textmode.
+
+Thu Jul 23 08:53:24 2009  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup/simple_markup/: Remove useless directory.
+	  [Bug #1343]
+
+Thu Jul 23 08:52:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_{state,left}): internal/debug use.
+
+Wed Jul 22 19:29:26 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_module.rb (TestModule#test_alias): warning
+	  message updated.
+
+Thu Jul 23 00:39:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (class.o): depends on vm_core.h.
+
+Thu Jul 23 00:10:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (init_leap_second_info): checks the result of gmtime to
+	  suppress warnings.
+
+Wed Jul 22 22:23:24 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_core.h (struct rb_iseq_t): add a new field line_no.  This field
+	  represents line number from which the original code of the iseq
+	  starts.  [ruby-dev:38698]
+
+	* iseq.c, compile.c: ditto.
+
+	* parse.y: line number hack (for Proc#source_location) is no longer
+	  needed.
+
+	* test/ruby/test_settracefunc.rb: line number of set_trace_func is now
+	  compatible with 1.8's.
+
+Wed Jul 22 22:16:48 2009  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* method.h: Commas at end of enum list not allowed as of C89
+
+	* vm_method.c (rb_add_method): avoid C++ comment
+
+	* vm_insnhelper.c (vm_call_cfunc): ditto.
+
+Wed Jul 22 20:42:52 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: define IPPROTO_IPV6 macro for recent Windows
+	  SDK.
+
+Wed Jul 22 19:32:10 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/mkconstants.rb: define IPV6_* constants only when INET6
+	  is defined.
+
+Wed Jul 22 19:23:04 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: ipv6 support is disabled by default on mswin.
+
+Wed Jul 22 17:41:08 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): drop IPv6 addresses
+	  if INET6 is not defined.
+
+Wed Jul 22 17:29:59 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: enable ipv6 support for win32.
+
+Wed Jul 22 16:38:39 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* hash.c (Hash::[]): rdoc. patch by Marc-Andre Lafortune.
+	  #1385.
+
+Wed Jul 22 10:00:53 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (WSAMSG): get rid of compile error on VC9 and mingw.
+
+Wed Jul 22 06:35:56 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_enum.rb (TestEnumerable#each): recursive join now
+	  raises ArgumentError.
+
+Wed Jul 22 02:33:57 2009  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb.rb: forget svn commit.
+
+Wed Jul 22 01:20:54 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bootstraptest/test_io.rb: block write may block for long time.
+
+Wed Jul 22 00:34:39 2009  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb.rb, lib/irb/init.rb, lib/irb/ext/save-history.rb: add
+	  IRB::irb_at_exit. no use finalizer saving history. [ruby-dev-38563]
+
+Tue Jul 21 23:47:38 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* random.c (rand_init): array length of random seed was broken, which
+	  causes memory error with srand(2**1000000-1).
+
+	* test/ruby/test_rand.c: test for above.
+
+Tue Jul 21 21:37:19 2009  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/cmd/help.rb: fixed irb's "help" command. [ruby-core:22310].
+
+	* lib/tracer.rb: no show lines unknown line number. [ruby-core:22096],
+	  no trace display  c-call and c-return as default.
+
+Tue Jul 21 16:24:41 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (vtdate2rbtime): VT_DATE variant object
+	  is converted to Time object now.
+
+	* test/win32ole/test_win32ole_variant.rb (test_s_new_with_nil,
+	  test_conversion_time2date, test_conversion_str2date,
+	  test_conversion_vt_date, test_set_value): ditto.
+
+	* ext/win32ole/win32ole.c (rbtime2vtdate): refactoring.
+
+Tue Jul 21 16:07:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tracer.rb: toplevel caller exists now.  [ruby-core:24454]
+
+Tue Jul 21 13:36:20 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.[ch] (recvmsg, sendmsg): new functions to support recvmsg/
+	  sendmsg like UNIX. these functions are experimental and not tested
+	  well. bug reports are welcome.
+
+Tue Jul 21 13:35:21 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (TEST_RUNNABLE): follow r24209.
+
+Tue Jul 21 12:45:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_enable, rb_gc_disable): should return Qtrue/Qfalse.
+
+Tue Jul 21 12:38:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_generic_ivar_memsize): should not remove generic
+	  instance variable table.
+
+Mon Jul 20 20:35:55 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_f_open): add rdoc about specifying ext_enc as *-bom.
+
+Mon Jul 20 19:00:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_dstr_fragments): reduced needless literal.
+
+	* parse.y (xstring, regexp, dsym, literal_concat, evstr2dstr):
+	  literal at the top of dstr is no longer needed if it is empty,
+	  since concatstrings and toregexp always create new strings.
+
+Mon Jul 20 12:51:39 2009  wanabe  <s.wanabe@gmail.com>
+
+	* lib/matrix.rb (Matrix#rank): revert a part of r20859 to avoid
+	  infinite loop.  [Bug #1020]
+
+Mon Jul 20 11:12:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, Makefile.in (TEST_RUNNABLE): tests are no
+	  runnable if cross-compiling.
+
+	* common.mk (btest*, test*): ditto.
+
+Mon Jul 20 10:55:18 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Addrinfo#family_addrinfo): fix a typo in
+	  error message.  patch by Nobuhiro IMAI.  [ruby-dev:38828]
+
+Sun Jul 19 22:22:54 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (float_to_r): an improvement.
+
+Sun Jul 19 20:41:24 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (make_patterns): do not use \d.
+
+	* rational.c (make_patterns): ditto.
+
+Sun Jul 19 17:32:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_read): should taint the result.  [ruby-dev:38826]
+
+Sun Jul 19 11:00:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (me_opts): fixed optimized method aliasing.
+	  [ruby-dev:38824]
+
+Sun Jul 19 10:54:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.accept_loop): rescue
+	  IO::WaitReadable instead of Errno::EWOULDBLOCK.
+	  (Socket.udp_server_loop_on): ditto.
+
+Sat Jul 18 23:44:59 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (reg_enc_error): use rb_enc_get to get the encoding of
+	  a Regexp object. REGEXP(re)->ptr->enc is the encoding of the
+	  regexp engine for patterns and target strings.
+	  [ruby-core:23208]
+
+Sat Jul 18 17:43:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_hash_cmp): got rid of overflow.
+
+Sat Jul 18 16:03:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/defines.h (TRUE, FALSE): for internal use.
+
+Sat Jul 18 11:39:49 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* random.c: workaround for VC++ 6.0.
+
+Sat Jul 18 09:16:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_random_{int32,real,bytes,int}): added functions for
+	  extension libraries.
+
+Sat Jul 18 09:07:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (fill_random_seed): use cryptographic service on Windows.
+
+Sat Jul 18 07:56:00 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fail to create a widget object for an unknown
+	  widget path.
+
+Sat Jul 18 07:06:31 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb,ext/tk/lib/tk/grid.rb: Bug fix on grid_slaves().
+	  Extend usage pattern of grid_column()/grid_row().
+
+Sat Jul 18 06:30:35 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h: include winsock headers in extern "C++" for
+	  C++ extension libraries.
+
+	* include/ruby/missing.h (vsnprintf): workaround for VC++.
+	  [ruby-core:23096]
+
+Sat Jul 18 00:23:47 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ext/readline/readline.c (Init_readline): rl_catch_signals does
+	  not exist on some platform like Mac OS X 10.5.
+
+	* ext/readline/extconf.rb: checks existence of rl_catch_signals.
+
+Fri Jul 17 22:37:22 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (socket_s_ip_address_list): drop inactive
+	  adapters.
+
+	* test/socket/test_{nonblock,addrinfo,socket}.rb: skip some tests on
+	  Windows.
+	  [ruby-core:23051]
+
+Fri Jul 17 22:29:21 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/readline/readline.c (Init_readline): use rl_catch_sigwinch only
+	  when existing.
+
+	* ext/readline/extconf.rb: check existence of rl_catch_sigwinch, and
+	  workaround for native Win32 readline port.
+
+Fri Jul 17 18:18:23 2009  Tanaka Akira  <akr@fsij.org>
+
+	* range.c (recursive_hash): extracted from range_hash.  reject
+	  recursive key.
+	  (range_hash): use recursive_hash.
+
+Fri Jul 17 18:11:32 2009  Tanaka Akira  <akr@fsij.org>
+
+	* struct.c (recursive_hash): extracted from rb_struct_hash.  reject
+	  recursive key.
+	  (rb_struct_hash): use recursive_hash.
+
+Fri Jul 17 16:45:22 2009  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (recursive_hash): reject recursive key.
+
+Fri Jul 17 15:20:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (DIGSPERINT): fix for LP64.
+
+	* random.c (fill_random_seed): /dev/urandom is not available on
+	  DOSISH systems.
+
+	* random.c (random_equal): new method Random#==.
+
+Fri Jul 17 13:35:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (rsock_getfamily): return AF_UNSPEC if getsockname
+	  is failed.  [ruby-core:24383]
+
+Fri Jul 17 01:22:57 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (recursive_join): raise ArgumentError for joining
+	  recursive array.
+
+	* array.c (ary_join_1): ditto.
+
+	* test/ruby/test_array.rb (TestArray#test_join2): test updated for
+	  recursive join.
+
+Thu Jul 16 23:32:16 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb,ext/tk/config_list.in: ignore paths which includes
+	  white space characters on Windows.[ruby-dev:38794]
+
+	* ext/tk/lib/tk.rb: works on Cygwin (limitation:: Tk.mainloop works on
+	  the main thread only).
+
+Thu Jul 16 20:58:18 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole.rb (test_s_codepage_changed,
+	  test_s_locale_set, test_s_locale_change): skip test if Japanese
+	  locale is not installed. [ruby-core:23806]
+
+	* test/win32ole/test_win32ole_variant.rb (test_conversion_str2cy):
+	  ditto.
+
+Thu Jul 16 19:02:28 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/lib/socket.rb (Socket.tcp_server_sockets_port0): servers
+	  is nil if an error occurs before setting it.
+
+Thu Jul 16 18:42:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (fill_random_seed): suppress a warning.
+
+Thu Jul 16 18:30:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_aset, recursive_hash): rejects recursive hash.
+	  [ruby-core:22921]
+
+Thu Jul 16 18:28:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* id.c (Init_id), vm.c (vm_exec): @#__ThrowState__ is no longer
+	  used.  [ruby-dev:38760]
+
+Thu Jul 16 17:41:28 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_alias): fix a case which try non-existing method alias.
+
+	* test/ruby/test_alias.rb: add a test.
+
+Thu Jul 16 16:00:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_reopen): discards read buffer.  [ruby-core:24240]
+
+Thu Jul 16 15:52:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_new, rb_bigzero_p), range.c (rb_range_values):
+	  added for random.c.
+
+	* random.c (rb_random_t): objectified.  [EXPERIMENTAL]
+	  [ruby-dev:30954]
+
+Thu Jul 16 14:08:44 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_sample): RDoc update.  a patch from Florian
+	  Frank.   [ruby-core:24347]
+
+Thu Jul 16 12:42:10 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_readline): rl_free_line_state
+	  before rl_cleanup_after_signal.
+
+	* ext/readline/readline.c (Init_readline): set rl_catch_signals and
+	  rl_catch_sigwinch as 0. [ruby-core:21884]
+
+Thu Jul 16 08:59:22 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* node.h, vm_core.h, variable.c: rename global_entry to rb_global_entry.
+
+	* compile.c, insns.def, iseq.c, vm_insnhelper.h: ditto.
+
+Thu Jul 16 08:57:44 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c: separate clearing cache entry code.
+
+Thu Jul 16 08:56:32 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* dir.c (push_glob): re-fix GC problem.
+
+Thu Jul 16 08:55:27 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: add method.h.
+
+Thu Jul 16 08:53:26 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* io.c (argf_free): free data body.
+
+Wed Jul 15 23:46:55 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h, vm_core.h: add rb_method_entry_t.  Remove nodes around
+	  method management.  This change affect some VM control stack structure.
+
+	* vm.c, vm_insnhelper.c, vm_method.c, vm_eval.c: ditto.  and make some
+	  refactoring.
+
+	* insns.def, class.c, eval.c, proc.c, vm_dump.c : ditto.
+
+	* vm_core.h, compile.c (iseq_specialized_instruction): remove
+	  VM_CALL_SEND_BIT.  use another optimization tech for Kernel#send.
+
+	* node.h: remove unused node types.
+
+	* ext/objspace/objspace.c (count_nodes): ditto.
+
+	* gc.c: add mark/free functions for method entry.
+
+	* include/ruby/intern.h: remove decl of
+	  rb_define_notimplement_method_id().  nobody can use it
+	  because noex is not opened.
+
+	* iseq.c (iseq_mark): fix to check ic_method is available.
+
+	* iseq.c (rb_iseq_disasm): fix to use rb_method_get_iseq().
+
+Wed Jul 15 23:45:11 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* dir.c (push_glob): fix GC problem.
+
+Wed Jul 15 17:33:52 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ext/purelib.rb: translates a fake path to rubygems in $" into
+	  an alternative in $: so that Kernel.#require does not load
+	  more rubygems.rb.
+	  Resolves many failures in test/rubygems/*.
+
+	* gem_prelude.rb (Gem.load_full_rubygems_library): supports case
+	  the rubygems to load is not in $(rubylibprefix).
+	  (Gem.path_to_full_rubygems_library): new method for the changes in
+	  purelib.rb and Gem.load_full_rubygems_library.
+	  (Gem.fake_rubygems_as_loaded): new method.
+
+Wed Jul 15 16:29:35 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (LIBPATHFLAG): path is already quoted in mkmf.rb.
+
+Wed Jul 15 06:21:50 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: --with-{tcl,tk}-dir doesn't work.[ruby-dev:38782]
+
+Wed Jul 15 04:22:54 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/cgi/test_cgi_multipart.rb (CGIMultipartTest#_prepare):
+	  set tempfile name with literal.
+
+Tue Jul 14 21:53:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/nonblock: moved from ext/io/wait/lib.
+
+Tue Jul 14 17:29:20 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_index_m): return nil if pos is out of string.
+	  [ruby-core:23660]
+
+Tue Jul 14 16:13:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_initialize): check if the descriptor can be accessed
+	  in the specified open mode.  [ruby-dev:38571]
+
+Tue Jul 14 09:26:14 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: Long-term-callback support isn't stable yet.
+	  So, disable the feature and waiting for improvement in the future.
+
+Tue Jul 14 01:28:17 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_m17n.rb (TestM17N#test_env): the encoding of
+	  ENV is now locale encoding.
+
+Tue Jul 14 01:24:56 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* marshal.c (r_object0): should return real object.
+
+Tue Jul 14 01:06:31 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enumerator.c (yielder_yield_i): use rb_proc_new instead of
+	  rb_iterate.  [ruby-dev:38518]
+
+	* README.EXT: rb_iterate is obsolete since 1.9; use rb_block_call
+	  instead.
+
+	* README.EXT.ja: ditto.
+
+Tue Jul 14 00:45:41 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_case.rb (TestCase#test_deoptimization):
+	  test for [ruby-core:23190].
+
+Mon Jul 13 22:49:50 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/prime.rb (Prime#prime_division): now decomposes
+	  negative integer into a decomposition with element [-1, 1].
+
+	* test/test_prime.rb: test for it.
+
+Mon Jul 13 22:28:03 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (pipe_open): handles leaked on win32 when an error occurs.
+
+Mon Jul 13 20:21:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): should not shadow outer variable not to
+	  return Qnil always.
+
+Mon Jul 13 19:41:05 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* marshal.c (r_object0): copy instance variables to new regexp.
+
+Mon Jul 13 17:49:11 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, compile.c: declare struct iseq_inline_cache_entry.
+	  Inline cache (IC) entries are no longer GC managed object.
+	  IC entries are freed when ISeq is freed.
+
+	* iseq.c: fix mark, free, memsize functions for above change.
+
+	* insns.def: remove rb_gc_write_barrier().
+
+	* vm_insnhelper.c (vm_method_search): ditto.
+
+	* tool/instruction.rb, template/insns_info.inc.tmpl (insn_iclen):
+	  added.
+
+Mon Jul 13 13:35:08 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm_insnhelper.c (getinstancevariable):
+	  fix to use inline cache.
+
+	* compile.c: fix to skip inline cache entry (IC).  IC is added
+	  automatically by compiler.
+
+	* insns.def, vm_insnhelper.h: fix IC positions.
+
+	* iseq.c: increment minor_version of ISeq because of above change.
+
+Mon Jul 13 08:01:00 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: New strategy for searching Tcl/Tk libraries.
+
+	* ext/tk/*: Support new features of Tcl/Tk8.6b1 and minor bug fixes.
+	     ( [KNOWN BUG] Ruby/Tk on Ruby 1.9 will not work on Cygwin. )
+
+	* ext/tk/*: Unify sources between Ruby 1.8 & 1.9.
+	            Improve default_widget_set handling.
+
+	* ext/tk/*: Multi-TkInterpreter (multi-tk.rb) works on Ruby 1.8 & 1.9.
+	     ( [KNOWN BUG] On Ruby 1.8, join to a long term Thread on Tk
+	        callbacks may freeze. On Ruby 1.9, cannot create a second
+	        master interpreter (creating slaves are OK); supported master
+	        interpreter is the default master interpreter only. )
+
+	* ext/tk/lib/tkextlib/*: Update supported versions of Tk extensions.
+	         Tcllib 1.8/Tklib 0.4.1  ==>  Tcllib 1.11.1/Tklib 0.5
+	         BWidgets 1.7            ==>  BWidgets 1.8
+	         TkTable 2.9             ==>  TkTable 2.10
+	         TkTreeCtrl 2005-12-02   ==>  TkTreeCtrl 2.2.9
+	         Tile 0.8.0/8.5.1        ==>  Tile 0.8.3/8.6b1
+	         IncrTcl 2005-02-14      ==>  IncrTcl 2008-12-15
+	         TclX 2005-02-07         ==>  TclX 2008-12-15
+	         Trofs 0.4.3             ==>  Trofs 0.4.4
+
+Mon Jul 13 01:18:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_timespec): rounds subsecond toward zero.
+
+Sun Jul 12 23:51:39 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* hash.c (env_str_new): use rb_locale_str_new instead of
+	  rb_tainted_str_new. rb_locale_str_new set string locale
+	  encoding and tainted.
+
+Sun Jul 12 23:56:40 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: added some shortcuts.
+
+	* rational.c: ditto.
+
+Sun Jul 12 23:30:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_to_integer, rb_check_to_integer): return Bignum
+	  as-is.
+
+Sun Jul 12 21:07:46 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: use k_exact_{zero,one}_p macro.
+
+	* rational.c: ditto.
+
+Sun Jul 12 20:42:58 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c (fix_divide): added an entry to rational.
+
+	* rational.c (rb_rational_reciprocal): added.
+
+	* complex.c (f_reciprocal): added.
+
+Sun Jul 12 02:24:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rand_init): use fixed buffer for small numbers.
+
+Sat Jul 11 14:43:34 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_io_m17n.rb (test_strip_bom): added.
+
+Sat Jul 11 07:11:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  array length is long.
+
+	* ext/readline/readline.c (readline_s_get_filename_quote_characters):
+	  missing type of self.
+
+Sat Jul 11 02:37:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (limited_rand): expands to long before shift so that
+	  the result does not overflow.
+
+Sat Jul 11 00:16:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rand_init): got rid of buffer overflow.
+
+Sat Jul 11 00:11:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_join_1): should recurse for element array.
+
+Fri Jul 10 23:10:11 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_strip_bom): ungetbyte third byte when UTF-16LE.
+
+Fri Jul 10 23:04:16 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_strip_bom): Fix condition of second byte of
+	  UTF-16LE/UTF-32LE.
+
+Fri Jul 10 21:45:30 2009  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/extconf.rb: checked rl_refresh_line in readline.
+
+	* ext/readline/readline.c (readline_s_refresh_line): add new
+	  method, a patch from Koichiro Ohba. see [ruby-list:45922].
+
+Fri Jul 10 21:00:05 2009  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/extconf.rb: checked rl_line_buffer and rl_point in
+	  readline.
+
+	* ext/readline/readline.c (readline_s_get_line_buffer): new method.
+	  (readline_s_get_point): new method.
+
+Fri Jul 10 16:30:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (recursive_join): use obj to tell if recursion occurs.
+	  [ruby-core:24150]
+
+	* enum.c (enum_join): reverted r23966.  [ruby-core:24196]
+
+Fri Jul 10 14:41:34 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* marshal.c (r_object0): set encoding only if the encoding
+	  is not US-ASCII.
+
+Fri Jul 10 14:44:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (struct MT): ruby already assumes int has 32bit a
+	  least, so no needs to use long.
+
+	* random.c (rand_init): copies data to be used really only.
+
+Fri Jul 10 11:41:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_obj_{untrust,untrusted,trust}): added
+	  prototypes.  [ruby-dev:38756]
+
+Fri Jul 10 10:25:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_superclass): checks for implicit
+	  argument passing before method search.  [ruby-core:24244]
+
+Fri Jul 10 07:22:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_data_type): typed.
+
+Thu Jul  9 23:28:48 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/io.h (FMODE_STRIP_BOM): new constant.
+
+	* io.c (io_encname_bom_p): judge whether the encoding name
+	  is *-bom or not.
+	  (parse_mode_enc): drop "-bom".
+	  (rb_io_modestr_fmode): set FMODE_STRIP_BOM if needed.
+	  (rb_io_extract_modeenc): ditto.
+	  (io_strip_bom): strip bom if exists.
+	  (io_set_encoding_by_bom): set encoding if there is bom.
+	  Set encoding and strip bom when modeenc string is "r:foo-bom"
+	  [ruby-dev:37236]
+
+Thu Jul  9 21:56:59 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* marshal.c (r_object0): replace \u by u when the regexp is
+	  made by Ruby 1.8. [ruby-dev:36750]
+
+Wed Jul  8 23:13:54 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_div): omitted zero division check.
+
+Wed Jul  8 21:00:37 2009  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/inspector.rb: forget svn add.
+
+Wed Jul  8 19:10:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_check_type): rejects typed data.
+
+Wed Jul  8 18:28:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c ({proc,binding,method}_data_type): typed.
+
+Wed Jul  8 16:47:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_basename): returns new string instead of
+	  shared string from FilePathStringValue().  [ruby-core:24199]
+
+Wed Jul  8 04:28:16 2009  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/.document:  Update with extensions that appear to have
+	  documentation.  [ruby-core:24181]
+
+Wed Jul  8 04:28:16 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h:
+	  rename
+	    "...TypeStruct" and "typed_struct" to
+	    "TypedData..." and "typeddata", respectively.
+	  rename
+	    rb_data_type_t#name to
+	    rb_data_type_t#wrap_struct_name.
+
+	* error.c, gc.c, iseq.c, vm.c: ditto.
+
+Tue Jul  7 20:23:27 2009  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* bin/irb, lib/irb.rb lib/irb/*: irb-0.9.6, extend inspect-mode
+
+Tue Jul  7 17:00:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_call_finalizer_at_exit): deal with typed struct like
+	  as obj_free().
+
+	* error.c (rb_typed_struct_is_kind_of): new function to see if the
+	  given typed struct.
+
+	* error.c (rb_check_typed_struct): new function to check typed
+	  struct.
+
+	* include/ruby/ruby.h (Check_TypedStruct, Data_Get_TypedStruct):
+	  new macro to check typed struct.
+
+Tue Jul  7 13:36:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (DEFINE_ENUMFUNCS): included function signature.
+
+	* enum.c (rb_enum_join): non-nil separator must be convertible to
+	  String.  [ruby-core:24172]
+
+Tue Jul  7 12:47:28 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (rb_enum_join): should propagate taint to the return
+	  value.  the change was overridden by r23967.  [ruby-core:24176]
+
+Tue Jul  7 11:56:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (GC_PROF_SET_MALLOC_INFO, GC_PROF_SET_HEAP_INFO): simplified.
+
+Tue Jul  7 10:12:37 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_ext_int_to_encs): Set external encoding if
+	  ext==intern. [ruby-dev:38278]
+
+Mon Jul  6 09:31:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (make_curry_proc): should propagate lambda-ness.
+	  [ruby-core:24127]
+
+	* proc.c (proc_hash): use long.
+
+Mon Jul  6 09:06:49 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_module.rb (TestModule#test_undef): adjust to
+	  message change.
+
+	* test/ruby/test_object.rb (TestObject#test_redefine_method_which_may_case_serious_problem): ditto.
+
+	* test/ruby/test_object.rb (TestObject#test_remove_method): ditto.
+
+Mon Jul  6 09:04:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_join): deals with self recursive objects to get rid
+	  of infinite recursion.  [ruby-core:24150]
+
+Mon Jul  6 08:00:10 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_to_a): should propagate taint to the return value.
+	  [ruby-core:24152]
+
+	* enum.c (enum_sort_by): ditto.
+
+Mon Jul  6 00:41:41 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: edited rdoc.
+
+Sun Jul  5 23:55:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): fix Time.local(2009,2,31) failure on 64bit
+	  time_t environment.
+
+Sun Jul  5 22:43:13 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_s_polar): now arg is optional.
+
+Sun Jul  5 20:40:35 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (float_arg): returns PI for -0.0.
+
+Sun Jul  5 14:04:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_threadptr_exec_event_hooks): new function to
+	  execute event hooks, with preserving errinfo.  [ruby-core:24118]
+
+Sun Jul  5 08:14:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_add_method, remove_method, rb_undef): fixed
+	  minor grammatical errors in warnings.  a patch from Run Pain
+	  Run Run at [ruby-core:24141].
+
+	* vm_method.c (Init_eval_method): registers notimplement_body as a
+	  mark-object.
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): restores the firs
+	  arg where is overwritten at funcall.  [ruby-core:24139]
+
+Sat Jul  4 08:20:03 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (dbl2ival): should raise FloatDomainError on Infinity
+	  and NaN as 1.8 does.  [ruby-dev:38726]
+
+Fri Jul  3 22:48:45 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): less number of guesses for hh:mm:60.
+
+Fri Jul  3 21:30:14 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_equal_p): removed.
+
+Fri Jul  3 21:07:29 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: renamed equal_p to eqeq_p.
+
+	* complex.c: ditto.
+
+	* complex.c (nucomp_equal_p): added.
+	  Complex(NaN).equal?(Complex(NaN)) should return true.
+
+Fri Jul  3 19:48:40 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: undef-ed some methods.  [ruby-core:24110]
+
+	* complex.c (Numeric#arg): NaN for NaN.  [ruby-core:24116]
+
+Fri Jul  3 18:35:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (struct rb_iseq_struct): fixed types.
+
+	* vm_core.h (ic_vmstat): VM state version is VALUE.
+
+Fri Jul  3 02:52:20 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): time guess strategy refined again.
+
+Fri Jul  3 00:36:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (find_time_t): time guess strategy refined.
+
+Thu Jul  2 11:16:25 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: added response to Net::IMAP::ResponseError.
+	  a patch from Eric Hodel in [ruby-core:24111].
+
+Thu Jul  2 08:04:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (num_exact): rb_check_to_integer() can deal with both of
+	  Fixnum and Bignum together.
+
+Thu Jul  2 07:53:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): fixed wrong variable.
+
+Thu Jul  2 05:37:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (num_exact): use to_r for T_FLOAT.
+
+Thu Jul  2 05:15:54 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (quo): return an integer if possible.
+
+Wed Jul  1 21:09:25 2009  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_time_num_new): declared.
+
+	* time.c (nsec2timev): extracted from time_new_internal.
+	  (time_new_internal): change argument to VALUE.
+	  (rb_time_new): follow the argument change.
+	  (rb_time_nano_new): ditto.
+	  (rb_time_num_new): new function.
+
+	* ext/socket/ancdata.c (ancillary_timestamp): use rb_time_num_new to
+	  represent struct bintime preciously.
+
+Wed Jul  1 08:46:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_encoding): encodings need extra depth.
+	  [ruby-core:24100]
+
+Wed Jul  1 06:47:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_grep): gets rid of type-punning calls.
+
+Wed Jul  1 06:36:28 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_join): add Enumerable#join.
+
+	* array.c (ary_join_1): recursive join for Enumerators (and
+	  objects with #to_a).
+
+	* array.c (rb_ary_join): performance tune.
+
+Tue Jun 30 18:19:07 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_hash): documentation fix.  a patch from
+	  Marc-Andre Lafortune.  [ruby-core:23943]
+
+	* object.c (rb_mod_cmp): ditto.
+
+	* range.c (range_eq): ditto.
+
+	* string.c (rb_str_partition, rb_str_rpartition): ditto.
+
+	* struct.c (rb_struct_s_def): ditto.
+
+Tue Jun 30 17:44:24 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (reg_match_pos): adjust offset based on characters, not
+	  bytes.  [ruby-dev:38722]
+
+	* string.c (rb_str_offset): new function.
+
+	* string.c (rb_str_index_m): no call to rb_reg_adjust_startpos().
+
+Tue Jun 30 16:57:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb: renamed to get rid of collision against
+	  instruction.rb on command line completion of shell.
+
+	* tool/mkconfig.rb (RbConfig.expand): get rid of exceptions on
+	  frozen strings unless really changed.
+
+	* tool/file2lastrev.rb: get rid of global variables.
+
+	* tool/compile_prelude.rb: use US-ASCII name.
+
+Tue Jun 30 16:46:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def, tool/instruction.rb: fixed types.
+
+Tue Jun 30 11:08:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/oniguruma.h, include/ruby/re.h, re.c, regcomp.c,
+	  regenc.c, regerror.c, regexec.c, regint.h, regparse.c: use long.
+
+Tue Jun 30 11:05:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_find_1): fixed index overrun.
+
+Tue Jun 30 08:42:34 2009  Eric Hodel  <drbrain@segment7.net>
+
+	* tool/instruby.rb: summary is required in a .gemspec.
+
+Tue Jun 30 01:35:12 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/strip-rdocs.rb: supports QT style doxy-comments.
+
+Tue Jun 30 01:24:10 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (log2, cbrt): added.  [experimental]
+
+Tue Jun 30 01:19:53 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_expt): do not use rb_fexpt.
+
+Mon Jun 29 22:50:10 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* Doxyfile.in: removed. merged into template/Doxyfile.template
+
+	* configure.in: new checking for dot and doxygen.
+
+	* template/Doxyfile.template: merged with Doxyfile.in.
+	  configured some options.
+
+	* common.mk (capi): use $(DOXYGEN) instead of "doxygen".
+	  (Doxyfile): removed a duplicate entry in the dependency.
+
+Mon Jun 29 21:01:31 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_expt): checks exactness.
+
+Mon Jun 29 20:29:11 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (float_to_r): always returns rational.
+
+Mon Jun 29 18:55:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_find_1): fix for files with dots.  [ruby-dev:38588]
+
+Mon Jun 29 17:14:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): should copy original encoding.
+	  [ruby-dev:38612]
+
+Sun Jun 28 23:10:55 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* gem_prelude.c (Gem.default_dir): follows the change on
+	  lib/rubygems/default.rb in r23879
+
+Sun Jun 28 23:32:11 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_div): raises ZeroDivisionError immediately
+	  when the given second argument is zero.
+
+	* rational.c (nurat_fdiv): never raise even if the given second
+	  argument is zero.
+
+	* rational.c (rb_raise_zerodiv): changed the message (zero to 0).
+
+Sun Jun 28 22:25:07 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_expt): convert to a float when the given power
+	  is a bignum.
+
+	* rational.c (nurat_expt): ditto.
+
+Sun Jun 28 21:16:48 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (sqrt): fixed an issue [ruby-list:45852].
+
+Sun Jun 28 19:48:29 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_expt): some improvements.
+
+	* rational.c (nurat_expt): ditto.
+
+Sun Jun 28 19:03:46 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/instruby.rb (:gem): registers the bundled version
+	  of minitest as a gem as rdoc or rake.
+	  c.f. [ruby-dev:38692].
+
+Sun Jun 28 19:02:07 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/rubygems.rb (ConfigMap[:rubylibprefix]): new entry.
+
+	* lib/rubygems/defaults.rb (Gem.default_dir): considers
+	  "--with-rubylibprefix" configure option.
+
+Sun Jun 28 09:21:00 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: renamed some static functions.
+
+	* rational.c: ditto.
+
+Sat Jun 27 19:06:22 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_addsub): new
+
+	* complex.c (nucomp_{add,sub}): use nucomp_addsub.
+
+	* complex.c (nucomp_divide): changed the algorithm.
+
+	* complex.c (nucomp_abs): added shortcuts.
+
+Sat Jun 27 16:56:33 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_cmp): use rb_num_coerce_cmp.
+
+Sat Jun 27 16:45:10 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: revised rdoc.
+
+	* rational.c: ditto.
+
+	* numeric.c: ditto.
+
+Sat Jun 27 13:44:48 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS, lib/rss/maker/base.rb, test/rss/test_maker_2.0.rb: add
+	  item.guid.permanent_link? and item.guid.permanent_link=.
+
+Sat Jun 27 13:41:00 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS: rss: 0.2.5 -> 0.2.7.
+
+	* lib/rss/rss.rb, test/rss/test_version.rb: 0.2.6 -> 0.2.7.
+
+Sat Jun 27 03:16:56 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_chr): use default_internal encoding as default
+	  destination encoding if set.  [ruby-core:23997]
+
+Sat Jun 27 03:09:04 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_rewind): need to rewind $. and ARGF.lineno.
+	  [ruby-core:24046]
+
+	* io.c (struct argf): refactoring on $. and ARGF.lineno behavior.
+
+Fri Jun 26 21:48:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (pty_getpty): check dup failure.
+
+Fri Jun 26 17:33:46 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_rubyopt):
+	  test suite add '.' to RUBYLIB.  remove checks.
+
+	* test/ruby/test_require.rb (TestRequire#test_tainted_loadpath):
+	  the default tempdir directory /tmp is world writable, so
+	  SecurityError would be raised.  check removed.
+
+Fri Jun 26 16:32:59 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (COMPILE_PRELUDE): need -I. before -rrbconfig.
+	  [ruby-dev:38714]
+
+Thu Jun 25 18:41:51 2009  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.4.2 r5269.
+	* test/minitest/*.rb: ditto.
+
+Thu Jun 25 17:58:39 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_binmode_m): should call rb_io_ascii8bit_binmode() to
+	  set its encoding to ASCII-8BIT.  [ruby-core:24029]
+
+Thu Jun 25 13:04:58 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* instruby.rb: '&' in sed s command's replacement is '\&' in ruby.
+	  [ruby-dev:38713]
+
+Thu Jun 25 06:50:23 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_each_codepoint): uninitialized local variable enc.
+
+Thu Jun 25 06:25:49 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (first_i): wrong condition for no argument #first.
+	  [ruby-core:24017]
+
+Wed Jun 24 20:19:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_s_now): new function.  Time.now don't take arguments.
+
+Wed Jun 24 16:08:03 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/resource.rb: CONFIG["TEENY"] is not ruby's version but API's
+	  one.  So need to use RUBY_VERSION instead.
+
+Wed Jun 24 16:07:04 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/mkexports.rb: rbconfig.rb exists at ".".
+
+Wed Jun 24 15:02:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_set_encode): show the erred file name instead of
+	  the file that requires it.  [ruby-core:24006]
+
+Wed Jun 24 11:41:20 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-style.el: It is too late to set c-file-style in
+	  c-mode-hook (at least on Emacs 23).  Call c-set-style instead.
+
+Tue Jun 23 21:28:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_get_path_check): check with given safe level.
+
+	* file.c (rb_find_file_ext_safe, rb_find_file_safe): ditto.
+
+	* safe.c (rb_insecure_operation): function to raise security
+	  error.
+
+Tue Jun 23 20:32:43 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gc.c: remove the definition of GC_DEBUG (debugging macro).
+
+Tue Jun 23 16:16:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ruby_version): defaults revision to 0 when no
+	  revision.h exists.
+
+Tue Jun 23 16:04:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext, rb_find_file): no needs to expand
+	  paths with tilde twice.
+
+	* load.c (rb_f_load): load the given path directly if not found in
+	  load_path.
+
+	* load.c (search_required): search file in specified safe level.
+
+	* load.c (rb_require_safe): path to load is already searched in
+	  search_required().
+
+Tue Jun 23 12:43:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: remove PACKAGE_* macros generated by autotools.
+	  [ruby-core:20938]
+
+Tue Jun 23 01:17:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c (process_options): don't specify .so for encdb here.
+	  "." is replaced by "_" in load_encoding.
+
+	* encoding.c (load_encoding): add .so here.
+
+Mon Jun 22 23:24:22 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_generic_ivar_memsize): typo fixed.  a patch from
+	  Kazuhiro NISHIYAMA.  [ruby-dev:38700]
+
+	* ext/objspace/objspace.c (memsize_of): ditto.
+
+Mon Jun 22 21:21:59 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: remove __CHECKER__ test.
+
+	* dir.c: ditto.
+
+	* dln.c: ditto.
+
+	* file.c: ditto.
+
+	* process.c: ditto.
+
+Mon Jun 22 17:15:38 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_each_codepoint): new method.
+	  [ruby-core:23949]
+
+	* ext/stringio/stringio.c (strio_each_codepoint): ditto.
+
+Mon Jun 22 16:26:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): removed "." from load_path.
+
+Mon Jun 22 16:14:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_arguments, iseq_compile_each): internal
+	  arrays must be hidden.  [ruby-dev:38613]
+
+	* vm.c (Init_top_self): ditto.
+
+Mon Jun 22 14:41:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options), enc/prelude.rb: encdb and transdb are
+	  extension libraries.
+
+	* ruby.c (process_options): set progname earlier.
+
+Mon Jun 22 13:50:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_fdatasync): new method IO#fdatasync.
+
+Sun Jun 21 22:33:05 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* load.c (Init_load): $: must be readonly.  [ruby-dev:38690]
+
+	* ruby.c (ruby_prog_init): $-W must be readonly.  [ruby-dev:38691]
+
+Sun Jun 21 10:47:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::Entry_#copy_file): open with
+	  default umask.  [ruby-core:23952]
+
+Sun Jun 21 10:46:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (enc_arg): default internal encoding may not be set.
+	  [ruby-core:23932]
+
+Sat Jun 20 21:11:43 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c (num_div): don't use num_floor which is actually
+	  flo_floor.
+
+	* numeric.c (num_modulo): don't call '%'.
+
+	* numeric.c (num_divmod): use num_modulo.
+
+	* numeric.c: defined '%'.
+
+	* rational.c (nurat_idiv,nurat_mod,nurat_divmod,nurat_rem): removed.
+
+Sat Jun 20 20:28:44 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: edited rdoc.
+
+	* numeric.c: ditto.
+
+Sat Jun 20 08:56:47 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: edited rdoc.
+
+	* rational.c: ditto.
+
+	* numeric.c: ditto.
+
+Sat Jun 20 07:17:52 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/monitor.rb (MonitorMixin::extend_object): should use
+	  #__send__ instead of #send to avoid possible name conflict.
+	  [ruby-core:23907]
+
+Sat Jun 20 06:56:31 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: edited rdoc.
+
+	* rational.c: ditto.
+
+Sat Jun 20 05:08:59 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: edited rdoc.
+
+	* rational.c: ditto.
+
+Sat Jun 20 04:30:35 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_abs): removed.
+
+Sat Jun 20 03:34:16 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: added rdoc.
+
+Fri Jun 19 23:43:38 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c: edited rdoc.
+
+Fri Jun 19 22:58:16 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: edited rdoc.
+
+Fri Jun 19 22:21:17 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c: edited rdoc.
+
+Fri Jun 19 21:56:01 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_expt): delegates to complex when self is
+	  negative.  because Float#** does not produce complex.
+
+Fri Jun 19 21:40:58 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c: edited rdoc.
+
+	* rational.c: ditto.
+
+Fri Jun 19 20:53:54 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_name_list): update RDoc.  [ruby-core:23926]
+
+Fri Jun 19 20:44:45 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: constant COMPLEX_NAME has been removed.
+
+	* rational.c: constant RATIONAL_NAME has been removed.
+
+Fri Jun 19 20:39:46 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: added rdoc.  a patch from Run Paint Run Run.
+
+Fri Jun 19 17:04:59 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_cmp): should always return nil for NaN.
+
+	* numeric.c (flo_cmp): handle infinite value specially using
+	  infinite? method internally.  [ruby-dev:38681]
+
+Fri Jun 19 09:28:45 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c (*_numerator,*_denominator): moved to rational.c.
+
+	* rational.c (*_numerator,*_denominator): moved from numeric.c.
+
+Fri Jun 19 08:14:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (big_lshift, big_rshift): return Bignum always without
+	  normalization.  [ruby-dev:38679]
+
+Thu Jun 18 22:31:38 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_s_convert): calls to_r when the given argument
+	  is non-integer.
+
+	* rational.c (nurat_s_convert): raises TypeError when the given
+	  argument is nil.
+
+	* complex.c (nucomp_s_convert): ditto.
+
+Thu Jun 18 20:32:11 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c (num_numerator, num_denominator): use
+	  to_r [ruby-core:23910].
+
+Thu Jun 18 16:21:05 2009  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.4.0 r5083.
+	* test/minitest/*.rb: ditto.
+
+Thu Jun 18 10:12:49 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): went infinity too
+	  early.  add BASE_FIG margin.  [ruby-dev:38673]
+
+Thu Jun 18 01:35:51 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_cmp): Infinity is greater than any bignum
+	  number.  [ruby-dev:38672]
+
+	* bignum.c (rb_big_cmp): ditto.
+
+Thu Jun 18 01:29:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): drive letter is ascii only.
+	  [ruby-dev:38612]
+
+Thu Jun 18 01:09:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ridir, RI_BASE_NAME): fixed for path expansion.
+	  [ruby-core:23876]
+
+Wed Jun 17 23:46:08 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_coerce): accepts Complex when the imag is
+	  exact zero.
+
+Wed Jun 17 21:25:54 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* bignum.c (rb_big_fdiv): checks whether the given second argument
+	  can be converted to float properly.
+
+	* numeric.c (fix_fdiv): calls rb_big_fdiv when the given second
+	  argument is a bignum.
+
+	* rational.c (nurat_fdiv): should calculate Float(x/y), not
+	  Float(x)/Float(y).
+
+Wed Jun 17 16:57:40 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* load.c (rb_f_require): RDoc updated.  a patch from Run Paint Run
+	  Run in [ruby-core:23833].
+
+	* load.c (rb_mod_autoload): ditto.  [ruby-core:23835]
+
+Wed Jun 17 14:37:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/test.rb (valid_syntax?): skips BOM.  [ruby-dev:38666]
+
+	* test/ruby/test_system.rb (TestSystem#valid_syntax?): ditto.
+
+Wed Jun 17 13:54:18 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/strscan/strscan.c (Init_strscan): remove obsolete
+	  matchedsize method, use matched_size instead.  [ruby-dev:38591]
+
+Wed Jun 17 12:37:37 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (ruby_thread_stack_overflow): call rb_exc_raise() on
+	  stack overflows in the signal handler, if sigaltstack is
+	  available.  On stack overflow (and with sigaltstack), the signal
+	  handler is more likely to have room to create an exception
+	  object.  [ruby-core:23813]
+
+Wed Jun 17 08:10:38 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace: added.  objspace library extends some methods to
+	  ObjectSpace module.
+
+Wed Jun 17 08:14:01 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_coerce): accepts Complex instances.
+
+	* rational.c (nurat_coerce): accepts Rational
+	  instances. [ruby-core:23859]
+
+Wed Jun 17 07:36:22 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httputils.rb (parse_form_data): escape boundary of
+	  multipart/form-data when embed in regexp.
+
+Wed Jun 17 07:24:26 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_memsize): added.
+
+	* io.c (rb_io_memsize): added.
+
+	* regcomp.c (onig_memsize): added.
+
+	* string.c (rb_str_memsize): added.
+
+	* transcode.c (rb_transcoding_memsize, rb_econv_memsize): added.
+
+	* variable.c (rb_geneic_ivar_memsize): added.
+
+Wed Jun 17 07:04:33 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_memsize): added.  Use RTypedData instead of RData
+	  for ISeq.
+
+	* vm.c (env_memsize, vm_memsize, thread_memsize): added.  Use
+	  RTypedData instead of RData for Env, VM, Thread.
+
+Wed Jun 17 06:48:28 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* st.c, include/ruby/st.h (st_memsize): added.  This function returns
+	  the memory usage of st_table.
+
+Wed Jun 17 06:19:06 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: New structure RTypedData, added.
+	  This structure includes more explicit type information for
+	  T_DATA objects.  If RData(obj)->dfree is immediate value `1' on
+	  T_DATA object obj, obj is needed to be accessed with RTYPEDDATA(obj)
+	  instead of RDATA(obj).  A RTypedData structure points the structure
+	  rb_typed_data_t.  rb_typed_data_t includes information such as the
+	  type name of this data, mark and free function what RData includes,
+	  and memsize function show how data consuming the memory size.
+	  Note that you do not need any change existing T_DATA objects.
+	  If you use RDataType instead of RData on T_DATA object,
+	  you can specify explicit type information.
+
+	* gc.c (rb_data_typed_object_alloc, rb_objspace_data_type_memsize,
+	  rb_objspace_data_type_name): added.
+
+Wed Jun 17 06:14:23 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix indent.
+
+Wed Jun 17 06:05:03 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_each_objects): New C API, added.
+
+Wed Jun 17 00:31:30 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_argf.rb (TestArgf#test_skip): updated test
+	  according to clarified behavior.
+
+Tue Jun 16 22:47:37 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (fptr_finalize): revert last change.  [ruby-dev:38648]
+
+	* io.c (fptr_finalize): skip close(2) for fd 0,1,2.
+
+Tue Jun 16 20:07:09 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (transcode_restartable0): refix can't build with VC9.
+
+Tue Jun 16 16:09:59 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* parse.y (parser_read_escape, parser_tokadd_escape):
+	  replace scan_oct as ruby_scan_oct.
+
+Tue Jun 16 06:40:31 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (fptr_finalize): should close stdin/stdout/stderr when
+	  closed explicitly.   [ruby-core:23853]
+
+	* io.c (argf_skip): should close only when current_file is available.
+
+Tue Jun 16 01:50:02 2009  Tanaka Akira  <akr@fsij.org>
+
+	* vm_eval.c (rb_call0): refine exception message for hidden objects.
+
+Mon Jun 15 22:35:31 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* bignum.c (rb_big2db): (-Float::MAX.to_i*2).to_f should return
+	  -HUGE_VAL (-Infinity).
+
+Mon Jun 15 18:48:41 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_each_line): should return self.  [ruby-core:23852]
+
+	* io.c (argf_each_byte, argf_each_char): ditto.
+
+Mon Jun 15 17:48:42 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (os_obj_of): invoke garbage collection before iteration, to
+	  avoid accessing half recycled object references.  [ruby-dev:38613]
+
+Mon Jun 15 11:04:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .gdbinit (rp, iseq): load dummy_gdb_enums on demand.
+	  [ruby-dev:38606]
+
+Sun Jun 14 14:57:57 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, vm_eval.c: add Thread.backtrace.
+
+	* test/ruby/test_thread.rb: add a test.
+
+Sun Jun 14 13:58:32 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* transcode.c (transcode_restartable0): revert last commit because
+	  this change cause SEGV at test-all.
+
+Sun Jun 14 10:49:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext, rb_find_file): canonicalize absolute
+	  paths.  [ruby-core:23845]
+
+	* file.c (rb_file_size): added rdoc.  a patch from Run Paint Run
+	  Run at [ruby-core:23839].
+
+Sun Jun 14 07:53:26 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_fdiv): use fdiv recursively.
+
+	* complex.c (nucomp_expt): reduced code.
+
+Sun Jun 14 03:37:09 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf8_mac.trans: remove wrong optimization.
+
+Sun Jun 14 01:53:00 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (transcode_restartable0): can't build with VC9.
+
+Sun Jun 14 01:23:41 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_to_f): use fdiv.
+
+Sat Jun 13 15:03:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_lock): show backtrace at circular require.
+
+	* load.c (rb_provide): assumes us-ascii only.
+
+	* load.c (rb_require_safe): FilePathValue() implies rb_str_new4().
+
+	* load.c (rb_mod_autoload): try conversion to path like as
+	  require.  [ruby-core:23834]
+
+Sat Jun 13 09:58:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_ivar): should not set internal encoding ivar as an
+	  ordinary ivar.  [ruby-dev:38596]
+
+Sat Jun 13 07:08:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_f_local_variables): now returns symbols.  a patch from
+	  Run Paint Run Run at [ruby-core:23828].
+
+Sat Jun 13 07:06:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_f_catch): updated rdoc about generalized argument,
+	  and the case without arguments.  [ruby-core:23827]
+
+Sat Jun 13 06:50:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/protocol.rb (Net::BufferedIO#rbuf_fill): TimeoutError is
+	  obsolete, use Timeout::Error instead.  [ruby-core:23821]
+
+Sat Jun 13 06:45:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_f_throw): fixed rdoc about exception.
+	  [ruby-core:23824]
+
+Fri Jun 12 14:56:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): associate the input encoding when
+	  copying an absolute path.  [ruby-dev:38594]
+
+Fri Jun 12 02:41:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_replace_shared): shared target must be frozen.
+	  [ruby-core:23727]
+
+Thu Jun 11 21:05:09 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (exp): omitted redundant function call.
+
+Thu Jun 11 17:49:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (gfCheckVal): never used.
+
+	* ext/bigdecimal/bigdecimal.c (VpInit): fixed format modifiers.
+
+	* ext/bigdecimal/bigdecimal.c (VPrint): constified.
+
+Thu Jun 11 15:27:17 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/prime.rb: documentation typo fixed.  a patch from okkez.
+	  [ruby-dev:38586]
+
+Wed Jun 10 18:15:17 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* dir.c (dir_s_getwd): directory path's encoding should be filesystem's
+	  one.
+
+	* lib/tmpdir.rb: ditto (but not finished yet.)
+
+Wed Jun 10 06:28:15 2009  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*: Upgrade to RubyGems 1.3.4 r2223.
+
+Tue Jun  9 22:38:09 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (log10): raised exception when the given number is
+	  a negative real.
+
+Tue Jun  9 15:13:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_s_glob): fixed rdoc.  a patch from Joseph Pecoraro a
+	  [ruby-core:23767].
+
+	* dir.c (sys_warning): get rid of type-punning function cast.
+
+	* dir.c (ruby_glob0): get rid of possible overflow.
+
+Tue Jun  9 10:58:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub (RMALL): need for distclean-rdoc.
+
+Tue Jun  9 01:07:33 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: rename functions which require a parameter
+	  "rb_thread_t *", the prefix to be rb_threadptr_ instead of
+	  rb_thread_.
+
+	* thread.c (rb_thread_add_event_hook(), rb_thread_remove_event_hook):
+	  change the parameter type from rb_thread_t * to VALUE.
+
+	* eval.c, eval_error.c, eval_intern.h, signal.c, vm_core.h, vm_eval.c:
+	  ditto.
+
+	* include/ruby/intern.h: remove decl of rb_thread_signal_raise() and
+	  rb_thread_signal_exit().
+
+Mon Jun  8 05:07:41 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): print fatal error
+	  message to stderr instead of using rb_bug().
+
+	* KNOWNBUGS.rb, bootstraptest/test_fork.rb: move a fixed test.
+
+Sun Jun  7 22:44:20 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (log): avoided redundant expression.
+
+Sat Jun  6 02:49:05 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_reject_bang): always check frozen status.
+	  [ruby-core:23715]
+
+	* hash.c (rb_hash_update): ditto.
+
+	* hash.c (rb_hash_reject_bang): call rb_hash_foreach() directly.
+
+	* hash.c (rb_hash_update_i): call st_insert() directly.
+
+	* hash.c (rb_hash_update_block_i): ditto.
+
+Fri Jun  5 07:12:32 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mkmf.rb (#link_command): should dup CONFTEST_C which is
+	  frozen.  ref  [ruby-core:23675].  [ruby-core:23702]
+
+Thu Jun  4 02:25:51 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/base64.rb: typo fixed.  a patch from okkez.   [ruby-dev:38564]
+
+Wed Jun  3 09:03:23 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_first): should check negative length.
+
+Tue Jun  2 17:32:40 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (first_i): Enumerator#first should consume only what is
+	  needed.   a patch from Marc-Andre Lafortune.  [ruby-core:23661]
+
+	* enum.c (enum_first): call to_int once for an argument.  based on
+	  a patch from Marc-Andre Lafortune.
+
+Tue Jun  2 13:27:21 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ripper/test_filter.rb: add tests. see [ruby-dev:37856]
+
+Tue Jun  2 07:44:43 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_gsub_bang): modify check at the beginning.
+	  [ruby-core:23662]  ref [ruby-core:23657]
+
+	* string.c (rb_str_rstrip_bang): ditto.  [ruby-core:23657]
+
+	* string.c (rb_str_chop_bang): ditto.
+
+	* string.c (rb_str_chomp_bang): ditto.
+
+	* string.c (rb_str_reverse_bang): modify check added.  [ruby-core:23671]
+
+Mon Jun  1 11:21:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_capture, fiber_store): reraise transferred error.
+
+	* cont.c (fiber_switch): transfers dead fiber error to the previous
+	  or root fiber if the current fiber is dead.  [ruby-core:23651]
+
+Mon Jun  1 10:41:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_exc_new2): optimization for literal.
+
+Mon Jun  1 07:20:02 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): returns Inf if
+	  exp is bigger than DBL_MANT_DIG.
+
+Sun May 31 23:28:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): checks for duplication of source
+	  files.
+
+Sun May 31 23:26:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (istrailinggarbage): fixed typo.
+
+Fri May 29 17:10:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (Init_Struct): made #to_s an alias to #inspect to
+	  reduce the result of recursive struct.  a patch from ujihisa a
+	  [ruby-dev:38554].
+
+Fri May 29 17:08:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (Init_Hash): made #to_s an alias to #inspect to reduce
+	  the result of recursive hash.  a patch from ujihisa a
+	  [ruby-core:23601].  [ruby-dev:38555]
+
+Fri May 29 09:30:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (collect_all): checks interrupts.  [ruby-core:23594]
+
+Thu May 28 07:39:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (THREAD_MODEL): reject unknown value and checks
+	  pthread.h only when pthread.  [ruby-core:23577]
+
+Thu May 28 03:47:46 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/optparse/opttest.rb: typo fixed.  [ruby-dev:38544]
+
+Thu May 28 03:43:10 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_strftime): update RDoc according to info from
+	  Marc-Andre Lafortune in [ruby-core:23575].  [ruby-core:23564]
+
+Thu May 28 02:40:54 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/uri: don't set @parser if it is DEFAULT_PARSER for marshaling
+	  URI objects between Ruby 1.8 and Ruby 1.9.
+	  [ruby-dev:38377]
+
+Wed May 27 23:00:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (struct parser_params): lex_gets_ptr should be long.
+
+Wed May 27 18:00:15 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigand_int): new function to calculate bignum and
+	  fixnum without allocating internal bignum.
+
+	* bignum.c (bigor_int): ditto.
+
+	* bignum.c (bigxor_int): ditto.
+
+	* bignum.c (bigand_int): even less object allocation.
+
+Wed May 27 14:29:55 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_encoding): more compact encoding information for
+	  US-ASCII and UTF-8.  [incompatible] [experimental]
+
+	* marshal.c (r_ivar): restore :E encoding information.
+
+Wed May 27 14:08:39 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* st.c (st_insert2): new function with processing new key,
+	  e.g. copy.
+
+	* hash.c (rb_hash_aset): use st_insert2() to reduce redundant
+	  st_lookup calls.
+
+Wed May 27 02:31:38 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_getc): the function for
+	  rl_getc_function must be a byte function.
+	  so use getbyte method. [ruby-dev:38535]
+
+Tue May 26 14:24:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c: fixed types.
+
+	* common.mk (bignum.o, numeric.o): depend on util.h.
+
+	* bignum.c, marshal.c: fixed types.
+
+	* numeric.c (infinite_value): use ruby_div0.
+
+	* include/ruby/util.h (ruby_div0): moved from marshal.c.
+
+Tue May 26 11:01:41 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* lib/mkmf.rb: use map! to replace strings in $objs array.
+
+Tue May 26 10:12:08 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bignew_1): inline memory allocation.
+
+	* bignum.c (bigtrunc): call rb_big_resize() only when needed.
+
+	* bignum.c (bigfixize): declare inline.
+
+Tue May 26 05:39:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (FilePathValue): prevent from GC.
+
+	* include/ruby/ruby.h (NUM2LONG): added GCC specific optimization.
+
+Tue May 26 03:41:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (rb_gmtime, rb_localtime): gmtime and localtime return
+	  NULL on error.  [ruby-core:23551]
+
+Tue May 26 03:38:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_each_char, rb_str_each_codepoint): string
+	  length must be long.
+
+Mon May 25 13:27:32 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* lib/mkmf.rb: dont use gsub! method for frozen string.
+
+Mon May 25 11:47:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb, tool/instruby.rb: removed redundant code.
+
+Mon May 25 09:34:09 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_hash): avoid calling rb_enc_str_asciionly_p().
+
+	* string.c (rb_str_replace): avoid redundant calling rb_str_new4().
+
+	* string.c (str_replace): factor out replacement from
+	  rb_str_replace() without type check nor discarding the
+	  destination contents.
+
+Mon May 25 08:06:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_partition): should use the converted result.  a
+	  patch from Marc-Andre Lafortune at [ruby-core:23540].
+
+	* string.c (rb_str_rpartition): ditto.
+
+Mon May 25 06:25:38 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_hash.rb (TestHash::test_equal2): recursive hashes
+	  are handled properly now. ref: [ruby-core:23402]
+
+	* test/ruby/test_m17n.rb (TestM17N#test_sprintf_p): test fixed
+
+Mon May 25 05:32:19 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI::HTTP_STATUS): typo fixed.  a patch from
+	  Nobuhiro IMAI.  [ruby-dev:38538]
+
+Sun May 24 22:48:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_exec_recursive_paired): new function for proper
+	  handling of recursive arrays.  [EXPERIMENTAL] [ruby-core:23402]
+
+	* array.c (rb_ary_equal, rb_ary_eql, rb_ary_cmp): use above.
+
+	* hash.c (hash_equal): ditto.
+
+Sun May 24 22:39:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (syserr_initialize): errno is int.
+
+Sun May 24 00:52:54 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* configure.in ($ridir): new configuration. [ruby-core:23520].
+	  c.f. [ruby-core:23519].
+
+	  (--with-ridir): new configure option.
+
+	* tool/instruby.rb (:doc, :rdoc): uses $ridir instead of
+	  a fixed path.
+
+	* lib/rdoc/ri/paths.rb: follows $ridir.
+
+	* Makefile.in: removes RIDATADIR which is no longer used.
+
+	* bcc32/Makefile.sub: generates the 'ridir' entry for RbConfig.
+	  removes RIDATADIR which is no longer used.
+
+	* win32/Makefile.sub: ditto.
+
+Sat May 23 23:52:33 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_each_char): return original string.
+	  [ruby-core:23499]
+
+	* string.c (rb_str_each_codepoint): protect string from
+	  modification.
+
+Sat May 23 21:48:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/handle.c (rb_dlhandle_s_sym): added a method to access
+	  using RTLD_NEXT.  [ruby-dev:38152]
+
+	* ext/dl/handle.c (Init_dlhandle): added constants DEFAULT and
+	  NEXT which correspond to RTLD_DEFAULT and RTLD_NEXT.
+
+Sat May 23 18:53:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/lib/dl/cparser.rb (DL::CParser#parse_struct_signature):
+	  splitting with regexp source string is obsolete.  a patch from
+	  Minwoo Lee at [ruby-core:23494].
+
+	* ext/dl/cptr.c (rb_dlptr_cmp): return signed value, and restrict
+	  to Fixnum.  [ruby-dev:38533]
+
+Fri May 22 23:22:53 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* missing/vsnprintf.c (errno): [BUG] fixes a compilation
+	  error on SIZEOF_LONG > SIZEOF_INT.
+	  (BSD_vfprintf): ditto.
+
+Fri May 22 23:20:48 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/default.mspec: follows runruby.rb's move at r23542.
+
+Fri May 22 21:38:55 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* NEWS: add Time#to_r.
+
+Fri May 22 20:29:01 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* config.guess: moved into tool/.
+
+	* config.sub: ditto.
+
+	* install-sh: ditto
+
+	* configure.in: follows the moves.
+
+	* LEGAL: ditto.
+
+Fri May 22 20:10:18 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* instruby.rb: moved into tool/.
+
+	* mkconfig.rb: ditto.
+
+	* rubytest.rb: ditto.
+
+	* runruby.rb: ditto.
+
+	* common.mk: follows the moves.
+
+	* configure.in: ditto.
+
+	* win32/Makefile.sub: ditto.
+
+Fri May 22 05:09:43 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_slice_bang): avoid call of rb_scan_args() unless
+	  it's really necessary.
+
+Thu May 21 22:17:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/id.h.tmpl, id.h (enum ruby_method_ids): added some IDs.
+
+	* debug.c (dummy_gdb_enums): added enum ruby_method_ids.
+
+	* .gdbinit (rp): improved output of Symbol.
+
+Thu May 21 21:07:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/vsnprintf.c (BSD_vfprintf): support for 'z' modifier.
+
+Thu May 21 18:55:33 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* gem_prelude.rb (Gem.default_dir and misc.): use rubylibprefix.
+	  follows the change in r23368.
+
+Thu May 21 12:07:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_EVENT_HOOKS_HAVE_CALLBACK_DATA):
+	  new macro for compatibility check.
+
+Thu May 21 01:43:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_long2int, RARRAY_LENINT): check long to
+	  cast to int.  [ruby-dev:38508]
+
+	* struct.c, vm_eval.c, vm_insnhelper.c: use RARRAY_LENINT.
+
+Wed May 20 21:00:27 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* rb_enc_get_index: allows an arbitrary RData as the argument but not
+	  only what points a rb_encoding.
+
+Wed May 20 20:54:37 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/.gitignore: ignores rubyspec/ and mspec/.
+
+Wed May 20 19:41:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (rb_struct_new): get rid of too large alloca.
+
+	* struct.c (rb_struct_hash): use long.
+
+Wed May 20 18:58:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c, vm_insnhelper.c: argument number is restricted to
+	  int, and fixed overflow.
+
+Wed May 20 18:34:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (zip_ary): should use long.
+
+	* enumerator.c (inspect_enumerator): should use long.
+
+Wed May 20 09:18:44 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_count): optimized for 1byte string count by
+	  avoiding tr_setup_table().
+
+Wed May 20 06:25:29 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_fast_mbclen): faster mbclen for strings known
+	  to be valid.
+
+	* string.c (enc_strlen): coderange specified version of
+	  rb_enc_strlen().  use rb_enc_fast_mbclen() if coderange is 7bit
+	  or valid.
+
+	* string.c (str_gsub): use rb_enc_fast_mbclen().
+
+	* string.c (rb_str_reverse, rb_str_split_m, rb_str_each_char,
+	  scan_once): ditto.
+
+Wed May 20 06:20:05 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#unlink): close first for Windows.  a
+	  patch from Florian Frank.  [ruby-core:23505]
+
+Wed May 20 00:13:38 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_codepoint_len): combine rb_enc_codepoint()
+	  and rb_enc_codelen() in one function to reduce calls.
+
+	* encoding.c (rb_enc_codepoint): compatibility function.
+
+	* sprintf.c (rb_str_format): use rb_enc_codepoint_len().
+
+	* string.c (rb_str_inspect, rb_str_upcase_bang,
+	  rb_str_downcase_bang, rb_str_capitalize_bang,
+	  rb_str_swapcase_bang, trnext, tr_trans, rb_str_delete_bang,
+	  rb_str_squeeze_bang, rb_str_count, rb_str_split_m,
+	  rb_str_each_line, rb_str_each_codepoint, rb_str_lstrip_bang,
+	  sym_printable): ditto.
+
+	* transcode.c (make_econv_exception): use rb_enc_mbc_to_codepoint()
+
+Wed May 20 00:05:52 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_method.c (rb_attr): should preserve encoding info.
+	  [ruby-dev:38498]
+
+Tue May 19 22:54:35 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* time.c (time_minus): always return a Float.  [ruby-dev:38446]
+
+	* time.c (time_to_r): new method.  [ruby-dev:38461]
+
+Tue May 19 13:59:35 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c (clone_method): add cast to remove warning from
+	  rb_gc_write_barrier().
+
+Tue May 19 13:54:15 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (struct_ivar_get): new function to avoid repeated
+	  rb_intern() calls.
+
+	* struct.c (rb_struct_iv_get): use struct_ivar_get()
+
+	* struct.c (num_members): ditto.
+
+	* struct.c (rb_struct_s_members): ditto.
+
+	* class.c (rb_singleton_class): cache symbol to reduce calls to
+	  rb_intern().
+
+Tue May 19 07:52:05 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/test_time.rb: make tests timezone independent.
+	  reported by zunda.  [ruby-dev:38492]
+
+Mon May 18 21:40:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pathname.rb (Pathname#sub): suppress a warning.  [ruby-dev:38488]
+
+Sun May 17 23:23:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (SRC_EXT): should be flat.
+	  http://twitter.com/_tad_/status/1825862632
+
+Sun May 17 23:05:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): VARIABLE_LIBPATH is always
+	  defined, see its value instead.
+
+Sun May 17 18:59:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.h (nd_line): NODE_LMASK is not needed.
+
+	* node.h (NOEX_SAFE): made int.
+
+Sun May 17 14:23:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (rb_parse_in_eval): returns true in true eval, not in
+	  main.  [ruby-dev:38382]
+
+	* parse.y (program): inherits dvars in eval or main.
+
+Sun May 17 14:02:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_script): sets also VM toplevel program name.
+
+	* ruby.c (process_options): no longer needs additional frame.
+
+	* vm.c (rb_vm_get_sourceline): should not access out of bound.
+
+Sun May 17 09:47:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (cmdline_options_init): initialize encodings.
+
+	* ruby.c (add_modules, require_libraries, process_sflag):
+
+	* ruby.c (process_sflag): not process twice.
+
+	* ruby.c (moreswitches): get rid of possible overflow.
+
+Sun May 17 09:31:05 2009  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (foletypelib_name): should return
+	  encoded name corresponding to WIN32OLE.codepage.
+
+Sun May 17 09:02:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): support for cygwin 1.7.  see
+	  [ruby-core:23241].
+	  gets rid of possible buffer overflow with realpath().
+
+	* ruby.c (set_arg0): get rids of overrun.
+
+Sat May 16 18:38:32 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb: add nil check.
+
+Sat May 16 18:36:01 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_maker_atom_feed.rb: suppress warnings.
+
+Sat May 16 18:33:15 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS: add RSS::Maker.supported?(version).
+
+Sat May 16 18:26:42 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb, test/test_parser_1.0.rb: fix foaf:Image
+	  element causes parse error even if ignore_unknown_element mode.
+
+Sat May 16 18:14:19 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker.rb, lib/rss/maker/0.9.rb,
+	  test/test_maker_*.rb: add RSS::Maker.supported?
+
+Sat May 16 18:12:39 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/content/*, lib/rss/dublincore/*: fix circular require.
+	* test/test_maker_atom_feed.rb,
+	  test/test_maker_atom_entry.rb: suppress warnings.
+
+Sat May 16 18:07:17 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/feed.rb, test/test_maker_atom_feed.rb:
+	  remove needless codes.
+
+Sat May 16 18:05:07 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/entry.rb: fix a typo.
+
+Sat May 16 18:02:57 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/feed.rb, test/test_maker_atom_entry.rb,
+	  test/test_maker_atom_feed.rb: fix duplicated dc:date.
+	  Reported by Kazuhiro NISHIYAMA. Thanks!!! [ruby-list:46014]
+
+Sat May 16 18:02:16 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb, lib/rss/maker/1.0.rb, lib/rss/maker/feed.rb,
+	  test/rss/test_maker_1.0.rb, test/rss/test_maker_atom_feed.rb:
+	  RSS 1.0 and Atom feed maker treat maker.channel.language as
+	  maker.channel.dc_language.
+
+Sat May 16 17:57:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/lib/dl/callback.rb (DL#remove_callback_internal): ignore
+	  unbound function.  [ruby-dev:38474]
+
+Sat May 16 17:51:11 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/rss_recent.rb, sample/rss/list_description.rb: use
+	  UTF-8.
+
+Sat May 16 17:47:55 2009  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb, test/rss/test_version.rb (RSS::VERSION):
+	  0.2.5 -> 0.2.6.
+
+Sat May 16 17:26:04 2009  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* iseq.c (rb_iseq_clone): use longlife object and insert write barrier.
+
+	* vm_insnhelper.c (vm_cref_push): ditto.
+
+	* vm_insnhelper.h (COPY_CREF): insert write barrier.
+
+Sat May 16 13:49:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_autoload_load): gets rid of false warning.
+	  [ruby-core:23466]
+
+Sat May 16 10:59:54 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/drb/dhasenc.rb: add magic comment for encoding.
+
+	* sample/mine.rb: ditto.
+
+	* ext/tk/sample/tcltklib/sample1.rb: ditto.
+
+Sat May 16 09:49:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (magic_comment_encoding): ignores unused emacs-style
+	  encoding comment, as like Vim styles.  [ruby-core:23470]
+
+Sat May 16 09:30:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/keywords (reserved_word): made inline function static.
+	  [ruby-core:23210]
+
+	* parse.y (rb_reserved_word): ordinary function for ripper.
+
+Sat May 16 09:19:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (magic_comment_encoding): use rb_compile_warning() to
+	  show the currently parsing file name.  [ruby-core:23469]
+
+Sat May 16 09:03:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ruby_version): now version.h includes
+	  include/ruby/version.h, so need to tell to cpp to see
+	  $(srcdir)/include.  [ruby-core:23468]
+
+Fri May 15 17:35:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_shift, rb_ary_shift_m): clears unused elements.
+	  [ruby-dev:38448]
+
+Fri May 15 15:15:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_autoload_load): checks if iv_tbl is valid.
+	  [ruby-dev:38456]
+
+Fri May 15 11:17:48 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.mak (-version-): now version.h includes
+	  include/ruby/version.h, so need to tell to cpp to check
+	  $(srcdir)/include.
+	  reported by KIMURA Koichi at http://www.kt.rim.or.jp/%7ekbk/zakkicho/09/zakkicho0905b.html#D20090514-6
+
+Thu May 14 16:13:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_getpwuid): use rb_uid_t.  [ruby-dev:38443]
+
+	* ext/stringio/stringio.c (strio_ungetbyte): encoding should no
+	  be effective.
+
+Thu May 14 10:17:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/test.rb (valid_syntax?): defaults to us-ascii.
+
+Wed May 13 22:34:31 2009  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c: add longlife garbage collection. [ruby-dev:38423]
+	  (NORMAL_HEAPS_USED): new macro.
+	  (LONGLIFE_ALLOCATE_HEAPS_MIN): ditto.
+	  (add_longlife_heaps_slot): new function.
+	  (rb_newobj_from_longlife_heap): ditto.
+	  (rb_newobj_longlife): ditto.
+	  (rb_node_newnode_longlife): ditto.
+	  (rb_gc_write_barrier): ditto.
+	  (remembered_set_recycle): ditto.
+	  (rb_gc_mark_remembered_set): ditto.
+	  (clear_mark_longlife_heaps): ditto.
+	  (gc_sweep_for_longlife): ditto.
+	  (assign_heap_slot): new argument to longlife heaps slot.
+	  (add_freelist): ditto.
+	  (gc_sweep): avoid longlife heap slot. set longlife_collection
+	  flag at add heap.
+	  (rb_gc_force_recycle): avoid mark object and remembered_set
+	  object.
+	  (garbage_collect): add longlife collection.
+	  (rb_gc_start): invoke longlife collection.
+	  (gc_profile_record_get): for longlife collection profile.
+	  (gc_profile_result): ditto.
+
+	* include/ruby/intern.h (rb_gc_write_barrier): declared.
+
+	* include/ruby/ruby.h (FL_REMEMBERED_SET): renamed from FL_RESERVED.
+
+	* debug.c (FL_REMEMBERED_SET): ditto.
+
+	* insns.def (setinlinecache): insert write barrier.
+
+	* vm_insnhelper.c (vm_method_search): ditto.
+
+	* set_relation (set_relation): use longlife object.
+
+	* vm.c (vm_define_method): ditto.
+
+	* vm_core.h (NEW_INLINE_CACHE_ENTRY): ditto.
+
+	* vm_method.c (rb_add_method): ditto.
+
+	* class.c (rb_add_method): ditto.
+
+	* node.h (NEW_NODE_LONGLIFE): new macro.
+	  (rb_node_newnode_longlife): declared.
+
+Wed May 13 15:23:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/version.h: extracted the extensions interface and
+	  the never-changeable info.
+
+Wed May 13 03:20:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_init_1): unused variable removed.
+
+Tue May 12 21:03:02 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: support fixed UTC offset.  [ruby-dev:38326]
+	  (leap_year_v_p): new macro.
+	  (TIME_FIXOFF_P): new macro.
+	  (TIME_SET_FIXOFF): new macro.
+	  (time_init_0): renamed from time_init.
+	  (time_set_utc_offset): new function.
+	  (vtm_add_offset): new function.
+	  (utc_offset_arg): new function.
+	  (time_init_1): new function.
+	  (time_init): call time_init_0 or time_init_1 according argc.
+	  (validate_utc_offset): new function.
+	  (time_localtime_m): new function.
+	  (time_fixoff): new function.
+	  (time_getlocaltime): take optional UTC offset argument.
+	  (time_get_tm): support fixed UTC offset time.
+	  (Init_Time): make Time#{initialize,localtime,getlocal} varargs.
+
+	* strftime.c (rb_strftime): vtm->zone can be NULL now.
+
+Tue May 12 18:23:40 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* yarvtest: removed because it's outdated.
+
+Mon May 11 21:46:20 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: add semicolons to some grammar rules not terminated
+	  with them.  a patch from Dave B in [ruby-core:23422].
+
+Mon May 11 20:08:33 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* io.c (Init_IO): add constant File::NOATIME.  [ruby-core:23194]
+
+Mon May 11 13:08:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb (rubylibdir): use rubylibprefix.  [ruby-dev:38426]
+
+Mon May 11 08:37:04 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_coerce): support
+	   coercing into Rational.   [ruby-core:23415]
+
+Mon May 11 04:39:45 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/smtp.rb (Net::SMTP#check_auth_args): should not change
+	  number of methods for the sake of compatibility.
+
+Sun May 10 11:36:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_instance_p): new function to check if
+	  the argument is an instance of DL::CFunc.
+
+	* ext/dl/cptr.c (rb_dlptr_initialize, rb_dlptr_s_malloc): checks
+	  if DL::CFunc.  [ruby-dev:38403].
+
+	* ext/dl/lib/dl/cparser.rb (DL::CParser#parse_signature): strips
+	  spaces.  based on a patch from Takashi Tamura in [ruby-dev:38398].
+
+	* ext/dl/lib/dl/value.rb (DL::ValueUtil#wrap_arg): block must be
+	  given if arg is not bound.  [ruby-dev:38404]
+
+	* ext/dl/cfunc.c (rb_dlcfunc_instance_p): new function to check if
+	  the argument is an instance of DL::CFunc.
+
+Sat May  9 19:57:00 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_float.rb (TestFloat#test_sleep_with_Float): add a
+	  test. see [ruby-core:23282]
+
+Sat May  9 19:23:46 2009  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: use subsec instead of nsec.
+
+Sat May  9 12:19:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c: 3rd argument of rb_hash_foreach() is VALUE.
+
+	* hash.c (rb_any_hash, recursive_hash): use VALUE for hash.
+
+Sat May  9 11:14:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_f_catch): gets rid of issue with gcc 4.4.  a patch
+	  from Alexey Froloff in [ruby-core:23398].  [ruby-core:22924]
+
+Fri May  8 19:38:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_link0): removes waste dSYM directories left
+	  when debug and universal-binary are enabled.
+
+	* lib/mkmf.rb (check_sizeof): fixed wrong recurring result for
+	  intrinsic types.
+
+Fri May  8 10:14:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, configure.in, win32/Makefile.sub (RUBY_BASE_NAME):
+	  program base name.  [ruby-dev:38241]
+
+	* configure.in (--with-soname): base name of shared library.
+	  [ruby-dev:38290]
+
+Fri May  8 10:07:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_flatten_bang): clears temporary array.
+
+Fri May  8 02:30:14 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (SortedSet#add): Do not require each newly added
+	  element to be Comparable but to respond to <=>. [ruby-dev:38371]
+
+Thu May  7 21:42:51 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rubyoptions.rb (test_indentation_check): add a test
+	  for indentation check.  [ruby-dev:38382]
+
+Thu May  7 16:40:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_flatten_bang): returns nil if nothing changed.
+	  a patch from Marc-Andre Lafortune in [ruby-core:23382].
+
+Thu May  7 14:26:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sample): negative sample number is invalid.
+	  [ruby-core:23374]
+
+Thu May  7 14:16:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c, include/ruby/encoding.h: fixed types.
+
+	* include/ruby/encoding.h (rb_enc_nth): long is used for index.
+
+Thu May  7 14:01:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (r_le): fixed types.
+
+	* range.c (range_eql): fixed rdoc.
+
+Thu May  7 13:10:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): should be rb_pid_t.
+
+Wed May  6 16:50:20 2009  Tanaka Akira  <akr@fsij.org>
+
+	* math.c (math_gamma): use a table for positive small integers.
+
+Wed May  6 09:27:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.ja: code for THINK_C does not exist already.  [Bug #1435]
+
+Wed May  6 05:33:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (CFLAGS, CXXFLAGS): strips extra spaces.
+
+Tue May  5 11:29:07 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/ifchange: Fix: arguments which begin with minus sign may
+	  parsed as options.  Because of older systems, don't use --
+	  but use parentheses.
+
+Tue May  5 10:42:28 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Update to JSON 1.1.4.
+
+Tue May  5 07:22:37 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c: NOMAP is now multibyte direct map.
+
+	* transcode.c: remove ASIS.
+
+	* transcode_data.h: ditto.
+
+	* tool/transcode-tb (ActionMap#generate_info): remove :asis.
+
+	* tool/transcode-tb (ActionMap#generate_info): add :nomap0.
+
+	* enc/trans/utf8_mac.trans: replace :asis by :nomap0.
+
+Sat May  2 22:53:02 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (SortedSet): Fix document. [Bug #1429]
+
+Sat May  2 10:34:29 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf8_mac-tbl.rb: don't use Unicode escape.
+
+	* enc/trans/utf8_mac.trans: follow above.
+
+Sat May  2 09:19:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (Init_Array): made #to_s an alias to #inspect to reduce
+	  the result of recursive array.  a patch from ujihisa at
+	  [ruby-dev:38362]
+
+Fri May  1 16:50:05 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#merge): Only directly use the passed objects
+	  @hash instance variable when self and the passed object are
+	  instances of the same class. [Bug #118]
+
+Fri May  1 16:44:11 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (SortedSet#add): Do not let an incomparable object
+	  in. [Bug #118]
+
+Fri May  1 13:18:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (_longjmp): never return.  see [ruby-core:23241]
+
+Fri May  1 01:31:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/extmk.rb: use RbConfig instead of Config.
+
+	* instruby.rb: ditto.
+
+	* lib/rubygems.rb: ditto.
+
+	* test/rubygems/test_config.rb: ditto.
+
+Thu Apr 30 21:23:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* runruby.rb: use RbConfig::CONFIG instead of Config::CONFIG.
+
+	* spec/default.mspec: ditto.
+
+	* yarvtest/yarvtest.rb: ditto.
+
+	* instruby.rb: ditto.
+
+	* benchmark/report.rb: ditto.
+
+	* benchmark/runc.rb: ditto.
+
+	* tool/eval.rb: ditto.
+
+	* test/rubygems/test_gem.rb: ditto.
+
+	* test/rubygems/test_config.rb: ditto.
+
+	* test/rubygems/test_gem_platform.rb: ditto.
+
+Thu Apr 30 18:18:13 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/http.rb: documentation typo fixed.  [ruby-core:23335]
+
+Thu Apr 30 15:27:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/utf8_mac.trans: get rid of a 1.9 feature for cross
+	  compile.
+
+Thu Apr 30 11:35:30 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encoding): Change filesystem_encoding of
+	  Mac OS X to UTF-8.
+
+Wed Apr 29 21:23:40 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/iso_2022_jp.h: add CP50221.
+
+	* enc/trans/iso2022.trans: add converter for CP50221.
+
+Wed Apr 29 15:22:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_join): recursive array has no meaning as path
+	  name.  [ruby-core:23329]
+
+Tue Apr 28 19:09:45 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/test_timeout.rb (TestTimeout#test_timeout): add a test.
+
+Tue Apr 28 07:13:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pp.rb (Struct#pretty_print): coerce to a string since
+	  anonymous class has name no longer.  [ruby-dev:38349]
+
+Mon Apr 27 16:47:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ruby_version): uses sed instead of grep if the
+	  result is need, to get rid of GREP_OPTIONS.  [ruby-dev:38336]
+
+Mon Apr 27 01:25:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_getc): use rl_getc_function if
+	  possible, to get rid of hang up at EOF without a newline.
+
+Sun Apr 26 23:19:32 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf8_mac.trans: Add converter for UTF8-MAC.
+
+	* enc/trans/utf8_mac-tbl.rb: ditto.
+
+	* test/ruby/test_econv.rb: tests for above.
+
+Sun Apr 26 22:17:02 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/transcode-tb (ActionMap#each_firstbyte):
+	  if :asis collides other mappings, use another.
+
+	* tool/transcode-tb (ActionMap#generate_info):
+	  add :asis for ASIS.
+
+Sun Apr 26 21:59:43 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (ASIS): added for multi byte direct map.
+
+	* transcode.c (transcode_restartable0): ditto.
+
+Sun Apr 26 20:33:12 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/transcode-tb (ActionMap#generate_node):
+	  Use ActionMap#gennode instead of generate_node
+	  because of initialization.
+
+Sun Apr 26 20:21:39 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .gitignore: added.
+
+Sun Apr 26 20:17:24 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rake.rb: use RbConfig::CONFIG instead of Config::CONFIG.
+
+	* lib/rbconfig/datadir.rb: ditto.
+
+Sun Apr 26 19:30:29 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_arg): unused variable removed.
+
+Sun Apr 26 18:35:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_reopen): keeps pathv for prep_stdio.  [ruby-dev:38131]
+
+Sun Apr 26 15:13:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#parse_in_order): do not make an
+	  option from non-option argument.  [ruby-dev:38333]
+
+Sat Apr 25 19:11:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func_daemon): use daemon(3) only on *BSD.
+
+	* process.c (proc_daemon): double fork to ensure not having ctty.
+	  [ruby-core:23305]
+
+Sat Apr 25 16:19:48 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (month_arg): extracted from time_arg.
+	  (validate_vtm): ditto.
+
+Sat Apr 25 16:03:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (TIME_COPY_GMT): new macro.
+	  (time_s_at): use TIME_COPY_GMT.
+	  (time_succ): ditto.
+
+Sat Apr 25 15:54:04 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_mload): use TIME_SET_UTC.
+
+Sat Apr 25 15:47:54 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_get_tm): take time_object instead of gmt.
+
+Sat Apr 25 15:39:44 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_succ): refactored to avoid gmt variable.
+	  (strftimev): use TIME_UTC_P.
+	  (time_strftime): ditto.
+
+Sat Apr 25 15:21:33 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (TIME_UTC_P): new macro.
+	  (TIME_SET_UTC): ditto.
+	  (TIME_LOCALTIME_P): ditto.
+	  (TIME_SET_LOCALTIME): ditto.
+	  (time_utc_p): use the above macro.
+	  (time_localtime): ditto.
+	  (time_localtime): ditto.
+	  (time_gmtime): ditto.
+	  (time_to_s): ditto.
+	  (time_add): ditto.
+	  (time_sec): ditto.
+	  (time_min): ditto.
+	  (time_hour): ditto.
+	  (time_mday): ditto.
+	  (time_mon): ditto.
+	  (time_year): ditto.
+	  (time_wday): ditto.
+	  (wday_p): ditto.
+	  (time_yday): ditto.
+	  (time_isdst): ditto.
+	  (time_zone): ditto.
+	  (time_utc_offset): ditto.
+	  (time_to_a): ditto.
+	  (strftimev): ditto.
+	  (time_strftime): ditto.
+	  (time_mdump): ditto.
+
+Thu Apr 23 01:30:37 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/zlib/zlib.c (Zlib::GzipFile#path): New method.
+
+Wed Apr 22 20:25:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_timespec): check out-of-range.  [ruby-core:23282]
+	  [Bug #1396]
+
+Wed Apr 22 19:33:13 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.parse): use year completion in Date._parse.
+
+Wed Apr 22 11:12:15 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (GMTIME, LOCALTIME): should set result if not have *_r().
+
+	* time.c (localtime_with_gmtoff): now always needed tmbuf.
+
+Wed Apr 22 10:38:47 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* dir.c (glob_helper): C99(gcc)-ism.
+
+	* time.c (find_time_t): GUESS macro needs the variable named ``result''
+	  always.
+
+Wed Apr 22 09:27:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (localtime_with_gmtoff): fixed cross function jump.
+
+Wed Apr 22 03:06:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time#rfc2822): pad leading zeros for year.
+	  (Time#httpdate): ditto.
+	  (Time#xmlschema): ditto.
+
+Wed Apr 22 02:10:48 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time#xmlschema): use subsec instead of nsec.
+
+Wed Apr 22 01:27:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_arg): use the year argument as-is.  [ruby-dev:38194]
+
+	* lib/time.rb (Time.parse): interpret small year 0..99 as 1950..2049.
+
+Wed Apr 22 00:32:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (find_time_t): constified.
+
+Wed Apr 22 00:11:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (leap_year_v_p): removed.
+
+Tue Apr 21 23:52:45 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: remove time_t restriction from Time class.
+
+	* timev.h: new file to define struct vtm.
+
+	* strftime.c: format struct vtm instead of struct tm.
+
+	* ext/syck/rubyext.c (mktime_do): don't use time_t;
+
+	[ruby-dev:38191]
+
+Tue Apr 21 09:25:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_path, dir_each, glob_helper): use readdir_r() if
+	  available.
+
+Tue Apr 21 09:20:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (init_mkmf): needs default library path even if
+	  cross compiling.
+
+Tue Apr 21 07:07:45 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_push): bypass rb_ary_store().
+
+Tue Apr 21 01:25:16 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigsub_int): subtraction without making internal
+	  bignum values.
+
+	* bignum.c (bigadd_int): ditto for addition.
+
+	* bignum.c (bigtrunc): declare inline.
+
+	* bignum.c (rb_quad_pack): fix condition.
+
+Tue Apr 21 01:13:42 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup (config.h): added TIMET2NUM and NUM2TIMET to match
+	  the change in time.c
+
+Mon Apr 20 20:29:04 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_split_m): faster processing on 7bit strings.
+
+	* string.c (ascii_isspace): faster isspace() for 7bit strings.
+
+Sun Apr 19 14:43:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): the order of local variables on stack is
+	  undefined.  should use outermost VALUE for ruby_init_stack.
+
+	* gc.c (ruby_get_stack_grow_direction, Init_stack): allows volatile
+	  pointer.
+
+	* thread_*.c (ruby_init_stack): ditto.
+
+Sun Apr 19 13:17:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gem_prelude.rb (Gem::QuickLoader#push_gem_version_on_load_path):
+	  check for requirement if the gem is installed.  a patch from
+	  Kyosuke MOROHASHI at [ruby-dev:38020].
+
+Sun Apr 19 01:39:17 2009  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (proc_seteuid_m): defined to use rb_f_notimplement if not
+	  implemented.
+	  (proc_setegid_m): ditto.
+
+Sun Apr 19 01:03:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (proc_setuid): use rb_f_notimplement if not implemented.
+	  (proc_setgid): ditto.
+
+Sat Apr 18 23:07:18 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/readline/readline.c: use rb_f_notimplement for methods not
+	  implemented.
+
+	* ext/openssl/ossl_engine.c: ditto.
+
+	* ext/openssl/ossl_config.c: ditto.
+
+	* ext/openssl/ossl_cipher.c: ditto.
+
+	* ext/openssl/ossl_pkcs5.c: ditto.
+
+	* ext/openssl/ossl_x509ext.c: ditto.
+
+	* ext/socket/socket.c: ditto.
+
+	* ext/socket/basicsocket.c: ditto.
+
+	* ext/socket/ancdata.c: ditto.
+
+	* ext/socket/unixsocket.c: ditto.
+
+	* ext/iconv/iconv.c: ditto.
+
+Sat Apr 18 21:07:34 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c: use rb_f_notimplement for methods not
+	  implemented.
+
+Fri Apr 17 01:51:17 2009  Tanaka Akira  <akr@fsij.org>
+
+	* node.h (rb_notimplement_body_p): declared.
+
+	* vm_method.c (Init_eval_method): suppress a warning.
+
+	* io.c (rb_io_fsync): use rb_f_notimplement if not implemented.
+	  (rb_io_close_on_exec_p): ditto.
+	  (rb_io_set_close_on_exec): ditto.
+	  (rb_io_fcntl): ditto.
+	  (rb_f_syscall): ditto.
+
+	* dir.c (dir_tell): ditto.
+	  (dir_seek): ditto.
+	  (dir_s_chroot): ditto.
+
+	* process.c (proc_getpgrp): ditto.
+	  (proc_setpgrp): ditto.
+	  (proc_getpgid): ditto.
+	  (proc_setpgid): ditto.
+	  (proc_setsid): ditto.
+	  (proc_getpriority): ditto.
+	  (proc_setpriority): ditto.
+	  (proc_getrlimit): ditto.
+	  (proc_setrlimit): ditto.
+	  (p_sys_setuid): ditto.
+	  (p_sys_setruid): ditto.
+	  (p_sys_seteuid): ditto.
+	  (p_sys_setreuid): ditto.
+	  (p_sys_setresuid): ditto.
+	  (p_sys_setgid): ditto.
+	  (p_sys_setrgid): ditto.
+	  (p_sys_setegid): ditto.
+	  (p_sys_setregid): ditto.
+	  (p_sys_setreuid): ditto.
+	  (p_sys_setresgid): ditto.
+	  (p_sys_issetugid): ditto.
+	  (proc_getgroups): ditto.
+	  (proc_setgroups): ditto.
+	  (proc_initgroups): ditto.
+	  (proc_daemon): ditto.
+	  (rb_proc_times): ditto.
+
+	* file.c (rb_file_s_lchown): ditto.
+	  (rb_file_s_link): ditto.
+	  (rb_file_s_symlink): ditto.
+	  (rb_file_s_readlink): ditto.
+	  (rb_file_s_truncate): ditto.
+	  (rb_file_truncate): ditto.
+
+Fri Apr 17 00:53:47 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/cgi/core.rb (read_multipart): When path is not defined,
+	  define local_path as a method always returning nil instead of
+	  aliasing.  This is because StringIO#path no longer exists.
+
+Fri Apr 17 00:45:09 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* dir.c (bracket): fix escape handling for range character in bracket
+	  of fnmatch pattern.  e.g., '[a\-c]' should not match 'b'.
+
+Thu Apr 16 23:09:03 2009  Tanaka Akira  <akr@fsij.org>
+
+	* class.c (rb_define_method_id): use rb_define_notimplement_method_id
+	  if rb_f_notimplement is given.
+	  (rb_define_protected_method): ditto.
+	  (rb_define_private_method): ditto.
+	  (rb_define_method): use rb_define_method_id.
+
+	* include/ruby/intern.h (rb_f_notimplement): declared.
+	  (rb_define_notimplement_method_id): declared.
+
+	* proc.c (method_inspect): show not-implemented.
+
+	* vm_method.c (notimplement_body): new variable.
+	  (rb_notimplement_body_p): new function.
+	  (rb_method_boundp): return false if not implemented.
+	  (rb_f_notimplement): new function.
+	  (rb_define_notimplement_method_id): new function.
+
+	* process.c (rb_f_fork): use rb_f_notimplement if not implemented.
+
+	* file.c (rb_file_s_lchmod): use rb_f_notimplement if not implemented.
+
+Wed Apr 15 20:24:49 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_flatten): flatten(0) works as Array#dup.
+	  [ruby-core:23168]
+
+	* test/ruby/test_array.rb: add a test for above.
+
+Wed Apr 15 11:53:35 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (fnmatch_helper): use rb_enc_precise_mbclen and
+	  fail if bytes are invalid. [ruby-dev:38307]
+
+Tue Apr 14 18:11:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (bracket): if same in bytes, path is matching.
+	  [ruby-dev:38305]
+
+Mon Apr 13 17:21:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): scan coderange incrementally.
+
+Mon Apr 13 11:35:55 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): optimize previous commit.
+	  [ruby-list:45954]
+
+Mon Apr 13 10:58:54 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): check encoding compatibility only on
+	  real parts.
+
+Sun Apr 12 19:54:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (bracket, fnmatch_helper): compare bytewise first, to get
+	  rid of invalid byte sequence.  [ruby-dev:38303]
+
+Sat Apr 11 08:45:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (Makefile): phony ruby target needs empty command.
+
+Fri Apr 10 11:32:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (up): updates timestamp file.
+
+Fri Apr 10 04:54:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (Makefile): info-program needs common.mk.
+
+Fri Apr 10 03:46:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_INSTALL_NAME): use --program-transform-name.
+
+	* instruby.rb, mkconfig.rb: deal with --program-transform-name
+	  better.  now supports s, y commands and single addressing.
+
+Thu Apr  9 23:59:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: don't override the rule for ruby.
+
+Wed Apr  8 21:58:12 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (rb_str_dump): buffer length plus one byte for null
+	  terminator.  [ruby-dev:38294]
+
+	* test/ruby/test_m17n.rb (test_str_dump): add a test for above.
+
+Wed Apr  8 20:08:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_rstrip_bang): should not sign-expand non-ascii.
+	  [ruby-core:23158]
+
+Wed Apr  8 17:29:29 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_chop_bang): reset coderange. [ruby-core:23155]
+
+Wed Apr  8 14:00:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (what_type?): fixed typo, and refined for member of
+	  aggregation types.
+
+	* lib/mkmf.rb (Logging.postpone): copy postponed output always.
+
+Wed Apr  8 09:45:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBY_SO): removed redundant additional version
+	  numbers.
+
+Tue Apr  7 13:35:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (info): shows configured names.
+
+	* configure.in (Makefile): works even if RUBY_INSTALL_NAME
+	  contains macro.
+
+	* configure.in (LIBRUBY_DLDFLAGS): compatibility version is
+	  ruby_version.
+
+	* configure.in (RUBY_REPLACE_TYPE): defines type modifier prefix
+	  for printf.
+
+Tue Apr  7 02:27:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_DEFINT): should pass includes to
+	  AC_CHECK_SIZEOF and RUBY_CHECK_SIZEOF.
+
+	* configure.in (CFLAGS, CXXFLAGS): need ARCH_FLAG for universal
+	  binary.
+
+Tue Apr  7 01:08:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_to_s): reduce fragments if no precision lost.
+	  c.f. [ruby-core:23075]
+
+Mon Apr  6 23:16:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (CFLAGS, CXXFLAGS): override with $cflags and
+	  $cxxflags if not given.  [ruby-core:23130]
+
+Mon Apr  6 19:26:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (utime_failed): refined the error message for EINVAL on
+	  DOSISH platforms, where it may fail depending on filesystems.
+	  see [ruby-dev:38277].
+
+Mon Apr  6 16:38:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (sys_fail2, rb_file_s_readlink, BUFCHECK, rmext),
+	  (rb_file_s_basename): get rid of overflow.
+
+Mon Apr  6 15:11:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (int_chr): checks overflow.
+
+Mon Apr  6 10:49:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/stringio/test_stringio.rb (test_path): StringIO#path is no
+	  longer defined.  [ruby-dev:38254]
+
+Mon Apr  6 10:47:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_path): removed.  [ruby-dev:38254]
+
+Sun Apr  5 18:02:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_fd_resize): does nothing on Win32.
+
+Sat Apr  4 17:05:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (RB_NUM_COERCE_FUNCS_NEED_OPID): macro to
+	  check compatibility.  [ruby-dev:38162]
+
+Sat Apr  4 07:38:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb (sitearch): default to arch.
+
+Fri Apr  3 14:02:42 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/irb/completion.rb (IRB::InputCompletor::Operators): Add
+	  overloadable negative operators.
+
+	* lib/irb/ruby-lex.rb (RubyLex#lex_init): Support overloadable
+	  negative operators.
+
+	* lib/irb/ruby-lex.rb (RubyLex#identify_identifier): Minus signs
+	  need to be escaped in regexp character class.
+
+	* misc/ruby-mode.el (ruby-font-lock-keywords, ruby-parse-partial):
+	  Support overloadable negative operators.
+
+Fri Apr  3 12:45:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: quotes arguments with spaces always.
+
+Thu Apr  2 14:50:06 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/getoptlong.rb: remove unused rdoc/usage from example.
+	   [ruby-core:23098]
+
+Thu Apr  2 07:42:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): updated.
+
+Wed Apr  1 20:37:49 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (rb_cv_fork_with_pthread): fail if the child process
+	  fail.
+
+Wed Apr  1 19:46:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): support for parallel make.  a
+	  patch from Takuto Matsuu at [ruby-dev:38220].
+
+Wed Apr  1 19:39:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBY_LDSHARED): use $(CC) instead of cc.
+	  a patch from Wataru Kimura at [ruby-dev:38225].
+
+Wed Apr  1 18:53:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): should not #include inside a
+	  function, since headers may have declarations.
+	  c.f. [ruby-core:23095]
+
+Wed Apr  1 18:44:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: ignores version numbers in config.status.  replaces
+	  all $$s in program_transform_name.
+
+Wed Apr  1 15:12:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (.c.i): use CPP instead of CC, since gcc -E can'
+	  work with multiple -arch options.
+
+Wed Apr  1 13:46:20 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (rb_thread_fd_select): new function to call select
+	  using rb_fdset_t.
+
+	* io.c (select_internal): use rb_thread_fd_select instead of
+	  rb_thread_select.  based on the patch from Kengo Matsuyama.
+	  [ruby-dev:38221]
+
+Wed Apr  1 13:16:19 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (rb_f_sleep): RDoc disambiguation.  [ruby-talk:332632]
+
+Tue Mar 31 15:17:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: fixed the help strings for the header and library
+	  dir switches, and updated to use AS_HELP_STRING.  patches from
+	  Richard Brown, c.f. [ruby-core:23067].
+
+Mon Mar 31 08:18:57 2009  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* test/csv/test_interface.rb, test/csv/test_serialization.rb:
+	  Trying more fixes some failing tests on Windows.
+
+Mon Mar 30 19:04:25 2009  Tanaka Akira  <akr@fsij.org>
+
+	* .gdbinit (rp): show negative fixnum correctly.
+
+Mon Mar 30 12:12:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (eval_string_with_cref): replaces the message if
+	  frozen.  [ruby-dev:38208]
+
+Mon Mar 30 01:00:20 2009  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* test/csv/test_interface.rb: Trying a fix for some failing tests
+	  on Windows.
+
+Sun Mar 29 08:59:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pathname.rb (Pathname#relative_path_from): compares path
+	  components according to system default case-sensitiveness.
+	  [ruby-core:22829]
+
+Sat Mar 28 11:10:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ruby.imp): all symbols in static library need to be
+	  exported, so that encoding-extensions can be loaded.  based on a
+	  patch from Yutaka Kanemoto <kinpoco AT gmail.com> in
+	  [ruby-talk:332282].
+
+Sat Mar 28 08:49:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): checks if named argument given twice.
+
+	* sprintf.c (GETNAMEARG): remembers named arg is used, to get rid
+	  of too many arguments warning.
+
+Sat Mar 28 03:16:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (link_command, cc_command, cpp_command): should no
+	  override extout defined in extmk.rb.
+
+Fri Mar 27 12:56:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (what_type?): checks more restrictively, and
+	  supports universal binary.
+
+Fri Mar 27 01:33:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_cmp): negate the result of reverse comparison.
+
+Fri Mar 27 01:19:50 2009  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_cmpint): FIX2INT may fail on LP64 platforms.
+
+Thu Mar 26 12:22:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb (String#is_binary_data?): TAB would be
+	  usually considered to be included in text data.
+
+	* lib/rdoc/parser.rb (RDoc::Parser.binary?): blksize may be nil
+	  and is irrelevant to whether a file is binary.  copied from
+	  above since TAB and newlines would be usually considered to be
+	  included in text data.
+
+Thu Mar 26 11:33:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/paths.rb (RDoc::RI::Paths): considers
+	  --program-prefix and --program-suffix.  reapplied r19923.
+
+Wed Mar 25 07:45:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (target_alias): replaces cpu with universal too.
+
+	* configure.in (target): replaces cpu with arch by --with-arch.
+
+	* mkconfig.rb (TOPDIR): chops arch in config instead of
+	  RUBY_PLATFORM which varies on universal_binary.
+
+	* mkconfig.rb (MAJOR, MINOR, TEENY): reads from version.h always.
+
+Tue Mar 24 19:23:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): need to include $4.
+
+Tue Mar 24 17:08:52 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* io.c (rb_io_inspect): Cannot access fptr->fd if fptr is NULL.
+	  This fixes a coredump caused by: ruby -e "class X < IO; def
+	  initialize; end; end; p X.new.inspect"
+
+Mon Mar 23 22:01:00 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_timespec): use NUM2TIMET.
+	  (time_s_at): ditto.
+
+Mon Mar 23 21:52:26 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* process.c (Init_process): Better patch for eliminating an
+	  "unused variable".
+
+Mon Mar 23 21:41:14 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/dbm/dbm.c (fdbm_initialize): Make the file variable
+	  volatile, because FilePathValue() currently does not protect the
+	  given variable from GC. (Probably it should)
+
+	* ext/sdbm/init.c (fsdbm_initialize): Ditto.
+
+Mon Mar 23 19:22:14 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* file.c (rb_file_s_lstat): Back out.
+
+Mon Mar 23 18:54:57 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* file.c (rb_file_s_stat, rb_file_s_lstat): Remove repeated type
+	  checks.
+
+Mon Mar 23 14:57:48 2009  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (init_heap): set default to heap slots length
+	  if HEAP_OBJ_LIMIT is larger than HEAP_MIN_SLOTS. [Bug #1310]
+	  (set_heaps_increment): increment next_heaps_length if
+	  next_heaps_length and heaps_used are same.
+
+Mon Mar 23 14:32:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn): use original command if not found.
+
+Mon Mar 23 06:51:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (link_so): replaces $(TARGET) with basename of the
+	  target.  [ruby-talk:330286]
+
+Sun Mar 22 14:51:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn): use NULL as
+	  application name for batch files.
+
+Sat Mar 21 15:54:41 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ssl.c (write_would_block): defined.
+	  (read_would_block): defined.
+	  (ossl_start_ssl): add nonblock argument.
+	  (ossl_ssl_connect): follow ossl_start_ssl change.
+	  (ossl_ssl_connect_nonblock): new method.
+	  (ossl_ssl_accept): follow ossl_start_ssl change.
+	  (ossl_ssl_accept_nonblock): new method.
+	  (ossl_ssl_read_internal): use write_would_block and
+	  read_would_block.
+	  (ossl_ssl_write_internal): ditto.
+
+Sat Mar 21 08:19:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_exec): prints error message only on platforms
+	  neither close-on-exec nor spawnv is supported.
+
+Sat Mar 21 08:17:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn): omit program name
+	  if actual program path is found.  [ruby-core:22960]
+
+Sat Mar 21 07:25:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): should pass program name even if multiple
+	  arguments are given.  fix for TestProcess::test_argv0.
+
+Sat Mar 21 02:37:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/lib/openssl/buffering.rb
+	  (OpenSSL::Buffering#write_nonblock): new method.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_write_nonblock): new method.
+	  (ossl_ssl_write_internal): defined.
+	  (ossl_ssl_write): use ossl_ssl_write_internal.
+
+Fri Mar 20 18:25:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (errmap): added ERROR_MOD_NOT_FOUND.
+
+Fri Mar 20 09:22:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (libprefix): must not append "lib" after $(libdir),
+	  when load_relative is not used on darwin.  a patch from Kenta
+	  Murata at [ruby-dev:38182].
+
+Thu Mar 19 20:29:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_mWaitReadable): defined.
+	  (rb_mWaitWritable): defined.
+	  (io_getpartial): extend IO::WaitReadable on EWOULDBLOCK and EAGAIN.
+	  (rb_io_write_nonblock): extend IO::WaitWritable on EWOULDBLOCK and
+	  EAGAIN.
+
+	* error.c (make_errno_exc): extracted from rb_sys_fail.
+	  (rb_mod_sys_fail): new function.
+
+	* include/ruby/ruby.h (rb_mod_sys_fail): declared.
+	  (rb_mWaitReadable): declared.
+	  (rb_mWaitWritable): declared.
+
+	* ext/socket/init.c (rsock_s_recvfrom_nonblock): extend
+	  IO::WaitReadable on EWOULDBLOCK and EAGAIN.
+	  (rsock_s_accept_nonblock): extend IO::WaitReadable on EWOULDBLOCK,
+	  EAGAIN, ECONNABORTED and EPROTO.
+
+	* ext/socket/socket.c (sock_connect_nonblock): extend IO::WaitWritable
+	  on EINPROGRESS.
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): extend
+	  IO::WaitWritable on EWOULDBLOCK and EAGAIN.
+	  (bsock_recvmsg_internal): extend IO::WaitReadable on EWOULDBLOCK and
+	  EAGAIN.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read_internal): raise SSLError
+	  extended by IO::WaitReadable/IO::WaitWritable on
+	  SSL_ERROR_WANT_READ/SSL_ERROR_WANT_WRITE.
+
+	* ext/openssl/ossl.c (ossl_make_error): extracted from ossl_raise.
+	  (ossl_exc_new): new function.
+
+	* ext/openssl/ossl.h (ossl_exc_new): declared.
+
+	* lib/net/protocol.rb (rbuf_fill): rescue IO::WaitReadable and
+	  IO::WaitWritable.
+
+	  [ruby-core:22539], [ruby-dev:38140]
+
+Thu Mar 19 18:49:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): allows qualified name.
+
+	* configure.in (RUBY_REPLACE_TYPE): checks more strictly.
+
+	* configure.in (struct stat.st_size, struct stat.st_blocks),
+	  (struct stat.st_ino): check for size.
+
+	* lib/mkmf.rb (check_sizeof): allows qualified name.
+
+	* file.c (rb_stat_ino, rb_stat_blocks): check by size.
+
+Wed Mar 18 16:59:48 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/test_syslog.rb (TestSyslog#test_open): check
+	  param after block again. because detect not to call block.
+
+Wed Mar 18 16:54:04 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* ext/socket/ancdata.c: not use pktinfo.ipi_spec_dst if not defined
+	* ext/socket/extconf.rb: define HAVE_IPI_SPEC_DST
+
+Wed Mar 18 16:38:11 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/test_syslog.rb (TestSyslog#test_open): check
+	  block parameter in block. [ruby-dev:38180]
+
+Wed Mar 18 12:48:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (report_bug): rb_bug can be caused by extension
+	  libraries.
+
+Wed Mar 18 02:41:33 2009  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (open_server_inaddr_any): fixed multiple network
+	  families problem. a patch from Charl Matthee at [ruby-core:21033].
+
+Tue Mar 17 21:42:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): creates .revision.time and passes
+	  CHDIR to create prerequisite files.
+
+Tue Mar 17 18:00:55 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (Enumerator#{each_,}{with_index,with_object}): Fix
+	  a bug where any parameter but the first one is dropped even if
+	  multiple values are yielded with. [Bug #1198]
+
+Tue Mar 17 14:25:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pathname.rb (Pathname#sub): set $~ in block.binding.
+	  [ruby-dev:38173]
+
+Tue Mar 17 13:48:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): added RUBY_COREDLL.
+
+	* ext/dl/handle.c (rb_dlhandle_initialize): returns msvcrt if libc
+	  or RUBY_COREDLL is given.  [ruby-core:22828]
+
+Tue Mar 17 10:29:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c, dln.c, parse.y, re.c, ruby.c, sprintf.c, strftime.c,
+	  string.c, util.c, variable.c: use strlcpy, memcpy and snprintf
+	  instead of strcpy, strncpy and sprintf.
+
+Mon Mar 16 17:15:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/dl/test_win32.rb (Win32API): enclosed by DL::TestWin32.
+	  [ruby-core:22827]
+
+Mon Mar 16 16:37:22 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c (eventloop_sleep, lib_eventloop_core),
+	  (lib_watchdog_core): tv_usec is not time_t.
+
+Mon Mar 16 12:30:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (REVISION_H): keeps timestamp of revision.h.
+	  [ruby-core:22900]
+
+	* tool/ifchange, win32/ifchange.bat: extended --timestamp option.
+
+Mon Mar 16 09:28:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.mk, win32/Makefile.sub: added preprocessing rules.
+
+Sun Mar 15 12:34:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/minitest/test_mini_test.rb: fixed tests depending on the
+	  detail of floating point representation.
+
+Sun Mar 15 12:01:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/installer.rb (Gem::Installer#shebang): fix for env
+	  shebang.
+
+Sun Mar 15 11:15:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h ({RSTRING,RBIGNUM}_EMBED_LEN_MAX): made int.
+
+	* include/ruby/ruby.h (OBJ_{TAINTED,UNTRUSTED,FROZEN}): return int.
+
+	* include/ruby/encoding.h (ENC_CODERANGE): ditto.
+
+Sun Mar 15 11:11:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_hash_uint, rb_hash_start, rb_hash_end): use VALUE
+	  rather than unsigned int.
+
+Sun Mar 15 11:00:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): stop the timer thread before exec.
+
+Sun Mar 15 09:17:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (.y.c): use SRC_FILE which contains slashes instead of
+	  backslashes.  [ruby-core:22891]
+
+Sun Mar 15 08:26:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (srcs-ext): creates ext/dl/callback/callback.c also.
+
+Sun Mar 15 03:29:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (SAVED_GROUP_ID, p_gid_switch): should be rb_gid_t.
+
+Sun Mar 15 02:53:13 2009  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_subpat): accept capture name.
+	  (rb_str_aref): follow above change.
+	  (rb_str_aref_m): pass the 2nd argument to rb_str_subpat.
+	  (rb_str_subpat_set): accept capture name.
+	  (rb_str_aset): follow above change.
+	  (rb_str_partition): ditto.
+	  (rb_str_aset_m): pass the 2nd argument to rb_str_subpat_set.
+
+	* include/ruby/intern.h (rb_reg_backref_number): declared.
+
+	* re.c (rb_reg_backref_number): defined.
+
+	  [ruby-core:21057]
+
+Sun Mar 15 02:09:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (bmcall): should not uninitialized variable.  a patch from
+	  pegacorn at [ruby-dev:38169].
+
+Sat Mar 14 18:25:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_scan_oct, ruby_scan_hex): use size_t.
+
+Sat Mar 14 18:18:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_proc_call, rb_node_arity, bmcall, curry): checks
+	  overflow.
+
+	* proc.c (rb_proc_parameters): unnamed_parameters() expects in
+	  not VALUE.
+
+Sat Mar 14 17:54:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_each_words): assume no string exceeds INT_MAX.
+
+Sat Mar 14 15:59:04 2009  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_spawn_internal): use int variable for status.
+
+Sat Mar 14 14:45:51 2009  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_f_system): use rb_pid_t for pid.
+	  (rb_spawn_internal): local variable renamed.
+
+Sat Mar 14 14:16:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strdup, Balloc, rv_alloc): use size_t.
+
+Sat Mar 14 13:53:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_qsort): the result of cmp must be signed, so ge
+	  rid of reuse of a variable.
+
+Sat Mar 14 10:56:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (init_funcname_len, dln_find_exe_r, dln_find_file_r): use
+	  size_t.
+
+	* file.c (rb_stat_inspect, file_expand_path): ditto.
+
+	* util.c (ruby_qsort): ditto.
+
+Sat Mar 14 10:39:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (CXX_EXT): checks for case-sensitive filesystem with
+	  FNM_SYSCASE rather than build_os.
+
+Fri Mar 13 23:03:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (wait_connectable0): remove unreachable code.
+
+Fri Mar 13 23:00:02 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): pass [include] to AC_CHECK_SIZEOF.
+
+Fri Mar 13 20:58:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* dln.c (dln_find_1): compare fspace in size_t world.
+
+Fri Mar 13 18:58:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (CFLAGS, CXXFLAGS): moved after warnflags.
+
+Fri Mar 13 18:10:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_throw_obj): inverted call flow.  [ruby-core:22872]
+
+Fri Mar 13 17:04:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (bool): not define to get rid of conflict
+	  against curses.
+
+Fri Mar 13 16:45:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/openssl_missing.h (i2d_of_void): cast for callbacks.
+	  [ruby-core:22860]
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_by_id): suppress a
+	  warning.
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_flush_sessions): time_t may
+	  be larger than long.
+
+	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_get_time),
+	  (ossl_ssl_session_get_timeout): use TIMET2NUM() to convert
+	  time_t.
+
+Fri Mar 13 15:10:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_set_value): should use
+	  OPENSSL_free instead of free.  a patch from Charlie Savage at
+	  [ruby-core:22858].
+
+Fri Mar 13 21:11:51 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/sdbm/_sdbm.c: should include "ruby/defines.h" as well for
+	  prototypes of compatibility functions on Win32 platform.
+	  [ruby-core:22870]
+
+Fri Mar 13 10:42:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (AC_HEADER_STDBOOL): added.
+
+	* include/ruby/ruby.h (NUM2INT, rb_special_const_p): returns true
+	  and false instead of Qtrue and Qfalse for platforms where VALUE
+	  is bigger than int.
+
+	* gc.c (gc_stress_set), ext/openssl/ossl_asn1.c (decode_bool): go
+	  rid of variables named `bool'.
+
+Fri Mar 13 10:16:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (struct stat.st_size): may be huge.
+
+Fri Mar 13 09:30:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (bigfixize): zero length Bignum is 0.
+
+Fri Mar 13 09:17:12 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils#fu_get_gid): stringify group
+	  argument before making regexp match.  [ruby-dev:38155]
+
+Fri Mar 13 08:06:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (rv_strdup): macro to duplicate nul-terminated string.
+	  [ruby-core:22852]
+
+Thu Mar 12 22:41:41 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl: suppress warnings.
+
+	* ext/openssl/ossl.h (OSSL_Debug): don't use gcc extension for
+	  variadic macro.
+
+Thu Mar 12 22:29:36 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (WARNFLAGS): warning 4996 is only in VC++8 or
+	  later.
+
+Thu Mar 12 22:14:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_def_const): use INT2NUM because
+	  OP_NETSCAPE_DEMO_CIPHER_CHANGE_BUG doesn't fit into Fixnum.
+
+Thu Mar 12 18:16:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c, bignum.c, dln.c, error.c, gc.c, io.c, marshal.c,
+	  numeric.c, pack.c, strftime.c, string.c, thread.c, transcode.c,
+	  transcode_data.h, util.c, variable.c, vm_dump.c,
+	  include/ruby/encoding.h, missing/crypt.c, missing/vsnprintf.c:
+	  suppress VC type warnings.  [ruby-core:22726]
+
+	* marshal.c (div0), numeric.c (infinite_value): new functions to
+	  get rid of VC division by 0 warnings.
+
+	* st.c: use st_index_t for indexes instead of int.
+
+	* vm.c (rb_vm_get_sourceline), vm_insnhelper.c (vm_throw): use
+	  rb_num_t.
+
+Thu Mar 12 09:30:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): if same size type is found, no
+	  more calculation is needed.
+
+	* configure.in (RUBY_DEFINT): falls back to RUBY_CHECK_SIZEOF if
+	  size is not immediate.
+
+Thu Mar 12 09:24:24 2009  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h: suppress warnings of overflow.
+
+Thu Mar 12 09:15:14 2009  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h: suppress warnings of overflow.
+
+Thu Mar 12 05:49:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (warnflags): added some default flags,
+	  pointer-arith, write-strings and shorten-64-to-32, but suppress
+	  unused-parameter and missing-field-initializers, if possible.
+
+	* configure.in (RUBY_CHECK_HUGE): checks whether a value range is
+	  larger than long.
+
+	* file.c (rb_stat_blocks): struct stat.st_blocks may be larger
+	  than long.
+
+	* io.c (copy_stream_fallback_body): off_t may be larger than long.
+
+	* iseq.c (rb_iseq_disasm): RSTRING_LEN() returns long.
+
+	* process.c (rb_f_sleep): time() needs time_t.
+
+	* random.c (fill_random_seed): enclosed conditionally used
+	  variables.
+
+	* thread_pthread.c (ruby_init_stack): range of rlim_cur may be
+	  larger than int.
+
+	* missing/vsnprintf.c (BSD_vfprintf): commented out code which has
+	  no effect at all.
+
+Thu Mar 12 02:10:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: strip spaces from ruby_version.
+	  Sun C 5.9 SunOS_i386 Build47_dlight 2007/05/22 generates
+	  `ruby_version= 1 "." 9 "." 1'.
+
+Thu Mar 12 01:18:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_seek): use long for seekdir().
+
+Thu Mar 12 01:08:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/ytab.sed: replaces backslashes with slash for nmake.
+
+Wed Mar 11 21:23:06 2009  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_mload): don't clear tm_mday.
+
+Wed Mar 11 20:13:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/handle.c (rb_dlhandle_sym): RTLD_NEXT is not for symbol
+	  name.   [ruby-dev:38150]
+
+Wed Mar 11 17:26:38 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/win32/lib/Win32API.rb: call by :stdcall as default.
+	  [ruby-core:22826]
+
+Wed Mar 11 13:27:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stack_type): uses VALUE which is able to be stored
+	  parser stack.
+
+Wed Mar 11 13:22:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (init_env): set TMPDIR if none of TMPDIR, TMP,
+	  TEMP is set.
+
+	* win32/win32.c (rb_w32_telldir, rb_w32_seekdir): should use long,
+	  as defined by SUSv3.
+
+Wed Mar 11 13:03:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/handle.c (rb_dlhandle_sym): fixed heap corruption.
+	  [ruby-core:22822]
+
+Wed Mar 11 06:12:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): expands libpath and removes
+	  last /lib.
+
+Wed Mar 11 04:47:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_to_i, time_hash): time_t may be bigger than long
+	  and int.
+
+	* time.c (time_timeval, rb_time_timeval, obj2nsec, time_strftime),
+	  (time_mdump, time_mload): suppress warnings.
+
+	* win32/Makefile.sub (config.h): added TIMET2NUM and NUM2TIMET.
+
+Wed Mar 11 04:29:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c: suppress warnings.  based on a patch from Charlie
+	  Savage at [ruby-core:22804].
+
+Wed Mar 11 04:22:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (sig_trap): suppress warnings.
+
+Tue Mar 10 20:07:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (WARNFLAGS): uses -wd4996 instead of
+	  CRTDEFFLAGS.  cf. [ruby-core:22725]
+
+Tue Mar 10 19:58:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (strcasecmp, strncasecmp): use _ prefixed
+	  versions to suppress warnings.
+
+	* ext/dl/callback/mkcallback.rb (DLTYPE[FLOAT]): cast to suppress
+	  warnings.  [ruby-core:22792]
+
+	* array.c, bignum.c, gc.c, numeric.c, string.c, util.c, insns.def,
+	  missing/crypt.c, missing/vsnprintf.c, : suppress warnings.
+	  [ruby-core:22792]
+
+Tue Mar 10 13:45:54 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (LDFLAGS): moved -link to TRY_LINK, LINK_SO
+	  and so on.  based on a patch by Charlie Savage at
+	  [ruby-core:22794]
+
+	* win32/Makefile.sub (config.h): passes LDFLAGS to rbconfig.rb so
+	  that extconf.rb could refer it.  [ruby-core:22725]
+
+Tue Mar 10 11:36:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (prepare_iseq_build): too few arguments to function
+	  rb_ary_tmp_new().
+
+Tue Mar 10 10:53:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_compile_with_option): argument may be converted.
+
+Tue Mar 10 04:56:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (MINIRUBY): keep macro into Makefile.
+
+	* configure.in (RUBY_CHECK_SIZEOF): fix for fallback.
+
+	* configure.in (MANGLED_PATH): defines for RUBYLIB_PREFIX.
+
+	* ruby.c (ruby_init_loadpath_safe): uses string as buffer.
+
+	* symbian/setup (config_h): defines MANGLED_PATH.
+
+Tue Mar 10 03:48:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/callback/depend: fix for parallel build.
+
+	* ext/dl/callback/extconf.rb: callback.h is no longer created.
+
+	* ext/dl/callback/mkcallback.rb: creates main source first.
+
+	* ext/dl/callback/extconf.rb ($distcleanfiles): added callback.c.
+
+Tue Mar 10 00:06:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (load_relative): new option to enable load path
+	  relative to libruby_so.
+
+	* ruby.c (ruby_init_loadpath_safe): added the case using dladdr().
+
+Mon Mar  9 16:49:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake: updated to rake code to rake-0.8.4 source code base.
+
+Mon Mar  9 16:03:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/installer.rb (Gem::Installer#initialize): env may
+	  not be under /usr/bin.
+
+	* lib/rubygems/installer.rb (Gem::Installer#shebang): uses /bin/sh
+	  if shebang has any options, since env shebang trick does no
+	  work with options.
+
+Mon Mar  9 15:19:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (manpages): use basename to compare and for Tempfile.
+
+Mon Mar  9 14:55:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (default gems): installs default gemspecs.
+
+Mon Mar  9 14:48:08 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (OPTFLAGS): simplified. see [ruby-core:22725]
+
+Mon Mar  9 14:33:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb: prints names of the directories where files are
+	  installed to.
+
+Mon Mar  9 02:25:54 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_proc.rb: add some tests.
+
+Mon Mar  9 01:38:00 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_enum.rb: add some tests.
+
+Mon Mar  9 01:12:37 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_object.rb: add a test for Object#method_missing.
+
+Mon Mar  9 01:11:17 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_file_exhaustive.rb: add tests for File#size and
+	  File.absolute_path.
+
+Sun Mar  8 23:02:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (LDFLAGS): always prepends -link.
+
+Sun Mar  8 11:39:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF): dirty hack to compute multiple
+	  sizeof, for apple's universal binary.
+
+Sat Mar  7 22:06:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (universal_binary): checks if sizes are same on all
+	  architectures.
+
+Sat Mar  7 18:45:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: checks if cd really succeeded.
+
+Sat Mar  7 14:24:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: don't use CONFIG_SHELL for testing shell feature in
+	  Makefile.
+
+Sat Mar  7 14:07:31 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: fix the location of cdcmd.
+
+Sat Mar  7 13:52:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: the shell which runs configure may be different from
+	  the shell for make.  configure likes bash but Makefile specifies
+	  SHELL=/bin/sh.
+
+Sat Mar  7 12:31:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/erb/test_erb.rb: reverted space sensitive test results.
+
+	* test/rdoc/test_rdoc_parser_c.rb: ditto.
+
+	* test/rdoc/test_rdoc_ri_default_display.rb: ditto.
+
+Fri Mar  6 23:35:59 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (SecureRandom.urlsafe_base64): add optional
+	  argument to add padding.
+
+Fri Mar  6 19:25:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_LIB_VERSION_STYLE): sets full or minor style
+	  not to show RUBY_VERSION_MAJOR and so on.
+
+	* version.h (RUBY_LIB_VERSION): makes from RUBY_LIB_VERSION_STYLE.
+
+	* configure.in, Makefile.in, win32/Makefile.sub (CHDIR): cd using
+	  physical directory.
+
+	* Makefile.in (update-mspec): split from update-rubyspec.
+
+	* win32/Makefile.sub (update-{rubyspec,mspec}): added.
+
+Fri Mar  6 14:28:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub (INSNS): moved variable
+	  definition before rule.
+
+Fri Mar  6 13:47:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake/loaders/makefile.rb (Rake::MakefileLoader#load): deals with
+	  escaped spaces.  incorporated from rake 0.8.4.
+
+	* lib/rake/testtask.rb (Rake::TestTask#define): passes each libs
+	  as each arguments with expanded.  incorporated from rake 0.8.4.
+
+Thu Mar  5 18:36:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_to_s): keeps enough precision for round trip.
+	  [ruby-core:22325]
+
+Thu Mar  5 17:42:06 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_fill): doc for Array#fill misses indication
+	  about negative value for the start argument.  [ruby-core:22497]
+
+Thu Mar  5 16:56:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tmpdir.rb (Dir.tmpdir): not use USERPROFILE, and ignores
+	  non-existent directory.
+
+Thu Mar  5 14:52:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_sysinit): initializes version info first.
+
+	* win32/win32.c (rb_w32_osid): always define for binary compatibility.
+
+Thu Mar  5 11:52:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (make_insns_rules): rule to make rules for insns.
+
+Thu Mar  5 11:27:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (miniruby.exe): workaround for a failure of
+	  mt.exe on Windows 7 beta.  [ruby-talk:330284]
+
+Thu Mar  5 10:32:17 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/dl/test_base.rb: Add dragonfly to libc and libm switch.
+
+Thu Mar  5 10:29:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_eql): extracted from rb_str_equal and rb_str_eql.
+
+	* string.c (rb_str_chomp_bang): keeps 7bit coderange.
+
+Thu Mar  5 10:10:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (enc_initialized_p): ENC_UNINITIALIZED
+	  is not defined already.
+
+	* include/ruby/encoding.h (ENC_FROM_ENCINDEX): removed because
+	  depending on a static variable.
+
+Thu Mar  5 04:42:52 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb: fix wrong test, and add a test for
+	  sort_by!.
+
+Thu Mar  5 04:32:38 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb: add some tests for coverage.
+
+Thu Mar  5 00:06:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/README.win32 (Requirement): added unicows.lib and dll.
+
+Wed Mar  4 23:29:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, win32/configure.bat (--enable-win95): new option
+	  to enable/disable win95 support.
+
+Wed Mar  4 21:21:51 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/thread.rb (SizedQueue#push): fix limit condition.
+	  [ruby-dev:38135]
+
+Wed Mar  4 20:27:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (mingw): checks if unicows.lib is available.
+
+	* include/ruby/win32.h (WIN95): moved to config.h
+
+	* win32/Makefile.sub (LIBS): links unicows.lib if available.
+
+	* win32/Makefile.sub (config.h): defines WIN95 only if unicows.lib
+	  is available.
+
+	* win32/setup.mak (-unicows-): checks if unicows.lib is available.
+
+Tue Mar  3 17:10:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): removes object files no longer used and
+	  files to be removed by old distclean.
+
+Tue Mar  3 16:51:51 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/resolv.rb: get rid of warning.
+
+	* ext/dl/lib/dl/func.rb: ditto.
+
+Tue Mar  3 16:25:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake), lib/mkmf.rb (create_makefile): $objs and
+	  $srcs are always Array or nil now.
+
+Tue Mar  3 15:54:11 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/{registry,resolv}.rb: use dl/import instead
+	  of Win32API.
+
+Tue Mar  3 15:53:20 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/lib/dl/func.rb (DL::Function.name): delegate to @cfunc.
+
+Tue Mar  3 15:40:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/dl.c (rb_dl_init_callbacks): autoloads callbacks.
+
+	* ext/dl/callback: split from dl.
+
+	* ext/dl/lib/dl/cparser.rb (DL::CParser::parse_signature): fixed
+	  variable name.
+
+	* ext/dl/lib/dl/pack.rb (DL::PackInfo): reduced redundant
+	  initialization.
+
+	* ext/dl/lib/dl/stack.rb (DL::Stack): ditto.
+
+	* ext/dl/lib/dl/value.rb (DL::ValueUtil#wrap_arg): NULL for nil.
+
+	* test/dl: moved from ext/dl/test.
+
+Tue Mar  3 14:57:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (main): passes $(MAKE) to mkmain_cmd
+
+	* ext/extmk.rb (command_output): uses arguments to invoke make.
+
+Tue Mar  3 01:56:03 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (SecureRandom.urlsafe_base64): delete padding.
+
+Mon Mar  2 21:39:08 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/sockport.h (SS_LEN): defined unconditionally.
+
+Mon Mar  2 21:26:42 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (anc_inspect_socket_creds): suppress warnings.
+
+Mon Mar  2 21:13:42 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (rsock_discard_cmsg_resource): defined
+	  unconditionally.
+
+Mon Mar  2 21:06:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_mload): don't use "sun" variable
+	  for Solaris.
+
+Mon Mar  2 21:01:36 2009  Tanaka Akira  <akr@fsij.org>
+
+	* file.c: include fcntl.h for O_RDONLY on Solaris.
+
+Mon Mar  2 19:43:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/uri/common.rb (URI::Parser#escape): escaped string is ascii
+	  only.
+
+	* lib/uri/common.rb (URI::Parser#unescape): converts bytewise.
+	  [ruby-dev:38005], [ruby-dev:38110]
+
+Mon Mar  2 16:40:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja (Macros): added HAVE_RUBY_*_H.
+
+Mon Mar  2 12:22:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (init_env): keeps wide chars as wide chars.
+
+Mon Mar  2 11:01:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/dl.h (dlerror): fixed on Windows.
+
+	* ext/dl/handle.c (rb_dlhandle_initialize): returns handle to
+	  libruby if nil is given on Windows as dlopen.
+
+	* ext/dl/handle.c (rb_dlhandle_sym): get rid of buffer overrun,
+	  and try ASCII WINAPI versions too.
+
+Mon Mar  2 10:29:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_longjmp), include/ruby/intern.h (ruby_cleanup): fixed
+	  prototypes.
+
+Mon Mar  2 00:56:45 2009  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/observer.rb: The module observable uses now a hash to
+	  store listeners. merged a patch from Alessandro Di Maria in
+	  [ruby-core:22560].
+
+Sun Mar  1 22:59:41 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c: suppress warnings.
+
+	* ext/socket/socket.c: ditto.
+
+	* ext/socket/unixsocket.c: ditto.
+
+Sun Mar  1 20:57:41 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ (rsock_getfamily): renamed from rb_sock_getfamily.
+
+Sun Mar  1 19:52:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/win32/lib/Win32API.rb (initialize): passes res
+	  arguments to DL.
+
+Sun Mar  1 19:38:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (zstream_run): suppress a warning.  a patch from
+	  Charlie Savage at [ruby-core:22607].
+
+Sun Mar  1 19:36:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_dtoa): allocates one more byte to get rid of buffer
+	  overrun.  a patch from Charlie Savage at [ruby-core:22604].
+
+Sun Mar  1 19:10:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (.y.c): removed extra backslash.  [ruby-core:22602]
+
+Sun Mar  1 19:02:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): fix for stdcall and missing
+	  argument numbers.  [ruby-core:22601]
+
+	* ext/dl/dl.h (DLSTACK_PROTO0_): added.
+
+	* ext/dl/mkcallback.rb (gencallback, gen_callback_file),
+	  (rb_dl_init_callbacks): omit stdcall functions unless supported.
+
+	* lib/rubygems/config_file.rb (SHGetFolderPath): stdcall.
+
+Sun Mar  1 17:27:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (gfDebug): uncommented out.
+	  [ruby-core:22600]
+
+Sun Mar  1 16:15:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_waitpid): use wait_each() on no waitpid platforms.
+	  [ruby-dev:38054]
+
+Sun Mar  1 16:01:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_daemon): stops and restarts timer thread because
+	  daemon(3) implies fork(2).  [ruby-dev:38055]
+
+Sun Mar  1 15:45:49 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ (rsock_make_hostent): renamed from sock_make_hostent.
+	  (rsock_addrinfo): renamed from sock_addrinfo.
+	  (rsock_getaddrinfo): renamed from sock_getaddrinfo.
+	  (rsock_socket): renamed from ruby_socket.
+	  (rsock_sock_s_socketpair): renamed from sock_s_socketpair.
+	  (rsock_connect): renamed from ruby_connect.
+
+	* ext/socket/socket.c (sock_listen): make it static.
+
+Sun Mar  1 15:29:31 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: add rsock_ prefix.
+
+Sat Feb 28 21:52:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* vm_eval.c (rb_iterate): use volatile to suppress warnings.
+
+	* eval.c (ruby_cleanup): ditto.
+
+Sat Feb 28 20:09:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* eval_error.c (error_print): use volatile to suppress warnings.
+
+	* vm_eval.c (eval_string_with_cref): ditto.
+
+	* thread.c (rb_exec_recursive): ditto.
+
+	* eval_jump.c (rb_exec_end_proc): ditto.
+
+	* eval.c (ruby_exec_node): ditto.
+	  (rb_longjmp): ditto.
+
+Sat Feb 28 18:28:58 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_unix_rights): check message type.
+
+Sat Feb 28 16:18:39 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/net/http.rb: suppress warnings of non-existing instance variable
+	  access for SSL.
+
+Sat Feb 28 14:50:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (BasicSocket#connect_address): returns ::1
+	  if the address is IPv4-mapped IPv6 address for 0.0.0.0 for MacOS X.
+
+Sat Feb 28 00:18:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c: suppress strict-aliasing warning with
+	  gcc version 4.4.0 20090219 (Red Hat 4.4.0-0.21) on Fedora 11 Alpha.
+
+	* ext/socket/socket.c: ditto.
+
+	* ext/socket/unixsocket.c: ditto.
+
+	* ext/socket/sockport.h (SS_LEN): defined.
+	  (SET_SS_LEN): ditto.
+
+Fri Feb 27 22:30:18 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): close FDs passed by
+	  SCM_RIGHTS unless :scm_rights=>true is given.
+	  (discard_cmsg): extracted from rsock_discard_cmsg_resource.
+
+Fri Feb 27 22:14:22 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/lib/openssl/buffering.rb: define Buffering module under
+	  OpenSSL.  [ruby-dev:37906]
+
+Fri Feb 27 22:09:24 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb: don't define assert_not_* if already
+	  overridden.
+
+Fri Feb 27 22:02:34 2009  Tanaka Akira  <akr@fsij.org>
+
+	* util.c: fix SEGV by test_time.rb with
+	  gcc version 4.4.0 20090219 (Red Hat 4.4.0-0.21) on Fedora 11 Alpha.
+
+Fri Feb 27 18:30:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (cdhash_type, iseq_set_sequence): should not call
+	  methods of the argument of case, to keep the semantics of
+	  case/when.  [ruby-dev:38079]
+
+Fri Feb 27 18:01:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (ruby_get_stack_grow_direction): no needs to use thread
+	  here, and not initialized yet.  [ruby-core:22439]
+
+Fri Feb 27 17:45:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_load_ok): checks if regular file, except for the
+	  platform disallows to open directories, e.g. dosish.
+	  [ruby-dev:38097], [Bug #1221]
+
+Fri Feb 27 14:39:40 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (flo_eq, flo_gt, flo_ge, flo_lt, flo_le, flo_eql): revert
+	  isnan() checks for VC++6.
+
+	* insns.def (opt_lt, opt_gt): ditto.
+
+Fri Feb 27 14:23:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (modifiable): check if really frozen.
+	  [ruby-core:22559]
+
+Thu Feb 26 23:14:46 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (BasicSocket#connect_address): new method.
+
+Thu Feb 26 19:29:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (hash_foreach_iter): fix for prototype.
+
+Thu Feb 26 18:36:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (INSNS, newline.c, miniprelude.c, prelude.c),
+	  (node_name.inc, known_errors.inc): unified from Makefile.in and
+	  win32/Makefile.sub.
+
+Thu Feb 26 13:30:34 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* util.c (valid_filename): use our own implementation of open(),
+	  close() and unlink().
+
+Thu Feb 26 13:23:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (cdhash_each): resurrects internal literals.
+
+Thu Feb 26 10:56:48 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (rb_w32_wopen): prototype forgotten.
+
+	* io.c (rb_sysopen_internal): shouldn't replace undefined chars in this
+	  case.
+
+Thu Feb 26 01:34:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: more MSG_* constants.
+
+Thu Feb 26 00:21:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (sockopt_level): extracted from sockopt_level_m.
+	  (sockopt_optname): extracted from sockopt_optname_m.
+	  (sockopt_data): apply StringValue.
+	  (sockopt_s_linger): new method.
+	  (sockopt_linger): new method.
+	  (inspect_linger): show onoff value if it is neither 0 nor 1.
+
+Wed Feb 25 23:23:03 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_unix_rights): raise
+	  NotImplementedError if no fd passing feature.
+	  (ancillary_timestamp): raise NotImplementedError if no timestamp
+	  feature.
+
+Wed Feb 25 23:18:53 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_s_unix_rights): new method.
+
+Wed Feb 25 23:01:26 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/unixsocket.c (unix_recv_io): prevent FD leak when 2 fd is
+	  sent on LP64 platform.
+
+	* ext/socket/rubysocket.h (rsock_discard_cmsg_resource): declared.
+
+	* ext/socket/ancdata.c (rsock_discard_cmsg_resource): renamed from
+	  discard_cmsg_resource.  export it.
+
+Wed Feb 25 17:31:32 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (sysopen_func, rb_sysopen_internal, rb_sysopen): open file
+	  by UTF-16'ed filename on Windows.
+
+	* io.c (rb_file_open_generic, rb_io_s_sysopen, rb_io_reopen,
+	  argf_next_argv): follow above change.
+
+	* io.c (rb_scan_open_args): no longer need to convert filepath here on
+	  Windows.
+
+	* win32/wio32.c (rb_w32_wopen): new function to open file by UTF-16'ed
+	  filename.
+
+	* win32/win32.c (rb_w32_open): call rb_w32_open().
+
+Wed Feb 25 15:05:35 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): use un.rb as cp instead of
+	  cmd.exe's copy command.
+
+	* lib/mkmf.rb (create_makefile): no longer need to convert path
+	  separator when copying file.
+
+Wed Feb 25 12:27:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_force_encoding): should clear coderange at changing
+	  encoding.  [ruby-core:22437]
+
+Wed Feb 25 12:06:09 2009    <nobu@ruby-lang.org>
+
+	* io.c (rb_io_getline_1): enables limit even if rs is given.
+	  [ruby-core:22434]
+
+Wed Feb 25 02:28:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_chomp_bang): coderange may change.
+	  [ruby-core:22414]
+
+Wed Feb 25 02:17:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_delete_bang): should recalculate coderange.
+	  [ruby-talk:329267]
+
+Wed Feb 25 00:41:21 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* class.c (rb_scan_args): Allow specifying the number of trailing
+	  mandatory arguments right after the number of optional arguments
+	  only if the number of leading mandatory arguments is not omitted.
+
+	* ext/socket/tcpserver.c (tcp_svr_init): Make use of it.
+
+Wed Feb 25 00:28:25 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_file_size): new method added.  [ruby-core:22088]
+
+Wed Feb 25 00:15:15 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/curses.c (curses_colors): new method added.  a patch
+	  from Alexander Beisig in [ruby-core:22331].
+
+	* ext/curses/curses.c (curses_color_pairs): ditto.
+
+Wed Feb 25 00:05:13 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* class.c (rb_scan_args): Revamp rb_scan_args() to compute the
+	  number of required and optional arguments precisely to prepare
+	  for a more informative error message.
+
+Tue Feb 24 23:58:52 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_index, rb_ary_rindex): Emit a warning that a
+	  given block is unused when an argument is specified.
+
+Tue Feb 24 22:00:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sockaddr_obj): convert fe80:1::1 to fe80::1%1.
+
+Tue Feb 24 19:56:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_timestamp): support SCM_BINTIME.
+
+Tue Feb 24 19:45:15 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_timestamp): new method.
+
+Tue Feb 24 19:01:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (main): split from exts and makes main program after
+	  building exts with miniruby, to get rid of overwriting running
+	  program.  [ruby-core:22339]
+
+	* Makefile.in, win32/Makefile.sub (RUNCMD, MKMAIN_CMD): macros to
+	  run script file.
+
+	* ext/extmk.rb (parse_args): added --command-output option which
+	  creates script file to make main program.
+
+Tue Feb 24 17:58:07 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ripper.c): exec can't work on Windows.
+
+	* ext/ripper/depend (.SUFFIXES): nmake doesn't have .y by default.
+
+	* Makefile.in (id.h): overwrites unconditionally.
+
+Tue Feb 24 13:13:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/vsnprintf.c (BSD_vfprintf): ptrdiff_t may be larger than
+	  long.
+
+Tue Feb 24 11:21:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (rb_disable_interrupt, rb_enable_interrupt): see
+	  USE_TRAP_MASK.
+
+Tue Feb 24 10:49:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_waitpid): last argument was missing.
+
+	* process.c (waitall_each): fixed typo.  a patch from shinichiro.h
+	  <shinichiro.hamaji AT gmail.com> at [ruby-dev:38054].
+
+Tue Feb 24 01:53:40 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_eq): remove unnecessary NaN check using isnan().
+	  comparison regarding NaN is false anyway.
+
+	* numeric.c (flo_gt, flo_ge, flo_lt, flo_le, flo_eql): ditto.
+
+Tue Feb 24 01:22:19 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bootstraptest/test_thread.rb: fix for environment where fork is not
+	  available.
+
+Tue Feb 24 01:19:38 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (thread_cleanup_func): unlock all locked mutexes even when
+	  forking.  [ruby-core:22269]
+
+Tue Feb 24 00:54:16 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (opt_minus): inline float operation.
+
+	* insns.def (opt_lt, opt_gt): ditto.
+
+Tue Feb 24 00:24:13 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_each): check #succ only when it is really
+	  needed.  small performance improvement.
+
+	* math.c (Need_Float): call rb_to_float() only when it is really
+	  needed.  small performance improvement.
+
+Tue Feb 24 00:19:33 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/pathname.rb (Pathname#foreachline): removed wrongly
+	  resurrected method.  [ruby-dev:38065]
+
+Mon Feb 23 20:39:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (inspect_timespec_as_abstime): new function to
+	  show struct timespec.
+	  (ancillary_inspect): use it for SCM_TIMESTAMPNS on GNU/Linux.
+
+Mon Feb 23 20:30:06 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (inspect_bintime_as_abstime): new function to
+	  show struct bintime.
+	  (ancillary_inspect): use it for SCM_BINTIME on FreeBSD.
+
+	* ext/socket/mkconstants.rb: define SCM_BINTIME.
+
+Mon Feb 23 20:18:49 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (inspect_timeval_as_abstime): use time_t
+	  variable to avoid a warning on environments which use long as
+	  tv_sec as NetBSD.
+
+Mon Feb 23 15:53:30 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_index, rb_ary_rindex): We know that argc is not
+	  zero when we reach here.
+
+	* array.c (rb_ary_initialize, rb_ary_index): Shrink a double
+	  space. (NFC)
+
+Mon Feb 23 14:47:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (mv): added description for the case of
+	  different partitions.  [ruby-talk:329072]
+
+Mon Feb 23 10:19:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (USE_SYMBOL_AS_METHOD_NAME): Module#methods,
+	  #singleton_methods and so on return Symbols.  [ruby-talk:328775]
+
+Mon Feb 23 10:15:35 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja: added macros for the Compatibilities.
+
+Mon Feb 23 03:19:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: define SO_TIMESTAMPNS and SCM_TIMESTAMPNS
+	  if available.
+
+Mon Feb 23 02:54:31 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_inspect): use SCM_TIMESTAMP instead
+	  of SO_TIMESTAMP.  SCM_TIMESTAMP != SO_TIMESTAMP on MacOS X.
+
+Mon Feb 23 02:44:42 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (inspect_timeval_as_abstime): new function.
+	  (ancillary_inspect): use it for SO_TIMESTAMP.
+
+Mon Feb 23 02:26:44 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_timeval_as_interval): renamed from
+	  inspect_timeval.
+
+Mon Feb 23 02:10:09 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_timeval): fix the size test.
+
+Sun Feb 22 22:42:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (tr_trans): should recalculate coderange.
+	  [ruby-core:22326] (reopened at [ruby-core:22328])
+
+Sun Feb 22 20:09:29 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_linger): message refined.
+
+Sun Feb 22 19:28:08 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (s_recvfrom): use rb_io_wait_readable after
+	  recvfrom.
+
+Sun Feb 22 18:40:04 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_errno): new function.
+	  (sockopt_inspect): use inspect_errno for SO_ERROR.
+
+Sun Feb 22 15:11:51 2009  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (Init_Regexp): define Regexp::FIXEDENCODING.  [ruby-dev:38066]
+
+Sun Feb 22 14:33:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (tr_trans): should not be affected by the encoding of
+	  replacement unless actually modified.  [ruby-talk:328967]
+	  [ruby-core:22326]
+
+Sun Feb 22 13:38:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/init.c (s_recvfrom_nonblock): handles EAGAIN too.
+
+	* ext/socket/init.c (s_accept_nonblock): ditto, and EPROTO depends
+	  on platforms.
+
+Sun Feb 22 13:03:12 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* eval.c (ruby_options): the evaluator now expects iseq instead of tree.
+
+	* ruby.c (ruby_process_options): ditto.
+
+Sun Feb 22 13:03:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (configure): unset PWD to use symlink to building
+	  directory.
+
+	* Makefile.in (up): cd doesn't need LC_TIME, VCS needs.
+
+Sun Feb 22 10:43:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (method_missing): should not pop cfp if missing method
+	  is method_missing.  [ruby-core:22298]
+
+	* vm_eval.c (rb_raise_method_missing): new function to directly
+	  raise NoMethodError.
+
+	* vm_insnhelper.c (vm_call_method): fixed the case method_missing
+	  is missing.
+
+Sun Feb 22 02:15:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): handle EMSGSIZE as
+	  well.
+
+Sun Feb 22 02:03:46 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): check max length
+	  overflow.
+
+Sun Feb 22 01:52:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): don't call
+	  discard_cmsg_resource when EMFILE.
+
+Sun Feb 22 01:48:51 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): gc when SCM_RIGHTS
+	  hit the file descriptor limit.
+
+Sun Feb 22 00:51:58 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_getpartial): error message describes what should be
+	  waited after nonblocking error.
+	  (rb_io_write_nonblock): ditto.
+
+	* ext/socket/init.c (s_recvfrom_nonblock): ditto.
+	  (s_accept_nonblock): ditto.
+
+	* ext/socket/socket.c (sock_connect_nonblock): ditto.
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): ditto.
+	  (bsock_recvmsg_internal): ditto.
+
+Sun Feb 22 00:31:42 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): check recvmsg error
+	  earlier.
+
+Fri Feb 20 23:28:11 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* util.c (rv_alloc, freedtoa): use our normal xmalloc()/xfree() because
+	  couldn't free the returned pointer from ruby_dtoa().
+
+	* missing/vsnprintf.c (cvt): receive buffer and use/return it instead
+	  of returning the pointer returned from BSD__dtoa().
+
+	* missing/vsnprintf.c (BSD_vfprintf): pass buf to cvt() as the buffer.
+	  [ruby-core:22184]
+
+Thu Feb 19 22:59:09 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (make_io_for_unix_rights): cmsg_len may be
+	  bigger than msg_controllen on 4.4BSD at least.
+	  freeze unix_rights array.
+
+Thu Feb 19 22:17:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): fix exception.
+
+Thu Feb 19 21:13:03 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_unix_rights): method renamed.
+
+Thu Feb 19 15:47:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in: extracted from Makefile.in.
+
+	* configure.in (fake.rb): prefixed with $(arch)-.
+
+Thu Feb 19 15:39:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspbres_verify): OCSP_basic_verify
+	  returns positive value on success, not non-zero.  [ruby-core:21762]
+
+Thu Feb 19 15:17:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tmpdir.rb (@@systmpdir): File.expand_path also joins paths.
+
+Thu Feb 19 15:14:25 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_each_group): defines only when Etc::Group is
+	  available.
+
+Thu Feb 19 15:11:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (translit_char): moved from ruby.c.
+
+	* load.c (load_ext): transliterates file separators and back if
+	  needed.
+
+	* symbian/setup (DLN_NEEDS_ALT_SEPARATOR): defined.
+
+Thu Feb 19 14:48:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, */Makefile.sub (LOAD_RELATIVE): moved from ruby.c
+
+Thu Feb 19 14:39:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (darwin): uses -unexported_symbol option of ld instead
+	  of objcopy
+
+	* configure.in (darwin): makes dylib instead of so.
+
+Thu Feb 19 14:22:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): added phony targets.
+
+	* lib/mkmf.rb (create_makefile): creates target directories before
+	  copying.  [ruby-core:21958]
+
+	* lib/mkmf.rb (create_makefile): removes directories in the depth
+	  order.
+
+Thu Feb 19 13:56:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_home_dir): checks if HAVE_PWD_H.  [ruby-dev:38049]
+
+Thu Feb 19 08:45:48 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* ext/socket/rubysocket.h: add "#include <sys/stat.h>"
+
+Thu Feb 19 03:42:48 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_rights): new method.
+	  (make_io_for_rights): new function to allocate
+	  IOs for FDs in SCM_RIGHTS message.
+	  (bsock_recvmsg_internal): use make_io_for_rights.  So the FDs can be
+	  closed by GC.
+
+Thu Feb 19 03:28:59 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* README.EXT, README.EXT.ja: Improve the document about
+	  rb_scan_args() even more.
+
+Thu Feb 19 03:12:51 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_file_exhaustive.rb: skip some tests if euid == 0.
+	  a patch from Lucas Nussbaum in [ruby-core:22209].
+
+Thu Feb 19 03:00:49 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* README.EXT, README.EXT.ja: Improve the document about
+	  rb_scan_args().
+
+Wed Feb 18 22:47:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (discard_cmsg_resource): new function to close
+	  file descriptors in control message.
+	  (bsock_recvmsg_internal): call discard_cmsg_resource before retrying
+	  recvmsg.
+
+Wed Feb 18 21:47:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): prevent misalignment.
+
+Wed Feb 18 21:34:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): prevent misalignment.
+
+Wed Feb 18 21:09:43 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): reduce code on
+	  environments which have no control message.
+
+Wed Feb 18 20:27:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (OBJCOPY): set ac_cv_prog_ac_ct_OBJCOPY to do nothing
+	  on cygwin and mingw because symbols are exported by def file.
+
+Wed Feb 18 15:35:36 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_chown): use chown unless HAVE_FCHOWN.
+
+	* io.c (pipe_open): sarg is always used unless HAVE_FORK.
+
+Wed Feb 18 15:23:34 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* bootstraptest/runner.rb: Use RUBY_DESCRIPTION if defined.
+
+Wed Feb 18 14:33:35 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_resurrect), string.c (rb_str_resurrect): new
+	  functions based on [ruby-dev:37983]
+
+	* insns.def (putstring, duparray): use rb_{ary,str}_resurrect().
+
+	* iseq.c (iseq_data_to_ary): needs to result TS_VALUE.
+
+Wed Feb 18 12:35:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_extname): fix for spaces before extension.
+	  [ruby-dev:38044]
+
+Wed Feb 18 12:00:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): suppress a warning.  named magic
+	  numbers.
+
+Wed Feb 18 10:29:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): do not delete main thread from
+	  living_threads.  [ruby-core:19385], [ruby-core:22158]
+
+Wed Feb 18 01:05:05 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl: avoid cyclic require.
+
+	* ext/openssl/lib/openssl/ssl-internal.rb: renamed from ssl.rb
+
+	* ext/openssl/lib/openssl/x509-internal.rb: renamed from x509.rb.
+
+	* lib/net/imap.rb: require openssl, instead of openssl/ssl.
+
+	* lib/net/pop.rb: require openssl, instead of openssl/ssl.
+
+	[ruby-dev:38018]
+
+Wed Feb 18 00:55:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (Etc::Passwd.each, Etc::Group.each): new methods.
+	  [ruby-dev:37999]
+
+Tue Feb 17 23:25:01 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): make the padding
+	  removing only on NetBSD.
+
+Tue Feb 17 19:39:04 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): should not remove the
+	  last padding of control messages, basically.
+
+Tue Feb 17 12:50:57 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_shared_replace): no need for
+	  str_make_independent.
+
+Tue Feb 17 05:41:08 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: fix to add "ensure" codes across "while" clause
+	  before "return" expression.  [ruby-dev:37967]
+
+	* bootstraptest/test_flow.rb: add a test.
+
+Tue Feb 17 01:53:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: generate rb_define_const directly for
+	  rdoc.
+
+	* ext/.document: add socket/constdefs.c.
+
+	* ext/socket/constants.c (sock_define_const): removed.
+	  (sock_define_uconst): ditto.
+	  (rb_mSockConst): new static variable.
+
+Mon Feb 16 23:14:51 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal) [OpenBSD]: don't remove
+	  the last padding to prevent sendmsg failure in test_udp_server in
+	  test/socket/test_socket.rb on OpenBSD 4.4.
+
+Mon Feb 16 21:59:32 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/setup: Update Symbian config to match recent changes in missing/alloca.c
+
+Mon Feb 16 17:04:14 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* class.c (rb_scan_args), README.EXT, README.EXT.ja: Add support
+	  for specifying the number of the trailing mandatory arguments.
+	  Update the documents accordingly. [ruby-dev:37995]
+
+Mon Feb 16 16:46:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* debug.c (set_debug_option): added rtc_error option.
+
+	* win32/Makefile.sub (CRTDEFFLAGS): separated from DEFS.
+
+	* win32/win32.c (rtc_error_handler): ignores RTC errors unless
+	  rtc_error debug option is given.
+
+	* win32/win32.c (rb_w32_sysinit): suppress useless CRT assertions.
+	  [ruby-core:22116]
+
+Sun Feb 15 21:43:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (OpenStruct#new_ostruct_member): checks if frozen.
+	  [ruby-talk:328195], [ruby-core:22142]
+
+Sun Feb 15 21:22:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions): aliases
+	  assert_not_ methods.
+
+Sun Feb 15 16:57:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (SecureRandom.urlsafe_base64): new method.
+
+Sun Feb 15 14:58:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* hash.c (hash_i): use Murmurhash.
+
+Sun Feb 15 11:45:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_define_hooked_variable): suppress false assertion
+	  with VC9.  [ruby-core:22115]
+
+Sun Feb 15 08:35:33 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/test_pp.rb: extract from lib/pp.rb.
+
+	* test/test_prettyprint.rb: extract from lib/prettyprint.rb.
+
+	* test/test_tsort.rb: extract from lib/tsort.rb.
+
+Sun Feb 15 06:34:22 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.parse): raise ArgumentError if Date._parse don't
+	  extract date information.  [ruby-core:20912]
+
+Sun Feb 15 04:48:08 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (rb_hash_uint32, rb_hash_uint, rb_hash_start, rb_hash_end),
+	  include/ruby/intern.h: add Murmurhash API.  [ruby-dev:37784]
+
+	* complex.c (nucomp_hash), array.c (rb_ary_hash), time.c (time_hash),
+	  string.c (rb_str_hash), object.c (rb_obj_hash), range.c
+	  (range_hash), struct.c (rb_struct_hash), hash.c (rb_any_hash),
+	  rational.c (nurat_hash): use Murmurhash.  [ruby-dev:37784]
+
+Sun Feb 15 03:50:21 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* hash.c (rb_hash): always return a fixnum value because a return
+	  value of rb_hash may be used as a hash value itself and bignums have
+	  no unique VALUE.
+
+	* test/ruby/test_hash.rb: add a test for above.
+
+Sun Feb 15 00:45:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_uniq_bang, rb_ary_uniq): unique by the result of
+	  given block.  [ruby-dev:37998]
+
+Sun Feb 15 00:39:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_resize_capa): should not overwrite outside embedded
+	  array.
+
+Sat Feb 14 20:18:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/test_time.rb: extracted from lib/time.rb.
+
+Sat Feb 14 19:20:15 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pathname.rb: obsolete methods removed.
+	  [ruby-core:21564]
+
+Sat Feb 14 15:46:01 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/pathname.rb (Pathname#binread): added.  [ruby-dev:37952]
+
+Sat Feb 14 13:14:18 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* iseq.c: remove nil parameter from Proc#parameters
+	  after rest appeared.
+
+Sat Feb 14 07:23:52 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_external_str_new_with_enc): change evaluation order
+	  for speed.
+
+Sat Feb 14 02:20:04 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/sockport.h (SET_SA_LEN): cast to void for suppressing a
+	  warning.
+
+Fri Feb 13 23:37:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/win32/lib/Win32API.rb (Win32API#initialize): DL.dlopen
+	  raises DLError.
+
+Fri Feb 13 21:13:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (Init_socket): define TCPServer#listen and
+	  UNIXServer#listen here for rdoc.
+
+Fri Feb 13 20:59:48 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (extract_ipv6_pktinfo): set sa_len for 4.4BSD.
+
+Fri Feb 13 15:34:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (simple_default_value): removed.  default values of optional
+	  parameters are not available still now.  [ruby-dev:37980]
+
+Fri Feb 13 15:11:11 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (eval_string_with_cref): use rb_vm_get_ruby_level_next_cfp()
+	  instead of vm_get_ruby_level_caller_cfp().  checking a upper frame
+	  is not enough.  [ruby-dev:37984]
+
+	* proc.c, vm_core.h: declare rb_vm_get_ruby_level_next_cfp()
+	  on vm_core.h.
+
+Fri Feb 13 15:01:40 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_alive_p): fix to return true instead of 0.
+	  [ruby-dev:37991]
+
+	* test/ruby/test_fiber.rb: add a test for Fiber#alive?
+
+Fri Feb 13 09:43:19 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func_flock): using flock defined in
+	  win32/win32.c on mingw.
+
+Fri Feb 13 01:45:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_sequence, compile_dstr_fragments),
+	  (iseq_compile_each): hides other internal objects.
+
+	* compile.c (iseq_compile_each): just freeze xstr.
+
+Fri Feb 13 00:48:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defineclass): try to autoload const to be overridden.
+
+Thu Feb 12 23:22:29 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_socket.rb (test_udp_server): filter out unreachable
+	  addresses.  [ruby-dev:37970]
+
+Thu Feb 12 22:51:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (hide_obj): OBJ_FREEZE() is not an expression.  a patch
+	  from nagachika <nagachika00 AT gmail.com> at [ruby-dev:37977].
+
+	* compile.c (insn_set_sc_state): fixed typo.
+
+Thu Feb 12 12:36:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/basicsocket.c (bsock_getpeereid): implemented for Solaris
+	  using getpeerucred.
+
+	* ext/socket/extconf.rb: check ucred.h and getpeerucred.
+
+	* ext/socket/rubysocket.h: include ucred.h if available.
+
+Thu Feb 12 19:42:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_array_, defined_expr, iseq_compile_each): hide
+	  and freeze internal literal objects, to prevent from modifying.
+	  [ruby-dev:37959]
+
+	* iseq.c (insn_operand_intern): copy internal literal objects.
+
+	* insns.def (putstring, duparray): ditto.
+
+	* string.c (rb_str_replace): exported.
+
+Thu Feb 12 17:17:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_ungetc): calculates new position
+	  before reallocation.  [Bug#1099]
+
+Thu Feb 12 16:50:27 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: a patch to build on GNU/kOpenSolaris from Rober
+	  Millan at [ruby-core:21888].
+
+Thu Feb 12 15:28:04 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* iseq.c (simple_default_value): allow plain strings as default
+	  values.
+
+Wed Feb 11 18:09:41 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (define_final, undefine_final): shouldn't add/remove finalizer
+	  of frozen objects.
+
+	* gc.c (undefine_final): should remove FL_FINALIZE flag.
+	  [ruby-dev:37964] (also see [ruby-dev:37959])
+
+Wed Feb 11 17:37:41 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.udp_server_sockets): call the block
+	  if given.  close the sockets when the block exits.
+	  (Socket.udp_server_loop): use udp_server_sockets in block form.
+
+Wed Feb 11 17:34:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.tcp_server_sockets): call the block
+	  if given.  close the sockets when the block exits.
+	  (Socket.tcp_server_loop): use tcp_server_sockets in block form.
+
+Wed Feb 11 17:01:52 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.unix_server_loop): use
+	  unix_server_socket with a block.
+
+Wed Feb 11 16:54:26 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.unix_server_socket): close the
+	  socket when the block exits.
+
+Wed Feb 11 16:50:59 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.unix_server_socket): call the block
+	  if given.  remove the socket file when the block exits.
+
+Wed Feb 11 16:44:20 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_s_ip_pktinfo): make 3rd argument
+	  optional.
+
+Wed Feb 11 15:47:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_make_shared): returns shared root array itself, and
+	  frozen array can be shared.
+
+Wed Feb 11 14:46:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (setspecial, putstring): fixed typos in rdoc.
+
+	* insns.def (toregexp): uses rb_ary_tmp_new(), and clears it after
+	  used.
+
+Wed Feb 11 14:08:23 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (anc_inspect_socket_rights): return 1 on
+	  success.  0 on failure.
+	  (anc_inspect_passcred_credentials): ditto.
+	  (anc_inspect_socket_creds): ditto.
+	  (anc_inspect_socket_creds): ditto.
+	  (anc_inspect_socket_creds): ditto.
+	  (anc_inspect_ip_recvdstaddr): ditto.
+	  (anc_inspect_ip_pktinfo): ditto.
+	  (anc_inspect_ipv6_pktinfo): ditto.
+	  (ancillary_inspect): check family.
+
+Wed Feb 11 12:55:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (anc_inspect_ip_pktinfo): make result bit
+	  succinct.
+	  (Init_ancdata): fix number of arguments.
+
+Wed Feb 11 11:47:41 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c (Init_ipsocket): undef getpeereid at IPSocket.
+
+Wed Feb 11 10:16:34 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: check getpeereid.
+
+	* ext/socket/basicsocket.c (bsock_getpeereid): new method.
+
+Wed Feb 11 09:58:59 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket::UDPSource#inspect): fix variable
+	  name.
+
+Wed Feb 11 00:38:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_gethostname): use HOST_NAME_MAX.
+
+	* ext/socket/raddrinfo.c (make_ipaddr): local variable renamed.
+
+Tue Feb 10 23:44:53 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (struct recvfrom_arg): use struct
+	  sockaddr_storage.
+	  (recvfrom_blocking): follow struct recvfrom_arg change.
+	  (s_recvfrom): ditto.
+	  (s_recvfrom_nonblock): use struct sockaddr_storage.
+
+	* ext/socket/socket.c (sock_accept): use struct sockaddr_storage.
+	  (sock_accept_nonblock): ditto.
+	  (sock_sysaccept): ditto.
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): use struct
+	  sockaddr_storage.
+
+Tue Feb 10 23:30:32 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/basicsocket.c (bsock_getpeername): use struct
+	  sockaddr_storage.
+	  (bsock_local_address): ditto.
+	  (bsock_remote_address): ditto.
+
+Tue Feb 10 21:26:33 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.udp_server_sockets): new method.
+	  (Socket.udp_server_loop_on): new method.
+	  (Socket.udp_server_loop): new method
+	  (Socket.ip_sockets_port0): extracted from tcp_server_sockets_port0.
+	  (Socket::UDPSource): new class.
+
+Tue Feb 10 21:14:43 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sockaddr_obj): fill pfamily.
+
+Tue Feb 10 21:09:23 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (sockopt_s_bool): new method.
+	  (sockopt_bool): new method.
+
+Tue Feb 10 20:58:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (sockopt_family_m): renamed from sockopt_family.
+	  (sockopt_level_m): renamed from sockopt_level.
+	  (sockopt_optname_m): renamed from sockopt_optname.
+	  (inspect_local_peercred): follow the renaming.
+	  (Init_sockopt): ditto.
+
+Tue Feb 10 20:56:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (anc_inspect_socket_creds): refactored to avoid
+	  a rb_str_cat2 call.
+
+Tue Feb 10 04:34:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (when): makes parentheses balanced.
+
+Tue Feb 10 01:28:49 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ip_cmsg_type_to_sym): renamed from
+	  cmsg_type_to_sym.
+
+Tue Feb 10 01:22:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (sockopt_inspect): refactored.
+
+Tue Feb 10 01:02:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: generate intern_family_noprefix.
+
+	* ext/socket/option.c (sockopt_inspect): use intern_family_noprefix
+	  not to show AF_ prefix.
+
+	* ext/socket/ancdata.c (ancillary_inspect): ditto.
+
+Mon Feb  9 23:21:29 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/constants.c (level_arg): use unknown_level_to_int for
+	  non internet protocol.
+	  (optname_arg): use only so_optname_to_int for non internet protocol.
+	  (cmsg_type_arg): use only scm_optname_to_int for non internet
+	  protocol.
+
+	* ext/socket/mkconstants.rb: generate unknown_level_to_int.
+	  rename iplevel_to_int to ip_level_to_int.
+
+Mon Feb  9 23:04:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: rename level_to_int to iplevel_to_int.
+	  rename intern_level to intern_iplevel.
+
+	* ext/socket/constants.c: follow the renaming.
+
+	* ext/socket/option.c: ditto.
+
+	* ext/socket/ancdata.c: ditto.
+
+Mon Feb  9 22:52:13 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (level_arg): add family argument.
+	  (optname_arg): ditto.
+	  (cmsg_type_arg): ditto.
+	  (rb_sock_getfamily): declared.
+
+	* ext/socket/constants.c (level_arg): add family argument.
+	  (optname_arg): ditto.
+	  (cmsg_type_arg): ditto.
+
+	* ext/socket/init.c (rb_sock_getfamily): defined.
+
+	* ext/socket/option.c (sockopt_initialize): give family for level_arg
+	  and optname_arg.
+	  (sockopt_s_int): ditto.
+
+	* ext/socket/basicsocket.c (bsock_setsockopt): ditto.
+	  (bsock_getsockopt): ditto.
+
+	* ext/socket/ancdata.c (ancillary_initialize): ditto.
+	  (ancillary_s_int): ditto.
+	  (ancillary_cmsg_is_p): ditto.
+	  (bsock_sendmsg_internal): ditto.
+	  (bsock_recvmsg_internal): use rb_sock_getfamily.
+
+Mon Feb  9 21:48:59 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_inspect): show address family.
+
+Mon Feb  9 20:19:36 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/basicsocket.c (bsock_getsockname): use sockaddr_storage.
+
+Mon Feb  9 20:11:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (load_conv_function51932): fixed for the
+	  case IMultiLanguage is not available.  [ruby-dev:37950]
+
+Mon Feb  9 20:04:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_int): return 1/0 on success/failure.
+	  (inspect_uint): ditto.
+	  (inspect_linger): ditto.
+	  (inspect_socktype): ditto.
+	  (inspect_timeval): ditto.
+	  (inspect_peercred): ditto.
+	  (inspect_local_peercred): ditto.
+	  (sockopt_inspect): refactored.
+
+Mon Feb  9 20:00:26 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: more constants for Solaris.
+
+Mon Feb  9 19:59:18 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/basicsocket.c (bsock_getsockopt): refine address family
+	  examination.
+
+Mon Feb  9 19:52:32 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (sockopt_initialize): argument conversion
+	  reordered.
+
+Mon Feb  9 19:46:22 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_initialize): add family argument.
+	  (ancdata_new): ditto.
+	  (ancillary_s_int): ditto.
+	  (ancillary_family): new function.
+	  (ancillary_family_m): new method.
+	  (ancillary_s_ip_pktinfo): follow ancdata_new change.
+	  (ancillary_s_ipv6_pktinfo): ditto.
+	  (bsock_recvmsg_internal): examine the socket address family.
+
+Mon Feb  9 11:31:09 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_encoding_set): ignore second argument when external and
+	  internal are same. [ruby-dev:37939]
+
+Mon Feb  9 09:05:12 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_local_peercred): check version.
+	  (sockopt_inspect): suppress warning.
+
+Mon Feb  9 02:04:03 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_peercred): struct ucred contains
+	  effective uid/gid.
+
+Mon Feb  9 00:44:45 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (sockopt_inspect): add ifdef guard for
+	  LOCAL_PEERCRED.
+
+Mon Feb  9 00:37:06 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_local_peercred): cr_uid is a effective
+	  uid, not a real uid.
+
+Mon Feb  9 00:30:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: check sys/param.h and sys/ucred.h.
+
+	* ext/socket/rubysocket.h: include sys/param.h and sys/ucred.h.
+
+	* ext/socket/option.c (inspect_local_peercred): new function to show
+	  LOCAL_PEERCRED socket option on FreeBSD.
+	  (sockopt_inspect): show as LOCAL_* socket option if AF_UNIX and level
+	  is 0.
+
+Mon Feb  9 00:01:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (sockopt_new): add family argument.
+
+	* ext/socket/option.c (sockopt_initialize): add vfamily argument.
+	  (sockopt_new): add family argument and record it in the object.
+	  (sockopt_family): new method.
+	  (sockopt_s_int): add vfamily argument.
+	  (sockopt_inspect): show family.
+
+	* ext/socket/basicsocket.c (bsock_getsockopt): check address family
+	  using getsockname.
+
+Sun Feb  8 23:37:17 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enumerator.c (enumerator_with_index): receives one argument which
+	  represents a start offset.  [ruby-dev:37921]
+
+Sun Feb  8 23:28:05 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* include/ruby/st.h, st.c: order entries by a linked list instead of
+	  a loop to fix iteration miss when hash is modified during iteration.
+	  [ruby-dev:37910]
+
+Sun Feb  8 23:22:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_peercred): new function to show
+	  SO_PEERCRED socket option on GNU/Linux.
+
+Sun Feb  8 22:44:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/backward/rubysig.h (rb_thread_blocking_region_begin),
+	  (rb_thread_blocking_region_end): marked as deprecated.
+
+	* include/ruby/backward/rubysig.h (TRAP_BEG): fix for C++.  a
+	  patch from Aman Gupta at [ruby-core:21934]
+
+Sun Feb  8 21:47:50 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: check struct cmsgcred.
+
+	* ext/socket/ancdata.c (anc_inspect_passcred_credentials): add
+	  "(ucred)".
+	  (anc_inspect_socket_creds): show struct cmsgcred too, for FreeBSD.
+
+Sun Feb  8 21:05:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/drb/extservm.rb (DRb::ExtServManager#invoke_service_command):
+	  detach spawned process.  [ruby-dev:37936]
+
+Sun Feb  8 20:30:29 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: check struct sockcred.
+
+	* ext/socket/ancdata.c (anc_inspect_socket_creds): new function to
+	  show SCM_CREDS on NetBSD.
+
+Sun Feb  8 19:05:24 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/backward/rubysig.h: extern "C" was missing.
+	  [ruby-core:21929]
+
+Sun Feb  8 18:46:15 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (anc_inspect_passcred_credentials): new
+	  function to show SCM_CREDENTIALS on GNU/Linux.
+
+Sun Feb  8 18:34:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_s_binread): ensures file path.  [ruby-dev:37940]
+
+Sun Feb  8 13:52:02 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (init_unix_addrinfo): add socktype argument.
+	  (addrinfo_initialize): follow init_unix_addrinfo change.
+	  (addrinfo_s_unix): add optional argument: socktype
+
+Sun Feb  8 13:09:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_APPEND_OPTION, RUBY_APPEND_OPTIONS),
+	  (RUBY_PREPEND_OPTION, RUBY_PREPEND_OPTIONS): add option(s)
+	  without duplication.
+
+	* configure.in (RUBY_DEFINE_IF): changed parameter order, now
+	  condition comes first.
+
+	* configure.in (universal_binary): checks architecture macros, and
+	  improved thin load paths.
+
+Sun Feb  8 09:41:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* runruby.rb: gets extout and arch from rbconfig.rb.
+
+Sat Feb  7 21:26:15 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_addrinfo.rb (test_family_addrinfo): don't use
+	  www.ruby-lang.org.
+	  http://d.hatena.ne.jp/nagachika/20090204/working_for_0f0e
+
+Sat Feb  7 18:02:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--with-arch): added new option to support
+	  universal binary.  replaced --enable-fat-binary option which
+	  didn't work actually.
+
+	* configure.in (RUBY_FUNC_ATTRIBUTE): added conditional test.
+
+	* configure.in (ac_cv_type_getgroups): declared because getgroups()
+	  fills rest of the buffer with garbage on Rosetta.
+
+	* configure.in (alloca): defines only for powerpc, but always
+	  create empty object to suppress ld warning.
+
+	* configure.in (LIBRUBY_DLDFLAGS): set compatibility version with
+	  TEENY.
+
+	* configure.in (CFLAGS, LDFLAGS): separates ARCH_FLAG.
+
+	* configure.in (arch): renamed to "universal" from "fat".
+
+	* Makefile.in (ARCH_FLAG): added.
+
+	* include/ruby/defines.h (WORDS_BIGENDIAN): uses
+	  AC_APPLE_UNIVERSAL_BUILD.
+
+	* missing/alloca.c (alloca): defines only if C_ALLOCA is defined.
+
+Sat Feb  7 12:31:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* runruby.rb: added --precommand and --show options.
+
+	* runruby.rb: added --cpu option.
+
+	* runruby.rb: skips version check in rbconfig.rb.
+
+Sat Feb  7 11:44:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_header): needs dependent headers if trying to
+	  compile.
+
+	* ext/socket/extconf.rb: net/if.h depends on other headers.
+
+Sat Feb  7 04:02:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* dir.c (dir_read): don't disable rdoc.
+
+Fri Feb  6 23:28:33 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_fread): use rb_io_wait_readable for retry
+	  avoid Errno::EINTR on ruby -e 'trap(:CHLD) {}; spawn("sleep 1"); STDIN.read'
+
+Fri Feb  6 22:36:11 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* thread_pthread.c (native_thread_create) [__SYMBIAN32__]: reduced pthread stack size.
+
+	* thread_pthread.c (thread_timer) [__SYMBIAN32__]: compiled out unsupported yet signal-related functionality.
+
+	* io.c (pipe_open) [__SYMBIAN32__]: fixed compile time error.
+
+Fri Feb  6 22:11:46 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* include/ruby/defines.h [__SYMBIAN32__]: included <sys/select.h> for fd_set definition
+
+Fri Feb  6 21:58:24 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/missing-pips.c: added a stub for missing PIPS function execl.
+
+	* process.c (rb_proc_exec) [__SYMBIAN32__]: removed conditional around execl function call.
+
+Fri Feb  6 20:37:42 2009  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (register_sigaltstack): ignore sigaltstack error.
+	  It fails on OpenBSD 4.4 when pthread library is linked.
+
+Fri Feb  6 18:18:07 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_readdir_with_enc): fallback to OS's conversion
+	  when ASCII-8BIT is passed.
+
+Fri Feb  6 17:19:23 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{dir.h, win32.c} (rb_w32_readdir_with_enc): new function to
+	  read entry with specified enc.
+
+	* win32/win32.c (readdir_internal): extract from rb_w32_opendir().
+
+	* win32/win32.c (opendir_internal): extract from rb_w32_readdir().
+
+	* dir.c (dir_read, dir_each): use new READDIR macro instead of readdir()
+	  to pass enc for special version of readdir, such as above.
+
+Fri Feb  6 12:11:24 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (process_options): set initial default_external before -r.
+
+Fri Feb  6 12:03:47 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (process_options): -K and -E in shebang should be reflect to
+	  default_external. [ruby-dev:37920]
+
+Fri Feb  6 07:52:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (chfunc): type fixed.
+
+Fri Feb  6 02:51:59 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_each_codepoint): update RDoc for
+	  String#codepoints.  a patch from Radoslaw Bulat in
+	  [ruby-core:21835]
+
+Fri Feb  6 01:09:13 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c (cont_mark, cont_capture, cont_restore_1): use #else instead
+	  of #elif.  a patch from NISHIMATSU Takeshi <t_nissie at yahoo.co.jp>
+	  in [ruby-list:45856].
+
+Thu Feb  5 20:28:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_run_exec_options_err): renamed from
+	  rb_run_exec_options.
+	  (rb_exec_err): renamed from rb_exec.
+	  (rb_fork_err): renamed from rb_fork.
+	  (rb_spawn_err): renamed from rb_spawn.
+	  (rb_run_exec_options): declared with 1.9.1 compatible signature.
+	  (rb_exec): ditto.
+	  (rb_fork): ditto.
+	  (rb_spawn): ditto.
+
+	* process.c (rb_run_exec_options_err): renamed from
+	  rb_run_exec_options.
+	  (rb_exec_err): renamed from rb_exec.
+	  (rb_fork_err): renamed from rb_fork.
+	  (rb_spawn_err): renamed from rb_spawn.
+	  (rb_run_exec_options): defined.
+	  (rb_exec): ditto.
+	  (rb_fork): ditto.
+	  (rb_spawn): ditto.
+
+	* io.c: follow above change.
+
+	* ext/pty/pty.c: follow above change.
+
+	  [ruby-dev:37893]
+
+Thu Feb  5 19:58:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: AddrInfo is renamed to Addrinfo.  [ruby-dev:37876]
+
+Thu Feb  5 16:18:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_SITE_LIB_PATH, RUBY_VENDOR_LIB_PATH): fix for
+	  NetBSD.
+
+Thu Feb  5 16:04:29 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): follow recent changes about paths.
+
+Thu Feb  5 14:26:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (MAJOR, MINOR, TEENY): uses RUBY_VERSION_*.
+
+	* mkconfig.rb (prefix): uses ruby_version in config.status.
+
+Thu Feb  5 12:24:18 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_uniq): gets rid of copying.
+
+Thu Feb  5 12:01:53 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_add_hash): split from ary_make_hash().
+
+	* array.c (ary_make_hash): hide a Hash used internally.
+
+	* array.c (ary_recycle_hash): clears internally used hash.  this
+	  name came from [ruby-dev:37908].
+
+	* array.c (rb_ary_diff, rb_ary_and, rb_ary_or, rb_ary_uniq_bang):
+	  recycle hashes.
+
+Thu Feb  5 11:21:35 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_LIB_VERSION): added for library version, to
+	  split from core version.  [ruby-dev:37748]
+
+	* configure.in (RUBY_LIB_PATH, etc): moved actual version
+	  dependent stuff to version.c.
+
+	* ruby.c (ruby_init_loadpath_safe): ditto.
+
+	* version.c (ruby_initial_load_paths): moved initial load path
+	  version depending on version from ruby.c.
+
+	* version.h (RUBY_VERSION_{MAJOR,MINOR,TEENY}): now mean library
+	  and API version, and reverted to 1.9.1.  [ruby-dev:37889]
+
+Thu Feb  5 07:39:33 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/readline/readline.c (Init_readline): remove_history(0) may be
+	  NULL.  [ruby-dev:37891]
+
+Thu Feb  5 03:55:22 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (LOCALTIME): should call tzset() before localtime_r().
+	  [ruby-dev:37896]
+
+Thu Feb  5 02:12:35 2009  Alexander Zavorine  <alexandre.zavorine@nokia.com>
+
+	* symbian/configure.bat: Updated Symbian configuration procedure to
+	  avoid creating files outside build directory
+
+	* symbian/pre-build: ditto
+
+	* symbian/setup: ditto
+
+	* symbian/rubyu.def: removed
+
+Thu Feb  5 01:44:02 2009  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_econv_exception): refine error message for
+	  undefined conversion.  [ruby-core:21828]
+
+Thu Feb  5 01:18:25 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_socketpair): make 3rd argument optional.
+
+	* ext/socket/unixsocket.c (unix_s_socketpair): follow the above
+	  change.
+
+	* ext/socket/rubysocket.h (sock_s_socketpair): ditto.
+
+Thu Feb  5 00:09:39 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_ipv6_to_ipv4): new method.
+
+Wed Feb  4 21:59:31 2009  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_econv_exception): show U+XXXX form for undefined
+	  conversion error from UTF-8.
+
+Wed Feb  4 21:57:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_dump): use MBCLEN_CHARFOUND_P properly.
+
+Wed Feb  4 21:55:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb: refine success message.
+
+Wed Feb  4 19:10:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (id.h): updates from parse.h.
+
+Wed Feb  4 11:45:06 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/mkexports.rb: shouldn't export DllMain.
+	  reported at http://pc11.2ch.net/test/read.cgi/tech/1233686068/21
+
+Wed Feb  4 10:12:05 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* missing/vsnprintf.c (BSD_vfprintf): should support 't' format
+	  modifier to handle PRIdPTRDIFF.  thanks for the info from
+	  Kazuhiro NISHIYAMA.  [ruby-core:21807]
+
+Wed Feb  4 01:28:46 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: fix struct in_pktinfo and struct in6_pktinfo
+	  detection.
+
+Wed Feb  4 00:32:59 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/etc/test_etc.rb(test_getpwnam, test_getgrgid, test_getgrnam):
+	  support an environment that has duplicative entries.  a patch from
+	  Tomoyuki Chikanaga <chikanag at nippon-control-system.co.jp> in
+	  [ruby-dev:37882].
+
+Wed Feb  4 00:17:52 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.tcp_server_sockets_port0): skip
+	  socket creation error.
+
+Tue Feb  3 23:37:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): resets EOF flag after parse.
+
+Tue Feb  3 23:13:34 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_backtrace): always returns non-nil array if lev is
+	  negative.  [ruby-core:21795]
+
+Tue Feb  3 21:19:06 2009  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/extconf.rb: checked rl_set_screen_size and
+	  rl_get_screen_size.
+
+	* ext/readline/readline.c (readline_s_set_screen_size): added
+	  Readline.set_screen_size.
+
+	* ext/readline/readline.c (readline_s_get_screen_size): added
+	  Readline.get_screen_size.
+
+Tue Feb  3 21:07:19 2009  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c (readline_s_set_completion_proc): set
+	  default if proc is nil. fix #1095
+
+Tue Feb  3 16:36:06 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* array.c (rb_ary_sort_by_bang): RDoc update.
+
+	* NEWS: add Array#sort_by!.
+
+Tue Feb  3 16:23:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.tcp_server_sockets_port0): new
+	  private function for allocating same port both IPv4 and IPv6.
+	  (Socket.tcp_server_sockets): use tcp_server_sockets_port0 for port 0.
+
+Tue Feb  3 14:12:10 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: validate data before sending to a server.
+	  [ruby-core:20320]
+
+Tue Feb  3 12:35:41 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (match_array): replace match_check().
+
+	* re.c (match_values_at): ditto.
+
+Tue Feb  3 12:09:08 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (hmac_md5): should use String#ord to get ascii
+	  code from the one-character string.
+
+Tue Feb  3 11:25:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_lock): warn for circular require.  [ruby-core:20794],
+	  [ruby-core:20797]
+
+Tue Feb  3 08:35:12 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.tcp_server_sockets): extracted from
+	  Socket.tcp_server_loop.
+	  (Socket.accept_loop): ditto.
+	  (Socket.unix_server_socket): extracted from Socket.unix_server_loop.
+	  (Socket.unix_server_loop): use Socket.accept_loop.
+
+Tue Feb  3 08:21:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_readpartial.rb (make_pipe): readpartial does not
+	  work in text mode.
+
+Tue Feb  3 08:18:26 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (RUBYDEF): uses mkexports.rb on cygwin too.
+
+	* win32/mkexports.rb (Exports::Cygwin): added.
+
+Tue Feb  3 08:10:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (Doxyfile): moved from Makefile.in.
+
+	* template/Doxyfile.tmpl: split from Doxyfile.in.
+
+Tue Feb  3 08:01:38 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write): use of cast expressions as lvalues
+	  is deprecated.
+
+Tue Feb  3 07:57:58 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/extconf.rb (in_pktinfo, in6_pktinfo),
+	  ext/socket/ancdata.c: defined in w32api/ws2tcpip.h on cygwin but
+	  cannot compile for some reason.
+
+Tue Feb  3 07:02:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/langinfo.c (nl_langinfo_codeset): accepts iso-8859
+	  fragment.  [ruby-core:21757]
+
+Tue Feb  3 07:01:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb (patchlevel): config.status may not contain
+	  PATCHLEVEL even if other version numbers exist.
+
+Mon Feb  2 23:43:00 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (Init_addrinfo): add AddrInfo#to_s as an
+	  alias of AddrInfo#to_sockaddr.
+
+	* ext/socket/option.c (Init_sockopt): add Socket::Option#to_s as an
+	  alias of Socket::Option#data.
+
+	  [ruby-dev:37873]
+
+Mon Feb  2 21:04:13 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_sort_by_bang): RDoc update.  [ruby-core:21742]
+
+Mon Feb  2 20:49:24 2009  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c: Introduce id_each to save rb_intern() and SYM2ID()
+	  calls.
+
+Mon Feb  2 19:55:51 2009  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_addrinfo.rb: use AddrInfo.getaddrinfo to generate
+	  IPv6 address.  AddrInfo.ip generates IPv4 address for IPv4 mapped
+	  IPv6 address if --with-lookup-order-hack=INET.
+	  [ruby-dev:37868]
+
+Mon Feb  2 19:17:16 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* doc/NEWS-1.9.1: typo fixed: collect_all -> collect.
+	   [ruby-core:21706]
+
+Mon Feb  2 17:23:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (dir.o, win32.o): depend on win32/dir.h.
+
+Mon Feb  2 17:13:46 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* mkconfig.rb: patchlevel is sometimes minus.
+
+	* win32/resource.rb: ditto.
+
+Mon Feb  2 17:11:23 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (cmsg_type_to_sym): add #ifdef for no IPv6
+	  environment.
+
+Mon Feb  2 17:05:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext): should not be infected from other
+	  load paths.
+
+Mon Feb  2 16:33:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_s_home): new method.   [ruby-core:21454]
+
+Mon Feb  2 16:06:10 2009  Tanaka Akira  <akr@fsij.org>
+
+	* version.h: bump up to 1.9.2 patchlevel -1.
+	  yugui recommend "-1" strongly.
+
+	* lib/rubygems/version.rb: accept negative patchlevel.
+
+Mon Feb  2 14:53:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_initialize): make 3rd argument, protocol,
+	  optional.
+
+Mon Feb  2 14:47:53 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_sort_by_bang): new method.  requested in
+	  [ruby-core:21709]
+
+Mon Feb  2 14:22:56 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/constants.c (cmsg_type_arg): INET6 check.
+
+Mon Feb  2 14:18:20 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c: use INET6 instead of IPV6.
+
+Mon Feb  2 12:47:47 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/{options,raddrinfo}.c: use INET6 instead of AF_INET6 for
+	  VC++6.
+
+Mon Feb  2 12:41:52 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* complex.c (f_signbit): regard NaN as a positive value.
+	  [ruby-dev:37861].
+
+Mon Feb  2 12:39:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/test.rb (valid_syntax?), test/ruby/test_system.rb
+	  (TestSystem::valid_syntax?): use catch and throw instead of
+	  return inside BEGIN block.
+
+Mon Feb  2 11:45:10 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (cmsg_type_arg): declared.
+	  (Init_ancdata): ditto.
+
+	* ext/socket/init.c (Init_socket_init): call Init_ancdata.
+
+	* ext/socket/constants.c (cmsg_type_arg): defined.
+
+	* ext/socket/depend: add dependency for ancdata.o.
+
+	* ext/socket/mkconstants.rb: generate scm_optname_to_int.
+	  more constants.
+
+	* ext/socket/extconf.rb: add ancdata.o.
+
+	* ext/socket/ancdata.c: new file.  new method
+	  BasicSocket#{sendmsg,sendmsg_nonblock,recvmsg,recvmsg_nonblock}
+
+Mon Feb  2 10:57:27 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, win32/dir.h (open_dir_handle, rb_w32_opendir,
+	  move_to_next_entry, rb_w32_readdir, check_valid_dir): change backend
+	  API from A to W.
+
+Mon Feb  2 10:48:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/basicsocket.c (bsock_setsockopt): accept Socket::Option
+	  object.
+	  (bsock_getsockopt): return Socket::Option object.
+
+	* ext/socket/option.c: new file.
+
+	* ext/socket/rubysocket.h (rb_cSockOpt): declared.
+	  (sockopt_new): ditto.
+	  (Init_sockopt): ditto.
+
+	* ext/socket/init.c (Init_socket_init): call Init_sockopt.
+
+	* ext/socket/depend: add dependency for option.o
+
+	* ext/socket/mkconstants.rb: generate intern_level, intern_so_optname,
+	  intern_ip_optname, intern_ipv6_optname, intern_tcp_optname,
+	  intern_udp_optname and intern_scm_optname.
+
+	* ext/socket/extconf.rb: add option.o.
+
+Mon Feb  2 09:49:39 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_backtrace_each): progname is not available at
+	  initializing phase.
+
+Mon Feb  2 08:12:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/xmlrpc/server.rb (Server#serve): gets rid of hardcoded
+	  platform names.
+
+	* lib/resolv.rb (Resolv::Hosts::DefaultFileName),
+	  (Resolv::DNS::Config.default_config_hash): tries win32/resolv on
+	  mswin64 too.
+
+	* lib/rubygems/specification.rb (Gem::Specification#ruby_code):
+	  added mswin64.
+
+	* lib/drb/extservm.rb (DRb::ExtServManager#invoke_service_command):
+	  spawn is better to start a process in background.
+
+	* ext/extmk.rb: uses FNM_SYSCASE.
+
+	* instruby.rb: installs win32.h on mswin64 platform.
+	  [ruby-core:21722]
+
+Mon Feb  2 07:36:13 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_backtrace_each): now takes an iterator function.
+
+	* vm_core.h (rb_make_backtrace, rb_backtrace_each): added
+	  prototypes.
+
+	* vm_dump.c (rb_vm_bugreport), vm_eval.c (rb_backtrace): gets rid
+	  of allocating objects.  [ruby-core:21619]
+
+	* vm_eval.c (rb_backtrace_each): new function which iterates over
+	  each backtrace info.
+
+Mon Feb  2 06:51:36 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encoding): Windows' filesystem encoding is
+	  sometimes ANSI code page and sometimes OEM code page. we should check
+	  whether code page is used.
+
+Sun Feb  1 21:27:55 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_ipv4_private_p): new method.
+	  (addrinfo_ipv4_loopback_p): ditto.
+	  (addrinfo_ipv4_multicast_p): ditto.
+
+Sun Feb  1 16:10:06 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (Switch#summarize): strips an equal sign from
+	  short option, and fills right sides in shorter lines.
+	  [ruby-talk:326414]
+
+Sun Feb  1 05:19:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_const_get_0), vm_insnhelper.c (vm_get_ev_const):
+	  avoids infinite self recursion autoload.  [ruby-core:21696]
+
+Sat Jan 31 22:50:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb (assert_equal): show class in failure
+	  message if meaningful.
+
+Sat Jan 31 22:38:46 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS#each_address): don't query IPv6 address
+	  if the host has no global IPv6 address.
+
+Sat Jan 31 22:29:05 2009  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (STR2CSTR): removed.
+	  (rb_str2cstr): removed.
+
+	* object.c (rb_str2cstr): removed.
+
+Sat Jan 31 20:07:59 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_ipv6_unspecified_p): new method.
+	  (addrinfo_ipv6_loopback_p): ditto.
+	  (addrinfo_ipv6_multicast_p): ditto.
+	  (addrinfo_ipv6_linklocal_p): ditto.
+	  (addrinfo_ipv6_sitelocal_p): ditto.
+	  (addrinfo_ipv6_v4mapped_p): ditto.
+	  (addrinfo_ipv6_v4compat_p): ditto.
+	  (addrinfo_ipv6_mc_nodelocal_p): ditto.
+	  (addrinfo_ipv6_mc_linklocal_p): ditto.
+	  (addrinfo_ipv6_mc_sitelocal_p): ditto.
+	  (addrinfo_ipv6_mc_orglocal_p): ditto.
+	  (addrinfo_ipv6_mc_global_p): ditto.
+
+Sat Jan 31 19:09:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_require_safe): raises when the path to be loaded is
+	  tainted.  [ruby-dev:37843]
+
+Sat Jan 31 18:08:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (optflags): defaulted to -O3 to get rid of slug of
+	  gcc 4.3.
+
+Sat Jan 31 18:03:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (List#accept, parse_in_order): n option is no
+	  longer needed.
+
+Sat Jan 31 14:12:43 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pathname.rb (Pathname#each_child): new method.
+
+Sat Jan 31 00:07:49 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb
+	  (Test::Unit::Assertions#assert_nothing_raised): suppress warning.
+	  [ruby-core:21312]
+
+Fri Jan 30 21:49:32 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pathname.rb (Pathname#realdirpath): new method based on the
+	  patch in [ruby-dev:36560] by Akinori MUSHA.
+	  [ruby-dev:36290]
+
+Fri Jan 30 18:04:23 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write): limit write size to 32KB if the file
+	  seems to be console.  [ruby-core:21613]
+
+Fri Jan 30 16:12:32 2009  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/curses/curses.c (Init_curses): Curses#crmode and
+	  Curses#nocrmode changes to the module function.
+	  fix #916
+
+Fri Jan 30 14:31:14 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/make-snapshot (prereq): remove enc.mk from tarball because
+	  BSD make checks $(srcdir)/enc.mk and try to run $(builddir)/enc.mk.
+
+Fri Jan 30 14:11:48 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/depend: extract compile rules to each target for nmake.
+
+Fri Jan 30 12:59:49 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (RUBYDEF): depends on makefiles.
+
+	* cygwin/GNUmakefile.in (RUBYDEF): adds DATA to non-function symbols
+	  which is not marked as T.
+
+Fri Jan 30 11:03:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (assert_respond_to): gets rid of
+	  overcounting.   [ruby-dev:37703]
+
+Fri Jan 30 02:55:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_init_by_convpath_i): make it static.
+
+Thu Jan 29 16:22:41 2009  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* load.c (rb_feature_provided): should not calculate len by pointer
+	  subtraction because feature may be a expanded path.
+	  [ruby-core:21267]
+
+Thu Jan 29 14:12:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (BASERUBY): erases RUBYOPT at the test.
+
+	* configure.in (OBJCOPY): not used on Windows.
+
+Thu Jan 29 13:16:11 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/securerandom.rb (SecureRandom.uuid): uses unpacked array
+	  instead of string, because String#[] returns one length string.
+
+Thu Jan 29 12:29:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake.rb (FileUtils#rake_system): no longer needs workaround
+	  on Windows.  [ruby-core:21339]
+
+	* lib/rake/win32.rb (Rake::Win32#rake_system): ditto.
+
+	* lib/rake/win32.rb (Rake::Win32#win32_system_dir): no longer
+	  needs environment variables other than APPDATA now.
+
+	* lib/rake.rb (Rake::Application#standard_system_dir): uses
+	  platform specific definition on Windows system.
+
+Thu Jan 29 12:18:54 2009  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* lib/securerandom.rb: new method SecureRandom#uuid
+
+Thu Jan 29 11:22:19 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (inspect_sockaddr): if defined AF_INET6,
+	  perhaps can inspect IPv6 addresses if not defined INET6.
+
+	* ext/socket/socket.c (socket_s_ip_address_list): support Windows XP
+	  or later. (Win2k or earlier is still not supported)
+
+Thu Jan 29 00:24:48 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/zlib/zlib.c (zstream_run): previous change didn't resolve the
+	  problem.
+
+Wed Jan 28 22:51:12 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/zlib/zlib.c (zstream_run): desperately guard the variable.
+	  [ruby-core:20576]
+
+Wed Jan 28 15:24:11 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* math.c: SEGV is caused by implicit rb_to_float declaration
+	  test_complexrational.rb: [BUG] in IA-64 architecture
+
+Tue Jan 27 20:02:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (socks_connect_blocking): moved from
+	  sockssocket.c.  [ruby-dev:37834]
+
+	* sockssocket.c (socks_init): don't omit "int" type.
+
+Tue Jan 27 14:41:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (RUBYDEF): needs DATA marks to export
+	  non-function symbols.  [ruby-core:21582]
+
+	* win32/mkexports.rb (Exports::Mingw#each_export): ditto.
+
+Tue Jan 27 12:59:55 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/ripper/depend: use VPATH.
+	  If a platform doesn't support VPATH, Ruby doesn't support it:
+	  for example MSYS. [ruby-core:21570]
+
+Tue Jan 27 12:21:17 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/ripper/depend: lex.c's correct path is ../../lex.c.
+
+Tue Jan 27 11:09:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_header): checks the header depending on
+	  platform.
+
+	* lib/mkmf.rb (have_header, find_header): use try_header.
+
+	* win32/Makefile.sub (try_header): uses try_compile to get rid of
+	  slow -E option of VC.
+
+Tue Jan 27 11:03:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (distclean-enc, realclean-enc): do not call clean of
+	  enc.mk twice or more.
+
+	* enc/depend (cleanobjs): added deffile.
+
+	* lib/mkmf.rb (create_makefile): removes deffile at clean instead
+	  of distclean.
+
+	* win32/Makefile.sub (miniruby, LIBRUBY_SO): removes lib and exp
+	  files.
+
+	* win32/Makefile.sub (clean, distclean): have moved to common.mk.
+
+	* win32/rmdirs.bat: omits `not empty' message.
+
+Tue Jan 27 10:15:33 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (struct rb_transcoding): added ary member for debug.
+
+Tue Jan 27 10:10:14 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Doxyfile.in (FILE_VERSION_FILTER, INPUT_FILTER): should not use
+	  ./miniruby directly for cross compiling.
+
+Tue Jan 27 04:02:53 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/ripper/depend: fixed dependency path to lex.c.
+	  [ruby-core:21570]
+
+Tue Jan 27 03:43:34 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_any_hash): fixed performance issues with nil, true,
+	  false as hash keys.  a patch from Matthias Waechter.
+	  [ruby-core:21568]
+
+Tue Jan 27 03:23:43 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): should point correct path to
+	  ruby.h and defines.h.   [ruby-core:21569]
+
+Mon Jan 26 16:00:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_aset): fixed arguments evaluation order.
+
+Mon Jan 26 15:49:42 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pathname.rb (chop_basename, prepend_prefix): use o option.
+
+Sun Jan 25 16:35:44 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (socket_s_ip_address_list): renamed from
+	  socket_s_list_ip_address.  [ruby-dev:37806]
+
+Sun Jan 25 12:17:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (RUBYDEF): needs read-only section too.
+
+Sun Jan 25 12:02:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (clean-ext): condition of EXTS was inverted.
+
+Sun Jan 25 11:50:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): fixed the variables order because
+	  converter proc refers the separator.
+
+Sun Jan 25 11:25:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub (RMDIRS): remove directory and
+	  parents.
+
+	* Makefile.in, win32/Makefile.sub (distclean-rdoc): added to remove
+	  temporary rdoc.
+
+	* Makefile.in, win32/Makefile.sub (clean-ext): skips non-existent
+	  directories.
+
+	* common.mk (clean, distclean): cleans rdoc.
+
+	* common.mk (clean-extout): removes extout directory.
+
+	* configure.in (RMDIRS, RMALL): added to clean extout.
+
+	* lib/fileutils.rb (FileUtils#rmdir): added :parents option.
+
+	* lib/mkmf.rb (create_makefile): cleans installed files at clean
+	  instead of distclean.
+
+	* lib/mkmf.rb (create_makefile): added clean-so and clean-rb.
+
+	* lib/mkmf.rb (def init_mkmf): added DISTCLEANDIRS.
+
+	* lib/un.rb (rmdir): added -p option.
+
+	* tool/rmdirs, win32/rmdirs.bat: removes directory and the parents.
+
+	* win32/rm.bat: added -r option.
+
+Sun Jan 25 09:09:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (join_path): use strlcat() to force link.
+
+	* dir.c (glob_helper): no strcpy() is needed since len is known.
+
+Sun Jan 25 06:44:58 2009  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_ssl.c: Server Name Indication support.
+	  new methods SSLContext#server_name_cb=, SSLSocket#hostname=.
+
+	* test/openssl/test_ssl.rb: Tests for above.
+
+Sat Jan 24 08:22:35 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): tools under the top source
+	  directory are not installed, so unusable outside extmk.rb.
+
+Fri Jan 23 17:24:31 2009  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* golf_prelude.rb (Enumerator#inspect): avoid warning.
+
+Fri Jan 23 15:12:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (mkintpath): new function which converts native path
+	  to format acceptable in Makefile.
+
+	* lib/mkmf.rb (configuration): leaves PATH_SEPARATOR unchanged.
+
+	* lib/mkmf.rb (configuration): converts srcdir, topdir and hdrdir.
+	  a patch by Alexey Borzenkov <snaury AT gmail.com> at
+	  [ruby-core:21448].
+
+	* lib/mkmf.rb (try_func): got rid of c-mode confusion.
+
+Fri Jan 23 13:26:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (trap_handler): also accepts symbols.  [ruby-dev:37823]
+
+Thu Jan 22 18:14:04 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): removes installed files under
+	  extout at distclean.
+
+Thu Jan 22 17:12:37 2009  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI.parse): generate only key on params hash
+	  if request have only key uri parameters.
+		(ex. index.cgi?aaa&bbb=1 # params=>{:aaa=>[],:bbb=>["1"]})
+
+	* test/cgi/test_cgi_core.rb: fix test for key only params.
+
+Thu Jan 22 16:29:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/installer.rb (Gem::Installer#initialize): fixed
+	  typos.
+
+	* test/rubygems/test_gem.rb (test_self_dir): removed false
+	  assertions.
+
+	* test/rubygems/test_gem.rb (test_self_set_paths): checks if paths
+	  are included.
+
+	* test/rubygems/test_gem_commands_install_command.rb
+	  (test_execute_remote): checks diagnostic message too.
+
+	* test/rubygems/test_gem_installer.rb (load): uses Gem.ruby.
+
+	* test/rubygems/gemutilities.rb (Gem.ruby): initializes from the
+	  environment variable to run without installation.
+
+	* test/rubygems/gemutilities.rb (RubyGemTestCase#util_build_gem):
+	  creates cache directory.
+
+Thu Jan 22 16:12:51 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* revert previous revision.  it's already out-of-date.
+
+Thu Jan 22 15:54:02 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h, win32/win32.c (rb_w32_is_valid_fd): new function
+	  to validate fd.
+
+	* io.c (rb_io_initialize): check fd with above function.
+
+Thu Jan 22 14:53:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_process.rb (MANDATORY_ENVS): needs RUBYLIB to run
+	  tests without install.
+
+Thu Jan 22 12:19:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (define_final): cannot define finalizer for immediate
+	  values.  [ruby-core:21500]
+
+	* gc.c (define_final): freezes or hides internal values.
+
+Thu Jan 22 11:33:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_time_timeval): made a real prototype.  a patch from
+	  Marcus Rueckert <darix AT opensu.se> at [ruby-core:21492].
+
+Wed Jan 21 21:43:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_ungetbyte, rb_io_ungetc): allows nil to reset EOF
+	  flag with ungetting nothing.
+
+	* ruby.c (load_file_internal): rests EOF flag to make possible to
+	  load from stdin after reading data.
+
+Wed Jan 21 17:17:18 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (ary_double_capa): a new function to expand array more
+	  aggressively.   [ruby-core:21460]
+
+	* array.c (rb_ary_store): use ary_double_capa().
+
+	* array.c (rb_ary_unshift_m): ditto.
+
+	* array.c (rb_ary_splice): ditto.
+
+Wed Jan 21 15:32:15 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_ungetbyte, rb_io_ungetc): clears EOF flag.
+
+Wed Jan 21 14:41:48 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* array.c (take_items): to_ary() raises ArgumentError if cannot to
+	  convert to Array. [ruby-dev:37797]
+
+Wed Jan 21 14:32:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (debug_lines): calls rb_intern() once.
+
+Wed Jan 21 13:58:17 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h (FMODE_EOF): EOF flag on TTY.
+
+	* io.c (io_set_eof): sets EOF flag for TTY.
+
+	* io.c (io_seek): clears EOF flag.
+
+	* io.c (io_fillbuf): returns EOF if already met EOF.  [ruby-dev:37798]
+
+	* io.c (io_fillbuf, io_fread, io_getpartial): sets EOF.
+
+Wed Jan 21 08:22:04 2009  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.3.1 r4532.
+	* test/minitest/*.rb: ditto.
+
+Tue Jan 20 20:16:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (socket_s_list_ip_address): new method.
+	  (sockaddr_obj): new function.
+
+	* ext/socket/rubysocket.h: include ifaddrs.h, sys/ioctl.h,
+	  sys/sockio.h, net/if.h if available.
+	  (addrinfo_new): declared.
+
+	* ext/socket/raddrinfo.c (addrinfo_new): exported.
+
+	* ext/socket/extconf.rb: check ifaddrs.h, sys/ioctl.h, sys/sockio.h,
+	  net/if.h and getifaddrs.
+
+Tue Jan 20 20:05:21 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (pseudo_AF_FTIP): moved from mkconstants.rb.
+
+	* ext/socket/mkconstants.rb: prepend header if -H is not given.
+
+Tue Jan 20 17:50:00 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_buf): use nkf_char.
+
+Tue Jan 20 16:17:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (ruby_Init_{Continuation_body,Fiber_as_Coroutine}): prefixed
+	  with ruby_ to export.
+
+	* ext/continuation/continuation.c, ext/fiber/fiber.c: ditto.
+
+Tue Jan 20 15:32:29 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (miniruby): renames and then removes, to get rid of
+	  EPERM on cygwin and mingw.
+
+	* Makefile.in ($(LIBRUBY_SO)): use wildcard option of objcopy.
+
+	* configure.in (DLDFLAGS): do not export all symbols.
+
+	* cygwin/GNUmakefile.in (RUBYDEF): rejects symbols prefixed with
+	  Init_.
+
+	* win32/mkexports.rb (Exports::Mingw): includes all symbols as
+	  well as mswin32, except for prefixed with Init_.
+
+Tue Jan 20 13:03:50 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (hash): fixed the tail bytes handling in the aligned
+	  access case.
+
+Tue Jan 20 09:26:05 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_buf_push): maybe a bug.
+
+	* ext/nkf/nkf-utf8/nkf.c (options): no need to support help option.
+
+Tue Jan 20 06:48:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (take_items), enum.c (enum_zip): tries to convert to
+	  array first.  [ruby-core:21442]
+
+Tue Jan 20 03:50:37 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Update nkf to 2.0.9.
+	  revert -s meaning as Shift_JIS, etc.
+
+Tue Jan 20 03:42:32 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (hash): fixed typo.  [ruby-dev:37791]
+
+Tue Jan 20 01:15:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (SOMAXCONN): defined.
+
+Mon Jan 19 22:31:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: use erb for generating code.
+
+Mon Jan 19 17:33:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (hash): defaulted to MurmurHash 2.0.
+
+Mon Jan 19 17:29:34 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/defines.h (RUBY_ALIAS_FUNCTION_TYPE): added.
+
+Mon Jan 19 17:24:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (vm_stack_dump_each): used only if debug mode.
+
+Mon Jan 19 16:32:35 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (hash): added MurmurHash 2.0.
+
+Mon Jan 19 14:31:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_execute_interrupts): needs
+	  rb_signal_buff_size to be declared.
+
+Mon Jan 19 13:34:25 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.c (require_libraries): reset th->parse_in_eval while
+	  loading libraries.  fixes [ruby-dev:37780]
+
+Mon Jan 19 11:46:39 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c, eval.c (rb_f_block_given_p): move definition of
+	  "iterator?" and "block_given?" to make static.
+
+	* vm.c (vm_get_ruby_level_caller_cfp): make it static.
+
+	* eval_intern.h, vm_insnhelper.c: move decl. of
+	  vm_get_ruby_level_caller_cfp()
+	  from eval_intern.h to vm_insnhelper.c.
+
+Mon Jan 19 11:27:39 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: add a prefix "rb_" to exposed functions
+	  vm_get_ruby_level_next_cfp(), rb_vm_make_env_object(),
+	  vm_stack_to_heap(), vm_make_proc(), vm_invoke_proc(),
+	  vm_get_sourceline(), vm_cref(), vm_localjump_error(),
+	  vm_make_jump_tag_but_local_jump(), vm_jump_tag_but_local_jump().
+	  This changes may affect only core because most of renamed functions
+	  require a pointer of not-exposed struct such as rb_thread_t or NODE.
+	  In short, they are core functions.
+
+	* cont.c, eval.c, eval_intern.h, load.c, proc.c, thread.c,
+	  vm_core.h, vm_dump.c, vm_eval.c, vm_exec.c, vm_insnhelper.c:
+	  ditto.
+
+Mon Jan 19 11:22:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/rubysocket.h (rb_cUNIXSocket, rb_cUNIXServer),
+	  (unixpath, unixaddr): UNIX domain sockets depend on sys/un.h.
+
+Mon Jan 19 11:02:30 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c: add a prefix "rb_vmdebug_" to
+	  vm_stack_dump_raw(), vm_stack_dump_raw_current(),
+	  vm_env_dump_raw(), vm_proc_dump_raw(), vm_stack_dump_th(),
+	  vm_debug_print_register(), vm_thread_dump_regs(),
+	  vm_debug_print_pre(), vm_debug_print_post(),
+	  vm_thread_dump_state().
+	  This change also may affect core (in fact, user of
+	  above functions may be only ko1).
+
+	* vm_core.h: ditto.
+
+	* vm_exec.h (SDR2): remove duplicate definition.
+
+Mon Jan 19 11:00:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_ip_port): use AF_INET6 only when
+	  defined, as well as addrinfo_ipv6_p().
+
+Mon Jan 19 10:43:38 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect_with_gvl): suppress warnings.
+
+Mon Jan 19 10:34:32 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/depend: workaround for nmake. files in depend without
+	  paths should converted by RULE_SUBST, but mkmf.rb cannot recognize
+	  macros currently.
+
+Mon Jan 19 09:53:43 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c:
+	  rename ruby_iseq_disasm_insn() -> rb_iseq_disasm_insn().
+	  rename ruby_iseq_disasm() -> rb_iseq_disasm().
+
+	* compile.c:
+	  rename ruby_iseq_compile() -> rb_iseq_compile_node().
+	  rename ruby_iseq_translate_threaded_code() ->
+	         rb_iseq_translate_threaded_code().
+	  rename ruby_insns_name_array() -> rb_insns_name_array().
+	  rename ruby_iseq_build_from_ary() -> rb_iseq_build_from_ary().
+
+	* iseq.c, compile.c: remove ruby_insn_make_insn_table() and make
+	  static function insn_make_insn_table().
+
+	* iseq.h, ruby.c, vm.c, vm_core.h, vm_eval.c, vm_dump.c,
+	  blockinlining.c: ditto.
+	  Rename strange "ruby_" prefix to "rb_" prefix.
+	  This changes may affect only core because renamed functions
+	  require a pointer of rb_iseq_t which is not exposed.
+
+Mon Jan 19 09:21:04 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/mkconstants.rb: define macros with default value in
+	  constdefs.h if not exist them.
+
+	* ext/socket/raddrinfo.c (addrinfo_mdump, addrinfo_mload): support
+	  UNIX socket only on platforms which support it.
+
+Mon Jan 19 08:56:53 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, vm_eval.c (rb_f_local_variables): move definition from eval.c
+	  to vm_eval.c because vm_collect_local_variables_in_heap() should
+	  be static function.
+
+	* vm.c (vm_collect_local_variables_in_heap): make it static.
+
+Mon Jan 19 04:06:10 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_load): renamed from ruby_iseq_load, since it is
+	  for C extensions or the ruby core.  [ruby-core:21407]
+
+Mon Jan 19 03:06:22 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (str_transcode0): fix: :xml option doesn't
+	  work on str.encode([options]) form without default_internal.
+
+Sun Jan 18 16:56:46 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_inspect_sockaddr): new
+	  method AddrInfo#inspect_sockaddr.
+	  (inspect_sockaddr): extracted from addrinfo_inspect.
+	  (addrinfo_inspect): use inspect_sockaddr.
+	  (Init_addrinfo): define the new method.
+
+Sun Jan 18 16:46:37 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_ip_address): new method
+	  AddrInfo#ip_address.
+	  (addrinfo_ip_port): new method AddrInfo#ip_port.
+	  (Init_addrinfo): define the methods above.
+
+Sun Jan 18 14:29:52 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/unixsocket.c: redundant #ifdef removed.
+
+Sun Jan 18 03:33:23 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_mdump): don't use symbol.
+	  (addrinfo_mload): ditto.
+
+Sun Jan 18 03:05:20 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_mdump): new method.
+	  (addrinfo_mload): new method.
+	  (Init_addrinfo): define the method above.
+
+	* ext/socket/constants.c (constant_arg): str_to_int's first argument
+	  constified.
+
+	* ext/socket/mkconstants.rb (gen_name_to_int_decl): generated
+	  function's first argument constified.
+	  (gen_name_to_int_func_in_guard): ditto.
+	  (ipproto_to_int): generated.
+
+	* ext/socket/rubysocket.h (IS_IP_FAMILY): moved from raddrinfo.c.
+
+Sun Jan 18 01:37:50 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_getnameinfo): accept AddrInfo object.
+
+	* ext/socket/raddrinfo.c (rb_check_sockaddr_string_type): defined.
+
+	* ext/socket/rubysocket.h (rb_check_sockaddr_string_type): declared.
+
+Sat Jan 17 22:01:15 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb: new file.
+
+Sat Jan 17 19:33:48 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* configure.in (VCSUP): fixed the cases for git-svn or git.
+
+	* win32/Makefile.sub (VCSUP): ditto.
+
+	* Makefile.in (up): `cd' is necessary for git and git-svn.
+
+Sat Jan 17 19:16:16 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: generate a header file for generated
+	  functions.
+
+	* ext/socket/rubysocket.h: include constdefs.h.  don't declare
+	  generated functions.
+
+	* ext/socket/constants.c: include constdefs.c instead of constants.h.
+
+	* ext/socket/depend: dependency updated.
+
+Sat Jan 17 17:58:22 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/xmp.rb: multilingualizes XMP::StringInputMethod.
+	  [ruby-core:21383].
+
+Sat Jan 17 15:01:22 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (make_inspectname): add a res argument to
+	  suppress numeric inspectname.
+	  (init_addrinfo_getaddrinfo): call make_inspectname here.
+	  (addrinfo_firstonly_new): follow make_inspectname change.
+	  (addrinfo_list_new): ditto.
+	  (addrinfo_initialize): follow init_addrinfo_getaddrinfo change.
+
+Sat Jan 17 14:52:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: don't cause an error for duplicate names.
+
+Sat Jan 17 12:46:17 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: split files for each class.
+
+	* ext/socket/rubysocket.h: common header.
+
+	* ext/socket/basicsocket.c: new file for BasicSocket.
+
+	* ext/socket/ipsocket.c: new file for IPSocket.
+
+	* ext/socket/tcpsocket.c: new file for TCPSocket.
+
+	* ext/socket/tcpserver.c: new file for TCPServer.
+
+	* ext/socket/sockssocket.c: new file for SOCKSSocket.
+
+	* ext/socket/udpsocket.c: new file for UDPSocket.
+
+	* ext/socket/unixsocket.c: new file for UNIXSocket.
+
+	* ext/socket/unixserver.c: new file for UNIXServer.
+
+	* ext/socket/socket.c: now for Socket.
+
+	* ext/socket/raddrinfo.c: new file for AddrInfo and name resolution.
+
+	* ext/socket/constants.c: new file for constants.
+
+	* ext/socket/init.c: new file for utilities.
+
+	* ext/socket/mkconstants.rb: export *_to_int.
+
+	* ext/socket/extconf.rb: add new object files.
+
+	* ext/socket/depend: add dependencies for new files.
+
+	* ext/.document: add new files.
+
+Sat Jan 17 11:12:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_restore_0): padding size doesn't need to be large
+	  if alloca is used.  suppress warnings.
+
+Sat Jan 17 11:12:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (vm_stack_dump_each): initialized at declarations.
+
+	* vm_dump.c (rb_vm_bugreport): constified to suppress a warning.
+
+Fri Jan 16 22:30:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c: move addrinfo code.
+
+Fri Jan 16 18:51:11 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (up): tell nmake that need to run command via
+	  shell.
+
+Fri Jan 16 18:43:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): includes all rules and expand
+	  configured values from the environment to create *.inc, sets RM
+	  for ripper.c, and needs chdir if absolute path is given with
+	  -exported option.
+
+Fri Jan 16 18:26:47 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (VALUE): use unsigned long or long long
+	  instead of uintptr_t, since many %lx and so on are still used.
+
+Fri Jan 16 17:33:59 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (Makefile): set VCS and VCSUP to support `git svn'
+	  and git.
+
+	* Makefile.in, win32/Makefile.sub (up): split from common.mk.
+
+Fri Jan 16 12:24:39 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (thread_start_func_2): call ruby_cleanup() if thread is
+	  main thread.  [ruby-dev:37624]
+
+Fri Jan 16 12:27:50 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub ($(INSNS), node_name.inc, known_errors.inc,
+	  miniprelude.c, newline.c): if dependencies is newer than targets or
+	  targets don't exist, try to create targets with BASERUBY. no need to
+	  detect errors there because the absence of BASERUBY is not abnormal.
+	  after the try, if the targets still don't exist, copy them from
+	  $(srcdir).
+
+Fri Jan 16 12:15:07 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRI_PTRDIFF_PREFIX): must define as string.
+	  if not, cause compile error in using PRI?VALUE.
+
+	* win32/Makefile.sub (config.h): add SIZEOF_INTPTR_T and
+	  SIZEOF_UINTPTR_T for SIZEOF_VALUE.
+
+Fri Jan 16 11:51:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbian/pre-build (COMSPEC): should not overridden.
+
+Fri Jan 16 11:47:56 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_parse_in_main): fixed typo.
+
+Fri Jan 16 11:41:23 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_REPLACE_TYPE): does not define rb_ prefixed
+	  name if no default type is given.
+
+	* configure.in (RUBY_DEFINT): checks size.
+
+	* include/ruby/ruby.h (PRI_PTRDIFF_PREFIX, PRI_SIZE_PREFIX):
+	  assumes usable if PRIdPTR is defined.
+
+Fri Jan 16 02:25:46 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_thread_execute_interrupts): if signal is already
+	  buffered, main thread should wait until timer thread delivers it.
+
+	* thread.c (timer_thread_function): should defer delivery of a signal
+	  if main thread does not yet trap a previous one.  [ruby-dev:37676]
+
+Fri Jan 16 02:05:55 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (addrinfo_s_ip): new method AddrInfo.ip.
+
+Fri Jan 16 01:42:50 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (call_trace_proc): as Matz said ([ruby-core:21183]),
+	  should skip rb_str_new2() if rb_sourcefile() returns NULL.
+	  rb_sourcefile() returns NULL if frame is toplevel of Fiber.
+	  [ruby-core:21161] [Bug #985]
+
+Fri Jan 16 01:09:37 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.c (process_options): decrement parse_in_eval to recognize
+	  parsing main or normal eval script.
+
+	* compile.c (rb_parse_in_main): return 1 if parsing main script.
+	 (if parse_in_eval is negative value, it means main script)
+
+	* parse.y (yycompile0): check rb_parse_in_main() to accumulate
+	  script text.  Bug #848 [ruby-core:20450]
+
+Fri Jan 16 00:57:34 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/debug.rb: as wanabe-san pointed out,
+	  set_trace_func should be completely final task in debug.rb.
+	  Bug #847 [ruby-core:20449]
+
+Fri Jan 16 00:27:03 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_inc_const_missing_count, ruby_vm_const_missing_count):
+	  added.
+
+	* vm_insnhelper.h: ditto.
+
+	* variable.c (rb_const_get_0), insns.def: Constants should not be
+	  cached if const_missing is called.  [ruby-core:21059] [Bug #967]
+
+	* bootstraptest/test_class.rb: add a test.
+
+Fri Jan 16 00:25:09 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: btest-ruby should receive option with OPTS.
+
+Fri Jan 16 00:21:10 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub ($(INSNS), node_name.inc, known_errors.inc,
+	  miniprelude.c, newline.c): shouldn't copy when $(srcdir) != ".", too.
+	  (this is a workaround.  we need more strict check.)
+
+	* win32/Makefile.sub ($(INSNS)): shouldn't remove BASERUBY.
+
+Fri Jan 16 00:19:55 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_with_gvl): suppress warnings.
+
+Thu Jan 15 20:44:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* tool/file2lastrev.rb: RUBY_REVISION must be an integer.
+
+Thu Jan 15 15:55:31 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (enc_set_default_encoding): should not treat US-ASCII
+	  special here.  a patch from Yui NARUSE in [ruby-dev:37769].
+	  [ruby-dev:37699]
+
+Thu Jan 15 14:27:27 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* LEGAL: Added information for missing/langinfo.c
+
+Thu Jan 15 14:17:57 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (stmt): explicit error for "Object::Far += foo 1" just
+	  as "Object::Far += 1".
+
+	* parse.y (command): moved return/break/next from command_call for
+	  better error message.
+
+	* parse.y (call_args): void value check added.
+
+Thu Jan 15 13:10:58 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (COMPILERFLAG): for enc/trans/gb18030.c.
+
+Thu Jan 15 13:10:09 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, win32/Makefile.sub (MISSING): added langinfo on mingw
+	  and mswin.
+
+	* encoding.c (rb_locale_charmap): use environments on mingw and mswin.
+
+	* missing/langinfo.c (nl_langinfo_codeset): MS-Windows Japanese
+	  environment uses Windows-31J derived from Shift_JIS, not EUC-JP.
+
+Thu Jan 15 12:10:39 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub ($(INSNS), node_name.inc, known_errors.inc,
+	  miniprelude.c, newline.c): copy only when $(srcdir) != pwd.
+
+Thu Jan 15 11:40:44 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (pipe_open): revert a part of previous commit. it may be a
+	  mistake.
+
+Thu Jan 15 10:45:52 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* symbian/README.SYMBIAN: symbian support added.  great appreciate
+	  to <alexandre.zavorine at symbian.com>.
+
+Thu Jan 15 00:06:12 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/gb18030.trans: get rid of a 1.9 feature for cross
+	  compile.  [ruby-core:21345]
+
+Wed Jan 14 23:57:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_locale_charmap): fallback to codepage if no
+	  locale is found.  [ruby-core:21110]
+
+	* missing/langinfo.c (nl_langinfo_codeset): returns NULL if no
+	  locale is found.
+
+Wed Jan 14 22:38:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (negative_size_allocation_error_with_gvl): abolish a warning.
+	  (negative_size_allocation_error): ditto.
+
+	* ext/openssl/ossl.c (ossl_raise): ditto.
+
+Wed Jan 14 20:05:05 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/gb18030.trans, gb18030-tbl.rb:
+	  new Chinese GB18030 transcoding (from Yoshihiro Kambayashi)
+
+	* test/ruby/test_transcode.rb: added tests for the above
+	  (from Yoshihiro Kambayashi)
+
+	* transcode_data.h, transcode.c, tool/transcode_tblgen.rb:
+	  added support for GB18030-specific 4-byte sequences
+	  (with Yoshihiro Kambayashi)
+
+Wed Jan 14 16:16:19 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/extconf.rb: check ncursesw earlier than ncurses to
+	  support UTF-8 strings.  non UTF-8 strings should be converted
+	  explicitly.  [ruby-core:21094]
+
+Wed Jan 14 14:42:30 2009  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* cont.c: fix prototype declare of register_stack_extend
+
+Wed Jan 14 13:19:21 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_aspawn): should not escape with caret
+	  unless using cmd.exe.
+
+Wed Jan 14 13:09:00 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/extconf.rb (gai_strerror): checks if available and if
+	  returns const pointer.
+
+	* ext/socket/getaddrinfo.c (gai_strerror): defines only if non
+	  available.  [ruby-core:21328]
+
+Wed Jan 14 12:39:16 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_wait_readable, rb_io_wait_writable): need rb_fd_ptr.
+
+	* io.c (appendline): reformed.
+
+	* io.c (rb_io_s_pipe): reduced nest of rb_ensure of main block.
+
+Tue Jan 13 22:56:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/addrinfo.h (AI_NUMERICSERV): defined.
+	  (AI_MASK): add AI_NUMERICSERV.
+
+	* ext/socket/getaddrinfo.c (getaddrinfo): support AI_NUMERICSERV.
+
+Tue Jan 13 21:45:53 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/ostruct.rb (OpenStruct#inspect): fixed the recursion check.
+	  Patch by Kornelius Kalnbach.  [ruby-core:20992].
+
+	* test/ostruct/test_ostruct.rb: test for inspect.
+	  Patch by Kornelius Kalnbach.  [ruby-core:20992].
+
+Tue Jan 13 21:44:30 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_reopen, rb_io_init_copy): should register fptr to pipe_list
+	  when copying pipe fptr.
+
+Tue Jan 13 21:38:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rb_getaddrinfo): use getaddrinfo with GVL if
+	  getaddrinfo.c/getnameinfo.c is used.  They are not thread safe
+	  because gethostbyname/gethostbyaddr is used.
+	  (rb_getnameinfo): ditto.
+
+	* extconf.rb: define GETADDRINFO_EMU if getaddrinfo.c/getnameinfo.c is
+	  used.
+
+Tue Jan 13 21:28:14 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* object.c (rb_obj_not_match): rdoc.
+	  Patch by Kornelius Kalnbach.  [ruby-core:20991]
+
+Tue Jan 13 18:21:44 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_close_read): call rb_io_fptr_cleanup() instead of
+	  fptr_finalize() because the fptr has special finalizer if it is a
+	  pipe.  [ruby-dev:37757] (3)
+
+Tue Jan 13 18:19:49 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_initialize): workaround for Windows. [ruby-dev:37686]
+	  (also see [ruby-dev:37721])
+
+Tue Jan 13 17:29:02 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/make_transdb.rb (converters): should not depend on the
+	  hash order for cross compile.
+
+Tue Jan 13 16:39:11 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/intern.h, thread.c, win32/Makefile.sub (rb_fdset_t,
+	  rb_fd_init, rb_fd_term, rb_fd_zero, rb_fd_set, rb_fd_clr, rb_fd_isset,
+	  rb_fd_select, rb_fd_ptr, rb_fd_max, HAVE_RB_FD_INIT): new type,
+	  functions, and macros for Windows.
+
+	* win32/win32.c (extract_fd, rb_w32_select): use rb_fdset_t to expand
+	  fd_array if needed.  [ruby-core:19946]
+
+	* win32/win32.c (copy_fd): new function for rb_w32_select().
+
+Tue Jan 13 12:31:54 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/file2lastrev.rb (get_revisions): fixes problem with
+	  svn on cygwin. [ruby-dev:37702].
+	  Patch by Kouhei Sutou.
+
+Tue Jan 13 11:58:04 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/input-method.rb: IRB did not prompt for MSwin32.
+	  [ruby-dev:37686].
+	  Patch by arton <artonx AT yahoo.co.jp>.
+
+Tue Jan 13 12:10:42 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/{getaddrinfo,getnameinfo}.c: need to include ws2tcpip.h
+	  on Windows.
+
+Tue Jan 13 09:49:49 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* ext/socket/socket.c: removed warning about unused variable 'rai'.
+
+Tue Jan 13 03:07:28 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* instruby.rb: require "tool/mdoc2man.rb" before chdir if needed.
+
+Tue Jan 13 02:54:54 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: use headers instead of "netdb.h" in checking
+	  getnameinfo() and getaddrinfo() because Windows doesn't have it.
+	  [ruby-dev:37757] (1)
+
+	* ext/socket/sockport.h (SA_LEN): use sockaddr_in6 when defined AF_INET6
+	  if INET6 is not defined.  winsock2's getaddrinfo() returns
+	  sockaddr_in6 if ipv6 is available.
+
+Tue Jan 13 02:21:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (internal_cmd_match): extracted from
+	  is_internal_cmd.
+
+	* win32/win32.c (argv_size, join_argv): escapes redirection, pipe
+	  and caret punctuations with carets.
+
+	* win32/win32.c (rb_w32_aspawn): ditto, and redirections and pipe
+	  have no meanings.
+
+Tue Jan 13 02:07:38 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: use Encoding.find("locale") instead of
+	  Encoding.locale_charmap
+
+Tue Jan 13 00:57:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_unpack_sockaddr_in): check too short
+	  sockaddr.
+	  (sock_s_unpack_sockaddr_un): ditto.
+
+Mon Jan 12 23:55:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (addrinfo_ip_unpack): new method
+	  AddrInfo#ip_unpack.
+	  (addrinfo_unix_path): new method AddrInfo#unix_path.
+	  (Init_socket): define above methods.
+
+Mon Jan 12 23:31:42 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (IS_IP_FAMILY): defined.
+	  (addrinfo_ip_p): use it.
+
+Mon Jan 12 17:23:05 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (assert_nothing_raised): removes the
+	  current trace to get rid of an issue of MiniTest::Unit#location.
+
+Mon Jan 12 16:49:20 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (install): erased a garbage.
+
+Mon Jan 12 16:45:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (open_dir_handle): extracted from rb_w32_opendir.
+
+	* win32/win32.c (winnt_stat): gets rid of strange behavior of
+	  GetFileAttributes().  [ruby-core:21269]
+
+Mon Jan 12 12:43:55 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (parse_args, install): added --strip option.
+
+	* instruby.rb (install_recursive): skips the directory if matched
+	  to a no_install pattern ending with a file separator.
+
+Mon Jan 12 12:33:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: don't use bytesize because 1.9 dependent.
+	  [ruby-core:21266]
+
+Mon Jan 12 11:59:19 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (ruby_xmalloc, ruby_xrealloc, ruby_xfree):
+	  enable to use them without GVL.
+	  if GC is invoked, acquire GVL during GC.
+	  if NoMemoryError is raised, acquire GVL and raise it.
+
+Mon Jan 12 10:39:19 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: fix comment (terminology: Global VM Lock).
+
+	* thread.c (blocking_region_begin, blocking_region_end):
+	  save and clear th->blocking_region_buffer.
+
+	* thread.c (rb_thread_call_with_gvl): check if it has GVL.
+
+	* thread.c (ruby_thread_has_gvl_p): added.
+
+	* vm_core.h: add decls.
+
+Mon Jan 12 10:21:11 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c: remove unused decl.
+
+Sun Jan 11 16:53:14 2009  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/erb.rb (PercentScanner): remove PercentScanner. fixed % after
+	  %> bug. [ruby-dev:37751] [Bug #997]
+
+	* test/erb/test_erb.rb: ditto
+
+Sun Jan 11 09:53:01 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/mkconstants.rb (c_str): get rid of a 1.9 feature for
+	  cross compile.  [ruby-core:21243]
+
+Sun Jan 11 09:47:30 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_s_create): set nil as the value if assoc length
+	  is not enough.  [ruby-core:21249]
+
+Sat Jan 10 21:17:28 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: don't generate unintended newlines.
+
+Sat Jan 10 20:50:02 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: refactored to nest properly in statically
+	  and dynamically.
+
+Sat Jan 10 20:26:17 2009  Tanaka Akira  <akr@fsij.org>
+
+	* version.h: make version string succinct:
+	  "ruby 1.9.1p5000 (2009-01-10 trunk 21414) [i686-linux]".
+
+Sat Jan 10 19:05:13 2009  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (cont_restore_1): should not be inlined.
+
+Fri Jan  9 21:52:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_define_const): use INT2NUM.
+
+Fri Jan  9 15:40:07 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/mkconstants.rb: define macro if default_value is passed.
+
+Fri Jan  9 13:20:08 2009  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: rescue abnormal Encoding.locale_charmap value.
+
+Fri Jan  9 12:29:46 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (test-sample), rubytest.rb: use runruby.rb
+
+Fri Jan  9 10:40:52 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_process.rb (TestProcess::MANDATORY_ENVS): some
+	  platforms need environments for shared objects.
+
+Thu Jan  8 23:19:38 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (bsock_shutdown): accept symbol/string as how.
+	  (shutdown_how_arg): new function.
+
+	* ext/socket/mkconstants.rb: generate shutdown_how_to_int.
+
+Thu Jan  8 23:16:37 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* instruby.rb: easy code contains fewer bugs.
+
+Thu Jan  8 22:59:30 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (gen_name_to_int_func): generate
+	  name_to_int functions, not only body.
+
+Thu Jan  8 22:44:10 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c: don't use AI_NUMERICSERV for platforms which
+	  not define it as old Windows.
+	  [ruby-dev:37736]
+
+Thu Jan  8 17:32:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb: should not depend on a library which does not exist
+	  in 1.8.
+
+Thu Jan  8 07:52:55 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (init_unix_addrinfo): don't return a value.
+
+Thu Jan  8 07:30:52 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI.redirectable?): reverted https redirection.
+
+Thu Jan  8 00:31:58 2009  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb (assert_no_match): don't count up as two
+	  assertions.
+
+Thu Jan  8 00:26:35 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (addrinfo_getnameinfo): use NI_DGRAM if socktype
+	  is SOCK_DGRAM.
+
+Thu Jan  8 00:16:22 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (host_str): add flags_ptr argument to specify
+	  AI_NUMERICHOST if host is numeric form.
+	  (port_str): add flags_ptr argument to specify AI_NUMERICSERV if port
+	  is numeric form.
+
+Wed Jan  7 22:24:12 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rb_cAddrInfo): new class AddrInfo.
+	  (s_recvfrom): return AddrInfo as address.
+	  (s_recvfrom_nonblock): ditto.
+	  (sock_accept): ditto.
+	  (sock_accept_nonblock): ditto.
+	  (sock_sysaccept): ditto.
+	  (bsock_send): accept AddrInfo as address argument.
+	  (sock_connect): ditto.
+	  (sock_connect_nonblock): ditto.
+	  (sock_bind): ditto.
+	  (sock_s_unpack_sockaddr_in): ditto.
+	  (sock_s_unpack_sockaddr_un): ditto.
+	  (bsock_local_address): new method BasicSocket#local_address.
+	  (bsock_remote_address): new method BasicSocket#remote_address.
+	  (addrinfo_initialize): new method AddrInfo#initialize.
+	  (addrinfo_inspect): new method AddrInfo#inspect.
+	  (addrinfo_afamily): new method AddrInfo#afamily.
+	  (addrinfo_pfamily): new method AddrInfo#pfamily.
+	  (addrinfo_socktype): new method AddrInfo#socktype.
+	  (addrinfo_protocol): new method AddrInfo#protocol.
+	  (addrinfo_to_sockaddr): new method AddrInfo#to_sockaddr.
+	  (addrinfo_canonname): new method AddrInfo#canonname.
+	  (addrinfo_ip_p): new method AddrInfo#ip?.
+	  (addrinfo_ipv4_p): new method AddrInfo#ipv4?.
+	  (addrinfo_ipv6_p): new method AddrInfo#ipv6?.
+	  (addrinfo_unix_p): new method AddrInfo#unix?.
+	  (addrinfo_getnameinfo): new method AddrInfo#getnameinfo.
+	  (addrinfo_s_getaddrinfo): new method AddrInfo.getaddrinfo.
+	  (addrinfo_s_tcp): new method AddrInfo.tcp.
+	  (addrinfo_s_udp): new method AddrInfo.udp.
+	  (addrinfo_s_unix): new method AddrInfo.unix.
+	  (Init_socket): define new class and methods.
+	  (sock_getaddrinfo): apply socktype hack regardless of ai_flags.
+	  (addrinfo_new): defined.
+	  (get_afamily): ditto.
+	  (fd_socket_addrinfo): ditto.
+	  (io_socket_addrinfo): ditto.
+	  (SockAddrStringValue): ditto.
+	  (SockAddrStringValuePtr): ditto.
+	  (sockaddr_string_value): ditto.
+	  (sockaddr_string_value_ptr): ditto.
+	  (rb_addrinfo_t): ditto.
+	  (addrinfo_mark): ditto.
+	  (addrinfo_free): ditto.
+	  (addrinfo_s_allocate): ditto.
+	  (IS_ADDRINFO): ditto.
+	  (check_addrinfo): ditto.
+	  (get_addrinfo): ditto.
+	  (alloc_addrinfo): ditto.
+	  (init_addrinfo): ditto.
+	  (addrinfo_new): ditto.
+	  (call_getaddrinfo): ditto.
+	  (init_addrinfo_getaddrinfo): ditto.
+	  (make_inspectname): ditto.
+	  (addrinfo_firstonly_new): ditto.
+	  (addrinfo_list_new): ditto.
+	  (init_unix_addrinfo): ditto.
+	  (ai_get_afamily): ditto.
+
+	* ext/socket/mkconstants.rb: generate intern_protocol_family,
+	  intern_socktype and intern_ipproto.
+
+	  [ruby-dev:37692]
+
+Wed Jan  7 22:13:03 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (man, gem): fixed errors.
+
+Wed Jan  7 17:14:40 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb: use installed libraries.  [ruby-core:21006]
+
+	* instruby.rb (gem), lib/rubygems/defaults.rb (Gem.default_dir):
+	  use version invariant configuration.
+
+Tue Jan  6 19:09:51 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_restore_0): streamlined to ensure O(1) time.  based on
+	  a patch by Brent Roman <brent AT mbari.org>.
+
+Tue Jan  6 00:34:25 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_close_before_exec): more heuristics to detect maximum fd.
+
+Mon Jan  5 17:59:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (cygwin): needs properly implemented nl_langinfo().
+	  [ruby-core:21110]
+
+	* missing/langinfo.c: added as suggested at [ruby-core:21015], from
+	  http://www.cl.cam.ac.uk/~mgk25/ucs/langinfo.c.
+
+	* missing/langinfo.c (nl_langinfo_0): fixed typos for SJIS.
+
+Mon Jan  5 17:38:40 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* encoding.c (rb_locale_charmap): use GetConsoleCP() instead of
+	  GetACP() because external encoding should be console's one.
+
+	* encoding.c (rb_filesystem_encoding): use GetOEMCP() instead of
+	  GetACP() because VFAT/FAT32 uses OEM CP.
+
+Mon Jan  5 16:26:48 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encoding): use ANSI codepage for file
+	  system on cygwin.
+
+	* encoding.c (rb_locale_charmap): reverted.  [ruby-core:21110]
+
+Mon Jan  5 16:15:00 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/curses/curses.c (free_window): use xfree instead of free.
+	  [ruby-dev:37200]
+
+Mon Jan  5 15:49:45 2009  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/gdbm/gdbm.c (rb_gdbm_fetch): remove needless cast.
+
+Mon Jan  5 12:52:08 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (init_env): use user profile folder than personal
+	  folder.
+
+Mon Jan  5 08:41:13 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_file_s_stat): need type check for non string values.
+	  [ruby-dev:37673]  fix: #964
+
+	* file.c (rb_file_s_size, rb_file_s_atime, rb_file_s_mtime,
+	  rb_file_s_ctime, rb_f_test): ditto.
+
+Mon Jan  5 08:17:56 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_step): should not add up errors on loops.
+	  [ruby-dev:37691]
+
+Mon Jan  5 07:58:37 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* random.c (rb_f_srand): update RDoc.  [ruby-core:21113]
+
+Mon Jan  5 06:39:56 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_max): max value from ... not defined for non
+	  Integer Numeric end values. [ruby-dev:37690] fix: #974
+
+Sun Jan  4 22:37:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rb_getaddrinfo): defined for address lookup without GVL.
+	  (struct getaddrinfo_arg): defined.
+	  (nogvl_getaddrinfo): defined.
+	  (rb_getnameinfo): defined for name lookup without GVL.
+	  (getnameinfo_arg): defined.
+	  (nogvl_getnameinfo): defined.
+
+Sun Jan  4 18:10:10 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/gbk.trans, gbk-tbl.rb:
+	  new Chinese GBK transcoding (from Yoshihiro Kambayashi)
+
+	* test/ruby/test_transcode.rb: added tests for the above
+	  (from Yoshihiro Kambayashi)
+
+Sun Jan  4 17:55:55 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb: added tests for GB2312
+	  (from Yoshihiro Kambayashi)
+
+	* enc/trans/chinese.trans: set valid byte patterns for
+	  GB2312 and GB12345
+
+Sun Jan  4 17:39:39 2009  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/big5.trans, big5-tbl.rb:
+	  new Chinese Big5 transcoding (from Yoshihiro Kambayashi)
+
+	* test/ruby/test_transcode.rb: added tests for the above
+	  (from Yoshihiro Kambayashi)
+
+Sun Jan  4 17:07:45 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems.rb (Gem.set_home, Gem.set_paths): should not create
+	  directories stealthily.  [ruby-core:20990]
+
+	* lib/rubygems.rb (Gem.find_home): expand_path deals with platform
+	  dependent environments.
+
+	* lib/rdoc/ri/paths.rb (RDoc::HOMEDIR): ditto.
+
+	* instruby.rb (gem): creates gem directories at installation.
+
+Sun Jan  4 15:41:37 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encoding, rb_locale_charmap): uses
+	  codepage on cygwin.  [ruby-core:20994]
+
+Sun Jan  4 11:58:43 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (ruby_float_step): extracted from num_step().
+
+	* range.c (range_step): uses ruby_float_step() for float range.
+	  [ruby-dev:37691]
+
+Sun Jan  4 11:11:31 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): does not use both of makefile.rb and
+	  extconf.rb at the same time.
+
+	* lib/mkmf.rb (DLLIB): depends on Makefile.  [ruby-core:21096]
+
+Sun Jan  4 09:27:41 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (constant_arg): constified.
+
+	* ext/socket/socket.c (optname_arg): cygwin does not have
+	  IPPROTO_IPV6.
+
+	* ext/socket/mkconstants.rb (ipv6_optname_to_int): ditto.
+
+Sun Jan  4 04:33:14 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c: redundant #ifdef removed.
+
+Sun Jan  4 03:30:18 2009  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (maygvl_copy_stream_wait_read): renamed to add prefix to
+	  express GVL state.
+	  (nogvl_copy_stream_wait_write): ditto.
+	  (nogvl_copy_stream_sendfile): ditto.
+	  (maygvl_copy_stream_read): ditto.
+	  (nogvl_copy_stream_write): ditto.
+	  (nogvl_copy_stream_read_write): ditto.
+	  (nogvl_copy_stream_func): ditto.
+
+Sun Jan  4 00:30:50 2009  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/erb.rb: merged r20850, r17881, r16811, r16763, r16748, r15829,
+	  r15794 and r15698 from ruby_1_8.
+
+	* test/erb/test_erb.rb: ditto.
+
+Sat Jan  3 22:24:36 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk, Makefile.in, win32/Makefile.sub (INSNS): move the macro
+	  definition from common.mk to {Makefile.in,win32/Makefile.sub}.
+	  [ruby-dev:37678]
+
+Sat Jan  3 15:30:09 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/chinese.trans: added for transcoding EUC-CN and GB12345.
+
+	* enc/trans/GB/: ditto.
+
+Sat Jan  3 15:26:30 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/euc-cn.c: renamed from gb2312.c because GB2312 is
+	  the name of a CSS.  it's encoding name is "EUC-CN".
+
+Sat Jan  3 03:00:42 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk, Makefile.in, win32/Makefile.sub ($(INSNS), node_name.inc,
+	  known_errors.inc, miniprelude.c, newline.c): move rules from common.mk
+	  to {Makefile.in,win32/Makefile.sub} for nmake.
+	  [ruby-core:20993]
+
+Fri Jan  2 23:51:41 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_sleep.rb (test_sleep_5sec): stop gc.
+
+Fri Jan  2 23:36:10 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c: don't apply socktype hack [ruby-core:184] for
+	  sock_s_getaddrinfo.
+	  (sock_getaddrinfo): add socktype_hack argument.
+	  (sock_addrinfo): call sock_getaddrinfo with socktype_hack.
+	  (sock_s_getaddrinfo): call sock_getaddrinfo without socktype_hack.
+	  [ruby-dev:37674]
+
+Fri Jan  2 23:33:38 2009  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/missing.h, sprintf.c: get rid of a warning of VC++.
+
+Fri Jan  2 22:25:46 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* process.c: use sigaction instead of sigset.  [ruby-core:21021]
+
+Fri Jan  2 22:22:04 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bootstraptest/test_proc.rb: suppress error message.
+
+Fri Jan  2 22:20:46 2009  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_process.rb (test_wait_and_sigchild): remove debug
+	  output.
+
+Fri Jan  2 17:45:19 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: rename family_to_str to intern_family.
+	  it returns ID now.
+
+	* ext/socket/socket.c (ipaddr): use intern_family.
+
+Fri Jan  2 17:20:31 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: use ID in the values of
+	  family_to_str_hash.  family_to_str returns a VALUE.
+
+	* ext/socket/socket.c (ipaddr): follow family_to_str change.
+
+Fri Jan  2 17:01:51 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: refactored to gather stringizer
+	  description.
+
+Fri Jan  2 15:30:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: make common prefix optional.
+
+Fri Jan  2 14:59:52 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: use hash for family_to_str to avoid
+	  linear search.  lenp argument removed.
+
+	* ext/socket/socket.c (ipaddr): call family_to_str without lenp
+	  argument.
+
+Fri Jan  2 14:33:12 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (gen_int_to_name): don't compare constants
+	  in preprocessor because the constants may be enum.
+
+Fri Jan  2 14:11:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (gen_int_to_name): add lenp argument.
+
+	* ext/socket/socket.c (ipaddr): follow argument change.
+
+Fri Jan  2 10:20:24 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/singleton.rb: fix documentation. [ruby-core:21038]
+
+Fri Jan  2 06:43:58 2009  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* signal.c (rb_signal_buff_size): defined for check whether signal
+	  is in the buffer or not.
+
+	* thread_pthread.c (thread_timer): don't exit the loop when signal
+	  is in the buffer. [ruby-dev:37637]
+
+Fri Jan  2 04:40:00 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: check duplicates.
+
+Fri Jan  2 02:39:08 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (optname_arg): defined.
+	  (bsock_setsockopt): use optname_arg.
+	  (bsock_getsockopt): ditto.
+
+	* ext/socket/mkconstants.rb: generate so_optname_to_int,
+	  ip_optname_to_int, ipv6_optname_to_int, tcp_optname_to_int and
+	  udp_optname_to_int.
+	  more constants.
+
+Fri Jan  2 02:08:36 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (constant_arg): extracted from family_arg.
+	  (family_arg): use constant_arg.
+	  (socktype_arg): ditto.
+	  (level_arg): ditto.
+
+Fri Jan  2 01:15:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (level_arg): defined.
+	  (bsock_setsockopt): use level_arg.
+	  (bsock_getsockopt): ditto.
+
+	* ext/socket/mkconstants.rb: generate level_to_int.
+
+Fri Jan  2 00:58:40 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (family_arg): accept symbols as well.
+	  (socktype_arg): ditto.
+
+Fri Jan  2 00:49:44 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_getaddrinfo): use socktype_arg.
+
+Fri Jan  2 00:12:27 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (family_arg): extracted from
+	  setup_domain_and_type.
+	  (socktype_arg): ditto.
+	  (udp_init): use family_arg.
+	  (sock_s_gethostbyaddr): ditto.
+	  (sock_s_getaddrinfo): ditto.
+	  (sock_s_getnameinfo): ditto.
+
+Thu Jan  1 22:27:55 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/default.mspec: fix for builddir != srcdir
+
+	* Makefile.in: ditto.
+
+Thu Jan  1 21:26:05 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_getnameinfo): use family_to_int.
+
+Thu Jan  1 21:08:34 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: check all alias possibility.
+
+Thu Jan  1 20:59:55 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: more possible aliases.
+
+Thu Jan  1 20:54:02 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: AF_ISO and AF_OSI may be aliased.
+
+Thu Jan  1 20:50:09 2009  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* process.c (before_exec): it makes no sense for a conditional
+	  expression to return a void value.
+
+Thu Jan  1 20:47:09 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_getaddrinfo): use family_to_int.
+
+Thu Jan  1 20:17:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c: include constants.h at top.
+
+Thu Jan  1 19:53:33 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: add valp argument for family_to_int and
+	  socktype_to_int.
+
+	* ext/socket/socket.c (setup_domain_and_type): use valp argument.
+
+Thu Jan  1 19:36:57 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: generate family_to_str.
+
+	* ext/socket/socket.c (ipaddr): use family_to_str.
+
+Thu Jan  1 17:37:12 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: generate socktype_to_int.
+
+	* ext/socket/socket.c (setup_domain_and_type): use socktype_to_int.
+
+Thu Jan  1 17:26:47 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (setup_domain_and_type): initialize ptr.
+
+Thu Jan  1 17:01:50 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: refine family_to_int.
+
+Thu Jan  1 16:48:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: generate family_to_int().
+
+	* ext/socket/socket.c (setup_domain_and_type): use family_to_int.
+
+Thu Jan  1 15:08:46 2009  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/file2lastrev.rb: supports git repositories which are cloned
+	  from a git-svn gateway.
+	  Patch by Hongli Lai. [ruby-core:21020]
+
+Thu Jan  1 16:08:11 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: generate init_constants function.
+
+	* ext/socket/socket.c (Init_socket): call init_constants.
+
+Thu Jan  1 16:03:04 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (Socket::AF_INET6): moved from socket.c.
+	  (Socket::PF_INET6): ditto.
+
+Thu Jan  1 15:27:07 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: use erb.
+
+Thu Jan  1 15:07:56 2009  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: add -o option.
+
+	* ext/socket/depend: use mkconstants.rb with -o option.
+
+Thu Jan  1 07:42:36 2009  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/singleton.rb: fix indentation of RDoc text.  [ruby-core:21029]
+
+Thu Jan  1 07:16:44 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (distclean-ext, realclean-ext): use EXTS as default.
+
+	* win32/Makefile.sub (distclean-ext, realclean-ext): try to remove ext
+	  directory.
+
+Thu Jan  1 06:56:28 2009  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/file2lastrev.rb: unset PWD.
+
+Wed Dec 31 23:37:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_socketpair): yield if a block is given.
+	  (io_call_close): defined.
+	  (io_close): defined.
+	  (pair_yield): defined.
+
+Wed Dec 31 19:35:57 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/README: follows the change of directory structure in rubyspec.
+
+	* spec/default.mspec: ditto. Also follows change of mspec command.
+
+	* Makefile.in: ditto.
+
+Wed Dec 31 17:16:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read_internal): show openssl error
+	  code in EWOULDBLOCK error.
+
+Wed Dec 31 15:45:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_body): don't check to_io because
+	  Zlib::GzipWriter#to_io returns the underlying IO.
+
+Wed Dec 31 14:52:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_digest.c (GetDigestPtr): use StringValueCStr
+	  instead of STR2CSTR.
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_initialize): ditto.
+	  (ossl_ec_group_initialize): ditto.
+
+Wed Dec 31 14:12:35 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_to_float): replaced by to_flo definition from
+	  math.c  [ruby-dev:37668]
+
+	* math.c (Need_Float): use rb_to_float().
+
+Wed Dec 31 13:49:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#read_nonblock):
+	  implemented.
+
+	* ext/openssl/ossl_ssl.c (rb_sys_fail_path): removed.
+	  (fcntl.h): don't include.
+	  (ossl_ssl_read_internal): defined.
+	  (ossl_ssl_read): use ossl_ssl_read_internal.
+	  (ossl_ssl_read_nonblock): use ossl_ssl_read_internal.
+	  (Init_ossl_ssl): define sysread_nonblock, instead of read_nonblock.
+
+Wed Dec 31 00:27:54 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_to_float): prohibit conversion from nil to float.
+	  [ruby-dev:37663]
+
+	* pack.c (pack_pack): replace all rb_Float() to rb_to_float().
+	  [ruby-dev:37663]
+
+Tue Dec 30 21:17:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: use sock_define_uconst for INADDR_*
+	  constants because Socket::INADDR_LOOPBACK should be 0x7f000001
+	  instead of -0xffffff.
+
+	* ext/socket/socket.c (sock_define_uconst): defined.
+
+Tue Dec 30 18:23:10 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.3.1 r4506.
+	* test/minitest/*.rb: ditto.
+
+Tue Dec 30 17:59:59 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c: Minor fixes and tweaks in documentation.
+
+Tue Dec 30 17:03:51 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/dl/test/test_import.rb: fix a prototype decl.
+	  pointed out by Takehiro Kubo [ruby-core:20971].
+
+Tue Dec 30 16:56:09 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_blocking_region): add a comment.
+
+	* thread.c (rb_thread_call_without_gvl): added as a alias of
+	  rb_thread_blocking_region().
+
+	* thread.c (rb_thread_call_with_gvl): added.
+
+	* vm_core.h (rb_thread_t#blocking_region_buffer): added for
+	  rb_thread_call_with_gvl().
+
+Mon Dec 29 23:41:42 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/dl/test/test_base.rb: add libc search logic.
+	  this patch is written by Takehiro Kubo.
+	  [ruby-core:20963] [Bug #932]
+
+	* ext/dl/dl.h: Add ",..." as the last argument.
+	  this patch is written by Takehiro Kubo.
+	  Bug #633 [ruby-core:19289]
+
+	* ext/dl/lib/dl/stack.rb: add add_padding() to calculate
+	  alignment.  this patch is written by Takehiro Kubo.
+	  Bug #633 [ruby-core:19289]
+
+	* ext/dl/test/test_func.rb: atof()'s return value is double.
+	  this patch is written by Takehiro Kubo.
+	  Bug #633 [ruby-core:19289]
+
+	* ext/dl/test/test_import.rb:
+	  - atof()'s return value is double.
+	  - The types of qsort's second and third argument are size_t.
+	  - fprintf()'s return value is int.
+	  this patch is written by Takehiro Kubo.
+	  Bug #633 [ruby-core:19289]
+
+Mon Dec 29 22:37:17 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* process.c (rb_waitpid): retries waitpid when EINTR.
+	  [ruby-core:19744].
+
+Mon Dec 29 23:18:52 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* bignum.c (rb_cstr_to_inum): changed an error message.
+
+	* complex.c (string_to_c_strict): ditto.
+
+	* rational.c (string_to_r_strict): ditto.
+
+Mon Dec 29 22:37:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): template f should not accept non float
+	  values.  [ruby-dev:37656]
+
+	* object.c (rb_to_float): new function to type check floats.
+
+Mon Dec 29 22:27:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* random.c (rb_f_rand): type check simplified.  strings are no
+	  longer allowed for argument.  [ruby-dev:37655]
+
+	* test/ruby/test_rand.rb (TestRand::o.to_int): need override
+	  to_int.
+
+Mon Dec 29 21:22:31 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c: Infinity.numerator returns self. [experimental]
+
+	* numeric.c: Infinity.denominator returns 1. [experimental]
+
+Mon Dec 29 20:57:14 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* man/ri.1: new manpage.
+
+Mon Dec 29 20:45:25 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: floor, ceil, truncate and round accept an extra
+	  argument like Float#round. [experimental]
+
+Mon Dec 29 18:24:49 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* Makefile.in (distclean-local): removes Doxyfile and
+	  run.gdb at distclean'ing.
+
+Mon Dec 29 17:53:24 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* bin/erb: accepts options in more flexible styles.
+
+Mon Dec 29 17:25:17 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/erb.rb (ERB): m17n of ERB. adds rdoc.
+	  fixes #712. c.f. [ruby-dev:37516].
+
+	* lib/erb.rb (ERB::Compiler#compile): recognizes magic comments.
+	  returns a pair of compiled script and its script encoding.
+
+	* lib/erb.rb (ERB#set_eoutvar): make generated scripts return a
+	  string in correct encoding.
+
+	* lib/erb.rb (ERB#def_method): use Kernel#eval for encoding-awareness
+	  of the evaluated string.
+
+	* bin/erb.rb (ERB::Main.run): adds -E and -U options.
+	  String is no longer Enumerable.
+
+	* man/erb.1: new manpage.
+
+	* test/erb/test_erb_m17n.rb: new test case for m17n features.
+
+Mon Dec 29 18:02:45 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/io/wait/wait.c (io_nread): returns number of bytes available
+	  for read.  response to feature request #936 in [ruby-core:20917].
+
+Mon Dec 29 17:52:16 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/io/wait/wait.c (io_ready_p): updated to follow RDoc.
+
+Mon Dec 29 16:52:15 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (s_recvfrom_nonblock): revert r21162.
+
+Mon Dec 29 16:16:20 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_set_eval_stack, vm_set_main_stack, vm_set_top_stack):
+	  check stack overflow.  [ruby-dev:37646]
+
+Sun Dec 28 18:36:33 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/rubygems/test_gem_command.rb
+	  (test_add_option_overlapping_common_and_local_options):
+	  follows r21066.
+
+Mon Dec 29 14:50:19 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* Makefile.in (ext-clean): ext-clean always fails.
+
+	* ext/dl/ripper/extconf.rb: adds y.output into the clean list.
+
+Mon Dec 29 11:58:39 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_mutex_trylock): return false if Mutex owned
+	  by current thread.  [ruby-core:20943]
+
+	* thread.c (rb_mutex_lock): check dead lock (recursive lock) here.
+
+	* test/ruby/test_thread.rb: add a test.
+
+Mon Dec 29 10:58:54 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_get_path): move encoding conversion of file path
+	  from rb_scan_open_args.
+
+	* io.c (rb_scan_open_args): ditto.
+
+Mon Dec 29 07:15:16 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* cont.c: small RDoc fix mentioned from <radek.bulat at gmail.com>
+	  in [ruby-core:20921].
+
+Mon Dec 29 03:43:25 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.c (process_options): set th->base_block only while
+	  it is needed.  [ruby-dev:37634] [Bug #939]
+
+	* ruby.c (require_libraries): clear th->base_block before
+	  require libraries.
+
+Sun Dec 28 21:33:52 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ext/dl/test_base.rb: add BSD's case.
+
+Sun Dec 28 21:09:12 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ext/dl/lib/dl/import.rb (DL::Importer#sizeof): follows
+	  a feature change in 1.9.
+
+Sun Dec 28 21:03:07 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ext/dl/lib/dl/import.rb (DL::Importer#extern): adds filename and
+	  line number to module_eval'ing for readability of backtrace.
+	  (DL::Importer#bind): ditto.
+
+	* ext/dl/lib/dl/struct.rb (DL::CStructBuilder#create): ditto.
+
+Sun Dec 28 19:11:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): do not use snprintf/vsnprintf in
+	  msvcrt.
+
+	* sprintf.c: not force to override snprintf/vsnprintf.
+	  [ruby-core:20824]
+
+Sun Dec 28 17:21:36 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rubygems/test_ext_configure_builder.rb (test_self_build_fail):
+	  Extends pattern for FreeBSD's error message.
+
+Sun Dec 28 17:10:13 2008  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* ext/dl/test/test_dl2.rb: modify strncpy, strcpy, qsort, types.
+	  Patch by Takehiro Kubo.
+	  Bug #633 [ruby-core:19289]
+	* ext/dl/test/test_base.rb: /lib/libc.so is x86_64 binary in x86_64 architecture.
+
+Sun Dec 28 12:53:10 2008  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* ext/dl/test/test_base.rb: use libc.dylib when the platform is darwin.
+
+Sun Dec 28 12:24:14 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (mutex_free, mutex_unlock): add qualifiers.
+
+Sun Dec 28 10:28:04 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (mutex_free): GC thread (main thread) has failed to unlock
+	  a mutex that is locked by another thread, which makes the mutex
+	  dangling in keeping_mutexes and causes [BUG] or stuck finally.
+	  Now unlocking is performed as locking thread.
+
+	* thread.c (mutex_unlock, rb_mutex_unlock, rb_mutex_unlock_all):
+	  mutex_unlock receives a thread.
+
+Sun Dec 28 05:44:44 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.3.1 r4505.
+	* test/minitest/*.rb: ditto.
+
+Sun Dec 28 00:43:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* runruby.rb: refactored to modify ENV as once.
+
+Sat Dec 27 22:41:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_info_formatting.rb (setup, teardown): uses
+	  mktmpdir and removes it after the test.
+
+	* test/cgi/test_cgi_session.rb (setup, teardown): ditto.
+
+Sat Dec 27 21:46:10 2008  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* ext/dl/test/test_base.rb: use lib64 when the architecture is x86_64.
+
+Sat Dec 27 20:26:59 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_argv_size): if an argument is empty, it's size
+	  is 2, not 0, because it will be converted to "".
+
+Sat Dec 27 19:40:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (after_exec): needs to reset before restart timer
+	  thread.
+
+	* thread.c (thread_start_func_2): stops timer thread if forked in
+	  the new thread.   [ruby-core:19385]
+
+	* thread.c (rb_thread_atfork, rb_thread_atfork_before_exec): DRY.
+
+Sat Dec 27 18:25:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_eval.c (rb_f_loop): return enumerator if no block given.
+	  [ruby-list:45747]
+
+Sat Dec 27 17:08:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .gdbinit (dummy_gdb_enums.special_consts): forces to load debug
+	  info on Mac OS X.
+
+	* .gdbinit (rp): added T_ZOMBIE.
+
+Sat Dec 27 16:18:02 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_frame_callee, rb_frame_caller): rb_frame_callee()
+	  should return method id on current frame.
+	  add rb_frame_caller() to get method id on parent frame.
+	  Bug #884 [ruby-dev:37446]
+
+	* eval.c (rb_f_method_name): use rb_frame_caller()
+	  instead of rb_frame_callee().
+
+Sat Dec 27 15:28:12 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_econv_exception): show source and destination
+	  encoding.  [ruby-dev:37285]
+
+Sat Dec 27 15:23:38 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_yield_with_cfunc): rename parameter name
+	  "blockptr" to "blockargptr".
+
+Sat Dec 27 15:15:56 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (revision.h): ignores failure of file2lastrev.
+
+Sat Dec 27 15:02:38 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.c (process_options): fix to untouch th->mild_compile_error.
+	  [ruby-dev:37621], [ruby-dev:37620]
+
+Sat Dec 27 14:48:26 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.h, iseq.c (rb_iseq_new_main): add a type ISEQ_TYPE_MAIN.
+	  [ruby-dev:37619]
+
+	* compile.c (rb_dvar_defined, ruby_iseq_compile): ditto.
+
+	* iseq.c (iseq_data_to_ary, iseq_load): ditto.
+
+	* compile.c (iseq_compile_each): fix to check ip->compile_data.
+
+Sat Dec 27 14:29:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI.redirectable?): permit https redirection.
+	  patch from Roman Shterenzon.  [ruby-core:20485]
+
+Sat Dec 27 13:36:55 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (get_errinfo): return th->errinfo value
+	  if normal errinfo place (dynamic local variable) is not found.
+	  fixes Bug #732 [ruby-dev:37046].
+
+	* bootstraptest/test_proc.rb: add a test.
+
+Sat Dec 27 13:10:43 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* man/irb.1: adds -v, -h, -E and -U.
+
+Sat Dec 27 11:41:45 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method, vm_call_cfunc): use original id instead of
+	  calling id when NODE_CFUNC or NODE_BMETHOD.
+	  fixes Bug #632 [ruby-core:19282].
+
+	* vm_eval.c (vm_call0, vm_call_super): ditto.
+
+	* vm_method.c (rb_add_method, rb_alias): store original id
+	  in nd_file field of NODE_METHOD.
+
+	* test/stringio/test_stringio.rb: add a test.
+
+Sat Dec 27 09:48:54 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (Init_VM): create and define TOPLEVEL_BINDING at first.
+
+	* vm.c (vm_set_main_stack, rb_iseq_eval_main): added.
+
+	* parse.y (rb_parser_compile_file): fix to check parse_in_eval flag.
+
+	* eval.c (ruby_exec_node): use rb_iseq_eval_main()
+	  instead of rb_iseq_eval().
+
+	* iseq.c (rb_iseq_new_main), vm_core.h: added.
+	  main script (specified by -e or script name) should be run
+	  under TOPLEVEL_BINDING using Kernel#eval.  Above changes
+	  simulate Kernel#eval behaviour.  [ruby-dev:37240]
+
+	* compile.c (make_name_for_block): skip iseq except block type.
+	  this fix is needed for [ruby-dev:37240], and also fixes
+	  [ruby-dev:35392].
+
+Sat Dec 27 09:14:17 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* cont.c: rdoc for Fiber. patch by Muhammad Ali.
+	  [ruby-core:20894]
+
+Sat Dec 27 05:38:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (after_fork): ignores a termination request in the
+	  parent process.  [ruby-dev:37447]
+
+Sat Dec 27 01:52:39 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb: Using a more robust transcoding scheme to produce
+	  ASCII compatible inspect() messages.  [ruby-dev:37591]
+
+Fri Dec 26 18:04:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (fptr_finalize): don't allocate objects if noraise.
+	  (finish_writeconv): add noalloc argument to be able to avoid
+	  object allocation.
+	  (finish_writeconv_arg): introduced again.
+	  (finish_writeconv_sync): follow the above change.
+
+Fri Dec 26 17:04:14 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/input-method.rb (IRB::StdioInputMethod#initialize):
+	  removed a 'p' for debugging.
+
+Fri Dec 26 15:50:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::List#summarize): gives priority
+	  to latter switches.  [ruby-dev:36692]
+
+	* lib/optparse.rb (OptionParser#summarize): do not append
+	  unnecessary line terminator.
+
+Fri Dec 26 14:01:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (fptr_finalize): close the IO object even if finish_writeconv or
+	  flush is failed.
+	  (finish_writeconv): don't raise.  return errno or exception.
+	  (finish_writeconv_arg): removed.
+	  (finish_writeconv_sync): follow finish_writeconv change.
+
+	* transcode.c (rb_econv_make_exception): new function.
+
+	* include/ruby/encoding.h (rb_econv_make_exception): declared.
+
+Thu Dec 25 20:57:01 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (str_transcode0): set encoding when String#encode was
+	  given explicit but the same destination and source encoding.
+
+Thu Dec 25 20:01:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_s_pipe): unused variable removed.
+
+Thu Dec 25 19:35:51 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (sym_ignore): remove useless symbol.
+
+Thu Dec 25 18:36:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (pipe_close): removed.
+	  (pipe_yield): defined.
+	  (rb_io_s_pipe): use pipe_yield.
+
+Thu Dec 25 17:49:45 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (decorate_convpath): show type of escaping for
+	  xml_attr_quote or some conversions.
+
+Thu Dec 25 17:06:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_initialize): check fd validity.  [ruby-dev:36646]
+
+Thu Dec 25 16:23:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (flush_before_seek): check io_fflush result.
+	  (rb_io_check_readable): ditto.
+	  (rb_io_flush): ditto.
+	  (rb_io_fsync): ditto.
+	  (remain_size): ditto.
+	  (rb_io_write_nonblock): ditto.
+	  (finish_writeconv): ditto.
+	  (fptr_finalize): ditto.
+	  (io_reopen): ditto.
+	  (rb_io_reopen): ditto.
+	  (copy_stream_body): ditto.
+
+Thu Dec 25 15:54:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_fflush): flush write buffer without write lock in
+	  finalizers.  [ruby-dev:37572]
+
+	* io.c (rb_io_fptr_finalize): clear write lock before finalizing.
+
+Thu Dec 25 15:07:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (fptr_finalize): close the IO object even if close(2) is failed.
+
+Thu Dec 25 14:51:43 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/parser.rb (RDoc::Parser.binary?): should read in binary mode.
+
+Thu Dec 25 14:32:23 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_method_search): fix control flow bug.
+	  (committed at r20981)
+
+Thu Dec 25 13:28:20 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_flush): fsync() after buffer is flushed on win32.
+	  [ruby-core:20043]
+
+Thu Dec 25 13:13:00 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_method_search): return rb_cObject if there is no
+	  super class.  [ruby-dev:37587]
+
+	* bootstraptest/test_method.rb: add tests for above.
+
+Thu Dec 25 12:49:12 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (proc_new): should use proc_dup() if block has Proc.
+
+	* vm.c (vm_make_proc_from_block): should use rb_cProc for block.
+
+	* vm.c (vm_make_proc): add an assertion.
+
+	* bootstraptest/test_proc.rb: add a test.
+
+Thu Dec 25 12:44:27 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_yield_with_cfunc): check block has Proc.
+
+Thu Dec 25 05:53:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/id.h.tmpl, id.h (ruby_method_ids): not depend on if
+	  token are defined as macros.  [ruby-dev:37553]
+
+	* template/id.h.tmpl, id.h (ruby_method_ids_check): enclosed in a
+	  struct.
+
+Thu Dec 25 01:52:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): sets native thread key.
+
+Thu Dec 25 01:09:23 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* test/readline/test_readline_history.rb: check the encoding that
+	  is in the Readline::HISTORY.
+
+Thu Dec 25 00:59:06 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* test/readline/test_readline_history.rb: did not check the
+	  encoding that is in the Readline::HISTORY. I will fix it.
+
+Wed Dec 24 22:36:06 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c, vm_dump.c: change message by rb_bug().
+
+Wed Dec 24 21:57:39 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_peephole_optimize): fix typo.
+
+Sun Dec 21 13:38:04 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (invoke_block_from_c): fix to point right cfp.
+
+	* vm.c (vm_make_proc, vm_make_proc_from_block), vm_core.h:
+	  remove unused parameter cfp.
+
+	* vm_insnhelper.c, proc.c (proc_new): ditto.
+
+Wed Dec 24 20:59:12 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c (exc_equal): == method should not raise Exception.
+	  [ruby-dev:37519]
+
+	* sample/test.rb: fix test.
+
+Wed Dec 24 20:33:45 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method): use class of method defined
+	  instead of receiver's class on bmethod.  fixes [ruby-core:20786]
+
+	* bootstraptest/test_method.rb: add a test for above.
+
+	* vm_insnhelper.c (vm_setup_method): remove unused parameter klass.
+
+	* vm_insnhelper.h (CALL_METHOD): ditto.
+
+	* insns.def, vm_eval.c: ditto.
+
+Wed Dec 24 20:15:50 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/file2lastrev.rb (get_revisions): fix to ignore end of line.
+
+Wed Dec 24 13:33:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (HAVE_LONG_LONG, HAVE_OFF_T): revised for autoconf
+	  2.62 or later.
+
+Wed Dec 24 06:29:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (OBJDUMP, OBJCOPY): autoconf list is not comma
+	  separated.
+
+	* configure.in (AC_HEADER_STDC, AC_CHECK_HEADERS, AC_CHECK_FUNCS):
+	  removed duplicated checks.
+
+Wed Dec 24 03:21:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el: added comment to mention newer version
+	  bundled with Emacs 23 or later.  a patch from Phil Hagelberg
+	  <phil at hagelb.org> in [ruby-core:20838].
+
+Tue Dec 23 23:49:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_init_copy): call io_seek only if io_tell succeeds.
+
+Tue Dec 23 22:57:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork): don't propagate an error message if error
+	  buffer not given.
+
+Tue Dec 23 21:55:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork): propagate an error message from child to parent.
+	  (rb_f_exec): show details of error in child process on
+	  exception.
+	  (save_redirect_fd): add error message arguments.
+	  (run_exec_dup2): ditto.
+	  (run_exec_close): ditto.
+	  (run_exec_open): ditto.
+	  (run_exec_dup2_child): ditto.
+	  (run_exec_pgroup): ditto.
+	  (run_exec_rlimit): ditto.
+	  (rb_run_exec_options): ditto.
+	  (rb_exec): ditto.
+	  (rb_exec_atfork): ditto.
+	  (rb_spawn_internal): ditto.
+	  (rb_spawn): ditto.
+	  (rb_f_system): follow arguments change.
+	  (proc_daemon): ditto.
+	  (rb_f_spawn): show details of error in child process on exception.
+
+	* io.c (popen_exec): add error message arguments.
+	  (pipe_open): show details of error in child process on exception.
+
+	* include/ruby/intern.h (rb_run_exec_options): add error message
+	  arguments.
+	  (rb_exec): ditto.
+	  (rb_fork): ditto.
+	  (rb_spawn): ditto.
+
+	* ext/pty/pty.c (chfunc): add error message arguments.
+
+Tue Dec 23 20:28:28 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* io.c: rdoc for File::open and 1.9 feature in file modes.
+
+	* transcode.c: rdoc for String#encode
+
+Tue Dec 23 19:51:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn): deals with quoted commands.
+
+Tue Dec 23 18:44:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c (rb_strftime): use locale insensitive functions for tr_TR
+	  locale.
+
+Tue Dec 23 17:38:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb (assert_equal): show small differences
+	  for Regexp and Float.
+
+Tue Dec 23 16:49:48 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn): support normal commands with arguments.
+
+Tue Dec 23 16:22:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_inspect): show fd number if there is no pathname.
+
+Tue Dec 23 15:48:55 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/file2lastrev.rb: shouldn't use single quote in shell's command
+	  line if you want to support Windows.
+
+Tue Dec 23 15:46:43 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/mathn.rb: Math.sqrt(NaN) should be NaN. [ruby-dev:37537]
+
+Tue Dec 23 15:46:47 2008  Eric Hodel  <drbrain@segment7.net>
+	* gem_prelude.rb:  Match full RubyGems behavior when a gem can't be
+	  found.
+
+Tue Dec 23 15:36:58 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/file2lastrev.rb: detects vcs directory properly on building
+	  outside of srcdir. [ruby-dev:37555] [ruby-dev:37561]
+
+Tue Dec 23 15:30:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (pipe_open): need to initialize args.
+
+Tue Dec 23 14:18:14 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/st.h: need to include defines.h because LONG_LONG is
+	  defined there.
+
+Tue Dec 23 13:52:05 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parser/c.rb:  Don't coalesce adjacent comment blocks.
+	  [bug#901]
+
+Tue Dec 23 12:00:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (pipe_close): fix coding style.  [ruby-dev:37554]
+
+Tue Dec 23 08:10:05 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/installer.rb: Fix documentation.  [bug#827]
+
+Tue Dec 23 08:04:58 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* gem_prelude.rb: Don't remove methods twice.  [bug#555]
+
+Tue Dec 23 00:16:48 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* sprintf.c (rb_str_format): fix buffer overflow.
+
+Mon Dec 22 19:31:19 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (revision.h): uses tool/file2lastrev.rb to support
+	  git-svn.
+
+	* version.h: changed version string as
+	  `ruby 1.9.1 (2008-12-22 patchlevel-5000 trunk 20912)
+	  [i386-darwin9.6.0]'.
+
+Mon Dec 22 16:32:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): encoding of packed string only from 'm',
+	  'M', and 'u' should be US-ASCII.  [ruby-dev:37284]
+
+	* pack.c (pack_pack): encoding of packed string only from 'U'
+	  should be UTF-8.  also upgrade US-ASCII strings to UTF-8.
+
+Mon Dec 22 15:56:49 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* Doxyfile.in: new file. Template of a configuration file.
+	  Introduces C-level API reference generation with Doxygen.
+
+	* tool/file2lastrev.rb: wrapper script that abstracts subversion
+	  and git-svn.
+
+	* tool/strip-rdoc.rb: filter for preventing doxygen from processing
+	  rdoc comments.
+
+	* configure.in: (Doxyfile): Doxyfile is generated together with
+	  Makefile.
+	  (PACKAGE): configuration $(PACKAGE) is necessary for $(docdir).
+
+	* instruby.rb: adds a new install target 'capi'
+
+	* common.mk (capi): new target that generates C API documents with
+	  Doxygen.
+	  (install-capi): new target.
+	  (pre-install-capi): ditto.
+	  (do-install-capi): ditto.
+	  (post-install-capi): ditto.
+
+Mon Dec 22 21:31:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_ord): Integer#ord implemented.
+
+Mon Dec 22 21:26:12 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_s_pipe): IO.pipe can take a block.
+	  (pipe_close): new function.
+
+Mon Dec 22 21:08:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb (assert_nothing_raised): increment
+	  assertion count.  [ruby-dev:37480]
+
+Mon Dec 22 19:10:44 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (Init_native_thread): need to call
+	  ruby_thread_set_native() here.
+
+Mon Dec 22 18:00:36 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: fix wrong flag value.
+
+Mon Dec 22 15:15:58 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/protocol.rb (Net::BufferedIO#rbuf_fill): avoid calling
+	  read_nonblock in rescue.  use retry instead.
+
+Mon Dec 22 14:35:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_external_str_new_with_enc): set ASCII-8BIT if
+	  encoding is US-ASCII and string contains 8bit characters.
+
+Mon Dec 22 12:26:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (rubydll.def), win32/mkexports.rb
+	  (Exports#exports): added VERSION.
+
+Mon Dec 22 12:25:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/vsnprintf.c (BSD_vfprintf): fix for test_sprintf_comb.rb, by
+	  wanabe in [ruby-dev:36935].
+
+Mon Dec 22 12:05:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (mingw): no longer uses snprintf and vsnprintf of
+	  msvcrt.  [Bug#6]
+
+	* win32/win32.c (rb_w32_vsnprintf, rb_w32_snprintf): removed.
+
+	* win32/Makefile.sub (config.h): vsnprintf exists in VC7 or later.
+
+	* win32/mkexports.rb (Exports#initialize): aliases
+	  rb_w32_vsnprintf and rb_w32_snprintf for binary compatibility.
+
+	* sprintf.c (rb_str_format): uses snprintf instead of sprintf.
+
+	* numeric.c (flo_to_s, rb_num2long, rb_num2ll): ditto.
+
+Mon Dec 22 10:59:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_upto): should return enumerator if called
+	  without block.  [ruby-dev:37526]
+
+Mon Dec 22 09:54:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (gettable_gen): the encoding of __FILE__ should be
+	  rb_filesystem_encoding().  [ruby-list:45733]
+
+	* parse.y (gettable_gen): __FILE__ should be ASCII-8BIT when
+	  filesystem encoding is US-ASCII and __FILE__ contains non 7bit
+	  characters.
+
+Mon Dec 22 02:54:56 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: temporary patch for trouble on TkAqua (> Tk8.4.9).
+
+	* ext/tk/sample/demos-jp/widget: set encoding to a demo-code string to
+	  avoid garbled text on the Code Window.
+
+Sun Dec 21 14:55:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (SPLAT_PROC): splat values by hand.
+
+Sun Dec 21 13:50:37 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* test/readline/test_readline.rb
+	  (TestReadline#test_some_characters_methods): checked encoding.
+
+Sun Dec 21 12:23:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (SPLAT_PROC): fix for regexp.  [ruby-dev:37514]
+
+Sun Dec 21 10:43:10 2008  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* sample/test.rb: rescue exception.
+
+Sat Dec 20 18:28:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_find_1): supplements an extension for executable
+	  files on DOSish platforms.
+
+	* io.c (pipe_open): use rb_w32_aspawn() for array form.
+
+	* win32/win32.c (rb_w32_pipe_exec): no longer used.
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn): deals with batch
+	  files and commands with extensions.  [ruby-core:20695]
+
+	* win32/win32.c (has_redirection): supports environment variables
+	  references.
+
+Sat Dec 20 15:34:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rake.rb (Rake::FileList#egrep): change open mode to "rb",
+	  i.e. default to binary.    [ruby-dev:37385]
+
+	* lib/rake.rb (Rake::FileList#egrep): allow specifying reading
+	  encoding, e.g. FileList['*.rb'].egrep(/require/, encoding:
+	  "ascii-8bit")
+
+	* lib/rake.rb (Rake::MultiTask): invoke_prerequisites should be
+	  private.  a patch from okkez in [ruby-dev:37399]
+
+Sat Dec 20 10:59:16 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/locale.rb (IRB::Locale#initialize)
+	  initializes instance variables to suppress warnings.
+	  typo fix to ignore LoadError: [ruby-dev:37508]
+
+Fri Dec 19 17:01:35 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_extract_encoding_option): "internal_encoding: nil"
+	  to specify no-transcoding.  and other corner case fixed.
+	  [ruby-dev:37496]
+
+	* hash.c (rb_hash_lookup2): new function to look-up hash with
+	  default value.
+
+Fri Dec 19 15:51:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): get rid of warning on DOSISH.
+
+Fri Dec 19 07:45:37 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (exc_equal): duck typing equal to make it transitive.
+	  [ruby-dev:34808]
+
+Thu Dec 18 19:31:54 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/irb/init.rb (IRB.opt_parse): (M17N) adds -U and -E as command
+	  line options. [ruby-dev:37161]. Fixes #711.
+	  improved long options.
+
+	* lib/irb/init.rb (IRB.set_encoding): new subroutine for IRB.opt_parse
+
+	* lib/irb/input-method.rb (IRB::StdioInputMethod): (M17N) regards
+	  scripts that user types as encoded in the external_encoding.
+
+	* lib/irb/input-method.rb (IRB::ReadlineInputMethod): ditto.
+
+	* lib/irb/input-method.rb (IRB::FileInputMethod): (M17N) respects
+	  magic comment.
+
+	* lib/irb/help.rb (IRB.print_usage): (M17N) respects magic comment
+	  in the resource file.
+
+	* lib/irb/lc/help-message: adds -U and -E.
+
+	* lib/irb/lc/ja/help-message: ditto. re-encoded from ISO-2022-JP into UTF-8.
+
+	* lib/irb/lc/ja/encoding_aliases.rb: new file. provides Japanese specific
+	  character encoding name table for backward compatibility.
+
+	* lib/irb/lc/ja/error.rb: re-encoded from ISO-2022-JP into UTF-8.
+	  magic comment.
+
+	* lib/irb/locale.rb: integrated with Ruby 1.9's M17N support.
+
+	* lib/irb/magic-file.rb: new file. utility to handle files with magic
+	  comment and opens in the correct encoding.
+
+	* lib/irb/ruby-lex.rb (RubyLex#each_top_level_statement): recovers
+	  character encoding for a statement after it lexed so that irb can
+	  eval the statement in correct encoding.
+
+	* lib/irb/src_encoding.rb: new file. utility.
+
+Thu Dec 18 21:12:28 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/session.rb: fix bug for ignore session_id option.
+	  report from [ruby-core:18635], [Bug #572]
+
+	* lib/cgi/core.rb: use Encoding#find when encoding set.
+
+	* test/cgi/test_cgi_session.rb: test for session_id specified.
+
+Thu Dec 18 17:00:56 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_aset): string key copying only happen if key is
+	  an exact instance of String, not a subclass. based on a patch
+	  from Mike Gold <mike.gold.4433 at gmail.com> in [ruby-talk:322667].
+	  [incompatible] [ruby-talk:322417]
+
+Thu Dec 18 16:48:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/optparse/version.rb: remove variable shadowing to stop
+	  warning.   [ruby-core:20612]
+
+	* lib/irb/completion.rb, lib/net/imap.rb, lib/prime.rb,
+	  lib/rinda/ring.rb, lib/racc/parser.rb,
+	  lib/shell/command-processor.rb, lib/yaml/yamlnode.rb: ditto.
+
+	* lib/racc/parser.rb: remove space before parentheses.
+
+	* lib/shell/command-processor.rb, lib/shell/process-controller.rb:
+	  use parentheses around arguments.
+
+	* lib/irb/ext/change-ws.rb, lib/rexml/validation/relaxng.rb,
+	  lib/yaml/baseemitter.rb: indentation fix.
+
+	* lib/matrix.rb: small cosmetic change.
+
+Thu Dec 18 08:15:04 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/xmlrpc/server.rb:  Restricting method inspection to show only
+	  non-inherited public methods.  [ruby-core:20603]
+
+	* lib/xmlrpc/server.rb:  Fixing method inspection so it doesn't
+	  trigger XMLRPC::FaultException when used.  [ruby-core:20604]
+
+Wed Dec 17 19:39:44 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (unix_recv_io): relax msg_controllen error
+	  check for x86_64-openbsd4.4.
+
+Wed Dec 17 19:37:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/extconf.rb: check util.h for OpenBSD.
+
+	* ext/pty/pty.c: include util.h if available.  fix variable name.
+
+Wed Dec 17 19:23:28 2008  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/matrix.rb: shut up warning. [ruby-dev:37481] [Bug #899]
+
+Wed Dec 17 15:44:28 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_pop_m): use enum ary_take_pos_flags.
+
+	* array.c (rb_ary_push_m): add modification check before actual
+	  operation.   [ruby-dev:37440]
+
+Wed Dec 17 15:35:39 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* configure.in (OBJDUMP, OBJCOPY):
+	  *BSD's ports and MacPorts prefix GNU binutils with 'g'.
+
+Wed Dec 17 11:39:39 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_hash):
+	  shut up warning.  see [ruby-dev:37437]
+
+Wed Dec 17 11:01:35 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_hash): gets rid of collisions between different
+	  encoding strings.  [ruby-core:20526]
+
+Wed Dec 17 09:50:19 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_metaclass.rb: removed codes for my debugging.
+	  sorry. [ruby-dev:37470]
+
+Tue Dec 16 22:55:11 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bin/testrb: set $0.
+
+Tue Dec 16 22:42:16 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit.setup_argv): sorry, fixed wrong commit.
+
+Tue Dec 16 22:15:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/minitest/unit.rb (MiniTest::Assertions#assert_instance_of):
+	  should assert by instance_of?, not ===.  [ruby-dev:37458]
+
+	* lib/minitest/unit.rb (MiniTest::Assertions#assert_instance_of):
+	  typo fixed.
+
+Tue Dec 16 21:59:29 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit.setup_argv): ALT_SEPARATOR support.
+
+Tue Dec 16 21:59:02 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpToString): reverted modification
+	  (that caused a bug) in r20359.  [ruby-dev:37370]
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_limit): comment update.
+	  [ruby-dev:37465]
+
+Tue Dec 16 20:34:44 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (getDevice): add nomesg argument.
+	  (get_device_once): add nomesg argument.  chmod slave tty to 0600
+	  if nomesg.  more error tests.
+	  (no_mesg): new function.
+	  (pty_open): make slave tty's mode 0600.
+
+Tue Dec 16 20:24:20 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (_parse): m17n compliant.
+
+Tue Dec 16 19:48:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/extconf.rb: check posix_openpt.
+
+	* ext/pty/pty.c (get_device_once): use posix_openpt if available.
+
+Tue Dec 16 19:43:53 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c: use strlcpy for error messages.
+
+Tue Dec 16 16:41:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt): returns dispatched result.
+
+	* parse.y (primary): also in ripper, saves in_def before restoring.
+
+Tue Dec 16 12:30:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_header_ucontext_h): do not use ucontext.h on Mac
+	  OS X 10.5 or later.  merged from MacPorts.
+
+	* configure.in (ac_cv_func_daemon): daemon(3) is deprecated on Mac OS
+	  X 10.5 or later.
+
+Tue Dec 16 11:37:07 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (set_arg0): use strlcpy() instead of strncpy().
+
+	* load.c (rb_feature_p): ditto.
+
+	* dln.c (dln_load): ditto.
+
+Tue Dec 16 09:14:28 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (block_call): block should not be given to yield.
+	  [ruby-core:20583]
+
+Mon Dec 15 23:48:39 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (sleep_timeval): cast tv_usec to long to shut up
+	  warnings on OSX.  [ruby-dev:37449]
+
+Mon Dec 15 23:34:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (pty_open): set FMODE_SYNC and FMODE_DUPLEX.
+
+Mon Dec 15 23:29:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (process_options): revive global sub, gsub, chop, chomp
+	  only when auto looping options (-p/-n) is specified.
+	  [ruby-core:20570]
+
+Mon Dec 15 22:48:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (pty_open): new method PTY.open.
+
+Mon Dec 15 21:24:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (get_device_once): use DEVICELEN instead of
+	  sizeof SlaveName.
+
+Mon Dec 15 21:01:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (chfunc): make it static.
+
+Mon Dec 15 20:59:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (child_info): add slavename.
+	  (chfunc): use slavename.
+	  (establishShell): set slavename.
+
+Mon Dec 15 20:16:46 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (EXTMK_ARGS): shouldn't use ``\"'' because cmd.exe eat
+	  ''\'' in such quotes.
+
+Mon Dec 15 19:07:31 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): define CANONICALIZATION_FOR_MATHN
+	  to follow r20683.
+
+Mon Dec 15 18:12:34 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rake/packagetask.rb (Rake::PackageTask): small document
+	  update from okkez in [ruby-dev:37443]
+
+Mon Dec 15 16:28:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_arg_complex): uses cfp from
+	  blockptr instead of the current cfp.  [ruby-core:20544]
+
+Mon Dec 15 14:56:59 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_metaclass.rb: new test case for metaclass hierarchy.
+
+	* class.c (make_metametaclass): new function. extracted from
+	  rb_make_metaclass.
+
+	* class.c (rb_make_metaclass): uses make_metametaclass when called for a
+	  metaclass.
+
+	* class.c (rb_singleton_class): creates a meta^(n+2)-class in
+	  addition to a meta^(n+1)-class when called for a meta^(n)-class.
+	  This is because the returned meta^(n+1) class must acts as an instance of
+	  Class, metaclass of Class, ..., meta^(n+1)-class of Class,
+	  Module, metaclass of Module, ..., meta^(n+1)-class of Module,
+	  Object, metaclass of Object, ..., meta^(n+2)-class of Object,
+	  BasicObject, metaclass of BasicObject, ..., meta^(n+2)-class of
+	  and BasicObject even when Class, Module, Object or BasicObject has
+	  not have its meta^(i)-class yet.
+
+Mon Dec 15 15:13:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* id.h, template/id.h.tmpl (ruby_method_ids): sees YYTOKENTYPE too.
+	  [ruby-dev:37436]
+
+Mon Dec 15 15:02:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (EXTMK_ARGS): needs MINIRUBY for cross-compile.
+	  [ruby-core:20131]
+
+Mon Dec 15 14:39:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call0): NODE_ZSUPER supported.  [ruby-core:20480]
+
+	* vm_eval.c (vm_call_super): uses method_missing().
+
+	* vm_eval.c (method_missing): get rid of too large alloca.
+
+	* vm_eval.c (rb_call0, method_missing): uses idMethodMissing.
+
+	* vm_method.c (rb_add_method, remove_method, rb_undef): uses
+	  id__send__.
+
+	* vm_method.c (Init_eval_method): removed IDs which are defined as
+	  immediate values.
+
+Mon Dec 15 11:35:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_backtrace): defaults to script name for C functions.
+
+Mon Dec 15 11:32:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (rubylib_mangled_path, rubylib_mangled_path2): cannot use
+	  locale encoding before load path is initialized
+
+	* ruby.c (ruby_init_loadpath_safe): ditto.
+
+	* ruby.c (process_options): loads encdb so that encodings can be
+	  loaded, then associates script name and load paths with the
+	  locale encoding.
+
+Mon Dec 15 10:00:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (btest-ruby): use RUNRUBY instead of MINIRUBY to load the
+	  shared library.  [Bug #849]
+
+Sun Dec 14 22:31:19 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (bigmul1_karatsuba): remove temporal bignum.
+
+	* bignum.c (bigsqr): call bigmul0(x, x) because it is faster than the
+	  original bigsqr at this point.
+
+	* bignum.c (rb_big_pow): a value returned from bigsqr is already
+	  truncated.
+
+Sun Dec 14 21:13:02 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (bigmul1_karatsuba): fix comment and refactoring.
+
+Sun Dec 14 14:53:41 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (bigmul1_balance, bigmul1_karatsuba): remove all
+	  rb_gc_force_recycle which causes memory leak.
+
+Sun Dec 14 14:26:11 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/README: directory structure changed
+
+Sun Dec 14 09:14:37 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* reverts r20713.
+
+Sun Dec 14 12:51:48 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (rb_big_mul): faster multiplication by Karatsuba method and
+	  twice faster square than normal multiplication. [ruby-dev:37392]
+
+	* random.c (rb_rand_internal): used by Bignum#*.
+
+	* test/ruby/test_bignum.rb: add some tests for above.
+
+Sun Dec 14 01:35:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (forked_child): new variable.
+	  (before_exec): don't call rb_thread_stop_timer_thread if
+	  forked_child.
+	  (after_exec): reset forked_child after rb_thread_start_timer_thread.
+	  (rb_fork): set forked_child just after fork in child.
+
+	* ext/pty/pty.c (chfunc): extracted from establishShell.
+	  (establishShell): use rb_fork.
+
+	  [ruby-dev:37418]
+
+Sat Dec 13 22:17:30 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (help): describes more targets.
+	  based on a patch by Michael Klishin. [ruby-core:20523].
+
+Sat Dec 13 18:34:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_printable): wrong condition for string iteration.
+
+Sat Dec 13 16:04:48 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* encoding.c (default_external): endless recursion during
+	  loading a locale encoding on some locale.
+	  fixed by Nobuyoshi Nakada.
+
+Sat Dec 13 15:55:37 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ruby.c (process_options): fixed default_internal is nil.
+	  (closes #862)
+
+Sat Dec 13 15:52:27 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_equal): remove documentation error "Otherwise,
+	  compares them as strings".  [ruby-dev:37398]
+
+	* string.c (sym_inspect): quote if symbol contains non-printable
+	  characters.  [ruby-dev:37398]
+
+Sat Dec 13 14:24:38 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/enc/test_utf16.rb: feature changed in r20626.
+	  follows it.
+
+Sat Dec 13 13:41:44 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_m17n_comb.rb: feature changed in r20626.
+	  follows it.
+
+Sat Dec 13 13:03:20 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: use rb_usascii_str_new2 instead of rb_str_new2.
+
+	* rational.c: ditto.
+
+Sat Dec 13 12:48:57 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_m17n.rb: feature changed in r20626.
+	  follows it.
+
+Sat Dec 13 12:38:16 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/minitest/test.rb (MinTest::Unit#location):
+	  skips user-defined assertions as MiniTest's.
+	  e.g. for test/ruby/test_m17n.rb.
+
+Sat Dec 13 09:17:33 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*.rb: Imported minitest 1.3.1 r4503.
+	* test/minitest/test_mini_test.rb: ditto.
+	* lib/miniunit/autorun.rb: added (as part of r4503).
+
+Sat Dec 13 08:35:14 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_canonicalization): renamed.
+
+	* ext/math/complex.c: followed the above change.
+
+	* rational.c (nurat_canonicalization): renamed.
+
+	* ext/math/rational.c: followed the above change.
+
+	* configure.in: defines a new macro.
+
+Sat Dec 13 01:10:03 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c
+	  (readline_s_get_completion_append_character): uses locale
+	  encoding but not ASCII-8BIT.
+
+Fri Dec 12 22:12:46 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c, rational.c: do not use RUBY_VERSION_CODE.
+
+Fri Dec 12 21:41:36 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c: r20662 reverted.
+
+Fri Dec 12 19:29:07 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* encoding.c (rb_enc_set_default_external): default_internal can be
+	  nil, but default_external cannot.
+
+	* encoding.c (rb_set_default_internal): adds rdoc.
+
+	* encoding.c (enc_find): ditto.
+
+Fri Dec 12 19:24:47 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (run, runruby, parse, gdb, gdb-ruby): you can also run
+	  other scripts than $(srcdir)/test.rb.
+
+Fri Dec 12 19:13:43 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c (readline_readline): changed the message
+	  of IOError to 'closed stdin' from 'stdin closed' if stdin was
+	  closed.
+
+Fri Dec 12 19:00:49 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c: used the ExportStringValue macro
+	  instead of the OutputStringValue macro. removed the
+	  OutputStringValue macro.
+
+Fri Dec 12 16:23:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (yycompile0): ruby_eval_tree_begin is always 0 when
+	  ruby_eval_tree is 0.
+
+Fri Dec 12 14:09:55 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* encoding.c (enc_get_default_encoding): removed.
+	  Generalizing rb_default_{external,internal}_encoding seems to be
+	  difficult.
+	  default_external cannot be NULL even before detected. [ruby-dev:37390]
+
+	* encoding.c (rb_default_external_encoding): has its own
+	  implementation again.
+
+	* encoding.c (rb_default_internal_encoding): ditto.
+
+	* gem_prelude.rb: added notice.
+
+	* ruby.c (rubylib_mangled_path, rubylib_mangled_path2): uses locale
+	  encoding but not ASCII-8BIT.
+
+	* ruby.c (process_options): refers less to default_external.
+
+Fri Dec 12 11:00:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_feature_p): load path must be expanded.
+	  [ruby-dev:37388]
+
+Fri Dec 12 10:28:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_version): greps only the line that defines version.
+
+	* version.h (RUBY_VERSION_CODE, RUBY_RELEASE_CODE): removed.
+
+Fri Dec 12 10:20:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c, rational.c: do not use RUBY_VERSION_CODE.
+
+Fri Dec 12 10:19:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_string_value_ptr, rb_to_id): do not use a side
+	  effect expression in RSTRING_PTR.
+
+	* string.c (rb_str_split_m): simplified the argument of
+	  RSTRING_LEN.
+
+Thu Dec 11 23:48:00 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (sqrt): should pass nan to the original method.
+
+Thu Dec 11 21:04:04 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker.rb, lib/rss/maker/0.9.rb, lib/rss/maker/base.rb:
+	  RSS::Maker.[] returns a maker class corresponds to passed version.
+
+	* test/rss/test_maker_*.rb: add tests for RSS::Maker.[].
+
+Thu Dec 11 20:05:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_aliases_enc_i): skips default internal.
+
+Thu Dec 11 19:31:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/assertions.rb: extracted from lib/test/unit.rb.
+	  lib/test/unit/testcase.rb: ditto.
+	  at asakusa.rb.  [ruby-core:20014]
+
+Thu Dec 11 19:23:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb (test_dup_many): extracted from test_dup.
+	  test in ruby subprocess.  at asakusa.rb.  [ruby-dev:35648]
+
+Thu Dec 11 19:16:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pathname.rb (Pathname#=~): undefed.  at asakusa.rb.
+	  [ruby-core:19690]
+
+Thu Dec 11 13:09:01 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (reg_enc_error): raise EncodingCompatibilityError for
+	  encoding incompatibility.  [ruby-core:18600]
+
+	* re.c (rb_reg_prepare_enc): more consistent error message.
+	  [ruby-core:18611]
+
+Thu Dec 11 13:36:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_set_default_encoding): allowed to set default
+	  encoding.
+
+	* encoding.c (rb_locale_encoding): removed Encoding::LOCALE.
+
+	* encoding.c (rb_enc_set_default_external): ditto.
+
+	* encoding.c (rb_enc_set_default_internal): ditto.
+
+	* encoding.c (rb_enc_set_default_internal): defines internal
+	  encoding as nil.
+
+Thu Dec 11 11:11:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/st.h (size_t): needs stddef.h or stdlib.h.
+	  [ruby-core:20339]
+
+Thu Dec 11 10:25:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_bignum.rb (TestBignum#test_convert): remove
+	  platform dependent call of Process.wait.  coverage should be
+	  accomplished by other method.  [ruby-dev:36062]
+
+Thu Dec 11 10:18:35 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/document.rb (REXML::Document#write): require
+	  rexml/formatters/transitive if transitive flag is on.  a patch
+	  from akira yamada in [ruby-dev:36230].  fix #553
+
+	* lib/rexml/element.rb (REXML::Element#write): ditto.
+
+	* lib/rexml/formatters/transitive.rb (REXML::Formatters::Transitive#write_element):
+	  add hack for IE.  [ruby-dev:36230]
+
+Thu Dec 11 02:37:22 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* math.c (domain_check): should not raise EDOM exception for NaN
+	  value if operand is also NaN.  affects acos, asin, acosh, atanh,
+	  log, log2, log10, sqrt, gamma, lgamma. [ruby-dev:37357]
+
+Thu Dec 11 02:23:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.pipeline_start): return an array of threads if a
+	  block is not given.
+
+Thu Dec 11 01:48:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_round): should be round
+	  to integer.  [ruby-dev:37355]
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_divmod): division should
+	  be integer.  [incompatible] [ruby-dev:37355]
+
+Thu Dec 11 01:21:58 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (rb_mutex_unlock): shut up warning.  a patch from
+	  Kazuhiro NISHIYAMA in [ruby-dev:37345]. fix #846
+
+Wed Dec 10 23:58:56 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: avoided warnings on cpp. [ruby-dev:37344]
+
+	* rational.c: ditto.
+
+Wed Dec 10 21:32:02 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker.rb (RSS::Maker.[]): add.
+	  Suggested by Akinori MUSHA. Thanks!!! [ruby-dev:37210]
+
+Wed Dec 10 12:56:32 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_rewind): If the enclosed object
+	  responds to a "rewind" method, call it; cf. [ruby-dev:37268]
+
+Wed Dec 10 12:46:52 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_next): Fix a typo: s/rewinded/rewound/.
+
+	* lib/prime.rb (Prime::OldCompatibility#each): Ditto.
+
+Wed Dec 10 11:25:53 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_initialize): raise RegexpError when encoding
+	  is dummy encoding. [ruby-dev:37091]
+
+Wed Dec 10 01:28:16 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (win32.obj): depend on headers.
+
+Tue Dec  9 18:35:35 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/single_byte.trans, cp850-tbl.rb, cp852-tbl.rb,
+	  cp855-tbl.rb, koi8-r-tbl.rb, koi8-u-tbl.rb, tis-620-tbl.rb:
+	  new single-byte transcodings (from Yoshihiro Kambayashi)
+
+	* test/ruby/test_transcode.rb: added tests for the above
+	  (from Yoshihiro Kambayashi), small cosmetic fixes
+
+Tue Dec  9 13:33:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each), gc.c (assign_heap_slot),
+	  (gc_mark_children), parse.y (vtable_alloc, vtable_free, vtable_add),
+	  proc.c (proc_to_s), thread.c (terminate_i, rb_thread_terminate_all),
+	  (thread_start_func_2, blocking_region_begin, blocking_region_end),
+	  (rb_thread_kill), thread_pthread.c (native_thread_create),
+	  (ubf_pthread_cond_signal), vm.c (check_env, thread_free), vm_dump.c
+	  (vm_env_dump_raw, vm_stack_dump_each, vm_thread_dump_state),
+	  (vm_call0): use void pointer for %p.
+
+	* cont.c (fiber_status), template/insns.inc.tmpl (ruby_vminsn_type),
+	  vm_insnhelper.h (BOP): ISO C forbids comma at end of enumerator
+	  list.
+
+	* vm_insnhelper.c (check_cfunc): use function pointer.
+
+Tue Dec  9 13:23:55 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/curses.c (curses_use_default_colors): add new
+	  method.  a patch from Giancarlo F Bellido
+	  <support at coaxialhost.com> in [ruby-core:20434].
+
+	* ext/curses/curses.c (curses_tabsize_set): ditto.
+
+	* ext/curses/curses.c (curses_tabsize_get): ditto.
+
+Tue Dec  9 12:08:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .gdbinit (rp): fixed typo for T_STRUCT.
+
+Tue Dec  9 11:50:46 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* pack.c (pack_pack): fixed [ruby-dev:37295].
+	  Array#pack always has returned an US-ASCII string when
+	  the given format ended with 'u', 'M' or 'm'.
+
+Tue Dec  9 03:21:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: fix for continued lines.  based on a patch from
+	  Marcus Rueckert <darix AT opensu.se> at [ruby-core:20420].
+
+Tue Dec  9 00:54:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): enc.mk in snapshot is dummy and should
+	  not deal with objects.  [ruby-core:20422]
+
+Tue Dec  9 00:17:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (clean-srcs): split out from clean.
+
+Mon Dec  8 23:34:18 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* pack.c (pack_pack): fixed odd act of 'm*', 'M*', and 'P*'.
+	  just ignores '*' in these cases.
+	  [ruby-dev:37289]
+
+Mon Dec  8 18:31:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack): fixed length for odd length string.
+	  [ruby-dev:37283]
+
+Mon Dec  8 11:28:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): added version, copyright, usage, yydebug,
+	  syntax to --dump option.
+
+Mon Dec  8 11:15:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): source-encoding option.
+
+Mon Dec  8 10:32:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (set_option_encoding_once): dry.
+
+	* ruby.c (proc_options): checks extra argument for -E/--encoding.
+
+Sun Dec  7 23:47:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBPATHFLAG): use numbered specifier if RPATHFLAG
+	  is set.  [ruby-talk:322136]
+
+Sun Dec  7 23:17:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (LIBRUBY_SO): localizes non-public symbols.
+
+	* configure.in (OBJDUMP, OBJCOPY): for dealing with binary files.
+
+Sun Dec  7 17:44:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.capture3): renamed from Open3.poutput3.
+	  (Open3.capture2): renamed from Open3.poutput2.
+	  (Open3.capture2e): renamed from Open3.poutput2e.
+
+Sun Dec  7 11:48:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.poutput3): :binmode option implemented.
+	  (Open3.poutput2): ditto.
+	  (Open3.poutput2e): ditto.
+
+Sat Dec  6 18:33:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): added RM and CP.  [ruby-dev:37288]
+
+Sat Dec  6 17:51:35 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* pack.c (pack_pack): 'u0' is not special differently from 'm0'.
+
+Sat Dec  6 14:35:06 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_pack): set US-ASCII and CODERANGE_7BIT
+	  when quoted-printable ('M') and base64 ('m').
+	  [ruby-dev:37284]
+
+	* pack.c (pack_unpack): ditto.
+
+Fri Dec  5 22:56:24 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_make_exception): single argument to raise may be
+	  string compatible object.
+
+Fri Dec  5 21:45:45 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_{to_s,inspect}): performance improvement.
+
+Fri Dec  5 21:42:44 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: inspect should not depend on to_s.
+
+Fri Dec  5 21:18:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_truncate): return
+	  integer if no optional argument given. [incompatible]
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_floor): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_ceil): ditto.
+
+Fri Dec  5 19:06:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.pipeline_start): new method.
+	  (Open3.pipeline): ditto.
+
+Fri Dec  5 18:55:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (run_exec_dup2): !save is false if Qnil.
+
+Fri Dec  5 18:07:32 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read, rb_w32_write, rb_w32_isatty): check
+	  whether fd is valid.
+
+Fri Dec  5 13:05:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_parameters): proc arguments are always optional.
+
+	* proc.c (get_proc_iseq, rb_proc_parameters): ditto.
+
+Fri Dec  5 12:38:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_sequence): uses rb_compile_warning() for
+	  warning at compilation time.
+
+Fri Dec  5 12:35:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (ruby_iseq_compile, ruby_iseq_translate_threaded_code),
+	  (ruby_insns_name_array, ruby_iseq_build_from_ary): prefixed with
+	  ruby_.
+
+	* iseq.c (ruby_iseq_load, ruby_insn_make_insn_table): ditto.
+
+Fri Dec  5 10:01:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_cmp_m): fixed rdoc.  pointed out by <Thomas
+	  C. Mitchell AT gmail.com> at [ruby-talk:321967]
+
+Fri Dec  5 07:58:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_binwrite): arg.offset should be updated after retry.
+
+Fri Dec  5 03:29:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_get_load_path): returns the load path without
+	  touching.
+
+	* load.c (rb_feature_provided): new function to return the loading
+	  path in addition to rb_provided().
+
+	* load.c (search_required): sets path if loading.
+
+	* variable.c (autoload_provided): load paths are expanded to check
+	  if loading.
+
+	* variable.c (autoload_node): keeps autoload mark while loading.
+	  [ruby-core:20235]
+
+	* variable.c (rb_const_get_0): loops while autoload mark is set.
+
+Fri Dec  5 01:37:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): ERROR_BROKEN_PIPE is not a real error
+	  at this point.
+
+	* io.c (pipe_open): use rb_w32_spawn() instead of rb_w32_pipe_exec()
+	  to use our own redirection scheme.
+
+Fri Dec  5 01:35:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_to_proc): use hidden object.
+
+Fri Dec  5 01:19:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): propagate taint status from format string to
+	  result string.
+
+Fri Dec  5 00:34:10 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (run_exec_dup2): need to sort by reverted order when
+	  restoring fds.
+
+Fri Dec  5 00:17:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_to_proc): caches Symbol procs, based on a patch from
+	  Shumpei Akai <admin AT flexfrank.net>.  [ruby-dev:37265]
+
+Thu Dec  4 23:29:34 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (waitpid): fix bug of checking child slot.
+
+	* win32/win32.c (FindChildSlotByHandle): new.
+
+Thu Dec  4 23:24:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.poutput3): new method.
+	  (Open3.poutput2): ditto.
+	  (Open3.poutput2e): ditto.
+
+Thu Dec  4 23:02:13 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/default.mspec: follows changes in rubyspec project.
+	  inherits configurations from ruby.1.9.mspec.
+
+Thu Dec  4 22:13:55 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_complex.rb: added some tests.
+
+	* test/ruby/test_rational.rb: ditto.
+
+Thu Dec  4 19:56:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.popen3): simplified.
+	  (Open3.popen_run): extracted from Open3.popen3.
+	  (Open3.popen2): new method.
+	  (Open3.popen2e): new method.
+	  (Open3.pipeline_rw): new method.
+	  (Open3.pipeline_r): new method.
+	  (Open3.pipeline_w): new method.
+	  (Open3.pipeline_run): new private method.
+
+Thu Dec  4 19:16:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (check_exec_fds): resolve cascaded child fd reference.
+
+Thu Dec  4 16:58:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rubygems/validator.rb (Gem#remove_leading_dot_dir): make
+	  this method private.  a patch from okkez in [ruby-dev:37245]
+
+Thu Dec  4 16:19:18 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read_nonblock):
+	  OpenSSL::SSL::SSLSocket should implement read_nonblock.  a patch
+	  from Aaron Patterson in [ruby-core:20277].  fix: #814 [ruby-core:20241]
+
+Thu Dec  4 16:16:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/gserver.rb: fixed type in sample code.  a report from Oleg
+	  Puchinin.
+
+Thu Dec  4 14:54:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rubygems/local_remote_options.rb (Gem#add_update_sources_option):
+	  little documentation fix.  a patch from okkez.  [ruby-dev:37271]
+
+Thu Dec  4 13:56:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/curses.c (window_getch): avoid ISPRINT() macro which
+	  has an issue with OpenSolaris.  [ruby-core:20189]
+
+	* ext/curses/curses.c (curses_getch): no ISPRINT(). [ruby-core:20294]
+
+	* signal.c (ruby_signal): EINVAL from sigaction(2) is not a bug.
+
+Thu Dec  4 11:40:56 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (inspect_enumerator): Implement #inspect.
+	  [ruby-dev:37248]-[ruby-dev:37263]
+
+Thu Dec  4 11:38:40 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* vm_method.c (rb_obj_respond_to): Remove a duplicated rdoc
+	  comment and fix a markup error.
+
+Thu Dec  4 06:04:16 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/menu.rb: TkOptionMenubutton.new fails to treat
+	  'parent' and 'variable' options on a Hash argument.
+
+Thu Dec  4 05:06:47 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: bug fix. use ::RubyVM instead of ::VM
+	  [ruby-list:45676]
+
+	* ext/tk/tcltklib.c: update RELEASE_DATE
+
+Thu Dec  4 01:37:47 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nurat_{to_s,inspect}): provides better representation
+	  for in-finite imag part.
+
+Thu Dec  4 01:22:41 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (f_signbit): NaN may be signed value.
+
+Wed Dec  3 23:59:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (EXEC_OPTION_DUP2_CHILD): defined.
+	  (check_exec_redirect_fd): check :in, :out and :err.
+	  (check_exec_redirect): check [:child, fd].
+	  (check_exec_fds): validate EXEC_OPTION_DUP2_CHILD array.
+	  (run_exec_dup2_child): new function.
+	  (rb_run_exec_options): call run_exec_dup2_child.
+
+Wed Dec  3 22:54:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (expr): keyword_not can continue across newline.
+	  [ruby-core:20252]
+
+Wed Dec  3 22:40:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (simple_default_value): returns simplest assignment only.
+	  [ruby-core:20237]
+
+Wed Dec  3 21:30:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (check_exec_redirect): accept :in, :out, :err as redirect
+	  target.
+
+Wed Dec  3 21:18:27 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_rational.rb: revert.
+
+Wed Dec  3 14:48:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c (ip_ruby_cmd, ip_invoke_with_position): must
+	  not access internal union directly.  [ruby-list:45670]
+
+Wed Dec  3 12:24:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_getc, rb_io_readchar): documentation correction from
+	  Emiel van de Laar.  [ruby-core:20212]
+
+	* ext/stringio/stringio.c (strio_readchar): ditto.
+
+Wed Dec  3 09:26:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/xpath.rb (REXML::XPath.first): apply documentation
+	  patch from Ken Bloom in [ruby-core:20213].
+
+	* lib/rexml/xpath.rb (REXML::XPath.each): ditto.
+
+Wed Dec  3 02:56:34 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rational.rb: add a test.
+
+Wed Dec  3 02:53:24 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_range.rb: add a test.
+
+Wed Dec  3 02:26:07 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_string.rb: add some tests.
+
+Wed Dec  3 02:04:21 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/pty/pty.c (Init_pty): fix typo.
+
+Tue Dec  2 19:22:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.popen3): merge hash options if given.
+
+Tue Dec  2 15:31:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/protocol.rb (Net::BufferedIO#rbuf_fill): use
+	  read_nonblock instead of sysread wrapped by timeout to boost
+	  performance.  a patch from Aaron Patterson in [ruby-core:20191].
+	  fix #806
+
+Mon Dec  1 23:23:52 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* set 1.9.1-p5000 into version number. [ruby-dev:36998]
+
+Mon Dec  1 15:48:47 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* signal.c (register_sigaltstack): no need to define on non-sigaltstack
+	  platform.
+
+Mon Dec  1 12:00:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (rb_fiber_start): calls with exact argument number.
+	  [ruby-core:20088]
+
+Sun Nov 30 21:41:10 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* man/rake.1: new manual page
+
+Sun Nov 30 18:01:50 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_regexp.rb (TestRegexp#test_parse_curly_brace):
+	  now accepts quantifier on anchors again by r20391.
+
+Sat Nov 29 23:56:44 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* man/irb.1 (EXAMPLES): new section
+
+Sat Nov 29 19:19:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regparse.c (is_invalid_quantifier_target): Perl and old Ruby
+	  accepts quantifier on anchors.  [ruby-core:20161]
+
+Sat Nov 29 18:28:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_getaddrinfo): should have updated for
+	  Mac OS X.  a patch from Shumpei Akai in [ruby-dev:37234]
+
+Sat Nov 29 00:18:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* cont.c (fiber_alloc): separate allocation and initialization.
+	  allow subclass to override #initialize.   [ruby-core:20086]
+
+Fri Nov 28 18:31:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_getaddrinfo): refactored to remove
+	  code duplication regarding getaddrinfo.
+
+Fri Nov 28 17:52:26 2008  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/forwardable.rb: should be usable def_single_delegator for
+	  Class and Module.
+
+Fri Nov 28 13:19:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (simple_default_value): extracts simplest default
+	  argument value.
+
+	* iseq.c (rb_iseq_parameters): returns parameter list.
+
+	* proc.c (get_proc_iseq, get_method_iseq): handles ifunc and
+	  bmethod.
+
+	* proc.c (rb_proc_parameters, rb_method_parameters): added
+	  Proc#parameters and Method#parameters.  [ruby-core:19759]
+
+Fri Nov 28 02:18:47 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_DoDivmod): bigdecimal
+	  division (including modulo) should raise ZeroDivisionError as
+	  integer division. [incompatible]
+
+Fri Nov 28 00:12:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flodivmod): floating point division should raise
+	  ZeroDivisionError as integer division. [incompatible]
+
+Thu Nov 27 23:54:37 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_mark): still needs to check stack depth during GC.
+
+	* gc.c (stack_check): ditto.
+
+Thu Nov 27 21:41:29 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* strftime.c (rb_strftime): should add padding for %%.
+
+Thu Nov 27 16:32:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (register_sigaltstack): stores alt stack for debug
+	  purpose.
+
+Thu Nov 27 16:12:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (ruby_sigaction_t): added.
+
+Thu Nov 27 15:59:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (ruby_stack_check): no check if using sigaltstack.
+
+	* signal.c (ALT_STACK_SIZE): default minimum size is insufficient
+	  for method calls.
+
+	* signal.c (sigsegv): handles stack overflow if possible.
+
+	* thread.c (ruby_thread_stack_overflow): helper function to raise
+	  sysstack_error.
+
+	* thread_pthread.c (ruby_stack_overflowed_p): checks for stack
+	  overflow.
+
+Thu Nov 27 10:40:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_div2): should return
+	  Integer for #div operation.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_div2): should raise
+	  ZeroDivisionError if divisor is zero.  [ruby-dev:37207]
+
+Wed Nov 26 23:15:47 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* strftime.c (STRFTIME): use rb_strftime() recursively, instead of
+	  platform's strftime().
+
+Wed Nov 26 22:46:23 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpException): bigdecimal zero
+	  division should raise FloatDomainError if mode
+	  VP_EXCEPTION_ZERODIVIDE is set.  [ruby-dev:37204]
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_mode): should handle
+	  VP_EXCEPTION_ZERODIVIDE.
+
+Wed Nov 26 15:16:07 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/gdbm/gdbm.c (rb_gdbm_nextkey): fix memory leak.
+
+Wed Nov 26 03:17:48 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_r): raise exception
+	  for nan/inf conversion.  [ruby-dev:37187] fix #793
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_i): ditto.
+
+Wed Nov 26 03:00:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpAlloc): avoid ALLOCA_N() to avoid
+	  segmentation fault caused by (insanely) long decimal values.
+	  [ruby-dev:37189] fix #794
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_dump, BigDecimal_to_i,
+	  BigDecimal_to_f, BigDecimal_to_s, BigDecimal_split,
+	  BigDecimal_inspect): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (VpToString): small performance
+	  improvement.
+
+Wed Nov 26 00:26:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* strftime.c (STRFTIME): should add padding for %[xXrR] etc.
+	  [ruby-dev:37185] fix: #792
+
+Tue Nov 25 16:26:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_times): taint (and untrust) status should be
+	  inherited by "ary * 0".  [ruby-dev:37024]
+
+Tue Nov 25 15:54:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* strftime.c (rb_strftime): should not swallow incomplete
+	  formatter, e.g. "%E".  [ruby-dev:37170] fix: #787
+
+	* strftime.c (rb_strftime): clear flags before processing unknown
+	  formatter, e.g. "%i".  [ruby-dev:37180]
+
+Tue Nov 25 10:35:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* strftime.c (rb_strftime): "%^P" should convert to upper case.
+	  [ruby-dev:37180]
+
+Tue Nov 25 07:51:18 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* strftime.c (FMT): use "%0d" formatter for zero padding, not "%.d".
+	  [ruby-dev:37168]  fix: #768
+
+	* strftime.c (rb_strftime): %s to use zero padding by default.
+	  [ruby-dev:37180]
+
+Tue Nov 25 03:37:42 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkextlib/blt/tabset.rb,
+	  ext/tk/lib/tkextlib/blt/tabnotebook.rb:
+	  fix many bugs. Now, those work properly.
+
+Tue Nov 25 03:26:04 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (num_step): treat infinite step specially.
+	  [ruby-dev:37157] fix: #781.
+
+Tue Nov 25 01:23:25 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (strftime): ignores '_' flag for %[LN].
+
+Tue Nov 25 00:08:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_fork): stops the timer thread during fork.
+	  [ruby-dev:37117]
+
+	* thread.c (rb_thread_start_timer_thread): timer thread needs
+	  system_working to be set.
+
+Mon Nov 24 23:27:28 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): The # flag should work with %a, %A, %b,
+	  %B, and %h.  [ruby-dev:37162]
+
+	* test/ruby/test_time.rb (test_strftime): ditto.
+
+Mon Nov 24 23:16:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (register_sigaltstack): should not add external
+	  variable (with some cosmetic changes).  [ruby-dev:37158]
+
+Mon Nov 24 22:57:25 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): A width specifier for %t and %n should
+	  work.  [ruby-dev:37160]
+
+	* test/ruby/test_time.rb (test_strftime): ditto.
+
+Mon Nov 24 22:07:07 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): The precision of %0N should be 9.
+	  [ruby-dev:37156]
+
+	* test/ruby/test_time.rb (test_strftime): ditto.
+
+Mon Nov 24 21:38:23 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): The default precision should be 1, not
+	  0.  [ruby-dev:37155]
+
+	* test/ruby/test_time.rb (test_strftime): ditto.
+
+Mon Nov 24 19:53:47 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (inspect): changed again.
+
+Mon Nov 24 18:35:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/time.rb: r20251 reverted.  The patched behavior do not round
+	  trip.  [ruby-core:19988]
+
+Sun Nov 23 16:04:05 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* signal.c (default_handler, Init_signal): compile error if
+	  USE_SIGALTSTACK is not defined.
+
+Sun Nov 23 00:04:14 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* signal.c (ALT_STACK_SIZE): 4KB is not enough on Mac OS X.
+	  Uses SIGSTKSZ.
+
+Sat Nov 22 21:29:54 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_method.rb (test_default_accessibility): test case for
+	  [ruby-dev:37124].
+
+Sat Nov 22 18:24:24 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_file_world_writable_p): should return nil for non
+	  world-writable files.
+
+Sat Nov 22 10:31:25 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkextlib/blt.rb, ext/tk/lib/tkextlib/blt/vector.rb:
+	  fix NameError bug.
+
+Sat Nov 22 03:41:22 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/pty/pty.c (get_device_once): abandon asynchronous exception
+	  that might cause serious problem if a program terminated early.
+	  asynchronous exception is a very bad thing anyway.  use
+	  Process.waitpid(pid) or PTY.check(pid) to poll program
+	  termination. if PTY.check is called with optional second
+	  argument being true, it raises an exception same as one from
+	  previous behavior.  [incompatible] fix: [ruby-core:19583]
+
+Fri Nov 21 22:24:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/curses.c (curses_escdelay_set): support ESCDELAY.  a
+	  patch from Giancarlo F Bellido <support at coaxialhost.com> in
+	  [ruby-core:19961].
+
+Fri Nov 21 22:17:15 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (usage): -W description updated.  [ruby-core:19858]
+
+Fri Nov 21 21:50:54 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (register_sigaltstack): use alternative stack for
+	  SIGSEGV to avoid uncaught stack overflow. based on a patch from
+	  Hiro Yoshioka <hyoshiok at miraclelinux.com> in [ruby-dev:37134].
+	  [ruby-dev:36993]
+
+Fri Nov 21 16:06:54 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm.c (thread_free): th->vm may be NULL when pthread_create
+	  failed for ENOMEM.  [ruby-dev:37095]
+
+Thu Nov 20 07:33:15 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/logger.rb (Logger): should handle the case that cvs/svn do
+	  not expand $Id keyword.  [ruby-core:19991]
+
+Thu Nov 20 07:27:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/minitest/unit.rb (MiniTest::Assertions#capture_io): adjust
+	  indentation to shut up warning.  [ruby-core:19993]
+
+Wed Nov 19 17:48:05 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/rubyext.c (rb_syck_mktime): return DateTime for a value
+	  out of range of Time.   [ruby-core:19919]
+
+Wed Nov 19 14:14:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* node.h (NOEX_MODFUNC): should be include NOEX_PRIVATE.
+	  [ruby-dev:37124]
+
+Wed Nov 19 03:01:04 2008  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb: fixed fails occasionally [ruby-dev:37119].
+	  thanks, shinichiro.h.
+
+Wed Nov 19 02:39:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (Init_syserr): moved function definition itself from
+	  template/known_errors.inc.tmpl.
+
+Wed Nov 19 01:10:52 2008  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (DRbConn::alive?): fixed NoMethodError problem
+	  from NaHi [ruby-dev:37110].
+
+Wed Nov 19 01:05:26 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* cont.c (cont_mark, cont_capture, cont_restore_1): speed up
+	  continuation by copying only valid YARV stack.  [ruby-dev:37106]
+
+Wed Nov 19 00:03:13 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/time.rb: according to RFC2822, -0000 means local time, +0000
+	  means UTC.  a patch from Chun Wang.   [ruby-core:19578]
+
+Tue Nov 18 23:02:23 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (proc_eq): equivalence check should not done by pointer
+	  comparison, but should be based on iseq contents.   [ruby-dev:37101]
+
+Tue Nov 18 20:30:08 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::FTP#open_socket): SOCKSsocket is obsolete.
+	  a patch from Alan Johnson <alan.wayne.johnson at gmail.com> in
+	  [ruby-core:19982].
+
+Tue Nov 18 16:18:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/logger.rb (ProgName): fixed for svn, based on a patch from
+	  Nobuhiro IMAI at [ruby-dev:37108].
+
+Tue Nov 18 15:56:55 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#read_request_line):
+	  use possessive match for path retrieval to avoid huge recursion
+	  for insanely long path.
+
+Tue Nov 18 15:50:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_desc): re might be NULL.
+
+	* regerror.c (onig_error_code_to_format): message updated.
+
+Tue Nov 18 09:42:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::make_switch): makes default
+	  conversion to splat.  [ruby-list:45645]
+
+Mon Nov 17 18:05:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (LIBS): fixed for disable-shared.  [ruby-dev:37103]
+
+Mon Nov 17 11:12:24 2008  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* bin/erb: remove -K option.
+
+Mon Nov 17 02:19:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBYARG_STATIC): fixed for --enable-rpath.
+
+Sat Nov 15 22:46:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_regsub): returns -1 unless ascii as well as
+	  rb_enc_ascget().  [ruby-dev:37097]
+
+Sat Nov 15 22:23:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LDSHARED): do not suppress debug info if higher debug
+	  level is given, on cygwin and mingw.
+
+Fri Nov 14 20:27:46 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* README.EXT, README.EXT.ja, vm_exec.c, vm_insnhelper.c: update
+	  filenames.
+
+Thu Nov 13 11:31:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_barrier_wait): gets rid of recursive deadlock.
+
+Thu Nov 13 06:08:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_barrier_{new,wait,release,destroy}): use Mutex so that
+	  circular requires fail with deadlock.  [ruby-core:19821]
+
+Wed Nov 12 07:16:01 2008  David Flanagan  <david@think32>
+
+	* ruby.c (set_internal_encoding_once): fix typo in error string
+
+Wed Nov 12 03:30:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_{external,locale}_str_new_cstr): optimized
+	  versions for literal strings.
+
+Wed Nov 12 03:28:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_require_safe): destroys barrier after successfully
+	  loaded, to get rid of loading same library again.  [ruby-core:19798]
+
+	* thread.c (rb_barrier_wait): can not wait destroyed barrier.
+
+	* thread.c (rb_barrier_destroy): destroys barrier so that no longer
+	  waited.
+
+Tue Nov 11 23:02:27 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* man/ruby.1 (RUBYOPT): wrote which option can appear.
+	  (--yydebug): wrote "do not use".
+	  (--dump): wrote "do not use".
+
+Tue Nov 11 21:22:28 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#check_path): REL_PATH should a
+	  symbol.  [ruby-core:19805]
+
+Tue Nov 11 20:49:29 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* configure.in: fix SEGV on Mac OS X 10.5.3 with pthread.
+	  a patch from Wataru Kimura in Bug #193 [ruby-core:17333].
+
+Tue Nov 11 20:39:27 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (inspect): reverted the previous change.
+
+Tue Nov 11 17:35:25 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_accept): secure fd before accept because if
+	  error causes in securing, cannot restore the state of accepted
+	  socket.
+	  fixed [ruby-core:19728]
+
+Tue Nov 11 14:40:40 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb: unnecessary setup method
+	  (setup_really_needed?) removed
+
+Tue Nov 11 14:25:25 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/single_byte.trans, macgreek-tbl.rb, macroman-tbl.rb,
+	  macromania-tbl.rb, macturkish-tbl.rb, macukraine-tbl.rb,
+	  ibm437-tbl.rb, ibm852-tbl.rb, ibm855-tbl.rb, ibm857-tbl.rb,
+	  ibm860-tbl.rb, ibm861-tbl.rb, ibm862-tbl.rb, ibm863-tbl.rb,
+	  ibm865-tbl.rb, ibm866-tbl.rb, ibm869-tbl.rb, ibm775-tbl.rb:
+	  new single-byte transcodings (from Yoshihiro Kambayashi)
+
+	* test/ruby/test_transcode.rb: added tests for the above
+	  (from Yoshihiro Kambayashi)
+
+Tue Nov 11 13:20:23 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_r): moved from
+	  bigdecimal/util, converted into C.  [ruby-dev:36805]
+
+Tue Nov 11 07:55:57 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (inspect): changed the format.
+
+Mon Nov 10 17:34:50 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/sdbm/test_sdbm.rb (TestSDBM#test_index): Reflect the method
+	  name change from #index to #key and remove a duplicated test
+	  method.
+
+Mon Nov 10 10:50:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator.delegating_block): should not refer
+	  DelegateClass specific @delegate_dc_obj.  a patch from Erik
+	  Hollensbe in [ruby-core:19671].
+
+Mon Nov 10 09:54:21 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (ifs_open_socket): should retry without proto_buffer
+	  if cannot find the suitable protocol. a patch from Heesob Park.
+	  fixed [ruby-core:19713]
+
+Sun Nov  9 13:04:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_initialize_m): avoid unnecessary array
+	  allocation.
+
+Sun Nov  9 04:10:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (control_frame_dump): suppresses finished method name.
+
+Sun Nov  9 00:30:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (thread_timer): checks working flags again.
+
+Sun Nov  9 00:02:01 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/session/pstore.rb: fix indentation.
+
+Sat Nov  8 23:47:45 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/session.rb (FileStore): use marshalized data.
+
+	* test/cgi/session_dir: add a session directory in test.
+
+	* test/cgi/test_cgi_session.rb: add a test.
+
+Sat Nov  8 21:57:03 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/session.rb: remove debug code.
+
+Sat Nov  8 21:33:53 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/session.rb: fix indentation.
+
+Sat Nov  8 18:11:14 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#do_CONNECT):
+	  use #bytesize instead of #size.  a patch submitted from
+	  raspberry lemon in [ruby-core:18571].
+
+	* lib/webrick/httpauth/digestauth.rb, lib/webrick/httpproxy.rb,
+	  lib/webrick/httprequest.rb, lib/webrick/httpservlet/cgi_runner.rb,
+	  lib/webrick/httpservlet/abstract.rb, lib/webrick/httpresponse.rb,
+	  lib/webrick/httpservlet/cgihandler.rb, lib/webrick/utils.rb: ditto.
+
+Sat Nov  8 17:31:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_ungetc): ungetc position should
+	  not be changed from 1.8.  [ruby-core:19632]
+
+Sat Nov  8 17:15:08 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#make_switch): 1.9 Symbols respond
+	  to :match.  a patch from keith cascio, <keith at ucla.edu>.
+	  [ruby-core:19730]
+
+Sat Nov  8 12:10:15 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/gdbm/gdbm.c (fgdbm_index): make #index warn like Hash.
+	  [ruby-dev:37039]
+
+	* ext/sdbm/init.c (Init_sdbm): typo fixed.  [ruby-dev:37039]
+
+Sat Nov  8 07:07:48 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: detect stdio buffer pointers for uClibc.  a patch
+	  from Brian Candler <B.Candler at pobox.com> in [ruby-core:12020].
+	  [ruby-core:19713]
+
+Sat Nov  8 06:51:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/dbm/dbm.c (fdbm_index): make #index warn like Hash.
+
+	* ext/dbm/dbm.c (fdbm_key): new method.
+
+	* ext/sdbm/init.c: ditto.
+
+Sat Nov  8 06:20:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/curses.c: curses encoding should obey locale.
+
+	* ext/curses/curses.c (curses_getch): 1.9 getch should return one
+	  character string for single byte string.  wchar_t support may
+	  follow in the future.
+
+Sat Nov  8 05:46:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h (rb_io_t): added write_lock to serialize.
+	  [ruby-core:19668]
+
+	* gc.c (gc_mark_children): mark write_lock.
+
+	* gc.c (rb_gc_call_finalizer_at_exit): Mutex also needs to be
+	  deferred.
+
+	* io.c (io_fflush, io_binwrite, fptr_finalize): serializes writing.
+
+Sat Nov  8 05:31:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* safe.c (safe_setter): add warning for $SAFE=3.
+
+Sat Nov  8 01:48:07 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): do not wait never
+	  coming signal if failed to create timer thread.
+
+	* thread_pthread.c (native_cond_timedwait): returns error code.
+
+	* thread_pthread.c (thread_timer): typo fixed.
+
+Sat Nov  8 01:14:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (thread_timer, rb_thread_create_timer_thread):
+	  handshakes properly.
+
+Fri Nov  7 22:51:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* inits.c (rb_call_inits): do not repeat.
+
+Fri Nov  7 22:46:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (thread_timer): initializes mutex each time.
+
+Fri Nov  7 21:23:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_win32.c (thread_start_func_1): use already gotten stack info.
+
+Fri Nov  7 12:26:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb.rb (IRB::Irb#eval_input): remove extra  @context.prompt_i.
+	  [ruby-core:19718]
+
+Fri Nov  7 11:55:30 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* nkf.c: update to r1.188. fixes for 16bit environment.
+
+Fri Nov  7 07:43:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_timer): pthread_cond_timedwait returns ETIMEDOUT
+	  when timed out.
+
+Fri Nov  7 07:32:55 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb: support Rational and Complex as 1.8
+	  does.  a patch from Hiroshi Moriyama in [ruby-dev:36899].
+
+Fri Nov  7 07:12:06 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (thread_timer): uses pthread_cond_timedwait always
+	  instead of pthread_kill.
+
+Fri Nov  7 06:58:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (Init_String): remove Symbol#===.  [ruby-dev:37026]
+
+Fri Nov  7 04:02:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_timer): uses timedwait on cygwin.
+
+Fri Nov  7 02:00:12 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/rexml/entity.rb (unnormalized): do not call
+	  document.record_entity_expansion if document is nil.
+	  see <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=502535>.
+	  Thanks, Naohisa Goto.
+
+	* test/rexml/test_document.rb: ditto.
+
+Fri Nov  7 01:24:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: binary encoding spec is no longer needed in 1.9.
+
+Thu Nov  6 22:21:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_stop_timer_thread): terminates timer thread
+	  immediately.
+
+Thu Nov  6 21:21:46 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fole_s_connect, fole_initialize,
+	  folevariant_initialize): check argument type of WIN32OLE.connect,
+	  WIN32OLE.new, WIN32OLE_VARIANT.new.
+
+	* test/win32ole/test_win32ole.rb (test_s_new_exc, test_s_connect_exc):
+	  ditto.
+
+	* test/win32ole/test_win32ole_variant.rb (test_s_new_exc): ditto.
+
+	* test/win32ole/test_win32ole_method.rb: add assertion of
+	  WIN32OLE_METHOD.new.
+
+Tue Nov  4 13:08:01 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit.setup_argv): now can specify ``-xname''
+	  instead of ``-x name''.
+
+Tue Nov  4 08:00:00 2008  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* test/zlib/test_zlib.rb(test_readchar): compare in same type.
+
+Tue Nov  4 03:28:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_join): do not repeat self in a recursive array.
+	  [ruby-dev:37019]
+
+Tue Nov  4 01:56:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_mutex_synchronize): fixed prototype.
+
+	* thread.c (rb_mutex_synchronize): fixed function name.
+
+Mon Nov  3 16:32:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing.h: needs RUBY_EXTERN instead of extern for mswin.
+
+Mon Nov  3 13:01:13 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* array.c (rb_ary_take_first_or_last): renamed rb_ary_partial more
+	  appropriately.
+	  (ary_take_pos_flags): new enum. makes the fourth argument of
+	  rb_ary_take_first_or_last more descriptive.
+	  (rb_ary_partial): renamed rb_ary_partial0 more appropriately.
+	  takes a new parameter klass for what class to instantiate.
+
+Mon Nov  3 13:00:26 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* range.c (rb_range_beg_len): rb_raise takes C values as arguments.
+	  [BUG] ranges turned into other value in a message of RangeError.
+
+Mon Nov  3 00:36:44 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_event.rb: rename test class.
+
+	* test/win32ole/test_win32ole_variant_outarg.rb: add in order
+	  to use ADO instead of Internet Explorer.
+	  remove test/win32ole/test_win32ole_variant_with_ie.rb.
+
+	* test/win32ole/orig_data.csv: ditto.
+
+Sun Nov  2 23:08:08 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_err_in_callback.rb: add teardown.
+
+Sun Nov  2 22:12:41 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_event_free): invoke IConnectionPoint::Unadvise
+	  before invoking IConnectionPoint::Release.
+
+	* test/win32ole/test_win32ole_event.rb, test/win32ole/err_in_callback.rb,
+	  test/win32ole/test_err_in_callback.rb: Use ActiveX Data Object
+	  Library instead of InternetExplorer.
+
+Sun Nov  2 22:06:55 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* win32/Makefile.sub: add RUNRUBYOPT. [ruby-dev:37009]
+
+Sun Nov  2 16:41:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (Init_zlib): defines readbyte instead redefining
+	  readchar.
+
+Sat Nov  1 07:09:40 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* gem_prelude.rb: Don't require rubygems/defaults.rb.
+
+Fri Oct 31 21:58:50 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/converter.rb (RSS::Converter): use String#encode.
+
+Fri Oct 31 21:28:14 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/webrick/httpauth/digestauth.rb
+	  (WEBrick::HTTPAuth::DigestAuth#initialize): fix typo.
+
+Fri Oct 31 18:51:42 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_array.rb (test_array_subclass): test for r20076.
+
+Fri Oct 31 18:42:19 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* array.c (ary_partial0): recovers the mistakenly changed feature at r19723.
+	  Array#[m,n] had returned an Array even for an instance of a
+	  subclass.
+
+Fri Oct 31 18:05:05 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/single_byte.trans, maccroatioan-tbl.rb,
+	  maccyrillic-tbl.rb, maciceland-tbl.rb: new single-byte
+	  transcodings (from Yoshihiro Kambayashi)
+
+	* test/ruby/test_transcode.rb: added tests for the above
+	  (from Yoshihiro Kambayashi)
+
+Fri Oct 31 12:51:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_globs): need taint check.  reported by steve
+	  <oksteev at gmail.com>
+
+Fri Oct 31 12:42:45 2008  wanabe  <s.wanabe@gmail.com>
+
+	* array.c (rb_ary_decrement_share): fix to work recycling
+	  shared-array without the following.  [ruby-dev:36991]
+
+	* array.c (ary_make_substitution): don't recycle substitution array.
+
+Fri Oct 31 12:27:52 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/default.mspec: supports that the build directory != srcdir.
+
+Fri Oct 31 11:35:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_set_default_internal): no alias "internal" if
+	  default internal is not set.  [ruby-dev:37000]
+
+Fri Oct 31 08:16:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): delays setting safe level.
+	  [ruby-dev:36997]
+
+Thu Oct 30 21:32:15 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_permutation): hide temporal array.
+
+Thu Oct 30 14:45:45 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/single_byte.trans: refactoring to make it easier
+	  to add more transcodings (with Yoshihiro Kambayashi)
+
+	* enc/trans/iso-8859-1-tbl.rb: new file to avoid having to
+	  treat ISO-8859-1 as special
+
+Thu Oct 30 14:10:46 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): templates a, A, Z should propagate encoding
+	  of original strings.  templates b, B, h, H brings ASCII-8BIT to
+	  the result.
+
+Thu Oct 30 10:34:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (moreswitches): splits option string and passes arguments.
+
+	* ruby.c (proc_options): checks if allowed in RUBYOPT.
+
+	* ruby.c (process_options): allows long style options in RUBYOPT.
+
+	* ruby.c (load_file_internal): ditto in shebang.  [ruby-dev:36979]
+
+Thu Oct 30 09:31:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_locale_encoding): makes an alias for locale.
+	  [ruby-dev:36976]
+
+	* encoding.c (rb_enc_set_default_external): ditto for external.
+
+	* encoding.c (rb_enc_set_default_internal): ditto for internal.
+
+Thu Oct 30 03:30:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (load_file_internal): should not set script encoding to
+	  ASCII-8BIT.
+
+	* ruby.c (load_file_internal): do not auto convert scripts even
+	  when default_internal is set.  [ruby-core:19579]
+
+Thu Oct 30 02:20:33 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_sort_bang): remove SEGV when replacing array with
+	  embedded one during sort.  [ruby-dev:36983]
+
+	* test/ruby/test_array.rb (test_sort!): add tests for above.
+
+Thu Oct 30 01:44:23 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb: add some tests.
+
+Thu Oct 30 01:32:15 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_splice): remove redundant check.
+
+Thu Oct 30 01:24:55 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (rb_str_cmp): fix condition which always returned true
+	  because ENCODING_GET returns unsigned long.
+
+Thu Oct 30 01:10:32 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb (test_permutation): add a test that replaces
+	  array during permutation.
+
+Wed Oct 29 23:31:34 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_check_dummy_enc): new function to check dummy
+	  encoding.
+
+	* string.c (rb_str_upcase_bang): case conversion functions should
+	  not be applicable to strings in dummy encoding.  [ruby-dev:36985]
+
+Wed Oct 29 23:57:29 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_sort_bang): replacing array during sort broke
+	  invariant of array.  [ruby-dev:36983]
+
+Wed Oct 29 21:06:46 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_rewind): should reset fptr->readconv if it's
+	  available.
+
+	* io.c (more_char): clear readconv at EOF.
+
+	* test/ruby/test_file.rb: should not read after EOF.  use rewind
+	  instead.
+
+Wed Oct 29 20:45:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/webrick/utils.rb (start_server): provide a reference to log of
+	  webrick.
+
+	* test/webrick/test_httpproxy.rb, test/webrick/test_httpauth.rb,
+	  test/webrick/test_cgi.rb, test/webrick/test_httpserver.rb,
+	  test/webrick/test_server.rb, test/webrick/test_filehandler.rb: use
+	  webrick log as an assertion message.
+
+Wed Oct 29 16:41:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack): set encoding from packing string and UTF-8 for
+	  "U".  [ruby-core:19590]
+
+Wed Oct 29 15:50:00 2008  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* complex.c (imp1, imp2): should declare type.
+	  [BUG] at IA-64
+
+Wed Oct 29 14:36:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (revision.h): have to be updated daily or by non-trivial
+	  commits.
+
+Wed Oct 29 13:18:02 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* README, README.ja: Update the portability section.
+
+Wed Oct 29 13:02:17 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in (rb_cv_lib_xpg4_needed): Drop legacy FreeBSD
+	  support regarding libxpg4.  Those old versions of FreeBSD
+	  shipped with the library don't have a working pthread library
+	  anyway.
+
+Wed Oct 29 11:50:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (dln-a-out): cannot make shared library nor work with
+	  ELF.  [ruby-core:19571]
+
+Wed Oct 29 11:00:25 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_open): shouldn't seek here.
+
+	* win32/win32.c (rb_w32_write): write to the end of the file when
+	  FAPPEND is specified.
+
+Wed Oct 29 09:30:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): cache common interned IDs.
+
+Wed Oct 29 00:43:39 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: uses macro in declarations.
+
+Wed Oct 29 00:26:00 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_open): need to seek to the end of the file when
+	  O_APPEND is specified.
+
+Wed Oct 29 00:08:05 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* math.c (rb_math_{atan2,cos,cosh,hypot,log,sin,sinh,sqrt}): added.
+
+	* complex.c: follows the above change.
+
+Tue Oct 28 23:29:06 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (kanji_convert): output unicode chars.
+	  [ruby-dev:36957]
+
+	* ext/nkf/nkf-utf8/nkf.c (numchar_getc): increase buffer size.
+	  reported and patched at [ruby-dev:36957] by mame.
+
+Tue Oct 28 23:03:46 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/test_proc.rb: filled all patterns for testing
+	  vm_yield_setup_args.
+
+Tue Oct 28 23:05:35 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: continues to support canonicalization *unofficially*
+	  for an odd library mathn for the time being (only 1.9.x).
+	  since grand mathn must be very very special library for us.
+
+	* rational.c: ditto.
+
+	* ext/mathn/*/*: follow the above changes.
+
+Tue Oct 28 21:31:55 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (parser_yylex): check EOF explicitly.
+
+Tue Oct 28 20:59:12 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (extract_binmode): new function to extract binmode/textmode
+	  options from hash.
+
+	* io.c (rb_io_extract_modeenc): use above function.
+
+	* io.c (rb_io_s_pipe): recognize binmode/textmode options.
+
+Tue Oct 28 20:15:49 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (make_readconv): now can specify the size of cbuf.
+
+	* io.c (read_all, appendline, io_getc, rb_io_ungetc): follow above
+	  change.
+
+Tue Oct 28 19:00:51 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pipe_exec): internal fds should be always
+	  binmode.
+
+Tue Oct 28 17:21:18 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* tool/make-snapshot.rb: merged from ruby_1_9_1.
+
+Tue Oct 28 16:56:31 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* gem_prelude.rb: merged from ruby_1_9_1 branch.
+	  considers --program-suffix and prefix configure options.
+
+	* lib/rubygems/defaults.rb: ditto.
+
+	* test/rubygems/test_gem.rb (@default_dir_re): ditto.
+
+Tue Oct 28 16:13:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: use String#bytesize.
+
+Tue Oct 28 09:51:48 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bootstraptest/test_thread.rb: ignore some exceptions.
+	  [ruby-dev:36951]
+
+Tue Oct 28 09:19:40 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (RUNRUBY): now ruby requires something from
+	  gem_prelude, so need to set library path.
+
+Mon Oct 27 22:29:11 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* vm_insnhelper.c (vm_yield_setup_args): supports optional parameters.
+	  Fixed [ruby-core:19503].
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): a new function. extracted
+	  from vm_yield_setup_args.
+
+	* vm_insnhelper.c (vm_yield_setup_block_args_complex): ditto.
+
+	* test/ruby/test_proc.rb: added tests for arguments on a Proc from
+	  Kernel#proc called.
+
+Mon Oct 27 20:03:05 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/mathn/complex/complex.c: no need to define rb_cComplex because
+	  it's already defined at $(srcdir)/complex.c.
+
+	* ext/mathn/rational/rational.c: no need to define rb_cRational because
+	  it's already defined at $(srcdir)/rational.c.
+
+Mon Oct 27 15:58:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file): preserves $..  [ruby-dev:36937]
+
+	* io.c (argf_init): initial value of $. should be 0.
+	  see [ruby-dev:36937].
+
+Mon Oct 27 15:38:38 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/minitest/test_mini_test.rb
+	  (test_assert_raises_triggered_different): the test failed when a
+	  file path in the backtrace contains something except [\w\/\.],
+	  e.g. hyphen, Japanese characters or backslash.
+
+	* test/minitest/test_mini_test.rb
+	  (test_assert_raises_triggered_subclass): ditto.
+
+Mon Oct 27 15:32:08 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/minitest/test_mini_test.rb: fixed that r19958 made
+	  test-all fail when the ruby was built at $(srcdir).
+
+Mon Oct 27 12:04:58 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/ruby/envutil.rb: reverted the changeset 19948 because it
+	  concealed unexpected behaviours of ruby.
+
+	* test/ruby/test_io.rb: ditto.
+
+Mon Oct 27 12:02:53 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/nkf/nkf.c (rb_nkf_convert): should specify type of variable.
+
+Mon Oct 27 11:34:56 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/minitest/test_mini_test.rb (test_filter_backtrace):
+	  fixed failure on Ruby built at out of $(srcdir).
+
+	* test/minitest/test_mini_test.rb (test_Filter_backtrace_unit_starts):
+	  ditto.
+
+Mon Oct 27 10:02:47 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (read_all): the 3rd argument maybe Qnil.
+
+Mon Oct 27 01:29:55 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/nkf/test_nkf.rb: add a test for [ruby-dev:36909].
+
+Mon Oct 27 01:05:48 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_convert): avoid GC.
+	  reported in [ruby-dev:36909] and patched [ruby-dev:36941] by mame.
+
+Sun Oct 26 21:28:42 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/json/test_json.rb: remove dependency on permutation gem.
+
+Sun Oct 26 19:18:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (blocking_region_{begin,end}): declared as inline.
+
+	* util.c (freedtoa): used only when MULTIPLE_THREADS is not defined.
+
+	* win32/win32.c (rb_w32_pipe): serial is DWORD.
+
+	* ext/etc/etc.c (sGroup): getgrent may not be available.
+
+Sun Oct 26 18:11:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/wait/{extconf.rb,wait.c}: needs sys/ioctl.h for fcntl on
+	  cygwin.
+
+Sun Oct 26 18:18:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gem_prelude.rb: disables debug and verbose flags to suppress failure
+	  messages.  interpolation does not occur inside single quotes.
+
+Sun Oct 26 18:42:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_in_out_err): disables builtin rubygems
+	  to get rid of the interference.
+
+	* test/ruby/test_io.rb (ruby): ditto.
+
+Sun Oct 26 13:30:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (read_all): use the given buffer to read when needs readconv.
+	  based on the patch by wanabe <s.wanabe AT gmail.com> at
+	  [ruby-dev:36931].
+
+Sun Oct 26 07:57:35 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*: Update to RubyGems 1.3.1 r1909.
+
+Sun Oct 26 07:35:56 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/unit.rb: Imported minitest 1.3.0 r4429.
+	* test/minitest/*: ditto.
+
+Sun Oct 26 02:16:29 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* configure.in ($MANTYPE): followed ruby.1, which had moved.
+
+	* instruby.rb (:man): ditto.
+
+	* mdoc2man.rb: fixed for the case Xr with 2 arguments.
+
+Sun Oct 26 01:42:45 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* man/irb.1: new manual page.
+
+Sun Oct 26 00:11:48 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* man/ruby.1: moved into man/. added some descriptions.
+
+	* man/goruby.1: new file.
+
+	* instruby.rb (:man): followed ruby.1.
+
+Sat Oct 25 20:05:22 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (gc_profile_result): bug fix. The millisecond is 1/1000.
+
+Sat Oct 25 11:12:40 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ruby.c (usage): updated the description of -E option.
+
+Sat Oct 25 09:54:10 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb:  Fixed a bug in read_to_char() that would slurp
+	  whole files if the encoding was invalid.  It will now read
+	  up to 10 bytes ahead to find a valid character boundary or
+	  give up.  [ruby-core:19465]
+
+	* test/csv/test_features.rb, test/csv/test_table.rb, test/csv/test_row.rb:
+	  Loosened some tests to check for a compatible? Encoding instead
+	  of an exact Encoding.  [ruby-core:19470]
+
+Sat Oct 25 07:42:49 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc*: Update to RDoc 2.2.2 r192.
+
+Sat Oct 25 04:00:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/us_ascii.c (us_ascii_mbc_enc_len): made static.  a patch by
+	  Tadashi Saito <shiba AT mail2.accsnet.ne.jp> at [ruby-dev:36916]
+
+	* signal.c (signal_buff): made static.  a patch by Tadashi Saito
+	  <shiba AT mail2.accsnet.ne.jp> at [ruby-dev:36915]
+
+	* util.c (freedtoa): made static to get rid of name clash.  a patch by
+	  Tadashi Saito <shiba AT mail2.accsnet.ne.jp> at [ruby-dev:36913]
+
+	* util.c (ruby_dtoa): added prefix, ditto.
+
+Sat Oct 25 03:43:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lambda): need to adjust lpar_beg for both of the parser and
+	  ripper.
+
+Sat Oct 25 01:29:34 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/mathn/rational/rational.c: undef duplicated macro definition
+	  first to shut up the warning.
+
+	* ext/mathn/rational/rational.c (rb_rational_new_mathn): specify
+	  omitted return value.
+
+Sat Oct 25 00:14:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sample): fixed sizes and randomness.
+
+Fri Oct 24 23:04:42 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* configure.in (sitedir): considers --program-prefix and
+	  --program-suffix.
+	  (vendordir): ditto.
+	  (rubyhdrdir): ditto.
+
+	* mkconfig.rb (CONFIG["rubylibdir"]): ditto.
+
+	* win32/Makefile.sub: ditto.
+
+	* instruby.rb (:rdoc): ditto.
+
+	* lib/rdoc/ri/paths.rb (RDoc::RI::Paths): ditto.
+
+Fri Oct 24 22:59:27 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb (test_sample): add tests for size of
+	  returned array and randomness.
+
+Fri Oct 24 21:50:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* id.h, template/id.h.tmpl (enum ruby_method_ids): reordered.
+
+Fri Oct 24 21:19:45 2008  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* ext/mathn/rational.c: change several global functions.
+
+Fri Oct 24 21:05:38 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/html.rb: allow symbolized key.
+
+	* test/cgi/test_cgi_tag_helper.rb: add a test.
+
+Fri Oct 24 20:54:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (exts): renamed to get rid of the existent directory.
+
+Fri Oct 24 20:45:41 2008  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* ext/mathn/rational.c: change several global functions to static
+	  function.
+
+Fri Oct 24 19:57:19 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_dump): should have removed commented out
+	  unnecessary code.  [ruby-cvs:27131]
+
+Fri Oct 24 19:16:36 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* id.h (enum ruby_method_ids): ISO C forbids comma at end of
+	  enumerator list.
+
+Fri Oct 24 18:29:00 2008  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/mathn.rb: moved into ext/mathn/rational/rational.c and
+	  ext/mathn/complex/complex.c.
+	* ext/mathn: new mathn ext-libraries.
+
+Fri Oct 24 18:21:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_array.rb (TestArray#test_join): should restore
+	  global variable after the test.  [ruby-dev:36896]
+
+	* test/ruby/test_hash.rb (TestHash#test_to_s): ditto.
+
+Fri Oct 24 17:43:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (lambda): need to adjust lpar_beg for ripper as well.
+	  [ruby-dev:36702]
+
+Fri Oct 24 16:13:12 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (read_multipart): change field value as String
+	  from StringIO or Tempfile when multipart parse except file field.
+	  add files method that can get uploaded files. [ruby-dev:36547]
+
+	* test/cgi/test_cgi_multipart.rb: fix the test for core.rb.
+
+Fri Oct 24 14:22:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ext): split out the target for extension library.
+
+	* common.mk (encs): depends on enc and trans.
+
+Fri Oct 24 10:23:24 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_signal.rb (TestSignal#test_exit_action): Thread
+	  may terminate before sysread. [ruby-dev:36835]
+
+Thu Oct 23 22:06:23 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* distruby.rb: removed. "make dist" uses tool/make-snapshot for the 1.9
+	  series. distruby.rb has been no longer used.
+
+Thu Oct 23 17:30:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c (rb_class_instance_methods): rdoc fixed.  [ruby-core:19458]
+
+Thu Oct 23 16:19:33 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/init.rb (IRB.load_modules): catch LoadError explicitly.
+	  a patch from Daniel Berger <Daniel.Berger at qwest.com> in
+	  [ruby-core:19451].  fix [ruby-core:19450]
+
+Thu Oct 23 16:16:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_getservbyport): check if the port range.
+	  [ruby-core:19460]
+
+Thu Oct 23 14:08:07 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_getservbyname): unuse unnecessary
+	  object.
+
+	* ext/socket/socket.c (sock_s_getservbyport): the port should be
+	  converted before the proto so that the #to_int of the former cannot
+	  alter the latter.
+
+Thu Oct 23 10:55:08 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_dump): use \u{ff} escape for UTF-8 encoding
+	  string.
+
+Thu Oct 23 09:26:22 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_getservbyport): cast to get rid of
+	  warning.
+
+Thu Oct 23 08:06:03 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_getservbyport): port should be
+	  network byte order.  a patch from Dave Thomas in [ruby-core:19454].
+
+Thu Oct 23 01:26:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_conv_enc_opts): new function to convert with
+	  specifying ecflags and ecopts.
+
+	* ext/zlib/zlib.c (gzfile_newstr): specify ecflags and ecopts for
+	  conversion using above function.
+
+	* ext/zlib/zlib.c (gzfile_newstr): use own rb_econv_t for dummy
+	  encoding to handling stateful encoding (e.g. iso-2022-jp).
+	  [ruby-dev:36857]
+
+	* ext/zlib/zlib.c (gzfile_getc): ditto.
+
+Thu Oct 23 01:24:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_tmpsrc): get rid of side effects.
+
+Thu Oct 23 00:10:01 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: separate data structure between rb_fiber_t and
+	  rb_context_t (rb_fiber_t includes rb_context_t).
+	  rb_fiber_t has double linked list of fibers in same threads.
+
+Wed Oct 22 17:25:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (pipe_open): remove unnecessary flush before fork.
+	  a patch from Kazuhiro NISHIYAMA <zn at mbf.nifty.com>
+	  in [ruby-dev:36840].
+
+Wed Oct 22 16:27:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_yield_setup_args): should not drop arguments
+	  corresponding to default parameters.
+
+Wed Oct 22 14:59:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_larglist): should not allow semicolon separated local
+	  variable declarations if formal argument list is not surrounded
+	  by parentheses, mostly because semicolon outside of parentheses
+	  appears to terminate the expression.
+
+Wed Oct 22 14:52:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_external_str_new_with_enc): no implicit strlen call.
+	  [ruby-dev:36854]
+
+	* string.c (rb_external_str_new_cstr): new function to create
+	  string from external NUL terminated C string.
+
+	* string.c (rb_locale_str_new_cstr): ditto.
+
+	* ext/readline/readline.c: now use rb_locale_str_new_cstr().
+
+	* test/sdbm/test_sdbm.rb (TestSDBM#test_delete_with_block):
+	  deleted key to the block may be a copy of specified key.
+
+	* test/dbm/test_dbm.rb (TestDBM#test_delete_with_block): ditto.
+
+Wed Oct 22 13:16:47 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (unescape_escaped_nonascii): back out the last change on
+	  the function.  [ruby-dev:36818]
+
+Wed Oct 22 07:09:19 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/zlib/zlib.c (rb_gzreader_ungetc): should be able to unget
+	  Fixnum.
+
+	* ext/stringio/stringio.c (strio_ungetc): should convert unget
+	  string.
+
+	* ext/stringio/stringio.c (strio_ungetbyte): new method.
+
+Wed Oct 22 05:46:25 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/vm.h: write a comment.
+
+	* common.mk: remove vm.h from any rules.
+	  vm.h is not needed on this version.
+
+	* vm_core.h: ditto.
+
+Wed Oct 22 05:39:39 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (arg_concat_gen): concat target node should be NODE_ARRAY.
+	  [ruby-core:19413]
+
+	* bootstraptest/test_method.rb: add tests.
+
+Wed Oct 22 00:29:13 2008  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb (test_core_03_notify): Fixed test failures
+	  [ruby-dev:36837].
+
+Wed Oct 22 00:22:06 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_external_str_new_with_enc): wrong condition to
+	  calculate strlen().
+
+	* ext/readline/readline.c: add encoding support.
+
+Tue Oct 21 23:12:24 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_write): should convert writing
+	  string to the encoding of the buffer.
+
+	* hash.c (rb_any_hash): typo fixed.
+
+	* ext/zlib/zlib.c (rb_gzwriter_write): oops, IO string conversion
+	  need to be done by to_s.
+
+Tue Oct 21 22:38:58 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (open_key_args): should adjust argc, argv in struct
+	  foreach_arg.
+
+Tue Oct 21 21:52:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/zlib/zlib.c (rb_gzwriter_write): conversion should be done
+	  using to_str, not to_s.
+
+	* ext/zlib/zlib.c (rb_gzwriter_write): need proper conversion
+	  according to gz encoding.
+
+	* ext/zlib/zlib.c (rb_gzreader_ungetc): convert string encoding
+	  before unget.
+
+Tue Oct 21 21:33:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (assert_nothing_raised): raise with backtrace.
+
+Tue Oct 21 20:12:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit.rb (assert_nothing_raised): set backtrace.
+
+Tue Oct 21 18:17:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/zlib/zlib.c: remove obsolete prototype macros.
+
+	* ext/zlib/zlib.c (struct gzfile): add encoding field to gzfile
+	  structure.
+
+	* ext/zlib/zlib.c (rb_gzreader_getc): now works on characters.
+
+	* ext/zlib/zlib.c (rb_gzreader_getbyte): new method to retrieve
+	  single byte.
+
+	* ext/zlib/zlib.c (rb_gzreader_readbyte): ditto.
+
+	* ext/zlib/zlib.c (rb_gzreader_each_byte): renamed from each_char
+
+	* ext/zlib/zlib.c (rb_gzreader_ungetc): allow unget strings.
+
+	* ext/zlib/zlib.c (rb_gzreader_ungetbyte): renamed from ungetc.
+
+Tue Oct 21 13:28:42 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* io.c, include/ruby/intern.h (rb_io_ascii8bit_binmode): externed.
+
+	* ext/socket/socket.c (init_sock): use rb_io_ascii8bit_binmode()
+	  instead of rb_io_binmode().
+
+Tue Oct 21 13:17:39 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* mdoc2man.rb: moved into tools/.
+
+	* instruby.rb: followed the change of mdoc2man.
+
+Tue Oct 21 11:34:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (strip_glibc_option, map_charset): check if
+	  encoding is a string.  based on the patch by Hiroshi Moriyama at
+	  [ruby-dev:36811].
+
+	* test/iconv/test_basic.rb (test_invalid_arguments): added tests.
+
+Tue Oct 21 10:40:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_file_open_internal): should initialize fmode before using.
+
+Tue Oct 21 08:56:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regparse.c (parse_char_class): CCV_SB is only for single
+	  byte.  [ruby-dev:36786]
+
+Tue Oct 21 01:49:55 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_extract_modeenc): plain rb/wb should set ASCII-8BIT
+	  to the external_encoding.
+
+	* io.c (rb_file_open_internal): ditto.
+
+	* io.c (NEED_WRITECONV): no conversion when the external_encoding
+	  is ASCII-8BIT.
+
+	* io.c (do_writeconv): skip ASCII-8BIT.
+
+Tue Oct 21 00:51:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_ascii8bit_binmode): renamed from rb_io_binmode.
+	  (rb_io_binmode): don't change encoding conversion.
+	  (rb_io_binmode_m): call rb_io_ascii8bit_binmode.
+
+Mon Oct 20 21:19:00 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (read_multipart): file's encoding is ascii-8bit
+	  from file field of multipart form.
+
+Mon Oct 20 20:16:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/zlib/zlib.c (Init_zlib): add getbyte as an alias to getc.
+	  [ruby-dev:36801]
+
+Mon Oct 20 17:31:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (exc_equal): should not compare recursively.
+	  [ruby-dev:36796]
+
+Mon Oct 20 16:48:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/ruby.h (ExportStringValue): new macro to convert
+	  string in internal encoding to external to export.
+
+	* string.c (rb_str_export): new function to do conversion to
+	  external encoding.
+
+	* ext/sdbm/init.c: encoding conversion support.
+
+	* ext/dbm/dbm.c: ditto.
+
+Mon Oct 20 15:42:02 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_locale_str_new): new function to convert string
+	  from locale to internal encoding.
+
+Mon Oct 20 09:47:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/runner.rb: search srcdir/test/arg at first to find test/ruby
+	  directory when "ruby" is specified.
+
+Mon Oct 20 08:20:01 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: support Encoding.default_internal.
+
+	* ext/tk/tcltklib.c: ditto.
+
+	* ext/tk/extconf.rb: improve the strategy for searching Tcl/Tk headers.
+
+Mon Oct 20 00:57:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit.rb (Test::Unit.setup_argv): call given block for
+	  filtering files.
+
+	* test/runner.rb: search srcdir/test/arg, srcdir/arg.
+
+	* bin/testrb: show usage if no files given.
+
+Sun Oct 19 21:19:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (srcs): removed ID_H_TARGET.
+
+	* tool/generic_erb.rb: always overwrites if no if-change option.
+	  sends the result to stdout if no output option.
+
+	* template/id.h.tmpl: shows which token differs.
+
+Sun Oct 19 18:15:15 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/single_byte.trans: adding WINDOWS-wwww encodings
+	  (wwww = 874/1250/1251/1253/1254/1255/1256/1257)
+	  (contributed by Yoshihiro Kambayashi)
+
+	* enc/trans/windows-wwww-tbl.rb: 8 new files
+	  (contributed by Yoshihiro Kambayashi)
+
+	* test/ruby/test_transcode.rb: added test_windows_wwww
+	  (contributed by Yoshihiro Kambayashi)
+
+Sun Oct 19 07:37:13 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (log2,cbrt,frexp,ldexp,hypot,erf,erfc,gamma,lgamma):
+	  should be also provided as module function.  [ruby-dev:36787]
+
+Sun Oct 19 07:25:08 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* keywords, lex.c.src, opt_insn_unif.def, opt_operand.def: moved rarely changed
+	  input files for code generators into defs/ directory.
+
+	* Makefile.in (lex.c): followed keywords and lex.c.src.
+
+	* common.mk (parser.o): followed keywords.
+	  (INSNS): followed opt_*.def
+
+	* tools/instruction.rb: followed opt_*.def.
+
+Sun Oct 19 12:28:01 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* test/csv/*:  Renamed tc_* files to test_* to fit in within Ruby's
+	  testing process.
+
+Sun Oct 19 00:15:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit.rb (assert_nothing_thrown): don't intern.
+
+Sat Oct 18 14:40:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (num_sadded): remove newly defined singleton method
+	  that should not exist after exception handling.  [ruby-dev:36569]
+
+Sat Oct 18 14:27:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (opt_block_arg): allow trailing comma after usual
+	  arguments.  not after block argument.
+
+Sat Oct 18 13:30:53 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_external_str_new): a new function to convert from
+	  external encoding to internal encoding.  if something went
+	  wrong, it returns a string with the external encoding.
+
+	* string.c (rb_external_str_new_with_enc): same as above besides
+	  you can specify the source encoding.
+
+	* ruby.c (ruby_set_argv): use rb_external_str_new()
+
+	* ruby.c (set_arg0, ruby_script): ditto.
+
+Sat Oct 18 04:08:18 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#initialize): now Tempfile.new takes
+	  keyword arguments to open().  [ruby-dev:36756]
+
+Sat Oct 18 17:10:10 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* tool/transcode-tblgen.rb: added set_valid_byte_pattern
+	  to reduce coupling between table generation script and
+	  specific encodings.
+
+	* enc/trans/single_byte.trans: using set_valid_byte_pattern
+
+Sat Oct 18 16:39:39 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c: in transcode_search_path, eliminated a warning
+	  on cygwin about pathlen potentially not being initialized
+
+Sat Oct 18 13:40:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit.rb (assert_nothing_thrown): implemented.
+
+Sat Oct 18 01:33:22 2008  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* common.mk: use added options for generic_erb.rb.
+
+Sat Oct 18 00:47:51 2008  wanabe  <s.wanabe@gmail.com>
+
+	* array.c: don't recycle shared-array while sort!.
+
+Fri Oct 17 22:04:38 2008  wanabe  <s.wanabe@gmail.com>
+
+	* array.c: recycle shared-array when it isn't referenced.
+
+Fri Oct 17 20:29:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_initialize_m): specify ARG_ENCODING_NONE instead of
+	  ARG_ENCODING_FIXED for Regexp.new("", nil, "n").  [ruby-dev:36761]
+
+	* test/ruby/test_regexp.rb (TestRegexp#test_initialize): test
+	  updated.
+
+Fri Oct 17 19:46:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/ifchange, win32/ifchange.bat: --timestamp option added.
+
+	* tool/generic_erb.rb: --timestamp, --output and --if-change options
+	  added.
+
+Fri Oct 17 18:33:12 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* array.c (ary_resize_capa): renamed RESIZE_CAPA.
+
+Fri Oct 17 18:16:51 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_sqrt_bigdecimal):
+	  test updated.  a patch from TAKANO Mitsuhiro <takano32 at jus.or.jp> in
+	  [ruby-dev:36736].
+
+Fri Oct 17 18:14:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (uncommon.mk): ignores failures at loading.
+
+	* test/rubygems/gemutilities.rb (RubyGemTestCase#build_rake_in):
+	  get rid of interference with other tests.
+
+Fri Oct 17 18:06:39 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* transcode.c (str_transcode0): String#encode without argument now
+	  behave as if :undef => :replace, :invalid => :replace specified.
+
+	* transcode.c (rb_econv_prepare_opts): should preserve options in
+	  any case.
+
+Fri Oct 17 11:48:18 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/socket/socket.c (init_sock): sockets should be binmode.
+
+	* test/socket/test_tcp.rb (test_encoding): ditto.
+
+Fri Oct 17 10:26:13 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/insns2vm.rb: remove -Kn option in shebang line because it's
+	  inconsistent with -Ks commandline option in common.mk.
+
+Fri Oct 17 01:26:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_restartable0): in_pos and out_pos never be
+	  NULL, now.
+
+Fri Oct 17 00:39:19 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/test/unit.rb: sort the order of executing tests.
+
+Fri Oct 17 00:24:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_binmode): reset encoding conversion.
+
+Fri Oct 17 00:16:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (rb_getc, rb_io_fread, rb_io_fwrite, rb_read_pending):
+	  remove deprecated functions which do not work.  [ruby-dev:36697]
+
+Thu Oct 16 23:48:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (TESTS): removed to re-enable test for minitest.
+
+Thu Oct 16 23:37:11 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_set_sequence): fix memory leak.
+
+Thu Oct 16 23:27:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit.rb (assert_equal): avoid incompatible character
+	  encodings error.
+
+Thu Oct 16 23:23:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit.rb (assert_equal): failure message changed.
+
+Thu Oct 16 22:52:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit: removed test/unit.
+
+	* lib/test/unit.rb: new compatibility layer using minitest.
+
+	* bin/testrb: use above.
+
+	* test/runner.rb: ditto.
+
+Thu Oct 16 14:34:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, enc/depend (enc, trans): targets for sources.
+
+Thu Oct 16 14:30:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (lib): installs all files other than README etc.
+
+Thu Oct 16 09:48:03 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_initialize_m): changed the message to clarify the
+	  third option argument is now ignored.  [ruby-dev:36753]
+
+Thu Oct 16 08:14:39 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (proc_options): no warning when default_external already
+	  set by -E.
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_rubyopt):
+	  put -K after -E to set script encoding.
+
+Thu Oct 16 06:20:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_transcode.rb (TestTranscode#test_errors):
+	  String#encode now works without any argument. [ruby-dev:36740]
+
+Wed Oct 15 23:48:22 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* .gdbinit (rp): REGEXP handling fixed.
+
+	* string.c (rb_str_rindex_m): need not to call rb_enc_check on
+	  regexp.
+
+	* re.c (unescape_escaped_nonascii): try ASCII-8BIT encoding for
+	  broken strings.
+
+Wed Oct 15 23:11:10 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (DelegateClass): restored 1.8 behavior for
+	  DelegateClass as well.  [ruby-dev:36739]
+
+Wed Oct 15 22:19:14 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* array.c (RESIZE_CAPA): check whether len is longer than capacity.
+
+	* array.c (rb_ary_compact_bang): resize ary before changing capacity.
+	  fixed [ruby-dev:36704]
+
+Wed Oct 15 16:57:30 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* version.h (RUBY_DESCRIPTION): remove unnecessary space.
+
+Tue Oct 14 23:18:15 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator): simplified and restored 1.8
+	  behavior.  [ruby-dev:35986]
+
+Tue Oct 14 21:50:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (TESTS): skips minitest by default, since it interferes
+	  with test/unit.
+
+Tue Oct 14 21:36:29 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (load_conv_function51932): check
+	  having IMultiLanguage2 or IMultiLanguage. [ruby-dev:36716]
+
+	* ext/win32ole/extconf.rb: ditto.
+
+Tue Oct 14 20:45:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (Init_IO): $FILENAME and $* must be read-only.  [ruby-dev:36698]
+
+	* variable.c (*_getter, *_setter, *_marker): made public.
+
+	* include/ruby/ruby.h (rb_gvar_*_{getter,setter,marker}): declared.
+
+Tue Oct 14 20:32:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/single_byte.trans (transcode_tblgen_singlebyte): renamed
+	  from transcode_tblgen_windows.
+	  (transcode_tblgen_iso8859): use transcode_tblgen_singlebyte.
+
+Tue Oct 14 19:32:14 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (process_options): -U should be allowed in RUBYOPT
+	  environment variable.  [ruby-dev:36720]
+
+Tue Oct 14 14:11:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): -U can be followed by other options.
+
+Tue Oct 14 13:30:30 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/single_byte.trans: added windows-1252
+
+	* enc/trans/windows-1252-tbl.rb: new file
+	  (contributed by Yoshihiro Kambayashi)
+
+	* tool/transcode-tblgen.rb: listed windows-1252 as '1byte'
+
+	* test/ruby/test_transcode.rb: added test_windows_1252
+	  (contributed by Yoshihiro Kambayashi)
+
+Tue Oct 14 12:22:32 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_variable.rb (TestVariable#test_variable): add
+	  a test of [ruby-dev:36698].
+
+Tue Oct 14 11:14:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* transcode.c (str_transcode0): should not raise error when
+	  default_internal is not set.  [ruby-core:19309]
+
+Tue Oct 14 11:14:26 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/net/pop.rb: check for invalid APOP timestamp. (CVE-2007-1558)
+	  [ruby-dev:36631]
+
+	* test/net/pop/test_pop.rb: ditto.
+
+Tue Oct 14 09:39:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (set_internal_encoding_once): check double contradicted
+	  specification of the encoding from command line.
+
+	* ruby.c (set_external_encoding_once): ditto.
+
+Tue Oct 14 08:28:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): allow reserved word to be keyword argument.
+
+Mon Oct 13 19:07:52 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (token_info_pop): show source filename. [ruby-dev:36710]
+
+Mon Oct 13 08:59:08 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (load_conv_function51932): support
+	  CP51932 on cygwin and mingw32. thanks to arton.
+
+	* test/win32ole/test_win32ole.rb (test_cp51932): ditto.
+
+Mon Oct 13 07:54:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (proc_options): add -U command line option to specify
+	  utf-8 as default_internal.
+
+Mon Oct 13 07:42:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* transcode.c (str_transcode0): String#encode() with no encoding
+	  specified transcode the string into Encoding.default_internal.
+	  inspired by [ruby-core:19298].
+
+Sun Oct 12 18:00:18 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (set_ole_codepage, ole_cp2encoding,
+	  ole_wc2mb, ole_vstr2wc, ole_mb2wc): support CP51932 (only mswin32).
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+Sun Oct 12 12:03:38 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, vm_insnhelper.h (ruby_vm_redefined_flag): apply optimization
+	  patch proposed by Paul Brannan.  [ruby-core:19171]
+
+Sun Oct 12 09:46:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime): suppressed warnings on cygwin.
+
+Sat Oct 11 19:49:35 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_prepare): use utf-8 encoding directly.
+
+Sat Oct 11 10:20:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (CLEANINGS): uses escaped form tabs to preventing the
+	  confusion.  [ruby-talk:317345]
+
+Sat Oct 11 08:51:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake.rb (Rake::Application#standard_exception_handling): removed
+	  unnecessary SystemExit, and exits with false instead of system
+	  dependent value.  [ruby-talk:317330]
+
+Sat Oct 11 03:54:05 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: optimize 'for' statement when one variable given.
+
+	* benchmark/bm_loop_for.rb: added.
+
+	* benchmark/bm_loop_times.rb: modified.
+
+Sat Oct 11 12:09:05 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv/csv.rb:  Added support for Encoding::default_internal.
+	* lib/csv/csv.rb:  Switched to new Hash syntax.
+
+Fri Oct 10 22:16:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (comment_at_top): needed for ripper too.
+
+Fri Oct 10 22:07:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby.h: updated macros for headers.  [ruby-core:19275]
+
+Fri Oct 10 19:21:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (magic_comment_encoding): warns when ignored.
+
+	* parse.y (parser_magic_comment): replaces '-' with '_'.
+
+	* parse.y (parser_yylex): allows magic comments indented and the
+	  second line or later.
+
+Fri Oct 10 17:26:50 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/ext/parser/parser.c (JSON_parse_string):
+	  associate encoding.
+
+Fri Oct 10 10:18:21 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/test/*: reverted back to test/unit.
+	* test/test/*: ditto
+	* lib/minitest/*: Imported minitest 1.3.0 r4333.
+
+Fri Oct 10 03:41:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): check if codepoint for %c is valid.
+	  [ruby-dev:36691]
+
+Fri Oct 10 01:55:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_names): minor improvement.
+
+Fri Oct 10 00:21:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (ParseError.filter_backtrace): removes internal
+	  backtrace.
+
+Fri Oct 10 00:11:16 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* transcode.c (rb_econv_add_converter): remove unnecessary NULL check.
+
+Thu Oct  9 23:53:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el (ruby-style-c-mode): buffer may not have file
+	  name.
+
+Thu Oct  9 23:30:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* process.c (run_exec_dup2): fix resource leak.
+
+Thu Oct  9 23:19:02 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* process.c (rb_run_exec_options): fix resource leak.
+
+Thu Oct  9 22:13:58 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_module.rb (test_remove_class_variable): add a test
+	  for Class#remove_class_variable.
+
+	* test/ruby/test_object.rb (test_remove_instance_variable): add a test
+	  for Object#remove_instance_variable.
+
+Thu Oct  9 22:04:38 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* variable.c (rb_mod_remove_cvar): fix r19711.
+
+Thu Oct  9 21:17:50 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/syck/syck.c (syck_lookup_sym): remove reading uninitialized
+	  variable.
+
+Thu Oct  9 16:34:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/ext/generator/extconf.rb: fix target.
+
+	* ext/json/ext/parser/extconf.rb: ditto.
+
+Thu Oct  9 14:37:59 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* include/ruby/ruby.h: embeds the elements of an array into its
+	  struct RArray for # of elements <= 3.
+
+	* array.c: ditto.
+
+	* gc.c (gc_mark_children): following the change of struct RArray.
+
+	* ext/tk/tcltklib.c (ip_ruby_cmp): ditto.
+
+	* parse.y (coverage): ditto.
+
+	* proc.c (curry): ditto.
+
+	* .gdbinit: ditto.
+
+Thu Oct  9 11:29:33 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (Init_Encoding): new instance method Encoding#names,
+	  returns its name and alias names.
+
+	* encoding.c (enc_names): defined for Encoding#names.
+
+	* encoding.c (enc_names_i): defined for enc_names.
+
+Thu Oct  9 08:47:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (rb_thread_wait_fd_rw): should not block by select if
+	  there's only one thread living.  fixed [ruby-dev:36646].
+
+Wed Oct  8 20:59:52 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_init_cp): initialize WIN32OLE.codepage
+	  according to Encoding.default_internal and
+	  Encoding.default_external.
+
+Wed Oct  8 17:02:21 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/pure/generator.rb (utf8_to_json):
+	  process each unpacked Unicode character.
+
+Wed Oct  8 15:00:22 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/pure/parser.rb
+	  (JSON::Pure::Parser#parse_string): force_encoding("UTF-8").
+
+Wed Oct  8 11:56:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: assert_nothing_thrown,
+	  assert_raise, assert_not_equal, assert_no_match, assert_not_nil,
+	  assert_not_same are coming back as alias.
+
+Wed Oct  8 11:18:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (autoload_delete, autoload_file): should not delete
+	  autoload table, since it may be shared with duplicated modules.
+	  [ruby-core:19181]
+
+Wed Oct  8 02:38:28 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_default_internal_encoding): merged a patch from
+	  Michael Selig <michael.selig at fs.com.au> in [ruby-core:18985].
+
+	* io.c (rb_io_ext_int_to_encs): ditto.
+
+	* ruby.c (proc_options): support default internal encoding in -E
+	  option.
+
+Wed Oct  8 00:03:39 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (today,now): should produce own instances.
+	  [ruby-talk:317020]
+
+Tue Oct  7 16:43:51 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/pure/generator.rb (utf8_to_json):
+	  force_encoding before gsub.
+
+Tue Oct  7 16:35:41 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/pure/generator.rb (utf8_to_json):
+	  downcase the result of Unicode escape.
+
+Mon Oct  6 16:44:52 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI::QueryExtension::initialize_query): fix the
+	  condition.
+
+	* test/cgi/test_cgi_core.rb: bug fix encoding.
+	  thanks to TAKANO Mitsuhiro <takano32 at jus.or.jp> .
+
+	* test/cgi/test_cgi_multipart.rb: temporary comment in.
+
+Mon Oct  6 15:22:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_mark_children): ignores T_ZOMBIE.
+
+Mon Oct  6 12:45:20 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_module.rb (TestModule#_wrap_assertion): add
+	  definition.  is this really needed?
+
+	* test/ruby/test_module.rb (TestModule#test_mod_constants): should
+	  not depend on global Module constants.
+
+Mon Oct  6 12:38:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Assertions#assert_nothing_raised):
+	  should take optional message argument.
+
+Mon Oct  6 12:18:23 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_comparable): string comparison should be
+	  transitive.  [ruby-dev:36484]
+
+	* test/ruby/test_m17n_comb.rb (TestM17NComb#test_str_eq): test
+	  updated.
+
+Mon Oct  6 09:00:58 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/bigdecimal/test_bigdecimal.rb (test_sqrt_bigdecimal): test
+	  updated.  a patch from TAKANO Mitsuhiro <takano32 at jus.or.jp>
+	  in [ruby-dev:36669].
+
+Sun Oct  5 23:34:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb ($config_h): now always defines for old libraries.
+
+Sun Oct  5 23:06:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_mod_modfunc): method undefined in included module
+	  may not have nd_body.  [ruby-core:18738]
+
+Sun Oct  5 21:28:58 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/rational.rb: resolved conflicts of aliases.
+
+	* lib/mathn.rb: avoided some warnings.
+
+Sun Oct  5 18:51:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: remove debug code introduced by previous change.
+
+Sun Oct  5 17:49:35 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/open-uri.rb (OpenURI.open_uri): set encoding to strio.
+
+Sun Oct  5 17:39:21 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_set_encoding): defined for
+	  StringIO#set_encoding.
+
+Sun Oct  5 11:24:42 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* eval.c: come back definition of environ.
+
+Sun Oct  5 09:19:49 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/utf8tbl.c: update table.
+
+Sat Oct  4 22:50:14 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* dln.c: Ruby no longer supports VMS.
+
+	* error.c: ditto.
+
+	* eval.c: ditto.
+
+	* eval_intern.h: ditto.
+
+	* include/ruby/defines.h: ditto.
+
+	* include/ruby/ruby.h: ditto.
+
+	* io.c: ditto.
+
+	* process.c: ditto.
+
+	* ruby.c: ditto.
+
+	* vms/config.h: removed.
+
+	* vms/vms.h: ditto.
+
+	* vms/vmsruby_private.c: ditto.
+
+	* vms/vmsruby_private.h: ditto.
+
+Sat Oct  4 22:44:23 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* dln.c: Ruby no longer supports Windows CE.
+
+	* eval.c: ditto.
+
+	* include/ruby/defines.h: ditto.
+
+	* include/ruby/win32.h: ditto.
+
+	* ruby.c: ditto.
+
+	* strftime.c: ditto.
+
+	* win32/Makefile.sub: ditto.
+
+	* win32/win32.c: ditto.
+
+	* ext/tk/extconf.rb: ditto.
+
+	* lib/fileutils.rb: ditto.
+
+	* test/fileutils/test_fileutils.rb: ditto.
+
+	* wince/*: removed.
+
+Sat Oct  4 22:35:06 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* dln.c: Ruby no longer supports MacOS 9 or before.
+
+	* eval.c: ditto.
+
+	* eval_intern.h: ditto.
+
+	* ext/extmk.rb: ditto.
+
+	* ext/tk/sample/tkextlib/treectrl/demo.rb: ditto.
+
+	* ext/tk/stubs.c: ditto.
+
+	* file.c: ditto.
+
+	* hash.c: ditto.
+
+	* include/ruby/defines.h: ditto.
+
+	* ruby.c: ditto.
+
+	* signal.c: ditto.
+
+	* vm_core.h: ditto.
+
+Sat Oct  4 22:30:28 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* dln.c: Ruby no longer supports MS-DOS.
+
+	* ext/sdbm/_sdbm.c: ditto.
+
+	* ext/sdbm/sdbm.h: ditto.
+
+	* gc.c: ditto.
+
+	* hash.c: ditto.
+
+	* include/ruby/defines.h: ditto.
+
+	* include/ruby/util.h: ditto.
+
+	* io.c: ditto.
+
+	* process.c: ditto.
+
+	* ruby.c: ditto.
+
+	* strftime.c: ditto.
+
+	* util.c: ditto.
+
+Sat Oct  4 22:14:15 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* djgpp/GNUmakefile.in: removed. Ruby no longer supports djgpp.
+
+	* djgpp/README.djgpp: ditto.
+
+	* djgpp/config.hin: ditto.
+
+	* djgpp/config.sed: ditto.
+
+	* djgpp/configure.bat: ditto.
+
+	* djgpp/mkver.sed: ditto.
+
+	* ext/Setup.dj: ditto.
+
+	* dln.c: removed djgpp supports.
+
+	* file.c: ditto.
+
+	* gc.c: ditto.
+
+	* io.c: ditto.
+
+	* process.c: ditto.
+
+	* ruby.c: ditto.
+
+	* signal.c: ditto.
+
+	* util.c: ditto.
+
+	* vm_core.h: ditto.
+
+	* lib/fileutils.rb: ditto.
+
+	* lib/mkmf.rb: ditto.
+
+	* ext/socket/socket.c: ditto.
+
+	* test/fileutils/test_fileutils.rb: ditto.
+
+	* test/ruby/test_env.rb: ditto.
+
+	* test/ruby/test_path.rb: ditto.
+
+Sat Oct  4 21:59:39 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* x68/_dtos18.c: removed. Ruby no longer supports human68k.
+
+	* x68/_round.c: ditto.
+
+	* x68/fconvert.c: ditto.
+
+	* x68/select.c: ditto.
+
+	* ext/Setup.x68: ditto.
+
+	* missing/x68.c: ditto.
+
+	* dln.c (dln_find_exe_r): removed human68k supports.
+	  (dln_find_1): ditto.
+
+	* lib/mkmf.rb: ditto.
+
+	* ext/extmk.rb (Init_ext): ditto.
+
+	* ext/socket/socket.c (init_sock): ditto.
+
+	* gc.c (GC_MALLOC_LIMIT): ditto.
+	  (rb_setjmp, rb_jmpbuf): ditto.
+	  (mark_current_machine_context): ditto.
+
+	* include/ruby/defines.h (PATH_ENV): ditto.
+
+	* io.c: ditto.
+
+	* process.c: ditto.
+
+	* ruby.c: ditto.
+
+	* test/ruby/test_env.rb: ditto.
+
+	* test/ruby/test_path.rb: ditto.
+
+	* LEGAL
+
+Sat Oct  4 19:02:36 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: no need to require the "lib/rational.rb".
+
+Sat Oct  4 19:02:59 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_enc_get): don't replicate encoding.
+	  [ruby-dev:36575]
+
+Sat Oct  4 18:47:22 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_external_encoding): defined.
+
+	* ext/stringio/stringio.c (strio_internal_encoding): defined.
+
+	* ext/stringio/stringio.c (Init_stringio): define above.
+
+Sat Oct  4 15:52:17 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/mini/test_mini_test.rb: recovered.
+	  It had been temporarily removed at r19645.
+
+	* test/mini/test_mini_mock.rb: ditto.
+
+	* test/mini/test_mini_spec.rb: ditto.
+
+	* lib/test/**/*: replaced by miniunit.
+	  miniunit had been temporarily reverted at r19643 but
+	  now recovered.
+
+Sat Oct  4 15:33:26 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (gdb-ruby): new target. it runs ruby under controlled by
+	  gdb.
+
+	* common.mk (dist): accepts RELNAME variable as the second argument of
+	  make-snapshot.rb.
+
+Sat Oct  4 12:17:46 2008  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/irb/ext/save-history.rb: change load_history using File.expand_path. see [ruby-dev:36660]. Thanks Kouhei Sutou.
+	* lib/irb/irb/context.rb: convert string Symbol of instance variable names in IRB:Context#inspect.
+
+Fri Oct  3 22:43:04 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ext/dl/extconf.rb ($distcleanfiles): added callback-?.c into
+	  the distclean list.
+
+Fri Oct  3 19:33:21 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/pure.rb (module JSON): remove Iconv dependency.
+
+	* ext/json/lib/json/pure/generator.rb (utf8_to_json): process as
+	  binary and remove Iconv dependency.
+
+	* ext/json/lib/json/pure/parser.rb (parse_string): ditto.
+
+Fri Oct  3 14:04:05 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{configure.bat,setup.mak,Makefile.sub): add --program-prefix
+	  configure option support.
+
+Thu Oct  2 21:22:43 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI::QueryExtension): delete MorphingBody
+	  and replace like as 1.8's in multipart reading.
+	  see [ruby-dev:36443],
+	  reference from CGIAlt http://cgialt.rubyforge.org/
+
+	* test/cgi/test_cgi_multipart.rb : fixed multipart test.
+
+Thu Oct  2 20:46:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_sub_bang): fix coderange.
+
+Wed Oct  1 22:54:09 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rexml/parsers/lightparser.rb, lib/debug.rb: remove unused when
+	  clauses.
+
+Wed Oct  1 22:43:59 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/xmlrpc, lib/rexml, test/ruby/test_array.rb,
+	  test/ruby/test_unicode_escape.rb, test/scanf/test_scanf.rb,
+	  test/rss/rss-assertions.rb: fix indentation to remove warning.
+
+Wed Oct  1 21:36:35 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_call_finalizer_at_exit): finalize deferred IO and Data.
+
+	* gc.c (rb_gc_call_finalizer_at_exit): defers IO finalization.
+	  [ruby-dev:36646]
+
+Wed Oct  1 19:11:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (marshal_dump): fixed for check_dump_arg.
+
+	* marshal.c (marshal_dump): initializes dump_arg before any funcall.
+	  [ruby-dev:36648]
+
+Wed Oct  1 00:54:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_split_m): should use rb_str_subseq() which use
+	  offset and length by bytes.  [ruby-dev:36641]
+
+Wed Oct  1 00:29:48 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_settracefunc.rb (test_call, test_class): line number
+	  was changed at r19592.
+
+Wed Oct  1 00:20:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (check_exec_redirect): don't depend on rb_stdout and
+	  rb_stderr.
+
+Tue Sep 30 23:06:58 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/mini/test_mini_test.rb: removed.
+	  Temporarily removed the tests for miniunit because of
+	  they depend on the miniunit version of test/unit,
+	  where test/unit was temporarily reverted to the state
+	  before importing miniunit.
+
+	* test/mini/test_mini_mock.rb: ditto.
+
+	* test/mini/test_mini_spec.rb: ditto.
+
+Tue Sep 30 23:04:30 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/test/**/*: recovered.
+	  Temporarily reverts the changes of lib/test/unit/* in
+	  r19502 and r19501, in order to release 1.9.0-5.
+
+Tue Sep 30 23:00:05 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/test/unit.rb: removed.
+	  Temporarily reverts the changes of lib/test/unit/* in
+	  r19543, r19534 and r19503, in order to release 1.9.0-5.
+
+	* lib/test/unit/assertions.rb: ditto.
+
+	* lib/test/unit/deprecate.rb: ditto.
+
+	* lib/test/unit/error.rb: ditto.
+
+	* lib/test/unit/testcase.rb: ditto.
+
+Tue Sep 30 20:17:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_extract_modeenc): should honor value of the
+	  textmode: and binmode: options.
+
+Tue Sep 30 17:04:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/bigdecimal/test_bigdecimal.rb (test_sqrt_bigdecimal): wrong
+	  precision specification.  a patch from TAKANO Mitsuhiro
+	  <takano32 at jus.or.jp> in [ruby-dev:36634].
+
+Tue Sep 30 16:53:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt): returns non zero.  [ruby-dev:36633]
+
+Tue Sep 30 16:52:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): add %<name>s style named argument
+	  support.  a patch from Yusuke Endoh <mame at tsg.ne.jp>
+	  in [ruby-dev:35851].
+
+	* sprintf.c (rb_str_format): add gettext style named format
+	  (%{name}) support.  inspired by [ruby-dev:35852].
+
+	* sprintf.c (GETNAMEARG): should raise KeyError exception when no
+	  named argument found.
+
+	* hash.c (rb_hash_fetch): export fetch function.
+
+Tue Sep 30 13:19:01 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/test_gem_installer.rb: remove extra call to setup to
+	  fix ENV['HOME'] pollution.
+
+Tue Sep 30 13:04:11 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/mini/test.rb: Updated to 1.3.0 r4257.
+	* test/mini/*: added from r4257.
+
+Tue Sep 30 07:46:07 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rdoc/test_rdoc_markup_attribute_manager.rb: Test with clean
+	  SPECIALS.
+	* test/rubygems/gemutilities.rb: Restore ENV['HOME'] after test.
+	* test/runner.rb: Pass -n argument down to miniunit.
+
+Tue Sep 30 01:02:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (token_info_push, token_info_pop): do nothing for evaled
+	  source.
+
+Tue Sep 30 00:47:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: constified.
+
+Mon Sep 29 23:32:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_subseq): don't use rb_str_drop_bytes for short
+	  string.
+
+Mon Sep 29 22:54:29 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/mathn.rb: added String#to_[rc].  moved def_canon.
+
+Mon Sep 29 20:41:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_subseq): use rb_str_drop_bytes if possible.
+
+Mon Sep 29 20:22:20 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/*.rb: imported additional tests and some adjustments.
+
+Mon Sep 29 20:13:05 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fev_initialize): initialization
+	  handler instance variable.
+
+	* test/win32ole/test_win32ole_typelib.rb (test_s_typelibs):
+	  fix the warning of shadowing outer local variable.
+
+Mon Sep 29 19:59:00 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (rescue_callback): use rb_write_error
+	  instead of fprintf.
+
+Mon Sep 29 01:51:54 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/rdoc/test_rdoc_info_formatting.rb (teardown): delete tmpdir for
+	  test.
+
+Mon Sep 29 00:30:20 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/html.rb (CGI::HtmlExtension::{radio_group, checkbox_group}):
+	  can specify the false as checked or selected value.
+	  [ruby-core:18306], [ruby-core:18307]
+
+	* test/cgi/test_cgi_tag_helper.rb: add a test.
+
+Sun Sep 28 23:08:38 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/html.rb (CGI::HtmlExtension::popup_menu):
+	  fix the calls #bytesize on array parameters. [ruby-core:18919]
+
+Sun Sep 28 22:44:26 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI::parse): ignore illegal query params.
+	  [ruby-dev:36618]
+
+Sun Sep 28 22:05:07 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_set_sequence): check duplicated when clause.
+	  [ruby-dev:36616]
+
+Sun Sep 28 19:04:03 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/*.rb: imported a date test suite ruby 1.9 limited ed.
+
+Sun Sep 28 18:40:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mini/test.rb (MINI_DIR): loaded path is expanded.
+
+	* lib/mini/test.rb (assert_match): get rid of confusion of
+	  ruby-mode.el.
+
+Sun Sep 28 12:36:09 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_exec.c (vm_exec_core): add an UNLIKELY() hint.
+
+	* vm_insnhelper.h (BASIC_OP_UNREDEFINED_P): add a LIKELY() hint.
+
+Sun Sep 28 12:04:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_new): sets empty string to coderange 7bit.
+	  [ruby-core:18993]
+
+Sun Sep 28 09:39:48 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/mathn.rb ({Fixnum,Bignum,Float}#**): may produce complex
+	  value.
+
+Sun Sep 28 09:05:53 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv/csv.rb:  Worked around some minor encoding changes in Ruby
+	  pointed out by Nobu.
+
+Sun Sep 28 08:37:12 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/mathn.rb: a hack to provide canonicalization.  This must be
+	  temporary, but this seems to be not bad for the time being.
+
+Sat Sep 27 06:22:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake.rb (Module#rake_extension, String#ext, String#pathmap): use
+	  built-in methods.
+
+	* lib/rake.rb (String#pathmap): fixed typo in rdoc.
+
+Sat Sep 27 04:28:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rmext): preceding dots are not a part of extension.
+
+	* file.c (rb_file_s_extname): ditto.
+
+Sat Sep 27 03:15:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_jump_tag_but_local_jump): exc is not used if val is nil.
+
+Sat Sep 27 03:12:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (host_str): numeric address should be unsigned.
+	  [ruby-core:18971]
+
+Fri Sep 26 22:46:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary, brace_block): fix for line number.
+
+	* proc.c (rb_proc_location, rb_method_location): new methods
+	  {Proc,Method,UnboundMethod}#source_location.  [ruby-core:18452]
+
+Fri Sep 26 21:36:33 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (RUBY_VM_CHECK_INTS_TH): add an UNLIKELY hint.
+
+Fri Sep 26 19:33:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_str_encode): renamed from
+	  rb_str_transcode.  [ruby-dev:36593]
+	  (rb_econv_has_convpath_p): renamed from rb_transcode_convertible.
+
+	* transcode.c: follow the renaming.
+
+	* io.c: ditto.
+
+Fri Sep 26 19:01:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/bigdecimal/test_bigdecimal.rb (test_inspect): fixed CPU bit
+	  size dependent test.  pointed out by TAKANO Mitsuhiro <takano32 AT
+	  jus.or.jp>.
+
+Fri Sep 26 18:57:33 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/pathname/test_pathname.rb: Fix use of deprecated methods.
+
+Fri Sep 26 18:39:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (dist): nothing is needed to run BASERUBY.
+
+	* tool/make-snapshot: new option -archname.
+
+Fri Sep 26 18:24:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el: safe customizable variables.
+
+Fri Sep 26 17:37:18 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* include/ruby/ruby.h (rb_mPrecision): removed.
+	  Precision module was temporarily removed at r19430.
+
+Fri Sep 26 17:25:10 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (dist): Changed its dependencies and action.
+
+Fri Sep 26 17:07:04 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* error.c (Init_syserr): moved to the template.
+	  (errno_missing): removed. fixed [ruby-dev:35958].
+
+	* defs/known_errors.def: added. extracted from Init_syserr.
+
+	* templates/known_errors.inc.tmpl: added. Template for Init_syserr.
+	  c.f. [ruby-dev:35958].
+
+	* tools/generic_erb.rb: added. general purpose mapper which maps
+	  data+template into source code.
+
+	* common.mk (error.$(OBJEXT), incs, known_errors.inc):
+	  Fixed dependencies.
+
+Fri Sep 26 17:02:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thlist_signal): clears the woken thread if nothing woke.
+
+	* thread.c (rb_barrier_wait): achieves the lock if no thread was
+	  waiting yet.
+
+Fri Sep 26 12:04:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/curses.c: should include <ruby/io.h>.
+
+	* ext/io/wait/wait.c: ditto.
+
+	* ext/openssl/ossl.h: ditto.
+
+Fri Sep 26 11:34:08 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* variable.c (rb_define_hooked_variable): cast to get rid of compiler
+	  warnings.
+
+Fri Sep 26 11:05:41 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* transcode.c, include/ruby/encoding.c (rb_transcode_convertible):
+	  new function. checking the existence of converter.
+
+Fri Sep 26 10:35:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (global_variable, struct trace_var): made function
+	  members more strict.
+
+Fri Sep 26 09:55:06 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby{io,sig}.h: moved to include/ruby/backward.
+
+Fri Sep 26 09:47:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/rubysig.h: old macros for backward compatibility.
+
+	* thread.c (BLOCKING_REGION): rewritten using helper functions.
+
+Fri Sep 26 03:03:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* vm_eval.c (Init_vm_eval): define module_eval, class_eval and eval
+	  here for rdoc.
+
+Fri Sep 26 00:40:30 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* range.c (range_inspect): fix SEGV for cyclic range object.
+	  [ruby-core:18835]
+
+Thu Sep 25 23:59:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: dependency updated.
+
+Thu Sep 25 21:46:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_eUndefinedConversionError): renamed from
+	  rb_eConversionUndefinedError.
+	  (rb_eConverterNotFoundError): renamed from rb_eNoConverterError.
+
+Thu Sep 25 21:23:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* pack.c (pack_pack, pack_unpack): 'm0' format (base64) complies with
+	  RFC 4648.  It adds no line feed when encoding, and raise ArgumentError
+	  if the encoded string contains non-alphabet (including CR and LF).
+
+	* lib/base64.rb: added.  This provides encoding/decoding method for
+	  Base64 in standard RFC 2045, Base64 in standard RFC 4648 and ``Base 64
+	  Encoding with URL and Filename SafeAlphabet'' in RFC 4648.
+
+	* test_pack.c, test/base64/test_base64.rb: add tests for above.
+
+Thu Sep 25 21:00:32 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: fix btest-* rules [ruby-dev:36528].
+	  and remove obsolete rules.
+
+Thu Sep 25 20:27:54 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* class.c (rb_make_metaclass): [BUG] Fixed a bus error
+	  on the case for metaclass of a class which includes a
+	  module.
+
+Thu Sep 25 19:34:33 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/*: Update some new tests to use build_rake_in.
+	* test/rubygems/gemutilities.rb: Fix broken merge of rake detection.
+
+Thu Sep 25 19:13:12 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*: Update to RubyGems 1.3.0 r1891.
+
+Thu Sep 25 18:40:42 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* class.c (rb_make_metaclass): Made class of class of class
+	  from Class into its own eigenclass. Now meta^(n)-class
+	  hierarchy regresses infinitely, again.
+	  (This feature was decided on developer-meeting-20080922.)
+
+Thu Sep 25 16:01:07 2008  Jim Weirich  <jim@tardis.local>
+
+	* lib/rake.rb: Update rake source to version 0.8.3.  This
+	  version includes some fixes for running Rake on windows.
+	  (1) better APPDATA/HOMExxx/USERPROFILE integration for
+	  system rakefiles, (2) Better handling of the :ruby command
+	  when installed in directory containing spaces.
+
+Thu Sep 25 11:22:51 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc*: Update to RDoc 2.2.1 r185.
+
+Thu Sep 25 02:08:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_mode_enc): make it static.
+
+Thu Sep 25 01:22:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mini/test.rb (Mini::Assertions#assert_raise): fixed typo.
+
+Wed Sep 24 23:39:00 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* test/cgi/test_cgi_multipart.rb : revert last change.
+	  ( deprecated is assert_raises. )
+	* test/cgi/test_cgi_core.rb      : ditto.
+	* test/cgi/test_cgi_header.rb    : ditto.
+
+Wed Sep 24 23:12:49 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI::new, CGI::{accept_charset,accept_charset=}) :
+	  accept parameters either in a hash,
+	  string as a block. add the encoding validation process.
+	* test/cgi/test_cgi_core.rb : test for query encoding check.
+
+Wed Sep 24 22:58:18 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_rstrip_bang): raise exception when the encoding of
+	  the string is dummy.
+
+	* string.c (rb_str_rstrip_bang): remove nul characters even if the
+	  encoding of the string is not single byte optimizable.
+	  fixed [ruby-core:18844], reported by Michael Selig.
+
+Wed Sep 24 19:01:45 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_rstrip_bang): workaround for VC++8 x64.
+
+Wed Sep 24 17:44:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (Init_Complex), rational.c (Init_Rational): undefines
+	  default #allocate methods which call the allocator.
+
+Wed Sep 24 17:02:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (Init_Complex), rational.c (Init_Rational): ID_ALLOCATOR
+	  differs from :allocate, and invisible in ruby level.
+
+Wed Sep 24 15:58:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_rstrip_bang): removing mixed spaces and nuls at
+	  the end of strings.  [ruby-dev:36497]
+
+Wed Sep 24 15:13:04 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* test/cgi/test_cgi_multipart.rb : test for miniunit.
+	* test/cgi/test_cgi_core.rb      : ditto.
+	* test/cgi/test_cgi_header.rb    : ditto.
+
+Wed Sep 24 14:38:29 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/tmpdir.rb: setup buffer with nul characters instead of spaces.
+	  fixed [ruby-dev:36493]
+
+Wed Sep 24 11:18:33 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_s_binread): offset argument was wrongly passed to
+	  internal IO#read call.  [ruby-core:18810]
+
+Wed Sep 24 07:38:23 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* vm_core.h: ruby/ruby.h should be included at the very first
+	  place in an entire compilation unit, as it includes
+	  ruby/config.h
+
+Wed Sep 24 06:41:24 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/test/*: removed test/unit.
+	* lib/test/*: added miniunit.
+	* lib/mini/*: ditto.
+
+Wed Sep 24 04:45:35 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/node.h, node.h: move node.h from include path.
+	  This change stop to install node.h because of saving ABI
+	  (node.h will be changed. Extensions should not depends on
+	   this file).
+
+	* blockinlining.c, class.c, compile.c, debug.h, enum.c,
+	  gc.c, iseq.c, parse.y, ruby.c, signal.c, variable.c,
+	  vm.c, vm_core.h, vm_dump.c: ditto.
+
+	* ext/ripper/depend: ditto.
+
+Wed Sep 24 04:09:13 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove unused frame values.
+
+Wed Sep 24 04:07:07 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* signal.c (signal_exec): execute rb_eval_cmd() directly.
+
+Wed Sep 24 03:08:58 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/uri/mailto.rb: fix indentation.
+	* lib/uri/generic.rb: ditto.
+	* lib/yaml.rb: ditto.
+	* lib/uri/cgi/core.rb: fix indentation and ambiguous argument warning.
+
+Tue Sep 23 23:39:25 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv/csv.rb:  Improved the idea of whitespace and word characters used
+	  in substitutions during header conversion as suggested by Michael Selig.
+
+Tue Sep 23 21:50:53 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_binwrite): allocate wbuf if nosync.
+
+Tue Sep 23 21:45:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_body): use io_binwrite instead of io_fwrite.
+
+Tue Sep 23 21:31:16 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_rational2.rb: updated.
+
+	* test/ruby/test_complex2.rb: added.
+
+	* test/ruby/test_complexrational.rb: added.
+
+Tue Sep 23 21:26:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_casecmp): make the ordering consistent with
+	  String#<=>.
+
+Tue Sep 23 20:52:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_binwrite): add nosync argument.
+	  (do_writeconv): extracted from io_fwrite.
+	  (io_fwrite): add nosync argument.  use do_writeconv.
+	  (io_write): add nosync argument.
+	  (io_write_m): new function for IO#write.
+	  (rb_p): don't append record separator.
+
+Tue Sep 23 20:24:41 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* signal.c (signal_exec): fix to use rb_proc_call().
+
+Tue Sep 23 20:06:46 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/formatters/pretty.rb (REXML::Formatters::Pretty#wrap):
+	  abandon wrapping if the line contains no space.  [ruby-dev:36045]
+
+Tue Sep 23 19:38:03 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (subtract): check tv_sec. reported by ko1.
+
+Tue Sep 23 19:21:03 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_s_canonicalize_internal): does no apply
+	  canonicalization rule anymore.
+
+	* rational.c (nurat_s_canonicalize_internal(_no_reduce)?): ditto.
+
+	* complex.c: removed class method new! and new.
+
+	* rational.c: ditto.
+
+Tue Sep 23 19:20:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_s_binread): a new method to read binary chunk from a
+	  file.  [ruby-core:18674]
+
+	* io.c (open_key_args): wrong permission specified for read-open.
+
+Tue Sep 23 18:24:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* hash.c (rb_obj_is_proc): declaration moved for rdoc.
+
+Tue Sep 23 18:07:55 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk (io.o): remove dependency for vm_core.h.
+
+Tue Sep 23 17:56:44 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_blocking_region): fix typo in a document.
+
+Tue Sep 23 17:50:35 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_blocking_region): write a document.
+
+Tue Sep 23 17:14:31 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, include/ruby/intern.h (rb_thread_interrupted): added.
+
+	* io.c: use VALUE of thead instead of rb_tread_t to check interrupts.
+
+Tue Sep 23 17:05:14 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c, vm_core.h: move decl of rb_iseq_clone() to class.c.
+
+Tue Sep 23 17:02:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* debug.h (ruby_set_debug_option): declared.
+
+	* main.c: include debug.h.
+
+	* common.mk (main.$(OBJEXT)): dependency updated.
+
+Tue Sep 23 16:59:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (ENVMATCH, ENVNMATCH): reduced same code.
+
+Tue Sep 23 16:55:11 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.h: fix comment and rename macro HAVE_RUBY_MVM_H to
+	  HAVE_RUBY_VM_H.
+
+Tue Sep 23 16:41:31 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/node.h, vm_core.h: move definition of
+	  RUBY_VM_METHOD_NODE to node.h.
+
+	* class.c, common.mk: remove useless inclusion.
+
+	* compile.h, iseq.h, vm_core.h: rename compile.h to iseq.h.
+	  move some definitions from vm_core.h to iseq.h.
+
+	* compile.c, iseq.c, vm.c: ditto.
+
+	* eval.c, compile.c: move some functions for parser
+	  from eval.c to compile.c.
+
+	* eval_intern.h, vm_core.h: move va_init_list() macro to
+	  vm_core.h.
+
+	* iseq.c (rb_iseq_new_top, rb_iseq_first_lineno): added.
+
+	* load.c, ruby.c: use rb_iseq_new_top() instead of
+	  rb_iseq_new() with ISEQ_TYPE_TOP constant directly.
+
+	* proc.c: use rb_iseq_first_lineno() instead of accessing
+	  iseq structure.
+
+Tue Sep 23 16:17:54 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/ripper/eventids2.c (token_to_eventid): supper __ENCODING__
+	  keyword.  [ruby-dev:36478]
+
+Tue Sep 23 15:30:53 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk, hash.c: remove unused inclusion.
+
+Tue Sep 23 09:52:07 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_safe.c, safe.c: rename eval_safe.c to safe.c.
+
+	* common.mk, eval.c, safe.c, inits.c: separate safe.c from eval.c and
+	  make Init_safe().
+
+Tue Sep 23 09:20:00 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: clean up
+	  - remove blockinlining.$(OBJEXT) to built
+	  - make ENCODING_H_INCLDUES variable (include/ruby/encoding.h)
+	  - make VM_CORE_H_INCLUDES variable (vm_core.h)
+	  - simplify rules.
+	  - make depends rule to output depend status using gcc -MM.
+
+	* include/ruby/mvm.h, include/ruby/vm.h: rename mvm.h to vm.h.
+
+	* include/ruby.h: ditto.
+
+	* load.c: add inclusion explicitly.
+
+	* enumerator.c, object.c, parse.y, thread.c, vm_dump.c:
+	  remove useless inclusion.
+
+	* eval_intern.h: cleanup inclusion.
+
+	* vm_core.h: rb_thread_t should be defined in this file.
+
+	* vm_evalbody.c, vm_exec.c: rename vm_evalbody.c to vm_exec.c.
+
+	* vm.h, vm_exec.h: rename vm.h to vm_exec.h.
+
+	* insnhelper.h, vm_insnhelper.h: rename insnhelper.h to vm_insnhelper.h.
+
+	* vm.c, vm_insnhelper.c, vm_insnhelper.h:
+	  - rename vm_eval() to vm_exec_core().
+	  - rename vm_eval_body() to vm_exec().
+	  - cleanup include order.
+
+	* vm_method.c: fix comment.
+
+Tue Sep 23 09:01:15 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/pathname.rb (each_filename): return Enumerator if no block
+	  given.
+
+	* test/pathname/test_pathname.rb: add a test for above.
+
+Tue Sep 23 08:25:56 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (env_aset): allow nil value to remove an entry.
+	  [ruby-dev:36465]
+
+	* test/ruby/test_env.rb (TestEnv#test_aset): test updated.
+
+Tue Sep 23 08:07:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-keyword-end-re): emacs21 support.  a
+	  patch from Hiroshi Moriyama <hiroshi at kvd.biglobe.ne.jp> in
+	  [ruby-dev:36471].
+
+	* misc/ruby-mode.el (ruby-in-ppss-context-p): ditto.
+
+	* misc/ruby-mode.el (ruby-here-doc-end-syntax):
+
+Tue Sep 23 02:07:52 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* test/readline/test_readline_history.rb
+	  (Readline::TestHistory#test_each): checked return value.
+
+Tue Sep 23 02:05:34 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c (Init_readline): used remove_history to
+	  check offset of history if unsupported clear_history.
+
+Mon Sep 22 17:39:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#open): should return a value from the
+	  block (if given).   [ruby-core:18394]
+
+Mon Sep 22 11:17:51 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* include/ruby/encoding.h (enum): ISO C forbids comma at end of
+	  enumerator list
+
+Mon Sep 22 10:15:24 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* LEGAL: updated file names removed and added.  [ruby-dev:36426]
+
+Mon Sep 22 02:39:05 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (get_default_external): add description.
+
+Sun Sep 21 21:52:41 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* math.c (math_log): should check argc.
+
+Sun Sep 21 21:20:24 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: added two macros.
+
+	* rational.c: ditto.
+
+Sun Sep 21 18:06:38 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_s_convert): checks argc.
+
+	* rational.c (nurat_s_convert): ditto.
+
+Sun Sep 21 10:19:04 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_s_canonicalize_internal): checks exactness of
+	  imag only.
+
+	* rational.c (nurat_s_convert): accepts non real value
+	  (Rational(a,b) as a/b).
+
+	* complex.c (nucomp_s_convert): refined.
+
+Sun Sep 21 09:37:57 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv/csv.rb:  Reworked CSV's parser and generator to be m17n.  Data
+	  is now parsed in the Encoding it is in without need for translation.
+	* lib/csv/csv.rb:  Improved inspect() messages for better IRb support.
+	* lib/csv/csv.rb:  Fixed header writing bug reported by Dov Murik.
+	* lib/csv/csv.rb:  Use custom separators in parsing header Strings as
+	  suggested by Shmulik Regev.
+	* lib/csv/csv.rb:  Added a :write_headers option for outputting headers.
+	* lib/csv/csv.rb:  Handle open() calls in binary mode whenever we can to
+	  workaround a Windows issue where line-ending translation can cause an
+	  off-by-one error in seeking back to a non-zero starting position after
+	  auto-discovery for :row_sep as suggested by Robert Battle.
+	* lib/csv/csv.rb:  Improved the parser to fail faster when fed some forms
+	  of invalid CSV that can be detected without reading ahead.
+	* lib/csv/csv.rb:  Added a :field_size_limit option to control CSV's
+	  lookahead and prevent the parser from biting off more data than
+	  it can chew.
+	* lib/csv/csv.rb:  Added readers for CSV attributes:  col_sep(), row_sep(),
+	  quote_char(), field_size_limit(), converters(), unconverted_fields?(),
+	  headers(), return_headers?(), write_headers?(), header_converters(),
+	  skip_blanks?(), and force_quotes?().
+	* lib/csv/csv.rb:  Cleaned up code syntax to be more inline with
+	  Ruby 1.9 than 1.8.
+
+Sun Sep 21 07:43:16 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: an instance method image has been removed and
+	  uses "imag" instead of "image".
+
+	* complex.c: two instance method re and im are removed.
+
+	* rational.c: follows the above changes.
+
+	* include/ruby/ruby.h: ditto.
+
+	* gc.c: ditto.
+
+	* lib/cmath.rb: ditto.
+
+	* lib/mathn.rb: ditto.
+
+	* lib/complex.rb: ditto. and provides some obsolete methods.
+
+Sun Sep 21 02:48:45 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/ext/parser/parse.c: use ruby_xfree().
+
+	* ext/json/ext/parser/parse.rl: ditto.
+
+	* ext/json/ext/parser/unicode.c: ditto.
+
+Sun Sep 21 02:40:20 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: import JSON v 1.1.3.
+
+Sat Sep 20 11:43:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/optparse.rb (summarize): separator "" should output new line.
+
+Sat Sep 20 08:07:34 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y: strings which contain only US-ASCII don't force to have
+	  US-ASCII encoding. [ruby-dev:36400]
+
+Sat Sep 20 07:59:31 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_desc): Regexps of ASCII Compatible encoding may
+	  contain non-ASCII characters. So in that case its encoding
+	  must keep original encoding.
+
+Sat Sep 20 07:44:56 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: prec.c is removed.
+
+Fri Sep 19 22:37:25 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: uses f_(in)?exact_p macro.
+
+	* rational.c: ditto.
+
+	* bignum.c (rb_big_pow): bignum**bignum - should calculate without
+	  rational.
+
+	* lib/complex.rb: should override Math module at most once.
+
+	* lib/mathn.rb: requires 'cmath' directly.
+
+Fri Sep 19 20:48:06 2008  Yuki Sonoda  <yugui@yugui.jp>
+
+	* prec.c: removed. Precision will be redesigned and be back again.
+	  c.f. [ruby-dev:36352].
+
+	* common.mk (COMMON_OBJS): removed prec.o.
+
+	* inits.c (rb_call_inits): removed Init_Precision.
+
+	* numeric.c (Init_Numeric): removed inclusion of Precision.
+	  removed #induced_from from each class.
+
+	* rational.c: ditto.
+
+	* ext/bigdecimal/bigdecimal.c: ditto.
+
+	* lib/rdoc/known_classes.rb: removed the entry for Precision.
+
+	* test/ruby/test_prec.rb: removed.
+
+	* test/ruby/test_integer.rb: removed tests for Precision.
+
+	* test/ruby/test_fixnum.rb: ditto.
+
+	* test/ruby/test_float.rb: ditto.
+
+	* test/ruby/test_rational.rb: ditto.
+
+	* test/ruby/test_complex.rb: ditto.
+
+	* test/bigdecimal/test_bigdecimal.rb: ditto.
+
+Fri Sep 19 19:43:40 2008  Yuki Sonoda  <yugui@yugui.jp>
+
+	* common.mk : Reverts the changeset 18994.
+	  Uses the official repository of rubyspec again.
+
+Fri Sep 19 17:41:56 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: applied OS/2 support patch from Brendan Oakley
+	  <gentux2 at gmail.com> in [ruby-core:18707].
+
+Fri Sep 19 09:29:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dln.c: newer BeOS support.  a patch from Pete Goodeve
+	  <pete.goodeve at computer.org> in [ruby-core:18712].
+
+Fri Sep 19 03:41:25 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: fix memory violation. mentioned by mame [ruby-dev:36373]
+
+Fri Sep 19 01:07:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* regexec.c (string_cmp_ic): add text_end argument.
+	  (slow_search): call enclen with real string end.
+	  (map_search): add text_end argument.
+
+Thu Sep 18 22:54:39 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_to_string): comment out
+	  fragments of unused code.
+
+Thu Sep 18 22:35:03 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpCtoV): 1E1000...000 is interpreted as
+	  Infinity.  [ruby-dev:36159]
+
+	* ext/bigdecimal/bigdecimal.c (VpPower): Infinity ** 1 returns
+	  Infinity instead of NaN.  [ruby-dev:36159]
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests for above.
+
+Thu Sep 18 22:20:08 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (Init_Exception): introduce EncodingError which is a
+	  superclass for all encoding related exception classes,
+	  e.g. Encoding::CompatibilityError.  [ruby-dev:36371]
+
+	* transcode.c (Init_transcode): ditto.
+
+Thu Sep 18 21:57:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_comparable): make ascii8bit string to be
+	  compatible with any other encoding.
+
+	* string.c (rb_str_cmp): use rb_str_comparable() instead of
+	  rb_enc_compatible() since <=> is a comparison anyway.
+
+Thu Sep 18 21:37:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* grapheme cluster implementation reverted.  [ruby-dev:36375]
+
+Thu Sep 18 20:50:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake.rb (Rake::Application#standard_exception_handling):
+	  replaced magic numbers for exit status.
+
+Thu Sep 18 20:37:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::ParseError#set_backtrace): omits
+	  OptionParser internal backtraces unless debug flag is set.
+
+Thu Sep 18 20:02:48 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/uri/common.rb (unescape): skip '%' to make String#hex work
+	  correctly.
+
+Thu Sep 18 19:51:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (big2str_karatsuba): free internal buffer eagerly.
+	  a patch from TOYOFUKU Chikanobu <nobu_toyofuku at nifty.com>
+	  in [ruby-dev:36217].
+
+	* bignum.c (rb_big2str0): ditto.
+
+Thu Sep 18 19:43:04 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* st.c (new_size): raise RuntimeError when st_table is too big.
+	  [ruby-dev:36354]
+
+Thu Sep 18 18:23:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (enc_arg): must take pointer argument to avoid GC
+	  problem.  StringValueCStr modifies the argument and it should be
+	  preserved while the string StringValueCStr returns is used.
+	  Since the string is used by caller, the modified argument should be
+	  hold by caller.  Actually
+	    GC.stress = true
+	    def (o=Object.new).to_str()
+	      "universal"+"_newline"
+	    end
+	    "\u3042".encode(o, "")
+	  causes curious warning:
+	    rb_define_const: invalid name `' for constant
+
+Thu Sep 18 17:32:44 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* transcode.c: add "Error" suffix for Encoding exception classes.
+	  a patch from Tadashi Saito <shiba at mail2.accsnet.ne.jp> in
+	  [ruby-dev:36346].
+
+	* encoding.c (Init_Encoding): rename EncodingCompatibilityError to
+	  Encoding::CompatibilityError.  [ruby-dev:36366]
+
+Thu Sep 18 17:18:35 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* transcode.c (enc_arg): revert last change.  too early exception
+	  raising.
+
+	* transcode.c (enc_arg): need not to take pointer argument.
+
+Sun Sep 18 02:22:15 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests to achieve over 90%
+	  test coverage of bigdecimal.
+
+Wed Sep 17 21:50:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_casecmp): don't use rb_enc_codepoint.
+
+Wed Sep 17 19:55:33 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_s_convert): accepts complex
+	  value (Complex(a,b) as a+bi).
+
+Wed Sep 17 19:16:47 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_io.rb (TestIO#test_dup): avoid infinite loop.
+	  [ruby-dev:36326]
+
+	* test/ruby/test_io.rb (TestIO#test_dup): remove needless open.
+	  [ruby-dev:35957]
+
+Wed Sep 17 17:44:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_extract_modeenc): process option mode: and perm: as
+	  well.
+
+	* io.c (open_key_args): move mode: and perm: related code to
+	  rb_io_extract_modeenc().
+
+Wed Sep 17 13:42:59 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/html.rb (checkbox_group,radio_group): bug fix
+	  use size instead of	bytesize.
+
+	* test/cgi/test_cgi_tag_helper.rb: test for checkbox_group,radio_group.
+
+Wed Sep 17 06:58:31 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c: provides predicate real? instead of scalar?.
+
+	* complex.c: follows the above change.
+
+	* lib/cmath.rb: ditto.
+
+Wed Sep 17 01:56:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io_m17n.rb: use __FILE__ instead of /dev/null.
+	  [ruby-dev:36327]
+
+Wed Sep 17 01:33:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h (OnigEncodingTypeST): add precise_ret
+	  argument for mbc_to_code.
+	  (ONIGENC_MBC_TO_CODE): provide NULL for precise_ret.
+	  (ONIGENC_MBC_PRECISE_CODEPOINT): defined.
+
+	* include/ruby/encoding.h (rb_enc_mbc_precise_codepoint): defined.
+
+	* regenc.h (onigenc_single_byte_mbc_to_code): precise_ret argument
+	  added.
+	  (onigenc_mbn_mbc_to_code): ditto.
+
+	* regenc.c (onigenc_single_byte_mbc_to_code): precise_ret argument
+	  added.
+	  (onigenc_mbn_mbc_to_code): ditto.
+
+	* string.c (count_utf8_lead_bytes_with_word): removed.
+	  (str_utf8_nth): removed.
+	  (str_utf8_offset): removed.
+	  (str_strlen): UTF-8 codepoint oriented optimization removed.
+	  (rb_str_substr): ditto.
+	  (enc_succ_char): use rb_enc_mbc_precise_codepoint.
+	  (enc_pred_char): ditto.
+	  (rb_str_succ): ditto.
+
+	* encoding.c (rb_enc_ascget): check length with
+	  rb_enc_mbc_precise_codepoint.
+	  (rb_enc_codepoint): use rb_enc_mbc_precise_codepoint.
+
+	* regexec.c (string_cmp_ic): add text_end argument.
+	  (match_at): check end of character after exact string matches.
+
+	* enc/utf_8.c (grapheme_table): defined for extended grapheme cluster
+	  boundary.
+	  (grapheme_cmp): defined.
+	  (get_grapheme_properties): defined.
+	  (grapheme_boundary_p): defined.
+	  (MAX_BYTES_LENGTH): defined.
+	  (comb_char_enc_len): defined.
+	  (mbc_to_code0): extracted from mbc_to_code.
+	  (mbc_to_code): use mbc_to_code0.
+	  (left_adjust_combchar_head): defined.
+	  (utf_8): use a extended grapheme cluster as a unit.
+
+	* enc/unicode.c (onigenc_unicode_mbc_case_fold): use
+	  ONIGENC_MBC_PRECISE_CODEPOINT to extract codepoints.
+	  (onigenc_unicode_get_case_fold_codes_by_str): ditto.
+
+	* enc/euc_jp.c (mbc_to_code): follow mbc_to_code field change.
+	  use onigenc_mbn_mbc_to_code.
+
+	* enc/shift_jis.c (mbc_to_code): ditto.
+
+	* enc/emacs_mule.c (mbc_to_code): ditto.
+
+	* enc/gbk.c (gbk_mbc_to_code): follow mbc_to_code field and
+	  onigenc_mbn_mbc_to_code change.
+
+	* enc/cp949.c (cp949_mbc_to_code): ditto.
+
+	* enc/big5.c (big5_mbc_to_code): ditto.
+
+	* enc/euc_tw.c (euctw_mbc_to_code): ditto.
+
+	* enc/euc_kr.c (euckr_mbc_to_code): ditto.
+
+	* enc/gb18030.c (gb18030_mbc_to_code): ditto.
+
+	* enc/utf_32be.c (utf32be_mbc_to_code): follow mbc_to_code field
+	  change.
+
+	* enc/utf_16be.c (utf16be_mbc_to_code): ditto.
+
+	* enc/utf_32le.c (utf32le_mbc_to_code): ditto.
+
+	* enc/utf_16le.c (utf16le_mbc_to_code): ditto.
+
+Wed Sep 17 01:17:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (enc_arg): raise exception when unknown encoding is
+	  given.
+
+Tue Sep 16 22:23:24 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI#header): performance improvement.
+	  From CGIAlt http://cgialt.rubyforge.org/
+
+	* test/cgi/test_cgi_header.rb: exception class fixed.
+
+Tue Sep 16 22:21:33 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_concat): fix rdoc. (codepoint is integer)
+
+	* string.c (rb_str_each_codepoint): use UINT2NUM.
+
+Tue Sep 16 21:48:55 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_desc): Regexp#inspect should be US-ASCII.
+
+Tue Sep 16 21:33:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_crypt): orthodox crypt() sees only first two bytes
+	  of salt.
+
+Tue Sep 16 19:18:40 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_marshal_{dump,load}): preserve instance
+	  variables.
+
+	* rational.c (nurat_marshal_{dump,load}): ditto.
+
+Tue Sep 16 18:28:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_gets_m): rdoc updated.  limit counts in bytes.
+	  [ruby-core:18617]
+
+	* io.c (rb_io_readlines, rb_f_gets): ditto.
+
+	* io.c (rb_io_readbyte): rdoc updated.  [ruby-core:18617]
+
+Tue Sep 16 11:55:16 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_each_codepoint): add new methods, #codepoints
+	  and #each_codepoint.  a patch from Michael Selig
+	  <michael.selig at fs.com.au> in [ruby-core:18532].
+
+Tue Sep 16 11:24:44 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/mkconstants.rb: add new constants.  a patch from
+	  IWAMURO Motonori <deenheart+ruby at gmail.com> in [ruby-dev:36309].
+
+Tue Sep 16 09:20:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#test_dup): exception should be
+	  raised from IO#dup.
+
+Tue Sep 16 08:35:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#test_dup): this test might fail if
+	  there are any garbage IO objects.
+
+Tue Sep 16 02:02:56 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_buf_cat_ascii): codepoint is unsigned int.
+
+	* string.c (rb_str_concat): ditto.
+
+	* string.c (str_cat_char): ditto.
+
+	* string.c (prefix_escape): ditto.
+
+Tue Sep 16 01:47:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (single_byte_optimizable): make function inline.  based
+	  on a patch from Michael Selig <michael.selig at fs.com.au> in
+	  [ruby-core:18532].
+
+	* string.c (str_modify_keep_cr): new function act as
+	  rb_str_modify(), but don't clear coderange
+
+	* string.c (rb_str_casecmp): specialized for single byte strings.
+
+	* string.c (rb_str_splice): preserve coderange.
+
+	* string.c (rb_str_slice_bang, rb_str_reverse_bang,
+	  rb_str_upcase_bang, rb_str_downcase_bang, tr_trans,
+	  rb_str_capitalize_bang, rb_str_swapcase_bang,
+	  rb_str_delete_bang, rb_str_chop_bang, rb_str_chomp_bang,
+	  rb_str_lstrip_bang, rb_str_rstrip_bang): ditto.
+
+	* string.c (rb_str_clear): preset coderange.
+
+	* string.c (rb_str_split_m): specialized for splitting with a
+	  string.
+
+Tue Sep 16 00:57:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_quote): use rb_enc_mbcput to generate ASCII
+	  incompatible characters properly.
+
+Tue Sep 16 00:51:59 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (Encoding#base_encoding): removed. [ruby-dev:36270]
+
+Tue Sep 16 00:40:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (STR1_LENGTH): defined.
+	  (makeSTR1LEN): defined.
+
+	* tool/transcode-tblgen.rb: use makeSTR1LEN.  generate STR1 for 4 to
+	  259 bytes.
+
+	* transcode.c (rb_transcoding): new field: output_index.
+	  (transcode_restartable0): use STR1_LENGTH.
+
+Mon Sep 15 23:52:45 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, include/ruby/ruby.h: rename rb_register_mark_object()
+	  to rb_gc_register_mark_object().
+
+	* eval.c, vm.c: initialize vm->mark_object_ary at
+	  Init_top_self().
+
+	* bignum.c, complex.c, encoding.c, ext/win32ole/win32ole.c,
+	  io.c, load.c, marshal.c, rational.c, ruby.c, vm.c:
+	  use rb_gc_register_mark_object() instead of
+	  rb_global_variable() or rb_gc_register_address().
+
+Mon Sep 15 23:37:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_opts): don't use to_sym.
+
+Mon Sep 15 23:28:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_get_index): don't return -1.
+
+	* io.c (rb_scan_open_args): call FilePathValue before encoding
+	  conversion.
+
+Mon Sep 15 22:11:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_squeeze_bang): specialized for 7bit characters in
+	  ASCII compatible strings.
+
+	* string.c (rb_str_count): ditto.
+
+	* string.c (tr_trans): preserve 7bit/valid coderange flag.
+
+	* string.c (rb_str_squeeze_bang): preserve previous coderange value.
+
+	* string.c (rb_str_lstrip_bang): ditto.
+
+	* string.c (rb_str_rstrip_bang): ditto.
+
+	* encoding.c (rb_default_external_encoding): preserve
+	  default_external_encoding in a static variable.
+
+	* string.c (single_byte_optimizable): check coderange first, to
+	  reduce number of calling rb_enc_from_index().
+
+Mon Sep 15 20:57:00 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/matrix.rb (Matrix#eql?): fixed [ruby-dev:36298].
+	  Reported by an anonymous user.
+
+	* lib/matrix.rb (Vector#eql?): ditto.
+
+	* (Matrix#compare_by_row_vectors): takes comparison
+	  strategy as an optional parameter.
+
+	* (Vector#compare_by): ditto.
+
+Mon Sep 15 14:34:32 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (RUBY_MAX_CHAR_LEN): defined.
+
+	* encoding.c (str_cat_char): use RUBY_MAX_CHAR_LEN.
+
+Mon Sep 15 13:53:33 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): accept other than strings and
+	  regexps. [ruby-core:18595]
+
+	* encoding.c (rb_enc_get_index): works files and encodings.
+
+Mon Sep 15 13:17:21 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_eql_p): new.
+
+	* complex.c (nucomp_hash): should use hash values of the elements.
+
+	* rational.c (nurat_hash): ditto.
+
+	* hash.c (rb_any_cmp): removed an unused variable.
+
+Mon Sep 15 11:11:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoder): resetsize_func and resetstate_func
+	  also returns ssize_t.
+
+	* enc/trans/iso2022.trans: follow the type change.
+
+Mon Sep 15 10:28:10 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* test/matrix/test_matrix.rb (setup): typo.
+	  (test_equality): misdefinition of the expected working.
+	  Reported by an anonymous user.
+	  (test_hash): added.
+
+	* test/matrix/test_vector.rb: ditto.
+
+Mon Sep 15 03:33:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h: return output functions ssize_t.
+
+	* transcode.c (transcode_restartable0): don't need to cast the result
+	  of output functions.
+
+	* enc/trans/newline.trans: follow the type change.
+
+	* enc/trans/escape.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+	* enc/trans/iso2022.trans: ditto.
+
+	* enc/trans/japanese.trans: ditto.
+
+Mon Sep 15 03:04:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h: output function takes output buffer size.
+
+	* transcode.c: give output buffer size for output functions.
+
+	* enc/trans/newline.trans: follow the type change.
+
+	* enc/trans/escape.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+	* enc/trans/iso2022.trans: ditto.
+
+	* enc/trans/japanese.trans: ditto.
+
+Mon Sep 15 02:37:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (str_encode): returns duplicated string if nothing
+	  changed.  [ruby-core:18578]
+
+Sun Sep 14 22:09:01 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb (CGI::parse): performance improvement
+
+Sun Sep 14 18:33:32 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: trivial changes.
+
+Sun Sep 14 16:15:22 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/matrix.rb (Vector#eql?): typo of the method name as "eqn?".
+	  (Vector#eqn?): removed. Defined by mistake.
+	  Fixes [ruby-dev:36294]. Reported by weda <weda AT
+	  issp.u-tokyo.ac.jp> and an anonymous user.
+
+	* test/matrix/test_matrix.rb: added.
+
+	* test/matrix/test_vector.rb: added.
+
+Sun Sep 14 16:07:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): associates the locale encoding with $0 as
+	  well as ARGV.
+
+Sun Sep 14 13:48:03 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* object.c (Init_Object): added metameta-class initialization for
+	  BasicObject, Object, and Module.
+	  The metameta-classes of them are now the metaclass of Class, as the
+	  metameta-class of Class is.
+
+	* object.c (boot_defmetametaclass): added.
+
+Sun Sep 14 10:10:43 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (f_{add,mul,sub}): omitted some shortcuts for preserving
+	  signed zero anyway.
+
+	* complex.c (nucomp_negate): new.
+
+Sun Sep 14 04:15:16 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h (OnigEncodingTypeST): add end argument for
+	  left_adjust_char_head.
+	  (ONIGENC_LEFT_ADJUST_CHAR_HEAD): add end argument.
+	  (onigenc_get_left_adjust_char_head): ditto.
+
+	* include/ruby/encoding.h (rb_enc_left_char_head): add end argument.
+
+	* regenc.h (onigenc_single_byte_left_adjust_char_head): ditto.
+
+	* regenc.c (onigenc_get_right_adjust_char_head): follow the interface
+	  change.
+	  (onigenc_get_right_adjust_char_head_with_prev): ditto.
+	  (onigenc_get_prev_char_head): ditto.
+	  (onigenc_step_back): ditto.
+	  (onigenc_get_left_adjust_char_head): ditto.
+	  (onigenc_single_byte_code_to_mbc): ditto.
+
+	* re.c: ditto.
+
+	* string.c: ditto.
+
+	* io.c: ditto.
+
+	* regexec.c: ditto.
+
+	* enc/euc_jp.c: ditto.
+
+	* enc/cp949.c: ditto.
+
+	* enc/shift_jis.c: ditto.
+
+	* enc/gbk.c: ditto.
+
+	* enc/big5.c: ditto.
+
+	* enc/euc_tw.c: ditto.
+
+	* enc/euc_kr.c: ditto.
+
+	* enc/emacs_mule.c: ditto.
+
+	* enc/gb18030.c: ditto.
+
+	* enc/utf_8.c: ditto.
+
+	* enc/utf_16le.c: ditto.
+
+	* enc/utf_16be.c: ditto.
+
+	* enc/utf_32le.c: ditto.
+
+	* enc/utf_32be.c: ditto.
+
+Sun Sep 14 03:43:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h (ONIGENC_STEP_BACK): add end argument.
+	  (onigenc_step_back): ditto.
+
+	* regenc.c (onigenc_step_back): add end argument.
+
+	* regexec.c: follow the interface change.
+
+Sun Sep 14 03:20:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h (onigenc_get_prev_char_head): add end
+	  argument.
+
+	* include/ruby/encoding.h (rb_enc_prev_char): ditto.
+
+	* regenc.c (onigenc_get_prev_char_head): add end argument.
+
+	* regparse.c: follow the interface change.
+
+	* regexec.c: ditto.
+
+	* string.c: ditto.
+
+	* parse.y: ditto.
+
+Sun Sep 14 02:04:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h
+	  (onigenc_get_right_adjust_char_head_with_prev): add end argument.
+
+	* regenc.c (onigenc_get_right_adjust_char_head_with_prev): use end
+	  argument.
+
+	* regexec.c (forward_search_range): follow the interface change.
+
+Sun Sep 14 01:38:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h (onigenc_get_right_adjust_char_head): add
+	  end argument.
+
+	* include/ruby/encoding.h (rb_enc_right_char_head): add end argument.
+
+	* regenc.c (onigenc_get_right_adjust_char_head): use end argument.
+
+	* re.c (rb_reg_adjust_startpos): follow the interface change.
+
+	* string.c (rb_str_index): ditto.
+
+	* regexec.c (backward_search_range): ditto.
+	  (onig_search): ditto.
+
+Sun Sep 14 00:43:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/ripper/depend (ripper.o): it depends on ruby headers.
+
+Sat Sep 13 19:27:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_encoding_set): don't need argc argument.
+	  (rb_io_s_pipe): update io_encoding_set call.
+	  (rb_io_set_encoding): ditto.
+
+Sat Sep 13 18:46:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_transcoding): align state field.
+
+Sat Sep 13 18:21:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (DECORATOR_P): renamed from SUPPLEMENTAL_CONVERSION.
+
+Sat Sep 13 18:05:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (make_writeconv): use rb_enc_name.
+	  (make_readconv): ditto.
+
+	* transcode.c (rb_econv_open_opts): ditto.
+	  (econv_args): ditto.
+	  (decorate_convpath): ditto.
+	  (econv_set_replacement): ditto.
+
+Sat Sep 13 11:15:38 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/rexml/document.rb: fixed typo.
+
+Sat Sep 13 10:53:13 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: refined.
+
+	* rational.c: ditto.
+
+Sat Sep 13 03:44:52 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_prepare_opts): raise ArgumentError if
+	  a broken string is specified as a replacement.
+
+Sat Sep 13 03:31:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* iseq.c (iseq_s_compile_file): use rb_file_open_str.
+
+	* io.c (rb_file_open_generic): call validate_enc_binmode.
+	  (rb_file_open_str): call FilePathValue.
+
+Sat Sep 13 02:58:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (validate_enc_binmode): extracted from rb_io_extract_modeenc.
+	  (rb_io_extract_modeenc): use validate_enc_binmode.
+	  (io_encoding_set): call validate_enc_binmode.
+
+Sat Sep 13 02:50:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_prepare_opts): initialize *opt and return 0
+	  if opthash is nil.
+
+Sat Sep 13 02:29:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/escape.trans: transcoder name renamed to use underscore.
+
+	* transcode.c: follow the renaming.
+
+Sat Sep 13 02:12:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_init_by_convpath_i): new function.
+	  (rb_econv_init_by_convpath): new function.
+	  (econv_init): use rb_econv_init_by_convpath.
+
+Sat Sep 13 01:11:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_args): take snamev_p and dnamev_p to prevent
+	  possible GC problem.
+
+Sat Sep 13 00:38:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (enc_arg): take arg as VALUE * to prevent possible GC
+	  problem.
+
+Fri Sep 12 23:19:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_add_transcoder_at): fix loop condition.
+
+Fri Sep 12 22:39:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (decorator_names): extracted from rb_econv_open.
+	  (rb_econv_open): use decorator_names.
+	  (econv_args): extracted from econv_init.
+	  (econv_init): use econv_args.
+	  (decorate_convpath): new function.
+	  (search_convpath_i): new function.
+	  (econv_s_search_convpath): new method.
+	  (Init_transcode): new method defined.
+
+Fri Sep 12 21:55:43 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_t): new field: num_allocated.
+	  (rb_econv_alloc): extracted from
+	  rb_econv_open_by_transcoder_entries.
+	  (rb_econv_add_transcoder_at): extracted from rb_econv_decorate_at
+	  and generalized
+	  (rb_econv_open_by_transcoder_entries): use rb_econv_alloc and
+	  rb_econv_add_transcoder_at.
+	  (rb_econv_add_converter): extracted from rb_econv_decorate_at.
+	  (rb_econv_decorate_at): use rb_econv_add_converter.
+
+Fri Sep 12 21:22:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_t): new field: started.
+	  (rb_econv_open_by_transcoder_entries): initialize started field.
+	  (rb_econv_convert): set started field.
+	  (rb_econv_insert_output): ditto.
+	  (rb_econv_decorate_at): check started field instead of num_finished.
+
+Fri Sep 12 20:28:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* regenc.c (onigenc_single_byte_code_to_mbc): follow OnigEncoding
+	  change.
+
+Fri Sep 12 20:00:22 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* regparse.c (parse_exp): fix resource leak.
+
+Fri Sep 12 01:21:22 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/syck/node.c (syck_replace_str2): fix wrong check to free ptr.
+
+Thu Sep 11 20:18:24 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_err_in_callback.rb (teardown): get rid
+	  of infinite loop.
+
+Thu Sep 11 19:12:56 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/oniguruma.h (OnigCodePoint): unsigned long to unsigned int.
+
+	* include/ruby/encoding.h (rb_enc_codepoint): ditto.
+
+	* encoding.c (rb_enc_codepoint): signed int to unsigned int.
+
+	* encoding.c (rb_enc_ascget): ditto.
+
+	* string.c (rb_str_casecmp): ditto.
+
+	* string.c (enc_succ_alnum_char): ditto.
+
+	* string.c (rb_str_inspect): ditto.
+
+	* string.c (rb_str_upcase_bang): ditto.
+
+	* string.c (rb_str_downcase_bang): ditto.
+
+	* string.c (rb_str_capitalize_bang): ditto.
+
+	* string.c (rb_str_swapcase_bang): ditto.
+
+	* string.c (struct tr): ditto.
+
+	* string.c (trnext): ditto.
+
+	* string.c (tr_trans): ditto.
+
+	* string.c (tr_setup_table): ditto.
+
+	* string.c (tr_find): ditto.
+
+	* string.c (rb_str_delete_bang): ditto.
+
+	* string.c (rb_str_squeeze_bang): ditto.
+
+	* string.c (rb_str_count): ditto.
+
+	* string.c (rb_str_split_m): ditto.
+
+	* string.c (rb_str_each_line): ditto.
+
+	* string.c (rb_str_lstrip_bang): ditto.
+
+	* string.c (rb_str_rstrip_bang): ditto.
+
+	* string.c (rb_str_intern): ditto.
+
+	* dir.c (char_casecmp): ditto.
+
+	* sprintf.c (rb_str_format): ditto.
+
+	* enc/emacs_mule.c (mbc_to_code): to be 32bit clean.
+
+	* enc/emacs_mule.c (code_to_mbc): ditto.
+
+	* enc/gb18030.c (mbc_to_code): ditto.
+
+	* enc/gb18030.c (code_to_mbc): ditto.
+
+Thu Sep 11 19:09:25 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/iconv/iconv.c: add attention for machine independent.
+
+Thu Sep 11 15:23:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigdivrem): adjust length for division and remainder.
+	  a patch from TOYOFUKU Chikanobu <nobu_toyofuku at nifty.com> in
+	  [ruby-dev:36231].
+
+Thu Sep 11 02:59:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_binwrite): extracted from io_fwrite.
+
+Thu Sep 11 02:25:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_encoding): new function.
+	  (make_encobj): new function.
+	  (econv_s_asciicompat_encoding): use make_encoding.
+	  (rb_econv_open_exc): use SUPPLEMENTAL_CONVERSION.
+	  (econv_convpath): use encoding object in the result.
+
+Thu Sep 11 02:14:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_convpath): new method.
+
+Wed Sep 10 23:00:43 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* tool/compile_prelude.rb: print "<internal:prelude>" instead of
+	  "prelude.rb" on stack trace.  [ruby-dev:36129]
+
+Wed Sep 10 21:19:58 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (big2str_karatsuba): remove unnecessary fixnum code.  a
+	  patch from TOYOFUKU Chikanobu <nobu_toyofuku at nifty.com> in
+	  [ruby-dev:36217].
+
+Wed Sep 10 21:09:32 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_err_in_callback.rb: InternetExplorer should
+	  be closed.
+
+	* test/win32ole/err_in_callback.rb: ditto.
+
+Wed Sep 10 18:25:19 2008  akira yamada  <akira@rice.p.arika.org>
+
+	* lib/uri/common.rb (URI::Parser): new class.
+
+	* lib/uri/mailto.rb, lib/uri/generic.rb: follow the above change.
+
+	* test/uri/test_parser.rb: added tests for URI::Parser.
+
+Wed Sep 10 10:35:32 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi/cookie.rb (CGI::Cookie#to_s): performance improvement
+	  from http://jp.rubyist.net/magazine/?0023-Cgirb.
+
+Wed Sep 10 10:12:29 2008  akira yamada  <akira@arika.org>
+
+	* lib/sync.rb (Sync_m#sync_exclusive): fixed
+	  typo. [ruby-dev:36233] [RubyForge#11680]
+
+Wed Sep 10 01:57:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (StrSet#hash): cache hash value.
+	  (ActionMap#hash): ditto.
+
+Wed Sep 10 01:26:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (econv_after_output): renamed from
+	  econv_output_followed_by_input.
+	  (ECONV_AFTER_OUTPUT): renamed from ECONV_OUTPUT_FOLLOWED_BY_INPUT.
+
+	* transcode.c: follow the renaming.
+
+	* io.c: ditto.
+
+Wed Sep 10 01:16:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (get_replacement_character): don't return ASCII
+	  incompatible replacements.
+	  (make_replacement): don't convert the result of
+	  get_replacement_character.
+
+Wed Sep 10 01:05:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/newline.trans (rb_universal_newline): swap src_encoding
+	  and dst_encoding.
+
+	* transcode.c (rb_econv_decorate_at): call get_transcoder_entry only
+	  once.
+	  (rb_econv_binmode): follow universal_newline change.
+
+Wed Sep 10 00:01:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): rename field: writeconv_stateless to
+	  writeconv_asciicompat.
+
+	* io.c: follow the renaming.
+
+	* gc.c: ditto.
+
+Tue Sep  9 23:55:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_fwrite): raise an error if ASCII incompatible string
+	  written for text mode IO without encoding conversion.
+	  (rb_io_extract_modeenc): binmode requirement changed.
+
+Tue Sep  9 21:59:48 2008  Takeyuki Fujioka  <xibbar@ruby-lang.org>
+
+	* lib/cgi*: split cgi.rb into four files. [ruby-dev:36041]
+
+Tue Sep  9 21:17:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_DECORATOR_MASK): merge
+	  ECONV_ENCODER_MASK and ECONV_DECORATOR_MASK.
+	  (ECONV_UNIVERSAL_NEWLINE_DECORATOR): renamed from
+	  ECONV_UNIVERSAL_NEWLINE_DECODER.
+	  (ECONV_CRLF_NEWLINE_DECORATOR): renamed from
+	  ECONV_CRLF_NEWLINE_ENCODER.
+	  (ECONV_CR_NEWLINE_DECORATOR): renamed from ECONV_CR_NEWLINE_ENCODER.
+	  (ECONV_XML_TEXT_DECORATOR): renamed from ECONV_XML_TEXT_ENCODER.
+	  (ECONV_XML_ATTR_CONTENT_DECORATOR): renamed from
+	  ECONV_XML_ATTR_CONTENT_ENCODER.
+	  (ECONV_STATEFUL_DECORATOR_MASK): renamed from
+	  ECONV_STATEFUL_ENCODER_MASK.
+	  (ECONV_XML_ATTR_QUOTE_DECORATOR): renamed from
+	  ECONV_XML_ATTR_CONTENT_DECORATOR.
+
+	* io.c: follow the renaming.
+
+	* transcode.c: ditto.
+
+Tue Sep  9 20:55:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open0): make it static.
+	  (rb_econv_open): place decorators at last.
+
+Tue Sep  9 20:40:15 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/openssl/test_ssl.rb (OpenSSL#test_client_session):
+	  Debian's openssl 0.9.8g-13 failed at assert(ssl.session_reused?),
+	  when use default SSLContext. [ruby-dev:36167]
+
+Tue Sep  9 20:34:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_t): last_trans_index removed.
+	  (rb_econv_open_by_transcoder_entries): follow the type change.
+	  (rb_econv_open0): ditto.
+	  (rb_econv_decorate_at): ditto.
+	  (rb_econv_binmode): ditto.
+	  (rb_econv_insert_output): simplified because there are no decorators
+	  at last.
+
+Tue Sep  9 20:21:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): reverse the order of decoders.
+
+Tue Sep  9 20:18:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open0): num_encoders and num_decoders removed.
+
+Tue Sep  9 03:59:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (make_writeconv): useless branch removed.
+
+Tue Sep  9 02:18:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_decorate_at_first): declared.
+	  (rb_econv_decorate_at_last): declared.
+
+	* transcode.c (rb_econv_open_by_transcoder_entries): initialize
+	  replacement_enc.  allocate outbuf for the last transcoder.
+	  (rb_econv_open0): extracted from rb_econv_open.
+	  (rb_econv_open): use rb_econv_open0 and decorate the result using
+	  rb_econv_decorate_at_first and rb_econv_decorate_at_last.
+	  (rb_econv_decorate_at): new function.
+	  (rb_econv_decorate_at_first): ditto.
+	  (rb_econv_decorate_at_last): ditto.
+	  (rb_econv_binmode): fix iteration end condition.
+	  (econv_init): don't set source_encoding_name and
+	  destination_encoding_name because they are set in rb_econv_open0.
+
+Tue Sep  9 01:10:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): make ec a local variable.
+
+Tue Sep  9 00:20:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_asciicompat_encoding): check decoder.
+
+Tue Sep  9 00:00:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_t): last_error.partial_input removed.
+
+Mon Sep  8 23:24:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_asciicompat_encoding): renamed
+	  from rb_econv_stateless_encoding to apply stateless ASCII
+	  incompatible encodings such as UTF-16BE.
+
+	* io.c (make_writeconv): use rb_econv_asciicompat_encoding.
+
+	* transcode_data.h (rb_transcoder_asciicompat_type_t): renamed from
+	  rb_transcoder_stateful_type_t.
+	  (rb_transcoder): use rb_transcoder_asciicompat_type_t.
+
+	* transcode.c: follow the type change.
+	  (asciicompat_encoding_i): renamed from stateless_encoding_i.
+	  (rb_econv_asciicompat_encoding): renamed from
+	  rb_econv_stateless_encoding.
+	  (econv_s_asciicompat_encoding): method renamed.
+
+	* tool/transcode-tblgen.rb: follow the type change.
+
+	* enc/trans/utf_16_32.trans: follow the type change.
+	  rb_from_UTF_16BE to UTF-8 is asciicompat_decoder.
+	  rb_from_UTF_16LE to UTF-8 is asciicompat_decoder.
+	  rb_from_UTF_32BE to UTF-8 is asciicompat_decoder.
+	  rb_from_UTF_32LE to UTF-8 is asciicompat_decoder.
+	  UTF-8 to rb_to_UTF_16BE is asciicompat_encoder.
+	  UTF-8 to rb_to_UTF_16LE is asciicompat_encoder.
+	  UTF-8 to rb_to_UTF_32BE is asciicompat_encoder.
+	  UTF-8 to rb_to_UTF_32LE is asciicompat_encoder.
+
+	* enc/trans/newline.trans: follow the type change.  universal newline
+	  decoder is asciicompat_converter.
+
+	* enc/trans/escape.trans: follow the type change.
+
+	* enc/trans/iso2022.trans: ditto.
+
+	* enc/trans/japanese.trans: ditto.
+
+Mon Sep  8 23:05:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_insert_output): "readagain" part should be
+	  after replacement.
+
+Mon Sep  8 22:30:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (dir_initialize): rename option name of Dir.open
+	  from :external_encoding to :encoding.
+
+Mon Sep  8 22:16:20 2008  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi.rb : obsolete regex "n" option. [ruby-dev:36130]
+
+Mon Sep  8 18:13:20 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (getppid): typo. [ruby-dev:36202]
+
+	* process.c (get_ppid): mention the return value on Windows.
+
+Mon Sep  8 18:15:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (ArrayCode): less string substitutions.
+
+Mon Sep  8 18:13:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* vm.c (rb_mRubyVMFrozenCore): registered for GC.
+
+	* re.c (rb_reg_preprocess_dregexp): fix GC problem on MacOS X with
+	  powerpc-apple-darwin8-gcc-4.0.1 (GCC) 4.0.1 (Apple Computer, Inc.
+	  build 5367).
+
+Mon Sep  8 18:09:07 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): define SIZE_MAX for VC++6/7.
+
+Mon Sep  8 17:46:09 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (filetime_to_unixtime): remove unused variable.
+	  [ruby-dev:36191]
+
+Mon Sep  8 13:47:39 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_io.rb (TestIO#test_dup): add open in block.
+	  see [ruby-dev:35957].
+
+Mon Sep  8 07:09:42 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: some adjustments.
+
+	* rational.c: ditto.
+
+Mon Sep  8 06:50:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (XLDFLAGS): added --enable-auto-import for cygwin and
+	  mingw.
+
+Mon Sep  8 02:04:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_set_replacement): renamed from
+	  rb_econv_set_replacemenet.
+
+	* transcode.c: follow the renaming.
+
+Mon Sep  8 01:10:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (transcode_tblgen): log message refined.
+
+Mon Sep  8 00:58:04 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: removed unused code.
+
+Mon Sep  8 00:04:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/iso2022.trans: upcase to iso-2022-jp.
+
+	* enc/emacs_mule.c: ditto.
+
+Sun Sep  7 23:46:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/iso2022.trans: stateless-iso-2022-jp is defined to avoid
+	  undefined conversion error between iso-2022-jp and the corresponding
+	  stateless encoding.
+
+	* enc/emacs_mule.c: replicate emacs-mule as stateless-iso-2022-jp.
+
+Sun Sep  7 20:03:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/escape.trans (hexstr): renamed from str1.
+
+Sun Sep  7 19:45:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/escape.trans: use transcode_tblgen.
+
+	* tool/transcode-tblgen.rb: generate an empty line after str1.
+
+Sun Sep  7 19:16:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (ActionMap#str_name): new method to
+	  generate a name base on string content.
+	  (ActionMap#gen_str): extracted from generate_info and use str_name.
+
+Sun Sep  7 18:28:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (ActionMap#generate_info): use a memo to
+	  avoid duplication for STR1.
+
+Sun Sep  7 18:10:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (STR1): defined for a string up to 255 bytes.
+	  (STR1_BYTEINDEX): defined.
+	  (makeSTR1): defined.
+
+	* tool/transcode-tblgen.rb: generate STR1.
+
+	* transcode.c (transcode_restartable0): interpret STR1.
+
+	* enc/trans/escape.trans (fun_so_escape_xml_chref): removed.  STR1 is
+	  used instead.
+
+Sun Sep  7 17:54:45 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* configure.in: Mac OS X's crypt(2) is broken with invalid salt.
+	  [ruby-dev:35899]
+	* string.c (rb_str_crypt): ditto.
+
+Sun Sep  7 17:29:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: o4 is usable only if the first byte is
+	  f0-f7.
+
+Sun Sep  7 12:44:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (make_writeconv): choose ASCII compatible encoding as
+	  intermediate encoding if stateful encoder exists.
+
+Sun Sep  7 12:09:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_XML_ATTR_CONTENT_ENCODER): defined.
+	  (ECONV_STATEFUL_ENCODER_MASK): defined.
+	  (ECONV_XML_ATTR_QUOTE_ENCODER): defined.
+	  (ECONV_XML_ATTR_ENCODER): removed.
+
+	* enc/trans/escape.trans (rb_escape_xml_attr_content): defined.
+	  (rb_escape_xml_attr_quote): defined.
+	  (rb_escape_xml_attr): removed.
+
+	* io.c (NEED_WRITECONV): writeconv is required if supplemental
+	  converter is used.
+	  (make_writeconv): apply stateful encoder in writeconv.
+
+	* transcode.c: follow the constant change.
+
+Sun Sep  7 07:24:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/*.el:  merged the following patches from Nathan Weizenbaum
+	  <nex342 at gmail.com>.   [ruby-core:18424]
+
+	* misc/ruby-mode.el: improve here-doc performance.
+
+	* misc/ruby-mode.el: don't crash the whole mode if the Subversion
+	  $ keyword isn't interpolated.
+
+	* misc/ruby-mode.el: don't highlight keywords when they're the
+	  beginning of non-keyword symbols.
+
+	* misc/ruby-mode.el, misc/ruby-electric.el: use regexp-opt where
+	  possible for more efficient regexps.
+
+	* misc/*.el: untabify, for internal consistency and consistency
+	  with standard Emacs elisp files.
+
+	* misc/ruby-mode.el: fix a variable-name error.
+
+	* misc/emacs-mode.el: don't set case-fold-search globally to nil;
+	  instead just set it when we need it.
+
+Sun Sep  7 06:31:51 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (file_expand_path): applied a patch from Nobuhiro Tachino
+	  in [ruby-dev:35948].  fix #491
+
+Sun Sep  7 03:37:05 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_block_optarg): allow default for block parameters as
+	  long as the value is primary.  a patch from Eric Mahurin
+	  <eric.mahurin at gmail.com> in [ruby-core:16880].
+
+Sun Sep  7 01:07:10 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (yylex): "1.upto 2 {|i| p i }" should be syntax error.
+	  [ruby-dev:36008]
+
+	* test/ruby/test_parse.rb (TestParse): update a test not to use
+	  recently fixed inconsistent syntax.
+
+Sun Sep  7 00:37:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_ERROR_HANDLER_MASK): defined.
+	  (ECONV_DECODER_MASK): defined.
+	  (ECONV_ENCODER_MASK): defined.
+
+	* io.c (make_writeconv): restrict ecflags for writeconv with
+	  ECONV_ERROR_HANDLER_MASK.
+
+Sat Sep  6 23:03:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_XML_TEXT_ENCODER): renamed from
+	  ECONV_HTML_TEXT_ENCODER.
+	  (ECONV_XML_ATTR_ENCODER): renamed from ECONV_HTML_ATTR_ENCODER.
+
+	* enc/trans/escape.trans: follow the renaming.
+
+	* transcode.c: ditto.
+
+Sat Sep  6 21:35:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/mkconstants.rb: add some IPV6 constants. [ruby-dev:36175]
+
+Sat Sep  6 20:10:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_description): show ECONV_HTML_TEXT_ENCODER and
+	  ECONV_HTML_ATTR_ENCODER.
+
+Sat Sep  6 20:06:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_binmode): check actual transcoders.
+
+Sat Sep  6 19:54:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): fix last_tc.
+
+Sat Sep  6 19:36:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_binmode): clear newline flags in writeconv_pre_ecflags.
+
+Sat Sep  6 18:47:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (str_transcode0): check ECONV_HTML_TEXT_ENCODER and
+	  ECONV_HTML_ATTR_ENCODER.
+
+Sat Sep  6 18:38:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (allocate_converted_string): fix overflow condition.
+
+Sat Sep  6 15:06:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (sym_html): new variable.
+	  (sym_text): ditto.
+	  (sym_attr): ditto.
+	  (econv_opts): check :html=>:text and :html=>:attr.
+	  (Init_transcode): initialize the above variables.
+
+Sat Sep  6 14:46:12 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_HTML_TEXT_ENCODER): new constant.
+	  (ECONV_HTML_ATTR_ENCODER): ditto.
+
+	* transcode.c (rb_econv_open): check ECONV_HTML_TEXT_ENCODER and
+	  ECONV_HTML_ATTR_ENCODER.
+	  (Init_transcode): Encoding::Converter::HTML_TEXT_ENCODER and
+	  Encoding::Converter::HTML_ATTR_ENCODER defined.
+
+Sat Sep  6 14:15:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (struct trans_open_t): defined to pass num_additional.
+	  (trans_open_i): use struct trans_open_t.
+	  (rb_econv_open): ditto.
+
+Sat Sep  6 13:43:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/escape.trans (fun_so_escape_html_attr): fix return type.
+
+Sat Sep  6 12:43:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): needless branch removed.
+
+Sat Sep  6 12:38:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): test condition simplified.
+
+Sat Sep  6 12:25:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (stateless_encoding_i): ignore supplemental conversions.
+	  Encoding::Converter.stateless_encoding("html-attr-escaped") should be
+	  nil.
+
+Sat Sep  6 12:19:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/escape.trans (escape_html_attr_init): new function.
+	  (fun_so_escape_html_attr): new function.
+	  (escape_html_attr_finish): new function.
+	  (rb_escape_html_attr): use them to quote the converted result.
+
+Sat Sep  6 07:54:36 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: uses f_real_p macro.
+
+Sat Sep  6 07:27:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): fail for ASCII incompatible with
+	  newline conversion.
+
+Sat Sep  6 07:24:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_extract_modeenc): raise an error for ASCII incompatible
+	  encoding without binmode.
+
+Sat Sep  6 07:12:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigdivrem1): optimization by skipping zeros at the
+	  tail of digits.  a patch from TOYOFUKU Chikanobu
+	  <nobu_toyofuku at nifty.com> in [ruby-dev:36169].
+
+Sat Sep  6 06:28:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/escape.trans: new file.
+
+Sat Sep  6 06:23:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (StrSet.parse): accept upper case
+	  hexadecimal digits.
+
+Sat Sep  6 05:37:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (output_hex_charref): upcase hexadecimal digits.
+
+Sat Sep  6 05:22:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_UNDEF_HEX_CHARREF): defined.
+
+	* transcode.c (output_hex_charref): new function.
+	  (rb_econv_convert): call output_hex_charref if
+	  ECONV_UNDEF_HEX_CHARREF.
+	  (Init_transcode): Encoding::Converter::UNDEF_HEX_CHARREF added.
+
+Sat Sep  6 03:52:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_convert): use ECONV_INVALID_MASK and
+	  ECONV_UNDEF_MASK.
+
+Sat Sep  6 06:05:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/ruby.h (DBL2NUM): renamed from DOUBLE2NUM.
+	  a patch from Tadashi Saito <shiba at mail2.accsnet.ne.jp>
+	  in [ruby-dev:36102].
+
+Sat Sep  6 04:44:06 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (Init_Proc): add '===' operator to use procs in when
+	  clause of the case statement.  inspired by <http://www.aimred.com/news/developers/2008/08/14/unlocking_the_power_of_case_equality_proc/>.
+
+Sat Sep  6 03:18:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (allocate_converted_string): add arguments for a buffer
+	  allocated by caller.
+	  (rb_econv_insert_output): provide caller allocated buffer to
+	  allocate_converted_string.
+
+Sat Sep  6 02:58:53 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (str_transcode_enc_args): local variables renamed.
+	  (str_transcode0): ditto.
+
+Sat Sep  6 02:23:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_s_stateless_encoding): new method.
+
+Sat Sep  6 02:01:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (enc_arg): extracted from str_transcode_enc_args.
+	  (str_transcode_enc_args): use enc_arg.
+
+Fri Sep  5 20:27:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_modestr_fmode): renamed from
+	  rb_io_mode_flags.
+	  (rb_io_modestr_oflags): renamed from rb_io_mode_modenum.
+	  (rb_io_oflags_fmode): renamed from rb_io_modenum_flags.
+	  (rb_io_mode_flags): defined as a macro.
+	  (rb_io_modenum_flags): ditto.
+
+	* io.c: follow the renaming with consistency.
+
+	* process.c (check_exec_redirect): call rb_io_modestr_oflags.
+
+	* ext/pty/depend: pty.o depends on io.h.
+
+Fri Sep  5 20:12:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/newline.trans (universal_newline_finish): new function.
+
+Fri Sep  5 20:07:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_mode_modenum): moved from
+	  include/ruby/intern.h.
+
+Fri Sep  5 19:59:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (MODENUM_MAX): removed.
+
+Fri Sep  5 18:37:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (copy_stream_fallback_body): use read method unless readpartial
+	  is available.  [ruby-dev:36124]
+
+Fri Sep  5 18:16:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_create): strips glibc style option before
+	  charset mapping.  retries without options if they seemed causing
+	  error, and warns.  [ruby-dev:36147]
+
+Fri Sep  5 03:09:48 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_data_to_ary): make it static.
+
+	* thread.c (thgroup_enclose): ditto.
+
+Fri Sep  5 02:56:37 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (thread_recycle_stack_slot, thread_recycle_stack_count):
+	  make it static.
+
+Fri Sep  5 02:40:38 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_critical): removed.
+
+Fri Sep  5 01:22:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (argf_mark): mark p->encs.ecopts.
+
+Fri Sep  5 00:45:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoder): rename fields:
+	  from_encoding -> src_encoding, to_encoding -> dst_encoding.
+
+	* transcode.c: follow the renaming.
+
+Fri Sep  5 00:22:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c: variables renamed for consistency.
+
+Fri Sep  5 00:05:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (encodes): make buff fixed length to avoid SEGV by
+	  ruby -e '["a"*10000000].pack("m1000000000")'
+
+Thu Sep  4 23:47:05 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_mode): set exception mode
+	  correctly.  In spite of BigDecimal.mode(BigDecimal::EXCEPTION_ALL,
+	  true), BigDecimal.new("NaN") did not raise an exception previously.
+
+Thu Sep  4 23:42:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_file_open_str): declared.
+
+	* io.c (rb_file_open_str): defined.
+
+	* ext/zlib/zlib.c (gzfile_s_open): use rb_file_open_str instead of
+	  rb_file_open.
+
+Thu Sep  4 23:18:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/newline.trans: record newline types met in universal
+	  newline decoder.
+
+Thu Sep  4 23:05:54 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/prime.rb (Prime::OldCompatibility#each): added compatibility to
+	    Ruby 1.8.7.
+	  (Prime#each): added more rdocs.
+	  (Prime#each): remembers the last value of the given block.
+
+Thu Sep  4 21:53:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_init): accept an integer as 3rd argument as well.
+
+Thu Sep  4 21:46:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (sym_partial_input): new variable.
+	  (econv_primitive_convert): accept a hash as 5th argument as well.
+
+Thu Sep  4 21:04:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (sym_universal_newline_decoder): new variable.
+	  (sym_crlf_newline_encoder): ditto.
+	  (sym_cr_newline_encoder): ditto.
+	  (econv_opts): check newline converter options.
+	  (econv_init): make 3rd argument hash/nil only.
+
+Thu Sep  4 21:03:28 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c: rename some debug functions.
+
+Thu Sep  4 20:57:54 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_check_trap_pending): added for compatibility.
+
+	* ext/tk/tcltklib.c, ext/tk/tkutil/tkutil.c: remove ruby/signal.h
+	  dependency.
+
+Thu Sep  4 20:30:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (PType): defined unconditionally.
+
+	* transcode.c (PType): don't define here.
+
+Thu Sep  4 20:19:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: variables renamed for consistency.
+
+Thu Sep  4 19:40:50 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (gc_profile_record_get): to static function.
+	  (gc_profile_result): ditto.
+	  (gc_profile_report): ditto.
+
+Thu Sep  4 19:20:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_enc_t): rename flags to ecflags.
+	  (rb_io_t): rename writeconv_pre_flags to writeconv_pre_ecflags.
+	  (MakeOpenFile): follow the renaming.
+
+	* io.c: follow the renaming.
+
+Thu Sep  4 19:10:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_INVALID_IGNORE): removed because
+	  it tend to cause security problem.  If the behaviour is really
+	  required, ECONV_INVALID_REPLACE with empty string can be used.
+	  For example, CVE-2006-2313, CVE-2008-1036, [ruby-core:15645],
+	  http://unicode.org/reports/tr36/
+	  (ECONV_UNDEF_IGNORE): ditto.
+
+	* transcode.c (rb_econv_convert): follow the above change.
+	  (econv_opts): ditto.
+	  (Init_transcode): ditto.
+
+Thu Sep  4 13:22:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (struct rb_vm_struct): replaced signal staff with trap
+	  staff.
+
+	* signal.c (signal_buff): per process resource now.
+
+	* signal.c (trap_list): moved to VM.
+
+	* signal.c (rb_get_next_signal): reverted.
+
+	* signal.c (rb_trap_exit): trap_pending_list was no longer used.
+
+	* thread.c (timer_thread_function): delivers buffered per-process
+	  signals to each VMs.
+
+	* vm.c (rb_vm_mark): marks trap_list.
+
+Thu Sep  4 13:01:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (struct sysopen_struct, rb_sysopen_internal, rb_sysopen):
+	  constified.
+
+	* io.c: expanded ARGF members macros.
+
+Thu Sep  4 10:43:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/transcode-tblgen.rb (citrus_decode_mapsrc): support older 1.8.
+
+Thu Sep  4 10:33:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_next_argv, argf_set_encoding): copy struct wise.
+
+Thu Sep  4 09:46:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_scan_open_args): follow rb_str_transcode change.
+
+Thu Sep  4 08:59:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (rb_find_file): fix GC problem on Debian GNU/Linux (IA64)
+	  with gcc (GCC) 4.1.2 20061115 (prerelease) (Debian 4.1.1-21).
+	  There is no register/memory to contain load_path.  A register (r35)
+	  contains &RARRAY_PTR(load_path), (char*)load_path + 32.
+
+Thu Sep  4 03:10:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): new fields: encs.ecopts and
+	  writeconv_pre_ecopts.
+	  (MakeOpenFile): initialize them.
+
+	* include/ruby/encoding.h (rb_str_transcode): take ecopts argument.
+	  (rb_econv_flags): removed.
+	  (rb_econv_prepare_opts): declared.
+	  (rb_econv_open_opts): declared.
+
+	* io.c (make_writeconv): use rb_econv_open_opts.
+	  (make_readconv): ditto.
+	  (io_fwrite): follow rb_str_transcode change.
+	  (rb_io_extract_modeenc): use rb_econv_prepare_opts.
+	  (rb_file_open_generic): initialize encs.ecopts.
+	  (rb_file_open_internal): ditto.
+	  (rb_io_reopen): ditto.
+	  (argf_ecopts): defined.
+	  (argf_next_argv): set encs.ecopts.
+	  (io_encoding_set): use rb_econv_prepare_opts.
+	  (argf_set_encoding): set argf_ecopts.
+
+	* gc.c (gc_mark_children): mark encs.ecopts and writeconv_pre_ecopts
+	  in T_FILE.
+
+	* transcode.c (transcode_loop): take ecopts argument. use
+	  rb_econv_open_opts.
+	  (rb_econv_flags): removed.
+	  (rb_econv_prepare_opts): defined.
+	  (rb_econv_open_opts): defined.
+	  (str_transcode0): take ecopts.
+	  (str_transcode): use rb_econv_prepare_opts.
+	  (rb_str_transcode): take ecopts.
+	  (econv_init): accept hash argument.
+	  (econv_insert_output): follow rb_str_transcode change.
+
+Thu Sep  4 01:30:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_set_replacemenet): declared.
+
+	* transcode.c (rb_econv_t): new fields: replacement_str,
+	  replacement_len, replacement_enc and replacement_allocated.
+	  (get_replacement_character): make len as size_t.
+	  (rb_econv_open_by_transcoder_entries): initialize the new fields.
+	  (rb_econv_close): deallocate replacement_str if it allocated.
+	  (make_replacement): new function.
+	  (output_replacement_character): use make_replacement.
+	  (rb_econv_set_replacemenet): defined.
+	  (econv_get_replacement): new method.
+	  (econv_set_replacement): new method.
+
+Thu Sep  4 01:12:03 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (filetime_to_timeval): new function, split from
+	  gettimeofday().
+
+	* win32/win32.c (gettimeofday): use above function.
+
+	* win32/win32.c (filetime_to_unixtime): ditto. [ruby-dev:36135]
+
+Thu Sep  4 01:00:57 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* transcode.c (transcode_restartable0): avoid VC++6's bug.
+
+Thu Sep  4 00:26:27 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (make_writeconv): follow akr's previous commit.
+
+Thu Sep  4 00:09:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_option_t): removed.  Since
+	  rb_econv_option_t has only one field, int flags, rb_econv_option_t is
+	  replaced by int.
+
+	* include/ruby/io.h: follow the above change.
+
+	* io.c: ditto.
+
+	* transcode.c: ditto.
+
+Thu Sep  4 00:04:59 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* win32/win32.c: fix ruby/signal.h depending codes.
+
+Thu Sep  4 00:01:45 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/etc/test_etc.rb (test_getpwuid, test_getgrgid): use
+	  Process.euid and egid instead of Etc.getlogin to identify current
+	  user which may be different from a login one (e.g., su command).
+
+Wed Sep  3 23:52:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/etc/test_etc.rb: check only typical use of
+	  setpwent/getpwent/endpwent and setgrent/getgrent/endgrent.
+
+Wed Sep  3 23:33:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_transcoding): moved from transcode_data.h.
+	  (TRANSCODING_READBUF): ditto.
+	  (TRANSCODING_WRITEBUF): ditto.
+	  (TRANSCODING_STATE_EMBED_MAX): ditto.
+	  (TRANSCODING_STATE): ditto.
+
+Wed Sep  3 23:03:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): remove stateful field.
+	  add state field.
+	  (TRANSCODING_STATE): defined.
+	  (rb_transcoder): add fields: state_size, state_init_func,
+	  state_fini_func.
+	  change rb_transcoding* argument to void*.
+
+	* transcode.c (transcode_restartable0): use TRANSCODING_STATE for
+	  first arguments of transcoder functions.
+	  (rb_transcoding_open_by_transcoder): initialize state field.
+	  (rb_transcoding_close): finalize state field.
+
+	* tool/transcode-tblgen.rb: provide state size/init/fini.
+
+	* enc/trans/newline.trans (universal_newline_init): defined.
+	  (fun_so_universal_newline): take void* as a state pointer.
+	  (rb_universal_newline): provide state size/init/fini.
+	  (rb_crlf_newline): ditto.
+	  (rb_cr_newline): ditto.
+
+	* enc/trans/iso2022.trans (iso2022jp_init): defined.
+	  (fun_si_iso2022jp_to_eucjp): take void* as a state pointer.
+	  (fun_so_iso2022jp_to_eucjp): ditto.
+	  (fun_so_eucjp_to_iso2022jp): ditto.
+	  (iso2022jp_reset_sequence_size): ditto.
+	  (finish_eucjp_to_iso2022jp): ditto.
+	  (rb_ISO_2022_JP_to_EUC_JP): provide state size/init/fini.
+	  (rb_EUC_JP_to_ISO_2022_JP): ditto.
+
+	* enc/trans/utf_16_32.trans (fun_so_from_utf_16be): take void* as a
+	  state pointer.
+	  (fun_so_to_utf_16be): ditto.
+	  (fun_so_from_utf_16le): ditto.
+	  (fun_so_to_utf_16le): ditto.
+	  (fun_so_from_utf_32be): ditto.
+	  (fun_so_to_utf_32be): ditto.
+	  (fun_so_from_utf_32le): ditto.
+	  (fun_so_to_utf_32le): ditto.
+	  (rb_from_UTF_16BE): provide state size/init/fini.
+	  (rb_to_UTF_16BE): ditto.
+	  (rb_from_UTF_16LE): ditto.
+	  (rb_to_UTF_16LE): ditto.
+	  (rb_from_UTF_32BE): ditto.
+	  (rb_to_UTF_32BE): ditto.
+	  (rb_from_UTF_32LE): ditto.
+	  (rb_to_UTF_32LE): ditto.
+
+	* enc/trans/japanese.trans (fun_so_eucjp2sjis): take void* as a state
+	  pointer.
+	  (fun_so_sjis2eucjp): ditto.
+	  (rb_eucjp2sjis): provide state size/init/fini.
+	  (rb_sjis2eucjp): provide state size/init/fini.
+
+Wed Sep  3 22:31:11 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/mathn.rb (Integer): moved into prime.rb.
+	  (Prime): ditto.
+
+	* lib/prime.rb (Integer): moved from mathn.rb.
+	  (Integer.each_prime): added.
+	  (Integer#prime?): added.
+	  (Prime): moved from mathn.rb.
+	    Its implementation was rewritten. see [ruby-dev:35863].
+	    And patched by Keiju ISHITSUKA <keiju@ishitsuka.com>,
+	    see [ruby-dev:36128].
+	  (Prime.new):                     obsolete.
+	  (Prime.instance):                added.
+	  (Prime.each):                    added.
+	  (Prime.int_from_prime_division): added.
+	  (Prime.prime_division):          added.
+	  (Prime.prime?):                  added.
+	    Patch by TOYOFUKU Chikanobu
+	    <nobu_toyofuku at nifty.com> in [ruby-dev:36067].
+	  (Prime.cache):                   removed.
+	  (Prime.primes):                  removed.
+	  (Prime.primes_so_far):           removed.
+	  (Prime#int_from_prime_division): added.
+	  (Prime#prime_division):          added.
+	  (Prime#prime?):                  added.
+	  (Prime#primes):                  removed.
+	  (Prime#primes_so_far):           removed.
+	  (Prime::PseudoPrmeGenerator):    added.
+	  (Prime::EratosthenesGenerator):  added.
+	  (Prime::TrialDivisionGenerator): added.
+	  (Prime::Generator23):            added.
+	  (Prime::TrialDivision):          added.
+	    Extracted from the previous implementation of Prime
+	    by Keiju ISHITSUKA.
+	  (Prime::EratosthenesSieve):      added.
+
+	* lib/.document (prime.rb): added
+
+	* lib/README (prime.rb): added
+
+	* test/test_prime.rb: added.
+
+Wed Sep  3 21:49:00 2008  David A. Black  <dblack@rubypal.com>
+
+	* lib/scanf.rb: fixed bug involving matching literal '['
+
+	* test/scanf/test_scanf.rb: added test for scanf.rb fix
+
+Wed Sep  3 21:31:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (WORDINDEX_SHIFT_BITS): defined.
+	  (WORDINDEX2INFO): defined.
+	  (INFO2WORDINDEX): defined.
+
+	* tool/transcode-tblgen.rb: use WORDINDEX2INFO.
+
+	* transcode.c: use INFO2WORDINDEX.
+
+Wed Sep  3 21:19:51 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* golf_prelude.rb: suppress warnings when goruby -v.
+
+Wed Sep  3 21:07:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (BL_MIN_BYTE): defined.
+	  (BL_MAX_BYTE): defined.
+	  (BL_OFFSET): defined.
+	  (BL_ACTION): defined.
+	  (transcode_restartable0): use BL_MIN_BYTE, BL_MAX_BYTE and
+	  BL_ACTION.
+
+Wed Sep  3 20:58:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (BYTE_ADDR): defined.
+	  (WORD_ADDR): ditto.
+	  (BL_BASE): use BYTE_ADDR and WORD_ADDR.
+	  (BL_INFO): use WORD_ADDR.
+
+Wed Sep  3 20:47:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoder): new field: byte_array_length and
+	  word_array_length.
+
+	* tool/transcode-tblgen.rb (transcode_generated_code): generate
+	  byte_array_length and word_array_length.
+
+Wed Sep  3 20:34:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (ArrayCode): new class.
+	  (ActionMap#gen_array_code): moved to ArrayCode.
+	  (ActionMap#numelt_array_code): ditto.
+	  (ActionMap#array_code_insert_at_last): ditto.
+	  (TRANSCODE_GENERATED_BYTES_CODE): use ArrayCode.
+	  (TRANSCODE_GENERATED_WORDS_CODE): ditto.
+
+Wed Sep  3 20:08:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/japanese.trans: new file.
+
+Wed Sep  3 20:04:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (ActionMap#gen_array_code): extracted from
+	  generate_lookup_node.
+	  (ActionMap#numelt_array_code): ditto.
+	  (ActionMap#array_code_insert_at_last): ditto.
+
+Wed Sep  3 20:01:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (parser_encoding_name): defined.
+	  (parser_tokadd_mbchar): show encoding in the message of
+	  "invalid multibyte char" error.
+
+Wed Sep  3 19:28:04 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_method_missing): C99.
+
+Wed Sep  3 19:18:12 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_thread.rb: fix test.
+	   [ruby-dev:35960]
+
+Wed Sep  3 17:48:32 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/make_transdb.rb: check $(srcdir)/enc/trans before
+	  enc/trans.
+
+	* enc/trans/make_transdb.rb: keep names_t.
+
+Wed Sep  3 16:54:46 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/signal.h: removed.
+
+	* common.mk, class.c, eval.c, eval_intern.h, file.c, gc.c, hash.c,
+	  io.c, process.c, signal.c: vm_core.h: ditto.
+	  Some unused external global variables are also removed.
+	 (rb_prohibit_interrupt, rb_trap_immediate, rb_trap_pending,
+	  rb_thread_critical)
+
+	* ext/openssl/ossl_ssl.c, ext/openssl/ossl_x509store.c,
+	  ext/readline/readline.c, ext/socket/depend,
+	  ext/socket/socket.c: ditto.
+
+Wed Sep  3 00:23:25 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_insnhelper.c (vm_method_missing): copy arguments to allocated
+	  memory from machine stack.  [ruby-dev:36064]
+
+Tue Sep  2 22:20:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (base_element): removed.
+	  (BYTE_LOOKUP): removed.
+	  (BYTE_LOOKUP_BASE): don't cast.
+	  (BYTE_LOOKUP_INFO): ditto.
+	  (PType): unsigned int, instead of uintptr_t.
+	  (rb_transcoding): change type of next_field, conv_tree_start and
+	  word_array.
+
+	* tool/transcode-tblgen.rb: generate word_array as array of unsigned
+	  int.
+
+	* transcode.c (transcode_restartable0): follow the above type change.
+
+Tue Sep  2 21:22:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: add prefix for byte_array and word_array.
+
+Tue Sep  2 20:57:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/make_transdb.rb: check foo.c only if foo.trans exists.
+
+Tue Sep  2 19:19:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/make_transdb.rb: error message improved.
+
+Tue Sep  2 14:05:50 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/trans/utf_16_32.trans (from_UTF_8): rename from to_UTF_16BE
+	  because it was not collect.
+
+Tue Sep  2 14:00:54 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* enc/emacs_mule.c (svn:executable): dropped executable bit.
+
+	* enc/make_encdb.rb (svn:executable): ditto.
+
+Tue Sep  2 12:58:58 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (COMPILERFLAG): new compiler flag to compile
+	  enc/trans/japanese*.c.
+
+Tue Sep  2 12:37:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/japanese_euc.trans: splitted from japanese.trans to avoid
+	  compiler limitation.  reported by usa.
+
+	* enc/trans/japanese_sjis.trans: ditto.
+
+Tue Sep  2 12:31:24 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (miniruby): new target.
+
+Tue Sep  2 11:47:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (struct load_arg): data is now st_table.
+
+	* marshal.c (mark_load_arg): marks data and compat_tbl.
+
+	* marshal.c (r_object0): no need to check if reentered.
+
+	* marshal.c (marshal_load): make the wrapper with mark_load_arg.
+
+Tue Sep  2 10:49:18 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (gettimeofday): easier calculation. use the definition
+	  of the Gregorian calendar.
+
+Tue Sep  2 10:33:12 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (distclean-golf): moved golf_prelude.c here from
+	    distclean-local.
+	  (GOLFPRELUDE): added.
+	  (distclean-local): moved golf_prelude.c into distclean-golf.
+
+Tue Sep  2 10:09:17 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (gettimeofday): calc tv_sec and tv_usec from system
+	  time by myself. [ruby-dev:36084]
+
+Tue Sep  2 04:00:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (o3): prevent sign extension on 64bit environment.
+	  (o4): ditto.
+
+Tue Sep  2 03:26:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: comment removed in generated code.
+
+Tue Sep  2 03:16:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: define TRANSCODE_TABLE_INFO in generated
+	  code.  use it in rb_transcoder.
+
+	* enc/trans/newline.trans: use TRANSCODE_TABLE_INFO.
+
+	* enc/trans/iso2022.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+Tue Sep  2 03:04:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: record infos and BYTE_LOOKUPs as index of
+	  word_array to avoid relocation.
+
+	* transcode.c (transcode_restartable0): add word_array to get infos
+	  and BYTE_LOOKUPs.
+
+	* transcode_data.h (BYTE_LOOKUP_INFO): change return type to
+	  uintptr_t.
+
+Tue Sep  2 02:48:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: don't need to cast offsets array.
+
+Tue Sep  2 02:36:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: record offsets array as index of
+	  byte_array to avoid relocation.
+
+	* transcode.c (transcode_restartable0): add byte_array to get offsets
+	  array.
+
+	* transcode_data.h (BYTE_LOOKUP_BASE): change return type to
+	  uintptr_t.
+	  (rb_transcoder): add fields: byte_array, word_array and word_size.
+
+	* enc/trans/newline.trans: follow rb_transcoder change.
+
+	* enc/trans/iso2022.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+Tue Sep  2 02:05:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: make infos arrays and BYTE_LOOKUPs into
+	  single array.
+
+Tue Sep  2 01:47:44 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (clean-golf): added.
+	  (distclean-golf): added.
+	  (realclean-golf): added.
+	  (clean): added clean-golf.
+	  (distclean): added distclean-golf.
+	  (realclean): added realclean-golf.
+
+Tue Sep  2 01:31:45 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (realclean-local): added newline.c.
+
+Tue Sep  2 01:19:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (BYTE_LOOKUP): change to uintptr_t array.
+	  (BYTE_LOOKUP_BASE): follow the type change.
+	  (BYTE_LOOKUP_INFO): ditto.
+	  (PType): ditto.
+	  (rb_transcoding): ditto.
+
+	* tool/transcode-tblgen.rb: follow the type change.
+
+	* transcode.c: ditto.
+
+	* enc/trans/newline.trans: ditto.
+
+	* enc/trans/iso2022.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+Mon Sep  1 23:32:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (BYTE_LOOKUP_BASE): abstract accessor for
+	  BYTE_LOOKUP.
+	  (BYTE_LOOKUP_INFO): ditto.
+
+	* transcode.c (transcode_restartable0): use BYTE_LOOKUP_BASE and
+	  BYTE_LOOKUP_INFO.
+
+Mon Sep  1 23:14:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: gather infos arrays and BYTE_LOOKUPs.
+
+Mon Sep  1 22:42:54 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/mkmf.rb (distclean): added RM_RF conftest.dSYM.
+	  Fixes distclean-ext problem on Mac OS X.
+	  (RM_RF): added.
+
+Mon Sep  1 22:37:06 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/rexml/document.rb: limit entity expansion.  Thanks, Luka
+	  Treiber, Mitja Kolsek, and Michael Koziarski.
+
+	* lib/rexml/entity.rb: ditto.
+
+	* test/rexml/test_document.rb: ditto.
+
+Mon Sep  1 22:20:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: make offsets arrays into single array.
+
+Mon Sep  1 21:46:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: gather offsets arrays at top.
+
+Mon Sep  1 21:09:04 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* ./: removed rubyspec from svn:ignore subversion property.
+	  rubyspec directory is no longer used.
+
+	* spec: added mspec and rubyspec to svn:ignore.
+
+Mon Sep  1 19:37:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: ValidEncoding['eucJP-ms'] defined.
+	  "\xA2\xAF".encode("utf-8", "eucJP-ms") should raise
+	  Encoding::ConversionUndefined, not Encoding::InvalidByteSequence.
+
+Mon Sep  1 18:27:51 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (gettimeofday): shouldn't use mktime(3) because it's
+	  buggy about handling summer time.
+	  reported by Yoshikawa <yoshixool AT gmail.com> at [ruby-dev:36071]
+
+Mon Sep  1 17:07:23 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/euc_jp.c (euc-jp-ms): euc-jp-ms is not an alias of EUC-JP
+	  but eucJP-ms.
+
+	* enc/trans/japanese.trans (eucJP-ms): eucJP-ms is the correct
+	  name of the encoding in Ruby. [ruby-dev:36070]
+
+Mon Sep  1 16:48:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/md5/md5init.c (md5), ext/digest/rmd160/rmd160init.c
+	  (rmd160) ext/digest/sha1/sha1init.c (sha1),
+	  ext/digest/sha2/sha2init.c (sha256, sha384, sha512): constified.
+
+Mon Sep  1 15:15:40 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.trans: fix mapping priority.
+	  IBM extended is prior than NEC selected IBM.
+
+Mon Sep  1 14:44:22 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/transcode-tblgen.rb (transcode_compile_tree): use the first
+	  mapping when some mappings are given for a character.
+	  [ruby-dev:36068]
+
+	* tool/transcode-tblgen.rb: expandtab.
+
+Mon Sep  1 14:40:18 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* README.EXT (Appendix A): fixed the name of class
+	  Enumerator. Enumerable::Enumerator was renamed just
+	  Enumerator at revision 18564.
+
+	* README.EXT (Appendix A): ditto.
+
+Mon Sep  1 14:37:58 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/transcode-tblgen.rb: set ERB source filename for error message.
+
+Mon Sep  1 14:37:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend: transdb.c may not present.
+
+Mon Sep  1 14:25:26 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.trans: fix Ruby 1.8 compatibility.
+
+	* enc/trans/japanese.trans: fix mapping priority. [ruby-dev:36068]
+
+Mon Sep  1 14:24:14 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_transcode.rb (test_windows_31j): added.
+
+Mon Sep  1 14:00:04 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* README.EXT (Data-types): fixed for current status.
+	  (Manipulating Ruby data): mentioned some more functions.
+	  (Class/module definition): ditto.
+	  (Global variables shared between C and Ruby):
+	    fixed prototypes for the getter/setter's of global variables.
+	  (Appendix A): mentioned some more files.
+
+	* README.EXT.ja: ditto.
+
+Mon Sep  1 11:31:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (copy_stream, fu_copy_stream0, copy_file): use
+	  IO.copy_stream to get rid of extraneous conversion.
+
+Mon Sep  1 02:55:15 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* strftime.c (rb_strftime): calc timezone offset by myself if system
+	  doesn't provide timezone info.
+
+Mon Sep  1 02:48:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: change "illegal" to "invalid".
+
+Mon Sep  1 02:31:16 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (transcode_generated_code): defined for
+	  generating table at once.
+	  (transcode_tblgen): returns an empty string.
+	  (transcode_generate_node): ditto.
+
+	* enc/trans/newline.trans: use transcode_generated_code.
+
+	* enc/trans/iso2022.trans: ditto.
+
+	* enc/trans/single_byte.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+	* enc/trans/japanese.trans: ditto.
+
+	* enc/trans/korean.trans: ditto.
+
+Mon Sep  1 02:10:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (citrus_decode_mapsrc): print logging
+	  message on STDERR.
+
+Mon Sep  1 01:24:09 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/eucjp-tbl.rb: replace by previous Citrus maps.
+
+	* enc/trans/sjis-tbl.rb: ditto.
+
+Mon Sep  1 01:18:26 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/transcode-tblgen.rb: add table generator from Citrus maps.
+
+	* enc/trans/japanese.trans: use Citrus maps.
+
+	* enc/trans/CP: add maps from Citrus.
+
+	* enc/trans/JIS: ditto.
+
+	* test/ruby/test_transcode.rb: Shift_JIS and EUC-JP doesn't support
+	  IBM extended characters.
+
+Mon Sep  1 00:19:45 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (docs): removed a useless make target.
+	  tool/makedoc.rb was removed at revision 13919.
+
+Sun Aug 31 20:49:42 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (numeric_abs2): new.
+
+Sun Aug 31 18:22:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_devel?): checks if the compiler works.
+
+Sun Aug 31 18:02:41 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* Temporarily uses git://github.com/yugui/rubyspec.git
+	  until specs for 1.9 is merged into
+	  git://github.com/rubyspec/rubyspec.git.
+
+Sun Aug 31 18:06:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode.c): set source_encoding and
+	  destination_encoding as encoding object.
+	  (ecerr_source_encoding): new method.
+	  (ecerr_destination_encoding): ditto.
+
+Sun Aug 31 17:58:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (Logging.log_open): opens in binary mode to get rid of
+	  extra CR.
+
+	* lib/mkmf.rb (try_func, try_var, have_struct_member),
+	  (scalar_ptr_type?, scalar_type?): use MAIN_DOES_NOTHING.
+
+	* lib/mkmf.rb (MAIN_DOES_NOTHING): defaults to ordinary main.
+
+	* win32/Makefile.sub (config): for MAIN_DOES_NOTHING for wince.
+
+Sun Aug 31 17:30:35 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/README (How to run): added.
+
+Sun Aug 31 17:28:25 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* spec/default.mspec (config): added.
+
+	* spec/default.mspec (target): replaced the built ruby
+	  to runruby.rb.
+	  Improved $LOAD_PATH on running specs.
+
+	* spec/default.mspec (flags): ditto.
+
+Sun Aug 31 17:16:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_econv_exception): rename instance variable names
+	  for storing encoding names.
+	  (ecerr_source_encoding_name): method renamed.
+	  (ecerr_destination_encoding_name): ditto.
+
+Sun Aug 31 16:57:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_putback): associate encoding to the result.
+
+Sun Aug 31 16:43:56 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/intern.h: rename RB_UBF_DFL to
+	  RUBY_UBF_IO and RUBY_UBF_PROCESS.
+	  Because there is no default (universal) unblocking function.
+
+	* ext/socket/socket.c, file.c, io.c, process.c, thread.c: ditto.
+
+Sun Aug 31 16:42:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_insert_output): raise ArgumentError on failure.
+
+Sun Aug 31 16:39:17 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_thread.rb: add a test.
+
+Sun Aug 31 16:34:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/single_byte.trans (us_ascii_map): don't define 8bit bytes.
+
+Sun Aug 31 14:27:27 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_last_error): new method.
+
+Sun Aug 31 14:17:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_convert): make two arguments,
+	  destination_byteoffset and destination_bytesize, optional.
+
+Sun Aug 31 14:12:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_econv_exception): error message simplified.
+
+Sun Aug 31 13:47:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/coverage/coverage.c (rb_mCoverage): removed unneeded static variable.
+
+	* ext/socket/socket.c (unix_recv_io): ditto.
+
+	* ext/socket/socket.c (mConst): ditto.
+
+Sun Aug 31 13:45:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspres_initialize): fix for
+	  initialization of r18168.
+
+Sun Aug 31 13:42:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/ripper/eventids2.c (ripper_init_eventids2): use rb_intern_const.
+
+	* ext/ripper/tools/generate.rb (generate_eventids1): ditto.
+
+	* ext/ripper/tools/generate.rb (generate_eventids2_table): ditto.
+
+Sun Aug 31 12:56:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/mkcallback.rb (rb_dl_init_callbacks): fix indent.
+
+Sun Aug 31 11:39:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/strscan/strscan.c (strscan_do_scan): fix usecnt decrement.
+
+Sun Aug 31 04:38:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_putback): make max argument optional.
+
+Sun Aug 31 04:35:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_putback): putback from last byte.
+
+Sun Aug 31 04:27:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_putback): renamed from econv_primitive_putback.
+
+Sun Aug 31 04:22:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_insert_output): renamed from
+	  econv_primitive_insert_output.
+
+Sun Aug 31 04:20:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_econv.rb: follow econv_primitive_errinfo change.
+
+Sun Aug 31 03:44:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_errinfo): 5th element (partial_input)
+	  removed from result.
+
+Sun Aug 31 01:53:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/mkcallback.rb: split callback.h to callback-[0-9].c to reduce
+	  memory consumption on compilation.  [ruby-dev:31898]
+
+	* ext/dl/depend: add dependency for callback-[0-9].[co].
+
+Sun Aug 31 01:27:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/mkcallback.rb (foreach_proc_entry): extracted.
+	  (gencallback): ditto.
+	  (gen_push_proc_ary): ditto.
+	  (gen_push_addr_ary): ditto.
+
+Sat Aug 30 23:51:01 2008  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_tuplebag.rb (test_has_expires_eh): fix failures
+	  [ruby-dev:35937].
+
+Sat Aug 30 23:38:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/test/test_import.rb: fix character code.
+
+	* ext/dl/test/test_func.rb: ditto.
+
+Sat Aug 30 22:23:31 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: added a static variable for nurat_to_f.
+
+Sat Aug 30 20:05:41 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_open, rb_w32_read, rb_w32_write): fallback to
+	  MSVCRT if text mode is specified. this case will not be used from
+	  ruby itself.
+
+Sat Aug 30 19:49:38 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): EOF is not error.
+	  ref [ruby-dev:36050]
+
+Sat Aug 30 18:17:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (struct token_info): constified.
+
+	* parse.y (token_info_get_column, token_info_has_nonspaces),
+	  (token_info_push, token_info_pop): constified.
+
+Sat Aug 30 15:43:03 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/openssl/ossl_config.c (Init_ossl_config): memory leak fixed.
+	  a patch <shinichiro.hamaji at gmail.com> in [ruby-dev:35880].
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_set_value): ditto.
+
+	* ext/strscan/strscan.c (strscan_do_scan): ditto.
+
+Sat Aug 30 14:58:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_initialize): fix for
+	  initialization of r18168.
+
+Sat Aug 30 14:47:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/mkcallback.rb: fix continuation line detection.
+
+Sat Aug 30 14:39:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/mkcallback.rb (rb_dl_init_callbacks): avoid GC problem which
+	  is caused by 'GC.stress=true; require "dl"'.
+
+Sat Aug 30 11:54:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (token_info_push): -w warns indentation level mismatch.
+	  based on a patch from Yukina Yamano presented at RubyKaigi'07.
+	  See <http://jp.rubyist.net/RubyKaigi2007/Log0609-LT06.html> and
+	  <http://www.logic-junction.com/products/rubyend.html> (Japanese).
+
+Sat Aug 30 10:46:44 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/iconv/iconv.c: remove include pragma for "ruby/intern.h".
+
+Sat Aug 30 10:43:03 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: remove include pragma for "ruby/intern.h".
+
+Sat Aug 30 10:41:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake.rb (Rake::Application#help): fix for OPTIONAL_ARGUMENT and
+	  multiple short options.  [ruby-dev:36051]
+
+Sat Aug 30 08:59:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_file_s_absolute_path),
+	  (rb_file_absolute_path): prototyped.
+
+Sat Aug 30 08:09:36 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_compatible_p): use TYPE not BUILTIN_TYPE.
+	  [ruby-dev:36048]
+
+Sat Aug 30 07:48:48 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_compatible_p): add SPECIAL_CONST_P check.
+	  [ruby-dev:36048]
+
+	* test/ruby/test_m17n.rb (test_compatible): fix test.
+
+Sat Aug 30 02:48:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_freeze): update rdoc to mention RuntimeError
+	  (not TypeError any longer) would be raised.  [ruby-dev:35982]
+
+Sat Aug 30 01:55:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigdivmod): remove redundant code.  a patch from
+	  TOYOFUKU Chikanobu <nobu_toyofuku at nifty.com> in [ruby-dev:36044].
+
+Sat Aug 30 01:37:20 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI#initialize): remove unused constants.  a patch
+	  from Takeyuki Fujioka in [ruby-dev:36046].
+
+Fri Aug 29 22:29:41 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c ({nucomp,numeric}_rect): new.
+
+	* complex.c: added some aliases
+	  (::rectangular, ::rect, #rectangular, #rect, #phase,
+	  #magnitude).
+
+	* complex.c (string_to_c_internal): should not strip any null
+	  bytes.
+
+	* rational.c (string_to_r_internal): ditto.
+
+	* rational.c (i_gcd): reverted to nurat 0.0.2's one.
+
+	* numeric.c: added an alias (#magnitude).
+
+	* bignum.c: ditto.
+
+	* test/ruby/test_complex.rb: added assertions.
+
+	* test/ruby/test_rational.rb: ditto.
+
+Fri Aug 29 19:46:02 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, include/ruby/ruby.h: rename T_DEFERRED to T_ZOMBIE.
+
+Fri Aug 29 18:22:27 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_compatible_p): raise TypeError when argument is not
+	  String nor Regexp.
+
+Fri Aug 29 18:20:56 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_index_m): only regexp uses byte offset.
+
+	* string.c (rb_str_rindex_m): ditto.
+
+Fri Aug 29 16:48:34 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method): copy arguments to allocated
+	  memory from machine stack.  [ruby-dev:36028]
+
+	* KNOWNBUGS.rb, bootstraptest/test_method.rb: move fixed test.
+
+Fri Aug 29 12:19:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_x509attr.c (ossl_x509attr_initialize): fix for
+	  initialization of r18168.
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_initialize): ditto.
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_initialize): ditto.
+
+Fri Aug 29 11:04:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): ASCII compatible string and US-ASCII
+	  regexp is compatible. [ruby-dev:36033]
+
+	* string.c (rb_str_index_m): second argument is position of character.
+	  by usa. [ruby-dev:36033]
+
+	* string.c (rb_str_rindex_m): ditto.
+
+Fri Aug 29 04:47:49 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read, rb_w32_write): more accurate handling
+	  of wait result.
+
+Fri Aug 29 02:59:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_finish): new method.
+
+Fri Aug 29 02:45:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_convert): new method.
+
+Fri Aug 29 02:38:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_convert): accept nil as
+	  destination_bytesize for unlimited destination size.
+
+Fri Aug 29 02:11:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_convert): accept nil as input for empty
+	  input.
+
+Fri Aug 29 02:03:56 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): supported %s and %P.
+
+	* time.c (time_strftime): ditto.
+
+	* test/ruby/test_time.rb (test_strftime): ditto.
+
+Fri Aug 29 01:57:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (sym_invalid_byte_sequence): new variable.
+	  (sym_undefined_conversion): ditto.
+	  (sym_destination_buffer_full): ditto.
+	  (sym_source_buffer_empty): ditto.
+	  (sym_finished): ditto.
+	  (sym_output_followed_by_input): ditto.
+	  (sym_incomplete_input): ditto.
+	  (econv_result_to_symbol): use above variables.
+	  (Init_transcode): initialize above variables.
+
+Fri Aug 29 00:53:40 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): fixed a bug of padding.
+
+	* test/ruby/test_time.rb: ditto.
+
+Fri Aug 29 00:19:54 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (set_pioinfo_extra): use MSVCRT's open() and close().
+
+Fri Aug 29 00:03:13 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* file.c (file_expand_path): check invalid access.
+
+Thu Aug 28 23:55:50 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): Time.mktime(2000).strftime("%-S") should
+	  return "0", not "".
+
+	* test/ruby/test_time.rb: ditto.
+
+Thu Aug 28 23:55:22 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_time.rb: suppress warning during test.
+
+Thu Aug 28 23:50:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_time.rb (test_strftime): make test for %a independent
+	  from local timezone.  [ruby-dev:35992]
+
+Thu Aug 28 23:39:49 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (rb_str_shared_replace): remove corrupt noembed string
+	  which causes SEGV.  [ruby-dev:35989]
+
+Thu Aug 28 23:36:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: dependency updated.
+
+Thu Aug 28 23:28:57 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* KNOWNBUGS.rb: add a test.  see [ruby-dev:36028]
+
+Thu Aug 28 21:43:05 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_open): overlapped file
+	  I/O support.
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_pipe): overlapped pipe
+	  I/O support.
+
+	* win32/win32.c (rb_w32_read, rb_w32_write): overlapped I/O support to
+	  enable canceling I/O.
+
+	* thread_win32.c (ubf_handle): remove workaround.
+
+Thu Aug 28 20:22:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_yield_setup_args): object with to_ary should
+	  behave like array.  a patch from Yusuke Endoh  <mame at tsg.ne.jp>
+	  in [ruby-dev:35988].  [ruby-dev:35977]
+
+	* object.c (convert_type): call less rb_intern() less frequently
+	  by using cache structure.
+
+	* vm_method.c (rb_method_node): fail earlier if no method found.
+
+Thu Aug 28 19:04:50 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bootstraptest/test_io.rb: no need to create real file.
+
+Thu Aug 28 09:26:53 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/cgi/test_cgi_multipart.rb: add test for multipart.
+	  Patch by Takeyuki Fujioka. [ruby-dev:36014]
+	  From CGIAlt http://cgialt.rubyforge.org/
+
+Thu Aug 28 09:22:01 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/cgi/test_cgi_modruby.rb: add test for mod_ruby adaptor.
+	  Patch by Takeyuki Fujioka. [ruby-dev:36013]
+	  From CGIAlt http://cgialt.rubyforge.org/
+
+Thu Aug 28 09:17:10 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/cgi/test_cgi_header.rb: add test for CGI::Header.
+	  Patch by Takeyuki Fujioka. [ruby-dev:36010] [ruby-dev:36011]
+	  From CGIAlt http://cgialt.rubyforge.org/
+
+Thu Aug 28 02:16:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): refcnt field removed.
+	  (MakeOpenFile): refcnt initialization removed.
+
+	* io.c (rb_io_fptr_finalize): don't check refcnt.
+	  (rb_io_close_read): don't use refcnt.
+
+Thu Aug 28 00:07:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_initialize): don't accept IO object.  [ruby-dev:35895]
+
+Wed Aug 27 23:28:51 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): WIN32OLE#[] and WIN32OLE#[]=
+	  accepts number argument.
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+Wed Aug 27 14:45:34 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/cgi/test_cgi_core.rb: add core test.
+	  Patch by Takeyuki Fujioka. [ruby-dev:36001] [ruby-dev:36002]
+	  From CGIAlt http://cgialt.rubyforge.org/
+
+Wed Aug 27 10:34:42 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::Cookie): remove delegate.
+	  Patch by Takeyuki Fujioka. [ruby-dev:35995]
+
+	* test/cgi/test_cgi_cookie.rb: added for above.
+	  Patch by Takeyuki Fujioka. [ruby-dev:35996]
+	  From CGIAlt http://cgialt.rubyforge.org/
+
+Wed Aug 27 01:13:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_loop): simplified.
+
+Wed Aug 27 01:03:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_result_t): enumeration constant:
+	  econv_incomplete_input.
+
+	* io.c (finish_writeconv): check econv_incomplete_input.
+
+	* transcode.c (transcode_restartable0): return econv_incomplete_input
+	  for unexpected end of source buffer.
+	  (trans_sweep): check econv_incomplete_input.
+	  (rb_trans_conv): ditto.
+	  (rb_econv_convert0): ditto.
+	  (rb_econv_convert): ditto.
+	  (transcode_loop): ditto.
+	  (make_econv_exception): change message for econv_incomplete_input.
+	  (econv_result_to_symbol): return :incomplete_input for
+	  econv_incomplete_input.
+	  (ecerr_incomplete_input): new method.
+
+Wed Aug 27 00:05:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): rename crbuf to cbuf.
+
+	* io.c: follow the renaming.
+
+Tue Aug 26 23:52:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_elem_t): move to transcode.c
+	  (rb_econv_t): defined as an incomplete type.
+
+	* transcode.c (rb_econv_elem_t): moved from encoding.h.
+	  (rb_econv_t): complete type defined.
+
+Tue Aug 26 22:44:12 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (time_asctime): workaround for MSVCRT's bug.
+	  based on a patch from wanabe <s.wanabe AT gmail.com> at
+	  [ruby-dev:35831] and a suggestion by KIMURA Koichi
+	  <kbk AT kt.rim.or.jp> at [ruby-dev:35832]
+	  fixed [ruby-dev:35555]
+
+Tue Aug 26 21:53:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): disable newline conversion for ASCII
+	  incompatible encodings.
+	  (str_transcode0): don't need disable newline conversion here.
+
+Tue Aug 26 21:44:39 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_binmode): binmode is effective only once.
+
+Tue Aug 26 19:43:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_file_s_expand_path): new class method for class File.
+
+	* file.c (file_expand_path): add absolute_path handling (no ~user
+	  expansion).  [ruby-core:18319]
+
+Tue Aug 26 19:27:54 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/ruby/test_enum.rb (TestEnumerable#test_each_with_object):
+	  Add a test for Enumerable#each_with_object.
+
+Tue Aug 26 19:25:44 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/ruby/test_enumerator.rb (TestEnumerator#test_with_object):
+	  Add (back) the test for Enumerator#with_object.
+
+Tue Aug 26 16:16:43 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/cgi/test_cgi_tag_helper.rb: add more tests for html3.
+	  Patch by Takeyuki Fujioka. [ruby-dev:35975]
+	  From CGIAlt http://cgialt.rubyforge.org/
+
+Tue Aug 26 14:43:10 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c: Activate Enumerator#with_object and add
+	  Enumerable#each_with_object. [experimental]
+
+Tue Aug 26 14:38:32 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_initialize),
+	  test/ruby/test_enumerator.rb: Add an ability to generate an
+	  enumerator from a block. [experimental] [ruby-dev:35903]
+
+Tue Aug 26 13:25:25 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb: added test_euc_jp
+	  (contributed by Yoshihiro Kambayashi)
+
+Tue Aug 26 11:54:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/cgi/test_cgi_tag_helper.rb: added by Takeyuki Fujioka.
+	  [ruby-dev:35971]
+
+Tue Aug 26 11:33:30 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi.rb (nOE_element_def): fix error when cascade methods which
+	  don't have arguments. Patch by Takeyuki Fujioka. [ruby-dev:35964]
+
+	* lib/cgi.rb (blockquote): String#collect (this depends on
+	  String#each) is no longer exists.
+	  Patch by Takeyuki Fujioka. [ruby-dev:35946]
+
+	* lib/cgi.rb (blockquote): ditto.
+
+Tue Aug 26 03:06:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (make_writeconv): copy invalid/undef flags.
+
+Tue Aug 26 02:47:08 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (dist): tool/make-snapshot is executable.
+
+	* tool/make-snapshot (package): supported release numbers 1.9.0-x.
+
+Tue Aug 26 02:43:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): new field: writeconv_pre_opts.
+
+	* io.c (make_writeconv): initialize writeconv_pre_opts.
+	  (io_fwrite): use writeconv_pre_opts.
+
+Tue Aug 26 01:48:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: test _WIN32 for CRLF platform.  (cygwin defines O_BINARY.)
+
+Tue Aug 26 01:29:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: force CRLF handling if RUBY_TEST_CRLF_ENVIRONMENT is defined.
+	  This is only for testing.
+
+Tue Aug 26 01:26:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (str_transcode0): disable newline conversion for ASCII
+	  incompatible encoding.
+
+Tue Aug 26 00:55:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (str_transcode0): don't short cut for newline conversion.
+
+Tue Aug 26 00:36:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_fwrite): add TEXTMODE_NEWLINE_ENCODER to option for
+	  rb_str_transcode.
+
+Tue Aug 26 00:24:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_econv_exception): show readagain part for invalid
+	  byte sequence exception.  store the part as an instance variable.
+	  (ecerr_readagain_bytes): new method to access the readagain part.
+
+Tue Aug 26 00:02:49 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpMult): fix double free.
+
+Mon Aug 25 23:59:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): make last_tc NULL if there are only
+	  additional transcoders.
+	  (econv_description): extracted from rb_econv_open_exc.
+	  (rb_econv_open_exc): use econv_description.
+	  (econv_inspect): use econv_description.
+
+Mon Aug 25 23:56:42 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32.c (init_stdhandle): set binmode.
+
+Mon Aug 25 23:38:17 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb(strftime): %[LN] can print with given
+	  arbitrary precision.
+
+	* lib/date/format.rb(strftime): optional flags and filed width
+	  should also affect %[nt].
+
+Mon Aug 25 23:01:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (defined_expr): default defined? should return
+	  "expression" without evaluating the expression.  [ruby-dev:35967]
+
+Mon Aug 25 22:46:01 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_thread.rb (test_status_and_stop_p): reduce the risk
+	  of race condition.
+
+Mon Aug 25 22:39:57 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/irb/extend-command.rb (def_extend_command): check number of
+	  arguments.  [ruby-dev:35074]
+
+	* lib/irb/ext/multi-irb.rb (search): check if a corresponding job is
+	  found.  [ruby-dev:35074]
+
+Mon Aug 25 22:29:13 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_io.rb (test_dup): fix typo.  see [ruby-dev:35958]
+
+Mon Aug 25 22:02:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open_by_transcoder_entries): initialize
+	  last_error.  num_trans may be zero.
+	  (rb_econv_convert0): num_trans may be zero.
+	  (rb_econv_putbackable): ditto.
+	  (rb_econv_putback): ditto.
+	  (rb_econv_convert): input_ptr and output_ptr may be NULL.
+
+Mon Aug 25 19:05:27 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (defined_expr): should handle NODE_{AND,OR} as
+	  "expression".  [ruby-dev:35951]
+
+Mon Aug 25 17:36:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_io_ungetbyte): added prototype.
+
+	* io.c (rb_io_gets): use rb_io_getline_1 for unget buffer.
+
+	* ruby.c (load_file): use rb_io_ungetbyte.
+
+	* transcode.c (rb_econv_substr_append): fix for buffer overrun.
+
+	* win32/enc-setup.mak (BUILTIN_TRANSOBJS): added.
+
+Mon Aug 25 14:22:21 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* parse.y (YYSTYPE): struct RVarmap has been no longer
+	  defined.
+	  see also r11717 by matz.
+
+Mon Aug 25 10:42:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_econv_open_exc): constified.
+
+Mon Aug 25 07:10:21 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/emacs_mule.c: fix ctype.
+
+Mon Aug 25 00:52:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (BUILTIN_TRANSSRCS): defined.
+	  (BUILTIN_TRANSOBJS): defined.
+
+	* enc/Makefile.in (BUILTIN_TRANSES): defined.
+
+	* enc/make_encmake.rb (BUILTIN_TRANSES): defined.
+
+	* enc/depend: don't generate rules for builtin transcoders.
+
+	* common.mk (COMMONOBJS): add BUILTIN_TRANSOBJS.
+	  (enc.mk): pass BUILTIN_TRANSOBJS.
+	  (newline.c): new rule.
+	  (newline.$(OBJEXT)): new rule.
+	  (srcs): newline.c added.
+
+	* Makefile.in (BUILTIN_TRANSSRCS): defined.
+	  (BUILTIN_TRANSOBJS): defined.
+
+	* transcode.c (Init_transcode): call Init_newline.
+
+Mon Aug 25 00:11:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (declare_transcoder): arguments order changed.
+	  (rb_declare_transcoder): call declare_transcoder just once.
+
+Sun Aug 24 19:48:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_t): use rb_econv_option_t.
+
+	* transcode.c: follow the rb_econv_t change.
+
+Sun Aug 24 19:40:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_init_copy): copy encs.
+
+Sun Aug 24 19:17:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_open): add an argument: vperm.
+	  (open_key_args): call rb_io_open with perm.
+
+Sun Aug 24 19:11:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (sym_invalid): removed.
+	  (sym_undef): removed.
+	  (sym_ignore): removed.
+	  (sym_replace): removed.
+
+Sun Aug 24 19:07:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_reopen): initialize fptr->encs.opts.
+
+Sun Aug 24 18:37:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_s_pipe): accept optional hash.
+	  (rb_io_set_encoding): ditto.
+	  (rb_io_extract_modeenc): use rb_econv_opts to initialize
+	  ecopts.
+	  (rb_file_open_generic): ditto.
+	  (rb_file_open_internal): ditto.
+	  (io_encoding_set): new argument: opt.
+	  (argf_set_encoding): copy fptr->encs.opts to argf_ecopts.
+
+	* transcode.c (rb_econv_opts): accept Qnil for initialization.
+
+Sun Aug 24 18:10:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_enc_t): add opts field.
+	  (FMODE_INVALID_MASK): removed.
+	  (FMODE_INVALID_IGNORE): ditto.
+	  (FMODE_INVALID_REPLACE): ditto.
+	  (FMODE_UNDEF_MASK): ditto.
+	  (FMODE_UNDEF_IGNORE): ditto.
+	  (FMODE_UNDEF_REPLACE): ditto.
+	  (MakeOpenFile): initialize the opts field.
+
+	* io.c (make_writeconv): use the opts field.
+	  (io_fwrite): ditto.
+	  (make_readconv): ditto.
+	  (argf_ecopts): new macro.
+	  (rb_io_extract_modeenc): fill the opts field.
+	  (argf_next_argv): ditto.
+	  (rb_file_open_generic): initialize the opts field.
+	  (rb_file_open_internal): ditto.
+	  (io_encoding_set): ditto.
+	  (argf_set_encoding): ditto.
+
+Sun Aug 24 18:02:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (struct argf): use struct rb_io_enc_t.
+	  (argf_enc): follow the struct argf change.
+	  (argf_enc2): ditto.
+
+Sun Aug 24 17:36:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_str_transcode): make 3rd argument
+	  rb_econv_option_t*.
+
+	* transcode.c (transcode_loop): take rb_econv_option_t* as a argument.
+	  (str_transcode0): ditto.
+	  (str_transcode): make rb_econv_option_t and call str_transcode0 with
+	  it.
+	  (rb_str_transcode): take rb_econv_option_t*.
+
+	* io.c (io_fwrite): follow the rb_str_transcode change.
+
+Sun Aug 24 16:47:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): make enc and enc2 as struct
+	  rb_io_enc_t;
+
+	* io.c: follow the structure change.
+	  (convconfig_t): defined by rb_io_enc_t.
+
+Sun Aug 24 16:19:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_opts): declared.
+
+	* transcode.c (rb_econv_opts): defined.
+
+	* io.c (rb_io_extract_modeenc): use rb_econv_opts.
+
+Sun Aug 24 16:06:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_option_t): defined.
+	  (rb_econv_open): 3rd arg changed.
+	  (rb_econv_open_exc): ditto.
+
+	* io.c (make_writeconv): use rb_econv_option_t.
+	  (make_readconv): ditto.
+	  (rb_econv_open): take rb_econv_option_t for options.
+	  (rb_econv_open_exc): ditto.
+	  (transcode_loop): use rb_econv_option_t.
+	  (econv_init): use rb_econv_option_t.
+
+Sun Aug 24 15:43:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_substr_append): associate dst with destination
+	  encoding when dst is created.
+
+Sun Aug 24 15:21:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_str_transcode): add ecflags argument.
+
+	* transcode.c (econv_opts): extracted from str_transcode.
+	  (str_transcode_enc_args): extracted from str_transcode.
+	  (str_transcode0): extracted from str_transcode.
+	  (str_transcode): use econv_opts, str_transcode_enc_args,
+	  str_transcode0.
+	  (rb_str_transcode): call str_transcode0.
+	  (econv_primitive_insert_output): give the additional argument for
+	  rb_str_transcode.
+
+	* io.c (make_writeconv): use invalid/undef flags.
+	  (io_fwrite): ditto.
+	  (rb_scan_open_args): give the additional argument for
+	  rb_str_transcode.
+
+Sun Aug 24 13:27:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (str_transcode): check last hash only if 0 < argc.
+
+Sun Aug 24 12:22:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_substr_append): renamed from
+	  rb_econv_string.
+	  (rb_econv_str_convert): declared.
+	  (rb_econv_substr_convert): declared.
+	  (rb_econv_str_append): declared.
+
+	* io.c (io_fwrite): use rb_econv_str_convert instead of
+	  rb_econv_string.
+
+	* transcode.c (rb_econv_substr_append): renamed from rb_econv_string.
+	  (rb_econv_str_append): new function.
+	  (rb_econv_substr_convert): ditto.
+	  (rb_econv_str_convert): ditto.
+
+Sun Aug 24 12:15:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#urlencode): str[0] returns char in 1.9.
+	  Patch by Junegunn Choi [ruby-core:18368]
+
+Sun Aug 24 12:02:39 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): return NULL on error.
+
+Sun Aug 24 11:40:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_open_exc): declared.
+
+	* transcode.c (rb_eNoConverter): new exception.
+	  (rb_econv_open_exc): new function.
+	  (transcode_loop): use rb_econv_open_exc.
+
+	* io.c (make_writeconv): use rb_econv_open_exc.
+	  (make_readconv): ditto.
+
+Sun Aug 24 11:26:54 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::unescapeHTML): consider ISO-8859-1.
+	  [ruby-dev:35936]
+
+Sun Aug 24 10:55:00 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/emacs_mule.c: support Emacs/Mule internal encoding.
+
+Sun Aug 24 08:03:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (daylight, timezone): not provided as dllexport on cygwin,
+	  right now.
+
+Sun Aug 24 06:39:05 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::unescapeHTML): more encoding sensible unescaping.
+	  [ruby-dev:35936]
+
+Sun Aug 24 04:23:19 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_compatible_p): raise TypeError when argument is Encoding.
+
+Sat Aug 23 23:25:47 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi.rb (CGI#out): remove auto encoding conversion and set
+	  LANGUAGE. [ruby-dev:35922]
+
+Sat Aug 23 20:01:29 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_process.rb (TestProcess#test_rlimit_value):
+	  add Errno::EINVAL. [ruby-dev:35900]
+
+Sat Aug 23 18:29:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_scan_open_args): use pop_last_hash.
+
+Sat Aug 23 16:59:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (FMODE_INVALID_MASK): defined.
+	  (FMODE_INVALID_IGNORE): defined.
+	  (FMODE_INVALID_REPLACE): defined.
+	  (FMODE_UNDEF_MASK): defined.
+	  (FMODE_UNDEF_IGNORE): defined.
+	  (FMODE_UNDEF_REPLACE): defined.
+
+	* io.c (sym_invalid): defined.
+	  (sym_undef): defined.
+	  (sym_ignore): defined.
+	  (sym_replace): defined.
+	  (make_readconv): specify ECONV_INVALID_* and ECONV_UNDEF_* if
+	  FMODE_INVALID_* and FMODE_UNDEF_* is set.
+	  (rb_io_extract_modeenc): check {:invalid, :undef} => {:replace,
+	  :ignore} for FMODE_INVALID_* and FMODE_UNDEF_*.
+
+Sat Aug 23 17:06:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime): support more flags.
+
+Sat Aug 23 15:14:10 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_opts.h: enable "OPT_TRACE_INSTRUCTION" on default.
+
+Sat Aug 23 14:59:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_INVALID_MASK): defined.
+	  (ECONV_INVALID_IGNORE): defined.
+	  (ECONV_INVALID_REPLACE): defined.
+	  (ECONV_UNDEF_MASK): defined.
+	  (ECONV_UNDEF_IGNORE): defined.
+	  (ECONV_UNDEF_REPLACE): defined.
+
+	* transcode.c (INVALID_IGNORE): removed.
+	  (INVALID_REPLACE): removed.
+	  (UNDEF_IGNORE): removed.
+	  (UNDEF_REPLACE): removed.
+	  (rb_econv_convert0): renamed from rb_econv_convert.
+	  (rb_econv_convert): defined to call rb_econv_convert0 with
+	  replace/ignore behavior moved from transcode_loop.
+	  (transcode_loop): replace/ignore behavior removed.
+
+Sat Aug 23 11:23:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_extract_modeenc): check :textmode and :binmode in option
+	  hash.
+
+Sat Aug 23 10:48:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (pty_getpty): follow rb_io_t's path -> pathv change.
+
+Sat Aug 23 10:42:52 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (check_pipe_command): extracted from rb_f_open and rb_io_open.
+	  (rb_f_open): use check_pipe_command.
+	  (rb_io_open): ditto.
+
+Sat Aug 23 10:13:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (read_all): fptr->enc2 is 0 if no conversion.
+	  (rb_io_getline_fast): ditto.
+	  (io_getc): ditto.
+
+Sat Aug 23 09:45:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): remove path field and add pathv field.
+	  (MakeOpenFile): initialize pathv as Qnil.
+
+	* gc.c: mark pathv field in T_FILE.
+
+	* io.c: follow the rb_io_t field change.
+
+	* file.c: ditto.
+
+	* ext/socket/socket.c: ditto.
+
+Sat Aug 23 01:42:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (FMODE_TEXTMODE): defined.
+
+	* include/ruby/encoding.h (rb_econv_t): new field: flags.
+	  (rb_econv_binmode): declared.
+
+	* io.c (io_unread): text mode hack removed.
+	  (NEED_NEWLINE_DECODER): defined.
+	  (NEED_NEWLINE_ENCODER): defined.
+	  (NEED_READCONV): defined.
+	  (NEED_WRITECONV): defined.
+	  (TEXTMODE_NEWLINE_ENCODER): defined for windows.
+	  (make_writeconv): setup converter with TEXTMODE_NEWLINE_ENCODER for
+	  text mode.
+	  (io_fwrite): use NEED_WRITECONV.  character code conversion is
+	  disabled if fptr->writeconv_stateless is nil.
+	  (make_readconv): setup converter with
+	  ECONV_UNIVERSAL_NEWLINE_DECODER for text mode.
+	  (read_all): use NEED_READCONV.
+	  (appendline): use NEED_READCONV.
+	  (rb_io_getline_1): use NEED_READCONV.
+	  (io_getc): use NEED_READCONV.
+	  (rb_io_ungetc): use NEED_READCONV.
+	  (rb_io_binmode): OS-level text mode test removed.  call
+	  rb_econv_binmode.
+	  (rb_io_binmode_m): call rb_io_binmode_m with write_io as well.
+	  (rb_io_flags_mode): return mode string including "t".
+	  (rb_io_mode_flags): detect "t" for text mode.
+	  (rb_sysopen): always specify O_BINARY.
+
+	* transcode.c (rb_econv_open_by_transcoder_entries): initialize flags.
+	  (rb_econv_open): if source and destination encoding is
+	  both empty string, open newline converter.  last_tc will be NULL in
+	  this case.
+	  (rb_econv_encoding_to_insert_output): last_tc may be NULL now.
+	  (rb_econv_string): ditto.
+	  (output_replacement_character): ditto.
+	  (transcode_loop): ditto.
+	  (econv_init): ditto.
+	  (econv_inspect): ditto.
+	  (rb_econv_binmode): new function.
+
+Fri Aug 22 21:18:40 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_div): now behaves as quo.
+
+	* complex.c (nucomp_s_generic_p): has been removed.
+
+	* complex.c (nucomp_to_s): adopts new form.
+
+	* complex.c (nucomp_inspect): ditto.
+
+	* complex.c (string_to_c_internal): ditto and supports polar form.
+
+	* complex.c (rb_complex_polar): new.
+
+	* rational.c (nurat_to_s): do not canonicalize.
+
+	* rational.c (nurat_inspect): adopts new form.
+
+	* rational.c (string_to_r_internal): ditto.
+
+	* include/ruby/intern.h: added a declaration.
+
+	* lib/complex.rb: added an obsolete class method.
+
+	* lib/cmath.rb: use scalar? instead of generic?.
+
+Fri Aug 22 20:06:46 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#shutdown):
+	  rescue Errno::ENOTCONN and close. [ruby-dev:35896]
+
+	* test/openssl/test_ssl.rb (OpenSSL#start_server): ditto.
+	  [ruby-dev:35897]
+
+	* lib/net/imap.rb (Net::IMAP#disconnect): ditto. [ruby-dev:35898]
+
+Fri Aug 22 19:58:27 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: no need to require the "lib/rational.rb" any more.
+
+Fri Aug 22 15:47:38 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/cgi.rb: use bytesize instead of size/length.
+
+Fri Aug 22 14:28:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime): supported flags and precision for most
+	  conversions.  [ruby-dev:35906]
+
+Fri Aug 22 14:04:04 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb: test_shift_jis:
+	  fixed comment strings (see r18291)
+
+Fri Aug 22 12:41:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_pid): use PIDT2NUM.
+
+Fri Aug 22 11:36:31 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (dir_enc_str_new): set US-ASCII to the path
+	  when the path is 7bit string and encoding is ASCII compatible.
+
+	* dir.c (push_glob): set file system encoding when argument encoding
+	  is US-ASCII.
+
+Fri Aug 22 11:30:38 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (dir_enc_str_new): renamed from dir_enc_str.
+
+	* dir.c (dir_read): use dir_enc_str_new.
+
+	* dir.c (dir_each): ditto.
+
+	* dir.c (push_pattern): ditto.
+
+Fri Aug 22 11:29:10 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: remove config.h dependency.
+
+Fri Aug 22 10:52:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (gmtime_r): check if reentrant versions are available.
+
+	* time.c (IF_HAVE_GMTIME_R, ASCTIME, GMTIME, LOCALTIME): use reentrant
+	  versions if available.
+
+Fri Aug 22 05:29:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (INT2NUM): just use a simple macro on LP64.
+	  (UINT2NUM): ditto.
+
+Fri Aug 22 05:10:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_file_open_generic): take filename as a VALUE.
+	  (rb_file_open_internal): ditto.
+	  (rb_io_open): ditto.
+	  (rb_file_open): pass filename as a VALUE to rb_file_open_internal.
+	  (rb_open_file): pass filename as a VALUE to rb_file_open_generic.
+	  (open_key_args): pass filename as a VALUE to rb_io_open.
+
+Fri Aug 22 04:33:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h: fix previous change for LP64.
+
+Fri Aug 22 03:19:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (NUM2LONG): make it inline function to evaluate
+	  the argument only once.
+	  `t = Object.new; def t.*(x) p x; 0 end; p Time.at(0, t)' did print x
+	  twice.
+	  (NUM2INT): ditto.
+	  (NUM2LL): ditto.
+	  (INT2NUM): make it inline function.
+	  (LONG2NUM): ditto.
+	  (UINT2NUM): ditto.
+	  (ULONG2NUM): ditto.
+
+Fri Aug 22 03:03:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_s_sysopen): mode can be a Bignum.
+
+Fri Aug 22 02:57:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_extract_modeenc): notify coerced to caller.
+	  (rb_io_initialize): mode may be a Bignum.
+
+Fri Aug 22 02:42:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_extract_modeenc): use rb_check_to_integer to test
+	  integer.
+
+Fri Aug 22 02:25:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* iseq.c (iseq_inspect): don't raise on uninitialized object.
+	  show real class name.
+
+Fri Aug 22 02:08:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_initialize): accept hash argument.
+
+Thu Aug 21 23:51:51 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): supported %F and %<precision>N.
+	  reverted config.h to ruby.h for Windows.
+
+	* test/ruby/test_time.rb (TestTime::test_strftime): added tests
+	  for %F and %N.
+
+	* time.c: documented %F and %N.
+
+Thu Aug 21 20:23:26 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_m17n_comb.rb (TestM17NComb#test_str_crypt): add
+	  key and salt to error message.
+
+Thu Aug 21 19:20:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (rb_stat_inspect): don't raise if self is not initialized.
+
+Thu Aug 21 19:17:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (pst_pid): use rb_attr_get to avoid warning on
+	  Process::Status.allocate.pid.
+	  (pst_inspect): don't raise if self is not initialized.
+
+Thu Aug 21 19:05:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_extract_encoding_option): if internal encoding is not
+	  specified, enc is external encoding.
+
+Thu Aug 21 14:22:50 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c: include ruby/config.h instead of ruby/ruby.h.
+
+	* common.mk: removed the old rule for missing/strftime.c.
+
+Thu Aug 21 09:29:01 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c (rb_strftime): return "UTC" instead of "GMT".
+
+	* test/ruby/test_time.rb (test_strftime): ditto.
+
+Thu Aug 21 07:59:04 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/iconv/test_option.rb (test_ignore_option): skip if iconv
+	  doesn't have transliterate.
+
+	* test/iconv/test_option.rb (test_translit_option): ditto.
+
+Thu Aug 21 06:12:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/scanf.rb (Scanf::FormatSpecifier#initialize): %i should accept
+	  single digit decimal.  [ruby-core:18355]
+
+Thu Aug 21 06:02:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_initialize_copy): Range is now a Struct.
+	  [ruby-core:18353]
+
+	* struct.c (rb_struct_init_copy): made public.
+
+Thu Aug 21 03:09:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_extract_encoding_option): enc2 is external encoding if
+	  external encoding and internal encoding is given.
+	  (pipe_open): add flags and convconfig argument to initialize
+	  fptr->{mode,enc,enc2}.
+	  (pipe_open_v): pass flags and convconfig from caller to pipe_open.
+	  (pipe_open_s): ditto.
+	  (pop_last_hash): new function.
+	  (rb_io_s_popen): use last hash as option to specify code conversion.
+	  (rb_io_open): specify flags and convconfig arguments for
+	  pipe_open_s.
+	  (rb_f_backquote): ditto.
+
+Thu Aug 21 02:27:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (prep_io): local variable renamed.
+	  (rb_io_fdopen): ditto.
+	  (prep_stdio): ditto.
+
+Thu Aug 21 01:58:59 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* configure.in: always check timezone().
+
+	* strftime.c (rb_strftime): prefer timezone/altzone rather than
+	  gettimeofday().  The second argument to gettimeofday() is ignored
+	  on Solaris.
+
+Thu Aug 21 02:03:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (open_key_args): simplified.
+
+Thu Aug 21 01:57:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (open_key_args): IO.foreach(path, rs, limit) didn't work.
+
+Thu Aug 21 01:31:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_file_sysopen_internal): unused function removed.
+	  (rb_file_sysopen): ditto.
+
+Thu Aug 21 01:09:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: use mode_t for the 3rd argument, permission, of open(2).
+
+Thu Aug 21 00:51:42 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* configure.in: removed strftime from AC_REPLACE_FUNCS().
+
+	* include/ruby/missing.h: removed prototype for strftime().
+
+	* missing/strftime.c: removed.
+
+	* time.c (time_to_s): use rb_strftime() instead of strftime(3).
+	  (time_zone): ditto.
+
+Thu Aug 21 00:49:50 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* strftime.c: win32 support.
+
+Thu Aug 21 00:20:05 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* strftime.c: new file.
+
+	* common.mk (COMMONOBJS): added strftime.$(OBJEXT).
+
+	* time.c (time_strftime): do not use strftime(3).  supported
+	  %L(millisecond) and %N(nanosecond).
+
+	* test/ruby/test_time.rb: added tests for %L and %N.
+
+Wed Aug 20 23:53:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_set_encoding): removed.
+	  (rb_io_open): set up encoding using new argument opt.
+	  (open_key_args): call rb_io_open with opt.  don't call
+	  io_set_encoding.
+
+Wed Aug 20 22:30:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_file_open_internal): use rb_io_flags_modenum.
+	  (rb_io_reopen): use rb_io_flags_modenum and rb_io_modenum_mode.
+	  (rb_io_stdio_file): ditto.
+
+Wed Aug 20 22:28:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_new): use the given class.
+
+	* vm.c (vm_make_proc): added an argument for the class.
+
+Wed Aug 20 22:24:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_take): get rid of extraneous iteration.
+
+Wed Aug 20 20:32:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_open_with_args): use rb_open_file instead of rb_io_open.
+
+Wed Aug 20 20:16:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_scan_open_args): extracted from rb_open_file.
+
+Wed Aug 20 19:22:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (convconfig_t): new type.
+	  (rb_io_extract_modeenc): new function.
+	  (rb_file_open_generic): new function.
+	  (rb_file_open_internal): use rb_file_open_generic.
+	  (rb_file_sysopen_internal): use rb_file_open_generic.
+	  (rb_open_file): use rb_io_extract_modeenc and rb_file_open_generic.
+	  (rb_io_open): call rb_file_open_internal instead of rb_file_open.
+
+Wed Aug 20 19:15:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (Init_IO): new constants: File::DSYNC, File::RSYNC and
+	  File::NOFOLLOW.
+
+Wed Aug 20 18:41:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_open_file): don't lookup :mode and :perm in opt.  it is
+	  useless because vmode and perm is overwritten by rb_scan_args
+	  anyway.
+
+Wed Aug 20 18:37:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (open_key_args): meaningless MEMCPY removed.
+
+Wed Aug 20 18:30:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (prep_io) [cygwin]: use FMODE_BINMODE instead of O_BINARY.
+
+Wed Aug 20 16:39:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sample): performance improvement for huge array.
+
+Wed Aug 20 12:28:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sample): get rid of infinite loop.  #455
+
+Wed Aug 20 06:09:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (FMODE_TRUNC): value changed because 0x100 is used
+	  as FMODE_NOREVLOOKUP in socket.c
+
+Wed Aug 20 05:19:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (parse_mode_enc): extracted from mode_enc.
+	  (io_extract_encoding_option): extracted from io_set_encoding.
+
+Wed Aug 20 04:17:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_flags_modenum): make it static.
+
+Wed Aug 20 03:36:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (FMODE_TRUNC): new constant.
+
+	* io.c (rb_io_mode_flags): set FMODE_TRUNC for "w".
+	  (rb_io_modenum_flags): set FMODE_TRUNC for O_TRUNC.
+	  (rb_io_flags_modenum): new function.
+	  (rb_io_mode_modenum): just use rb_io_mode_flags and
+	  rb_io_flags_modenum.
+
+Wed Aug 20 02:36:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (fopen): macro for vms removed.  fopen is not used now.
+
+Wed Aug 20 02:27:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open): fix memory leak.
+	  (rb_econv_close): ditto.
+	  fixed by shinichiro.h.  [ruby-dev:35880]
+
+Tue Aug 19 21:50:43 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_fopen): declaration removed.
+
+	* io.c (rb_fopen): unused function removed.
+
+Tue Aug 19 21:14:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (FMODE_READWRITE): define as
+	  FMODE_READABLE|FMODE_WRITABLE.
+
+Tue Aug 19 21:13:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_flags_mode): use hexadecimal in error message.
+	  (rb_io_modenum_mode): ditto.
+	  (rb_io_initialize): ditto.
+
+Tue Aug 19 18:23:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_flags_mode): add a prefix 0 to octal in error message.
+	  (rb_io_modenum_mode): ditto.
+
+Tue Aug 19 18:05:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (FMODE_*): sorted and describe the values in
+	  hexadecimal.
+
+Tue Aug 19 17:32:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_initialize): store the
+	  result of d2i_X509_NAME into DATA_PTR(self).
+
+Tue Aug 19 04:54:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rake.rb (FileUtils#ruby): takes care of space containing path.
+	  [ ruby-Bugs-21591 ]
+
+Tue Aug 19 01:32:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_open_file): encoding in mode string was ignored if perm is
+	  specified.
+
+Tue Aug 19 01:02:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_open_file): don't access argv[-1] by
+	  File.allocate.instance_eval { initialize }.
+
+Tue Aug 19 00:56:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_inspect): don't raise for uninitialized Regexp.
+
+Tue Aug 19 00:34:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_check_readable): side effect for STDIN removed.
+	  (rb_io_external_encoding): ditto.
+
+Mon Aug 18 23:27:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_ungetbyte): renamed from io_ungetc.
+	  (rb_io_ungetbyte): new method.
+	  (rb_io_ungetc): push back into character buffer if enc2 is set.
+
+Mon Aug 18 22:41:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (id_encode): removed.
+
+Mon Aug 18 22:30:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (make_writeconv): if enc and enc2 is set, convert
+	  string.encoding to enc2.
+
+	* include/ruby/io.h: comment changed.
+
+Mon Aug 18 21:02:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): new fields: writeconv,
+	  writeconv_stateless and writeconv_initialized.
+	  (MakeOpenFile): initialize them.
+
+	* include/ruby/encoding.h (rb_econv_stateless_encoding): declared.
+	  (rb_econv_string): declared.
+
+	* io.c (make_writeconv): new function.
+	  (io_fwrite): use econv.
+	  (make_readconv): fix error message.
+	  (finish_writeconv): new function.
+	  (fptr_finalize): call finish_writeconv.
+	  (clear_writeconv): new function.
+	  (clear_codeconv): new function to call both clear_readconv and
+	  clear_writeconv.
+	  (rb_io_fptr_finalize): call clear_codeconv instead of
+	  clear_readconv.
+	  (mode_enc): ditto.
+	  (io_set_encoding): ditto.
+	  (argf_next_argv): ditto.
+	  (io_encoding_set): ditto.
+
+	* gc.c (gc_mark_children): mark writeconv_stateless in T_FILE.
+
+	* transcode.c (stateless_encoding_i): new function.
+	  (rb_econv_stateless_encoding): ditto.
+	  (rb_econv_string): ditto.
+
+Mon Aug 18 17:23:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (clear_readconv): extracted from rb_io_fptr_finalize.
+	  (mode_enc): call clear_readconv.
+	  (io_set_encoding): ditto.
+	  (argf_next_argv): ditto.
+	  (io_encoding_set): ditto.
+
+Mon Aug 18 16:54:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (mode_enc): modify enc and enc2 consistently.
+	  (io_set_encoding): ditto.
+	  (io_encoding_set): ditto.
+
+Mon Aug 18 13:21:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_search_path): prevent infinite loop by
+	  searching a path for self conversion.
+
+Mon Aug 18 12:45:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_enc_str): code conversion removed.
+	  (io_enc_str_converted): removed because it is identical to
+	  io_enc_str now.
+
+Mon Aug 18 12:12:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_shift_crbuf): add strp argument to append into existing
+	  string.
+	  (read_all): use econv if enc2 is set.
+	  (io_getc): follow the io_shift_crbuf change.
+
+Mon Aug 18 10:35:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_enc_str_converted): new function.
+	  (make_readconv): extracted from io_getc.
+	  (more_char): ditto.
+	  (appendline): use econv via make_readconv and more_char for code
+	  conversion.
+	  (prepare_getline_args): don't convert record separator.
+	  (rb_io_getline_1): don't use rb_io_getline_fast if enc2 is set.
+	  (io_getc): use make_readconv and more_char.
+
+Mon Aug 18 08:27:44 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* common.mk: fix error in uncommon.mk.
+
+Mon Aug 18 06:10:08 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: config.h depends config.status.
+
+Mon Aug 18 03:59:43 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (appendline): appendline cannot check character boundary.
+	  (rb_io_getline_1): relax limit until character boundary.
+
+Mon Aug 18 02:25:11 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#eof?): added lacked method.
+
+Sun Aug 17 21:50:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (prepare_getline_args): io.gets(10,nil) should cause TypeError.
+
+Sun Aug 17 15:58:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: not check config.h.
+
+	* lib/mkmf.rb (init_mkmf): add include/ruby/backward to $INCFLAGS.
+
+	* include/ruby/backward/{st,util}.h: added for backward compatibility.
+	  [ruby-dev:35811]
+
+Sun Aug 17 13:40:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_putback): new method.
+
+Sun Aug 17 13:23:53 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_putbackable): declared.
+	  (rb_econv_putback): ditto.
+
+	* transcode.c (rb_econv_putbackable): implemented.
+	  (rb_econv_putback): ditto.
+
+	* io.c (io_getc): put back bytes if possible.
+
+Sun Aug 17 12:00:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (make_econv_exception): add several instance variables
+	  to exception object.
+	  (ecerr_source_encoding): new method:
+	  Encoding::ConversionUndefined#source_encoding and
+	  Encoding::InvalidByteSequence#source_encoding.
+	  (ecerr_destination_encoding): new method:
+	  Encoding::ConversionUndefined#destination_encoding and
+	  Encoding::InvalidByteSequence#destination_encoding.
+	  (econverr_error_char): new method:
+	  Encoding::ConversionUndefined#error_char.
+	  (econverr_error_bytes): new method:
+	  Encoding::ConversionUndefined#error_bytes.
+
+Sun Aug 17 11:43:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (struct MT): packed Mersenne Twister staffs.
+
+	* random.c (struct RandSeed): packed random seed staffs.
+
+Sun Aug 17 08:38:26 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/iconv/test_option.rb (test_ignore_option): skip if iconv
+	  doesn't have transliterate.
+
+	* test/iconv/test_option.rb (test_translit_option): ditto.
+
+Sun Aug 17 01:29:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): new fields: readconv, crbuf, crbuf_off,
+	  crbuf_len, crbuf_capa.
+	  (MakeOpenFile): initialize them.
+
+	* io.c (io_shift_crbuf): new function.
+	  (io_getc): use econv.
+	  (rb_io_fptr_finalize): finalize readconv and crbuf.
+
+Sun Aug 17 00:02:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_check_error): declared.
+
+	* transcode.c (make_econv_exception): new function.
+	  (transcode_loop): use make_econv_exception.
+	  (rb_econv_check_error): defined.
+
+Sat Aug 16 15:23:16 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_elem_t): fields removed: from and
+	  to.
+	  (rb_econv_t): new fields: source_encoding_name and
+	  destination_encoding_name.
+
+	* transcode.c (rb_econv_open_by_transcoder_entries): initialize the
+	  new fields.
+	  (rb_econv_open): set up the new fields.
+	  (econv_inspect): use the new fields.
+
+Sat Aug 16 14:22:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_t): add fields: in_buf_start,
+	  in_data_start, in_data_end, in_buf_end and last_trans_index.
+	  (rb_econv_output): removed.
+	  (rb_econv_insert_output): declared.
+	  (rb_econv_encoding_to_insert_output): declared.
+
+	* enc/trans/newline.trans (rb_universal_newline): stateful_type
+	  changed.
+
+	* transcode.c (transcode_restartable0): initialize inchar_start,
+	  tc->recognized_len and next_table at beginning of the loop.
+	  (rb_econv_open_by_transcoder_entries): initialize new fields.
+	  (rb_econv_open): setup last_trans_index.
+	  (trans_sweep): last out_buf_start can be non-NULL now.
+	  (rb_econv_convert): check last out_buf_start and in_buf_start at
+	  first.
+	  (rb_econv_output_with_destination_encoding): removed.
+	  (econv_just_convert): removed.
+	  (rb_econv_output): removed.
+	  (econv_primitive_output): method removed.
+	  (rb_econv_encoding_to_insert_output): new function.
+	  (allocate_converted_string): new function.
+	  (rb_econv_insert_output): new function.
+	  (econv_primitive_insert_output): new method.
+	  (output_replacement_character): use rb_econv_insert_output.  unused
+	  arguments removed.
+
+Sat Aug 16 09:20:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_intern_const): tiny optimization.
+
+	* include/ruby/ruby.h (SSIZET2NUM, NUM2SSIZET, SSIZE_MAX, SSIZE_MIN):
+	  macros for ssize_t.
+
+Sat Aug 16 08:11:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoder_stateful_type_t): defined.
+	  (rb_transcoder): add field: stateful_type.
+
+	* tool/transcode-tblgen.rb: generate stateful_type field as
+	  stateless_converter.
+
+	* enc/trans/iso2022.trans: follow rb_transcoder change.
+
+	* enc/trans/newline.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+Fri Aug 15 23:07:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_just_convert): extracted from rb_econv_output.
+	  (rb_econv_output): use econv_just_convert.
+	  (econv_primitive_output): new method.
+
+Fri Aug 15 19:57:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_output): add str_encoding
+	  argument.
+
+	* transcode.c (get_replacement_character): add repl_enc_ptr argument.
+	  (rb_econv_output_with_destination_encoding): renamed from
+	  rb_econv_output and make it static.
+	  (rb_econv_output): convert str and call
+	  rb_econv_output_with_destination_encoding.
+	  (output_replacement_character): follow above interface change.
+
+Fri Aug 15 19:33:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_drop_bytes): use memmove.
+
+Fri Aug 15 18:33:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_t): add error_tc in last_error.
+
+	* transcode.c (rb_econv_convert): fill error_tc.
+
+Fri Aug 15 18:10:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_t): new field: last_error.
+
+	* transcode.c (rb_trans_conv): new argument: result_position_ptr.
+	  (rb_econv_convert): fill last_error.
+	  (econv_result_to_symbol): extracted from econv_primitive_convert.
+	  (econv_primitive_errinfo): new method.
+
+Fri Aug 15 17:39:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (parse.{c,h}): creates in the dependency order.
+
+Fri Aug 15 16:59:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (exec_event_hooks): skips RubyVM::FrozenCore.
+
+	* vm.c (Init_VM): get rid of SEGV in a trace proc.
+
+Fri Aug 15 09:33:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (conv_init): check empty name.
+
+Fri Aug 15 09:03:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_init): don't create dummy encoding if
+	  rb_econv_open is failed.
+	  (make_dummy_encoding): new function extracted from make_encoding.
+	  (make_encoding): removed.
+
+Fri Aug 15 01:07:16 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk ({$(srcdir)}.y.c): escape backslash.
+
+Fri Aug 15 01:05:39 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_convert): set destination_buffer
+	  encoding.
+
+Fri Aug 15 00:52:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_output): declared.
+
+	* transcode_data.h (rb_transcoder): add resetsize_func field.
+
+	* enc/trans/iso2022.trans (iso2022jp_reset_sequence_size): defined.
+	  (rb_EUC_JP_to_ISO_2022_JP): provide resetsize_func.
+
+	* tool/transcode-tblgen.rb: set NULL for resetsize_func.
+
+	* transcode.c (rb_econv_output): new function for inserting output.
+	  (output_replacement_character): use rb_econv_output.
+	  (transcode_loop): check return value of
+	  output_replacement_character.
+
+Thu Aug 14 23:47:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ECONV_UNIVERSAL_NEWLINE_DECODER): defined.
+	  (ECONV_CRLF_NEWLINE_ENCODER): ditto.
+	  (ECONV_CR_NEWLINE_ENCODER): ditto.
+	  (ECONV_PARTIAL_INPUT): ditto.
+	  (ECONV_OUTPUT_FOLLOWED_BY_INPUT): ditto.
+
+	* transcode.c: use ECONV_* defined as above.
+
+Thu Aug 14 23:35:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c: local variable renamed.
+
+Thu Aug 14 23:22:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_econv_result_t): moved from
+	  transcode_data.h.
+	  (rb_econv_elem_t): ditto.
+	  (rb_econv_t): ditto.  source_encoding and destination_encoding field
+	  is added.
+	  (rb_econv_open): declared.
+	  (rb_econv_convert): ditto.
+	  (rb_econv_close): ditto.
+
+	* transcode.c (rb_econv_open_by_transcoder_entries): initialize
+	  source_encoding and destination_encoding field as NULL.
+	  (rb_econv_open): make it external linkage.
+	  (rb_econv_close): ditto.
+	  (rb_econv_convert): ditto.  renamed from rb_econv_conv.
+	  (make_encoding): new function.
+	  (econv_init): use make_encoding and store rb_encoding* in
+	  rb_econv_t.
+	  (econv_source_encoding): new method
+	  Encoding::Converter#source_encoding.
+	  (econv_destination_encoding): new method
+	  Encoding::Converter#destination_encoding.
+
+Thu Aug 14 22:44:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_econv_result_t): change enumeration
+	  constant's prefix: transcode_ to econv_.
+
+	* transcode.c: follow the constant prefix change.
+
+Thu Aug 14 21:34:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_init): accept Encoding object as source_encoding
+	  and destination_encoding.
+
+Thu Aug 14 21:10:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_filesystem_encoding): use default external encoding
+	  for Unix.
+
+	* dir.c (dir_initialize): don't cache fs_encoding.
+
+Thu Aug 14 20:58:57 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_open_by_transcoder_entries): renamed from
+	  rb_trans_open_by_transcoder_entries.
+
+Thu Aug 14 20:56:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_max_output): removed.
+
+Thu Aug 14 20:52:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: check unexpected actions.
+
+Thu Aug 14 20:25:52 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (transcode_invalid_byte_sequence): renamed from
+	  transcode_invalid_input.
+	  (transcode_destination_buffer_full): renamed from transcode_obuf_full.
+	  (transcode_source_buffer_empty): renamed from transcode_ibuf_empty.
+	  (rb_econv_result_t): renamed from rb_trans_result_t.
+	  (rb_econv_elem_t): renamed from rb_trans_elem_t.
+	  (rb_econv_t): renamed from rb_trans_t.
+
+	* transcode.c (UNIVERSAL_NEWLINE_DECODER): renamed from
+	  UNIVERSAL_NEWLINE.
+	  (CRLF_NEWLINE_ENCODER): renamed from CRLF_NEWLINE.
+	  (CR_NEWLINE_ENCODER): renamed from CR_NEWLINE.
+	  (rb_econv_open): renamed from rb_trans_open.
+	  (rb_econv_close): renamed from rb_trans_close.
+
+Thu Aug 14 19:41:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (load_transcoder): unused function removed.
+	  (rb_transcoding_open): ditto.
+
+Thu Aug 14 17:57:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_enc_ispunct): added.
+
+	* common.mk (COMMONOBJS), inits.c (rb_call_inits): id.c is now
+	  included from parse.c.
+
+	* id.c (Init_id), id.h (ruby_method_ids): added IDs used by VM.
+
+	* parse.y (global_symbols): added rooms for VM IDs.
+
+	* parse.y (rb_intern3, rb_id2str): single punctuation symbol is now
+	  same as char code.
+
+Thu Aug 14 17:46:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (union tmpyystype): no longer needed, since YYSTYPE is
+	  defined in parse.h now.
+
+Thu Aug 14 17:27:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* version.c (Init_version): add RUBY_ENGINE constant.
+
+Thu Aug 14 15:34:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_econv_conv): new function.  it doesn't consume
+	  input too much, even for multilevel conversion.
+	  (transcode_loop): use rb_econv_conv.
+	  (econv_primitive_convert): ditto.
+
+Thu Aug 14 15:27:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/defines.h (RUBY_ALIAS_FUNCTION): fallback definition.
+
+Thu Aug 14 15:08:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_trans_result_t): new enumeration constant:
+	  transcode_output_followed_by_input.
+
+	* transcode.c (OUTPUT_FOLLOWED_BY_INPUT): new flag.
+	  (transcode_restartable0): suspend when output followed by input if
+	  OUTPUT_FOLLOWED_BY_INPUT is specified.
+	  (trans_sweep): check OUTPUT_FOLLOWED_BY_INPUT.
+	  (rb_trans_conv): support OUTPUT_FOLLOWED_BY_INPUT.
+	  (econv_primitive_convert): return :output_followed_by_input for
+	  transcode_output_followed_by_input.
+
+Thu Aug 14 14:57:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (getrusage_time): should return the value.
+
+	* thread.c (rb_thread_priority_set): get rid of C99 feature.
+
+	* transcode.c (transcode_search_path): ditto.
+
+	* vm.c (REWIND_CFP): get rid of statement expressions.
+
+Thu Aug 14 14:13:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (Init_VM): hide FrozenCore.
+
+Thu Aug 14 12:20:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (load_transcoder): suppress warning.
+
+	* missing/vsnprintf.c (BSD_vfprintf): ditto.
+
+Thu Aug 14 12:01:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (gnumake): check for GNU make.
+
+	* Makefile.in (uncommon.mk): includes filtered common.mk.
+
+	* common.mk (parse.h): workaround for nmake.
+
+Thu Aug 14 06:09:12 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_execute_interrupts): switch event
+	  should be occur only once.
+
+Thu Aug 14 05:36:36 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: parse.h depends on parse.c.
+
+	* common.mk: hash.o depends on $(ID_H_INCLUDES).
+
+Thu Aug 14 02:45:16 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_file_exhaustive.rb (TestFileExhaustive#setup):
+	  set group of tmpdir.  [ruby-dev:35633]
+
+Thu Aug 14 15:50:30 2008  Giuseppe Bilotta  <giuseppe.bilotta@gmail.com>
+
+	* hash.c (set_default): fix rdoc.  #441
+
+Thu Aug 14 02:23:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_convert): add output_byteoffset
+	  argument.
+
+Thu Aug 14 00:43:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_gcc_function_alias): checks alias attribute.
+
+	* string.c (alias_func): replaced with RUBY_ALIAS_FUNCTION.
+
+Wed Aug 13 23:42:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (parse.h): fake rule.
+
+Wed Aug 13 23:15:58 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_thread.rb (test_list): call Thread.pass to run t1.
+	  fixed [ruby-core:18264].
+
+Wed Aug 13 23:05:51 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* string.c (alias_func): changed to 'weak, alias' from 'alias' for
+	  Mac OSX. (closes #429)
+
+Wed Aug 13 19:40:12 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: count only freed object as freelist size.
+
+Wed Aug 13 18:34:22 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* marshal.c (marshal_load): set arg.untrust properly.
+
+Wed Aug 13 17:47:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_inspect): new method.
+
+Wed Aug 13 17:35:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_restartable0): several local variables
+	  removed.
+
+Wed Aug 13 17:35:23 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_set_default_proc): add new method.  a patch from
+	  Giuseppe Bilotta.  #419
+
+Wed Aug 13 17:31:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_insnhelper.c (caller_setup_args): should ensure if the value
+	  from to_proc is a real Proc.
+
+Wed Aug 13 17:28:29 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* common.mk: regint.h no longer includes vm_core.h.
+
+Wed Aug 13 17:26:44 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): add fields: writebuf_off,
+	  writebuf_len and writebuf.
+	  (TRANSCODING_WRITEBUF): new macro.
+
+	* transcode.c (transcode_restartable0): output until output buffer is
+	  really full.
+	  (rb_transcoding_open_by_transcoder): initialize writebuf_len,
+	  writebuf_off and writebuf.
+	  (rb_transcoding_close): finalize writebuf.
+
+Wed Aug 13 17:18:37 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_check_ints): added.  please note that
+	  this function may cause ruby's thread switching.
+
+	* include/ruby/intern.h: ditto.
+
+	* regint.h: use rb_thread_check_ints() instead of
+	  RUBY_CHECK_INTS() directly.
+
+Wed Aug 13 17:01:36 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* golf_prelude.rb, lib/set.rb (TC_Set#test_each),
+	  test/readline/test_readline_history.rb (Readline#test_each__enumerator),
+	  test/ruby/test_array.rb (TestArray#test_collect),
+	  test/ruby/test_enumerator.rb (TestEnumerator#test_initialize):
+	  Enumerable::Enumerator is now called Enumerator.
+
+	* lib/rinda/tuplespace.rb (Rinda::TupleBag#initialize): Use
+	  enum_for instead of hardcoding Enumerable::Enumerator.
+
+Wed Aug 13 16:40:57 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, vm_core.h: add manual priority support
+	  using time slice.  if you enable USE_NATIVE_THREAD_PRIORITY
+	  macro, this mechanism is ignored.  [ruby-dev:33124]
+
+	* thread_pthread.c, thread_win32.c: ditto.
+
+	* test/ruby/test_thread.rb: fix test parameter.
+
+Wed Aug 13 16:02:14 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* object.c (rb_obj_untrusted): new method Object#untrusted?.
+	  (rb_obj_untrust): new method Object#untrust.
+	  (rb_obj_trust): new method Object#trust.
+
+	* array.c, debug.c, time.c, include/ruby/ruby.h, re.c, variable.c,
+	  string.c, io.c, dir.c, vm_method.c, struct.c, class.c, hash.c,
+	  ruby.c, marshal.c: fixes for Object#untrusted?.
+
+	* test/ruby/test_module.rb, test/ruby/test_array.rb,
+	  test/ruby/test_object.rb, test/ruby/test_string.rb,
+	  test/ruby/test_marshal.rb, test/ruby/test_hash.rb: added tests for
+	  Object#untrusted?.
+
+Wed Aug 13 16:13:58 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_m17n.rb: follow EncodingCompatibilityError.
+
+	* test/ruby/test_mixed_unicode_escapes.rb: ditto.
+
+	* test/ruby/enc/test_utf16.rb: ditto.
+
+Wed Aug 13 16:05:50 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_cmp): retry with right hand operand if its not a
+	  time object.  [ruby-dev:35011]
+
+Wed Aug 13 15:51:22 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c: Apply a temporary fix to fix build on OS X.  A real
+	  fix will follow later.
+
+Wed Aug 13 15:25:09 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (Init_Enumerator): Rename Enumerable::Enumerator to
+	  just Enumerator.
+
+Wed Aug 13 15:13:57 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* common.mk (prelude.c, golf_prelude.c): Using $(COMPILE_PRELUDE)
+	  requires $(RBCONFIG).
+
+Wed Aug 13 15:08:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_init): make flags argument optional.
+	  (econv_primitive_convert): ditto.
+
+Wed Aug 13 14:55:27 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* common.mk (realclean-local): Remove parse.h as well.
+
+	* common.mk: Add missing dependencies regarding vm_core.h, id.h,
+	  parse.c (parse.h), etc..
+
+Wed Aug 13 14:45:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/newline.trans (rb_crlf_newline): new transcoder.
+	  (rb_cr_newline): new transcoder.
+
+	* transcode.c (trans_open_i): one more extra room for input newline
+	  converter.
+	  (rb_trans_open): crlf newline and cr newline implemented.
+	  (Init_transcode): Encoding::Converter::CRLF_NEWLINE and
+	  Encoding::Converter::CR_NEWLINE defined.
+
+Wed Aug 13 14:22:16 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/newline.trans: new file.
+
+	* transcode_data.h (rb_trans_t): add last_tc field.
+
+	* transcode.c (UNIVERSAL_NEWLINE): defined.
+	  (CRLF_NEWLINE): defined.
+	  (CR_NEWLINE): defined.
+	  (rb_trans_open_by_transcoder_entries): initialize last_tc.
+	  (trans_open_i): allocate one more room for output newline converter.
+	  (rb_trans_open): universal newline implemented.
+	  (more_output_buffer): take max_output argument instead ts.
+	  (output_replacement_character): take tc argument instead of ts.
+	  (transcode_loop): use last_tc field.
+	  (econv_init): add flags argument for rb_trans_open.
+	  (Init_transcode): Encoding::Converter::UNIVERSAL_NEWLINE defined.
+
+Wed Aug 13 14:00:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (parse.c): generates parse.h together.
+
+	* id.c, id.h: use constants for parser tokens.
+
+Wed Aug 13 13:47:05 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: rewind cfp to show proper backtrace.
+	  [ruby-dev:35820]
+
+Wed Aug 13 13:09:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: fix for clean up.
+
+Wed Aug 13 11:45:51 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* vm_core.h: Declare rb_iseq_clone, which is used in class.c.
+
+Wed Aug 13 11:39:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_search_path): return the length of converter
+	  path.
+	  (load_transcoder_entry): renamed from load_transcoder.
+	  (load_transcoder): new function for loading transcoder by encoding
+	  names.
+	  (rb_transcoding_open_by_transcoder): extracted from
+	  rb_transcoding_open.
+	  (rb_transcoding_open): use load_transcoder and
+	  rb_transcoding_open_by_transcoder.
+	  (rb_trans_open_by_transcoder_entries): new function.
+	  (trans_open_i): construct entries array.
+	  (rb_trans_open): use rb_trans_open_by_transcoder_entries.
+
+Wed Aug 13 10:45:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: try to find exported directory.
+
+Wed Aug 13 10:16:35 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* doc/NEWS: Mention the Readline.vi_editing_mode?,
+	  Readline.emacs_editing_mode?, Readline::HISTORY[] and
+	  Readline::HISTORY.clear change.
+
+Wed Aug 13 08:03:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (econv_primitive_convert): add output_size argument.
+
+Wed Aug 13 07:41:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_trans_conv): report last transcode_obuf_full.
+	  (econv_max_output): new method Encoding::Converter#max_output.
+
+Wed Aug 13 02:46:01 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* error.c (rb_eEncCompatError): add Exception.
+
+	* include/ruby/ruby.h: ditto.
+
+	* encoding.c (rb_enc_check): use rb_eEncCompatError.
+
+	* string.c (rb_enc_cr_str_buf_cat): ditto.
+
+	* string.c (rb_str_sub_bang): ditto.
+
+	* string.c (rb_str_hex): ditto.
+
+	* string.c (rb_str_oct): ditto.
+
+Wed Aug 13 02:36:47 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (str_transcode): fix error message.
+
+Tue Aug 12 23:42:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_cEncodingConverter): new class Encoding::Converter.
+	  (econv_free): new function.
+	  (econv_s_allocate): ditto.
+	  (econv_init): ditto.
+	  (check_econv): ditto.
+	  (econv_primitive_convert): new method.
+	  (Init_transcode): define Encoding::Converter.
+
+Tue Aug 12 23:16:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_splice_0): call rb_str_drop_bytes, not rb_str_drop.
+
+Tue Aug 12 19:11:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: fix for enc.mk.
+
+Tue Aug 12 19:08:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_drop_bytes): new function to drop first bytes.
+
+Tue Aug 12 18:58:48 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, vm_insnhelper.c (vm_define_method): move
+	  function place.
+
+Tue Aug 12 18:56:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c: more descriptive aliases of rb_str_new[2-5].
+	  [ruby-dev:35615]
+
+Tue Aug 12 18:51:29 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* gc.c (getrusage_time): Returned effective value on Windows.
+
+Tue Aug 12 18:51:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_trans_open): free ts before raise.
+
+Tue Aug 12 18:46:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimalCmp): should not ignore
+	  <=> comparison.  [ruby-dev:35732]
+
+Tue Aug 12 18:43:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/make_transdb.rb: *.erb.c is not used anymore.
+
+Tue Aug 12 18:32:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_trans_elem_t): new field: from and to.
+
+	* transcode.c (trans_open_i): just record from and to.
+	  (rb_trans_open): load transcodings.
+
+Tue Aug 12 18:32:03 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/syslog/syslog.c (mSyslog_open): Use of Check_SafeStr() is
+	  deprecated.
+	  (mSyslogConstants_LOG_MASK, mSyslogConstants_LOG_UPTO): Use
+	  NUM2INT() instead of FIX2INT() when the type of an given object
+	  is not determined.
+
+Tue Aug 12 18:14:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* tool/transcode-tblgen.rb (#transcode_tblgen): slight message
+	  improvement.
+
+Tue Aug 12 17:16:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (trans_open_i): check the result of rb_transcoding_open.
+
+Tue Aug 12 16:23:32 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (gc_profile_result): use rb_str_catf.
+
+Tue Aug 12 16:13:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (TRANSCODE_ERROR): removed.
+
+	* tool/transcode-tblgen.rb: 8bit byte of ASCII-8BIT is a valid
+	  (but unique to ASCII-8BIT) character.
+
+	* transcode.c (rb_eConversionUndefined): new error.
+	  (rb_eInvalidByteSequence): ditto.
+
+Tue Aug 12 16:08:51 2008  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/pop.rb: POP3Command should export @socket since POP
+	  calls #socket.
+
+Tue Aug 12 16:06:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c, util.c (quorem, nrv_alloc, dtoa): enabled floating point
+	  support.
+
+Tue Aug 12 15:52:45 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (gc_profile_result): use sprintf.
+
+Tue Aug 12 15:37:40 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_intern): should check symbol table overflow.
+
+Tue Aug 12 15:31:04 2008  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (send_request_with_body): Content-Length should
+	  be byte length of string. (closes #203)
+
+Tue Aug 12 15:17:06 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/depend: (transvpath_prefix): prefix has no extension, so replace
+	  %s with "".
+
+Tue Aug 12 15:03:43 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* test/readline/test_readline.rb (TestReadline#test_safe_level_4):
+	  tested Readline.vi_editing_mode? and Readline.emacs_editing_mode?.
+
+Tue Aug 12 14:25:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/Makefile.in (.SUFFIXES): renamed to .trans.
+
+	* enc/make_encmake.rb: added --encs and --no-encs options.
+
+	* enc/depend (TRANSVPATH): fix for nmake.
+
+Tue Aug 12 14:19:53 2008  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/https.rb: suppress syntax warning.
+
+Tue Aug 12 14:15:51 2008  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (Net::HTTP#initialize): initialize net/https
+	  variables to suppress syntax warning.
+
+Tue Aug 12 14:15:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cptr.c (rb_dlptr_s_to_ptr): fixed shadowing variable.
+
+Tue Aug 12 10:25:14 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_delete): Array#delete to return deleted element.
+
+Tue Aug 12 10:18:01 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/win32ole/test_err_in_callback.rb: do not require 'mkmf'
+	  unless WIN32OLE is defined.
+
+Tue Aug 12 07:41:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoder): add resetstate_func field for
+	  resetting a state of stateful encoding.
+
+	* enc/trans/iso2022.trans (rb_EUC_JP_to_ISO_2022_JP): specify
+	  finish_eucjp_to_iso2022jp for resetstate_func.
+
+	* tool/transcode-tblgen.rb: specify NULL for resetstate_func.
+
+	* transcode.c (output_replacement_character): call resetstate_func
+	  before appending the replacement character.
+
+Tue Aug 12 07:19:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (get_replacement_character): extracted from
+	  output_replacement_character.
+
+Tue Aug 12 07:00:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoder): typedef at first.
+
+Tue Aug 12 06:48:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (rb_trans_conv): find second last error.
+
+Tue Aug 12 00:43:44 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_trans_result_t): new type.
+	  (rb_trans_elem_t): new type.
+	  (rb_trans_t): new type.
+
+	* transcode.c (transcode_dispatch_cb): removed.
+	  (transcode_dispatch): removed.
+	  (rb_transcoding_result_t): moved to rb_trans_result_t in
+	  transcode_data.h.
+	  (transcode_restartable0): goto follow_info when FUNsi.
+	  (rb_transcoding_open): use get_transcoder_entry.
+	  (rb_trans_open): new function.
+	  (rb_trans_conv): ditto.
+	  (rb_trans_close): ditto.
+	  (trans_open_i): ditto.
+	  (trans_sweep): ditto.
+	  (more_output_buffer): take rb_trans_t instead of rb_transcoding as
+	  an argument.
+	  (transcode_loop): take from_encoding and to_encoding instead of tr
+	  as arguments.  use rb_trans_open/rb_trans_conv/rb_trans_close.
+	  (str_transcode): don't use transcode_dispatch.
+
+Mon Aug 11 20:37:05 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* iseq.c (rb_iseq_clone): should preserve cref_stack link.
+
+Mon Aug 11 20:27:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): add taint check.
+
+	* ext/dl/dl.c (rb_dl_malloc): add rb_secure(2).
+
+	* ext/dl/dl.c (rb_dl_realloc): ditto.
+
+	* ext/dl/dl.c (rb_dl_free): ditto.
+
+	* ext/dl/dl.c (rb_dl_ptr2value): ditto.
+
+Mon Aug 11 20:11:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (getrusage_time): works only if RUSAGE_SELF is defined right
+	  now.
+
+	* gc.c (gc_profile_result): C99.
+
+Mon Aug 11 18:57:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_sample): rename #choice to #sample.  in
+	  addition, sample takes optional argument, a la #first.
+
+Mon Aug 11 18:28:02 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c: added GC::Profiler.
+
+Mon Aug 11 17:26:16 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/README.ja: added API document for
+	  Readline.vi_editing_mode? and Readline.emacs_editing_mode?.
+
+	* ext/readline/extconf.rb: checked rl_editing_mode variable in
+	  Readline library.
+
+	* ext/readline/readline.c (readline_s_emacs_editing_mode_p): added
+	  Readline.emacs_editing_mode? method.
+	  (readline_s_vi_editing_mode_p): added Readline.vi_editing_mode?
+	  method.
+	  (Init_readline): ditto.
+
+Mon Aug 11 16:56:40 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* test/readline/test_readline.rb: added test for Readline's class
+	  methods.
+
+Mon Aug 11 16:39:23 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c (clone_method): should copy cbase in cref as well.
+	  [ruby-dev:35116]
+
+	* iseq.c (iseq_mark): mark original iseq object.
+
+	* iseq.c (iseq_free): do not free internal data if they have
+	  original iseq to belong.
+
+	* iseq.c (rb_iseq_clone): a new function to clone iseq value.
+
+Mon Aug 11 16:34:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/iso2022.trans: renamed from iso2022.erb.c.
+
+	* enc/trans/single_byte.trans: ditto.
+
+	* enc/trans/utf_16_32.trans: ditto.
+
+	* enc/trans/korean.trans: ditto.
+
+	* enc/trans/japanese.trans: ditto.
+
+	* enc/depend: follow the renaming.
+
+	* tool/build-transcode: ditto.
+
+Mon Aug 11 15:09:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (rb_cv_broken_glibc_ia64_erfc): renamed from
+	  rb_broken_glibc_ia64_erfc.
+	  [ruby-core:18228]
+
+Mon Aug 11 10:04:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): rename fields.
+	  readlen -> recognized_len.
+	  feedlen -> readagain_len.
+
+	* transcode.c: follow the field renaming.
+
+Sun Aug 10 22:34:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_restartable0): invalid handling simplified.
+	  (transcode_restartable): use PARTIAL_INPUT for converting buffered
+	  input.
+
+Sun Aug 10 22:24:24 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* encoding.c: include util.h.  [ruby-dev:35715]
+
+	* common.mk: ditto.
+
+Sun Aug 10 20:59:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): new field: flags.
+
+	* transcode.c (load_transcoder): extracted from transcode_dispatch_cb.
+	  (rb_transcoding_result_t): renamed from transcode_result_t.
+	  (rb_transcoding_open): new function.
+	  (rb_transcoding_convert): ditto.
+	  (rb_transcoding_close): ditto.
+	  (transcode_loop): use rb_transcoding_open, rb_transcoding_convert
+	  and rb_transcoding_close.
+	  (str_transcode): don't need rb_transcoding.
+
+Sun Aug 10 18:09:16 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: record checksum of
+	  transcode-tblgen.rb itself.
+
+Sun Aug 10 12:34:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* config.guess ({powerpc-apple,i586-pc}-haiku): re-applied r18403.
+	  [ruby-core:18110]
+
+	* config.{guess,sub}: updated to automake-1.10.1.
+
+Sun Aug 10 11:45:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/Makefile.in (make-workdir): use MAKEDIRS.
+
+	* enc/depend: makes target directory before compile/link.
+
+	* tool/transcode-tblgen.rb: creates target directory.
+
+Sun Aug 10 11:30:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c: rename my_transcoding to tc and my_transcoder to tr.
+
+Sun Aug 10 11:15:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): add feedlen field.
+
+	* transcode.c (transcode_restartable0): renamed from
+	  transcode_restartable.
+	  save input buffer into feed buffer if next character is started the
+	  point before input buffer.  for example, "\x00\xd8\x01" then "\x02"
+	  in UTF-16LE.  \x02 causes invalid and next character is started from
+	  \x01.
+	  (transcode_restartable): new function to call
+	  transcode_restartable0.  if feed buffer is not empty, convert it at
+	  first.
+
+Sun Aug 10 11:02:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (extconf): use MAKEDIRS.
+
+Sun Aug 10 09:35:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb.rb (IRB::Irb#inspect): instance_variables returns symbols
+	  instead of strings now.  [ruby-dev:34930]
+
+Sun Aug 10 09:22:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_char_start): refactored to remove readlen
+	  argument.
+	  (transcode_restartable): refactored to remove readlen variable.
+
+Sun Aug 10 08:56:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (STACK_LEVEL_MAX, ruby_stack_length): returns size_t.
+	  [ruby-core:18207]
+
+	* configure.in (rb_cv_missing__dtos18, rb_cv_missing_fconvert),
+	  ext/digest/md5/md5.c, ext/openssl/ossl.c (main),
+	  ext/socket/extconf.rb (ipv6, wide-getaddrinfo): main should be int.
+
+	* main.c (main), win32/winmain.c, wince/wincemain.c (WinMain): envp is
+	  no longer used so long time.  based on a patch from Peter Bowen at
+	  [ruby-core:18208]
+
+Sat Aug  9 22:05:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_loop): take destination and resize function
+	  as parameters.
+	  (more_output_buffer): ditto.
+	  (str_transcoding_resize): argument changed from rb_transcoding* to
+	  VALUE.
+	  (str_transcode): call transcode_loop with destination string and its
+	  resize function.
+
+	* transcode_data.h (rb_transcoding): move ruby_string_dest and
+	  flush_func to transcode_loop parameters.
+
+Sat Aug  9 21:29:45 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: encs depend on transdb.h
+
+Sat Aug  9 21:10:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_rshift_sign, rb_cv_binary_elf): get rid of
+	  AC_TRY_RUN.
+
+Sat Aug  9 16:33:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): new fields: next_info and next_byte.
+
+	* transcode.c (transcode_restartable): save/restore next_info and
+	  next_byte.
+	  sync readlen and in_p when invalid.
+
+Sat Aug  9 15:10:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode.c (transcode_restartable): my_transcoder argument removed.
+	  (transcode_loop): my_transcoder argument removed.
+
+Sat Aug  9 14:39:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): add fields for restartable
+	  transcoding.
+	  (rb_transcoder): add max_input field.
+	  from_unit_length field is renamed to input_unit_length.
+
+	* tool/transcode-tblgen.rb: generate max_input field.
+
+	* enc/trans/iso2022.erb.c: follow rb_transcoder change.
+
+	* enc/trans/utf_16_32.erb.c: ditto.
+
+	* transcode.c (PARTIAL_INPUT): new constant.
+	  (transcode_char_start): new function.
+	  (transcode_result_t): new type.
+	  (transcode_restartable): new function.
+	  (more_output_buffer): new function.
+	  (transcode_loop): use transcode_restartable.
+
+Sat Aug  9 13:35:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* stable/ext/socket/socket.c (NI_MAXHOST, NI_MAXSERV): fixed invalid
+	  preprocessor directives.  a patch from Peter Bowen at
+	  [ruby-core:18211].
+
+Sat Aug  9 06:37:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/make_encdb.rb, enc/trans/make_transdb.rb: skip nonexistent
+	  directory.  [ruby-dev:35802]
+
+Sat Aug  9 01:07:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/trans/utf_16_32.erb.c (fun_so_from_utf_32le): implemented.
+	  (fun_so_to_utf_32le): implemented.
+	  [ruby-dev:35777]
+
+Sat Aug  9 00:42:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoder): from_unit_length field added.
+	  from_utf8 field removed.
+
+	* tool/transcode-tblgen.rb: generate offsets range.
+	  follow rb_transcoder change.
+
+	* transcode.c (transcode_loop): don't use from_utf8.
+	  make invalid region from_unit_length wise.
+
+	* enc/trans/iso2022.erb.c: follow rb_transcoder and
+	  transcode_generate_node change.
+
+	* enc/trans/utf_16_32.erb.c: follow rb_transcoder and
+	  transcode_generate_node change.
+	  explicit :invalid map removed.
+
+Fri Aug  8 23:29:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (TRANSCSRCS): needs rule_subst to apply.
+
+	* io.c (shutdown): not define if defined already.
+
+Fri Aug  8 22:47:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (opt_eq_func): large function to make inline.
+
+	* bcc32/Makefile.sub (config.h): cannot compile a big inline function.
+
+Fri Aug  8 21:09:10 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (enk.mk): give not RUBY but MINIRUBY.
+
+Fri Aug  8 17:03:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (shutdown, sched_yield, pthread_attr_setinheritsched):
+	  check for Haiku.
+
+	* eval_intern.h, io.c, thread_pthread.c: use autoconfisticated results.
+
+Fri Aug  8 16:52:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (enc.mk): mkmf.rb requires rbconfig.rb.
+
+	* common.mk (srcs-enc): renamed from transcodes.
+
+	* enc/Makefile.in (make-workdir): creates object directories.
+
+	* common.mk (encdb.h): see both $(srcdir)/enc and enc.
+
+	* enc/make_encdb.rb: ditto.
+
+	* enc/trans/make_transdb.rb: fix for the case no transdirs are given.
+
+	* enc/trans/make_transdb.rb: converts only one transcoders for each
+	  basename.
+
+Fri Aug  8 10:53:52 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: randomize source port and transaction id.
+	  CVE-2008-1447.
+
+	* lib/resolv-replace.rb (UDPSocket#bind): don't resolv host if host is
+	  "".
+
+Fri Aug  8 04:20:14 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: see both $(srcdir)/enc/trans and enc/trans.
+
+	* enc/trans/make_transdb.rb: ditto.
+
+Fri Aug  8 00:05:02 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/.document: added readline/readline.c.
+
+	* ext/readline/readline.c: changed Copyright. added RDoc.
+
+	* ext/readline/README.ja: fixed typo.
+
+	* ext/readline/README: contents was moved RDoc.
+
+Thu Aug  7 23:43:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* transcode_data.h (rb_transcoding): new field "stateful".
+	  (rb_transcoder): preprocessor and postprocessor field removed.
+	  change arguments of func_ii, func_si, func_io and func_so.
+	  new field "finish_func".
+
+	* tool/transcode-tblgen.rb: make FUNii, FUNsi and FUNio
+	  generatable.
+
+	* transcode.c (transcoder_lib_table): removed.
+	  (transcoder_table): change structure.
+	  (transcoder_key): removed because the above structure change.
+	  (make_transcoder_entry): new function.
+	  (get_transcoder_entry): ditto.
+	  (rb_register_transcoder): follow the structure change.
+	  (declare_transcoder): ditto.
+	  (transcode_search_path): new function for breadth first search to
+	  find a list of converters.
+	  (transcode_search_path_i): new function.
+	  (transcode_dispatch_cb): ditto.
+	  (transcode_dispatch): use transcode_search_path.
+	  (transcode_loop): follow the argument change.
+	  (str_transcode): preprocessor and postprocessor stuff removed.
+
+	* enc/trans/iso2022.erb.c: new file.  ISO-2022-JP conversion
+	  re-implemented.
+
+	* enc/trans/japanese.erb.c: ISO-2022-JP stuff removed.
+
+	* enc/trans/utf_16_32.erb.c: follow argument change of FUNso.
+
+	[ruby-dev:35798]
+
+Thu Aug  7 22:55:44 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/README.ja: updated API document for Readline module.
+
+Thu Aug  7 20:52:08 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (login): raise FTPReplyError if passwd or acct
+	  is not supplied.  fixed [ruby-core:18058].
+
+Thu Aug  7 18:01:44 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-mode.el (ruby-imenu-create-index-in-block): Fix the
+	  regexp to only pick definition lines properly.  `module_function'
+	  is not a definition of a module named `_function'.
+
+Thu Aug  7 17:47:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend: add transdb.c.
+
+Thu Aug  7 16:28:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.mk, {bcc,win}32/Makefile.sub (clean-ext): do not
+	  use miniruby.
+
+Thu Aug  7 14:17:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend: removed needless explicit commands.
+
+	* lib/mkmf.rb, {bcc32,win32}/Makefile.sub (CLEANLIBS, CLEANOBJS):
+	  moved clean targets to platform makefiles.
+
+Thu Aug  7 13:12:30 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/getaddrinfo.c (gai_strerror): ignore only on Haiku.
+	  a patch from <zn at mbf.nifty.com>, [ruby-dev:35796].
+
+Thu Aug  7 06:31:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/process.rb (test_popen_fork): skip a test which is
+	  freeze in FreeBSD.
+
+Thu Aug  7 06:05:48 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* config.guess: add powerpc-apple-haiku, i586-pc-haiku.
+	  see [ruby-core:18110]
+
+Thu Aug  7 05:51:05 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* eval_intern.h: Add support to Haiku. see [ruby-core:18110]
+
+	* include/ruby/defines.h: ditto.
+
+	* configure.in: ditto.
+
+	* thread_pthread.c: ditto.
+
+	* io.c: ditto.
+
+	* lib/mkmf.rb: ditto.
+
+	* ext/socket/getaddrinfo.c: ditto.
+
+	* ext/socket/extconf.rb: ditto.
+
+	* ext/socket/socket.c: ditto.
+
+	* ext/socket/addrinfo.h: ditto.
+
+	* ext/socket/getnameinfo.c: ditto.
+
+Thu Aug  7 05:43:32 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: mkdir enc/trans before build tables.
+
+Thu Aug  7 05:18:30 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/depend: enc/*.c is source but enc/trans/*.c is generated.
+
+Thu Aug  7 05:14:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regenc.c (code_to_mbc): raise RangeError Integer#chr
+	  when more than 255 is given with single byte encoding.
+	  [ruby-dev:35789]
+
+Thu Aug  7 05:06:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (run_final): runs finalizers with the object terminated.
+
+	* gc.c (rb_gc_call_finalizer_at_exit): keeps finalized objects from
+	  linking to freelist.
+
+Thu Aug  7 04:43:14 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/depend: for build in other than srcdir.
+
+Wed Aug  6 22:40:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (transcode_generate_node): code
+	  argument removed.
+
+Wed Aug  6 21:25:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (chain_finalized_object): deletes finalizers to be invoked from
+	  finalizer_table.
+
+	* gc.c (rb_gc_call_finalizer_at_exit): warns when could not invoke
+	  finalizers.
+
+Wed Aug  6 20:56:43 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/depend: transcode table generation depends on
+	  tool/transcode-tblgen.rb.
+
+Wed Aug  6 20:48:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_call_finalizer_at_exit): self-referencing finalizers
+	  cannot be invoked.  [ruby-dev:35681]
+
+Wed Aug  6 20:44:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: distinguish UNDEF and INVALID.
+	  [ruby-dev:35709]
+
+	* transcode.c (transcode_loop): don't need rb_enc_mbclen now.
+
+Wed Aug  6 14:40:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (transdb.h): requires transcoders.
+
+	* enc/depend (srcs): target for transcoders.
+
+Wed Aug  6 14:04:08 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/depend: replace not only $(<:...) but also $<.
+
+Wed Aug  6 13:54:30 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): export BASERUBY.
+
+	* enc/depend: avoid GNU make'ism.
+
+Wed Aug  6 07:59:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb (ActionMap#eql?): use == to compare @map.
+
+Wed Aug  6 07:45:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/transcode-tblgen.rb: show generating tables in verbose mode.
+	  (transcode_generate_node): call ActionMap#generate_node with showing
+	  table name.
+
+	* enc/trans/utf_16_32.erb.c: use transcode_generate_node.
+
+Wed Aug  6 06:55:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): propagates fatal error and system
+	  exit to the main thread.
+
+Wed Aug  6 05:31:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_to_encoding_index, rb_to_encoding): check if the name
+	  is ascii compatible as well as Encoding.find.
+
+	* transcode.c (str_encode): no need to duplicate first.
+
+Wed Aug  6 05:08:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): reset to real class.
+
+	* file.c (rb_find_file_ext, rb_find_file): ditto.
+
+	* io.c (io_reopen): ditto.
+
+Wed Aug  6 03:56:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (transcodes), tool/build-transcode: generates transcode
+	  sources.
+
+	* enc/trans/{japanese,korean,single_byte,utf_16_32}.c: to be
+	  autogenerated now.
+
+	* enc/depend: added rules for .c from .erb.c.
+
+Tue Aug  5 20:46:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/build-transcode: new file.
+
+	* tool/transcode-tblgen.rb: new file.
+
+	* enc/trans/make_transdb.rb: exclude *.erb.c.
+
+	* enc/depend: exclude *.erb.c.
+
+	* enc/trans/utf_16_32.erb.c: new file.
+
+	* enc/trans/single_byte.erb.c: new file.
+
+	* enc/trans/japanese.erb.c: new file.
+
+	* enc/trans/korean.erb.c: new file.
+
+	* enc/trans/iso-8859-2-tbl.rb: new file.
+
+	* enc/trans/iso-8859-3-tbl.rb: new file.
+
+	* enc/trans/iso-8859-4-tbl.rb: new file.
+
+	* enc/trans/iso-8859-5-tbl.rb: new file.
+
+	* enc/trans/iso-8859-6-tbl.rb: new file.
+
+	* enc/trans/iso-8859-7-tbl.rb: new file.
+
+	* enc/trans/iso-8859-8-tbl.rb: new file.
+
+	* enc/trans/iso-8859-9-tbl.rb: new file.
+
+	* enc/trans/iso-8859-10-tbl.rb: new file.
+
+	* enc/trans/iso-8859-11-tbl.rb: new file.
+
+	* enc/trans/iso-8859-13-tbl.rb: new file.
+
+	* enc/trans/iso-8859-14-tbl.rb: new file.
+
+	* enc/trans/iso-8859-15-tbl.rb: new file.
+
+	* enc/trans/eucjp-tbl.rb: new file.
+
+	* enc/trans/sjis-tbl.rb: new file.
+
+	* enc/trans/euckr-tbl.rb: new file.
+
+	* enc/trans/utf_16_32.c: regenerated.
+
+	* enc/trans/single_byte.c: regenerated.
+
+	* enc/trans/japanese.c: regenerated.
+
+	* enc/trans/korean.c: regenerated.
+
+	[ruby-dev:35730]
+
+Tue Aug  5 18:02:53 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/io/nonblock/test_flush.rb (TestIONonblock#test_flush):
+	  rescue some exceptions. [ruby-dev:35638]
+
+Tue Aug  5 16:40:06 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (chdir): handle 5xx errors correctly.
+	  fixed [ruby-core:18057].
+
+Tue Aug  5 16:38:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h, bcc32/Makefile.sub (config.h): bcc 5.8 has
+	  stdint.h.
+
+Tue Aug  5 16:13:05 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (disconnect): do not refer to SSL::SSLSocket for
+	  environments without OpenSSL. fixed [ruby-dev:35755].
+
+Tue Aug  5 14:19:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_stat_mode): generalized st_mode mask.
+
+Tue Aug  5 12:43:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (retry_sendfile, retry_read): ENOSYS and EWOULDBLOCK are not
+	  defined on every platforms.
+
+Tue Aug  5 12:34:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode_data.h (TRANSCODE_ERROR): common transcode failure
+	  exception, would be changed later.
+
+	* enc/trans/japanese.c (UNSUPPORTED_MODE): unsupported mode transition
+	  exception.
+
+Tue Aug  5 03:29:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): respect overridden <=> for String and
+	  Fixnum.  [ruby-core:17708]
+
+	* include/ruby/node.h (NOEX_BASIC): basic definition method flag.
+
+	* include/ruby/intern.h, vm_method.c (rb_method_basic_definition_p):
+	  new function to check if the method is not redefined after the
+	  initialization.
+
+	* vm_method.c (rb_obj_respond_to): use rb_method_basic_definition_p.
+
+Mon Aug  4 20:39:06 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* Makefile.in (update-rubyspec): renamed the rubyspec directory
+	  "spec". Changed directory structure.
+
+	* Makefile.in (test-rubyspec): ditto.
+
+	* spec/README: described the structure of "spec" directory.
+
+	* spec/default.mspec: configured for Matz's Ruby Implementation.
+
+Mon Aug  4 19:21:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_tmp_new): added.
+
+	* vm_eval.c (vm_call_super): fixed typo, and get rid of too large
+	  alloca.  [ruby-core:17922]
+
+Mon Aug  4 16:48:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (deferred_nodes, compstmt, arg, fixup_nodes, range_op): fix
+	  up fixnum range literal in conditional as automagical line number
+	  comparison.  [ruby-core:12124], [ruby-dev:35731]
+
+Mon Aug  4 14:41:25 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* lib/net/smtp.rb (Net::SMTP::rcptto): fix a typo. a patch from
+	  Masao Takaku <masao at nii.ac.jp>
+	  fix [ruby-dev:35489].
+
+Mon Aug  4 14:08:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (dump_ensure), process.c (run_exec_dup2),
+	  string.c (rb_str_replace), transcode.c (transcode_dispatch): fixed
+	  memory leaks.  based on patches from shinichiro.h <shinichiro.hamaji
+	  AT gmail.com> at [ruby-dev:35751].
+
+Sun Aug  3 19:32:52 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (hash2named_arg): refactoring.
+
+	* ext/win32ole/win32ole.c (ole_invoke, fole_respond_to, ev_on_event,
+	  fev_off_event): accepts Symbol argument.
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+	* test/win32ole/test_win32ole_event.rb: ditto.
+
+Sun Aug  3 10:41:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_shared_replace): fixed memory leak.  a patch from
+	  shinichiro.h <shinichiro.hamaji AT gmail.com> at [ruby-dev:35742]
+
+Sat Aug  2 22:55:41 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_err_in_callback.rb: remove temporary files.
+	  some refactoring.
+
+Sat Aug  2 15:51:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (yylex): 8 and 9 in octal integer should cause compile
+	  error.  [ruby-dev:35729]
+
+Sat Aug  2 01:06:10 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.c: add U+FF5E to EUC-JP.
+	  [ruby-dev:35720] [ruby-dev:35722]
+
+Fri Aug  1 23:49:44 2008  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/extconf.rb: checked to have clear_history in
+	  readline library.
+	* ext/readline/readline.c (hist_get, hist_each, Init_readline):
+	  The offset specified for the argument of history_get() might be
+	  different in GNU Readline and libedit. If use libedit, it was
+	  corrected that the computational method of the offset specified
+	  for the argument of history_get() when the Readline module was
+	  initialized was decided.
+	  (hist_get, hist_set): If use libedit, accesses first an input
+	  content in history when specifies the negative offset for the
+	  argument of history_get() or replace_history_entry(). Then
+	  checks the offset is negative in ruby.
+	  (rb_remove_history): When compiling, it corrects it to warning
+	  when libedit is used.
+	  (hist_clear, Init_readline): added Readline::HISTORY.clear
+	  method. [ruby-dev:35551]
+	* test/readline/test_readline_history.rb: added unit test for
+	  Readline::HISTORY.
+
+Fri Aug  1 23:26:45 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (transcode_loop): undefined character is replaced with
+	  only one character. [ruby-dev:35709]
+
+Fri Aug  1 23:26:22 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	Merged r16430(akr), r16431(akr), r16433(akr), r16469(nobu), and
+	r17874(akr) from ruby_1_8.
+
+	* Makefile.in (update-rubyspec): added
+
+	* Makefile.in (test-rubyspec): added
+
+Fri Aug  1 23:16:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.c: add support for CP51932,
+
+Fri Aug  1 22:59:40 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.c: add U+FF0C,
+
+Fri Aug  1 21:49:41 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/sdbm/test_sdbm.rb: add some tests.
+
+Fri Aug  1 21:36:00 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gc.c (allocate_heaps, assign_heap_slot, rb_newobj_from_heap):
+	  reset during_gc before raising NoMemoryError.
+
+Fri Aug  1 21:29:56 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (Init_BareVM): check failure of malloc().
+
+Fri Aug  1 20:55:27 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.c (to_SHIFT_JIS_EF_BF_offsets): add U+FFF3,
+	  U+FFF4, U+FFF5.
+
+	* enc/trans/japanese.c (to_SHIFT_JIS_EF_BF_infos): ditto.
+
+	* enc/trans/japanese.c (to_EUC_JP_EF_BF_infos): added.
+
+	* enc/trans/japanese.c (to_EUC_JP_EF_BF): added.
+
+	* enc/trans/japanese.c (to_EUC_JP_EF_infos): change size.
+	  [ruby-dev:35714]
+
+Fri Aug  1 18:27:15 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* parse.y (parser_yylex): removed an useless conditional, and magic
+	  comment are ignored unless at the first of line.
+
+	* test/ruby/test_m17n.rb (test_magic_comment_vim): added.
+
+	* test/ruby/test_m17n.rb (test_magic_comment_at_various_positions):
+	  added.
+
+Fri Aug  1 14:54:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_seekdir): no need to rewind to seek forward.
+
+Fri Aug  1 05:31:08 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (output_replacement_character):
+	  rename from _get_replacement_character.
+
+	* transcode.c (output_replacement_character):
+	  fix replacement on UTF-32{BE,LE}. [ruby-dev:35705]
+
+	* transcode.c (transcode_loop): ditto.
+
+	* test/ruby/test_transcode.rb (test_invalid_replace):
+	  add for above.
+
+Fri Aug  1 01:01:49 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (rb_proc_call_with_block): reduce comparison.
+
+Thu Jul 31 22:17:35 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpMult): prevent memory leak.
+
+Thu Jul 31 20:05:56 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_transcode.rb (test_unicode_public_review_issue_121):
+	  fix option1 and 3.
+
+Thu Jul 31 19:54:57 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (get_replacement_character): use U+FFFD as replacement
+	  character when convert to Unicode.
+
+	* test/ruby/test_transcode.rb (test_unicode_public_review_issue_121):
+	  rename from test_public_review_issue_121.
+
+	* test/ruby/test_transcode.rb (test_unicode_public_review_issue_121):
+	  enable option2.
+
+Thu Jul 31 17:00:10 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (get_replacement_character): fix: invalid byte sequence
+	  is always replaced "\x00?".
+
+Thu Jul 31 16:37:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rubygems/test_gem_ext_configure_builder.rb
+	  (test_self_build_fail): remove extra newline. [ruby-dev:35704]
+
+Thu Jul 31 15:11:11 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb: added test_shift_jis
+	  (contributed by Yoshihiro Kambayashi) and
+	  test_public_review_issue_121
+	  (see http://www.unicode.org/review/pr-121.html)
+
+Thu Jul 31 13:18:30 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* include/ruby/ruby.h (struct RString): size of ary must be
+	  RSTRING_EMBED_LEN_MAX + 1.
+
+Thu Jul 31 12:23:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_unpack): upper half of hexdigits has never been used.
+
+Thu Jul 31 11:31:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/syck.h (ASSERT): fix typo at r18176.
+
+	* ext/syck/rubyext.c (rb_syck_compile): expression in ASSERT() has no
+	  effect unless debug mode.
+
+Thu Jul 31 10:51:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (ruby_iseq_disasm_insn): suppress warnings on platforms which
+	  int size differs from pointer size.
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_get_asn1type): ditto
+
+	* ext/syck/rubyext.c (rb_syck_err_handler),
+	  (syck_default_error_handler): ditto.
+
+Thu Jul 31 02:21:39 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/syck/rubyext.c (rb_syck_compile): remove meaningless branch
+	  that misleads into thinking that the variable sav may be
+	  uninitialized.
+
+Thu Jul 31 01:25:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_mutex_unlock_all): mutex is no longer a ruby object.
+
+Thu Jul 31 01:18:07 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (magic_comment_encoding): remove meaningless null check.
+
+Thu Jul 31 01:09:28 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpIsRoundMode): fix tautology
+	  condition.
+
+Thu Jul 31 00:58:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_unpack): reduced static variables.
+
+Thu Jul 31 00:10:20 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (rb_proc_call_with_block): prevent null reference.
+
+Wed Jul 30 22:19:13 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (vtable_free): remove meaningless null check.
+
+Wed Jul 30 22:08:25 2008  Tanaka Akira  <akr@fsij.org>
+
+	* dir.c (struct dir_data): change path field char * to VALUE.
+	  (mark_dir): new function for mark path field.
+	  (free_dir): follow the path field change.
+	  (dir_s_alloc): ditto.
+	  (dir_initialize): ditto.
+	  (dir_s_open): ditto.
+	  (dir_inspect): ditto.
+	  (dir_path): return (duplicate of) the path field to preserve
+	  encoding.  [ruby-dev:35685]
+
+Wed Jul 30 22:06:56 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (sym_inspect): remove dead code.
+
+Wed Jul 30 21:32:52 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ChangeLog: fix wrong method name and add reference.
+
+Wed Jul 30 21:30:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* regparse.c (name_add): fix memory leak.
+
+Wed Jul 30 21:08:06 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/syck/syck.c (syck_free_parser): fix memory leak by
+	  YAML::Syck.compile.
+
+	* regparse.c (parse_exp): fix memory leak at Regexp.new("x{1,1}").
+
+Wed Jul 30 17:48:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_select): recalc the rest of timeout for each
+	  iterations.  [ruby-core:18015]
+
+Tue Jul 29 23:37:37 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (io_ungetc): raise NotImplementedError when ungetc is called
+	  against dummy encoding IO.  [ruby-dev:35686]
+
+	* io.c (rb_io_getline_1): ditto when gets with delimiter is called.
+
+	* io.c (io_getc): ditto when getc is called.
+
+	* test/ruby/test_io_m17n.rb (test_terminator_stateful_conversion,
+	  test_getc_stateful_conversion, test_ungetc_stateful_conversion):
+	  these tests should raise NotImplementedError.
+
+Tue Jul 29 22:55:34 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_io.rb (pipe): run reader thread and writer thread.
+
+Tue Jul 29 21:38:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (Init_bigdecimal): fix typo.
+
+Tue Jul 29 21:35:59 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/etc/test_etc.rb (test_passwd): age field may be string under
+	  some environments.
+
+Tue Jul 29 17:54:35 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (char_casecmp): fix: return 0 if either of characters is NUL.
+
+Tue Jul 29 13:17:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/etc/test_etc.rb (test_getpwuid): fix for users whose uid is
+	  shared.
+
+Tue Jul 29 05:37:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext, rb_find_file): explicit relative path
+	  which starts with "./" or "../" should be searched from cwd
+	  instead of load path.  [ruby-dev:35673]
+
+Tue Jul 29 02:39:46 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* math.c (math_atanh): raise EDOM on FreeBSD when atanh(1).
+
+	* math.c (math_log): ditto.
+
+	* math.c (math_log2): ditto.
+
+	* math.c (math_log10): ditto.
+
+	* test/ruby/test_math.rb: test for above.
+
+Tue Jul 29 01:41:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* dir.c (struct dir_data): intenc field removed.
+	  (dir_s_alloc): intenc initialization removed.
+	  (dir_initialize): :internal_encoding option removed.  dirname code
+	  conversion removed.
+	  (dir_enc_str): code conversion removed.
+	  [ruby-dev:35661]
+
+Mon Jul 28 21:32:17 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/: use PNG instead of zlib as binary data.
+
+Mon Jul 28 21:24:33 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.[ch] (cond_every_entry, rb_thread_cond_struct): reverted
+	  r18239 because r18245 made the changes unnecessary.
+
+	* thread.c (rb_mutex_struct): define after including thread_{pthread,
+	  win32}.c.
+
+Mon Jul 28 21:00:10 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_require.rb (test_require_too_long_filename):
+	  Kernel#require does not use dln_find_file_r (at r18242).
+
+Mon Jul 28 20:17:03 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_core.h: move the definition of struct rb_mutex_struct.
+
+	* thread.c: ditto.
+
+Mon Jul 28 18:58:46 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (mutex_unlock): fix typo.
+
+Mon Jul 28 18:15:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext, rb_find_file): not to split load path with
+	  path separator.  [ruby-Bugs-21356]
+
+Mon Jul 28 18:14:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (overlapped_socket_io, fcntl, rb_w32_close): must not
+	  pass a pointer to int which is smaller than st_data_t on mswin64.
+
+Mon Jul 28 16:49:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (CreateChild, overlapped_socket_io): suppress
+	  warnings.
+
+Mon Jul 28 16:06:36 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (MAXPATHLEN): define before use.
+
+Mon Jul 28 16:01:12 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.[ch] (cond_every_entry, rb_thread_cond_struct): moved
+	  the definitions from .c to .h because rb_thread_cond_struct is used
+	  in vm_core.h.
+
+Mon Jul 28 14:29:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (load_lib): use dln_find_file_r instead of dln_find_file.
+
+Mon Jul 28 00:18:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_core.h, thread.c: It is now prohibited to use Data_Get_Struct in
+	  *_free against an object that is going to be free'ed.  So, change type
+	  of thread_t#keeping_mutexes from VALUE to mutex_t.
+
+	* vm.c: remove mark to keeping_mutexes.
+
+Sun Jul 27 23:32:42 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/openssl/test_ssl.rb (server_loop): rescue Errno::EINVAL and
+	  Errno::ECONNABORTED.
+
+Sun Jul 27 22:11:57 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bootstraptests/method.rb: increase RLIMIT_STACK size to 4M+8Kbytes
+	  because FreeBSD fails this less than that.
+
+Sun Jul 27 21:45:59 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_children, obj_free): T_DEFERRED should not be appear.
+
+	* gc.c (gc_sweep, finalize_list): fix to decrement heap_slot#limit
+	  after executing finalizer.
+
+Sun Jul 27 14:48:37 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: add a type T_DEFERRED.
+
+	* gc.c: fix deferred finalizer system.  finalize processes of
+	  T_DATA and T_FILE are executed after gc process.
+	  And fix to use BUILTIN_TYPE() instead of seeing flag.
+
+	* thread.c, vm_core.h: add RUBY_VM_SET_FINALIZER_INTERRUPT()
+	  and check interrupt_flag at rb_thread_execute_interrupts().
+
+	* thread.c (mutex_mark): fix to mark next_mutex.
+
+	* vm.c (rb_thread_mark): fix to mark keeping_mutexes.
+
+Sun Jul 27 09:15:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.h (dln_find_exe, dln_find_file): deprecated, use reentrant
+	  versions instead.
+
+Sun Jul 27 09:02:32 2008  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/tuplespace.rb: merged from 1.8.
+
+	* test/rinda/test_rinda.rb: merged from 1.8.
+
+Sat Jul 26 22:45:18 2008  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* sample/exyacc.rb: fixed NoMethodError(Kernel#sub!).
+	  replaced use of special variables with explicit IO
+	  operations.
+
+Sat Jul 26 21:17:18 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (Init_win32ole): add
+	  WIN32OLE_EVENT#handler=, WIN32OLE_EVENT#handler
+
+	* test/win32ole/test_win32ole_event.rb: ditto.
+
+Sat Jul 26 07:44:14 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (add_event_call_back): remove unused
+	  variable.
+
+Fri Jul 25 23:48:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_sweep, obj_free, run_final): defer finalizers of IO and
+	  Data.  [ruby-dev:35578]
+
+Fri Jul 25 23:35:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils#split_header_value):
+	  reduce backtrack.  based on a fix by Christian Neukirchen
+	  <chneukirchen AT gmail.com>.
+
+Fri Jul 25 21:55:38 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/enc/test_koi8.rb: move from test/ruby/test_koi8.rb.
+
+Fri Jul 25 21:09:32 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke, add_event_callback,
+	  rescue_callback): refactoring.
+
+Fri Jul 25 20:52:44 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/err_in_callback.rb: add test of raising
+	  exception in WIN32OLE_EVENT callback.
+
+	* test/win32ole/test_err_in_callback.rb: ditto.
+
+Fri Jul 25 20:43:57 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (Init_win32ole): add
+	  WIN32OLE_EVENT#off_event.
+
+	* test/win32ole/test_win32ole_event.rb: ditto.
+
+	* test/win32ole/test_win32ole_event.rb: some refactoring.
+
+Fri Jul 25 19:50:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regint.c (xmalloc, xrealloc, xfree): not to use ruby managed memory.
+
+Fri Jul 25 15:52:40 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_invoke_proc): skip setting safe_level if
+	  it from bmethod.  This change makes test/ruby/test_proc.rb pass.
+
+Fri Jul 25 10:00:00 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb: refactoring/cleanup of
+	  test_iso_2022_jp(_1)
+
+Fri Jul 25 04:12:57 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_convert): output of mime encode is US-ASCII.
+	  [ruby-list:45257]
+
+Fri Jul 25 02:43:11 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* sample/coverage.rb: move from lib/coverage.rb because this remains in
+	  an early phase of development.
+
+Fri Jul 25 00:10:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext, rb_find_file): converts Windows style path
+	  to Cygwin path.  [ruby-dev:35647]
+
+Thu Jul 24 16:30:21 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (exit_handler): use st_free_table() to free socklist.
+
+	* win32/win32.c (rb_w32_pipe_exec, rb_w32_accept, rb_w32_socket,
+	  rb_w32_socketpair): should check and release fd and sockets/handles
+	  if an error occurs in rb_w32_open_osfhandle().
+
+Thu Jul 24 16:05:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (overlapped_socket_io): avoid warnings.
+
+	* thread_win32.c (ubf_handle): refactoring.
+
+Thu Jul 24 07:01:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c: got rid of improper casts.
+
+Wed Jul 23 23:19:15 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/socket/test_tcp.rb (test_recvfrom): replace an irrelevant test
+	  for old behavior.
+
+Wed Jul 23 21:38:16 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_marshal.rb: suppress warning during test.
+
+Wed Jul 23 21:35:53 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_dir.rb: suppress warning during test.
+
+Wed Jul 23 18:27:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/{stubs,tcltklib}.c, ext/tk/tkutil/tkutil.c: fix warnings
+	  about constness and signedness.
+
+Wed Jul 23 17:04:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/openssl_missing.h (d2i_of_void): define for older
+	  versions.  [ruby-dev:35637]
+
+Wed Jul 23 13:53:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/curses/extconf.rb: use try_static_assert.
+
+Wed Jul 23 10:06:19 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/zlib/test_zlib.rb (TestZlibDeflate#test_params): suppress a
+	  finalizer warning.
+
+Wed Jul 23 06:25:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck: suppress warnings more.
+
+Wed Jul 23 03:19:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (struct input_code.name, input_codename):
+	  constified.
+
+Wed Jul 23 03:02:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck: suppress warnings.
+
+Wed Jul 23 00:34:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl: suppress warnings.
+
+Tue Jul 22 23:01:34 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/syck/rubyext.c, ext/syck/yaml2byte.c, ext/syck/emitter.c,
+	  ext/syck/syck.c, ext/syck/handler.c, ext/syck/syck.h: suppress GCC
+	  warning.
+
+Tue Jul 22 20:42:24 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimalCmp): BigDecimal#<=> should
+	  return nil if an argument can't be coerced into BigDecimal.
+
+	* ext/bigdecimal/bigdecimal.h, ext/bigdecimal/bigdecimal.c
+	  (VpIsNegDoubleZero, VpItoV): comment out unused functions.
+
+Tue Jul 22 20:33:54 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/rdoc/test_rdoc_parser_ruby.rb (teardown): close tempfile.
+
+Tue Jul 22 19:38:38 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_transcode.rb: added two comments
+
+Tue Jul 22 18:08:34 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja: mention about FIX2LONG and NUM2LONG.
+	  see [ruby-dev:35197]
+
+Tue Jul 22 17:53:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (insn_data_to_s_detail), file.c (rb_stat_inspect),
+	  iseq.c (ruby_iseq_disasm_insn, ruby_iseq_disasm),
+	  process.c (pst_message), re.c (match_inspect): use rb_str_catf.
+
+	* dir.c (dir_inspect), iseq.c (iseq_inspect, insn_operand_intern): use
+	  rb_sprintf.
+
+	* error.c (rb_name_error, rb_raise, rb_loaderror, rb_fatal): use
+	  rb_vsprintf.
+
+Tue Jul 22 17:20:25 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (init_func): new function to get API's address which
+	  is often used and not supported on all Windows.
+
+	* win32/win32.c (overlapped_socket_io): shouldn't use overlapped I/O if
+	  CancelIo() is not supported.
+
+Tue Jul 22 16:47:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h, sprintf.c (rb_str_catf, rb_str_vcatf): new
+	  functions.  [ruby-dev:35597]
+
+	* string.c (rb_str_capacity): new function to return the capacity.
+
+Tue Jul 22 16:08:58 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (connect_blocking, socks_connect_blocking,
+	  ruby_connect): cast.
+
+Tue Jul 22 11:05:08 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_dir.rb: use realpath of tmpdir. [ruby-dev:35481]
+
+	* test/ruby/test_process.rb: ditto.
+
+Tue Jul 22 09:51:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el: fix here-doc strings with inner quotes.  patches
+	  by Nathan Weizenbaum <nex342 AT gmail.com> from [ruby-core:17615]
+	  through [ruby-core:17910].
+
+Tue Jul 22 04:26:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_str_buf_new2): optimization for literals.
+
+	* string.c (str_buf_cat): returns VALUE.
+
+Tue Jul 22 03:34:01 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc*: Update to RDoc 2.1.0 r112.
+
+Tue Jul 22 02:51:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_str_buf_cat2, rb_str_cat2): optimization
+	  for literals.
+
+Tue Jul 22 02:50:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja (1.5 Manipulating Ruby data): fix the
+	  prototype of rb_vsprintf, and added rb_str_cat2.  [ruby-dev:35622]
+
+Mon Jul 21 17:15:38 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_filesystem_encoding): use locale encoding on Unix.
+	  [ruby-dev:35617]
+
+Mon Jul 21 15:29:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/vsnprintf.c (struct __sbuf, FILE): use size_t.
+
+	* sprintf.c (ruby__sfvwrite): ditto.
+
+Mon Jul 21 13:55:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja (1.5 Manipulating Ruby data): mentioned
+	  rb_sprintf and rb_vsprintf.  [ruby-dev:35611]
+
+Mon Jul 21 10:25:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (prepare_getline_args): check if rs is a string when non-nil
+	  lim is given.  [ruby-dev:35610]
+
+Mon Jul 21 04:55:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* dir.c (dir_initialize): use rb_convert_type instead of
+	  rb_check_convert_type to prevent SEGV by Dir.new(".", true).
+	  (dir_initialize): use FilePathValue before rb_enc_get(dirname) to
+	  prevent SEGV by Dir.new(0).
+
+Mon Jul 21 04:42:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_s_union): useless rb_enc_get call removed to prevent
+	  SEGV by Regexp.union("", nil).
+
+Sun Jul 20 22:50:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): NODE_POSTEXE should set each end
+	  procs only once.  [ruby-dev:35596]
+
+Sun Jul 20 16:00:37 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (EVENTSINK_Invoke): use rb_protect
+	  instead of rb_rescue2. [ruby-dev:35595]
+
+Sun Jul 20 01:23:24 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (EVENTSINK_Invoke): little refactoring.
+
+	* ext/win32ole/win32ole.c (EVENTSINK_GetIDsOfNames): return
+	  ITypeInfo::GetIDsOfNames().
+
+Sat Jul 19 09:31:35 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (EVENTSINK_Invoke): little refactoring.
+
+	* ext/win32ole/win32ole.c: add document of inspect methods.
+
+Sat Jul 19 06:08:43 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc*: Import RDoc r104.  Various make test-all fixes.
+
+Sat Jul 19 00:27:58 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (check_uint, rb_num2uint, rb_fix2uint): fixed wrong check
+	  about 64bit positive value.
+
+Fri Jul 18 23:23:37 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (EVENTSINK_Invoke): avoid cfp consistency
+	  error when exception raised in event callback.
+
+Fri Jul 18 14:52:14 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (socklist): table for registering socket options
+	  (currently only O_NONBLOCK).
+
+	* win32/win32.c (StartSockets, exit_handler): alloc/free socklist.
+
+	* win32/win32.c (is_socket): use socklist.
+
+	* win32/win32.c (rb_w32_accept, rb_w32_socket, rb_w32_socketpair):
+	  register new socket to socklist.
+
+	* win32/win32.c (rb_w32_close): remove closing socket from socklist.
+
+	* win32/win32.c (fcntl): register socket options.
+
+	* win32/win32.c (overlapped_socket_io): send to/recv from socket with
+	  overlapped operation if the socket is not nonblocking mode.
+	  [experimental]
+
+	* win32/win32.c (rb_w32_send, rb_w32_sendto, rb_w32_recv,
+	  rb_w32_recvfrom): use overlapped_socket_io().
+
+	* win32/win32.c (open_ifs_socket): set overlapped mode. this is the
+	  default mode of winsock's socket(), so lacking it is an old bug.
+
+Fri Jul 18 09:44:30 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/*: Import RDoc r101.
+
+Thu Jul 17 23:45:55 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/rdoc/test_rdoc_c_parser.rb (teardown): close tempfile.
+
+Thu Jul 17 21:08:49 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_process.rb (test_getpriority, test_setpriority): use
+	  PRIO_PROCESS instead of PRIO_USER.
+
+Thu Jul 17 20:41:42 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* pack.c (pack_unpack): fix v and V with big endian.
+	  [1].pack("V").unpack("V") was [4294967296].
+
+Thu Jul 17 20:35:03 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* pack.c (pack_pack): fix i! with big endian.  [1].pack("i!") was
+	  "\0\0\0\0".
+
+Thu Jul 17 16:48:40 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_connect): select() for connect() has
+	  mean only when the socket is non-blocking.
+
+Thu Jul 17 10:55:24 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* mkconfig.rb: Simplify expression for RDoc.
+
+Thu Jul 17 10:21:15 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (check_uint, rb_num2uint, rb_fix2uint): strict check.
+	  fixed [ruby-dev:33683]
+
+Thu Jul 17 04:19:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): wake up joining threads.
+
+	* thread.c (sleep_forever, sleep_timeval): return when interrupted.
+	  [ruby-dev:35542]
+
+	* thread.c (timer_thread_function): restore main thread status.
+	  [ruby-core:17270]
+
+Thu Jul 17 01:27:38 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (appendline): remove invalid access.
+
+Wed Jul 16 18:04:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (signal_exec, trap_handler): trap accepts a string as
+	  command.  [ruby-dev:35533]
+
+Wed Jul 16 00:04:30 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_in_out_err): new
+	  method.
+
+	* test/ruby/test_argf.rb: use assert_in_out_err instead of
+	  EnvUtil.rubyexec.
+
+	* test/ruby/test_module.rb: ditto.
+
+	* test/ruby/test_require.rb: ditto.
+
+	* test/ruby/test_objectspace.rb: ditto.
+
+	* test/ruby/test_object.rb: ditto.
+
+	* test/ruby/test_string.rb: ditto.
+
+	* test/ruby/test_method.rb: ditto.
+
+	* test/ruby/test_variable.rb: ditto.
+
+	* test/ruby/test_io.rb: ditto.
+
+	* test/ruby/test_rubyoptions.rb: ditto.
+
+	* test/ruby/test_exception.rb: ditto.
+
+	* test/ruby/test_class.rb: ditto.
+
+	* test/ruby/test_thread.rb: ditto.
+
+Tue Jul 15 22:34:03 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (ruby_version): follow changes in configure.in.
+
+Tue Jul 15 21:58:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/{,tkutil/}extconf.rb: ruby/ruby.h no longer needs to be
+	  checked.
+
+	* ext/tk/{tcltklib.c,tkutil/tkutil.c}: check macros for each headers.
+
+Tue Jul 15 21:45:41 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/tk/extconf.rb, ext/tk/tcltklib.c, ext/tk/tkutil/tkutil.c: Do
+	  not test ruby/ruby.h, which makes OS X's gcc pick the wrong
+	  header file from Ruby.framework.
+
+Tue Jul 15 21:31:26 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in (--with-ruby-version): Add a new option to specify
+	  the ruby version string for version specific directories.
+	  [ruby-dev:35490]
+
+	* mkconfig.rb: Definition of ruby_version is now determined by the
+	  configure script.
+
+Tue Jul 15 18:14:20 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_connect, s_accept): check before readable/
+	  writable by select() instead of wrapping in blocking region.
+
+	* ext/socket/socket.c (bsock_send, s_recvfrom, udp_send, unix_send_io,
+	  unix_recv_io): should check readable/writable before calling blocking
+	  functions.
+	  see [ruby-dev:35446]
+
+Tue Jul 15 18:12:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (ubf_handle): cancel blocking IO if it can (only
+	  Vista). see [ruby-dev:35446]
+
+	* win32/win32.c (errmap): add ERROR_OPERATION_ABORTED as EINTR.
+
+Mon Jul 14 20:35:21 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_event.rb (teardown): fix typo.
+
+Mon Jul 14 18:47:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (transcode_loop): constified.
+
+	* transcode.c (str_transcode): rb_str_set_len() sets a delimiter.
+
+	* transcode_data.h (rb_transcoder): constified preprocessor and
+	  postprocessor input.
+
+	* enc/trans/japanese.c: ditto.
+
+Sun Jul 13 05:37:50 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRI_PTRDIFF_PREFIX, PRI_SIZE_PREFIX): typo.
+
+	* {bcc32,win32}/Makefile.sub: (SIZEOF_SIZE_T, SIZEOF_PTRDIFF_T): added.
+
+Sat Jul 12 23:54:55 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (LIBRUBY_DLDFLAGS): import library which created
+	  with DLL is broken. save import library which created by lib.exe and
+	  install it.
+	  fixed the problem mentioned at the postscript of [ruby-dev:35448]
+
+Sat Jul 12 23:24:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c: use PRIuSIZE.
+
+Sat Jul 12 22:41:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRI_PTRDIFF_PREFIX, PRI_SIZE_PREFIX): fixed typo.
+
+Sat Jul 12 22:30:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (HAVE_RUBY_RUBY_H): defines macros so that
+	  extconf.rb do not need to check if headers exist under separated
+	  directory.  [ruby-dev:35437]
+
+	* include/{ruby,rubyio,rubysig}.h, include/ruby/intern.h: use
+	  "ruby/..." instead of <ruby/...>.
+
+Sat Jul 12 22:17:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (sizeof ptrdiff_t): check for size of ptrdiff_t.
+
+	* include/ruby/ruby.h (PRI?PTRDIFF, PRI?SIZE): printf conversion
+	  specifiers for ptrdiff_t and size_t/ssize_t.
+
+	* insns.def (leave), marshal.c (long_toobig), transcode.c
+	  (str_transcode), vm_dump.c (control_frame_dump, stack_dump_each),
+	  (debug_print_register, debug_print_pre): t and z length modifiers
+	  are C99.
+
+Sat Jul 12 16:02:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client#do_rpc): requires
+	  webrick/cookie.  [ ruby-Bugs-21139 ]
+
+Sat Jul 12 09:25:07 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_event.rb: add test
+	  for WIN32OLE_EVENT#on_event_with_outargs
+
+	* test/win32ole/test_win32ole_event.rb(teardown): calling
+	  WIN32OLE_EVENT.message_loop
+
+Sat Jul 12 01:54:13 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_select): shouldn't pass non-socket handle to
+	  original select().
+
+Fri Jul 11 23:05:40 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/zlib/test_zlib.rb: add a test for Zlib::Deflate#params.
+
+Fri Jul 11 22:58:28 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/zlib/zlib.c (rb_deflate_params): flush before deflateParams.
+	  [ruby-core:17675]
+
+Fri Jul 11 22:09:01 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.rb, win32/Makefile.sub (PLATFORM): use $(PLATFORM)
+	  instead of $(OS) because ENV["OS"] is used in test-all (drb).
+
+Fri Jul 11 20:51:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_wait_readable, rb_io_wait_writable): check if the file
+	  descriptor is closed.
+
+	* thread.c (rb_thread_wait_fd_rw): ditto.
+
+Fri Jul 11 16:16:43 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_accept, rb_w32_socket, rb_w32_socketpair):
+	  prohibit inheritance of sockets, too. [experimental]
+
+Fri Jul 11 14:39:49 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): the measures for Vista is no longer
+	  unnecessary.
+
+Fri Jul 11 06:16:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (process.o): depends on util.h.
+
+Fri Jul 11 05:07:46 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (pipe): prohibit inheritance.
+	  fixed: [ruby-dev:35421]
+
+Fri Jul 11 00:56:46 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (thread_create_core): fix GC problem.
+	  [ruby-core:17669]
+
+Thu Jul 10 22:06:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pstore.rb (PStore#transaction): return the result from the
+	  block.  [ruby-core:17718]
+
+Thu Jul 10 21:15:49 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_typelib.rb: add some illegal argument
+	  test.
+
+	* test/win32ole/test_win32ole_type.rb: ditto.
+
+	* test/win32ole/test_win32ole_event.rb: ditto.
+
+	* test/win32ole/test_win32ole_param.rb: ditto.
+
+	* test/win32ole/test_win32ole_method.rb: ditto.
+
+Thu Jul 10 19:38:35 2008  wanabe  <s.wanabe@gmail.com>
+
+	* test/ruby/envutil.rb (assert_normal_exit): r17993 revert.
+
+Thu Jul 10 18:29:41 2008  wanabe  <s.wanabe@gmail.com>
+
+	* test/ruby/envutil.rb (assert_normal_exit): finish writing script
+	  before spawn("ruby") to avoid blocking in win32.
+
+Thu Jul 10 17:20:50 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (insert): follow recent changes of globbing.
+
+Thu Jul 10 14:09:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/coverage/coverage.c (rb_coverage_start): return nil.
+
+Thu Jul 10 12:41:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_wait_for): fixed variable name.
+
+Thu Jul 10 12:09:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_thread_t), vm.c (rb_thread_mark), process.c
+	  (rb_last_status_get, rb_last_status_set, rb_last_status_clear):
+	  moved last_status from rb_vm_t.  [ruby-dev:35414]
+
+	* vm.c (th_init2): initialize last_status with nil.
+
+Thu Jul 10 12:09:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_wait_for): wait until timed out only when
+	  sleeping with timeout.
+
+Wed Jul  9 22:41:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (sleep_timeval): wait until timed out.  [ruby-core:17270]
+
+Wed Jul  9 20:58:16 2008  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_fill): don't raise even if length is negative.
+	  [ruby-core:17483], [ruby-core:17661]
+
+Wed Jul  9 20:18:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_va_args_macro): check for __VA_ARGS__.
+
+	* thread.c (thread_debug): show source name and line if possible.
+
+	* thread_{pthread,win32}.c (rb_thread_create_timer_thread): needs more
+	  stack for debug.
+
+Wed Jul  9 11:13:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/profiler.rb (Profiler__#print_profile): sort in the descending
+	  order of cumulative time.
+
+Wed Jul  9 11:11:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (struct glob_args, rb_glob_caller, rb_glob2, push_pattern),
+	  (glob_brace): make consistent prototypes.
+
+	* dir.c (push_glob): set enc in the caller of rb_glob_caller as well
+	  as rb_glob2.
+
+Wed Jul  9 09:12:11 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (options): use input_endian.
+
+Wed Jul  9 01:38:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_succ): alphabets or numerics mutually enclosing
+	  non-alphanumeric characters can carry up.  e.g., "1.999".succ should
+	  be "2.000".
+
+Wed Jul  9 00:12:31 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_set_coverages, rb_reset_coverages): enable and disable
+	  coverage measurement.
+
+	* thread.c (rb_get_coverages): rename and move from vm.c.
+
+	* vm.c (rb_vm_get_coverages): ditto.
+
+	* iseq.c (prepare_iseq_build): ditto.
+
+	* thread.c (clear_coverage): ditto.
+
+	* parse.y (coverage): ditto.
+
+	* ext/coverage/coverage.c: use above functions, add new method
+	  Coverage.start and fix rdoc .
+
+Tue Jul  8 23:02:35 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (find_default_source): bug fix when
+	  OLE object does not have default source interface.
+
+	* test/win32ole/test_win32ole_event.rb: ditto.
+
+Tue Jul  8 22:56:23 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_enable_coverages): hide coverage array by setting 0 to
+	  klass during measurement.
+
+	* parse.y (coverage, yycompile0): ditto.
+
+	* iseq.c (prepare_iseq_build): use rb_hash_lookup instead of
+	  rb_hash_aref.
+
+	* thread.c (rb_coverage_result): restore klass of coverage array
+	  and return it.
+
+	* thread.c (update_coverage): check whether its klass is 0.
+
+Tue Jul  8 22:28:25 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/debug.rb, lib/profile.rb: fix to use RubyVM.
+
+	* lib/rdoc/parsers/parse_c.rb: ditto.
+
+Tue Jul  8 21:45:22 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (rb_vm_mark): mark the last element of special_exceptions.
+
+Tue Jul  8 19:55:40 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (find_default_source): try to
+	  find COCLASS when WIN32OLE object is not COCLASS.
+
+	* test/win32ole/test_win32ole_event.rb: ditto
+
+Tue Jul  8 13:38:22 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.h: fix to skip inserting a trace insn.
+
+Tue Jul  8 11:41:17 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* dir.c: shouldn't use ruby object in globbing, because glob service
+	  routines are called before initializing ruby on some platforms (ex.
+	  windows).
+
+Tue Jul  8 10:08:40 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (Next): use rb_enc_mbclen. [ruby-dev:35390]
+
+Tue Jul  8 07:59:40 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (Next): use rb_enc_precise_mbclen.
+
+Tue Jul  8 02:27:23 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c: preserve encoding of strings in glob and fnmatch.
+
+	* include/ruby/ruby.h: related changes.
+
+Tue Jul  8 00:22:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_succ): limit carrying in an alphanumeric region if
+	  exists.  [ruby-dev:35094]
+
+Mon Jul  7 20:39:28 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(Init_win32ole): add
+	  WIN32OLE_TYPE#source_ole_types, WIN32OLE_TYPE#default_ole_types,
+	  WIN32OLE_TYPE#default_event_sources.
+
+	* test/win32ole/test_win32ole_type.rb: ditto.
+
+Mon Jul  7 19:45:22 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_dir.rb (test_chroot_nodir): add Errno::EPERM.
+
+Mon Jul  7 17:12:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ipaddr.rb (IPAddr#initialize): get rid of ArgumentError in
+	  IPAddr#to_range.  a patch from okkez <okkez000 AT gmail.com> in
+	  [ruby-dev:35091].
+
+Mon Jul  7 01:24:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_extname): fix for file name with spaces.
+	  [ruby-talk:307404]
+
+Mon Jul  7 00:59:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack): prior STACK_END_ADDRESS if
+	  found.  [ruby-core:17624]
+
+Sun Jul  6 23:48:06 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (bsock_send, s_recvfrom, ruby_connect, s_accept),
+	  (udp_send, unix_send_io, unix_recv_io): blocking region support.
+
+Sun Jul  6 18:34:35 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_type.rb (test_initialize): add
+	  more assertions.
+
+Sun Jul  6 10:12:21 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/test/unit/collector/objectspace.rb
+	  (Test::Unit::Collector::ObjectSpace::NAME): fix a typo.
+
+Sun Jul  6 00:56:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (host_str): fix type mismatch in rb_raise
+	  format and argument.
+	  (port_str): ditto.
+	  (unix_recv_io): ditto.
+	  (sock_s_unpack_sockaddr_un): ditto.
+
+Sat Jul  5 23:42:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (NUM2INT): cast to int.
+	  (FIX2INT): ditto.
+	  (NUM2UINT): cast to unsigned int.
+	  (FIX2UINT): ditto.
+
+Sat Jul  5 23:10:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_pipe): new function for handling EMFILE and ENFILE
+	  error of pipe().
+	  (UPDATE_MAXFD_PIPE): removed.
+	  (pipe_open): use rb_pipe.
+	  (rb_io_s_pipe): ditto.
+
+	* process.c (pipe_nocrash): use rb_pipe.
+
+	* include/ruby/intern.h (rb_pipe): declared.
+
+Sat Jul  5 22:22:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_initialize): NUM2INT() returns int.
+
+	* thread.c (timer_thread_function), thread_pthread.c (thread_timer),
+	  thread_win32.c (timer_thread_func), thread_{pthread,win32}.c
+	  (rb_thread_create_timer_thread): passing VM.
+
+Sat Jul  5 20:53:18 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_word.rb: check word installed.
+
+Sat Jul  5 16:12:54 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c: revert. before lazy sweep.
+
+Sat Jul  5 09:55:44 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE#ole_respond_to?
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+Sat Jul  5 08:48:05 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (unescape_nonascii): add has_property argument not to
+	  raise error by /\p{Hiragana}\u{3042}/ in EUC-JP script.
+	  (rb_reg_preprocess): use has_property argument to make regexp
+	  encoding fixed.
+
+Sat Jul  5 08:29:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (unescape_nonascii): make regexp fixed_encoding if \p is used.
+	  fixed [ruby-core:17279].
+
+Fri Jul  4 23:12:53 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (d2time): fix the bug of VT_DATE
+	  to String conversion when negative value.
+
+	* test/win32ole/test_win32ole_variant.rb: ditto.
+
+Fri Jul  4 22:15:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit/testcase.rb: collect descendants of
+	  Test::Unit::TestCase using inherited.
+
+	* lib/test/unit/autorunner.rb: don't use ObjectSpace.each_object.
+
+	* lib/test/unit/collector/dir.rb: ditto.
+
+	* lib/test/unit/collector/objectspace.rb: ditto.
+
+	[ruby-core:17126]
+
+Fri Jul  4 20:43:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/oniguruma.h (OnigEncoding): removed auxiliary_data.
+
+	* include/ruby/encoding.h (ENC_DUMMY_P): moved dummy encoding flag to
+	  rb_encoding from Encoding instance.
+
+	* encoding.c (rb_encoding_list): list of Encoding instances.
+
+	* encoding.c (struct rb_encoding_entry): moved base encoding from
+	  instance variable.
+
+Fri Jul  4 17:51:07 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (check_uint, rb_num2uint, rb_fix2uint): proper check.
+
+Fri Jul  4 14:17:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::FTP#sendport): use divmod.  [ruby-core:17557]
+
+Fri Jul  4 11:08:37 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (garbage_collect_force): sweep is completely ended.
+
+	* gc.c (os_obj_of): invoke garbage_collect_force() when freelist none.
+
+Fri Jul  4 05:01:26 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (rb_num2uint, rb_fix2uint): typo.
+
+Fri Jul  4 02:21:06 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (check_uint, rb_num2uint, rb_fix2uint): also needs checking
+	  negative value. see [ruby-dev:33683]
+
+Thu Jul  3 23:26:36 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* include/ruby/intern.h: remove prototypes about coverage.
+
+	* iseq.c (prepare_iseq_build): add prototype.
+
+	* parse.y (coverage): ditto.
+
+	* thread.c (clear_coverage): ditto.
+
+	* thread.c (update_coverage): use rb_sourceline.
+
+	* thread.c (rb_get_coverages): rename and move to vm.c.
+
+	* vm.c (rb_vm_get_coverages): ditto.
+
+	* ext/coverage/coverage.c: add rdoc.
+
+Thu Jul  3 21:51:21 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/coverage/coverage.c, ext/coverage/extconf.rb: eliminate
+	  COVERAGE__ and introduce coverage.so instead.  How to measure
+	  coverage: (1) require "coverage.so", (2) require or load Ruby source
+	  file, and (3) Coverage.result will return the same hash as COVERAGE__.
+	  [ruby-dev:35324]
+
+	* thread.c (rb_enable_coverages): start coverage measurement by using
+	  rb_add_event_hook.
+
+	* thread.c (rb_get_coverages): returns current results of coverage
+	  measurement.
+
+	* include/ruby/intern.h: add prototype for above two functions.
+
+	* vm_core.h, vm.c: add field of coverages to rb_vm_t.
+
+	* insns.def (trace): remove special handling for COVERAGE__.
+
+	* iseq.c (prepare_iseq_build): switch COVERAGE__ to
+	  rb_get_coverages().
+
+	* parse.y (coverage): ditto.
+
+	* thread.c (clear_coverage): ditto.
+
+	* lib/coverage.rb: use coverage.so instead of COVERAGE__.
+
+Thu Jul  3 21:20:45 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (thread_initialize): NUM2INT returns long.
+
+Thu Jul  3 21:06:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (Init_eval): typo fixed in r17833.
+
+Thu Jul  3 19:44:44 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (Init_win32ole): remove duplicate line.
+
+Thu Jul  3 16:08:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (erfc): erfc of glibc comes with Debian GNU/Linux Etch
+	  on IA64 is broken.  erfc(10000.0) aborts.
+	  use missing/erf.c instead.
+	  http://sources.redhat.com/ml/libc-hacker/2005-08/msg00008.html
+
+Thu Jul  3 12:49:39 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/smtp.rb (Net::SMTP::start): use 'localhost' instead of
+	  'localhost.localdomain'.  [ruby-dev:35333]
+
+	* lib/net/smtp.rb (Net::SMTP::SMTP.start): ditto.
+
+Thu Jul  3 07:06:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (SET_LC_MESSAGES): LC_MESSAGES must be C.
+
+Thu Jul  3 07:02:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (Init_eval), gc.c (Init_GC), proc.c (Init_Proc): freeze
+	  messages of preallocated special exceptions also.
+
+Thu Jul  3 04:39:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_during_gc): VALUE cache is irrelevant.
+
+Thu Jul  3 01:44:01 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* regint.h (GET_ALIGNMENT_PAD_SIZE, ALIGNMENT_RIGHT): cast pointer to
+	  uintptr_t instead of unsigned int.
+
+Thu Jul  3 01:23:13 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* sprintf.c: include ieeefp.h to refer to isinf.
+
+	* ext/bigdecimal/bigdecimal.c: ditto.
+
+	* ext/json/ext/generator/generator.c: ditto.
+
+	* rational.c: ditto.
+
+Thu Jul  3 01:01:57 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* missing/tgamma.c (tgamma): remove unused variable.
+
+Thu Jul  3 00:18:00 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: avoid creating Ruby object during
+	  GC. thanks to arton <artonx AT yahoo.co.jp>. [ruby-dev:35313]
+
+	* test/win32ole/test_win32ole_event.rb: ditto.
+
+Thu Jul  3 00:09:31 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* gc.c: add rb_during_gc(). based on a patch from arton <artonx AT
+	  yahoo.co.jp> at [ruby-dev:35313].
+
+	* include/ruby/intern.h: ditto.
+
+Wed Jul  2 09:49:10 2008  Narihiro Nakamura  <authorNari@gmail.com>
+
+	* gc.c (gc_lazy_sweep): use lazy sweep algorithm for response
+	  performance gain.
+
+	* gc.c (garbage_collect_force): mark and lazysweep invoke, after
+	  erasing all mark.
+
+	* gc.c (GC_NOT_LAZY_SWEEP): not lazy sweep flag. for debug.
+
+Wed Jul  2 03:42:44 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_settracefunc.rb: fix expected traces for
+	  RubyVM::FrozenCore's event and r17744.
+
+Wed Jul  2 03:10:41 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.h, insns.def: reduce insn operand of "trace".
+
+	* include/ruby/ruby.h: add RUBY_EVENT_COVERAGE event.
+
+Wed Jul  2 02:02:34 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c, vm.c: fix to refer to next ruby level cfp to make binding.
+
+Wed Jul  2 01:58:19 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* insns.def (trace): C99ism.
+
+Wed Jul  2 01:53:40 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* Add coverage measurement constant COVERAGE__.  This constant is not
+	  for casual use.  Usage: (1) assign {} to COVERAGE__, (2) require or
+	  load Ruby source file, and (3) COVERAGE__["sourcefilepath"] will
+	  return an array whose elements represent number of executions per
+	  line of source code.
+
+	* vm_core.h: add field of coverage array to iseq.
+
+	* iseq.c (prepare_iseq_build): ditto.
+
+	* insns.def (trace): update coverage array.
+
+	* parse.y (coverage): create and initialize coverage array.
+
+	* compile.h (ADD_TRACE): add trace instruction to update coverage
+	  array.
+
+	* thread.c (clear_coverage): delete coverage array when forking.
+	  Otherwise, double count of coverage may occur.
+
+	* lib/coverage.rb: sample coverage measurement tool.
+
+	* error.c: distinguish explicitly between parse_in_eval and
+	  mild_compile_error.
+
+	* load.c: ditto.
+
+	* vm_eval.c: ditto.
+
+Tue Jul  1 21:32:43 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/test/unit/ui/console/testrunner.rb: prevent destructive
+	  modification to $0.
+
+	* test/rubygems/gemutilities.rb (build_rake_in): move from
+	  test_gem_ext_rake_builder.rb.
+
+	* test/rubygems/test_gem_ext_rake_builder.rb: ditto.
+
+	* test/rubygems/test_gem_installer.rb: override Gem.ruby and
+	  ENV["rake"].
+
+	* test/rubygems/test_gem_uninstaller.rb: ditto.
+
+Tue Jul  1 21:13:17 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, vm.c, insns.def: call FrozenCore.set_postexe method
+	  instead to use "postexe" insn.
+
+	* id.c, id.h: add a prepared id for above.
+
+Tue Jul  1 21:09:58 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* lib/mkmf.rb (create_tmpsrc): we need to include COMMON_HEADERS,
+	  namely inclusion of ruby.h, because _GNU_SOURCE is now defined
+	  there (if any) and socket.so requires it on Linux systems.
+
+Tue Jul  1 20:55:07 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* eval.c (rb_interrupt): trick to suppress GCC warning.
+
+	* sprintf.c (rb_str_format): ditto.
+
+Tue Jul  1 20:44:36 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/instruction.rb: RubyVM is not module.
+
+Tue Jul  1 19:31:24 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* ext/digest/digest.c (rb_digest_instance_inspect): constified.
+
+	* variable.c (rb_path2class): field precision should have type int.
+
+Tue Jul  1 19:01:00 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/shift_jis.c (code_is_ctype): HALF WIDTH KATAKANA is
+	  a character.
+
+Tue Jul  1 17:56:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/vsnprintf.c (BSD__uqtoa): constified.
+
+Tue Jul  1 17:50:44 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/make_encdb.h: always add ';' at the end of line.
+
+Tue Jul  1 17:44:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (COMPILE_OPTION_FALSE), time.c (timegm_noleapsecond),
+	  thread.c (eKillSignal, eTerminateSignal),
+	  missing/vsnprintf.c (BSD_vfprintf): constified.
+
+Tue Jul  1 17:37:43 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* enc/ascii.c: ISO C does not allow extra ';' outside of a
+	  function
+
+	* enc/us_ascii.c: ditto.
+
+	* enc/utf_8.c: ditto.
+
+	* enc/big5.c: ditto.
+
+	* enc/euc_jp.c: ditto.
+
+	* enc/euc_kr.c: ditto.
+
+	* enc/euc_tw.c: ditto.
+
+	* enc/gb2312.c: ditto.
+
+	* enc/gbk.c: ditto.
+
+	* enc/iso_8859_1.c: ditto.
+
+	* enc/iso_8859_2.c: ditto.
+
+	* enc/iso_8859_3.c: ditto.
+
+	* enc/iso_8859_4.c: ditto.
+
+	* enc/iso_8859_5.c: ditto.
+
+	* enc/iso_8859_6.c: ditto.
+
+	* enc/iso_8859_7.c: ditto.
+
+	* enc/iso_8859_8.c: ditto.
+
+	* enc/iso_8859_9.c: ditto.
+
+	* enc/iso_8859_10.c: ditto.
+
+	* enc/iso_8859_11.c: ditto.
+
+	* enc/iso_8859_13.c: ditto.
+
+	* enc/iso_8859_14.c: ditto.
+
+	* enc/iso_8859_15.c: ditto.
+
+	* enc/iso_8859_16.c: ditto.
+
+	* enc/koi8_r.c: ditto.
+
+	* enc/shift_jis.c: ditto.
+
+	* enc/utf_16be.c: ditto.
+
+	* enc/utf_32be.c: ditto.
+
+	* enc/utf_32le.c: ditto.
+
+	* enc/windows_1251.c: ditto.
+
+	* process.c (run_exec_rlimit): ISO C90 forbids mixed declarations
+	  and code
+
+	* include/ruby/ruby.h (enum ruby_special_consts): ISO C forbids
+	  comma at end of enumerator list
+
+	* include/ruby/ruby.h (enum ruby_value_type): ditto.
+
+	* eval_intern.h (enum): ditto.
+
+	* vm_core.h (enum rb_thread_status): ditto.
+
+	* parse.y (enum lex_state_e): ditto.
+
+	* parse.y (enum string_type): ditto.
+
+	* process.c (enum): ditto.
+
+	* ruby.c (enum dump_flag_bits): ditto.
+
+	* ruby.c (enum disable_flag_bits): ditto.
+
+	* enc/gb18030.c (enum): ditto.
+
+Tue Jul  1 17:21:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regexec.c (stack_double): use MatchStackLimitSize atomically.
+
+	* regparse.c (onig_free_shared_cclass_table): OnigTypeCClassTable
+	  needs atomicity
+
+	* regsyntax.c: constified all predefined OnigSyntaxTypes.
+
+Tue Jul  1 16:57:44 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* include/ruby/st.h (struct st_table): type of bit-field
+	  'num_entries' is a GCC extension
+
+	* include/ruby/ruby.h (rb_intern): prefix __extension__ for
+	  braced-groups within expressions.
+
+	* include/ruby/intern.h (rb_usascii_str_new2): ditto.
+
+	* include/ruby/intern.h (rb_tainted_str_new2): ditto.
+
+	* include/ruby/intern.h (rb_str_new2): ditto.
+
+Tue Jul  1 15:01:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/ext/builder.rb (Gem::Ext::Builder.make),
+	  (Gem::Ext::Builder.run): EXIT_SUCCESS may be 0 or may not.
+
+	* test/rubygems/test_gem_ext_rake_builder.rb (build_rake_in): override
+	  Gem.ruby and ENV["rake"].
+
+	* runruby.rb: bin/rake does not exist in archdir where architecture
+	  depend script (i.e. rbconfig.rb) exists.
+
+Tue Jul  1 13:19:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_fill): check if beg is too big.
+
+Tue Jul  1 12:01:16 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def, vm.c, vm_core.h: remove some insns
+	  (undef, alias, definemethod).
+	  Call RubyVM::FrozenCore's singleton method instead.
+	  Add "putiseq" and "putspecialobject" instructions.
+
+	* id.c, id.h: add ids for above.
+
+	* tool/parse.rb: "VM" no longer exists.  Use RubyVM instead.
+
+Tue Jul  1 03:28:16 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/test_ext_configure_builder.rb:  Apply locale-free
+	  patch by Yusuke ENDOH.  [ruby-core:17444].
+
+	* runruby.rb: Set ENV['rake'].  Patch by Yusuke ENDOH
+	  [ruby-core:17442].
+
+Tue Jul  1 01:07:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension.read_multipart): blanks inside
+	  double quotes are allowed.  [ruby-list:45140]
+
+Tue Jul  1 00:59:43 2008  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (num_coerce): call rb_Float(x) first.  don't depend on
+	  evaluation order of function arguments.
+
+Tue Jul  1 00:49:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_newobj): abort GC phase before rb_bug.
+
+Mon Jun 30 23:15:07 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/openssl/test_ssl.rb (start_server): shutdown TCPServer before
+	  close.
+
+Mon Jun 30 23:01:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RUBY_H_INCLUDES): common headers which are included with
+	  ruby.h together.
+
+Mon Jun 30 22:57:50 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_settracefunc.rb (test_raise): reveal an exception
+	  hided by rescue modifier.
+
+Mon Jun 30 22:49:32 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb (test_aset): access with too big index
+	  raises not ArgumentError but IndexError now.
+
+Mon Jun 30 22:30:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb (recursive_collect): r15662 reverted.
+
+Mon Jun 30 22:27:39 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/stringio/stringio.c (strio_getline): fix for nil and "" as
+	  separator.  [ruby-dev:34591]
+
+Mon Jun 30 22:21:30 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (argf_each_line): pass args to each_line. [ruby-dev:34958]
+
+Mon Jun 30 22:12:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (chain_finalized_object): should not delete from finalizer
+	  table until run.
+
+	* gc.c (rb_gc_call_finalizer_at_exit): deferred_final_list may be
+	  empty first.
+
+Mon Jun 30 18:57:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_finalize_deferred): allow object allocation in finalizers.
+
+	* gc.c (rb_gc_call_finalizer_at_exit): ditto.
+
+Mon Jun 30 14:41:36 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (rb_newobj): prohibit call of rb_newobj() during gc when
+	  USE_VALUE_CACHE is not defined (normal case).
+
+Mon Jun 30 10:28:33 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syslog/syslog.c (syslog_write): syslog operations should be
+	  protected from $SAFE level 4.  a patch from Keita Yamaguchi
+	  <keita.yamaguchi at gmail.com>.
+
+	* ext/syslog/syslog.c (mSyslog_close): ditto.
+
+	* ext/syslog/syslog.c (mSyslog_set_mask): ditto.
+
+Mon Jun 30 03:01:35 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/net/http/test_http.rb (_test_request__file): specify encoding
+	  explicitly.
+
+	* test/net/http/utils.rb: ditto.
+
+Mon Jun 30 02:31:07 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_utf8_encindex): defined.
+
+	* include/ruby/encoding.h (rb_utf8_encindex): ditto.
+
+Mon Jun 30 02:14:34 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, vm_core.h,vm_core.h, vm_dump.c, iseq.c: rename class name
+	  VM -> RubyVM, and rename rb_cVM -> rb_cRubyVM.
+	  "VM" is too short name for class.
+
+	* test/ruby/test_method.rb, test/ruby/test_settracefunc.rb: ditto.
+
+	* include/ruby/ruby.h: rb_cRubyVM, rb_cEnv, rb_cISeq should not be
+	  exposed.
+
+Mon Jun 30 02:10:32 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* process.c (Init_process): fix to avoid a warning.
+
+Mon Jun 30 01:52:05 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_parse.rb: remove tests for open_args.
+
+Sun Jun 29 23:01:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_format_m): make tmp volatile to avoid possible GC
+	  problem.
+
+Sun Jun 29 18:01:30 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb, lib/rss/utils.rb: merge documents from ruby_1_8.
+
+Sun Jun 29 17:44:23 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb (RSS::ListenerMixin#known_class): define to
+	  work with ruby 1.8.x too.
+
+Sun Jun 29 17:41:42 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb (RSS::Maker::RSSBase#to_feed): raise
+	  exception not return nil if RSS::Maker.make can't get required
+	  information.
+
+	* test/rss/rss-assertions.rb: follow the above change.
+
+Sun Jun 29 17:37:23 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb (RSS::Maker::RSSBase#make): require block.
+
+	* test/rss/test_maker_{0.9,1.0,2.0}.rb: follow the above change.
+
+Sun Jun 29 17:33:34 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb, lib/rss/maker/itunes.rb: don't use
+	  instance_eval to initialize variables. (speed up)
+
+Sun Jun 29 17:31:15 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb, test/rss/test_version.rb (RSS::VERSION):
+	  0.2.4 -> 0.2.5.
+
+Sun Jun 29 11:36:20 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* dir.rb: fix resource leak.
+
+Sun Jun 29 09:43:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tmpdir.rb (@@systmpdir): prior LOCAL_APPDATA if possible, and
+	  should be clean.  based on a patch from arton <artonx AT
+	  yahoo.co.jp> at [ruby-dev:35269]
+
+Sun Jun 29 07:53:08 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (date2time_str): fix the overflow in
+	  some situation. [ruby-bugs-20793]
+
+Sat Jun 28 21:25:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (struct RRegexp): new field usecnt.  replace
+	  str and len by src.
+
+	* gc.c (gc_mark_children): mark src field of regexp.
+	  (obj_free): don't free str field.
+
+	* re.c (REG_BUSY): removed.
+	  (rb_reg_initialize): prohibit re-initialize regexp.
+	  (rb_reg_search): use usecnt to prevent freeing regexp currently
+	  using.  this prevents SEGV by:
+	    r = /\A((a.)*(a.)*)*b/
+	    r =~ "ab" + "\xc2\xa1".force_encoding("euc-jp")
+	    t = Thread.new { r =~ "ab"*8 + "\xc2\xa1".force_encoding("utf-8")}
+	    sleep 0.2
+	    r =~ "ab"*8 + "\xc2\xa1".force_encoding("euc-jp")
+
+Sat Jun 28 21:15:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_str_new2, rb_tainted_str_new2,
+	  rb_usascii_str_new2): use inline versions only for constant
+	  literals.
+
+Sat Jun 28 13:12:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/rubygems/test_gem.rb (test_self_path_APPLE_GEM_HOME): don't use
+	  fixed /tmp/apple_gem_home directory.
+
+Sat Jun 28 08:40:18 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: check String encoding when
+	  converting String to VT_BSTR in OLE.
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+Sat Jun 28 01:08:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_timespec): fix rounding negative float.
+
+Fri Jun 27 21:38:57 2008  Tanaka Akira  <akr@fsij.org>
+
+	* struct.c: __size__ removed.  use the length of __members__ instead.
+	  (num_members): new function.
+
+Fri Jun 27 21:19:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/inlinetest.rb (InlineTest.in_progname): workaround for frozen
+	  $0.  [ruby-dev:35261]
+
+	* lib/test/unit/ui/console/testrunner.rb (TestRunner#finished): ditto.
+
+Fri Jun 27 17:45:17 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/erb.rb: adjust line number for magic comment.
+
+	* test/erb/test_erb.rb: add tests for def_method.
+
+Fri Jun 27 14:29:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (primary): empty not should call '!' on nil.
+	  cf [ruby-dev:35227]
+
+Fri Jun 27 14:25:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_eval_body): if thrown exception is frozen, reraise it to
+	  create a new instance.
+
+Fri Jun 27 13:29:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_str_new2, rb_tainted_str_new2,
+	  rb_usascii_str_new2): use with-length versions with strlen to
+	  optimize strlen, if optimized.
+
+Fri Jun 27 12:28:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/un.rb (mkmf): new command to create makefile.
+
+Fri Jun 27 11:06:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/un.rb (wait_writable): added help message.
+
+Fri Jun 27 06:52:54 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* configure.in: need a ",".
+
+	* process.c (rb_fork): check CANNOT_FORK_WITH_PTHREAD macro.
+
+Fri Jun 27 06:50:56 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (vm_call_super): fix to call method_missing.
+	  [ruby-core:15719], [ruby-core:17340]
+
+Fri Jun 27 00:00:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (set_arg0, ruby_prog_init): freeze $0.  a patch from Keita
+	  Yamaguchi <keita.yamaguchi at gmail.com>.
+
+Thu Jun 26 23:58:29 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (rb_str_index_m, rb_str_rindex_m, rb_str_include): fix
+	  rdoc.
+
+Thu Jun 26 17:43:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_f_trace_var): should not be allowed at safe level 4.
+	  a patch from Keita Yamaguchi <keita.yamaguchi at gmail.com>.
+
+Thu Jun 26 11:04:30 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems: Update to RubyGems 1.2.0 r1824.  Incorporates patch by
+	  Yusuke ENDOH [ruby-core:17353].
+
+Thu Jun 26 00:48:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (primary): not operand might be empty.  [ruby-dev:35227]
+
+Wed Jun 25 21:54:34 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (primary): make functional-style not operator to act
+	  like function.  see <http://d.hatena.ne.jp/ku-ma-me/20080624/p1>.
+
+Wed Jun 25 15:28:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_fill): not depend on unspecified behavior at integer
+	  overflow.  reported by Vincenzo Iozzo <snagg AT openssl.it>.
+
+Wed Jun 25 13:42:44 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/erb.rb (ERB::Compiler:Buffer#new): push magic comment first.
+
+	* lib/erb.rb (ERB::Compiler#compile): fix for broken input.
+
+Wed Jun 25 12:10:01 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/erb.rb (ERB::Compiler#compile): magic comment needs LF.
+
+Wed Jun 25 09:31:11 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/erb.rb (ERB::Compiler#compile): output magic comment.
+
+Tue Jun 24 22:14:36 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_eval.c (eval_string_with_cref): preserve parse_in_eval even if
+	  exception raised.
+
+Tue Jun 24 22:09:18 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(ole_invoke): fix memory leak.
+	  [ruby-bugs-20792]
+
+Tue Jun 24 17:20:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_fork_with_pthread): check after check for
+	  pthread library, and define the macro when checked only.
+
+Tue Jun 24 17:04:39 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/io/wait/extconf.rb: check sys/socket.h for cygwin.
+
+Tue Jun 24 16:51:51 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_reopen): remove cygwin handling because it seems to be for
+	  C's stdio.
+	  fixed [ruby-dev:35183]
+
+Tue Jun 24 11:12:33 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h, win32/win32.c (rb_w32_getppid): now support
+	  getppid() on win32 (but only Win2k or later).
+
+	* process.c (get_ppid): remove win32 special logic.
+
+Tue Jun 24 09:40:47 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (init_sock): socket is binmode on platforms
+	  which support binmode.
+
+Tue Jun 24 00:21:53 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_build_from_ary): initialize arg_opts, a patch from
+	  Adam Strzelecki <ono at java.pl> in [ruby-core:17220].
+
+Tue Jun 24 00:10:53 2008  wanabe  <s.wanabe@gmail.com>
+
+	* compile.c (iseq_build_from_ary): fix expression to obtain
+	  iseq->local_size and iseq->local_table_size.  [ruby-dev:35205]
+
+Mon Jun 23 11:31:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mathn.rb (Rational::power2): removed incomplete method.
+	  see [ruby-dev:35195].   [ruby-core:17293]
+
+Sun Jun 22 14:16:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/extconf.rb (have_readline_func): readline on Mac OS X
+	  needs headers to detect some functions.
+
+Sun Jun 22 09:51:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_class_superclass): rdoc improvement, a patch from
+	  Gaston Ramos <ramos.gaston AT gmail.com> in [ruby-core:17371].
+
+Sun Jun 22 09:22:32 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* README.EXT: translated README.EXT.ja
+
+Sun Jun 22 00:42:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_osid, rb_w32_osver, CreateChild): XP is
+	  is different from Vista about pipe handle inheritance.
+	  fixed [ruby-core:17367], reported by Lars Christensen <larsch at
+	  belunktum.dk>
+
+Sun Jun 22 00:38:45 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* README.EXT.ja: add note about mark and free.
+
+Sun Jun 22 00:01:36 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_binmode_p, argf_binmode_p, Init_IO): new method
+	  IO#binmode? and ARGF.binmode?  [ruby-dev:35148]
+
+Sat Jun 21 17:33:50 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn): no longer support P_WAIT.
+
+Sat Jun 21 16:46:09 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (native_sleep): must block reentrance when accessing
+	  th->unblock.
+	  fixed [ruby-core:17341], reported by Bill Kelly <billk at cts.com>
+
+Sat Jun 21 16:29:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (call_args2, open_args): removed.
+
+	* parse.y (parser_yylex): unified warnings at space between method
+	  name and argument parenthesis.  [ruby-dev:33943]
+
+Sat Jun 21 16:21:07 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_round): get rid of overflow.
+
+Sat Jun 21 15:57:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc#collect_first_comment): skip
+	  magic comment.
+
+Sat Jun 21 15:54:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if fork works with pthread.
+
+Sat Jun 21 15:31:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: ported to ruby.
+
+	* tool/make-snapshot: fixed digests.
+
+Sat Jun 21 04:36:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/jacobian.rb (Jacobian::dfdxi):
+	  typo fixed (raize -> raise).  [ruby-list:45101]
+
+	* enumerator.c (enum_each_cons): typo in RDoc fixed.
+
+Sat Jun 21 00:45:34 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* tool/make-snapshot: do not use sha256sum; use BASERUBY instead
+
+	* common.mk (dist): use tool/make-snapshot instead
+
+Fri Jun 20 16:34:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (Init_process): Process::Status#to_int removed.
+	  (PST2INT): defined.
+	  (pst_to_s): use PST2INT.
+	  (pst_inspect): ditto.
+	  (pst_equal): ditto.
+	  (pst_bitand): ditto.
+	  (pst_rshift): ditto.
+	  (pst_wifstopped): ditto.
+	  (pst_wstopsig): ditto.
+	  (pst_wifsignaled): ditto.
+	  (pst_wtermsig): ditto.
+	  (pst_wifexited): ditto.
+	  (pst_wexitstatus): ditto.
+	  (pst_success_p): ditto.
+	  (pst_wcoredump): ditto.
+	  (rb_f_system): ditto.
+
+Fri Jun 20 15:40:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_store, rb_ary_splice): not depend on unspecified
+	  behavior at integer overflow.
+
+	* string.c (str_buf_cat): ditto.
+
+Fri Jun 20 12:39:55 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_detach_process): store detached process ID in the
+	  thread local storage.  moved from lib/open3.rb.
+
+Fri Jun 20 11:57:46 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_sub_bang): should preserve replacement points
+	  since they may be altered in the yielded block.
+
+Fri Jun 20 11:07:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_memhash): randomize hash to avoid algorithmic
+	  complexity attacks.
+	  (rb_str_hash): use rb_memhash.
+
+	* include/ruby/intern.h (rb_reset_random_seed): declared.
+
+	* thread.c (rb_thread_atfork): call rb_reset_random_seed.
+
+	* inits.c (rb_call_inits): call Init_RandomSeed at first.
+
+	* random.c (seed_initialized): defined.
+	  (fill_random_seed): extracted from random_seed.
+	  (make_seed_value): extracted from random_seed.
+	  (rb_f_rand): initialize random seed at first.
+	  (initial_seed): defined.
+	  (Init_RandomSeed): defined.
+	  (Init_RandomSeed2): defined.
+	  (rb_reset_random_seed): defined.
+	  (Init_Random): call Init_RandomSeed2.
+
+Wed Jun 18 21:52:38 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* array.c (ary_new, rb_ary_initialize, rb_ary_store,
+	  rb_ary_splice, rb_ary_times): integer overflows should be
+	  checked. based on patches from Drew Yao <ayao at apple.com>
+	  fixed CVE-2008-2726
+
+	* string.c (rb_enc_cr_str_buf_cat): fixed unsafe use of alloca,
+	  which led memory corruption. based on a patch from Drew Yao
+	  <ayao at apple.com> fixed CVE-2008-2726
+
+Fri Jun 20 03:26:00 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_f_fork): NetBSD 4.0 or later can fork.
+
+Fri Jun 20 03:19:39 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/testunit/collector/test_dir.rb: r15825 made it unnecessary to
+	  change String to Symbol.
+
+	* test/testunit/collector/test_objectspace.rb: ditto.
+
+Fri Jun 20 03:14:31 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*, test/rubygems/*:  Update to RubyGems 1.1.1 r1784 (pre
+	  1.2).
+
+Fri Jun 20 03:01:59 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c: try to remove false positive of deadlock detection (second
+	  trial).
+
+Fri Jun 20 02:16:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mathn.rb (Rational::power2): typo fixed.  [ruby-core:17293]
+
+Fri Jun 20 02:11:01 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): should preserve last successful match
+	  data.  [ruby-dev:35182]
+
+Fri Jun 20 01:07:28 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* KNOWNBUGS.rb, bootstraptest/pending.rb: move a bug (?) to pending.
+
+Fri Jun 20 00:40:08 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (proc_new): fix to return Proc object if block is already
+	  in heap.  [ruby-core:15711]
+
+	* bootstraptest/test_proc.rb: add a test.
+
+Fri Jun 20 00:18:04 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_win32.c (native_sleep): fix to decrement sleeper count.
+
+Thu Jun 19 23:48:45 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/net/http/test_http.rb: compare encodings of two strings before
+	  comparing themself, which suppress too big error output.
+
+Thu Jun 19 23:46:35 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/json/ext/parser/parser.rl, ext/json/ext/parser/parser.c: JSON
+	  text SHALL be encoded in Unicode.
+
+Thu Jun 19 23:17:56 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c, thread_win32.c, vm_core.h: try to remove false positive of
+	  deadlock detection.
+
+Thu Jun 19 21:38:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): check if compile before showing message.
+
+Thu Jun 19 21:35:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: supported multiple snapshots.
+
+Thu Jun 19 20:37:00 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/net/pop.rb (Net::POP3#set_all_uids): speed
+	  up. a patch from <m-sumi AT techfirm.co.jp> [ruby-list:45047]
+
+Thu Jun 19 17:44:15 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/etc/etc.c (Init_etc): define constant aliases Etc::Passwd
+	  and Etc::Group.  [ruby-dev:35150]
+
+Thu Jun 19 17:37:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_alloc): specify 'inline' modifier.
+
+	* string.c (str_alloc): remove cSymbol hack that no longer
+	  necessary.
+
+	* string.c (scan_once): avoid retrieving encoding info unless
+	  necessary.
+
+Thu Jun 19 17:19:14 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_scan): String#scan should preserve last
+	  successful match data.  [ruby-dev:35106]
+
+Thu Jun 19 16:49:36 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* missing/acosh.c (atanh): should set ERANGE to errno if parameter
+	  is the boundary case. fixed [ruby-dev:35155]
+
+Thu Jun 19 16:06:01 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkextlib/tile/treeview.rb: cannot configure tags.
+
+Thu Jun 19 11:48:33 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_enumerator.rb: fix to skip "with_memo" test.
+
+Thu Jun 19 11:40:55 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_throw): fix "return" process from "lambda".
+
+	* bootstraptest/test_proc.rb: add a test.
+
+	* bootstraptest/pending.rb: add a pending bug.
+
+Thu Jun 19 00:33:40 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/etc/test_etc.rb: avoid infinite loop.  [ruby-dev:35158]
+
+Wed Jun 18 23:07:19 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* gc.c (rb_newobj): prohibit call of rb_newobj() during gc.
+	  a patch from Sylvain Joyeux in [ruby-core:12099].
+
+Wed Jun 18 21:08:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (verbose_setter, opt_W_getter): fixed prototypes.
+
+Wed Jun 18 19:20:00 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (opt_W_getter): use ruby_verbose directly instead of parameter
+	  because ruby_verbose is not a real variable, so the address of
+	  parameter is not collect.
+
+Wed Jun 18 18:31:03 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (errmap): add some pipe errors.
+
+	* win32/win32.c (rb_w32_write): set errno when CRT's errno is EINVAL
+	  for pipe errors.
+
+Wed Jun 18 18:09:08 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (poll_child_status): set EINVAL to errno when
+	  GetExitCodeProcess() fails with ERROR_INVALID_HANDLE.
+
+Wed Jun 18 15:01:18 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_open_file): fs_encoding and fname_encoding is
+	  rb_encoding *.
+	  fixed [ruby-dev:35151]
+
+Wed Jun 18 14:30:06 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_open_file): not rb_enc_get_index but rb_enc_get.
+
+Wed Jun 18 13:49:58 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (pipe): now pipe is textmode. although this
+	  change is experimental, it will be spec if no compatibility problem
+	  is reported.
+
+Wed Jun 18 12:05:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object, marshal_dump, r_object0, marshal_load): search
+	  public methods only.  [ruby-core:17283]
+
+	* object.c (convert_type): ditto.
+
+	* lib/singleton.rb (Singleton#_dump): conversion method should be
+	  public.
+
+Wed Jun 18 10:18:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_passwd, etc_group): fixed rdoc.  a patch from
+	  okkez <okkez000 AT gmail.com> in [ruby-dev:35141].
+
+Wed Jun 18 08:58:16 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/*:  Fix errors for 1.9.
+
+	* gem_prelude.rb:  Only remove methods from gem_prelude.rb when
+	  loading real RubyGems.
+
+Wed Jun 18 07:03:30 2008  Eric Hodel  <drbrain@egment7.net>
+
+	* lib/rubygems/*:  Update to RubyGems r1778 (pre 1.2).
+
+Wed Jun 18 04:27:58 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* KNOWNBUGS.rb, bootstraptest/pending.rb: move pending bug.
+
+Wed Jun 18 04:24:20 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, vm_insnhelper.c: fix escape process with "break" and "return"
+	  syntax in "lambda".  [ ruby-Bugs-19304 ], [ruby-core:17164]
+
+	* KNOWNBUGS.rb, bootstraptest/test_proc.rb: add/move solved test.
+
+Wed Jun 18 01:51:10 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: cannot access class variable from
+	  singleton method.
+
+Wed Jun 18 00:03:33 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* dir.c (Init_Dir): dir_foreach() takes variable argument.
+
+Tue Jun 17 23:04:24 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/net/telnet.rb:  Fixing Telnet#wairfor() which was broken by
+	  changes to the Kernel::Integer() method.  [ruby-core:17272]
+
+Tue Jun 17 23:02:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (opt_W_getter): made a hooked variable.
+
+Tue Jun 17 22:04:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enc/euc_jp.c (property_name_to_ctype): core dumped when sizeof(int)
+	  differs from sizeof(long).  [ruby-dev:35131]
+
+	* enc/shift_jis.c (property_name_to_ctype): ditto.
+
+	* enc/unicode.c (onigenc_unicode_property_name_to_ctype): ditto.
+
+Tue Jun 17 20:32:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (miniruby$(EXEEXT)): $(PREP) isn't always same as
+	  miniruby, and tests, debug, etc have no meaning when
+	  cross-compiling.
+
+Tue Jun 17 18:39:11 2008  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* common.mk: fixed dependencies on miniruby.
+
+Tue Jun 17 18:11:01 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (pipe): expand pipe buffer size.
+
+Tue Jun 17 17:07:35 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): no need to inherit handles here because
+	  spawn'ed child cannot detect that STDIN is closed.
+
+Tue Jun 17 06:32:55 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (dir_data): add intenc and extenc.
+
+	* dir.c (dir_s_alloc): ditto.
+
+	* dir.c (dir_initialize): now accept internal_encoding and
+	  external_encoding.
+
+	* dir.c (dir_s_open): changed for dir_initialize.
+
+	* dir.c (dir_open_dir): ditto.
+
+	* dir.c (dir_foreach): changed for dir_open_dir.
+
+	* dir.c (dir_entries): changed for dir_open_dir.
+
+	* dir.c (dir_enc_str): defined.
+
+	* dir.c (dir_path): use dir_enc_str.
+
+	* dir.c (dir_read): ditto.
+
+	* dir.c (dir_each): ditto.
+
+Tue Jun 17 06:28:57 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_set_encoding): defined.
+
+	* io.c (rb_open_file): convert path on Windows and Mac OS X.
+
+	* io.c (open_key_args): use io_set_encoding and now accept
+	  internal_encoding and external_encoding.
+
+Tue Jun 17 06:26:28 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_filesystem_encoding): defined.
+
+	* include/ruby/encoding.h (rb_filesystem_encoding): added.
+
+Tue Jun 17 06:24:28 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/syck/rubyext.c: add encoding header.
+
+Tue Jun 17 01:52:50 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: SEGV when exit.
+
+	* ext/tk/lib/tk.rb: add a check for safety to Tk.exit.
+
+	* ext/tk/sample/irbtkw.rbw: freezes when receives SIGINT.
+
+Mon Jun 16 21:58:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_each, strio_readlines): IO#each and
+	  IO#readlines do not affect $_.  [ruby-core:17277]
+
+Mon Jun 16 18:52:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/thwait.rb (ThreadsWait): Exception2MessageMapper no longer has
+	  extend_to method.  [ruby-core:17267]
+
+Mon Jun 16 14:46:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/e2mmap.rb (E2MM.def_e2message): typo.
+
+Mon Jun 16 09:43:27 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_with_object, Init_Enumerator):
+	  Temporarily back out with_memo, for which we need a better name.
+
+Mon Jun 16 07:14:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_readline, strio_each)
+	  (strio_readlines): set lastline.  [ruby-core:17257]
+
+Mon Jun 16 01:49:39 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_f_block_given_p): fix to skip class frame.
+	  [ruby-core:14813]
+
+	* KNOWNBUGS.rb, bootstraptest/test_method.rb: move solved test.
+
+Mon Jun 16 01:48:08 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c (vm_stack_dump_raw): disable verbose debug output.
+
+Mon Jun 16 01:33:08 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, thread.c: rename global_interpreter_lock to
+	  global_vm_lock.
+
+Sun Jun 15 18:40:35 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/syck/rubyext.c (yaml_org_handler): associate encoding.
+
+	* ext/syck/rubyext.c (syck_genericresolver_node_import): ditto.
+
+Sun Jun 15 18:17:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_vm_t), vm.c (rb_vm_mark): moved preallocated special
+	  exceptions.
+
+	* eval.c (Init_eval), gc.c (Init_GC), proc.c (Init_Proc): freeze
+	  preallocated special exceptions.
+
+	* eval.c (rb_longjmp): duplicate the thrown exception to set backtrace
+	  if it was frozen.
+
+	* gc.c (rb_memerror): raise nomem_error without backtrace if failed to
+	  make backtrace.
+
+Sat Jun 14 22:52:35 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_sysread): should not raise at empty
+	  read.  a patch from Arthur Schreiber at [ruby-core:17245].
+
+Sat Jun 14 16:55:46 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): no need to expand root path which has no
+	  short file name.  [ruby-dev:35095]
+
+Sat Jun 14 11:59:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.h (STACK_UPPER): moved from gc.c
+
+	* thread.c, thread_{pthread,win32}.c (ruby_init_stack,
+	  ruby_thread_init_stack): moved stack initialization from gc.c.
+
+Sat Jun 14 11:57:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.h (STACK_UPPER): moved from gc.c
+
+	* thread.c, thread_{pthread,win32}.c (ruby_init_stack,
+	  ruby_thread_init_stack): moved stack initialization from gc.c.
+
+Sat Jun 14 07:52:53 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (ruby_initial_gc_stress): defined.
+	  (ruby_initial_gc_stress_ptr): defined.
+
+	* debug.c (set_debug_option): use ruby_initial_gc_stress_ptr for
+	  gc_stress option.
+
+Sat Jun 14 00:09:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (ruby_gc_stress): moved to rb_objspace_t.
+
+	* gc.c (gc_stress_get, gc_stress_set): VM local attribute.
+
+	* signal.c (sigsegv): ditto.
+
+Fri Jun 13 21:55:48 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_equal_p): Rational(0,x) and 0 are equivalent,
+	  anyway.
+
+Fri Jun 13 21:26:39 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (string_to_c, nucomp_s_convert): preserve the current
+	  backref.
+
+	* rational.c (string_to_r, nurat_s_convert): ditto.
+
+	* include/ruby/intern.h (rb_match_busy): added a declaration.
+
+Fri Jun 13 18:08:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.xmlschema): don't accept decimal dot without
+	  fractional digits.  fractional digits handling simplified.
+
+Fri Jun 13 17:20:40 2008  wanabe  <s.wanabe@gmail.com>
+
+	* complex.c (string_to_c_internal): save and restore backref.
+	  fixed [ruby-dev:34991]
+
+Fri Jun 13 17:06:20 2008  wanabe  <s.wanabe@gmail.com>
+
+	* rational.c (string_to_r_internal): save and restore backref.
+	  fixed [ruby-dev:34990]
+
+Fri Jun 13 14:41:26 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* README.EXT.ja: update about Fixnum. reported in
+	   <http://www.tmtm.org/ja/tdiary/?date=20080611#p01>
+
+	* README.EXT.ja: describe about StringValueCStr().
+
+	* README.EXT: ditto.
+
+Fri Jun 13 14:24:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_memerror): exit with EXIT_FAILURE instead of magic number.
+
+	* gc.c (ruby_stack_check): STACK_LENGTH should be less than
+	  STACK_LEVEL_MAX.
+
+Fri Jun 13 12:55:37 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (sort_by_i): use NODE_DOT2 instead of NODE_MEMO to avoid
+	  extra calls to is_pointer_to_heap() in GC.
+
+	* enum.c (enum_zip): ditto.
+
+Fri Jun 13 00:41:58 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_thread.rb: add a test.
+
+Thu Jun 12 23:30:11 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (mutex_unlock): fix cond_notified consistency.
+
+Thu Jun 12 22:19:45 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (native_sleep): fixed previous commit.
+
+Thu Jun 12 21:59:17 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c, vm_core.h, vm.c, thread_pthread.c, thread_win32.c: add
+	  deadlock detection.  [ruby-dev:35044]
+
+	* bootstraptest/test_thread.rb: add tests for above.
+
+Thu Jun 12 21:39:55 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: refactoring.
+
+	* rational.c: ditto.
+
+Thu Jun 12 17:11:24 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regint.h: undefine USE_CAPTURE_HISTORY which is mentioned as
+	  unsupported in the Onigiruma document.
+
+Thu Jun 12 13:36:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (rb_intern): use rb_intern2 with strlen for
+	  constant symbols to optimize strlen.
+
+Thu Jun 12 08:47:51 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (read_all): should use io_read_encoding(), not
+	  io_input_encoding().
+
+	* io.c (rb_io_getline_1): reduce calling of io_read_encoding().
+
+	* string.c (rb_str_scan): need not to restore $~ value, so avoid
+	  pinning match object.
+
+Thu Jun 12 02:49:40 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/stringio/stringio.c (strio_init): rewind when reopened.
+
+Thu Jun 12 02:43:27 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_zip): ANSI style.
+
+Thu Jun 12 02:25:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_reopen): clear read buffer.
+
+Thu Jun 12 00:56:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk: check proper conditions.  [ruby-dev:35047]
+
+Wed Jun 11 23:33:13 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_fread): bypass buffered read if reading buffer is empty.
+
+	* io.c (remain_size): do not add extra one byte.
+
+Wed Jun 11 12:15:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_normal_exit): hide stderr output
+	  when success.
+
+Wed Jun 11 09:26:43 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (encs): need to pass miniruby path for windows.
+
+Wed Jun 11 05:53:20 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, eval_intern.h (PASS_PASSED_BLOCK):
+	  set a VM_FRAME_FLAG_PASSED flag to skip this frame when
+	  searching ruby-level-cfp.
+
+	* eval.c, eval_intern.h, proc.c: fix to check cfp.  if there is
+	  no valid ruby-level-cfp, cause RuntimeError exception.
+	  [ruby-dev:34128]
+
+	* vm_core.h, vm_evalbody.c, vm.c, vm_dump.c, vm_insnhelper.c,
+	  insns.def: rename FRAME_MAGIC_* to VM_FRAME_MAGIC_*.
+
+	* KNOWNBUGS.rb, bootstraptest/test*.rb: move solved bugs.
+
+Wed Jun 11 05:55:31 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: SEGV when tcltk-stubs is enabled.
+
+	* ext/tk/tcltklib.c: avoid error on a shared object.
+
+	* ext/tk/extconf.rb: support --with-tcltkversion
+
+	* ext/tk/README.tcltklib: add document about --with-tcltkversion
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/multi-tk.rb, ext/tk/lib/remote-tk.rb:
+	  not work on $SAFE==4
+
+	* ext/tk/lib/multi-tk.rb: Object#methods returns Symbols on Ruby1.9.
+
+	* ext/tk/lib/tk/timer.rb: add TkTimer#at_end(proc) to register the
+	  procedure which called at end of the timer.
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tk/itemfont.rb, ext/tk/lib/font.rb:
+	  support __IGNORE_UNKNOWN_CONFIGURE_OPTION__ about font options.
+
+	* ext/tk/lib/*: treat __IGNORE_UNKNOWN_CONFIGURE_OPTION__
+
+	* ext/tk/lib/tkextlib/iwidgets/scrolledcanvas.rb,
+	  ext/tk/lib/tkextlib/iwidgets/scrolledlistbox.rb,
+	  ext/tk/lib/tkextlib/iwidgets/scrolledtext.rb: bug fix.
+
+	* ext/tk/lib/tk/text.rb: typo. call a wrong method.
+
+	* ext/tk/lib/tk/itemconfig.rb: ditto.
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tk/itemconfig.rb,
+	  ext/tk/lib/tk/canvas.rb: support alias names of option keys.
+
+	* ext/tk/lib/tk/grid.rb: lack of module-method definitions.
+
+	* ext/tk/lib/tk/pack.rb, ext/tk/lib/tk/grid.rb: increase supported
+	  parameter patterns of configure method.
+
+	* ext/tk/lib/tk.rb: add TkWindow#grid_anchor, grid_column, grid_row.
+
+	* ext/tk/lib/tk/wm.rb: methods of Tk::Wm_for_General module cannot
+	  pass the given block to methods of Tk::Wm module.
+
+	* ext/tk/lib/tk/wm.rb: Wm#overrideredirect overwrites argument to
+	  an invalid value.
+
+	* ext/tk/lib/tk.rb: fix memory (object) leak bug.
+
+	* ext/tk/tcltklib.c, ext/tk/tkutil/tkutil.c: fix memory leak.
+
+	* ext/tk/sample/demos-jp/aniwave.rb, ext/tk/sample/demos-en/aniwave.rb:
+	  bug fix.
+
+	* ext/tk/lib/tkextlib/blt/component.rb,
+	  ext/tk/lib/tkextlib/tile/tentry.rb,
+	  ext/tk/lib/tkextlib/tile/treeview.rb: ditto.
+
+	* ext/tk/lib/tkextlib/tile/tpaned.rb: improve TPaned#add.
+
+	* ext/tk/sample/demos-jp/widget, ext/tk/sample/demos-en/widget,
+	  ext/tk/sample/demos-jp/style.rb, ext/tk/sample/demos-en/style.rb,
+	  ext/tk/sample/demos-jp/bind.rb, ext/tk/sample/demos-en/bind.rb:
+	  bug fix.
+
+	* ext/tk/sample/ttk_wrapper.rb: ditto.
+
+	* ext/tk/sample/ttk_wrapper.rb: support "if __FILE__ == $0" idiom.
+
+	* ext/tk/sample/tktextio.rb: add binding for 'Ctrl-u' at console mode.
+
+	* ext/tk/lib/tkextlib/tile.rb, ext/tk/lib/tkextlib/tile/style.rb,
+	  ext/tk/sample/ttk_wrapper.rb: improve treating and control themes.
+	  add Tk::Tile.themes and Tk::Tile.set_theme(theme).
+
+	* ext/tk/lib/tkextlib/tile.rb: lack of autoload definitions.
+
+	* ext/tk/lib/tkextlib/tile/tnotebook.rb: cannot use kanji (not UTF-8)
+	  characters for headings.
+
+	* ext/tk/lib/tkextlib/tkDND/shape.rb: wrong package name.
+
+	* ext/tk/tkutil/tkutil.c: improve handling callback-subst-keys.
+	  Now, support longnam-keys (e.g. '%CTT' on tkdnd-2.0; however, still
+	  not support tkdnd-2.0 on tkextlib), and symbols of parameters (e.g.
+	  :widget=>'%W', :keycode=>'%k', '%x'=>:x, '%X'=>:root_x, and so on;
+	  those are attributes of event object). It means that Ruby/Tk accepts
+	  not only "widget.bind(ev, '%W', '%k', ...){|w, k, ...| ... }", but
+	  also "widget.bind(ev, :widget, :keycode, ...){|w, k, ...| ... }".
+	  It is potentially incompatible, when user passes symbols to the
+	  arguments of the callback block (the block receives the symbols as
+	  strings). I think that is very rare case (probably, used by Ruby/Tk
+	  experts only). When causes such trouble, please give strings instead
+	  of such symbol parameters (e.g. call Symbol#to_s method).
+
+	* ext/tk/lib/tk/event.rb, ext/tk/lib/tk/validation.rb,
+	  ext/tk/lib/tkextlib/blt/treeview.rb,
+	  ext/tk/lib/tkextlib/winico/winico.rb: ditto.
+
+	* ext/tk/tkutil/tkutil.c: strings are available on subst_tables on
+	  TkUtil::CallbackSubst class (it is useful on Ruby 1.9).
+
+	* ext/tk/lib/tk/spinbox.rb, ext/tk/lib/tkextlib/iwidgets/hierarchy.rb,
+	  ext/tk/lib/tkextlib/iwidgets/spinner.rb,
+	  ext/tk/lib/tkextlib/iwidgets/entryfield.rb,
+	  ext/tk/lib/tkextlib/iwidgets/calendar.rb,
+	  ext/tk/lib/tkextlib/blt/dragdrop.rb,
+	  ext/tk/lib/tkextlib/tkDND/tkdnd.rb,
+	  ext/tk/lib/tkextlib/treectrl/tktreectrl.rb,
+	  ext/tk/lib/tkextlib/tktable/tktable.rb: disable code piece became
+	  unnecessary by reason of the changes of ext/tk/tkutil/tkutil.c.
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/multi-tk.rb: change strategy to define
+	  the constant WITH_ENCODING.
+
+	* ext/tk/lib/tk.rb: fix bug on Tk::Encoding.tk_encoding_names.
+
+Wed Jun 11 03:40:37 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/find.rb (Find#find): Return an enumerator if no block is
+	  given.
+
+Wed Jun 11 01:28:12 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/intern.h, proc.c: revert rb_proc_call() and
+	  create rb_proc_call_with_block() instead.
+
+	* include/ruby/ruby.h, eval_jump.c, thread.c, vm_insnhelper.c:
+	  rb_blockptr should not be exposed.
+
+Tue Jun 10 21:07:19 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_float.rb: add tests. [ruby-dev:35009]
+
+Tue Jun 10 20:55:57 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_s_convert): need not to initialize optional
+	  argument for rb_scan_args().
+
+Tue Jun 10 20:13:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: negate default of --without-ext if --with-ext is
+	  given.
+
+	* ext/extmk.rb: negate default of --without-ext.
+
+Tue Jun 10 17:43:35 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_shuffle_bang): update RDoc.  [ruby-dev:35034]
+
+Tue Jun 10 17:30:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_obj_instance_exec, rb_mod_module_exec):
+	  added prototypes.
+
+Tue Jun 10 17:00:29 2008  wanabe  <s.wanabe@gmail.com>
+
+	* util.c (ruby_strtod): ruby_strtod don't allow a trailing
+	  decimal point like "7.". [ruby-dev:34835] [ruby-dev:35009]
+
+Tue Jun 10 13:48:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* rational.c (nurat_s_convert): need not to initialize optional
+	  argument for rb_scan_args().
+
+Tue Jun 10 12:58:07 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/wait/wait.c (FIONREAD_POSSIBLE_P): suppress warnings.
+
+Tue Jun 10 12:43:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (CONST_ID_CACHE): fixed statement expression.
+
+Tue Jun 10 11:25:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (rb_argv0): revised for ext/tk.
+
+	* include/ruby/encoding.h: not to use varargs.h since requiring C89.
+
+Tue Jun 10 00:50:51 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* include/ruby/ruby.h, vm_core.h: add a type rb_blockptr.
+
+	* vm_insnhelper.c (vm_yield_with_cfunc): vm_yield_with_cfunc receives
+	  blockptr and passes it to iterating block.
+
+	* proc.c (rb_proc_call), include/ruby/intern.h: rb_proc_call receives
+	  blockptr.  "rb_proc_call(self, args, blockptr)" in C corresponds to
+	  "self.call(*args, &block)" in Ruby.
+
+	* proc.c (proc_call): pass blockptr to block that is written in C.
+
+	* proc.c (curry): receive blockptr and pass it to original proc.
+	  [ruby-core:15551]
+
+	* vm.c (invoke_block_from_c): fix for change of vm_yield_with_cfunc.
+
+	* thread.c (call_trace_proc), eval_jump.c (rb_call_end_proc): fix for
+	  change of rb_proc_call.
+
+Tue Jun 10 00:10:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (test-knownbug): give $(OPTS) for bootstraptest/runner.rb.
+
+Mon Jun  9 23:10:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (Init_stack): don't declare.  it is a macro now.
+
+Mon Jun  9 22:46:47 2008  wanabe  <s.wanabe@gmail.com>
+
+	* compile.c : treat []&&= in virtually the same way as []||=.
+	  [ruby-dev:34679]
+
+Mon Jun  9 21:17:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (clean): remove build directories.
+
+	* test_knownbug.rb -> KNOWNBUGS.rb: renamed.
+
+	* common.mk: apply above change.
+
+Mon Jun  9 21:14:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): set flags.
+
+Mon Jun  9 21:09:02 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb -> ./test_knownbug.rb: moved.
+
+	* common.mk: add a rule "test-knownbug".
+
+Mon Jun  9 21:00:32 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_s_convert): can accept Complex('i').
+	  [ruby-dev:34991]
+
+Mon Jun  9 18:25:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (CONST_ID): constant ID cache for non-gcc.
+
+	* *.c: no cache in init functions.
+
+Mon Jun  9 17:56:30 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#delete_if): Call to_a.
+	  (SortedSet#delete_if, TC_SortedSet#test_sortedset): Use super to
+	  yield elements in sorted order; [ruby-core:17144] by Arthur
+	  Schreiber.
+	  (SortedSet#each, SortedSet#each, TC_Set#test_each)
+	  (TC_SortedSet#test_sortedset): Return self; [ruby-dev:35002] by
+	  Arthur Schreiber.
+
+Mon Jun  9 17:47:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (fptr_finalize): close IO object if fd is already closed.
+	  (rb_p): call rb_io_write just once.
+
+Mon Jun  9 15:37:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (require_libraries): req_list may be NULL.  [ruby-dev:35008]
+
+Mon Jun  9 14:18:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (struct rb_vm_struct): moved src_encoding_index,
+	  ruby_debug, ruby_verbose, and rb_progname.
+
+	* ruby.c (rb_argv0): no longer used.
+
+	* ruby.c (struct cmdline_options): moved setids and req_list, and the
+	  latter is now an array, to prevent memory leak.
+
+	* ruby.c (cmdline_options_init): added.
+
+	* ruby.c (add_modules, require_libraries, init_ids, forbid_setid): use
+	  struct cmdline_options.
+
+	* vm.c (vm_init2): initialize src_encoding_index.
+
+	* vm.c: getters/setters for ruby_{debug,verbose}.
+
+Mon Jun  9 09:54:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (Init_stack): make to call ruby_init_stack.
+
+Mon Jun  9 08:12:40 2008  wanabe  <s.wanabe@gmail.com>
+
+	* vm_insnhelper.c, vm.c, proc.c : revert r17021. [ruby-dev:34997]
+
+Mon Jun  9 03:12:23 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/pending.rb: move/remove solved issues.
+
+	* bootstraptest/test_class.rb: ditto.
+
+Mon Jun  9 02:32:58 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/zlib/zlib.c (rb_deflate_init_copy): Copy buffers as well.
+	  [ruby-list:45018]
+
+Sun Jun  8 22:22:20 2008  wanabe  <s.wanabe@gmail.com>
+
+	* vm_insnhelper.c, vm.c, proc.c (proc_call): allow call method with
+	  block that both is written in C. [ruby-dev:34273] [ruby-core:15551]
+
+	*  proc.c (curry): use proc_call instead of rb_proc_call.
+	  [ruby-dev:34273] [ruby-core:15551]
+
+Sun Jun  8 21:50:27 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/zlib/test_zlib.rb: add tests to achieve over 90% test coverage
+	  of zlib.
+
+Sun Jun  8 20:12:47 2008  wanabe  <s.wanabe@gmail.com>
+
+	* vm_insnhelper.c (vm_throw): regard break as return in lambda.
+	  [ruby-dev:34646]
+
+Sun Jun  8 19:17:59 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add a build option "CALC_EXACT_MALLOC_SIZE".
+	  This option enables to calculate exact size of current
+	  allocated size by malloc().  You can access these information
+	  with GC.malloc_allocated_size and GC.malloc_allocations.
+	  This option consume additional memory as a header of each memory
+	  object.  This option also helps to find out xmalloc()/xfree()
+	  consistency.  If you get trouble with this option, some extension
+	  using "free()" instead of "xfree()".
+	  This options is disabled by default.
+
+Sun Jun  8 18:15:38 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c, bignum.c, cont.c, dir.c, dln.c, encoding.c, enumerator.c,
+	  enumerator.c (enumerator_allocate), eval_jump.c, file.c, hash.c,
+	  io.c, load.c, pack.c, proc.c, random.c, re.c, ruby.c, st.c,
+	  string.c, thread.c, thread_pthread.c, time.c, util.c, variable.c,
+	  vm.c, gc.c:
+	  allocated memory objects by xmalloc (ruby_xmalloc) should be
+	  freed by xfree (ruby_xfree).
+
+	* ext/curses/curses.c, ext/dbm/dbm.c, ext/digest/digest.c,
+	  ext/gdbm/gdbm.c, ext/json/ext/parser/parser.c,
+	  ext/json/ext/parser/unicode.c, ext/openssl/ossl_cipher.c,
+	  ext/openssl/ossl_hmac.c, ext/openssl/ossl_pkey_ec.c,
+	  ext/sdbm/init.c, ext/strscan/strscan.c, ext/zlib/zlib.c:
+	  ditto.
+
+Sun Jun  8 01:15:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* hash.c (hash_i): make Hash#hash order insensitive.
+	  (rb_hash_dup): use DUPSETUP.
+
+Sat Jun  7 23:47:35 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/zlib/zlib.c (rb_deflate_initialize, Init_zlib): Fix up
+	  initialize_copy; [ruby-list:45016].
+
+Sat Jun  7 22:15:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (VENDOR_DIR): use LIBDIR instead of PREFIX as well as
+	  SITE_DIR.  a patch from Richard Brown <rbrown AT exherbo.org> in
+	  [ruby-core:17129].
+
+Sat Jun  7 21:37:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_f_open), re.c (rb_reg_search), transcode.c (str_transcode):
+	  suppress warnings.
+
+	* util.c (quorem, rv_alloc, nrv_alloc): only used in dtoa().
+
+Sat Jun  7 16:06:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (CFLAGS, CXXFLAGS): append default flags.
+
+Sat Jun  7 01:23:59 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_open_file, rb_io_s_sysopen): fmode should be unsigned int.
+	  fixed [ruby-dev:34979]
+
+Fri Jun  6 23:46:19 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_callee_setup_arg): check simple flag before
+	  calling setup_arg function().  this change reduce function call.
+
+Fri Jun  6 21:51:46 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (COMMON_HEADERS): include ws2tcpip.h.
+
+	* ext/socket/addrinfo.h (addrinfo, getaddrinfo, getnameinfo,
+	  freehostent, freeaddrinfo): undef before define because these are
+	  macros in some versions of Windows SDK.
+
+	  merged from ruby_1_8.
+
+Fri Jun  6 18:25:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/iconv/utils.rb (default_test): override not to croak.
+
+Fri Jun  6 16:41:45 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h: include ws2tcpip.h. fixed [ruby-Bugs-20528]
+
+Fri Jun  6 15:05:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (count_objects): clear hash after counting objects.
+
+Fri Jun  6 12:43:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_dir.rb (TestDir::test_glob): glob file names not sorted.
+
+Fri Jun  6 00:05:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.xmlschema): don't use float.  fix
+	  http://rubyforge.org/tracker/index.php?func=detail&group_id=426&atid=1698&aid=20504
+
+Thu Jun  5 23:56:18 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_gc.rb: add tests to achieve over 90% test coverage of
+	  gc.c.
+
+	* test/ruby/test_objectspace.rb: ditto.
+
+	* test/ruby/test_marshal.rb: ditto.
+
+Thu Jun  5 23:40:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* gc.c (rb_objspace_alloc): this function is needed only when
+	  ENABLE_VM_OBJSPACE macro is defined.
+
+	* vm.c: ditto.
+
+Thu Jun  5 23:31:21 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/stringio/test_stringio.rb: add tests to achieve over 95% test
+	  coverage of stringio.
+
+	* test/strscan/test_stringscanner.rb: ditto for strscan.
+
+Thu Jun  5 23:25:34 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval.c (ruby_finalize_0): clear trace_func after executing END
+	  procs.
+
+	* thread.c: fix typo.
+
+Thu Jun  5 22:50:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (os_obj_of): heaps may be modified in yield.
+
+Thu Jun  5 21:46:50 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* st.c (st_reverse_foreach): comment out unused function.
+
+	* util.c (dtoa): ditto.
+
+Thu Jun  5 20:30:46 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_initialize):
+	  Add a null check for ssl; submitted by akira yamada
+	  in [ruby-dev:34950].
+
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): Define OP_NO_TICKET if
+	  SSL_OP_NO_TICKET is present; submitted by akira yamada
+	  in [ruby-dev:34944].
+
+	* test/openssl/test_ssl.rb (OpenSSL#test_server_session): Add a
+	  workaround for the case where OpenSSL is configured with
+	  --enable-tlsext; submitted by akira yamada in [ruby-dev:34944].
+
+Thu Jun  5 20:24:15 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (thread_set_trace_func_m): fix check for proc argument.
+
+Thu Jun  5 20:17:29 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/rexml/document.rb (REXML::Document:write): leaky modification
+	  trans -> transitive.  [ruby-dev:32040], r13686
+
+	* lib/rexml/text.rb (Text.check): fix check for illegal character.
+
+Thu Jun  5 14:03:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_create): find encoding without options.
+
+Thu Jun  5 07:48:32 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (hash): should be "static".
+
+Thu Jun  5 01:47:18 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_transcode.rb: add tests for iso-2022-jp.
+
+Thu Jun  5 01:27:42 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_process.rb: add tests.
+
+Wed Jun  4 23:10:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/zlib/zlib.c (rb_gzfile_set_mtime): fix typo.
+
+Wed Jun  4 18:53:39 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_alloc): RDoc updated.  a patch from Gaston
+	  Ramos <ramos.gaston at gmail.com> in [ruby-core:17073].
+
+Wed Jun  4 18:36:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rdoc.rb: massive spelling correction patch from Evan Farrar
+	  <evanfarrar at gmail.com> in [ruby-doc:1382] applied.
+
+Wed Jun  4 17:52:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_iconv): fix for length argument and now
+	  allows range.  [ruby-core:17092]
+
+Wed Jun  4 15:45:41 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_with_index, enumerator_with_memo): Fix
+	  grammo in rdoc.
+
+Wed Jun  4 13:06:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (CFLAGS, CXXFLAGS): include additional flags to
+	  CFLAGS and CXXFLAGS while configuration.
+
+Tue Jun  3 23:06:35 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/strscan/strscan.c (strscan_scan_full, strscan_search_full): fix
+	  document.
+
+Tue Jun  3 22:37:26 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/strscan/strscan.c (strscan_exist_p): fix document.
+
+Tue Jun  3 22:33:29 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_dir.rb: add tests to achieve over 90% test coverage
+	  of dir.c.
+
+	* test/ruby/test_encoding.rb: add tests for dummy?, name_list and
+	  aliases.
+
+	* test/ruby/test_marshal.rb: add some tests.
+
+Tue Jun  3 22:25:51 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/etc/test_etc.rb: new tests for etc.
+
+Tue Jun  3 19:35:02 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_with_memo): New method: with_memo().
+
+Tue Jun  3 20:04:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (miniruby$(EXEEXT)): miniruby cannot be
+	  written by miniruby itself.
+
+Tue Jun  3 19:33:22 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_init_copy): Take care of
+	  initialize_copy as well as initialize.
+
+Tue Jun  3 16:06:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): fix for non-existent files and SFN of
+	  symlinks.  [ruby-talk:303736]
+
+Tue Jun  3 15:12:01 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#classify): Back out the `group_by' alias.
+	  Better think twice.
+
+Tue Jun  3 15:00:22 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#collect, Set#select): Back out.  I thought it
+	  was consistent but turned out to be wrong.
+
+Tue Jun  3 13:41:08 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#collect, Set#select): Override Enumerable
+	  methods and make them return a set. [ruby-core:17055]
+	  (Set#delete_if, Set#collect!, Set#reject!, Set#classify)
+	  (Set#divide, Set#delete_if): Return an enumerator if no block is
+	  given.
+	  (Set#classify): Define an alias `group_by' to override that of
+	  Enumerable.
+
+Tue Jun  3 13:35:40 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (run_exec_pgroup): C99 ism.
+
+Tue Jun  3 12:51:57 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_allocate, enumerator_ptr): Properly
+	  detect if the object is initialized and raise error when
+	  appropriate.
+	  (enumerator_initialize): Fix a typo in rdoc. [ruby-core:17052]
+
+Tue Jun  3 01:21:51 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_method.rb: add a test.
+
+Tue Jun  3 00:26:48 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* marshal.c (w_object): add a check for modification of array during
+	  its dump.
+
+Mon Jun  2 22:27:57 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enc/iso_8859_5.c: Large omicron should lowercase to small omicron.
+
+	* test/ruby/test_big5.rb, test/ruby/test_cp949.rb,
+	  test/ruby/test_euc_jp.rb, test/ruby/test_euc_kr.rb,
+	  test/ruby/test_euc_tw.rb, test/ruby/test_gb18030.rb,
+	  test/ruby/test_gbk.rb, test/ruby/test_iso_8859.rb,
+	  test/ruby/test_koi8.rb, test/ruby/test_shift_jis.rb,
+	  test/ruby/test_windows_1251.rb: new tests for encoding.
+
+	* test/ruby/test_utf16.rb, test/ruby/test_utf32.rb,
+	  test/ruby/test_regexp.rb: add tests.
+
+Mon Jun  2 21:56:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_file.rb: add tests for uninitialized object.
+
+	* test/ruby/test_class.rb: ditto.
+
+	* test/ruby/test_thread.rb: ditto.
+
+Mon Jun  2 21:44:15 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* re.c: fix SEGV by Regexp.allocate.names, Match.allocate.names, etc.
+
+	* test/ruby/test_regexp.rb: add tests for above.
+
+	* io.c: fix SEGV by IO.allocate.print, etc.
+
+	* test/ruby/test_io.rb: add tests for above.
+
+Mon Jun  2 19:17:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_argf.rb (teardown): remove renamed temporary files.
+
+Mon Jun  2 18:51:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/un.rb (wait_writable): wait until target files can be
+	  written actually.
+
+	* win32/Makefile.sub (LDSHARED_0, LINK_SO): get rid of failure of
+	  mt.exe.
+
+Mon Jun  2 16:26:17 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/delegate.rb (Delegator::MethodDelegation#respond_to):
+	  respond_to? should now take optional second argument; submitted
+	  by Jeremy Kemper <jeremy at bitsweat.net> in [ruby-core:17045].
+
+Mon Jun  2 16:14:18 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/erb.rb (ERB::Compiler::TrimScanner#scan_line): Oops.  This
+	  change did not apply to trunk.  Backed out.
+
+Mon Jun  2 16:08:24 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/erb.rb (ERB::Compiler::TrimScanner#scan_line): Fix a bug
+	  where tokens are not yielded one by one.
+
+	* test/erb/test_erb.rb (TestERBCore#_test_01)
+	  (TestERBCore#test_02_safe_04): The expected value should come
+	  first for assert_equal().
+
+Mon Jun  2 13:06:38 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* mkconfig.rb: hide build path from rbconfig.rb.
+
+Mon Jun  2 08:46:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strtod, dtoa): initialize more variables for error
+	  handling.
+
+Mon Jun  2 04:55:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* suppress warnings on cygwin, mingw and mswin.
+
+Mon Jun  2 04:35:32 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/gb18030.c (gb18030_code_to_mbc): add 0x80000000
+	  for 4bytes character.
+
+Mon Jun  2 03:52:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (set_arg0): reverted used variable definition.
+
+Mon Jun  2 03:23:25 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/gb18030.c (gb18030_mbc_to_code): mask by 0x7FFFFFFF
+	  because OnigCodePoint will be used as 32bit signed int.
+	  Masking by 0x7FFFFFFF is ok on GB18030;
+	  Minimum 4bytes character is 0x81308130.
+
+Sun Jun  1 22:29:35 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* rational.c (string_to_r_internal): use rb_isdigit.
+
+	* marshal.c (long_toobig): use %zd.
+
+	* ruby.c (set_arg0): move unused variable definition.
+
+Sun Jun  1 12:18:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (DEFINE_INSN): subtract of pointers is ptrdiff_t.
+	  this is not int on 64bit system.
+
+	* vm_dump.c (control_frame_dump): ditto.
+
+	* vm_dump.c (stack_dump_each): ditto.
+
+	* vm_dump.c (debug_print_register): ditto.
+
+	* vm_dump.c (debug_print_pre): ditto.
+
+	* transcode.c (str_transcode): ditto.
+
+Sun Jun  1 10:32:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_normal_exit): show coredump status.
+
+Sat May 31 23:33:34 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* README, README.ja: Add a note about default C flags.
+
+Sat May 31 23:02:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (count_objects): clear given hash.
+
+Sat May 31 20:28:10 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_regexp.rb: add tests.
+
+Sat May 31 19:11:39 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/utf_16{be,le}.c (utf16{be,le}_code_to_mbc):
+	  fix codepoint to bytes.
+
+Sat May 31 18:28:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* suppress warnings with -Wwrite-string.
+
+Sat May 31 18:26:33 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_delete_if): should return enumerator if no block
+	  is given.  [ruby-dev:34901]
+
+Sat May 31 15:58:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, configure.in (warnflags): defaulted to -Wall
+	  -Wno-parentheses with gcc.  [ruby-dev:34810]
+
+Sat May 31 15:17:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/mvm.h: new header file for MVM, and moved rb_vm_t and
+	  rb_thread_t from vm_core.h.
+
+Sat May 31 12:02:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_normal_exit): show pid when fail.
+
+Fri May 30 23:55:56 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_rubyoptions.rb: add a test of RUBY_DESCRIPTION.
+
+Fri May 30 22:47:17 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_regexp.rb: add tests.
+
+Fri May 30 22:40:53 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_signal.rb: add tests to achieve over 80% test
+	  coverage of signal.c.
+
+Fri May 30 22:28:03 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* signal.c (esignal_signo): fix SignalException#signo which returned
+	  nil absolutely.
+
+	* signal.c (esignal_init): always prepend "SIG" to a string that is
+	  returned by SignalException#signm.
+
+Fri May 30 22:17:39 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_argf.rb: rename a conflicting method name.
+
+	* test/ruby/test_string.rb: ditto.
+
+	* test/ruby/test_io.rb: ditto.
+
+Fri May 30 22:14:37 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (defined_expr): fix SEGV by defined?([1]).
+
+Fri May 30 12:18:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (prelude.c): simply depends on PREP.  [ruby-dev:34877]
+
+	* enc/make_encdb.rb, enc/trans/make_transdb.rb: ditto.
+
+Fri May 30 10:55:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (struct rb_unblock_callback), thread.c
+	  (set_unblock_function), thread_{pthread,win32}.c (native_sleep):
+	  extracted from struct rb_thread_struct.
+
+	* thread.c (reset_unblock_function): not check interrupts at leaving
+	  blocking region.  [ruby-dev:34874]
+
+Fri May 30 06:09:31 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/utf_8.c: add UTF8-MAC (UTF-8-MAC).
+
+Fri May 30 04:17:13 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enum.c (enum_count, count_all_i, Init_Enumerable),
+	  array.c (rb_ary_count): If no argument or block is given, count
+	  the number of all elements.
+
+Fri May 30 03:12:18 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_s_rand, ossl_bn_s_pseudo_rand):
+	  Int should be enough here.
+
+Fri May 30 02:35:00 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_s_rand, ossl_bn_s_pseudo_rand),
+	  ext/openssl/ossl_pkey_dh.c (ossl_dh_s_generate)
+	  (ossl_dh_initialize),
+	  ext/openssl/ossl_pkey_dsa.c (ossl_dsa_s_generate),
+	  ext/openssl/ossl_rand.c (ossl_rand_bytes)
+	  (ossl_rand_pseudo_bytes, ossl_rand_egd_bytes),
+	  ext/openssl/ossl_x509store.c (ossl_x509stctx_set_error): Do not
+	  use FIX2INT() without checking the value type.  Use NUM2INT()
+	  instead; found by akr in [ruby-dev:34890].
+
+Fri May 30 02:08:20 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* signal.c (esignal_init): handle a non-integer argument correctly,
+	  allowing SignalException.new(:INT).
+
+Fri May 30 00:59:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_regexp.rb: add tests.
+
+Thu May 29 22:51:05 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_require.rb: add a test for load with wrap flag, to
+	  achieve 100% test coverage of eval_jump.c.
+
+Thu May 29 22:47:53 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_argf.rb: new tests for ARGF, to achieve over 85% test
+	  coverage of file.c.
+
+	* test/ruby/test_io.rb: add tests.
+
+Thu May 29 22:41:48 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (argf_readchar): raise EOFError, synchronizing IO#readchar.
+
+Thu May 29 22:29:39 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* io.c (argf_external_encoding, argf_internal_encoding): fix SEGV by
+	  ARGF.external_encoding.
+
+Thu May 29 17:52:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/extconf.rb: search zlib1, and regard mswin32 later than VC6
+	  as WIN32.  [ruby-core:16984]
+
+Wed May 28 18:05:28 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_nitems, Init_Array): Axe Array#nitems().
+	  cf. [ruby-dev:34676]-[ruby-dev:34713]
+
+Wed May 28 17:50:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/mkexports.rb (Exports#objdump, Exports#each_line): extracted.
+
+Wed May 28 17:41:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (MKPREP): appended $(RBCONFIG).
+
+	* common.mk (enc.mk, prelude.c): not depend on $(RBCONFIG) on mswin32
+	  to get of compiling twice each time.
+
+	* win32/Makefile.sub (prelude.c): not depend on $(PREP).
+
+Wed May 28 17:37:07 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/mkexports.rb (Exports::Mswin#each_export): speed up.
+
+Wed May 28 16:41:59 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_slice_bang): Call rb_ary_modify_check() at the
+	  beginning. [rubyspec]
+
+Wed May 28 16:12:44 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb (WEBrick::HTTPServlet::CGIHandler#do_GET):
+	  Set the HTTP status code to 302 if a Location header field is
+	  present and the status code is not valid as a client
+	  redirection.  cf. RFC 3875 6.2.3, 6.2.4.
+
+Wed May 28 15:53:52 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/trans/japanese.c (to_SHIFT_JIS_EF_infos): typo.
+
+Wed May 28 15:18:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/singleton.rb (SingletonClassMethods): _load should be public.
+
+Wed May 28 13:30:43 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.c: add workaround for Unicode to CP932.
+	  U+2015->0x815C, U+2225->0x8161, U+FF0D->0x817C, U+FF3C->0x815F,
+	  U+FF5E->0x8160, U+FFE0->0x8191, U+FFE1->0x8192, U+FFE2->0x81CA
+
+Wed May 28 12:52:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object, marshal_dump, r_object0, marshal_load): search
+	  private methods too.  [ruby-dev:34671]
+
+	* object.c (convert_type): ditto.
+
+Wed May 28 08:42:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c: "%" is required before PRI?VALUE.
+
+Tue May 27 22:10:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.c (error_handle): SystemExit and SignalException throws
+	  TAG_RAISE but not TAG_FATAL.
+
+	* thread.c (rb_thread_execute_interrupts): delay interrupts during
+	  raising exceptions.  [ruby-dev:34855]
+
+Tue May 27 20:18:30 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_slice_bang): Return an empty array instead of
+	  nil when pos is valid and len is adjusted from a valid value to
+	  zero; caught by RubySpec.
+
+Tue May 27 19:12:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (MKPREP), common.mk, win32/Makefile.sub (prelude.c): get
+	  rid of depending PREP with nmake.
+
+	* common.mk (encs): depends on libruby.
+
+Tue May 27 19:00:22 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/stringio/stringio.c (strio_each_char, Init_stringio): Add
+	  StringIO#{each_char,chars}.
+	  (Init_stringio): Fix StringIO#bytes.
+
+Tue May 27 17:54:35 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/stringio/stringio.c (strio_each_byte): Return self instead
+	  of nil as the rdoc says.
+
+Tue May 27 15:36:27 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (check_int): use PRIxVALUE format specifier.
+
+	* numeric.c (check_uint, rb_num2fix, int_chr): ditto.
+
+	* numeric.c (num_fdiv): fallback to_f should always return float
+	  result.  should not use #quo that may return rational.
+
+	* numeric.c (num_div): should raise ZeroDivisionError.
+
+	* numeric.c (fix_divide): ditto.
+
+	* test/ruby/test_numeric.rb (TestNumeric::test_divmod): avoid
+	  ZeroDivisionError in tests.
+
+Tue May 27 13:14:53 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enum.c (enum_to_a): Pass arguments through to #each().
+	  (enum_sort): Follow the enum_to_a signature change.
+	  (enum_reverse_each): Add #reverse_each().
+
+Tue May 27 13:12:37 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* io.c (Init_IO): Define ARGF.{lines,bytes,chars}.
+
+Tue May 27 12:06:37 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (BUFCHECK): wrong condition. [ruby-core:16921]
+
+	* file.c (file_expand_buf): shouldn't use buflen for length of string.
+
+Mon May 26 18:24:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (BUFCHECK): no resize if enough room.
+
+	* file.c (file_expand_path): use BUFCHECK.
+
+Mon May 26 17:48:42 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (struct enumerator, enumerator_init)
+	  (enumerator_init_copy, enumerator_each): Eliminate iter.
+	  (enumerator_ptr): Do not hardcode the class name.
+	  (enumerator_with_index): Delay variable initialization after
+	  RETURN_ENUMERATOR().
+
+Mon May 26 17:23:49 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (file_expand_path): add more space for '/'.
+
+	* file.c (file_expand_path): should reset address of p after calling
+	  rb_str_resize(). [ruby-dev:34800]
+
+Mon May 26 16:49:55 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode): use run-hooks if run-mode-hook is
+	  not available.  a patch from Kazuhiro NISHIYAMA <zn at mbf.nifty.com>
+	  in [ruby-dev:34853].
+
+Mon May 26 16:41:35 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (ntfs_tail): filename which starts with '.' is valid.
+
+	* file.c (file_expand_path): cygwin symlink support.
+
+Mon May 26 07:15:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): rb_make_backtrace has no arguments.
+
+Mon May 26 01:17:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_normal_exit): signal description
+	  refined.
+
+Mon May 26 00:52:52 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* hash.c (env_each_key, env_each_value, env_reject_bang)
+	  (rb_env_clear, env_replace): Omit duplicated secure level check.
+
+Mon May 26 00:37:16 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* hash.c (env_each_value): Do not call env_values() twice.
+
+Sun May 25 17:54:36 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_compile): set local_table for
+	  ISEQ_TYPE_DEFINED_GUARD.
+
+Sun May 25 17:52:25 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_build_body): remove side effect from
+	  VM::InstructionSequence.load.
+
+Sun May 25 04:30:45 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_module.rb (remove_json_mixins): change judgment
+	  condition.
+
+Sun May 25 03:54:39 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_module.rb (test_ancestors, test_included_modules):
+	  ignore json mixins.
+
+Sun May 25 02:37:25 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.c: renamed from vm_method.c.  "vm_method.c" is included
+	  by "vm.c".
+
+	* vm_eval.c: added.  Some codes are moved from "eval.c"
+
+	* common.mk: fix for above changes.
+
+	* compile.c: make a vm_eval(0)
+
+	* eval.c, eval_error.c, eval_intern.h, eval_jump.c, proc.c, vm.c,
+	  id.c, id.h, vm_core.h, vm_dump.c, vm_evalbody.c, vm_insnhelper.c,
+	  blockinlining.c: fix for above changes.  and do some refactoring.
+	  this changes improve rb_yield() performance.
+
+Sat May 24 22:32:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* util.c (ruby_strtod): clear errno at the top of our own
+	  implementation of strtod(3).  [ruby-dev:34834] [ruby-dev:34839]
+
+Sat May 24 15:26:16 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (iseq_set_exception_table, NODE_WHILE, NODE_NEXT): remove
+	  special handling that decrements sp in CATCH_TYPE_NEXT for NODE_WHILE.
+
+	* vm.c (vm_eval_body), vm_insnhelper.c (vm_throw): remove unused code.
+
+Sat May 24 08:13:34 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (rb_str_transcode): argc is 1, and argv is &to.
+
+Fri May 23 17:55:11 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_slice_bang): Be consistent with Array#slice()
+	  and String#slice!().  Just return nil when a negative length or
+	  out of boundary index is given instead of raising an exception
+	  via internal functions.
+
+Fri May 23 16:44:34 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (Init_Enumerator): Override
+	  Enumerable::Enumerator#each_with_index with #with_index.
+
+Fri May 23 12:23:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_num_t): moved form vm.h.
+
+	* tool/instruction.rb (RubyVM::Instruction#sp_increase_c_expr),
+	  tool/instruction.rb (RubyVM::VmBodyGenerator#make_header_operands):
+	  omit unused variables.
+
+Fri May 23 08:47:02 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (exc_equal): == operator should be transitional.
+	  [ruby-dev:34808]
+
+	* error.c (syserr_eqq): === should be able to handle delegated
+	  objects as well.
+
+Fri May 23 06:15:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_compile_with_option): get rid of segv.
+
+Fri May 23 02:29:14 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (opt_gt|ge|lt|le): use values directly to compare.
+
+Fri May 23 01:15:09 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, eval_intern.h, include/ruby/intern.h, include/ruby/ruby.h,
+	  vm.c, vm_core.h, vm_insnhelper.c: remove pointless "const".
+
+Thu May 22 23:45:17 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* compile.c (get_destination_insn, get_next_insn, get_prev_insn):
+	  peephole optimization should not ignore ISEQ_ELEMENT_ADJUST.
+
+Thu May 22 20:20:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (check_dump_arg, check_load_arg): check if reentered.
+	  [ruby-dev:34802]
+
+Thu May 22 20:14:28 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* iseq.c (iseq_load, iseq_data_to_ary): support
+	  ISEQ_TYPE_DEFINED_GUARD.
+
+Thu May 22 19:01:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_get_ruby_level_cfp): moved from eval_intern.h.
+
+	* vm.c (sdr, nsdr): define methods only if VMDEBUG is defined.
+
+Thu May 22 17:18:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_compact_bang): fix reallocation size.
+
+Thu May 22 15:20:20 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval_intern.h, vm_core.h, include/ruby/intern.h, include/ruby/ruby.h,
+	  vm.c: need to add const to prototypes, of course.
+
+Thu May 22 13:24:43 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, vm.c, vm_core.h, vm_insnhelper.c: specify "const".
+
+	* vm_opts.h: add a OPT_TOKEN_THREADED_CODE macro.
+
+Thu May 22 12:51:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* insns.def (newhash): fix a variable definition: "const k".
+
+Thu May 22 12:40:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (flatten): check if reentered.  [ruby-dev:34798]
+
+Thu May 22 11:39:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_normal_exit): capture stdout and stderr
+	  of the child process.
+
+Thu May 22 08:28:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (flatten): free memo hash table before raising exception.
+	  [ruby-dev:34789]
+
+Thu May 22 06:30:10 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* array.c (flatten): fix memory leak.
+
+Thu May 22 06:21:34 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_str_caseeql): added.
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_enc_find_index): use nkf_str_caseeql.
+
+Thu May 22 05:45:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (proc_dup): should copy safe_level from src proc
+	  properly.  a patch from Keita Yamaguchi
+	  <keita.yamaguchi at gmail.com>
+
+Thu May 22 02:46:08 2008  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: do not use Thread#raise. [ruby-dev:34739]
+
+Thu May 22 00:30:06 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_require.rb: new tests for library requiring, to
+	  achieve over 90% test coverage of dln.c.
+
+	* test/ruby/test_class.rb: add tests to achieve over 90% test coverage
+	  of class.c.
+
+	* test/ruby/test_module.rb: ditto.
+
+Thu May 22 00:15:44 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm_insnhelper.c: specify "const".
+
+Wed May 21 23:20:21 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_eval.rb: fix syntax.
+
+Wed May 21 17:46:17 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_enc_find_index):
+	  use strcasecmp. [ruby-dev:34787]
+
+Wed May 21 16:48:22 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_compact_bang): avoid forceful realloc.
+
+Wed May 21 07:42:28 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_usascii_str_new): use rb_str_new.
+
+	* string.c (rb_enc_str_new): ditto.
+
+	* string.c (rb_usascii_str_new2): use rb_str_new2.
+
+Wed May 21 07:22:01 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c, include/ruby/encoding.h
+	  (rb_enc_associate, rb_enc_associate_index):
+	  returns obj. [ruby-dev:34778]
+
+Wed May 21 04:20:20 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_ascii8bit_encoding): use ENCINDEX_ASCII.
+
+	* encoding.c, include/ruby/encoding.h (rb_ascii8bit_encindex):
+	  added.
+
+	* encoding.c (rb_locale_encoding): use rb_usascii_encoding().
+
+Wed May 21 01:45:58 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file_exhaustive.rb (setup): workaround for Windows
+	  Vista.
+
+	* test/ruby/envutil.rb (rubyexec): now Open3.open3 is supported on
+	  Windows.
+
+	* test/ruby/test_process.rb: use ``||'' instead of ``;'' because
+	  cmd.exe not support it.
+
+Wed May 21 01:28:47 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c, include/ruby/encoding.h (rb_str_transcode):
+	  C API of encoding conversion for Ruby object.
+	  VALUE rb_str_transcode(VALUE str, VALUE to).
+
+	* transcode.c (str_encode, str_encode_bang):
+	  rename from rb_tr_transcode or rb_str_transcode_bang.
+
+Tue May 20 23:26:05 2008  Yusuke Endoh	<mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb: fix tests for 64bit CPU.
+
+Tue May 20 20:59:56 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (rb_nkf_convert) (nkf_enc_without_bom):
+	  reverted. nkf-utf8/nkf.c should be independent of ruby.
+
+	* ext/nkf/nkf.c (options):
+	  moved from nkf-utf8/nkf.c.
+	  override nkf's original settings for Unicode BOM.
+
+Tue May 20 13:20:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_convert), ext/nkf/nkf-utf8/nkf.c
+	  (nkf_enc_without_bom): BOM is not a part of encodings.
+
+	* ext/nkf/nkf.c (Init_nkf), ext/nkf/nkf-utf8/nkf.c (options):
+	  UTF-{16,32} without endian have no sense.
+
+Tue May 20 12:13:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options, process_options): --dump option.
+
+Tue May 20 11:36:06 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRI[diouxX]VALUE): printf format for VALUE.
+
+	* gc.c (assign_heap_slot): suppress a warning.
+
+Tue May 20 03:42:43 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, vm_insnhelper.c: fix cref in instance_eval
+	  and cvar_base search protocol.
+
+	* bootstraptest/test_knownbug.rb, test_eval.rb: move solved test
+	  and add new tests.
+
+	* test/ruby/test_eval.rb: fix tests for spec.
+
+Tue May 20 01:43:44 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: fix a test.
+	  "block_given?" returns true if "yield" can be used.
+
+Tue May 20 01:07:19 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (assignable_gen): when "self = 1" was evaluated, unnecessary
+	  error message was output, which might cause null pointer access.
+
+Tue May 20 08:38:56 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_enc_strlen_cr): need to set ENC_CODERANGE_7BIT if
+	  search_nonascii() fails.    [ruby-dev:34751]
+
+	* string.c (rb_str_reverse): preserve coderange info if the
+	  receiver is 7bit string.
+
+	* string.c (rb_str_reverse_bang): ditto.
+
+	* string.c (rb_str_reverse_bang): should have called
+	  single_byte_optimizable before rb_str_modify() that clears
+	  coderange info.
+
+	* string.c (tr_trans): handle single bytes more eagerly.
+
+Mon May 19 23:32:12 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (invoke_block_from_c): fix call flow.
+
+Mon May 19 23:19:35 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* regexec.c (slow_search): check the case when the length is 1.
+	  The behavior of memcmp is undefined if the third argument is 0.
+
+Mon May 19 21:07:48 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (native_thread_apply_priority):
+	  fix argument range check.  [ruby-dev:33124]
+
+Mon May 19 18:22:35 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/openssl/ossl_pkcs5.c (ossl_pkcs5_pbkdf2_hmac): Fix the type
+	  of md; pointed out by Takahiro Kambe <taca at back-street.net>
+	  in [ruby-dev:34748].
+
+Mon May 19 17:23:55 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regparse.c (PINC): use optimized enclen() instead of
+	  ONIGENC_MBC_ENC_LEN().
+
+	* regparse.c (PFETCH): ditto.
+
+	* regparse.c (PFETCH): small optimization.
+
+	* regexec.c (slow_search): single byte encoding optimization.
+
+	* regenc.h (enclen): avoid calling function when encoding's
+	  min_len == max_len.
+
+	* re.c (rb_reg_regsub): rb_enc_ascget() optimization for single
+	  byte encoding.
+
+	* re.c (rb_reg_search): avoid allocating new re_registers if we
+	  already have MatchData.
+
+	* re.c (match_init_copy): avoid unnecessary onig_region_free()
+	  before onig_region_copy.
+
+	* encoding.c (rb_enc_get_index): remove implicit enc_capable check
+	  each time.
+
+	* encoding.c (rb_enc_set_index): ditto.
+
+	* encoding.c (enc_compatible_p): small refactoring.
+
+	* include/ruby/encoding.h (rb_enc_dummy_p): inline
+	  rb_enc_dummy_p() and export related code.
+
+Mon May 19 14:32:03 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h: fix strange change by version.h update tool.
+
+Mon May 19 14:18:13 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: move solved tests.
+
+	* bootstraptest/test_eval.rb, test_literal.rb, test_syntax.rb,
+	  test_thread.rb: ditto.
+
+	* test/ruby/test_m17n.rb, test_proc.rb, test_sprintf.rb,
+	  test_string.rb, test/ruby/test_struct.rb: ditto.
+
+Mon May 19 13:23:03 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_spawn_internal): set last_status when status == -1
+	  because there is no path to set it on win32. this patch is derived
+	  from [ruby-core:16787], submitted by Luis Lavena <luislavena at
+	  gmail.com>
+
+Mon May 19 11:32:47 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, insns.def, eval.c, vm_insnhelper.c: fix CREF handling.
+	  VM value stack frame of block contains cref information.
+	 (dfp[-1] points CREF)
+
+	* compile.c, eval_intern.h, eval_method.c, load.c, proc.c,
+	  vm_dump.h, vm_core.h: ditto.
+
+	* include/ruby/ruby.h, gc.c: remove T_VALUES because of above
+	  changes.
+
+	* bootstraptest/test_eval.rb, test_knownbug.rb: move solved test.
+
+Sun May 18 22:26:51 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservlet/filehandler.rb: should normalize path
+	  name in path_info to prevent script disclosure vulnerability on
+	  DOSISH filesystems. (fix: CVE-2008-1891)
+	  Note: NTFS/FAT filesystem should not be published by the platforms
+	  other than Windows. Pathname interpretation (including short
+	  filename) is less than perfect.
+
+	* lib/webrick/httpservlet/abstract.rb
+	  (WEBrick::HTTPServlet::AbstractServlet#redirect_to_directory_uri):
+	  should escape the value of Location: header.
+
+	* lib/webrick/httpservlet/cgi_runner.rb: accept interpreter
+	  command line arguments.
+
+Sun May 18 02:54:46 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* pack.c (pack_pack): check errno to detect error of ruby_strtoul.
+
+	* pack.c (pack_unpack): ditto.
+
+	* test/ruby/test_pack.rb: add a test for above.
+
+Sat May 17 23:53:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): fix for short file name on Cygwin.
+
+Sat May 17 18:03:52 2008  Yuki Sonoda (Yugui) <yugui@yugui.jp>
+
+	* vm.c (Init_VM): removed the definition of Thread#initialize,
+	  which is overwritten in Init_Thread and is never used.
+
+Sat May 17 14:01:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): should not free shared pointer, and set
+	  shared.  [ruby-dev:34732]
+
+Sat May 17 12:34:54 2008  Yuki Sonoda (Yugui) <yugui@yugui.jp>
+
+	* thread_pthread.c (Init_native_thread): Kernel#.sleep used never to
+	  sleep on Mac OS X. Reported by arton <artonx AT yahoo.co.jp>.
+
+	* thread_pthread.c (native_sleep): added error checks.
+
+Sat May 17 11:29:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_extname): first dot is not an extension name.
+
+Sat May 17 03:21:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): stop memory leak.  [ruby-dev:34726]
+
+	* re.c (rb_reg_search): need to free allocated buffer in re_register.
+
+	* regexec.c (onig_region_new): more pedantic malloc check.
+
+	* regexec.c (onig_region_resize): ditto.
+
+	* regexec.c (STATE_CHECK_BUFF_INIT): ditto.
+
+	* regexec.c (onig_region_copy): use onig_region_resize.
+
+Fri May 16 12:48:33 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* math.c (to_flo): rb_Float() accepts even strings for input.
+
+	* complex.c (nucomp_to_f): fix wrong message.
+
+	* complex.c (nucomp_to_r): ditto.
+
+	* object.c (rb_Float): do not check NaN for error.  NaN is a part
+	  of valid float values.
+
+Thu May 15 23:36:09 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_string.rb: add tests to achieve over 90% test
+	  coverage of string.c.
+
+	* test/ruby/test_m17n.rb: ditto.
+
+	* test/ruby/test_symbol.rb: ditto.
+
+	* test/ruby/test_pack.rb: ditto.
+
+Thu May 15 23:01:06 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (tr_find): String#delete returned wrong result when multiple
+	  utf-8 arguments are passed.
+
+	* test/ruby/test_m17n.rb (test_delete): add a test for above.
+
+Thu May 15 22:37:56 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (ripper_warningS): now used.
+
+Thu May 15 15:33:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): support for alternative data stream
+	  and ignored trailing garbage of NTFS.
+
+	* file.c (rb_file_s_basename): ditto.
+
+	* file.c (rb_file_s_extname): ditto.
+
+Thu May 15 13:43:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): no need for forceful warning when
+	  converting to float.  overflow is a nature of float values.
+
+	* parse.y (parser_yylex): ditto.
+
+Thu May 15 13:23:20 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_prepare_enc): error condition was updated for non
+	  ASCII compatible strings.
+
+Thu May 15 12:19:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/openssl/openssl_missing.c (HMAC_CTX_copy): adopted
+	  prototype change in openssl bundled with newer OpenBSD.
+	  a patch from Takahiro Kambe <taca at back-street.net> in
+	  [ruby-dev:34691].
+
+Wed May 14 22:09:25 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ChangeLog: fix typo.
+
+Wed May 14 21:49:14 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_object.rb: new tests to achieve over 90% test
+	  coverage of object.c, eval.c and eval_method.c.
+
+	* test/ruby/test_module.rb: ditto.
+
+	* test/ruby/test_trace.rb: ditto.
+
+	* test/ruby/test_integer.rb: ditto.
+
+	* test/ruby/test_float.rb: ditto.
+
+	* test/ruby/test_method.rb: ditto.
+
+	* test/ruby/test_variable.rb: ditto.
+
+	* test/ruby/test_eval.rb: ditto.
+
+	* test/ruby/test_exception.rb: ditto.
+
+	* test/ruby/test_class.rb: ditto.
+
+Wed May 14 12:46:37 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (insn_operand_intern): remove Qundef related code.
+
+Wed May 14 12:42:36 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_count): Override Enumerable#count for better
+	  performance.
+
+Wed May 14 11:29:06 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: add a "putcbase" instruction.
+
+	* compile.c, insns.def: fix to use putcbase instruction for
+	  class search.  Qundef should not be used.
+
+Wed May 14 07:49:35 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): defer calling of rb_frame_self() until it
+	  become really necessary.
+
+	* eval.c (rb_call): ditto.
+
+Wed May 14 00:55:56 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_io_m17n.rb: remove a duplicative method.
+
+	* test/ruby/test_utf16.rb: rename a conflicting method name.
+
+	* test/ruby/test_array.rb: ditto.
+
+	* test/ruby/test_file_exhaustive.rb: ditto.
+
+	* test/ruby/test_hash.rb: ditto.
+
+	* test/ruby/test_env.rb: ditto.
+
+	* test/ruby/test_fixnum.rb: ditto.
+
+	* test/ruby/test_rational.rb: ditto.
+
+Wed May 14 00:45:58 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval_method.c (rb_add_method): fix check for warning when
+	  Object#initialize is redefined. (same as 1.8)
+
+Tue May 13 23:32:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_yield): use rb_yield_values2.
+
+	* enum.c (DEFINE_ENUMFUNCS): macro to define enumerator and yielding
+	  functions.
+
+	* enum.c (enum_all_func, enum_any_func, enum_one_func,
+	  enum_none_func): reduced duplicate code.
+
+Tue May 13 15:09:38 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c: Update rdoc.
+	  (enumerator_initialize): Discourage the use.
+	  (enum_each_slice, enum_each_cons, enumerator_each)
+	  (enumerator_with_index): Add a note about a call without a block.
+
+Tue May 13 08:25:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_f_gets): re-enable rdoc.
+	  (rb_f_readline): ditto.
+	  (rb_f_readlines): ditto.
+
+Tue May 13 07:56:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_cat): fixed buffer overrun reported by
+	  Christopher Thompson <cthompson at nexopia.com> in [ruby-core:16746]
+
+Mon May 12 23:37:57 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (collect_local_variables_in_env): remove unnecessary check
+	  which causes: x=1;proc{local_variables}.call #=> []
+
+	* test/ruby/test_variable.rb: add a test for above.
+
+Mon May 12 23:05:24 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c, include/ruby/intern.h (rb_run_exec_options): externed.
+
+	* process.c (save_redirect_fd, save_env_i, save_env, run_exec_dup2,
+	  run_exec_open, run_exec_pgroup, run_exec_rlimit, rb_run_exec_options):
+	  save parent's process environments.
+
+	* process.c (rb_spawn_internal): remove calling run_exec_options()
+	  because cannot restore after spawn.
+
+	* io.c (pipe_open): ditto.
+
+	* test/ruby/test_process.rb (test_execopts_env): upcase environment
+	  variable name for case insensitive platforms.
+
+	* win32/win32.c (init_env): set USER environment variable only when
+	  USERNAME is available.
+
+Mon May 12 22:23:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/date.rb (once): use Object#object_id instead of Symbol#to_i.
+
+Mon May 12 21:34:46 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/envutil.rb (rubybin): return expanded rubyexe instead of
+	  expanded ruby if available.
+
+Mon May 12 20:19:55 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enum.c (grep_i): Be aware of multiple values;
+	  fix [ruby-dev:34653].
+	  (grep_iter_i): Ditto.
+	  (count_i): Ditto.
+	  (find_i): Ditto.
+	  (find_index_i): Ditto.
+	  (find_all_i): Ditto.
+	  (reject_i): Ditto.
+	  (inject_i): Ditto.
+	  (inject_op_i): Ditto.
+	  (partition_i): Ditto.
+	  (group_by_i): Ditto.
+	  (first_i): Ditto.
+	  (sort_by_i): Ditto.
+	  (all_i): Ditto.
+	  (all_iter_i): Ditto.
+	  (any_i): Ditto.
+	  (any_iter_i): Ditto.
+	  (one_i): Ditto.
+	  (one_iter_i): Ditto.
+	  (none_i): Ditto.
+	  (none_iter_i): Ditto.
+	  (min_i): Ditto.
+	  (min_ii): Ditto.
+	  (max_i): Ditto.
+	  (max_ii): Ditto.
+	  (minmax_i): Ditto.
+	  (minmax_ii): Ditto.
+	  (min_by_i): Ditto.
+	  (max_by_i): Ditto.
+	  (minmax_by_i): Ditto.
+	  (member_i): Ditto.
+	  (take_i): Ditto.
+	  (take_while_i): Ditto.
+	  (drop_i): Ditto.
+	  (drop_while_i): Ditto.
+	  (cycle_i): Ditto.
+
+	* enum.c (each_with_index): Update rdoc.  each_with_index() takes
+	  arguments that are passed through to each(), and a hash preserves
+	  key order.
+
+Mon May 12 19:05:24 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_spawn_internal): remove calling run_exec_options()
+	  because cannot restore after spawn. we'll fix this later.
+
+Mon May 12 18:16:44 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_spawn_internal): need to call run_exec_options() before
+	  spawn if the platform doesn't have fork. [ruby-dev:34647]
+
+Mon May 12 15:20:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (ruby_vm_xmalloc): increase malloc_increase only if malloc
+	  succeeds.  failed malloc size can be huge.  it may increase
+	  malloc_limit too big which cause less GC and memory full.
+	  (ruby_vm_xrealloc): ditto.
+	  (rb_objspace): make params.limit and params.increase size_t.
+
+Mon May 12 15:04:58 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_prepare_re): made non static with small refactoring.
+
+	* ext/strscan/strscan.c (strscan_do_scan): should adjust encoding
+	  before regex searching.
+
+Mon May 12 13:57:19 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (is_defined): add NODE_OP_ASGN_{OR,AND}.  "defined?(a||=1)"
+	  should not operate assignment.  [ruby-dev:34645]
+
+Mon May 12 13:29:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigzero_p): check from MSB to LSB.  [ruby-dev:34649]
+
+Mon May 12 12:32:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RUBYOPT): affected BASERUBY too.  [ruby-talk:301514]
+
+Mon May 12 12:27:55 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (assign_heap_slot): fix condition for number of objects in
+	  a heap.
+
+Mon May 12 12:24:48 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_to_i): really removed.  [ruby-dev:34641]
+
+Mon May 12 11:15:55 2008  Yuki Sonoda (Yugui) <yugui@yugui.jp>
+
+	* gc.c (assign_heap_slot):  put the binary search routine in order.
+
+Mon May 12 10:52:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_gems), gem_prelude.rb: check if Gem is defined
+	  instead of Gem::Enable.
+
+	* gem_prelude.rb (load_full_rubygems_library, const_missing): prevent
+	  infinite recursion.  [ruby-dev:34539]
+
+Sun May 11 23:19:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (all_iter_i, any_iter_i): reduced duplicated code.
+
+Sun May 11 22:54:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/runner.rb (main): leave -I options for purelib.rb
+	  untouched.
+
+	* bootstraptest/runner.rb (main): handle relative path -r options.
+
+Sun May 11 19:04:06 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_thread.rb: kill and join temporal threads that are
+	  created in each test.
+
+Sun May 11 17:58:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_process.rb (TestProcess#with_stdin): defined.
+	  (TestProcess#test_argv0_noarg): don't use redirect_fds.
+	  [ruby-dev:34647]
+
+Sun May 11 17:57:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (MINIRUBY): should not include extension library path.
+
+Sun May 11 14:40:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (SIZET2NUM): new macro.
+	  (NUM2SIZET): new macro.
+
+	* gc.c (struct rb_objspace): use size_t for increment, length and
+	  used for 64bit.
+	  (allocate_heaps): ditto.
+	  (assign_heap_slot): ditto.
+	  (set_heaps_increment): ditto.
+	  (gc_mark_all): ditto.
+	  (is_pointer_to_heap): ditto.
+	  (free_unused_heaps): ditto.
+	  (gc_sweep): ditto.
+	  (os_obj_of): ditto.
+	  (rb_gc_call_finalizer_at_exit): ditto.
+	  (count_objects): ditto.
+
+Sun May 11 13:14:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (thread_cleanup_func_before_exec): extracted from
+	  thread_cleanup_func not to touch pthread data.
+	  pthread_cond_destroy in forked process may cause deadlock on
+	  Debian GNU/Linux Etch on x86, x86-64 and IA64.
+	  this doesn't cause resource leak because the process will exec soon.
+	  (terminate_atfork_before_exec_i): defined.
+	  (rb_thread_atfork_before_exec): defined.
+
+	* include/ruby/intern.h (rb_thread_atfork_before_exec): declared.
+
+	* process.c (rb_exec_atfork): call rb_thread_atfork_before_exec
+	  instead of rb_thread_atfork.
+
+	* io.c (popen_exec): call rb_thread_atfork_before_exec instead of
+	  rb_thread_atfork.
+
+Sat May 10 22:14:03 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (tr_trans): single '^' does not mean negation.
+	  [ruby-dev:34632]
+
+	* string.c (tr_trans): should check src size, not str size.
+	  [ruby-dev:34637]
+
+	* string.c (tr_trans): should not turn on modify flag if no
+	  modification happens.  [ruby-dev:34631]
+
+Sat May 10 18:11:18 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_each_line): zero length record separator should
+	  split a string into paragraphs.  [ruby-dev:34586]
+
+	* string.c (rb_str_each_line): RDoc updated.
+
+Sat May 10 11:36:20 2008  Tanaka Akira  <akr@fsij.org>
+
+	* vm.c (env_mark): mark env->block.self.  prevent SEGV when GC occur
+	  in prepare_iseq_build with gcc version 3.4.6 [FreeBSD] 20060305 on
+	  FreeBSD/amd64.
+
+Fri May  9 19:16:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (timeofday): use monotonic clock.  based on a patch
+	  from zimbatm <zimbatm at oree.ch> in [ruby-core:16627].
+
+Fri May  9 07:47:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* cont.c (cont_restore_0): dynamic stack direction code should be
+	  consistent with static one.  [ruby-talk:301152]
+
+Fri May  9 00:03:50 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (arg): operator assignment "a += b rescue c" should be
+	  parsed as "a += (b rescue c)" just like normal assignment.
+	  [ruby-talk:301000]
+
+Thu May  8 18:14:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_and): bit-wise operation should not take float
+	  values.  [ruby-dev:34612]
+
+	* bignum.c (rb_big_or): ditto.
+
+	* bignum.c (rb_big_xor): ditto.
+
+Thu May  8 17:44:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, ext/extmk.rb, lib/mkmf.rb: use absolute path for RUBYOPT.
+
+	* file.c (rb_find_file_ext): guard load_path from GC.
+	  gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) optimizes
+	  load_path by holding only RARRAY_LEN(load_path) and
+	  RARRAY_PTR(load_path) in registers on IA64 GNU/Linux Etch.
+
+Thu May  8 16:41:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (MINIRUBY), common.mk (RUBYOPT): add purelib.rb.
+	  [ruby-core:16642]
+
+Thu May  8 16:00:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): ! and ? at the bottom are no longer part
+	  of valid symbol names.  [ruby-dev:34590]
+
+Thu May  8 15:36:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (rb_gc_save_machine_context): call FLUSH_REGISTER_WINDOWS
+	  to mark the register stack from GC on another thread.
+
+Thu May  8 15:14:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): freeze temporary array.
+
+Thu May  8 13:19:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_thread_mark): mark stat_insn_usage only when ptr is not
+	  null.
+
+Thu May  8 10:44:04 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (sort_reentered): reentered check may be called from
+	  Array#sort.
+
+Thu May  8 09:51:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (sort_1, sort_2): check for reentered and if elements are
+	  accessible.  [ruby-core:16679]
+
+Thu May  8 06:43:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_find_exe_r, dln_find_file_r): reentrant versions.
+
+	* file.c (rb_find_file_ext, rb_find_file), process.c (proc_exec_v),
+	  (rb_proc_exec, proc_spawn_v, proc_spawn), ruby.c (process_options):
+	  use reentrant versions.
+
+Thu May  8 06:27:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_key_p): thread local storage stores ID.
+
+Thu May  8 01:10:03 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (tr_trans): should squeeze properly.  [ruby-dev:34587]
+
+	* string.c (tr_trans): had a bug in treating multi-byte character
+	  replacement.
+
+	* string.c (rb_str_delete_bang): need not to do anything for empty
+	  strings.
+
+	* test/ruby/test_m17n_comb.rb (TestM17NComb::test_str_delete): add
+	  test for empty receiver.
+
+Wed May  7 20:19:18 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (process_options, ruby_set_argv): set encoding of rb_argv
+	  after Init_prelude() because cannot load encoding extensions before
+	  it.
+
+Wed May  7 20:00:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (bit_coerce): float should not be a valid operand of
+	  bitwise operations.  [ruby-dev:34583]
+
+Wed May  7 19:35:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (rb_thread_key_p): should always convert symbol to ID.
+	  [ruby-dev:34588]
+
+Wed May  7 19:30:34 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_divide): float division should floor() before
+	  rounding into integer.  [ruby-dev:34584]
+
+Wed May  7 18:02:01 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_to_i): remove obsolete method.  preparation for
+	  symbol GC.
+
+	* numeric.c (fix_to_sym): ditto.
+
+	* numeric.c (fix_id2name): ditto.
+
+Wed May  7 17:43:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_puts_ary): check recursion first.  [ruby-dev:34580]
+
+Wed May  7 17:41:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_eval_body): initialize retval.  [ruby-dev:34576]
+
+Wed May  7 13:02:56 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_fdiv): flo.fdiv(NaN) should result NaN.
+
+	* numeric.c (num_quo): renamed and moved from bignum.c.
+	  [ruby-dev:34582]
+
+	* bignum.c (rb_big_fdiv): update RDoc description
+
+	* rational.c (nurat_s_new_m): small refactoring.
+
+	* bignum.c (rb_big2dbl): no need for forceful warning when
+	  converting to float.  overflow is a nature of float values.
+
+Wed May  7 00:54:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzreader_gets): may cause infinite loop.
+	  a patch from Kouya <kouyataifu4 at gmail.com> in
+	  [ruby-reference-manual:762].
+
+Tue May  6 02:08:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/io/nonblock/test_flush.rb: don't set Thread.abort_on_exception.
+
+	* test/net/imap/test_imap.rb: ensure disconnecting imap to terminate
+	  receiver thread.
+
+Tue May  6 00:29:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* iseq.c (insn_operand_intern): should handle Qundef embedded in
+	  operand.  [ruby-core:16656]
+
+Tue May  6 00:00:02 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): should call compile_cpath() for
+	  modules as well.  [ruby-dev:34585]
+
+	* insns.def (defineclass): add undef handling.
+
+Mon May  5 23:49:40 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (defineclass): was using wrong variable.  [ruby-dev:34592]
+
+Mon May  5 20:07:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_fflush): IO#flush problem within threads.  a patch from
+	  <s.wanabe at gmail.com> in [ruby-dev:34595].
+
+Mon May  5 19:58:44 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (defined_expr): protect some expression from
+	  segmentation fault.  a patch from wanabe <s.wanabe at gmail.com>
+	  in [ruby-dev:34593].
+
+Mon May  5 19:49:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_s_def): Struct.new(0) should not SEGV.
+	  based on the patch from wanabe <s.wanabe at gmail.com> in
+	  [ruby-dev:34594].
+
+	* struct.c (make_struct): call to_str on name object.
+
+Mon May  5 17:17:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (ruby_cleanup): wrap ruby_finalize_0 by SAVE_ROOT_JMPBUF to
+	  avoid SEGV by at_exit { Fiber.new{}.resume } on IA64.
+
+Mon May  5 12:12:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_slice_bang): should adjust length before making
+	  sub-array.
+
+Mon May  5 11:36:14 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_dup): should dupe corresponding information.
+	  [ruby-dev:34581]
+
+Mon May  5 11:13:50 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (compile_cpath): use Qundef to denote cbase lookup.
+
+	* insns.def (defineclass): Qundef is passed for cbase.
+
+	* insns.def (setconstant): ditto.
+
+	* vm_insnhelper.c (vm_check_if_namespace): use rb_inspect()
+	  instead of rb_obj_as_string() for better description.
+
+Mon May  5 02:10:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (set_heaps_increment): fix memory allocation strategy by
+	  determining heaps_inc from heaps_used, not objects_delta.
+	  (struct rb_objspace): delta removed.  change increment, length and
+	  used to long for LP64.
+	  (objects_delta): removed.
+	  (allocate_heaps): add next_heaps_length argument.
+	  (init_heap): renamed from add_heap.
+	  (garbage_collect): use heaps_increment in dont_gc.
+
+Sun May  4 21:09:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/getoptlong.rb: use $stderr instead of $deferr.
+
+Sun May  4 16:04:28 2008  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (obj2nsec): fix string argument.
+
+Sun May  4 14:29:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (rb_obj_respond_to): check the result of respond_to? method
+	  by RTEST.
+
+Sun May  4 12:57:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_each_line): return original string.
+
+Sat May  3 20:57:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_normal_exit):
+	  new method.
+
+Sat May  3 18:10:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_timespec): raise TypeError for nil and other objects
+	  which has no divmod method.
+
+Fri May  2 23:59:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (internal_read_func, internal_write_func): split from
+	  internal_io_func.
+
+Fri May  2 23:55:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* variable.c (rb_define_hooked_variable): guard *var from GC to
+	  prevent collecting argf under RUBY_DEBUG=gc_stress.
+
+Fri May  2 17:29:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_step): call to_int if step is not a numeric
+	  value.  [ruby-dev:34575]
+
+Fri May  2 16:10:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_step): do not forcefully convert steps into
+	  integers.  [ruby-dev:34571]
+
+Fri May  2 14:52:33 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el: move fontifying code from hook.  a patch from
+	  Phil Hagelberg <phil at hagelb.org> in [ruby-core:16636].
+
+Fri May  2 14:10:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_step): step may be bignum.
+
+Fri May  2 13:52:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (Init_Regexp): remove MatchData#select.  [ruby-dev:34563]
+
+Thu May  1 23:59:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_divide), numeric.c (fix_divide): check for result
+	  domain.  [ruby-dev:34559]
+
+Thu May  1 23:57:06 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/net/telnet.rb:  This patch from Brian Candler adds a FailEOF mode which
+	  can be activated to have net/telnet raise EOFError exceptions when the
+	  remote connection is closed.  The default behavior remains unchanged though.
+
+Thu May  1 23:43:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_step): check if step can be converted to an integer.
+	  [ruby-dev:34558]
+
+	* range.c (range_step): allow float step bigger than zero but less
+	  than one.  [ruby-dev:34557]
+
+Thu May  1 23:20:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_divide): return an integer for idiv.
+	  [ruby-dev:34553]
+
+Thu May  1 20:47:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_s_create): should access converted hash value.
+	  [ruby-dev:34555]
+
+Thu May  1 20:31:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_parse.rb (TestParse::test_void_expr_stmts_value):
+	  shut up warning.
+
+	* rational.c (nurat_to_f): no need for forceful warning when
+	  converting to float.  overflow is a nature of float values.
+
+Thu May  1 16:10:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_delete_if): return enumerator if no block given.
+	  [ruby-dev:34554]
+
+Wed Apr 30 21:36:40 2008  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/erb.rb (url_encode): [ruby-dev:34497] ERB::Util#url_encode
+	  bug fix. Reported by rubikitch.
+
+	* test/erb/test_erb.rb: ditto
+
+Wed Apr 30 20:11:36 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/net/telnet.rb: Fixing a bug where line endings would not be properly
+	  escaped when the two character ending was broken up into separate TCP
+	  packets.  Issue reported and patched by Brian Candler.
+
+Wed Apr 30 18:03:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_load_path), vm_core.h (rb_vm_t): moved to VM.
+
+	* load.c (rb_get_load_path): returns absolute load path.
+
+	* load.c (load_path_getter): $LOAD_PATH getter.
+
+	* file.c (rb_find_file_ext, rb_find_file), ruby.c (push_include,
+	  ruby_init_loadpath): use the accessor.
+
+	* vm.c (rb_vm_mark): mark load_path.
+
+Wed Apr 30 17:47:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_search): use local variable.  a patch from wanabe
+	  <s.wanabe AT gmail.com> in [ruby-dev:34537].  [ruby-dev:34492]
+
+Wed Apr 30 16:10:18 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval_intern.h: specify the values of the enumeration constants
+	  explicitly.  [ruby-dev:34489]
+
+Wed Apr 30 12:32:39 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (check_exec_redirect_fd): prohibit duplex IO.
+	  (check_exec_fds): record maxhint even if close_others is not
+	  specified.
+	  (rb_exec_arg_fixup): renamed from rb_exec_arg_fix.
+
+Mon Apr 28 20:24:27 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_marshal_load): checks the given
+	  denominator. [ruby-dev:34536]
+
+Mon Apr 28 14:21:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (POSFIXABLE): use FIXNUM_MAX+1 instead of
+	  FIXNUM_MAX to make it possible to convert to double accurately on
+	  environments with 64bit VALUE and 64bit double.
+	  It assumes FLT_RADIX is 2.
+	  fix RubyForge bug #14102.
+
+Mon Apr 28 12:48:57 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_exec_arg_addopt, rb_exec_arg_addopt): now can specify
+	  close_exec on having no fork environment (but still meaningless).
+
+Mon Apr 28 11:11:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (run_exec_options): don't call FIX2INT for nil.
+
+Mon Apr 28 11:11:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (method_name): should return symbols instead of strings.
+	  [ruby-dev:34531]
+
+Mon Apr 28 09:02:43 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_exec_arg_init): declared.
+	  (rb_exec_arg_addopt): declared.
+	  (rb_exec_arg_fix): declared.
+	  (rb_exec_initarg): removed.
+	  (rb_exec_getargs): removed.
+	  (rb_exec_initarg2): removed.
+
+	* io.c (struct popen_arg): make execarg as a pointer.
+	  (popen_exec): follow popen_arg change.
+	  (pipe_open): add eargp argument.  extract argc and argv from eargp.
+	  use rb_exec_arg_addopt to add redirect options.
+	  (pipe_open_v): set up struct rb_exec_arg.
+	  (pipe_open_s): set up struct rb_exec_arg.
+
+	* process.c (rb_exec_arg_addopt): new function extracted from
+	  check_exec_options_i.
+	  (check_exec_options_i): use rb_exec_arg_addopt.
+	  (rb_check_exec_options): opthash is always a hash now.
+	  (rb_exec_getargs): make it static.
+	  (rb_exec_fillarg): renamed from rb_exec_initarg2.  don't set up
+	  redirect_fds.
+	  (rb_exec_arg_init): new function.
+	  (rb_exec_arg_fix): new function.
+	  (rb_f_exec): use rb_exec_arg_init and rb_exec_arg_fix.  use
+	  rb_exec_arg_addopt to set close_others option.
+	  (run_exec_options): make close_others by default.
+	  (rb_spawn_internal): use rb_exec_arg_init and rb_exec_arg_fix.  use
+	  rb_exec_arg_addopt to set close_others option.
+
+Sun Apr 27 18:59:04 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_expt): use f_rational_new2. [ruby-dev:34524]
+
+Sun Apr 27 15:23:40 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_count): add a GC.count method.  This method returns
+	  a GC invoking count.
+
+Sun Apr 27 12:20:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_vm_t), gc.c (rb_objspace, rb_newobj), vm.c
+	  (Init_BareVM): per-VM object space support, which is disabled now.
+
+	* gc.c (rb_objspace_alloc), vm.c (Init_BareVM): should not use ruby
+	  malloc here.
+
+	* gc.c (garbage_collect, etc): performance improvement by passing the
+	  reference instead of referring the global variable in each functions.
+
+Sun Apr 27 08:06:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (ruby_set_argv): ARGV should be locale encoding.
+	  [ruby-list:44861]
+
+Sun Apr 27 01:46:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.popen3w): removed.
+	  (Open3.popen3): notice wait_thr.
+
+Sun Apr 27 01:13:05 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc, test/rdoc:  Update to RDoc 2.0.0 r56.
+
+Sat Apr 26 21:30:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_hash_dup): declared.
+
+	* hash.c (rb_hash_dup): new function.
+
+	* process.c (rb_spawn_internal): don't modify option hash.
+
+Sat Apr 26 18:36:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c, signal.c, thread.c, thread_win32.c, include/ruby/intern.h:
+	  suppress warnings.
+
+Sat Apr 26 17:42:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (builtin_types), gc.c (count_objects): added Complex and
+	  Rational.
+
+Sat Apr 26 17:35:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_eNOERROR): renamed.
+
+Sat Apr 26 17:30:11 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, gc.c: remove T_BLOCK.
+
+	* include/ruby/ruby.h: re-number T_xxx.
+
+Sat Apr 26 17:31:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_cProcessTms, rb_cProcessStatus): renamed.
+
+	* error.c (builtin_types), signal.c (siglist), st.c (primes),
+	  struct.c (ref_func), time.c (months): constified.
+
+Sat Apr 26 13:00:41 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb: double fork is replaced by spawn with Process.detach.
+	  (Open3.popen3w): new method to access the thread returned by
+	  Process.detach.
+
+Sat Apr 26 00:47:43 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_spawn_internal): new function to specify
+	  default_close_others.
+	  (rb_spawn): specify default_close_others true.
+	  (rb_f_system): call rb_spawn_internal with default_close_others as
+	  false.
+
+Sat Apr 26 12:26:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_each): use INT2FIX() for fixnum values.
+
+Fri Apr 25 17:56:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (free_unused_heaps): preserve last used heap segment to
+	  reduce malloc() call.
+
+Fri Apr 25 17:54:10 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (HEAP_SIZE): use smaller heap segment (2K) for more chance
+	  to be freed.  based on patch from authorNari <authornari at gmail.com>.
+
+	* gc.c (rb_newobj_from_heap): eventually allocate heap segments.
+
+Fri Apr 25 15:35:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_spawn): rb_exec_initarg() returns new argc and argv in
+	  earg.
+
+Fri Apr 25 12:37:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (flatten): returns an instance of same class.
+	  [ruby-core:16554]
+
+Fri Apr 25 10:52:27 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h: define mode_t for umask.
+
+	* process.c (check_exec_options_i, check_exec_fds, run_exec_options):
+	  support "close_others" only when fork(2) is available.
+
+Fri Apr 25 00:16:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: include sys/stat.h for umask.
+
+Thu Apr 24 23:25:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_env_clear): declared.
+	  (rb_io_mode_modenum): declared.
+	  (rb_close_before_exec): declared.
+	  (struct rb_exec_arg): add options and redirect_fds field.
+	  (rb_check_argv): removed.
+	  (rb_exec_initarg): declared.
+	  (rb_exec_getargs): declared.
+	  (rb_exec_initarg2): declared.
+	  (rb_fork): add third argument: fds.
+
+	* io.c (max_file_descriptor): new static variable to record maximum
+	  file descriptor ruby used.
+	  (UPDATE_MAXFD): new macro.
+	  (UPDATE_MAXFD_PIPE): new macro.
+	  (rb_io_mode_modenum): externed.
+	  (rb_sysopen): update max_file_descriptor.
+	  (rb_close_before_exec): new function.
+	  (popen_exec): redirection removed because it is done by extended
+	  spawn mechanism.
+	  (pipe_open): generate a hash for spawn options to specify
+	  redirections.
+	  (pipe_open_v): use rb_exec_getargs.
+	  (pipe_open_s): use rb_exec_getargs.
+	  (rb_io_initialize): update max_file_descriptor.
+
+	* process.c (hide_obj): new function.
+	  (check_exec_redirect_fd): new function.
+	  (check_exec_redirect): new function.
+	  (check_exec_options_i): new function.
+	  (check_exec_fds): new function.
+	  (rb_check_exec_options): new function.
+	  (check_exec_env_i): new function.
+	  (rb_check_exec_env): new function.
+	  (rb_exec_getargs): new function.
+	  (rb_exec_initarg2): new function.
+	  (rb_exec_initarg): new function.
+	  (rb_f_exec): use rb_exec_initarg.
+	  (intcmp): new function.
+	  (run_exec_dup2): new function.
+	  (run_exec_close): new function.
+	  (run_exec_open): new function.
+	  (run_exec_pgroup): new function.
+	  (run_exec_rlimit): new function.
+	  (run_exec_options): new function.
+	  (rb_exec): call run_exec_options.
+	  (move_fds_to_avoid_crash): new function.
+	  (pipe_nocrash): new function.
+	  (rb_fork): use pipe_nocrash to avoid file descriptor conflicts.
+	  (rb_spawn): use rb_exec_initarg.
+	  (rlimit_resource_name2int): extracted from rlimit_resource_type.
+	  (rlimit_type_by_hname): new function.
+	  (rlimit_type_by_lname): new function.
+	  (rlimit_resource_type): use rlimit_type_by_hname.
+	  (proc_daemon): add fds argument for rb_fork.
+
+	* hash.c (rb_env_clear): renamed from env_clear and externed.
+
+	[ruby-dev:34086]
+
+Thu Apr 24 23:00:58 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_thread.rb: fix typos.
+
+	* test/ruby/envutil.rb (rubyexec): move Open3.popen3 call into timeout
+	  block.
+
+Thu Apr 24 22:34:52 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_comparable.rb: new tests for Comparable, to achieve
+	  100% test coverage of compar.c.
+
+Thu Apr 24 17:19:01 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (process_options): set safe_level before loading script.
+	  [ruby-dev:34421]
+
+Thu Apr 24 14:15:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_find_1): prior files with extensions to files sans
+	  extensions.  [ruby-core:16517]
+
+Thu Apr 24 00:26:06 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/ri/descriptions.rb: fixed wrong class nestings.
+
+Thu Apr 24 00:20:01 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_settracefunc.rb: add a test for set_trace_func.
+
+	* test/ruby/envutil.rb: move "rubyexec" method from test_rubyoptions.rb.
+
+	* test/ruby/test_rubyoptions.rb: use rubyexec in envutil.rb.
+
+	* test/ruby/test_thread.rb: add tests to achieve over 90% test coverage
+	  of thread.c.
+
+Wed Apr 23 15:28:52 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/gdbm/test_gdbm.rb (TestGDBM#test_s_open_no_create): failed
+	  notice moved from comment to assertion message. [ruby-dev:29127]
+
+Wed Apr 23 11:49:54 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#each, SortedSet#each, TC_Set#test_each): Return
+	  an enumerator if no block is given.
+
+Wed Apr 23 00:36:03 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/openssl/test_ssl.rb (start_server): add timeout to server.join.
+
+Wed Apr 23 00:18:45 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_symbol.rb (TestSymbol#test_to_proc): Improve
+	  tests of Symbol#to_proc.
+
+Tue Apr 22 22:40:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/drb/drb.rb (DRb::DRbServer::check_insecure_method): should
+	  check method names by symbols, not by strings.  a patch from
+	  Kazuhiro NISHIYAMA <zn at mbf.nifty.com> in [ruby-dev:34487].
+
+Tue Apr 22 22:15:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el (ruby-style-{case,label}-indent): up list from
+	  indentation point.
+
+Tue Apr 22 21:09:05 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (score_table_A0, score_table_F0):
+	  type of content is unsigned char.
+
+	* ext/nkf/nkf-utf8/nkf.c (push_broken_buf): 'c' is nkf_char.
+
+	* ext/nkf/nkf-utf8/nkf.c (push_broken_buf): enc is 0 or pointer.
+
+	* ext/nkf//nkf.c (options): type of option is unsigned char.
+
+Tue Apr 22 20:51:58 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (z_conv): characters must be nkf_char.
+
+Tue Apr 22 19:23:05 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_initialize): Remove an undocumented
+	  feature (passing a block to the constructor) that's broken.
+	  This is not what I intended.
+
+Tue Apr 22 17:54:05 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* vm_core.h (exec_event_hooks): ``inline'' is a type modifier, not
+	  a type itself.
+
+Tue Apr 22 16:24:27 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_buf_cat): do not use C++ comments.
+
+Tue Apr 22 16:23:53 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* configure.in: use AC_USE_SYSTEM_EXTENSIONS.
+
+Tue Apr 22 16:23:16 2008  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* vm_evalbody.c (DECL_SC_REG): use __asm__ instead.
+
+Tue Apr 22 16:18:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_evalbody.c (DECL_SC_REG): typo fixed.
+
+Tue Apr 22 15:25:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (struct timespec): needs time.h according to POSIX.
+
+Tue Apr 22 13:19:48 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (rb_thread_stop_timer_thread): should clear
+	  timer_thread_id after stopping it.
+
+Tue Apr 22 13:12:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_join): remove the current thread from the join list
+	  of the target thread.
+
+Tue Apr 22 12:03:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_get_ev_const): search from the base klass if it
+	  is given.
+
+Tue Apr 22 09:58:13 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c: avoid warnings.
+
+Tue Apr 22 09:56:51 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (eaccess): workaround for recent msvcrt's behavior.
+	  [ruby-core:16460]
+
+Mon Apr 21 19:08:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_body): call rb_io_check_readable and
+	  rb_io_check_writable.
+
+Mon Apr 21 17:45:27 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/dbm/dbm.c (fdbm_each_value, fdbm_each_key, fdbm_each_pair):
+	  GDBM#{each,each_pair,each_key,each_value}: Return an enumerator
+	  if no block is given.
+
+	* ext/gdbm/gdbm.c (fgdbm_each_value, fgdbm_each_key,
+	  fgdbm_each_pair): GDBM#{each,each_pair,each_key,each_value}:
+	  Return an enumerator if no block is given.
+
+	* ext/openssl/ossl_config.c (ossl_config_each):
+	  OpenSSL::Config#each: Return an enumerator if no block is given.
+
+	* ext/readline/readline.c (hist_each): Readline::HISTORY#each:
+	  Return an enumerator if no block is given.
+
+	* ext/sdbm/init.c (fsdbm_each_value, fsdbm_each_key,
+	  fsdbm_each_pair): SDBM#{each,each_pair,each_key,each_value}:
+	  Return an enumerator if no block is given.
+
+	* ext/stringio/stringio.c (strio_each_byte, strio_each):
+	  StringIO#{each,each_line,each_byte}: Return an enumerator if no
+	  block is given.
+
+	* ext/stringio/stringio.c (Init_stringio): Add #lines and #bytes,
+	  which are aliases to #each_line and #each_byte, respectively.
+
+	* ext/win32ole/win32ole.c (fole_each): WIN32OLE#each: Return an
+	  enumerator if no block is given.
+
+	* ext/zlib/zlib.c (rb_gzreader_each_byte, rb_gzreader_each):
+	  Zlib::GzipReader#{each,each_line,each_byte}: Return an
+	  enumerator if no block is given.
+
+	* ext/zlib/zlib.c (Init_zlib): Add Zlib::GzipReader#lines and
+	  #bytes, which are aliases to #each_line and #each_byte,
+	  respectively.
+
+Mon Apr 21 17:01:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_compile_with_option): check if src is a string.
+	  [ruby-core:16453]
+
+Mon Apr 21 16:06:47 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_init): preserve the method name in ID.
+
+	* enumerator.c (enumerator_each): need not to call rb_to_id().
+
+	* enumerator.c (enumerator_with_index): ditto.
+
+Mon Apr 21 11:00:27 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (defined_expr): capture exception during defined?
+	  evaluation.  a patch from wanabe <s.wanabe at gmail.com> in
+	  [ruby-dev:34461].  [ruby-core:16010]
+
+Mon Apr 21 10:06:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c: should include <errno.h> to refer to errno.
+
+Mon Apr 21 09:58:04 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (rb_strftime): check errno to detect strftime(3)'s error.
+	  this is workaround for recent version of MSVCRT.
+	  [ruby-dev:34456]
+
+Mon Apr 21 08:54:30 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (ruby_xmalloc): use size_t for malloc argument instead of long.
+
+Sun Apr 20 21:00:21 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c, include/ruby/ruby.h: Export rb_cEnumerator.
+
+Sun Apr 20 20:47:50 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c: Resolve the method every time an enumeration
+	  method is run, not once when the enumerator is initialized as it
+	  was before, so that method_missing() and method (re)definition
+	  afterwards are both in effect; pointed out in: [ruby-core:16441]
+
+Sun Apr 20 15:11:00 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_rbuf_to_dst): removed.
+	  (copy_stream_fallback_body): don't bypass write method.
+	  (copy_stream_body): simplified.
+
+Sun Apr 20 15:01:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (struct iseq_compile_data): moved label_no from
+	  new_label_body().
+
+	* compile.c (iseq_set_exception_table): allocates catch_table only
+	  when entries exist.
+
+	* compile.c (struct iseq_link_element, struct iseq_insn_data): made
+	  enum for debuggers.
+
+Sun Apr 20 14:44:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): fix for splat in when and rescue.
+	  a patch from wanabe <s.wanabe AT gmail.com> in [ruby-dev:34429].
+	  [ruby-core:14537]
+
+Sun Apr 20 13:55:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_fallback): write directly (bypassing write method)
+	  if possible.
+
+Sun Apr 20 12:49:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_fallback): read directly (bypassing readpartial
+	  method) if possible.
+
+Sun Apr 20 04:45:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_body): use readpartial and write method for
+	  non-IOs such as StringIO and ARGF.
+
+Fri Apr 18 20:57:33 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_array.rb: add tests to achieve over 95% test coverage
+	  of array.c.
+
+Fri Apr 18 17:37:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_mark_locations): get rid of underflow.
+
+	* gc.c (mark_current_machine_context): check if the main thread stack
+	  position may shrink under the initialized position.  [ruby-core:16436]
+
+Thu Apr 17 22:20:52 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enc/trans/utf_16_32.c (fun_so_to_utf_16be, fun_so_to_utf_16le): add
+	  parentheses to remove warnings of gcc.
+
+	* io.c (rb_io_getc): remove unused variables.
+
+	* compile.c (NODE_NEXT, NODE_REDO): remove unused labels.
+
+	* ext/nkf/nkf.c (rb_nkf_convert): remove unused variables.
+
+	* ext/syck/rubyext.c (syck_resolver_initialize,
+	  syck_resolver_detect_implicit, syck_emitter_emit): remove unused
+	  variables.
+
+Thu Apr 17 20:12:47 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rubyoptions.rb (test_search): enable some assertions.
+
+	* test/ruby/test_rubyoptions.rb: flunk message in win32.
+
+Thu Apr 17 16:07:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (ruby): run in C locale.
+
+	* test/ruby/test_rubyoptions.rb (test_encoding): --encoding does not
+	  affect source code.
+
+Thu Apr 17 00:45:41 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_process.rb (test_rlimit_nofile): reset RLIMIT_NOFILE
+	  before exit (for gcov).
+
+	* test/ruby/test_rubyoptions.rb: new tests for option of ruby
+	  interpreter, to achieve over 95% test coverage of ruby.c.
+
+Wed Apr 16 02:40:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): preludes and parser need to run in safe
+	  level 0.  [ruby-dev:34407]
+
+Wed Apr 16 02:26:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): dln_find_file returns the pointer to a
+	  static buffer, so should copy it.  [ruby-dev:34409]
+
+Tue Apr 15 23:08:46 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/xmlrpc/client.rb: fix cookie handling. [ruby-dev:34403]
+
+	* test/xmlrpc/test_cookie.rb: add a test for the above fix.
+
+Tue Apr 15 19:20:14 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c: #undef rb_argv moved before #define.
+
+Tue Apr 15 18:02:17 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_argv): replaced with rb_get_argv().
+	  [ruby-Bugs-19514]
+
+Tue Apr 15 17:10:59 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/net/http.rb, lib/net/smtp.rb, lib/net/pop.rb: update
+	  URLs of Japanese documents.
+
+Tue Apr 15 16:45:14 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_symbol.rb (TestSymbol#test_to_proc): add tests.
+
+Tue Apr 15 15:38:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-encoding-map): added shift-jis for older
+	  versions.
+
+	* misc/ruby-mode.el (ruby-mode-set-encoding):
+	  coding-system-to-mime-charset is not a standard function.
+	  [carbon-emacs:795]
+	  fix for the case that magic comment exists but coding system is
+	  absent.
+
+	* misc/ruby-mode.el (ruby-mode): use write-contents-functions or
+	  write-contents-hooks for older versions.
+
+Tue Apr 15 07:21:21 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_div): [ruby-dev:34357]
+
+	* complex.c (nucomp_abs): use hypot.
+
+	* complex.c (nucomp_quo): do not force conversion.
+
+	* test/ruby/test_complex.rb: omitted some meaningless tests.
+
+Mon Apr 14 23:25:50 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_objectspace.rb: add a test for
+	  ObjectSpace.count_objects.
+
+Mon Apr 14 22:44:24 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (SET_EXTERNAL_ENCODING): avoid call rb_enc_check() on
+	  half-baked result string.
+
+	* re.c (rb_reg_search): make search reentrant.  [ruby-dev:34223]
+
+	* test/ruby/test_parse.rb (TestParse::test_global_variable):
+	  should preserve $& variable.
+
+Mon Apr 14 17:23:27 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* hash.c (rb_hash_delete_if, rb_hash_reject_bang, env_delete_if,
+	  env_reject_bang): Return an enumerator if no block is given.
+
+Mon Apr 14 14:33:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c, compile.h (compile_debug): made runtime option.
+
+	* debug.c (ruby_debug_print_indent): returns if debug_level exceeds
+	  the threshold.
+
+	* debug.c (ruby_debug_printf): printf to stderr.
+
+	* iseq.c (make_compile_option, make_compile_option_value): added
+	  debug_level option.
+
+	* vm_core.h (rb_compile_option_t): added debug_level.
+
+	* vm_core.h (struct iseq_compile_data): added node_level.
+
+Mon Apr 14 12:52:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (Init_stack): use ruby_init_stack.  [ruby-dev:34350]
+
+	* gc.c (rb_objspace_t): packed globals.  [ruby-dev:34348]
+
+	* gc.c (finalizers): removed.  [ruby-dev:34349]
+
+Mon Apr 14 11:30:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (ary_new): new integer overflow check condition.
+	  suggested by TOYOFUKU Chikanobu <nobu_toyofuku at nifty.com> in
+	  [ruby-dev:34156].
+
+	* array.c (rb_ary_initialize): ditto.
+
+Mon Apr 14 00:51:40 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_parse.rb: add tests to achieve over 95% test coverage
+	  of parse.y.
+
+Sun Apr 13 23:53:58 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enum.c (enum_cycle): Make Enumerable#cycle do a finite loop when
+	  the number of cycles is specified.
+
+	* array.c (rb_ary_cycle): Ditto for Array#cycle.
+
+Sun Apr 13 18:52:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (lock_func): should not check interrupts in
+	  blocking region.  [ruby-dev:34378]
+
+Sat Apr 12 12:41:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_exec_node, ruby_run_node), ruby.c (process_options):
+	  use iseq instead of NODE.
+
+	* gc.c (source_filenames): removed.
+
+	* include/ruby/intern.h, parse.y (yycompile, parser_mark, parser_free,
+	  ripper_initialize): rb_source_filename() is no longer used.
+
+	* compile.c, compile.h (ERROR_ARGS), parse.y (node_newnode, fixpos,
+	  parser_warn, e_option_supplied, warn_unless_e_option, range_op,
+	  cond0): nd_file is no longer used.
+
+Sat Apr 12 12:17:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* prelude.rb (require_relative): move require_relative from
+	  lib/require_relative.rb.  [ruby-core:16356]
+
+	* lib/require_relative.rb: removed.
+
+Sat Apr 12 05:55:57 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*, test/rubygems*:  Update to RubyGems 1.1.1 r1701.
+
+Sat Apr 12 03:13:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): set external encoding.
+
+	* file.c (rb_file_s_basename, rb_file_s_dirname, rb_file_s_extname):
+	  copy encoding.
+
+Fri Apr 11 17:35:06 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (count_i): modified to shut warning up.
+
+Fri Apr 11 17:25:09 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enum.c (count_i, count_iter_i, enum_count, enum_find_index):
+	  Reduce code.
+
+Fri Apr 11 17:06:01 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (find_index_i): modified to shut warning up.
+
+	* enum.c (find_index_iter_i): ditto.
+
+Fri Apr 11 16:44:43 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enum.c (enum_find_index): Add support for find_index(obj);
+	  [ruby-dev:34313].
+
+	* array.c (rb_ary_index): Define find_index as an alias to index.
+
+Fri Apr 11 16:42:33 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/yaml/store.rb (YAML::load): modified to support empty
+	  database.
+
+Fri Apr 11 08:05:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_object): add volatile to avoid potential GC bug.  a
+	  patch from Tomoyuki Chikanaga <chikanag at nippon-control-system.co.jp>
+	  in [ruby-dev:34312].
+
+Thu Apr 10 23:08:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/pstore.rb (PStore::dump, PStore::load): allow subclass
+	  overriding.  [ruby-dev:34305]
+
+	* lib/yaml/store.rb (YAML::Store::marshal_dump_supports_canonical_option?):
+	  add a method to support faster PStore.
+
+Thu Apr 10 20:36:45 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/rdebug.el, misc/README: Remove rdebug.el as per request
+	  from the maintainer and mention the ruby-debug project at
+	  RubyForge in README; bug#19043.
+
+Thu Apr 10 19:41:00 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* eval.c (rb_f_loop): Mention StopIteration in the document.
+
+Thu Apr 10 19:23:55 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_pop_m, rb_ary_shift_m): Update documents for
+	  #pop() and #shift().
+
+	* array.c (rb_ary_slice_bang): Update document.  Assigning
+	  array[*args]= nil no longer removes elements.
+
+Thu Apr 10 16:58:44 2008  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (w_object): TYPE_USERDEF assigns id for ivars first.
+	  [ruby-dev:34159] by nagachika.
+
+Thu Apr 10 15:03:47 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/generator.rb: removed obsolete library.  [ruby-core:16233]
+
+	* test/test_generator.rb: removed as well.  [ruby-dev:34306]
+
+	* lib/pstore.rb: replaced by Hongli Lai's faster version.
+
+Thu Apr 10 10:27:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (native_sleep): sleep_cond is initialized at
+	  creation.  [ruby-Patches-19361].
+
+Wed Apr  9 14:43:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (lock_func): optimized and checks for interrupt_flag.
+	  based on a patch from Sylvain Joyeux in [ruby-Patches-19361] and
+	  [ruby-Patches-19362].
+
+Wed Apr  9 12:12:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_thread.rb: new tests from Sylvain Joyeux in
+	  [ruby-Patches-19361].
+
+Tue Apr  8 21:36:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_mutex_sleep): ensures to re-acquire at waking up.
+	  [ruby-Patches-19361]
+
+Tue Apr  8 11:00:14 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/complex.rb: remove Math first before overwriting by CMath.
+
+Tue Apr  8 10:34:10 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* load.c (rb_require_safe): should check fname path after $SAFE is
+	  properly set.  [ruby-dev:34268]
+
+	* re.c (rb_reg_quote): should always copy the quoting string.
+	  [ruby-core:16235]
+
+Tue Apr  8 10:30:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (prelude.c): depends on enc/prelude.rb.
+
+	* enc/prelude.rb: fixed initial library names.
+
+Tue Apr  8 03:39:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_provided): check expanded path for relative path
+	  features, loading or loaded features are already expanded in 1.9.
+
+	* variable.c (rb_autoload_load): no needs to check if provided before
+	  rb_require_safe.  [ruby-dev:34266]
+
+Mon Apr  7 22:41:21 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* numeric.c: cancelled recent changes (except to remove rdiv).
+
+	* bignum.c: ditto.
+
+	* bignum.c: added rb_big_idiv.
+
+Mon Apr  7 15:51:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_init_db): moved to enc/encdb.c.
+
+	* transcode.c (init_transcoder_table): moved to enc/trans/transdb.c.
+
+	* enc/depend (enc/encdb.o enc/trans/transdb.o): depend on
+	  corresponding headers.
+
+	* common.mk (COMMONOBJS): moved transcode.o from OBJS
+
+Mon Apr  7 12:26:32 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: add a known-bug.
+
+Mon Apr  7 12:15:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_each_func): should not leave a variable
+	  uninitialized, which could cause SEGV.
+
+	* range.c (range_step): removed duplicated and unreachable code.
+
+Mon Apr  7 02:12:27 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_intern): need not to check if tainted.
+	  [ruby-dev:34219]
+
+Sun Apr  6 09:45:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_tell): check if closed.  [ruby-core:16223]
+
+Sat Apr  5 23:17:20 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb: new.
+
+	* lib/complex.rb: depends lib/cmath.rb.
+
+	* lib/rational.rb: added rdiv.
+
+	* complex.c: removed some math functions.
+
+Sat Apr  5 05:50:57 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/parse_rb.rb: Fix uninitialized variable warnings.
+
+	* lib/rdoc/generator/html.rb: ditto.
+
+	* lib/rdoc/options.rb: Fix shadowed variable warning.
+
+	* lib/webrick/httprequest.rb: Fix redefined method warning.
+
+Sat Apr  5 02:13:52 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_cmpint): moved from compar.c, to check bignum
+	  zero.
+
+Fri Apr  4 23:24:06 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_memsearch_qs): wrong boundary condition.
+
+	* re.c (rb_memsearch_qs_utf8): ditto.
+
+Fri Apr  4 14:11:36 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_memsearch_qs): wrong boundary condition.  a patch from
+	  wanabe <s.wanabe AT gmail.com> in [ruby-dev:34248].
+
+Fri Apr  4 05:57:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/pop.rb (Net::POP3::do_finish): clear @n_mails and
+	  @n_bytes as well.  [ruby-core:16144]
+
+Fri Apr  4 01:59:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_step): add step for each iteration if begin and
+	  end are numeric.  [ruby-core:15990]
+
+Fri Apr  4 00:42:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (Init_Bignum): rdiv method removed.  [ruby-dev:34242]
+
+	* complex.c (nucomp_quo): ditto.
+
+	* numeric.c (num_rdiv): ditto.
+
+	* rational.c (nurat_div): ditto.
+
+	* complex.c (nucomp_fdiv): fdiv implementation restored.
+
+	* numeric.c (num_quo): RDoc updated.
+
+Thu Apr  3 21:51:45 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_int_check): function for DRY real check.
+
+	* complex.c (nucomp_{add,sub,mul,div,expt}): use rb_num_coerce_bin().
+
+Thu Apr  3 19:59:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defineclass): check if cbase is a class or a module.
+	  [ruby-core:16118]
+
+Thu Apr  3 14:42:11 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (INSNS): add insns_info.inc.
+
+	* common.mk (INSNS): make incs separately for nmake.
+
+Thu Apr  3 13:20:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (endb.h, transdb.h, prelude.c): depend on $(PREP) and
+	  check if really changed.  [ruby-core:16102]
+
+	* Makefile.in, common.mk, configure.in, {win32,bcc32}/Makefile.sub
+	  (MINIOBJS, ARCHMINIOBJS): separated.
+
+Thu Apr  3 09:00:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_sequence, iseq_insns_unification,
+	  insn_data_to_s_detail): constified.
+
+	* iseq.c (insn_operand_intern, ruby_iseq_disasm_insn): ditto.
+
+	* template/{insns_info,opt_sc,optunifs}.inc.tmpl: ditto.
+
+	* tool/instruction.rb (OptUnifsIncGenerator): ditto.
+
+Thu Apr  3 08:46:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_include): add RDoc to describe that comparison
+	  for numeric is done according magnitude of values.
+	  [ruby-core:15907]
+
+Wed Apr  2 22:29:35 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* rational.c (nurat_int_check): function for DRY integer check.
+
+	* numeric.c (num_rdiv): should always return rational number.
+
+	* rational.c (nurat_add, nurat_sub, nurat_mul, nurat_fdiv,
+	  nurat_cmp): use rb_num_coerce_bin().
+
+	* rational.c (nurat_division): does / and rdiv.
+
+	* .gdbinit (rp): no longer use rb_p().
+
+Wed Apr  2 06:52:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* .gdbinit (rp): supports rational and complex numbers.  it's
+	  cheating since it uses rb_p().
+
+Wed Apr  2 06:24:06 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/node.h: add new constants for rb_call()'s scope.
+
+	* eval.c (iterate_method): use CALL_* scope constant to specify
+	  proper scope value.
+
+	* eval.c (rb_each, rb_apply, rb_funcall, rb_funcall2, rb_funcall3):
+	  ditto.
+
+Tue Apr  1 21:19:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* rational.c: need to include <float.h> just once.
+
+Tue Apr  1 16:40:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (big2dbl): more precise conversion at edge cases.
+	  [ruby-dev:34195]
+
+Tue Apr  1 14:43:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: get rid of empty expansion.
+
+	* configure.in: _setjmp is available but _longjmp is not on mingw.
+
+Tue Apr  1 09:41:22 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc,win}32/Makefile (config.h): need to define RUBY_SETJMP, etc.
+
+Tue Apr  1 07:31:58 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems* test/rubygems*: Import RubyGems 1.1.0.
+
+Tue Apr  1 03:20:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_SETJMP, RUBY_LONGJMP, RUBY_JMP_BUF): prefers
+	  _setjmp over setjmp and sigsetjmp.  [ruby-core:16023]
+	  __builtin_setjmp cannot handle a variable.
+
+	* configure.in (--with-setjmp-type): new option to override the
+	  default rule in the above.
+
+	* eval_intern.h (ruby_setjmp, ruby_longjmp), gc.c (rb_setjmp),
+	  vm_core.h (rb_jmpbuf_t): use RUBY_SETJMP, RUBY_LONGJMP and
+	  RUBY_JMP_BUF.
+
+Tue Apr  1 01:55:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::Config.default_config_hash): requires
+	  win32/resolv to use Win32::Resolv.  [ruby-dev:34138]
+
+Tue Apr  1 01:40:58 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: adopted the ruby's style.
+
+	* rational.c: ditto.
+
+Tue Apr  1 00:17:35 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: revert.
+
+Mon Mar 31 18:57:36 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc,win}32/Makefile.sub (config.h): define ssize_t.
+
+	* io.c (copy_stream_body): some platform don't have O_NOCTTY.
+
+Mon Mar 31 18:42:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check for ssize_t.  [ruby-dev:34184]
+
+Mon Mar 31 14:45:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (num_quo): should convert its operand to Rational.
+
+	* rational.c (string_to_r_strict): should raise TypeError.
+
+	* bignum.c (Init_Bignum): should not redefine Bignum#div.
+	  Numeric#div will do.  [ruby-dev:34066]
+
+Mon Mar 31 04:05:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_getc): set coderange while getting characters.
+
+Sun Mar 30 23:16:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (proc_dup): should copy is_lambda attribute as well.
+	  [ruby-talk:296244]
+
+Sun Mar 30 15:33:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: IO.copy_stream implemented.  [ruby-dev:33843]
+
+	* thread.c (rb_fd_select): new function.
+
+	* configure.in (sys/sendfile.h): check the header file.
+	  (sendfile): check the function.
+	  (pread): check the function.
+
+Sat Mar 29 14:18:41 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/*: full update Ruby/Tk to support Ruby(1.9|1.8) and Tc/Tk8.5.
+
+	* ext/tk/lib/tkextlib/tile.rb: [incompatible] remove TileWidgets'
+	  instate/state/identify method to avoid the conflict with standard
+	  widget options. Those methods are renamed to ttk_instate/ttk_state/
+	  ttk_identify (tile_instate/tile_state/tile_identify are available
+	  too). Although I don't recommend, if you really need old methods,
+	  please define "Tk::USE_OBSOLETE_TILE_STATE_METHOD = true" before
+	  "require 'tkextlib/tile'".
+
+	* ext/tk/lib/tkextlib/tile.rb: "Tk::Tile::__Import_Tile_Widgets__!"
+	  is obsolete. It outputs warning. To control default widget set,
+	  use "Tk.default_widget_set = :Ttk".
+
+	* ext/tk/lib/tk.rb: __IGNORE_UNKNOWN_CONFIGURE_OPTION__ method and
+	  __set_IGNORE_UNKNOWN_CONFIGURE_OPTION__!(mode) method are defined
+	  as module methods of TkConfigMethod. It may help users to wrap old
+	  Ruby/Tk scripts (use standard widgets) to force to use Ttk widgets.
+	  Ttk widgets don't have some options of standard widgets which are
+	  control the view of widgets. When set ignore-mode true, configure
+	  method tries to ignore such unknown options with no exception.
+	  Of course, it may raise other troubles on the GUI design.
+	  So, those are a little danger methods.
+
+	* ext/tk/lib/tk/itemconfig.rb: __IGNORE_UNKNOWN_CONFIGURE_OPTION__
+	  method and __set_IGNORE_UNKNOWN_CONFIGURE_OPTION__!(mode) method
+	  are defined as module methods of TkItemConfigMethod as the same
+	  purpose as TkConfigMethod's ones.
+
+	* ext/tk/sample/ttk_wrapper.rb: A new example. This is a tool for
+	  wrapping old Ruby/Tk scripts (which use standard widgets) to use
+	  Ttk (Tile) widgets as default.
+
+	* ext/tk/sample/tkextlib/tile/demo.rb: use ttk_instate/ttk_state
+	  method instead of instate/state method.
+
+	* ext/tk/lib/tk/root, ext/tk/lib/tk/namespace.rb,
+	  ext/tk/lib/tk/text.rb, ext/tk/lib/tkextlib/*: some 'instance_eval's
+	  are replaced to "instance_exec(self)".
+
+	* ext/tk/lib/tk/event.rb: bug fix on KEY_TBL and PROC_TBL (?x is not
+	  a character code on Ruby1.9).
+
+	* ext/tk/lib/tk/variable.rb: support new style of operation argument
+	  on Tcl/Tk's 'trace' command for variables.
+
+	* ext/tk/sample/demos-jp/widget, ext/tk/sample/demos-en/widget: bug fix
+
+	* ext/tk/sample/demos-jp/textpeer.rb,
+	  ext/tk/sample/demos-en/textpeer.rb: new widget demo.
+
+	* ext/tk/tcltklib.c: decrease SEGV troubles (probably)
+
+	* ext/tk/lib/tk.rb: remove Thread.critical access if Ruby1.9
+
+	* ext/tk/lib/tk/multi-tk.rb: support Ruby1.9 (probably)
+
+	* ext/tk/lib/tkextlib/tile.rb: add method to define Tcl/Tk command
+	  to make Tcl/Tk theme sources (based on different version of Tile
+	  extension) available.
+	  (Tk::Tile::__define_LoadImages_proc_for_compatibility__)
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tk/wm.rb: support dockable frames
+	  (Tcl/Tk8.5 feature). 'wm' command can treat many kinds of widgets
+	  as toplevel widgets.
+
+	* ext/tk/lib/tkextlib/tile/style.rb: ditto.
+	  (Tk::Tile::Style.__define_wrapper_proc_for_compatibility__)
+
+	* ext/tk/lib/tk/font.rb: add actual_hash and metrics_hash to get
+	  properties as a hash. metrics_hash method returns a boolean value
+	  for 'fixed' option. But metrics method returns numeric value
+	  (0 or 1) for 'fixed' option, because of backward compatibility.
+
+	* ext/tk/lib/tk/timer.rb: sometimes fail to set callback procedure.
+
+	* ext/tk/lib/tk.rb: add Tk.sleep and Tk.wakeup method. Tk.sleep
+	  doesn't block the eventloop. It will be better to use the method
+	  in event callbacks.
+
+	* ext/tk/sample/tksleep_sample.rb: sample script about Tk.sleep.
+
+Thu Mar 27 20:44:22 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (f_lcm): removed.
+
+	* rational.c (rb_lcm, rb_gcdlcm): added.
+
+	* lib/complex.rb (gcd, lcm, gcdlcm): removed.
+
+	* lib/rational.rb (gcd, lcm, gcdlcm): ditto.
+
+Wed Mar 26 18:11:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_mod_constants): rdoc updated.  a patch from
+	  Florian Gilcher <flo AT andersground.net> in [ruby-core:16009].
+
+Wed Mar 26 00:55:28 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rand.rb: add tests to achieve over 95% test coverage
+	  of random.c.
+
+Wed Mar 26 00:28:55 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_rational.rb: add tests to achieve over 90% test
+	  coverage of rational.c.
+
+	* test/ruby/test_complex.rb: ditto for complex.c.
+
+Tue Mar 25 19:34:05 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bootstraptest/test_knownbug.rb: add tests.  [ruby-dev:34128]
+
+Tue Mar 25 19:09:04 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (ary_new): fix size check.  [ruby-dev:34123]
+
+	* array.c (rb_ary_take, rb_ary_drop): check negative size and use
+	  NUM2LONG instead of FIX2LONG.  [ruby-dev:34123]
+
+	* enum.c (enum_take, enum_drop): check negative size.
+
+	* test/ruby/test_array.rb: add tests for above.
+
+Tue Mar 25 16:32:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): checks if the word is empty.
+
+	* ruby.c (process_options): typo fixed.  [ruby-dev:34122]
+
+Tue Mar 25 15:26:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (defined_expr): false short-circuit destination label may
+	  be needed.  [ruby-talk:295296]
+
+	* compile.c (iseq_compile_each): put nil if false short-circuit is
+	  created.
+
+	* compile.c (compile_massign_opt): no need to use alloca.
+
+Mon Mar 24 19:23:52 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* parse.y (debug_lines): Always prepare a new array for each
+	  file's SCRIPT_LINES__ storage, instead of appending source lines
+	  every time a file is re-loaded; submitted by Rocky Bernstein in
+	  #18517.
+
+Mon Mar 24 10:25:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: sitearch should use target_cpu.  [ruby-core:15986]
+
+Sun Mar 23 02:51:57 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rlimit_resource_value): use NUM2RLIM.
+
+Sun Mar 23 02:28:01 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: fixed. [ruby-dev:34109]
+
+	* rational.c: ditto.
+
+Fri Mar 21 21:32:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_f_gets, rb_f_readline, rb_f_readlines): delegates to ARGF
+	  as well as puts and putc.  [ruby-dev:34100]
+
+Fri Mar 21 21:26:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::Hosts): should not use win32/resolv on cygwin.
+	  [ruby-dev:29945], [ruby-dev:34095]
+
+	* lib/win32/registry.rb (Win32::Registry.expand_environ): try upcased
+	  name too for cygwin.  [ruby-dev:29945]
+
+	* lib/win32/resolv.rb (Win32::Resolv.get_hosts_path): use expand_path.
+
+Fri Mar 21 21:10:00 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb: Say that I am the current maintainer.
+
+	* lib/set.rb: Ditto.
+
+	* lib/shellwords.rb: Ditto.
+
+	* ext/syslog/syslog.txt: Ditto.
+
+Fri Mar 21 09:24:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (open_for_install): write block result and rewrite only
+	  if changed from existing file.
+
+Fri Mar 21 08:29:33 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_to_f): rearrangement.
+
+Fri Mar 21 06:44:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* rational.c (nurat_to_f): C99.
+
+Fri Mar 21 01:40:27 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* complex.c (nucomp_sub, nucomp_expt): call corresponding functions.
+
+Fri Mar 21 01:21:43 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* missing/tgamma.c: include config.h before math.h.  [ruby-dev:34075]
+
+Thu Mar 20 21:46:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_getline): use receiver.
+
+Thu Mar 20 21:20:19 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: some improvements (include Shin-ichiro HARA's
+	  effort).
+
+	* complex.c: some improvements.
+
+	* test/ruby/test_rational2.rb: new.
+
+Thu Mar 20 00:21:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_initialize_copy): get rid of segfault.
+
+	* io.c (argf_tell, argf_seek_m, argf_set_pos, argf_rewind,
+	  argf_fileno, argf_to_io, argf_eofl, argf_getc, argf_getbyte,
+	  argf_readchar, argf_readbyte, argf_each_line): use receiver.
+
+Wed Mar 19 23:52:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_putc, rb_io_puts): output directly if the receiver is
+	  rb_stdout to get rid of infinite recursion.  [ruby-dev:34059]
+
+Wed Mar 19 22:27:41 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c: added rb_gcd.
+
+	* complex.c: use rb_gcd.
+
+Wed Mar 19 18:37:00 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: revert.
+
+	* rational.c: revert.
+
+Wed Mar 19 17:31:20 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_intern.h (TH_EXEC_TAG): need not to FLUSH_REGISTER_WINDOWS.
+	  [ruby-core:15871], [ruby-dev:34088]
+
+Wed Mar 19 14:53:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (nucomp_to_s, nucomp_inspect): get rid of making
+	  unnecessary intermediate objects.
+
+	* complex.c (make_patterns, string_to_c): do not treat successive
+	  underscores as a part of numeric like as literals.  [ruby-dev:34085]
+
+	* rational.c (make_patterns, string_to_r): ditto.
+
+Wed Mar 19 14:36:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_cstr_to_inum): treat successive underscores as
+	  nondigit.  [ruby-dev:34089]
+
+Wed Mar 19 14:08:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_check_encoding): should not load autoloaded encoding
+	  directly, instead use rb_enc_find_index() which deal with alias and
+	  replica.  [ruby-core:15957]
+
+Wed Mar 19 11:49:47 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* regint.h (include): include ruby.h instead of defines.h and config.h.
+
+Wed Mar 19 10:17:12 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regint.h (CHECK_INTERRUPT_IN_MATCH_AT): add interrupt check
+	  during match.  [ruby-talk:295002]
+
+Tue Mar 18 16:24:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (literal_concat_gen): bail out at different encoding.
+
+Tue Mar 18 04:00:27 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_memsearch_ss): simple shift search.
+
+	* re.c (rb_memsearch_qs): quick search.
+
+	* re.c (rb_memsearch_qs_utf8): quick search for UTF-8 string.
+
+	* re.c (rb_memsearch_qs_utf8_hash): hash functions for above.
+
+	* re.c (rb_memsearch): use above functions.
+
+	* string.c (rb_str_index): give enc to rb_memsearch.
+
+	* include/ruby/intern.h (rb_memsearch): move to encoding.h.
+
+	* include/ruby/encoding.h (rb_memsearch): move from intern.h.
+
+	* common.mk (PREP): add dependency.
+
+Mon Mar 17 22:23:54 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_take, rb_ary_take_while, rb_ary_drop,
+	  rb_ary_drop_while): new methods. [ruby-dev:34067]
+
+	* test/ruby/test_array.rb: add tests for above.
+
+Mon Mar 17 17:11:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode): should use `run-mode-hooks' instead
+	  of calling `run-hooks' directly to run the mode hook.  patch from
+	  Chiyuan Zhang <pluskid AT gmail.com> in [ruby-core:15915]
+
+Mon Mar 17 16:41:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: unset GREP_OPTIONS.  [ruby-core:15918]
+
+Sun Mar 16 18:07:07 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/utf_16_32.c: bug fix (some invalid UTF-8 sequences
+	  were legal)
+
+	* test/ruby/test_transcode.rb: test for above bug
+
+Sun Mar 16 17:28:07 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (LIBRUBY_SO): add dependency to $(BUILTIN_ENCOBJS).
+
+Sun Mar 16 08:51:41 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* include/ruby/intern.h: added some declarations.
+
+	* include/ruby/ruby.h: ditto.
+
+	* common.mk: added some entries.
+
+	* configure.in: added a check for signbit.
+
+	* lib/complex.rb: nearly all of core definitions have been removed.
+
+	* lib/rational.rb: ditto.
+
+	* lib/mathn.rb: some trivial adjustments.
+
+	* complex.c: new.
+
+	* rational.c: ditto.
+
+	* numeric.c (flo_{quo,rdiv}, fix_fdiv): added.
+
+	* numeric.c ({num,int}_{numerator,denominator}): ditto.
+
+	* bignum.c (rb_big_fdiv): ditto.
+
+	* numeric.c (fix_{quo,pow}): now may yield rational number.
+
+	* bignum.c (rb_big_{quo,pow}): ditto.
+
+	* numeric.c (rb_{int,flo}_induced_from): now can accept rational.
+
+	* gc.c (gc_mark_children, obj_free): now detects complex and rational.
+
+	* inits.c (rb_call_inits): now calls Init_{Complex,Rational}.
+
+	* test/ruby/test_complex.rb: new.
+
+	* test/ruby/test_rational.rb: ditto.
+
+Sat Mar 15 17:48:48 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_associate_index): pass unnecessary enc_capable().
+
+	* string.c (rb_str_cmp): reduce invocation of rb_enc_compatible().
+
+Fri Mar 14 17:04:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (inttypes.h): includes always if available.
+
+	* string.c, ext/digest/defs.h: moved inttypes.h to ruby.h.
+
+Fri Mar 14 16:59:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_LIB_PREFIX): fix for prefix.
+
+Fri Mar 14 16:35:11 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::Cookie::initialize): performance patch from
+	  Makoto Kuwata <kwa@kuwata-lab.com> in [ruby-dev:34048].
+
+Fri Mar 14 15:49:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_LIB_PREFIX): use libdir.
+
+Fri Mar 14 14:24:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/defs.h: inttypes.h is still needed.
+
+Fri Mar 14 11:34:12 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc,win}32/Makefile.sub: follow below changes.
+
+Fri Mar 14 11:24:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-encoding-map, ruby-use-encoding-map): added
+	  to customize.
+
+Fri Mar 14 10:37:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (int8_t, uint8_t,  int16_t,  uint16_t int32_t,
+	  uint32_t int64_t, uint64_t, int128_t, uint128_t,
+	  intptr_t, uintptr_t): check if defined.
+
+	* win32/Makefile.sub: follow configure.in.
+
+	* ext/digest/defs.h: remove checks for uint8_t, uint32_t and uint64_t.
+
+Fri Mar 14 10:12:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_VARTYPE): should not indent preprocessor
+	  directives.
+
+Fri Mar 14 10:03:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (UNALIGNED_WORD_ACCESS): IA64 cannot access unaligned word.
+
+Thu Mar 13 21:00:50 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_slice_bang): should not use rb_ary_subseq()
+	  which shares internal pointer.  splice modifies the receiver
+	  right after subseq.  [ruby-dev:34005]
+
+	* bootstraptest/test_struct.rb: some test moved from test to shut
+	  warning up.
+
+Thu Mar 13 19:42:43 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc,win}32/Makefile.sub (config.h): define uint32_t.
+
+Thu Mar 13 14:14:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* trunk/configure.in (AC_CHECK_HEADERS): stdint.h is not needed to
+	  check.
+
+	* trunk/configure.in (rb_cv_type_uint32_t): unquoted.  [ruby-dev:34030]
+
+	* trunk/string.c (hash): use inttypes.h instead of stdint.h.
+
+Thu Mar 13 10:42:46 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_divmod): should return integer division.  [ruby-dev:34006]
+
+	* enum.c (zip_ary): wrong boundary condition.
+
+	* test/ruby/test_numeric.rb (TestNumeric::test_num2long): bit-and
+	  should not raise RangeError.
+
+Thu Mar 13 03:12:48 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/cmd/help.rb: should be updated for new ri structure.
+	  [ruby-core:15825]
+
+	* lib/rdoc/ri/driver.rb (RDoc::initialize): allow options to be optional.
+
+	* lib/rdoc/ri/driver.rb (RDoc::class_cache): map_dirs may be
+	  empty.
+
+	* lib/rdoc/ri/driver.rb (RDoc::get_info_for): revive get_info_for
+	  method.  maybe broken.
+
+	* lib/rdoc/ri/util.rb (RDoc::initialize): should not use RiError
+	  no more.
+
+Thu Mar 13 01:45:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (stdint.h): check if presence.
+
+	* configure.in (uint32_t): check if defined.
+
+	* string.c (hash): fix for portability.  [ruby-dev:34020]
+
+Wed Mar 12 17:33:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): fix for a mere underscore.
+
+Wed Mar 12 14:47:07 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (rb_thread_raised_set): use generic flags.
+
+	* eval.c (rb_longjmp): clear all raised flags.
+
+	* eval.c (stack_check): leave clearing flag to rb_longjmp.
+
+	* gc.c (rb_memerror): use thread raised flag instead of static flag.
+
+Tue Mar 11 23:38:39 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_combination): argument check before creating
+	  Enumerator.
+
+	* array.c (rb_ary_permutation): ditto.
+
+	* enum.c (enum_zip): optimize if all arguments are arrays.
+
+Tue Mar 11 19:48:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_coerce): try conversion before type check.
+	  [ruby-core:15838]
+
+Tue Mar 11 12:39:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (clean-local): WINMAINOBJ is Windows specific.
+
+Tue Mar 11 10:19:10 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (hash): replaced by MurmurHash described in
+	  <http://murmurhash.googlepages.com/>.
+
+Tue Mar 11 09:52:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_comparable): empty strings in any encoding are
+	  compatible each other.
+
+Tue Mar 11 00:46:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (usage): remove some unimportant lines to fit -h message
+	  in a page.  [ruby-dev:34018]
+
+Mon Mar 10 17:11:00 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_local_variables): local_variables should return an
+	  array of symbols.  [ruby-dev:34008]
+
+	* vm.c (collect_local_variables_in_env): ditto.
+
+Mon Mar 10 15:53:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* version.c (MKSTR): make US-ASCII.  [ruby-dev:34010]
+
+Mon Mar 10 02:08:21 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_index): if t == s + pos, the character beginning
+	  from s + pos is valid.
+
+Sun Mar  9 13:51:21 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/generator.rb: Restore missing line to #params.  Patch by
+	  Lincoln Stoll <lstoll at lstoll.net>
+
+Sun Mar  9 09:52:00 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/code_objects.rb: Remove debugging Kernel#p.  Patch by
+	  Lincoln Stoll <lstoll at lstoll.net>
+	* lib/rdoc/generator/html.rb: Fully qualify AllReferences.  Patch by
+	  Lincoln Stoll <lstoll at lstoll.net>
+	* lib/rdoc/ri/writer.rb: Fix 1.8 backwards compatibility.
+
+Sat Mar  8 18:50:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (isdirsep): backslash is valid path separator on cygwin too.
+
+Sat Mar  8 06:53:48 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (search_nonascii): Use VALUE instead of unsigned long
+	  because VALUE can be the fastest unsigned integer type.
+	  On LLP64 unsigned long isn't the fastest.
+	* string.c (str_strlen): ditto.
+	* string.c (str_utf8_nth): ditto.
+	* string.c (count_utf8_lead_bytes_with_ulong): ditto.
+
+	* string.c (count_utf8_lead_bytes_with_word): renamed.
+
+Fri Mar  7 21:27:43 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c: fix indent.
+
+Fri Mar  7 21:12:19 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (power_cache_init, power_cache_get_power0, Init_Bignum):
+	  delayed initializing power cache per base.  [ruby-dev:34003]
+
+Fri Mar  7 20:30:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_restore_0): fixed typo.  [ruby-core:15821]
+
+Fri Mar  7 19:56:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: rdoc added.  [ruby-Patches-9762]
+
+Thu Mar  6 17:26:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): space flag is in effect for Inf/NaN too.
+	  [ruby-dev:34002]
+
+Thu Mar  6 15:44:20 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): casting double to long is undefined
+	  if the integer part of double is out of the range of long.
+
+Thu Mar  6 15:11:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): ignore 0 flag for NaN and Inf.
+	  [ruby-dev:33994]
+
+Thu Mar  6 15:05:25 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32}/Makefile.sub (RUNRUBY): use $(PROGRAM) instead of
+	  ruby$(EXEEXT).
+	  suggested by KIMURA Koichi <kimura.koichi at canon.co.jp>.
+	  [ruby-dev:34000]
+
+Thu Mar  6 14:46:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* missing/lgamma_r.c (loggamma): return 0 for 1 and 2.
+
+	* test/ruby/test_math.rb: accept errors by functions under missing/.
+
+Thu Mar  6 14:29:44 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (rb_str_transcode_bang): set coderange.
+
+	* transcode.c (rb_str_transcode): use rb_str_transcode_bang.
+
+Thu Mar  6 14:00:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/missing.h (cbrt): add declaration.
+
+Thu Mar  6 11:14:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-add-log-current-method): use ruby style
+	  method name format.
+
+Thu Mar  6 11:12:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): no need of loop.
+
+Thu Mar  6 08:30:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_mod_freeze): call rb_class_name() directly.
+	  [ruby-core:15802]
+
+Thu Mar  6 04:32:06 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (count_utf8_lead_bytes_with_ulong): fix shift size.
+	  [ruby-dev:33993]
+
+	* string.c (str_utf8_nth) fix wrong counting.
+
+Thu Mar  6 00:34:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): size_t returned from strlen() can be
+	  unsigned.
+
+Thu Mar  6 00:31:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (make_struct): preserve encoding of struct name.
+
+Wed Mar  5 22:49:20 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (is_utf8_lead_byte, count_utf8_lead_bytes_with_ulong):
+	  defined for UTF-8 optimization.
+
+	* string.c (str_strlen): use is_utf8_lead_byte and
+	  count_utf8_lead_bytes_with_ulong.
+
+	* string.c (str_utf8_nth) ditto.
+
+Wed Mar  5 17:53:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_flock): returns false on EAGAIN if non-blocking.
+	  [ruby-core:15795]
+
+Wed Mar  5 17:43:43 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c (transcode_loop): Adjusted detection of invalid
+	  (ill-formed) UTF-8 sequences. Fixing potential security issue, see
+	  http://www.unicode.org/versions/Unicode5.1.0/#Notable_Changes.
+
+	* test/ruby/test_transcode.rb: Added two tests for above fix.
+
+Wed Mar  5 14:00:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_to_s): avoid rb_scan_args() when no argument
+	  given.
+	* bignum.c (rb_big_to_s): ditto.
+	* enum.c (enum_first): ditto.
+	* eval_jump.c (rb_f_catch): ditto.
+	* io.c (rb_obj_display): ditto.
+	* class.c (rb_obj_singleton_methods): ditto.
+	* object.c (rb_class_initialize): ditto.
+	* random.c (rb_f_srand): ditto.
+	* range.c (range_step): ditto.
+	* re.c (rb_reg_s_last_match): ditto.
+	* string.c (rb_str_to_i): ditto.
+	* string.c (rb_str_each_line): ditto.
+	* string.c (rb_str_chomp_bang): ditto.
+	* string.c (rb_str_sum): ditto.
+
+	* string.c (str_modifiable): declare inline.
+	* string.c (str_independent): ditto.
+
+Wed Mar  5 11:50:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/debug.rb: require 'continuation' to implement "restart"
+	  command.  [ruby-dev:33992]
+
+	* lib/debug.rb (Context::debug_command): remove local variable
+	  shadowing to shut up warnings.  [ruby-dev:33992]
+
+	* lib/debug.rb (Context::display_list): ditto.
+
+	* lib/debug.rb (Context::resume): ditto.
+
+	* lib/debug.rb (Context::get_thread): no longer use #index for Hash.
+
+Tue Mar  4 21:35:59 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb.rb (IRB::Irb::eval_input): SyntaxError should not be
+	  considered as IRB bug.  [ruby-dev:33991]
+
+	* lib/irb/workspace.rb (IRB::WorkSpace::filter_backtrace): should
+	  filter 'irb.rb' as well for context mode 2 and 3.
+
+Tue Mar  4 19:10:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_aset): should not copy key string when
+	  compare_by_identity is set.  [ruby-dev:33604]
+
+	* hash.c (hash_equal): two hash tables are different when internal
+	  comparison table differ.  [ruby-dev:33989]
+
+Tue Mar  4 16:29:06 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): disallow non digits '0o' expression.
+
+Tue Mar  4 14:35:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (open_key_args): use rb_io_open_with_args instead of rb_f_open.
+	  [ruby-core:15763]
+
+Tue Mar  4 13:41:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (add_heap): fix previous change.  [ruby-dev:33988]
+
+Tue Mar  4 10:21:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (add_heap): use binary search to find the place to insert the
+	  new heap slot.  [ruby-dev:33983]
+
+Tue Mar  4 05:30:31 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (open_key_args): use rb_io_open instead of rb_f_open.
+	  [ruby-core:15746]
+
+Mon Mar  3 23:28:37 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservlet/filehandler.rb: should normalize path
+	  separators in path_info to prevent directory traversal
+	  attacks on DOSISH platforms.
+	  reported by Digital Security Research Group [DSECRG-08-026].
+
+	* lib/webrick/httpservlet/filehandler.rb: pathnames which have
+	  not to be published should be checked case-insensitively.
+
+Mon Mar  3 17:25:45 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (add_heap): sort heaps array in ascending order to use
+	  binary search.
+
+	* gc.c (is_pointer_to_heap): use binary search to identify object
+	  in heaps.  works better when number of heap segments grow big.
+
+Mon Mar  3 17:15:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_regsub): remove too strict encoding check.
+	  [ruby-dev:33966]
+
+Mon Mar  3 16:14:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_any_hash): shrinks all results in Fixnum range.
+	  [ruby-core:15713]
+
+Sun Mar  2 23:03:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_ungetc): reduce redundant call.
+
+Sun Mar  2 10:13:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file): parse shebang in us-ascii.  a patch from
+	  sheepman <sheepman AT sheepman.sakura.ne.jp> in [ruby-dev:33955]
+
+Sun Mar  2 00:08:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): check for successive underscores.
+	  [ruby-dev:33952]
+
+Sat Mar  1 17:59:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (struct argf): packed ARGF stuffs.
+
+	* ruby.c (proc_options): use ruby_set_inplace_mode().
+
+Sat Mar  1 17:51:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb (recursive_collect): do not always
+	  include all test_*.rb.
+
+Sat Mar  1 14:14:17 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* development snapshot 1.9.0-1 released.
+
+Sat Mar  1 13:46:26 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* tool/make-snapshot: make prereq uses MINIRUBY.
+
+	* tool/make-snapshot: allow packaging like 1.9.0-1 by second
+	  command-line argument.
+
+Sat Mar  1 13:11:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/allpairs.rb: new file for all pairs method.
+
+	* test/ruby/test_m17n_comb.rb: use allpairs.rb to reduce test cases.
+
+	* test/ruby/test_sprintf_comb.rb: ditto.
+
+Sat Mar  1 12:34:21 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_inspect): use rb_str_inspect() instead of
+	  rb_str_dump().  [ruby-dev:33946]
+
+Sat Mar  1 12:15:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_method.c (rb_get_method_body): ent->method may be freed by
+	  GC.  [ruby-dev:31819]
+
+	* thread.c (remove_event_hook): should not access freed memory.
+	  [ruby-dev:31820]
+
+Sat Mar  1 10:31:19 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (read_all, rb_io_getline_fast): encoding is io_input_encoding.
+
+Sat Mar  1 10:09:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (tr_setup_table, rb_str_split_m, rb_str_chomp_bang):
+	  simplified with rb_enc_ascget().  [ruby-dev:33944]
+
+Sat Mar  1 10:01:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_coderange_scan_restartable): should not return
+	  offset in the middle of a character.
+
+	* string.c (rb_str_coderange_scan_restartable): should not return
+	  invalid cr value.
+
+Sat Mar  1 09:36:08 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): "%#.0o" should keep prefix where
+	  "%#.0x" should not.
+
+Sat Mar  1 02:35:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (big2str_find_n1): check integer overflow.
+
+Sat Mar  1 00:29:07 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_dummy_p): bootstrap encodings can not be dummy.
+
+	* encoding.c (rb_enc_ascget): no needs to call rb_enc_precise_mbclen()
+	  twice.
+
+Fri Feb 29 23:14:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_m17n_comb.rb (TestM17NComb::test_str_chomp): test
+	  updated.
+
+Fri Feb 29 20:58:09 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_iterator.rb (TestIterator::test_enumerator):
+	  adjust test for zip behavior reversion.
+
+Fri Feb 29 20:25:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_chomp_bang): now works on UTF-16.
+
+	* string.c (tr_setup_table): negation should work on non ASCII
+	  compatible strings as well.
+
+	* string.c (rb_str_split_m): awk split should work on non ASCII
+	  compatible strings as well.
+
+Fri Feb 29 18:08:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_strftime): format should be ascii compatible.
+
+	* parse.y (rb_intern3): non ASCII compatible symbols.
+
+	* re.c (rb_reg_regsub): add encoding check.
+
+	* string.c (rb_str_chomp_bang): ditto.
+
+	* test/ruby/test_utf16.rb (TestUTF16::test_chomp): raises exception.
+
+Fri Feb 29 15:16:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_rpartition): calculation was done in byte indexing.
+
+	* test/ruby/test_m17n_comb.rb (TestM17NComb::test_str_start_with):
+	  allow start_with? matching on broken strings.
+
+Fri Feb 29 15:12:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (opt_block_param): command can start just after block param
+	  definition.  [ruby-list:44479]
+
+Fri Feb 29 03:22:19 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_time.rb (test_readers): fix typo.
+	  (test_strftime): "UTC" is also ok for time.gmtime.strftime("%Z").
+
+Fri Feb 29 02:50:07 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_new): remove encoding assumption of empty string.
+
+	* hash.c ( rb_f_getenv, env_fetch, env_inspect): result of ENV should
+	  be always ASCII-8BIT.
+
+	* object.c (nil_to_s): nil.to_s should be US-ASCII.
+
+Fri Feb 29 02:24:22 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: forgot to update RELEASE_DATE
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tk/text.rb,
+	  ext/tk/lib/tkextlib/iwidgets/scrolledtext.rb: remove adhoc check
+	  of Ruby's features (use existence of some classes instead of
+	  comparing with RUBY_VERSION)
+
+	* ext/tk/lib/tk/root.rb, ext/tk/lib/tk/autoload.rb: make TkRoot
+	  (Tk::Root) unswitchable
+
+	* ext/tk/lib/multi-tk.rb: partial bug fix (still not work!!)
+
+Thu Feb 28 23:37:12 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI::Meta#meta_setup_encoding): use ASCII-8BIT
+	  for charset unspecified non-text data.
+
+Thu Feb 28 22:19:14 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_capable): IMMEDIATE_P doesn't include Qnil and Qfalse.
+	  use SPECIAL_CONST_P.
+
+Thu Feb 28 19:45:52 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (enc_find): check type of argument and convert to String
+	  if it is StringValue. [ruby-cvs:22866]
+
+Thu Feb 28 18:07:52 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI::Meta#meta_setup_encoding): setup encoding
+	  by charset.
+	  (OpenURI::Meta#meta_add_field): call meta_setup_encoding when
+	  content-type.
+
+Thu Feb 28 15:29:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_getline_fast): scan coderange.
+
+Thu Feb 28 14:36:46 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_str_copy): removed.
+
+Thu Feb 28 13:51:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (stack_check): made flag per threads.
+
+	* thread.c (rb_thread_set_raised, rb_thread_reset_raised): prefixed.
+
+Thu Feb 28 11:43:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_flock): immediately returns on EAGAIN if
+	  non-blocking.  [ruby-core:15672]
+
+Thu Feb 28 11:23:50 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_getline_1): get rid of segfault.  [ruby-dev:33938]
+
+Thu Feb 28 11:19:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_reverse_bang): removed unused variables.
+
+	* include/ruby/encoding.h (rb_str_coderange_scan_restartable): added
+	  prototype.
+
+	* string.c (rb_str_coderange_scan_restartable, rb_str_times): removed
+	  unused variables.
+
+	* string.c (rb_str_reverse_bang): ditto
+
+	* string.c (rb_enc_str_copy): unused now.  may be used in future?
+
+Thu Feb 28 03:03:32 2008  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/*: make default widget set
+	  switchable between Tk (standard Tcl/Tk widget set) and
+	  Ttk (Tile). Initial default widget set is Tk. Now, toplevel
+	  widget classes are removed and defined as aliases.
+	  For example, "TkButton" is an alias of the "Tk::Button" class.
+	  Those aliases are replaced when switching default widget set.
+	  "Tk.default_widget_set=" is the method for switching default
+	  widget set. "Tk.default_widget_set = :Ttk" defines Ttk (Tile)
+	  widget set as default. It means that "TkButton" denotes
+	  "Tk::Tile::Button" class. And then, "TkButton.new" creates
+	  a Tk::Tile::Button widget. Of course, you can back to use
+	  standard Tk widgets as the default widget set by calling
+	  "Tk.default_widget_set = :Tk", whenever you want. Based on
+	  the feature, you can use Ttk widget styling engine on your
+	  old Ruby/Tk application without modifying its source, if you
+	  don't use widget options unsupported on Ttk widgets (At first,
+	  call "Tk.default_widget_set = :Ttk", and next load and run
+	  your application).
+	  This is one step for supporting Tcl/Tk8.5 features.
+
+Wed Feb 27 22:55:42 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_coderange_scan_restartable): coderange scanning
+	  for partial read.
+
+	* io.c (read_all): set coderange when not convert encoding.
+
+Wed Feb 27 03:55:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb, enc/make_encmake.rb: load current mkmf.rb even if
+	  cross-compiling.
+
+	* ext/extmk.rb, enc/make_encmake.rb, lib/mkmf.rb: need to be 1.8
+	  compatible for cross-compiling.
+
+Tue Feb 26 16:53:13 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-calculate-indent): should distinguish
+	  comment and # in strings.  [ruby-dev:33874]
+
+Tue Feb 26 16:41:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (combi_len, rb_ary_product): check for overflow.
+	  [ruby-Bugs-18355]
+
+Tue Feb 26 16:38:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (recursive_cmp): compare minimal length parts.
+
+Tue Feb 26 16:06:00 2008  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_{ec,dh,dsa,rsa}.c: Remove useless warnings.
+
+	* ext/openssl/ossl_asn1.c: Simplify code.
+
+	* ext/openssl/ossl_ssl_session.c Fix compiler warnings.
+	  Undefine #id if SSL_SESSION_get_id is not supported.
+
+Tue Feb 26 15:50:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (value_expr_gen): removed inappropriate warning.
+	  [ruby-core:15660]
+
+Tue Feb 26 15:43:42 2008  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (tokadd_escape): refactored.  [ruby-core:15657]
+
+Tue Feb 26 15:30:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_eql, rb_ary_cmp): get rid of stack overflow with
+	  self-recursive constructs.  [ruby-Bugs-18356]
+
+Tue Feb 26 01:16:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (ROBJECT_NUMIV): renamed from ROBJECT_LEN.
+	  (ROBJECT_IVPTR): renamed from ROBJECT_PTR.
+
+	* variable.c: follow the above renaming.
+
+	* object.c: ditto.
+
+	* gc.c: ditto.
+
+	* marshal.c: ditto.
+
+Mon Feb 25 17:30:29 2008  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* ext/openssl/digest.c ext/openssl/lib/openssl/digest.rb:
+	  Commit patch #9280 from Akinori MUSHA.
+	  Simplify the OpenSSL::Digest class and make use of the
+	  existing Digest framework.
+	  Enhance performance.
+
+Mon Feb 25 15:33:29 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (big2str_karatsuba): initialize cache if not initialized.
+
+	* bignum.c (Init_Bignum): delayed initializing cache.
+	  [ruby-dev:33930]
+
+Mon Feb 25 13:40:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (Init_process): share bignum objects for RLIM_INFINITY,
+	  RLIM_SAVED_MAX and RLIM_SAVED_CUR if they are equal.
+
+Mon Feb 25 10:41:41 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* encoding.c (Encoding#dummy): minor grammatical fixes
+	  in rdoc documentation.
+
+Mon Feb 25 00:01:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (clean-local): should be double-colon.
+
+Sun Feb 24 23:39:59 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, {bcc,win}32/Makefile.sub (clean-local): remove
+	  intermediate files.
+
+	* cygwin/GNUmakefile.in (clean-local): remove def file.
+
+Sun Feb 24 06:49:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* debug.c (ruby_set_debug_option): separated ruby_each_words().
+
+	* util.c (ruby_each_words): extracted from ruby_set_debug_option().
+
+	* ruby.c (enable_option, disable_option): allow all for all known
+	  features.
+
+	* ruby.c (proc_options): generalized enable/disable options.
+
+	* ruby.c (ruby_init_gems): take enabled flag.  [ruby-core:14840]
+
+	* ruby.c (process_options): added --disable-rubyopt flag.
+
+	* include/ruby/util.h (ruby_each_words): prototype.
+
+Sun Feb 24 05:25:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): check if argument for -E exists.
+
+Sun Feb 24 05:09:43 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el (ruby-style-label-indent): fix for labels inside
+	  blocks in switch.
+
+Sun Feb 24 03:52:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (valid_filename): use O_EXCL to get rid of clobbering
+	  existing files in race conditions.
+
+Sat Feb 23 21:36:13 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (ole_init_cp): should return value.
+
+Sat Feb 23 20:16:05 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_sublen): removed.
+
+	* string.c (rb_str_reverse, rb_str_reverse_bang): use
+	  single_byte_optimizable.
+
+Sat Feb 23 19:25:18 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_copy_for_substr): renamed from
+	  rb_enc_cr_str_copy.
+
+	* string.c: use rb_enc_cr_str_copy_for_substr and keep coderange.
+
+Sat Feb 23 18:50:17 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_encoding2cp): remove US-ASCII
+	  mapping.
+
+Sat Feb 23 01:09:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rlimit_resource_type): new function.
+	  (rlimit_resource_value): new function.
+	  (proc_getrlimit): use rlimit_resource_type to accept
+	  symbol and string as resource type.
+	  (proc_setrlimit): use rlimit_resource_type and rlimit_resource_value
+	  to accept symbol and string as resource type and values.
+
+Fri Feb 22 21:12:42 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_copy): check string's coderange is 7bit or
+	  valid.
+
+Fri Feb 22 19:50:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (BIGZEROP): fix for longer Bignum zeros.  [ruby-Bugs-17454]
+
+Fri Feb 22 15:47:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_mbclen): return minlen instead of 1 when
+	  a character is not found properly.
+
+	* string.c (rb_enc_strlen): round up string length with fixed
+	  multibyte encoding such as UTF-32.
+	  (rb_enc_strlen_cr): ditto.
+	  (rb_str_substr): fix substring with fixed multibyte encoding.
+	  (rb_str_justify): check number of characters.
+
+Fri Feb 22 12:11:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): string of ascii incompatible encoding
+	  should be escaped and returned as US-ASCII encoding.
+
+Fri Feb 22 11:16:55 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_substr): copy encoding although empty string.
+
+Fri Feb 22 04:48:22 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_times): empty string's coderange is CODERANGE_7BIT.
+
+	* string.c (rb_str_substr): ditto.
+
+	* encoding.c (rb_enc_compatible): empty string is compatible with not
+	  only nonasciicompatible strings. [ruby-dev:33895]
+
+Thu Feb 21 17:15:15 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c: Added basic support for passing options to String#encode
+	  via a hash. Currently only one option, with one value, is supported:
+	  invalid: :ignore (dropping invalid byte sequences instead of
+	  producing an error). Option naming is not yet stable!
+
+	* test/ruby/test_transcode.rb: Added a single test for invalid: :ignore
+	  option. Not more tests because most data does not yet distinguish
+	  between INVALID and UNKNOWN.
+
+Thu Feb 21 16:35:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_unshift_m): expands enough for argc.  [ruby-dev:33880]
+
+Thu Feb 21 14:49:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_set_encoding): uses current_file after check if next
+	  input is available.
+
+Thu Feb 21 14:13:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_putc): invoke stdout method so that redefining putc
+	  may take effect.   [ruby-talk:291844]
+
+	* io.c (rb_f_puts): ditto.
+
+Thu Feb 21 11:10:49 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c: replace rb_enc_copy by rb_enc_cr_str_copy or
+	  rb_enc_cr_str_exact_copy.
+
+Thu Feb 21 10:35:04 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_enc_asciicompat): dummy encoding is not
+	  ascii compatible. [ruby-dev:33878]
+
+Thu Feb 21 00:01:34 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RPATHFLAG): -R option of HP-UX ld is not for runtime
+	  load path.  [ruby-list:44600]
+
+Wed Feb 20 23:55:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_map_errno): exported.
+
+Wed Feb 20 23:28:43 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/readline/extconf.rb (rl_event_hook): workaround for native
+	  windows.
+
+Wed Feb 20 19:42:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_associate_index): doesn't clear coderange
+	  when new encoding equals to old one.
+
+Wed Feb 20 19:15:38 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_str_copy): added for wrapper for rb_enc_copy.
+	  this also copy coderange when ptr and len is equal.
+
+	* string.c (rb_enc_cr_str_copy): added for wrapper for rb_enc_copy.
+	  this always copy coderange.
+
+	* string.c (str_replace_shared): use rb_enc_str_copy.
+
+	* string.c (str_new3): don't rb_enc_copy because encoding is copied
+	  at str_replace_shared.
+
+Wed Feb 20 13:08:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (parse_args): added --dir-mode, --script-mode and
+	  --cmd-type options.  [ruby-dev:33816]
+
+	* instruby.rb (parse_args): added bin-arch and bin-comm to install
+	  type, for compiled files and script files.
+
+	* instruby.rb (parse_args): deal with make style command line macros,
+	  and count as long style options if prefixed with INSTALL_.
+
+	* instruby.rb (makedirs): use $dir_mode.  [ruby-dev:33805]
+
+	* instruby.rb (open_for_install): set file mode, which is now
+	  permission mode instead of access mode.
+
+	* instruby.rb (bin-comm): installs scripts with replacing shebang
+	  lines.
+
+Wed Feb 20 10:04:22 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (open_key_args): set arg->io even if no options passed.
+	  [ruby-dev:33072]
+
+Tue Feb 19 21:11:49 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_strlen_cr): get length with coderange scan.
+
+	* string.c (str_strlen): use rb_enc_strlen_cr. [ruby-dev:33849]
+
+Tue Feb 19 20:49:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_raise_jump): moved adjustment for control frame.
+
+Tue Feb 19 18:34:32 2008  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (STACK_LENGTH) [SPARC] : 0x80 offset removed.  [ruby-dev:33857]
+
+Tue Feb 19 14:27:32 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_event): prevent polling.  based on
+	  a patch from error errorsson in [ruby-Bugs-17675].
+
+Tue Feb 19 11:14:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_exec_node): no thread starts inside iseq compilation.
+
+	* eval.c (rb_f_raise): skip current control frame.  [ruby-core:15589]
+
+	* insns.def (opt_div): raise as the ordinary method. [ruby-core:15589]
+
+Mon Feb 18 15:16:30 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_each_line): fix newline size.
+
+Mon Feb 18 13:06:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/irb/locale.rb (IRB::Locale#lc2kconv): check ja_JP.EUC-JP as well.
+
+Mon Feb 18 11:51:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (re_warn): defined to restore warnings for /[a-c-e]/, etc.
+
+Mon Feb 18 10:17:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/lib/expect.rb (IO#expect): check if peer is closed.
+	  [ruby-Bugs-17940]
+
+Mon Feb 18 00:33:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_regsub): don't repeat repl twice with
+	  "X".sub!(/./, sprintf("\\%c", 255)).
+
+Sun Feb 17 23:06:55 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/cgi.rb (CGI::escapeHTML): use gsub with Hash. [ruby-dev:33828]
+
+Sun Feb 17 21:38:21 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (ENC_CODERANGE_AND): fix broken case. [ruby-dev:33826]
+
+	* string.c (rb_str_times): fix broken case. [ruby-dev:33826]
+
+Sun Feb 17 20:45:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_prepare_re): add enable_warning parameter.
+	  (rb_reg_adjust_startpos): disable warning by rb_reg_prepare_re.
+	  (rb_reg_search): follow rb_reg_prepare_re parameter change.
+
+Sun Feb 17 20:12:41 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_regexp.rb: add tests to achieve over 90% test
+	  coverage of re.c.
+
+Sun Feb 17 15:25:08 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (ENC_CODERANGE_AND): added.
+
+	* string.c (rb_str_plus, rb_str_times): keep coderange.
+
+	* parse.y (STR_NEW0) use rb_usascii_str_new.
+
+Sun Feb 17 14:07:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (str_strlen): rb_enc_strlen doesn't fail.
+
+Sun Feb 17 13:03:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (str_sublen): use rb_enc_strlen.
+
+Sun Feb 17 12:17:52 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/{euc_jp.c,gbk.c,iso_8859_1.c,iso_8859_11.c,iso_8859_13.c,
+	  iso_8859_2.c,iso_8859_6.c,iso_8859_7.c,iso_8859_8.c,iso_8859_9.c,
+	  shift_jis.c,windows_1251.c}: add document about encodings.
+
+	* enc/cp949.c: divided into new file.
+
+Sun Feb 17 10:59:04 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_quote): return US-ASCII string consistently.
+
+Sun Feb 17 09:17:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_times): reduce loop overhead.
+
+Sun Feb 17 03:37:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/re.h (struct rmatch_offset): new struct for character
+	  offsets.
+	  (struct rmatch): new struct.
+	  (struct RMatch): reference struct rmatch.
+	  (RMATCH_REGS): new macro.
+
+	* re.c (match_alloc): initialize struct rmatch.
+	  (pair_byte_cmp): new function.
+	  (update_char_offset): update character offsets.
+	  (match_init_copy): copy regexp and character offsets.
+	  (match_sublen): removed.
+	  (match_offset): use update_char_offset.
+	  (match_begin): ditto.
+	  (match_end): ditto.
+	  (rb_reg_search): make character offset updated flag false.
+	  (match_size): use RMATCH_REGS.
+	  (match_backref_number): ditto.
+	  (rb_reg_nth_defined): ditto.
+	  (rb_reg_nth_match): ditto.
+	  (rb_reg_match_pre): ditto.
+	  (rb_reg_match_post): ditto.
+	  (rb_reg_match_last): ditto.
+	  (match_array): ditto.
+	  (match_aref): ditto.
+	  (match_values_at): ditto.
+	  (match_inspect): ditto.
+
+	* string.c (rb_str_subpat_set): use RMATCH_REGS.
+	  (rb_str_sub_bang): ditto.
+	  (str_gsub): ditto.
+	  (rb_str_split_m): ditto.
+	  (scan_once): ditto.
+
+	* gc.c (obj_free): free character offsets.
+
+Sun Feb 17 03:13:40 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/resource.rb: made version infos confirm to OS spec.
+
+	* {bcc32,win32}/Makefile.sub (*.rc): add dependency.
+
+Sat Feb 16 20:49:34 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_substr): optimized for UTF-8.
+
+Sat Feb 16 18:13:53 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_compatible): check encoding incapable arguments.
+
+Sat Feb 16 20:12:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (match_inspect): avoid SEGV with MatchData.allocate.inspect.
+
+Sat Feb 16 19:04:17 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_strlen): revert r15507. [ruby-dev:33810]
+
+Sat Feb 16 18:25:14 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_strlen): little more optimization.
+	  (rb_enc_nth): remove needless variable 'c'.
+
+Sat Feb 16 18:00:13 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_compatible): empty strings are always compatible.
+
+	* string.c (rb_enc_cr_str_buf_cat): ditto.
+
+Sat Feb 16 16:14:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_enc_strlen): UTF-8 character count moved to str_strlen.
+	  (str_strlen): UTF-8 character count is only applicable for valid
+	  UTF-8 string.  [ruby-dev:33807]
+
+Sat Feb 16 13:16:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_sub_bang): stringize replacing hash values.
+	  (str_gsub): ditto.
+
+Sat Feb 16 13:01:33 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_strlen): add search_nonascii like character
+	  counter for UTF-8.
+
+Sat Feb 16 11:53:35 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_strlen): moved to string.c.
+
+	* string.c (rb_enc_strlen): use search_nonascii.
+	  (str_strlen): don't use search_nonascii.
+
+Sat Feb 16 11:45:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/require_relative.rb: check require_relative call in eval.
+
+Sat Feb 16 08:00:01 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (process_options): set default_external before loading
+	  libraries. [ruby-dev:33801]
+
+Sat Feb 16 05:49:54 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/iso_8859_{4,13}.c: Windows-1257 is replica of ISO-8859-13.
+
+	* string.c (single_byte_optimizable): rb_enc_mbminlen must be 1
+	  when rb_enc_mbmaxlen is 1.
+
+Sat Feb 16 03:43:18 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_nth): moved to string.c.
+
+	* string.c (rb_enc_nth): moved from string.c.  use search_nonascii
+	  for ASCII compatible string.
+	  (str_nth): wrong optimization removed to fix
+	  "a".force_encoding("EUC-JP").slice!(0,10) returns
+	  "a\x00\x00\x00\x00\x00\x00\x00\x00\x00"
+
+Sat Feb 16 00:21:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (rb_range_beg_len): check if responds to "begin" and "end"
+	  methods for non-Range object.
+
+Fri Feb 15 20:29:42 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_init_cp): initialize WIN32OLE.codepage
+	  according to Encoding.default_external.
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+Fri Feb 15 19:31:23 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/node.h (NODE_FL_NEWLINE): renamed from NODE_NEWLINE
+	  to denote its a flag.  [ruby-core:15529]
+
+Fri Feb 15 18:23:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_sub_bang, str_gsub): allows hash for replacement.
+
+Fri Feb 15 17:12:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_strlen): use search_nonascii() for performance.
+
+	* string.c (str_nth): ditto.
+
+Fri Feb 15 16:22:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (open_key_args): allow specifying both :mode and :encoding.
+
+Fri Feb 15 15:34:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_getbyte): new method.
+	  (rb_str_setbyte): new method.
+
+Fri Feb 15 15:29:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/require_relative.rb: new file.
+
+Fri Feb 15 15:23:12 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_convert): check upper bound.  a patch from
+	  Daniel Luz at [ruby-Bugs-17910].
+
+Fri Feb 15 10:35:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_quote): set US-ASCII for ASCII-only string.
+	  [ruby-dev:33785]
+
+Fri Feb 15 10:27:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {win,bcc}32/Makefile.sub (config.h): added HAVE_FTRUNCATE.
+	  [ruby-dev:33786]
+
+Fri Feb 15 09:44:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_compile_gen): reg_fragment_setenc might not raise an
+	  exception before rb_reg_compile.
+
+Fri Feb 15 07:37:40 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/paths.rb: Preserve compatibility with 1.8.
+
+Fri Feb 15 02:42:25 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ftruncate): check if available.
+
+	* file.c (rb_file_truncate): check if ftruncate instead of truncate.
+
+Fri Feb 15 02:40:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (sigsetmask): check when signal semantics is not POSIX.
+
+	* signal.c (USE_TRAP_MASK): set true if sigprocmask or sigsetmask is
+	  available.
+
+Thu Feb 14 23:56:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.c (error_print): append a newline to rest lines.
+
+	* parse.y (reg_compile_gen): appends error message from
+	  rb_reg_compile() to one from reg_fragment_setenc().
+
+Thu Feb 14 21:00:14 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_reopen): check STDIN, STDOUT and STDERR mode according to
+	  stdio streams.
+
+Thu Feb 14 16:07:40 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_math.rb: actual-expected argument ordering for
+	  test_math.rb fixed.  a patch from Tadashi Saito
+	  <shiba AT mail2.accsnet.ne.jp> in [ruby-dev:33770].
+
+Thu Feb 14 16:02:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_utime): inhibits with secure level 2 or higher.
+
+Thu Feb 14 12:30:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_preprocess_dregexp): use non-preprocessed regexp source
+	  for result.
+
+Thu Feb 14 01:43:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout::timeout): made sensitive to location on the
+	  stack.  [ruby-core:15458]
+
+Thu Feb 14 00:49:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (INSTRUBY_ARGS): pass mode to install.  [ruby-dev:33766]
+
+	* instruby.rb (parse_args): added --data-mode and --prog-mode options.
+
+Thu Feb 14 00:02:19 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* eval.c (eval): allow to eval in a binding that has a singleton method.
+	  [ruby-dev:33763]
+
+	* test/ruby/test_proc.rb: add tests to achieve over 70% test coverage
+	  of proc.c.
+
+	* test/ruby/test_method.rb: ditto.
+
+Wed Feb 13 22:46:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pathname.rb (Pathname#sub_ext): new method.  [ruby-list:44608]
+
+Wed Feb 13 21:50:32 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (proc_curry): new method. [ruby-dev:33676]
+
+	* test/ruby/test_proc.rb: add tests for above.
+
+Wed Feb 13 20:48:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (RObject): add iv_index_tbl for shortcut of
+	  RCLASS_IV_INDEX_TBL(rb_obj_class(obj)).
+	  (ROBJECT_IV_INDEX_TBL): defined.
+
+	* object.c (init_copy): initialize iv_index_tbl in struct RObject.
+
+	* variable.c (ivar_get): use ROBJECT_IV_INDEX_TBL.
+	  (rb_ivar_defined): ditto.
+	  (obj_ivar_each): ditto.
+	  (rb_obj_remove_instance_variable): ditto.
+	  (rb_ivar_set): update iv_index_tbl in struct RObject.
+
+Wed Feb 13 16:21:48 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb: revert r15442. 2nd argument of String#sub parse
+	  escapes. [ruby-dev:33726]
+
+	* bootstraptest/test_method.rb, enc/depend, instruby.rb, lib/mkmf.rb,
+	  mkconfig.rb: revert r15443. ditto.
+
+Wed Feb 13 11:20:26 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/depend: fix typo.
+
+	* lib/mkmf.rb: revert r15443. "\\1#{sep}\\2" is wrong if sep is ended
+	  with "\\".
+
+Wed Feb 13 08:57:21 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup/inline.rb:  Allow inline markup to have a leading
+	  '#' or '\', or trailing punctuation.  i.e. *#freeze?*, *\foo?*.
+
+Wed Feb 13 07:21:23 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/to_html_hyperlink.rb:  Moved linking to to_html.rb, move
+	  crossref to to_html_crossref.rb
+
+Wed Feb 13 04:15:44 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (arg_concat_gen, arg_append_gen): optimize for array push.
+
+	* parse.y (arg_concat_gen): optimize for array concat.
+
+	* parse.y (arg_add_gen): removed since identical to arg_append_gen.
+
+Tue Feb 12 21:04:51 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (exc_list): should use mrhs if non array.
+
+Tue Feb 12 20:32:50 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/rational.rb (floor, ceil, truncate, round): do not use
+	  definitions of Numeric.
+
+	* lib/rational.rb (to_i): should returns truncated self.
+
+	* lib/complex.rb (numerator): requires
+	  Integer#{numerator,denominator}.
+
+	* lib/complex.rb (quo): do not use definition of Numeric.
+
+	* lib/complex.rb (>, >=, <, <=, between?, div, divmod, modulo,
+	  floor, ceil, truncate, round): undef'ed.
+
+	* lib/mathn.rb (Rational#inspect): removed.
+
+Tue Feb 12 16:48:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (args, mrhs): flattens literal array splats.
+
+	* parse.y (exc_list): splat literal array.
+
+Tue Feb 12 15:27:19 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bootstraptest/runner.rb, bootstraptest/test_method.rb, enc/depend,
+	  instruby.rb, lib/mkmf.rb, lib/test/unit/util/procwrapper.rb,
+	  mkconfig.rb, sample/test.rb, template/vm.inc.tmpl,
+	  test/ruby/test_stringchar.rb: fixes around String#gsub.
+
+Tue Feb 12 15:11:47 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/lib/json/pure/generator.rb,
+	  ext/json/lib/json/pure/parser.rb, ext/openssl/lib/openssl/x509.rb,
+	  ext/win32ole/sample/olegen.rb, lib/date/format.rb, lib/irb/context.rb,
+	  lib/irb/workspace.rb, lib/net/http.rb, lib/net/imap.rb,
+	  lib/rdoc/generator.rb, lib/rdoc/markup/to_html.rb,
+	  lib/rdoc/markup/to_latex.rb, lib/rdoc/parsers/parse_c.rb,
+	  lib/rdoc/ri/formatter.rb, lib/rexml/parsers/baseparser.rb,
+	  lib/rexml/quickpath.rb, lib/rexml/text.rb, lib/rss/parser.rb,
+	  lib/uri/common.rb, lib/uri/generic.rb, lib/webrick/httpresponse.rb,
+	  lib/webrick/httpservlet/filehandler.rb, lib/yaml/baseemitter.rb,
+	  lib/yaml/encoding.rb: performance tuning around String#gsub.
+
+Tue Feb 12 12:16:45 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_hash_cmp): lighter version of rb_str_cmp() for
+	  hash comparison function.
+
+	* hash.c (rb_any_cmp): use rb_str_hash_cmp().
+
+	* string.c (rb_str_casecmp): should return nil for incompatible
+	  comparison.
+
+Tue Feb 12 12:13:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* instruby.rb: specify file mode to install.  a patch from
+	  pegacorn  <subscriber.jp AT gmail.com> in [ruby-dev:33699].
+
+Tue Feb 12 11:38:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (rb_num_coerce_bin): add ID argument to specify
+	  caller's method name.  [ruby-dev:33663]
+
+	* numeric.c (rb_num_coerce_cmp): ditto.
+
+	* numeric.c (rb_num_coerce_relop): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (DoSomeOne): add function name argument.
+
+Tue Feb 12 10:25:02 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/rdoc.rb:  Wrap parse_files' read in version check for
+	  backwards compatibility.
+
+Tue Feb 12 10:15:14 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (load_file): enc must effect source encoding.
+	  [ruby-core:15496]
+
+Tue Feb 12 10:16:47 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/paths.rb:  Restore require rubygems check.
+
+Tue Feb 12 02:42:27 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_include): specialize single character string
+	  case (e.g. (?a ..?z).include(?x)) for performance.
+	  [ruby-core:15481]
+
+	* string.c (rb_str_upto): specialize single character case.
+
+	* string.c (rb_str_hash): omit coderange scan for performance.
+
+	* object.c (rb_check_to_integer): check Fixnum first.
+
+	* object.c (rb_to_integer): ditto.
+
+	* string.c (rb_str_equal): inline memcmp to avoid unnecessary
+	  rb_str_comparable().
+
+	* parse.y (rb_intern2): use US-ASCII encoding.
+
+	* parse.y (rb_intern_str): ditto.
+
+Mon Feb 11 17:21:18 2008  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::VERSION), test/rss/test_version.rb:
+	  0.2.3 -> 0.2.4.
+
+	* lib/rss/maker.rb, lib/rss/maker/, test/rss/test_maker_2.0.rb:
+	  fixed a bug that RSS::Maker.make("0.9")'s item doesn't make some
+	  elements if description is missed.
+	  Reported by Michael Auzenne. Thanks!!!
+
+	* lib/rss/maker/0.9.rb, test/rss/test_maker_0.9.rb:
+	  RSS::Maker.make("0.9") generates RSS 0.92 not RSS 0.91.
+
+Mon Feb 11 10:43:31 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (load_file): the encoding of DATA follows the source
+	  file encoding. [ruby-dev:33693]
+
+Mon Feb 11 06:50:42 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_pack.rb: fix tests for 64bit CPU.
+
+	* test/ruby/test_bignum.rb: ditto.
+
+	* test/ruby/test_file_exhaustive.rb: ditto.
+
+	* test/ruby/test_integer.rb: ditto.
+
+	* test/ruby/test_time.rb: ditto.
+
+	* test/ruby/test_numeric.rb: ditto.
+
+	* test/ruby/test_fixnum.rb: ditto.
+
+Mon Feb 11 00:18:57 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/benchmark.rb (Job::Benchmark#item): fix typo.
+
+Sun Feb 10 21:58:32 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (encdb, transdb): depend on $(PREP).
+
+Sun Feb 10 16:58:20 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*, test/rubygems*, gem_prelude.rb: Import RubyGems
+	  r1601.  [ruby-core:15381]
+
+Sun Feb 10 15:07:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {bcc32,win32,wince}/Makefile.sub (MISSING): added cbrt.obj.
+
+Sun Feb 10 12:58:33 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/code_objects.rb: Make some attributes accessible for reuse.
+	* lib/rdoc/generator/html.rb: Pull out ContextUser classes and related
+	  methods for reuse.
+	* lib/rdoc/generator.rb: Move ContextUser classes to
+	  RDoc::Generator::Context for reuse.
+	* lib/rdoc/rdoc.rb: Make RDoc::RDoc initialization a little easier.
+	* lib/rdoc/options.rb: Make RDoc::Options easier to use without
+	  parsing an ARGV.
+	* lib/rdoc/markup/to_*.rb: Subclass RDoc::Markup::Formatter.
+	* lib/rdoc/markup/formatter.rb: Add RDoc::Markup::Formatter to make
+	  RDoc markup conversion easier.
+	* lib/rdoc/markup/fragments.rb: Make RDoc::Markup::ListItem easier to
+	  test.
+	* lib/rdoc/markup/to_html_hyperlink.rb: Pulled out of the HTML
+	  generator for easier reusability.
+	* lib/rdoc/markup.rb: Fix bug with labeled lists containing bullet
+	  lists.
+	* lib/rdoc/generators/html/html.rb: Fix Constant display.
+
+Sat Feb  9 23:44:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* missing/tgamma.c (tgamma): use lgamma_r if available.
+
+Sat Feb  9 23:22:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/extconf.rb: simplified the condition.
+
+Sat Feb  9 21:20:28 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_math.rb: add tests for Math.gamma, Math.lgamma and
+	  Math.cbrt, and use assert_in_delta instead of assert.
+
+Sat Feb  9 18:34:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* math.c (math_cbrt): new method Math.cbrt.
+
+	* configure.in (cbrt): check for replacement functions.
+
+	* missing/cbrt.c: new file.
+
+Sat Feb  9 17:51:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): use strtod() for more
+	  precision.  [ruby-talk:290296]
+
+	* ext/bigdecimal/bigdecimal.c (BASE_FIG): made constant.
+
+	* ext/bigdecimal/extconf.rb: ditto.  [ruby-dev:33658]
+
+Sat Feb  9 12:06:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* missing/tgamma.c (tgamma): add error check.
+
+Sat Feb  9 11:47:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* math.c (math_gamma): add error check.
+	  (math_lgamma): ditto.
+
+Sat Feb  9 11:09:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* missing/lgamma_r.c (lgamma_r): return HUGE_VAL for non-positive
+	  integers.
+
+Sat Feb  9 10:03:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_new4): copy encoding from orig, instead of shared
+	  one.
+
+Sat Feb  9 01:01:38 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (lchmod_internal): fix warning cast from pointer to integer of
+	  different size.
+
+Sat Feb  9 00:44:52 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb.rb (IRB::Irb::eval_input): rescues Interrupt and other than
+	  SystemExit and SignalException.  [ruby-core:15359]
+
+Fri Feb  8 23:51:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* missing/lgamma_r.c (lgamma_r): use smaller argument for sin function.
+
+Fri Feb  8 22:10:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI.open_http): rescue URI::InvalidURIError by
+	  URI.parse for location URI.
+
+Fri Feb  8 19:22:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (rb_str_derive): uses rb_str_subseq() for byte
+	  length.  [ruby-dev:33653]
+
+	* ext/iconv/iconv.c (iconv_convert): added toidx argument to set
+	  encoding of successfully converted string.  [ruby-dev:33221]
+
+Fri Feb  8 15:09:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (xsystem): expand macros like as make.
+
+Fri Feb  8 09:27:57 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/ri/driver.rb (read_yaml): remove SM* for compatibility.
+
+Fri Feb  8 00:07:24 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_hash.rb: follow the change of Hash#flatten.
+
+	* test/ruby/test_time.rb: add tests to achieve over 70% test coverage
+	  of time.c.
+
+	* test/ruby/test_prec.rb: ditto over 90% for prec.c.
+
+Thu Feb  7 19:11:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_dup): reverted unneeded change.  [ruby-dev:33634]
+
+	* string.c (rb_str_replace): makes frozen shared string before
+	  sharing.
+
+Thu Feb  7 16:33:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_reopen): don't change access mode for stdin, stdout and
+	  stderr.  [ruby-core:15360]
+
+Thu Feb  7 16:33:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_replace_shared): replaces string with sharing.
+
+	* string.c (rb_str_new4, rb_str_associate, rb_str_associated): allows
+	  associated strings shared.
+
+	* string.c (rb_str_dup, rb_str_substr, rb_str_replace): shares memory.
+	  [ruby-core:15400]
+
+Thu Feb  7 15:42:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_end_with): compares with the suffix.
+
+Thu Feb  7 15:03:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/korean.c: add support for CP949 by Park Ji-In.
+	  [ruby-dev:33626]
+
+Thu Feb  7 11:11:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* missing/lgamma_r.c (lgamma_r): some compilers don't permit dividing
+	  by literal 0.0. use const variable instead.
+
+	* {bcc32,win32,wince}/Makefile.sub (MISSING): add lgamma_r.obj and
+	  tgamma.obj.
+
+Thu Feb  7 10:39:21 2008  Tanaka Akira  <akr@fsij.org>
+
+	* math.c (math_gamma): new method Math.gamma.
+	  (math_lgamma): new method Math.lgamma.
+
+	* include/ruby/missing.h (tgamma): declared unless HAVE_TGAMMA.
+	  (lgamma_r): declared unless HAVE_LGAMMA_R.
+
+	* configure.in (tgamma): check for replacement functions.
+	  (lgamma_r): ditto.
+
+	* missing/tgamma.c: new file.  based on gamma.c from
+	  "C-gengo niyoru saishin algorithm jiten" (New Algorithm handbook
+	  in C language) (Gijyutsu hyouron sha, Tokyo, 1991)
+	  by Haruhiko Okumura.
+
+	* missing/lgamma_r.c: ditto.
+
+	* LEGAL (missing/tgamma.c): describe as public domain.
+	  (missing/lgamma_r.c): ditto.
+
+Thu Feb  7 09:05:57 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_enc_from_index): BINARY does not
+	  have in-bound encoding index.
+
+Thu Feb  7 04:26:28 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/korean.c: add EUC-KR conversion support by Park Ji-In.
+	  [ruby-dev:33621]
+
+Wed Feb  6 01:47:39 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_flatten): do not flatten recursively by default.
+	  [ruby-dev:33603]
+
+Wed Feb  6 00:50:19 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* insns.def (adjuststack): never use INC_SP with minus value because
+	  some compilers cannot deal it correctly. use DEC_SP instead.
+
+Wed Feb  6 00:48:41 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_hash.rb: add tests to achieve over 90% test coverage
+	  of hash.c.
+
+	* test/ruby/test_env.rb: ditto.
+
+Wed Feb  6 00:24:49 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* hash.c (env_rassoc): remove access to free'd environment on mswin32.
+
+Tue Feb  5 21:57:34 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm.c (rb_call_super): pass a passed block when super is called via
+	  rb_call_super.  [ruby-dev:33598]
+
+Tue Feb  5 11:14:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (INSTALL_DIRS, install_dirs): added BINDIR.
+
+	* lib/mkmf.rb (install_files): rejects files matching to
+	  $NONINSTALLFILES.
+
+	* lib/mkmf.rb (init_mkmf): defaults $NONINSTALLFILES to backup and
+	  temporary files.
+
+Mon Feb  4 21:52:06 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (DelegateClass): use define_method instead of
+	  module_eval to improve performance.  [ruby-dev:33586]
+
+Mon Feb  4 16:44:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (darwin): NSIG is not defined if _XOPEN_SOURCE > 500L.
+	  [ruby-dev:33584]
+
+Mon Feb  4 14:51:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_enc_symname2_p): support "!", "!=" and "!~".
+	  [ruby-dev:33592]
+
+Mon Feb  4 13:58:42 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator.preserved, DelegateClass.methods): extend
+	  shouldn't be delegated. [ruby-dev:32987], etc.
+
+Mon Feb  4 08:59:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::[]): no more transition
+	  extend(CGI::Value).  a patch from <tommy AT tmtm.org> in
+	  [ruby-dev:33583].
+
+Sun Feb  3 21:13:13 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_numeric.rb: forgot to add this (at r15360).
+
+	* test/ruby/test_file_exhaustive.rb: add tests to achieve over 80% test
+	  coverage of file.c.
+
+Sat Feb  2 20:06:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/benchmark.rb (Benchmark::realtime): make Benchmark#realtime
+	  a bit faster.  a patch from Alexander Dymo <dymo AT ukrpost.ua> in
+	  [ruby-core:15337].
+
+Sat Feb  2 17:40:21 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (time_cmp): Time.<=> no longer supports comparison with
+	  numeric. [ruby-core:15332]
+
+Sat Feb  2 09:53:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (darwin): disabled fat-binary support which confuses
+	  configure much, since ``universal'' implies hidden cross-compiling.
+	  TODO: ruby and libruby.bundle might be possible to bound with `lipo'
+	  after builds for each archs.  Anyway, config.h and rbconfig.rb must
+	  be separated definitely at least.
+
+Sat Feb  2 09:28:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (limited_big_rand): fix buffer overflow when SIZEOF_BDIGITS
+	  is 2.  fixed by Kenta Murata.  [ruby-dev:33565]
+
+Fri Feb  1 21:42:37 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (darwin): _XOPEN_SOURCE is necessary to make ucontext_t
+	  consistent with the library implementation of MacOS X 10.5.
+	  [ruby-dev:33461]
+
+	* configure.in (darwin): ucontext on PowerPC MacOS X 10.5 is broken.
+
+Fri Feb  1 11:44:22 2008  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb (C_ESC): use octal escape to avoid
+	  "\x09for (;;) ..." to be interpret the first character 0x9f.
+
+Thu Jan 31 23:06:42 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_bignum.rb: suppress warnings during test.
+
+	* test/ruby/test_enum.rb: ditto.
+
+	* test/ruby/test_integer.rb: add tests to achieve over 90% test
+	  coverage of numeric.c.
+
+	* test/ruby/test_float.rb: ditto.
+
+	* test/ruby/test_fixnum.rb: ditto.
+
+	* test/ruby/test_numeric.rb: ditto.
+
+	* test/ruby/test_pack.rb: add tests to achieve over 90% test coverage
+	  of pack.c.
+
+Thu Jan 31 17:30:42 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* marshal.c (r_object0): no need to call r_entry for immediate values.
+
+Thu Jan 31 15:46:30 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/ri/formatter.rb (output): add accessor.
+
+	* lib/rdoc/ri/display.rb (page): replace @formatter.output instead of
+	  $stdout.
+
+Thu Jan 31 15:06:50 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* marshal.c (r_object0): call r_entry/r_leave to call proc when
+	  TYPE_FIXNUM, TYPE_NIL, TYPE_TRUE, TYPE_FALSE, TYPE_SYMBOL.
+
+Thu Jan 31 14:03:38 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/ri/display.rb (display_method_list, display_class_list):
+	  use @formatter.raw_print_line instead of puts.
+
+	* lib/rdoc/ri/driver.rb (select_methods): new method to collect all
+	  instance/class methods which match with passed pattern.
+
+	* lib/rdoc/ri/driver.rb (run): use class_cache's result directly
+	  instead of select_classes' because it's removed now.
+
+	* lib/rdoc/ri/driver.rb (run): search methods when passed name is not
+	  class name. [ruby-core:15309]
+
+Thu Jan 31 08:31:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ext/extmk.rb, instruby.rb): inlined $(MAKE) so that can
+	  be executed even with -n.
+
+Thu Jan 31 06:24:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_close_read): replaces fptr with the tied writer if
+	  duplex.
+
+	* io.c (rb_io_close_write): unties the tied IO for writing if duplex.
+	  [ruby-dev:33532]
+
+Thu Jan 31 02:22:04 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (open_key_args): allow encoding key to take two encoding
+	  names.  a patch from <rubikitch AT ruby-lang.org>.  [ruby-dev:33540]
+
+Thu Jan 31 02:15:49 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (dsym): allow empty symbols.  [ruby-core:15248]
+
+Thu Jan 31 00:01:51 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (select_internal): fix SEGV by `select [STDIN],nil,[STDIN]'.
+	  fixed by Petr Chromec.
+	  http://rubyforge.org/tracker/index.php?func=detail&aid=17275&group_id=426&atid=1698
+
+Wed Jan 30 17:32:49 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/*.c: add GB12345, UCS-{2,4}{BE,LE}.
+
+Wed Jan 30 14:32:18 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/ri/driver.rb (cache_file_for): shouldn't use `:' in filename.
+
+Wed Jan 30 14:27:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_succ): use wrapped character as a carry for
+	  ASCII incompatible encoding.
+
+Wed Jan 30 12:26:59 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/utf_16be.c (UTF16_IS_SURROGATE_FIRST): avoid branch.
+	  (UTF16_IS_SURROGATE_SECOND): ditto.
+	  (UTF16_IS_SURROGATE): defined.
+	  (utf16be_mbc_enc_len): validation implemented.
+
+	* enc/utf_16le.c (UTF16_IS_SURROGATE_FIRST): avoid branch.
+	  (UTF16_IS_SURROGATE_SECOND): ditto.
+	  (UTF16_IS_SURROGATE): defined.
+	  (utf16le_mbc_enc_len): validation implemented.
+
+Wed Jan 30 12:06:43 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* bignum.c (rb_cstr_to_inum): '0_2' is a valid representation.
+
+Wed Jan 30 11:57:50 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bootstraptest/runner.rb: fix -I../../hoge case.
+
+Wed Jan 30 01:25:16 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_range.rb: add tests to achieve over 90% test coverage
+	  of range.c.
+
+Wed Jan 30 00:09:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/euc_tw.c (euctw_mbc_enc_len): validation implemented.
+
+Tue Jan 29 22:58:11 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_enumerator.rb: add tests to achieve over 90% test
+	  coverage of enumerator.c.
+
+	* test/ruby/test_enum.rb: add for enum.c.
+
+Tue Jan 29 22:29:48 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enumerator.c: fix documents.
+
+Tue Jan 29 22:27:11 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* range.c: fix SEGV by ("a" .. "z").step(2 ** 30) { }.
+
+Tue Jan 29 21:59:16 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/euc_tw.c (euctw_islead): 0x8e is a leading byte.
+
+Tue Jan 29 21:55:35 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c: move object allocation out of blocking_region.
+	  [ruby-dev:33139]
+
+Tue Jan 29 20:37:36 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/make_transdb.rb: add for make transdb.h.
+
+	* dmytranscode.c: add for miniruby.
+
+	* enc/gbk.c (gbk_left_adjust_char_head, gbk_is_allowed_reverse_match):
+	  fix odd regexp match. [ruby-dev:33502]
+
+Tue Jan 29 20:17:06 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32}/Makefile.sub (MINIOBJS): add dmytranscode.$(OBJEXT).
+
+Tue Jan 29 19:39:40 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in, common.mk: fix rule for dmytranscode.o.
+
+Tue Jan 29 19:03:16 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/japanese.c (rb_to_Windows_31J): to 'Windows-31J'.
+
+	* common.mk: add rules for transdb.h.
+
+	* transcode.c (init_transcoder_table): use transdb.h.
+
+Tue Jan 29 18:05:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (encdb_{replicate,alias,dummy,declare}): define only if
+	  NO_ENCDB_H is not defined.
+
+Tue Jan 29 17:54:11 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/gbk.c (EncLen_gbk): too short. [ruby-dev:33497]
+
+Tue Jan 29 17:25:22 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dmyencoding.c, encoding.c (enc_init_db, NO_ENCDB_H):
+	  miniruby doesn't use encdb.
+
+	* common.mk: encdb.h use miniruby.
+
+Tue Jan 29 17:37:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/gb18030.c (gb18030_mbc_enc_len): validation implemented.
+
+Tue Jan 29 17:01:07 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/ifchange: remove $temp when unchanged.
+
+Tue Jan 29 16:59:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* insns.def (toregexp): generate a regexp from strings instead of one
+	  string.
+
+	* re.c (rb_reg_new_ary): defined for toregexp.  it concatenates
+	  strings after each string is preprocessed.
+
+	* compile.c (compile_dstr_fragments): split from compile_dstr.
+	  (compile_dstr): call compile_dstr_fragments.
+	  (compile_dregx): defined for dynamic regexp.
+	  (iseq_compile_each): use compile_dregx for dynamic regexp.
+
+	  [ruby-dev:33400]
+
+Tue Jan 29 16:25:26 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk, ext/extmk.rb: always make encdb.h.
+
+Tue Jan 29 12:53:39 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/gbk.c: add GBK, CP936 and CP949.
+
+	* enc/euc_kr.c: remove CP949.
+
+	* enc/euc_cn.c: remove CP936 and rename to gb2312.c
+
+	* enc/gb2312.c: GB2312 is preferred MIME name.
+
+Tue Jan 29 03:01:29 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (reg_fragment_setenc_gen): US-ASCII script special code.
+
+	* parse.y (reg_fragment_check_len, reg_compile_gen): no need such
+	  trick.
+	  [ruby-dev:33399]
+
+	* test/ruby/test_m17n.rb (test_regexp_usacii_literal): add tests.
+
+Tue Jan 29 01:38:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk ($(srcdir)/revision.h): no need to show ifchange execution
+	  because ifchange echos updated or unchanged.
+
+Tue Jan 29 01:26:23 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (up): use last changed revision.
+
+	* common.mk (up): force to update revision.h.
+
+Tue Jan 29 00:12:17 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): should be US-ASCII.
+
+Tue Jan 29 00:10:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode-set-encoding): updates magic comment.
+
+Mon Jan 28 23:47:52 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (rb_id2str, ripper_initialize, Init_ripper):
+	  use rb_usascii_str_new2. [ruby-dev:33449]
+
+Mon Jan 28 19:37:08 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (ole_cp2encoding): new function.
+
+	* ext/win32ole/win32ole.c (ole_wc2vstr, ole_variant2val, fole_missing):
+	  set encoding to result.
+
+	* ext/win32ole/win32ole.c (fole_s_set_code_page, Init_win32ole): set
+	  default encoding.
+	  [ruby-dev:33433]
+
+Mon Jan 28 11:17:28 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c, parse.y, re.c: use rb_ascii8bit_encoding.
+
+Mon Jan 28 17:54:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/utf_7.h: add dummy encoding UTF-7 and its alias CP65000.
+
+Mon Jan 28 17:41:19 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/utf_8.c: add alias CP65001.
+
+Mon Jan 28 15:33:23 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/big5.c (big5_mbc_enc_len): validation implemented.
+
+Mon Jan 28 13:02:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/euc_kr.c (euckr_mbc_enc_len): validation implemented.
+
+Mon Jan 28 11:24:49 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (parser_str_new): encoding of UTF-8 literal string in
+	  US-ASCII script is UTF-8. [ruby-dev:33406]
+
+Mon Jan 28 10:25:59 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_m17n.rb (test_magic_comment): add test.
+
+Mon Jan 28 09:34:54 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (help): use double quotes for nmake.
+
+Mon Jan 28 00:39:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_set_encode): check if encoding is ASCII compatible.
+
+Mon Jan 28 01:21:15 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_open_file): should check NUL in path.
+	  <http://www.rubyist.net/~matz/20080125.html#c01>.
+
+	* io.c (rb_io_s_popen): ditto.
+
+	* io.c (rb_io_reopen): ditto.
+
+	* io.c (next_argv): ditto.
+
+Sun Jan 27 23:33:35 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): fix for left justify flag.
+
+	* sprintf.c (rb_str_format): zero-precision zero bug revised.
+	  [ruby-dev:33419]
+
+Sun Jan 27 23:20:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h: precise mbclen API redesigned to avoid
+	  inline functions.
+	  (onigenc_mbclen_charfound): removed.
+	  (onigenc_mbclen_needmore): removed.
+	  (onigenc_mbclen_recover): removed.
+	  (ONIGENC_MBCLEN_CHARFOUND): removed.
+	  (ONIGENC_MBCLEN_CHARFOUND_P): defined.
+	  (ONIGENC_MBCLEN_CHARFOUND_LEN): defined.
+	  (ONIGENC_MBCLEN_INVALID): removed.
+	  (ONIGENC_MBCLEN_INVALID_P): defined.
+	  (ONIGENC_MBCLEN_NEEDMORE): removed.
+	  (ONIGENC_MBCLEN_NEEDMORE_P): defined.
+	  (ONIGENC_MBCLEN_NEEDMORE_LEN): defined.
+	  (ONIGENC_MBC_ENC_LEN): use onigenc_mbclen_approximate.
+
+	* regenc.c (onigenc_mbclen_approximate): defined.
+
+	* include/ruby/encoding.h (MBCLEN_CHARFOUND): removed.
+	  (MBCLEN_INVALID): removed.
+	  (MBCLEN_NEEDMORE): removed.
+	  (MBCLEN_CHARFOUND_P): defined.
+	  (MBCLEN_INVALID_P): defined.
+	  (MBCLEN_NEEDMORE_P): defined.
+	  (MBCLEN_CHARFOUND_LEN): defined.
+	  (MBCLEN_NEEDMORE_LEN): defined.
+
+	* encoding.c: use new API.
+
+	* re.c: ditto.
+
+	* string.c: ditto.
+
+	* parse.y: ditto.
+
+Sun Jan 27 22:55:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (value_expr_gen): reverted r12880.  [ruby-dev:33388]
+
+Sun Jan 27 22:33:39 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): fix for octal with precision.
+	  [ruby-dev:33411]
+
+Sun Jan 27 22:31:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-mode-set-encoding): automatically insert
+	  encoding magic comment.
+
+	* misc/ruby-mode.el (ruby-mode): set ruby-mode-set-encoding to buffer
+	  local before-save-hook.
+
+Sun Jan 27 19:51:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_inspect): avoid exception by
+	  "\#\xa1".force_encoding("euc-jp").inspect.
+
+Sun Jan 27 19:07:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_succ): warning suppressed.
+
+Sun Jan 27 18:18:13 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (help): show major targets.
+
+Sun Jan 27 17:54:48 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c: raise error when no output encoding is given.
+
+Sun Jan 27 17:20:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_succ): don't increment/decrement codepoint.
+
+Sun Jan 27 16:03:42 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (RubyLex#buf_input): use chars.to_a.
+
+Sun Jan 27 16:27:22 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8: update nkf.
+
+Sun Jan 27 16:25:27 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_source): set encoding as regexp encoding.
+
+Sun Jan 27 05:56:39 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_preprocess): force fixed encoding when ASCII
+	  incompatible source string.
+
+Sat Jan 26 23:46:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): zero-precision zero should be empty.
+	  [ruby-dev:33363]
+
+	* sprintf.c (rb_str_format): not prepend octal prefix to negative or
+	  zero value.  [ruby-dev:33363], [ruby-dev:33367]
+
+Sat Jan 26 23:42:15 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assignable_gen, keyword_to_name): __ENCODING__ was missing.
+
+Sat Jan 26 19:08:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (w_object): dump string encoding in USERDEF.
+	  [ruby-dev:33401]
+
+Sat Jan 26 17:42:23 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): validate argument expr of "next"
+	  statement.
+
+	* bootstraptest/test_syntax.rb: add a test.
+
+Sat Jan 26 17:22:46 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, compile.h: fix to calculate correct stack depth
+	   at each instruction.
+
+Sat Jan 26 09:41:02 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rexml/doctype.rb, test/rss/test_maker_itunes.rb: replace
+	  multi-byte string.
+
+	* test/json/{test_json.rb, test_json_unicode.rb}:
+	  add magic comment.
+
+Sat Jan 26 09:30:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_usascii_encindex): added prototype.
+
+	* include/ruby/intern.h (rb_usascii_str_new, rb_usascii_str_new2):
+	  ditto.
+
+Sat Jan 26 09:17:13 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_new): set US-ASCII and ENC_CODERANGE_7BIT when
+	  empty string (len == 0).
+
+Sat Jan 26 03:41:53 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (parser_initialize): set default script encoding as US-ASCII.
+
+	* ruby.c (load_file): ditto.
+
+	* ruby.c (process_options): set script encoding of -e from locale
+	  except when -K is specified.
+
+	* ruby.c (load_file): set script encoding of stdin from locale except
+	  when -K is specified. [ruby-dev:33375]
+
+Sat Jan 26 02:51:06 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, compile.h: fix stack pointer issues.
+	  calculate correct stack depth at compile time.
+
+	* insns.def (emptstack): remove it and add a new insn "adjuststack".
+
+	* bootstraptest/test_knownbug.rb: move/remove fixed test.
+
+	* bootstraptest/test_syntax.rb: ditto.
+
+Sat Jan 26 00:17:18 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_usascii_new{,2}: defined.
+	  (rb_str_new): set US-ASCII and ENC_CODERANGE_7BIT when empty
+	  string.
+
+	* encoding.c (rb_usascii_encoding, rb_usascii_encindex): defined.
+	  (rb_enc_inspect, enc_name, rb_locale_charmap, rb_enc_name_list_i):
+	  use rb_str_ascii_new.
+
+	* array.c (recursive_join, inspect_ary): ditto.
+
+	* object.c (nil_to_s, nil_inspect, true_to_s, false_to_s,
+	  rb_mod_to_s): ditto.
+
+	* hash.c (inspect_hash, rb_hash_inspect, rb_f_getenv, env_fetch,
+	  env_clear, env_to_s, env_inspect): ditto.
+
+	* numeric.c (flo_to_s, int_chr, rb_fix2str): ditto.
+
+	* bignum.c (rb_big2str): ditto.
+
+	* file.c (rb_file_ftype, rb_file_s_dirname, rb_file_s_extname,
+	  file_inspect_join, Init_file): ditto.
+
+	* test/ruby/test_ruby_m17n.rb: add checks for encoding of string.
+
+Sat Jan 26 01:35:46 2008  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (r_byte): use getbyte instead of getc.
+	  (marshal_load): ditto.
+	  [ruby-dev:33264]
+
+Sat Jan 26 00:43:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_getline_fast): don't care ASCII incompatible encoding.
+	  (prepare_getline_args): generate a newline according to IO encoding
+	  when necessary.
+	  (rb_io_getline_1): call rb_io_getline_fast only for ASCII
+	  compatible encoding.
+
+Fri Jan 25 21:49:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_buf_cat_ascii): use rb_enc_cr_str_buf_cat.
+
+Fri Jan 25 19:38:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (version.$(OBJEXT)): depends on $(srcdir)/revision.h.
+
+	* common.mk (revision.h): extracts revision number with ``svn info''.
+
+	* common.mk (up): target to update from the repository.
+
+	* Makefile.in, {win,bcc}32/Makefile.sub (IFCHANGE): tool to update a
+	  file if changed.
+
+	* tool/ifchange: for unixen.
+
+	* win32/ifchange.bat: some fix
+
+Fri Jan 25 17:12:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file): set default to ASCII-8BIT explicitly if -K
+	  option is not given.
+
+Fri Jan 25 16:31:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_str_buf_cat_ascii): declared.
+
+	* string.c (rb_str_buf_cat_ascii): defined.
+
+	* re.c (rb_reg_s_union): use rb_str_buf_cat_ascii to support ASCII
+	  incompatible encoding.
+
+Fri Jan 25 16:11:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options, load_file, rb_load_file): propagates script
+	  encoding by -K to libraries.  [ruby-dev:33156]
+
+Fri Jan 25 15:56:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (cmdline_arguments): split argc and argv from cmdline_options.
+
+	* ruby.c (process_options): not set encoding of -e option from -E
+	  option if they are not compatible.
+
+Fri Jan 25 13:15:23 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (proc_options, process_options, load_file): shouldn't effect
+	  --encoding to script encoding. [ruby-dev:33169]
+
+Fri Jan 25 10:31:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* */*.bat: set svn:mime-type to text/batch.
+
+Thu Jan 24 23:23:06 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* enum.c (enum_one, enum_take_while, enum_drop_while): fix documents.
+
+Thu Jan 24 21:46:24 2008  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (reg_fragment_setenc_gen): associate ASCII-8BIT only if
+	  str has only ASCII characters.
+
+Thu Jan 24 20:46:17 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_m17n.rb: follow to the following changes.
+
+Thu Jan 24 20:21:07 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parser.y (parser_str_new): automatically update string literal's
+	  encoding from US-ASCII to ASCII-8BIT when script encoding is US-ASCII
+	  and the string includes non-ascii bytes. [ruby-dev:33348]
+
+	* parser.y (reg_fragment_check_gen, reg_compile_gen): automatically
+	  update regexp literal's encoding from US-ASCII to ASCII-8BIT when
+	  script encoding is US-ASCII, the regexp has no kcode option and the
+	  regexp includes non-ascii bytes. [ruby-dev:33353]
+
+Thu Jan 24 19:36:22 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic::inspect): use Kernel#to_s instead
+	  object_id with printf.  [ruby-dev:33347]
+
+Thu Jan 24 19:29:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (remove_sign_bits): returns pointer to the first char to
+	  be used, instead of copying.
+
+	* sprintf.c (rb_str_format): negative indicator dots should come
+	  before sign digits always.  [ruby-dev:33224]
+
+Thu Jan 24 18:19:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_enc_is_newline): parenthesized arguments.
+
+Thu Jan 24 18:14:14 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* re.c (rb_reg_fixed_encoding_p): no need to treat ASCII-8BIT specially.
+
+Thu Jan 24 16:53:06 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* re.c (rb_reg_initialize): 7bit clean regexp should be US-ASCII.
+	  [ruby-dev:33346]
+
+Thu Jan 24 16:31:04 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_getline_fast): the end point of left_char_head()
+	  must be the last character.  [ruby-cvs:22445]
+
+Thu Jan 24 16:24:25 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (reg_fragment_setenc_gen): recognize regexp with option n as
+	  as ASCII-8BIT instead of US-ASCII. [ruby-dev:33339]
+
+Thu Jan 24 15:44:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (collect_bang_i): use rb_ary_store() to avoid potential
+	  memory corruption.  a patch from Yusuke Endoh <mame@tsg.ne.jp>
+	  in [ruby-dev:33328].
+
+	* array.c (ITERATE): remove unnecessary macro.
+
+	* array.c (sort_1): remove ary_sort_check().  in-place sort keep
+	  original elements even when it's modified.
+
+	* array.c (sort_2): ditto.
+
+Thu Jan 24 15:09:40 2008  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (make_time_t): revert round trip test.  [ruby-dev:33058]
+
+Thu Jan 24 11:14:56 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_enc_cr_str_buf_cat): ASCII incompatible encoding is
+	  not compatible with any other encoding.
+
+Thu Jan 24 07:34:34 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (STR_NEW0): set encoding as US-ASCII.
+
+Thu Jan 24 03:47:41 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rexml/text.rb, lib/rubygems/open-uri.rb, lib/open-uri.rb,
+	  test/logger/test_logger.rb, test/ruby/test_regexp.rb:
+	  fix tests. [ruby-dev:33336]
+
+Thu Jan 24 03:23:44 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_each_line): use memchr(3) for faster newline
+	  search.
+
+	* io.c (appendline): remove unused arguments
+
+	* io.c (rb_io_getline_fast): make much simpler (and faster).
+
+Thu Jan 24 02:13:07 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* insns.def (expandarray): fix stack inc.  [ruby-dev:32892]
+
+	* bootstraptest/test_knownbug.rb, test_massign.rb: move a fixed test.
+
+Thu Jan 24 01:00:34 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.{c, h} (rb_usascii_encoding): added.
+
+	* parse.y (parser_str_new, rb_intern3): ascii only string literal is
+	  US-ASCII.
+
+	* ruby.c (proc_optionc): -Kn means ASCII-8BIT.
+
+Wed Jan 23 23:54:40 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* sprintf.c: fix comment.  [ruby-dev:33275]
+
+	* math.c: fix comment.  [ruby-dev:33276]
+
+Wed Jan 23 22:47:34 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_struct.rb: add tests to achieve over 90% test
+	  coverage of struct.c.
+
+	* test/ruby/test_sprintf.rb: ditto for sprintf.c.
+
+	* test/ruby/test_math.rb: ditto for math.c.
+
+Wed Jan 23 22:14:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/japanese.c (rb_from_Windows_31J, rb_to_Windows_31J):
+	  provisional workaround for Windows-31J.  [ruby-dev:33320]
+
+Wed Jan 23 15:25:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_strftime): copy encoding from format.  [ruby-dev:33303]
+
+Wed Jan 23 15:04:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_make_independent): should set length.
+
+	* string.c (rb_str_associate): hide associated array from ObjectSpace.
+
+	* string.c (rb_str_associated): return associated array with freezing
+	  instead of false.  [ruby-dev:33282]
+
+	* string.c (rb_str_freeze): freeze associated array together.
+
+Wed Jan 23 13:39:48 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_prepare_re): fix SEGV by
+	  /a/ =~ "aa".force_encoding("utf-16be").
+
+Wed Jan 23 11:53:26 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_mod_check, str_nth, str_offset): constified.
+
+	* string.c (rb_str_dump): dump in ASCII-8BIT always.
+
+Wed Jan 23 10:18:10 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_method.c (rb_export_method): set ruby_vm_redefined_flag for
+	  visibility change as well.  reported by K.Kosako in
+	  http://d.hatena.ne.jp/kkos/20080122#1201012720.
+
+Tue Jan 22 22:26:23 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_bignum.rb: change some tests because rational
+	  redefines Bignum#quo and Bignum#**.
+
+Tue Jan 22 20:58:15 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): need to output sodir rule.
+
+Tue Jan 22 19:37:16 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): lib files shouldn't depend on install
+	  dir because if the dir is newer than lib files, lib files will be
+	  always copied.
+
+Tue Jan 22 17:52:52 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/utf_16_32.c: Streamline parentheses, add more
+	  'static' qualifiers.
+
+Tue Jan 22 12:57:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in (MINIRUBY): remove -I$(EXTOUT)/$(arch) from
+	  MINIRUBY since miniruby might not be able to load DLL.
+
+	* test/ruby/test_m17n.rb: move tests from bootstrap test.
+
+	* encoding.c (enc_find): should check name if ASCII compatible.
+
+	* string.c (rb_str_end_with): should check character boundary.
+
+	* encoding.c (rb_enc_compatible): encoding must be ASCII
+	  compatible before checking ENC_CODERANGE_7BIT.
+
+	* encoding.c (rb_enc_compatible): wrong compatibility condition.
+	  [ruby-dev:33273]
+
+Tue Jan 22 09:26:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_each_char): iterates over a shadow.
+	  [ruby-dev:33243]
+
+Tue Jan 22 08:59:52 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/formatter.rb:  Indent labeled lists like note lists.
+
+	* test/rdoc/test_rdoc_ri_overstrike_formatter.rb:  Added.
+
+	* test/rdoc/test_rdoc_ri_formatter.rb:  Added tests.
+
+Tue Jan 22 04:40:28 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (rb_intern3): do not call rb_enc_mbclen() if *m is
+	  ASCII.  [ruby-talk:287225]
+
+	* string.c (rb_str_each_line): use rb_enc_is_newline() to gain
+	  performance if the record separator ($/) is not modified.
+
+Tue Jan 22 01:15:51 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ChangeLog: format-time-string under C locale. [ruby-dev:33261]
+
+Tue Jan 22 00:45:12 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_bignum.rb: add tests for bignum.c.
+
+Tue Jan 22 00:30:08 2008  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (big_shift): fix a bug that caused infinite loop when
+	  left shifting.
+
+Mon Jan 21 20:09:38 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (marshal_load): initialize the cache.
+
+Mon Jan 21 19:42:42 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c, enc/trans/utf_16_32.c, test/ruby/test_transcode.rb:
+	  added UTF-32BE and UTF-32LE conversions.
+
+Mon Jan 21 14:36:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (str_transcode): initialize transcoder in
+	  rb_transcoding.  [ruby-dev:33234]
+
+	* transcode_data.h (rb_transcoding): transcoder constified.
+
+Mon Jan 21 12:50:02 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c, gc.c (setjmp): sigsetjmp is a macro on cygwin.
+
+Mon Jan 21 12:35:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (transcode_loop, str_transcoding_resize): use unsigned
+	  char.  [ruby-dev:33232]
+
+	* transcode_data.h (rb_transcoding, rb_transcoder): removed callback
+	  parameters.
+
+	* enc/trans/japanese.c: ditto.
+
+	* enc/trans/utf_16_32.c: parenthesized bit-or operands.
+
+Mon Jan 21 11:59:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_each_char): move forward.  [ruby-dev:33231]
+
+Mon Jan 21 06:40:03 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (transcode_dispatch): constified return value.
+
+	* transcode_data.h (rb_transcoding): include pointer to rb_transcoder
+	  and auxiliary data.
+
+	* transcode_data.h (rb_transcoder): all callback functions should have
+	  their own parameters.
+
+	* enc/trans/{japanese,single_byte}.c: constified.
+
+Mon Jan 21 03:45:49 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_each_char): advance offset before get next char
+	  length.  [ruby-dev:33211]
+
+Sun Jan 20 20:00:20 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c, enc/trans/utf_16_32.c, test/ruby/test_transcode.rb:
+	  added UTF-16LE conversions.
+
+	* fixed changelog for last commit
+
+Sun Jan 20 17:54:00 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* added changelog for last commit
+
+Sun Jan 20 15:08:08 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/trans/utf_16_32.c: new file, currently implementing
+	  UTF-16BE conversions only.
+
+	* test/ruby/test_transcode.rb: Added tests for UTF-16BE;
+	  made check_both_ways() use force_encoding differently.
+
+	* transcode_data.h, transcode.c: Support for more conversion
+	  functions.
+
+Sun Jan 20 13:06:01 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_each_char): performance improvement, and stop if
+	  shortened in the block.  [ruby-dev:33189]
+
+Sun Jan 20 09:12:26 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: use -Ks when read insns.def. [ruby-dev#33185]
+
+	* parse.y: fix -e and stdin strings aren't set encoding.
+
+Sun Jan 20 05:12:44 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/make_encdb.rb: fix duplication check.
+
+Sun Jan 20 05:03:46 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ascii.c: remove definition of replica KOI8-U.
+
+Sun Jan 20 00:33:59 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/koi8_u.c: added.
+
+	* regenc.c, enc/utf_8.c, enc/unicode.c, enc/gb18030.c: add ARG_UNUSED.
+
+Sat Jan 19 22:41:39 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (coderange_scan): don't call mbclen functions for ASCII
+	  characters with ASCII compatible encoding.
+
+Sat Jan 19 21:00:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rdoc/template.rb (RDoc): defined to avoid uninitialized constant
+	  error by `./ruby test/rubygems/test_gem_server.rb'.
+
+Sat Jan 19 20:41:29 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (enc_new): don't free rb_encoding to avoid SEGV by
+	  `miniruby -e exit' on x86_64 GNU/Linux.
+
+Sat Jan 19 18:40:19 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (once): use an instance variable which points a hash
+	  as cache. [experimental]
+
+Sat Jan 19 17:21:29 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: parse's hints as an
+	  experimental function has been removed.
+
+Sat Jan 19 11:21:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (sigsetjmp): check if available.
+
+	* eval.c, gc.c (setjmp): do not use _setjmp if sigsetjmp is available.
+
+Sat Jan 19 11:10:11 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: Remove wrong assumptions about Cygwin.  a patch from
+	  Corinna Vinschen in [ruby-Bugs-17018].
+
+Sat Jan 19 09:23:14 2008  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_set_safe_array): should not use
+	  recursive calling.
+
+	* test/win32ole/test_win32ole_variant.rb: ditto.
+
+Sat Jan 19 08:58:47 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup: Remove ListBase and Line constants.
+
+	* lib/rdoc/ri: Allow output IO to be specified.
+
+	* test/rdoc/parser/test_parse_c.rb: Move up one level, fixed.
+
+	* test/rdoc/parser/test_rdoc_markup_attribute_manager.rb: Renamed to
+	  match new class name, updated to match new classes.
+
+	* test/rdoc/test_rdoc_ri_formatter.rb: Start of RI formatting tests.
+
+	* test/rdoc/test_rdoc_ri_attribute_manager.rb: Start of
+	  RDoc::RI::AttributeManager tests.
+
+	* test/rdoc/test_simple_markup.rb: Moved to match new class name.
+
+Sat Jan 19 08:35:36 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_prepare): get encoding from the first line.
+	  [ruby-dev:33168]
+
+	* ruby.c (load_file): set encoding to input with set_encoding.
+
+Sat Jan 19 03:46:42 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (thread_create_core): prohibit thread creation in the
+	  frozen thread group.  a patch in [ruby-dev:33176] from sheepman
+	  <sheepman AT sheepman.sakura.ne.jp>.
+
+	* thread.c (thread_create_core): should inherit ThreadGroup from
+	  the current thread.
+
+Sat Jan 19 00:37:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): set result encoding for wider width.
+
+Sat Jan 19 00:13:19 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (w32_wait_events): shouldn't invoke interrupt handle
+	  by timer_thread.
+
+Fri Jan 18 23:49:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_create_core): set thread group before creating
+	  thread.
+
+Fri Jan 18 20:19:51 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (ripper_initialize): too early to set parser->enc.
+
+Fri Jan 18 20:03:05 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.mak (BASERUBY): nmake cannot execute ruby correctly
+	  if the path of ruby.exe is quoted.
+
+	* win32/setup.mak ($(ARCH)): if a macro is appended by $(APPEND),
+	  a space will be inserted on the top of the line.
+
+Fri Jan 18 17:56:09 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h, insnhelper.h, thread_pthread.h, vm_core.h, vm_opts.h:
+	  prefixed include guards with RUBY.
+
+	* id.h: added include guard.
+
+	* regenc.h, regint.h, regparse.h: prefixed include guards with
+	  ONIGURUMA.
+
+Fri Jan 18 15:57:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_cleanup_func): ignore errors from destroying mutex
+	  of dead thread.  [ruby-core:15069]
+
+	* thread_pthread.c, thread_win32.c (native_thread_destroy): ditto.
+
+Fri Jan 18 15:56:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_name_list_i, rb_enc_aliases_enc_i): freeze
+	  element strings to be returned.
+
+Fri Jan 18 14:36:34 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_m17n.rb (test_str_dump): added test for
+	  String#dump. [ruby-dev:33142]
+
+Fri Jan 18 12:25:13 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (load_encoding): check if successfully loaded.
+
+	* encoding.c (rb_enc_find_index): use original encoding name to
+	  replicate loaded encoding instead alias.
+
+Fri Jan 18 09:43:02 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* re.c (rb_char_to_option_kcode): Regexp switch `s' should mean
+	  Windows-31J, as wells as `-Ks'.
+
+Fri Jan 18 09:22:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_initialize): explicitly call rb_ascii8bit_encoding().
+
+	* parse.y (parser_prepare): lex_input may not be have encoding (e.g. IO).
+
+	* parse.y (rb_parser_compile_string): set encoding from input string.
+
+	* encoding.c (rb_enc_find_index): use ASCII-8BIT if loading known
+	  encoding failed.
+
+	* parse.y (ripper_initialize): move parser->enc initialization.
+
+	* encoding.c (rb_enc_aliases_enc_i): exclude non alias names from
+	  Encoding.aliases.
+
+	* encoding.c (rb_enc_find_index): use original encoding name to
+	  load DLL.
+
+Fri Jan 18 07:06:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (Init_IO): stdin/stdout may not be duplex.
+
+Fri Jan 18 04:27:57 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* sample/rdoc/markup/rdoc2latex.rb:  Fix for new namespacing.
+
+	* lib/rdoc/markup/to_latex.rb:  Fix namespacing.
+
+Fri Jan 18 02:02:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/runner.rb (assert_valid_syntax): added.
+
+	* bootstraptest/test_knownbug.rb: added test for [ruby-list:44479]
+
+Fri Jan 18 01:48:06 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): check argument number to
+	  attr_reader.  [ruby-core:15120]
+
+Fri Jan 18 00:49:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_check_readable): flush tied write IO too.
+
+	* io.c (Init_IO): tie stdin with stdout.  [ruby-core:15107]
+
+Fri Jan 18 00:23:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_free): removed since rb_encoding may be used while
+	  cleanup.
+
+Fri Jan 18 00:17:19 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/euc_cn.c: split from enc/euc_kr.c.
+
+Fri Jan 18 00:03:54 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_init): use default external encoding
+	  if nothing is given.  a patch from sheepman <sheepman AT
+	  sheepman.sakura.ne.jp> in [ruby-dev:33159].
+
+Thu Jan 17 23:56:20 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (encdb.h): give output file name to make_encdb.rb.
+
+	* encoding.c (enc_table): simplified.
+
+	* encoding.c (enc_register_at): lazy loading.  [ruby-dev:33013]
+
+	* regenc.h (ENC_DUMMY): added.
+
+	* enc/make_encdb.rb: now emits macros only.
+
+	* enc/iso_2022_jp.h: split from encoding.c.
+
+Thu Jan 17 21:48:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_char_to_option_kcode): fixed typo.
+
+Thu Jan 17 21:01:25 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (Date::Infinity#<=>): didn't work.  A patch from
+	  Dirkjan Bussink <d.bussink AT gmail.com> [ruby-core:15098].
+	  This is a bug obviously.  However it didn't affect the library's
+	  functions.
+
+	* lib/date.rb, lib/date/format.rb: some trivial changes.
+
+Thu Jan 17 13:07:18 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_dump): preserve the encoding of source string
+	  if it is ASCII compatible.  otherwise, add '.force_encoding()'
+	  for ugly work around.  maybe we should implement some other way
+	  to keep non ASCII encoding in dumped string.  [ruby-dev:33142]
+
+Thu Jan 17 10:30:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_fwrite): always flush IO on tty, even without newlines.
+	  [ruby-core:15107]
+
+Wed Jan 16 22:45:27 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_register_at): make own copy.  [ruby-dev:33136]
+
+Wed Jan 16 18:03:10 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open, rb_io_s_popen): clear temporary object to release
+	  and prevent from GC.
+
+Wed Jan 16 17:55:07 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (fix_quo): typo. a patch from Shin-ichiro HARA
+	  <sinara AT blade.nagaokaut.ac.jp> in [ruby-dev:33130]
+
+Wed Jan 16 17:36:26 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/test_delegate.rb: add new test file for delegate.rb.
+
+Wed Jan 16 16:14:00 2008  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ruby.1: Fix grammar.
+
+Wed Jan 16 15:26:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (sys_fail2): get rid of unlimited alloca.
+
+	* io.c (mode_enc, pipe_open, rb_io_s_popen): ditto.
+
+	* load.c (rb_feature_p): ditto.
+
+	* object.c (rb_cstr_to_dbl): ditto.
+
+	* io.c (mode_enc): fixed uninitialized variable.
+
+Wed Jan 16 12:51:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_str_tmp_new, rb_str_shared_replace):
+	  prototype moved.
+
+Tue Jan 15 23:52:51 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/*: add ARG_UNUSED.
+
+	* enc/koi8_u.c: added.
+
+Tue Jan 15 23:00:08 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/utf_{16,32}{be,le}.c: remove some ARG_UNUSED.  replace struct
+	  OnigEncodingST by OnigEncoding.
+
+Tue Jan 15 22:30:43 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (ENC_REGISTER): use &OnigEncoding*.
+	  (ENCINDEX_UTF_8): renamed from ENCINDEX_UTF8.
+	  (rb_enc_init): use ENC_REGISTER.
+
+	* include/ruby/oniguruma.h (OnigEncodingUTF8, ONIG_ENCODING_UTF8):
+	  removed.
+
+	* enc/*.c: remove use of &encoding_*; use enc argument instead.
+
+Tue Jan 15 18:44:46 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enc/utf_8.c: remove use of ONIG_ENCODING_UTF8 altogether; use
+	  enc argument instead.
+
+Tue Jan 15 18:05:26 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/utf_8.c (ONIG_ENCODING_UTF8): reverted.
+
+Tue Jan 15 18:01:55 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (MKFILES): add dependencies.
+
+Tue Jan 15 18:00:16 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enc/utf_8.c (OnigEncodingDefine): encoding name should be kept
+	  unchanged.
+
+Tue Jan 15 17:53:05 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/utf_8.c: renamed as IANA name.
+
+	* enc/Makefile.in: ditto.
+
+Tue Jan 15 16:59:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): encoding libraries cannot be loaded until
+	  load path is set.
+
+Tue Jan 15 15:09:28 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/setup.mak: strip out empty lines from CPP output.
+
+Tue Jan 15 14:57:38 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {win,bcc}32/setup.mak (-basic-vars-): expand BASERUBY to full path
+	  to get rid of ./ruby.exe.
+
+	* win32/enc-setup.mak: workaround for Borland make.
+
+Tue Jan 15 14:44:53 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_locale_charmap): use ASCII-8BIT in miniruby.
+
+Tue Jan 15 13:54:41 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32}/Makefile.sub (RUNRUBY): need to set archdir when
+	  invoking ruby.
+
+Tue Jan 15 13:43:18 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (us_ascii.o): add dependencies. [ruby-dev:33111]
+
+Tue Jan 15 03:41:42 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (eval): check if backtrace is empty.  [ruby-core:15040]
+
+Tue Jan 15 01:28:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: simplified dummy objects dependencies.
+
+Tue Jan 15 01:19:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (OBJS): moved encoding.o from COMMONOBJS.
+
+	* common.mk (dmyencoding.o): added.  [ruby-dev:33099]
+
+	* configure.in, {win,bcc}32/Makefile.sub (MINIOBJS): added
+	  dmyencoding.o.
+
+	* dmyencoding.c (rb_locale_charmap): returns nil for miniruby.
+
+Tue Jan 15 00:05:50 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (appendline): specifying limit should not generate broken
+	  byte sequence.  strings should be rounded.  [ruby-dev:33088]
+
+Mon Jan 14 23:33:02 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb (Kconv.tolocale): argument is str.
+
+Mon Jan 14 23:31:05 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in (setup): add -I$(EXTOUT)/$(arch) to MINIRUBY.
+
+	* bootstraptest/runner.rb (main): expand -I directory path.
+
+Mon Jan 14 23:28:10 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/enc-setup.mak (BUILTIN_ENCOBJS): depends on enc/Makefile.in.
+
+Mon Jan 14 22:48:16 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_char_to_option_kcode): use rb_enc_find_index() instead
+	  of using fixed index value.
+
+	* enc/Makefile.in (encsrcdir): make US-ASCII built-in.
+
+Mon Jan 14 22:25:02 2008  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* golf_prelude.rb: Shorter method name completion.  Same method
+	  used for const missing. do_while and do_until added.  Enumerator
+	  gains all of Array's abilities. Ex:
+	    '123'.m{|i|i*2} #=> "112233"
+	    '123'.pe #=> '123'.perm*' ' #=> "123 132 213 231 312 321"
+	  base on a patch from Darren Smith <darrenks AT ml1.net>.
+
+Mon Jan 14 21:10:02 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enc/us_ascii.c: wrong alias name: ANSI_X3.4-1986.
+
+	* rubytest.rb: add -I#{srcdir} to load encoding DLL.
+
+Mon Jan 14 18:53:58 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: clear thread structure.
+	  (TODO: survey that child process should clear mutex or not).
+
+	* bootstraptest/test_knownbug.rb, test_thread.rb: move a fixed test.
+
+Mon Jan 14 18:43:38 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb: add "flunk" method.
+
+	* bootstraptest/test_knownbug.rb: fix to use flunk.
+
+Mon Jan 14 18:10:59 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.h: remove dangerous assembler sentence.
+
+Mon Jan 14 18:06:37 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_locale_encoding): return US-ASCII when charmap is
+	  nil.
+
+Mon Jan 14 16:12:58 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/shellwords.rb: scape should be an alias to shellescape.  a
+	  patch from Masahiro Kawato <m-kawato AT mwb.biglobe.ne.jp> in
+	  [ruby-dev:33060].
+
+Mon Jan 14 16:09:16 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.1: a patch to describe --encoding.  a patch from Yugui
+	  <yugui AT yugui.sakura.ne.jp> in [ruby-dev:33079].
+
+	* ruby.c: ditto.
+
+Mon Jan 14 13:49:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_prepare_re): initialize error message buffer.
+	  (rb_reg_search): ditto.
+	  (rb_reg_check_preprocess): ditto.
+	  (rb_reg_new_str): ditto.
+	  (rb_enc_reg_new): ditto.
+	  (rb_reg_compile): ditto.
+	  (rb_reg_initialize_m): ditto.
+	  (rb_reg_s_union_m): ditto.
+
+Mon Jan 14 12:33:07 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup*:  Renamespace from SM::SimpleMarkup to
+	  RDoc::Markup.
+
+Mon Jan 14 10:45:45 2008  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/ascii.c: Exchanged order of arguments for one ENC_ALIAS
+
+Mon Jan 14 09:19:07 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/time.rb: do not reference Time directly from the inside of
+	  definitions. [ruby-dev:33059]
+
+Mon Jan 14 05:44:44 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/*.c: add replicas and aliases.
+
+	* enc/make_encdb.h: add duplicate and undefined check.
+
+Mon Jan 14 02:03:05 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/oniguruma.h: remove ONIG_ENCODING_* and OnigEncoding*
+	  which are not builtin.
+
+	* regenc.{c,h} (onigenc_mb2_code_to_mbclen, onigenc_mb4_code_to_mbclen):
+	  fix prototype.
+
+	* enc/big5.c, enc/euc_kr.c, enc/euc_tw.c, enc/gb18030.c,
+	  enc/koi8_r.c, enc/windows_1251.c: imported from Oniguruma.
+
+Sun Jan 13 22:47:28 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/make_encdb.h: sort encoding names by original name.
+
+	* encoding.c, enc/*.c: define replicas and aliases.
+
+Sun Jan 13 20:24:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c: add documents.
+
+Sun Jan 13 18:41:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (Init_Encoding): moved initialization from encdb.h.
+
+	* enc/make_encdb.rb (enc_name_list): constified.
+
+	* enc/make_encdb.rb (enc_init_db): moved some functions to encoding.c.
+
+Sun Jan 13 13:53:00 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file): local variable was not initialized when -x flag
+	  is given.
+
+	* ruby.c (load_file): script files should not be affected by locale.
+	  [ruby-dev:33054]
+
+Sun Jan 13 12:01:32 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/generators*: Reorganize RDoc generators.
+
+Sun Jan 13 11:41:11 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (ENCINDEX_EUC_JP, ENCINDEX_SJIS): removed.
+	  (rb_enc_init): EUC-JP and Shift_JIS are not builtin now.
+
+	* enc/Makefile.in: ditto.
+
+	* common.mk: ditto.
+
+	* ruby.c (proc_options): ditto.
+
+	* enc/shift_jis.c, enc/euc_jp.c: fixes for remove from builtin.
+
+Sun Jan 13 10:21:40 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_table): packed all enc_table stuff.
+
+Sun Jan 13 09:58:17 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_init): revert removing SJIS.
+
+	* enc/sjis.c: move to enc/shift_jis.c, to make encoding name equal to
+	  filename for convenience of loading lib.
+
+	* enc/shift_jis.c: moved from enc/sjis.c.
+
+	* common.mk: follows enc/shift_jis.c.
+
+	* enc/Makefile.in: ditto.
+
+Sun Jan 13 09:22:33 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (incs): includes encdb.h.
+
+Sun Jan 13 09:17:29 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {bcc,win}32/Makefile.sub (MV): use move instead of ren.  [ruby-Bugs-17019]
+
+Sun Jan 13 01:52:31 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enc/make_encdb.rb: should work on Ruby 1.8.  [ruby-dev:33069]
+
+	* common.mk (encdb.h): pass enc dir from outside to make_encdb.rb.
+
+Sun Jan 13 00:01:07 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/make_encdb.rb: added. search enc/*.c and make encoding database.
+
+	* regenc.h (ENC_REPLICATE, ENC_ALIAS): added for defining replica
+	  encoding and encoding alias.
+
+	* encoding.c (rb_enc_init): move alias definitions to enc/*.c.
+	  (rb_enc_find_index): search original of replica and alias when no
+	  encoding library.
+	  (rb_enc_name_list, rb_enc_aliases_enc_i, rb_enc_aliases_str_i,
+	   rb_enc_aliases, Encoding.name_list, Encoding.aliases): added.
+	  (Init_Encoding): init encdb.
+
+	* enc/ascii.c, enc/us_ascii.c, enc/euc_jp.c, enc/sjis.c:
+	  add replica encoding and encoding alias definition.
+
+	* common.mk (dist-clean-local): add rule for remove encdb.h.
+
+Sat Jan 12 18:27:41 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_define_alloc_func, rb_undef_alloc_func): should
+	  define/undef on a singleton class.  [ruby-core:09959]
+
+Sat Jan 12 12:44:36 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c: rdoc update.
+
+Sat Jan 12 12:01:49 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: tuning for performance.
+
+Sat Jan 12 11:29:45 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/test_proc.rb: fixed wrong expected result.  pointed
+	  out by Kornelius "murphy" Kalnbach <murphy AT rubychan.de> in
+	  [ruby-core:15022].
+
+Sat Jan 12 04:38:38 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (process_options): -e'script' is locale encoding by default.
+	  (load_file): ruby script from stdin is locale encoding by default.
+
+Sat Jan 12 04:31:59 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: fix bug: -m was -m0.
+
+Fri Jan 11 23:22:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (string.c): call rb_str_buf_append to update encoding of
+	  str1, even if str2 is empty.
+
+Fri Jan 11 20:20:16 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_mark): needs to mark the receiver too.  a patch from
+	  Chris Heath <chris AT heathens.co.nz> in [ruby-core:14983].
+	  [ruby-core:14885]
+
+Fri Jan 11 18:28:49 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/usage.rb: Removed.
+
+	* lib/getoptlong.rb: Update example to not use lib/rdoc/usage.rb.
+
+Fri Jan 11 18:17:10 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/driver.rb (read_yaml): Follow namespace change
+	  [ruby-core:14964].
+
+Fri Jan 11 16:55:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_append): performance improvement.
+
+Fri Jan 11 12:35:56 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: moved broken syscall checks from process.c etc.
+
+	* defines.h (WORDS_BIGENDIAN): honor __BIG_ENDIAN__ than the result of
+	  configure.
+
+	* lib/rdoc/options.rb (check_diagram): more precise check, darwin
+	  is not Windows but mingw is on it.
+
+Fri Jan 11 09:59:05 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: update to r1.163.
+
+	* ext/nkf/nkf.c: ASCII's canonical name is US-ASCII.
+
+	* ext/nkf/lib/kconv.rb (Kconv.isjis): force_encoding('BINARY').
+
+Fri Jan 11 09:23:31 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (set_base_encoding): must use rb_enc_dummy_p.
+
+Fri Jan 11 06:13:14 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_to_encoding_index, rb_to_encoding): disallow nil.
+	  [ruby-dev:33003]
+
+Fri Jan 11 01:08:01 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (rb_mutex_unlock): proper error message for unlocking
+	  mutex that is not locked.  a patch from Yusuke ENDOH
+	  <mame at tsg.ne.jp> in [ruby-dev:33010].
+
+Thu Jan 10 18:00:41 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* prelude.rb (Mutex::synchronize): capture exception from unlock.
+	  [ruby-dev:32935]
+
+Thu Jan 10 10:15:03 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_encoding_set): IO.pipe("euc-jp", nil) should work as
+	  IO.pipe("euc-jp", nil).  [ruby-dev:33000]
+
+	* io.c (io_encoding_set): handle nil for v1.
+
+Thu Jan 10 02:41:22 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_binmode): should not alter encoding.  [ruby-dev:32918]
+
+	* io.c (io_read_encoding): need not to return ASCII-8BIT for
+	  binary IO.
+
+Wed Jan  9 22:04:17 2008  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/mathn.rb (Prime#each): returns an enumerator if no block
+	  given. [ruby-dev:32815]
+
+Wed Jan  9 22:03:26 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_replicate): replica of dummy is a dummy.
+
+Wed Jan  9 20:55:38 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/e2mmap.rb (Exception2MessageMapper::Raise): define fail.
+	  [ruby-dev:32854]
+
+Wed Jan  9 20:35:42 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httprequest.rb: support X-Forwarded-* header fields.
+	  WEBrick::HTTPRequest#{host,port,request_uri} is derived having
+	  regards to X-Forwarded-Proto and X-Forwarded-Host.
+
+	* lib/webrick/httprequest.rb
+	  (WEBrick::HTTPRequest#server_name?): new method.
+	  (WEBrick::HTTPRequest#remote_ip?): new method.
+	  (WEBrick::HTTPRequest#ssl?): new method.
+
+Wed Jan  9 18:24:39 2008  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* golf_prelude.rb (Array#to_s): alias to join.
+
+	* golf_prelude.rb (FalseClass#to_s): return "".
+
+Wed Jan  9 16:59:54 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_enc_cr_str_buf_cat): fix self appending.
+
+Wed Jan  9 15:54:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): give priority command line encoding option
+	  to RUBYOPT, and enable -E option in RUBYOPT.
+
+	* ruby.c (load_file): deal with encoding option in shebang line if
+	  nothing in command line and RUBYOPT.
+
+Wed Jan  9 14:55:36 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (yycompile0): remove setting parser->enc because it is set
+	  in parser_prepare() by previous change of parser_prepare().
+
+Wed Jan  9 14:52:18 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_buf_cat, rb_str_buf_append): deal with self
+	  appending.
+
+Wed Jan  9 14:44:57 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (parser_prepare): set parser->enc from lex_input for ripper.
+
+Wed Jan  9 13:45:52 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/server.rb (WEBrick::HTTPServer#start):
+	  :DoNotReverseLookup option had not been performed.
+
+Wed Jan  9 13:03:34 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_buf_cat): do not recalculate coderange
+	  value if it's given from outside.
+
+Wed Jan  9 08:42:01 2008  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* enum.c:  Updating the documentation of Enumerable#zip to reflect
+	  the recent changes Matz made to the method.
+
+Wed Jan  9 01:35:10 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/Makefile.in (BUILTIN_ENCS): UTF-{16,32}{BE,LE} are not builtin.
+
+Tue Jan  8 23:55:15 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_init): UTF-{16,32}{BE,LE} are not builtin.
+
+Tue Jan  8 22:33:03 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c, Makefile.in, include/ruby/oniguruma.h,
+	  enc/Makefile.in: fix rules for UTF-{16,32}{BE,LE}.
+
+Tue Jan  8 20:02:08 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win{32,ce}/Makefile.sub: merged.
+
+Tue Jan  8 19:48:15 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/driver.rb: Speed up Marshal.load. Fix bug with nested
+	  classes' methods.
+
+Tue Jan  8 19:17:29 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/*: Clean up namespacing of RI's classes.
+
+Tue Jan  8 18:05:35 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* bin/ri, lib/rdoc/ri/*: Replace with Ryan Davis' cached ri.
+
+Tue Jan  8 17:32:07 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_zip): honor length of the receiver, not the
+	  shortest length.  [ruby-core:14738]
+
+	* enum.c (enum_zip): returns array not enumerator for no block
+	  form.  [ruby-core:14738]
+
+	* enumerator.c (next_ii): do not ignore multiple values yielded.
+
+	* array.c (rb_ary_zip): faster version without creating generators.
+
+Tue Jan  8 15:47:43 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/utf{16,32}_{be,le}.c: use &OnigEncodingName(*) instead of
+	  ONIG_ENCODING_*.
+
+Tue Jan  8 15:40:31 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regenc.c (onigenc_strlen_null, onigenc_str_bytelen_null): suppressed
+	  warnings.
+
+	* regenc.h, enc/unicode.c (onigenc_unicode_ctype_code_range): added
+	  encoding argument.
+
+	* enc/utf{16,32}_{be,le}.c: added init functions.
+
+	* enc/utf{16,32}_{be,le}.c: imported from Oniguruma 5.9.1.
+
+Tue Jan  8 15:03:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (str_gsub): avoid appending empty pre-match substr.
+
+Tue Jan  8 13:05:57 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_build_from_ary), iseq.c (iseq_load): fix for format change.
+
+Tue Jan  8 07:56:11 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_buf_append): fix append itself.
+
+Tue Jan  8 01:13:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (STR_ENC_GET): defined.  same as rb_enc_get without
+	  enc_capable.
+	  (coderange_scan): ASCII-8BIT test refined.
+	  (rb_enc_cr_str_buf_cat): new internal function to accumulate
+	  strings with encoding.
+	  (rb_enc_str_buf_cat): use rb_enc_cr_str_buf_cat.
+	  (rb_str_buf_append): ditto
+	  (str_gsub): use rb_str_buf_append.
+	  (rb_str_hash): use ENCODING_GET.
+	  (rb_str_comparable): ditto.
+	  (rb_str_cmp): compare encoding index, not rb_encoding address.
+
+Mon Jan  7 20:37:55 2008  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb: external encoding of
+	  tempfiles is set to "ASCII-8BIT".
+
+Mon Jan  7 19:39:50 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/template.rb:  Use ERB instead of custom template language.
+
+	* lib/rdoc/generators/template/html/old_html.rb:  Remove.
+
+	* lib/rdoc/generators/template/*: Convert to ERB.
+
+Mon Jan  7 19:11:30 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (Init_String): sym_match arity spec was wrong.  a patch
+	  from Hiroyuki Iwatsuki <don at na.rim.or.jp> in [ruby-dev:32957].
+
+Mon Jan  7 18:10:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (str_gsub): move rb_enc_get(str) to out of loop.
+
+Mon Jan  7 15:52:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h (OnigEncodingType): new member
+	  ruby_encoding_index to avoid linear search in rb_enc_to_index.
+
+	* include/ruby/encoding.h (rb_enc_to_index): macro defined to use
+	  ruby_encoding_index.
+
+	* encoding.c (rb_enc_to_index): removed.
+	  (enc_register_at): initialize ruby_encoding_index member.
+
+Mon Jan  7 16:10:35 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/tokenstream.rb:  Namespace under RDoc.
+
+Mon Jan  7 16:06:09 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/dot.rb:  Namespace under RDoc.
+
+	* lib/rdoc/diagram.rb:  Clean up formatting.
+
+Mon Jan  7 15:51:35 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/options.rb:  Convert to OptionParser, clean up -h output,
+	  namespace under RDoc.
+	* lib/rdoc/*:  Namespace RDoc::Options.
+
+Mon Jan  7 15:42:46 2008  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk, Makefile.in, */Makefile.sub (distclean-local): move
+	  removing rule of ext/ripper/y.output from common.mk to Makefiles
+	  that depend on platforms.
+
+Mon Jan  7 13:54:57 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_preprocess): fix fixed_enc condition.
+
+Mon Jan  7 11:51:49 2008  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/generators/ri_generator.rb:  Merge documentation from the
+	  same class on output.  Fixes bug where documentation could
+	  disappear.
+
+	* lib/rdoc/options.rb:  Fix typo.
+
+	* lib/rdoc/generators/*: Clean up some namespacing and make RDoc
+	  consistent.
+
+Mon Jan  7 11:44:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_internal_get_index): extracted from
+	  rb_enc_get_index.
+	  (rb_enc_internal_set_index): extracted from rb_enc_associate_index
+
+	* include/ruby/encoding.h (ENCODING_SET): work over ENCODING_INLINE_MAX.
+	  (ENCODING_GET): ditto.
+	  (ENCODING_IS_ASCII8BIT): defined.
+	  (ENCODING_CODERANGE_SET): defined.
+
+	* re.c (rb_reg_fixed_encoding_p): use ENCODING_IS_ASCII8BIT.
+
+	* string.c (rb_enc_str_buf_cat): use ENCODING_IS_ASCII8BIT.
+
+	* parse.y (reg_fragment_setenc_gen): use ENCODING_IS_ASCII8BIT.
+
+	* marshal.c (has_ivars): use ENCODING_IS_ASCII8BIT.
+
+Mon Jan  7 02:14:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (coderange_scan): avoid rb_enc_to_index.
+	  (rb_enc_str_buf_cat): ditto.
+	  (str_cat_char): use rb_enc_str_buf_cat.
+	  (rb_str_inspect): ditto.
+
+Mon Jan  7 01:36:49 2008  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (coderange_scan): optimize ASCII-8BIT string.
+	  (rb_enc_str_buf_cat): don't call coderange_scan if possible.
+
+Mon Jan  7 01:05:45 2008  Tanaka Akira  <akr@fsij.org>
+
+	* lib/erb.rb (ERB::Revision): cut off locale dependent string in Date
+	  keyword.
+
+Mon Jan  7 00:48:02 2008  Tanaka Akira  <akr@fsij.org>
+
+	* Date keyword removed to avoid inclusion of locale dependent
+	  string.  [ruby-dev:32940]
+
+Sun Jan  6 21:14:12 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_initialize_str): forbid raw non ASCII character
+	  for ASCII-8BIT regexp in non ASCII-8BIT script.
+
+Sun Jan  6 18:19:12 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_enc_str_buf_cat): declared.
+
+	* string.c (coderange_scan): extracted from rb_enc_str_coderange.
+	  (rb_enc_str_coderange): use coderange_scan.
+	  (rb_str_shared_replace): copy encoding and coderange.
+	  (rb_enc_str_buf_cat): new function for linear complexity string
+	  accumulation with encoding.
+	  (rb_str_sub_bang): don't conflict substituted part and replacement.
+	  (str_gsub): use rb_enc_str_buf_cat.
+	  (rb_str_clear): clear coderange.
+
+	* re.c (rb_reg_regsub): use rb_enc_str_buf_cat.
+
+Sun Jan  6 17:55:44 2008  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* lib/securerandom.rb: Add Win32 support.
+
+Sun Jan  6 09:32:58 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: introduced some constants
+	  (for internal use) and aliases (minute and second).
+
+	* sample/cal.rb: trivial adjustments.
+
+Sun Jan  6 01:38:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_initialize_str): /\x80/n is not an error even if script
+	  encoding is EUC-JP.
+
+Sun Jan  6 00:48:12 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::DNS#each_address): get A record and then AAAA
+	  record. [ruby-dev:32925]
+
+Sat Jan  5 21:48:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* vm_insnhelper.c (vm_callee_setup_arg): it is not inlinable because
+	  alloca is used.
+
+Sat Jan  5 16:50:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_resize): copy if old data is not empty
+	  [ruby-core:14785]
+
+Sat Jan  5 13:04:24 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): no need for intermediate object.
+
+Sat Jan  5 11:48:19 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (Init_Encoding): alias csWindows31J to Windows-31J.
+	  IE6 accepts csWindows31J but Windows-31J.
+
+Sat Jan  5 02:21:10 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (rb_intern): memorize interned ID for constant
+	  string, using gcc's __builtin_constant_p and statement expression.
+
+Sat Jan  5 02:14:45 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (trnext): should enable backslash escape.
+
+Sat Jan  5 01:50:32 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_eval): move instance_eval and instance_exec to
+	  BasicObject.  [ruby-core:14747]
+
+	* lib/delegate.rb: should preserve new methods in BasicObject.
+
+Sat Jan  5 01:46:29 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (proc_options): update according to the last API revert.
+
+Sat Jan  5 01:30:30 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h, re.c (rb_reg_new): keep interface same as
+	  1.8.  [ruby-core:14583]
+
+	* include/ruby/intern.h, re.c (rb_reg_new_str): renamed, and defines
+	  HAVE_RB_REG_NEW_STR macro to tell if it is available.
+
+	* include/ruby/encoding.h (rb_enc_reg_new): added.
+
+	* insns.def (toregexp), marshal.c (r_object0): use rb_reg_new_str().
+
+	* re.c (rb_reg_regcomp, rb_reg_s_union): ditto.
+
+Fri Jan  4 23:08:48 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_arg): use converted object.  [ruby-core:14759]
+
+Fri Jan  4 16:24:58 2008  Tanaka Akira  <akr@fsij.org>
+
+	* ext/digest/lib/digest/hmac.rb (Digest::HMAC#initialize): use
+	  String#bytesize to avoid test errors on EUC-JP environment.
+
+Fri Jan  4 14:00:50 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_prepare_re): check string encoding.  Oniguruma doesn't
+	  support invalid encoding.
+
+Fri Jan  4 10:22:25 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_search): avoid inner loop for reverse search.
+
+	* regexec.c: unset USE_MATCH_RANGE_MUST_BE_INSIDE_OF_SPECIFIED_RANGE
+	  which is turned on since oniguruma 5.9.1.
+
+Fri Jan  4 02:53:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/euc_jp.c: remove eucjp_ prefix.  breakpoint can be specified as
+	  euc_jp.c:mbc_enc_len.  avoid needless conflict by merge.
+
+	* enc/sjis.c: remove sjis_ prefix.
+
+	* enc/utf8.c: remove utf8_ prefix.
+
+	* enc/iso_8859_1.c: remove iso_8859_1_ prefix.
+
+	* enc/iso_8859_2.c: remove iso_8859_2_ prefix.
+
+	* enc/iso_8859_3.c: remove iso_8859_3_ prefix.
+
+	* enc/iso_8859_4.c: remove iso_8859_4_ prefix.
+
+	* enc/iso_8859_5.c: remove iso_8859_5_ prefix.
+
+	* enc/iso_8859_6.c: remove iso_8859_6_ prefix.
+
+	* enc/iso_8859_7.c: remove iso_8859_7_ prefix.
+
+	* enc/iso_8859_8.c: remove iso_8859_8_ prefix.
+
+	* enc/iso_8859_9.c: remove iso_8859_9_ prefix.
+
+	* enc/iso_8859_10.c: remove iso_8859_10_ prefix.
+
+	* enc/iso_8859_11.c: remove iso_8859_11_ prefix.
+
+	* enc/iso_8859_13.c: remove iso_8859_13_ prefix.
+
+	* enc/iso_8859_14.c: remove iso_8859_14_ prefix.
+
+	* enc/iso_8859_15.c: remove iso_8859_15_ prefix.
+
+	* enc/iso_8859_16.c: remove iso_8859_16_ prefix.
+
+Fri Jan  4 02:47:06 2008  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_search): iterate onig_match for reverse mode.
+
+Fri Jan  4 01:20:21 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32.h: only VC6 needs extern "C++" for math.h.  [ruby-talk:285660]
+
+Fri Jan  4 00:54:43 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/oniguruma.h: Oniguruma 5.9.1 merged.
+
+Fri Jan  4 00:20:47 2008  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_ungetc): move data in buffer if it is required to store the
+	  argument.
+
+Thu Jan  3 21:56:07 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (st_strcasecmp): declared for STRCASECMP.
+	  (st_strncasecmp): declared for STRNCASECMP.
+
+Thu Jan  3 20:24:48 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_jump.c (rb_f_catch): Restore cfp if caught thrown object.
+
+Thu Jan  3 19:45:57 2008  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_eval.rb, test_knownbug.rb: move a fixed test.
+
+Thu Jan  3 18:39:12 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c: (rb_tolower, rb_toupper): body was exchanged.
+
+Thu Jan  3 17:54:01 2008  Tanaka Akira  <akr@fsij.org>
+
+	* regenc.h (onigenc_ascii_is_code_ctype): put back.
+
+Thu Jan  3 17:33:09 2008  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_isalnum): defined.
+	  (rb_isalpha): ditto.
+	  (rb_isblank): ditto.
+	  (rb_iscntrl): ditto.
+	  (rb_isdigit): ditto.
+	  (rb_isgraph): ditto.
+	  (rb_islower): ditto.
+	  (rb_isprint): ditto.
+	  (rb_ispunct): ditto.
+	  (rb_isspace): ditto.
+	  (rb_isupper): ditto.
+	  (rb_isxdigit): ditto.
+	  (rb_tolower): ditto.
+	  (rb_toupper): ditto.
+
+	* include/ruby/ruby.h: don't include include/ruby/encoding.h.
+	  (rb_isascii): defined.
+	  (rb_isalnum): declared.
+	  (rb_isalpha): ditto.
+	  (rb_isblank): ditto.
+	  (rb_iscntrl): ditto.
+	  (rb_isdigit): ditto.
+	  (rb_isgraph): ditto.
+	  (rb_islower): ditto.
+	  (rb_isprint): ditto.
+	  (rb_ispunct): ditto.
+	  (rb_isspace): ditto.
+	  (rb_isupper): ditto.
+	  (rb_isxdigit): ditto.
+	  (rb_tolower): ditto.
+	  (rb_toupper): ditto.
+	  (ISASCII): simplified.
+	  (ISPRINT): ditto.
+	  (ISSPACE): ditto.
+	  (ISUPPER): ditto.
+	  (ISLOWER): ditto.
+	  (ISALNUM): ditto.
+	  (ISALPHA): ditto.
+	  (ISDIGIT): ditto.
+	  (ISXDIGIT): ditto.
+	  (TOUPPER): ditto.
+	  (TOLOWER): ditto.
+
+	* include/ruby/encoding.h (rb_isascii): removed.
+	  (rb_isalnum): ditto.
+	  (rb_isalpha): ditto.
+	  (rb_isblank): ditto.
+	  (rb_iscntrl): ditto.
+	  (rb_isdigit): ditto.
+	  (rb_isgraph): ditto.
+	  (rb_islower): ditto.
+	  (rb_isprint): ditto.
+	  (rb_ispunct): ditto.
+	  (rb_isspace): ditto.
+	  (rb_isupper): ditto.
+	  (rb_isxdigit): ditto.
+	  (rb_tolower): ditto.
+	  (rb_toupper): ditto.
+
+	* common.mk: dependency updated.
+
+Thu Jan  3 15:10:26 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_isascii): simplified.
+	  (rb_isalnum): call onigenc_ascii_is_code_ctype without indirect call.
+	  (rb_isalpha): ditto.
+	  (rb_isblank): ditto.
+	  (rb_iscntrl): ditto.
+	  (rb_isdigit): ditto.
+	  (rb_isgraph): ditto.
+	  (rb_islower): ditto.
+	  (rb_isprint): ditto.
+	  (rb_ispunct): ditto.
+	  (rb_isspace): ditto.
+	  (rb_isupper): ditto.
+	  (rb_isxdigit): ditto.
+
+	* include/ruby/oniguruma.h (onigenc_ascii_is_code_ctype): declaration
+	  moved from regenc.h.
+
+Thu Jan  3 14:37:17 2008  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (parser_magic_comment): use STRNCASECMP.
+	  (set_file_encoding): ditto.
+
+Thu Jan  3 11:44:37 2008  Tanaka Akira  <akr@fsij.org>
+
+	* time.c: don't mention an obsolete library, ParseDate.
+
+Thu Jan  3 11:28:58 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (fptr_finalize): clear errno first.  [ruby-talk:284492]
+
+Thu Jan  3 05:02:36 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/us_ascii.c: add us_ascii_ prefix for functions to ease
+	  setting breakpoint when debugging.
+
+	* enc/euc_jp.c: add eucjp_ prefix.
+
+	* enc/sjis.c: add sjis_ prefix.
+
+	* enc/iso_8859_1.c: add iso_8859_1_ prefix.
+
+	* enc/iso_8859_2.c: add iso_8859_2_ prefix.
+
+	* enc/iso_8859_3.c: add iso_8859_3_ prefix.
+
+	* enc/iso_8859_4.c: add iso_8859_4_ prefix.
+
+	* enc/iso_8859_5.c: add iso_8859_5_ prefix.
+
+	* enc/iso_8859_6.c: add iso_8859_6_ prefix.
+
+	* enc/iso_8859_7.c: add iso_8859_7_ prefix.
+
+Thu Jan  3 02:44:34 2008  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (conv_digit): use ISDIGIT, ISLOWER and ISUPPER.
+
+Wed Jan  2 23:50:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* util.c (ruby_strtoul): "0x", "+" and "-" is not a valid integer.
+	  end of integer should be just after "0", the beginning, the
+	  beginning respectively.
+
+Wed Jan  2 15:23:15 2008  Tanaka Akira  <akr@fsij.org>
+
+	* util.c (ruby_strtoul): locale independent strtoul is implemented to
+	  avoid "i".to_i(36) cause 0 under tr_TR locale on Debian GNU/Linux
+	  4.0 (Etch).
+	  This is newly implemented, not a copy of missing/strtoul.c.
+
+	* include/ruby/ruby.h (ruby_strtoul): declared.
+	  (STRTOUL): defined to use ruby_strtoul.
+
+	* bignum.c, pack.c, ext/socket/socket.c: use STRTOUL.
+
+	* configure.in (strtoul): don't check.
+
+	* missing/strtoul.c: removed.
+
+	* include/ruby/missing.h (strtoul): removed.
+
+	* common.mk (strtoul.o): removed.
+
+	* LEGAL (missing/strtoul.c): removed.
+
+Wed Jan  2 14:41:08 2008  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (strcasecmp.o): removed.
+	  (strncasecmp.o): removed.
+
+	* include/ruby/missing.h (strcasecmp): removed.
+	  (strncasecmp): removed.
+
+Wed Jan  2 11:34:57 2008  Tanaka Akira  <akr@fsij.org>
+
+	* missing/strcasecmp.c: removed.  Ruby don't use locale dependent
+	  strcasecmp.
+
+	* missing/strncasecmp.c: ditto.
+
+	* configure.in: don't check strcasecmp and strncasecmp.
+
+	* LEGAL: missing/strcasecmp.c and missing/strncasecmp.c removed.
+
+Wed Jan  2 10:13:54 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* sample/time.rb: use Process.times instead of Time.times.
+
+Wed Jan  2 09:09:53 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* sample/goodfriday.rb: examples for date are enough.  retired.
+
+Wed Jan  2 08:58:54 2008  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* sample/cal.rb: just updated with the newest version.
+
+Wed Jan  2 01:19:31 2008  Tanaka Akira  <akr@fsij.org>
+
+	* enc/depend: dependency updated.
+
+Wed Jan  2 00:14:41 2008  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/{nkf.c, utf8tbl.c}: Update nkf.
+
+	* ext/nkf/nkf.c: fix documents.
+
+	* ext/nkf/lib/kconv.rb: fix documents.
+	  (Kconv.is*): use valid_encoding?.
+	  (Kconv.isjis): defined.
+
+Tue Jan  1 23:17:03 2008  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: dependency updated.
+
+Tue Jan  1 21:11:33 2008  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_isascii): defined.
+	  (rb_isalnum): ditto.
+	  (rb_isalpha): ditto.
+	  (rb_isblank): ditto.
+	  (rb_iscntrl): ditto.
+	  (rb_isdigit): ditto.
+	  (rb_isgraph): ditto.
+	  (rb_islower): ditto.
+	  (rb_isprint): ditto.
+	  (rb_ispunct): ditto.
+	  (rb_isspace): ditto.
+	  (rb_isupper): ditto.
+	  (rb_isxdigit): ditto.
+	  (rb_tolower): ditto.
+	  (rb_toupper): ditto.
+
+	* include/ruby/st.h (st_strcasecmp): declared.
+	  (st_strncasecmp): ditto.
+
+	* st.c (type_strcasehash): use st_strcasecmp instead of strcasecmp.
+	  (st_strcasecmp): defined.
+	  (st_strncasecmp): ditto.
+
+	* include/ruby/ruby.h: include include/ruby/encoding.h.
+	  (ISASCII): use rb_isascii.
+	  (ISPRINT): use rb_isprint.
+	  (ISSPACE): use rb_isspace.
+	  (ISUPPER): use rb_isupper.
+	  (ISLOWER): use rb_islower.
+	  (ISALNUM): use rb_isalnum.
+	  (ISALPHA): use rb_isalpha.
+	  (ISDIGIT): use rb_isdigit.
+	  (ISXDIGIT): use rb_isxdigit.
+	  (TOUPPER): defined.
+	  (TOLOWER): ditto.
+	  (STRCASECMP): ditto.
+	  (STRNCASECMP): ditto.
+
+	* dir.c, encoding.c, file.c, hash.c, process.c, ruby.c, time.c,
+	  transcode.c, ext/readline/readline.c: use locale insensitive
+	  functions.  [ruby-core:14662]
+
+Tue Jan  1 17:50:47 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_mode_enc): encoding spec is not allowed in binary mode.
+	  [ruby-dev:32913]
+
+Tue Jan  1 14:41:56 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml: 1.9 patch from Sam Ruby mentioned in his blog:
+	  <http://intertwingly.net/blog/2007/12/31/Porting-REXML-to-Ruby-1-9>
+	  [ruby-core:14639]
+
+Tue Jan  1 14:15:04 2008  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_substr): offset movement bug.  a patch from
+	  Vincent Isambart <vincent.isambart at gmail.com> in
+	  [ruby-core:14647].  [ruby-core:14644]
+
+Tue Jan  1 01:29:04 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_to_encoding): raises for non-nil, non-encoding,
+	  non-string object.  [ruby-core:14634]
+
+Tue Jan  1 01:04:06 2008  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): rejects dummy encoding.
+
+Mon Dec 31 23:53:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options, process_options): delays finding encoding
+	  until load_path is set.
+
+Mon Dec 31 23:27:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_resize): embeds if ptr is null.  [ruby-dev:32819]
+
+Mon Dec 31 23:17:22 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_service):
+	  call do_XXX which correspond with request method.
+	  (WEBrick::HTTPProxyServer#do_CONNECT,do_GET,do_POST,do_HEAD): added.
+
+	* test/webrick/test_httpproxy.rb: add test for WEBrick::HTTPProxyServer.
+
+Mon Dec 31 22:53:29 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread_pthread.c (native_sleep): timespec tv_sec may overflow on
+	  some platform.  a patch from zunda <zunda616e AT yahoo.co.jp> in
+	  [ruby-dev:32904].
+
+Mon Dec 31 19:35:20 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (IS_7BIT): removed.
+	  (single_byte_optimizable): new function to test optimizationability
+	  using single byte string.
+	  (str_strlen): use single_byte_optimizable instead of
+	  is_ascii_string.
+	  (str_nth): rename argument: asc -> singlebyte.
+	  (str_offset): ditto.
+	  (rb_str_substr): use single_byte_optimizable instead of IS_7BIT.
+	  (rb_str_index): ditto.
+	  (rb_str_rindex): ditto.
+	  (rb_str_splice): ditto.
+	  (rb_str_justify): ditto.
+
+Mon Dec 31 07:39:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* main.c, goruby.c (RUBY_MAIN_INIT): removed.
+
+	* goruby.c (goruby_run_node): run after ruby_init_loadpath() so that
+	  require works, and protect the call.
+
+Mon Dec 31 06:50:38 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: not use -I$(srcdir)/lib with $(MINIRUBY) for cross
+	  compiling.
+
+	* configure.in, {win,bcc}32/Makefile.sub (MINIRUBY): -I$(srcdir)/lib
+	  moved.
+
+Mon Dec 31 06:08:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_enc_sprintf, rb_enc_vsprintf): prototyped.
+
+	* sprintf.c (rb_enc_sprintf, rb_enc_vsprintf): new functions to format
+	  arguments with encoding.
+
+Sun Dec 30 23:48:00 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* golf_prelude.rb (String#/): define / as split, as association of
+	  Array#*.
+
+Sun Dec 30 23:19:06 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* golf_prelude.rb (@@golf_hash): for performance improvement.
+
+Sun Dec 30 22:44:50 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (_valid_time?): I'm not sure to recommend such an
+	  expression.  but anyway it is acceptable now.  [ruby-core:14580]
+
+Sun Dec 30 21:54:26 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (program, yycompile0): too early to drop lex_lastline in
+	  rules.
+
+Sun Dec 30 19:23:23 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bootstraptest/test_knownbug.rb: support DOSISH.
+
+Sun Dec 30 17:43:54 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (Init_Encoding): registered rb_encoding differs from
+	  ONIG_ENCODINGs.
+
+Sun Dec 30 13:56:15 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (program): clear input strings after all process.
+
+	* parse.y (parser_nextc, parser_yylex): should not drop lex_lastline
+	  while lex_p is valid.  [ruby-dev:32896]
+
+Sun Dec 30 10:54:49 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: rm largefile.h.
+
+	* common.mk: clean golf, conf*, preludes, and so on.
+
+	* enc/depend: silent and ignore error for rm.
+
+	* enc/Makefile.in: should define prefix and exec_prefix.
+
+Sun Dec 30 06:31:11 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* encoding.c (Init_encoding): register Windows-31J and its alias.
+	  [ruby-dev:32843]
+
+	* ruby.c (proc_options): -Ks options means Windows-31J, not Shift_JIS.
+
+Sun Dec 30 06:27:38 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (depend_rules): need to convert `/' to `\' for windows
+	  native commands.
+
+Sun Dec 30 01:43:56 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/Makefile.in (DLDFLAGS): like as extensions.  [ruby-core:14567]
+
+Sat Dec 29 23:48:13 2007  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_fflush): don't retry when wbuf modified by other threads.
+
+Sat Dec 29 22:44:30 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_regsub): returns the given string itself if nothing
+	  changed.
+
+	* string.c (rb_str_sub_bang): keeps code-range as possible.
+
+	* string.c (str_gsub): adjusts code-range.  [ruby-core:14566]
+
+Sat Dec 29 21:54:37 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (clean, distclean, realclean): should include clean-enc
+	  and others.  [ruby-dev:32887]
+
+Sat Dec 29 13:29:29 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/test_knownbug.rb: add a test reported by
+	  Kazuhiro NISHIYAMA.  [ruby-dev:32819].
+	  add a test reported by Frederick Cheung.  [ruby-core:14556].
+
+	* test/ruby/test_m17n.rb (test_gsub): add a test reported by
+	  Sam Ruby.  [ruby-core:14566]
+
+Sat Dec 29 04:46:58 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert_throws):
+	  throw won't raise NameError nor ThreadError but ArgumentError on 1.9.
+	  (Test::Unit::Assertions#assert_not_throws): ditto.
+
+	* test/testunit/test_assertions.rb: add assertions for throwing some
+	  objects other than Symbol.
+
+Sat Dec 29 03:10:12 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_unread): fix typo.
+
+Sat Dec 29 02:18:45 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_unread): adhoc workaround for non-binary mode of some DOSish
+	  platforms. this is not perfect and safety, but works with most cases.
+
+Fri Dec 28 23:53:18 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ext/strscan/strscan.c (str_new): new function for allocate an string
+	  with encoding propagation.
+	  (extract_range): use str_new.
+	  (extract_beg_len): ditto.
+	  (strscan_peek): ditto.
+	  (strscan_rest): ditto.
+
+Fri Dec 28 20:18:42 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* golf_prelude.rb (Object.say): derived from Perl 5.10.
+
+Fri Dec 28 19:39:34 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* encoding.c (rb_locale_encoding): should check return value from
+	  rb_locale_charmap().
+
+	* ruby.c (locale_encoding): removed.
+
+	* ruby.c (process_options): use rb_locale_encoding() instead of
+	  locale_encoding().
+
+	* ext/readline/readline.c (readline_readline): use locale encoding
+	  instead of input IO's encoding. [ruby-dev:32872]
+
+Fri Dec 28 19:29:07 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_readline, readline_s_set_input):
+	  use mReadline directly because self is not always same.
+
+Fri Dec 28 19:11:28 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_locale_encoding): defined.
+
+	* include/ruby/encoding.h (rb_locale_encoding): declared.
+
+Fri Dec 28 18:45:29 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_readline): set encoding to result.
+
+	* ext/readline/readline.c (readline_s_set_input, Init_readline): save
+	  input IO to hidden instance variable.
+
+Fri Dec 28 01:55:04 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c (transcode_dispatch): reverted some of the changes
+	  in r14746.
+
+	* transcode.c, enc/trans/single_byte.c: Added conversions to/from
+	  US-ASCII and ASCII-8BIT (using data tables).
+
+	* enc/trans/single_byte.c: Some spacing/ordering changes due to
+	  automatic data file generation.
+
+	* transcode_data.h, transcode.c: Preliminary code for using
+	  micro-conversion functions.
+
+	* test/ruby/test_transcode.rb: Added some tests for US-ASCII and
+	  ASCII-8BIT conversions.
+
+Fri Dec 28 17:33:44 2007  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (make_time_t): verify mktime and timegm result.
+
+Fri Dec 28 16:36:33 2007  NARUSE, Yui  <naruse@airemix.com>
+
+	* lib/resolv.rb (Resolv::DNS#each_address): now returns IPv6 address.
+
+Fri Dec 28 16:10:00 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/dot/dot.rb: Move to lib/rdoc/dot.rb. Fix namespacing.
+
+	* lib/rdoc/diagram.rb: Update for 1.9.
+
+Fri Dec 28 15:38:29 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup/sample/: Move to sample/rdoc/markup directory.
+
+Fri Dec 28 15:15:12 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/irb/completion.rb: Remove garbage ("X=1").
+
+Fri Dec 28 15:12:05 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc, test/rdoc: Move RDoc tests out of lib/.
+
+Fri Dec 28 15:10:47 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (set_base_encoding, enc_base_encoding): renamed
+	  based_encoding as base_encoding.
+
+Fri Dec 28 13:57:49 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* golf_prelude.rb (Integer#each): use alias simply.
+
+Fri Dec 28 13:45:21 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* golf_prelude.rb (Object.const_missing): No need to delegate to
+	  superclass.  Just raise a NameError when none matches.
+
+Fri Dec 28 13:18:47 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb, test/rss/test_version.rb: 0.2.2 -> 0.2.3.
+
+	* lib/rss/parser.rb, test/rss/test_parser.rb: supported "-" in tag name.
+	  Reported by Ray Chen. Thanks.
+
+Fri Dec 28 13:07:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (os_obj_of): returns an enumerator if no block given.  based on
+	  a patch from Yugui <yugui AT yugui.sakura.ne.jp>.  [ruby-dev:32828]
+
+Fri Dec 28 11:46:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/ytab.sed: skip yydestruct hack unless yymsg exists, for bison
+	  1.8 series.  [ruby-dev:32825]
+
+Fri Dec 28 11:39:02 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* golf_prelude.rb (Object.quine): need to join because SCRIPT_LINES__[]
+	  returns an array of lines.
+
+Fri Dec 28 11:16:53 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* golf_prelude.rb (Object.quine): get the script itself.
+
+Fri Dec 28 10:06:54 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* golf_prelude.rb (Object.const_missing): Auto-complete constants.
+
+Fri Dec 28 01:55:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (transcode_dispatch): allows transcoding from/to
+	  ASCII-8BIT.
+
+Fri Dec 28 01:47:25 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* golf_prelude.rb (Integer): Integer is now enumerable on goruby.
+
+Fri Dec 28 01:27:47 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rdoc/rdoc.rb (parse_files): don't depend on the default external
+	  encoding.
+
+Fri Dec 28 00:01:57 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (golf_prelude.c): use MINIRUBY instead of BASERUBY because
+	  tool/compile_prelude.rb requires rbconfig.rb.
+
+Thu Dec 27 23:56:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: should not use the libraries under the source directory
+	  at cross compiling.
+
+Thu Dec 27 23:43:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (EXTOBJS): uses ruby.rc always for other than
+	  rubyw.exe.
+
+Thu Dec 27 22:31:37 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rubygems/commands/update_command.rb (do_rubygems_update): use
+	  portable and safely ENV operation. reported in
+	  <http://slashdot.jp/developers/comments.pl?sid=384937&cid=1273085>.
+
+Thu Dec 27 21:47:04 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* mkconfig.rb (prefix): archdir is "1.9.0", not "1.9". reported in
+	  <http://slashdot.jp/developers/comments.pl?sid=384937&cid=1273085>.
+
+Thu Dec 27 17:57:30 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y, transcode_data.h, transcode.c, lib/weakref.rb,
+	  lib/irb/ruby-lex.rb, lib/irb/lc/error.rb, enc/trans/japanese.c:
+	  change "illegal" to "invalid" in a context which doesn't against
+	  a law.
+
+Thu Dec 27 16:37:06 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_s_union): show encodings in error message.
+
+Thu Dec 27 15:25:16 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_codelen): show codepoint in error message.
+
+	* include/ruby/encoding.h (rb_enc_codelen): comment it returns
+	  positive integer.
+
+	* string.c (rb_str_concat): rb_enc_codelen doesn't return 0.
+
+Thu Dec 27 15:18:44 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_codelen): error message refined.
+
+Thu Dec 27 15:11:27 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_check): show encodings in error message.
+
+Thu Dec 27 15:02:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_casecmp): fixed using a wrong variable.
+	  [ruby-list:44402]
+
+Thu Dec 27 14:34:38 2007  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_fflush): checks wbuf modification by other threads.
+	  not perfect.  it need locks.
+
+Thu Dec 27 10:44:03 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c: a patch to support IRIX from Andrew
+	  Thompson <andrew@hijacked.us> in [ruby-core:14447].
+
+Thu Dec 27 02:25:45 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_tmpsrc): retry to create file if Errno::EACCES
+	  occurs. this is a workaround for mswin32.
+
+Wed Dec 26 22:47:31 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::DNS::Name.==): fix for other is array of
+	  Resolv::DNS::Label::Str.
+
+	* lib/resolv.rb (Resolv::DNS::MessageEncoder#put_label): String#string
+	  is not defined, so replace to_s.
+
+	* lib/resolv.rb (Resolv::IPv6#to_name): ip6.int is obsoleted by
+	  ip6.arpa.
+
+Wed Dec 26 21:27:02 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (_xmlschema): some improvements.
+
+	* lib/date/format.rb (_parse): a new hint compfunc. [experimental]
+
+Wed Dec 26 17:31:08 2007  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_fflush): check closed fptr after rb_write_internal to avoid
+	  SEGV on MacOS X.
+
+Wed Dec 26 16:10:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (Init_String): defines chars method.
+
+Wed Dec 26 14:38:43 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* instruby.rb: install goruby if exists.
+
+Wed Dec 26 13:55:02 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb:  Cleaned up some code with Ruby 1.9 idioms.
+
+Wed Dec 26 13:29:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (tmpbuf): use rb_str_tmp_new().
+
+Wed Dec 26 00:57:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/ext/generator/generator.c (Init_generator): requires
+	  json/common.rb for GeneratorError, when static linked.  a patch from
+	  Kenta Murata <muraken AT gmail.com> in [ruby-dev:32789].
+
+Tue Dec 25 23:33:55 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* development version 1.9.0 released.
+
+Tue Dec 25 23:25:29 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/node.rb (REXML::Node::indent): should initialize rv
+	  variable.  a patch from Tadayoshi Funaba <tadf AT dotrb.org> in
+	  [ruby-dev:32783].
+
+Tue Dec 25 23:16:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): encoding option in shebang and RUBYOPT did not
+	  work, do not store alloca()ed string in a parent scope struct.
+
+Tue Dec 25 22:56:52 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): keep this file.
+
+Tue Dec 25 22:55:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (TIMEZONE_VOID): typo.
+
+Tue Dec 25 22:45:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns2vm.rb: add encoding option to shebang.
+
+Tue Dec 25 22:13:51 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/pending.rb: add pending issue.
+
+Tue Dec 25 22:12:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: remove Thread.critical(=).
+
+Tue Dec 25 21:44:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: add version number.
+
+Tue Dec 25 21:32:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix stack consistency error
+	  (break is compiled to throw instead of jump insn).
+	  these problems are reported by Yusuke ENDOH <mame AT tsg.ne.jp>
+
+	* bootstraptest/test_knownbug.rb, test_syntax.rb: move fixed test.
+
+Tue Dec 25 21:32:44 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (struct parser_params): make parser_ruby_sourcefile common
+	  field.  it is used by node_newnode.
+	  new field parser_ruby_sourcefile_string for ripper.
+	  (parser_initialize): initialize parser_ruby_sourcefile in ripper.
+	  (ripper_initialize): initialize parser_ruby_sourcefile_string.
+
+Tue Dec 25 21:26:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (parse.c): depends on tool/ytab.sed.
+
+	* tool/ytab.sed: hack for bison 2.1.
+
+Tue Dec 25 20:24:58 2007  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_ssl.c: Only show a warning if the default
+	  DH callback is actually used.
+
+	* ext/openssl/ossl_rand.c: New method: random_add().
+
+Tue Dec 25 20:24:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot: argument check, and cleanup exported directory.
+
+Tue Dec 25 20:07:13 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* tool/make-snapshot: more portable.
+
+Tue Dec 25 19:01:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.h (rb_enc_mbc_to_codepoint): wrapper for
+	  ONIGENC_MBC_TO_CODE().
+
+	* string.c (rb_str_succ): deal with invalid sequence as binary.
+
+Tue Dec 25 18:40:46 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c: all methods need $SAFE < 1.
+
+	 vm.c: comment out debug functions.
+
+Tue Dec 25 18:37:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (appendline): move RS comparison to rb_io_getline_1().
+
+Tue Dec 25 18:27:51 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_each_line): don't call rb_enc_codepoint with empty
+	  string.
+
+Tue Dec 25 18:06:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_inspect): don't call rb_enc_codepoint with empty
+	  string.  fix '#'.inspect.
+
+	* encoding.c (rb_enc_codepoint): raise on empty string.
+
+Tue Dec 25 17:48:28 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm.c (rb_frame_method_id_and_class): new function to get the
+	  method id and class of the current frame.
+
+Tue Dec 25 17:32:04 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/mkmf.rb (create_makefile): Add a missing dependency on the
+	  target directory for each .rb file.  This will hopefully fix
+	  parallel make (-jN).  Tested on FreeBSD.
+
+Tue Dec 25 16:51:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/japanese.c (rb_{from,to}_{SHIFT_JIS,EUC_JP}): inversed
+	  from_encoding and to_encoding.
+
+Tue Dec 25 16:41:57 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* golf_prelude.rb (h): add new method for all golfers.
+
+Tue Dec 25 16:37:12 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/japanese.c (rb_to_EUC_JP): fixed typo.
+
+Tue Dec 25 16:34:58 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/depend: add dependencies. [ruby-dev:32760]
+
+Tue Dec 25 16:26:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, thread.c: rename is_ruby_native_thread() to
+	  ruby_native_thread_p().
+
+	* ext/tk/tcltklib.c: apply it.
+
+Tue Dec 25 16:15:15 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (clean-enc): clean encoding objects.
+
+Tue Dec 25 16:04:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, goruby.c, golf_prelude.rb: for golfers.
+
+	* main.c (main): hook for embedding applications.
+
+	* tool/compile_prelude.rb: can change initialize function name.
+
+Tue Dec 25 15:59:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_register): do not use based_encoding to check if
+	  dummy encoding.
+
+Tue Dec 25 15:55:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_succ): fix for string with non-alphanumeric chars.
+
+Tue Dec 25 15:42:49 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_external_encoding): should return nil for
+	  pass-through write IO.  [ruby-dev:32740]
+
+Tue Dec 25 15:24:57 2007  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (appendline): initialize rslen to 1 if rsptr is 0.
+	  rslen is the length of the delimiter.
+	  if only delim is given, it should be 1.
+	  [ruby-dev:32746]
+
+Tue Dec 25 15:21:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (transcode_dispatch): fix for multistep transcode.
+
+Tue Dec 25 15:07:51 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/trans/single_byte.c (Init_single_byte): renamed.
+
+Tue Dec 25 15:00:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_yield): when multiple values yielded from #each
+	  pack them into an array.  [ruby-dev:32708]
+
+	* enum.c: all method but all?, any?, one? and none? passed packed
+	  multiple values to the block.
+
+	* enum.c (collect_all): should pack all values.  [ruby-core:14410]
+
+Tue Dec 25 14:57:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (COMMONOBJS): transcode_data_*.c moved under enc/trans.
+
+	* transcode_data.h (rb_transcoding, rb_transcoder): prefixed.
+
+	* transcode.c (rb_register_transcoder, rb_declare_transcoder): split
+	  declaration and registration.  [ruby-dev:32704]
+
+	* transcode.c (transcode_dispatch): autoload pre-declared transcoder.
+
+	* transcode.c (str_transcode): use rb_define_dummy_encoding().
+
+	* transcode.c (Init_transcode): initialize transcoder tables.
+
+	* enc/trans/single_byte.c, enc/trans/japanese.c: moved from top.
+
+Tue Dec 25 14:20:13 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mkmf.rb (map_dir): should generate path including $top_srcdir.
+
+Tue Dec 25 14:09:16 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb:  Fixed test failures caused by changes to Ruby.
+
+	* test/csv/tc_serialization, test/csv/tc_csv_parsing, test/csv/tc_features:
+	  Fixed test failures caused by changes to Ruby.
+
+Tue Dec 25 14:11:57 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_encoding_set): missing return type.
+
+Tue Dec 25 14:03:48 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* test/rinda/test_rinda.rb (MockClock#{_forward, forward, sleep}):
+	  Change default value of n as @reso from nil.  If default value is
+	  nil, n.+ is not defined.
+
+Tue Dec 25 13:54:01 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io_m17n.rb (test_pipe): fixed.
+	  [ruby-dev:32743]
+
+Tue Dec 25 13:44:51 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_wait_fd_rw): should check EBADF on select().
+
+Tue Dec 25 13:30:03 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c, thread_pthread.h, thread_win32.c,
+	  thread_win32.c: make some functions static functions.
+	  a patch from Tadashi Saito <shiba AT mail2.accsnet.ne.jp>
+	  in [ruby-core:14407]
+
+Tue Dec 25 13:23:13 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io_m17n.rb (test_write_noenc): don't mix text and
+	  binary mode.  [ruby-dev:32743]
+
+Tue Dec 25 13:13:09 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* README.EXT.ja, dir.c, eval.c, eval_intern.h, lex.c.src,
+	  lex.c.blt, keywords, load.c, thread.c: more ANSI'ize.
+	  a patch from Tadashi Saito <shiba AT mail2.accsnet.ne.jp>
+	  in [ruby-dev:32725]
+
+Tue Dec 25 13:07:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, thread.c, cont.c: add RUBY_VM_SET_INTERRUPT(),
+	  RUBY_VM_SET_TIMER_INTERRUPT(), RUBY_VM_INTERRUPTED().
+
+	* thread.c, thread_pthread.c, thread_win32.c: fix to ignore time slice
+	  event until sleep.
+
+	* bootstraptest/test_thread.rb: add a test for time limited join test.
+
+Tue Dec 25 12:42:59 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (Init_VM): remove unused code.
+	  [ruby-dev:32732]
+
+Tue Dec 25 12:32:32 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c: Moving a static counter from inside register_transcoder()
+	  and register_functional_transcoder() to outside the functions, renaming
+	  from n to next_transcoder_position. Fixes 3) in [ruby-dev:32715].
+
+Tue Dec 25 12:22:17 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* sample/from.rb: follow Ruby 1.9 libraries.
+
+Tue Dec 25 12:21:56 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (method_inspect): preserve encoding of the method name.
+
+Tue Dec 25 12:07:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (BASERUBY): delayed error until BASERUBY is used.
+
+Tue Dec 25 11:48:35 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/README: removed obsoleted files: dbmtest.rb,
+	  getopts.test, mrshtest.rb, regx.rb.
+
+Tue Dec 25 11:45:34 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb:  Import the FasterCSV source as the new CSV class.
+
+	* test/csv/*:  Added all applicable tests from FasterCSV.
+
+Tue Dec 25 11:33:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (report_bug): uses ruby_description.
+
+Tue Dec 25 11:20:38 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix stack consistency error.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>  [ruby-dev:32720]
+
+	* bootstraptest/test_syntax.rb: add 2 tests for above.
+
+Tue Dec 25 11:14:20 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c, vm_core.h: comment out unused fields.
+
+Tue Dec 25 11:02:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: check frame is FINAL when creating env.
+	  [ruby-core:14395]
+
+	* bootstraptest/test_block.rb: add a test for above.
+
+Tue Dec 25 09:12:13 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/:  Enable RDoc debugging only with $DEBUG_RDOC.
+
+Tue Dec 25 08:37:43 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/csv.rb, test/csv/test_csv.rb:  Removed in preparation for
+	  FasterCSV code import.
+
+Tue Dec 25 08:27:43 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Fix test failures.
+
+	* test/rubygems/test_gem.rb:  Fix test failure.
+
+Tue Dec 25 06:23:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb, test_literal.rb: move fixed test.
+
+Tue Dec 25 06:19:04 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* sample/biorhythm.rb: follow Ruby 1.9 libraries.
+
+Tue Dec 25 06:15:01 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: add dummy toplevel frame.
+
+Tue Dec 25 05:44:56 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Fix uninitialized variable warning.
+	  [ruby-talk:284582]
+
+	* lib/irb/output-method.rb:  Remove unused #foo method.
+	  [ruby-talk:284582]
+
+Tue Dec 25 05:24:12 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile): clear local table if node == 0.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>  [ruby-dev:32530]
+
+	* vm.c: clear VM stack.
+
+Tue Dec 25 04:23:32 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (rb_id2str): fill klass of returned string as rb_cString.
+	  some strings are allocated before rb_cString is created.
+	  This prevents a "called on terminated object" error by
+	  ObjectSpace.each_object(Module) {|m| p m.name }.
+
+Tue Dec 25 03:51:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix stack consistency bug.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>
+
+Tue Dec 25 03:19:47 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* tool/make-snapshot: must create configure and lex.c.
+
+Tue Dec 25 03:16:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_s_pipe): now takes up to two arguments.  allow its
+	  external/internal encoding by Encoding objects.
+
+	* io.c (rb_io_set_encoding): new method to set encoding of the IO.
+
+	* io.c (argf_set_encoding): ditto.
+
+Tue Dec 25 03:08:53 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* pack.c (pack_pack): use NUM2LONG instead of NUM2INT.
+
+	* numeric.c (fix_lshift, fix_aref): use SIZEOF_LONG instead of
+	  SIZEOF_VALUE.
+
+	* bignum.c (big2ulong, rb_big_aref): ditto.
+
+Tue Dec 25 02:55:26 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/rexml/element.rb (REXML::Elements#each): yield in each
+	  should be called with one parameter. [ruby-dev:32708]
+
+Tue Dec 25 02:15:39 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): add a "pop" insn after break
+	  to fix stack consistency error.  [ruby-core:14385]
+
+	* bootstraptest/test_syntax.rb: add tests for above.
+
+	* bootstraptest/test_knownbug.rb: remove fixed bug.
+
+Tue Dec 25 01:54:36 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* id.c (Init_id): remove several unused symbols. [ruby-core:14362]
+
+	* compile.c (iseq_specialized_instruction): do not use
+	  VM_CALL_SEND_BANG flag any longer.
+
+Tue Dec 25 01:42:41 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rdoc/rdoc.rb (parse_files): interpret coding cookie.
+
+Tue Dec 25 01:38:04 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (method_name): preserve Symbol's encoding.
+
+	* numeric.c (fix_id2name): ditto.
+
+Tue Dec 25 01:19:18 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_enc_left_char_head): new utility macro.
+
+	* include/ruby/encoding.h (rb_enc_right_char_head): ditto.
+
+	* io.c (appendline): does multibyte RS search in the function.
+
+	* io.c (prepare_getline_args): RS may be nil.
+
+	* io.c (rb_io_getc): should process character based on external
+	  encoding, when transcoding required.
+
+Tue Dec 25 01:07:57 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/irb/output-method.rb: translate a comment to English to
+	  avoid mix of EUC-JP comment and UTF-8 Date keyword.
+	  svn substitute Date keyword with UTF-8 weekday on UTF-8 locale.
+
+Tue Dec 25 00:27:28 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservley/cgihandler.rb
+	  (WEBrick::HTTPServlet::CGIHandler#do_GET): m17nized.
+
+Mon Dec 24 23:55:29 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/cgi.rb (CGI::escape): m17nized.
+	  (CGI::unescape): ditto.
+	  (CGI::escapeHTML): ditto.
+	  (CGI::unescapeHTML): ditto.
+
+Mon Dec 24 23:32:24 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* transcode_data_japanese.c (select_iso_2022_mode): '\e' is not valid.
+
+Mon Dec 24 23:13:09 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/rdoc/diagram.rb (RDoc::Diagram#initialize): use fileuitls
+	  instead of ftools.
+
+Mon Dec 24 23:04:57 2007  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/ftools.rb: removed obsoleted lib.  use fileutils instead (by eban).
+
+	* lib/rdoc/rdoc.rb, lib/rdoc/generators/*, lib/rake.rb: let it use
+	  fileutils instead of ftools.
+
+	* lib/shell/command-processor.rb: removed unused references to ftools.
+
+	* lib/parsedate.rb: removed.  see [ruby-core:12535], [ruby-dev:31969].
+
+	* lib/README: updated.
+
+Mon Dec 24 23:01:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI::Buffer): use Meta ===.  [ruby-core:14295]
+
+Mon Dec 24 22:46:42 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* transcode.c: register_functional_transcoder() added.
+	  (init_transcoder_table(: register ISO-2022-JP.
+	  (str_transcode): add preprocessor and postprocessor.
+
+	* transcode_data_japanese.c: add ISO-2022-JP support.
+
+	* transcode_data.h: moved transcoder and transcoding definition from
+	  transcode.c.
+
+Mon Dec 24 20:29:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/io/nonblock/test_flush.rb: fix test for 1.9.
+
+Mon Dec 24 20:23:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/rinda/test_rinda.rb: revert last commit because this test seems
+	  to have timing problem to halt all tests.
+
+Mon Dec 24 20:18:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/rinda/test_rinda.rb: enable rinda test.
+
+Mon Dec 24 20:16:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* instruby.rb: fix rdoc install dir.
+
+Mon Dec 24 18:37:32 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_prepare_re): show regexp encoding in the error message.
+
+Mon Dec 24 18:23:32 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (rb_exc_raise): ANSI style.
+	  (rb_exc_fatal): ditto.
+	  (rb_raise_jump): ditto.
+	  (rb_jump_tag): ditto.
+	  (rb_block_given_p): ditto.
+
+	* variable.c (original_module): ditto.
+
+Mon Dec 24 18:05:09 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (Init_ISeq): disable ISeq.load() because there is no verifier.
+
+	* iseq.c, proc.c: add ISeq.disasm(method).
+
+Mon Dec 24 18:06:03 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval_method.c (Init_eval_method): extracted from Init_eval
+	  for rdoc to find rb_mod_remove_method, rb_mod_undef_method and
+	  rb_mod_alias_method.
+
+	* eval.c (Init_eval): call Init_eval_method.
+
+Mon Dec 24 17:59:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_lock): reverted.
+
+	* thread.c (rb_barrier_wait): check for recursive wait.
+
+Mon Dec 24 17:50:54 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (function_call_may_return_twice_jmp_buf): removed.
+	  (function_call_may_return_twice_false): removed.
+	  [ruby-core:14335]
+
+Mon Dec 24 17:40:57 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (node_name.inc, miniprelude.c, prelude.c): nmake cannot
+	  handle target vpath in other than implicit rules.
+
+Mon Dec 24 17:20:34 2007  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/{mailread.rb,getopts.rb,parsearg.rb}: removed.
+	  see [ruby-core:12535], [ruby-dev:31969].
+
+Mon Dec 24 17:12:57 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h, random.c, array.c:
+	  change exported name.
+	  genrand_int32 -> rb_genrand_int32.
+	  genrand_real -> rb_genrand_real.
+	  [ruby-core:14335]
+
+Mon Dec 24 17:06:37 2007  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* {lib,test}/{soap,wsdl,xsd}: removed soap4r along to the discussion
+	  at ruby-core and ruby-dev.  see [ruby-core:12535], [ruby-dev:31969].
+
+Mon Dec 24 17:06:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_feature_p): returns loading path name too.
+
+	* load.c (search_required): returns path too if feature is being
+	  loaded.  [ruby-dev:32048]  [TODO: refactoring]
+
+Mon Dec 24 16:29:12 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* sample/openssl: reviewed and remove dependency on getopts.rb.
+
+Mon Dec 24 16:18:57 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* mkconfig.rb: add teeny to CONFIG['ruby_version'].
+
+Mon Dec 24 15:55:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/compile.rb, getrev.rb, runruby.rb: remove unused tools.
+
+Mon Dec 24 15:42:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, cygwin/GNUmakefile.in, */Makefile.sub,
+	  djgpp/config.hin.: version dependent directory names now contain
+	  teeny.
+
+Mon Dec 24 15:29:13 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/serb.rb: removed.
+
+Mon Dec 24 13:55:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_proc_new), vm.c (invoke_block): removed u3.state magic.
+	  [ruby-core:14310]
+
+	* test/ruby/test_symbol.rb (test_to_proc): a test from Frederick
+	  Cheung <frederick.cheung AT gmail.com>.
+
+Mon Dec 24 13:43:36 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_atfork): should not leave living_threads
+	  referring freed table while allocating new table.
+
+Mon Dec 24 12:49:54 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, configure.in, lib/mkmf.rb, */Makefile.sub: specify
+	  compiled output file name explicitly.
+
+	* enc/Makefile.in, enc/depend: now makes compiler to put generated
+	  files under directories corresponding to the each source.
+	  enc/trans supported.
+
+	* enc/make_encmake.rb: evaluates depend file before Makefile.in so
+	  that the former can influence to CONFIG.
+
+Mon Dec 24 12:35:03 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win{32,ce}/Makefile.sub (MFLAGS): defaulted to -l.
+
+Mon Dec 24 12:08:10 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* /, ext/:  Add svn:ignore for OS X Xcode 3's conftest.dSYM
+	  directories.
+
+Mon Dec 24 11:56:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: should not pass MAKEFLAGS to recursive make.
+	  + normal make: MFLAGS are set and command line options and macros
+	    are all passed silently.
+	  + GNU make: ditto, and all options and macros in MAKEFLAGS are in
+	    effect.
+	  + nmake: MFLAGS is not set and MAKEFLAGS has only options without
+	    hyphen, no macros exist in any variables.
+	  + Borland make: ditto, and command line macros cannot override
+	    macros in makefile, so passing them is vain.
+
+	* {bcc32,win{32,ce}}/Makefile.sub (SET_MAKE): set MFLAGS which is not
+	  set by default, to get rid of chaotic situation of MFLAGS/MAKEFLAGS.
+
+Mon Dec 24 11:32:44 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_comparable): fixed to keep transitivity.
+	  [ruby-dev:32693]
+
+Mon Dec 24 11:20:31 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/ri_options.rb:  Fix display of GEMDIRS, make command
+	  examples match ri's name.
+
+	* lib/rdoc/ri/ri_paths.rb:  Only allow latest ri dirs in ri output.
+
+Mon Dec 24 10:49:04 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/uri/mailto.rb, lib/uri/common.rb:  Fix Regexp warnings.  Patch
+	  #16524 from Kornelius Kalnbach, [ruby-core:14302].
+
+Mon Dec 24 10:37:38 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* gem_prelude.rb:  Remove methods from Gem, not QuickLoader, to fix
+	  warnings.
+
+Mon Dec 24 09:45:45 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c, transcode_data_one_byte.c, transcode_data_japanese.c:
+	  added rb_ prefix to external data symbols.
+
+Mon Dec 24 05:32:22 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_inject): updated documentation.  a patch from Keita
+	  Yamaguchi <keita.yamaguchi AT gmail.com> in [ruby-dev:32686].
+
+	* README.EXT: updated.  a patch from Tadashi Saito
+	  <shiba AT mail2.accsnet.ne.jp> in [ruby-core:14328].
+
+	* array.c (rb_ary_at): updated documentation. a patch from Tadashi
+	  Saito <shiba AT mail2.accsnet.ne.jp> in [ruby-core:14330].
+
+Mon Dec 24 05:13:04 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (tr_trans): should associate new encoding if modified.
+
+Mon Dec 24 04:04:12 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/net/http/test_https.rb: should rescue LoadError.
+
+Mon Dec 24 03:57:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c, vm.h: fix to support sparc machine.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>
+
+Mon Dec 24 03:35:19 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: remove additional "-".
+
+Mon Dec 24 02:59:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_check_readable): should not fill fptr->enc always.
+	  read-write IO (e.g. socket) does not work.  [ruby-dev:32685]
+
+	* io.c (io_read_encoding): retrieve reading encoding.
+
+	* io.c (prepare_getline_args): convert RS to external encoding.
+
+	* string.c (str_new_shared): was setting embedding flag of wrong
+	  string object.  [ruby-dev:32685]
+
+	* io.c (io_enc_str): should preserve default_external encoding.
+
+	* io.c (appendline): should do multibyte aware RS search.
+
+Mon Dec 24 02:06:35 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_open): documentation update.
+
+	* io.c (rb_io_s_pipe): ditto.
+
+	* io.c (io_fwrite): wrong encoding destination.
+
+	* io.c (rb_io_external_encoding): should return the encoding of
+	  the file reading.
+
+	* io.c (rb_io_internal_encoding): should return the encoding of
+	  read string.
+
+Mon Dec 24 01:46:43 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_s_pipe): allow specifying read-side encoding.
+
+	* io.c (io_enc_str): wrong encoding destination.
+
+Mon Dec 24 01:03:17 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_comparable): comparison including broken
+	  coderange strings do not consider encoding.
+
+Mon Dec 24 00:57:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (open_key_args): IO direct methods (foreach, readlines,
+	  read) now takes keyword argument: encoding, mode, open_args.
+
+Mon Dec 24 00:52:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_s_read): encoding argument reverted.
+
+	* io.c (mode_enc): independent function to share code.
+
+	* io.c (rb_io_internal_encoding): new method.
+
+Mon Dec 24 00:47:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb (TestBeginEndBlock::test_endblockwarn):
+	  rename endblockwarn.rb to endblockwarn_rb to avoid unnecessary
+	  warning in make test.
+
+Sun Dec 23 23:03:13 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_codepoint): implemented to raise invalid
+	  encoding.
+
+	* include/ruby/encoding.h (rb_enc_codepoint): macro is replaced as a
+	  declaration.
+
+Sun Dec 23 19:45:22 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.httpdate): fix 2 digits year for 20xx.
+	  reported by Tadayoshi Funaba.  [ruby-dev:32687]
+
+Sun Dec 23 19:33:42 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/open-uri.rb:  Fix method redefined warning.  [ruby-core:14304]
+
+Sun Dec 23 18:31:49 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_enc_get): use rb_define_dummy_encoding.
+
+	* ext/nkf/nkf.c (Init_nkf): use rb_nkf_enc_get("ASCII").
+
+	* ext/nkf/nkf-utf8/nkf.c: Update 1.161.
+
+	* ext/nkf/nkf-utf9/config.h: default output encoding is now UTF-8.
+
+	* ext/nkf/lib/kconv.rb (Kconv.kconv): replace Encoding#name by
+	  Encoding#to_s.
+
+Sun Dec 23 18:02:52 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/gem_open_uri.rb:  Fix version check.
+
+Sun Dec 23 17:24:48 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_init): add eucJP as an alias of EUC-JP.
+
+Sun Dec 23 17:00:23 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.httpdate): use Time.utc for
+	  "day-of-week, dd-mon-yy HH::MM:SS GMT" format.
+
+Sun Dec 23 16:12:40 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Fix 1.9 warnings.
+
+Sun Dec 23 15:28:37 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems, test/rubygems:  Fix new 1.9 warnings.
+
+Sun Dec 23 14:43:10 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* gem_prelude.rb:  Use require to load rubygems.rb so the correct path
+	  is in $LOADED_FEATURES on RubyGems upgrade.
+
+Sun Dec 23 11:26:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_call): use exact argument array interface.
+	  [ruby-core:14279]
+
+Sun Dec 23 11:01:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_binmode_m): removed C99ism.
+
+Sun Dec 23 10:23:23 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode_data_one_byte.c: Better (and more honest) optimization.
+
+	* transcode_data_japanese.c: First optimization step.
+
+Sun Dec 23 09:07:02 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h, encoding.c, re.c, io.c, parse.y, numeric.c,
+	  ruby.c, transcode.c, ext/nkf/nkf.c: rename rb_ascii_encoding to
+	  rb_ascii8bit_encoding.  rb_ascii_encoding is ambiguous with
+	  ASCII-8BIT and US-ASCII.
+
+Sun Dec 23 03:35:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* version.h: use patchlevel if revision is not set.
+
+	* {bcc32,win{32,ce}}/setup.mak (-version-): skip including revision.h.
+
+	* common.mk (srcs): new target to generated sources.
+
+	* common.mk (encs, ext/ripper/ripper.c): MAKEFLAGS needs -.
+
+	* enc/depend, enc/make_encmake.rb: use erb.
+
+Sun Dec 23 01:56:18 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_mode_enc): do not set encoding unless explicitly
+	  specified.
+
+	* io.c (rb_io_check_readable): fill fptr->enc by default_external
+	  if it's empty.
+
+	* io.c (io_enc_str): fptr->enc is always set for reading IO (by
+	  rb_io_check_readable(fptr)).
+
+Sun Dec 23 01:18:06 2007  David Flanagan  <david@davidflanagan.com>
+
+	* io.c, io.h: temporary patch to partially implement
+	  transcode-on-read and transcode-on-write
+
+Sun Dec 23 00:48:05 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/net/imap/test_imap.rb: added tests for SSL.
+
+Sat Dec 22 21:10:53 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_genericresolver_node_import): should
+	  not set instance variable "@kind" before initializing it.
+	  [ruby-dev:32677]
+
+Sat Dec 22 19:52:23 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_objectspace.rb: fix condition.
+
+Sat Dec 22 19:17:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/probeprofiler/: removed.
+
+Sat Dec 22 19:14:38 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* process.c (rb_f_fork): Unsupport Kernel.fork() on NetBSD.
+
+Sat Dec 22 18:20:13 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (version.o): add dependency. [ruby-dev:32680]
+
+Sat Dec 22 17:45:11 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c (Init_nkf): use rb_ascii_encoding() for
+	  rb_nkf_enc_get("US-ASCII").
+	  * if use rb_nkf_enc_get("US-ASCII"), ruby will crash - this is bug?
+
+Sat Dec 22 17:39:03 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb ($extmk): fixed broken condition.
+
+Sat Dec 22 17:35:59 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8.c: Update nkf.c rev:1.157.
+
+	* ext/nkf/nkf.c (rb_nkf_enc_get): replicate proper based encoding.
+
+	* ext/nkf/kconv.c (Kconv#kconv, to*): use self.encoding as from_enc
+	  when from_enc isn't given.
+
+Sat Dec 22 17:06:50 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/net/ssl.rb (OpenSSL::SSL::SSLContext.build): removed.
+
+	* ext/openssl/lib/net/ssl.rb (OpenSSL::SSL::SSLContext#set_params):
+	  new method to set suitable SSL parameters.
+
+	* lib/net/pop.rb, lib/net/http.rb, lib/net/imap.rb,
+	  test/openssl/test_ssl.rb: follow above change.
+
+	* test/net/http/test_https.rb: refine error case.
+
+Sat Dec 22 16:58:49 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (encode_utf7): accept UTF-8 strings.
+
+	* lib/net/imap.rb (decode_utf7): return UTF-8 strings.
+
+Sat Dec 22 15:56:36 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* transcode_data_japanese: typo.
+
+Sat Dec 22 15:54:54 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* ChangeLog: Information for last patch got lost, fixed
+
+	* test/ruby/test_transcode.rb: Added simple tests for
+	  EUC-JP and Shift_JIS and tests for ASCII-only range
+
+Sat Dec 22 15:45:45 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode_data_one_byte: slightly optimized
+
+	* transcode_data_japanese: new data file for EUC-JP and SHIFT_JIS
+	  (not yet optimized; tests to follow; data from
+	  http://nkf.sourceforge.jp/ucm/{SJIS|eucJP}-nkf.ucm)
+
+	* common.mk, transcode.c: Adjusted for transcode_data_japanese
+
+Sat Dec 22 15:30:13 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* */Makefile.sub (MFLAGS): define unless defined.
+
+Sat Dec 22 15:17:40 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb ($extmk): set true only when under ext/ or tool/.
+
+Sat Dec 22 15:14:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (encs, ext/ripper/ripper.c): needs MFLAGS.
+
+	* configure.in (STRINGIZE): stringizing macro.
+
+	* include/ruby/defines.h (STRINGIZE): fallback.
+
+	* tool/make-snapshot: new file.
+
+	* version.c (ruby_description, ruby_copyright): string constants for
+	  -v option.
+
+Sat Dec 22 15:03:37 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (initialize): fixed documentation.
+
+Sat Dec 22 15:01:16 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/pop.rb (enable_ssl): use OpenSSL::SSL::SSLContext.build
+	  instead of SSLContext.new (default verify mode is now
+	  OpenSSL::SSL::VERIFY_PEER).
+
+Sat Dec 22 14:45:21 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: shouldn't freeze nil. [ruby-dev:32677]
+
+Sat Dec 22 14:27:27 2007  Tanaka Akira  <akr@fsij.org>
+
+	* regenc.c (onigenc_ascii_is_code_ctype): moved from enc/ascii.c.
+
+	* regenc.h (onigenc_ascii_is_code_ctype): declared.
+
+	* enc/ascii.c: use onigenc_ascii_is_code_ctype.
+
+	* enc/us_ascii.c: new file for US-ASCII.
+
+Sat Dec 22 14:30:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_named_capture_assign_iter): allows non-ascii names and
+	  get rid of reserved word IDs.
+
+Sat Dec 22 14:18:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_float.rb, test_sprintf.rb: fix test place.
+
+Sat Dec 22 14:17:02 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_objectspace.rb: skip frozen string.
+
+Sat Dec 22 14:02:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_permutation): add volatile to avoid GC problem.
+
+Sat Dec 22 11:47:42 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_obj_instance_eval): use class of immediate objects.
+
+	* test/ruby/test_eval.rb: fix a test.
+
+Sat Dec 22 11:37:06 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* encoding.c (rb_locale_charmap): win32 support.
+
+Sat Dec 22 11:31:14 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* gem_prelude.rb:  Place bin dir before lib dir so gem bin stubs work.
+
+Sat Dec 22 11:05:44 2007  Jim Weirich  <jim@tardis.local>
+
+	* lib/rake.rb (Rake): Added Rake and related libraries to the
+	  source code base.
+
+Sat Dec 22 10:30:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/insns2vm.rb: moved from lib/vm/instruction.rb.
+
+Sat Dec 22 10:25:44 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (reg_named_capture_assign_iter): captured name should
+	  not be reserved word.  a patch from Keita Yamaguchi
+	  <keita.yamaguchi AT gmail.com> in [ruby-dev:32675].
+
+	* parse.y (reg_named_capture_assign_iter): just ignore the
+	  captures that do not have valid local variable name.
+
+Sat Dec 22 10:19:08 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_method_name): now __method__ and __callee__ are
+	  aliases.  [ruby-core:14244]
+
+Sat Dec 22 08:29:56 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_buf_append): improvement for non-broken coded
+	  strings.
+
+Sat Dec 22 06:30:04 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_fork.rb: skip if fork is not unsupported.
+
+	* bootstraptest/test_io.rb: skip if require failed.
+
+Sat Dec 22 06:09:12 2007  David Flanagan  <david@davidflanagan.com>
+
+	* io.c: fix typo in rdoc comment
+
+Sat Dec 22 05:09:43 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (parser_str_new, rb_intern3): rb_default_encoding() renamed.
+
+	* ext/nkf/nkf.c (rb_nkf_putchar): ditto.
+
+Sat Dec 22 03:54:18 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_ascii_encoding): renamed from previous
+	  rb_default_encoding().
+
+Sat Dec 22 02:49:02 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (command): block from cmd_brace_block was ignored.
+	  [ruby-dev:32644]
+
+	* re.c (rb_reg_prepare_re): stop ENCODING_NONE warning if the
+	  encoding of the str is ASCII-8BIT.
+
+Sat Dec 22 01:52:11 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_ungetc): avoid buffer relocation, which might cause
+	  serious problem under concurrent situation.
+
+Sat Dec 22 01:35:41 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (ARG_ENCODING_NONE): defined for /.../n option.
+	  (REG_ENCODING_NONE): ditto.
+	  (rb_char_to_option_kcode): return ARG_ENCODING_NONE for n.
+	  (rb_reg_prepare_re): warn /ascii/n =~ "non-ascii".
+	  (rb_reg_initialize): set REG_ENCODING_NONE from ARG_ENCODING_NONE.
+
+Sat Dec 22 01:23:10 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/json/test_json_addition.rb (test_core): do not use Time.now
+	  because JSON can't hold nsec.
+
+Sat Dec 22 01:10:30 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/tk/sample/tkextlib/vu/canvSticker2.rb,
+	  ext/tk/sample/demos-{en,jp}/bind.rb: fix typo. [ruby-dev:32668]
+
+Sat Dec 22 00:56:43 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/test.rb (valid_syntax): force_encoding input script.
+
+Fri Dec 21 23:48:38 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (depend_rules): suffixes list broken. fixed.
+
+Fri Dec 21 20:18:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_mul0): remove unused variable.
+
+	* bignum.c (bigdivrem): ditto.
+
+Fri Dec 21 20:13:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_catch_obj, rb_throw_obj): prototyped.
+
+	* include/ruby/intern.h (rb_fiber_alive_p): prototyped.
+
+Fri Dec 21 20:09:18 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_succ): retry increasing until valid char is found.
+
+Fri Dec 21 20:00:02 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_replicate): now creates first class encoding.
+
+	* encoding.c (rb_define_dummy_encoding): always based on the default
+	  encoding.
+
+	* encoding.c (rb_enc_dummy_p): check if dummy.
+
+	* encoding.c (enc_inspect): shows if dummy.
+
+	* encoding.c (Init_Encoding): added dummy? method
+
+	* include/ruby/encoding.h (ENCODING_INLINE_MAX): increased.
+
+Fri Dec 21 18:40:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* io.c: write() should be in blocking region.
+
+	* bootstraptest/test_io.rb, test_knownbug.rb: move a fixed test.
+
+Fri Dec 21 17:56:30 2007  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: provisional support on Ruby-VM.
+
+	* ext/tk/MANUAL_tcltklib.eng, ext/tk/MANUAL_tcltklib.eucj:
+	  modify document about new functions.
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tk/labelframe.rb,
+	  ext/tk/lib/tk/frame.rb, ext/tk/lib/tk/toplevel.rb,
+	  ext/tk/lib/tk/scrollbar.rb, ext/tk/lib/tk/message.rb,
+	  ext/tk/lib/tk/listbox.rb, ext/tk/lib/tk/text.rb,
+	  ext/tk/lib/tk/scale.rb, ext/tk/lib/tk/entry.rb,
+	  ext/tk/lib/tk/ttk_selector.rb, ext/tk/lib/tk/menu.rb,
+	  ext/tk/lib/tk/label.rb, ext/tk/lib/tk/spinbox.rb,
+	  ext/tk/lib/tk/textmark.rb, ext/tk/lib/tk/winpkg.rb,
+	  ext/tk/lib/tk/checkbutton.rb, ext/tk/lib/tk/panedwindow.rb,
+	  ext/tk/lib/tk/texttag.rb, ext/tk/lib/tk/root.rb,
+	  ext/tk/lib/tk/textimage.rb, ext/tk/lib/tk/radiobutton.rb,
+	  ext/tk/lib/tk/package.rb, ext/tk/lib/tk/macpkg.rb,
+	  ext/tk/lib/tk/composite.rb, ext/tk/lib/tk/autoload.rb,
+	  ext/tk/lib/tk/canvas.rb, ext/tk/lib/tk/button.rb,
+	  ext/tk/lib/tk/textwindow.rb,
+	  ext/tk/lib/tkextlib/iwidgets/scrolledtext.rb,
+	  ext/tk/lib/tkextlib/tile/style.rb,
+	  ext/tk/lib/tkextlib/tile/tscrollbar.rb,
+	  ext/tk/lib/tkextlib/tile/tpaned.rb, ext/tk/lib/tkextlib/tile.rb,
+	  ext/tk/extconf.rb: support Tcl/Tk8.5 (partial, not complete).
+
+	* ext/tk/sample/demos-jp/widget,
+	  ext/tk/sample/demos-jp/pendulum.rb,
+	  ext/tk/sample/demos-jp/bind.rb,
+	  ext/tk/sample/tkextlib/vu/canvSticker2.rb,
+	  ext/tk/sample/demos-en/pendulum.rb,
+	  ext/tk/sample/demos-en/bind.rb: remove $KCODE and minor bug fix.
+
+Fri Dec 21 17:49:06 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_str_transcode_bang): returns self if no conversion.
+	  [ruby-dev:32662]
+
+Fri Dec 21 17:44:47 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*:  Update to RubyGems 1.0.1, r1581
+
+Fri Dec 21 17:32:49 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/pending.rb: renamed from featurebug.rb.
+	  This file contains bugs which is known but will not be
+	  fixed in days.
+
+Fri Dec 21 17:31:22 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* debug.c (dummy_gdb_enums.various): added ENCODING and CODERANGE
+	  constants.
+
+	* .gdbinit: use enum constants.
+
+Fri Dec 21 17:28:17 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/featurebug.rb: added.
+
+	* bootstraptest/test_knownbug.rb: move a feature bug.
+
+Fri Dec 21 17:25:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_atfork): fix to mark thread object.
+	  [ruby-dev:32404]
+
+	* bootstraptest/test_knownbug.rb, test_fork.rb: move a fixed test.
+
+Fri Dec 21 17:07:13 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.h: extern variable should not be initialized.
+
+	* thread_pthread.c: add a parameter.
+
+Fri Dec 21 16:50:43 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (Init_Encoding): use enc_name as to_s.
+	  (enc_inspect): renamed from enc_to_s.  add "#" at beginning.
+
+Fri Dec 21 16:37:43 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/config.h (MIME_DECODE_DEFAULT, X0201_DEFAULT):
+	 defined as FALSE.  nkf and kconv don't decode MIME encoded string
+	 and don't convert JIS X 0201 Katakana.
+
+	* test/nkf/test_kconv.rb: fix tests.
+
+Fri Dec 21 16:33:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_iter_i): adjusted for rb_block_call_func.
+
+	* include/ruby/ruby.h (rb_block_call_func): function to be called back
+	  as block.
+
+Fri Dec 21 16:25:25 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk, transcode_data_iso_8859.c: renamed to
+	  transcode_data_one_byte.c.
+
+Fri Dec 21 16:10:30 2007  Shigeo Kobayashi  <shigeo@tinyforest.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpMidRound): Round method bug
+	  pointed by Ryan Platte fixed(Patch to the patch from "NATORI
+	  Shin").  [ruby-talk:273360]
+
+Fri Dec 21 16:06:13 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (append_utf8): use rb_utf8_encoding() instead of
+	  rb_enc_find("utf-8").
+
+Fri Dec 21 15:59:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_init): use enc_register_at() directly.
+
+	* encoding.c (rb_utf8_encoding): returns utf-8 encoding.
+
+	* include/ruby/encoding.h (rb_utf8_encoding): prototyped.
+
+	* parse.y (UTF8_ENC): uses rb_utf8_encoding().
+
+Fri Dec 21 15:31:59 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_s_read): allow specifying encoding explicitly.
+
+	* io.c (rb_io_binmode): specifies encoding to ASCII-8BIT (binary).
+
+	* io.c (rb_io_s_read): IO should be in binary mode when offset is
+	  specified.
+
+	* encoding.c (rb_to_encoding): returns default encoding if no
+	  corresponding encoding found.
+
+Fri Dec 21 15:24:22 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (initialize): accept service name.  changed
+	  the default value of the old style +verify+ argument to true.
+
+Fri Dec 21 15:15:44 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (rb_garbage_collect): new function for debugging.
+
+Fri Dec 21 15:16:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_to_encoding_index): should return error instead of
+	  exception even if type is incorrect.
+
+Fri Dec 21 14:58:27 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_init_copy): prohibit cloning of
+	  generators since Fibers cannot be copied.
+
+	* enumerator.c (enumerator_init_copy): typo fixed.
+
+Fri Dec 21 14:46:07 2007  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (Init_IO): define IO::BINARY even if O_BINARY is not exist.
+
+Fri Dec 21 14:01:14 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_system.rb (TestSystem::valid_syntax): apply
+	  ASCII-8BIT encoding explicitly.
+
+	* re.c (rb_reg_prepare_re): add encoding name in the message.
+
+Fri Dec 21 13:54:05 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c: change "character encodings differ" error messages.
+
+Fri Dec 21 13:46:58 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_register): set encoding constant.
+
+	* encoding.c (rb_enc_find_index): replace non-alphanumeric chars with
+	  underscores, so that initialize function can be called.
+
+	* encoding.c (rb_enc_find_index): extension libraries have lower case
+	  names conventionally.
+
+	* ruby.c (proc_options, process_options): finds encoding after
+	  load_path is initialized.
+
+Fri Dec 21 13:10:57 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_external_encoding): new method.
+
+	* encoding.c (rb_enc_from_encoding): returns Qnil for NULL
+	  encoding.
+
+	* io.c (rb_io_external_encoding): should fill delayed
+	  initialization for STDIN.
+
+Fri Dec 21 13:09:11 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_locale_charmap): return nil if no locale information.
+
+Fri Dec 21 12:55:39 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/runit, lib/rubyunit.rb, test/testunit/runit: removed.
+
+Fri Dec 21 12:45:49 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_chomp_bang): avoid unnecessary loop using
+	  ONIGENC_LEFT_ADJUST_CHAR_HEAD().
+
+Fri Dec 21 12:32:08 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (Init_File): File.exists? revived.
+
+	* dir.c (Init_Dir): Dir.exists? again.
+
+Fri Dec 21 12:26:36 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: remove "illegal".
+
+Fri Dec 21 12:22:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* keywords, parse.y (__ENCODING__): represent script encoding.
+
+Fri Dec 21 12:16:50 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_is_ascii_only_p): new method ascii_only?.
+
+Fri Dec 21 12:11:57 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_mode_enc): set default external encoding if no
+	  encoding specified explicitly.
+
+Fri Dec 21 12:00:34 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: check langinfo.h and locale.h.
+
+	* encoding.c: use langinfo.h only if available.
+
+	* main.c: use locale.h only if available.
+
+Fri Dec 21 11:47:56 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c: include locale.h
+	  (rb_locale_charmap): new method Encoding.locale_charmap for
+	  nl_langinfo(CODESET).
+
+	* include/ruby/encoding.h (rb_locale_charmap): declared.
+
+	* main.c (main): call setlocale with LC_CTYPE.
+
+	* ruby.c (locale_encoding): use rb_locale_charmap.
+
+Fri Dec 21 11:35:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, vm_dump.c: fix typo.  Reported by Yuki Mitsui.
+
+Fri Dec 21 11:28:00 2007  Tanaka Akira  <akr@fsij.org>
+
+	* regerror.c, string.c, io.c, lib/getoptlong.rb, lib/net/imap.rb,
+	  compile.c, sprintf.c, parse.y, ext/win32ole/win32ole.c,
+	  ext/tk/sample/demos-en/entry3.rb, ext/tk/lib/tcltk.rb,
+	  ext/openssl/ossl_bn.c, numeric.c, vm.c,
+	  benchmark/bm_so_meteor_contest.rb, bignum.c, ruby.c: don't "illegal"
+	  for non law violation context.
+
+Fri Dec 21 11:23:24 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (enc.mk): depends on $(RBCONFIG) instead of rbconfig.rb.
+
+	* encoding.c (Init_Encoding): ISO-8859-1 is no longer a replica.
+
+	* regenc.h (OnigEncodingDefine): names of extension and encoding can
+	  differ.
+
+	* enc/Makefile.in: always shared.
+
+	* enc/depend (deffile): should not upcase.
+
+	* enc/{ascii,euc_jp,sjis,utf8,iso_8859_{1..16}}.c: fix for Init.
+
+Fri Dec 21 09:26:48 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb: use erb.
+
+Fri Dec 21 08:07:35 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_next): should call next_init() if fiber
+	  is dead already. [ruby-dev:32459]
+
+Fri Dec 21 01:21:49 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLContext.build):
+	  enable CRL checking by default.
+
+Fri Dec 21 01:20:56 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/net/http.rb (Net::HTTP#connect): use
+	  OpenSSL::SSL::SSLContext.build instead of SSLContext.new (default
+	  verify mode is now OpenSSL::SSL::VERIFY_PEER).
+
+	* lib/net/https.rb: SSL parameters are defined by attr_accessor.
+
+	* test/net/http/test_https.rb: add test for HTTPS features.
+
+Fri Dec 21 01:11:37 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* io.c (select_internal): should return original value.
+
+Fri Dec 21 00:26:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (tr_trans): wrong encoding check for tree strings.
+
+	* test/ruby/test_m17n.rb (TestM17N::test_tr_s): "invalid mbstring
+	  sequence" is not an error to be tested.
+
+	* test/ruby/test_m17n.rb (TestM17N::test_tr): ditto.
+
+Thu Dec 20 19:29:07 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (initialize): the second argument is an option
+	  hash now.  use SSLContext.build to specify SSL parameters.
+
+Thu Dec 20 19:11:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (native_thread_apply_priority): check
+	  _POSIX_PRIORITY_SCHEDULING for OpenBSD.
+
+Thu Dec 20 18:33:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* configure.in: add libthr for FreeBSD.
+
+Thu Dec 20 18:17:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk, *.ci: renamed to *.c.
+
+	* eval_load.c: renamed to load.c.
+
+Thu Dec 20 17:36:01 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*:  Import RubyGems 1.0.0, r1575.
+
+Thu Dec 20 17:18:38 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: support Proc#binding.
+
+	* sample/test.rb: add a test.
+
+Thu Dec 20 17:15:15 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* pack.c: Slight change to documentation ('character' ->
+	  'byte (C char)'). [ruby-core:13126], see also [ruby-core:13998].
+
+Thu Dec 20 17:07:54 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (enc.mk): depends on rbconfig.rb.
+
+	* regenc.h (OnigEncodingDefine): external encoding definition macro.
+
+	* enc/Makefile.in: fix for linking.
+
+	* enc/depend, enc/make_encmake.rb: fix for Windows.
+
+	* enc/{ascii,euc_jp,sjis,utf8,iso_8859_{1..16}}.c: renamed.
+
+Thu Dec 20 16:42:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (find_line_no): return 0 if not found.
+
+Thu Dec 20 16:04:17 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci, vm.c, vm_core.h: change interface of
+	  vm_invoke_block() to specify block ptr.  [ruby-talk:266422]
+
+	* cont.c, eval_jump.ci, insns.def, proc.c, signal.c, thread.c:
+	  apply above change.
+
+	* bootstraptest/test_knownbug.rb: move fixed bug.
+
+	* bootstraptest/test_block.rb: ditto. and add a test.
+
+Thu Dec 20 15:47:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/iso_8859_{1..16}.c: renamed.
+
+Thu Dec 20 09:59:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_expandarray): fix sp increase place.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp> in [ruby-dev:32581].
+
+	* bootstraptest/test_massign.rb: add a test for above.
+
+	* bootstraptest/test_syntax.rb: fix last committed test.
+
+Thu Dec 20 09:47:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_syntax.rb: add a test.
+
+Thu Dec 20 09:40:51 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each/NODE_RETURN): fix stack consistency.
+
+Thu Dec 20 09:42:11 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (big2str_orig): access beyond memory region cause crash
+	  on interrupt.  a patch from Yusuke ENDOH <mame AT tsg.ne.jp> in
+	  [ruby-dev:32651].  [ruby-dev:32641]
+
+Thu Dec 20 09:06:54 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_index): wrong starting position.
+
+Thu Dec 20 06:34:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): add pop after throw as return.
+
+	* bootstraptest/test_knownbug.rb, test_syntax.rb: move resolved test.
+
+	* vm_core.h, iseq.c, compile.h: add debug output code.
+
+Thu Dec 20 04:57:18 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): remove unused retry entry.
+
+Thu Dec 20 04:15:41 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* */Makefile.sub (DEFS, RM): output to config.status.
+
+Thu Dec 20 02:59:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* common.mk (encs): create encoding directory.
+
+Thu Dec 20 02:50:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/iso8859_{1..16}.c: adjust for ruby.
+
+Thu Dec 20 02:28:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/iso8859_{1..16}.c: imported from Onigiruma 5.9.0.
+
+Thu Dec 20 02:23:22 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/Makefile.in (RM): added.
+
+	* enc/depend (encs): sort in alpha-numeric order.
+
+	* enc/depend (clean, distclean): added.
+
+Thu Dec 20 01:10:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): should handle upper level eval iseq
+	  from break/next, and COMPILE_ERROR() breaks only one block.
+	  [ruby-dev:31372]
+
+Thu Dec 20 00:07:36 2007  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/drbtest.rb (test_07_public_private_protected_missing):
+	  followed current Ruby specification.
+
+Wed Dec 19 23:57:37 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_inspect, dir_path, dir_tell): check for frozen and closed
+	  is not needed.  [ruby-dev:32640]
+
+Wed Dec 19 22:59:52 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_sublen): adjust position if position is not at the
+	  head of a character.
+
+	* string.c (rb_str_chomp_bang): check if match start at the head
+	  of a character.
+
+	* string.c (rb_str_chomp_bang): wrong adjust condition.
+
+	* string.c (rb_str_rindex): comparison length should be based on
+	  bytes, not characters.
+
+	* string.c (rb_str_rindex_m): too much adjustment.
+
+	* re.c (reg_match_pos): pos adjustment should be based on
+	  characters.
+
+	* test/ruby/test_m17n.rb (TestM17N::test_str_insert): test updated
+	  to check negative offset behavior.
+
+	* string.c (rb_str_each_line): should consider rslen.
+
+	* string.c (rb_str_buf_append): should propagate encoding.
+
+	* string.c (rb_str_each_line): ditto.
+
+	* test/ruby/test_m17n.rb (TestM17N::test_str_each_line): should
+	  check encoding as well.
+
+	* test/ruby/test_m17n.rb (TestM17N::test_str_each_line): empty
+	  array can not propagate encoding;  should not check.
+
+Wed Dec 19 21:42:18 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_regsub): should set checked encoding.
+
+	* string.c (rb_str_sub_bang): applied r14212 too.
+
+Wed Dec 19 20:40:01 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (bigmul1): C99ism.
+
+	* bignum.c (bigdivrem1): need dummy return value.
+
+Wed Dec 19 19:18:06 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Updated.
+
+	* ext/nkf/nkf.c (rb_nkf_enc_get): added.
+	  (find encoding or replicate default encoding)
+
+	* ext/nkf/nkf.c (NKF::<ENCODING>): redefine encoding constant.
+
+	* ext/nkf/lib/kconv.rb (Kconv::<ENCODING>): redefined as Encoding.
+
+	* ext/nkf/lib/kconv.rb: refactoring.
+
+Wed Dec 19 19:11:08 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_mul0): blocking check for bigger numbers.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp> in [ruby-dev:32632].
+
+	* bignum.c (bigdivrem): ditto.
+
+Wed Dec 19 17:34:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): remove "retry" in block.
+	  ("iter{retry}" cause syntax error)
+	  Currently, "begin; ...; rescue; iter{retry}; end" cause
+	  syntax error too.
+
+	* bootstraptest/test_jump.rb: ditto.
+
+	* lib/drb/invokemethod.rb: ditto.
+
+	* sample/drb/darrayc.rb: ditto.
+
+	* sample/test.rb: ditto.
+
+	* test/drb/drbtest.rb: ditto.
+
+	* test/ruby/test_iterator.rb: ditto.
+
+	* sample/test.rb: add a 'test' directory on the SYSTEM test.
+
+Wed Dec 19 17:12:59 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb, test_block.rb:
+	  move fixed bug.
+
+	* bootstraptest/test_m17n.rb: added.
+
+Wed Dec 19 16:59:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (errinfo_place): skip if error is Fixnum.  [ruby-dev:32608]
+
+	* bootstraptest/test_exception.rb, test_known_bug.rb: move fixed bug.
+
+Wed Dec 19 16:31:58 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_named_capture_assign_iter): remove C99 dependency.
+
+	* parse.y (reg_named_capture_assign_iter): get rid of creating
+	  unnecessary ID.
+
+	* parse.y (rb_enc_symname2_p): check for non-nul-terminated string.
+
+Wed Dec 19 15:37:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insnhelper.ci (vm_yield_with_cfunc): call cfunc with
+	   (argv[0], data, argc, argv) to pass all arguments.
+
+	* enumerator.c (enumerator_each_i): adapted to new calling
+	  convention.
+
+Wed Dec 19 15:13:20 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_justify): should propagate encoding from pad
+	  string too.
+
+Wed Dec 19 13:57:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (TIMEZONE_VOID): check whether timezone requires zero
+	  arguments.  [ruby-dev:32631]
+
+Wed Dec 19 13:22:14 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c (NKF::_ENCODING): removed.
+
+	* ext/nkf/nkf.c (rb_nkf_kconv): renamed to rb_nkf_convert.
+
+	* ext/nkf/nkf.c (rb_nkf_convert): set encoding.
+
+	* ext/nkf/nkf.c (rb_nkf_guess1): removed.
+
+	* ext/nkf/nkf.c (rb_nkf_guess2): renamed to rb_nkf_guess.
+
+	* ext/nkf/nkf.c (rb_nkf_guess):
+	  guess method now returns encoding object.
+
+	* ext/nkf/nkf-utf8/nkf.c: Update to nkf 2.0.8 2007-12-19.
+
+Wed Dec 19 10:52:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_cstr_to_inum): an underscore succeeding after octal
+	  prefix is allowed.  [ruby-core:14139]
+
+Wed Dec 19 00:09:19 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_cstr_to_inum): wrong radix check.  a patch from
+	  Yusuke ENDOH <mame AT tsg.ne.jp> in [ruby-dev:32628].
+
+	* bignum.c (big2str_find_n1): ditto.
+
+Tue Dec 18 23:53:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): fix for segfault.  [ruby-dev:31372]
+
+Tue Dec 18 23:44:32 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/net/http/utils.rb: split TestNetHTTPUtils module from
+	  test/net/http/test_http.rb. and start HTTP server in each test case.
+
+Tue Dec 18 23:27:51 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#accept_client):
+	  should rescue Errno::EINVAL from TCPServer#accept. this exception
+	  might occur if the server socket is not in ready to listen.
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#accept_client):
+	  don't call TCPServer#close if the :ShutdownSocketWithoutClose is set.
+
+	* lib/webrick/config.rb (WEBrick::Config::General): add new parameter
+	  :ShutdownSocketWithoutClose.
+
+Tue Dec 18 22:51:47 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLServer#shutdown):
+	  new method which calls TCPSocket#shutdown of the underlying socket.
+
+Tue Dec 18 22:11:50 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/rss/parser.rb, lib/rss/atom.rb, lib/rss/rss.rb,
+	  test/rss/rss-assertions.rb, test/rss/test_atom.rb: use
+	  pack/unpack("m") instead of base64 library.
+
+	* lib/webrick/httpproxy.rb: use delete("\n") instead of chomp/chop
+	  because the result of pack("m") might be multi-line.
+
+Tue Dec 18 22:12:35 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci, vm.c: rewrite sp manipulation around method/block
+	  invocation.  [ruby-dev:32547]
+
+Tue Dec 18 22:11:23 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/sspi.rb: use pack/unpack("m") instead of
+	  base64 library which was already removed.
+
+Tue Dec 18 21:09:23 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (invoke_block): merge 2 stack overflow checks.
+
+Tue Dec 18 20:58:35 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insnhelper.ci, insns.def, object.c, vm.c, vm.h:
+	  optimize !@, != method invocation.
+
+	* id.c, id.h: ditto.
+
+	* bootstraptest/test_syntax.rb: add tests for above.
+
+Tue Dec 18 18:10:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: add issues.
+
+Tue Dec 18 20:22:44 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (arg tMATCH arg): call reg_named_capture_assign_gen if regexp
+	  literal is used.
+	  (reg_named_capture_assign_gen): assign the result of named capture
+	  into local variables.
+	  [ruby-dev:32588]
+
+	* re.c: document the assignment by named captures.
+
+Tue Dec 18 18:09:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_splice): propagate encoding.
+
+	* string.c (rb_str_subpat_set): ditto.
+
+Tue Dec 18 17:27:12 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_freeze): preserve frozen state of immediate
+	  values in internal hash table, a la generic_ivar.
+
+	* object.c (rb_obj_frozen_p): check immediate values too.
+
+	* variable.c (generic_ivar_set): add frozen check fro immediate
+	  values.
+
+Tue Dec 18 17:04:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_str_transcode_bang, rb_str_transcode): set new
+	  encoding even if no conversion is done because of 7bit only.
+	  [ruby-dev:32591]
+
+Tue Dec 18 15:43:59 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (encs, ext/ripper/ripper.c): other options must come
+	  before MAKEFLAGS in GNU make.
+
+Tue Dec 18 15:19:55 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/parse_rb.rb: Don't call private fail anymore.
+
+Tue Dec 18 15:17:26 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (encs, ext/ripper/ripper.c): pass MAKEFLAGS.
+
+Tue Dec 18 14:45:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (op_tbl): remove duplication to avoid symbol aliases.
+
+Tue Dec 18 14:39:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_nth): need not to raise out-of-range exception.
+
+	* test/ruby/test_m17n.rb (TestM17N::test_str_aref_len): removed
+	  debug print.
+
+Tue Dec 18 14:05:23 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend: get rid of target expanded as empty for nmake.
+
+Tue Dec 18 07:56:57 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* proc.c (rb_obj_public_method): Object#public_method to retrieve
+	  public method object.
+
+	* proc.c (rb_mod_public_instance_method): Module#public_instance_method
+	  to retrieve public instance method from class / module.
+
+	* proc.c (mnew): visibility check added.
+
+	* eval_error.ci (rb_print_undef): add rb_ prefix.
+
+	* eval_error.ci (rb_print_undef): add visibility in the error
+	  message.
+
+Tue Dec 18 05:54:26 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/Env.rb, lib/base64.rb, lib/importenv.rb, lib/eregex.rb: removed.
+
+	* lib/ping.rb, lib/readbytes.rb: removed
+
+Tue Dec 18 02:30:56 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (BUILTIN_ENCS): removed.
+
+	* common.mk (enc.mk): pass BUILTIN_ENCS from command line.
+
+	* enc/depend: ditto.
+
+	* enc/make_encmake.rb: ditto.
+
+Tue Dec 18 01:46:48 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): need to clear output buffer to avoid
+	  broken encoding compatibility check.
+
+Tue Dec 18 01:40:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (depend_rules): inserts ruby to only headers.
+
+Tue Dec 18 01:21:19 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_encode_length): chomp eol style modifiers.
+
+	* parse.y (parser_magic_comment): ditto.
+
+	* parse.y (set_file_encoding): ditto.
+
+Tue Dec 18 01:15:44 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (encs): added dependencies.
+
+	* enc/Makefile.in, enc/depend, enc/make_encmake.rb: moved serb code.
+
+	* lib/mkmf.rb (depend_rules): now takes content string, not file name.
+
+	* win32/enc-setup.mak: overrides default target.
+
+Tue Dec 18 00:26:12 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_initialize): raise error if non-Unicode fixed
+	  encoding option is specified for regexp literals with \u{}
+	  escapes.
+
+	* string.c (rb_str_squeeze_bang): should squeeze multibyte
+	  characters as well.
+
+Mon Dec 17 21:41:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/enc-setup.mak: extracts BUILTIN_ENCOBJS.
+
+	* tool/compile_prelude.rb: needs srcdir.
+
+Mon Dec 17 21:24:04 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* common.mk (miniprelude.c): add -I$(srcdir).
+
+Mon Dec 17 20:53:27 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_set_ssl_version):
+	  new method OpenSSL::SSL::SSLContext#ssl_version to wrap
+	  SSL_CTX_set_ssl_version.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_get_verify_result):
+	  new method OpenSSL::SSL::SSLSocket#verify_result to wrap
+	  SSL_get_verify_result.
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLContext.build):
+	  new method to build OpenSSL::SSL::SSLContext with Hash parameters.
+	  this method provides safety default parameters than SSLContext.new.
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL.verify_certificate_identity):
+	  new module function: pull out identity verification process
+	  from OpenSSL::SSL::SSLSocket#post_connection_check.
+
+Mon Dec 17 18:42:23 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (scan_once): need no encoding compatibility check.
+	  it's done inside of re_reg_search().
+
+	* string.c (rb_str_split_m): ditto.
+
+	* re.c (rb_reg_regsub): ditto.
+
+Mon Dec 17 17:50:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_index): check if substring is broken.
+
+	* string.c (rb_str_rindex): ditto.
+
+	* string.c (rb_str_succ): should carry over.
+
+Mon Dec 17 17:47:26 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (encs): new target to compile external encodings.
+
+	* enc/Makefile.in: became a serb template.
+
+	* enc/make_encmake.rb: creates enc.mk from enc/Makefile.in using serb.
+
+	* lib/mkmf.rb (relative_from): moved from ext/extmk.rb.
+
+	* lib/mkmf.rb ($extmk): true if under to top source directory, not
+	  only ext.
+
+	* lib/mkmf.rb (depend_rules): extracted from create_makefile.
+
+	* tool/serb.rb (serb): splitted from tool/compile_prelude.rb.
+
+Mon Dec 17 17:32:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (MAKEFILES): removed enc/Makefile.
+
+	* configure.in (EXTERNAL_ENCOBJS, ENCSOS): removed.
+
+	* enc/Makefile.in (BUILTIN_ENCS): includes .c suffix.
+
+	* enc/depend: splitted from Makefile.in.
+
+	* {bcc32,win32,wince}/setup.mak (-encs-): extracts BUILTIN_ENCOBJS.
+
+Mon Dec 17 17:07:53 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_enc_str_asciionly_p): use rb_enc_str_coderange.
+
+Mon Dec 17 16:39:25 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_enc_str_coderange): set ENC_CODERANGE_BROKEN using
+	  rb_enc_precise_mbclen.
+	  (rb_str_valid_encoding_p): just check coderange is
+	  ENC_CODERANGE_BROKEN or not.
+
+Mon Dec 17 16:04:16 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#gets): added second
+	  optional argument to specify maximum length limit.
+
+Mon Dec 17 16:02:30 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httprequest.rb, lib/webrick/cgi.rb: Request-Line or
+	  header fields should be read with maximum length. [ruby-talk:231745]
+
+Mon Dec 17 14:03:39 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (ENC_CODERANGE_VALID): rename from
+	  ENC_CODERANGE_8BIT.
+
+	* string.c (rb_enc_str_coderange): follow the renaming.
+
+Mon Dec 17 13:56:53 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (tr_find): wrong condition fixed.
+
+	* sprintf.c (rb_str_format): check encoding based on result, not
+	  the format string.
+
+	* string.c (rb_str_upto): add encoding check.
+
+Mon Dec 17 12:21:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (RUNRUBY): added RUNRUBYOPT.
+
+Mon Dec 17 11:38:59 2007  Tanaka Akira  <akr@fsij.org>
+
+	* thread_win32.ci (native_thread_create): initialize
+	  th->machine_stack_maxsize as rb_gc_stack_maxsize.
+
+Sun Dec 16 17:07:35 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c (transcode_loop): removed special case (-1)
+	  for undefined conversions.
+
+	* transcode_data_iso_8859.c: Changed from character constants
+	  ('\xC2') to integer constants (0xC2) for shorter files and
+	  better readability; eliminated duplicated tables; changed
+	  from -1 offset to actual UNDEF entry (not yet distinguishing
+	  UNDEF and ILLEGAL correctly).
+
+	* test/ruby/test_transcode.rb: added a test for UNDEF conversion.
+
+Sun Dec 16 14:51:59 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (MAKEFILES): should be enc/Makefile, not GNUmakefile.
+	  [ruby-dev:32609]
+
+	* configure.in (BUILTIN_ENCS): removed escapes for OpenBSD.
+
+Sat Dec 15 23:58:46 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/test_pair.rb (SSLPair#ssl_pair): join the thread, even
+	  on an error.
+
+Sat Dec 15 23:50:31 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc/webrick_testing.rb: join webrick server thread.
+
+Sat Dec 15 22:27:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (BUILTIN_ENCS): splitted command line instead of
+	  semicolons for Solaris.
+
+Sat Dec 15 21:38:24 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/timeout.rb: join the background thread to make sure it is dead.
+
+Sat Dec 15 20:20:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (enc/Makefile): add external encoding objects list.
+
+	* common.mk (BUILTIN_ENCOBJS): renamed from ENCOBJS.
+
+	* configure.in (BUILTIN_ENCS): [] needs to be enclosed because of m4.
+
+	* Makefile.in (BUILTIN_ENCOBJS): substituted by autoconf.
+
+	* enc/Makefile.in: new file to compile external encoding sources.
+
+	* encoding.c (rb_enc_find_index): auto-load external encoding objects
+	  as "ext/ENCODING_NAME".  [ruby-dev:32606]
+
+Sat Dec 15 13:04:30 2007  Tanaka Akira  <akr@fsij.org>
+
+	* vm_core.h (rb_thread_t): new member machine_stack_maxsize and
+	  machine_register_stack_maxsize.
+
+	* gc.c (rb_gc_stack_maxsize): new global variable for the thread size
+	  of the main thread.
+	  (STACK_LEVEL_MAX): use machine_stack_maxsize of current thread.
+	  (ruby_stack_check): check IA64 register stack.
+	  (ruby_set_stack_size): set rb_gc_stack_maxsize.
+	  (Init_stack): set rb_gc_stack_maxsize.
+
+	* thread_pthread.ci (native_thread_create): initialize
+	  th->machine_stack_maxsize and th->machine_register_stack_maxsize.
+
+	* vm.c (Init_BareVM): initialize th->machine_stack_maxsize and
+	  th->machine_register_stack_maxsize.
+
+	* thread_win32.ci (native_thread_create): initialize
+	  th->machine_stack_maxsize.  not tested.  just a guess at all.
+
+	[ruby-dev:32604]
+
+Sat Dec 15 12:58:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_register, rb_enc_replicate, rb_enc_alias): check
+	  if already registered.
+
+Sat Dec 15 01:57:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rdoc/options.rb (Options::parse): do not access $KCODE any
+	  longer.  [ruby-core:14079]
+
+	* lib/irb/init.rb (IRB::IRB.parse_opts): ditto.
+
+	* lib/cgi.rb (CGI::CGI): ditto.
+
+Fri Dec 14 18:18:31 2007  Tanaka Akira  <akr@fsij.org>
+
+	* thread_pthread.ci (native_thread_create): twice the stack size.
+	  512KB is not enough to complete test-all on Debian GNU/Linux on
+	  IA64.
+
+Fri Dec 14 16:10:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_p): RDoc update.  a patch from murphy <murphy AT rubychan.de>.
+	  [ruby-core:14010]
+
+Fri Dec 14 16:06:18 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_cmp): encoding aware comparison.
+
+	* string.c (rb_str_casecmp): ditto.
+
+Fri Dec 14 15:25:30 2007  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c (encoding_equal): new function.
+
+	* transcode.c (str_transcode, transcode_dispatch): added two-step
+	  conversion logic via UTF-8.
+
+	* transcode.c: some minor formatting fixes
+
+	* transcode_data.h, transcode_data_iso_8859.c: Shortened
+	  extremely frequently used macros to shorten file length.
+
+	* test/ruby/test_transcode.rb: Fixed name of test class;
+	  added setup method to ensure all necessary encodings exist;
+	  split tests into more test methods; added tests; fixed ordering
+	  of arguments in assert_equal to have expected result first.
+
+Fri Dec 14 13:47:54 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ruby.imp): fix for circular dependency.  a patch from
+	  Yutaka Kanemoto <kinpoco AT gmail.com> in [ruby-dev:32590].
+
+	* regint.h, st.c, ext/json/ext/generator/generator.c: suppress
+	  warnings on AIX.
+
+Fri Dec 14 12:36:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_VARTYPE): check if a variable is defined
+	  and its type.
+
+	* configure.in (timezone, altzone): check for recent cygwin.
+
+	* missing/strftime.c (strftime): fix for timezone.  [ruby-dev:32536]
+
+	* lib/mkmf.rb (try_var): should fail for functions.
+
+	* ext/readline/extconf.rb: should use have_func for functions instead
+	  of have_var.
+
+Fri Dec 14 10:25:56 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/e2mmap.rb (Exception2MessageMapper::E2MM.Raise): $! no
+	  longer modifiable in 1.9.
+
+Fri Dec 14 08:17:24 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (rb_protect): restore root_jmpbuf even if proc exits by
+	  break such as dbm.delete_if { break }.
+
+Fri Dec 14 02:55:41 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_nth): direct jump if string is 7bit only.  great
+	  performance boost for worst case.
+
+	* string.c (str_strlen): direct size if string is 7bit only.
+
+Fri Dec 14 02:29:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): 1st argument (typically the
+	  receiver) would have higher priority in encoding detection.
+
+Fri Dec 14 02:05:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_synchronized): should check if initialized.
+	  [ruby-dev:32585]
+
+Fri Dec 14 00:54:40 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_initialize): embedded string may override encoding
+	  of the regular expression.
+
+	* re.c (rb_reg_initialize): fix encoding of regular expression if
+	  embedded string has its own encoding specified.
+
+Thu Dec 13 22:16:46 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): encoding should never fall back
+	  to ASCII-8BIT unless both encodings are ASCII-8BIT.
+
+Thu Dec 13 20:31:28 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (rb_str_shared_replace): make str noembed after free.
+
+Thu Dec 13 20:09:09 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (rb_protect): restore root_jmpbuf to avoid SEGV by
+	  'IO.pipe; [].each.next' with gcc version 3.3.5 (Debian 1:3.3.5-13)
+	  on IA64.
+
+Thu Dec 13 17:51:54 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_split_m): need not to check encoding if regexp
+	  is empty.
+
+	* string.c (rb_str_justify): associate encoding of original to the
+	  result.
+
+	* string.c (rb_str_chomp_bang): need to check encoding of record
+	  separator.
+
+	* string.c (str_gsub): should copy encoding to the result.
+
+	* sprintf.c (rb_str_format): ditto.
+
+	* string.c (rb_str_succ): should not enter infinite loop for
+	  non-ASCII, non-alphanumeric character at the bottom.
+
+Thu Dec 13 17:03:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): should swap encoding indexes too.
+
+Thu Dec 13 16:41:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): should not judge compatibility
+	  based on rb_enc_asciicompat().
+
+Thu Dec 13 13:09:03 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/io.h (MakeOpenFile): fptr->enc should be
+	  initialized to zero.  [ruby-dev:32569]
+
+Thu Dec 13 08:56:01 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_getc): use default external encoding if fptr->enc is
+	  not set.  [ruby-dev:32565]
+
+	* lib/rubygems/package.rb (Gem::TarReader::Entry::rewind): typo fixed.
+	  [ruby-dev:32565]
+
+Thu Dec 13 08:24:16 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (read_all): should associate default external encoding.
+
+	* io.c (io_read): should NOT associate default external encoding.
+
+Wed Dec 12 23:22:58 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c, regerror.c, string.c, parse.y, ruby.c, file.c:
+	  use capital letter for \xHH notation.  [ruby-dev:32511]
+
+Wed Dec 12 22:21:34 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_mode_enc): allow specifying external encoding in
+	  open mode, e.g. open(path, "r:utf-8").
+
+Wed Dec 12 21:26:03 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_method.ci (rb_alias): no need to skip aliasing when new
+	  equals to old.  [ruby-core:13990]
+
+Wed Dec 12 16:34:53 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_check_readable): set default external encoding to
+	  STDIN.
+
+	* io.c (io_enc_str): associate encoding to output string.
+
+Wed Dec 12 12:44:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (expr): 'not' and '!' should act as conditional
+	  expression.   [ruby-dev:32548]
+
+Wed Dec 12 12:11:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_regsub): should copy encoding.
+
+	* string.c (rb_str_sub_bang, str_gsub): should check and copy encoding
+	  to be replaced.
+
+Tue Dec 11 23:04:17 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): RDoc update.  a patch from Gary Wright
+	  <radar2002 AT gmail.com>.  [ruby-core:13998]
+
+	* pack.c (pack_unpack): ditto.
+
+Tue Dec 11 16:37:47 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_ascget): renamed from rb_enc_get_ascii.
+
+	* include/ruby/encoding.h: follow the renaming.
+
+	* re.c: ditto.
+
+Tue Dec 11 16:19:26 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, */Makefile.sub (CP, MV): added.
+
+	* common.mk (.y.c): not discard the old target until successfully
+	  created.
+
+Tue Dec 11 15:20:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (Init_Object): singleton_method_{added,removed,undefined}
+	  hooks should be defined for BasicObject.  [ruby-dev:32531]
+
+	* eval.c (Init_eval): method_missing should be defined for all
+	  objects; moved to BasicObject.
+
+Tue Dec 11 14:27:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (transcode_loop): get rid of SEGV at sequence can not be
+	  converted.
+
+	* transcode.c (rb_str_transcode_bang): copy encoding.  [ruby-dev:32532]
+
+	* test/ruby/test_transcode.rb: added tests from Martin Duerst <duerst
+	  AT it.aoyama.ac.jp>.  [ruby-dev:32532]
+
+Tue Dec 11 12:05:51 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_get_ascii): add an argument to provide the
+	  length of the returned character.
+
+	* include/ruby/encoding.h (rb_enc_get_ascii): add the argument.
+
+	* re.c (rb_reg_expr_str): modify rb_enc_get_ascii call.
+	  (rb_reg_quote): ditto.
+	  (rb_reg_regsub): ditto.
+
+Tue Dec 11 09:40:21 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/oniguruma.h (ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE):
+	  parenthesize an argument.
+
+Tue Dec 11 02:23:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_method_missing): RDoc update patch from Hugh Sasse
+	  <hgs AT dmu.ac.uk>.  [ruby-core:12932]
+
+Tue Dec 11 01:51:34 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP::get): now supports gzip
+	  content-encoding.  a patch from Hugh Sasse <hgs AT dmu.ac.uk>.
+	  [ruby-core:13451]
+
+Tue Dec 11 01:21:21 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (shadowing_lvar_gen): no duplicate error for "_".
+
+Mon Dec 10 22:08:47 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* array.c (rb_ary_slice_bang): If an invalid range is given, do
+	  not raise an exception but return nil just like slice() does.
+
+Mon Dec 10 21:47:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (str_transcode): allow non-registered encodings.
+	  [ruby-dev:32520]
+
+Mon Dec 10 21:00:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_slice_bang): should return nil if position out
+	  of range.  a patch from Akinori MUSHA <knu AT iDaemons.org>.
+	  [ruby-dev:32518]
+
+Mon Dec 10 19:02:52 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_match): should calculate offset by converted
+	  operand.  [ruby-cvs:21416]
+
+Mon Dec 10 18:28:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/uri/common.rb (URI::REGEXP::PATTERN): typo in REG_NAME
+	  regular expression.  a patch from Ueda Satoshi
+	  <s-ueda AT livedoor.jp>.  [ruby-dev:32514]
+
+Mon Dec 10 17:46:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_tmp_new): creates hidden temporary buffer.
+
+	* transcode.c (transcoding): added a pointer to function to flush.
+
+	* transcode.c (transcode_loop): do not use string internal.
+	  [ruby-dev:32512]
+
+	* transcode.c (str_transcode): allow Encoding objects.
+
+	* transcode_data.h (BYTE_LOOKUP): use actual struct name.
+
+Mon Dec 10 16:52:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_insert): should not add length in bytes to index in
+	  chars.
+
+Mon Dec 10 14:33:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_public_send): rename invoke_method to public_send.
+	  it now invokes public method only no matter how it's called.
+
+Mon Dec 10 14:00:43 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* transcode.c: new file to provide encoding conversion features.
+	  code contributed by Martin Duerst.
+
+Mon Dec 10 13:50:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_search): return byte offset.  [ruby-dev:32452]
+
+	* re.c (rb_reg_match, rb_reg_match2, rb_reg_match_m): convert byte
+	  offset to char index.
+
+	* string.c (rb_str_index): return byte offset.  [ruby-dev:32472]
+
+	* string.c (rb_str_split_m): calculate in byte offset.
+
+Mon Dec 10 09:56:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm1_neq.rb, bm_vm1_not.rb: added.
+
+Mon Dec 10 07:48:14 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): wrong token was generated.  [ruby-dev:32498]
+
+	* object.c (rb_obj_not_match): wrong test.
+
+Mon Dec 10 06:44:47 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_expr_str): use \xHH instead of \OOO.
+
+	* regerror.c (to_ascii): ditto.
+	  (onig_snprintf_with_pattern): ditto.
+	  (onig_snprintf_with_pattern): ditto.
+
+	* string.c (rb_str_inspect): ditto.
+	  (rb_str_dump): ditto.
+
+	* parse.y (parser_yylex): ditto.
+
+	* ruby.c (proc_options): ditto.
+
+	* file.c (rb_f_test): ditto.
+
+	[ruby-dev:32495]
+
+Mon Dec 10 06:41:00 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_names): new method Regexp#names.
+	  (rb_reg_named_captures): new method Regexp#named_captures
+	  (match_regexp): new method MatchData#regexp.
+	  (match_names): new method MatchData#names.
+
+	* lib/pp.rb (MatchData#pretty_print): show names of named captures.
+
+	[ruby-dev:32493]
+
+Mon Dec 10 01:35:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (expr): redefinable not (!) operator.
+
+	* parse.y (arg): ditto.
+
+	* object.c (rb_obj_not): new method "!".
+
+	* object.c (rb_obj_not_equal): new method "!=".
+
+	* object.c (rb_obj_not_match): new method "!~".
+
+Sun Dec  9 22:31:36 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_s_last_match): accept named capture's name.
+
+Sun Dec  9 15:57:53 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (match_backref_number): new function for converting a backref
+	  name/number to an integer.
+	  (match_offset): use match_backref_number.
+	  (match_begin): ditto.
+	  (match_end): ditto.
+	  (name_to_backref_number): raise IndexError instead of RuntimeError.
+	  (match_inspect): show capture index.
+
+Sun Dec  9 14:59:15 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h (CHECK_STACK_OVERFLOW): reserve frame size.
+	  [ruby-dev:32485]
+
+Sun Dec  9 14:38:25 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_thread_mark): use rb_gc_mark_maybe() for
+	  VM stack specified by mark_stack_len.  [ruby-dev:32462]
+
+	* insnhelper.ci: clear vm stack extended by opt value.
+
+Sun Dec  9 14:08:47 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (FilePathStringValue): defined.  similar to
+	  FilePathValue but no taint check.
+
+	* file.c (rb_get_path_no_checksafe): implementation of
+	  FilePathStringValue.
+	  (rb_file_s_basename): use FilePathStringValue.
+	  (rb_file_s_dirname): ditto.
+	  (rb_file_s_extname): ditto.
+	  (rb_file_s_split): ditto.
+	  (rb_file_join): ditto.
+
+	* dir.c (file_s_fnmatch): ditto.
+
+Sun Dec  9 12:49:34 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (append_utf8): check unicode range.
+
+Sun Dec  9 12:39:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi.rb (read_multipart): exclude blanks from header values.
+	  [ruby-list:44327]
+
+Sun Dec  9 12:18:19 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_get_path): use the original object if to_path method is
+	  not defined.  [ruby-dev:32473]
+
+	* io.c (rb_f_open): call to_open on non-string objects, instead of
+	  to_str.  [ruby-dev:32473]
+
+Sun Dec  9 12:12:23 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (tr_find): returns true if no characters to be removed is
+	  specified.
+
+Sun Dec  9 12:03:16 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_magic_comment): delimits with a semicolon.
+
+Sun Dec  9 11:29:23 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (tr_trans): get rid of segfaults when has multibytes but
+	  source sets have no multibytes.
+
+Sun Dec  9 04:01:28 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_mbclen): return 1 if underlying implementation
+	  returns a length longer than e-p.
+	  (rb_enc_precise_mbclen): return needmore if underlying
+	  implementation returns a length longer than e-p.
+
+Sat Dec  8 17:59:40 2007  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (posix_signal): return value.
+
+Sat Dec  8 17:22:16 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/openssl/test_pkcs7.rb: Remove redundant module namespace.
+
+Sat Dec  8 17:07:10 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c (proc_options): make rb_raise format as a string literal to
+	  avoid warning.
+
+Sat Dec  8 16:18:16 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_check_preprocess): new function for validating regexp
+	  fragment.
+
+	* parse.y (regexp): invoke reg_fragment_check.
+	  (reg_fragment_check): defined.
+	  (reg_fragment_check_gen): defined.
+
+Sat Dec  8 11:06:29 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_mbclen): make it never fail.
+	  (rb_enc_nth): don't check the return value of rb_enc_mbclen.
+	  (rb_enc_strlen): ditto.
+	  (rb_enc_precise_mbclen): return needmore(1) if e <= p.
+	  (rb_enc_get_ascii): new function for extracting ASCII character.
+
+	* include/ruby/encoding.h (rb_enc_get_ascii): declared.
+
+	* include/ruby/regex.h (ismbchar): removed.
+
+	* re.c (rb_reg_expr_str): use rb_enc_get_ascii.
+	  (unescape_escaped_nonascii): use rb_enc_precise_mbclen to determine
+	  the termination of escaped non-ASCII character.
+	  (unescape_nonascii): use rb_enc_precise_mbclen.
+	  (rb_reg_quote): use rb_enc_get_ascii.
+	  (rb_reg_regsub): use rb_enc_get_ascii.
+
+	* string.c (rb_str_reverse) don't check the return value of
+	  rb_enc_mbclen.
+	  (rb_str_split_m): don't call rb_enc_mbclen with e <= p.
+
+	* parse.y (is_identchar): use ISASCII.
+	  (parser_ismbchar): removed.
+	  (parser_precise_mbclen): new macro.
+	  (parser_isascii): new macro.
+	  (parser_tokadd_mbchar): use parser_precise_mbclen to check invalid
+	  character precisely.
+	  (parser_tokadd_string): use parser_isascii.
+	  (parser_yylex): ditto.
+	  (is_special_global_name): don't call is_identchar with e <= p.
+	  (rb_enc_symname_p): ditto.
+
+	  [ruby-dev:32455]
+
+	* ext/tk/sample/tkextlib/vu/canvSticker2.rb: remove coding cookie
+	  because the encoding is not UTF-8.  [ruby-dev:32475]
+
+Fri Dec  7 20:21:35 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/net/ftptls.rb, ext/openssl/lib/net/telnets.rb:
+	  half-finished libraries are discontinued.
+
+Fri Dec  7 15:44:40 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb: use Hash for recursion check as inspect.
+
+Fri Dec  7 15:04:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (flatten): some performance improvements, based on a patch
+	  from Yusuke ENDOH <mame AT tsg.ne.jp> in [ruby-core:13877].
+	  [ruby-core:13851]
+
+	* thread.c (rb_exec_recursive): use Hash instead of Array for
+	  performance improvement.  [ruby-core:13898]
+
+	* thread.c (recursive_pop): use object ID.
+
+Thu Dec  6 19:52:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (arg): typo fixed ("!" -> "|") in the ripper code.
+
+Thu Dec  6 19:48:41 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (arg): tUPLUS no longer works as identity operation any
+	  more.  inspired by [ruby-talk:265532].
+
+Thu Dec  6 18:22:11 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_precise_mbclen): new function for mbclen with
+	  validation.
+
+	* include/ruby/encoding.h (rb_enc_precise_mbclen): declared.
+	  (MBCLEN_CHARFOUND): new macro.
+	  (MBCLEN_INVALID): new macro.
+	  (MBCLEN_NEEDMORE): new macro.
+
+	* include/ruby/oniguruma.h (OnigEncodingTypeST): replace mbc_enc_len
+	  by precise_mbc_enc_len.
+	  (ONIGENC_PRECISE_MBC_ENC_LEN): new macro.
+	  (ONIGENC_CONSTRUCT_MBCLEN_CHARFOUND): new macro.
+	  (ONIGENC_CONSTRUCT_MBCLEN_INVALID): new macro.
+	  (ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE): new macro.
+	  (ONIGENC_MBCLEN_CHARFOUND): new macro.
+	  (ONIGENC_MBCLEN_INVALID): new macro.
+	  (ONIGENC_MBCLEN_NEEDMORE): new macro.
+	  (ONIGENC_MBC_ENC_LEN): use ONIGENC_PRECISE_MBC_ENC_LEN.
+
+	* enc/euc_jp.c: validation implemented.
+
+	* enc/sjis.c: ditto.
+
+	* enc/utf8.c: ditto.
+
+	* string.c (rb_str_inspect): use rb_enc_precise_mbclen for invalid
+	  encoding.
+	  (rb_str_valid_encoding_p): new method String#valid_encoding?.
+
+	* io.c (rb_io_getc): use rb_enc_precise_mbclen.
+
+	[ruby-dev:32438]
+
+Thu Dec  6 01:37:23 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regparse.c (i_apply_case_fold): fix for negative character class.  a
+	  patch from Yusuke ENDOH <mame AT tsg.ne.jp> in [ruby-core:13884].
+
+Thu Dec  6 01:00:38 2007  NARUSE, Yui <naruse@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_s_list): support NetBSD/Citrus iconv.
+
+Wed Dec  5 16:18:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_proc_s_new): call initialize.  [ruby-core:13824]
+
+	* proc.c (rb_proc_location): return file name and line number where
+	  the proc is defined.
+
+	* thread.c (thread_s_new): call initialize.  [ruby-core:13835]
+
+	* thread.c (thread_initialize): split initialize method.
+
+Wed Dec  5 15:25:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_intern3): fix to changing encoding to default, and
+	  uncommented r13835, which is rare but not impossible.
+
+Wed Dec  5 15:15:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (step_i, range_step): support non-fixnum steps.
+	  [ruby-talk:282100]
+
+Wed Dec  5 14:25:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix typo.
+
+Wed Dec  5 13:41:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (yycompile): get rid of tracing while parsing.
+	  [ruby-dev:31351]
+
+	* thread.c (ruby_suppress_tracing): added a new parameter, which
+	  directs to call func always.
+
+Tue Dec  4 19:56:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_convert): should not set encoding unless
+	  the target encoding is supported.  [ruby-dev:32451]
+
+Tue Dec  4 17:34:17 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb (tojis, tosjis, toeuc, toutf8):
+	  set encoding. [ruby-dev:32447]
+
+Tue Dec  4 17:07:25 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/json.rb, lib/json/*: moved to ext/json/lib.
+
+Tue Dec  4 16:34:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_create): achieve target encoding.
+
+	* ext/iconv/iconv.c (iconv_convert, iconv_finish, iconv_iconv,
+	  iconv_conv): set result string encoding.  [ruby-dev:32446]
+
+	* ext/iconv/iconv.c (iconv_initialize, iconv_s_open): set encoding to
+	  Iconv instance.
+
+Tue Dec  4 14:34:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): reverted c flag.
+
+Tue Dec  4 11:23:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_cstr_to_inum): trailing spaces may exist at squeezing
+	  preceding 0s.  [ruby-core:13873]
+
+Mon Dec  3 11:51:53 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/Win32API/*: removed or moved to ext/dl/win32.
+
+	* ext/dl/win32/*: new. [ruby-dev:32387]
+
+Sun Dec  2 22:08:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_mbchar): fix for ASCII chars.  [ruby-dev:32432]
+
+	* parse.y (parser_parse_string, parser_here_document): prevent false
+	  error messages.
+
+Sun Dec  2 20:43:22 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (unescape_escaped_nonascii): fix mbclen argument.
+
+Sun Dec  2 15:47:57 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_mbchar): check insufficient multibyte char.
+	  [ruby-dev:32429]
+
+Sun Dec  2 15:42:16 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb, test/rss/test_version.rb: 0.2.1 -> 0.2.2.
+
+	* lib/rss/maker/itunes.rb: fixed new_itunes_category.
+	* lib/rss/maker/taxonomy.rb: new_taxo_topic -> new_topic because
+	  of consistency.
+
+	* test/rss/test_maker_itunes.rb, test/rss/test_itunes.rb: removed
+	  needless UTF-8 characters.
+
+Sun Dec  2 15:18:37 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_callee_setup_arg): fix error message.
+	  [ruby-dev:32430]
+
+Sun Dec  2 09:12:48 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (regexp): fix /#{}\xa1\xa2/e to be EUC-JP.
+	  (reg_fragment_setenc_gen): extracted from reg_compile_gen.
+
+Sun Dec  2 01:39:51 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_uv_to_utf8): declared.
+
+	* re.c (rb_reg_preprocess): new function for dynamic regexp with
+	  \u{} such as Regexp.new("\\u{6666}").
+	  (rb_reg_prepare_re): preprocess regexp for recompiling.
+	  (read_escaped_byte): new function.
+	  (unescape_escaped_nonascii): new function.
+	  (append_utf8): new function.
+	  (unescape_unicode_list): new function.
+	  (unescape_unicode_bmp): new function.
+	  (unescape_nonascii): new function.
+	  (rb_reg_initialize): preprocess regexp.
+
+	* pack.c (rb_uv_to_utf8): renamed from uv_to_utf8.
+
+	* parse.y (STR_NEW3): take func instead of has8 and hasmb.
+	  (parser_str_new): use default coderange mechanism except for regexp.
+	  (parser_tokadd_utf8): copy regexp source as-is.
+	  (parser_read_escape): UTF-8 stuff removed.
+	  (parser_tokadd_escape): has8bit and hasmb removed.
+	  (parser_tokadd_string): fix 8-bit single byte character with \u.
+	  (parser_parse_string): has8bit and hasmb removed.
+	  (parser_here_document): has8bit and hasmb removed.
+	  (parser_yylex): call parser_tokadd_utf8 instead of read_escape for
+	  UTF-8 character.
+
+Wed Dec  2 01:00:07 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/xmlrpc/server.rb (XMLRPC::Server#server): Improve signal handling so
+	  pressing control-c in the controlling terminal or sending SIGTERM stops
+	  the XML-RPC server.
+
+Sat Dec  1 23:04:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c: rename primary_encoding -> default_external (encoding).
+
+Sat Dec  1 19:52:57 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (Time#to_datetime): use nsec instead of usec.
+
+	* lib/date.rb (DateTime#to_time): second minute as an argument to
+	  Time::utc contains fractional part in rational; hence Time
+	  object may keep resolution at most nanosecond.
+
+Sat Dec  1 14:36:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: move fixed bugs.
+
+	* test/ruby/test_sprintf.rb: ditto.
+
+	* test/yaml/test_yaml.rb: ditto.
+
+Sat Dec  1 13:24:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_yield_with_cfunc): fix to passing argc on third
+	  parameter of IFUNC.  [ruby-dev:32329]
+
+	* enumerator.c: fix to pass exact number of argument.
+
+	* eval.c (rb_yield_values2): added.
+
+	* include/ruby/ruby.h: ditto.
+
+	* bootstraptest/test_knownbug.rb: move a fixed test.
+
+	* bootstraptest/test_block.rb: ditto.
+
+Sat Dec  1 10:45:56 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_open): use to_open for every non-string object.  path
+	  object may use method_missing.
+
+Sat Dec  1 09:44:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (concatarray, splatarray): use to_a instead of
+	  to_splat.
+
+	* insnhelper.ci (caller_setup_args): ditto.
+
+Sat Dec  1 03:34:32 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (newline_node): always remove NODE_BEGIN.
+
+Fri Nov 30 23:48:07 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (ruby_signal): use SA_SIGINFO if available.
+	  [ ruby-Patches-6418 ]
+
+Fri Nov 30 22:52:54 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (trap_signm): SIGVTALRM no longer used for green
+	  thread.  [ruby-talk:281318]
+
+	* signal.c (ruby_sig_finalize): do not install SIG_DFL handler if
+	  previous handler is sighandler().
+
+Fri Nov 30 21:02:15 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/json.rb, lib/json/add/{core.rb, rails.rb},
+	  test/json/test_json_rails.rb: additional files of JSON 1.1.2.
+	  [ruby-dev:32405]
+
+Fri Nov 30 19:33:38 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/rubyext.c (rb_syck_mktime): avoid segmentation fault.
+	  [ruby-core:13735]
+
+Fri Nov 30 19:05:55 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_count): precise argument number check.
+
+	* enum.c (enum_count): return Enumerator if no block given.
+
+Fri Nov 30 16:42:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_take_while): returns Enumerator if no block given.
+
+	* enum.c (enum_drop_while): ditto.
+
+Thu Nov 29 16:59:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt): remove unnecessary NODE_BEGIN.  [ruby-core:13814]
+
+Thu Nov 29 06:45:48 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_eql): recursive comparison should be based on
+	  eql?  [ruby-core:13803]
+
+Wed Nov 28 18:08:00 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json, lib/json, test/json: Update to JSON 1.1.2.
+	  (RubyForge#15447)
+
+	* math.c: fix typo.
+
+Wed Nov 28 16:29:35 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_invoke_block): should splat args.
+	  [ruby-dev:32392]
+
+	* test/ruby/test_yield.rb: add tests for above.
+
+Wed Nov 28 14:43:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): use dldflags instead of DLDFLAGS to
+	  get rid of mixing $LDFLAGS and $ARCH_FLAG.
+
+	* lib/mkmf.rb (configuration): ditto.
+
+	* lib/mkmf.rb (create_makefile): support for extensions which has no
+	  shared object.
+
+Wed Nov 28 02:42:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (big2str_find_n1): removed extraneous element.
+	  [ruby-dev:32351], [ruby-dev:32365]
+
+	* bignum.c (big2str_find_n1): returns necessary digits now.
+
+	* sprintf.c (remove_sign_bits): extends sign bit first.
+
+Tue Nov 27 15:53:43 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): "when *[],1" dumps core.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>.  [ruby-dev:32350]
+
+	* bootstraptest/test_syntax.rb: add a test for above.
+
+Tue Nov 27 15:40:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): "a[*b] += 1" dumps core.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>.  [ruby-dev:32354]
+
+	* bootstraptest/test_syntax.rb: add a test for above.
+
+Tue Nov 27 12:47:23 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def: change return value of "defined?"
+	  for $&, $1, ... .  If such variables are defined,
+	  return "global-variable".
+
+	* test/ruby/test_defined.rb: add tests.
+
+	* bootstraptest/test_syntax.rb: fix a test.
+
+Tue Nov 27 11:54:46 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: fix typo.
+
+Tue Nov 27 11:23:20 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test_beginendblock.rb: add loop to wait signal.
+	  [ruby-dev:32332]
+
+Tue Nov 27 11:14:57 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h, encoding.c, re.c, string.c, parse.y:
+	  rename ENC_CODERANGE_SINGLE to ENC_CODERANGE_7BIT.
+	  rename ENC_CODERANGE_MULTI to ENC_CODERANGE_8BIT.
+	  Because single byte 8bit character, such as Shift_JIS 1byte katakana,
+	  is represented by ENC_CODERANGE_MULTI even if it is not multi byte.
+
+Tue Nov 27 10:45:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_method_missing): fix stack trace.
+
+	* bootstraptest/test_knownbug.rb: move solved tests.
+
+	* bootstraptest/test_method.rb, test/ruby/test_regexp.rb: ditto.
+
+Tue Nov 27 09:57:42 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c: fix to allow dsym for alias/undef.
+	  [ruby-dev:32355]
+
+	* bootstraptest/test_method.rb: add tests for above.
+
+Mon Nov 26 23:18:46 2007  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/extserv.rb (initialize, stop_service): synchronize with
+	  ExtServManager.
+
+	* test/drb/test_drb.rb (TestDRbEval): ignored.
+
+Mon Nov 26 17:32:16 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (Init_Regexp): new method Regexp#fixed_encoding?
+	  [ruby-dev:32361]
+
+Mon Nov 26 13:28:14 2007  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/complex.rb: be able to create Complex(0, -0.0). [ruby-list:44268]
+
+Mon Nov 26 11:24:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_fixed_encoding_p): extracted from rb_reg_prepare_re and
+	  rb_reg_s_union.
+	  (rb_reg_s_union): refactored.
+
+Mon Nov 26 10:44:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_read_internal, rb_sysopen_internal): remove C99 dependency.
+
+Sun Nov 25 22:21:35 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/encoding.h (rb_enc_str_asciionly_p): declared.
+	  (rb_enc_str_asciicompat_p): defined.
+
+	* re.c (rb_reg_initialize_str): use rb_enc_str_asciionly_p.
+	  (rb_reg_quote): return ascii-8bit string if the argument is
+	  ascii-only to generate encoding generic regexp if possible.
+	  (rb_reg_s_union): fix encoding handling.  [ruby-dev:32094]
+
+	* string.c (rb_enc_str_asciionly_p): defined.
+
+Sun Nov 25 12:12:03 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* gem_prelude.rb: Import fast-loading gem_prelude.rb from RubyGems.
+
+	* lib/rubygems*: Import RubyGems r1516.
+
+Sat Nov 24 23:25:52 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_eval.rb (TestEval::test_instance_eval_cvar):
+	  updated not to modify class variable of Object class.
+
+Fri Nov 23 17:34:24 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* io.c: add rb_read_internal() as blocking function.
+
+Fri Nov 23 17:33:39 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: fix comment.
+
+Fri Nov 23 17:26:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: move solved tests.
+
+	* bootstraptest/test_io.rb, test_marshal.rb, test_objectspace.rb:
+	  ditto.
+
+	* test/ruby/test_integer.rb, test_regexp.rb: ditto.
+
+Fri Nov 23 15:59:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* struct.c (rb_struct_alloc_noinit): new function.
+	  (rb_struct_define_without_accessor): add allocator to the arguments.
+
+	* range.c (range_alloc): re-introduced using rb_struct_alloc_noinit.
+
+Fri Nov 23 15:27:43 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (REG_CASESTATE): unused macro removed.
+	  (rb_reg_prepare_re): check encoding difference.
+	  (rb_reg_initialize): check 8bit byte.
+
+	* parse.y (parser_tokadd_escape): fix has8bit.
+
+	  [ruby-dev:32113]
+
+Fri Nov 23 15:16:48 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_f_global_variables): variable names should not
+	  duplicate.  [ruby-dev:32344]
+
+Fri Nov 23 13:34:08 2007  Tanaka Akira  <akr@fsij.org>
+
+	* struct.c (rb_struct_define_without_accessor): new function.
+
+	* range.c (range_alloc): removed.
+	  (Init_Range): use rb_struct_define_without_accessor.
+
+	  based on [ruby-dev:32327].
+
+Fri Nov 23 11:01:54 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (match_begin): should return offset by character.
+	  [ruby-dev:32331]
+
+	* re.c (match_end): ditto.
+
+	* re.c (rb_reg_search): ditto.
+
+Fri Nov 23 10:44:24 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (defined_expr): defined(method(x)) dumped core.  a
+	  patch from Yusuke ENDOH <mame AT tsg.ne.jp>.  [ruby-dev:32335]
+
+Wed Nov 21 18:03:49 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: fix to recycle thread data (VM stack).
+
+	* thread.c: ditto.
+
+	* benchmark/bm_vm3_thread_create_join.rb: add loop count.
+
+Wed Nov 21 18:02:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: add path to trunk/lib if driver runner is
+	  in build directory.
+
+Wed Nov 21 16:39:21 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/fileutils/fileasserts.rb (assert_equal_timestamp): new assert
+	  to test tv_sec only for filestamp resolution portability.
+	  (assert_same_entry): use assert_same_entry for mtime comparison.
+
+Wed Nov 21 14:55:13 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_permutation): add gc guard codes.
+
+Wed Nov 21 11:16:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insnhelper.ci (vm_search_normal_superclass): rename function.
+
+	* insnhelper.ci (vm_search_superclass): ditto.
+
+	* proc.c (struct METHOD): rename rklass -> rclass.
+
+Wed Nov 21 03:12:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (rb_f_system): returns nil on execution failure.
+	  [ruby-core:13715]
+
+Wed Nov 21 01:04:12 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (nil_plus): remove unused function.  [ruby-core:13737]
+
+Tue Nov 20 21:46:46 2007  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_mload): ignore invalid digits in submicro.
+
+Tue Nov 20 20:33:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: rename RFloat#double_value -> float_value.
+
+	* numeric.c, parse.y: ditto.
+
+Tue Nov 20 19:36:21 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.h, vm_core.h: decl of rb_gc_save_machine_context()
+	  should be at vm_core.h.
+
+	* include/ruby/ruby.h, intern.h: remove type rb_thread_t.
+
+	* include/ruby/intern.h: change rb_unblock_function_t,
+	  rb_unblock_function_t.
+
+	* file.c, process.c: apply above changes.
+
+	* thread.c, thread_pthread.ci, thread_win32.ci: ditto.
+
+	* io.c: support blocking open (2). [ruby-core:13614]
+
+Tue Nov 20 17:10:11 2007  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_close_on_exec_p): new method IO#close_on_exec?.
+	  (rb_io_set_close_on_exec): new method IO#close_on_exec=.
+	  [ruby-dev:32323]
+
+Tue Nov 20 16:24:31 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (gc_mark_children): obj->as.file.fptr may be 0 for T_FILE.
+
+Tue Nov 20 15:09:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_read_escape): has8bit flag may be set with control
+	  escape.  [ruby-core:13722]
+
+	* parse.y (parser_prepare): set begging after BOM if exists.
+	  [ruby-core:13718]
+
+Tue Nov 20 14:55:37 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*: Update to RubyGems 0.9.5.
+
+Tue Nov 20 13:00:44 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h win32/win32.c (rb_w32_pipe_exec): use dual fd
+	  instead of socketpair when mode is RDWR.
+
+	* io.c (pipe_open): pass &write_fd to rb_w32_pipe_exec().
+
+	* io.c (popen_redirect): define only when HAVE_FORK.
+
+Tue Nov 20 12:12:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (rb_io_t): add tied_io_for_writing member.
+
+	* io.c: use tied_io_for_writing for duplex popen.
+
+	* gc.c: mark tied_io_for_writing.
+
+	* common.mk: gc.o depends io.h.
+
+	  [ruby-dev:32205]
+
+Tue Nov 20 11:59:33 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/drb/test_drb.rb: rename TestRubyYield to TestDRbRubyYield to
+	  avoid name crash with test/ruby/test_yield.rb.
+	  TestRuby18Yield is renamed to TestDRbRuby18Yield too.
+
+Tue Nov 20 03:24:42 2007  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/extservm.rb: merged from ruby_1_8 branch.
+
+	* lib/drb/acl.rb: ditto.
+
+	* lib/drb/ssl.rb: ditto.
+
+	* lib/drb/unix.rb: ditto.
+
+	* lib/drb/drb.rb: ditto.
+
+	* lib/drb/observer.rb: ditto.
+
+	* lib/drb/invokemethod.rb: ditto.
+
+	* test/drb/test_drbssl.rb: ditto.
+
+	* test/drb/test_drb.rb: ditto.
+
+	* test/drb/drbtest.rb: ditto.
+
+	* test/drb/test_drbunix.rb: ditto.
+
+Tue Nov 20 00:52:46 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/fileutils/fileasserts.rb (assert_equal_time): show nsec if
+	  assertion fails but time.to_s equals.
+	  (assert_same_entry): use assert_equal_time.
+
+	* test/fileutils/test_fileutils.rb (test_install): use
+	  assert_equal_time.
+
+Mon Nov 19 18:46:49 2007  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (utime_internal): fallback utimensat to utimes.
+
+Mon Nov 19 17:51:27 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: check struct timespec, clock_gettime, utimensat,
+	  struct stat.st_atim,
+	  struct stat.st_atimespec,
+	  struct stat.st_atimensec,
+	  struct stat.st_mtim,
+	  struct stat.st_mtimespec,
+	  struct stat.st_mtimensec,
+	  struct stat.st_ctim,
+	  struct stat.st_ctimespec,
+	  struct stat.st_ctimensec.
+
+	* include/ruby/missing.h: provide struct timespec if not available.
+
+	* time.c: support nanosecond-resolution using struct timespec.
+	  (time_nsec): new method: Time#nsec and Time#tv_nsec.
+
+	* include/ruby/intern.h: provide rb_time_nano_new.
+
+	* file.c (utime_internal): use utimensat if available.
+	  (rb_file_s_utime): refactored.
+	  (rb_f_test): use stat_atime, stat_mtime, stat_ctime.
+	  (rb_stat_cmp): check tv_nsec.
+	  (stat_atimespec): new function.
+	  (stat_atime): ditto.
+	  (stat_mtimespec): ditto.
+	  (stat_mtime): ditto.
+	  (stat_ctimespec): ditto.
+	  (stat_ctime): ditto.
+	  (rb_stat_atime): use stat_atime.
+	  (rb_file_s_atime): ditto.
+	  (rb_file_atime): ditto.
+	  (rb_stat_mtime): use stat_mtime.
+	  (rb_file_s_mtime): ditto.
+	  (rb_file_mtime): ditto.
+	  (rb_file_ctime): use stat_ctime.
+	  (rb_file_s_ctime): ditto.
+	  (rb_stat_ctime): ditto.
+
+	* variable.c (rb_copy_generic_ivar): clear clone's instance variables
+	  if obj has no instance variable.
+
+	* marshal.c (w_object): dump instance variables of generated string
+	  for TYPE_USERDEF, even if original object has instance variables.
+
+	* lib/time.rb (Time#xmlschema): use nsec instead of usec.
+
+	[ruby-dev:32306]
+
+Mon Nov 19 17:48:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_class_superclass): should not raise exception for
+	  BasicObject.  [ruby-Bugs-15668]
+
+Mon Nov 19 16:04:08 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_permutation): gives all permutations of elements
+	  if no argument given.  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>.
+	  [ruby-dev:32309]
+
+Mon Nov 19 02:44:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): alias and undef accept dsyms as well
+	  as literals.  [ruby-dev:32308]
+
+Mon Nov 19 02:31:36 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_method.ci (rb_add_method): no redefinition warning for undef.
+
+Mon Nov 19 01:53:11 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_read_escape): disallow control and meta modifiers
+	  for non-ASCII characters.  [ruby-core:13685]
+
+Sun Nov 18 20:47:41 2007  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (mark_dump_arg): it may be called after dump_ensure.
+
+Sun Nov 18 18:27:47 2007  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_minus): fix Time.at(2**60+1) - Time.at(2**60).
+
+Sun Nov 18 17:28:49 2007  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_arg): show actual year in 2-3 digits year warning.
+	  (time_mdump): show actual year in "year too big to marshal" error.
+
+Sun Nov 18 14:03:44 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_method.ci (rb_alias): do not call hook functions until
+	  initialization finishes.  [ruby-talk:279538]
+
+Sun Nov 18 09:09:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (String#tr_cpp): make preprocessor identifiers.
+
+Sun Nov 18 05:19:46 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/mkmf.rb (have_struct_member): define HAVE_type_member.
+
+Sat Nov 17 23:51:29 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): bug fix. [ruby-talk:279100]
+
+Sat Nov 17 23:21:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): should clear parser->tokp as well.
+	  [ruby-dev:32250]
+
+	* parse.y: remove NEED_ASSOC that break test_parser_events.
+
+	* parse.y (parser_yylex): should not decrement line numbers at the
+	  end of file.
+
+	* file.c (rb_find_file_ext): search .rb files first through in the
+	  loadpath.
+
+Fri Nov 16 23:31:18 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_odd_p): new method added.  a patch from Tadashi
+	  Saito <shiba AT mail2.accsnet.ne.jp>.  [ruby-dev:32305]
+
+	* bignum.c (rb_big_even_p): ditto.
+
+Fri Nov 16 17:41:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (Document-class): moved the simplest example to
+	  the top.
+
+	* ext/iconv/iconv.c (iconv_s_iconv): Document-method: needs class
+	  prefix for class method.  [ruby-core:13542]
+
+	* ext/iconv/iconv.c (iconv_iconv): also instance method needs to be
+	  qualified.
+
+Fri Nov 16 16:26:57 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* include/ruby/ruby.h: added some declarations for event hooks.
+
+	* lib/profile.rb: set VM::InstructionSequence.compile_option.
+
+Fri Nov 16 11:16:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb (String#is_binary_data?): use Integer#fdiv.
+
+Fri Nov 16 03:36:01 2007  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c: Node#value defined twice.
+
+	* lib/yaml/: several method redefinitions causing warnings.
+
+Fri Nov 16 03:01:00 2007  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/types.rb: Likewise, pass self to YAML::quick_emit.
+
+Fri Nov 16 02:51:59 2007  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml.rb (quick_emit): use combination of object_id and hash to
+	  identify repeated object references, since GC will reuse memory of
+	  objects during output of YAML. [ruby-Bugs-8548] [ruby-Bugs-3698]
+
+Thu Nov 15 19:49:03 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/curses/extconf.rb: check macro if cannot find func.
+	  [ruby-list:44224]
+
+Thu Nov 15 18:04:06 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb: fix TMP_RUBY_PREFIX for relative load path
+	  environment.
+
+Thu Nov 15 17:28:21 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb: absolute path may not start with a slash.
+	  pointed by usa.
+
+Thu Nov 15 17:07:54 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb: fix first substitution.
+	  use constant for prefix.
+	  pointed by Richard Kilmer.
+
+Thu Nov 15 14:29:56 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (prereq): added auto generated sources.  [ruby-dev:32280]
+
+Thu Nov 15 12:31:13 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb: use constant for prefix.
+
+Thu Nov 15 12:24:39 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb: use simple template system for source
+	  code generation.
+
+Thu Nov 15 12:19:14 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session::FileStore::restore): use
+	  lockfile for exclusive locks.  a patch from <tommy AT tmtm.org>.
+	  [ruby-dev:32296]
+
+Thu Nov 15 12:14:53 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* tool/compile_prelude.rb (c_esc): need to escape closing brace.
+
+Thu Nov 15 11:52:16 2007  Tanaka Akira  <akr@fsij.org>
+
+	* tool/compile_prelude.rb: adjust RbConfig::CONFIG paths relative
+	  to the installation path.
+
+Thu Nov 15 11:25:20 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c (usage): fix typo on --disable-gems option.
+	  pointed by Richard Kilmer.
+
+Wed Nov 14 16:16:17 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/net/http/test_https_proxy.rb
+	  (HTTPSProxyTest::test_https_proxy_authentication): initialize
+	  local variable 't' first.  [ruby-dev:32253]
+
+Wed Nov 14 15:39:24 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/socket/test_socket.rb: update not to use 1.8 assignment to
+	  external local variable in the block parameters.  [ruby-dev:32251]
+
+	* test/strscan/test_stringscanner.rb: avoid $KCODE, and use
+	  String#force_encoding().  [ruby-dev:32251]
+
+Wed Nov 14 14:04:42 2007  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk, Makefile.in: rename prelude.c to miniprelude.c.
+	  rename ext_prelude.c to prelude.c
+
+	* win32/Makefile.sub: ditto.
+
+	* bcc32/Makefile.sub: ditto.
+
+Wed Nov 14 07:09:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c, compile.c, compile.h, debug.c, debug.h,
+	  id.c, insnhelper.h, insns.def, thread.c, thread_pthread.ci,
+	  thread_pthread.h, thread_win32.ci, thread_win32.h, vm.h,
+	  vm_dump.c, vm_evalbody.ci, vm_opts.h: fix comments and
+	  copyright year.
+
+Wed Nov 14 07:07:51 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/makedocs.rb, template/insnstbl.html: removed.
+
+Wed Nov 14 02:50:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (parse.c): dependency also needs vpath.
+
+	* common.mk (node_name.inc, prelude.c): VPATH in nmake does not
+	  work for targets of explicit rules.
+
+Wed Nov 14 02:11:38 2007  Tanaka Akira  <akr@fsij.org>
+
+	* missing/isinf.c (isinf): don't define if the macro is defined.
+
+	* configure.in: no need to set ac_cv_func_isinf=yes on non-gcc
+	  solaris.
+
+Wed Nov 14 01:34:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (round): fallback definition.
+
+	* numeric.c (flo_divmod, flo_round): use round() always.
+	  [ruby-dev:32269]
+
+Wed Nov 14 00:33:49 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: introduce 2 macros:
+	  RFLOAT_VALUE(v), DOUBLE2NUM(dbl).
+	  Rename RFloat#value -> RFloat#double_value.
+	  Do not touch RFloat#double_value directly.
+
+	* bignum.c, insns.def, marshal.c, math.c, numeric.c, object.c,
+	  pack.c, parse.y, process.c, random.c, sprintf.c, string.c,
+	  time.c: apply above changes.
+
+	* ext/dl/mkcallback.rb, ext/json/ext/generator/generator.c:
+	  ditto.
+
+Wed Nov 14 00:15:37 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/ytab.sed: get rid of GNU sed feature.  a patch from Laurent
+	  Sansonetti <laurent.sansonetti AT gmail.com> in [ruby-core:13470].
+
+Tue Nov 13 21:41:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (parse.c), ext/ripper/depend (ripper.c): process after
+	  bison with sed.  [ruby-dev:32204]
+
+	* ruby.c (proc_options): use yydebug in cmdline_options.
+
+	* ruby.c (process_options): set yydebug flag of parser.
+
+	* parse.y (yydebug): moved into struct parser_params.
+
+	* parse.y (rb_parser_get_yydebug, rb_parser_set_yydebug): parser
+	  generic methods.
+
+	* */Makefile.sub (parse.c): moved to common.mk.
+
+	* tool/ytab.sed: comment out yydebug definition, and substitute
+	  yyerror with parser_yyerror.
+
+Tue Nov 13 16:33:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flodivmod): work around for infinity.
+
+	* numeric.c (flo_divmod): work around for platforms have no round().
+	  [ruby-dev:32247]
+
+Tue Nov 13 15:26:33 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lex.c.blt: moved from lex.c.
+
+	* lex.c.src: copied from keywords.  This is the source of lex.c.blt.
+
+	* Makefile.in (lex.c): use lex.c.blt if keywords is same as lex.c.src.
+
+	* win32/Makefile.sub (lex.c): re-introduce copy rule.
+
+	* bcc32/Makefile.sub (lex.c): ditto.
+
+	* wince/Makefile.sub (lex.c): ditto.
+
+Tue Nov 13 15:21:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_specialized_instruction): check argc.
+
+Tue Nov 13 14:44:32 2007  why the lucky stiff  <why@ruby-lang.org>
+
+	* test/yaml/test_yaml.rb: fixed the failing YAML Struct test
+	  at ko1's request.
+
+Tue Nov 13 02:57:04 2007  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
+
+	* numeric.c (flo_divmod): round to the nearest integer.
+	  [ ruby-Bugs-14540 ]
+
+Tue Nov 13 00:36:16 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_settracefunc.rb: fixed tests for set_trace_func.
+
+Mon Nov 12 19:47:29 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (call_trace_proc): should return value.
+
+Mon Nov 12 19:45:18 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc,win}32/Makefile.sub (miniruby): use $(COMMONOBJS) and $(DMYEXT)
+	  instead of $(LIBRUBY_A).
+
+Mon Nov 12 18:32:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {bcc,win}32/Makefile.sub (MINIOBJS): added prelude.$(OBJEXT).
+
+Mon Nov 12 17:13:23 2007  Tanaka Akira  <akr@fsij.org>
+
+	* Makefile.in, common.mk: add prelude.o to MINIOBJS.
+
+Mon Nov 12 16:52:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): rdoc about srcprefix.  a patch from
+	  Daniel Berger <djberg96 AT gmail.com> in [ruby-core:13378].
+
+Mon Nov 12 16:48:09 2007  Tanaka Akira  <akr@fsij.org>
+
+	* Makefile.in, common.mk: add ext_prelude.o to OBJS.
+
+Mon Nov 12 13:57:39 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (MINIDLNOBJS): removed.
+	  (MINIOBJS): set to dln.o if dmydln.o is not used.
+
+	* Makefile.in (miniruby): use MINIOBJS instead of MINIDLNOBJS.
+
+Mon Nov 12 13:53:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-parse-partial): handle stringified
+	  symbols properly using ruby-forward-string.
+
+Mon Nov 12 12:17:59 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (MINIDLNOBJS): defined.
+
+	* Makefile.in (miniruby): use MINIDLNOBJS and COMMONOBJS instead of
+	  MINIOBJS and OBJS to avoid linking both dmydln.o and dln.o.
+
+Sun Nov 11 20:32:45 2007  Tanaka Akira  <akr@fsij.org>
+
+	* {win32,wince,bcc32}/Makefile.sub: delete lex.c rule.
+
+Sun Nov 11 19:40:52 2007  Tanaka Akira  <akr@fsij.org>
+
+	* Makefile.in (lex.c): simplified.
+
+Sun Nov 11 18:31:48 2007  Tanaka Akira  <akr@fsij.org>
+
+	* Makefile.in (lex.c): touch lex.c if gperf failed but lex.c exists.
+	  Although this may cause non-updated lex.c,
+	  svn co may generate keywords newer than lex.c especially on
+	  a file system which can record fractional mtime such as XFS.
+
+Sun Nov 11 17:32:46 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* insnhelper.ci (vm_call_method): pass mn->nd_clss to
+	  vm_call_cfunc() instead of klass.
+
+	* vm.c (rb_thread_method_id_and_klass): traverse parent_iseq.
+
+	* thread.c (call_trace_proc): use rb_thread_method_id_and_klass().
+
+Sun Nov 11 16:54:25 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lex.c: renamed from lex.c.blt.
+
+	* Makefile.in (lex.c): use find command to check mtime.
+
+Sun Nov 11 05:34:13 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* bin/gem: Add forgotten gem command.
+
+Sat Nov 10 23:50:31 2007  Tanaka Akira  <akr@fsij.org>
+
+	* string.c (tr_trans): cast to unsigned char after dereference
+	  a pointer to a char to avoid SEGV with "\377".tr("a", "b").
+	  on FreeBSD/amd64.
+
+Sat Nov 10 23:08:53 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in, common.mk, Makefile.in: don't generate
+	  libminiruby-static.a.
+
+Sat Nov 10 19:46:54 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in, common.mk, Makefile.in: generate libminiruby-static.a
+	  which contains prelude.o for miniruby.
+
+Sat Nov 10 18:10:07 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gem_prelude.rb: new file for gem libraries.  currently empty.
+
+	* common.mk: generate ext_prelude.c by prelude.rb and gem_prelude.rb.
+	  ruby (not miniruby) is linked with ext_prelude.o instead of
+	  prelude.o.
+
+	* inits.c (rb_call_inits): don't call Init_prelude.
+
+	* ruby.c: support --disable-gems option.
+	  (ruby_init_gems): new function to define Gem::Enable and
+	  invoke Init_prelude.
+	  (process_options): call ruby_init_gems just after
+	  ruby_init_loadpath.
+
+	* tool/compile_prelude.rb: support multiple files.
+
+Sat Nov 10 17:27:55 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* thread.c (call_trace_proc): don't call ID2SYM() for ID_ALLOCATOR
+	  to avoid SEGV.
+
+Sat Nov 10 16:37:07 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems: Import RubyGems revision 1493.
+
+	* lib/rubygems.rb: ditto.
+
+	* lib/ubygems.rb: ditto.
+
+	* lib/rbconfig/datadir.rb: ditto.
+
+	* test/rubygems: ditto.
+
+Sat Nov 10 16:34:21 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/soap/property.rb: Don't override Enumerable#inject for 1.9.
+
+Sat Nov 10 14:43:30 2007  David Flanagan  <davidflanagan@ruby-lang.org>
+
+	* parse.y: use ASCII encoding for string literals that are
+	  7-bit clean, fixing regression from my previous patch
+
+Sat Nov 10 13:18:54 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {bcc32,win32}/Makefile.sub: vendor_ruby support.
+
+Fri Nov  9 23:33:16 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_nextc): added single line read forward buffer.
+
+	* parse.y (parser_yylex): adjust line number for fluent interface.
+
+Fri Nov  9 22:04:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.h (FRAME_MAGIC_MASK_BITS): bits of FRAME_MAGIC_MASK.
+
+	* insnhelper.ci (vm_call_cfunc, vm_cfunc_flags): use shift operations.
+
+Fri Nov  9 21:46:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (eval): should be volatile value for GC.
+
+Fri Nov  9 17:48:22 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (locale_encoding): guesstimate encoding from environment
+	  variables.  [ruby-core:13315]
+
+	* ruby.c (process_options): set primary encoding from environment.
+
+Fri Nov  9 16:51:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_each_byte): should update rbuf_off and rbuf_len for
+	  each iteration.  [ruby-dev:31659][ruby-dev:32192]
+
+	* variable.c (rb_cvar_set): cvar assignment obey same rule to cvar
+	  reference.  [ruby-dev:32192]
+
+Fri Nov  9 15:52:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_check_encoding, rb_set_primary_encoding): ENCODING
+	  is no longer in FL_USERS flags.
+
+Fri Nov  9 15:20:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_squeeze_bang): initialize squeezing table if no
+	  arguments given.
+
+Fri Nov  9 13:57:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (each_with_index_i): use rb_yield_values() for
+	  compatibility with Enumerator#with_index().  a patch from Yusuke
+	  ENDOH <mame AT tsg.ne.jp>.  [ruby-dev:32195]
+
+Fri Nov  9 13:45:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_iterator.rb (TestIterator::IterTest::each):
+	  #each_pair is now alias to #each.  [ruby-dev:32192]
+
+	* test/ruby/test_iterator.rb (TestIterator::test_assoc_yield):
+	  ditto
+
+Fri Nov  9 12:56:36 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_load.c (loaded_feature_path): check with type of given feature.
+
+Fri Nov  9 12:43:02 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_basicinstructions.rb: updated for new class
+	  behavior.  [ruby-dev:32192]
+
+	* encoding.c (enc_name): Encoding should not rely on ENCODING in
+	  the FL_USERS flags.
+
+	* encoding.c (rb_enc_from_encoding): do not call rb_enc_associate
+	  for encoding itself.
+
+	* encoding.c (enc_register_at): ditto.
+
+	* marshal.c (r_ivar): do not set real instance variable for
+	  encoding data associated.
+
+Fri Nov  9 10:43:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (send_internal): use self in the previous frame to check for
+	  protected methods.  [ruby-core:13254]
+
+	* insnhelper.ci (vm_call_method): send! method has gone.
+
+Fri Nov  9 10:38:13 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* marshal.c (w_object): should be SPECIAL_CONST_P() instead of
+	  IMMEDIATE_P().
+
+Fri Nov  9 10:29:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_invoke_method): check if invoked in function style.
+	  [ruby-core:13245]
+
+	* insnhelper.ci (vm_call_cfunc, vm_cfunc_flags): stores and returns VM
+	  calling flags.
+
+	* vm.c (rb_vm_cfunc_funcall_p): returns if the current method is
+	  invoked in function style.
+
+Fri Nov  9 10:10:21 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: add rb_context_t#type.
+
+Fri Nov  9 10:05:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.c (set_arg0): fix breaking environ bugs.
+
+Fri Nov  9 07:26:04 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* random.c: update MT URL.[ruby-core:13305].
+
+Thu Nov  8 17:09:55 2007  David Flanagan  <davidflanagan@ruby-lang.org>
+
+	* object.c: improve docs for Object.tap
+
+	* ChangeLog: fix bogus dates on my previous entries
+
+Thu Nov  8 15:13:56 2007 David Flanagan <davidflanagan@ruby-lang.org>
+
+	* parse.y: fix segfault with \x escapes in regexps
+	  delete unused #if 0 code regions from previous patch
+
+Thu Nov  8 12:12:10 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (parser_read_escape): remove C99/gcc-ism.
+
+Thu Nov  8 07:54:22 2007 David Flanagan <davidflanagan@ruby-lang.org>
+
+	* parse.y: patch, based on Nobu's, work to support \u escapes
+	           also modifications for better coderange detection
+
+	* test/ruby/test_unicode_escapes.rb: test cases
+
+	* test/ruby/test_mixed_unicode_escapes.rb: mixed encoding test cases
+
+Thu Nov  8 07:14:37 2007 David Flanagan <davidflanagan@ruby-lang.org>
+
+	* parse.y (rb_intern3): commented out broken code that prevented
+	  correct interning of multi-byte symbols.  Without this patch
+	  :x==:x is false when x is a multi-byte character.
+
+Thu Nov  8 07:04:31 2007 David Flanagan <davidflanagan@ruby-lang.org>
+
+	* string.c (tr_setup_table, tr_trans): fix test failures
+	  in test/ruby/test_string.rb
+
+Wed Nov  7 15:07:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_each_with_index): make different arrays at each
+	  iteration.  [ruby-dev:32181]
+
+Wed Nov  7 05:17:24 2007  David Flanagan <davidflanagan@ruby-lang.org>
+
+	* eval.c: fix typo in invoke_method documentation
+
+Wed Nov  7 03:52:26 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_product): core dumped with non array arguments.
+	  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>.  [ruby-dev:32180]
+
+Wed Nov  7 03:32:38 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/encodings/SHIFT-JIS.rb (REXML::Encoding): place -x for
+	  nkf conversion.  a patch from <moonwolf AT moonwolf.com>.
+	  [ruby-dev:32183]
+
+Wed Nov  7 02:59:49 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_each_index): should return meaningful value.
+
+Tue Nov  6 16:37:47 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_load.c (loaded_feature_path): need to expand relative paths.
+
+	* eval_load.c (rb_feature_p): check if the feature is loading with
+	  load path.  [ruby-dev:31932]
+
+	* eval_load.c (load_lock): check the result of barrier waiting.
+
+	* thread.c (rb_barrier_wait): check if owned by the current thread.
+
+	* thread.c (rb_barrier_release): ditto.
+
+Mon Nov  5 08:01:22 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_eval): move #send to Kernel module from BasicObject.
+
+Mon Nov  5 05:17:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Switch::summarize): fix for long form
+	  option with very long argument.  a patch from Kobayashi Noritada
+	  <nori1 AT dolphin.c.u-tokyo.ac.jp> in [ruby-list:44179].
+
+Mon Nov  5 01:20:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (call_args): remove "parenthesize argument(s) for future
+	  version" warning.  when I added this warning, I had a plan to
+	  reimplement the parser that is simpler than the current one.
+	  since we abandoned the plan, warning no longer required.
+
+Mon Nov  5 01:02:56 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (HTTPHeader#initialize): provide default
+	  User-Agent to fix 500 error on some corrupted HTTP servers.
+	  [ruby-core:13135]
+
+Mon Nov  5 00:32:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_send): allow send/__send__ to call methods of all
+	  visibility again.  we no longer provide __send, __send!.
+
+	* eval.c (rb_invoke_method): new method to honor private
+	  visibility.  if it's invoked in a function call style, it calls
+	  private methods as well (previous 1.9 send behavior).
+
+Mon Nov  5 00:24:24 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub: vendor_ruby support.
+
+	* configure.in (RUBY_LIB): duplicated.
+
+Mon Nov  5 00:01:33 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_quote): quote \v as well.
+
+Sun Nov  4 23:51:59 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_initialize_m): use StringValuePtr instead of
+	  StringValueCStr because \0 exists when Regexp.new("\0").
+
+Sun Nov  4 08:11:19 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (count_objects): count TOTAL.
+
+Sun Nov  4 03:58:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (tr_setup_table): use C array for characters that fit
+	  in a byte to gain performance.
+
+	* string.c (rb_str_delete_bang): ditto.
+
+	* string.c (rb_str_squeeze_bang): ditto.
+
+	* string.c (rb_str_count): ditto.
+
+	* string.c (tr_trans): ditto.
+
+Sun Nov  4 00:06:40 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (count_objects): ObjectSpace.count_objects implemented.
+	  [ruby-core:12301]
+
+Sat Nov  3 22:49:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_each_pair): make Hash#each to be alias to
+	  Hash#each_pair for compatibility and clarity.
+
+	* hash.c (env_each_pair): ditto.
+
+Sat Nov  3 22:41:05 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: --with-vendor-hdrdir implemented.
+
+	* lib/mkmf.rb: check --vendor argument.
+
+	* README.EXT: explain --vendor option for extconf.rb
+
+	* README.EXT.ja: ditto.
+
+Sat Nov  3 20:30:48 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: --with-vendordir implemented.
+
+	* mkconfig.rb: add config to vendorlibdir and vendorarchdir.
+
+	* instruby.rb: make vendor library directories.
+
+	* ruby.c: insert vendor library directories into load path.
+
+Fri Nov  2 20:55:49 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/content.rb, lib/rss/content/, lib/rss/maker/content.rb,
+	  test/rss/test_content.rb, test/rss/test_maker_content.rb,
+	  test/rss/rss-testcase.rb (RSS::TestCase): supported
+	  content:encoded with RSS 2.0.
+	  Suggested by Sam Lown. Thanks.
+
+Fri Nov  2 20:47:04 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb, test/rss/test_version.rb: 0.2.0 -> 0.2.1.
+
+Thu Nov  1 21:56:45 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (Init_Exception): make NameError to be subclass of
+	  StandardError again.
+
+	* error.c (Init_Exception): make SecurityError to be subclass of
+	  Exception, since it's too important to be handled implicitly.
+
+Thu Nov  1 14:51:39 2007 David Flanagan <davidflanagan@ruby-lang.org>
+	* enum.c (take_while_i, drop_while_i) add RTEST to handle nil return
+
+Thu Nov  1 02:12:50 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (prereq): update the path of prelude.c.
+
+	* common.mk (prelude.c): rollback a part of r13675, because it is not
+	  documented and causes build error.
+
+Thu Nov  1 01:52:23 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enum.c (enum_drop): fix typo.
+
+Thu Nov  1 01:51:01 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_core.h (ruby_current_thread): RUBY_EXTERN'ed for probeprofiler.
+
+Thu Nov  1 00:46:30 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/configure.bat, setup.mak: now can recognize OS even if
+	  the ``--target'' option of configure is omitted.
+
+	* win32/README.win32: update the descriptions about compiler.
+
+Wed Oct 31 03:13:41 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_take_while): separate with-block form.
+
+	* enum.c (drop_while_i): ditto.
+
+	* enum.c (enum_butfirst): abandon butfirst method.  reverted.
+
+Tue Oct 30 10:03:43 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_butfirst): add a new method to iterates over
+	  elements but first n.  RDoc need to be updated.
+
+	* enumerator.c (Init_Enumerator): remove unnecessary symbol
+	  initialization.
+
+Mon Oct 29 18:42:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (bvar): block-local variable can shadow outer variable.
+	  [ruby-core:13036]
+
+Mon Oct 29 17:58:16 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_substr): performance improvement.  [ruby-dev:31806]
+
+Mon Oct 29 17:20:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_replicate): new function to replicate encoding.
+
+	* encoding.c (enc_based_encoding): Encoding#base_encoding returns
+	  based encoding of replica.
+
+Mon Oct 29 17:18:36 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): ASCII encoding is compatible with
+	  ASCII-compatible encoding, even for non-string objects.
+
+Sun Oct 28 21:50:02 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: :redirect option implemented to disable redirects.
+	  (OpenURI::HTTPRedirect): new exception class for redirection.
+
+Fri Oct 26 17:38:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (int_chr): take an optional encoding parameter.
+	  [ruby-core:12816]
+
+Fri Oct 26 17:14:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_pow): returns 1.0 for 0**0.0.
+
+	* numeric.c (fix_pow): returns infinity for 0**-1.  [ruby-dev:32084]
+
+Fri Oct 26 15:00:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el (ruby-style-{case,label}-indent): adjust for
+	  labels inside switch block.
+
+Fri Oct 26 05:48:57 2007  David Flanagan <davidflanagan@ruby-lang.org>
+	* array.c: raise IndexError for negative length in rb_ary_fill
+
+Wed Oct 25 07:12:03 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/net/telnet.rb (Net::Telnet#login): Allowing "passphrase" in
+	  addition to "password" for Telnet login prompts. [ruby-Bugs-10746]
+
+Wed Oct 25 06:34:11 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/net/telnet.rb (Net::Telnet#login): Making the password prompt
+	  pattern case insensitive. [ruby-Bugs-10746]
+
+Fri Oct 26 04:21:36 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/pop.rb (Net::POP3::do_start): type fixed.  a patch from
+	  Dan Zwell <dzwell AT zwell.net> in [ruby-core:12941].
+
+Fri Oct 26 01:48:28 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_assoc): check and convert inner arrays (assocs)
+	  using #to_ary.
+
+	* hash.c (rb_hash_s_create): check and convert argument hash
+	  using #to_hash.
+
+	* hash.c (rb_hash_s_create): Hash#[] now takes assocs as source of
+	  hash conversion.
+
+Thu Oct 25 16:46:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): dot at the head of the line denote line
+	  continuation from previous one to support fluent interface.
+	  [experimental]
+
+	* misc/ruby-mode.el (ruby-calculate-indent): support fluent dot.
+
+Thu Oct 25 14:19:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_tell, rb_io_seek): check errno too.  [ruby-dev:32093]
+
+Thu Oct 25 13:59:53 2007 David Flanagan <davidflanagan@ruby-lang.org>
+
+	* parse.y (parser_tokspace): increment tokidx
+	  fixes test failure at [test/ruby/test_stringchar.rb:72]
+
+Thu Oct 25 09:49:49 2007  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri.rb, lib/uri/ldaps.rb: added LDAPS
+	  scheme. [ruby-dev:31896]
+
+Wed Oct 25 06:23:14 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/xmlrpc/parser.rb (XMLRPC::Convert::dateTime): Fixing a bug that
+	  caused time zone conversion to fail for some ISO 8601 date formats.
+	  [ruby-Bugs-12677]
+
+Wed Oct 25 04:59:28 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client#do_rpc): Explicitly start
+	  the HTTP connection to support keepalive requests. [ruby-Bugs-9353]
+
+Wed Oct 25 04:46:53 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client#do_rpc): Improving the error
+	  message for Content-Type check failures. [ruby-core:12163]
+
+Wed Oct 25 03:45:08 2007  James Edward Gray II  <jeg2@ruby-lang.org>
+
+	* lib/xmlrpc/utils.rb (XMLRPC::ParseContentType#parse_content_type):
+	  Making Content-Type checks case insensitive. [ruby-Bugs-3367]
+
+Wed Oct 24 17:09:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokspace): make space in token buffer.
+
+	* parse.y (parser_yylex): fix encoding of single character literal.
+
+Tue Oct 23 13:44:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (call_args2): nd_head of NODE_BLOCK_PASS should be a list.
+	  [ruby-core:12850]
+
+Tue Oct 23 10:42:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): encoding set by command line option takes
+	  priority over the encoding in the source, as the primary encoding.
+
+Mon Oct 22 19:24:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (proc_options): allow space after -E (encoding) option.
+
+Mon Oct 22 11:03:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_check_encoding): returns index now.
+
+	* encoding.c (rb_enc_compatible): check if two objects have compatible
+	  encodings.
+
+	* encoding.c (enc_compatible_p): added Encoding.compatible?.
+
+	* include/ruby/encoding.h (rb_enc_compatible): prototype.
+
+Sun Oct 21 18:29:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_default, rb_enc_primary): return pointers to
+	  rb_encoding of default and primary respectively.  [ruby-core:12795]
+
+	* encoding.c (set_primary_encoding): removed primary_encoding setter.
+
+Sat Oct 20 13:17:01 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_cycle): hide temporary array from ObjectSpace.
+	  [ruby-core:12762]
+
+Sat Oct 20 11:49:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_get_path): returns frozen string.
+
+	* file.c (rb_file_s_chown, rb_file_s_lchown): use uid_t and gid_t.
+
+Fri Oct 19 20:08:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_id_encoding): returns ID "encoding".
+
+	* marshal.c (w_encoding): dump encoding name.
+
+	* marshal.c (r_ivar): load encoding.
+
+Fri Oct 19 16:41:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_regx_options, reg_compile_gen): relaxened encoding
+	  matching rule.
+
+	* re.c (rb_reg_initialize): always set encoding of Regexp.
+
+	* re.c (rb_reg_initialize_str): fix encoding for non 7bit-clean
+	  strings.
+
+	* re.c (rb_reg_initialize_m): use ascii encoding for 'n' option.
+
+Fri Oct 19 11:09:56 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): set primary encoding from the parser
+	  always.  [ruby-core:12758]
+
+	* ruby.c (load_file): should not discard the parser parameter.
+
+Fri Oct 19 10:55:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_last): removed unused variables.
+
+Thu Oct 18 17:08:57 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_find_index): update RDoc.  a patch from David Flanagan
+	  <david AT davidflanagan.com> in [ruby-core:12710].
+
+	* enum.c (enum_take, enum_drop): ditto.
+
+	* enum.c (enum_cycle): should not cause infinite loop for empty
+	  arrays.  [ruby-core:12710]
+
+	* range.c (Init_Range): typo fixed.
+
+Thu Oct 18 16:39:34 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/source.rb (REXML::SourceFactory::SourceFactory): more
+	  duck typed.  better performance on JRuby.
+	  http://headius.blogspot.com/2007/10/another-performance-discovery-rexml.html
+
+Thu Oct 18 09:33:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_first): takes first n element if argument is
+	  given.  [ruby-core:12697]
+
+	* range.c (range_last): returns last n elements if argument is
+	  given.
+
+	* array.c (rb_ary_subseq, rb_ary_last): export.
+
+Wed Oct 17 17:39:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): fixed reversed condition.  [ruby-core:12722]
+
+Wed Oct 17 13:54:28 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_s_union): the last check was not complete.
+
+Wed Oct 17 11:30:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_from_encoding, rb_enc_register): associate index
+	  to self.
+
+	* encoding.c (enc_capable): Encoding objects are encoding capable.
+
+	* re.c (rb_reg_s_union): check if encoding matching by exact encoding
+	  objects.
+
+Wed Oct 17 06:18:06 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_alias, rb_enc_find_index): changed
+	  enc_table_alias to a name-to-index hash.
+
+	* encoding.c (rb_enc_init): use upper case names for aliases to use as
+	  constant names.
+
+	* encoding.c (enc_find): allow symbols.
+
+	* encoding.c (Init_Encoding): define encoding constants.
+
+	* st.c (strcasehash): fix wrong code range condition.
+
+Wed Oct 17 05:07:18 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_primary_encoding): added Encoding.primary_encoding.
+
+	* parse.y (rb_parser_encoding): added.
+
+	* ruby.c (proc_options): added -E and --encoding options.
+
+	* ruby.c (process_options): set primary encoding from command line
+	  option if set, or source encoding.
+
+	* include/ruby/encoding.h (rb_enc_from_encoding,
+	  rb_get_primary_encoding, rb_set_primary_encoding): prototypes.
+
+	* include/ruby/node.h (rb_parser_encoding): prototype.
+
+Wed Oct 17 03:37:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_desc): set encoding.
+
+	* re.c (rb_reg_s_union): check encodings.
+
+	* enc/utf8.c (utf8_code_to_mbclen): 0xfe and 0xff are valid Unicode to
+	  be encoded to 2bytes in UTF-8.  [ruby-core:12700]
+
+Wed Oct 17 02:50:49 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_ord): use encoding.
+
+Wed Oct 17 01:57:06 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_initialize_m): allow binary encoding option.
+	  [ruby-dev:32083]
+
+Tue Oct 16 19:48:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_s_union): check for encoding of original object.
+
+Tue Oct 16 18:28:51 2007  Tanaka Akira  <akr@fsij.org>
+
+	* debug.c: use enum for constants for gdb if possible.
+
+Tue Oct 16 18:20:10 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c, debug.c: move debug enum and constants to debug.c.
+
+Tue Oct 16 18:16:15 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c (RUBY_ENCODING_SHIFT): added as enum.
+
+	* .gdbinit (rp): show encoding and coderange for strings.
+
+Tue Oct 16 14:48:38 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_regx_options): check if regexp encoding option
+	  matches to current encoding.
+
+	* re.c (char_to_option, rb_char_to_option_kcode): 'n' is not kcode
+	  option now.
+
+	* re.c (rb_reg_to_s, rb_reg_error_desc): copy encoding rather than
+	  append as an option.
+
+	* re.c (make_regexp, rb_reg_prepare_re): use encoding of Regexp and
+	  String instead of kcode.
+
+	* re.c (rb_reg_initialize): set fixed option if none is set.
+
+	* re.c (rb_reg_regcomp): ditto.
+
+	* re.c (rb_reg_equal): check if encodings are equal.
+
+	* re.c (rb_reg_initialize_m): encoding option is obsolete.
+
+	* re.c (rb_kcode, rb_get_kcode, rb_set_kcode): removed.
+
+	* re.c (Init_Regexp): removed Regexp#kcode method.
+
+	* ruby.c (proc_options): allow long encoding name.
+
+Tue Oct 16 14:03:09 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_s_union): encoding of all regexp objects should
+	  match.  [ruby-dev:32076]
+
+Tue Oct 16 13:49:41 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (enc_to_s): rename function.
+
+Tue Oct 16 13:25:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_new4): should copy encoding.  a patch from NARUSE,
+	  Yui <naruse AT airemix.com>.  [ruby-dev:32076]
+
+Tue Oct 16 01:31:23 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_inject): RDoc update.  a patch from David Flanagan
+	  <david AT davidflanagan.com> in [ruby-core:12710].
+
+Tue Oct 16 01:25:40 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (Init_Encoding): define #to_s to show encoding name
+	  in to_s representation as well as #inspect.
+
+Mon Oct 15 13:24:08 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_round): should be number but not rounding factor.
+	  [ruby-dev:32060]
+
+Mon Oct 15 11:45:12 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_bytes0): refined length check.  [ruby-dev:32059]
+
+Mon Oct 15 10:24:19 2007  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (pst_to_s): returns a string such as "pid 10220 exit 1"
+	  instead of "256".  [ruby-dev:32053]
+	  (pst_inspect): change format
+	  "#<Process::Status: pid=10220,exited(1)>" to
+	  "#<Process::Status: pid 10220 exit 1>".
+
+Mon Oct 15 09:58:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_bytes0): check if source has enough data.
+	  [ruby-dev:32054]
+
+Mon Oct 15 01:15:09 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (s_accept_nonblock): make accepted fd
+	  nonblocking.  [ruby-talk:274079]
+
+Sun Oct 14 17:31:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_obj_encoding): rdoc update.  a patch from David
+	  Flanagan <david AT davidflanagan.com>.  [ruby-core:12664]
+
+	* encoding.c (enc_dump, enc_load): marshaling feature.  a patch from
+	  David Flanagan.  [ruby-core:12665]
+
+	* encoding.c (Init_Encoding): undefine allocator of Encoding.
+	  [ruby-core:12665], [ruby-core:12666]
+
+	* test/ruby/test_encoding.rb: tests for Encoding from David Flanagan
+	  [ruby-core:12665]
+
+Sun Oct 14 11:09:09 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb (PP::PPMethods#pp_hash): don't sort keys because hash is
+	  ordered.
+	  (ENV.pretty_print): call pp_hash with sorted hash.
+
+Sun Oct 14 04:08:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (AC_SYS_LARGEFILE): keep results also in command
+	  options, to vail out of mismatch.  [ruby-list:44114]
+
+	* mkconfig.rb, lib/mkmf.rb (configuration): add DEFS.
+
+Sun Oct 14 03:55:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/mkexports.rb: deal with __fastcall name decorations.
+	  [ruby-list:44111]
+
+Sun Oct 14 02:20:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_cEncoding): new Encoding class.
+
+	* encoding.c (rb_to_encoding, rb_to_encoding_index): helper functions.
+
+	* encoding.c (rb_obj_encoding): return Encoding object now.
+
+	* gc.c (garbage_collect): mark Encoding objects.
+
+	* inits.c (rb_call_inits): call Init_Encoding.
+
+	* string.c (rb_str_force_encoding): accept Encoding object as well as
+	  encoding name.
+
+	* include/ruby/encoding.h (rb_to_encoding_index, rb_to_encoding):
+	  prototypes.
+
+Sun Oct 14 01:03:30 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI.open_http): fix :ssl_ca_cert option.
+
+Sat Oct 13 21:23:21 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (foletype_s_ole_classes,
+	  foletype_s_typelibs): refactoring.
+
+	* test/win32ole/test_win32ole_type.rb: add some test.
+
+	* ext/win32ole/win32ole.c (Init_win32ole): change method name
+	  WIN32OLE_TYPELIB.ole_types from WIN32OLE_TYPELIB.ole_classes.
+
+	* test/win32ole/test_win32ole_typelib.rb: ditto.
+
+	* test/win32ole/test_folderitem2_invokeverb.rb: check create
+	  shortcut string more strictly (This test is invoked in Japanese
+	  Windows environment).
+
+Sat Oct 13 09:11:58 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (set_file_encoding): case-insensitive search, a patch from
+	  David Flanagan <david AT davidflanagan.com> [ruby-core:12629]
+
+Sat Oct 13 09:02:16 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {bcc,win}32/mkexports.rb: explicit data.  [ruby-list:44108]
+
+Sat Oct 13 00:17:49 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/document.rb (REXML::Document::write): leaky
+	  modification trans -> transitive.  [ruby-dev:32040]
+
+Sat Oct 13 00:00:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: encoding specifier should work if the line matches
+	  /coding[:=] ?/, a la Python PEP-263, so that VIM comments like
+	  "# vim: set fileencoding=<encoding name>" should be recognized.
+
+Fri Oct 12 15:04:54 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (magic_comments): add "encoding" as same as "coding".
+
+	* parse.y (set_file_encoding): special file encoding handling.
+
+	* parse.y (parser_yylex): ditto.
+
+Fri Oct 12 12:44:11 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_combination): fixed memory corruption due to too
+	  small memory allocation
+
+	* array.c (rb_ary_product): accessing out of memory bounds.
+	  condition fixed.
+
+Fri Oct 12 11:22:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (match_values_at): make #select to be alias to #values_at
+	  to adapt RDoc description.  [ruby-core:12588]
+
+Thu Oct 11 21:10:17 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/node.h (NOEX_LOCAL): remove unused local visibility.
+
+	* class.c (ins_methods_push): ditto.
+
+	* class.c (rb_class_local_methods): method removed.
+
+Thu Oct 11 14:29:31 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* */Makefile.sub (COMMON_MACROS): workaround for old SDK's bug.
+	  [ruby-core:12584]
+
+Thu Oct 11 06:35:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, */Makefile.sub (VPATH): add enc directory.
+
+	* common.mk (ENCOBJS): encoding objects.
+
+	* enc: directory for encodings.
+
+Thu Oct 11 00:04:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/oniguruma.h (OnigEncodingTypeST): add OnigEncoding
+	  parameter to every function members.
+
+	* include/ruby/oniguruma.h (OnigEncodingTypeST): add auxiliary
+	  data member to provide user defined data for an encoding.
+
+Wed Oct 10 23:32:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_s_quote): no longer takes optional second argument
+	  that has never been documented.
+
+Wed Oct 10 15:39:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* encoding.c (rb_enc_init): don't alias iso-8859-1 to ascii.
+
+	* ascii.c (OnigEncodingASCII): change the name US-ASCII to ASCII-8BIT.
+
+Wed Oct 10 14:31:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_enc_str_coderange): fixed check for non-ascii.
+
+Tue Oct  9 21:35:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_permutation, rb_ary_combination): missing type
+	  names.
+
+	* array.c (rb_ary_permutation): used buffer should be t1.
+
+	* array.c (rb_ary_permutation): use frozen shared hidden array.
+	  [ruby-dev:31985]
+
+Tue Oct  9 16:58:10 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c: remove to_a completely.
+
+	* array.c (tmpbuf): keep DRY to clear klass of temporary objects.
+
+Tue Oct  9 16:33:32 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_permutation, rb_ary_combination, rb_ary_product):
+	  hide internal buffer objects.  [ruby-dev:31982]
+
+Tue Oct  9 16:00:32 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (parser_read_escape, parser_tokadd_escape): check code range.
+	  [ruby-dev:31980]
+
+Tue Oct  9 15:40:24 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (STR_NEW3): check for if single byte sequence.
+
+Mon Oct  8 20:06:29 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/net/imap.rb, lib/net/smtp.rb, lib/net/pop.rb: hostname should
+	  be verified against server's identity as presented in the server's
+	  certificate. [ruby-dev:31960]
+
+	* ext/openssl/lib/net/telnets.rb, ext/openssl/lib/net/ftptls.rb: ditto.
+
+Sun Oct  7 22:37:47 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_taxonomy.rb, test/rss/test_parser_1.0.rb,
+	  test/rss/test_image.rb, test/rss/rss-testcase.rb: ensured
+	  declaring XML namespaces.
+
+Sun Oct  7 22:00:01 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/node.h: make node flags as VALUE type.
+	  enum ruby_node_flags removed.
+
+	* ruby.c: define RUBY_NODE_* as const for gdb.
+
+Sun Oct  7 18:57:12 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h: enum ruby_value_flags removed.  [ruby-dev:31959]
+
+	* ruby.c: define RUBY_FL_* as const VALUE for gdb.
+
+Sun Oct  7 17:50:14 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/net/http.rb: remove enable_post_connection_check flag.
+
+	* lib/open-uri.rb: ditto.
+
+Sun Oct  7 15:48:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (opt_eq): fix to use rb_str_equal().
+
+Sat Oct  6 23:14:54 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_to_i): update RDoc since base can be any value
+	  between 2 and 36.  [ruby-talk:272879]
+
+Sat Oct  6 16:24:02 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (cont_free): check Fiber or Continuation.
+
+	* bootstraptest/test_knownbug.rb: remove a fixed test.
+
+Sat Oct  6 14:56:02 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_register): returns new index or -1 if failed.
+
+	* encoding.c (rb_enc_alias): check if original name is registered.
+
+	* encoding.c (rb_enc_init): register in same order as kcode options in
+	  re.c.  added new aliases.
+
+	* string.c (rb_str_force_encoding): check if valid encoding name.
+
+Sat Oct  6 14:32:30 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (opt_eq): get rid of gcc bug.
+
+Sat Oct  6 02:34:18 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/defines.h: no longer provide DEFAULT_KCODE.
+
+Fri Oct  5 21:24:59 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (rb_reg_s_union_m): Regexp.union accepts single argument which
+	  is an array of patterns.  [ruby-list:44084]
+
+Fri Oct  5 16:42:27 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_not_match): new method.
+
+Fri Oct  5 16:15:52 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in: Turn on --enable-pthread by default for FreeBSD
+	  5.2.1-RELEASE and later, and remove pthread support for older
+	  versions which has never worked perfectly.
+
+Fri Oct  5 16:11:50 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* time.c (time_to_s): Fix documentation.  Time format changed.
+
+Fri Oct  5 04:02:39 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (in_addr, in6_addr, addr_mask): Make some minor
+	  code optimization.
+
+Fri Oct  5 03:25:51 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (<=>): Implement IPAddr#<=> and make IPAddr
+	  comparable.
+
+	* lib/ipaddr.rb (succ): Implement IPAddr#succ.  You can now create
+	  a range between two IPAddr's, which (Range) object is
+	  enumerable.
+
+	* lib/ipaddr.rb (to_range): A new method to create a Range object
+	  for the (network) address.
+
+Fri Oct  5 03:14:45 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (coerce_other): Support type coercion and make &,
+	  |, == and include? accept a string or an integer instead of an
+	  IPAddr object as the argument.
+
+	* lib/ipaddr.rb (initialize): Give better error messages.
+
+	* lib/ipaddr.rb: Improve documentation.
+
+Thu Oct  4 20:45:53 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (Init_process): win32 has our own WNOHANG definition, so
+	  remove unnecessary #ifdef guard.
+
+Thu Oct  4 20:17:19 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* array.c (rb_ary_permutation, rb_ary_product): support non C99
+	  compilers.
+
+Thu Oct  4 17:33:18 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (kcode_setter): Perl-ish global variable `$=' no longer
+	  effective.
+
+	* io.c (Init_IO): remove obsolete variables: $defout, $deferr.
+
+	* re.c (Init_Regexp): remove obsolete const alias: MatchingData.
+
+	* time.c (Init_Time): remove obsolete Time::times.
+
+	* re.c (ignorecase_setter): change warning message.
+
+	* re.c (ignorecase_getter): now gives warning.
+
+	* string.c (rb_str_cmp_m): update RDoc document.
+
+	* re.c (kcode_setter): restore erroneously removed setter.
+
+Thu Oct  4 16:28:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_obj_encoding): returns encoding of the given object.
+
+	* parse.y (reg_compile_gen): copy encoding from source string if
+	  non-empty.
+
+	* re.c (Init_Regexp): new method Regexp#encoding.
+
+	* string.c (str_encoding): moved to encoding.c
+
+Thu Oct  4 15:49:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_permutation): remove C99 dependency.
+	  [ruby-dev:31934]
+
+	* array.c (rb_ary_product): ditto.
+
+Wed Oct  3 23:37:17 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/nkf/nkf.c, bin/ri, bin/irb: fixed typos in doc and comments.
+	  a patch from Eugene Ossintsev  <eugoss AT gmail.com>.
+	  [ruby-core:12375]
+
+Wed Oct  3 17:56:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: enable specify label to executable.
+	  (-e "ruby1::/path/to/ruby1; ruby2::/path/to/ruby2; ...")
+
+Wed Oct  3 16:58:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_str_new, parser_yylex, rb_intern3): set code-range
+	  bits.
+
+	* parse.y (parser_tokadd_string): check code-range.
+
+	* parse.y (parser_parse_string, parser_here_document): ditto.
+
+	* parse.y (parser_set_encode): check if valid encoding.
+
+Wed Oct  3 15:43:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_cvar_set): check whether class variable is
+	  defined in superclasses.  root classes have higher priority.
+	  removes lower class variable entry from IV_TBL (if it's defined
+	  in classes, not modules).
+
+	* variable.c (rb_cvar_get): ditto.
+
+Wed Oct  3 10:06:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_process_options): push frame with program name.
+	  [ruby-core:12351]
+
+Tue Oct  2 20:16:55 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (init_env): refactoring. remove unused code.
+
+Tue Oct  2 12:30:40 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_product): generalized product, now takes
+	  arbitrary number of arrays.  a patch from David Flanagan
+	  <david AT davidflanagan.com>.  [ruby-core:12346]
+
+Tue Oct  2 08:25:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_permutation): implementation contributed from
+	  David Flanagan.  [ruby-core:12344]
+
+	* array.c (rb_ary_combination): RDoc update to clarify.  a patch
+	  from David Flanagan.  [ruby-core:12344]
+
+	* array.c (rb_ary_permutation): small dirty hack by Matz to avoid
+	  arrays on stack.
+
+Tue Oct  2 07:01:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (proc_dup): proc->block.proc should be self.
+
+	* bootstraptest/test_knownbug.rb, test_method.rb:
+	  move a fixed test.
+
+Mon Oct  1 16:17:44 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/test_method.rb: use assert_normal_exit to test
+	  [ruby-dev:31818].
+
+Mon Oct  1 15:57:49 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (id2ref): skip ICLASS.
+
+Mon Oct  1 15:29:35 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_normal_exit): use `` instead of
+	  system.
+
+Mon Oct  1 15:17:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (id2ref): T_VALUES is less than T_BLOCK.  [ruby-dev:31911]
+
+Mon Oct  1 10:58:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (require_libraries): use require method instead of calling
+	  rb_require directly.  [ruby-dev:31322]
+
+Mon Oct  1 10:52:30 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_options), ruby.c (proc_options, process_options): not
+	  call exit(2) directly.  [ruby-dev:31912]
+
+	* eval.c (ruby_run_node): deal with direct exit code.
+
+Sun Sep 30 17:12:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_append): always set encoding, and coderange
+	  cache bits.
+
+	* include/ruby/encoding.h (ENC_CODERANGE_SET): fixed a bug not to
+	  set cache bits.
+
+Sun Sep 30 11:52:11 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (pretty): don't show beginning empty line.
+
+Sun Sep 30 11:32:34 2007  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c: use #ifdef for test LONG_LONG_VALUE.
+
+Sun Sep 30 04:30:55 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: use SIZEOF_LONG instead of SIZEOF_ULONG which is not
+	  defined.
+
+Sun Sep 30 04:03:43 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (Init_Regexp): test DEFAULT_KCODE in C code because
+	  KCODE_EUC, etc. are enum.
+
+Sun Sep 30 00:55:40 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* variable.c (obj_ivar_each): get rid of warning.
+
+Sat Sep 29 17:45:22 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* main.c (main): use platform-independent per-process initialization.
+	  [ruby-dev:31900]
+
+	* ruby.c (ruby_sysinit): new function for per-process initialization.
+
+	* include/ruby/ruby.h (RUBY_GLOBAL_SETUP): toplevel setup declaration.
+
+	* include/ruby/win32.h, win32/mkexports.rb: alias NtInitialize
+	  ruby_sysinit.
+
+	* win32/win32.c (rb_w32_sysinit): renamed from NtInitialize.
+
+Sat Sep 29 17:31:04 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_combination): new method to give all combination
+	  of elements from an array.  [ruby-list:42671]
+
+	* array.c (rb_ary_product): a new method to get all combinations
+	  of elements from two arrays.  can be extended to combinations of
+	  n-arrays, e.g. a.product(b,c,d).  anyone volunteer?
+
+	* array.c (rb_ary_permutation): empty function body to calculate
+	  permutations of array elements.  need volunteer.
+
+Sat Sep 29 17:14:44 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_leave): move proc invocation from r_entry() to
+	  avoid potential crash.
+
+Sat Sep 29 12:28:08 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_normal_exit): new method.
+
+	* bootstraptest/test_knownbug.rb: add test for Marshal.load.
+
+Sat Sep 29 10:12:20 2007  Tanaka Akira  <akr@fsij.org>
+
+	* variable.c (rb_ivar_set): fix class instance variable.
+
+	* object.c (rb_class_real): cl argument may be 0.
+
+Sat Sep 29 09:12:02 2007  Tanaka Akira  <akr@fsij.org>
+
+	* object.c (rb_class_real): use BUILTIN_TYPE instead of TYPE.
+	  access flags directly instead of FL_TEST.
+	  they are enough because cl argument is a class.
+
+Sat Sep 29 08:57:15 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (strcasecmp): needed for type_strcasehash.
+
+Sat Sep 29 06:47:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (struct cmdline_options): static variables packed.
+
+Sat Sep 29 05:29:30 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_fdopen): create IO object from fd.
+
+	* parse.y (yycompile): use encoding of the source as default.
+
+	* ruby.c (proc_options, load_file): ditto.
+
+Sat Sep 29 04:27:08 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_alias): allow encodings multiple aliases.
+
+	* encoding.c (rb_enc_find_index): search the encoding which has the
+	  given name and return its index if found, or -1.
+
+	* st.c (type_strcasehash): case-insensitive string hash type.
+
+	* string.c (rb_str_force_encoding): force encoding of self.  this name
+	  comes from [ruby-dev:31894] by Martin Duerst.  [ruby-dev:31744]
+
+	* include/ruby/encoding.h (rb_enc_find_index, rb_enc_associate_index):
+	  prototyped.
+
+	* include/ruby/encoding.h (rb_enc_isctype): direct interface to ctype.
+
+	* include/ruby/st.h (st_init_strcasetable): prototyped.
+
+Sat Sep 29 03:53:26 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: Thread local storage should be fiber local.
+
+	* bootstraptest/test_knownbug.rb, test/ruby/test_fiber.rb:
+	  move a fixed test.
+
+Fri Sep 28 23:15:31 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insnhelper.ci (vm_call_method): allow send! to call protected
+	  methods as well.  [ruby-core:12280]
+
+Fri Sep 28 22:33:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_so_fasta.rb: added.
+
+	* benchmark/bm_so_k_nucleotide.rb: added.
+
+	* benchmark/bm_so_reverse_complement.rb: added.
+
+	* benchmark/make_fasta_output.rb: added.
+
+	* benchmark/prepare_so_k_nucleotide.rb: added.
+
+	* benchmark/prepare_so_reverse_complement.rb: added.
+
+Fri Sep 28 19:14:51 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: fix notations.
+
+	* benchmark/bm_loop_whileloop.rb: ditto.
+
+	* benchmark/bm_loop_whileloop2.rb: ditto.
+
+	* benchmark/bm_app_uri.rb: added.
+
+	* benchmark/bm_vm1_ivar_set.rb: ditto.
+
+	* benchmark/bm_so_binary_trees.rb: added from Computer Language
+	  Benchmarks Game (http://shootout.alioth.debian.org/).
+
+	* benchmark/bm_so_fannkuch.rb: ditto.
+
+	* benchmark/bm_so_mandelbrot.rb: ditto.
+
+	* benchmark/bm_so_meteor_contest.rb: ditto.
+
+	* benchmark/bm_so_nbody.rb: ditto.
+
+	* benchmark/bm_so_nsieve.rb: ditto.
+
+	* benchmark/bm_so_nsieve_bits.rb: ditto.
+
+	* benchmark/bm_so_partial_sums.rb: ditto.
+
+	* benchmark/bm_so_pidigits.rb: ditto.
+
+	* benchmark/bm_so_spectralnorm.rb: ditto.
+
+Fri Sep 28 16:22:52 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_core.h (rb_vm_struct): fix typo: bufferd -> buffered.
+
+Fri Sep 28 15:47:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: fix to output benchmark results
+	  to file "bmlog-#{Time.now.strftime('%Y%m%d-%H%M%S')}.#{$$}".
+
+	* benchmark/bm_io_file_create.rb: remove useless codes.
+
+	* benchmark/bm_vm2_eval.rb: added.
+
+Fri Sep 28 15:05:24 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h: export rb_ivar_foreach.
+
+	* include/ruby/ruby.h: modify struct RObject and RClass for optimizing
+	  T_OBJECT space.  [ruby-dev:31853]
+	  (ROBJECT_LEN, ROBJECT_PTR)
+	  (RCLASS_IV_TBL, RCLASS_M_TBL, RCLASS_SUPER, RCLASS_IV_INDEX_TBL)
+	  (RMODULE_IV_TBL, RMODULE_M_TBL, RMODULE_SUPER): abstract accessor
+	  defined.
+
+	* variable.c: support the modified RObject and RClass.
+
+	* object.c: ditto.
+
+	* class.c: ditto.
+
+	* gc.c: ditto.
+
+	* marshal.c: ditto.
+
+	* eval_method.ci: use the abstract accessor.
+
+	* insns.def: ditto.
+
+	* proc.c: ditto.
+
+	* struct.c: ditto.
+
+	* eval.c: ditto.
+
+	* error.c: ditto.
+
+	* vm.c: ditto.
+
+	* insnhelper.ci: ditto.
+
+	* ext/digest/digest.c: ditto.
+
+Fri Sep 28 13:20:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_getline_fast, rb_io_getline_1): set encoding to the
+	  result string, as well as getc.
+
+Fri Sep 28 12:51:42 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_erb.rb: added.
+
+	* benchmark/bm_io_file_(create|read|write).rb: added.
+
+Fri Sep 28 12:49:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: fix file selection algorithm.
+
+Fri Sep 28 02:05:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_comparable): need not to check asciicompat here.
+
+	* encoding.c (rb_enc_check): ditto.
+
+	* string.c (rb_enc_str_coderange): tuned a bit; no broken check.
+
+	* encoding.c (rb_enc_check): new encoding comparison criteria.
+
+Thu Sep 27 17:36:28 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/REAMDE.win32: follow recent changes.
+
+Thu Sep 27 16:01:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: fix to output some helpful messages.
+
+Thu Sep 27 15:44:51 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: added.
+
+	* common.mk: fix to use above driver.
+
+	* benchmark/prepare_so_count_words.rb: added.
+
+	* benchmark/bm_so_count_words.rb: fix benchmark process.
+
+Thu Sep 27 15:42:34 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/fiber/fiber.c: modify prototype declaration.
+	  [ruby-core:12247]
+
+Thu Sep 27 09:42:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (strings, xstring, regexp, dsym): empty strings have
+	  US-ASCII encoding.
+
+Thu Sep 27 07:39:13 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_finish): new method.
+
+	* bootstraptest/test_knownbug.rb: add test for [ruby-dev:31866] using
+	  assert_finish.
+
+Thu Sep 27 04:46:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_associate_index): deal with ASCII compatible
+	  flags.
+
+	* encoding.c (rb_enc_check): allow ASCII compatible strings.
+
+	* parse.y (rb_intern_str): use ASCII encoding for ASCII string.
+
+	* string.c (rb_enc_str_coderange): check for code-range.
+
+	* string.c (rb_str_modify): clear code-range flags.
+
+	* string.c (rb_str_hash, rb_str_eql): ASCII compatible strings are
+	  comparable.
+
+	* include/ruby/encoding.h: added code-range flags.
+
+Thu Sep 27 04:40:47 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_mark_set): new function to mark keys.
+
+	* marshal.c (struct dump_arg, struct load_arg): added wrappers to mark
+	  data and compat_tbl entries.  [ruby-dev:31870]
+
+Thu Sep 27 03:17:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_waitpid): no needs to poll.  [ruby-dev:31871]
+
+Wed Sep 26 20:11:46 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: remove a fixed test.
+
+Wed Sep 26 20:00:12 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: move fixed tests.
+
+	* bootstraptest/test_method.rb: ditto.
+
+	* test/ruby/test_io.rb: ditto.
+
+Wed Sep 26 19:36:26 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (eval): fix to check stack overflow.
+	  [ruby-dev:31850]
+
+	* eval_intern.h, vm.h: move CHECK_STACK_OVERFLOW() macro.
+
+Wed Sep 26 19:27:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_throw): fix to move increment point.
+	  [ruby-dev:31840]
+
+Wed Sep 26 19:23:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: Fiber as SemiCoroutine on default. [ruby-core:12146]
+
+	* ext/fiber/fiber.c: enable Fiber#transfer.
+
+Wed Sep 26 18:38:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_check): check for ASCII-compatibilities.
+
+	* parse.y (parser_tokadd_string, parser_parse_string,
+	  parser_here_document, parser_yylex): set encoding to US-ASCII.
+
+	* parse.y (rb_enc_symname_p): check if valid with encoding.
+
+	* parse.y (rb_intern3): let symbols have encoding.
+
+	* string.c (rb_str_hash): add encoding index.
+
+	* string.c (rb_str_comparable, rb_str_equal, rb_str_eql): check if
+	  compatible encoding.
+
+	* string.c (sym_inspect): made encoding aware.
+
+	* insns.def (opt_eq): compare with encoding.
+
+	* include/ruby/encoding.h (rb_enc_asciicompat): check if ASCII
+	  compatible.
+
+	* include/ruby/encoding.h (rb_enc_get_index): added prototype.
+
+	* include/ruby/intern.h (rb_str_comparable, rb_str_equal): ditto.
+
+Wed Sep 26 15:01:16 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_method.ci (rb_get_alloc_func): cast to suppress a warning.
+
+	* eval_method.ci (remove_method): local variable to be initialized.
+
+Wed Sep 26 08:36:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (ext/extinit.o): use $(OUTFLAG) as well as other
+	  objects.  [ruby-Bugs-14228]
+
+Wed Sep 26 05:12:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yyerror): limit error message length.
+	  [ruby-dev:31848]
+
+Tue Sep 25 15:11:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_ungetc): reallocate internal buffer if pushing data
+	  excess capacity.  [ruby-dev:31650]
+
+Tue Sep 25 13:43:03 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_method.ci (remove_method): should not remove undef place
+	  holder.  [ruby-dev:31816], [ruby-dev:31817]
+
+Tue Sep 25 09:51:31 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_longjmp): source file information may be NULL.
+	  [ruby-dev:31849]
+
+	* eval.c (ruby_finalize_0): clear trace_func before finalization.
+
+Mon Sep 24 22:36:13 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_equal): should handle recursive array.
+
+	* hash.c (hash_equal): should handle recursive hash.
+
+Mon Sep 24 22:14:13 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/weakref.rb (WeakRef): remove debug print.  [ruby-dev:31799]
+
+	* hash.c (hash_i): avoid too frequent hash conflict where key and
+	  value are same.  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>.
+	  [ruby-dev:31802]
+
+Mon Sep 24 17:56:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/strscan/strscan.c (strscan_initialize, strscan_getch): use the
+	  encoding of the target string instead of setting to StringScanner
+	  instance.  [ruby-dev:31831]
+
+Mon Sep 24 16:52:11 2007  Urabe Shyouhei  <shyouhei@ruby-lang.org>
+
+	* lib/net/http.rb: fix typo.
+
+Mon Sep 24 06:49:15 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/net/http.rb: an SSL verification (the server hostname should
+	  be matched with its certificate's commonName) is added.
+	  this verification can be skipped by
+	  "Net::HTTP#enable_post_connection_check=(false)".
+	  suggested by Chris Clark <cclark at isecpartners.com>
+
+	* lib/net/open-uri.rb: use Net::HTTP#enable_post_connection_check to
+	  perform SSL post connection check.
+
+	* ext/openssl/lib/openssl/ssl.c
+	  (OpenSSL::SSL::SSLSocket#post_connection_check): refine error message.
+
+Sun Sep 23 09:05:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (os_obj_of, os_each_obj): hide objects to be finalized.
+	  [ruby-dev:31810]
+
+Sun Sep 23 08:58:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_method.ci (rb_attr): should not use alloca for unknown size
+	  input.  [ruby-dev:31818]
+
+	* parse.y (rb_intern_str): prevent str from optimization.
+
+Sun Sep 23 06:16:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_method.ci (remove_method): check for undefined method.
+	  [ruby-dev:31816]
+
+Sun Sep 23 05:42:35 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* lib/rdoc/options.rb (Options::check_diagram): dot -V output
+	  changed. [ ruby-Bugs-11978 ], Thanks Florian Frank.
+
+Sat Sep 22 06:02:11 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::List::summarize): use each_line if
+	  defined rather than each.  [ruby-Patches-14096]
+
+Sat Sep 22 05:19:49 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_init): separate from strio_initialize
+	  to share with strio_reopen properly.  [ruby-Bugs-13919]
+
+Fri Sep 21 14:51:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_mark_rest): copy just used part.
+
+	* gc.c (gc_mark_children): mark u1 and u2 of NODE_ARGSCAT.
+
+	* gc.c (os_obj_of): hide T_VALUES too.  [ruby-dev:31804]
+
+	* gc.c (run_final): freeze temporary argument array.
+
+Fri Sep 21 04:58:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el (ruby-style-label-indent): fix for function top
+	  level labels.
+
+Fri Sep 21 02:11:22 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_match_m): evaluate a block if match.  it would make
+	  condition statement much shorter, if no else clause is needed.
+
+	* string.c (rb_str_match_m): ditto.
+
+Fri Sep 21 02:02:34 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (hash_equal): should call rb_eql when argument eql is set.
+
+Thu Sep 20 17:28:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (popen_exec), process.c (rb_spawn): stop other threads before
+	  exec.  [ruby-core:08262]
+
+Tue Sep 18 22:08:42 2007  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/matrix.rb: fix a coerce bug of Vector. [ruby-core: 12190]
+
+Mon Sep 17 21:06:03 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (HTTP::GenericRequest#initialize): check path is
+	  not nil. [ruby-dev:31149]
+
+Mon Sep 17 21:00:51 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y (ripper): rename event: arglist_* -> args_*.
+
+	* parse.y (ripper): rename event: restparam -> rest_param.
+
+	* parse.y (ripper): rename event: constpath_* -> const_path_*.
+
+	* parse.y (ripper): rename event: topconst_* -> top_const_*.
+
+	* parse.y (ripper): rename event: iter_block -> method_add_block.
+
+	* parse.y (ripper): support block local parameter declaration.
+
+	* parse.y (ripper): introduce new macro params_new.
+
+	* ext/ripper/lib/ripper/sexp.rb: should not dispose event
+	  arguments whose name ends with "_new" but arity != 0.
+
+Sat Sep 15 23:36:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (struct parser_params): common members in the parser and
+	  ripper must be placed at each same location.
+
+Sat Sep 15 18:25:15 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* string.c (rb_str_rstrip_bang): fixed too much rstrip. [ruby-dev:31786]
+
+Sat Sep 15 17:32:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_initialize): set default encoding.  [ruby-dev:31787]
+
+	* ruby.c (load_file): make new parse instance after processing shebang
+	  line options.
+
+Sat Sep 15 17:04:08 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_associate_index, rb_enc_get_index): check if
+	  object is encoding capable.  [ruby-dev:31780]
+
+	* string.c (rb_str_subpat_set): check for if the argument is a String.
+
+Sat Sep 15 13:31:21 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss.rb, lib/rss/, test/rss/:
+	  - 0.1.9 -> 0.2.0.
+	  - supported Slash module.
+
+Fri Sep 14 22:20:01 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fev_unadvise): no needs to reset
+	  event handlers.
+
+Fri Sep 14 17:28:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_io.rb: tests which cause SEGV should not be
+	  added.
+
+	* bootstraptest/test_knownbug.rb: add above test to known bug.
+
+Fri Sep 14 16:24:04 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_expandarray): assignment should be placed
+	  after creating new array.
+
+Fri Sep 14 16:17:30 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb: add a stress test (-s).
+
+Fri Sep 14 16:14:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.h, eval_intern.h: move some macros to eval_intern.h.
+
+	* eval_jump.ci (rb_f_throw): fix to use NEW_THROW_OBJECT().
+
+	* eval.c (rb_f_loop): remove additional macro.
+
+Fri Sep 14 16:12:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_expandarray): should be volatile value for GC.
+
+Thu Sep 13 15:42:03 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): inline cache entries are overwritten
+	  in iseq_build_body().
+
+Thu Sep 13 14:00:15 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (brace_block): should use compstmt.  patch submitted by
+	  Kirill A. Shutemov <k.shutemov AT gmail.com>  [ruby-core:12154].
+
+Thu Sep 13 13:47:08 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_id2str): fixed typo.
+
+Wed Sep 12 23:12:22 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (proc_options): -W should be allowed in RUBYOPT
+	  environment variable.  [ruby-core:12118]
+
+Wed Sep 12 15:19:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_s_sysopen): should not use alloca for unknown size
+	  input.  [ruby-dev:31775]
+
+	* parse.y (rb_id2str): ditto.
+
+	* marshal.c (w_float): use snprintf instead of sprintf.
+
+Tue Sep 11 17:28:00 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/tempfile.rb (Tempfile::make_tmpname): Allow to specify a
+	  suffix for a temporary file name.
+
+	* lib/tempfile.rb (Tempfile::make_tmpname): Make temporary file
+	  names less predictable by including a random string.
+	  [inspired by: akr]
+
+Tue Sep 11 17:25:59 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb: Add shellescape() and shelljoin().
+
+	* lib/shellwords.rb: Rename shellwords() to shellsplit() and make
+	  the former an alias to the latter.
+
+	* lib/shellwords.rb: Add escape(), split(), join() as class
+	  methods, which are aliases to their respective long names
+	  prefixed with `shell'.
+
+	* lib/shellwords.rb: Add String#shellescape(), String#shellsplit()
+	  and Array#shelljoin() for convenience.
+
+Mon Sep 10 15:48:31 2007  Tanaka Akira  <akr@fsij.org>
+
+	* range.c: represent initialized state using EXCL instead of FL_USER3.
+
+	* range.c (range_dumper): make uninitialized range dumpable.
+	  (range_loader): make uninitialized range loadable.
+
+Mon Sep 10 13:44:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_cycle): avoid infinite loop for empty array.
+	  based on a patch from David Flanagan.  [ruby-core:12085]
+
+Mon Sep 10 01:21:23 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_object0): struct allocation first to check if a
+	  class is a struct.  compatibility check should come next.
+
+Mon Sep 10 01:05:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_step): fixed integer overflow.  [ruby-dev:31763]
+
+Sun Sep  9 08:57:27 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (_strptime): now also attaches an element
+	  which denotes leftover substring if exists.
+
+Sun Sep  9 01:59:08 2007  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (r_object0): check T_STRUCT type for structs.
+
+Sun Sep  9 01:17:05 2007  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (r_object0): don't call user-defined initialize for
+	  T_STRUCT objects.
+
+	* include/ruby/intern.h (rb_struct_initialize): declared.
+
+	* struct.c (rb_struct_initialize): export.
+
+Sat Sep  8 23:55:56 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval_method.ci (rb_get_alloc_func): new function to get allocation
+	  function.
+
+	* include/ruby/intern.h (rb_alloc_func_t): declared.
+	  (rb_define_alloc_func): declared.
+	  (rb_marshal_define_compat): declared.
+
+	* range.c: use T_STRUCT for Range.
+
+	* inits.c: move Init_marshal() prior to Init_Range() because
+	  Init_Range calls rb_marshal_define_compat which needs
+	  marshal's compat_allocator_tbl initialized.
+
+	* marshal.c: support marshal format compatibility layer designed for
+	  marshaling T_STRUCT Range using T_OBJECT format.
+	  (rb_marshal_define_compat): defined.
+
+	[ruby-dev:31710]
+
+Sat Sep  8 10:05:14 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_s_members): check if __members__ is an
+	  array to prevent segmentation fault.  [ruby-dev:31759]
+
+Sat Sep  8 09:33:09 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (str[fp]time): now check specifications more
+	  strictly.
+
+Sat Sep  8 02:56:31 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_fiber.rb (TestFiber::test_throw): uncaught throw
+	  now raise ArgumentError exception.
+
+Sat Sep  8 02:45:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_s_members): should raise TypeError instead
+	  of call rb_bug().  [ruby-dev:31709]
+
+	* marshal.c (r_object0): no nil check require any more.
+
+Sat Sep  8 01:46:36 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* regenc.c, regenc.h (onigenc_single_byte_mbc_enc_len): should take
+	  two arguments. [ruby-dev:31754]
+
+Fri Sep  7 00:58:16 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: fix typo.
+
+Fri Sep  7 00:28:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_getc): forgot to commit rb_enc_mbclen() fix.
+
+	* ext/stringio/stringio.c (strio_getc): rb_enc_mbclen() fix.
+
+	* ext/stringio/stringio.c (strio_ungetc): ditto.
+
+Thu Sep  6 22:57:01 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::Entry_::copy): prevent self copy of
+	  directories.
+
+	* lib/fileutils.rb (FileUtils::fu_each_src_dest0): use try_convert.
+
+	* lib/fileutils.rb (FileUtils::fu_update_option): ditto.
+
+Thu Sep  6 21:36:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/oniguruma.h (OnigEncodingTypeST): add end parameter
+	  to mbc_enc_len.
+
+	* euc_jp.c (mbc_enc_len), sjis.c (mbc_enc_len),
+	  utf8.c (utf8_mbc_enc_len): ditto.
+
+	* encoding.c (rb_enc_mbclen): add end parameter.
+
+Thu Sep  6 21:31:49 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_cycle): typo in rdoc.  a patch from Yugui
+	  <yugui@yugui.sakura.ne.jp>.  [ruby-dev:31748]
+
+Thu Sep  6 12:42:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_succ, rb_str_chop_bang, rb_str_chop): m17n support.
+	  [ruby-dev:31734]
+
+Wed Sep  5 22:02:27 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_subseq): need integer overflow check.
+	  [ruby-dev:31736]
+
+	* array.c (rb_ary_splice): ditto.  [ruby-dev:31737]
+
+	* array.c (rb_ary_fill): ditto.  [ruby-dev:31738]
+
+	* string.c (rb_str_splice): integer overflow for length.
+	  [ruby-dev:31739]
+
+Tue Sep  4 20:43:44 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_EVENT#unadvise.
+
+	* test/win32ole/test_win32ole_event.rb: ditto.
+
+Mon Sep  3 15:37:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (vm.o): depends on st.h too.
+
+Sun Sep  2 23:38:29 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/st.h (struct st_table): make num_entries bitfield
+	  instead of num_bins for speed.  num_entries has less access.
+
+Sun Sep  2 00:37:57 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (_parse): improved parsing of ordinal dates.
+
+	* lib/date/format.rb (_parse): use named character classes in some
+	  regular expressions.
+
+Sat Sep  1 23:44:26 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_jump.ci (rb_f_throw): wrap tag and TAG_THROW in a NODE_MEMO
+	  node to make throw instruction to work well.
+
+Sat Sep  1 20:56:07 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (struct RBignum): embed digits in RBignum for
+	  small bignums.
+
+	* bignum.c: RBignum embedded digits implemented.
+
+	* include/ruby/intern.h: declare rb_big_resize.
+
+	* gc.c: don't free embedded digits.
+
+	* numeric.c: replace direct bignum field accessor by abstract field
+	  accessor such as RBIGNUM(val)->sign to RBIGNUM_SIGN(val).
+
+	* sprintf.c: ditto.
+
+	* compar.c: ditto.
+
+	* marshal.c: ditto.
+
+	* random.c: ditto.
+
+	* .gdbinit: support embedded small bignums.
+
+	[ruby-dev:31689]
+
+Sat Sep  1 19:59:43 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_event_free): IConnectionPoint should be
+	  released.
+
+	* ext/win32ole/win32ole.c (EVENTSINK_Destructor): ITypeInfo should be
+	  released.
+
+	* ext/win32ole/win32ole.c (fev_initialize): refactoring.
+
+Sat Sep  1 16:26:09 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/socket/test_tcp.rb (test_recvfrom): same as mswin32 on mswin64.
+
+Sat Sep  1 14:24:23 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_numcmp, st_numhash): use st_data_t instead of long, because
+	  the former may be larger than the latter.
+
+	* include/ruby/st.h (CHAR_BIT): get rid of magic number.
+
+	* include/ruby/st.h (rb_index_t): use st_data_t for the platforms it
+	  is larger than int.
+
+Sat Sep  1 10:43:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_jump.ci (rb_f_catch): generate new tag object if no argument is
+	  given.  backported from MatzRuby.  [ruby-dev:31609]
+
+	* eval_jump.ci (rb_catch): call #catch without arguments if tag
+	  string is NULL.
+
+	* eval_jump.ci (rb_f_throw): allow throwing non-symbol object.
+
+	* eval_jump.ci (rb_catch_obj): new function to wait throw with arbitrary
+	  object.
+
+	* eval_jump.ci (rb_throw_obj): new function to throw arbitrary object.
+
+	* variable.c (check_autoload_table): prevent multiple calls from
+	  RSTRING_PTR().
+
+Fri Aug 31 07:12:24 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (SQRT_LONG_MAX): use SIZEOF_LONG instead of SIZEOF_VALUE
+	  because SIZEOF_VALUE > SIZEOF_LONG on some platforms.
+
+Fri Aug 31 04:18:03 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_norm_arg, f_arg_item): not croak already erred names.
+	  [ruby-dev:31687]
+
+	* parse.y (assignable_gen): ignore already erred names.
+
+	* parse.y (shadowing_lvar_gen): always make new block local variable
+	  when shadowing outer local variable.  [ruby-dev:31507]
+
+Thu Aug 30 19:40:33 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* regenc.h: check RUBY_EXTERN before including config.h and defines.h
+
+	* common.mk: update header dependency.
+
+Thu Aug 30 14:06:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_obj_define_method): reverted.  [ruby-talk:266637]
+
+Thu Aug 30 13:49:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: initialize $top_srcdir always.  [ruby-dev:31682]
+
+	* lib/mkmf.rb (try_const, have_const): check for a const is defined.
+	  [ruby-core:04422]
+
+Thu Aug 30 08:00:12 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h: declare rb_hash_tbl.
+
+	* include/ruby/ruby.h (RHash): delay st_table allocation.
+	  rename tbl field to ntbl to detect direct reference to the st_table
+	  as a compile error.
+	  (RHASH_TBL): abstract accessor defined.
+	  (RHASH_ITER_LEV): ditto.
+	  (RHASH_IFNONE): ditto.
+	  (RHASH_SIZE): ditto.
+	  (RHASH_EMPTY_P): ditto.
+	  (hash_alloc0, hash_alloc): unified because hash_alloc doesn't
+	  allocate st_table now.
+
+	* hash.c: delay st_table allocation.
+
+	* gc.c: replace tbl by ntbl.
+
+	* array.c: replace direct field accessor by abstract field accessor
+	  such as RHASH(hash)->tbl to RHASH_TBL(hash).
+
+	* marshal.c: ditto.
+
+	* insns.def: ditto.
+
+	* ext/iconv/iconv.c: ditto.
+
+	* ext/json/ext/generator/generator.c: ditto.
+
+	* ext/json/ext/parser/parser.c: ditto.
+
+	* ext/json/ext/parser/parser.rl: ditto.
+
+	* ext/syck/rubyext.c: ditto.
+
+	* ext/tk/tkutil/tkutil.c: ditto.
+
+	[ruby-dev:31678]
+
+Wed Aug 29 18:36:06 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: add :ftp_active_mode option.  [ruby-dev:31677]
+
+Wed Aug 29 14:55:28 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): should not use mbclen2() which has broken API.
+
+	* re.c: remove rb_reg_mbclen2().
+
+Wed Aug 29 12:48:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (aref_args): args may not be a list.  [ruby-dev:31592]
+
+Wed Aug 29 11:30:10 2007  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/st.h (struct st_table): add entries_packed 1-bit
+	  bitfield.  decrease num_bins 1-bit.
+
+	* st.c: pack numhash which have 5 or less entries in bins.
+	  (st_init_table_with_size): setup entries_packed flag.
+	  (st_clear): support packed mode.
+	  (st_lookup): ditto.
+	  (st_insert): ditto.
+	  (st_add_direct): ditto.
+	  (st_copy): ditto.
+	  (st_delete): ditto.
+	  (st_foreach): ditto.
+	  (st_reverse_foreach): ditto.
+	  (unpack_entries): new function for converting to unpacked mode.
+
+	[ruby-list:43954]
+
+Wed Aug 29 10:46:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/defines.h (flush_register_windows): call "ta 0x03"
+	  even on Linux/Sparc.  [ruby-dev:31674]
+
+Tue Aug 28 19:16:00 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_type_progid): progid getted by
+	  ProgIDFromCLSID should be freed by CoTaskMemFree. Thanks, arton.
+
+	* test/win32ole/test_win32ole.rb (test_raise_message): set negative
+	  compareMode value to raise WIN32OLERuntimeError.
+
+	* test/win32ole/test_win32ole_type.rb (test_implemented_ole_types):
+	  support some environment which returns IShellDispatch5 instead
+	  of IShellDispatch.
+
+Tue Aug 28 15:42:09 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_subseq): retrieve substring based on byte offset.
+
+	* string.c (rb_str_rindex_m): was confusing character offset and
+	  byte offset.
+
+Tue Aug 28 14:23:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_splice_0): should check to modify.  [ruby-dev:31665]
+
+Tue Aug 28 14:21:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_prepare): set parser encode too when BOM exists.
+
+Tue Aug 28 00:51:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: fix Mutex to be interruptable lock.
+
+	* thread_win32.ci, thread_win32.h, thread_pthread.ci, thread_pthread.h:
+	  prepare native_cond_*() which are based on pthread_cond_*() spec.
+
+	* prelude.rb: fix Mutex#synchronize method.
+
+	* vm_core.h, include/ruby/intern.h: change unblock function interface
+	  (to pass some user data).
+
+	* file.c, process.c: ditto.
+
+	* benchmark/bm_vm2_mutex.rb: add a benchmark for mutex.
+
+	* benchmark/bm_vm3_thread_mutex.rb: add a benchmark for mutex
+	  with contention.
+
+	* benchmark/run.rb: fix to remove ENV['RUBYLIB'] for matzruby.
+
+	* test/ruby/test_thread.rb: add a test.
+
+	* common.mk: fix benchmark options.
+
+Mon Aug 27 23:14:02 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_rstrip_bang): wrong strip point.  [ruby-dev:31652]
+
+	* string.c (rb_str_each_line): should swallow sequence of newlines
+	  if rs (optional argument) is an empty string.  [ruby-dev:31652]
+
+Mon Aug 27 22:39:08 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c (rb_enc_codelen): raises invalid sequence exception
+	  if ONIGENC_CODE_TO_MBCLEN() returns zero.  [ruby-dev:31661]
+
+	* encoding.c (rb_enc_mbclen): check invalid sequence.
+
+Mon Aug 27 20:27:59 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_type_progid): fix the bug.
+	  [ruby-dev:31576]
+
+	* test/win32ole/test_win32ole_type.rb (test_initialize):
+	  remove duplicate assertions.
+
+Mon Aug 27 19:10:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_getlogin): update documentation to note
+	  security issue.  [ruby-Bugs-11821]
+
+Mon Aug 27 15:56:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_encoding): return the encoding of a Symbol.
+
+Mon Aug 27 15:33:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (IEEE_BIG_ENDIAN): use configured value.  [ruby-dev:31623]
+
+	* util.c (Llong): set to LONG_LONG if available.
+
+Mon Aug 27 13:11:56 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (tr_trans): wrong condition for mbmaxlen==1 strings.
+	  [ruby-dev:31652]
+
+Mon Aug 27 00:41:13 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_each_byte): caused infinite loop.  [ruby-dev:31652]
+
+	* io.c (rb_io_getc): should return nil at EOF, not EOFError.
+
+	* lib/delegate.rb (SimpleDelegator::__setobj__): use raise
+	  argument to specify backtrace.
+
+	* test/ruby/test_fnmatch.rb (TestFnmatch::bracket_test):
+	  String#include? no longer works for Fixnum.  use #chr.
+	  [ruby-dev:31652]
+
+Sun Aug 26 12:27:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: fix to remove Fiber.new until fiber.so is not loaded.
+
+	* test/ruby/test_continuation.rb: fix to use resume.
+
+Sun Aug 26 06:51:46 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_wc2mb, reg_enum_key): allocate
+	  buffer should be NULL terminated.
+
+Sun Aug 26 06:04:13 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_setup_method): reorder code for branch prediction.
+
+Sun Aug 26 05:54:49 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_call_method): fix to relaxant safe level check
+	  ($SAFE > 2).  [ruby-core:11998]
+
+	* bootstraptest/test_method.rb: add tests for above.
+
+Sun Aug 26 05:52:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_fiber.rb: fix to require 'continuation'.
+
+Sat Aug 25 23:52:14 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_next): message changed.
+
+Sat Aug 25 23:22:14 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/encoding.h: remove unused rb_enc_ismbchar().
+
+Sat Aug 25 22:50:14 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_event_free, EVENTSINK_Destructor,
+	  fev_initialize): remove the connection ole_event_free and
+	  EVENTSINK_Destructor.
+
+Sat Aug 25 17:52:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c, include/ruby/intern.h: export rb_str_length().
+
+	* insns.def: use rb_str_length() in opt_length.
+
+Sat Aug 25 17:48:51 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: rename FIBER_STACK_SIZE to FIBER_VM_STACK_SIZE.
+
+Sat Aug 25 17:05:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (swallow): removed condition using an unset variable.
+
+	* parse.y, re.c: re-applied revision 13092.
+
+	* string.c (rb_str_splice): return from void function.
+
+	* include/ruby/encoding.h (rb_enc_str_new): prototype added.
+
+Sat Aug 25 11:45:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* encoding.c: provide basic features for M17N.
+
+	* parse.y: encoding aware parsing.
+
+	* parse.y (pragma_encoding): encoding specification pragma.
+
+	* parse.y (rb_intern3): encoding specified symbols.
+
+	* string.c (rb_str_length): length based on characters.
+	  for older behavior, bytesize method added.
+
+	* string.c (rb_str_index_m): index based on characters.  rindex as
+	  well.
+
+	* string.c (succ_char): encoding aware succeeding string.
+
+	* string.c (rb_str_reverse): reverse based on characters.
+
+	* string.c (rb_str_inspect): encoding aware string description.
+
+	* string.c (rb_str_upcase_bang): encoding aware case conversion.
+	  downcase, capitalize, swapcase as well.
+
+	* string.c (rb_str_tr_bang): tr based on characters.  delete,
+	  squeeze, tr_s, count as well.
+
+	* string.c (rb_str_split_m): split based on characters.
+
+	* string.c (rb_str_each_line): encoding aware each_line.
+
+	* string.c (rb_str_each_char): added.  iteration based on
+	  characters.
+
+	* string.c (rb_str_strip_bang): encoding aware whitespace
+	  stripping.  lstrip, rstrip as well.
+
+	* string.c (rb_str_justify): encoding aware justifying (ljust,
+	  rjust, center).
+
+	* string.c (str_encoding): get encoding attribute from a string.
+
+	* re.c (rb_reg_initialize): encoding aware regular expression
+
+	* sprintf.c (rb_str_format): formatting (i.e. length count) based
+	  on characters.
+
+	* io.c (rb_io_getc): getc to return one-character string.
+	  for older behavior, getbyte method added.
+
+	* ext/stringio/stringio.c (strio_getc): ditto.
+
+	* io.c (rb_io_ungetc): allow pushing arbitrary string at the
+	  current reading point.
+
+	* ext/stringio/stringio.c (strio_ungetc): ditto.
+
+	* ext/strscan/strscan.c: encoding support.
+
+Sat Aug 25 10:59:19 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: separate Continuation and Fiber from core.
+
+	* ext/continuation/*, ext/fiber/*: ditto.
+
+	* include/ruby/ruby.h: remove rb_cFiber.
+
+	* include/ruby/intern.h: add the rb_fiber_new() declaration.
+
+	* enumerator.c (next_init): fix to use rb_fiber_new().
+
+	* test/ruby/test_enumerator.rb: remove next? tests.
+
+	* test/ruby/test_continuation.rb: add a require 'continuation'.
+
+	* test/ruby/test_fiber.rb: add a require 'fiber'.
+
+Sat Aug 25 10:20:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (prelude.o): depends on vm_core.h now.
+
+	* common.mk (prelude.c): depends on tool/compile_prelude.rb too.
+
+	* common.mk (prereq): updates all auto-generated sources.
+
+	* tool/compile_prelude.rb: separated dynamic and static portions.
+
+Sat Aug 25 10:05:17 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* prelude.rb: add Thread.exclusive.  This class method
+	  is different from 1.8's.  Thread.exclusive only does
+	  synchronize with VM global mutex.
+
+Sat Aug 25 09:39:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_s_try_convert): more document description.
+
+	* re.c (rb_reg_s_try_convert): typo fixed.
+
+Sat Aug 25 08:54:12 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* id.h, id.c: remove idFuncall.
+
+	* compile.c (iseq_specialized_instruction): ditto.
+
+Sat Aug 25 08:47:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/compile_prelude.rb: fix to include "vm_core.h".
+
+Sat Aug 25 03:49:14 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/sentence.rb (Sentence): include Enumerable.
+	  (Sentence#each): defined.
+
+	* test/ruby/test_assignment.rb: use Sentence#expand.
+
+Sat Aug 25 03:08:57 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* prelude.rb: fix Mutex#synchronize definition.
+
+Sat Aug 25 02:08:45 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_s_try_convert): a new class method to convert
+	  object or nil if it's not target-type.  this mechanism is used
+	  to convert types in the C implemented methods.
+
+	* hash.c (rb_hash_s_try_convert): ditto.
+
+	* io.c (rb_io_s_try_convert): ditto.
+
+	* re.c (rb_reg_s_try_convert): ditto.
+
+	* string.c (rb_str_s_try_convert): ditto.
+
+Sat Aug 25 00:49:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_loop_generator.rb: added.
+
+Sat Aug 25 00:22:31 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* prelude.rb: added.  run this script on startup.
+
+	* tool/compile_prelude.rb: compile prelude.rb to C string.
+	  (prelude.rb -> prelude.c)
+
+	* common.mk: fix to build with prelude.c.
+
+	* inits.c (rb_call_inits): ditto.
+
+	* thread.c (Init_Thread): move definition of Mutex#synchronize
+	  to prelude.rb.
+
+Sat Aug 25 00:08:43 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_massign_opt): fix to skip massign optimization
+	  with global variables.
+
+	* bootstraptest/test_massign.rb: add some tests for above.
+
+Fri Aug 24 18:42:03 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_dir.rb (TestDir::setup): ?c now makes a string.
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner::initialize):
+	  initialize @workdir to stop warning.
+
+Fri Aug 24 18:30:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/drb/drb.rb (DRb::DRbServer::InvokeMethod::perform_without_block):
+	  replace funcall by send!.  other files in the distribution as well.
+
+Fri Aug 24 17:06:56 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_send_bang): abandon the name funcall for private
+	  aware method call.
+
+Fri Aug 24 15:27:12 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_next): stop pre-fetching.
+
+	* enumerator.c (Init_Enumerator): remove next? method.
+
+Fri Aug 24 15:14:57 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_loop): now handles StopIteration exception.
+
+Thu Aug 23 20:31:31 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: optimize simple massign.
+
+Thu Aug 23 20:02:25 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (reg_get_value): use RegQueryValueEx instead
+	  of RegQueryValueEx.
+
+	* ext/win32ole/win32ole.c (typelib_file_from_clsid): fix the bug
+	  that the function always returns Qnil.
+
+	* test/win32ole/test_win32ole_type.rb (test_initialize): add some test.
+
+Thu Aug 23 17:25:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (group_member): check if presents.
+
+	* configure.in (XCFLAGS): add _GNU_SOURCE on linux.
+
+	* file.c (group_member): use system routine if available.
+
+	* process.c: moved _GNU_SOURCE macro to Makefile.
+
+Thu Aug 23 16:59:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_massign), insnhelper.ci (vm_throw): not use C99
+	  comment.
+
+	* cont.c (rb_cont_call, fiber_switch, rb_fiber_resume, rb_fiber_yield):
+	  suppress warnings.
+
+	* cont.c (rb_fiber_start): change on non-volatile variable between
+	  setjmp and longjmp may not has an effect.
+
+	* enumerator.c (sym_call): initialize first.
+
+	* enumerator.c (enum_iter): typedefed.
+
+	* enumerator.c (next_i): suppress a warning.
+
+Thu Aug 23 16:04:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def, parse.y: fix massign order.  This change
+	  causes performance problem.  Try vm1_swap benchmark.
+	  [ruby-dev:31522]
+
+	* insns.def, insnhelper.ci: move process body of expandarray insn to
+	  vm_expandarray().
+
+	* bootstraptest/test_knownbug.rb, bootstraptest/test_massign.rb:
+	  move a solved test.
+
+Thu Aug 23 15:51:19 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_norm_arg): ripper has no shadowing check.
+
+	* parse.y (f_block_arg): dispatch blkarg_mark.
+
+Thu Aug 23 15:48:26 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (make_masgn_lhs, iseq_compile_each): fixed indent.
+
+	* compile.c (iseq_translate_threaded_code),
+	  vm_evalbody.ci (get_insns_address_table),
+	  template/vmtc.inc.tmpl (insns_address_table): constified.
+
+	* vm_evalbody.ci (vm_eval),
+	  template/insns_info.inc.tmpl (insn_stack_increase, insn_ret_num):
+	  suppress warnings.
+
+Thu Aug 23 13:19:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_norm_arg): check also nested arguments.  [ruby-dev:31502]
+
+Thu Aug 23 00:06:48 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* util.c: updated for ANSI C only.  applied a patch from
+	  <snakagawa AT infoteria.co.jp>.  [ruby-dev:31591]
+
+Thu Aug 23 00:04:45 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* missing/strtod.c: removed.  [ruby-dev:31588]
+
+	* LEGAL: updated.
+
+Wed Aug 22 15:59:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb:
+	  fix typo of filename (test_knownbug.rb).
+
+Wed Aug 22 14:04:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_knownbug.rb: added.  This file will contain
+	  test cases which point out known bug.  If bug is fixed, tests
+	  should move to the suitable place.
+
+	* bootstraptest/test_massign.rb: move a test which show known bug
+	  to test_knownbug.rb.
+
+Wed Aug 22 13:02:26 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (in_temporary_working_directory):
+	  don't remove the directory specified by --dir.
+
+Wed Aug 22 05:51:23 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* util.c (ruby_strtod): replaced by the implementation by David
+	  M. Gay inspired by William D. Clinger's paper "How to Read Floating
+	  Point Numbers Accurately" [Proc. ACM SIGPLAN '90, pp. 92-101].
+	  a patch from Satoshi Nakagawa <snakagawa AT infoteria.co.jp>.
+	  [ruby-dev:31582]
+
+	* test/ruby/test_float.rb (TestFloat::test_float): add test for
+	  precision.
+
+Wed Aug 22 03:51:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: add Fiber#resume and Fiber.yield.
+	  and Fiber::Core class to realize Coroutine.
+
+	* include/ruby/intern.h: declare rb_fiber_yield(), rb_fiber_resume().
+
+	* enumerator.c: use above api.
+
+	* test/ruby/test_fiber.rb: fix and add tests for above changes.
+
+Tue Aug 21 21:09:48 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb (Dir.mktmpdir): make directory suffix specifiable.
+	  suggested by knu.  [ruby-dev:31568]
+
+Tue Aug 21 15:00:23 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* st.c (st_clear): reset num_entries too.
+
+Tue Aug 21 13:57:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (st_foreach_func, rb_foreach_func): typedefed.
+
+	* hash.c (rb_hash_clear): use st_clear() unless iterating.
+
+	* st.c (struct st_table_entry): add new members, fore and back, to
+	  iterate in inserted order.
+
+	* include/ruby/st.h (struct st_table): ditto.
+
+Tue Aug 21 04:09:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm2_case.rb: add a new benchmark.
+	  YARV optimize case/when syntax.  If every conditions
+	  are literal (such as Symbol, Fixnum, String), dispatch
+	  calc order will be O(1).
+
+Tue Aug 21 04:08:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_enumerator.rb: fix last commit.
+
+Tue Aug 21 03:59:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test_enumerator.rb (enum_test): fix to return sorted
+	  array (for Hash test).
+
+Tue Aug 21 03:55:20 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* enumerator.c (next_i): fix to return with Fiber#yield at
+	  the end of each block.  [ruby-dev:31470]
+
+	* enumerator.c (enumerator_next_p): call init_next if not
+	  initialized.  [ruby-dev:31514]
+
+	* test/ruby/test_enumerator.rb: add tests for Enumerator.
+
+Mon Aug 20 23:28:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (Init_String): remove Symbol.intern and Symbol#dump.
+	  [ruby-dev:31525]
+
+	* dir.c (dir_foreach): return Enumerator if no block given.
+	  [ruby-dev:31525]
+
+	* io.c (rb_io_s_foreach): argument count check before making
+	  Enumerator.  [ruby-dev:31525]
+
+Mon Aug 20 23:17:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (struct rb_thread_struct): removed first_func_arg and
+	  reuse first_args instead.
+
+Sun Aug 19 13:31:40 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c: use VALGRIND_MAKE_MEM_UNDEFINED to detect use of collected
+	  objects if valgrind is available.  It cannot detect first 2 words
+	  because they are used as the free list.
+
+Sun Aug 19 13:13:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each/NODE_RESBODY): fix to add
+	  additional nop to prevent tailcall optimization.
+
+	* vm_opts.h: clean up comments.
+
+Sun Aug 19 12:58:39 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_call_bmethod): fix to propagate information
+	  that this proc is "from Method".  [ruby-dev:31490]
+
+	* proc.c (method_proc, rb_mod_define_method): ditto.
+
+	* vm.c (vm_invoke_proc_core): removed.
+
+	* vm_core.h: ditto.
+
+Sun Aug 19 12:36:11 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/sentence.rb: new method Sentence().
+
+Sun Aug 19 12:32:39 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_callee_setup_arg): fix to mark enough VM stack.
+	  [ruby-dev:31492]
+
+Sat Aug 18 19:02:34 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_yield.rb (TestYieldGen): relax array size check
+	  in nested parameters.
+	  [ruby-dev:31485]
+
+Sat Aug 18 17:44:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c (Init_tcltklib): use rb_set_end_proc().
+
+Sat Aug 18 17:40:11 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (is_ruby_native_thread): check properly.  [ruby-dev:31166]
+
+Sat Aug 18 16:44:15 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_call_bmethod),
+	  vm.c (vm_invoke_proc_core): fix to do not restore
+	  $SAFE when proc invoked by bmethod.
+
+	* vm_core.h: ditto.
+
+Sat Aug 18 16:44:49 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.ci (ruby_error_print): call error_print.
+
+	* eval_jump.ci, process.c (rb_exit, rb_f_exit, rb_f_abort): moved.
+
+Sat Aug 18 15:59:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (detach_process_watcher): should not pass the pointer
+	  to an auto variable to the thread to be created.  pointed and
+	  fix by KUBO Takehiro <kubo at jiubao.org>  [ruby-dev:30618]
+
+Sat Aug 18 15:52:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insnhelper.ci (vm_send_optimize): use MEMMOVE to shift values inside
+	  an array.
+
+Sat Aug 18 15:45:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insnhelper.ci (vm_yield_setup_args): rsize and psize should not be
+	  negative.
+
+Sat Aug 18 14:35:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insnhelper.ci (vm_callee_setup_arg, vm_send_optimize,
+	  vm_yield_setup_args): bulk copy for arguments.
+
+Sat Aug 18 13:55:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix to allow self.x=
+	  if x= is private.
+
+	* bootstraptest/test_method.rb: add a test for above.
+
+Sat Aug 18 14:05:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_compile_gen): obtain error info from errinfo.
+
+	* re.c (rb_reg_error_desc): make RegexpError for initialization error.
+
+	* re.c (rb_reg_compile): return nil and set errinfo if error.
+
+Sat Aug 18 13:23:01 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c: $! should not be writable.
+
+	* eval.c (rb_rubylevel_errinfo): added.  rb_errinfo() returns
+	  rb_thread_t#errinfo.  rb_rubylevel_errinfo() returns $! value.
+
+Sat Aug 18 13:14:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_callee_setup_arg): fix to check arguments
+	  correctly. [ruby-dev:31472]
+
+Sat Aug 18 13:13:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_exception.rb: add escape character ("\") for
+	  editors.
+
+Sat Aug 18 12:42:50 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_yield.rb (TestYieldGen): add test for yielding to
+	  lambda using lambda parameter passing emulator.
+
+Sat Aug 18 12:24:30 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/test.rb, test/ruby/test_system.rb(valid_syntax?): keep
+	  comment lines first.
+
+Sat Aug 18 11:44:59 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_set_arguments), insnhelper.ci
+	  (vm_callee_setup_arg, vm_yield_setup_args):
+	  fix to cause raise on "lambda{|a|}.call(1, 2)".
+	  [ruby-dev:31464]
+
+	* bootstraptest/test_block.rb: add tests for above.
+
+Sat Aug 18 01:12:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (frame_func_id): return proper method ID.
+
+Fri Aug 17 22:43:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): massign should return rvalue(s).
+
+Fri Aug 17 22:10:19 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_basicinstructions.rb: fix old tests.
+	  class variables should be inherited.
+
+Fri Aug 17 21:20:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_call), eval_method.ci (rb_add_method, rb_alias),
+	  insnhelper.ci (vm_call_method): fix to save safelevel for
+	  method node.
+
+	* include/ruby/node.h: ditto.
+
+	* bootstraptest/test_method.rb: add a test for above.
+
+Fri Aug 17 16:02:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_terminate_all): fix to ignore
+	  exceptions.
+
+	* thread.c (thread_start_func_2): fix
+	  abort_on_exception process.  [ruby-dev:31394]
+
+Fri Aug 17 14:38:36 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (in_temporary_working_directory): use
+	  Dir.mktmpdir to create and remove temporary directory.
+	  (Dir.mktmpdir): define if not available.
+	  [ruby-dev:31431]
+
+Fri Aug 17 03:07:37 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix throw insn option of next.
+
+Fri Aug 17 01:25:23 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_set_arguments), insnhelper.ci
+	  (vm_callee_setup_arg, vm_yield_setup_args): fix
+	  block parameter problems. [ruby-dev:31437], [ruby-dev:31440]
+
+	* bootstraptest/test_block.rb: add a test of [ruby-dev:31440].
+
+Fri Aug 17 01:24:12 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (ruby_iseq_disasm): fix to show arg_simple value.
+
+Fri Aug 17 01:21:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (throw): insert a RUBY_VM_CHECK_INTS(). [ruby-dev:31361]
+
+Thu Aug 16 20:40:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bigtrunc): RBIGNUM(x)->len may be zero.  out of bound
+	  access.  [ruby-dev:31404]
+
+	* sprintf.c (rb_str_format): small float should not call
+	  rb_dbl2big().
+
+Thu Aug 16 22:10:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix next/redo stack consistency.
+	  [ruby-dev:31373]
+
+	* bootstraptest/test_syntax.rb: add tests for above.
+
+	* sample/test.rb: fix to use __FILE__ instead of $0 to know basedir.
+
+Thu Aug 16 21:14:06 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (BASERUBY): need AC_SUBST.  [ruby-dev:31438]
+
+Thu Aug 16 19:18:26 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/sentence.rb: Sentence class implemented
+	  based on sentgen.rb
+
+	* test/ruby/sentgen.rb: removed.
+
+	* test/ruby/test_assignment.rb: use sentence.rb.
+
+	* test/ruby/test_yield.rb: block parameter passing emulator
+	  implemented.
+
+Thu Aug 16 16:48:47 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (BASERUBY): check if base ruby is runnable first.
+	  [ruby-core:11900]
+
+Thu Aug 16 16:46:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (aix): enable shared by default.
+
+	* configure.in (aix): for 64bit-mode AIX.  [ruby-dev:31401]
+	  + use CC for LDSHARED if non-gcc,
+	  + moved -G option from *LDFLAGS to LDSHARED,
+	  + set -brtl only in XLDFLAGS.
+
+Thu Aug 16 13:06:08 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_lshift): make shift offset long type.
+	  (big_rshift): ditto.
+	  (rb_big_lshift): ditto.
+	  (big_rshift): ditto.
+	  [ruby-dev:31434]
+
+Thu Aug 16 06:29:08 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_readpartial): argf_forward needs argc and argv.
+
+Thu Aug 16 02:47:39 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_m_yield): added.  use this function
+	  for Fiber#yield instead of rb_fiber_yield.
+
+Thu Aug 16 00:36:52 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb (Dir.mktmpdir): new method.  [ruby-dev:31416]
+
+Wed Aug 15 18:57:30 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (os_live_obj): fix to skip T_VALUES.
+
+	* sample/test.rb: add an ObjectSpace test.
+
+Wed Aug 15 16:49:04 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* inits.c (rb_call_inits): change initializing order.
+	  [ruby-dev:31420]
+
+Wed Aug 15 16:44:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (ARGF_FORWARD): wrongly compares with current_file with
+	  rb_stdout.  should be rb_stdin.  [ruby-cvs:20177]
+
+Wed Aug 15 14:59:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_initialize, argf_each_line, argf_each_byte): suppress
+	  warnings.
+
+Wed Aug 15 14:22:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: remove "//" type comment.
+
+Wed Aug 15 13:42:15 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: fix rules around f_margs.  "make test" passes all tests.
+
+	* bootstraptest/test_block.rb: add some tests for above.
+
+Wed Aug 15 13:50:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_delete_key): delete the entry without calling block.
+
+	* hash.c (rb_hash_shift): should consider iter_lev too.
+
+	* hash.c (delete_if_i): use rb_hash_delete_key() so that the block
+	  isn't called twice.  [ruby-core:11556]
+
+Wed Aug 15 13:39:25 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* process.c (proc_geteuid): fix strange conversion.  [ruby-dev:31417]
+
+Wed Aug 15 01:05:55 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_check): don't call newtest.
+	  (assert_equal): call newtest.
+	  (assert_match): ditto.
+
+Tue Aug 14 21:43:39 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/runner.rb (assert_check): new method.
+	  (assert_match): new method.
+	  (assert_equal): use assert_check.
+	  (pretty): give failure description as an argument.
+
+	* bootstraptest/test_exception.rb: use assert_match to describe the
+	  test for [ruby-dev:31407].  [ruby-dev:31412]
+
+Tue Aug 14 19:53:15 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (Init_Proc), eval.c (Init_eval), eval_intern.h: move
+	  init place of exception_error.
+
+	* inits.c: ditto.
+
+	* eval.c (Init_eval): set exception_error#throwed_state as TAG_FATAL.
+	  [ruby-dev:31407]
+
+	* bootstraptest/test_exception.rb: add a test for above.
+
+Tue Aug 14 19:51:20 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: change test order (test -> btest).
+
+Tue Aug 14 00:04:27 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: 0.1.8 -> 0.1.9.
+
+	* test/rss/test_version.rb: followed the above change.
+
+	* lib/rss/parser.rb: fixed a bug that handles unintended elements.
+	  Thanks to Takuo Yonezawa. [ruby-list:43841]
+
+Mon Aug 13 17:23:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_clear_trace_func, rb_thread_stop_timer_thread):
+	  declarations for forward references.
+
+	* eval.c (rb_longjmp, eval): use local variable.
+
+	* eval.c (rb_longjmp): string object not to be optimized.
+
+Mon Aug 13 13:21:58 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: make ftp passive mode to avoid NAT problem.
+	  [ruby-dev:31377]
+
+Mon Aug 13 09:18:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_getc): should returns
+	  one-character string.
+
+	* ext/stringio/stringio.c: remove unnecessary prototypes.
+
+	* ext/stringio/stringio.c (strio_getbyte): new method.
+
+	* ext/stringio/stringio.c (strio_readbyte): new method.
+
+	* ext/stringio/stringio.c (strio_ungetc): should take a string as
+	  an input.
+
+Mon Aug 13 08:19:43 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_close): always close via method.
+
+	* io.c (Init_IO): remove obsolete Kernel#getc.
+
+Mon Aug 13 05:03:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_raise): check if target thread is
+	  thrown by another thread or not.  [ruby-dev:31371]
+
+	* bootstraptest/test_thread.rb: add a test for above.
+
+Mon Aug 13 04:35:30 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_peephole_optimize): fix peephole optimization
+	  bug. [ruby-dev:31360]
+
+	* bootstraptest/test_syntax.rb: add a test for above.
+
+Mon Aug 13 04:02:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c (debug_print_pre): fix to show control frame count.
+
+	* insns.def (opt_call_c_function): fix operand type.
+
+	* lib/vm/instruction.rb: ditto.
+
+	* insnhelper.ci (vm_push_frame, vm_pop_frame): fix to show
+	  control stack status on if VMDEBUG == 2.
+
+	* vm.h: add a comment about VMDEBUG.
+
+	* iseq.c (find_prev_line_no): fix to skip bug report if
+	  line is not found.
+
+	* lib/vm/instruction.rb: fix to use build_string() on
+	  source code generators.
+
+Mon Aug 13 03:57:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* template/yasmdata.rb.tmpl: fix type and name.
+
+Sat Aug 11 23:27:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_each_line): should use #each_line, not #each.
+
+	* io.c (argf_each_line): simplified.
+
+	* io.c (argf_getline): should handle non T_FILE object in ARGV.
+
+	* io.c (argf_each_byte): each_byte should yield bytes not
+	  one-character strings.  [ruby-dev:31374]
+
+Sat Aug 11 07:24:55 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: reverted some wrongly erased "o" options
+	  (pointed out by nobu).
+
+Sat Aug 11 00:01:12 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): not re-raise to main thread if it is
+	  joining the current thread.
+
+Fri Aug 10 23:54:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_create_core): inherit the priority of creating
+	  thread.  submitted at [ruby-core:11873] by David Flanagan <david AT
+	  davidflanagan.com>.  [ruby-core:11876]
+
+Fri Aug 10 05:12:19 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): let abort_on_exception work.
+	  [ruby-core:11873]
+
+Fri Aug 10 04:47:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (internal_id_gen): internal ID must be bigger than
+	  tLAST_TOKEN.
+
+Thu Aug  9 16:04:55 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt): check if rhs has value before assignment instead
+	  inside node_assign_gen().  [ruby-dev:31293]
+
+	* parse.y (call_bin_op_gen, call_uni_op_gen): split call_op_gen.
+
+Thu Aug  9 14:01:36 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_marg_list): renamed from f_marg_head.
+
+	* parse.y (f_margs): allow multiple mandatory arguments after a splat.
+	  [ruby-dev:31153]
+
+Thu Aug  9 02:02:02 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_cycle): typo fixed.  a patch from Kazuhiro
+	  NISHIYAMA <zn AT mbf.nifty.com>.  [ruby-dev:31362]
+
+Wed Aug  8 19:17:40 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/dl/cptr.c (rb_dlptr_s_to_ptr): no longer check
+	  HAVE_RB_IO_STDIO_FILE.
+
+Wed Aug  8 15:52:01 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_next_p): should check correctly even when
+	  e.next has not been called before.
+
+	* enumerator.c (enumerator_next): raise StopIteration (name taken
+	  from Python) instead of IndexError.
+
+	* enum.c (enum_zip): catch StopIteration exception.
+
+	* enumerator.c (enumerator_with_index): return Enumerator if no
+	  block is given.
+
+	* test/ruby/test_iterator.rb (TestIterator::test_enumerator): add
+	  test for enumerators.
+
+Wed Aug  8 11:48:37 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): should not use RTEST for non-VALUE.
+
+Wed Aug  8 11:25:25 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): should preserve sign mark.
+
+Wed Aug  8 11:02:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (btest-miniruby, test-sample): split the test target
+	  so that -k option works.
+
+Tue Aug  7 14:58:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (establishShell): handshaking before close slave
+	  device.  [ruby-talk:263410]
+
+	* ext/pty/pty.c (MasterDevice, SlaveDevice, deviceNo): constified.
+
+	* ext/pty/pty.c (SlaveName): removed static buffer.
+
+	* ext/pty/expect_sample.rb: support for autologin.
+
+Tue Aug  7 13:58:03 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ruby_special_consts): added RUBY_SPECIAL_SHIFT.
+
+	* .gdbinit: some improvements.
+
+Tue Aug  7 13:28:32 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (node_name.inc): use $? instead of $< for nmake.
+	  [ruby-dev:31356]
+
+Tue Aug  7 12:45:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func_isinf): set yes also on OpenSolaris.
+	  [ruby-Bugs-12859]
+
+Tue Aug  7 12:31:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): fix for win32 platforms.
+
+Tue Aug  7 02:58:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): make Bignum#to_s even faster.  a patch
+	  from Kenta Murata <muraken AT gmail.com>.  [ruby-dev:31354]
+
+Tue Aug  7 01:42:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_zip): zip no longer converts arguments into
+	  arrays, uses enumerators.
+
+Tue Aug  7 01:27:47 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* cont.c (rb_fiber_yield): change argument ordering.  export.
+
+	* cont.c (rb_fiber_current): export
+
+	* include/ruby/intern.h: export several functions from cont.c.
+
+	* enumerator.c (enumerator_next): new method to implement external
+	  iterator (generator) using fiber.
+
+	* enumerator.c (enumerator_next_p): new method to check whether
+	  any element is left in the generator sequence.
+
+	* enumerator.c (enumerator_rewind): a new method to rewind the
+	  generator sequence.
+
+Tue Aug  7 01:15:24 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_cycle): new method to cycle enumerable forever.
+
+Tue Aug  7 00:05:38 2007  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* irb/ruby-lex.rb: support for '\c'. [ruby-talk:263508]
+
+Mon Aug  6 20:29:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci, insns.def: move some statements to functions.
+
+	* vm.c, vm.h, vm_evalbody.ci: fix include/typedef places.
+
+Mon Aug  6 18:41:12 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/vm/instruction.rb (make_header_analysys): fix last commit.
+
+Mon Aug  6 18:33:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/vm/instruction.rb (make_header_analysys): add to separate
+	  header addition process.
+
+Mon Aug  6 17:36:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rexml/encodings/{ISO-8859-15,CP-1252}.rb: fixed invalid syntax.
+
+Mon Aug  6 16:57:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* sample/test.rb: fix to output file name if it contains
+	  invalid syntax.
+
+Mon Aug  6 16:41:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (value_expr_gen): fix to cause "void value expression"
+	  when jump expression such as "next" are shown on value_expr().
+	  [ruby-dev:31119]
+
+	* bootstraptest/test_syntax.rb: fix to above change.
+
+Mon Aug  6 14:36:30 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: fix a f_marg rule. [ruby-dev:31160]
+
+Mon Aug  6 14:29:30 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb (assert_equal): add additional
+	  message parameter.
+
+Mon Aug  6 13:34:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (INSNS): not chdir to srcdir.
+
+	* common.mk (node_name.inc): auto-generate node name list.
+
+	* iseq.c (ruby_node_name): ditto.
+
+	* iseq.c (iseq_s_compile_option_get, Init_ISeq): added a new
+	  method VM::InstructionSequence::compile_option.
+
+	* lib/vm/instruction.rb (RubyVM::SourceCodeGenerator): --destdir
+	  option.
+
+	* tool/node_name.rb: to auto-generate node name list.
+
+Sun Aug  5 11:51:39 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss, sample/rss, test/rss:
+	- 0.1.7 -> 0.1.8.
+	- supported <itunes:XXX>.
+	- reverted backward incompatibility API changes introduced 0.1.7.
+
+Sun Aug  5 04:56:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open_v, pipe_open_s): separate array and string
+	  cases.  [ruby-dev:31344]
+
+Fri Aug  3 11:05:54 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): save all CONFIG values.
+
+	* ext/extmk.rb (extmake): remove mkmf.log at clean, and extconf.h at
+	  distclean, respectively.
+
+	* ext/extmk.rb: remove rdoc at clean, and installed list file at
+	  distclean, respectively.
+
+Fri Aug  3 07:09:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: more verbose message.  [ruby-Bugs-12766]
+
+	* lib/mkmf.rb (have_type): suppress a warning with -Wall.
+
+	* lib/mkmf.rb (find_type): new method.
+
+Fri Aug  3 00:00:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (big2str_table): base cannot be 0 or 1.
+
+Thu Aug  2 23:42:57 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_compile_gen): set error if failed to compile regexp
+	  literal.  [ruby-dev:31336]
+
+	* re.c (option_to_str, arg_kcode, opt_kcode): options conversion
+	  between int and string.
+
+	* re.c (rb_reg_compile): should not use regexp which could not get
+	  initialized.  [ruby-dev:31333]
+	  return error message to let the parser know it.
+
+	* re.c (rb_reg_compile): append regexp options to error message.
+	  [ruby-dev:31334]
+
+Thu Aug  2 22:05:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): faster Bignum#to_s using Karatsuba
+	  algorithm.  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>
+	  in [ruby-dev:31312], slightly modified by Kenta Murata
+	  <muraken AT gmail.com> in [ruby-dev:31339].
+
+Thu Aug  2 13:46:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): should not check positional number as
+	  width.  [ruby-core:11838]
+
+Wed Aug  1 12:40:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* generic.rb (URI::Generic::merge_path): behave as RFC 3986.
+	  [ruby-talk:252052]
+
+Tue Jul 31 23:38:09 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (_parse): now interprets slashed numerical
+	  dates as a big endian (except dd/mm/yyyy). [experimental]
+
+Mon Jul 30 11:16:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_aref): check for Bignum index range.
+	  [ruby-dev:31271]
+
+Sat Jul 28 09:35:41 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/digest/lib/digest.rb (Digest::self.const_missing): avoid
+	  infinite recursive const_missing call.  [ruby-talk:262193]
+
+Thu Jul 26 20:40:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_eqq): call_super() in === does not work well
+	  since Enumerable#=== has different behavior.  [ruby-dev:31296]
+
+Thu Jul 26 13:57:45 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (load_1, dln_find_1): constified.
+
+	* dln.c (conv_to_posix_path): removed.
+
+	* ruby.c (usage): constified.
+
+	* ruby.c (rubylib_mangled_path, rubylib_mangled_path2): return
+	  VALUE instead of a pointer to static buffer.
+
+	* ruby.c (push_include_cygwin): fixed buffer overflow.
+	  [ruby-dev:31297]
+
+	* ruby.c (ruby_init_loadpath): not convert built-in paths.
+
+Tue Jul 24 10:37:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_f_p): return nil if no argument.  [ruby-dev:31285]
+
+Tue Jul 24 01:05:07 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regint.h (USE_MATCH_RANGE_IS_COMPLETE_RANGE): undef to achieve old
+	  rindex behavior.  [ruby-dev:31265]
+
+Mon Jul 23 18:37:14 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* include/ruby/oniguruma.h: upgrade to Oniguruma 5.9.0.  fixes
+	  some memory violation.  [ruby-dev:31070]
+
+Sun Jul 22 20:09:49 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (Date._parse): now accepts some new
+	  hints. [experimental]
+
+	* lib/parsedate.rb: followed the changes on
+	  lib/date/format.rb. [experimental]
+
+Sun Jul 22 16:06:56 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (is_ruby_native_thread): made an int function as
+	  well as version 1.8.
+
+	* include/ruby/ruby.h (is_ruby_native_thread): moved prototype
+	  from intern.h as well as version 1.8.
+
+Sun Jul 22 14:33:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_rename): deleted code to get rid of a bug of
+	  old Cygwin.
+
+	* file.c (rb_file_truncate): added prototype of GetLastError()
+	  on cygwin.  [ruby-dev:31239]
+
+	* include/ruby/intern.h (is_ruby_native_thread): prototype.
+
+	* missing/strftime.c (strftime): fix printf format and actual
+	  arguments.
+
+	* ext/Win32API/Win32API.c (Win32API_initialize): ditto.
+
+	* ext/tk/tcltklib.c (ip_finalize): ditto.
+
+	* ext/win32ole/win32ole.c (lcid_installed): ditto.
+
+	* ext/socket/getnameinfo.c: include stdio.h always.
+
+Sat Jul 21 21:39:12 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb (Date._parse): now can take some
+	  hints (its aim must be mainly determination of endianness of
+	  date). [experimental]
+
+	* lib/date.rb, lib/date/format.rb (Date._parse): now completes
+	  truncated year as default action. [experimental]
+
+	* lib/date.rb, lib/date/format.rb: added ::iso8601, ::rfc3339,
+	  ::xmlschema, ::rfc2822, ::httpdate, ::jisx0301, #xmlschema,
+	  #httpdate. [experimental]
+
+Sat Jul 21 17:48:26 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: inverted rules order.
+
+	* thread_win32.ci (w32_create_thread): bcc does not have
+	  _beginthreadex().
+
+	* lib/mkmf.rb (create_makefile): make OBJS depend on RUBY_EXTCONF_H
+	  only if extconf.h is created.
+
+	* bcc32/Makefile.sub: headers have moved.
+
+	* bcc32/{Makefile.sub,configure.bat,setup.mak: configure_args
+	  support.
+
+	* bcc32/setup.mak: check runtime version.
+
+	* win32/win32.c (rb_w32_open_osfhandle): prototype has changed
+	  in bcc 5.82.
+
+	* {win32,wince,bcc32}/setup.mak (-version-): no RUBY_EXTERN magic.
+
+	* win32/resource.rb: include patchlevel number.
+
+Sat Jul 21 12:06:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (init_mkmf): should remove mkmf.log too.
+
+Sat Jul 21 01:45:03 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (Date._parse): completes calendar week based year.
+
+	* lib/date/format.rb (Date._parse): detects year of ordinal date in
+	  extended format.
+
+Fri Jul 20 16:30:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_sequence): raise SyntaxError instead of rb_bug
+	  since this function can be called from VM::InstructionSequence.load.
+
+	* compile.c (insn_set_sc_state, iseq_set_sequence_stackcaching): ditto.
+
+Fri Jul 20 16:11:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c, compile.h (DECL_ANCHOR, INIT_ANCHOR): split not to
+	  initialize aggregations with dynamic values.  [ruby-talk:259306]
+
+	* eval.c (rb_protect): not to initialize aggregations with dynamic
+	  values.  [ruby-talk:259306]
+
+	* gc.c (mark_current_machine_context): ditto.
+
+	* thread.c (thgroup_list, call_trace_func): ditto.
+
+	* vm.c (vm_init_redefined_flag): ditto.
+
+Fri Jul 20 15:22:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_config.c (ossl_config_set_section): do not
+	  initialize aggregations with dynamic values.  [ruby-talk:259306]
+
+Fri Jul 20 10:39:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt, mlhs_node, lhs, arg, var_ref): return dummy
+	  NODE_BEGIN after errors.  [ruby-dev:31100], [ruby-dev:31118]
+
+	* parse.y (remove_begin): keep empty NODE_BEGIN, instead of null.
+	  [ruby-dev:31252], [ruby-dev:31263]
+
+Fri Jul 20 09:50:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/runner.rb (get_result_string): check $?.coredump?
+	  first.
+
+	* bootstraptest/runner.rb (cleanup_coredump, check_coredump): see
+	  stackdump file too.
+
+Thu Jul 19 20:39:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (value_expr_gen): warn for empty expression ().
+	  [ruby-dev:31252]
+
+Thu Jul 19 19:24:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.ci (get_backtrace): check the result more.
+	  [ruby-dev:31261] [ruby-bugs-12398]
+
+Thu Jul 19 14:38:45 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_lshift, rb_big_rshift): separated functions
+	  to get rid of infinite recursion.  fixed calculation in edge
+	  cases.  [ruby-dev:31244]
+
+	* numeric.c (rb_fix_lshift, rb_fix_rshift): ditto.
+
+Wed Jul 18 16:57:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_pow): refine overflow check.  [ruby-dev:31242]
+
+Wed Jul 18 09:19:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_parser_append_print, rb_parser_while_loop): moved check
+	  for node to the head.
+
+	* ruby.c (proc_options): do nothing for -p/-n options if tree is null.
+	  submitted by Yusuke ENDOH <mame AT tsg.ne.jp> at [ruby-dev:31243].
+
+Wed Jul 18 08:47:09 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_succ): Time#succ should return a time object in the
+	  same timezone mode to the original.  [ruby-talk:260256]
+
+Mon Jul 16 23:07:51 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/base64.rb (Base64::b64encode): should not specify /o option
+	  for regular expression.  [ruby-dev:31221]
+
+Mon Jul 16 22:57:53 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): make %u behave like %d for negative
+	  values, since decimal format does not work with preceding dots.
+	  [ruby-core:11575]
+
+Mon Jul 16 18:29:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_rindex_m): accept string-like object convertible
+	  with #to_str method, as well as rb_str_index_m.  [ruby-core:11692]
+
+Mon Jul 16 07:17:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (getspecial): lfp_svar_get() requires int for special
+	  global variables.
+
+Mon Jul 16 05:45:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): more checks for format argument.
+	  [ruby-core:11569], [ruby-core:11570], [ruby-core:11571],
+	  [ruby-core:11573]
+
+Mon Jul 16 00:26:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_pow): removed invariant variable.  [ruby-dev:31236]
+
+Sun Jul 15 22:24:37 2007  pegacorn  <subscriber.jp AT gmail.com>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): adjust format.  [ruby-dev:31222]
+
+	* ext/digest/digest.c (rb_digest_instance_update,
+	  rb_digest_instance_finish, rb_digest_instance_reset,
+	  rb_digest_instance_block_length): %s in rb_raise() expects char*.
+	  [ruby-dev:31222]
+
+	* ext/openssl/ossl.h: include ossl_pkcs5.h.  [ruby-dev:31231]
+
+	* ext/openssl/ossl_pkcs5.h: new file for PKCS5.  [ruby-dev:31231]
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_to_s): use ossl_raise()
+	  instead of rb_raise().  [ruby-dev:31222]
+
+	* ext/sdbm/_sdbm.c: DOSISH platforms need io.h.  [ruby-dev:31232]
+
+	* ext/syck/syck.h: include stdlib.h for malloc() and free().
+	  [ruby-dev:31232]
+
+	* ext/syck/syck.h (syck_parser_set_input_type): prototype added.
+	  [ruby-dev:31231]
+
+	* win32/win32.c: include mbstring.h for _mbspbrk().  [ruby-dev:31232]
+
+	* include/ruby/win32.h (rb_w32_getcwd): prototype added.
+	  [ruby-dev:31232]
+
+Sun Jul 15 21:07:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (bigtrunc): do not empty Bignum.  [ruby-dev:31229]
+
+Sun Jul 15 19:05:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_cstr_to_inum): check leading non-digits.
+	  [ruby-core:11691]
+
+Sun Jul 15 04:42:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (get2comp): do nothing for empty Bignum.  [ruby-dev:31225]
+
+Sat Jul 14 22:49:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_pow): integer power calculation: 0**n => 0,
+	  1**n => 1, -1**n => 1 (n: even) / -1 (n: odd).
+
+	* test/ruby/test_fixnum.rb (TestFixnum::test_pow): update test
+	  suite.  pow(-3, 2^64) gives NaN when pow(3, 2^64) gives Inf.
+
+Sat Jul 14 18:46:35 2007  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: add --with-valgrind.
+
+	* gc.h (SET_MACHINE_STACK_END): new macro to replace
+	  rb_gc_set_stack_end.  it find out accurate stack boundary by
+	  asm using gcc on x86.
+
+	* thread.c (rb_gc_set_stack_end): don't define if asm-version
+	  SET_MACHINE_STACK_END is available.
+
+	* gc.c (mark_current_machine_context): extracted from garbage_collect.
+	  it use SET_MACHINE_STACK_END to not scan out of stack area.
+	  it notify conservative GC information to valgrind if
+	  --with-valgrind.
+
+Sat Jul 14 14:04:06 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (sort_by_cmp): check if reentered.  [ruby-dev:24291]
+
+Sat Jul 14 11:08:28 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/json/ext/generator/generator.c (check_max_nesting): wrong
+	  format specifier.  a patch from pegacorn <subscriber.jp AT gmail.com>.
+	  [ruby-dev:31217]
+
+Sat Jul 14 02:27:43 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_pow): overflow detection using FIT_SQRT_LONG().
+	  [ruby-dev:31215]
+
+Sat Jul 14 02:05:53 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (opt_div): LONG2FIX() may not work for corner cases,
+	  use LONG2NUM() instead.  [ruby-dev:31210]
+
+Sat Jul 14 00:34:01 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_round): should not return false, but self.
+	  [ruby-dev:31212]
+
+Fri Jul 13 18:31:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/intern.h: remove unused function declarations.
+
+	* include/ruby/ruby.h: ditto.
+
+Fri Jul 13 17:32:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm.c (vm_free): clear free'ed living_threads field.
+	  [ruby-dev:31163]
+
+	* insns.def (opt_succ): use cast to shut a warning up.
+
+Fri Jul 13 16:10:00 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (URI::Generic#find_proxy): use ENV.to_hash to access
+	  http_proxy environment variable to avoid case insensitive
+	  environment search.
+
+Fri Jul 13 15:02:15 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): enclose command line except for
+	  command.com which can not handle quotes.  [ruby-talk:258939]
+
+Fri Jul 13 11:33:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_max, range_min): return nil for empty set as well as
+	  1.8 and Enumerable.  [ruby-dev:31198]
+
+Fri Jul 13 11:28:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (bvar): semicolon was lost for ripper description.
+	  [ruby-dev:31140]
+
+Fri Jul 13 11:25:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (program, stmt, primary): reduced duplicated code.
+
+	* parse.y (dsym): convert also literals containing NUL to
+	  symbol.
+
+	* parse.y (debug_lines): use rb_hash_lookup() to get rid of
+	  call of Hash#default.
+
+	* parse.y (ripper_warningS): unused in ripper right now.
+
+Fri Jul 13 10:33:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_flatten_bang): check argument if valid
+	  integer.  [ruby-dev:31197]
+
+Fri Jul 13 10:10:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (link_command, cc_command, cpp_command): do not expand
+	  ::CONFIG which is an alias of MAKEFILE_CONFIG.
+
+Thu Jul 12 21:38:02 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* insns.def (opt_succ): use LONG_MAX as maximum Fixnum VALUE.
+	  [ruby-dev:31199]
+
+Thu Jul 12 18:42:18 2007  Tanaka Akira  <akr@fsij.org>
+
+	* range.c (range_max): use FIX2LONG instead of FIX2INT to avoid
+	  RangeError by ((-0x80000001)...(-0x80000001)).max on LP64.
+
+	* insns.def (opt_plus): use FIX2LONG instead of FIX2INT to avoid
+	  RangeError by 0x3fffffffffffffff+1 on LP64.
+
+	* insns.def (opt_succ): don't use 0x80000000 which assumes 32bit VALUE.
+	  use FIX2LONG instead of FIX2INT.
+	  [ruby-dev:31190]
+
+Thu Jul 12 17:03:15 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (rb_struct_init_copy): disallow changing the size.
+	  [ruby-dev:31168]
+
+Thu Jul 12 12:58:21 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c: remove "yarv" prefix.
+
+	* array.c, numeric.c: ditto.
+
+	* insnhelper.ci, insns.def, vm_evalbody.ci: ditto.
+
+	* yarvcore.c: removed.
+
+	* yarvcore.h: renamed to core.h.
+
+	* cont.c, debug.c, error.c, process.c, signal.c : ditto.
+
+	* ext/probeprofiler/probeprofiler.c: ditto.
+
+	* id.c, id.h: added.
+
+	* inits.c: ditto.
+
+	* compile.c: rename internal functions.
+
+	* compile.h: fix debug flag.
+
+	* eval.c, object.c, vm.c: remove ruby_top_self.
+	  use rb_vm_top_self() instead.
+
+	* eval_intern.h, eval_load: ditto.
+
+	* gc.c: rename yarv_machine_stack_mark() to
+	  rb_gc_mark_machine_stack().
+
+	* insnhelper.h: remove unused macros.
+
+	* iseq.c: add iseq_compile() to create iseq object
+	  from source string.
+
+	* proc.c: rename a internal function.
+
+	* template/insns.inc.tmpl: remove YARV prefix.
+
+	* thread.c: use rb_iseq_eval() and rb_str_new2().
+
+	* vm.c (rb_iseq_eval): added.
+
+	* vm.c: move some functions from yarvcore.c.
+
+	* vm_dump.c: fix to remove compiler warning.
+
+Thu Jul 12 12:24:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (opt_succ): fixed typo.  [ruby-dev:31189]
+
+Thu Jul 12 10:30:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): moved prototye from thread_*.ci.
+
+	* thread_pthread.ci (thread_start_func_2): not use a directive
+	  inside a macro argument.  [ruby-talk:258763]
+
+	* thread.c (thread_join): pthread_t may not be pointer.
+
+	* thread_pthread.ci (ubf_select_each): ditto.
+
+Thu Jul 12 05:32:28 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (FIX2ULONG): drop sign bit for LLP64 platform.
+
+Tue Jul 10 19:34:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c (rb_hash_lookup): added.  this function is similar to
+	  rb_hash_aref(), but doesn't call Hash#default when no entry
+	  exists.
+
+	* include/ruby/intern.h: ditto.
+
+	* insnhelper.ci (lfp_svar_get): use rb_hash_lookup().
+
+Tue Jul 10 19:16:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, insnhelper.ci, vm.c: change cref index (-1 -> 2).
+
+Tue Jul 10 18:49:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (exec_under): add proper casts.
+
+Tue Jul 10 16:58:16 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, insnhelper.ci: fix svar interface.
+
+	* compile.c (iseq_compile_each), yarvcore.h: fix to use new
+	  svar interface for flip flop.
+
+	* eval.c: ditto.
+
+	* insns.def: ditto.
+
+	* include/ruby/intern.h: remove "rb_svar()" declaration.
+
+Tue Jul 10 16:52:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (rb_iseq_compile): formatted if/else to switch statement.
+
+Tue Jul 10 15:57:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/test_flip.rb: new test for flip-flop operator.
+
+Tue Jul 10 14:50:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bcc32/{Makefile.sub,setup.mak}: remove surplus slash from srcdir.
+
+Mon Jul  9 02:17:36 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* cont.c (cont_restore_1): workaround for x64-mswin64's SEH.
+
+Sun Jul  8 02:08:53 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/json.rb, lib/json/, ext/json/: import JSON 1.1.1
+
+Sat Jul  7 21:59:29 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb (PP::PPMethods#pp_hash): sort condition changed:
+	  all keys have a same class which is kind of Comparable.
+
+Sat Jul  7 17:12:37 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: use rb_bug() instead of rb_compile_error().
+
+Sat Jul  7 16:12:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: fix node construction (around f_margs).
+	  [ruby-dev:31143]
+
+	* bootstraptest/test_block.rb: add a test for above.
+
+	* insnhelper.ci: fix indent.
+
+Sat Jul  7 15:36:50 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb (PP::PPMethods#pp_hash): sort if
+	  all keys are strings, symbols or integers.
+
+Sat Jul  7 15:30:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_yield_setup_args), vm.c, insns.def:
+	  fix to pass nil as block parameter to yielded block.
+	  [ruby-dev:31147]
+
+	* bootstraptest/test_block.rb: add a test for above.
+
+Fri Jul  6 19:55:10 2007  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb.rb: typo. Thanks, Giles Bowkett.
+
+	* lib/irb/completion.rb: support Ruby1.9 changing return value
+	  String to Symbol for Object#methods, etc. [ruby-dev:31148].
+
+Fri Jul  6 18:20:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb: fix load path.
+
+	* common.mk: fix "test" rule to run with "btest".
+
+	* rubytest.rb, sample/test.rb: fix to show tests progress.
+
+Fri Jul  6 15:37:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_iterator.rb: fix test to 1.9 spec.
+
+Fri Jul  6 15:21:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_interrupt): suppress a gcc's officious warning.
+
+Fri Jul  6 14:57:12 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (keyword_to_name): constified.
+
+	* ext/ripper/eventids2.c (token_to_eventid): ditto.
+
+Fri Jul  6 14:50:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/runner.rb: added --quiet option.
+
+Fri Jul  6 14:35:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_marg): wrap f_norm_arg assignment node with NODE_LIST.
+	  [ruby-dev:31141]
+
+Fri Jul  6 12:15:01 2007  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/sentgen.rb: new file.
+
+	* test/ruby/test_assignment.rb: tests implemented using assignment
+	  generator and emulator.
+
+Fri Jul  6 03:06:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: remove unused code.
+
+	* compile.c (compile_massign): fix to invoke to_splat on
+	  splat rhs (example: *a = *nil). [ruby-dev:31136]
+
+	* bootstraptest/test_massign.rb: add tests for above.
+
+	* compile.c (iseq_compile_each): disable excess optimization.
+	  [ruby-dev:31126]
+
+Fri Jul  6 02:08:25 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: fix to invoke nil.to_splat on NODE_ARGSCAT.
+	  [ruby-dev:31138].
+
+	* bootstraptest/test_literal.rb: add tests for above.
+
+Thu Jul  5 19:45:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h: rename rb_control_frame_t#magic to flag.
+
+	* vm.h: add VM_FRAME_TYPE() and VM_FRAME_FLAG().
+
+	* cont.c, insnhelper.ci, insns.def, vm.c, vm_dump.c,
+	  vm_evalbody.ci, yarvcore.c: apply above changes.
+
+Thu Jul  5 19:16:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_basicinstructions.rb: remove an assertion using
+	  unsupported hash literal (such as {1, 2}).
+
+	* test/ruby/test_hash.rb: ditto.
+
+Thu Jul  5 19:12:22 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c: Qfalse is VALUE, not pointer.
+
+Thu Jul  5 18:42:01 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): add break catch point.
+
+	* insns.def (throw): support correct "break" and "return".
+	  this commit achieve that "make test" passes all tests.
+
+	* vm.c: ditto.
+
+Thu Jul  5 18:44:12 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (mlhs_basic): use mlhs_post after tSTAR.
+	  [ruby-dev:31109]
+
+Thu Jul  5 18:27:58 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/{intern,ruby}.h, compile.[ch], error.c, eval.c,
+	  eval_load.c, gc.c, iseq.c, main.c, parse.y, re.c, ruby.c,
+	  yarvcore.[ch] (ruby_eval_tree, ruby_sourcefile, ruby_sourceline,
+	  ruby_nerrs): purge global variables.
+
+	* ruby.c (proc_options): moved do_print and do_loop options
+	  handling from ruby_process_options().
+
+Thu Jul  5 16:37:34 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (int_pow): fix previous nubu's commit.
+
+	* test/ruby/test_fixnum.rb: new test.
+
+Thu Jul  5 15:56:06 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (int_pow): even number multiplication never be negative.
+
+Thu Jul  5 10:42:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/{node,ruby}.h, ruby.c: added enum constants for gdb
+	  support.  [ruby-dev:31066]
+
+	* .gdbinit: some improvements.
+
+Thu Jul  5 10:13:08 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (global_symbols.last_id): reduce unused ID numbers.
+
+	* include/ruby/st.h, st.c (st_init_table, st_init_table_with_size):
+	  constified.
+
+Wed Jul  4 23:36:27 2007  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/webrick/httpauth/authenticator.rb
+	  (WEBrick::HTTPAuth::Authenticator#check_scheme): auth-scheme must be
+	  treated as a case-insensitive token according to RFC 2617 section 1.2.
+
+Wed Jul  4 18:30:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (mlhs_inner): new rule.  [ruby-dev:31132]
+
+Wed Jul  4 05:11:57 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (set_relation): added.
+
+Wed Jul  4 04:58:30 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (caller_setup_args): fix to show correct class
+	  on an error message (ex: m(&1)). [ruby-dev:31101]
+
+Wed Jul  4 04:30:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_array, iseq_compile_each): fix about array
+	  generation in void context. [ruby-dev:31102]
+
+	* bootstraptest/test_literal.rb: add a test for above.
+
+Wed Jul  4 04:07:00 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_array): ignore NODE_ZARRAY.
+	  [ruby-dev:31110]
+
+	* bootstraptest/test_method.rb: add a test for above.
+
+Wed Jul  4 04:04:02 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.h: fix debug print level.
+
+Wed Jul  4 03:52:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): support v[&b]= type method call.
+	  [ruby-dev:31094]
+
+	* bootstraptest/test_method.rb: add a test for above.
+
+Wed Jul  4 03:43:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_massign): fix massign compilation
+	  (example: a, *v, (*x) = ...).  [ruby-dev:31107]
+
+	* bootstraptest/test_massign.rb: add tests for above.
+
+Tue Jul  3 23:12:29 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/onigiruma.h (ONIG_EXTERN): use RUBY_EXTERN if defined.
+
+	* regenc.h: include ruby/defines.h.
+
+	* regint.h: x64-mswin64 support.
+
+Tue Jul  3 13:47:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (cont_save_machine_stack): clear saved_thread.machine_stack*.
+
+Mon Jul  2 21:45:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: rename iseq_translate_direct_threaded_code()
+	  to iseq_translate_threaded_code().
+
+	* eval_intern.h, yarvcore.h: mv EXEC_EVENT_HOOK() and
+	  exec_event_hooks() to yarvcore.h.
+
+	* insnhelper.ci, vm.c: mv yarv_finish_insn_seq to vm.c.
+
+	* insns.def (opt_call_c_function): fix to use RESTORE_REGS().
+
+	* iseq.c (rb_iseq_build_for_ruby2cext): fix to allocate iseq.
+
+Mon Jul  2 11:59:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defineclass): suppress a warning.
+
+	* insns.def (opt_call_c_function): should raise the thrown exception
+	  instead of returning it.
+
+Mon Jul  2 08:53:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h, yarvcore.h: move declaration of sysstack_error
+	  to yarvcore.h.
+
+	* iseq.c: fix symbol name (:toplevel -> :top).
+
+	* lib/vm/instruction.rb, template/vm.inc.tmpl: replaceable
+	  current file name.
+
+Mon Jul  2 05:29:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, iseq.c: fix iseq some of load/store process.
+
+Mon Jul  2 03:09:36 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, compile.c, insnhelper.ci, iseq.c, vm.c:
+	  rename structure names and field names.
+
+	* insnhelper.h, insns.def: add GET_CONST_INLINE_CACHE().
+
+	* iseq.c: add rb_iseq_build_for_ruby2cext().
+
+	* yarvcore.h, vm.h: move declaration of rb_insn_func_t
+	  to yarvcore.h.
+
+Sun Jul  1 03:25:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.h, vm.h: some refactoring.
+	  remove useless comments, etc.
+
+Sun Jul  1 03:02:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h: some refactoring on rb_iseq_t.
+	  rename some variable names, add comments, etc.
+
+	* compile.c, iseq.c, proc.c, vm.c: ditto.
+
+Sun Jul  1 02:57:57 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.h: rename insn_func_type to rb_insn_func_type.
+
+	* vm_evalbody.ci: ditto.
+
+	* insns.def: add opt_call_native_compiled instruction
+	  instead of opt_call_native_compiled.
+
+Sat Jun 30 00:17:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): return non-valid token for an invalid
+	  instance/class variable name.  a patch from Yusuke ENDOH
+	  <mame AT tsg.ne.jp>.  [ruby-dev:31095]
+
+Fri Jun 29 23:38:06 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmts): fix for ripper.
+
+Fri Jun 29 21:55:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: fix to show line number of blank block.
+	  [ruby-dev:31093]
+
+Fri Jun 29 20:51:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/cgi/session.rb (create_new_id): don't cut off md5.hexdigest to
+	  follow Ruby 1.8.
+
+Fri Jun 29 17:10:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* debug.h: constified.
+
+	* debug.c (ruby_set_debug_option): separated from main.c.
+
+	* gc.c (ruby_gc_stress), signal.c (ruby_enable_coredump): prefixed.
+
+Fri Jun 29 16:39:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (proc_new): fix to return a proc object
+	  which block is contained ([ruby-dev:31056]).
+
+Fri Jun 29 15:43:59 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_pow): get rid of division by zero.  reported by
+	  Yusuke ENDOH <mame AT tsg.ne.jp> [ruby-dev:31040]
+
+	* numeric.c (int_round): do nothing when rounding by zeroth digit.
+	  check underflow.  [ruby-dev:31043]
+
+Fri Jun 29 15:32:00 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* configure.in: add fastcall attribute check.
+
+Fri Jun 29 14:51:24 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (assoc_list): remove expanded hash literal (no splat).
+
+	* lib/webrick/httpstatus.rb (WEBrick::HTTPStatus::EOFError): adapt
+	  to new syntax.
+
+Fri Jun 29 14:48:18 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/insns2vm.rb, lib/vm/instruction.rb: move process body
+	  to lib/vm/instruction.rb.
+
+	* common.mk: fix aotc rule.
+	  experimental. bin/ruby2cext is not added yet.
+
+Fri Jun 29 11:23:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (dsym): return non-null NODE even if yyerror().  based on a
+	  patch from Yusuke ENDOH <mame AT tsg.ne.jp>.  [ruby-dev:31085]
+
+Thu Jun 28 23:29:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (assoc_list): odd number check only for NODE_ARRAY.
+	  [ruby-dev:31082]
+
+Thu Jun 28 22:24:33 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* win32/Makefile.sub: define FUNC_FASTCALL macro.
+
+	* vm.h: fix to use FUNC_FASTCALL macro.
+	  TODO: add FUNC_FASTCALL macro by configure.
+
+Thu Jun 28 19:38:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: fix to remove -Wall warnings on gcc.
+
+	* compile.c (make_name_with_str): removed. use rb_sprintf() instead.
+
+Thu Jun 28 18:53:01 2007  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_hash): fix hash area.
+
+Thu Jun 28 15:00:06 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_getline): local variable to be
+	  initialized.  [ruby-dev:31077]
+
+Thu Jun 28 11:30:39 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (rb_obj_id): use SIGNED_VALUE instead of long.
+
+Thu Jun 28 05:01:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk (run.gdb): fix to load $(srcdir)/.gdbinit
+
+	* vm.c (rb_vm_set_finish_env): add a cast.
+
+	* vm.h: support __fastcall for MSVC.
+
+Thu Jun 28 02:12:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb: fix to untouch $:.
+
+Thu Jun 28 02:03:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (setup_args): change parameter type.
+
+Thu Jun 28 02:03:39 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (rb_intern2): unconstify cast.
+
+Thu Jun 28 01:44:31 2007  Tanaka Akira  <akr@fsij.org>
+
+	* parse.y (rb_intern2): don't allocate a string object at first.
+	  [ruby-dev:31064]
+
+Thu Jun 28 01:24:02 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb: fix to show file name.
+
+	* bootstraptest/test_*.rb: add bootstrap tests.
+
+Thu Jun 28 01:22:15 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/node.h, parse.y, gc.c, iseq.c: remove NODE_CREF.
+
+Thu Jun 28 01:19:43 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix popped backref and others.
+	  ([ruby-dev:31068]).
+
+	* compile.c (iseq_compile_each): remove needless statements.
+
+Wed Jun 27 23:51:33 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c: remove unused functions.
+
+Wed Jun 27 20:46:05 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h, win32/Makefile.sub, win32/configure.bat,
+	  win32/mkexports.rb, win32/setup.mak, win32/win32.c: import
+	  x64-mswin64 port.
+
+Wed Jun 27 20:31:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_translate_direct_threaded_code): fix prototype
+	  function name.
+
+	* vm.h: add correct cast.
+
+Wed Jun 27 17:08:42 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_evalbody.ci: support OPT_CALL_THREADED_CODE.
+
+	* insns.def, vm.c, vm.h: ditto.
+
+	* vm.h: add VM_CFP_CNT() and VM_SP_CNT().
+
+Wed Jun 27 04:23:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix type error.
+
+Wed Jun 27 03:26:15 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_massign), insns.def (expandarray): support
+	  postarg with massign (a, *b, c = ...).
+
+	* bootstraptest/test_massign.rb: add tests for above.
+
+	* compile.h: fix debug macro names.
+
+Wed Jun 27 00:18:41 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_clear): need to check STR_EMBED_P() before
+	  free()ing memory.  a patch from Yusuke ENDOH <mame AT tsg.ne.jp>.
+	  [ruby-dev:31062]
+
+Tue Jun 26 16:39:01 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_getgroups): use GIDT2NUM for rb_gid_t.
+
+Tue Jun 26 16:28:24 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_wait_fd_rw): terminate fdset.
+
+Tue Jun 26 16:26:58 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* regint.h: IL32LLP64 support.
+
+Tue Jun 26 16:22:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (ruby_node_name): update node names.
+
+Tue Jun 26 15:21:20 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h: IL32LLP64 support.
+
+	* bignum.c (bigfixize, rb_cstr_to_inum): ditto.
+
+	* insns.def (opt_plus, opt_minus, opt_mult): ditto.
+
+Tue Jun 26 15:04:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_s_new): revert initializing VM stack.
+
+	* yarvcore.c (th_init2): ditto.
+
+	* vm.c, vm.h: fix to stop using Qundef on VM stack.  According to
+	  this change, VM stack should not include Qundef value.
+
+	* insns.def (putundef): removed.
+
+	* compile.c (iseq_compile_each): ditto.
+
+	* eval.c (eval): fix spacing.
+
+Tue Jun 26 04:03:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (vm_yield_with_cfunc), proc.c: fix Method#to_proc
+	  to return lambda Proc ([ruby-dev:31021], [ruby-dev:31037]).
+
+Tue Jun 26 03:46:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_s_new): fix to clear rb_thread_t#tag.
+	  [ruby-dev:30995]
+
+Tue Jun 26 03:38:31 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_s_new), yarvcore.c (th_init2): fix to clear
+	  VM stack ([ruby-dev:31046]).
+
+Tue Jun 26 03:15:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: rename setup_arg() to setup_args().
+	  fix to use setup_args() at processing NODE_YIELD.
+
+Tue Jun 26 02:50:24 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (setup_arg): support kind of "m(*ary, x)" method call.
+	  ([ruby-dev:31048]).
+
+Tue Jun 26 00:28:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci, vm.c: complete block parameter support.
+	  post arguments, optional arguments, block argument.
+
+	* compile.c, parse.y: fix {|a|} parameter.
+
+	* insnshelper.ci, insns.def: revert caller_setup_args() option
+	  (need_block_check) parameter.
+
+Mon Jun 25 20:18:44 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (RVALUE): in RVALUE and RBasic, flags must be the same type.
+
+Mon Jun 25 18:02:55 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/pty/extconf.rb: skip wince and win64.
+
+Mon Jun 25 17:59:32 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/node.h (NODE_LMASK, nd_line): shouldn't use int and/or
+	  long carelessly.
+
+Mon Jun 25 11:36:35 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.h: add RUBY_ prefix to debug macros.
+
+	* cont.c, proc.c, yarvcore.c,
+
+	* gc.c: define ruby_gc_debug_indent variable to debug mark/free.
+
+	* vm.c, insnhelper.ci: rename some functions to vm_* or rb_vm_*.
+	  move some functions, definitions, declarations to suitable files.
+
+	* eval.c, yarvcore.h, eval_error.ci, insnhelper.ci: ditto.
+
+Mon Jun 25 09:45:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.ci, eval_jump.ci, eval_method.ci, eval_safe.ci: c-mode.
+
+Mon Jun 25 05:27:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c: remove ruby_current_node and change eval() prototype.
+	  fix to use rb_sourcefile/line() instead of ruby_sourcefile/line.
+
+	* error.c, eval_error.ci, eval_load.c, eval_safe.ci, gc.c,
+	  include/ruby/intern.h, parse.y, process.c, ruby.c: ditto.
+
+	* vm.c: fix spaces.
+
+Mon Jun 25 04:20:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_*.h: rename to eval_*.ci.
+
+	* common.mk: ditto.
+
+	* eval_error.ci: remove ruby_set_current_source().
+
+	* error.c, eval.c, ruby.c: ditto.
+
+	* eval_safe.c, proc.c: remove unused macros.
+
+Mon Jun 25 03:37:20 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.ci (caller_setup_args): add need_block_check option.
+
+	* insns.def: ditto.
+
+	* yarvcore.h: add GetCoreDataFromValue().
+
+Mon Jun 25 02:14:30 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* call_cfunc.ci: removed.
+
+	* insnhelper.ci: added. this function includes all functions that
+	  vm insns need.
+
+	* common.mk: ditto.
+
+	* insnhelper.h, vm.h, vm.c: move some declaration.
+
+	* gc.h: remove GC_CHECK() macro because GC.stress is more useful.
+
+	* compile.c, iseq.c, vm_dump: ditto.
+
+	* gc.h, thread.c: move a prototype declaration.
+
+	* debug.c, debug.h: rename some functions.
+
+	* compile.h: ditto.
+
+Mon Jun 25 00:45:02 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (invokesuper): fix error message.
+
+Mon Jun 25 00:14:13 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: some refactoring.
+	  * rename th_* to vm_*.
+	  * remove unused variables functions.
+	  * add prototypes.
+
+	* blockinlining.c, compile.c, cont.c, eval.c, eval_intern.h,
+	  eval_jump.h, eval_load.c, inits.c, insns.def, iseq.c, parse.y,
+	  proc.c, process.c, signal.c, thread.c, vm.c, vm_dump.c,
+	  vm_evalbody.ci, yarvcore.c, yarvcore.h: ditto.
+
+Sun Jun 24 22:32:00 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.h (rb_add_method): fix to check 0.
+
+Sun Jun 24 22:00:17 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insn_send.ci: removed.
+
+	* common.mk: ditto.
+
+	* vm.c (vm_call_bmethod), isnsn.def: added.  fix to use this
+	  function instead of using goto.
+
+	* vm.c (vm_call_bmethod): renamed from th_invoke_bmethod().
+
+	* vm.c (vm_method_missing): renamed from eval_method_missing().
+
+	* vm_evalbody.ci: remove tmp_* variables.
+
+	* insnhelper.h: add some macros.
+
+	* insns.def: forbid zsuper from method defined by define_method().
+
+	* test/ruby/test_super.rb: ditto.
+
+Sun Jun 24 20:01:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_macro.def: removed.
+
+	* insn_send.ci: added.  this file includes send instruction body.
+
+	* common.mk: ditto.
+
+	* insns.def: ditto.
+
+	* tool/insns2vm.rb: ditto.
+
+	* vm.c: ditto.
+
+Sun Jun 24 19:30:37 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.h (RESTORE_REGS): add do/while(0) around macro.
+
+	* vm.c, vm_macro.def: remove macro_eval_invoke_func() and
+	  add vm_setup_method().  use it instead.
+
+Sun Jun 24 19:02:33 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, vm_macro.def : remove macro_eval_invoke_cfunc() and
+	  add vm_call_cfunc().
+
+Sun Jun 24 17:54:13 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm.c: add/fix stack overflow check.
+
+Sun Jun 24 17:28:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.h: change CHECK_STACK_OVERFLOW() to throw exception.
+
+	* vm.c (caller_setup_arg), vm_macro.def: remove
+	  macro_eval_setup_send_arguments and add caller_setup_arg().
+
+	* insns.def: ditto.
+
+	* bootstraptest/test_method.rb: add splat arg tests.
+
+Sun Jun 24 16:35:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_to_s): used a variable before initialized.
+
+Sun Jun 24 16:05:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (callee_setup_arg): added.  support correct post arg.
+
+	* vm_macro.def (macro_eval_invoke_func): fix to use
+	  callee_setup_arg.
+
+	* compile.c (set_arguments): adjust for above changes.
+
+	* compile.c (iseq_compile_each): ditto.
+
+	* iseq.c (ruby_iseq_disasm): ditto.
+
+	* yarvcore.h: add rb_iseq_t#post_arg_start and arg_size.
+
+	* bootstraptest/test_method.rb: add post arg tests.
+
+Sun Jun 24 16:10:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_to_s): suppress warning, and reduced duplicated code.
+
+Sun Jun 24 15:33:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb: set default directory to
+	  '/tmp/bootstraptest.tmpwd' and add --dir option.
+	  fix to output driver and target information.
+
+	* common.mk: fix to run btest on BASERUBY and
+	  add OPTS to pass option ("make btest OPTS=...").
+
+Sun Jun 24 03:05:00 2007  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_minmax): fix SEGV by [].minmax.
+
+Sat Jun 23 17:18:19 2007  Tanaka Akira  <akr@fsij.org>
+
+	* re.c (match_inspect): MatchData#inspect implemented.
+
+Sat Jun 23 15:00:16 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_minmax): new method to get the minimum and maximum
+	  values from the enumerable at once.
+
+	* enum.c (enum_minmax_by): ditto.
+
+Sat Jun 23 01:25:40 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_assoc): new method.
+
+	* hash.c (rb_hash_rassoc): ditto.
+
+	* hash.c (rb_hash_flatten): ditto.
+
+Fri Jun 22 23:55:59 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_upto): add optional argument to specify
+	  exclusiveness.
+
+	* range.c (range_step): use String#upto with optional argument.
+
+	* range.c (range_each): ditto.
+
+Fri Jun 22 19:55:51 2007  Tanaka Akira  <akr@fsij.org>
+
+	* proc.c (proc_to_s): revert the change from %p to %lx at YARV
+	  merge time.
+
+Fri Jun 22 19:33:49 2007  Tanaka Akira  <akr@fsij.org>
+
+	* proc.c (proc_to_s): show is_lambda.
+
+Thu Jun 21 20:36:12 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_sort): remove hash specific implementation.
+
+Thu Jun 21 20:28:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_select): returns new hash, not assoc array.
+	  [ruby-core:11504]
+
+	* hash.c (env_select): ditto.
+
+Thu Jun 21 23:08:19 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (rb_intern2): ID_JUNK test based on len, not by NUL.
+
+Thu Jun 21 19:42:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* yarvcore.c (rb_thread_mark): mark also thrown_errinfo.
+
+Thu Jun 21 17:13:44 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_intern2): name may not be NUL-terminated.
+
+Wed Jun 20 08:27:57 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval_error.h (error_print): show full stacktrace on
+	  non-SystemStackError.
+
+Wed Jun 20 04:45:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_intern2): use rb_intern2 to intern without trailing
+	  equal sign.
+
+	* parse.y (rb_intern2, ripper_id2sym): fixed indent.
+
+Tue Jun 19 10:55:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_load.c (load_ext, rb_require_safe): pass VALUE instead of
+	  pointer.  [ruby-Bugs-11659]
+
+Mon Jun 18 08:47:54 2007  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* ext/openssl/{extconf.rb,ossl_ssl_session.c}:
+	  Fix ruby-Bugs-11513.
+
+	* ext/openssl/ossl_pkey_ec.c
+	  New methods EC::Point.[eql,make_affine!,invert!,on_curve?,infinity?]
+	  By default output the same key form as the openssl command.
+
+	* ext/openssl/ossl_rand.c
+	  New method Random.status?
+
+	* test/openssl/test_ec.rb
+	  New tests.
+
+Mon Jun 18 17:04:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_load.c (rb_require_safe, ruby_init_ext): load with ruby level
+	  cfp.  [ruby-core:10779]
+
+	* eval_intern.h, vm.c (rb_vm_call_cfunc): new function to call a
+	  function with ruby level cfp.
+
+Mon Jun 18 16:57:24 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (yycompile): disable trace while creating ruby_debug_lines.
+	  [ruby-talk:253586]
+
+	* thread.c (ruby_suppress_tracing): new function to call a function
+	  with suppressing trace.
+
+	* lib/debug.rb, lib/tracer.rb: for YARV.
+
+Mon Jun 18 13:54:36 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): return EXIT_FAILURE if any exceptions occurred
+	  in at_exit blocks.  [ruby-core:11263]
+
+Mon Jun 18 02:49:16 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (env_mark): fix to mark block.proc.
+
+	* vm.c (th_make_proc_from_block): set created proc to block->proc.
+
+Mon Jun 18 02:48:12 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c (vm_stack_dump_raw): hide VM stack trace.
+
+Mon Jun 18 02:43:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* signal.c (sigsegv): clear gc_stress flag on SEGV.
+
+Mon Jun 18 01:14:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_path2class): get rid of dangling pointer caused by
+	  optimized out value.
+
+	* variable.c (rb_global_entry, rb_f_untrace_var, rb_alias_variable,
+	  rb_generic_ivar_table, generic_ivar_get, generic_ivar_set,
+	  generic_ivar_defined, generic_ivar_remove, rb_mark_generic_ivar,
+	  rb_free_generic_ivar, rb_copy_generic_ivar,
+	  rb_obj_instance_variables): suppress warnings.
+
+Sun Jun 17 11:11:07 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (rb_method_missing): avoid a warning "too many arguments
+	  for format string" on "./ruby -ve 'def m() super end; m'".
+
+Sat Jun 16 22:24:17 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (garbage_collect): re-introduce ruby_current_node marking code.
+	  [ruby-dev:31005]
+
+Sat Jun 16 21:37:43 2007  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (gc_sweep): re-introduce heap extension strategy change.
+	  [ruby-dev:31005]
+
+Fri Jun 15 22:59:37 2007  Tanaka Akira  <akr@fsij.org>
+
+	* .gdbinit: new file to ease debugging using gdb.
+
+Fri Jun 15 22:33:55 2007  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (default_handler): func argument removed.
+	  (trap_handler): support SYSTEM_DEFAULT.  call default_handler
+	  internally.
+	  (sig_trap): don't call default_handler.
+	  [ruby-dev:30999]
+
+Fri Jun 15 22:33:29 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (realclean): separate local and ext.
+
+	* ext/extmk.rb: not remove unrelated directories.
+
+Fri Jun 15 20:50:02 2007  Tanaka Akira  <akr@fsij.org>
+
+	* keywords: enclose C code in declaration section by %{ and %} to
+	  avoid extra semicolon after #ifdef RIPPER.
+	  pointed by eban.
+
+Fri Jun 15 18:56:52 2007  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (trap_handler): trap("SIGSEGV", "DEFAULT") may cause wrong
+	  trap error because SIG_DFL may be zero.
+
+Fri Jun 15 15:55:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (setconstant, toregexp): fix to mark object correctly.
+
+Fri Jun 15 13:24:18 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c: exchange semantics of Hash#each and Hash#each_pair.
+	  pointed out by [ruby-dev:30997].
+
+	* test/ruby/test_iterator.rb: ditto.
+
+	* test/ruby/test_yield.rb: ditto.
+
+Fri Jun 15 12:38:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_iterator.rb: remove debug code (GC.stress=true).
+
+Fri Jun 15 12:25:33 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (th_yield_setup_args): |v| should work as |v,|.
+	  ex) def m;yield 1, 2; end; m{|v| p v} #=> 1
+
+	* parse.y: apply above change for "for" statement.
+
+	* test/ruby/test_assignment.rb: ditto
+
+	* test/ruby/test_basicinstructions.rb: ditto.
+
+	* test/ruby/test_iterator.rb: ditto.
+
+	* test/ruby/test_yield.rb: ditto.
+
+	* compile.c (iseq_compile_each): fix debug.
+
+Fri Jun 15 12:22:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (ruby_finalize_1): rb_thread_t#errinfo should be clear with
+	  Qnil.
+
+Fri Jun 15 12:20:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_cont_call): forbid cross fiber continuation call.
+
+	* test/ruby/test_fiber.rb: ditto.
+
+Fri Jun 15 12:14:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* sample/test.rb: fix to show line information whether test succeeds.
+
+Thu Jun 14 17:16:05 2007  Tanaka Akira  <akr@fsij.org>
+
+	* eval_load.c (Init_load): delay allocating an array for rb_load_path
+	  to avoid GC problem in very early stage.
+	  (RUBY_GC_STRESS causes GC in such stage.)
+
+	* variable.c (rb_gc_mark_global_tbl): rb_global_tbl may be 0 in
+	  very early stage.
+
+	* thread.c (thread_cleanup_func) [IA64]: clear register stack position.
+	  (thread_start_func_2) [IA64]: record the beginning of register
+	  stack using extra argument.
+	  (rb_gc_save_machine_context) [IA64]: record the end of register
+	  stack.
+
+	* gc.c [IA64] (SET_STACK_END): record the end of register stack.
+	  (garbage_collect) [IA64]: use recorded register stack area for
+	  GC marking.
+	  (yarv_machine_stack_mark) [IA64]: GC mark from the register stack
+	  area.
+
+	* yarvcore.c [IA64] (rb_gc_register_stack_start): defined.
+	  (Init_VM): store th->self on stack to fix GC problem.
+	  (Init_yarv) [IA64]: initialize the beginning of register stack.
+
+	* yarvcore.h (struct rb_thread_struct) [IA64]: new members for
+	  register stack area.
+
+	* thread_pthread.ci (thread_start_func_1) [IA64]: call
+	  thread_start_func_2 with the end of register stack.
+
+	* cont.c (struct rb_context_struct) [IA64]: new members for register
+	  stack area.
+	  (cont_mark) [IA64]: GC mark from register stack area.
+	  (cont_free) [IA64]: free saved register stack.
+	  (cont_save_machine_stack) [IA64]: record the position and contents
+	  of the register stack.
+	  (cont_capture): store cont->self on stack to fix GC problem.
+	  (cont_restore_1) [IA64]: restore the register stack.
+	  [IA64] (register_stack_extend): new function.
+	  (cont_restore_0) [IA64]: call register_stack_extend instead of
+	  cont_restore_1.
+
+	[ruby-dev:30982]
+
+Thu Jun 14 17:09:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser): handle more
+	  extensions.  [ruby-dev:30972]
+
+Thu Jun 14 14:40:42 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb: document updated.
+	  suggested by NaHi.  [ruby-dev:30966]
+
+Wed Jun 13 22:42:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (garbage_collect): update IA64 register stack code.
+	  [ruby-dev:30971]
+
+Wed Jun 13 06:05:12 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (darwin): prohibit loading extension libraries to
+	  miniruby.
+
+Tue Jun 12 21:50:51 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (call_args): no allow splat after assocs.  takes
+	  consistency over compatibility.
+
+	* parse.y (call_args2): ditto
+
+Tue Jun 12 14:53:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (Logging.quiet, Logging.message): added quiet flag and
+	  use it.  [ruby-core:10909]
+
+	* lib/mkmf.rb (find_header): use header names in the message.
+
+Sun Jun 10 18:37:13 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/probeprofiler/probeprofiler.c: clean warnings.
+
+Sun Jun 10 18:32:24 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/isinf.c, missing/dup2.c, missing/strtod.c, missing/x68.c,
+	  missing/alloca.c: use "ruby/config.h".
+
+Sun Jun 10 17:49:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (install_recursive): should check parent directories of
+	  the destination.  [ruby-dev:30947]
+
+Sun Jun 10 16:59:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (do_block, brace_block): fix line numbers.  [ruby-dev:30831]
+
+Sun Jun 10 16:57:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (install_recursive): add :glob option rather than
+	  using FNM_DOTMACH.
+
+	* instruby.rb (ext-comm): make header directory first.
+
+Sun Jun 10 16:10:58 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb: typo.
+
+Sun Jun 10 16:07:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (install_recursive): skip .svn directories.
+
+Sun Jun 10 15:44:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): rubyhdrdir was missing.
+
+Sun Jun 10 15:26:36 2007  Tanaka Akira  <akr@fsij.org>
+
+	* Makefile.in: use --output-file for gperf to not leave lex.c.tmp.
+
+Sun Jun 10 15:11:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub (XCFLAGS): -I. is needed for *.inc.
+
+	* ext/extmk.rb: prepend also topdir to mflags at last.
+
+Sun Jun 10 13:47:36 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb (test_should_propagate_signaled):
+	  get rid of invoking shell.  [ruby-dev:30942]
+
+Sun Jun 10 12:56:46 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby: moved public headers.
+
+	* instruby.rb (install_recursive): skip backup files.
+
+	* instruby.rb (ext-comm): install only current platform headers.
+
+Sun Jun 10 10:42:04 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb: renamed from lib/secrand.rb.
+	  suggested by NaHi.  [ruby-dev:30934]
+
+Sat Jun  9 06:40:05 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/secrand.rb: rename SecRand() to SecRand.random_number.
+	  suggested by NaHi.  [ruby-dev:30934]
+
+Fri Jun  8 16:34:20 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ext/zlib/zlib.c (gzfile_s_open): use FilePathValue to support
+	  to_path.
+
+Fri Jun  8 16:11:00 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_jump.h: th->errinfo should clear with nil.
+
+Fri Jun  8 14:53:28 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (call_args): allow splat argument after unpacked
+	  assocs like 1.8 does.
+
+	* parse.y (call_args): ditto.
+
+Fri Jun  8 14:26:18 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/secrand.rb: new file for secure random interface.
+
+	* lib/cgi/session.rb: use secrand for generating cookies.
+
+Fri Jun  8 12:44:37 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {win32,wince}/Makefile.sub: add lex.c rule.
+
+Fri Jun  8 11:54:18 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lex.c.blt: moved from lex.c.
+
+	* Makefile.in: use lex.c.blt if gperf is not available.
+	  [ruby-list:8212], [ruby-list:8214], [ruby-list:24667],
+	  [ruby-talk:120857], [ruby-dev:28102]
+
+Thu Jun  7 21:38:39 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_execute_interrupts): invoke ensure when
+	  main thread exits.
+
+Thu Jun  7 19:02:48 2007  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb: call original "method" method instead of redefined one.
+
+Thu Jun  7 17:20:57 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (prepare_iseq_build): freeze filename and name string.
+
+	* variable.c: freeze class name string.
+
+Thu Jun  7 12:48:33 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (cont_restore_1): fix to check root fiber [ruby-dev:30911].
+
+	* test/ruby/test_fiber.rb: add a test.
+
+Thu Jun  7 07:24:36 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/json/common.rb: Ponder offering parse! method.
+
+	* lib/json/editor.rb: be a bit more robust while loading data.
+
+	* ext/json/ext/{generator,parser}/extconf.rb:
+	  add a have_header directive for st.h
+
+	* test/json: fix some tests.
+
+Thu Jun  7 03:29:18 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test_fiber.rb: add a test (Continuation and Fiber).
+
+Thu Jun  7 03:17:24 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (cont_new): add debug message.
+
+	* cont.c (cont_restore_1): copy stack information from fiber.
+
+	* cont.c (rb_fiber_s_new): fix to mark created fiber.
+
+	* test/ruby/test_fiber.rb: add some tests around Thread and Fiber.
+
+	* yarvcore.c (thread_free): fix to skip freeing stack if root fiber
+	  is available.
+
+Thu Jun  7 01:03:20 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h, eval.c (ruby_init): remove POP_TAG_INIT().
+
+	* cont.c (rb_fiber_start): remove zero-clearing tag.
+
+Wed Jun  6 20:23:46 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (invokeblock): fix of splat argument.
+	  (splat same as normal method dispatch)
+
+Wed Jun  6 16:27:25 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* insns.def: fixed indentation.
+
+Wed Jun  6 10:58:23 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_yield): fix to check Qundef.
+
+Wed Jun  6 10:57:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_continuation.rb: add a test for last commit.
+
+Wed Jun  6 10:55:42 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_cont_call): forbid calling dead fiber with
+	  Continuation#call.
+
+Wed Jun  6 10:50:01 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix around yield arguments
+	  (with NODE_ARGSCAT).
+
+Wed Jun  6 02:50:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_start): clear th->tag and check error to fix
+	  [ruby-dev:30888] and [ruby-dev:30889].
+
+	* eval_intern.h: fix rb_fiber_start() prototype.
+
+	* test/ruby/test_fiber.rb: add tests for above.
+
+Wed Jun  6 02:40:20 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* insnhelper.h, insns.def (DEC_SP): shouldn't use unary minus operator
+	  in pointer operation. some compilers (such as VC++8 x64) cannot deal
+	  it with expected way.
+
+Wed Jun  6 02:19:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (new_yield), compile.c (iseq_compile_each): fix
+	  passing parameter.
+
+	* eval.c, eval_jump.h: simplify rb_yield*.
+
+	* proc.c (proc_mark): fix to mark proc->block.proc.
+
+	* proc.c (Init_Proc): add Proc#lambda?
+
+	* test/ruby/test_lambda.rb: add some tests.
+
+	* vm.c (invoke_block): fix to check lambda block or not.
+
+	* vm.c (th_yield_setup_args): fix to check arguments size
+	  when lambda block.
+
+Tue Jun  5 16:30:38 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_p): returns arguments to intervene.  [ruby-dev:29736]
+
+Tue Jun  5 14:07:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (invokeblock): check block is created by lambda
+	  or Proc.new.
+
+	* vm.c (block_proc_is_lambda): added.
+
+Tue Jun  5 14:47:52 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/pp.rb (PP::PPMethods::seplist): revert last change to work
+	  around wrapper bug.  [ruby-dev:30840]
+
+Tue Jun  5 14:11:15 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (kanji_convert): Fix guess fallback.
+
+Tue Jun  5 13:32:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c, dir.c, eval.c, eval_jump.h, eval_method.h, numeric.c,
+	  pack.c, parse.y, re.c, thread.c, vm.c, vm_dump.c, call_cfunc.ci,
+	  thread_pthread.ci, thread_win32.ci: fixed indentation.
+
+	* call_cfunc.ci: protoized.
+
+	* thread_win32.ci: fixed typo.
+
+Tue Jun  5 13:17:11 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* call_cfunc.ci, compile.c, dir.c, eval.c, eval_jump.h, numeric.c,
+	  pack.c, re.c, thread.c, thread_win32.ci, vm.c, vm_dump.c: fixed
+	  indentation.
+
+Mon Jun  4 21:15:45 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/json.rb, lib/json, ext/json, test/json:
+	  import JSON library.
+
+	* ext/nkf: import nkf.c rev:1.124
+	  Support CP10001.
+
+Mon Jun  4 20:52:58 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_round): should not just truncate.
+
+Sat Jun  2 16:48:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (Fiber#pass): rename to Fiber#yield.  Block parameter
+	  of fiber body receive first yield values.
+	  e.g.: Fiber.new{|x| p x}.yield(:ok) #=> :ok
+
+	* cont.c: rename rb_context_t#retval to rb_context_t#value.
+
+	* test/ruby/test_fiber.rb: ditto.
+
+Sat Jun  2 16:45:21 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (Init_Proc): remove a line break.
+
+Sat Jun  2 01:27:27 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_round): small optimization to handle bignums.
+
+Fri Jun  1 13:02:35 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* insnhelper.h (INC_SP): shouldn't cast ``x'' to unsigned type because
+	  it might be a negative value.
+
+	* insnhelper.h, insns.def: shouldn't use unary minus operator in index
+	  operator. some compilers (such as VC++8 x64) cannot deal it with
+	  expected way.
+
+Fri Jun  1 11:33:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_round): should convert self to Float.
+	  [ruby-dev:30860]
+
+Fri Jun  1 02:01:13 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_round): now takes optional argument to specify
+	  number of digits, like round() in Python/PHP.
+
+	* numeric.c (num_round): ditto.
+
+Fri Jun  1 01:58:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (each_with_index_i): should work well with continuation.
+	  a patch from sheepman <sheepman AT sheepman.sakura.ne.jp>.
+	  [ruby-dev:30846]
+
+Thu May 31 17:27:53 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/benchmark.rb (Benchmark::Job::item): avoid modifying the
+	  argument unintentionally.  [ruby-talk:253676]
+
+Wed May 30 14:43:00 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (cont_capture): store all local variables in heap
+	  ([ruby-dev:30832]).
+
+	* vm.c (th_stack_to_heap): added.
+
+	* test/ruby/test_continuation.rb: add a test for above.
+
+	* eval_intern.h (th_get_ruby_level_cfp): fix to clean code.
+
+Wed May 30 13:32:34 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (ResponseParser#next_token): fixed
+	  error message.
+
+	* lib/net/imap.rb (ResponseParser#parse_error): fixed
+	  the condition not to refer to @token.symbol unexpectedly.
+	  Thanks, Dick Monahan.
+
+Wed May 30 13:24:33 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::FTP#transfercmd): skip 2XX
+	  responses for some FTP servers.
+
+Wed May 30 04:18:37 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval_cmd): just return if no exceptions.
+	  [ruby-dev:30820]
+
+Wed May 30 02:14:25 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (interrupt_init): needs to pass nil for Interrupt.
+	  [ruby-core:11038]
+
+	* signal.c (trap): fixed segfaults.  [ruby-dev:30830]
+
+Wed May 30 00:50:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_source_filename, obj_free): suppress warnings.
+
+	* gc.c (garbage_collect, yarv_machine_stack_mark): fixed typo.
+	  http://bugs.debian.org/426267
+
+Wed May 30 00:24:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (open_args, arg_ambiguous, parser_warning): should not use
+	  rb_warning in the parser.
+
+Tue May 29 12:31:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_opendir): removed duplicated code.
+
+Tue May 29 10:55:24 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: fix bug around Continuation and Fiber.
+
+	* test/ruby/test_continuation.rb: add tests for Continuation.
+
+Tue May 29 10:54:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_opendir, rb_w32_readdir): eliminate magic
+	  numbers.
+
+Mon May 28 10:27:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c: fixed a function name.
+
+Mon May 28 03:56:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: support Fiber.  Check test/ruby/test_fiber.rb for detail.
+	  Fiber is known as "Micro Thread", "Coroutine", and other terms.
+	  At this time, only Fiber#pass is supported to change context.
+	  I want to know more suitable method name/API for Fiber (... do you
+	  know more suitable class name instead of Fiber?) as "suspend/resume",
+	  "call", "yield", "start/kick/stop/restart", ....
+
+	* eval.c, eval_intern.h, thread.c, yarvcore.c, yarvcore.h: ditto.
+
+Sat May 26 00:38:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_exec_internal): do nothing if no code.
+
+	* compile.c (rb_iseq_compile): check node if NULL before check
+	  nd_type.  [ruby-talk:252956]
+
+Sat May 26 00:05:22 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb (test_should_propagate_signaled):
+	  skip tests for exitstatus and termsig on the platforms where
+	  signals not supported.
+
+Fri May 25 16:04:47 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* yarvcore.c (Init_VM): wrap already initialized structs to use
+	  it directly.
+
+Fri May 25 11:09:47 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* regint.h (include): on some platform, defines.h redefines
+	  SIZE_OF_LONG_LONG so shouldn't re-include config.h after included
+	  defines.h.
+
+	* regint.h (vsnprintf): ruby on windows already have vsnprintf macro.
+
+Thu May 24 12:07:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: check across trap violation.
+
+	* eval.c, yarvcore.h: ditto.
+
+Thu May 24 11:46:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, yarvcore.c: fix to mark VM structure on startup.
+
+	* yarvcore.h: disable USE_CACHED_VALUE.
+
+Thu May 24 01:54:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c: support callcc which everyone love.
+	  incomplete. please give me bug reports.
+
+	* common.mk, inits.c, thread.c: ditto.
+
+	* yarvcore.c: export thread_mark().
+
+	* yarvcore.h: disable value cache option.
+
+	* eval_intern.h: set th_get_ruby_level_cfp to inline.
+
+Wed May 23 15:39:02 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* common.mk: add a rule for regsyntax.c.
+
+Wed May 23 10:31:53 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* oniguruma.h: updated to Oniguruma 5.7.0.
+
+	* regsyntax.c, unicode.c: new files along with Oniguruma 5.x.
+
+Wed May 23 06:51:46 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* lib/cgi.rb (CGI#[]): get rid of exceptions being raised.
+	  [ruby-dev:30740], Thanks Kentaro KAWAMOTO.
+
+Wed May 23 05:49:49 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb, ext/purelib.rb, lib/mkmf.rb, runruby.rb: clear default
+	  load path to get rid of load pre-installed extensions/libraries.
+	  [ruby-core:11017]
+
+Tue May 22 16:37:36 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (set_arg0): support RSTRING_LEN on HP-UX.  a patch from
+	  WATANABE Tetsuya <Tetsuya.WATANABE AT nifty.com>.  [ruby-dev:30806]
+
+Mon May 21 13:40:00 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, vm_macro.def: support tail call optimization
+	  (on default, this feature is not enabled).
+
+	* iseq.c, compile.c, vm_opts.h: add "tailcall_optimization"
+	  option.
+
+	* sample/test.rb (test_ok): fix to adjust tailcall stack layout.
+
+	* insns.def, vm.c, compile.c, yarvcore.c, yarvcore.h:
+	  add opt_gt, opt_le instructions.
+
+Mon May 21 03:34:06 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb: CRAM-MD5 authentication did not work.
+	  [ruby-dev:30770]
+
+Sat May 19 10:26:01 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (Date._parse): detects some OFX dates
+	  (Of course not fully).
+
+Sat May 19 03:08:05 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_inject): minor improvement.  [ruby-dev:30792]
+
+	* enum.c (one_i): no needs to iterate once the result became false.
+
+	* enum.c (enum_one): fix for an example.
+
+	* enum.c (one_iter_i, none_iter_i): DRY.;
+
+Sat May 19 01:07:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_inject): it is now can work without block.  you
+	  have to specify two argument method name as the first argument.
+
+	* enum.c (Init_Enumerable): reduce is new alias to inject.
+
+Sat May 19 01:05:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (Init_File): method definition mismatch.
+
+Fri May 18 16:44:04 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (Init_File): add to_path method to File objects.
+
+Fri May 18 11:12:39 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (DllMain, ruby_init_loadpath): use DLL instance handle given
+	  to DllMain instead of VirtualQuery so that loadpath becomes relative
+	  from the DLL on WinCE too.
+
+Thu May 17 17:03:11 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el (ruby-style-label-indent): for yacc rules.
+
+Thu May 17 13:30:27 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_arg): remove typo from ripper description.
+
+Thu May 17 13:23:38 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y, compile.c (set_arguments): fix to support in-paren
+	  parameter (ex: def foo((a, b))).
+
+Thu May 17 13:01:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (ruby_iseq_disasm): fix to show post arg info.
+
+Thu May 17 12:56:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* debug.c (ruby_debug_node): fix to show node line.
+
+Wed May 16 21:48:44 2007  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb (Logger::Application): remove meaningless logdev
+	  attribute and added logger attribute instead.  [ruby-core:11143]
+	  also added Logger#formatter rdoc comment.
+
+Tue May 15 16:40:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb (test_endblockwarn): now parser
+	  warnings emit source names and line numbers.
+
+Tue May 15 15:01:38 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (init_stdhandle): stderr should be without buffering,
+	  but mswin32 use buffering when stderr is not connected to tty.
+
+Mon May 14 02:12:07 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_zip): a.zip(b,c) should return an array, not
+	  enumerator.
+
+	* array.c (rb_ary_zip): a.zip(b,c) should return array with size
+	  truncated to the size of its shortest argument array.
+	  [incompatible]
+
+Mon May 14 01:54:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_choice): should return nil when the array is
+	  empty.
+
+Sat May 12 18:26:36 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (tokens): forgot to add strip. [ruby-core:11120]
+
+	* test/net/http/test_http.rb: test Net::HTTP.post_form.
+
+Fri May 11 15:27:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_data_to_ary): internal IDs must not be exposed.
+	  [ruby-core:11073]
+
+	* parse.y (internal_id_gen): now returns scope local ID instead of
+	  global one.
+
+Thu May 10 15:15:53 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_super.rb: add tests.
+
+Thu May 10 15:14:05 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_s_conv): rdoc fix.
+
+Thu May 10 15:09:23 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (POINTER_P): pointer may be larger than long.
+
+	* parse.y (vtable_size, vtable_included, vtable_tblcpy,
+	  vtable_to_tbl): constified.
+
+Thu May 10 10:13:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_priority): rdoc fix; the initial value is
+	  inherited from the creating thread.  [ruby-core:10607]
+
+Wed May  9 12:28:57 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (Init_Bignum), numeric.c (Init_Numeric): added fdiv as
+	  aliases of quo.  [ruby-dev:30771]
+
+Tue May  8 23:39:09 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (Date._parse): revised treatment of
+	  hyphened/separatorless dates.
+
+	* lib/date/format.rb: some trivial adjustments.
+
+Tue May  8 20:23:07 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: reverted.
+
+Tue May  8 19:32:18 2007  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/rational.rb: fix high-precision Rationals cannot be
+	  converted to Floats. [ruby-Bugs:10502], [ruby-core:11069],
+	  [ruby-dev:30743]
+
+Mon May  7 10:59:55 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/image.rb, test/rss/test_image.rb: fixed Image module
+	  namespace URI. reported by Dmitry Borodaenko. Thanks.
+
+Sun May  6 18:44:11 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (Net::HTTP.post_form): allow an Array of String
+	  for pairs argument. [ruby-Bugs:10340]
+
+	* lib/net/http.rb (Net::HTTP#set_form_data): ditto.
+
+Sun May  6 17:54:36 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: Connection header field might include both of
+	  "keep-alive" token and "close" token. [ruby-core:10818]
+
+Sat May  5 16:26:33 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/date/format.rb (Format::Bag#method_missing): get rid of
+	  modifying original argument.  [ruby-core:11090]
+
+Thu May  3 22:20:08 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, defines.h, eval_load.c (rb_feature_p, rb_provided,
+	  search_required, rb_require_safe), ext/extmk.rb: Fix
+	  a bug where a statically linked extension cannot be autoloaded.
+	  [ruby-dev:30023] / [ruby-dev:30239]
+
+	* thread.c: added an internal class, Barrier.
+
+	* thread.c: copied rdocs from fastthread.
+
+	* yarvcore.h (struct rb_vm_struct): moved loading_table from global.
+
+Thu May  3 18:10:12 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_evalbody.ci, insns.def, vm.c, tool/insns2vm.rb (rb_num_t):
+	  renamed to get rid of name clash.  [ruby-dev:30504]
+
+	* yarvcore.c (ruby_thread_init): ditto.
+
+Wed May  2 18:52:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, yarvcore.h, yarvcore.c, insns.def: fix to mark VM stack
+	  in correct range.
+
+Wed May  2 17:13:26 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_quo): now calculate in integer.  [ruby-dev:30753]
+
+Wed May  2 15:14:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.h: add redefine checks ([ruby-dev:30751]).
+
+Wed May  2 11:22:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: use Qtrue instead of 2.
+
+	* vm.c, insns.def: support "lambda" calling convention.
+
+Wed May  2 06:46:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c, parse.y, ruby.h (rb_compile_warn, rb_compile_warning): warn
+	  for compilation.  the parser should no longer use rb_warn() and
+	  rb_warning().  [ruby-dev:30121]
+
+Wed May  2 05:45:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assoc): result of assoc_new needs to be an assoc.
+
+Wed May  2 05:40:43 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_pow): improvement by calculating from MSB and using
+	  factorization.  <http://yowaken.dip.jp/tdiary/20070426.html#p01>
+
+Tue May  1 18:45:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* sample/test.rb: import matzruby's sample/test.rb.
+
+Tue May  1 17:46:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_choice): a new method to choose an element
+	  randomly from an array.
+
+	* array.c (rb_ary_choice): fixed mistake from RDoc.
+
+Tue May  1 13:59:18 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (proc_arity): fix an arity bug ([ruby-core:11060]).
+
+Tue May  1 13:12:49 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, compile.c (set_arguments): support post arguments.
+
+	* test/ruby/test_method.rb: add tests for above.
+
+	* test/ruby/test_proc.rb: ditto.
+
+	* proc.c: fix an arity bug ([ruby-core:11029]).
+
+	* vm.c, vm.h, insns.def, vm_dump.h: fix bmethod process.
+
+	* vm.c: support block argument on block parameter.
+
+Fri Apr 27 17:05:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (int_pow): bugfix of overflow detection.
+
+	* numeric.c (int_pow): rb_big_pow() may return other than Bignum.
+
+Fri Apr 27 01:51:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: support multiple splat (e.g, [a, *b, *c, e, *f]).
+
+Fri Apr 27 00:03:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_pow): truncate all zero BDIGITs. [ruby-dev:30733]
+
+Thu Apr 26 17:31:00 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_pow): reduce multiplying for even number.
+
+	* numeric.c (int_pow): calculate power in Fixnum as possible.
+	  [ruby-dev:30726]
+
+Thu Apr 26 17:18:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y: fixes for ripper.
+
+	* parse.y (primary): reduced duplicated code.
+
+	* parse.y (f_arg_item): should not override by meaningless value.
+
+	* parse.y (f_arg, assocs): should not use $$ before assigned.
+
+	* parse.y (assoc_list): dispatch assoclist_from_args for assocs as
+	  well as args.
+
+	* parse.y (assoc): return assoc if dispatched result is $1.
+
+Thu Apr 26 13:54:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-style.el: new file.  C/C++ style for ruby source code.
+
+Wed Apr 25 19:49:16 2007  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (unix_send_io, unix_recv_io): use CMSG_DATA to
+	  align file descriptor appropriately.
+
+Wed Apr 25 15:23:29 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (NtInitialize, exit_handler): add initializing and
+	  cleanup of critical section object for select.
+
+	* win32/win32.c (do_select): block reentrance.
+
+	* win32/win32.c (rb_w32_select): 0 sec polling of socket. this is
+	  workaround because winsock cannot do select at same socket at the
+	  same time by two or more threads.
+
+Wed Apr 25 14:10:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/probeprofiler/probeprofiler.c: fix function name and
+	  return value.
+
+Wed Apr 25 12:42:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h: remove rb_control_frame_t#callee_id.
+
+	* vm_macro.def: ditto.
+
+	* eval_intern.h (exec_event_hooks): fix to check event flags
+
+	* eval_intern.h (EXEC_EVENT_HOOK): fix to re-check event flags.
+
+	* ext/probeprofiler : added.  this profiler is sampling based
+	  profiler.
+
+	* vm.c: add rb_thread_current_status() API for probeprofiler.
+
+	* thread.c (rb_thread_execute_interrupts): add comments.
+
+Wed Apr 25 10:36:03 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (PUSH_TAG): no argument now.
+
+	* eval.c, eval_error.h, eval_jump.h, eval_load.c, proc.c, thread.c:
+	  ditto.
+
+	* thread.c (alloc_event_hook, rb_thread_remove_event_hook): should
+	  return value.
+
+Tue Apr 24 09:33:57 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (do_stat, do_lstat, do_opendir): should not warn ENOTDIR.
+	  [ruby-talk:248288]
+
+Mon Apr 23 22:14:42 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb ($ruby): add extout directory to include path.
+	  [ruby-core:11003]
+
+	* lib/mkmf.rb (libpathflag): not to append RPATHFLAG to current
+	  directory.
+
+	* lib/mkmf.rb (init_mkmf): add current directory to default
+	  library path with highest priority.  [ruby-core:10960]
+
+	* lib/mkmf.rb (LINK_SO): LIBPATH to be placed before DLDFLAGS.
+
+Fri Apr 20 16:05:22 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBPATHFLAG, RPATHFLAG): no needs to be quoted,
+	  it is done by libpathflag in mkmf.rb.
+
+Fri Apr 20 12:27:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb: fix to override conv proc.
+
+Fri Apr 20 12:21:28 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): fixed access to out of bound, and inverted
+	  the order of errinfos.
+
+Fri Apr 20 10:33:23 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h: add prototypes of rb_sourceline() and
+	  rb_sourcefile().
+
+Fri Apr 20 02:37:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): re-send signal.  [ruby-dev:30516]
+
+	* eval_error.h (error_handle): no message when exiting by signal.
+
+	* intern.h (rb_thread_signal_raise, ruby_default_signal): prototypes.
+
+	* signal.c (esignal_init): takes a signal number and an optional
+	  signal name.
+
+	* signal.c (interrupt_init): pass SIGINT always.
+
+	* signal.c (ruby_default_signal): invoke system default signal
+	  handler.
+
+	* signal.c (rb_f_kill): use NUM2PIDT instead of NUM2INT.
+
+	* signal.c (rb_signal_exec, trap): handle SIGTERM.  [ruby-dev:30505]
+
+	* thread.c (rb_thread_signal_raise): now takes signal number instead
+	  of signal name.
+
+	* thread.c (rb_thread_signal_exit): since rb_make_exception() calls
+	  #exception method, rb_class_new_instance() is not needed here.
+
+	* yarvcore.h (struct rb_vm_struct), eval_jump.h (terminate_process):
+	  exit_code is no longer stored in VM.
+
+Thu Apr 19 18:37:49 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, node.h, thread.c, yarvcore.[ch], eval_intern.h:
+	  support set_trace_func (incomplete.  id and klass
+	  don't be passed).  And support Thread#set_trace_func
+	  which hook only specified thread and Thread#add_trace_func
+	  which add new trace func instead of replace old one.
+	  C level API was modified.  See thread.c (logic) and
+	  yarvcore.h (data structures).
+
+	* vm.c, vm_macro.def: add hook points.
+
+	* compile.c, insns.def: fix "trace" instruction.
+
+	* iseq.c, vm_macro.h: add compile option "trace_instruction".
+
+	* test/ruby/test_settracefunc.rb: hook "c-return" of set_trace_func.
+
+Thu Apr 19 20:57:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (symbol): symbols should be followed by EXPR_ENDARG.
+
+	* parse.y (dsym): ditto.
+
+	* parse.y (parser_yylex): strings should be followed by
+	  EXPR_ENDARG.
+
+	* parse.y (parser_yylex): ditto for numbers.
+
+	* parse.y (parser_yylex): EXPR_ENDARG after ']' and '}'.
+
+Thu Apr 19 17:46:36 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/optparse.rb: fix to override conv proc.
+
+Wed Apr 18 10:41:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strtod): exponent is radix 10.  [ruby-talk:248272]
+
+Wed Apr 18 02:50:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* yarvcore.c (th_init2): push initial blockptr value for
+	  rb_block_given_p() outside ruby_exec().  [ruby-core:10923]
+
+Wed Apr 18 02:30:24 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LDFLAGS): prepend -L. instead appending it to
+	  XLDFLAGS.  [ruby-core:10933]
+
+	* configure.in (Makefile): remove $U for automake from MISSING.
+	  [ruby-talk:248171]
+
+Mon Apr 16 22:56:01 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/pty/expect_sample.rb: avoid symbolic link representation for
+	  expect.  a patch from Kazuhiro NISHIYAMA <zn at mbf.nifty.com>.
+	  [ruby-dev:30714]
+
+Mon Apr 16 22:51:11 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample: replace TRUE, FALSE with true, false respectively.
+	  a patch from Kazuhiro NISHIYAMA <zn at mbf.nifty.com>.
+	  [ruby-dev:30713]
+
+Mon Apr 16 17:08:02 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (make_switch): do not clobber converter if pattern
+	  has no convert method.  reported by sheepman in [ruby-dev:30709].
+
+Mon Apr 16 16:49:32 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_seek): consistent behavior with
+	  IO#seek.  patch by sheepman in [ruby-dev:30710].
+
+Mon Apr 16 16:34:08 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): should set command_start after block
+	  starting "do"s and braces.  [ruby-core:10916]
+
+Mon Apr 16 10:51:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_each_with_index): each_with_index to forward
+	  arguments to each.  [ruby-core:10921]
+
+Mon Apr 16 10:43:10 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_arg): should allow to specify 24:00.
+	  [ruby-core:10915]
+
+Sun Apr 15 09:12:54 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: added some zone names.
+
+	* lib/date/format.rb (_parse): now interprets doted numerical
+	  dates as a big endian (except dd.mm.yyyy).
+
+Thu Apr 12 17:13:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (mutex_try_lock): check and set owner thread.
+
+	* thread_pthread.ci: fix to show error code in error message.
+
+Thu Apr 12 17:11:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_rescue2): restore cfp ([ruby-dev:30582]).
+
+Thu Apr 12 16:06:48 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_protect): restore cfp ([ruby-dev:30671]).
+
+Thu Apr 12 16:04:31 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): check node->nd_state == 1, not !0.
+
+Wed Apr 11 16:35:16 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.[ch] (rb_w32_enter_critical, rb_w32_leave_critical): no
+	  need to reject reentrance. removed.
+
+	* rubysig.h (RUBY_CRITICAL): follow above changes.
+
+	* rubysig.h (TRAP_BEG, TRAP_END): no need to save errno.
+
+Tue Apr 10 17:02:17 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_fclose, rb_w32_close): need to save errno
+	  before calling original fclose()/close().
+
+Tue Apr 10 16:14:22 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.ci (w32_wait_events): check whether interrupt_event is
+	  valid handle or not.
+
+	* thread_win32.ci (native_thread_destroy): clear interrupt_event when
+	  close it.
+
+Tue Apr 10 15:53:17 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_pthread.ci (native_thread_create): initialize sleep_cond.
+	  fixed: [ruby-dev:30675]
+
+Mon Apr  9 18:48:57 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (do_select): ubf_select() is not necessary. interrupt is
+	  checked in the loop.
+
+Mon Apr  9 18:27:26 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (do_select): use ubf_select() as UBF on windows.
+
+	* win32/win32.c (do_select): shouldn't call catch_interrupt() here.
+	  fixed: [ruby-dev:30674], reported by wanabe.
+
+Mon Apr  9 09:24:32 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (disconnect): call shutdown for
+	  SSLSocket. Thanks, Technorama Ltd.
+
+Sun Apr  8 13:28:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (defined_expr): test arguments of NODE_CALL and so
+	  on as well as NODE_ATTRASGN.  [ruby-core:10886]
+
+Fri Apr  6 10:56:29 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_get_group): get rid of
+	  warning. we are aware of it.
+
+Fri Apr  6 04:00:24 2007  Technorama Ltd.  <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_{bn,x509{attr,cert,name,store}}.c:
+	  Add documentation.
+
+Thu Apr  5 17:59:19 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (defined_expr): support for assignment.
+	  [ruby-core:10867]
+
+	* compile.h (ADD_CATCH_ENTRY): removed temporary variable.
+
+Thu Apr  5 15:13:34 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_alloc): should
+	  return value.
+
+Thu Apr  5 14:58:49 2007  Technorama Ltd. <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_pkcs5.c: New module.
+
+	* ext/openssl/ossl_{cipher,digest,pkcs7,pkcs12}.c:
+	  Remove redundant module namespace.
+
+	* ext/openssl/lib/openssl/{cipher,digest}.rb
+	  Add backwards compatible classes for rearranged classes.
+
+	* ext/openssl/ossl_{pkcs7,pkcs12}.c: Add documentation.
+
+Thu Apr  5 00:42:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_notimplement), io.c (pipe_open): removed definite
+	  articles and UNIX manual section from messages.  [ruby-dev:30690]
+
+Wed Apr  4 17:09:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): refined the message of NotImplementedError.
+	  [ruby-dev:30685]
+
+Wed Apr  4 12:29:02 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* error.c (rb_notimplement): should show the name of this func,
+	  not callee.
+
+Wed Apr  4 10:18:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (popen_exec): should not close close-on-exec FDs.
+	  [ruby-dev:30679]
+
+	* io.c (pipe_open): raise NotImplementedError for command "-" on
+	  platforms where fork(2) is not available.  [ruby-dev:30681]
+
+Tue Apr  4 04:17:18 2007  Technorama Ltd. <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_ssl.c: Add documentation.
+
+Tue Apr  3 16:22:24 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/extconf.rb: check for functions added in 1.9.
+
+	* ext/openssl/ruby_missing.h: check per features instead by
+	  checking version code.  [ruby-core:10845]
+
+Tue Apr  3 16:02:44 2007  Technorama Ltd. <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_bn.c: More documentation.
+
+	* ext/openssl/lib/ossl_{pkey,pkey_ec}.[ch]: Add elliptic curves.
+
+Tue Apr  3 15:50:41 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (s_recv, s_recvfrom): some systems (such as
+	  windows) doesn't set fromlen if the socket is connection-oriented.
+	  reported by Bram Whillock in [ruby-core:10512] [ruby-Bugs#9061]
+
+Tue Apr  3 09:36:55 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ruby_missing.h: need to include version.h to check
+	  RUBY_VERSION_CODE.
+
+Mon Apr  3 07:10:12 2007  Technorama Ltd. <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_{ssl.[ch],ssl_session.c},
+	  ext/openssl/lib/openssl/lib/openssl/ssl.rb:
+	  New SSL::Session class.  Add session cb's, getter/setters,
+	  config, and statistics methods.
+
+Mon Apr  3 04:00:23 2007  Technorama Ltd. <oss-ruby@technorama.net>
+
+	* ext/openssl/{ossl.[ch],ossl_pkey.c} Add documentation.
+
+	* ext/openssl/ossl_hmac.c Add reset method.
+
+	* ext/openssl/ossl_cipher.c (Cipher#update) Take additional
+	  buffer argument.
+
+	* ext/openssl/{ossl_bio.c,ossl_ssl.c,ruby_missing.h}
+	  compatibility with 1.8.
+
+Mon Apr  2 21:55:12 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (throw), thread.c, yarvcore.h (throwed_errinfo): fixed
+	  typo.
+
+Fri Mar 30 11:46:51 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_cmdvector): fixed buffer size. reported by
+	  wanabe [ruby-dev:30672]
+
+	* win32/win32.c (init_env, insert, rb_w32_get_environ): use strdup
+	  instead of malloc + strlcpy. suggested by nobu [ruby-dev:30673]
+
+Fri Mar 30 02:29:04 2007  Technorama  <oss-ruby@technorama.net>
+
+	* ext/openssl/ossl_{bn,cipher,digest,hmac,rand,pkey_{dh,dsa,rsa}}.c:
+	  Add Documentation for various methods.
+
+	* ext/openssl/lib/openssl/cipher.rb: Ditto
+
+	* ext/openssl/ossl_bn.c: add lshift! and rshift! methods.
+
+	* ext/openssl/ossl_digest.c: GetDigestPtr() also accept a string.
+
+Fri Mar 23 11:28:24 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (init_env, insert, cmdglob, rb_w32_cmdvector,
+	  rb_w32_opendir, rb_w32_readdir, rb_w32_strerror, rb_w32_stati64,
+	  rb_w32_get_environ): use strlcpy() and strlcat().
+
+	* win32/win32.c (rb_w32_opendir): use realloc() instead of xrealloc().
+
+	* win32/win32.c (rb_w32_closedir): check NULL before free pointers.
+
+Fri Mar 23 00:24:52 2007  Keiju Ishitsuka <keiju@ruby-lang.org>
+
+	* lib/shell: commit miss(support for ruby 1.9(YARV) thread model).
+
+Thu Mar 22 13:32:17 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (LIBS): remove an unnecessary library.
+
+Thu Mar 22 10:27:58 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_bignum.rb (test_to_s): add tests for Bignum#to_s.
+
+Wed Mar 21 20:38:06 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* marshal.c (w_short, w_long, w_object): get rid of VC++ warnings.
+
+Wed Mar 21 20:05:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, parse.y, eval.c, intern.h, iseq.c, lex.c, node.h,
+	  proc.c, vm.c, vm_macro.def, vm_macro.def, yarvcore.c, yarvcore.h,
+	  debug.c, debug.h: merge half-baked-1.9 changes.  The biggest change
+	  is to change node structure around NODE_SCOPE, NODE_ARGS.  Every
+	  scope (method/class/block) has own NODE_SCOPE node and NODE_ARGS
+	  represents more details of arguments information.  I'll write a
+	  document about detail of node structure.
+
+Wed Mar 21 17:04:30 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): round up for the most significant digit.
+	  [ruby-core:10686]
+
+Tue Mar 21 08:20:00 2007  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* test/testunit/collector/test_dir.rb: Fixed test/unit tests that
+	  were breaking due to Module#public_instance_methods now
+	  returning a Symbol instead of a String.
+
+	* test/testunit/collector/test_objectspace.rb: Ditto.
+
+Tue Mar 20 22:54:50 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_extended): erroneous check condition when dump
+	  method is defined.  [ruby-core:10646]
+
+Tue Mar 20 21:36:47 2007  Keiju Ishitsuka <keiju@ruby-lang.org>
+
+	* lib/shell.rb, lib/shell: support for ruby 1.9(YARV) thread model.
+
+Tue Mar 20 16:36:08 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* distruby.rb: Add zip generation.
+
+Tue Mar 20 16:20:49 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_callee_name): add __method__ and __callee__ again.
+	  __callee__ need to rework to adopt YARV.  [ruby-core:10671]
+
+Tue Mar 20 11:09:00 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb: Revise rdoc.
+
+	* lib/set.rb (Set#freeze, Set#taint, Set#untaint): Implement
+	  Set#freeze, Set#taint, and Set#untaint; requested by: Dan
+	  Hutchings <dan AT moltoagitato.com> in [ruby-bugs:PR#9359].
+
+Tue Mar 20 09:13:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_fork): flush stdouts always before fork(2).
+	  fixed: [ruby-dev:30612]
+
+Tue Mar 20 01:38:48 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): store the result of first_func
+	  as well as first_proc.
+
+	* thread.c (thread_create_core): block is not used if first_func
+	  is given.
+
+Mon Mar 19 16:58:52 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* missing/{strlcat,strlcpy}.c, missing.h: new functions.
+
+	* LEGAL: add copyright notice about above files.
+
+	* configure.in: check whether strlcat and strlcpy are exist or not.
+
+	* {bcc32,win32,wince}/Makefile.sub: use above files.
+
+Mon Mar 19 14:12:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/matrix.rb (Matrix::inverse_from): adding partial pivoting to
+	  the Gauss-Jordan algorithm, making it stable.  a patch from
+	  Peter Vanbroekhoven.  [ruby-core:10641]
+
+Mon Mar 19 12:13:36 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regparse.c, etc.: K&R to ANSI code cleanup patch from Stefan
+	  Huehner <stefan at huehner.org>.  [ruby-core:10543]
+
+Mon Mar 19 11:27:13 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/protocol.rb (rbuf_read): extend buffer size for speed.
+
+Sun Mar 18 08:31:51 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/dir.h, win32/win32.c (rb_w32_opendir, rb_w32_readdir,
+	  rb_w32_closedir): get rid of possible buffer-overflows.
+
+Sat Mar 17 19:10:39 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss, test/rss:
+	- supported Atom.
+	- bumped version 0.1.6 to 0.1.7.
+	* sample/rss/convert.rb: added new sample.
+
+Fri Mar 16 22:32:20 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/pop.rb: change default verification mode from
+	  VERIFY_PEER to VERIFY_NONE because most POPS server does not have
+	  true certification.
+
+Fri Mar 16 22:19:24 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE#ole_activex_initialize,
+	  a little bit supporting ActiveX control. [ruby-talk:241188]
+
+Fri Mar 16 22:16:58 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: merge Ruby-SSPI patch contributed by Justin
+	  Bailey.
+
+	* ext/Win32API/lib/win32/sspi.rb: new file.
+
+Wed Mar 14 12:30:00 2007  Shigeo Kobayashi  <shigeo@tinyforest.jp>
+
+	* ext/bigdecimal/bigdecimal.c: BigDecimal("-.31") is now
+	  treated as ("-0.31") not as ("0.31").
+
+Tue Mar 13 19:04:30 2007  Keiju Ishitsuka <keiju@ruby-lang.org>
+
+	* lib/sync.rb: support for ruby 1.9(YARV) thread model.
+
+Tue Mar 13 09:25:10 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (clear-installed-list): separated from install-prereq.
+
+Tue Mar 13 07:23:20 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb (mon_try_enter): call @mon_muetx.try_lock.
+	  Thanks, Keiju ISHITSUKA. [ruby-dev:30507]
+
+Tue Mar 13 02:42:58 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/cgi.rb (CGI::header): IIS >= 5.0 does not need the nph
+	  assumption any more; submitted by MIYASAKA Masaru <alkaid AT
+	  coral.ocn.ne.jp> in [ruby-dev:30537].
+
+Mon Mar 12 10:53:28 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/openssl/ossl_asn1.c (Init_ossl_asn1): Let rdoc know about
+	  externally defined modules; submitted by Technorama
+	  Ltd. <oss-ruby AT technorama.net> in [ruby-bugs:PR#4704].
+
+	* ext/openssl/ossl_bn.c (Init_ossl_bn): Ditto.
+
+	* ext/openssl/ossl_cipher.c (Init_ossl_cipher): Ditto.
+
+	* ext/openssl/ossl_digest.c (Init_ossl_digest): Ditto.
+
+	* ext/openssl/ossl_hmac.c (Init_ossl_hmac): Ditto.
+
+	* ext/openssl/ossl_pkey.c (Init_ossl_pkey): Ditto.
+
+	* ext/openssl/ossl_pkey_dh.c (Init_ossl_dh): Ditto.
+
+	* ext/openssl/ossl_pkey_dsa.c (Init_ossl_dsa): Ditto.
+
+	* ext/openssl/ossl_pkey_rsa.c (Init_ossl_rsa): Ditto.
+
+	* ext/openssl/ossl_rand.c (Init_ossl_rand): Ditto.
+
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): Ditto.
+
+Sun Mar 11 18:42:01 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-mode.el (ruby-block-end-re): Support for the
+	  experimental ';;' terminator had been dropped.
+
+Sun Mar 11 05:45:46 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/README, misc/rdebug.el: Add rdebug.el, Emacs ruby-debug
+	  interface based on rubydb3x.el; submitted by Martin Nordholts
+	  <enselic AT gmail.com> in [ruby-bugs:PR#9023].
+
+Sat Mar 10 07:20:28 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_TYPELIB#library_name,
+	  WIN32OLE_TYPELIB#visible?.
+
+	* test/win32ole/test_win32ole_typelib.rb: ditto.
+
+Thu Mar  8 09:17:59 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* compile.c: iseq_compile -> rb_iseq_compile.
+
+	* iseq.c: ditto.
+
+	* intern.h: provide function prototype of Init_jump.
+
+	* eval_jump.h (Init_jump): declare function type.
+
+	* thread.c: platform-dependent functions should be surrounded by #ifdef.
+
+	* iseq.c (iseq_data_to_ary): remove unused variable.
+
+	* compile.c (set_arguments): ditto.
+
+	* thread.c (set_unblock_function): ditto.
+
+	* thread_pthread.ci: reduce printf warning.
+
+	* vm_dump.c: ditto.
+
+Tue Mar  6 16:35:04 2007  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/shell/process-controller.rb: fix thread synchronization problem for [ruby-dev:30477].
+
+Tue Mar  6 11:53:25 2007  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/irbtkw.rbw: fails to exit process.
+
+Tue Mar  6 10:23:09 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* runruby.rb: added --pure (turned on by default) and --debugger
+	  options.
+
+Mon Mar  5 09:19:33 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/timeout.rb (Timeout.timeout): should return the block value
+	  always.
+
+	* lib/timeout.rb (Timeout.timeout): should yield sec argument
+	  always.
+
+	* lib/timeout.rb (Timeout.timeout): fix document.
+
+Mon Mar  5 09:16:40 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb: support automatic STARTTLS.
+
+	* lib/net/smtp.rb: check server advertisement.
+
+	* lib/net/smtp.rb: introduce new class SMTP::Response.
+
+	* lib/net/smtp.rb (getok): should not use sprintf.
+
+	* lib/net/smtp.rb (get_response): ditto.
+
+	* lib/net/protocol.rb: reduce syntax warning on 1.9.
+
+Mon Mar  5 07:13:28 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb: reconstruct SMTPS/STARTTLS interface.  New
+	  interface is incompatible from current 1.9 interface at all.
+
+	* lib/net/smtp.rb: All SSL-related class methods are removed; use
+	  instance methods instead.
+
+	* lib/net/smtp.rb: rename methods: *ssl -> *tls (with alias
+	  "ssl").
+
+	* lib/net/smtp.rb: rename methods: *tls -> *starttls.
+
+Mon Mar  5 01:36:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex), win32/win32.c (rb_w32_utime): fixed
+	  indentation broken at YARV merger.
+
+Sun Mar  4 23:41:14 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_stat_uid, rb_stat_gid, eaccess): use rb_uid_t and
+	  rb_gid_t instead of int.
+
+	* file.c (rb_stat_s_utime): fixed a commit miss for the platforms
+	  where utimes() does not exist.
+
+	* lib/fileutils.rb (touch): ditto.
+
+Sun Mar  4 14:46:56 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* util.c (push_element): should return a int value.
+
+Sun Mar  4 01:01:25 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#^, Set#&): Correct documentation.  Those methods
+	  return sets, not arrays; noted by Oliver Frank Wittich <nietz AT
+	  mangabrain.de>.
+
+Sat Mar  3 22:54:33 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (touch): last #touch change causes error when
+	  :mtime option was not given.
+
+Sat Mar  3 22:51:29 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mv): could not move directory between
+	  different file systems. [ruby-dev:30411]
+
+Sat Mar  3 22:37:02 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_utime): allow nil to set the current time.
+
+	* lib/fileutils.rb (touch): ditto, and added :mtime and :nocreate
+	  options.  fixed: [ruby-talk:219037]
+
+Sat Mar  3 15:52:26 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* object.c (instance_variable_get): Restore rdoc markups lost in
+	  the last commit.
+
+Fri Mar  2 21:17:14 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (get_ptr_of_variant, ole_set_safe_array,
+	  ole_val2ptr_variant, ole_val2olevariantdata, ole_variant2val,
+	  ): fix some bugs of WIN32OLE_VARIANT.new when variant type is
+	  VT_ARRAY|VT_BSTR or VT_BYREF.
+
+	* ext/win32ole/win32ole.c (folevariant_s_array, folevariant_initialize):
+	  WIN32OLE_VARIANT#[], WIN32OLE_VARIANT#[]=, WIN32OLE_VARIANT#value=
+	  is defined as instance method of WIN32OLE_VARIANT.
+
+	* test/win32ole/test_win32ole_variant.rb: add some test for
+	  VT_ARRAY, VT_BYREF variant type.
+
+Fri Mar  2 07:58:24 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_ivar_set): RDoc updated according to a
+	  suggestion from Brian Candler <B.Candler AT pobox.com>.
+	  [ruby-core:10469]
+
+Thu Mar  1 21:38:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt, arg): should not omit lhs of OP_ASGN1 even if
+	  empty.  [ruby-dev:30455]
+
+Thu Mar  1 02:55:25 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (get_digest_base_metadata): Allow inheriting
+	  Digest::Base subclasses, which was unintentionally made
+	  impossible while restructuring Digest classes.
+
+Thu Mar  1 02:05:17 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb (patchlevel): read from version.h.
+
+Wed Feb 28 21:15:00 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (ac_cv_func_fcntl): fcntl support for MinGW.
+
+	* missing/flock.c: workaround for MinGW.
+
+Wed Feb 28 20:51:32 2007  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* pack.c (pack_unpack): properly ignore non-base64 octets such as
+	  UTF-8 encoded BOMs; submitted by SOUMA Yutaka <holon@radastery.jp>
+	  to fix [ruby-core:10437]
+
+Wed Feb 28 18:31:51 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/extconf.rb: no need to check unistd.h and sys/time.h.
+	  they are already checked at configure.
+	  reported by KOBAYASHI Yasuhiro [ruby-list:43225]
+
+Wed Feb 28 18:23:43 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb ($DEFLIBPATH): default library paths ($(topdir), etc)
+	  should be the first elements of library paths list.
+	  reported by KOBAYASHI Yasuhiro [ruby-list:43225]
+
+Wed Feb 28 10:33:58 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, configure.in, */Makefile.sub (THREAD_MODEL): system
+	  specific thread model.
+
+	* compile.h, regint.h, vm.h, array.c: removed unnecessary #include.
+
+Wed Feb 28 04:03:03 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (take_i): small cosmetic / documentation patch from
+	  Tadashi Saito <shiba AT mail2.accsnet.ne.jp>. [ruby-dev:30446]
+
+Wed Feb 28 01:20:18 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/{dbm,gdbm}/test_{dbm,gdbm}.rb: shouldn't use host_os. use
+	  target_os instead. reported by KOBAYASHI Yasuhiro [ruby-list:43225]
+
+Wed Feb 28 00:08:11 2007  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
+
+	* mkconfig.rb (RbConfig): add CONFIG['PATCHLEVEL']
+
+	* common.mk: new target dist
+
+	* distruby.rb: new file
+
+Tue Feb 27 22:18:45 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (--enable-auto-image-base): avoid the necessity to
+	  rebase the shared libs as much as possible;
+	  submitted by Corinna Vinschen <spam at vinschen.de> in
+	  [ruby-talk:240964].
+
+Tue Feb 27 21:36:47 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* util.c (__crt0_glob_function): use ruby_glob() instead of rb_globi().
+
+Tue Feb 27 21:33:04 2007  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (ac_cv_func_setrlimit): workaround for djgpp.
+
+Tue Feb 27 20:35:28 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (error.c, process.c): depend on yarvcore.h and rubysig.h.
+
+Tue Feb 27 19:26:31 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/nkf/nkf.c (rb_str_resize, rb_nkf_kconv, rb_nkf_guess1,
+	  rb_nkf_guess2): Silence warnings regarding char * vs. unsigned
+	  char * mismatch; submitted by Lyle Johnson
+	  <lyle.johnson@gmail.com> in [ruby-core:10416].
+
+Tue Feb 27 19:15:01 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/base64.rb (Base64::b64encode): Fix documentation; submitted
+	  by David Symonds <dsymonds@gmail.com> in [ruby-core:10432].
+
+Tue Feb 27 18:59:42 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_load): Silence warnings
+	  regarding char * vs. unsigned char * mismatch; submitted by Lyle
+	  Johnson <lyle.johnson@gmail.com> in [ruby-core:10416].
+
+	* ext/digest/sha1/sha1ossl.c (SHA1_Finish): Ditto.
+
+	* ext/digest/rmd160/rmd160ossl.c (RMD160_Finish): Ditto.
+
+	* ext/digest/digest.c (rb_digest_base_finish,
+	  rb_digest_base_update): Ditto.
+
+Tue Feb 27 18:12:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_take): new method.  [ruby-dev:30407]
+
+	* enum.c (enum_drop): ditto.
+
+Tue Feb 27 07:47:24 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, vm.h: rename th_invoke_yield() to th_yield().
+
+	* blockinlining.c: ditto.
+
+	* eval.c: ditto.
+
+	* vm.c, insns.def: rename th_invoke_yield_cfunc()
+	  to th_yield_with_cfunc().
+
+	* yarvcore.h, yarvcore.c: rename theYarvVM to ruby_current_vm and
+	  yarvCurrentThread to ruby_current_thread.  remove yarvVMArray.
+
+Tue Feb 27 00:45:23 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_optimization.rb: restore method before calling
+	  assert_equal.
+
+Mon Feb 26 00:58:39 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h: add rb_thread_t#top_wrapper, top_self.
+
+	* eval_load.c (rb_load): support eval in wrapper module
+	  (load(file, true)).
+
+	* eval.c: ditto.
+
+	* eval_jump.h: ditto.
+
+	* iseq.c: ditto.
+
+	* vm.c: ditto.
+
+	* yarvcore.c: ditto.
+
+	* insns.def: add a empty line.
+
+Mon Feb 26 00:54:36 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: change "gdb" rule.  You can debug miniruby with
+	  $(srcdir)/test.rb on gdb by this rule (type "make gdb").
+	  If you write break points to "breakpoints.gdb" on $srcdir,
+	  gdb runs with this file.
+
+Sun Feb 25 11:46:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* win32/Makefile.sub: enable -Zi (debug) option.
+
+Sun Feb 25 11:38:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.h: define RUBY_VM macro and remove NATIVETHREAD* macros.
+
+	* intern.h: ditto.
+
+	* signal.c (posix_signal): remove unused function
+	  posix_nativethread_signal().
+
+Sun Feb 25 11:31:13 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_run): fix to ANSI style.
+
+Sun Feb 25 11:09:16 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* bootstraptest/runner.rb: show source code in error message.
+
+Sun Feb 25 09:39:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h:
+	  rename:
+	    rb_iseq_t#file_name -> filename
+	    rb_iseq_t#local_tbl -> local_table
+	  add:
+	    rb_iseq_t#local_table_size
+
+	* compile.c: separate local_table_size and local_size
+	  (local variable size)
+
+	* blockinlining.c: apply above rename.
+
+	* compile.h: ditto.
+
+	* eval.c: ditto.
+
+	* iseq.c: ditto.
+
+	* proc.c: ditto.
+
+	* vm.c: ditto.
+
+	* vm_dump.c: ditto.
+
+Sun Feb 25 10:27:17 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* bootstraptest/runner.rb: add lib/ to load path.
+
+Sat Feb 25 10:16:50 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* rubyio.h (HAVE_RB_IO_T): macro to tell if rb_io_t is defined.
+
+Sat Feb 24 19:39:16 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* common.mk: new target "btest", to run bootstraptests.
+
+Sat Feb 24 19:30:40 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c, gc.c, io.c, ruby.h, rubyio.h, win32/win32.h (rb_io_t):
+	  renamed from OpenFile.
+
+	* ext/dl/cptr.c, ext/io/wait/wait.c, ext/openssl/ossl.h,
+	  ext/openssl/ossl_bio.c, ext/openssl/ossl_ssl.c, ext/pty/pty.c,
+	  ext/readline/readline.c, ext/socket/socket.c: ditto.
+
+Sat Feb 24 19:28:23 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* bootstraptest/runner.rb: new option -v,--verbose.
+
+Sat Feb 24 18:55:50 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* yarvtest/test_method.rb: removed (merged to bootstraptest).
+
+	* yarvtest/test_class.rb: ditto.
+
+Sat Feb 24 18:44:39 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* bootstraptest/test_class.rb: new file.
+
+	* bootstraptest/test_method.rb: add tests.
+
+Sat Feb 24 18:44:30 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h (rb_thread_blocking_region): add prototype.
+
+	* file.c (rb_thread_flock, rb_file_flock): use UBF feature.
+
+	* process.c (rb_waitpid_blocking, rb_waitpid): use UBF feature.
+
+	* thread.c (rb_thread_debug): added runtime debugging flag.
+
+	* thread.c (BLOCKING_REGION): restore previous UBF.
+
+	* thread.c (rb_thread_blocking_region): default UBF to interrupt
+	  in system dependent way by RB_UBF_DFL.
+	  + ubf_select() on posix system
+	  + ubf_handle() on Win32
+	  + none on cygwin
+
+	* thread_win32.ci (rb_w32_wait_events_blocking): blocking version.
+
+	* win32/win32.c (waitpid): use rb_w32_wait_events_blocking().
+
+Sat Feb 24 17:45:48 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y (f_arg, opt_f_block_arg): ripper should export VALUE.
+
+Sat Feb 24 16:52:55 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* bootstraptest/runner.rb: fix syntax error.
+
+Sat Feb 24 16:51:09 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* bootstraptest/runner.rb: new option --help.
+
+Sat Feb 24 16:47:33 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* bootstraptest: new test suite.
+
+	* bootstraptest/runner.rb: new file.
+
+	* bootstraptest/test_literal.rb: new file.
+
+	* bootstraptest/test_method.rb: new file.
+
+Sat Feb 24 16:29:15 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (StartSocket): remove unnecessary code.
+
+Sat Feb 24 16:04:30 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (struct local_vars): remove unused nofree member from
+	  struct.
+
+	* parse.y (parser_free): ditto.
+
+Sat Feb 24 15:57:19 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/thread.rb (ConditionVariable#broadcast): use Mutex
+	  instead of Thread.exclusive.
+
+	* lib/monitor.rb (MonitorMixin#mon_exit): unset @mon_owner
+	  before calling Mutex#unlock.
+
+Sat Feb 24 15:51:45 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y (program): remove useless assignment to reduce warning.
+
+Sat Feb 24 15:41:51 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (lambda): remove unused clause from the rule to stop
+	  warning.
+
+Sat Feb 24 15:41:22 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/thread.rb: do not redefine Mutex#synchronize.
+
+Sat Feb 24 15:14:02 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb: rewritten using Mutex/ConditionVariable.
+
+Sat Feb 24 13:25:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/soap/mapping/factory.rb: catch up with spec changes (return
+	  Symbols instead of Strings).
+
+	* lib/soap/mapping/mapping.rb: ditto.
+
+Sat Feb 24 10:49:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y, node.h, compile.c: change node tree structure.  a purpose
+	  of this change is to unify argument structure of method and block.
+	  this change prohibits duplicate block parameter name.
+	  new argument information:
+	    NODE_ARGS	  [m: int, o: NODE_OPT_ARG, ->]
+	    NODE_ARGS_AUX [r: ID, b: ID, ->]
+	    NODE_ARGS_AUX [Pst: id, Plen: int, init: NODE*]
+	  optarg information:
+	    NODE_OPT_ARGS [idx, expr, ->]
+
+	* vm_macro.def: ditto.
+
+	* gc.c: ditto.
+
+	* iseq.c: ditto.
+
+	* compile.h: fix debug function name.
+
+	* test/ripper/test_scanner_events.rb: |_,_,foo| -> |_1,_2,foo|
+
+	* test/ruby/test_lambda.rb: disable test temporarily.
+
+Sat Feb 24 10:46:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/testunit/test_testcase.rb: catch up with current instance
+	  variable spec.
+
+Sat Feb 24 10:32:59 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: change vm_macro.def rule.
+
+Sat Feb 24 10:38:05 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/racc/cparse/cparse.c (cparse_params_mark): remove useless
+	  rb_gc_mark.  Thanks Tomoyuki Chikanaga. [ruby-dev:30405]
+
+Sat Feb 24 07:31:35 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_VARIANT.array,
+	  WIN32OLE_VARIANT#value=, refactoring.
+
+	* test/win32ole/test_win32ole_variant.rb: add some test for
+	  WIN32OLE_VARIANT.array, WIN32OLE_VARIANT#value=.
+
+Fri Feb 23 18:37:55 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_yield.rb: new test.
+
+	* yarvtest/test_yield.rb: removed (moved to test_yield.rb).
+
+Fri Feb 23 18:27:17 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (rb_thread_polling): check interrupts here.
+
+	* thread_win32.ci (w32_wait_events): rename from w32_wait_event(), and
+	  now receive multiple event handles.
+
+	* win32/win32.c (wait_events, rb_w32_main_context): removed.
+
+	* thread_win32.ci (rb_w32_wait_events): new function.
+
+	* thread_win32.ci, win32/win32.c (rb_w32_sleep, rb_w32_Sleep): move
+	  from win32/win32.c to thread_win32.ci, and use w32_wait_events().
+
+Fri Feb 23 18:13:22 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_optimization.rb: new test (merges test_opts.rb).
+
+	* yarvtest/test_opts.rb: removed.
+
+Fri Feb 23 16:59:39 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_assignment.rb: merge yarvtest/test_massign.
+
+	* yarvtest/test_massign.rb: removed (merged to
+	  test_assignment.rb).
+
+Fri Feb 23 15:58:20 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* signal.c (sighandler): need to re-install sighandler on some
+	  platforms.
+
+Fri Feb 23 15:05:57 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (set_pioinfo_extra): simplified.
+
+Fri Feb 23 14:23:20 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_literal.rb: new test.
+
+Fri Feb 23 12:40:12 2007  James Edward Gray II  <james@grayproductions.net>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client::do_rpc): Make the
+	  Content-Length parameter optional for responses in
+	  xmlrpc/client.rb; suggested by Daniel Berger
+	  <Daniel.Berger@qwest.com> and approved by the maintainer.
+
+	* lib/xmlrpc/create.rb (XMLRPC::Create::conv2value): Add DateTime
+	  support to xmlrpc; approved by the maintainer.
+
+Fri Feb 23 12:24:46 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y (lambda): add ripper event.  This fixes bus error on
+	  "make test-all".
+
+	* ext/ripper/extconf.rb: do not stop build.
+
+Fri Feb 23 12:16:05 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: remove dyna_check_gen() prototype.
+
+Fri Feb 23 11:41:21 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y, compile.c, gc.c, insns.def, intern.h, iseq.c, node.h,
+	  object.c, string.c, variable.c, vm_macro.def: revert private
+	  instance variable feature, which is postponed until next major
+	  release.
+
+	* marshal.c: TYPE_SYMBOL2 removed; MARSHAL_MINOR reverted back to
+	  8th version.
+
+Fri Feb 23 10:53:21 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* thread_pthread.ci (native_mutex_lock): do not call
+	  pthread_mutex_trylock().
+
+Fri Feb 23 10:31:16 2007  Minero Aoki  <aamine@loveruby.net>
+
+	* dln.c: use dlopen on Mac OS X 10.3 or later.
+
+Fri Feb 23 10:03:49 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_ord): need not to check string length;  ord
+	  returns a codepoint for the first character in the string.
+
+Wed Feb 21 22:29:45 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* numeric.c (fix_equal): A bit more optimization.
+
+Wed Feb 21 17:40:37 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_equal): remove FIX2LONG() to optimize.  suggested
+	  in http://t-a-w.blogspot.com/2007/02/making-ruby-faster.html.
+	  [ruby-talk:240223]
+
+	* numeric.c (fix_cmp): ditto.
+
+Wed Feb 21 09:14:04 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_load.c (rb_require_safe): should restore safe level.
+
+Tue Feb 20 21:19:29 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (installed_code_page_proc,
+	  ole_variant2val): small refactoring.
+
+Tue Feb 20 15:11:42 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, vm.c, yarvcore.h: move definition of rb_call_super() to
+	  vm.c from eval.c.  change th_call_super() to static function.
+
+Tue Feb 20 15:08:25 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/io/nonblock/test_flush.rb: YARV doesn't raise any errors if
+	  another thread close IO object which current thread is blocking with.
+
+Tue Feb 20 15:03:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (do_select, rb_thread_wait_fd_rw): raise sys error if
+	  errno is not 0 and EBADF.
+
+Mon Feb 19 22:15:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_REPLACE_TYPE): cache convertible type info.
+
+	* intern.h (rb_detach_process): use rb_pid_t instead of pid_t.
+
+	* ruby.h (PIDT2NUM, NUM2PIDT, UIDT2NUM, NUM2UIDT, GIDT2NUM, NUM2GIDT):
+	  defaulted to conversion using long.
+
+Mon Feb 19 17:14:28 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (unix_peeraddr): wrong syscall name in error
+	  message for #peeraddr. a patch from Sam Roberts
+	  <sroberts at uniserve.com>.  [ruby-core:10366]
+
+Sun Feb 18 22:56:07 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_getgrgid): missed to replace a macro.
+
+Sun Feb 18 19:33:00 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: updated based on date2 4.0.3.
+
+Sun Feb 18 13:11:51 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (pid_t, uid_t, gid_t): check if defined.
+
+	* intern.h, process.c, rubyio.h, ext/etc/etc.c, ext/pty/pty.c: use
+	  rb_{pid,uid,gid}_t instead of plain int.  [ruby-dev:30376]
+
+	* ext/etc/extconf.rb (PIDT2NUM, NUM2PIDT, UIDT2NUM, NUM2UIDT, GIDT2NUM,
+	  NUM2GIDT): moved to configure.in.
+
+Fri Feb 16 21:34:33 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* object.c (rb_obj_ivar_set/get/defined): fix to check :@_v/C id.
+
+	* test/testunit/test_testcase.rb: fix to use instance_variable_get()
+	  to access @_result.
+
+Fri Feb 16 20:59:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* intern.h: add a prototype of rb_sym_to_s().
+
+Fri Feb 16 19:24:22 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* math.c (math_log): update document to mention second optional
+	  argument for logarithm base.
+
+Fri Feb 16 19:19:21 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (mrhs): need to append by arg_append().
+	  [ruby-talk:239385]
+
+Fri Feb 16 11:18:21 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/.document: Apply patch for irb, e2mmap and README by Hugh Sasse
+	  <hgs at dmu.ac.uk> from [ruby-core:10135]
+
+	* lib/prettyprint.rb: Suppress RDoc for PrettyPrint test suite.
+
+Thu Feb 15 20:48:36 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (set_pioinfo_extra): new function for VC++8 SP1
+	  workaround. [ruby-core:10259]
+
+	* win32/win32.c (NtInitialize): call above function.
+
+Thu Feb 15 16:25:54 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/uri/generic.rb (URI::Generic::userinfo): Considering how
+	  `scheme://user:@...', `scheme://:password@...' and
+	  `scheme://:@...' are parsed, an empty user name or password
+	  should be allowed and represented as it is.
+
+Thu Feb 15 01:52:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.(c|h), yarvcore.(c|h) (yarvGlobalStateVersion): rename to
+	  ruby_vm_global_state_version.
+
+Thu Feb 15 01:50:26 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/fileutils/test_fileutils.rb (check_singleton): fix to use
+	  symbol instead of string.
+
+	* test/io/nonblock/test_flush.rb: enable tests.
+
+	* test/xmlrpc/test_webrick_server.rb: ditto.
+
+Thu Feb 15 01:43:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/delegate.rb: catch up with class local variable (@_v) spec.
+
+	* lib/singleton.rb: ditto.
+
+Wed Feb 14 22:52:43 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_variant2val): VC++6 does not
+	  support VT_I8, VT_UI8.
+
+Wed Feb 14 22:10:21 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* configure.in: change stack limit to 2MB from 32MB.
+
+	* win32/Makefile.sub: ditto.
+
+Wed Feb 14 21:39:36 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest.rb (Digest::self.const_missing): Drop
+	  autoloads for sha2 classes in favor of handling in
+	  const_missing(), to work around a problem exposed on OS X.
+
+Wed Feb 14 21:19:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.ci (native_thread_create): adjust 4KB (page size)
+	  alignment.
+
+Wed Feb 14 21:12:36 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.ci (CHECK_ERR): call rb_bug()
+	  instead of printf() and exit().
+
+Wed Feb 14 16:48:56 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/date/format.rb (Date::Format::Bag::method_missing): need not
+	  to use instance variables corresponding each method; use Hash
+	  instead.
+
+Wed Feb 14 13:12:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (reg_operand): allow symbols to be operands for regular
+	  expression matches.
+
+	* string.c (Init_String): allow Symbol#===.
+
+	* lib/date/format.rb (Date::Format::Bag::to_hash): string
+	  added prefixes.
+
+Wed Feb 14 12:58:38 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (do_select): fix to iterate select().
+	  on cygwin/mswin32, iterate in unblocking region.
+
+	* thread.c (rb_thread_select): don't iterate on this function.
+	  (iterate in do_select).
+
+Wed Feb 14 11:39:18 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (set_unblock_function): fix function interface.
+
+Wed Feb 14 11:12:02 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_load.c, yarvcore.h: use rb_vm_t#loaded_features instead of
+	  rb_features (global variable).
+
+	* yarvcore.c: mark rb_vm_t#loaded_features.
+
+Wed Feb 14 08:46:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compile.c (defined_expr): no longer distinguish ordinary local
+	  variables and in-block local variables in defined? value.
+
+Wed Feb 14 03:14:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic::userinfo): should support
+	  empty password.  [ruby-core:10290]
+
+	* lib/uri/generic.rb (URI::Generic::set_password): password can be
+	  cleared by nil.  [ruby-core:10290]
+
+Wed Feb 14 03:10:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (struct local_vars): no need to warn out-of-scope
+	  variables; remove dnames member from struct.
+
+Wed Feb 14 03:04:10 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: RVarmap no longer used as yytype; removed.
+
+	* parse.y (dyna_push_gen): no longer need return value.
+
+	* parse.y (dyna_pop_gen): no longer need argument.
+
+	* parse.y (local_push_gen): initialize nofree.
+
+Wed Feb 14 00:30:07 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (init_unixsock): path may contain NUL for
+	  abstract unix sockets.  [ruby-core:10288]
+
+Tue Feb 13 02:21:12 2007  Sam Roberts  <sroberts@uniserve.com>
+
+	* io.c (rb_f_syscall): Fix buffer overflow with syscall
+	  arguments.  [ruby-bugs:PR#8541]
+
+Mon Feb 12 13:57:30 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_variant2val): support VT_I8, VT_UI8.
+
+Mon Feb 12 11:48:52 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/dublincore.rb, lib/rss/maker/dublincore.rb: dc_rightses
+	  -> dc_rights_list. dc_rightses still exists for backward
+	  compatibility. [ruby-core:8350]
+
+	* test/rss/test_maker_dc.rb: added tests for dc_rights_list.
+
+Sun Feb 11 22:40:17 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_ptrtype2val): ole_type, ole_type_detail
+	  should not return "VARIANT,VARIANT".
+
+Sun Feb 11 22:11:05 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/xml-stylesheet.rb (RSS::XMLStyleSheet#initialize):
+	  kept backward compatibility.
+
+Sun Feb 11 22:10:08 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb (RSS::ListenerMixin#start_else_element): used
+	  const_defined? instead of constants.include?.
+
+Sun Feb 11 18:47:14 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_val2olevariantdata):
+	  WIN32OLE_VARIANT#new accepts nil when variant type is VT_ARRAY.
+
+	* test/win32ole/test_win32ole_variant.rb: ditto.
+
+	* ext/win32ole/win32ole.c: small refactoring.
+
+Sun Feb 11 07:42:25 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::read_multipart): Properly parse
+	  a quoted-string in a Content-Disposition value.
+
+Sat Feb 10 20:21:29 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE#ole_query_interface.
+	  thanks to Mikael Pahmp.
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+Sat Feb 10 17:46:52 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: targets which depend on yarvcore.h now depend on
+	  rubysig.h too.
+
+	* yarvcore.h (rb_vm_t): use rb_atomic_t instead of int.
+
+Sat Feb 10 00:13:11 2007  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fix typo (TkConfigMethod::__confinfo_cmd,
+	  __conv_keyonly_opts).
+
+Fri Feb  9 12:33:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_win32.ci (w32_show_error_message): renamed to w32_error.
+	  this function do rb_bug().
+
+	* thread_win32.ci (w32_set_event, w32_reset_event, w32_close_handle,
+	  w32_resume_thread): added. fix to use these functions instead calling
+	  win32api directly.
+
+	* thread_win32.ci (w32_create_thread): create suspend thread
+	  (caller must call w32_resume_thread()).
+
+Fri Feb  9 11:03:40 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_readpartial.rb: tests are working on mswin32/cygwin.
+
+Fri Feb  9 05:08:17 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, thread_pthread.ci, thread_win32.ci (thread_start_func_1):
+	  move cleanup function to thread_start_func_2().
+
+	* thread.c, thread_pthread.ci, thread_win32.ci:
+	  add more destruct functions.
+	  (native_thread_destroy() and native_mutex_destroy())
+
+	* thread_pthread.ci, thread_pthread.h: make native_mutex_* functions
+	  (check error, etc), it's not macro any more.
+
+	* thread_win32.ci (thread_start_func_1): store some values before
+	  running thread (to release these after running thread).
+
+	* thread_win32.ci (native_thread_create): fix spaces.
+
+Thu Feb  8 22:44:04 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_set_safe_array, ole_variant2val,
+	  ole_val_ary2variant_ary): fix WIN32OLE_VARIANT.new bug when
+	  1st argument is empty array, and when 2nd argument is
+	  VT_ARRAY|VT_BYREF.
+
+	* test/win32ole/test_win32ole_variant.rb: ditto.
+
+Thu Feb  8 22:39:09 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvtest/yarvtest.rb: check target command names.
+
+Thu Feb  8 22:31:45 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_clone.rb: fix to current spec
+	  (Module should not be occur many times in ancestors).
+
+Thu Feb  8 22:26:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_string.rb: ("foo" == :foo) and ("foo" === :foo)
+	  should be false.
+
+	* ChangeLog: fix last messages.
+
+Thu Feb  8 22:24:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_module.rb: fix to use Symbol instead of String.
+
+	* test/ruby/test_module.rb: remove space before argument parentheses.
+
+Thu Feb  8 22:02:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/marshaltestlib.rb: eval(sym) -> eval(sym.to_s)
+
+Thu Feb  8 21:35:16 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_propertyputref.rb (setup): fix typo.
+
+	* test/win32ole/test_win32ole_event.rb: should not use
+	  InternetExplorer.gohome to test.
+
+Thu Feb  8 21:02:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (GVL_UNLOCK_RANGE): rename to BLOCKING_REGION().
+
+	* thread.c (rb_thread_run_parallel): rename to
+	  rb_thread_blocking_region().
+
+Thu Feb  8 15:48:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, thread.c: fix to use pthread on cygwin.
+
+	* yarvcore.h, thread.c: move GVL_UNLOCK_BEGIN() and GVL_UNLOCK_END()
+	  from yarvcore.h to thread.c.
+
+	* thread.c: change GVL_UNLOCK_RANGE() arguments
+	  (adding ubf as 2nd argument).
+
+	* thread.c: fix to use polling in select on cygwin and mswin32.
+
+	* thread.c, thread_pthread.ci, thread_win32.ci, yarvcore.h:
+	  rename:
+	  * rb_thread_t#interrupt_function -> unblock_function
+	  * rb_interrupt_function_t -> rb_unblock_function
+	  * some interrupt function name -> ubf_*
+	  * yarv_* -> *
+
+Thu Feb  8 16:08:02 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: fix to use RUNRUBY instead of BASERUBY if possible.
+
+	* common.mk ($(INSNS) rule): remove $(PROGRAM) first.
+
+Thu Feb  8 15:43:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* process.c: fix to use rb_status_line_set/get/clear().
+
+	* eval_intern.h: fix line break.
+
+Thu Feb  8 15:00:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c, error.c, eval.c, eval_error.h, eval_intern.h,
+	  eval_jump.h, eval_load.c, eval_safe.h, gc.c, proc.c, signal.c,
+	  thread.c, thread_pthread.ci, thread_win32.ci, vm.c, vm.h,
+	  vm_dump.c, vm_evalbody.ci, yarvcore.c, yarvcore.h:
+	  fix typo (rb_thead_t -> rb_thread_t).
+
+	* eval_intern.h: remove unused definitions.
+
+	* common.mk: fix around vm_opts.h path
+	  and remove harmful argument passed to insns2vm.rb.
+
+Thu Feb  8 03:11:47 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::unescapeHTML): invalid decoding for single
+	  unescaped ampersand.  a patch from Tietew
+	  <tietew+ruby-dev at tietew.net> in [ruby-dev:30292].
+	  fixed: [ruby-dev:30289]
+
+Wed Feb  7 23:25:31 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (specific_eval): suppress warning.
+
+	* thread_win32.h: undefine _WIN32 on cygwin.  [ruby-dev:30303]
+
+Wed Feb  7 22:41:34 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h: remove UNSUPPORTED() macro.
+
+	* thread.c: fix to define Continuation methods
+	  (they only do rb_notimplement()).
+
+Wed Feb  7 22:33:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h, yarvcore.h: remove unused macro definition.
+
+Wed Feb  7 22:30:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c: fixed to use ANSI function style.
+
+Wed Feb  7 09:35:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* this commit is a result of refactoring. only renaming functions,
+	  moving definitions place, add/remove prototypes, deleting
+	  unused variables and removing yarv.h.
+	  This commit doesn't change any behavior of ruby/vm.
+
+	* yarv.h, common.mk: remove yarv.h (contents are moved to yarvcore.h).
+
+	* error.c, eval_intern.h: include yarvcore.h instead yarv.h
+
+	* rename some functions:
+	  * debug.[ch]: debug_*() -> ruby_debug_*()
+	  * iseq.c: iseq_*() -> rb_iseq_*(), ruby_iseq_disasm()
+	  * iseq.c: node_name() -> ruby_node_name()
+	  * vm.c: yarv_check_redefinition_opt_method() ->
+	          rb_vm_check_redefinition_opt_method()
+
+	* some refactoring with checking -Wall.
+
+	* array.c: remove rb_ary_ptr() (unused) and remove unused
+	  local variables.
+
+	* object.c: add a prototype of rb_mod_module_exec().
+
+	* eval_intern.h (ruby_cref): set it inline.
+
+	* eval_load.c (rb_load), yarvcore.c: yarv_load() -> rb_load_internal().
+
+	* parse.y: add a prototype of rb_parse_in_eval() (in eval.c).
+
+	* process.c: add a prototype of rb_thread_stop_timer_thread() (in thread.c).
+
+	* thread.c: remove raw_gets() function (unused) and fix some format
+	  mismatch (format mismatches have remained yet. this is todo).
+
+	* thread.c (rb_thread_wait_fd_rw): fix typo on label name.
+
+	* thread_pthread.ci: comment out codes with USE_THREAD_CACHE.
+
+	* vm.c (rb_svar, rb_backref_get, rb_backref_get,
+	  rb_lastline_get, rb_lastline_set) : moved from yarvcore.c.
+
+	* vm.c (yarv_init_redefined_flag): add a prototype and rename
+	  yarv_opt_method_table to vm_opt_method_table.
+
+	* vm.c (rb_thread_eval): moved from yarvcore.c.
+
+	* yarvcore.c: remove unused global variables and fix to use nsdr().
+
+Wed Feb  7 03:39:32 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c, compile.c, compile.h, error.c, eval.c,
+	  eval_intern.h, eval_jump.h, eval_load.c, eval_method.h,
+	  eval_safe.h, gc.c, insnhelper.h, insns.def, iseq.c, proc.c,
+	  process.c, signal.c, thread.c, thread_pthread.ci, thread_win32.ci,
+	  vm.c, vm.h, vm_dump.c, vm_evalbody.ci, vm_macro.def,
+	  yarv.h, yarvcore.h, yarvcore.c: change type and macro names:
+	  * yarv_*_t -> rb_*_t
+	  * yarv_*_struct -> rb_*_struct
+	  * yarv_tag -> rb_vm_tag
+	  * YARV_* -> RUBY_VM_*
+
+	* proc.c, vm.c: move functions about env object creation
+	  from proc.c to vm.c.
+
+	* proc.c, yarvcore.c: fix rb_cVM initialization place.
+
+	* inits.c: change Init_ISeq() order (after Init_VM).
+
+	* ruby.h, proc.c: change declaration place of rb_cEnv
+	  from proc.c to ruby.c.
+
+Tue Feb  6 22:06:45 2007  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/{nkf.c,utf8tbl.c}:
+	  imported nkf 2007-01-28.
+	  * Fixed: can't decode MIME encode JIS string.
+	  * Fixed: Fullwidth-halfwidth conversion.
+	  * Support DoCoMo's and Softbank's EMOJI
+	  * Support CP932, CP5022x, eucJP-ms UDC
+	  * Support UTF-32 encoding
+	  * Support beyond BMP
+	  [ruby-dev:29700] [ruby-dev:29922] [ruby-dev:30144]
+
+Tue Feb  6 20:36:19 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb, lib/rss/parser.rb: followed current Ruby
+	  specification. [ruby-dev:30274]
+
+Tue Feb  6 20:29:44 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb, lib/rss/parser.rb: removed needless code for
+	  backward compatibility.
+
+Tue Feb  6 18:43:17 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb: moved fixes for EPIPE to the correct
+	  place. [ruby-core:10204]
+
+Tue Feb  6 16:38:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_opts.h: set properties:
+	  svn:keywords:  Author Date Id Revision
+	  svn:eol-style: native
+
+Tue Feb  6 15:55:46 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (ivar_i): need to support class local instance
+	  variables.
+
+Tue Feb  6 15:44:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): fix setting is_local flag.
+
+	* yarvtest/test_class.rb: add a test for class local instance variable.
+
+Tue Feb  6 14:15:34 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def: remove (get|set)instancevariable2 and add a
+	  operand is_local to (get|set)instancevariable.
+
+	* yarvtest/test_class.rb: add a test for class local instance variable.
+
+	* parse.y (rb_decompose_ivar2): remove unused variable oid.
+
+	* tool/insns2vm.rb: remove needless require.
+
+Tue Feb  6 11:18:41 2007  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb: check the control connection on EPIPE.
+	  Thanks, Simon Williams. [ruby-core:9547]
+
+Tue Feb  6 11:03:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* complement last commit.
+
+	* common.mk (*.inc): use VPATH.
+
+	* vm_opts.h: renamed from vm_opts.h.base.
+
+Tue Feb  6 10:02:41 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/insns2vm.rb: use vm_opts.h in VPATH.
+
+Tue Feb  6 03:47:58 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: support Binding#eval.
+
+	* yarvtest/test_eval.rb: add a test for above change.
+
+Tue Feb  6 03:13:33 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: refactoring (remove K&R style, move Binding stuffs from
+	  Init_Proc() to Init_Binding()).
+
+Tue Feb  6 01:07:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* intern.h: prepare rb_last_status_get() and rb_last_status_set().
+	  Use these functions instead of rb_last_status ([ruby-dev:30264]).
+
+	* process.c: define above functions.
+
+	* ext/pty/pty.c: use above functions.
+
+	* io.c (pipe_finalize): ditto.
+
+	* process.c: ditto.
+
+Mon Feb  5 21:26:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.h: add a prototype of rb_id2str().
+
+Mon Feb  5 21:06:50 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_thread.c, common.mk: remove eval_thread.c.
+
+	* yarvcore.c: rename cYarvThread to rb_cThread.
+
+	* gc.c: remove YARV_* prefix.
+
+	* gc.h: add an include guard and prototype of rb_gc_set_stack_end().
+
+	* inits.c: fix to ANSI prototype style and reorder Init_*().
+
+	* io.c (pipe_finalize): TODO: comment out last_status.
+
+	* process.c, yarvcore.h: fix to use yarv_vm_t#last_status instead of
+	  rb_last_status and make last_status_get() to access $?.
+
+	* yarvcore.c (vm_mark): mark yarv_vm_t#last_status.
+
+	* ruby.h: add declarations of rb_cISeq and rb_cVM.
+
+	* thread.c: move eval_thread.c codes to thread.c and remove yarv_*
+	  function prefix.
+
+	* thread.c (thread_start_func_2): use yarv_thread_t#first_func if
+	  it is not null.
+
+	* vm.c: fix copyright year.
+
+	* yarvcore.c (Init_vm): rename to Init_VM().
+
+Mon Feb  5 04:09:48 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_frame_callee): check if prev_cfp can be accessible.
+	  a patch from Yoshinori Sano <yoshinori.sano at gmail.com> in
+	  [ruby-dev:30252].  solves [ruby-dev:30200] and [ruby-core:9856].
+
+Sun Feb  4 20:34:41 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/rss-assertions.rb: removed needless code for backward
+	  compatibility.
+
+Sun Feb  4 02:22:59 2007  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::read_multipart): Remove a debug
+	  print.
+
+Sat Feb  3 23:51:58 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (rb_compose_ivar2): function to create a new ivar2
+	  symbol from a symbol and a class.  back-ported from matzruby.
+
+	* parse.y (rb_decompose_ivar2): reverse function of
+	  rb_compose_ivar2().
+
+	* marshal.c (w_symbol): support class local instance variables.
+
+	* marshal.c (r_object0): ditto.
+
+	* compile.c (defined_expr): ditto.
+
+	* compile.c (iseq_compile_each): ditto.
+
+	* insns.def: add two new instructions: getinstancevariable2 and
+	  setinstancevariable2.
+
+Sat Feb  3 23:21:13 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (setclassvariable): remove unnecessary operand.
+
+	* compile.c (iseq_compile_each): ditto.
+
+	* common.mk (insns_info.inc): add dependency for insns_info.inc.
+
+Sat Feb  3 14:32:58 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_val2olevariantdata, ole_val2variant):
+	  fix the bug of WIN32OLE_VARIANT.new when variant type is
+	  VT_ARRAY.
+
+	* ext/win32ole/sample/excel1.rb: rewrite using WIN32OLE_VARIANT.
+
+	* test/win32ole/test_win32ole.rb: add some test.
+
+	* test/win32ole/test_win32ole_variant.rb: ditto.
+
+Sat Feb  3 03:35:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, */Makefile.sub, common.mk (vmasm): generalized.
+
+	* common.mk (runruby, benchmark, benchmark-each, tbench): use
+	  PROGRAM for the file to be built.
+
+	* proc.c (yarv_proc_alloc): needs return.
+
+	* call_cfunc.ci, compile.c, compile.h, debug.h, eval.c,
+	  eval_error.h, eval_jump.h, eval_load.c, eval_thread.c, gc.c,
+	  insnhelper.h, insns.def, iseq.c, main.c, numeric.c, parse.y,
+	  range.c, ruby.h, signal.c, thread.c, thread_win32.ci, vm.c,
+	  vm.h, vm_dump.c, vm_evalbody.ci, yarvcore.c, yarvcore.h:
+	  fixed indents and non-C90 comments.
+
+	* regenc.h: revert to before YARV.
+
+	* lib/mkmf.rb (create_makefile): make object files depend on
+	  extconf.h even if depend file exists.
+
+Fri Feb  2 23:39:42 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* common.mk (bin): add more dependency.  a patch from Tadashi
+	  Saito <shiba at mail2.accsnet.ne.jp>.  [ruby-dev:30245]
+
+Fri Feb  2 18:44:31 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* insns.def (setclassvariable): remove warn argument.
+
+Fri Feb  2 18:36:40 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* common.mk (compile.$(OBJEXT)): add config.h to vm.c and
+	  compile.c.
+
+Fri Feb  2 18:27:54 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c: remove duplicated global variables rb_cProc and
+	  rb_cBinding.  [ruby-dev:30242]
+
+Fri Feb  2 00:13:44 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (SYMBOL_P): make Symbol immediate again for performance.
+
+	* string.c: redesign symbol methods.
+
+Thu Feb  1 23:25:21 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_id2str): store Strings for operator symbols.
+	  [ruby-dev:30235]
+
+Thu Feb  1 21:04:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (assignable_gen): no need to generate NODE_CVDECL.
+
+	* compile.c (iseq_compile_each): no NODE_CVDECL.
+
+Thu Feb  1 20:53:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm.c (eval_get_cvar_base): destination for class variable access
+	  is now strictly innermost surrounding class or module.  warned
+	  if accessed from toplevel.
+
+	* variable.c (rb_cvar_get): new class variable look-up scheme:
+	  1) look up in the class.  2) if the class is singleton attached
+	  to a class (i.e. metaclass) then start look up in the attached
+	  class and its ancestors. 3) otherwise, look-up in ancestors of
+	  the class.
+
+	* eval.c (cvar_cbase): destination for class variable access is
+	  the class/module that holds the method, or cbase outside of
+	  methods.
+
+Thu Feb  1 20:31:41 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_cvar_set): remove warn argument.
+
+Wed Jan 31 14:52:09 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_iterator.rb (TestIterator::test_block_given_within_iterator):
+	  add new test.  [ruby-core:10125]
+
+Tue Jan 30 17:01:21 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_sub_bang): calling rb_str_modify() should be just
+	  before actually modifying the string.
+	  fixed: [ruby-dev:30211] (originally reported by zunda)
+
+Tue Jan 30 13:24:06 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_pred): add Integer#pred corresponding
+	  Integer#succ.  [RCR#5]
+
+Tue Jan 30 12:05:35 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: autoconf 2.61 support.  [ruby-core:10016]
+
+Mon Jan 29 23:52:32 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* tool/compile.rb: replace YARVCore by VM class.
+
+Mon Jan 29 17:52:44 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/date/format.rb (Date::Format::Bag::method_missing): add
+	  prefix to avoid making t class-local instance variable.
+
+Mon Jan 29 21:32:37 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE.locale=, WIN32OLE.locale,
+	  WIN32OLE_VARIANT#vartype.
+
+	* test/win32ole/test_win32ole.rb: add test for WIN32OLE.locale=,
+	  WIN32OLE.locale.
+
+	* test/win32ole/test_win32ole_variant.rb: add test for
+	  WIN32OLE_VARIANT#vartype.
+
+Mon Jan 29 14:14:35 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* tool/parse.rb: replace YARVCore by VM class.
+	  http://d.hatena.ne.jp/ysano2005/20070128
+
+Sun Jan 28 08:41:49 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: refactoring.
+
+Sat Jan 27 18:36:33 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_val2olevariantdata): bug fix.
+	  WIN32OLE_VARIANT.new check that 1st argument should T_ARRAY
+	  when variant type is VT_ARRAY.
+
+	* test/win32ole/test_win32ole_variant.rb: add some test.
+
+Fri Jan 26 23:55:56 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: bug fix of WIN32OLE_VARIANT when variant
+	  type is VT_BYREF|VT_VARIANT.
+
+	* test/win32ole/test_win32ole_variant_with_ie.rb: ditto.
+
+Fri Jan 26 12:03:39 2007  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb (TkConfigMethod#__confinfo_cmd,
+	  __conv_keyonly_optkeys): make them private [ruby-dev:30074].
+
+	* ext/tk/lib/tk/txtwin_abst.rb: fix typo [ruby-dev:30073].
+
+	* ext/tk/lib/tk/canvas.rb (TkCanvas#scan_dragto): lack of an argument.
+
+	* ext/tk/lib/tk/canvas.rb: clarify the including module name
+	  [ruby-dev:30080].
+
+	* ext/tk/lib/tk/scrollable.rb: change primary name of modules
+	  [ruby-dev:30080].
+
+Fri Jan 26 07:48:57 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_init_copy): need to copy internal
+	  structure on clone and dup.  [ruby-dev:30192]
+
+Wed Jan 24 20:34:51 2007  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/ruby/test_iterator.rb: removed a needless workaround.
+
+Wed Jan 24 18:05:39 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-font-lock-syntactic-keywords): fix
+	  regexp font-lock bug.  [ruby-talk:235758]
+
+Tue Jan 23 18:26:12 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::read_multipart): use == instead
+	  of ===.  [ruby-dev:30176]
+
+Tue Jan 23 15:39:25 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* Makefile.in, common.mk, configure.in, */{Makefile.sub, configure.bat,
+	  setup.mak}: add --with-baseruby configure option.
+
+Mon Jan 22 14:57:25 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c: fix errors in socket sample code.
+	  [ruby-core:09992]
+
+Sat Jan 20 21:05:18 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fole_s_set_code_page): WIN32OLE.codepage=
+	  accepts installed codepage.
+
+	* test/win32ole/test_win32ole.rb (test_s_codepage_changed): ditto.
+
+Sat Jan 20 11:18:49 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke, ole_invoke2, ole_propertyput):
+	  modify WIN32OLERuntimeError message.
+
+	* test/win32ole/test_win32ole.rb: ditto.
+
+Sat Jan 20 06:45:21 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval_proc.c (method_receiver): add new method to get the bound
+	  receiver of the method object.  [ruby-talk:234949]
+
+	* eval_proc.c (method_name): new method to get the name of a
+	  method.
+
+	* eval_proc.c (method_owner): a new method to get the class or
+	  module that defines the method.
+
+Fri Jan 19 17:12:23 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (Init_win32ole): add WIN32OLE_VARIANT::Empty,
+	  WIN32OLE_VARIANT::Null, WIN32OLE_VARIANT::Nothing.
+
+	* test/win32ole/test_win32ole_variant.rb: ditto.
+
+	* test/win32ole/test_nil2vtempty.rb(test_openSchema): ditto.
+
+Fri Jan 19 06:53:38 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (olevariant_free): fix memory leak.
+
+	* ext/win32ole/win32ole.c (ole_val2olevariantdata):
+	  WIN32OLE_VARIANT.new accepts nil as first argument for some VARIANT
+	  TYPE.
+
+	* test/win32ole/test_win32ole_variant.rb: ditto.
+
+Wed Jan 17 17:31:28 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* some refactoring around yarvcore and proc.
+
+	* eval_proc.c: renamed to proc.c.
+
+	* common.mk: ditto.
+
+	* yarvcore.h, yarvcore.c: rename or remove some global variables
+	  removed: mYarvCore, mYarvInsns
+	  renamed: cYarvISeq -> rb_cISeq,
+	           cYarvProc -> rb_cProc, cYarvBinding -> rb_cBinding
+	  ::YarvCore module is removed and ::YarvCore::VM class becomes ::VM.
+	  And change/remove some functions which added with YARV.
+
+	* compile.c: ditto.
+
+	* eval.c: ditto.
+
+	* iseq.c: ditto.
+
+	* vm.c: ditto.
+
+	* inits.c: rename Init_yarvcore to Init_vm.
+
+	* yarvcore.c, proc.c: move some functions and initialization
+	  from yarvcore.c to proc.c.
+
+	* intern.h, proc.c: add global function rb_binding_new(void).
+
+Tue Jan 16 17:49:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (eval_search_super_klass): rename to search_super_klass() and
+	  use it by th_call_super().
+
+	* insns.def: ditto.
+
+Tue Jan 16 17:48:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: fix ruby script path.
+
+Tue Jan 16 17:39:44 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (invoke_block): fix to specify self.
+
+Tue Jan 16 12:12:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* env.h: removed
+
+Tue Jan 16 12:00:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_proc.c (rb_proc_new): added.
+
+	* string.c (sym_to_proc): supported.
+
+	* vm.c (invoke_block, th_invoke_yield, th_invoke_proc): fix to support
+	  rb_proc_new.
+
+	* yarvcore.c: add a test code.
+
+Sat Jan 13 23:24:59 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_free, ole_type_free,
+	  olemethod_free, olevariable_free, oleparam_free,
+	  ole_event_free): fix memory leak.  [ruby-core:09846]
+
+Wed Jan 10 00:10:23 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_replace): use ptr and len of orig instead of
+	  shared.  fixed: [ruby-dev:30116]
+
+Tue Jan  9 17:48:38 2007  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_find_file): should not call fpath_check() with NULL.
+	  fixed: [ruby-core:09867]
+
+Tue Jan  9 12:29:20 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_getpwuid, etc_getgrgid): fix to correctly
+	  convert uid/gid from VALUE.
+
+	* ext/etc/etc.c (etc_getpwuid): ditto.
+
+Tue Jan  9 03:54:38 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_upto): String#upto from empty string makes
+	  infinite loop.  [ruby-core:09864]
+
+	* string.c (rb_str_upto): use RSTRING_LEN().
+
+Sun Jan  7 18:36:05 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_stop_timer_thread(), rb_thread_reset_timer_thread(),
+	  rb_thread_start_timer_thread()): added.
+
+	* thread_pthread.ci: add a native_thread_join() and move
+	  rb_thread_reset_timer_thread() definition to thread.c.
+
+	* thread_win32.ci: ditto
+
+	* process.c: fix before_exec(), after_exec() to stop timer thread
+	  (and restart timer thread if exec failed).  and fix to reset
+	  timer thread information when forked child process starts
+	  (to fix [ruby-core:09822]).
+
+Sun Jan  7 18:28:17 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: add a "compare" rule and fix MATZRUBY variable
+
+Sun Jan  7 17:47:16 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole.rb: add test for WIN32OLE#[],
+	  WIN32OLE#[]=.
+
+	* ext/win32ole/win32ole.c: update comment for rdoc of
+	  WIN32OLE#[] and WIN32OLE#[]=.
+
+Sun Jan  7 12:13:26 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#find_class_comment):
+	  Look for class and module comments above rb_define_class and
+	  rb_define_module.  Patch by Daniel Berger <djberg96 at gmail.com>
+
+Sun Jan  7 10:32:12 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#handle_constants):
+	  Properly handle escaping of : in comments.
+	* test/rdoc/parsers/test_parse_c.rb:
+	  Test RDoc::C_Parser#do_classes and Rdoc::C_Parser#find_class_comment.
+
+Sun Jan  7 09:31:18 2007  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: updated based on date2 4.0.1.
+
+Sat Jan  6 18:46:34 2007  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_getline_fast, rb_io_getline): increase lineno
+	  when met the delimiter or EOF.  fixed: [ruby-dev:30081]
+
+	* io.c (prepare_getline_args, rb_io_getline_1): split
+	  preparation of arguments and reading.  [ruby-dev:30085]
+
+Sat Jan  6 13:48:36 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (send): fix to optimize send() with Symbol.
+
+	* yarvtest/test_method.rb: add another test.
+
+Sat Jan  6 13:43:55 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: add PHONY dependency to some rules
+
+Sat Jan  6 11:50:33 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_read): fix wrong replacement.  [ruby-dev:30070]
+
+Sat Jan  6 09:10:52 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: support direct method dispatch with "send" or "funcall".
+	  This means that "obj.send :m" skips "BasicObject#send" invocation
+	  (method frame creation, etc) and "obj.m" invokes directly.
+	  If you make backtrace, there are no entries of "send" method.
+
+	* compile.c (iseq_specialized_instruction): fix to support above
+
+	* eval.c: ditto (remove "static" from rb_f_send and rb_f_funcall
+
+	* yarvcore.c: ditto (add a external IDs for compiler)
+
+	* yarvcore.h: ditto (add a VM_CALL_SEND_BIT macro)
+
+	* yarvtest/test_method.rb: add tests for above changes
+
+	* eval.c: remove unused "Kernel#send" declaration
+
+Sat Jan  6 08:29:17 2007  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (Init_win32ole): add
+	  WIN32OLE::VARIANT::VT_EMPTY, WIN32OLE::VARIANT::VT_NULL
+
+	* test/win32ole/test_win32ole_variant_m.rb (test_variant): ditto.
+
+Sat Jan  6 03:49:22 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb: change option format
+
+	* common.mk: ditto
+
+Fri Jan  5 22:21:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_pentomino.rb: use Array#dup instead of
+	  Array#clone
+
+	* benchmark/bmx_temp.rb: removed
+
+	* benchmark/run.rb: use run.rb instead of run_rite.rb
+
+	* common.mk: ditto
+
+	* benchmark/run_rite.rb: removed
+
+	* common.mk: use $(srcdir)/test.rb to run a test program
+	  with "make run"
+
+	* benchmark/bmx_temp.rb: removed and
+	  set svn:ignore (bmx_*.rb) to benchmark/
+
+	* test.rb: set svn:ignore
+
+Fri Jan  5 21:03:08 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvtest/yarvtest.rb: fix to compare results
+
+Fri Jan  5 20:52:56 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, compile.h: add ADD_CALL_RECEIVER() macro.
+
+	* insns.def (send): use GET_SELF() direct if FCALL.
+
+	* eval.c (rb_f_send): check method dispatch type to permit
+	  invoking private method when dispatch type is FCALL/VCALL
+
+	* insns.def (opt_ltlt): remove useless statement.
+
+	* vm.h: remove unused macros.
+
+Fri Jan  5 20:50:31 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run_rite.rb: fix to use readlines instead of
+	  read(...).lines (because 1.8 doesn't have String#lines).
+
+Fri Jan  5 20:28:19 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_win32.ci (rb_thread_reset_timer_thread):
+	  added ([ruby-dev:30086]).
+
+Fri Jan  5 20:20:36 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: add .SUFFIXES rule
+
+Fri Jan  5 15:58:15 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.h (rb_alias): fix to check search result
+
+Fri Jan  5 13:59:53 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.h (rb_add_method): fix to check old_node
+
+Fri Jan  5 12:03:07 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each, set_block_local_tbl):
+	  support NODE_LAMBDA (partly).
+
+	* sample/test.rb: restore test of NODE_LAMBDA
+
+	* test/ruby/test_lambda.rb: ditto
+
+Fri Jan  5 12:31:23 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* thread_pthread.ci (native_sleep): fix tv_nsec overflow.
+
+Thu Jan  4 20:01:29 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: rename yarv-test-[all/each] to compare-test[/-each].
+	  purpose of "compare-test" rule is to compare ruby (trunk) and
+	  matzruby (branches/matzruby) binary in miniruby level.  MATZRUBY
+	  parameter means an path to miniruby of matzruby binary.  to do this
+	  comparison test, you should build matzruby branch.
+
+	* yarvtest/yarvtest.rb: fix to use command line option as
+	  command names to be compared.
+
+	* yarvtest/runner.rb: remove a debug output.
+
+Thu Jan  4 19:12:27 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: fix to use test.rb script in build directory.
+	  ($(srcdir)/test.rb -> test.rb)
+
+Thu Jan  4 17:28:05 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler#initialize):
+	  Arrays could not be modified in its each block. [ruby-dev:30063]
+
+Thu Jan  4 16:57:14 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* yarv_version.h: removed.
+
+	* common.mk: remove yarv_version.h from rules
+
+	* yarvcore.h (Init_yarvcore): remove useless constants
+
+Thu Jan  4 17:00:06 2007  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (Init_ossl_asn1):
+	  OpenSSL::ASN1::ASN1Data#value,#tag,#tag_class and
+	  OpenSSL::ASN1::BitString#unused_bits should be public.
+
+Thu Jan  4 13:45:10 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.ci: fix last changes around PTHREAD_STACK_MIN.
+
+Thu Jan  4 13:42:47 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: restore changes.
+
+Thu Jan  4 10:33:54 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.ci: fix to skip using PTHREAD_STACK_MIN.
+	  [ruby-dev:30063]
+
+Thu Jan  4 10:30:11 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run_rite.rb (bm): fix to use lines.
+
+Wed Jan  3 18:49:15 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_getline): lineno update condition was wrong.
+	  [ruby-dev:30065]
+
+	* io.c (rb_io_getline_fast): ditto.
+
+Wed Jan  3 11:36:51 2007  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (ruby_dup): start GC on ENOMEM as well.
+
+Tue Jan  2 10:29:54 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c: fix to compile on YARV
+	  ruby_errinfo -> rb_errinfo()
+
+Mon Jan  1 08:07:06 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/tk/tcltklib.c: fix to compile on YARV
+	  ruby_errinfo -> rb_errinfo(),
+	  ruby_safe_level -> rb_safe_level().
+
+Mon Jan  1 07:57:17 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* test/drb/test_drbssl.rb: fix to skip drb tests.
+
+Mon Jan  1 06:13:11 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/c_parser.rb: Make Rdoc accessible.  Update constant
+	  value information.
+
+	* ext/bigdecimal/bigdecimal.c: Update constant comments to provide
+	  values for RDoc.
+
+Mon Jan  1 06:05:55 2007  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#handle_constants):
+	  Allow RDoc comment to give friendly value for rb_define_const.  Patch
+	  by Daniel Berger <djberg96 at gmail.com>, [ruby-patches-7499].
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#handle_constants): Fix
+	  whitespace handling in constant comments.
+
+Mon Jan  1 00:00:00 2007  Koichi Sasada  <ko1@atdot.net>
+
+	* Merge YARV
+
+Sun Dec 31 16:22:48 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c: Fix Array#reject.
+
+Sun Dec 31 00:46:25 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date2.rb: removed.
+
+Sun Dec 31 00:15:13 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: updated based on date2 4.0.
+
+Sat Dec 30 04:38:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_each_with_index): reuse array for yield parameters.
+
+	* enum.c (enum_min, enum_max): ditto.
+
+Sat Dec 30 04:25:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_inject): reuse array for yield parameters.
+
+Sat Dec 30 02:54:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_gets): accepts limit argument.
+
+	* ext/stringio/stringio.c (strio_readline, strio_each,
+	  strio_readlines): ditto.
+
+Sat Dec 30 02:22:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_getline): add limit capability.
+
+	* io.c (rb_io_gets_m): accepts limit argument.  [ruby-talk:231563]
+
+	* io.c (rb_io_readline, rb_io_readlines, rb_io_each_line, argf_getline):
+	  ditto.
+
+	* io.c (appendline): add limit capability.
+
+	* io.c (rb_io_getline_fast, rb_io_getline): ditto.
+
+	* io.c (rb_io_getline): small refactoring for DRY.
+
+	* io.c (rb_io_s_foreach, rb_io_s_readlines): small refactoring.
+
+Thu Dec 28 15:27:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::Cookie::initialize): use Array() again.
+	  [ruby-core:09781]
+
+Wed Dec 27 20:52:32 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: remove WIN32OLE::PROPERTY class.
+
+	* ext/win32ole/tests/testWIN32OLE.rb: ditto.
+
+Wed Dec 27 10:04:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_Array): returns 1-element array if the argument
+	  does not have to_ary nor to_a.
+
+Tue Dec 26 21:02:14 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_folderitem2_invokeverb.rb: The argument
+	  of Shell.NameSpace should not be file path.
+
+Tue Dec 26 06:13:08 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/bigdecimal/bigdecimal.c: remove useless method
+	  BigDecimal#!=. [ruby-dev:30050]
+
+Thu Dec 21 15:37:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_slice_bang): rdoc description bug fixed.
+	  [ruby-core:09754]
+
+Wed Dec 20 12:54:31 2006  Koichi Sasada  <ko1@atdot.net>
+
+	* Convert CVS repository to Subversion repository.
+
+Mon Dec 18 08:36:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::Cookie::initialize): Array(string) no longer
+	  works.  [ruby-core:09738]
+
+Fri Dec 15 00:19:53 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (COMPILE_RULES): latter rule has higher priority.
+
+	* lib/mkmf.rb (create_makefile): remove static library before update,
+	  to get rid of sludge of Borland tlib.exe.
+
+Thu Dec 14 18:29:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/readline/readline.c: NetBSD editline does not have
+	  rl_username_completion_function() and rl_completion_matches().
+	  a patch from Takahiro Kambe <taca at back-street.net>.
+	  [ruby-dev:30008]
+
+Thu Dec 14 18:20:43 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/locale.rb (IRB::Locale::puts): typo fixed.  a patch from
+	  NAKAMURA Usaku <usa@ruby-lang.org>.  [ruby-dev:30012]
+
+Tue Dec 12 23:33:53 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (Switch#parse_arg, Switch#conv_arg): splat failures.
+
+Mon Dec 11 11:51:10 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/sha2/lib/sha2.rb: Moved one level up from under
+	  the superfluous subdirectory digest/.
+
+Mon Dec 11 11:46:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_define_const): typo fixed.
+
+Mon Dec 11 09:36:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_aset): index double decode problem.
+	  [ruby-core:09695]
+
+Sat Dec  9 21:39:24 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): keep the exception till after END blocks.
+	  [ruby-core:09675]
+
+Sat Dec  9 11:22:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/locale.rb (IRB::Locale::search_file): use File.exist?
+	  instead of File.exists?.  a patch from Yutaka Kanemoto
+	  <kinpoco at gmail.com> in [ruby-dev:30000].
+
+Fri Dec  8 18:11:18 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/optparse.rb: cannot put :nodoc: before method definition.
+	  put after it.
+
+Fri Dec  8 17:00:13 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bin/rdoc: use File.exist? instead of File.exists?.
+
+Thu Dec  7 23:50:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (Init_Object): new method Dir.exist?(path).
+	  [ruby-core:09663]
+
+	* file.c (Init_File): remove File.exists?; use File.exist?
+	  instead.
+
+	* file.c: rename functions to test_* to rb_file_*_p.
+
+Thu Dec  7 09:29:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/weakref.rb (WeakRef::__setobj__): should support
+	  marshaling.  [ruby-talk:228508]
+
+Wed Dec  6 23:58:36 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.mk (NULLCMD): moved for platforms that empty
+	  command does not run.  fixed: [ruby-dev:29994]
+
+	* win32/win32.c (init_stdhandle): redirect unopened IOs to NUL.
+	  [ruby-core:09572]
+
+Tue Dec  5 19:01:42 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (SITE_DIR): fixed to empty RUBY_SITE_LIB in config.h on
+	  NetBSD.  fixed: [ruby-dev:29358]
+
+Tue Dec  5 18:38:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/matrix.rb (Matrix::rank): use quo method to avoid integer
+	  division problem.  [ruby-core:09644]
+
+	* lib/matrix.rb (Matrix::rank_e): ditto.
+
+Tue Dec  5 00:59:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-parse-partial): need to parse "/=" as
+	  self assignment operator, not regex.  [ruby-talk:227324]
+
+Tue Dec  5 00:19:14 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h, object.c, variable.c (rb_mod_constants): added an optional
+	  flag to search ancestors, which is defaulted to true, as well as
+	  const_defined? and const_get.  [ruby-dev:29989]
+
+Mon Dec  4 23:49:28 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (install_recursive): get rid of warning.
+
+	* lib/optparse.rb (CompletingHash#match): get rid of splat failure.
+
+Mon Dec  4 19:16:39 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb: Do alias << update.
+
+Mon Dec  4 10:48:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (OFFT2NUM): use LONG2NUM() if sizeof(long) equals to
+	  sizeof(off_t).
+
+Mon Dec  4 08:32:25 2006  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::read_multipart): should quote
+	  boundary. JVN#84798830
+
+Sun Dec  3 16:16:53 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb: Fix problems with update
+	  timing. [Reported by: oss-ruby@technorama.net]
+
+Sat Dec  2 07:33:53 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils::FormData::<<):
+	  HTTPUtils::parse_header() takes a string.  [ruby-dev:29931]
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils::parse_header):
+	  String does no longer have each method.
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils::parse_form_data):
+	  ditto.
+
+Sat Dec  2 07:09:04 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ocsp.c: OpenSSL::OCSP::OSCPError should be
+	  subclass of OpenSSL::OpenSSLError. [ruby-dev:29980]
+
+Fri Dec  1 16:31:53 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: shouldn't run the killed thread at callback.
+	  [ruby-talk: 227408]
+
+Tue Nov 28 17:25:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (ary_iter_check): should check modification (size
+	  change) during iteration.
+
+	* array.c (rb_ary_initialize, rb_ary_shift, rb_ary_unshift,
+	  rb_ary_splice, rb_ary_reverse, rb_ary_sort, rb_ary_delete,
+	  rb_ary_delete_at, rb_ary_reject_bang, rb_ary_replace,
+	  rb_ary_clear, rb_ary_fill, rb_ary_uniq_bang, rb_ary_compact,
+	  rb_ary_shuffle): add iteration check.
+
+Mon Nov 27 09:00:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_ord): typo fixed.  reported from Kornelius
+	  Kalnbach <murphy@rubychan.de>.  [ruby-core:09621]
+
+Sun Nov 26 16:36:46 2006  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* version.h: addition of RUBY_PATCHLEVEL.
+	* version.c: ditto.
+
+Wed Nov 22 16:00:49 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: support --with-X11/--without-X11 option.
+
+	* ext/tk/README.tcltklib: add description about --with-X11-* option
+	  [ruby-talk:225166] and --with-X11/--without-X11 option.
+
+	* ext/tk/tkutil/extconf.rb: able to be called manually
+	  [ruby-talk:225950].
+
+Sat Nov 18 23:39:20 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_tap): a new method.  [ruby-talk:224013]
+
+Wed Nov 15 23:22:54 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (test_grpowned, rb_stat_grpowned): should honor
+	  supplementary group IDs.  [ruby-core:09546]
+
+Tue Nov  7 18:35:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (formal_assign): need to pack rest arg information in
+	  argc.
+
+Tue Nov  7 18:05:01 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/itemconfig.rb: minor bug fix.
+
+Tue Nov  7 17:52:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c (rb_include_module): revert duplicate inclusion of
+	  modules.  [ruby-dev:29793]
+
+Tue Nov  7 17:18:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (method_missing): update old argument adjustment.
+
+Tue Nov  7 16:41:21 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (when_check): need to splat for NODE_ARGSCAT as well.
+	  [ruby-dev:29860]
+
+Mon Nov  6 22:23:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (Init_String): remove duplicated definition of
+	  Symbol#to_s.
+
+Mon Nov  6 18:54:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (svalue_to_avalue): need to splat but no error.
+
+	* eval.c: new macros - YIELD_CALL, YIELD_VALUES.
+
+	* eval.c (rb_yield_values): specify YIELD_VALUES.
+
+	* eval.c (rb_yield_0): use new macros.
+
+	* eval.c (proc_invoke): slightly modified to separate YIELD_CALL
+	  and YIELD_VALUES from YIELD_ARY_ARGS.
+
+	* object.c (Init_Object): add nil.to_splat => [].
+
+Mon Nov  6 15:41:55 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/itemconfig.rb: ext/tk/lib/tk/itemconfig.rb: bug
+	  fix on 'itemconfiginfo' method, and modify to make it easy to
+	  override 'itemconfiginfo' method.
+
+	* ext/tk/lib/tkextlib/tile/treeview.rb: support Tile 0.7.8.
+
+	* ext/tk/lib/tkextlib/version.rb: [new] add Tk::Tkextlib_RELEASE_DATE
+	  to get the information from scripts.
+
+	* ext/tk/lib/tk.rb: load 'tkextlib/version.rb', and update RELEASE_DATE
+
+	* ext/tk/lib/tkextlib/SUPPORT_STATUS: update.
+
+	* ext/tk/sample/editable_listbox.rb: [new] the listbox with editable
+	  items. It's one of the example about usage of Place geometry manager.
+
+	* ext/tk/sample/tktextio.rb: improve the functions of TkTextIO class.
+	  Those are required by 'irbtkw.rbw'.
+
+	* ext/tk/sample/irbtkw.rbw: [new] IRB on Ruby/Tk. It doesn't need any
+	  real console. IRB works on a text widget without I/O blocking. That
+	  is, thread switching on IRB will work properly, even if on Windows.
+
+Mon Nov  6 00:42:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (arg_dup_check): vid may be nameless internal id.
+
+Sun Nov  5 19:52:19 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: updated based on date2 3.9.7.
+
+Sat Nov  4 13:09:31 2006  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: accept NOMODSEQ. [ruby-core:9002]
+
+Fri Nov  3 00:16:37 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_getnameinfo__aix): AF_INET6 workaround
+	  for AIX.  a patch from Yutaka Kanemoto <kinpoco AT gmail.com>.
+	  [ruby-dev:29744]
+
+Thu Nov  2 14:19:44 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#^): Fix XOR operation against a container that
+	  holds duplicate values. [ruby-core:9372]
+
+Thu Nov  2 10:00:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c: class Symbol is no longer subclass of String. also
+	  covers [ruby-core:09366]
+
+Thu Nov  2 08:21:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/xmlrpc/create.rb (XMLRPC::Create::conv2value): Symbol should
+	  come earlier than String.
+
+	* lib/soap/mapping/rubytypeFactory.rb (RubytypeFactory::obj2soap):
+	  ditto.
+
+	* lib/set.rb (TC_Set::test_s_new): strings are no longer
+	  Enumerable
+
+	* lib/soap/property.rb (Property::load): ditto.
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils::parse_header): ditto.
+
+	* lib/soap/mimemessage.rb (MIMEMessage::Headers::parse): ditto.
+
+Thu Nov  2 09:08:04 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c: revert lfree shift/unshift boost patch to avoid unknown
+	  memory error.
+
+Wed Nov  1 23:24:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (struct RArray): revert embedding ptr in RVALUE.
+
+	* array.c: ditto.
+
+Wed Nov  1 23:01:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (hash): use Bob Jenkins' hash algorithm.
+
+Wed Nov  1 02:22:31 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb (Digest::HMAC::update): Minor
+	  optimization.
+
+	* ext/digest/digest.c (rb_digest_instance_equal): Allow comparing
+	  a digest instance with another of a different class.
+
+Wed Nov  1 01:05:13 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (rb_call0): fixed bug of zsuper with both of opt and rest.
+	  fixed: [ruby-list:42928]
+
+Tue Oct 31 17:03:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_dup): duplicate the class of original time.
+	  [ruby-core:09357]
+
+	* lib/time.rb (Time::make_time, Time::rfc2822, Time::httpdate):
+	  should respect subclasses.  [ruby-core:09357]
+
+Tue Oct 31 16:25:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (ary_shared_first): should address offset after
+	  ary_shared_array().  [ruby-core:09358]
+
+Mon Oct 30 23:40:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (miniruby): add XLDFLAGS.
+
+	* configure.in (aix): use -bE option for miniruby.  [ruby-dev:29698]
+
+	* dir.c (glob_helper): get rid of possible memory leak.
+
+	* win32/win32.c (cmdglob, rb_w32_cmdvector, rb_w32_opendir,
+	  rb_w32_get_environ): not to use GC before initialization.
+
+Mon Oct 30 19:28:02 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): use better approximation.
+
+Mon Oct 30 18:35:33 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big2str0): wrong allocation length.  a patch from
+	  U.Nakamura <usa at garbagecollect.jp> [ruby-dev:29710]
+
+Mon Oct 30 12:34:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): fix commit miss.  [ruby-dev:29707]
+
+Mon Oct 30 11:15:40 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): should preserve leading zero
+	  information for negative %b and %x.  [ruby-talk:221347]
+
+Sun Oct 29 19:51:31 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* regexec.c: invalid offset value was used in STATE_CHECK_BUFF_INIT().
+
+Sat Oct 28 20:13:18 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* oniguruma.h: Version 4.4.5
+
+	* regint.h: ditto.
+
+	* regerror.c: ditto.
+
+	* regexec.c: ditto.
+
+	* regcomp.c ditto.
+
+	* regparse.c ditto.
+
+Sat Oct 28 07:56:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_object0): missing break.  [ruby-core:09345]
+
+Fri Oct 27 17:30:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enum_each_cons): move RETURN_ENUMERATOR() after
+	  argument check.
+
+Thu Oct 26 21:05:48 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_verify): should clear error.
+	  (fix http://bugs.debian.org/394336)
+
+	* ext/openssl/ossl_ns_spki.c (ossl_spki_initialize): ditto.
+
+Thu Oct 26 15:23:47 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c: remove by_slice and by_cons.
+
+Thu Oct 26 15:12:12 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/digest/digest.c (Init_digest): typo.
+
+Wed Oct 25 17:16:05 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/digest/test_digest_hmac.rb: added.
+
+Wed Oct 25 16:34:31 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/test.sh: make this script work again.
+
+Wed Oct 25 07:59:42 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: updated based on date2 3.9.6.
+	  [ruby-core:09323]
+
+Wed Oct 25 00:58:19 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/mkexports.rb, win32/resource.rb: use unique variable names.
+
+Tue Oct 24 19:18:53 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_by_slice): new method added.
+
+	* enumerator.c (enumerator_by_cons): ditto.
+
+Tue Oct 24 18:56:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enum_each_slice, enum_each_cons): returns
+	  Enumerable::Enumerator if no block is given.  [ruby-dev:29246]
+
+	* enumerator.c: remove methods: enum_with_index, enum_slice,
+	  enum_cons.  [ruby-dev:29246]
+
+Tue Oct 24 18:51:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_zip): add RETURN_ENUMERATOR() to zip method.
+
+Mon Oct 23 04:30:04 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_object0): use return value from proc given as the
+	  second argument to Marshal#load() to allow value replacement in
+	  the restoring data.
+
+Sun Oct 22 14:48:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (Init_signal): avoid duplicated installation of SIGCHLD
+	  handler.
+
+Sun Oct 22 16:47:56 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_substr): should be infected with only original
+	  string, but not the shared string.  fixed: [ruby-core:09152]
+
+	* string.c (rb_str_new4): keep shared string untainted when original
+	  string is tainted.  fixed: [ruby-dev:29672]
+
+Sun Oct 22 07:55:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_upcase, rb_str_downcase, rb_str_downcase,
+	  rb_str_upcase_bang, rb_str_downcase_bang, rb_str_swapcase_bang):
+	  add RDoc description that case conversion to be effective only
+	  in ASCII region.
+
+Sun Oct 22 05:20:34 2006  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
+
+	* configure.in: alloca is broken; use C_ALLOCA instead.
+	  [ruby-dev:29416]
+
+Sat Oct 21 17:50:40 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest.rb: Follow the framework updates.
+
+Fri Oct 20 22:00:43 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb: Complete half-boiled updates.
+
+	* ext/digest/sha2/lib/digest/sha2.rb: Fix #initialize_clone().
+
+Fri Oct 20 20:28:37 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest: Prefix C constants with RUBY_ and C type names with
+	  rb_ to avoid name clash in writing extensions.
+
+	* ext/digest: Introduce Digest::Class and Digest::Instance for
+	  ease of implementing subclasses and add-ons, inspired by
+	  gotoyuzo.
+
+	* ext/digest: The Digest::Instance module now requires and assumes
+	  that any instance be resettable and clonable, and add some
+	  convenient instance methods such as "new()", for creating a new
+	  copy, parameter taking "digest()" and "hexdigest()", for instant
+	  calculation.  These methods make digest instances work just like
+	  digest classes.
+
+	* ext/digest/sha2/lib/digest/sha2.rb:
+	  Add the Digest::SHA2 class to wrap up SHA2 variants: SHA256,
+	  SHA384 and SHA512, hoping this module would make a decent
+	  example of a digest subclass written in Ruby.
+
+	* ext/digest/lib/digest.rb: Adjust autoload entries for SHA2
+	  classes.
+
+	* ext/digest/lib/digest/hmac.rb: Follow the framework updates.
+
+Fri Oct 20 10:47:43 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb: fixed the bug of handling COMMON_MACROS.
+
+Fri Oct 20 08:42:38 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (NULLCMD): dummy command.
+
+	* bcc32/Makefile.sub (post-install-*): Borland make cannot ignore
+	  command-less double-colon rules.  [ruby-dev:29676]
+
+Fri Oct 20 00:37:07 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bcc32/Makefile.sub ($(LIBRUBY_SO)): execute pre-link hook.
+
+	* ext/extmk.rb: workaround for Borland make.
+
+Wed Oct 18 23:02:40 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_shift): shorten copy size.  fixed: [ruby-list:42907]
+
+	* signal.c (Init_signal): handle SIGTERM.  fixed: [ruby-list:42895]
+
+	* win32/win32.c (rb_w32_utime): allow NULL to set the current time.
+	  [ruby-talk:219248]
+
+Wed Oct 18 13:25:50 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_each_line): String#lines now works when a block
+	  is given.  in other words, lines become an alias to each_line.
+	  [ruby-core:09218]
+
+	* string.c (rb_str_each_byte): ditto for bytes in place of lines.
+
+Wed Oct 18 00:55:33 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): use particular enums.  [ruby-core:09221]
+
+Tue Oct 17 22:03:08 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/erb.rb: String#each was removed, use #each_line instead.
+
+Tue Oct 17 12:27:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (ary_shared_array): should set NOEMBED flag for a copied
+	  array.
+
+Tue Oct 17 08:04:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_lines): now takes optional argument for the
+	  line separator.
+
+	* io.c (rb_io_lines, rb_io_bytes): new methods.
+
+Mon Oct 16 23:33:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_unshift_m): a bug in lfree shift length
+	  calculation.
+
+Mon Oct 16 08:30:43 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: *OBJS are not needed for extension libraries.
+
+	* {bcc32,wince,win32}/Makefile.sub (config.status): fixed typo,
+	  missing comma.
+
+Mon Oct 16 00:44:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_unpack): execute block if given with unpacked value
+	  instead of creating an array.  an idea from Tim Bray.
+
+Sun Oct 15 01:03:08 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb (Collector::Dir#collect): append base
+	  directory but not prepend.
+
+	* lib/test/unit/collector/dir.rb (Collector::Dir#collect_file): do not
+	  join with dot.  fixed: [ruby-core:09179]
+
+Sat Oct 14 23:39:50 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (singleton): no need to re-create NODE_SELF() again.
+	  [ruby-core:09177]
+
+Sat Oct 14 23:25:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_warning, parser_warn): some error message may
+	  contain format specifiers.  a patch from Akinori MUSHA <knu at
+	  iDaemons.org>.  [ruby-dev:29657]
+
+	* regparse.c (onig_rb_warning, onig_rb_warn): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (VpException): ditto.
+
+	* ext/dl/handle.c (rb_dlhandle_initialize): ditto.
+
+	* ext/gdbm/gdbm.c (rb_gdbm_fatal): ditto.
+
+Sat Oct 14 08:15:42 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c, ext/digest/digest.h,
+	  ext/digest/md5/md5init.c, ext/digest/rmd160/rmd160init.c,
+	  ext/digest/sha1/sha1init.c, ext/digest/sha2/sha2init.c:
+	  Introduce API versioning.
+
+	* ext/digest/digest.c, ext/digest/digest.h,
+	  ext/digest/md5/md5init.c, ext/digest/rmd160/rmd160init.c,
+	  ext/digest/sha1/sha1init.c, ext/digest/sha2/sha2init.c: Remove
+	  the constants DIGEST_LENGTH and BLOCK_LENGTH and turn them into
+	  instance methods digest_length() and block_length().  Class
+	  methods with the same names are also provided, which take extra
+	  parameters for a digest method.
+
+	* ext/digest/lib/digest/hmac.rb: Completely redesign the somewhat
+	  bizarre API, now that Digest classes can take hashing
+	  parameters.
+
+Sat Oct 14 05:54:05 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c: Improve RDoc documentation further more.
+
+Sat Oct 14 04:33:33 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c: Improve RDoc documentation.
+
+	* ext/digest/digest.c (Init_digest, rb_digest_base_s_digest,
+	  rb_digest_base_s_hexdigest): Make Digest::Base::digest() and
+	  Digest::Base::hexdigest() take extra arguments, which are passed
+	  through to the constructor in an internal call.
+
+	* ext/digest/bubblebabble/bubblebabble.c
+	  (rb_digest_base_s_bubblebabble): Ditto for
+	  Digest::Base::bubblebabble().
+
+Sat Oct 14 00:55:08 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bcc32/Makefile.sub (post-install-ext): no longer needed.
+
+	* bcc32/configure.bat: get rid of a quirk of Borland make, which
+	  sets empty macro in command line to "1".
+
+Fri Oct 13 22:49:02 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: updated based on date2 3.9.5.
+
+Fri Oct 13 21:00:01 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest.rb (Digest): Try to auto-load non-standard
+	  digest modules when a specified digest class is missing.
+
+	* ext/digest/lib/digest.rb: Define Digest(name) for ease of
+	  dynamically selecting a hashing algorithm.
+
+Fri Oct 13 20:53:37 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (Init_digest): Digest::Base.new() does no
+	  longer take an initial string to feed.  This change allows
+	  subclasses to take hashing parameters.  A statement such as
+	  ``md = Digest::MD5.new(s)'' can be easily rewritten as
+	  ``md = Digest::MD5.new << s'' or
+	  ``md = Digest::MD5.new.update(s)''.
+
+Fri Oct 13 20:51:55 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c, ext/digest/md5/md5init.c,
+	  ext/digest/rmd160/rmd160init.c, ext/digest/sha1/sha1init.c,
+	  ext/digest/sha2/sha2init.c: Add RDoc documentation.
+
+	* ext/digest/digest.txt, ext/digest/digest.txt.ja: Removed in
+	  favor of embedded RDoc documentation.
+
+Fri Oct 13 20:38:12 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/bubblebabble, ext/digest/digest.c: Rip BubbleBabble
+	  support out of the base class and have a separate module named
+	  digest/bubblebabble.
+
+Fri Oct 13 19:53:59 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_base_equal): Again, should call
+	  digest() of a subclass instead of the one defined in the base
+	  class.
+
+Fri Oct 13 18:19:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c: Class#inherited RDoc added.  a patch from Daniel
+	  Berger <djberg96 at gmail.com>  [ruby-core:08942]
+
+Fri Oct 13 02:42:00 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_base_equal): Should call digest()
+	  of a subclass instead of the one defined in the base class.
+
+Fri Oct 13 02:30:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb (Collector::Dir#collect): prepend
+	  base directory to load path.
+
+	* lib/test/unit/collector/dir.rb (Collector::Dir#collect_file): should
+	  use the given File-like interface, but not File directly.
+
+	* test/testunit/collector/test_dir.rb (TestDir::FileSystem): implement
+	  File-like methods correctly.
+
+Fri Oct 13 01:48:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/date.rb (Date::self.complete_hash): need to check if g is
+	  nil before dereference.  [ruby-core:09116]
+
+Fri Oct 13 01:05:58 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_partition): RDoc update.  a patch from
+	  Mauricio Fernandez <mfp at acm.org>.  [ruby-core:09160]
+
+	* hash.c (rb_hash_compare_by_id): ditto.
+
+Fri Oct 13 00:34:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_mod_cvar_defined): wrong id check.  a patch from
+	  Mauricio Fernandez <mfp at acm.org>.  [ruby-core:09158]
+
+	* object.c (rb_mod_cvar_get): typo fixed.  [ruby-core:09168]
+
+	* object.c (rb_mod_cvar_set): ditto.
+
+Thu Oct 12 22:58:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_compare_by_id): somehow we lost renaming from
+	  Hash#identical.  [ruby-core:09163]
+
+Thu Oct 12 18:25:40 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/tk/tkutil/tkutil.c (cbsubst_table_setup): need to handle new
+	  character literal (1 char string).
+
+	* lib/mkmf.rb: shut up some warnings from tk's extconf.rb.
+
+Thu Oct 12 02:15:24 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb: Make use of String#bytes.
+
+Thu Oct 12 02:12:31 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (get_digest_base_metadata): Use an instance
+	  variable of a class object instead of a class variable for
+	  metadata.  This change is only crucial for ruby 1.8 because
+	  class variables are inherited to subclasses prior to 1.9, but
+	  applying it also to 1.9 will assure compatibilities.
+
+	* ext/digest/md5/md5init.c (Init_md5): Ditto.
+
+	* ext/digest/rmd160/rmd160init.c (Init_rmd160): Ditto.
+
+	* ext/digest/sha1/sha1init.c (Init_sha1): Ditto.
+
+	* ext/digest/sha2/sha2init.c (Init_sha2): Ditto.
+
+Wed Oct 11 21:36:47 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_base_alloc,
+	  rb_digest_base_equal): Simplify the equality check and just
+	  compare resulted digests since state-level equality should
+	  not be so significant.
+
+	* ext/digest/digest.h: Ditto.
+
+	* ext/digest/*/*.[ch]: Ditto.
+
+Wed Oct 11 17:11:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_obj_define_method): add half boiled RDoc document.
+
+Wed Oct 11 16:57:46 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_replace): should shift lfree pointer before
+	  calling xfree.
+
+Wed Oct 11 15:07:42 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb: Add digest/hmac, which implements
+	  HMAC keyed-hashing algorithm.
+
+Wed Oct 11 15:03:55 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_base_reset): Do not make
+	  recursive calls, but call initialize() when reset() is not
+	  defined in a subclass.
+
+Wed Oct 11 14:56:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/digest/sha1/sha1ossl.h: libssl 0.9.8c-3 defines no
+	  SHA_BLOCK_LENGTH.
+
+Wed Oct 11 14:03:31 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (rb_digest_base_reset, Init_digest): Add
+	  Digest::Base#reset.
+
+	* ext/digest/digest.h: Update the header comment.
+
+	* ext/digest/md5/md5ossl.h, ext/digest/md5/md5init.c (Init_md5):
+	  Define DIGEST_LENGTH and BLOCK_LENGTH.
+
+	* ext/digest/rmd160/rmd160init.c (Init_rmd160): Ditto.
+
+	* ext/digest/sha1/sha1init.c (Init_sha1): Ditto.
+
+	* ext/digest/sha2/sha2init.c (Init_sha2): Ditto.
+
+	* ext/digest/depend, ext/digest/extconf.rb: Use $INSTALLFILES
+	  rather than adding make targets. [Pointed out by: nobu]
+
+Tue Oct 10 16:39:08 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (hexdigest_str_new, bubblebabble_str_new):
+	  Perform StringValue() checks properly.
+
+Tue Oct 10 13:21:21 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/sha1/depend, ext/digest/sha2/depend: Remove obsolete
+	  dependencies.
+
+Mon Oct  9 23:46:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/parsedate.rb: documentation patch from Konrad Meyer
+	  <konrad.meyer@gmail.com>.  [ruby-doc:1238]
+
+	* lib/open3.rb, lib/ping.rb: ditto.
+
+Mon Oct  9 23:40:58 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb, lib/fileutils.rb, lib/mkmf.rb, lib/optparse.rb,
+	  lib/shellwords.rb: get rid of shadowing outer local variable.
+
+Mon Oct  9 22:56:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/encoding.rb (REXML::Encoding::check_encoding): spaces
+	  are allowed around equal sign.  [ruby-core:09032]
+
+	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser): ditto.
+
+Mon Oct  9 01:56:34 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_obj_define_method): add new method
+	  Kernel#define_singleton_method.  [ruby-list:42851]
+
+Sat Oct  7 23:53:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_scan): small documentation fix.
+	  [ruby-core:09007]
+
+Sat Oct  7 23:44:33 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_rshift): a bug in right shift of negative
+	  bignums.  [ruby-core:09020]
+
+Sat Oct  7 23:33:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (formal_assign): packed post splat arguments may conflict
+	  with normal arguments.  [ruby-core:09021]
+
+	* eval.c (rb_call0): ditto.
+
+Sat Oct  7 11:53:04 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_initialize): since module_eval no longer passes
+	  self, use module_exec instead.  fixed: [ruby-dev:29637]
+
+Sat Oct  7 00:27:58 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c (rb_include_module): remove unnecessary check.
+	  [ruby-talk:218402]
+
+Fri Oct  6 15:19:59 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/depend: Fix header installation when the build
+	  directory is different from srcdir. [Pointed out by: eban]
+
+Fri Oct  6 09:56:31 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32,wince}/Makefile.sub (config.status): shouldn't use
+	  copy command instead of install. use -run install.
+
+Fri Oct  6 06:53:46 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_yield_0): small refactoring.
+
+	* parse.y (bparam_item): fixed bugs in handling parenthesized LHS.
+
+Fri Oct  6 04:47:07 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/depend: Install digest.h.
+
+Fri Oct  6 04:27:40 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/md5.rb, ext/digest/lib/sha1.rb: Remove those
+	  compatibility stub libraries.
+
+	* sample/openssl/c_rehash.rb: Use digest/md5 instead of obsolete md5.
+
+Fri Oct  6 04:09:51 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c: Make hexdigest() always call digest() internally.
+
+	* ext/digest/digest.c: Add bubblebabble().
+
+Fri Oct  6 02:38:42 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c: Allow subclassing in Ruby.
+
+Fri Oct  6 02:06:10 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.c (hexdigest_str_new): Add a string size check.
+
+Thu Oct  5 19:28:35 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/digest.[ch]: Since the argument order of
+	  hash_final_func_t was inconsistent with others, change it and
+	  rename to hash_finish_func_t to avoid confusion.
+
+	* ext/digest/digest.[ch]: Remove and eliminate the use of
+	  hash_end_func_t.  Implement hexdigest conversion in the base
+	  class.
+
+	* ext/digest/md5/md5.c, ext/digest/md5/md5.h,
+	  ext/digest/md5/md5init.c, ext/digest/md5/md5ossl.c,
+	  ext/digest/md5/md5ossl.h: Remove MD5_End() and change
+	  MD5_Final() to MD5_Finish().
+
+	* ext/digest/rmd160/depend, ext/digest/rmd160/extconf.rb,
+	  ext/digest/rmd160/rmd160.c, ext/digest/rmd160/rmd160.h,
+	  ext/digest/rmd160/rmd160hl.c, ext/digest/rmd160/rmd160init.c,
+	  ext/digest/rmd160/rmd160ossl.c, ext/digest/rmd160/rmd160ossl.h:
+	  Remove unused functions RMD160_End(), RMD160_File(),
+	  RMD160_Data() and change RMD160_Final() to RMD160_Finish().
+
+	* ext/digest/sha1/extconf.rb, ext/digest/sha1/sha1.c,
+	  ext/digest/sha1/sha1.h, ext/digest/sha1/sha1hl.c,
+	  ext/digest/sha1/sha1init.c, ext/digest/sha1/sha1ossl.c,
+	  ext/digest/sha1/sha1ossl.h: Likewise.
+
+	* ext/digest/sha2/extconf.rb, ext/digest/sha2/sha2.c,
+	  ext/digest/sha2/sha2.h, ext/digest/sha2/sha2hl.c,
+	  ext/digest/sha2/sha2init.c: Likewise.
+
+Wed Oct  4 18:47:25 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkextlib/*: bugfix and update
+	  (see ext/tk/ChangeLog.tkextlib).
+
+Wed Oct  4 17:25:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call): check protected visibility based on real self,
+	  not ruby_frame->self.  [ruby-talk:217822]
+
+Wed Oct  4 15:46:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (block_param): should interpret single parenthesized
+	  left hand side expression.
+
+Wed Oct  4 08:52:30 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/optparse/test_getopts.rb: changed the class name of test case
+	  to get rid of conflict with test_optparse.rb.
+
+Tue Oct  3 21:04:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (dyna_in_block): inline using macro.
+
+	* parse.y (mlhs): simplifies the rule a bit.
+
+	* parse.y (block_param): restrict block parameters to be local
+	  variables only.
+
+	* test/ruby/test_iterator.rb (TestIterator::test_nested_iterator):
+	  update test suite to conform the last change.
+
+Tue Oct  3 02:31:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (splat_value): use "to_splat" instead of "to_ary" to
+	  prepare splat values as an array.
+
+	* array.c (Init_Array): define to_splat.
+
+	* range.c (range_to_splat): new method.
+
+	* enumerator.c (enumerator_to_splat): ditto.
+
+Tue Oct  3 01:36:47 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_lines): returns an Enumerator instead of an
+	  array of lines.
+
+	* string.c (rb_str_bytes): a new method.
+
+Mon Oct  2 23:47:55 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner::COLLECTORS):
+	  base directory should be lower precedence.  fixed: [ruby-dev:29622]
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner#options): typo.
+
+	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir#collect_file):
+	  load expanded path.  fixed: [ruby-dev:29621]
+
+Mon Oct  2 15:47:55 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* instruby.rb: batfile should be CRLF'ed.
+
+Mon Oct  2 01:24:26 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (test-all): separate directory where running test cases
+	  from source tree.
+
+	* lib/test/unit/autorunner.rb (options): added --basedir, --workdir
+	  and --load-path options.
+
+	* lib/test/unit/collector/dir.rb (recursive_collect, collect_file):
+	  base directory support.
+
+Sun Oct  1 23:56:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.mk, ext/extmk.rb, win{32,ce}/Makefile.in: keep
+	  LIBRUBY_SO unless need to be removed.
+
+Sun Oct  1 23:12:19 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#make_switch): pass arguments directly.
+
+Sat Sep 30 15:11:26 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: updated based on date2 3.9.4.
+
+Fri Sep 29 13:18:24 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest.rb (Digest): Require digest.so and fix the
+	  breakage.  Point out by NAKAMURA Usaku in [ruby-dev:29619].
+
+Fri Sep 29 12:11:04 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* jcode.rb (succ!): call original succ! if $KCODE == 'n'.
+	  fixed: [ruby-talk:216845]
+
+Fri Sep 29 11:43:40 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_func): revert fallback checking undeclared function.
+	  fixed: [ruby-core:08949]
+
+Fri Sep 29 09:56:56 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: extout is needed for also clean.
+	  fixed: [ruby-core:08944]
+
+	* lib/optparse.rb (OptionParser::Switch#conv_arg): unsplat by
+	  Proc#call if no conversion is given.
+
+Thu Sep 28 23:59:31 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.h (struct thread): declare win32_exception_list on cygwin and
+	  win32 regardless if it is implemented.  Provisional fix for
+	  [ruby-core:08917].
+
+Thu Sep 28 20:49:20 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/tmpdir.rb: use return value of getdir.call for length.
+
+Wed Sep 27 22:08:16 2006  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/md5/md5init.c (Init_md5): Now that we have digest.rb,
+	  require "digest" rather than "digest.so".
+
+	* ext/digest/rmd160/rmd160init.c (Init_rmd160): Ditto.
+
+	* ext/digest/sha1/sha1init.c (Init_sha1): Ditto.
+
+	* ext/digest/sha2/sha2init.c (Init_sha2): Ditto.
+
+Wed Sep 27 21:21:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_startwith): rename startwith? to start_with?,
+	  endwith? to endwith?, respectively.  [ruby-talk:216685]
+
+Wed Sep 27 13:29:01 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::TagMaker::nOE_element_def): replace to_s by
+	  join.  some other methods as well.  [ruby-dev:29613]
+
+Wed Sep 27 01:04:49 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_func): check function pointer first and macro next.
+
+	* lib/mkmf.rb (have_type): simplified with typedef and sizeof.
+
+Wed Sep 27 00:08:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_shift): shift/unshift performance boost patch,
+	  based on the patch from Eric Mahurin <eric_mahurin at yahoo.com>.
+	  [ruby-core:05861]
+
+	* array.c (rb_ary_unshift_m): ditto.
+
+	* array.c (ary_make_shared): ditto.
+
+	* array.c (RESIZE_CAPA): ditto.
+
+	* array.c (rb_ary_free): new function to free memory.  code moved
+	  from gc.c.
+
+	* string.c (rb_str_free): ditto.
+
+Tue Sep 26 23:57:03 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#getopts): use strings as key.
+	  fixed: [ruby-dev:29614]
+
+Tue Sep 26 15:29:55 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {win32,wince}/Makefile.sub (CPP): check predefined value.
+
+Tue Sep 26 07:55:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_shift): should not move memory region if array
+	  body is shared.  a patch from Kent Sibilev <ksruby at gmail.com>.
+	  [ruby-core:08922]
+
+Mon Sep 25 23:10:46 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (rb_push_glob): need not to check by FilePathValue().
+	  [ruby-dev:29599]
+
+	* dir.c (dir_globs): ditto.
+
+Mon Sep 25 22:26:26 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_path_end): skip root directory.  fixed: [ruby-core:08913]
+
+	* lib/mkmf.rb (rm_f): get rid of NUL.
+
+	* lib/mkmf.rb (init_mkmf): set default $LDFLAGS.  Patch by Michal
+	  Suchanek <hramrach at centrum.cz>.  [ruby-talk:216256]
+
+Mon Sep 25 15:06:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/test.rb: "print nil" now prints empty string.
+
+	* test/ruby/test_system.rb (TestSystem::test_system): ditto.
+
+Mon Sep 25 11:26:25 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (recursive_hash): remove unused local variable.
+
+	* parse.y (parser_yylex): ditto.
+
+	* parse.y (rb_gc_mark_symbols): fix unmatched prototype .
+
+	* file.c (rb_get_path): check NUL byte in the path string.
+
+Mon Sep 25 08:14:43 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_shift): should clear shifting top element.
+	  [ruby-talk:216055]
+
+	* array.c (rb_ary_shift): avoid creating shared object if array
+	  size is small.
+
+Mon Sep 25 08:11:35 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* random.c (rb_f_rand): RDoc typo fix.  a patch from Frederick
+	  Cheung <fred at 82ask.com>.  [ruby-talk:216047]
+
+Sun Sep 24 21:19:24 2006  Guy Decoux  <ts@moulon.inra.fr>
+
+	* gc.c (gc_mark_children): NODE_POSTEXE holds Ruby VALUE.
+	  [ruby-core:08912]
+
+Sun Sep 24 22:28:20 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* runruby.rb: extension library scripts moved into common directory.
+
+Sun Sep 24 12:10:04 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: updated based on date2 3.9.3.
+
+Sun Sep 24 06:55:36 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_print): no special handling for nil as well as puts.
+	  fixed: [ruby-dev:29586]
+
+Sun Sep 24 06:25:53 2006  why the lucky stiff  <why@ruby-lang.org>
+
+	* eval.c (rb_thread_save_context, rb_thread_restore_context):
+	  sandbox hook to save and restore sandbox state.
+
+	* eval.c (thread_no_ensure): added THREAD_NO_ENSURE thread flag.
+
+	* eval.c (rb_thread_kill_bang): Thread#kill! uses the above flag
+	  to circumvent ensure, in order to prevent endless loops.
+	  contributed by MenTaLguY. [ruby-core:08768]
+
+	* eval.c (rb_thread_kill): fix Thread#kill docs, which returns
+	  the thread object in all cases.
+
+	* node.h: expose the rb_jmpbuf_t and rb_thread_t structs, along
+	  with the thread flags.  used by the sandbox extension.
+
+	* ruby.h: extern rb_eThreadError, so sandbox can swap it.
+
+Sat Sep 23 21:34:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::read_multipart): CGI content
+	  may be empty.  a patch from Jamis Buck <jamis at 37signals.com>.
+
+Sat Sep 23 20:54:28 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* oniguruma.h: Version 4.4.4
+
+	* regexec.c: ditto.
+
+	* regcomp.c ditto.
+
+Sat Sep 23 08:35:53 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rdoc/ri/ri_options.rb: prevent NameError.  [ruby-dev:29597]
+
+Sat Sep 23 01:02:57 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: updated based on date2 3.9.2.
+
+Fri Sep 22 18:07:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_partition): no need to call rb_call_super(),
+	  since String is no longer includes Enumerable.
+
+Fri Sep 22 17:33:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_eql): new method to be used by Hash.
+
+	* hash.c (rb_hash_hash): ditto.
+
+Fri Sep 22 06:53:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_hash): use rb_memhash().
+
+	* numeric.c (flo_hash): simplified.  klass need not to affect
+	  resulting hash value.
+
+Fri Sep 22 02:06:26 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .cvsignore: ignore timestamp files and installed list file.
+
+Fri Sep 22 01:36:34 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb: include FileUtils unconditionally.
+
+Fri Sep 22 00:36:05 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (Init_Numeric): fix_odd_p and fix_even_p are for Fixnum.
+	  patch from Ondrej Bilka <neleai at seznam.cz>.  [ruby-core:08904]
+
+Thu Sep 21 22:56:20 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (no-install): not install rdoc actually.
+
+	* common.mk (install-doc, no-install-doc): use instruby.rb.
+
+	* instruby.rb: rdoc installation.
+
+	* ext/extmk.rb: expand ruby executable names.
+
+Thu Sep 21 20:19:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_new3): embed shorter strings more eagerly.
+
+Thu Sep 21 17:44:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_startwith): a new method to check if a string
+	  starts with given prefix.
+
+	* string.c (rb_str_endwith): the opposite of String#startwith?.
+
+Thu Sep 21 16:29:02 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* rubytest.rb: use each_line instead of each.
+
+Thu Sep 21 15:06:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_odd_p): a new method to check even or odd.
+	  [RCR#337]
+
+	* numeric.c (int_even_p): ditto.
+
+Thu Sep 21 13:55:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_getpwuid): uid integer should be wrapped in
+	  uid_t value.  [ruby-core:08897]
+
+	* ext/etc/etc.c (etc_getpwuid): uid_t may be bigger than plain
+	  'int' type.
+
+Thu Sep 21 10:07:09 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_partition): RDoc typo fixed.  [ruby-core:08898]
+
+	* string.c (rb_str_rpartition): fixed separation seek bug.
+
+Thu Sep 21 09:38:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_lines): new method to split a string into lines.
+
+	* string.c (Init_String): Strings are no longer Enumerable.  use
+	  each_line or lines method explicitly.
+
+	* string.c (Init_String): remove each method. use each_lines.
+
+Wed Sep 20 23:17:41 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (pre-install-doc): create data directory before install.
+
+	* lib/mkmf.rb (dir_re): fixed typo.
+
+	* lib/mkmf.rb (install_dirs): remove extra slash.
+
+Wed Sep 20 22:41:45 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_mul): typo again.  patch from Tadashi Saito
+	  <shiba at mail2.accsnet.ne.jp>.  fixed: [ruby-core:08893]
+
+Wed Sep 20 19:32:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_partition): a new method to separate the string
+	  by a separator.  taken from Python 2.5.
+
+	* string.c (rb_str_rpartition): ditto.
+
+Wed Sep 20 09:49:40 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32,wince}/Makefile.sub (INSTALLED_LIST): need to define
+	  this macro to install.
+
+Wed Sep 20 09:43:10 2006  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: allow extra spaces in responses.
+	  Thanks, Tom Soderlund.
+
+Wed Sep 20 09:25:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/gdbm/gdbm.c: add RDoc documentation. a patch from Peter
+	  Adolphs <futzilogik at users dot sourceforge dot net>.
+	  [ruby-doc:1223]
+
+Tue Sep 19 00:42:15 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_ivar_defined, rb_mod_cvar_defined): new methods,
+	  Kernel#instance_variable_defined? and Module#class_variable_defined?.
+	  [ruby-dev:29587]
+
+	* lib/date/format.rb (Date::Bag#method_missing): use new method,
+	  instance_variable_defined? to check if an instance variable is
+	  defined.  fixed: [ruby-dev:29554]
+	  -- This didn't fix anything.
+
+Tue Sep 19 00:07:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_eql): fail early to gain performance.
+
+	* string.c (sym_hash): cache hash value in aux.shared if possible.
+
+	* gc.c (rb_obj_id): no need to treat symbols specially.
+
+	* lib/fileutils.rb (FileUtils::FileUtils): singleton_methods() no
+	  longer return an array of strings, but of symbols.
+
+	* lib/delegate.rb (DelegateClass): ditto.
+
+Mon Sep 18 15:29:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_s_glob): restore GC protection volatile variable.
+	  [ruby-dev:29588]
+
+	* re.c (rb_reg_regcomp): ditto.
+
+Mon Sep 18 12:16:48 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_mul): get rid of shift overflow.
+
+Mon Sep 18 10:47:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_s_glob): remove unused variable.
+
+	* math.c (math_log): ditto.
+
+	* re.c (rb_reg_regcomp): ditto.
+
+	* eval.c (break_jump): ditto.
+
+	* eval.c (rb_thread_yield_0): remove unused function.
+
+Sun Sep 17 23:44:58 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc#document): scan only files modified
+	  after the previous generation.
+
+Sun Sep 17 17:42:13 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (install-doc): reverted.
+
+	* instruby.rb: stores file name list without destdir prefix.
+
+	* lib/rdoc/generators/ri_generator.rb: do not chdir twice.
+
+Sun Sep 17 10:42:10 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_mul): fixed typo.  fixed: [ruby-core:08885]
+
+Sat Sep 16 19:47:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* README.EXT: should mention new macros: RSTRING_PTR, RSTRING_LEN,
+	  RARRAY_PTR, RARRAY_LEN.
+
+	* README.EXT.ja: ditto.
+
+Sat Sep 16 16:39:23 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.in, instruby.rb, ext/extmk.rb, lib/mkmf.rb:
+	  use instruby.rb to install extensions instead of ext/extmk.rb.
+
+	* instruby.rb: store installed list into the file.
+
+	* ext/dbm/extconf.rb: allow multiple candidates for dbm-type.
+
+	* ext/io/wait/extconf.rb: suspicious checking_for.
+
+	* ext/pty/pty.c (establishShell): parent pid is not used.
+
+	* ext/pty/pty.c (freeDevice): not used.
+
+	* lib/mkmf.rb (checking_for): improved the messages.
+
+Sat Sep 16 11:03:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (ary_shared_first): should create embedded copies
+	  instead of sharing memory region for smaller arrays.
+
+Sat Sep 16 09:37:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (inspect_struct): do not display a class name for
+	  anonymous struct.  The member fields are sufficient.
+
+Fri Sep 15 20:22:15 2006  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-8/nkf.c: imported nkf 2.0.8 rev.110.
+	  * Fix: check_bom cuts \xfe\xff\xXX\xXX of UTF-32LE.
+	  * Add support --ic=UTF-32.
+	  * Fix: can't guess UTF-16 and UTF-32.
+	  * Fix: can't decode beyond BMP of UTF-16LE.
+
+	* ext/nkf/nkf.c (guess): Support UTF-32.
+
+	* ext/nkf/lib/kconv.rb (kconv): Support UTF-32.
+
+	* ext/nkf/lib/kconv.rb (to_utf32): new method.
+
+Fri Sep 15 05:23:24 2006  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-8/nkf.c: imported nkf 2.0.8 2006-09-15.
+	  Add support for U+10000 - U+10FFFF
+	  Add support UTF-32
+
+Fri Sep 15 00:03:07 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/digest/lib/digest.rb (Digest::Base.file): open a file in binary
+	  mode.  suggested by Kazuhiro NISHIYAMA.  [ruby-dev:29579]
+
+Thu Sep 14 17:21:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_mul): avoid bignum multiplication as far as
+	  possible.  a patch from Ondrej Bilka <neleai at seznam.cz>.
+	  [ruby-core:08825]
+
+Thu Sep 14 16:34:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_intern): allow zero length symbols.
+	  [ruby-core:08861]
+
+Thu Sep 14 16:11:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_intern): raise SecurityError only when $SAFE
+	  level is greater than zero.  [ruby-core:08862]
+
+	* parse.y (rb_interned_p): new function to check if a string is
+	  already interned.
+
+	* string.c (str_to_id): use rb_str_intern().
+
+Thu Sep 14 14:37:45 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/digest/lib/digest.rb (Digest::Base.file): new method.
+	  [ruby-dev:29572]
+
+Thu Sep 14 08:30:02 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/digest/digest.c (rb_digest_base_inspect): new method.
+	  [ruby-dev:29573]
+
+Thu Sep 14 01:13:56 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (ruby_init_stack): decrease "stack level too deep" in Windows.
+	  [ruby-dev:29569]
+
+Thu Sep 14 01:02:25 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/digest/lib/digest.rb: new file.
+	  [ruby-dev:28689]
+
+Wed Sep 13 18:43:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* README.EXT: English adjustment.  [ruby-core:08851] and
+	  [ruby-core:08852]
+
+Wed Sep 13 18:25:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-parse-partial): better here-doc support.
+	  a patch from Marshall T. Vandegrift <llasram at gmail.com>.
+	  [ruby-core:08804]
+
+Wed Sep 13 16:43:36 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_intern): prohibit interning tainted string.
+
+Wed Sep 13 01:14:02 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#getopts): works with pre-registered
+	  options.  [ruby-core:08826]
+
+Tue Sep 12 03:58:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_compare_by_identity): rename Hash#identical to
+	  Hash#compare_by_identity.
+
+Mon Sep 11 16:52:37 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_identical): a new method to make a hash to
+	  compare keys by their identity.
+
+	* hash.c (rb_hash_identical_p): new method to tell if a hash is
+	  identical or not.
+
+	* st.c (st_numcmp, st_numhash): export hash type functions.
+
+Mon Sep 11 11:42:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rexml/source.rb (REXML::Source::encoding): should not
+	  convert the body twice.  [ruby-core:08828]
+
+	* lib/rexml/encoding.rb (REXML::Encoding::encoding):
+	  Encoding#encoding= to return boolean value to tell if the body
+	  is really converted or not.
+
+	* lib/rexml/encoding.rb (REXML::Encoding::encoding): Specific
+	  conversion library (e.g. rexml/encodings/UTF-16.rb) to have
+	  higher preceding.
+
+	* lib/rexml/encodings/UTF-16.rb (REXML::Encoding::decode_utf16):
+	  UTF-16#decode_utf16 should work strings without BOM.
+
+Mon Sep 11 07:39:44 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_equal): "sym == str" should compare them as
+	  strings.  [ruby-dev:29554]
+
+Sun Sep 10 22:59:43 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (parse_args): remove splat.
+
+Sun Sep 10 20:25:30 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: updated based on date2 3.9.1.
+
+Sun Sep 10 09:41:29 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c: ISPRINT() needs ctype.h
+
+Sun Sep 10 09:19:47 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb: splat parsed arguments.
+
+Tue Jan 10 09:18:03 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_require_safe): prevent extension from loading twice.
+	  fixed: [ruby-dev:29523]
+
+Sat Sep  9 23:55:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_f_test): test(0) should not have any special
+	  meaning.  [ruby-dev:29425]
+
+	* file.c (rb_f_test): properer error message.
+
+Sat Sep  9 14:08:38 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/test/unit/testcase.rb (Test::Unit::TestCase#run): Rescue
+	  Exception in Test::Unit::TestCase#run.  [ruby-core:08783]
+
+Sat Sep  9 04:55:59 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pstore.rb: open all in binary mode, and get rid of the quirk of
+	  msvcrt.  fixed: [ruby-dev:29518]
+
+Sat Sep  9 04:47:45 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub (MINIRUBY): append MINIRUBYOPT.
+
+	* mkconfig.rb, ext/extmk.rb, lib/mkmf.rb, win32/mkexports.rb: suppress
+	  warnings with $VERBOSE.
+
+	* win32/resource.rb: only file which has more than one icon is DLL.
+
+Fri Sep  8 16:53:30 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_alloc): should allocate a String object, even when
+	  asked to allocate a Symbol object.  [ruby-dev:29529]
+
+Fri Sep  8 16:36:27 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): follow Array#to_s.
+
+	* lib/mkmf.rb (create_makefile): ditto.
+
+	* win32/resource.rb: ditto.
+
+Fri Sep  8 10:00:12 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cookie.rb (WEBrick::Cookie.parse_set_cookies): new
+	  method to parse multiple cookies per Set-Cookie header.
+	  Thanks to Aaron Patterson <aaron_patterson at speakeasy.net>.
+	  [ruby-core:08802]
+
+Fri Sep  8 08:59:30 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub, win32/configure.bat win32/setup.mak: program
+	  name transform.
+
+Fri Sep  8 08:25:39 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb: suppress `assigning void value' warning.
+
+Fri Sep  8 01:16:34 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (Init_Array): #to_s to be an alias to #inspect.
+	  [ruby-dev:29520]
+
+	* hash.c (Init_Hash): ditto.
+
+	* lib/mkmf.rb (create_makefile): replace "print array" by
+	  "print *array".
+
+	* mkconfig.rb: ditto.
+
+Thu Sep  7 21:02:56 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (nil_to_s): returns the empty string again.
+	  [ruby-dev:29520]
+
+Thu Sep  7 23:27:05 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (path_check_0, fpath_check): disable path check on cygwin.
+	  [ruby-talk:213074]
+
+Thu Sep  7 02:03:45 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_to_s): adopt new date format using digits
+	  e.g. "2006-09-07 02:03:45 +9000".
+
+Thu Sep  7 01:54:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (sym_equal): override.  check equivalence.
+
+Wed Sep  6 13:25:04 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (symbols_i): need to initialize early-created symbols.
+	  [ruby-dev:29496]
+
+Wed Sep  6 12:05:19 2006  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb (Kconv::toeuc): remove -m0 [ruby-dev:29505]
+
+Tue Sep  5 22:06:43 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c: use rb_ary_new3() since RARRAY_LEN() is not l-value.
+
+	* ext/tk/tkutil/tkutil.c: use RARRAY_PTR() and RARRAY_LEN() and etc.
+	  fixed: [ruby-dev:29473]
+
+Tue Sep  5 06:47:22 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_to_s): variable declaration after an execution
+	  statement.
+
+Tue Sep  5 05:49:41 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (path_check_0): check if sticky bit is set on parent
+	  directories for executable path.  fixed: [ruby-dev:29415]
+
+Tue Sep  5 05:03:46 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_plus): addition in Fixnum will never overflow
+	  long.  a patch from Ondrej Bilka <neleai at seznam.cz>.
+	  [ruby-core:08794]
+
+	* numeric.c (fix_minus): ditto.
+
+	* bignum.c (rb_big_pow): eagerly truncate resulting bignum.
+	  [ruby-core:08794]
+
+Mon Sep  4 23:15:34 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_to_s): make it conform to RFC2822 date format.
+	  [ruby-dev:29467]
+
+Mon Sep  4 21:43:57 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dbm/extconf.rb: create makefile according to the result of check
+	  for dbm header.  fixed: [ruby-dev:29445]
+
+Mon Sep  4 21:39:42 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: updated based on date2 3.9.
+
+Mon Sep  4 21:14:20 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_strftime): include nul character.  fixed: [ruby-dev:29422]
+
+Mon Sep  4 16:39:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::out): specify -x option for nkf.
+
+	* lib/cgi.rb (CGI::out): should not convert utf-8 implicitly using
+	  NKF.  it is too Japanese centric.
+
+Mon Sep  4 14:23:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/dbm/extconf.rb (db_check): remove debug print.
+
+Mon Sep  4 06:46:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (rb_id2sym): intern if id is attrset_id.
+	  [ruby-dev:29420] [ruby-dev:29447]
+
+Mon Sep  4 01:25:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_local_variables): list symbols.
+
+	* struct.c (rb_struct_s_members_m): ditto.
+
+	* variable.c (ivar_i): ditto.
+
+	* variable.c (gvar_i): ditto.
+
+	* variable.c (cv_i): ditto.
+
+Sun Sep  3 20:47:02 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.h (SYMBOL_P): Qnil and Qfalse are not Symbol.
+
+Sun Sep  3 15:32:44 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: get rid of nil.to_s.
+
+Sun Sep  3 06:24:38 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (ruby_connect): sockerrlen should be socklen_t.
+
+Sun Sep  3 04:40:42 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: check arpa/inet.h for ntohs.
+
+	* ext/socket/socket.c: include arpa/inet.h if available.
+
+Sat Sep  2 23:59:58 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (Init_String): undef Symbol#new.
+
+	* struct.c (rb_struct_s_def): wrong symbol detection.
+
+Sat Sep  2 23:59:37 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_to_id): a bug caused by premature optimization.
+
+Sat Sep  2 23:53:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (Init_Object): move symbol related code to string.c
+
+	* string.c (Init_String): Symbol as subclass of String.
+
+	* parse.y (rb_intern2): handle symbol as strings.
+
+	* string.c (str_new): substring of symbols are mere strings, not
+	  symbols.
+
+Sat Sep  2 23:37:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (struct RArray): embed small arrays.
+	  (RARRAY_LEN): defined for accessing array members.
+	  (RARRAY_PTR): ditto.
+
+	* array.c: use RARRAY_LEN and RARRAY_PTR.
+
+Sat Sep  2 13:23:01 2006  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (ia64.o): use the compiler driver to assemble ia64.s
+	  to use appropriate ABI.
+
+Sat Sep  2 12:06:35 2006  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/generator.rb (SOAP::SOAPGenerator#encode_tag): do not dump
+	  XML attribute which value is nil.  value "" and nil both were dumped
+	  as 'attr="value"'.  [ruby-dev:29395]
+
+Sat Sep  2 12:00:32 2006  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb (CSV::IOReader#initialize): use String#[](pos, len)
+	  instead of String#[](idx) to check utf BOM.  follows String#[](idx)
+	  behavior change of 1.9.
+
+Sat Sep  2 11:47:58 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): should handle when in else clause.  a patch
+	  from Eric Hodel <drbrain at segment7.net>.  [ruby-core:08662]
+
+	* parse.y (primary): wrap with NODE_CASE.  [ruby-core:08663]
+
+Fri Sep  1 22:07:04 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (RSTRING_EMBED_LEN_MASK): uses 5 bits to support 64bit
+	  environment.  [ruby-dev:29369]
+
+Fri Sep  1 22:02:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_resize): should copy embedded string to
+	  malloc'ed buffer.  a patch from <nobu at ruby-lang.org> in
+	  [ruby-dev:29369].  fixed: [ruby-dev:29368]
+
+	* string.c (rb_str_ord): use %ld specifier since STRING_LEN() is a
+	  long.  [ruby-dev:29369]
+
+Fri Sep  1 21:41:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (socks_init): typo fixed.  a patch from Sven
+	  Klemm <sven at c3d2.de>.  [ruby-core:08770]
+
+Fri Sep  1 14:22:42 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* array.c (rb_ary_shuffle): RDoc fixed.
+
+Fri Sep  1 13:52:57 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/font.rb: TkFont#current_configinfo() doesn't work
+	  on Tcl/Tk8.x.
+
+Fri Sep  1 09:32:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (RubyLex::getc): should not push nil into
+	  reading buffer (@readed).  reported in
+	  <http://jarp.does.notwork.org/diary/200608c.html#200608311>.
+
+Thu Aug 31 23:59:03 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): follow nil.to_s.
+
+Thu Aug 31 20:50:46 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): follow nil.to_s.
+
+	* win32/resource.rb: ditto.
+
+Thu Aug 31 20:21:47 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (search_required): use RSTRING_PTR and RSTRING_STR.
+
+	* file.c (test_identical, rb_file_s_truncate): ditto.
+
+	* io.c (pipe_open, rb_io_reopen): ditto.
+
+	* object.c (nil_plus): ditto.
+
+	* process.c (proc_spawn_n, rb_spawn): ditto.
+
+	* util.c (ruby_add_suffix): ditto.
+
+	* ext/Win32API/Win32API.c (Win32API_initialize): ditto.
+
+	* ext/dl/cptr.c (rb_dlptr_s_to_ptr): ditto.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_create_ext): ditto.
+
+	* ext/tk/stubs.c, ext/tk/tcltklib.c, ext/tk/tkutil/tkutil.c: ditto.
+
+	* ext/win32ole/win32ole.c (ole_val2olevariantdata): ditto.
+
+Thu Aug 31 18:23:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (struct RString): embed small strings.
+	  (RSTRING_LEN): defined for accessing string members.
+	  (RSTRING_PTR): ditto.
+
+	* string.c: use RSTRING_LEN and RSTRING_PTR.
+
+Thu Aug 31 17:16:19 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_shuffle_bang): new method.
+
+	* array.c (rb_ary_shuffle): ditto.
+
+	* random.c (genrand_real): ditto.
+
+	* random.c (genrand_int32): export the function.
+
+	* random.c (Init_Random): initialize random seed at the
+	  beginning.
+
+Thu Aug 31 13:12:06 2006  why the lucky stiff  <why@ruby-lang.org>
+
+	* eval.c (ruby_init): rename top_cref to ruby_top_cref and export,
+	  along with ruby_cref, for use by the sandbox. [ruby-core:08762]
+
+	* node.h: ditto.
+
+Wed Aug 30 12:01:57 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_hash): improve collision.
+
+	* string.c (rb_memhash): new generic function to calculate hash value
+	  for memory chunk.
+
+Tue Aug 29 19:10:10 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_s_create): fixed memory leak, based on the patch
+	  by Kent Sibilev <ksruby at gmail.com>.  fixed: [ruby-talk:211233]
+
+Mon Aug 28 11:29:46 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* eval.c, parse.y: Revert.
+	* ext/.document: Add digest.c.
+	* ext/digest/digest.c: Make RDoc show up.
+	* ext/io/wait.c: Fix call-seq in RDoc.
+
+Mon Aug 28 08:03:20 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/.document: Add C files with RDoc.
+	* ext/digest/digest.c: Convert to RDoc.
+	* ext/io/wait.c: ditto.
+	* lib/rdoc/parsers/parse_rb.rb: Fix typo.  Submitted by
+	  <calamitas at gmail.com>.  [ruby-core:08724]
+
+Mon Aug 28 07:21:47 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* file.c (File#size?): Fix documentation submitted by Rick Ohnemus.
+	  ruby-Bugs-5529.  [ruby-core:08725]
+
+Sun Aug 27 21:41:23 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* oniguruma.h: Version 4.4.0
+
+	* regint.h: ditto.
+
+	* regparse.h: ditto.
+
+	* regexec.c: ditto.
+
+	* regcomp.c ditto.
+
+	* regparse.c: ditto.
+
+Sat Aug 26 08:03:03 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: updated based on date2 3.8.2.
+
+Fri Aug 25 21:15:22 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* common.mk: add regint.h and oniguruma.h to dependence.
+
+	* ext/strscan/depend: ditto.
+
+Fri Aug 25 20:35:57 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/wsdl/document/echo.rb: removed.
+
+	* test/wsdl/document/test_rpc.rb: remove echo.rb after test.
+	  [ruby-dev:29337]
+
+Fri Aug 25 17:02:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_sweep): typo fixed.
+
+Fri Aug 25 16:05:50 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (sym_call): check if the receiver is given.
+
+Fri Aug 25 01:10:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_Integer): Integer(nil) should raise TypeError.
+	  [ruby-talk:210205]
+
+	* object.c (nil_to_s): no longer returns empty string but "nil".
+	  [ruby-talk:210205]
+
+	* lib/mkmf.rb: avoid COMMON_HEADERS being nil.
+
+Wed Aug 23 00:25:14 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rexml/source.rb (REXML::IOSource#initialize): encoding have to
+	  be set with the accessor.  fixed: [ruby-list:42737]
+
+Tue Aug 22 19:21:00 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb: parameter `to_addrs' might be an Array,
+	  .flatten is required. [ruby-dev:29316]
+
+Tue Aug 22 18:47:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::handle_method):
+	  rdoc documents C module methods as instance methods. a patch in
+	  [ruby-core:08536].
+
+Tue Aug 22 12:35:57 2006  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb (Kconv::RegexpEucjp): fix regexp for
+	  euc-jp [ruby-dev:29344]
+
+Sun Aug 20 11:46:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_step): also return an enumerator object if no block
+	  is given.
+
+Sat Aug 19 16:47:51 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (hash2named_arg): accept hash argument
+	  of symbol key.
+
+	* test/win32ole/test_win32ole.rb
+	  ditto.
+
+Sat Aug 19 11:28:08 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_rename): use errno if set properly.
+	  fixed: [ruby-dev:29293]
+
+Fri Aug 18 01:05:57 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::out): specify -m0 to disable MIME decode.  a
+	  patch from Fujioka <fuj at rabbix.jp>.  [ruby-dev:29284]
+
+Thu Aug 17 19:15:16 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_stat_[rRwWxX]): check for super user.
+	  fixed: [ruby-core:08616]
+
+Thu Aug 17 14:47:06 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: added rdoc by Daniel Berger.  [ruby-core:08177]
+
+Wed Aug 16 17:46:59 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_byte): IO#getc returns one byte string now.
+	  fixed: [ruby-dev:29255]
+
+Wed Aug 16 17:22:44 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (pre-install-local): remove unnecessary code.
+	  [ruby-dev:29249]
+
+Wed Aug 16 11:45:36 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_setuid, proc_setgid, proc_seteuid, proc_setegid):
+	  get rid of bogus implementations on Mac OS X.
+
+Wed Aug 16 11:09:26 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (set_arg0): fill argv other than the first with an empty
+	  string instead of NULL.
+
+Tue Aug 15 11:21:08 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb: support SMTP/SSL.  Thanks Kazuhiro NISHIYAMA.
+
+	* lib/net/smtp.rb: new method SMTP.use_ssl?
+
+	* lib/net/smtp.rb: new method SMTP.enable_ssl.
+
+	* lib/net/smtp.rb: new method SMTP.disable_ssl.
+
+	* lib/net/smtp.rb: new method SMTP.default_ssl_port.
+
+	* lib/net/smtp.rb: new method SMTP.default_tls_port.
+
+	* lib/net/smtp.rb: now SMTP#enable_tls accepts a SSLContext
+	  object, instead of a verity and cert.  [FEATURE CHANGE]
+
+	* lib/net/smtp.rb: new method SMTP.ssl_context.
+
+	* lib/net/smtp.rb: new method SMTP.default_ssl_context.
+
+	* lib/net/smtp.rb: export SMTP.authenticate.
+
+	* lib/net/smtp.rb: export SMTP.auth_plain.
+
+	* lib/net/smtp.rb: export SMTP.auth_login.
+
+	* lib/net/smtp.rb: export SMTP.auth_cram_md5.
+
+	* lib/net/smtp.rb: export SMTP.starttls.
+
+	* lib/net/smtp.rb: export SMTP.helo.
+
+	* lib/net/smtp.rb: export SMTP.ehlo.
+
+	* lib/net/smtp.rb: export SMTP.mailfrom.
+
+	* lib/net/smtp.rb: export SMTP.rcptto.
+
+	* lib/net/smtp.rb: export SMTP.rcptto_list.
+
+	* lib/net/smtp.rb: export SMTP.data.
+
+	* lib/net/smtp.rb: export SMTP.quit.
+
+Sat Aug 12 22:33:06 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* string.c (String#split): Describe grouping behavior.  Patch by Jan
+	  Svitok <jan.svitok at gmail.com>.  [ruby-core:08603]
+
+Sun Aug 13 12:08:02 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c: ANSIfied.  [ruby-core:08601]
+
+Sat Aug 12 15:55:32 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, bcc32/Makefile.sub, win32/Makefile.sub, win32/dir.h,
+	  win32/win32.c, win32/win32.h: large file support for win32.
+
+Fri Aug 11 15:39:25 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#find_body): Make RDoc
+	  ignore C function prototypes.  Patch by Tilman Sauerbeck
+	  <tilman at code-monkey.de>.  [ruby-core:8574]
+	* lib/yaml/tag.rb: Replace nodoc with stopdoc so Module methods get
+	  documented.
+
+Wed Aug  9 16:53:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/smtp.rb (Net::SMTP::auth_cram_md5): use ord to retrieve
+	  bytes from strings.  a patch from WATANABE Tetsuya
+	  <Tetsuya.WATANABE at nifty.com>.  [ruby-dev:29240]
+
+Tue Aug  8 23:49:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/extend-command.rb (IRB::ExtendCommandBundle): pacify
+	  RDoc.  a patch from Eric Hodel <drbrain at segment7.net>.
+	  [ruby-core:08522]
+
+Tue Aug  8 19:26:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::DNS::Message::MessageDecoder::get_string):
+	  affected by str[0] returns 1 char string.  [ruby-dev:29223]
+
+	* lib/resolv.rb (Resolv::DNS::Message::MessageDecoder::get_labels):
+	  ditto.
+
+Tue Aug  8 12:28:43 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (arg): allow newlines before ternary colon.  [ruby-dev:29189]
+
+Mon Aug  7 17:56:59 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c, ext/digest/rmd160/rmd160ossl.c,
+	  ext/digest/sha1/sha1ossl.c, ext/readline/readline.c: move
+	  inclusion of config.h to pacify AIX.  a patch from Yutaka
+	  Kanemoto <kinpoco at gmail.com>.  [ruby-dev:29197]
+
+Mon Aug  7 15:55:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/syck.c (syck_move_tokens): should avoid negative
+	  memmove.  [ruby-list:42625]
+
+Mon Aug  7 14:37:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in, common.mk: AIX link issue.  a patch from Yutaka
+	  Kanemoto <kinpoco at gmail.com>.  [ruby-dev:29190]
+
+	* ext/socket/socket.c: AIX socket support.  [ruby-dev:29190]
+
+Mon Aug  7 12:05:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dln.c, eval.c, gc.c, ruby.h: shut up AIX alloca warning.
+	  a patch from Yutaka Kanemoto <kinpoco at gmail.com>.
+	  [ruby-dev:29191]
+
+Sun Aug  6 20:34:24 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (str[fp]time): %[EO]U didn't denote %U.
+
+Sun Aug  6 17:12:12 2006  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_reopen): STDERR.reopen(open("/dev/tty", "w")) should not
+	  clear FMODE_PREP in STDERR.
+
+Sat Aug  5 22:53:41 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* oniguruma.h: Version 4.2.2
+
+	* regint.h: ditto.
+
+	* regparse.h: ditto.
+
+	* regexec.c: ditto.
+
+	* regcomp.c ditto.
+
+	* regerror.c: ditto.
+
+	* regparse.c: ditto.
+
+Sat Aug  5 17:07:43 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (top_local_setup): local_vars[-1] should point
+	  ruby_scope itself to protect local_tbl from garbage collection.
+	  [ruby-dev:29049]
+
+Sat Aug  5 13:49:43 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb (str[fp]time): "%\n" means "\n".
+
+Fri Aug  4 12:13:22 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* lib: Clean up files for RDoc.
+	* lib/.document: Include most of the standard library in RDoc
+	  generation.
+	* lib/rdoc/ri/ri_formatter.rb: Don't unescape HTML in HtmlFormatter.
+	  Submitted by <ksruby at gmail.com>.  [ruby-core:08392].
+	* lib/drb/ssl.rb: Close socket on SSLError [ruby-core:7197]
+
+Fri Aug  4 18:59:49 2006  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/{init.rb,ruby-lex.rb,slex.rb}: can't input '\c' for
+	  [ruby-core: 7122]. and support for ruby1.8.X
+
+Fri Aug  4 14:02:14 2006  James Edward Gray II  <james@grayproductions.net>
+
+	* lib/date/format.rb (__strptime, strftime): allow multi-line patterns
+	  in Date#strftime the same as Time#strftime accepts.
+	  fixed: [ruby-core:08466]
+
+Fri Aug  4 13:56:51 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack): check argument overrun for 'P'.  based on a
+	  patch by rucila <rucila at yahoo.cojp>.  fixed: [ruby-dev:29182]
+
+Fri Aug  4 02:42:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): a bug in %c type check.
+
+Fri Aug  4 01:28:19 2006  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_reopen): STDERR.reopen(File.open("/dev/null", "w")) should
+	  not fclose stderr.
+
+Thu Aug  3 15:16:44 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_include): should always call Enumerable#include?
+	  (not #===) for non numeric end points.  [ruby-core:08477]
+	  [ruby-core:08496]
+
+Mon Jul 31 16:51:40 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (exit_handler): new function; release winsock and
+	  environment work area.
+
+	* win32/win32.c (NTInitialize): setup exit_handler.
+
+	* win32/win32.c (StartSockets): use exit_handler.
+
+	* win32/win32.c (rb_w32_getenv): use GetEnvironmentStrings() instead
+	  of GetEnvironmentVariable(), because the latter cannot distinguish
+	  whether a null environment variable exists or not.
+	  fixed: [ruby-talk:205123]
+
+Mon Jul 31 16:15:13 2006  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_process.rb (TestProcess#test_rlimit_nofile):
+	  setrlimit may fail with EINVAL.
+	  reported by MIYAMUKO Katsuyuki.  [ruby-dev:29174]
+
+Mon Jul 31 09:22:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h: use ifdef (or defined) for macro constants that may or
+	  may not be defined to shut up gcc's -Wundef warnings.
+	  [ruby-core:08447]
+
+Mon Jul 31 13:38:13 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPReuqest#parse_uri): improve
+	  for the value of IPv6 address in the Host: header field.
+
+Sun Jul 30 23:26:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_call0): trace call/return of method defined from block.
+	  fixed: [ruby-core:08329]
+
+	* eval.c (rb_trap_eval): make the current thread runnable to deal with
+	  exceptions which occurred within the trap.  fixed: [ruby-dev:27729]
+
+	* lib/cgi/session.rb, lib/cgi/session/pstore.rb: suppress warnings.
+	  fixed: [ruby-talk:204896]
+
+Sat Jul 29 06:12:06 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: freeze ip_name for security reason.
+
+Sat Jul 29 01:23:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/logger.rb: improves the amount of documentation that Rdoc
+	  picks up when processing logger.rb by moving the require
+	  statement back before the comment block.  a patch from Hugh
+	  Sasse <hgs at dmu.ac.uk>.  [ruby-core:08422]
+
+Fri Jul 28 17:18:03 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/curses/curses.c (NUM2CH, CH2FIX): use single char strings.
+
+Fri Jul 28 14:09:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call): fixed typo in cache look-up.  [ruby-dev:29167]
+
+Fri Jul 28 10:41:35 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call): a bug in method cache look-up.
+	  http://www.rubyist.net/~matz/20060720.html#c04
+
+Fri Jul 28 10:19:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): documentation update patch from Jacob
+	  Fugal <lukfugl at gmail.com>.  [ruby-core:08418]
+
+Fri Jul 28 09:41:45 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_to_s): fixed typo.  [ruby-dev:29162]
+
+Fri Jul 28 00:26:47 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* math.c (domain_check): ANSI style function arguments
+
+	* math.c (math_log): too few argument to domain_check().
+
+Thu Jul 27 21:19:54 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* math.c (domain_check): a new function to check domain error
+	  explicitly for systems that return NaN like FreeBSD.
+	  [ruby-core:07019]
+
+	* math.c (math_acos, math_asin, math_acosh, math_atanh, math_log,
+	  math_log10, math_sqrt): use domain_check().
+
+	* math.c (math_sqrt): fix documentation flaw.
+
+Thu Jul 27 22:21:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_to_s): fixed format mismatch.
+
+Thu Jul 27 18:12:12 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* time.c: need to declare time_utc_offset.
+
+Thu Jul 27 17:01:01 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_close): always calls "close" method of the receiver.
+	  [ruby-core:6911] [ruby-core:8112]
+
+Thu Jul 27 16:41:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/openssl/ossl.h: move <ruby.h> inclusion point to shut up
+	  Solaris compiler.  [ruby-core:08114]
+
+	* time.c (time_to_s): use +0900 style timezone string for local time.
+	  [ruby-dev:29143]
+
+Wed Jul 26 22:20:59 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: add support for as and ASFLAGS.  [ruby-dev:29138]
+
+Wed Jul 26 21:59:33 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (Net::HTTP#post, request_post, request): should
+	  set Content-Type: x-www-form-urlencoded by default.
+
+	* lib/net/http.rb (Net::HTTPHeader#content_type): should return
+	  nil when there's no Content-Type.
+
+	* lib/net/http.rb (Net::HTTPHeader#sub_type): should return nil
+	  when there's no sub Content-Type (e.g. "Content-Type: text").
+
+	* lib/net/http.rb (Net::HTTPHeader#type_params): wrongly failed
+	  when there's no Content-Type.
+
+Wed Jul 26 18:38:13 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c (strscan_do_scan): always return nil if
+	  p->curr exceeds string size.
+
+Wed Jul 26 18:33:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_eval): rename #invoke_method and
+	  #invoke_functional_method to __send and __send! respectively.
+
+	* eval.c (remove_method): prohibit removing __send and __send!.
+
+	* eval.c (rb_undef): prohibit undef'ing  __send and __send!.
+
+	* eval.c (rb_eval): prohibit redefining  __send and __send!.
+
+	* lib/delegate.rb (Delegator): preserve __send.
+
+Wed Jul 26 18:14:19 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (getDevice): retry once after GC on failure.
+	  [ruby-core:08282]
+
+Wed Jul 26 17:43:20 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c (strscan_do_scan):
+	  StringScanner.new("").scan(//) should return "". [ruby-Bugs:4361]
+
+Wed Jul 26 17:28:16 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): prepend ".." to %u for negative bignum,
+	  but not "-".  fixed: [ruby-core:08167]
+
+Wed Jul 26 16:39:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_scan): add string modification check.
+	  [ruby-core:7216]
+
+Wed Jul 26 16:06:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::read_multipart): check
+	  multipart boundary end.  a patch from Fujioka <fuj at rabbix.jp>
+	  [ruby-dev:28470]
+
+Wed Jul 26 01:02:59 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: suppress warnings by automake 1.8 or later.
+
+Tue Jul 25 14:46:14 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): typo.
+
+Tue Jul 25 13:14:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (rb_proc_times): rename hz to hertz to avoid name
+	  crash on AIX.  [ruby-dev:29126]
+
+Mon Jul 24 22:03:40 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (backtrace): skip frames successive on node and method name.
+
+Mon Jul 24 15:51:52 2006  Tanaka Akira  <akr@fsij.org>
+
+	* ext/readline/readline.c (readline_readline): rl_deprep_term_function
+	  may be NULL with libedit.  reported by Ryan Davis.  [ruby-dev:29070]
+
+Mon Jul 24 15:19:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): revert last change.  [ruby-dev:29112]
+	  [ruby-core:08374]
+
+Sun Jul 23 22:59:49 2006  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_unix.rb: disabled on cygwin.
+	  reported by Kouhei Yanagita.  [ruby-dev:29080]
+
+Fri Jul 21 23:57:26 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): script is never used while recursing.
+
+Fri Jul 21 21:21:08 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_call0): include funcalled methods in caller list.
+	  fixed: [ruby-core:08290]
+
+Fri Jul 21 17:52:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): "9_e8" should consider "_e8" as
+	  trailing garbage so that it should return 9.0.  [ruby-dev:29088]
+
+Fri Jul 21 12:11:00 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb, lib/mkmf.rb (with_destdir): remove drive letter before
+	  prepending destdir on DOSISH.
+
+Fri Jul 21 04:17:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call): try local method look-up first for fcall, then
+	  normal method look-up.  [ruby-talk:202564]
+
+	* eval.c (rb_get_method_body): save local method cache separately.
+
+	* eval.c (search_method): export info whether method is local or
+	  not.
+
+Thu Jul 20 20:27:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_mod_attr): make Module#attr to be an alias to
+	  attr_reader.  [RCR#331]
+
+Thu Jul 20 15:07:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h: export classes/modules to implement sandbox.
+	  [ruby-core:08283]
+
+Wed Jul 19 19:40:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_yield_0): should check args_args before lambda
+	  argument check.  [ruby-dev:29029]
+
+Tue Jul 18 23:53:59 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_f_system): shouldn't block SIGCHLD if it's not
+	  exist.
+
+Tue Jul 18 22:10:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (rb_f_system): block SIGCHLD during the process
+	  execution, like glibc system(3) does.  [ruby-talk:202361]
+
+Tue Jul 18 23:10:43 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (open_ifs_socket): should not use plain malloc.
+
+	* win32/win32.c (rb_w32_opendir): should not use plain realloc.
+
+Tue Jul 18 18:05:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_float.rb (TestFloat::test_strtod): update test to
+	  conform strtod change.
+
+Tue Jul 18 16:52:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (yield_under_i): argument should be passed in avalue
+	  form.  [ruby-dev:29044]
+
+Tue Jul 18 15:49:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_unpack): propagate association array to copied
+	  string.  [ruby-core:08223]
+
+	* pack.c (pack_unpack): return referenced string itself if it has
+	  same length as specified.  a patch from <nobu at ruby-lang.org>
+	  in [ruby-core:08225].
+
+	* pack.c (pack_pack): taint 'p' packed strings.
+
+Tue Jul 18 15:19:07 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* intern.h (st_foreach_safe): fix prototype.
+
+	* node.h (NODE_LMASK): bigger than long on LLP64.
+
+	* missing/vsnprintf.c (BSD__uqtoa): new function to support LLP64.
+	  all changes are derived from [ruby-dev:29045]
+
+Tue Jul 18 14:03:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer::unmount): remove
+	  inpect argument from sprintf.  [ruby-dev:29039]
+
+Tue Jul 18 10:53:37 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): limit out-of-range message.
+
+	* util.c (ruby_strtod): return end pointer even if ERANGE occurred.
+	  fixed: [ruby-dev:29041]
+
+Mon Jul 18 00:43:05 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strtod): stop at dot not followed by digits.
+	  fixed: [ruby-dev:29036]
+
+Tue Jul 18 00:01:27 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: remove LIBRUBY_SO if static linked extensions exist.
+
+Mon Jul 17 23:30:46 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_msvcrt): defaulted to msvcrt.  Workaround for a
+	  bug of cygwin 1.5.20.
+
+Mon Jul 17 22:55:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/io/wait/wait.c (io_ready_p): protoize.
+
+Mon Jul 17 13:43:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (define_swapx): should not use plain malloc.
+
+	* ext/curses/curses.c (curses_getmouse): ditto.
+
+Mon Jul 17 12:58:41 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: should use ac_cv_lib_dl_dlopen=no on MinGW.
+
+Mon Jul 17 11:47:35 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* st.c: still need to include config.h on some platforms.
+
+Sat Jul 15 01:09:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* st.c (malloc): use xmalloc/xcalloc instead of plain
+	  malloc/calloc, to detect memory allocation failure.  see
+	  <http://www.nongnu.org/failmalloc/>.
+
+Fri Jul 14 13:08:13 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: add methods for new features of latest Tcl/Tk8.5.
+
+	* ext/tk/lib/tk/namespace.rb: ditto.
+
+Fri Jul 14 02:30:12 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/monitor.rb: document patch from Hugh Sasse <hgs at dmu.ac.uk>.
+	  [ruby-core:08205]
+
+Fri Jul 14 00:10:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_pop): may cause realloc oscillation.  a patch
+	  from MORITA Naoyuki <mlgetter at kidou.sakura.ne.jp>.
+	  [ruby-dev:29028]
+
+Thu Jul 13 22:23:56 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/composite.rb: improve handling of the classname on the
+	  option database for the widget class which includes TkComposite.
+
+Thu Jul 13 00:40:57 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (FIX2LONG): returns integer of size of VALUE.
+	  [ruby-dev:29024]
+
+	* ruby.h (FIX2ULONG): ditto.
+
+Wed Jul 12 20:05:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_args): allow post mandatory arguments after optional
+	  arguments.  [ruby-dev:29014]
+
+	* parse.y (new_args_gen): allow post_args without rest_args.
+
+	* eval.c (formal_assign): ditto.
+
+	* parse.y (new_args_gen): check post argument duplication.
+
+Tue Jul 11 20:58:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h: export rb_cMethod.  [ruby-talk:201259]
+
+Tue Jul 11 19:13:33 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: remove restriction on the class of
+	  pseudo-toplevel.
+
+Tue Jul 11 18:00:57 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: security fix.
+
+Tue Jul 11 17:28:08 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_dump): need to extend len for \b.
+
+Tue Jul 11 15:29:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_int2big): use SIGNED_VALUE.  [ruby-dev:29019]
+
+	* bignum.c (rb_int2inum, rb_uint2inum): use VALUE sized integer.
+
+	* bignum.c (rb_big2long, rb_big2ulong): ditto.
+
+	* numeric.c (rb_num2long, rb_num2ulong): ditto.
+
+	* numeric.c (check_int, check_uint): ditto.
+
+	* bignum.c (rb_quad_pack): typo fixed.
+
+Tue Jul 11 13:40:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bignorm): sizeof(long) may be smaller than
+	  sizeof(VALUE).  [ruby-dev:29013]
+
+	* ruby.h (FIXNUM_MAX): fixnum may be bigger than long.
+
+	* ruby.h (SIGNED_VALUE): signed integer of size of VALUE.
+
+Mon Jul 10 23:37:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/soap/rpc/proxy.rb (Proxy::Operation::response_doc): remove
+	  splat star from return statements.
+
+	* lib/soap/rpc/proxy.rb (Proxy::Operation::response_obj): retrieve
+	  the first value from the result array if response has only one
+	  value.
+
+Mon Jul 10 22:00:00 2006  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: Allows '_' to appear within
+	  digits.  [ruby-dev:28872]
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb: Bug in to_r reported by
+	  [ruby-list:42533] fixed.
+
+Mon Jul 10 19:22:19 2006  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (gc_sweep): expand heap earlier.
+	  reported by MORITA Naoyuki.  [ruby-dev:28960]
+
+Mon Jul 10 18:59:34 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/font.rb: sorry. mistaken to patch.
+
+Mon Jul 10 18:46:52 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: make SEGV risk lower at exit.
+
+	* ext/tk/lib/tk.rb: ditto.
+
+	* ext/tk/lib/multi-tk.rb: fail to call function-style methods on slave
+	  interpreters. The strategy (MultiTkIp_PseudoToplevel_Evaluable) to
+	  fix the problem is a little tricky. You may have to take care of
+	  conflicting with it.
+
+	* ext/tk/lib/tk.rb: a little change for the pseudo-toplevel strategy.
+
+	* ext/tk/lib/tk/font.rb: ditto.
+
+	* ext/tk/lib/tk/msgcat.rb: ditto.
+
+	* ext/tk/lib/tkextlib/itk/incr_tk.rb: ditto.
+
+	* ext/tk/sample/demos-en/widget: fail to call function-style methods
+	  on sample scripts. To fix it, a strategy which similar to the way
+	  on MultiTiIp is used. Please take care when re-write and re-run a
+	  demo script on the Widget-Demo code viewer.
+
+	* ext/tk/sample/demos-jp/widget: ditto.
+
+Mon Jul 10 17:32:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/test.rb: update test suites.
+
+	* test/ruby/test_assignment.rb (TestAssignment::test_yield): ditto.
+
+	* test/ruby/test_iterator.rb (TestIterator::test_itertest): ditto.
+
+Mon Jul 10 14:43:47 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call): remove erroneously restored prot_tag->blkid
+	  initialization.  [ruby-dev:28997]  [ruby-dev:29000]
+
+Mon Jul 10 13:58:08 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* signal.c (install_nativethread_sighandler): commented out.
+
+Mon Jul 10 09:29:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_clear_cache_for_remove): clear entries for included
+	  module.  fixed: [ruby-core:08180]
+
+Mon Jul 10 02:22:58 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): should not overwrite block information in
+	  current frame.  [ruby-dev:28957]
+
+	* eval.c (rb_yield_0): retrieve proper block object from the frame
+	  record.
+
+	* eval.c (proc_alloc): return preserved block object if it's
+	  available.
+
+Mon Jul 10 01:48:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* st.h (st_data_t): use pointer sized integer for st_data_t.
+	  [ruby-dev:28988]
+
+Sun Jul  9 18:06:47 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_constant): fix for value 1 at cross compiling.
+
+	* lib/mkmf.rb (create_makefile): prevent substitution of macro
+	  definition.  fixed: http://www.yotabanana.com/lab/20060624.html#p02
+
+Sun Jul  9 07:58:48 2006  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/rdoc/parsers/parse_f95.rb: massive overhaul from Yasuhiro
+	  Morikawa including new file suffixes, function support, public
+	  variables and constants, derived-types, defined operators and
+	  assignments, namelists, and subroutine and function
+	  arguments. Truly massive.
+
+	* lib/rdoc/diagram.rb: diagrams are now cached.
+
+	* lib/irb/completion.rb: fixed a crasher when completing against
+	  an unnamed class/module.
+
+	* lib/rdoc/parsers/parse_c.rb: private comment (--/++) support in
+	  C-file rdoc.
+
+	* lib/debug.rb: minor clarification in help.
+
+	* lib/pp.rb: minor clarification on exception.
+
+Sun Jul  9 00:54:11 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (next_jump): deal with destination of next.
+	  fixed: [ruby-core:08169]
+
+Fri Jul  7 17:49:16 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_ord): extract lower byte.  fixed: [ruby-dev:28980]
+
+	* lib/jcode.rb (String#succ!): fix for 1.9.  fixed: [ruby-dev:28979]
+
+Fri Jul  7 14:05:03 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): define FUNC_STDCALL/FUNC_CDECL.
+	  from [ruby-dev:28970].
+
+Fri Jul  7 00:38:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_default): should not call default procedure if
+	  no key is given.  [ruby-list:42541]
+
+Thu Jul  6 23:30:04 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_proc_times): use sysconf(_SC_CLK_TCK) value prior to
+	  HZ and CLK_TCK.  fixed: [ruby-talk:200293]
+
+Thu Jul  6 21:50:06 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/racc/cparse/cparse.c: sync with original code, rev 1.8.
+
+	* ext/racc/cparse/cparse.c: should mark CparseParams objects.
+
+	* lib/racc/parser.rb: sync with original code, rev 1.8.
+
+	* lib/racc/parser.rb: update coding style.
+
+Wed Jul  5 05:28:45 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (block_param): should allow block argument after splat
+	  and post splat args.
+
+Wed Jul  5 01:12:19 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/ruby/test_lambda.rb (TestLambdaParameters::test_lambda_as_iterator):
+	  -> style block no longer available.  [ruby-dev:28958]
+
+Tue Jul  4 21:48:56 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (proc_options): suppress warning on DOSISH.
+
+Tue Jul  4 15:12:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call): should not set prot_tag->blkid since it would
+	  never catch breaks at this level.  [ruby-dev:28922]
+
+Tue Jul  4 04:48:36 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c: ruby 1.9 HEAD 64 bit warnings clean up from
+	  <ville.mattila at stonesoft.com>.   [ruby-core:08120]
+
+Mon Jul  3 19:04:38 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_make_menu_embeddable): help to make a menu
+	  widget embeddable (pack, grid, and so on) like as a general widget.
+	  However, an embeddable menu may require to be defined some event
+	  bindings for general use.
+
+	* ext/tk/lib/tk/event.rb: [bug fix] Tk.callback_break and
+	  Tk.callback_continue don't work on MultiTkIp.
+
+	* ext/tk/lib/multi-tk.rb: ditto.
+
+	* ext/tk/lib/tk.rb: lack of Tk.callback_return.
+
+	* ext/tk/lib/tk/menu.rb: improve creating clone menus.
+
+Mon Jul  3 14:42:06 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/extconf.rb (PW_UID2VAL, PW_GID2VAL): defaulted to conversion
+	  from int, and sys/types.h needs to be included before grp.h.
+	  fixed: [ruby-dev:28938]
+
+Mon Jul  3 10:44:01 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (popen_exec): close file descriptors other than standard I/Os.
+	  fixed: [ruby-dev:28924]
+
+Mon Jul  3 05:15:29 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/openssl/test_asn1.c: String#[]= doesn't accept Integer.
+
+Mon Jul  3 01:14:15 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_inspect): encode \b (\010) for escape.
+	  [ruby-dev:28927]
+
+	* string.c (rb_str_dump): ditto.
+
+Sun Jul  2 19:03:30 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/racc/cparse/cparse.c: sync with original code, rev 1.7.
+
+	* ext/racc/cparse/cparse.c: must require version.h to get
+	  RUBY_VERSION_CODE.
+
+Sun Jul  2 18:42:27 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/racc/cparse/cparse.c: sync with original source code, rev
+	  1.6.
+
+	* ext/racc/cparse/cparse.c: do not use rb_iterate to give a block
+	  to the method, use rb_block_call instead. [ruby-dev:28445]
+
+Sun Jul  2 11:22:03 2006  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (io_reopen): STDOUT.reopen(filename, "w+") didn't work.
+	  (rb_io_reopen): STDOUT.reopen(File.open(filename, "w+")) didn't work.
+
+Sat Jul  1 23:55:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (PUSH_FRAME): initialize frame->self.  [ruby-dev:28911]
+
+Sat Jul  1 17:00:42 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/webrick/utils.rb: use Proc#yield instead of Proc#call.
+	  [ruby-dev:28914]
+
+Sat Jul  1 15:15:49 2006  Tanaka Akira  <akr@m17n.org>
+
+	* test/socket/test_nonblock.rb: add timeout to send/receive
+	  an empty UDP packet.
+	  [ruby-dev:28820]
+
+Fri Jun 30 23:46:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: should test isinf for Solaris with GCC compiler.
+	  a patch from <ville.mattila at stonesoft.com>.  [ruby-core:07791]
+
+	* configure.in: -shared patch from Andrew Morrow
+	  <andrew.c.morrow at gmail.com>.  [ruby-core:08100]
+
+Fri Jun 30 19:35:41 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils._escape): should
+	  use String#ord to get ascii code from the one-character string.
+	  [ruby-dev:28901]
+
+Thu Jun 29 23:56:01 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_mark_children): a bug in NODE_BLOCK_PASS marking.
+	  [ruby-dev:28908]
+
+Thu Jun 29 23:04:36 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: use ARGSPUSH instead of ARGSCAT to prevent too much
+	  splat expansion.
+
+	* eval.c (when_check): need to handle ARGSPUSH as well.
+
+	* eval.c (block_orphan): lambda and proc from method are always
+	  orphan.
+
+	* gc.c (gc_mark_children): proper marking for NODE_LAMBDA.
+
+Thu Jun 29 22:47:30 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (SETUP_ARGS0): avoid GC problem.
+	  [ruby-dev:28902]
+
+Thu Jun 29 18:58:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_version): fix patch
+	  failure.
+
+Thu Jun 29 18:00:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: add RDoc document.  a patch from
+	  mathew <meta at pobox.com>.   [ruby-core:07050]
+
+Wed Jun 28 14:53:09 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/optparse.rb: RDoc patch from Robin Stocker <robin@nibor.org>
+	  [ruby-core:08087]
+
+Wed Jun 28 23:23:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): underscores should appear only
+	  between digits.  [ruby-dev:28891]
+
+Wed Jun 28 19:04:34 2006  Tanaka Akira  <akr@m17n.org>
+
+	* test/socket/test_unix.rb: test_seqpacket_pair removed.
+	  [ruby-dev:28846]
+
+Wed Jun 28 13:51:21 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (when_check): arbitrary values are allowed after splats.
+	  fixed: [ruby-dev:28879]
+
+Wed Jun 28 09:16:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (primary): remove meaningless else-only case statement
+	  syntax.
+
+Wed Jun 28 08:08:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): problem to handle else part.  [ruby-dev:28873]
+
+Wed Jun 28 01:48:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): support splat in when expression list.
+	  [ruby-dev:28822]
+
+	* eval.c (when_check): a new auxiliary function for case match.
+
+	* eval.c (when_cond): ditto.
+
+Wed Jun 28 01:05:37 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_cstr_to_dbl): should not skip '_' at the beginning
+	  of a string.  [ruby-dev:28830]
+
+	* bignum.c (rb_cstr_to_inum): ditto.
+
+Tue Jun 27 23:03:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c: RDoc update for =~ method.  a patch from Alex Young
+	  <alex at blackkettle.org>.  [ruby-core:08068]
+
+Tue Jun 27 22:47:18 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: forgot to update TCLTKLIB_RELEASE_DATE.
+
+	* ext/tk/lib/tk.rb (tk_tcl2ruby): [bug fix] sometimes fail to convert
+	  a tcl string to a ruby object if the tcl string includes "\n".
+
+Tue Jun 27 20:05:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (pipe_open): backout unnecessary fix on 2006-06-26.
+	  [ruby-dev:28865]
+
+	* eval.c (rb_yield_0): exact argument number check now done only
+	  for lambda Proc.
+
+	* eval.c (rb_yield_0): add check for number of arguments, if
+	  there's one lambda block parameter.
+
+Tue Jun 27 16:04:05 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* win32/win32.h: define isascii on MinGW for msvcrt compatibility.
+
+	* configure.in: set ac_cv_header_sys_time_h=no on MinGW
+	  for msvcrt compatibility.
+
+Tue Jun 27 11:36:02 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (setup_passwd, setup_group): allow bignum uid, gid and
+	  so on.  [ruby-talk:199102]
+
+Tue Jun 27 10:46:53 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (rb_yield_0): avoid core dump.  [ruby-dev:28840]
+
+Mon Jun 26 11:03:00 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri: Add options to limit the ri search path.
+
+Tue Jun 27 01:31:59 2006  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/socket.c (bsock_recv_nonblock): new method
+	  BasicSocket#recv_nonblock.
+	  (udp_recvfrom_nonblock): renamed from ip_recvfrom_nonblock.
+	  IPSocket#recvfrom_nonblock is moved to UDPSocket#recvfrom_nonblock.
+	  (unix_recvfrom_nonblock): removed.
+	  UNIXSocket#recvfrom_nonblock is removed.
+
+Tue Jun 27 00:52:40 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/ripper/eventids2.c (token_assoc): added tCHAR, which is not
+	  under 256 now.  fixed: [ruby-dev:28832]
+
+Mon Jun 26 23:42:57 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (call_trace_func): no check for argument number of the
+	  callback.  fixed: [ruby-dev:28812]
+
+Mon Jun 26 18:37:44 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_delete): fix SEGV when a slave-ip is
+	  deleted on callback.
+
+Mon Jun 26 15:40:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_accept): revert to avoid ambiguity of
+	  argument evaluation order.  [ruby-dev:28861]
+
+	* ext/socket/socket.c (sock_accept_nonblock): ditto.
+
+Mon Jun 26 10:47:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (pipe_open): avoid closing uninitialized file descriptors.
+	  a patch from <tommy at tmtm.org> [ruby-dev:28600]
+
+Sun Jun 25 23:02:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, mkconfig.rb: catch-up for latest autoconf.
+
+Sun Jun 25 17:44:16 2006  Tanaka Akira  <akr@m17n.org>
+
+	* parse.y (paren_args): wrap $2 by escape_Qundef because it may be
+	  Qundef.  [ruby-dev:28843]
+
+Sun Jun 25 17:18:33 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(ole_invoke): support some kind of
+	  method of word. [ruby-Bugs#3237]
+
+	* test/win32ole/test_word.rb: ditto.
+
+Sat Jun 24 23:48:08 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: replace terminal token names with more descriptive
+	  name, i.e. kEND to keyword_end.  [ruby-list:42477]
+
+Sat Jun 24 23:37:41 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (rb_eval): use rb_ary_new2 instead of rb_ary_new4 to avoid
+	  GC problem.
+	  (rb_yield_values): use rb_ary_new2 instead of rb_ary_new4.
+
+	* array.c (rb_ary_new4): don't set len as n if contents is not
+	  initialized.  make it safe with GC.
+
+	  [ruby-dev:28826]
+
+Fri Jun 23 23:35:32 2006  Tanaka Akira  <akr@m17n.org>
+
+	* ruby.h, lib/drb/drb.rb, lib/drb/invokemethod.rb: remove Values class.
+	  [ruby-dev:28805]
+
+Fri Jun 23 17:27:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_block_pass): removed.
+
+	* eval.c (rb_thread_start_1): use rb_proc_yield() instead of
+	  rb_block_pass().  fixed: [ruby-dev:28794]
+
+Thu Jun 22 11:52:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTPResponse): duplicated error 501;
+	  HTTPInternalServerError should be error 500.  [ruby-core:08037]
+
+Thu Jun 22 11:47:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_mod_name): returns nil for anonymous modules.
+	  [ruby-talk:198440]
+
+Thu Jun 22 10:31:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_aref): "abc"[3] should not return an empty
+	  string but nil.  [ruby-dev:28786]
+
+Thu Jun 22 05:15:58 2006  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/socket.c (sock_s_socketpair): try GC only once.
+	  [ruby-dev:28778]
+
+Wed Jun 21 21:20:31 2006  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (jd_to_commercial): now works fine even if in
+	  mathn-ized context.
+
+Wed Jun 21 17:29:57 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/getaddrinfo.c (freeaddrinfo, get_name): fixed typo.
+
+	* ext/tk/tcltklib.c (tcl_eval, tcl_global_eval): ditto.
+
+	* ext/zlib/zlib.c (rscheck): constified.
+
+Wed Jun 21 17:18:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/pp.rb (PP::PPMethods::seplist): should have preserved
+	  original reference to the array.  [ruby-dev:28747]
+
+Wed Jun 21 14:35:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (block_param): do not use multiple assignment for a sole
+	  block parameter.  [ruby-dev:28710]
+
+	* eval.c (rb_yield_0): pass a raw yielded value to a sole block
+	  parameter if a value is passed by yield.
+
+	* eval.c (proc_invoke): args may not be an array.
+
+	* eval.c (rb_proc_yield): pass original value without wrapping
+	  it in an array.
+
+Wed Jun 21 14:06:47 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (method_call): remove (fn)(args) style lambda
+	  invocation, add fn.(args) instead.
+
+Wed Jun 21 08:39:54 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/xmlrpc/create.rb (XMLRPC::Create::conv2value): merge Date
+	  and Time processing.  [ruby-core:08033]
+
+Wed Jun 21 03:01:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c, file.c, etc.: code-cleanup patch from Stefan Huehner
+	  <stefan at huehner.org>.  [ruby-core:08029]
+
+Wed Jun 21 01:40:25 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reswords): modifier token is no longer returned in fname
+	  state.  fixed: [ruby-dev:28775]
+
+Tue Jun 20 23:28:34 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (parse_args): provisional catch-up for the recent changes.
+
+	* lib/optparse.rb (OptionParser::List#summarize, OptionParser#order!): ditto.
+
+Tue Jun 20 11:07:55 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): intercept break and return from lambda
+	  Proc objects.  [ruby-dev:28742]
+
+	* eval.c (proc_invoke): remove unnecessary YIELD_PROC_CALL flag.
+
+	* eval.c (YIELD_EXACT_ARGS): renamed from YIELD_LAMBDA_CALL, which
+	  is no longer related to the behavior turned on by this flag.
+
+	* eval.c (return_jump): no need to care about PROT_YIELD.
+
+	* eval.c (break_jump): no jump to toplevel PROT_THREAD tag.
+
+	* eval.c (rb_yield_0): fix confusion between lambda (which is a
+	  property of a proc) and pcall (which depends on whether it's
+	  called via yield or call).
+
+	* eval.c (rb_thread_yield): no need to specify YIELD_LAMBDA_CALL.
+
+	* eval.c (rb_block_pass): update blkid in prot_tag.
+
+Mon Jun 19 23:40:59 2006  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	  * ext/nkf/lib/kconv.rb: remove default -m0 and fix document.
+
+	  * ext/nkf/nkf-8/{nkf.c, config.h, utf8tbl.c, utf8tbl.h}:
+	    imported nkf 2.0.7.
+
+Mon Jun 19 17:02:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/test.rb (proc_return3): return within non lambda block
+	  should terminate surrounding method.  [ruby-dev:28741]
+
+Mon Jun 19 13:22:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (unix_sysaccept): typo fixed.
+
+	* ext/socket/socket.c (sock_connect): remove an unused local
+	  variable tmpaddr.
+
+Mon Jun 19 02:10:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (tcp_accept_nonblock): forgot to remove
+	  abandoned hacks.  [ruby-dev:28740]
+
+Mon Jun 19 00:00:17 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: fix bug: initialize improper tables.
+
+Sun Jun 18 20:28:43 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fole_methods): The return value
+	  of WIN32OLE#ole_methods should include PROPERTYPUTREF methods.
+
+	* ext/win32ole/win32ole.c (fole_put_methods): The return value
+	  of WIN32OLE#ole_put_methods should include PROPERTYPUTREF methods.
+
+	* test/win32ole/test_ole_methods.rb: ditto.
+
+	* ext/win32ole/win32ole.c (ole_propertyput): support
+	  PROPERTYPUTREF. [ruby-talk:183042]
+
+	* test/win32ole/test_propertyputref.rb: ditto.
+
+Sat Jun 17 23:42:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_eval): add aliases invoke_method and
+	  invoke_functional_method corresponding send and funcall
+	  respectively.  [ruby-talk:197512]
+
+	* parse.y (parser_yylex): returns the most typical keyword token
+	  on EXPR_FNAME.  [ruby-core:7995]
+
+	* ext/socket/socket.c: protoize.
+
+Sat Jun 17 22:17:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mathn.rb (Integer::prime_division): raise ZeroDivisionError
+	  on zeros.  [ruby-dev:28739]
+
+Sat Jun 17 14:53:32 2006  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Kernel#Pathname): new method.
+
+Sat Jun 17 02:01:00 2006  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb (Kernel#pretty_inspect): defined for pretty printed
+	  string.
+
+Fri Jun 16 01:41:00 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_proc_arity): get rid of segfault for mere splat.
+
+	* gc.c (gc_mark_children): NODE_BLOCK_PASS needs u3 to be marked.
+
+Thu Jun 15 22:06:56 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (then): remove ':' from 'then' and 'do' rules.
+
+Wed Jun 14 18:00:20 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* enum.c (enum_any): Documentation typo.
+
+Wed Jun 14 15:01:09 2006  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser#warn): Don't print
+	  warnings when -q is set.
+
+Wed Jun 14 16:11:37 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (rb_f_method_name, rb_f_callee_name): document typo.
+
+Wed Jun 14 15:19:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (env_aset): raise TypeError on nil with more descriptive
+	  message.  [ruby-core:07990]
+
+Tue Jun 13 17:22:19 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (Init_socket): remove obsolete constants:
+	  IPsocket, TCPsocket, SOCKSsocket, TCPserver, UDPsocket,
+	  UNIXsocket, UNIXserver.
+
+Tue Jun 13 09:07:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (formal_assign): post splat arguments should have had
+	  higher priority than optional arguments, since they are
+	  mandatory.  [ruby-dev:28715]
+
+	* eval.c (VIS_MASK): broken. should be 15.  [ruby-dev:28715]
+
+	* io.c (argf_getc): should return one-character string.
+	  [ruby-dev:28715]
+
+	* io.c (rb_io_readchar): ditto.
+
+Sun Jun 11 23:20:07 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (sym_call): disallow to call private methods.
+
+	* lib/optparse.rb (OptionParser::Arguable#getopts): pass self to the
+	  parser.
+
+Sun Jun 11 09:56:41 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.h (write): not need to define on bcc.
+
+Sun Jun 11 08:30:33 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#getopts): new methods.
+
+Sun Jun 11 07:27:11 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/ri/ri_writer.rb: use String#ord.
+
+Sun Jun 11 04:38:20 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (sym_to_proc): imported Symbol#to_proc from ActiveSupport.
+
+Sat Jun 10 18:02:40 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/newton.rb (Newton::nlsolve): typo
+	  fixed: raize -> raise.  [ruby-talk:196608]
+
+Sat Jun 10 17:49:53 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_ord): new method.
+
+	* parse.y (rbracket): allow optional newline before closing
+	  brackets.
+
+Sat Jun 10 15:12:29 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (rb_f_method_name, rb_f_callee_name): new functions.
+	  new global method `__method__' and `__callee__'.
+
+Sat Jun 10 10:13:13 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/getoptlong.rb (GetoptLong#set_options): receive arguments
+	  as Array.
+
+	* lib/irb/slex.rb: use Proc#yield.
+
+	* lib/rdoc/markup/simple_markup/inline.rb: follow the new behavior
+	  of String#[].
+
+	* lib/rdoc/ri/ri_writer.rb: ditto.
+
+Sat Jun 10 08:17:23 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (log2): may be a macro.
+
+	* parse.y (args, block_param, f_args): pass f_post_arg to #params.
+
+	* util.c (powersOf10): constified.
+
+	* ext/readline/readline.c: include extconf.h first.
+
+	* ext/ripper/eventids2.c: removed tLAMBDA_ARG.
+
+	* ext/tk/tcltklib.c (lib_fromUTF8_core): removed conflict.
+
+	* ext/tk/tkutil/tkutil.c (cbsubst_get_subst_arg): rb_id2name() is
+	  defined as const now.
+
+	* ext/win32ole/win32ole.c (fole_missing): ditto.
+
+	* lib/mkmf.rb (create_makefile): force to create extconf header.
+
+	* lib/optparse.rb (order!): use Proc#yield.
+
+Sat Jun 10 06:53:22 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (CALLARGS): remove last semicolon. C90 compiler doesn't
+	  allow any lines (even if they're empty) within variable
+	  declarations.
+
+Fri Jun  9 09:56:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): allow %c to print one character
+	  string (e.g. ?x).
+
+Thu Jun  8 14:00:02 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.[ch] (rb_w32_read, rb_w32_write): new functions.
+	  use recv() and send() when fd is socket. fixed: [ruby-dev:28694]
+
+Wed Jun  7 16:22:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile::make_tmpname): put dot between
+	  basename and pid.  [ruby-talk:196272]
+
+Wed Jun  7 16:16:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (do_block): remove -> style block.
+
+	* parse.y (parser_yylex): remove tLAMBDA_ARG.
+
+Wed Jun  7 14:51:22 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (errmap): add some winsock errors.
+
+Wed Jun  7 09:14:44 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): binding for the return event hook should have
+	  consistent scope.  [ruby-core:07928]
+
+Tue Jun  6 23:25:49 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): return behavior should depend whether it
+	  is surrounded by a lambda or a mere block.
+
+Mon Jun  5 18:12:12 2006  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/socket.c (sock_s_unpack_sockaddr_in): reject
+	  non-AF_INET/AF_INET6 sockaddr.
+	  (sock_s_unpack_sockaddr_un): reject non-AF_UNIX sockaddr.
+	  [ruby-dev:28691]
+
+Sun Jun  4 20:40:19 2006  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/socket.c: fix sockaddr_un handling.
+	  [ruby-dev:28677]
+
+Sat Jun  3 23:53:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (formal_assign): handles post splat arguments.
+
+	* eval.c (rb_call0): ditto.
+
+Sat Jun  3 13:10:41 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* st.c (strhash): use FNV-1a hash.
+
+Fri Jun  2 20:01:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): removed experimental ';;' terminator.
+
+Fri Jun  2 19:00:40 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: use create_header.
+
+	* ext/openssl/ossl.h, ext/openssl/openssl_missing.h:
+	  include RUBY_EXTCONF_H.
+
+Fri Jun  2 17:16:52 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (CLEANINGS): remove extconf.h by distclean if created.
+
+Fri Jun  2 00:11:19 2006  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/socket.c (s_recvfrom): alen may be zero with UNIXSocket
+	  too.  (tested on NetBSD 3.0)
+	  (s_recvfrom_nonblock): extracted from sock_recvfrom_nonblock.
+	  (sock_recvfrom_nonblock): use s_recvfrom_nonblock.
+	  (ip_recvfrom_nonblock): new method: IPSocket#recvfrom_nonblock
+	  (unix_recvfrom_nonblock): new method: UNIXSocket#recvfrom_nonblock
+	  (s_accept_nonblock): extracted from sock_accept_nonblock.
+	  (sock_accept_nonblock): use s_accept_nonblock.
+	  (tcp_accept_nonblock): new method: TCPServer#accept_nonblock
+	  (unix_accept_nonblock): new method: UNIXServer#accept_nonblock
+
+Thu Jun  1 19:12:37 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_cmdvector): backslashes inside single-quotes
+	  no longer has special meanings.  fixed: [ruby-list:42311]
+
+Thu Jun  1 17:55:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_node_arity): should be aware of post splat arguments.
+
+	* eval.c (rb_proc_arity): ditto.
+
+Thu Jun  1 16:17:26 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_getcwd): runtime's getcwd() will not success
+	  if the length of the cwd is longer than MAX_PATH.
+	  fixed [ruby-list:42335]
+
+Thu Jun  1 16:07:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_args): syntax rule enhanced to support arguments
+	  after the splat.
+
+	* parse.y (mlhs_basic): ditto for multiple assignments
+
+	* parse.y (block_param): ditto for block parameters.
+
+	* parse.y (f_post_arg): mandatory formal arguments after the splat
+	  argument.
+
+	* parse.y (new_args_gen): generate nodes for mandatory formal
+	  arguments after the splat argument.
+
+	* eval.c (rb_eval): dispatch mandatory formal arguments after the
+	  splat argument.
+
+Thu Jun  1 11:33:32 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_getcwd): set errno if not set.
+	  fixed [ruby-list:42346]
+
+Thu Jun  1 00:45:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (args): allow more than one splat in the argument list.
+
+Wed May 31 18:38:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (method_call): allow aref [] to accept all kind of
+	  method argument, including assocs, splat, and block argument.
+
+	* eval.c (SETUP_ARGS0): prepare block argument as well.
+
+Tue May 30 18:13:53 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mathn.rb (Integer): remove Integer#gcd2. [ruby-core:07931]
+
+Mon May 29 22:40:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (error_line): print receivers true/false/nil specially.
+
+	* eval.c (rb_proc_yield): handles parameters in yield semantics.
+
+	* eval.c (nil_yield): gives LocalJumpError to denote no block
+	  error.
+
+	* io.c (rb_io_getc): now takes one-character string.
+
+Sat May 27 22:46:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): save and restore block in the current frame.
+	  fixed: [ruby-core:07833], [ruby-talk:191639]
+
+Sat May 27 11:29:46 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): remove extinit files if no statically linked
+	  extensions.
+
+Fri May 26 19:56:46 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_hash): use FNV-1a hash from Fowler/Noll/Vo
+	  hashing algorithm.
+
+Fri May 26 09:05:11 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.h, lib/mkmf.rb (create_header): clear command line options for
+	  macros moved to extconf.h.
+
+	* ext/extmk.rb (extract_makefile, extmk): made RUBY_EXTCONF_H and
+	  EXTSTATIC permanent.
+
+	* ext/{dbm,digest/*,socket,zlib}/extconf.rb: used $defs and $INCFLAGS.
+
+	* {bcc32,win32,wince}/Makefile.sub (COMPILE_C, COMPILE_CXX): added
+	  $(INCFLAGS).
+
+	* lib/mkmf.rb (configuration): add $defs unless extconf.h was created.
+
+Thu May 25 01:52:07 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (pkg_config): particular config commands support.
+
+	* ext/extmk.rb: deal with $static set in extconf.rb.
+
+	* mkconfig.rb: merge multiple entries to an entry with multiple lines.
+
+	* lib/mkmf.rb: allow a series of commands to link.
+
+	* win32/Makefile.sub: embed manifests.
+
+	* win32/setup.mak: suffix OS name by runtime version.
+
+Wed May 24 23:52:11 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_install_sh): ignore dummy install-sh.
+	  [ruby-talk:193876]
+
+Wed May 24 17:55:13 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_aref): str[0] now returns 1 character string,
+	  instead of a fixnum.	[Ruby2]
+
+	* parse.y (parser_yylex): ?c now returns 1 character string,
+	  instead of a fixnum.	[Ruby2]
+
+	* string.c (rb_str_aset): no longer support fixnum insertion.
+
+Wed May 24 03:10:44 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb
+	  (OpenSSL::SSL::SocketForwarder#setsockopt,getsockopt): typo fixed.
+
+Mon May 22 16:32:03 2006  Tanaka Akira  <akr@m17n.org>
+
+	* rubyio.h (rb_io_set_nonblock): declared.
+
+	* io.c (rb_io_set_nonblock): new function.
+	  (io_getpartial): nonblocking read support.
+	  (io_read_nonblock): new method: IO#read_nonblock.
+	  (io_write_nonblock): new method: IO#write_nonblock.
+
+	* ext/socket/socket.c (s_accept): retry for EWOULDBLOCK.
+	  revert [ruby-talk:113807].
+	  (sock_connect_nonblock): new method: Socket#connect_nonblock.
+	  (sock_accept_nonblock): new method: Socket#accept_nonblock.
+	  (sock_recvfrom_nonblock): new method: Socket#recvfrom_nonblock.
+
+	  [ruby-core:7917]
+
+Mon May 22 15:57:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (umethod_bind): should not update original class.
+	  [ruby-dev:28636]
+
+Mon May 22 13:38:57 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (ev_const_get): should support constant access from
+	  within instance_eval().  [ruby-dev:28327]
+
+Sun May 21 09:50:31 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* regexec.c: add STK_NULL_CHECK_END to IS_TO_VOID_TARGET().
+	  [ruby-list:42234]
+
+Thu May 18 22:37:20 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/config.rb (WEBrick::Config::HTTP): add new parameters,
+	  :InputBufferSize and :OutputBufferSize.
+
+	* lib/webrick/utils.rb (WEBrick::Utils.timeout): add new timeout
+	  method. this implementation is expected to be compatible with
+	  timeout.rb and faster than timeout.rb.
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#_read_data):
+	  Timeout.timeout is replaced by WEBrick::Utils.timeout.
+
+	* lib/webrick/httprequest.rb: WEBrick::HTTPRequest::BUFSIZE is
+	  replaced by config[:InputBufferSize].
+
+	* lib/webrick/httpresposne.rb: WEBrick::HTTPResponse::BUFSIZE is
+	  replaced by config[:OutputBufferSize].
+
+	* lib/webrick/server.rb: get rid of unnecessary require.
+
+	* test/webrick/test_utils.rb: test for WEBrick::Utils.timeout.
+
+Thu May 18 17:51:32 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_timeval): should round for usec floating
+	  number.  [ruby-core:07896]
+
+	* time.c (time_add): ditto.
+
+Thu May 18 00:42:12 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb, lib/mkmf.rb: use BUILD_FILE_SEPARATOR in Makefiles.
+
+Wed May 17 17:55:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (sys_warning): should not call a vararg function
+	  rb_sys_warning() indirectly.	[ruby-core:07886]
+
+Tue May 16 17:23:19 2006  Shin-ichiro HARA  <sinara@blade.nagaokaut.ac.jp>
+
+	* numeric.c (flo_divmod): the first element of Float#divmod should
+	  be an integer. [ruby-dev:28589]
+
+	* test/ruby/test_float.rb: add tests for divmod, div, modulo and remainder.
+
+Tue May 16 15:34:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_initialize): should not allow modifying literal
+	  regexps.  frozen check moved from rb_reg_initialize_m as well.
+
+Tue May 16 09:20:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_initialize): should not modify untainted objects in
+	  safe levels higher than 3.
+
+	* re.c (rb_memcmp): type change from char* to const void*.
+
+	* dir.c (dir_close): should not close untainted dir stream.
+
+	* dir.c (GetDIR): add tainted/frozen check for each dir operation.
+
+Mon May 15 21:37:12 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* re.c (rb_reg_prepare_re): don't use onig_recompile().
+
+Mon May 15 17:42:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_symbol_arg):
+	  typo fixed.  a patch from Florian Gross <florg at florg.net>.
+
+Sat May 13 16:14:05 2006  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb (PP.mcall): new method.
+	  (Struct#pretty_print): call Kernel#class and Struct#members even if
+	  overridden.
+	  (Struct#pretty_print_cycle): ditto.
+	  [ruby-core:7865]
+
+Fri May 12 15:54:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (EXEC_EVENT_HOOK): trace_func may remove itself from
+	  event_hooks.	no guarantee for arbitrary hook deletion.
+	  [ruby-dev:28632]
+
+Thu May 11 19:57:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* util.c (ruby_strtod): differ addition to minimize error.
+	  [ruby-dev:28619]
+
+Thu May 11 18:30:11 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_cipher.c (add_cipher_name_to_ary): should return
+	  value. [ruby-dev:28627]
+
+Thu May 11 18:10:43 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* util.c (ruby_strtod): should not raise ERANGE when the input
+	  string does not have any digits.  [ruby-dev:28629]
+
+Wed May 10 23:40:21 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* oniguruma.h: Version 4.0.3
+
+	* regexec.c: ditto.
+
+Mon May  8 09:10:31 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: add check for OBJ_NAME_do_all_sorted.
+
+	* ext/openssl/ossl_cipher.c (ossl_s_ciphers): new method
+	  OpenSSL::Cipher.ciphers. it returns all the cipher names.
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_init): refine warning message.
+
+	* ext/openssl/lib/openssl/cipher.rb: reimplement without eval() and
+	  add constants AES128, AES192, AES256. [ruby-dev:28610]
+
+	* ext/openssl/lib/openssl/digest.rb: reimplement without eval().
+
+	* test/openssl/test_cipher.rb, test_digest: fix about reimplemented
+	  features.
+
+	* sample/openssl/cipher.rb: rewrite all.
+
+Sun May  7 03:09:51 2006  Stephan Maka  <stephan@spaceboyz.net>
+
+	* lib/resolv.rb (Resolv::DNS::Requester::ConnectedUDP#initialize):
+	  Use AF_INET6 for nameservers containing colons.
+
+Sat May  6 23:40:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): should restore old ruby_frame->block.
+	  thanks to ts <decoux at moulon.inra.fr>.  [ruby-core:07833]
+	  also fix [ruby-dev:28614] as well.
+
+Sat May  6 00:38:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (trap): sig should be less then NSIG.  Coverity found
+	  this bug.  a patch from Kevin Tew <tewk at tewk.com>.
+	  [ruby-core:07823]
+
+Thu May  4 22:13:22 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* math.c (math_log2): add new method inspired by
+	  [ruby-talk:191237].
+
+	* math.c (math_log): add optional base argument to Math::log().
+	  [ruby-talk:191308]
+
+Thu May  4 02:24:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/emitter.c (syck_scan_scalar): avoid accessing
+	  uninitialized array element.	a patch from Pat Eyler
+	  <rubypate at gmail.com>.  [ruby-core:07809]
+
+	* array.c (rb_ary_fill): initialize local variables first.  a
+	  patch from Pat Eyler <rubypate at gmail.com>.	 [ruby-core:07810]
+
+	* ext/syck/yaml2byte.c (syck_yaml2byte_handler): need to free
+	  type_tag.  a patch from Pat Eyler <rubypate at gmail.com>.
+	  [ruby-core:07808]
+
+Wed May  3 02:12:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (make_hostent_internal): accept ai_family
+	  check from Sam Roberts <sroberts at uniserve.com>.
+	  [ruby-core:07691]
+
+Mon May  1 17:58:16 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (add_event_call_back): should not
+	  delete event handler when the event name is not entried.
+
+Mon May  1 08:32:10 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_param_ole_type): should return
+	  "unknown type" string when ITypeInfo::GetFuncDesc failed.
+
+Sat Apr 29 22:43:37 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_decode0): should initialize
+	  flag. [ruby-core:07785]
+
+Fri Apr 28 10:53:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* util.c (ruby_strtod): should not cut off 18 digits for no
+	  reason.  [ruby-core:07796]
+
+Thu Apr 27 01:38:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_fill): internalize local variable "beg" to
+	  pacify Coverity.  [ruby-core:07770]
+
+Wed Apr 26 16:59:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_unpack): now supports CRLF newlines.  a patch from
+	  <tommy at tmtm.org>.	[ruby-dev:28601]
+
+Wed Apr 26 16:55:19 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* applied code clean-up patch from Stefan Huehner
+	  <stefan at huehner.org>.  [ruby-core:07764]
+
+Tue Apr 25 18:00:05 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (delete_slaves): maybe increment the reference
+	  count of a NULL Tcl_Obj [ruby-core:07759].
+
+Tue Apr 25 07:55:31 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/jcode.rb (String::tr_s): should have translated non
+	  squeezing character sequence (i.e. a character) as well.  thanks
+	  to Hiroshi Ichikawa <gimite at gimite.ddo.jp> [ruby-list:42090]
+
+Fri Apr 21 15:19:13 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (lib_eventloop_ensure): avoid dereferencing
+	  freed pointer [ruby-core:07744] and memory leak.
+
+Fri Apr 21 12:14:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c: document update patch from Sam Roberts
+	  <sroberts at uniserve.com>.  [ruby-core:07701]
+
+Thu Apr 20 08:43:54 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mathn.rb (Integer): need not to remove gcd2.  a patch from
+	  NARUSE, Yui <naruse at airemix.com>.	[ruby-dev:28570]
+
+Wed Apr 19 13:55:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (arg): too much NEW_LIST()
+
+	* eval.c (SETUP_ARGS0): remove unnecessary access to nd_alen.
+
+Wed Apr 19 11:57:04 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): use ARGSCAT for NODE_OP_ASGN1.
+	  [ruby-dev:28585]
+
+	* parse.y (arg): use NODE_ARGSCAT for placeholder.
+
+Wed Apr 19 11:13:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/getoptlong.rb (GetoptLong::get): RDoc update patch from
+	  mathew <meta at pobox.com>.  [ruby-core:07738]
+
+Wed Apr 19 10:13:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_const_set): raise error when no target klass is
+	  supplied.  [ruby-dev:28582]
+
+Tue Apr 18 17:40:37 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: add a binding to a container for a slave IP.
+
+	* ext/tk/lib/tk.rb: update RELEASE_DATE.
+
+	* ext/tk/tcltklib.c: forget to reset a Tcl interpreter.
+
+	* ext/tk/stubs.c: fix potential bugs about handling rb_argv0.
+
+Mon Apr 10 01:03:10 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* prec.c (prec_prec_f): documentation patch from
+	  <gerardo.santana at gmail.com>.  [ruby-core:07689]
+
+Sat Apr  8 02:34:34 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_pow): second operand may be too big even if
+	  it's a Fixnum.  [ruby-talk:187984]
+
+Sat Apr  8 02:12:38 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* README.EXT: update symbol description.  [ruby-talk:188104]
+
+Sat Apr  8 18:06:28 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_METHOD#inspect,
+	  WIN32OLE_PARAM#inspect.
+
+	* test/win32ole/test_win32ole_method.rb: ditto.
+
+	* add test/win32ole/test_win32ole_param.rb.
+
+Fri Apr  7 22:11:30 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(foletypelib_initialize): WIN32OLE_TYPELIB.new
+	  accepts OLE file.
+
+	* test/win32ole/test_win32ole_typelib.rb(test_initialize): ditto.
+
+Thu Apr  6 23:28:47 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* COPYING: explicitly note GPLv2.  [ruby-talk:187922]
+
+Thu Apr  6 16:43:06 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* intern.h (rb_obj_instance_exec, rb_mod_module_exec): add declaration.
+
+Thu Apr  6 11:18:37 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/panedwindow.rb: lack of arguments. [ruby-core:7681]
+
+Thu Apr  6 01:04:47 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: fix SEGV when embedding to an application.
+	  [ruby-core:7600]
+
+	* ext/tk/tcltklib.c: fix SEGV at exit. [ruby-talk:186489]
+
+	* ext/tk/tkutil/tkutil.c: follow to changing specification of
+	  instance_eval on ruby-1.9.x.
+
+	* ext/tk/lib/tk.rb: ditto.
+
+	* ext/tk/lib/multi-tk.rb: ditto.
+
+	* ext/tk/lib/tk.rb: remove warning about redefinition of methods.
+
+	* ext/tk/lib/tk/variable.rb: remove warning about unseting Tcl
+	  variables.
+
+Wed Apr  5 00:22:54 2006  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb: use a subclass for instantiation except
+	  methods take pathname argument.  suggested by Evan Phoenix.
+	  [ruby-core:7618]
+
+Tue Apr  4 22:15:41 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: remove some obsolete syntax rules (unparenthesized
+	  method calls in argument list).
+
+Sat Apr  1 15:11:27 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_TYPE#inspect,
+	  WIN32OLE_VARIABLE#inspect
+
+	* remove ext/win32ole/tests/testOLEVARIABLE.rb, testOLETYPE.rb
+	  testOLETYPELIB.rb.
+
+	* testall.rb: ditto.
+
+	* add test/win32ole
+
+Fri Mar 31 14:24:55 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_with_index): removed suspicious return
+	  statement.
+
+Wed Mar 29 23:06:48 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): change the behavior of
+	  WIN32OLE#[], WIN32OLE#[]=. These methods invoke DISPID_VALUE.
+
+	* ext/win32ole/sample/excel2.rb: ditto.
+
+	* ext/win32ole/tests/testWIN32OLE.rb: ditto.
+
+Wed Mar 29 10:11:31 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_each_char_to_hex, encode_fallback_subchar,
+	  e2w_conv): support C90 compiler.
+
+Mon Mar 27 22:45:37 2006  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/{nkf.c, utf8tbl.c}: imported nkf 2.0.6.
+	  * Add --ic / --oc option and mapping tables.
+	  * Add fallback option.
+	  * Add --no-best-fit-chars option.
+	  * Fix some bugs.
+
+	* ext/nkf/nkf.c (nkf_split_options): added for parse option string.
+
+	* ext/nkf/lib/kconv.rb (Kconv.to*): add -m0.
+	  Note that Kconv.to* still imply -X.
+
+	* ext/nkf/test.rb: Removed.  Obsolete by test/nkf.
+
+	* ext/.document: enabled documents in nkf and kconv
+
+	* ext/nkf/nkf.c, ext/nkf/lib/kconv.rb: Add rdoc.
+
+Mon Mar 27 03:17:21 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): insecure calling should be checked for non
+	  NODE_SCOPE method invocations too.
+
+	* eval.c (rb_alias): should preserve the current safe level as
+	  well as method definition.
+
+Sun Mar 26 22:02:51 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* re.c: refactoring for options.
+
+	* parse.y: ditto.
+
+Fri Mar 24 21:11:02 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* re.c (match_aref): RDoc description updated.
+
+	* string.c (rb_str_sub): ditto.
+
+	* string.c (rb_str_gsub): ditto.
+
+Fri Mar 24 17:20:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (rb_f_sleep): remove RDoc description about SIGALRM
+	  which is not valid on the current implementation. [ruby-dev:28464]
+
+Thu Mar 23 21:40:47 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* re.c (rb_reg_regsub): prohibit \1, \2 ...\9 in replaced string
+	  for named regex pattern.
+
+Thu Mar 23 21:06:23 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* oniguruma.h: Version 4.0.2
+
+	* regparse.c: ditto.
+
+	* regcomp.c ditto.
+
+	* regerror.c: ditto.
+
+Thu Mar 23 10:47:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (method_missing): should support argument splat in
+	  super.  a bug in combination of super, splat and
+	  method_missing.  [ruby-talk:185438]
+
+Thu Mar 23 00:01:32 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* re.c (rb_reg_regsub): add back reference by name \k<name> in
+	  replace string.
+
+	* re.h: add regexp argument to rb_reg_regsub().
+
+	* string.c (rb_str_sub_bang): ditto.
+
+	* string.c (str_gsub): ditto.
+
+Tue Mar 21 22:14:01 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* re.c (match_alloc): initialize member regexp.
+
+	* re.c (match_aref): add String and Symbol argument. [ruby-dev:28448]
+
+	* re.h: add member regexp to RMatch.
+
+	* gc.c (gc_mark_children): add gc_mark() to regexp member.
+
+Mon Mar 20 12:05:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: Solaris SunPro compiler -rapth patch from
+	  <kuwa at labs.fujitsu.com>.  [ruby-dev:28443]
+
+Mon Mar 20 11:12:38 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (folevariant_value): could not compile
+	  with C90 compiler.
+
+Mon Mar 20 09:40:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: remove enable_rpath=no for Solaris.
+	  [ruby-dev:28440]
+
+Sun Mar 19 09:46:30 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_val2olevariantdata): change behavior
+	  of converting	 OLE Variant object with VT_ARRAY|VT_UI1 and Ruby
+	  String object.
+
+	* ext/win32ole/win32ole.c (folevariant_value): ditto.
+
+	* ext/win32ole/tests/testOLEVARIANT.rb: ditto.
+
+Wed Mar 15 16:51:11 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): support libraries without *.so.
+
+Wed Mar 15 16:39:29 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c, ext/openssl/ossl_nsspki.c: should use
+	  "rb_str_new(0, 0)" to make empty string.
+
+Sun Mar 12 17:02:10 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(ole_val2olevariantdata): support VT_ARRAY in
+	  WIN32OLE_VARIANT.new().
+
+	* ext/win32ole/tests/testOLEVARIANT.rb: ditto.
+
+	* ext/win32ole/tests/testOLEPARAM.rb: test method name should not be
+	  duplicated.
+
+Sat Mar 11 14:24:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.1: a clarification patch from David Lutterkort
+	  <dlutter at redhat.com>.  [ruby-core:7508]
+
+Sun Mar  5 18:40:58 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb: do not repeat command options.
+
+Sun Mar  5 18:35:03 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (send_request_with_body): #content_type never
+	  return false, use #main_type instead. [ruby-core:07476]
+
+Sat Mar  4 15:26:40 2006  Tanaka Akira  <akr@m17n.org>
+
+	* gc.c (id2ref): fix symbol test.
+
+Sat Mar  4 01:08:07 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rdoc/ri/ri_paths.rb (RI::Paths): adding paths from rubygems
+	  directories.	a patch from Eric Hodel <drbrain at segment7.net>.
+	  [ruby-core:07423]
+
+Fri Mar  3 17:59:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_clear_cache_by_class): clearing wrong cache.
+
+Fri Mar  3 21:22:42 2006  Tanaka Akira  <akr@m17n.org>
+
+	* lib/fileutils.rb (FileUtils.cp_r): implement :remove_destination
+	  option.
+
+	* ext/extmk.rb: use :remove_destination to install extension libraries
+	  to avoid SEGV.  [ruby-dev:28417]
+
+Fri Mar  3 14:41:04 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/dl/.cvsignore: ignore callback.h.
+
+	* ext/ripper/.cvsignore: ignore eventids2table.c.
+
+	* ext/socket/.cvsignore: ignore constants.h.
+
+Thu Mar  2 18:58:18 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_thread_fd_writable): should not re-schedule output
+	  from KILLED thread (must be error printing).
+
+Thu Mar  2 09:12:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_flatten_bang): allow specifying recursion
+	  level.  [ruby-talk:182170]
+
+	* array.c (rb_ary_flatten): ditto.
+
+Thu Mar  2 08:02:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (add_heap): a heap_slots may overflow.  a patch from Stefan
+	  Weil <weil at mail.berlios.de>.
+
+Wed Mar  1 17:13:37 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call): use separate cache for fcall/vcall
+	  invocation.
+
+	* eval.c (rb_eval): NODE_FCALL, NODE_VCALL can call local
+	  functions.
+
+	* eval.c (rb_mod_local): a new method to specify newly added
+	  visibility "local".
+
+	* eval.c (search_method): search for local methods which are
+	  visible only from the current class.
+
+	* class.c (rb_class_local_methods): a method to list local methods.
+
+Thu Mar  2 17:54:45 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* gc.c: committed magic for reducing RVALUE size on windows. (24->20byte)
+	  [ruby-core:7474]
+
+Thu Mar  2 14:12:26 2006  Tanaka Akira  <akr@m17n.org>
+
+	* gc.c: align VALUE with sizeof(RVALUE) globally.
+	  (is_pointer_to_heap): check alignment out of loop.
+	  (id2ref): avoid collision between symbols and objects.
+	  (rb_obj_id): ditto.  moved from object.c.
+	  [ruby-talk:178364] [ruby-core:7305]
+
+Thu Mar  2 12:55:16 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (filetime_to_unixtime): should set tm_isdst to -1.
+	  stat() didn't treat daylight saving time property on WinNT.
+	  [ruby-talk:182100]
+
+Wed Mar  1 00:15:51 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/rdoc/parsers/parse_rb.rb (read_escape): could not handle /\^/.
+	  merged Mr. Ishizuka's lib/irb/ruby-lex.rb 's patch rev 1.29.
+	  [ruby-talk:181631] [ruby-dev:28404]
+
+Tue Feb 28 19:32:14 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (Init_Object): add BasicObject class as a top level
+	  BlankSlate class.
+
+Mon Feb 27 00:19:16 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (SYM2ID): should not cast to signed long.
+	  [ruby-core:07414]
+
+Fri Feb 24 20:21:38 2006  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/drbtest.rb (add_service_command): quote pathnames in the
+	  server's command line for space contained directory names.
+	  Thanks, arton. [ruby-dev:28386]
+
+Fri Feb 24 12:10:07 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* instruby.rb: install *.exe.manifest and *.dll.manifest if exist.
+	  It's for VC++8.
+
+Fri Feb 24 11:17:45 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (NtInitialize): need to set a handler for VC++8.
+
+Thu Feb 23 22:39:59 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/Makefile.sub: use borlndmm.dll if possible. bcc32's RTL internal
+	  memory manager cannot handle large memory block properly.
+	  ex: 10000.times { "" << "." * 529671; GC.start } # crash
+	  [ruby-dev:28230]
+
+Thu Feb 23 13:23:03 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* eval.c (SETUP_ARGS0): fixed memory corruption. [ruby-dev:28360]
+
+Wed Feb 22 21:16:55 2006  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#each_filename): use split_names properly.
+
+Wed Feb 22 16:24:05 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/webrick/test_cgi.rb: should support platforms which search
+	  library path from the interpreter's path.
+	  And, support test without install incidentally.
+
+Wed Feb 22 14:21:03 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bignorm): x may not be a bignum.  [ruby-dev:28367]
+
+Wed Feb 22 09:22:40 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_alloc): add proper check for creation of a lambda
+	  without a block.
+
+Tue Feb 21 02:07:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_arglist): should set command_start = Qtrue for
+	  command body.  [ruby-talk:180648]
+
+Mon Feb 20 22:30:17 2006  Tanaka Akira  <akr@m17n.org>
+
+	* mkconfig.rb: alias Config to RbConfig for compatibility.
+
+Mon Feb 20 18:21:41 2006  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (rb_io_reopen): flush before reopening a file.
+	  reported by Mathieu Bouchard.  [ruby-core:7396]
+
+Mon Feb 20 17:29:50 2006  Tanaka Akira  <akr@m17n.org>
+
+	* mkconfig.rb: generate RbConfig instead of Config.
+
+	* instruby.rb, rubytest.rb, runruby.rb, bcc32/Makefile.sub,
+	  ext/extmk.rb, ext/dl/extconf.rb, ext/iconv/charset_alias.rb,
+	  lib/mkmf.rb, lib/rdoc/ri/ri_paths.rb,
+	  lib/webrick/httpservlet/cgihandler.rb,
+	  test/dbm/test_dbm.rb, test/gdbm/test_gdbm.rb,
+	  test/ruby/envutil.rb, test/soap/calc/test_calc_cgi.rb,
+	  test/soap/header/test_authheader_cgi.rb, test/soap/ssl/test_ssl.rb,
+	  win32/mkexports.rb, win32/resource.rb: Use RbConfig instead of
+	  Config.
+
+Mon Feb 20 13:46:19 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/find.rb: should raise ENOENT if root entry does not exist,
+	  without opening it. [ruby-dev:28345]
+
+Mon Feb 20 12:27:53 2006  Kent Sibilev  <ksruby@gmail.com>
+
+	* lib/rational.rb (Integer::gcd): small typo fix.
+	  [ruby-core:07395]
+
+Mon Feb 20 10:03:59 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_const_get_0): Object should have been the lowest
+	  in const lookup precedence.  [ruby-dev:28343]
+
+Mon Feb 20 09:17:11 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator): should not delegate "funcall".
+
+Mon Feb 20 09:13:42 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb (WEBrick::HTTPServlet::CGIHandler):
+	  qualify the access for Config constant.  [ruby-dev:28338]
+
+	* lib/resolv.rb (Resolv::DNS::Resource::IN::A): qualify
+	  ClassValue.  [ruby-dev:28338]
+
+Mon Feb 20 01:05:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rational.rb (Integer::gcd): replaced by gcd4 in
+	  [ruby-core:07390].  [ruby-core:07377]
+
+Mon Feb 20 00:57:02 2006  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.h (OSSL_Debug): should not use __func__.
+	  [ruby-dev:28339]
+
+Mon Feb 20 00:13:49 2006  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb: add :ssl_verify_mode option.
+	  suggested by Will Glynn.
+
+	* lib/open-uri.rb: add :ssl_ca_cert option.
+
+Sun Feb 19 04:46:29 2006  Guy Decoux  <ts@moulon.inra.fr>
+
+	* eval.c: initial value for block_unique must be 1.
+	  [ruby-talk:180420]
+
+Sat Feb 18 23:58:26 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tracer.rb (Tracer::Tracer.add_filter): turn on tracer mode
+	  only when caller() level size is one.  [ruby-core:07389]
+
+	* lib/rdoc/parsers/parse_rb.rb: need not to require "tracer".
+	  [ruby-core:07389]
+
+	* sample/rtags.rb: ditto.
+
+Sat Feb 18 21:16:27 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_obj_instance_eval): RDoc description updated.  a
+	  patch from Ozgur Murat Homurlu <ozgurmurath at gmail.com>.
+	  [ruby-core:07381]
+
+Sat Feb 18 01:01:17 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_const_get_0): skip ruby_wrapper in const search
+	  to give it lower priority (just above Object).  need not to
+	  change rb_const_defined_0() since it's only a precedence matter;
+	  they are defined anyway.
+
+Sat Feb 18 00:22:39 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tracer.rb: merged a minor clarification patch from Daniel
+	  Berger <Daniel.Berger at qwest.com>.  [ruby-core:07376]
+
+Fri Feb 17 17:30:20 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (ev_const_get): simplified using rb_const_get_fallback().
+
+	* eval.c (ev_const_defined): adopt to ev_const_get() using
+	  rb_const_defined_fallback().
+
+	* variable.c (rb_const_get_fallback): new function to implement
+	  constant search.
+
+	* variable.c (rb_const_defined_fallback): new function to
+	  implement constant definition check.
+
+	* variable.c (rb_const_get_0): adopt to new behavior.  constants
+	  are looked up in the order of: current class, super classes (but
+	  Object), lexically external classes/modules, and Object.
+
+	* variable.c (rb_const_defined_0): ditto.
+
+Fri Feb 17 11:20:53 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* util.c (ruby_strtod): Float("1e") should fail. [ruby-core:7330]
+
+	* pack.c (EXTEND32): unpack("l") did not work where sizeof(long) != 4.
+	  [ruby-talk:180024]
+
+	* pack.c (pack_unpack): fixed integer overflow on template "w".
+	  [ruby-talk:180126]
+
+Fri Feb 17 09:39:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_thread_wait_for): sleep should always sleep for
+	  specified amount of time.  [ruby-talk:180067]
+
+Wed Feb 15 16:52:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): NODE_OP_ASGN1 should allow splat in its
+	  argument list.  [ruby-core:07366]
+
+	* parse.y (arg): avoid unnecessary extra argument.
+	  [ruby-core:07366]
+
+	* eval.c (rb_eval): honor visibility on OP_ASGN1 and
+	  OP_ASGN2. [ruby-core:07366]
+
+Wed Feb 15 15:20:23 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (error_line): remove void control path.  [ruby-dev:28335]
+
+Wed Feb 15 10:09:51 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (yield_under_i): should not pass self as an argument to
+	  the block for instance_eval.  [ruby-core:07364]
+
+Wed Feb 15 09:20:35 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_obj_instance_eval): should be no singleton classes for
+	  true, false, and nil.  [ruby-dev:28186]
+
+Tue Feb 14 20:26:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_each): return self if no block is
+	  given.  [yarv-dev:882]
+
+Tue Feb 14 18:48:33 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (DMETHOD_P): accessing wrong frame.  [ruby-dev:28181]
+
+	* eval.c (proc_invoke): preserve FRAME_DMETH flag.
+
+Tue Feb 14 15:15:22 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/zlib/zlib.c: suppress warning on test/zlib. [ruby-dev:28323]
+
+Tue Feb 14 13:47:22 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_utime): drop read-only attribute before
+	  changing file time.
+
+Tue Feb 14 13:38:01 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): should define HAVE_LONG_LONG with
+	  VC++8.
+
+Tue Feb 14 11:42:38 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (search_time_t): support non 32bit time_t environments.
+
+	* win32/Makefile.sub (config.h): VC++8 has ``long long'' type.
+
+	* win32/Makefile.sub (config.h): VC++8's time_t is 64bit value.
+
+Mon Feb 13 18:01:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (copy_node_scope): remove duplicated semicolons at end.
+	  a patch from KIMURA Koichi <kimura.koichi at canon.co.jp>.
+	  [ruby-dev:28332]
+
+	* eval.c (VIS_MODE): remove unnecessary argument.
+	  [ruby-dev:28332]
+
+Mon Feb 13 13:49:48 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_parse_string): mention "regexp" in a error
+	  message.  a patch from Mauricio Fernandez <mfp at acm.org>
+	  [ruby-core:07340]
+
+Mon Feb 13 00:01:32 2006  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* oniguruma.h: Version 4.0.1
+
+	* regparse.c (onig_free_shared_cclass_table): fix memory leaks.
+
+	* regcomp.c (optimize_node_left): change from IS_POSIXLINE() to IS_MULTILINE().
+
+	* regint.h: rename ANCHOR_ANYCHAR_STAR_PL to ANCHOR_ANYCHAR_STAR_ML.
+
+	* regparse.h: ditto.
+
+	* regexec.c: ditto.
+
+Sat Feb 11 21:57:29 2006  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE.create_guid.
+
+	* ext/win32ole/tests/testWIN32OLE.rb: ditto.
+
+Sat Feb 11 01:57:44 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_autoload): check if ruby_cbase is nil (during
+	  instance_eval for objects cannot have singleton classes,
+	  e.g. fixnums and symbols).  [ruby-dev:28178]
+
+Fri Feb 10 12:31:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): should support NODE_ZSUPER in NODE_ITER.
+	  [ruby-dev:28326]
+
+	* eval.c (ZSUPER_ARGS): support macro.
+
+Wed Feb  8 10:26:06 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (rb_gc_call_finalizer_at_exit): turn on during_gc while
+	  invoking finalizers.
+
+	* gc.c (rb_gc_finalize_deferred): ditto.
+
+Tue Feb  7 23:03:13 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/zlib/zlib.c: should not access ruby objects in finalizer.
+	  [ruby-dev:28286]
+
+Tue Feb  7 18:42:00 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_write_error2): use fwrite(3) if rb_stderr is not
+	  updated or is already freed.  [ruby-dev:28313]
+
+Mon Feb  6 16:02:51 2006  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* file.c (rb_thread_flock): ERROR_NOT_LOCKED is not an error on Cygwin.
+	  In such situation, flock() should return 0.
+
+Mon Feb  6 14:36:29 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (error_line): include the class name of a surrounding
+	  method in error position description.
+
+Mon Feb  6 00:14:57 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_find_index): a new method Enumerable#find_index.
+	  [ruby-talk:178495]
+
+Sun Feb  5 23:29:31 2006  Tanaka Akira  <akr@m17n.org>
+
+	* ruby.h (struct RStruct): embed 3 or less elements structs.
+	  (RSTRUCT_LEN): defined for accessing struct members.
+	  (RSTRUCT_PTR): ditto.
+
+	* struct.c: use RSTRUCT_LEN and RSTRUCT_PTR.
+	  (struct_alloc): allocate small structs in embedded format.
+	  (rb_struct_init_copy): ditto.
+
+	* gc.c (gc_mark_children): use RSTRUCT_LEN and RSTRUCT_PTR.
+	  (obj_free): ditto.
+
+	* marshal.c (w_object): use RSTRUCT_LEN and RSTRUCT_PTR.
+
+Sun Feb  5 21:01:49 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* numeric.c (fix_to_s): removed workaround for radix 2. Historically,
+	  rb_fix2str could only handle radix 8, 10, 16. (Rev1.37) But for now,
+	  it can handle radix 2..36. [ruby-Bugs#3438] [ruby-core:7300]
+
+Sun Feb  5 18:49:00 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (add_field, get_fields): keep 1.8.2
+	  compatibility.  This patch is contributed by Rob Pitt.
+
+Sun Feb  5 16:33:50 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/mkmf.rb (create_makefile): Kernel#sub! was removed on HEAD.
+
+Sun Feb  5 14:26:54 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/pstore.rb: should return default value if name is not found.
+	  [ruby-core:7304]
+
+	* lib/pstore.rb: should raise PStore::Error if not in transaction.
+
+Sat Feb  4 22:51:43 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c: apply the FreeBSD getcontext/setcontext workaround
+	  only before FreeBSD 7-CURRENT.
+
+Sat Feb  4 21:10:06 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (LK_ERR): ERROR_NOT_LOCKED is not an error.
+	  In such situation, flock() should return 0.
+
+Sat Feb  4 15:52:56 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* numeric.c (fix_to_s): (2**32).to_s(2) fails with exception where
+	  sizeof(int) == 4 < sizeof(long). [ruby-core:7300]
+
+Sat Feb  4 15:02:05 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* oniguruma.h: merge Oniguruma 4.0.0  [ruby-dev:28290]
+
+Fri Feb  3 19:25:53 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ruby.h: fixed prototype.
+
+	* ext/syck/rubyext.c: defined symbol ID as global variable as others.
+
+Fri Feb  3 17:57:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c: unify ruby_class (for method definition) and ruby_cbase
+	  (for constant reference).
+
+Fri Feb  3 15:02:10 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/syck.c (syck_move_tokens): should reset p->cursor or etc
+	  even if skip == 0. This causes buffer overrun.
+	  (ex: YAML.load('--- "..' + '\x82\xA0' * 511 + '"'))
+
+Fri Feb  3 00:01:31 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/emitter.c (syck_emitter_write): should not set '\0' on
+	  emitter's marker. if marker points to the end of buffer, this is
+	  buffer overrun. (ex: YAML.dump("." * 12288))
+
+Thu Feb  2 17:13:01 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser#get_tk): added
+	  support of :'string' style Symbol.
+
+Thu Feb  2 16:01:24 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): use TMP_ALLOC() instead of allocating
+	  a temporary array object.
+
+	* eval.c (eval): need not to protect $SAFE value.
+	  [ruby-core:07177]
+
+	* error.c (Init_Exception): change NameError to direct subclass of
+	  Exception so that default rescue do not handle it silently.
+
+Thu Feb  2 14:45:53 2006  Ville Mattila  <ville.mattila@stonesoft.com>
+
+	* configure.in: The isinf is not recognized by autoconf
+	  library guesser on solaris 10. [ruby-core:7138]
+
+Wed Feb  1 22:01:47 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* configure.in, hash.c (ruby_setenv): use setenv(3) and unsetenv(3)
+	  where they are supported. modifying environ variable seems to
+	  segfault solaris 10. [ruby-core:7276] [ruby-dev:28270]
+
+	* ruby.c (set_arg0): if use setenv(3), environ space cannot be used
+	  for altering argv[0].
+
+Tue Jan 31 14:46:28 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_select): update RDoc description.
+	  [ruby-core:7254]
+
+Tue Jan 31 11:58:51 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: add MultiTkIp#eval and bg_eval.
+
+	* ext/tk/lib/tk/namespace.rb: TkNamespace#eval was enbugged at the
+	  last commit. Now it will return a proper object.
+
+Tue Jan 31 08:07:02 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (int_upto): return an enumerator if no block is
+	  attached to the method.
+
+	* numeric.c (int_downto): ditto.
+
+	* numeric.c (int_dotimes): ditto.
+
+	* enum.c (enum_first): new method Enumerable#first to take first n
+	  elements from an enumerable.
+
+	* enum.c (enum_group_by): new method Enumerable#group_by that
+	  groups enumerable values according to their block values.
+
+Tue Jan 31 00:08:22 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c (syck_resolver_transfer): workaround for SEGV.
+	  ex: ruby -ryaml -e 'YAML.load("!map:B {}")' [ruby-core:7217]
+
+Sat Jan 28 07:49:30 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/rdoc/usage.rb: support "a:0:33" style caller[-1]. In this case
+	  file name is "a:0". I don't know this really happens though...
+	  [ruby-Bugs:3344]
+
+Thu Jan 26 15:55:52 2006  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c: turn on do_not_reverse_lookup by default.
+
+Wed Jan 25 22:29:04 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, dln.c, file.c, intern.h, missing.h (eaccess): use
+	  system routine if provided.  fixed: [ruby-core:07195]
+
+Sun Jan 22 23:27:13 2006  Go Noguchi  <gonoguti@yahoo.co.jp>
+
+	* lib/test/unit/autorunner.rb (process_args): ignore arguments after
+	  '--' so that test scripts can handle them.  fixed: [ruby-dev:28258]
+
+Sun Jan 22 22:09:52 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (POST_GETCONTEXT): define separately from PRE_GETCONTEXT on
+	  IA64 to avoid reusing variable address.
+
+Sun Jan 22 20:03:35 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (ruby_setjmp): define PRE_GETCONTEXT and POST_GETCONTEXT
+	  instead of FUNCTION_CALL_MAY_RETURN_TWICE.
+	  define PRE_GETCONTEXT to clear carry flag for workaround of
+	  FreeBSD/i386 getcontext/setcontext bug.
+	  [ruby-dev:28263]
+
+Thu Jan 19 22:19:18 2006  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mv): should remove file after copying.
+	  [ruby-dev:28223]
+
+Wed Jan 18 23:37:06 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (FUNCTION_CALL_MAY_RETURN_TWICE): don't clobber %l7 of SPARC
+	  if enable-shared.
+	  (ruby_setjmp): call FUNCTION_CALL_MAY_RETURN_TWICE after getcontext
+	  too.
+	  reported by Pav Lucistnik and Marius Strobl.
+	  http://lists.freebsd.org/pipermail/freebsd-sparc64/2006-January/003739.html
+
+Tue Jan 17 23:59:56 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_get, rb_mod_const_defined): added optional
+	  flag to search ancestors, which is defaulted to true.
+	  fixed: [ruby-talk:175899]
+
+	* eval.c (rb_mod_method_defined): ditto.
+
+Tue Jan 17 11:31:47 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.mak (MAKE): workaround for nmake 8.
+
+Tue Jan 17 11:06:19 2006  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub: invoke .bat via shell. workaround for nmake 8.
+
+Mon Jan 16 10:13:38 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/emitter.c (syck_emit_seq, syck_emit_map, syck_emit_item):
+	  should output complex key mark even if map's key is empty seq/map.
+	  [ruby-core:7129]
+
+Sat Jan 14 03:38:54 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* file.c (rb_file_s_chmod): avoid warning where sizeof(int) !=
+	  sizeof(void*).
+
+Fri Jan 13 19:26:15 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/rdoc/diagram.rb:
+	    - properly quote bare element attributes
+	    - terminates dangling elements (e.g. <img>, <br>, <link>, etc)
+	    - converts "CVS" to the more HTML-friendly acronym element
+	    - adds missing type attributes to style elements
+
+	  based on Paul Duncan's patch <pabs@pablotron.org> [ruby-core:7028]
+
+	* lib/rdoc/generators/html_generator.rb: ditto.
+	* lib/rdoc/generators/template/html/hefss.rb: ditto.
+	* lib/rdoc/generators/template/html/html.rb: ditto.
+	* lib/rdoc/generators/template/html/kilmer.rb: ditto.
+
+Thu Jan 12 11:53:08 2006  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tkballoonhelp.rb: [bug fix] couldn't add to a widget
+	  which is constructed with TkComposite module.
+	  [new feature] support 'command' option which is called just before
+	  popping up the balloon help.
+
+Wed Jan 11 00:12:29 2006  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/erb.rb (ERB::Compiler): add instance variable @insert_cmd to
+	  change <%='s behavior.
+
+Tue Jan 10 19:42:33 2006  Tanaka Akira  <akr@m17n.org>
+
+	* gc.c (garbage_collect): mark ruby_current_node.
+	  if an exception is raised in a finalizer written in C called by
+	  rb_gc_call_finalizer_at_exit, ruby_set_current_source may use
+	  collected ruby_current_node and mark_source_filename may corrupt
+	  memory.
+
+Tue Jan 10 13:30:34 2006  akira yamada  <akira@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_resolver_transfer): should be able to load
+	  !ruby/object:Bignum syntax 1.8.3 dumped. [ruby-core:6159]
+
+Tue Jan 10 12:47:41 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/yaml/rubytypes.rb (Fixnum): Bignum could not be loaded in
+	  ruby 1.8.3/1.8.4. [ruby-core:6115]
+
+	* lib/yaml/rubytypes.rb (Numeric): Subclass of Numeric could not
+	  be dumped properly. [ruby-core:7047]
+
+Tue Jan 10 12:00:48 2006  Aaron Schrab  <aaron @nospam@ schrab.com>
+
+	* lib/yaml/rubytypes.rb (Symbol#yaml_new): YAML loading of quoted
+	  Symbols broken. [ruby-Bugs:2535]
+
+Tue Jan 10 07:26:52 2006  Tanaka Akira  <akr@m17n.org>
+
+	* gc.c (gc_stress): renamed from always_gc and enabled by default.
+	  (gc_stress_get): new function for GC.stress.
+	  (gc_stress_set): new function for GC.stress=.
+
+Mon Jan  9 19:58:56 2006  arton  <artonx@yahoo.co.jp>
+
+	* ext/zlib/extconf.rb: zlib compiled DLL version 1.2.3 distributed by
+	  http://www.zlib.net/ has zdll.lib. [ruby-dev:28209]
+
+Mon Jan  9 14:25:00 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/Makefile.sub (OPTFLAGS): I have experienced trouble on y- flag,
+	  (VisualC++6) so use -O2b2xg- if  $(MSC_VER) < 1400. [ruby-core:7040]
+
+Mon Jan  9 14:25:00 2006  Kero van Gelder  <rubyforge @nospam@ kero.tmfweb.nl>
+
+	* lib/webrick/httpservlet/filehandler.rb: fixed typo. [ruby-core:7075]
+
+Sun Jan  8 14:15:27 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (GCC_VERSION_BEFORE): check __INTEL_COMPILER.
+	  Intel C++ Compiler defines __GNUC__.
+	  http://www.intel.com/software/products/compilers/clin/docs/ug_cpp/lin1077.htm
+
+Sat Jan  7 15:40:07 2006  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (singleton): get rid of segfault on syntax error.
+	  fixed: [ruby-core:07070]
+
+Sat Jan  7 06:24:18 2006  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (rb_fd_isset): compare the result of FD_ISSET with 0 to
+	  avoid FreeBSD bug.  FreeBSD defines FD_ISSET as just a bitmap of
+	  unsigned long.  So returning the value from rb_fd_isset discards
+	  upper 32bits on LP64 environment.
+	  http://www.freebsd.org/cgi/query-pr.cgi?pr=ia64/91421
+
+Fri Jan  6 02:20:18 2006  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in: don't force getcontext on IA64.
+
+	* eval.c (ruby_setjmp): add an argument for just before getcontext.
+	  (THREAD_SAVE_CONTEXT): call rb_thread_save_context just
+	  before getcontext.
+	  [ruby-dev:28205]
+
+Sun Jan  1 15:28:46 2006  Tanaka Akira  <akr@m17n.org>
+
+	* missing.h (isinf): avoid macro expansion
+	  "extern int isinf(double);" to
+	  "extern int ((sizeof(double)==sizeof(float))?_Isinff(double):_Isinf(double));" on
+	  HP-UX.
+
+Sun Jan  1 14:42:54 2006  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_seekdir): should not segfault even if passed
+	  the location which rb_w32_telldir didn't return. (and should change
+	  `bits' position) [ruby-core:7035]
+
+	* win32/dir.h: ditto. (stores `loc' instead of `bitpos')
+
+	* test/ruby/test_dir.rb: added.
+
+Sat Dec 31 22:57:00 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_save_context): should not recycle scope object used
+	  in a thread.  fixed: [ruby-dev:28177]
+
+Sat Dec 31 19:50:38 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c: attribute name was truncated with Rev1.64.
+
+Sat Dec 31 11:53:16 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/generator.rb: (Generator#initialize): should kill @loop_thread
+	  before starting new thread. (occurs when called via Generator#rewind)
+	  [ruby-dev:28184]
+
+Fri Dec 30 18:22:42 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (garbage_collect): mark objects referred from aborting threads.
+	  [ruby-dev:28190]
+
+	* win32/Makefile.sub: VC++8 support.
+
+Fri Dec 30 15:17:35 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/generator.rb (Generator#initialize): ensured to stop @loop_thread.
+	  Mr. Tanaka pointed out one Thread.pass is not enough. [ruby-dev:28185]
+
+Fri Dec 30 12:20:57 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/generator.rb (Generator#initialize): fixed dead lock. this occurred
+	  when end? was called before @loop_thread was stopped. [ruby-core:7029]
+
+Fri Dec 30 01:04:52 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/generator.rb: should work with another thread. (more robust code)
+	  [ruby-dev:28177]
+
+Thu Dec 29 23:59:37 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_gc_mark_threads): keep unmarked threads which won't wake
+	  up alone, and mark threads in the loading table.  [ruby-dev:28154]
+
+	* eval.c (rb_gc_abort_threads), gc.c (gc_sweep): kill unmarked
+	  threads.  [ruby-dev:28172]
+
+Thu Dec 29 17:02:07 2005  Tanaka Akira  <akr@m17n.org>
+
+	* test/ruby/envutil.rb (EnvUtil.rubybin): search "ruby" instead of
+	  "miniruby".  [ruby-dev:28140]
+
+Thu Dec 29 14:35:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_mod_define_method): should save safe_level in the
+	  proc object.  [ruby-dev:28146]
+
+Thu Dec 29 11:22:34 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/generator.rb: reimplemented Generator class with Thread instead of
+	  callcc, in order to fix memory leak. [ruby-dev:28142]
+
+Wed Dec 28 14:10:05 2005  Tanaka Akira  <akr@m17n.org>
+
+	* ia64.s: remove .pred.safe_across_calls directive.
+	  reported by WATANABE Tetsuya.  [ruby-dev:28141]
+
+Wed Dec 28 01:32:39 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (struct thread): add bstr_max.
+	  (rb_thread_save_context): use realloc instead of REALLOC_N
+	  to avoid GC.
+
+Tue Dec 27 23:59:53 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (CompletingHash#match): fix for 1.9.
+
+Tue Dec 27 16:59:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/drb/drbtest.rb (DRbService::self.ext_service): increase
+	  timeout limit.  a patch from Kazuhiro NISHIYAMA
+	  <zn at mbf.nifty.com>. [ruby-dev:28132]
+
+Tue Dec 27 14:17:55 2005  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in: define IA64 for portability.  (HP aC++/ANSI C doesn't
+	  define __ia64__.)
+	  don't check libunwind stuff.
+	  check __libc_ia64_register_backing_store_base.
+
+	* defines.h: declare rb_ia64_bsp and rb_ia64_flushrs.
+	  (flush_register_windows): call rb_ia64_flushrs on IA64.
+
+	* ia64.s: new file for IA64.
+	  it is separated from C program files because
+	  Intel C++ Compiler for IA64 doesn't support inline assembly.
+
+	* common.mk (ia64.$(OBJEXT)): new target.
+
+	* ruby.h (RUBY_INIT_STACK): defined.
+	  (ruby_init_stack): declared for RUBY_INIT_STACK.
+
+	* main.c (main): precedes RUBY_INIT_STACK before ruby_init.
+
+	* gc.c (rb_gc_register_stack_start): new global variable on IA64.
+	  (garbage_collect): simplify register stack marking code.
+	  don't use libunwind.
+	  (Init_stack): initialize rb_gc_register_stack_start.
+	  (ruby_init_stack): new function for RUBY_INIT_STACK.
+
+	* eval.c (struct thread): add bstr_pos member for original position of
+	  register stack.
+	  (rb_thread_save_context): simplify register stack saving code.
+	  don't use libunwind.
+	  (rb_thread_restore_context_0): new function.  moved from
+	  rb_thread_restore_context except the stack position checking code.
+	  don't use libunwind for IA64 register stack.
+	  (register_stack_extend): new function.
+	  (stack_extend): make it self-recursive with
+	  the stack position checking code in old rb_thread_restore_context.
+	  (rb_thread_restore_context): just call stack_extend.
+	  (flush_register_windows): removed.
+
+	  [ruby-dev:28127]
+
+Tue Dec 27 14:09:39 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* process.c: new method Process.exec. [ruby-dev:28107]
+
+Tue Dec 27 08:22:15 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLSocket#post_connection_check):
+	  treat wildcard character in commonName. [ruby-dev:28121]
+
+Mon Dec 26 08:50:36 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (ev_const_get): fixed a bug in constant reference during
+	  instance_eval.  [yarv-dev:707]
+
+	* eval.c (ev_const_defined): ditto.
+
+	* lib/yaml.rb (YAML::add_domain_type): typo fixed.  a patch from
+	  Joel VanderWerf <vjoel at path.berkeley.edu>.
+	  [ruby-talk:165285] [ruby-core:6995]
+
+Fri Dec 23 10:30:23 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/digest/sha2/sha2.c (ULL): support AIX C.  a patch from
+	  Kailden <kailden at gmail.com>.  [ruby-core:06984]
+
+Wed Dec 21 16:47:35 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* file.c (w32_io_info): should return handle because FileIndex is
+	  valid only while file is open. [ruby-dev:28088]
+
+Wed Dec 21 12:12:21 2005  Tanaka Akira  <akr@m17n.org>
+
+	* test/pathname/test_pathname.rb (test_kernel_open): use
+	  File.identical?.
+	  [ruby-talk:171804]
+
+Tue Dec 20 22:41:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (eval_under_i): evaluate source in caller's frame.
+	  [ruby-dev:28076]
+
+Tue Dec 20 12:53:23 2005  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_emitter_reset): to ensure compatibility
+	  with previous Ruby versions, documents are no longer headless.
+
+Tue Dec 20 12:33:01 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c (syck_node_transform): ruby object holding
+	  explicitly freed SyckNode caused SEGV. [ruby-dev:28067]
+
+	... I think syck GC problem was solved now!
+
+Tue Dec 20 01:46:48 2005  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (rb_f_backquote): fix a GC problem on
+	  IA64 with gcc 4.0.3 20051216 (prerelease) -O3.
+
+Mon Dec 19 23:32:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (XCFLAGS): separated as well as win32/Makefile.sub.
+
+	* main.c (always_gc): dllimport is required for VC to import a DLL
+	  symbol.  fixed: [ruby-dev:28051]
+
+	* parse.y (rb_symname_p): fixed wrong validation.  [ruby-dev:28047]
+
+Mon Dec 19 23:09:24 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* st.c: uses malloc instead of xmalloc to avoid GC. syck uses st_insert
+	  in gram.c to insert node from rb_syck_bad_anchor_handler into
+	  SyckParser's hash table. if GC occurs in st_insert, it's not under
+	  SyckParser's mark system yet. so RString can be released wrongly.
+
+	* ext/syck/syck.h (S_FREE): small hack. no need to check if pointer is
+	  NULL or not before S_FREE.
+
+	* ext/syck/rubyext.c (syck_parser_assign_io): rb_check_string_type can
+	  return new RString. if so, it becomes unreachable from GC after
+	  returns syck_parser_assign_io, and can be freed by GC. (dangling
+	  in syck io system) so extends its life time till syck_parse is called.
+
+	* ext/syck/rubyext.c (syck_parser_s_alloc): always allocates bonus,
+	  so no need to check if NULL, and "volatile VALUE hash"
+	  is not needed. (bonus->port was not protected in syck_emitter_reset)
+
+	* ext/syck/rubyext.c (syck_mark_parser): ditto.
+
+	* ext/syck/rubyext.c (syck_parser_load): ditto.
+
+	* ext/syck/rubyext.c (syck_parser_load_documents): ditto.
+
+	* ext/syck/rubyext.c (syck_emitter_s_alloc): ditto.
+
+	* ext/syck/rubyext.c (syck_mark_emitter): ditto.
+
+	* ext/syck/rubyext.c (syck_emitter_reset): ditto.
+
+	* ext/syck/rubyext.c (syck_scalar_value_set): "should set newly
+	  allocated memory instead of RString's internal storage" stuff again.
+	  by this, should call syck_free_node instead of rb_syck_free_node.
+
+	* ext/syck/rubyext.c (syck_node_type_id_set): ditto.
+
+	... I believe syck GC problem was solved by this.
+
+Mon Dec 19 12:20:59 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (FUNCTION_CALL_MAY_RETURN_TWICE): activate only
+	  before gcc 4.0.3 on SPARC and IA64.
+
+Mon Dec 19 11:37:47 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c: sorry, I reverted my "should set newly
+	  allocated memory instead of RString's internal storage" stuff.
+	  node allocated in rubyext.c seems to be freed by rb_syck_free_node
+	  not syck_free_node, and it won't free data.str->ptr and type_id.
+
+	 (I still think this is unsafe because RString(foo)->ptr becomes
+	  dangling pointer when RString is modified or freed, but anyway
+	  I misunderstood, so go back to original code for now)
+
+Sat Dec 17 21:50:41 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c (syck_emitter_reset): should initialize
+	  emitter->bonus->oid. otherwise rb_gc_mark crashes.
+
+	* ext/syck/rubyext.c (syck_mark_parser): should mark anchor nodes
+	  because they hold ruby objects. (ie: rb_syck_bad_anchor_handler)
+
+Sat Dec 17 11:00:17 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/rubyext.c (rb_syck_compile): avoid potential memory
+	  leak.
+
+	* ext/syck/rubyext.c (syck_set_ivars): avoid potential memory
+	  leak by explicit symbol allocation.
+
+Sat Dec 17 03:57:01 2005  Tanaka Akira  <akr@m17n.org>
+
+	* bignum.c (rb_big_rshift): fix a GC problem on
+	  IA64 with gcc 4.0.3 20051216 (prerelease).
+
+Sat Dec 17 03:30:23 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (bmcall): fix a GC problem by tail call on
+	  IA64 with gcc 4.0.3 20051216 (prerelease).
+
+Fri Dec 16 17:53:45 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c (rb_syck_compile): fixed memory leak.
+
+	* ext/syck/rubyext.c: should protect global variable from GC.
+
+Fri Dec 16 11:44:43 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c (syck_resolver_tagurize): fixed memory leak.
+
+	* ext/syck/rubyext.c (syck_node_type_id_set): should set newly
+	  allocated memory instead of RString's internal storage.
+
+	* ext/syck/rubyext.c (syck_scalar_value_set): ditto.
+
+	... these fixes won't fix [ruby-dev:27839]. more work is needed.
+
+Fri Dec 16 04:38:55 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator::method_missing): should delegate
+	  block as well.
+
+Thu Dec 15 19:57:12 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::MorphingBody): fix criteria to
+	  use Tempfile.  A fix from Zev Blut <rubyzbibd at ubit.com>.
+	  [ruby-core:06076]
+
+	* string.c: remove global functions work on $_.
+
+Thu Dec 15 12:35:14 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tmpdir.rb: merged RDoc patch from Eric Hodel <drbrain at
+	  segment7.net>.  [ruby-core:06894]
+
+Thu Dec 15 01:33:31 2005  Tanaka Akira  <akr@m17n.org>
+
+	* ext/zlib/zlib.c (zstream_run): fix a GC problem by tail call on
+	  x86_64 with gcc 4.0.3 20051111 (prerelease) (Debian 4.0.2-4)
+
+Wed Dec 14 23:50:20 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_c.rb (find_class_comment): fix for class
+	  document with prototypes.  [ruby-core:06863]
+
+Wed Dec 14 23:39:53 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (has_magic): glob names contain alphabets to enable case fold
+	  search.  [ruby-dev:27735]
+
+	* dir.c (Init_Dir): FNM_SYSCASE which is default case fold flag.
+	  [ruby-dev:23296]
+
+Wed Dec 14 12:01:26 2005  Tanaka Akira  <akr@m17n.org>
+
+	* marshal.c (r_object0): fix a GC problem for reading a bignum on
+	  IA64 with gcc 3.3.5 (Debian 1:3.3.5-13).
+
+Tue Dec 13 12:23:47 2005  Tanaka Akira  <akr@m17n.org>
+
+	* re.c (rb_reg_regcomp): fix a GC problem on x86_64 with
+	  gcc 3.3.5 (Debian 1:3.3.5-13).
+
+Tue Dec 13 01:44:16 2005  Tanaka Akira  <akr@m17n.org>
+
+	* array.c (rb_ary_diff): fix a GC problem on IA64 with
+	  gcc 3.3.5 (Debian 1:3.3.5-13).
+	  When rb_ary_push is called, there was no register which contains
+	  `hash' but `&RHASH(hash)->tbl' instead.
+
+Tue Dec 13 00:08:09 2005  Tanaka Akira  <akr@m17n.org>
+
+	* sprintf.c (rb_str_format): fix a GC problem.
+	  [ruby-dev:28001]
+
+Mon Dec 12 15:51:22 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/openssl/test_ssl.rb (test_parallel): call GC.start to close
+	  unused files. [ruby-dev:27981]
+
+Mon Dec 12 09:58:09 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_cover): new method Range#cover? added.  the
+	  method name might be changed.  thanks to takano32 at
+	  http://www.rubyist.net/~matz/20051210.html#c08 for name
+	  suggestion.  [ruby-talk:167182]
+
+Mon Dec 12 00:33:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/digest/digest.c (rb_digest_base_s_digest): add volatile to
+	  protect temporary context object.  [ruby-dev:27979]
+
+	* ext/iconv/iconv.c (Init_iconv): rb_gc_register_address() should
+	  be called before actual variable initialization.
+	  [ruby-dev:27986]
+
+Sun Dec 11 23:54:07 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/*: update to support libraries in ActiveTcl8.4.12.0
+	  (see ext/tk/ChangeLog.tkextlib).
+
+	* ext/tk/sample/scrollframe.rb: add a new sample.
+
+Sun Dec 11 22:07:58 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb (test_remote_array_and_hash): pseudo remote
+	  objects are protected against GC. [ruby-dev:27911]
+
+Sat Dec 10 01:06:06 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/matrix.rb: add Matrix#determinant_e, Matrix#rank_e.
+	  [ruby-dev:27820] and related thread.
+
+Sat Dec 10 00:31:42 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (calling_scope_t): gave names to magic numbers for rb_call().
+	  [ruby-dev:27978]
+
+Fri Dec  9 23:31:02 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rexml/encoding.rb (encoding=): give priority to particular
+	  conversion to iconv.  [ruby-core:06520]
+
+Fri Dec  9 23:16:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_include): return false unless included in numeric
+	  range.  fixed: [ruby-dev:27975]
+
+Thu Dec  8 02:07:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (umethod_bind): adjust invoking class for module method.
+	  [ruby-dev:27964]
+
+Thu Dec  8 00:40:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (call_trace_func): klass parameter should be a
+	  class/module that defines calling method.  [ruby-talk:169307]
+
+Wed Dec  7 17:10:27 2005  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* sprintf.c (rb_f_sprintf): [ruby-dev:27967]
+
+Wed Dec  7 16:39:18 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_include): use discrete membership for non Numeric
+	  values, for example, String.
+
+	* numeric.c (num_scalar_p): new method. [ruby-dev:27936]
+
+	* lib/complex.rb (Complex#scalar?): ditto.
+
+Wed Dec  7 15:31:35 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): integer overflow check added.
+
+	* sprintf.c (GETASTER): ditto.
+
+Wed Dec  7 01:02:04 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/README.macosx-aqua: [new document] tips to avoid the known
+	  bug on platform specific dialogs of Tcl/Tk Aqua on MacOS X.
+
+	* ext/tk/tcltklib.c: fix bug on switching threads and waiting on the
+	  deleted interpreter on vwait and tkwait command.
+
+	* ext/tk/lib/multi-tk.rb: kill the meaningless loop for the deleted Tk
+	  interpreter.
+
+	* ext/tk/sample/demos-jp/image3.rb: [bug fix] wrong argument.
+
+	* ext/tk/sample/demos-en/image3.rb: ditto.
+
+	* ext/tk/sample/demos-jp/menu.rb: fix message for MacOS X.
+
+	* ext/tk/sample/demos-jp/menu8x.rb: ditto.
+
+	* ext/tk/sample/demos-en/menu.rb: ditto.
+
+Tue Dec  6 16:48:40 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (ruby_xmalloc2): change check condition for integer
+	  overflow.  [ruby-dev:27399]
+
+	* gc.c (ruby_xrealloc2): ditto.
+
+Tue Dec  6 16:37:57 2005  Yuya Nishida  <yuya@j96.org>
+
+	* eval.c (exec_under): avoid accessing ruby_frame->prev.
+	  [ruby-dev:27948]
+
+Fri Dec  2 19:06:06 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (Compare): should not fold double byte alphabet on win9x.
+
+Thu Dec  1 00:50:33 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_funcall2): allow to call protected methods.
+	  fixed: [ruby-dev:27890]
+
+Wed Nov 30 23:52:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (struct parser_params): fields common to ripper must be
+	  placed at each same offset.
+
+	* parse.y (NEWHEAP, ADD2HEAP): set count after pointer was set.
+	  fixed: [ruby-dev:27896]
+
+Wed Nov 30 13:43:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-expr-beg): support $! at the end of
+	  expression.   [ruby-dev:27868]
+
+Tue Nov 29 23:57:05 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (struct parser_params): heap must be placed at same offset
+	  also in ripper.y.  fixed: [ruby-dev:27846]
+
+	* parse.y (yycompile): prevent vparser from tail call optimization.
+	  fixed: [ruby-dev:27851]
+
+	* parse.y (parser_mark): value needs to be marked.
+	  fixed: [ruby-dev:27845]
+
+Tue Nov 29 22:45:30 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/observer.rb (notify_observers): follow change of observer.rb.
+	  fixed: [ruby-core:6796]
+
+Mon Nov 28 20:24:22 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb (PP::PPMethods#object_address_group): mask an address with
+	  word size.
+
+	* lib/pp.rb (PP::PPMethods#object_address_group): adjust address format.
+
+Mon Nov 28 18:55:22 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (init_inetsock_internal): remove setting
+	  SO_REUSEADDR option on server socket on Cygwin.
+	  fixed: [ruby-core:6765] ([ ruby-Bugs-2872 ])
+
+Mon Nov 28 13:11:45 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_strerror): remove all CR and LF. (avoid broken
+	  error message on bccwin32 + winsock)
+
+Mon Nov 28 09:15:50 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/mkmf.rb (create_makefile): should not change sodir with
+	  dir.gsub!. (bccwin32 failed to install third party extensions)
+	  [ruby-dev:27834]
+
+Sun Nov 27 05:37:20 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb: use File.basename to decompose pathnames.
+	  experimental Windows support.
+
+Sun Nov 27 00:56:13 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/wsdl/xmlSchema/complexContent.rb: missing
+	  ComplexContent#elementformdefault method.
+
+Sat Nov 26 19:57:45 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* dln.c (conv_to_posix_path): should initialize posix.
+
+Fri Nov 25 20:34:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/xmlrpc/datetime.rb (DateTime::to_a): comparison with non
+	  array-convertible object must return false.
+
+Fri Nov 25 14:34:09 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_max): treat end exclusion without iteration if
+	  the end value is an integer.  [ruby-talk:167433]
+
+Fri Nov 25 12:52:57 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: added backward compatibility codes.
+	* lib/rss/parser.rb: ditto.
+	* test/rss/test_parser.rb: ditto.
+	* test/rss/test_2.0.rb: ditto.
+
+	* test/rss/test_content.rb: use #__send__ instead of #funcall for
+	  no private method.
+
+Fri Nov 25 12:39:56 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: improved type conversion.
+	* lib/rss/1.0.rb: ditto.
+	* lib/rss/0.9.rb: ditto.
+	* lib/rss/2.0.rb: ditto.
+	* lib/rss/image.rb: ditto.
+	* lib/rss/syndication.rb: ditto.
+
+	* test/rss/test_2.0.rb: added type conversion tests.
+	* test/rss/test_accessor.rb: ditto.
+	* test/rss/test_to_s.rb: ditto.
+	* test/rss/test_syndication.rb: ditto.
+	* test/rss/test_setup_maker_2.0.rb: ditto.
+	* test/rss/test_setup_maker_1.0.rb: ditto.
+	* test/rss/test_setup_maker_0.9.rb: ditto.
+	* test/rss/test_maker_sy.rb: ditto.
+	* test/rss/test_maker_image.rb: ditto.
+	* test/rss/test_maker_2.0.rb: ditto.
+	* test/rss/test_maker_0.9.rb: ditto.
+	* test/rss/test_image.rb: ditto.
+
+	* test/rss/test_maker_1.0.rb: use assert instead of assert_equal.
+
+	* test/rss/rss-assertions.rb: improved type conversion assertions.
+
+Fri Nov 25 10:38:20 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/image.rb: added Image prefix.
+
+	* lib/rss/maker/image.rb: ditto.
+
+Fri Nov 25 10:33:02 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_2.0.rb: added RSS 2.0 tests.
+
+	* test/rss/rss-assertions.rb: extended XML stylesheet assertion.
+
+	* lib/rss/0.9.rb: added initialize method.
+
+	* test/rss/test_1.0.rb: cleanup.
+
+Fri Nov 25 10:29:48 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_min): use <=> comparison rather than iteration.
+	  [ruby-talk:167420]
+
+	* range.c (range_max): ditto.
+
+Thu Nov 24 01:31:44 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* file.c (w32_io_info): CreateFile failed on Win9x if file was already
+	  opened. (FILE_SHARE_READ was needed, but actually I don't understand
+	  the flags of CreateFile well...)
+
+Wed Nov 23 23:52:35 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (num_div): use floor rather than rb_Integer().
+	  [ruby-dev:27674]
+
+Wed Nov 23 22:34:15 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb: added entity handling type predicate.
+	* lib/rss/rexmlparser.rb: ditto.
+	* lib/rss/xmlparser.rb: ditto.
+	* lib/rss/xmlscanner.rb: ditto.
+
+	* lib/rss/xmlscanner.rb: more robust entity handling.
+
+	* test/rss/test_parser.rb: added an entity handling test.
+
+Wed Nov 23 20:59:01 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: add Tk.pkgconfig_list and Tk.pkgconfig_get
+	  [Tk8.5 feature].
+
+	* ext/tk/lib/tk/text.rb: supports new indices modifiers on a Text
+	  widget [Tk8.5 feature].
+
+	* ext/tk/lib/tk/virtevent.rb: add TkNamedVirtualEvent.
+
+	* ext/tk/lib/tk/autoload.rb: ditto.
+
+	* ext/tk/lib/tk/event.rb: add :data key for virtual events [Tk8.5
+	  feature].
+
+Wed Nov 23 18:52:45 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* file.c (w32_io_info): should not call GetFileInformationByHandle
+	  for pipe.
+
+	* file.c (w32_io_info): checks return value from rb_w32_get_osfhandle.
+
+	* file.c (w32_io_info): now can identify directory on WinNT.
+
+Wed Nov 23 18:46:53 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/: use #__send__ instead of #send.
+	* test/rss/: ditto.
+
+Wed Nov 23 18:32:56 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_taxonomy.rb: use #reject directory.
+
+Wed Nov 23 18:26:00 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/taxonomy.rb: changed class or module prefix to
+	  Taxonomy from Taxo.
+	* lib/rss/maker/taxonomy.rb: ditto.
+
+Wed Nov 23 18:21:11 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/taxonomy.rb: implemented taxonomy module for RSS
+	  Maker.
+	* lib/rss/taxonomy.rb: supported RSS Maker.
+	* lib/rss/maker.rb: added taxonomy module support.
+
+	* lib/rss/rss.rb: adjusted to other element API.
+	* lib/rss/1.0.rb: adjusted to other element API but backward
+	  compatibility is reserved.
+	* lib/rss/0.9.rb: ditto.
+
+	* test/rss/test_maker_taxo.rb: added test case for taxonomy module
+	  for RSS Maker.
+	* test/rss/test_setup_maker_1.0.rb: added tests for taxo:topic.
+
+	* test/rss/test_setup_maker_1.0.rb: added backward compatibility
+	  test.
+	* test/rss/test_setup_maker_0.9.rb: ditto.
+	* test/rss/test_setup_maker_2.0.rb: ditto.
+
+	* test/rss/rss-testcase.rb: added convenience method for setting
+	  up taxo:topic.
+	* test/rss/rss-assertions.rb: added assertion for taxo:topic.
+
+	* sample/rss/blend.rb: followed new API.
+
+Wed Nov 23 17:42:24 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: fixed a indentation bug.
+
+	* lib/rss/taxonomy.rb: fixed <taxo:topic> #to_s bug.
+
+	* test/rss/test_taxonomy.rb: added a #to_s test.
+
+Wed Nov 23 03:40:49 2005  Guy Decoux  <ts@moulon.inra.fr>
+
+	* re.c (KR_REHASH): should cast to unsigned for 64bit CPU.
+	  [ruby-core:06721]
+
+Wed Nov 23 07:26:44 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: check for X509V3_EXT_nconf_nid.
+
+	* ext/openssl/ossl_x509ext.c (MakeX509ExtFactory): should use
+	  OPENSSL_malloc to allocate X509V3_CTX.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_create_ext): use
+	  X509V3_EXT_nconf_nid to avoid SEGV (and to build extensions which
+	  values are placed in separate section).
+
+	* test/openssl/test_x509ext.rb: new file.
+
+Wed Nov 23 01:22:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (test_identical): test if two files are identical.
+
+	* file.c (rb_f_test): support DOSISH systems where st_ino is not
+	  reliable.  fixed: [ruby-core:06672]
+
+	* win32.h, win32.c (rb_w32_osid): check the running platform.
+
+Tue Nov 22 23:52:06 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb: match incomplete (in current enconding) multibyte
+	  string.  http://inamode6.tokuhirom.dnsalias.org/show/1551
+
+Tue Nov 22 18:36:11 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (winnt_stat): set mapped errno instead of ENOENT.
+
+Tue Nov 22 14:36:54 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_file_s_basename): skip slashes just after UNC top slashes.
+
+	* test/ruby/test_path.rb (test_dirname, test_basename): follow new
+	  spec. and add new tests.
+
+Tue Nov 22 13:30:15 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_stat): Dir.chdir('//server/shared');
+	  p Dir.glob('*') should work on WinNT. (implemented our own
+	  stat(2) on WinNT) [ruby-list:41552] [ruby-dev:27711]
+
+Tue Nov 22 02:31:53 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkextlib/tile.rb: bug fix (Tk::Tile::USE_TTK_NAMESPACE
+	  is not defined).
+
+Tue Nov 22 01:45:21 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_basename): DOSISH_UNC is defined on cygwin but
+	  DOSISH is not.  fixed: [ruby-dev:27797]
+
+Mon Nov 21 22:50:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_path_skip_prefix, rb_file_s_basename): UNC without path
+	  should not be splitted.  fixed: [ruby-dev:27776] [ruby-dev:27786]
+
+	* parse.y (dsym): prohibit empty symbol literal by interpolation.
+	  fixed: [ruby-talk:166529]
+
+Mon Nov 21 16:03:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/setup.mk: findstr doesn't exist on win9x.
+	  fixed: [ruby-dev:27756]
+
+Sun Nov 20 21:39:27 2005  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* regparse.c (fetch_token_in_cc): tok->escaped should be
+	  initialized.  [ruby-dev:27763]
+
+Sun Nov 20 22:34:06 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_symname_p): [ not followed by ] is not valid symbol.
+	  fixed: [ruby-talk:166520]
+
+Sat Nov 19 19:57:54 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::ln): ln documentation fix.
+	  [ruby-core:06661]
+
+Sat Nov 19 08:19:38 2005  Zach Dennis  <zdennis@mktec.com>
+
+	* ext/socket/socket.c: Socket Documentation. [ruby-core:6552]
+
+Sat Nov 19 07:34:32 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/font.rb: remove dependency on Ruby's version (1.8
+	  or 1.9).
+
+	* ext/tk/lib/tkextlib/ICONS/icons.rb: ditto.
+
+	* ext/tk/sample/tkextlib/treectrl/demo.rb: ditto.
+
+Fri Nov 18 18:07:05 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_file_s_dirname): should use skipprefix for UNC path.
+	  pointed out by nobu ([ruby-dev:27744]). fixed: [ruby-core:5076]
+
+Fri Nov 18 17:35:09 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: add restriction to access the entried
+	  command table and manipulate other IPs (for reason of security).
+	  Now, a IP object can be controlled by only its master IP or the
+	  default IP.
+
+	* ext/tk/lib/remote-tk.rb: add restriction to manipulate.
+
+	* ext/tk/tcltklib.c (ip_is_slave_of_p): add TclTkIp#slave_of?(ip)
+	  to check manipulability.
+
+	* ext/tk/lib/tk.rb: bug fix on handling of Tcl's namespaces.
+
+	* ext/tk/lib/tk/namespace.rb: ditto.
+
+Fri Nov 18 16:47:33 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_file_s_dirname): added checks for some patterns with drive
+	  letter. fixed: [ruby-dev:27738]
+
+	* test/ruby/test_path.rb (test_dirname): added tests for above
+	  patterns.
+
+Fri Nov 18 12:19:16 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.h (S_IFIFO): r,w = IO.pipe; r.stat.pipe? now
+	  returns true on VisualC++6.
+
+Thu Nov 17 17:58:00 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/1.0.rb: added convenience method 'resources'.
+
+	* lib/rss/taxonomy.rb: ditto.
+
+	* test/rss/rss-assertions.rb: added test for 'resources'.
+
+	* test/rss/test_taxonomy.rb: ditto.
+
+Thu Nov 17 17:53:30 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/taxonomy.rb: implemented taxonomy module.
+
+	* test/rss/test_taxonomy.rb: added tests for taxonomy support.
+
+Thu Nov 17 17:40:19 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/1.0.rb: added rdf:Bag.
+
+Thu Nov 17 13:52:00 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: removed needless argument 'prefix'.
+
+	* lib/rss/parser.rb: ditto.
+
+Wed Nov 16 23:24:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (static-ruby): overridable.
+
+	* ext/extmk.rb (parse_args): force to link extensions statically only
+	  if static is given for extstatic.
+
+	* ext/extmk.rb (RUBY, RUBYW): overridable.
+
+Wed Nov 16 01:29:31 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/trackback.rb: added TrackBack prefix.
+
+	* lib/rss/maker/trackback.rb: ditto.
+
+Wed Nov 16 01:26:13 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::VERSION): 0.1.5 -> 0.1.6.
+
+	* test/rss/test_version.rb (RSS::TestVersion#test_version): ditto.
+
+Tue Nov 15 23:54:24 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_load_ok): eaccess() returns 0 on success.
+	  fixed: [ruby-dev:27713]
+
+Tue Nov 15 16:36:03 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* array.c (rb_ary_fill): previous commit disabled this usage:
+
+	    a = [0,1,2,3,4,5,6,7,8,9]
+	    a.fill {|i| a[i] * 10} #=> [nil, nil, ...., nil]
+
+	  previous commit has the advantage of early garbage collection, but
+	  potentially this would break some script. so I reverted behavior.
+
+Tue Nov 15 16:15:23 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (file_load_ok): use eaccess() instead of actually opening
+	  the file.  [ruby-talk:156378]
+
+	* lib/jcode.rb (String::reverse): add new methods.
+	  [ruby-list:41245]
+
+Tue Nov 15 15:49:34 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* array.c (rb_ary_fill): tail elements were vanished when the middle
+	  part of array was filled. (ie: [0,1,2,3,4].fill(-1,2,1) => [0,1,-1])
+
+	* test/ruby/test_array.rb (test_fill): added.
+
+Tue Nov 15 14:39:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_fill): should adjust array length correctly when
+	  an array is expanded in the fill process.  [ruby-core:06625]
+
+Mon Nov 14 23:49:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_readlink): ERANGE will occur only on GPFS.
+	  [ruby-dev:27699]
+
+Mon Nov 14 17:36:22 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_first): RDoc update from Daniel Berger
+	  <djberg96@yahoo.com>.  [ruby-core:06577].
+
+Sun Nov 13 10:55:24 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/uri/common.rb (escape): regard second argument as a
+	  character set. [ruby-dev:27692]
+
+Sat Nov 12 08:36:40 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, eval.c, intern.h: check fd_mask type.
+
+	* configure.in (socketpair): need to be checked.
+
+Fri Nov 11 19:53:47 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* eval.c, intern.h: failed to compile where NFDBITS is defined but
+	  howmany() is not defined. [ruby-dev:27680]
+
+	* io.c (is_socket): failed to compile where S_ISSOCK is not defined.
+
+	* io.c (pipe_open): failed to compile where socketpair is not supported.
+
+Fri Nov 11 08:20:56 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (OUTFLAG): keep trailing spaces.  [ruby-dev:27666]
+
+	* mkconfig.rb: substitution references added.
+
+Fri Nov 11 07:39:49 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* configure.in: undef HAVE_LINK on BeOS. (link(2) always returns
+	  EINVAL, and this causes error in test/fileutils.)
+
+	* file.c: override chown(2) and fchown(2) on BeOS. (these functions
+	  should not change user/group id if -1 is passed as corresponding
+	  argument, and this causes error in test/fileutils too)
+	  [ruby-dev:27672]
+
+Thu Nov 10 21:05:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/shellwords.rb: fix for blank but not empty string.
+	  fixed: [ruby-dev:27663]
+
+Wed Nov  9 08:39:38 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/shellwords.rb: refactored.  [ruby-core:06581]
+
+Tue Nov  8 17:35:53 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* intern.h, eval.c (rb_thread_signal_raise): constified.
+
+	* signal.c: cosmetic change.
+
+Tue Nov  8 15:32:27 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/drb/ssl.rb (DRb::SSLConfig#accept): fixed typo.
+	  [ruby-dev:27560] [ruby-core:4627]
+
+Mon Nov  7 20:54:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c: iconvctl() support.  [EXPERIMENTAL]
+
+Mon Nov  7 16:23:23 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl.h: need to include winsock2.h before including
+	  windows.h by some openssl headers.
+
+Mon Nov  7 13:43:51 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/stubs.c (_nativethread_consistency_check): use simpler
+	  (low cost) way to check whether the Tcl interpreter was compiled
+	  with threads enabled of not.
+
+	* ext/tk/tcltklib.c: reduce warnings.
+
+	* ext/tk/tkutil/tkutil.c: ditto.
+
+Mon Nov  7 00:06:12 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/yaml.rb: removed :nodoc: to generate Kernel doc. [ruby-core:6324]
+
+Sun Nov  6 23:39:13 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (Iconv::BrokenLibrary): exception when detected a
+	  bug of underlying library.
+
+Sun Nov  6 21:43:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/stubs.c (ruby_tcl_create_ip_and_stubs_init): should touch
+	  interpreter after initialization is done. [ruby-dev:27638]
+
+Sun Nov  6 20:13:27 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_readlink): readlink(2) on AIX fails with ERANGE if
+	  buffer size is less than required.  fixed: [ruby-dev:27634]
+
+Sat Nov  5 13:42:50 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, cygwin/GNUmakefile.in (mingw): use def file to alias
+	  symbols.  [ruby-dev:27532]
+
+	* bcc32/mkexports.rb, win32/mkexports.rb: make aliases in DLL.
+
+	* win32/win32.c, win32/win32.h: replace symbols only when RUBY_EXPORT
+	  is defined.
+
+Thu Nov  3 07:57:39 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/open-uri.rb (open_loop): find_proxy should return nil when
+	  proxy does not exist. [ruby-dev:27630]
+
+Wed Nov  2 20:25:28 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: ext/tk/extconf.rb: change the check parameter
+	  for Win32.
+
+Wed Nov  2 19:03:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_rbUpdateObjCmd, ip_rb_threadUpdateObjCmd):
+	  passed improper flags to DoOneEvent().
+
+	* ext/tk/tkutil/tkutil.c: use rb_obj_respond_to() instead of
+	  rb_respond_to().
+
+Tue Nov  1 14:20:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call_super): should call method_missing if super is
+	  called from Kernel method.
+
+	* eval.c (exec_under): frame during eval should preserve external
+	  information.
+
+Tue Nov  1 10:48:49 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: should check ERR_peek_last_error().
+	  [ruby-dev:27597]
+
+	* ext/openssl/ossl.c (ossl_raise): ditto.
+
+Mon Oct 31 17:34:46 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: use proper option for Sun linker. A patch from
+	  Shinya Kuwamura <kuwa at labs.fujitsu.com>.  [ruby-dev:27603]
+
+Mon Oct 31 05:46:08 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_update): input data must
+	  not be empty. [ruby-talk:161220]
+
+	* test/openssl/test_cipher.rb: add test for Cipher#update("").
+
+Mon Oct 31 05:38:26 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb
+	  (WEBrick::HTTPServlet::CGIHandler#do_GET): the value of Set-Cookie:
+	  header field should be splited into each cookie.  [ruby-Bugs:2199]
+
+	* lib/webrick/cookie.rb (WEBrick::Cookie.parse_set_cookie): new method
+	  to parse the value of Set-Cookie: header field.
+
+	* test/webrick/test_cookie.rb, test/webrick/test_cgi.rb,
+	  test/webrick/webrick.cgi: add some test for cookie.
+
+Mon Oct 31 02:33:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_rshift): RDoc fix.  [ruby-core:6351]
+
+	* util.h (strtod): add #undef for platforms defines strtod()
+	  macro.   [ruby-dev:27563]
+
+Mon Oct 31 02:31:41 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_float.rb (test_precision): test by assert_in_delta.
+	  [ruby-dev:27575]
+
+Sat Oct 29 01:58:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/etc/etc.c: document update from mathew <meta@pobox.com>.
+	  [ruby-core:06473]
+
+	* ext/fcntl/fcntl.c: ditto.
+
+Sat Oct 29 16:56:03 2005  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: added seven predicates sunday? to saturday?.
+
+	* lib/date.rb: added two methods {prev,next}_month,
+	  that are almost same as << and >>.
+
+Thu Oct 27 20:34:43 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_allocate): allow subclassing.
+
+Thu Oct 27 16:45:31 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (scan_once): wrong condition to use mbclen2().
+	  [ruby-dev:27535]
+
+Thu Oct 27 11:53:17 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* missing.h, missing/memcmp.c, missing/memmove.c:
+	  ANSI compatible interface.
+
+Wed Oct 26 09:15:48 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/implicit.c (syck_type_id_to_uri): should return
+	  newly allocated memory. otherwise, type_id will be freed
+	  twice. [ruby-dev:27384] [ruby-core:6385]
+
+Wed Oct 26 01:58:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_EXTERN): macro to export symbols in shared
+	  library.  [ruby-core:05528]
+
+	* defines.h, {bcc32,win32,wince}/Makefile.sub (RUBY_EXTERN): moved to
+	  configuration pass.
+
+	* ext/extmk.rb (extmake): RUBY_EXTERN for static linked extensions.
+
+Tue Oct 25 20:06:59 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ruby.h (Qfalse, Qtrue, Qnil, Qundef): make sure these immediate
+	  values have VALUE type. there is an environment where sizeof(VALUE)
+	  != sizeof(int) like IA64. if 32bit integer (Qtrue) is passed to ANYARGS
+	  and received by 64bit integer (VALUE), upper bits may have garbage value.
+	  [ruby-dev:27513]
+
+Tue Oct 25 15:32:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rational.rb: applied documentation patch from Gavin Sinclair
+	  <gsinclair@gmail.com>.  [ruby-core:06364]
+
+	* lib/irb.rb (IRB::Irb::eval_input): handle prompts with newlines
+	  in irb auto-indentation mode.  [ruby-core:06358]
+
+Tue Oct 25 14:21:46 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* gc.c (garbage_collect): sorry, previous commit was incorrect.
+	  [ruby-core:6386]
+
+Tue Oct 25 13:40:16 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* gc.c (garbage_collect): return now whether there're rooms for new
+	  objects, rather than whether GC run. fixed: [ruby-core:6376]
+
+Tue Oct 25 02:12:08 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/markup/simple_markup.rb (SM::SimpleMarkup::LABEL_LIST_RE):
+	  reduce redundant backtrack.  [ruby-talk:161771]
+
+Tue Oct 25 00:35:33 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/*: RDoc documentation from Eric Hodel
+	  <drbrain@segment7.net> added.
+
+Mon Oct 24 21:14:29 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, io.c: use sys/syscall.h if syscall.h is not available.
+	  [ruby-core:06247]
+
+Mon Oct 24 20:38:25 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/Win32API/lib/win32/resolv.rb (get_info): support multiple DNS.
+	  fixed: [ruby-list:40058], [ruby-dev:27479]
+
+Mon Oct 24 11:01:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/canvas.rb (TkCanvasItemConfig::__item_val2ruby_optkeys):
+	  typo fixed.  [ruby-talk:162187]
+
+	* ext/tk/lib/tk/menu.rb (TkMenuEntryConfig::__item_val2ruby_optkeys):
+	  ditto.  [ruby-core:06359]
+
+Mon Oct 24 07:57:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/tk/lib/tk/canvas.rb (TkCanvasItemConfig::__item_val2ruby_optkeys):
+	  typo fixed.  [ruby-talk:162187]
+
+	* ext/tk/lib/tk/menu.rb (TkMenuEntryConfig::__item_val2ruby_optkeys):
+	  ditto.  [ruby-core:06359]
+
+	* lib/matrix.rb (Matrix::initialize): use funcall instead of send
+	  to allow private methods to be called.  A report from
+	  Jean-Claude Arbaut <jcarbaut@laposte.net>.  [ruby-core:06359]
+
+Mon Oct 24 00:41:18 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_sunday): added predicate methods for the days of the
+	  week.  [ruby-list:41340]
+
+Sun Oct 23 07:11:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: improve messages [ruby-core:06325].
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tk/canvas.rb, ext/tk/lib/tk/entry.rb,
+	  ext/tk/lib/tk/frame.rb, ext/tk/lib/tk/image.rb,
+	  ext/tk/lib/tk/itemconfig.rb, ext/tk/lib/tk/labelframe.rb,
+	  ext/tk/lib/tk/listbox.rb, ext/tk/lib/tk/menu.rb,
+	  ext/tk/lib/tk/radiobutton.rb, ext/tk/lib/tk/scale.rb,
+	  ext/tk/lib/tk/spinbox.rb, ext/tk/lib/tk/text.rb,
+	  ext/tk/lib/tk/toplevel.rb: improve conversion of option values.
+
+	* ext/tk/lib/tkextlib/*: ditto.
+
+	* ext/tk/lib/tkextlib/*: update to support ActiveTcl8.4.11.2.
+
+	* ext/tk/lib/tkextlib/trofs/*: support Trofs 0.4.3.
+
+	* ext/tk/lib/tkextlib/tile/*: support Tile 0.7.2.
+
+	* ext/tk/lib/tkextlib/vu/*: support vu 2.3.0.
+
+	* ext/tk/lib/tkextlib/tcllib/*: support Tcllib 1.8 (Tklib 0.3).
+
+Sat Oct 22 23:54:07 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb, lib/mkmf.rb (with_config): support --with-extension
+	  options.  [ruby-dev:27449]
+
+Sat Oct 22 14:25:43 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* util.[hc] (ruby_add_suffix): constified.
+
+	* util.[hc] (ruby_scan_{oct,hex}): fixed typo. (renamed from
+	  scan_{oct,hex})
+
+	* util.c: almostly ANSI styled. (except for functions depending on
+	  macro and K&R technique)
+
+Sat Oct 22 13:26:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (sym_inspect), parse.y (parser_yylex, rb_symname_p): check
+	  if valid as a symbol name more strictly.  [ruby-dev:27478]
+
+	* test/ruby/test_symbol.rb: tests for [ruby-core:03573].
+
+	* time.c (rb_strftime): removed meaningless volatile modifiers, and
+	  concatenate successive nul characters at once.  [ruby-dev:27472]
+
+	* ext/tk/lib/tk/font.rb, ext/tk/lib/tkextlib/ICONS/icons.rb,
+	  ext/tk/sample/tkextlib/treectrl/demo.rb, lib/net/imap.rb,
+	  lib/rss/parser.rb, test/rss/test_content.rb,
+	  test/rss/test_dublincore.rb, test/rss/test_syndication.rb,
+	  test/rss/test_trackback.rb, test/ruby/test_eval.rb,
+	  test/socket/test_socket.rb, test/socket/test_udp.rb:
+	  Object#fcall was renamed as Object#funcall.
+
+Sat Oct 22 10:08:28 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* missing.h, missing/*.c: SUSv3 compatible strcasecmp and strncasecmp,
+	  ANSI compatible strtol and strtoul, and ANSI styled other functions.
+
+Fri Oct 21 19:16:08 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* rubysig.h (CHECK_INTS): fixed typo. (I believe bit-or is improper)
+
+Fri Oct 21 17:49:32 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bin/erb (ERB::Main::run): typo fixed.  [ruby-core:06337]
+
+Fri Oct 21 15:42:28 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* intern.h, struct.c (rb_struct_iv_get): constified.
+
+	* marshal.c: avoid one VC++6 warning for implicit conversion
+	  from int to char.
+
+	* ruby.h: ANSI styled.
+
+	* bcc32/Makefile.sub (HAVE_HYPOT): added.
+
+	* ext/socket/extconf.rb: BeOS is only one platform should call
+	  closesocket, so check __BEOS__ macro directly. (I was worried
+	  accidentally HAVE_CLOSESOCKET is defined on windows again because
+	  it has it)
+
+	* ext/socket/{getaddrinfo.c,socket.c}: ditto.
+
+	... these are all cosmetic changes.
+
+Fri Oct 21 15:23:23 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bignum.c (bignew_1): conversion from `int' to `char' discards
+	  upper bits, (ie. (char)0xff00 -> 0) so it's better to test if
+	  nonzero and set 0 or 1 instead of simply casting ... as a flag usage.
+	  (but I believe this won't cause actual bug in current implementation)
+	  [ruby-dev:27055]
+
+	* time.c: should use LONG_LONG instead of `long long'.
+
+Thu Oct 20 22:22:49 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parser.y (struct parser_params): parser never modify input string.
+
+	* ext/ripper/tools/preproc.rb (prelude): do not append surplus
+	  newlines to fix line numbers.
+
+Thu Oct 20 11:41:57 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* class.c, eval.c, hash.c, st.c, variable.c: changed /* ??? */ stuff
+	  protoize generated to ANYARGS.
+
+Thu Oct 20 11:18:11 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* eval.c, file.c, ruby.c: removed strchr, strrchr, strstr definition
+	  because they are defined in missing.h.
+
+	* missing.h, missing/strchr.c, missing/strstr.c: ANSI styled.
+
+Thu Oct 20 09:36:06 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/mkmf.rb (create_makefile): Borland make seems not to allow
+	  empty dependency list. If this change is not good, please correct
+	  it.
+
+Thu Oct 20 07:55:09 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): get rid of a restriction
+	  of Borland make.  fixed: [ruby-dev:27460]
+
+	* ext/ripper/depend: ditto.
+
+Wed Oct 19 23:58:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): do not create unnecessary empty
+	  directories.  fixed: [ruby-dev:27451]
+
+Wed Oct 19 08:28:32 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_join): elements may contain null pointer strings.
+	  report and fixed by Lloyd Zusman (hippoman): [ruby-core:06326]
+
+Wed Oct 19 02:34:33 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c, eval.c, gc.c, parse.y, regparse.c, sjis.c, time.c:
+	  made internal symbols static.  [ruby-dev:27435]
+
+Tue Oct 18 10:58:27 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/depend, ext/dl/extconf.rb, ext/socket/depend,
+	  ext/socket/extconf.rb: shouldn't define DESTCLEANFILES in depend,
+	  use $distcleanfiles in extconf.rb.
+
+	* win32/Makefile.sub (distclean-local): should remove .config.h.time.
+
+Mon Oct 17 09:42:50 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* mkconfig.rb: fixup configure_args for mswin32 configure.
+
+	* win32/configure.bat (srcdir, target): ditto.
+
+Mon Oct 17 05:01:50 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* env.h: move struct METHOD and struct BLOCK from eval.c to
+	  support NodeWrap and ParseTree.
+
+Sun Oct 16 22:16:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: omit non-existing directories.
+
+Sun Oct 16 14:40:54 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/rinda.rb (Rinda::Tuple#initialize): check remote hash
+	  tuple. fixed: [ruby-list:41227]
+
+	* test/rinda/test_rinda.rb: test it.
+
+Sun Oct 16 03:38:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* rubysig.h (CHECK_INTS): prevent signal handler to run during
+	  critical section.  [ruby-core:04039]
+
+	* eval.c (load_wait): need not to call rb_thread_schedule()
+	  explicitly.  [ruby-core:04039]
+
+	* eval.c (rb_thread_schedule): clear rb_thread_critical.
+	  [ruby-core:04039]
+
+Sun Oct 16 00:13:14 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/configure.bat: remove unnecessary line which prevents
+	  creating Makefile.
+
+Sat Oct 15 23:52:07 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb: (getbinaryfile): allow nil for localfile, and
+	  returns retrieved data if localfile is nil.
+
+	* lib/net/ftp.rb: (gettextfile): ditto.
+
+Sat Oct 15 19:51:29 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* bin/erb: typo fixed, again. thanks, Doug Kearns.
+
+Fri Oct 14 23:09:31 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (MKFILES): update MKFILES if configure files get
+	  changed.
+
+	* win32/configure.bat, win32/setup.mak (configure_args): store
+	  arguments to configure files.
+
+Fri Oct 14 22:05:45 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (ioctl): should set errno.
+
+Fri Oct 14 16:39:37 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/config.rb (Config::FileHandler): :UserDir should be nil.
+	  It is harmful to permit the access to ~/public_html by default.
+	  suggested by Hiroyuki Iwatsuki.
+
+Fri Oct 14 04:58:38 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_obj_instance_exec): create instance_exec and
+	  module_exec which pass arguments to the block.
+
+	* eval.c (rb_f_funcall): rename fcall to funcall to follow
+	  tradition.
+
+Thu Oct 13 23:29:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (HEAPCNT): bison allocates indivisible size.
+	  fixed: [ruby-core:06261]
+
+	* io.c, pack.c, ext/syck/rubyext.c, ext/syck/syck.h, missing/isinf.c:
+	  get rid of warnings.  fixed: [ruby-core:06247]
+
+Wed Oct 12 12:51:56 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.c (Init_openssl): should call
+	  OpenSSL_add_ssl_algorithms().
+
+Wed Oct 12 11:08:54 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* file.c (rb_f_test): typo in RDoc comments.
+
+Tue Oct 11 21:41:58 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_FUNC_ATTRIBUTE): check prefixed attribute form
+	  first.  [ruby-dev:27398]
+
+	* array.c, enum.c, eval.c, util.c: safer function pointer usage.
+	  fixed: [ruby-core:06143]
+
+	* util.h (qsort): removed the definition incompatible to ANSI.
+	  fixed: [ruby-core:06147]
+
+	* eval.c (rb_obj_respond_to): check if obj responds to the given
+	  method with the given visibility.  [ruby-dev:27408]
+
+	* eval.c (rb_respond_to): conform to Object#respond_to?.  [ruby-dev:27411]
+
+Tue Oct 11 00:01:21 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* st.c (st_free_table): do not call free() but xfree().
+	  [ruby-core:06205]
+
+Sat Oct  8 19:49:42 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (Init_Binding): add Binding#dup method.  [yarv-dev:666]
+
+	* io.c (rb_io_init_copy): clear PREP flag for copied IO.
+	  fixed: [ruby-dev:27371]
+
+	* parse.y (rb_parser_malloc, rb_parser_free): manage parser stack on
+	  heap.  [ruby-list:41199]
+
+	* parse.y (ripper_initialize): use rb_respond_to().
+
+	* ext/ripper/depend (check): get rid of re-generating ripper.y always.
+
+	* ext/iconv/charset_alias.rb: parse config.charset_alias file directly.
+
+	* ext/nkf/lib/kconv.rb (Kconv.conv): get rid of nil.to_a.
+
+	* lib/scanf.rb (Scanf::FormatSpecifier#letter, #width): use matched
+	  substring directly.
+
+	* test/ruby/test_assignment.rb, test/ruby/test_iterator.rb: followed
+	  change of sample/test.rb.
+
+	* test/net/http/test_http.rb: removed superfluous splatting stars.
+
+Fri Oct  7 16:41:43 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (splat_value): call rb_Array() to convert svalue to
+	  values.  [ruby-dev:27397]
+
+Fri Oct  7 09:54:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::Cookie::parse): Cookies from Nokia devices may
+	  not be parsed correctly.  A patch from August Z. Flatby
+	  (augustzf) in [ruby-Patches-2595].  [ruby-core:06183]
+
+Thu Oct  6 22:51:30 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_Array): Array() to raise error for objects without
+	  to_ary, nor to_a.
+
+	* object.c (nil_to_a): revert NilClass#to_a.
+
+Thu Oct  6 20:10:38 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c (strscan_free): remove useless code.
+	  [ruby-dev:26368] [ruby-dev:27389]
+
+Thu Oct  6 01:02:19 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (rb_range_beg_len): should return Qfalse for non-range
+	  object.
+
+Wed Oct  5 04:42:38 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/xmlrpc/server.rb (XMLRPC::Server#initialize): should mount the
+	  servlet on "/".
+
+Wed Oct  5 04:06:49 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/xmlrpc/server.rb (XMLRPC::Server#serve): delete wrong call
+	  of "join".
+
+Mon Oct  3 00:04:00 2005  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* pack.c (EXTEND16): [ruby-dev:27383]
+
+Sat Oct  1 23:55:24 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (do_select, rb_w32_select): brush up.
+
+Sat Oct  1 12:57:02 2005  Tanaka Akira  <akr@m17n.org>
+
+	* bignum.c (rb_big_rand): removed.  [ruby-dev:25405]
+
+Sat Oct  1 01:46:51 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI.open_loop): prohibit multiple proxy
+	  options.
+
+Thu Sep 29 10:26:18 2005  Tanaka Akira  <akr@m17n.org>
+
+	* ext/dl/cptr.c (rb_dlptr_s_to_ptr): abolish sizeof(FILE).
+	  [ruby-dev:27317]
+
+Thu Sep 29 10:15:14 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (:proxy_http_basic_authentication): new option.
+
+Thu Sep 29 07:22:05 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_f_send): underscores need to be escaped.
+	  fixed by Doug Kearns.  [ruby-core:06053]
+
+Thu Sep 29 00:57:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ev_const_get), variable.c (rb_const_get_0): retry only when
+	  autoload succeeded.
+
+	* variable.c (rb_autoload_load): now return true if autoload
+	  succeeded.  fixed: [ruby-dev:27331]
+
+Wed Sep 28 23:40:04 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_stat_inspect): constified.
+
+Wed Sep 28 15:12:28 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI#start): req.query_string should
+	  refer to the value of QUERY_STRING. [ruby-list:41186]
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#query_string=):
+	  add new method.
+
+Wed Sep 28 10:45:44 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: cannot compile with Tcl/Tk8.0.x [ruby-dev:27335].
+
+Wed Sep 28 07:56:52 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/yaml/basenode.rb (YAML::BaseNode::match_segment): fix typo.
+	  [ruby-dev:27237], [ruby-core:05854]
+
+	* lib/yaml/tag.rb (Module#yaml_as): suppress warnings.
+
+	* lib/yaml/types.rb (YAML::PrivateType, YAML::DomainType): ditto.
+
+Wed Sep 28 03:16:41 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* rubysig.h: fixed build problem with --enable-pthread on platforms
+	  which don't have setitimer().
+
+Mon Sep 26 22:32:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (set_trace_func): add rb_secure(4) to prevent adding
+	  tracing function.
+
+Mon Sep 26 20:59:28 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* parse.y: changed to ANSI function style.
+
+Sun Sep 25 12:02:04 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* bin/erb: typo fixed.
+
+Sun Sep 25 11:54:11 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/tuplespace.rb (Rinda::TemplateEntry::initialize): pull
+	  up method. Tabs converted to spaces.
+
+Sun Sep 25 09:34:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* parse.y: replaced `foo _((boo))' with `foo(boo)'.
+
+Sun Sep 25 08:19:53 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rss/test_content.rb, test/rss/test_syndication.rb: use fcall
+	  instead of send in order to override visibility.
+
+Sun Sep 25 01:46:43 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-calculate-indent): arrange deep-indent
+	  closing parenthesis at same column as the opening.
+
+Sun Sep 25 01:33:41 2005  Tanaka Akira  <akr@m17n.org>
+
+	* process.c (proc_setrlimit): make the third argument (rlim_max)
+	  optional.
+
+Sun Sep 25 00:42:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-expr-beg): deal with heredoc separately.
+	  fixed: [ruby-list:41168]
+
+	* misc/ruby-mode.el (ruby-calculate-indent): not to deepen indent
+	  level for continuous line inside parentheses.
+	  http://nabeken.tdiary.net/20050915.html#p02
+
+Sat Sep 24 21:19:39 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c: document enhancement.
+
+	* ext/strscan/strscan.c: update copyright year.
+
+	* ext/strscan/strscan.c: update coding style.
+
+Sat Sep 24 20:20:05 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/net/http/test_http.rb (teardown): Net::HTTP.version_1_1 breaks
+	  many other tests; ensure that Net::HTTP is version 1.2 after test.
+	  [ruby-dev:27312]
+
+Sat Sep 24 11:44:28 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/net/http/test_http.rb: new file.
+
+Sat Sep 24 08:54:05 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (cd): no longer accept :noop option, related
+	  code is useless. [ruby-core:05858] [ruby-Bugs:2494]
+
+Sat Sep 24 08:30:00 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#sub): new method.
+
+Sat Sep 24 08:29:36 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb: fix visibility of FileUtils::NoWrite, Verbose,
+	  DryRun. [ruby-core:05954]
+
+	* test/fileutils/test_nowrite.rb: test it.
+
+	* test/fileutils/test_dryrun.rb: new file.
+
+	* test/fileutils/test_verbose.rb: new file.
+
+Sat Sep 24 07:59:01 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* sample/ripper/colorize.rb: removed (replaced by ruby2html.rb).
+
+	* sample/ripper/ruby2html.rb: added.
+
+Sat Sep 24 06:35:15 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper: no longer generates .rb files.
+
+	* parse.y (Init_ripper): ripper_init_eventids*() takes 1 argument,
+	  self (class Ripper).
+
+	* ext/ripper/depend: target removed: `lib/ripper/core.rb'.
+
+	* ext/ripper/depend: new target `eventids2table.c'.
+
+	* ext/ripper/depend: new target `check'.
+
+	* ext/ripper/eventids2.c: include eventids2table.c.
+
+	* ext/ripper/eventids2.c: initialize SCANNER_EVENT_TABLE.
+
+	* ext/ripper/extconf.rb: update $cleanfiles list.
+
+	* ext/ripper/tools/generate.rb: no longer generate ripper/core.rb.
+
+	* ext/ripper/tools/generate.rb: new mode `check'.
+
+	* ext/ripper/tools/generate.rb: new mode `eventids2table'.
+
+	* ext/ripper/lib/ripper/core.rb.in: removed.
+
+	* ext/ripper/lib/ripper/core.rb: added.
+
+	* ext/ripper/lib/ripper/filter.rb: update copyright year.
+
+	* ext/ripper/lib/ripper/lexer.rb: ditto.
+
+	* ext/ripper/lib/ripper/sexp.rb: ditto.
+
+Sat Sep 24 02:40:20 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb: document update from James Edward Gray II
+	  <james@grayproductions.net>.  [ruby-core:05942]
+
+Sat Sep 24 02:05:51 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (proc_daemon): should restrict execution on levels
+	  higher than $SAFE=2.  suggested by URABE Shyouhei
+	  <shyouhei@ice.uec.ac.jp>.
+
+Fri Sep 23 20:10:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/ripper/tools/generate.rb, ext/ripper/tools/preproc.rb: StringIO
+	  is not available for miniruby.  fixed: [ruby-dev:27307]
+
+Fri Sep 23 17:36:48 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: avoid core dump with WIN32OLE_EVENT.
+	  [ruby-dev:27133]
+
+Fri Sep 23 16:27:39 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/forwardable.rb: replaced by new implementation from
+	  <Daniel.Berger@qwest.com>.  [ruby-core:05899]
+
+Fri Sep 23 07:07:47 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ripper/depend: use --output option instead of redirect;
+	  nmake does not remove a target when the target file is created by
+	  redirect. [ruby-dev:26466]
+
+	* test/ripper/tools/preproc.rb: new option --output.
+
+Fri Sep 23 06:57:52 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ripper/tools/generate.rb: check parser event arity.
+
+	* test/ripper/tools/generate.rb: detect crash of parser-event-IDs
+	  and scanner-event-IDs.
+
+Fri Sep 23 06:01:30 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_file.rb: check File#chown(nil,nil).
+	  [ruby-dev:27140]
+
+Fri Sep 23 05:57:23 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper: refactoring code generation tools. [ruby-dev:27247]
+	  [ruby-dev:27273]
+
+	* ext/ripper/depend: use generate.rb.
+
+	* ext/ripper/lib/ripper/core.rb: removed.
+
+	* ext/ripper/tools/generate-eventids1.rb: removed (code moved to
+	  generate.rb).
+
+	* ext/ripper/tools/generate-ripper_rb.rb: removed (code moved to
+	  generate.rb).
+
+	* ext/ripper/tools/list-parse-event-ids.rb: removed (code moved to
+	  generate.rb).
+
+	* ext/ripper/tools/list-scan-event-ids.rb: removed (code moved to
+	  generate.rb).
+
+	* ext/ripper/lib/ripper/core.rb: removed.
+
+	* ext/ripper: refactoring tests. [ruby-dev:27273]
+
+	* ext/ripper/test/check-event-arity.rb: removed (code moved to
+	  tools/generate.rb).
+
+	* ext/ripper/test/check-event-coverage.rb: removed (code moved to
+	  test/ripper/test_parser_events.rb).
+
+	* ext/ripper/test/check-scanner-event-coverage.rb: removed (code
+	  moved to test/ripper/test_scanner_events.rb).
+
+	* ext/ripper/test/list-called-events.rb: removed.
+
+	* ext/ripper/test/src_rb: removed.
+
+	* ext/ripper/test/validate.rb: removed.
+
+	* test/ripper/test_scanner_events.rb: check event coverage.
+
+	* ext/ripper/lib/ripper/core.rb.in: update copyright year.
+
+Thu Sep 22 23:40:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (find_executable0): default path if environment is not
+	  set.  [ruby-dev:27281]
+
+	* ext/ripper/extconf.rb (have_command): replaced with find_executable.
+
+Thu Sep 22 17:31:48 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/readline/test_readline.rb (TestReadline::replace_stdio):
+	  merged the patch of [ruby-dev:25232] instead of [ruby-dev:25223].
+	  (merged from ruby_1_8 branch)
+
+Wed Sep 21 23:30:44 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): generalized nmake dependent code.
+
+Wed Sep 21 14:16:30 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/ripper/depend (SUFFIXES): no longer needed.
+
+	* ext/ripper/depend (c): avoid nmake problem. fixed [ruby-dev:27191]
+
+Wed Sep 21 08:52:25 2005  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/token.c: correctly compute indentation of a block
+	  scalar's parent node. [ruby-talk:150620]
+
+Wed Sep 21 08:20:24 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja: add new features.
+
+Wed Sep 21 07:43:58 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (default_argv, Arguable#options): defaults strings
+	  to be parsed to Arguable instance.
+
+Wed Sep 21 02:44:09 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (path_check_0): disallow sticky world writable directory
+	  in PATH (and $LOAD_PATH).  [ruby-dev:27226]
+
+Wed Sep 21 00:32:22 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_idiv): 1.div(1.0) should return integer value.
+	  [ruby-dev:27235]
+
+Tue Sep 20 22:25:43 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/wait/lib/nonblock.rb: disable on platforms non-blocking flag
+	  is not available.  fixed: [ruby-dev:27187]
+
+	* file.c (rb_stat_inspect): protoized function pointer.
+
+Tue Sep 20 18:23:04 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (thread_mark): mark th->last_status.  [ruby-dev:27179]
+
+Tue Sep 20 18:20:33 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/yaml.rb: require 'yaml/constants'.  [ruby-core:5776]
+
+Tue Sep 20 17:48:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client::do_rpc): add charset
+	  information to content-type header.[ruby-core:5127]
+
+	* lib/xmlrpc/server.rb (CGIServer::serve): ditto.
+
+	* lib/xmlrpc/server.rb (ModRubyServer::serve): ditto.
+
+	* lib/xmlrpc/server.rb (WEBrickServlet::service): ditto.
+
+Tue Sep 20 17:26:42 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/webrick/test_cgi.rb: set ENV["PATH"] to CGIEnvPath on
+	  windows. bcc32's runtime is not installed into system directory,
+	  so it cannot be found without this setting. [ruby-dev:27166]
+
+Tue Sep 20 17:14:10 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* io.c: PIPE_BUF is not defined on BeOS. use _POSIX_PIPE_BUF instead.
+	  [ruby-dev:27185]
+
+Tue Sep 20 17:10:38 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/dbm/test_dbm.rb (TestDBM::test_s_open_error): remove
+	  test_s_open_error test to detect duplicate open.
+	  [ruby-dev:27202]
+
+Tue Sep 20 15:39:40 2005  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/emitter.c (syck_scan_scalar): prevent indicators from
+	  appearing alone or at the end of plain scalars. [ruby-core:5826]
+
+	* ext/syck/emitter.c (syck_emit_scalar): treat typed scalar nodes
+	  as complex keys.
+
+	* lib/syck.h: version 0.60.
+
+	* lib/yaml/basenode.rb (YAML::BaseNode#at): transform keys during
+	  key searches.
+
+	* ext/syck/rubyext.c: loading of binary-typed nodes.  prevent
+	  emission of plain strings that look like symbols, but which aren't.
+
+Tue Sep 20 05:48:26 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/xmlrpc/test_webrick_server.rb (setup_http_server):
+	  should not include 'webrick/https' unless 'use_ssl' because
+	  it fails where openssl is not installed.
+
+Tue Sep 20 01:24:45 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (splat_value): use to_a to splat non Array object.
+
+	* object.c (nil_to_a): remove nil.to_a.  [experimental]
+
+Tue Sep 20 01:01:41 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mathn.rb (Fixnum): remove debug print.
+
+	* lib/rational.rb (Rational): ditto.
+
+Tue Sep 20 00:34:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_close): call rb_io_close() directly if io is a T_FILE
+	  object.  [ruby-dev:27156]
+
+Mon Sep 19 18:58:10 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* file.c (rb_file_chown): should accept nil. [ruby-dev:27171]
+
+Mon Sep 19 18:29:54 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (file_expand_path): allow pathnames to expand.
+	  [ruby-dev:27152]
+
+Mon Sep 19 15:12:15 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/depend: do not make ripper/core.rb. [ruby-dev:26462]
+
+Mon Sep 19 14:49:19 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/eventids2.c: add prefix `t' to tLAMBDA-related lexer
+	  events.
+
+	* ext/ripper/lib/ripper/core.rb: updated.
+
+Mon Sep 19 14:39:46 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y (do_block): do_block event dispatches 2 args.
+	  [ruby-dev:26964]
+
+	* ext/ripper/lib/ripper/core.rb: updated.
+
+	* ext/ripper/tools/list-parser-event-ids.rb: check arity mismatch.
+
+Mon Sep 19 07:45:37 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_pkey.h, ossl_pkey_rsa.c, ossl_pkey_dsa.c:
+	  an instance variable "private" is added to OpenSSL::PKey class.
+	  this ivar is a flag that shows whether there is a private key
+	  in the instance.
+
+	* ext/openssl/ossl_engine.c: (ossl_engine_load_privkey): set private
+	  key flag.
+
+	* test/openssl/test_pkey_rsa.rb: add test about private detection.
+
+Mon Sep 19 06:38:03 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb: method renaming: collect_methods ->
+	  collect_method.
+
+Mon Sep 19 05:58:59 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb: use module_function instead of single extend.
+
+	* test/fileutils/test_fileutils.rb: test existence of singleton
+	  methods.
+
+Mon Sep 19 05:32:41 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (remove_entry_secure): does not use chdir(2).
+
+Mon Sep 19 03:17:48 2005  Tanaka Akira  <akr@m17n.org>
+
+	* file.c (rb_thread_flock): wrap the flock system call by
+	  TRAP_BEG/TRAP_END to enable signals.  [ruby-dev:27122]
+
+	* ext/socket/socket.c (bsock_send): wrap the sendto and send system
+	  call by TRAP_BEG/TRAP_END to enable signals when writing to a socket
+	  which is full.  [ruby-dev:27132]
+
+	* io.c (rb_io_syswrite): wrap the write system call by
+	  TRAP_BEG/TRAP_END to run signal handler in syswrite method.
+	  [ruby-dev:27134]
+
+Mon Sep 19 01:07:38 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (Init_Numeric): should define Fixnum#div.
+	  [ruby-dev:27129]
+
+	* file.c (rb_thread_flock): wrap flock(2) by TRAP_BEG and
+	  TRAP_END.  [ruby-dev:27122]
+
+	* file.c (rb_file_join): call FilePathValue() to all Pathnames to
+	  join.  [ruby-dev:27127]
+
+	* file.c (rb_get_path): call StringValueCStr() to ensure no nul
+	  bytes in path strings.
+
+	* gc.c (garbage_collect): need value for return.  [ruby-dev:27127]
+
+Sun Sep 18 02:10:47 2005  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb: remove comments that are bungling up
+	  the rdoc and ri output.  output symbols as plain scalars.
+
+	* ext/syck/rubyext.c (syck_emitter_reset): emit headless
+	  documents always.
+
+	* ext/syck/emitter.c (syck_scan_scalar): quote scalars with any
+	  kind of surrounding line space, tabs or spaces alike.
+
+	* ext/syck/token.c: accept tabs as whitespace, not for indentation,
+	  but strip from plain scalars.
+
+	* test/yaml/test_yaml.rb: remove outdated tests.
+
+Sun Sep 18 01:10:37 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (garbage_collect): return false if no GC run.
+
+Sat Sep 17 23:25:04 2005  sheepman  <sheepman@sheepman.sakura.ne.jp>
+
+	* lib/mathn.rb (Rational::inspect): should preserve original
+	  operand.  [ruby-core:05806]
+
+Sat Sep 17 23:20:27 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::Cookie): should handle multiple values for a
+	  cookie name.  [ruby-talk:156140]
+
+	* test/socket/test_tcp.rb (TestTCPSocket::test_recvfrom): typo
+	  fixed.  [ruby-dev:27123]
+
+Sat Sep 17 20:58:56 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_select): fixed deadlock bug.
+	  because select(2) modifies its fd_set arguments, it must be
+	  restored sometimes.
+
+	* win32/win32.c (rb_w32_select): performance improvement when
+	 'always readable/writable handles' and sockets are passed.
+	  sockets should be polled every time.
+
+	    require "net/http"
+
+	    Thread.new {
+	      loop do
+		STDOUT.write(".") # busy on console (this is worst case though)
+	      end
+	    }
+
+	    # socket operation took long time. (sometimes timed out)
+	    Net::HTTP.start("www.ruby-lang.org") do |http|
+		http.get("/cgi-bin/cvsweb.cgi/ruby/array.c?rev=1.179")
+	    end
+
+Sat Sep 17 14:54:40 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_readpartial.rb (test_open_pipe, test_with_stdio):
+	  these tests are working now, so turned on. (windows)
+
+Sat Sep 17 14:18:15 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_select): I hope performance problem was
+	  solved.
+
+Sat Sep 17 13:45:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_select): console support is back.
+	  but still has performance problem because I loosely took 1 second
+	  for wait time. I'll fix it later. (The reason I drastically changed
+	  the code is that I wanted to implement the fileset management as
+	  single function, and I was worried that if pipe or console
+	  was always available, socket may not be processed any time)
+
+Sat Sep 17 11:24:16 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_select): select for socket didn't work.
+	  this caused deadlock in drb test. this happened because GetFileType
+	  for socket handle returns FILE_TYPE_PIPE. Of course, it's not a
+	  pipe. So socket handle didn't reach winsock's select function.
+
+	* win32/win32.c (rb_w32_select): read for pipe was still blocked
+	  even if writer handle was closed.
+
+	    r,w = IO.pipe
+
+	    Thread.new {
+	      sleep 3; puts "------- 1"
+	      w.puts("foo")
+	      sleep 3; puts "------- 2"
+	      w.puts("boo")
+	      sleep 3; puts "------- 3"
+	      w.close
+	    }
+
+	    until r.eof? # should break by w.close but didn't.
+	      puts r.gets
+	    end
+
+	* win32/win32.c (rb_w32_select): temporary reverted console support
+	  but it'll be back soon.
+
+Sat Sep 17 10:42:13 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string and bg_eval_string
+	  should call Kernel.eval on caller's safe-level instead of slave's
+	  safe-level (Of course, the given script should be evaluated on
+	  slave's safe-level).
+
+Sat Sep 17 09:45:26 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_substr): should propagate taintness even for
+	  empty strings.  [ruby-dev:27121]
+
+	* string.c (rb_str_aref): should infect result if range argument
+	  is tainted.  [ruby-dev:27121]
+
+Sat Sep 17 08:35:39 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb (RSS::Maker::ItemsBase#normalize): fixed
+	  strange RSS::Maker::Item#max_size behavior.
+	  Thanks to Kazuhiko <kazuhiko@fdiary.net>.
+
+	* test/rss/test_maker_1.0.rb (RSS::TestMaker10#test_items): ditto.
+
+Sat Sep 17 08:02:53 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: supported DIGEST-MD5.  Thanks, Mathieu Arnold.
+
+	* lib/net/imap.rb: use fcall instead of send.  Thanks, Satoru
+	  Takabayashi.
+
+Fri Sep 16 22:45:49 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_extname): empty string for path name ending with a
+	  period.  fixed: [ruby-core:05651]
+
+	* file.c (rb_file_join): smarter behavior at edge cases.
+	  fixed: [ruby-core:05706]
+
+	* gc.c (rb_memerror, ruby_xmalloc, ruby_xrealloc, rb_newobj): just
+	  abandon if no memory available, when interpreter is not running.
+	  [ruby-dev:27104]
+
+	* gc.c (garbage_collect): return whether GC could run.
+
+	* dir.c (rb_push_glob): fix delimiter bug.  fixed: [ruby-dev:27105]
+
+	* dir.c (dir_s_aref, dir_s_glob): allow multiple patterns.
+	  [ruby-dev:27110]
+
+	* win32/win32.c (cmdglob): enable brace expansion.
+
+Fri Sep 16 18:34:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/node.c (syck_replace_str): was using return from the
+	  void function.  a patch from MIYAMUKO Katsuyuki
+	  <miyamuko at mtb.biglobe.ne.jp>.  [ruby-dev:27111]
+
+Fri Sep 16 14:48:48 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: fix typo on MultiTkIp#bg_eval_string
+
+Fri Sep 16 12:02:12 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_resolver_transfer): remove C++ style
+	  comment (//).  [ruby-core:05793]
+
+Fri Sep 16 00:17:03 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/logger/test_logger.rb: unintentionally overwritten changes by
+	  Usa.  reverted.
+
+Fri Sep 16 00:03:11 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#initialize): should set
+	  $stdout.binmode.
+
+Thu Sep 15 23:25:21 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/{soap,wsdl,xsd}, test/{soap,wsdl,xsd}: imported soap4r/1.5.5.
+
+	  #nnn is a ticket number at http://dev.ctor.org/soap4r
+
+	  * SOAP
+
+	    * allow to configure an envelope namespace of SOAP request. (#124)
+		TemporaryNamespace = 'http://www.w3.org/2003/05/soap-envelope'
+		@client.options["soap.envelope.requestnamespace"] =
+		  TemporaryNamespace
+		@client.options["soap.envelope.responsenamespace"] =
+		  TemporaryNamespace
+		@client.do_proc(...)
+
+	    * let SOAP request XML indent space configurable.  see
+	      "soap.envelope.no_indent" option. (#130)
+
+	    * let external CES configurable.
+	      ex. client["soap.mapping.external_ces"] = 'SJIS'.  $KCODE is used
+	      by default. (#133)
+		external CES ::= CES used in Ruby object of client and server
+		internal CES ::= CES used in SOAP/OM
+
+	    * add iso-8859-1 external CES support. (#106)
+
+	    * fixed illegal 'qualified' handling of elements.  it caused
+	      ASP.NET interoperability problem. (#144)
+
+	    * added 'soap.envelope.use_numeric_character_reference' (boolean)
+	      option to let query XML use numeric character reference in XML,
+	      not plain UTF-8 character.  !GoogleSearch server seems to not
+	      allow plain UTF-8 character since 2005-08-15 update. (#147)
+
+	    * SOAP::Header::SimpleHeader (de)serialization throws an exception
+	      on !SimpleHeader.on_(in|out)bound when header is a String.  so we
+	      could not use a simple single element headerItem.  fixed.  thanks
+	      to emil. (#129)
+
+	    * out parameter of rpc operation did not work.  (#132)
+
+	    * follow HTTP redirect only if using http-access2.  (#125) (#145)
+
+	    * add a workaround for importing an WSDL whose path begins with
+	      drive letter.  (#115)
+
+	  * WSDL
+
+	    * SOAP Data which is defined as a simpletype was not mapped
+	      correctly to Ruby obj when using wsdl2ruby.rb generated classdef
+	      file. (#123)
+
+	    * rpc/literal support. (#118)
+
+	    * re-implemented local element qualify/unqualify control.  handles
+	      elementFormDefault and form in WSDL.  (#119)
+
+	    * Array of an element which has simpleType causes a crash. (#128)
+
+	    * parameterOrder may not contain return part so it can be shorter
+	      than parts size.  Thanks to Hugh.  (#139)
+
+	  * Samples
+
+	    * added !BasicAuth client sample. (#117)
+
+	    * added Base64 client/server sample.
+
+	    * added Flickr SOAP interface client sample. (#122)
+
+	    * added !SalesForce client sample. (#135)
+
+	    * updated Thawte CA certificate for !GoogleAdWords sample.
+
+	    * updated a client script with the newer version made by Johan.
+	      thanks!
+
+	    * shortened long file names. (#120)
+
+	    * fixed typo in authheader sample. (#129)
+
+	    * updated deprecated method usage.  (#138)
+
+Thu Sep 15 22:40:27 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_signal.rb (test_exit_action): skip the test using
+	  fork on fork-less platforms.
+
+Thu Sep 15 13:54:33 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb: add :read_timeout option.
+	  [ruby-core:4848]
+
+Thu Sep 15 11:39:18 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/dialog.rb: If a dialog does not show up yet,
+	  TkDialogObj#name raises an exception. [ruby-talk:156109]
+
+Thu Sep 15 11:01:58 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pipe_exec): remove unnecessary CloseHandle().
+
+	* win32/win32.c (extract_console_fd, peek_console): new functions.
+
+	* win32/win32.c (rb_w32_select): check consoles by polling them.
+
+Thu Sep 15 00:18:24 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/net/telnet.rb (Net::Telnet::waitfor): replace sysread with
+	  readpartial.  [ruby-talk:127641]
+
+Wed Sep 14 23:28:28 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (collect_file_fd): rename from extract_file_fd.
+
+	* win32/win32.c (extract_pipe_fd, peek_pipe): new functions.
+
+	* win32/win32.c (rb_w32_select): check pipes by polling them.
+
+Wed Sep 14 22:40:26 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (ruby_glob): glob function not using ruby exception system.
+
+Wed Sep 14 17:24:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c: changed `foo (*bar)_((boo))' to `foo (*bar)(boo)`.
+
+	* enumerator.c, eval.c, gc.c, intern.h, io.c, process.c, ruby.c,
+	  ruby.h, signal.c: ditto.
+
+Wed Sep 14 15:06:22 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bignum.c: changed `foo _((boo))' to `foo(boo)`. [ruby-dev:27056]
+
+	* defines.h, dir.c, dln.h, enumerator.c, env.h, error.c, eval.c, file.c,
+	  gc.c, hash.c, inits.c, intern.h, io.c, lex.c, marshal.c, missing.h,
+	  node.h, numeric.c, pack.c, process.c, re.h, ruby.c, ruby.h, rubyio.h,
+	  rubysig.h, signal.c, sprintf.c, st.h, string.c, struct.c, time.c,
+	  util.c, util.h, variable.c: ditto.
+
+Tue Sep 13 22:09:40 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb (Logger): added formatter accessor to logger for
+	  dictating the way in which the logger should format the messages it
+	  displays.  Thanks to Nicholas Seckar (cf. [ruby-talk:153391]) and
+	  Daniel Berger.
+
+	* lib/logger.rb (Logger): added VERSION constant.
+
+	* lib/logger.rb: removed document for LogDevice. It is an
+	  implementation detail and is not a public interface.
+
+	* test/logger/test_logger.rb: added tests.
+
+Tue Sep 13 21:47:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (BEGIN_CALLARGS): pop halfly pushed status.
+	  fixed: [ruby-dev:26881]
+
+Tue Sep 13 20:24:37 2005  Tanaka Akira  <akr@m17n.org>
+
+	* ruby.h (PRINTF_ARGS): new macro for printf style argument checking.
+
+Tue Sep 13 15:41:29 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: wrote documentation of HTTPRequest/HTTPResponse
+	  classes.
+
+Tue Sep 13 14:27:47 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* string.c, missing.h: failed to build on powerpc-apple-darwin7.9.0
+	  because of crypt argument's constness mismatch. (I hope this works)
+	  (http://mput.dip.jp/autobuild/ruby-trunk/log/20050913T110001.gz)
+
+Tue Sep 13 12:33:05 2005  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml.rb: reworking YAML::Stream to use the new
+	  emitter.
+
+	* lib/yaml/stream.rb: ditto.
+
+	* lib/yaml/rubytypes.rb: added Object#yaml_new.
+
+	* lib/yaml/tag.rb: the tag_subclasses? method now
+	  shows up in the class.  allow taguri to be set using an accessor.
+	  continue support of Object#to_yaml_type.
+
+	* ext/syck/rubyext.c: new emitter code.  yaml_new and yaml_initialize
+	  get called, should they be present.  consolidated all the diaspora of 	  internal node types into the family below YAML::Syck::Node -- Map,
+	  Seq, Scalar -- all of whom are SyckNode structs pointing to
+	  Ruby data.  moved Object#yaml_new into the node_import and made it
+	  the default behavior.  the target_class is always called with
+	  yaml_new, prepended a parameter, which is the klass.  loaded nodes
+	  through GenericResolver show their style.
+	  new Resolver#tagurize converts type ids to taguris.
+
+	* ext/syck/implicit.re: were 'y' and 'n' seriously omitted??
+
+	* ext/syck/emitter.c: renovated emitter, walks the tree in advance.
+	  consolidated redundant block_styles struct into
+	  the scalar_style struct.  (this means loaded nodes can now
+	  be sent back to emitter and preserve at least its very basic
+	  formatting.)
+
+	* ext/syck/gram.c: headless documents of any kind allowed.
+
+	* ext/syck/node.c: new syck_replace_str methods and syck_empty_*
+	  methods for rewriting node contents, while keeping the ID
+	  and other setup info.  added syck_seq_assign.
+
+	* ext/syck/syck.h: reflect block_styles and new node functions.
+
+Tue Sep 13 08:09:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (new_ostruct_member): Object#send no longer call
+	  private methods.  [ruby-dev:27044]
+
+	* test/rss/test_dublincore.rb, test/rss/test_trackback.rb,
+	  test/ruby/test_eval.rb, test/socket/test_socket.rb: ditto.
+
+	* test/ruby/test_lambda (test_call_with_block): lambda makes new scope
+	  for formal block parameter.
+
+Tue Sep 13 01:17:45 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_save_safe_level): no need to restrict safe level
+	  memoize in $SAFE>=3.  [ruby-dev:27050]
+
+Tue Sep 13 00:02:33 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (apply2files): stricter callback definition.
+
+	* file.c (rb_path_check): constified.
+
+Mon Sep 12 20:53:06 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/openssl/test_pkcs7.rb (test_enveloped): skip this test
+	  to avoid a bug of PKCS7_encrypt() (only if ext/openssl is
+	  compiled with OpenSSL-0.9.7d or earlier versions).
+	  http://www.mail-archive.com/openssl-dev@openssl.org/msg17376.html
+
+Mon Sep 12 20:32:00 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.[hc] (rb_w32_argv_size, ...): reverted my latest change
+	  to avoid incompatible pointer warning. (mingw32)
+
+Mon Sep 12 19:58:53 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dln.c: avoid warning of const to non-const conversion.
+	  [ruby-dev:27041]
+
+	* eval.c, io.c, ruby.c: ditto.
+
+Mon Sep 12 19:26:29 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* array.c: moved to ANSI function style from K&R function style.
+	  (used protoize on windows, so still K&R remains on #ifdef part of
+	   other platforms. And  `foo _((boo))' stuff is still there)
+	   [ruby-dev:26975]
+
+	* bignum.c, class.c, compar.c, dir.c, dln.c, dmyext.c, enum.c,
+	  enumerator.c, error.c, eval.c, file.c, gc.c, hash.c, inits.c,
+	  io.c, main.c, marshal.c, math.c, numeric.c, object.c, pack.c,
+	  prec.c, process.c, random.c, range.c, re.c, regcomp.c, regenc.c,
+	  regerror.c, regexec.c, regparse.c, regparse.h, ruby.c, signal.c,
+	  sprintf.c, st.c, string.c, struct.c, time.c, util.h, variable.c,
+	  version.c: ditto.
+
+Mon Sep 12 14:03:33 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* test/dbm/test_dbm.rb: remove locking test, which may not be
+	  supported on some platforms.  [ruby-dev:27030]
+
+Sun Sep 11 23:23:02 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (starttls): supported the STARTTLS command.
+
+Sun Sep 11 22:18:07 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* bin/erb (ERB::Main#run): set ERB#filename so that it is used
+	  when reporting syntax/runtime errors. Tabs converted to spaces.
+
+Sat Sep 10 22:34:19 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c, bignum.c: protoize.
+
+Sat Sep 10 00:23:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (splat_value): simpler and consistent array conversion
+	  for argument splat.  [yarv-dev:599]
+
+Fri Sep  9 16:45:25 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_times): make empty strings to keep taintness,
+	  and a little improvement.  [ruby-dev:26900]
+
+	* ext/iconv/iconv.c (iconv_try), ext/iconv/extconf.rb: get rid of meta
+	  characters in command line option.  fixed: [ruby-talk:155369]
+
+	* ext/iconv/iconv.c: protoized.
+
+Thu Sep  8 14:58:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* merged a patch from Takahiro Kambe <taca at back-street.net> to
+	  support DragonFly BSD.  [ruby-dev:26984]
+
+Thu Sep  8 13:14:57 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
+
+	* missing/strchr.c (strrchr): fixed a bug in detecting NUL in a
+	  string.  [ruby-dev:26985]
+
+Wed Sep  7 17:29:27 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_by_id):
+	  OpenSSL::Engine.by_id calls given block before calling
+	  ENGINE_init (block parameter is the return value of this method
+	  itself).  this functionality is useful to load dynamic shared
+	  engines.
+
+		require "openssl"
+		pkcs11 = OpenSSL::Engine.by_id("dynamic"){|e|
+		  e.ctrl_cmd("SO_PATH", "/usr/lib/opensc/engine_pkcs11.so")
+		  e.ctrl_cmd("LIST_ADD", "1")
+		  e.ctrl_cmd("LOAD")
+		}
+		pkcs11.ctrl_cmd("PIN", "secret")
+		key = pkcs11.load_private_key
+
+	* ext/openssl/ossl_engine.c (ossl_engine_ctrl_cmd): new method
+	  OpenSSL::Engine#ctrl_cmd. it wraps ENGINE_ctrl_cmd_string.
+
+	* ext/openssl/ossl_engine.c (ossl_engine_get_cmds): new method
+	  OpenSSL::Engine#cmds. it returns engine command definitions.
+
+Wed Sep  7 15:48:37 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (asn1str_to_str): new function.
+
+	* ext/openssl/ossl_pkcs7.c: new class OpenSSL::PKCS7::RecipientInfo.
+	  this class wraps PKCS7_RECIP_INFO struct.
+
+	* ext/openssl/ossl_pkcs7.c: OpenSSL::PKCS7::Signer is renamed to
+	  OpenSSL::PKCS7::SignerInfo. ("Signer" remains as an alias of
+	  SignerInfo.)
+
+	* test/openssl/test_pkcs7.rb: new file.
+
+Wed Sep  7 12:55:08 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb: abolish mod === tempfile to avoid a problem
+	  [ruby-dev:26967].
+
+Wed Sep  7 10:45:15 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_switch): convert all exceptions to
+	  SystemExit.  fixed: [ruby-core:05724]
+
+	* eval.c (rb_thread_terminated): show backtrace before propagate
+	  exceptions to main thread.
+
+Wed Sep  7 09:21:41 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.[hc] (rb_w32_utime): constified.
+
+	* win32/win32.h (rb_w32_stat): added prototype.
+
+	* win32/win32.[hc] (rb_w32_argv_size,rb_w32_join_argv,rb_w32_aspawn):
+	  changed `char *const *' to `const char *const *'. (constify string)
+
+Wed Sep  7 08:35:04 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, configure.in (MINIOBJS): miniruby on HP-UX can not load
+	  extension libraries.
+
+	* bignum.c (bignew_1, bigadd): K&R style argument actually can't be
+	  defined as char.
+
+	* missing/vsnprintf.c: ANSI compiler supports const keyword.
+
+	* ext/digest/sha2/extconf.rb: reject platforms which has inttypes.h
+	  but no 64bit integer.
+
+	* lib/mkmf.rb (what_type?): guesstimate type.
+
+	* ext/etc/etc.c (setup_passwd), ext/etc/extconf.rb: pw_age might be
+	  char*.  fixed: [ruby-core:05470]
+
+Wed Sep  7 08:32:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_times): should taint empty strings as well.
+
+	* object.c (Init_Object): make class_variable_{get,set} public.
+	  [ruby-dev:26965]
+
+Mon Sep  5 22:28:46 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt, mlhs_node, lhs, arg, method_call): aref_args might be
+	  nothing.  fixed: [ruby-dev:26952]
+
+	* ext/ripper/eventids2.c: added new tokens.  fixed: [ruby-dev:26952]
+
+Mon Sep  5 17:03:07 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/find.rb: should raise ENOENT if root entry does not exist.
+	  [ruby-list:41054]
+
+	* lib/ostruct.rb: a patch from Florian Gross <florgro at gmail.com>
+	  merged to allow recursive inspect (and to_s) for OpenStruct.
+	  [ruby-core:05532]
+
+Mon Sep  5 08:20:19 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/observer.rb: a patch from nornagon <nornagon at gmail.com>
+	  merged to allow arbitrary names for update methods.
+	  [ruby-core:05416]
+
+Mon Sep  5 07:01:12 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/openssl/lib/openssl/buffering.rb (Buffering#do_write):
+	  should clear data from the buffer which already been output.
+
+Sun Sep  4 15:01:35 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y (f_arg): Ripper should not do semantic check.
+	  [ruby-dev:26948]
+
+Sat Sep  3 23:52:35 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_fcall): new method to avoid inefficiency of
+	  obj.instance_eval{send(...)} tricks.
+
+Sat Sep  3 13:59:31 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#descend): Pathname.new("./a/b/c").descend
+	  didn't yield "."
+	  (Pathname#ascend): ditto.
+
+Fri Sep  2 23:51:54 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_arg): f_norm_arg is a VALUE in ripper, not an ID.
+	  fixed: [ruby-dev:26942]
+
+	* lib: do not use __send__ to access private methods.  [ruby-dev:26935]
+
+Thu Sep  1 17:11:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): wrong condition for $SAFE restoration.
+
+Thu Sep  1 14:12:45 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: On Tcl8.5, MultiTkIp#invoke_hidden doesn't
+	  work (gives wrong order of arguments).
+
+	* ext/tk/lib/multi-tk.rb: add MultiTkIp#invoke_hidden_on_namespace
+	  to support '-namespace' option of 'interp invokehidden' command
+	  on Tcl8.5.
+
+Wed Aug 31 14:41:30 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (OPTFLAGS): default global optimization to
+	  disabled for all VC++ versions.  fixed: [ruby-dev:26897]
+
+Wed Aug 31 10:36:09 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* process.c (proc_detach, proc_setmaxgroups): missing argument type
+	  declaration. (I recommend ANSI-style function)
+
+Wed Aug 31 06:59:01 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_scan): already String#scan behaves differently
+	  regarding if block is given.
+
+Tue Aug 30 23:49:34 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c, dir.c, enum.c, hash.c, io.c, range.c, string.c, struct.c:
+	  let enumerable methods return Enumerator.  [ruby-dev:26924]
+
+	* intern.h (RETURN_ENUMERATOR): utility macro for enumerable methods.
+
+Tue Aug 30 23:25:45 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/debug.rb: no need to restart at exit.
+
+Tue Aug 30 23:20:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_rescue2): initialization miss.  fixed: [ruby-dev:26917]
+
+	* lib/mkmf.rb (xsystem, xpopen): no longer expand by Config.
+
+	* lib/mkmf.rb (link_command, cc_command, cpp_command): expand
+	  variables at once, and quote hdrdir.  fixed: [ruby-core:05680]
+
+	* lib/mkmf.rb (libpathflag): quote paths.
+
+Tue Aug 30 19:34:27 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/digest/md5/md5ossl.h, ext/digest/rmd160/rmd160ossl.h,
+	  ext/digest/sha1/sha1ossl.h: include <stddef.h> to avoid
+	  error in compilation with OpenSSL-0.9.8. [ruby-list:41068]
+
+Tue Aug 30 16:19:40 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/init.rb: bug fix. [ruby-dev: 26920]
+
+Tue Aug 30 16:13:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_count): new method.  [ruby-dev:26895]
+
+Tue Aug 30 12:45:15 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_send): do not call private methods if the receiver
+	  is specified.  [ruby-talk:153672]
+
+Mon Aug 29 19:47:18 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/rdoc/usage.rb: improper exceptions. [ruby-dev:26870]
+
+	* lib/rdoc/usage.rb: support the case when non-ruby code exists before
+	  shebang. (this is needed when ri.bat is executed on windows)
+
+Mon Aug 29 18:58:05 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/init.rb: make IRB -I option that is same behavior for ruby.
+	  [ruby-dev:26872]
+
+	* lib/irb/locale.rb: support to print help message when OS locale is
+	  ja_JP.utf-8. [ruby-dev:26872]
+
+Mon Aug 29 01:43:05 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#descend): new method.
+	  (Pathname#ascend): ditto.
+
+Mon Aug 29 00:35:09 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/time.rb: require 'date/format' instead of 'parsedate'.
+	  (Time.parse): extract fractional seconds using Date._parse.
+	  (Time.strptime): extract fractional seconds using Date._strptime.
+	  [ruby-talk:153859]
+
+Sat Aug 27 20:13:31 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/curses/curses.c ({curses,window}_clrtoeol): added. suggested
+	  by Reyn Vlietstra.
+
+	* ext/curses/curses.c: chtype in curses is not `char', rather `long'.
+	  [ruby-Bugs:2298]
+
+	* ext/curses/view.rb: String =~ String is deprecated.
+
+Thu Aug 25 15:48:58 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/win32ole/win32ole.c: suppress warnings. (win32)
+
+Wed Aug 24 11:01:26 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/logger/test_logger.rb (test_shifting_size): should close log
+	  device before unlink, since some platform cannot unlink opened
+	  file.
+
+Tue Aug 23 06:07:02 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/digest.rb: added SHA224, SHA256, SHA384 and SHA512.
+	  these features are enabled if this library is compiled with
+	  OpenSSL 0.9.8 or later.
+
+	* test/openssl/test_digest.rb: add test for new digests.
+
+Tue Aug 23 05:47:04 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ns_spki.c (ossl_spki_initialize): try to decode
+	  the argument as a string.
+
+	* ext/openssl/ossl_ns_pki.c (ossl_spki_to_der): new method.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_initialize): should
+	  set @time to avoid warning.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_set_default_paths,
+	  X509_STORE_add_cert, X509_STORE_add_crl): should raise error if
+	  wrapped functions fails.
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_set_ciphers): fix error message.
+
+	* ext/openssl/ossl_x509req.c (ossl_x509req_set_attributes): get rid
+	  of unused variable.
+
+	* test/openssl/test_ns_spki.rb: add new file.
+
+	* test/openssl/test_x509store.rb: add test for error.
+
+Tue Aug 23 01:11:40 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* sprintf.c (ruby__sfvwrite): should move `buf' to the end of
+	  `result'. [ruby-dev:26859]
+
+Mon Aug 22 23:51:19 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: ONIG_OPTION_CAPTURE_GROUP conflicts with
+	  RE_OPTION_ONCE.  [ruby-dev:26852]
+
+Mon Aug 22 20:11:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/vsnprintf.c (BSD__sprint): needs to call vwrite function
+	  pointer.  fixed: [ruby-dev:26854]
+
+Sat Aug 20 23:55:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): update paren_nest for brackets [].
+
+Sun Aug 21 00:10:23 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/wsdl/xmlSchema/importer.rb (WSDL::XMLSchema::Importer#fetch): add
+	  a workaround for importing an WSDL whose path begins with drive
+	  letter.  [ruby-dev:26242]
+
+Sat Aug 20 22:05:25 2005  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* regexec.c (code_is_in_cclass_node): check code size.
+	  [ruby-dev:26840]
+
+Sat Aug 20 22:37:13 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb (write, shift_log?, shift_log): file shifting race
+	  condition bug fixed.  [ruby-dev:26764]
+
+	* test/logger/test_logger.rb: tests.
+
+Fri Aug 19 18:13:39 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/time.rb (Time.apply_offset): fix a problem with last day of
+	  month.  reported by Lucas Nussbaum.  [ruby-talk:152866]
+
+Thu Aug 18 11:05:36 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (socketpair_internal): need to call open_ifs_socket()
+	  to create sockets instead of winsock's socket().
+	  fixed: [yarv-dev:581]
+
+Wed Aug 17 23:58:05 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (terminate_process): take String message.
+
+	* eval.c (rb_thread_switch): propagate the exception caused thread
+	  termination directly.  fixed: [ruby-core:05552]
+
+Wed Aug 17 21:20:05 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb: ensure that symbol_to_option is private_class_method
+	  and all other methods are module_function
+	  fixed: [ruby-dev:26808]
+
+Wed Aug 17 00:05:46 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_add_method): preserve safe level in the environment
+	  where a method is defined .
+
+	* eval.c (rb_call0): restore preserved safe level in the method
+	  execution.
+
+	* parse.y (lambda): need separate block variable stack
+	  manipulation and lpar_beg maintenance.  based on a patch found
+	  in [ruby-core:05551] from Mauricio Fernandez <mfp at acm.org>.
+
+	* parse.y (parser_yylex): adjust lpar_beg after tLAMBEG and
+	  kDO_LAMBDA. [ruby-core:05551]
+
+Mon Aug 15 07:24:38 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h (rb_check_to_integer): add declaration.
+
+	* object.c (rb_to_integer, rb_check_to_integer): argument constified.
+
+Mon Aug 15 00:38:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_rescue2): reduce PUSH_TAG() as well as NODE_RESCUE.
+	  [ruby-dev:26800]
+
+	* range.c (range_check, range_init): reduce useless exceptions.
+
+Mon Aug 15 00:34:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (yycompile): remove unreachable code.  [yarv-dev:570]
+
+Sat Aug 13 22:16:12 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (remove_entry_secure): forgot final chdir.
+
+Sat Aug 13 22:07:49 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (remove_entry_secure): uses chdir(2) and check
+	  if current directory is correct. [ruby-dev:26100] [ruby-dev:26226]
+
+Sat Aug 13 21:11:05 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_VARIANT class.
+
+	* ext/win32ole/tests/testall.rb: ditto.
+
+	* ext/win32ole/tests/testOLEVARIANT.rb: ditto.
+
+Sat Aug 13 18:51:26 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_block_pass): distinguish current block from others.
+	  fixed: [ruby-dev:26274]
+
+	* ext/stringio/stringio.c (strio_set_string): disallow nil.
+	  http://www.rubyist.net/~nobu/t/20050811.html#c05
+
+Sat Aug 13 08:01:59 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb: Kconv.kconv is now alias of Kconv.conv
+	* ext/nkf/lib/kconv.rb: remove nkf dependent symbols from SYMBOL_TO_OPTION
+
+Fri Aug 12 17:06:53 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_larglist): allow optional arguments even when
+	  parentheses are omitted.  based on Nobu's patch from
+	  http://www.rubyist.net/~nobu/t/20050805.html
+
+	* parse.y (parser_yylex): update & maintain lpar_beg for detect
+	  lambda parameters.
+
+Thu Aug 11 23:29:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c: keep holding string after closed.
+
+Thu Aug 11 20:48:40 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
+
+	* numeric.c (fix_equal, fix_cmp, fix_gt, fix_ge, fix_lt, fix_le):
+	  reduce coercing when a method knows about a operand type.
+	  [ruby-dev:26789]
+
+Thu Aug 11 13:01:48 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss: fixed sort bug. [ruby-list:41018]
+
+	* lib/rss/1.0.rb (RSS::RDF::Channel#setup_maker_attributes):
+	  removed self.
+
+	* lib/rss/maker/base.rb (RSS::Maker::ItemsBase#<=>): use #date
+	  instead of @date.
+	  (RSS::Maker::Base::self.def_array_element): added #size.
+
+	* lib/rss/maker/1.0.rb
+	  (RSS::Maker::RSS10::Channel#to_rss,
+	   RSS::Maker::RSS10::Items::Item#to_rss): cleared dc_dates set
+	  upped by using #date.
+
+	* lib/rss/maker/dublincore.rb
+	  (RSS::Maker::ChannelBase, RSS::Maker::ItemsBase::ItemBase):
+	  fixed opposite alias.
+
+	* test/rss/test_setup_maker_1.0.rb
+	  (RSS::TestSetupMaker10::test_setup_maker_items_sort): added some
+	  tests for RSS::Maker::ItemsBase#do_sort.
+
+Wed Aug 10 12:01:20 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb: simplifies Delegator classes; SimpleDelegator
+	  now uses method_missing for all methods.
+
+Wed Aug 10 10:38:50 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_mul0): multiply two numbers (x, y) without
+	  normalizing the result.  x should be a big number.
+	  [ruby-dev:26778]
+
+	* bignum.c (rb_big_pow): use rb_big_mul0() instead of
+	  rb_big_mul().
+
+	* array.c (rb_ary_or, rb_ary_and, rb_ary_plus, rb_ary_diff):
+	  revert the change on 2005-08-03.  Set operation on other item
+	  should have in separate methods.
+
+	* parse.y (shadowing_lvar_gen): warn when arguments shadows
+	  external local variables.
+
+	* parse.y (f_opt): optional arguments should not clobber external
+	  local variables.
+
+	* parse.y (f_rest_arg): rest arguments should not clobber external
+	  local variables.
+
+Wed Aug 10 10:29:40 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fix bug on handling __ruby2val_optkeys().
+
+	* ext/tk/lib/tk/itemconfig.rb: fix bug on handling
+	  __item_ruby2val_optkeys().
+
+	* ext/tk/lib/tk/canvas.rb: didn't check __item_ruby2val_optkeys().
+
+	* ext/tk/lib/tkextlib/blt/component.rb: ditto.
+
+Tue Aug  9 21:53:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (formal_assign): let default values override
+	  arguments to zsuper.  fixed: [ruby-dev:26743]
+
+Tue Aug  9 20:30:19 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
+
+	* bignum.c (rb_big_coerce): allow bignum x bignum coercing.
+	  [ruby-dev:26778]
+
+Tue Aug  9 15:12:04 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: remove dangerous 'rb_jump_tag's.
+
+	* ext/tk/lib/tk.rb: add __val2ruby_optkeys and __ruby2val_optkeys to
+	  help to convert option values between ruby and tcl.
+
+	* ext/tk/lib/tk/itemconfig.rb: add __item_val2ruby_optkeys and
+	  __item_ruby2val_optkeys to help to convert option values between
+	    ruby and tcl.
+
+	* ext/tk/lib/tk/radiobutton.rb: use __ruby2val_optkeys for 'variable'
+	  option (for the reason of backward compatibility).
+
+	* ext/tk/lib/tk/composite.rb: clarify the arguments of super().
+
+	* ext/tk/lib/tk/spinbox.rb: ditto.
+
+	* ext/tk/lib/tk/text.rb: ditto.
+
+	* ext/tk/lib/tk/validation.rb: ditto.
+
+	* ext/tk/lib/tkextlib/*: support to treat tkvariable-type
+	  configure options.
+
+Tue Aug  9 08:24:05 2005  Mauricio Fernandez  <mfp@acm.org>
+
+	* parse.y (f_block_arg), eval.c (rb_yield_0): deal with dynamic
+	  variable lambda arguments.  [ruby-core:05540]
+
+Mon Aug  8 22:13:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (assign): deal with new block argument.
+	  fixed: [ruby-core:05536]
+
+	* eval.c (rb_node_arity): follow change of NODE_ARGS.
+	  fixed: [ruby-dev:26761]
+
+Mon Aug  8 21:28:13 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_fnmatch.rb: separated from test_file.rb.
+
+Mon Aug  8 20:40:35 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_method.rb: added.  [ruby-dev:26761]
+
+Mon Aug  8 01:26:37 2005  Mauricio Fernandez  <mfp@acm.org>
+
+	* parse.y (f_larglist): mistake in syntax rule.  [ruby-core:05535]
+
+Mon Aug  8 05:16:55 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.c (ossl_raise): should use ERR_peek_last_error
+	  to get last error on the current thread. And should report
+	  if errors are on the stack while OpenSSL.debug is true.
+
+	* ext/openssl/ossl.c (ossl_get_errors): new method for debugging
+	  this library.
+
+Mon Aug  8 05:15:19 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpproxy.rb (HTTPProxyServer#initialize),
+	  lib/webrick/httpserver.rb (HTTPServer#initialize),
+	  lib/webrick/httpservlet/cgihandler.rb (CGIHandler#initialize),
+	  lib/webrick/httpservlet/erbhandler.rb (ERBHandler#initialize),
+	  lib/webrick/httpservlet/filehandler.rb(DefaultFileHandler#initialize):
+	  super (called with no arguments) takes default value of optional
+	  arguments. [ruby-dev:26743]
+
+	* lib/webrick/httputils.rb: add a media-type "text/html" for .xhtml.
+
+Sun Aug  7 23:52:39 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_super.rb: added optional arg tests.  [ruby-dev:26743]
+	  the tests expects 1.8 behavior at this time.
+
+Sat Aug  6 12:35:24 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/{tk.rb,tk/itemconfig.rb}: configure creates
+	  TkVariable if key name is 'variable' or 'textvariable'
+	  by default. [ruby-dev:26749]
+
+	* ext/tk/lib/tk/{label,radiobutton}.rb: removed its own
+	  {variable,textvariable} function.
+
+	* ext/tk/lib/tk/variable.rb: retains backward compatibility.
+
+Fri Aug  5 12:48:31 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c: fixed memory leak when tk_funcall raised
+	  exception. (copies argv into heap in tk_funcall instead of
+	  caller)
+
+Fri Aug  5 12:36:40 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): need to convert path separator
+	  before invoking install command.
+
+Fri Aug  5 08:08:05 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (return_jump): fix "can't across thread" error message
+	  when no thread associated.
+	  http://www.namikilab.tuat.ac.jp/~sasada/diary/200507.html#d31
+
+Fri Aug  5 00:25:12 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c: refactoring - extract ruby string <->
+	  tcl object conversion as get_str_from_obj and get_obj_from_str.
+
+Fri Aug  5 00:19:33 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (Init_Enumerator): provided features should have
+	  extensions.
+
+	* eval.c (rb_feature_p): returns type of the feature instead of
+	  extension.
+
+	* eval.c (search_required): ruby library should be prior to statically
+	  linked extensions.  fixed: [ruby-dev:26711]
+
+	* eval.c (formal_assign): returns position of rest arguments variable.
+
+	* parse.y (f_rest_arg): use anonymous variable for rest arguments.
+	  fixed: [ruby-dev:26647]
+
+	* extmk.rb (extmake): needs to be wrapped in an Array.
+
+Thu Aug  4 20:03:18 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
+
+	* numeric.c (Init_Numeric): do not share implementation among
+	  Fixnum#/ and Fixnum#div.  [ruby-core:05531]
+
+Thu Aug  4 18:38:36 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: cannot compile for Tcl7.6/Tk4.2.
+
+	* ext/tk/tcltklib.c: add nativethread consistency check.
+
+	* ext/tk/stubs.c: ditto.
+
+	* ext/tk/lib/tk.rb: forgot to define TclTkIp.encoding and encoding=
+	  when Tcl is 7.6 or 8.0.
+
+	* ext/tk/lib/tk/wm.rb: support to make some methods as options of
+	  root or toplevel widget. [ruby-talk:150336]
+
+	* ext/tk/lib/tk/root.rb: ditto.
+
+	* ext/tk/lib/tk/toplevel.rb: ditto.
+
+	* ext/tk/lib/tkextlib/SUPPORT_STATUS: update RELEASE_DATE
+
+Thu Aug  4 13:30:15 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_div): should not convert the result into
+	  integer.  [ruby-core:05524]
+
+Thu Aug  4 08:03:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): should not modify $mflags for each
+	  extensions.
+
+Thu Aug  4 00:25:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, Makefile.in, {bcc32,win32,wince}/Makefile.sub: integrated
+	  macro definitions.
+
+	* bcc32/Makefile.sub: LIBRUBY_SO should use DLDOBJS, not EXTOBJS.
+
+	* {win32,wince}/Makefile.sub: separate config.h for compiler versions.
+
+Thu Aug  4 00:24:59 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c: replacing is no longer needed.
+
+Wed Aug  3 21:59:16 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb: TkVariable#trace didn't work on
+	  TkVariable retrieved from TkVariable.new_hash.ref. [ruby-dev:26721]
+
+Wed Aug  3 12:40:28 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
+
+	* numeric.c (fix_plus): reduce coercing when a method knows about
+	  a operand type.  [ruby-dev:26723]
+
+	* numeric.c (fix_minus, fix_mul, fix_quo, fix_div, fix_mod,
+	  fix_divmod, fix_pow): ditto.
+
+	* bignum.c (rb_big_div, rb_big_modulo): export to reduce
+	  coercing.
+
+Wed Aug  3 10:13:52 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in, {bcc32,win32,wince}/Makefile.sub (HAVE_SNPRINTF,
+	  HAVE_VSNPRINTF): use win32/win32.c's implementation instead of
+	  missing/vsnprintf.c's.
+
+	* win32/win32.[ch] (rb_w32_snprintf, rb_w32_vsnprintf): reverted.
+
+Wed Aug  3 10:05:08 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check vsnprintf() and snprintf().
+
+	* sprintf.c, missing/vsnprintf.c: made vsnprintf() and snprintf()
+	  private.  fixed: [ruby-dev:26651]
+
+Wed Aug  3 08:22:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_connect): revert [ruby-talk:111654]
+	  changes at 2004-09-07.  [ruby-dev:26656]
+
+Wed Aug  3 06:53:35 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_or): wraps the operand in an array if it is not
+	  an array.  [ruby-talk:150495] [EXPERIMENTAL]
+
+	* array.c (rb_ary_and, rb_ary_plus, rb_ary_diff): ditto.
+
+Tue Aug  2 10:23:12 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c: use Tcl_[GS]etVar2Ex instead of
+	  Tcl_Obj[GS]etVar2. (avoid Tcl_NewStringObj on supported platforms)
+
+	* ext/tk/tcltklib.c: use ip_{get,set,unset}_variable2_core from
+	  ip_{get,set,unset}_variable.
+
+	* ext/tk/tcltklib.c: replaced Tcl_Panic with rb_bug.
+
+Tue Aug  2 01:40:38 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/ping.rb (Ping.pingecho): should rescue StandardError.
+	  [ruby-dev:26677]
+
+Mon Aug  1 19:02:23 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c: refactoring - replaced rb_ivar_defined &
+	  rb_ivar_get with single rb_attr_get call.
+
+Mon Aug  1 18:44:08 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c (Tcl_GetStringResult): refactoring - define
+	  alternative macro on Tcl7.x or earlier.
+
+Mon Aug  1 13:53:55 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c (deleted_ip): refactoring - interpreter
+	  deletion check. [ruby-dev:26664]
+
+Mon Aug  1 01:08:21 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (check_insecure_method): use private_methods and
+	  protected_methods instead of respond_to? to check method visibility.
+	  [ruby-dev:26616]
+
+	* test/drb/drbtest.rb: ditto.
+
+	* test/drb/ut_drb.rb: ditto.
+
+Sat Jul 30 18:49:44 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_TYPE#ole_typelib,
+	  WIN32OLE_TYPE#implemented_ole_types.
+
+	* ext/win32ole/tests/testOLETYPE.rb: ditto.
+
+Fri Jul 29 16:12:02 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/context.rb: fix `irb --readline` option. [ruby-dev:40955]
+
+Fri Jul 29 09:59:38 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_call0): fix calling zsuper from a method with anonymous
+	  rest argument.  [ruby-dev:26639]
+
+	* eval.c (rb_yield_0): push yielded node instead of yielding.
+	  fixed: [yarv-dev:549]
+
+Thu Jul 28 21:49:17 2005  IWATSUKI Hiroyuki  <don@na.rim.or.jp>
+
+	* parse.y (rb_parser_end_seen_p): exclude from ripper.
+	  <http://moonrock.jp/~don/d/200507.html#d28_t2>
+
+	* sprintf.c (clearerr): remove standard macro before re-definition.
+	  <http://moonrock.jp/~don/d/200507.html#d28_t3>
+
+Thu Jul 28 18:09:55 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/stubs.c: When --enable-tcltk-stubs, the initialize
+	  routine creates a Tcl/Tk interpreter and deletes it. However,
+	  init cost of Tk's MainWindow is not so small. And that makes it
+	  impossible to use libraries written with Tcl functions only on
+	  an environment without a graphical display. This changes support
+	  delaying initialization of Tk_Stubs until the script needs Tk.
+
+	* ext/tk/stubs.h: New file. Define prototypes and return codes of
+	  functions on stubs.c.
+
+	* ext/tk/tcltklib.c: Support delaying initialization of Tk_Stubs
+	  until the script needs Tk.
+
+	* ext/tk/tcltklib.c: Show friendly error messages for errors on
+	  initialization.
+
+	* ext/tk/tcltklib.c: Avoid SEGV on ip_finalize() when ruby is
+	  exiting and $DEBUG is true. (Not fix. If you know the reason of
+	  why, please fix it.)
+
+	* ext/tk/tkutil/tkutil.c (ary2list, ary2list2): bug fix on handling
+	  of encoding.
+
+	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string and bg_eval_string
+	  don't work properly.
+
+	* ext/tk/lib/tk.rb: Forget extending Tk::Encoding module to Tk.
+
+	* ext/tk/lib/tk/variable.rb: TkVarAccess fails to initialize the
+	  object for an element of a Tcl's array variable.
+
+Thu Jul 28 17:23:37 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_larglist): allow block argument in lambda parameter
+	  list without parenthesis.
+
+Thu Jul 28 17:14:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (each_i): typo fixed.  [ruby-dev:26622]
+
+Thu Jul 28 15:04:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_arg): better argument name duplication check
+
+	* parse.y (new_args_gen): factored out name duplication check for
+	  optional and rest arguments.
+
+	* parse.y (new_bv_gen): allow shadowing outer local variables;
+	  warning remains.
+
+Thu Jul 28 13:46:06 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ripper_warningS): the argument was omitted.
+	  [ruby-dev:26621]
+
+Thu Jul 28 11:30:57 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_larglist): allow bv_decl at the end of lambda
+	  argument list.  [EXPERIMENTAL]
+
+	* parse.y (new_bv_gen): allow local variable shadowing, with
+	  warning in verbose mode.
+
+Wed Jul 27 23:23:54 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (obj_free): make message format consistent with one from
+	  gc_mark().  [ruby-talk:149668]
+
+	* sprintf.c (quad_t): prepare quad_t as well.  [ruby-talk:149668]
+
+Wed Jul 27 22:11:37 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/tdiary_plugin: removed. because the plugin
+	  is imported in the tDiary plugin packages.
+
+Wed Jul 27 19:11:53 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (cd): follow :noop option change. (This patch
+	  is contributed by Doug Kearns)
+
+Wed Jul 27 16:25:59 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (lambda): Perl6 style -> lambda expression. [NEW]
+	  [VERY EXPERIMENTAL]
+
+Wed Jul 27 10:43:14 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (id2ref): must not assign pointers to long int.  use
+	  LONG_LONG instead if SIZEOF_LONG < SIZEOF_VOIDP.
+	  [ruby-talk:149645]
+
+	* ruby.h: use LONG_LONG to simplify the change.
+	  [ruby-talk:149645]
+
+Wed Jul 27 10:59:02 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_each): rewinddir(3) before iteration.
+	  [ruby-talk:149628]
+
+Wed Jul 27 02:34:58 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_throw): replace all '0x%lx' by '%p'.
+	  [ruby-talk:149553]
+
+	* missing/vsnprintf.c (BSD_vfprintf): '%p' need to handle 64bit
+	  size pointer.  [ruby-talk:149553]
+
+Tue Jul 26 22:41:28 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/lib/ripper/sexp.rb: new method Ripper.sexp_raw.
+
+	* ext/ripper/lib/ripper/sexp.rb (Ripper.sexp): returns more
+	  readable tree.  This is suggested by Kirill A. Shutemov.
+
+Tue Jul 26 22:05:12 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: merge a patch contributed by Daniel Berger,
+	  with some modification. (RubyForge #2128)
+
+Tue Jul 26 18:11:33 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h: support LLP64 model.  [ruby-talk:149524]
+
+Tue Jul 26 12:57:40 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/openssl_missing.c: include <openssl/engine.h> before
+	  <openssl/x509_vfy.h> to avoid compilation error of mswin32.
+	  suggested by NAKAMURA Usaku.
+
+Mon Jul 25 23:48:55 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.[ch]: (rb_w32_vsnprintf, rb_w32_snprintf): removed.
+
+Mon Jul 25 21:30:46 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: Borland MAKE doesn't look for file names which have paths
+	  from VPATH.  fixed: [ruby-dev:26604]
+
+	* ruby.h (NORETURN, DEPRECATED): moved just after config.h.
+
+	* {win32,wince}/Makefile.sub: vsnprintf() is in missing now.
+
+	* {bcc32,win32,wince}/Makefile.sub: moved CPPFLAGS only for ruby
+	  source to XCFLAGS.
+
+Mon Jul 25 14:10:02 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: fix en-bugged part in the last commit.
+
+Mon Jul 25 13:45:18 2005  NAJIMA Hiroki  <najima@mickey.ai.kyutech.ac.jp>
+
+	* io.c: check HAVE_SYS_IOCTL_H before including the header.
+	  [ruby-dev:26610]
+
+Sat Jul 23 16:48:12 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_load): should check
+	  OPENSSL_NO_STATIC_ENGINE.
+
+Sat Jul 23 11:46:30 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (rb_fd_select): the all three fd_sets must be long enough for
+	  select.  fixed: [ruby-talk:149059]
+
+Sat Jul 23 10:01:41 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_vsprintf, rb_sprintf): new functions return new String,
+	  using missing/vsnprintf.c.  [ruby-dev:26580]
+
+	*  missing/vsnprintf.c: made the output changeable.
+
+Fri Jul 22 21:06:08 2005  Tadashi Saito  <shiba@mail2.accsnet.ne.jp>
+
+	* bignum.c (rb_big_eq): reduce isnan().  [ruby-dev:26600]
+
+	* numeric.c (flo_eq, flo_gt, flo_ge, flo_lt, flo_le): ditto.
+
+Fri Jul 22 15:02:39 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: moved copyright description to lib/rss.rb.
+
+	* lib/rss.rb: added for convenience.
+
+	* sample/rss/re_read.rb: added #to_s sample.
+
+	* sample/rss/blend.rb: use 'require "rss"' instead of
+	  'require "rss/*"'.
+	* sample/rss/list_description.rb: ditto.
+	* sample/rss/rss_recent.rb: ditto.
+	* sample/rss/tdiary-plugin/rss-recent.rb: ditto.
+
+	* sample/rss/tdiary-plugin/rss-recent.rb: 0.0.6 -> 0.0.7.
+
+Fri Jul 22 14:37:43 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb (RSS::Parser#initialize): accept HTTP/FTP
+	  URI and local file path too.
+
+	* test/rss/test_parser.rb (RSS::TestParser#test_parse): test
+	  for the above.
+
+Fri Jul 22 07:01:42 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tkutil/tkutil.c (tk_conv_args): forget to revert
+	  thread_critical and gc_disable when raise ArgumentError.
+
+	* ext/tk/lib/remote-tk.rb: RemoteTkIp doesn't need to include TkUtil.
+
+	* ext/tk/tcltklib.c: add TclTkIp#has_mainwindow? method.
+
+	* ext/tk/lib/tk.rb: add Tk.has_mainwindow? method.
+
+	* ext/tk/lib/multi-tk.rb: add MultiTkIp#has_mainwindow? method.
+
+	* ext/tk/lib/remote-tk.rb: add RemoteTkIp#has_mainwindow? method.
+
+	* ext/tk/lib/multi-tk.rb: slave IP fail to exit itself when $SAFE==4.
+
+	* ext/tk/lib/multi-tk.rb: remove constants from MultiTkIp module to
+	  avoid access from external.
+
+	* ext/tk/lib/multi-tk.rb: check_root flag is ignored on slave IPs'
+	  mainloop.
+
+	* ext/tk/lib/multi-tk.rb: hang-up Tk.mainloop called on a slave IP
+	  with $SAFE==4.
+
+	* ext/tk/lib/multi-tk.rb: MultiTkIp#bg_eval_proc doesn't work
+	  properly.
+
+	* ext/tk/lib/multi-tk.rb: add MultiTkIp#set_cb_error(proc) and
+	  cb_error(exc) to log errors at callbacks on safe slave IPs.
+
+	* ext/tk/lib/multi-tk.rb: fail to get an available slave IP object
+	  when call Tk.mainloop in the block which is given to new_* method,
+	    because cannot finish initialize while the root widget is alive.
+
+	* ext/tk/lib/multi-tk.rb: fail to control a slave IP when Tk.mainloop
+	  runs on the IP.
+
+Thu Jul 21 01:00:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/{nkf.c,utf8tbl.c,config.h}:
+	  import 1.76
+	  [ruby-dev:26592] nkf constification
+
+Wed Jul 20 19:18:52 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (S_ISREG): need to define S_ISREG before it is used first.
+
+Wed Jul 20 18:33:15 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* io.c (wsplit_p): patch for the environment where
+	  fcntl(F_GETFL, O_NONBLOCK) is not supported. in that case,
+	  set FMODE_WSPLIT without fcntl check. [ruby-dev:26566]
+
+Wed Jul 20 18:07:11 2005  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (rb_io_ctl): update FMODE_WSPLIT_INITIALIZED and FMODE_WSPLIT
+	  by F_SETFL.
+
+Wed Jul 20 10:04:51 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_class_path): need to adjust snprintf() len for
+	  terminating NUL.  [ruby-dev:26581]
+
+Wed Jul 20 03:58:52 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/socket/socket.c: sorry, BeOS also uses HAVE_CLOSESOCKET,
+	  so reverted.
+
+	* ext/socket/extconf.rb: should not define HAVE_CLOSESOCKET
+	  on windows.
+
+Wed Jul 20 03:12:21 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/socket/socket.c: should not undef close() on win32.
+	  it's defined to rb_w32_close(), otherwise handle leaks.
+	  [ruby-Bugs-2131]
+
+Wed Jul 20 00:48:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (syserr_initialize): don't use str before StringValue()
+	  check.  [ruby-dev:26579]
+
+Tue Jul 19 22:47:29 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (syserr_initialize): add 1 byte for snprintf() size for
+	  NUL at the end.  [ruby-dev:26574]
+
+Tue Jul 19 17:16:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (trap): remove sigexit(); handle "EXIT" via sig_exec().
+	  [ruby-dev:26440]
+
+	* io.c (rb_io_inspect): replace sprintf() with "%s" format all
+	  over the place by snprintf() to avoid integer overflow.
+
+Tue Jul 19 14:10:50 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c: rbtk_eventloop_depth is used as int.
+
+	* ext/tk/tcltklib.c: rbtk_pending_exception is tested with
+	  NIL_P, so should assign Qnil instead of 0 (Qfalse).
+
+	* ext/tk/tcltklib.c (ip_invoke_real): fixed memory leak when
+	  ip is deleted.
+
+Tue Jul 19 13:19:46 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb: For symmetry, add TkVariable#string. It
+	  returns a string even if the default value type of the TkVariable
+	  object is not "string".
+
+Mon Jul 18 21:39:18 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* eval.c (rb_call0): make the pointer to NODE volatile
+	  instead of NODE itself.
+
+Mon Jul 18 14:32:21 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (rb_call0): make body volatile to avoid possible optimization
+	  problem.
+	  [ruby-dev:26195]
+
+Mon Jul 18 12:23:27 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/io/wait/wait.c: wrong backport from trunk, and compile error on
+	  platforms fd_set is not a bit set.  fixed: [ruby-dev:26562]
+
+Mon Jul 18 09:36:25 2005  Tanaka Akira  <akr@m17n.org>
+
+	* rubyio.h (FMODE_WSPLIT, FMODE_WSPLIT_INITIALIZED): new constant.
+
+	* io.c (wsplit_p): new function.
+	  (io_fflush): split writing data by PIPE_BUF if wsplit_p is true in
+	  multi-threaded mode.
+	  (io_fwrite): ditto.
+	  [ruby-dev:26540]
+
+Mon Jul 18 05:00:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: import nkf.c 1.73
+	  fix: TestKconv 1F
+
+Sun Jul 17 13:46:54 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/wait/extconf.rb, ext/io/wait/wait.c: Win32 platforms support.
+
+Sat Jul 16 23:43:16 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (Init_Enumerator): wrong argument specs.
+	  [ruby-core:05481]
+
+Sat Jul 16 15:52:50 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.[hc]: constified socket functions. [ruby-dev:26553]
+
+Fri Jul 15 23:59:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_c.rb (handle_class_module): handle a
+	  module enclosed in a built-in module.  fixed: [ruby-talk:148239]
+
+	* lib/rdoc/parsers/parse_c.rb (find_body): allow macros as methods.
+
+	* lib/rdoc/parsers/parse_c.rb (find_call_seq): allow :nodoc: modifier
+	  in C.  [ruby-core:04572]
+
+Fri Jul 15 23:20:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (Init_Enumerator): use an internal directly.
+
+Fri Jul 15 07:58:10 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#accept_client):
+	  sockets should be non-blocking mode. [ruby-dev:26405]
+
+	* lib/webrick/utils.rb (WEBrick::Utils.set_non_blocking): new method.
+
+Fri Jul 15 00:11:36 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enumeratorize): create new enumerator for current method if
+	  no block is given.
+
+	* enumerator.c: moved from ext/enumerator.
+
+Thu Jul 14 18:27:35 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_strerror): should return correct message
+	  for ENAMETOOLONG and ENOTEMPTY. (bcc32) [ruby-dev:26533]
+
+	* win32/win32.c (rb_w32_strerror): stripped CR LF on the tail.
+	  (bcc32) [ruby-dev:26533]
+
+Thu Jul 14 00:45:42 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* LEGAL (ext/nkf/nkf-utf8): updated from nkf1.7 to nkf-utf8.
+
+Wed Jul 13 22:44:00 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y: remove static variables.  [ruby-dev:26530]
+
+Wed Jul 13 19:36:29 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_mkdir): should set EEXIST (not EACCES)
+	  if file or directory already exists. (bcc32) [ruby-dev:26508]
+
+	* win32/win32.c (rb_w32_rmdir): should set ENOTDIR (not EINVAL)
+	  if it is not directory. (bcc32, win32)
+
+	* win32/win32.c (rb_w32_rmdir, rb_w32_unlink): restore
+	  FILE_ATTRIBUTE_READONLY flag on function failure.
+
+Wed Jul 13 12:40:00 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: TclTkLib.do_one_event doesn't work.
+
+	* ext/tk/lib/tk.rb: Tk.thread_update is available.
+
+Tue Jul 12 23:32:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: keep curdir unexpanded.
+
+Mon Jul 11 23:50:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c, intern.h (rb_proc_call, rb_obj_method, rb_method_call):
+	  export.
+
+	* ext/enumerator/enumerator.c (enumerator_with_index): [EXPERIMENTAL]
+	  added a new method Enumerator#with_index.  [ruby-talk:147728]
+
+Mon Jul 11 08:31:29 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regparse.c (fetch_escaped_value): mask values following \c in
+	  regexp.  fixed: [ruby-dev:26500]
+
+Sun Jul 11 05:18:17 2005  Michael Neumann  <mneumann@ruby-lang.org>
+
+	* lib/xmlrpc/server.rb (XMLRPC::Server): Switch from GServer over to
+	  WEBrick. This makes file lib/xmlrpc/httpserver.rb obsolete (at least it is
+	  no further used by the XML-RPC library).
+
+Mon Jul 11 02:50:23 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#request_line):
+	  mistook to merge the patch of [ruby-dev:26235] at
+	  revision 1.11.
+
+Sun Jul 10 23:58:04 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#unlink): try Dir.unlink first to
+	  avoid unlink a directory by root.
+	  cf. [ruby-dev:26237]
+
+Sun Jul 10 12:47:01 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/debug.rb (debug_command): added a deficient format specifier.
+	  fixed: [ruby-core:05419]
+
+Sat Jul  9 22:02:37 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_method_dispid): convert dispid
+	  in Ruby and C by INT2NUM and NUM2INT.
+
+	* ext/win32ole/win32ole.c (ole_invoke2): ditto.
+
+	* ext/win32ole/test/testWIN32OLE.rb: ditto.
+
+	* ext/win32ole/test/testOLEMETHOD.rb: ditto.
+
+Fri Jul  8 15:45:04 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::VERSION): 0.1.4 -> 0.1.5.
+
+	* test/rss/test_version.rb (RSS::TestVersion#test_version):
+	  ditto.
+
+	* lib/rss/0.9.rb (RSS::Rss::Channel::Item::Category):
+	  domain attribute of <category> is optional. Thanks to
+	  Chris Lee <clee@kde.org>.
+
+	* test/rss/test_parser.rb (RSS::TestParser#test_category20):
+	  adjusted test case.
+
+Wed Jul  6 18:45:53 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_pattern_match): now returns nil.
+	  [ruby-core:05391]
+
+Mon Jul  4 14:35:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sample/svr.rb: service can be stopped by ill-behaved client; use
+	  tsvr.rb instead.
+
+Mon Jul  4 13:25:21 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* missing/erf.c: original erf.c by prof. Okumura is confirmed to
+	  be public domain.  reverted BSD implementation.
+
+Wed Jul  6 11:15:21 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (open_ifs_socket): new function.
+
+	* win32/win32.c (StartSockets, rb_w32_socket): use open_ifs_socket()
+	  instead of socket().
+	  all changes are derived from [ruby-core:5388].
+
+Wed Jul  6 00:15:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/{nkf.c,utf8tbl.c,config.h}:
+	  imported nkf.c 1.70 (support UTF-8-MAC)
+
+	* ext/nkf/lib/kconv.rb: add :utf8mac and :internalunicode
+
+Tue Jul  5 23:44:06 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb: expand source library path.
+
+Tue Jul  5 23:27:14 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (sort_2): get rid of yet another bcc's bug.
+	  fixed: [ruby-core:05152]
+
+	* eval.c (rb_thread_save_context): must not switch contexts during
+	  re-allocating stack.  fixed: [ruby-core:05219]
+
+Tue Jul  5 15:15:10 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tkutil.c: fix typo.
+
+Tue Jul  5 14:52:56 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: bug fix on treating Unicode strings.
+
+	* ext/tk/tcltklib.c: add methods to treat encoding mode.
+
+	* ext/tk/MANUAL_tcltklib.eng: add description of TclTkLib#encoding,
+	  encoding_system, and so on.
+
+	* ext/tk/MANUAL_tcltklib.eucj: ditto.
+
+	* ext/tk/tkutil/tkutil.c: fail to create a Tcl's list string from
+	  an array including multiple kind of encoded strings.
+
+	* ext/tk/lib/tk.rb: ditto.
+
+	* ext/tk/lib/multi-tk.rb: 2nd arg of _{to|from}UTF8 is omissible.
+
+	* ext/tk/lib/remote-tk.rb: ditto.
+
+	* ext/tk/lib/tk.rb: override TclTkLib#encoding and encoding= to
+	  use TkCore::INTERP.encoding and encoding=.
+
+	* ext/tk/lib/tk.rb: when "require 'tk'" and $KCODE=='NONE', check
+	  DEFAULT_TK_ENCODING to decide Ruby/Tk's system encoding mode.
+
+	* ext/tk/lib/tk/encodedstr.rb: check both of Tk.encoding and
+	  Tk.encoding_system. Tk.encoding has higher priority.
+
+	* ext/tk/lib/tk/optiondb.rb: ditto.
+
+	* ext/tk/lib/tk/spinbox.rb: ditto.
+
+	* ext/tk/lib/tk/validation.rb: ditto.
+
+	* ext/tk/lib/tk/namespace.rb: arguemnts for TclTkIp#_merge_tklist
+	  should be UTF-8 strings.
+
+Mon Jul  4 19:29:32 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/set.rb: test change to follow revision 1.28. (duck typing?)
+
+Mon Jul  4 11:23:50 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/{dbm,gdbm,sdbm}/test_{dbm,gdbm,sdbm}.rb: skip some tests
+	  which using fork on fork-less platforms.
+
+Sun Jul  3 23:26:30 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/wsdl/document/test_rpc.rb: compare formatted time string of
+	  Time objects instead of comparing Time objects itself to avoid
+	  unintended conflict of usec part.  [ruby-dev:26220]
+
+Sat Jul  2 22:41:04 2005  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/socket.c (unix_send_io, unix_recv_io): support x86-64 and
+	  IA64.
+
+Sat Jul  2 17:06:23 2005  Tanaka Akira  <akr@m17n.org>
+
+	* defines.h (FLUSH_REGISTER_WINDOWS): defined for IA64.
+	  (flush_register_windows): declare flush_register_windows.
+
+	* eval.c (flush_register_windows): new function.
+
+	* ruby.h (NOINLINE): move up to be effective in defines.h.
+
+Sat Jul  2 15:19:41 2005  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in: check select_large_fdset.
+
+	* eval.c: use select_large_fdset to support large file descriptors
+	  on Solaris.  [ruby-dev:26404]
+
+Fri Jul  1 17:55:08 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_neg): may be accessing bogus pointer value.
+
+Fri Jul  1 15:50:12 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* missing/erf.c: need to include some headers for some platforms.
+
+	* win32/win32.h (copysign, scalb): define for compatibility with
+	  other platforms. [ruby-dev:26430]
+
+Fri Jul  1 15:37:42 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* missing/crypt.c: modified to make it compilable on platforms
+	  other than BSD.  [ruby-dev:26430]
+
+	* missing/erf.c: ditto.  code from <exp.c> merged.
+
+Fri Jul  1 12:44:56 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI.open_http): refine post_connection_check
+	  call.
+
+Fri Jul  1 11:34:08 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* missing/crypt.c: replaced with 4.4BSD version.
+
+	* missing/erf.c: ditto.
+
+	* missing/vsnprintf.c: removed the third provision from the old
+	  BSD license.  [ruby-core:05177]
+
+Fri Jul  1 01:45:21 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_min, enum_max): must not return Qundef.
+	  fixed: [ruby-core:05299]
+
+Fri Jul  1 00:18:40 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator::respond_to): respond_to? must check
+	  destination object.  [ruby-talk:146894]
+
+Thu Jun 30 23:52:12 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (trap): non-string trap hander was ignored.
+	  fixed: [ruby-dev:26417]
+
+Thu Jun 30 19:00:21 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (RubyLex::identify_number): alternative implements
+	  for [ruby-dev:26410]. And support a numeric form of 0d99999.
+
+Thu Jun 30 17:28:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (RubyLex::identify_number): should not treat
+	  plain zero as an octal number.  [ruby-dev:26410]
+
+Thu Jun 30 15:13:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): pre-evaluate argument for unambiguous
+	  evaluation order.  [ruby-dev:26383]
+
+Thu Jun 30 14:48:23 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/net/http.rb (Net::HTTP#connect, Net::HTTP#request): should
+	  not send proxy username and password to origin servers.
+	  [ruby-dev:25673]
+
+	* lib/net/http.rb (Net::HTTP::ProxyDelta#edit_path): should not
+	  send HTTPS scheme URL to origine servers. [ruby-dev:25689]
+
+Thu Jun 30 09:53:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator::method_missing): forward unknown
+	  method to the destination.  suggested by
+	  <christophe.poucet@gmail.com>.  [ruby-talk:146776]
+
+Wed Jun 29 00:03:20 2005  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* regparse.c (fetch_token): avoid warning of unused goto tag.
+	  [ruby-dev:26389]
+
+Tue Jun 28 21:59:29 2005  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* dir.c, eval.c, parse.y, process.c, ruby.c: avoid warning "unused
+	  variable" [ruby-dev:26387]
+
+	* dir.c (glob_helper): avoid warning "enumeration value `RECURSIVE'
+	  not handled in switch" [ruby-dev:26392]
+
+Tue Jun 28 01:52:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb: add Kconv::VERSION
+	* ext/nkf/lib/kconv.rb (conv): can process arrayed options
+	* ext/nkf/nkf-utf8/nkf.c: imported Revision 1.69
+	* ext/nkf/nkf-utf8/utf8tbl.c: imported Revision 1.9
+
+Sat Jun 25 23:30:51 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (detach_process_watcher): terminate process watcher
+	  thread right after rb_waitpid() succeed.  [ruby-talk:146430]
+
+Sat Jun 25 17:12:20 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.parse_query): should
+	  discard if key=val pair is empty. patch from Gary Wright.
+
+Sat Jun 25 15:49:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_min, enum_max, enum_min_by, enum_max_by): do not ignore
+	  nil as the first element.
+
+Sat Jun 25 15:13:54 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/set.rb (Set#==): [ruby-dev:25206] (ported from ruby_1_8 branch)
+
+Sat Jun 25 11:37:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb: remove constants
+	  Iconv_Shift_JIS, Uconv_EUC_JP, Iconv_UTF8
+	* ext/nkf/lib/kconv.rb: add module functions to Kconv
+	  conv, {eucjp, shiftjis, utf8}?, guess_as_symbol
+	* ext/nkf/lib/kconv.rb: add instance methods to String
+	  conv, {eucjp, shiftjis, utf8}?
+	* ext/nkf/lib/kconv.rb: add aliases Kconv.to_* and String#to_*
+
+Fri Jun 24 17:00:00 2005  Shigeo Kobayashi  <shigeo@tinyforest.jp>
+
+	* ext/bigdecimal/bigdecimal.c: patch from "NATORI Shin"
+	  (u-tokyo.ac.jp) applied to fix rounding bug.
+
+Fri Jun 24 13:17:45 2005  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/common.rb, lib/uri/generic.rb: fixed typo in documents and
+	  replaced some existent domain name with "example.com".
+
+Fri Jun 24 12:23:19 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fix typo on Tk.grid_propagate.
+
+	* ext/tk/lib/tk.rb: Tk.event_generate and TkWindow#event_generate
+	  accept TkEvent::Event object as context argument.
+
+	* ext/tk/lib/tk/event.rb: add TkEvent::Event#valid_fields and
+	  valid_for_generate to get field parameters of event_generate.
+
+Thu Jun 23 23:55:59 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* runruby.rb: should load built rbconfig.rb.
+
+Thu Jun 23 16:53:15 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/canvastag.rb: TkcGroup.new cannot include given items.
+	  TkcGroup#exclude calls wrong method.
+	  Add alias TkcGroup#add [ruby-talk:146049].
+
+	* ext/tk/lib/tk/canvas.rb: TkCanvas#dtag and some subcommands of
+	  TkCanvas#addtag fail to treat a TkcTag argument.
+
+	* ext/tk/lib/tk/event.rb: add TkEvent::Event#generate to help to send
+	  current event to other widgets.
+
+Mon Jun 20 18:44:04 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (FUNCTION_CALL_MAY_RETURN_TWICE): DUMMY_SETJMP is replaced
+	  because setjmp is not enough to fix getcontext and SPARC register
+	  window problem.
+
+Mon Jun 20 17:15:51 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dbm/dbm.c (fdbm_closed): new method DBM#closed?
+
+	* ext/gdbm/gdbm.c (fgdbm_closed): new method GDBM#closed?
+
+	* ext/sdbm/init.c (fsdbm_closed): new method SDBM#closed?
+
+	* test/dbm/test_dbm.rb, test/gdbm/test_gdbm.rb, test/sdbm/test_sdbm.rb
+	  (teardown): close all db objects before deleting data files.
+
+	* win32/win32.{ch} (unlink): hook runtime function to change
+	  file attribute before unlinking.
+	  merge from 1.8, see [ruby-dev:26360]
+
+Mon Jun 20 02:15:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (define_final): document fix: finalizers never get called
+	  before target object is destroyed.
+
+Mon Jun 20 01:26:49 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/openssl_missing.c, ext/openssl/ossl.h,
+	  ext/openssl/ossl_asn1.c, ext/openssl/ossl_bio.c,
+	  ext/openssl/ossl_pkcs12.h, ext/openssl/ossl_x509req.c: avoid
+	  compiler warnings. suggested by Michal Rokos.
+
+Sun Jun 20 00:22:02 2005  Michael Neumann  <mneumann@ruby-lang.org>
+
+	* lib/xmlrpc/utils.rb: Patch by Nobuhiro IMAI fixes the following
+	  problem: Default value modification on
+	  Module#public_instance_methods (false -> true) breaks
+	  s.add_handler(XMLRPC::iPIMethods("sample"), MyHandler.new) style
+	  security protection.
+
+	* lib/xmlrpc/client.rb: Aliased XMLRPC::Client#new2 as
+	  XMLRPC::Client#new_from_uri, and #new3 as #new_from_hash.
+
+Sun Jun 19 14:09:07 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (run_final): reduce unnecessary object allocation during
+	  finalization.
+
+	* gc.c (rb_gc_call_finalizer_at_exit): deferred finalizers list should
+	  be cleared before calling them.  fixed: [ruby-talk:145790]
+
+Sat Jun 18 01:15:36 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_readline): do not set
+	  rl_{in,out}stream.
+
+	* ext/readline/readline.c (readline_s_set_input): new method.
+
+	* ext/readline/readline.c (readline_s_set_output): new method.
+
+	* lib/irb/input-method.rb: set Readline.input and Readline.output.
+
+Fri Jun 17 13:01:40 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/time.rb (Time.parse): fix previous leap seconds support.
+	  (Time.rfc2822): ditto.
+	  (Time.xmlschema): ditto.
+
+Thu Jun 16 15:41:32 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (load_file): '!' is already read. reported by gotoyuzo.
+
+Thu Jun 16 15:09:38 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c (ip_rb_threadVwaitCommand): Tcl_Release
+	  was missing.
+
+Thu Jun 16 13:34:48 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: add Tk.getMultiple{Open|Save}File() which return
+	  an Array of selected files.
+
+Thu Jun 16 12:53:24 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/time.rb (Time.parse): "Fri Jan  1 08:59:60 +0900 1999" was
+	  parsed as "Fri Jan 01 09:00:00 JST 1999" even on an environment
+	  which supports leap seconds.
+	  (Time.rfc2822): ditto.
+	  (Time.xmlschema): ditto.
+
+Thu Jun 16 00:13:41 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb (Resolv::DNS::Resource#ttl): new attribute.
+	  (Resolv::DNS::Resource#==): ignore @ttl.
+	  (Resolv::DNS::Resource#hash): ditto.
+	  (Resolv::DNS::Message::MessageDecoder#get_rr): save TTL in a
+	  Resource object.
+	  based on [ruby-core:5190] by Eric Hodel.
+
+Wed Jun 15 18:26:39 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: support "tk inactive" sub-command [for Tcl/Tk8.5a3]
+
+	* ext/tk/lib/tk/namespace.rb: support "namespace path" sub-command and
+	  'namespace ensemble' sub-command [for Tcl/Tk8.5a3]
+
+Tue Jun 14 02:02:43 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tkutil/tkutil.c: add TkUtil::CallbackSubst.subst_arg(m, ...)
+	  & _define_attribute_aliases(hash) to get substitution-argument from
+	  attributes (e.g. subst_arg(:x,:y,:num,:button) --> "%x %y %b %b ").
+
+	* ext/tk/lib/tk/event.rb: use _define_attribute_aliases().
+
+Mon Jun 13 13:03:08 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* hash.c (ruby_setenv): fixed SEGV. [ruby-dev:26186]
+
+Mon Jun 13 01:54:20 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (sigexit): call rb_thread_signal_exit() instead of
+	  rb_exit().  [ruby-dev:26347]
+
+	* eval.c (rb_thread_signal_exit): a new function to exit on main
+	  thread.
+
+	* eval.c (rb_thread_switch): exit status should be retrieved from
+	  ruby_errinfo.
+
+	* eval.c (rb_f_exit): ensure exit(0) should call
+	  exit(EXIT_SUCCESS).
+
+Mon Jun 13 01:20:02 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (rb_gc_mark_threads): curr_thread may not be part of the
+	  thread list.  [ruby-dev:26312]
+
+Sat Jun 11 22:34:44 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y: missing arg_paren event.  This patch is contributed by
+	  Mitchell N Charity.
+
+Fri Jun 10 23:55:17 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (unknown_node): show more information.  [ruby-dev:26196]
+
+Fri Jun 10 23:35:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* missing/mkdir.c: remove. [ruby-core:05177]
+
+Fri Jun 10 22:54:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing.h: fd_set stuffs need sys/types.h.  fixed: [ruby-core:05179]
+
+Thu Jun  9 23:58:12 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/Win32API/Win32API.c (Win32API_Call): disable global
+	  optimization. fixed: [ruby-core:05143]
+
+Thu Jun  9 23:35:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_inject): default the result value to Qundef to use
+	  first element as initial value if not given.
+
+Thu Jun  9 19:55:41 2005  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (ruby_longjmp): new macro to call longjmp, setcontext, etc.
+	  (ruby_setjmp): new macro to call setjmp, getcontext, etc.
+	  (ruby_setjmp): call setjmp before getcontext to avoid IA64 register
+	  stack problem.
+	  [ruby-talk:144939]
+
+	* gc.c (Init_stack): remove IA64_MAGIC_STACK_LIMIT.
+
+Thu Jun  9 18:24:16 2005  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in, eval.c, gc.c: use libunwind only on HP-UX.
+	  [ruby-dev:26297]
+
+Thu Jun  9 14:46:32 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (env_aset): do not treat nil as key-removing value.
+	  [ruby-list:40865]
+
+	* parse.y (method_call): allow aref expression ([]) to take a
+	  block.
+
+	* parse.y (block_dup_check): a function to check duplication of
+	  a block argument and an actual block.
+
+Thu Jun  9 11:55:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (SimpleDelegator::__setobj__): need check for
+	  recursive delegation.  [ruby-core:04940]
+
+Thu Jun  9 11:50:43 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb: add underscore aliases CGI::escape_html,
+	  CGI::unescape_html, CGI::escape_element, CGI::unescape_element.
+	  [ruby-core:05058]
+
+Wed Jun  8 18:47:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-expr-beg): fix looking point drift.
+
+Wed Jun  8 12:25:59 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_nitems): add the block feature to Array#nitems.
+	  suggested by Bertram Scharpf <lists@bertram-scharpf.de> in
+	  [ruby-talk:134083].
+
+Wed Jun  8 11:11:34 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (get2comp): revert all prior changes, and calculate
+	  proper 2's complement for negative numbers.
+
+Wed Jun  8 08:33:10 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_min_by, enum_max_by): return nil if no iteration.
+	  fixed: [ruby-dev:26245]
+
+	* eval.c (rb_need_block): ensure a block is given.
+
+	* eval.c (backtrace): skip successive frames sharing same node.
+
+Wed Jun  8 01:27:06 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (bignorm): fixed a bug in normalizing negative numbers
+	  reported from Honda Hiroki <hhonda@ipflex.com>.  normalizing
+	  should not trim leading zeros from negative numbers.
+
+	* bignum.c (rb_cstr_to_inum): must remove leading zeros for this
+	  case.
+
+Wed Jun  8 00:15:08 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_getaddrinfo__aix): merged a patch from
+	  KUBO Takehiro <kubo at jiubao.org> to support AIX.  [ruby-list:40832]
+
+Wed Jun  8 00:09:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb (Array::to_yaml): merged a patch from
+	  Tilman Sauerbeck <tilman at code-monkey.de>.  [ruby-core:05055]
+
+	* lib/yaml/rubytypes.rb (Hash::to_yaml): ditto.
+
+Wed Jun  8 00:00:01 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/curses/curses.c (curses_insertln): merged a patch from
+	  TAKAHASHI Tamotsu <ttakah at lapis.plala.or.jp>.  [ruby-ext:02305]
+
+Tue Jun  7 19:34:15 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/init.rb (IRB::IRB.rc_file_generators): more flexible
+	  IRB.rc_file_generators.  [ruby-core:05163]
+
+Tue Jun  7 18:39:31 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/thread.rb: RDoc documentation from Eric Hodel
+	  <drbrain@segment7.net> added.  [ruby-core:05148]
+
+Tue Jun  7 18:30:04 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): add .SUFFIXES from depend file.
+	  fixed: [ruby-dev:26294]
+
+Tue Jun  7 17:20:39 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): allow ';;' to be block terminator in
+	  place of 'end'.  [highly experimental]
+
+	* misc/ruby-mode.el (ruby-block-end-re): allow ';;' to be a
+	  negative indent trigger.  [highly experimental]
+
+	* parse.y (parser_yylex): small error fixed.
+
+Tue Jun  7 16:45:49 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): "respond_to?:foo" should be interpreted
+	  as "respond_to? :foo" at the command level.  [ruby-talk:144303]
+
+Tue Jun  7 16:32:53 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): raise exception on debug mode (-d),
+	  not verbose mode (-v/-w).  [ruby-core:05123]
+
+	* sprintf.c (rb_f_sprintf): warn always on verbose mode.
+
+Tue Jun  7 10:30:49 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: slave-ip fails to call procedures
+	  delegated by master-ip.
+
+Mon Jun  6 16:35:18 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/ripper/depend: add .y to .SUFFIXES for nmake.
+
+Sun Jun  5 23:00:35 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/console.rb: create console when required
+
+	* ext/tk/sample/tkextlib/tile/demo.rb: fix TypeError & create Console
+
+Sun Jun  5 10:23:52 2005  Tanaka Akira  <akr@m17n.org>
+
+	* signal.c (ruby_signal): don't set SA_RESTART.
+	  [ruby-dev:26276]
+
+Sat Jun  4 14:55:18 2005  Tanaka Akira  <akr@m17n.org>
+
+	* test/dbm/test_dbm.rb: merged from ext/dbm/testdbm.rb.
+
+	* test/gdbm/test_gdbm.rb: merged from ext/gdbm/testgdbm.rb.
+
+	* test/sdbm/test_sdbm.rb: renamed from ext/sdbm/testsdbm.rb with
+	  modification to use test/unit.
+
+Fri Jun  3 23:23:02 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h (rb_fdset_t): deal with fd bit sets over FD_SETSIZE.
+	  fixed: [ruby-dev:26187]
+
+	* eval.c (rb_fd_init, rb_fd_term, rb_fd_zero, rb_fd_set, rb_fd_clr,
+	  rb_fd_isset, rb_fd_copy): ditto.
+
+	* io.c (rb_io_wait_readable, rb_io_wait_writable, rb_f_select): ditto.
+
+	* ext/io/wait/wait.c (io_wait): ditto.
+
+	* ext/socket/socket.c (wait_connectable, unix_recv_io): ditto.
+
+Fri Jun  3 14:06:12 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: fix typo.
+
+Thu Jun  2 23:42:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y: pragma support on ripper.  [ruby-dev:26266]
+
+Thu Jun  2 00:02:16 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* struct.c: accessing >10 member caused segmentation fault.
+	  [ruby-dev:26247]
+
+	* test/ruby/test_struct.rb: test it.
+
+Wed Jun  1 11:30:09 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/Makefile.sub: can use single quote character in DESTDIR.
+	  [ruby-dev:26205]
+
+	* bcc32/Makefile.sub: Dir.glob in 1.9 doesn't treat \ as path separator.
+	  [ruby-dev:26254]
+
+Wed Jun  1 00:11:06 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (method_call): new experiment: "(expr)(args...)" to
+	  invoke "expr.call(args...)".  [EXPERIMENTAL]
+
+Tue May 31 23:43:41 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (command): revert implicit "call" for local variables.
+
+Tue May 31 15:52:45 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): should
+	  break the loop if the socket reached to EOF. [ruby-talk:142285]
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): send response
+	  without reading the whole request body if keep-alive is disabled.
+	  [experimental]
+
+Mon May 30 23:48:29 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/macpkg.rb: add PACKAGE_NAME information of Tcl/Tk
+	  Extension.
+
+	* ext/tk/lib/tk/msgcat.rb: ditto.
+
+	* ext/tk/lib/tk/winpkg.rb: ditto.
+
+	* ext/tk/lib/tkextlib/*: ditto.
+
+Sat May 28 16:39:21 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/openssl/test_x509store.rb: add test for expired CRL
+	  and refine some assertions.
+
+Sat May 28 05:15:44 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509store.c (ossl_x509stctx_set_time): should
+	  not set internal flag directory.
+
+Sat May 28 02:00:11 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#request_line):
+	  ENV["REQUEST_URI"] is better to get correct Request-URI
+	  than ENV["SCRIPT_NAME"] + ENV["PATH_INFO"].  [ruby-dev:26235]
+
+Fri May 27 16:32:04 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/mkmf.rb: use the semicolon as the path separator
+	  in the environment of MSYS.  fixed: [ruby-dev:26232]
+
+Thu May 26 20:31:21 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (remove_entry_secure): add documentation.
+
+	* lib/fileutils.rb (remove_entry_secure): should not invoke
+	  unlink(2) against a directory.
+
+Thu May 26 08:29:19 2005  Akiyoshi, Masamichi  <akiyoshi@hp.com>
+
+	* vms/vmsruby_private.c, vms/vmsruby_private.h: private routines
+	  for VMS port are added.
+
+	* eval.c (ruby_init): change to call VMS private intialization routine.
+
+Thu May 26 07:39:07 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (rm_r): use lchown(2), not chown(2).
+	  [ruby-dev:26226]
+
+	* lib/fileutils.rb (cd): remove :noop option. (feature change)
+
+	* lib/fileutils.rb (cp_r): should copy symlink as symlink, for
+	  also tree root. (feature change)
+
+	* lib/fileutils.rb (cp_r): new option :dereference_root.
+
+	* lib/fileutils.rb: new method remove_entry.
+
+	* lib/fileutils.rb: new method remove_entry_secure.
+
+	* lib/fileutils.rb: add documentation.
+
+Thu May 26 06:08:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: add shortcut-methods of tk_call + tk_split_list
+
+Wed May 25 20:06:27 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: TkComm#tk_split_*list fail to split a kind of SJIS
+	  strings. To avoid the trouble, add arguments to control converting
+	  encoding, and do split on a UTF8 string.
+
+	* ext/tk/lib/multi-tk.rb: modify to attend encoding.
+
+	* ext/tk/lib/remote-tk.rb: ditto.
+
+	* ext/tk/lib/tk/itemconfig.rb: ditto.
+
+	* ext/tk/lib/tk/listbox.rb: ditto.
+
+	* ext/tk/lib/tk/namespace.rb: ditto.
+
+	* ext/tk/lib/tk/panedwindow.rb: ditto.
+
+	* ext/tk/lib/tk/text.rb: ditto.
+
+	* ext/tk/lib/tk/textmark.rb: ditto.
+
+	* ext/tk/lib/tk/texttag.rb: ditto.
+
+	* ext/tk/lib/tk/variable.rb: ditto.
+
+	* ext/tk/lib/tk/winfo.rb: ditto.
+
+	* ext/tk/lib/tkextlib/iwidgets/scrolledlistbox.rb: ditto.
+
+	* ext/tk/lib/tkextlib/iwidgets/scrolledtext.rb: ditto.
+
+	* ext/tk/lib/tk.rb: add TkWindow#lower_window/raise_window and
+	  Tk#lower_window/raise_window by reason of method-name conflict
+
+	* ext/tk/lib/tk/canvas.rb: bug fix on TkCanvas#delete when given
+	  non-TkcItem arguments.
+
+	* ext/tk/lib/tkextlib/iwidgets/scrolledcanvas.rb: ditto.
+
+Wed May 25 19:48:12 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (rm_r): does chown(2). [ruby-dev:26199]
+
+Wed May 25 12:59:48 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI::Meta::RE_QUOTED_STRING): a content of
+	  quoted-string should be zero or more characters.
+
+Tue May 24 23:42:16 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (fix_pow): support Fixnum ** Float case directly
+	  without coercing.  [ruby-talk:142697] [ruby-talk:143054]
+
+Tue May 24 16:57:24 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.c (require_libraries): caused SEGV when continuation jumped
+	  in to the required library code.
+
+Tue May 24 17:45:59 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/readline/test_readline.rb: do not test libedit.
+	  fixed: [ruby-dev:26217]
+
+Tue May 24 06:45:31 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-font-lock-syntactic-keywords): string
+	  literals to be matched non-greedy.
+
+Tue May 24 00:39:14 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/calc: method name 'set' was able to crash with a class Set.
+	  [ruby-dev:26210]
+
+	* test/wsdl/document/test_rpc.rb: dateTime comparison failed under
+	  TZ=right/Asia/Tokyo (with leap second.) [ruby-dev:26208]
+
+Mon May 23 16:23:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: Framework support on MacOS X Tiger.
+
+	* ext/tk/README.tcltklib: add description of Framework support options.
+
+Mon May 23 15:07:34 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub ($(PROGRAM)): add dependency on $(LIBRUBY_SO).
+	  [experimental]
+
+Mon May 23 12:21:37 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (make_regexp): should not return junk address during
+	  compile time.  [ruby-dev:26206]
+
+Sun May 22 21:54:06 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/{soap,wsdl,xsd}, test/{soap,wsdl,xsd}: imported soap4r/1.5.4.
+
+	  == SOAP client and server ==
+
+	  === for both client side and server side ===
+
+	  * improved document/literal service support.
+	    style(rpc,document)/use(encoding, literal) combination are all
+	    supported.  for the detail about combination, see
+	    test/soap/test_style.rb.
+
+	  * let WSDLEncodedRegistry#soap2obj map SOAP/OM to Ruby according to
+	    WSDL as well as obj2soap.  closes #70.
+
+	  * let SOAP::Mapping::Object handle XML attribute for doc/lit service.
+	    you can set/get XML attribute via accessor methods which as a name
+	    'xmlattr_' prefixed (<foo name="bar"/> -> Foo#xmlattr_name).
+
+	  === client side ===
+
+	  * WSDLDriver capitalized name operation bug fixed.  from
+	    1.5.3-ruby1.8.2, operation which has capitalized name (such as
+	    KeywordSearchRequest in AWS) is defined as a method having
+	    uncapitalized name. (converted with GenSupport.safemethodname
+	    to handle operation name 'foo-bar').  it introduced serious
+	    incompatibility; in the past, it was defined as a capitalized.
+	    define capitalized method as well under that circumstance.
+
+	  * added new factory interface 'WSDLDriverFactory#create_rpc_driver'
+	    to create RPC::Driver, not WSDLDriver (RPC::Driver and WSDLDriver
+	    are merged).  'WSDLDriverFactory#create_driver' still creates
+	    WSDLDriver for compatibility but it warns that the method is
+	    deprecated.  please use create_rpc_driver instead of create_driver.
+
+	  * allow to use an URI object as an endpoint_url even with net/http,
+	    not http-access2.
+
+	  === server side ===
+
+	  * added mod_ruby support to SOAP::CGIStub.  rename a CGI script
+	    server.cgi to server.rb and let mod_ruby's RubyHandler handles the
+	    script.  CGIStub detects if it's running under mod_ruby environment
+	    or not.
+
+	  * added fcgi support to SOAP::CGIStub.  see the sample at
+	    sample/soap/calc/server.fcgi.  (almost same as server.cgi but has
+	    fcgi handler at the bottom.)
+
+	  * allow to return a SOAPFault object to respond customized SOAP fault.
+
+	  * added the interface 'generate_explicit_type' for server side
+	    (CGIStub, HTTPServer).  call 'self.generate_explicit_type = true'
+	    if you want to return simplified XML even if it's rpc/encoded
+	    service.
+
+	  == WSDL ==
+
+	  === WSDL definition ===
+
+	  * improved XML Schema support such as extension, restriction,
+	    simpleType, complexType + simpleContent, ref, length, import,
+	    include.
+
+	  * reduced "unknown element/attribute" warnings (warn only 1 time for
+	    each QName).
+
+	  * importing XSD file at schemaLocation with xsd:import.
+
+	  === code generation from WSDL ===
+
+	  * generator crashed when there's '-' in defined element/attribute
+	    name.
+
+	  * added ApacheMap WSDL definition.
+
+	* sample/{soap,wsdl}: removed.
+
+Sun May 22 19:11:35 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLServer#initialize):
+	  should initialize session id context. [ruby-core:4663]
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): add session id support.
+
+Sun May 22 12:30:58 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h, parse.y (ruby_pragma): removed.  fixed: [ruby-dev:26198]
+
+	* parse.y (parser_pragma): pragma name was ignored.
+
+Sun May 22 02:39:57 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (rm_r): new option :secure to avoid
+	  time-to-check-to-time-to-use security problem.  [ruby-dev:26100]
+
+	* lib/fileutils.rb (remove_file, remove_dir): try chmod(700) only
+	  on Windows.
+
+	* lib/fileutils.rb: does not depend on find.rb.
+
+	* lib/fileutils.rb: new method chmod_R.
+
+	* lib/fileutils.rb (chown_R): did not work.
+
+Sat May 21 10:23:21 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/Makefile.sub: tds files were not deleted when DESTDIR
+	  included '\' path delimiter. [ruby-dev:26193]
+
+Fri May 20 15:52:18 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  return 2 items if completion_proc returns only 1 item (for libedit).
+
+Fri May 20 01:24:33 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/extconf.rb: check rl_vi_editing_mode() and
+	  rl_emacs_editing_mode().
+
+Thu May 19 23:33:09 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/readline.c: supported libedit. fixed: [ruby-core:4858]
+
+	* ext/readline/extconf.rb: added new option --enable-libedit.
+
+	* test/readline/test_readline.rb: added assertions for
+	  Readline::HISTORY.
+
+	* lib/irb/input-method.rb: do not use Readline::HISTORY.pop.
+
+Wed May 18 23:42:25 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (exc_exception): reverted to call Exception#initialize
+	  directly.  fixed: [ruby-dev:26177]
+
+Wed May 18 17:38:51 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* dir.c (glob_helper): check whether path is "" before calling
+	  do_opendir.  [ruby-dev:26183]
+
+Wed May 18 13:40:48 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (NtInitialize): fix typo.
+
+Wed May 18 11:07:47 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): get rid of using String.  [ruby-dev:26180]
+
+	* eval.c (ruby_options), win32/win32.c (NtInitialize): move argument
+	  intialization back.  [ruby-dev:26180]
+
+Tue May 17 11:49:18 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (unixtime_to_filetime): use localtime() instead of
+	  gmtime() when using FileLocalTimeToFileTime().
+
+Mon May 16 22:42:52 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.h, {bcc32,win32,wince}/Makefile.sub: moved rb_[ugp]id_t
+	  to get rid of redefinition warnings on mingw.
+
+	* class.c (rb_class_init_copy): singleton class is disallowed to copy,
+	  from its definition.  fixed: [ruby-talk:142749]
+
+	* parse.y (pragma_encoding): add prototype to suppress false warning
+	  by VC.
+
+	* process.c (proc_spawn_v): use rb_w32_aspawn on Win32.
+
+Mon May 16 03:29:01 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.{h,c}: define rb_[pgu]id_t.
+
+Mon May 16 00:21:02 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#unlink): use SystemCallError instead of
+	  Errno::EISDIR because EISDIR is not portable.
+	  [ruby-core:5001]
+
+Sun May 15 22:28:10 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (DRbObject#method_missing): use raise(exception).
+	  [ruby-dev:26164]
+
+Sun May 15 18:56:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, ruby.h: define rb_[pgu]id_t macros instead of typedefs
+	  to get rid of types which might not be defined yet.  [ruby-dev:26165]
+
+Sun May 15 14:35:46 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#unlink): unlink a symlink to a directory
+	  was failed.  [ruby-core:4992]
+
+Sun May 15 09:57:30 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (unixtime_to_filetime): deal with DST.
+	  [ruby-talk:141817]
+
+Sat May 14 23:59:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (exc_exception, {exit,name_err,syserr}_initialize): call
+	  Execption#initialize.  fixed: [ruby-talk:142593]
+
+Sat May 14 23:56:41 2005  Erik Huelsmann  <ehuels@gmail.com>
+
+	* configure.in: Check for the availability of pid_t, gid_t and uid_t and
+	  remove AC_TYPE_UID_T.  fixed: [ruby-core:04745]
+
+	* defines.h: Remove pid_t typedef.
+
+	* ruby.h: Define rb_pid_t, rb_gid_t and rb_uid_t in accordance with
+	 the available system types.
+
+	* process.c: Change instances of pid_t and gid_t to their rb_*
+	 counterparts.
+
+	* ext/pty/pty.c: Change pid_t to rb_pid_t.
+
+	* vms/config.h: Define HAVE_{P,G,U}ID_T to 1.
+
+	* win32/Makefile.sub: Remove #define for {g,u}id_t.
+
+	* win32/win32.c: Change pid_t to rb_pid_t.
+
+	* wince/Makefile.sub: Remove #define for {g,u}id_t.
+
+	* wince/sys/types.h: Remove definitions of {p,g,u}id_t.
+
+Sat May 14 11:47:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h (ruby_pragma): prototype.  [ruby-core:04881]
+
+	* parse.y (parser_pragma): parse Emacsen hack.
+
+	* parse.y (parser_prepare): deal with specific syntax at the top.
+
+	* ruby.c (load_file): read the first line iff it started with shebang.
+
+Fri May 13 23:44:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: keep srcdir unexpanded.
+
+	* lib/mkmf.rb (create_makefile): quote topdir and hdrdir if necessary.
+	  fixed: [ruby-core:04932]
+
+	* lib/mkmf.rb (configuration), {bcc32,win32,wince}/Makefile.sub: make
+	  also INSTALL_PROG and INSTALL_DATA system dependent.
+	  fixed: [ruby-core:04931]
+
+Fri May 13 23:32:55 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (unknown_node): add volatile directive to prototype.
+
+Fri May 13 17:50:49 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* variable.c (generic_ivar_get): rb_attr_get should not warn.
+	  [ruby-dev:26010]
+
+Thu May 12 17:41:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: follow nkf 2.0.5
+
+Thu May 12 16:50:40 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/rdoc/parsers/parse_c.rb: more readability for mixing
+	  progress "c..." and warning message.
+
+Thu May 12 15:50:56 2005  Tilman Sauerbeck  <tilman@code-monkey.de>
+
+	* lib/rdoc/parsers/parse_c.rb: show parsing progress for C files.
+	  [ruby-core:4341]
+
+Thu May 12 09:53:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* version.c (ruby_show_version): flush for non-tty stdout.
+
+Thu May 12 01:23:55 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (unknown_node): ignore broken NODE to get rid of accessing
+	  possibly inaccessible address.  fixed: [ruby-dev:26122]
+	  should emit more useful information like [ruby-dev:26126], though.
+
+Wed May 11 15:58:39 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (break_jump): break should not cross functions.
+	  [ruby-list:40818]
+
+Wed May 11 10:41:54 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/tempfile.rb (Tempfile#unlink): fixed typo.
+
+Wed May 11 01:03:36 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (TMP_ALLOC): use macro NEW_NODE() to get rid of warnings on
+	  platforms which have no alloca().  fixed: [ruby-talk:141301]
+
+Sun May  8 23:17:47 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/timer.rb: fix typo.
+
+Sun May  8 21:00:50 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* hash.c (Init_Hash): remove custom "hash" and "eql?".
+	  (ported from 1.8) [ruby-dev:26132]
+
+Sun May  8 16:50:25 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/profiler.rb: fixed "undefined method `[]' for nil:NilClass"
+	  [ruby-core:4775] [ruby-talk:140401] [ruby-dev:26118]
+
+Sat May  7 22:58:00 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_var): no libs argument is given.
+
+Fri May  6 08:08:37 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c:rb_hash_hash_i() should be static.  [ruby-core:04815]
+
+	* re.c should include regint.h for declarations of oniguruma
+	  functions.  [ruby-core:04815]
+
+Sun May  1 09:15:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_sflag): replace '-' in variable names with '_'.
+	  [ruby-dev:26107]
+
+	* eval.c (rb_eval), parse.y (arg): reduce fixnum range literal at
+	  parser.  fixed: [ruby-dev:26113]
+
+Sat Apr 30 11:59:25 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_FUNC_ATTRIBUTE): check for function attribute.
+	  [ruby-dev:26109]
+
+	* eval.c, gc.c: moved noinline to configure.in.
+
+	* rubyio.h (DEPRECATED): moved to configure.in.
+
+	* ruby.h (DEPRECATED, NOINLINE): default definition.
+
+	* win{32,ce}/Makefile.sub (config.h): deprecated and noinline for
+	  __declspec() are available for VC++7 or later.
+
+Sat Apr 30 06:57:39 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb: new methods WEBrick::CGI#[], WEBrick::CGI#logger
+	  and WEBrick::CGI#config.  these are necessary to use an instance of
+	  WEBrick::CGI as the first argument of HTTPServlet#get_instance.
+	  (suggested by Tatsuki Sugiura)
+
+	* lib/webrick/cgi.rb
+	  (WEBrick::CGI#initalize): set a dummy to @config[:ServerSoftware]
+	  if SERVER_SOFTWARE environment variable is not given.
+	  (WEBrick::CGI#start): req.path_info must be a String.
+	  (WEBrick::CGI::Socket#request_line): treat REQUEST_METHOD, PATH_INFO
+	  and SCRIPT_NAME to run in console.
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.escape_path): should
+	  not use String#split("/"). it removes trailing empty path component.
+
+Thu Apr 28 08:21:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (set_arg0): use also environment variable space for setting
+	  $0.  [ruby-core:04774]
+
+Wed Apr 27 23:42:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (OPTFLAGS): default global optimization to
+	  disabled only for VC++6.
+
+Tue Apr 26 22:58:00 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_invoke_core): call Tcl's "::unknown"
+	  command when can't get information of target command.
+
+Mon Apr 25 13:54:55 2005  speakillof  <speakillof@yahoo.co.jp>
+
+	* lib/rexml/encodings/SHIFT-JIS.rb: encoding and decoding were
+	  swapped. [ruby-core:4772]
+
+Mon Apr 25 01:18:43 2005  Tanaka Akira  <akr@m17n.org>
+
+	* oniguruma.h (OnigWarnFunc): add a variadic argument.
+	  [ruby-core:4751]
+
+Sat Apr 23 19:49:21 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c (ip_RubyExitCommand): exit with status code
+	  via TclTkIp#_eval didn't work. [ruby-talk:139390]
+
+Sat Apr 23 11:45:29 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_provided): should check also path name to be loaded.
+	  fixed: [ruby-dev:26093]
+
+Fri Apr 22 16:55:35 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tcltklib.c (ip_set_exc_message): fixed memory leak.
+
+	* ext/tk/tcltklib.c: eTkCallbackReturn was not initialized.
+
+Thu Apr 21 06:45:28 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_incpush_expand, proc_options): expand relative path
+	  given with -I option.  [ruby-dev:26090]
+
+	* configure.in, lib/mkmf.rb, {bcc32,win32,wince}/Makefile.sub: improve
+	  C++ support.  [ruby-dev:26089]
+
+Thu Apr 21 01:53:09 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: add rdoc.
+
+Thu Apr 21 00:07:50 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): support platforms have file separator
+	  other than /.
+
+	* {bcc32,win32,wince}/Makefile.sub (BUILD_FILE_SEPARATOR): separator
+	  of building platform.
+
+	* {bcc32,win32,wince}/Makefile.sub (CP, INSTALL): use COPY command.
+
+Wed Apr 20 23:22:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.mk: miniruby depends on MINIOBJS.
+
+	* dmydln.c (dln_load): dummy function to raise LoadError.
+
+	* cygwin/GNUmakefile.in, {bcc32,win32,wince}/Makefile.sub: miniruby
+	  can't load extensions on Windows.
+
+Wed Apr 20 23:01:35 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/ifchange.bat: delete testing files.
+
+Wed Apr 20 22:54:54 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: new method Net::HTTP.post_form.
+
+	* lib/net/http.rb: new method Net::HTTPHeader#set_form_data and
+	  its alias #form_data=.
+
+	* lib/net/http.rb: Net::HTTPHeader#add_header -> add_field
+	  (adjustted to Ruby 1.8).
+
+Wed Apr 20 10:53:30 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_rb.rb (lex_init): use IRB module.
+	  [ruby-core:04737]
+
+Wed Apr 20 07:27:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {bcc32,win32,wince}/configure.bat, {bcc32,win32,wince}/setup.mak:
+	  add extout option.
+
+	* bcc32/setup.mak: make configuration variables overridable.
+
+Tue Apr 19 23:37:09 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/ftools.rb (File.safe_unlink): do not modify a symlinked file.
+
+Tue Apr 19 23:02:40 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (search_required): deal with features with path too.
+
+	* intern.h (rb_file_expand_path): prototype.  fixed: [ruby-dev:26082]
+
+Tue Apr 19 08:38:07 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (search_required, rb_require_safe): expand path in
+	  rb_features.  [ruby-dev:26079]
+
+	* file.c (rb_find_file_ext): return absolute path.
+
+	* ext/extmk.rb: expand path for ext/**/extconf.rb.
+
+	* eval.c (search_required): handle static linked extensions.
+
+Mon Apr 18 15:37:35 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_attr): attribute name check added.
+
+	* numeric.c (flo_plus): small typo fix.
+
+Mon Apr 18 11:25:14 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/zlib/zlib.c (zstream_run): fixed SEGV. [ruby-core:4712]
+
+Sun Apr 17 23:57:49 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake, parse_args): do not expand destdir.
+
+	* ext/extmk.rb (relative_from): treat mere drive letter as an absolute
+	  path.
+
+Sat Apr 16 17:01:16 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/tdiary_plugin/rss-recent.rb (rss_recent_cache_rss):
+	  use the first date information of items as site date information
+	  if channel doesn't have date information.
+
+Sat Apr 16 15:27:03 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_PROG_INSTALL): not add -p option to INSTALL.
+	  files need timestamps to be kept are only ar-archive on a few
+	  platforms, and be installed by instruby.rb but not INSTALL.
+	  fixed: [ruby-core:04721]
+
+	* mkconfig.rb: purge autoconf value variables.
+
+Sat Apr 16 10:33:48 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/Makefile.sub: quick hack... prepend DESTDIR.
+	  still have restriction on DESTDIR ("", "/", "e:")
+
+Sat Apr 16 03:59:42 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: check for OPENSSL_cleanse.
+
+	* ext/openssl/openssl_missing.h: ditto.
+
+Fri Apr 15 22:40:19 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/tests/testWIN32OLE.rb: add test for WIN32OLE.codepage=
+
+	* ext/win32ole/tests/testOLETYPELIB.rb: correct expected message.
+
+Fri Apr 15 22:04:07 2005  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(ole_invoke): retry after converting Qnil
+	  to VT_EMPTY.
+
+Thu Apr 14 19:05:06 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper] (regexp): dispatch regexp option.
+	  [ruby-Bugs:1688]
+
+	* ext/ripper/lib/core.rb: regenerated (interface changed).
+
+Thu Apr 14 18:59:43 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (remove_file): ignore exceptions caused by
+	  chmod.
+
+	* lib/fileutils.rb (remove_dir): try to get rights to rmdir.
+	  [ruby-Bugs:1502]
+
+Thu Apr 14 18:51:02 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb, lib/irb/slex.rb: bug fix of [ruby-Bugs-1745]
+
+	* lib/irb/ext/loader.rb, lib/irb/ext/save-history.rb:
+	  fix location of @RCS_ID
+
+	* lib/irb/cmd/help.rb: a lost of release IRB 0.9.5.
+
+Thu Apr 14 15:10:30 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* lib/irb/notifier.rb, lib/irb/output-method.rb, lib/irb/ext/history.rb
+	  fixed warning of 'ruby -w'
+
+Thu Apr 14 05:35:45 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* doc/irb/irb.rd.ja: a lost  of release IRB 0.9.5.
+
+	* lib/irb/slex.rb: bug fix by [ruby-core:04707].
+
+Thu Apr 14 00:20:31 2005  Keiju Ishitsuka  <keiju@ruby-lang.org>
+
+	* bin/irb lib/irb.rb lib/irb/...: IRB 0.9.5.
+
+Wed Apr 13 23:40:21 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::VERSION): 0.1.3 -> 0.1.4.
+
+	* lib/rss/rss.rb (RSS::Element#converter): fixed converter
+	  transmission bug.
+
+Wed Apr 13 22:12:16 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#order!): call handlers iff matches
+	  non-switch.
+
+Wed Apr 13 21:20:35 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (mingw32): extract msvcr*.dll from objdump result.
+
+Wed Apr 13 19:25:31 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (mingw32): use actual runtime DLL name as ruby DLL
+	  name and default load path.
+
+	* win32/Makefile.sub, win32/setup.mak: ditto.
+
+Tue Apr 12 19:30:36 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#make_switch, OptionParser#order!):
+	  added non-option and end-of-args handler.  [ruby-talk:136878]
+	  [EXPERIMENTAL]
+
+Tue Apr 12 15:33:09 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_finalize): better modification than the
+	  previous commit [ruby-dev:26029].
+
+Tue Apr 12 12:38:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_finalize): fix SEGV when Tcl_GlobalEval()
+	  modifies the argument string to eval.
+
+Tue Apr 12 02:21:55 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_finalize): add existence check of
+	  Tcl commands before calling Tcl_GlobalEval().
+
+Mon Apr 11 23:36:04 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: [druby-ja:123] fix: When reference of my object is
+	  loaded, the object is tainted.
+
+	* test/drb/test_drb.rb: ditto.
+
+Mon Apr 11 22:18:23 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* dir.c, file.c (lstat): avoid warnings for mingw.
+
+Mon Apr 11 20:11:06 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (ip_finalize): adhoc patch to avoid SEGV when exit
+	  on Tcl/Tk8.3.x.
+
+Mon Apr 11 15:24:20 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): shouldn't output hdrdir twice.
+
+Sat Apr  9 18:20:31 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/image.rb: support to create TkImage object without
+	  creating a new image object on Tk.
+
+	* ext/tk/lib/tk/menu.rb: use TkCommandNames on create_self()
+
+	* ext/tk/lib/tk/root.rb: TkRoot.to_eval() returns '.'.
+
+	* ext/tk/lib/tk/text.rb: add methods to create a TkText::IndexString
+	  from (x, y) coords.
+
+	* ext/tk/lib/tkextlib/tile/: add demo and update support status.
+
+Sat Apr  9 14:42:29 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/tdiary_plugin/rss-recent.rb: supported configuration
+	  via Web browser.
+
+Fri Apr  8 20:17:48 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): hdrdir needs to be defined also in
+	  Config::CONFIG.
+
+	* lib/mkmf.rb (configuration, create_makefile): get rid of recursive
+	  macro reference.
+
+Fri Apr  8 01:55:20 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/demos-{en,jp}/goldberg.rb: reduced window size.
+	  [ruby-dev:25992]
+
+Thu Apr  7 23:58:40 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): keep directory names in Makefile as macros.
+
+	* lib/mkmf.rb (configuration, create_makefile): ditto.
+
+	* lib/mkmf.rb (CXX_EXT): separate C++ extensions.
+
+Thu Apr  7 17:24:17 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_call0): "return" event hook should be always executed
+	  if event_hooks is set.
+
+Thu Apr  7 14:33:09 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_maker_dc.rb (test_date): added a test for #date=
+	  and #dc_date=.
+
+Thu Apr  7 11:49:53 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/dublincore.rb: _really_ supported multiple Dublin
+	  Core items.
+
+	* test/rss/rss-assertions.rb (assert_multiple_dublin_core): added
+	  an assertion for testing multiple Dublin Core items.
+
+	* test/rss/test_maker_dc.rb (test_rss10_multiple): added a test
+	  for making multiple Dublin Core items.
+
+Wed Apr  6 16:06:30 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_env.rb (test_key): should test ENV.key instead of
+	  ENV.index. [ruby-dev:25994]
+
+Tue Apr  5 16:01:12 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/*: refactored.
+	  - gave a name to 'x'.
+	  - undef_method -> remove_method for avoiding a warning in ruby 1.6.
+
+Tue Apr  5 15:45:33 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/tdiary_plugin/rss-recent.rb:
+	  new option: @options['rss-recent.use-image-link']:
+	  use image as link instread of text if available.
+
+	* sample/rss/tdiary_plugin/rss-recent.rb (RSS_RECENT_VERSION):
+	  0.0.5 -> 0.0.6.
+
+Tue Apr  5 15:15:26 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/dublincore.rb: supported multiple Dublin Core items.
+
+	* lib/rss/parser.rb: added class name registry for complex model
+	  elements. (ex. have childlen elements, have some attributes and
+	  a child element and so on.)
+
+	* lib/rss/maker/base.rb: added default current_element implementation.
+
+	* lib/rss/maker/dublincore.rb: supported multiple Dublin Core
+	  items.
+
+	* lib/rss/maker/image.rb: supproted new Dublin Core API.
+
+	* lib/rss/trackback.rb (RSS::TrackBackUtils.new_with_value_if_need):
+	  moved to RSS::Utils.
+
+	* lib/rss/utils.rb (RSS::Utils.new_with_value_if_need):
+	  moved from RSS::TrackBackUtils.
+
+	* lib/rss/maker/image.rb: fixed invalid argument of
+	  add_need_initialize_variable bug.
+
+	* lib/rss/maker/trackback.rb: ditto.
+
+	* lib/rss/rss.rb (Hash#merge): added for ruby 1.6.
+
+	* lib/rss/rss.rb (RSS::BaseModel.date_writer): changed to accept nil
+	  for date value.
+
+	* test/test_dublincore.rb: added tests for plural accessor and
+	  multiple Dublin Core items.
+
+	* test/test_setup_maker_1.0.rb: fixed swapped actual and expected
+	  values.
+
+Mon Apr  4 23:17:52 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb (TkComm#array2tk_list): accept enc-mode argument to
+	  decide whether convert encoding of each element or not.
+
+	* ext/tk/lib/tk/variable.rb (TkVariable#value=): fail to convert the
+	  encoding of array elements when assign an array to an TkVariable
+	  object.
+
+Mon Apr  4 10:26:48 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/lib/tk/dialog.rb: fixed typo.
+
+Sat Apr  2 23:38:54 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (CP, INSTALL): get rid of less portable options.
+
+	* lib/mkmf.rb (configuration, create_makefile): correct configuration
+	  variable.
+
+	* wince/configure.bat, wince/setup.mak: add prefix, extstatic and
+	  rdoc options.
+
+	* lib/mkmf.rb (create_makefile): ensure library directories get made
+	  before copying libraries there.
+
+Sat Apr  2 16:59:46 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: forgot to update RELEASE_DATE
+
+	* ext/tk/lib/tk/variable.rb: fix namespace trouble when autoloading
+
+	* ext/tk/lib/tk/palette.rb: define Tcl variable 'tkPalette' as global
+
+	* ext/tk/lib/tk/dialog.rb: use array2tk_list method when calling
+	  Tk.ip_eval.
+
+	* ext/tk/lib/tk/autoload.rb: add autoload entry 'TkDialogObj' and
+	  'TkWarningObj'
+
+Sat Apr  2 13:23:17 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* hash.c (env_key): ENV.index is deprecated as well as Hash#index.
+	  use ENV.key instead. [ruby-dev:25974]
+
+Sat Apr  2 02:19:11 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb (TkWindow.initialize): accept 'without_creating'
+	  option without 'widgetname' option to allow creating a widget object
+	  which is used as an argument of Tcl/Tk's widget allocation commands.
+
+	* ext/tk/lib/tk/image.rb (TkImage.initialize): accept 'imagename'
+	  option to create a image object by the given name.
+
+Thu Mar 31 22:23:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (SRC_EXT): exclude just case different suffixes on case
+	  insensitive file system platforms.
+
+	* README.EXT, README.EXT.ja (Appendix C): utility functions.
+
+Thu Mar 31 14:08:43 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_load): should return
+	  value. [ruby-dev:25971]
+
+Thu Mar 31 11:07:50 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb: @@setter -> @@setters.
+
+	* lib/rss/parser.rb
+	  (RSS::BaseListener.register_uri)
+	  (RSS::BaseListener.uri_registered?)
+	  (RSS::BaseListener.install_get_text_element):
+	  swapped the first argument and the second argument.
+
+	* lib/rss/taxonomy.rb: swapped the first argument and the second
+	  argument for RSS::BaseListener.install_get_text_element.
+	* lib/rss/image.rb: ditto.
+	* lib/rss/syndication.rb: ditto.
+	* lib/rss/dublincore.rb: ditto.
+	* lib/rss/parser.rb: ditto.
+	* lib/rss/1.0.rb: ditto.
+	* lib/rss/2.0.rb: ditto.
+	* lib/rss/0.9.rb: ditto.
+	* lib/rss/content.rb: ditto.
+
+Thu Mar 31 11:00:36 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb
+	  (RSS::BaseListener.install_setter)
+	  (RSS::BaseListener.register_uri): changed fallback way.
+
+Thu Mar 31 08:25:40 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RUBYOPT): clear for the environment RubyGems installed.
+
+	* common.mk (clean-local): keep $(PREP) files till distclean.
+
+	* common.mk (check): do all tests.
+
+Thu Mar 31 06:00:20 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_load): should not raise
+	  error even if the specified engine could not be loaded. (Dynamic
+	  engines don't have fixed name to load.)
+
+Wed Mar 30 17:41:48 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: add TclTkIp#_create_console() method to create
+	  a Tcl/Tk's console window.
+
+	* ext/tk/lib/multi-tk.rb: support TclTkIp#_create_console() method.
+
+	* ext/tk/lib/remote-tk.rb: ditto.
+
+	* ext/tk/lib/tk/console.rb: ditto.
+
+	* ext/tk/lib/tk.rb: update RELEASE_DATE
+
+	* ext/tk/sample/demo-*/check2.rb: use 'return' in the Proc object.
+
+	* ext/tk/sample/tkextlib/**: ditto.
+
+Tue Mar 29 22:20:49 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb: use DRbObject.new_with instead of reinit.
+	  [ruby-dev:25961]
+
+Tue Mar 29 00:04:57 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: move method DRbObject#reinit to DRbObject.new_with.
+	  extract method DRbObject.prepare_backtrace. add DRb.regist_server,
+	  remove_server, fetch_server. change server in thread variable if
+	  in-proc server. [druby-ja:113]
+
+	* lib/drb/gw.rb: ditto.
+
+Mon Mar 28 20:53:44 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): nothing to be removed when no file
+	  was deleted.
+
+	* ext/extmk.rb (extmake): restore srcdir.
+
+Mon Mar 28 08:39:49 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_create): Iconv::Failure requires 3
+	  arguments.  (pointed out by NaHi)
+
+Sun Mar 27 00:56:58 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (remove_file): ignore Errno::E* if force option
+	  is set. [ruby-dev:25944]
+
+Sat Mar 26 22:51:33 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb (_callback_entry_class?): add for checking whether
+	  a class is available for a callback entry.
+
+	* ext/tk/lib/tk.rb (after_cancel): add Tk.after_cancel(afterID) method.
+
+	* ext/tk/lib/tk.rb (array2tk_list): change from private module method
+	  of TkComm to public module method.
+
+	* ext/tk/lib/tk.rb (cget): add check that slot argument is not
+	  empty string.
+
+	* ext/tk/lib/tk.rb (configinfo): ditto.
+
+	* ext/tk/lib/tk/itemconfig.rb (itemcget): add check that slot argument
+	  is not empty string.
+
+	* ext/tk/lib/tk/itemconfig.rb (itemconfiginfo): ditto.
+
+	* ext/tk/lib/tk/entry.rb: add TkEntry#icursor and icursor= (alias of
+	  cursor and cursor= method).
+
+	* ext/tk/lib/tk/font.rb: improve font treatment when the font name is
+	  empty string.
+
+	* ext/tk/lib/tk/variable.rb: add :variable, :window and :procedure
+	  type.
+
+	* ext/tk/lib/tk/variable.rb: improve treatment of array-type
+	  tkvariable.
+
+	* ext/tk/lib/tkextlib/blt.rb: add commands for zooming.
+
+	* ext/tk/lib/tkextlib/blt/*: bug fix.
+
+	* ext/tk/lib/tkextlib/treectrl/tktreectrl.rb: bug fix and add methods
+	  to call TreeCtrl commands for bindings.
+
+	* ext/tk/sample/tkextlib/blt/*: new sample scripts.
+
+	* ext/tk/sample/tkextlib/treectrl/*: ditto.
+
+Fri Mar 25 10:53:16 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (WIN32_LEAN_AND_MEAN): removed because a lot of
+	  troubles.  [ruby-list:40721]
+
+Thu Mar 24 23:10:44 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (macro_defined?): try to compile for an old compiler
+	  which doesn't bail out at #error directive.  [ruby-dev:25818]
+
+	* lib/mkmf.rb (check_sizeof): refine logging messages.
+
+Wed Mar 23 19:08:10 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/utils.rb (WEBrick::Utils.create_listeners):
+	  - should raise ArgumentError if no port is specified.
+	  - even if the specified port is 0, all TCPServers should be
+	    initialized with the port given to the first one.
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#initialize): if :Port
+	  parameter is 0, it should be updated with the port number which
+	  ectually listened.
+
+Wed Mar 23 16:12:40 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* parse.y (primary): fix lineno of rescue and ensure.
+
+Wed Mar 23 00:39:05 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_settracefunc.rb (test_event): added tests for
+	  "class" and "end" and "raise".
+
+Sun Mar 20 22:51:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (mkmf_failed): check if Makefile is created without
+	  create_makefile.
+
+Sat Mar 19 23:48:10 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-expr-beg): returned true always.
+	  fixed: [ruby-list:40683]
+
+Sat Mar 19 00:41:02 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/font.rb: add some TkFont class methods to get font
+	  information without creating a TkFont object.
+
+	* ext/tk/lib/tkextlib/treectrl/tktreectrl.rb: bug fix and define some
+	  classes for components of Tk::TreeCtrl
+
+Thu Mar 17 17:42:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (make_struct): allow non local-id field
+	  names. [ruby-core:04575]
+
+	* struct.c (inspect_struct): ditto.
+
+Wed Mar 16 23:39:13 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_settracefunc.rb: added test for c-return.
+
+Wed Mar 16 22:57:43 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_call0): call_cfunc() should be protected.
+
+	* eval.c (rb_add_event_hook): use K&R style.
+
+	* eval.c (rb_remove_event_hook): ditto.
+
+Wed Mar 16 22:03:15 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_add_event_hook): new function to add a hook function for
+	  interpreter events.
+
+Wed Mar 16 18:08:32 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): reorganize "return" event post.
+
+	* eval.c (return_jump): no need to post "return" event here.
+
+Tue Mar 15 23:49:19 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (Init_iconv): InvalidEncoding also should include
+	  Iconv::Failure.
+
+Tue Mar 15 23:12:36 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (recursive_check, recursive_push): more restrictive check.
+	  fixed: [ruby-dev:25916]
+
+Tue Mar 15 16:38:31 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tkutil/tkutil.c (ary2list): give wrong arguments to hash2kv()
+
+Mon Mar 14 19:39:33 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/timer.rb (TkTimer): forgot to clear @return_value
+	  when restarting
+
+	* ext/tk/lib/tk/sample/cd_timer.rb: new sample of TkRTTimer
+
+Mon Mar 14 12:21:03 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/timer.rb (TkRTTimer): forgot to reset the callback
+	  time. So, 'continue' do all callbacks between 'stop' and 'continue'.
+
+Mon Mar 14 08:14:56 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (str_to_id): raise ArgumentError for NUL containing
+	  strings.
+
+Mon Mar 14 00:13:49 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/timer.rb (TkRTTimer): correct calculation of offset
+	  value. get a little better accuracy.
+
+	* ext/tk/sample/demos-en/widget: use a binding with no local variables
+	  when eval a sample script.
+
+	* ext/tk/sample/demos-en/bind.rb: ditto.
+
+	* ext/tk/sample/demos-en/tcolor: ditto.
+
+	* ext/tk/sample/demos-jp/widget: ditto.
+
+	* ext/tk/sample/demos-jp/bind.rb: ditto.
+
+	* ext/tk/sample/demos-jp/tcolor: ditto.
+
+Sun Mar 13 22:19:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (recursive_pop): raise TypeError instead of fatal error.
+	  fixed: [ruby-dev:25843]
+
+Sun Mar 13 10:09:17 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb: remove test_gc. [ruby-dev:25871]
+
+Sun Mar 13 02:32:54 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_tmp_dh_callback): should get DH
+	  parameter from the current SSL object.
+
+Sun Mar 13 02:09:03 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_pkey_dh.c (ossl_create_dh): fix typo.
+	  patch from IWATSUKI Hiroyuki. [ruby-dev:25867]
+
+	* ext/openssl/ossl_ssl.c (ossl_tmp_dh_callback): ditto.
+	  (ossl_call_tmp_dh_callback): ditto
+
+Fri Mar 11 03:24:59 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (primary): wrong var node was set for NODE_LAMBDA.
+	  [ruby-core:04555]
+
+Thu Mar 10 19:10:29 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (lib_eventloop_ensure): mis-delete a timer handler
+	  when exit from a recursive called eventloop
+
+	* ext/tk/lib/tk/timer.rb: new TkRTTimer class, which can works for a
+	  realtime operation
+
+	* ext/tk/sample/tkrttimer.rb: sample of TkRTTimer class
+
+	* ext/tk/lib/tk/textmark.rb: move  TkTextMark#+ and TkTextMark#- to
+	  TkText::IndexModMethods
+
+	* ext/tk/lib/tk/text.rb: improve TkTextMark#+ and TkTextMark#-, and
+	  add them to TkText::IndexModMethods module
+
+	* ext/tk/sample/tktextio.rb: add test part of "seek by text index
+	  modifiers"
+
+Thu Mar 10 08:10:11 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (make_regexp): need to free internal regexp structure when
+	  compilation fails.  [ruby-talk:133228]
+
+Thu Mar 10 01:08:20 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (bv_decl): remove initialize rule from block local
+	  variable declaration.
+
+Wed Mar  9 23:55:34 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb (PP::PPMethods#guard_inspect_key): support
+	  __recursive_key__.  [ruby-dev:25821]
+
+Wed Mar  9 19:42:21 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c: OpenSSL::SSL::SSLContexts suports callbacks:
+	  - SSLContext#client_cert_cb is a Proc. it is called when a client
+	    certificate is requested by a server and no certificate was yet
+	    set for the SSLContext. it must return an Array which includes
+	    OpenSSL::X509::Certificate and OpenSSL::PKey::RSA/DSA objects.
+	  - SSLContext#tmp_dh_callback is called in key exchange with DH
+	    algorithm. it must return an OpenSSL::PKey::DH object.
+
+	* ext/openssl/ossl_ssl.c:
+	  (ossl_sslctx_set_ciphers): ignore the argument if it's nil.
+	  (ossl_start_ssl, ossl_ssl_write): call rb_sys_fail if errno isn't 0.
+	  [ruby-dev:25831]
+
+	* ext/openssl/ossl_pkey.c
+	  (GetPrivPKeyPtr, ossl_pkey_sign): should call rb_funcall first.
+	  (DupPrivPKeyPtr): new function.
+
+	* ext/openssl/ossl_pkey_dh.c: add default DH parameters.
+
+	* ext/openssl/ossl_pkey.h: ditto.
+
+	* ext/openssl/lib/openssl/cipher.rb: fix typo. [ruby-dev:24285]
+
+Wed Mar  9 18:09:51 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (gettable_gen): warns if VCALL name is used as
+	  out-of-scope block local variable.  [EXPERIMENTAL]
+
+	* parse.y (opt_bv_decl): add explicit block local variable
+	  declaration.  raises error for name conflicts.  [EXPERIMENTAL]
+
+Wed Mar  9 13:37:57 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tktextio.rb: fix bug of handling 'end' position.
+	  support initial text, overwrite setting and pos_gravity control.
+
+Tue Mar  8 18:16:55 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tktextio.rb: New sample script. TkTextIO class in this
+	  sample supports to use a text widget as if it is a I/O stream (such
+	  like as StringIO class).
+
+Tue Mar  8 13:39:25 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/mkconstants.rb: workaround for some of 4.4BSD-Lite
+	  derived OSs.
+
+Tue Mar  8 12:36:17 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c: document from Sam Roberts
+	  <sroberts@uniserve.com> for getsockopt and setsockopt is merged.
+	  [ruby-doc:824]
+
+Tue Mar  8 10:48:53 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (rb_exec_recursive): declaration should precede statements
+	  before C99.
+
+Tue Mar  8 10:05:40 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (errno_missing): Errno.const_missing to allow references
+	  to SyscallError exceptions not defined on the platform.
+	  [ruby-core:04522]
+
+	* error.c (Init_syserr): Errno::NOERROR(0) for fallback exception.
+
+Tue Mar  8 01:19:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: follow nkf 1.66
+	  fixed: [ruby-dev:25828]
+
+Mon Mar  7 21:29:40 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#start): should
+	  restore @token if accept failure. suggested by Dominique Brezinski.
+	  [ruby-core:04518]
+
+	* sample/webrick/httpsd.rb: fix typo in comment. suggested by
+	  Kazuhiko Shiozaki.
+
+Mon Mar  7 21:01:37 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_require_safe): get actual path string under safe level
+	  when requested.  fixed: [ruby-dev:25815]
+
+Mon Mar  7 16:46:02 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_start_ssl, ossl_ssl_read,
+	  ossl_ssl_write): need to set errno on Win32 platform.
+
+Mon Mar  7 14:55:43 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (block_pass): should not push unique number if a block is
+	  not an orphan.  [ruby-dev:25808]
+
+Mon Mar  7 14:13:23 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#initialize):
+	  should set @eof and @rbuffer.
+
+Mon Mar  7 10:28:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (inspect_obj): unintended space removal.
+	  [ruby-dev:25810]
+
+	* eval.c (rb_exec_recursive): should not use NODE in disclosed
+	  context.  [ruby-dev:25812]
+
+	* io.c (rb_f_open): need not to check if to_open value is a
+	  T_FILE.  [ruby-dev:25812]
+
+Mon Mar  7 01:21:01 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tkutil/tkutil.c: follow the change of st.c (committed
+	  at Fri, 4 Mar 2005 15:47:47 +0900 by matz)
+
+Mon Mar  7 00:01:55 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: fail to call TclTkLib.mainloop when $SAFE==4
+
+Sun Mar  6 13:04:10 2005  Dee Zsombor  <zsombor@ruby-lang.org>
+
+	* misc/ruby-electric.el: added.
+	* misc/Readme: updated.
+
+Sun Mar  6 11:47:10 2005  Sam Roberts  <sroberts@uniserve.com>
+
+	* lib/pp.rb: rdoced.  [ruby-core:4490]
+
+Sun Mar  6 11:36:37 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb (File::Stat#pretty_print): Etc.getpwuid and Etc.getgrgid
+	  may return nil.  [ruby-talk:129826]
+	  reported by Daniel Berger.
+
+Sun Mar  6 06:34:31 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_start_ssl): should wait for that
+	  the underlying IO become readable or writable if the error was
+	  SSL_ERROR_WANT_READ or SSL_ERROR_WANT_WRITE. [ruby-dev:25795]
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read, ossl_ssl_write): ditto.
+
+	* ext/openssl/lib/openssl/buffering.rb
+	  (Buffering#consume_rbuf): pointless eof flag resetting is deleted.
+	  (Buffering#read): should return an empty string if the specified
+	  size is zero.
+	  (Buffering#readpartial): new method.
+	  (Buffering#readline): fix typo.
+	  (Buffering#getc): return the first character of string correctly.
+	  (Buffering#readchar): fix typo.
+	  (Buffering#eof?): should read again it the input buffer is empty.
+	  (Buffering#do_write): should rescue Errno::EAGAIN.
+	  (Buffering#puts): use "\n" as the output field separator.
+
+	* ext/openssl/extconf.rb: get rid of GNUmakefile generation.
+
+	* text/openssl/test_pair.rb: test for IO like methods.
+
+	* test/ruby/ut_eof.rb: test about empty file.
+
+Sat Mar  5 17:48:31 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (rb_glob): fixed mismatch of argument.
+
+	* dir.c (fnmatch): removed unnecessary code. (by string.c 1.219)
+
+	* win32/win32.c (NtInitialize): ditto. (by numeric.c 1.117)
+
+Sat Mar  5 16:50:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: follow nkf 1.65
+
+Sat Mar  5 16:29:26 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: freeze callback-entry objects
+
+	* ext/tk/lib/tkextlib/tile.rb: support tile-0.6
+
+Sat Mar  5 12:52:08 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): allow putting spaces between target
+	  and colon in depend file.
+
+Sat Mar  5 02:41:00 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (eaccess): workaround for VC++8 runtime.
+
+	* win32/win32.c (ioinfo): VC++8 support.
+
+Fri Mar  4 19:39:55 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser#do_includes): replace
+	  also locally defined modules.
+
+	* ext/iconv/iconv.c: rdocified.
+
+	* ext/strscan/strscan.c: moved misplaced rdoc.
+
+Fri Mar  4 16:11:20 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (rb_exec_recursive): matched the declaration to prototype.
+
+	* ext/curses/curses.c: don't need to check HAVE_WCOLOR_SET excluding
+	  window_color_set().
+
+	* ext/tk/tcltklib.c: fixed commit mistakes.
+
+Fri Mar  4 12:45:17 2005  Tilman Sauerbeck  <tilman@code-monkey.de>
+
+	* lib/rdoc/parsers/parse_c.rb: allow whitespace after function names.
+	  [ruby-core:4296]
+
+	* lib/rdoc/parsers/parse_simple.rb: adds support for private comments
+	  in the "simple" parser. [ruby-core:4301]
+
+Fri Mar  4 12:45:17 2005  Charles Mills  <cmills@freeshell.org>
+
+	* lib/rdoc/parsers/parse_c.rb: adds support for constants
+	  (rb_define_const), accessors (rb_define_attr), and makes a
+	  couple fixes. [ruby-core:4307]
+
+Fri Mar  4 12:45:17 2005  Florian Gross  <florgro@gmail.com>
+
+	* lib/rdoc/parsers/parse_rb.rb: Logic for def Builtin.method() end
+	  [ruby-core:4302]
+
+Fri Mar  4 12:45:17 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c: replace rb_protect_inspect() and rb_inspecting_p() by
+	  rb_exec_recursive() in eval.c.
+
+	* eval.c (rb_exec_recursive): new function.
+
+	* array.c (rb_ary_join): use rb_exec_recursive().
+
+	* array.c (rb_ary_inspect, rb_ary_hash): ditto.
+
+	* file.c (rb_file_join): ditto.
+
+	* hash.c (rb_hash_inspect, rb_hash_to_s, rb_hash_hash): ditto.
+
+	* io.c (rb_io_puts): ditto.
+
+	* object.c (rb_obj_inspect): ditto
+
+	* struct.c (rb_struct_inspect): ditto.
+
+Fri Mar  4 10:15:30 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/set.rb (SortedSet::setup): a hack to shut up warning.
+	  [ruby-talk:132866]
+
+Fri Mar  4 09:37:12 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (install-nodoc, pre-install-doc, post-install-doc):
+	  fix some omissions.
+
+Fri Mar  4 08:09:12 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/time.rb (Time::strptime): add new function.  inspired by
+	  [ruby-talk:132815].
+
+	* lib/parsedate.rb (ParseDate::strptime): ditto.
+
+Fri Mar  4 07:07:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: follow nkf 1.63
+
+Thu Mar  3 23:24:00 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: follow nkf 1.62
+
+Thu Mar  3 18:47:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* {bcc32,win32,wince}/Makefile.sub (config.h): check if affected
+	  when makefiles are modified.
+
+	* {bcc32,win32,wince}/Makefile.sub (config.status): add variables
+	  for tests.
+
+	* win32/ifchange.bat: try to update a file only if modified.
+
+	* win32/resource.rb: more descriptions.
+
+	* common.mk: add {pre,post}-install targets.
+
+	* instruby.rb (install?): install particular part.
+
+	* bcc32/Makefile.sub (post-install-ext): remove debug information
+	  files after installation.
+
+	* ext/tk/tcltklib.c (ip_rbUpdateCommand, ip_rb_threadUpdateCommand):
+	  get rid of warnings with Tcl/Tk 8.3 or former.
+
+Thu Mar  3 11:49:51 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/tdiary_plugin/rss-recent.rb: added site information.
+
+Wed Mar  2 19:53:44 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (parse_args): return false if nothing matched.
+
+Wed Mar  2 17:15:08 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c (lib_eventloop_core): fix typo
+
+Wed Mar  2 16:59:50 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* eval.c (ruby_native_thread_kill): call pthread_kill() to send a
+	  signal to ruby's native thread
+
+	* ruby.h: add definition of ruby_native_thread_kill()
+
+	* signal.c (sigsend_to_ruby_thread): send the signal to ruby's
+	  native thread ([ruby-dev:25744], [ruby-dev:25754]), and set
+	  signal mask to the current native thread
+
+Wed Mar  2 16:03:08 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: enforce thread-check and exception-handling to
+	  avoid SEGV trouble.
+
+	* ext/tk/tkutil/tkutil.c: fix a bug on converting a SJIS string array
+	  to a Tcl's list string.
+
+	* ext/tk/tcltklib.c: wrap Tcl's original "namespace" command to
+	  protect from namespace crash.
+
+	* ext/tk/lib/multi-tk.rb: enforce exception-handling.
+
+	* ext/tk/lib/multi-tk.rb: catch IRB_EXIT to work on irb.
+
+	* ext/tk/lib/tk.rb: ditto.
+
+	* ext/tk/tcltklib.c: add TclTkLib.mainloop_thread?
+
+	* ext/tk/lib/multi-tk.rb: (bug fix) callback returns a value.
+
+	* ext/tk/lib/tk/canvas.rb (delete): bug fix when multiple arguments.
+
+	* ext/tk/lib/clock.rb: fix 'no method error'.
+
+	* ext/tk/lib/clock.rb (self.clicks): accept a Symbol argument.
+
+	* ext/tk/lib/variable.rb: be able to set default_value_type; :numeric,
+	  :bool, :string, :symbol, :list, :numlist or nil (default; same to
+	  :string). If set a type, TkVariable#value returns a value of the
+	  type.
+
+	* ext/tk/lib/tkextlib/tclx/tclx.rb: add Tk::TclX.signal to warn the
+	  risk of using TclX extension's 'signal' command.
+
+	* ext/tk/sample/irbtk.rb: irb with Ruby/Tk.
+
+	* ext/tk/sample/demos-*/anilabel.rb: bug fix on 'show code'
+
+	* ext/tk/sample/demos-*/aniwave.rb: new Ruby/Tk animation demo.
+
+	* ext/tk/sample/demos-*/pendulum.rb: ditto.
+
+	* ext/tk/sample/demos-*/goldberg.rb: ditto.
+
+	* ext/tk/sample/demos-*/widget: add entries of animation demos.
+
+Wed Mar  2 12:21:18 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): [EXPERIMENTAL] NODE_LAMBDA implemented.
+	  [ruby-dev:25780]
+
+	* node.h (NODE_LAMBDA): for literal Proc object.
+
+	* parse.y (expr): interpret mere do...end block as proc object.
+
+	* parse.y (primary): ditto, for brace block.
+
+Tue Mar  1 21:16:54 2005  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* regcomp.c (optimize_node_left): uninitialized member
+	  (OptEnv.backrefed_status) was used. [ruby-dev:25778]
+
+Tue Mar  1 16:50:37 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regparse.c: move st_*_strend() functions from st.c.  fixed some
+	  potential memory leaks.
+
+Tue Mar  1 00:40:35 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/tuplespace.rb (Rinda::TupleSpace): improved keeper thread.
+
+	* test/rinda/test_rinda.rb: ditto.
+
+Mon Feb 28 23:10:13 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/socket/socket.c (Init_socket): IPv6 is not supported although
+	  AF_INET6 is defined on bcc32. (rev1.108 again)
+
+	* ext/socket/mkconstants.rb: ditto.
+
+Mon Feb 28 21:55:49 2005  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* ext/strscan/strscan.c (strscan_s_allocate):
+	  use onig_region_init().
+
+	* ext/strscan/strscan.c (adjust_registers_to_matched):
+	  use onig_region_set().
+
+Mon Feb 28 15:12:06 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/socket/socket.c (Init_socket): ported more Socket::Constants
+	  from ruby_1_8, and made it easy to add new constants.
+	  [ruby-dev:25771]
+
+	* ext/socket/depend: ditto.
+
+	* ext/socket/mkconstants.rb: ditto. (added)
+
+Mon Feb 28 11:42:23 2005  Ian Macdonald  <ian@caliban.org>
+
+	* exception error messages updated.  [ruby-core:04497]
+
+Mon Feb 28 09:03:09 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (Init_socket): add bunch of Socket
+	  constants.  Patch from Sam Roberts <sroberts@uniserve.com>.
+	  [ruby-core:04409]
+
+Sun Feb 27 05:55:38 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: fix typo. [ruby-core:04494]
+
+Sat Feb 26 16:58:20 2005  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* parse.y, re.c, regex.h, LEGAL, ext/strscan/strscan.c:
+	  remove oniggnu.h (GNU regex API).
+
+Wed Feb 23 22:08:16 2005  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* st.c, st.h: imported additional file changes on
+	  Oniguruma 3.7.0.
+
+Wed Feb 23 21:45:29 2005  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* ascii.c, euc_jp.c, oniggnu.h, oniguruma.h, regcomp.c,
+	  regenc.c, regenc.h, regerror.c, regexec.c, regint.h,
+	  regparse.c, regparse.h, sjis.c, utf8.c: imported Oni Guruma
+	  3.7.0.
+
+Wed Feb 23 15:04:32 2005  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/generic.rb (split_userinfo): should split ":pass" into ""
+	  and "pass".  [ruby-dev:25667]
+
+Wed Feb 23 08:00:18 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_s_create): no need for negative argc check.
+	  [ruby-core:04463]
+
+	* array.c (rb_ary_unshift_m): ditto.
+
+Wed Feb 23 01:53:29 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (initialize): handle certs correctly. Thanks,
+	  NABEYA Kenichi.
+
+Wed Feb 23 00:37:34 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/mkmf.rb (mkmf_failed): fixed typo.
+
+Tue Feb 22 23:52:45 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, lib/mkmf.rb: use simple commands if available.
+
+	* mkconfig.rb: remove autoconf internal variables from rbconfig.rb.
+
+	* lib/mkmf.rb (create_makefile): substitute implicit rules in depend
+	  file.
+
+	* {bcc32,win32,wince}/Makefile.sub (COMPILE_RULES, RULE_SUBST):
+	  include $(topdir) and $(hdrdir) to search path.
+
+Tue Feb 22 23:51:45 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c: get rid of warnings caused by a bug of VC.
+
+Tue Feb 22 23:50:26 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (ruby_signal, ruby_nativethread_signal): must be valid as
+	  expressions, not only statements.
+
+Tue Feb 22 12:54:13 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_thread_start_0): update curr_thread before raising
+	  TAG_THREAD.  [ruby-dev:25712]
+
+Tue Feb 22 07:24:57 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): identifier after dot must not be a variable.
+
+Mon Feb 21 18:31:12 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* signal.c: Standard signal handlers ignore signals on non-Ruby native
+	  threads. When a handler is entried with ruby_signal() (like as the
+	  standard signal handlers), the handler for the signal is marked as
+	  it cannot accept non-Ruby native threads. If a handler can treat all
+	  signals on all native threads, please use ruby_nativethread_signal()
+	  to entry it.
+
+Sun Feb 20 00:48:48 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::FTP#buffer_open): access mechanism
+	  re-implemented according to RFC 1738.
+	  reported by Guillaume Marcais.  [ruby-talk:131650]
+
+Sat Feb 19 18:46:56 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (DRbObject#respond_to?): take two arguments.
+	  [ruby-dev:25722]
+
+	* test/drb/drbtest.rb: ditto.
+
+Sat Feb 19 13:52:02 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb: call OpenSSL::SSL::SSLSocket#post_connection_check
+	  after connection is made.
+
+Sat Feb 19 13:31:28 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): remove no longer existing installed
+	  files.
+
+	* lib/mkmf.rb (install_dirs): return installation directory list.
+
+	* lib/mkmf.rb (create_makefile): reverted wrongly removed lines.
+
+Sat Feb 19 01:28:56 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/newton.rb: resolved LoadError.
+	  [ruby-dev:25685]
+
+	* ext/bigdecimal/sample/linear.rb: ditto.
+
+	* ext/bigdecimal/sample/nlsolve.rb: ditto.
+
+	* ext/bigdecimal/lib/bigdecimal/nlsolve.rb: removed because this file
+	  is sample script and same file exists in ext/bigdecimal/sample.
+
+Fri Feb 18 17:14:00 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/xmlrpc/parser.rb (XMLRPC::FaultException): make it subclass
+	  of StandardError class, not Exception class.  [ruby-core:04429]
+
+Fri Feb 18 04:06:41 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (fcall_gen): lvar(arg) will be evaluated as
+	  lvar.call(arg) when lvar is a defined local variable. [new]
+
+Thu Feb 17 22:15:34 2005  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* ext/strscan/strscan.c: calls Oniguruma API directly.
+
+Thu Feb 17 21:53:12 2005  K.Kosako  <sndgk393 AT ybb.ne.jp>
+
+	* common.mk, LEGAL: remove reggnu.c.
+
+Thu Feb 17 21:53:12 2005  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* gc.c, re.c: now ruby calls Oniguruma API directly, bypassing
+	  GNU compatible APIs.
+
+Thu Feb 17 20:09:23 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/drb/drb.rb (DRbServer.default_safe_level): fix typo.
+
+Thu Feb 17 20:09:23 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/digest/test_digest.rb: separate test case for each algorithms.
+	  [ruby-dev:25412]
+
+Thu Feb 17 14:31:52 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_class_initialize): call inherited method before
+	  calling initializing block.
+
+	* eval.c (rb_thread_start_1): initialize newly pushed frame.
+	  fixed: [ruby-dev:25707]
+
+Thu Feb 17 13:46:00 2005  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/collector.rb (collect_file): now deletes paths added
+	  to $LOAD_PATH instead of restoring it verbatim.
+
+	* lib/test/unit/autorunner.rb (AutoRunner.run): fixed so that
+	  'ruby -rtest/unit -rtest1 -rtest2 -e0' will use the objectspace
+	  collector again. Also tried to simplify the calling convention.
+
+	* test/runner.rb: adjusted for new AutoRunner semantics.
+
+	* lib/test/unit.rb: ditto.
+
+Thu Feb 17 04:21:47 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/open3.rb (Open3::popen3): $? should not be EXIT_FAILURE.
+	  fixed: [ruby-core:04444]
+
+Thu Feb 17 00:31:21 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/test_drb.rb, ut_safe1.rb: port from 1.8
+
+Thu Feb 17 00:02:27 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (is_defined): NODE_IASGN is an assignment.
+
+Wed Feb 16 23:54:14 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_start_1): outer block variables wasn't linked to
+	  threads.  fixed: [ruby-dev:25700]
+
+Wed Feb 16 15:11:43 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::Nonblock#initialize):
+	  native win32 platform doesn't have F_GETFL.
+
+Wed Feb 16 02:47:45 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read, ossl_ssl_write): should
+	  call rb_sys_fail instead of raising SSLError if SSL_ERROR_SYSCALL
+	  occurred.
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#fill_rbuff):
+	  should rescue Errno::EAGAIN.
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#each): fix typo.
+	  suggested by Brian Ollenberger.
+
+	* ext/openssl/lib/openssl/ssl.rb: set non-blocking flag to the
+	  underlying IO.
+
+Tue Feb 15 22:14:34 2005  sheepman  <sheepman@tcn.zaq.ne.jp>
+
+	* ext/readline/readline.c (Readline.readline): use rl_outstream
+	  and rl_instream.  [ruby-dev:25699]
+
+Mon Feb 14 23:58:17 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb (RSS::ListenerMixin::tag_end):
+	  fixed invalid namespace handling bug.
+
+Mon Feb 14 13:12:38 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb
+	  (OpenSSL::SSL::SSLSocket#post_connection_check): new method.
+
+Mon Feb 14 00:10:17 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (DRbServer): add default_safe_level, safe_level,
+	  config[:safe_level] ([druby-ja:120])
+
+	* test/drb/test_drb.rb, ut_eval.rb, ut_safe1.rb: ditto.
+
+Sun Feb 13 23:13:46 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/dublincore.rb (RSS::DublicCoreModel#date{,=}): added
+	  convenient methods.
+
+	* lib/rss/0.9.rb (RSS::Rss::Channel#date{,=}): ditto.
+
+	* lib/rss/2.0.rb (RSS::Rss::Channel::Item#date{,=}): ditto.
+
+	* test/rss/: added tests for the convenient methods.
+
+Sun Feb 13 23:12:47 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_start_0): restore prot_tag before rewinding.
+
+Sun Feb 13 16:56:52 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI.start): should set reason-phrase
+	  to the value of status header field. ([ruby-dev:40617])
+
+Sun Feb 13 11:38:40 2005  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regparse.c (type_cclass_hash): (Thanks Nobu) fixed
+	  overrun. ([ruby-dev:25676]).
+
+Sun Feb 13 10:53:08 2005  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* oniggnu.h, oniguruma.h, regcomp.c, st.c: imported
+	  Oni Guruma 3.6.0.
+
+Sun Feb 13 01:33:19 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/erb.rb (ERB::Util.h, u): make it module_function.
+
+Sat Feb 12 22:17:11 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (TAG_THREAD): to start a new thread.
+
+	* eval.c (ruby_init, ruby_options, ruby_cleanup, rb_protect,
+	  rb_load_protect, rb_thread_start_0): make thread anchor.
+
+	* eval.c (proc_alloc): clone proc object if klass is not Proc or
+	  created in different thread.
+
+	* eval.c (rb_block_pass): call a function with a block.  [new]
+
+	* eval.c (rb_f_throw): raise NameError in main thread.
+
+Sat Feb 12 17:29:19 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI.open_loop): send authentication only for
+	  the URI directly specified.
+
+Sat Feb 12 15:07:23 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rand_init): suppress warning.
+
+Sat Feb 12 14:10:24 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI.open_http): reject userinfo explicitly.
+
+Sat Feb 12 13:54:03 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb: support https if the platform provides CA
+	  certificates.
+
+Sat Feb 12 06:18:28 2005  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
+
+	* ext/etc/etc.c (Init_etc): sGroup needs HAVE_ST_GR_PASSWD check
+	  [ruby-dev:25675]
+
+Fri Feb 11 17:37:50 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_set_default_paths):
+	  new method OpenSSL::X509::Store#set_default_paths.
+
+Fri Feb 11 11:33:53 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::HTTP#proxy_open): new option supported:
+	  :http_basic_authentication.
+	  suggested by Kent Sibilev.  [ruby-core:4392]
+
+Fri Feb 11 06:30:07 2005  George Ogata  <g_ogata@optushome.com.au>
+
+	* misc/ruby-mode.el: ignore parenthesis inside heredoc.
+	  [ruby-core:04415]
+
+Fri Feb 11 04:54:13 2005  Tilman Sauerbeck  <tilman@code-monkey.de>
+
+	* lib/rdoc/generators/html_generator.rb: [ruby-core:04412]
+
+	* lib/rdoc/generators/ri_generator.rb: ditto.
+
+Thu Feb 10 13:52:42 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in, win32/Makefile.sub (LIBS, COMMON_HEADERS): use
+	  winsock2 on mswin32/mingw.
+
+	* ext/socket/extconf.rb: ditto.
+
+	* win32/win32.c (StartSockets): ditto.
+
+	* win32/win32.h: ditto.
+
+Thu Feb 10 12:09:16 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): default to true if not compiled
+	  previously.
+
+	* ext/extmk.rb (extmake): create dummy makefile if extconf failed.
+
+Thu Feb 10 12:07:10 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (init_stdhandle): assign standard file handles.
+
+	* bcc32/Makefile.sub (COMMON_LIBS): add libraries included in
+	  import32.lib.
+
+	* lib/mkmf.rb (create_makefile): restrict prefixing with srcdir to
+	  rule lines, add search path to implicit rules, and set Borland make
+	  special macros for search path.
+
+	* win32/win32.c, win32/win32.h (read): avoid a BCC runtime bug.
+
+Thu Feb 10 00:47:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (make_struct): fixed: [ruby-core:04402]
+
+Wed Feb  9 16:33:05 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (wait_connectable): fixed wrong condition.
+
+Wed Feb  9 14:42:28 2005  URABE Shyouhei  <shyouhei@ice.uec.ac.jp>
+
+	* eval.c (scope_dup): add volatile not to optimize tbl.
+
+Wed Feb  9 10:02:02 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/tk/make-tkutil, ext/tk/tkutil/subconf.rb: no longer used.
+
+	* ext/tk/tkutil/extconf.rb: need to compile tkutil. [ruby-dev:25607]
+
+Wed Feb  9 08:07:08 2005  Paul Duncan  <pabs@pablotron.org>
+
+	* ext/curses/curses.c (window_color_set): [ruby-core:04393]
+
+Tue Feb  8 23:48:36 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: reject :instance_eval, :class_eval, :module_eval
+	  [druby-ja:117]
+
+Tue Feb  8 22:38:28 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* keywords, parse.y: separate EXPR_VALUE from EXPR_BEG.
+	  fixed: [ruby-core:04310], [ruby-core:04368]
+
+Tue Feb  8 13:06:12 2005  Sam Roberts  <sroberts@uniserve.com>
+
+	* ext/socket/socket.c (Init_socket): SO_REUSEPORT added.
+	  [ruby-talk:130092]
+
+Tue Feb  8 00:19:02 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb (Resolv::DNS::Name#subdomain_of?): new method.
+	  (Resolv::DNS::Name#inspect): ditto.
+	  Suggested by Sam Roberts.  [ruby-talk:129086]
+
+Mon Feb  7 23:14:11 2005  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (io_getc): flush rb_stdout before read from stdin, which is
+	  connected to a tty.  [ruby-core:4378]
+
+	* rubyio.h (FMODE_TTY): renamed from FMODE_LINEBUF.
+
+Mon Feb  7 10:06:30 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c: [ruby-doc:818]
+
+Mon Feb  7 02:13:05 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb (sockaddr_storage): winsock2.h have the
+	  definition of struct sockaddr_storage, but socket.c doesn't
+	  include it because this version of ruby still has binary level
+	  compatibility with winsock1.
+
+Mon Feb  7 01:22:50 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): extract previously collected
+	  informations from existing Makefile.
+
+	* ext/socket/extconf.rb: check if getaddrinfo() works fine only when
+	  wide-getaddrinfo option is not given.  fixed: [ruby-dev:25422]
+
+	* ext/tk/extconf.rb: separate tkutil configuration.
+
+	* lib/mkmf.rb ($extmk): check if under ext directory.
+
+	* lib/mkmf.rb (Logging.postpone): allow recursive operation.
+
+	* lib/mkmf.rb (try_constant): make sure if really a constant, reduce
+	  the number of times of compile.
+
+	* lib/mkmf.rb (have_macro, have_var, byte_order): new functions.
+
+	* lib/mkmf.rb (find_library): allow directory list with separators.
+
+	* lib/mkmf.rb (arg_config): manage provided configuration options.
+
+	* lib/mkmf.rb (dir_config): accept arrays of directory names as
+	  default values.
+
+	* lib/mkmf.rb (with_cppflags, with_cflags, with_ldflags): keep flags
+	  modified if the block returned true.
+
+Sun Feb  6 19:20:05 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (stack_extend): add prototype because VC++8 doesn't
+	  accept __declspec(noinline) with K&R style function definitions.
+
+Sun Feb  6 13:56:19 2005  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (new_with_hash): changed messages of exception.
+
+	* lib/date/format.rb (str[fp]time): undocumented conversion
+	  specifications %[1-3] are now deprecated.
+
+Sun Feb  6 11:27:37 2005  Tanaka Akira  <akr@m17n.org>
+
+	* ext/dl/dl.c (Init_dl): function declaration should precede
+	  statements before C99.
+
+Sun Feb  6 03:24:20 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb (Resolv::DNS::Resource::TXT): multiple strings was not
+	  handled.
+	  (Resolv::DNS::Resource::TXT#strings): new method to return all
+	  strings.
+	  (Resolv::DNS::Message::MessageEncoder#put_string_list): new method.
+	  (Resolv::DNS::Message::MessageDecoder#get_string_list): ditto.
+	  based on [ruby-talk:129732] by Sam Roberts.
+
+Sat Feb  5 02:24:06 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ripper/test_scanner_events.rb: fix test.
+
+Fri Feb  4 18:44:35 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/lib/ripper/lexer.rb: last Lexer fix was incomplete;
+	  test all green.
+
+Fri Feb  4 15:57:06 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (open_args): fix too verbose warnings for the space
+	  before argument parentheses.  [ruby-dev:25492]
+
+	* parse.y (parser_yylex): ditto.
+
+Fri Feb  4 14:33:25 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/lib/ripper/filter.rb: ripper/tokenizer ->
+	  ripper/lexer. [ruby-dev:25632]
+
+Fri Feb  4 00:24:15 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss: supported Image module.
+	  http://web.resource.org/rss/1.0/modules/image/
+
+Thu Feb  3 23:42:36 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_close, strio_close_read, strio_close_write):
+	  should return nil instead of self as well as IO.  [ruby-dev:25623]
+
+	* ext/stringio/stringio.c (strio_extend, strio_putc): fill with zero
+	  extended portion.  [ruby-dev:25626]
+
+Thu Feb  3 16:12:57 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (parser_yylex): the first expression in the parentheses
+	  should not be a command.  [ruby-dev:25492]
+
+Thu Feb  3 03:31:20 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: follow original v 1.57
+
+	* ext/nkf/nkf-utf8/utf8tbl.c: follow original v 1.8
+
+	* ext/nkf/nkf-utf8/config.h: follow original v 1.7
+
+Wed Feb  2 23:52:53 2005  sheepman  <sheepman@tcn.zaq.ne.jp>
+
+	* ext/stringio/stringio.c (strio_truncate): should MEMZERO an extended
+	  part.  [ruby-dev:25618]
+
+Wed Feb  2 21:56:01 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::Element#convert): added.
+
+	* lib/rss/rss.rb: convert -> need_convert.
+
+	* lib/rss/1.0.rb: ditto.
+
+	* lib/rss/0.9.rb: ditto.
+
+	* lib/rss/2.0.rb: ditto.
+
+	* lib/rss/trackback.rb: ditto.
+
+Wed Feb  2 03:30:58 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/lib/ripper/tokenizer.rb -> lexer.rb.
+
+	* ext/ripper/lib/ripper/lexer.rb: new method Ripper.slice.
+	  [experimental]
+
+	* ext/ripper/lib/ripper/sexp.rb: new file. [experimental]
+
+	* ext/ripper/lib/ripper.rb: require ripper/lexer and ripper/sexp.
+
+Tue Feb  1 21:49:24 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (DRb::DRbObject#respond_to?): check marshal_dump and
+	  _dump.
+
+Tue Feb  1 00:20:23 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, configure.in: made EXTOUT configurable.
+
+	* ext/extmk.rb (extmake), lib/mkmf.rb: keep topdir as relative style.
+
+	* lib/mkmf.rb: make extensions in depth order.  [ruby-dev:25522]
+
+	* configure.in (aix): fix linker flags on AIX.  [ruby-talk:125460]
+
+Mon Jan 31 13:16:39 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/extconf.rb: add tkutil configuration step (remove old schema)
+
+	* ext/tk/depend: remove the information of tkutil
+
+	* ext/tk/make-tkutil: sub-part of Makefile to compile tkutil
+
+	* ext/tk/tkutil/tkutil.c: move tkutil.c to subdirectory
+
+	* ext/tk/tkutil/subconf.rb: configuration file for tkutil.c
+
+	* ext/tk/tkutil/depend: ditto
+
+Mon Jan 31 13:13:35 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tcltklib.c: add invalid namespace check
+
+	* ext/tk/lib/multi-tk.rb: add invalid_namespace? method
+
+	* ext/tk/lib/remote-tk.rb: ditto
+
+Mon Jan 31 10:29:18 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/context.rb (IRB::Context::initialize): [ruby-core:04330]
+
+Mon Jan 31 09:44:03 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (Init_Object): remove Object#type.  [ruby-core:04335]
+
+Sat Jan 29 09:42:12 2005  Sam Roberts  <sroberts@uniserve.com>
+
+	* lib/resolv.rb (Resolv::DNS::Resource::IN::SRV): Added RFC2782 SRV
+	  resource record for specifying location of services.
+
+Sat Jan 29 00:10:33 2005  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* ascii.c, euc_jp.c, hash.c, oniggnu.h, oniguruma.h, regcomp.c,
+	  regenc.c, regenc.h, regerror.c, regexec.c, reggnu.c, regint.h,
+	  regparse.c, regparse.h, sjis.c, st.c, st.h, utf8.c: imported
+	  Oni Guruma 3.5.4.
+
+Fri Jan 28 17:16:55 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb (Resolv::DNS::Config.parse_resolv_conf):
+	  parse options line for ndots option.
+	  (Resolv::Hosts#lazy_initialize): return self.
+	  (Resolv::DNS#lazy_initialize): ditto.
+	  (Resolv::DNS::Config#lazy_initialize): ditto.
+	  Suggested by Sam Roberts.
+
+Thu Jan 27 17:15:03 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/tk/extconf.rb: support new tk scheme on bccwin32.
+	  fixed: [ruby-dev:25546]
+
+Thu Jan 27 13:18:03 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* st.c (st_foreach): report success/failure by return value.
+	  [ruby-Bugs-1396]
+
+Thu Jan 27 00:12:19 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb (setup): support BSD style
+	  directory group inheritance (again).
+
+Thu Jan 27 00:02:40 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb (setup): support BSD style
+	  directory group inheritance. [ruby-dev:25440]
+
+	* test/fileutils/fileasserts.rb (assert_same_entry): show entry
+	  difference.
+
+Wed Jan 26 17:12:50 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: forgot to initialize parser struct.  [ruby-dev:25492]
+
+	* parse.y (parser_yylex): no tLABEL on EXPR_BEG.
+	  [ruby-talk:127711]
+
+Wed Jan 26 14:12:58 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/Setup*: remove tcltklib.
+
+Wed Jan 26 12:45:16 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/tk/extconf.rb: support new tk scheme on mswin32.
+	 fixed: [ruby-dev:25535]
+
+Wed Jan 26 10:45:19 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (flock_winnt, flock_win95): unlock file even if
+	  LOCK_NB is specified.
+
+Tue Jan 25 23:10:48 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk: merge tcltklib for Ruby/Tk installation control
+
+	* ext/tcltklib: remove
+
+Tue Jan 25 17:05:15 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (proc_options): correct -T option in RUBYOPT.
+	  fixed: [ruby-dev:25512]
+
+Tue Jan 25 14:05:52 2005  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: fix SEGV bug; trouble on canceling remained
+	  after scripts [ruby-dev:25479]: NULL current namespace when deleting
+	    Tk interpreter [ruby-talk:126225]
+
+	* ext/tcltklib/extconf.rb: bug fix; TCL_ENABLE_THREAD flag is inverted
+	  [ruby-talk:126360]
+
+	* ext/tcltklib/extconf.rb: add yet another native-thread check
+
+	* ext/tk/tkutil.c: fix SEGV bug; NULL string pointer when finalize
+	  Ruby interpreter
+
+	* ext/tk/lib/multi-tk.rb: avoid warning for deleted safeTk ip frame
+
+	* ext/tk/lib/tk/bindtag.rb: bug fix; new method of named bindtag
+	  doesn't return the created object [ruby-dev:25479]
+
+	* ext/tk/lib/tk/menu.rb: bug on treating arguments [ruby-dev:25479]
+
+	* ext/tk/lib/tk.rb: bug fix; cannot accept a callback ID string for
+	  a command argument [ruby-dev:25479]
+
+	* ext/tk/lib/multi-tk.rb: ditto
+
+	* ext/tk/lib/tk/*.rb: ditto
+
+	* ext/tk/lib/tkextlib/*.rb: ditto
+
+	* ext/tk/sample/demos-jp/anilabel.rb: new demo script
+
+	* ext/tk/sample/demos-en/anilabel.rb: ditto
+
+	* ext/tk/sample/tkHTML/ss.rb: local variable scope bug fix
+	  [ruby-dev:25479]
+
+Mon Jan 24 16:00:53 2005  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb (guess_old): not use NKF.guess_old
+	  but NKF.guess1. fixed: [ruby-dev:25491]
+
+Mon Jan 24 15:44:25 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* document updates - [ruby-core:04296], [ruby-core:04301],
+	  [ruby-core:04302], [ruby-core:04307]
+
+Sun Jan 23 12:38:01 2005  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/wsdlDriver.rb: from 1.5.3-ruby1.8.2, operation which has
+	  capitalized name (such as KeywordSearchRequest in AWS) is defined as
+	  a method having uncapitalized name. (converted with
+	  GenSupport.safemethodname to handle operation name 'foo-bar').  it
+	  introduced serious incompatibility; in the past, it was defined as a
+	  capitalized.
+
+	  define capitalized method as well under that circumstance.
+
+Sun Jan 23 05:24:42 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_to_der): should call
+	  GetOCSPReq at first.
+
+Sat Jan 22 22:59:08 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/ssl.rb (accept): rescue SSLError. [druby-ja:110]
+
+Sat Jan 22 22:27:28 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/unix.rb: fail if UNIXFileOwner is set. [druby-ja:111]
+
+Fri Jan 21 20:07:02 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb (Resolv::DNS::Config.resolv): don't raise ResolvError.
+	  reported by Sam Roberts.  [ruby-talk:127133]
+
+Fri Jan 21 17:09:44 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (decode_utf7): use pack("U*") to encode UTF-8.
+
+	* lib/net/imap.rb (encode_utf7): use unpack("U*") to decode UTF-8.
+
+	* test/net/imap/test_imap.rb: added tests for Net::IMAP.
+
+Fri Jan 21 16:58:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (rb_push_glob): should work for NUL delimited patterns.
+
+	* dir.c (rb_glob2): should aware of offset in the pattern.
+
+Fri Jan 21 13:58:37 2005  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (u8tou16): fixed typo. fixed: [ruby-list:40546]
+
+Fri Jan 21 00:37:09 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/syck/rubyext.c (syck_parser_bufsize_set): avoid VC++ warning
+	  "local variable 'size' used without having been initialized".
+
+Thu Jan 20 11:42:02 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_new4): should propagate taintedness.
+
+	* env.h: rename member names in struct FRAME; last_func -> callee,
+	  orig_func -> this_func, last_class -> this_class.
+
+	* struct.c (rb_struct_set): use original method name, not callee
+	  name, to retrieve member slot.  [ruby-core:04268]
+
+	* time.c (time_strftime): protect from format modification from GC
+	  finalizers.
+
+Thu Jan 20 02:01:10 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (Init_Object): remove rb_obj_id_obsolete()
+
+Wed Jan 19 18:02:19 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/ipaddr.rb (to_s, test_to_s): too many colons with some cases.
+
+Wed Jan 19 01:16:30 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb (Resolv::DNS::Config.parse_resolv_conf): ignore
+	  domain and search directive without an argument.
+	  reported by Sam Roberts.  [ruby-talk:126781]
+
+Mon Jan 17 23:33:46 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (aix): fix typo.  [ruby-talk:126401]
+
+Mon Jan 17 07:08:51 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c: suppress warnings.
+
+	* lib/irb/extend-command.rb (IRB::ContextExtender.def_extend_command):
+	  ditto.
+
+	* lib/irb/ext/history.rb (IRB::Context::set_last_value): ditto.
+
+	* lib/irb/ext/history.rb (IRB::Context::eval_history): ditto.
+
+	* lib/irb/locale.rb (IRB::Locale::real_load): ditto.
+
+	* lib/irb/slex.rb (SLex::Node::create_subnode): remove garbage.
+
+Mon Jan 17 00:09:42 2005  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/uri/common.rb (PORT): typo fix. fixed: [ruby-core:04256]
+
+Sat Jan 15 14:57:22 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): ignore trailing CRs at the end of short
+	  options as well as long options.  fixed: [ruby-core:04232]
+
+Sat Jan 15 13:44:22 2005  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::VERSION): 0.1.2 -> 0.1.3.
+
+	* lib/rss/rss.rb: accept inheritance. [ruby-talk:126104]
+
+Wed Jan 12 12:29:28 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_mod_define_method): incomplete subclass check.
+	  [ruby-dev:25464]
+
+	* class.c (rb_make_metaclass): class of metaclasses should be
+	  plain Class.  [ruby-list:40524]
+
+Tue Jan 11 20:58:52 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (remain_size): use buffered data instead of unreading to avoid
+	  inconsistency of text mode.  fixed: [ruby-dev:25446]
+
+Tue Jan 11 09:37:53 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* numeric.c (Init_Numeric): turn off floating point exceptions
+	  on bcc32. "1e300".to_f had crashed by overflow.
+
+Mon Jan 10 15:28:51 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI::Socket#request_line): should
+	  escape SCRIPT_NAME and PATH_INFO before being parsed as a URI.
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils#escape_path): add
+	  new method to escape URI path component.
+
+	* lib/webrick/ssl.rb (WEBrick::Config::SSL): the default value
+	  of :SSLEnable is false.
+
+	* test/webrick/{test_cgi.rb,webrick.cgi}: new file.
+
+	* test/webrick/utils.rb: require "webrick/https.h".
+
+Mon Jan 10 01:22:55 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (rb_data_object_alloc): klass may be NULL.
+	  [ruby-list:40498]
+
+Sun Jan  9 14:12:17 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_f_select): IO list could be altered.  [ruby-dev:25312]
+
+Sun Jan  9 04:08:40 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* test/webrick/test_server.rb (test_daemon): delete an assertion
+	  which has possibility to fail by race condition.
+
+Sun Jan  9 03:22:46 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb (test_copy_entry): copy_entry
+	  copies only file type, not mtime. [ruby-dev:25383]
+
+Sat Jan  8 04:38:47 2005  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml.rb: Kernel#y requires an argument.
+
+Fri Jan  7 21:12:29 2005  TAMURA Takashi  <sheepman@tcn.zaq.ne.jp>
+
+	* random.c (rand_init): use ALLOC_N instead of ALLOCA_N
+	  [ruby-dev:25426]
+
+Fri Jan  7 20:01:31 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_service):
+	  should delete trailing LF from the result of pack("m*").
+
+	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_connect):
+	  - should delete trailing LF from the result of pack("m*").
+	  - clear Request-Line not to send the response by HTTPServer#run.
+
+	* lib/webrick/httputils (WEBrick::HTTPUtils.parse_qvalues):
+	  refine regexp (and change the name of a local variable).
+
+	* lib/webrick/server.rb (WEBrick::Daemon.start): prepared stdio
+	  don't allow changing its mode.
+
+	* test/webrick/*, sample/webrick/httpproxy.rb: add new files.
+
+Fri Jan  7 18:03:35 2005  Tanaka Akira  <akr@m17n.org>
+
+	* gc.c (mark_locations_array): avoid core dump with -O3.
+	  [ruby-dev:25424]
+
+Thu Jan  6 20:29:18 2005  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/zlib/zlib.c (zstream_end): should return value.
+
+Thu Jan  6 19:59:03 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.c (rb_w32_close): didn't close socket handle.
+	  [ruby-dev:25414]
+
+	* win32/win32.c (rb_w32_open_osfhandle): bcc32's _open_osfhandle
+	  never set EMFILE.
+
+Thu Jan  6 17:22:41 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* random.c (random_seed): O_NONBLOCK isn't defined on some
+	  platforms. [ruby-dev:25417]
+
+Thu Jan  6 13:45:35 2005  Tanaka Akira  <akr@m17n.org>
+
+	* lib/time.rb: recognize +00:00 and GMT as a localtime.
+
+Thu Jan  6 07:58:28 2005  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/usage.rb (RDoc::RDoc.usage_no_exit): Allow for colons
+	  in path names on DOS machines. (thanks to Johan Nilsson)
+
+Thu Jan  6 00:02:35 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb: use MockClock.sleep instead of Kernel.sleep
+	  [ruby-dev:25387]
+
+Wed Jan  5 20:16:32 2005  Tanaka Akira  <akr@m17n.org>
+
+	* random.c (limited_big_rand): didn't work if SIZEOF_BDIGITS == 2.
+	  [ruby-dev:25408]
+
+	* random.c (random_seed): refined.
+
+Wed Jan  5 16:39:54 2005  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* parse.y (BITSTACK_POP): workaround for bcc32 compiler's bug.
+	  shift assignment operator '>>=' for __int64 in struct may
+	  generate collapsed code. [ruby-dev:25342]
+
+	* win32/win32.[ch]: failed to compile on bcc32 (and probably wince)
+	  [ruby-dev:25306]
+
+Wed Jan  5 12:49:39 2005  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_initialize): Thread objects cannot be initialized
+	  again.  fixed: [ruby-core:04067]
+
+Wed Jan  5 02:30:11 2005  Tanaka Akira  <akr@m17n.org>
+
+	* random.c (init_by_array): imported from mt19937ar-cok.tgz.
+	  (genrand_int32): ditto.
+	  (genrand_real): replaced with genrand_res53 in mt19937ar-cok.
+	  (rand_init): support bignum for longer seed.
+	  (random_seed): generate longer seed.
+	  (make_mask): new function.
+	  (limited_rand): ditto.
+	  (limited_big_rand): ditto.
+	  (rb_f_rand): call limited_rand and limited_big_rand.
+	  [ruby-dev:25403]
+
+Tue Jan  4 23:25:29 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_rand): should return positive random number.
+	  [ruby-dev:25401]
+
+Tue Jan  4 21:25:43 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/{test_drbssl.rb,test_drbunix.rb,ut_drb.rb}: use
+	  DRbService.ext_service. reduce sleep.
+
+Mon Jan  3 14:01:54 2005  Tanaka Akira  <akr@m17n.org>
+
+	* random.c (random_seed): don't use /dev/urandom if it is not
+	  character device.
+
+Mon Jan  3 11:37:42 2005  Tanaka Akira  <akr@m17n.org>
+
+	* random.c (random_seed): use /dev/urandom if available.
+	  [ruby-dev:25392]
+
+Tue Jan  4 11:15:29 2005  TAMURA Takashi  <sheepman@tcn.zaq.ne.jp>
+
+	* bignum.c (rb_big_rand): do not use rb_big_modulo to generate
+	  random bignums.  [ruby-dev:25396]
+
+Mon Jan  3 11:03:37 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/test_drb.rb: move TestDRbReusePort to new file.
+	  [ruby-dev:25238]
+
+	* test/drb/drbtest.rb: change timeout.
+
+	* test/drb/ignore_test_drb.rb: new file.
+
+Mon Jan  3 07:27:46 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpauth/htpasswd.rb (WEBrick::Htpasswd#reload):
+	  raise NotImplementedError if password is encrypted by digest
+	  algorithms. This patch is contributed by sheepman. [ruby-list:40467]
+
+	* lib/webrick/httpauth/digestauth.rb
+	  (WEBrick::HTTPAuth::DigestAuth#_authenticate): fix digest calculation.
+	  This patch is contributed by sheepman. [ruby-list:40482]
+
+	* lib/webrick/{httpauth.rb,httpauth/basicauth.rb,httpproxy.rb}: use
+	  pack/unpack-template char "m" instead of lib/base64.rb to do base64
+	  encoding/decoding. fixed: [ruby-dev:25336]
+
+	* test/webrick/test_httpauth.rb: new file.
+
+Sun Jan  2 15:42:10 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: add lazy stop_service.
+
+	* lib/drb/extserv.rb: ditto.
+
+Sun Jan  2 01:17:17 2005  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/drbtest.rb: add method DRbService.ext_service.
+
+	* test/drb/test_drb.rb: ditto.
+
+	* test/drb/test_drbssl.rb: ditto.
+
+Sat Jan  1 20:23:02 2005  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (argf_readpartial): new method ARGF.readpartial.
+	  (io_getpartial): extracted from io_readpartial.
+	  (io_readpartial): call io_getpartial.
+
+Sat Jan  1 17:44:54 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (each_capitalized): should join header field
+	  value.  This patch is contributed sheepman [ruby-list:40478]
+
+	* test/net/http/test_httpheader.rb: test it.
+
+Sat Jan  1 16:21:29 2005  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (copy_stream): use read/write instead of
+	  sysread/syswrite, which allows duck typing. [ruby-dev:25369]
+
+	* lib/fileutils.rb (copy_stream): does NOT support nonblocking IO.
+	  [ruby-dev:25370]
+
+	* lib/fileutils.rb (copy_entry): could not copy symlink.
+
+	* test/fileutils/test_fileutils.rb: test copy_entry, copy_file,
+	  copy_stream.
+
+Sat Jan  1 04:20:23 2005  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ns_spki.c (ossl_spki_set_challenge): should call
+	  StringValue before GetSPKI. fixed: [ruby-dev:25359].
+
+Sat Jan  1 01:13:28 2005  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_autoload): [ruby-dev:25373]
+
+Fri Dec 31 14:10:43 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_formatter.rb (RI::TextFormatter::display_flow_item):
+	  Fix problem if heading contains formatting.
+
+Fri Dec 31 00:08:02 2004  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in (HAVE_RLIM_T): removed because not used.
+
+Thu Dec 30 22:45:39 2004  Tanaka Akira  <akr@m17n.org>
+
+	* rubyio.h: don't deprecate rb_read_check.
+
+	* io.c (STDIO_READ_DATA_PENDING): reverted from old READ_DATA_PENDING
+	  to check stdio read buffer.
+	  (rb_read_check): use STDIO_READ_DATA_PENDING.
+	  (rb_read_pending): ditto.
+	  (rb_getc): ditto.
+
+Thu Dec 30 05:39:35 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y: eliminate unused members in struct parser_params.
+	  [ruby-dev:25258] (again)
+
+	* parse.y: make parser_new() static.
+
+Thu Dec 30 00:41:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (svalue_to_avalue): [ruby-dev:25366]
+
+	* string.c (rb_str_justify): [ruby-dev:25367]
+
+Wed Dec 29 11:07:07 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/template/html/kilmer.rb: Update to use new
+	  sections.
+
+Tue Dec 28 22:31:46 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_justify): create buffer string after argument type
+	  conversion.  fixed: [ruby-dev:25341]
+
+Tue Dec 28 17:18:17 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/net/telnet.rb (preprocess): remove NULL unless binmode.
+	  fixed: [ruby-list:40320]
+
+Tue Dec 28 15:41:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (reinit): should initialize all static
+	  variables.  fixed: [ruby-list:40445]
+
+Tue Dec 28 15:25:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/nkf/lib/kconv.rb (Kconv::RegexpEucjp): second byte is up to
+	  0xfe.
+
+	* ext/nkf/lib/kconv.rb (Kconv#kconv): should handle UTF8 and UTF16
+	  properly.
+
+Tue Dec 28 13:35:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (rb_deflate_s_deflate, rb_inflate_s_inflate): ensure
+	  freeing internal zstreams.  fixed: [ruby-dev:25309]
+
+	* ext/zlib/zlib.c (rb_deflate_init_copy): replace rb_deflate_clone.
+
+Mon Dec 27 20:02:14 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: fix SEGV bug when deleting Tk interp
+
+	* ext/tk/lib/multi-tk.rb: ditto
+
+Mon Dec 27 16:54:05 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509name.c (Init_ossl_x509name): should use
+	  rb_hash_new to get exactly a Hash. fix [ruby-dev:25325].
+
+Mon Dec 27 15:29:12 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb (cp_r): tested wrong file name.
+	  [ruby-dev:25339]
+
+Mon Dec 27 15:15:18 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mv): should raise error when moving a
+	  directory to the (empty) directory. [ruby-talk:124368]
+
+	* lib/fileutils.rb (mv): wrongly did not overwrite file on Win32
+	  platforms.
+
+Mon Dec 27 14:36:20 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (NUM2RLIM, RLIM2NUM): Without SIZEOF_RLIM_T is not error.
+	  fixed: [ruby-dev:25346]
+
+Sun Dec 26 16:21:39 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (Net::IMAP::PlainAuthenticator): added a new class
+	  to support the PLAIN authentication mechanism. Thanks, Benjamin
+	  Stiglitz.
+
+Sat Dec 25 01:28:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_select): [ruby-dev:25312]
+
+Fri Dec 24 23:27:18 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/image.rb: TkPhotoImage#cget bug fix
+
+Fri Dec 24 03:06:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_reopen, rb_io_reopen): prohibit to change access mode for
+	  special IO ports.  [ruby-dev:25225]
+
+	* io.c (next_argv): reduce use of stdio.
+
+Fri Dec 24 02:22:53 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_loader_transfer): check type conversion.
+
+	* ext/syck/rubyext.c (syck_parser_assign_io, rb_new_syck_node): duck
+	  typing.
+
+	* ext/syck/rubyext.c (syck_parser_s_alloc, syck_parser_initialize):
+	  allocation framework.
+
+	* ext/syck/rubyext.c (syck_emitter_s_alloc, syck_emitter_initialize):
+	  ditto.
+
+Fri Dec 24 01:21:00 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkextlib/blt.rb: add BLT extension support
+
+	* ext/tk/lib/tkextlib/blt/*.rb: ditto
+
+	* ext/tk/lib/tkextlib/blt/tile/*.rb: ditto
+
+Thu Dec 23 23:43:24 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_setgroups): check if the argument length is
+	  modified.  fixed: [ruby-dev:25285]
+
+	* process.c (SIZEOF_RLIM_T): err if size of rlim_t is not set.
+
+Thu Dec 23 19:08:41 2004  Tanaka Akira  <akr@m17n.org>
+
+	* rubyio.h: rename FMODE_UNSEEKABLE to FMODE_DUPLEX.
+
+	* io.c (io_check_tty): extracted function to set FMODE_LINEBUF and
+	  FMODE_DUPLEX.
+
+Thu Dec 23 13:13:33 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: define TclTkLib::COMPILE_INFO and
+	  RELEASE_DATE
+
+	* ext/tcltklib/extconf.rb: ditto
+
+	* ext/tk/tkutil.c: define TkUtil::RELEASE_DATE
+
+	* ext/tk/lib/tk.rb: define Tk::RELEASE_DATE
+
+Thu Dec 23 00:16:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (bsdi): use $(CC) for LDSHARED.  fixed [ruby-dev:25270]
+
+Wed Dec 22 11:14:55 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_mode_modenum): replace O_ACCMODE with O_RDWR.
+	  fixed: [ruby-dev:25273]
+
+Wed Dec 22 08:34:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/sym.c (rb_dlsym_initialize): extract internal pointers after
+	  all argument conversion.  fixed: [ruby-dev:25271]
+
+Tue Dec 21 16:15:21 2004  Michael Neumann  <mneumann@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb: use "" instead of "." if prefix argument is
+	  nil in proxy methods. nil is default value.
+
+	* test/xmlrpc/test_webrick_server.rb, test/xmlrpc/webrick_testing.rb:
+	  use threads instead of forking. this should fix issue #1208
+	  (http://rubyforge.org/tracker/?func=detail&atid=1698&aid=1208&group_id=426).
+	  removed testing of SSL enabled servlet as this hangs.
+
+Wed Dec 22 00:05:10 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/*, test/soap/*, sample/soap/authheader/*: eval cleanup.
+
+Tue Dec 21 22:07:41 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_decode_all): use rb_str_new4
+	  to avoid SEGV.
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_traverse, ossl_asn1_decode,
+	  ossl_asn1_decode_all): temporary value should be marked volatile.
+
+Tue Dec 21 12:42:34 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_traverse, ossl_asn1_decode):
+	  use rb_str_new4 to avoid SEGV. fix [ruby-dev:25261]
+
+	* test/openssl/test_asn1.rb: add tests for OpenSSL::ASN1.
+
+Tue Dec 21 12:10:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/grid.rb: rescue bug of 'grid configure' on Tcl/Tk8.3-
+
+Mon Dec 20 22:52:29 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* added samples for the previous soap4r's commit.
+
+Mon Dec 20 22:56:39 2004  Tanaka Akira  <akr@m17n.org>
+
+	* gc.c (set_stack_end): gcc noinline attribute is available since
+	  gcc-3.1.
+
+Mon Dec 20 22:40:31 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* added files:
+	  * lib/soap/mapping/wsdl*.rb
+	  * lib/wsdl/soap/element.rb
+	  * lib/wsdl/xmlSchema/simpleContent.rb
+
+	* modified files:
+	  * lib/soap/*
+	  * lib/wsdl/*
+	  * lib/xsd/*
+	  * test/soap/*
+	  * test/wsdl/*
+	  * test/xsd/*
+
+	* summary
+	  * imported from the soap4r repository.  Version: 1.5.3-ruby1.8.2
+
+	  * added several XSD basetype support: nonPositiveInteger,
+	    negativeInteger, nonNegativeInteger, unsignedLong, unsignedInt,
+	    unsignedShort, unsignedByte, positiveInteger
+
+	  * HTTP client connection/send/receive timeout support.
+
+	  * HTTP client/server gzipped content encoding support.
+
+	  * improved WSDL schema definition support; still is far from
+	    complete, but is making step by step improvement.
+
+Mon Dec 20 14:45:19 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/net/https.rb: delete descriptions about key_file and cert_file.
+	  fixed: [ruby-dev:25243]
+
+	* ext/openssl/lib/net/telnets.rb: ditto.
+
+Mon Dec 20 14:07:02 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: supports new features of Tcl/Tk8.5a2
+
+	* ext/tk/lib/tk/clock.rb: ditto
+
+	* ext/tk/lib/tk/text.rb: ditto
+
+	* ext/tk/lib/tk/panedwindow.rb: ditto
+
+Mon Dec 20 13:51:40 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_getservbyport): [ruby-talk:124072]
+
+Mon Dec 20 10:51:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (special_local_set): prevent the parser object from GC.
+	  fixed: [ruby-dev:25252]
+
+Mon Dec 20 03:30:40 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session#initialize): empty session id was
+	  used if request had no session key.  fixed: [ruby-core:03981]
+
+Mon Dec 20 01:51:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (make_struct): [ruby-dev:25249]
+
+Mon Dec 20 00:16:54 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/encodings/SHIFT_JIS.rb: fixed LoadError bug.
+	  [ruby-core:3958]
+
+Sun Dec 19 17:24:59 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (enable_rpath): use rpath flag to embed the library
+	  path into extensions on ELF environment.  [ruby-dev:25035]
+
+Sun Dec 19 11:01:25 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb: use standalone runner for -e.
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner#options): accept
+	  multiple -p and -x options.
+
+	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir#recursive_collect):
+	  ditto.
+
+Sat Dec 18 16:36:23 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (rb_deflate_s_deflate, rb_inflate_s_inflate):
+	  disallow interrupt by type conversion.  fixed: [ruby-dev:25226]
+
+Sat Dec 18 15:09:02 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/webrick/httpauth.rb,
+	  lib/webrick/httpauth/{basicauth.rb,digestauth.rb}: use
+	  pack/unpack-template char "m" instead of lib/base64.rb to do base64
+	  encoding/decoding.
+
+Sat Dec 18 10:51:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_open_dir): new function.  [ruby-dev:25242]
+
+	* io.c (rb_f_open): add type check for return value from to_open.
+
+Fri Dec 17 16:44:26 2004  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in (ac_cv_sizeof_rlim_t): set 8 for BSD/OS.
+	  Reported by OHARA Shigeki.  [ruby-dev:25236]
+
+Fri Dec 17 16:28:12 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fix bug on setting up system encoding
+
+	* ext/tk/lib/tk/event.rb: fix error on require process
+
+	* ext/tk/lib/tk/font.rb: fix abnormal termination error on Windows
+
+	* ext/tk/lib/tk/virtevent.rb: TkVirtualEvent::PreDefVirtEvent.new()
+	  accepts event-sequence arguments
+
+	* ext/tk/lib/tk/text.rb: fail to dump embedded images
+
+	* ext/tk/lib/tk/text.rb: tag_nextrange and tag_prevrange returns wrong
+	  types of values
+
+	* ext/tk/lib/tk/texttag.rb: nextrange and prevrange returns wrong
+	  types of values
+
+	* ext/tk/lib/tk/text.rb: add TkText::IndexModMethods module and
+	  TkText::IndexString class to treat text index modifiers
+
+	* ext/tk/lib/tk/texttag.rb: use TkText::IndexModMethods module
+
+	* ext/tk/lib/tk/textmark.rb: ditto
+
+	* ext/tk/lib/tk/textimage.rb: ditto
+
+	* ext/tk/lib/tk/textwindow.rb: ditto
+
+	* ext/tk/lib/tk/textimage.rb: wrong gravity of text mark for embedded
+	  image
+
+	* ext/tk/lib/tk/textwindow.rb: wrong gravity of text mark for
+	  embedded window
+
+Fri Dec 17 13:33:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session#initialize): control adding
+	  session_id hidden fields.  fixed: [ruby-talk:123850]
+
+Fri Dec 17 00:01:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_proc_arity, rb_node_arity, rb_mod_method_arity,
+	  rb_obj_method_arity): new functions to obtain method arity.
+	  [ruby-dev:25143]
+
+Thu Dec 16 23:31:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): create RUBYARCHDIR also when no extension
+	  is installed.  fixed: [ruby-dev:25215]
+
+Thu Dec 16 22:36:57 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/test_drb.rb: adjust and reduce sleep.
+
+Thu Dec 16 18:37:08 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.c (ossl_raise): refine message format.
+
+Thu Dec 16 16:29:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/demos-en/widget: modify version check for
+	  supporting features
+
+Thu Dec 16 16:03:50 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/bindtag.rb: bug fix [ruby-talk: 123667]
+
+	* ext/tk/lib/tk/timer.rb: accept :idle for the interval argument
+
+	* ext/tk/lib/tk.rb: add TkComm._callback_entry?()
+
+	* ext/tk/lib/multi-tk.rb: add MultiTkIp.cb_entry_class
+
+	* ext/tk/lib/tk/canvas.rb: use TkComm._callback_entry?()
+
+	* ext/tk/lib/tk/canvastag.rb: ditto
+
+	* ext/tk/lib/tk/dialog.rb: ditto
+
+	* ext/tk/lib/tk/optiondb.rb: ditto
+
+	* ext/tk/lib/tk/text.rb: ditto
+
+	* ext/tk/lib/tk/texttag.rb: ditto
+
+	* ext/tk/lib/tk/textwindow.rb: ditto
+
+	* ext/tk/lib/tk/timer.rb: ditto
+
+	* ext/tk/lib/tk/validation.rb: ditto
+
+	* ext/tk/lib/tkextlib/*: ditto
+
+Thu Dec 16 04:02:28 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/extconf.rb: bison is not needed if ripper.c exists.
+	  [ruby-dev:25191]
+
+Thu Dec 16 03:27:10 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: remove junk.
+
+Thu Dec 16 00:57:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/syck/rubyext.c (rb_syck_io_str_read): [ruby-core:03973]
+
+Thu Dec 16 00:43:29 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: changed default binded address family to use an
+	  available address family of host name. [druby-ja:101]
+
+	* lib/drb/ssl.rb: ditto
+
+Wed Dec 15 17:47:17 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#start_thread):
+	  should log about all accepted socket. [ruby-core:03962]
+
+	* lib/webrick/accesslog.rb (WEBrick::AccessLog#setup_params):
+	  "%%" and "%u" are supported. [webricken:135]
+
+	* lib/webrick/httpservlet/filehandler.rb
+	  (WEBrick::HTTPServlet::FileHandler#check_filename):
+	  :NondisclosureName is acceptable if it is Enumerable.
+
+	* lib/webrick/config.rb (WEBrick::Config::FileHandler):
+	  default value of :NondisclosureName is [".ht*", "*~"].
+
+Wed Dec 15 16:10:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_obj_id_obsolete): warn always.
+
+Wed Dec 15 15:31:02 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/set.rb (Set#==): [ruby-dev:25206]
+
+Wed Dec 15 14:32:18 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_fdisset): check whether the handle is valid.
+
+Wed Dec 15 10:30:37 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/openssl/ossl_digest.c (ossl_digest_initialize): [ruby-dev:25198]
+
+Tue Dec 14 19:17:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* utf8.c (utf8_is_mbc_ambiguous): [ruby-talk:123561]
+
+	* utf8.c (utf8_mbc_to_normalize): ditto.
+
+Tue Dec 14 17:08:15 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_close): need to reset osfhnd().
+
+Tue Dec 14 14:03:57 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.c (ossl_raise): avoid buffer overrun.
+	  [ruby-dev:25187]
+
+Tue Dec 14 12:36:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session::initialize): generate new
+	  session if given session_id does not exist.  [ruby-list:40368]
+
+Tue Dec 14 08:47:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_eval): should mark ruby_eval_tree.  [ruby-dev:25189]
+
+Mon Dec 13 18:13:52 2004  Tanaka Akira  <akr@m17n.org>
+
+	* gc.c (set_stack_end): new function to obtain stack end address.
+	  set_stack_end obtains a stack end address by an address of local
+	  variable in the function.
+	  (SET_STACK_END, STACK_END): use set_stack_end.  don't use alloca.
+	  This makes the conservative garbage collector to scan a stack frame
+	  of the garbage_collect function itself.  This is required because
+	  callee-save registers may be stored in the frame.
+	  [ruby-dev:25158]
+
+Mon Dec 13 02:45:51 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/curses/curses.c (window_subwin): call NUM2INT() before
+	  GetWINDOW(). fixed: [ruby-dev:25161]
+
+Mon Dec 13 00:58:02 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (cleanpath_aggressive): make it private.
+	  (cleanpath_conservative): ditto.
+	  Suggested by Daniel Berger.  [ruby-core:3914]
+
+Sun Dec 12 21:32:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_super.rb (TestSuper#test_define_method): now methods
+	  from procs can call super.
+
+Sun Dec 12 10:35:10 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/template/html/html.rb (RDoc::Page): Don't
+	  show an accessor's r/w flag if none was specified
+
+Sun Dec 12 10:14:03 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc::parse_files): Never exclude files
+	  explicitly given on the command line.
+
+Sat Dec 11 21:10:16 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: add DRbRemoteError. [ruby-list:40348],
+	  [ruby-list:40390]
+
+	* test/drb/drbtest.rb: ditto.
+
+	* test/drb/ut_drb.rb: ditto.
+
+Sat Dec 11 13:08:28 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/optparse/subcommand.rb: a sample for sub commands like
+	  cvs.  contributed by Minero Aoki.
+
+Fri Dec 10 08:39:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_listen): get OpenFile just before calling
+	  listen(2).
+
+Thu Dec  9 16:28:35 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/sdbm/init.c (GetDBM): typo.
+
+Thu Dec  9 16:21:51 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI#setup_header): avoid
+	  SecurityError. [ruby-dev:24970]
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): should wait
+	  for reading request till data arrive. [ruby-talk:121068]
+
+Thu Dec  9 14:38:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_inspect): escape # which starts an expression
+	  substitution.  fixed: [ruby-core:03922]
+
+	* string.c (rb_str_dump): not escape # which isn't a substitution.
+
+Thu Dec  9 12:31:53 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* io.c (pipe_open): should set prog if argc != 0.
+
+Thu Dec  9 10:54:36 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/dbm/dbm.c (fdbm_select): [ruby-dev:25132]
+
+	* ext/sdbm/init.c: ditto.
+
+	* ext/gdbm/gdbm.c: ditto.
+
+Thu Dec  9 10:19:18 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/socket/test_socket.rb (test_setsockopt): use SO_LINGER instead
+	  of SO_BINDTODEVICE.  fixed: [ruby-dev:25133]
+
+Thu Dec  9 03:08:36 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (ip_init): set root-win title to "ruby" when
+	  the running script is '-e one-liner' or '-' (stdin).
+
+	* ext/tcltklib/extconf.rb: add find_library("#{lib}#{ver}",..) for
+	  stub libs
+
+	* ext/tk/lib/tk/textmark.rb: TkTextMarkCurrent and TkTextMarkAnchor
+	  have a wrong parent class.
+
+	* ext/tk/lib/tk/dialog.rb: rename TkDialog2 --> TkDialogObj and
+	  TkWarning2 --> TkWarningObj (old names are changed to alias names)
+
+	* ext/tk/lib/tk/dialog.rb: bug fix of treatment of 'prev_command'
+	  option and hashes for configuration
+
+	* ext/tk/lib/tk/dialog.rb: add TkDialogObj#name to return the
+	  button name
+
+	* ext/tk/lib/tk/radiobutton.rb: rename enbugged method value() ==>
+	  get_value() and value=(val) ==> set_value(val).
+
+	* ext/tk/lib/tk/menu.rb: add TkMenu.new_menuspec
+
+	* ext/tk/lib/tk/menu.rb: add alias (TkMenuButton = TkMenubutton,
+	  TkOptionMenuButton = TkOptionMenubutton)
+
+	* ext/tk/lib/tk/event.rb: new method aliases (same as option keys of
+	  event_generate) for Event object
+
+	* ext/tk/lib/tk/font.rb: configinfo returns proper types of values
+
+	* ext/tk/lib/tk.rb: bind methods accept subst_args + block
+
+	* ext/tk/lib/tk/canvas.rb: ditto
+
+	* ext/tk/lib/tk/canvastag.rb: ditto
+
+	* ext/tk/lib/tk/frame.rb: ditto
+
+	* ext/tk/lib/tk/text.rb: ditto
+
+	* ext/tk/lib/tk/texttag.rb: ditto
+
+	* ext/tk/lib/tk/toplevel.rb: ditto
+
+	* ext/tk/lib/tkextlib/*: ditto and bug fix
+
+Wed Dec  8 23:54:29 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/template/html/html.rb (RDoc::Page): Typo
+	  meant that h2 tag was invisible.
+
+Wed Dec  8 22:10:02 2004  Tanaka Akira  <akr@m17n.org>
+
+	* rubyio.h, io.c, ext/dl/dl.c, ext/pty/pty.c, ext/socket/socket.c:
+	  create FILE object only when required: popen(3) and DL's IO#to_ptr.
+	  [ruby-dev:25122]
+
+	* io.c (rb_io_binmode): use setmode for Human68k.  [ruby-dev:25121]
+
+Wed Dec  8 20:13:06 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* process.c (rb_spawn): support for DJGPP.
+
+	* lib/mkmf.rb (VPATH): specify the implicit path separator for DJGPP.
+
+Wed Dec  8 17:48:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): merge Guy Decoux's argument preserve
+	  patch in [ruby-core:03874].
+
+Wed Dec  8 17:37:33 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pipe_exec): need to close original socket
+	  handle.
+
+Wed Dec  8 14:31:36 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): [ruby-dev:25104]
+
+Wed Dec  8 13:49:46 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pipe_exec): must close original handle
+	  before exec. fixed: [ruby-dev:25112]
+
+Wed Dec  8 11:46:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (string_content): get rid of segfault at empty evstr.
+	  fixed: [ruby-dev:25113]
+
+Wed Dec  8 03:26:51 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_bio.c (ossl_obj2bio): should not use fptr->f.
+	  [ruby-dev:25101]
+
+Wed Dec  8 03:26:41 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* runruby.rb: prepend LIBRUBY_SO to LD_PRELOAD as well as rubytest.rb.
+
+Wed Dec  8 01:35:44 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (is_socket): reorder of function definitions.
+
+Wed Dec  8 00:44:31 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (pipe_open): win32 bidirectional pipe support.
+
+	* win32/win32.[ch] (rb_w32_pipe_exec): ditto.
+
+	* win32/win32.[ch] (socketpair): new function. POSIX socketpair
+	  emulation.
+
+	* win32/win32.c (socketpair_internal): ditto.
+
+Wed Dec  8 00:25:07 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_version.rb: added version check test.
+	  [ruby-dev:25053]
+
+Tue Dec  7 15:40:38 2004  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (io_fwrite): avoid context switch before writing to stderr.
+	  [ruby-dev:25080]
+
+	* rubyio.h: refine deprecated declaration.
+
+	* configure.in, file.c, io.c: remove useless check: fseeko, etc.
+
+Tue Dec  7 13:42:07 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* dir.c (dir_s_mkdir): win32 special processing doesn't need any
+	  longer.
+
+	* win32/win32.[ch] (rb_w32_mkdir): new function. POSIX.1 compatible
+	  interface.
+
+	* win32/win32.[ch] (rb_w32_rmdir): new function.
+
+Tue Dec  7 00:27:37 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (proc_setgroups): [ruby-dev:25081]
+
+Mon Dec  6 23:07:57 2004  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in: check -lsocket for socketpair and shutdown.
+	  reported by Ville Mattila.  [ruby-core:03903]
+
+Mon Dec  6 23:00:45 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (ac_cv_sizeof_rlim_t): setup for DJGPP.
+
+	* io.c (is_socket, shutdown): define dummy macros for DJGPP.
+
+	* process.c: use SIZEOF_RLIM_T instead of HAVE_RLIM_T for DJGPP.
+
+Mon Dec  6 21:19:40 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (is_socket): fix typos. [ruby-core:03900]
+
+Mon Dec  6 20:13:28 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (is_socket): new function.
+
+	* io.c (rb_io_close_read, rb_io_close_write): use is_socket().
+
+	* io.c (rb_io_fptr_finalize): need to check fptr->f before calling
+	  rb_io_fptr_cleanup().
+
+	* io.c (pipe_open): win32 pipe support (experimental).
+
+	* win32/win32.[ch] (rb_w32_pipe_exec): return file descriptors
+	  instead of FILE structure objects.
+
+	* win32/win32.[ch] (rb_w32_is_socket): new function.
+
+Mon Dec  6 19:40:40 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* Makefile.in (.y.c): simplify the rule.
+
+Mon Dec  6 18:08:10 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_eqq): document fix.  [ruby-talk:122541]
+
+Mon Dec  6 17:49:30 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (run_trap_eval): add prototype for Microsoft compiler.
+
+Mon Dec  6 17:32:38 2004  Tanaka Akira  <akr@m17n.org>
+
+	* rubyio.h, intern.h, io.c, file.c, process.c, ext/socket/socket.c,
+	  ext/pty/pty.c, ext/io/wait/wait.c, ext/openssl/ossl_ssl.c:
+	  Use own buffering mechanism instead of stdio.  [ruby-dev:25056]
+
+	* io.c, ext/stringio/stringio.c, test/ruby/ut_eof.rb:
+	  EOF flag removed.
+
+Mon Dec  6 17:15:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/socket/test_socket.rb (TestBasicSocket#test_setsockopt):
+	  BasicSocket#setsockopt dumps core.  [ruby-dev:25039]
+
+	* test/socket/test_tcp.rb (TestTCPSocket#test_recvfrom):
+	  TCPSocket#recvfrom dumps core.  [ruby-dev:24705]
+
+	* test/socket/test_udp.rb (TestUDPSocket#test_connect):
+	  UDPSocket#connect dumps core.  [ruby-dev:25045]
+
+	* test/socket/test_udp.rb (TestUDPSocket#test_bind):
+	  UDPSocket#bind dumps core.  [ruby-dev:25057]
+
+Mon Dec  6 09:59:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_fread): take VALUE argument.
+
+	* ext/socket/socket.c (sock_connect): use rb_str_new4().
+	  [ruby-dev:25052]
+
+	* eval.c (rb_yield_0): [ruby-dev:25051]
+
+Mon Dec  6 01:32:31 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_public_encrypt,
+	  ossl_rsa_public_decrypt, ossl_rsa_private_encrypt,
+	  ossl_rsa_private_decrypt): should take an optional argument
+	  to specify padding mode. [ruby-talk:122539]
+
+	* ext/openssl/ossl_pkey_rsa.c (Init_ossl_rsa): add new constants
+	  PKCS1_PADDING, SSLV23_PADDING, NO_PADDING and PKCS1_OAEP_PADDING
+	  under OpenSSL::PKey::RSA.
+
+	* test/openssl/test_pkey_rsa.rb: new file.
+
+Sun Dec  5 19:39:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Completion#complete): new parameter
+	  to direct case insensitiveness.
+
+	* lib/optparse.rb (OptionParser#order!): ignore case only for long
+	  option.  [ruby-dev:25048]
+
+Sun Dec  5 00:54:32 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* mkconfig.rb: setup library paths before requiring library.
+	  [ruby-core:03892]
+
+Sat Dec  4 22:54:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_write): remove rb_str_locktmp().  [ruby-dev:25050]
+
+	* io.c (io_fwrite): takes VALUE string as an argument.
+	  [ruby-dev:25050]
+
+	* ext/socket/socket.c (sock_connect): remove rb_str_locktmp().
+	  [ruby-dev:25050]
+
+	* ext/socket/socket.c (udp_connect): [ruby-dev:25045]
+
+	* ext/socket/socket.c (udp_bind): ditto.
+
+	* ext/socket/socket.c (udp_send): ditto.
+
+	* ext/socket/socket.c (bsock_send): ditto.
+
+	* ext/socket/socket.c (s_recvfrom): ditto.
+
+	* hash.c (rb_hash_hash): should provide "hash" method where "eql?"
+	  is redefined.  [ruby-talk:122482]
+
+Sat Dec  4 21:29:05 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb: (In previous commit) new method chown.
+
+	* lib/fileutils.rb: (In previous commit) new method chown_R.
+
+	* lib/fileutils.rb: (In previous commit) new method chmod_R
+	  wrongly added.  Removed now.
+
+Sat Dec  4 20:45:52 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mkdir, mkdir_p): should chmod explicitly.
+	  [ruby-core:03881]
+
+Sat Dec  4 18:54:09 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: removed empty lines from output.
+
+Sat Dec  4 18:49:09 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::VERSION): 0.1.1 -> 0.1.2
+
+	* lib/rss/rss.rb: #item=/#set_item and so on are obsolete.
+
+Sat Dec  4 14:28:56 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/code_objects.rb (RDoc::Context::Section::set_comment):
+	  Section comments may now be bracketed by lines which are
+	  ignored. You can now write
+	      # -----------
+	      # :section: Dave's Section
+	      # comment material
+	      # -----------
+	   The lines before :section: are removed, and identical lines at the end are
+	   also removed if present.
+
+Sat Dec  4 00:35:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (bsock_setsockopt): [ruby-dev:25039]
+
+Fri Dec  3 12:25:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.h: fix prototype for C++.
+
+Fri Dec  3 01:55:24 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: widget configuration by TkWindow#method_missing
+	  returns proper object. "widget.option = val" returns val, and
+	  "widget.option(val)" returns self.
+
+	* ext/tk/lib/tk/font.rb: TkFont#replace accepts only one font argument.
+
+	* ext/tk/lib/tk/radiobutton.rb: add TkRadiobutton#value and
+	  TkRadiobutton#value=(val).
+
+	* ext/tk/lib/tk/spinbox.rb: callback substitution support on
+	  command option.
+
+	* ext/tk/sample/demos-en/widget: bug fix (wrong image height)
+
+	* ext/tk/sample/demos-jp/widget: ditto.
+
+Fri Dec  3 00:21:05 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): prepare to pass a block from "call" method
+	  to a Proc generated by Method#to_proc.  [ruby-dev:25031]
+
+	* eval.c (rb_yield_0): actually passes a block given to "call".
+
+	* object.c (convert_type): use rb_respond_to() again.  this fix is
+	  based on [ruby-dev:25021]
+
+	* eval.c (rb_respond_to): funcall respond_to? if it's redefined.
+	  [ruby-dev:25021]
+
+Thu Dec  2 15:13:53 2004  Michael Neumann  <mneumann@ruby-lang.org>
+
+	* test/xmlrpc/test_parser.rb, test/xmlrpc/data/*.expected: Expected
+	  values are now stored in YAML instead of using #inspect. This fixes
+	  false hash order.
+
+Fri Dec  3 00:11:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_file_initialize): [ruby-dev:25032]
+
+Thu Dec  2 16:41:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_protect): prevent continuations created inside from being
+	  called from the outside.  [ruby-dev:25003]
+
+	* eval.c (rb_callcc, rb_cont_call): prohibit calling from different
+	  signal contexts.  [ruby-dev:25022]
+
+Thu Dec  2 10:45:02 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/readline/test_readline.rb: fix for NetBSD.
+
+Thu Dec  2 09:57:24 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/ostruct.rb (OpenStruct::Marshaler): OpenStruct can be
+	  marshaled again.  [ruby-core:03862]
+
+Thu Dec  2 09:30:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (thread_mark): mark thread group.  [ruby-dev:25020]
+
+	* eval.c (thgroup_add): check whether the argument is really a Thread.
+
+Thu Dec  2 07:57:16 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_ctl): [ruby-dev:25019]
+
+Wed Dec  1 06:13:00 2004  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c: add constant NKF::VERSION
+
+	* ext/nkf/nkf.c(guess): this becomes an alias of guess2
+
+	* ext/nkf/test.rb: add --no-cp932
+
+	* ext/nkf/nkf-utf8/nkf.c: original nkf2 revision 1.47
+
+Wed Dec  1 02:21:02 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* signal.c (sighandler): call handler immediately only for default
+	  handlers.  [ruby-dev:25003]
+
+Tue Nov 30 23:49:12 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): errno should be preserved for rb_sys_fail() when
+	  fork failed.
+
+Tue Nov 30 16:18:50 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_fread): need not to null terminate.  [ruby-dev:24998]
+
+	* eval.c (rb_eval): should check previous frame for ZSUPER.
+
+	* io.c (read_all): remove unnecessary rb_str_resize().
+	  [ruby-dev:24996]
+
+	* io.c (io_readpartial): ditto.
+
+	* io.c (io_read): ditto.
+
+Tue Nov 30 14:58:33 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* instruby.rb (install): add arguments explicitly to "super".
+
+Tue Nov 30 00:49:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (PUSH_FRAME): flags should have been initialized.
+
+	* eval.c (rb_eval): [ruby-core:03856]
+
+	* io.c (rb_io_sysread): use temporary lock.  [ruby-dev:24992]
+
+Tue Nov 30 00:12:57 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regparse.c: now handles many alternatives (over 500000)
+	  in regexp. [ruby-dev:24773]
+
+Mon Nov 29 16:06:04 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_write): insufficiently filled string
+	  being extended when overwriting.  [ruby-core:03836]
+
+Mon Nov 29 15:59:05 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/ostruct.rb (OpenStruct::method_missing): check method
+	  duplication for -d.
+
+	* lib/ostruct.rb (OpenStruct::initialize): ditto.
+
+Mon Nov 29 15:22:28 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/io/nonblock/test_flush.rb: abandon tests when io/nonblock is
+	  not supported.
+
+Mon Nov 29 13:37:54 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (fptr_finalize): must not use FILE after fclose().
+	  [ruby-dev:24985]
+
+Mon Nov 29 13:13:13 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): push back the last space before next
+	  loop because CharNext() eats it.
+
+Mon Nov 29 03:08:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (convert_type): [ruby-core:03845]
+
+	* eval.c (rb_funcall_rescue): new function.
+
+	* object.c (rb_Array): avoid using rb_respond_to().
+
+	* object.c (rb_Integer): ditto.
+
+	* eval.c (get_backtrace): no conversion for nil.
+
+	* parse.y (reduce_nodes): empty body should return nil.
+
+Mon Nov 29 01:18:18 2004  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (rb_io_check_writable): call io_seek regardless of
+	  NEED_IO_SEEK_BETWEEN_RW.  [ruby-dev:24986]
+
+Sun Nov 28 15:57:58 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::Element#tag): not use block_given? for
+	  working with ruby 1.6 again.
+
+	* lib/rss/{0.9,2.0,trackback}.rb, lib/rss/maker/base.rb:
+	  undef -> remove_method for working with ruby 1.6 again.
+
+Sun Nov 28 15:51:40 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::NotSetError): added.
+
+	* lib/rss/maker/{1.0,0.9,2.0}.rb: changed RSS Maker to raise
+	  RSS::NotSetError if required values of maker.channel are not
+	  set. [ruby-talk:120061]
+
+	* test/rss/test_maker_{1.0,0.9,2.0}.rb: changed tests to check RSS
+	  Maker raises or not.
+
+Sun Nov 28 12:14:47 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regparse.c (fetch_token): fixed test failure on HP-UX ia64
+	  ([ruby-dev:24859]).
+
+Sun Nov 28 12:08:15 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regparse.c, test/ruby/test_regexp.rb: fixed problem with UTF-8
+	  characters that have U+00FE or invalid characters.
+
+Sun Nov 28 12:07:04 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regexec.c, test/ruby/test_regexp.rb: fixed segmentation fault
+	  ([ruby-dev:24887]).
+
+Sun Nov 28 12:05:48 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regcomp.c, regint.h: fixed PLATFORM_UNALIGNED_WORD_ACCESS
+	  problem ([ruby-dev:24802] and [ruby-core:3733])
+
+Sat Nov 27 23:43:39 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_initialize): [ruby-dev:24972]
+
+Sat Nov 27 21:43:39 2004  Tanaka Akira  <akr@m17n.org>
+
+	* io.c: avoid data lost with nonblocking fd and
+	  stdio buffering in sync mode.  [ruby-dev:24966]
+	  based on matz's patch [ruby-dev:24967]
+	  (io_fwrite): new primitive writing function which writes
+	  directly if sync mode.
+	  (rb_io_fwrite): wrapper for io_fwrite now.
+	  (io_write): call io_fwrite instead of rb_io_fwrite.
+
+Sat Nov 27 17:43:21 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/{0.9,1.0,2.0,trackback,xml-stylesheet}.rb: added
+	  #setup_maker.
+
+	* test/rss/test_setup_maker_*.rb: added tests for #setup_maker.
+
+	* lib/rss/maker/base.rb(RSS::Maker::Items#max_size=): supported
+	  output item size limitation.
+
+	* sample/rss/blend.rb: added sample for RSS Maker.
+
+Sat Nov 27 17:41:35 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/0.9.rb: supported RSS::Maker.make("0.91"). Now,
+	  "0.9" is just alias of "0.91."
+
+	* test/rss/test_maker_0.9.rb: make("0.9") -> maker("0.91").
+
+	* test/rss/test_to_s.rb: ditto.
+
+Sat Nov 27 17:21:30 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/list_description.rb: untabified.
+
+	* sample/rss/rss_recent.rb: ditto.
+
+Sat Nov 27 14:44:15 2004  Kent Sibilev  <ksibilev@bellsouth.net>
+
+	* lib/cgi/session.rb (CGI::Session::initialize): [ruby-core:03832]
+
+Sat Nov 27 09:41:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_fread): old rb_io_fread with file closing checking.
+	  (rb_io_fread): wrapper for io_fread now.
+	  [ruby-dev:24964]
+
+Fri Nov 26 18:02:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: Tk.destroy uses TkWindow#epath
+
+	* ext/tk/lib/tk/image.rb: bug fix
+
+	* ext/tk/lib/tk/wm.rb: add 'iconphoto' method(Windows only)
+
+	* ext/tk/lib/tkextlib/*: some methods uses TkWindow#epath
+
+Fri Nov 26 14:29:39 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_initialize): uninitialized fd was checked to see open
+	  mode.  [ruby-dev:24963]
+
+	* io.c (rb_io_initialize): uninitialized fd was used.  [ruby-dev:24962]
+
+Fri Nov 26 13:49:06 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_initialize): should retrieve flags from copying file
+	  descriptor. [ruby-dev:24961]
+
+	* eval.c (method_missing): raise TypeError for classes do not
+	  have allocators.  [ruby-core:03752]
+
+	* lib/erb.rb: add RDoc by James Edward Gray II.  [ruby-core:03786]
+
+Fri Nov 26 13:29:02 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::look_for_directives_in): Break
+	  out of preprocessing when we find a :section: directive (previously cleared out the
+	  comment, but this apparently now generates an error in gsub!)
+
+Fri Nov 26 00:17:40 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_read): move StringValue() check before GetOpenFile().
+	  [ruby-dev:24959]
+
+Thu Nov 25 20:14:57 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/thwait.rb (ThreadsWait#join_nowait): abnormally terminated
+	  threads should be also processed.  [ruby-talk:121320]
+
+Thu Nov 25 18:06:37 2004  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in: AC_CHECK_SIZEOF(rlim_t) to include stdio.h to fix
+	  problem with autoconf 2.52 or earlier.
+	  revert AC_PREREQ to 2.50.
+	  [ruby-core:3809]
+
+Thu Nov 25 07:59:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: AC_PREREQ(2.53) [ruby-core:03800]
+
+	* io.c (read_all): stringify non-nil buffer argument, and always
+	  taint the result.  [ruby-dev:24955]
+
+Wed Nov 24 01:01:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_read): integer conversion should be prior to
+	  GetOpenFile().  [ruby-dev:24952]
+
+	* configure.in, io.c: cancel [ ruby-Patches-1074 ].
+
+Tue Nov 23 08:09:50 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/menu.rb: improve usability of TkOptionMenubutton
+
+Tue Nov 23 02:00:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_file_chown): integer conversion should be prior to
+	  GetOpenFile().  [ruby-dev:24949]
+
+Tue Nov 23 00:10:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_file_chown): integer conversion should be prior to
+	  GetOpenFile().  [ruby-dev:24947]
+
+	* file.c (rb_file_truncate): ditto.
+
+	* file.c (rb_file_s_truncate): ditto.
+
+	* dir.c (dir_seek): use NUM2OFFT().
+
+	* misc/ruby-mode.el (ruby-non-block-do-re): [ruby-core:03719]
+
+Mon Nov 22 22:33:02 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::parse_require): Don't use names
+	  of variables or constants when parsing 'require'
+
+Mon Nov 22 00:13:35 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_seek): should retrieve dir_data after NUM2INT().
+	  [ruby-dev:24941]
+
+Sat Nov 20 23:57:33 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/README (et al): Add a new directive, :section:, and
+	  change the output format to accommodate. :section: allows to to
+	  group together methods, attributes, constants, etc under
+	  headings in the output. If used, a table of contents is
+	  generated.
+
+Sat Nov 20 23:56:54 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/options.rb (Options::parse): Force --inline-source if
+	  --one-file option given
+
+Sat Nov 20 23:55:19 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_splice): should place index wrapping after
+	  possible modification.  [ruby-dev:24940]
+
+Sat Nov 20 23:25:12 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* io.c (rb_io_getline): f.gets("") did not work. [ruby-core:03771]
+
+	* test/ruby/test_io.rb (test_gets_rs): test it.
+
+Sat Nov 20 22:55:09 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* test/runner.rb (CROSS_COMPILING): need to require rbconfig.rb before
+	  using CROSS_COMPILNG.
+
+Sat Nov 20 20:42:42 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/depend: fix ripper.o dependency.
+
+Sat Nov 20 17:48:29 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* io.c (io_reopen): work around problem with Cygwin fseeko
+	  returning ESPIPE.
+
+Sat Nov 20 05:34:24 2004  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: original nkf.c rev:1.40
+
+	* ext/nkf/test.rb: add test for mime encode/decode
+
+Sat Nov 20 01:45:04 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* test/xmlrpc/test_webrick_server.rb: move `requrie "webrick/https"'
+	  into #setup_http_server method to avoid soap test errors.
+
+Sat Nov 20 01:37:34 2004  Johan Holmberg  <holmberg@iar.se>
+
+	* eval.c (error_print): nicer traceback at interrupt.
+	  [ruby-core:03774]
+
+Sat Nov 20 00:07:16 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): internal buffer should not be listed by
+	  ObjectSpace.each_object() by String#gsub.  [ruby-dev:24931]
+
+Fri Nov 19 22:44:43 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb: better support for -p/-x option.
+
+Fri Nov 19 17:46:56 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/0.9.rb (RSS::Maker::RSS09::Image#have_required_values):
+	  changed /rss/channel/image to be optional. [ruby-Bugs:1047]
+
+	* test/rss/test_maker_0.9.rb: added tests for the above.
+
+Fri Nov 19 17:18:17 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::VERSION): 0.1.0 -> 0.1.1.
+
+	* lib/rss: #to_s used #tag.
+
+	* test/rss/test_to_s.rb: added.
+
+	* lib/rss/maker.rb (RSS::Maker.make): changed API. It's not
+	  received modules which is used as the second argument.
+
+	* lib/rss/xml-stylesheet.rb (RSS::XMLStyleSheet#alternate):
+	  changed return value type which is not String but Boolean.
+
+	* lib/rss/2.0.rb (RSS::Rss::Channel#ttl): changed return value
+	  type which is not String but Integer.
+
+	* lib/rss/0.9.rb (RSS::Rss::Channel): <skipDays> has <day>s and
+	  <skipHours> has <hour>s.
+
+	* lib/rss/maker/0.9.rb (RSS::Maker::RSS09::Channel): ditto.
+
+	* lib/rss/0.9.rb (RSS::Rss::Channel::Item): <item> has <category>s.
+
+	* lib/rss/maker/2.0.rb (RSS::Maker::Rss20::Channel::Item): ditto.
+
+	* lib/rss/2.0.rb (RSS::Rss::Channel): <channel> has <category>s.
+
+	* lib/rss/maker/2.0.rb (RSS::Maker::RSS20::Channel): ditto.
+
+	* lib/rss/trackback.rb: parent element has <trackback:about>s.
+
+	* lib/rss/maker/trackback.rb: ditto.
+
+Fri Nov 19 11:10:16 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb: add support for directory name
+	  with -p/-x options.
+
+	* test/testunit/collector/test_dir.rb: ditto.
+
+	* lib/xmlrpc/datetime.rb (XMLRPC::DateTime#==): should use Array()
+	  instead of to_a.
+
+Fri Nov 19 10:32:36 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_s_set_completion_append_character):
+	  accept nil.  [ruby-core:03765]
+
+Fri Nov 19 01:20:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session::FileStore::initialize): raise
+	  exception if data corresponding to session specified from the
+	  client does not exist.
+
+Fri Nov 19 00:59:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): internal buffer should not be listed by
+	  ObjectSpace.each_object().  [ruby-dev:24919]
+
+Thu Nov 18 23:42:36 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/depend: Never regenerate lib/ripper/core.rb
+	  automatically. [ruby-dev:24911]
+
+Thu Nov 18 20:47:24 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.[ch] (rb_w32_isatty): new function to replace MSVCRT's
+	  isatty because it never sets errno.
+
+Thu Nov 18 18:41:08 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_stringchar.rb (test_bang): added.
+
+	* string.c (rb_str_upcase_bang, rb_str_capitalize_bang)
+	  (rb_str_swapcase_bang): missing rb_str_modify().
+
+Thu Nov 18 17:05:01 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_rest_arg): store rest args into invisible local variable
+	  in order to get rid of SEGV at ZSUPER.  [ruby-dev:24913]
+
+Thu Nov 18 15:39:52 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_f_getenv): prohibit for $SAFE=4. [ruby-dev:24908]
+
+Thu Nov 18 14:58:42 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/readline.c: check $SAFE.
+
+	* test/readline/test_readline.rb: added tests for readline.
+
+Thu Nov 18 00:21:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (proc_getpgrp): prohibit for $SAFE=2.
+	  [ruby-dev:24899]
+
+	* process.c (get_pid): ditto.  [ruby-dev:24904]
+
+	* process.c (get_ppid): ditto.
+
+	* array.c (rb_ary_delete): defer rb_ary_modify() until actual
+	  modification.  [ruby-dev:24901]
+
+Thu Nov 18 10:10:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c, rubyio.h (rb_io_modenum_flags): exported.
+
+	* ext/stringio/stringio.c (strio_initialize): allow Fixnum as mode as
+	  well as IO.new does.  [ruby-dev:24896]
+
+Wed Nov 17 23:47:30 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_settracefunc.rb: added.  [ruby-dev:24884]
+
+Wed Nov 17 18:59:16 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* process.c (proc_getrlimit, proc_setrlimit): add rb_secure(2) to
+	  methods of Process.{getrlimit,setrlimit}
+
+Wed Nov 17 13:56:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (newline_node): should not use FL_SET. [ruby-dev:24874]
+
+	* parse.y (string_content): should not use FL_UNSET.
+
+	* node.h (NODE_NEWLINE): remove unused bit to utilize flag field
+	  in nodes.
+
+Wed Nov 17 13:05:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (rb_push_glob): fix overrun.  [ruby-dev:24886]
+
+Wed Nov 17 11:48:17 2004  Michael Neumann  <mneumann@ruby-lang.org>
+
+	* lib/xmlrpc/parser.rb, test/xmlrpc/test_features.rb: fixed "assigning
+	  to constants" warnings
+
+Wed Nov 17 09:38:18 2004  Johan Holmberg  <holmberg@iar.se>
+
+	* re.c (rb_reg_initialize_m): should raise exception instead of
+	  compile error.  [ruby-core:03755]
+
+Wed Nov 17 03:42:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_splice): move rb_str_modify() after
+	  StringValue(), which may alter the receiver.  [ruby-dev:24878]
+
+	* error.c (rb_error_frozen): now raise RuntimeError instead of
+	  TypeError.
+
+Tue Nov 16 21:22:47 2004  Michael Neumann  <mneumann@ruby-lang.org>
+
+	* lib/xmlrpc/server.rb (CGIServer): fixed bug when client sends
+	  "Content-typ: text/xml; ..."
+
+Tue Nov 16 23:45:07 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_divmod): protect float values from GC by
+	  assignment to local variables.  [ruby-dev:24873]
+
+Tue Nov 16 14:31:54 2004  Michael Neumann  <mneumann@ruby-lang.org>
+
+	* test/xmlrpc/*: imported and refactored original test cases.
+
+	* test/xmlrpc/test_webrick_server.rb, test/xmlrpc/webrick_testing.rb:
+	  added test case that starts up a WEBrick XML-RPC server and performs
+	  some tests on it (both http and https servers are started).
+
+	* lib/xmlrpc/create.rb (XMLWriter::each_installed_writer),
+	  lib/xmlrpc/parser.rb (XMLParser::each_installed_parser):
+	  added methods to simply original test cases
+
+	* lib/xmlrpc/parser.rb, lib/xmlrpc/datetime.rb: applied patch by
+	  MoonWolf <moonwolf@moonwolf.com> to allow parsing datetime.iso8601
+	  (e.g. 20041105T01:15:23Z).
+
+	* lib/xmlrpc/server.rb: fixed issue #998
+	  (http://rubyforge.org/tracker/?func=detail&atid=1700&aid=998&group_id=426)
+
+	* lib/xmlrpc/create.rb, lib/xmlrpc/utils.rb: when marshalling/loading
+	  user-defined data structures, use Class#allocate instead of defining
+	  an empty #initialize method. module XMLRPC::Marshallable is now only
+	  used for tagging.
+
+	* lib/xmlrpc/.document, lib/xmlrpc/README.rdoc: added howto
+
+Tue Nov 16 16:26:12 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32,wince}/setup.mak (-epilogue-): remove config.h and
+	  config.status to force updating them.
+
+Tue Nov 16 16:20:45 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): position was ignored when a
+	  buffer was passed.  http://www.yo.rim.or.jp/~nov/d/?date=20041116#p03
+
+Tue Nov 16 13:35:54 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner::options): add
+	  new option --exclude (-x) to skip some tests. [ruby-core:3363],
+	  [ruby-dev:24865]
+
+	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir.exclude):
+	  ditto.
+
+Tue Nov 16 11:19:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner::options): use
+	  Regexp conversion.
+
+Tue Nov 16 01:41:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_mod_check): frozen check should be separated.
+	  [ruby-core:3742]
+
+	* array.c (rb_ary_update): pedantic check to detect
+	  rb_ary_to_ary() to modify the receiver.  [ruby-dev:24861]
+
+Mon Nov 15 18:58:05 2004  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in: check rlim_t more portably. [ruby-core:3735]
+
+Mon Nov 15 11:50:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-special-char-p, ruby-parse-partial): handle
+	  operator symbols.  [ruby-talk:120177]
+
+Mon Nov 15 08:58:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_Proc): make proc as an alias to Proc.new.
+	  [ruby-dev:24848]
+
+Mon Nov 15 00:46:03 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_yield_0): lambda{}.call(1) should raise exception.
+	  [ruby-talk:120253]
+
+Mon Nov 15 00:33:40 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_clear): avoid revealing NULL pointer.
+	  [ruby-dev:24766]
+
+	* string.c (str_gsub): add paranoid check. [ruby-dev:24827]
+
+	* string.c (str_mod_check): check frozen status as well.
+	  [ruby-dev:24801]
+
+Sun Nov 14 18:59:03 2004  Tanaka Akira  <akr@m17n.org>
+
+	* process.c (proc_getrlimit): new function for Process.getrlimit.
+	  (proc_setrlimit): new function for Process.setrlimit.
+	  [ruby-dev:24834]
+
+	* configure.in: check rlim_t and its size.  check setrlimit.
+
+	* ruby.h (NUM2ULL): new macro.
+
+Sun Nov 14 13:27:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pp.rb (PP#object_address_group): remove odd number of 'f'
+	  prefixed to negative address.
+
+Sun Nov 14 10:48:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/mathn.rb (Integer::gcd2): faster implementation by
+	  <erlercw@siu.edu>. [ruby-talk:120232]
+
+Sun Nov 14 08:46:33 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/logger/test_logger.rb: Logger just expects
+	  Logger#datetime_format to be used for Time#strftime independently of
+	  locale. [ruby-dev:24828]
+
+Fri Nov 12 17:32:07 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bcc32/README.bcc32, win32/README.win32: need bison instead of
+	  byacc.
+
+Fri Nov 12 15:15:06 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (ruby_options): now we cannot call rb_glob() before
+	  ruby_init(), so call rb_w32_cmdvector() at ruby_options().
+
+	* win32/win32.{c,h} (rb_w32_cmdvector): rename make_cmdvector() and
+	  export it.
+
+Fri Nov 12 14:08:01 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/event.rb: remove $LOADED_FEATURES trick
+
+	* ext/tk/lib/tk.rb: ditto
+
+Fri Nov 12 00:31:05 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/gdbm/gdbm.c (fgdbm_store): StringValue() may alter string
+	  pointer.  [ruby-dev:24783]
+
+Thu Nov 11 17:58:19 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (rb_glob): should have called rb_glob_caller().
+	  [ruby-dev:24773]
+
+Thu Nov 11 16:56:10 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file.rb (test_truncate_wbuf): we want to test
+	  only File#truncate, not behavior of seek(2).
+
+Thu Nov 11 01:52:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (syserr_initialize): use stringified object.
+	  [ruby-dev:24768]
+
+Wed Nov 10 22:49:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/delegate.rb (SimpleDelegator::dup): wrong number of
+	  arguments.
+
+	* lib/delegate.rb (DelegateClass::dup): ditto.
+
+Wed Nov 10 19:47:55 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): path is a string object now.
+
+Wed Nov 10 12:31:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT (Example): extconf.rb is indispensable now.
+
+Wed Nov 10 03:33:36 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: fix SEGV when compiled with Tcl/Tk8.3.x
+	  or older
+
+	* ext/tk/lib/tkextlib/tile/style.rb: bug fix
+
+Tue Nov  9 22:24:07 2004  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf: original nkf.c rev:1.38
+
+	* ext/nkf/nkf.c: fix bug: can't parse long-name options
+
+	* ext/nkf/test.rb: fix bug: mime tests fail
+
+Tue Nov  9 14:27:18 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Officious): moved from DefaultList.
+
+Tue Nov  9 00:50:06 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/rdoc.rb: Change version numbering of RDoc and ri
+
+Tue Nov  9 01:05:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (rb_glob2): do not allocate buffer from heap to avoid
+	  memory leaks.  use string object for buffering instead.
+	  [ruby-dev:24738]
+
+	* dir.c (join_path): ditto.
+
+	* io.c (io_read): external input buffer may be modified even after
+	  rb_str_locktmp().  [ruby-dev:24735]
+
+	* dir.c (fnmatch): p or s may be NULL.  [ruby-dev:24749]
+
+Tue Nov  9 00:36:26 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/extservm.rb: add DRb::ExtServManager#uri=.
+	  [ruby-dev:24743]
+
+Mon Nov  8 22:20:19 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_class):
+	  Fix bug where parent class wasn't being detected if the
+	  child class was defined using the A::B notation.
+
+Mon Nov  8 00:14:13 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: add setup for mingw32 cross compiling.
+	  [ruby-talk:119413]
+
+Sun Nov  7 23:49:26 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: bind-event methods accept multi substitution
+	  arguments.
+
+	* ext/tk/lib/tk/canvas.rb: ditto.
+
+	* ext/tk/lib/tk/canvastag.rb: ditto.
+
+	* ext/tk/lib/tk/text.rb: ditto.
+
+	* ext/tk/lib/tk/texttag.rb: ditto.
+
+	* ext/tk/lib/tkextlib: ditto.
+
+Sat Nov  6 20:40:16 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: rename WIN32OLE#ole_obj_help to
+	  WIN32OLE#ole_type. alias ole_obj_help to ole_type.
+
+	* ext/win32ole/tests/testWIN32OLE.rb: ditto.
+
+Sat Nov  6 11:18:59 2004  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb (_parse): checks whether zone was given.
+
+Sat Nov  6 00:46:27 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_locktmp): check STR_TMPLOCK flag before
+	  locking.  [ruby-dev:24727]
+
+Fri Nov  5 19:07:16 2004  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf: follow CVS Head of original nkf.
+
+Fri Nov  5 18:12:42 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/scrollable.rb: divide Scrollable module into
+	  X_Scrollable and Y_Scrollable
+
+	* ext/tk/lib/tk/entry.rb: include X_Scrollable instead of Scrollable
+
+	* ext/tk/lib/tk/autoload.rb: define autoload for X_Scrollable and
+	  Y_Scrollable
+
+Fri Nov  5 16:05:32 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: TkComm._at() supprts both of "@x,y" and "@x"
+
+Fri Nov  5 13:22:58 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/text.rb: sorry. bug fix again.
+
+Fri Nov  5 13:17:54 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/text.rb: bug fix
+
+Fri Nov  5 08:52:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_mark): stricter GC stack check.
+
+Fri Nov  5 08:34:43 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): should have removed rb_str_unlocktmp(str).
+	  [ruby-dev:24708]
+
+	* ext/socket/socket.c (s_recvfrom): buffer modification check.
+	  [ruby-dev:24708]
+
+Thu Nov  4 23:54:21 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regexec.c, regparse.c, regint.h: fixed conflicts between
+	  vendor branch.
+
+Thu Nov  4 23:41:55 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* ascii.c, euc_jp.c, oniggnu.h, oniguruma.h, regcomp.c,
+	  regenc.c, regenc.h, regerror.c, regexec.c, reggnu.c,
+	  regint.h, regparse.c, regparse.h, sjis.c, utf8.c:
+	  imported Oni Guruma 3.4.0.
+
+	* parse.y, re.c: Now mbclen() takes unsigned char as
+	  its argument.
+
+Thu Nov  4 21:25:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): string modify check no longer based on
+	  tmplock.  [ruby-dev:24706]
+
+Thu Nov  4 21:13:48 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(typelib_file_from_typelib): search "win16"
+	  entry to get library path.
+
+	* ext/win32ole/win32ole.c(oletypelib_path): ditto.
+
+	* ext/win32ole/win32ole.c(ole_typedesc2val): add VT_LPWSTR, VT_LPSTR,
+	  VT_ERROR case.
+
+Thu Nov  4 15:02:14 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb: forget to initialize instance_variables
+	  of TkVarAccess objects
+
+Thu Nov  4 09:11:35 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_mark): enable GC stack checking.
+
+Thu Nov  4 03:11:33 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): lock strings temporarily.  [ruby-dev:24687]
+
+	* ext/socket/socket.c (s_recvfrom): tmplock input buffer.
+	  [ruby-dev:24705]
+
+Wed Nov  3 22:24:17 2004  Daigo Moriwaki  <techml@sgtpepper.net>
+
+	* lib/webrick/httpauth/digestauth.rb: use Base64.encode64 to
+	  avoid warnings.
+
+Wed Nov  3 17:19:59 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_uniq_bang): do not push frozen string from hash
+	  table.  [ruby-dev:24695]
+
+	* array.c (rb_ary_and): ditto.
+
+	* array.c (rb_ary_or): ditto.
+
+Wed Nov  3 17:02:48 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: support to use different Tcl commands between
+	  configure and configinfo
+
+	* ext/tk/lib/font.rb: ditto.
+
+	* ext/tk/lib/itemconfig.rb: support to use different Tcl commands
+	  between item_configure and item_configinfo
+
+	* ext/tk/lib/itemfont.rb: ditto.
+
+	* ext/tk/extconf.rb: install SUPPORT_STATUS
+
+	* ext/tk/lib/tkextlib: some bug fixes (see ext/tk/ChangeLog.tkextlib)
+
+Wed Nov  3 15:38:28 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/*.rb: removed tab width configuration headers.
+
+	* test/rss/test_maker_{0.9,1.0}.rb: sort -> do_sort.
+
+	* lib/rss/maker/*.rb: changed API to RSS version independence.
+
+	* lib/rss/maker/base.rb
+	  (RSS::Maker::XMLStyleSheets::XMLStyleSheet): checked required
+	  (pseudo) attributes.
+
+	* lib/rss/maker/base.rb (RSS::Maker::Items): sort -> do_sort.
+
+	* lib/rss/rss.rb (RSS::BaseModel.install_date_element): avoided
+	  warning.
+
+	* lib/rss/0.9.rb (RSS::Rss#textinput): added convenience method.
+
+Tue Nov  2 16:35:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/enumerator/enumerator.c (each_cons_i): pass copy of an
+	  internal consequent array.  [ruby-talk:118691]
+
+Tue Nov  2 14:54:02 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (pipe_open): need to set cmd if argc == 0 (win32).
+
+Tue Nov  2 01:20:09 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (pipe_open): IO.popen should take array as 1st argument for
+	  a command line.  [ruby-dev:24678]
+
+	* eval.c (proc_invoke): nail down dyna_var node when Proc object
+	  or continuation is created.  [ruby-dev:24671]
+
+	* io.c (rb_io_s_popen): do not expand argv array.  [ruby-dev:24670]
+
+Mon Nov  1 22:25:56 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/base.rb: changed xml-stylesheet's API of RSS Maker
+	  like to item's one.
+
+	* lib/rss/xml-stylesheet.rb (RSS::XMLStyleSheet#guess_type): fixed
+	  regular expression bug.
+
+	* test/rss/test_maker_xml-stylesheet.rb: updated tests for
+	  xml-stylesheet.
+
+Mon Nov  1 13:59:28 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/extmk.rb (MANIFEST): do not use anymore, use extconf.rb instead.
+
+	* ext/enumerator/extconf.rb, ext/fcntl/extconf.rb,
+	  ext/stringio/extconf.rb: added.
+
+	* MANIFEST, ext/**/MANIFEST: removed.
+
+	* README.EXT, README.EXT.ja: remove MANIFEST stuff.
+
+Mon Nov  1 11:52:18 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (pipe_open): avoid conflict of variable name. [ruby-dev:24662]
+
+Mon Nov  1 11:46:19 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* process.c (rb_f_exec): should check whether prog is NULL.
+
+Mon Nov  1 09:37:19 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker.rb: added entry point of RSS Maker.
+
+Mon Nov  1 03:14:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_get_method_body): store ICLASS in the cache.
+	  [ruby-core:03672]
+
+	* eval.c (rb_provided): should return true for loading library
+	  too for autoloading.  [ruby-core:03655]
+
+Mon Nov  1 01:14:52 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_open): create copy of popen specifier.  [ruby-dev:24656]
+
+	* string.c (rb_str_locktmp): lock string temporarily.
+
+	* string.c (str_independent): add tmplock check.
+
+	* io.c (io_write): lock output string temporarily.
+	  [ruby-dev:24649]
+
+	* io.c (io_write): use rb_str_locktmp().
+
+	* io.c (read_all): ditto.
+
+Sun Oct 31 23:37:00 2004  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c: on NetBSD don't use setruid() and setrgid().
+
+Sun Oct 31 23:12:10 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker/*.rb: added RSS Maker.
+
+	* test/rss/test_maker_*.rb: added tests for RSS Maker.
+
+Sun Oct 31 16:58:12 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE.codepage, WIN32OLE.codepage=.
+
+	* ext/win32ole/tests/testWIN32OLE.rb: ditto.
+
+Sun Oct 31 14:35:26 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: add "\075?UTF-8?Q?" for Gmail.
+
+Sun Oct 31 14:18:56 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: emit lexer-event values to the parser
+	  (still incomplete).
+
+Sat Oct 30 15:24:41 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_TYPELIB class. add
+	  WIN32OLE#ole_typelib method.
+
+	* ext/win32ole/tests/testOLETYPELIB.rb: add WIN32OLE_TYPELIB class.
+
+Sat Oct 30 06:53:24 2004  Peter Vanbroekhoven  <peter.vanbroekhoven@cs.kuleuven.ac.be>
+
+	* eval.c (rb_eval): NODE_XSTR should pass copy of literal string.
+
+Sat Oct 30 00:19:40 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_update): a[n,m]=nil no longer works as element
+	  deletion.
+
+	* enum.c (enum_sort_by): protect continuation jump in.
+	  [ruby-dev:24642]
+
+	* eval.c (rb_eval), gc.c (gc_mark_children), node.h (NEW_ALIAS,
+	  NEW_VALIAS), parse.y (fitem): allow dynamic symbols to
+	  NODE_UNDEF and NODE_ALIAS.
+
+Fri Oct 29 21:27:51 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_check_initialized): new function to check uninitialized
+	  object.  [ruby-talk:118234]
+
+	* file.c (rb_file_path), io.c (rb_io_closed): check if initialized.
+
+Fri Oct 29 19:05:33 2004  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf: follow nkf2.0.
+
+Fri Oct 29 17:18:22 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (ripper_s_allocate): add prototype for Microsoft compiler.
+
+	* range.c (range_step, range_each): need cast.
+
+Fri Oct 29 16:34:19 2004  Daiki Ueno  <ueno@unixuser.org>
+
+	* misc/ruby-mode.el (ruby-parse-partial): Parse the rest of the
+	  line after opening heredoc identifier.  [ruby-dev:24635]
+
+Fri Oct 29 11:35:04 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_parser_append_print, rb_parser_while_loop): body node
+	  can be empty.  [ruby-dev:24628]
+
+Fri Oct 29 10:00:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_thread_start_0): forget to free some memory chunks.
+	  [ruby-core:03611]
+
+	* eval.c (ruby_cleanup): ruby_finalize_1 may cause exception,
+	  should be wrapped by PUSH_TAG/POP_TAG().  [ruby-dev:24627]
+
+Thu Oct 28 08:42:02 2004  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (argf_forward): use ANSI style.
+	  (argf_read): call argf_forward with argv argument.
+	  [ruby-dev:24624]
+
+Wed Oct 27 09:17:30 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_gsub): use a string object for exception safeness.
+	  [ruby-dev:24601]
+
+Wed Oct 27 07:38:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* node.h (NODE_TYPESHIFT): allow 4 more bits for line numbers.
+	  [ruby-talk:117841]
+
+	* ruby.h (FL_ABLE): nodes are not subject for flag operations.
+
+	* io.c (ARGF_FORWARD): should have specified argv explicitly,
+	  since we no longer have frame->argv saved.  [ruby-dev:24602]
+
+Tue Oct 26 23:30:39 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/code_objects.rb (RDoc::Context::add_class_or_module):
+	  Restore correct :nopdoc: behavior with nested classes and modules.
+
+Tue Oct 26 18:21:29 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (RESIZE_CAPA): check string attribute before modifying
+	  capacity member of string structure.  [ruby-dev:24594]
+
+Tue Oct 26 11:33:26 2004  David G. Andersen  <dga@lcs.mit.edu>
+
+	* ext/zlib/zlib.c (gzreader_gets): use memchr() to to gain
+	  performance.  [ruby-talk:117701]
+
+Tue Oct 26 10:56:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): raise ArgumentError for extra
+	  arguments, unless (digit)$ style used.
+
+Mon Oct 25 18:35:39 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* win32/win32.c (isUNCRoot): should check NUL after '.'.
+	  [ruby-dev:24590]
+
+	* win32/win32.c (isUNCRoot): fixed buffer overrun.
+
+Mon Oct 25 08:03:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (get_backtrace): ignore illegal backtrace.  [ruby-dev:24587]
+
+Sun Oct 24 00:40:50 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_load, search_required, rb_require_safe, rb_require): use
+	  frozen shared string to avoid outside modification.  [ruby-dev:24580]
+
+Sat Oct 23 23:40:34 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_fptr_finalize): leave stdin/stdout/stderr open in
+	  interpreter termination.  [ruby-dev:24579]
+
+Sat Oct 23 22:18:32 2004  Guy Decoux  <ts@moulon.inra.fr>
+
+	* eval.c (frame_free): Guy Decoux solved the leak problem.
+	  Thanks.  [ruby-core:03549]
+
+Sat Oct 23 00:20:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/zlib/zlib.c (zstream_append_input): clear klass for z->input
+	  to avoid potential vulnerability.
+
+	* ext/zlib/zlib.c (zstream_run): always use zstream_append_input()
+	  to avoid SEGV.  [ruby-dev:24568]
+
+Fri Oct 22 12:02:28 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_alias): was warning for wrong condition.
+	  [ruby-dev:24565]
+
+Fri Oct 22 10:36:37 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#meta_vars):
+	  should check if path_info is not nil.
+
+Fri Oct 22 00:22:31 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/zlib/zlib.c (zstream_shift_buffer): should restore class
+	  field of a buffer.  [ruby-dev:24562]
+
+Fri Oct 22 00:20:33 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_include): should not treat char as negative value.
+	  [ruby-dev:24558]
+
+Thu Oct 21 21:32:30 2004  IWATSUKI Hiroyuki  <don@na.rim.or.jp>
+
+	* lib/pstore.rb (PStore#transaction): Use the empty content when a
+	  file is not found.  [ruby-dev:24561]
+
+Thu Oct 21 19:06:15 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpresponse.rb (WEBrick::HTTPResponse#send_body_io):
+	  ensure to close @body. (http://bugs.debian.org/277520)
+
+Thu Oct 21 13:11:31 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* io.c (pipe_open): variable name "fpw" is conflicted.
+
+Thu Oct 21 00:36:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_alias): should warn on method discarding.
+	  [ruby-dev:24546]
+
+	* ext/zlib/zlib.c (zstream_expand_buffer_into): hide internal
+	  string buffer by clearing klass.  [ruby-dev:24548]
+
+	* parse.y (lex_getline): should not touch ruby_debug_lines if
+	  RIPPER is defined.  [ruby-dev:24547]
+
+Wed Oct 20 19:45:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_gsub): 	reentrant check.  [ruby-dev:24432]
+
+Wed Oct 20 12:42:53 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_getline): rs modification check should not interfere
+	  in the loop.
+
+Wed Oct 20 10:31:33 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (lex_getline): should update ruby_debug_lines.
+
+Wed Oct 20 04:17:55 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/dbm/dbm.c (fdbm_delete_if): should check if deleting element
+	  is a string.  [ruby-dev:24490]
+
+	* ext/sdbm/init.c (fsdbm_delete_if): ditto.
+
+Wed Oct 20 01:37:18 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_times): Array#* should return an instance of
+	  the class of right operand.  [ruby-dev:24526]
+
+	* ext/zlib/zlib.c (zstream_detach_buffer): should not expose
+	  class-less object to Ruby world. [ruby-dev:24530]
+
+	* eval.c (proc_dup): provide Proc#dup as well.  [ruby-talk:116915]
+
+	* eval.c (ruby_exec): stack marking position may be higher than
+	  expected.  thanks to Guy Decoux.  [ruby-core:03527]
+
+Wed Oct 20 00:25:41 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (search_required): required name must not be changed before
+	  loading.  [ruby-dev:24492]
+
+Tue Oct 19 23:59:46 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_require_safe): provide the feature after loaded.
+	  [ruby-list:40085]
+
+Tue Oct 19 22:43:12 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_attr): If
+	  we come across 'attr' in a context where it isn't
+	  followed by a symbol, just issue a warning.
+
+Tue Oct 19 20:32:50 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole.c(ole_invoke): retrieve the result value when
+	  retrying the IDispatch::invoke.
+
+Tue Oct 19 17:24:11 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (read_all): block string buffer modification during
+	  rb_io_fread() by freezing it temporarily. [ruby-dev:24479]
+
+	* dir.c (rb_push_glob): block call at once the end of method.
+	  [ruby-dev:24487]
+
+	* ext/enumerator/enumerator.c (enum_each_slice): remove
+	  rb_gc_force_recycle() to prevent potential SEGV.
+	  [ruby-dev:24499]
+
+	* ext/zlib/zlib.c (zstream_expand_buffer): hide internal string
+	  buffer by clearing klass.  [ruby-dev:24510]
+
+Tue Oct 19 08:47:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_upto): method result must be checked.  [ruby-dev:24504]
+
+	* eval.c (error_print): ditto.  [ruby-dev:24519]
+
+Mon Oct 18 23:37:05 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): check inheritance by the internal function.
+	  [ruby-dev:24515]
+
+Mon Oct 18 11:29:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_flags_mode, rb_io_mode_flags): distinguish whether file
+	  not existing is created.  [ruby-dev:24505]
+
+Mon Oct 18 07:26:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_truncate): discard read buffer before truncation.
+	  [ruby-dev:24197]
+
+Mon Oct 18 01:56:03 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#initialize):
+	  initial value of accpet-* should be array.
+
+Mon Oct 18 00:42:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_getservbyaname): protocol string
+	  might be altered.  [ruby-dev:24503]
+
+	* string.c (rb_str_upto): check if return value from succ is a
+	  string.  [ruby-dev:24504]
+
+Sun Oct 17 23:03:48 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/timer.rb: TkTimer#start and restart accept a block
+
+Sun Oct 17 12:53:46 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fole_func_methods): correct argument mismatch.
+	* ext/win32ole/win32ole.c (fole_get_methods): ditto.
+	* ext/win32ole/win32ole.c (fole_put_methods): ditto.
+	* ext/win32ole/tests/testWIN32OLE.rb: add test for WIN32OLE#ole_func_methods
+	  WIN32OLE#ole_get_methods, WIN32OLE#ole_put_methods
+
+Sat Oct 16 14:45:28 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/0.9.rb (RSS::Rss#to_s): removed garbage.
+
+Sat Oct 16 13:42:49 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/: untabified.
+	* test/rss/: untabified.
+	* lib/rss/0.9.rb (RSS::Rss#to_s): inent -> indent.
+
+Sat Oct 16 13:34:56 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss: supported prety print.
+	* test/rss/test_1.0.rb: added test for calculating default indent size.
+
+Sat Oct 16 10:56:36 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): install-rb is needed for statically
+	  linked extensions.  [ruby-dev:24491]
+
+Fri Oct 15 18:07:08 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509store.c
+	  (ossl_x509stctx_initialize): setup OpenSSL::X509::StoreContext with
+	  ossl_x509stctx_* functions instead of X509_STORE_CTX_*.
+	  (ossl_x509store_set_time): add OpenSSL::X509::Store#time=.
+	  (ossl_x509stctx_set_time): add OpenSSL::X509::StoreContext#time=.
+
+	* test/openssl/ossl_x509store.rb: test certificate validity times.
+
+Fri Oct 15 18:04:35 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/timer.rb: TkTimer.new(interval, loop){ ... } is
+	  acceptable. Add TkTimer.start ( == new + start ).
+
+Fri Oct 15 12:43:09 2004  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (Init_stack): make prototype declaration consistent with
+	  the definition in gc.c.
+
+Thu Oct 14 13:33:59 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: added link to Tutorial.
+
+Tue Oct 12 21:22:50 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/config.rb:
+	  add WEBrick::Config::FileHandler[:AcceptableLanguages].
+
+	* lib/webrick/httpservlet/filehandler.rb
+	  (WEBrick::HTTPServlet::FileHandler#set_filename): search files
+	  having suffix of language-name which Accept-Language header field
+	  includes if :AcceptableLanguages options is present.
+
+	* lib/webrick/httpservlet/filehandler.rb
+	  (WEBrick::HTTPServlet::FileHandler#get_servlet): new method to
+	  search servlet correspond to the suffix of filename.
+
+	* lib/webrick/httprequest.rb: add attributes access methods: accept,
+	  accept_charset, accept_encoding, accept_language, content_length
+	  and content_type.
+
+	* lib/webrick/httpresponse.rb: add attribute access methods:
+	  content_length, content_length=, content_type and content_type=.
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.mime_types):
+	  use the second suffix to detect media type. (the first suffix
+	  may be a language name.)
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.parse_qvalues):
+	  add method to parse Accept header field. it returns an Array of
+	  values sorted by the qvalues.
+
+Tue Oct 12 15:05:32 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* io.c (MODE_BINMODE, MODE_BINARY): fixed reversed condition.
+
+Mon Oct 11 17:51:34 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_popen): get mode string via rb_io_flags_mode() to
+	  avoid mode string modification.  [ruby-dev:24454]
+
+	* io.c (rb_io_getline_fast): should take delim as unsigned char to
+	  distinguish EOF and '\377'.  [ruby-dev:24460]
+
+	* io.c (rb_io_getline): add check for RS modification.
+	  [ruby-dev:24461]
+
+	* enum.c (enum_sort_by): use qsort() directly instead using
+	  rb_iterate().  [ruby-dev:24462]
+
+	* enum.c (enum_each_with_index): remove rb_gc_force_recycle() to
+	  prevent access to recycled object (via continuation for
+	  example).  [ruby-dev:24463]
+
+Mon Oct 11 13:48:20 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/*: untabify
+
+Sun Oct 10 12:32:08 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::parse_require): Allow 'require'
+	  to be used as a variable name
+
+Sun Oct 10 02:49:14 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/lib/ripper/filter.rb: require ripper/tokenizer.
+
+	* ext/ripper/lib/ripper/filter.rb (parse): argument is optional.
+
+Sun Oct 10 02:43:13 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: \n between two comments disappeared.
+
+Sat Oct  9 21:23:37 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/converter.rb: changed to try to use Iconv for default
+	  conversion.
+
+	* lib/rss/rss.rb: 0.0.9 -> 0.1.0.
+
+Sat Oct  9 19:50:36 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_getline): should not treat char as negative value.
+	  [ruby-dev:24460]
+
+Sat Oct  9 00:25:39 2004  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (rb_io_fread): rb_thread_wait_fd() was lost.
+	  [ruby-dev:24457]
+
+Fri Oct  8 21:36:56 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (fu_traverse): return value of Dir.entries is
+	  reliable. (pass $SAFE=1)
+
+	* lib/fileutils.rb (remove_dir): return value of Dir.foreach is
+	  reliable. (pass $SAFE=1)
+
+Fri Oct  8 09:49:32 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): pointer modification check before each
+	  iteration.  [ruby-dev:24445]
+
+Fri Oct  8 01:13:05 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/optiondb.rb: make it more secure
+
+Thu Oct  7 23:47:57 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/scrollbar.rb: When 'set' operation, a scrollbar
+	  cannot propagate view port information from the source widget
+	  (that calls 'set') to other assigned widgets.
+
+Thu Oct  7 17:36:25 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: When CHILDKILLED and so on, Tk.errorCode returns
+	  a Fixnum for 2nd element (it's pid) of the return value.
+
+Thu Oct  7 12:55:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (io_read): should freeze buffer before thread context
+	  switch. [ruby-dev:24442]
+
+	* pack.c (pack_unpack): string conversion should at the top of the
+	  method.  [ruby-dev:24439]
+
+	* io.c (io_read): buffer should be frozen only after the length
+	  check.  [ruby-dev:24440]
+
+Thu Oct  7 02:56:43 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c: use FMODE_APPEND.
+
+Thu Oct  7 01:05:33 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: add Tk.errorInfo and Tk.errorCode
+
+Thu Oct  7 00:08:37 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_s_sysopen): preserve path in the buffer allocated by
+	  ALLOCA_N() to prevent modification.  [ruby-dev:24438]
+
+Wed Oct  6 09:21:00 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_mode_flags): preserve append mode flag.
+	  [ruby-dev:24436]
+
+	* io.c (rb_io_modenum_mode): do not use external output buffer.
+
+	* string.c (rb_str_justify): differ pointer retrieval to prevent
+	  padding string modification.  [ruby-dev:24434]
+
+	* range.c (range_each_func): allow func to terminate loop by
+	  returning RANGE_EACH_BREAK.
+
+	* range.c (member_i): use RANGE_EACH_BREAK. [ruby-talk:114959]
+
+Tue Oct  5 09:53:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_fopen): mode string copy at the lowest level.
+
+	* io.c (rb_io_flags_mode): requires output buffer no more.  no
+	  allocation needed.
+
+	* array.c (rb_ary_index): takes a block to compare items in an
+	  array.  [ruby-talk:113069] [Ruby2]
+
+	* array.c (rb_ary_rindex): ditto.
+
+Mon Oct  4 14:03:40 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_file_open_internal, rb_io_reopen): fname might be altered
+	  while GC.  [ruby-dev:24408]
+
+Mon Oct  4 12:53:45 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/optiondb.rb: support definition of command
+	  resources on widgets
+
+	* ext/tk/lib/tk/image.rb: bug fix
+
+Sun Oct  3 21:16:05 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (TEXT_REGEXP): allow 8-bit characters for the german
+	  version of Microsoft Exchange Server.
+
+	* lib/net/imap.rb (RTEXT_REGEXP): ditto.
+
+	* lib/net/imap.rb (CTEXT_REGEXP): ditto.
+
+Sat Oct  2 20:34:05 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (local_vars): moved to struct parser_params.
+	  [ruby-dev:24391]
+
+	* parser.y (stmts): remove suspicious NODE_BEGIN.  [ruby-dev:24390]
+
+	* node.h (NEW_DVAR): extra semicolon.
+
+Sat Oct  2 00:42:20 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_byte): retrieve pointer from string value for each
+	  time.  [ruby-dev:24404]
+
+	* marshal.c (r_bytes0): ditto.
+
+	* enum.c (sort_by_i): re-entrance check added.  [ruby-dev:24399]
+
+	* io.c (io_read): should freeze all reading buffer.
+	  [ruby-dev:24400]
+
+	* string.c (rb_str_sum): should use bignums when bits is greater
+	  than or equals to sizeof(long)*CHAR_BITS. [ruby-dev:24395]
+
+	* eval.c (specific_eval): defer pointer retrieval to prevent
+	  unsafe sourcefile string modification.  [ruby-dev:24382]
+
+	* string.c (rb_str_sum): wrong cast caused wrong result.
+	  [ruby-dev:24385]
+
+	* enum.c (enum_sort_by): hide temporary array from
+	  ObjectSpace.each_object.  [ruby-dev:24386]
+
+	* string.c (rb_str_sum): check was done with false pointer.
+	  [ruby-dev:24383]
+
+	* string.c (rb_str_sum): string may be altered.  [ruby-dev:24381]
+
+Fri Oct  1 11:40:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_eval): defer pointer retrieval to prevent unsafe
+	  sourcefile string modification.  [ruby-dev:24373]
+
+	* io.c (io_read): block string buffer modification during
+	  rb_io_fread() by freezing it temporarily. [ruby-dev:24366]
+
+	* io.c (rb_io_s_popen): mode argument may be altered.
+	  [ruby-dev:24375]
+
+	* file.c (rb_file_s_basename): ext argument may be altered.
+	  [ruby-dev:24377]
+
+	* enum.c (enum_sort_by): use NODE instead of 2 element arrays.
+	  [ruby-dev:24378]
+
+	* string.c (rb_str_chomp_bang): StringValue() may change the
+	  receiver.  [ruby-dev:24371]
+
+Fri Oct  1 11:25:20 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/grid.rb: revive TkGrid.grid
+
+	* ext/tk/lib/tk/pack.rb: revive TkPack.pack
+
+	* ext/tk/lib/tk/place.rb: revive TkPlace.place
+
+Thu Sep 30 00:50:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (ip_init): bug fix
+
+	* ext/tk/tkutil.c (get_eval_string_core): accept a Regexp object
+
+	* ext/tk/lib/multi-tk.rb: fix bug on 'exit' operation
+
+	* ext/tk/lib/tk/text.rb: 'tksearch' accepts a Regexp object as a
+	  matting pattern argument
+
+Wed Sep 29 10:58:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (sort_by_i): internally used object must not be changed
+	  outside.  [ruby-dev:24368]
+
+Mon Sep 27 21:25:12 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_call0): invoke finalizers periodically.
+
+	* gc.c (gc_sweep): defer running finalizers.  [ruby-dev:24354]
+
+	* gc.c (rb_gc_finalize_deferred): run deferred finalizers.
+
+Mon Sep 27 15:01:59 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: missing ';'.
+
+Mon Sep 27 13:46:45 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h, struct.c (rb_struct_s_members, rb_struct_members): public
+	  accessors.  [ruby-dev:24342]
+
+	* marshal.c (w_object, r_object0): use accessors.
+
+Mon Sep 27 09:14:03 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_delete): comparison may change the capacity.
+	  [ruby-dev:24348]
+
+	* array.c (rb_ary_fill): fill should honor length argument.
+	  [ruby-dev:24346]
+
+	* array.c (rb_ary_replace): should not use ptr from shared array.
+	  [ruby-dev:24345]
+
+	* ext/socket/socket.c (s_accept): don't retry for EWOULDBLOCK.
+	  [ruby-talk:113807]
+
+Sun Sep 26 08:05:10 2004  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: provides {Time,Date,DateTime}#to_{time,date,datetime}.
+
+	* sample/cal.rb: uses getoptlong instead of getopts.
+
+Sat Sep 25 18:39:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (flatten): element size might change during comparison.
+	  [ruby-dev:24343]
+
+Sat Sep 25 01:52:49 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_s_members): wrong call of struct_members.
+	  [ruby-dev:24333]
+
+Fri Sep 24 16:09:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): propagate DVAR_DONT_RECYCLE on termination
+	  to avoid double call to rb_gc_force_recycle(). [ruby-dev:24311]
+
+Fri Sep 24 08:29:45 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (rb_parser_append_print): should handle prelude.
+	  [llama@u01.gate0]
+
+	* parse.y (rb_parser_while_loop): ditto.
+
+	* array.c (rb_ary_subseq): original object might be modified after
+	  sharing data creation.  [ruby-dev:24327]
+
+	* array.c (rb_ary_replace): ditto.
+
+	* array.c (ary_make_shared): freeze shared array. [ruby-dev:24325]
+
+	* struct.c (struct_members): always check struct size and size of
+	  members list in the class.  [ruby-dev:24320]
+
+Thu Sep 23 19:48:14 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/Makefile.dev: removed.
+
+	* ext/ripper/ripper.rb.in: moved to lib/ripper/core.rb.in.
+
+	* ext/ripper/lib/ripper/core.rb: new file.
+
+	* ext/ripper/lib/ripper/core.rb.in: new file.
+
+	* ext/ripper/tools/generate-ripper_rb.rb: change comment.
+
+	* test/ripper/*.rb: on__scan event removed.
+
+	* test/ripper/*.rb: event name is changed: on__XXX -> on_XXX.
+
+Thu Sep 23 09:29:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_sub_bang): check if string is not modified
+	  during iteration.  [ruby-dev:24315]
+
+	* hash.c (rb_hash_rehash): replace st_foreach() by its deep
+	  checking counterpart.  [ruby-dev:24310]
+
+Wed Sep 22 14:21:54 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: on__scan event removed.
+
+	* parse.y [ripper]: event name is changed: on__XXX -> on_XXX.
+
+	* ext/ripper/eventids2.c: ditto.
+
+	* ext/ripper/ripper.rb.in: ditto.
+
+	* ext/ripper/lib/ripper.rb: sync with ripper.rb.in.
+
+	* ext/ripper/lib/ripper/tokenizer: ditto.
+
+	* ext/ripper/lib/ripper/filter: new file.
+
+	* sample/ripper/colorize.rb: new file.
+
+	* sample/ripper/strip-comment.rb: new file.
+
+Wed Sep 22 13:50:49 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (hash_alloc): was using tbl pointer without
+	  initialization.
+
+Wed Sep 22 13:38:12 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_rehash): add iteration check.  [ruby-dev:24301]
+
+	* st.c (st_foreach): add deep check.
+
+	* hash.c (rb_hash_fetch): returns KeyError instead of IndexError.
+
+	* hash.c (env_fetch): ditto.
+
+Wed Sep 22 13:02:02 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_call_handler): workaround for Ctrl-C.
+
+Wed Sep 22 09:04:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y: remove global variables ruby_eval_tree and
+	  ruby_eval_tree_begin.
+
+	* array.c (rb_ary_collect_bang): element size might change during
+	  comparison.  [ruby-dev:24300]
+
+	* array.c (rb_ary_reject_bang): ditto. [ruby-dev:24300]
+
+	* array.c (rb_ary_eql): ditto. [ruby-dev:24300]
+
+Wed Sep 22 00:11:12 2004  Dave Thomas  <dave@pragprog.com>
+
+	* process.c: Add documentation for fork()
+
+Tue Sep 21 18:29:49 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_uniq_bang): element size might change during
+	  comparison.  [ruby-dev:24298]
+
+Mon Sep 20 17:46:51 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/lib/ripper/tokenizer.rb: fix typo.
+
+Mon Sep 20 17:38:43 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ripper/test_scanner_events.rb: tokens must be reordered.
+
+	* ext/ripper/lib/ripper/tokenizer.rb: ditto.
+
+Mon Sep 20 16:58:16 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: do not delay dispatching.
+
+	* ext/ripper/lib/ripper/tokenizer.rb: sort tokens by right order.
+
+Mon Sep 20 15:17:47 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/lib/ripper/tokenizer.rb: new file.
+
+Mon Sep 20 15:13:52 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ripper/test_scanner_events.rb: test #lineno and #column.
+
+Mon Sep 20 14:50:17 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: delayed heredocument events should be
+	  dispatched after EOF.
+
+Mon Sep 20 14:39:42 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: adjust lineno and columns for multi-line
+	  strings.
+
+	* parse.y [ripper]: delay heredocument events until seeing
+	  end-of-line.
+
+	* parse.y [ripper]: event on__heredoc_contentn ->
+	  on__tstring_content.
+
+	* ext/ripper/eventids2.c: ditto.
+
+	* ext/ripper/lib/ripper.rb: sync with eventids2.c.
+
+	* test/ripper/test_scanner_events.rb: test it.
+
+	* ext/ripper/tools/generate-ripper_rb.rb: show basename of input.
+
+	* ext/ripper/Makefile.dev: support objdir build.
+
+Mon Sep 20 13:22:55 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: remove Ripper#pos.
+
+	* parse.y [ripper]: Ripper#column should return the column of the
+	  current token.
+
+Mon Sep 20 12:02:41 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: unify old_lex_p and token_head.
+
+	* test/ripper/test_scanner_events.rb: now \r\n is saved correctly.
+
+	* parse.y: new macro lex_goto_eol() for next change.
+
+Mon Sep 20 11:01:55 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: adjust line number for heredoc.  [ruby-dev:24272]
+
+Mon Sep 20 04:49:22 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/ripper.rb.in: new const Ripper::PARSER_EVENT_TABLE.
+
+	* ext/ripper/ripper.rb.in: new const Ripper::SCANNER_EVENT_TABLE.
+
+	* ext/ripper/lib/ripper.rb: sync with ripper.rb.in.
+
+Mon Sep 20 04:13:00 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ripper/test_scanner_events.rb: test spaces before heredoc
+	  mark.
+
+Mon Sep 20 03:46:54 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: spaces before heredoc marker was lost.
+	  [ruby-dev:24272]
+
+	* keywords: rb_reserved_word() should be defined only in ruby
+	  core.  [ruby-dev:24272]
+
+	* lex.c: sync with keywords.
+
+	* ext/ripper/ripper.rb.in (parse): fix typo.
+
+	* ext/ripper/lib/ripper.rb: sync with ripper.rb.in.
+
+Mon Sep 20 03:37:59 2004  Tanaka Akira  <akr@m17n.org>
+
+	* ext/zlib/zlib.c (gzfile_read_raw): call readpartial at first.
+	  (Zlib::GzipReader#readpartial): new method.
+
+Mon Sep 20 00:24:19 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_sort_by): do not use qsort directly.  use
+	  rb_ary_sort_bang() instead.  [ruby-dev:24291]
+
+	* enum.c (enum_sort_by): pedantic type check added.
+	  [ruby-dev:24291]
+
+	* hash.c (rb_hash_foreach_iter): check iter_lev after each
+	  iteration.  [ruby-dev:24289]
+
+	* array.c (rb_ary_and): element size might change during
+	  comparison.  [ruby-dev:24290]
+
+	* array.c (rb_ary_or): ditto. [ruby-dev:24292]
+
+	* array.c (rb_ary_equal): wrong fix. [ruby-dev:24286]
+
+Sat Sep 18 15:02:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_equal): element size might change during
+	  comparison.  [ruby-dev:24254]
+
+	* array.c (rb_ary_diff): ditto. [ruby-dev:24274]
+
+	* array.c (rb_ary_select): ditto. [ruby-dev:24278]
+
+	* array.c (rb_ary_delete): ditto. [ruby-dev:24283]
+
+	* array.c (rb_ary_rindex): ditto. [ruby-dev:24275]
+
+	* array.c (rb_ary_initialize): element size might change during
+	  initializing block.  [ruby-dev:24284]
+
+Sat Sep 18 14:10:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_s_chdir): avoid memory leak and unnecessary chdir to
+	  the original directory when exception has caused in changing
+	  directory or within block.  thanks to Johan Holmberg
+	  <holmberg@iar.se> [ruby-core:03446]
+
+Fri Sep 17 20:29:33 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y: add prototypes for Microsoft compiler.
+
+	* ext/ripper/depend (parse.obj): lex.c exists at hdrdir.
+
+	* {bcc32,win32,wince}/Makefile.sub (YACC, YFLAGS, parse.c):
+	  use bison.
+
+Fri Sep 17 17:11:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c (rb_ary_delete): element comparison might change array
+	  size. [ruby-dev:24273]
+
+	* parse.y: make ruby parser reentrant. merge ripper parser to the
+	  real one.  this change makes ruby require bison.
+
+	* file.c (rb_file_truncate): clear stdio buffer before truncating
+	  the file.  [ruby-dev:24191]
+
+	* ext/digest/digest.c: use rb_obj_class() instead of CLASS_OF
+	  which might return singleton class.  [ruby-dev:24202]
+
+Fri Sep 17 16:07:09 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: improve exit operation
+
+Fri Sep 17 15:01:57 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: fix SEGV when (thread_)vwait or
+	  (thread_)tkwait
+
+	* ext/tk/lib/tk.rb: add alias wait_window to wait_destroy
+
+	* ext/tk/lib/multi-tk.rb: support calling 'mainloop' on slave
+	  interpreters (however, the 'real' eventloop must be run on the
+	  Default Master IP)
+
+	* ext/tk/lib/remote-tk.rb: follow the changes of ext/tk/lib/multi-tk.rb
+
+	* ext/tk/sample/remote-ip_sample2.rb: ditto
+
+	* ext/tk/sample/tkoptdb-safeTk.rb: ditto
+
+Thu Sep 16 18:12:13 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI#start): should set REMOTE_USER
+	  to request.user attribute.
+
+	* lib/webrick/httpservlet/filehandler.rb
+	  (WEBrick::HTTPServlet::FileHandler#initialize): should expand
+	  the pathname of document root directory.
+
+Thu Sep 16 15:49:28 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_intern): protect string argument from GC.
+	  [ruby-core:03411]
+
+Wed Sep 15 20:22:23 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tkoptdb-safeTk.rb: fix a bug depend on the changes
+	  of MultiTkIp
+
+Tue Sep 14 23:54:11 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string was en-bugged by
+	  the previous changes.
+
+Tue Sep 14 23:45:44 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_formatter.rb (RI::TextFormatter::TextFormatter.for):
+	  Add Eric Hodel's simpleformatter.
+
+Tue Sep 14 22:11:08 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: Add rdoc.
+
+Tue Sep 14 20:24:49 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y [ripper]: space event is on__sp, not on__lwsp.
+	  [ruby-dev:24257]
+
+	* ext/ripper/eventids2.c: ditto.
+
+	* ext/ripper/lib/ripper.rb: ditto.
+
+	* ext/ripper/depend (ripper.o): No action is needed.
+	  [ruby-dev:24260]
+
+	* ext/ripper/depend: Borland make does not accept pipes in
+	  Makefile rules. [ruby-dev:24589]
+
+	* ext/ripper/depend: separate rules for developpers.
+
+	* ext/ripper/Makefile.dev: new file.
+
+	* ext/ripper/MANIFEST: add Makefile.dev.
+
+	* ext/ripper/tools/generate-eventids1.rb: read from file, not
+	  stdin.
+
+	* ext/ripper/extconf.rb: clean ripper.E.
+
+	* ext/ripper/tools/generate-ripper_rb.rb: #include ids1/ids2
+	  function was lost.
+
+	* ext/ripper/tools/generate-ripper_rb.rb: SCANNER_EVENTS wrongly
+	  contained parser events.
+
+	* ext/ripper/lib/ripper.rb: ditto.
+
+Tue Sep 14 16:59:37 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: fix SEGV
+
+	* ext/tk/lib/multi-tk.rb: improve safe-level handling of argument proc
+
+	* ext/tk/sample/multi-ip_sample.rb: rename of old 'safe-tk.rb'
+
+	* ext/tk/sample/safe-tk.rb: new sample script
+
+Mon Sep 13 21:33:40 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_to_s): add optional
+	  second argument to specify the output format (see also
+	   X509_NAME_print_ex).
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_init): new constants:
+	  OpenSSL::X509::Name::COMPAT, OpenSSL::X509::Name::RFC2253,
+	  OpenSSL::X509::ONELINE, OpenSSL::X509::MULTILINE.
+
+	* ext/openssl/lib/openssl/x509.rb (OpenSSL::X509::Name::RFC2253DN):
+	  new module to provide the parse for RFC2253 DN format.
+
+	* ext/openssl/lib/openssl/x509.rb (OpenSSL::X509::Name.parse_rfc2253):
+	  new method to parse RFC2253 DN format.
+
+Mon Sep 13 19:16:33 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* eval.c (blk_copy_prev): need frame_dup().  [ruby-dev:24103]
+
+Mon Sep 13 16:23:27 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: MultiTkIp.new_master and new_slave accept
+	  safe-level value argument
+
+Mon Sep 13 10:48:37 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_getpid): don't need to use _getpid() on
+	  mswin32 and mingw32.
+
+Mon Sep 13 10:22:05 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* object.c (nil_inspect): fix typo.
+
+Mon Sep 13 09:29:58 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/depend: (nmake hack) prepend "./" to ripper.c to
+	  avoid {$(srcdir)}.
+
+Mon Sep 13 06:43:42 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper/tools/list-parse-event-ids.rb: does not use getopts.
+
+	* ext/ripper/tools/list-scan-event-ids.rb: ditto.
+
+Mon Sep 13 02:42:28 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/Setup: add ripper.
+
+	* ext/Setup.atheos: ditto.
+
+	* ext/Setup.dj: ditto.
+
+	* ext/Setup.emx: ditto.
+
+	* ext/Setup.nt: ditto.
+
+	* ext/Setup.x68: ditto.
+
+Mon Sep 13 02:26:31 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/ripper: ripper extension added.
+
+	* ext/ripper/MANIFEST: new file.
+
+	* ext/ripper/README: new file.
+
+	* ext/ripper/depend: new file.
+
+	* ext/ripper/extconf.rb: new file.
+
+	* ext/ripper/eventids2.c: new file.
+
+	* ext/ripper/ripper.rb.in: new file.
+
+	* ext/ripper/lib/ripper.rb: new file.
+
+	* ext/ripper/test/check-event-arity.rb: new file.
+
+	* ext/ripper/test/check-event-coverage.sh: new file.
+
+	* ext/ripper/test/check-scanner-event-coverage.rb: new file.
+
+	* ext/ripper/test/list-called-events.rb: new file.
+
+	* ext/ripper/test/src_rb: new file.
+
+	* ext/ripper/test/validate.rb: new file.
+
+	* ext/ripper/tools/generate-eventids1.rb: new file.
+
+	* ext/ripper/tools/generate-param-macros.rb: new file.
+
+	* ext/ripper/tools/generate-ripper_rb.rb: new file.
+
+	* ext/ripper/tools/list-parse-event-ids.rb: new file.
+
+	* ext/ripper/tools/list-scan-event-ids.rb: new file.
+
+	* ext/ripper/tools/preproc.rb: new file.
+
+	* ext/ripper/tools/strip.rb: new file.
+
+	* test/ripper: ripper tests added.
+
+	* test/ripper/dummyparser.rb: new file.
+
+	* test/ripper/test_parser_events.rb: new file.
+
+	* test/ripper/test_scanner_events.rb: new file.
+
+Mon Sep 13 01:03:02 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: improve control of preserv/release tcltkip
+
+	* ext/tcltklib/tcltklib.c: store original 'exit' command
+
+	* ext/tk/tkutil.c: fix(?) SEGV
+
+Mon Sep 13 00:22:53 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y: fix file header.
+
+Mon Sep 13 00:20:39 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y: ripper merged.
+
+	* lex.c: ditto.
+
+	* keywords: ditto.
+
+Sun Sep 12 23:53:17 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* util.c (ruby_strdup): remove unnecessary code. (xmalloc never
+	  returns NULL.)
+
+	* util.c (ruby_getcwd): fix memory leak on failure.
+
+Sun Sep 12 02:41:58 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: add TclTkIp#allow_ruby_exit? and
+	  allow_ruby_exit=
+
+	* ext/tk/lib/multi-tk.rb: ditto.
+
+	* ext/tk/lib/remote-tk.rb: ditto.
+
+	* ext/tcltklib/MANUAL.euc: ditto.
+
+	* ext/tcltklib/MANUAL.eng: ditto.
+
+	* ext/tcltklib/tcltklib.c: fix some reasons of SEGV
+
+	* ext/tk/tkutil.c: ditto.
+
+	* ext/tk/lib/multi-tk.rb: ditto.
+
+	* ext/tk/lib/tk/timer.rb: ditto.
+
+Sat Sep 11 16:09:46 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb: Fix up cross-file class merging.
+
+Fri Sep 10 20:18:05 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tcltklib/tcltklib.c (lib_merge_tklist): fix suspicious
+	  pointer conversion.
+
+Fri Sep 10 19:16:24 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/Makefile.sub: bccwin32 port starts to use RTL dll.
+	  (need to rebuild all) [ruby-dev:24138]
+
+	* win32/win32.{h,c}: ditto.
+
+Fri Sep 10 15:55:59 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mkdir_p): should pass mode argument to
+	  Dir.mkdir. [ruby-dev:24242]
+
+	* test/fileutils/test_fileutils.rb: test it.
+
+Fri Sep 10 02:43:54 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/template/kilmer.rb: James Buck's
+	  patch for call-seq.
+
+Thu Sep  9 13:58:56 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (ip_init): change flag value for setting
+	  'argv' and 'argv0' variable
+
+	* ext/tk/lib/remote-tk.rb: follow changes of multi-tk.rb
+
+Thu Sep  9 11:46:18 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_classes): Allow
+	  spaces aroun parameter to define_method_under (James Buck)
+
+Wed Sep  8 18:44:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_write): zero fill a gap if exsts.
+	  [ruby-dev:24190]
+
+Wed Sep  8 15:19:49 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (ip_init): cannot create a IP at level 4
+
+	* ext/tk/lib/multi-tk.rb: improve 'exit' operation, security check,
+	  and error treatment
+
+	* ext/tk/lib/multi-tk.rb: allow a trusted slave IP to create slave IPs
+
+	* ext/tk/lib/tk/listbox.rb: add TkListbox#value, value=, clear,	and
+	  erase
+
+	* ext/tk/lib/tk/text.rb: add TkText#clear and erase
+
+Tue Sep  7 15:17:49 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_connect): break immediately if a
+	  socket is non-blocking.  [ruby-talk:111654]
+
+Tue Sep  7 12:48:22 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32,wince}/Makefile.sub (config.h): add fcntl.
+
+	* win32/win32.[ch] (fcntl): ditto.
+
+	* win32/win32.c (rb_w32_connect): support nonblocking mode.
+
+	* ext/socket/socket.c (wait_connectable, ruby_connect): support
+	  nonblocking connect on various platforms.
+
+Mon Sep  6 11:00:47 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (dir_s_chdir): the patch to shut up false warning when
+	  exception occurred within a block.  a patch was given from Johan
+	  Holmberg <holmberg at iar.se>.  [ruby-core:03292]
+
+Mon Sep  6 10:57:40 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/lib/tk/menu.rb(TkOptionMenubutton#insert): call correct method
+
+Mon Sep  6 07:51:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (cvar_cbase): singletons should refer to outer cvar scope.
+	  [ruby-dev:24223]
+
+	* eval.c (rb_load): should preserve previous ruby_wrapper value.
+	  [ruby-dev:24226]
+
+Sat Sep  4 01:14:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (cvar_cbase): class variables cause SEGV in
+	  instance_eval() for fixnums and symbols. [ruby-dev:24213]
+
+Fri Sep  3 17:47:58 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (make_struct): remove redefining constant when
+	  conflict.  [ruby-dev:24210]
+
+Fri Sep  3 11:31:44 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: Tk.after makes TkCore::INTERP.tk_cmd_tbl grow
+	  [ruby-dev:24207]
+
+Fri Sep  3 02:12:48 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: fix typo [ruby-talk:111266]
+
+	* ext/tk/lib/tk/text.rb: fix typo
+
+	* ext/tk/lib/multi-tk.rb: improve safe-level treatment on slave IPs
+
+Fri Sep  3 01:54:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): extact target prefix from Makefiles.
+
+	* ext/extmk.rb: already built-in libraries satisfy dependencies.
+	  [ruby-dev:24028]
+
+Wed Sep  1 21:16:50 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/lib/tk/spinbox.rb: fix typo
+
+Wed Sep  1 19:28:37 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_proc_exec): label cannot precede variable declarations.
+
+Tue Aug 31 18:20:49 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/tk/tkutil.c (cbsubst_init): fix memory leak
+
+	* ext/tk/tkutil.c (cbsubst_get_all_subst_keys): fix SEGV
+
+Tue Aug 31 16:04:22 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (ip_delete): when a tcltkip is deleted,
+	  destroy its root widget
+
+Tue Aug 31 12:30:36 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (del_root): fix SEGV
+
+Mon Aug 30 21:50:14 2004  Dave Thomas  <dave@pragprog.com>
+
+	* object.c: Add RDoc for Module.included.
+
+Mon Aug 30 23:11:06 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_driver.rb (and others): ri now merges documentation
+	  if it finds the same class in multiple places.
+
+Mon Aug 30 22:40:30 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: 'restart' method accepts arguments
+
+Mon Aug 30 15:10:46 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (GNU/k*BSD): fixed FTBFS on GNU/k*BSD. [ruby-dev:24051]
+
+Sun Aug 29 14:08:56 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: compile error on bcc32 [ruby-dev:24081]
+
+	* ext/tk/lib/multi-tk.rb: MultiTkIp#eval_string does not work
+
+Sat Aug 28 23:04:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_and): protect parameters from GC.
+	  [ruby-talk:110664]
+
+Fri Aug 27 12:13:50 2004  Tanaka Akira  <akr@m17n.org>
+
+	* ext/stringio/stringio.c (Init_stringio): add StringIO#readpartial as
+	  an alias for StringIO#sysread.
+
+Fri Aug 27 10:14:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_proc_exec): strip trailing spaces.  [ruby-dev:24143]
+
+	* win32/win32.c (CreateChild): ditto.
+
+Thu Aug 26 04:38:29 2004  Dave Thomas  <dave@pragprog.com>
+
+	* eval.c (return_jump): Minor typo in error message. Now reads
+	  "return can't jump across threads".
+
+Wed Aug 25 15:18:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_longjmp): Exception#to_str is no longer defined.
+
+Wed Aug 25 11:39:10 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (exc_equal): exceptions are equal if they share same
+	  class, message and backtrace. [ruby-talk:110354]
+
+	* error.c (name_err_mesg_equal): ditto.
+
+Tue Aug 24 16:41:48 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session::FileStore#initialize): do not
+	  use a session id as a filename.
+
+	* lib/cgi/session/pstore.rb (CGI::Session::PStore#initialize): ditto.
+
+	* lib/cgi/session/pstore.rb (CGI::Session::PStore#initialize): use
+	  Dir::tmpdir.
+
+Tue Aug 24 14:32:17 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session::FileStore#initialize): untaint
+	  session id after check.
+
+Tue Aug 24 08:57:51 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509attr.c (ossl_x509attr_initialize): d2i
+	  functions may replace the pointer indicated by the first argument.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_initialize): ditto.
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_initialize): ditto.
+
+Mon Aug 23 12:43:32 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/resolv.rb (Config.default_config_hash): when multiple domains
+	  are set, Win32::Resolv.get_resolv_info returns Array.
+
+Sun Aug 22 16:27:38 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read):
+	  - should return an empty string if specified length to read is 0.
+	  - should check for pending data and wait for fd before reading.
+	  - call underlying IO's sysread if SSL session is not started.
+	  [ruby-dev:24072], [ruby-dev:24075]
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_write):
+	  - call underlying IO's syswrite if SSL session is not started.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_pending): new method
+	  OpenSSL::SSL#pending.
+
+	* ext/openssl/lib/openssl/buffering.rb: should not use select.
+
+Sun Aug 22 01:10:36 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_connect):
+	  should call :ProxyContentHandler before finishing CONNECT.
+
+Sat Aug 21 06:41:16 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tcltklib/extconf.rb (find_tcl, find_tk): find stub library.
+
+	* lib/mkmf.rb (arg_config, with_config): deal with '-' and '_'
+	  uniformly.  [ruby-dev:24118]
+
+Fri Aug 20 14:49:42 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_check_writable): no need to check read buffer if
+	  already changed to write mode.
+
+Fri Aug 20 11:46:43 2004  UENO Katsuhiro  <katsu@blue.sky.or.jp>
+
+	* ext/zlib/zlib.c: GzipReader#ungetc caused crc error.
+
+Thu Aug 19 16:29:45 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: Fail to treat a hash value of 'font' option.
+
+	* ext/tk/lib/tk.rb: bindinfo cannot return '%' substiturion information.
+
+	* ext/tk/lib/menu.rb: typo bug.
+
+Thu Aug 19 15:15:24 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (free_dir): fix memory leak.  reported by yamamoto
+	  madoka.
+
+Thu Aug 19 09:19:27 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in, win32/Makefile.sub (LIBS): need to link shell32
+	  library for SH* functions on mswin32 and mingw32.
+
+	* wince/Makefile.sub (LIBS): need to link ceshell library for SH*
+	  functions on mswince.
+
+Thu Aug 19 03:07:00 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/baseemitter.rb: folding now handles double-quoted strings,
+	  fixed problem with extra line feeds at end of folding, whitespace
+	  opening scalar blocks.
+
+	* lib/yaml/rubytypes.rb: subtelties in handling strings with
+	  non-printable characters and odd whitespace patterns.
+
+Wed Aug 18 23:44:20 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/protocol.rb (rbuf_fill): OpenSSL::SSLSocket has its own
+	  buffer, select(2) might not work. [ruby-dev:24072]
+
+Wed Aug 18 17:10:12 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/tcltklib/stubs.c (ruby_tcltk_stubs): need to call
+	  Tcl_FindExecutable() for Tcl/Tk 8.4.
+
+Wed Aug 18 12:52:55 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_obj_instance_eval): evaluates under special singleton
+	  classes as for special constants.
+
+Wed Aug 18 11:22:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (init_env): initialize HOME and USER environment
+	  variables unless set.
+
+Wed Aug 18 10:17:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (bind_eval): new method. [RCR 251]
+
+	* string.c (rb_str_clear): new method. [ruby-dev:24104]
+
+Tue Aug 17 17:20:59 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_reopen): should clear allocated OpenFile.  pointed
+	  out by Guy Decoux. [ruby-core:03288]
+
+Tue Aug 17 01:36:32 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/usage.rb: Remove extra indent. Tidy 'ri' option
+	  parsing so RDoc::usage plays better with OptionParser.
+
+Sat Aug 14 02:48:16 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/usage.rb: Added. Allows command line programs
+	  to report usage using their initial RDoc comment.
+
+Sat Aug 14 01:25:48 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/token.c: re2c no longer compiled with bit vectors.  caused
+	  problems for non-ascii characters. [ruby-core:03280]
+	* ext/syck/implicit.c: ditto.
+	* ext/syck/bytecode.c: ditto.
+
+Fri Aug 13 12:55:20 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.parse_range_header):
+	  fix regex for range-spec.
+
+	* lib/webrick/httpservlet/filehandler.rb
+	  (WEBrick::HTTPServlet::DefaultFileHandler#make_partial_content):
+	  multipart/byteranges response was broken.
+
+	* lib/xmlrpc/server.rb: refine example code.
+
+Thu Aug 12 10:54:17 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* rubyio.h (rb_eof_error): should mark as NORETURN.
+
+	* win32/win32.c (make_cmdvector): adjust escaped successive
+	  double-quote handling.
+
+Thu Aug 12 01:53:10 2004  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (read_buffered_data): extracted from rb_io_fread.
+	  (io_readpartial): new method IO#readpartial.
+	  [ruby-dev:24055]
+
+Wed Aug 11 17:17:50 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (RPATHFLAG): stop setting RPATHFLAG on Interix.
+
+Mon Aug  9 15:03:20 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservlet/erbhandler.rb
+	  (WEBrick::HTTPServlet::ERBHandler#do_GET): should select media type
+	  by suffix of script filename.
+
+Mon Aug  9 12:51:43 2004  Dave Thomas  <dave@pragprog.com>
+
+	* dir.c (dir_s_glob): Roll in Austin Ziegler's Dir.glob and
+	  fnmatch updates.
+
+Mon Aug  9 06:33:06 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (cp_r): copies symlink to symlink, except
+	  root entries of cp_r.
+
+	* lib/fileutils.rb: new method FileUtils.copy_entry.
+
+	* test/fileutils/test_fileutils.rb: more cp_r tests.
+
+Sun Aug  8 00:43:31 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/implicit.c: added sexagecimal float#base60.
+
+	* ext/syck/rubyext.c (yaml_org_handler): ditto.
+
+	* lib/token.c: indentation absolutely ignored when processing flow
+	  collections.  plain scalars are trimmed if indentation follows in
+	  an ambiguous flow collection.
+
+Sat Aug  7 03:08:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (proc_daemon): new method.  should be modified for
+	  platforms without /dev/null.
+
+Sat Aug  7 00:50:01 2004  Tanaka Akira  <akr@m17n.org>
+
+	* ext/zlib/zlib.c: Zlib::GzipReader#read(0) returns "" instead of nil.
+
+Wed Aug  4 13:26:00 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_bytes0): optimize out read(0). [ruby-talk:108276]
+
+Tue Aug  3 13:49:12 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/namespace.rb: bug fix
+
+	* ext/tk/lib/tkextlib/treectrl/tktreectrl.rb: add Tk::TreeCtrl.loupe
+
+Mon Aug  2 23:33:48 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_file.rb (test_fnmatch): added more tests.
+
+Mon Aug  2 18:04:21 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/msgcat.rb (set_translation): bug fix (fail to set
+	  trans_str to the same as src_str when trans_str is not given.)
+
+Mon Aug  2 17:40:44 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (blk_free): fixed serious memory leak. [ruby-dev:24013]
+
+Mon Aug  2 11:53:06 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/code_objects.rb (RDoc::Context::find_symbol): Fix infinite recursion
+	  looking up some top level symbols (batsman)
+
+Mon Aug  2 11:48:29 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_methods): Allow '.'s in
+	  variable names to support SWIG generated files (Hans Fugal)
+
+Sat Jul 31 23:08:00 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (is_defined): stupid mistakes fixed. [ruby-dev:24006]
+
+Sat Jul 31 17:39:47 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-expr-beg, ruby-parse-partial,
+	  ruby-calculate-indent, ruby-move-to-block, ruby-forward-sexp,
+	  ruby-backward-sexp): keywords must match word-wise.
+
+Sat Jul 31 13:37:51 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (is_defined): avoid unnecessary method invocations.
+
+Sat Jul 31 05:35:37 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb: exceptions were using an older
+	  YAML.object_maker. [ruby-core:03080]
+
+	* ext/syck/token.c (sycklex_yaml_utf8): using newline_len to
+	  handline CR-LFs.  "\000" was showing up on folded blocks which
+	  stopped at EOF.
+
+Sat Jul 31 01:25:11 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (is_defined): call is_defined() before invoking
+	  rb_eval().  [ruby-talk:107867]
+
+Fri Jul 30 16:10:54 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (lib_fromUTF8_core): raise ArgumentError when
+	  the unknown encoding name is given.
+
+	* ext/tcltklib/tcltklib.c (lib_toUTF8_core): ditto.
+
+	* ext/tk/lib/tk.rb (Tk::Encoding.encoding_convertfrom): bug fix.
+
+	* ext/tk/lib/tk.rb (Tk::Encoding.encoding_convertto): ditto.
+
+Wed Jul 28 18:59:17 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::initialize): remove at_exit code for CGI_PARAMS
+	  and CGI_COOKIES.  they will no longer be used.
+
+Wed Jul 28 15:44:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): should call rb_call_super() directly for
+	  visibility overriding.  [ruby-dev:23989]
+
+Wed Jul 28 01:04:44 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* env.h: remove argv from ruby_frame.
+
+	* eval.c (rb_eval): no more copy on write.
+
+	* eval.c (assign): ditto.
+
+	* eval.c (rb_call0): can receive *rest by specifying negative
+	  argc.  (-1 means 0 arg and *rest, -2 means 1 arg and *rest...)
+
+	* eval.c (rb_call0): properly set frame's argc counter.
+
+	* gc.c (rb_gc_mark_frame): need not to mark frame's argv
+
+	* gc.c (run_final): wrong order of data. [ruby-dev:23984]
+
+Tue Jul 27 07:05:04 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): copy on write for argument local variable
+	  assignment.
+
+	* eval.c (assign): ditto.
+
+	* eval.c (rb_call0): update ruby_frame->argv with the default
+	  value used for the optional arguments.
+
+	* object.c (Init_Object): "===" calls rb_obj_equal() directly.
+	  [ruby-list:39937]
+
+Mon Jul 26 11:22:55 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils.escape): should
+	  escape space.
+
+Sun Jul 25 10:56:28 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* win32/win32.{h,c} (rb_w32_{f,fd,fs}open): workaround for bcc32's
+	  {f,fd,fs}open bug. set errno EMFILE and EBADF. [ruby-dev:23963]
+
+	* test/drb/drbtest.rb: fix method duplication.
+
+Sat Jul 24 13:32:47 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (rb_range_beg_len): returns Qnil only when "beg" points
+	  outside of a range.  No boundary check for "end".
+
+Fri Jul 23 16:40:25 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (define_final): should not disclose NODE* to Ruby world.
+	  [ruby-dev:23957]
+
+Fri Jul 23 08:52:22 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (disconnected?): new method.
+
+Thu Jul 22 16:41:54 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi/session.rb (CGI::Session::FileStore#update): sets the
+	  permission of the session data file to 0600.
+
+	* lib/cgi/session/pstore.rb (CGI::Session::Pstore#initialize):
+	  ditto.
+
+Mon Jul 19 00:53:46 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb
+	  (WEBrick::HTTPServlet::CGIhandler#do_GET): set SystemRoot environment
+	  variable to CGI process on Windows native platforms. [ruby-dev:23936]
+
+Sun Jul 18 16:14:29 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/msgcat.rb (TkMsgCatalog.callback): bug fix
+	  ( wrong number of argument )
+
+Sun Jul 18 08:13:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): remove extra sign digit.
+
+Sun Jul 18 03:19:14 2004  Akinori MUSHA  <knu@iDaemons.org>
+
+	* dir.c (bracket): use NULL instead of 0.
+
+Sun Jul 18 02:35:30 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (receive_responses): return if a LOGOUT response
+	  received.
+
+Sat Jul 17 23:59:01 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (send_string_data): wait command continuation
+	  requests before sending octet data of literals.
+
+Sat Jul 17 23:54:59 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/variable.rb: TkVariable#ref returns a TkVariable object
+
+Sat Jul 17 22:04:44 2004  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/ldap.rb: method hierarchical? should be in URI::LDAP.
+
+Sat Jul 17 18:29:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt): not to show same error messages twice.
+
+Sat Jul 17 14:18:11 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_match_m): String#match should also take
+	  optional argument.  [ruby-core:03205]
+
+	* re.c (rb_reg_match_m): add optional second argugment "pos" to
+	  specify match start point.  [ruby-core:03203]
+
+Sat Jul 17 13:13:32 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (RubyLex::identify_string): %s string do not
+	  process expression interpolation.  [ruby-talk:106691]
+
+Sat Jul 17 05:26:27 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/diagram.rb: Incorporate Micheal Neumann's
+	  client-side imagemao patch
+
+Sat Jul 17 01:57:03 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (THREAD_ALLOC): th->thread should be initialized to NULL.
+	  [ruby-talk:106657]  The solution was found by Guy Decoux.
+
+Fri Jul 16 22:30:28 2004  Michael Neumann  <mneumann@ntecs.de>
+
+	* file.c (rb_stat_dev_major): new methods File::Stat#dev_major and
+	  #dev_minor. [ruby-core:03195]
+
+Fri Jul 16 11:17:38 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (exit_initialize): use EXIT_SUCCESS instead of 0.
+	  [ruby-dev:23913]
+
+	* error.c (exit_success_p): new method SystemExit#success?.
+	  [ruby-dev:23912]
+
+	* error.c (syserr_initialize): initialization for subclasses.
+	  [ruby-dev:23912]
+
+Thu Jul 15 23:53:38 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#warn, OptionParser#abort): Exception
+	  no longer has to_str method.
+
+Thu Jul 15 22:59:48 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/extconf.rb: added dir_config for curses, ncurses,
+	  termcap.
+
+Thu Jul 15 20:44:46 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* class.c: rdoc patch
+
+Thu Jul 15 14:12:34 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml.rb (YAML::load_file, YAML::parse_file): added.
+
+	* ext/syck/token.c: re2c compiled with bit vectors now.
+	* ext/syck/implicit.c: ditto.
+	* ext/syck/bytecode.c: ditto.
+
+Thu Jul 15 10:15:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/, ext/tcltklib/: bug fix
+
+	* ext/tk/lib/tk.rb: better operation for SIGINT when processing
+	  callbacks.
+	* ext/tk/lib/tk/msgcat.rb: ditto.
+	* ext/tk/lib/tk/variable.rb: ditto.
+	* ext/tk/lib/tk/timer.rb: ditto.
+
+	* ext/tk/lib/tk/validation.rb: add Tk::ValidateConfigure.__def_validcmd
+	  to define validatecommand methods easier
+
+	* ext/tk/lib/tk.rb (_genobj_for_tkwidget): support autoload Tk ext
+	  classes
+
+	* ext/tk/lib/tk/canvas.rb and so on: remove the parent widget type
+	  check for items (e.g. canvas items; depends on the class) to
+	  avoid some troubles on Tk extension widget class definition.
+
+	* ext/tk/lib/tkextlib/: add Iwidget and TkTable extension support
+
+	* ext/tk/sample/tkextlib/: add samples of Iwidget and TkTable
+
+Wed Jul 14 23:49:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (enum_min_by): new method Enum#min_by.  added Enum#max_by
+	  as well.
+
+Wed Jul 14 18:05:21 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1cons_to_der): fix type of
+	  argument. [ruby-dev:23891]
+
+	* test/openssl/test_x509store.rb: prune tests for CRL checking
+	  unless X509::V_FLAG_CRL_CHECK is defined.
+
+Wed Jul 14 12:20:05 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* util.c (ruby_strtod): should not convert string in the form of
+	  "-I.FE-X" which both "I" and "F" are omitted. [ruby-dev:23883]
+
+	* test/ruby/test_float.rb (test_strtod): add test for bug fix.
+
+Wed Jul 14 00:33:48 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* array.c: rdoc patch. merged patch from Johan Holmberg
+	  <holmberg@iar.se> [ruby-core:3170]
+
+Tue Jul 13 19:39:12 2004  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#merge_path):
+	  "URI('http://www.example.com/foo/..') + './'" should return
+	  "URI('http://www.example.com/')".  [ruby-list:39838]
+	  "URI('http://www.example.com/') + './foo/bar/..'" should return
+	  "URI('http://www.example.com/foo/')".  [ruby-list:39844]
+
+	* test/uri/test_generic.rb (TestGeneric#test_merge): added tests.
+
+Tue Jul 13 15:48:56 2004  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/mkmf.rb (init_mkmf): Do not add $(libdir) to $LIBPATH in
+	   extmk mode.
+
+	* lib/mkmf.rb (dir_config): Prepend a new library path instead of
+	  appending so it is tried first.
+
+Tue Jul 13 00:50:48 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb: Support call-seq: for Ruby files.
+
+Mon Jul 12 21:20:51 2004  Dave Thomas  <dave@pragprog.com>
+
+	* html_generator.rb: Support hyperlinks of the form {any text}[xxx]
+	  as well as stuff[xxx]
+
+Sat Jul 10 09:30:24 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/marshal/test_struct.rb: use qualified built-in class name
+	  (::Struct) to avoid name crash.
+
+Sat Jul 10 04:21:56 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: better operation for SIGINT when processing
+	  callbacks.
+	* ext/tk/lib/tk/msgcat.rb: ditto.
+	* ext/tk/lib/tk/variable.rb: ditto.
+	* ext/tk/lib/tk/timer.rb: ditto.
+
+	* ext/tk/lib/tk/validation.rb (__def_validcmd):  add a module
+	  function of Tk::ValidateConfigure to define validatecommand
+	  methods easier
+
+Fri Jul  9 22:18:59 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* array.c, enum.c, pack.c: rdoc patch from Johan Holmberg
+	  <holmberg@iar.se> [ruby-core:3132] [ruby-core:3136]
+
+	* numeric.c: rdoc patch.
+
+Fri Jul  9 19:26:39 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::HTTPS#proxy_open): raise ArgumentError to
+	  notice https is not supported.
+
+Fri Jul  9 14:28:54 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_raise): accept third argument as well as
+	  Kernel#raise, and evaluate the arguments to create an exception in
+	  the caller's context.  [ruby-talk:105507]
+
+Fri Jul  9 01:47:08 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib: bug fix
+	* ext/tk/lib/tkextlib/itcl: add [incr Tcl] support
+	* ext/tk/lib/tkextlib/itk: add [incr Tk] support
+	* ext/tk/lib/tkextlib/iwidgets: midway point of [incr Widgets] support
+	* ext/tk/sample/tkextlib/iwidgets: very simple examples of
+	  [incr Widgets]
+
+Thu Jul  8 19:27:16 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_stack_end_address): detect stack end address
+	  variable supplied by system.  [ruby-core:03115]
+
+	* gc.c (Init_stack): use system provided address if possible.
+
+Thu Jul  8 00:05:23 2004  akira yamada  <akira@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile::initialize): got out code of
+	  generating tmpname.  [ruby-dev:23832][ruby-dev:23837]
+
+Wed Jul  7 02:31:41 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/{rss,parser,0.9,1.0,2.0}.rb: supported RSS 0.9x/2.0
+	  validation and validation which disregard order of elements.
+	* test/rss/test_parser.rb: added tests for RSS 0.9x/2.0
+	  validation.
+	* test/rss/{test_trackback,rss-testcase}.rb: fixed no good method
+	  name.
+
+Wed Jul  7 00:48:34 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/tk/lib/tkextlib/tktrans.rb,
+	  ext/tk/lib/tkextlib/treectrl.rb: fix syntax errors.
+
+Tue Jul  6 18:38:45 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib: improve framework of developping Tcl/Tk extension
+	  wrappers
+
+Mon Jul  5 23:56:42 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/{trackback,syndication,dublincore,content}.rb: worked
+	  with ruby 1.6 again.
+
+	* test/rss/rss-assertions.rb: ditto.
+
+Mon Jul  5 22:54:39 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/uri/common.rb (Kernel#URI): new global method for parsing URIs.
+
+Mon Jul  5 09:02:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_yield, rb_f_catch): 4th argument to rb_yield_0()
+	  is a set of bit flags.  [ruby-dev:23859]
+
+Mon Jul  5 01:20:17 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* lib/drb/drb.rb(DRbConn self.open): If socket pool is full, close
+	  the socket whose last-access-time is oldest. (and add new one)
+	  [ruby-dev:23860]
+
+Sun Jul  4 12:24:50 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb: added copyright header.
+
+Sat Jul  3 22:25:27 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* added files:
+	  * lib/soap/header/*
+	  * lib/soap/rpc/httpserver.rb
+	  * lib/wsdl/soap/cgiStubCreator.rb
+	  * lib/wsdl/soap/classDefCreator.rb
+	  * lib/wsdl/soap/classDefCreatorSupport.rb
+	  * lib/wsdl/soap/clientSkeltonCreator.rb
+	  * lib/wsdl/soap/driverCreator.rb
+	  * lib/wsdl/soap/mappingRegistryCreator.rb
+	  * lib/wsdl/soap/methodDefCreator.rb
+	  * lib/wsdl/soap/servantSkeltonCreator.rb
+	  * lib/wsdl/soap/standaloneServerStubCreator.rb
+	  * lib/wsdl/xmlSchema/enumeration.rb
+	  * lib/wsdl/xmlSchema/simpleRestriction.rb
+	  * lib/wsdl/xmlSchema/simpleType.rb
+	  * lib/xsd/codegen/*
+	  * lib/xsd/codegen.rb
+	  * sample/soap/authheader/*
+	  * sample/soap/raa2.4/*
+	  * sample/soap/ssl/*
+	  * sample/soap/swa/*
+	  * sample/soap/whois.rb
+	  * sample/wsdl/raa2.4/*
+	  * test/soap/header/*
+	  * test/soap/ssl/*
+	  * test/soap/struct/*
+	  * test/soap/swa/*
+	  * test/soap/wsdlDriver/*
+	  * test/wsdl/multiplefault.wsdl
+	  * test/wsdl/simpletype/*
+	  * test/wsdl/test_multiplefault.rb
+
+	* modified files:
+	  * lib/soap/baseData.rb
+	  * lib/soap/element.rb
+	  * lib/soap/generator.rb
+	  * lib/soap/netHttpClient.rb
+	  * lib/soap/parser.rb
+	  * lib/soap/property.rb
+	  * lib/soap/soap.rb
+	  * lib/soap/streamHandler.rb
+	  * lib/soap/wsdlDriver.rb
+	  * lib/soap/encodingstyle/handler.rb
+	  * lib/soap/encodingstyle/literalHandler.rb
+	  * lib/soap/encodingstyle/soapHandler.rb
+	  * lib/soap/mapping/factory.rb
+	  * lib/soap/mapping/mapping.rb
+	  * lib/soap/mapping/registry.rb
+	  * lib/soap/mapping/rubytypeFactory.rb
+	  * lib/soap/mapping/wsdlRegistry.rb
+	  * lib/soap/rpc/cgistub.rb
+	  * lib/soap/rpc/driver.rb
+	  * lib/soap/rpc/proxy.rb
+	  * lib/soap/rpc/router.rb
+	  * lib/soap/rpc/soaplet.rb
+	  * lib/soap/rpc/standaloneServer.rb
+	  * lib/wsdl/data.rb
+	  * lib/wsdl/definitions.rb
+	  * lib/wsdl/operation.rb
+	  * lib/wsdl/parser.rb
+	  * lib/wsdl/soap/definitions.rb
+	  * lib/wsdl/xmlSchema/complexContent.rb
+	  * lib/wsdl/xmlSchema/complexType.rb
+	  * lib/wsdl/xmlSchema/data.rb
+	  * lib/wsdl/xmlSchema/parser.rb
+	  * lib/wsdl/xmlSchema/schema.rb
+	  * lib/xsd/datatypes.rb
+	  * lib/xsd/qname.rb
+	  * sample/soap/sampleStruct/server.rb
+	  * sample/wsdl/amazon/AmazonSearch.rb
+	  * sample/wsdl/amazon/AmazonSearchDriver.rb
+	  * test/soap/test_property.rb
+	  * test/soap/calc/test_calc_cgi.rb
+	  * test/wsdl/test_emptycomplextype.rb
+
+	* summary
+	  * add SOAP Header mustUnderstand support.
+
+	  * add HTTP client SSL configuration and Cookies support (works
+	    completely with http-access2).
+
+	  * add header handler for handling sending/receiving SOAP Header.
+
+	  * map Ruby's anonymous Struct to common SOAP Struct in SOAP Object
+	    Model.  it caused error.
+
+	  * add WSDL simpleType support to restrict lexical value space.
+
+Sat Jul  3 17:19:44 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/tk/lib/tkextlib/tkDND.rb: fix syntax error.
+
+Thu Jul  1 18:36:08 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tcltklib: bug fix
+
+	* ext/tk/lib/tk: bug fix and add Tcl/Tk extension support libraries
+
+Thu Jul  1 18:31:31 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pstore.rb (PStore#transaction): get rid of opening in write mode
+	  when read only transaction.  [ruby-dev:23842]
+
+Thu Jul  1 00:44:42 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_encrypt, ossl_cipher_decrypt):
+	  re-implemnt (the arguments for this method is ).
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_pkcs5_keyivgen): new method
+	  OpenSSL::Cipher::Cipher#pkcs5_keyivgen. it calls EVP_BytesToKey().
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_set_key_length): new method
+	  OpenSSL::Cipher::Cipher#key_len=.
+
+Wed Jun 30 19:48:09 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: check for EVP_CIPHER_CTX_copy, ENGINE_add,
+	  EVP_CIPHER_CTX_set_padding, EVP_CipherFinal_ex, EVP_CipherInit_ex,
+	  EVP_DigestFinal_ex and EVP_DigestInit_ex.
+
+	* ext/openssl/openssl_missing.c (EVP_CIPHER_CTX_copy): new function.
+
+	* ext/openssl/openssl_missing.h (EVP_DigestInit_ex, EVP_DigestFinal_ex,
+	  EVP_CipherInit_ex, EVP_CipherFinal_ex, HMAC_Init_ex): new macro for
+	  OpenSSL 0.9.6.
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_alloc, ossl_cipher_initialize,
+	  ossl_cipher_copy, ossl_cipher_reset, ossl_cipher_encrypt,
+	  ossl_cipher_decrypt, ossl_cipher_final, ossl_cipher_set_key,
+	  ossl_cipher_set_iv): replace all EVP_CipherInit and
+	  EVP_CipherFinal into EVP_CipherInit_ex and EVP_CipherFinal_ex.
+	  and EVP_CIPHER_CTX_init should only be called once.
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_set_padding): check for
+	  EVP_CIPHER_CTX_set_padding.
+
+	* ext/openssl/ossl_cipher.c (Init_ossl_cipher): Cipher#<< is deprecated.
+
+	* ext/openssl/ossl_digest.c: replace all EVP_DigestInit and
+	  EVP_DigestFinal into EVP_DigestInit_ex and EVP_DigestFinal_ex.
+	  and EVP_MD_CTX_init should only be called once.
+
+	* ext/openssl/ossl_digest.c (digest_final): should call
+	  EVP_MD_CTX_cleanup to avoid memory leak.
+
+	* ext/openssl/ossl_hmac.c (ossl_hmac_initialize): repalce HMAC_init
+	  into HMAC_init_ex. and HMAC_CTX_init is moved to ossl_hmac_alloc.
+
+	* ext/openssl/ossl_hmac.c (hmac_final): should call
+	  HMAC_CTX_cleanup to avoid memory leak.
+
+	* test/openssl/test_cipher.rb, test/openssl/test_digest.rb,
+	  test/openssl/test_hmac.rb: new file.
+
+Wed Jun 30 16:59:39 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_file.rb (test_fnmatch): some tests for File.fnmatch
+	  are added.
+
+Wed Jun 30 11:38:51 2004  Mikael Brockman  <phubuh@phubuh.org>
+
+	* parse.y (primary): should not be NULL.  [ruby-core:03098]
+
+Wed Jun 30 02:41:10 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_emitter_new): set buffer after
+	  Data_Wrap_Struct to avoid possible GC. [ruby-talk:104835]
+
+Tue Jun 29 10:31:19 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval_cmd, rb_thread_trap_eval): restore safe level.
+
+	* gc.c (define_final, run_final): preserve and restore safe level for
+	  finalizers.  [ruby-core:03058]
+
+	* signal.c (signal_exec, rb_trap_exit, trap): preserve and restore
+	  safe level for signal handlers.  [ruby-dev:23829]
+
+Mon Jun 28 14:57:56 2004  Jeff Mitchell  <quixoticsycophant@yahoo.com>
+
+	* configure.in, lib/mkmf.rb (LIBPATHFLAG): use double quotes due to
+	  DOSISH compilers.  [ruby-core:03107]
+
+Mon Jun 28 00:35:29 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* sample/drb/*.rb: using 'DRb.thread.join' instead of 'gets'
+
+Sun Jun 27 22:36:47 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/tdiary_plugin/rss-recent.rb: supported Hiki.
+
+Sat Jun 26 15:17:11 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_mod_class_variables): class variables are no longer
+	  inherited.  [ruby-dev:23808]
+
+Sat Jun 26 11:07:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (aix): -b must come at the start of the command line,
+	  and -e must not appear while testing libraries.  [ruby-talk:104501]
+
+	* lib/mkmf.rb (find_header, dir_config): quote directory names if
+	  necessary.  [ruby-talk:104505]
+
+Sat Jun 26 00:13:08 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_fopen, rb_fdopen, rb_io_reopen): setvbuf() may return
+	  positive value on failure.  [ruby-dev:23792]
+
+Fri Jun 25 18:07:15 2004  Michal Rokos  <michal@ruby-lang.org>
+
+	* gc.c: bring back _stklen for DJGPP [ruby-core:3084]
+
+Fri Jun 25 15:33:01 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/extconf.rb: check stricter.  [ruby-talk:104501]
+
+Fri Jun 25 01:58:01 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* {bcc32,win32,wince}/setup.mak: remove RUBY_EXTERN lines when
+	  including version.h. [ruby-talk:104456]
+
+Thu Jun 24 14:23:29 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_fread): return already read data when system call is
+	  interrupted.  [ruby-talk:97206]
+
+Thu Jun 24 01:25:21 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* version.h: added declarations of ruby_version,
+	  ruby_release_date, ruby_platform.
+
+Thu Jun 24 01:07:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_sockaddr): Socket#gethostbyname()
+	  should give us packed address, not struct sockaddr.
+	  [ruby-core:03053]
+
+Wed Jun 23 22:19:10 2004  Dave Thomas  <dave@pragprog.com>
+
+	* ext/socket/socket.c (sock_s_gethostbyaddr): Work around problem
+	  with OS X not returning 'from' parameter to recvfrom for
+	  connection-oriented sockets.
+
+Wed Jun 23 22:16:16 2004  Michal Rokos  <michal@ruby-lang.org>
+
+	* io.c: io_seek()'s retval should be checked [ruby-core:03045]
+
+Wed Jun 23 21:48:27 2004  Michal Rokos  <michal@ruby-lang.org>
+
+	* time.c: Fix indentation.
+
+	* main.c: Remove _stklen, and _CRT_glob. Move _stacksize for
+	  __human68k__ to gc.c where the others are.
+
+	* gc.c: put _stacksize in place and clean the #ifdefs macros.
+
+Wed Jun 23 17:37:54 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: added new option --ssl.
+
+Wed Jun 23 01:45:27 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RubyLex::identify_quotation):
+	  Fix problem with the 'r' being dropped from %r{xxx}
+
+Wed Jun 23 00:10:17 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_hresult2msg): remove trailing
+	  CRs and LFs. (doesn't depend on CR+LF) [ruby-dev:23749]
+
+Wed Jun 23 00:00:25 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (return_jump, break_jump): raise unexpected local jump
+	  exception directly.  [ruby-dev:23740]
+
+	* io.c (rb_io_initialize): should check fcntl result.  [ruby-dev:23742]
+
+Tue Jun 22 23:35:43 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): support FZERO and FSPACE with NaN/Inf.
+
+	* test/ruby/test_sprintf.rb (test_nan, test_inf): add tests.
+
+Tue Jun 22 21:11:36 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (OLE_FREE): should not call CoFreeUnuse-
+	  dLibraries().
+
+	* ext/win32ole/win32ole.c (ole_event_free): ditto.
+
+	* ext/win32ole/win32ole.c (ole_hresult2msg): truncate error message
+	  before CR.
+
+Tue Jun 22 19:24:59 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): unify output of NaN, Inf and -Inf with
+	  "%f" or etc on all platform. [ruby-dev:23704], [ruby-dev:23747]
+
+Tue Jun 22 15:28:12 2004  Michal Rokos  <michal@ruby-lang.org>
+
+	* compar.c: Remove explicit NIL_P() checks since rb_cmpint() does it
+	  again in the exactly same manner.
+
+Tue Jun 22 01:32:40 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_pkey_dh.c (ossl_dh_initialize): should create
+	  empty pkey object if no argument is passed. [ruby-talk:103328]
+
+	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_initialize): ditto.
+
+	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_initialize): ditto.
+
+	* ext/openssl/ossl_pkey_dh.c: add new methods: OpenSSL::PKey::DH#p,
+	  OpenSSL::PKey::DH#p=, OpenSSL::PKey::DH#g, OpenSSL::PKey::DH#g=,
+	  OpenSSL::PKey::DH#pub_key, OpenSSL::PKey::DH#pub_key=,
+	  OpenSSL::PKey::DH#priv_key and OpenSSL::PKey::DH#priv_key=.
+
+	* ext/openssl/ossl_pkey_dsa.c: add new methods: OpenSSL::PKey::DSA#p,
+	  OpenSSL::PKey::DSA#p=, OpenSSL::PKey::DSA#q, OpenSSL::PKey::DSA#q=,
+	  OpenSSL::PKey::DSA#g, OpenSSL::PKey::DSA#g=,
+	  OpenSSL::PKey::DSA#pub_key, OpenSSL::PKey::DSA#pub_key=,
+	  OpenSSL::PKey::DSA#priv_key and OpenSSL::PKey::DSA#priv_key=.
+
+Mon Jun 21 09:24:51 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_opendir): should set errno if error occurs
+	  when calling OS API.
+
+Sun Jun 20 21:12:54 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (binary=): send TYPE commands only once.
+
+Sat Jun 19 13:27:01 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (method_call): allow changing $SAFE.  [ruby-dev:23713]
+
+	* eval.c (proc_set_safe_level, proc_invoke, rb_mod_define_method): not
+	  set $SAFE for methods defined from Proc.  [ruby-dev:23697]
+
+Sat Jun 19 01:10:12 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* sample/rss/tdiary_plugin/rss-recent.rb: added more information.
+
+Fri Jun 18 23:12:22 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (proc_save_safe_level, rb_set_safe_level, safe_setter): limit
+	  safe level.
+
+Wed Jun 16 23:05:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_mod_freeze): prepare string representation before
+	  freezing. [ruby-talk:103646]
+
+Wed Jun 16 19:57:24 2004  Michal Rokos  <michal@ruby-lang.org>
+
+	* test/ruby/test_array.rb: extend testcase to check #first, #last,
+	  #shift, #unshift, #pop, #push
+
+Wed Jun 16 16:05:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_new): move alloc behind checks.  [ruby-core:02982]
+
+	* array.c (rb_ary_pop_m, rb_ary_shift_m): take arg to behave as push
+	  and unshift.
+
+	* array.c (rb_ary_first, rb_ary_last): make shared array for result
+	  array, and correct doc for Array#first(n) and Array#last(n)
+
+	* array.c (rb_ary_select): not accept any arg.
+
+Wed Jun 16 16:03:59 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_class_inherited_p): singleton class inherits Class
+	  rather than its object's class.  [ruby-dev:23690]
+
+Wed Jun 16 16:01:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (stack_grow_direction): memoize the direction.
+
+	* gc.c (Init_stack): should always move to end of VALUE.
+
+Tue Jun 15 12:10:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: bug fix (TkWindow#grab)
+
+Mon Jun 14 18:23:27 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/remote-tk.rb: bug fix
+
+Sun Jun 13 00:23:04 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/extconf.rb: [EXPERIMENTAL] MacOS X (darwin) support
+
+	* ext/tcltklib/tcltklib.c: fix thread trouble on callback proc, and
+	  eliminate warning about instance variable access
+
+	* ext/tk/lib/tk/menubar.rb: improve supported menu_spec
+
+	* ext/tk/lib/tk/menuspec.rb: [add] menu_spec support library
+
+	* ext/tk/lib/tk/root.rb: add menu_spec support
+
+	* ext/tk/lib/tk/text.rb: bug fix
+
+	* ext/tk/lib/tk/toplevel.rb: add menu_spec support
+
+	* ext/tk/sample/menubar?.rb: [add] sample of menu_spec usage
+
+Sat Jun 12 14:15:20 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c: RDOC for File::FNM_CASEFOLD was missed.
+
+Sat Jun 12 11:15:53 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (target_os): strip -gnu suffix on Linux.
+
+Fri Jun 11 22:08:50 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* array.c: remove #indexes, #indices.
+
+	* hash.c: ditto.
+
+	* ext/dbm/dbm.c: remove #indexes, #indices, "values_at" warning
+	  from #select.
+
+	* ext/gdbm/gdbm.c: ditto.
+
+	* ext/sdbm/init.c: ditto.
+
+	* ext/dbm/dbm.c (Init_dbm): set VERSION constant as "unknown" when
+	  DB_VERSION_STRING is not available.
+
+Thu Jun 10 19:19:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/sdbm/init.c (fsdbm_store): sdbm should use StringValue().
+	  [ruby-talk:103062]
+
+Wed Jun  9 16:09:01 2004  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic::merge,
+	  URI::Generic::route_from): accepts non-hierarchical URI.
+	  [ruby-dev:23631]
+
+	* test/uri/test_generic.rb (TestGeneric::test_route,
+	  TestGeneric::test_merge): added tests for above changes.
+
+Wed Jun  9 15:39:55 2004  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in: Add support for DragonFly BSD.
+
+Wed Jun  9 15:07:06 2004  Akinori MUSHA  <knu@iDaemons.org>
+
+	* config.guess, config.sub: Update to a more recent version as of
+	  2004-01-20.
+
+Wed Jun  9 11:20:05 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c: remove unused functions and variables.
+
+Wed Jun  2 20:16:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_new4): should share shared instance if it already
+	  exists.  [ruby-dev:23665]
+
+Wed Jun  2 12:41:53 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_gets_m): set lastline ($_) even when read line is
+	  nil.  [ruby-dev:23663]
+
+Fri May 28 11:20:31 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): bad influence on frame node.
+
+	* eval.c (eval): reverted wrongly removed condition.  [ruby-dev:23638]
+
+Thu May 27 21:37:50 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#initialize): fix pathname initialization
+	  by pathname.
+
+Thu May 27 20:02:09 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* io.c (rb_io_fwrite): check all case errno != 0 [ruby-dev:23648]
+
+Thu May 27 15:54:02 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (MDTM_REGEXP): fix for demon's ftp server.
+	  Thanks, Rutger Nijlunsing.
+
+Thu May 27 14:53:13 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* io.c (rb_io_fwrite): workaround for bcc32's fwrite bug.
+	  add errno checking.  [ruby-dev:23627]
+
+	* io.c (rb_io_fwrite): should check if errno == ENOENT, too.
+
+Thu May 27 11:25:03 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/csv/test_csv.rb: illegal require module name (../lib/csv.rb).
+
+Wed May 26 23:12:13 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb (CSV.read, CSV.readlines): added.  works as IO.read and
+	  IO.readlines in CSV format.
+
+	* lib/csv.rb (CSV.parse): [CAUTION] behavior changed.  in the past,
+	  CSV.parse accepts a filename to be read-opened (it was just a
+	  shortcut of CSV.open(filename, 'r')).  now CSV.parse accepts a
+	  string or a stream to be parsed e.g.
+	  CSV.parse("1,2\n3,r") #=> [['1', '2'], ['3', '4']]
+
+	* test/csv/test_csv.rb: follow above changes.
+
+Wed May 26 14:19:42 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval, eval): make line number consistent on eval with
+	  Proc.  [ruby-talk:101253]
+
+Wed May 26 13:59:17 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::skip_for_variable): Allow for
+	  'do' after for statement
+
+Wed May 26 13:56:03 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb (Generators::MarkUp::style_url): Fix
+	  relative path to code CSS file
+
+Wed May 26 13:14:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_init_copy): copy also positions.  [ruby-talk:100910]
+
+Wed May 26 00:00:00 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/syck.c (syck_new_parser): clear parser on init.
+	  thanks, ts. [ruby-core:02931]
+
+	* ext/syck/token.c (sycklex_yaml_utf8): buffer underflow.
+	  thanks, ts. [ruby-core:02929]
+
+	* lib/yaml/baseemitter.rb (indent_text): simpler flow block code.
+
+Tue May 25 11:54:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_yield_0, proc_invoke, proc_arity): allow passing a block
+	  to a Proc.  [ruby-dev:23533]
+
+	* parse.y (block_par, block_var): ditto.
+
+Tue May 25 01:50:17 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_i2d_ASN1_TYPE, ossl_ASN1_TYPE_free):
+	  workaround for the versions earlier than OpenSSL-0.9.7.
+
+Mon May 24 10:46:26 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/rdoc/generators/template/html/html.rb: SYSTEM identifiers
+	  must be absolute URIs
+
+Sun May 23 04:53:50 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/pstore.rb (transaction): allow overriding dump and load.
+	  [ruby-dev:23567]
+
+	* lib/yaml/store.rb: follow lib/pstore.rb's change.
+
+Sat May 22 11:54:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* MANIFEST: add test/openssl/test_x509store.rb.
+
+	* ext/tk/MANIFEST: add recent files.
+
+Sat May 22 05:37:11 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/remote-tk.rb: (NEW library) control Tk interpreters
+	  on the other processes by Tcl/Tk's 'send' command
+
+Fri May 21 09:22:05 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method_parameters):
+	  Add ()'s around parameters that don't have them
+
+Fri May 21 02:21:11 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb: fixed a few bugs around multi char record/field separator.
+
+	* test/csv/test_csv.rb: added boundary test for above feature.
+
+Thu May 20 17:02:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (check_sizeof): define result size.  [ruby-core:02911]
+
+	* lib/mkmf.rb (create_header): macro name should not include equal
+	  sign.
+
+Thu May 20 14:35:52 2004  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/socket.c: check SCM_RIGHTS macro addition to
+	  the msg_control field to test existence of file descriptor passing
+	  by msg_control.
+
+Thu May 20 12:38:06 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_eq): always check if operands are NaN.
+	  [ruby-list:39685]
+
+Thu May 20 12:34:39 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_visibility):
+	  At Ryan Davis' suggestion, honor visibility modifers if guarded by a
+	  statement modifier
+
+Thu May 20 12:22:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_type): do not check pointer to incomplete type,
+	  which always get compiled.
+	  [ruby-list:39683]
+
+Wed May 19 23:45:43 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/inlinetest.rb (InlineTest::loadtest): requiring library with
+	  replaced $0 can make $0 == __FILE__ block be evaluated twice.
+
+	* test/ruby/envutil.rb (EnvUtil::rubybin): give priority to
+	  environment variable.  [ruby-dev:23538]
+
+Wed May 19 11:08:10 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: change permition of TkObject#tk_send from
+	  private to public
+
+Wed May 19 02:29:36 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: support TRACE.
+
+Wed May 19 02:21:53 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: do not use class variables.
+
+Tue May 18 21:21:43 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb: writes lines with "\n" when row separator is not given.
+	  formerly it was "\r\n".
+
+	* lib/csv.rb: [CAUTION] API change
+
+	  * CSV::Row removed.  a row is represented as just an Array.  since
+	    CSV::Row was a subclass of Array, it won't hurt almost all programs
+	    except one which depended CSV::Row#match.
+
+	  * CSV::Cell removed.  a cell is represented as just a String or
+	    nil(NULL).  this change will cause widespread destruction.
+
+	      CSV.open("foo.csv", "r") do |row|
+		row.each do |cell|
+		  if cell.is_null       # Cell#is_null
+		    p "(NULL)"
+		  else
+		    p cell.data		# Cell#data
+		  end
+		end
+	      end
+
+	    must be just;
+
+	      CSV.open("foo.csv", "r") do |row|
+		row.each do |cell|
+		  if cell.nil?
+		    p "(NULL)"
+		  else
+		    p cell
+		  end
+		end
+	      end
+
+	* lib/csv.rb: [CAUTION] record separator(CR, LF, CR+LF) behavior
+	  change.  CSV.open, CSV.parse, and CSV,generate now do not force
+	  opened file binmode.  formerly it set binmode explicitly.
+
+	  with CSV.open, binmode of opened file depends the given mode
+	  parameter "r", "w", "rb", and "wb".  CSV.parse and CSV.generate open
+	  file with "r" and "w".
+
+	  setting mode properly is user's responsibility now.
+
+	* lib/csv.rb: accepts String as a fs (field separator/column separator)
+	  and rs (record separator/row separator)
+
+	* lib/csv.rb: added CSV.foreach(path, rs = nil, &block).  CSV.foreach
+	  now does not handle "| cmd" as a path different from IO.foreach.
+	  needed?
+
+	* test/csv/test_csv.rb: updated.
+
+Tue May 18 14:24:20 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml.rb: added rdoc to beginning of lib.
+
+Tue May 18 14:00:46 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.h (NEW_DSTR): adjust list length.
+
+	* parse.y (literal_concat): ditto.
+
+Tue May 18 09:30:25 2004  SASADA Koichi  <ko1@atdot.net>
+
+	* eval.c (rb_method_node): search cache entry first.
+
+Mon May 17 16:04:06 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* numeric.c (flo_to_s): it's preferable that "p 0.0" outputs "0.0"
+	  instead of "0.0e+00". [ruby-dev:23480]
+
+	* numeric.c (flo_to_s): it's preferable that "p 0.00000000000000000001"
+	  outputs "1.0e-20" instead of "9.999999999999999e-21". (the precision
+	  is considered, but there is assumption DBL_DIG == 15 in current
+	  implementation)
+
+Mon May 17 10:13:33 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (setup_domain_and_type): honor duck typing.
+	  [ruby-dev:23522]
+
+	* ext/socket/socket.c (sock_s_getnameinfo): ditto.
+
+Mon May 17 00:36:21 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/baseemitter.rb (indent_text): was forcing a mod value
+	  of zero at times, which kept some blocks from getting indentation.
+
+Mon May 17 00:07:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/drb/drb.rb: Cosmetic documentation changes.
+
+Sun May 16 20:55:49 2004  Tanaka Akira  <akr@m17n.org>
+
+	* ext/dbm/dbm.c (fdbm_initialize): accept optional 3rd argument to
+	  specify an open flag.
+	  (Init_dbm): define open flags: DBM::READER, DBM::WRITER, DBM::WRCREAT
+	  and DBM::NEWDB.
+
+Sat May 15 17:52:24 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_float.rb(test_strtod): Add test for signed 0.000...1
+
+Sat May 15 14:20:13 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/syck/depend: add ruby's headers.
+
+Sat May 15 13:38:33 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/MANIFEST, ext/syck/depend: new file.
+
+	* lib/yaml/rubytypes.rb: range of exponential floats.  [ruby-core:02824]
+
+	* test/yaml/test_yaml.rb: tests for strings start with colon and some
+	  round trip.
+
+Sat May 15 12:04:58 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml.rb: removed fallback to pure Ruby parser.
+
+	* lib/yaml/baseemitter.rb (node_text): rewriting folded scalars.
+
+	* ext/syck/syck.h: reports style of scalars now, be they plain, block
+	  single-, or double-quoted.
+
+	* ext/syck/syck.c: ditto.
+
+	* ext/syck/gram.c: ditto.
+
+	* ext/syck/node.c: ditto.
+
+	* ext/syck/token.c: ditto.
+
+	* ext/syck/rubyext.c (yaml_org_handler): symbols loaded only
+	  if scalar style is plain.
+
+	* test/yaml/test_yaml.rb (test_perl_regexp): updated test to
+	  match new regexp serialization.
+
+Sat May 15 01:41:34 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (eval): forgot to restore $SAFE value before evaluating
+	  compiled node.  [ruby-core:02872]
+
+Sat May 15 01:33:12 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* range.c (range_each_func): terminates loop if generating value
+	  is same to @end.  [ruby-talk:100269]
+
+Fri May 14 22:08:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_new4): should not reuse frozen shared string if
+	  the original is not an instance of String. [ruby-talk:100193]
+
+Fri May 14 21:29:26 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_mdump): preserve GMT bit in the marshal data.
+	  [ruby-talk:100213]
+
+Fri May 14 18:37:49 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/canvas.rb: improve coords support for canvas
+	  items. Now, supports all of the followings.
+	       TkcLine.new(c, 0, 0, 100, 100, :fill=>'red')
+	       TkcLine.new(c, [0, 0, 100, 100], :fill=>'red')
+	       TkcLine.new(c, [0, 0], [100, 100], :fill=>'red')
+	       TkcLine.new(c, [[0, 0], [100, 100]], :fill=>'red')
+	       TkcLine.new(c, :coords=>[0, 0, 100, 100], :fill=>'red')
+	       TkcLine.new(c, :coords=>[[0, 0], [100, 100]], :fill=>'red')
+
+Fri May 14 13:30:39 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* test/ruby/test_float.rb: Add test for util.c revision 1.42.
+
+Fri May 14 12:13:46 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* util.c (ruby_strtod): strtod("0", &end); => end should point '\0'.
+	 [ruby-dev:23498]
+
+Thu May 13 15:47:30 2004  akira yamada  <akira@ruby-lang.org>
+
+	* lib/net/telnet.rb (Net::Telnet::login): "options" can specify
+	  regexps for login prompt and/or password prompt.
+
+Thu May 13 14:17:57 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c (yaml_org_handler): some empty strings were
+	  loaded as symbols.
+
+Thu May 13 11:04:08 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack): always add with null for 'Z'.
+
+	* pack.c (pack_unpack): terminated by null for 'Z'.  [ruby-talk:98281]
+
+Wed May 12 19:59:43 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_type, check_sizeof): replace unusable characters.
+	  [ruby-talk:99788]
+
+Wed May 12 17:41:42 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb (Resolv::DNS::Config): make it configurable without
+	  external file such as /etc/resolv.conf.
+
+Wed May 12 14:37:27 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509name.c: attribute value of DC (short name of
+	  domainComponent) should be IA5String.
+
+Wed May 12 13:20:19 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk/composite.rb: improve configure methods (based on
+	  the proposal of [ruby-talk:99671]).
+
+Wed May 12 11:51:08 2004  Dave Thomas  <dave@pragprog.com>
+
+	* class.c (rb_obj_singleton_methods): fix rdoc
+
+Tue May 11 07:09:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (is_defined): do not protect exception during receiver
+	  evaluation.
+
+Mon May 10 22:28:14 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/protocol.rb (each_crlf_line): remove junk line.
+
+Mon May 10 21:44:42 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb: Change scheme for
+	  looking up symbols in  HTML generator.
+
+Mon May 10 16:45:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (eval): warning during eval should not cause deadlock.
+	  [ruby-talk:98651]
+
+	* eval.c (rb_eval): raise TypeError exception for superclass
+	  mismatch.  [ruby-list:39567]
+
+Mon May 10 12:11:37 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb: Hack to search parents
+	  for unqualified constant names.
+
+Mon May 10 01:18:15 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/pop.rb (logging): append "\n".
+
+Sun May  9 23:38:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/net/ftp.rb: ported documentation improvement from 1.8 branch
+
+	* lib/net/imap.rb:   ditto
+
+	* lib/net/pop.rb:    ditto
+
+	* lib/net/smtp.rb:   ditto
+
+	* lib/net/telnet.rb: ditto
+
+Sun May  9 23:34:51 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_float.rb: added test_strtod to test Float("0").
+
+Sun May  9 13:24:24 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/yaml/store.rb: use FileUtils::copy.
+
+Sun May  9 12:34:26 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regex.c: removed unused file.
+
+Sat May  8 10:53:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_open): open should not ignore block when "to_open"
+	  method is used.  [ruby-dev:23478]
+
+Fri May  7 22:07:39 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mv): new option `force'. [ruby-talk:99457]
+
+	* lib/fileutils.rb: new method for command option reflection:
+	  FileUtils.commands, .options, .have_option?, .options_of,
+	  .collect_methods.
+
+	* lib/fileutils.rb: module Verbose, NoWrite, DryRun do not have
+	  option flags @fileutils_verbose and @fileutils_noop, they make no
+	  sense.
+
+Fri May  7 21:50:21 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::parse_include): Allow
+	  multiple arguments to 'include'
+
+Fri May  7 21:03:51 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (fu_list): Array() breaks pathes including "\n".
+	  [ruby-core:02843]
+
+	* test/fileutils/test_fileutils.rb (mkdir): test "\n" in path.
+
+Fri May  7 20:53:25 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/dbm/dbm.c (fdbm_modify): typo fixed. [ruby-dev:23473]
+
+Fri May  7 11:17:27 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* util.c (ruby_strtod): "0.0000000000000000001" should be converted
+	  to 1.0e-19 instead of 0.0. (leading zeros aren't significant digits)
+	  [ruby-talk:99318] [ruby-dev:23465]
+
+Thu May  6 22:27:32 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* ext/socket/socket.c (ippaddr): use NUMERICHOST if can not resolve
+	  hostname.
+
+Thu May  6 22:09:29 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/tkutil.c (get_eval_string_core): bug fix. [ruby-dev:23466]
+
+Thu May  6 14:22:29 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb (to_yaml): added instance variable handling
+	  for Ranges, Strings, Structs, Regexps.
+
+	* lib/yaml/rubytypes.rb (to_yaml_fold): new method for setting a
+	  String's flow style.
+
+	* lib/yaml.rb (YAML::object_maker): now uses Object.allocate.
+
+	* ext/syck/gram.c: fixed transfer methods on structs, broke it
+	  last commit.
+
+Thu May  6 14:38:02 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (rb_push_glob): simplified code (not change behavior)
+
+Thu May  6 13:32:44 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: get rid of side effect of Config.expand, patched by
+	  <tttt01@infoseek.jp>  (ruby-bugs:PR#597)
+
+Thu May  6 11:40:28 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (string): accept NIL.
+
+	* lib/net/imap.rb (body_type_basic): allow body-fields omissions.
+
+Thu May  6 01:59:04 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb (Generators::HtmlMethod::params):
+	  Don't include the &block parameter if we have explicit
+	  yield parameters.
+
+Wed May  5 03:52:31 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/ring.rb: use recv instead of recvfrom.
+
+Wed May  5 00:38:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/gserver.rb: documented
+	* lib/xmlrpc/README.txt: introduced for documentation purposes
+
+Mon May  3 09:47:24 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method_or_yield_parameters):
+	  Fix parsing bug if yield called within 1 line block
+
+Sun May  2 21:56:48 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb (rm_f, rm_r): test :force flag.
+
+Sun May  2 01:04:38 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib, ext/tk: renewal Ruby/Tk
+
+Fri Apr 30 20:08:41 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* time.c (SIZEOF_TIME_T): support SIZEOF_TIME_T == SIZEOF_INT.
+
+Wed Apr 28 01:26:11 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* oniguruma.h, regparse.c: imported Oni Guruma 2.2.8.
+
+Wed Apr 28 01:16:23 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* oniguruma.h, regparse.c: imported Oni Guruma 2.2.7.
+
+Tue Apr 27 14:43:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: LIBURUBY_A is needed for extconf.rb even when
+	  cross-compiling.
+
+Tue Apr 27 13:33:50 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (string_content): turn off NODE_NEWLINE flag to avoid
+	  unnecessary line trace for inlined expression.
+	  (ruby-bugs PR#1320)
+
+Tue Apr 27 08:15:13 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb: passing Range tests.
+
+	* ext/syck/syck.h: version 0.44.
+
+	* ext/syck/gram.c: transfers no longer open an indentation.
+	  fixed transfers which precede blocks.
+
+	* ext/syck/token.c: ditto.
+
+	* ext/syck/syck.c: fixed segfault if an anchor has been released already.
+
+	* ext/syck/node.c (syck_free_members): organized order of free'd nodes.
+
+	* ext/syck/rubyext.c (syck_emitter_write_m): test for proper string with
+	  StringValue.
+
+Mon Apr 26 23:56:54 2004  Daniel Kelley  <news-1082945587@dkelley.gmp.san-jose.ca.us>
+
+	* README.EXT, README.EXT.ja: fixed wrong function signature.
+	  [ruby-talk:98349]
+
+Mon Apr 26 21:40:09 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/code_objects.rb (RDoc::Context::add_alias): Only alias
+	  to instance methods.
+
+Sun Apr 25 18:26:23 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (ac_cv_func_fork): set to no on DJGPP.
+
+Sat Apr 24 14:32:03 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* re.c: applied stack error handling patch. [ruby-dev:23431]
+
+Sat Apr 24 10:38:31 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/markup/simple_markup.rb (SM::SimpleMarkup::group_lines):
+	  Fix bug where consecutive headings are merged.
+
+Fri Apr 23 23:24:47 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: $hdrdir should not contain macros, for backward
+	  compatibility.  [bruby-dev:28]
+
+	* lib/mkmf.rb (create_makefile): in the case of extout, just copy
+	  script files, without comparison.
+
+Fri Apr 23 16:38:46 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb: sync taint/freeze flag between
+	  a pathname object and its internal string object.
+
+Fri Apr 23 14:52:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt, arg, aref_args): should not make sole splat into
+	  array, in aref_args other than aref with op_asgn.
+
+Fri Apr 23 14:14:38 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/resolv.rb: don't use Regexp#source to embed regexps.
+	  [ruby-dev:23432]
+
+Thu Apr 22 18:25:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, ext/extmk.rb: make ext and .ext get removed by distclean.
+
+Thu Apr 22 10:07:01 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* */Makefile.sub (distclean-local): should remove $(RBCONFIG).
+
+Thu Apr 22 04:17:57 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_mod_define_method): allow binding methods to modules.
+	  [ruby-dev:23410]
+
+	* parse.y (aref_args): should pass expanded list.  [ruby-core:02793]
+
+Thu Apr 22 01:12:57 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (flo_to_s): tweak output string based to preserve
+	  decimal point and to remove trailing zeros.  [ruby-talk:97891]
+
+	* string.c (rb_str_index_m): use unsigned comparison for T_FIXNUM
+	  search.  [ruby-talk:97342]
+
+Wed Apr 21 23:04:42 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/rinda.rb, test/rinda/test_rinda.rb: check Hash tuple size.
+
+Wed Apr 21 20:05:00 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::HTTP#proxy_open): set Host: field explicitly.
+	  [ruby-list:39542]
+
+Wed Apr 21 18:39:46 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb: merge SMTP-TLS patch.  This patch is
+	  contributed by Daniel Hob. [ruby-core:02789]
+
+Wed Apr 21 18:23:45 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb: change coding style: def m( a ) -> def m(a).
+
+Wed Apr 21 18:01:47 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/pop.rb: do not use class variables.
+
+	* lib/net/pop.rb (do_start): ensure to clean up connection when
+	  authentication failed.
+
+Wed Apr 21 17:23:59 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (HTTP#connect): CONNECT must precede SSL connect.
+	  [ruby-dev:23379]
+
+	* lib/net/http.rb (HTTP.new): class variables are not inherited
+	  now.
+
+Wed Apr 21 15:56:43 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/ui/console/testrunner.rb (test_started): restore $0
+	  after changing process title.  [ruby-talk:97426]
+
+Wed Apr 21 10:18:06 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* process.c(rb_spawn): fix SEGV at "p system('command line here')"
+	 (may happen only in bccwin32) [ruby-dev:23380]
+
+Mon Apr 19 20:58:44 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c: Updated RDocs.
+
+Mon Apr 19 18:11:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_equal): returns true if two hashes have same set
+	  of key-value set.  [ruby-talk:97559]
+
+	* hash.c (rb_hash_eql): returns true if two hashes are equal and
+	  have same default values.
+
+Mon Apr 19 08:19:11 2004  Doug Kearns  <djkea2@mugca.its.monash.edu.au>
+
+	* dln.c, io.c, pack.c, lib/benchmark.rb, lib/cgi.rb, lib/csv.rb,
+	  lib/date.rb, lib/ftools.rb, lib/getoptlong.rb, lib/logger.rb,
+	  lib/matrix.rb, lib/monitor.rb, lib/set.rb, lib/thwait.rb,
+	  lib/timeout.rb, lib/yaml.rb, lib/drb/drb.rb, lib/irb/workspace.rb,
+	  lib/net/ftp.rb, lib/net/http.rb, lib/net/imap.rb, lib/net/pop.rb,
+	  lib/net/telnet.rb, lib/racc/parser.rb, lib/rinda/rinda.rb,
+	  lib/rinda/tuplespace.rb, lib/shell/command-processor.rb,
+	  lib/soap/rpc/soaplet.rb, lib/test/unit/testcase.rb,
+	  lib/test/unit/testsuite.rb: typo fix.
+
+Mon Apr 19 08:14:18 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_body): Allow for
+	  #ifdef HAVE_PROTOTYPES
+
+Fri Apr 16 17:04:07 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_equal): always returns true or false, never
+	  returns nil. [ruby-dev:23404]
+
+Fri Apr 16 12:38:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/drb/drb.rb (DRb::DRbUnknown::initialize): Exception#to_str is
+	  deprecated.
+
+	* lib/drb/drb.rb (DRb::DRbServer::InvokeMethod::perform): multiple
+	  value class changed.
+
+	* lib/drb/invokemethod.rb (DRb::DRbServer::InvokeMethod18Mixin::block_yield):
+	  ditto.
+
+Fri Apr 16 08:27:08 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: skip linking when libraries to be preloaded not
+	  compiled.  [ruby-list:39561]
+
+Thu Apr 15 19:57:11 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (pst_success_p): new method Process::Status#success?.
+	  [ruby-dev:23385]
+
+	* rubytest.rb: do nothing while cross-compiling, return status in
+	  system independent style.
+
+Thu Apr 15 19:26:54 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (rb_push_glob): Dir.glob() should return nil if block is given.
+	  (http://www.ruby-lang.org/ja/man/index.cgi?cmd=view;name=Dir)
+
+	* dir.c (push_braces): Dir.glob() should handle '{ }' nested more than
+	  3 times.
+
+	* dir.c (push_braces, rb_push_glob): Dir.glob() should handle escaped
+	  '{' and '}' and ','.
+
+	  [ruby-dev:23376]
+
+Thu Apr 15 17:12:13 2004  Tanaka Akira  <akr@m17n.org>
+
+	* ext/gdbm/gdbm.c (Init_gdbm): define GDBM::READER, GDBM::WRITER,
+	  GDBM::WRCREAT and GDBM::NEWDB.
+	  (fgdbm_initialize): use specified read/write flag.
+
+Wed Apr 14 13:06:09 2004  Doug Kearns  <djkea2@mugca.its.monash.edu.au>
+
+	* array.c, enum.c, eval.c, file.c, io.c, numeric.c, object.c, prec.c,
+	  process.c, re.c, string.c: typos in RDoc comments.  [ruby-core:02783]
+
+Wed Apr 14 11:29:56 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* numeric.c (flo_eq): workaround for bcc32's bug.
+	  (ruby-bugs-ja:PR#594)
+
+Wed Apr 14 11:06:38 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::scan): Changed
+	  behavior of :enddoc: -- it now unconditionally terminates
+	  processing of the current file.
+
+Wed Apr 14 10:57:40 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* defines.h: include <net/socket.h> to get fd_set definition in BeOS.
+
+Tue Apr 13 23:00:55 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/rinda.rb: change pattern matching.
+	  a === b -> a == b || a === b. [druby-ja:98]
+
+	* test/rinda/test_rinda.rb: ditto.
+
+Tue Apr 13 21:50:57 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bcc32/Makefile.sub (PHONY): Borland make disallows empty command
+	  rules.
+
+Tue Apr 13 17:55:16 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (begin_transport): should not overwrite HTTP
+	  request header. [ruby-list:39543]
+
+Tue Apr 13 16:48:00 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/pop.rb: merge POP3S patch.  This patch is contributed by
+	  Daniel Hobe.
+
+Tue Apr 13 02:56:29 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* common.mk: changed the order of ascii.c alphabetically.
+
+Mon Apr 12 19:11:21 2004  Eric Hodel  <drbrain@segment7.net>
+
+	* gc.c (rb_gc_copy_finalizer): typo.  [ruby-core:02774]
+
+Mon Apr 12 18:45:58 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_init_i): should return
+	  a value.
+
+Mon Apr 12 10:39:50 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (rb_glob2, rb_glob, push_globs, push_braces, rb_push_glob):
+	  fix memory leak. (leaked when block was interrupted)
+
+Sun Apr 11 19:10:13 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (require_libraries): restore source file/line after
+	  statically linked extensions initialized.  [ruby-dev:23357]
+
+Sun Apr 11 10:47:04 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/code_objects.rb (RDoc::TopLevel::add_class_or_module): Toplevel
+	  classes and modules are a special case too... (handle extending existing
+	  classes with or without :enddoc:)
+
+Sat Apr 10 23:51:13 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/code_objects.rb (RDoc::Context::add_to): Implementation of :enddoc:
+	  made one too many assumptions...
+
+Sat Apr 10 00:00:19 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/markup/simple_markup/inline.rb: Fix problem
+	  with \_cat_<b>dog</b>
+
+Fri Apr  9 17:05:21 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (has_magic, find_dirsep): incomplete '[' matches no character
+	  in Dir.glob. (follows File.fnmatch's behavior)
+
+	* dir.c (fnmatch_helper): incomplete escape is ignored in File.fnmatch.
+	  (follows Dir.glob's behavior)
+
+	* dir.c (find_dirsep): '/' between '[' and ']' is ignored in Dir.glob.
+	  (follows File.fnmatch with File::FNM_PATHNAME 's behavior)
+
+	* dir.c (find_dirsep): escaped slash '\/' loses its meaning as
+	  directory separator in Dir.glob.
+
+	  [ruby-dev:23291]
+
+Thu Apr  8 20:25:19 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): skip uncompiled extensions.
+
+	* lib/mkmf.rb (create_makefile): emit no rules for static library if
+	  $static is nil, e.g., outside of ext/.
+
+	* lib/test/unit/ui/console/testrunner.rb (test_started): show test
+	  name via $0.
+
+	* runruby.rb: set environments to use the compiled binary.
+
+	* test/runner.rb: do nothing while cross-compiling.
+
+	* test/drb/drbtest.rb, test/soap/calc/test_calc_cgi.rb: use envutil to
+	  know ruby binary, and restore $: after require.
+
+	* test/ruby/envutil.rb: give priority to RUBY environment variable to
+	  use just compiled binary and libraries.
+
+Thu Apr  8 19:03:33 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_binmode): inverted condition.  [ruby-dev:23349]
+
+Thu Apr  8 18:22:00 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_s_list): return encoding list if no block
+	  is given.  [ruby-dev:23063]
+
+Wed Apr  7 15:29:24 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): use NUM2INT() instead of num2i32().
+
+Wed Apr  7 12:32:02 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb, lib/rss/1.0.rb: accepted rdf:resource or
+	  resource attribute in rdf:li.
+	* test/rss/test_parser.rb: added test for above change.
+
+	* lib/rss/dublincore.rb: reverted style.
+
+	* lib/rss/xmlparser.rb: normalized XMLParser class hierarchy.
+
+Wed Apr  7 10:43:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, common.mk, */Makefile.sub (ext/extinit.o): OUTFLAG
+	  doesn't work for object files on VC.
+
+	* */Makefile.sub (config.h): need SIZEOF_TIME_T now.
+
+Wed Apr  7 00:24:34 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/rinda.rb: fix hash tuple bug.
+
+	* lib/rinda/tuplespace.rb: ditto.
+
+	* test/rinda/test_rinda.rb
+
+Tue Apr  6 18:24:18 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_get_path): get path string via "to_path" method if
+	  path object is not a string. [Ruby2]
+
+	* gc.c (rb_gc_call_finalizer_at_exit): do not free threads in the
+	  exit finalizers.
+
+	* io.c (rb_io_reopen): should use rb_io_check_io().
+
+Tue Apr  6 16:46:09 2004  Tanaka Akira  <akr@m17n.org>
+
+	* configure.in: check the size of time_t.
+
+	* time.c (time_add): new function.
+	  (time_plus): use time_add.
+	  (time_minus): use time_add.
+
+Tue Apr  6 13:11:48 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (raise_socket_error): never return.
+
+	* ext/socket/socket.c (make_hostent): must return value.
+
+Tue Apr  6 00:14:43 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (Init_Exception): remove Exception#to_str. [Ruby2]
+
+	* eval.c (error_print): should no call "to_str" anymore use
+	  "message" method instead.
+
+	* io.c (rb_f_open): Kernel#open() calls "to_open" if the first
+	  argument responds to it. [Ruby2]
+
+Tue Apr  6 00:13:43 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/rinda.rb: add require 'drb/drb'
+
+Mon Apr  5 22:25:32 2004  Tanaka Akira  <akr@m17n.org>
+
+	* test/zlib/test_zlib.rb: new file.
+	  (TestZlibGzipWriter#test_new_nil): test for [ruby-dev:23228].
+
+Mon Apr  5 22:16:23 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y (assoc_list): {a: 1, b: 2} should be allowed.
+	  [ruby-dev:23328]
+
+Mon Apr  5 19:43:40 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regexec.c: imported Oni Guruma 2.2.6.
+
+Mon Apr  5 19:39:10 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* regparse.c, oniguruma.h: imported Oni Guruma 2.2.6.
+
+Mon Apr  5 12:12:09 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (raise_socket_error): some platforms don't have
+	  EAI_SYSTEM.
+
+Mon Apr  5 08:18:23 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/rdoc.rb: Remove leading ./ from file names so that cross
+	  references work properly.
+
+Sun Apr  4 14:01:20 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/options.rb (Options::parse): Allow multiple -x options to
+	  RDoc. Fix bug where files weren't being excluded properly
+
+Sat Apr  3 09:36:38 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/syck.h: version 0.43.
+
+Sat Apr  3 08:28:47 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/lib/gram.c: allow root-level inline collections.
+	  [ruby-talk:94922]
+
+	* lib/yaml/rubytypes.rb (Symbol#to_yaml): emit symbols as implicits.
+	  [ruby-talk:94930]
+
+Fri Apr  2 19:28:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bcc32/Makefile.sub (OUTFLAG): needed for static-linked-ext.
+
+Fri Apr  2 18:00:05 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): extract necessary variables for static link
+	  from Makefile.
+
+	* lib/mkmf.rb (create_makefile): save preload and libpath for next
+	  compile.
+
+Fri Apr  2 17:27:17 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (top_include): include in the wrapped load is done for
+	  the wrapper, not for a singleton class for wrapped main.
+	  [ruby-dev:23305]
+
+Fri Apr  2 15:13:44 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_eq): use temporary double variable to save the
+	  result (internal float register may be bigger than 64 bits, for
+	  example, 80 bits on x86).  [ruby-dev:23311]
+
+Fri Apr  2 14:35:26 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (block_pass): should generate unique identifier of the
+	  pushing block.  [ruby-talk:96363]
+
+Fri Apr  2 11:36:20 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* eval.c (Init_load): make $LOADED_FEATURES built-in.
+	  [ruby-dev:23299]
+
+	* ruby.c (ruby_prog_init): make $PROGRAM_NAME built-in.
+
+	* lib/English.rb: remove $LOADED_FEATURES and $PROGRAM_NAME.
+
+Fri Apr  2 07:31:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c: mistakingly removed do_not_reverse_lookup.
+	  [ruby-list:39475]
+
+	* ext/socket/socket.c (make_hostent): fix memory leak, based on
+	  the patch from HORIKAWA Hisashi <vzw00011@nifty.ne.jp>.
+
+Thu Apr  1 22:55:33 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb: Allow rdoc comments in
+	  =begin rdoc/=end
+
+	* lib/rdoc/parsers/parse_rb.rb: Fix problem with comment in
+	  top-level method being taken as file comment.
+
+Thu Apr  1 22:55:04 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_options.rb: Fix undefined variable warning.
+
+Thu Apr  1 19:58:37 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/mapping/{factory.rb,registry.rb}: fixed illegal mapped URI
+	  object with soap/marshal.
+	  added URIFactory class for URI mapping.  BasetypeFactory checks
+	  instance_variables when original mapping is not allowed (ivar must
+	  be empty).  Instance of URI have instance_variables but it must be
+	  llowed whenever original mapping is allowed or not.
+
+Wed Mar 31 19:06:23 2004  Tanaka Akira  <akr@m17n.org>
+
+	* time.c (year_leap_p): new function.
+	  (timegm_noleapsecond): ditto.
+	  (search_time_t): use timegm_noleapsecond instead of
+	  mktime for first guess.
+
+Wed Mar 31 12:05:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (DelegateClass): define internal methods of the
+	  result class, but not metaclass of the caller.  [ruby-talk:96156]
+
+	* intern.h: provide proper prototypes.  [ruby-core:02724]
+
+	* ruby.h: missing.h is now prerequisite to intern.h.
+
+Wed Mar 31 11:17:16 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): raises RangeError if uv is out of UTF8 value
+	  range.  [ruby-dev:23281]
+
+	* io.c (rb_io_binmode): stdio buffer should be empty when calling
+	  IO#binmode.  [ruby-talk:96155]
+
+Tue Mar 30 20:25:34 2004  Tanaka Akira  <akr@m17n.org>
+
+	* time.c (search_time_t): limit guess range by mktime if it is
+	  available.  [ruby-dev:23274]
+
+Tue Mar 30 18:19:00 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): fix SEGV at retry in iterator's receiver.
+	  [ruby-dev:23227]
+
+Mon Mar 29 20:17:16 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_exec): follow older behavior if close-on-exec is not
+	  available.
+
+	* process.c (rb_fork): protect from exceptions while waiting failed
+	  process, if status is given.
+
+Sun Mar 28 16:25:37 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (clean-local, distclean-local): remove
+	  work files.
+
+	* win32/Makefile.sub (clean-local): ditto.
+
+Sun Mar 28 14:23:02 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/pop.rb: def m( arg ) -> def m(arg).
+
+Sun Mar 28 14:09:13 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/pop.rb (auth): failed when account/password include "%".
+	  [ruby-talk:95933]
+
+Sat Mar 27 21:40:41 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb: permit extra semicolon in content-type field.
+
+Sat Mar 27 10:40:48 2004  Tanaka Akira  <akr@m17n.org>
+
+	* (lib/pp.rb, lib/prettyprint.rb): define seplist in PP::PPMethods
+	  instead of PrettyPrint.
+
+Sat Mar 27 01:47:09 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: trim tail space of each line.  no user visible change.
+
+	* lib/rss/dublincore.rb: fixed class definition mismatch.
+
+	* sample/openssl/gen_csr.rb: fixed wrong usage text.
+
+Thu Mar 25 23:15:24 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_options.rb (RI::Options::show_version):
+	  Add --version option
+
+Thu Mar 25 21:45:00 2004  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: Bug in + and - reported by Bret Jolly
+	  fixed.
+
+Thu Mar 25 21:01:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* version.c (ruby_show_copyright): obtain copyright year from
+	  RUBY_RELEASE_YEAR.
+
+	* win32/resource.rb: ditto.
+
+Thu Mar 25 19:37:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/resource.rb: default rubyw icon to ruby.ico, and let DLL also
+	  include them.
+
+	* win32/resource.rb: include winver.h for older WindowsCE.
+
+Thu Mar 25 14:01:03 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, */Makefile.sub (lib, dll): phony targets.
+
+	* configure.in (ruby, miniruby): ditto.
+
+	* cygwin/GNUmakefile.in (rubyw): ditto.
+
+Thu Mar 25 04:16:18 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_options.rb (RI::Options): Add the --list-names option,
+	  which dumps our all known names
+
+Thu Mar 25 03:57:47 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_util.rb (NameDescriptor::initialize): No longer
+	  allow nested classes to be designated using "."--you must
+	  now use "::"
+
+Thu Mar 25 02:00:18 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/template/html/one_page_html.rb (Page):
+	  Fix to work with C modules.
+
+Wed Mar 24 20:49:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/uri.rb: Documented (thanks Dmitry V. Sabanin).
+	* lib/uri/common.rb: Ditto.
+	* lib/uri/ftp.rb: Ditto.
+	* lib/uri/generic.rb: Ditto.
+	* lib/uri/http.rb: Ditto.
+	* lib/uri/https.rb: Ditto.
+	* lib/uri/ldap.rb: Ditto.
+	* lib/uri/mailto.rb: Ditto.
+
+Wed Mar 24 18:48:05 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb ($ruby, $topdir, $hdrdir): should not be affected by
+	  DESTDIR after installed.
+
+	* lib/mkmf.rb (dummy_makefile): default file lists to be cleaned.
+
+Wed Mar 24 12:32:56 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::handle_class_module):
+	  Don't document methods if we don't know for sure the
+	  class or module.
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_class):
+	  Don't store documentation for singleton classes if we
+	  don't know the real class.
+
+Wed Mar 24 11:11:26 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb (Generators::HTMLGenerator::load_html_template):
+	  Allow non-RDoc templates by putting a slash in the template name
+
+Wed Mar 24 10:05:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile::_close): should not clear @tmpname
+	  until the file is really removed. [ruby-core:02684]
+
+Wed Mar 24 04:12:44 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_mod_cvar_get): new method Module#class_variable_get.
+
+	* object.c (rb_mod_cvar_set): ditto (Module#class_variable_set).
+
+Tue Mar 23 17:45:22 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_thread_atfork): 1.9 warns no more for thread
+	  termination.  [ruby-dev:23212]
+
+Tue Mar 23 14:46:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, */Makefile.sub, common.mk (clean-local, distclean-local):
+	  separate files under directories due to directory separator.
+
+	* */Makefile.sub (MKFILES): common.mk and */Makefile.sub should not be
+	  removed.
+
+	* win32/Makefile.sub, wince/Makefile.sub: $* cannot appear in explicit
+	  rules.
+
+	* cygwin/GNUmakefile.in: some mingw stuffs were missed.
+
+	* lib/mkmf.rb (create_makefile): Borland make wrongly removes braces
+	  from command lines.
+
+	* bcc32/Makefile.sub: needs bcc32/mkexports.rb.
+
+Mon Mar 22 08:21:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, */Makefile.sub, common.mk: extract common portions.
+
+	* Makefile.in, cygwin/GNUmakefile.in, */Makefile.sub (RBCONFIG):
+	  separated time stamp file for rbconfig.rb.
+
+	* configure.in: append common.mk to Makefile.
+
+	* mkconfig.rb: keep mtime of rbconfig.rb if unchanged.
+
+	* win32/rm.bat: remove multiple files.
+
+	* wince/mkconfig_wce.rb: use fake.rb instead.
+
+Sun Mar 21 22:17:35 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#virtual_host):
+	  sort @virtual_hosts in address, port, host order.
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#lookup_server):
+	  hostname should not be match if :ServerAlias is not given.
+
+Sun Mar 21 21:11:16 2004  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/shell/*: bug fix for Shell#system(command_line_string).
+
+Sun Mar 21 21:04:42 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ruby.1: add -width option to .Bl for old groff.
+
+Sun Mar 21 18:57:37 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/*: Test::Unit::TestCase -> RSS::TestCase and
+	  Test::Unit::Assertions -> RSS::Assertions.
+
+Sun Mar 21 18:48:20 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/{rss,dublincore,syndication}.rb: handled W3CDTF correctly.
+
+Sun Mar 21 18:15:29 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_xml-stylesheet.rb: added tests for xml-stylesheet.
+
+	* lib/rss/xml-stylesheet.rb: added xml-stylesheet parsing
+	  function.
+
+Sat Mar 20 23:51:03 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* eval.c (rb_require_safe): preserve old ruby_errinfo.
+	  [ruby-talk:95409]
+
+	* eval.c (rb_f_raise): should not clear backtrace information if
+	  exception object already have one.
+
+Sat Mar 20 21:21:03 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/extmk.rb: rm -rf $extout, not extout.
+
+Sat Mar 20 15:25:36 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/template/html/html.rb (RDoc::Page): Force
+	  page background to white.
+
+Sat Mar 20 09:33:36 2004  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb, lib/date/format.rb: _parse() now accepts fractional
+	  part of second minute that follows a comma or a full stop.
+
+Fri Mar 19 21:06:21 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (assoc_list): allow {sym: val} style Hash. [Ruby2]
+	  this change is done by Nobuyoshi Nakada <nobu@ruby-lang.org>.
+
+Fri Mar 19 15:15:15 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_cvar_set): class variables become private to the
+	  particular class/module. [Ruby2]
+
+	* variable.c (rb_cvar_get): ditto.
+
+	* variable.c (rb_cvar_defined): ditto.
+
+Fri Mar 19 11:31:32 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb ($beos, $solaris): add OS flags.
+
+	* lib/mkmf.rb (RUBY): / is not recognized as path separator on
+	   nmake/bmake. [ruby-list:39388]
+
+	* lib/mkmf.rb (CLEANLIBS, CLEANOBJS): should remove *.exp with *.so.
+
+Fri Mar 19 01:55:57 2004  Mauricio Fernandez  <batsman.geo@yahoo.com>
+
+	* io.c (rb_io_sync): need not to check writable. [ruby-core:02674]
+
+Thu Mar 18 19:47:44 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* instruby.rb, rubytest.rb: do not depend on srcdir.
+
+Thu Mar 18 18:50:06 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: no longer embed srcdir and compile_dir into
+	  rbconfig.rb.
+
+	* ext/extmk.rb, lib/mkmf.rb: obtain top_srcdir and topdir from library
+	  paths.
+
+Thu Mar 18 17:46:35 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: do not undef :to_a.
+
+Thu Mar 18 16:22:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_eq): avoid false positive by using scope and
+	  dyna_vars.  no longer use frame.uniq.
+
+	* eval.c (proc_arity): arity is now defined as number of
+	  parameters that would not be ignored. i.e. Proc.new{}.arity
+	  returns zero.  update test suites too.
+
+Thu Mar 18 15:27:25 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c: remove specialized version of rb_Array().  use simple
+	  one defined in object.c.
+
+	* object.c (Init_Object): remove Kernel#to_a.
+
+	* enum.c (enum_zip): use "to_a" instead of "to_ary".
+
+Wed Mar 17 00:22:03 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* oniguruma.h: imported Oniguruma 2.2.5.
+	* regparse.c: ditto.
+
+Tue Mar 16 11:14:17 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (fnmatch_helper): File.fnmatch('\.', '.') should return true.
+	  (Rev1.112 lost compatibility)
+
+	* dir.c (fnmatch_helper): File.fnmatch('\/', '/', File::FNM_PATHNAME)
+	  should return true. (Rev1.112 lost compatibility)
+
+	* dir.c (fnmatch): File.fnmatch('**/.boo', '.foo/.boo',
+	  File::FNM_PATHNAME) should return false because of leading period.
+
+Mon Mar 15 17:01:07 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (exc_initialize): calling 'to_str' each time just for
+	  type checking is too heavy.  [ruby-core:02661]
+
+Mon Mar 15 10:14:51 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SocketForwarder):
+	  add do_not_reverse_lookup.
+
+Mon Mar 15 07:39:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_yield_0): should not re-submit TAG_BREAK if this
+	  yield is not break destination. [ruby-dev:23197]
+
+Sun Mar 14 22:07:38 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_raise): err at unstarted thread.  (PR#1302)
+
+Sat Mar 13 14:56:32 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/ut_drb.rb: use 'druby://localhost:0'. [ruby-dev:23078]
+
+	* test/drb/ut_eval.rb: ditto.
+
+	* test/drb/ut_large.rb: ditto.
+
+	* test/drb/ut_safe1.rb: ditto.
+
+	* test/drb/ut_drb_drbssl.rb: use 'drbssl://localhost:0'.
+
+Fri Mar 12 23:52:56 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (fnmatch): directory recursion '**/' can be used with
+	  File::FNM_PATHNAME. [ruby-dev:22901]
+
+	* dir.c (fnmatch, fnmatch_helper): only '/' is accepted as path
+	  separator even in DOSISH environment. [ruby-dev:22974]
+	  [ruby-list:39337]
+
+	* dir.c (fnmatch_helper): faster '*' matching.
+
+Fri Mar 12 20:19:16 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_noreturn): default for platforms not support
+	  prototypes.
+
+	* ruby.c (ruby_init_loadpath): buffer for path name should have
+	  MAXPATHLEN.
+
+	* lib/mkmf.rb (configuration): include topdir and hdrdir in VPATH.
+
+	* lib/mkmf.rb (create_makefile): default dependency rule.
+
+Fri Mar 12 07:35:36 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/config.rb (WEBrick::Config::General): add
+	  :DoNotReverseLookup.
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#accept): call
+	  do_not_reverse_lookup for each socket if :DoNotReverseLookup
+	  is set.  [ruby-code:02357]
+
+Wed Mar 10 22:26:25 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (remove_dir): should handle symlink correctly.
+	  This patch is contributed by Christian Loew.  [ruby-talk:94635]
+
+Wed Mar 10 16:28:42 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (return_jump): set return value to the return
+	  destination.  separated from localjump_destination().
+
+	* eval.c (break_jump): break innermost loop (or thread or proc).
+
+	* eval.c (rb_yield_0): set exit_value for block break.
+
+Wed Mar 10 16:00:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* struct.c (rb_struct_s_def): Struct::new executes block with
+	  generated struct class. [ruby-talk:02606]
+
+Wed Mar 10 15:58:43 2004  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* eval.c (eval): Only print backtrace if generating the backtrace
+	  doesn't generate an exception.  [ruby-core:02621]
+
+Wed Mar 10 10:15:16 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (opt_W_getter): get rid of warning.
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub:
+	  fixed dependency.
+
+Tue Mar  9 13:04:26 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_ungetc): raise IOError instead of calling
+	  rb_sys_fail().  [ruby-talk:23181]
+
+Tue Mar  9 10:03:40 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub:
+	  replaced regex.c entry with Oniguruma files.
+
+Tue Mar  9 01:09:46 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* Makefile.in: replaced regex.c entry with Oniguruma files.
+
+Mon Mar  8 23:16:07 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: HTTPHeader did not initialized correctly.
+
+	* lib/net/http.rb (connect): does same debug output.
+
+Mon Mar  8 21:38:18 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (add_header): remove warning. [ruby-dev:23170]
+
+Mon Mar  8 21:09:39 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (range): Cancel change for incomplete '['. More discussion
+	  is needed.
+
+Mon Mar  8 19:35:13 2004  akira yamada  <akira@arika.org>
+
+	* lib/uri/common.rb (URI::REGEXP::PATTERN::HOSTPORT): (?:#{PORT})
+	  -> (?::#{PORT}).  [ruby-dev:23170]
+
+Mon Mar  8 15:03:24 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (range): treat incomplete '[' as ordinary character (like
+	  has_magic does). fix buffer overrun at incomplete escape like '[\'.
+
+Mon Mar  8 13:35:32 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* regparse.c (parse_exp): need to separate initialization for bcc32.
+	  [ruby-dev:23169]
+
+	* oniguruma.h (ONIG_EXTERN): check __GNUC__ instead of __CYGWIN__.
+
+Mon Mar  8 01:05:55 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/config.rb (WEBrick::Config::HTTP): rename :RequestHander
+	  to :RequestCallback and add new option :ServerAlias.
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): use
+	  :RequestCallback and warn if :RequestHandler is in server's option.
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): should print
+	  error message for WEBrick::HTTPSataus::Error.
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#lookup_server):
+	  lookup for hostname from :ServerAlias if the req.host is not match
+	  to :ServerName.
+
+	* lib/webrick/httpservlet.rb (WEBrick::HTTPServlet::CGIHandler#do_GET):
+	  use $?.exitstatus and refine log message.
+
+Sun Mar  7 16:22:26 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* Makefile.in (lex.c): use $? instead of $<.
+
+	* lib/pstore.rb (commit_new): use FileUtils.copy_stream for Cygwin.
+	  [ruby-dev:23157]
+
+Sun Mar  7 05:34:42 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: HTTPHeader keeps its header fields as an array.
+
+	* lib/net/http.rb: new method HTTPHeader#add_header, get_fields.
+
+	* lib/net/http.rb: new method HTTPHeader#content_length=.
+
+	* lib/net/http.rb: new method HTTPHeader#content_type, main_type,
+	  sub_type, type_params, content_type=, set_content_type.
+
+	* lib/net/http.rb (HTTPHeader#basic_encode): result of pack(m) may
+	  contain multiple LFs.
+
+Sun Mar  7 03:11:00 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: new method Net::HTTPRequest#body(=).
+
+	* lib/net/http.rb: new method Net::HTTPRequest#body_stream(=).
+
+Sun Mar  7 02:06:07 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: spin off https code again.
+
+	* lib/net/https.rb: new file.
+
+	* ext/openssl/lib/net/https.rb: removed.  moved to net/https with
+	  slight modifications.
+
+	* ext/openssl/lib/net/protocols.rb: removed.  merged with net/http.
+
+	* lib/net/protocol.rb: new class BufferedIO.
+
+	* lib/net/protocol.rb: InternetMessageIO < BufferedIO.
+
+	* lib/net/protocol.rb: BufferedIO.new takes an IO.
+
+	* lib/net/smtp.rb: follow InternetMessageIO's change.
+
+	* lib/net/pop.rb: ditto.
+
+Sun Mar  7 00:55:03 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/protocol.rb: remove method: InternetMessageIO#address,
+	  port, ip_address, read_timeout(=), socket.
+
+	* lib/net/protocol.rb: simplify code.
+
+	* lib/net/protocol.rb: apply latest coding style.
+
+Sat Mar  6 15:15:05 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/strscan/depend: depends on re.h and regex.h.
+
+	* ext/strscan/strscan.c: no version check needed.
+
+	* ext/strscan/strscan.c (strscan_init_copy): struct re_registers must
+	  not be bitwise copied.
+
+Sat Mar  6 11:14:33 2004  David Black  <dblack@wobblini.net>
+
+	* lib/scanf.rb: refixed the previous fix in IO#block_scanf
+
+Sat Mar  6 10:49:40 2004  David Black  <dblack@wobblini.net>
+
+	* lib/scanf.rb: fixed a logic glitch in IO#block_scanf
+
+Sat Mar  6 02:00:19 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: net/https is merged.
+
+	* ext/openssl/lib/net/https.rb: ditto.
+
+Sat Mar  6 00:39:21 2004  Kazuo Saito  <ksaito@uranus.dti.ne.jp>
+
+	* oniggnu.h: imported from Oniguruma library.
+	* oniguruma.h: ditto.
+	* regcomp.c: ditto.
+	* regenc.c: ditto.
+	* regenc.h: ditto.
+	* regerror.c: ditto.
+	* regex.c: ditto.
+	* regexec.c: ditto.
+	* reggnu.c: ditto.
+	* regint.h: ditto.
+	* regparse.c: ditto.
+	* regparse.h: ditto.
+	* ascii.c: ditto.
+	* euc_jp.c: ditto.
+	* sjis.c: ditto.
+	* utf8.c: ditto.
+
+	* MANIFEST: added Oniguruma files listed above.
+
+	* LEGAL: added Oniguruma license.
+
+	* regex.h: now includes oniggnu.h.
+
+	* re.c: applied Oniguruma patch.
+
+Fri Mar  5 23:13:08 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: support WebDAV methods, PROPPATCH, LOCK,
+	  UNLOCK, OPTIONS, PROPFIND, DELETE, MOVE, COPY, MKCOL.
+	  This patch is contributed by Tatsuki Sugiura.
+
+Fri Mar  5 20:58:37 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: Net::HTTPResponse#response is obsolete.
+	  [ruby-core:02592]
+
+	* lib/net/http.rb: Net::HTTPResponse#header is obsolete.
+
+	* lib/net/http.rb: Net::HTTPResponse#read_header is obsolete.
+
+Fri Mar  5 20:10:57 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c: new method StringScanner#initialize_copy
+	  to allow #dup and #clone.
+
+	* test/strscan/test_strscan.rb: test StringScanner#dup.
+
+Fri Mar  5 19:42:09 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (HTTPResponse#to_ary): should return an object
+	  which does not respond to #to_ary.  It causes infinite loop in
+	  puts. [ruby-core:02578]
+
+Fri Mar  5 00:51:35 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/test/unit.rb: Move RDoc documentation so that you can
+	  now say 'ri Test::Unit'
+
+Thu Mar  4 22:31:40 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in: miniruby is not needed for cross compile.
+
+	* configure.in (PREP): miniruby for native compile.
+
+Thu Mar  4 11:46:32 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/drb/extservm.rb (DRb::ExtServManager#invoke_service_command):
+	  detach server processes to get rid of zombies.
+
+Thu Mar  4 10:41:25 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ruby.h (T_MASK): save 1 bit in flags bits by shifting T_xxx
+	  values.
+
+Thu Mar  4 08:08:36 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c: get rid of warnings.
+
+	* lib/rss/taxonomy.rb: ditto.
+
+	* lib/rdoc/ri/ri_formatter.rb: ditto.
+
+	* test/ruby/test_assignment.rb: ditto.
+
+Thu Mar  4 01:17:28 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/ri/ri_display.rb (DefaultDisplay::page): wait until the
+	  pager terminates.
+
+Wed Mar  3 13:10:56 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (method_hash): new method. [ruby-talk:93968]
+
+	* eval.c (proc_eq): do not compare dyna_vars.
+
+	* eval.c (proc_hash): new method.
+
+	* eval.c (rb_yield_0): protect break/return from within orphan (or
+	  lambda) Proc object.
+
+Wed Mar  3 09:52:05 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb ($topdir): use compile_dir only when not installed yet.
+	  [ruby-talk:94098]
+
+Wed Mar  3 01:18:52 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/converter.rb: handled Uconv::Error.
+
+	* lib/rss/dublincore.rb: DublincoreModel -> DublinCoreModel
+
+Wed Mar  3 00:59:30 2004  David Black  <dblack@wobblini.net>
+
+	* lib/scanf.rb: soak_up_spaces only ungetc's non-space last
+	  character
+
+	* lib/scanf.rb: IO#block_scanf now returns partial last iteration
+	  array if format string matches partly
+
+Tue Mar  2 16:30:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): erred program name should be reported by
+	  exceptions, instead of the first argument.
+
+	* process.c (rb_spawn): ditto.
+
+	* process.c (proc_spawn_v): use first argument as program name.
+
+	* win32/win32.c (rb_w32_aspawn): ditto.
+
+	* win32/win32.c (CreateChild): search executable file if no program
+	  name given.
+
+	* lib/drb/extservm.rb (invoke_service_command): use Process.spawn.
+	  [ruby-dev:23103]
+
+	* lib/rdoc/ri/ri_display.rb (setup_pager): use IO.popen.
+	  [ruby-dev:23086], [ruby-dev:23103]
+
+	* lib/rdoc/diagram.rb (convert_to_png): ditto.
+
+	* lib/rdoc/generators/chm_generator.rb (compile_project): ditto.
+
+Tue Mar  2 12:24:03 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub, wince/Makefile.sub (config.h): shouldn't check
+	  defined? NORETURN. [ruby-dev:23100]
+
+Tue Mar  2 11:28:40 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_iterator.rb (test_ljump): cannot use
+	  assert_nothing_raised due to passing block.
+
+Tue Mar  2 06:23:14 2004  David Black  <dblack@wobblini.net>
+
+	* lib/scanf.rb: fixed Kernel#scanf to propagate code block
+
+Mon Mar  1 23:25:40 2004  David Black  <dblack@wobblini.net>
+
+	* lib/scanf.rb: Partial fix so STDIN#scanf works with new
+	 STDIN#pos behavior
+
+Mon Mar  1 19:42:05 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/setup.mak: configure's default is "--enable-install-doc"
+
+	* win32/setup.mak: ditto.
+
+Mon Mar  1 12:24:10 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_alias):
+	  Allow aliases to have parentheses
+
+Sun Feb 29 23:14:53 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_class):
+	  Handle :nodoc: on singleton classes.
+
+Sat Feb 28 21:50:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bcc32/Makefile.sub, bcc32/README.bcc32, bcc32/configure.bat,
+	  bcc32/setup.mak: new configure scheme. use ``configure --prefix=dir''
+	  instead of ``make DESTDIR=dir install''.
+	  --with-static-linked-ext support on bccwin32. [ruby-dev:23034]
+
+Sat Feb 28 21:50:20 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/setup.mak: "configure --disable-install-doc" is now working.
+
+	* win32/setup.mak: ditto.
+
+Sat Feb 28 15:09:49 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/configure.bat: append missing label ":exit".
+
+	* bcc32/configure.bat: fix typo.
+
+Sat Feb 28 10:31:03 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* MANIFEST: add test_erb.rb
+
+	* lib/erb.rb, test/erb/test_erb.rb: don't forget filename,
+	  if both filename and safe_level given. [ruby-dev:23050]
+
+Sat Feb 28 01:08:40 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (yylex): should not allow symbol for invalid global
+	  variable (e.g. `:$-)`). [ruby-core:02518]
+
+Fri Feb 27 20:37:09 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): no orphan block check is needed when pcall
+	  is true.
+
+	* eval.c (localjump_destination): update localjump condition.
+
+Fri Feb 27 02:10:49 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (localjump_destination): lambda should not interfere
+	  return from the yielded block.
+
+Fri Feb 27 00:53:49 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb, test/drb/drbtest.rb: require drb/eq.rb by default
+
+Thu Feb 26 12:15:02 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (make_cmdvector): adjust successive double-quote
+	  handling.
+
+Thu Feb 26 09:42:56 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (delete_if_i): use st_delete_safe() (via
+	  rb_hash_delete()) instead of returning ST_DELETE.
+
+Thu Feb 26 02:35:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_f_exec): get rid of SEGV when exec failed for command
+	  in single string.
+
+Wed Feb 25 21:17:33 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (obj_free), io.c (rb_io_fptr_finalize), rubyio.h (OpenFile):
+	  sharing OpenFile.
+
+	* io.c (rb_io_initialize): accept IO instance.  [ruby-dev:22195]
+
+Wed Feb 25 21:16:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (with_destdir): should return the given argument if no
+	  DESTDIR is given.
+
+	* instruby.rb: use path name expansion of cmd.exe.
+
+Wed Feb 25 20:44:45 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi-lib.rb, lib/getopts.rb, lib/importenv.rb, lib/parsearg.rb:
+	  warn with caller position.
+
+	* test/rss/test_content.rb, test/rss/test_dublincore.rb,
+	  test/rss/test_syndication.rb, test/rss/test_trackback.rb: use cgi
+	  instead of cgi-lib.
+
+Tue Feb 24 18:42:03 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (glob_helper): '**/' should not match leading period
+	  unless File::FNM_DOTMATCH is set. (like '*/') [ruby-dev:23014]
+
+Tue Feb 24 18:03:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_file.rb (test_fnmatch): test for dir.c:1.108.
+
+Tue Feb 24 17:07:17 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (fnmatch): File.fnmatch with File::FNM_PATHNAME was broken
+	  for the pattern including '*' followed by '/'.
+
+Tue Feb 24 13:22:21 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc::normalized_file_list): Attempt to get better
+	  heuristics on which files to include and exclude. Now only include
+	  non-standard files if they are explicitly named in ARGV.
+
+Tue Feb 24 07:23:30 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb: Deal with :stopdoc: when
+	  choosing a default main page to display (ie. don't select a page
+	  if we don't have documentation for it).
+
+Tue Feb 24 06:40:14 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RubyLex::identify_identifier): Handle
+	  class variables in code listings
+
+Tue Feb 24 06:32:27 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_aliases): Handle
+	  aliases in C files.
+
+Tue Feb 24 06:16:22 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/rdoc.rb (RDoc::RDoc::document): Now create op dir _before_
+	  parsing files.
+
+Tue Feb 24 06:08:47 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_constant):
+	  Start collecting text of constant values earlier: was missing
+	  values in output if there was no space after '='
+
+Tue Feb 24 06:08:25 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb: Escape contant values.
+
+Tue Feb 24 03:45:06 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_config.c (ossl_config_each): add new method
+	  OpenSSL::Config#each. it iterates with section name, field name
+	  and value.
+
+	* ext/openssl/ossl_config.c (Init_ossl_config): include Enumerable.
+
+Mon Feb 23 09:09:44 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* instruby.rb (parse_args): use optparse instead of getopts.
+
+	* instruby.rb (DOSISH): embedded path in batch files should not be
+	  prefixed by DESTDIR.  [ruby-core:02186]
+
+Sun Feb 22 14:58:04 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: $extstatic is Array or nil now.  [ruby-talk:93383]
+
+	* Makefile.in, bcc32/Makefile.sub, win32/Makefile.sub: terminate options.
+
+	* lib/mkmf.rb (init_mkmf): $INCFLAGS also should be lazy-evaluated.
+
+Sun Feb 22 13:05:37 2004  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/mailto.rb (URI::MailTo::to_s): should include fragment.
+
+Sun Feb 22 12:58:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: use optparse instead of getopts.
+
+	* Makefile.in, bcc32/Makefile.sub, win32/Makefile.sub: ditto.
+
+Sun Feb 22 09:51:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* re.c: corrected documentation format (rb_reg_initialize_m)
+
+Sat Feb 21 22:41:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* ext/zlib/zlib.c: documented, but needs more effort.
+
+Sat Feb 21 14:33:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: prefer relative path.  [ruby-talk:93037]
+
+Sat Feb 21 11:12:08 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/os2.c, missing/x68.c: typo fix.  pointed out by greentea.
+
+Fri Feb 20 19:11:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (OpenStruct#initialize_copy): should not share
+	  members.  [ruby-dev:22966]
+
+Fri Feb 20 18:59:47 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/init.rb (IRB::IRB.parse_opts): add -I option to
+	  irb. [ruby-dev:39243]
+
+Fri Feb 20 12:55:27 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): fix typo.
+
+	* win32/win32.c (CreateChild): first argument to CreateProcess() must
+	  have path, not just basename.
+
+Thu Feb 19 23:24:16 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb (Generators::HtmlClass::build_attribute_list):
+	  Support visibility modifiers for attributes
+
+Thu Feb 19 22:39:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/ostruct.rb: documented
+
+Thu Feb 19 22:39:04 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/rinda/test_rinda.rb: DRb.start_service only once in testsuites.
+	  DRb.start_service could handle this.
+
+Thu Feb 19 22:24:04 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/mapping/rubytypeFactory.rb: should not dump singleton class.
+	  [ruby-dev:22588]
+	  c = class << Object.new; class C; self; end; end; SOAPMarshal.dump(c)
+
+Thu Feb 19 18:08:18 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* ext/strscan/strscan.c: improved documentation
+
+Thu Feb 19 18:08:18 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c, win32/win32.h: fixed prototypes.
+
+	* win32/win32.c (wait): same as waitpid() with -1.
+
+Thu Feb 19 02:34:28 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/markup/simple_markup/preprocess.rb (SM::PreProcess::include_file):
+	  Only strip comment markers if all lines start with comments.
+
+Thu Feb 19 03:05:49 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c: StringScanner#restsize is obsolete;
+	  use #rest_size instead.
+
+	* ext/strscan/strscan.c: StringScanner#matchedsize is obsolete;
+	  use #matched_size instead.
+
+Thu Feb 19 02:42:19 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c: don't use rb_eval_string, it defines
+	  classes under the module when required in module clauses.
+	  [ruby-dev:22951]
+
+Thu Feb 19 02:37:28 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c: merge documentation from 1.8 branch.
+	  Thanks Gavin Sinclair.
+
+Thu Feb 19 00:20:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/cgi-lib.rb:   deprecated after 1.8.1
+
+	* lib/getopts.rb:   ditto
+
+	* lib/importenv.rb: ditto
+
+	* lib/parsearg.rb:  ditto
+
+Thu Feb 19 00:11:05 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/markup/simple_markup/preprocess.rb (SM::PreProcess::handle):
+	  Strip extraneous space from filenames in :include:
+
+Wed Feb 18 22:53:41 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/unix.rb: remove O_NONBLOCk, thanks \ay
+
+Wed Feb 18 22:42:19 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb: improt test_rinda.rb
+
+Wed Feb 18 22:03:11 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/*: should not depend on $KCODE.
+
+Wed Feb 18 18:07:09 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_sprintf.rb: added tests.
+
+Wed Feb 18 17:18:01 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c: need to include <olectl.h> on Cygwin.
+
+Wed Feb 18 10:40:38 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): sign bit extension should not be done
+	  if FPLUS flag is specified.  [ruby-list:39224]
+
+	* sprintf.c (rb_f_sprintf): do not prepend dots for negative
+	  numbers if FZERO is specified.  [ruby-dev:39218]
+
+Wed Feb 18 10:23:34 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* sprintf.c (rb_f_sprintf): clean up.
+
+Tue Feb 17 23:40:34 2004  Guy Decoux  <ts@moulon.inra.fr>
+
+	* sprintf.c (rb_f_sprintf): preserve original val for
+	  format_integer. [ruby-talk:92975]
+
+Tue Feb 17 23:28:45 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/marshal/test_marshal.rb, test/ruby/test_marshal.rb: do $:
+	  trick to share the testcase test/ruby/marshaltestlib.rb.
+
+Tue Feb 17 23:13:23 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/marshaltestlib.rb: common marshal testcase added.
+
+	* test/ruby/test_marshal.rb: use above testsuite.
+
+	* test/soap/marshal/test_marshal.rb: ditto.
+
+	* test/soap/marshal/cmarshal.rb: removed (not used).
+
+Tue Feb 17 19:34:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): $extout_prefix doesn't vary for libraries.
+
+	* ext/extmk.rb (extmake): remove compile directory if empty.
+
+	* ext/extmk.rb (parse_args) lib/mkmf.rb (create_makefile): move
+	  initialization of $extout_prefix from lib/mkmf.rb.  [ruby-dev:22928]
+
+	* ext/extmk.rb: clear ext and extout directory when cleaning.
+
+	* lib/mkmf.rb (CLEANLIBS): should be under $(arch) directory.
+
+Tue Feb 17 18:02:10 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c: ScanError may be (wrongly) garbage
+	  collected. (thanks Gavin Sinclair)
+
+	* ext/strscan/strscan.c: move ::ScanError to StringScanner::Error.
+	  ::ScanError is also defined for backward compatibility.
+
+	* ext/strscan/strscan.c: #peep is obsolete, use #peek.
+
+	* ext/strscan/strscan.c: #empty? is obsolete, use #eos?.
+
+	* ext/strscan/strscan.c: #clear is obsolete, use #terminate.
+
+	* ext/strscan/strscan.c: #getbyte is obsolete, use #get_byte.
+
+Tue Feb 17 12:12:47 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (parse_args): delay expanding $(extout) until invoking
+	  make.
+
+	* lib/mkmf.rb (CLEANLIBS): should remove files have specific
+	  extensions.
+
+Tue Feb 17 11:33:30 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rss/rexmlparser.rb: REXML version may be 4 digits.
+
+Tue Feb 17 10:45:59 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/syck/rubyext.c (syck_emitter_end_object): takes only one arg.
+
+Tue Feb 17 07:48:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rexml/encodings/SHIFT_JIS: wrong library name.
+
+Tue Feb 17 01:35:28 2004  Tanaka Akira  <akr@m17n.org>
+
+	* eval.c (rb_eval): care that another thread replace NODE_DREGX_ONCE
+	  to NODE_LIT.  [ruby-dev:22920]
+
+Tue Feb 17 01:20:57 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb: new module FileUtils::DryRun.
+
+Mon Feb 16 23:28:14 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb: document reduction.  [ruby-core:02429]
+
+	* test/yaml/test_yaml.rb: added 0..1 test with "0".."1" on display.
+	  it should be defined that the specification about what kind of Range
+	  is supported in ruby's custom type in YAML.
+
+Mon Feb 16 22:22:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/generator.rb: corrected doc format
+
+	* lib/rinda/rinda.rb: added documentation (from Hugh Sasse)
+
+	* lib/rinda/tuplespace.rb: ditto
+
+	  [Note: rinda files actually committed Wed Feb 18 07:27:00 2004]
+
+Mon Feb 16 20:28:52 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/Makefile.sub: show more warnings. (referring to mingw)
+
+	* bcc32/setup.mak: ditto.
+
+Mon Feb 16 18:35:58 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): should create *.pdb on ext/,
+	  not .ext/.
+
+	* win32/Makefile.sub (config.status): convert the name of import
+	  library.
+
+	* lib/mkmf.rb (create_makefile): now don't need to remove
+	  $(TARGET).lib.
+
+Mon Feb 16 15:45:22 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check functions, fork spawnv.
+
+	* io.c (rb_io_s_popen): accept argv not only single command line.
+
+	* process.c (rb_proc_exec_n): export.
+
+	* process.c (rb_check_argv): check if arguments are safe to invoke.
+
+	* process.c (rb_fork): retry to fork.
+
+	* process.c (rb_spawn): spawn child process asynchronously.
+
+	* process.c (rb_f_system): raise an exception if the command could not
+	  execute.
+
+	* win32/win32.c (rb_w32_argv_size): count necessary size for joined
+	  arguments.
+
+	* win32/win32.c (rb_w32_join_argv): join arguments with quoting.
+
+	* win32/win32.c (rb_w32_pipe_exec, rb_w32_spawn, rb_w32_aspawn):
+	  accept program name adding to command line.
+
+Mon Feb 16 15:18:33 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/racc/parser.rb: add note for Racc full package.
+
+Mon Feb 16 15:13:01 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/racc/cparse/README: new file.
+
+	* ext/racc/cparse/MANIFEST: add README.
+
+Mon Feb 16 12:29:10 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_s_list): new method Iconv.list
+	  (libiconv only).
+
+Mon Feb 16 10:29:52 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (CompareImpl): File.fnmatch and Dir.glob get better performance
+	  in Win32. This is achieved by calling downcase() for single-byte
+	  characters. (CharLower() is slower than downcase())
+
+Mon Feb 16 02:14:29 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* bcc32/Makefile.sub: should warn suspicious pointer conversion.
+
+	* bcc32/setup.mak: ditto.
+
+Sun Feb 15 20:56:22 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (push_braces): remove wrong const. [ruby-dev:22891]
+
+Sun Feb 15 20:41:15 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* sample/soap/calc/httpd.rb, sample/soap/exchange/httpd.rb,
+	  sample/soap/sampleStruct/httpd.rb, sample/wsdl/googleSearch/httpd.rb:
+	  use soap/property instead of getopts for configuring DocumentRoot
+	  and port# of httpd.  see samplehttpd.conf below.
+
+	* sample/soap/calc/samplehttpd.conf,
+	  sample/soap/exchange/samplehttpd.conf,
+	  sample/soap/sampleStruct/samplehttpd.conf,
+	  sample/wsdl/googleSearch/samplehttpd.conf: added.
+
+Sun Feb 15 19:13:33 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/tuplespace.rb: read(tpl,0), raise RequestExpiredError
+	  if not found.
+
+Sun Feb 15 15:48:57 2004  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add IDispatch wrapper in val2variant.
+	  Thanks, arton.
+
+Sun Feb 15 15:23:29 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* ruby.h, dir.c (rb_glob): add const.
+
+Sun Feb 15 01:46:05 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/mkmf.rb: absolute path of ruby is assigned to $(RUBY).
+	  [ruby-dev:22870]
+
+Sat Feb 14 23:59:11 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* mkconfig.rb: use fileutils.rb instead of ftools.rb.
+
+	* bcc32/Makefile.sub, win32/Makefile.sub (config.h): define
+	  STACK_GROW_DIRECTION.
+
+	* bcc32/Makefile.sub (config.h): add newer checks.
+
+	* wince/Makefile.sub (config.h): define NEED_IO_SEEK_BETWEEN_RW.
+
+Sat Feb 14 23:26:27 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/un.rb: use OptionParser instead of getopts.
+
+Sat Feb 14 11:28:14 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* sample/drb/*: import lib/drb/sample
+
+Sat Feb 14 11:14:12 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: add pretty_print, thanks gotoken.
+
+Fri Feb 13 21:51:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/fileutils.rb: slightly improved documentation (sync with 1.8)
+
+Fri Feb 13 19:57:01 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rss/test_trackback.rb: added tests for TrackBack with RSS
+	  2.0.
+
+	* test/rss/common.rb: added methods make RSS 2.0.
+
+	* lib/rss/trackback.rb: TrackBack API is decided.
+
+	* lib/rss/rss.rb: RSS::VERSION 0.0.7 -> 0.0.8.
+
+	* lib/rss/parser.rb, lib/rss/rss.rb: replaced $DEBUG by RSS::DEBUG.
+
+	* lib/rss/2.0.rb: removed RSS 2.0 URI. Because RSS 2.0 doesn't
+	  have URI.
+
+Fri Feb 13 14:41:00 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: en-bugged at last commit (Feb 11 23:24:22 2004)
+
+Fri Feb 13 12:26:37 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb: rescue SystemCallError instead
+	  of EINVAL.  File.link may raise EACCES on network file systems.
+
+Fri Feb 13 05:18:58 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb: File.link raises EINVAL on
+	  Win9x. [ruby-dev:22713]
+
+Thu Feb 12 21:49:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/ftools.rb: documented
+
+Thu Feb 12 21:19:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/base64.rb: added and tidied documentation
+
+Thu Feb 12 20:45:01 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/protocol.rb (WriteAdapater#puts): must append "\n" to
+	  the string, don't prepend. (ruby-bugs:PR#1280)
+
+Thu Feb 12 20:31:43 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_tmpsrc): cpp32 of Borland C++ ignores #error
+	  directives in DOS line-ending files at all.
+
+Thu Feb 12 15:23:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rparen): ignore preceding newlines to right parentheses.
+	  (ruby-bugs:PR#1221) [ruby-dev:22858]
+
+Thu Feb 12 14:17:43 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: set ac_cv_func_link to yes to enable link() on MinGW.
+	  [ruby-dev:22241]
+
+Thu Feb 12 13:32:49 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (link): raise NotImplementedError on Win9X.
+	  contributed by Tietew. [ruby-dev:22713]
+
+	* win32/win32.c, win32/win32.h (link): add const.
+
+Thu Feb 12 09:56:19 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/lib/tk.rb (TkComm::tk_split_list): suppress a warning.
+
+Thu Feb 12 02:23:56 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb: use assert_raise instead of assert_raises.
+
+	* lib/pp.rb: ditto.
+
+	* lib/time.rb: ditto.
+
+	* lib/tsort.rb: ditto.
+	  use TSortHash and TSortArray instead of Hash and Array in test.
+
+Wed Feb 11 23:24:22 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: properly treat a Tcl/Tk's string with escaping
+	  special characters.
+
+Tue Feb 10 20:49:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (method_proc): return bound Proc object.  [ruby-dev:22854]
+
+	* eval.c (rb_mod_define_method): bind method body itself for Method
+	  object.
+
+	* node.h (NODE_DMETHOD): deprecated.
+
+	* object.c (rb_class_inherited_p): export.
+
+Tue Feb 10 16:43:50 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (umethod_bind): purge unused check.  [ruby-dev:22850]
+
+Tue Feb 10 14:33:08 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_match): raise TypeError when both arguments are
+	  strings. [ruby-dev:22851]
+
+	* string.c (rb_str_match2): removed.
+
+	* Makefile.in, bcc32/Makefile.sub, win32/Makefile.sub,
+	  wince/Makefile.sub (string.c): now not depend on version.h.
+
+Mon Feb  9 17:46:07 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, bcc32/Makefile.sub, win32/Makefile.sub, configure.in,
+	  runruby.rb: run rdoc, test and so on with compiled extension
+	  libraries.  [ruby-dev:22688]
+
+	* ext/extmk.rb, lib/mkmf.rb: make extension libraries in separated
+	  directory, similar to the actual directory structure.
+
+	* lib/fileutils.rb (FileUtils.copy_file): use the mode of the original
+	  file to create new file.
+
+	* lib/rdoc/ri/ri_paths.rb (RI::Paths::SYSDIR): get rid of unexpected
+	  influence by environment variable.
+
+	* bcc32/configure.bat, win32/configure.bat: add install-doc options.
+
+	* win32/win32.c, win32/win32.h (rb_w32_fstat): fix Borland C runtime
+	  bug which returns wrong mode.  [ruby-dev:22846]
+
+Mon Feb  9 16:30:12 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (detach_process_watcher): return the last status.
+	  [ruby-dev:22841]
+
+Sun Feb  8 16:46:08 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pp.rb (PP::PPMethods::object_address_group): suppress negative
+	  sign for higher heap areas.
+
+Sun Feb  8 16:18:27 2004  akira yamada  <akira@ruby-lang.org>
+
+	* test/yaml/test_yaml.rb (YAML_Unit_Tests::test_range_cycle):
+	  added tests.
+
+Sun Feb  8 15:51:57 2004  akira yamada  <akira@ruby-lang.org>
+
+	* test/ruby/test_file.rb (TestFile::test_fnmatch): added tests for
+	  File.fnmatch. [ruby-dev:22815][ruby-dev:22819]
+
+Sun Feb  8 15:41:45 2004  akira yamada  <akira@ruby-lang.org>
+
+	* test/yaml/test_yaml.rb (YAML_Unit_Tests::test_range_cycle):
+	  added tests. [ruby-core:02306] [ruby-core:02311]
+
+Sun Feb  8 14:24:35 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (HTTP#request): should not overwrite Connection
+	  header. (ruby-bugs:PR#1274)
+
+Sun Feb  8 10:11:21 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (glob_helper): Dir.glob('**/') did not work. [ruby-dev:22832]
+
+Sun Feb  8 00:29:26 2004  Hirokazu Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (fnmatch): File.fnmatch('*?', 'a') should return true.
+	  [ruby-dev:22815]
+
+	* dir.c (fnmatch): File.fnmatch('\[1\]' , '[1]') should return true.
+	  [ruby-dev:22819]
+
+	* dir.c: Did some styles (no change to behavior)
+
+Sat Feb  7 19:56:11 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb/init.rb (IRB.rc_files): yield possible rc file names.
+
+	* lib/irb/input-method.rb (IRB::ReadlineInputMethod::initialize):
+	  load and save history automatically.  [ruby-core:02352]
+
+Fri Feb  6 22:48:16 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb (gen_url): Support
+	  https in RDoc hyperlinks
+
+Fri Feb  6 22:41:22 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/pp.rb (PPInspectTest#test_to_s_with_iv): rollback the previous
+	  commit.  [ruby-dev:22813]
+
+Fri Feb  6 22:22:50 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/pp.rb (PPInspectTest#test_to_s_with_iv): remove instance
+	  variable which is defined in the test.
+
+Fri Feb  6 18:54:18 2004  akira yamada  <akira@ruby-lang.org>
+
+	* test/ruby/test_proc.rb (TestProc::test_eq): added a
+	  test.  [ruby-dev:22599]
+
+Fri Feb  6 18:26:00 2004  akira yamada  <akira@ruby-lang.org>
+
+	* test/ruby/test_proc.rb (TestProc::test_eq): added tests for
+	  Proc#==.  [ruby-dev:22592], [ruby-dev:22601]
+
+Fri Feb  6 10:12:06 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (bsock_do_not_reverse_lookup): control reverse
+	  lookup for every instance.  [ruby-core:02346]
+
+Fri Feb  6 09:15:11 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb/extend-command.rb: add irb_help command.  [ruby-talk:91610]
+
+	* lib/irb/cmd/help.rb (IRB::ExtendCommand::Help): show RDoc.
+
+Fri Feb  6 00:48:37 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/prettyprint.rb (PrettyPrint#first?): obsoleted.
+
+Thu Feb  5 23:56:55 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/prettyprint.rb (PrettyPrint#seplist): added.
+
+	* lib/pp.rb (PPMethods#pp_object): use seplist.
+	  (PPMethods#pp_hash): ditto.
+	  (Array#pretty_print): ditto.
+	  (Struct#pretty_print): ditto.
+	  (MatchData#pretty_print): ditto.
+
+	* lib/set.rb (Set#pretty_print): use seplist.
+
+Wed Feb  4 22:39:46 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_stat_mode): should not sign-expand, so backout.
+
+Wed Feb  4 02:12:06 2004  Tanaka Akira  <akr@m17n.org>
+
+	* file.c (test_l): fix wrong method name in document.
+	  (test_S): ditto.
+	  (test_b): ditto.
+	  (test_c): ditto.
+	  (test_suid): ditto.
+	  (test_sgid): ditto.
+	  (test_sticky): ditto.
+
+Tue Feb  3 22:36:25 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/2.0.rb, lib/rss/content.rb, lib/rss/dublincore.rb,
+	  lib/rss/rss.rb, lib/rss/syndication.rb: removed warnings.
+
+	* lib/rss/converter.rb: removed handling load error of nkf.
+
+	* test/rss/test_syndication.rb, test/rss/test_trackback.rb,
+	  test/rss/test_dublincore.rb, test/rss/test_content.rb: replaced
+	  'require "rss/parser"' by 'require "rss/1.0"'.
+
+	* test/rss/test_parser.rb, test/rss/test_accessor.rb: removed
+	  'require "rss/parser"'.
+
+Tue Feb  3 11:23:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reduce_nodes): remove tail returns.  [ruby-talk:90934]
+
+Tue Feb  3 08:04:57 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb (Struct#pretty_print_cycle): follow 1.8 style.
+
+Mon Feb  2 22:06:31 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (block_append, new_evstr, void_expr0): remove no longer used
+	  labels.
+
+Mon Feb  2 18:45:50 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* dln.c (dln_load): don't specify RTLD_GLOBAL on Interix,
+	  because it caused SEGV when running runner.rb.
+
+Mon Feb  2 01:54:00 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb (Struct#pretty_print): make it 1.8 style.
+	  (Numeric#pretty_print, FalseClass#pretty_print)
+	  (TrueClass#pretty_print, Module#pretty_print): fix pp for objects
+	  with instance variables.  [ruby-talk:91157]
+
+	* lib/open-uri.rb (URI::Generic#find_proxy): return nil on loopback
+	  address.
+
+	* lib/resolv-replace.rb (BasicSocket#send): don't replace because
+	  it has no hostname argument.
+	  (IPSocket.getaddress): raise SocketError instead of
+	  Resolv::ResolvError for errors.
+	  (TCPSocket#initialize, UDPSocket#bind, UDPSocket#connect)
+	  (SOCKSSocket#initialize): use IPSocket.getaddress instead of
+	  Resolv.getaddress.
+	  (UDPSocket#send): recognize 3 arguments form.  try all addresses on
+	  4 arguments form.
+
+Sun Feb  1 23:00:00 2004  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal.c: Bug in BigDecimal("1e#{n}").add BigDecimal('.5'),n)
+	  reported and fixed by Javier Goizueta.
+
+Sun Feb  1 18:21:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	  From ruby_1_8 branch:
+	* lib/test/unit.rb: rearranged documentation for RDoc's sake.
+	* lib/matrix.rb: improved documentation.
+	* lib/net/http.rb: slight documentation formatting improvement.
+
+Sun Feb  1 05:30:06 2004  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::Generic#find_proxy): warn HTTP_PROXY.
+	 raise an error on non-http proxy URI.
+	 (OpenURI::Buffer#<<): make a tempfile binmode.  [ruby-talk:90793]
+
+Sun Feb  1 00:57:41 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/parser.rb (RSS::Parser): added @@default_parser. Used
+	  XML parser became selectable.
+	* test/rss/test_parser.rb: added tests for
+	  RSS::Parser.default_parser.
+
+Sat Jan 31 02:28:15 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (RPATHFLAG): set to -Wl,-R like NetBSD on Interix.
+
+Sat Jan 31 01:09:41 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: leading 0 padding of timestamp usec part.
+
+Fri Jan 30 18:53:23 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (KR_REHASH): wrong hash value on sizeof(long) > sizeof(int).
+
+Thu Jan 29 23:11:57 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (DLEXT2): removed.  Ruby does not treat
+	  ".dll" as a extension library anymore.
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub (DLEXT2):
+	  ditto.
+
+	* util.c (mblen): fix overrun.  [ruby-dev:22672]
+
+Thu Jan 29 22:41:53 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb: Allow 'link:' in Tidylinks.
+	  THis means you can write "see f1[link:files/f1_rb.html]".
+
+Thu Jan 29 22:24:47 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* sample/openssl/gen_csr.rb: follow OpenSSL::X509::Name change.
+	  ASN.1 type of subject DN elements were wrong.
+
+Thu Jan 29 22:19:51 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/*: remove $: trick.  [ruby-dev:22763]  use test/runner.rb to
+	  run test.
+
+Thu Jan 29 19:28:16 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (Request#initialize): reject only when a path is
+	  empty. [ruby-dev:22771]
+
+Thu Jan 29 18:54:08 2004  H.Yamamoto  <ocean@m2.ccsnet.ne.jp>
+
+	* dir.c (glob_helper): infinite loop bug in win32 code.
+	  [ruby-dev:22770]
+
+Thu Jan 29 17:03:49 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c: merge tuning from H.Yamamoto <ocean@m2.ccsnet.ne.jp>.
+	  [ruby-dev:22761]
+
+Thu Jan 29 11:32:14 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/rss/test_*: do $: trick while searching a module in the current
+	  directory.
+
+	* test/xsd/test_xmlschemaparser.rb, test/wsdl/test_emptycomplextype.rb,
+	  test/soap/helloworld/test_helloworld.rb,
+	  test/soap/calc/{test_calc.rb,test_calc2.rb}: do File.expand_path
+	  before using __FILE__.
+
+	* test/yaml/test_yaml.rb: assert_equals -> assert_equal.
+
+Thu Jan 29 01:56:02 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c: usec round-tripping skew. [ruby-core:2305]
+
+	* lib/yaml/rubytypes.rb: character Range now round-trips. [ruby-core:2306]
+
+	* test/yaml/test_yaml.rb: add Time and Range tests.
+
+Thu Jan 29 00:00:46 2004  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss: rss/parser.rb is always required.
+
+Wed Jan 28 15:09:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rss/*.rb: remove "test/" prefix.
+
+Wed Jan 28 13:07:02 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/extconf.rb: include iconv.h for libiconv.  [ruby-dev:22715]
+
+Wed Jan 28 12:43:07 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/rss: rss library imported. [ruby-dev:22726]
+
+Wed Jan 28 04:29:41 2004  Eric Schwartz  <emschwar@fc.hp.com>
+
+	* lib/cgi/session.rb: use LOCK_SH to read, and a few other
+	  improvements.  [ruby-core:02328]
+
+Tue Jan 27 15:00:14 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el: better support for general delimited
+	  strings.  [ruby-dev:22695]
+
+Tue Jan 27 11:04:40 2004  FUKUMOTO Atsushi  <fukumoto@nospam.imasy.or.jp>
+
+	* ext/socket/socket.c (s_recvfrom): sending length should be an
+	  invariant while retrying on EAGAIN.  [ruby-talk:89962]
+
+Tue Jan 27 10:31:28 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (set_argv): fix condition.
+
+Tue Jan 27 02:26:31 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (WEBrick:HTTPUtils::parse_header):
+	  refine regex for header-name.
+
+Mon Jan 26 22:53:04 2004  Dave Thomas  <dave@pragprog.com>
+
+	* io.c: Remove documentation references to $defout.
+
+Mon Jan 26 14:41:46 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/weakref.rb (WeakRef::initialize): set up @__id before
+	  calling "super".
+
+	* lib/delegate.rb (Delegator::initialize): preserve
+	  singleton_method_added method [ruby-dev:22685]
+
+	* lib/delegate.rb (Delegator::initialize): use Kernel::raise
+	  instead of mere raise.  [ruby-dev:22681]
+
+Mon Jan 26 12:45:23 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: define CONST84 when TCL_MAJOR_VERSION == 7
+
+Mon Jan 26 11:30:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: Makefiles should depend on also rbconfig.rb.
+	  (ruby-bugs:PR#1256)
+
+	* ext/win32ole/win32ole.c (set_argv): set real arguments to
+	  WIN32OLE::ARGV.  [ruby-list:39073]
+
+Sun Jan 25 18:25:26 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_add_entry): third
+	  argument become optional.
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_initialize): ditto.
+
+	* ext/openssl/ossl_x509name.c (Init_x509name): emailAddress and
+	  domainComponent should be IA5String.
+
+Sun Jan 25 01:45:38 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): support
+	  virtual host.
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#virtual_host): add
+	  new method to register virtual hosting server.
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#lookup_server): add
+	  new method to lookup virtual hosting server.
+
+Sat Jan 24 13:06:26 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509hame.c (ossl_x509name_initialize): change
+	  second argument. it expected to be a Hash not an Integer.
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_add_entry): add new
+	  function for OpenSSL::X509::Name#add_entry.
+
+	* ext/openssl/ossl_x509name.c (Init_ossl_x509name): add constants
+	  OpenSSL::X509::Name::DEFAULT_OBJECT_TYPE and OBJECT_TYPE_TEMPLATE.
+
+	* ext/openssl/lib/openssl/x509.rb (OpenSSL::X509::Name#initialize):
+	  second argument takes OBJECT_TYPE_TEMPLATE by default.
+
+Fri Jan 23 02:26:30 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (num2i32): pack should not raise RangeError.
+	  [ruby-dev:22654]
+
+	* pack.c (pack_pack): do not auto convert nil to zero.
+
+Thu Jan 22 22:54:53 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (BEG_REGEXP): allow 8-bit characters in quoted
+	  strings for Novell GroupWise Internet Agent.
+
+	* lib/net/imap.rb (DATA_REGEXP): ditto.
+
+Thu Jan 22 18:35:49 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/e2mmap.rb (VERSION): remove unnecessary version checking.
+
+Thu Jan 22 16:21:02 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (string_content): reset lexical states at the beginning of
+	  string contents.  [ruby-list:39061]
+
+Thu Jan 22 08:08:50 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (opt_rescue): use NODE_ERRINFO() instead of
+	  NODE_GVAR("$!"), to avoid confusion from variable aliasing.
+	  [ruby-talk:90074]
+
+	* version.c (Init_version): remove obsolete constants VERSION
+	  etc.  [ruby-dev:22643]
+
+Thu Jan 22 01:46:32 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (newline_node): do not use NODE_NEWLINE node anymore,
+	  use NEWLINE flag instead.
+
+Thu Jan 22 01:12:12 2004  Siena.  <siena@faculty.chiba-u.jp>
+
+	* missing/os2.c (chdir, getcwd):
+	  use _chdir2 and _getcwd2 supporting multiple drives in OS/2 with EMX.
+
+Thu Jan 22 00:33:52 2004  Siena.  <siena@faculty.chiba-u.jp>
+
+	* configure.in: check availability of link(). [ruby-dev:22237]
+	* file.c (rb_file_s_link): raise an exception when link() is unavailable.
+	* missing/os2.c (link): removed. File#link isn't supported.
+	* bcc32/Makefile.sub: define HAVE_LINK to enable link(). [ruby-dev:22241]
+	* win32/Makefile.sub: ditto.
+
+Thu Jan 22 00:26:25 2004  Siena.  <siena@faculty.chiba-u.jp>
+
+	* ChangeLog: typo: RUBY_MBCHAR_MAX was RUBY_MBCHAR_MAXSIZE.
+
+Thu Jan 22 00:12:51 2004  Siena.  <siena@faculty.chiba-u.jp>
+
+	* defines.h: define RUBY_MBCHAR_MAX instead of MB_CUR_MAX.
+	* dir.c (Next, emx_mblen): use RUBY_MBCHAR_MAX for mblen().
+	* file.c (CharNext): ditto.
+	* ruby.c (translate_char): ditto.
+	* util.c (__crt0_glob_function): ditto.
+
+Thu Jan 22 00:10:01 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/base64.rb: :nodoc: the Deprecated module
+
+Wed Jan 21 23:52:39 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: Interix(SFU) support.
+
+Wed Jan 21 23:03:45 2004  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: remove O_NONBLOCK, thanks \ay
+	* lib/drb/extserv.rb: typo
+
+Wed Jan 21 17:57:56 2004  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (envelope): allow NIL.
+	* lib/net/imap.rb (body): ditto.
+	* lib/net/imap.rb (number): ditto.
+	* lib/net/imap.rb (ensure_nz_number): show a detailed error
+	  message.
+
+Wed Jan 21 16:44:15 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (merge_libs): squeeze successive same libraries.
+	  [ruby-dev:22652]
+
+Wed Jan 21 16:10:36 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/base64.rb: enclosed in a module.  [ruby-core:02285]
+
+Wed Jan 21 16:01:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/rmd160/extconf.rb: have_library appends found library.
+
+Wed Jan 21 11:36:00 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_gethostbyname): returns host if
+	  ai_canonname is NULL. (ruby-bugs PR#1243)
+
+	* parse.y (block_append): update nd_end for "real" head node.
+	  [ruby-list:39058]
+
+Tue Jan 20 14:48:28 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: should check <openssl/conf_api.h> instead
+	  of OPENSSL_VERSION_NUMBER. [ruby-list:39056]
+
+Tue Jan 20 14:43:17 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/base64.rb: Add RDoc
+
+Tue Jan 20 14:25:51 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/abbrev.rb: Add RDoc
+
+Tue Jan 20 13:22:39 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb: Document aliases at
+	  top-most level.
+
+	* lib/English.rb: Document English.rb.
+
+Tue Jan 20 04:41:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_marshal.rb (MarshalTestLib::test_exception): test
+	  for [ruby-dev:22604].
+
+	* test/ruby/test_marshal.rb (MarshalTestLibtest_singleton): test
+	  for [ruby-dev:22588].
+
+Tue Jan 20 02:38:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_class): should not dump singleton class.
+	  [ruby-dev:22631]
+
+Tue Jan 20 02:49:22 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: add check for OpenSSL version.
+	  [ruby-list:39054]
+
+Mon Jan 19 23:56:20 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (name_err_mesg_to_str): inverted condition for result of
+	  inspection.  [ruby-dev:22628]
+
+Mon Jan 19 22:24:28 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* sample/exyacc.rb: escape '}' to avoid warning.
+
+	* lib/rdoc/parsers/parse_c.rb: escape '{' and '}' to avoid warnings.
+
+Mon Jan 19 21:28:06 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/defs.h, win32/win3.c, win32/win32.h, file.c: remove
+	  useless casts for Borland C.
+
+Mon Jan 19 17:39:38 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (NameError::message): internal use only.
+
+	* eval.c (rb_method_missing): use hidden constant.
+
+Mon Jan 19 16:30:53 2004  akira yamada  <akira@ruby-lang.org>
+
+	* test/ruby/test_time.rb: added tests for [ruby-dev:22614] and
+	  [ruby-dev:22617].
+
+Mon Jan 19 13:09:21 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/extmk.rb, win32/Makefile.sub, win32/configure.bat,
+	  win32/setup.mak: --with-static-linked-ext support on mswin32.
+
+Mon Jan 19 06:49:07 2004  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date.rb: zone was wrong when it was behind UTC.
+	  Thanks Mark J. Reed.
+
+	* lib/date/format.rb: %z is now always replaced by four digits
+	  with a leading plus or minus sign.
+
+	* sample/cal.rb: added a class, anyway.
+
+Mon Jan 19 01:08:39 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* class.c, error.c, eval.c, intern.h, object.c, variable.c:
+	  do not set path if it is a singleton class.  [ruby-dev:22588]
+
+	* lib/cgi.rb (CGI::QueryExtension): give extended string, not a
+	  delegater object.
+
+Sun Jan 18 23:59:44 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/charset_alias.rb: prefer us_EN locale encodings or
+	  former.  [ruby-dev:22609]
+
+	* ext/iconv/iconv.c (iconv_create): raise InvalidEncoding
+	  exception when EINVAL.
+
+Sun Jan 18 23:16:34 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c, error.c, file.c, io.c, numeric.c, object.c, re.c, struct.c,
+	  time.c: marked init_copy functions nodoc.
+
+Sun Jan 18 20:47:35 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ruby.c: use translate_char() on Cygwin.
+
+Sun Jan 18 20:00:16 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/wsdlDriver.rb, lib/wsdl/soap/operation.rb: add support of
+	  "parts" attribute of soap:body element in WSDL.
+
+	* lib/wsdl/xmlSchema/schema.rb: friendly warning message for
+	  simpleType element which is not supported for now.
+
+	* test/wsdl/soap/{soapbodyparts.wsdl,test_soapbodyparts.wsdl}: new
+	  files.
+
+Sun Jan 18 16:46:48 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_overflow_p): should return results.  [ruby-dev:22614]
+
+Sun Jan 18 12:07:24 2004  Siena.  <siena@faculty.chiba-u.jp>
+
+	* test/ruby/test_time.rb: new test case to test Time#[+-].
+
+	* time.c (time_plus, time_minus): fix RangeError for a negative
+	  argument in environments whose time_t is unsigned. [ruby-dev:22608]
+
+Sun Jan 18 02:33:26 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* defines.h (_WIN32): undef _WIN32 on Cygwin before defining DOSISH.
+
+Sun Jan 18 00:23:55 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (class2path): check anonymous class/module before
+	  checking referable, and allow singleton classes.
+
+Sat Jan 17 23:58:51 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (NameError::Message): new class for lazy evaluation of
+	  message to ensure replaced before marshalling.  [ruby-dev:22604]
+
+	* eval.c (rb_method_missing): use NameError::Message.
+
+Sat Jan 17 21:49:50 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/time.rb (test_rfc2822, test_rfc3339, test_encode_xmlschema):
+	  should not expect that all platforms handle negative time_t value.
+
+Fri Jan 16 23:53:09 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (proc_eq): compare also arguments and environment
+	  (including local variables).  [ruby-dev:22590]
+
+Fri Jan 16 14:33:35 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (class2path): get class path and check referable.
+	  [ruby-dev:22588]
+
+Thu Jan 15 12:58:26 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_read): do not append EOF.  (ruby-bugs-ja:PR#585)
+
+	* io.c (rb_io_fwrite): ad-hockery hack to get rid of HP-UX stdio
+	  weird behavior.  [ruby-dev:22424]
+
+Wed Jan 14 21:13:06 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/inlinetest.rb (InlineTest::eval_part): eval under the top
+	  level environment.
+
+Wed Jan 14 17:54:17 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/inlinetest.rb (InlineTest::loadtest): require instead of
+	  load, to get rid of multiple loading.
+
+Wed Jan 14 13:30:52 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/extconf.rb: wrapper iconv.rb is dependent on platform.
+
+Wed Jan 14 09:32:02 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* MANIFEST: add test/net/test_httpheader.rb. (commit miss?)
+
+Wed Jan 14 00:58:35 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_Proc): move SystemStackError from under
+	  StandardError to Exception.  [ruby-talk:89782]
+
+Tue Jan 13 18:03:02 2004  Ian Macdonald  <ian@caliban.org>
+
+	* file.c (rb_stat_wr, rb_stat_ww): New functions
+	  implementing new methods (File::Stat#world_readable?,
+	  File::Stat#world_writable?).
+
+Tue Jan 13 16:53:25 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c: omission of Date library code caused
+	  test suite failure. [ruby-core:2251]
+
+Tue Jan 13 16:50:03 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: use $0 as the default application class name.
+
+Tue Jan 13 14:48:00 2004  Ian Macdonald  <ian@caliban.org>
+
+	* lib/pathname.rb: New methods (Pathname#world_readable?,
+	  Pathname#world_writable?).
+
+Tue Jan 13 14:48:01 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (primary): allow no "when" case. [ruby-dev:22578]
+
+	* ruby.h (rb_class_of): reduce branch. [ruby-dev:22577]
+
+	* ruby.h (rb_type): ditto.
+
+Tue Jan 13 14:26:59 2004  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/test/unit/ui/testrunnerutilities.rb (TestRunnerUtilities):
+	  moved run method which allows output level.  [ruby-dev:22554]
+
+Tue Jan 13 13:04:24 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/test_*.rb: Pathname#parent -> Pathname#dirname.
+
+Tue Jan 13 11:38:58 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/yaml/test_yaml.rb (YAML_Unit_Tests::test_spec_type_{int,float}):
+	  fix syntax error.
+
+Tue Jan 13 07:52:40 2004  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/bytecode.c: turn off default implicit typing.
+
+	* ext/syck/implicit.c: detect base60 integers.
+
+	* ext/syck/rubyext.c: handle base60, as well as hex and octal
+	  with commas.  implicit typing of ruby symbols.
+
+	* test/yaml/test_yaml.rb: add test.
+
+Tue Jan 13 04:29:52 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_driver.rb (RiDriver::report_method_stuff):
+	  Show fully-qualified class names in class list.
+
+Tue Jan 13 01:24:17 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (test_wr): Rdoc fix.  [ruby-core:02225]
+
+Tue Jan 13 01:04:37 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_paths.rb (RI::Paths): First attempt at
+	  incorporating DESTDIR in the rdoc installation.
+
+Mon Jan 12 23:26:21 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): fix position after FCALL.  [ruby-dev:22574]
+
+Mon Jan 12 18:00:11 2004  Ian Macdonald  <ian@caliban.org>
+
+	* file.c (test_wr, test_ww): New functions implementing new
+	  methods (File::world_readable?, File::world_writable?).
+
+	* file.c (S_IRUGO, S_IGUGO): New macros.
+
+Mon Jan 12 12:07:22 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_methods):
+	  Someone changed the "// in eval.c" comments to "/*...*/" style,
+	  so the parsing of the source file name broke.
+
+	* object.c: Remove spurious space in TrueClass documentation.
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_body): Fix
+	  bad regexp: if the code before a documented method contained
+	  a comment that wasn't terminated by whitespace, that comment
+	  and all intervening code was included in the following
+	  method's documentation.
+
+	* lib/rdoc/ri/ri_formatter.rb (RI::HtmlFormatter::break_to_newline):
+	  HTML formats need explicit line breaks.
+
+Mon Jan 12 02:24:07 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_formatter.rb (RI::HtmlFormatter): Add HTML
+	  generation support to ri (Elliot Hughes)
+
+Sun Jan 11 23:54:41 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* env.h (ruby_frame, ruby_scope, ruby_in_eval, ruby_class,
+	  ruby_dyna_vars): export.  [ruby-dev:22566]
+
+Sun Jan 11 02:35:53 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (make_hostent): a bug in brace position.
+
+	* configure.in: install rdoc by default. if you do not want to
+	  install rdoc, specify --disable-install-doc.
+
+Sun Jan 11 02:07:47 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_options.rb (RI::Options::OptionList::OptionList):
+	  Also accept command line options via the 'RI' environment variable.
+
+Sat Jan 10 21:27:41 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (eval): need to add message delimiter. [ruby-dev:22561]
+
+Sat Jan 10 01:54:50 2004  Eric Sunshine  <sunshine@sunshineco.com>
+
+	* defines.h (__NeXT__): Ensure that all standard S_IRUSR, S_IWGRP,
+	  S_IRWXO, etc. macros are defined since future code might require
+	  them (even though present code only requires a subset).
+
+	* defines.h (__NeXT__): Bug fix: WORDS_BIGENDIAN was not being set
+	  correctly on Rhapsody when -arch compiler flag was used (via
+	  configure's --enable-fat-binary option).
+
+Sat Jan 10 23:01:41 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBPATHFLAG, RPATHFLAG): enclose paths with single
+	  quotes.  [ruby-dev:22564]
+
+	* lib/mkmf.rb (libpathflag): do not enclose with quotes always.
+
+	* {bcc32,win32,wince}/Makefile.sub (LIBPATHFLAG): quoted.
+
+Sat Jan 10 22:46:18 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_inspect): new method, Dir#inspect.  [ruby-dev:22562]
+
+Fri Jan  9 17:36:51 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/socket.c (make_hostent): getaddrinfo(3) on BSD do not
+	  fill ai_canonname if serv is not supplied.  (ruby-bugs PR#1243)
+
+Fri Jan  9 13:14:59 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb: do not ignore exceptions(LoadError
+	  and SystemExitError) while loading a testcase.  smell of bug.
+
+	* test/testunit/collector/test_dir.rb: add new test of the LoadError.
+
+	* test/drb/{test_drbssl.rb,test_drbunix.rb}: do not define testcase if
+	  openssl is not installed.
+
+	* test/testunit/collector/test_dir.rb: assert_raises -> assert_raise.
+
+Fri Jan  9 11:52:16 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* rubysig.h: <errno.h> is needed to use errno which may be a macro.
+
+Fri Jan  9 11:20:24 2004  Siena.  <siena@faculty.chiba-u.jp>
+
+	* ext/extmk.rb (extmake): should not reduce necessary libraries.
+	  [ruby-dev:22440]
+
+	* lib/mkmf.rb (merge_libs): merge libraries according to
+	  dependency.
+
+Fri Jan  9 10:05:23 2004  Siena.  <siena@faculty.chiba-u.jp>
+
+	* lib/mkmf.rb (libpathflag): use single quotes.  [ruby-dev:22440]
+
+Thu Jan  8 23:49:21 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (RDOCTARGET): new macro.  if you want to install
+	  rdoc documentation, you need to run configure with
+	  --enable-install-doc.
+
+Thu Jan  8 21:17:43 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_pkey.c (ossl_pkey_to_der): removed; it returns
+	  public key only.
+
+	* ext/openssl/ossl_pkey_dh.c (ossl_dh_to_der): new function for
+	  OpenSSL::PKey::DH#to_der.
+
+	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_to_der): new function for
+	  OpenSSL::PKey::DSA#to_der.
+
+	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_to_der): new function for
+	  OpenSSL::PKey::RSA#to_der.
+
+Thu Jan  8 18:25:29 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): should not recurse in exceptional status.
+
+Thu Jan  8 16:51:04 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/wsdl/datetime/test_datetime.rb: fixed a stupid testcase which
+	  dumps "E" at month-end.
+
+Thu Jan  8 11:20:01 2004  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* eval.c, object.c, process.c, re.c: don't use C++ style comments.
+
+Thu Jan  8 08:46:14 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c (yaml_org_handler): lazy-load Date for
+	  static-ext.
+
+Thu Jan  8 07:06:30 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: preserve order in Setup.  [ruby-dev:22503]
+
+	* ext/extmk.rb: move dependent libraries just after depended
+	  libraries.
+
+	* ext/digest/*/extconf.rb: depend on digest.
+
+Thu Jan  8 04:36:17 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (WEBrick::CGI#initialize): should create
+	  @config[:Logger] if it was not given.
+
+Wed Jan  7 22:28:12 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): fix memory leak.
+
+Wed Jan  7 21:15:07 2004  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* sample/webrick/*: new files.
+
+	* MANIFEST: add sample/webrick/*
+
+Wed Jan  7 20:51:51 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/net/test_httpheader.rb: new file.
+
+	* MANIFEST: add test/net/test_httpheader.rb.
+
+Wed Jan  7 20:42:06 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (HTTPHeader#content_length): should return nil
+	  unless header exists. [ruby-dev:22519]
+
+Wed Jan  7 14:26:05 2004  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb (TkPanedWindow): use epath for embedded windows.
+
+	* ext/tk/lib/tktext.rb: use epath for embedded windows.
+
+	* ext/tk/lib/tkcanvas.rb: use epath for window items.
+
+Wed Jan  7 14:24:04 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/{attachment.rb,mimemessage.rb}: added from soap4r/1.5.2.
+
+Wed Jan  7 13:00:18 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/ri/ri_driver.rb: Fix problem where ri was
+	  being too eager to find matches of ambiguous method
+	  names (such as "ri Thread.join" would return both
+	  Thread.join and ThreadsWait.join)
+
+Wed Jan  7 12:35:41 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/debug.rb: revert command parse regexps.  [ruby-list:39014] by
+	  Shirai,Kaoru.
+
+Wed Jan  7 08:21:04 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parserfactory.rb: Check for shebang
+	  line in files that would otherwise be treated as
+	  plain text.
+
+Tue Jan  6 22:13:34 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_mod_modfunc): should break if m has no super class.
+	  [ruby-dev:22498]
+
+Tue Jan  6 21:51:37 2004  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (fptr_finalize): should save errno just after failure.
+	  [ruby-dev:22492]
+
+Tue Jan  6 20:51:10 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb(Logger#msg2str): no special treatment for the object
+	  which responds to :to_str.
+
+	* lib/logger.rb(LogDevice#initialize): remove type checking if the
+	  given object is a String.  Kernel.open handles it correctly.
+
+	* test/logger/test_logger.rb: follow above change (ArgumentError ->
+	  TypeError.)
+
+Tue Jan  6 14:53:14 2004  Dave Thomas  <dave@pragprog.com>
+
+	* bin/ri: split out the display side, making it pluggable. Added
+	  new ri_driver and ri_display files in lib/rdoc/ri.
+
+Tue Jan  6 11:29:43 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/inlinetest.rb, test/{test_generator.rb,test_ipaddr.rb,
+	  test_pathname.rb,test_pp.rb,test_prettyprint.rb,test_set.rb,
+	  test_time.rb,test_tsort.rb: added.
+
+Tue Jan  6 09:38:27 2004  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* import soap4r/1.5.2;
+
+	* lib/soap/{attachment.rb,baseData.rb,encodingstyle/soapHandler.rb}:
+	  introduce SOAPExternalReference class as a referenct to SOAPEnvelope
+	  external content.
+
+	* lib/soap/{attachment.rb,mimemessage.rb}: great SwA (SOAP messages
+	  with Attachments) support code by Jamie Herre.
+
+	* lib/soap/{element.rb,marshal.rb,parser.rb,processor.rb,
+	  streamHandler.rb,wsdlDriver.rb}: SwA support.
+
+	* lib/soap/rpc/{cgistub.rb,driver.rb,element.rb,proxy.rb,router.rb,
+	  soaplet.rb}: SwA support and refactoring.
+
+	* lib/soap/generator.rb, lib/soap/mapping/mapping.rb: follow
+	  SOAPReference#initialize signature change.
+
+	* lib/soap/mapping/factory.rb: deleted unused methods.
+
+	* lib/soap/mapping/rubytypeFactory.rb: do no ignore case while xsi:type
+	  string <-> Ruby class name matching.
+
+	* lib/xsd/datatypes.rb: check the smallest positive non-zero
+	  single-precision float exactly instead of packing with "f".
+	  [ruby-talk:88822]
+
+	* test/soap/test_basetype.rb, test/xsd/test_xsd.rb: use 1.402e-45, not
+	  1.4e-45.  1.4e-45 is smaller than 2 ** -149...
+
+	* test/soap/test_basetype.rb, test/soap/marshal/test_marshal.rb,
+	  test/xsd/test_xsd.rb: use "(-1.0 / (1.0 / 0.0))" instead of "-0.0".
+
+	* test/soap/test_streamhandler.rb: revert to the previous test that
+	  warns "basic_auth unsupported under net/http".
+
+Tue Jan  6 06:37:53 2004  Dave Thomas  <dave@pragprog.com>
+
+	* bin/rdoc: Add --ri-system switch
+
+	* lib/.document: Update with list of files that seem to have
+	  documentation
+
+	* lib/test/unit.rb: Reorder comment to make it RDoc friendly.
+
+	* Makefile.in: add install-nodoc target, and make it
+	  generate RDoc on default install.
+
+	* lib/rdoc/ri/ri_options.rb (RI::Options::parse): Add
+	  --doc-dir option to ri.
+
+Tue Jan  6 00:04:40 2004  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method_or_yield_parameters):
+	  fix parsing if there are braces in a method parameter list
+
+Tue Jan  6 01:01:04 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/dir.h, win32/win32.c: fix patch miss.
+
+	* win32/Makefile.sub: fix file dependency.
+
+Mon Jan  5 20:32:00 2004  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/logger.rb: enhanced documentation.
+
+Mon Jan  5 18:58:47 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c: merge tuning from H.Yamamoto <ocean@m2.ccsnet.ne.jp>.
+	  [ruby-dev:22486]
+
+	* pack.c (pack_unpack): unpack requires big endian offet (OFF16B
+	  and OFF32B).  The patch is from Minero Aoki in [ruby-dev:22489]
+
+	* pack.c (OFF16B): add big-endian offset again.
+
+Mon Jan  5 03:00:53 2004  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_pack.rb: new test test_unpack_N.
+
+Mon Jan  5 01:47:53 2004  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): remove duplicated object files
+	  from $objs on DOSISH platforms.
+
+Sat Jan  3 02:44:48 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* rubysig.h (TRAP_END): preserve errno before switching context.
+	  [ruby-core:02137]
+
+Sat Jan  3 01:18:08 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c: merge tuning from H.Yamamoto <ocean@m2.ccsnet.ne.jp>.
+	  [ruby-dev:22476]
+
+Fri Jan  2 14:54:11 2004  Dave Thomas  <dave@pragprog.com>
+
+	* bin/ri: Add new --classes option, and arrange for
+	  help messages to be paged too.
+
+	* bin/rdoc: Add statistics.
+
+	* process.c: (MG) Added Process documentation
+
+	* lib/rdoc/ri/ri_formatter.rb (RI::AttributeFormatter::wrap):
+	  Fix problem with labels not displaying in RI labeled
+	  lists using BS and ANSI modes.
+
+Fri Jan  2 01:50:13 2004  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_eof): ARGF.eof? should not have any side effect.
+	  [ruby-dev:22469]
+
+Thu Jan  1 09:03:20 2004  Dave Thomas  <dave@pragprog.com>
+
+	* bin/ri (report_class_stuff): Fix problem with ambiguous nested
+	  classes not matching.
+
+Wed Dec 31 17:25:17 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_each_byte): should return self.  [ruby-dev:22465]
+
+Wed Dec 31 15:05:00 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/pathname.rb: Corrected small coding error.
+
+Wed Dec 31 15:00:00 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/pathname.rb: Completed documentation.
+
+Wed Dec 31 11:20:34 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::do_methods): Make
+	  file referenced in "// in sss.c" relative to current file.
+
+Wed Dec 31 11:17:37 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/generators/html_generator.rb: Fix problem when
+	  a public method was aliased, but the alias is then
+	  made private, and hence doesn't appear in RDoc output.
+
+Wed Dec 31 01:33:05 2003  Dave Thomas  <dave@pragprog.com>
+
+	* array.c, error.c, eval.c, io.c, prec.c, range.c, re.c,
+	  string.c, time.c: Add RDoc for Kernel functions, and tidy.
+
+Tue Dec 30 19:39:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_readline): should raise EOFError at the end of
+	  files.  [ruby-dev:22458]
+
+	* io.c (argf_read): should concatenate input files when length
+	  argument is nil. [ruby-dev:22450]
+
+	* io.c (argf_read): should update supplied string buffer (2nd
+	  argument) even when IO#read is called multiple times.
+
+	* io.c: should initialize lineno by zero. [ruby-dev:22460]
+
+Tue Dec 30 12:30:30 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/code_objects.rb (RDoc::Context::find_symbol): If a
+	  class and a method have the same name, finding Xxx.abc was trying
+	  to find 'abc' in method 'Xxx', not class 'Xxx'.
+
+Tue Dec 30 08:32:32 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_rb.rb (RDoc::RubyParser::parse_method):
+	  Handle undoing nsting of yield parameters correctly for:
+
+	    def each_entry(&b) Dir.foreach(@path) {|f| yield P.new(f) } end
+
+Tue Dec 30 07:30:00 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/pathname.rb: Added documentation.
+
+Mon Dec 29 20:08:17 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (GenericRequest#initialize): check if path
+	  begins with '/'.
+
+	* lib/net/http.rb: def m( arg ) -> def m(arg)
+
+Mon Dec 29 12:51:02 2003  Dave Thomas  <dave@pragprog.com>
+
+	* eval.c: Add RDoc for Kernel global functions.
+
+Mon Dec 29 11:00:16 2003  Dave Thomas  <dave@pragprog.com>
+
+	* array.c: Tidy up RDoc loose ends.
+
+Mon Dec 29 05:05:51 2003  Dave Thomas  <dave@pragprog.com>
+
+	* struct.c, random: Add RDoc comments
+
+Mon Dec 29 02:25:00 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/optparse.rb: Improved documentation.
+
+Mon Dec 29 02:20:54 2003  Dave Thomas  <dave@pragprog.com>
+
+	* eval.c: Add RDoc for class Proc, Method, UnboundMethod
+
+Mon Dec 29 02:20:26 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* instruby.rb: fix install directory if destdir and compile_dir are
+	  not in the same drive.
+
+	* ext/extmk.rb: ditto. [ruby-list:39009]
+
+	* win32/Makefile.sub, win32/README.win32, win32/configure.bat,
+	  win32/setup.mak: new configure scheme. use ``configure --prefix=dir''
+	  instead of ``nmake DESTDIR=dir install''.
+
+Mon Dec 29 00:41:44 2003  Dave Thomas  <dave@pragprog.com>
+
+	* math.c: Add RDoc comments
+
+Sun Dec 28 20:19:11 2003  Tanaka Akira  <akr@m17n.org>
+
+	* ext/stringio/stringio.c (strio_sysread): StringIO.new.sysread didn't
+	  raise EOFError.
+
+	* ext/zlib/zlib.c (gzreader_gets): don't increment lineno when
+	  gzfile_read_all returns "".
+
+Sun Dec 28 15:25:08 2003  Dave Thomas  <dave@pragprog.com>
+
+	* class.c,object.c,parse.y,sprintf.c,variable.c: Document classes
+	  Object, Module, etc...
+
+Sun Dec 28 11:55:29 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/csv/test_csv.rb: generate bom.csv and mac.csv files on the fly.
+	  [ruby-talk:88852]
+
+	* test/csv/{bom.csv,mac.csv}: removed.
+
+Sun Dec 28 08:56:51 2003  Dave Thomas  <dave@pragprog.com>
+
+	* eval.c: Thead[Group] RDoc (thanks to MG)
+
+Sun Dec 28 03:50:05 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_override_comment):
+	  Escape method names used in regexp
+
+Sun Dec 28 01:46:02 2003  Dave Thomas  <dave@wireless_3.local.thomases.com>
+
+	* lib/rdoc/ri/ri_formatter.rb (RI::TextFormatter::display_flow_item):
+	  Add support for rules in 'ri' output.
+
+Sun Dec 28 01:35:35 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_body):
+	  Sometimes the Ruby source aliases two otherwise
+	  unrelated methods (for example Kernel#object_id and
+	  Kernel#hash are both the same C function). Provide a
+	  facility to allow the methods to be documented
+	  separately.
+
+Sun Dec 28 01:05:31 2003  Dave Thomas  <dave@pragprog.com>
+
+	* marshal.c, signal.c: RDoc collemts added by Elliott Hughes
+
+Sun Dec 28 00:46:25 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/parsers/parse_c.rb (RDoc::C_Parser::find_class_comment):
+	  Some Ruby source uses lower-case class names for the
+	  Init_Xxx C function name.
+
+Sat Dec 27 23:41:46 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: fix "test: too many arguments" error.
+
+Sat Dec 27 15:32:40 2003  Dave Thomas  <dave@wireless_3.local.thomases.com>
+
+	* time.c: Add RDoc comments for Time class.
+
+Sat Dec 27 15:07:26 2003  Dave Thomas  <dave@pragprog.com>
+
+	* object.c: Add RDoc comments for Symbol class.
+
+Sat Dec 27 14:39:53 2003  Dave Thomas  <dave@pragprog.com>
+
+	* numeric.c (Init_Numeric): Add RDoc comments.
+
+Sat Dec 27 00:44:00 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (next_argv): warn always for stdin on inplace edit mode.
+
+	* io.c (read_all): need to check string value.
+
+	* io.c (argf_read): allow ARGF.read(nil).  [ruby-dev:22433]
+
+Fri Dec 26 23:02:09 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_backquote): need not to check nil result.
+	  [ruby-core:02078]
+
+	* io.c (rb_io_getline): should return nil when read_all gives
+	  empty string, even when nil rs is specified. [ruby-core:02077]
+
+Fri Dec 26 18:33:54 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if getcontext and setcontext are available.
+
+	* eval.c: use presence of getcontext/setcontext.
+
+Fri Dec 26 16:40:53 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (PathnameTest#test_plus): add 2 assertions.
+
+Fri Dec 26 14:05:13 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_pack.rb: new test test_pack_N.
+
+Fri Dec 26 12:53:26 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): add sign check for 'i', and 'l'.
+	  [ruby-dev:22427]
+
+	* bignum.c (rb_quad_pack): add range check for 'quad int'.
+
+Fri Dec 26 10:58:58 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* MANIFEST: add vms/config.h and remove vms/config.h_in.
+
+Fri Dec 26 10:42:00 2003  AKIYOSHI, Masamichi  <masamichi.akiyoshi@hp.com>
+
+	* io.c: [VMS] "rfm=stmlf" is specified for open() and fopen().
+
+Thu Dec 25 22:29:53 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_update): don't return any value.
+
+Thu Dec 25 15:30:17 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_update): call rb_str_modify().
+
+Thu Dec 25 05:08:09 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (search_required): search actual file name once when no
+	  extension specified.
+
+Thu Dec 25 04:00:44 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* stable version 1.8.1 released.
+
+Thu Dec 25 00:17:53 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* configure.in: check for nanosleep, -lrt if required.
+	  [ruby-core:02059]
+
+	* eval.c (thread_timer): use select(2) if nanosleep(2) is not
+	  available.
+
+	* eval.c: check __stub_getcontext for glibc on some platforms.
+	  [ruby-list:38984]
+
+Wed Dec 24 23:48:04 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/test_basetype.rb, test/soap/marshal/test_marshal.rb
+	  test/xsd/test_xsd.rb: use "(-1.0 / (1.0 / 0.0))" instead of "-0.0"
+	  to express -0.0.  [ruby-talk:88786]
+
+Wed Dec 24 23:29:30 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/tsort.rb (test_orphaned_break): removed.
+
+Wed Dec 24 20:53:06 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tkmulticolumnlist.rb: new sample
+
+	* ext/tk/sample/tkmultilistframe.rb: bug fix
+
+Wed Dec 24 20:37:37 2003  Eric Sunshine  <sunshine@sunshineco.com>
+
+	* configure.in (LDSHARED): Fixed typographical error in assignment of
+	  LDSHARED for Rhapsody which caused linking of extension modules to
+	  fail.
+
+Wed Dec 24 17:51:18 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_thread_flock): enable thread support again.
+
+Wed Dec 24 16:46:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (catch_timer): do not call rb_thread_schedule() inside to
+	  avoid pthread_mutex_lock() deadlock.  interrupts to system calls
+	  are detected by TRAP_END via EINTR error.
+
+	* eval.c (thread_timer): do not post signal unless it is
+	  absolutely necessary.
+
+	* rubysig.h (TRAP_END): add CHECK_INTS to switch thread.
+
+	* regex.c (re_compile_pattern): check if nextp is smaller than
+	  pend.  [ruby-dev:22372]
+
+	* eval.c (umethod_bind): remove method overridden check.
+	  [ruby-dev:22366]
+
+Wed Dec 24 16:13:05 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read): should check for error
+	  status by SSL_get_error().
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_write): ditto.
+
+Wed Dec 24 14:23:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): clear the buffer argument
+	  when returning nil.  [ruby-dev:22363]
+
+	* test/ruby/ut_eof.rb (TestEOF::test_eof_0, TestEOF::test_eof_1):
+	  add buffer argument tests.
+
+Wed Dec 24 14:07:55 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: Modules are allowed to rescue.
+
+	* lib/test/unit/autorunner.rb: show output_level in order.
+
+	* lib/test/unit/collector/dir.rb: get rid of successive same
+	  directories in load path.
+
+	* test/testunit/test_assertions.rb (test_assert_nothing_raised,
+	  test_assert_raise): test for modules.
+
+Wed Dec 24 13:43:34 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (authenticate): remove "\n" from base64 encoded
+	  strings.
+
+Wed Dec 24 11:26:41 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb: should not create any
+	  files or directories in current directory.  [ruby-talk:88724]
+
+Wed Dec 24 10:29:53 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): never return nil at
+	  unlimited read.  [ruby-dev:22334]
+
+	* ext/stringio/stringio.c (strio_read): support second
+	  argument.  [ruby-dev:22350]
+
+Wed Dec 24 09:38:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (arg): should return 0 after error.  [ruby-dev:22360]
+
+Wed Dec 24 00:56:54 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (read_all): do not return nil at the end of file.
+	  [ruby-dev:22334]
+
+	* io.c (argf_read): do not depend on nil at eof behavior of
+	  IO#read().
+
+	* eval.c (rb_thread_join): dup exception before re-raising it.
+
+	* io.c (rb_io_eof): call clearerr() to prevent side effect.  this
+	  patch is supplied by Masahiro Sakai <sakai@tom.sfc.keio.ac.jp>.
+	  [ruby-dev:22234]
+
+	* pack.c (OFF16): get offset for big endian machines.
+
+	* pack.c (pack_pack): use OFF16 instead of OFF16B.
+	  [ruby-dev:22344]
+
+	* pack.c (pack_unpack): ditto.
+
+Tue Dec 23 22:47:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_io_check_readable): set FMODE_RBUF always, even if
+	  NEED_IO_SEEK_BETWEEN_RW is not defined. [ruby-dev:22340]
+
+	* io.c (rb_io_check_writable): clear FMODE_RBUF before writing
+	  something.
+
+Tue Dec 23 22:25:00 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/optparse.rb: incomplete RDoc documentation added in place of
+	  existing RD comments.  Tabs converted to spaces.
+
+Tue Dec 23 19:44:47 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/test_streamhandler.rb (test_basic_auth): removed.
+	  soap4r + basic_auth is not officially supported in ruby/1.8.1 even
+	  though soap4r + basic_auth + http-access2 should run fine.
+
+Tue Dec 23 19:42:59 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_ungetc): raise an exception at unread stream to
+	  avoid unspecified behavior.  [ruby-dev:22330]
+
+	* test/ruby/test_system.rb (test_syntax): glob relatively from
+	  __FILE__.
+
+Tue Dec 23 18:09:40 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (pack_pack): remove unnecessary negative value check.
+	  [ruby-dev:22329]
+
+Tue Dec 23 17:26:55 2003  KONISHI Hiromasa  <konishih@fd6.so-net.ne.jp>
+
+	* bcc32/Makefile.sub (config.h): bcc has finite(). [ruby-list:38940]
+
+Tue Dec 23 16:08:16 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/rexml/encodings/US-ASCII.rb: typo.  [ruby-talk:88650]
+
+	* test/ruby/test_system.rb: num of asserts depended on running dir.
+
+	* test/xsd/test_noencoding.rb: rexml + without iconv/uconv cannot
+	  handle euc-jp.  install iconv, uconv or xmlscan.
+
+Tue Dec 23 14:13:51 2003  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic::check_userinfo,
+	  URI::Generic::check_user, URI::Generic::check_password): tests
+	  conflicts/depends with other components closely.
+
+	* test/uri/test_generic.rb (TestGeneric::test_set_component):
+	  added tets.
+
+Tue Dec 23 11:08:34 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/xsd/test_noencoding.rb: rescue Errno::EINVAL and do not test.
+	  "euc-jp" might not be in supported encoding name list.
+	  [ruby-talk:88650]
+
+Tue Dec 23 06:10:31 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb (CGI): add support for mod_ruby.
+
+	* lib/webrick/cgi.rb (CGI::Socket): add check for existence of
+	  OpenSSL module in all HTTPS related methods.
+
+	* lib/webrick/cgi.rb (CGI::Socket#cipher): should create similar
+	  value to OpenSSL::SSLSocket#cipher.
+
+	* lib/webrick/httpresponse.rb (HTTPResponse#setup_header): should
+	  set "connection: close" if @keep_alive is false.
+
+	* lib/webrick/https.rb (HTTPrequest#meta_vars): add supprt for
+	  SSL_PROTOCOL, SSL_CIPHER_USEKEYSIZE and SSL_CIPHER_ALGKEYSIZE.
+
+Mon Dec 22 23:00:05 2003  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic::check_opaque): fixed typo.
+
+Mon Dec 22 21:59:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (map_charset): always ensure code is a String.
+
+Mon Dec 22 21:15:29 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mod_init_copy): always copy singleton class.
+	  [ruby-dev:22325]
+
+Mon Dec 22 20:44:36 2003  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#route_from): accepts urls which
+	  has no host-part.
+
+	* test/uri/test_generic.rb (TestGeneric::test_route): added a test.
+
+Mon Dec 22 20:38:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi.rb: reduce eval.
+
+	* lib/cgi.rb (CGI::QueryExtension::read_multipart): alias path to
+	  local_path.  [ruby-list:38883]
+
+Mon Dec 22 20:09:31 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/test_property.rb: remove duplicated test method.
+
+Mon Dec 22 18:22:04 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bcc32/Makefile.sub, win32/Makefile.sub (config.h): remove
+	  HAVE_ISINF definition to follow previous commits of missing.h
+	  and win32/win32.h.
+
+Mon Dec 22 17:23:42 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func_setitimer): moved from defines.h
+
+	* defines.h, rubysig.h, signal.c: removed macro handling which
+	  should be done in configure.
+
+	* configure.in (intrinsics.h): check if present.
+
+	* ruby.h: include intrinsics.h if available.
+
+	* bignum.c, marshal.c: include ieeefp.h if available.
+
+	* missing.h (isinf): define as a macro if finite() and isnan()
+	  are available.  [ruby-core:02032]
+
+Mon Dec 22 17:07:31 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in (mingw): set isnan, finite and isinf to yes.
+
+Mon Dec 22 13:40:19 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/property.rb: passing block by reference.
+
+Mon Dec 22 00:32:43 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_with_disable_interrupt): use ENABLE_INTS instead of
+	  ALLOW_INTS which may switch context.  [ruby-dev:22319]
+
+	* ext/syck/emitter.c (syck_emitter_write): str bigger than
+	  e->bufsize causes buffer overflow.  [ruby-dev:22307]
+
+Sun Dec 21 17:29:00 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_check_inheritable): new function.  [ruby-dev:22316]
+
+	* intern.h: add prototype.
+
+	* eval.c (superclass): use rb_check_inheritable().
+
+	* object.c (rb_class_initialize): check argument validity.
+
+Sun Dec 21 16:25:10 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#+): re-implemented to resolve ".." in
+	  beginning of the argument.
+	  (Pathname#join): concatenate from the last argument.
+	  (Pathname#parent): just use Pathname#+.
+
+Sun Dec 21 00:12:37 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: add new methods (TkScrollbar#assign, assign_list)
+
+	* ext/tk/sample/tkmultilistframe.rb: use TkScrollbar#assign method
+
+Sat Dec 20 21:59:03 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httprequest.rb (HTTPRequest#meta_vars): refine regexp.
+
+	* lib/webrick/cgi.rb (CGI#start): NPH scripts return status line
+	  instead of Status: header field.
+
+	* lib/webrick/cgi.rb (CGI::Socket): refine some coditions.
+
+Sat Dec 20 16:07:14 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Completion::complete): wrong
+	  Regexp for word boundary.  pointed out by Gavin Sinclair.
+
+	* lib/optparse.rb (OptionParser::make_switch): [no-] prefix was
+	  missing.
+
+Sat Dec 20 11:40:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/yaml.rb (YAML::YAML): adjust Marshal version.
+
+Sat Dec 20 03:56:02 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_with_disable_interrupt): prohibit thread context
+	  switch during proc execution.  [ruby-dev:21899]
+
+Sat Dec 20 02:41:02 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/cgi.rb: add file. (yet another CGI library)
+
+	* MANIFEST: add lib/webrick/cgi.rb.
+
+Sat Dec 20 02:18:31 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-calculate-indent): proper indentation
+	  inside of parentheses.  [ruby-dev:22308]
+
+Fri Dec 19 21:24:22 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httprequest.rb (HTTPRequest#meta_vars): should not set
+	  HTTP_CONTENT_TYPE and HTTP_CONTENT_LENGTH.
+
+	* lib/webrick/https.rb (HTTPRequest#parse): should check presence
+	  of cert() method to detect SSLSocket.
+
+Fri Dec 19 22:56:46 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/property.rb (SOAP::Property#load): new method for loading
+	  property value into existing property tree.
+
+	* test/soap/test_property.rb: add test.
+
+Fri Dec 19 19:21:49 2003  akira yamada  <akira@ruby-lang.org>
+
+	* lib/runit/cui/testrunner.rb (RUNIT::CUI::TestRunner::run):
+	  should use Test::Unit::UI::{PROGRESS_ONLY,VERBOSE}.
+
+Fri Dec 19 17:36:49 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tkmultilistbox.rb: bug fix
+
+	* ext/tk/sample/tkmultilistframe.rb: new sample script
+
+Fri Dec 19 03:44:27 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (parse_form_data): should return an
+	  empty Hash if the body is empty.
+
+Thu Dec 18 21:47:35 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): should remove deffile if it's
+	  made by miniruby. based on nobu's patch.
+
+Thu Dec 18 21:44:21 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (stack_extend): ignore inline optimization on VC7.
+
+	* win32/Makefile.sub (OS, RT): can override.
+
+	* win32/Makefile.sub (LDFLAGS): ditto. shouldn't use pdb:none
+	  option. based on Tietew's patch [ruby-dev:22289]
+
+Thu Dec 18 16:38:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (fnmatch): unlike find_dirsep(), rb_path_next() never
+	  return NULL.
+
+Thu Dec 18 15:27:59 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/ipaddr.rb (IPSocket::getaddress): merge usa's patch.
+	  [ruby-dev:21678]
+
+Wed Dec 17 15:15:30 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/cgi.rb (CGI::QueryExtension::Value::[]): should work like
+	  String#[] if more than one arguments are specified.
+
+	* lib/delegate.rb: avoid using common instance name as "@obj".
+
+	* lib/cgi.rb (CGI::QueryExtension::Value): Value is no longer
+	  subclass of String, but DelegateClass(String).
+
+	* ext/curses/extconf.rb: restore function check for init_color.
+	  [ruby-list:38905]
+
+	* Makefile.in: need to specify $(MAINLIBS) for the miniruby
+	  generation rule.
+
+	* configure.in: better FreeBSD -lc_r support.
+
+Wed Dec 17 00:16:14 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/strscan/strscan.c: new method
+	  StringScanner#beginning_of_line? (alias #bol?)
+
+	* ext/strscan/strscan.c: new method StringScanner#concat and #<<.
+
+	* ext/strscan/strscan.c: StringScanner#new(str) does not duplicate
+	  nor freeze STR (allow destructive modification).
+
+	* test/strscan/test_stringscanner.rb: test new methods above.
+
+	* test/strscan/test_stringscanner.rb: test destructive string
+	  modification.
+
+Tue Dec 16 21:20:47 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pp.rb: don't use local variable `pp'.
+
+	* lib/prettyprint.rb: ditto.
+
+Tue Dec 16 13:20:43 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: condition bug of if statement on
+	  {pack,grid}_propagate methods
+
+Tue Dec 16 03:17:29 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb: comments in strings. [ruby-talk:88012]
+
+	* test/yaml/test_yaml.rb: add test.
+
+Tue Dec 16 01:14:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (catch_timer): check rb_thread_crtical in main native
+	  thread.
+
+	* eval.c (thread_timer): just sends signals periodically, to
+	  prevent main native thread from receiving them in critical
+	  section.  [ruby-core:01959]
+
+Mon Dec 15 13:32:22 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (check_dirname): check string safety and remove extraneous
+	  trailing directory separators.  [ruby-dev:22279]
+
+	* file.c: renamed and externalized rb_path_next,
+	  rb_path_skip_prefix, rb_path_last_separator, rb_path_end.
+
+	* intern.h: prototypes for rb_path_next, rb_path_skip_prefix,
+	  rb_path_last_separator, rb_path_end.
+
+Mon Dec 15 09:27:46 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_pkcs12.c (ossl_pkcs12_initialize): first argument
+	  of rb_protect should take an argument of VALUE.
+
+Sun Dec 14 18:46:48 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/socket/socket.c (Init_socket): IPv6 is not supported although
+	  AF_INET6 is defined on MinGW.
+
+	* lib/ipaddr.rb (AF_INET6): workaround in the environment which does
+	  not support IPv6.
+
+Sat Dec 13 18:55:16 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/charset_alias.rb: preserve original order.
+
+	* ext/iconv/extconf.rb: remove wrapper file at clean.
+
+Sat Dec 13 18:09:42 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (thread_timer): use timer by sub-thread and nanosleep.
+	  [ruby-talk:87519]
+
+	* gc.c (Init_stack): no stack adjustment for THREAD_SAFE.
+
+Sat Dec 13 17:17:59 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (proc_alloc): cache the created object at first time.
+	  [ruby-talk:61288], [ruby-dev:22240]
+
+Sat Dec 13 09:01:23 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check ucontext.h.
+
+	* eval.c: use getcontext/setcontext() instead of setjmp/longjmp()
+	  on ia64 or with native thread enabled.  [ruby-core:01932]
+
+Sat Dec 13 03:09:14 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* lib/yaml/rubytypes.rb: anonymous struct fix. [ruby-core:01946]
+
+	* test/yaml/test_yaml.rb: add test.
+
+Fri Dec 12 22:36:44 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb: add Cell#to_str and Cell#to_s for /.../ =~ aCell,
+	  "#{aCell}" and so on.
+
+	* test/csv/test_csv.rb: add tests.
+
+Fri Dec 12 19:33:06 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mkdir): remove trailing `/' from pathes.
+
+	* lib/fileutils.rb (rmdir): ditto. [ruby-dev:22238]
+
+	* lib/fileutils.rb (rmdir_r): ditto.
+
+	* lib/fileutils.rb (fu_copy_dir): check if it is a directory after
+	  mkdir(2).
+
+Fri Dec 12 06:06:09 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (proc_invoke): fix class name in warning message for
+	  define_method.  [ruby-dev:22235]
+
+Thu Dec 11 21:24:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_pkcs12.[ch]: new files. add OpenSSL::PKCS12.
+
+	* ext/openssl/ossl.[ch]: ditto.
+
+	* ext/openssl/MANIFEST: add ossl_pkcs12.[ch].
+
+Thu Dec 11 20:54:28 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mkdir_p): remove trailing `/' befere mkdir(2).
+	  mkdir("nonexistdir/") does not work on NetBSD/Alpha 1.6.1.
+
+	* lib/fileutils.rb (fu_list): call to_str for all arguments.
+
+Thu Dec 11 20:07:01 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/ftools.rb (makedirs): sync with fileutils.
+
+Thu Dec 11 19:53:03 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mkdir_p): catch all SystemCallErrors.
+	  (mkdir("C:\") causes EACCESS on Windows 2000/NTFS)
+
+Thu Dec 11 19:08:02 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (mkdir_p): check if it is a directory after
+	  mkdir(2) instead of before mkdir(2), to avoid race condition.
+	  [ruby-talk:87730]
+	  Refer: mkinstalldirs sh script, GNU mkdir(1) (coreutils 5.0)
+
+Thu Dec 11 18:49:30 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb: def m( arg ) -> def m(arg).
+
+Thu Dec 11 11:39:43 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ieeefp.h), numeric.c: needed for finite() on
+	  Solaris.  [ruby-core:01921]
+
+	* file.c (rb_stat_inspect): adjust format specifier.
+
+	* parse.c (arg_prepend): nodetype() is for debug use.
+
+	* ruby.h (ISASCII, etc): cast to int to get rid of warning.
+
+	* ruby.h (alloca.h): include even in GCC.  [ruby-core:01925]
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValue): adjust format
+	  specifier.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_prec, BigDecimal_coerce,
+	  BigDecimal_divmod): use rb_assoc_new() to suppress memory usage.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_split): ditto.
+
+	* ext/dl/sym.c (rb_dlsym_guardcall): guard itself should be
+	  volatile.
+
+	* ext/iconv/iconv.c (iconv_convert): ensure actual parameter with
+	  format specifier.
+
+	* ext/pty/pty.c (MasterDevice, SlaveDevice, deviceNo): do not
+	  define unless used.
+
+	* ext/pty/pty.c (getDevice): get rid of warning.
+
+	* ext/socket/socket.c (port_str, sock_s_getaddrinfo,
+	  sock_s_getnameinfo): FIX2INT() now returns long.
+
+	* ext/socket/socket.c (init_inetsock_internal): uninitialized
+	  variable.
+
+	* ext/syck/rubyext.c (syck_parser_assign_io): add prototype.
+
+	* ext/syck/rubyext.c (rb_syck_mktime, yaml_org_handler): use
+	  ISDIGIT() instead of isdigit() to avoid warnings and for
+	  platforms which don't support non-ascii charater.
+
+Wed Dec 10 19:28:56 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): set EOF flag at short read.
+	  [ruby-dev:22223], [ruby-dev:22224]
+
+Wed Dec 10 18:07:25 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/erb.rb: new method ERB#filename(=). [ruby-dev:22208]
+
+Wed Dec 10 17:54:51 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): do not set EOF flag when
+	  requested length is zero.  [ruby-dev:22214]
+
+Wed Dec 10 17:17:18 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (read_all): should return given string even if data read is
+	  empty.  [ruby-dev:22207]
+
+Wed Dec 10 17:16:06 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): adjust behavior at reading
+	  beyond EOF to IO.  [ruby-dev:22205]
+
+	* test/ruby/ut_eof.rb (TestEOF::Seek): test behaviors at reading
+	  beyond EOF.
+
+	* test/ruby/test_file.rb, test/stringio/test_stringio.rb: include
+	  TestEOF::Seek test case.
+
+Wed Dec 10 15:01:19 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/monitor/test_monitor.rb (test_cond): use Queue#deq
+	  instead of sleep.
+
+Wed Dec 10 14:45:39 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/pty/pty.c (HAVE_SYS_IOCTL_H): need to include <sys/ioctl.h>
+	  for TIOCSCTTY on *BSD.  based on gotoyuzo's patch.
+	  (ruby-bugs:PR#1211)
+
+	* ext/pty/pty.c (establishShell): should close descriptors if fork
+	  failed.
+
+Wed Dec 10 12:53:05 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* win32/win32.h: define execv() using do_aspawn().
+
+	* process.c (proc_exec_v): remove #ifdef's which stopped needing.
+
+Tue Dec  9 23:32:23 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tkcanvas.rb, ext/tk/lib/tkdialog.rb,
+	  ext/tk/lib/tkentry.rb, ext/tk/lib/tkscrollbox.rb, ext/tk/lib/tktext.rb,
+	  ext/tk/sample/tkalignbox.rb, ext/tk/sample/tkcombobox.rb,
+	  ext/tk/sample/tkmultilistbox.rb, ext/tk/sample/tkoptdb.rb, ext/tk/sample/tktextframe.rb,
+	  ext/tk/sample/demos-en/dialog1.rb, ext/tk/sample/demos-en/dialog2.rb,
+	  ext/tk/sample/demos-jp/dialog1.rb, ext/tk/sample/demos-jp/dialog2.rb:
+	  overrided instance methods, which are private methods on the super
+	  class, are changed to 'private'
+
+Tue Dec  9 19:53:02 2003  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#route_from0): make case insensitive
+	  for host-part.
+
+	* test/uri/test_generic.rb (test_route): added tests for the above
+	  change.
+
+Tue Dec  9 14:10:48 2003  Tanaka Akira  <akr@m17n.org>
+
+	* io.c (rb_io_check_readable): don't call io_seek if EOF flag is set,
+	  to avoid clearing EOF flag.
+	  (rb_io_check_writable): ditto.
+
+Tue Dec  9 02:53:55 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tkalignbox.rb: new sample script
+
+Tue Dec  9 00:45:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: renamed #assert_raises to #assert_raise
+	  and made the former call the latter. [ruby-core:01890]
+
+	* test/testunit/test_assertions.rb: ditto.
+
+Tue Dec  9 00:07:35 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/rpc/standaloneServer.rb: add 'shutdown' and 'status'
+	  methods as delegates to WEBrick.
+
+	* test/soap/calc/{test_calc.rb,test_calc2.rb},
+	  test/soap/helloworld/test_helloworld.rb,
+	  test/wsdl/datetime/test_datetime.rb, test/wsdl/raa/test_raa.rb:
+	  follow the change.
+
+Mon Dec  8 22:48:03 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/autorunner.rb: remove dependency to a particular
+	  runner.  [ruby-core:01901], [ruby-list:38869]
+
+	* lib/test/unit/ui/testrunnerutilities.rb: moved output level
+	  constants from Console.
+
+	* lib/test/unit/ui/console/testrunner.rb: ditto.
+
+	* lib/test/unit/ui/{fox,gtk,gtk2,tk}/testrunner.rb (initialize):
+	  accept output_level.
+
+Mon Dec  8 15:03:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/syck.c (syck_io_str_read): get rid of buffer overflow.
+
+Mon Dec  8 13:02:11 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/uri/common.rb: new method URI.regexp. [ruby-dev:22121]
+
+	* test/uri/test_common.rb: add test for URI.regexp.
+
+Mon Dec  8 12:44:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c: define swap16 and swap32 only if they are not
+	  defined. OpenBSD defines these macros. [ruby-dev:22181]
+
+Sun Dec  7 20:54:17 2003  Tanaka Akira  <akr@m17n.org>
+
+	* ext/iconv/iconv.c (map_charset): make case sensitive.
+	  ext/iconv/charset_alias.rb (charset_alias): don't ignore
+	  config.charset's information.  sort aliases.
+
+Sat Dec  6 22:58:03 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_start_ssl): new function to wrap
+	  SSL_connect and SSL_accept; if SSL_connect (or SSL_accept) returned
+	  but not finished the handshake process, we should retry it.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_connect): call ossl_start_ssl.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_accept): ditto.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read): allow signal traps.
+
+Sat Dec  6 21:45:10 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* io.c (flush_before_seek): flush before seek on any platform.
+
+	* configure.in: ditto.
+
+Sat Dec  6 17:23:00 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/soap.rb(SOAP::Env.getenv): allow upcase environment variable
+	  as well as downcase one.
+
+	* lib/soap/netHttpClient.rb(SOAP::NetHttpClient#proxy=): check URI.
+
+Fri Dec  5 23:22:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions::assert_raises,
+	  Test::Unit::Assertions::assert_nothing_raised): use the last
+	  argument as message unless class object.
+
+	* test/testunit/test_assertions.rb (test_assert_raises): test for
+	  multiple exception list.  [ruby-core:01891]
+
+	* test/testunit/test_assertions.rb (test_assert_nothing_raised): test
+	  for non-exception classes.
+
+Fri Dec  5 22:23:04 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/netHttpClient.rb: proxy support did not work.  fixed.
+
+	* lib/soap/property.rb: add class methods for loading property from
+	  stream/file/propertyfile.  propertyfile is a file which is located at
+	  somedir in $:.
+
+	* lib/soap/soap.rb, lib/soap/wsdlDriver.rb, lib/soap/rpc/driver.rb,
+	  lib/wsdl/importer.rb: load property from propertyfile 'soap/property'
+	  e.g. /usr/local/lib/ruby/site_ruby/1.8/soap/property.
+
+	* test/soap/test_property.rb, test/soap/test_streamhandler.rb: new file.
+
+Fri Dec  5 17:26:23 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_exec_end_proc): maintain tmp_end_procs.
+	  [ruby-dev:22154]
+
+Fri Dec  5 13:36:59 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_exec_end_proc): should not clear end_procs and
+	  ephemeral_end_procs before execution. [ruby-dev:22144]
+
+	* eval.c (rb_obj_extend): call Module#extended hook after
+	  extended_object.  [ruby-list:38866]
+
+	* object.c (Init_Object): Module#extended defined.
+
+Fri Dec  5 13:17:30 2003  Tanaka Akira  <akr@m17n.org>
+
+	* test/ruby/test_pipe.rb: use IO.pipe instead of IO.popen.
+
+Fri Dec  5 11:54:45 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): follow IO#read.
+
+	* test/ruby/ut_eof.rb, test/ruby/test_file.rb, test/ruby/test_pipe.rb,
+	  test/stringio/test_stringio.rb: add EOF test.
+
+Fri Dec  5 02:49:35 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions::assert_raises):
+	  allow multiple exception list.  [ruby-core:01884]
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions::assert_nothing_raised):
+	  check whether arguments are subclass of Exception.
+
+Thu Dec  4 23:54:00 2003  Rick Ohnemus  <rick.ohnemus@systemware.com>
+
+	* dln.c (aix_loaderror): should not use member named 'errno' which
+	  might be a macro (e.g. on AIX).
+
+Thu Dec  4 23:32:26 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (read_all): do not depend on lseek position.
+	  [ruby-dev:22026]
+
+Thu Dec  4 22:37:26 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): preserve $! value when retry happens in the
+	  rescue clause.  [ruby-talk:86697]
+
+Thu Dec  4 21:50:07 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/drb/drb.rb (DRb::DRbMessage::send_request, send_reply):
+	  should rescue errors and re-raise DRbConnError on write too.
+	  [ruby-dev:22132]
+
+Thu Dec  4 16:41:17 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (exc_list): allow expanding list.  [ruby-dev:22134]
+
+Thu Dec  4 14:09:24 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb (test_cp): test if the error is
+	  kind of SystemCallError.  It is needless details that which errno
+	  is set on each systems.
+
+Thu Dec  4 13:24:13 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb: use Object#__send__ instead of Object#send.
+
+Thu Dec  4 13:17:45 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/streamHandler.rb: support latest released version of
+	  http-access2.
+
+Thu Dec  4 13:04:44 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/soap.rb: add SOAP::Env module for environment repository
+	  such as HTTP_PROXY.
+
+	* lib/soap/property.rb: property implementation.
+
+	* lib/soap/streamHandler.rb, lib/soap/wsdlDriver.rb,
+	  lib/soap/rpc/driver.rb: use soap/property.rb.
+
+	* lib/wsdl/importer.rb, lib/soap/wsdlDriver.rb, lib/soap/rpc/driver.rb:
+	  use SOAP::Env.
+
+	* lib/soap/netHttpClient.rb: add basic_auth, ssl_config, and cookie
+	  management interface, but ignored for now.
+
+	* lib/xsd/charset.rb: add XSD::Charset.encoding= interface to set
+	  wiredump charset explicitly.  it was fixed to 'utf-8' when iconv or
+	  uconv module was found.
+
+Thu Dec  4 10:43:58 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/sym.c (rb_dlsym_guardcall): __declspec(noinline) is VC7
+	  feature.
+
+Thu Dec  4 10:27:12 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: update hyperlink to the Japanese document.
+
+Thu Dec  4 09:12:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (asn1time_to_time): should check that
+	  the underlying value of ASN1_TIME isn't NULL. [ruby-core:01881]
+
+Thu Dec  4 08:29:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/server.rb (GenericServer#start): should rescue
+	  Exception to avoid unexpected aborting. [ruby-core:01853]
+
+	* lib/webrick/server.rb (GenericServer#start_thread): should check
+	  that peeraddr isn't nil before printing.
+
+	* lib/webrick/httpresponse.rb (HTTPResponse#start_thread): should
+	  rescue Exception to avoid unexpected aborting of thread.
+
+Thu Dec  4 03:48:59 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#link, Pathname#symlink): obsoleted.
+	  (Pathname#make_link, Pathname#make_symlink): new method.
+
+Thu Dec  4 01:45:24 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (argf_read): should not terminate on empty string; wait
+	  until real EOF.  [ruby-dev:21969]
+
+	* io.c (argf_read): should adjust length to read, when length is
+	  specified and read spans command line argument files.
+
+Wed Dec  3 19:38:36 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: correct fcntl parameter. [ruby-dev:22120]
+
+Wed Dec  3 13:49:07 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: 'format'==>'Kernel.format' (avoid override trouble)
+
+	* ext/tk/lib/tkafter.rb: ditto.
+
+	* ext/tk/lib/tkcanvas.rb: ditto.
+
+	* ext/tk/lib/tkdialog.rb: ditto.
+
+	* ext/tk/lib/tktext.rb: ditto.
+
+Wed Dec  3 13:28:13 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (lex.c): try gperf first, and copy from the source
+	  directory if failed.  [ruby-dev:22123]
+
+	* ext/extmk.rb (MTIMES): let makefiles depend to mkmf.rb.
+
+	* lib/mkmf.rb (configuration): DLDFLAGS was duplicated.
+
+Tue Dec  2 23:18:12 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: wrote the warning about HTTP_PROXY environment
+	  variable.
+
+Tue Dec  2 21:31:42 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bin/testrb: new test runner.  [ruby-core:01845]
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner.run,
+	  Test::Unit::AutoRunner#process_args): take test list to run and
+	  options.
+
+	* lib/test/unit/autorunner.rb (Test::Unit::AutoRunner::RUNNERS,
+	  Test::Unit::AutoRunner#run): should not exit inside a library,
+	  just return the result instead.
+
+	* lib/test/unit.rb: ditto.
+
+	* test/runner.rb: exit with the test result.
+
+Tue Dec  2 20:18:48 2003  Eric Sunshine  <sunshine@sunshineco.com>
+
+	* configure.in (AC_PROG_YACC): AC_DEFINE(OLD_YACC) if Yacc is found
+	  instead of Bison or byacc.
+
+	* parse.y: If OLD_YACC is defined, ensure that YYMAXDEPTH is at least
+	  10000 (Bison's default) since some old versions of Yacc define it as
+	  low as 150 by default, which is too low for Ruby to parse some files,
+	  such as date/format.rb.  Among other issues, the parse problem causes
+	  "make test" to fail.
+
+Tue Dec  2 20:03:20 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb: check if Pathnames are usable
+	  for arguments.
+
+Tue Dec  2 04:22:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: fixed #assert_no_match message.
+
+	* test/testunit/test_assertions.rb: ditto.
+
+Tue Dec  2 00:43:00 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/syck.c: string buffering bug.  decrementing by full
+	  max_size now. [ruby-core:01834]
+
+Mon Dec  1 21:33:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* numeric.c (num_sadded): prohibit singleton method definition for
+	  Numerics.  fill yet another gap between Fixnum and Bignum.
+
+Mon Dec  1 17:33:47 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (htov16): converts endian using swap16. htov32(), hton16,
+	  hton32 as well. [ruby-talk:85377]
+
+	* pack.c (swap16): swap 2 bytes no matter how big short is on the
+	  platform.  swap32() is also prepared.
+
+	* numeric.c (rb_num2int): returns long to preserve information.
+	  rb_fix2int(), rb_num2uint(), rb_fix2uint() as well.
+	  [ruby-talk:85377]
+
+	* numeric.c (rb_num2uint): should not check for value range if the
+	  source value is negative.
+
+Mon Dec  1 17:14:34 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/optparse/opttest.rb: added.
+
+Mon Dec  1 16:10:52 2003  Dave Thomas  <dave@pragprog.com>
+
+	* lib/rdoc/rdoc.rb: (etc) initial merge into main tree.
+
+Mon Dec  1 14:17:49 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (fu_each_src_dest0): call #to_str to allow
+	  Pathname for arguments. [ruby-core:01795]
+
+	* test/fileutils/test_fileutils.rb: does much strict test on
+	  "same" files detecting.
+
+Mon Dec  1 09:28:14 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
+	  (XCFLAGS): re-export $(XCFLAGS).
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
+	  (ARCH_FLAG): export $(ARCH_FLAG) (perhaps empty value).
+
+Mon Dec  1 01:03:27 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/mkmf.rb (TRY_LINK, link_command): added support for DLDFLAGS
+	  and ARCH_FLAG.  [ruby-dev:22085]
+
+Sun Nov 30 20:18:07 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: keep ARCH_FLAG separate. export ARCH_FLAG.
+	  [ruby-core:01819]
+
+	* Makefile.in: add ARCH_FLAG to CFLAGS.
+
+	* Makefile.in: add @CPPFLAGS@ to CPPFLAGS.
+
+	* lib/mkmf.rb (link_command, cc_command): use ARCH_FLAG.
+
+	* lib/mkmf.rb (configuration): add ARCH_FLAG to DLDFLAGS.
+
+	* Makefile.in: add ARCH_FLAG to DLDFLAGS.
+
+	* configure.in: should put getcwd in AC_CHECK_FUNCS, not
+	  AC_REPLACE_FUNCS.  [ruby-core:01826]
+
+Sun Nov 30 18:22:48 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: do not override CCDLDFLAGS, LDFLAGS, XLDFLAGS,
+	  DLDFLAGS and LDSHARED.
+
+	* configure.in: XCFLAGS for compiling ruby itself.  ARCH_FLAG is
+	  reflected in CFLAGS.
+
+	* lib/mkmf.rb: ditto.  do not import XCFLAGS from config.status.
+
+Sun Nov 30 17:37:36 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: bug fix [ruby-talk:86746]
+
+Sun Nov 30 13:02:00 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/encodingstyle/soapHandler.rb: refactoring - Simplifying
+	  Conditional Expressions.
+
+	* lib/wsdl/soap/definitions.rb: refactoring - Move Method.
+
+	* test/xsd/{test_noencoding.rb,noencoding.xml}: new files.  test for
+	  encoding unspecified XML file parsing.
+
+	* test/wsdl/{test_fault.rb,map,datetime}: new files.  test of
+	  SOAPFault, dateTime and Apache's Map.
+
+Sun Nov 30 09:35:14 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_update): get rid of SEGV at just allocated String.
+	  [ruby-core:01812]
+
+Fri Nov 28 23:19:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_mark): explicitly check mark recursion levels, instead
+	  of unreliable stack length.
+
+Fri Nov 28 22:49:56 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/rinda.rb: fix TupleSpaceProxy#read, read_all.
+
+Fri Nov 28 21:44:40 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb (test_ln_s): should be a file, not
+	  a directory for FreeBSD.
+
+Fri Nov 28 19:37:56 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_has_value, env_index): must match exactly.
+
+	* test/ruby/test_env.rb (test_has_value, test_index): condition for
+	  aboves.
+
+Fri Nov 28 17:59:20 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_env.rb: add tests for ENV.
+
+Fri Nov 28 17:47:46 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (DRbMessage#load): rescue Errno::* and raise
+	  DRbConnError.
+
+Fri Nov 28 15:41:15 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#realpath): obsolete the force_absolute
+	  argument.
+
+Fri Nov 28 14:41:52 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/streamHandler.rb: drop unused http parameters.
+
+	* lib/soap/encodingstyle/soapHandler.rb, lib/soap/mapping/factory.rb,
+	  lib/soap/mapping/mapping.rb, lib/soap/mapping/registry.rb,
+	  lib/wsdl/soap/complexType.rb: ApacheSOAP's map support was broken
+	  under WSDL dynanic client environment.  fixed.
+
+	* test/wsdl/raa/*: add tests.
+
+	* lib/xsd/datatypes.rb: dateTime precision bug fix (at least, I hope.)
+	  bug of soap4r.  XSDDateTimeImple.to_time passed a Float to
+	  Time.local/Time.gm as an usec, and NUM2LONG(rb_num2long for Float)
+	  causes rounding error.
+
+	* test/soap/test_basetype.rb, test/xsd/test_xsd.rb: add tests.
+
+Fri Nov 28 04:15:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (method_arity): used wrong Proc object.  [ruby-talk:86504]
+
+Fri Nov 28 00:47:29 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_f_exit), process.c (rb_f_exit_bang): treat true as
+	  success, false as failure.  [ruby-dev:22067]
+
+	* eval.c (rb_f_abort, rb_thread_switch), process.c (rb_f_system): use
+	  ANSI macro instead of hard coded value.
+
+	* eval.c (rb_f_exit), process.c (rb_f_exit_bang): use VALUEs not but
+	  TYPEs.
+
+Thu Nov 27 22:05:48 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* eval.c, gc.c: FreeBSD/ia64 currently does not have a way for a
+	  process to get the base address for the RSE backing store, so
+	  hardcode it for the moment.
+	  [submitted by: Marcel Moolenaar <marcel@FreeBSD.org>]
+
+Thu Nov 27 17:36:42 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkafter.rb: bug fix on TkTimer#cancel_on_exception=(mode).
+	  TkTimer#wait receives the exception of the callback.
+	  The exception is kept on @return_value.
+
+Thu Nov 27 16:58:48 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_stat): remove _fullpath() for NUL: device.
+
+Wed Nov 26 15:38:47 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb (test_ln_s): should take the
+	  existing symbolic link for OpenBSD.
+
+Wed Nov 26 04:48:42 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/token.c: removed YYTOKTMP references which
+	  were causing buffer overflows on large block scalars,
+	  comments, quoted scalars and plain scalars.
+
+	* ext/syck/rubyext.c: dynamic changing of buffer size.
+
+	* ext/syck/syck.h: default buffer size of 4k.
+
+Wed Nov 26 00:55:30 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpresponse.rb: add HTTPResponse#keep_alive=.
+
+	* lib/webrick/httpserver.rb (HTTPServer#run): should pass the
+	  request's keep_alive flag to the response.
+
+Tue Nov 25 21:41:35 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* defines.h (ENV_IGNORECASE): should define when DOSISH without
+	  human68k. [ruby-dev:22047]
+
+	* hash.c (env_has_value, env_index): don't ignore case of value.
+	  [ruby-dev:22048]
+
+Tue Nov 25 21:39:37 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (path_check_1): honor sticky bits always.
+	  [ruby-talk:86273]
+
+Tue Nov 25 20:02:14 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb: do test in more deep
+	  directory.
+
+	* test/fileutils/test_nowrite.rb: ditto.
+
+Tue Nov 25 19:04:23 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::Generic#find_proxy): ENV case sensitivity test
+	  refined.
+
+Tue Nov 25 18:13:30 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb: chdir Dir.tmpdir before each
+	  test. [ruby-dev:22045]
+
+	* test/fileutils/test_nowrite.rb: ditto.
+
+Tue Nov 25 17:52:11 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::Generic#find_proxy): use http_proxy under CGI
+	  if the environment variable is case sensitive.
+
+Tue Nov 25 16:41:33 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/wsdl/multiplefault.wsdl, test/wsdl/test_multiplefault.rb:
+	  removed.  this test requires extra libraries in soap4r/1.5.*.
+
+Tue Nov 25 16:24:42 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/**/*.rb, lib/wsdl/**/*.rb, lib/xsd/**/*.rb: changed license;
+	  GPL2 -> Ruby's.
+
+	* lib/soap/rpc/driver.rb, lib/soap/wsdlDriver.rb,
+	  lib/soap/streamHandler.rb: add interface to streamhandler.
+
+	* lib/soap/marshal.rb: raise error if parse fails.
+
+	* lib/soap/netHttpClient.rb: add https support.  Patched by
+	  Oliver M. Bolzer.
+
+	* lib/soap/netHttpClient.rb: dump HTTP response message body by itself.
+
+	* lib/soap/rpc/driver.rb, lib/soap/rpc/proxy.rb,
+	  lib/soap/wsdlDriver.rb: add driver#mandatorycharset interface to foce
+	  using charset for parsing response from buggy server.
+
+	* lib/soap/encodingstyle/soapHandler.rb: support Apache Axis's half
+	  typed multi-ref array.
+
+	* lib/soap/mapping/factory.rb, lib/soap/mapping/registry.rb: map
+	  SOAPStruct which has multi-accessors which name are the same, to an
+	  array.
+
+	* lib/soap/rpc/element.rb: fixed illegal parameter order.
+
+	* lib/soap/rpc/element.rb: element name of response message could have
+	  the name other than 'return'.
+
+	* lib/wsdl/operation.rb, lib/wsdl/operationBinding.rb,
+	  lib/wsdl/soap/classDefCreator.rb, lib/wsdl/soap/methodDefCreator.rb,
+	  lib/wsdl/soap/methodDefCreatorSupport.rb: WSDL/1.1 allows plural
+	  fault definition in a operation. [ruby-talk:84948]
+
+	* test/wsdl/multiplefault.wsdl, test/wsdl/test_multiplefault.rb: add
+	  test for above fix.
+
+	* lib/wsdl/soap/complexType.rb: support WSDL array definition with
+	  maxOccures="unbound".
+
+	* lib/xsd/charset.rb: use cp932 under emx.  Patched by
+	  Siena. / SHINAGAWA, Norihide in [ruby-dev:21972]
+
+	* lib/xsd/xmlparser/parser.rb: set @charset nil by default.  Nil means
+	  'follow encoding declaration in XML'.
+
+	* sample/soap/digraph.rb, sample/wsdl/amazon/wsdlDriver.rb,
+	  sample/wsdl/googleSearch/sampleClient.rb,
+	  sample/wsdl/googleSearch/wsdlDriver.rb,
+	  test/wsdl/test_emptycomplextype.rb,
+	  test/wsdl/marshal/test_wsdlmarshal.rb,
+	  test/xsd/test_xmlschemaparser.rb: use File.open(...) { |f| f.read }
+	  instead of File.open(...).read. [ruby-dev:21964]
+
+	* test/wsdl/emptycomplextype.wsdl, test/wsdl/test_emptycomplextype.rb:
+	  simplify the test case.
+
+	* test/wsdl/axisArray/*: add tests for axis's array encoding.
+
+Tue Nov 25 16:15:29 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ruby.h: don't treat Cygwin as Windows.
+
+Tue Nov 25 15:18:28 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* configure.in: change default value of --enable-pthread (default: no)
+
+Tue Nov 25 07:31:16 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): allow newlines just before right argument
+	  parenthesis.  (ruby-bugs:PR#1221)
+
+Mon Nov 24 23:32:06 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI.open_loop, URI::HTTP#proxy_open): use
+	  catch/throw for redirection instead of exception.
+	  (OpenURI.open_loop, OpenURI.redirectable?): restrict redirection.
+
+Mon Nov 24 19:59:48 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (URI::Generic#find_proxy): use CGI_HTTP_PROXY
+	  instead of HTTP_PROXY in the CGI environment.
+
+Mon Nov 24 19:32:55 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/etc/extconf.rb: check for pw_passwd in struct passwd and
+	  gr_passwd in struct group for DJGPP.
+
+	* ext/etc/etc.c: ditto.
+
+	* ext/Setup.dj: support for curses, etc, zlib.
+
+Mon Nov 24 17:00:00 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb: validate option names.
+	  :content_length_proc and :progress_proc option implemented.
+
+Mon Nov 24 14:53:10 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
+	  (XCFLAGS): output empty value instead of `-DRUBY_EXPORT'.
+
+Sat Nov 22 23:09:45 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: set enable_pthread to no on MinGW.
+
+Sat Nov 22 22:56:20 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* configure.in: add --enable-pthread option (default: yes)
+
+Sat Nov 22 22:48:46 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: add Tk.grab_release and fix bug of TkComposite
+
+	* ext/tk/lib/tkafter.rb: bug fix of TkAfter#start
+
+	* ext/tk/sample/tkcombobox.rb: new sample script
+
+	* ext/tcltklib/tcltklib.c: add native thread check
+
+Sat Nov 22 18:49:47 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/curses/curses.c (window_nodelay): nodelay() of NetBSD's
+	  libcruses returns no value, just like keypad().
+
+Sat Nov 22 17:36:36 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
+	  (HAVE_GETCWD): output to config.h.
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub
+	  (XCFLAGS): output to config.status.
+
+Sat Nov 22 13:10:10 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (have_st_ino?): djgpp has valid st_ino.
+
+Sat Nov 22 11:28:48 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (Init_stack): stack region is far smaller than usual if
+	  pthread is used.
+
+Sat Nov 22 07:30:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/util/backtracefilter.rb: fixed a bug that occurred
+	  when an exception had no backtrace.
+
+	* test/testunit/util/test_backtracefilter.rb: ditto.
+
+Fri Nov 21 16:44:18 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tkentry.rb: fix the encoding trouble of percent
+	  substitutions on validatecommand option of TkEntry widget
+
+	* ext/tk/lib/tk.rb: fix bug on {pack|grid}_propagate() method
+
+Fri Nov 21 16:12:11 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ruby.1: Fix markups and grammar.
+
+Fri Nov 21 14:49:42 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* ruby.1: wrote about ruby related environment variables.
+
+Fri Nov 21 12:28:03 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_extended): singleton methods should not be checked
+	  when dumping via marshal_dump() or _dump(). [ruby-talk:85909]
+
+Fri Nov 21 01:40:00 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* configure.in: check <pthread.h>
+
+	* ruby.h: include pthread.h if existence.
+	  define is_ruby_native() macro when not HAVE_NATIVETHREAD
+
+	* eval.c: undef is_ruby_native() function when not HAVE_NATIVETHREAD
+
+Fri Nov 21 00:43:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: use #__send__ instead of #send.
+
+	* lib/test/unit/testcase.rb: ditto.
+
+Thu Nov 20 19:19:22 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in: don't find the Cygwin's pthread library on MinGW.
+
+Thu Nov 20 19:15:50 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (have_st_ino?): emx (OS/2 with EMX) does not
+	  have st_ino (always 0). [ruby-dev:21972]
+
+	* lib/fileutils.rb (rename_cannot_overwrite_file?): emx does not
+	  allow overwriting files by rename(2).
+
+	* test/fileutils/test_fileutils.rb: windows? ->
+	  have_drive_letter?, have_file_perm?
+
+Thu Nov 20 17:50:58 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/tkballoonhelp.rb: new sample script
+
+	* ext/tk/sample/tkmultilistbox.rb: ditto
+
+	* ext/tk/sample/tktextframe.rb: ditto
+
+Thu Nov 20 13:37:34 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ruby.h: define is_ruby_native_thread() for no native thread
+	  environment
+
+	* eval.c: ditto
+
+Thu Nov 20 12:42:47 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* configure.in: always check existence of the pthread library
+
+	* ruby.h: define macros for ruby's native thread check
+
+	* eval.c: add ruby's native thread check
+
+	* gc.c: ditto
+
+Wed Nov 19 14:45:18 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (to_ary): print more friendly warning message.
+
+Wed Nov 19 14:32:08 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (fu_same?): add djgpp and wince.
+
+	* lib/fileutils.rb (cannot_overwrite_file?): add wince.
+
+Wed Nov 19 11:04:47 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/fileutils.rb (cannot_overwrite_file?, have_st_ino?): bccwin32
+	  is same as mswin32.
+
+Wed Nov 19 07:54:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit.rb: do not run tests if $! is set.
+
+	* lib/test/unit/assertionfailederror.rb: extend StandardError instead
+	  Exception (irb catches the former but not the latter).
+
+Tue Nov 18 23:31:36 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* missing/memmove.c (memmove): take void *, not char *.
+
+	* missing.h (memmove): ditto.
+
+	* missing.h (strchr, strrchr): return char *, not int.
+
+Tue Nov 18 22:20:10 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (fu_same?): temporal fix for windows.
+
+Tue Nov 18 19:05:04 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (fu_same?): check by inode instead of path
+	  name, to detect two hard links pointing to the same content.
+
+	* test/fileutils.rb: did not create correctly looped symlinks.
+
+Tue Nov 18 18:23:05 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): behave as IO at empty string.
+	  [ruby-dev:21939], [ruby-dev:21941]
+
+	* ext/stringio/stringio.c (strio_getc, strio_getline): set EOF flag.
+
+	* ext/stringio/stringio.c (strio_rewind, strio_seek, strio_ungetc):
+	  clear EOF flag.
+
+	* test/stringio/test_stringio.rb: imported from [ruby-dev:21941].
+
+Tue Nov 18 14:06:35 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/fileutils.rb (fu_each_src_dest): raise if src==dest.
+	  [ruby-talk:85344] [ruby-core:01699]
+
+	* lib/fileutils.rb: use Object#is_a? instead of Class#=== to allow
+	  e.g. remote objects for receivers.
+
+	* lib/fileutils.rb: FileTest -> File.
+
+	* lib/fileutils.rb: put parentheses for arguments of File.xxxx?
+
+	* test/fileutils/test_fileutils.rb (test_cp): test "cp a a".
+
+	* test/fileutils/test_fileutils.rb (test_mv): test "mv a a".
+
+	* test/fileutils/test_fileutils.rb (test_ln): test "ln a a".
+
+	* test/fileutils/test_fileutils.rb (test_ln_s): test "ln_s a a".
+
+	* test/fileutils/test_fileutils.rb (test_install): test "install a a".
+
+	* test/fileutils/fileasserts.rb: new method assert_symlink.
+
+	* test/fileutils/fileasserts.rb: assert_is_directory -> assert_directory.
+
+Mon Nov 17 19:38:49 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (getcwdofdrv): avoid using getcwd() directly, use
+	  my_getcwd() instead.
+
+	* merged NeXT, OpenStep, Rhapsody ports patch from Eric Sunshine
+	  <sunshine@sunshineco.com>.  [ruby-core:01596]
+
+Mon Nov 17 10:50:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Completion::complete): allow least
+	  common completion for three or more candidates.
+
+Mon Nov 17 09:41:38 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/ui/tk/testrunner.rb,
+	  lib/test/unit/ui/gtk/testrunner.rb:
+	  run GUI main loop in sub thread.
+
+	* lib/test/unit/ui/gtk2/testrunner.rb: imported from rough.
+
+	* lib/test/unit/autorunner.rb (keyword_display): sort keywords.
+
+Sun Nov 16 18:10:57 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): iterator should return value from next inside
+	  begin/rescue/end.  (ruby-bugs:PR#1218)
+
+Sun Nov 16 13:26:07 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_object): LINK check earlier than anything else,
+	  i.e. do not dump TYPE_IVAR for already dumped objects.
+	  (ruby-bugs:PR#1220)
+
+	* eval.c (rb_eval): call "inherited" only when a new class is
+	  generated; not on reopening.
+
+	* eval.c (eval): prepend error position in evaluating string to
+	  "mesg" attribute string only when it's available and is a
+	  string.
+
+Sun Nov 16 12:16:10 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/protocol.rb: logging response body. [experimental]
+	  [ruby-list:38800]
+
+Sun Nov 16 10:49:38 2003  Gavin Sinclair  <gsinclair@soyabean.com.au>
+
+	* lib/thread.rb (Thread.exclusive): wrap method definition in
+	  class Thread to enable rdoc to process.
+
+Sun Nov 16 09:45:23 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (set_debug_output): warn if method is called
+	  after #start.  [ruby-dev:38798]
+
+Sun Nov 16 04:41:33 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (eval): do not re-raise exception to avoid unnecessary
+	  exception copying, instead modify exception and internal
+	  information to adjust eval().
+
+	* eval.c (backtrace): can return the current frame information
+	  only if lev < -1.
+
+Sat Nov 15 22:16:42 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* /ext/openssl/ossl_x509ext.c (ossl_x509extfactory_create_ext):
+	  refine error message.
+
+Sat Nov 15 10:05:40 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI.open_loop, OpenURI::HTTP#proxy_open):
+	  refactored to support options.
+	  (Buffer): maintain size by this class.
+
+Sat Nov 15 07:40:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_method_node): new API to retrieve method body.
+
+Fri Nov 14 13:21:30 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: fix (en-bugged at 2003/11/07)
+
+	* ext/tk/lib/tkdialog.rb: TkDialog.new accepts a parent widget
+	  argument [ruby-talk:85066]
+
+Thu Nov 13 20:53:35 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (Kernel[#.]open): hard coded URI schemes removed.
+	  [ruby-ext:02251]
+
+Thu Nov 13 19:17:00 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* lib/test/unit/ui/tk/testrunner.rb: use grid and panedwindow
+	  (if available)
+
+Thu Nov 13 17:56:41 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/open-uri.rb (OpenURI.open_uri): use File::RDONLY.
+	  reported by Take_tk <ggb03124@nifty.ne.jp>.
+	  [ruby-ext:02245]
+
+Thu Nov 13 16:45:53 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509req.c (ossl_x509req_to_der): add function for
+	  X509::Request#to_der.
+
+Thu Nov 13 11:31:14 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Completion#complete): prior shorter
+	  name to containing longer name.
+
+Thu Nov 13 06:08:54 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: stop freezing some classes
+
+	* ext/tk/lib/multi-tk.rb: ditto.
+
+Wed Nov 12 17:32:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (assert_throws, assert_nothing_thrown):
+	  uncaught throw in sub thread raises ThreadError.
+
+	* lib/test/unit/ui/tk/testrunner.rb (setup_ui): "expand" is not
+	  necessary.
+
+Wed Nov 12 14:09:43 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/monitor/test_monitor.rb: fix the timing problem by Queue.
+
+Wed Nov 12 12:59:44 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/monitor/test_monitor.rb: added.
+
+Wed Nov 12 10:14:28 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb: refactored. Thanks, Gennady Bystritsky.
+
+Wed Nov 12 06:11:39 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.c (ossl_x509_sk2ary, ossl_x509crl_sk2ary):
+	  add functions to convert STACK into Array.
+
+	* ext/openssl/ossl.h: add prototypes.
+
+	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_set_certificates,
+	  ossl_pkcs7_get_certificates, ossl_pkcs7_get_crls,
+	  ossl_pkcs7_set_crls): add functions for PKCS7#certificates=
+	  PKCS7#certificates, PKCS7#crls= and PKCS7#crls.
+
+Wed Nov 12 00:47:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/ui/testrunnermediator.rb: should require 'test/unit'.
+
+Tue Nov 11 23:54:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/ui/gtk/testrunner.rb: added a rescue clause to handle
+	  the case when the requested font is not available.
+
+Tue Nov 11 22:44:08 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (appendline): file may not end with newline.  a bug if
+	  READ_DATA_PENDING_PTR is defined. [ruby-talk:84925]
+
+Tue Nov 11 10:42:41 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: raise an exception when creating TkWindow
+	  object, because TkWindow class is an abstract class.
+
+Tue Nov 11 03:30:43 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/ext/openssl/ossl_conf.c (ossl_config_get_value): return nil
+	  if the specified value doesn't exist.
+
+	* lib/ext/openssl/ossl_conf.c (ossl_config_get_section): return
+	  a empty hash if the specified section doesn't exist.
+
+Mon Nov 10 11:40:29 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb (wait): return true on signal/broadcastfalse and
+	  false on timeout. Thanks Gennady Bystritsky.
+
+Mon Nov 10 00:07:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): primary_value may be 0 when syntax error.
+	  [ruby-talk:84893]
+
+Sun Nov  9 02:05:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: un-deprecated #assert_not_nil to
+	  maintain symmetry with #assert_nil. Also added better output for
+	  #assert_kind_of.
+
+	* test/testunit/tc_assertions.rb: ditto.
+
+Sat Nov  8 18:50:20 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/wsdl/raa/*: add new testcase for WSDL loading, parsing and
+	  reading.
+
+	* test/soap/marshal/*: backport from soap4r/1.5.1.  all differences are
+	  for ruby/1.6.
+
+	* lib/soap/*: backport from soap4r/1.5.1.  all differences are for
+	  ruby/1.6.
+
+	* lib/wsdl/data.rb, lib/wsdl/xmlSchema/data.rb: move definition of
+	  ArrayTypeAttrName from ::WSDL::XMLSchema::* to ::WSDL::*.
+	  [ruby-talk:84813]
+
+	* lib/wsdl/soap/definitions.rb: element name typo in custom exception
+	  struct definition which is needed for wsdlDriver; camelCase ->
+	  underscore_name.
+
+Sat Nov  8 13:49:50 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* configure.in: improvement of pthread check
+
+Sat Nov  8 13:28:46 2003  Takaaki Tateishi  <ttate@ttsky.net>
+
+	* ext/dl/sym.c: Add DL.win32_last_error and DL.last_error.
+	  Thanks, Kaoru Shirai.
+
+Sat Nov  8 06:19:38 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: To fix 'pthread-enabled Tcl/Tk' problem,
+	  TclTkIp#_eval calls Tcl_Eval() on the mainloop thread only
+	  (queueing a handler to the EventQueue).
+
+	* ext/tcltklib/README.1st: edit the description of '--with-pthread-ext'
+
+Fri Nov  7 23:23:04 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (Pathname#+): if self or the argument is `.', return
+	  another.
+	  (Pathname#parent): if self is `.', return `..'.
+	  (Pathname#children): if self is `.', don't prepend self for a
+	  pathname in a result.
+	  (Pathname#join): re-implemented using Pathname#+.
+	  (Pathname#find): if self is `.', remove `./' prefix of yielding
+	  pathname.
+
+Fri Nov  7 10:23:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (make_hostent): get rid of SEGV on aliases
+	  lookup failure.  (ruby-bugs:PR#1215)
+
+Fri Nov  7 04:08:05 2003  UENO Katsuhiro  <katsu@blue.sky.or.jp>
+
+	* ext/zlib/zlib.c (Init_zlib): define Zlib::GzipReader#each_line as
+	  an alias of Zlib::GzipReader#each.
+
+Fri Nov  7 01:03:16 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_load): save and restore rb_prohibit_interrupt.
+	  [ruby-dev:21857]
+
+Thu Nov  6 18:05:07 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_inspect): show the path also at a closed file.
+	  [ruby-dev:21851]
+
+Thu Nov  6 11:42:07 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_set_string, strio_reopen): check
+	  tainted.
+
+	* ext/stringio/stringio.c (strio_copy, strio_ungetc, strio_write,
+	  strio_putc): add infection.
+
+	* ext/stringio/stringio.c (strio_path): just nil.  [ruby-dev:21846]
+
+	* ruby.c (proc_options): reserve searched script path in the
+	  source file name table.  [ruby-list:38765]
+
+	* lib/optparse.rb (OptionParser::Completion#complete): default not to
+	  ignore case on completion.  [ruby-talk:84726]
+
+	* win32/win32.c (make_cmdvector): process backslashes even if a quote
+	  is not enclosed.
+
+Wed Nov  5 23:49:45 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* sample/openssl/gen_csr.rb: there (at least) is a CA which does not
+	  accept DN in UTF8STRING format.  it's a sample.
+
+Wed Nov  5 22:55:16 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* configure.in, eval.c, signal.c: : add '--with-pthread-ext'
+	  option to fix the pthread trouble on 'tcltklib'
+
+	* ext/tcltklib/README.1st: add the description of '--with-pthread-ext'
+
+	* ext/tk/lib/tktext.rb: add TkText#text_copy, text_cut, text_paste
+	  to support Tcl/Tk8.4's tk_textCopy, tk_textCut, tk_textPaste
+
+	* ext/tk/lib/tk.rb: add TkMenu#set_focus support Tcl/Tk's
+	  tk_menuSetFocus
+
+Wed Nov  5 17:33:45 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_load): allow interrupt during loaded program
+	  evaluation.  [ruby-dev:21834]
+
+	* hash.c (rb_hash_fetch): always warn if default argument and a
+	  block are supplied at the same time. [ruby-dev:21842]
+
+	* hash.c (env_fetch): ditto.
+
+	* array.c (rb_ary_fetch): ditto.
+
+Wed Nov  5 19:08:47 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Switch::PlacedArgument::parse):
+	  do not remove next argument if empty value is placed.
+
+	* test/optparse: added.
+
+Wed Nov  5 17:05:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/ui/gtk/testrunner.rb: typo.
+
+Wed Nov  5 11:13:32 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c: add #include "version.h". this file still depends on it.
+
+	* Makefile.in, bcc32/Makefile.sub, win32/Makefile.sub,
+	  wince/Makefile.sub: add version.h dependency to string.c.
+
+Wed Nov  5 09:14:23 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb: revert to the previous revision.
+
+Wed Nov  5 08:39:51 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/https.rb (HTTPRequest#parse): set @client_cert_chain.
+
+	* lib/webrick/https.rb (HTTPRequest#meta_vars): create
+	  SSL_CLIENT_CERT_CHAIN_n from @client_cert_chain.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_get_peer_cert_chain): return nil
+	  if no cert-chain was given.
+
+Tue Nov  4 23:44:48 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, wince/Makefile.sub:
+	  remove needless version.h dependency.
+
+Tue Nov  4 23:38:43 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* class.c, hash.c, string.c: remove #include "version.h".
+
+	* Makefile.in: remove needless version.h dependency.
+
+Tue Nov  4 06:54:52 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (read_all): fptr->f may be NULL, if IO is closed in the
+	  signal handler.
+
+	* io.c (io_read): ditto.
+
+	* string.c (get_pat): remove 1.8.0 warning code.
+
+	* string.c (rb_str_match): extend warning until 1.8.2.
+
+	* string.c (rb_str_match2): ditto.
+
+	* class.c (class_instance_method_list): remove 1.8.0 warnings.
+	  method_list now recurs.  [ruby-dev:21816]
+
+	* class.c (rb_obj_singleton_methods): ditto.
+
+	* array.c (rb_ary_select): remove select with block.
+	  [ruby-dev:21824]
+
+	* hash.c (rb_hash_select): ditto.
+
+	* hash.c (env_select): ditto.
+
+	* re.c (match_select): ditto.
+
+	* struct.c (rb_struct_select): ditto.
+
+Mon Nov  3 22:53:21 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/racc/parser.rb: synchronize with Racc 1.4.4.
+
+	* ext/racc/cparse/cparse.c: ditto.
+
+	* ext/racc/cparse/cparse.c (parse_main): should abort when
+	  the length of LR state stack <=1, not ==0.
+
+Mon Nov  3 08:50:47 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* process.c (check_uid_switch): remove duplicated error messages.
+
+	* process.c (check_gid_switch): ditto.
+
+Sun Nov  2 02:28:33 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/ssl.rb: new option :SSLExtraChainCert.
+
+Sun Nov  2 01:02:04 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c (rb_str_hash): Update the HASH_PERL alternative hash
+	  algorithm in sync with Perl 5.8.
+
+	* st.c (strhash): Ditto.
+
+Sat Nov  1 18:21:09 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_peer_cert_chain): add new method
+	  SSLSocket#peer_cert_chain.
+
+	* ext/openssl/ossl_x509req.c (GetX509ReqPtr): new function
+	  which returns underlying X509_REQ.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_set_issuer_cert,
+	  ossl_x509extfactory_set_subject_cert, ossl_x509extfactory_set_crl,
+	  ossl_x509extfactory_set_subject_req, ossl_x509extfactory_set_config):
+	  use underlying C struct without duplication not to leak momory.
+
+Sat Nov  1 01:49:03 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/mapping/factory.rb: mark marshalled basetype objects when
+	  @allow_original_mapping is true.  multi-referencing basetype node is
+	  prohibited in SOAP/1.1 encoding but soap4r's original ruby object
+	  mapping requires basetype to be marked to detect self referencing
+	  loop.  e.g. o = 1; o.instance_eval { @iv = o }  soap4r's original
+	  mapping is only used through soap/marshal API.
+
+	* test/soap/marshal/test_marshal.rb: add tests for self referencing
+	  immutable objects.
+
+	* test/soap/calc/test_calc_cgi.rb: fix test name.
+
+Fri Oct 31 22:26:29 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
+
+	* wince/string_wce.c (strrchr): should decrement pointer.
+
+	* wince/Makefile.sub: correct a range of isdigit().
+
+Fri Oct 31 12:55:24 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in, lib/mkmf.rb: add RPATHFLAG for NetBSD.
+	  [ruby-dev:21791]
+
+	* bcc32/Makefile.sub, win32/Makefile.sub, win32/Makefile.sub: ditto.
+
+Fri Oct 31 01:38:14 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* wince/Makefile.sub, win32/Makefile.sub (.y.c): allow white spaces
+	  at the beginning of line to remove by sed. (ruby-bugs-ja:PR#580)
+
+Fri Oct 31 01:02:24 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* compar.c (cmp_equal): protect exceptions from <=> comparison
+	  again.  returns nil if any exception or error happened during
+	  comparison.
+
+	* eval.c (search_required): should update *featurep when DLEXT2 is
+	  defined. (ruby-bugs-ja:PR#581)
+
+Thu Oct 30 23:41:04 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb: add DRbArray
+
+	* lib/drb/invokemethod.rb: fix Hash#each problem. [ruby-dev:21773]
+
+	* lib/drb/unix.rb: add LoadError. [ruby-dev:21743]
+
+Thu Oct 30 23:19:11 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/generator.rb: better XML pretty printing.
+
+	* lib/soap/encodingstyle/soapHandler.rb: remove unnecessary namespace
+	  assignment in the element which has "encodingStyle" attribute, and
+	  add necessary namespace assignment for "arrayType" attribute.
+
+	* test/soap/calc/test_calc_cgi.rb: take over $DEBUG to ruby process
+	  through CGI.
+
+Thu Oct 30 22:59:39 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/yaml2byte.c: HASH const too long.  Thanks, matz.
+
+Thu Oct 30 19:13:53 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/syck/MANIFEST: Add yamlbyte.h.
+
+Thu Oct 30 14:25:31 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (READ_DATA_BUFFERED): new macro to detect whether stdio
+	  buffer filled.
+
+	* io.c (rb_io_fptr_cleanup): move path deallocation to
+	  rb_io_fptr_finalize (finalizer called by GC).
+
+Thu Oct 30 13:23:39 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (logop): left may be NULL. [ruby-talk:84539]
+
+	* eval.c (rb_eval): NODE_CASE nd_head may be NULL.
+
+Thu Oct 30 10:14:51 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/test/unit/autorunner.rb: make fox runner work.
+
+Thu Oct 30 09:32:26 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_f_system): fixed lack of security check before
+	  calling do_spawn() on win32. [ruby-talk:84555]
+
+Thu Oct 30 02:46:35 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): single array value to normal Proc#call
+	  (i.e. not via lambda call), should be treated just like yield.
+	  [ruby-dev:21726]
+
+Thu Oct 30 02:25:48 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#initialize):
+	  add new method to inherit @sync from @io.sync.
+
+	* ext/openssl/lib/net/protocols.rb (SSLIO#ssl_connect): no need to
+	  set sync flag explicitly.
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_initialize): call super.
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): set extra chain
+	  certificates in @extra_chain_cert.
+
+Wed Oct 29 22:02:04 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/drb/drbtest.rb: use rbconfig.rb to make the path of ruby
+	  interpreter to exec, instead of test/ruby/envutil.rb,
+
+Wed Oct 29 19:58:59 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/tcltklib/tcltklib.c (CONST84): define CONST84 when it is not
+	  defined and TCL_MAJOR_VERSION >= 8.
+
+	* ext/tcltklib/tcltklib.c (VwaitVarProc, WaitVariableProc,
+	  rb_threadVwaitProc): use CONST84 instead of CONST.
+
+	* ext/tcltklib/tcltklib.c (ip_rbTkWaitCommand,
+	  ip_rb_threadTkWaitCommand): use CONST84 always.
+
+Wed Oct 29 17:27:05 2003  Tanaka Akira  <akr@m17n.org>
+
+	* re.c (rb_reg_s_union, Init_Regexp): new method `Regexp.union'.
+
+	* lib/pathname.rb (realpath): examine Dir.pwd because it may have
+	  symlinks.
+
+Wed Oct 29 17:16:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_longjmp): must not disturb original jump.
+	  [ruby-dev:21733]
+
+Wed Oct 29 15:28:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_Proc): taint preallocated exception object
+	  sysstack_error. [ruby-talk:84534]
+
+Wed Oct 29 11:27:39 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (ret_args): node may be NULL. [ruby-talk:84530]
+
+Tue Oct 28 15:20:12 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/tcltklib/tcltklib.c (VwaitVarProc, ip_rbVwaitObjCmd,
+	  WaitVariableProc, WaitVisibilityProc, WaitWindowProc,
+	  ip_rbTkWaitObjCmd, ip_rbTkWaitCommand, rb_threadVwaitProc,
+	  rb_threadWaitVisibilityProc, rb_threadWaitWindowProc,
+	  ip_rb_threadVwaitObjCmd, ip_rb_threadTkWaitObjCmd): prototype;
+	  avoid VC++ warnings.
+
+Mon Oct 27 19:19:55 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_longjmp): ignore reentering error while warning.
+	  [ruby-dev:21730]
+
+Mon Oct 27 00:23:50 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (ip_ruby): bug fix on Win : hang-up when
+	  calling 'exit' in the Tk callback procedure. [ruby-list:38656]
+
+Sat Oct 25 09:18:04 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_method_missing): protect exception from within
+	  "inspect".  (ruby-bugs:PR#1204)
+
+Fri Oct 24 23:26:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* hash.c (rb_hash_each): Hash#each should yield single value.
+	  [ruby-talk:84420]
+
+	* hash.c (env_each): ditto for ENV.each.
+
+Thu Oct 23 20:25:32 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/server.rb (GenericServer#start): should rescue
+	  IOError from IO::accept. [ruby-dev:21692]
+
+Thu Oct 23 17:59:36 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): initialize stack bottom for embedding.
+	  [ruby-dev:21686]
+
+	* ext/dl/extconf.rb: move list of files to clean from DEPEND file,
+	  to get rid of macro redefinitions.
+
+Thu Oct 23 13:44:00 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y: integrate operations for stack_type.  [ruby-dev:21681]
+
+Thu Oct 23 00:41:45 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/calc/*, test/soap/helloworld/*: set logging threshold
+	  to ERROR.
+
+Wed Oct 22 12:53:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir#collect_file):
+	  ignore tests which raised LoadError.
+
+	* test/drb/drbtest.rb, test/ruby/test_beginendblock.rb,
+	  test/ruby/test_system.rb: avoid requiring same file twice.
+
+	* test/drb/test_drbssl.rb, test/drb/test_drbunix.rb: should not use
+	  ARGV unless invoked directly.  do not create test cases unless
+	  required libraries are available.
+
+Wed Oct 22 02:31:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): should not ignore exit_value in END
+	  execution. [ruby-dev:21670]
+
+Tue Oct 21 23:16:26 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): call finalizers and exit procs before
+	  terminating threads.
+
+	* eval.c (ruby_cleanup): preserve ruby_errinfo before ruby_finalize_0().
+
+Tue Oct 21 15:57:11 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/collector/dir.rb (Test::Unit::Collector::Dir#collect_file):
+	  prepend the directory of target file to the load path.
+
+Tue Oct 21 15:08:53 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (do_spawn, do_aspawn): should wait child process even
+	  if callded with P_OVERLAY.
+
+	* win32/win32.c (do_spawn, do_aspawn): should return child's exit
+	  status to parent.
+
+Tue Oct 21 00:35:02 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/calc/*, test/soap/helloworld/*: catch the exception from
+	  test server thread and recover.
+
+Tue Oct 21 00:22:57 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/*: import drb/runit.
+
+Mon Oct 20 23:55:47 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): set current node after arguments evaluation.
+	  [ruby-dev:21632]
+
+	* eval.c (rb_yield_0): set current node and keep it at local jump.
+
+Mon Oct 20 22:01:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_cleanup): keep thread group for main thread.
+	  [ruby-dev:21644]
+
+Mon Oct 20 18:28:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_catch): backout.
+
+Mon Oct 20 17:31:46 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (PUSH_FRAME): generate unique number to be TAG_JUMP()
+	  destination.
+
+	* eval.c (localjump_destination): use unique number in ruby_frame
+	  for localjump destination.
+
+Mon Oct 20 11:31:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_signal.rb (test_signal): restore old trap.
+
+Mon Oct 20 11:00:46 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (gc_sweep): loosen page free condition to avoid add_heap()
+	  race condition. [ruby-dev:21633]
+
+	* gc.c (gc_sweep): do not update malloc_limit when malloc_increase
+	  is smaller than malloc_limit.
+
+Mon Oct 20 09:45:12 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/debug.rb (debug_command): remove debug print.
+
+Sun Oct 19 13:12:30 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (foreachline, dir_foreach): add obsolete warning.
+
+Sun Oct 19 00:14:22 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/calc/*, test/soap/helloworkd/*: changed port# of test
+	  server. (17171)
+
+Sat Oct 18 23:01:32 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* missing/acosh.c (DBL_MANT_DIG): typo fix(ifdef -> ifndef).
+
+Sat Oct 18 05:48:59 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c: YAML::Syck::compile method.
+
+	* ext/syck/syck.c: Buffer edge bug.
+
+	* ext/syck/yaml2byte.c: YAML to bytecode converter.
+
+	* ext/syck/yamlbyte.h: Ditto.
+
+	* ext/syck/bytecode.c: Bytecode parser fixes to empty collections
+	  and empty strings.
+
+	* ext/syck/token.c: Ditto.
+
+Fri Oct 17 23:07:38 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/enumerator/enumerator.c, ext/enumerator/enumerator.txt:
+	  Provide Kernel#to_enum as an alias for Kernel#enum_for.  Maybe
+	  this is a better name.
+
+Fri Oct 17 23:00:30 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/generator.rb: Add rdoc documentation.
+
+Fri Oct 17 22:16:42 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb: Reword and fix Overview.
+
+	* lib/set.rb: It is not necessary to require
+	  'test/unit/ui/console/testrunner'.
+
+Fri Oct 17 11:15:22 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_range.rb: added.
+
+	* MANIFEST: add test/ruby/test_range.rb.
+
+Fri Oct 17 03:21:23 2003  William Sobel  <will.sobel@barra.com>
+
+	* ext/socket/socket.c (make_hostent): h_aliases may be NULL.
+	  (ruby-bugs:PR#1195)
+
+	* ext/socket/socket.c (sock_s_gethostbyaddr): ditto.
+
+Fri Oct 17 00:12:41 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: (bug fix) instance variable @frame was used
+	  without initializing on TkComposite module.
+
+Thu Oct 16 23:51:04 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: If $DEBUG == true and some exception is caused
+	  in a callback operation, Ruby/Tk shows a (verbose) backtrace
+	  information on the callback process.
+
+Thu Oct 16 17:09:19 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/debug.rb (DEBUGGER__::Context::debug_command): do not call
+	  debug_silent_eval() when $1 is not set. (ruby-bugs:PR#1194)
+
+Thu Oct 16 16:54:57 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (rb_str_upto): ("a"..."a").to_a should return [].
+	  [ruby-core:01634]
+
+Thu Oct 16 16:40:51 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb:
+	  Add Tk::EncodedString and Tk::UTF8_String class to support
+	  characters using the \uXXXX escape to the UNICODE string.
+
+	* ext/tk/sample/{demos-en,demos-jp}/unicodeout.rb
+	  new demo-scripts (samples of Tk::UTF8_String)
+
+	* ext/tk/sample/{demos-en,demos-jp}/widget
+	  add entries for 'unicodeout.rb'
+
+Thu Oct 16 08:38:06 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/digest/test_digest.rb (test_eq): show failed class.
+
+	* test/ruby/test_iterator.rb (test_break, test_return_trace_func):
+	  test localjump destination.
+
+Wed Oct 15 20:22:31 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/netHttpClient.rb: use URI::HTTP#request_uri instead of
+	  instance_eval('path_query').  [ruby-list:38575]
+
+Wed Oct 15 17:24:45 2003  URABE Shyouhei  <root@mput.dip.jp>
+
+	* lib/cgi.rb (CGI::Cookie): tiny typo fix.
+
+Wed Oct 15 15:00:54 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_run): just return FAILURE instead of parse error
+	  count.  [ruby-list:38569]
+
+Wed Oct 15 13:17:02 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/digest/digest.c (rb_digest_base_alloc): need to initialize
+	  buffer. [ruby-dev:21622]
+
+Wed Oct 15 11:23:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_object): dump extended modules as well.
+
+	* marshal.c (r_object0): TYPE_USRMARSHAL should restore extended
+	  modules before invoking marshal_load.  these two fixes are done
+	  by Masatoshi Seki <m_seki@mva.biglobe.ne.jp>.
+
+Wed Oct 15 09:30:34 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/enumerator/enumerator.c (enumerator_each): avoid VC++ warning.
+
+	* ext/syck/syck.h: include stdio.h for definition of FILE.
+
+Wed Oct 15 08:09:07 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/bytecode.c: Checkin of YAML bytecode support.
+
+	* ext/syck/gram.c: Ditto.
+
+	* ext/syck/syck.c: Ditto.
+
+	* ext/syck/token.c: Ditto.
+
+	* ext/syck/handler.c: Ditto.
+
+	* ext/syck/handler.c: Now using 'tag' rather than 'taguri' in type URIs.
+
+	* ext/syck/rubyext.c: Ditto (on both counts).
+
+Wed Oct 15 05:05:53 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/generator.rb: A new library which converts an internal
+	  iterator to an external iterator.
+
+	* lib/abbrev.rb: A new library which creates an abbreviation table
+	  from a list.
+
+Wed Oct 15 04:31:51 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/sample/demos-en/entry3.rb, ext/tk/sample/demos-jp/entry3.rb:
+	  new demo-scripts
+
+	* ext/tk/sample/demos-en/widget, ext/tk/sample/demos-jp/widget:
+	  add entries for 'entry3.rb'
+
+Wed Oct 15 04:31:47 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/digest/test_digest.rb: Moved from ext/digest/test.rb.
+
+Wed Oct 15 03:53:20 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/tk.rb: fixed trouble on auto-load Tcl commands (enbug
+	  on the last commit).
+
+Wed Oct 15 00:25:00 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (yylex): argument parentheses preceded by spaces should
+	  be warned; not error.  [ruby-talk:84103]
+
+Wed Oct 15 00:20:15 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c: replace Tcl/Tk's vwait and tkwait to
+	  switch on threads smoothly and avoid seg-fault.
+
+	* ext/tcltklib/tcltklib.c: add TclTkIp._thread_vwait and
+	  _thread_tkwait for waiting on a thread. (Because Tcl/Tk's vwait
+	  and tkwait command wait on an eventloop.)
+
+	* ext/tk/lib/multi-tk.rb: support TclTkIp._thread_vwait and
+	  _thread_tkwait.
+
+	* ext/tk/lib/tk.rb: now, TkVariable#wait has 2 arguments.
+	  If 1st argument is true, waits on a thread. If false, waits on
+	  an eventloop. If 2nd argument is true, checks existence of
+	  rootwidgets. If false, doesn't. Default is wait(true, false).
+
+	* ext/tk/lib/tk.rb: add TkVariable#tkwait(arg) which is equal to
+	  TkVariable#wait(arg, true). wait_visibility and wait_destroy
+	  have an argument for waiting on a thread or an eventloop.
+
+	* ext/tk/lib/tk.rb: improve of accessing Tcl/Tk's special variables.
+
+	* ext/tk/lib/tkafter.rb: support 'wait on a thread' and 'wait on
+	  an eventloop'.
+
+Wed Oct 15 00:10:24 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/baseData.rb: Introduce SOAPType as the common ancestor of
+	  SOAPBasetype and SOAPCompoundtype.
+
+	* lib/soap/generator.rb, lib/soap/element.rb, lib/soap/encodingstyle/*:
+	  Encoding methods signature change.  Pass SOAPGenerator as a parameter.
+
+	* lib/soap/mapping/*, test/soap/marshal/test_marshal.rb: Refactoring
+	  for better marshalling/unmarshalling support.  Now I think SOAP
+	  marshaller supports all kind of object graph which is supported by
+	  Ruby's original marshaller.  Of course there could be bugs as always.
+	  Find it.  :-)
+
+	* lib/soap/rpc/standaloneServer.rb: Set severity threshould to INFO.
+	  DEBUG is too noisy.
+
+	* lib/xsd/datatypes.rb: DateTime#of is obsoleted.  Use DateTime#offset.
+
+	* test/wsdl/emptycomplextype.wsdl, test/xsd/xmlschema.xml: Avoid
+	  useless warning.
+
+Tue Oct 14 19:09:35 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_finalize_0): return the given exit status unless
+	  SystemExit got raised.
+
+Tue Oct 14 11:53:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h (ruby_stop): never return.
+
+	* ruby.h (ruby_run): ditto.
+
+Tue Oct 14 04:43:55 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (realpath): make ELOOP check bit more robust.
+	  (children): prepend self by default.
+	  (chroot): obsoleted.
+
+Tue Oct 14 02:29:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_require_safe): segfault after loading .so.
+
+Tue Oct 14 02:05:23 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/Setup*, ext/enumerator/*: Add ext/enumerator, a helper
+	  module for the Enumerable interface.
+
+Mon Oct 13 23:55:59 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* test/ruby/envutil.rb: use Config::CONFIG["ruby_install_name"],
+	  not "ruby".
+
+Mon Oct 13 23:57:29 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_feature_p): match by classified suffix.
+
+	* eval.c (rb_require_safe): require library in the specified safe
+	  level.
+
+	* variable.c (rb_autoload, rb_autoload_load): restore safe level
+	  when autoload was called.  [ruby-dev:21338]
+
+	* intern.h: prototypes; rb_require_safe.
+
+	* test/runner.rb: accept non-option arguments.
+
+Mon Oct 13 20:49:51 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_new4): should not preserve FL_TAINT status in the
+	  internal shared string. [ruby-dev:21601]
+
+	* string.c (rb_str_new4): ditto.
+
+	* eval.c: use EXIT_SUCCESS and EXIT_FAILURE for exit values.
+
+	* process.c: ditto. [ruby-list:38521]
+
+Mon Oct 13 19:51:02 2003  Koji Arai  <jca02266@nifty.ne.jp>
+
+	* lib/debug.rb (debug_command): should enter emacs mode when
+	  assigned any value to the environment variable "EMACS".
+	  On Meadow, (getenv "EMACS") is "meadow".
+
+Sun Oct 12 14:45:03 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/win32ole/extconf.rb: check "windows.h", not "windows".
+	  [ruby-talk:84051]
+
+Sat Oct 11 20:41:03 2003  Corinna Vinschen  <corinna@vinschen.de>
+
+	* file.c (eaccess): Use access(2) on Cygwin.
+
+Sat Oct 11 17:09:21 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/rexml/quickpath.rb (REXML::QuickPath::match):
+	  escape '[' to avoid warning.
+
+Sat Oct 11 16:08:41 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (realpath): check existence of the file.
+
+	* lib/pathname.rb (realpath): re-implemented.
+	  (realpath_root?, realpath_rec): removed
+
+Sat Oct 11 10:19:39 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb: handle exceptions correctly. Thanks, Gennady
+	  Bystritsky.
+
+Fri Oct 10 07:50:54 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (is_defined): inheritance line adjustment as like as
+	  rb_call_super().
+
+Fri Oct 10 01:19:00 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_initialize): add
+	  optional argument to specify the DirectoryString type
+	  (ASN1::UTF8STRING by default). RFC3280 deprecates PrintableString
+	  for DirectoryString, and strongly requires to use UTF8String for
+	  all certificates issued after December, 31 2003.
+
+	* ext/openssl/lib/openssl/x509.rb (X509::Name::parse): ditto.
+
+Thu Oct  9 23:50:21 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_start_0): prevent thread from GC.
+	  [ruby-dev:21572]
+
+Thu Oct  9 19:11:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_start_0): non-volatile should be restored from
+	  volatile.
+
+Thu Oct  9 17:43:36 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (proc_save_safe_level, proc_get_safe_level,
+	  proc_set_safe_level): save/restore safe level 1..4.
+
+Thu Oct  9 16:33:23 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (r_object0): remove unnecessary iv restoration for
+	  USRMARSHAL. [ruby-dev:21582]
+
+	* marshal.c (w_object): dump generic instance variables from
+	  a string from '_dump'.
+
+	* variable.c (rb_generic_ivar_table): return 0 if obj's FL_EXIVAR
+	  is not set.
+
+	* time.c (time_dump): copy instance variables to dumped string, to
+	  be included in the marshaled data.
+
+	* bignum.c (rb_big2ulong): add range check to ensure round trip.
+
+Thu Oct  9 15:45:27 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* pack.c (uv_to_utf8): change message to "out of range", since
+	  negative values are not "too big". [ruby-dev:21567]
+
+Thu Oct  9 14:05:38 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_set_end_proc, rb_exec_end_proc): restore safe level.
+	  [ruby-dev:21557]
+
+Thu Oct  9 10:51:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_yield_0): no error if block is empty.
+
+Thu Oct  9 06:43:33 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (localjump_error): id should be ID.
+
+	* eval.c (rb_eval): nd_rval is set in copy_node_scope().
+
+	* eval.c (rb_yield_0): unused variable.
+
+	* eval.c (rb_yield_0): nothing to do for empty node.
+
+	* eval.c (call_end_proc, proc_invoke): adjust backtrace in END.
+	  [ruby-dev:21551]
+
+	* eval.c (rb_thread_start_0): set the value by break as the result.
+	  [ruby-dev:21552]
+
+	* eval.c (rb_thread_start_0, rb_thread_raise, rb_callcc): save
+	  variables across THREAD_SAVE_CONTEXT.
+
+Thu Oct  9 12:05:46 2003  Eric Sunshine  <sunshine@sunshineco.com>
+
+	* configure.in: revived NextStep, OpenStep, and Rhapsody ports which
+	  had become unbuildable; enhanced --enable-fat-binary option so that
+	  it accepts a list of desired architectures (rather than assuming a
+	  fixed list), or defaults to a platform-appropriate list if user does
+	  not provide an explicit list; made the default list of architectures
+	  for MAB (fat binary) more comprehensive; now uses -fno-common even
+	  when building the interpreter (in addition to using it for
+	  extensions), thus allowing the interpreter to be embedded into a
+	  plugin module of an external project (in addition to allowing
+	  embedding directly into an application); added checks for
+	  <netinet/in_systm.h> (needed by `socket' extension) and getcwd(); now
+	  ensures that -I/usr/local/include is employed when extensions'
+	  extconf.rb scripts invoke have_header() since extension checks on
+	  NextStep and OpenStep will fail without it if the desired resource
+	  resides in the /usr/local tree; fixed formatting of --help message.
+
+	* Makefile.in: $(LIBRUBY_A) rule now deletes the archive before
+	  invoking $(AR) since `ar' on Apple/NeXT can not "update" MAB archives
+	  (see configure's --enable-fat-binary option); added rule for new
+	  missing/getcwd.c.
+
+	* defines.h: fixed endian handling during MAB build (see configure's
+	  --enable-fat-binary option) to ensure that all portions of the
+	  project see the correct WORDS_BIGENDIAN value (some extension modules
+	  were getting the wrong endian setting); added missing constants
+	  GETPGRP_VOID, WNOHANG, WUNTRACED, X_OK, and type pid_t for NextStep
+	  and OpenStep; removed unnecessary and problematic HAVE_SYS_WAIT_H
+	  define in NeXT section.
+
+	* dir.c: do not allow NAMLEN() macro to trust dirent::d_namlen on
+	  NextStep since, on some installations, this value always resolves
+	  uselessly to zero.
+
+	* dln.c: added error reporting to NextStep extension loader since the
+	  previous behavior of failing silently was not useful; now ensures
+	  that NSLINKMODULE_OPTION_BINDNOW compatibility constant is defined
+	  for OpenStep and Rhapsody; no longer includes <mach-o/dyld.h> twice
+	  on Rhapsody since this header lacks multiple-include protection,
+	  which resulted in "redefinition" compilation errors.
+
+	* main.c: also create hard reference to objc_msgSend() on NeXT
+	  platforms (in addition to Apple platforms).
+
+	* lib/mkmf.rb: now exports XCFLAGS from configure script to extension
+	  makefiles so that extensions can be built MAB (see configure's
+	  --enable-fat-binary option); also utilize XCFLAGS in cc_command()
+	  (but not cpp_command() because MAB flags are incompatible with
+	  direct invocation of `cpp').
+
+	* ext/curses/extconf.rb: now additionally checks for presence of these
+	  curses functions which are not present on NextStep or Openstep:
+	  bkgd(), bkgdset(), color(), curs(), getbkgd(), init(), scrl(), set(),
+	  setscrreg(), wattroff(), wattron(), wattrset(), wbkgd(), wbkgdset(),
+	  wscrl(), wsetscrreg()
+
+	* ext/curses/curses.c: added appropriate #ifdef's for additional set of
+	  curses functions now checked by extconf.rb; fixed curses_bkgd() and
+	  window_bkgd() to correctly return boolean result rather than numeric
+	  result; fixed window_getbkgd() to correctly signal an error by
+	  returning nil rather than -1.
+
+	* ext/etc/etc.c: setup_passwd() and setup_group() now check for null
+	  pointers before invoking rb_tainted_str_new2() upon fields extracted
+	  from `struct passwd' and `struct group' since null pointers in some
+	  fields are common on NextStep/OpenStep (especially so for the
+	  `pw_comment' field) and rb_tainted_str_new2() throws an exception
+	  when it receives a null pointer.
+
+	* ext/pty/pty.c: include "util.h" for strdup()/ruby_strdup() for
+	  platforms such as NextStep and OpenStep which lack strdup().
+
+	* ext/socket/getaddrinfo.c: cast first argument of getservbyname(),
+	  gethostbyaddr(), and gethostbyname() from (const char*) to non-const
+	  (char*) for older platforms such as NextStep and OpenStep.
+
+	* ext/socket/socket.c: include "util.h" for strdup()/ruby_strdup() for
+	  platforms such as NextStep and OpenStep which lack strdup(); include
+	  <netinet/in_systm.h> if present for NextStep and OpenStep; cast first
+	  argument of gethostbyaddr() and getservbyname() from (const char*) to
+	  non-const (char*) for older platforms.
+
+	* ext/syslog/syslog.c: include "util.h" for strdup()/ruby_strdup() for
+	  platforms such as NextStep and OpenStep which lack strdup().
+
+Wed Oct  8 22:19:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit.rb: removed installation instructions.
+
+	* lib/test/unit/ui/testrunnermediator.rb: moved the run flag to a more
+	  central location.
+
+	* lib/test/unit.rb: ditto.
+
+	* lib/test/unit.rb: extracted the running code in to AutoRunner.
+
+	* lib/test/unit/autorunner.rb: added.
+
+	* lib/test/unit/collector/objectspace.rb: extracted common test
+	  collection functionality in to a module.
+
+	* lib/test/unit/collector.rb: ditto; added.
+
+	* test/testunit/collector/test_objectspace.rb: ditto.
+
+	* lib/test/unit/collector/dir.rb: added. Supports collecting tests out
+	  of a directory structure.
+
+	* test/testunit/collector/test_dir.rb: added.
+
+	* test/runner.rb: simplified to use the new capabilities.
+
+Tue Oct  7 15:23:09 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb: add tests for nested BEGIN/END.
+
+	* test/ruby/beginmainend.rb: add tests for nested BEGIN/END.
+
+	* test/ruby/endblockwarn.rb: new file added to test of END-in-method
+	  warning.
+
+Tue Oct  7 12:23:47 2003  Tanaka Akira  <akr@m17n.org>
+
+	* ext/fcntl/fcntl.c (Init_fcntl): define Fcntl::O_ACCMODE.
+
+	* ext/socket/extconf.rb: useless assignment removed.
+
+Tue Oct  7 09:13:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb (test_endinmethod): END{} is now
+	  allowed in eval.
+
+Tue Oct  7 04:15:25 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt): should not expand mrhs if lhs is solely starred.
+
+Tue Oct  7 02:57:53 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (stmt): rhs of multiple assignment should not be
+	  expanded using "to_a". [ruby-dev:21527]
+
+Tue Oct  7 01:42:34 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_get_asn1type): use appropriate
+	  free function for ASN1_OBJECT.
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1obj_get_sn): add new function for
+	  ASN1::ObjectId#sn; it returns short name text representation of OID.
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1obj_get_ln): add new function for
+	  ASN1::ObjectId#ln; it returns long name text representation of OID.
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1obj_get_oid): add new function for
+	  ASN1::ObjectId#oid; it returns numerical representation of OID.
+
+Mon Oct  6 22:59:46 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb (IOReader, BasicWriter): call binmode when a given IO
+	  respond_to?(:binmode).  record separator was wrong when you gave
+	  text mode IO to Reader.parse and Writer.generate.
+
+	* test/csv/test_csv.rb: add tests for above change.
+
+Mon Oct  6 16:23:38 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): wrong method name in the message.
+
+Mon Oct  6 16:02:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (stmt): END in method should cause warning.
+	  [ruby-dev:21519]
+
+Mon Oct  6 15:17:23 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_iterator.rb (test_block_argument_without_paren):
+	  added. (follows sample/test.rb)
+
+Mon Oct  6 11:57:06 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb, test/ruby/beginmainend.rb: added
+	  test for eval-ed BEGIN END order.
+
+Mon Oct  6 09:19:54 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_object): should pass "weak" value to next level.
+	  [ruby-dev:21496]
+
+	* eval.c (proc_alloc): should not use cached object if klass is
+	  different. [ruby-talk:83685]
+
+Sun Oct  5 23:27:09 2003  Tanaka Akira  <akr@m17n.org>
+
+	* ext/socket/extconf.rb: check recvmsg even if sendmsg is exists.
+
+	* ext/socket/socket.c (thread_read_select): restored.
+
+	* lib/pathname.rb: version information is added in document.
+
+Sun Oct  5 23:07:03 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_f_END): block should be given.  [ruby-dev:21497]
+
+Sun Oct  5 22:51:23 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/ext/openssl/extconf.rb: add check for some engine functions
+	  unavailable in OpenSSL-0.9.6.
+
+	* lib/ext/openssl/ossl_engine.c: ditto.
+
+Sun Oct  5 17:56:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): fix evaluation order.  [ruby-list:38431]
+
+Sun Oct  5 15:05:06 2003  akira yamada  <akira@ruby-lang.org>
+
+	* test/uri/*: translated RUNIT to Test::Unit.
+
+Sun Oct  5 14:37:39 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/xsd/datatypes.rb: Rational -> Decimal string bug fix.
+
+	* test/soap/marshal/test_marshal.rb: ditto.
+
+	* test/soap/calc/test_calc_cgi.rb: add Config::CONFIG["EXEEXT"] to
+	  RUBYBIN.
+
+Sun Oct  5 13:47:22 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_beginendblock.rb, test/ruby/beginmainend.rb: add tests
+	  about scope, order and allowed syntax.
+
+Sun Oct  5 11:54:29 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/envutil.rb: added.  split "rubybin" from test_system.rb.
+
+	* test/ruby/test_system.rb: use envutil.rb
+
+	* test/ruby/test_beginendblock.rb: added.
+
+	* test/ruby/beginmainend.rb: added.  used in test_beginendblock.rb.
+
+Sun Oct  5 11:23:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* test/testunit/runit/test_testresult.rb: removed some unnecessary
+	  cruft.
+
+Sun Oct  5 11:14:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/rubyunit.rb: aliasing TestCase into the top level is
+	  problematic.
+
+	* lib/runit/assert.rb: fixed a couple of bugs caused by recent
+	  refactoring in Test::Unit.
+
+	* test/testunit/runit/*: added.
+
+Sun Oct  5 10:55:29 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/open-uri.rb (URI::Generic#find_proxy): no_proxy support did not
+	  work.  [ruby-dev:21484]
+
+Sun Oct  5 09:52:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: will use pp for output if available.
+	  Can be disabled by setting Assertions.use_pp = false.
+
+	* test/testunit/test_assertions.rb: made a small change to exception
+	  formatting.
+
+Sun Oct  5 07:42:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: made small improvements to assertion
+	  messages. Deprecated Assertions#assert_not_nil; use #assert instead.
+
+	* test/testunit/test_assertions.rb: ditto.
+
+	* test/testunit/util/test_procwrapper.rb: use #assert instead of
+	  #assert_not_nil.
+
+Sun Oct  5 04:10:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: refactored message building.
+
+Sun Oct  5 03:40:22 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.h: global symbols should be declared
+	  as external.
+
+Sun Oct  5 03:03:20 2003  akira yamada  <akira@ruby-lang.org>
+
+	* test/ruby/test_exception.rb (test_else): added.
+
+Sun Oct  5 02:12:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: changed assertion messages to rely more
+	  heavily on #inspect. Added backtrace filtering for exceptions in
+	  assertion messages.
+
+	* test/testunit/test_assertions.rb: ditto.
+
+Sun Oct  5 02:12:00 2003  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/acl.rb, lib/drb/ssl.rb: added.
+
+	* lib/drb/drb.rb: exit from a thread using 'break'.
+
+Sat Oct  4 21:49:14 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* gc.c (Init_stack): the type of space is changed to unsigned int
+	  from double.  [ruby-dev:21483]
+
+Sat Oct  4 17:52:59 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/netHttpClient.rb: follow http-access2.  hosts which matches
+	  ENV['no_proxy'] or ENV['NO_PROXY'] are not proxyed.
+	  - [,:] separated. ("ruby-lang.org:rubyist.net")
+	  - no regexp. (give "ruby-lang.org", not "*.ruby-lang.org")
+	  - if you want specify host by IP address, give full address.
+	    ("192.168.1.1, 192.168.1.2")
+
+	* lib/soap/rpc/cgistub.rb: return "Status: XXX MMM" line.
+
+	* test/runner.rb: give testsuite name.
+
+Sat Oct  4 15:16:02 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_object): instance variable dump do not cause error
+	  for objects that cannot be dumped, if they traversed from
+	  marshal_dump.  they are just ignored.
+
+	* gc.c (Init_stack): cast "space" (doble value) into unsigned
+	  int.  should run on PowerPC.
+
+	* eval.c (rb_eval): should not execute else part if any exception
+	  is caught. [ruby-dev:21482]
+
+	* parse.y (f_args): should allow unparenthesized block argument.
+
+	* parse.y (f_rest_arg): should allow unparenthesized rest
+	  argument.
+
+Sat Oct  4 14:59:51 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (initialize): raise ArgumentError if argument has
+	  '\0' character.
+	  (relative_path_from): new method.
+	  (each_entry): new method for replacement of dir_foreach.
+	  (foreach, foreachline, dir_foreach, chdir): obsoleted.
+
+Sat Oct  4 12:58:48 2003  akira yamada  <akira@ruby-lang.org>
+
+	* test/uri/* (6 files): added.
+
+Sat Oct  4 12:44:45 2003  akira yamada  <akira@ruby-lang.org>
+
+	* lib/uri/ftp.rb, lib/uri/mailto.rb: renamed to #to_s from #to_str.
+
+Sat Oct  4 07:33:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/testsuite.rb: changed #<< to return self, and added
+	  #delete.
+
+	* test/testunit/test_testsuite.rb: ditto. Also slightly refactored
+	  #test_size.
+
+	* lib/test/unit/collector/objectspace.rb: collector now preserves the
+	  hierarchy of suites.
+
+	* test/testunit/collector/test_objectspace.rb: ditto.
+
+Sat Oct  4 04:48:49 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c: default keys handled.
+
+	* ext/syck/syck.h: lowered default buffer size to 16k for increased
+	  performance.
+
+	* test/yaml: checkin of basic unit tests.
+
+Sat Oct  4 04:24:19 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/extconf.rb: add check for X509V3_set_nconf.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_set_config):
+	  cannot implement if X509V3_set_nconf doesn't exist.
+
+Sat Oct  4 02:12:44 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/xsd/datatypes.rb: dump sign by itself.  under the problematic
+	  platform, sprintf("%+.10g", -0.0) => +0.  sigh.
+
+	* sample/wsdl/amazon/*: update schema ver2 to ver3.
+
+Sat Oct  4 01:33:46 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb (initialize): duplicate and freeze argument.
+	  (to_s): return duplicated string.
+	  (children): new method.
+	  (each_line): new alias to foreachline.
+
+Fri Oct  3 16:13:19 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_asn1.c: add DER encoder and decoder.
+
+	* ext/openssl/ossl_asn1.h: add OpenSSL::ASN1 module.
+
+	* ext/openssl/ossl.c (Init_openssl): call Init_ossl_asn1.
+
+	* ext/openssl/extconf.rb: check if X509_ATTRIBUTE has field "single".
+
+	* ext/openssl/ossl_x509attr.c (ossl_x509attr_set_value): accept
+	  DER encoded data argument.
+
+	* ext/openssl/ossl_x509attr.c (ossl_x509attr_get_value): return
+	  DER encoded data in OpenSSL::ASN1 types.
+
+Fri Oct  3 13:02:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit.rb: refactored to use optparse.
+
+	* lib/test/unit.rb: added support for selecting the output
+	  level from the command-line.
+
+	* lib/test/unit.rb: added a command-line switch to stop processing
+	  the command-line, allowing arguments to be passed to tests.
+
+	* lib/test/unit.rb: changed the method for specifying a runner or a
+	  filter from the command-line.
+
+	* lib/test/unit/collector/objectspace.rb: fixed a bug causing all
+	  tests to be excluded when the filter was set to an empty array.
+
+	* test/testunit/collector/test_objectspace.rb: ditto.
+
+Fri Oct  3 08:14:32 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb (RubyLex::identify_identifier): support
+	  'class ::Foo' syntax. [ruby-talk:83514]
+
+Fri Oct  3 08:01:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: added a default message for #assert,
+	  #assert_block, and #flunk.
+
+	* test/testunit/test_assertions.rb: ditto.
+
+	* lib/test/unit/failure.rb: failures now show a better trace of where
+	  they occurred.
+
+	* test/testunit/test_failure.rb: ditto (added).
+
+	* lib/test/unit/testcase.rb: ditto.
+
+	* test/testunit/test_testcase.rb: ditto.
+
+	* lib/test/unit/util/backtracefilter.rb: added.
+
+	* test/testunit/util/test_backtracefilter.rb: added.
+
+	* lib/test/unit/error.rb: changed to use BacktraceFilter and improved
+	  output.
+
+	* test/testunit/test_error.rb: ditto.
+
+Thu Oct  2 20:33:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_failure_initialize): conform with
+	  orthodox initialization method.
+
+	* ext/iconv/iconv.c (iconv_fail): initialize exception instance
+	  from the class, and do not share instance variables with the
+	  others.  [ruby-dev:21470]
+
+Thu Oct  2 18:20:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (Init_Time): define initialize.  [ruby-dev:21469]
+
+Thu Oct  2 17:39:38 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_engine.c: add a new module OpenSSL::Engine.
+	  it supports OpenSSL hardware cryptographic engine interface.
+
+	* ext/openssl/ossl_engine.h: ditto.
+
+	* ext/openssl/MANIFEST: add ossl_engine.c and ossl_engine.h.
+
+	* ext/openssl/extconf.rb: add check for openssl/engine.h.
+
+	* ext/openssl/ossl.c: call Init_ossl_engine().
+
+	* ext/openssl/ossl.h: include openssl/engine.h.
+
+	* ext/openssl/ossl_pkey_{rsa,dsa,dh}.c: check if underlying
+	  EVP_PKEY referes engine.
+
+Thu Oct  2 17:22:37 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_load): restore instance variables (if any) before
+	  loading from marshaled data.
+
+Thu Oct  2 14:19:15 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (iconv_fail): now yield erred substring, and
+	  set error object to $!.
+
+	* ext/iconv/iconv.c (iconv_convert): error handler block should
+	  return appended part and the rest.  if rest is nil, the
+	  conversion stops.
+
+Thu Oct  2 12:00:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_const_defined_0): look up constants in Object as
+	  well.  [ruby-dev:21458]
+
+	* test/ruby/test_defined.rb (TestDefined::test_defined): test for
+	  constants.
+
+Thu Oct  2 11:17:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb: should not capture an
+	  AssertionFailedError unless explicitly requested.
+
+	* test/testunit/test_assertions.rb: ditto.
+
+	* test/testunit/collector/test_objectspace.rb: fixed a test failure
+	  caused by methods being returned in different orders on different
+	  platforms by moving test sorting from TestSuite into the locations
+	  where suites are constructed. [ruby-talk:83156]
+
+	* lib/test/unit/testcase.rb: ditto.
+
+	* lib/test/unit/testsuite.rb: ditto.
+
+	* lib/test/unit/collector/objectspace.rb: ditto.
+
+Thu Oct  2 03:25:01 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval.c (rb_thread_raise): prototype; avoid VC++ warning.
+
+Thu Oct  2 01:37:34 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* time.c (time_mdump): new marshal dumper. _dump is still
+	  available for compatibility.
+
+	* time.c (time_mload): new marshal loader.
+
+	* marshal.c (w_object): preserve instance variables for objects
+	  with marshal_dump.
+
+	* marshal.c (r_object0): restore instance variables before calling
+	  marshal_load.
+
+	* error.c (rb_warn_m): always return nil.
+
+Thu Oct  2 01:32:46 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_block_given_p): real required condition is
+	  ruby_frame->prev->iter == ITER_CUR.
+
+	* eval.c (rb_block_given_p): ditto.
+
+	* eval.c (block_pass): update ruby_frame->iter only when previous
+	  value is ITER_NOT.
+
+Thu Oct  2 01:02:35 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_const_defined_at): should exclude constants from
+	  Object when TYPE(klass) == T_MODULE *and* exclude is on.
+	  [ruby-dev:21458]
+
+	* variable.c (rb_const_get_0): do not lookup constants from Object
+	  when TYPE(klass) == T_MODULE *and* exclude is on.
+
+Thu Oct  2 00:21:11 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/logger/test_logger.rb: unlinking file before close causes
+	  problem under win32 box.
+
+	* lib/xsd/datatypes.rb(XSDFloat, XSDDouble): add +/- sign explicitly
+	  when stringified and embedded into XML instance.  Ruby's sprintf may
+	  format -0.0 as "0.0" (no minus sign) depending on underlying C
+	  sprintf implementation.
+
+	* test/xsd/test_xsd.rb, test/soap/test_basetype.rb: follow above change.
+
+	* test/soap/calc/*: give httpd config param "CGIInterpreter".
+	  "/usr/bin/env ruby" thing does not work under non-Unix boxes.
+
+Thu Oct  2 00:25:21 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (ruby_signal_name): adjust to the prototype.
+
+	* process.c (pst_inspect): ditto.
+
+	* ext/etc/etc.c (etc_getgrent, Init_etc): typo.
+
+Wed Oct  1 20:49:41 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (heaps): manage slots and limits together.  [ruby-dev:21453]
+
+	* gc.c (add_heap): should not clear heaps slot even if realloc()
+	  failed.
+
+Wed Oct  1 20:36:49 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* MANIFEST: add wince/mkconfig_wce.rb.
+
+Wed Oct  1 17:22:33 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/etc/etc.c: add new functions: setpwent, getpwent, endpwent,
+	  setgrent, getgrent, endgrent.
+
+	* ext/socket/socket.c (sock_s_gethostbyname): do not reverse lookup.
+
+Wed Oct  1 17:01:30 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_load): Object scope had priority over required file
+	  scope.  [ruby-dev:21415]
+
+Wed Oct  1 14:09:53 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
+
+	* wince/mkconfig_wce.rb: sorry, forget to commit.
+
+Wed Oct  1 10:08:42 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
+
+	* wince/setup.mak: add sigmarionIII SDK support.
+
+	* wince/Makefile.sub: ditto.
+
+	* wince/mkexports.rb: fix linker error in SH4.
+
+	* wince/mkconfig_wce.rb: camouflage RUBY_PLATFORM for compiling ext.
+
+Wed Oct  1 08:02:52 2003  Takaaki Uematsu  <uema2x@jcom.home.ne.jp>
+
+	* wince/time_wce.c (time): add zero check.
+
+Tue Sep 30 16:11:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* Makefile.in: copy lex.c from $(srcdir) if it's not the current
+	  directory.  [ruby-dev:21437]
+
+Tue Sep 30 11:29:23 2003  Tanaka Akira  <akr@m17n.org>
+
+	* process.c (pst_inspect): describe stopped process "stopped".
+
+Tue Sep 30 09:31:56 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/runner.rb: glob for directories.
+
+Tue Sep 30 09:11:43 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): while/until should not capture break unless
+	  they are destination of the break.
+
+Tue Sep 30 03:12:02 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (finish): revert to 1.93.
+
+	* lib/net/pop.rb (finish): revert to 1.60.
+
+	* lib/net/smtp.rb (finish): revert to 1.67.
+
+	* lib/net/http.rb (do_start): ensure to close socket if failed to
+	  start session.
+
+	* lib/net/pop.rb (do_start): ditto.
+
+	* lib/net/smtp.rb (do_start): ditto.
+
+	* lib/net/smtp.rb: SMTP#started? wrongly returned false always.
+
+Tue Sep 30 02:54:49 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_iterator.rb: new test
+	  test_break__nested_loop[123].
+
+Mon Sep 29 23:39:13 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb (finish): does not raise IOError even if
+	  !started?, to allow closing socket which was opened before
+	  session started.
+
+	* lib/net/pop.rb (finish): ditto.
+
+	* lib/net/smtp.rb (finish): ditto.
+
+Mon Sep 29 19:06:51 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* ext/win32ole/extconf.rb: add windows.h checking.
+	  (ruby-bugs:PR#1185)
+
+Mon Sep 29 16:18:30 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: check if the given logdevice object respond_to :write
+	  and :close, not is_a? IO.  duck duck.
+
+	* test/logger/test_logger.rb: self IO.pipe reading/writing may be
+	  locked by the flood.  use tempfile.
+
+	* lib/wsdl/xmlSchema/data.rb: wrong constant reference.
+
+Mon Sep 29 16:11:23 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb: clean up temporary symlink.
+	  Patched by NaHi.  [ruby-dev:21420]
+
+Mon Sep 29 11:16:55 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_thread_atfork): wrong format specifier.
+	  [ruby-dev:21428]
+
+	* process.c (pst_inspect): better description.
+
+Mon Sep 29 02:31:44 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/utils.rb (Utils::su): use setgid and setuid to
+	  set real and effective IDs. and setup group access list by
+	  initgroups.
+
+Sun Sep 28 11:14:19 2003  Koji Arai  <jca02266@nifty.ne.jp>
+
+	* ext/digest/digest.c (Init_digest): `copy_object' was deprecated.
+	  `initialize_copy' should be defined.
+
+	* ext/stringio/stringio.c (Init_stringio): ditto.
+
+Sat Sep 27 18:25:13 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/xsd/charset.rb: XSD::Charset.is_ces did return always true under
+	  $KCODE = "NONE" environment.  check added.
+
+	* test/xsd/test_xsd.rb: add tests for above fix.
+
+Sat Sep 27 15:58:50 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/rpc/cgistub.rb: make logging severity threshold higher.
+
+	* lib/soap/rpc/standaloneServer.rb: defer WEBrick server start to give
+	  a chance to reset logging severity threshold.
+
+	* test/soap/calc/test_*, test/soap/helloworld/test_helloworld.rb: run
+	  silent.
+
+Sat Sep 27 09:44:18 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/test_fileutils.rb: clear all errors on Windows.
+	  [ruby-dev:21417]
+
+	* test/fileutils/test_nowrite.rb: ditto.
+
+Sat Sep 27 04:57:07 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_file.rb: new file.  only asserts unlink-before-close
+	  behaviour now.
+
+	* test/soap/marshal/test_digraph.rb: should close before unlink.
+	  unlink-before-close pattern is not needed here.
+
+Sat Sep 27 03:32:37 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/soap/*, test/wsdl/*, test/xsd/*: move TestCase classes into
+	  each module namespace.  TestMarshal in
+	  test/soap/marshal/test_marshal.rb crashed with
+	  test/ruby/test_marshal.rb.
+
+Sat Sep 27 01:30:59 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_connect): on win32, type of the 4th
+	  argument of getsockopt is char *.
+
+Fri Sep 26 18:35:40 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/resolv-replace.rb: 1.8 compliance.  [ruby-talk:82946]
+
+Fri Sep 26 17:39:27 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_marshal.rb: add test for ruby's objects.
+
+Fri Sep 26 09:52:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defines.h (flush_register_windows): use volatile only for gcc on
+	  Solaris.  [ruby-dev:21403]
+
+	* lib/mkmf.rb (xsystem): use system directly to honor shell meta
+	  charaters.
+
+Fri Sep 26 00:10:13 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/README: updated.
+
+Thu Sep 25 17:48:10 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl.c (ossl_buf2str): fix type of 1st argument for
+	  rb_protect.
+
+	* ext/openssl/ossl_hmac.c (ossl_hmac_digest): should return meaningful
+	  value.
+
+Thu Sep 25 09:00:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/ostruct.rb: Added OpenStruct#==.
+
+	* test/ostruct/test_ostruct.rb: Added.
+
+Thu Sep 25 07:55:26 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c, ext/openssl/ossl_pkey_dsa.c,
+	  ext/openssl/ossl_pkey_rsa.c, ext/bigdecimal/bigdecimal.h: must
+	  not use C++ or C99 style comment yet.  (ruby-bugs:PR#1184)
+
+Thu Sep 25 00:23:22 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* MANIFEST: add SOAP4R.
+
+Thu Sep 25 00:13:15 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/soap/* (29 files): SOAP4R added.
+
+	* lib/wsdl/* (42 files): WSDL4R added.
+
+	* lib/xsd/* (12 files): XSD4R added.
+
+	* test/soap/* (16 files): added.
+
+	* test/wsdl/* (2 files): added.
+
+	* test/xsd/* (3 files): added.
+
+	* sample/soap/* (27 files): added.
+
+	* sample/wsdl/* (13 files): added.
+
+Wed Sep 24 02:08:11 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb: conform to mswin32.
+	  [ruby-talk:82735], [ruby-talk:82748], [ruby-talk:82818]
+
+Tue Sep 23 23:10:16 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: add Logger#<<(msg) for writing msg without any
+	  formatting.
+
+	* test/logger/test_logger.rb: ditto.
+
+Tue Sep 23 20:47:51 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* error.c (rb_warn_m): should not warn if -W0 is specified.
+	  [ruby-talk:82675]
+
+Mon Sep 22 21:28:57 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* MANIFEST: updated.
+
+Mon Sep 22 19:22:26 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* configure.in (AC_CHECK_FUNCS): add setuid and setgid.
+
+Mon Sep 22 12:34:55 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* util.c (ruby_strtod): skip preceding zeros before counting
+	  digits in the mantissa. (ruby-bugs:PR#1181)
+
+Sun Sep 21 04:12:36 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_initialize): the argument
+	  should be a String.
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspres_initialize): ditt.
+
+	* ext/openssl/ossl_x509attr.c (ossl_x509attr_initialize): ditto.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_initialize): ditto.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_set_value): ditto.
+
+Sat Sep 20 11:49:05 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: typo fixed.
+
+	* test/logger/test_logger.rb: new file.
+
+Fri Sep 19 11:39:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* test/testunit/*: Added.
+
+	* lib/test/unit.rb: Documentation update.
+
+	* lib/test/unit/ui/console/testrunner.rb (TestRunner#initialize):
+	  Ditto.
+
+	* lib/test/unit.rb: Factored out an ObjectSpace collector.
+
+	* lib/test/unit/collector/objectspace.rb: Ditto.
+
+	* sample/testunit/*: Added.
+
+Fri Sep 19 01:00:48 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/log.rb (BasicLog#log): get rid of as ineffectual
+	  condition.
+
+	* lib/webrick/log.rb (BasicLog#format): add "\n" to message.
+
+Thu Sep 18 22:43:20 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): should push PROT_PCALL tag for orphans.
+
+	* eval.c (proc_invoke): should update "result" for orphans.
+
+Thu Sep 18 20:33:03 2003  Tietew  <tietew-ml-ruby-list@tietew.net>
+
+	* parse.y (str_xquote): do not prepend escapes in
+	  backqoute literals.  [ruby-list:38409]
+
+Thu Sep 18 20:30:17 2003  Tanaka Akira  <akr@m17n.org>
+
+	* lib/pathname.rb: update document.
+
+Thu Sep 18 15:27:05 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/logger.rb: new file.  Logger, formerly called devel-logger or
+	  Devel::Logger.
+
+	* sample/logger/*: new file.  samples of logger.rb.
+
+Wed Sep 17 23:41:45 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (localjump_destination): should not raise ThreadError
+	  exception for "break". [ruby-dev:21348]
+
+	* eval.c (proc_invoke): use result instead of prot_tag->retval.
+	  retval is no longer propagated to the ancestors.
+
+Wed Sep 17 20:34:00 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (tokadd_string, parse_string, yylex): escaped terminator
+	  is now interpreted as is.  [ruby-talk:82206]
+
+Wed Sep 17 18:52:36 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/fileutils/fileassertions.rb: new file.
+
+	* test/fileutils/test_fileutils.rb: new file.
+
+	* test/fileutils/test_nowrite.rb: new file.
+
+Wed Sep 17 18:51:02 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/strscan/test_stringscanner.rb: require test/unit.
+
+Wed Sep 17 18:35:34 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* test/strscan/test_stringscanner.rb: new file.
+
+Wed Sep 17 18:03:30 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl: all files are reviewed to simplify and avoid memory leak.
+
+	* ext/openssl/extconf.rb: add check for assert.h.
+
+	* ext/openssl/ossl.c (ossl_buf2str): new function to convert
+	  C buffer to String and free buffer.
+
+	* ext/openssl/ossl.c (ossl_x509_ary2sk): new function to convert
+	  Array of OpenSSL::X509 to STACK_OF(X509) with exception safe.
+
+	* ext/openssl/ossl.c (ossl_to_der, ossl_to_der_if_possible): new
+	  functions to convert object to DER string.
+
+	* ext/openssl/ossl.h: ditto.
+
+	* ext/openssl/ossl_bio.c (ossl_membio2str): new function to convert
+	  BIO to String object and free BIO.
+
+	* ext/openssl/ossl_bio.h: ditto.
+
+	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_to_der): add for "to_der".
+
+	* ext/openssl/ossl_x509name.c (ossl_x509name_to_der): ditto.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_to_der): ditto.
+
+	* ext/openssl/ossl_x509ext.c (create_ext_from_array): removed
+	  and reimplement in openssl/x509.rb.
+
+	* ext/openssl/ossl_x509attr.c: reimplemented and disable some
+	  method temporarily. this class doesn't work fine without ASN.1
+	  data support;-) I'll rewrite in near future.
+
+	* ext/openssl/lib/openssl/x509.c (X509::Attribute): get rid off
+	  unused code.
+
+	* ext/openssl/lib/openssl/x509.c (X509::ExtensionFactory): refine all.
+
+Tue Sep 16 22:25:06 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/csv/test_csv.rb: add negative tests of row_sep.
+
+Tue Sep 16 18:02:36 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* regex.c (re_compile_pattern): should not translate character
+	  class range edge. [ruby-list:38393]
+
+Tue Sep 16 16:47:56 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* MANIFEST: add test/csv/mac.csv.
+
+	* win32/Makefile.sub, bcc32/Makefile.sub (test): add phony NUL target.
+
+Mon Sep 15 19:02:52 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/csv.rb: add extra pamameter to specify row(record) separater
+	  character.  To parse Mac's CR separated CSV, do like this.
+	    CSV.open("mac.csv", "r", ?,, ?\r) { |row| p row.to_a }
+	  The 3rd parameter in this example ?, is for column separater and the
+	  4th ?\r is for row separater.  Row separater is nil by default.  Nil
+	  separater means "\r\n" or "\n".
+
+	* test/csv/test_csv.rb: add tests for above feature.
+
+	* test/csv/mac.csv: added.  Sample CR separated CSV file.
+
+Fri Sep 12 22:41:48 2003  Michal Rokos  <m.rokos@sh.cvut.cz>
+
+	* ext/openssl/ossl.c: move ASN.1 stuff to ossl_asn1.[ch]
+
+	* ext/openssl/ossl.c: move BIO stuff to ossl_bio.[ch]
+
+	* ext/openssl/ossl_asn1.[ch]: new files
+
+	* ext/openssl/ossl_bio.[ch]: new files
+
+Fri Sep 12 12:30:41 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* intern.h (rb_disable_super, rb_enable_super): replace with dummy
+	  expressions instead of prototypes.  the functions remain yet for
+	  binary compatibility.  [ruby-talk:81758]
+
+Fri Sep 12 12:09:54 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* bignum.c (rb_big_and): convert argument using 'to_int'.
+
+	* bignum.c (rb_big_or): ditto.
+
+	* bignum.c (rb_big_xor): ditto.
+
+Fri Sep 12 07:06:14 2003  David Black  <dblack@superlink.net>
+
+	* lib/scanf.rb: Took out useless @matched_item variable; some small
+	  refactoring.
+
+Thu Sep 11 08:43:44 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_require): allow "require" on $SAFE>0, if feature
+	  name is not tainted.
+
+	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser::stream):
+	  Supports StringIO.
+
+Wed Sep 10 22:47:30 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.h: add a workaround for win32 platform.
+	  libeay32.dll doesn't export functions defined in conf_api.h.
+
+	* ext/openssl/ossl_config.c (ossl_config_initialize): ditto.
+
+	* ext/openssl/ossl_config.c (ossl_config_add_value): ditto.
+
+	* ext/openssl/ossl_config.c (set_conf_section_i): should check
+	  if the argument is Array.
+
+Wed Sep 10 22:41:54 2003  Tietew  <tietew@tietew.net>
+
+	* eval.c (win32_get_exception_list): avoid VC7 warning.
+	  [ruby-win32:577]
+
+Tue Sep  9 10:39:51 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (struct tag): dst should be VALUE.
+
+	* eval.c (localjump_destination): stop at the scope where the current
+	  block was created.  [ruby-dev:21353]
+
+Tue Sep  9 05:17:04 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_config.rb: avoid compile error in OpenSSL-0.9.6.
+
+Tue Sep  9 02:41:35 2003  Michal Rokos  <m.rokos@sh.cvut.cz>
+
+	* ext/openssl/ossl_config.c: Refine compatibility.
+
+Tue Sep  9 01:50:45 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httpserver.rb (HTTPServer#access_log): add "\n" to
+	  the message.
+
+	* lib/webrick/log.rb (BasicLog#log): add "\n" only if needed.
+
+Mon Sep  8 22:15:33 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tk/lib/multi-tk.rb: modify security check at creating
+	  a new interpreter
+
+Mon Sep  8 20:00:12 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb, lib/optparse/version.rb: search also all
+	  capital versions.
+
+Mon Sep  8 19:26:33 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl.h: include openssl/conf.h and openssl/conf_api.h.
+
+	* ext/openssl/ossl_config.c: refine all with backward compatibility.
+
+	* ext/openssl/ossl_config.h: export GetConfigPtr() and DupConfigPtr().
+
+	* ext/openssl/ossl_x509.c: added new constants under X509 module.
+	  DEFAULT_CERT_AREA, DEFAULT_CERT_DIR, DEFAULT_CERT_FILE,
+	  DEFAULT_CERT_DIR_ENV, DEFAULT_CERT_FILE_ENV and DEFAULT_PRIVATE_DIR.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509extfactory_free): don't free
+	  the members of the struct. it's left to GC.
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509_set_config): add for config=.
+
+	* ext/openssl/ossl_x509ext.c (Xossl_x509extfactory_initialize):
+	  add attr readers: issuer_certificate, subject_certificate,
+	  subject_request, crl and config.
+
+Mon Sep  8 18:26:41 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/accesslog.rb (AccessLog::setup_params): use req.port
+	  instead of config[:Port] or req.request_uri.port.
+
+	* lib/webrick/httprequest.rb (HTTPRequest#meta_vars): ditto.
+
+	* lib/webrick/httpservlet/filehandler.rb (FileHandler#dir_list): ditto.
+
+	* lib/webrick/config.rb: :Listen option never be used.
+
+	* lib/webrick/server.rb (GenericServer#initialize): don't use :Listen
+	  option and add warning message.
+
+	* lib/webrick/log.rb (BasicLog#<<): shortcut of log(INFO, ...).
+
+	* lib/webrick/httpserver.rb (HTTPServer#accesslog): use << for logging.
+
+Sun Sep  7 16:08:28 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (lib_mainloop_core): fixed signal-trap bug
+
+	* ext/tk/lib/*.rb: Ruby/Tk works at $SAFE == 4
+
+Sat Sep  6 02:26:34 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_*.rb: assert_same, assert_match, and so on.
+
+Sat Sep  6 18:45:46 2003  Mauricio Fernandez  <batsman.geo@yahoo.com>
+
+	* parse.y (assignable): call rb_compile_error(), not rb_bug().
+	  [ruby-core:01523]
+
+Sat Sep  6 17:40:41 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ruby_missing.c: rid of unnecessary backward
+	  compatibility stuff. and remove DEFINE_ALLOC_WRAPPER from
+	  all sources.
+
+	* ext/openssl/ossl_x509ext.c (X509::Extension.new): new method.
+
+	* ext/openssl/ossl_x509ext.c (X509::Extension#oid=): new method.
+
+	* ext/openssl/ossl_x509ext.c (X509::Extension#value=): new method.
+
+	* ext/openssl/ossl_x509ext.c (X509::Extension#critical=): new method.
+
+Sat Sep  6 01:23:22 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (CreateChild): need to quote cmd if RUBYSHELL is set.
+
+	* win32/win32.c (CreateChild): fix condition about whether to call
+	  shell or not.
+
+Sat Sep  6 00:36:20 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
+
+	* Makefile.in (test): phony target.
+
+	* lib/mkmf.rb (have_library, find_library): configure by library
+	  name.
+
+	* lib/optparse.rb (OptionParser#order, #permute, #parse): allow an
+	  array as argument.
+
+	* test/ruby/test_*.rb: moved invariants to left side in
+	  assert_equal, and use assert_nil, assert_raises and so on.
+
+	* win32/win32.c (isInternalCmd): distinguish command.com and
+	  cmd.exe.
+
+	* win32/win32.c (make_cmdvector): a character just after wildcard
+	  was ignored.  [ruby-core:01518]
+
+Fri Sep  5 20:27:08 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_*.rb: replace 'assert(a == b)' with assert_equal(a, b)'
+
+Fri Sep  5 18:00:51 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/x509.rb: new method X509::Name::parse.
+
+	* ext/openssl/ossl_digest.c: add ossl_digest_new().
+
+	* ext/openssl/ossl_digest.h: ditto.
+
+	* ext/openssl/ossl_cipher.c: add ossl_cipher_new().
+
+	* ext/openssl/ossl_cipher.h: ditto.
+
+Fri Sep  5 15:32:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-font-lock-maybe-here-docs): should not
+	  search delimiter forward if found in backward.
+
+Fri Sep  5 13:32:48 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/runner.rb: arguments should be keys.
+
+Fri Sep  5 12:09:55 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* test/ruby/test_system.rb (test_system): check existence of ruby
+	  interpreter.
+
+Fri Sep  5 11:32:17 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (--version): fix assignment/reference order.
+
+	* lib/optparse.rb (OptionParser#help): new; OptionParser#to_s may
+	  be deprecated in future.
+
+	* lib/optparse/version.rb (OptionParser#show_version): hide Object.
+
+	* test/runner.rb: fix optparse usage.
+
+	* test/runner.rb: glob all testsuits if no tests given.
+
+Fri Sep  5 10:42:58 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/runner.rb: added.  gets testcases from command line and runs it.
+
+	* test/ruby/test_gc.rb: remove useless part which was for dumping test
+	  result.
+
+Fri Sep  5 09:28:59 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby/test_gc.rb: added.  splitter.rb which I made to split
+	  sample/test.rb into test/ruby/test_* kindly removed GC test (the
+	  last section in the original test) to reduce things to be worried.
+
+Fri Sep  5 03:00:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_iterator.rb (test_block_in_arg): add no block
+	  given tests.
+
+	* test/ruby/test_iterator.rb (test_ljump): uncomment LocalJumpError
+	  test.
+
+Fri Sep  5 01:10:11 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/ruby: tests for ruby itself.
+
+	* test/ruby/test_*.rb: split sample/test.rb into 28 test/unit testcases.
+	  some tests could not be translates...  search '!!' mark to see it.
+
+	* test/csv/test_csv.rb: should require 'csv', not '../lib/csv'.  test
+	  runner should set load path correctly.
+
+Fri Sep  5 01:03:59 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/csv/test_csv.rb: close opened files for CSV::IOBuf explicitly.
+	  opened file cannot be removed under win32 box.
+
+Thu Sep  4 23:59:40 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (tokadd_string): newlines have no special meanings in
+	  %w/%W, otherwise they are ignored only when interpolation is
+	  enabled.  [ruby-dev:21325]
+
+Thu Sep  4 19:38:25 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* ext/io/wait/.cvsignore: added.
+
+	* ext/openssl/.cvsignore: added.
+
+Thu Sep  4 19:28:24 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* sample/openssl: added.  Sample of standard distribution library
+	  should be locate in sample/{module_name}/*.
+
+	* ext/openssl/sample/*: removed.  move to sample/openssl/*.
+
+Thu Sep  4 18:02:15 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/csv/test_csv.rb: use remove_const to reduce warnings.  use
+	  Dir.tmpdir to locate working files.
+
+Thu Sep  4 17:41:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-here-doc-beg-re): underscore also is
+	  valid delimiter.
+
+	* misc/ruby-mode.el (ruby-here-doc-end-match): must quote
+	  arbitrary string to use as regexp.
+
+	* misc/ruby-mode.el (ruby-font-lock-maybe-here-docs): must not
+	  call `ruby-here-doc-end-match' unless `ruby-here-doc-beg-re'
+	  matched.
+
+Thu Sep  4 15:40:07 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test/csv/test_csv.rb: run on test/unit original layer.
+
+Thu Sep  4 12:54:50 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/token.c: headerless documents with root-level spacing now
+	  honored.
+
+Thu Sep  4 00:06:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (mark_frame_adj): need to adjust argv pointer if using
+	  system's alloca. [ruby-core:01503]
+
+Wed Sep  3 21:33:20 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* test: add test directory.  Test::Unit aware testcases and needed
+	  files should be located in this directory.  dir/file name convention;
+	    test/{module_name}/test_{testcase_name}.rb
+	    test/{module_name}/{needed_files}
+	  someday, someone will write testrunner which searches test_*.rb and
+	  run testcases automatically.
+
+	* test/csv/*: add testcase for lib/csv.rb.
+
+Wed Sep  3 01:37:09 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (rb_f_gets): should call next_argv() before type check
+	  current_file. [ruby-list:38336]
+
+Tue Sep  2 20:37:15 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/net/protocols.rb (SSLIO#ssl_connect): warning
+	  for skipping server verification.
+
+Tue Sep  2 23:36:57 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (proc_invoke): should retrieve retval when pcall is true.
+
+Tue Sep  2 14:09:20 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ext/socket/extconf.rb: check s6_addr8 in in6_addr (Tru64 UNIX).
+	  the patch is submitted by nmu <nmu@users.sourceforge.jp>.
+
+	* ext/socket/getaddrinfo.c (getaddrinfo): should use in6_addr8 on
+	  some platforms.
+
+	* ext/socket/getnameinfo.c (getnameinfo): ditto.
+
+Tue Sep  2 14:02:19 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* ext/tcltklib/tcltklib.c (ip_invoke): fixed bug on passing a exception
+
+	* ext/tk/lib/{tk.rb, tkcanvas.rb, tkfont.rb, tktext.rb}:
+	  bug fix and improvement of font control
+
+Tue Sep  2 09:51:36 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): should not handle exceptions within rescue
+	  argument.  [ruby-talk:80804]
+
+Tue Sep  2 00:44:37 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_memsearch): fix overrun.  [ruby-talk:80759]
+
+Tue Sep  2 00:41:27 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv/iconv.c (map_charset): use lower case keys.
+
+	* ext/iconv/iconv.c (iconv_fail): just yield error and return the
+	  result if a block is given.
+
+	* ext/iconv/iconv.c (iconv_convert): yield error and append the
+	  result if a block is given.
+
+	* ext/iconv/charset_alias.rb (charset_alias): optional third
+	  argument.
+
+	* ext/iconv/charset_alias.rb (charset_alias): use CP932 instead of
+	  SHIFT_JIS on cygwin.
+
+Mon Sep  1 18:34:25 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval): make tail recursion in ELSE clause of
+	  RESCUE a jump.
+
+Mon Sep  1 18:00:02 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (aref_args): forgot to call NEW_SPLAT(). reported by
+	  Dave Butcher.
+
+	* eval.c (Init_Thread): protect thgroup_default.  suggested by Guy
+	  Decoux in [ruby-talk:80623]
+
+Mon Sep  1 16:59:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_thread_switch): add RESTORE_EXIT; exit by another
+	  thread termination.
+
+	* eval.c (rb_thread_start_0): should not error_print() within
+	  terminated thread, because $stderr used by it might be
+	  overriden now.  [ruby-dev:21280]
+
+Sun Aug 31 22:46:55 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* eval.c (TAG_DST()): take no argument.
+
+	* process.c (p_gid_sw_ensure): return VALUE.
+
+Sun Aug 31 22:27:10 2003  Hidetoshi NAGAI  <nagai@dumbo.ai.kyutech.ac.jp>
+
+	* process.c (p_gid_sw_ensure): lack of function type
+
+Sun Aug 31 12:25:06 2003  Nobuyoshi Nakada  <nobu.nokada@softhome.net>
+
+	* lib/optparse.rb: --version takes an optional argument; "all" or
+	  a list of package names.
+
+Sun Aug 31 10:17:02 2003  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/date/format.rb: yyyy/mm is not an acceptable format.
+
+	* lib/time.rb: follow above.
+
+Sat Aug 30 14:25:43 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_iter_break): should not call TAG_JUMP directly.
+
+Sat Aug 30 03:58:21 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (struct BLOCK): remove BLOCKTAG, use scope instead.
+
+	* eval.c (POP_TAG): no longer propagate retval.  retval is now set
+	  directly by localjump_destination().
+
+	* eval.c (localjump_destination): new function to cast
+	  return/break local jump.
+
+	* eval.c (rb_yield_0): stop TAG_RETURN/TAG_BREAK escaping.
+
+Fri Aug 29 22:35:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* bigdecimal.c *.html: The 2nd arg. for add,sub,mult, and div is 0,
+	  then result will be the same as +,-,*,/ respectively.
+
+Fri Aug 29 17:30:15 2003  Hidetoshi NAGAI  <nagai@ai.kyutech.ac.jp>
+
+	* process.c: bug fix
+
+	* process.c: add rb_secure(2) to methods of Process::{UID,GID,Sys}
+
+	* process.c: deny handling IDs during evaluating the block given to
+	  the Process::{UID,GID}.switch method
+
+	* ext/tcltklib/tcltklib.c: some methods have no effect if on slave-IP
+
+	* ext/tcltklib/tcltklib.c: can create a interpreter without Tk
+
+	* ext/tcltklib/tcltklib.c: bug fix on handling exceptions
+
+	* ext/tcltklib/MANUAL.euc: modify
+
+	* ext/tk/lib/tk.rb: freeze some core modules
+
+	* ext/tk/lib/multi-tk.rb: more secure
+
+	* ext/tk/lib/tk.rb: TkVariable.new(array) --> treat the array as the
+	  Tk's list
+
+	* ext/tk/lib/tk.rb: improve accessibility of TkVariable object
+
+	* ext/tk/lib/tk.rb, ext/tk/lib/tkfont.rb, ext/tk/lib/tkcanvas.rb,
+	  ext/tk/lib/tktext.rb: fix bug of font handling
+
+	* ext/tk/lib/tkfont.rb: TkFont.new() accepts compound fonts
+
+Thu Aug 28 22:07:12 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_autoload_load): call const_missing if autoloading
+	  constant is not defined to allow hook.
+
+	* eval.c (rb_eval): use rb_const_get_from() instead of
+	  rb_const_get_at().
+
+	* eval.c (is_defined): forgot to check NODE_COLON3.
+
+Thu Aug 28 17:30:24 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* variable.c (rb_const_get_0): should check constants defined in
+	  included modules, if klass is Object. [ruby-talk:79302]
+
+	* numeric.c (check_uint): check should be done using UINT_MAX, not
+	  INT_MAX. this fix is submitted by Lyle Johnson
+	  <lyle@knology.net> in [ruby-core:01486]
+
+Thu Aug 28 05:02:52 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (singleton): typo fixed (ruby-bugs-ja:PR#562)
+
+Thu Aug 28 02:37:45 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_eval): *a = [1,2] now assigns [[1,2]] to a.
+	  consistent with *a = [1], which set [[1]] to a.
+
+	* node.h: merge NODE_RESTARY to NODE_SPLAT.
+
+	* parse.y: rules simplified a bit by removing NODE_RESTARY.
+
+	* sample/test.rb: updated for new assignment behavior.
+
+Wed Aug 27 22:33:24 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_bug): should not use other methods; this function is
+	  not for ordinary use.  [ruby-dev:21259]
+
+Wed Aug 27 15:07:57 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/smtp.rb (check_response): AUTH CRAM-MD5 returns 334
+	  response. [ruby-list:38279]
+
+Wed Aug 27 05:10:15 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (map_errno): support winsock error.
+
+	* win32/win32.c (pipe_exec, CreateChild, poll_child_status, waitpid,
+	  kill, link, rb_w32_rename, unixtime_to_filetime, rb_w32_utime):
+	  pass errno to map_errno().
+
+	* win32/win32.c (rb_w32_select, rb_w32_accept, rb_w32_bind,
+	  rb_w32_connect, rb_w32_getpeername, rb_w32_getsockname,
+	  rb_w32_getsockopt, rb_w32_ioctlsocket, rb_w32_listen, rb_w32_recv,
+	  rb_w32_recvfrom, rb_w32_send, rb_w32_sendto, rb_w32_setsockopt,
+	  rb_w32_shutdown, rb_w32_socket, rb_w32_gethostbyaddr,
+	  rb_w32_gethostbyname, rb_w32_gethostname, rb_w32_getprotobyname,
+	  rb_w32_getprotobynumber, rb_w32_getservbyname, rb_w32_getservbyport,
+	  rb_w32_fclose, rb_w32_close): use map_errno().
+
+	* win32/win32.h: add winsock errors.
+
+Tue Aug 26 23:53:23 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* lib/ostruct.rb (OpenStruct::method_missing): prohibit modifying
+	  frozen OpenStruct. [ruby-talk:80214]
+
+Tue Aug 26 20:03:50 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_tmpsrc): add the hook for source.
+	  [ruby-list:38122]
+
+Tue Aug 26 15:59:53 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* implicit.c (syck_type_id_to_taguri): corrected detection of
+	  x-private types.
+
+Sun Aug 24 01:02:48 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): performance improvement.
+	  [ruby-talk:79748]
+
+Sat Aug 23 23:41:16 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* file.c (rb_file_s_expand_path): avoid calling rb_scan_args() for
+	  apparent cases. [ruby-talk:79748]
+
+Sat Aug 23 18:56:53 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_putchar): should use rb_str_resize() to just
+	  resize a string, rb_str_cat() disallows NULL.  [ruby-dev:21237]
+
+Sat Aug 23 16:48:41 2003  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ruby-lex.rb: bug fix for "foo" !~ /bar/. [ruby-talk:79942]
+
+Sat Aug 23 15:59:58 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_eval, rb_iterate, block_pass): reduce PUSH/POP_TAG and
+	  EXEC_TAG() for retry.  [ruby-dev:21216]
+
+Sat Aug 23 02:32:33 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_yield_splat): should check if "values" is array.
+
+	* enum.c (each_with_index_i): typo.
+
+Fri Aug 22 17:07:05 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enum.c (inject_i): use rb_yield_values.
+
+	* enum.c (each_with_index_i): ditto.
+
+	* eval.c (rb_yield_splat): new function to call "yield *values".
+
+	* string.c (rb_str_scan): use rb_yield_splat().
+
+Fri Aug 22 06:13:22 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/rubyext.c: refactoring of the transfer method
+	  dispatch.  added yaml_org_handler for faster dispatch of
+	  transfers to base types.
+
+	* lib/yaml/rubytypes.rb: removed handling of builtins from
+	  Ruby library.
+
+	* ext/syck/token.c: quoted and block scalars are now implicit !str
+
+	* ext/syck/implicit.c: empty string detected as !null.
+
+Fri Aug 22 01:00:31 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (block_pass): improve passing current block.
+
+Fri Aug 22 00:13:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: Int. overflow bug in multiplication
+	  fixed, and VpNmlz() speed up.
+
+Wed Aug 20 16:44:49 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (ruby_connect): many systems seem to have
+	  a problem in select() after EINPROGRESS.  [ruby-list:38080]
+
+Wed Aug 20 01:31:17 2003  why the lucky stiff  <why@ruby-lang.org>
+
+	* ext/syck/syck.h: Parser definition problems on HP-UX.
+	  [ruby-talk:79389]
+
+	* ext/syck/handler.c (syck_hdlr_get_anchor): Memory leak.
+
+	* ext/syck/syck.s (syck_io_file_read): Bad arguments to fread.
+
+	* ext/syck/rubyext.c: Tainting issues.
+
+Tue Aug 19 23:20:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c .h .html: to_s("+") implemented.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb: E implemented.
+
+Tue Aug 19 11:19:33 2003  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* io.c (next_argv): should not call GetOpenFile() if rb_stdout is
+	  not a IO (T_FILE).
+
+Tue Aug 19 07:47:10 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/ssl.rb: new file; SSL/TLS enhancement for GenericServer.
+
+	* lib/webrick/https.rb: SSLSocket handling is moved to webrick/ssl.rb.
+
+	* lib/webrick/compat.rb (File::fnmatch): remove old migration code.
+
+	* lib/webrick/httpserver.rb (HTTPServer#run): ditto.
+
+	* lib/webrick/server.rb (GenericServer#listen): the body of this
+	  method is pull out as Utils::create_lisnteners.
+
+	* lib/webrick/utils.rb (Utils::create_lisnteners): new method.
+
+	* lib/webrick/server.rb (GenericServer#start): should rescue
+	  unknown errors. and refine comments.
+
+	* ext/openssl/lib/openssl/ssl.rb (SSLServer#accept): should close
+	  socket if SSLSocket raises error.
+
+Tue Aug 19 07:47:09 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/ossl_ssl.c: sync_close is moved to SSLSocket as
+	  a builtin.
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#close): ditto.
+
+	* ext/openssl/lib/openssl/buffering.rb (Buffering#puts): should
+	  add a return to the tails of each line.
+
+	* ext/openssl/lib/openssl/ssl.rb: new class OpenSSL::SSL::SSLServer.
+
+	* ext/openssl/lib/net/protocols.rb (SSLIO#ssl_connect): use sync_close.
+
+	* ext/openssl/sample/echo_svr.rb: use SSLServer.
+
+	* ext/openssl/sample/echo_cli.rb: add example of SSLSocket#sync_close.
+
+Tue Aug 19 01:24:34 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/curses/curses.c (_XOPEN_SOURCE_EXTENDED): Mac OS X standard
+	  headers are inconsistent at this macro.  [ruby-core:01432]
+
+	* ext/curses/extconf.rb: check if _XOPEN_SOURCE_EXTENDED breaks.
+
+	* ext/tcltklib/stubs.c: Status macro in X11/Xthreads.h bothers
+	  winspool.h
+
+	* instruby.rb: make list at first instead of iterator.
+	  [ruby-talk:79347]
+
+Mon Aug 18 11:23:11 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): preserve raw order for **.
+
+Sun Aug 17 23:39:55 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/extconf.rb (HAVE_VA_ARGS_MACRO): need to compile.
+
+Sun Aug 17 17:10:03 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (SSLSocket#sync_close=): add a
+	  method to specify if the underlying IO will be closed in
+	  SSLSocket#close.
+
+	* ext/openssl/lib/openssl/buffering.rb: add forwarders to
+	  setsockopt, getsockopt and fcntl.
+
+	* ext/openssl/lib/net/protocols.rb: enable sync for SSLSocket.
+
+Sun Aug 17 11:32:04 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): should not force to remake Makefile when
+	  installation and so on.
+
+Sat Aug 16 23:58:18 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_symbol, w_object): get rid of warnings.
+
+	* re.c (rb_memsearch): ditto.
+
+	* time.c (time_dump): ditto.
+
+	* ext/extmk.rb (extmake): not continue making when extconf.rb
+	  failed.
+
+	* ext/openssl/extconf.rb: check __VA_ARGS__ macro more precisely.
+
+	* ext/openssl/ossl.h: remove version.h dependency.
+
+	* ext/openssl/ruby_missing.h: ditto.
+
+	* lib/mkmf.rb (pkg_config): use --libs output except with
+	  only-L for other options.  [ruby-list:38099]
+
+	* lib/mkmf.rb (create_makefile): separate rule for static
+	  library from shared object.
+
+	* win32/Makefile.sub, bcc32/Makefile.sub, wince/Makefile.sub:
+	  define exec_prefix and libdir.
+
+Fri Aug 15 23:15:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c .h: Bug in combination of limit & div
+	  method fixed.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb: atan() & sqrt() added.
+
+Fri Aug 15 12:01:44 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (HUGE_ST_INO): check whether struct stat.st_ino
+	  is larger than long.  [ruby-dev:21194]
+	  http://www.geocities.co.jp/SiliconValley-PaloAlto/1409/ruby/beos.html
+
+	* error.c (syserr_eqq): errno might exceed Fixnum limit.
+
+	* error.c (Init_Exception): moved base initialization from
+	  init_syserr().
+
+	* inits.c (rb_call_inits): postpone initializing errnos until
+	  Bignum is available.
+
+Fri Aug 15 12:01:43 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/curses/curses.c (_XOPEN_SOURCE_EXTENDED): needed to let
+	  keyname() and so on be declared.
+
+	* ext/curses/curses.c (curses_resizeterm, window_resize):
+	  arguments conflicted with macros in term.h.
+
+	* ext/curses/curses.c (Curses module methods): ensure
+	  initialized.  [ruby-dev:21191]
+
+Fri Aug 15 02:08:53 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (id2ref): recycle check should be done by klass == 0.
+	  [ruby-core:01408]
+
+Fri Aug 15 01:34:23 2003  Michal Rokos  <m.rokos@sh.cvut.cz>
+
+	* ext/openssl/ossl_pkey.c: move generate_cb here
+
+	* ext/openssl/ossl_pkey_{dh|dsa|rsa}.c: adapt to this cb
+
+	* ext/openssl/openssl_missing.[ch]: add (0.9.6x, x<j) missing BN funcs
+
+	* ext/openssl/ossl_bn.c: use supplied funcs from openssl_missing.c
+
+Fri Aug 15 00:38:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: Bug in div method fixed.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb: Newly added.
+
+	* ext/bigdecimal/sample/pi.rb: Changed so as to use math.rb.
+
+Thu Aug 14 21:19:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (Init_Thread): Continuation#[] added.  [ruby-talk:79028]
+
+Thu Aug 14 20:03:34 2003  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (OLE_FREE): should not call
+	  ole_message_loop.
+
+	* ext/win32ole/win32ole.c (ole_event_free): ditto.
+
+	* ext/win32ole/win32ole.c (ole_initialize): stop calling
+	  OleUninitialize at exit.
+
+Thu Aug 14 11:27:37 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (rb_data_object_alloc): check type of 1st argument.
+	  [ruby-dev:21192]
+
+Thu Aug 14 00:21:14 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (mlhs_node): should allow "::Foo" (colon3) as lhs.
+
+	* parse.y (lhs): ditto.
+
+	* parse.y (yylex): should return tCOLON3 right after kCLASS.
+	  [ruby-talk:78918]
+
+	* error.c (exc_initialize): was converting argument to string too
+	  eagerly.  Only check was needed. [ruby-talk:78958]
+
+Wed Aug 13 23:31:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c .h .html: Ambiguity of
+	  BigDecimal::limit removed.
+
+Wed Aug 13 19:21:34 2003  Christian Neukirchen  <chneukirchen@yahoo.de>
+
+	* lib/webrick/https.rb (HTTPServer#run): should set syncing-mode
+	  to SSLSocket. [ruby-talk:78919]
+
+Wed Aug 13 18:13:49 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (POP_BLOCK): turn on BLOCK_LEFT flag when leaving block.
+
+	* eval.c (proc_invoke): unpack return/break destination when block
+	  is already left.
+
+Wed Aug 13 15:58:31 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* object.c (rb_class_s_alloc): add function prototype to avoid VC++
+	  warning.
+
+Wed Aug 13 13:50:59 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/Win32API/Win32API.c (Win32API_initialize): should pass some
+	  class to first argument of Data_Wrap_Struct(). (ruby-bugs:PR#1109)
+
+Tue Aug 12 16:55:11 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in: static link libraries to LIBRUBY_SO with static linked
+	  ext.  [ruby-dev:21157]
+
+	* ext/extmk.rb (extmake): sort extension library initialization order.
+
+	* ext/extmk.rb (extmake): compact $extlibs.
+
+Tue Aug 12 02:48:56 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (THREAD_SAVE_CONTEXT): should explicitly turn off the
+	  flag before calling getcontext(2).
+
+	* eval.c (struct thread): add member to save backing store on
+	  IA64. (ruby-bugs PR1086)
+
+	* eval.c (thread_mark): mark IA64 backing store region.
+
+	* eval.c (thread_free): free saved IA64 backing store.
+
+	* eval.c (rb_thread_save_context): save IA64 backing store as well.
+
+	* eval.c (rb_thread_restore_context): restore IA64 backing store.
+
+	* eval.c (THREAD_ALLOC): initialize IA64 members.
+
+Mon Aug 11 22:31:50 2003  NAKAMURA, Hiroshi  <nahi@ruby-lang.org>
+
+	* lib/debug.rb(debug_command): inspection command should inspect
+	  resulting value even if it's nil.  [ruby-dev:21180] by OMAE, jun
+	  <jun66j5@ybb.ne.jp>.
+
+	* lib/debug.rb(debug_command): incomplete regexp.
+
+Mon Aug 11 17:33:07 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call_super): do not use rb_block_given_p() for
+	  check. [ruby-talk:78656]
+
+	* eval.c (BEGIN_CALLARGS): push ITER_NOT only when ITER_PRE.
+
+Sun Aug 10 10:43:05 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* ext/openssl/lib/openssl/buffering.rb: increase BLOCK_SIZE
+	  from 1k to 16k bytes. [ruby-talk:78603]
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_s_alloc): enable
+	  partial write to allow interruption in SSLSocket#write.
+
+Sun Aug 10 00:34:16 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* cygwin/GNUmakefile: remove unnecessary '--drive-name=$(CC)'
+	  for ccache.
+
+Sat Aug  9 10:36:21 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* marshal.c (w_object): do not dump generic instance variable when
+	  marshal_dump is defined.
+
+Sat Aug  9 00:35:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal.c: F style output(like 1234.56789) implemented
+	  to to_s method.
+	* ext/bigdecimal_??.html: F style output(like 1234.56789)
+	  implemented to to_s method.
+
+Fri Aug  8 12:33:17 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* bcc32/Makefile.sub: rubyw.exe should be a Windows GUI program.
+	  add the -aa option to WLDFLAGS.
+
+Fri Aug  8 11:29:26 2003  Koji Arai  <jca02266@nifty.ne.jp>
+
+	* marshal.c (w_object): should set `c_arg' at first.
+
+Fri Aug  8 03:22:28 2003  GOTOU Yuuzou  <gotoyuzo@notwork.org>
+
+	* lib/webrick/httputils.rb (FormData#list): should not take
+	  a side effect for the receiver.
+
+Thu Aug  7 14:40:37 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* cygwin/GNUmakefile: better --disbale-shared option support.
+
+	* cygwin/GNUmakefile: add forwarding DLL target for cygwin.
+
+Thu Aug  7 14:21:05 2003  Corinna Vinschen  <vinschen@redhat.com>
+
+	* configure.in: Fix Cygwin specific naming of libraries to
+	  be net distribution compliant. (ruby-bugs:PR#1077)
+	  cygwin-ruby18.dll -> cygruby18.dll
+
+Thu Aug  7 12:51:38 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_at_exit): should not be called without a block.
+	  block_given check added.
+
+Thu Aug  7 06:46:06 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_call0): forgot to pop ruby_class.
+
+	* eval.c (rb_call0): update ruby_class as well as ruby_cref.
+	  (ruby-bugs-ja:PR#540)
+
+Thu Aug  7 04:52:50 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_yield_0): remove ruby_frame->cbase and unify to
+	  ruby_cref.  [ruby-talk:78141]
+
+Thu Aug  7 04:19:15 2003  Akinori MUSHA  <knu@iDaemons.org>
+
+	* gc.c: FreeBSD/ia64's mcontext_t is a bit different from that of
+	  Linux/ia64.  This makes gc.c compile but miniruby coredumps for
+	  the moment.
+
+Thu Aug  7 00:15:00 2003  Shigeo Kobayashi  <shigek@ruby-lang.org>
+
+	* ext/bigdecimal.c: Comparison results adjusted to Float's.
+	* ext/bigdecimal.c: Use rb_num_coerce_????(x,y) instead of own.
+
+Wed Aug  6 22:58:00 2003  Nathaniel Talbott  <ntalbott@ruby-lang.org>
+
+	* lib/test/unit/testcase.rb: Added equality checking.
+	* lib/test/unit/testsuite.rb: Added equality checking.
+	* lib/test/unit/assertions.rb: Fixed a warning.
+
+Wed Aug  6 17:28:10 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): pass LIBPATH to make ruby.  [ruby-dev:21137]
+
+	* ext/extmk.rb (extmake): set library name as source file name in
+	  Init_ext().  [ruby-dev:21137]
+
+	* lib/mkmf.rb (Logging::postpone): postpone logging messages after
+	  heading message as the result of the block.
+
+	* lib/mkmf.rb (macro_defined?): append newline to src unless ended
+	  with it.
+
+	* lib/mkmf.rb (have_library): treat nil function name as "main".
+	  (ruby-bugs:PR#1083)
+
+	* lib/mkmf.rb (pkg_config): should append additional libraries to
+	  $libs but not $LIBS.  [ruby-dev:21137]
+
+	* ext/io/wait/extconf.rb: check DOSISH macro instead of platform.
+
+	* ext/digest/sha1/extconf.rb: have_library already appends library
+	  name.
+
+Wed Aug  6 17:23:57 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c: initialize /* OK */ variables by Qnil to stop warnings.
+
+Wed Aug  6 04:58:32 2003  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/Setup*: add io/wait and openssl.
+
+Wed Aug  6 01:13:38 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* eval.c (rb_f_autoload): use ruby_cbase instead of ruby_class.
+
+	* eval.c (rb_f_autoload_p): ditto.
+
+	* class.c (rb_mod_init_copy): no longer implements independent
+	  clone and dup methods.  override "initialize_copy" instead.
+	  [ruby-core:01352]
+
+	* object.c (rb_class_s_alloc): define Class allocation function.
+	  this makes Classes to follow clone framework that uses
+	  initialize_copy.
+
+	* object.c (rb_class_initialize): separate instantiation and
+	  initialization.
+
+	* object.c (rb_obj_alloc): prohibit instantiation from
+	  uninitialized class.
+
+	* object.c (rb_class_superclass): check uninitialized class.
+
+	* array.c (rb_ary_fill): wrong index processing with block.  this
+	  fix was done by Koji Arai <JCA02266@nifty.ne.jp> [ruby-list:38029]
+
+	* marshal.c (w_object): should preserve generic ivar for nil,
+	  true, false, symbols, and fixnums.
+
+	* marshal.c (w_uclass): base_klass check should be done after
+	  rb_class_real().
+
+Wed Aug  6 01:18:50 2003  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/net/http.rb: update document.
+
+	* lib/net/pop.rb: ditto.
+
+	* lib/net/protocol.rb: ditto.
+
+Wed Aug  6 00:48:37 2003  Koji Arai  <jca02266@nifty.ne.jp>
+
+	* marshal.c (w_object): should recommend marshal_dump rather than
+	  _dump_data.
+
+Tue Aug  5 17:58:57 2003  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* lib/fileutils.rb (install): should preserve timestamp only.
+
+Tue Aug  5 17:31:59 2003  Ian Macdonald  <ian@caliban.org>
+
+	* lib/shell/command-processor.rb (Shell::CommandProcessor::rmdir):
+	  simple typo.
+
+Tue Aug  5 15:47:34 2003  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_load): should preserve current source file/line.
+
+Tue Aug  5 10:04:42 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* string.c (str_new4): ptr may refer to null_str.
+
+Mon Aug  4 17:25:18 2003  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* stable version 1.8.0 released.
+
+For the changes before 1.8.0, see doc/ChangeLog-1.8.0
+
+Local variables:
+add-log-time-format: (lambda ()
+  (let* ((time (current-time))
+	 (system-time-locale "C")
+	 (diff (+ (cadr time) 32400))
+	 (lo (% diff 65536))
+	 (hi (+ (car time) (/ diff 65536))))
+  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
+indent-tabs-mode: t
+tab-width: 8
+end:
diff --git a/doc/ChangeLog-2.0.0 b/doc/ChangeLog-2.0.0
new file mode 100644
index 0000000000..a1a79b8dca
--- /dev/null
+++ b/doc/ChangeLog-2.0.0
@@ -0,0 +1,24015 @@
+Fri Feb  8 19:56:54 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* array.c (rb_ary_dup): reverted r39004.  see [Bug #7768], and
+	  release manager finally decided to revert it.
+
+Fri Feb  8 16:09:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_ensure): preserve errinfo across ensure proc before
+	  JUMP_TAG().  [ruby-core:52022] [Bug #7802]
+
+Fri Feb  8 16:08:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_separately): check also terminating
+	  signal not only if core dumped.
+
+Fri Feb  8 13:12:04 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/generator/darkfish.rb:  Set encoding on output template to
+	  user-specified encoding.
+	* test/rdoc/test_rdoc_generator_darkfish.rb:  Test for above.
+
+	* lib/rdoc.rb:  Bump version
+
+Fri Feb  8 11:53:33 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/security/policy.rb:  Raise proper exceptions when
+	  verifying unsigned gems (instead of crashing).
+	* test/rubygems/test_gem_security_policy.rb:  Tests for the above.
+
+Fri Feb  8 10:44:44 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/test_gem_dependency_installer.rb:  Improve coverage of
+	  --install-dir feature of gem install.
+
+Fri Feb  8 10:11:09 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/config_file.rb:  Add missing require for
+	  user_interaction.rb
+
+	* lib/rubygems/dependency_installer.rb:  Minor refactor for clarity.
+
+Fri Feb  8 09:35:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#configuration): set all ruby names.
+	  hdrdir now needs RUBY_VERSION_NAME.
+
+Fri Feb  8 08:58:26 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/package/old.rb:  Fix loading old format gems on ruby
+	  1.8.  This commit is only so trunk and rubygems master have the same
+	  code.
+
+Fri Feb  8 08:53:27 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: fixing string quotation
+	  when dumping Ruby strings. Thanks Ingy
+
+	* test/psych/test_psych.rb: appropriate tests.
+
+	* test/psych/test_yaml.rb: ditto
+
+Fri Feb  8 08:50:42 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: change output reference
+	  ids to be sequential numbers.
+
+Fri Feb  8 07:47:56 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/package/old.rb:  Disallow installation of old-format
+	  gems when a security policy is active.
+	* test/rubygems/test_gem_package_old.rb:  Test for above.
+
+Fri Feb  8 07:34:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/net/http.rb (HTTP.post_form): Fix module scope in documentation
+	  Patch by David Albert [Bug #7794] [ruby-core:51955]
+
+Fri Feb  8 07:33:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* compar.c (cmp_equal): Document ignored exception and return false
+	  By Makoto Kishimoto [Bug #7790] [ruby-dev:46925] [ruby-dev:46910]
+
+Fri Feb  8 07:17:00 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/dependency_installer.rb:  Only install local gems if
+	  they end in '.gem'.  Fixes github rubygems issue #407.
+	* test/rubygems/test_gem_dependency_installer.rb:  Test for the above.
+
+Fri Feb  8 00:02:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (obj2gid): use getgrnam_r() only if getgrnam_r() and
+	  _SC_GETGR_R_SIZE_MAX is available.
+	  MirOS BSD (MirBSD 10 GENERIC#1382 i386) have getgrnam_r() but
+	  no _SC_GETGR_R_SIZE_MAX.
+	  (obj2uid): use getpwnam_r() only if getpwnam_r() and
+	  _SC_GETPW_R_SIZE_MAX is available.
+	  This is consistency for obj2gid.
+	  MirOS BSD have neither getpwnam_r() nor _SC_GETPW_R_SIZE_MAX.
+
+Thu Feb  7 22:01:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: define linker for shared library on MirOS BSD.
+
+Thu Feb  7 21:09:23 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_config_file.rb
+	  (TestGemConfigFile#test_check_credentials_permissions): skip on
+	  Windows. see [Bug #7784] [ruby-core:51864] and r39070.
+
+Thu Feb  7 20:52:40 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): added variables which were
+	  missing at r39130.
+
+Thu Feb  7 15:33:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#merge_libs): insert following reversal
+	  ordered elements just after the duplicated element, not overwriting
+	  successive elements.  [ruby-core:50314] [Bug #7467]
+
+Thu Feb  7 14:56:15 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/package.rb:  Ensure digests are generated for signing.
+	* test/rubygems/test_gem_package.rb:  Test for the above.
+
+	* lib/rubygems/security/policy.rb:  Ensure digests are present when
+	  verifying a gem and match the number of signatures bidirectionally.
+	* test/rubygems/test_gem_security_policy.rb:  Test for the above.
+
+	* lib/rubygems.rb:  Documentation improvements (by zzak)
+
+Thu Feb  7 05:52:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/pty/README: Remove static documentation file
+	* ext/pty/pty.c: Add License to PTY module overview
+
+Thu Feb  7 02:31:10 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* vm_insnhelper.c: attr_writer should return its argument [Bug #7773]
+
+	* test/ruby/test_basicinstructions.rb: Test for above
+
+Thu Feb  7 01:35:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/security.rdoc: Link to japanese version of CVE page patch by
+	  nagachika
+
+Wed Feb  6 23:30:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/pty/README.expect: Removed static documentation file
+	* ext/pty/lib/expect.rb: Documentation for IO#expect
+
+Wed Feb  6 22:25:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* hash.c (env_reject_bang): hide keys array from ObjectSpace
+	* hash.c (env_select_bang): ditto
+
+Wed Feb  6 17:33:01 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (multiarch): add option to move architecture dependent
+	  directories.  [Feature #6111]
+
+	* template/ruby.pc.in: add arch dependent paths.
+
+	* configure.in (rubyarchhdrdir, sitearchhdrdir, vendorarchhdrdir): add
+	  options to customize architecture dependent header directories.
+
+	* configure.in (rubyarchprefix, sitearchdir, vendorarchdir): add
+	  options to customize architecture dependent library directories.
+
+	* template/ruby.pc.in, tool/mkconfig.rb, tool/rbinstall.rb: use
+	  configured values.
+
+	* tool/mkconfig.rb: expand rubyarchdir to extract prefix.
+
+	* configure.in (RUBY_VERSION_NAME), template/ruby.pc.in: add
+	  substitution and define.
+
+	* configure.in, version.c: parametric architecture name for paths.
+
+	* configure.in (shvar_to_cpp): convert sh variable references
+	  by replacing with string literal forms in cpp.
+
+Wed Feb  6 17:05:26 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Import RDoc 4.0.0.rc.2
+
+Mon Feb  4 02:22:49 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_process.rb (test_setsid): ensure to call
+	  Process.wait(). Reported by George Koehler. Thanks.
+
+Mon Feb  4 02:18:00 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_process.rb (test_setsid): skip when platform is
+	  OpenBSD. Contributed from George Koehler.
+	  [Bug #7789] [ruby-core:51889]
+
+Wed Feb  6 13:35:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_method_entry_location, rb_{mod,obj}_method_location): new
+	  functions to obtain source location of method definition.
+
+	* vm_method.c (rb_obj_respond_to): show the location of old style
+	  respond_to? method.
+
+Wed Feb  6 13:03:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/security.rdoc: Add link to CVEs on ruby-lang.org/en/security
+
+Wed Feb  6 12:49:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* NEWS: Add note about removal of CSV::load and CSV::dump from r39077
+
+Wed Feb  6 05:57:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/racc/parser.rb: Hide copyright notice from Racc doc
+
+Wed Feb  6 05:50:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/rubygems/*: Removed outdated documentation files
+	* lib/rubygems/LICENSE.txt: Include license file
+	* lib/rubygems.rb: Move Gem module documentation so rdoc can parse it
+	  and link to LICENSE.txt
+	* lib/rubygems/*: Hide useless documentation from Gem module rdoc
+
+Wed Feb  6 03:45:19 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/security.rdoc: Remove documentation for unsafe CSV.load which
+	  was deleted in r39077
+
+Wed Feb  6 03:27:19 2013  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb: Remove the dangerous serialization feature.
+
+Wed Feb  6 00:56:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb.rb: Remove example from restrictions, it works [Github #246]
+	  Based on patch by Ryunosuke SATO
+
+Wed Feb  6 00:46:53 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (rb_vm_stack_to_heap): call rb_vm_get_binding_creatable_next_cfp
+	  instead of rb_vm_get_ruby_level_next_cfp to prevent a segfault by
+	  calling Kernel#callcc. See r39067 for more details.
+	  [ruby-dev:46908] [ruby-trunk - Bug #7774]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Tue Feb  5 18:48:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* doc/security.rdoc: add regex, eval and drb sections
+
+Tue Feb  5 17:24:02 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/servlet.rb:  Fixed root search paths, filesystem paths
+	  instead of HTTP paths were returned.
+	* test/rdoc/test_rdoc_servlet.rb:  Test for above.
+
+Tue Feb  5 16:37:00 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/config_file.rb:  Ignore permissions check on windows.
+	  Windows writes 0600 file as 0644 permissions making the check
+	  useless.
+
+Tue Feb  5 16:25:25 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_obj_respond_to): drop optional include_all flag if
+	  respond_to? method is defined in old style.  [Bug #7722]
+
+Tue Feb 05 15:04:34 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_binding_new_with_cfp): permit to create binding object
+	  of IFUNC frame.
+	  When `rb_binding_new_with_cfp()' is called, VM finds out the first
+	  normal (has iseq) frame and create a binding object of this frame
+	  and create Env objects. `ep's of related frames are updated
+	  (`ep's point Env object managed spaces).
+	  However, `ep' of skipped IFUNC frame was not updated and
+	  old invalid `ep' was remained. It causes serious problems.
+	  To solve this issue, permit IFUNC to create binding.
+	  (Maybe there is no problem on it)
+	  [ruby-dev:46908] [ruby-trunk - Bug #7774]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+	* vm.c (rb_vm_get_binding_creatable_next_cfp), vm_core.h: added.
+
+	* vm_trace.c: fix to use `rb_vm_get_binding_creatable_next_cfp()'.
+
+Tue Feb  5 14:43:15 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Fix error message, patch by pypypy [Bug #7777]
+
+Tue Feb  5 14:36:04 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c (fix_pow): Handle special cases when base is 0, -1 or +1
+	  [Bug #5713] [Bug #5715]
+
+	* rational.c (nurat_expt): ditto
+
+Tue Feb  5 13:27:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (rawmode_opt): use default values by `stty
+	  raw`.
+
+Tue Feb  5 12:50:47 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* range.c: Use div instead of / for bsearch
+
+	* test/ruby/test_range.rb: Test showing bug when requiring mathn
+
+Tue Feb  5 12:48:38 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: Use to_enum for Enumerable methods returning
+	  Enumerators.
+	  This makes Lazy#cycle no longer needed, so it was removed.
+	  Make Enumerator#chunk and slice_before return lazy Enumerators.
+	  [Bug #7715]
+
+	* internal.h: Remove ref to rb_enum_cycle_size; no longer needed
+
+	* enum.c: Make enum_cycle_size static.
+
+	* test/ruby/test_lazy_enumerator.rb: Test for above
+
+Tue Feb  5 12:48:10 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: Finalize and document Lazy.new. [Bug #7248]
+	  Add Lazy#to_enum and simplify Lazy#size.
+
+	* test/ruby/test_lazy_enumerator.rb: tests for above
+
+Tue Feb  5 11:35:35 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/push_command.rb:  Fixed credential download for
+	  `gem push --host`
+	* lib/rubygems/gemcutter_utilities.rb:  ditto.
+	* test/rubygems/test_gem_commands_push_command.rb:  Test for the above.
+	* test/rubygems/test_gem_gemcutter_utilities.rb:  ditto.
+
+	* lib/rubygems/config_file.rb:  Abort if the `gem push` credentials
+	  file has insecure permissions.
+	* test/rubygems/test_gem_config_file.rb:  Test for the above.
+
+	* lib/rubygems/ext/builder.rb:  Do not look for Gemfile, Isolate, etc.
+	  while building gem extensions.
+
+	* lib/rubygems/package.rb:  Unset spec and files list if a gem's
+	  signatures cannot be verified.
+	* test/rubygems/test_gem_package.rb:  Test for the above.
+
+	* lib/rubygems/specification.rb:  Reduce use of eval.
+	* lib/rubygems/test_case.rb:  ditto.
+
+	* test/rubygems/test_gem_specification.rb:  Test setting
+	  specification_version for legacy gems.  Dup Gem.ruby before
+	  untainting in case it's frozen.
+
+	* lib/rubygems.rb:  Reduce use of eval.  Only read files when looking
+	  for Gemfile, Isolate, etc.
+	* test/rubygems/test_gem.rb:  Test for the above.
+
+Tue Feb  5 10:15:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/security.rdoc: Wrap security guide at 80 columns
+
+Tue Feb  5 10:15:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/security.rdoc: Grammatical error on security guide
+	  Patch by Josh Bassett [Github fixes #245]
+
+Tue Feb  5 10:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/racc/parser.rb: Update #do_parse and #yyparse from upstream
+	  See [Github tenderlove/racc@7d954b5]
+
+Tue Feb  5 09:55:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/racc: Merge Racc documentation downstream, add grammar ref file
+
+Tue Feb  5 08:03:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb.rb, lib/irb/ext/save-history.rb: Add documentation on how to
+	  enabled irb history [ruby-core:51347] [Bug #7679]
+
+Tue Feb  5 07:35:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb.rb, lib/irb/context.rb: Add documentation on how to enable
+	  auto-indentation and autocompletion using irbrc and irb_context
+	  [ruby-core:51209] [Bug #7642] and [ruby-core:51348] [Bug #7680]
+
+Tue Feb  5 05:20:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/standard_library.rdoc: Document list of libraries and extensions
+	  and their purpose or short description
+	* lib/README: Remove lib/README in favor of doc/standard_library.rdoc
+
+Tue Feb  5 04:40:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/json/lib/json.rb: Move module overview definition for rdoc
+
+Tue Feb  5 03:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/tracer.rb: Move class overview definition and reformat
+
+Mon Feb  4 15:10:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (rawmode_opt): initialize options for the
+	  case all options are not given.
+
+Mon Feb  4 12:44:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c (control_frame_dump): capitalize prefix of `ep'
+	  if `ep' points an env object.
+
+Mon Feb  4 04:20:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/English.rb: Add English module for RDoc to parse, then
+	  remove_const to avoid confusion. Include full list of aliases and
+	  their associated global variable.
+
+Mon Feb  4 02:40:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/yaml.rb (YAML::EngineManager): Documentation for #yamler and
+	  #yamler= for using the removed Syck gem as the YAML::ENGINE
+
+Sun Feb  3 16:54:27 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/io-console.gemspec: bump.  [Bug #7762]
+
+	* test/io/console/test_io_console.rb (test_stringio_getch): use more
+	  descriptive assertions.
+
+	* ext/io/console/console.c (rawmode_opt): min is minimum characters,
+	  not tenths.
+
+Sun Feb  3 16:13:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* doc/security.rdoc: add first cut at a Ruby security document
+
+Sun Feb  3 10:25:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* random.c: Document range argument for Kernel#rand.
+	  [ruby-core:51794] [Bug #7770]
+
+Sun Feb  3 10:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* numeric.c: Document Float constants [ruby-core:51484] [Bug #7709]
+
+Sun Feb  3 09:38:44 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/profiler.rb (PROFILE_CALL_PROC, PROFILE_RETURN_PROC): add b_call
+	  and b_return to profile block calls.
+
+	* lib/profiler.rb (PROFILE_CALL_PROC, PROFILE_RETURN_PROC): split
+	  PROFILE_PROC for call and return events.
+
+Sat Feb  2 14:32:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/minitest/mock.rb, lib/minitest/hell.rb: nodoc top-level module
+
+Sat Feb  2 14:05:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/debug.rb: Documentation for DEBUGGER__ class methods based on
+	  patch by Vincent Batts [ruby-core:51253]
+
+Sat Feb  2 13:37:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/net/smtp.rb: Fix rdoc title for Net::SMTP
+
+Sat Feb  2 13:32:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/net/pop.rb: Fix rdoc title for Net::POP3
+
+Sat Feb  2 13:00:11 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/gserver.rb (GServer#start): fix a timing issue.  patch from
+	  Charles Nutter.  [Bug #7081]
+
+Sat Feb  2 12:36:54 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/fileutils.rb (copy_entry, wrap_traverse): preserve attributes of
+	  directories on FileUtils.cp_r.  The fix was proposed by Jan
+	  Wedekind.  [Bug #7246]
+
+	* test/fileutils/test_fileutils.rb: add a test for above.
+
+Sat Feb  2 12:30:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/uri/ftp.rb (URI::FTP.new2): nodoc method from r39013 [Bug #7301]
+
+Sat Feb  2 12:15:36 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/uri/ftp.rb (URI::FTP.new2): remove the rdoc because it is not
+	  well tested yet.  [Bug #7301]
+
+Sat Feb  2 12:07:41 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ChangeLog: Forgot to add a reference to the ChangeLog of the
+	  previous commit.
+
+Sat Feb  2 12:05:18 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/fileutils.rb: chmod/chmod_R with a string mode (e.g., "+x")
+	  caused error in verbose mode.  [Bug #7373]
+
+	* test/fileutils/test_fileutils.rb: add a test for above.
+
+Sat Feb  2 11:44:42 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/English.rb: Remove some confusing words from rdoc.  [Bug #7406]
+
+Sat Feb  2 10:17:12 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* NEWS: add keyword arguments.
+
+Sat Feb  2 07:45:44 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c (proc_curry): Fix arity check [Bug #5747]
+
+	* test/ruby/test_proc.rb: Test for above
+
+Sat Feb  2 07:44:15 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c: Add {*}_min_max_arity and refactor.
+	  [Bug #7765]
+
+	* test/ruby/test_proc.rb: Fix wrong test
+
+Fri Feb  2 00:46:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* marshal.c: add security considerations to marshal overview, refer to
+	  overview from Marshal.load documentation [#7759]
+
+Fri Feb  1 23:04:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* array.c (rb_ary_dup): make returned array the same class as the original
+	  array [Bug #7768] [ruby-core:51792]
+	* test/ruby/test_array.rb (class TestArray): add test
+
+Fri Feb  1 16:35:34 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): prohibit setting instance variables of
+	  existing class/module.
+
+Fri Feb  1 14:34:29 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/readline/extconf.rb, ext/readline/readline.c: check
+	  RL_PROMPT_START_IGNORE and RL_PROMPT_END_IGNORE directly in
+	  readline.c.  Patch by Zachary Scott.  [Bug #7397] [ruby-core:49561]
+
+Thu Jan 31 21:55:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* marshal.c (marshal_load): Add documentation warning against using
+	  Marshal.load on untrusted data [Bug #7759] [ruby-core:51765]
+
+Thu Jan 31 16:33:27 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (local_push_gen): no assigned but unused variable warnings
+	  in eval as well as -e.  [Feature #7730] [ruby-core:51580]
+
+Wed Jan 30 12:30:08 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_signal.rb (test_trap_puts): Fix typo. "sync"
+	  should be "STDOUT.sync".
+
+Thu Jan 31 15:39:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* string.c (rb_str_aset_m): Documentation for String#[]= fix
+	  Raises an IndexError if Regexp match is out of range.
+	  Github fixes #243 Patch by Dmtiriy Budnik
+
+Thu Jan 31 13:54:44 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (rsock_unix_sockaddr_len): return
+	  sizeof(sa_family_t) if path is empty.  see "Autobind Feature" in
+	  unix(7) for details.
+
+	* ext/socket/lib/socket.rb (unix_socket_abstract_name?): treat an
+	  empty path as an abstract name.
+
+	* test/socket/test_unix.rb: related test.
+
+Wed Jan 30 20:58:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/basicsocket.c (bsock_getsockname): ignore truncated
+	  part of socket address.
+	  (bsock_getpeername): ditto.
+	  (bsock_local_address): ditto.
+	  (bsock_remote_address): ditto.
+
+	* ext/socket/unixsocket.c (unix_path): ditto.
+	  (unix_addr): ditto.
+	  (unix_peeraddr): ditto.
+
+	* ext/socket/init.c (cloexec_accept): ditto.
+
+Wed Jan 30 17:08:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (fstat): revert r37337, which uses _fstati64()
+	  instead of fstati64() on mingw32.  [Bug #7276]
+
+Wed Jan 30 15:26:37 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/socket/unixsocket.c (rsock_init_unixsock): use rb_inspect()
+	  because rb_sys_fail_str() fails if its argument contains NUL.
+
+	* test/socket/test_unix.rb: related test.
+
+Wed Jan 30 15:21:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): show the most important message, Crash
+	  Report log information, first.
+
+Wed Jan 30 15:00:05 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c (rb_ary_bsearch): Raise TypeError on bad return from block
+
+	* range.c (range_bsearch): ditto
+
+	* test/ruby/test_array.rb (class): Test for above
+
+	* test/ruby/test_range.rb (class): ditto
+
+Wed Jan 30 14:46:28 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* range.c: Restrict bsearch to integers [#7728]
+
+	* test/ruby/test_range.rb: Test for above
+
+Wed Jan 30 14:10:52 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c (rb_ary_bsearch): Return enumerator if no block [#7725]
+
+	* range.c (range_bsearch): ditto
+
+	* test/ruby/test_array.rb: Test for above
+
+	* test/ruby/test_range.rb: ditto
+
+Wed Jan 30 13:53:43 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Take conjugate for inner product
+	  [rubyspec:5a01ad5719f2] [ruby-dev:46101]
+
+Wed Jan 30 13:22:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (local_push_gen): warn assigned but unused variables also in
+	  toplevel, except for -e option.  [Feature #7730] [ruby-core:51580]
+
+Wed Jan 30 13:17:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (cont_restore_thread): svar should be separate per fibers.
+	  [ruby-core:51331] [Bug #7678]
+
+Wed Jan 30 07:15:04 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* re.c (reg_operand): Simplify and reuse error handling [Bug #7539]
+
+	* test/ruby/test_regexp.rb: Test for above
+
+Wed Jan 30 07:00:16 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* object.c: Improve error for failed implicit conversions [Bug #7539]
+
+	* error.c: Adapt rdoc
+
+	* test/ruby/test_object.rb: Test for above
+
+Tue Jan 29 21:40:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/net/http/generic_request.rb (encode_multipart_form_data): remove
+	  tempfile explicitly.
+
+Tue Jan 29 19:27:18 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* array.c: Improve documentation about
+	  comparison by hash for concerned methods. [ruby-core:51266]
+
+Tue Jan 29 17:03:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: fix issue of rb_debug_inspector_open().
+	  The order of making binding should be stack (frame) top to bottom.
+	  [Bug #7635]
+	  And also fix issue of collecting klass. Collecting klass is same
+	  as TracePoint#defined_class.
+	  (previous version, it returns T_ICLASS (internal objects).
+
+	* test/-ext-/debug/test_debug.rb: add a test.
+
+	* ext/-test-/debug/extconf.rb, init.c, inspector.c: ditto.
+
+	* vm_backtrace.c: remove magic number and add enum CALLER_BINDING_*.
+
+	* vm_backtrace.c, include/ruby/debug.h: add new C api (experimental)
+	  rb_debug_inspector_frame_self_get().
+
+	* vm.c, vm_core.h, vm_trace.c: move decl. of
+	  rb_vm_control_frame_id_and_class() and constify first parameter.
+
+Tue Jan 29 16:50:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_trace.c (rb_tracepoint_enable, rb_tracepoint_disable): check safe
+	  level as well as set_trace_func.
+
+	* vm_trace.c (set_trace_func, thread_{add,set}_trace_func_m): check
+	  safe level as well as 1.8.
+
+Tue Jan 29 16:49:19 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_mod_method_arity): return original arity of the method if
+	  aliased because of visibility change, like as Method#arity.
+
+Tue Jan 29 12:05:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_marshal.rb: remove temporally files early.
+
+	* test/ruby/test_process.rb: ditto.
+
+	* test/psych/test_exception.rb: ditto.
+
+Tue Jan 29 09:26:20 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_s_pack_sockaddr_un): calculate the
+	  correct address length of an abstract socket.
+
+	* test/socket/test_unix.rb: related test.
+
+Mon Jan 28 18:02:16 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_backtrace.c (rb_debug_inspector_frame_{class,binding,iseq}_get):
+	  use long as index as well as RARRAY_LEN().
+
+Mon Jan 28 17:51:38 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_separately): imply no core dump.
+
+Mon Jan 28 12:32:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/fcntl/fcntl.c: update document.  use "file descriptor" instead
+	  of "file handle" because it is not used other Ruby documents and
+	  it is confusing with Windows file handle.
+	  correct F_DUPFD behavior.
+
+Sat Jan 26 22:39:12 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): dump instance variables of the result of
+	  marshal_dump not the original object.  [ruby-core:51163] [Bug #7627]
+
+	* complex.c (nucomp_marshal_dump): need to copy instance variables.
+
+	* rational.c (nurat_marshal_dump): ditto.
+
+Sat Jan 26 13:35:56 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/fcntl/fcntl.c:  Document Fcntl constants
+
+Sat Jan 26 12:54:40 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* hash.c (rb_env_size):  Restored documentation for ENV.size
+
+	* lib/drb/drb.rb:  Documented DRb::DRb#run.
+
+	* lib/erb.rb (class ERB):  Improved documentation of ERb.
+
+	* transcode.c:  Documented Encoding::Converter constants.
+
+Sat Jan 26 10:09:57 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/accesslog.rb:  Improved WEBrick documentation.
+	* lib/webrick/cgi.rb:  ditto.
+	* lib/webrick/config.rb:  ditto.
+	* lib/webrick/cookie.rb:  ditto.
+	* lib/webrick/httpauth/authenticator.rb:  ditto.
+	* lib/webrick/httpauth/basicauth.rb:  ditto.
+	* lib/webrick/httpauth/digestauth.rb:  ditto.
+	* lib/webrick/httpproxy.rb:  ditto.
+	* lib/webrick/httprequest.rb:  ditto.
+	* lib/webrick/httpresponse.rb:  ditto.
+	* lib/webrick/https.rb:  ditto.
+	* lib/webrick/httpserver.rb:  ditto.
+	* lib/webrick/httpservlet/cgihandler.rb:  ditto.
+	* lib/webrick/httpservlet/filehandler.rb:  ditto.
+	* lib/webrick/httpservlet/prochandler.rb:  ditto.
+	* lib/webrick/httputils.rb:  ditto.
+	* lib/webrick/httpversion.rb:  ditto.
+	* lib/webrick/log.rb:  ditto.
+	* lib/webrick/server.rb:  ditto.
+	* lib/webrick/ssl.rb:  ditto.
+	* lib/webrick/utils.rb:  ditto.
+	* lib/webrick/version.rb:  ditto.
+
+Sat Jan 26 08:29:33 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/socket/raddrinfo (rsock_unix_sockaddr_len): renamed from
+	  rsock_unixpath_len, because it returns not the length of the path,
+	  but the length of a socket address for the path.
+
+Sat Jan 26 01:12:23 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (test_ioctl_linux): skip if a platform is
+	  not x86 because linux ioctl request number depend on cpu arch.
+	  At least, alpha, mips, sparc and ppc have a different number.
+	  [Bug #7718] [ruby-core:51544]
+
+Fri Jan 25 19:14:24 2013  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: use TlsAlloc instead of __declspec(thread)
+	  to avoid SEGV if win32ole.so loaded with LoadLibrary in Windows
+	  XP or earlier.
+
+Fri Jan 25 16:47:31 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (rsock_unixpath_len, init_unix_addrinfo),
+	  ext/socket/unixsocket.c (unixsock_connect_internal,
+	  rsock_init_unixsock): calculate the correct address length of
+	  an abstract socket.  Without this fix, sizeof(struct sockaddr_un)
+	  is specified as the length of an abstract socket for bind(2) or
+	  connect(2), so the address of the socket is filled with extra NUL
+	  characters.  See unix(7) for details.
+
+	* ext/socket/lib/socket.rb (unix_server_socket): don't access the
+	  file system if the platform is Linux and path starts with NUL,
+	  which means that the socket is an abstract socket.
+
+	* test/socket/test_unix.rb: related test.
+
+Fri Jan 25 13:02:27 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/drb/drb.rb:  Updated documentation based on patch from Vincent
+	  Batts.  [ruby-trunk - Bug #7714]
+	* lib/drb/ssl.rb:  ditto.
+
+Fri Jan 25 12:23:29 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/drb/drb.rb:  Improved documentation by adding or hiding methods.
+	* lib/drb/eq.rb:  ditto.
+	* lib/drb/extserv.rb:  ditto.
+	* lib/drb/gw.rb:  ditto.
+	* lib/drb/invokemethod.rb:  ditto.
+	* lib/drb/observer.rb:  ditto.
+	* lib/drb/ssl.rb:  ditto.
+	* lib/drb/timeridconv.rb:  ditto.
+	* lib/drb/unix.rb:  ditto.
+
+	* sample/drb/gw_cu.rb:  Fixed bug in DRb gateway sample.
+
+Fri Jan 25 12:01:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: modify a comment about rb_iseq_t::local_size.
+	  A patch by davidbalbert (David Albert) [Bug #6750]
+
+Fri Jan 25 10:36:31 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mkmf.rb:  Documented MakeMakefile constants.  Hide implementation
+	  details from RDoc
+
+Fri Jan 25 10:04:07 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/compatibility.rb:  Hide compatibility shims from RDoc
+
+	* lib/rubygems/config_file.rb:  Hide RbConfig use from RDoc
+
+	* lib/rubygems/test_case.rb:  Added note to use realpath when 1.8
+	  support is dropped.
+
+Fri Jan 25 09:14:43 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/generator/darkfish.rb:  Fixed debug message.  RDoc bug #174
+	  by Thomas Leitner.
+
+	* lib/rdoc/store.rb:  Fixed deletion of ri attribute data when a class
+	  was loaded then saved.  RDoc bug #171 by Thomas Leitner.
+	* test/rdoc/test_rdoc_store.rb:  Test for above.
+
+Thu Jan 24 19:55:25 2013  Shota Fukumori  <her@sorah.jp>
+
+	* NEWS (yaml): Write about bundled libyaml.
+
+Thu Jan 24 16:54:34 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/calling_methods.rdoc:  Added a Method Lookup section.
+	* doc/syntax/refinements.rdoc (Method Lookup):  Clarified that
+	  refinement methods are looked up in classes, not instances.
+
+Thu Jan 24 16:49:17 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enum.c (enum_zip): Fix error message
+
+	* array.c (take_items): Same, for Array#zip
+
+Thu Jan 24 16:47:26 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c (lazy_zip): raise error for bad arguments
+	  [Bug #7706]
+
+Thu Jan 24 16:05:08 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: Optimize Lazy#zip when passed only arrays
+	  [Bug #7706]
+
+Thu Jan 24 15:21:17 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: Fix state handling for Lazy#zip,{drop_take}{_while}
+	  [bug #7696] [bug #7691]
+
+Thu Jan 24 11:43:47 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* eval.c (f_current_dirname): Add documentation about "__dir__
+	  returns always an absolute path". [Bug #7729]
+
+Thu Jan 24 10:28:30 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS (RDoc):  Added mention of page support and markdown support.
+
+Thu Jan 24 09:40:13 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/refinements.rdoc:  Added refinements document based on
+	  the specification from the wiki.
+	* doc/syntax.rdoc:  Added link to refinements document.
+
+Wed Jan 23 16:29:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn_flags): fix missing
+	  initialization.  pointed out by phasis68 (Heesob Park) at
+	  [ruby-core:51579].  [Bug #7721]
+
+Wed Jan 23 16:18:04 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#try_constant): fix for large unsigned.
+
+	* lib/mkmf.rb (MakeMakefile#try_constant): fix for larger constants.
+
+	* test/mkmf/test_constant.rb: tests for try_constant.
+	  TODO: define check_constant and use it.
+
+Wed Jan 23 13:35:37 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (ruby_init_stack): ignore `STACK_END_ADDRESS'
+	  if Ruby interpreter is running on co-routine.
+	  [Feature #2294]
+	  https://bugs.ruby-lang.org/issues/2294#note-18
+
+Wed Jan 23 12:28:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_spawn, rb_w32_aspawn_flags): check the results
+	  of acp_to_wstr() which can return NULL.  [ruby-core:51557] [Bug #7721]
+
+Wed Jan 23 10:40:49 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/assignment.rdoc (Implicit Array Assignment):  Clarify
+	  that "left-hand side" means "of the assignment".  Suggested by Jorge
+	  Dias.
+	* doc/syntax/assignment.rdoc (Multiple Assignment):  ditto.
+
+Wed Jan 23 10:34:47 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/assignment.rdoc (Local Variables and Methods):  Fixed
+	  example showing caching of a method's results into a local variable.
+	  Added not about using an explicit receiver to call a method that
+	  matches a local variable.  Suggested by markov_twain on twitter.
+
+Wed Jan 23 10:20:08 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/README:  Fixed typo.  Patch by Pradeep Sahoo.
+	  Fixes #240 on github
+
+Wed Jan 23 09:53:39 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/servlet.rb:  Fixed display of site and home documentation.
+	  Fixes rdoc issue #170 by Thomas Leitner.
+	* test/rdoc/test_rdoc_servlet.rb:  Test for above.
+
+	* lib/rdoc/code_object.rb:  Split #initialize_visibility from
+	  #initialize for reuse when loading a stored object.
+	  Fixes rdoc issue #171 by Thomas Leitner.
+
+	* lib/rdoc/any_method.rb:  Initialize visibility for #display?  For
+	  rdoc issue #171
+	* lib/rdoc/attr.rb:  ditto.
+	* lib/rdoc/class_module.rb:  ditto.
+	* lib/rdoc/top_level.rb:  ditto.
+	* test/rdoc/test_rdoc_any_method.rb:  Test for above.
+	* test/rdoc/test_rdoc_attr.rb:  ditto.
+	* test/rdoc/test_rdoc_class_module.rb:  ditto.
+	* test/rdoc/test_rdoc_constant.rb:  ditto.
+	* test/rdoc/test_rdoc_top_level.rb:  ditto.
+
+Wed Jan 23 06:43:26 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/test_case.rb:  Use Dir.tmpdir for rubygems tests instead
+	  of ./tmp/test.  Fixes [ruby-trunk - Bug #7717]
+
+Tue Jan 22 22:58:03 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-curlies): Fix the bug where
+	  an open curly inserted in a string is always replaced with a
+	  hash sign.
+
+Mon Jan 21 15:41:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb: BASERUBY is transient at core build.
+
+Mon Jan 21 13:51:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb ($extmk): traverse parent directories for the case
+	  srcdir is a symlink.
+
+Sun Jan 20 23:55:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object, r_object0): separate respond_to checks and
+	  calling, and get back to the old behavior for 2.0.  [Bug #7564]
+
+Sun Jan 20 22:24:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/vpath.rb (VPath#def_options): hack for msys make, which
+	  converts a command line argument to non-msys command seems like a
+	  path list automagically.  [Bug #7710] [ruby-core:51489]
+
+Sat Jan 19 11:35:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* struct.c (Struct.new): Document Struct.new with block
+	  Patch by Hiroyuki Iwatsuki [Bug #7674]
+
+Sat Jan 19 09:52:46 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/miscellaneous.rdoc:  Added section on defined?
+
+Sat Jan 19 09:27:31 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/assignment.rdoc (Local Variables and Methods):  Made it
+	  more clear that local variables are created by the parser, not
+	  execution.  Thanks to John Hawthorn.
+
+Sat Jan 19 09:15:58 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/assignment.rdoc:  Improved links
+	* doc/syntax/methods.rdoc:  ditto.
+
+	* doc/syntax.rdoc:  Added link to assignment document
+
+Sat Jan 19 08:47:33 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/assignment.rdoc:  Added a syntax document on assignment.
+
+Fri Jan 18 14:11:01 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc:  Added Array Decomposition.
+
+Fri Jan 18 12:54:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (gem): Gem.ensure_gem_subdirectories makes
+	  subdirectories group-writable, so make them with $dir_mode.
+
+Fri Jan 18 11:24:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (ole_initialize): uninitialize OLE at thread
+	  ends.  [Bug #2618] [ruby-core:27634]
+
+	* ext/win32ole/win32ole.c (ole_initialize): initialize OLE for each
+	  threads.  [Bug #2618] [ruby-core:27634]
+
+Thu Jan 17 22:10:35 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rubygems/ext/builder.rb (Gem::Ext::Builder.make): Remove
+	  .time dependency from *.rb install target. It causes needless
+	  *.rb install. [Bug #7698] [ruby-core:51437]
+	  Reported by Tadashi Saito. Thanks!!!
+	* test/rubygems/test_gem_installer.rb
+	  (TestGemInstaller#test_install_extension_and_script): Add a test
+	  for the above change.
+
+Thu Jan 17 21:08:20 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* eval.c: Fix a typo in ruby_finalize() documentation.
+
+Thu Jan 17 20:28:18 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* object.c: Typo in Kernel#hash documentation.
+	  Patch by zed_0xff [Github Fixes #237]
+
+Thu Jan 17 10:48:56 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: use constants rather than
+	  calculating Inf and NaN.
+
+Thu Jan 17 10:21:05 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/miscellaneous.rdoc:  Added Ending an Expression and
+	  indentation.
+
+Thu Jan 17 09:30:21 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/installer.rb:  Untaint string when checking output
+	  for $SAFE=1
+
+	* lib/rubygems/specification.rb:  Keep previously loaded specs as
+	  active.  This prevents double loading when refreshing the gem list.
+	* test/rubygems/test_gem.rb:  Test for above
+
+	* lib/rubygems.rb:  Bump version to 2.0.0.rc.2
+
+Thu Jan 17 09:08:37 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/control_expressions.rdoc:  Added ? : ternary if
+
+Thu Jan 17 08:36:04 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/miscellaneous.rdoc:  Added documentation for alias, undef,
+	  BEGIN, END.
+	* doc/syntax/modules_and_classes.rdoc (Constants):  Fixed unwrapped
+	  paragraph with trailing whitespace.
+	* doc/syntax/modules_and_classes.rdoc (Scope):  Added section pointing
+	  to alias and undef documentation.
+	* doc/syntax.rdoc:  Added link to miscellaneous section.
+
+Thu Jan 17 07:50:26 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/control_expressions.rdoc (Flip-Flop):  Added a section on
+	  the flip-flop.
+
+Thu Jan 17 06:59:51 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/control_expressions.rdoc (if Expressions):  Fixed markup
+	  error.  Fixes #235 on github by FlyingFoX.
+
+Thu Jan 17 06:53:58 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/literals.rdoc (Strings):  Fixed typo.  Fixes #236 on
+	  github by Doug Yun.
+
+Wed Jan 16 18:45:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/lib/win32ole.rb: use TracePoint to hook all thread
+	  creation not only by Thread.new and to get rid of interference with
+	  svar scope.  [Bug #7681] [ruby-core:51365]
+
+Wed Jan 16 09:35:53 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* .document:  Removed extra space
+	* lib/irb/lc/.document:  Hide help-message
+	* lib/minitest/.document:  Hide README.txt
+	* lib/rake/lib/.document:  Hide project.rake
+	* lib/rdoc/generator/template/json_index/.document:  Hide JavaScript
+	  files
+	* lib/rubygems/ssl_certs/.document:  Hide PEM files.
+
+Wed Jan 16 03:54:28 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/control_expressions.rdoc:  Omit optional "then" for if and
+	  unless expressions.  Improved description of "a if a = 0.zero?"
+	  NameError.  Note that "do" for for loop is optional.
+
+Wed Jan 16 03:28:47 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/calling_methods.rdoc:  Link to defining methods.
+	* doc/syntax/methods.rdoc:  Link to calling methods, fixed typo.
+
+Wed Jan 16 03:15:00 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc:  Fixed link
+	* doc/syntax/modules_and_classes.rdoc:  Fixed link
+	* doc/syntax.rdoc:  Updated with links to the recently added pages
+
+Wed Jan 16 03:05:50 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/control_expressions.rdoc (redo Statement):  Added note
+	  about retry.
+	* doc/syntax/exceptions.rdoc:  Added retry statement
+
+Tue Jan 15 23:12:34 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/vpath.rb (VPath#list): default separator to PATH_SEPARATOR from
+	  configure.in for make, not same name constant of File for use in ruby.
+
+Tue Jan 15 22:30:04 2013  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ext/save-history.rb: identify rightly a status of a
+	  history file that already exists [Bug #7694]. Thanks Nobuhiro IMAI
+	  for this patch.
+
+Tue Jan 15 15:55:28 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/control_expressions.rdoc:  Added description of control
+	  expressions in ruby.
+
+Tue Jan 15 13:33:00 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc (Method Names):  Added method names including
+	  operator methods.
+	* doc/syntax/methods.rdoc (Return Values):  Added note that assignment
+	  methods ignore return values.
+	* doc/syntax/precedence.rdoc:  Added document describing precedence.
+
+Tue Jan 15 11:49:31 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc (Block Argument):  Added section on block
+	  argument.  Thanks to Andy Lindeman.
+
+Tue Jan 15 10:54:59 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/calling_methods.rdoc (Arguments):  Added improved
+	  introduction to arguments including passing style and lazy
+	  evaluation.  Thanks to Matt Aimonetti.
+	* doc/syntax/calling_methods.rdoc (Positional Arguments):  Added
+	  description for sending a message to a method with *arguments
+	* doc/syntax/calling_methods.rdoc (Default Positional Arguments):
+	  Added description.  Thanks to Andy Lindeman.
+	* doc/syntax/calling_methods.rdoc (Block Local Arguments):
+	  Added description of block locals.  Thanks to Justin Collins.
+	* doc/syntax/calling_methods.rdoc (Hash to Keyword Arguments):  Added
+	  section describing ** operator.  Thanks  to Justin Collins.
+
+Tue Jan 15 10:40:18 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* test_lazy_enumerator: Test that map & flat_map also require a block
+
+Tue Jan 15 09:22:47 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (gvl_init): Reset gvl.wait_yield explicitly when
+	  fork()ing. Patch by Apollon Oikonomopoulos. Thanks!
+	  [Bug #7693][ruby-core:51424]
+
+Tue Jan 15 09:27:56 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/calling_methods.rdoc (Receiver):  Added :: as pointed out
+	  by Tony Arcieri
+	* doc/syntax/calling_methods.rdoc (Block Argument):  Fixed { } block
+	  examples as pointed out by David Copeland.
+
+Tue Jan 15 09:10:29 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc (Array/Hash Argument):  Moved above Keyword
+	  Arguments
+	* doc/syntax/methods.rdoc (Keyword Arguments):  Described ** for
+	  gathering arbitrary keyword arguments.
+
+Tue Jan 15 08:56:37 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/calling_methods.rdoc:  Added document describing method
+	  calls.
+
+Tue Jan 15 07:39:21 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/top_level.rb:  Fixed extension trimming for page names in
+	  RDoc HTML output.  [ruby-trunk - Bug #7673]
+	* test/rdoc/test_rdoc_top_level.rb:  Test for above.
+
+Mon Jan 14 23:06:41 2013  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ext/save-history.rb: outputs its history to
+	  owner-only-readable file and change the permission of a file that
+	  already exists [Bug #7694]. Thanks Nobuhiro IMAI for bug reports.
+
+Mon Jan 14 17:12:48 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_flat_map_func): flat_map should call each only
+	  when the value of a block returns a forcable object.
+	  [ruby-core:51401] [Bug #7690]
+
+	* enumerator.c (lazy_flat_map): add documentation.
+
+	* test/ruby/test_lazy_enumerator.rb: related test.
+
+Mon Jan 14 16:42:28 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: Require block for Lazy#{take|drop}_while [Bug #7692]
+
+Mon Jan 14 14:41:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_s): use CRuby style.
+
+Mon Jan 14 14:39:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c: use `RB_TYPE_P(x, t)` instead of
+	  `TYPE(x) == t`.
+
+Mon Jan 14 10:18:56 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: Fix size for Enumerator::Lazy#flat_map
+
+Mon Jan 14 07:12:52 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix/lup_decomposition: Fix bugs with LUP Decomposition of
+	  rectangular matrices. [rubyspec:ba849801a85]
+
+Mon Jan 14 06:46:53 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (add_ctype_to_cc): don't check dup warn on adding
+	  negative ctype to cclass. [Bug #7471] [ruby-core:50344]
+
+Mon Jan 14 06:06:03 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix/eigenvalue_decomposition: Backport bugfix of Jama 1.0.3
+	  [rubyspec:df87040be371]
+
+Sun Jan 13 16:45:00 2013  Zachary Scott <zachary@zacharyscott.net>
+
+	* ext/psych/yaml/scanner.c: Fix typos, patch by James Dabbs
+	  [Github tenderlove/psych#118]
+
+Sun Jan 13 15:00:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_sub):
+	  need to specify precision for converting Rational and Float.
+	  [ruby-dev:46544] [Bug #7404]
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_mult): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_divide): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_DoDivmod): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_divremain): ditto.
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests for the above fixes.
+
+Sun Jan 13 14:48:55 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix/eigenvalue_decomposition: Fix eigensystem with complex
+	  eigenvectors. Patch by pypypy567.
+	  [Bug #7208] [ruby-dev:46251] [rubyspec:242f8e55bd]
+
+	* lib/matrix/lup_decomposition.rb: Fix error for rectangular matrices
+	  [bug#7620] [ruby-core:51118] [rubyspec:41f833ee2]
+
+Sun Jan 13 14:06:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb.rb, lib/prime.rb: Typos in overview
+	  Patch by Ershad K [Github Fixes #234]
+
+Sun Jan 13 13:40:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/open3.rb : Typo in Open3 overview. Patch by zed_0xff
+	  [Github Fixes #233]
+
+Sat Jan 12 17:42:00 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* numeric.c (do_coerce): fix for the exceptions which the coerce
+	  method raises.  The optimization done by r38756 is preserved.
+	  [Bug #7645] [ruby-core:51213]
+
+Sat Jan 12 16:12:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/setup.mak (-runtime-): see msvcrt from link header on mswin
+	  instead of running testing executable file, for cross compiling.
+
+Sat Jan 12 08:58:47 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: merge key values that
+	  contain something besides a hash should be left in tact.
+
+	* test/psych/test_merge_keys.rb: test for change
+
+Sat Jan 12 07:52:47 2013  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_set_byref): support VT_UI8|VT_BYREF,
+	  VT_I8|VT_BYREF in cygwin and mingw.
+
+	* ext/win32ole/win32ole.c (ole_variant2val): ditto.
+
+	* test/win32ole/test_win32ole_variant.rb (test_s_new_with_i8_byref):
+	  ditto.
+
+	* test/win32ole/test_win32ole_variant.rb (test_s_new_with_ui8_byref):
+	  ditto.
+
+Sat Jan 12 02:45:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* man/ruby.1 (options): include --*-encoding from r38784
+
+Fri Jan 11 23:34:48 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ruby.c (usage): sort --*-encoding in help. (same order of -E)
+
+Fri Jan 11 16:56:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb: use configured libdir value to fix
+	  --enable-load-relative on systems where libdir is not default value,
+	  overridden in config.site files.  [ruby-core:47267] [Bug #6903]
+
+Fri Jan 11 11:59:32 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (dispose_string): use rb_str_free for freeing string in
+	  parse.y. by Sokolov Yura <funny.falcon@gmail.com>
+	  https://github.com/ruby/ruby/pull/87 fix GH-87
+
+Fri Jan 11 09:56:22 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* insns.def (defineclass): private constants should not be accessed
+	  by scoped module definitions.  The bug was introduced in r38495.
+
+	* test/ruby/test_module.rb: related test.
+
+Fri Jan 11 02:11:59 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/rbconfig/obsolete.rb (respond_to_missing?): use send because
+	  RbConfig.respond_to_missing? is now private.
+
+Thu Jan 10 22:00:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (VM_DEBUG_BP_CHECK): set 0 as default.
+	  This flag specifies checking BP consistency on each frame popping.
+	  Now, we don't have any trouble on it, so I remove it.
+	  If you feel any bugs about VM execution, then set it to 1.
+
+Thu Jan 10 21:03:05 2013 TAKANO `takano32' Mitsuhiro <tak@no32.tk>
+
+	* cont.c: define FIBER_USE_NATIVE as 0 in ia64.
+
+Thu Jan 10 19:39:05 2013  TAKANO `takano32' Mitsuhiro <tak@no32.tk>
+
+	* thread.c: fix RB_GC_SAVE_MACHINE_REGISTER_STACK define for ia64.
+
+Thu Jan 10 17:45:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in, win32/Makefile.sub ($(MKFILES)): continue if Makefile
+	  unchanged.
+
+Thu Jan 10 16:31:20 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_super_method): raise a TypeError
+	  instead of a NotImplementedError if self is not an instance of the
+	  current class.  [ruby-dev:39772] [Bug #2402]
+
+Thu Jan 10 16:47:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/extconf.rb (find_tcltk_header): use have_header instead of
+	  try_cpp, which is incredibly slow with VC.
+
+Thu Jan 10 15:55:28 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* numeric.c (do_coerce): remove an unused variable.
+
+Thu Jan 10 15:35:55 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* tool/gen_dummy_probes.rb: remove comments in probes.d to fix the
+	  compilation error introduced by r38755.
+
+Thu Jan 10 11:15:04 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* numeric.c (do_coerce): speed optimization by using rb_check_funcall
+	  instead of rb_rescue + rb_funcall.
+	  This fix is based on the patch by Benoit Daloze.
+	  [Bug #7645] [ruby-core:51213]
+
+Thu Jan 10 11:15:04 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* probes.d: updating probes to be more symmetrical, adding
+	  documentation.
+
+	* load.c: ditto
+
+Thu Jan 10 04:23:07 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: strip trailing dots from
+	  floats so that Float() will not raise an exception.
+
+	* test/psych/test_numeric.rb: test to ensure "1." can be loaded
+
+	* test/psych/test_string.rb: make sure "1." can round trip
+
+Thu Jan 10 03:38:40 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ascii only binary strings
+	  will be dumped as unicode.  Thanks Paul Kunysch!
+
+	* test/psych/test_string.rb: appropriate test
+
+Thu Jan 10 03:29:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_array_): modify wrong optimization.
+	  A script "[print(1)]; print(2)" should output "12".
+	  However, the compiler had eliminated "[print(1)]" expression
+	  because it is void expression (unused array).
+	  Of course, side-effect should be remained.
+	  This issue is reported by Masaya Tarui.
+
+	* bootstraptest/test_literal.rb: add a test.
+
+Wed Jan  9 22:07:42 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* load.c (load_lock): if thread shield is destroyed and there is no
+	  waiting thread, insert new thread shield into load_table.
+	  [Bug #7530] [ruby-core:50645]
+
+Wed Jan  9 21:43:32 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* load.c (load_lock): revert r38744. it should acquire new thread
+	  shield.
+
+Wed Jan  9 15:40:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (th_init, ruby_thread_init): initialize root_svar with Qnil,
+	  since lep_svar_place() expects uninitialized svar to be nil, not 0.
+
+Wed Jan  9 13:20:23 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* test/ruby/test_require.rb: improve test for r38744.
+	  fix to use Tempfile instead of temporary file in current directory.
+	  the patch is from nobu (Nobuyoshi Nakada).
+
+Wed Jan  9 09:53:23 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* load.c (load_lock): fix not to delete thread shield twice.
+	  it may break the shield locked by another thread.
+	  [Bug #7530] [ruby-core:50645]
+
+	* test/ruby/test_require.rb: a test for above.
+
+Wed Jan  9 02:13:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RBasic): to be aligned on a VALUE size
+	  boundary.  [Bug #7647]
+
+Tue Jan  8 14:41:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_iseq_t): move flip_cnt from struct iseq_compile_data,
+	  because it has same life span as enclosing iseq.  [Bug #7671]
+	  [ruby-core:51296]
+
+Mon Jan  7 23:43:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (rmpd_power_by_big_decimal):
+	  add RB_GC_GUARD to prevent the intermediate object is GCed too early.
+	  This patch was made by Yusuke Endoh.  [Bug #7044] [ruby-core:47632]
+
+	* test/bigdecimal/test_bigdecimal.rb: add a reproduction test for
+	  the issue [Bug #7044]
+
+Mon Jan  7 21:40:36 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (Init_eval_method): main.public and main.private
+	  should be private.
+
+	* proc.c (Init_Proc): main.define_method should be private.
+
+	* test/ruby/test_module.rb: related test.
+
+Mon Jan  7 20:48:47 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (Init_eval): main.include should be private.
+	  [ruby-core:51293] [Bug #7670]
+
+	* test/ruby/test_module.rb (test_top_include_is_private): a new test
+	  for the above change.
+
+Mon Jan  7 20:29:50 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* NEWS: remove description about `require "refinement"'.
+
+Mon Jan  7 20:15:49 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (Init_eval): enable Refinements by default.
+	  [ruby-core:51286] [Bug #7667]
+
+	* eval.c (rb_mod_refine, top_using): show a warning when
+	  Module#refine or main.using is called at the first time.
+
+	* ext/refinement/*: removed the extension library "refinement".
+
+	* test/ruby/test_refinement.rb: fix for the above changes.
+
+Mon Jan  7 17:34:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h (RUBY_EVENT_SPECIFIED_LINE): make it special.
+	  This flag is not contained by RUBY_EVENT_TRACEPOINT_ALL.
+	  This event is experimental one. It is possible to remove/rename
+	  flag name after 2.0.1.
+
+	* vm_trace.c (get_event_id): return :line if SPECIFIED_LINE was
+	  occurred. `:specified_line' never been returned.
+
+Mon Jan  7 15:42:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_kwrest): allow bare kwrest_mark as valid syntax.  its
+	  semantics is still undefined.  [Bug #7662] [ruby-core:51269]
+
+Mon Jan  7 15:31:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_kwrest): reject duplicated kwrest argument name.
+
+Mon Jan  7 15:24:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks_orig): pop tag before
+	  JUMP_TAG() if frame is `finish' frame.
+	  Without this patch, there is an inconsistency between control
+	  frame stack and tags stack.
+	  [Bug #7668]
+
+	* test/ruby/test_settracefunc.rb: add a test for above.
+
+Mon Jan  7 15:21:48 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* Makefile.in, common.mk (fake, yes-fake, no-make): these dependencies
+	  are not platform dependent.
+
+	* win32/Makefile.sub ($(arch)-fake.rb): workaround.
+
+Mon Jan  7 12:09:24 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_arg_complex, vm_yield_setup_block_args):
+	  set keyrest hash after making rest array, so that the last element
+	  will not be overwritten.  [ruby-core:51278] [Bug #7665]
+
+Mon Jan  7 09:37:24 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: add a NEWS entry about RubyVM.
+
+Sun Jan  6 19:06:57 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* win32/Makefile.sub: Fix build with VC.
+	  Patch by Charlie Savage. Fixes [ruby-core:51261]
+
+Sun Jan  6 18:43:48 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* bootstraptest/test_io.rb: add a test for [ruby-dev:46834].
+
+	* io.c (rb_cloexec_fcntl_dupfd) Use an emulation with dup(2) when
+	  fcntl(2) and/or F_DUPFD is unavailable.
+	  Suggested by akr.
+
+	* configure.in (HAVE_FCNTL): NativeClient does not provide fcntl(2).
+
+Sun Jan  6 11:11:26 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/modules_and_classes.rdoc:  Fixed typo.
+
+Sun Jan  6 05:35:18 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/modules_and_classes.rdoc:  Added singleton classes
+	  documentation.
+
+Sun Jan  6 02:22:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/webrick/httpservlet/abstract.rb (WEBrick::HTTPServlet): Typo in
+	  example. Patch by shlensky [Fixes #232 on github]
+
+Sat Jan  5 21:15:10 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/generic_request.rb:
+	  Amazon ECA API and GTE/1.3 disallow requests whose host has port
+	  number if its port number equals to default port number of the
+	  scheme. [Bug #7650]
+
+Sat Jan  5 13:58:59 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/modules_and_classes.rdoc:  Improved description of methods
+	  on a module or class as suggested by Tobias Buhlmann
+
+Sat Jan  5 13:38:07 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_enumerate_lines): fix invalid byte sequence error
+	  when a separator is passed. The patch is from yoshidam (Yoshida
+	  Masato).
+	  [Bug #7646] [ruby-dev:46827]
+
+	* test/ruby/test_string.rb: a test for above.
+
+Sat Jan  5 12:25:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_in_out_err):
+	  check stdout and stderr both.
+
+Sat Jan  5 10:21:54 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/modules_and_classes.rdoc:  Added documentation of syntax
+	  for Modules and Classes.
+	* doc/syntax/methods.rdoc:  Moved some text to the Modules and
+	  Classes syntax document.
+
+Sat Jan  5 08:38:27 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc:  Added return values and scope sections,
+	  slightly modified from the original patch.  Fixes #227 from github by
+	  Dave Brown.
+
+Sat Jan  5 08:21:41 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_cloexec_fcntl_dupfd): improve #ifdef condition.
+	* io.c (rb_maygvl_fd_fix_cloexec): ditto.
+
+Sat Jan  5 07:54:59 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/cleanup_command.rb:  Clean all possible gems
+	  using multiple passes.  Fixes RubyGems bug #422.  Refactored for
+	  maintainability.
+	* test/rubygems/test_gem_commands_cleanup_command.rb:  Test for above.
+
+Sat Jan  5 05:04:39 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* gc.c (vm_xrealloc): add a few comment why we avoid realloc(ptr,0).
+
+Fri Jan  4 20:17:06 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* Makefile.in (RBCONFIG): Moved from common.mk in order to use the
+	  variable in Makefile.in.
+
+	* win32/Makefile.sub (RBCONFIG): Ditto.
+
+Fri Jan  4 19:45:50 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* common.mk (run, parse): Use BTESTRUBY instead of MINIRUBY to handle
+	  cross-compiling cases, e.g. NativeClient.
+
+Fri Jan  4 17:58:16 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* common.mk (yes-btest): btest depends on also $(arch)-fake.rb and
+	  rbconfig.rb on building for NativeClient.
+
+	* Makefile.in (fake): Avoid generating $(arch)-fake.rb unless cross
+	  compiling.
+
+	* configure.in (CROSS_COMPILING): New substitution.
+
+Fri Jan  4 16:26:45 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/doctor.rb:  Process directories in order in case the
+	  filesystem doesn't.  [ruby-trunk - Bug #7618]
+
+	  Process specifications before other directories in case of bugs.
+	* test/rubygems/test_gem_doctor.rb:  Test for above.
+
+	* lib/rubygems.rb:  Updated version.
+
+	* test/rubygems/test_require.rb:  Fixed double require of
+	  benchmark.rb.  RubyGems bug #420.
+
+	* test/rubygems/test_gem_commands_check_command.rb:  Fixed unused
+	  variable warnings.
+	* test/rubygems/test_gem_commands_query_command.rb:  ditto
+	* test/rubygems/test_gem_installer.rb:  ditto
+
+Fri Jan  4 15:05:25 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/cross_reference.rb:  Fixed matching of C#=== or #===.  RDoc
+	  bug #164
+	* test/rdoc/test_rdoc_cross_reference.rb:  Test for above.
+
+	* lib/rdoc/parser/changelog.rb:  Fixed parsing of dates.  RDoc bug #165
+	* test/rdoc/test_rdoc_parser_changelog.rb:  Test for above.
+
+	* lib/rdoc/parser.rb:  Fixed parsing multibyte files with incomplete
+	  characters at byte 1024.  [ruby-trunk - Bug #6393]
+	  Fixed handling of -E.  [ruby-trunk - Bug #6392]
+	* test/rdoc/test_rdoc_options.rb:  Test for above.
+	* test/rdoc/test_rdoc_parser.rb:  ditto.
+	* test/rdoc/test_rdoc_parser_c.rb:  ditto.
+	* test/rdoc/test_rdoc_parser_changelog.rb:  ditto.
+	* test/rdoc/test_rdoc_parser_markdown.rb:  ditto.
+	* test/rdoc/test_rdoc_parser_rd.rb:  ditto.
+	* test/rdoc/test_rdoc_rdoc.rb:  ditto.
+
+	* lib/rdoc/tom_doc.rb:  Fixed parsing of [] in TomDoc arguments list.
+	  RDoc bug #167
+	* test/rdoc/test_rdoc_tom_doc.rb:  Test for above.
+
+	* lib/rdoc.rb:  Update version.
+
+Fri Jan  4 11:51:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/forwardable.rb: Fix rdoc parameters for ::def_single_delegator.
+	  Patch by Vladimir Andrijevik [Github Fixes #230]
+
+Fri Jan  4 00:35:11 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	Fix failures on btest for NativeClient.
+	* bootstraptest/runner.rb (nacl?): New method to distinguish NaCl
+	  cross build.
+
+	* bootstraptest/test_io.rb: Skip unsupported operations.
+
+	* bootstraptest/test_literal.rb: ditto.
+
+Fri Jan  4 00:29:40 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* io.c (rb_cloexec_fcntl_dupfd): Fix failures in
+	  bootstrap_test/test_io.rb.  NativeClient does not support F_DUPFD
+	  but supports dup2(2).
+
+Thu Jan  3 17:46:50 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/element.rb (REXML::Elements#add): Remove too much
+	  "elements" in document. Sorry...
+
+Thu Jan  3 17:42:32 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/element.rb (REXML::Elements#each): Add missing
+	  "elements" in document. [ruby-talk:402713]
+	  Reported by Wesley Rishel. Thanks!!!
+
+Thu Jan  3 15:13:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/psych/lib/psych.rb (Psych.load): Return value of
+	  Psych::SyntaxError.message should be same as example.
+	  Patch by Ippei Obayashi [ruby-core:51193] [Bug #7636]
+
+Thu Jan  3 14:58:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/forwardable.rb (SingleForwardable): Fix example in overview
+	  Patch by Vladimir Andrijevik [Github Fixes #231]
+
+Thu Jan  3 14:32:47 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* configure.in (OBJCOPY):  Fixes build error for NativeClient.
+	  Avoid disabling OBJCOPY for NativeClient.
+
+	* thread_pthread.c (rb_reserved_fd_p): USE_SLEEPY_TIMER_THREAD is
+	  always defined.  Fixes compilation error for NativeClient.
+
+Wed Jan 02 03:09:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/zlib/zlib.c (Zlib::GzipReader): Fix typo by zed_0xff
+	  [Fixes Github #229]
+
+Wed Jan 02 02:29:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* hash.c (rb_hash_update): Revert documentation from r38672
+	  See: https://github.com/ruby/ruby/pull/228#issuecomment-11791013
+
+Wed Jan 02 02:16:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* hash.c (rb_hash_update): Documentation for Hash#merge and shallow
+	  copies Patch by Yorick Peterse [Fixes Github #228]
+
+Mon Dec 31 15:10:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm_backtrace.c: Add documentation for Kernel#caller_locations,
+	  Kernel#caller, and Thread::Backtrace::Location
+
+Mon Dec 31 13:05:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* test/ruby/test_backtrace.rb: Add test for r37957 [Feature #7434]
+
+Sun Dec 30 23:33:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (simple_re_meta): escaped closing parenthesis has different
+	  meaning.  [Bug #7610] [ruby-core:51088]
+
+Sun Dec 30 12:09:47 2012  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* configure.in: use 4 argument form of AC_CHECK_HEADERS to force
+	  autoconf to use compiler's result
+
+Sun Dec 30 10:58:04 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/ruby/test_keyword.rb: add a test for passing hash
+	  as a last argument. [ruby-dev:46712] [Bug #7529]
+
+Sun Dec 30 10:51:29 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_insnhelper.c: set keyword hash on Proc/block calls.
+	  [ruby-core:51172] [Bug #7630]
+
+	* test/ruby/test_keyword.rb: add tests for above.
+
+Sat Dec 29 21:57:11 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/completion.rb: treat rightly completion for symbol on irb
+	  [Bug #7632].
+
+Sat Dec 29 21:51:30 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/curses/curses.c (window_cury, window_curx, window_maxy,
+	  window_maxx, window_begy, window_begx): use RB_UNUSED_VAR()
+	  to suppress unused-but-set-variable warnings.
+
+Sat Dec 29 16:45:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* iseq.c (RubyVM::InstructionSequence): rdoc formatting
+
+Sat Dec 29 15:28:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* iseq.c (RubyVM::InstructionSequence):	Add rdoc for new iseq features
+	  added from r38085, this includes ::of, #path, #absolute_path,
+	  #label, #base_label, #first_lineno, and #inspect
+
+Sat Dec 29 14:06:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* iseq.c (rb_iseq_line_trace_all, rb_iseq_line_trace_specify): Add
+	  rdoc for experimental C level api of iseq, from r38076
+
+Sat Dec 29 11:37:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_clone): attach clone to its singleton class during
+	  cloning singleton class so that singleton_method_added will be called
+	  on it.  based on the patch by shiba (satoshi shiba)[Bug #5283] in
+	  [ruby-dev:44477].  [Bug #5283]
+
+Sat Dec 29 10:10:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (crt_externs.h): use standard macro AC_CHECK_HEADERS.
+
+Fri Dec 28 23:12:44 2012  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* configure.in: check for the whether crt_externs.h is present when compiling
+	  for darwin (this header is missing in the iOS SDK)
+	* eval_intern.h: check HAVE_CRT_EXTERNS_H before including crt_externs.h, if
+	  not defined, include missing/crt_externs.h instead
+	* hash.c: ditto
+	* missing/setproctitle.c: ditto
+	* missing/crt_externs.h: declare _NSGetEnviron() function and define environ
+	  for iOS
+
+Fri Dec 28 21:40:36 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/context.rb: IRB::Context#new: Check from JobManager
+	  inside IRB namespace [Bug #7628]. Thanks rafaelfranca for bug
+	  report and its patch.
+
+Fri Dec 28 17:06:17 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-curlies): Automatically
+	  indent closing curly brackets when
+	  ruby-electric-newline-before-closing-bracket is true.
+
+Fri Dec 28 11:50:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): pass single argument to
+	  single optional parameter unchanged without splatting.  [Bug #7621]
+	  [ruby-dev:46801]
+
+Fri Dec 28 11:17:47 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* proc.c (method_eq): fix the documentation to refer to owner.
+	  [ruby-core:51105] [Bug #7613]
+
+	* test/ruby/test_method.rb (test_alias_onwer): new test to confirm
+	  that `a == b' returns false if owners of a and b are different.
+
+Fri Dec 28 07:07:43 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* def/id.def: use split(/^/) instead of String#lines to support
+	  Ruby 1.8.5 as BASERUBY.
+
+Thu Dec 27 21:56:56 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* variable.c (rb_mod_remove_const): fix segv caused by r38558.
+
+Tue Dec 28 01:13:48 2012  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb: Added more Hash methods to CSV::Row.
+
+Thu Dec 27 23:27:15 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ruby-lex.rb: make lex_state to EXPR_END when next token
+	  is an operator after SYMBEG [Bug #6378].
+
+Thu Dec 27 21:30:21 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ruby-lex.rb: allow to handle recursive heredocs on
+	  irb[Bug #5648].
+
+Thu Dec 27 20:45:29 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* ext/stringio/stringio.c (strio_getline): fix not to raise TypeError
+	  when limit is nil.
+	  [Bug #7232] [ruby-core:48531]
+
+	* test/stringio/test_stringio.rb: a test for above.
+
+Thu Dec 27 21:08:23 2012  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* vm_core.h (VM_DEFINECLASS_TYPE): explicit cast to enum type to avoid 64->32
+	  shorten warning
+
+Thu Dec 27 20:11:29 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* ext/stringio/stringio.c (strio_ungetc): raise IOError instead of RuntimeError
+	  if the string is frozen.
+	  [Bug #7231] [ruby-core:48530]
+
+	* ext/stringio/stringio.c (strio_ungetbyte): ditto.
+
+	* test/stringio/test_stringio.rb: a test for above.
+
+Wed Dec 26 23:55:18 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/context.rb: fix IRB::Inspector#keys_with_inspector [Bug #7598]
+
+Wed Dec 26 23:26:15 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/context.rb: IRB::Context#use_readline= has been obsolete
+	  [Bug #6339].
+
+Wed Dec 26 21:32:46 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/context.rb: make a correct prompt from
+	  IRB.conf[:IRB_NAME] on irb [Bug #6338]. Patched by sho-h.
+
+Wed Dec 26 21:09:19 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ext/math-mode.rb: make not able to change math-mode
+	  after irb starting [Bug #6302]. Patched by sho-h.
+
+Wed Dec 26 12:52:36 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/init.rb: change default debug level for
+	  irb[ruby-dev:46805], [Bug #6301].
+
+Wed Dec 26 11:54:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: enable -fPIE when checking -pie for fixing
+	  OpenBSD build error. Patch by George Koehler. Thank you!
+	  [Bug #7606] [ruby-core:51082]
+
+Wed Dec 26 07:31:24 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_copy_for_substr): empty string is always
+	  valid or 7bit.
+
+	* string.c (rb_str_enumerate_lines, rb_str_chop): reduce duplicated
+	  code.
+
+	* string.c (rb_str_enumerate_chars): prevent shared copy from GC.
+
+Wed Dec 26 01:31:16 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/init.rb, lib/irb/context.rb: fix conf.debug_level=
+	  [Bug #6301] and fix irb command option: -- irb_debug_level for irb.
+
+Wed Dec 26 00:59:18 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ruby-lex.rb: improve RubyLex performance for large files
+	  [Bug #5202]. Patch by ryanmelt.
+
+Tue Dec 25 22:21:06 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/output-method.rb: raise right exception when
+	  IRB::OutputMethod#print don't defined [Bug #6657].
+
+Tue Dec 25 22:06:33 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks_and_pop_frame):
+	  pop a frame before JUMP_TAG() if exception occurred.
+	  This change fix bug of Ruby 1.9.
+	  [ruby-core:51128] [ruby-trunk - Bug #7624]
+
+	* vm_core.h (EXEC_EVENT_HOOK_AND_POP_FRAME): add to use
+	  `rb_threadptr_exec_event_hooks_and_pop_frame()'.
+
+	* vm.c (vm_exec): use EXEC_EVENT_HOOK_AND_POP_FRAME() while
+	  exception handling. While exception handling, if an exception
+	  is raised in hooks, need to pop current frame and raise this
+	  raised exception by hook.
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Tue Dec 25 21:08:53 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/init.rb, lib/irb/lc/ja/error.rb, lib/irb/lc/error.rb:
+	  raise exception when illegal RC_NAME_GENERATOR defined [Bug #6455].
+
+Tue Dec 25 19:22:17 2012  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/workspace.rb: define method to private on top-level irb
+	  [Bug #5776]. Patch by davidbalbert.
+
+Tue Dec 25 19:09:51 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c, include/ruby/intern.h (rb_big_eql): exported.
+
+	* thread.c (recursive_check): object_id maybe a Bignum, not Fixnum on
+	  LLP64.  see also r38493 and r38548.
+	  reported by Heesob Park at [ruby-core:51083] [Bug #7607], and patched
+	  by shirosaki at [ruby-core:51095]
+
+Tue Dec 25 18:53:35 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, eval_intern.h (CHECK_STACK_OVERFLOW): move
+	  CHECK_STACK_OVERFLOW() to vm_core.h and rename to
+	  CHECK_VM_STACK_OVERFLOW().
+	  This change is only move and rename.
+
+	* tool/instruction.rb: catch up above changes.
+
+	* vm.c, vm_insnhelper.c: ditto.
+
+	* vm_insnhelper.c (vm_stackoverflow): add a function to unify
+	  raising vm stackoverflow exception.
+
+Tue Dec 25 16:16:54 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (RUBY_VM_THREAD_VM_STACK_SIZE): change default
+	  VM stack size (128 KB or  256 KB -> 512 KB or 1024 KB).
+	  This re-sizing corrects smaller value introduced at r38478.
+	  Newer value is same VM stack size of Ruby 1.9.
+	  [ruby-dev:46797] [ruby-trunk - Bug #7603]
+
+Tue Dec 25 13:38:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (compile_err_append, compile_warn_print, warn_print): use
+	  rb_write_error_str() instead of writing to rb_stderr directly.
+
+	* io.c (rb_write_error_str): a stopgap measure not to unblock GVL.
+	  warning from require seems to still have race condition errors.
+
+Tue Dec 25 00:59:29 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.h (NODE_OP_CDECL), compile.c (iseq_compile_each),
+	  parse.y (stmt, arg): allow scoped constant op-assignment.
+	  [ruby-core:40154] [Bug #5449]
+
+Mon Dec 24 04:56:48 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/generic_request.rb (Net::HTTPGenericRequest):
+	  set content-length to zero on empty post requests
+	  by Gregory Ostermayr <gregory.ostermayr@gmail.com>
+	  https://github.com/ruby/ruby/pull/201 fix GH-201
+
+Sun Dec 23 19:09:16 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: rename methods:
+	    from Thread.async_interrupt_timing to Thread.handle_interrupt,
+	    from Thread.async_interrupted? to Thread.pending_interrupt?.
+	  Also rename option from `defer' to `never'.
+	  [ruby-core:51074] [ruby-trunk - Feature #6762]
+
+	* vm_core.c, thread.c: rename functions and data structure
+	  `async_errinfo' to `pending_interrupt'.
+
+	* thread.c: add global variables sym_immediate, sym_on_blocking and
+	  sym_never.
+
+	* cont.c, process.c, vm.c, signal.c: ditto.
+
+	* lib/sync.rb, lib/thread.rb: catch up this renaming.
+
+	* test/ruby/test_thread.rb: ditto.
+
+Sun Dec 23 17:57:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/profiler.rb (Profiler__::PROFILE_PROC, print_profile): store
+	  profile data per threads for concurrent-execution.
+	  [ruby-core:22046] [Bug #1152]
+
+	* lib/profiler.rb (Profiler__::Wrapper): support calling singleton
+	  methods of an instance of BasicObject.
+
+	* lib/profiler.rb (Profiler__::PROFILE_PROC): use TracePoint.
+
+Sun Dec 23 16:13:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/erb.rb: typos for ERB::new link
+
+Sun Dec 23 16:06:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/erb.rb: Document ERB::new trim_mode '-' for lines ending in -%>
+	  [ruby-core:51084] [Bug #7608]
+
+Sun Dec 23 15:25:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb/ruby-lex.rb: Add handling for %i and %I quoting to irb
+	  Patch by flori [ruby-core:49550] [Bug #7392] [Github Issue #157]
+
+Sun Dec 23 15:05:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_check_funcall_with_hook): rb_check_funcall with hook
+	  which is called before calling method_missing or target method.
+
+	* marshal.c (w_object, r_object0): use rb_check_funcall_with_hook
+	  instead of respond_to? and call.
+
+Sun Dec 23 14:52:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* re.c (rb_reg_eqq): doc: #=== is not a synonym for #=~, added example
+	  [ruby-dev:46746] [Bug #7571]
+
+Sun Dec 23 14:35:13 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (BLOCKING_REGION): if fail_if_interrupted is false ignore
+	  the result of blocking_region_begin(), since it always is true in
+	  that case.  suppress "uninitialized" warnings.
+
+Sun Dec 23 09:34:07 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/check_command.rb:  Added --doctor and --dry-run
+	  options to clean up after failed uninstallation.
+	* test/rubygems/test_gem_commands_check_command.rb:  Test for above.
+
+	* lib/rubygems/commands/push_command.rb:  Allow pushes from RubyGems
+	  2.0.0.preview3
+
+	* lib/rubygems/commands/update_command.rb:  Use Gem.ruby_version
+
+	* lib/rubygems/dependency.rb:  Update style.
+
+	* lib/rubygems/installer.rb:  Ensure installed gem specifications will
+	  be useable.  Refactor.
+	* test/rubygems/test_gem_installer.rb:  ditto.
+
+	* lib/rubygems/validator.rb:  Fixed bug with unreadable files.
+
+	* lib/rubygems.rb:  Fixed broken methods.
+	* test/rubygems/test_gem.rb:  Test for above.
+
+	* test/rubygems/test_gem_commands_push_command.rb:  Fixed overridden
+	  Gem.latest_rubygems_version
+
+Sun Dec 23 01:52:01 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* io.c (rb_io_lines, rb_io_bytes, rb_io_chars, rb_io_codepoints):
+	  Deprecate IO#{lines,bytes,chars,codepoints} and those of ARGF.
+	  [Feature #6670]
+
+	* ext/stringio/stringio.c (strio_lines, strio_bytes, strio_chars)
+	  (strio_codepoints): Deprecate
+	  StringIO#{lines,bytes,chars,codepoints}. [Feature #6670]
+
+	* ext/zlib/zlib.c (rb_gzreader_lines, rb_gzreader_bytes):
+	  Deprecate Zlib::GzipReader#{lines,bytes}. [Feature #6670]
+
+Sat Dec 23 01:35:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/optparse.rb: Documentation for OptionParser to remove 'shadowed
+	  outer local variable' from example and make obvious ARGV with
+	  non-option arguments.
+	  Patch by Marcus Stollsteimer [ruby-core:47460] [Bug #6997]
+
+Sat Dec 23 00:08:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* include/ruby/intern.h: add the prototype declaration of
+	  rb_num_coerce_bit.
+
+	* numeric.c (rb_num_coerce_bit): the new coerce function for bitwise
+	  binary operation.
+
+	* bignum.c (rb_big_and): use coerce to convert the argument, which isn't
+	  a Fixnum nor a Bignum, to the corresponding Integer object so that
+	  bitwise operations can support Integer-mimic objects.
+	  [Bug #1792] [ruby-core:39491]
+
+	* bignum.c (rb_big_or): ditto.
+
+	* bignum.c (rb_big_xor): ditto.
+
+	* numeric.c (bit_coerce): ditto.
+
+	* numeric.c (fix_and): ditto.
+
+	* numeric.c (fix_or): ditto.
+
+	* numeric.c (fix_xor): ditto.
+
+	* test/ruby/test_integer.rb: add tests for the above changes.
+
+	* test/ruby/test_bignum.rb: ditto.
+
+Sun Dec 23 00:04:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (QUOTE, QUOTE_ID): quote unprintable chars in strings and
+	  IDs. [Bug #7574] [ruby-dev:46749]
+
+	* string.c (rb_str_quote_unprintable): ditto.
+
+Sat Dec 22 23:59:18 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_compile_error, rb_compile_warn, rb_compile_warning),
+	  (rb_warn, rb_warning): support PRIsVALUE.
+
+Sat Dec 22 22:04:58 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* cont.c (rb_fiber_start): unify conditions.
+
+Sat Dec 22 21:47:55 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_wait_writable): use rb_thread_check_ints() instead
+	  of rb_thread_fd_writable().
+	* io.c (rb_io_wait_readable): ditto.
+
+Sat Dec 22 20:31:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_get): symbol cannot be nested constant name.
+
+Sat Dec 22 19:26:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_get): check more strictly.  [ruby-dev:46748]
+	  [Bug #7573]
+
+Wed Dec 19 02:34:48 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* cont.c (rb_fiber_start): in case of jump with TAG_FATAL,
+	  enqueue error into async_errinfo_queue, because you cannot call
+	  TH_TAG_JUMP() in this function. [ruby-dev:45218]  [Bug #5993]
+
+	* thread.c (rb_threadptr_execute_interrupts): now INT2FIX(TAG_FATAL)
+	  can be popped from async_errinfo_queue.
+
+	* vm.c (rb_vm_make_jump_tag_but_local_jump): revert r38441.
+	  rb_vm_make_jump_tag_but_local_jump() shouldn't return exception
+	  in case of state == TAG_FATAL.
+
+	* test/ruby/test_fiber.rb (test_exit_in_fiber): fix a test to illuminate
+	  Thread.exit should terminate current Thread.
+
+Sat Dec 22 13:15:08 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* gc.c (obj_id_to_ref): add a macro to treat Bignum object id.
+	  This follows the change r38493.
+
+	* gc.c (id2ref): fix for working fine with Bignum object id on x64
+	  Windows.
+	* gc.c (wmap_finalize): ditto.
+
+Sat Dec 22 11:30:21 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* struct.c (make_struct): remove junk ID check to allow members who
+	  have junk name like "foo\000".
+	* test/ruby/test_struct.rb: Test for above.
+	  [Bug #7575] [ruby-dev:46750]
+
+Sat Dec 22 05:34:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Requests may be created with a URI which sets the
+	  Host header.  Responses contain the requested URI for easier redirect
+	  following.  [ruby-trunk - Feature #6482]
+	* lib/net/http/generic_request.rb:  ditto.
+	* lib/net/http/response.rb:  ditto.
+	* NEWS (net/http):  Updated for above.
+	* test/net/http/test_http.rb:  Tests for above.
+	* test/net/http/test_http.rb:  ditto.
+	* test/net/http/test_httpresponse.rb:  ditto.
+
+Sat Dec 22 02:35:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb/slex.rb(#match): Typo, should be D_DETAIL
+	  [ruby-core:51071] [Bug#7600]
+
+Sat Dec 22 02:29:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb/input-method.rb, lib/irb.rb: Typo in
+	  InputMethod#readable_atfer_eof? to #readable_after_eof?
+	  [ruby-core:51069] [Bug #7599]
+
+Sat Dec 22 02:19:38 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_dump.c (rb_vm_bugreport): revert r38533.
+	* addr2line.c (fill_lines): add ELF sanity check.
+	  [Bug #7597] [ruby-dev:46786]
+
+Sat Dec 22 02:05:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb/inspector.rb, lib/irb/context.rb: Move IRB::INSPECTORS and
+	  class methods to IRB::Inspector [ruby-core:51067][Bug #7598]
+
+Sat Dec 22 00:28:46 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* object.c (rb_obj_hash): shouldn't assume object_id can be long.
+	  based on a patch by Heesob Park at [ruby-core:51060].
+	  cf. [Backport #7454]
+
+Fri Dec 21 23:15:25 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* ext/fiddle/lib/fiddle/struct.rb (Fiddle::CStructEntity#set_ctypes):
+	  CPtr -> Pointer.
+	* test/fiddle/test_c_struct_entry.rb
+	  (Fiddle::TestCStructEntity#test_aref_pointer):
+	  Added the test for the above.
+
+Fri Dec 21 23:12:05 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* ext/fiddle/lib/fiddle/struct.rb (Fiddle::CStructEntity#set_ctypes):
+	  CPtr -> Pointer.
+	* test/fiddle/test_c_struct_entry.rb
+	  (Fiddle::TestCStructEntity#test_aref_pointer_array):
+	  Added the test for the above.
+
+Fri Dec 21 22:43:36 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* ext/fiddle/lib/fiddle/import.rb (Fiddle::Importer#sizeof):
+	  CPtr -> Pointer.
+	* test/fiddle/test_import.rb (Fiddle::TestImport#test_sizeof):
+	  Added the test for the above.
+
+Fri Dec 21 22:34:17 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_iseq.rb: disable a test which checks features
+	  removed at r38532.
+
+Fri Dec 21 22:02:00 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{dir.h,win32.c} (rb_w32_readdir): removed old rb_w32_readdir()
+	  and renamed from rb_w32_readdir_with_enc().
+	  [ruby-core:24864] [Feature #1927]
+
+	* dir.c (READDIR): follow above change.
+
+Fri Dec 21 21:12:54 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): commentout addr2line call temporarily
+	  in order to avoid segv. anyone can fix addr2line?
+	  [Bug #7597] [ruby-dev:46786]
+
+Fri Dec 21 20:38:28 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (Init_ISeq): remove definition of the following methods:
+	  ISeq#line_trace_all and ISeq#line_trace_specify because they are
+	  half baked.
+	  C APIs are remained as experimental. These functions will be
+	  renamed, removed their parameters may be changed.
+	  You can use these methods by C exts. Please give us your comments.
+
+Fri Dec 21 20:21:04 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (tracepoint_new): add code to support specified thread.
+	  But not tested and this feature is not supported officially.
+
+Fri Dec 21 19:37:15 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.c (process_options): need to acquire env from TOPLEVEL_BINDING
+	  each time.
+	  `bind->env' may update after `eval()'.
+	   [Bug #7536]
+
+Fri Dec 21 18:46:50 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/debug.h, vm_core.h: define rb_trace_arg_t at
+	  include/ruby/debug.h (move from vm_core.h).
+
+Fri Dec 21 17:48:15 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, vm_trace.c: fix multi-threading bug for tracing.
+	  Move `trace_arg' from rb_tp_t::trace_arg to rb_thread_t::trace_arg.
+	  `trace_arg' may changed by multiple threads.
+	  rb_thread_t::trace_arg can represent rb_thread_t::trace_running
+	  (null or non-null) and rb_thread_t::trace_running is removed.
+	  After that, `rb_tp_t' is not needed to check tracing or not
+	  (A running thread knows tracing or not). This is why I remove
+	  tp_attr_check_active() and make new function get_trace_arg().
+
+	  And this modification disable to work the following code:
+	    TracePoint.trace{|tp|
+	      Thread.new{p tp.event} # access `tp' from other threads.
+	    }
+	  I believe nobody mix threads at trace procedure.
+	  This is current limitation.
+	  [Bug #7590]
+
+	* cont.c (fiber_switch, rb_cont_call): use rb_thread_t::trace_arg
+	  instead of rb_thread_t::trace_running.
+
+	* test/ruby/test_settracefunc.rb: add a multi-threading test.
+
+Fri Dec 21 16:38:08 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/id.h.tmpl (ID2ATTRSET): compile time constant macro for
+	  ID_ATTRSET.
+
+	* defs/id.def (KeywordError): check duplication.
+
+	* defs/id.def: support for other scope IDs,
+	  ID_{INSTANCE,GLOBAL,CONST,CLASS}.
+
+Fri Dec 21 14:45:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb.rb, lib/irb/*: Documentation for IRB
+
+Fri Dec 21 11:31:02 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake/*:  Updated to rake 0.9.6
+	* doc/rake/*:  ditto
+	* test/rake/*:  ditto
+
+Fri Dec 21 08:56:34 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* vm_trace.c (rb_suppress_tracing): remove unused variable 'vm_tracing'
+
+Fri Dec 21 01:01:45 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* lib/irb/completion.rb (CompletionProc): support completion of
+	  instance variables. [ruby-dev:46710] [Bug #7520]
+
+Thu Dec 20 20:58:25 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* vm_trace.c (rb_suppress_tracing): bugfix for vm->trace_running
+	  counter. And if tracing is already true, vm_trace_running ops is
+	  skipped to control overflow.
+
+Thu Dec 20 18:29:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RTEST, NIL_P): make bare expressions without
+	  outermost parentheses.
+
+Thu Dec 20 17:29:00 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* NEWS: fix the description for Refinements.
+
+Thu Dec 20 16:53:59 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_core.h (rb_vm_defineclass_type_t),
+	  compile.c (iseq_compile_each), insns.def (defineclass): change the
+	  meaning of the third operand of defineclass as follows:
+	    lower 3bits: the type of the defineclass
+	                   0 = class, 1 = singleton class, 2 = module
+	    4th bit: a flag represents whether the defineclass is scoped
+	               0 = not scoped (e.g., class Foo)
+		       1 = scoped (e.g., class Bar::Baz)
+	    5th bit: a flag represents whether the superclass is specified
+	               0 = not specified (e.g., class Foo)
+	               1 = specified (e.g., class Bar < Foo)
+	  If the superclass is specified and is not a class, a TypeError
+	  should be raised.  [ruby-dev:46747] [Bug #7572]
+
+	* test/ruby/test_class.rb: related test.
+
+Thu Dec 20 16:52:37 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* NEWS: announce AEAD encryption support in the OpenSSL extension.
+
+Thu Dec 20 16:40:13 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (nonspecial_obj_id): VALUE is not compatible with Fixnum on
+	  LLP64 platform, such as 64bit Windows.
+	  reported by Heesob Park at [ruby-core:50255] [Bug #7454], and the
+	  fix is suggested by akr.
+
+Thu Dec 20 16:39:04 2012  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/ossl_cipher.c: fix errors for installations that do not
+	  feature Authenticated Encryption.
+	* ext/openssl/extconf.rb: detect presence of EVP_CTRL_GCM_GET_TAG to
+	  determine whether Authenticated Encryption can be used.
+	  [Feature #6980] [ruby-core:47426]
+
+Thu Dec 20 15:55:46 2012  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/ossl.c: do not use FIPS_mode_set if not available.
+	* test/openssl/utils.rb: revise comment about setting FIPS mode to
+	  false.
+	* test/openssl/test_fips.rb: remove tests that cause errors on
+	  ruby-ci.
+	  [Feature #6946] [ruby-core:47345]
+
+Thu Dec 20 15:22:59 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parser/ruby.rb:  Ignore methods defined on constants to
+	  prevent modules with the names of constants from appearing in the
+	  documentation.
+	* test/rdoc/test_rdoc_parser_ruby.rb:  Test for the above.
+
+Thu Dec 20 15:00:33 2012  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/ossl_cipher.c: add support for Authenticated Encryption
+	  with Associated Data (AEAD) for OpenSSL versions that support the
+	  GCM encryption mode. It's the only mode supported for now by OpenSSL
+	  itself. Add Cipher#authenticated? to detect whether a chosen mode
+	  does support Authenticated Encryption.
+	* test/openssl/test_cipher.rb: add tests for Authenticated Encryption.
+	  [Feature #6980] [ruby-core:47426] Thank you, Stephen Touset for
+	  providing a patch!
+
+Thu Dec 20 12:56:53 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup/to_html.rb (class RDoc):  Added current heading and
+	  top links to headings.
+	* lib/rdoc/generator/template/darkfish/rdoc.css:  ditto
+	* test/rdoc/test_rdoc_generator_markup.rb:  Test for above
+	* test/rdoc/test_rdoc_markup_to_html.rb:  ditto
+
+	* test/rdoc/test_rdoc_comment.rb:  Removed trailing whitespace.
+
+Thu Dec 20 11:05:26 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_valid_syntax): move from
+	  test_syntax.rb.
+
+	* test/ruby/envutil.rb (assert_normal_exit): validate syntax before
+	  running because this assertion passes even if the code fails by
+	  SyntaxError.
+
+Thu Dec 20 10:29:58 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkey_dh.rb: revert special treatment of
+	  FIPS-capable installations since FIPS mode is now disabled for the
+	  tests.
+
+Thu Dec 20 10:23:12 2012  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/ossl.c: add OpenSSL.fips_mode= to allow enabling FIPS
+	  mode manually.
+	* test/openssl/utils.rb: turn off FIPS mode for tests. This prevents
+	  OpenSSL installations with FIPS mode enabled by default from raising
+	  FIPS-related errors during the tests.
+	* test/openssl/test_fips.rb: add tests for FIPS-capable OpenSSL
+	  installations.
+	  [Feature #6946] [ruby-core:47345]
+
+Thu Dec 20 06:59:52 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: support variable VM/Machine stack sizes.
+	  Specified by the following environment variables:
+	    - RUBY_THREAD_VM_STACK_SIZE: vm stack size used at thread creation.
+	      default: 128KB (32bit CPU) or 256KB (64bit CPU).
+	    - RUBY_THREAD_MACHINE_STACK_SIZE: machine stack size used at thread
+	      creation. default: 512KB or 1024KB.
+	    - RUBY_FIBER_VM_STACK_SIZE: vm stack size used at fiber creation.
+	      default: 64KB or 128KB.
+	    - RUBY_FIBER_MACHINE_STACK_SIZE: machine stack size used at fiber
+	      creation. default: 256KB or 256KB.
+	  This values are specified at launched timing. You can not change
+	  these values at running time.
+	  Environ variables are only *hints* because:
+	    - They are aligned to 4KB.
+	    - They have minimum values (depend on OSs).
+	    - Machine stack settings are ignored by some OSs.
+	  Default values especially fiber stack sizes are increased.
+	  This change affect Fiber's behavior:
+	    (1) You can run more complex program on a Fiber.
+	    (2) You can not make many (thousands) Fibers because of
+	        lack of address space (on 32bit CPU).
+	  If (2) bothers you,
+	    (a) Use 64bit CPU with big memory, or
+	    (b) Specify RUBY_FIBER_(VM|MACHINE)_STACK_SIZE correctly.
+	  You need to choose correct stack size carefully. These values
+	  are completely rely on systems (OS/compiler and so on).
+	  [Feature #4614], [Bug #7212]
+
+	* vm_core.h (rb_vm_t::default_params): add to record above settings.
+
+	* vm.c (RubyVM::DEFAULT_PARAMS): add new constant to see
+	  above setting.
+
+	* thread_pthread.c: support RUBY_THREAD_MACHINE_STACK_SIZE.
+
+	* cont.c: support RUBY_FIBER_(VM|MACHINE)_STACK_SIZE.
+
+	* test/ruby/test_fiber.rb: add tests for above.
+
+	* test/ruby/test_thread.rb: ditto.
+
+Thu Dec 20 06:25:44 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_fiber.rb: remove a strange single quote character.
+	  With this character, this script exits by SyntaxError.
+
+Thu Dec 20 01:03:00 2012  Zachary Scott  <zachary@zacharyscott>
+
+	* ext/.document: Add missing ext modules to .document
+	  Patch by Ryunosuke SATO [Fixes Github #224]
+
+Wed Dec 19 23:52:16 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (load_file_internal): use original C string as the filename
+	  for parser.
+	  reported by whiteleaf at [ruby-list:49085] [ruby-dev:46738]
+	  [Bug #7562]
+
+Wed Dec 19 23:36:12 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* marshal.c (marshal_dump, marshal_load): fix SEGV during make rdoc
+	  and test failure in TestMarshal#test_gc and test_context_switch
+	  on SPARC Solaris 10 compiled with Oracle Solaris Studio 12.3.
+	  [Bug #7591] [ruby-dev:46772]
+
+Wed Dec 19 19:34:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_get): nul byte is invalid as constant name.
+
+Wed Dec 19 17:54:18 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks): get rid of race
+	  condition. [Bug #7589] [ruby-dev:46763]
+
+Wed Dec 19 16:30:28 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/literals.rdoc:  Added 0d decimal format.  Thanks Nobu!
+
+Wed Dec 19 16:19:36 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc:  Fixed typo.  Thanks to Josh Susser.
+
+Wed Dec 19 16:18:22 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/query_command.rb:  Refactored to improve
+	  maintainability.
+	* test/rubygems/test_gem_commands_query_command.rb:  Note default gems
+	  in gem list details.
+
+	* lib/rubygems/uninstaller.rb:  Detect all gems for uninstallation.
+	  This allows duplicate installs of default gems to be removed.
+	* lib/rubygems/specification.rb:  Allow use of ::each_spec.
+	* lib/rubygems/test_case.rb:  Added install_default_gems.
+	* test/rubygems/test_gem_commands_uninstall_command.rb:  Moved test
+	  down to the uninstaller tests.
+	* test/rubygems/test_gem_uninstaller.rb:  Test for uninstallation of
+	  default gems and duplicate default gems.
+
+Wed Dec 19 15:23:50 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc:  Add () around keyword arguments example for
+	  consistency.  Thanks to Josh Susser.
+
+Wed Dec 19 01:51:24 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* vm.c (rb_vm_jump_tag_but_local_jump): remove unnecessary 2nd
+	  argument.
+
+	* load.c (rb_load_internal): ditto.
+
+	* eval_intern.h (rb_vm_jump_tag_but_local_jump): ditto.
+
+Tue Dec 18 18:57:58 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_wait_writable): don't use rb_thread_wait_fd()
+	  because it is for waiting until io readable.
+
+	* io.c (rb_io_wait_writable): always use rb_thread_fd_writable()
+	  instead of bare rb_wait_for_single_fd(). we shouldn't ignore
+	  return value.
+	* io.c (rb_io_wait_readable): ditto. always use rb_thread_wait_fd().
+
+Tue Dec 18 18:55:33 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_wait_fd_rw): fix infinite loop bug.
+	  rb_wait_for_single_fd() never return positive number.
+
+Tue Dec 18 17:24:40 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/encoding.rb:  Do not remove #! line from document when
+	  setting encoding.  This allows ruby executables to be parsed as ruby
+	  files.
+	* test/rdoc/test_rdoc_encoding.rb:  Test for above.
+
+	* lib/rdoc/parser.rb:  Set the parser file name of ruby executables
+	  correctly.
+	* test/rdoc/test_rdoc_parser.rb:  Test for above.
+
+Tue Dec 18 16:46:15 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/literals.rdoc:  Used simplified heredoc example that
+	  doesn't include method definition.  Added heredoc with backticks.
+
+Tue Dec 18 16:38:51 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/store.rb:  Work around RDoc stores from older versions of
+	  RDoc.
+	* test/rdoc/test_rdoc_store.rb:  Test for above.
+
+Tue Dec 18 16:31:20 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ruby_lex.rb:  Return a TkHEREDOC instead of a TkSTRING when
+	  the heredoc identifier is followed by a line-end.  This allows proper
+	  display of some HEREDOCs in source view.
+	* lib/rdoc/ruby_token.rb:  Added TkHEREDOC
+	* test/rdoc/test_rdoc_ruby_lex.rb:  Test for above.
+
+Tue Dec 18 09:45:14 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* vm.c (rb_vm_make_jump_tag_but_local_jump): take care of the case
+	  TAG_JUMP() with TAG_FATAL (ex. rb_fatal()). [ruby-core:50917]
+	  [Bug #7570]
+
+	* test/ruby/test_fiber.rb (test_fatal_in_fiber): add a test for above.
+
+	* ext/-test-/fatal/extconf.rb, ext/-test-/fatal/rb_fatal.c: ditto.
+
+Tue Dec 18 13:17:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm_trace.c (tracepoint_attr_defined_class): Clean up rdoc for
+	  TracePoint#defined_class
+
+Tue Dec 18 12:15:59 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/specification.rb:  Fixed ruby output of requirements
+	  with multiple version specifiers.
+	* test/rubygems/test_gem_ext_cmake_builder.rb:  Only look for specific
+	  lines in cmake output.  Should fix [ruby-trunk - Bug #7579]
+
+Tue Dec 18 11:45:26 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/literals.rdoc:  Added 0o octal integers.
+
+Tue Dec 18 12:28:52 2012  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* test/openssl/test_ssl.rb: Use :TLSv1_2_client explicitly in
+	  test_tls_v1_2 to prevent upstream bug.
+	  [Bug #7197] [ruby-dev:46240]
+
+Tue Dec 18 11:52:34 2012  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/lib/ssl.rb: Enable insertion of empty fragments as a
+	  countermeasure for the BEAST attack by default. The default options
+	  of OpenSSL::SSL:SSLContext are now:
+	  OpenSSL::SSL::OP_ALL & ~OpenSSL::SSL::OP_DONT_INSERT_EMPTY_FRAGMENTS
+	  [Bug #5353] [ruby-core:39673]
+
+	* test/openssl/test_ssl.rb: Adapt tests to new SSLContext default.
+
+	* NEWS: Announce the new default.
+
+Tue Dec 18 06:36:12 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: remove `VM_METHOD_TYPE_CFUNC_FRAMELESS' method type.
+	  This method type is for optimized CFUNC such as Fixnum#+ and so on.
+	  This feature is half-baked and no way to use them.
+	  [Background]
+	  Now, VM has opt_plus instructions to optimize `+' methods for
+	  some Classes (such as Fixnum, Float (flonum)). We call this
+	  type of instructions as `specialized instructions'.
+	  This simple technique improve simple program dramatically.
+	  However, we can make specialized instructions for only several
+	  types (classes) and selectors (method names) because a large
+	  instruction will be slow. In other words, this technique has no
+	  extensibility.
+	  To overcome this problem, VM_METHOD_TYPE_CFUNC_FRAMELESS was
+	  introduced (r37198). This type is a variant of CFUNC, but called
+	  their functions directly without building a method frame.
+	  Any CFUNC method can be defined as frameless methods if a method
+	  is not needed to make method frame. Frameless methods are faster
+	  as specialized instructions (a bit slower, but no need to care).
+	  No problem described at
+	  http://charlie.bz/blog/why-do-singleton-methods-make-ruby-slow
+	  because this technique doesn't see class, but see method body
+	  itself. Alias is also no problem.
+	  [Problem]
+	  However, we can't set frameless method type for polymorphic methods
+	  such as Array#[]. Necessity for method frame depends on which
+	  parameter type. For example, Fixnum#+ needs method frame if
+	  coerce is needed. Current VM_METHOD_TYPE_CFUNC_FRAMELESS is not
+	  flexible and need more tuning to introduce it.
+	  Expected behavior of frameless method type may be:
+	    result = optimized_cfunc(params); /* call optimized cfunc */
+	    if (result == Qundef) { result = normal_cfunc(); }
+	  This is why I say this feature is half-baked.
+	  We need to learn primitive method in Smalltalk more.
+	  (I heard this name at RubyConf Taiwan this month. Thanks!)
+	  [Conclusion]
+	  I remove this feature and consider again.
+	  This feature goes to next minor (2.1?).
+	  Nobody may use this feature and there is no compatibility issue.
+
+	* proc.c (rb_method_entry_arity): ditto.
+
+	* vm_eval.c, vm_insnhelper.c, vm_method.c: ditto.
+
+Tue Dec 18 04:58:22 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (fill_id_and_klass): TracePoint#defined_class returns
+	  singleton class. `set_trace_func' passed attached class (which is
+	  attached/modified by singleton class) by 6th block parameter if it
+	  is singleton class. Previous behavior follows this spec.
+	  However, this method named `defined_class' should return singleton
+	  class directly because singleton methods are defined in singleton
+	  class. There are no compatible issue because TracePoint is introduced
+	  after 2.0.
+	  But compatibility with `set_trace_func' is broken. This means that
+	  you can not replace all `set_trace_func' code with TracePoint
+	  without consideration of this behavior.
+	  [Bug #7554]
+
+	* test/ruby/test_settracefunc.rb: change a test to catch up
+	  an above change.
+
+Tue Dec 18 03:03:10 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: speed up node mapping so
+	  common cases are evaluated first.  Thanks Kevin Menard!
+
+Tue Dec 18 02:35:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/optparse.rb: Remove 'developer documentation' section from rdoc
+	  Patch by Marcus Stollsteimer [ruby-core:50526][Bug #7504]
+
+Tue Dec 18 02:35:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/matrix.rb (#lup): typo in example [ruby-core:50946][Bug #7582]
+
+Mon Dec 17 18:03:34 2012  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* class.c (rewrite_cref_stack, clone_method): rewrite a method's cref
+	  stack when cloning into a new class to allow lexical const lookup to
+	  work as expected [ruby-core:47834] [Bug #7107]
+	* test/ruby/test_class.rb (class TestClass): related test
+
+Mon Dec 17 13:56:55 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_flush_buffer_sync2): avoid to return 0. because
+	  rb_thread_call_without_gvl2 uses 0 internally.
+	* io.c (io_flush_buffer_async2): adapt the above.
+
+Mon Dec 17 12:05:32 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/methods.rdoc:  Added a description of singleton methods.
+
+Mon Dec 17 11:35:57 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/.document:  Added doc/syntax
+
+Mon Dec 17 11:25:32 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax.rdoc:  Added syntax guide table of contents
+	* doc/syntax/exceptions.rdoc:  Syntax guide for exceptions
+	* doc/syntax/literals.rdoc:  Syntax guide for literals
+	* doc/syntax/methods.rdoc:  Syntax guide for methods
+
+Mon Dec 17 07:59:40 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Updated VERSION
+
+	* test/rubygems/test_gem_installer.rb:  Fixed ambiguous first argument
+	  warning.
+
+	* test/rubygems/test_gem_rdoc.rb:  RDoc generation depends on installed
+	  version of RDoc.
+
+Sun Dec 16 02:04:41 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (rb_sigaltstack_size): cast sysconf() return value
+	  explicitly. Fix compile error on Mac OS X.
+
+Sun Dec 16 00:39:43 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* cont.c (rb_fiber_start): don't enqueue Qnil to async_errinfo_queue.
+	  rb_vm_make_jump_tag_but_local_jump() could return Qnil (ex. when
+	  finished by Thread.exit). [ruby-dev:45218] [Bug #5993]
+
+	* test/ruby/test_fiber.rb (test_exit_in_fiber): add test for it.
+
+Sat Dec 15 23:56:51 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/fiddle/pointer.c (rb_fiddle_ptr2cptr): fix error message
+	  forgotten to be changed from DL to Fiddle.
+
+Sat Dec 15 23:14:32 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (default_handler): remove rb_register_sigaltstack()
+	  call. sigaltstack was already registered when creating threads.
+
+Sat Dec 15 23:08:56 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (rb_sigaltstack_size): new. calculate stack size for
+	  sigsegv handler. enlarge value when x86 or x86_64 on Linux.
+	  Linux has very small MINSIGSTKSZ size (2048 bytes) and
+	  our sigsegv routine need 5KiB at least. [Bug #7141]
+	* internal.h: add declaration of rb_sigaltstack_size().
+	* vm_core.h: remove ALT_STACK_SIZE definition.
+
+	* signal.c (rb_register_sigaltstack): replace ALT_STACK_SIZE with
+	  rb_sigaltstack_size();
+	* gc.c (Init_heap): ditto.
+	* vm.c (th_init): ditto.
+
+Sat Dec 15 18:24:21 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (f_round_common): should check overflow.
+
+Sat Dec 15 18:00:00 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (float_rationalize): reduced.
+
+Sat Dec 15 14:18:44 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (finish_writeconv): uses rb_write_internal2 if
+	  fptr->write_lock have.
+
+Sat Dec 15 13:57:08 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_owned_p): remove static.
+	* io.c (io_flush_buffer): don't hold mutex if already have.
+	  Now recursive lock may occur when following scenario.
+	  fptr_finalize -> finish_writeconv_sync -> finish_writeconv
+	  -> io_fflush.
+
+Sat Dec 15 13:38:30 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_flush_buffer): uses io_flush_buffer_async2 instead of
+	  io_flush_buffer_async.
+	* io.c (io_flush_buffer_async2): new helper function for
+	  io_flush_buffer. It uses rb_thread_call_without_gvl2() instead
+	  of rb_thread_io_blocking_region.
+	* io.c (io_flush_buffer_sync2): new helper function for
+	  io_flush_buffer_async2.
+
+Sat Dec 15 13:04:26 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (internal_write_func2): new helper function for rb_write_internal2().
+	* io.c (rb_write_internal2): new function. it uses
+	  rb_thread_call_without_gvl2() instead of rb_thread_io_blocking_region().
+	* io.c (rb_binwrite_string): uses rb_write_internal2 instead of
+	  rb_write_internal. [Bug #7134]
+
+Sat Dec 15 12:55:29 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_wait_writable): add to call rb_thread_wait_fd()
+	  likes rb_io_wait_readable.
+
+Sat Dec 15 11:54:50 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_wait_writable): don't call rb_thread_fd_writable()
+	  when EINTR. EINTR mean signal interrupt was happen. We don't
+	  need any wait.
+
+Sat Dec 15 11:53:36 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_wait_fd_rw): remove silly rb_thread_alone()
+	  check.
+
+Sat Dec 15 10:22:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_polling): revert but deprecate.
+
+	* include/ruby/intern.h (rb_thread_polling): deprecate.
+
+Sat Dec 15 08:37:01 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* test/rubygems/test_gem_ext_cmake_builder.rb (test_self_build):
+	  get rid of false positive.
+
+Sat Dec 15 08:05:56 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_thread.rb (test_uninitialized, test_backtrace,
+	test_thread_timer_and_interrupt, test_thread_join_in_trap,
+	test_thread_join_current, test_thread_join_main_thread,
+	test_main_thread_status_at_exit, test_thread_status_in_trap,
+	test_thread_status_raise_after_kill, test_mutex_owned,
+	test_mutex_owned2): move these tests from TestThreadGroup class
+	  to TestThread because they are not thread group tests.
+
+	* test/ruby/test_thread.rb (test_thread_status_raise_after_kill):
+	  add t.join.
+
+	* test/ruby/test_threadgroup.rb: new file. moved ThreadGroup test
+	  from test_thread.rb.
+
+Sat Dec 15 08:02:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_thread.rb (TestThread::Thread::new.): remove
+	  th.abort_on_exception change. Test template shouldn't change
+	  global flag. It prevent to test a normal case.
+
+Sat Dec 15 06:15:14 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* configure.in (HAVE_GCC_ATOMIC_BUILTINS):  Set -march=i486 to enable
+	  __sync_val_compare_and_swap.  Patch by KOSAKI Motohiro.
+	  [ruby-trunk - Bug #7485]
+
+Sat Dec 15 03:42:34 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/.document: add fiddle/pointer.c, fiddle/handle.c, and
+	  fiddle/win32/lib as documentation.
+
+Sat Dec 15 03:06:40 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (rb_file_flock): use rb_thread_wait_for() instead of
+	  rb_thread_polling(). When getting EAGAIN, we need to wait a
+	  while even if no multi threading.
+	* thread.c (sleep_for_polling, rb_thread_polling) removed.
+
+Sat Dec 15 00:03:31 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (rb_f_kill): remove rb_thread_polling() because this
+	  has no good effect and makes meaningless 100ms delay. 1)
+	  when sending signal to another process, waiting has just silly.
+	  2) when sending signal to current process, 100ms is often not
+	  enough time to wait. It depend on kernel behavior. And,
+	  rb_thread_polling() doesn't make sense anyway. When rb_thread_alone()
+	  is true, it doesn't wait at all and Process.kill() users don't
+	  expect threading changes Process.kill() behavior. [Bug #7560]
+
+Fri Dec 14 17:10:57 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_params): parser_tokline to track the line number at
+	  which token started.  [ruby-dev:46737] [Bug #7559]
+
+	* parse.y (fcall): operation with starting line number.
+
+	* parse.y (command, primary, method_call): point method name line.
+
+	* parse.y (gettable_gen): return token line for __LINE__.
+
+Fri Dec 14 16:56:59 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_super_method): remove volatile introduced
+	  in r38365.
+
+	* vm_insnhelper.c (vm_call_method): use __forceinline to prevent
+	  VC++ to make vm_call_general and vm_call_super_method as the same
+	  method.  Thanks, Heesob Park.  [Bug #7556] [ruby-core:50867]
+
+Fri Dec 14 14:59:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_separately):
+	  take file and line by using caller_locations if not given.
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_separately):
+	  count assertions in separated tests.
+
+Fri Dec 14 14:16:42 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/rubygems_hook.rb:  Fixed generation of documentation.
+	  Disabled rdoc generation by default to match RubyGems defaults.
+	  Reduced diff with RubyGems::RDoc.
+	* test/rdoc/test_rdoc_rubygems_hook.rb:  Tests for the above.
+	* test/rubygems/test_gem_rdoc.rb:  ditto.
+
+	* lib/rdoc/store.rb:  Removed useless variable assignment
+
+Fri Dec 14 13:58:40 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/rdoc_command.rb:  When overwriting
+	  documentation, remove existing documentation first.
+
+	* lib/rubygems/server.rb:  Fixed documentation links.
+	* test/rubygems/test_gem_server.rb:  Test for the above.
+
+	* lib/rubygems/rdoc.rb:  Reduced diff with RDoc::RubyGemsHook
+	* test/rubygems/test_gem_rdoc.rb:  ditto
+
+Fri Dec 14 04:08:05 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_separately):
+	  added to execute given test source on separate process,
+	  catch its resulted exception and raise it on main process.
+
+Fri Dec 14 07:43:44 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: quote strings that begin
+	  with non-word characters.  Thanks Alex Tambellini!
+	* test/psych/test_yaml.rb: appropriate test case
+
+Thu Dec 13 23:14:17 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_super_method): a workaround for the
+	  failure of TestRefinement#test_refine_recursion in Windows.
+	  See [ruby-core:50871] for details.
+
+Thu Dec 13 23:10:52  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* object.c (Init_Object): use rb_mod_init_copy for Class#initialize_copy
+	* class.c (rb_class_init_copy): rename to class_init_copy_check, performs type
+	  checks on arguments to prevent reinitialization of initialized class
+	  [ruby-core:50869] [Bug #7557]
+	* class.c (rb_mod_init_copy): use class_init_copy_check if receiver is T_CLASS
+	* test/ruby/test_class.rb (class TestClass): related test
+
+Thu Dec 13 16:53:10 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/class_module.rb:  Fixed duplicate comments for classes and
+	  modules from C.
+	* test/rdoc/test_rdoc_class_module.rb:  Test for the above.
+
+	* lib/rdoc/parser/c.rb:  Reload C variable names to allow proper
+	  updates of an ri store for C files.
+	* lib/rdoc/rdoc.rb:  ditto.
+	* lib/rdoc/store.rb:  ditto.
+	* test/rdoc/test_rdoc_parser_c.rb:  Test for the above.
+	* test/rdoc/test_rdoc_store.rb:  ditto.
+
+Thu Dec 13 14:20:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb*: merge doc from doc/irb/ird.rd and improve overall
+	  documentation of IRB
+	* doc/irb/irb.rd: remove stale documentation
+
+Thu Dec 13 14:10:00 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* marshal.c (r_entry0): don't taint classes and modules because
+	  Marshal.load just returns the dumped classes and modules.
+	  [Bug #7325] [ruby-core:49198]
+
+	* test/ruby/test_marshal.rb: related test.
+
+Thu Dec 13 14:10:13 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_require.rb (TestRequire#test_loaded_features_encoding):
+	  need to check compatibility, not equality of encodings.
+
+Thu Dec 13 14:02:15 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_file_join): check encoding compatibility before joining
+	  strings.
+
+Thu Dec 13 13:06:27 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (umethod_bind): allow another form of method transplanting
+	  from a module via UnboundMethod.  [ruby-core:34267][Feature #4254]
+
+Thu Dec 13 12:07:25 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_UNUSED_VAR): new macro to suppress
+	  warnings for unused variables.
+
+	* ext/bigdecimal/bigdecimal.c (ENTER): use RB_UNUSED_VAR() to
+	  suppress annoying warnings by -Wunused-but-set-variable in gcc 4.6.
+
+Thu Dec 13 11:22:33 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: remove "VM_METHOD_TYPE__MAX" from rb_method_type_t.
+	  rb_method_type_t is not a number and "_MAX" causes misunderstanding.
+
+	* proc.c (rb_method_entry_arity): ditto.
+
+	* vm_eval.c (vm_call0_body): ditto.
+
+	* vm_insnhelper.c (vm_call_method): ditto.
+
+Wed Dec 12 21:40:45 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tmpdir.rb (Dir::Tmpname#create): deal with a prefix name which
+	  starts with tilde as a plain name, not expanding as home directory.
+	  [ruby-core:50793] [Bug #7547]
+
+Wed Dec 12 19:48:59 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: merge JSON 1.7.5.
+	  fix tests and other fixes.
+
+Wed Dec 12 18:30:29 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* class.c (rb_prepend_module): move refined methods from the origin
+	  of a class to the class, because refinements should have priority
+	  over prepended modules.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Wed Dec 12 18:27:09 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (zone_str): lookup or insert by using st_update() at once.
+
+Wed Dec 12 15:30:11 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: add -fno-omit-frame-pointer if libexecinfo is used.
+	  At least on FreeBSD ruby will crash on getting C backtrace
+	  when it is compiled with other than -O0.
+
+	* vm_dump.c: enable backtrace on FreeBSD even if with optimizations.
+
+Wed Dec 12 16:08:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_rdoc.rb (TestRDocRDoc#test_normalized_file_list_non_file_directory):
+	  use File::NULL for portability if possible.
+
+Wed Dec 12 16:07:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* method.h (rb_method_flag_t): name a magic number for NOEX_SAFE and
+	  NOEX_WITH as NOEX_SAFE_SHIFT_OFFSET.
+
+	* method.h (rb_method_type_t, method_optimized_type): C89 forbids a
+	  comma after the last element in enum.
+
+	* proc.c (rb_method_entry_arity), vm_eval.c (vm_call0_body),
+	  vm_insnhelper.c (vm_call_method): add VM_METHOD_TYPE__MAX case.
+
+Wed Dec 12 14:16:35 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/class_module.rb:  Added RDoc::ClassModule#documented? which
+	  checks comment_location.  Hide RDoc::ClassModule#comment=.
+	* test/rdoc/test_rdoc_class_module.rb:  Test for above.
+
+	* lib/rdoc/generator/template/darkfish/_sidebar_table_of_contents.rhtml:
+	  Fix display of the table of contents in the sidebar.
+
+	* lib/rdoc/generator/template/darkfish/table_of_contents.rhtml:
+	  Use #comment_location when displaying classes or modules.
+
+	* test/rdoc/test_rdoc_store.rb:  Use comment_location.
+
+Wed Dec 12 13:40:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_getivar): no uninitialized instance variables
+	  warnings for non-object if attr method.
+
+Wed Dec 12 06:43:37 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* iseq.c (rb_iseq_parameters): fix limit for optional arguments.
+
+	* test/ruby/test_keyword.rb: tests for above.
+
+	* vm_core.h (struct rb_iseq_struct): update documentation
+	  with keyword arguments. [Bug #7540] [ruby-core:50735]
+
+Wed Dec 12 03:45:41 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_exec): pass exceptions while handling an exception.
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks): propagate exceptions.
+	  revert r38293 partially.
+
+Wed Dec 12 03:09:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sample/test.rb (Progress#initialize): add --verbose option and show
+	  messages in one line unless --verbose is given.
+
+Wed Dec 12 01:47:02 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_using_refinement): make the method table of an iclass
+	  for a refinement that of the refinement, not that of the origin of
+	  the refinement, which is set by rb_include_class_new().  This
+	  change is needed to make module prepend into a refinement work
+	  properly.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Wed Dec 12 01:05:04 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/make-snapshot: add --disable-rubygem to both MINIRUBY and RUBY.
+	  On making miniprelude.c, it seems use MINIRUBY. this fixes #7541
+	  but rubygems also needs to be fixed for older rubies.
+
+Wed Dec 12 00:32:11 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/dl/test_func.rb (test_name_with_block, test_bind, test_qsort1):
+	  call unbind to release the callback closure because maximum number
+	  of callbacks is limited to DL::MAX_CALLBACK (== 5) with pure DL
+	  without Fiddle.
+
+Wed Dec 12 00:13:34 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/dl/lib/dl/func.rb (DL::Function#unbind, #bound?): suppress
+	  NoMethodError when Fiddle is available. [ruby-core:50756] [Bug #7543]
+	* test/dl/test_func.rb (test_bound*, test_unbind*): tests for the above.
+
+Tue Dec 11 19:38:37 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/fiddle/function.c (Fiddle::Function.new): new keyword argument
+	  :name to set the name attribute.
+	* ext/fiddle/lib/fiddle/import.rb (import_function, bind_function):
+	  set function name by using the :name keyword argument.
+	  Re-fixes r38243. [ruby-core:50566]
+	* test/fiddle/test_function.rb (test_name): test for the :name keyword
+	  argument and Fiddle::Function#name.
+
+Tue Dec 11 16:57:33 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* common.mk:  Added --pages-dir to rdoc creation.  Now doc/ items show
+	  up at top-level.
+	* .document:  Moved doc/* entries to doc/.document
+	* doc/.document:  ditto
+
+Tue Dec 11 16:44:37 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/options.rb:  Added --page-dir option for moving pages in
+	  doc/ to the top-level.
+	* lib/rdoc/rdoc.rb:  ditto.
+	* test/rdoc/test_rdoc_options.rb:  Test for the above.
+	* test/rdoc/test_rdoc_rdoc.rb:  ditto.
+
+Tue Dec 11 15:24:05 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/pathname/lib/pathname.rb:  Hide private methods from RDoc.
+
+Tue Dec 11 15:11:29 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/make-snapshot (BASERUBY): add --disable-gem to avoid load gems.
+	  [Bug #7541] [ruby-core:50736]
+
+Tue Dec 11 12:00:19 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/dl/win32/extconf.rb: Fix typo
+	  by Santiago Pastorino <santiago@wyeworks.com>
+	  https://github.com/ruby/ruby/pull/221 fix GH-221
+
+Tue Dec 11 01:53:37 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix: alias {row|column}_size to {row|column}_count and use
+	  the latter.
+	  [Bug #7369] [ruby-core:49409]
+
+Tue Dec 11 00:26:58 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* fix the behavior when a module is included into a refinement.
+	  This change is a little tricky, so it might be better to prohibit
+	  module inclusion to refinements.
+
+	* include/ruby/ruby.h (RMODULE_INCLUDED_INTO_REFINEMENT): new flag
+	  to represent that a module (iclass) is included into a refinement.
+
+	* class.c (include_modules_at): set RMODULE_INCLUDED_INTO_REFINEMENT
+	  if klass is a refinement.
+
+	* eval.c (rb_mod_refine): set the superclass of a refinement to the
+	  refined class for super.
+
+	* eval.c (rb_using_refinement): skip the above superclass (the
+	  refined class) when creating iclasses for refinements.  Otherwise,
+	  `using Refinement1; using Refinement2' creates iclasses:
+	  <Refinement2> -> <RefinedClass> -> <Refinement1> -> RefinedClass,
+	  where <Module> is an iclass for Module, so RefinedClass is
+	  searched before Refinement1.  The correct iclasses should be
+	  <Refinement2> -> <Refinement1> -> RefinedClass.
+
+	* vm_insnhelper.c (vm_search_normal_superclass): if klass is an
+	  iclass for a refinement, use the refinement's superclass instead
+	  of the iclass's superclass.  Otherwise, multiple refinements are
+	  searched by super.  For example, if a refinement Refinement2
+	  includes a module M (i.e., Refinement2 -> <M> -> RefinedClass,
+	  and if refinements iclasses are <Refinement2> -> <M>' ->
+	  <Refinement1> -> RefinedClass, then super in <Refinement2> should
+	  use Refinement2's superclass <M> instead of <Refinement2>'s
+	  superclass <M>'.
+
+	* vm_insnhelper.c (vm_search_super_method): do not raise a
+	  NotImplementError if current_defined_class is a module included
+	  into a refinement.  Because of the change of
+	  vm_search_normal_superclass(), the receiver might not be an
+	  instance of the module('s iclass).
+
+	* test/ruby/test_refinement.rb: related test.
+
+Mon Dec 10 18:35:25 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_without_refinements): use
+	  rb_resolve_refined_method() to search superclasses if
+	  me->def->orig_me is 0.  This change fixes make test-all
+	  TESTS="json ruby/test_refinement.rb".
+
+	* test/ruby/test_refinement.rb: related test.
+
+Mon Dec 10 17:59:07 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/fiddle/win32/*: library ports from DL to Fiddle.
+
+	* ext/dl/win32/extconf.rb: check fiddle.  often case dl compiled prior
+	  to fiddle, so this change is no meaning.  in most cases, simply
+	  fiddle/win32 overwrite dl/win32.
+
+Mon Dec 10 15:23:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks): exceptions in event
+	  hooks should not propagate outside.
+
+Mon Dec 10 15:11:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): count flip-flop state in local iseq
+	  not in each iseqs, so that the keys can be other than hidden
+	  strings.  [ruby-core:47253] [Bug #6899]
+
+	* vm_insnhelper.c (lep_svar_get, lep_svar_set, vm_getspecial): store
+	  flip-flop states in an array instead of a hash.
+
+	* iseq.c (set_relation): main iseq also can has local scope.
+
+Mon Dec 10 10:36:12 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* lib/irb/magic-file.rb: set a encoding, which is detected from
+	  the file to read, to the internal encoding.
+	  [Bug #4281][ruby-dev:43036]
+
+Mon Dec 10 09:40:19 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/ext/cmake_builder.rb:  Added a builder for cmake.
+	* lib/rubygems/ext.rb:  ditto.
+	* lib/rubygems/installer.rb:  ditto.
+	* test/rubygems/test_gem_ext_cmake_builder.rb:  Test for above.
+
+Mon Dec 10 09:13:08 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/package.rb:  Omit directories when packaging gems like
+	  RubyGems 1.8.x
+	* test/rubygems/test_gem_package.rb:  Test for above.
+
+Sun Dec  9 17:36:59 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_opt_send): Kernel#send should not use
+	  refinements.
+
+	* proc.c (mnew): Kernel#method, Kernel#public_method,
+	  Module#instance_method, and Module#public_instance_method should
+	  not use refinements.
+
+	* vm_method.c (rb_method_boundp): Kernel#respond_to? should not use
+	  refinements.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sun Dec  9 06:19:04 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markdown/entities.rb:  Added documentation.
+
+	* lib/rdoc/parser/ruby.rb:  Updated style
+
+	* lib/rdoc/ruby_lex.rb:  Parse characters up to \u{FFFFF}
+	* test/rdoc/test_rdoc_ruby_lex.rb:  Test for above.
+
+Sat Dec  8 22:38:35 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refine): don't override Module#include.  It's
+	  unnecessary now because refinements are activated only in refine
+	  blocks.
+
+Sat Dec  8 22:33:26 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c: remove Module#refinements.
+
+	* test/ruby/test_refinement.rb: remove tests for Module#refinements.
+
+Sat Dec  8 13:17:55 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (top_using): raise a RuntimeError if using is called in a
+	  module definition or a method definition.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sat Dec  8 15:01:35 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/cleanup_command.rb:  Skip default gems when
+	  cleaning up.
+	* test/rubygems/test_gem_commands_cleanup_command.rb:  Test for above.
+
+	* lib/rubygems/commands/query_command.rb:  Fixed listing remote gems.
+
+	* lib/rubygems/dependency_installer.rb:  Ignore non-files when looking
+	  for local gems.
+	* test/rubygems/test_gem_dependency_installer.rb:  Test for above.
+
+	* lib/rubygems/uninstaller.rb:  The user must confirm uninstalling gems
+	  that have dependencies.
+	* test/rubygems/test_gem_uninstaller.rb:  Test for above.
+
+	* lib/rubygems.rb (module Gem):  Updated version.
+
+	* test/rubygems/*.pem:  Updated to run in FIPS mode.
+	* test/rubygems/test_gem_security.rb:  ditto.
+	* test/rubygems/test_gem_security_signer.rb:  ditto.
+
+Sat Dec  8 12:34:01 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_normal_superclass): super in a
+	  refinement always uses the refined class as its superclass.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sat Dec  8 11:59:59 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refine): raise an ArgumentError if a given
+	  block is of a Proc object.
+
+	* vm_insnhelper.c (vm_call_method): store refined methods in inline
+	  cache to improve performance.  It's safe now because blocks cannot
+	  be yielded with different refinements in the new specification.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sat Dec  8 11:17:53 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refine), vm_eval.c (rb_yield_refine_block):
+	  Module#refine activates all refinements defined in that module
+	  only in a given block.
+
+	* string.c (sym_to_proc, sym_call): don't use refinements.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sat Dec  8 09:24:42 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_x509name.c:  Completed documentation for
+	  OpenSSL::X509::Name.
+
+Sat Dec  8 07:57:12 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (iow_size): return size of internal object
+	  for ObjectSpace.memsize_of().
+
+	* test/objspace/test_objspace.rb: add a test.
+
+Tue Dec 08 02:39:23 2012  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb: A fix for row comparison by Stephen Wattam.  [Bug #7528]
+
+Sat Dec  8 01:27:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): no need to
+	  check all reports.
+
+Sat Dec  8 00:10:34 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_eval.c (yield_under, eval_under): do not activate refinements
+	  of the receiver in module_eval and instance_eval.
+
+	* eval.c (ruby_Init_refinement): undef Class#refine.
+
+	* eval.c (ruby_Init_refinement): remove Module#using.
+
+	* eval.c (ruby_Init_refinement): main.using should be private.
+
+	* eval.c (rb_mod_refine): the argument of Module#refine should not
+	  be a module.
+
+	* insns.def (defineclass): do not activate refinements in a class or
+	  module.
+
+Fri Dec  7 23:42:11 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/refinement/refinement.c: include ruby/ruby.h instead of the
+	  declaration of rb_warn().
+
+Fri Dec  7 16:07:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/etc.rd: Removed stale documentation file
+	* ext/etc/etc.c:  Merged documentation from doc/etc.rd and updated
+	  rdoc, added documentation for Etc::Passwd and Etc::Group
+
+Fri Dec  7 16:00:57 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): no need to
+	  retry skipped test.  this fix makes 40% faster the whole test-all
+	  with -j5 on Windows.
+
+Fri Dec  7 14:22:29 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markup/to_joined_paragraph.rb:  Completed documentation
+	* lib/rdoc/parser/c.rb:  ditto
+	* lib/rdoc/parser/changelog.rb:  ditto
+	* lib/rdoc/servlet.rb:  ditto
+	* lib/rdoc/store.rb:  ditto
+
+	* lib/rdoc/store.rb:  Improved HTML error page.  Completed
+	  documentation
+
+	* lib/rdoc/parser/ruby.rb:  Fixed bug attaching a comment to A::B = 42
+	* test/rdoc/test_rdoc_parser_ruby.rb:  Test for above
+
+	* test/rdoc/test_rdoc_comment.rb:  Removed garbage
+
+Fri Dec  7 14:03:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout#timeout): since async_interrupt_timing
+	  re-raises a deferred exception, replace the timeout exception with
+	  Timeout::Error after it.  [Bug #7503]
+
+Fri Dec  7 13:07:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/forwardable.rd: Remove stale documentation file
+	* lib/forwardable.rb: Merge documentation from doc/forwardable.rd
+
+Fri Dec  7 09:47:35 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (time_mdump): dump timezone string to private instance variable
+	  on marshaling.
+
+	* time.c (time_mload): load timezone string from private instance
+	  variable named 'zone'.
+
+Fri Dec  7 01:15:07 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/fiddle/lib/fiddle/function.rb (Fiddle::Function#name): new
+	  attribute needed to switch Win32::Registry from DL to Fiddle.
+
+	* ext/fiddle/lib/fiddle/import.rb (import_function, bind_function):
+	  set function name to the returned Fiddle::Function object.
+
+Fri Dec  7 00:11:44 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_refinement.rb: fix some tests to use neither
+	  Module#using nor Module#module_eval.
+
+Thu Dec  6 23:27:50 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (ruby_Init_refinement): a new function to enable
+	  Refinements with a warning "Refinements are experimental...".
+
+	* ext/refinement/refinement.c, ext/refinement/extconf.rb: a new
+	  extension library to enable Refinements.
+
+Thu Dec  6 18:23:05 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* revised r37993 to avoid SEGV/ILL in tests.  In r37993, a method
+	  entry with VM_METHOD_TYPE_REFINED holds only the original method
+	  definition, so ci->me is set to a method entry allocated in the
+	  stack, and it causes SEGV/ILL.  In this commit, a method entry
+	  with VM_METHOD_TYPE_REFINED holds the whole original method entry.
+	  Furthermore, rb_thread_mark() is changed to mark cfp->klass to
+	  avoid GC for iclasses created by copy_refinement_iclass().
+
+	* vm_method.c (rb_method_entry_make): add a method entry with
+	  VM_METHOD_TYPE_REFINED to the class refined by the refinement if
+	  the target module is a refinement.  When a method entry with
+	  VM_METHOD_TYPE_UNDEF is invoked by vm_call_method(), a method with
+	  the same name is searched in refinements.  If such a method is
+	  found, the method is invoked.  Otherwise, the original method in
+	  the refined class (rb_method_definition_t::body.orig_me) is
+	  invoked.  This change is made to simplify the normal method lookup
+	  and to improve the performance of normal method calls.
+
+	* vm_method.c (EXPR1, search_method, rb_method_entry),
+	  vm_eval.c (rb_call0, rb_search_method_entry): do not use
+	  refinements for method lookup.
+
+	* vm_insnhelper.c (vm_call_method): search methods in refinements if
+	  ci->me is VM_METHOD_TYPE_REFINED.  If the method is called by
+	  super (i.e., ci->call == vm_call_super_method), skip the same
+	  method entry as the current method to avoid infinite call of the
+	  same method.
+
+	* class.c (include_modules_at): add a refined method entry for each
+	  method defined in a module included in a refinement.
+
+	* class.c (rb_prepend_module): set an empty table to
+	  RCLASS_M_TBL(klass) to add refined method entries, because
+	  refinements should have priority over prepended modules.
+
+	* proc.c (mnew): use rb_method_entry_with_refinements() to get
+	  a refined method.
+
+	* vm.c (rb_thread_mark): mark cfp->klass for iclasses created by
+	  copy_refinement_iclass().
+
+	* vm.c (Init_VM), cont.c (fiber_init): initialize th->cfp->klass.
+
+	* test/ruby/test_refinement.rb (test_inline_method_cache): do not skip
+	  the test because it should pass successfully.
+
+	* test/ruby/test_refinement.rb (test_redefine_refined_method): new
+	  test for the case a refined method is redefined.
+
+Thu Dec  6 17:29:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_here_document): flush string content between new
+	  line and :string_embexpr.  [ruby-core:48703] [Bug #7255]
+
+Thu Dec  6 16:35:21 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rake/helper.rb:  Load envutil correctly.  Removed useless rescue
+	  for signal propagation tests
+	* lib/rake/file_utils.rb:  Prefer the built ruby.
+	* test/rake/test_rake_functional.rb:  ditto
+
+Thu Dec  6 15:20:34 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/context.rb:  Don't warn for duplicate methods while loading.
+	* test/rdoc/test_rdoc_context.rb:  Test for above.
+
+Thu Dec  6 14:26:22 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/command_manager.rb:  Removed string concatenation
+	  syntax.  [Bug #6265]
+	* lib/rubygems/commands/install_command.rb:  ditto
+	* lib/rubygems/commands/uninstall_command.rb:  ditto
+	* lib/rubygems/indexer.rb:  ditto
+	* lib/rubygems/security/policy.rb:  ditto
+	* lib/rubygems/security.rb:  ditto
+	* lib/rubygems/uninstaller.rb:  ditto
+	* test/rubygems/test_gem_commands_cert_command.rb:  ditto
+	* test/rubygems/test_gem_package.rb:  ditto
+	* test/rubygems/test_gem_security.rb:  ditto
+	* test/rubygems/test_gem_security_policy.rb:  ditto
+
+Thu Dec  6 14:10:08 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/package.rb:  Set rubygems_version before validation.
+	  Fixes issue with bundler.
+	* test/rubygems/test_gem_package.rb:  Test for above.
+
+	* lib/rubygems/remote_fetcher.rb:  Only update the cache when we have
+	  permission.  [ruby-trunk - Bug #7509]
+	* lib/rubygems/source.rb (class Gem):  ditto
+	* test/rubygems/test_gem_remote_fetcher.rb:  Test for above.
+	* lib/rubygems/test_utilities.rb:  ditto
+
+	* lib/rubygems/specification.rb:  Derive base_dir properly for default
+	  gems.  [ruby-trunk - Bug #7496]
+	* test/rubygems/test_gem_specification.rb:  Test for above.
+
+	* lib/rubygems.rb:  Untaint Dir.pwd when searching for gemdeps files
+	  for operation under $SAFE=1
+
+Thu Dec 06 12:07:11 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: TracePoint#enable should not cause an error
+	  when it is already enabled. TracePoint#disable is too.
+	  [ruby-core:50561] [ruby-trunk - Bug #7513]
+
+	* test/ruby/test_settracefunc.rb: add tests.
+
+Thu Dec  6 07:19:58 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc*:  Improved display of ChangeLog files as HTML.
+	* test/rdoc*:  Test for above.
+
+Thu Dec  6 04:34:19 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_uninterruptible): helper function for providing
+	  temporary async_interrupt_timing(Object => :defer)
+
+	* io.c (rb_f_p): use rb_uninterruptible.
+	* io.c (rb_f_p_internal): helper function for rb_f_p().
+	* io.c (struct rb_f_p_arg): new struct for rb_f_p_internal.
+
+	* test/ruby/test_thread.rb (test_async_interrupt_and_p): test for
+	  the above.
+
+Thu Dec  6 04:27:10 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_binwrite): check interrupt before io issue.
+	* test/ruby/test_thread.rb (test_async_interrupt_and_io):
+	  test for the above.
+
+Thu Dec  6 01:10:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_method_call_status): use Qundef as no self instead of
+	  the current self.
+
+	* vm_eval.c (send_internal): public_send does not consider how it is
+	  called, as mentioned in r14173.  patched by charliesome (Charlie
+	  Somerville).  [ruby-core:50489] [Bug #7499]
+
+Wed Dec  5 23:50:23 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (getrusage_time): uses clock_gettime() with
+	  CLOCK_PROCESS_CPUTIME_ID when available, which provides a 1ns
+	  precision on linux. [ruby-core:50495] [Bug #7500]
+	  patched by Aman Gupta.
+
+Wed Dec  5 22:46:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_vm_make_proc): save the proc made from the given block so
+	  that it will not get collected.  [ruby-core:50545] [Bug #7507]
+
+Wed Dec  5 22:13:57 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/dl/lib/dl/func.rb (DL::Function#bind): When Fiddle is used,
+	  @ptr should be updated. This fixes SEGV raised in DL::Function#call
+	  after calling DL::Function#bind. [Bug #7516] [ruby-dev:46708]
+
+	* test/dl/test_func.rb (test_bind): test for the above
+
+Wed Dec  5 18:53:00 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* thread.c (rb_thread_s_async_interrupt_timing): have to check ints
+	  before jumping out.
+	* test/ruby/test_thread.rb (test_async_interrupt_with_return): add test
+	  rescue has to catch a queued async exception at the time of return.
+	* test/ruby/test_thread.rb (test_async_interrupt_with_break): add test
+	  rescue has to catch a queued async exception at the time of break.
+
+Wed Dec  5 16:54:28 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/memory_status.rb: suppress warning.
+	  A patch from NAKAMURA Usaku.
+
+Wed Dec  5 16:06:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parser/changelog.rb:  Parse more ChangeLog file variations.
+	* test/rdoc/test_rdoc_parser_changelog.rb:  Test for above.
+
+Wed Dec  5 12:17:11 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/dl/lib/dl/func.rb (DL::Function#initialize, DL::Function#bind):
+	  ABI should be set by using CFunc#calltype even when Fiddle is used.
+	  When Fiddle is used and a block is given, name should not be ignored.
+	  [ruby-core:50562] [Bug #7514]
+
+	* ext/dl/lib/dl/import.rb (DL::Importer#bind_function): should respect
+	  abi and name when Fiddle is used.
+
+	* test/dl/test_func.rb (test_name_with_block): test for "name" method
+	  with giving a block.
+
+Wed Dec  5 11:55:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/shell.rd, doc/shell.rd.ja: Removed stale doc files
+	* lib/shell.rb, lib/shell/*: Merge and updates docs from doc/shell.rd*
+
+Wed Dec  5 11:42:38 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_settracefunc.rb: disable trace.
+
+Wed Dec  5 11:37:37 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#macro_defined?): use clearly different
+	  strings from conflict markers.
+
+Wed Dec  5 04:25:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/README: Add rdoc modeline directive and formatting libs
+
+Wed Dec  5 04:04:02 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* test/ruby/test_thread.rb (test_async_interrupt_blocking): bugfix
+	  about deferred check
+
+Wed Dec  5 03:35:37 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* vm_core.h (RUBY_VM_CHECK_INTS_BLOCKING): check async queue everytime.
+	* thread.c (sleep_forever): check RUBY_VM_CHECK_INTS_BLOCKING first.
+	* thread.c (sleep_timeval): ditto.
+	* test/ruby/test_thread.rb (test_async_interrupt_blocking): add a test
+	  exceptions are correctly deferred and raised on :on_blocking context.
+
+Wed Dec  5 02:36:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, defs/id.def, template/id.c.tmpl: generate id.c as well as id.h.
+
+Wed Dec  5 00:56:21 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_owned_p): new method that return current
+	  thread have the target mutex or not. [Feature #7505] [ruby-dev:46697]
+	* test/ruby/test_thread.rb (test_mutex_owned, test_mutex_owned2):
+	  test for the above.
+	* NEWS: new for the above.
+
+Wed Dec  5 00:05:47 2012  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/erb.rb (make_compiler, add_put_cmd, add_insert_cmd): extract
+	  methods.
+
+Tue Dec  4 18:21:04 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/ruby/memory_status.rb (Memory): use fiddle/types if available.
+
+	* test/ruby/memory_status.rb (Memory::Win32): :stdcall is needed on
+	  x86 WIN32. This commit partly reverts r38054.
+
+Tue Dec  4 18:05:58 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/fiddle/lib/fiddle/types.rb: copied from ext/dl/lib/dl/types.rb
+	  and modified for Fiddle, needed for migration from DL to Fiddle.
+
+Tue Dec  4 17:57:09 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/fiddle/lib/fiddle/import.rb (import_function, bind_function):
+	  should respect call_type for migration from DL to Fiddle.
+	  [Bug #7484] [ruby-core:50405]
+
+Tue Dec  4 16:54:00 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* .document:  Added ChangeLog and doc/ChangeLog-* as documentation
+
+Tue Dec  4 16:47:46 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parser/changelog.rb:  Added a ChangeLog parser to RDoc.
+	* lib/rdoc/parser.rb:  ditto
+	* test/rdoc/test_rdoc_parser_changelog.rb:  Test for above.
+
+Tue Dec  4 16:23:32 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (path2class, path2module): use PRIsVALUE.
+
+	* marshal.c (w_object, marshal_dump, marshal_load): use
+	  rb_check_funcall if possible.
+
+	* marshal.c (w_object, marshal_dump, r_object0, marshal_load): use
+	  RB_GC_GUARD() (directly or indirectly) instead of volatile.
+
+	* variable.c (rb_path_to_class): prevent the argument from GC.
+
+Tue Dec 04 13:55:07 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_opts.h: enable optimization - operand unification.
+	  Operand unification technique enable to combine
+	  an instruction and specific operands and make new
+	  instruction.
+
+	* defs/opt_operand.def: add several configuration
+	  of operand unification.
+
+	* insns.def: use `int' instead to suppress warning.
+
+Mon Dec  3 17:58:53 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y: replace parser->enc with current_enc.
+
+Tue Dec  4 08:33:46 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* README.EXT:  Converted to RDoc format
+	* README.EXT.ja:  ditto
+
+Tue Dec  4 08:32:10 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/driver.rb:  Fixed ri page display for files with
+	  extensions.
+	* test/rdoc/test_rdoc_ri_driver.rb:  Test for above
+
+Tue Dec  4 04:11:50 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* .document:  Add NEWS for `ri ruby:NEWS`
+	* NEWS:  Set format as rdoc
+	* doc/NEWS-1.8.7:  ditto
+	* doc/NEWS-1.9.1:  ditto
+	* doc/NEWS-1.9.2:  ditto
+	* doc/NEWS-1.9.3:  ditto
+
+Mon Dec  3 20:37:22 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_exec.c: check VM_COLLECT_USAGE_DETAILS.
+
+Mon Dec  3 20:28:02 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_specialized_instruction):
+	  change condition of using `opt_send_simple'.
+	  More method invocations can be simple.
+
+Mon Dec  3 20:03:38 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_objectspace.rb: skip RuntimeError
+	  which says a message "can't modify frozen File".
+	  Is that correct behavior?
+
+Mon Dec 03 20:00:19 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_exec.c: vm_analysis_insn should be static.
+
+Mon Dec  3 19:10:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (Init_Random), rational.c (Init_Rational): make marshal
+	  methods private.  [Feature #6539]
+
+Mon Dec  3 18:29:27 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.h: iseq_catch_table_entry::catch_type should be
+	  Fixnum because they are pushed into Array in a compiler.
+	  [Bug #7502]
+
+	* test/ruby/test_objectspace.rb: add a test of this issue.
+
+Mon Dec  3 18:25:16 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/id.h.tmpl (preserved_ids): "empty?" is not an attribute name.
+
+Mon Dec  3 16:23:09 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c (vm_backtrace_to_ary): check negative size (2nd arg).
+
+Mon Dec  3 15:50:33 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el (ruby-mode-set-encoding): Unbreak by
+	  fixing a typo, s/set/setq/.
+
+Mon Dec  3 14:14:19 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): joke shouldn't use id.h defined ids.
+
+	* id.c (Init_id): ditto.
+
+	* common.mk: fix dependency.
+
+Mon Dec  3 12:43:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-block-end-re, ruby-delimiter)
+	  (ruby-mode-syntax-table, ruby-parse-partial, ruby-beginning-of-indent):
+	  merge from Emacs.
+
+	* misc/ruby-mode.el (ruby-calculate-indent): fix indentation of
+	  argument lines in parentheses.  [Bug #5140]
+
+Mon Dec  3 07:52:41 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parser.rb:  Improved modeline support.  Patch by nobu.
+	* test/rdoc/test_rdoc_parser.rb:  Test for above.
+
+Sun Dec  3 00:06:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_new): stop checking string
+	  taintness.  [Bug #5508] [ruby-core:40510]
+
+Sun Dec  2 19:26:47 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* thread.c (RB_GC_SAVE_MACHINE_CONTEXT, rb_gc_save_machine_context):
+	  extract rb_gc_save_machine_context to RB_GC_SAVE_MACHINE_CONTEXT.
+	  NOTE: machine_regs and machine_stack_end must be set in current scope.
+
+Sun Dec  2 18:46:24 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c, enum.c, insns.def, io.c, numeric.c, parse.y, process.c,
+	  range.c: use prepared IDs.
+	  A patch from charliesome (Charlie Somerville).
+	  [Bug #7495]
+
+	* common.mk: add dependency to id.h.
+
+	* common.mk: replace ID_H_INCLUDES with id.h.
+
+Sun Dec  2 16:48:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/weakref.rb (rdoc): Clean up usage, add example,
+	  note ArgumentError on WeakRef.new
+
+Sun Dec  2 16:45:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* gc.c (WeakMap): Add doc for internal reference, use lib/weakref.rb
+
+Sun Dec  2 07:24:23 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/parser.rb:  Parse files with a -*- rdoc -*- modeline
+	* test/rdoc/test_rdoc_parser.rb:  Test for above
+
+Sun Dec  2 06:02:00 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* gc.h (SET_MACHINE_STACK_END): add volatile for preventing
+	  harmful optimization.  [ruby-dev:46665] [Bug #7468]
+
+Sun Dec  2 05:01:58 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (rb_iseq_line_trace_each): iterate `line' event only.
+
+	* test/ruby/test_iseq.rb: add a test for this change.
+
+Sun Dec  2 02:46:04 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: add TracePoint#inspect.
+
+	* test/ruby/test_settracefunc.rb: add a test for this change.
+
+Sat Dec  1 21:18:19 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_backtrace.rb: add a test for
+	  Thread::Backtrace::Location#inspect.
+	  BTW, tests for `caller_locations' are not enough.
+	  Any volunteers are welcome.
+
+Sat Dec  1 21:06:58 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c (location_inspect_m): add
+	  Thread::Backtrace::Location#inspect.
+	  It same as loc_obj.to_s.inspect.
+
+Sat Dec  1 19:24:09 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_puts): recurse for the argument itself, not converted
+	  array elements.  [ruby-core:42444] [Bug #5986]
+
+Sat Dec  1 19:01:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object, r_object0): call private marshal methods.
+	  [Feature #6539]
+
+Sat Dec  1 18:52:22 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/cleanup_command.rb:  Fix cleanup command for
+	  multiple gems.  [ruby-trunk - #7481] by Kouhei Sutou
+	* test/rubygems/test_gem_commands_cleanup_command.rb:  Test for above.
+	* lib/rubygems.rb:  Autoload Gem::Source to prevent test failures
+
+Sat Dec  1 18:17:00 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (Init_Complex), time.c (Init_Time): make marshal methods
+	  private.  [Feature #6539]
+
+	* object.c (Init_Object): make remove_instance_variable public.
+	  [Feature #6539]
+
+	* id.c (Init_id), template/id.h.tmpl: add initialize_{copy,clone,dup}
+	  and respond_to_missing?.
+
+	* vm_method.c (rb_method_entry_make): make above methods private.
+	  [Feature #6539]
+
+Sat Dec  1 16:40:22 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_thread.rb: move ConditionVariable related test
+	  into test/thread/test_cv.rb.
+	* test/thread/test_cv.rb: new file.
+	* test/thread/test_cv.rb (test_condvar_empty_signal): new tests.
+	* test/thread/test_cv.rb (test_condvar_empty_broadcast): ditto.
+
+Sat Dec  1 15:14:25 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_thread.rb (test_cv_wait_deadlock): enable
+	  cv deadlock test.
+
+Sat Dec  1 14:23:33 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/thread.rb (ConditionVariable): use hash instead of array for
+	  @waiters.
+	* test/thread/test_queue.rb (test_sized_queue_and_wakeup): remove
+	  a test because @waiters no longer have a chance to duplicated. Now it's
+	  a hash.
+
+Sat Dec  1 17:16:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-electric.el (ruby-electric-curlies): use kill-region
+	  instead of interactive command delete-backward-char.
+
+Sat Dec  1 17:12:55 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/inf-ruby.el (inferior-ruby-mode): fix the
+	  compilation-shell-minor-mode configuration.  a patch by
+	  j2petkov (Jean-Christophe Petkovich) in [ruby-core:46518].
+	  [Bug #6742]
+
+Sat Dec  1 15:05:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): use NAMLEN() to tell the length of d_name
+	  instead of strlen(), which can access beyond the boundary.
+
+Sat Dec  1 13:48:13 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/specification.rb:  Don't add default gems to $LOAD_PATH
+	  as they are already there.
+
+Sat Dec  1 12:22:17 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* re-added r38053 that is reverted by r38061. Problems by r38053
+	  are resolved by r38096. r38096 removed GEM_SKIP configuration.
+
+	  The below is ChangeLog of r38053:
+
+	* defs/default_gems: Add base directory column.
+
+	* tool/rbinstall.rb:
+	  - Install .gemspecs of default gem to
+	    #{GEM_HOME}/specifications/default/.
+	  - Update files parameter of .gemspecs by relative path from
+	    library directory.
+
+Sat Dec  1 11:09:12 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* variable.c (rb_class_path_no_cache): add a function to get the class
+	  path without caching the computed path. Some classes are frozen, and
+	  will raise an exception without this.
+
+	* probes.d (cmethod-entry, cmethod-return): separate cmethods from
+	  regular methods to match set trace func.
+
+	* probes_helper.h: refactor macros.  Fix probes to avoid calling
+	  #inspect when profiling.
+
+	* insns.def: update for use with new macros.
+
+	* vm_eval.c: ditto
+
+	* vm_insnhelper.c: ditto
+
+	* test/dtrace/test_singleton_function.rb: fix test for new output.
+
+	* test/dtrace/test_cmethod.rb: test the cmethod probes.
+
+Sat Dec  1 09:44:16 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rdoc/test_rdoc_options.rb:  Windows drive letters are
+	  case-insensitive.
+
+Sat Dec  1 09:42:13 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Search for gem deps file up the directory tree.
+	* test/rubygems/test_gem.rb:  Test for above.
+
+Sat Dec  1 09:33:32 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/runner.rb:  Set GEM_HOME, GEM_PATH and GEM_SKIP to empty set.
+	  With default_gem support in RubyGems GEM_SKIP prevents loading of
+	  built-in gems.
+
+Sat Dec  1 07:16:17 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* compile.c (ADD_CATCH_ENTRY): add a cast to fix SEGV with x64 mingw
+	  on Windows 8. Without cast, 0 might be non zero value at higher bits
+	  in rb_ary_new3().
+	  [ruby-core:50258] [Bug #7456]
+
+Sat Dec  1 04:07:57 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (parser.utf8): remove unused property.
+
+	* parse.y (UTF8_ENC): remove unused macro.
+
+	* parse.y (parser_tokadd_utf8): use rb_utf8_encoding() directly.
+
+Sat Dec  1 03:49:45 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/sync.rb (Sync_m#sync_synchronize): add Thread.async_interrupt_timing
+	  for protecting from async interrupt.
+	* lib/sync.rb (Sync_m#sync_lock): ditto.
+
+Sat Dec  1 03:38:04 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/thread.rb (ConditionVariable#broadcast): s/RuntimeError/StandardError/
+	* lib/thread.rb (ConditionVariable#signal): ditto.
+
+Sat Dec  1 03:29:52 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/thread.rb (SizedQueue#pop): rewrite by using ConditionVariable.
+	* lib/thread.rb (SizedQueue#push): ditto.
+	* lib/thread.rb (SizedQueue#max): ditto.
+	* lib/thread.rb (Queue#pop): ditto.
+	* lib/thread.rb (Queue#push): ditto.
+
+	* lib/thread.rb (SizedQueue#num_waiting): adopt the above changes.
+	* lib/thread.rb (SizedQueue#initialize): ditto.
+	* lib/thread.rb (Queue#num_waiting): ditto.
+	* lib/thread.rb (Queue#initialize): ditto.
+	* test/thread/test_queue.rb (test_sized_queue_and_wakeup): ditto.
+
+Sat Dec  1 03:45:47 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (Thread.async_interrupt_timing): fix RDoc.
+	  :never is not used any more.
+
+Sat Dec  1 02:56:19 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c: add RubyVM::InstructionSequence (ISeq) inspection methods.
+	  * ISeq#path returns path of this ISeq written.
+	  * ISeq#absolute_path returns absolute path.
+	  * ISeq#label returns label (method name and so on).
+	  * ISeq#base_label returns base label (see Thread::Backtrace::Location).
+	  * ISeq#first_lineno returns first line number of this ISeq.
+	  * ISeq.of(obj) returns ISeq object which obj (Proc or Method)
+	    is contains.
+
+	* test/ruby/test_iseq.rb: add tests.
+
+Sat Dec  1 02:58:51 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* include/ruby/ruby.h (rb_event_flag_t):  Maintain integer precision
+	  for clang error (VALUE aka unsigned long vs unsigned int)
+
+Sat Dec  1 02:53:18 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/test_gem_dependency_installer.rb:  Use Gem.read_binary
+	  instead of File.binread for ruby 1.8 compatibility in the rubygems
+	  source repository.  Updates r38075
+
+Sat Dec  1 02:33:20 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_threadptr_interrupt_mask, async_interrupt_timing_func):
+	  merge into them into rb_thread_s_async_interrupt_timing.
+	* thread.c (rb_thread_s_async_interrupt_timing): ditto.
+
+Sat Dec  1 02:11:47 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_threadptr_interrupt_mask): add argument check.
+	* thread.c (async_interrupt_timing_arg_check_i): helper function
+	  for the above.
+	* test/ruby/test_thread.rb (test_async_interrupt_timing_invalid_argument):
+	  test for the above.
+
+Sat Dec  1 01:19:34 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/thread.rb (ConditionVariable#broadcast): protect from
+	  async interrupt by using Thread.async_interrupt_timing.
+	* lib/thread.rb (ConditionVariable#signal): ditto.
+	* lib/thread.rb (ConditionVariable#wait): ditto.
+
+Sat Dec  1 02:04:23 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_in_out_err):
+	  raise if assert_in_out_err misused.
+
+Sat Dec  1 02:08:16 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rdoc/test_rdoc_rubygems_hook.rb
+	  (TestRDocRubygemsHook#test_setup_unwritable): 1. check the existence
+	  of the file(directory) before touch it.  2. remove test
+	  file(directory) after the test.  see [ruby-core:50388].
+
+Sat Dec  1 01:51:06 2012  Koichi Sasada  <ko1@atdot.net>
+
+	[EXPERIMENTAL]
+	* iseq.c: add following two methods.
+	  * ISeq#line_trace_all returns all line traces (line numbers)
+	  * ISeq#line_trace_specify(pos, set) set `pos'th line event to
+	    specified_line event (if set is true).
+	  These features are introduced for debuggers (mainly to make
+	  breakpoint).
+
+	* iseq.h: add decl. of C APIs.
+
+	* test/ruby/test_iseq.rb: add tests.
+
+	* vm_trace.c: add `specified_line' event.
+
+	* include/ruby/ruby.h: ditto.
+
+Sat Dec  1 01:49:52 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_dependency_installer.rb: gems are of course
+	  binary files, so use a binary reading method when reading it.
+	  see [ruby-core:50388].
+
+Sat Dec  1 01:21:07 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rubygems/command.rb (Gem::Command#get_all_gem_names_and_versions):
+	  who assumes that the pathname of a gem never contains ':' ?
+	  yes, on Unixen pathnames can contain ':', and on Windows they almost
+	  certainly contain ':'.  see [ruby-core:50388].
+
+	* lib/rubygems/requirement.rb (Gem::Requirement::PATTERN_RAW): extract
+	  the regexp to match the version specifier from PATTERN to use in
+	  above method.
+
+Sat Dec  1 00:48:19 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/fiddle/extconf.rb, ext/fiddle/function.c
+	  (Fiddle::Function::STDCALL): FFI_STDCALL is not a macro, but an
+	  enumeration. [ruby-core:50398] [Bug #7483]
+
+Sat Dec  1 00:08:55 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_installer.rb
+	  (TestGemInstaller#test_check_executable_overwrite_other_non_gem):
+	  on Windows, rubygems always generate a wrapper .bat file when
+	  installing a file into bin, so testing no-overwrite a wrapper file
+	  and a non-wrapper file is nonsense.  see [ruby-core:50388].
+
+Fri Nov 30 23:39:58 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_installer.rb
+	  (TestGemInstaller#test_check_executable_overwrite_default_bin_dir):
+	  if the executable to be overwritten was generated by rubygems, the
+	  error message differs from the only copied one's.
+	  see [ruby-core:50388].
+
+Fri Nov 30 23:27:26 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_ext_ext_conf_builder.rb
+	  (TestGemExtExtConfBuilder::test_class_make): reading with binary mode
+	  of course introduce \r on Windows.  see [ruby-core:50388].
+
+Fri Nov 30 23:11:37 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rubygems/specification.rb
+	  (Gem::Specification.validate_permissions): don't check executability
+	  of the source on Windows.  they will be wrapped to .bat files when
+	  installing.  see [ruby-core:50388].
+
+Fri Nov 30 22:44:14 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (rb_vm_struct): add thread_destruct_lock field.
+	* thread.c (Init_Thread): ditto.
+	* thread.c (rb_vm_gvl_destroy): ditto.
+
+	* thread.c (thread_start_func_2): make sure vm->running_thread
+	  don't point to dead thread.
+	* thread.c (timer_thread_function): close a race against thread
+	  destruction. [Bug #4911][ruby-dev:43859]
+
+	* vm_core.h (rb_thread_set_current): reset running time of
+	  current thread instead of previous thread. We no longer
+	  assume previous running thread still live.
+
+Fri Nov 30 21:57:43 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* revert r38053 because it causes too many test failures.
+	  if you've already installed r38053 or later, remove the installed
+	  lib/ruby/gems/2.0.0 directory and reinstall this revision or later.
+
+Fri Nov 30 21:07:56 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit/parallel.rb (Test::Unit::Worker.run): wrap LoadError
+	  because it's Gem::LoadError sometimes.  see [Bug #6882]
+
+Fri Nov 30 20:47:44 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: TracePoint#self returns invoking/exiting thread object
+	  at thread_begin/end event.
+
+	* test/ruby/test_settracefunc.rb: fix test.
+
+Fri Nov 30 19:55:17 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/memory_status.rb (Memory::Win32): use fiddle instead of dl,
+	  but I doubt fiddle is not compatible with dl. (if you are interested,
+	  see the diff.)  [ruby-core:50194] [Bug #7443]
+
+Fri Nov 30 19:37:44 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* defs/default_gems: Add base directory column.
+
+	* tool/rbinstall.rb:
+	  - Install .gemspecs of default gem to
+	    #{GEM_HOME}/specifications/default/.
+	  - Update files parameter of .gemspecs by relative path from
+	    library directory.
+
+Fri Nov 30 19:30:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm_trace.c:
+	  tracepoint_attr_return_value (TracePoint#return_value):
+	    include `:b_return` for method doc
+	  tracepoint_enable_m, tracepoint_disable_m (#enable/#disable):
+	    don't have block argument, document block scope
+
+Fri Nov 30 18:52:56 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (tracepoint_disable_m, tracepoint_enable_m):
+	  fix block parameter.
+	  No argument should be given to a block which is passed
+	  to TracePoint#enable (and disable).
+
+Fri Nov 30 18:23:26 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: rename Thread.control_interrupt
+	  to Thread.async_interrupt_timing.
+	  The option name `:never' is also changed to  `:defer'.
+	  [ruby-core:50375] [ruby-trunk - Feature #6762]
+
+	* thread.c: remove Thread.check_interrupt.
+	  This method is difficult to understand by name.
+
+	* thread.c: add Thread.async_interrupted?.
+	  This method check any deferred async interrupts.
+
+	* test/ruby/test_thread.rb: change tests for above.
+
+Fri Nov 30 18:24:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm_trace.c: Documentation for TracePoint API
+	  [ruby-core:47243] [Feature #6895]
+
+Fri Nov 30 17:43:50 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_cmp_m): try to compare with to_str result if
+	  possible before calling <=> method.  [ruby-core:49279] [Bug #7342]
+
+	* string.c (rb_str_cmp_m): use rb_check_funcall instead of respond_to
+	  and call.
+
+	* string.c (rb_str_cmp_m): return fixed value, one of -1,0,+1 always.
+
+Fri Nov 30 16:19:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): get rid of calling methods in sigsegv
+	  handler.  based on a patch by charliesome (Charlie Somerville)
+	  [ruby-core:49573] [Bug #7402]
+
+Fri Nov 30 16:05:44 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS:  Added RubyGems 2.0.0
+
+Fri Nov 30 15:24:37 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): fix false usage of local variable, it cannot
+	  appear in fname state [ruby-core:49659] [Bug #7408]
+
+Fri Nov 30 15:20:12 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/package.rb:  Load YAML for building gems.
+	* test/rubygems/test_gem_commands_contents_command.rb:  Sort expected
+	  output of default gem contents.  Re-fixes r38004 and r38005.
+
+Fri Nov 30 15:15:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm_trace.c (set_trace_func): Formatting of params and events
+
+Fri Nov 30 14:45:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/net/http.rb: Net::HTTP::Patch to list of HTTP Request Classes
+	  Patch by Ryunosuke SATO [Fixes #217 on github]
+
+Fri Nov 30 14:05:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/cgi.rb: CGI example for HTML generation
+	  Patch by Marcus Stollsteimer [ruby-core:50303] [Bug #7465]
+
+Fri Nov 30 13:52:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* time.c: Documentation improvements, grammar and formatting
+	  Patch by Bernd Homuth [ruby-core:49203] [Bug #7326]
+
+Fri Nov 30 13:48:33 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc.rb:  Set version to 4.0.0.preview2
+	* lib/rubygems.rb:  Set version to 2.0.0.preview2
+
+Fri Nov 30 13:11:53 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/setup_command.rb:  Remove old files on install
+	  of RubyGems.  (not by rbinstall.rb).
+	* test/rubygems/test_gem_commands_setup_command.rb:  Test for above.
+
+Fri Nov 30 12:47:59 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/abbrev.rb (Abbrev#abbrev): A fixed string prefix pattern
+	  should only match the beginning of each word, not the beginning
+	  of every line in it.
+
+	* lib/abbrev.rb (Abbrev#abbrev): Stop using a regexp that causes a
+	  false warning. [Bug #7471]
+
+Fri Nov 30 12:30:55 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/test_abbrev.rb: Add tests for lib/abbrev.rb.
+
+Fri Nov 30 12:27:51 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/spec_fetcher.rb:  Allow prerelease spec fetching to fail
+	  for bundler.
+	* test/rubygems/test_gem_spec_fetcher.rb:  Test for above.
+
+Fri Nov 30 12:20:53 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake/backtrace.rb:  Removed duplication in
+	  Rake::Backtrace::SUPPRESSED_PATHS
+	* test/rake/test_rake_backtrace.rb:  Skip tests when tmpdir is in the
+	  suppressed pattern.
+
+Fri Nov 30 11:07:45 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* revert r37993 to avoid SEGV in tests.
+
+Fri Nov 30 10:38:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/ri/driver.rb:  Relaxed matching for pages to be more
+	  user-friendly.
+	* test/rdoc/test_rdoc_ri_driver.rb:   Test for above.
+
+Fri Nov 30 09:50:16 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markdown.rb:  Fixed warnings with -w
+
+Fri Nov 30 09:38:22 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_GC_GUARD_PTR): add note.
+
+	* vm_backtrace.c (backtrace_to_str_ary): use RB_GC_GUARD() instead of
+	  RB_GC_GUARD_PTR() which has no effect.
+	  (backtrace_to_location_ary): ditto.
+	  (vm_backtrace_to_ary): ditto.
+
+Fri Nov 30 09:22:52 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/contents_command.rb:  Sort output from command.
+	  Replaces r38004, r38005
+	* test/rubygems/test_gem_commands_contents_command.rb:  ditto.
+
+	* lib/rubygems/defaults.rb:  Use Gem.path_separator for jruby support.
+	* lib/rubygems/path_support.rb:  ditto
+
+Fri Nov 30 08:34:03 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/generator/darkfish.rb:  Silenced warning
+	* test/rdoc/test_rdoc_rdoc.rb:  ditto
+
+	* lib/rdoc/markup/parser.rb:  Use byteslice when available for
+	  performance
+	* test/rdoc/test_rdoc_markup_parser.rb:  Test for above
+	* lib/rdoc/test_case.rb:  ditto
+
+	* lib/rdoc/parser/ruby.rb:  Fixed bug parsing yield({})
+	* test/rdoc/test_rdoc_parser_ruby.rb (end):
+
+	* lib/rdoc/rubygems_hook.rb:  Skip default gems.  Display generator
+	  name properly.
+	* test/rdoc/test_rdoc_rubygems_hook.rb:  Test for above
+
+	* lib/rdoc/servlet.rb:  Fixed typo.
+
+Fri Nov 30 08:09:56 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c : remove a unused function.
+
+Fri Nov 30 07:46:42 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (rb_objspace_call_finalizer): finalize_deferred may free up
+	  a object which is reachable from a part after this function,
+	  e.g. ruby_vm_destruct(). [ruby-dev:46647] [Bug #7452]
+
+	* test/ruby/test_gc.rb (test_finalizing_main_thread): add a test
+	  for above.
+
+Fri Nov 30 07:43:44 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_interrupted): avoid warning of
+	  implicit conversion.
+
+	* thread.c (rb_threadptr_execute_interrupts): ditto.
+
+Fri Nov 30 07:34:28 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: add GC guards.
+
+Fri Nov 30 07:21:33 2012  Koichi Sasada  <ko1@atdot.net>
+
+	[EXPERIMENTAL: NEED DISCUSS]
+	* vm_trace.c: add events
+	  * :thread_begin - hook at thread beginning.
+	  * :thread_end - hook at thread ending.
+	  * :b_call - hook at block enter.
+	  * :b_return - hook at block leave.
+	  This change slow down block invocation.
+	  Please try and give us feedback until 2.0 code freeze.
+
+	* include/ruby/ruby.h: ditto.
+
+	* compile.c (rb_iseq_compile_node): ditto.
+
+	* insns.def: ditto.
+
+	* thread.c: ditto.
+
+	* vm.c: ditto.
+
+	* include/ruby/debug.h: add a comment.
+
+	* test/ruby/test_settracefunc.rb: add a tests.
+
+Fri Nov 30 06:56:30 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* test/minitest/*: Imported minitest 4.3.2 (r8027)
+
+Fri Nov 30 04:16:29 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake/*:  Updated to rake 0.9.5
+	* test/rake/*:  ditto.
+	* NEWS:  ditto.
+
+Fri Nov 30 02:53:47 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* vm.c: add a return hook when a method raises an exception.
+
+	* probes_helper.h: look up klass and method if none are provided.
+
+	* eval.c: update macro usage.
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* test/dtrace/test_function_entry.rb: test for change.
+
+Fri Nov 30 02:27:12 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* compile.c (compile_array_): refix r37991 remove assertion:
+	  it is true only if type == COMPILE_ARRAY_TYPE_HASH.
+	  [ruby-dev:46658] [Bug #7466]
+
+	* vm.c (m_core_hash_from_ary): add assertion instead of above.
+
+	* vm.c (m_core_hash_merge_ary): ditto.
+
+Thu Nov 29 19:15:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_array_): hash elements must be paired even for
+	  literal elements.  [ruby-dev:46658] [Bug #7466]
+
+Thu Nov 29 22:39:35 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/openssl/ossl_ssl.c (ssl_npn_encode_protocol_i): fix byte order
+	  issue on big-endian architecture [ruby-core:50292] [Bug #7463]
+
+Thu Nov 29 22:23:31 2012  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* test/openssl/test_cipher.rb (test_ctr_if_exists): add CTR mode test
+	  if underlying OpenSSL supports it. See #4408
+
+Thu Nov 29 21:42:16 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_make): add a method entry with
+	  VM_METHOD_TYPE_REFINED to the class refined by the refinement if
+	  the target module is a refinement.  When a method entry with
+	  VM_METHOD_TYPE_UNDEF is invoked by vm_call_method(), a method with
+	  the same name is searched in refinements.  If such a method is
+	  found, the method is invoked.  Otherwise, the original method in
+	  the refined class (rb_method_definition_t::body.orig_def) is
+	  invoked.  This change is made to simplify the normal method lookup
+	  and to improve the performance of normal method calls.
+
+	* vm_method.c (EXPR1, search_method, rb_method_entry),
+	  vm_eval.c (rb_call0, rb_search_method_entry): do not use
+	  refinements for method lookup.
+
+	* vm_insnhelper.c (vm_call_method): search methods in refinements if
+	  ci->me is VM_METHOD_TYPE_REFINED.  If the method is called by
+	  super (i.e., ci->call == vm_call_super_method), skip the same
+	  method entry as the current method to avoid infinite call of the
+	  same method.
+
+	* class.c (include_modules_at): add a refined method entry for each
+	  method defined in a module included in a refinement.
+
+	* class.c (rb_prepend_module): set an empty table to
+	  RCLASS_M_TBL(klass) to add refined method entries, because
+	  refinements should have priority over prepended modules.
+
+	* proc.c (mnew): use rb_method_entry_with_refinements() to get
+	  a refined method.
+
+	* test/ruby/test_refinement.rb (test_inline_method_cache): do not skip
+	  the test because it should pass successfully.
+
+	* test/ruby/test_refinement.rb (test_redefine_refined_method): new
+	  test for the case a refined method is redefined.
+
+Thu Nov 29 17:45:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_const_set): show namespace in warning messages.
+	  [Feature #7190]
+
+Thu Nov 29 17:31:53 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems.rb (Gem.load_yaml): return if Kernel#gem is not defined
+	  yet. This causes crash if test-all requires libraries in a certain
+	  order. A simple reproducible code is
+	  ruby --disable-gem -e'require"yaml";require"minitest/autorun"'
+
+Thu Nov 29 17:19:26 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/tracer.rb:  Updated to match removal of custom_require from
+	  RubyGems.
+	* test/test_tracer.rb:  ditto.  Improved failure message if the test
+	  fails
+
+Thu Nov 29 17:15:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* gc.c: Documentation for GC, GC::Profiler, ObjectSpace, and
+	  ObjectSpace::WeakMap [ruby-core:50245] [Bug #7449]
+
+Thu Nov 29 17:12:26 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/generic_erb.rb, tool/id2token.rb: add --path-separator option
+	  for mingw where make and built ruby live in different world.
+
+	* tool/vpath.rb: extract from tool/instruction.rb.
+
+Thu Nov 29 17:11:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/io/wait/test_io_wait.rb (TestIOWait#fill_pipe):
+	  Errno::EWOULDBLOCK may not be the same as Errno::EAGAIN.  patch by
+	  phasis68 (Heesob Park) at [ruby-core:49894].  [Bug #7420]
+
+Thu Nov 29 17:03:38 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/test_case.rb:  Determine path to certificates to avoid
+	  build-dir problems.
+	* test/rubygems/test_gem_security_signer.rb:  Use predetermined paths
+	  to avoid build-dir problems.
+
+Thu Nov 29 16:18:14 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/test_case.rb:  Disable loading of keys and certificates
+	  outside rubygems or ruby tests as the files are not available (or
+	  necessary).
+
+Thu Nov 29 16:14:41 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c (rb_debug_inspector_open): use RARRAY_LENINT() for
+	  int variable.
+
+Thu Nov 29 15:59:55 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/debug.h: add rb_debug_inspector_* APIs.
+
+	* vm_backtrace.c: ditto.
+
+	* common.mk: add dependency from vm_backtrace.o to
+	  include/ruby/debug.h.
+
+	* proc.c (rb_binding_new_with_cfp): constify.
+
+	* vm.c (rb_vm_get_ruby_level_next_cfp): constify.
+
+	* vm_core.h, vm_trace.c: move decls.
+
+Thu Nov 29 15:56:14 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/test_case.rb (RDoc::TestCase#verbose_capture_io):
+	  defined for asserts of warnings.
+
+	* test/rdoc: use verbose_capture_io on asserts of warnings.
+	  they failed when tests was run with RUBYOPT=-W0.
+
+Thu Nov 29 15:53:38 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/psych/extconf.rb: added --enable-bundled-libyaml option.  this
+	  enforces using bundled libyaml.
+
+Thu Nov 29 15:51:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems*:  Updated to RubyGems 2.0
+	* test/rubygems*:  ditto.
+
+	* common.mk (prelude):  Updated for RubyGems 2.0 source rearrangement.
+
+	* tool/change_maker.rb:  Allow invalid UTF-8 characters in source
+	  files.
+
+Thu Nov 29 15:38:14 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/debug.h: provide rb_tracearg_*() APIs,
+	  instead of rb_tracepoint_attr_*().
+	  These APIs are for debuggers/profilers.
+	  They will be explained in another docs sometime.
+
+	* vm_trace.c: ditto.
+
+Thu Nov 29 15:10:45 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/minitest/test_minitest_unit.rb: restore orig_verbose only
+	  if it is set. This broke rdoc's tests.
+	  http://u64.rubyci.org/~chkbuild/ruby-trunk/log/20121129T050102Z.diff.html.gz
+
+Thu Nov 29 14:56:30 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_tracepoint_attr_method_id):
+	  rename TracePoint#id to TracePoint#method_id.
+
+	* include/ruby/debug.h: ditto.
+
+	* test/ruby/test_settracefunc.rb: ditto,
+
+Thu Nov 29 14:49:10 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_tracepoint_attr_defined_class):
+	  rename TracePoint#klass to TracePoint#defined_class.
+	  [ruby-core:50187] Re: [ruby-trunk - Feature #6895]
+
+	* include/ruby/debug.h: ditto.
+
+	* test/ruby/test_settracefunc.rb: ditto.
+
+Thu Nov 29 14:27:57 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): prepare Symbol objects at first time
+	  to make it fast.
+
+Thu Nov 29 14:02:15 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): GC.stat supports new information
+	  * total_allocated_object: total allocated object number.
+	  * total_freed_object: total freed object number.
+	  Above two numbers are only accumulated and they will
+	  overflow (return to 0). Please use them as a hint.
+
+Thu Nov 29 12:13:54 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 4.3.2 (r8026)
+	* test/minitest/*: ditto
+
+Thu Nov 29 11:06:06 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_start_func_2): remove unused code. When
+	  th->safe_level == 4, th->errinfo never be thrown. So, to
+	  create new exception makes no sense.
+
+Thu Nov 29 10:29:53 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: same as a last patch.
+
+Thu Nov 29 10:24:25 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: use `long' for return values of `NUM2LONG()'.
+
+Thu Nov 29 09:52:08 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (do_select): suppress warning (uninitialized value warning)
+	  with UNINITIALIZED_VAR().
+
+Thu Nov 29 09:36:09 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (ruby_cleanup): delay THREAD_KILLED timing.
+	  It should be located just before rb_thread_terminate_all().
+
+Thu Nov 29 09:10:17 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c (vm_backtrace_to_ary): support range argument
+	  like Array#[].
+	  [ruby-core:50092] [ruby-trunk - Feature #7434]
+	  Test and document is not available. Please help us.
+
+Thu Nov 29 06:46:33 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_start_func_2): small cleanups.
+
+Thu Nov 29 06:37:08 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_start_func_2): remove unused code.
+	  this function never be used for main thread.
+
+Thu Nov 29 06:27:55 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_start_func_2): remove unused code.
+	  errinfo = th->errinfo; and errinfo = rb_errinfo(); are
+	  the same.
+
+Thu Nov 29 05:26:32 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c (backtrace_to_str_ary2): rename to backtrace_to_str_ary.
+
+	* vm_backtrace.c (rb_backtrace_to_str_ary): use `backtrace_to_str_ary()'.
+
+	* vm_backtrace.c (backtrace_to_frame_ary): rename to
+	  backtrace_to_location_ary.
+
+Thu Nov 29 05:19:25 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* probes.d: Change function-entry probe to method-entry.
+	* insns.def: ditto
+	* probes_helper.h: ditto
+	* test/dtrace/test_function_entry.rb: ditto
+	* test/dtrace/test_singleton_function.rb: ditto
+	* vm.c: ditto
+	* vm_eval.c: ditto
+	* vm_insnhelper.c: ditto
+
+Thu Nov 29 04:45:17 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: rename Class name from
+	  ::RubyVM::Backtrace and ::RubyVM::Backtrace::Location
+	  to ::Thread::Backtrace and ::Thread::Backtrace::Location.
+
+Wed Nov 28 23:52:02 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* NEWS (Thread) remove incompatible changes about trap.
+
+Wed Nov 28 23:39:01 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_mutex_sleep): fix to allow spurious wakeup.
+	  [ruby-dev:46654] [ruby-trunk - Bug #7455]
+
+	* NEWS: write about spurious wakeup.
+
+Wed Nov 28 22:57:23 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_win32.c: catch up latest change of BLOCKING_REGION.
+
+Wed Nov 28 22:54:21 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method): fix undefined behavior.
+	  Should not access scope local variable from outer scope.
+
+Wed Nov 28 22:20:55 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* test/ruby/test_thread.rb (test_thread_status_in_trap): change test for
+	  thread status in trap. now can accept Thread#join and Thread#value in trap.
+
+Wed Nov 28 21:58:47 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/thread.h (rb_thread_call_without_gvl2): change
+	  meaning of function.
+	  This function is called with same parameters of
+	  `rb_thread_call_without_gvl()'.
+	  However, if interrupts are detected, then return immediately.
+
+	* thread.c: implement `rb_thread_call_without_gvl2()'.
+
+Wed Nov 28 21:31:21 2012  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* thread.c (thread_join_sleep): check spurious wakeup by itself for
+	corresponding status change in trap context.
+	* vm_core.h (struct rb_thread_struct): add rb_thread_list_t and use as join_list for
+	reentry by trap context.
+	* thread.c (thread_start_func_2): ditto.
+	* thread.c (remove_from_join_list): ditto.
+	* thread.c (rb_thread_atfork): ditto.
+	* thread.c (thread_join): ditto. & remove trap handler check.
+	* thread.c (sleep_forever): add argument : spurious_check.
+	* thread.c (sleep_timeval): ditto.
+	* thread.c (rb_thread_sleep_forever): set spurious_check.
+	* thread.c (rb_thread_sleep_deadly): ditto.
+	* thread.c (sleep_for_polling): ditto.
+	* thread.c (rb_thread_wait_for): ditto.
+	* thread.c (sleep_wait_for_interrupt): bypass spurious_check.
+
+Wed Nov 28 21:23:18 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/psych/yaml/emitter.c (yaml_emitter_write_indicator): constify.
+
+	* ext/psych/yaml/emitter.c (yaml_emitter_write_block_scalar_hints):
+	  ditto.
+
+	* ext/psych/extconf.rb: mingw32 also needs macros for win32, not
+	  only mswin32.
+
+	* ext/psych/extconf.rb: compile sources in the source directory
+	  without copying by using VPATH.
+
+Wed Nov 28 21:18:57 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#each_compile_rules): splat $(*VPATH*) for
+	  each VPATH elements.
+
+Wed Nov 28 16:40:14 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (enum rb_thread_status): remove THREAD_TO_KILL
+	* vm_core.h (struct rb_thread_struct): add to_kill field
+	* thread.c (terminate_i): convert THREAD_TO_KILL to to_kill.
+	* thread.c (rb_threadptr_to_kill): ditto.
+	* thread.c (rb_thread_kill): ditto.
+	* thread.c (rb_thread_wakeup_alive): ditto.
+	* thread.c (thread_list_i): ditto.
+	* thread.c (static const char): ditto.
+	* thread.c (thread_status_name): ditto.
+	* thread.c (rb_thread_status): ditto.
+	* thread.c (rb_thread_inspect): ditto.
+	* vm_backtrace.c (thread_backtrace_to_ary): ditto.
+
+	* thread.c (rb_threadptr_execute_interrupts): fix thread status
+	  overwritten issue. [Bug #7450] [ruby-core:50249]
+
+	* test/ruby/test_thread.rb (test_hread_status_raise_after_kill):
+	  test for the above.
+	* test/ruby/test_thread.rb (test_thread_status_in_trap): test for
+	  thread status in trap.
+	* test/ruby/test_thread.rb (test_status_and_stop_p): remove
+	  Thread.control_interrupt unsafe test. Thread#kill no longer
+	  changes thread status. Instead of, Thread#kill receiver changes
+	  their own status when receiving kill signal.
+
+Wed Nov 28 16:21:46 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (struct rb_mutex_struct): add allow_trap field.
+	* internal.h (rb_mutex_allow_trap): added.
+	* thread.c (rb_mutex_lock, rb_mutex_unlock): check mutex->allow_trap.
+	* thread.c (mutex_sleep): remove trap check because it uses
+	  rb_mutex_lock and rb_mutex_unlock internally.
+	* thread.c (rb_mutex_allow_trap): new helper function for the above.
+
+	* io.c (io_binwrite): mark fptr->write_lock as writable in trap.
+
+	* test/ruby/test_signal.rb (test_trap_puts): test for the above.
+
+Wed Nov 28 16:59:12 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: remove Proc#== and Proc#eql?.
+	  Proc objects compared with their object ids.
+	  [Bug #4559]
+
+	* test/ruby/test_proc.rb: remove related test.
+
+Wed Nov 28 16:41:04 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/servlet.rb:  Add support for serving documentation from a
+	  subdirectory.
+	* lib/rdoc/generator/darkfish.rb:  ditto
+	* test/rdoc/test_rdoc_servlet.rb:  Test for above
+	* test/rdoc/test_rdoc_servlet.rb:  ditto
+
+Wed Nov 28 15:37:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: fix r37924: run only on i[3-6]86-linux.
+
+Wed Nov 28 15:31:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (FMODE_SYNCWRITE): removed unused macro.
+
+Wed Nov 28 15:19:25 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: revert r37911, r37906 and r37904 which break build
+	  with non-gcc and/or non-IA32 compilers, e.g. Solaris Studio,
+	  Fujitsu C Compiler. [ruby-dev:46646] [Bug #7451]
+
+Wed Nov 28 14:50:55 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/psych/extconf.rb: copy sources into build directory,
+	  not into srcdir.
+
+Wed Nov 28 14:34:06 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_lock): moved trap context check from
+	  rb_mutex_trylock because try_lock have no change to make
+	  a deadlock.
+	* thread.c (rb_mutex_trylock): ditto.
+	* NEWS: news for the above.
+
+Wed Nov 28 13:39:54 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_s_new): uses main_thread->status instead of
+	  th->inhibit_thread_creation for preventing thread creation.
+	* vm_core.h (rb_vm_struct): remove inhibit_thread_creation field.
+	* thread.c (rb_thread_terminate_all): ditto.
+
+Wed Nov 28 13:27:29 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/extconf.rb: use embedded libyaml if no system libyaml is
+	  found. [ruby-core:49463]
+	* ext/psych/lib/psych.rb: updating to psych 2.0.0
+	* ext/psych/lib/psych/deprecated.rb: updated docs
+	* ext/psych/psych.gemspec: updated to psych 2.0.0
+	* ext/psych/psych.h: fixing header file include for rename
+	* ext/psych/psych_emitter.c: renamed to avoid libyaml conflict.
+	* ext/psych/psych_emitter.h: ditto
+	* ext/psych/psych_parser.c: ditto
+	* ext/psych/psych_parser.h: ditto
+	* ext/psych/psych_to_ruby.c: ditto
+	* ext/psych/psych_to_ruby.h: ditto
+	* ext/psych/psych_yaml_tree.c: ditto
+	* ext/psych/psych_yaml_tree.h: ditto
+	* ext/psych/yaml/LICENSE: embedding libyaml 0.1.4
+	* ext/psych/yaml/api.c: ditto
+	* ext/psych/yaml/config.h: ditto
+	* ext/psych/yaml/dumper.c: ditto
+	* ext/psych/yaml/emitter.c: ditto
+	* ext/psych/yaml/loader.c: ditto
+	* ext/psych/yaml/parser.c: ditto
+	* ext/psych/yaml/reader.c: ditto
+	* ext/psych/yaml/scanner.c: ditto
+	* ext/psych/yaml/writer.c: ditto
+	* ext/psych/yaml/yaml.h: ditto
+	* ext/psych/yaml/yaml_private.h: ditto
+
+Wed Nov 28 12:54:59 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_join): A trap handler check was moved from
+	  thread_join_m because Thread#value should be raised an exception
+	  too.
+	* thread.c (thread_join_m): remove trap handler check.
+	* test/ruby/test_thread.rb (test_thread_join_in_trap): add test
+	  for thread#value.
+	* NEWS: documentation fix for the above.
+
+Wed Nov 28 11:07:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/fiddle/closure.c: Documentation for Fiddle
+	* ext/fiddle/lib/fiddle/import.rb: ditto
+	* ext/fiddle/lib/fiddle/value.rb: ditto
+	* ext/fiddle/lib/fiddle/pack.rb: ditto
+	* ext/fiddle/lib/fiddle/cparser.rb: ditto
+	* ext/fiddle/lib/fiddle/struct.rb: ditto
+	* ext/fiddle/lib/fiddle/function.rb: ditto
+
+Wed Nov 28 09:15:51 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* ext/strscan/strscan.c: Added #charpos for multibyte string position.
+	* test/strscan/test_stringscanner.rb: ditto
+
+Wed Nov 28 09:00:34 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/fiddle.c: adding alignment constants for compatibility
+	  with DL.
+	* ext/fiddle/fiddle.h: ditto
+	* ext/fiddle/lib/fiddle/cparser.rb: importing the C parser for DL
+	  backwards compatibility.
+	* ext/fiddle/lib/fiddle/import.rb: importing the import DSL for DL
+	  backwards compatibility.
+	* ext/fiddle/lib/fiddle/pack.rb: importing structure pack for DL
+	  backwards compatibility.
+	* ext/fiddle/lib/fiddle/value.rb: ditto
+	* ext/fiddle/lib/fiddle/struct.rb: importing struct DSL for DL backwards
+	  compatibility.
+	* test/dl/test_c_struct_entry.rb: importing tests
+	* test/dl/test_c_union_entity.rb: ditto
+	* test/dl/test_cparser.rb: ditto
+	* test/dl/test_import.rb: ditto
+	* test/fiddle/test_c_struct_entry.rb: ditto
+	* test/fiddle/test_c_union_entity.rb: ditto
+	* test/fiddle/test_cparser.rb: ditto
+	* test/fiddle/test_import.rb: ditto
+
+Wed Nov 28 08:56:00 2012  Zachary Scott <zachary@zacharyscott.net>
+
+	* doc/globals.rdoc: Add documentation file for magic globals
+	[ruby-core:29048] [Bug #3022]
+
+Wed Nov 28 08:55:00 2012  Zachary Scott <zachary@zacharyscott.net>
+
+	* .document: Add README's to be included with docs
+
+Wed Nov 28 08:26:00 2012  Zachary Scott <zachary@zacharyscott.net>
+
+	* ext/dl/lib/dl.rb: Deprecation notice for DL
+
+Wed Nov 28 08:25:00 2012  Zachary Scott <zachary@zacharyscott.net>
+
+	* ext/fiddle/closure.c: Documentation for Fiddle
+	* ext/fiddle/pointer.c: ditto
+	* ext/fiddle/function.c: ditto
+	* ext/fiddle/lib/fiddle.rb: ditto
+	* ext/fiddle/fiddle.c: ditto
+	* ext/fiddle/handle.c: ditto
+
+Wed Nov 28 04:53:40 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/handle.c: Make Fiddle independent of DL, copy DL::Handle
+	  to Fiddle::Handle.
+	* ext/fiddle/pointer.c: Make Fiddle independent of DL, copy
+	  DL::Pointer to Fiddle::Pointer.
+	* test/fiddle/test_func.rb: relevant tests
+	* test/fiddle/test_handle.rb: ditto
+	* test/fiddle/test_pointer.rb: ditto
+	* ext/dl/lib/dl/struct.rb: use Fiddle::Pointer if available
+	* ext/fiddle/extconf.rb: check for dlfcn.h
+	* ext/fiddle/fiddle.c: add constants for sizeof() things
+	* ext/fiddle/fiddle.h: include dlfcn.h
+	* ext/fiddle/function.c: expose a C function for creating new
+	  Fiddle::Function objects.
+	* ext/fiddle/lib/fiddle.rb: include constants for dl backwards compat
+	* ext/fiddle/lib/fiddle/function.rb: read the pointer from the
+	  function for dl backwards compat.
+	* test/dl/test_callback.rb: check the addresses of the pointers rather
+	  than their types.
+	* test/fiddle/helper.rb: remove dependency on dl
+	* test/fiddle/test_closure.rb: ditto
+	* test/fiddle/test_fiddle.rb: ditto
+
+Wed Nov 28 03:03:28 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (opt-dir): don't use non portable flag -E of sed.
+
+Wed Nov 28 02:55:35 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (ARCH_FLAG): __sync_val_compare_and_swap_4 needs
+	  -march=$target_cpu on at least linux gcc 4.1.
+	  patched by KOSAKI Motohiro
+
+Tue Nov 27 22:03:09 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c (rb_str_enumerate_chars, rb_str_enumerate_codepoints)
+	  (rb_str_enumerate_lines): Dummy initialization of ary has been
+	  replaced with UNINITIALIZED_VAR().
+
+Tue Nov 27 21:29:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (bigdivrem): optimize the way to retry calculation of
+	  bigdivrem so that the calculation is started from the point where
+	  the last interruption was occurred.
+
+	* bignum.c (bigdivrem1): ditto.
+
+	* test/ruby/test_bignum.rb: add a test case for rb_bigdivrem in the
+	  case that an interruption is occurred during bigdivrem1 is running.
+
+Tue Nov 27 19:56:43 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_make_env_object): make Proc object if Env is possible
+	  to point block.
+	  [ruby-core:41038] [ruby-trunk - Bug #5634]
+
+	* vm.c (rb_vm_make_proc): No need to make Proc object here.
+
+	* bootstraptest/test_proc.rb: add tests.
+
+Tue Nov 27 18:51:06 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ruby_atomic.h (ATOMIC_CAS): added for Solaris and other platforms.
+	* ruby_atomic.h, signal.c (NEED_RUBY_ATOMIC_OPS): renamed from
+	  NEED_RUBY_ATOMIC_EXCHANGE.
+	* signal.c (ruby_atomic_compare_and_swap): naive, non-atomic
+	  compare-and-swap implementation only used for platforms without
+	  valid support for atomic operations.
+
+Tue Nov 27 17:43:46 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/*:  Added --root option for building documentation outside
+	  the source directory.
+	* test/rdoc/*:  ditto
+	* common.mk (rdoc):  Added --root to rdoc rule
+
+Tue Nov 27 16:24:45 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rdoc/test_rdoc_ri_paths.rb:  Fixed duplicate path bug which
+	  caused windows failures.
+
+Tue Nov 27 16:06:00 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rdoc/test_rdoc_generator_darkfish.rb:  Updated tests for windows
+	* test/rdoc/test_rdoc_options.rb:  ditto
+	* test/rdoc/test_rdoc_parser.rb:  ditto
+	* test/rdoc/test_rdoc_rdoc.rb:  ditto
+	* test/rdoc/test_rdoc_ri_driver.rb:  ditto
+	* test/rdoc/test_rdoc_servlet.rb:  ditto
+
+Tue Nov 27 15:13:51 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* common.mk (rdoc):  Set --debug for rdoc generation in case of bugs
+
+Tue Nov 27 14:56:45 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/rubygems_hook.rb:  Updated for (upcoming) RubyGems 2
+	  import.
+	* test/rdoc/test_rdoc_rubygems_hook.rb:  ditto
+
+Tue Nov 27 13:59:29 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* NEWS: add improvements of the garbage collector.
+
+Tue Nov 27 13:27:46 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc*:  Updated to RDoc 4.0 (pre-release)
+	* bin/rdoc:  ditto
+	* test/rdoc*:  ditto
+	* NEWS:  Updated with RDoc 4.0 information
+
+Tue Nov 27 12:17:11 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_terminate_all): retry broadcast only when
+	  an exception is raised.
+
+Tue Nov 27 12:02:25 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_terminate_all): broadcast terminate event
+	  not only an interrupt exception but any exceptions.
+
+Tue Nov 27 10:55:09 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* eval.c (ruby_cleanup): set thread status to THREAD_KILLED
+	  for preventing thr.raise.
+	* test/ruby/test_thread.rb (test_main_thread_status_at_exit):
+	  test for the above.
+
+Tue Nov 27 10:31:29 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_terminate_all): suppress a warning.
+
+Tue Nov 27 09:29:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_join): raises ThreadError if target thread
+	  is a main thread.
+	* test/ruby/test_thread.rb (test_thread_join_main_thread):
+	  test for the above.
+	* NEWS: news for the above.
+
+Tue Nov 27 09:24:47 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_join): raises ThreadError if target thread
+	  is a current thread.
+	* test/ruby/test_thread.rb (test_thread_join_current):
+	  test for the above.
+	* NEWS: news for the above.
+
+Tue Nov 27 09:59:16 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): close mkmf.log for each libraries so that
+	  failure messages are not mixed.
+
+Tue Nov 27 09:58:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/*/extconf.rb, ext/openssl/extconf.rb: get git rid of
+	  post-1.8 feature require_relative for cross compilation.
+	  [ruby-core:50160] [Bug #7439]
+
+Tue Nov 27 09:17:59 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: add TracePoint.
+
+Tue Nov 27 08:16:03 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: rename TracePoint#file and TracePoint#line
+	  to TracePoint#path and TracePoint#lineno respectively.
+	  They are consistent to RubyVM::Backtrace::Location.
+
+	* include/ruby/debug.h: ditto.
+
+	* vm_core.h: ditto.
+
+	* test/ruby/test_settracefunc.rb: ditto.
+
+Tue Nov 27 08:04:26 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_terminate_all): broadcast eTerminateSignal
+	  again when Ctrl-C was pressed. [Feature #1952] [ruby-dev:39107]
+
+Tue Nov 27 07:58:03 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: add members to rb_trace_arg_t:
+	  * `klass_solved' represents klass and id is checked.
+	  * `line' represents line calculated from cfp.
+	  * `file' represents line calculated from cfp.
+
+	* vm_trace.c: fix to use above data structures.
+	  No need to calculate klass and id, line and file
+	  pairs for each trace points.
+
+Tue Nov 27 07:47:09 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_terminate_all): add RUBY_VM_CHECK_INTS_BLOCKING().
+	  Otherwise the loop in this function behave as busy loop because
+	  native_sleep() return immediately when RUBY_VM_INTERRUPTED() is true.
+
+Tue Nov 27 04:12:49 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): git rid of post-1.8 features for cross
+	  compilation.  [ruby-core:50160] [Bug #7439]
+
+Tue Nov 27 00:13:41 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_trylock, rb_mutex_unlock, mutex_sleep):
+	  raises ThreadError if called from trap handler as Thread#join.
+	* NEWS: news for the above.
+
+Mon Nov 26 23:55:33 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* NEWS: update for Thread#join incompatible change.
+
+Mon Nov 26 22:44:24 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_terminate_all): use native_sleep() instead
+	  of rb_thread_schedule(). Otherwise, it consume 100% cpu meaninglessly.
+	  [Bug #5368] [ruby-dev:44546]
+	* thread.c (thread_start_func_2): last sub-thread wakes up main thread.
+
+Mon Nov 26 21:16:04 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (RUBY_VM_SET_TIMER_INTERRUPT, RUBY_VM_SET_INTERRUPT)
+	(RUBY_VM_SET_FINALIZER_INTERRUPT, RUBY_VM_SET_TRAP_INTERRUPT)
+	(RUBY_VM_INTERRUPTED): use enum symbol instead of immediate value.
+	* thread.c (thread_join_m, rb_threadptr_execute_interrupts): ditto.
+	* signal.c (signal_exec): ditto.
+
+Mon Nov 26 20:23:20 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_join_m): use th->interrupt_mask instead of
+	  th->in_trap.
+
+	* vm_core.h (struct rb_thread_struct): remove in_trap member.
+	* signal.c (signal_exec): ditto.
+	* thread.c (thread_create_core): ditto.
+	* thread.c (Init_Thread): ditto.
+
+Mon Nov 26 20:23:49 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_argf.rb (TestArgf#test_chars): since marshal data is
+	  binary, shouldn't pass via text mode.  use base64 encoded data.
+
+Mon Nov 26 19:45:18 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ruby_atomic.h (ATOMIC_CAS): new macro for compare-and-exchange.
+
+	* vm_core.h (struct rb_thread_struct): add interrupt_mask member.
+	* thread.c (thread_create_core, Init_Thread): initialize
+	  th->thread_mask.
+
+	* vm_core.h (RUBY_VM_INTERRUPTED_ANY): new macro for avoiding
+	  bare th->interrupt_flag.
+	* vm_core.h (RUBY_VM_INTERRUPTED, RUBY_VM_INTERRUPTED): check
+	  th->interrupt_mask.
+	* thread.c (set_unblock_function, rb_thread_schedule): replace
+	  th->interrupt_flag with RUBY_VM_INTERRUPTED_ANY()
+
+	* signal.c (signal_exec): set up thread->interrupt_mask for
+	  preventing recursive trap handler.
+	* vm_core.h (RUBY_VM_CHECK_INTS, RUBY_VM_CHECK_INTS_BLOCKING): ditto.
+
+	* thread.c (rb_threadptr_execute_interrupts):
+	  don't process interrupt if it is masked.
+	  [Bug #6009] [ruby-core:42524]
+
+Mon Nov 26 19:43:42 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (make_compile_option_value): add trace_instruction option.
+	  a patch by davidbalbert (David Albert).
+	  [Bug #6786]
+
+Mon Nov 26 19:10:53 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_thread.rb: try to `join' each 100
+	  threads.
+	  This benchmark seems consuming long time on travis-ci
+	  several times (and make `failure').
+
+Mon Nov 26 18:22:56 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: specify label `built-ruby'.
+
+	* benchmark/driver.rb: quote path.
+
+Mon Nov 26 18:26:28 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (signal_exec): add volatile to make sure setjmp safe.
+
+Mon Nov 26 18:19:47 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (signal_exec): suppress "warning: variable 'signum'
+	  might be clobbered by 'longjmp' or 'vfork'" warning.
+
+Mon Nov 26 18:15:47 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: accept multiple `-e'.
+	  You don't need to use `;' separation character.
+	  [ruby-core:50139] [ruby-trunk - Bug #7380]
+
+Mon Nov 26 17:10:04 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* string.c (rb_str_enumerate_chars, rb_str_enumerate_codepoints)
+	(rb_str_enumerate_lines): suppress "may be used uninitialized in
+	  this function" warning.
+
+Mon Nov 26 17:00:12 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (rb_thread_struct): added 'in_trap' member for marking
+	  running trap handler.
+	* signal.c (signal_exec): turn on in_trap when running trap.
+	* thread.c (Init_Thread, thread_create_core): initialize in_trap
+	  when creating new threads.
+	* thread.c (thread_join_m): raise ThreadError when running trap
+	  handler.Bug [#6416][ruby-core:44956]
+	* test/ruby/test_thread.rb (test_thread_join_in_trap): new test
+	  for the above.
+
+Mon Nov 26 16:36:13 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (argf_each_codepoint): add missing ARGF#codepoints [Bug #7438]
+
+Mon Nov 26 15:50:29 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bignum.c (bigdivrem): restart calculation when bigdivrem1 was
+	  interrupted by signal. Otherwise, ruby script may see a garbage
+	  value.
+
+Mon Nov 26 15:33:02 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bignum.c (big_div_struct): added volatile to 'stop' member.
+	  Otherwise, "if (bds->stop)" check in bigdivrem1 don't read
+	  memory and ignore interrupt.
+	* bignum.c (bigdivrem, rb_big_stop): ditto.
+
+Mon Nov 26 12:11:09 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (DLNOBJ): missing in r37821.
+
+Mon Nov 26 10:50:19 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_process.rb (test_setsid): added a few wait for
+	  preventing that Process.getsid(io.pid) makes Errno::ESRCH.
+
+Sun Nov 25 22:34:00 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* array.c (ary_resize_smaller): new function to resize array.
+
+	* array.c (rb_ary_delete): refactoring to extract a function.
+
+	* array.c (rb_ary_delete_same): refactoring.
+	  It renames function, reduces duplicated code and removes unused
+	  code.
+
+	* gc.c (wmap_final_func): follow the above change.
+
+	* internal.h (rb_ary_delete_same): ditto.
+
+Sun Nov 25 22:27:33 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* array.c: fixes for the updated documentation in r35858:
+	  Typos and #take/#drop accept to take/drop 0 elements.
+
+Sun Nov 25 19:43:29 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* NEWS: add a news about iconv.
+
+Sun Nov 25 03:49:23 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb, test/test_set.rb: Move tests embedded in lib/set.rb
+	  to test/test_set.rb.
+
+Sun Nov 25 03:44:50 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c (rb_str_each_line, rb_str_lines): String#lines now
+	  returns an array instead of an enumerator.  Passing a block is
+	  deprecated but still supported for backwards compatibility.
+	  Based on the patch by yhara. [Feature #6670]
+
+	* string.c (rb_str_each_char, rb_str_chars): Ditto for
+	  String#chars.
+
+	* string.c (rb_str_each_codepoint, rb_str_codepoints): Ditto for
+	  String#codepoints.
+
+	* string.c (rb_str_each_byte, rb_str_bytes): Ditto for
+	  String#bytes.
+
+	* NEWS: Add notes for the above changes.
+
+Sun Nov 25 02:07:37 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_warning)
+	  (Test::Unit::Assertions#assert_warn), test/ruby/envutil.rb,
+	  test/ruby/test_enumerator.rb, test/ruby/test_io_m17n.rb,
+	  test/ruby/test_regexp.rb, test/ruby/test_syntax.rb:
+	  assert_warn() and assert_warning() are reversed.
+
+Sat Nov 24 21:08:50 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* gc.c (wmap_final_func): rename variables to clarify the meaning.
+	  In wmap2obj the key is WeakRef and the value is referenced object.
+	  In obj2wmap the key is referenced object and the value is an array
+	  of WeakRef.
+
+	* gc.c (wmap_finalize): ditto.
+	  [ruby-core:49044] [Bug #7304]
+
+Sat Nov 24 21:01:55 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* array.c (rb_ary_delete_same_obj): new function for WeakRef.
+	  This deletes same objects as item argument in the array.
+
+	* internal.h (rb_ary_delete_same_obj): add a declaration.
+
+	* gc.c (wmap_final_func): remove WeakRef object reference from the
+	  array. rb_ary_delete() is not usable because it uses rb_equal() to
+	  compare object references.
+
+	* gc.c (wmap_finalize): remove recycled object references from weak
+	  map hash properly. How to get object reference from object id was
+	  wrong. st_delete() doesn't work properly if key and value arguments
+	  are same. The key of obj2wmap is referenced object and the value of
+	  obj2wmap is WeakRef array.
+
+	* gc.c (wmap_aset): obj2wmap should contain WeakRef array in the
+	  definition.
+
+	* test/test_weakref.rb
+	  (TestWeakRef#test_not_reference_different_object,
+	   TestWeakRef#test_weakref_finalize): add tests for above.
+	  [ruby-core:49044] [Bug #7304]
+
+Sat Nov 24 19:44:41 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (unicode_iconv_combine): returning flags are
+	  nkf_char.
+
+Sat Nov 24 19:29:18 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_rubyoptions.rb (test_usage, test_usage_long):
+	  reduced, renamed.
+
+Sat Nov 24 13:10:14 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_tracepoint_attr_raised_exception): should not check
+	  value before event checking.
+
+	* vm_trace.c (rb_tracepoint_attr_return_value): ditto.
+
+	* test/ruby/test_settracefunc.rb: add tests for TracePoint#return_value
+	  and TracePoint#raised_exception.
+
+Sat Nov 24 12:47:27 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_process.rb (TestProcess#test_setsid): "./ruby-trunk"
+	  doesn't work on all environments. EnvUtil.rubybin would be suitable.
+
+Sat Nov 24 12:28:04 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* array.c (rb_ary_aref): fix Segmentation fault at TestArray#test_aref
+	  on x64 mingw. Variable argument of rb_scan_args() should be a pointer
+	  (VALUE *), but 0 of variable argument seems not equal to null pointer
+	  on x64 mingw.
+
+Sat Nov 24 11:47:14 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (proc_getsid): adds new method for getting session id.
+	  Contributed from fumiyas (Fumiyasu SATOH). Thank you!
+	  [Feature #6757] [ruby-dev:45977]
+	* configure.in: adds getsid check.
+	* test/ruby/test_process.rb (TestProcess#test_setsid): new test
+	  for the above.
+	* NEWS: news for the above.
+
+Sat Nov 24 10:59:14 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_create_core): don't use th->thread_id before
+	  initialized.
+
+Sat Nov 24 00:00:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options, process_options, ruby_process_options): take
+	  care of the case argc is 0, and check if argv has NULL.
+	  [ruby-core:49889] [Bug #7423]
+
+Sat Nov 24 00:00:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--disable-dln): option to disable dynamic linking
+	  feature.  [ruby-core:37676] [Feature #4946]
+
+Fri Nov 23 23:44:41 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#pkg_config): strip all white spaces for
+	  mingw64+MSYS pkg-config which erroneously emits extra newlines.
+	  [ruby-core:47998] [Bug #7163]
+
+Fri Nov 23 17:31:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (usage): wrap description lines if options are too long.
+
+Fri Nov 23 11:13:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_usage2): refine
+	  assertion.
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_usage2): now
+	  --help option is for modern terminals.
+
+Fri Nov 23 10:45:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/socket/ipsocket.c (IPSocket#peeraddr): Fix example
+	  [ruby-core:46429] [Bug #6732]
+
+Fri Nov 23 02:40:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/cgi/core.rb: Documentation for CGI#header alias
+	  Based on a patch by Marcus Stollsteimer
+	  [ruby-core:49585] [Bug #7405]
+
+Thu Nov 22 23:48:10 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (is_swept_object): extract from is_dead_object().
+
+	* gc.c (rb_gcdebug_print_obj_condition): add the function for debug.
+	  This function shows some conditions of given object (e.g.,
+	  marked, in heap, swept).
+
+	* gc.c (rb_gcdebug_sentinel): add the function for debug.
+	  This function allow check to inadvertently free up an object.
+
+Thu Nov 22 23:45:18 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* array.c (rb_ary_shared_with_p): fix cache validity check.
+	  If #pop or #shift has been called against $: or $", the array will
+	  be still shared with the snapshot. We check array length for cache
+	  validity.
+	  [ruby-core:49518] [Bug #7383]
+
+	* test/ruby/test_require.rb
+	  (TestRequire#test_require_with_array_pop,
+	   TestRequire#test_require_with_array_shift): add tests for above.
+
+Thu Nov 22 21:48:48 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk, win32/Makefile.sub (probes.dmyh): now be made in current
+	  (=build) directory if build from the repository.
+
+Thu Nov 22 21:34:51 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/ruby/test_rubyoptions.rb: added a test.
+
+Thu Nov 22 20:32:07 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (string_to_c_strict, string_to_c): check NUL.
+	* rational.c (string_to_r_strict, string_to_r): ditto.
+
+Thu Nov 22 20:21:45 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* Makefile.in (.dmyh.h): removed $(VPATH). GNU make don't
+	  recognize suffix rule with VPATH.
+
+Thu Nov 22 18:11:27 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk, Makefile.in, win32/Makefile.sub (.dmyh.h): nmake merges
+	  explicit rules for same target, but not merges explicit rules and
+	  implicit rules -- always explicit rules win.  So, need to add an
+	  explicit rule for probes.h.  [Bug #7421] [ruby-core:49839]
+
+Thu Nov 22 18:01:28 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (probes.o): add -C to ignore #include in probes.d.
+
+	* probes.d: include vm_opts.h instead of vm_core.h.
+
+	* vm_opts.h (VM_COLLECT_USAGE_DETAILS): move definition from vm_core.h.
+
+Thu Nov 22 17:45:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8: Merge b0a6577a521d1bba5e19853f95d5c4b9be1072b5.
+	  Support JIS X 0213 and some bugfixes.
+
+Thu Nov 22 17:39:37 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* tool/gen_dummy_probes.rb: don't change #include, #if and #endif
+	  lines. [Bug #7370]
+
+Thu Nov 22 16:58:26 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* Makefile.in: run preprocessor when making probe.h
+	* probes.d: define probe insn and insn__operand only when
+	  VM_COLLECT_USAGE_DETAILS is 1. [Bug #7370]
+
+Thu Nov 22 16:20:49 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm.c: Don't define vm_collect_usage_operand() and
+	  static void vm_collect_usage_insn() when disabling
+	  VM_COLLECT_USAGE_DETAILS. (refix r37796)
+
+Thu Nov 22 15:26:02 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_insnhelper.h: partly revert r37631 (DTrace support).
+	  "vm usage information is always collected, so uncomment the
+	  functions." causes performance impact. [Bug #7370]
+	  Off course this revert disables related probes.
+
+Thu Nov 22 14:14:36 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* re.c (rb_memsearch_ss): Apply performance improvement to short
+	  byte array search for platforms without memmem(3).
+	  [Feature #6311] [ruby-dev:45530]
+
+Thu Nov 22 12:52:19 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/ruby/test_string.rb (TestString#test_index): Add some
+	  corner cases to tests for String#index, which might fail if ruby
+	  directly used a buggy memmem(3) implementation.
+
+Thu Nov 22 08:06:42 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* test/ruby/test_gc.rb (test_profiler_clear): fix wrong method
+	  calls [Bug #7419] [ruby-core:49828].
+
+Thu Nov 22 02:22:33 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* NEWS: edited (order etc).
+
+Wed Nov 21 22:52:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/wait/wait.c (io_wait_readable): add alias wait_readable.
+
+	* ext/io/wait/wait.c (io_wait_writable): this is easier to use than
+	  IO.select for a single IO object and is immune to the
+	  limitations/inefficiency of select() on platforms where poll/ppoll
+	  is available.  patched by Eric Wong.  [Feature #4646]
+
+Wed Nov 21 22:27:52 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (garbage_collect): remove a duplicative probe.
+
+Wed Nov 21 22:08:48 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_profile_clear): realloc profile records if its size is
+	  higher than the threshold, GC_PROFILE_RECORD_DEFAULT_SIZE * 2.
+
+Wed Nov 21 21:53:29 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_to_c): added.
+
+Wed Nov 21 21:35:38 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* include/ruby/util.h: removed extra semicolon in definition of
+	  macro.
+	* compile.c: ditto.
+	* cont.c: ditto.
+	* math.c: ditto.
+	* node.c: ditto.
+	* parse.y: ditto.
+
+Wed Nov 21 18:46:37 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (read_digits): due to a bit tighter rb_cstr_to_inum().
+
+Wed Nov 21 16:13:37 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_so_nsieve_bits.rb: add an encoding pragma because
+	  this benchmark using strings (literals) as binary sequence.
+	  Now, they are UTF-8 strings. [ruby-dev:46578]
+
+Wed Nov 21 00:57:43 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (Init_File): null device definition uses rb_define_const
+	  instead of rb_file_const.
+
+Wed Nov 21 00:28:18 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_m17n_comb.rb (test_str_crypt): Use RbConfig to get
+	  libc's directory. Patched by Vit Ondruch [ruby-core:49763] [Bug #7312]
+
+Tue Nov 20 23:28:26 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* marshal.c: add marshal readahead. marshalized Array, Hash and Struct
+	  have size at least number of its elements, marshal readahead will
+	  read the certain readable length and buffer when it needs more bytes.
+	  marshal readahead prevents many calls to IO#getbyte and IO#read,
+	  then it enables performance improvement.
+	  [ruby-dev:45637] [Feature #6440]
+
+Tue Nov 20 22:35:02 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (.d.h): replace char * to const char * because somehow
+	  current dtrace removes const of function declaration in probes.d.
+
+Tue Nov 20 21:41:04 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/debug.h: introduced.
+	  Debugging/profiling features will be located.
+
+	* vm_trace.c: expose C-level TracePoint APIs.
+	  Note that they are experimental.
+
+	* vm_trace.c, include/ruby/debug.h: rename `rb_hook_flag_t'
+	  to `rb_event_hook_flag_t'.
+	  Macro names `RUBY_HOOK_FLAG_*' are also renamed to
+	  `RUBY_EVENT_HOOK_FLAG_*'.
+
+	* debug.h, vm_debug.h: rename debug.h to vm_debug.h.
+
+	* common.mk: ditto.
+
+	* debug.c, main.c, vm_core.h: ditto.
+
+Tue Nov 20 21:12:37 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (RUBY_DTRACE_AVAILABLE): only check dtrace availability.
+
+	* configure.in (RUBY_DTRACE_POSTPROCESS): restore.
+
+Tue Nov 20 21:22:44 2012  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_type.rb (test_implemented_ole_types):
+	  IShellDispatch6 bundled in Windows 8. Thanks to phasis68 (Heesob
+	  Park).  [ruby-core:49580][Bug #7403]
+
+Tue Nov 20 21:06:41 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: some improvements.
+	* rational.c: ditto.
+
+Tue Nov 20 21:01:16 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (incs): BSD make cannot deal with non-prefixed dependency
+	  and prefixed target.
+
+Tue Nov 20 20:10:23 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_bsearch): fix rdoc bug (O(n log n) -> O(log n)).
+	  Patch by Charlie Somerville.  [ruby-core:49661] [Bug #7409]
+
+	* range.c (range_bsearch): ditto.
+
+Tue Nov 20 19:02:44 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: rename and add TracePoint APIs.
+	  (1) TracePoint.new(...){...} creates a new trace point
+	      but does not make it enable.
+	  (2) TracePoint.trace(...){...} creates a new trace point
+	      and enable it (same as old behavior).
+	  (3) TracePoint#enable make it enable (renamed from TracePoint#retrace).
+	      If block given, when enable only in block.
+	  (4) TracePoint#disable make it disable (renamed from TracePoint#untrace).
+	      If block given, when disable only in block.
+	  (5) TracePoint#enabled? returns this trace is enable or not.
+	  [Feature #6895]
+
+	* test/ruby/test_settracefunc.rb: add tests.
+
+Tue Nov 20 18:35:05 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: add two methods:
+	  (1) TracePoint#return_value which returns return
+	      value on the :return and :c_return event.
+	  (2) TracePoint#raised_exception which returns raised exception
+	      value on the :raise event.
+	  Each methods raise RuntimeError if it is called at unsupported
+	  event.
+	  Please review and give us feedback until next preview
+	  release (Dec/2012) of Ruby 2.0.0.
+	  [Feature #6895]
+
+	* insns.def, vm.c, vm_eval.c, vm_insnhelper.c, eval.c, thread.c:
+	  ditto.
+
+	* vm_trace.c, vm_core.h: move definition of rb_trace_arg_t from
+	  vm_trace.c to vm_core.h.
+	  Caller fills rb_trace_arg_t and pass the pointer of this variable.
+
+	* test/ruby/test_settracefunc.rb: fix tests to test this change.
+
+Tue Nov 20 17:31:12 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: fix dtrace didn't work on darwin.
+
+	* configure.in (RUBY_DTRACE_AVAILABLE): unify RUBY_DTRACE_POSTPROCESS
+	  and RUBY_DTRACE_BSD_BROKEN.
+
+Tue Nov 20 15:20:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* file.c (File.extname): Documentation for extname on dotfiles and
+	files ending with a dot. Also, added example for this.
+	[ruby-core:47852] [Bug #7112]
+
+Tue Nov 20 14:19:20 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_signal.rb (TestSignal#test_signame): fix windows
+	  test failure. Process.kill on windows can't send a signal to
+	  another process.
+
+Tue Nov 20 13:58:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (Init_IO): moved all rb_file_const() into file.c.
+	* file.c (Init_File): replace with rb_file_const() with
+	  rb_define_const() because RDoc don't care rb_file_const.
+	  [Bug #5530]
+
+Tue Nov 20 12:35:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ruby.c (usage_msg): Fix typo [ruby-core:49205] [Bug #7327]
+
+Tue Nov 20 12:35:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* file.c (File::NULL): Document File::NULL constant
+	  [ruby-core:49384] [Bug #7365]
+
+Tue Nov 20 12:05:15 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): Windows 8 fixed one of a bug of console
+	  API.
+	  based on the patch by Heesob Park at [ruby-core:49479] [Bug #7379]
+
+Tue Nov 20 11:14:33 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (--enable-dtrace): always call RUBY_DTRACE_BSD_BROKEN
+	  for portability.
+	  As the note, FreeBSD 8 has DTrace as the optional
+	  feature (it is enabled by the build option), but doesn't have USDT.
+	  FreeBSD 9 has USDT but they are still optional. FreeBSD 10 will
+	  enable them by default.
+	  The variable $rb_cv_prog_dtrace_g is "yes" only on FreeBSD 9 with
+	  optional DTrace or FreeBSD 10. If it is "no", you cannot know
+	  whether it doesn't need -G or DTrace is disabled. (by checking
+	  error code, you can know)
+
+Mon Nov 19 22:55:48 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in (--enable-dtrace): change help message
+
+Tue Nov 20 11:05:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/tracer.rb (Tracer.trace_func): printf to stdout
+	  Patch by Michal Fojtik [ruby-core:45219] [Bug #6490]
+
+Mon Nov 19 21:24:18 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* vm_dump.c: not to include probes.h because the code does not depend
+	  on it.
+	* common.mk (vm_dump.$(OBJEXT)): remove dependency on probes.h
+
+Tue Nov 20 10:14:22 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_s_check_interrupt): removed redundant
+	  GET_THREAD().
+
+Tue Nov 20 10:12:46 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_threadptr_async_errinfo_active_p): added a small
+	  comment.
+
+Tue Nov 20 10:05:56 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_blocking_region_end): replaced GET_THREAD()
+	  with ruby_thread_from_native(). We don't have GVL here.
+
+Tue Nov 20 09:56:15 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_threadptr_execute_interrupts) removed.
+	* thread.c (rb_threadptr_execute_interrupts_common) renamed to
+	  rb_threadptr_execute_interrupts. I.e. unified
+	  rb_threadptr_execute_interrupts and rb_threadptr_execute_interrupts_common.
+	* thread.c (rb_thread_schedule, rb_thread_execute_interrupts) s/_common//.
+
+Tue Nov 20 09:48:34 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (rb_get_next_signal): removed pointless signal
+	  disabling. pthread_sigmask() only changes current thread
+	  mask.
+
+Tue Nov 20 09:36:55 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (install_sighandler): added comments why we need
+	  rb_disable_interrupt().
+
+Tue Nov 20 09:31:33 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (rb_disable_interrupt, rb_enable_interrupt): removed
+	  USE_TRAP_MASK.
+	* signal.c (trap_arg, trap_ensure): removed.
+	* signal.c (trap, sig_trap): removed pointless signal disabling.
+	  We don't need it because we no longer run trap handler on signal
+	  handler context.
+
+Tue Nov 20 09:20:49 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* prelude.rb: Moved Mutex#synchronize to
+	* thread.c (rb_mutex_synchronize_m): here. [Bug #4266]
+
+Tue Nov 20 08:36:15 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (sig_signame): implements Signal.signame method
+	[Feature #5613]
+	* test/ruby/test_signal.rb (test_signame): adds test for above
+	* NEWS: add an item about above
+
+Mon Nov 19 16:30:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (rb_struct_each_pair): yield associated pairs so that
+	  an unsplat argument can get both, for consistency with Hash,
+	  OpenStruct, and etc.  [ruby-dev:46533] [Bug #7382]
+
+Mon Nov 19 16:17:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBS): libelf is need on only FreeBSD.
+
+Mon Nov 19 16:08:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RUBYLIB): purelib option in runruby.rb is deprecated
+	  since r28841, so set to an empty list to get rid of a bogus path in
+	  child processes, which caused an insecure operation exception in
+	  test/ruby/test_encoding.rb:test_unsafe.
+
+Mon Nov 19 15:40:50 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (script): add OPTS=-v, requested by @_ko1.
+
+Mon Nov 19 15:09:07 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: fix didn't enable_dtrace=yes on auto.
+
+	* configure.in: see enable_dtrace for adding libelf on FreeBSD.
+
+	* common.mk: VPATH is not needed.
+
+Mon Nov 19 14:55:51 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: add `Thread#backtrace_locations' method.
+	  This method is similar to `caller_locations' method for
+	  specific method.
+	  And fix to accept `level' and `n' parameters for `Thread#backtrace'
+	  and `Thread#backtrace_locations'.
+	  `caller' (and `caller_locations') do not return `caller' method
+	  frame.
+	  However, `Thread#backtrace' (and `Thread#backtrace_locations')
+	  return `Thread#backtrace' method frame itself
+	  if `Thread.current.backtrace' was called.
+
+	* vm_backtrace.c: ditto.
+
+	* internal.h: ditto.
+
+	* test/ruby/test_backtrace.rb: add tests.
+
+Mon Nov 19 14:54:32 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* Makefile.in, common.mk (probes.h): moved to common.mk and changed to
+	  see $(srcdir).
+
+	* common.mk (probes.dmyh): now created it on $(srcdir) always.
+
+	* win32/Makefile.sub (.SUFFIXES): removed common suffix.
+
+Mon Nov 19 10:00:10 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (.SUFFIX): bsdmake needs .SUFFIX is defined before use.
+
+	* common.mk: fix path of probes.dmyh.
+
+	* common.mk (vm_dump.o): depend probes.h.
+
+	* configure.in: FreeBSD's USDT requires libelf.
+
+Mon Nov 19 01:11:59 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* vm_core.h, probes_helper.h (RUBY_DTRACE_FUNC_ENTRY_HOOK,
+	  RUBY_DTRACE_FUNC_RETURN_HOOK): move from vm_core.h to new file
+	  probes_helper.h for narrowing dependency to probes.h.
+	* common.mk (VM_CORE_H_INCLUDES): remove dependency to probes.h.
+	* common.mk (vm.$(OBJEXT)): add dependency to probes_helper.h.
+	* vm.c, vm_insnhelper.c: include probes_helper.h.
+
+Sun Nov 18 16:33:00 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in, Makefile.in, common.mk: support DTrace on Solaris 10,
+	  based on r26235 by Yugui. On Solaris 10, low optimization level
+	  may also be needed, e.g. optflags="-xO1" or "-xO0" with Oracle
+	  SolarisStudio 12.3 cc.
+
+	* configure.in (--enable-dtrace): new option to enable/disable
+	  DTrace support. By default, trying to enable if dtrace command
+	  is found on the system. It is disabled when cross compiling.
+
+	* configure.in (RUBY_DTRACE_POSTPROCESS): new macro. checks whether
+	  the dtrace on the system needs postprocessing with "dtrace -G".
+	  The postprocessing is needed on Solaris 10 and other platforms.
+
+	* configure.in (RUBY_DTRACE_BSD_BROKEN): new macro. checks whether
+	  the dtrace supports USDT.
+
+	* configure.in (DTRACE): move after RUBY_DTRACE_POSTPROCESS.
+
+	* configure.in (LD): On Solaris, /usr/ccs/bin/ld is preferred.
+
+	* configure.in, Makefile.in, common.mk (DTRACE_OBJ): new macro for
+	  DTrace probe object generated by postprocessing with "dtrace -G".
+
+	* Makefile.in, common.mk (probes.$(OBJEXT)): DTrace probe object
+	  generated by the postprocessing. New file probes.stamp is for
+	  rebuilding related objects that may be modified by "dtrace -G".
+
+	* configure.in, Makefile.in, common.mk (DTRACE_GLOMMED_OBJ): new
+	  macro for DTrace static library hacks.
+
+	* configure.in, Makefile.in (LIBRUBY_A_OBJS): ditto.
+
+	* Makefile.in, common.mk (ruby-glommed.$(OBJEXT)): new target with
+	  rule for DTrace static library hacks.
+
+	* common.mk (DTRACE_DEPENDENT_OBJS): objects depended on probes.h.
+
+Sun Nov 18 09:31:47 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (read_comp): mathn compliant.
+	* rational.c (read_num): ditto.
+
+Sun Nov 18 02:50:12 2012  Luis Lavena <luislavena@gmail.com>
+
+	* win32/file.c (replace_to_long_name): correct logic around wildcard
+	  characters detection and ensure wide-chars are used as pattern.
+	  [ruby-core:49451] [Bug #7374]
+
+Sun Nov 18 02:02:46 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (read_comp): modified handling of polar form.
+
+Sun Nov 18 00:50:44 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (read_comp): fixed handling of polar form.
+
+Sun Nov 18 00:14:46 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (string_to_c_strict, string_to_c): rewrote without regexp.
+	* rational.c (string_to_r_strict, string_to_r): ditto.
+
+Sat Nov 17 23:53:05 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (make_patterns): should not accept extra sign.
+
+Sat Nov 17 21:45:12  Luis Lavena  <luislavena@gmail.com>
+
+	* win32/file.c (replace_to_long_name): skip expansion for all wildcard
+	  characters.
+	  [ruby-core:49451] [Bug #7374]
+
+	* test/ruby/test_file_exhaustive.rb: add more assertions to test.
+
+Sat Nov 17 12:14:50 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: use literal YAML style
+	  when emitting multi-line strings. Thanks @atambo
+
+	* test/psych/test_yaml.rb: test for change.
+
+Sat Nov 17 12:03:41 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: avoid raising exceptions when
+	  parsing Floats and Integers. Thanks riffraff [ruby-core:44426]
+	* test/psych/test_numeric.rb: associated test
+
+Sat Nov 17 11:34:31 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* st.c (st_update): pass the key in st_table so that we can free
+	  memory of the key in st_table when deleting.
+	  [ruby-core:49220] [Bug #7330]
+
+	* test/-ext-/st/test_update.rb
+	  (Bug::StTable#test_pass_objects_in_st_table): add a test.
+
+Sat Nov 17 11:26:36 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/core_ext.rb: move Kernel#y so that it can
+	  manually be required as 'psych/y'.
+
+	* ext/psych/lib/psych/y.rb: ditto
+
+Sat Nov 17 08:13:48 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/abbrev.rb: fix r37113. Correct examples, fix style
+	  and show explicit dependency (require 'abbrev').
+
+Sat Nov 17 07:35:15 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* win32/file.c (replace_to_long_name): skip automatic path expansion
+	  when wildcard character is used.  [ruby-core:49451] [Bug #7374]
+
+	* test/ruby/test_file_exhaustive.rb: add a test for above.
+
+Sat Nov 17 00:50:23 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* marshal.c (w_object): add flonum to arg->data to keep reference index
+	  consistency. [ruby-core:49323] [Bug #7348]
+
+	* test/ruby/test_marshal.rb: add a test for above.
+
+Sat Nov 17 00:40:25 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (incs): dist files need probes.dmyh.
+
+	* common.mk (probes.dmyh): depends on generator script.
+
+	* Makefile.in, common.mk, configure.in, win32/Makefile.sub (probes.h):
+	  select generating with dtrace or copying dummy file by suffix rules.
+
+Fri Nov 16 19:24:10 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_call_without_gvl2): change the parameter of
+	  `func' from `int *skip_interrupt' to `VALUE *flags'.
+	  If (flags & RUBY_CALL_WO_GVL_FLAG_SKIP_CHECK_INTS) is not zero,
+	  then skip checking interrupt.
+	  [ruby-core:46547]
+
+	* include/ruby/thread.h: ditto.
+
+Fri Nov 16 18:59:05 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (no-dtrace-probes.h): dmyprobes.h is in srcdir.
+
+	* common.mk (dmyprobes.h): ditto.
+
+Fri Nov 16 17:57:15 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (probes.h): split build commands for dtrace-available
+	  and unavailable platforms.
+
+	* Makefile.in (incs): probes.h is a platform dependent file, so it
+	  cannot be a part of prereq target.  move it to all-incs.
+
+	* configure.in (DTRACE): move after AC_PROG_CC since cross_compiling
+	  is set in it.
+
+	* configure.in (DTRACE): ignore non-prefixed version if
+	  cross-compiling.
+
+	* Makefile.in, win32/Makefile.sub (probes.h): fix copying dmyprobes.h
+	  path when outplace-build.
+
+Fri Nov 16 15:27:36 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* lib/net/pop.rb (POP3.certs): fix typo in comment.
+	  patch from no6v (Nobuhiro IMAI) <nov@yo.rim.or.jp>.
+	  [ruby-dev:46519] [Bug #7355]
+
+Fri Nov 16 12:36:47 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rake/helper.rb (Rake::TestCase#setup): revert r37669.
+	  @orig_PWD should be the original pwd.
+
+	* test/rake/test_*.rb: don't use @orig_PWD to load libraries.
+	  It should be specified with relative path from the file.
+
+Fri Nov 16 10:22:52 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rake/helper.rb (Rake::TestCase#setup): @orig_PWD shouldn't be
+	  Dir.pwd when the build directory is different from source directory.
+
+Fri Nov 16 09:41:08 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rake/version.rb:  workaround fix to build. see #7366
+	  [ruby-dev:46522]
+
+Fri Nov 16 07:23:18 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake*:  Updated to rake 0.9.4
+	  http://rake.rubyforge.org/doc/release_notes/rake-0_9_4_rdoc.html for
+	  a list of changes in 0.9.4.
+
+	* test/rake*:  ditto
+
+	* NEWS:  ditto
+
+Fri Nov 16 06:58:52 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake*:  Updated to rake 0.9.3.  See
+	  http://rake.rubyforge.org/doc/release_notes/rake-0_9_3_rdoc.html for
+	  a list of changes in 0.9.3.
+
+	* test/rake*:  ditto
+
+	* bin/rake:  ditto
+
+	* NEWS:  ditto
+
+Thu Nov 15 22:39:32 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* range.c (range_bsearch): fix some bugs: a documentation bug, a wrong
+	  condition, missed break in switch/case, and workaround for GCC
+	  optimization. See [ruby-core:49364] in detail.  A great patch from
+	  Heesob Park. [Bug #7352] [Feature #4766]
+
+	* array.c (rb_ary_bsearch): fix similar bug (missed break).
+
+	* test/ruby/test_range.rb: add two test cases for above.
+
+Thu Nov 15 22:41:57 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_exec.h (GENTRY): GENTRY should be pointer size.
+	  A patch by yoshidam (Yoshida Masato) [Bug #7332].
+
+Thu Nov 15 13:20:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* man/erb.1: Fix grammar and copyright dates
+	* man/goruby.1: ditto
+	* man/irb.1: ditto
+	* man/rake.1: ditto
+	* man/ri.1: ditto
+	* man/ruby.1: ditto
+	  Patch by Arthur Gunn
+	  [Fixes Github #210]
+
+Thu Nov 15 11:35:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* man/ruby.1: Grammar edits for man page
+	  Based on a patch by Michael Endsley [Fixes Github #183]
+
+Thu Nov 15 00:47:20 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* array.c (rb_ary_bsearch): add Array#bsearch for binary search.
+	  [ruby-core:36390] [Feature #4766]
+
+	* test/ruby/test_array.rb: add a test for above.
+
+	* range.c (range_bsearch): add Range#bsearch for binary search.
+	  [ruby-core:36390] [Feature #4766]
+
+	* test/ruby/test_range.rb: add a test for above
+
+	* NEWS: added the two new methods.
+
+Wed Nov 14 13:25:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/fileutils.rb (chmod): Add "X" to modes, convert format to table
+	  [ruby-core:48965] [Bug #7288]
+
+Wed Nov 14 11:51:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/csv.rb (init_comments): Document private method #init_comments.
+	  Based on a patch from Vincent Batts [ruby-core:49172] [Bug #7319]
+
+Wed Nov 14 00:54:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* Makefile.in (probes.h): create from probes.d
+
+Tue Nov 13 18:44:01 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_cfunc_with_frame): don't use ci after
+	  EXEC_EVENT_HOOK because ci can be overridden.
+
+	* vm_eval.c: ditto.
+
+	* method.h: change invoker's parameters types.
+
+	* vm_method.c (call_cfunc_invoker_func): ditto.
+
+Tue Nov 13 18:01:54 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_using): raise an ArgumentError if cyclic using is
+	  detected.  based on the patch by Charlie Somerville.
+	  [ruby-core:49092] Bug #7308
+
+	* test/ruby/test_refinement.rb: related test.
+
+Tue Nov 13 17:40:04 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (vm_insnhelper.c): this target is useless and causes
+	  ruby always need rebuild.
+
+Tue Nov 13 17:35:49 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (insn_data_to_s_detail): remove debug lines.
+
+Tue Nov 13 17:28:47 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_caller_setup_args): save and restore
+	  ci->argc and ci->blockptr before and after method invocations
+	  because these method dispatches can override call_info.
+
+	* bootstraptest/test_method.rb: add tests for this fix.
+
+Tue Nov 13 16:38:02 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (dmyprobes.h): always create for make dist.
+
+	* Makefile.in (probes.h): create or copy dmyprobes.h
+
+	* win32/Makefile.sub: only do copy dmyprobes.h.
+
+Tue Nov 13 15:37:21 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (.SUFFIX): .SUFFIX is needed here for .d.h on bsd make.
+
+Tue Nov 13 15:34:35 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk Makefile.in win32/Makefile.sub (.d.h): it's not common.
+
+Tue Nov 13 12:27:11 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: disable dtrace because it doesn't work on FreeBSD.
+
+	* common.mk (clean-local): rm probes.h.
+
+	* common.mk (parse.o): depend $(PROBES_H_INCLUDES).
+
+	* common.mk (.d.h): moved from Makefile.in and use BASERUBY.
+
+	* tool/gen_dummy_probes.rb: reimplemented with ruby because sed is not
+	  available on Windows Microsoft VC++ environment.
+
+Tue Nov 13 12:30:26 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/README.win32: added mention about build directory.  currently
+	  we can not build ruby in win32 directory.  this problem is reported
+	  by Masahiro Kitajima <katonbo@katontech.com>.
+
+Tue Nov 13 11:03:47 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* re.c (rb_memsearch_ss): performance improvement by using memmem(3) if
+	  possible. [ruby-dev:45530] [Feature #6311]
+
+	* configure.in: check existence of memmem(3) and that it is not broken.
+
+Tue Nov 13 06:50:02 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* probes.d: add DTrace probe declarations. [ruby-core:27448]
+
+	* array.c (empty_ary_alloc, ary_new): added array create DTrace probe.
+
+	* compile.c (rb_insns_name): allowing DTrace probes to access
+	  instruction sequence name.
+
+	* Makefile.in: translate probes.d file to appropriate header file.
+
+	* common.mk: declare dependencies on the DTrace header.
+
+	* configure.in: add a test for existence of DTrace.
+
+	* eval.c (setup_exception): add a probe for when an exception is
+	  raised.
+
+	* gc.c: Add DTrace probes for mark begin and end, and sweep begin and
+	  end.
+
+	* hash.c (empty_hash_alloc): Add a probe for hash allocation.
+
+	* insns.def: Add probes for function entry and return.
+
+	* internal.h: function declaration for compile.c change.
+
+	* load.c (rb_f_load): add probes for `load` entry and exit, require
+	  entry and exit, and wrapping search_required for load path search.
+
+	* object.c (rb_obj_alloc): added a probe for general object creation.
+
+	* parse.y (yycompile0): added a probe around parse and compile phase.
+
+	* string.c (empty_str_alloc, str_new): DTrace probes for string
+	  allocation.
+
+	* test/dtrace/*: tests for DTrace probes.
+
+	* vm.c (vm_invoke_proc): add probes for function return on exception
+	  raise, hash create, and instruction sequence execution.
+
+	* vm_core.h: add probe declarations for function entry and exit.
+
+	* vm_dump.c: add probes header file.
+
+	* vm_eval.c (vm_call0_cfunc, vm_call0_cfunc_with_frame): add probe on
+	  function entry and return.
+
+	* vm_exec.c: expose instruction number to instruction name function.
+
+	* vm_insnhelper.c: add function entry and exit probes for cfunc
+	  methods.
+
+	* vm_insnhelper.h: vm usage information is always collected, so
+	  uncomment the functions.
+
+Mon Nov 12 19:14:50 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in (isinf, isnan): isinf() and isnan() are macros on
+	  DragonFly which cannot be found by AC_REPLACE_FUNCS().  This
+	  workaround enforces the fact that they exist on DragonFly.
+
+Mon Nov 12 15:59:38 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_core.h (rb_call_info_t::refinements), compile.c (new_callinfo),
+	  vm_insnhelper.c (vm_search_method): revert r37616 because it's too
+	  slow.  [ruby-dev:46477]
+
+	* test/ruby/test_refinement.rb (test_inline_method_cache): skip
+	  the test until the bug is fixed efficiently.
+
+Mon Nov 12 14:28:01 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/mkexports.rb (each_export): skip garbage generated by VS2012's
+	  nmake.
+	  reported and patched by Yoshida Masato at [Bug #7333] [ruby-dev:46484]
+
+Sun Nov 11 18:58:55 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/test_date_{parse,strptime}.rb: changed the format of
+	  some extra messages.
+
+Sun Nov 11 18:41:34 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (date__parse): revised the tight parser
+	  (about handling of apostrophes).
+
+Sun Nov 11 15:39:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_s_create): just warn for wrong elements now.
+	  [ruby-dev:46440] [Bug #7300]
+
+	* hash.c (rb_hash_s_create): refine error messages.
+
+	* error.c (rb_builtin_class_name): share for above.
+
+Sun Nov 11 15:12:18 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (top_using): remove Kernel#using, and add main.using instead.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sun Nov 11 13:41:01 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_using_refinement, rb_mod_using, f_using): clear method
+	  cache only when using is called explicitly.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sun Nov 11 12:56:34 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* lib/pstore.rb (PStore): fix not to replace ThreadError raised in
+	  #transaction block with PStore::Error.
+	  [ruby-core:39238] [Bug #5269]
+
+Sun Nov 11 11:36:19 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_core.h (rb_call_info_t::refinements), compile.c (new_callinfo):
+	  add a new field for inline method cache.
+
+	* vm_insnhelper.c (vm_search_method): check rb_call_info_t::refinements
+	  not to confuse inline method cache when module_eval is used with
+	  refinements.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sun Nov 11 08:45:45 2012  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* ruby.c: removed a comma before "before"
+
+Sat Nov 10 23:02:31 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: move immutable fields from struct heaps_slot and struct
+	  sorted_heaps_slot into struct heaps_header.
+	  Based on a patch from Sokolov Yura [Feature #6199][ruby-core:43592]
+
+Sat Nov 10 19:28:16 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: modified doc.
+	* rational.c: ditto.
+
+Sat Nov 10 18:20:10 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c: edited about era.
+
+Sat Nov 10 12:13:41 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* tool/rbinstall.rb: Don't install *.gemspec under lib/.
+	  [ruby-core:48966] [Bug #7289]
+	  Reported by Vit Ondruch. Thanks!!!
+
+Sat Nov 10 00:49:26 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ruby_atomic.h: renamed from atomic.h to avoid header file name
+	  conflict on Solaris 10. [ruby-dev:46414] [Bug #7287]
+
+	* gc.c, signal.c, vm_core.h, common.mk: reflect the rename from
+	  atomic.h to ruby_atomic.h.
+
+Sat Nov 10 00:46:57 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* atomic.h: Revert r37491 which is a temporary workaround.
+
+Sat Nov 10 00:33:31 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* siphash.h: check configure macros before include newer headers.
+
+Fri Nov  9 23:33:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/erb.rb (ERB#run, ERB#result): eval under isolated bindings for
+	  safe concurrent use.  [ruby-core:47638] [Bug #7046]
+
+Fri Nov  9 23:05:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (BYTE_ORDER): define using configured WORDS_BIGENDIAN.
+
+	* siphash.c (sip_init_state): use union to suppress warnings by gcc
+	  4.7.
+
+Fri Nov  9 19:40:03 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_splice): fix r37583 doesn't consider the case when
+	  beg > array length.
+
+Fri Nov  9 16:11:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_memhash): use siphash.
+
+Fri Nov  9 16:08:46 2012  Sokolov Yura funny-falcon  <funny.falcon@gmail.com>
+
+	* array.c: speedup Array#unshift by using space in shared array.
+	  [Feature #6638]
+	  - when array owns its shared array (ARY_SHARED_NUM == 1), and there
+	    is enough space then try unshift values directly into shared
+	    array.
+	  - when resulting array is big (~>64 items) then make it shared with
+	    enough room for future #unshifts, and then insert into shared
+	    array.
+
+	* array.c (rb_ary_splice): use shared array in rb_ary_slice.
+	  [Feature #6638]
+	  - use ary_ensure_room_for_push when rb_ary_slice used to add at the
+	    end of array, cause rb_ary_concat use rb_ary_slice.
+
+	* array.c (ary_ensure_room_for_push): make array really suitable for
+	  queue.  [Feature #6638]
+	  when array is shared (which happens after Array#shift), and
+	  ARY_SHARED_NUM == 1 (which is very often when array used as queue),
+	  then make rb_ary_push push directly into shared array.
+
+	* array.c (rb_ary_modify): steal shared array's container when
+	  ARY_SHARED_NUM == 1.  [Feature #6638]
+	  - Do not allocate new memory in rb_ary_modify when ARY_SHARED_NUM == 1
+	    and length almost same.
+	  - Store ARY_CAPA instead of RARRAY_LEN in ary_make_shared, to make
+	    it useful.
+	  - Fix rb_ary_sort_bang accordantly.
+
+Fri Nov  9 16:00:00 2012  Zachary Scott  <zzak@zacharyscott.net>
+
+	* ext/bigdecimal/bigdecimal.c: Documentation for BigDecimal
+	  Based on a patch from Vincent Batts [ruby-core:49047] [Bug #7305]
+
+Fri Nov  9 15:25:42 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb (Shellwords#shellescape): Add back my original
+	  real world example with some enhancement.
+
+	* lib/shellwords.rb (Shellwords#shelljoin): Undo part of the
+	  previous rdoc change.  This new example using a string-only
+	  array was not in line with the description.
+
+Fri Nov  9 12:58:13 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_crypt): crypt(3) may return NULL.
+	  Latest glibc (2.16?) crypt(3) actually returns NULL. [Bug #7312]
+
+Fri Nov  9 12:07:06 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* include/ruby/ruby.h (alloca), eval_intern.h (alloca), gc.c
+	  (alloca): Make alloca() globally available by moving the
+	  ultimate ifdef's to ruby/ruby.h.  Gcc hides its builtin alloca()
+	  when compiling with -ansi, and linking thus fails on platforms
+	  that lack their own alloca() implementation in libc, which
+	  include OpenBSD and some ports of NetBSD.  We use alloca()
+	  everywhere including from within third party C extensions, so
+	  alloca() must be made globally available. [Bug #7307]
+
+	* addr2line.c (alloca): Replace the alloca() part with the
+	  ultimate ifdef's. [Bug #7307]
+
+Fri Nov  9 09:30:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* io.c (IO#new):
+	  Fix indentation from r37444
+	  [ruby-core:48052] [Bug #7179]
+
+Fri Nov  9 07:36:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (bigmul0): enable big_mul_toom3.
+	  [ruby-core:48552] [Bug #7242]
+
+	* bignum.c (bigmul1_toom3): fix incorrect calculation.
+	  the patch is made by Heesob Park.
+	  [ruby-core:48552] [Bug #7242]
+
+Fri Nov  9 05:33:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (bigmul0): disable big_mul_toom3 temporarily.
+	  [ruby-core:48552] [Bug #7242]
+
+	* test/ruby/test_bignum.rb (test_mul_large_numbers):
+	  add a test for bigmul1_toom3 suggested in [Bug #7242].
+
+Thu Nov  8 21:57:59 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* re.c (rb_memsearch): performance improvement by using memchr().
+	  [ruby-dev:45397] [Feature #6173]
+
+Thu Nov  8 19:02:50 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/protocol.rb (Net::InternetMessageIO#each_crlf_line):
+	  treat \r as newline as mame pointed. [ruby-dev:46425] [Bug #7278]
+
+Thu Nov  8 11:32:11 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in (--with-opt-dir): Avoid nesting of double quotes
+	  inside backquotes, since some traditional shells like PD KSH
+	  (which OpenBSD's /bin/sh bases on) fail to parse them.
+
+Thu Nov  8 09:34:00 2012  Kenta Murata  <mrkn@cookpad.com>
+
+	* numeric.c: Add a caution that the results of the comparing
+	  operations of two NaNs are undefined.
+	  [#1720] [ruby-dev:38725] [ruby-core:36966]
+
+Thu Nov  8 04:45:21 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* include/ruby/intern.h: Restore rb_enumeratorize as it was before
+	  r37497 and introduce rb_enumeratorize_with_size instead. [#7302]
+
+	* enumerator.c: ditto.
+
+Wed Nov  7 15:22:37 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (ruby_float_step): fix r37514: it yielded with NaN
+	  if the unit is infinity.
+
+Wed Nov  7 15:46:12 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/webrick.rb: fix typo. reported by Rohit Arondekar.
+	  https://github.com/ruby/ruby/pull/211
+
+Wed Nov  7 15:34:12 2012  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb: alias CGI#http_header to CGI#header.
+
+Wed Nov  7 12:49:39 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refine): set RMODULE_IS_REFINEMENT to a created
+	  refinement module, and don't override method_added.
+
+	* vm_method.c (rb_method_entry_make): check redefinition of
+	  optimized methods when a method is added to a refinement module.
+	  [ruby-core:48970] [Bug #7290]
+
+	* test/ruby/test_refinement.rb: related test.
+
+Wed Nov  7 11:48:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-additional.el (ruby-mode-set-encoding): now encoding needs
+	  to be set always explicitly actually.  [Feature #6679]
+
+Wed Nov  7 09:15:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_get): avoid inadvertent symbol creation.
+
+Wed Nov  7 07:52:50 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (rb_enum_cycle_size): prefix with rb.
+
+Wed Nov  7 04:32:15 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* test/ruby/test_file_exhaustive.rb: Remove FIXME skip on Windows
+	  specific test because the test in question was already fixed.
+
+Wed Nov  7 03:45:12 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* ext/zlib/extconf.rb: Recognize zlibwapi as linking library.
+	  Patch by Daniel Berger.
+
+	  [ruby-core:44979] [Feature #6421]
+
+Wed Nov  7 02:06:40 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: New method #size; constructor accepts size.
+	                Have #to_enum accept a block
+	                Warn when using deprecated form of constructor
+	                Support #size for enumerators created from enumerators
+	                Support for lazy.{map|flat_map|...}.size.
+
+	* include/ruby/intern.h: RETURN_SIZED_ENUMERATOR for support of
+	  sized enumerators.
+
+	* array.c: Support for various enumerator.size.
+
+	* enum.c: ditto.
+
+	* hash.c: ditto.
+
+	* numeric.c: ditto.
+
+	* range.c: ditto.
+
+	* string.c: ditto.
+
+	* struct.c: ditto.
+
+	* vm_eval.c: ditto.
+
+Tue Nov  6 20:40:28 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* object.c (rb_mod_const_get): Fix constant missing exception class
+	  and message to maintain backwards compatibility. Constant search
+	  should start at Object when constant starts with '::'
+
+	* test/ruby/test_module.rb: test for fixes
+
+Tue Nov  6 16:50:00 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* lib/tempfile.rb (Tempfile#inspect): fix confusing #inspect.
+	  previous Tempfile#inspect says it is a File, but actually
+	  it is not a File.
+
+	    t = Tempfile.new("foo") #=> #<File:/tmp/foo20121106-31970-1ffbum0>
+	    t.is_a? File #=> false
+
+	  now Tempfile#inspect returns like:
+
+	    t = Tempfile.new("foo")
+	    #=> #<Tempfile:/tmp/foo20121106-31970-1ffbum0>
+
+	  [ruby-core:47544] [Bug #7027]
+
+Tue Nov  6 16:22:30 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* atomic.h: add #include <sys/atomic.h> for the workaround of
+	  header file name conflict of atomic.h with /usr/include/atomic.h
+	  on Solaris 10. [ruby-dev:46414] [Bug #7287]
+
+Tue Nov  6 14:38:00 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/win32ole/test_win32ole.rb: now source encoding is UTF-8, so
+	  binary strings in old scripts are dangerous.
+
+Tue Nov  6 14:25:09 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/protocol.rb (Net::InternetMessageIO#each_crlf_line):
+	  don't use /n in universal regexp. [ruby-dev:46394] [Bug #7278]
+
+Tue Nov  6 09:42:26 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_b): Add String#b, returning a copied string
+	  whose encoding is ASCII-8BIT. [ruby-dev:45992] [Feature #6767]
+
+Tue Nov  6 09:37:57 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (load_file_internal): set default source encoding as
+	  UTF-8 instead of US-ASCII. [ruby-core:46021] [Feature #6679]
+
+	* parse.y (parser_initialize): set default parser encoding as
+	  UTF-8 instead of US-ASCII.
+
+Tue Nov  6 05:48:06 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_require.rb
+	  (TestRequire#test_require_to_path_redefined_in_load_path,
+	   TestRequire#test_require_to_str_redefined_in_load_path):
+	  Suppress method redefined warning when test-all with RUBYOPT=-w.
+
+Thu Nov  8 00:24:14 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/curses/view.rb: Do not fail if the file to view is shorter
+	  than the screen height.
+
+Mon Nov  5 11:40:00 2012  Mark Somerville  <mark@scottishclmibs.com>
+
+	* thread_pthread.c (rb_reserved_fd_p): fix typo in macro check
+	  that prevented the ifdef ever being true.
+	  [Bug #7281] [ruby-core:48940]
+
+Mon Nov  5 23:28:57 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* file.c (rb_get_path_check_to_string): extract from
+	  rb_get_path_check(). We change the spec not to call to_path of
+	  String object.
+
+	* file.c (rb_get_path_check_convert): extract from rb_get_path_check().
+
+	* file.c (rb_get_path_check): follow the above change.
+
+	* file.c (rb_file_expand_path_fast): remove check_expand_path_args().
+	  Instead we call it in load.c.
+
+	* file.c (rb_find_file_ext_safe): use rb_get_expanded_load_path() to
+	  reduce expand cost.
+
+	* file.c (rb_find_file_safe): ditto.
+
+	* internal.h (rb_get_expanded_load_path): add a declaration.
+
+	* internal.h (rb_get_path_check_to_string, rb_get_path_check_convert):
+	  add declarations.
+
+	* load.c (rb_construct_expanded_load_path): fix for compatibility.
+	  Same checks in rb_get_path_check() are added. We don't replace
+	  $LOAD_PATH and ensure that String object of $LOAD_PATH are frozen.
+	  We don't freeze non String object and expand it every time. We add
+	  arguments for expanding load path partially and checking if load path
+	  have relative paths or non String objects.
+
+	* load.c (load_path_getcwd): get current working directory for checking
+	  if it's changed when getting load path.
+
+	* load.c (rb_get_expanded_load_path): fix for rebuilding cache properly.
+	  We check if current working directory is changed and rebuild expanded
+	  load path cache. We expand paths which start with ~ (User HOME) and
+	  non String objects every time for compatibility. We make this
+	  accessible from other source files.
+
+	* load.c (rb_feature_provided): call rb_get_path() since we changed
+	  rb_file_expand_path_fast() not to call it.
+
+	* load.c (Init_load): initialize vm->load_path_check_cache.
+
+	* vm.c (rb_vm_mark): mark vm->load_path_check_cache for GC.
+
+	* vm_core.h (rb_vm_struct): add vm->load_path_check_cache to store data
+	  to check load path cache validity.
+
+	* test/ruby/test_require.rb (TestRequire): add tests for require
+	  compatibility related to cached expanded load path.
+	  [ruby-core:47970] [Bug #7158]
+
+Mon Nov  5 23:26:05 2012  Greg Price  <price@mit.edu>
+
+	* load.c (rb_get_expanded_load_path): cache the expanded load
+	  path.  This saves 4KB of allocation and some stats for every
+	  element of the load path (so nearly a MB in my Rails app)
+	  on every require.
+
+	* load.c (rb_construct_expanded_load_path): ensure that $LOAD_PATH
+	  entries are frozen strings.  The user must mutate $LOAD_PATH
+	  itself rather than its individual entries.
+
+	* vm_core.h (rb_vm_struct): add fields.
+
+	* vm.c (rb_vm_mark): mark new fields.
+
+	* ruby.c (process_options): modify $LOAD_PATH directly rather than
+	  its elements.
+	  Patch by Greg Price.
+	  [ruby-core:47970] [Bug #7158]
+
+Mon Nov  5 23:24:42 2012  Greg Price  <price@mit.edu>
+
+	* load.c (rb_feature_p, rb_provide_feature): index $LOADED_FEATURES
+	  so that require isn't so slow.
+
+	* load.c (rb_provide_feature, get_loaded_features_index): ensure
+	  that $LOADED_FEATURES entries are frozen strings.  The user
+	  must mutate $LOADED_FEATURES itself rather than its individual
+	  entries.
+
+	* load.c (reset_loaded_features_snapshot): add a new function to reset
+	  vm->loaded_features_snapshot.
+
+	* load.c (get_loaded_features_index_raw): add a new function to get
+	  the loaded-features index.
+
+	* load.c (features_index_add_single): add a new function to add to the
+	  loaded-features index a single feature.
+
+	* load.c (features_index_add): add a new function to add to the
+	  loaded-features index all the required entries for `feature`.
+
+	* vm_core.h (rb_vm_struct): add fields.
+
+	* vm.c (rb_vm_mark): mark new fields.
+
+	* include/ruby/intern.h (rb_hash_clear): declare function.
+
+	* hash.c (rb_hash_clear): make function non-static.
+	  Patch by Greg Price.
+	  [ruby-core:47970] [Bug #7158]
+
+Mon Nov  5 23:23:51 2012  Greg Price  <price@mit.edu>
+
+	* array.c (rb_ary_shared_with_p): new function.
+	  Expose whether two arrays are shared (read-only, C only).
+
+	* include/ruby/intern.h (rb_ary_shared_with_p): declare.
+	  Patch by Greg Price.
+	  [ruby-core:47970] [Bug #7158]
+
+Mon Nov  5 23:21:14 2012  Greg Price  <price@mit.edu>
+
+	* load.c (loaded_feature_path): clarify and briefly comment
+	  function.  These clarifications have no effect on the behavior
+	  of the function.
+
+	* load.c (rb_feature_p): explain the search loop.  Especially
+	  useful because the logic is complicated as described in the
+	  second paragraph.
+	  Patch by Greg Price.
+	  [ruby-core:47970] [Bug #7158]
+
+Mon Nov  5 22:45:03 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* ext/dl/win32/lib/Win32API.rb (Win32API#call): use 64bit pointer for x64
+	  Windows. This would fix
+	  TestSecureRandom#test_s_random_bytes_without_openssl error.
+	  [ruby-core:47451] [Bug #6990]
+
+Mon Nov  5 22:09:26 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* cygwin/GNUmakefile.in (uncommon.mk): link *.res.o.
+	  EXTOBJES is defined in uncommon.mk. *.res.o setting should be below
+	  uncommon.mk.
+	  [ruby-core:48858] [Bug #7277]
+
+Mon Nov  5 11:35:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (native_thread_init, native_thread_destroy):
+	  removed HAVE_PTHREAD_CONDATTR_INIT check because this silly
+	  #ifdef makes use-uninitialized-var issue and (2) native_cond_initialize()
+	  already have a right platform and caller don't need any additional care.
+	  [Bug #6825]
+
+Mon Nov  5 10:57:59 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi/core.rb: check if Tempfile is defined before use it.
+
+	* lib/cgi/core.rb: remove tempfiles only if tempfiles exist
+
+Mon Nov  5 12:17:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/uri/http.rb (URI::HTTP.build): Fix example
+	  Patch by Carina C. Zona
+	  [Fixes #209 Github]
+
+Mon Nov  5 09:55:05 2012  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb: remove tempfile more early.
+
+Sun Nov  4 20:29:46 2012  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi.rb, lib/cgi/*/rb: rename CGI#header to CGI#http_header,
+	  add and update HTML5 tag generator.  [Bug #7110]
+	  Patch provided by Marcus Stollsteimer, thank you !
+
+Sun Nov  4 11:47:39 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* lib/fileutils.rb (module FileUtils): repatch [ruby-core:39622]
+	  [Feature #5337]. improve performance of FileUtils.compare_stream.
+	  [ruby-core:47545] [Feature #7028]
+
+Sun Nov  4 11:27:54 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (recursive_equal): fix to return true when self and other
+	  are resized to same size and the current index become out of
+	  range.
+
+	* test/ruby/test_array.rb: add a test for the above.
+
+Sun Nov  4 10:19:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (file_s_fnmatch): match with expanding braces if FNM_EXTGLOB
+	  is set.  [ruby-core:40037] [Feature #5422]
+
+Sat Nov  3 23:38:15 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: modified doc.
+	* rational.c: ditto.
+
+Sat Nov  3 22:38:55 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: modified doc.
+
+Sat Nov  3 18:35:55 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (rb_vm_rewrite_ep_in_errinfo, vm_rewrite_ep_in_errinfo):
+	  merge code and remove `rb_vm_rewrite_ep_in_errinfo'.
+
+Sat Nov  3 18:15:24 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c, proc.c: avoid unnecessary `rb_vm_rewrite_ep_in_errinfo'
+	  calls.
+
+Sat Nov  3 17:53:43 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* bin/testrb: Use only Test::Unit::AutoRunner in test-unit gem
+	  compatible API to be available by both test/unit bundled in Ruby
+	  and test-unit gem.
+	* lib/test/unit.rb (Test::Unit::AutoRunner): Move codes from testrb.
+
+Sat Nov  3 14:56:21 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (parse_eu): should capture apostrophe too.
+
+Sat Nov  3 14:46:15 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (date__parse): revised the tight parser.
+
+Sat Nov  3 14:43:42 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/xmldecl.rb (REXML::XMLDecl#content): Add missing \A
+	  and \z.
+
+Sat Nov  3 14:42:55 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/output.rb (REXML::Output#initialize): Use normalized
+	  encoding name.
+
+Sat Nov  3 14:41:17 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/output.rb (REXML::Output): Don't output BOM in middle
+	  of the output string.
+	* test/rexml/test_document.rb: Add a test for the above change.
+
+Sat Nov  3 14:09:55 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS: Add an item about REXML::Document#write.
+
+Sat Nov  3 13:46:49 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb: Fix tests that expect encoding name
+	  isn't normalized.
+
+Sat Nov  3 13:26:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* dir.c (Dir#glob):
+	  Documentation for pattern section, backslash subsection
+	  Patch by Eric Bouchut
+	  [ruby-core:48528] [Bug #7230]
+
+Sat Nov  3 13:26:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* io.c (IO#new):
+	  Documentation for IO#open modes and formatting
+	  [ruby-core:48052] [Bug #7179]
+
+Sat Nov  3 13:01:02 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_encoding.rb (EncodingTester#test_in_different_out):
+	  Fix a test that expects encoding in XML declaration is changed by
+	  Output's encoding. It is dropped feature.
+
+Sat Nov  3 12:49:45 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb (REXML::Document#write): Document encoding
+	  option. Now different encoding between XML file's encoding and
+	  XML declaration's encoding is support.
+	  [Feature #4872] (work in progress)
+	* lib/rexml/xmldecl.rb (REXML::XMLDecl#write): Always use XMLDecl's
+	  encoding.
+	* test/rexml/test_document.rb: Update tests for the above change.
+
+Sat Nov  3 12:18:35 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (recursive_equal): fix not to make invalid pointers when
+	  self and other are resized to same size in #== of their elements.
+	  [ruby-dev:46373] [Feature #6177]
+
+Sat Nov  3 12:06:15 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_xml_declaration.rb (TestXmlDeclaration#test_*):
+	  Remove needless prefix from test names.
+
+Sat Nov  3 12:04:52 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_xml_declaration_parent_child.rb: Rename to ...
+	* test/rexml/test_xml_declaration.rb: ... this.
+
+Sat Nov  3 11:43:00 2012  Zachary Scott  <zachary@zacharyscott.net>
+
+	* hash.c (rb_hash_delete): Correct grammar
+	  Patch by Glenn Oppegard
+	  [Fixes #208 Github]
+
+Sat Nov  3 11:28:28 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* NEWS: add a news about GC::Profiler.raw_data.
+
+Sat Nov  3 11:01:32 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* NEWS: add a news about rb_newobj_of() and NEWOBJ_OF().
+
+Sat Nov  3 10:17:41 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* eval.c (f_current_dirname): add the new method for Kernel.
+	  This method almost same as File.dirname(__FILE__). One
+	  different behavior is it returns nil when __FILE__ returns nil.
+	  [Feature #3346]
+
+	* NEWS:  ditto
+
+	* test/ruby/test_method.rb: related test.
+
+Sat Nov  3 09:03:34 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_refinement.rb (test_new_method_by_send,
+	  test_new_method_by_method_object): add tests for Kernel#send and
+	  Kernel#method with refinements.
+
+	* test/ruby/test_refinement.rb (test_symbol_to_proc): add a test
+	  calling a proc created by Symbol#to_proc outside the scope where
+	  a refinement is closed over.
+
+Sat Nov  3 04:14:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_vm_rewrite_ep_in_errinfo): rewrite all catch points in
+	  errinfo, not only the topmost frame.  based on the patch by
+	  ktsj (Kazuki Tsujimoto) in [ruby-dev:45656]. [Bug #6460]
+
+Fri Nov  2 20:11:17 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#timestamp_file): remove @ which looks like
+	  configure variables.
+
+	* lib/mkmf.rb (MakeMakefile#timestamp_file): use .-. instead of !, a
+	  special character of NMAKE and BSD make.  [Bug #7265]
+
+Fri Nov  2 17:55:39 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb (_run_parallel): Delete status line before showing
+	  results. Patch by Hiroshi Shirosaki. [Bug #6897] [ruby-core:47250]
+
+	* lib/test/unit.rb (_run_parallel): Fix strange result when disabled retrying.
+	  Patch by Hiroshi Shirosaki. [Bug #6897] [ruby-core:47250]
+
+Fri Nov  2 17:52:12 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* object.c (rb_mod_to_s): Module#{to_s,inspect}, when invoked on
+	  a refinement, returns a string in the format #<refinement:C@M>,
+	  where C is a refined class and M is a module at which the refinement
+	  is defined.
+
+	* eval.c (rb_mod_refine): store information on a refinement for the
+	  above change.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Fri Nov  2 16:57:52 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* vm_dump.c (rb_vm_bugreport): Because of many log directories,
+	  making directory lists readable.
+
+Fri Nov  2 16:44:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* vm_dump.c (rb_vm_bugreport): add ~/Library/Logs/DiagnosticReports
+	  in the locations list of crash reports.
+
+Fri Nov  2 14:52:52 2012  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (recursive_equal): performance improvement by avoiding
+	  overhead to call rb_ary_elt().
+	  [ruby-dev:45412] [Feature #6177]
+
+Fri Nov  2 14:47:53 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* string.c (sym_to_proc, sym_call): A Proc created by Symbol#to_proc
+	  should close over the current refinements.
+	  [ruby-dev:46345] [Bug #7261]
+
+	* vm_eval.c (rb_call0, rb_search_method_entry,
+	  rb_funcall_passing_block_with_refinements): add a new argument
+	  `refinements' for the above changes.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Fri Nov  2 08:24:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (top_define_method): new method, main.define_method.
+	  [ruby-core:45715] [Feature #6609]
+
+	* eval.c (top_include): fix a warning message, main is not a class or
+	  module.
+
+Fri Nov  2 04:41:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#timestamp_file): use ! instead of %, a GNU
+	  make special character.
+
+Fri Nov  2 04:40:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_process.rb (test_execopts_preserve_env_on_exec_failure):
+	  use never existing file in the current temporary directory.
+
+Fri Nov  2 04:23:20 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: add feature to tag preview/rc.
+
+Fri Nov  2 03:23:37 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/mkmf.rb: fix for if config["libdir"] is nil.
+
+Thu Nov  1 23:06:01 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/make-snapshot: fix wrong regexp for releasing preview.
+	  patched by mame.
+
+Thu Nov  1 22:27:11 2012  Koichi Sasada <ko1@atdot.net>
+
+	* NEWS: add a news about objspace,
+	  ObjectSpace.reachable_objects_from.
+
+Thu Nov  1 21:57:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_new),
+	  test/bigdecimal/test_bigdecimal.rb:
+	  Fix exception message of BigDecimal constructor with a Float.
+
+Thu Nov  1 21:52:20 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_add),
+	  test/bigdecimal/test_bigdecimal.rb:
+	  need to specify precision for converting Rational and Float.
+	  [ruby-core:48045] [Bug #7176]
+
+Thu Nov  1 21:42:20 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_process.rb: Revert r37404.  My ubuntu box has
+	  actually the directory named "/nonexistent"...  Sorry.
+
+Thu Nov  1 21:28:28 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* test/ruby/test_process.rb: Process.exec raised EACCES on Linux
+	  3.5.0-17-generic.  This is a temporal fix to rescue that exception.
+	  Needs kosaki's review.
+
+Thu Nov  1 21:19:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (set_relation): parent_iseq need to be set regardless iseq
+	  type.  fix r37397.
+
+Thu Nov  1 19:47:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (RUBY_STACK_MIN): may not a compile time constant.
+	  fix r37079.  [ruby-dev:46322] [Bug #7247]
+
+Thu Nov  1 16:44:36 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* NEWS: add note for Module#refine, Module#refinements,
+	  Module#using, and Kernel#using.
+
+Thu Nov  1 14:41:47 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_using_module): using should be used indirectly.
+	  [ruby-dev:46326] [Feature #7251]
+
+Wed Oct 31 18:17:38 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_profile_record): don't define unused variables when
+	  GC_PROFILE_MORE_DETAIL is 0.
+
+Wed Oct 31 18:10:53 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_prof_mark_timer_stop): count is not initialized.
+
+Wed Oct 31 09:28:24 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* thread.c (rb_thread_call_without_gvl2):  Note that ubf() may or may
+	  not be called with the GVL.  Hinted that rb_thread_call_with_gvl()
+	  can be used to access ruby functionality.
+
+Wed Oct 31 09:06:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* thread.c (rb_thread_call_without_gvl2):  Update documentation to
+	  natural English.
+	* thread.c (rb_thread_call_with_gvl):  ditto.
+
+Wed Oct 31 02:53:07 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/dl/lib/dl/struct.rb: fix strange require order. [ruby-dev:45702]
+
+	* ext/dl/lib/dl/value.rb: ditto
+
+	* test/dl/test_c_struct_entry.rb: remove strange require order from
+	  tests.
+
+	* test/dl/test_c_union_entity.rb: ditto
+
+Tue Oct 30 23:59:32 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refine): fix the error message when no block is
+	  given.  [ruby-dev:46319] [Bug #7244]
+
+	* test/ruby/test_refinement.rb: related test.
+
+Tue Oct 30 19:27:48 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (redirect_dup2): set standard handles when new fd is stdio,
+	  because if there is no allocated console at the moment Windows does
+	  not automatically associate it for child process's standard handle.
+	  this is adhoc workaround.
+	  reported by Martin Thiede at [ruby-core:48542] [Bug #7239].
+
+	* io.c (rb_cloexec_dup2): ditto.
+
+Tue Oct 30 03:08:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rbconfig/obsolete.rb (Config): re-introduce warnings for a
+	  lame-duck.  [ruby-core:46836] [Bug #6809]
+
+Tue Oct 30 02:20:10 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* thread.c: added Thread#thread_variable_(get|set),
+	  Thread#thread_variable?, and Thread#thread_variables for operating
+	  on variables that are local to threads. [ruby-core:47790]
+
+	* vm.c: ditto
+
+	* test/ruby/test_thread.rb: tests for thread variables.
+
+Mon Oct 29 18:22:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_close): close separately per each
+	  instances, as well as IO.
+
+Mon Oct 29 10:22:00 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/handlers/recorder.rb: added a class for
+	  recording YAML parse and emit events.
+
+	* ext/psych/lib/psych/handler.rb: adding a list of events so that
+	  handler classes can more easily be meta-programmed.
+
+	* test/psych/handlers/test_recorder.rb: tests for the change.
+
+Mon Oct 29 05:48:52 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb: Add [] and []=, base on a patch by Thomas Sawyer.
+	  Also accept {Open}Struct as argument to new.
+	  Add #eql?, #hash & #each_pair
+	  Protect new_ostruct_member
+
+Mon Oct 29 03:20:58 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Fix determinant_e [ruby-dev:46305] [Bug #7228]
+
+Sun Oct 28 23:52:25 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb: Add tests for parsing XML encoded
+	  by UTF-8 with BOM.
+
+Sun Oct 28 23:47:09 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/source.rb: Move encoding detection code to base class.
+	* lib/rexml/encoding.rb: Remove needless encoding detection code.
+
+Sun Oct 28 21:40:13 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb: Fix a bug that UTF-8 is used
+	  for UTF-16XX encoded XML that doesn't have encoding="UTF-16" in
+	  XML declaration.
+	* test/rexml/test_document.rb: Add tests for the above change.
+
+Sun Oct 28 21:37:34 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb: Group tests that they parse
+	  UTF-16XX encoded XML that has encoding="UTF-16" in XML declaration.
+
+Sun Oct 28 21:25:11 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/source.rb (REXML::IOSource#initialize): Reduce
+	  @line_break initialize code. It should be done only in #encoding=.
+	* lib/rexml/parsers/baseparser.rb: Don't set UTF-16 encoding to
+	  source by encoding="UTF-16" in XML declaration because UTF-16XX
+	  source encoding should be set in Source#initialize or
+	  IOSource#initialize. They should handle BOM. Parser should not
+	  consider about it.
+
+Sun Oct 28 21:18:37 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb: Add tests for parsing XML encoded
+	  by UTF-16 with BOM.
+
+Sun Oct 28 19:12:11 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (iso8601_{ext,bas}_time): should not match
+	  empty string.
+
+Sun Oct 28 18:51:33 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (date__parse): revised the tight parser.
+
+Sun Oct 28 15:41:50 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb (REXML::Document#write): Add :encoding option
+	  to support custom XML encoding.
+	  [Feature #4872] (work in progress)
+	* test/rexml/test_document.rb: Add tests for the above change.
+
+Sun Oct 28 15:00:19 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb (REXML::Document#write): Remove needless
+	  indent in document.
+
+Sun Oct 28 14:59:14 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb (REXML::Document#write): Accept options
+	  Hash as argument.
+	* test/rexml/test_document.rb: Add tests for the above change.
+
+Sun Oct 28 14:09:44 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb (REXML::Document#write): Fix wrong usage
+	  in document.
+
+Sun Oct 28 14:03:48 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb (REXML::Document#write): Fix wrong method
+	  names in document.
+
+Sun Oct 28 10:12:15 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: `tree` should return the
+	  same thing on every call.
+
+	* test/psych/visitors/test_yaml_tree.rb: related test.
+
+Sun Oct 28 10:05:03 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: YAML Tree object should
+	  be able to take an emitter object as it's output.
+
+	* test/psych/visitors/test_yaml_tree.rb: related test.
+
+Sun Oct 28 08:23:16 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* bignum.c (bignew_1): Bignum instances are frozen.
+	  Feature #3222
+
+	* include/ruby/ruby.h: Fixnum instances are also frozen.
+
+	* class.c (singleton_class_of): check Bignum before
+	  singleton checking.
+
+	* test/ruby/test_bignum.rb: add a test.
+
+	* test/ruby/test_fixnum.rb: ditto.
+
+	* test/ruby/marshaltestlib.rb, test/ruby/test_eval.rb,
+	  test/ruby/test_object.rb: catch up above changes.
+
+Sun Oct 28 04:38:06 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_define_method): remove type and frozen checking.
+	  Checking is done in `rb_singleton_class()'.
+
+Sun Oct 28 00:49:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assign_in_cond): warn for static content object assignments
+	  in conditional statements.  [ruby-dev:43083] [Feature #4299]
+
+Sat Oct 27 23:33:41 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* gc.c (gc_profile_result, gc_profile_report): use internal structures
+	  to avoid allocations and progressively print the output for #report.
+	  [ruby-core:47163] [Bug #6865]
+
+Sat Oct 27 11:01:10 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* numeric.c (rb_float_new_in_heap), include/ruby/ruby.h:
+	  make all Float objects frozen.
+	  [ruby-dev:46081] [ruby-trunk - Feature #6936]
+	  Most part of patch by NARUSE, Yui  <naruse@ruby-lang.org>.
+
+	* class.c (singleton_class_of): raise TypeError when
+	  trying to define a singleton method on Float objects.
+
+	* vm.c (vm_define_method): ditto.
+
+	* test/ruby/marshaltestlib.rb: catch up above changes.
+
+	* test/ruby/test_class.rb: ditto.
+
+	* test/test_pp.rb: ditto.
+
+Sat Oct 27 10:50:53 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* object.c (rb_mod_const_get): make sure the constant name is
+	  converted to a string before searching. [ruby-core:48405]
+
+Sat Oct 27 10:12:13 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* iseq.c (rb_iseq_compile_with_option): Instead of testing
+	  respond_to, just check if the argument is actually a file,
+	  because by calling user-defined gets something weired can
+	  happen.  Patch by Glass_saga. [ruby-dev:40202] [Bug #2861]
+
+	* parse.y (ripper_initialize): ditto.
+
+Sat Oct 27 10:07:57 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (enum lex_state_e): [EXPERIMENTAL] lex_state as bit field /
+	  IS_lex_state() macro.  based on the patch by Dave B in
+	  [ruby-core:23503].  [Feature #1493]
+
+Sat Oct 27 10:05:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (fstat): use _fstati64() instead of fstati64()
+	  on mingw32.
+
+Sat Oct 27 06:28:33 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* object.c (rb_mod_const_get): const_get accepts qualified constant
+	  strings.  e.g. Object.const_get("Foo::Bar::Baz") [ruby-core:41404]
+
+	* test/ruby/test_module.rb: tests for new behavior
+
+Fri Oct 26 13:24:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (literal_concat_gen): merge fixed strings across
+	  concatenated literals, after an interpolation.
+
+Thu Oct 25 17:48:54 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (has_redirection): should use shell (cmd.exe) when
+	  the commandline contains '&'.
+	  reported by Roger Pack at [ruby-core:47912] [Bug #7143], and
+	  patched by Heesob Park at [ruby-core:47931].
+
+Thu Oct 25 15:00:08 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, class.c: remove (revert)
+	  `rb_add_method_cfunc_frameless()' API.
+	  This API is not mature to become an official API.
+	  For example, we can not use this API with
+	  `rb_define_private_method()'.
+
+	* method.h, vm_method.c (rb_add_method_cfunc_frameless): removed.
+
+Thu Oct 25 13:35:07 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/mkconfig.rb: remove string literal concatenation.
+
+Wed Oct 24 18:49:16 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/objspace/objspace.c (type2sym, count_objects_size): use enum
+	  instead of size_t which may be larger than actual values.
+
+Wed Oct 24 17:41:24 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: add `-x' or `--exclude' option
+	  to specify exclude benchmark name pattern.
+	  You can specify "-x foo" if you want to exclude the benchmarks
+	  if the name of benchmark contains `foo'.
+
+Wed Oct 24 11:57:24 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_prepare_free_objects): rename to match the behavior of
+	  this function.
+
+Wed Oct 24 11:55:19 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (reachable_object_from_i): change data
+	  structure of the result of reachable objects. Keys of table
+	  contains object_id of each reachable objects. Value of table
+	  is an object itself or an instance of InternalObjectWrapper.
+	  To avoid duplication, we use st_table and object_id keys.
+
+	* ext/objspace/objspace.c (type2sym): bug fix.
+	  Should use `i' instead of `type'.
+
+Wed Oct 24 10:33:09 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect, gc_marks): move the location of
+	  clear and restore rb_objspace_t::mark_func_data
+	  from garbage_collect() to gc_marks().
+
+Wed Oct 24 10:17:45 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (Init_objspace): add a new method
+	  `ObjectSpace::InternalObjectWrapper#internal_object_id' which returns
+	  an object id of a wrapped internal object.
+
+Wed Oct 24 08:55:04 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (ObjectSpace.reachable_objects_from):
+	  internal object support.
+	  If given object `obj' has references to internal objects
+	  (such as T_NODE objects), then this method returns instances of
+	  `ObjectSpace::InternalObjectWrapper' instead of that internal objects.
+	  This instance contains a reference to an internal object and you can
+	  check the type of internal object using
+	  `ObjectSpace::InternalObjectWrapper#type' method.
+	  Rdoc of `InternalObjectWrapper' is not prepared yet.
+
+	* gc.c (rb_objspace_reachable_objects_from), gc.h: change
+	  an interface of 'rb_objspace_reachable_objects_from()'
+
+	* gc.c, gc.h: add two APIs
+	  - rb_objspace_markable_object_p(obj): check markable or not.
+	  - rb_objspace_internal_object_p(obj): check internal or not.
+
+Wed Oct 24 05:52:36 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method): remove `default' and
+	  add a case for `VM_METHOD_TYPE_UNDEF'.
+
+Wed Oct 24 05:41:18 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_error.c (error_print), vm_eval.c (eval_string_with_cref),
+	  vm_trace.c (rb_suppress_tracing): use TH_PUSH_TAG() instead of
+	  PUSH_TAG().
+
+Wed Oct 24 05:17:52 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (vm_call0_body): remove RUBY_VM_CHECK_INTS()
+	  after method invocation using rb_call0().
+
+	* vm_eval.c (vm_call0_body): remove default section on top of
+	  switch statement and add cases for `VM_METHOD_TYPE_CFUNC_FRAMELESS'
+	  and `VM_METHOD_TYPE_UNDEF'.
+
+	* vm_eval.c (vm_call0_body): remove useless brackets.
+
+Tue Oct 23 22:34:49 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (thread_raise_m):  check interrupts after Thread#raise
+	  if a target thread is the current thread because the behavior
+	  of Thread.current.raise is expected to perform same as
+	  Kernel.raise (by rubyspec).
+
+Tue Oct 23 17:08:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (usage, process_options): show more info in --help.
+	  [EXPERIMENTAL] [ruby-core:48072] [Bug #7184]
+
+Tue Oct 23 14:20:43 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-electric.el using variable `last-command-event' instead of
+	  obsolete `last-command-char', so that work with Emacs trunk.
+	  a patch by Victor Deryagin <vderyagin AT gmail.com>.
+
+Tue Oct 23 14:06:47 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (visibility_option): visibility attribute is not
+	  available before GCC 4, so do not use -fvisibility option in that
+	  case.  [ruby-core:48147] [Bug #7205]
+
+Tue Oct 23 12:57:29 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, vm_insnhelper.c, vm_eval.c (OPT_CALL_CFUNC_WITHOUT_FRAME):
+	  add a new optimization and its macro `OPT_CALL_CFUNC_WITHOUT_FRAME'.
+	  This optimization makes all cfunc method calls `frameless', which
+	  is faster than ordinal cfunc method call.
+	  If `frame' is needed (for example, it calls another method with
+	  `rb_funcall()'), then build a frame. In other words, this
+	  optimization delays frame building.
+	  However, to delay the frame building, we need additional overheads:
+	    (1) Store the last call information.
+	    (2) Check the delayed frame building before the frame is needed.
+	    (3) Overhead to build a delayed frame.
+	  rb_thread_t::passed_ci is storage of delayed cfunc call information.
+	  (1) is lightweight because it is only 1 assignment to `passed_ci'.
+	  To achieve (2), we modify GET_THREAD() to check `passed_ci' every
+	  time. It causes 10% overhead on my environment.
+	  This optimization only works for cfunc methods which do not need
+	  their `frame'.
+	  After evaluation on my environment, this optimization does not
+	  effective every time. Because of this evaluation results, this
+	  optimization is disabled at default.
+
+	* vm_insnhelper.c, vm.c: add VM_PROFILE* macros to measure behaviour
+	  of VM internals. I will extend this feature.
+
+	* vm_method.c, method.h: change parameters of the `invoker' function.
+	  Receive `func' pointer as the first parameter.
+
+Tue Oct 23 06:21:05 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c: just get the constant defined in Ruby.
+
+	* ext/psych/lib/psych/syntax_error.rb: Psych::SyntaxError now inherits
+	  from StandardError rather than SyntaxError.  Thanks Eric Hodel!
+
+	* test/psych/test_exception.rb: tests for change.
+
+Tue Oct 23 06:17:36 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: Cache symbols while
+	  tokenizing.  Thanks Kevin Menard!
+
+Tue Oct 23 06:15:40 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: Updated the RegExp to catch
+	  Strings earlier in the tokenization process.  Thanks Kevin Menard!
+
+Tue Oct 23 06:12:39 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Handle nil tags specially
+	  to avoid slow method_missing calls.  Thanks Kevin Menard!
+
+Tue Oct 23 06:07:57 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: Ignore bad timestamps.  If
+	  something looks like a timestamp but has an invalid component, treat
+	  it as a string instead of throwing an ArgumentError.
+	  Thanks Rhett Sutphin!
+
+	* test/psych/test_scalar_scanner.rb: appropriate tests.
+
+Tue Oct 23 06:04:07 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: Fix scalar_scanner to
+	  understand strings starting with an underscore and containing only
+	  digits.  Thanks Kelley Reynolds.
+
+	* test/psych/test_scalar_scanner.rb: test for fix
+
+Tue Oct 23 06:00:41 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: Changed comment in psych.rb to update new
+	  home page for libyaml.  Thanks to Carolyn Ann.
+
+Sun Oct 21 19:12:59 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_core.h (rb_vm_t::trace_running): add a new field
+	  `trace_running' to store vm global tracing status.
+
+	* vm_trace.c: fix SEGV bug. event_hook was free'd
+	  even when the hook is still used in another thread.
+	  [ruby-dev:46141] [Bug #7032]
+
+Sun Oct 21 19:12:42 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_core.h (rb_vm_t::trace_flag): remove `trace_flag'
+	  which is no longer used.
+
+Sun Oct 21 18:34:27 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (date__parse): uses more tight parser if
+	  defined TIGHT_PARSER.  now inactivated; because it introduces
+	  incompatibilities and it is a bit slow.
+
+Sat Oct 20 15:35:06 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* include/ruby/ruby.h: add C APIs.
+	  VALUE rb_newobj_of(VALUE klass, VALUE flags)
+	  #define NEWOBJ_OF(obj,type,klass,flags)
+	  These allow to change a allocation strategy depending on klass
+	  or flags.
+
+	* gc.c: ditto
+
+	* array.c: use new C API.
+	* bignum.c: ditto
+	* class.c: ditto
+	* complex.c: ditto
+	* ext/socket/ancdata.c: ditto
+	* ext/socket/option.c: ditto
+	* hash.c: ditto
+	* io.c: ditto
+	* marshal.c: ditto
+	* numeric.c: ditto
+	* object.c: ditto
+	* random.c: ditto
+	* range.c: ditto
+	* rational.c: ditto
+	* re.c: ditto
+	* string.c: ditto
+	* struct.c: ditto
+	  [Feature #7177][Feature #7047]
+
+Sat Oct 20 12:50:00 2012  Zachary Scott	<zachary@zacharyscott.net>
+
+	* ext/socket/socket.c: Documentation for Socket
+	  Based on a patch by David Albert
+	  [Bug #7105] [ruby-core:47828]
+
+Sat Oct 20 11:00:00 2012  Zachary Scott	<zachary@zacharyscott.net>
+
+	* lib/open-uri.rb: Documentation for OpenURI
+
+Sat Oct 20 06:18:34 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* hash.c (initialize_copy): unset the default proc if there isn't one
+	  for the target hash, call to_hash, check frozen status.
+
+Fri Oct 19 22:22:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_vm_jump_tag_but_local_jump): pass through thrown objects.
+	  [ruby-dev:46234] [Bug #7185]
+
+	* vm_eval.c (rb_eval_cmd): if state is non-zero, val should be nil and
+	  rb_vm_jump_tag_but_local_jump() just jump tag.
+
+Fri Oct 19 22:11:55 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* pack.c (pack_unpack): set encoding of the
+	  'H','h','B' and 'B' modifiers to US-ASCII.
+
+	* test/ruby/test_pack.rb: tests for the above.
+	  [ruby-core:47653][Bug #7050]
+
+	* test/test_securerandom.rb: tests for SecureRandom.hex
+	  from tenderlove. [ruby-core:46792][Bug #6799]
+
+Fri Oct 19 19:29:11 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h (rb_method_cfunc_t::invoker): add new field (func ptr)
+	  `invoker'. `invoker' function invoke cfunc body
+	  (rb_method_cfunc_t::func).
+	  `invoker' is set at method definition timing.
+	  With this change, the big `switch' (branch) in `call_cfunc()'
+	  is no longer needed.
+	  However, the performance benefit is only a bit.
+
+	* vm_core.h (rb_call_info_t::aux::func): add a new field to store
+	  cfunc body function pointer.
+
+	* vm_method.c (call_cfunc_invoker_func): add a new function which
+	  returns a suitable invoke function.
+
+	* vm_method.c (setup_method_cfunc_struct): added.
+
+	* vm_method.c (rb_add_method): fix to set `invoker'.
+
+	* vm_eval.c (vm_call0_body): catch up above changes.
+
+	* vm_insnhelper.c (call_cfunc): removed.
+
+	* vm_insnhelper.c (vm_call_cfunc): fix to call cfunc body
+	  with `invoker' function.
+
+Fri Oct 19 16:55:58 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, vm_eval.c: use TH_PUSH_TAG() instead of PUSH_TAG().
+
+Fri Oct 19 11:13:55 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: remove unexpected `output'.
+	  (commit miss)
+
+Fri Oct 19 10:24:03 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_search_method): remove needless local variable.
+
+Fri Oct 19 10:22:26 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bmx_temp.rb: removed.
+	  This file should not be in repository.
+
+Fri Oct 19 10:20:10 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: add new option `--ruby-arg [ARG]'
+	  which is passed as a launch parameter for each ruby's execution.
+	  ($ ruby [ARG] [File])
+
+Thu Oct 18 18:42:35 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (opt_send_simple): move the location of
+	  `opt_send_simple' to the place near `send' definition.
+	  (to take care about icache locality).
+
+Thu Oct 18 18:29:25 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (send): remove unused condition.
+	  This condition will be true after r37258.
+
+	* vm_insnhelper.c (vm_caller_setup_args): remove `UNLIKELY' on
+	  checking blockiseq (it seems `LIKELY').
+
+Thu Oct 18 17:31:58 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (opt_send_simple): introduce new instruction used
+	  when no need to care about block and splat.
+
+	* compile.c: use the `opt_send_simple' instruction.
+
+Thu Oct 18 16:44:07 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_add_method_cfunc, rb_add_method_cfunc_frameless):
+	  check arity earlier at definition time.
+
+Thu Oct 18 15:11:31 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: add `inline' keyword to several functions.
+	  Compilers (gcc) are conservative than I expected.
+
+Thu Oct 18 15:01:15 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: add a decl. of
+	  `rb_define_frameless_method()'.
+
+Thu Oct 18 14:31:17 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (new_callinfo): set a temporary index of callinfo
+	  (used in `iseq_set_sequence()') to rb_call_info_t::aux::index.
+	  rb_call_info_t::argc is initialized by same value of
+	  rb_call_info_t::orig_argc.
+
+Thu Oct 18 14:11:08 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (rb_define_frameless_method): rename from
+	  rb_define_method_fast(). Defined method with this C API
+	  does not make a method frame. It is bit lightweight than
+	  ordinal C functions. Now only 0 or 1 argc are permitted.
+
+	* method.h (VM_METHOD_TYPE_CFUNC_FRAMELESS): rename macro name
+	  from VM_METHOD_TYPE_CFUNC_FAST.
+
+	* vm_insnhelper.c, vm_method.c: rename related functions.
+
+	* proc.c (rb_method_entry_arity): catch up above changes.
+
+Thu Oct 18 10:30:34 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assignable_gen): fail if yyerror occurred.  fix a bug in
+	  r36973.
+
+Thu Oct 18 09:23:03 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* hash.c (initialize_copy): duping should rehash the hash.
+
+	* test/ruby/test_hash.rb: added a test to ensure rehash.
+
+Wed Oct 17 21:16:47 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* common.mk (WPROGRAM): need same dependencies as PROGRAM.
+
+	* cygwin/GNUmakefile.in (uncommon.mk): move include position
+	  below WPROGRAM definition to be defined in uncommon.mk.
+
+	* ext/extmk.rb (all, static): fix make rubyw.exe failure with make -jN.
+	  If make of ruby.exe and rubyw.exe run in parallel, link dll and link
+	  exe run in parallel, which causes link failure on mingw. To fix this,
+	  we make ruby.exe and rubyw.exe in one make process.
+	  [ruby-core:48007] [Bug #7165]
+
+Wed Oct 17 16:25:34 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm2_method_missing.rb: add a benchmark to measure
+	  performance of invoking `method_missing'.
+
+Wed Oct 17 16:23:17 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_getivar): fix to use `aux.index' instead of
+	  `aux.opt_pc'.
+
+Wed Oct 17 16:03:54 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method_missing): make a refactoring
+	  about method_missing process. Use `vm_call_method()' to invoke
+	  `method_missing' method instead of `rb_funcall2()'.
+	  In `vm_call_method()', set fastpath to `vm_call_method_missing()'
+	  if it can be cached.
+
+	* vm_core.h (rb_call_info_t): add new field
+	  `rb_call_info_t::aux::missing_reason' to pass the reason to
+	  `vm_call_method_missing()'.
+
+Wed Oct 17 15:33:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (opt-dir): allow multiple directories separated by
+	  $PATH_SEPARATOR as well as dir_config in mkmf.rb. [ruby-core:47868]
+	  [Bug #7120]
+
+Wed Oct 17 15:08:13 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: fix Net::IMAP::ResponseParser to accept
+	  message/delivery-status ([ruby-core:47920] [Bug #7146]),
+	  message/rfc822 attachments ([ruby-core:47921] [Bug #7147]), and
+	  (BODY ("MIXED")) ([ruby-core:47951] [Bug #7153]).
+	  patched by Tony Arkles.
+
+	* test/net/imap/test_imap_response_parser.rb: related test.
+
+Wed Oct 17 11:04:48 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_hash.rb (TestHash#test_dup_equality): added a new test
+	  to show the problem of r37232.
+
+Wed Oct 17 10:48:40 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_method): fix a build error that occurs
+	  when OPT_INLINE_METHOD_CACHE is 0.
+
+Wed Oct 17 08:46:47 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm2_dstr.rb: add a benchmark to measure
+	  performance of dynamic generated string ("foo#{bar}baz").
+
+Wed Oct 17 08:32:46 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_dstr_fragments): use `putobject' instead of
+	  `putstring' for all of strings used by NODE_DSTR because
+	  ruby users can not grab this string.
+	  For example, the string object of "baz" in "foo#{bar}baz"
+	  is located by `putobject' (users can not touch "baz" object
+	  directly). This change reduces GC pressure.
+	  This improvement is suggested by Aaron Patterson.
+
+Wed Oct 17 08:02:57 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_threadptr_interrupt_mask): fix to check interrupt
+	  after interrupt_mask changed.
+
+Wed Oct 17 06:42:47 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method): fix to return value immediately.
+	  Remove CHECK_INTS() after that method dispatch.
+
+Wed Oct 17 06:25:56 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* hash.c (initialize_copy): copy the underlying st_table on dup,
+	  rather than copying the hash key by key. [ruby-core:48009]
+
+	* test/ruby/test_hash.rb: relevant tests for initialize_copy
+
+Wed Oct 17 06:17:44 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_iseq_setup_2): separate tailcall and normal
+	  method frame setup functions.
+	  Add checking interrupts at the tailcall setup function.
+
+Wed Oct 17 05:35:37 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm1_yield.rb: add a benchmark to measure `yield'
+	  (invoke empty block) performance.
+
+	* benchmark/bm_vm2_method_with_block.rb: add a benchmark to measure
+	  method invocation with empty block.
+
+Wed Oct 17 05:05:07 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_invoke_block): vm_caller_setup_args() can skip
+	  when splat flag is not set.
+
+Wed Oct 17 01:53:47 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_getivar, vm_setivar): support index inline cache
+	  with rb_call_info_t to speedup `attr' getter and setter.
+	  Cached index is stored in rb_call_info_t::aux::index.
+	  `index' == 0 means not cached. `index' > 0 means cached and cached
+	  index is `index - 1'.
+
+	* insns.def ((get|set)instancevariable): use new wrapper functions
+	  vm_(get|set)instancevariable() defined in vm_insnhelper.c.
+
+	* vm_core.h (rb_call_info_t::aux): introduce new union data because
+	  opt_pc can share with index.
+
+Tue Oct 16 22:24:44 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb (show_results): Show speedup ratio
+	  with first executables score at last of results
+	  if two or more executables are given.
+
+Tue Oct 16 21:59:01 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: some refactoring.
+	  (1) Remove `average differential'.
+	      In this benchmark driver, We should not care about `average'.
+	      We use fastest score because this score should not include
+	      any disturbances (affections of background process, etc).
+	      If you care about timing affect, I recommend `median'
+	      score with more than 5 examinations rather than simple
+	      `average' score (`average' score was affected by error scores).
+	  (2) Show log file name.
+	  (3) Change default directory from './' to driver's directory.
+
+Tue Oct 16 14:56:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_join): need to check again after any conversion run.
+	  [ruby-core:48012] [Bug #7168]
+
+Tue Oct 16 12:52:14 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#assert_file):
+	  rename from file_assertion.
+
+Tue Oct 16 11:30:18 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_join): hide the result under construction until
+	  return.
+
+	* file.c (rb_file_join): check nul-byte only for strings, since
+	  FilePathStringValue() does it.  [ruby-core:48012] [Bug #7168]
+
+	* file.c (rb_file_join): path names must be ASCII-compatible.
+	  [ruby-core:48012] [Bug #7168]
+
+	* file.c (check_path_encoding): new function to ensure path name
+	  encoding to be ASCII-compatible.
+
+Tue Oct 16 09:40:04 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_regexp.rb
+	  (TextRegexp#test_raw_hyphen_and_tk_char_type_after_range): use
+	  Regexp.new instead of literal to ignore a parser warning.
+
+Tue Oct 16 09:30:30 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_regexp.rb
+	  (TextRegexp#test_raw_hyphen_and_tk_char_type_after_range): ignoring
+	  warnings are already set in setup method.
+
+Tue Oct 16 06:44:06 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (VM_CALLEE_SETUP_ARG): fix wrong condition.
+
+Tue Oct 16 06:29:18 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method): disable CI_SET_FASTPATH() if
+	  this method call needs splat argument because cached functions
+	  (vm_call_attrset, vm_call_ivar, vm_call_cfunc_fast_(unary|binary))
+	  do not check an arity.
+
+	* bootstraptest/test_method.rb: add a test to check an above issue.
+
+Tue Oct 16 06:15:44 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: introduce new method type VM_METHOD_TYPE_CFUNC_FAST.
+	  This method is similar to VM_METHOD_TYPE_CFUNC methods, but
+	  called cfunc without building new frame (does not push new control
+	  frame). If error is occurred in cfunc, the backtrace only shows
+	  caller frame and upper.
+	  This kind of methods can be added by rb_define_method_fast().
+	  This feature is similar to specialized instructions (opt_plus, etc),
+	  but more flexible (but a bit slower).
+
+	* class.c (rb_define_method_fast): added.
+	  Maybe it will be renamed soon.
+
+	* vm_insnhelper.c (vm_call_method): support method type
+	  VM_METHOD_TYPE_CFUNC_FAST.
+
+	* proc.c (rb_method_entry_arity): catch up new method type.
+
+	* vm_method.c (rb_add_method_cfunc_fast): added.
+
+Tue Oct 16 02:32:29 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h (CI_SET_FASTPATH): add new parameter `enabled'.
+	  If `enable' is 0 then CI_SET_FASTPATH() doesn't work.
+	  And add new configuration option OPT_CALL_FASTPATH. If this macro
+	  was defined by 0, then CI_SET_FASTPATH() doesn't work any more.
+
+	* vm_insnhelper.c (vm_call_method): Pass `0' for `enabled' parameter
+	  of CI_SET_FASTPATH if this method is protected.
+
+Tue Oct 16 02:17:35 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (VM_CALL_*): rename VM_CALL_*_BIT
+	  to VM_CALL_* (remove `_BIT' suffix).
+	  Add comments on each macros.
+	  Remove unused macro VM_CALL_TAILRECURSION_BIT.
+
+	* compile.c, iseq.c, insns.def, vm_insnhelper.c: ditto.
+
+Mon Oct 15 22:14:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (Test::Unit::Assertions#file_assertion):
+	  rewrite file assertions.
+
+Mon Oct 15 09:41:17 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (VM_CALLEE_SETUP_ARG): skip CI_SET_FASTPATH() if
+	  it was called from vm_yield_setup_args().
+
+Mon Oct 15 05:20:13 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h CI_SET_FASTPATH: introduce new macro
+	  `CI_SET_FASTPATH(ci, func)'. This macro set `ci->call' as `func'.
+	  `func' (ci->call) is called at the last of `send'
+	  (and `invokesuper') instruction.
+	  `CI_SET_FASTPATH' does not set `ci->call' when the method
+	  (stored in `ci->me') is `protected'.
+
+	* vm_insnhelper.c (vm_call_method): use `CI_SET_FASTPATH'.
+	  After several checking (visibility, argc checking), the result of
+	  checking can be reused until re-definition of this method
+	  with inline method cache.
+
+	  Note that this optimization is now experimental.
+	  If you find any problem about it, please tell us.
+
+Mon Oct 15 04:51:55 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: refactoring.
+	  - move all `call' related functions to the last of file.
+	  - make functions for respective method types in vm_call_method().
+	    (all functions have same function parameters)
+
+	* vm_core.h: add `opt_pc' field in `rb_call_info_t'
+	  as temporal variable.
+
+	* vm_eval.c (vm_call0_body): catch up above changes.
+
+Mon Oct 15 03:51:46 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm1_attr_ivar(_set).rb: added (for method dispatch speed).
+
+	* benchmark/bm_vm1_float_simple.rb: added (for flonum/float).
+
+Mon Oct 15 02:51:16 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (vm_call0_body): add new function.
+	  `vm_call0()' makes call_info struct and calls `vm_call0_body()'
+	  with this struct. In near future, `vm_call0()' will be removed
+	  because all of `vm_call0()' users setup call_info struct by itself.
+
+Mon Oct 15 01:38:06 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (send, invokesuper, invokeblock, opt_*), vm_core.h:
+	  use only a `ci' (rb_call_info_t) parameter instead of using
+	  parameters such as `op_id', 'op_argc', `blockiseq' and flag.
+	  These information are stored in rb_call_info_t at the compile
+	  time.
+	  This technique simplifies parameter passing at related
+	  function calls (~10% speedups for simple method invocation at
+	  my machine).
+	  `rb_call_info_t' also has new function pointer variable `call'.
+	  This `call' variable enables to customize method (block)
+	  invocation process for each place. However, it always call
+	  `vm_call_general()' at this changes.
+	  `rb_call_info_t' also has temporary variables for method
+	  (block) invocation.
+
+	* vm_core.h, compile.c, insns.def: introduce VM_CALL_ARGS_SKIP_SETUP
+	  VM_CALL macro.  This flag indicates that this call can skip
+	  caller_setup (block arg and splat arg).
+
+	* compile.c: catch up above changes.
+
+	* iseq.c: catch up above changes (especially for TS_CALLINFO).
+
+	* tool/instruction.rb: catch up above changes.
+
+	* vm_insnhelper.c, vm_insnhelper.h: ditto. Macros and functions
+	  parameters are changed.
+
+	* vm_eval.c (vm_call0): ditto (it will be rewritten soon).
+
+Sun Oct 14 12:30:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (rb_f_sub, rb_f_gsub): pass the given block.
+	  [ruby-core:47967] [Bug #7157]
+
+Sat Oct 13 23:15:39 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* regparse.c (parse_char_class): should match with a hyphen after a
+	  range in a character class.
+
+	* test/ruby/test_regexp.rb (TestRegexp#test_char_class): fixed wrong
+	  test.
+
+	* test/ruby/test_regexp.rb (TestRegexp#check): now can accept the
+	  error message.
+
+	* test/ruby/test_regexp.rb
+	  (TextRegexp#test_raw_hyphen_and_tk_char_type_after_range): renamed
+	  because the previous name was wrong.
+
+	* test/ruby/test_regexp.rb
+	  (TextRegexp#test_raw_hyphen_and_tk_char_type_after_range): added
+	  more test pattern.
+
+Sat Oct 13 03:01:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (realpath_rec): prevent link from GC while link_names refers
+	  the content.
+
+Sat Oct 13 01:37:48 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_regexp.rb
+	  (TestRegexp#test_raw_hyphen_and_type_char_after_range): added new
+	  test. ref [ruby-core:47115] [Backport #6853]
+
+Fri Oct 12 21:55:08 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* include/ruby/win32.h (rb_w32_pow): set floating point precision
+	  for mingw-w64 x86 pow(). This improves the precision of pow() on
+	  Windows XP for TestFloat#test_round_with_precision failure.
+	  [ruby-core:47911] [Bug #7142]
+
+Fri Oct 12 21:37:25 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/webrick/test_cgi.rb (TestWEBrickCGI#test_cgi): skip a test
+	  depending on locale on Windows. ENV[] doesn't work properly if
+	  console code page is not equal to file system encoding.
+	  [ruby-core:47910] [Bug #7140]
+
+Fri Oct 12 20:40:29 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (posix_sh_cmds): the command name of colon is ":".
+
+Fri Oct 12 18:18:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_get_path_check): path name must not contain NUL bytes.
+
+Fri Oct 12 16:06:20 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/merger.rb: now can merge revision(s) without --ticket again.
+
+Fri Oct 12 14:10:41 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (dir_config, init_mkmf): use configured libdir value as
+	  default library path.  [ruby-core:43726] [Bug #6207]
+
+Fri Oct 12 05:25:00 2012  Zachary Scott	<zzak@ruby-lang.org>
+
+	* lib/timeout.rb (timeout):
+	  Remove paragraph on wrong implementation detail.
+	  [ruby-core:47739] [Bug #7088]
+
+Thu Oct 11 23:09:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_sub{seq,pos,str}, rb_str_each_{line,codepoint}):
+	  prevent String copies from GC.  [ruby-core:47881] [Bug #7135]
+
+Thu Oct 11 07:40:50 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* iseq.c (insn_operand_intern): cast op to rb_call_info_t* before
+	  compare with iseq->callinfo_entries whose type is rb_call_info_t*.
+
+Thu Oct 11 03:37:08 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_block.rb: add tests for block with super.
+
+Thu Oct 11 02:54:07 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c: fix debug prints to catch up recent changes
+	  such as VM data structures.
+
+Thu Oct 11 02:50:34 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (insn_operand_intern): add support disasm TS_CALLINFO
+	  operands.
+
+Wed Oct 10 15:12:48 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_process.rb (TestProcess#test_execopts_gid): skip on
+	  windows because the platform does not have Process.group method.
+	  patched by Jon Forums in [ruby-core:47878] [Bug #7133].
+
+Tue Oct  9 23:18:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_file, assert_file_not): more
+	  descriptive assertions for File predicates.
+
+Tue Oct  9 18:01:37 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sample): use rb_random_ulong_limited, since
+	  precision of long may be larger than double.
+
+	* random.c (rb_random_ulong_limited): new function to return a random
+	  value from 0 upto limit as unsigned long, similarly to
+	  rb_genrand_ulong_limited but with arbitrary RNG object.
+
+Tue Oct  9 17:13:27 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_execarg_addopt, rb_execarg_run_options): add :uid and
+	  :gid options.  [ruby-core:47414] [Feature #6975]
+
+Tue Oct  9 14:36:11 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_free): fix memory leak.
+	  rb_iseq_t::callinfo_entries should be freed.
+
+Tue Oct  9 14:28:18 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (rb_call_info_t): add new type `rb_call_info_t'.
+	  This data structure contains information including inline method
+	  cache. After that, `struct iseq_inline_cache_entry' does not
+	  need to contain inline cache for method invocation.
+	  Other information will be added to this data structure.
+
+	* vm_core.h (rb_iseq_t): add `callinfo_entries' and `callinfo_size'
+	  members to `rb_iseq_t'.
+
+	* insns.def, compile.c: Use CALL_INFO instead of IC.
+
+	* tool/instruction.rb: support CALL_INFO as operand type.
+
+	* vm_insnhelper.c, vm_insnhelper.h: ditto.
+
+Sun Oct  7 23:54:33 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/zlib/zlib.c (zstream_run_func): don't call inflate() when
+	  z->stream.avail_in == 0. it return Z_BUF_ERROR.
+	  but deflate() could be called with z->stream->avail_in == 0 because
+	  it has hidden buffer in z->stream->state (opaque structure).
+	  fix for gem install error. [ruby-dev:46149] [Bug #7040]
+
+Mon Oct  8 23:55:41 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refinements): new method Module#refinements.
+
+	* test/ruby/test_refinement.rb: add new tests for the above changes.
+
+Mon Oct  8 23:02:19 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c, gc.c, iseq.c, node.h, vm_insnhelper.c, vm_insnhelper.h,
+	  vm_method.c: rename omod and overlaid modules to refinements.
+
+	* eval.c (hidden_identity_hash_new): renamed from identity_hash_new.
+
+Sun Oct  7 04:50:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* lib/abbrev.rb: Documentation examples for Abbrev.
+	  [ruby-core:47442] [Bug #6985]
+
+Sun Oct  7 04:50:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* thread.c (rb_thread_aref):
+	  Grammar in Thread documentation.
+	  Patch by Steve Klabnik [ruby-core:47799] [Bug #7099]
+
+Sun Oct  7 04:37:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* string.c (rb_str_match):
+	  Clarify behavior for captured strings and local variable assignment
+	  Patch by Marcus Stollsteimer [ruby-core:47668] [Bug #7062]
+
+Sat Oct  6 18:31:36 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_opts.h (OPT_GLOBAL_METHOD_CACHE): new build option to
+	  enable/disable global method caching. [ruby-dev:46203] [Bug #7111]
+
+	* vm_method.c (rb_method_entry_get_with_omod): don't use global
+	  method cache if OPT_GLOBAL_METHOD_CACHE is 0.
+
+Sat Oct  6 16:32:04 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (search_method): check omod only once for performance.
+
+Sat Oct  6 09:42:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/encdb.c, enc/utf_16_32.h (ENC_DUMMY_UNICODE): endian-less wide
+	  UTF encodings are dummy but Unicode.
+
+	* encoding.c (rb_encdb_set_unicode): set Unicode flag.
+
+	* template/encdb.h.tmpl: allow ENC_DUMMY variants.
+
+	* encoding.c (rb_enc_unicode_p): oniguruma provides Unicode flag.
+
+Fri Oct  5 17:18:42 JST 2012  TAKANO Mitsuhiro  <tak@no32.tk>
+
+	* template/Doxyfile.tmpl: remove SHOW_DIRECTORIES and
+	  HTML_ALIGN_MEMBERS lines. They have been obsolete in
+	  Doxygen version 1.8.2.
+
+Fri Oct  5 15:26:18 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c: add ObjectSpace#reachable_objects_from.
+	  This method returns an array of objects referenced by given object.
+	  If given object is special objects such as true/false/nil/Fixnum etc
+	  then it returns nil. See rdoc for details.
+	  [ruby-core:39772]
+
+	* test/objspace/test_objspace.rb: add a test for this method.
+
+	* gc.c: add rb_objspace_reachable_objects_from().
+	  To make this function, add several member `mark_func_data'
+	  to rb_objspace_t.  If mark_func_data is not null, then
+	  gc_mark() calls mark_func_data::mark_func.
+
+	* gc.h: export rb_objspace_reachable_objects_from().
+
+Thu Oct  4 23:40:04 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (init_heap): call init_mark_stack before to allocate
+	  altstack. This change avoid the stack overflow at the signal
+	  handler on 32bit, but I don't understand reason... [Feature #7095]
+
+Thu Oct  4 22:39:27 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (getlocal, setlocal): remove old getlocal/setlocal
+	  instructions and rename getdaynmic/setdynamic instructions
+	  to getlocal/setlocal.
+
+	* compile.c: ditto.
+
+	* iseq.c: remove TS_DINDEX.
+
+	* vm_exec.h (dindex_t): remove type definition of `dindex_t'.
+
+	* tool/instruction.rb: ditto.
+
+Thu Oct  4 21:44:17 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_analysis_insn|operand|register): use st_insert
+	  instead of using rb_hash_aset() because rb_hash_aset()
+	  check $SAFE.
+
+Thu Oct  4 21:15:26 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (VM_COLLECT_USAGE_DETAILS): make new VM usage analysis
+	  hooks (old macro name is COLLECT_USAGE_ANALYSIS).
+	  This feature is only for VM developers.  (I'm not sure I can use
+	  `VM developers' (the plural form) in this sentence).
+	  If VM_COLLECT_USAGE_DETAILS is not 0, VM enables the following
+	  usage collection features:
+	    (1) instruction: collect instruction usages.
+	    (2) operand: collect operand usages.
+	    (3) register: collect register usages.
+	  The results are stored in
+	    RubyVM::USAGE_ANALYSIS_INSN for (1, 2),
+	    RubyVM::USAGE_ANALYSIS_INSN_BIGRAM for (1) and
+	    RubyVM::USAGE_ANALYSIS_REGS for (3).
+	  You can stop collecting usages with
+	    RubyVM::USAGE_ANALYSIS_INSN_STOP(),
+	    RubyVM::USAGE_ANALYSIS_OPERAND_STOP(),
+	    RubyVM::USAGE_ANALYSIS_REGISTER_STOP()
+	  for (1), (2), (3) respectively.
+	  You can also change the hook functions by setting
+	  C level global variables
+	    `ruby_vm_collect_usage_func_(insn|operand|register)'
+	  for (1), (2), (3) respectively.
+	  See codes for more details.
+
+	* tool/instruction.rb: fix macro names.
+
+	* iseq.c (insn_operand_intern): make it export (used in vm.c).
+	  fix to skip several processes if not needed (pointer is 0).
+
+	* vm_dump.c: move codes for collection features to vm.c.
+
+	* vm_exec.h: rename macro and function names.
+
+	* vm_insnhelper.h: ditto.
+
+Thu Oct  4 18:59:14 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_settracefunc.rb (test_tracepoint):
+	  remove unused test case.
+	  (this test case is redefined by newer tests)
+
+Thu Oct  4 17:24:51 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (rb_objspace_call_finalizer): call gc_mark_stacked_objects
+	  at suitable point.
+
+Thu Oct  4 16:31:29 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_objspace_call_finalizer): mark self-referencing finalizers
+	  before run finalizers, to fix SEGV from btest on 32bit.
+
+	* gc.c (gc_mark_stacked_objects): extract from gc_marks().
+
+Thu Oct  4 11:43:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack): round stack limit to page size
+	  boundary to calculate stack size more precisely.  [ruby-dev:46174]
+	  [Bug #7084]
+
+Wed Oct  3 19:51:57 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: Use the non-recursive marking instead of recursion. The
+	  recursion marking of CRuby needs checking stack overflow and the
+	  fail-safe system, but these systems not good at partial points,
+	  for example, marking deep tree structures. [ruby-dev:46184]
+	  [Feature #7095]
+
+	* configure.in (GC_MARK_STACKFRAME_WORD): removed. It's used by
+	  checking stack overflow of marking.
+
+	* win32/Makefile.sub (GC_MARK_STACKFRAME_WORD): ditto.
+
+Wed Oct  3 15:33:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack): use getrlimit() for the main
+	  thread on Mac OS X, since pthread_get_stack{addr,size}_np()
+	  return the default value always, but not the ulimit value.
+	  [ruby-dev:46174] [Bug #7084]
+
+Wed Oct  3 11:43:15 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_reopen): improvement to accept optional arguments.
+	  a patch by Glass_saga (Masaki Matsushita) in [ruby-core:47806].
+	  [Feature #7103]
+
+Wed Oct  3 04:36:11 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_add_file):  Added
+	  documentation
+	* ext/openssl/ossl_x509store.c (ossl_x509store_set_default_paths):
+	  ditto
+	* ext/openssl/ossl_x509store.c (ossl_x509store_add_cert):  ditto
+
+Wed Oct  3 02:23:37 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* error.c (exc_to_s, name_err_to_s, name_err_mesg_to_str): do not
+	  taint messages.
+
+Tue Oct  2 16:47:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (identity_hash_new): hide internal hashes for refinements.
+
+	* eval.c (rb_mod_refine): no default value.
+
+Mon Oct  1 22:54:02 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (identity_hash_new): new function to create a new identity
+	  hash.
+
+	* eval.c (rb_overlay_module, rb_mod_using, rb_mod_refine): use
+	  identity_hash_new().
+
+Mon Oct  1 02:34:53 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in (--with-opt-dir): Make this also work on DLDFLAGS
+	  so LIBRUBY_SO links fine with libexecinfo installed in a
+	  non-system directory.
+
+Sun Sep 30 23:32:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* vm_dump.c (rb_vm_bugreport): add /Library/Logs/DiagnosticReports
+	  in the list of locations of crash reports.
+
+Sun Sep 30 21:18:03 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_concat): use memcpy to copy a string which contains
+	  NUL characters. [ruby-core:47751] [Bug #7090]
+
+Sat Sep 29 19:41:53 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/envutil.rb (EnvUtil#invoke_ruby): kill child process
+	  before Timeout::Error is raised. rmdir of mktmpdir fails with
+	  EACCES if child process is alive on Windows.
+
+	* test/thread/test_queue.rb (TestQueue): increase timeout.
+	  This test takes long time on Windows XP.
+
+Sat Sep 29 19:41:33 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/net/http/test_http.rb (TestNetHTTP#test_proxy_address):
+	  clear environment variables. If http_proxy environment variable was
+	  set, the test failed.
+
+	* test/net/http/test_http.rb (TestNetHTTP#test_proxy_port): ditto.
+
+Sat Sep 29 19:41:11 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/drb/drbtest.rb (DRbCore#teardown):
+	  Use Process.kill :KILL on Windows because Process.kill :INT silently
+	  fails on Windows 7 and raises EINVAL on Windows XP for spawned
+	  process with new_pgroup: false.
+
+	* test/drb/drbtest.rb (DRbAry#teardown): ditto.
+
+Sat Sep 29 19:40:32 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_unicode_escape.rb (TestUnicodeEscape#test_basic):
+	  set script encoding to work with LANG=C. It would work on both
+	  Windows and Unix. Refix of r37051.
+
+Sat Sep 29 11:21:06 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (rb_vm_using_modules): use using_modules before
+	  klass to fix method lookup order, and use klass even if klass is
+	  not a module to make refinements in class_eval invoked on classes
+	  work.
+
+	* eval.c (rb_using_module): accept a class as the second argument.
+
+	* eval.c (rb_mod_using, f_using): raise a TypeError if the argument
+	  is not a module.
+
+	* test/ruby/test_refinement.rb: add new tests for the above changes.
+
+Sat Sep 29 02:18:57 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_unicode_escape.rb (TestUnicodeEscape#test_basic):
+	  Use ruby only on Windows since the test fails on Unix with LANG=C.
+	  [ruby-core:47709] [Bug #7076]
+
+Fri Sep 28 22:19:31 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_unicode_escape.rb (TestUnicodeEscape#test_basic):
+	  echo command doesn't work properly against non-ascii character on
+	  Windows with chcp 437. Instead we use ruby.
+	  [ruby-core:47709] [Bug #7076]
+
+Fri Sep 28 17:54:31 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_setup_method): refactoring.
+	  Remove src_argc and use iseq->arg_size directly.
+
+Fri Sep 28 17:26:27 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/installer.rb (check_that_user_bin_dir_is_in_path):
+	  test_generate_bin_bindir_with_user_install_warning(TestGemInstaller)
+	  fails on Windows with msys bash. It makes comparing paths
+	  case-insensitive.
+	  pick from upstream to fix a failure of test-all [ruby-core:47711]
+	 https://github.com/rubygems/rubygems/commit/c474edb2f3704206f04da1c8c6cf9fb079d84abe
+
+Fri Sep 28 15:44:45 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (search_method): copy refinement iclasses to search
+	  superclasses correctly.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Fri Sep 28 15:15:41 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (opt_checkenv): remove unused instruction `opt_checkenv'.
+
+	* compile.c (iseq_compile_each): ditto.
+
+	* node.h: remove unused node `NODE_OPTBLOCK'.
+
+	* ext/objspace/objspace.c, gc.c (gc_mark_children): ditto.
+
+Fri Sep 28 13:14:34 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: now VM_DEBUG_BP_CHECK should be 1.
+
+Fri Sep 28 12:51:54 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove rb_control_frame_t::bp (bp: base pointer).
+	  `bp' can be calculate by `sp' (stack pointer) of previous frame.
+	  Now, `bp_check' field is remained for debug.  You can eliminate
+	  this field by setting  VM_DEBUG_BP_CHECK as 0.
+
+	* vm_insnhelper.c (vm_base_ptr): add `vm_base_ptr(cfp).
+	  This function calculates base pointer from cfp.
+
+	* vm_insnhelper.c (vm_setup_method): push `recv' value on top of
+	  value stack (before method parameters).
+	  This change is for keeping consistency with normal method dispatch.
+
+	* insns.def: fix to use vm_base_ptr().
+
+	* vm.c (vm_exec): ditto.
+
+	* vm_dump.c: remove `bp' related dumps.
+
+	* cont.c (fiber_init): fix to check VM_DEBUG_BP_CHECK.
+
+Fri Sep 28 10:40:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_reopen): accept File::Constants as well as mode string.
+	  based on the patch by Glass_saga (Masaki Matsushita) in
+	  [ruby-core:47694].  [Feature #7067]
+
+Thu Sep 27 18:36:51 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_overlay_module, rb_mod_refine): accept a module as the
+	  argument of Module#refine.
+
+	* vm_method.c (search_method): if klass is an iclass, lookup the
+	  original module of the iclass in omod in order to allow
+	  refinements of modules.
+
+	* test/ruby/test_refinement.rb: add tests for the above changes.
+
+Thu Sep 27 18:12:20 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/syslog/lib/syslog/logger.rb: add a formatter to the
+	  Syslog::Logger object. [Bug #7065]
+	* test/syslog/test_syslog_logger.rb: ditto.
+
+Wed Sep 26 16:39:57 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: add new instruction `opt_empty_p' for optimize `empty?'
+	  method.  Apply a patch proposed at [ruby-dev:46120]
+	  [ruby-trunk - Feature #6972] by Glass_saga (Masaki Matsushita).
+
+	* compile.c (iseq_specialized_instruction), vm.c, vm_insnhelper.h:
+	  ditto.
+
+	* id.c, template/id.h.tmpl: ditto.
+
+	* test/ruby/test_optimization.rb: test for this changes.
+
+Tue Sep 25 09:59:26 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (invokesuper): klass in cfp is not valid in at_exit and
+	  END blocks.  [ruby-core:47680] [Bug #7064]
+
+Tue Sep 25 08:11:11 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* iseq.c (rb_iseq_defined_string): the index of defined_strings must
+	  be the value of type - 1.
+
+Mon Sep 24 17:36:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (defined_expr), insns.def (defined): share single frozen
+	  strings.  [EXPERIMENTAL] [ruby-core:47558][Feature #7035]
+
+	* iseq.c (rb_iseq_defined_string): make expression strings.
+
+Mon Sep 24 11:22:36 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: add --ticket option to add ticket number.
+
+Sun Sep 23 21:51:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (String#unspace): unescape with backslashes.  normal
+	  makes need to escape spaces with backslashes.  nmake is not the
+	  case.  [Bug #7036]
+
+	* lib/mkmf.rb (create_makefile): use timestamp file dependencies for
+	  directories.
+
+	* lib/mkmf.rb: unexpand macros.
+
+	* lib/mkmf.rb (LIBPATHFLAG): no needs to escape library path here.
+
+	* lib/mkmf.rb (MakeMakefile#configuration): make prefix paths
+	  internal to deal with in Makefile.
+
+	* lib/mkmf.rb (MakeMakefile#mkintpath): not a global function now.
+
+Sun Sep 23 02:33:37 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* complex.c: Fix examples of r36993.
+	  Keep the simple definition, mathematics define the result.
+	  Based on patch by Robin Dupret. Fixes #188 on github.
+
+Sat Sep 22 07:15:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/ripper/lib/ripper.rb:
+	  Match sample output to Ripper.sexp from current trunk version.
+	  [Bug #6929]
+
+Thu Sep 20 23:05:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (native_cond_initialize): destroy condattr
+	  after using it. Patch by Stanislav Sedov. Thank you.
+	  [Bug #7041] [ruby-core:47619]
+
+Thu Sep 20 22:53:02 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (native_cond_initialize): clean up #ifdef condition.
+
+Thu Sep 20 16:42:44 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/drb/ssl.rb (DRb::DRbSSLSocket::SSLConfig::DEFAULT): add
+	  SSLTmpDhCallback for configuration option.
+
+	* lib/drb/ssl.rb (setup_ssl_context): copy the value of tmp_dh_callback.
+
+	* test/drb/ut_array_drbssl.rb: set tmp_dh_callback to suppress warning.
+
+	* test/drb/ut_drb_drbssl.rb: ditto.
+
+Thu Sep 20 10:56:08 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/drb/ut_drb.rb: revert a part of r36987, and get rid of a warning
+	  with another method.  if the substitution is removed, the ExtSrv
+	  object will be GC'ed and some tests will be blocked.
+
+Thu Sep 20 07:20:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* complex.c: Examples for Complex Documentation.
+	  Patch by Robin Dupret.
+	  Fixes #184 on github.
+
+Thu Sep 20 07:15:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/ripper/lib/ripper.rb: Documentation for Ripper.
+	  +:void_stmt+ is meaningless
+	  [Bug #6929] [ruby-core:47507]
+
+Thu Sep 20 07:05:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* lib/csv.rb (Object#CSV, Array#to_csv, String#parse_csv):
+	  Examples and documentation for CSV.
+	  [Bug #6880] [ruby-core:47218]
+
+Thu Sep 20 00:42:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (take_items), enum.c (enum_zip): raise TypeError at
+	  non-enumerable objects, not NoMethodError.  [ruby-dev:46145]
+	  [Bug #7038]
+
+	* vm_eval.c (rb_check_block_call): check_funcall variant with block
+	  function.
+
+Tue Sep 18 17:51:29 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_attrs): add npn_select_db to
+	  suppress warning: instance variable @npn_select_cb not initialized
+
+Sun Sep 16 17:47:00 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* tool/change_maker.rb:  Update svn detection for subversion 1.7's
+	  single .svn directory.
+
+Sun Sep 16 11:39:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_set_read_length): if the read length equals to the buffer
+	  string size then nothing to do.  or ensure the string modifiable
+	  before setting the length only when the former is shorter.  based on
+	  the patch in [ruby-core:47541] by Hiroshi Shirosaki.
+	  [ruby-core:46586] [Bug #6764]
+
+Sun Sep 16 08:57:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (strict_warnflags): separate strict flags from
+	  warnflags only for core.  [ruby-dev:46105]
+
+Sun Sep 16 08:16:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .editorconfig: add.  [ruby-core:47548] [Feature #7030]
+
+Sat Sep 15 01:56:40 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Merge upstream: 50a383c84.
+	  [ruby-dev:46128] [Bug #7005]
+
+Sat Sep 15 00:20:04 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf.c (rb_nkf_convert): suppress warning.
+
+Fri Sep 14 04:05:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* array.c (rb_ary_diff, rb_ary_uniq):
+	  Enhance documentation for array uniqueness
+	  Based on a patch by Robin Dupret
+	  [Bug #6872] [ruby-core:47209]
+
+Fri Sep 14 03:30:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* array.c (rb_ary_select):
+	  Update documentation for Array#select
+	* enum.c (enum_find_all, enum_reject):
+	  Update documentation for Enumerable#find_all and Enumerable#reject
+	  Based on a patch by Jeff Saracco
+	  [Bug #6908] [ruby-core:47285] [Fixes #166 on github]
+
+Fri Sep 14 00:20:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* signal.c (rb_f_kill):
+	  Update documentation for Process.kill to reflect kill(2)
+	  Patch by Richo Healey
+
+Thu Sep 13 21:40:49 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* lib/securerandom.rb (SecureRandom.random_bytes):
+	  Use 64bit value as pointer for Windows x64 to fix SystemCallError.
+
+	* lib/securerandom.rb (SecureRandom.lastWin32ErrorMessage):
+	  Set proper encoding to avoid invalid byte sequence error.
+	  [ruby-core:47451] [Bug #6990]
+
+Thu Sep 13 11:20:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* lib/optparse.rb: Remove unreachable email address from documentation
+	  [Bug #6996] [ruby-core:47459]
+
+Thu Sep 13 11:20:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* lib/xmlrpc.rb: Documentation for XMLRPC
+	* lib/xmlrpc/datetime.rb: ditto.
+	* lib/xmlrpc/parser.rb: ditto.
+	* lib/xmlrpc/client.rb: ditto.
+	* lib/xmlrpc/utils.rb: ditto.
+	* lib/xmlrpc/README.rdoc: ditto.
+	* lib/xmlrpc/create.rb: ditto.
+	* lib/xmlrpc/base64.rb: ditto.
+	* lib/xmlrpc/config.rb: ditto.
+	* lib/xmlrpc/httpserver.rb: ditto.
+	* lib/xmlrpc/server.rb: ditto.
+	* lib/xmlrpc/marshal.rb: ditto.
+	* lib/xmlrpc/README.txt: ditto.
+	  [Bug #6909] [ruby-core:47286]
+
+Thu Sep 13 10:22:11 2012  Takashi Toyoshima  <toyoshim@gmail.com>
+
+	* configure.in: Don't use PIE on Haiku because loader support is not
+	  enough.
+
+Thu Sep 13 08:20:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* lib/shellwords.rb: Documentation for Shellwords.
+
+Thu Sep 13 08:00:00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/ripper/lib/ripper.rb: Documentation for Ripper.
+	* ext/ripper/lib/ripper/lexer.rb: ditto.
+	* ext/ripper/lib/ripper/sexp.rb: ditto.
+	* ext/ripper/lib/ripper/filter.rb: ditto.
+	* ext/ripper/lib/ripper/core.rb: ditto.
+	  [Bug #6929] [ruby-core:47309]
+
+Wed Sep 12 22:59:07 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_method_missing, vm_call_method): reuse arguments
+	  on the VM stack and get rid of ALLOCA.
+
+Wed Sep 12 22:45::00 2012  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/pathname/lib/pathname.rb: Documentation for Pathname.
+	* ext/pathname/pathname.c: ditto.
+	  [Bug #6947] [ruby-core:47354]
+
+Mon Sep 10 10:19:34 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/depend: fixed wrong change in a part of r34802.
+
+Sun Sep  9 22:02:50 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/socket/basicsocket.c (rsock_bsock_send):
+	  avoid unnecessary select() calls before doing I/O
+	  Patch by Eric Wong. [Feature #4538] [ruby-core:35586]
+	* ext/socket/init.c (rsock_s_recvfrom): ditto.
+	* ext/socket/init.c (rsock_s_accept): ditto.
+	* ext/socket/udpsocket.c (udp_send): ditto.
+	* io.c (io_fflush): ditto.
+	* io.c (io_binwrite): ditto.
+	* io.c (rb_io_syswrite): ditto.
+
+Mon Sep 10 01:38:51 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (nogvl_close, maygvl_close, nogvl_fclose, maygvl_fclose):
+	  suppress integer <-> pointer cast warnings.
+	  [Feature #4570] [ruby-core:35711]
+
+Mon Sep 10 01:36:00 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_close): notify fd close before releasing gvl.
+	* io.c (fptr_finalize): modify fptr->mode before releasing gvl.
+	  remove unnecessary rb_thread_fd_close().
+	  [Feature #4570] [ruby-core:35711]
+
+Mon Sep 10 00:16:34 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c: exec() requires to be single threaded also on Haiku.
+	  by Takashi Toyoshima <toyoshim@gmail.com>
+	  https://github.com/ruby/ruby/pull/178
+
+Sun Sep  9 21:21:15 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/thread.rb (Queue#pop): Fixed double registration issue when
+	  mutex.sleep is interrupted. [Bug #5258] [ruby-dev:44448]
+	* lib/thread.rb (SizedQueue#push): ditto.
+
+	* test/thread/test_queue.rb (test_sized_queue_and_wakeup,
+	test_queue_pop_interrupt, test_sized_queue_pop_interrupt,
+	test_sized_queue_push_interrupt): new tests.
+
+Sun Sep  9 20:20:31 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/sync.rb (Sync_m#sync_lock): Fixed wakeup/raise unsafe code.
+	  Patched by Masaki Matsushita. [Bug #5355] [ruby-dev:44521]
+
+	* test/thread/test_sync.rb (test_sync_lock_and_wakeup,
+	  test_sync_upgrade_and_wakeup, test_sync_lock_and_raise):
+	  new test.
+
+Sun Sep  9 18:39:46 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/intern.h (rb_thread_blocking_region): Added
+	  a comment of recommended alternative way.
+
+Sun Sep  9 18:37:05 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/sync.rb (Sync_m): Removed RCS_ID.
+
+Sun Sep  9 18:21:03 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (test_advise_pipe): new test to check
+	  io.advise() against anonymous io object don't make crash.
+	  made by Eric Wong. [Bug #6081] [ruby-core:42880]
+
+Sun Sep  9 16:47:12 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (nogvl_close, maygvl_close, nogvl_fclose, maygvl_fclose):
+	  new functions.
+	* io.c (fptr_finalize): release GVL if possible.
+	  Patched by Eric Wong. [Feature #4570] [ruby-core:35711]
+
+Sun Sep  9 16:08:48 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (io_bufread): removed unnecessary rb_thread_wait_fd().
+	  Patch by Eric Wong. [Bug #6629] [ruby-core:45789]
+	* io.c (rb_io_sysread): ditto.
+	* io.c (copy_stream_fallback_body): ditto.
+
+Sun Sep  9 15:21:52 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_lock): stop multiple threads use
+	  pthread_cond_timedwait() concurrently. [Bug #6278] [ruby-core:44275]
+
+Sat Sep  8 18:52:22 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (struct rb_classext_struct): move allocator function into
+	  rb_classext_t from ordinary method table.  [ruby-dev:46121]
+	  [Feature #6993]
+
+	* object.c (rb_obj_alloc): call allocator function directly.
+
+	* vm_method.c (rb_define_alloc_func, rb_undef_alloc_func)
+	  (rb_get_alloc_func): use allocator function in rb_classext_t.
+
+Fri Sep  7 01:21:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake), lib/mkmf.rb (have_framework): fix splitting
+	  options with an argument, not using NUL as special character.
+	  [ruby-core:47447] [Bug #6987]
+
+Thu Sep  6 14:49:49 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .gdbinit (rp): FLONUM support.
+
+	* include/ruby/ruby.h (ruby_special_consts): define FLONUM constants
+	  always, so that they are available from gdb.
+
+	* include/ruby/ruby.h (RB_FLOAT_TYPE_P): merge FLONUM and non-FLONUM
+	  versions.  inline TYPE() comparison and FLONUM_P() should be
+	  optimized away on non-FLONUM.
+
+Thu Sep  6 08:20:55 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 3.4.0 (r7762)
+	* test/minitest/*: ditto
+
+Wed Sep  5 19:20:53 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (rb_warn4S): renamed from rb_warn4(), because the case in
+	  r36911 takes a string.
+
+	* parse.y (rb_warn4S): use ripper_warnS() for ripper.
+
+	* parse.y (ripper_warnS): now it is used.
+
+Wed Sep  5 15:51:52 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (notifications): [experimental] update notification
+	  template.
+
+Wed Sep  5 15:21:12 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (rb_warn4): added as a rb_warn variant to warn with explicit
+	  source file name and line in parse.y.
+
+	* parse.y (warn_unused_var): use rb_warn4 to suppress warning on ripper.
+
+Wed Sep  5 13:30:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_make_pattern): names under recursive need to be single
+	  basenames to match for each name.  [ruby-core:47418] [Bug #6977]
+
+Tue Sep  4 20:55:17 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/envutil.rb (EnvUtil#invoke_ruby): show Timeout::Error
+	  instead of IOError if the timeout has expired.
+
+	* test/test_pstore.rb
+	  (PStoreTest#test_pstore_files_are_accessed_as_binary_files):
+	  increase timeout because this test is slow on Windows.
+	  [ruby-core:47402] [Bug #6965]
+
+Tue Sep  4 11:28:57 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* vm_eval.c (ruby_eval_string_from_file_protect): initializer
+	  element is not computable at load time.
+
+Tue Sep  4 07:48:35 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_asn1_rb:
+	  test/openssl/test_ssl_session.rb:
+	  test/openssl/test_x509name.rb:
+	  test/openssl/test_buffering.rb:
+	  test/openssl/test_x509cert.rb:
+	  test/openssl/test_ssl.rb: Refactor code that leads to warnings on
+	  Ruby CI.
+
+Tue Sep  4 07:02:56 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/utils.rb: Use DSS1 as DSA signature digest for all
+	  OpenSSL versions < 1.0.0.
+	  [Feature #6946] [ruby-core:47405]
+
+Mon Sep  3 21:22:37 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_float_value): suppress warnings.
+	  [ruby-core:47406][Bug #6971]
+
+Mon Sep  3 14:49:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/matrix.rb (Vector#magnitude): accumulate squares of absolute
+	  values to fix for complex vector.  [ruby-dev:46100] [Bug #6966]
+
+Mon Sep  3 10:09:36 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/extconf.rb: Detect OpenSSL_FIPS macro
+	  ext/openssl/ossl.c: Expose OpenSSL::OPENSSL_FIPS constant to
+	  indicate whether OpenSSL runs in FIPS mode.
+	  test/openssl/test_pkey_dh.rb: Generate 256 bit keys for
+	  non-FIPS installations to improve test performance (e.g. for
+	  rubyci).
+	  test/openssl/utils.rb: Replace DSS1 as certificate signature
+	  digest with SHA1 for FIPS installations when using DSA by
+	  introducing TestUtils::DSA_SIGNATURE_DIGEST.
+	  test/openssl/test_x509cert.rb:
+	  test/openssl/test_x509crl.rb:
+	  test/openssl/test_x509req.rb: Use DSA_SIGNATURE_DIGEST
+	  NEWS: Introduce OpenSSL::OPENSSL_FIPS
+
+	  These changes allow running the OpenSSL tests in FIPS mode
+	  while keeping a high performance for non-FIPS installations.
+	  Introduction of OpenSSL::OPENSSL_FIPS allows for applications
+	  to react to special requirements when using OpenSSL in FIPS mode.
+	  [Feature #6946] [ruby-core:47345]
+
+Sun Sep  2 21:46:28 2012  Martin Bosslet <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/utils.rb: Use a cached DH key instead of generating a
+	  new one each time.
+
+Sun Sep  2 05:41:28 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/ssl.rb (WEBrick::Config::SSL): add new key
+	  SSLTmpDhCallback to set SSLContext#tmp_dh_callback.
+
+	* lib/webrick/ssl.rb (WEBrick::GenericServer#setup_ssl_context):
+	  follow above.
+
+Sat Sep  1 18:50:50 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (#initialize_copy, #eql): Use instance_variable_get
+	  instead of instance_eval.
+
+Fri Aug 31 21:47:56 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/test/unit/test-unit.gemspec: Make test/unit default gem.
+	  [Feature #6875] [ruby-dev:46051]
+
+Fri Aug 31 18:35:02 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/extconf.rb: Check existence of OPENSSL_NPN_NEGOTIATED.
+	  ext/ossl_ssl.c: Support Next Protocol Negotiation. Protocols to be
+	  advertised by the server can be set in the SSLContext by using
+	  SSLContext#npn_protocols=, protocol selection on the client is
+	  supported by providing a selection callback with
+	  SSLContext#npn_select_cb. The protocol that was finally negotiated
+	  is available through SSL#npn_protocol.
+	  test/openssl/test_ssl.rb: Add tests for Next Protocol Negotiation.
+	  NEWS: add news about NPN support.
+	  [Feature #6503] [ruby-core:45272]
+
+Fri Aug 31 17:38:43 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#{each,reject!,select!}, SortedSet#each): Pass
+	  the original block through instead of creating one that only
+	  yields the passed argument.
+
+Fri Aug 31 16:23:20 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb: Introduce several new error classes where only
+	  ArgumentError and StandardError were used.  IPAddr::Error is
+	  their common ancestor class that inherits from ArgumentError for
+	  backward compatibility.  Submitted by Jon Daniel.  Fixes #173 on
+	  GitHub.
+
+Fri Aug 31 14:51:27 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/bigdecimal/test_bigdecimal.rb (TestBigDecimal#test_to_f): added
+	  for previous commit.
+
+Fri Aug 31 14:32:05 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): use self's sign to
+	  determine 0.0 and Inf's sign instead of internal double value's.
+	  Reported by phasis68 (Heesob Park) at [ruby-core:47381] [Bug #6955]
+
+Fri Aug 31 14:31:17 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/id.h.tmpl, tool/id2token.rb: make id.h independent from
+	  parse.h, and make parse.c dependent on it instead.
+
+Fri Aug 31 14:27:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): fix race conditions at install-ext.
+	  target files need to depend on destination directory timestamp
+	  files, not phony targets.
+
+Fri Aug 31 14:03:45 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_trace.c (clean_hooks): do not access freed memory.
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks): fix uninitialized state
+	  when no events is executed.
+
+Thu Aug 30 18:21:51 2012  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_close): call rb_last_status_clear.
+
+Thu Aug 30 16:17:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_to_f): check underflow since
+	  strtod() sets errno to ERANGE at underflow too.  [ruby-core:47342]
+	  [Bug #6944]
+
+Thu Aug 30 12:44:43 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#{<,>,<=,>=}): Define comparison operators as
+	  shorthand for the {proper_}{subset?,superset?} methods (finally).
+	  Given a push by Alexander E. Fischer.
+
+Thu Aug 30 09:21:01 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/ftp.rb (URI::FTP#initialize): raise InvalidURIError if "//"
+	  is not present [ruby-core:47344] [Bug #6945]
+
+Thu Aug 30 07:45:12 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* test/ruby/test_file_exhaustive.rb: fix test introduced in r36811 for
+	  posix environments where HOME is not defined.  [ruby-core:47322]
+
+Wed Aug 29 23:42:59 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_last_status_clear): declared.
+
+	* process.c (rb_last_status_clear): exported.
+	  (rb_f_system): call rb_last_status_clear.
+
+	* io.c (rb_f_backquote): call rb_last_status_clear.
+
+Wed Aug 29 22:01:15 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_f_system): check failures of waitpid.
+	  [ruby-talk:398687]
+
+Wed Aug 29 15:03:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBDIR_BASENAME): use configured libdir value to fix
+	  --enable-load-relative on systems where libdir is not default value,
+	  overridden in config.site files.  [ruby-core:47267] [Bug #6903]
+
+	* ruby.c (ruby_init_loadpath_safe): ditto.
+
+Wed Aug 29 14:34:41 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: SIZE_MAX is defined in stdint.h, so r36755 breaks
+	  32bit FreeBSD. [ruby-core:47360] [Bug #6948]
+
+Wed Aug 29 04:50:04 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/utils.rb
+	  test/openssl/test_pair.rb
+	  test/openssl/test_pkey_dh.rb: Use 1024 bit DH parameters to satisfy
+	  OpenSSL FIPS requirements. Patch by Vit Ondruch.
+	  [Bug #6938] [ruby-core:47326]
+
+Tue Aug 28 22:31:49 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* insns.def (checkmatch): suppress warnings. [ruby-core:47339]
+	  [Bug #6930]
+
+Tue Aug 28 20:03:54 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: Fixing Haiku R1/alpha3 build with gcc-4.4.4.
+	  - omit ANSI standard flags to compile socket extension where
+	    anonymous union is required.
+	  - remove redundant -be flags.
+	  by Takashi Toyoshima <toyoshim@gmail.com>
+	  https://github.com/ruby/ruby/pull/168
+
+Tue Aug 28 11:32:37 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* nacl/GNUmakefile.in (.rbconfig.time): r36828 was incomplete.
+	  It did not run correctly on clean build.
+
+Tue Aug 28 09:25:20 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (Makefile): make to depend on common.mk, to
+	  stop and force to re-run make process when common.mk is changed.
+
+Mon Aug 27 20:19:49 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/etc/test_etc.rb (TestEtc#test_getgrgid): fix for non unique GID.
+	  No unixen systems guarantee that GID is unique. Etc.getgrgid would
+	  not return the first entry in the order of Etc.group for shared GID.
+	  [ruby-core:47312] [Bug #6935]
+
+Mon Aug 27 18:19:36 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h (rb_float_value): optimize it.
+	  This technique was pointed by shinichiro.hamaji
+	  <http://shinh.skr.jp/m/?date=20120825#p02>.
+
+Mon Aug 27 15:08:25 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* common.mk (vm_trace.o): Added a missing dependency.
+
+Sun Aug 26 09:29:32 2012  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* nacl/GNUmakefile.in (package): make package should install
+	  example.html for nacl build
+
+	Patch by Takashi Toyoshima <toyoshim AT gmail.com>.
+
+Sun Aug 26 09:22:33 2012  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* nacl/GNUmakefile.in (CC, LD, NM, AR, AS, RANLIB, OBJDUMP, OBJCOPY)
+	  Rewrites these variables instead of PATH.
+	  NaCl port uses a toolchain which is specified by NACL_SDK_ROOT
+	  environment variable. Originally, NaCl build added the toolchain
+	  under the NACL_SDK_ROOT to the PATH. But updating PATH doesn't work
+	  on Mac.
+	  (RBCONFIG): Replaces configs with the variable updates above.
+
+	* configure.in: Thus it is no longer necessary to check $PATH.
+
+	Based on a patch by Takashi Toyoshima <toyoshim AT gmail.com>.
+
+Sun Aug 26 16:53:00 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (checkmatch): suppress warnings.  [ruby-core:47310]
+	  [Bug #6930]
+
+	* vm_core.h (VM_FRAME_TYPE_FINISH_P): ditto.
+
+Fri Aug 24 15:42:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): use timestamp for destination
+	  directories to make them before making or copying files there.
+	  [ruby-dev:46067] [Bug #6904]
+
+Fri Aug 24 12:40:15 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* configure.in (mingw): add shlwapi to the list of dependency
+	  libs for Windows.
+	* win32/Makefile.sub (EXTSOLIBS): ditto.
+
+	* internal.h: declare internal functions rb_w32_init_file,
+	  rb_file_expand_path_internal and rb_file_expand_path_fast.
+
+	* file.c (Init_File): invoke Windows initialization rb_w32_init_file
+
+	* win32/file.c (rb_file_load_path_internal): new function.
+	  Windows-specific implementation that replaces file_expand_path.
+	  [Bug #6836][ruby-core:46996]
+
+	* win32/file.c (rb_w32_init_file): new function. Initialize codepage
+	  cache for faster conversion encodings lookup.
+
+	* file.c (file_expand_path): rename to rb_file_expand_path_internal.
+	  Conditionally exclude from Windows.
+
+	* file.c (rb_file_expand_path_fast): new function. delegates to
+	  rb_file_expand_path_internal without performing a hit to the
+	  filesystem.
+
+	* file.c (file_expand_path_1): use rb_file_expand_path_internal without
+	  path expansion (used by require).
+	* file.c (rb_find_file_ext_safe): ditto.
+	* file.c (rb_find_file_safe): ditto.
+
+	* load.c (rb_get_expanded_load_path): use rb_file_expand_path_fast.
+	* load.c (rb_feature_provided): ditto.
+
+	* file.c (rb_file_expand_path): use rb_file_expand_path_internal with
+	  path expansion.
+	* file.c (rb_file_absolute_path): ditto.
+
+	* test/ruby/test_file_exhaustive.rb: new tests to exercise
+	  rb_file_expand_path_internal implementation and compliance with
+	  existing behaviors.
+
+Fri Aug 24 07:35:24 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http/backward.rb (class Net):  Restored Net::HTTPSession to
+	  fix backwards-compatibility with ancient Net::HTTP.  [Bug #6889]
+
+Thu Aug 23 20:58:55 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* common.mk: support `make id.h` without `rm .id.h.time` after
+	  `rm id.h`.
+
+Thu Aug 23 20:48:45 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_fixnum.rb (TestFixnum#test_singleton_method): new test.
+
+	* test/ruby/test_bignum.rb (TestBignum#test_singleton_method): ditto.
+
+	* test/ruby/test_float.rb (TestFloat#test_singleton_method): ditto.
+
+	* test/ruby/test_symbol.rb (TestSymbol#test_singleton_method): ditto.
+
+Thu Aug 23 20:34:32 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* class.c (singleton_class_of): flonum can't have singleton class.
+
+	* vm.c (vm_define_method): flonum can't have singleton method.
+
+Thu Aug 23 19:18:33 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (win32/*): macro RUBY_H_INCLUDES is not defined there,
+	  so need to move dependency rules under the definition of it.
+
+Thu Aug 23 19:16:20 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub: refactoring.  remove unused rules, and update
+	  some rules which are not used usually to fit current macros.
+
+Thu Aug 23 16:46:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_find_file_ext_safe, rb_find_file_safe): default to
+	  US-ASCII for encdb and transdb.
+
+	* load.c (search_required): keep encoding of feature name.  set
+	  loading path to filesystem encoding.  [Bug #6377][ruby-core:44750]
+
+	* ruby.c (add_modules, require_libraries): assume default external
+	  encoding as well as ARGV.
+
+Thu Aug 23 16:20:04 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: introduce flonum technique for
+	  64bit CPU environment (sizeof(double) == sizeof(VALUE)).
+	  flonum technique enables to avoid double object creation
+	  if the double value d is in range about between
+	  1.72723e-77 < |d| <= 1.15792e+77 or 0.0.
+	  flonum Float value is immediate and their lowest two bits
+	  are b10.
+	  If flonum is activated, then USE_FLONUM macro is 1.
+	  I'll write detailed in this technique on
+	  https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/Flonum_tech
+
+	* benchmark/bmx_temp.rb: add an benchmark for simple
+	  Float calculation.
+
+	* gc.c (id2ref, rb_obj_id): add flonum Float support.
+
+	* include/ruby/intern.h: move decl of rb_float_new(double)
+	  to include/ruby/ruby.h.
+
+	* insns.def, vm.c, vm_insnhelper.c: add flonum optimization
+	  and simplify source code.
+
+	* vm_insnhelper.h (FLONUM_2_P): added.
+
+	* marshal.c: support flonum output.
+
+	* numeric.c (rb_float_new_in_heap): added.
+
+	* parse.y: support flonum.
+
+	* random.c: ditto.
+
+Thu Aug 23 16:12:40 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): add dependency to header files when
+	  depend files don't exist.  now we can remove simple (and often
+	  wrong) depend files in most cases.
+
+Thu Aug 23 16:02:20 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/date/depend: add dependency to $(ruby_headers).
+
+Thu Aug 23 12:51:39 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* insns.def (invokesuper): reverted r36640 partially to make super
+	  in a thread work correctly. [ruby-core:47284] [Bug #6907]
+
+	* test/ruby/test_super.rb: related test.
+
+Thu Aug 23 12:30:20 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/configure.bat: support --with(out)?-ext(ensions) options.
+
+Thu Aug 23 11:52:04 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: Fixing Haiku build.
+	  - -lbe is not required for linking
+	  - stack protector doesn't work for now because of the default gcc's
+	    bug
+	  by Takashi Toyoshima <toyoshim@gmail.com>
+	  https://github.com/ruby/ruby/pull/167
+
+	* signal.c (ruby_signal): haiku doesn't have SIGBUS.
+
+Thu Aug 23 11:32:44 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/open-uri/test_open-uri.rb (TestOpenURI#test_read_timeout): this
+	  test expects that the server thread will be killed in sleep, but 0.01
+	  sec is too short to reach there.
+
+Thu Aug 23 10:49:28 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: use the value of --with-opt-dir on building ruby
+	  itself. [ruby-dev:46064] [Bug #6900]
+
+Thu Aug 23 10:36:35 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (ID_H_TARGET): revert a part of r36724 and r36751.  they
+	  break mswin build from clean source.
+
+Thu Aug 23 02:37:35 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/syck: removed. Fixes [ruby-core:43360]
+
+	* test/syck: removed.
+
+	* lib/yaml.rb: only require psych, show a warning if people try to set
+	  the engine to syck.
+
+Thu Aug 23 01:46:53 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* insns.def: search up the cf stack for an object that is an instance
+	  of the recipient class.  Fixes [ruby-core:47186]
+
+	* test/ruby/test_super.rb: related test.
+
+Wed Aug 22 19:46:24 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: [ruby-core:47266].
+
+Wed Aug 22 19:41:19 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: [ruby-core:47226].
+
+Wed Aug 22 16:57:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): extract least ruby headers list as
+	  ruby_headers, so depend files can use default dependency
+	  explicitly.
+
+Wed Aug 22 15:27:50 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_setup_method): fix last commit of
+	  vm_insnhelper.c (r36771). [ruby-dev:46065] [Bug #6901]
+	  Should not disable tail call opt on FINISH_FRAME.
+	  This flag should be propagated correctly.
+
+Wed Aug 22 14:05:23 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: support TracePoint. [ruby-trunk - Feature #6895]
+
+	* test/ruby/test_settracefunc.rb: add tests for above.
+
+	* proc.c (rb_binding_new_with_cfp): add an internal function.
+
+	* vm.c (rb_vm_control_frame_id_and_class): add an internal function.
+
+	* vm_trace.c: add rb_add_event_hook2() and rb_thread_add_event_hook2().
+	  Give us the good name for them!
+
+Wed Aug 22 11:38:16 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (before_script): Turned out that make -j is broken.
+
+Wed Aug 22 11:23:35 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_setup_method): should not enable tail call
+	  optimization for frames with VM_FRAME_FLAG_FINISH.
+	  [ruby-dev:46065] [Bug #6901]
+
+Wed Aug 22 11:20:47 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/test_case.rb: run test with psych if exist.
+
+Thu Aug 16 12:09:51 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* nacl/pepper_main.c (init_loadpath): Pushes the correct load path on
+	  other architectures than x86_64. Fixes #6873.
+
+Wed Aug 15 19:37:33 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* configure.in (ac_cv_func_shutdown): shutdown(2) has a dummy
+	  implementation but has no declaration and does not work in
+	  NativeClient SDK pepper_20.
+
+Wed Aug 15 19:29:29 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* common.mk (vm_backtrace.o): Added missing dependencies.
+
+	* ext/nkf/depend (nkf.o): ditto.
+
+	* ext/ripper/depend (ripper.o) ditto.
+
+Wed Aug 22 07:27:00 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi/util.rb (CGI.escapeHTML): use &#39;
+	  [ruby-core:47221] [Bug #6861]
+
+Tue Aug 21 21:59:22 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/observer.rb: fix typo. https://github.com/ruby/ruby/pull/162 by
+	  unsymbol (Philip Cunningham).
+
+Tue Aug 21 20:30:06 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/fileutils/test_fileutils.rb (TestFileUtils#teardown):
+	  do not assume cwd is TMPROOT and never remove current directory.
+	  [ruby-core:47224][Bug #6884]
+
+Tue Aug 21 17:29:56 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* addr2line.c (fill_lines): need check and cast of the file size of
+	  target binary because there are some platforms which off_t > size_t.
+
+Tue Aug 21 17:07:58 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (compiler): [experimental] clang support.
+
+Tue Aug 21 15:44:27 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/lib/dl/func.rb (DL::Function#bind): fixes an error in
+	  test/dl/test_import.rb (DL::TestImport#test_carried_function)
+	  introduced by r36718.
+	  the instance of the anonymous class which wraps the block should have
+	  same methods and instance variables of self.
+
+Tue Aug 21 14:29:22 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (scriptbin.mk): no need to include twice.
+
+Tue Aug 21 10:52:08 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/unit/test.rb (Test::Unit::ProxyError): new exception class to
+	  wrap exceptions raised in workers in parallel test mode.
+
+	* test/unit/parallel.rb (Test::Unit::Worker#puke): use above wrapper
+	  exception.
+	  [Bug #6882] [ruby-dev:46054]
+
+Tue Aug 21 10:40:06 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test_continuation.rb (tracing_with_thread_set_trace_func):
+	  fix to use Thread#set_trace_func(nil), not set_trace_func(nil).
+
+Tue Aug 21 09:32:41 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 3.3.0 (r7676)
+	* test/minitest/*: ditto
+
+Tue Aug 21 09:05:32 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/testunit/tests_for_parallel/ptest_forth.rb: added a test case
+	  which causes an error.
+
+	* test/testunit/test_parallel.rb: follow above change.
+	  see [Bug #6882]
+
+Tue Aug 21 05:43:00 2012  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb: Fixes #161 on github
+	* lib/csv.rb: You can now specify a pattern for :skip_lines.
+	  Matching lines will not be passed to the CSV parser.
+	* lib/csv.rb: Patch by Christian Schwartz.
+
+Tue Aug 21 05:25:41 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* re.c (rb_reg_initialize_m): Forgot to update output for or'd-options
+	  example.
+
+Tue Aug 21 05:18:03 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* re.c (rb_reg_initialize_m): Update example to show that regexp
+	  options use | an not || to avoid confusion.
+
+Mon Aug 20 23:02:27 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y: more descriptive token names in syntax error messages.
+
+Mon Aug 20 20:36:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): follow iclasses as klass in cfp
+	  but not included modules.  [ruby-core:47241] [Bug #6891]
+
+	* vm_insnhelper.c (vm_call_bmethod): pass defined_class to follow
+	  proper ancestors.  [ruby-core:47241] [Bug #6891]
+
+Mon Aug 20 11:40:27 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* common.mk: fix failed to make with -j2.
+	  https://gist.github.com/3397935
+
+Mon Aug 20 10:51:01 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb, lib/test/unit/parallel.rb:
+	  generate error message (String) in parallel.rb instead of
+	  marshalling Exception. Fixes [Bug #6882] [ruby-dev:46054]
+
+Sun Aug 19 01:24:32 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* enum.c: fix docs. https://github.com/ruby/ruby/pull/129 by
+	  richardkmichael (Richard Michael).
+
+Sun Aug 19 00:47:26 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/fileutils.rb: fix typo.
+	  https://github.com/ruby/ruby/pull/155 by simonc (Simon COURTOIS).
+
+Sat Aug 18 09:57:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend: fix inplace-build condition.  enc.mk is generated with
+	  setting $srcdir to enc, but pwd is still top build directory.
+	  [ruby-core:47236] [Bug #6888]
+
+Fri Aug 17 23:28:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_any_to_s, rb_obj_inspect): preserve encodings of class
+	  name and instance variable names.
+
+Fri Aug 17 12:39:33 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/lib/dl/func.rb (DL::Function#bind): allow to return/break from
+	  the callback method. (Fiddle already allows it.)
+	  [Bug #6389] [ruby-dev:45604]
+
+Thu Aug 16 19:54:24 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c, vm_core.h: simplify tracing mechanism.
+
+	  (1) add rb_hook_list_t data structure which includes
+	      hooks, events (flag) and `need_clean' flag.
+	      If the last flag is true, then clean the hooks list.
+	      In other words, deleted hooks are contained by `hooks'.
+	      Cleanup process should run before traversing the list.
+	  (2) Change check mechanism
+	      See EXEC_EVENT_HOOK() in vm_core.h.
+	  (3) Add `raw' hooks APIs
+	      Normal hooks are guarded from exception by rb_protect().
+	      However, this protection is overhead for too simple
+	      functions which never cause exceptions.  `raw' hooks
+	      are executed without protection and faster.
+	      Now, we only provide registration APIs.  All `raw'
+	      hooks are kicked under protection (same as normal hooks).
+
+	* include/ruby/ruby.h: remove internal data definition and
+	  macros.
+
+	* internal.h (ruby_suppress_tracing), vm_trace.c: rename
+	  ruby_suppress_tracing() to rb_suppress_tracing()
+	  and remove unused function parameter.
+
+	* parse.y: fix to use renamed rb_suppress_tracing().
+
+	* thread.c (thread_create_core): no need to set RUBY_VM_VM.
+
+	* vm.c (mark_event_hooks): move definition to vm_trace.c.
+
+	* vm.c (ruby_vm_event_flags): add a global variable.
+	  This global variable represents all of Threads and VM's
+	  event masks (T1#events | T2#events | ... | VM#events).
+	  You can check the possibility kick trace func or not
+	  with ruby_vm_event_flags.
+	  ruby_vm_event_flags is maintained by vm_trace.c.
+
+	* cont.c (fiber_switch, rb_cont_call): restore tracing status.
+	  [Feature #4347]
+
+	* test/ruby/test_continuation.rb: ditto.
+
+Thu Aug 16 19:15:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_class_initialize): forbid inheriting uninitialized
+	  class.  another class tree not based on BasicObject cannot exist.
+	  [ruby-core:47148][Bug #6863]
+
+Thu Aug 16 11:52:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/-ext-/test_printf.rb (Test_SPrintf#test_{taint,untrust}): use
+	  plain object so that the results of to_s and inspect are infected.
+	  [ruby-dev:46053] [Bug #6881]
+
+Thu Aug 16 09:46:07 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c: remove unnecessary macros to check traditional C.
+	  https://github.com/ruby/ruby/pull/46 by lateau (Daehyub Kim).
+
+	* vsnprintf.c: remove K&R.
+
+Wed Aug 15 20:47:49 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* object.c (rb_obj_inspect): Kernel#inspect: do not call #to_s. A class
+	  can now benefit from the nice default #inspect even if it defines #to_s.
+	  Also, there is no more unexpected change in #inspect result.
+
+	* NEWS: Add note about the change.
+
+	* bignum.c, io.c, numeric.c, object.c, proc.c, vm.c (Init_*):
+	  Adapt internal structures (by aliasing #inspect to #to_s) so they
+	  don't rely on the removed behavior (#inspect calling overridden #to_s).
+
+	* test/ruby/test_object.rb (test_inspect): add tests for Kernel#inspect.
+
+	* lib/pp.rb (class PP): do not call #to_s anymore, as #inspect
+	  no more does (mame).
+
+	* test/test_pp.rb (class PPInspectTest): remove related assertion (mame).
+	  [ruby-core:43238][Feature #6130]
+
+	* test/drb/drbtest.rb (DRbCore#teardown, DRbAry#teardown):
+	  adapt DRb tests with the new change (shirosaki).
+	  [ruby-core:47182][Bug #6866]
+
+Wed Aug 15 18:05:37 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#failed): need to delete the
+	  status line if the status is skipped and -q is specified.
+
+Wed Aug 15 16:26:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (ruby__sfvextra): the result should be infected by the
+	  given strings.
+
+	* sprintf.c (ruby__sfvwrite): set buffer length and exclude
+	  uninitialized garbage to get correct coderange.
+
+Wed Aug 15 16:20:09 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ID_H_TARGET): make timestamp file of id.h so that the
+	  header will not be remade repetitively.
+
+Wed Aug 15 11:39:53 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: separate trace_func related functions from
+	  thread.c.
+
+	* thread.c: ditto.
+
+	* common.mk: add vm_trace.o.
+
+	* inits.c: call Init_vm_trace().
+
+Tue Aug 14 16:25:46 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/erb/test_erb.rb (test_html_escape): add assertions for the
+	  cases where the argument is not a String.
+
+Tue Aug 14 16:03:31 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (check_valid_dir): reject "..." as directory name.
+	  [Bug #6851]
+
+Tue Aug 14 16:02:51 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file_exhaustive.rb
+	  (TestFileExhaustive#test_stat_dotted_prefix): added.
+
+Tue Aug 14 15:39:09 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file_exhaustive.rb
+	  (TestFileExhaustive#test_stat_drive_root): added.
+
+Tue Aug 14 10:38:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/erb.rb (ERB::Util.html_escape): fix r36687: call to_s before
+	  passing it to CGI.escapeHTML.
+
+Mon Aug 13 13:13:19 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/erb.rb (ERB::Util.html_escape): use CGI.escapeHTML to escape
+	  single quotes. [ruby-core:47138] [Bug #6861]
+
+Sun Aug 12 11:57:20 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (invoke_block_from_c): fix unintentional block passing.
+	  [ruby-dev:45071] [Bug #5832]
+
+Fri Aug 10 08:41:28 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* gc.c (gc_malloc_allocated_size):  RDoc does not process macros, so
+	  mention this method is only available when ruby is built with
+	  CALC_EXACT_MALLOC_SIZE
+	* gc.c (gc_malloc_allocations):  ditto
+
+Thu Aug  9 23:46:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkrunnable.rb: see build_os instead of target arch for
+	  cross-compiling.
+
+	* configure.in (MINIRUBY): use real path for include path.
+
+	* template/fake.rb.in (builddir): remove duplications
+
+Thu Aug  9 20:03:11 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_file_exhaustive.rb
+	  (TestFileExhaustive#test_stat_special_file): add a test.
+	  GetFileAttributesExW fails to get attributes of special files
+	  such as pagefile.sys.
+
+	* win32/win32.c (check_valid_dir): for performance, check the path
+	  by FindFirstFileW only if the path contains "..."
+
+	* win32/win32.c (winnt_stat): use GetFileAttributesExW instead of
+	  FindFirstFileW since GetFileAttributesExW is faster.
+	  Based on the patch by Dusan D. Majkic.
+	  [ruby-core:47083] [Feature #6845]
+
+Thu Aug  9 18:33:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): show version only once even if -v and
+	  --version are given together.
+	  http://twitter.com/d6rkaiz/status/233491797085671424
+
+Thu Aug  9 12:37:22 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/openssl/test_config.rb (OpenSSL#test_constants): skip this
+	  test if platform is Mac OS X or Windows. [Bug #6830]
+
+Wed Aug  8 22:51:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (eval_under): singletons other than special constants
+	  don't need cref-scope hack.
+
+Wed Aug  8 22:45:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (.y.h): split from .y.c rule to manage dependency on
+	  parse.h.  [ruby-core:46741] [Bug #6789]
+
+	* common.mk (id.h): keep old file unless changed.
+
+Wed Aug  8 17:11:20 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (ADD_INSNL): make ADD_INSNL as alias of ADD_INSN1.
+
+Wed Aug  8 17:08:14 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstrap/test_exception.rb: fix a last committed test.
+
+Wed Aug  8 16:27:58 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def (checkmatch):
+	  remove checkincludearray instruction and
+	  add new instruction checkmatch.
+	  This change is to solve
+	  [Bug #4438] "rescue args type check omitted".
+
+	* iseq.c: increment ISEQ_MAJOR_VERSION because removal of
+	  checkincludearray instruction.
+
+	* vm_core.h: add several definitions for
+	  the checkmatch instruction.
+
+	* vm_insnhelper.c (check_match): added.
+
+	* bootstraptest/test_exception.rb: add a test.
+
+	* test/ruby/test_exception.rb: ditto.
+
+Wed Aug  8 05:51:20 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* proc.c (method_clone):  Added documentation.  Patch by Robin Dupret.
+	  Fixes #152 on github.
+
+Tue Aug  7 20:19:29 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (Init_readline): rl_catch_signals=0 returns
+	  back. Without this, on FreeBSD9 and readline 6.2 irb can't catch ^C.
+	  [Bug #5423]
+
+Tue Aug  7 20:12:39 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_exec.c, insns.def (leave): solve problems on
+	  OPT_CALL_THREADED_CODE.
+	  Catch up finish frame structure on OPT_CALL_THREADED_CODE.
+
+	* vm_core.h: add rb_thread_t#retval for temporary space on
+	  OPT_CALL_THREADED_CODE.
+
+	* vm.c (th_init): clear rb_thread_t#retval as Qundef.
+
+	* vm_dump.c (rb_vmdebug_debug_print_pre): fix debug print format.
+
+Tue Aug  7 11:58:27 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_require.rb (TestRequire#test_require_twice): added.
+
+Tue Aug  7 11:35:37 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_redefine_opt_method): use RCLASS_ORIGIN to avoid
+	  SEGV when a module-prepended class is refined.
+
+Tue Aug  7 10:46:37 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file_exhaustive.rb
+	  (TestFileExhaustive#test_expand_path*): refactoring.  split the method
+	  into some chunks of the same kind of tests.
+
+Tue Aug  7 00:31:09 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_special_singleton_class_of): utility function.
+
+	* vm_eval.c (eval_under): special deal for class variable scope with
+	  instance_eval.
+
+	* vm_eval.c (rb_obj_instance_eval, rb_obj_instance_exec): allow method
+	  definition in instance_eval of special constants.  [ruby-core:28324]
+	  [Bug #2788]
+
+Tue Aug  7 00:23:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (CVAR_LOOKUP): split into helper functions.
+
+Mon Aug  6 19:15:11 2012  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_variant.rb: setting WIN32OLE.locale
+	  to pass some assertion.  Thanks to Hiroshi Shirosaki.
+	  [ruby-core:46873][Bug #6814]
+
+Mon Aug  6 15:54:50 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* internal.h, class.c, eval.c, insns.def: find the appropriate
+	  receiver for super called in instance_eval.  If such a receiver is
+	  not found, raise NoMethodError. [ruby-dev:39772] [Bug #2402]
+
+Mon Aug  6 14:54:38 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* include/ruby/ruby.h, eval.c, vm_insnhelper.c: fix typo.
+
+Mon Aug  6 13:13:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call_super): since cfp->klass is always class or
+	  iclass, no search from method entry.
+
+	* insns.def (defined): now should use klass in the current control
+	  frame to search superclass, not me->klass.  reported by naruse.
+
+Mon Aug  6 11:19:19 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/etc/test_etc.rb (TestEtc#test_getpwuid): `s' is never set to nil.
+
+Mon Aug  6 11:08:48 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/syslog/test_syslog_logger.rb: skip unless Syslog module is
+	  available.
+
+Mon Aug  6 00:40:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_log): fix format specifier.
+
+Mon Aug  6 00:39:24 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (NUM2ULONG): optimize by inline as well as
+	  NUM2LONG, and cast to unsigned long explicitly for the platforms
+	  where SIZEOF_VALUE is larger than SIZEOF_LONG.
+
+	* include/ruby/ruby.h (NUM2SSIZET): fix type to cast.
+
+Sun Aug  5 21:10:36 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c : if ENABLE_VM_OBJSPACE is 1, rest_sweep is not defined.
+	  remove unused declarations. [ruby-core:47004] [Bug #6837]
+
+Sun Aug  5 19:31:57 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: just move functions and so on. I don't touch any internal
+	  implementation.
+
+Sun Aug  5 13:22:29 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: use gcc-4.2 prior to clang, gcc, and cc if exist for
+	  the use of Snow Leopard's old clang. see also r36594, r36610, r36611.
+
+Sun Aug  5 06:55:10 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_{core,strftime}.c: [ruby-core:46990].
+
+Sat Aug  4 22:56:20 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: use inline functions instead of macros, and close up
+	  related codes for the profiler.
+
+Sat Aug  4 20:37:56 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_mark_children): use gc_mark_ptr instead of marking
+	  a object directly.
+
+Sat Aug  4 10:02:03 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_alias.rb (test_super_in_aliased_module_method):
+	  add a test case for [ruby-dev:46028], which fails in 1.8.
+
+Sat Aug  4 01:56:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_normal_superclass): no longer needs
+	  receiver, klass is always unique in the ancestors now.
+
+Sat Aug  4 01:27:40 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* insns.def (invokesuper): reverted r36612 so that super in an
+	  aliased method will not call the same method.
+
+Fri Aug  3 19:26:10 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* insns.def (invokesuper): don't skip the same class.  instead, use
+	  rb_method_entry_get_with_omod() to avoid infinite loop when
+	  super is used with refinements. [ruby-core:30450] [Bug #3351]
+
+Fri Aug  3 19:21:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: use clang prior to gcc only when self-compiling on
+	  darwin.  search default compilers on other platforms.  [Bug #6816]
+
+Fri Aug  3 17:25:49 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: move RUBY_MINGW32 after AC_PROG_CC.
+	  RUBY_MINGW32 uses AC_TRY_CPP and it sets CC and CPP. [Bug #6816]
+
+	* configure.in: don't use AC_PROG_CC in AS_CASE.
+
+Fri Aug  3 17:21:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/runner.rb: get rid of loading previously installed gems.
+	  [ruby-dev:46025]
+
+Fri Aug  3 16:40:01 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (notifications): [experimental] IRC notifications.
+
+Thu Aug  2 20:32:29 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_using): new method Module#using. [experimental]
+
+	* eval.c (rb_mod_refine): new method Module#refine. [experimental]
+
+	* eval.c (f_using): new method Kernel#using. [experimental]
+
+Thu Aug  2 20:08:02 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* class.c, insns.def, method.h, proc.c, vm.c, vm_core.h, vm_eval.c,
+	  vm_insnhelper.c, vm_insnhelper.h, vm_method.c: add klass to
+	  rb_control_frame_t to implement super correctly.
+
+Thu Aug  2 13:23:08 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (AC_PROG_CC): AC_PROG_CC tries clang at first on
+	  darwin. [Bug #6816]
+
+Thu Aug  2 11:39:25 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: return true or false. Patch by Dirkjan Bussink. [Bug #6821]
+
+	* test/ruby/test_gc.rb: add test-case for this bug.
+
+Thu Aug  2 10:51:12 2012  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/lib/openssl/digest.rb
+	  test/openssl/test_digest.rb: Add Digest module function to OpenSSL
+	  module and test it. Patch provided by Eric Hodel.
+	  [ruby-core:46908][Feature #6819]
+
+Wed Aug  1 22:29:12 2012  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/digest/digest.c (hexencode_str_new): return an ASCII string
+
+	* test/digest: tests for all kind of digests encodings
+	  [ruby-core:46792][Bug #6799]
+
+Wed Aug  1 05:50:53 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_encoding):
+	  Fix test_encoding failure on Windows.
+	  With chcp 65001, 1252 and 437, test_encoding failed. Test result
+	  depends on locale because LANG environment variable doesn't affect
+	  locale on Windows.
+	  [ruby-core:46872] [Bug #6813]
+
+Wed Aug  1 00:33:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (include_class_new): fix duplication of prepended module.
+	  since m_tbl of prepended module is always zero, copy from its
+	  copy iclass of original.
+
+Tue Jul 31 18:22:34 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (classname): tell if found name is permanent.  search
+	  tmp_classpath only if class id is set.  [ruby-core:42865][Bug #6078]
+
+	* variable.c (rb_class_path): duplicate found temporary path.
+
+	* variable.c (rb_set_class_path_string, rb_set_class_path): set class
+	  id to find classpath.
+
+Tue Jul 31 10:36:12 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: updated to released version.
+
+	* ext/psych/psych.gemspec: ditto
+
+Tue Jul 31 06:18:06 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* time.c (time_sec):  Remove extra wording about leap seconds and refer
+	  directly to Wikipedia's leap second page for further information.
+	  [Bug #6749]
+
+Mon Jul 30 23:01:47 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/platform.rb (Gem::Platform#initialize): Support pattern
+	  like x86_64-netbsd6.99.7.
+
+Mon Jul 30 21:00:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (find_class_path): no retry when preferred is given.
+
+	* variable.c (classname): if classid is set try it to find full
+	  qualified class path, and then try arbitrary class path.  try
+	  tmp_classpath at last even if enclosing namespace is anonymous.
+	  fix r36574.  [ruby-core:42865][Bug #6078]
+
+	* variable.c (rb_set_class_path_string, rb_set_class_path): set
+	  tmp_classpath instead of classpath if the name is not permanent.
+
+Mon Jul 30 14:24:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c: store anonymous class path in tmp_classpath but not in
+	  classpath.  [ruby-core:42865][Bug #6078]
+
+Mon Jul 30 13:11:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DLDFLAGS): on Darwin, deprecate -flat_namespace to get
+	  rid of huge imported symbols table.
+
+	* configure.in (LIBRUBY_RELATIVE): libruby_so is not made when
+	  disable-shared, so no absolute path is used for it and executable
+	  file is runnable anywhere.
+
+Mon Jul 30 01:30:10 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* common.mk: add a dependency. [ruby-core:46741] [Bug #6789]
+
+Sun Jul 29 15:44:47 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* thread.c (thread_create_core): hide th->async_errinfo_mask_stack from
+	  ObjectSpace.each_object. refix of r36539.
+
+Sun Jul 29 23:57:27 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/option.c (inet_ntop): use rb_w32_inet_ntop, instead of
+	  inet_ntop directly, which is unavailable on older version Windows.
+
+	* win32/win32.c (rb_w32_inet_ntop): type should be const.
+
+Sun Jul 29 14:20:34 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* thread.c (Init_Thread): does not need to set klass
+	  explicitly.
+
+Sun Jul 29 06:21:04 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* win32/win32.c: suppress warning redeclared on mingw64.
+	  *_s functions are declared if MINGW_HAS_SECURE_API is defined.
+	  Follow up r36556.
+
+Sun Jul 29 00:28:46 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: remove unused initialization.
+
+Sat Jul 28 16:26:09 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* win32/win32.c (gmtime_r): use _gmtime64_s() with x86_64-w64-mingw32.
+
+	* win32/win32.c (localtime_r): use _localtime64_s() with
+	  x86_64-w64-mingw32. Since FileTimeToSystemTime() seems not work with
+	  large value under x64. Mingw-w64 doesn't have these declaration.
+	  [ruby-core:46780] [Bug #6794]
+
+Fri Jul 27 18:25:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_check_io): make public.
+
+	* process.c (check_exec_redirect): try conversion to IO on redirect
+	  parameters.  [ruby-core:44181] [Bug #6269]
+
+Fri Jul 27 17:58:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CPPOUTFILE): get rid of variable conflict so
+	  CPPFLAGS is not duplicated.  [ruby-core:43097] [Bug #6119]
+
+Fri Jul 27 12:12:36 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/mkexports.rb: should not export DllMain().
+	  reported by luis at [ruby-core:46743] [Bug #6790], solved by
+	  Heesob Park, and confirmed by nobu.
+
+Thu Jul 26 14:51:29 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/net/http/test_https.rb (TestNetHTTPS#test_session_reuse):
+	  localhost is not (always) 127.0.0.1.  Don't expect that.
+
+Thu Jul 26 07:18:38 2012    <kanemoto@ruby-lang.org>
+
+	* ext/json/fbuffer/fbuffer.h: avoid compilation error on AIX by
+	   -ansi -std=iso9899:199409 (r36038). [ruby-core:46744] [Bug #6791].
+
+Thu Jul 26 00:42:23 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* thread.c (thread_create_core, Init_Thread): hide
+	  th->async_errinfo_queue and th->async_errinfo_mask_stack from
+	  ObjectSpace.each_object.
+
+Wed Jul 25 17:41:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c, rational.c: compatible marshal loader for compatibilities
+	  with 1.8.  [ruby-core:45775] [Bug #6625]
+
+Wed Jul 25 17:17:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* atomic.h: prefer GCC atomic builtins than Windows APIs, if possible,
+	  since they are generic.
+
+Wed Jul 25 11:16:57 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/.document:  Removed.  All files in net/ should be included in
+	  RDoc.
+
+Wed Jul 25 10:00:23 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/testunit/test_redefinition.rb: broken class/method names.
+
+Wed Jul 25 09:26:32 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi/html.rb: Use << instead of +=.
+	  `a += b` is syntax sugar of `a = a + b`; it creates a new string
+	  object. `a << b` is concatenation and doesn't create new object.
+
+Wed Jul 25 09:16:26 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi/html.rb (element_init): suppress redefine warning.
+	  Don't define methods if they are already defined.
+
+Wed Jul 25 09:05:38 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Added SSL session reuse across connections for a
+	  single instance to speed up connection.  [Feature #5341]
+	* NEWS:  ditto
+	* test/net/http/test_https.rb:  Tests for #5341
+
+Wed Jul 25 06:54:24 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/re.rdoc:  Fix spelling
+
+Wed Jul 25 06:49:12 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* re.c (rb_reg_s_last_match):  Update $~ to reference Regexp
+	  documentation about "special global variables".  [Bug #6723]
+
+Wed Jul 25 06:28:56 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* iseq.c:  Added documentation.  Patch by David Albert.  [Bug #6785]
+
+Wed Jul 25 03:05:06 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* parse.y: added symbols and qsymbols productions for %i and %I
+	  support. %i{ .. } returns a list of symbols without interpolation,
+	  %I{ .. } returns a list of symbols with interpolation.  Thanks to
+	  Josh Susser for inspiration of this feature. [Feature #4985]
+
+	* ext/ripper/eventids2.c: added ripper events for %i and %I.
+
+	* test/ripper/test_parser_events.rb: ripper tests
+
+	* test/ripper/test_scanner_events.rb: ditto
+
+	* test/ruby/test_array.rb: test for %i and %I behavior
+
+Tue Jul 24 23:34:43 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* include/ruby/win32.h (rb_w32_pow): add new function.
+	  We use powl() instead of broken pow() for x64-mingw32. This workaround
+	  fixes test failures related to floating point numeric.
+	  [ruby-core:46686] [Bug #6784]
+
+Tue Jul 24 15:01:24 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_socket, rb_w32_socketpair): remember the family
+	  in the high word of socklist value.
+
+	* win32/win32.c (overlapped_socket_io, recvmsg, sendmsg, setfl): follow
+	  above changes.
+
+	* win32/win32.c (rb_w32_getsockname): set remembered family to the
+	  argument when OS's function fails.
+
+Tue Jul 24 12:35:13 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_dir_m17n.rb: remove a garbage.
+
+	* test/ruby/test_dir_m17n.rb: convert from ascii-8bit to other encoding
+	  with 8bit bytes always fails.
+
+Tue Jul 24 12:32:18 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_dir_m17n.rb: sorry, typo.
+
+Tue Jul 24 12:13:26 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_dir_m17n.rb: refactoring. RE should be in the left side
+	  of the =~ operator, and compare the result with nil is meaningless.
+
+Tue Jul 24 11:35:20 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_pack.rb (test_pack_unpack_M): was redefined
+	  accidentally.
+
+Tue Jul 24 09:31:18 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Updated to RubyGems 1.8.24, a bugfix release.
+
+Tue Jul 24 08:30:15 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* test/ruby/test_dir_m17n.rb (create_and_check_raw_file_name): add new
+	  helper method to ease encoding testing. Patch by Oleg Sukhodolsky.
+	  [ruby-core:46589][Bug #6765]
+
+	* test/ruby/test_dir_m17n.rb (test_filename_extutf8): use filesystem
+	  encoding when reading entries and comparing.
+
+	* test/ruby/test_dir_m17n.rb (test_filename_utf8_raw_name): removed.
+
+	* test/ruby/test_dir_m17n.rb (test_filename_utf8_raw_jp_name): split test.
+
+Tue Jul 24 08:09:30 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* test/win32ole/test_win32ole_method.rb (is_ruby64?): Correct platform
+	  used to identify mingw-w64 (x64-mingw32). Patch by Hiroshi Shirosaki.
+	  [ruby-core:46651][Bug #6782]
+
+Tue Jul 24 07:22:58 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* time.c (time_sec):  Updated description of leap seconds for accuracy.
+	  Based on patch by Marcus Stollsteimer.  [Bug #6749]
+
+Tue Jul 24 07:03:11 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* string.c (rb_str_sub):  Fixed wording of documentation to match the
+	  replacement operation.  Minor cleanup of markup.  [Bug #6719]
+	* string.c (rb_str_sub_bang):  Minor wording change for clarity, minor
+	  cleanup of markup.
+
+Mon Jul 23 23:58:40 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/Makefile.in (TARGET_NAME, TARGET_ENTRY): needed for EXTDLDFLAGS
+	  on some platforms.  [ruby-core:46600] [Bug #6768]
+
+	* enc/depend: no longer needs tweaking DLDFLAGS for TARGET names.
+
+Mon Jul 23 22:48:19 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: use respond_to? to test Tempfile.
+	  [ruby-dev:45995] [Bug #6781] reported by hsbt (Hiroshi SHIBATA).
+
+Mon Jul 23 14:43:34 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBPATHENV): LIBPATH is used on AIX, but not
+	  SHLIB_PATH which was carelessly copied from HP/UX.  suggested by
+	  Perry Smith at [ruby-core:46397].  [Bug #6728]
+
+Mon Jul 23 01:55:08 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/uri/test_generic.rb (URI#test_find_proxy): add tests with
+	  empty *_proxy env variables.
+
+Mon Jul 23 01:47:26 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/uri/test_generic.rb (URI#with_env): unset proxy related env
+	  variables.  [Bug #6774]
+
+	* test/uri/test_generic.rb (URI#test_find_proxy): fix failures
+	  when proxy related env variables already set.  [Bug #6774]
+
+Sun Jul 22 23:58:48 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (rb_threadptr_execute_interrupts_common): increase
+	  running_time_us on THREAD_TO_KILL like on THREAD_RUNNABLE.
+	  This cause not to switch from a thread which is to be killed
+	  on FreeBSD and Mac OS X. see also the test.
+	  This issue maybe exist for long time but happens after r36430.
+
+Sat Jul 21 06:21:45 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: fixes for r36476. [Feature #6546]
+	 http://u64.rubyci.org/~chkbuild/ruby-trunk/log/20120720T030101Z.diff.html.gz
+
+	* lib/net/http.rb (Net::HTTP.newobj): return back for compatibility.
+
+	* lib/net/http.rb (Net::HTTP.new): set default_port if proxy port is
+	  not given.
+
+	* lib/net/http.rb (Net::HTTP#initialize): ditto.
+
+	* lib/net/http.rb (Net::HTTP#proxy?): return true or false.
+
+	* lib/net/http.rb (Net::HTTP#proxy_address): check proxy_uri is not nil.
+
+	* lib/net/http.rb (Net::HTTP#proxy_port): ditto.
+
+Sat Jul 21 23:12:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack): STACK_GROW_DIR_DETECTION is
+	  necessary on platforms with unknown stack direction.  [Bug #6761]
+
+Sat Jul 21 15:13:42 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/testcase.rb (method_added): refactoring.
+
+Sat Jul 21 14:06:41 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit/testcase.rb: warn when test_* method is redefined.
+	  Patch by mame (Yusuke Endoh). [Feature #2643] [ruby-core:27790]
+
+	* test/testunit/test_redefinition.rb: Test for above.
+
+	* test/testunit/test4test_redefinition.rb: Ditto.
+
+Sat Jul 21 08:41:14 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/logger.rb:  Updated example in Logger comment to match other
+	  examples and fixed a bug.  Patch by Marcus Stollsteimer.
+	  [Bug #6759]
+
+Fri Jul 20 17:20:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_random_real): refine error message.
+
+Fri Jul 20 11:03:17 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS:  Updated net/http for automatic proxy detection (#6546) and
+	  automatic gzip and deflate compression (#6492, #6494).
+
+Fri Jul 20 10:55:38 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Net::HTTP now automatically detects and uses
+	  proxies from the environment.  A proxy may also be specified as
+	  before.
+
+	  Net::HTTP::Proxy still creates anonymous classes, but these classes
+	  are only used to store configuration information.  When an HTTP
+	  instance is created the configuration is now copied.
+
+	  Additionally, Net::HTTP::ProxyDelta is no longer used by Net::HTTP
+
+	  [Feature #6546]
+	* lib/open-uri.rb:  Moved URI::Generic#find_proxy to uri/generic.
+	* lib/uri/generic.rb:  Imported find_proxy from open-uri.
+	* test/open-uri/test_open-uri.rb:  Moved proxy-discovery tests to URI.
+	* test/uri/test_generic.rb:  Imported proxy-discovery tests from
+	  open-uri.
+	* test/net/http/test_http.rb:  Added tests for proxy behavior.
+
+Fri Jul 20 09:34:11 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/socket/test_socket.rb:  Ignore IPv6 unique local addresses on OS
+	  X (iCloud Back to my Mac addresses) for test_udp_socket since they do
+	  not act as loopback addresses.  [Bug #6692]
+
+Fri Jul 20 09:32:14 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/socket/raddrinfo.c (addrinfo_ipv6_unique_local_p):  Added
+	  Addrinfo#ipv6_unique_local? to detect RFC 4193 unique local
+	  addresses.  Part of #6692
+	* ext/socket/rubysocket.h:  Add IN6_IS_ADDR_UNIQUE_LOCAL macro if
+	  missing.
+	* test/socket/test_addrinfo.rb:  Test for ipv6_unique_local?
+
+Fri Jul 20 07:40:32 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http/response.rb:  Automatically inflate gzip and
+	  deflate-encoded response bodies.  [Feature #6942]
+	* lib/net/http/generic_request.rb:  Automatically accept gzip and
+	  deflate content-encoding for requests.  [Feature #6494]
+	* lib/net/http/request.rb:  Updated documentation for #6494.
+	* lib/net/http.rb:  Updated documentation for #6492 and #6494, removed
+	  Content-Encoding handling now present in Net::HTTPResponse.
+	* test/net/http/test_httpresponse.rb:  Tests for #6492
+	* test/net/http/test_http_request.rb:  Tests for #6494
+	* test/open-uri/test_open-uri.rb (test_content_encoding):  Updated test
+	  for automatic content-encoding handling.
+
+Fri Jul 20 03:42:54 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c: use #ifdef, not #if.
+
+Thu Jul 19 15:08:40 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_s_control_interrupt,
+	            rb_thread_s_check_interrupt): added for
+	  Thread.control_interrupt and Thread.check_interrupt.
+	  See details on rdoc.
+	  I'll make an ticket for this feature.
+
+	* test/ruby/test_thread.rb: add a test for Thread.control_interrupt.
+
+	* thread.c (rb_threadptr_raise): make a new exception object
+	  even if argc is 0.
+
+	* thread.c (rb_thread_kill): kill thread immediately if target thread
+	  is current thread.
+
+	* vm_core.h (RUBY_VM_CHECK_INTS_BLOCKING): added.
+	  CHECK_INTS while/after blocking operation.
+
+	* vm_core.h (RUBY_VM_CHECK_INTS): require rb_thread_t ptr.
+
+	* cont.c (fiber_switch): use replaced RUBY_VM_CHECK_INTS().
+
+	* eval.c (ruby_cleanup): ditto.
+
+	* insns.def: ditto.
+
+	* process.c (rb_waitpid): ditto.
+
+	* vm_eval.c (vm_call0): ditto.
+
+	* vm_insnhelper.c (vm_call_method): ditto.
+
+Thu Jul 19 22:46:48 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb: remove temporally files early.
+
+Thu Jul 19 15:38:35 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* variable.c (rb_mod_class_variables): return inherited variables
+	  except when the optional argument is set to false.
+	  [ruby-dev:44034] [Bug #4971]
+
+	* variable.c (rb_mod_constants): fix typo in documentation.
+
+Thu Jul 19 14:30:43 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h: move mark function declarations that should be private.
+
+Thu Jul 19 14:18:22 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/init.c (rsock_init_sock): need to update max fd on all
+	  platforms.
+
+Thu Jul 19 14:15:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_gc_mark_threads): remove deprecated function.
+
+Thu Jul 19 13:28:03 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/net/http/test_http.rb (TestNetHTTPLocalBind#test_bind_to_local*):
+	  re-enable the tests because now it's OK on windows.
+
+Thu Jul 19 13:26:25 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: now enable IPv6 by default on mswin.
+
+Thu Jul 19 09:33:46 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/emitter.c (initialize): allow a configuration object to be
+	  passed to the constructor so that mutation isn't required after
+	  instantiation.
+
+	* ext/psych/lib/psych/handler.rb: add configuration object
+
+	* ext/psych/lib/psych/visitors/emitter.rb: use configuration object if
+	  extra configuration is present.
+
+Thu Jul 19 08:20:25 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file.rb: remove temporally files early.
+
+Thu Jul 19 07:37:41 2012  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/drbtest.rb: fixed: can't delete unix domain sockets problem.
+
+Thu Jul 19 03:41:20 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bignum.c: Added #include <strings.h> for ffs(). Patch by Perry
+	  Smith. Thank you. [Bug #6748]
+
+Thu Jul 19 01:56:02 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/intern.h (rb_num_zerodiv): Added NORETURN.
+	  Patched by Xi Wang. [Bug #6736]
+
+Wed Jul 18 23:57:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack): round down too long uuencode width.  folding
+	  width in uuencode format cannot be longer than 63 bytes.
+
+Wed Jul 18 23:04:18 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/dbm/dbm.c (fdbm_empty_p): fix wrong condition introduced in r36438.
+
+	* ext/sdbm/init.c (fsdbm_empty_p): ditto.
+
+Wed Jul 18 23:08:57 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_beginendblock.rb: remove temporally files early.
+
+Wed Jul 18 22:43:02 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_autoload.rb: remove temporally files early.
+
+Wed Jul 18 21:59:46 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_argf.rb: use temporally directory.
+
+Wed Jul 18 19:41:19 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/test_config.rb: remove temporally files early.
+
+Wed Jul 18 17:45:26 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_builtin_type_name): map by index.
+
+Wed Jul 18 16:17:40 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_framework): get rid of separating -framework
+	  option and its argument and dealing with the argument as a library
+	  or an object name.  if $LDFLAGS were an array...
+
+Wed Jul 18 16:09:10 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/curses/extconf.rb: support PDCurses.  patched by Luis Lavena.
+	  [ruby-core:46485] [Feature #6735]
+
+Wed Jul 18 15:50:25 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* parse.y (primary): allow an empty grouped expression as the
+	  operand of the not operator (e.g., not ()).
+	  [ruby-core:45976] [Bug #6674]
+
+	* parse.y (parser_yylex): show no warning for a grouped expression
+	  as the operand of the not operator (e.g., not (a)) or as an
+	  argument of a method call without parentheses (e.g., foo (a)).
+	  [ruby-core:39050] [Bug #5214]
+
+Wed Jul 18 15:33:21 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_thread_call_without_gvl2): added.
+	  it can skip last CHECK_INTS.  See document for more details.
+	  Document about it was updated a bit.
+
+	* include/ruby/thread.h (decl. of rb_thread_call_without_gvl2): added.
+
+	* thread.c (rb_thread_call_with_gvl): remove "EXPERIMENTAL!"
+	  warning from a document.
+
+Wed Jul 18 14:53:21 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (EXTDLDFLAGS): split options for each extension
+	  libraries, and unused in ruby.pc.  [Bug #6734]
+
+	* lib/mkmf.rb (MakeMakefile#configuration): add EXTDLDFLAGS.
+
+Wed Jul 18 14:47:23 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c: fix last commit miss.
+
+Wed Jul 18 14:16:51 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_threadptr_async_errinfo_*): manage async errors queue.
+	  Async events such as an exception throwed by Thread#raise,
+	  Thread#kill and thread termination (after main thread termination)
+	  will be queued to th->async_errinfo_queue.
+	  - clear: clear the queue.
+	  - enque: enque err object into queue.
+	  - deque: deque err object from queue.
+	  - active_p: return 1 if the queue should be checked.
+	  rb_thread_t#thrown_errinfo was removed.
+
+	* vm_core.h: add declarations of rb_threadptr_async_errinfo_*.
+	  remove rb_thread_t#thrown_errinfo field and
+	  add rb_thread_t#async_errinfo_queue (queue body: Array),
+	      rb_thread_t#async_errinfo_queue_checked (flag),
+	      rb_thread_t#async_errinfo_mask_stack(Array, not used yet).
+
+	* vm.c (rb_thread_mark): fix a mark function.
+
+	* cont.c (rb_fiber_start): enque an error.
+
+	* process.c (after_fork): clear async errinfo queue.
+
+Wed Jul 18 14:25:55 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* pack.c: (ditto) bitwise operations are not char.  Apply explicit
+	  casts on them.
+
+Wed Jul 18 12:59:50 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* encoding.c (load_encoding): explicit cast  to suppress  warning.
+	  Though the  cast truncates some bits, from  heuristic analysis I
+	  believe it is OK to do so here.
+
+	* bignum.c (rb_cstr_to_inum): ditto.
+
+Wed Jul 18 12:55:54 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/benchmark.rb: Fix Benchmark.benchmark output with an empty
+	  caption.  patched by Benoit Daloze. [ruby-core:45719] [Bug #6610]
+
+Wed Jul 18 10:00:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/debug.rb:  Added toplevel documentation.  Based on patch by Oscar
+	  Del Ben.  [Bug #6743], fixes #146 on github.
+
+Wed Jul 18 09:33:59 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/win32ole/test_win32ole_event.rb (TestWIN32OLE_EVENT): use
+	  standard skip method to skip tests.
+
+Wed Jul 18 09:26:45 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/logger.rb:  Updated typos and output to match modern Logger
+	  output.  Patch by Marcus Stollsteimer.  [Bug #6738]
+
+Wed Jul 18 07:59:29 2012  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/util.rb (CGI.escapeHTML,unescapeHTML): Add &apos; for HTML5
+	  escaping.
+	  [Feature #6620]
+
+Tue Jul 17 22:17:13 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: call io.close! for Tempfile.
+
+Tue Jul 17 16:41:32 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* proc.c (rb_proc_arity): return normal value (not -n-1) if it is not
+	  a labmda, or it is a labmda and no arg_opts. [Bug #5694]
+
+Tue Jul 17 03:56:34 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: strings with YAML anchors
+	  are properly referenced. Patched by Joe Rafaniello via Github:
+	    https://github.com/tenderlove/psych/pull/69
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ditto
+	* test/psych/test_alias_and_anchor.rb: test for change
+
+Mon Jul 16 23:20:24 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_float_cmp): use FIXNUM_MIN and FIXNUM_MAX,
+	  instead of LONG_MIN and LONG_MAX.
+
+Mon Jul 16 22:50:41 2012  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (flo_to_s): use the exponential form if the integer part
+	  is longer than or equal DBL_DIG.
+	  [ruby-dev:45960] [ruby-trunk - Bug #6741]
+
+Mon Jul 16 22:01:00 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/readline/readline.c: fixed docs. [Bug #6740][ruby-core:46501]
+	  patched by Nobuhiro IMAI.
+
+Mon Jul 16 19:24:01 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_float_eq): new function.
+	  (rb_big_eq): use rb_integer_float_eq.
+
+	* internal.h (rb_integer_float_eq): declared.
+
+	* numeric.c (flo_eq): use rb_integer_float_eq.
+	  (fix_equal): ditto.
+
+Mon Jul 16 19:02:31 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_float_cmp): rename a local variable.
+
+Mon Jul 16 18:40:26 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_float_cmp): renamed from rb_big_float_cmp.
+
+	* internal.h: follow the above change.
+
+	* numeric.c: ditto.
+
+Mon Jul 16 17:57:54 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_float_cmp): compare an integer and float precisely.
+	  [ruby-core:31376] [Bug #3589] reported by Tomasz Wegrzanowski.
+
+Mon Jul 16 17:29:45 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_float_cmp): support fixnum for argument x.
+
+	* numeric.c (fix_equal): use rb_big_float_cmp.
+	  (fix_cmp): ditto.
+	  (fix_gt): ditto.
+	  (fix_ge): ditto.
+	  (fix_lt): ditto.
+	  (fix_le): ditto.
+	  (flo_eq): ditto.
+	  (flo_cmp): use rb_big_float_cmp for fixnum argument.
+	  (flo_gt): ditto.
+	  (flo_ge): ditto.
+	  (flo_lt): ditto.
+	  (flo_le): ditto.
+
+Mon Jul 16 17:05:53 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/fileutils/test_fileutils.rb: add test for FileUtils#uptodate?
+
+Mon Jul 16 16:56:12 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/fileutils.rb (FileUtils.uptodate?): remove useless parameter.
+	  patched by Oscar Del Ben.[Bug #6708][ruby-core:46256]
+
+Mon Jul 16 15:37:56 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_eq): use rb_big_float_cmp.
+
+Mon Jul 16 15:00:45 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_big_float_cmp): declared.
+
+	* bignum.c (rb_big_float_cmp): extracted from rb_big_cmp and big_op.
+	  (rb_big_cmp): use rb_big_float_cmp.
+	  (big_op): ditto.
+
+	* numeric.c (flo_cmp): use rb_big_float_cmp.
+	  (flo_gt): ditto.
+	  (flo_ge): ditto.
+	  (flo_lt): ditto.
+	  (flo_le): ditto.
+
+Mon Jul 16 14:14:21 2012  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (enum big_op_t): new type.
+	  (big_op): use enum big_op_t.
+	  (big_gt): ditto.
+	  (big_ge): ditto.
+	  (big_lt): ditto.
+	  (big_le): ditto.
+
+Sat Jul 14 18:18:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_get_values_at): fill with nil out of range.
+	  [ruby-core:43678] [Bug #6203]
+
+Sat Jul 14 17:17:55 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* cont.c (cont_restore_0): improve docs. [Bug #6706][ruby-core:46243]
+	  patched by Oscar Del Ben via https://github.com/ruby/ruby/pull/140
+
+Sat Jul 14 17:08:13 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_s_create): raise an exception, when input elements
+	  are not one or two elements arrays.  [ruby-core:39945] [Bug #5406]
+
+Sat Jul 14 16:16:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): use
+	  Array#uniq!.
+
+	* lib/test/unit.rb (Test::Unit::Runner#deal): deal tasks to workers.
+
+	* lib/test/unit.rb (Test::Unit::Runner#quit_workers): close and kill
+	  all workers.
+
+	* lib/test/unit.rb (Test::Unit::Runner#delete_worker): delete dead
+	  worker from working set.
+
+	* lib/test/unit.rb (Test::Unit::Runner#launch_worker): add new worker
+	  to working set.
+
+	* lib/test/unit.rb (Test::Unit::Runner#launch_worker): extract.
+
+	* lib/test/unit.rb (Test::Unit::Runner#start_watchdog): extract.
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): move
+	  initializations with nothing to release outside begin/ensure.
+
+Sat Jul 14 16:04:24 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_join): should not infected by separator if it is not
+	  used.  [ruby-core:42161][Bug #5902]
+
+Sat Jul 14 02:31:55 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_thread_blocking_region): fix declarations
+	  prototypes without arguments in C++ have different meanings than C.
+
+Thu Jul 12 12:32:26 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/runner.rb: skip default gems to get rid of loading old versions
+	  before installation.
+
+Thu Jul 12 11:44:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_new_frozen): since the result object should have
+	  same tainted/untrusted bits with the original object, return new
+	  object if the shared object unmatch.  [ruby-core:39745][Bug #5374]
+
+Thu Jul 12 10:46:39 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/net/http/test_http.rb (TestNetHTTPLocalBind#test_bind_to_local*):
+	  cannot cross between network interfaces on Windows, so skip this test
+	  until we find better test.
+
+Thu Jul 12 08:48:33 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 3.2.0 (r7598)
+	* test/minitest/*: ditto
+
+Thu Jul 12 05:11:41 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defined): use method entry and id in cfp for proper
+	  superclass, since klass in iseq is shared by dynamically defined
+	  methods from the same block.  [ruby-core:45831][Bug #6644]
+
+Thu Jul 12 01:49:07 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#connect): use local_host and local_port
+	  if specified. patched by Ricardo Amorim [Feature #6617]
+
+Wed Jul 11 17:36:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: append ENCOBJS to DLDOBJS but not EXTSOLIBS which is
+	  not a target, to compile enc/encinit.c.
+
+Wed Jul 11 12:38:20 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_point_mul): nonstatic initializer
+	  of an aggregate type is a C99ism.
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_point_mul): get rid of VC++
+	  warnings.
+
+Mon Jul  9 16:11:30 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* vm_eval.c (rb_eval_string_from_file,
+	  rb_eval_string_from_file_protect): new functions to replace
+	  rb_compile_main_from_string() and ruby_eval_main().
+
+	* nacl/pepper_ruby.c: Follows the change in vm_eval.c
+
+Mon Jul  9 14:05:42 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	Reverts a half of r36079. As we discussed on ruby-dev@ and IRC,
+	we do not need to disclose intermediate representation of program.
+	The program embedding CRuby should use rb_eval_string family.
+	* include/ruby/ruby.h (ruby_opaque_t): removed.
+	  (ruby_compile_main_from_file, ruby_compile_main_from_string,
+	   ruby_eval_main): removed.
+
+	* eval.c (ruby_eval_main_internal): became ruby_exec_internal() again.
+	  (ruby_eval_main): removed.
+
+	* ruby.c (PREPARE_PARSE_MAIN) reverted.
+	  (parse_and_compile_main, ruby_compile_main_from_file,
+	   ruby_compile_main_from_string): removed
+
+Wed Jul 11 10:16:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby.h (HAVE_RUBY_THREAD_H): to show ruby/thread.h to be
+	  available.  fixup of r36355.
+
+Wed Jul 11 03:26:47 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c:  Added streaming support to inflate processing.
+	  This allows zlib streams to be processed without huge memory growth.
+	  [Feature #6612]
+	* NEWS:  ditto
+	* ext/zlib/zlib.c (zstream_expand_buffer):  Uses rb_yield when a block
+	  is given for streaming support.  Refactored to use
+	  zstream_expand_buffer_into to remove duplicate code.
+	* ext/zlib/zlib.c (zstream_expand_buffer_protect):  Added wrapper
+	  function to pass jump state back through GVL-free section to allow
+	  zstream clean-up before terminating the ruby call.
+	* ext/zlib/zlib.c (zstream_expand_buffer_without_gvl):  Acquire GVL to
+	  yield processed chunk of output stream.
+	* ext/zlib/zlib.c (zstream_detach_buffer):  When a block is given,
+	  returns Qnil mid-stream and yields the output buffer at the end of
+	  the stream.
+	* test/zlib/test_zlib.rb:  Updated tests
+
+Tue Jul 10 22:57:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/thread.h: new header file for thread stuff.
+
+	* thread.c (rb_thread_call_without_gvl): export.  [Feature#4328]
+	  returns void* instead of VALUE.  [Feature #5543]
+
+	* thread.c (rb_thread_blocking_region): deprecate.  [ruby-core:46295]
+
+Tue Jul 10 10:48:59 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (NT, NtInitialize): removed unused old macros.
+
+Tue Jul 10 10:43:37 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in: removed --enable/disable-win95 options. (see r36342)
+
+Tue Jul 10 00:44:41 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/ruby.h: Removed RUBY_GLOBAL_SETUP completely. It is
+	  no meaning definition since r24894.
+	* main.c: ditto.
+	* nacl/pepper_main.c: ditto.
+
+Mon Jul  9 23:59:36 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* dln.c: Simplify and make consistent an ifdef for Mac OS X.
+	* ext/socket/rubysocket.h: ditto.
+	* ext/tk/stubs.c: ditto.
+	* io.c: ditto.
+	* process.c: ditto.
+	* signal.c: ditto.
+	* vm_dump.c: ditto.
+
+Mon Jul  9 17:37:35 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (win95_stat): removed unnecessary macro.
+
+Mon Jul  9 17:22:16 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/configure.bat, win32/setup.mak, win32/Makefile.sub: omitted
+	  Win9x support.  removed --enable/disable-win95 options.
+
+	* include/ruby/win32.h, file.c, win32/win32.c: ditto.
+
+	* win32/README.win32: ditto.
+
+Mon Jul  9 13:28:34 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DLDFLAGS): use TARGET_ENTRY to specify an entry point
+	  instead of TARGET which may contain non-identifier characters.
+
+	* lib/mkmf.rb (create_makefile): add TARGET_NAME which is the first
+	  part consists of only word characters.  [ruby-core:46248][Bug #6709]
+
+Sun Jul  8 07:36:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (shadowing_lvar_gen, warn_unused_var): no warnings for
+	  variables starting with _.  [ruby-core:46160][Feature #6693]
+
+Sat Jul  7 23:07:30 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* test/csv/test_features.rb: add require for Tempfile.
+	* test/csv/test_serialization.rb: ditto.
+
+Fri Jul  6 06:49:50 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c (rb_ary_aref):  Added a description of the behavior of
+	  index positioning.  [Bug #6680]
+	* array.c (rb_ary_aset):  ditto.  Reordered sentences for clarity.
+	* string.c (rb_str_aref_m):  Added a description of the behavior of
+	  index positioning
+
+Fri Jul  6 05:38:44 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* string.c (rb_str_bytesize):  Improve documentation.  Patch by Oscar
+	  Del Ben from github issue #138.
+	* string.c (rb_str_empty):  ditto.
+	* string.c (rb_str_times):  ditto.
+	* string.c (rb_str_dump):  ditto.
+	* string.c (rb_str_center):  ditto.
+
+Fri Jul  6 04:05:59 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (zstream_expand_buffer_without_gvl):  Use
+	  ruby_xrealloc() to avoid crash with CALC_EXACT_MALLOC_SIZE.
+
+Thu Jul  5 17:32:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h: move ThreadShield declarations from intern.h.
+
+Thu Jul  5 16:00:24 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (ThreadShield): rename from Barrier.
+
+Thu Jul  5 15:14:50 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/runner.rb (show_progress): refine error output. do not
+	  count non-empty error message, but just warn.
+
+	* bootstraptest/runner.rb (error): show errors immediately if tty.
+
+Thu Jul  5 12:28:11 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/net/http/test_httpresponses.rb: Add a test file for
+	  Net::HTTPResponses and put a test case for the previous bug.
+
+Thu Jul  5 06:33:52 2012  Mark Dodwell <mark@mkdynamic.co.uk>
+
+	* lib/net/http/responses.rb: Fix 4xx classes to inherit correctly
+	  from Net::HTTPClientError. [Bug #6700]
+
+Wed Jul  4 21:55:35 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (proc_options): warn only if -K and -w option is specified.
+	  see also r36274 [Feature #5206]
+
+Wed Jul  4 21:41:44 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* gc.c, atomic.h (ATOMIC_SIZE_*): moved from gc.c to atomic.h
+	  [ruby-dev:45909]
+
+Wed Jul  4 19:13:15 2012  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole.rb (test_s_codepage_changed):
+	  FileSystemObject only supports ANSI or UTF-16LE encoding.
+	  Patch by h.shirosaki (Hiroshi Shirosaki) [ruby-trunk - Bug #6650]
+
+Wed Jul  4 11:52:12 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (ATOMIC_SIZE_*): 64bit Windows support.
+
+Wed Jul  4 11:11:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_frame_callee, rb_f_callee_name): fix to return the
+	  called id.  this longstanding bug has been caused and blocked by
+	  the structure of old rb_control_frame_t and rb_iseq_t.
+
+	* vm_insnhelper.c (vm_push_frame): set proper method entry.
+
+Wed Jul  4 08:29:31 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c (rb_ary_aref):  Updated documentation to indicate the
+	  starting index is an index into the array or string.  Updated
+	  examples to show behavior of indexes at the end of an array or
+	  string.  Based on patch by Marcus Stollsteimer.  [Bug #6680]
+	* array.c (rb_ary_aset):  ditto.
+	* string.c (rb_str_aref):  ditto.  Also added descriptive argument
+	  names to call-seq section.
+
+Wed Jul  4 07:05:59 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* test/zlib/test_zlib.rb (test_inflate_partial_input):  Added test for
+	  inflating incomplete zlib streams.
+
+Tue Jul  3 23:14:16 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* gc.c (ATOMIC_SIZE_EXCHANGE): fix function name on Solaris [Bug #6689]
+	  [ruby-dev:45904]
+
+Tue Jul  3 16:07:49 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (vm_malloc_fixup, vm_xrealloc, vm_xfree, after_gc_sweep): use
+	  atomic operations to update malloc_params.
+
+Tue Jul  3 14:50:16 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (zstream_run_func):  Don't exit run loop for buffer
+	  error.  [Feature #6615]
+	* ext/zlib/zlib.c:  Fix style to match existing functions.
+
+Tue Jul  3 12:05:51 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): also needed the workaround for VC8
+	  for x64.  [ruby-dev:45875] [Bug #6676]
+	  reported by aves_ramphastos (Seigo Ishigane)
+
+Tue Jul  3 11:56:46 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (zstream_detach_buffer):  Refactored tainting of
+	  output string, moving it from the callee to zstream_detach_buffer.
+	* ext/zlib/zlib.c (rb_zstream_finish):  ditto
+	* ext/zlib/zlib.c (rb_zstream_flush_next_out):  ditto
+	* ext/zlib/zlib.c (rb_deflate_deflate):  ditto
+	* ext/zlib/zlib.c (rb_deflate_flush):  ditto
+	* ext/zlib/zlib.c (rb_inflate_inflate):  ditto
+
+Tue Jul  3 11:16:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (runnable): make symbolic links to run in build directory.
+
+Tue Jul  3 10:46:06 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ruby.c (proc_options): warn if -K option is specified. [Feature #5206]
+
+Tue Jul  3 06:12:13 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (Init_Object):  Added RDoc location pointers for
+	  Kernel#methods, Kernel#protected_methods, Kernel#private_methods and
+	  Kernel#public_methods.  [Bug #6666]
+
+Tue Jul  3 06:02:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (zstream_run):  Process zlib streams without GVL.
+	  [Feature #6615]
+	* NEWS:  ditto.
+
+Mon Jul  2 22:13:04 2012  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (rb_thread_aref): add explanation for why Thread#[] and
+	  Thread#[]= are fiber-local and not thread-local.
+	  reported by Julien A.  [ruby-core:41606] [ruby-trunk - Bug #5750]
+
+Mon Jul  2 21:25:55 2012  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (timew_out_of_timet_range): specialization for
+	  SIZEOF_TIME_T == SIZEOF_INT64_T.
+
+Mon Jul  2 17:06:32 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_include_module): include modules after the origin.
+
+	* class.c (include_modules_at): skip prepended modules.
+
+	* class.c (rb_prepend_module): now basic.klass in ICLASS refers the
+	  old original class/module.  [ruby-dev:45868][Bug #6662]
+
+	* class.c (rb_mod_ancestors): ditto.
+
+	* vm_method.c (search_method): search method entry from the origin
+	  iclass.
+
+Mon Jul  2 05:54:58 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: [ruby-core:46058].
+
+Mon Jul  2 05:35:43 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (d_lite_marshal_load): accepts old dump.
+
+Mon Jul  2 03:21:53 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* README.EXT.ja: fixed args of have_struct_member() ,
+	  create_makefile() same as r35977. however, mkmf.rb include
+	  no Japanese-docs, so Appendix C was not removed. [Bug #6597]
+
+Fri Jun 29 05:08:41 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/test/unit/parallel.rb: workaround fix for rubygems.
+	  RubyGems can't find rake if the source directory is not equal to
+	  the directory which is running the test. [Bug #6604]
+
+Thu Jun 28 20:33:15 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* test/win32ole/test_win32ole.rb (test_s_codepage_changed):
+	  FileSystemObject only supports ANSI or UTF-16LE encoding.
+	  Patch by bosko (Bosko Ivanisevic) [ruby-trunk - Bug #6650]
+
+Thu Jun 28 09:27:09 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* class.c (class_instance_method_list): consider prepended Class/Module
+	  when recur != 0. [ruby-dev:45863] [Bug #6660]
+
+	* test/ruby/test_module.rb (test_prepend_instance_methods_false): add
+	  a test for it.
+
+
+Thu Jun 28 06:12:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mod_ancestors): fix ancestors order.
+	  [ruby-core:45919][Bug #6658] [ruby-dev:45861][Bug #6659]
+
+Wed Jun 27 21:28:59 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/racc/parser.rb: NotImplementError is not exist.
+
+	* lib/irb/output-method.rb (IRB::OutputMethod#print): ditto.
+
+Wed Jun 27 21:31:13 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_prepend_module): ancestors of prepending module also
+	  should be included.  [ruby-core:45914][Bug #6654]
+
+Wed Jun 27 21:01:32 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (class_instance_method_list): m_tbl in prepended
+	  class/module is NULL.  [ruby-core:45915][Bug #6655]
+
+Wed Jun 27 16:48:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_prepend_module): prepend module into another module.
+
+	* eval.c (rb_mod_prepend): new method Module#prepend.  [Feature #1102]
+
+Wed Jun 27 09:15:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (is_popen_fork): check if fork and raise NotImplementedError if
+	  unavailable.
+
+	* io.c (rb_io_s_popen): allow environment variables hash and exec
+	  options as flat parameters, not in an array arguments.
+	  [Feature#6651] [EXPERIMENTAL]
+
+	* process.c (rb_execarg_extract_options): extract exec options, but no
+	  exceptions on non-exec options and returns them as a Hash.
+
+	* process.c (rb_execarg_setenv): set environment variables.
+
+Tue Jun 26 16:57:14 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.c (register_cached_thread_and_wait):
+	  return immediately if malloc() failed.
+	  [ruby-core:43960] [ruby-trunk - Bug #6235]
+
+	* thread_pthread.c (USE_THREAD_CACHE): check already defined or not.
+
+Tue Jun 26 10:01:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_s_popen): revert r36213 "popen: shell commands with
+	  envvar" because it disabled to let single command bypass shell.
+
+Mon Jun 25 17:49:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mix_module): revert Module#mix.
+
+Mon Jun 25 16:57:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_mod_define_method): allow method transplanting from a
+	  module to either class or module.  [ruby-core:34267][Feature #4254]
+
+Mon Jun 25 11:34:45 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* internal.h: use rb_pid_t instead of pid_t because of there is no
+	  definition of pid_t here on Windows.
+
+Mon Jun 25 00:25:01 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in (for stack end address): remove human68k specific
+	  check. It is no longer supported.
+
+Sun Jun 24 23:02:17 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): merge win32 code using spawnv().
+
+Sun Jun 24 22:53:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (check_exec_fds): separate check_exec_fds_1() since
+	  nonstatic initializer of an aggregate type is not allowed by C89.
+
+Sun Jun 24 07:47:17 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): options field removed.
+
+	* process.c: follow the rb_execarg change.
+
+Sat Jun 23 23:48:21 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (proc_spawn_cmd): unused variable removed to suppress a
+	  warning.
+	  (save_env): ditto.
+
+	  [ruby-core:45797] reported by Luis Lavena.
+
+Sat Jun 23 23:19:31 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add new_pgroup_given and new_pgroup_flag
+	  fields.
+
+	* process.c (EXEC_OPTION_NEW_PGROUP): removed.
+	  (proc_spawn_cmd): take a struct rb_execarg argument.
+	  use the new fields.
+	  (rb_execarg_addopt): use the new fields.
+	  (rb_spawn_process): follow the proc_spawn_cmd change.
+
+	  [ruby-core:45794] [ruby-trunk - Bug #6633] reported by Luis Lavena.
+
+Sat Jun 23 20:26:36 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add fd_dup2, fd_close, fd_open,
+	  fd_dup2_child fields.
+
+	* process.c (EXEC_OPTION_DUP2): removed.
+	  (EXEC_OPTION_CLOSE): removed.
+	  (EXEC_OPTION_OPEN): removed.
+	  (EXEC_OPTION_DUP2_CHILD): removed.
+	  (mark_exec_arg): mark the new fields.
+	  (check_exec_redirect1): change condition for default option.
+	  (check_exec_redirect): take a struct rb_execarg argument.
+	  use the new fields.
+	  (rb_execarg_addopt): follow the check_exec_redirect change.
+	  (check_exec_fds): use the new fields.
+	  (save_redirect_fd): ditto.
+
+Sat Jun 23 19:01:18 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_execarg_fixup): fix envopts condition.
+
+Sat Jun 23 18:44:13 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (check_exec_redirect1): extracted from
+	  check_exec_redirect.
+
+Sat Jun 23 17:22:02 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (save_env): don't use EXEC_OPTION_UNSETENV_OTHERS.
+	  (rb_execarg_run_options): ditto.
+
+Sat Jun 23 17:04:08 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add env_modification field.
+
+	* process.c (EXEC_OPTION_ENV): removed.
+	  (mark_exec_arg): mark env_modification field.
+	  (rb_exec_fillarg): update the new field, instead of options array.
+	  (rb_execarg_fixup): use the new field.
+	  (save_env): ditto.
+	  (rb_execarg_run_options): ditto.
+
+Sat Jun 23 16:27:01 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add rlimit_limits field.
+
+	* process.c (EXEC_OPTION_RLIMIT): removed.
+	  (mark_exec_arg): mark rlimit_limits field.
+	  (rb_execarg_addopt): update the new fields, instead of options array.
+	  (run_exec_rlimit): use the new field.
+	  (rb_execarg_run_options): clear sarg using MEMZERO.  use the new
+	  field.
+
+Sat Jun 23 14:29:25 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add chdir_given and chdir_dir fields.
+
+	* process.c (EXEC_OPTION_CHDIR): removed.
+	  (mark_exec_arg): mark chdir_dir field.
+	  (rb_execarg_addopt): update the new fields, instead of options array.
+	  (rb_execarg_run_options): use the new fields.
+
+Sat Jun 23 13:20:47 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add close_others_given, close_others_do and
+	  close_others_maxhint fields.
+
+	* process.c (EXEC_OPTION_CLOSE_OTHERS): removed.
+	  (rb_execarg_addopt): update the new fields, instead of options array.
+	  (check_exec_fds): take eargp as an argument.  update the
+	  close_others_maxhint field.
+	  (rb_execarg_fixup): follow the argument change of check_exec_fds.
+	  (rb_execarg_run_options): use the new fields.
+
+Sat Jun 23 10:41:59 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add unsetenv_others_given and
+	  unsetenv_others_do fields.
+
+	* process.c (EXEC_OPTION_UNSETENV_OTHERS): removed.
+	  (rb_execarg_addopt): update the new fields, instead of options array.
+	  (rb_execarg_fixup): use the new fields.
+
+Sat Jun 23 09:35:47 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: use the variable name "soptions" for sargp->options.
+
+Sat Jun 23 09:17:49 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: use the name "sargp" for struct rb_execarg variables
+	  consistently for saving process attributes.
+
+	* io.c: ditto.
+
+Sat Jun 23 07:59:57 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: use the name "eargp" for struct rb_execarg variables
+	  consistently except for saving process attributes.
+
+	* io.c: ditto.
+
+	* ext/pty/pty.c: ditto.
+
+Wed Jun 20 18:27:03 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* common.mk: Add missing dependencies.
+
+Fri Jun 22 20:27:39 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg): add pgroup_given and pgroup_pgid fields.
+
+	* process.c (EXEC_OPTION_PGROUP): removed.
+	  (rb_execarg_addopt): update the new fields, instead of options array.
+	  (run_exec_pgroup): take a struct rb_execarg argument.  refer the new
+	  fields.
+	  (rb_execarg_run_options): follow run_exec_pgroup change.
+
+Fri Jun 22 18:48:51 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* README.EXT, README.EXT.ja: use "sval" for the third argument
+	  name of Data_Wrap_Struct().
+	  Suggested by @satoh_fumiyasu. Thanks!!!
+
+Fri Jun 22 18:04:26 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c, vm_eval.c: set th->base_block properly.
+	  th->base_block is information for (a) parsing, (b) compiling
+	  and (c) setting up the frame to execute the program passed by
+	  `eval' method.  For example, (1) parser need to know up-level
+	  variables to detect it is variable or method without paren.
+	  Befor (a), (b) and (c), VM set th->base_block by passed bindng
+	  (or previous frame information).  After execute (a), (b) and (c),
+	  VM should clear th->base_block.  However, if (a), (b) or (c)
+	  raises an exception, then th->base_block is not cleared.
+	  Problem is that the uncleared value th->balo_block is used for
+	  irrelevant iseq compilation.  It causes SEGV or critical error.
+	  I tried to solve this problem: to clear them before exception,
+	  but finally I found out that it is difficult to do it (Ruby
+	  program can be run in many places).
+	  Because of this background, I set th->base_block before
+	  compiling iseq and restore it after compiling.
+	  Basically, th->base_block is dirty hack (similar to global
+	  variable) and this patch is also dirty.
+
+	* bootstraptest/test_eval.rb: add a test for above.
+
+	* internal.h: remove unused decl.
+
+	* iseq.c (rb_iseq_compile_with_option): add base_block parameter.
+	  set th->base_block before compilation and restore it after
+	  compilation.
+
+	* ruby.c (require_libraries): pass 0 as base_block instead of
+	  setting th->base_block
+
+	* tool/compile_prelude.rb (prelude_eval): apply above changes.
+
+	* vm.c, vm_eval.c: ditto.
+
+	* vm_core.h: add comments.
+
+Fri Jun 22 18:19:38 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: pass struct rb_execarg value instead of its options
+	  field for saving process attribute changing functions.
+	  (save_redirect_fd): take a struct rb_execarg argument.
+	  (run_exec_dup2): ditto.
+	  (run_exec_close): ditto.
+	  (run_exec_open): ditto.
+	  (run_exec_dup2_child): ditto.
+	  (run_exec_pgroup): ditto.
+	  (run_exec_rlimit): ditto.
+	  (save_env): ditto.
+	  (rb_execarg_run_options): follow the above functions change.
+
+Fri Jun 22 17:55:48 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_backtrace.rb: decrease recursion depth
+	  to reduce consuming stack size.
+
+Fri Jun 22 13:36:50 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_init, random_load): cannot initialize frozen object
+	  again, nor with tainted/untrusted object.  [Bug #6540]
+
+Fri Jun 22 13:32:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_check_copyable): new function, to ensure the target is
+	  not frozen and the source is not tainted nor untrusted.
+
+Fri Jun 22 05:55:20 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* eval.c (ruby_cleanup):  Fixed typo.  Patch by Trever Dawe.
+	  Fixes #131 (github).  [ruby-trunk - Bug #6619]
+
+Thu Jun 21 21:16:58 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_execarg_addopt): take a VALUE argument instead of
+	  struct rb_execarg.
+	  (rb_exec_arg_addopt): follow the rb_execarg_addopt change.
+	  (check_exec_options_i): ditto.
+
+	* io.c (pipe_open): follow the rb_execarg_addopt change.
+
+	* internal.h (rb_execarg_addopt): follow the definition change.
+
+Thu Jun 21 20:34:19 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_fillarg): take a VALUE argument instead of
+	  struct rb_execarg.
+	  (rb_check_exec_options): ditto.
+	  (check_exec_options_i): ditto.
+
+Thu Jun 21 19:48:05 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_async_signal_safe): use rb_execarg_run_options
+	  instead of rb_run_exec_options_err.
+	  (rb_spawn_process): ditto.
+
+Thu Jun 21 19:02:43 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_fillarg): take a VALUE argument instead of
+	  struct rb_execarg.
+	  (rb_execarg_init): follow the rb_exec_fillarg change.
+
+Thu Jun 21 18:36:43 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_execarg_init): take a VALUE argument instead of
+	  struct rb_execarg.
+	  (rb_execarg_new): follow the rb_execarg_init change.
+	  (rb_exec_arg_init): ditto.
+
+	* internal.h (rb_execarg_init): follow the definition change.
+
+Thu Jun 21 17:20:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (new_args_tail_gen): fix GC problem of keyword rest
+	  argument.  the wrapped struct should be bound to the wrapping node
+	  before assignment of child nodes, to get rid of the case the
+	  children are referred by only the struct pointer which is not a
+	  subject of GC.  [ruby-core:45744]
+
+Thu Jun 21 07:06:52 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c (err_append): rename err_append() to compile_err_append()
+	  and move definition body.  err_append() is used only by compiling.
+
+Thu Jun 21 06:21:54 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_execarg_fixup): take a VALUE argument instead of
+	  struct rb_execarg.
+
+	* internal.h (rb_execarg_fixup): follow the definition change.
+
+	* io.c (pipe_open): follow rb_execarg_fixup change.
+
+	* ext/pty/pty.c (establishShell): ditto.
+
+Wed Jun 20 21:25:37 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (struct rb_execarg): add umask_given and umask_mask
+	  fields.
+
+	* process.c (STATIC_ASSERT): removed.
+	  (rb_execarg_addopt): follow the rb_execarg change.
+	  (rb_execarg_run_options): ditto.
+
+Wed Jun 20 20:38:23 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (struct rb_execarg) moved and renamed from
+	  struct rb_exec_arg in intern.h.
+
+	* include/ruby/intern.h (struct rb_exec_arg): refer Data object which
+	  contains struct rb_execarg.
+
+	* process.c: use struct rb_execarg instead of struct rb_exec_arg
+	  except functions declared in intern.h.
+	  (rb_exec_arg_addopt): extract a pointer to struct rb_execarg from
+	  struct rb_exec_arg.
+	  (rb_exec_arg_init): ditto.
+	  (rb_exec_arg_fixup): ditto.
+	  (rb_run_exec_options_err): ditto.
+	  (rb_run_exec_options): ditto.
+	  (rb_exec_err): ditto.
+	  (rb_exec): ditto.
+
+	* io.c: use struct rb_execarg instead of struct rb_exec_arg.
+
+	* ext/pty/pty.c: ditto.
+
+Wed Jun 20 19:13:25 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg_new): declared.
+	  (rb_execarg_get): ditto.
+
+	* process.c (mark_exec_arg): new function.
+	  (free_exec_arg): ditto.
+	  (memsize_exec_arg): ditto.
+	  (exec_arg_data_type): defined.
+	  (rb_execarg_new): new function.
+	  (rb_execarg_get): ditto.
+	  (rb_f_exec): use rb_execarg_new.
+	  (rb_spawn_internal): ditto.
+	  (rb_f_spawn): ditto.
+
+	* io.c (pipe_open_v): use rb_execarg_new.
+	  (pipe_open_s): ditto.
+
+	* ext/pty/pty.c (establishShell): use rb_execarg_new.
+
+Wed Jun 20 16:36:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/setproctitle.c (environ): use (*_NSGetEnviron()) instead of
+	  environ on Darwin for namespace cleanness, same as [ruby-core:00537].
+	  [ruby-core:45615] [Bug #6576]
+
+Wed Jun 20 11:33:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_execarg_addopt): always make Fixnum, and ignore higher
+	  bits in too large umask value.
+
+Wed Jun 20 11:24:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): deal with
+	  sudden-death of workers.
+
+Mon Jun 18 20:34:20 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* time.c (init_leap_second_info): fix non-ANSI function declaration.
+
+Mon Jun 18 20:29:04 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ruby.c (rb_f_sub): use ansi style declaration.
+	* ruby.c (rb_f_gsub): ditto.
+	* ruby.c (rb_f_chomp): ditto.
+
+Mon Jun 18 20:26:23 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c (rb_random_int32): get rid of "warning: constant 0x100000000
+	  is so big it is long" warning.
+
+Mon Jun 18 20:07:23 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* dir.c (dir_initialize): get rid of "unused return: argc = rb_scan_args()"
+	  warning.
+
+Mon Jun 18 19:31:20 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/missing.h: include math.h before checking INFINITY
+	  and NAN. Otherwise, strange macro redefinition will occur.
+
+Mon Jun 18 19:12:37 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* array.c (ary_reverse): use ansi style declaration.
+
+Tue Jun 19 18:43:50 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/backward/rubysig.h: fix visibility.  [Bug #6607]
+
+Tue Jun 19 17:51:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_execarg_run_options): do not call any methods in the
+	  async-signal-safe function.  mask has been checked with NUM2MODET()
+	  already and converted with LONG2NUM().
+
+Tue Jun 19 11:59:56 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (Init_readline): don't set 0 to
+	  rl_catch_signals and rl_catch_sigwinch. [Bug #5423]
+
+Tue Jun 19 11:52:59 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_s_get_special_prefixes): suppress
+	  warning: uninitialized instance variable.
+
+Tue Jun 19 11:43:16 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_getc): fix editline compatibility
+	  broken by r36123. [Bug #6601]
+
+Mon Jun 18 17:10:08 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_subpos): split from rb_str_substr.  returns
+	  adjusted position for substring.
+
+Mon Jun 18 10:42:57 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_getc): deal with ESC just followed
+	  by ASCII as meta prefix in incremental search mode.  based on the
+	  patch from rctay (Tay Ray Chuan) at [ruby-core:45682].  [Bug #6601]
+
+Sun Jun 17 22:23:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (rb_file_directory_p): move documentation for Dir.exist? from
+	  file.c so that the proper description will be shown instead of the
+	  documentation of File.directory?.  [ruby-core:45685]
+
+Sun Jun 17 16:21:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_win32.h (rb_thread_lock_t): make a union for USE_WIN32_MUTEX.
+	  this internal is used only in thread_win32.c, but has to be complete
+	  to define rb_thread_t.
+
+	* thread_win32.c (native_mutex_lock, native_mutex_destroy): fix for
+	  USE_WIN32_MUTEX.
+
+	* thread_win32.c (native_cond_timedwait_ms): rename reserved pattern
+	  name.  user defined symbols should not start with __.
+
+Sat Jun 16 19:24:01 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: define date_sg_t.
+
+Sat Jun 16 18:46:57 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_tmx.h: offset in struct tmx_funcs is now int.
+	* ext/date/date_strftime.c: ditto.
+	* ext/date/date_core.c: ditto.
+
+Sat Jun 16 18:31:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_setup): set running state in the normal case before
+	  popping a tag.
+
+Sat Jun 16 07:46:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): format workers
+	  results in the parent.
+
+Sat Jun 16 07:12:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/runruby.rb (File.realpath): return real path of expanded path.
+	  [Bug #6598]
+
+Sat Jun 16 07:12:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/runner.rb (main): ignore -j option for compatibility
+	  with test/unit.
+
+Sat Jun 16 07:11:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#puke): modify only result and
+	  drop useless reports, not override entirely.
+
+	* lib/test/unit/parallel.rb (Test::Unit::Worker#_run_suite): report
+	  unformatted results.  formatting messages is not a workers task.
+
+	* lib/test/unit/parallel.rb (Test::Unit::Worker#puke): store raw
+	  results.
+
+Sat Jun 16 01:27:14 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: bumping psych to 1.3.3
+	* ext/psych/psych.gemspec: ditto
+
+Fri Jun 15 20:54:28 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* vm_backtrace.c (backtrace_collect): rename from backtreace_collect.
+
+Fri Jun 15 19:22:13 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove VM_FRAME_MAGIC_FINISH (finish frame type).
+	  Before this commit:
+	    `finish frame' was place holder which indicates that VM loop
+	    needs to return function.
+	    If a C method calls a Ruby methods (a method written by Ruby),
+	    then VM loop will be (re-)invoked.  When the Ruby method returns,
+	    then also VM loop should be escaped.  `finish frame' has only
+	    one instruction `finish', which returns VM loop function.
+	    VM loop function executes `finish' instruction, then VM loop
+	    function returns itself.
+	    With such mechanism, `leave' instruction (which returns one
+	    frame from current scope) doesn't need to check that this `leave'
+	    should also return from VM loop function.
+	    Strictly, one branch can be removed from `leave' instruction.
+	  Consideration:
+	    However, pushing the `finish frame' needs costs because
+	    it needs several memory accesses.  The number of pushing
+	    `finish frame' is greater than I had assumed.  Of course,
+	    pushing `finish frame' consumes additional control frame.
+	    Moreover, recent processors has good branch prediction,
+	    with which we can ignore such trivial checking.
+	  After this commit:
+	    Finally, I decide to remove `finish frame' and `finish'
+	    instruction.  Some parts of VM depend on `finish frame',
+	    so the new frame flag VM_FRAME_FLAG_FINISH is introduced.
+	    If this frame should escape from VM function loop, then
+	    the result of VM_FRAME_TYPE_FINISH_P(cfp) is true.
+	    `leave' instruction checks this flag every time.
+	    I measured performance on it.  However on my environments,
+	    it improves some benchmarks and slows some benchmarks down.
+	    Maybe it is because of C compiler optimization parameters.
+	    I'll re-visit here if this cause problems.
+
+	* insns.def (leave, finish): remove finish instruction.
+
+	* vm.c, vm_eval.c, vm_exec.c, vm_backtrace.c, vm_dump.c:
+	  apply above changes.
+
+Fri Jun 15 19:11:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#puke): always add skipped
+	  results to the report for parallel test.  [Bug #6595]
+
+Fri Jun 15 09:01:35 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* nacl/pepper_main.c: Removed an unnecessary and erroneous inclusion.
+
+Thu Jun 14 22:59:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CPPOUTFILE): check if output is really sent to
+	  specified file to tell if -o option works.  [ruby-dev:45742]
+	  [Bug#6591]
+
+	* configure.in (RUBY_CPPOUTFILE): check if output file is actually
+	  created.  [ruby-dev:45742] [Bug#6591]
+
+Thu Jun 14 22:10:50 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (proc_exec_sh): don't strip leading spaces of the script.
+
+Thu Jun 14 15:54:02 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_file_s_basename, rb_file_s_dirname): documentation fix.
+	  File.basename and File.dirname support File::ALT_SEPARATOR.
+
+Thu Jun 14 11:10:10 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* nacl/pepper_main.c: Applies the new embedding API to pepper_ruby.
+
+Thu Jun 14 10:44:41 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* include/ruby/ruby.h: Grouped APIs for embedding CRuby interpreter.
+	  (ruby_setup, ruby_compile_main_from_file,
+	  ruby_compile_main_from_string, ruby_eval_main,
+	  ruby_set_script_name): new APIs to embed CRuby.
+	  (ruby_opaque_t) Opaque pointer to an internal data, to NODE or iseq
+	  in particular.
+
+	* eval.c (ruby_setup): Similar to ruby_init but returns an error code
+	  instead of exit(3) on error.
+	  (ruby_eval_main): Similar to ruby_exec_node but returns the
+	  evaluation result.
+	  (ruby_eval_main_internal): renamed from ruby_exec_internal.
+
+	* ruby.c (toplevel_context): new helper function.
+	  (PREPARE_EVAL_MAIN): moved.
+	  (process_options): refactored with new functions.
+	  (parse_and_compile_main) new helper function.
+	  (ruby_compile_main_from_file, ruby_compile_main_from_string) new API
+	  (ruby_set_script_name): new API.
+
+
+Thu Jun 14 10:39:48 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* eval.c: Add doxygen comments.
+
+	* ruby.c: ditto.
+
+	* thread_pthread.c: ditto
+
+	* version.c: ditto.
+
+	* vm_core.h: ditto.
+
+Thu Jun 14 10:16:07 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: revert r36071 and add NetBSD to blacklist of -ansi.
+
+Thu Jun 14 07:59:12 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (get_stack): Linux is the only OS which includes
+	  the size of guard page into the stack size.
+
+Thu Jun 14 06:21:00 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/drb/drb.rb:  Replace broken links to the English DRb book.
+	  Patch by Zachary Scott.  [ruby-trunk - Bug #6544]
+
+Thu Jun 14 06:17:47 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/observer.rb:  Update broken link to the Programming Ruby book.
+	  Patch by Zachary Scott.  [ruby-trunk - Bug #6536]
+	* lib/drb/drb.rb:  ditto.
+
+Thu Jun 14 05:23:05 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (PFETCH_READY): suppress Wunused-but-set-variable.
+
+	* regparse.c (is_onechar_cclass): restructured to clarify that c is
+	  used iff found == 1.
+
+Thu Jun 14 02:54:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: use -fbuiltin with -ansi -std=iso9899:199409.
+	  This prevents errors introduced by disabling builtin functions,
+	  which is the sub-effect of -ansi/-std.
+	  Now NetBSD can use -ansi -std=iso9899:199409.
+	  Maybe mingw, cygwin and darwin can also.
+
+Thu Jun 14 02:53:30 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in: don't remove macros. now name2ctype uses macros.
+
+	* tool/enc-unicode.rb: add comment why it uses Hash#index.
+
+	* enc/unicode/{name2ctype.kwd,name2ctype.src,name2ctype.h.blt}:
+	  update to follow the current name2ctype.h.
+	  FYI current Unicode version is 6.1.
+
+Thu Jun 14 00:16:59 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/net/http/responses.rb, lib/webrick/httpstatus.rb: Add HTTP
+	  response codes added in RFCs 2817 and 4918.  [ruby-core:45547]
+	  [Feature #6569]
+
+	* lib/net/http/responses.rb: Rename Net::HTTPMultipleChoice to
+	  Net::HTTPMultipleChoices, leaving the former as alias to the
+	  latter for backward compatibility.  [ruby-core:45547]
+	  [Feature #6569]
+
+	* lib/net/http/responses.rb: Add comments about unused,
+	  still-in-draft and private extension response codes.
+	  [ruby-core:45547] [Feature #6569]
+
+Wed Jun 13 22:44:32 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/dl/test_func.rb (test_qsort1, test_qsort2): use TYPE_SIZE_T
+	  for size_t variables. [ruby-dev:45733] [Bug #6584]
+
+Wed Jun 13 22:18:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: remove -ansi and -std options for lgamma_r() and
+	  finite().
+
+Wed Jun 13 21:46:34 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: cygwin does not provide some declarations in strict
+	  ANSI mode.
+
+Wed Jun 13 20:19:59 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_internal): move a variable declaration.
+
+Wed Jun 13 17:54:38 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* regparse.c (PFETCH_READY):  this  line was to  suppress warning,
+	  but  did emit  warnings if  -Wuninitialized was  set.  Assigning
+	  NULL instead if pfetch_prev should suffice the situation.
+
+Wed Jun 13 17:51:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: cygwin needs C99 for some stuff, e.g.,
+	  pthread_attr_setstacksize, sched_yield.
+
+Wed Jun 13 17:50:43 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (.c.i): add warnflags to make the result consistent with
+	  compilation.
+
+Wed Jun 13 15:12:07 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* configure.in: On Windows  platforms, system provided headers are
+	  VC++ optimized.  That is, C++ habits are often contaminated into
+	  various  headers.  Most  frequent  situation is  the  use of  //
+	  comments.  We bypass ANSI  C mode for them.  Otherwise extension
+	  libs cannot include those headers.
+
+Wed Jun 13 13:39:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h: get rid of C99 style one line comments.
+
+Wed Jun 13 13:39:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_alias_internal): use strdup defined as macro.
+
+Wed Jun 13 10:20:27 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_exec_fillarg): get rid of SIZE_T_MAX which may need
+	  more headers.
+
+	* process.c (rb_exec_fillarg): fix array element size.  "continue" and
+	  "readonly" exceeded the size.
+
+	* process.c (rb_exec_fillarg): use shell if the first word is reserved
+	  or special built-in name.
+	 http://pubs.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html
+
+	* process.c (rb_exec_fillarg): treat '=' only in the first word.  if
+	  the first word does not contain '=', it is the command name and
+	  environment assignments cannot be anymore.
+
+Tue Jun 12 23:45:36 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/mkmf.rb: add dummy clean-static target to prevent errors for the
+	  case real clean-static target doesn't exist.
+
+Tue Jun 12 22:49:42 2012  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* process.c (rb_exec_arg_fixup): fix compile error
+
+Tue Jun 12 21:40:13 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_fillarg): treat '=' character as a meta
+	  character to detect assignments preceding command name.
+
+Tue Jun 12 20:29:19 2012  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_exec_arg_init): deprecated.
+	  (rb_exec_arg_addopt): ditto.
+	  (rb_exec_arg_fixup): ditto.
+	  (rb_run_exec_options): ditto.
+	  (rb_run_exec_options_err): ditto.
+
+	* internal.h (rb_execarg_init): declared.
+	  (rb_execarg_addopt): ditto.
+	  (rb_execarg_fixup): ditto.
+	  (rb_execarg_run_options): ditto.
+
+	* process.c: call rb_execarg_addopt, rb_execarg_fixup,
+	  rb_execarg_run_options, rb_execarg_init.
+	  (rb_execarg_addopt): renamed from rb_exec_arg_addopt.
+	  (rb_exec_arg_addopt): stub to call rb_execarg_addopt.
+	  (rb_execarg_init): renamed from rb_exec_arg_init.
+	  (rb_exec_arg_init): stub to call rb_execarg_init.
+	  (rb_execarg_fixup): renamed from rb_exec_arg_fixup.
+	  (rb_exec_arg_fixup): stub to call rb_execarg_fixup.
+	  (rb_execarg_run_options): renamed from rb_run_exec_options_err.
+	  (rb_run_exec_options_err): stub to call rb_execarg_run_options.
+	  (rb_run_exec_options): call rb_execarg_run_options.
+
+	* io.c: call rb_execarg_addopt, rb_execarg_fixup,
+	  rb_execarg_run_options, rb_execarg_init.
+
+	* ext/pty/pty.c (establishShell): call rb_execarg_init and
+	  rb_execarg_fixup.
+
+Tue Jun 12 18:39:59 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* configure.in: enable strict ANSI mode by default in case of GCC,
+	  requested by _ko1.
+
+Tue Jun 12 06:40:23 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_fillarg): detect '#' as a meta character.
+
+Mon Jun 11 22:15:44 2012  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_proc_exec_n): deprecated.
+	  (rb_exec): ditto.
+	  (rb_exec_err): ditto.
+	  (rb_fork): ditto.
+	  (rb_fork_err): ditto.
+
+Mon Jun 11 18:49:52 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: on checking libexecinfo, don't specify /use/local.
+	  On FreeBSD people must specify --with-opt-dir or --with-execinfo-dir.
+
+Mon Jun 11 12:14:37 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove lfp (local frame pointer) and rename
+	  dfp (dynamic frame pointer) to ep (environment pointer).
+	  This change make VM `normal' (similar to other interpreters).
+	  Before this commit:
+	    Each frame has two env pointers lfp and dfp.  lfp points
+	    local environment which is method/class/toplevel frame.
+	    lfp[0] is block pointer.
+	    dfp is block local frame. dfp[0] points previous (parent)
+	    environment pointer.
+	    lfp == dfp when frame is method/class/toplevel.
+	    You can get lfp from dfp by traversing previous environment
+	    pointers.
+	  After this commit:
+	    Each frame has only `ep' to point respective environment.
+	    If there is parent environment, then ep[0] points parent
+	    environment (as dfp).  If there are no more environment,
+	    then ep[0] points block pointer (as lfp).  We call such ep
+	    as `LEP' (local EP).  We add some macros to get LEP and to
+	    detect LEP or not.
+	    In short, we replace dfp and lfp with ep and LEP.
+	  rb_block_t and rb_binding_t member `lfp' and `dfp' are removed
+	  and member `ep' is added.
+	  rename rb_thread_t's member `local_lfp' and `local_svar' to
+	  `root_lep' and `root_svar'.
+	  (VM_EP_PREV_EP(ep)): get previous environment pointer.  This macro
+	    assume that ep is not LEP.
+	  (VM_EP_BLOCK_PTR(ep)): get block pointer.  This macro assume
+	    that ep is LEP.
+	  (VM_EP_LEP_P(ep)): detect ep is LEP or not.
+	  (VM_ENVVAL_BLOCK_PTR(ptr)): make block pointer.
+	  (VM_ENVVAL_BLOCK_PTR_P(v)): detect v is block pointer.
+	  (VM_ENVVAL_PREV_EP_PTR(ptr)): make prev environment pointer.
+	  (VM_ENVVAL_PREV_EP_PTR_P(v)): detect v is prev env pointer.
+
+	* vm.c: apply above changes.
+	  (VM_EP_LEP(ep)): get LEP.
+	  (VM_CF_LEP(cfp)): get LEP of cfp->ep.
+	  (VM_CF_PREV_EP(cfp)): utility function VM_EP_PREV_EP(cfp->ep).
+	  (VM_CF_BLOCK_PTR(cfp)): utility function VM_EP_BLOCK_PTR(cfp->ep).
+
+	* vm.c, vm_eval.c, vm_insnhelper.c, vm_insnhelper.h, insns.def:
+	  apply above changes.
+
+	* cont.c: ditto.
+
+	* eval.c, eval_intern.h: ditto.
+
+	* proc.c: ditto.
+
+	* thread.c: ditto.
+
+	* vm_dump.c: ditto.
+
+	* vm_exec.h: fix function name (on vm debug mode).
+
+Mon Jun 11 11:52:18 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* compile.c (iseq_set_sequence): nonstatic initializer of an
+	  aggregate type is a C99ism.
+
+	* compile.c (enum compile_array_type_t): comma at the end of enum
+	  list is a C99ism.
+
+	* vm_backtrace.c (enum LOCATION_TYPE): ditto.
+
+Mon Jun 11 06:31:33 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_proc_exec_n): revert the function removed at r35889.
+
+Mon Jun 11 06:20:50 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): assign return
+	  value to the variable err.
+
+Mon Jun 11 06:17:06 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (native_cond_initialize): fix typo in r36022.
+	  this cause a failure on FreeBSD 8.2 amd64.
+	 http://fbsd.rubyci.org/~chkbuild/ruby-trunk/log/20120610T130201Z.diff.html.gz
+
+Mon Jun 11 05:21:57 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* .gdbinit (SDR): add SDR function.  It's only for VM debugging.
+
+Sun Jun 10 21:50:45 2012  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* nacl/nacl_config.rb: Fixed for 32bit hosts.
+
+Sun Jun 10 20:23:14 2012  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	Fixes threading on NativeClient.
+
+	* thread_pthread.c (timer_thread_sleep): Extracted out a function from
+	  thread_timer(). Added an alternative implementation for platforms
+	  that lacks select(2) or pipe(2).
+	  (rb_thread_create_timer_thread, native_cond_initialize,
+	  native_cond_destroy): Replaced wrong HAVE_XXX checks.
+
+	* configure.in (pthread_attr_init): New check.
+
+Sun Jun 10 21:30:11 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_without_timer_thread): renamed from rb_exec_err.
+	  (rb_exec_err): new stub function to call
+	  rb_exec_without_timer_thread.
+	  (rb_f_exec): call rb_exec_without_timer_thread.
+	  (rb_exec): call rb_exec_without_timer_thread.
+
+Sun Jun 10 21:13:10 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork): call rb_fork_internal instead of rb_fork_err.
+
+Sun Jun 10 20:55:59 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_ruby): call rb_fork_internal directly.
+
+Sun Jun 10 20:19:40 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_ruby): new function.
+	  (rb_f_fork): use rb_fork_ruby instead of rb_fork.
+	  (rb_daemon): ditto.
+
+	* io.c (pipe_open): use rb_fork_ruby instead of rb_fork.
+
+	* internal.h (rb_fork_ruby): declared.
+
+Sun Jun 10 18:58:16 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/net/http/response.rb: Remove a duplicated rdoc and leave a
+	  pointer.
+
+	* lib/net/http/responses.rb: Add RFC numbers to base on.
+
+Sun Jun 10 18:31:42 2012  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* configure.in (RUBY_NACL): Warns if $PATH does not contain the path
+	  to NativeClient SDK. PATH variable redefinition in GNUmakefile does
+	  not work for GNU make 3.81.
+
+Sun Jun 10 17:54:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.h (IS_STACK_DIR_UPPER): utility macro.
+
+	* thread_pthread.c (get_stack): seems stack size does not include
+	  guard size on Mac OS X.
+
+	* thread_pthread.c (ruby_init_stack): adjust stack size for offset of
+	  addr from the bottom.
+
+Sun Jun 10 15:49:47 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (retry_fork): call after_fork except in a child process.
+	  (rb_fork_internal): restrict after_fork call condition.
+
+Sun Jun 10 14:19:33 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: NetBSD 6 adds libexecinfo but it only works on amd64.
+	 http://www.mail-archive.com/source-changes-full@netbsd.org/msg38729.html
+
+Sun Jun 10 12:43:23 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_f_exec): call rb_exec_async_signal_safe except on
+	  Mac OS X.  cf. the comment in before_exec_non_async_signal_safe.
+
+Sun Jun 10 12:15:18 2012  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (popen_exec): don't call rb_thread_atfork_before_exec.  use
+	  rb_exec_async_signal_safe instead of rb_exec_err.
+	  (pipe_open): use rb_fork_async_signal_safe instead of rb_fork_err.
+
+Sun Jun 10 11:44:57 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_internal): call after_fork only unless
+	  chfunc_is_async_signal_safe.
+
+Sun Jun 10 11:33:01 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey_ec.c
+	  test/openssl/test_pkey_ec.rb: Add support for EC_POINT_mul.
+	  Patch provided by Sambasiva Suda. Thanks!
+	  [ruby-core:44408][ruby-trunk - Feature #6310]
+
+Sun Jun 10 10:48:15 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* lib/openssl/ssl.rb: Use a simple random number to generate the
+	  session id. MD5, as was used before, causes problems when
+	  using a FIPS version of OpenSSL. Issue was found by Jared
+	  Jennings, thank you!
+	  [ruby-trunk - Bug #6137]
+
+Sun Jun 10 10:27:34 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* NEWS: Add note about the new private key export behavior.
+
+Sun Jun 10 10:24:51 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_async_signal_safe): exported.
+
+	* ext/pty/extconf.rb: modify $INCFLAGS to include internal.h
+
+	* ext/pty/pty.c: include internal.h.
+	  (chfunc): don't call rb_thread_atfork_before_exec.  use
+	  rb_exec_async_signal_safe instead of rb_f_exec.
+	  (establishShell): set up earg.  use rb_fork_async_signal_safe
+	  instead of rb_fork_err.
+
+	* internal.h (rb_exec_async_signal_safe): declared.
+	  (rb_fork_async_signal_safe): declared.
+
+Sun Jun 10 10:21:37 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl.c
+	  ext/openssl/ossl_pkey_rsa.c
+	  ext/openssl/ossl_pkey_dsa.c
+	  ext/openssl/ossl_pkey_ec.c: Forbid export passwords that are less
+	  than four characters long, as OpenSSL itself does not allow this.
+	  Issue found by Eric Hodel.
+	* ext/openssl/ossl_pkey_ec.c: Add export as an alias of to_pem,
+	  following the PKey interface contract.
+	* test/openssl/test_pkey_dsa.rb
+	  test/openssl/test_pkey_rsa.rb
+	  test/openssl/test_pkey_ec.rb: Add tests that assert correct
+	  behaviour when dealing with passwords that are less than four
+	  characters long.
+	  [ruby-core: 42281][ruby-trunk - Bug #5951]
+
+Sun Jun 10 10:14:26 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_f_exec): use rb_exec_arg_prepare.
+
+Sun Jun 10 06:43:51 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: split after_exec into async-signal-safe part and rest.
+	  (after_exec_async_signal_safe): extracted from after_exec.
+	  (after_exec_non_async_signal_safe): ditto.
+	  (after_exec): call them.
+	  (rb_exec_async_signal_safe): call after_exec_async_signal_safe.
+	  (rb_exec_err): call after_exec_non_async_signal_safe instead of
+	  after_exec.
+
+Sun Jun 10 06:21:10 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* NEWS: document new features of Ruby OpenSSL.
+
+Sun Jun 10 03:09:41 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl.c: Fix error in example. Patch by David Albert.
+
+	  Add/extend existing documentation. Examples now also cover RSA
+	  signatures and PBKDF2.
+	  [ruby-core: 45154][ruby-trunk - Bug #6475]
+
+
+Sun Jun 10 01:41:45 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: Introduce SSLContext#renegotiation_cb and
+	  remove SSLContext#disable_client_renegotiation and related
+	  functionality introduced in r35797. The new callback approach
+	  gives clients maximum flexibility to decide on their own what to
+	  do on renegotiation attempts.
+	  Add documentation for SSL module and SSLError.
+	* test/openssl/test_ssl.rb: Add a test for
+	  SSLContext#renegotiation_cb.
+
+Sun Jun 10 01:37:18 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_internal): initialize exc.
+
+Sun Jun 10 00:19:25 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: don't use non async-signal-safe functions in a child
+	  process before exec, for invoking a command.
+	  (rb_exec_atfork): call rb_exec_async_signal_safe only.
+	  (retry_fork): take chfunc_is_async_signal_safe argument.  call
+	  before_fork and after_fork only unless chfunc_is_async_signal_safe.
+	  (send_child_error): take chfunc_is_async_signal_safe argument.
+	  send an exception only unless chfunc_is_async_signal_safe.
+	  (recv_child_error): take chfunc_is_async_signal_safe argument.
+	  receive an exception only unless chfunc_is_async_signal_safe.
+	  (rb_fork_internal): renamed from rb_fork_err and take
+	  chfunc_is_async_signal_safe argument.
+	  use rb_protect only unless chfunc_is_async_signal_safe.
+	  (rb_fork_err): call rb_fork_internal with false as
+	  chfunc_is_async_signal_safe.
+	  (rb_fork_async_signal_safe): call rb_fork_internal with true as
+	  chfunc_is_async_signal_safe.
+	  (rb_spawn_process): call rb_fork_async_signal_safe instead of
+	  rb_fork_err.
+
+Sat Jun  9 23:57:03 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_err): rewrite a complex "if" statement.
+
+Sat Jun  9 23:44:29 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (before_exec_async_signal_safe): extracted from
+	  before_exec.
+	  (before_exec_non_async_signal_safe): ditto.
+	  (before_exec): call before_exec_async_signal_safe and
+	  before_exec_non_async_signal_safe.
+	  (rb_exec_async_signal_safe): call before_exec_async_signal_safe.
+	  (rb_exec_err): call before_exec_non_async_signal_safe instead of
+	  before_exec.
+
+Sat Jun  9 23:36:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_load, insn_operand_intern, rb_iseq_disasm)
+	  (rb_iseq_parameters): use rb_id2str() instead of rb_id2name() to
+	  keep encoding.
+
+	* string.c (rb_str_symname_p): new function that checks if the string
+	  is valid as a symbol name.  split from sym_inspect().
+
+Sat Jun  9 22:27:05 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (retry_fork): rewrite a complex "for" statement by
+	  simple statements.
+
+Sat Jun  9 21:50:04 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (retry_fork): extracted from rb_fork_err.
+	  (send_child_error): ditto.
+	  (recv_child_error): ditto.
+
+Sat Jun  9 17:21:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_load): type is a symbol, and invalid as ID in common.
+
+Sat Jun  9 10:57:14 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_async_signal_safe): extracted from rb_exec_err.
+
+Sat Jun  9 09:31:07 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: simplified because close_others option is always
+	  enabled by default.
+	  (rb_f_exec): don't need to set the option.
+	  (rb_exec_arg_prepare): don't need to set the option.  don't need
+	  default_close_others argument.
+	  (rb_spawn_internal): don't need to give default_close_others
+	  argument for rb_exec_arg_prepare.  don't need default_close_others
+	  argument.
+	  (rb_spawn_err): don't need to give default_close_others
+	  argument for rb_spawn_internal.
+	  (rb_spawn): don't need to give default_close_others
+	  argument for rb_spawn_internal.
+	  (rb_f_system): don't need to give default_close_others argument for
+	  rb_spawn_internal.
+	  (rb_f_spawn): don't need to give default_close_others argument for
+	  rb_exec_arg_prepare.
+
+Sat Jun  9 09:00:58 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_proc_exec): call before_exec() here addition to
+	  rb_exec_err.
+
+Sat Jun  9 08:30:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack): use stack info if possible.
+
+Sat Jun  9 08:21:32 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* README.EXT (prepare extconf.rb):  Added note to see MakeMakefile for
+	  documentation of extconf.rb functions.  Patch by Zachary Scott.
+	  [ruby-trunk - Feature #6522]
+	* README.EXT (Appendix C):  Removed in favor of MakeMakefile.
+	  Patch by Zachary Scott.
+	* lib/mkmf.rb:  Merged documentation from README.EXT Appendix C.  Patch
+	  by Zachary Scott.
+
+Sat Jun  9 08:16:47 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/re.rdoc:  Completed wording in the description of the =~ operator.
+	  [ruby-trunk - Bug #6529]
+
+Sat Jun  9 08:09:38 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* string.c (rb_str_start_with):  Removed "p" from start_with? examples
+	  to match other String method examples.  [ruby-trunk - Bug #6553]
+	* string.c (rb_str_end_with):  Updated end_with? to use code markup
+	  instead of italic.
+
+Sat Jun  9 07:56:03 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/benchmark.rb:  Updated formatting of Benchmark documentation for
+	  consistency.  [ruby-trunk - Bug #6533]
+
+Sat Jun  9 07:46:26 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/delegate.rb:  Added documentation for Delegator#!.  Patch by
+	  Zachary Scott.  [ruby-trunk - Feature #6534]
+
+Sat Jun  9 07:39:50 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http/responses.rb:  Add RFC 6585 response codes.  Patch by
+	  Sangil Jung.  [ruby-trunk - Feature #6480]
+	* lib/net/http/response.rb:  ditto
+	* lib/net/http.rb:  ditto
+	* lib/webrick/httpstatus.rb:  ditto
+
+Sat Jun  9 01:24:28 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_err): before_exec() call moved from proc_exec_cmd
+	  and proc_exec_sh.
+	  (rb_proc_exec): ditto.
+
+Sat Jun  9 01:11:07 2012  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_exec_arg_init): declaration changed to
+	  return a value.
+
+	* process.c (rb_exec_arg_init): return a value.
+
+Fri Jun  8 23:44:14 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: don't check the availability of FD_CLOEXEC.  It should
+	  be available if fork() is available.
+
+	* io.c: ditto.
+
+Fri Jun  8 22:39:32 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_err): revert r35955.  The condition needs !chfunc
+	  to close ep[0] and ep[1].  The catched exception is re-raised
+	  immediately after that if status is not NULL.
+
+Fri Jun  8 19:43:33 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_err): after_exec() call moved from proc_exec_cmd
+	  and proc_exec_sh.
+	  (rb_proc_exec): ditto.
+
+Fri Jun  8 19:00:59 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (ARGV_COUNT): unused macro removed.
+	  (ARGV_SIZE): ditto.
+	  (ALLOC_ARGV): ditto.
+	  (ALLOC_ARGV_WITH_STR): ditto.
+
+Fri Jun  8 16:19:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/runner.rb (src_testdir): expand real path so that
+	  TestGem#test_self_find_files does not fail by aliased load path when
+	  srcdir contains a symbolic link.
+
+	* tool/runruby.rb (srcdir): ditto.
+
+Fri Jun  8 12:04:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_fork_err): error state in the child process is prior
+	  to exceptions in proc_syswait().
+
+	* process.c (rb_fork_err): determine status on errors.
+
+	* ext/pty/pty.c (establishShell): reraise exception if something
+	  raised during sleep.
+
+	* ext/pty/pty.c (establishShell): now needs status to protect from
+	  exceptions in rb_fork_err().
+
+Thu Jun  7 22:13:05 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_err): Fix the condition to use rb_protect.
+
+Thu Jun  7 20:29:12 2012  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h: rb_exec_arg and related stuff moved back from
+	  internal.h
+
+Thu Jun  7 15:53:03 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* .gdbinit: add function `trace_machine_instructions' to trace
+	  in native machine assemble.
+	  See https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/MachineInstructionsTraceWithGDB
+	  for more details.
+
+Wed Jun  6 21:31:21 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (proc_exec_cmd) renamed from proc_exec_v.
+	  (proc_exec_sh): renamed from rb_proc_exec_e.
+	  (proc_spawn_cmd_internal): renamed from proc_spawn_v.
+	  (proc_spawn_cmd): renamed from proc_spawn_n.
+	  (proc_spawn_sh): renamed from proc_spawn.
+
+Wed Jun  6 21:18:47 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (try_with_sh): please take care of the macro defined by
+	  you.
+
+Wed Jun  6 20:45:08 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (proc_exec_v): don't call dln_find_exe_r here because it
+	  is not async-signal-safe and proc_exec_v is called in a child
+	  process.
+	  command_abspath field of rb_exec_arg.
+	  (rb_exec_fillarg): call dln_find_exe_r and set command_abspath.
+	  (rb_exec_err): Give the absolute path of the invoking command for
+	  proc_exec_v, instead of the command name.
+
+	* internal.h: add command_abspath field for rb_exec_arg.
+
+Wed Jun  6 20:08:01 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (try_with_sh): take envp argument.
+	  (exec_with_sh): ditto.  use it for execve.
+	  (proc_exec_v): provide envp for try_with_sh.
+
+Wed Jun  6 13:25:04 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_wrap_io_handle): new API.
+	  this API wraps an I/O handle (HANDLE or SOCKET) and returns fd.
+	  the second parameter should be combination of O_*, for example,
+	  O_RDWR | O_BINARY | O_NOINHERIT.
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_unwrap_io_handle): new
+	  API.  this API unwraps an I/O handle and close the fd (not closes
+	  the handle itself).
+
+	[Feature #4906] [ruby-core:37227]
+
+Wed Jun  6 13:18:26 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_close): of course, console handle is not socket.
+
+Wed Jun  6 12:37:43 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_run_exec_options_err): allocate a temporary buffer for
+	  run_exec_dup2() for restoring fds on non-fork environments.
+
+Wed Jun  6 09:45:21 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/dl/test_c_{struct_entry,union_entity}.rb: sorry, typos.
+
+Wed Jun  6 05:27:54 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_fillarg): check use_shell field before accessing
+	  a union field.
+
+Wed Jun  6 04:58:44 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_spawn_process): prog variable is not used for Unix.
+
+Wed Jun  6 00:20:37 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_exec_arg_init): change return type to void.
+
+	* process.c (rb_exec_arg_init): don't return a value.
+	  (rb_exec_arg_prepare): ditto.
+	  (rb_spawn_process): don't take the prog argument.  extract the
+	  information from earg.
+	  (rb_spawn_internal): follow rb_spawn_process change.
+	  (rb_f_spawn): ditto.
+
+	* io.c (pipe_open): don't take the prog argument.  extract the
+	  information from eargp.
+	  (pipe_open_v): follow pipe_open change.
+	  (pipe_open_s): ditto.
+
+Tue Jun  5 23:51:33 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_exec_arg): use union to represent command invocation
+	  with/without shell.
+
+	* process.c: follow the rb_exec_arg change.
+
+	* io.c (pipe_open): ditto.
+
+Tue Jun  5 22:28:46 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: rb_exec_arg and related stuff moved from intern.h
+
+	* include/ruby/intern.h (rb_proc_exec_n): removed.
+
+Tue Jun  5 21:57:22 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_arg_fixup): allocate a temporary buffer for
+	  run_exec_dup2 here because it should be async-signal-safe.
+	  (run_exec_dup2): use the temporary buffer.
+	  (run_exec_dup2_tmpbuf_size): new function.
+
+	* include/ruby/intern.h (rb_exec_arg): add dup2_tmpbuf field.
+
+Tue Jun  5 20:13:15 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_init_copy): should check if trusted too.
+
+Tue Jun  5 19:59:13 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (strtok): declaration removed because it is not used.
+
+Tue Jun  5 19:33:51 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (proc_spawn): don't detect simple command line here
+	  because rb_exec_fillarg already did.
+
+Tue Jun  5 19:21:10 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_fillarg): bail out a loop eagerly.
+
+Tue Jun  5 19:15:14 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: add comments about async-signal-safe.
+
+	* io.c: ditto.
+
+Tue Jun  5 09:25:10 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* io.c:  Edited documentation for IO and File open and new and
+	  Kernel#open for consistency and clarity.
+
+Mon Jun  4 21:53:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_sysinit): let the system not display the
+	  critical-error-handler message box and the Windows Error Reporting
+	  dialog.  [ruby-core:45389] [Bug #6535]
+
+Mon Jun  4 19:36:25 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_fillarg): allocate one more element before
+	  beginning in argv_str for try_with_sh.
+
+	* internal.h (ARGVSTR2ARGC): adjust for the above change.
+	  (ARGVSTR2ARGV): ditto.
+
+Mon Jun  4 19:17:06 2012  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (ARGVSTR2ARGC): defined.
+	  (ARGVSTR2ARGV): defined.
+
+	* process.c (proc_exec_v): use ARGVSTR2ARGV.
+	  (rb_spawn_process): use ARGVSTR2ARGC and ARGVSTR2ARGV.
+
+	* io.c (pipe_open): use ARGVSTR2ARGV.
+
+Mon Jun  4 16:13:00 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h: remove magical code "lfp[0] & 0x02".
+	  Current VM doesn't use this bit.
+
+	* vm_core.h (RUBY_VM_GET_BLOCK_PTR): added.
+
+	* eval.c (rb_block_given_p): use RUBY_VM_GET_BLOCK_PTR().
+
+	* vm_eval.c (rb_f_block_given_p): ditto.
+
+Mon Jun  4 15:39:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (constat_apply): apply VT100 functions.
+	  [ruby-core:44958] [Feature #6418]
+
+	* win32/win32.c (constat_parse): parse some VT100 escape sequence.
+
+Mon Jun  4 14:06:12 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_exec_err): should preserve errno.
+
+Mon Jun  4 13:10:11 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/dl/test_c_{struct_entry,union_entity}.rb: broken require.
+
+Mon Jun  4 12:01:21 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_backtrace.rb: fix test.
+	  Windows path includes `:' character.
+
+Mon Jun  4 11:42:39 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (rb_location_t): fix type and field name.
+	  (1) rename rb_location_t to rb_iseq_location_t.
+	  (2) rename field names of rb_iseq_location_t to adjust
+	      RubyVM::Backtrace::Location methods.
+	      (2-1) filename -> path
+	      (2-2) filepath -> absolute_path
+	      (2-3) basename -> base_label
+	      (2-4) name -> label
+	  (3) rename filed name rb_iseq_location_t#line_no to
+	      rb_iseq_location_t#first_lineno to clear purpose of this field.
+	  (4) The field names rb_binding_t#(filename|line_no) are also renamed
+	      to rb_binding_t#(path|first_lineno).
+
+	* compile.c: apply above changes.
+
+	* iseq.c: ditto.
+
+	* proc.c: ditto.
+
+	* vm*.c: ditto.
+
+Mon Jun  4 11:40:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): also load TYPE_USRMARSHAL, TYPE_DATA using
+	  compatible loader.
+
+Mon Jun  4 11:33:42 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_run_exec_options_err): restore save_env() call for
+	  non-fork environments.
+
+	* process.c (rb_exec_err): restore environments after the failure of
+	  exec to fix [ruby-core:44093] [Bug #6249] on non-fork environments
+
+Mon Jun  4 10:42:04 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (pipe_open): follow up changes in r35889.
+
+	* process.c (proc_spawn_n): now uses char ** instead of VALUE *.
+
+	* process.c (rb_spawn_process): prog is now VALUE of String, not char *.
+
+Mon Jun  4 06:12:43 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): remove old warning for _alloc.
+
+Mon Jun  4 04:24:06 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* marshal.c: experimental test aborted.
+	* complex.c: ditto.
+	* rational.c: ditto.
+	* include/ruby/intern.h: ditto.
+
+Mon Jun  4 00:45:18 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_spawn_process): fix for Windows.  not tested.
+
+Mon Jun  4 00:11:51 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_proc_exec_e): don't use ISSPACE().  \f, \r and \v
+	  are not word separator in Bourne shell.
+
+Sun Jun  3 23:47:30 2012  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_exec_arg): remove argc and argv fields.
+	  add use_shell, argv_str and argv_buf fields.
+
+	* process.c (rb_proc_exec_e): don't split shell command line arguments
+	  here to avoid memory allocation in a child process.
+	  (rb_exec_fillarg): split shell command line arguments here.
+	  (proc_exec_v): takes argv_str argument instead of argv.
+	  (rb_proc_exec_ne): removed.
+	  (rb_proc_exec_n): removed.
+	  (rb_run_exec_options_err): don't initialize the removed fields.
+	  (rb_exec_err): don't initialize the removed fields.
+	  call proc_exec_v directly instead of rb_proc_exec_ne.
+	  (rb_spawn_process): use use_shell field.
+
+Sun Jun  3 21:53:00 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* GPL: update text of GPLv2. [ruby-core:44488] [Bug #6328]
+	  http://www.gnu.org/licenses/gpl-2.0.txt
+
+Sun Jun  3 21:22:52 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_exec_getargs): remove rb_exec_arg argument.
+
+Sun Jun  3 21:14:26 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* marshal.c: calls directly rb_{Complex,Rational}_marshal_load().
+	  But now disabled. [experimental]
+	* complex.c: followed the above.
+	* rational.c: ditto.
+	* include/ruby/intern.h: ditto.
+
+Sun Jun  3 21:18:17 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_check_argv): use rb_str_new_frozen instead of
+	  rb_str_new4.
+
+Sun Jun  3 20:10:52 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_proc_exec_e): extended version of rb_proc_exec() to
+	  call execle().
+	  (rb_proc_exec): use rb_proc_exec_e().
+	  (rb_exec_err): use rb_proc_exec_e().
+
+Sun Jun  3 19:47:18 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (vm_living_thread_num): suppress a warning.
+
+Sun Jun  3 17:23:52 2012  Tanaka Akira  <akr@fsij.org>
+
+	* use execve() to preserve environment variables when exec method is
+	  failed.  [ruby-core:44093] [ruby-trunk - Bug #6249]
+
+	* include/ruby/intern.h (rb_exec_arg): add envp_str and envp_buf field
+	  to store envp of execve().
+
+	* process.c (proc_exec_v): takes envp_str as an argument and use it
+	  for execve().
+	  (rb_proc_exec_ne): extended version of rb_proc_exec_n().
+	  (rb_proc_exec_n): use rb_proc_exec_ne().
+	  (rb_proc_exec): follow proc_exec_v() change.
+	  (fill_envp_buf_i): new function.
+	  (rb_exec_arg_fixup): set up envp_str and envp_buf.
+	  (save_env_i): removed.
+	  (save_env): removed.
+	  (rb_run_exec_options_err): don't modify environment variables.
+	  (rb_exec_err): use rb_proc_exec_ne().
+
+Sun Jun  3 16:33:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c: revert r35879 "now marshal_{load|dump} are external."
+
+	* complex.c (nucomp_marshal__{dump,load}): should use rb_marshal_{dump,load}.
+
+	* rational.c (nurat_marshal__{dump,load}): ditto.
+
+Sun Jun  3 14:13:58 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: checks whether the object is frozen or not.
+
+Sun Jun  3 14:00:51 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: wrote Complex#_dump and Complex::load.  But now
+	  disabled (due to compatibility) [experimental].
+
+	* rational.c: wrote Rational#_dump and Rational::load. ditto.
+
+Sun Jun  3 10:23:32 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_marshal_load): [ruby-core:45394]
+	* rational.c (nurat_marshal_load): ditto.
+
+Sun Jun  3 03:15:46 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (onig_number_of_names): suppress a warning.
+
+Sun Jun  3 01:36:52 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: change names.
+	  (1) Class name: RubyVM::FrameInfo -> RubyVM::Backtrace::Location.
+	  (2) Method name: RubyVM::FrameInfo.caller ->
+	      Kernel.caller_locations.
+	  (3) Instance methods of
+	      RubyVM::FrameInfo (RubyVM::Backtrace::Location)
+	      (3-1) name -> label
+	      (3-2) basename -> base_label (basename is confusing with
+		    File.basename)
+	      (3-3) line_no -> lineno (We have already similar name
+		    File#lineno, commented by kou [ruby-dev:45686]).
+	      (3-4) filename -> path.
+	      (3-5) filepath -> absolute_path.
+	      (3-5) iseq -> removed (we will make other APIs to access iseq
+		    and other information of frame for debugging).
+
+	* test/ruby/test_backtrace.rb: apply above changes.
+	  And apply comment from kou [ruby-dev:45686].
+
+Sun Jun  3 00:49:11 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: fix to build vm_backtrace.c only itself (vm_backtrace.c
+	  is no longer included from vm.c).  I hope this separation reduce
+	  compile time of vm.c.
+
+	* internal.h: ditto.
+
+	* vm.c, vm_core.h, vm_dump.c, vm_eval.c: ditto.
+
+	* vm_eval.c: some functions (callee, etc) moved to vm_backtrace.c.
+
+Sun Jun  3 00:20:53 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: added.  Separate backtrace related functions to
+	  this file.
+
+	* vm.c, common.mk: ditto.
+
+Sat Jun  2 18:09:02 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb: Inhibit zero-filled octets in an IPv4 address in
+	  all platforms. [ruby-dev:45671]
+
+	* lib/ipaddr.rb: Allow the x:x:x:x:x:x:d.d.d.d form not limited to
+	  IPv4 mapped/compatible addresses.  This change also makes it
+	  possible for the parser to understand IPv4 mapped and compatible
+	  IPv6 addresses in non-compressed form.
+
+	* lib/ipaddr.rb: Stop exposing IPSocket.valid*? methods which were
+	  only usable on non-IPv6-ready platforms.
+
+Sat Jun  2 16:59:00 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_enc_cr_str_buf_cat): don't reset coderange as unknown.
+	  the condition 'ptr_a8 && str_cr != ENC_CODERANGE_7BIT' means not
+	  unknown, str is also ASCII-8BIT because str_encindex == ptr_encindex,
+	  and nont (str_cr == ENC_CODERANGE_UNKNOWN) and
+	  str_cr != ENC_CODERANGE_7BIT means str_cr is valid because ASCII-8BIT
+	  can't be broken. [ruby-dev:45688] [Bug #6509]
+
+Sat Jun  2 07:04:48 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/re.rdoc (Performance):  Replaced incorrect example of reducing
+	  backtracking through anchoring with reduced backtracking through a
+	  range.  [ruby-trunk - Bug #6525]
+
+Sat Jun  2 06:34:15 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/re.rdoc (Performance):  Removed useless sample output from final
+	  performance example and switched from #match to #=~ for consistency.
+	  [ruby-trunk - Bug #6524]
+
+Fri Jun  1 09:30:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (class_or_module_required): extract check for class or
+	  module.
+
+Fri Jun  1 08:50:47 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c:  Updated Array documentation formatting.  Patch by Zachary
+	  Scott.  [ruby-trunk - Feature #6517]
+
+Fri Jun  1 06:57:10 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/dl/lib/dl/struct.rb (DL::CStructEntity#set_ctypes):  Refactored
+	  #set_ctypes using newer ruby features to simplify its implementation.
+	* test/dl/test_c_struct_entry.rb (class DL):  Test to verify
+	  refactoring.
+
+Fri Jun  1 06:40:25 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (Init_Object):  Restored Kernel documentation based on
+	  Pickaxe book documentation.  Patch by Zachary Scott.
+	  [ruby-trunk - Feature #6521]
+
+Fri Jun  1 06:29:42 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (rb_equal):  Let Object be a link in #=== documentation.
+	  Patch by Zachary Scott.  [ruby-trunk - Feature #6518]
+
+Thu May 31 09:27:06 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/dl/lib/dl/struct.rb (DL::CStructEntity::size):  Refactored ::size
+	  to remove unused variables and simplify using newer ruby features.
+	* test/dl/test_c_struct_entry.rb:  Test to validate refactoring
+
+Thu May 31 08:40:34 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/dl/lib/dl/struct.rb (DL::CUnionEntity#set_ctypes):  Refactored
+	  #set_types to reuse DL::CUnionEntity::size
+	* test/dl/test_c_union_entity.rb:  Added test
+
+Thu May 31 08:20:14 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/dl/lib/dl/struct.rb (DL::CUnionEntity::size):  Fixed ::size to
+	  return the size of the union.
+	* test/dl/test_c_union_entity.rb:  Test for DL::CUnionEntity::size
+
+Thu May 31 07:45:43 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/dl:  Added documentation.  Patch by Vincent Batts.
+	  [ruby-trunk - Bug #6496]
+
+Wed May 30 16:30:00 2012  Kenta Murata  <mrkn@cookpad.com>
+
+	* ext/bigdecimal/lib/bigdecimal/jacobian.rb,
+	  ext/bigdecimal/lib/bigdecimal/newton.rb:
+	  fix documentation comments.
+	  Patch by alperakgun from github.com/shyouhei/ruby/pull/8
+
+Wed May 30 16:20:00 2012  Kenta Murata  <mrkn@cookpad.com>
+
+	* ext/bigdecimal/lib/bigdecimal/jacobian.rb (Jacobian#dfdxi):
+	  fix jacobian to get stuck in an infinite loop when a solution is not
+	  found due to forget to increment nRetry counter.
+	  Patch by alperakgun from github.com/shyouhei/ruby/pull/8
+
+Wed May 30 10:58:31 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (utc_offset_arg): utc offset can be precision in seconds.
+	  e.g. old Europe/Lisbon (c.f. [ruby-dev:40066])
+
+Wed May 30 06:20:29 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* error.c (exc_set_backtrace):  Updated documentation to indicate
+	  set_backtrace allows a string as well as an array of strings.
+	  [ruby-trunk - Bug #6501]
+
+Tue May 29 17:28:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): support GNU extension triple
+	  colons modifier.  [EXPERIMENTAL]
+
+	* strftime.c (rb_strftime_with_timespec): check conversion with locale
+	  modifier.
+
+	* strftime.c (rb_strftime_with_timespec): colons are valid only for
+	  'z' and must come just before it.
+
+Mon May 28 16:56:55 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_prepare_run): StatusLineOutput
+	  needs job_status to be :replace.
+
+Mon May 28 13:35:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (do-install-*): fix dependencies.  based on the patch by
+	  nagachika at [ruby-dev:45683].  [Bug #6506]
+
+Mon May 28 12:03:04 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (obj_free): doesn't free a method table if it doesn't
+	  exist. [ruby-dev:44436]
+	* test/ruby/test_gc.rb (class TestGc): added the test case for
+	  this issue.
+
+Sun May 27 23:37:48 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm1_lvar_init.rb: added.
+	  This benchmark measures a initialize time of non-used variable.
+
+	* benchmark/bm_vm1_lvar_set.rb: added.
+	  This benchmark measures a local variables initialization time.
+
+	* benchmark/bm_vm2_bigarray.rb: added.
+	  This benchmark mesures a big array literal creation time.
+
+	* benchmark/bm_vm2_bighash.rb: added.
+	  This benchmark mesures a big hash literal creation time.
+
+	* benchmark/bm*: change notation "i=0" to "i = 0".
+
+Sun May 27 13:33:26 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: fix to continue benchmarks when
+	  an error is occurred.
+
+Sun May 27 11:27:50 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_prepare_run): fix operator
+	  precedence, so that platform and TERM should be counted.
+
+Sun May 27 10:02:33 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: allows %Ok and %Ol.
+
+Sun May 27 09:29:20 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: modified doc.
+
+Sat May 26 19:04:34 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: added description.
+
+Sat May 26 18:14:57 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: reduced the code.
+
+Sat May 26 18:08:59 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* time.c: modified doc.
+	* ext/date/date_core.c: ditto.
+
+Sat May 26 17:05:45 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (backtrace_*): change type of lev and n from size_t to int.
+	  Also set type of rb_backtrace_t#backtrace_size to int.
+	  A patch from nobu.
+
+	* vm_eval.c: ditto.
+
+Sat May 26 16:26:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (realpath_rec): UNC prefix does not end with path separator,
+	  so new separator is needed after it.
+
+Sat May 26 15:29:22 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_backtrace.rb (test_caller_lev):
+	  decrease recursion size.
+
+Sat May 26 13:50:48 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: add Kernel#caller's second argument.
+
+Sat May 26 13:40:29 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (RubyVM::FrameInfo): add a class to access each frame
+	  information.  You don't need to parse strings from caller().
+	  FrameInfo has the following methods:
+	    FrameInfo#name: method name, class name, etc with decorations.
+	    FrameInfo#basename: name without decorations.
+	    FrameInfo#line_no: line number.
+	    FrameInfo#filename: file name.
+	    FrameInfo#filepath: full filepath.
+	    FrameInfo#iseq: iseq if it is iseq frame (defined by ruby script)
+	    FrameInfo#to_s: return caller() method style string.
+	  RubyVM::FrameInfo.caller(n, lev) returns array of FrameInfo objects.
+	  The name "RubyVM::FrameInfo.caller" is long and ambiguous (it is
+	  confusing with Kernel::caller() method), we need to change the name
+	  before Ruby 2.0 release. Good names or comments are welcome.
+
+	* test/ruby/test_backtrace.rb: add a test for above change.
+
+Sat May 26 12:18:09 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (frame_info_to_str): add `break'.
+
+	* vm.c (backtrace_object): remove lev and n parameter.
+	  backtrace_object always returns all of backtrace information.
+
+	* vm.c (rb_backtrace_to_str_ary): fix to use backtrace_object().
+	  This change improve performance of caller(lev, n).
+
+	* benchmark/bm_vm3_backtrace.rb: added to check above improvement.
+	  FYI: measurement on my laptop, 1.9.3p229 needs 5.125 sec,
+	  and current trunk only needs 0.299sec.
+
+Sat May 26 11:05:09 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_frame_info_t): keep previous ISEQ frame info for CFUNC
+	  frame info.  And fix to cache a calculated line_no of ISEQ frame
+	  info.
+
+Sat May 26 09:54:53 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: Allow disabling client-side renegotiation.
+	* test/openssl/test_ssl.rb: Simple tests for this.
+
+	  Client-side renegotiation is still considered problematic, even
+	  when used in the context of secure renegotiation (RI, RFC 5746).
+	  The changes allow users to either completely disable client
+	  renegotiation on the server, or to specify a maximum number of
+	  handshakes allowed in total. The number of total handshakes is
+	  counted in a callback set as SSL_set_info_callback. If the
+	  maximum number of handshakes is exceeded an error will be raised
+	  We do not support renegotiation in the OpenSSL extension, therefore
+	  this feature can only be tested externally.
+	  The feature is opt-in, the default setting will be to allow
+	  unlimited client renegotiation, as was the case before.
+
+Fri May 25 23:38:58 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_ssl.rb: Clarify the intention of errors to be
+	  expected. Two errors are possible when connection is refused due
+	  to a protocol version that was explicitly disallowed,
+	  OpenSSL::SSL::SSLError or Errno::ECONNRESET, depending on the
+	  OpenSSL version in use.
+
+Fri May 25 22:19:40 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: Revert r35583
+	* test/openssl/test_ssl.rb: Handle ECONNRESET in code instead to avoid
+	  the test failing in Ruby CI [1]
+
+	[1] http://u64.rubyci.org/~chkbuild/ruby-trunk/log/20120507T190102Z.log.html.gz#test-all
+
+Fri May 25 19:51:36 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (rb_f_caller): caller() method accepts second optional
+	  argument `n' which specify how many frames should return.
+	  For example, `caller(0, 1)' returns only one frame information
+	  which calls caller() method.  If there are less than n frame
+	  information, then all frame information are returned.  If n is 0,
+	  then always return [].
+	  This fix is part of [ruby-dev:42345] [Ruby 1.9-Feature#3917].
+	  However, performance and features are not enough.
+	  RDoc is also not available.
+
+	* test/ruby/test_backtrace.rb: add a test for above.
+
+Fri May 25 17:05:07 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (oldbt_init, vm_backtrace_str_ary): arg->data should
+	  be initialized before calling `backtrace_each()'.
+
+Fri May 25 16:11:27 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* trunk/ext/-test-/printf/printf.c: change function names because of
+	  conflict with msvcrt.  fixed build error of mswin.
+
+Fri May 25 10:52:52 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: refactoring backtrace related functions.
+	  (1) unify similar functions (rb_backtrace_each() and
+	      backtrace_object()).  backtrace_each() is a unified function.
+	      variation:
+	      a) backtrace_object(): create backtrace object.
+	      b) vm_backtrace_str_ary(): create bt as an array of string.
+	      c) vm_backtrace_print(): print backtrace to specified file.
+	      d) rb_backtrace_print_as_bugreport(): print backtrace on
+	                                            bugreport style.
+	  (2) remove rb_backtrace_each().  Use backtrace_each() instead.
+	  (3) change the type of lev parameter to size_t.
+	      a) lev == 0 means current frame (exception, etc use it).
+	      b) lev == 1 means upper frame (caller(0) use it).
+
+	* vm_core.h, vm_dump.c, vm_eval.c: ditto.
+
+	* vm.c (backtrace_object(), vm_backtrace_str_ary()): fix to return a
+	  correct size of caller(lev) array.
+	  Let n be a "caller(0).size" then ln as caller(lev).size should be
+	  (n - lev).  However, the previous implementation returns a wrong
+	  size array (ln > n - lev).  [ruby-dev:45673]
+
+	* test/ruby/test_backtrace.rb: add tests for backtrace.
+
+Fri May 25 08:51:39 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* enum.c (enum_count):  Enumerable#count no longer uses #size when
+	  counting elements.  Patch by Nobuhiro IMAI.  [ruby-trunk - Bug #6473]
+
+Fri May 25 01:15:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (ruby__sfvextra): [EXPERIMENTAL] use inspect instead of
+	  to_s if plus flag is given.
+
+	* vsnprintf.c (BSD_vfprintf): pass sign flag.
+
+Fri May 25 00:37:22 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_indexer.rb (setup, teardown): save @tempdir
+	  to remove it properly.  [Bug #5348]
+
+Thu May 24 23:36:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): [EXPERIMENTAL] object representation in
+	  rb_enc_vsprintf(). [Feature #5896]
+
+Thu May 24 15:33:01 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_method_defined_by): removed.
+	  nobu pointed out that rb_method_basic_definition_p() is enough
+	  for last commit.
+
+	* error.c, eval_error.c: change for above.
+
+Thu May 24 14:30:13 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: add RubyVM::Backtrace object (btobj).
+	  Backtrace information contains an array consists of location
+	  information for each frames by string.
+	  RubyVM::Backtrace object is lightweight backtrace information,
+	  which contains complete information to generate traditional style
+	  backtrace (an array of strings) with faster generation.
+	  If someone accesses to backtrace information via
+	  Exception#backtrace, then convert a RubyVM::Backtrace object to
+	  traditional style backtrace.
+	  This change causes incompatibility on marshal dumped binary
+	  of Exception.  If you have any trouble on it, please tell us
+	  before Ruby 2.0 release.
+	  Note that RubyVM::Backtrace object should not expose Ruby level.
+
+	* error.c, eval.c, vm_eval.c: ditto.
+
+	* internal.h: ditto.
+
+	* eval_error.c: fix to skip "set_backtrace" method invocation in
+	  creating an exception object if it call a normal set_backtrace
+	  method (defined by core).
+
+	* test/ruby/test_settracefunc.rb: fix for above change.
+
+	* vm_method.c (rb_method_defined_by): added.  This function
+	  checks that the given object responds with the given method
+	  by the given cfunc.
+
+	* benchmark/bm_vm2_raise1.rb, benchmark/bm_vm2_raise2.rb:
+	  add to measure exception creation speed. raise1 create
+	  exception objects from shallow stack frame.  raise2 create
+	  exception objects from deep stack frame.
+
+Thu May 24 12:07:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_strip_bom): check EOF.  [Bug #6487][ruby-core:45203]
+
+Wed May 23 22:06:14 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/header.rb (Net::HTTPHeader#range): fix broken parser of
+	  HTTP Range request. Old one can't parse invalid specs and multiple
+	  specs correctly.
+
+Wed May 23 10:18:54 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (finish_overlapped_socket, overlapped_socket_io):
+	  replace ECONNABORTED to EPIPE in send, sendto and sendmsg to improve
+	  BSD socket compatibility.  this change removes a failure on the test
+	  of net/ftp.
+
+Wed May 23 05:35:58 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Broke up Net::HTTP into individual files.
+	  [ruby-trunk - Feature #6435]
+	* lib/net/http/backward.rb:  ditto.
+	* lib/net/http/response.rb:  ditto.
+	* lib/net/http/exceptions.rb:  ditto.
+	* lib/net/http/responses.rb:  ditto.
+	* lib/net/http/generic_request.rb:  ditto.
+	* lib/net/http/header.rb:  ditto.
+	* lib/net/http/request.rb:  ditto.
+	* lib/net/http/proxy_delta.rb:  ditto.
+	* lib/net/http/requests.rb:  ditto.
+
+Wed May 23 05:15:11 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* class.c (rb_mod_init_copy):  Clear the cached inspect string of a
+	  dup'd anonymous module or class.  [ruby-trunk - Bug #6454]
+	* test/ruby/test_module.rb (class TestModule):  ditto
+
+Tue May 22 16:49:15 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: add a data type rb_location_t to store iseq location
+	  information.
+	  rb_location_t#filename, filepath, name and line_no was moved from
+	  rb_iseq_t.  rb_location_t#basename is a new field which is
+	  similar to `name' field without any decoration.
+	  `name' field contains some decoration such as `block in foo'.
+	  `basename' only contains `foo'.
+	  rb_iseq_t contains memory object of rb_location_t.
+
+	* iseq.c: setup rb_location_t for each rb_iseq_t memory objects.
+
+	* compile.c, proc.c, vm.c, vm_dump.c, vm_eval.c, vm_insnhelper.c,
+	  vm_method.c: support about it.
+
+Tue May 22 00:45:05 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* struct.c (rb_struct_members): Refactoring.  As Struct#members had
+	  returned an array of String, the old code was needed to convert
+	  Symbols to Strings.  But it is almost unnecessary because the
+	  method now returns an array of Symbols.  A patch by Masaki
+	  Matsushita <glass.saga at gmail dot com>  [Feature #6218]
+	  [ruby-dev:45451]
+
+Mon May 21 19:20:25 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::FTP#retrbinary): close only if conn is not nil
+	  because transfercmd may fail and return nil.
+
+	* lib/net/ftp.rb (Net::FTP#retrlines): ditto.
+
+Mon May 21 15:10:28 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/syslog/syslog.c: Classify constants and macros into several
+	  sub-modules. (Syslog::Priority, Syslog::Level, Syslog::Option
+	  and Syslog::Macros)
+
+	* ext/syslog/syslog.c (mSyslog_inspect): Use rb_sprintf().
+
+	* ext/syslog/syslog.c (mSyslog_inspect): Make sure self is a
+	  module before calling rb_class2name().
+
+Mon May 21 12:44:11 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (install): It seems tcl/tk is skipped in Travis
+	  CI. Trying to fix the situation.
+
+Mon May 21 12:11:25 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (ENCOBJS): add dependencies.
+
+	* enc/make_encmake.rb (target_encodings): extract dependencies.
+
+Mon May 21 11:26:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::FTP#transfercmd): rescue shutdown.
+
+Sun May 20 23:00:11 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): reopen $stdout to NULL, since setting
+	  $stdout cannot affect child processes.
+
+Sun May 20 21:36:39 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/shift_jis.c (code_to_mbclen): return
+	  ONIGERR_INVALID_CODE_POINT_VALUE if the code is invalid.
+
+	* string.c (tr_next): increment character until the code
+	  is a valid character. [ruby-dev:45652] [Bug #6450]
+
+Sun May 20 12:25:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (LIBRUBY_SO): link EXTSOLIBS too.
+
+	* ext/extmk.rb (mf.macro): use EXTSOLIBS instead of SOLIBS to get rid
+	  of discard libraries needed by default.  [Bug #6462]
+
+Sat May 19 19:04:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (command_output): ENCOBJS is needed for all linked
+	  ruby, if --disable-shared and --with-static-linked-ext.
+
+	* ext/extmk.rb (command_output): dmyext is needed as DLDOBJS if no
+	  static linked extensions.
+
+	* Makefile.in, common.mk (PROGRAM): no extension libraries.
+
+	* common.mk (build-ext): pass macros for libruby.so.
+
+	* ext/extmk.rb (command_output): link extension libraries and encoding
+	  libraries into libruby.so, not ruby executable.
+
+	* ext/extmk.rb (command_output): fold long macro lines.
+
+	* Makefile.in (LIBEXT): add macro.
+
+	* configure.in (ENCOBJS, EXTOBJS): use LIBEXT, not hardcoded suffix.
+
+	* Makefile.in (LIBRUBY_A): fix typo.  re-applying r35242.
+
+Sat May 19 04:46:53 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/openssl/extconf.rb: Use Logging::message instead of message.
+	* ext/zlib/extconf.rb: ditto.
+
+Fri May 18 18:13:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#configuration): keep space at end of
+	  OUTFLAG and COUTFLAG.  [ruby-dev:45650]
+
+Fri May 18 17:39:42 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): Added error
+	  check when failing fcntl(). [Bug #6147] [ruby-dev:45364]
+
+Fri May 18 17:41:00 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): link archives only, skip script only
+	  extension libraries.
+
+Fri May 18 17:25:33 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c: bump up fiber machine stack size when running on 64bit
+	  arch. [Bug #6344] [ruby-dev:45554]
+
+Fri May 18 15:20:56 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic.build): duplicate args before adding
+	  new items. (don't change arguments)
+
+	* lib/uri/generic.rb (URI::Generic.build): use URI::Generic::COMPONENT
+	  if this method is called from URI::Generic.
+
+	* lib/uri/generic.rb (URI::Generic.build2): escape only if the item is
+	  a String.
+
+	* lib/uri/generic.rb (URI::Generic.build2): use DEFAULT_PARSER because
+	  it doesn't have parser method. [Bug #6420]
+
+Fri May 18 15:54:07 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/zlib/extconf.rb: Use an exception instead of bare puts.
+
+Fri May 18 15:53:05 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/psych/extconf.rb: Use an exception instead of bare abort.
+
+Fri May 18 15:51:32 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/fiddle/extconf.rb: Use an exception instead of bare abort.
+
+Fri May 18 15:49:35 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/readline/extconf.rb: Use an exception instead of bare exit.
+
+Fri May 18 15:38:11 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/ripper/extconf.rb: Use an exception instead of bare
+	  Logging.message.
+
+Fri May 18 15:23:06 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/openssl/extconf.rb: Clarify a message when hit Apple
+	  OpenSSL issue.
+
+Fri May 18 15:14:32 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/extmk.rb: Show a message when extconf.rb raised an exception.
+	* ext/openssl/extconf.rb: Use exception raising instead of message
+	  and/or abort. We want to display error message to console _and_
+	  logging into mkmf.log.
+
+Fri May 18 06:14:07 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/syslog/lib/syslog/logger.rb:  Added Syslog::Logger which was
+	  ported from the SyslogLogger gem.  [ruby-trunk - Feature #5096]
+	* NEWS:  ditto.
+	* test/syslog/test_syslog_logger.rb:  ditto.
+
+Fri May 18 01:28:21 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (transcode_string): fix encoding index names.
+	  Thanks markizko for reporting.
+
+Thu May 17 23:03:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: fix function name to be checked, to initialize
+	  rb_thread_cond_t properly.
+
+	* thread_pthread.c (native_cond_initialize, native_cond_destroy):
+	  fix macro name.
+
+Thu May 17 12:53:07 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* thread.c, thread_pthread.c: Moved pthread-specific preprocessor
+	  hacks to thread_pthread.c
+
+Thu May 17 12:18:47 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* io.c: Fix a mistake on merging the patch in the previous commit.
+
+Thu May 17 11:33:07 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	Imports Ruby's port to NativeClient (a.k.a NaCl).
+	Patch by Google Inc. [ruby-core:45073].
+
+	* configure.in (RUBY_NACL): New M4 func to configure variables for
+	  NaCl.
+	  (RUBY_NACL_CHECK_PEPPER_TYPES): New M4 func to check the old names
+	  of Pepper interface types.
+	  (BTESTRUBY): New variable to specify which ruby should be run on
+	  "make btest". NaCl can run the built binary by sel_ldr, but it need
+	  rbconfig.rb. So this variable is distinguished from $MINIRUBY.
+
+	* thread_pthread.c: Disabled some features on NaCl.
+
+	* io.c: ditto.
+
+	* process.c: ditto.
+
+	* signal.c: ditto.
+
+	* file.c: ditto.
+
+	* missing/flock.c: ditto.
+
+	* nacl/pepper_main.c: An example implementation of Pepper application
+	  that embeds Ruby.
+
+	* nacl/example.html: An example of web page that uses the Pepper
+	  application.
+
+	* nacl/nacl-config.rb: Detects variants of NaCl SDK.
+
+	* nacl/GNUmakefile.in: Makefile template for NaCl specific build
+	  process.
+
+	* nacl/package.rb: script for packaging a NaCl-Ruby embedding
+	  application.
+
+	* nacl/reate_nmf.rb: Wrapper script of create_nmf.py
+
+	* dln.c (dln_load): Added a hack to call on NaCl.
+
+	* util.c (ruby_getcwd): Path to the current directory is not available
+	  on NaCl.
+
+Thu May 17 10:54:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/extconf.rb: add -l options to $libs not $LDFLAGS,
+	  to be passed to EXTLIBS in exts.mk.
+
+	* enc/encinit.c.erb: use %-lines to adjust indent in the generated file.
+
+	* lib/mkmf.rb (MakeMakefile#have_framework): combine -framework option
+	  and its argument with an equal sign not to be separated in merge_libs.
+
+	* ext/tk/extconf.rb: ditto.
+
+	* ext/extmk.rb: EXTLDFLAGS also needs to be passed.
+
+Wed May 16 15:44:22 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* configure.in: Fix an unbalanced quote.
+
+Wed May 16 15:43:10 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/extmk.rb (exts.mk): use double quotes instead of single quotes
+	  for commandline because it's not recognized as quotes on Windows.
+
+Wed May 16 15:15:55 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LD): enclose with single quotes but not double quotes
+	  not to expand command substitution.
+
+Wed May 16 14:19:51 2012  Yuki Yugui Sonoda  <yugui@google.com>
+
+	Supports static linking of extensions and encodings again.
+	Fixes --with-static-linked-ext.
+
+	Patch by Google Inc. [ruby-core:45073].
+
+	* Makefile.in (ENCOBJS, EXTOBJS): New variables to specify static
+	  linked libraries. Also reintroduces extinit.o, introduces encinit.o
+	  introduces encinit.o
+
+	* common.mk: Builds static libraries rather than shared objects if
+	  specified.
+
+	* configure.in (LD): new substitution.
+
+	* enc/depend: Supports static linked libraries
+	  (libencs, libenc, libtrans): New target.
+
+	* enc/encinit.c.erb: new template to generate the initialization of
+	  statically linked encodings.
+
+	* enc/make_encmake.rb (--module): new flag to specify whether static
+	  or dynamic.
+
+	* transcode_data.h (TRANS_INIT): New macro to get rid of the name
+	  collision of encoding initializers and transcoder initializers.
+
+	* ext/extmk.rb: Fixes the behavior on $extstatic is true.
+
+	* lib/mkmf.rb (clean-static): new target to clean up static linked
+	  libraries.
+
+	* ruby.c (process_options): New initializes statically linked
+	  encodings here.
+
+Wed May 16 14:30:43 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c: fixed a merge mistake of r33878, reported by nobu via IRC.
+
+Wed May 16 06:59:41 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: should also be aware of flags on
+	  complex specifier.
+
+Wed May 16 05:11:29 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: fix a bug with string
+	  subclass dumping and loading.
+
+	* test/psych/test_array.rb: pertinent tests
+
+	* test/psych/test_string.rb: ditto
+
+Wed May 16 01:31:21 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: convert omap tagged maps to
+	  Psych::Omap objects rather than hashes. [Bug #6425]
+
+	* test/psych/test_omap.rb: pertinent test.
+
+Wed May 16 01:15:45 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: keep a reference to
+	  custom coders so that GC does not impact dumped yaml reference ids.
+
+Tue May 15 23:59:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Options#setup_options): add --color option.
+
+	* lib/test/unit.rb (Test::Unit::Runner#_prepare_run): defer color code
+	  initialization to regard --color option.
+
+Mon May 14 16:28:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_arglist): should reset lexical states after empty
+	  argument list with no parenthesis as well as parenthesized list,
+	  so that reserved name method definition work.  [ruby-dev:45626]
+	  [Bug #6403]
+
+Mon May 14 00:14:24 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* enumerator.c (lazy_take_func, lazy_take): multiple calls of
+	  force/to_a method to Enumerator::Lazy#take should return same
+	  results. [ruby-dev:45634] [Bug #6428]
+
+	* test/ruby/test_lazy_enumerator.rb (test_take_recycle): add test for
+	  above.
+
+Sun May 13 23:38:31 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* test/ruby/test_io.rb (test_flush_in_finalizer1): don't use IO.for_fd
+	  to close IO objects. it create IO object with already closed fd, and
+	  cause occasional Errno::EBADF in following tests. [ruby-core:45020]
+	  [Bug #6228]
+
+Sun May 13 23:32:16 2012  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO): revert r35631. it broke the intent of
+	  test_flush_in_finalizer1.  [ruby-core:43951] [Bug #6228]
+
+Sun May 13 22:46:36 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/etc/etc.c (passwd_ensure): move endpwent() call from
+	  passwd_iterate to close /etc/passwd on exception.
+	  (group_ensure): move endgrent() call from group_iterate to close
+	  /etc/group on exception.
+
+Sun May 13 18:10:43 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: removed unused code and changed the style.
+
+Sun May 13 17:37:56 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: refactored.
+
+Sun May 13 06:40:12 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* test/ruby/test_io.rb (class TestIO):  Disable GC during IO tests to
+	  avoid file descriptors being GC'ed. Suggestion by Tomoyuki Chikanaga
+	  [ruby-core:43951][Bug #6228]
+
+Sat May 12 07:00:16 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/sdbm/init.c:  Added documentation.  Patch by Justin Collins,
+	  cleanup by Zachary Scott.  [ruby-trunk - #6410]
+
+Sat May 12 06:02:03 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/fileutils.rb (cp_r):  Fixed cp_r example.  Patch by TJ Koblentz
+	  from pull request #114.  [ruby-trunk - Bug #6411]
+
+Sat May 12 05:23:06 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (rb_threadptr_execute_interrupts_common):
+	  test_signal_requiring of test/ruby/test_signal.rb fail if the sub
+	  process is killed on waiting IO in lex_io_gets in rb_load_file in
+	  rb_load_internal in require.
+	  This is because
+	  (1) the process receive the killing signal in
+	      rb_thread_io_blocking_region in rb_read_internal in lex_io_gets.
+	  (2) set th->errinfo as INT2FIX(TAG_FATAL) at
+	      rb_threadptr_execute_interrupts_common.
+	  (3) escape rb_load_file in rb_load_internal and jump to EXEC_TAG()
+	      without set loaded as TRUE.
+	  (4) call first rb_exc_raise(GET_THREAD()->errinfo); because loaded
+	      is FALSE as above. this errinfo should be an exception object
+	      but this is INT2FIX(TAG_FATAL).
+	  Don't call first rb_exc_raise if GET_THREAD()->errinfo is Fixnum.
+
+Fri May 11 14:23:11 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): begin/end block should be isolated from outside.
+	  [ruby-dev:45631][Bug #6419]
+
+Fri May 11 14:09:47 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (PUSH): to prevent VALUE from GC,
+	  must not cast it to unsigned long, which may be shorter than
+	  VALUE, and the result can be mere garbage.
+
+Fri May 11 09:51:07 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#failed): no unnecessary
+	  newlines if no reports to be displayed.
+
+Thu May 10 10:55:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/minitest/test_minitest_mock.rb: Correct requiring path to
+	  metametameta.rb.
+
+	* test/minitest/test_minitest_unit.rb: Correct requiring path to
+	  metametameta.rb.
+
+Thu May 10 10:18:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lex_state_name): returns name for lex_state_e, for debug
+	  use.
+
+Wed May  9 16:36:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#pkg_config): check if libs resulted from
+	  pkg-config works actually.
+
+Wed May  9 16:01:38 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (decode_utf7, encode_utf7): refactored by
+	  Nobuyoshi Nakada, to use String#encode.
+
+Wed May  9 13:26:25 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_remote_fetcher.rb: skip OpenSSL dependent
+	  tests if not available.
+
+Wed May  9 08:09:38 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 3.0.0 (r7435)
+	* test/minitest/*: ditto
+	* test/rubygems/*: Imported fixes for buggy use of assert_match
+	  and deprecated assert_block
+	UNBUNCH YOUR PANTIES. THE TESTS DO NOT RUN CLEAN ON OSX.
+
+Wed May  9 06:28:59 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* re.c (rb_reg_equal):  Removed incorrect example for Regexp#== with
+	  "n" option.  [ruby-talk - Bug #6415]
+
+Wed May  9 06:23:33 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: reverted.
+
+Wed May  9 04:31:26 2012  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/rinda/ring.rb (lookup_ring_any): fix Rinda::RingFinger.primary
+	  hungs forever. [ruby-talk:395364]
+
+Tue May  8 21:09:00 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* include/ruby/win32.h (FD_SET): change function to macro.
+	  To avoid buffer overflow when smaller FD_SETSIZE is used in ext
+	  libraries.
+
+	* win32/win32.c (rb_w32_fdset): this function is not used anymore.
+	  But we leave this for compatibility.
+
+	* win32/win32.c (rb_w32_select_with_thread): fix SEGV when smaller
+	  FD_SETSIZE is used in ext libraries. Dereference of fd_set pointer
+	  causes SEGV.
+
+	* test/-ext-/win32/test_fd_setsize.rb(TestFdSetSize): add tests for
+	  above.
+	* ext/-test-/win32/fd_setsize/depend: ditto.
+	* ext/-test-/win32/fd_setsize/extconf.rb: ditto.
+	* ext/-test-/win32/fd_setsize/fd_setsize.c: ditto.
+
+	  [ruby-core:44588] [Bug #6352]
+
+Tue May  8 20:44:46 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* io.c (io_unread): fix IO#pos with mode 'r' bug on Windows.
+	  If the end of reading buffer is CR, io_unread() needs to unread one
+	  more byte.
+	  [ruby-core:44874] [Bug #6401]
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#test_pos_with_buffer_end_cr):
+	  add a test for above.
+
+Tue May  8 13:38:17 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/date/date_core.c: improving introduction in Date/DateTime
+	  documentation. patched by Daniel Kaufman via Github.
+		https://github.com/ruby/ruby/pull/110
+
+Tue May  8 13:36:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (POSTLINK): default to : command to get rid of flag
+	  only command, since BSD make does not work with it.
+
+Tue May  8 13:35:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (MiniTest#run_test): remove exact trace and get rid
+	  of IndexError, which could caused by modified $@ sometimes.
+
+Tue May  8 11:21:27 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/minitest/metametameta.rb (MetaMetaMetaTestCase#assert_report):
+	  support drive letter on Windows.  yes, the original code is metameta.
+
+Tue May  8 08:54:48 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/abbrev.rb:  Fixed typo in abbrev pattern documentation.  Based on
+	  patch by Mark Rushakoff.  [ruby-trunk - #6346]
+
+Tue May  8 07:44:18 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_start_ssl): remove useless rb_sys_fail
+	  before ossl_raise. this cause a test failure on Linux.
+	 http://u64.rubyci.org/~chkbuild/ruby-trunk/log/20120507T190102Z.log.html.gz
+
+Tue May  8 05:35:18 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (Init_Object):  Added reference to variable.c where
+	  public_constant and private_constant documentation lives.  [#6381]
+
+Tue May  8 04:47:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#output): prefer local output to
+	  get rid of unexpected side effect in test/minitest/metametameta.rb.
+
+	* lib/test/unit.rb (MiniTest#run_test): show the running test in $0.
+
+	* lib/test/unit.rb (Test::Unit::StatusLineOutput): new class to output
+	  in status line.
+
+	* test/testunit/test_hideskip.rb (TestHideSkip#test_hideskip):
+	  MiniTest#puke now reports Skipped messages only if verbose mode.
+
+	* test/testunit/test_sorting.rb (TestTestUnitSorting#test_sorting):
+	  ditto.
+
+	* lib/test/unit.rb (Test::Unit::Runner#puke): modify only result and
+	  drop useless reports, not override entirely.
+
+	* bootstraptest/runner.rb (exec_test, show_progress): show rotators
+	  and pass/fail counts.
+
+	* sample/test.rb (PROGRESS): refine output.
+
+Tue May  8 02:34:26 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/minitest/unit.rb (assert_match): refix of r35563.
+	  r35563 breaks the intention of the original change.
+	 https://github.com/seattlerb/minitest/commit/68858105b2eb11c85105ffac5f32b662c59397f3
+	* lib/minitest/unit.rb (refute_match): ditto.
+
+Mon May  7 21:19:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Merge JSON 1.7.1.
+	 https://github.com/flori/json/commit/e5b9a9465c1159fae533bca320d950b772bcb4ac
+
+Mon May  7 22:54:22 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: add support for option flags
+	  OpenSSL::SSL::OP_NO_TLSv1_1
+	  OpenSSL::SSL::OP_NO_TLSv1_2
+	  to allow blocking specific TLS versions. Thanks to Justin Guyett for
+	  pointing this out to me.
+	* test/openssl/test_ssl.rb: add tests to assert correct behavior when
+	  blocking certain versions of TLS/SSL both on server and client side.
+	  Also refactored tests to reduce boilerplate code a little.
+	* test/openssl/utils.rb: rescue Errno::ECONNRESET for tests where
+	  client rejects the connection because a forbidden protocol version
+	  was used.
+
+Mon May  7 20:14:15 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (random_bytes): call to_int method for the
+	  argument at first.
+
+Mon May  7 17:54:12 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/minitest/unit.rb (assert_match): replace matcher only if both
+	  matcher and obj are String. fix r35541. [Bug #6405]
+	  DON'T COMMIT IF YOU CAN'T RUN TEST.
+	  FIX AS SOON AS POSSIBLE YOU BREAK TESTS.
+	  patched by ayumin.
+	  https://github.com/seattlerb/minitest/pull/124
+
+	* lib/minitest/unit.rb (refute_match): ditto.
+
+Mon May  7 13:41:00 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (PROGRAM), configure.in (POSTLINK): sign built program
+	  using RUBY_CODESIGN identity.
+
+Mon May  7 13:03:55 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (body_type_attachment): parse body type
+	  "ATTACHMENT". [ruby-core:44849] [Bug #6397]
+
+Mon May  7 10:49:36 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (Init_bigdecimal): define IDs before
+	  they are used. [ruby-core:44900] [Bug #6406]
+
+Mon May  7 10:27:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/rmd160/rmd160.c (RMD160_Update): fix for huge data.
+
+Mon May  7 10:23:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fileutils/fileasserts.rb: use assert_equal, assert_match, and so on.
+
+	* test/ruby/enc/test_utf16.rb, test/ruby/enc/test_utf32.rb,
+	  test/ruby/test_io_m17n.rb (assert_str_equal): ditto.
+
+	* test/rubygems/test_gem_remote_fetcher.rb
+	  (assert_data_from_{server,proxy}): ditto.
+
+	* test/test_pstore.rb (test_thread_safe): ditto.
+
+Mon May  7 10:16:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_installer.rb (TestGemInstaller#test_dir): fix
+	  argument order.  expected value must come first.
+
+Mon May  7 09:14:11 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: support TLSv1.1 & TLSv1.2. Add
+	  SSLContext#version to inspect the version that was negotiated for
+	  a given connection.
+	* ext/openssl/extconf.rb: detect TLS 1.1 & 1.2 support.
+	* test/openssl/test_ssl.rb: add tests for TLS 1.1 & 1.2 given they
+	  are supported by the native OpenSSL being used.
+
+Sun May  6 21:34:29 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_encoding_set): suppress warnings. [ruby-dev:45627]
+	  this tmp1 is not required after r35538.
+
+	* addr2line.c: suppress warnings.
+
+Sun May  6 18:39:39 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): remove unused variable `size'.
+
+Sun May  6 14:50:03 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb: show actual read length in an error message.
+
+Sat May  5 06:43:10 2012  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 2.12.1 (r7323)
+	* test/minitest/*: ditto
+
+Sat May  5 01:47:33 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/zlib/test_zlib.rb (test_inflate): add a test for Zlib.inflate.
+	  patched by headius (Charles Nutter). [ruby-core:44859] [Bug #6398]
+
+	* test/zlib/test_zlib.rb (test_deflate): add a test for Zlib.deflate.
+
+Sat May  5 00:53:55 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (parse_mode_enc): remove warnings 'Ignoring internal encoding'.
+	  [ruby-core:44455] [Bug #6324]
+
+	* io.c (io_encoding_set): ditto.
+
+Fri May  4 07:19:02 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/parser.rb (RDoc.binary?): fix wrong regexp.
+	  [ruby-core:44798] [Bug #6393]
+
+Fri May  4 01:33:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/parser.rb (RDoc.alias_extension): a real file is irrelevant
+	  to aliasing.  [ruby-core:44796][Bug #6392]
+
+	* lib/rdoc/parser.rb (RDoc.zip?): non-existent file will not be a zip
+	  file.
+
+	* lib/rdoc/parser.rb (RDoc.can_parse_by_name): accept aliased
+	  extension file names.
+
+	* lib/rdoc/parser.rb (RDoc.binary?): binary read data may have
+	  incomplete multibyte sequence.  [ruby-core:44798][Bug #6393]
+
+Wed May  2 23:55:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::RequireFiles#non_options): expand
+	  real path to get rid of loading same files via symlinks.
+
+Wed May  2 23:26:04 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* cont.c (rb_fiber_m_transfer): improve sample code in Fiber#transfer
+	  documentation. emphasize the difference between transfer and resume.
+
+Wed May  2 23:21:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): allow spaces between lambda arrow and
+	  parenthesis.  [ruby-dev:45605][Feature #6390]
+
+Wed May  2 19:06:30 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* cont.c (rb_fiber_m_transfer): Improved Fiber documentation.
+	  patched by Anuj Dutta. [ruby-core:44540][Bug #6343]
+
+Wed May  2 13:06:37 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README, README.ja: reformatted using rdoc markup.  based on the
+	  patches by zzak (Zachary Scott) in [Feature #6388].
+
+	* README, README.ja: updated the author's mail address.
+
+Wed May  2 09:46:09 2012  Kouji Takao  <kouji@takao7.net>
+
+	* ext/readline/readline.c (Readline.special_prefixes=)
+	  (Readline.special_prefixes): new function. An original patch was
+	  created by nagachika. [Feature #5784]
+
+Tue May  1 22:18:45 2012  Kouji Takao  <kouji@takao7.net>
+
+	* ext/readline/readline.c (Readline.pre_input_hook)
+	  (Readline.insert_text, Readline.redisplay): new function. An
+	  original patch was created by nagachika. [Feature #5785]
+
+Tue May  1 15:46:48 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: "$(Q)-..." doesn't work on nmake.
+
+Tue May  1 15:32:10 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: replace '@' prefix to '$(Q)' to control build
+	  process outputs.
+
+Tue May  1 14:17:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/deprecation.rb (OpenSSL.check_func): check if header is
+	  available for macro compatibility.
+
+Tue May  1 10:53:54 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_settracefunc.rb: ignore traces from another threads
+	  because Kernel.set_trace_func affects other threads.
+
+Tue May  1 06:04:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/sha2/sha2.c (REVERSE32): explicitly cast since unsigned
+	  long may be larger than sha2_word32.
+
+	* ext/digest/sha2/sha2.c (SHA{256,512,384}_{Final,End}): should clear
+	  whole content, not pointer size.
+
+	* ext/digest/*/extconf.rb: use pkg_config to use same library with
+	  openssl.  [ruby-core:44755][Bug #6379]
+
+	* ext/openssl/deprecation.rb: extract check for broken Apple OpenSSL.
+
+Tue May  1 05:02:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (optflags): disable unsafe optimizations.
+	  [ruby-core:44679][Bug #6370]
+
+Mon Apr 30 23:36:49 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/fileutils.rb (copy_metadata): use File.lchown and File.lchmod to
+	  update meta data of symlinks.
+
+Mon Apr 30 23:05:53 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/test_continuation.rb (tracing_with_set_trace_func): don't
+	  call Continuation from other threads. [ruby-dev:45596] [Bug #6382]
+
+Mon Apr 30 20:10:04 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/zlib/extconf.rb: detect z_crc_t type which will be defined
+	  since zlib-1.2.7.
+
+	* ext/zlib/zlib.c (rb_zlib_crc_table): use z_crc_t if available.
+
+Mon Apr 30 09:02:15 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/openssl/lib/openssl/ssl.rb: add hostname to "hostname does not
+	  match server cert." error. patched by Wes Morgan via Github.
+	  https://github.com/ruby/ruby/pull/122
+
+Mon Apr 30 04:43:53 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/json/yaml_events.rb: implicit styles should not
+	  be changeable for JSON events.
+
+Sun Apr 29 06:12:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assoc, parser_yylex): add syntax to splat keyword hash.
+	  [ruby-core:44591][Feature #6353]
+
+	* compile.c (compile_array_): generate keyword splat insns.
+
+	* vm.c (m_core_hash_merge_kwd): merge keyword hash into intermediate
+	  hash.  leftward argument is prior currently.
+
+Fri Apr 27 12:34:23 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): should convert a Bignum value to
+	  unsigned long long on Win64.
+	  [ruby-core:44636][Bug #6364] reported by raylinn@gmail.com (ray linn)
+
+Fri Apr 27 10:58:17 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/readline/test_readline.rb (setup): avoid affected by user's
+	  inputrc file.  [ruby-dev:45584][Bug #6357]
+
+Fri Apr 27 01:45:05 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (rb_threadptr_execute_interrupts_common):
+	  handle timer_interrupt only on the first loop for the case to avoid
+	  the infinite loop like following case:
+	  * there is 2 Ruby threads (3 pthreads)
+	  (1) main thread is waiting at gvl_yield:112 (native_cond_wait)
+	  (2) sub thread works
+	  (3) sub thread waits at gvl_yield:133 (native_mutex_unlock)
+	  (4) main thread works
+	  (5) main thread goes to gvl_acquire_common
+	  (6) main thread call rb_wakeup_timer_thread
+	  (7) timer thread set timer interrupt to the main thread
+	  (8) main thread works
+	  (9) main thread waits at gvl_acquire_common:64 (native_cond_wait)
+	  (10) sub tread works
+	  (11) set sub thread as the current thread
+	  (12) run Ruby thread
+	  (13) ...100ms
+	  (14) sub thread goes to rb_threadptr_execute_interrupts_common
+	  (15) sub thread call rb_thread_schedule_limits
+	  (16) sub thread call gvl_release_common
+	  (17) sub threads waits at gvl_yield:121 (native_cond_wait)
+	  (18) main threads works
+	  (19) main thread back to gvl_yield
+	  (20) set main thread as the current thread
+	  (21) main thread call gvl_yield
+	  (22) main thread waits at gvl_yield:112 (native_cond_wait)
+	  As described above, the main thread can't escape from
+	  rb_threadptr_execute_interrupts_common.
+	  See extended memo: http://bugs.ruby-lang.org/projects/ruby-trunk/wiki/R35480_ExtendedMemo
+
+Fri Apr 27 07:15:07 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_pack_sockaddr_un): support the longest
+	  path in sockaddr_un, really.
+	  reported by nagachika.
+	 http://d.hatena.ne.jp/nagachika/20120426/ruby_trunk_changes_35474_35476
+
+Thu Apr 26 12:28:06 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (init_unix_addrinfo): support the longest
+	  path in sockaddr_un.
+	  (inspect_sockaddr): ditto.
+	  (addrinfo_mdump): ditto.
+	  (addrinfo_mload): ditto.
+	  (rsock_unixpath_str): new function.
+	  (rsock_unixpath): removed.
+	  (rsock_unixaddr): use rsock_unixpath_str.
+
+	* ext/socket/socket.c (sock_s_pack_sockaddr_un): support the longest
+	  path in sockaddr_un.
+	  (sock_s_unpack_sockaddr_un): ditto.
+	  (sock_s_gethostbyaddr): unused variable removed.
+
+	* ext/socket/unixsocket.c (rsock_init_unixsock): support the longest
+	  path in sockaddr_un.
+
+	* ext/socket/rubysocket.h (rsock_unixpath_str): declared.
+	  (rsock_unixpath): removed.
+
+	* test/socket/test_unix.rb: comment out test_nul because abstract unix
+	  sockets may contain NULs.
+
+Thu Apr 26 01:32:33 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/optparse/test_summary.rb (test_summary_containing_space): add
+	  test for r35467. OptionParser#to_a shouldn't split banner by spaces.
+
+Wed Apr 25 23:02:46 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (init_unix_addrinfo): refine error message
+	  format.
+	  (addrinfo_mload): show more information on "too long AF_UNIX path"
+	  error.
+	  (addrinfo_unix_path): ditto for "too short AF_UNIX address" and
+	  "too long AF_UNIX address" error.
+
+Wed Apr 25 05:46:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#to_a): split for each lines.
+	  [ruby-dev:45568][Bug #6348]
+
+Tue Apr 24 21:57:53 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (init_unix_addrinfo): show actual path length
+	  when it is too long for Unix socket.
+
+	* ext/socket/unixsocket.c (rsock_init_unixsock): ditto.
+
+	* ext/socket/socket.c (sock_s_pack_sockaddr_un): ditto.
+
+Tue Apr 24 21:43:58 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/net/smtp.rb (check_continue): raise an error with an explanatory
+	  message.  [ruby-core:35854] [Feature #4598]
+
+Tue Apr 24 21:11:31 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#to_a): should split by end-of-line,
+	  and MUST TEST IT, MUST RUN THE TEST, MUST VERIFY BEFORE BACKPORT.
+	  [ruby-dev:45568][Bug #6348]
+
+Tue Apr 24 19:59:31 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/euc_jp.c: added EUC-JP-2004 and its alias EUC-JISX0213.
+	  [ruby-dev:45571] [Feature #6349]
+	  Requested by Kyouhei Yanagita <yanagi@shakenbu.org>.
+
+	* enc/trans/japanese_euc.trans: ditto.
+
+	* enc/trans/JIS/JISX0213-[12]%UCS@{BMP,SIP}.src: JIS X 0213:2004 ->
+	  Unicode mapping table from NetBSD.
+
+	* enc/trans/JIS/UCS@{BMP,SIP}%JISX0213-[12].src: Unicode -> JIS X
+	  0213:2004 mapping table from NetBSD.
+
+	* tool/transcode-tblgen.rb: added SIP support.
+
+	* test/ruby/test_transcode.rb: tests of above changes.
+
+Tue Apr 24 18:12:13 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: fix to output warning when the same literals
+	  are available as a condition of same case clause.
+	  And remove information ('#n') because we can find duplicated
+	  condition with explicit line numbers.
+	  [ruby-core:38343] [Ruby 1.9 - Bug #5068]
+
+	* test/ruby/test_syntax.rb: add a test for above.
+
+Tue Apr 24 17:03:51 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (waitpid): need to check the return value of
+	  FindChildSlotByHandle() before passing poll_child_status().
+	  this fixed a SEGV in test-all.  reported by ko1 via IRC.
+
+Tue Apr 24 16:04:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): EXPR_BEG by keywords is a start point of
+	  commands.  [ruby-dev:45563][Bug #6347]
+
+	* parse.y (superclass): ditto for superclass.
+
+	* parse.y (parser_parse_string, parser_here_document): ditto for
+	  string interpolation.
+
+	* parse.y (parser_yylex): ditto for singleton class.
+
+Tue Apr 24 15:51:41 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#to_a): should split by end-of-line
+	  [ruby-dev:45568][Bug #6348]
+
+	* lib/optparse.rb (OptionParser#to_a): String#to_a is no longer
+	  defined.  [ruby-dev:45568][Bug #6348]
+
+Tue Apr 24 12:46:50 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* hash.c, object.c, struct.c, lib/ostruct.rb: add to_h methods.
+	  [Feature #6276]
+
+Tue Apr 24 10:54:34 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/drb/drbtest.rb ({DRbCore,DRbAry}#teardown): cannot pass SIGTERM
+	  to another process on Windows, so use SIGINT instead.
+
+Tue Apr 24 00:25:39 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (rb_check_deadlock): refine an error message of deadlock
+	  detection.  [ruby-core:44336] [Bug #6288]
+
+Tue Apr 24 00:14:42 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (primary): remove wrong "fixpos" that caused incorrect
+	  source_location of blocks.  [ruby-core:42232] [Bug #5930]
+
+	* test/ruby/test_proc.rb: add a test for above.
+
+Mon Apr 23 22:56:08 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/iconv: deprecated.  [Feature #6322]
+
+Mon Apr 23 22:07:00 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_unix.rb (bound_unix_socket): make temporary
+	  filename shorter for less possibility of Unix socket path over
+	  107 bytes when TMPDIR has long path.
+
+Mon Apr 23 20:35:49 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (szInternalCmds, internal_match, internal_cmd_match):
+	  get rid of a segmentation fault with GCC 4.7.0.
+	  reported by raylinn@gmail.com (ray linn) at [ruby-core:44505]
+	  [Bug #6333], and patched by mame.
+
+	* test/ruby/test_system.rb (TestSystem#test_system): test for it.
+
+Mon Apr 23 20:11:02 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/drb/ssl.rb: generate 1024 bits RSA key instead of 512 bits.
+	  OpenSSL 1.0.1 rejects 512 bits RSA key for TLS1.2 with SHA512.
+	 http://rt.openssl.org/Ticket/Display.html?id=2769&user=guest&pass=guest
+	  reported by Bohuslav Kabrda.
+	  [ruby-core:43844] [ruby-trunk - Bug #6221]
+
+Mon Apr 23 19:54:33 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/drb/drbtest.rb: rescue Errno::ESRCH for Process.kill.
+	  reported by NARUSE, Yui.  [ruby-dev:45551]
+
+Mon Apr 23 14:16:45 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .gdbinit (rb_ps_vm): follow st_table's packing change.
+
+Mon Apr 23 10:43:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: disable rubygems not to load rbconfig.rb before
+	  fake.rb.  [ruby-core:44492][Bug #6329]
+
+Sun Apr 22 20:26:06 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/drb/extservm.rb (DRb::ExtServManager): don't use /bin/sh to
+	  invoke service subprocess.  mark detach threads for clean up.
+
+	* test/drb/drbtest.rb: clean up the service subprocess in teardown.
+
+	* test/drb/test_drb.rb: set @service_name for teardown.
+
+	* test/drb/test_drbunix.rb: ditto.
+
+	* test/drb/test_drbssl.rb: ditto.
+
+	[ruby-dev:45547]
+
+Sun Apr 22 07:51:29 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/drb/ssl.rb: close accepted TCP socket if SSL accept is failed.
+	  [ruby-dev:45541]
+
+Sat Apr 21 14:36:49 2012  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb: fix sticks on some tests problem
+	  [Bug #6272]
+
+Fri Apr 20 12:24:04 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/ssl_certs/AddTrustExternalCARoot.pem:  Removed to avoid
+	  conflict with ca-bundle.pem
+	* lib/rubygems/ssl_certs/VerisignClass3PublicPrimaryCertificationAuthority-G2.pem:
+	  ditto.
+	* lib/rubygems/ssl_certs/Entrust_net-Secure-Server-Certification-Authority.pem:
+	  ditto.
+
+Fri Apr 20 08:07:06 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 1.8.23 which contains security
+	  fixes:
+
+	  RubyGems now disallows redirection from HTTPS to HTTP.
+
+	  RubyGems now verifies SSL connections.
+
+	  See https://github.com/rubygems/rubygems/blob/1.8/History.txt for
+	  changes since 1.8.22.
+	* test/rubygems:  ditto.
+
+Thu Apr 19 16:33:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): fix carry-up bug and
+	  overwrite '+' with '-' if negative offset less than a hour.
+	  [ruby-core:44447][Bug #6323]
+
+Thu Apr 19 09:39:57 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/-test-/win32/dln/extconf.rb: need import library for ordinal
+	  entry even on mingw.  [ruby-core:44441][Bug #6320]
+
+Thu Apr 19 09:35:15 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* random.c (random_init):  Clarify that the default seed is
+	  Random.new_seed, not zero.  Based on patch by Roger Pack.
+	  [ruby-trunk - Bug #6313]
+	* random.c (rb_f_srand):  ditto.
+
+Thu Apr 19 08:59:02 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/curses/curses.c (window_nodelay):  Fixed call-seq of nodelay to
+	  include the '='.
+
+	  Improved description window.nodelay=.
+
+Thu Apr 19 08:47:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* io.c (io_readpartial):  Document the output buffer parameter is
+	  overwritten with the read contents even when non-empty.
+	  Patch by yu nobuoka.  [ruby-trunk - Bug #6285]
+	* io.c (io_read_nonblock):  ditto.
+	* io.c (io_read):  ditto.
+	* io.c (rb_io_sysread):  ditto.
+	* io.c (argf_read):  ditto.
+	* io.c (argf_readpartial):  ditto.
+	* ext/stringio/stringio.c (strio_read):  ditto.
+	* test/ruby/test_argf.rb (class TestArgf):  Add test for existing
+	  behavior of read outbuf.
+	* test/ruby/test_io.rb (class TestIO):  ditto.
+	* test/stringio/test_stringio.rb (class TestStringIO):  ditto.
+
+Wed Apr 18 22:58:55 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DOT, DOXYGEN): use AC_CHECK_PROGS instead of
+	  AC_CHECK_PROG which needs the third argument.  [ruby-core:44433]
+	  [Bug #6316]
+
+	* configure.in (PKG_CONFIG): fix condition to skip older version
+	  of pkg-config.  continue in backticks does not affect outside.
+
+Wed Apr 18 13:59:40 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/file.c (INVALID_FILE_ATTRIBUTES): define for old SDK.
+
+Wed Apr 18 10:22:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): add an interim digit for
+	  the timezone offset which is less than an hour.
+
+Wed Apr 18 09:58:29 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/version.rb:  Fixed init_with warning by calling into
+	  yaml_initialize (for syck) from psych's init_with
+
+Wed Apr 18 09:03:43 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 1.8.22 plus r33517 and r35337 which
+	  were ported to the rubygems git repository.
+
+	  See https://github.com/rubygems/rubygems/blob/1.8/History.txt for
+	  changes since 1.8.11.
+
+	* test/rubygems:  ditto.
+
+Tue Apr 17 22:18:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): fix padding of time zone
+	  offset.  [ruby-dev:43287][Bug #4458]
+
+Tue Apr 17 13:11:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (rb_w32_check_imported): skip ordinal entries.  based on a
+	  patch by phasis68 (Heesob Park) at [ruby-core:44381].
+	  [ruby-core:44371][Bug #6303]
+
+Mon Apr 16 18:22:14 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* spec/default.mspec: expand relative path for ruby_exe which uses
+	  them with Dir.chdir; it breaks relative paths, for example
+	  core/kernel/exec_spec.rb.
+
+Mon Apr 16 16:22:40 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (gmtime_r, localtime_r): POSIX compliant reentrant
+	  versions.
+
+	* configure.in (RUBY_MSVCRT_VERSION): define on mingw too.
+
+	* win32/Makefile.sub (config.h): prefix RT_VER with RUBY and make
+	  more descriptive to get rid of potential conflict.
+
+Mon Apr 16 15:19:39 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (NO_RUBY_VENDOR_LIB): fix missing comma.
+
+Mon Apr 16 12:17:12 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb (hermitian?): Bug fix, patch by George Koehler
+	  [Bug #6290] [rubyspec:4b9573d7613]
+
+Mon Apr 16 09:42:50 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rubygems/remote_fetcher.rb (Gem::RemoteFetcher#download): should
+	  use File.identical? to check the identity of the files.
+	  this fixed an error of a test on Windows.
+
+Sat Apr 14 12:55:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (UNREACHABLE): gcc 4.4 eliminates unreachable code
+	  if -O3 is given.
+
+	* win32/win32.c (child_result): dropped colon.
+
+Sat Apr 14 10:45:18 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#start):
+	  partially revert r35315.
+
+	* test/webrick/test_server.rb (test_start_exception):
+	  received signal is delivered to the main thread, so it is needed to
+	  emulate it. patched by Eric Hodel. [ruby-core:44348] [Feature #6236]
+
+Sat Apr 14 09:35:45 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* variable.c (trace_ev):  Removed "not reached" comment as this line is
+	  reached.
+	* variable.c (rb_obj_remove_instance_variable):  Replaced "not reached"
+	  comment with the UNREACHABLE macro.
+	* variable.c (rb_mod_const_missing):  ditto.
+	* variable.c (rb_mod_remove_cvar):  ditto.
+	* enum.c (first_i):  ditto.
+	* string.c (rb_str_aref):  ditto.
+	* string.c (str_byte_aref):  ditto.
+	* string.c (rb_to_id):  ditto.
+	* io.c (rb_io_fmode_modestr):  ditto.
+	* io.c (rb_io_oflags_modestr):  ditto.
+	* pack.c (num2i32):  ditto.
+	* vm_eval.c (rb_method_missing):  ditto.
+	* vm_eval.c (rb_f_throw):  ditto.
+	* dir.c (dir_read):  ditto.
+	* win32/win32.c (child_result):  ditto.
+	* struct.c (rb_struct_getmember):  ditto.
+	* struct.c (rb_struct_set):  ditto.
+	* struct.c (rb_struct_aref_id):  ditto.
+	* eval.c (rb_f_raise):  ditto.
+	* process.c (rb_f_exit_bang):  ditto.
+	* process.c (rb_f_exit):  ditto.
+	* process.c (rb_f_abort):  ditto.
+	* ext/-test-/iter/break.c (iter_break_value):  ditto.
+	* ext/pty/pty.c (pty_check):  ditto.
+	* ext/openssl/ossl_pkey.c (ossl_pkey_new):  ditto.
+	* ext/readline/readline.c (rb_remove_history):  ditto.
+	* ext/stringio/stringio.c (strio_unimpl):  ditto.
+	* numeric.c (num_sadded):  ditto.
+	* numeric.c (num_init_copy):  ditto.
+	* numeric.c (rb_num2ll):  ditto.
+	* numeric.c (rb_num2ull):  ditto.
+	* vm_insnhelper.c (call_cfunc):  ditto.
+	* ruby.c (opt_W_getter):  ditto.
+	* bignum.c (rb_big_coerce):  ditto.
+	* file.c (rb_f_test):  ditto.
+
+Sat Apr 14 08:38:20 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* encoding.c (rb_enc_codepoint_len):  Use UNREACHABLE to avoid "control
+	  reaches end of non-void function" warnings.  [ruby-trunk - Bug #6066]
+	* re.c (name_to_backref_number):  ditto.
+	* object.c (rb_Float):  ditto.
+	* io.c (io_readpartial):  ditto.
+	* io.c (io_read_nonblock):  ditto.
+	* pack.c (rb_uv_to_utf8):  ditto.
+	* proc.c (rb_method_entry_arity):  ditto.
+	* vm_method.c (rb_f_notimplement):  ditto.
+	* struct.c (rb_struct_aset_id):  ditto.
+	* class.c (rb_scan_args):  ditto.
+	* process.c (rlimit_resource_type):  ditto.
+	* process.c (rlimit_resource_value):  ditto.
+	* process.c (p_uid_switch):  ditto.
+	* process.c (p_gid_switch):  ditto.
+	* ext/digest/digest.c (rb_digest_instance_update):  ditto.
+	* ext/digest/digest.c (rb_digest_instance_finish):  ditto.
+	* ext/digest/digest.c (rb_digest_instance_reset):  ditto.
+	* ext/digest/digest.c (rb_digest_instance_block_length):  ditto.
+	* ext/bigdecimal/bigdecimal.c (BigDecimalCmp):  ditto.
+	* ext/dl/handle.c (rb_dlhandle_close):  ditto.
+	* ext/tk/tcltklib.c (pending_exception_check0):  ditto.
+	* ext/tk/tcltklib.c (pending_exception_check1):  ditto.
+	* ext/tk/tcltklib.c (ip_cancel_eval_core):  ditto.
+	* ext/tk/tcltklib.c (lib_get_reltype_name):  ditto.
+	* ext/tk/tcltklib.c (create_dummy_encoding_for_tk_core):  ditto.
+	* ext/tk/tkutil/tkutil.c (tk_hash_kv):  ditto.
+	* ext/openssl/ossl_ssl.c (ossl_ssl_session_reused):  ditto.
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_dsa_verify_asn1):  ditto.
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_point_is_at_infinit):  ditto.
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_point_is_on_curve):  ditto.
+	* ext/fiddle/conversions.c (generic_to_value):  ditto.
+	* ext/socket/raddrinfo.c (rsock_io_socket_addrinfo):  ditto.
+	* ext/socket/socket.c (sock_s_getnameinfo):  ditto.
+	* ext/ripper/eventids2.c (ripper_token2eventid):  ditto.
+	* cont.c (return_fiber):  ditto.
+	* dmydln.c (dln_load):  ditto.
+	* vm_insnhelper.c (vm_search_normal_superclass):  ditto.
+	* bignum.c (big_fdiv):  ditto.
+	* marshal.c (r_symlink):  ditto.
+	* marshal.c (r_symbol):  ditto.
+
+Fri Apr 13 17:12:09 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* hash.c (inspect_i): keep string's coderange.
+
+Fri Apr 13 15:26:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_aset, rb_hash_update, rb_hash_update_by): use
+	  st_update() to reduce evaluation of hash values.
+
+Fri Apr 13 15:17:36 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#stop): fix r35303;
+	  this method is to deny new connections, not shutdown yet.
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#start):
+	  re-raise exception only when the exception is Interrupt (^C).
+
+Thu Apr 12 19:51:45 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: added some notes.
+
+Wed Apr 11 17:16:49 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (compile_array, compile_array_):
+	  Divide big array (or hash) literals into several blocks and
+	  concatenate them.  There was a problem that a big array (hash)
+	  literal causes SystemStackError exception (stack overflow)
+	  because VM push all contents of the literal onto VM stack to
+	  make an array (or hash).  To solve this issue, we make several
+	  arrays (hashes) and concatenate them to make a big array (hash)
+	  object.  [ruby-dev:37701] [Bug #982]
+
+	* compile.c (iseq_compile_each, setup_args): use modified
+	  compile_array.
+
+	* vm.c (m_core_hash_from_ary, m_core_hash_merge_ary,
+	  m_core_hash_merge_ptr): added for above change.
+
+	* id.c (Init_id), parse.y: add core method ids.
+
+	* bootstraptest/test_literal.rb: add simple tests.
+
+	* bootstraptest/test_eval.rb: remove rescue clause to catch
+	  SystemStackError exception.
+
+	* test/ruby/test_literal.rb: add tests to check no stack overflow.
+
+Thu Apr 12 07:10:37 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/uri/generic.rb (module URI):  URI now downcases the scheme to
+	  follow RFC 2396 section 3.1.  [ruby-trunk - Feature #4551]
+	* test/uri/test_generic.rb (class URI):  Test for above
+
+Thu Apr 12 06:15:44 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/protocol.rb (module Net):  Added ReadTimeout to match
+	  OpenTimeout.  ReadTimeout is now raised by rbuf_fill instead of
+	  Timeout::Error to help users distinguish what type of timeout
+	  occurred.  [ruby-trunk - Feature #6088]
+	* lib/net/pop.rb (module Net):  Updated documentation for ReadTimeout
+	  and OpenTimeout.
+	* lib/net/http.rb (module Net):  ditto
+	* lib/net/smtp.rb (module Net):  ditto
+	* lib/net/telnet.rb (module Net):  Net::ReadTimeout is now raised in
+	  waitfor to match Net::Protocol.
+	* test/net/http/test_http.rb:  Updated Timeout::Error expectation to
+	  Net::ReadTimeout.
+	* test/net/ftp/test_ftp.rb:  ditto
+
+Thu Apr 12 05:27:01 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/server.rb (module WEBrick::GenericServer):  A server
+	  will now continue only when a StandardError subclass is raised.  For
+	  other exception types the error will be logged at the fatal level and
+	  the server will safely stop.  Based on a patch by Alex Young.
+	  [ruby-trunk - Feature #6236]
+	* test/webrick/test_server.rb:  Test for new exception handling
+	  behavior.  Join the server thread instead of busy-waiting for it to
+	  shut down to remove race conditions.
+
+Thu Apr 12 03:50:44 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit:Runner::Worker#_run_suites):
+	  call GC.start before running the test suites.
+
+Wed Apr 11 22:31:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_check_id_cstr): new function to check if ID is
+	  registered with NUL-terminated C string.
+
+	* sprintf.c (rb_str_format): avoid inadvertent symbol creation.
+
+Wed Apr 11 20:28:36 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* io.c (rb_io_eof): use eof() instead of io_fillbuf(). It's because
+	  io_unread() doesn't work properly when reading CRLF with read(length)
+	  and mode 'r'.
+	  [ruby-core:44189][Bug #6271]
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#test_read_crlf_and_eof):
+	  test for above.
+
+Wed Apr 11 07:38:33 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/digest/sha2/lib/sha2.rb (Digest#block_length):  Fixed method name
+	  in documentation examples.  Patch by naleski via
+	  https://github.com/ruby/ruby/pull/115
+
+Wed Apr 11 07:33:13 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* pack.c (pack_pack):  Warn when an invalid character is found in the
+	  format string when $VERBOSE is true.  [ruby-trunk - Feature #5219]
+	* pack.c (pack_unpack):  ditto
+	* test/ruby/test_pack.rb (class TestPack):  Test for warnings on
+	  invalid format characters.
+
+Wed Apr 11 06:11:10 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* string.c (rb_str_tr):  Documented use of \ to escape characters.
+	  [ruby-trunk - Bug #6161]
+	* string.c (rb_str_count):  ditto
+
+Wed Apr 11 05:14:51 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/abbrev.rb:  Clarified that Abbrev.abbrev returns a Hash instead
+	  of an Array.  Patch by Andrei Bocan.  [ruby-trunk - Bug #6107]
+
+Wed Apr 11 03:02:24 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/ripper/lib/ripper/sexp.rb: fix spelling. patched by
+	  Jonathan Hinkle via https://github.com/ruby/ruby/pull/116
+
+Tue Apr 10 19:07:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_enc_raise): new function to raise an exception with
+	  the message in the given encoding.  patched by now (Nikolai
+	  Weibull) at [ruby-core:41160].  [Feature #5650]
+
+Tue Apr 10 18:19:32 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#send_request_with_body_stream):
+	  use IO.copy_stream for requests using body_stream.
+	  patched by Eric Wong. [ruby-core:40898] [Feature #5605]
+
+Tue Apr 10 16:53:21 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c: add prototype declarations for older Mac OS X.
+	  [ruby-core:43376][Bug #6170]
+
+Tue Apr 10 15:35:21 2012  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_set_sequence): show a hint if there are duplicated
+	  "when" clauses.  [ruby-core:41502] [ruby-trunk - Feature #5716]
+
+Tue Apr 10 09:57:00 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* string.c (rb_str_split_m):  Documented behavior of split on the empty
+	  string.  [ruby-trunk - Feature #3575]
+
+Tue Apr 10 09:48:31 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (rb_deflate_s_deflate):  Fixed ruby example replacing
+	  NO_FLUSH with FINISH.  [ruby-trunk - Bug #6273]
+
+Mon Apr  9 23:10:26 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (isUNCRoot, winnt_stat): support long UNC.
+	  [ruby-core:30623][Feature #3399]
+
+Mon Apr  9 15:16:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (string_content, parser_yylex): count brace nesting to
+	  dispatch embexpr_end.  [ruby-core:43775][Bug #6211]
+
+Mon Apr  9 13:06:58 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* hash.c (rb_hash_set_default_proc): Accept nil, patch by Run Paint
+	  [Feature #4234]
+
+	* test/ruby/test_hash.rb: test for above.
+
+Mon Apr  9 08:01:15 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: gets the value with range() consistently.
+	* ext/date/date_strftime.c (range): now just replaces the given item.
+
+Mon Apr  9 06:58:01 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c (nucomp_expt): [ruby-core:44170].
+
+Mon Apr  9 02:52:03 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* complex.c (nucomp_expt): the result of f_complex_new2 may be a fixnum
+	  with mathn. [ruby-core:44170] [Bug #6267]
+
+Sun Apr  8 22:46:01 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/generator/generator.c (generate_json_bignum):
+	  add RB_GC_GUARD.
+	 http://fb.rubyci.org/~chkbuild/ruby-trunk/log/20120407T210301Z.diff.html.gz
+
+Sun Apr  8 07:26:40 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler): get keys
+	  and fetch values from it to prevent @timeout_info's error
+	  "can't add a new key into hash during iteration".
+
+Sun Apr  8 06:51:57 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_unread): cast as long the value for extra_max.
+	  [ruby-core:44137] [Bug #6257]
+
+Sun Apr  8 06:46:48 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httpresponse.rb (WEBrick::HTTPResponse#send_body_io):
+	  use readpartial to get data even if the response is streaming data and
+	  each data is smaller than @buffer_size.
+	  patched by yu nobuoka. [ruby-dev:45471] [Bug #6230]
+
+Sat Apr  7 22:35:36 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* include/ruby/win32.h (rb_w32_aspawn_flags): add the declaration of
+	  new function.
+
+	* process.c (enum): add EXEC_OPTION_PGROUP and move the position
+	  above for the usage in proc_spawn_n().
+
+	* process.c (proc_spawn_n): add an argument to pass new option
+	  `new_pgroup`. The option specifies CREATE_NEW_PROCESS_GROUP flag to
+	  CreateProcessW(). This flag is necessary for the usage of
+	  Process.kill on the subprocess on Windows.
+
+	* process.c (rb_exec_arg_addopt): ditto.
+
+	* process.c (rb_spawn_process): ditto.
+
+	* process.c (documentation for rb_f_spawn): add documentation for new
+	  option `new_pgroup` of spawn.
+
+	* test/ruby/test_process.rb (TestProcess#test_execopts_new_pgroup):
+	  add tests for option `new_pgroup`.
+
+	* test/ruby/test_thread.rb
+	  (TestThreadGroup#test_thread_timer_and_interrupt):
+	  add option `new_pgroup: true` to spawn on Windows. It's needed for
+	  Process.kill on a subprocess.
+
+	* win32/win32.c (CreateChild): add an argument to pass
+	  dwCreationFlags of CreateProcessW().
+
+	* win32/win32.c (rb_w32_spawn): ditto.
+
+	* win32/win32.c (rb_w32_aspawn_flags): add new function to pass
+	  dwCreationFlags.
+
+	* win32/win32.c (rb_w32_aspawn): refactor to move the content to
+	  rb_w32_aspawn_flags().
+	  [ruby-core:43245][Bug #6131]
+
+Sat Apr  7 22:32:00 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_thread.rb
+	  (TestThreadGroup#test_thread_timer_and_interrupt): skip on Windows.
+	  Process.kill cannot kill a subprocess if CREATE_NEW_PROCESS_GROUP
+	  flag is not specified in a call to CreateProcessW().
+
+	* win32/win32.c (CreateChild): revert the usage of
+	  CREATE_NEW_PROCESS_GROUP flag for compatibility.
+	  [ruby-core:43245][Bug #6131]
+
+Sat Apr  7 10:28:40 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: bumping up psych version to match release.
+	* ext/psych/psych.gemspec: ditto
+
+Sat Apr  7 02:07:00 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c: fall back to any encoding if the external
+	  encoding is wrong.  [ruby-core:44163]
+	* test/psych/test_encoding.rb: fix test
+
+Fri Apr  6 16:24:24 2012  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* struct.c (documentation for rb_struct_members_m):
+	  fix 'array of strings' to 'array of symbols'
+	  [ruby-core:44152][Bug #6264]
+
+Fri Apr  6 14:27:04 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* Makefile.in ($(LIBRUBY_A)): fix typo.
+
+Thu Apr  5 13:26:15 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/alloca.c (xmalloc, xfree): use ruby version, not
+	  depending on RUBY_LIB_PREFIX.  [ruby-dev:45492][Bug #6255]
+
+Wed Apr  4 13:06:39 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/ftp/ftp.rb (Net::FTP#close): restore original read_timeout.
+
+Wed Apr  4 10:33:31 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/ftp/ftp.rb (Net::FTP#close): ignore exceptions from shutdown and
+	  read on closing.
+
+Wed Apr  4 01:48:35 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/ftp/ftp.rb (Net::FTP#close): close socket more gracefully.
+
+	* lib/ftp/ftp.rb (Net::BufferedSocket#shutdown): added.
+
+	* test/net/ftp/test_ftp.rb (FTPTest#create_ftp_server): wait socket
+	  with shutdown and read.
+
+Tue Apr  3 19:00:52 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/net/ftp/test_ftp.rb (FTPTest#create_ftp_server): should wait
+	  a little before closing socket because if the client call
+	  Net::FTP#getmultiline the socket is suddenly closed by the server in
+	  the getline loop.
+
+Tue Apr  3 18:33:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (setreuid, setregid): suppress warnings.
+	  [ruby-core:43374][Bug #6169]
+
+Tue Apr  3 10:18:27 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (inspect_enumerator): suppress uninitialized
+	  instance variable warnings.  [ruby-dev:45449][Bug #6214]
+	  patched by no6v (Nobuhiro IMAI).
+
+Mon Apr  2 13:25:08 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse/ac.rb: autoconf-like options.
+
+Mon Apr  2 10:34:00 2012  eregon  <eregontp@gmail.com>
+
+	* string.c (rb_str_start_with, rb_str_end_with): raise an error if
+	  an argument is not convertible to a String.
+	  [ruby-core:40623][Bug #5536]
+
+Mon Apr  2 03:35:25 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/server.rb (WEBrick::GenericServer): close socket only if
+	  the socket is not closed yet.
+
+Sun Apr  1 23:03:18 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/ftp.rb (Net::BufferedSocket): should delegate send() to @io
+	  for Net::FTP#abort and Net::FTP#status.
+
+Sun Apr  1 00:41:56 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb: fixed the domain name in examples.
+
+Sat Mar 31 21:39:45 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler): dup to prevent
+	  @timeout_info's "can't add a new key into hash during iteration".
+
+Sat Mar 31 14:22:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (hash_default_value): extract from rb_hash_aref(), to be
+	  shared with rb_hash_shift(), so that overriding Hash#default
+	  will be respected.
+
+Sat Mar 31 14:16:02 2012  Sokolov Yura (funny-falcon)  <funny.falcon@gmail.com>
+
+	* hash.c: do not allocate st_table when it is not necessary.
+
+Sat Mar 31 13:42:39 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (read_timeout=, open_timeout=): supported timeout.
+
+Sat Mar 31 13:20:40 2012  Sokolov Yura (funny-falcon)  <funny.falcon@gmail.com>
+
+	* hash.c: remove unnecessary checks for Qundef in hash iterations.
+	  since hash use st_foreach_check for iterations, such checks are
+	  needless.
+
+Sat Mar 31 12:05:01 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_x509cert.c: Fix doc typo.
+
+Sat Mar 31 10:13:24 2012  Sokolov Yura (funny-falcon)  <funny.falcon@gmail.com>
+
+	* st.c (st_foreach_check, st_foreach): remove ancient check.  This
+	  check are from initial ordered hash commit when first entry were
+	  created with entry->fore = entry->back = entry.
+
+	* st.c (st_delete): use real_entries in st_delete for packed tables
+
+Sat Mar 31 07:53:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_foreach_check): remove the entry by replacing with never
+	  when ST_DELETE.
+
+	* hash.c (st_foreach_safe): since table is not for VALUE, Qundef is
+	  not special value, so use 0 instead.  therefore this function can be
+	  applied to only st_table which 0 is invalid as keys, e.g., IDs.
+
+	* hash.c: Qundef cannot be passed from st_foreach_check().
+
+	* hash.c, marshal.c, object.c, variable.c: fix callback argument types
+	  of iterators.
+
+Thu Mar 29 23:50:15 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_update): pass pointer to key to the callback function.
+
+Thu Mar 29 16:36:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_update): add existing parameter to the callback function.
+
+Thu Mar 29 16:35:32 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (terminal_width, del_status_line, put_status):
+	  extract as methods.
+
+Thu Mar 29 10:20:18 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkcs7.c: fix crash when parsing garbage data.
+	* test/openssl/test_pkcs7.rb: assert correct behavior for it.
+	  Thanks to Matt Venables for reporting the issue.
+	  [ruby-core:43250][Bug #6134]
+
+Thu Mar 29 10:16:05 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread_win32.c (TIME_QUANTUM_USEC): 10ms(= old setting) [experimental]
+	  cf. [Bug #6098]
+
+Thu Mar 29 10:12:12 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (rb_threadptr_execute_interrupts_common): use defined
+	  TIME_QUANTUM_USEC instead of a magic number.  there is no meanings
+	  to use different values for checking interval of interruption and
+	  thread switching limits.
+	  cf. [Bug #6098]
+
+Thu Mar 29 09:26:17 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_x509cert.rb: exclude test that fails when issuing
+	  a certificate with RSA signature and DSS1 digest for earlier
+	  OpenSSL versions when used in conjunction with OpenSSL 1.0.1.
+	  Thanks, Vit Ondruch, for reporting the issue.
+	  [ruby-core:42949][Bug #6089]
+
+Thu Mar 29 08:25:35 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* NEWS: add note about unified behavior of encoding nil values in
+	  instances of OpenSSL::ASN1::ASN1Data.
+
+Thu Mar 29 07:45:36 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: raise TypeError when trying to encode nil
+	  values for Primitive instances.
+	* test/openssl/test_asn1.rb: Assert consistent behavior when
+	  encoding nil values: Primitives raise TypeError, Constructives
+	  raise NoMethodError.
+	  Fixes [ruby-core:43009][Bug #6102]
+
+Wed Mar 28 16:39:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (obj2uid, obj2gid): allow strings as input user/group id.
+	  [ruby-core:40923][Feature #5610]
+
+Wed Mar 28 15:06:18 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_symreal): default to ASCII-8BIT for non-ascii symbols,
+	  otherwise it should be converted to US-ASCII in rb_intern_str() if
+	  possible.  [ruby-core:43762][Bug #6209]
+
+Wed Mar 28 08:44:24 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: updating version to match gem
+	* ext/psych/psych.gemspec: ditto
+	* ext/psych/lib/psych/visitors/to_ruby.rb: fixing deprecation warning
+
+Tue Mar 27 23:44:11 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_unread): fixed memory leak.  report by nagachika via IRC.
+
+Tue Mar 27 22:44:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (verconf.h): separate load path specific stuff from
+	  config.h.
+
+Tue Mar 27 22:43:40 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub: fix config.h path to include.
+
+Tue Mar 27 17:08:08 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* win32/win32.c (check_if_dir): fix memory leak.
+
+Tue Mar 27 13:13:51 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_new_empty): should copy also the encoding as an
+	  empty substring.  [ruby-dev:45441][Bug #6206]
+
+Mon Mar 26 23:43:04 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (parse227, parse228, parse229): don't use $~.
+
+Mon Mar 26 23:34:40 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (parse227, parse228, parse229): don't use local
+	  variables defined by named capture for other Ruby implementations
+	  such as Rubinius.
+
+Mon Mar 26 23:19:03 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (parse_pasv_port): refactored.
+
+Mon Mar 26 19:49:49 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/net/ftp/test_ftp.rb: add the test, which was forgotten in the
+	  previous commit.
+
+Mon Mar 26 19:37:27 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (parse227, parse228, parse229): refactored.
+
+Mon Mar 26 11:46:23 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (inspect_enumerator): show method arguments of
+	  lazy enumerators correctly.
+
+Mon Mar 26 13:51:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (check_if_dir, check_if_wdir): fix for Visual C++
+	  not to use S_ISDIR().  [Feature #2408][ruby-core:26925]
+
+	* ruby.c (load_file_internal): ditto.
+
+Mon Mar 26 11:46:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): bail out if the script is a directory.
+	  [Feature #2408][ruby-core:26925]
+
+	* win32/win32.c (rb_w32_open, rb_w32_wopen): check if the file is a
+	  directory when access denied, to set errno to EISDIR.
+
+Sun Mar 25 18:13:14 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (tr_setup_table): fix multiple non latin argument for
+	  non latin (over 256 characters) tr-like methods.
+	  [ruby-core:43371] [Bug #6167]
+
+Sun Mar 25 00:46:06 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator (lazy_initialize): set the instance variable "receiver"
+	  to include the receiver to the return value of inspect on a lazy
+	  enumerator directly created by Enumerator::Lazy.new.
+
+	* enumerator (RETURN_LAZY): don't set the instance variable "receiver".
+
+Sat Mar 24 23:59:00 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator (enumerator_inspect): include the original receiver and
+	  method name of Enumerator::Lazy in the result of inspect.
+	  [ruby-core:43345] [Bug #6159]
+
+	* enumerator (InitVM_Enumerator): don't use rb_define_alias for
+	  some methods such as collect in order to make rb_frame_this_func()
+	  return the correct method names.
+
+Sat Mar 24 22:22:18 2012  Sambasiva Rao Suda  <sambasivarao@gmail.org>
+
+	* time.c (time_init_1): Time.new will accept seconds as string or
+	  int.  [ruby-core:43569][Bug #6193]
+
+Fri Mar 23 15:12:12 2012  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c (documentation for str_encode): Explain
+	  that transcoding to the same encoding is a no-op
+	  (i.e. no exceptions, no replacements,...).
+	  [ruby-core:43557][Bug #6190]
+
+Fri Mar 23 13:19:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_str_to_inum): must be ASCII compatible encoding as
+	  well as String#hex and String#oct.  [ruby-core:43566][Bug #6192]
+
+	* string.c (rb_must_asciicompat): check if ASCII compatible.
+
+Thu Mar 22 23:14:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (str_encode_bang, encoded_dup): if nothing was
+	  transcoded, just set encoding but leave coderange unchanged as
+	  force_encoding.  [ruby-core:43557][Bug #6190]
+
+Thu Mar 22 22:30:44 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* io.c (static int io_fflush): add the definition.
+	  Use it in set_binary_mode_with_seek_cur().
+
+	* io.c (set_binary_mode_with_seek_cur): refactoring to split the
+	  content into io_unread(). Fix the possibility of buffer overflow.
+
+	* io.c (io_unread): add new implementation for Windows. Previous one
+	  caused invalid cursor position using IO#pos with OS text mode. New
+	  one fixes the bug.
+
+	* test/ruby/test_io_m17n.rb
+	  (TestIO_M17N#test_pos_dont_move_cursor_position): add a test for
+	  above bug.
+	  [ruby-core:43497] [Bug #6179]
+
+Thu Mar 22 19:55:08 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_fstat, rb_w32_fstati64): convert FILETIME
+	  to time_t directly, not to be affected by TZ unnecessarily.
+
+	* win32/win32.c (unixtime_to_filetime): convert time_t to FILETIME
+	  simply.
+
+Thu Mar 22 13:43:31 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey_rsa.c (rsa_generate): fix argument type.
+	  [Bug #6094]
+
+Thu Mar 22 11:14:10 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#test_pos_with_getc): updated.
+	  see [ruby-core:43550]
+
+Wed Mar 21 17:57:57 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c: Merge Onigmo 3d855b30d574536d3ae600260208c6624ae4791c.
+	  [Bug#6143] [Bug#6144] [Bug#6145]
+
+Wed Mar 21 17:01:55 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#test_pos_with_getc): added.
+	  see [Bug #6179][ruby-core:43518]
+
+Mon Mar 19 17:18:51 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_flat_map_func): convert the block value to
+	  Array if it doesn't respond to each.  [ruby-core:43334]
+	  [Bug #6155]
+
+Mon Mar 19 16:34:14 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enum.c (zip_i): variadic argument needs explicit cast on the
+	  platforms where VALUE is longer than int.
+
+Mon Mar 19 15:36:41 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (enumerable_lazy): add an example of take and first
+	  to the documentation.  [ruby-core:43344] [Bug #6158]
+	  add the description of the behavior when a block is given to zip
+	  or cycle.
+
+Mon Mar 19 15:20:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_specialized_instruction): DRY and replace chain
+	  of if-else with switch for special instructions.  based on a
+	  patch by Vasfed.  https://github.com/ruby/ruby/pull/105
+
+Mon Mar 19 15:05:54 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/test_pty.rb: same as r29280, skip tests when PTY allocation
+	  failed (that's not our fault).
+
+Sun Mar 18 23:21:17 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (aligned_free): fix condition for free.  memalign() and
+	  posix_memalign() are not defined together normally.
+
+Sun Mar 18 18:31:45 2012  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* gc.c (aligned_malloc, aligned_free): added fallback implementations
+	  for platforms like OSX Leopard.
+
+Sun Mar 18 17:17:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_pow): estimate result bit size more precisely.
+	  [ruby-core:30735][Feature #3429]
+
+Sun Mar 18 17:17:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (free_method_entry_i): method entry may be in
+	  unlinked_method_entry_list.  [ruby-core:43383][Bug #6171]
+
+Sun Mar 18 15:27:31 2012  Tanaka Akira  <akr@fsij.org>
+
+	* compile.c: typo fix by Run Paint Run Run.
+	  [ruby-core:28368] [Bug #2824]
+
+Sun Mar 18 10:01:02 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/profiler.rb: support calling singleton methods of
+	  an instance of BasicObject.
+
+Sat Mar 17 06:56:58 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c:  Fix indentation of Class#inherited example.
+
+Sat Mar 17 01:46:05 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* string.c (trnext): fix bug with string ending with '\\'.
+	  [ruby-dev:45374][Bug #6160]
+
+	* test/ruby/test_string.rb (TestString#test_delete): test for
+	  above.
+
+Fri Mar 16 20:06:24 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (trnext): should advance char-wise.
+	  [ruby-core:43335][Bug #6156]
+
+Fri Mar 16 17:42:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (block_append_gen): fix unreachable warning line number.
+	  should warn at the code, not jump.
+
+Fri Mar 16 17:33:08 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_take): allocate buffer array before iteration, as well
+	  as enum_first did.
+
+	* enum.c (enum_first): remove duplication.
+
+Fri Mar 16 14:43:18 2012  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* load.c (ruby_init_ext): don't free the given pointer itself.
+	  It is not guaranteed even that the pointer is on heap.
+
+Fri Mar 16 14:37:57 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_eval.c (rb_mod_module_eval): fix the documentation of
+	  class_eval to mention class variable lookup.  [ruby-core:40649]
+	  [Bug #5544]
+
+Fri Mar 16 14:27:11 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_eval.c (rb_mod_module_eval): fix the documentation of
+	  class_eval to mention constant lookup.  [ruby-core:41718]
+	  [Bug #5777]
+
+Fri Mar 16 14:10:45 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (initialize): raise Net::IMAP::Error when the
+	  connection is closed without a greeting response.
+	  [ruby-core:40938] [Bug #5616]
+
+Fri Mar 16 13:50:12 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (rfc822_text): ignore [] after RFC822.
+	  [ruby-core:40945] [Bug #5620]
+
+Fri Mar 16 12:00:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (argument_error): use line number at the beginning
+	  of lambda, not the first code of its body.
+	  [ruby-core:43314][Bug #6151]
+
+	* iseq.c (rb_iseq_first_lineno): constified.
+
+Fri Mar 16 11:20:07 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_take): don't enumerate an extra value.
+	  [ruby-dev:45370] [Bug #6152]
+
+Fri Mar 16 06:30:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_zip_func): variadic argument needs explicit cast
+	  on the platforms where VALUE is longer than int.
+
+	* enumerator.c (lazy_init_iterator): no need to check overflow twice.
+
+Fri Mar 16 05:47:09 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* enumerator.c (lazy_init_iterator):  Fix type error (int vs long).
+
+Thu Mar 15 23:13:36 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enum.c (rb_enum_values_pack): rename from enum_values_pack, and
+	  remove static.
+
+	* enumerator.c (lazy_init_iterator, lazy_init_yielder,
+	  lazy_select_func, lazy_reject_func, lazy_grep_func): handle
+	  multiple values correctly.
+
+	* enumerator.c (lazy_grep): change the behavior when a block is
+	  given, to be consistent with Enumerable#grep.
+
+Thu Mar 15 19:12:31 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_zip): rescue StopIteration returned by
+	  Enumerator#next.
+
+Thu Mar 15 18:19:53 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_zip, lazy_cycle): Enumerator::Lazy#{zip,cycle}
+	  should be eager when a block is given, to be consistent with
+	  Enumerable#{zip,cycle}.
+
+Thu Mar 15 17:45:27 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (InitVM_Enumerator): renamed Enumerable::Lazy to
+	  Enumerator::Lazy.
+
+Thu Mar 15 16:37:38 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (enumerable_lazy): added cycle to the documentation.
+
+Thu Mar 15 15:37:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): fix warning line number.
+
+Thu Mar 15 15:19:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_cycle): check argument number overflow before
+	  creating temporary array.
+
+Thu Mar 15 15:04:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strtod): no need to check same digit for hexdigit
+	  twice.  [ruby-dev:45363][Bug #6146]
+
+	* parse.y (sym_check_asciionly): check ascii compatibility before
+	  scanning for code range.
+
+	* parse.y (intern_str): set to us-ascii if ascii only.
+	  [ruby-dev:45363][Bug #6146]
+
+	* file.c (ruby_enc_find_basename): allow NULL as alllen.
+	  [ruby-dev:45363][Bug #6146]
+
+Thu Mar 15 14:49:31 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_conv_enc_opts): default to original encoding.
+
+Thu Mar 15 13:47:17 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (env_str_new, rb_f_getenv, env_fetch): use rb_str_conv_enc()
+	  instead of rb_str_encode() to simplify the code.
+
+Thu Mar 15 12:44:50 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_ugetenv): new API to
+	  accept and to return UTF-8 strings.
+
+	* win32/win32.c (rb_w32_getenv): follow above change.
+
+	* win32/win32.c (rb_w32_get_environ): returns UTF-8 environment area.
+
+	* hash.c (env_str_new, rb_f_getenv, env_fetch): follow above changes.
+	  [Bug #5570] [ruby-core:40737]
+
+Thu Mar 15 10:57:27 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_cycle): add Enumerable::Lazy#cycle.
+
+Thu Mar 15 10:31:40 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_arity.rb (TestArity#err_mess): use assert_raise.
+
+Thu Mar 15 07:03:52 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* vm_eval.c (check_funcall):  Raise ArgumentError if respond_to?
+	  requires more than three arguments. [Bug #6000]
+
+	* test/ruby/test_object.rb (class TestObject):  Test for respond_to?
+	  requiring more than three arguments.
+
+Thu Mar 15 06:08:06 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* include/ruby/intern.h: Add rb_check_arity, rb_error_arity [#6085]
+
+	* array.c: Use rb_check_arity / rb_error_arity
+
+	* class.c: ditto
+
+	* enumerator.c: ditto
+
+	* eval.c: ditto
+
+	* file.c: ditto
+
+	* hash.c: ditto
+
+	* numeric.c: ditto
+
+	* proc.c: ditto
+
+	* process.c: ditto
+
+	* random.c: ditto
+
+	* re.c: ditto
+
+	* signal.c: ditto
+
+	* string.c: ditto
+
+	* struct.c: ditto
+
+	* transcode.c: ditto
+
+	* vm_eval.c: ditto
+
+	* vm_insnhelper.c: ditto & implementation of rb_error_arity
+
+	* test/ruby/test_arity.rb: tests for above
+
+Thu Mar 15 06:08:05 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* vm_insnhelper.c: improve number of arguments error in case of
+	  optional parameters (issue #6085)
+
+	* include/ruby/intern.h: define UNLIMITED_ARGUMENTS
+
+	* test/ruby/test_arity.rb:  test for above
+
+Thu Mar 15 00:58:04 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (enumerable_lazy): fix the documentation of
+	  Enumerable#lazy.
+
+Wed Mar 14 22:01:06 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_init_iterator): break when Qundef is returned
+	  to make obj.drop(3).take(2) work properly.
+
+	* enumerator.c (lazy_take_while): add Enumerable::Lazy#take_while.
+
+	* enumerator.c (lazy_drop): add Enumerable::Lazy#drop.
+
+	* enumerator.c (lazy_drop_while): add Enumerable::Lazy#drop_while.
+
+	* enumerator.c (InitVM_Enumerator): add Enumerable::Lazy#force as an
+	  alias of to_a.
+
+Wed Mar 14 19:28:40 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_take): add Enumerable::Lazy#take.
+
+Wed Mar 14 18:40:36 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c: use long for array indices.
+
+Wed Mar 14 18:25:18 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c: moved the comment of StopIteration.
+
+Wed Mar 14 17:55:29 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (flodivmod): must go through the same pass if HAVE_FMOD or
+	  not.  this is a bugfix of r35013.
+
+Wed Mar 14 16:41:55 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/test_tmpdir.rb (TestTmpdir#test_world_writable): skip on Windows.
+
+Wed Mar 14 15:09:23 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c: fix flodivmod for cornercases [Bug #6044]
+	  add ruby_float_mod
+
+	* insns.def (opt_mod): use ruby_float_mod
+
+	* internal.h: declare ruby_float_mod
+
+	* test/ruby/test_float.rb: tests for above
+
+	* test/ruby/envutil.rb: create helper assert_is_minus_zero
+
+Wed Mar 14 10:44:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_grep_func): should use === instead of =~, as
+	  well as Enumerable#grep
+
+Wed Mar 14 08:15:54 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_flat_map_func): use each for non-Array objects.
+
+Wed Mar 14 08:06:35 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_zip): add Enumerable::Lazy#zip.
+
+	* enumerator.c (lazy_lazy): just returns self.
+
+Wed Mar 14 07:48:36 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (datetime_s_now): [ruby-core:43256].
+
+Tue Mar 13 22:00:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_arguments): keyword rest arg without keyword args.
+
+	* node.c (dump_node): dump kw_rest_arg too.
+
+	* parse.y (block_param, f_arg): more kwrest patterns.
+	  [ruby-core:42455][Bug #5989]
+
+	* parse.y (new_args_gen): no extra kw_rest_arg if no keyword rest arg.
+
+Tue Mar 13 15:17:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (block_param, f_args): add rules for the case arguments
+	  begin with kwrest.  [ruby-core:42455][Bug #5989]
+
+Tue Mar 13 12:37:53 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_encoding_set): always warn if external encoding and internal
+	  encoding are identical. [ruby-core:40727] [Bug #5568]
+
+Tue Mar 13 12:37:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c: add ObjectSpace::WeakMap.  [ruby-dev:44565][Bug #5350]
+
+	* lib/weakref.rb: use WeakMap instead of _id2ref.
+
+Tue Mar 13 10:59:48 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (prepare): skip if basedir is not defined.
+	  [ruby-core:39135][Bug #5238]
+
+	* tool/rbinstall.rb (CONFIG.[]): check for mandatory
+	  configurations.
+
+Tue Mar 13 00:09:18 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (enumerable_lazy): added documentation.
+
+Mon Mar 12 20:19:25 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb (Dir::tmpdir): test the current directory suitable for
+	  temporary directory.
+
+Mon Mar 12 20:08:16 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/fileutils.rb (fu_have_symlink?): specify TypeError for rescue
+	  clause.
+
+Mon Mar 12 19:23:13 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_find_encoding): new function find encoding from
+	  arbitrary object as a pointer to rb_encoding, and return NULL if
+	  not found.
+
+	* io.c (io_encoding_set): just warn unsupported encodings, but not
+	  exception.  [ruby-core:40726] [Bug #5567]
+
+Mon Mar 12 19:03:32 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (Init_eval_method): respond_to? and
+	  respond_to_missing? are public.
+
+Mon Mar 12 14:56:52 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.h (NEW_YIELD), parse.y (new_yield_gen): array-values flags
+	  has been already obsolete.  patch by Thomas Enebo.
+	  [ruby-core:41929][Bug #5847]
+
+Mon Mar 12 12:44:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (Init_eval_method): copy basic methods to Exception.
+	  [ruby-core:40287][Bug #5473]
+
+Mon Mar 12 10:13:36 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_jump.c (rb_exec_end_proc): remember the latest exit status.
+	  [ruby-core:43173][Bug #5218]
+
+Mon Mar 12 07:33:12 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb: update document for changing
+	  FileUtils.remove_entry_secure to FileUtils.remove_entry.
+
+	* NEWS: add incompatibility note for lib/tmpdir.rb.
+
+Mon Mar 12 07:19:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tmpdir.rb (Dir.tmpdir): should not use world-writable but
+	  non-sticky directory.
+
+	* lib/tmpdir.rb (Dir.mktmpdir): check the parent directory.
+
+Mon Mar 12 07:04:11 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (Init_Random): removed rb_Random_DEFAULT and register as
+	  mark-object instead of global variable.
+
+Mon Mar 12 07:03:32 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_s_rand): ensure default PRNG is re-initialized
+	  after fork.  patched by Eric Wong.  [ruby-core:41209][Bug #5661]
+
+Sun Mar 11 23:57:29 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_unpack): when unpack('M') occurs an illegal byte
+	  sequence, output the "=" character and the following character in
+	  the decoded data without any transformation.
+	  [ruby-dev:44875] [Bug #5635]
+
+Sun Mar 11 22:32:43 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Merge 164a75c8bd2007d32c4d7665d53140d8fc126dcd.
+	  [ruby-core:41917] [Bug #5846]
+
+Sun Mar 11 17:10:04 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Put error message into STDERR if failed to launch
+	  worker (job) process. [ruby-dev:44802] [Bug #5577]
+
+	* lib/test/unit/parallel.rb: If failed to increment_io, exit with code
+	  2. [ruby-dev:44802] [Bug #5577]
+
+Sun Mar 11 15:46:45 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* io.c: fix rdoc of `IO.binwrite` to show same as `IO.write` except
+	  it opens file with mode "wb:ASCII-8BIT". [Bug #5782] [ruby-core:42592]
+
+Sat Mar 10 23:52:28 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c: pack tables also generic keys.  patched by Sokolov Yura at
+	  https://github.com/ruby/ruby/pull/84
+
+	* st.c: add st_foreach_check for fixing iteration over packed table
+	  and st_delete_safe.  patched by Sokolov Yura at
+	  https://github.com/ruby/ruby/pull/84
+
+	* st.c: fix packed num_entries on delete_safe.  patched by Sokolov
+	  Yura at https://github.com/ruby/ruby/pull/84
+
+Fri Mar  9 14:29:32 2012  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enumerator.c (lazy_flat_map): add Enumerable::Lazy#flat_map.
+
+Fri Mar  9 06:29:22 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb (load, parse): stop parsing or loading after
+	  the first document has been parsed.
+
+	* test/psych/test_stream.rb: pertinent tests.
+
+Fri Mar  9 06:17:05 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb (parse_stream, load_stream): if a block is
+	  given, documents will be yielded to the block as they are parsed.
+	  [ruby-core:42404] [Bug #5978]
+
+	* ext/psych/lib/psych/handlers/document_stream.rb: add a handler that
+	  yields documents as they are parsed
+
+	* test/psych/test_stream.rb: corresponding tests.
+
+Fri Mar  9 00:35:03 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_initialize, enumerable_lazy): no additional
+	  arguments.
+
+Fri Mar  9 00:30:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c: add Enumerable#lazy. based on the patch by
+	  Innokenty Mikhailov at <https://github.com/ruby/ruby/pull/101>
+	  [ruby-core:37164] [Feature #4890]
+
+Fri Mar  9 00:25:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_each, generator_each): pass arguments to
+	  the block with yielder.
+
+Fri Mar  9 00:25:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_cat): new function to concat objects into array.
+
+Thu Mar  8 16:44:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .gdbinit (rb_numtable_entry): update for recent refactoring of
+	  st_table.
+
+Wed Mar  7 22:41:50 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* lib/xmlrpc/client.rb (module XMLRPC): fix typo.
+
+	* test/xmlrpc/test_client.rb (test_async_call): add test for
+	  XMLRPC::Client#call_async to check above fix.
+
+Wed Mar  7 16:30:24 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_load_fail): should honor encoding.
+
+	* load.c (load_failed): ditto.
+
+Wed Mar  7 12:26:25 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_load_fail): use path as a string, not char*.
+
+	* internal.h: (rb_load_fail): moved from ruby/intern.h.
+
+	* ruby.c (load_file_internal): fname cannot be NULL.
+
+Wed Mar  7 08:32:43 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* error.c (rb_loaderror_with_path): Adding the missing file as an
+	  instance variable to the LoadError exception.
+	  [ruby-core:39079]
+
+	* load.c: call rb_loaderror_with_path so that the missing path is
+	  added to the exception.
+
+	* ruby.c: call rb_loaderror rather than raising our own LoadError
+	  exception.
+
+	* include/ruby/intern.h: add declaration for rb_loaderror_with_path.
+
+	* test/ruby/test_require.rb: add supporting test for LoadError#path
+	  method.
+
+Wed Mar  7 08:28:00 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/xmlrpc/parser.rb: support i8 types. Thanks Stas Kelvich!
+	  [ruby-core:29246] [Feature #3090]
+
+	* test/xmlrpc/test_client.rb: supporting test
+
+Wed Mar  7 07:43:29 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/xmlrpc/client.rb: assume servers that do not send a Content-Type
+	  header are sending 'text/xml'. Thanks Nathan Leavitt!
+	  [ruby-core:41204] [Bug #5660]
+
+	* test/xmlrpc/test_client.rb: supporting test
+
+Wed Mar  7 07:39:28 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* test/xmlrpc/test_client.rb: adding a test for performing an XMLRPC
+	  call.
+	* test/xmlrpc/data/blog.xml: supporting XML document for the response.
+
+Tue Mar  6 16:24:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_string): escape simple regexp meta
+	  character terminators.
+
+Tue Mar  6 10:11:43 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (set_rawmode): clear ECHOE and ECHOK
+	  bits too.
+
+	* ext/io/console/console.c (echo_p): ignore ECHOE and ECHOK bits.
+	  [ruby-dev:45309] [Bug #6116]
+
+	* ext/io/console/console.c (console_raw): fix rdoc.
+
+	* ext/io/console/console.c (console_set_echo): mentioned about
+	  platform dependency.
+
+Tue Mar  6 07:18:10 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/xmlrpc/client.rb: switch net/http post2 calls to modern
+	  `request_post` methods.
+
+Tue Mar  6 02:31:20 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/core_ext.rb: only extend Kernel if IRB is loaded
+	  in order to stop method pollution.
+
+Tue Mar  6 01:34:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (block_call): rules for block_call after block_call.
+	  based on a patch by pasberth https://github.com/ruby/ruby/pull/102
+	  [ruby-dev:45308][Bug #6115]
+
+Tue Mar  6 01:24:13 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (block_command, block_call): simplified rules.
+
+Mon Mar  5 18:28:35 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_regexp.rb (TestRegexp#test_source): fix typo.
+	* test/ruby/test_regexp.rb (TestRegexp#test_equal): ditto.
+
+Mon Mar  5 17:11:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/lib/syck/rubytypes.rb (Exception.yaml_new): fix bug
+	  that causes YAML serialization problem for Exception.
+	  Exception#initialize doesn't use visible instance variable for
+	  the exception message, so call the method with the message.
+	  patched by Jingwen Owen Ou <jingweno AT gmail.com>.
+	  http://github.com/ruby/ruby/pull/41
+
+Mon Mar  5 16:50:22 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_sleep.rb (TestSleep#test_sleep_5sec): syntax error.
+
+	* test/ruby/test_sleep.rb (TestSleep#test_sleep_5sec): call uname
+	  only on linux because it's a workaround for linux only.
+
+Mon Mar  5 12:44:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (unpack_entries): chain entries directly.  based on a patch
+	  by Sokolov Yura <funny.falcon AT gmail.com>.
+
+	* st.c (unpack_entries): use union instead of casted pointer.
+	  patched by Sokolov Yura <funny.falcon AT gmail.com>.
+
+	* st.c: use PACKED_ENT and FIND_ENTRY.  patched by Sokolov
+	  Yura <funny.falcon AT gmail.com>.
+
+	* st.c (unpack_entries): reallocate bins if packed array size
+	  is not same as initial bins size.  based on a patch by
+	  Sokolov Yura <funny.falcon AT gmail.com>.
+
+Mon Mar  5 11:51:48 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb: remove description about
+	  BigMath#log. patched by Sho Hashimoto [ruby-dev:45307] [Bug #6112]
+
+	* string.c (str_byteslice): fix typo.
+
+Sun Mar  4 23:21:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_string): regexp engine doesn't need
+	  terminators to be escaped.  [ruby-core:40364][Bug #5484]
+
+Sat Mar  3 22:51:46 2012  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_run_exec_options_err): chdir at last to interpret
+	  relative pathnames from the current directory of the parent process.
+
+Sat Mar  3 12:20:44 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strftime.c: reassigned some variables.
+
+Sat Mar  3 12:12:16 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_{parse,strptime}.c: [ruby-dev:45303].
+
+Sat Mar  3 10:09:21 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/xmlrpc/client.rb (initialize): net/http defaults to 1_2 in 1.8+,
+	  so we can safely remove the call to enable it.
+
+Sat Mar  3 08:42:25 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/xmlrpc/client.rb (new2): use URI for uri parsing.
+	* test/xmlrpc/test_client.rb: test that query params are passed to the
+	  client constructor.
+
+Sat Mar  3 08:20:10 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/xmlrpc/client.rb (new2): raises an ArgumentError on bad
+	  arguments.
+	* test/xmlrpc/test_client.rb: tests for bad uris
+
+Sat Mar  3 08:08:11 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/xmlrpc/client.rb (new2): fix custom port specification when an
+	  SSL uri is used.
+	* test/xmlrpc/test_client.rb: tests for XMLRPC::Client.new2
+
+Sat Mar  3 08:03:29 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/syck/rubyext.c (mktime_do): use ISDIGIT().
+	  [ruby-core:43060] [Bug #6108]
+
+	* ext/syck/token.c (sycklex_yaml_utf8): cast as unsigned char.
+	  [ruby-core:43060] [Bug #6108]
+
+Sat Mar  3 06:57:14 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ruby_pc): make configurable.  [Bug #6051]
+
+Fri Mar  2 17:49:03 2012  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* .travis.yml (branches): Enable TravisCI for ruby_1_9_3.
+
+Fri Mar  2 17:13:33 2012  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* test/ruby/test_array.rb (test_combination2): Make the test case for
+	  [ruby-core:29240] more descriptive.
+	  cf. http://bugs.jruby.org/6518
+
+Fri Mar  2 16:37:11 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (file_expand_path): use wcscasecmp().
+
+Fri Mar  2 16:36:31 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (native_cond_timeout): cast explicitly to suppress
+	  a warning.
+
+Fri Mar  2 16:35:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (pipe_open): cmd is no longer used if fork is available.
+
+Thu Mar  1 16:13:18 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (rb_file_const, rb_file_load_ok): moved functions for
+	  internal use only.
+
+Thu Mar  1 15:40:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/makedirs.bat: new command to make intermediate
+	  directories, and not to report any errors if the directory
+	  already exists.
+
+	* win32/Makefile.sub (MAKEDIRS): enable command extensions.
+
+Thu Mar  1 01:25:43 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (is_onechar_cclass): optimize character class
+	  Merge Onigmo 27278c12e6674043cc8affca6507e20e119a86ee.
+
+	* regparse.c (is_onechar_cclass): [bug] unexpected match occurs when a
+	  char class contains no char
+
+	* enc/unicode.c (init_case_fold_table): define the sizes of case
+	  folding tables in casefold.h
+
+Wed Feb 29 16:11:34 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (MAKEDIRS): use mkdir of cmd.exe instead of ruby.
+	  [Bug #6103] [ruby-core:43012]
+
+	* win32/README.win32: added a notice about command extension of cmd.exe.
+
+Wed Feb 29 15:39:39 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#test_readpartial_locktmp): skip on
+	  windows because of the platform restriction.
+
+Wed Feb 29 15:38:50 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/memory_status.rb (Memory): syntax error.
+
+Wed Feb 29 13:06:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/memory_status.rb: use /proc/self/status if it is in
+	  the expected format.
+
+Wed Feb 29 06:14:51 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: reverted r34825.
+
+Tue Feb 28 23:20:01 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* Makefile.in (PLATFORM_DIR): add a variable for `win32` directory.
+	* Makefile.in (clean-platform): add new target.
+	  It cleans `win32` directory.
+
+	* common.mk (clean): add a dependency for `win32` directory.
+	* common.mk (distclean): ditto.
+	* common.mk (distclean-platform): add new target.
+	  It cleans `win32` directory.
+	* common.mk ($(PLATFORM_D)): add new target to make `win32` directory.
+	* common.mk (win32/win32.$(OBJEXT)): move win32.o into `win32`
+	  directory.
+	* common.mk (win32/file.$(OBJEXT)): add new target for win32/file.c.
+
+	* configure.in: move win32.o into `win32` directory and add
+	  win32/file.o to MISSING.
+
+	* file.c (file_load_ok, rb_file_load_ok): replace static
+	  file_load_ok() with public rb_file_load_ok().
+	  It's to link Windows implementation in win32/file.c.
+	* file.c (rb_find_file_ext_safe): ditto.
+	* file.c (rb_find_file_safe): ditto.
+
+	* win32/file.c (rb_file_load_ok): new file. Add Windows specific
+	  optimized implementation of rb_file_load_ok(). We created a
+	  separated file to avoid too many #ifdef macro which is unreadable.
+
+	* win32/Makefile.sub (PLATFORM_DIR): add a variable for `win32`
+	  directory.
+	* win32/Makefile.sub (MISSING): move win32.obj into `win32`
+	  directory and add win32/file.obj to MISSING.
+	* win32/Makefile.sub (MAKEDIRS): replace MINIRUBY with BASERUBY.
+	  It's because miniruby doesn't exist when making `win32` directory.
+	* win32/Makefile.sub (clean-platform): add new target to clean `win32`
+	  directory.
+	* win32/Makefile.sub ({$(srcdir)}.c{}.obj): make it not match
+	  win32/file.c to build properly.
+	* win32/Makefile.sub (win32/win32.$(OBJEXT)): move win32.obj into
+	 `win32` directory.
+
+	  Patch created with Luis Lavena.
+	  [ruby-core:42480] [Feature #5999]
+
+Tue Feb 28 20:27:25 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: [ruby-core:42998]
+
+Tue Feb 28 18:47:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_binwrite, rb_io_syswrite): use shared frozen source
+	  strings.
+
+	* io.c (io_fread, io_getpartial, rb_io_sysread): set buffer size
+	  after check if readable, which can cause thread switch.
+	  [ruby-dev:45297][Bug #6099]
+
+Tue Feb 28 17:16:01 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/time.rb (Time#xmlschema): use strftime specifiers instead of
+	  fractional exponential calculation which yields undesirable
+	  result.  [ruby-core:42997][Bug #6100]
+
+Tue Feb 28 14:15:29 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/protocol.rb: Add OpenTimeout subclass of Timeout::Error
+	* lib/net/pop.rb:  Modernize Timeout usage.  Patch by Eric Wong.
+	  Use Net::OpenTimeout instead of Timeout::Error.  [Bug #5765]
+	* lib/net/http.rb:  ditto
+	* lib/net/smtp.rb:  ditto
+	* lib/net/telnet.rb:  ditto
+
+Tue Feb 28 13:51:12 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Retry HTTP requests for additional network errors.
+	  Introduce OpenTimeout subclass of Timeout::Error.  [Bug #6001]
+	* test/net/http/test_http.rb:  Reduce timeout to 0.01s for faster test
+	* test/net/http/test_https.rb:  ditto
+
+Tue Feb 28 11:44:49 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (debugflags): check if -ggdb is accepted.
+	  [ruby-core:42875][Bug #6080]
+
+Tue Feb 28 10:28:51 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: default open YAML files with utf8 external
+	  encoding. [ruby-core:42967]
+	* test/psych/test_tainted.rb: ditto
+
+Mon Feb 27 23:46:09 2012  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (opt_bv_decl): allow newline at the end.  [ruby-dev:45292]
+
+Mon Feb 27 20:43:05 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_set_pos): add rdoc about textmode.
+
+	* test/ruby/test_io.rb (TestIO#test_setpos): use binmode.
+
+Mon Feb 27 17:00:15 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c (rb_str_crypt): Update rdoc and state that this
+	  function is system dependent.  Reviewed by nobu, thanks to
+	  @takai.
+
+Mon Feb 27 17:03:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValueWithPrec): since methods
+	  can be overridden, so should not make an assumption on the type
+	  of results.  [ruby-core:42969][Bug #6093]
+
+Mon Feb 27 10:54:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_cppflags, try_cflags, try_ldflags): replace the
+	  target flags if the given flag is accepted.
+
+Mon Feb 27 10:53:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_specification.rb (test_self_from_yaml_syck_default_key_bug):
+	  ignore the test for too old versions.
+
+Mon Feb 27 10:53:12 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit.rb (Test::Unit::Runner#puke): skips with no
+	  messages should be trivial.
+
+Mon Feb 27 10:50:23 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c, process.c, time.c, ext: use rb_sys_fail_str instead of
+	  rb_sys_fail.
+
+Mon Feb 27 10:48:49 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/extconf.rb: suppress useless deprecation warnings
+	  from OpenSSL added by Apple.
+
+Sun Feb 26 23:29:49 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (add_code_range_to_buf0): wrong condition of duplicated
+	  warnings.
+
+Sun Feb 26 11:26:44 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): call on special object instead of
+	  self.  since stabby lambda is a syntax, so it should not be
+	  affected by the context.  [ruby-core:42349][Bug #5966]
+
+	* insns.def (send): no special deal for FCALL.  self should be put
+	  on TOS instead.
+
+Sun Feb 26 05:35:43 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* error.c (report_bug): use buf and snprintf to avoid consuming stack.
+	  [ruby-dev:45272] [Bug #6058]
+
+Sat Feb 25 17:41:19 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb (headers): try ambiguous headers at last.
+
+Sat Feb 25 17:07:15 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/fileutils.rb: use chomp(?/) instead of sub to optimize and avoid
+	  to regexping invalid string.
+
+Sat Feb 25 16:18:24 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* complex.c (nucomp_marshal_load): raise error on invalid data.
+	  reported by John Firebaugh [ruby-core:42860] [Bug #6076]
+
+Sat Feb 25 14:46:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/dl.c (Init_dl): support intrinsic types, size_t, ptrdiff_t
+	  and intptr_t.  [ruby-core:42460][Feature #5992]
+
+	* ext/fiddle/fiddle.c (Init_fiddle): ditto.
+
+	* ext/dl/lib/dl/cparser.rb (DL::CParser#parse_ctype): ditto.
+
+Sat Feb 25 11:08:28 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c (Init_curses): use rb_define_const once for
+	  Curses::VERSION.
+
+	* ext/dbm/dbm.c (Init_dbm): ditto for DBM::VERSION.
+
+Sat Feb 25 10:34:22 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c (Init_curses): make Curses::VERSION
+	  understandable without context.
+
+	* ext/dbm/dbm.c (Init_dbm): ditto for DBM::VERSION.
+
+Sat Feb 25 07:53:58 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_string): insert a backslash only if
+	  quoted by single quotes.  [ruby-dev:45281] [Bug #6069]
+
+Sat Feb 25 07:53:49 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_inspect), io.c (rb_io_inspect): keep encoding of path.
+	  [Bug #6072]
+
+Sat Feb 25 07:53:40 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_initialize): keep path in original encoding.
+
+	* error.c (syserr_initialize): prefer the encoding of message over
+	  locale.  [ruby-dev:45279][Bug #6071]
+
+Sat Feb 25 06:55:29 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (utime_internal): fix a variable missed to replace.
+	  [ruby-core:42864] [Bug #6077]
+
+Fri Feb 24 18:21:55 2012  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* test/zlib/test_zlib.rb (TestZlibGzipReader#test_encoding): Add
+	  encoding testcases for GzipReader#read.  read() emits
+	  Encoding.default_external in contrast to read(size) emits BINARY.
+	  See also: http://bugs.jruby.org/6208
+
+Fri Feb 24 17:56:39 2012  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/ruby/test_literal.rb (TestRubyLiteral#test_special_const):
+	  test for https://bugs.php.net/bug.php?id=61095
+
+Fri Feb 24 16:48:29 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c, file.c, io.c (rb_sys_fail_path): use rb_sys_fail_str.
+
+	* error.c: new functions to deal exceptions with string instances.
+
+	* dir.c, file.c, io.c: use rb_sys_fail_path.
+
+Fri Feb 24 15:49:07 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (__builtin_unreachable): check for clang.
+	  [ruby-core:42849]
+
+	* include/ruby/ruby.h (UNREACHABLE): fallback definition.
+
+Fri Feb 24 13:54:33 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c: prevent a memory leak by protecting calls to
+	  handler callbacks.
+	* test/psych/test_parser.rb: test to demonstrate leak.
+
+Fri Feb 24 12:07:34 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/net/http.rb: Fix documentation. Patched from Florian Mhun
+	  via http://github.com/ruby/ruby/pull/96
+
+Fri Feb 24 11:48:07 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* string.c (rb_str_prepend): Fix documentation for String#prepend.
+	  Patched from Franck Verrot via http://github.com/ruby/ruby/pull/98
+	  and Andrew Horsman via http://github.com/ruby/ruby/pull/55
+
+Fri Feb 24 10:08:33 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb (Net::HTTP#transport_request):  Fix infinite loop
+	  upon EOFError or Errno::ECONNRESET where count is reset to 0.
+	* test/net/http/test_http.rb (class TestNetHTTPKeepAlive):  Test for
+	  above.
+
+Fri Feb 24 09:05:40 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* complex.c (Init_Complex):  Document Complex::I.  Patch by Sylvain
+	  Daubert.  [Feature #5623]
+
+Fri Feb 24 08:52:09 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_string, parser_yylex): insert a backslash
+	  if the next character is non-ascii.  [ruby-dev:45278] [Bug #6069]
+
+Fri Feb 24 08:13:20 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/profiler.rb:  Add Profiler documentation by Gonzalo Rodriguez.
+	  [Bug #5816]
+
+Fri Feb 24 08:08:38 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c: set parser encoding based on the YAML input
+	  rather than user configuration.
+	* test/psych/test_encoding.rb: corresponding tests.
+	* test/psych/test_parser.rb: ditto
+	* test/psych/test_tainted.rb: ditto
+
+Fri Feb 24 08:02:52 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* hash.c (Init_Hash):  Add section on how objects are used as Hash keys
+	  and how to use custom classes as Hash keys.
+
+Fri Feb 24 07:36:11 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (rb_obj_eql):  Improve equality documentation by adding an
+	  example of equal? vs == and recommending eql? be aliased to == when
+	  overridden.
+
+Fri Feb 24 07:21:15 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (rb_obj_hash):  Added note that the hash value is not
+	  deterministic on Marc-Andre's suggestion.  Expanded description of
+	  the purpose of the hash method.  [Bug #6068]
+
+Thu Feb 23 23:01:21 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: unused macro removed.
+
+Thu Feb 23 22:26:53 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/test_curses.rb: new file.
+
+Thu Feb 23 19:57:56 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/rain.rb: trap SIGHUP, SIGINT, SIGQUIT and SIGTERM only.
+
+Thu Feb 23 19:56:48 2012  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (sig_trap): show signal name on error.
+
+Thu Feb 23 12:21:48 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: use DBM_SUFFIX only to detect header of
+	  Berkeley DB.
+
+Thu Feb 23 10:00:18 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* io.c (rb_io_f_sync):  Fix double-negative typo.  [ruby-trunk - #5837]
+
+Thu Feb 23 09:57:21 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* load.c (rb_f_require):  Add note to require for scope of items in the
+	  loaded file.  [ruby-trunk - #5910]
+
+Thu Feb 23 03:58:08 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb (delete_field): Bug fix so previous value is
+	  returned. Patch by Nick Recobra [Bug #6063]
+
+Thu Feb 23 02:33:00 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_extract_modeenc): fail only if conflicting
+	  text/binary modes given explicitly.  [ruby-dev:45268][Bug #6055]
+
+Wed Feb 22 23:27:08 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/iconv/test_option.rb: enabled.  [ruby-core:42802][Bug #6061]
+
+Wed Feb 22 21:45:56 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c: use defined() to suppress a warning.
+
+Wed Feb 22 21:44:29 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: refactored.
+
+Wed Feb 22 20:42:28 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: try to distinguish curses_version is a
+	  function or variable.
+
+	* ext/curses/curses.c (Init_curses): refine Curses::VERSION.
+
+Wed Feb 22 19:47:03 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: show the chosen header and library.
+
+Wed Feb 22 19:22:31 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* reverted 34739 for test/date.
+
+Wed Feb 22 19:08:55 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: refactored.
+
+Wed Feb 22 18:44:41 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb (setup_options): add option "--retry" as opposite
+	  for "--no-retry"
+
+Wed Feb 22 18:34:02 2012  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb (setup_options): add option "--show-skip" to
+	  cancel "--hide-skip" (-q)
+
+Wed Feb 22 17:36:22 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_s_foreach): argument check before making Enumerator.
+	  [ruby-dev:31525]
+
+Wed Feb 22 17:07:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_s_foreach): return enumerator including keyword
+	  arguments.  [ruby-dev:45267][Bug #6054]
+
+Wed Feb 22 12:15:16 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: allow llvm-gcc because it work fine with r34278.
+
+Wed Feb 22 10:57:08 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (fetch_token): don't use // comment.
+
+Wed Feb 22 10:32:11 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/mkmf/test_framework.rb: try CoreFoundation framework, than
+	  Cocoa which is dependent on QuickTime SDK which has separated
+	  since Xcode 4.3.
+
+Wed Feb 22 10:18:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (test-all, test-ruby): more dependencies.
+
+Wed Feb 22 06:48:55 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* file.c (rb_f_test):  Fix formatting of Kernel#test rdoc.
+
+Wed Feb 22 06:12:15 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: check DBM_SUFFIX for Mac OS X.
+	  Its ndbm.h doesn't include db.h.
+
+Wed Feb 22 06:02:42 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/dbm.c (fdbm_initialize): disable Berkeley DB error messages.
+
+	* ext/dbm/extconf.rb: check DBC type for above.
+
+	[ruby-dev:45269]
+
+Tue Feb 21 20:23:47 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_any_hash): treat Qundef like as other special constants.
+
+	* hash.c (hash_foreach_iter): fix signature.
+
+Tue Feb 21 19:39:34 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c (Init_curses): use curses_version() for
+	  Curses::VERSION.
+
+Tue Feb 21 18:21:25 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c : remove gc_clear_mark_on_sweep_slots() and use
+	  rest_sweep() instead of it, because some dead objects might be
+	  marked in next the mark phase by false pointers.
+	  [ruby-core:42672]
+
+Tue Feb 21 16:08:17 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_hash_proc): get wrapped pointer properly.  [Bug #6048]
+
+Tue Feb 21 14:41:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/ruby.pc.in: added rubylibprefix, {rubylib,vendor,site}dir
+	  and {ruby,vendor,site}archdir.  [ruby-core:42766][Feature #6052]
+
+Tue Feb 21 09:13:25 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c (method_hash, proc_hash): Fix {Unbound}Method#hash
+	  [Bug #6048]. Isolate hash computation for proc
+
+	* internal.h: Declaration for above
+
+	* vm_method.c (rb_method_definition_hash): Computation for
+	  hash part of a method definition
+
+	* method.h: Declaration for above
+
+	* test/ruby/test_method.rb: Test for above
+
+Tue Feb 21 02:56:15 2012  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* enumerator.c (enumerator_rewind): update the documentation.
+	  fixed: #6053
+
+Mon Feb 20 23:38:35 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/depend: ignore mktable.c because it's not encoding library.
+	  [ruby-core:42760] [Bug #6049]
+
+Mon Feb 20 21:40:53 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: fold too long lines.
+
+Mon Feb 20 21:16:48 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* lib/fileutils.rb: revert a line modified accidentally at r34669.
+	  This fixes mingw test errors in TestDir_M17N.
+	  [ruby-core:42728] [Feature #4970]
+
+Mon Feb 20 21:09:27 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/curses.c (Init_curses): define Curses::VERSION.
+
+Mon Feb 20 21:08:00 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: restore $libs and $defs for each
+	  header/library choice.
+
+Mon Feb 20 19:57:26 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: weaken header/library consistency check if db is
+	  "ndbm".  It seems several (possibly historical) distributions
+	  provide libndbm.  However the content of libndbm vary: Berkeley DB,
+	  GDBM or even 4.3BSD NDBM.  (Mandriva, Tru64 UNIX, OpenSuSE,
+	  SCO OpenServer, ...)
+	  "ndbm" is not searched automatically now (dblib doesn't contain it)
+	  but configure --with-dbm-type=ndbm choose libndbm and ndbm.h.
+
+Mon Feb 20 19:15:57 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: refine variable names.
+
+Mon Feb 20 15:50:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if -fstack-protector is really available.
+
+Sun Feb 19 23:43:38 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: show header and library found.
+
+Sun Feb 19 23:01:01 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/dbm.c (Init_dbm): refine DBM::VERSION definition.
+
+	* ext/dbm/extconf.rb: provide RUBYDBM_GDBM_HEADER macro.
+
+Sun Feb 19 17:07:27 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/dbm/test_dbm.rb (test_dbmfile_suffix): check pag and dir is
+	  empty for 4.3BSD ndbm.
+
+Sun Feb 19 03:00:30 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/dbm/test_dbm.rb (test_dbmfile_suffix): check magic numbers.
+
+Sun Feb 19 01:05:41 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: detect GDBM's ndbm.h by testing dbm_clearerr is
+	  an empty macro.
+
+Sun Feb 19 00:25:55 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: don't choose 'dbm' if _GDB_H_ is defined which
+	  is available since GDBM 1.9 because 'gdbm_compat' is appropriate
+	  choice since GDBM 1.8.1.
+
+Sat Feb 18 23:27:00 2012  Kenta Murata  <mrkn@mrkn.jp>
+
+	* random.c: remove a duplicated comment.
+
+Sat Feb 18 18:43:13 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb (dblib): prefer recent GDBM over older GDBM.
+	  (have_declared_libvar): new function to check a declared variable
+	  exists in a library.
+	  (have_undeclared_libvar): renamed from renamed from have_libvar.
+	  (headers.db_check2): check that GDBM version variable if GDBM header
+	  is chosen.
+
+	* ext/dbm/dbm.c (Init_dbm): use HAVE_DECLARED_LIBVAR_GDBM_VERSION and
+	  HAVE_UNDECLARED_LIBVAR_GDBM_VERSION macro.
+
+Sat Feb 18 13:53:01 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/dbm/test_dbm.rb (test_dbmfile_suffix): DBM::VERSION should
+	  be Berkeley DB if foo.db is created by DBM.open.
+
+Sat Feb 18 13:40:37 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/dbm/test_dbm.rb (test_dbmfile_suffix): test dbm file suffix.
+
+Sat Feb 18 12:50:59 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/dbm.c (DBM::VERSION): define it by detecting _GDBM_H_ or
+	  _DBM_IOERR.
+
+Sat Feb 18 07:52:45 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/enc-unicode.rb: don't use 1.9 feature on tools.
+
+Sat Feb 18 02:48:39 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/fileutils.rb: refactored FileUtil methods to use the
+	  `define_command` API.  Patch from 7rans <transfire@gmail.com>
+	* test/fileutils/test_dryrun.rb: corresponding test refactoring
+	* test/fileutils/test_nowrite.rb: ditto
+	* test/fileutils/test_verbose.rb: ditto
+
+Fri Feb 17 21:39:36 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: remove dbm.
+
+Fri Feb 17 21:18:39 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: refine header/library mismatch detection.
+	  check only for ndbm.h except libc.  check _GDBM_H_ for gdbm.
+	  check _DBM_IOERR for the original ndbm.
+
+Fri Feb 17 20:30:44 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: don't check libdbm.  It is not a ndbm
+	  implementation.  (libdbm in Version 7 Unix is database library
+	  for single database per process.)
+
+Fri Feb 17 15:38:53 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Merge Onigmo-5.13.1. [ruby-dev:45057] [Feature #5820]
+	  https://github.com/k-takata/Onigmo
+	  cp reg{comp,enc,error,exec,parse,syntax}.c reg{enc,int,parse}.h
+	  cp oniguruma.h
+	  cp tool/enc-unicode.rb
+	  cp -r enc/
+
+Fri Feb 17 15:20:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_each_slice): arrays to be yielded can be newly
+	  created in the block.
+
+	* enum.c: move work variables to objects not to let called blocks
+	  access stack area out of scope.  [Bug #5801]
+
+Fri Feb 17 12:35:55 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: remove borders from the commit message which is used
+	  when the commit doesn't change ChangeLog.
+
+Fri Feb 17 11:50:11 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (btest, btest-ruby, test-sample test-knownbugs)
+	  (test-all, test-ruby): depend on prog.
+
+Fri Feb 17 09:56:22 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_header): log the content of header.
+
+Fri Feb 17 09:44:55 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/transcode-tblgen.rb (import_ucm): don't use \h because the
+	  script should work with ruby 1.8.
+
+	* tool/enc-unicode.rb: ditto.
+
+Fri Feb 17 07:33:29 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (id_lshift): use constant ID.
+
+Fri Feb 17 07:30:53 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: refactored to split too long conditions.
+
+Fri Feb 17 00:23:25 2012  Tanaka Akira  <akr@fsij.org>
+
+	* test/dbm/test_dbm.rb: fix skip condition for libgdbm 1.8.0 or prior.
+	  reported by Bohuslav Kabrda.
+	  [ruby-core:42685] [ruby-trunk - Bug #6036]
+
+Fri Feb 17 00:04:21 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: check _DB_H_ macro unavailable except
+	  Berkeley DB library.
+
+Thu Feb 16 05:41:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (splatarray): make new array if flag is set.
+
+	* compile.c (iseq_compile_each): make new array with
+	  splat. [ruby-core:21901][Feature #1125]
+
+Thu Feb 16 00:14:04 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/abbrev.rb (Array#abbrev): add missing '"' in documentation.
+
+Wed Feb 15 22:20:19 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* cont.c (rb_fiber_reset_root_local_storage): add a new function to
+	  restore rb_thread_t::local_storage.
+
+	* cont.c (rb_obj_is_fiber): add a new function to tell finalizer to
+	  prevent fibers from destroy.
+
+	* gc.c (rb_objspace_call_finalizer): don't sweep fibers at finalizing
+	  objspace.
+
+	* internal.h (rb_fiber_reset_root_local_storage, rb_obj_is_fiber):
+	  add prototypes.
+
+	* vm.c (ruby_vm_destruct): reset main thread's local_storage before
+	  free main thread. rb_thread_t::local_storage is replaced by fiber's
+	  local storage when forked from fiber, and it should be already freed
+	  when the fiber was destroyed. [ruby-core:41456] [Bug #5700]
+
+	* test/ruby/test_fiber.rb (test_fork_from_fiber): add test for fork
+	  from fiber.
+
+Wed Feb 15 19:57:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/closure.c (callback): deal with unsigned integers.
+	  [ruby-core:42458][Bug #5991][Bug #6022]
+
+	* ext/fiddle/conversions.c (value_to_generic, generic_to_value):
+	  ditto.
+
+	* ext/fiddle/closure.c (callback): same as r34506.
+
+Wed Feb 15 17:41:31 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_strsetbuf): call rb_str_modify to make str independent
+	  before calling rb_str_set_len for r34580.
+
+Wed Feb 15 12:30:10 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (Init_zlib):  Added Zlib::TEXT and note that
+	  Zlib::ASCII is deprecated in zlib 1.2.3 and newer.
+
+Wed Feb 15 12:24:40 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c:  Move constant descriptions to constants.  Remove
+	  extra comment block at the top of Init_zlib().
+
+Wed Feb 15 12:30:46 2012  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb: Create getters and setters after dup.
+	  [Bug #6028] [rubyspecs:0380bcc]
+
+Wed Feb 15 10:59:52 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (HEAP_BITMAP_LIMIT): HEAP_BITMAP_LIMIT is computed on the
+	  basis of HEAP_SIZE because it must covers a whole heap block.
+	  [ruby-trunk - Bug #6006]
+
+Wed Feb 15 09:27:45 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (Init_zlib):  Added Zlib::FIXED and Zlib::RLE
+	  strategies.
+	* NEWS:  Add note about the new Zlib constants.
+
+Wed Feb 15 09:11:36 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c:  Improve documentation.  [ruby-trunk - Bug #5948]
+
+Wed Feb 15 07:28:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* encoding.c (Init_Encoding):  Add IO example of internal and external
+	  encoding.  Fixed a typo in the force_encoding example.  [#5949]
+
+Wed Feb 15 06:58:21 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* encoding.c (Init_Encoding):  Add Encoding documentation.
+	  [ruby-trunk - Bug #5949]
+	* encoding.c (rb_set_default_external):  Fix typo in documentation.
+
+Tue Feb 14 20:22:11 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (CEILDIV): rename to a appropriate name.
+
+Tue Feb 14 18:07:20 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (assign_heap_slot): SEGV happens cause on 64-bit platform
+	  sometime there should be `objs-=2` instead of `objs--`.
+	  [Bug #6006]
+	  patched by Sokolov Yura. https://github.com/ruby/ruby/pull/92
+
+Tue Feb 14 16:00:30 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_setstrbuf): cut down the buffer if longer.
+
+Tue Feb 14 15:06:37 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (build_message): skip escaped
+	  question marks.
+
+Tue Feb 14 12:10:04 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (autoload_const_set, autoload_require): fix
+	  signatures.
+
+Tue Feb 14 05:23:40 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* process.c (proc_wait):  Change typo "SystemError" to
+	  "SystemCallError".  [ruby-trunk - Bug #5962]
+	* process.c (proc_wait2):  ditto
+
+Tue Feb 14 05:18:24 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* enumerator.c:  Document use of Enumerator.new for creating a lazy
+	  enumeration for filtering/chaining.  [ruby-trunk - Feature #707]
+
+Mon Feb 13 23:01:50 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* vm_method.c (rb_method_boundp):
+	  obj.respond_to?(:a_protected_method) should return false because
+	  calling a protected method may cause NoMethodError if called
+	  from outside the class inheritance tree.  Kernel#respond_to? is
+	  mostly used to test if it is safe to call a method, so the false
+	  positive should be avoided. [ruby-dev:40461] [ruby-dev:41739]
+	  [ruby-dev:41837]
+
+Mon Feb 13 21:52:06 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (HEAP_OBJ_LIMIT, HEAP_BITMAP_LIMIT): HEAP_OBJ_LIMIT used
+	  `sizeof(struct heaps_slot)` while heap is currently allocated
+	  with `struct heaps_header`.
+	  HEAP_BITMAP_LIMIT were calculated from
+	  `HEAP_OBJ_LIMIT/sizeof(uintptr_t)` - one Byte for each object,
+	  not one Bit. [Bug #6006]
+	  patched by Sokolov Yura. https://github.com/ruby/ruby/pull/92
+
+Mon Feb 13 18:30:32 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_setstrbuf): defer resizing buffer string until data is
+	  read actually.
+
+Mon Feb 13 10:24:39 2012  Loren Segal  <lsegal@soen.ca>
+
+	* io.c (Init_IO): use directive hack to make ARGF documentable
+	  in other tools.  [ruby-core:42515][Bug #6007]
+
+Sun Feb 12 20:43:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_event_hook_func_t): add argument names.
+
+Sun Feb 12 16:30:23 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* tool/merger.rb (#default_merge_branch): Add support for
+	  Subversion 1.7 which adopted a whole new working directory
+	  structure.
+
+Sun Feb 12 15:14:41 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* benchmark/driver.rb: suppress unused/shadowing variable warnings.
+
+Sun Feb 12 03:14:40 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* vm_eval.c (check_funcall):  Call respond_to? with matching arity for
+	  legacy single-argument implementations.  [ruby-trunk - Bug #6000]
+
+Sat Feb 11 12:04:05 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (defined_expr): guard the whole expression.
+	  [ruby-dev:45021][Bug#5786]
+
+Sat Feb 11 08:34:42 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (rb_inflate_add_dictionary):  Added
+	  Zlib::Inflate#add_dictionary to allow users to pre-specify
+	  for using during #inflate.  [ruby-trunk - Feature #5937]
+
+Sat Feb 11 08:23:02 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (do_inflate):  Inflate more data if buffered data
+	  exists.  Allows Zlib::Inflate#set_dictionary to work.
+	  [ruby-trunk - Bug #5929]
+
+Sat Feb 11 06:00:48 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dir.c (fnmatch): The * needs to be escaped to avoid formatting in
+	  fnmatch comment.
+	  patched by @dalton. https://github.com/ruby/ruby/pull/91
+
+Fri Feb 10 03:41:31 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c: removed external encoding setter, allow parser
+	  to be reused.
+	* ext/psych/lib/psych/parser.rb: added external encoding setter.
+	* test/psych/test_parser.rb: test parser reuse
+
+Fri Feb 10 01:30:41 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/dl.h (ALIGN_OF): use offsetof().
+
+	* ext/dl/dl.h (DLALIGN): round up at once and get rid of overflow.
+
+Fri Feb 10 00:47:07 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_no_memory_leak): new assertion to
+	  check memory leak by invoking child ruby process and watch its
+	  memory size.
+
+Thu Feb  9 23:41:44 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/pathname/test_pathname.rb (test_binread): add assertion to
+	  check encoding.
+
+Thu Feb  9 16:48:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/dl.c (Init_dl): fix mangled document.
+
+Thu Feb  9 16:10:34 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/memory_status.rb (Memory::Win32): 64bit support.
+
+Thu Feb  9 16:08:55 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/lib/value.rb (DL::ValueUtil.{unsigned_value,signed_value}):
+	  currently pack/unpack does not accept "q!" and "Q!".
+
+Thu Feb  9 16:01:29 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/fiddle/conversions.c (value_to_generic): src is not guaranteed as
+	  a Bignum if the type is LONG_LONG.  it may be a Fixnum if the value
+	  is small.
+
+Thu Feb  9 11:32:36 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/lib/types.rb: Win64 support.
+
+Thu Feb  9 04:12:29 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/pathname/test_pathname.rb: not read but binread.
+	  patched by Benoit Daloze, [ruby-core:42440] [Bug #5984]
+
+Wed Feb  8 22:29:59 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_modify_expand): fix memory leak.
+
+Wed Feb  8 14:06:59 2012  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c: Add SSL constants and allow to unset SSL
+	  option to prevent BEAST attack. See [Bug #5353].
+
+	  In OpenSSL, OP_DONT_INSERT_EMPTY_FRAGMENTS is used to prevent
+	  TLS-CBC-IV vulnerability described at
+	  http://www.openssl.org/~bodo/tls-cbc.txt
+	  It's known issue of TLSv1/SSLv3 but it attracts lots of attention
+	  these days as BEAST attack. (CVE-2011-3389)
+
+	  Until now ossl sets OP_ALL at SSLContext allocation and call
+	  SSL_CTX_set_options at connection.  SSL_CTX_set_options updates the
+	  value by using |= so bits set by OP_ALL cannot be unset afterwards.
+
+	  This commit changes to call SSL_CTX_set_options only 1 time for each
+	  SSLContext. It sets the specified value if SSLContext#options= are
+	  called and sets OP_ALL if not.
+
+	  To help users to unset bits in OP_ALL, this commit also adds several
+	  constant to SSL such as
+	  OpenSSL::SSL::OP_DONT_INSERT_EMPTY_FRAGMENTS.  These constants were
+	  not exposed in Ruby because there's no way to unset bits in OP_ALL
+	  before.
+
+	  Following is an example to enable 0/n split for BEAST prevention.
+
+	    ctx.options = OP_ALL & ~OP_DONT_INSERT_EMPTY_FRAGMENTS
+
+	* test/openssl/test_ssl.rb: Test above option exists.
+
+Wed Feb 8 13:12:02 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_x509name.c: Use the numerical representation of
+	  unrecognized OIDs instead of the sn "UNDEF".
+
+	* test/openssl/test_x509name.rb: Add tests for the fixed behavior.
+
+	  Patch provided by Paul Kehrer, thank you!
+	  [ruby-core:41769] [Feature #5787]
+
+Wed Feb  8 09:49:58 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: don't abort, update first.
+
+Wed Feb 8 09:47:33 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: Call INT2NUM only once for GeneralString.
+	  Thanks to Mantas Mikulenas for noticing and providing a patch!
+	  [ruby-core:42358] [Bug #5972]
+
+Wed Feb 8 09:19:00 2012  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_cipher.c: Add warning about key as IV.
+
+Tue Feb  7 20:08:12 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* error.c (exc_inspect): Fix typo. patch from Trent Ogren
+	  via https://github.com/ruby/ruby/pull/90
+
+Tue Feb  7 19:37:35 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c: refactor packed entries using structs.
+
+Tue Feb  7 14:52:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_update): table can be unpacked in the callback.
+
+	* st.c (st_foreach): should not yield same pair when checking
+	  after unpacking.
+
+Mon Feb  6 21:55:13 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: abort if the working directory is dirty.
+
+	* tool/merger.rb: update the working directory after commit.
+
+Mon Feb  6 00:16:27 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): return ASCII-8BIT even if 2nd string
+	  is ascii only string. [ruby-core:42354] [Bug #5968]
+
+Fri Feb  3 07:16:47 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick.rb:  Moved proxy rewriting to WEBrick::HTTPProxy.
+	* lib/webrick/httpproxy.rb:  Add examples of creating a proxy server
+	  and response rewriting using HTTPProxy.
+
+Fri Feb  3 06:53:22 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_x509store.c:  Add class documentation for
+	  OpenSSL::X509::Store
+
+Thu Feb  2 22:28:13 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/net/http/test_https_proxy.rb
+	  (HTTPSProxyTest#test_https_proxy_authentication):
+	  add workaround to avoid to hang up without openssl.
+	  see [ruby-dev:45021][Bug #5786]
+
+	* test/resolv/test_dns.rb (TestResolvDNS#test_query_ipv4_address):
+	  ditto.
+
+Thu Feb  2 21:48:18 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb: use meaningful names.
+
+Thu Feb  2 21:38:52 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb, test/rexml/test_namespace.rb:
+	  fix the default xml namespace URI validation.
+	  [ruby-dev:45169] [Bug #5956]
+	  Reported by Miho Hiramatsu. Thanks!!!
+
+Thu Feb  2 17:51:02 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (argf_next_argv): reset ARGF.next_p on ARGV.replace.
+	  r34409 breaks replacing ARGV.
+	  [ruby-dev:45160] [Bug #5952]
+
+Thu Feb  2 16:21:01 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/net/http/test_http.rb (TestNetHTTPKeepAlive#*): remove debug
+	  output.
+
+Thu Feb  2 01:24:34 2012  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* parse.y (debug_lines, coverage): set file path encoding for coverage
+	  result.  [ruby-dev:44950]
+
+Wed Feb  1 14:38:31 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/tempfile.rb (Tempfile#unlink, Tempfile::Remover#call): Just
+	  call File.unlink and ignore ENOENT because existence check
+	  before unlinking does not help in terms of race condition.
+
+	* lib/tempfile.rb (Tempfile#unlink, Tempfile::Remover#call): My
+	  comment about thread safeness is obsolete.
+
+Wed Feb  1 09:50:10 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/re.rdoc (Repetition): fix typo.  reported by Ori Avtalion
+	  and patched by Zachary Scott.  [Bug #5947]
+
+Wed Feb  1 06:38:54 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_close): skip stdin, which should be readable again.
+	  [ruby-dev:45160] [Bug #5952]
+
+	* io.c (argf_readlines): reinitialize after all read to be
+	  readable again.
+
+Tue Jan 31 21:27:43 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* configure.in (HEAP_ALIGN_LOG): HEAP_ALIGN_LOG should be page
+	  size in OpenBSD. [ruby-core:42158][Bug #5901]
+
+	* gc.c : avoid to redefine.
+
+Tue Jan 31 14:27:22 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (EnvUtil.invoke_ruby): yield also child pid
+	  in block form.
+
+Mon Jan 30 19:08:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call0): should pass block to enumerators.  patched
+	  by Kazuki Tsujimoto.  [ruby-dev:44961][Bug #5731]
+
+	* vm_eval.c (method_missing), vm_insnhelper.c (vm_call_method):
+	  ditto.  patched by satoshi shiba.
+
+Mon Jan 30 12:31:05 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (append_fspath): need to set the encoding to result always.
+
+Mon Jan 30 10:38:37 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/irb/test_completion.rb: skip if cannot load irb/completion
+	  (maybe readline does not exist).
+
+Sun Jan 29 22:47:19 2012  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* tool/config.{guess,sub}: updated to automake-1.11.2.
+
+Sun Jan 29 12:17:56 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  respect encodings.  [Bug #5941]
+
+Sat Jan 28 09:33:33 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* win32/win32.c (rb_w32_read): fix an issue that $stdin.read doesn't
+	  terminate by CTRL-C on Windows.
+	  [ruby-dev:45149] [Bug #5812]
+
+Sat Jan 28 08:18:11 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_thread.rb
+	  (TestThreadGroup#test_thread_timer_and_interrupt): skip exit status
+	  assertion because we cannot get signal status on Windows.
+
+	* win32/win32.c (CreateChild): create process group to receive the
+	  signal by GenerateConsoleCtrlEvent().
+
+	* win32/win32.c (kill): use CTRL_BREAK_EVENT instead of CTRL_C_EVENT
+	  if a process group is specified. CTRL_C_EVENT signal cannot be
+	  generated for process groups for the specification.
+	  [ruby-dev:45149] [Bug #5812]
+
+Sat Jan 28 07:46:03 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* thread_win32.c (rb_w32_wait_events_blocking): use
+	  ruby_thread_from_native() instead of GET_THREAD() because
+	  GET_THREAD() doesn't always return the current thread and
+	  WaitForMultipleObjects() at rb_w32_read() doesn't return by
+	  Thread#kill. This fixes TestQueue#test_thr_kill failure on
+	  Windows.
+
+	* thread_win32.c (rb_w32_wait_events): use ruby_thread_from_native()
+	  instead of GET_THREAD() for consistency with the above change.
+
+	* thread_win32.c (rb_w32_sleep): ditto.
+
+	* thread_win32.c (rb_w32_Sleep): ditto.
+	  [ruby-dev:45149] [Bug #5812]
+
+Sat Jan 28 07:28:48 2012  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/zlib/test_zlib.rb (TestZlibGzipReader#test_reader_wrap): set
+	  binmode explicitly for fixing test error on Windows. This is consistent
+	  with r34243.
+	  [ruby-dev:45149] [Bug #5812]
+
+Sat Jan 28 05:53:34 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb/completion.rb (IRB::InputCompletor::CompletionProc):
+	  ignore non-string name modules.  [ruby-core:42244][Bug #5938]
+
+Fri Jan 27 16:31:45 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (HEAP_ALIGN, HEAP_ALIGN_MASK): DRY, let compiler calculate
+	  from HEAP_ALIGN_LOG.
+
+Thu Jan 26 11:03:37 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/matrix.rb:  Clean up extra whitespace in output documentation.
+
+Thu Jan 26 03:24:02 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (io_getch): default delegating method
+	  for StringIO.  https://github.com/nobu/io-console/issues/4
+
+	* ext/stringio/stringio.c: moved some methods to hidden modules.
+
+Wed Jan 25 13:27:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_basename): ignore non-ascii extension in
+	  different encoding, which cannot match.
+
+	* file.c (rmext): no extension to strip if empty string.
+
+	* file.c (rb_enc_path_next, rb_enc_path_skip_prefix)
+	  (rb_enc_path_last_separator, rb_enc_path_end)
+	  (ruby_enc_find_basename, ruby_enc_find_extname): encoding-aware
+	  path handling functions.
+
+	* file.c (rb_home_dir, file_expand_path, rb_realpath_internal)
+	  (rb_file_s_basename, rb_file_dirname, rb_file_s_extname)
+	  (rb_file_join): should respect the encodings of arguments than
+	  file system encoding.  [ruby-dev:45145] [Bug #5919]
+
+	* dir.c (check_dirname, ruby_glob0): ditto.
+
+	* ext/pathname/pathname.c (path_sub_ext): ditto.
+
+Tue Jan 24 14:20:42 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_iter_break_value): new function to break a block with
+	  the value.  [ruby-dev:45132] [Feature #5895]
+
+Tue Jan 24 12:58:41 2012  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* object.c (rb_Hash): add Kernel#Hash conversion method like
+	  Array() or Float().  a patch from Run Paint Run Run.  Fix #3131
+
+Tue Jan 24 11:38:05 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.encode_www_form_component): initialize on
+	  requiring to support JRuby, which runs parallel multithreads.
+	  [ruby-core:42222] [Bug #5925]
+
+	* lib/uri/common.rb (URI.decode_www_form_component): initialize on
+
+Mon Jan 23 20:33:11 2012  Jason Kay  <geniture@me.com>
+
+	* lib/net/http.rb (Net::HTTP#connect): Writing entire packet at
+	  once to avoid incomplete transmission.  Current code using
+	  writeline was causing sub-optimal conversing with a proxy due to
+	  the connect tunnel request headers being split over multiple
+	  packets. The modification I made allows the connect request to
+	  be written as one packet, avoiding problems and optimizing the
+	  conversation.
+
+	  https://github.com/ruby/ruby/pull/72
+	  [Feature #5460]
+
+Mon Jan 23 17:06:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/mailto.rb (URI::MailTo.build): follow Array#to_s change of
+	  Ruby 1.9; use Array#join. [Bug #5840]
+
+Mon Jan 23 16:42:28 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (extract_binmode): raise an exception if binmode/textmode
+	  is specified with both vmode and opthash.
+	  [ruby-core:42199] [Bug #5918]
+
+Mon Jan 23 16:35:27 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_extract_modeenc): set ASCII-8BIT if binmode is specified
+	  with opthash. [ruby-core:42197] [Bug #5917]
+
+Mon Jan 23 10:08:00 2012  Kenta Murata  <mrkn@cookpad.com>
+
+	* test/cgi/test_cgi_util.rb (test_cgi_escape_preserve_encoding):
+	  add a test for CGI::escape to preserve encoding.
+
+	* test/cgi/test_cgi_util.rb (test_cgi_unescape_preserve_encoding):
+	  add a test for CGI::unescape to preserve encoding.
+
+Mon Jan 23 00:45:34 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/rdoc-mode.el (rdoc-imenu-create-index): Add imenu support
+	  to rdoc-mode.
+
+	* misc/rdoc-mode.el (rdoc-mode): Fix regexp patterns containing
+	  "\s " where CR/LF is not supposed to match.
+
+Sun Jan 22 15:41:26 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_intern3): split to registration check and new
+	  registration.
+
+	* parse.y (rb_intern_str): make interned string shared with the
+	  given string.
+
+	* parse.y (rb_intern3, rb_intern_str): check the coderange first.
+
+Sat Jan 21 22:21:07 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (FIXNUM_P): simple flag should be int.
+
+Sat Jan 21 21:51:19 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (rb_enc_compatible): fix segv on symbols.
+	  [ruby-core:42204] [Bug #5921]
+
+Sat Jan 21 11:43:45 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_chdir, check_dirname): get rid of optimization-out.
+
+Fri Jan 20 20:47:37 2012  Kenta Murata  <mrkn@cookpad.com>
+
+	* lib/cgi/util.rb (CGI.escape): support a string with invalid byte
+	  sequence.  [Bug #5913]
+
+	* test/cgi/test_cgi_util.rb
+	  (test_cgi_escape_with_invalid_byte_sequence): test for the above
+	  change.
+
+Fri Jan 20 17:37:37 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm.c (vm_exec): remove workaround for LLVM because r34278 fixes it.
+
+	* vm_insnhelper.c (vm_call_cfunc): ditto.
+
+Fri Jan 20 14:31:43 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#transport_request): retry a idempotent
+	  request automatically. [ruby-dev:45030] [Bug #5790]
+	  [ruby-core:41821] [Bug #5813]
+
+	* lib/net/http.rb (Net::HTTP#keep_alive_timeout=): added to specify
+	  the second to reconnect the TCP connection on Keep-Alive.
+	  The default value is 2 second because current servers uses 2 sec.
+	  http://ftp-admin.blogspot.com/2009/09/keepalivetimeout2.html
+
+	* lib/net/http.rb (Net::HTTP#begin_transport): reconnect TCP
+	  connection on keep-alive timeout.
+
+Thu Jan 19 07:53:09 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strptime.c: moved detector of leftover.
+
+Thu Jan 19 07:10:47 2012  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c: [ruby-core:42173].
+
+Wed Jan 18 18:11:02 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/rdoc-mode.el (rdoc-mode): Add provide so that requiring
+	  this library succeeds.
+
+Wed Jan 18 18:06:45 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/curses/curses.c (cWindow, cMouseEvent): made typed data.
+
+Wed Jan 18 12:49:15 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Added support for loading
+	  subclasses of String with ivars
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: Added support for dumping
+	  subclasses of String with ivars
+	* test/psych/test_string.rb: corresponding tests
+
+Wed Jan 18 10:39:47 2012  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Added ability to load array
+	  subclasses with ivars.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: Added ability to dump
+	  array subclasses with ivars.
+	* test/psych/test_array.rb: corresponding tests
+
+Tue Jan 17 17:18:41 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (SPT_TYPE): enable as SPT_REUSEARGV on Darwin.
+
+	* missing/setproctitle.c (ruby_init_setproctitle): changed prefix.
+
+Tue Jan 17 12:32:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (aligned_malloc, aligned_free): covered missing defined
+	  operators and fixes for cygwin.
+
+Tue Jan 17 10:54:46 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (do_hash): it's the time to remove cast to unsigned int.
+
+Tue Jan 17 07:30:12 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* st.c (unpack_entries): Fix r34310: on unpacking, the position of
+	  a hash must be do_hash-ed value.
+
+	* st.c (add_packed_direct): ditto.
+
+Mon Jan 16 16:41:53 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (Regexp): fix incorrect options when casting to
+	  a Regexp, and suppress encoding option warnings.
+	  https://github.com/ruby/ruby/pull/82
+
+Mon Jan 16 11:22:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_chsize): no need to get the current file size.
+
+Mon Jan 16 00:41:33 2012  Sokolov Yura  <funny.falcon@gmail.com>
+
+	* st.c: st use function instead of macro.  In my current
+	  environment (Ubuntu 11.04 32bit gcc-4.5.2) it gives 4%
+	  performance improvement.
+
+	  https://github.com/ruby/ruby/pull/77
+
+Sun Jan 15 14:09:48 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* object.c (rb_inspect): raise the result is not compatible with
+	  the default external encoding. [ruby-core:42095] [Bug #5848]
+	  If the default external encoding is ASCII compatible, the encoding of
+	  inspected result must be compatible with it.
+	  If the default external encoding is ASCII incompatible,
+	  the result must be ASCII only.
+
+Sun Jan 15 13:21:50 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/parser/parser.rl (json_string_unescape): workaround fix
+	  for over optimization of GCC 4.7. [ruby-core:42085] [Bug #5888]
+	  http://gcc.gnu.org/bugzilla/show_bug.cgi?id=51862
+
+Sat Jan 14 22:24:09 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/callback/mkcallback.rb (gencallback): suppress unused
+	  variables.
+
+Sat Jan 14 21:56:43 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_data_to_ary): check line info table boundary.  line
+	  number 0 means no line number info is needed.  [ruby-dev:45130]
+	  [Bug #5894]
+
+Sat Jan 14 18:24:13 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* error.c (exc_equal): clear rb_thread_t::errinfo when ignore
+	  an exception under rb_protect(). [ruby-core:41979] [Bug #5865]
+
+Sat Jan 14 12:02:55 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_enc_vsprintf): relaxed the restriction.  since the
+	  implementation deeply depends on plain char, so wchar_t based
+	  encodings are not supported.
+
+Sat Jan 14 12:00:20 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (exc_equal): ignore exceptions during implicit
+	  conversion.  [ruby-core:41979] [Bug #5865]
+
+Sat Jan 14 05:58:54 2012  Eric Hodel  <drbrain@segment7.net>
+
+	* io.c (rb_io_s_read):  Fix formatting of open_args comment.  Reported
+	  by Adam Prescott.
+
+Fri Jan 13 18:41:19 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_enc_vsprintf): can be used for ASCII compatible
+	  encodings only.
+
+Fri Jan 13 18:29:06 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_mutex_unlock_th): simplified.
+
+	* thread.c (rb_barrier_waiting): fix potential overflows.
+
+Fri Jan 13 17:23:38 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_unlock): update loading table at once.
+
+Fri Jan 13 16:44:45 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (exc_equal): try implicit conversion for delegator.
+	  [ruby-core:41979] [Bug #5865]
+
+Fri Jan 13 03:46:53 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb (Shellwords#shellescape): shellescape() now
+	  stringifies the given object using to_s.
+
+	* lib/shellwords.rb (Shellwords#shelljoin): shelljoin() accepts
+	  non-string objects in the given array, each of which is
+	  stringified using to_s.
+
+	* lib/shellwords.rb: Fix rdoc markups.
+
+Fri Jan 13 03:38:36 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb (Shellwords#shellsplit): Fix a bug where
+	  consecutive backslashes in double quotes are all removed except
+	  the one at the tail.
+
+Fri Jan 13 03:28:00 2012  Luis Lavena  <luislavena@gmail.com>
+
+	* ext/socket/extconf.rb (if ipv6): only define _WIN32_WINNT if was not
+	  previously defined. This solve warnings with multiple defines in
+	  command line with GCC 4.6.1
+
+Thu Jan 12 18:44:31 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: fix r33904 and revert r33905.  initialize global
+	  variables with init_mkmf before initializing constants.
+	  [ruby-dev:45124] [Bug #5879]
+
+Thu Jan 12 13:51:00 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c (cont_restore_0): prevent optimizing out `sp'. sp is used for
+	  reserving a memory space with ALLOCA_N for restoring machine stack
+	  stored in cont->machine_stack, but clang optimized out it (and
+	  maybe #5851 is also caused by this).
+	  This affected TestContinuation#test_check_localvars.
+
+	* cont.c (cont_restore_1): revert workaround introduced in r32201.
+
+Thu Jan 12 02:14:43 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* object.c: Added examples for Object#is_a? and
+	  Object#instance_of? patched from Manoj Kumar.
+	  [Bug #5880] [ruby-core:42057]
+
+Thu Jan 12 00:57:48 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/mkmf.rb: verbose-mode can use by RM, RMDIRS, etc.
+	  (e.g. make V=1 realclean)
+
+Wed Jan 11 23:40:21 2012  Naohisa Goto  <ngoto@gen-info.osaka-u.ac.jp>
+
+	* string.c (rb_str_concat): set array element after definition
+	  to fix compile error with Fujitsu C Compiler 5.6 on Solaris 10
+	  on Sparc.  [Bug #5878] [ruby-dev:45123]
+
+Wed Jan 11 22:52:51 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* gc.c (ruby_mimmalloc): don't set allocated size to header.
+	  ruby_mimmalloc() doesn't increment allocated_size/allocations and
+	  decrement them in ruby_xfree() cause inconsistency.
+
+	* gc.c (ruby_xfree): don't decrement allocated_size/allocations if
+	  allocated size record is 0.
+
+Wed Jan 11 22:36:43 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/readline/test_readline.rb (test_completion_proc_empty_result):
+	  ensure clearance of Readline's line_buffer after the test.
+
+Tue Jan 10 21:57:38 2012  Hiroshi Shirosaki <h.shirosaki@gmail.com>
+
+	* ext/dbm/dbm.c (Init_dbm): fix a build error on mswin32.
+	  use `extern __declspec(dllimport)` for dll link with VC.
+	  [ruby-core:41996] [Bug #5869]
+
+Tue Jan 10 15:31:55 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm.c (vm_exec): refix r34162; suppress warning and add description.
+
+Tue Jan 10 15:13:58 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  use rb_memerror().
+
+Tue Jan 10 12:49:42 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c: in fact, i686-linux doesn't need to define _XOPEN_SOURCE 600.
+
+Tue Jan 10 12:44:11 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (ruby_mimmalloc): defined for objects need not rb_objspace,
+	  but should return pointer suitable for ruby_xfree;
+	  main vm and main thread.
+	  patched by Sokolov Yura. https://github.com/ruby/ruby/pull/79
+
+	* internal.h: ditto.
+
+	* vm.c (Init_BareVM): use ruby_mimmalloc.
+
+	* ext/dl/cfunc.c: #include <ruby/util.h>.
+
+	* ext/syslog/syslog.c: use xfree because it is allocated by
+	  ruby_strdup.
+
+Tue Jan 10 12:13:56 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  fix compile error.
+
+Tue Jan 10 10:41:11 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  empty completion result does not mean memory error.
+
+Tue Jan 10 02:19:22 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/test_io.rb (test_autoclose_true_closed_by_finalizer,
+	  test_autoclose_true_closed_by_finalizer): skip if IO objects are
+	  not recycled yet. [ruby-dev:45098] [Bug #5850]
+
+Tue Jan 10 00:41:28 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* lib/tempfile.rb (Tempfile#_close): clear @tempfile and @data[1] even
+	  when exception is raised at @tempfile.close. [ruby-dev:45113]
+
+	* lib/tempfile.rb (Tempfile#unlink): fix a typo.
+
+Tue Jan 10 00:32:17 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* gc.c (run_finalizer): clear rb_thread_t::errinfo when ignore
+	  an exception under rb_protect(). [ruby-dev:45113]
+
+Mon Jan  9 23:37:43 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  fix typos.
+
+Mon Jan  9 20:55:34 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c : don't embed struct heaps_slot to a heap block because it
+	  can causes copy-on-write of memory page on heap block when its
+	  free_next is rewritten.
+
+Mon Jan  9 20:26:33 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_entries): add document suggested by
+	  the thread [ruby-core:41959] [Bug #5859].
+
+Mon Jan  9 20:14:13 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (family_addrinfo): don't require protocol
+	  equality.  For example, protocol 0 and IPPROTO_TCP is not problem
+	  for TCP.
+
+Mon Jan  9 20:08:52 2012  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (family_addrinfo): return the given
+	  addrinfo object.
+	  Patch by Ippei Obayashi.  [ruby-dev:45095] [Bug #5845]
+
+Mon Jan  9 19:40:20 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/zlib/test_zlib.rb (TestZlibGzipWriter#test_writer_wrap): set
+	  binmode explicitly.
+
+Mon Jan  9 14:42:41 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: free_slots is changed Singly linked list. clear
+	  free_slots before sweep.
+
+Mon Jan  9 07:46:17 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c: i686-linux needs to define _XOPEN_SOURCE 600 for posix_memalign.
+
+Mon Jan  9 04:24:59 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (rb_objspace_free): global_List is allocated with xmalloc.
+	  patched by Sokolov Yura.  https://github.com/ruby/ruby/pull/78
+
+	* dln_find.c: remove useless replacement of free.
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  strings for readline must allocated with malloc.
+
+	* process.c (run_exec_dup2): use free; see also r20950.
+
+	* re.c (onig_new_with_source): use malloc for oniguruma.
+
+	* vm.c (ruby_vm_destruct): use free for VMs.
+
+	* vm.c (thread_free): use free for threads.
+
+Mon Jan  9 04:24:59 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* dln_find.c: remove useless replacement of free.
+
+	* ext/readline/readline.c (filename_completion_proc_call):
+	  matches should use xfree.
+
+	* ext/readline/readline.c (username_completion_proc_call): ditto.
+
+Mon Jan  9 01:12:35 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (rb_enc_uint_char): raise RangeError when added codepoint
+	  is invalid. [Feature #5855] [Bug #5863] [Bug #5864]
+
+	* string.c (rb_str_concat): ditto.
+
+	* string.c (rb_str_concat): set encoding as ASCII-8BIT when the string
+	  is US-ASCII and the argument is an integer greater than 127.
+
+	* regenc.c (onigenc_mb2_code_to_mbclen): rearrange error code.
+
+	* enc/euc_jp.c (code_to_mbclen): ditto.
+
+	* enc/shift_jis.c (code_to_mbclen): ditto.
+
+Sun Jan  8 20:31:45 2012  Narihiro Nakamura  <narihiro@netlab.jp>
+
+	* gc.c : consider header bytes which are used by malloc.
+
+Sun Jan  8 11:54:43 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (aligned_free): support MinGW. Patch by Hiroshi Shirosaki.
+
+Sun Jan  8 11:43:05 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (slot_sweep): add a assertion instead of a debug print.
+
+Sun Jan  8 01:18:19 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/-ext-/old_thread_select/test_old_thread_select.rb:
+	  avoid platform bug. [Bug #5858] [ruby-dev:45108]
+
+Sun Jan  8 00:46:34 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* gc.c: get rid of implicit narrowing conversion.
+
+Sun Jan  8 00:10:10 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: check posix_memalign(3) and memalign(3).
+
+	* gc.c (aligned_malloc): use configure's result instead of
+	  _POSIX_C_SOURCE and _XOPEN_SOURCE because they can't be used
+	  to check availability at least on FreeBSD.
+
+Sat Jan  7 22:25:50 2012  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: use Bitmap Marking algorithm to avoid copy-on-write of
+	  memory pages. See [ruby-dev:45085] [Feature #5839]
+	  [ruby-core:41916].
+
+	* include/ruby/ruby.h : FL_MARK rename to FL_RESERVED1.
+
+	* node.h : ditto.
+
+	* debug.c : ditto.
+
+	* object.c (rb_obj_clone): FL_MARK move to a bitmap.
+
+	* class.c (rb_singleton_class_clone): ditto.
+
+Sat Jan  7 00:47:07 2012  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* configure.in: always define CANONICALIZATION_FOR_MATHN.
+	  [ruby-dev:45100] [Bug #5852]
+
+Fri Jan  6 23:11:20 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* include/ruby/version.h: RUBY_API_VERSION 2.0.0
+
+Fri Jan  6 12:24:11 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* object.c (rb_inspect): raises Encoding::CompatibilityError if the
+	  result is incompatible with the default external encoding.
+	  [ruby-core:41931] [Bug #5848]
+
+Thu Jan  5 15:26:15 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (check_valid_dir): strict checking of root.
+	  GetDriveType() succeeds with non root directory as the argument,
+	  even if MSDN says that the API needs the root directory.
+	  this patch fixes a failure of test/ruby/test_file_exhaustive.rb.
+
+Thu Jan  5 12:15:55 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_file_join): separator is appended by array length - 1
+	  times. patched by Benoit Daloze [ruby-core:41901] [Bug #5841]
+
+Thu Jan  5 11:47:54 2012  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI::Parser#initialize_regexp):
+	  use \A \z instead of ^ $. [Bug #5843]
+
+Wed Jan  4 17:55:53 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* array.c (rb_ary_sample): add example for Array#sample
+	  based on patch from https://github.com/ruby/ruby/pull/74
+
+Wed Jan  4 14:24:33 2012  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_nth_len): count ascii-only run at the end.  this
+	  bug appears only when single-byte-optimization is disabled due
+	  to unknown coderange.  [ruby-core:41896] [Bug #5836]
+
+Wed Jan  4 11:32:07 2012  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (check_valid_dir): special case for a root directory.
+	  Reported by Masateru OKAMOTO at [Bug #5819].
+
+Wed Jan  4 00:19:54 2012  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb: use private instead of _xxx
+	  method name. This is Ruby code not Python code.
+	  refs #5696
+
+Tue Jan  3 23:57:37 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/rexml/parsers/baseparser.rb: rexml BaseParser uses
+	  instance_eval unnecessarily on listener add.
+	  patch from Charles Nutter. [Bug #5696] [ruby-core:41437]
+
+Tue Jan  3 20:44:13 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* README: add comment for Git user. patch from Arun Agrawal.
+	* README.ja: ditto.
+
+Tue Jan  3 15:58:22 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* thread.c: changed documentation for "thread-local" variables.
+	  patch from Julien Ammous.
+
+Tue Jan  3 15:50:12 2012  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* process.c: Fix typo. patch from Aviv Ben-Yosef.
+
+Tue Jan  3 13:43:37 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* tool/merger.rb: allow r0123 style revision number.
+
+Tue Jan  3 11:17:55 2012  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* tool/merger.rb (#version_up): version.h date should be Japanese
+	  locale date.
+
+Mon Jan  2 22:08:00 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* tool/file2lastrev.rb (VCS::detect): Add support for Subversion
+	  1.7 which adopted a whole new working directory structure.
+
+	* tool/file2lastrev.rb (VCS::detect): Simply use .each instead of
+	  .sort.reverse_each which looks too arbitrary.  If you want SVN
+	  to be tried first, then you just have to register it first as it
+	  is right now.
+
+Mon Jan  2 20:53:36 2012  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (random_bytes): use IO#read instead of
+	  IO#readpartial to make the intent more clear.
+
+Mon Jan  2 15:26:39 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/ruby/test_object.rb (test_send_with_block): add a normal case.
+
+Mon Jan  2 15:18:54 2012  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/ruby/test_object.rb (test_send_with_block): moved from
+	  bootstraptest/test_flow.rb.
+
+Mon Jan  2 15:10:11 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/test/unit/parallel.rb: use pack("m0") instead of
+	  pack("m").gsub("\n","").
+	* lib/test/unit.rb (Test::Unit::Runner::Worker#run): ditto.
+
+Mon Jan  2 15:05:09 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/test/unit.rb (Test::Unit::Runner::Worker#run): use
+	  File.basename with suffix instead of gsub.
+
+Mon Jan  2 14:55:28 2012  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/test/unit.rb (Test::Unit::Runner#_run_parallel): find may
+	  return nil and nil can not dup.
+
+Sun Jan  1 12:23:10 2012  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb (Shellwords#shellescape): Drop the //n flag
+	  that only causes warnings with no real effect.  [Bug #5637]
+
+Sat Dec 31 06:28:37 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (rb_barrier_waiting): save the number of waiting threads
+	  in RBASIC()->flags. [ruby-dev:45002] [Bug #5768]
+
+	* thread.c (rb_barrier_wait): increment and decrement around
+	  rb_mutex_lock, and use rb_barrier_waiting().
+
+	* thread.c (rb_barrier_release): use rb_barrier_waiting().
+
+	* thread.c (rb_barrier_destroy): ditto.
+
+Mon Dec 26 17:20:10 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm.c (vm_exec): add guard to prevent optimization for LLVM clang.
+
+Fri Dec 30 17:01:12 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* vm_eval.c (rb_f_send): fix obj.send() documentation issue.
+	 [Bug #5125] [ruby-core:38633]
+
+Thu Dec 29 22:36:16 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* lib/test/unit.rb (Test::Unit::Runner::Worker#_run_parallels): fix
+	  premature exit when all workers' status are :ready or :prepare.
+	  [ruby-dev:45061] [Bug #5822]
+
+Thu Dec 29 01:51:13 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* include/ruby/ruby.h: fix #error pragma. LLP64 platform is supported.
+
+	* include/ruby/st.h: ditto.
+
+Wed Dec 28 11:22:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::Entry_#entries): use utility method
+	  instead of typoed regexp.  [ruby-core:41829] [Bug #5817]
+
+Wed Dec 28 02:08:04 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* vm_insnhelper.c (unknown_keyword_error): add GC guard to prevent
+	  intermediate object from GC.
+
+Tue Dec 27 22:34:54 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb (Worker#close): "closing IO if IO is closed"
+	  should be "closing IO if IO isn't closed"
+
+Tue Dec 27 22:04:27 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (st_update): new function to lookup the given key and
+	  update the value.  [ruby-dev:44998]
+
+Tue Dec 27 21:17:33 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.h (rb_args_info): change pre_args_num and post_args_num as
+	  int, to match with rb_iseq_t.
+
+	* parse.y (new_args_gen): check overflow.
+
+Mon Dec 26 22:38:35 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_insnhelper.c (unknown_keyword_error): make it kind a error
+	  message when unknown keyword is given.  It require more work.
+	  See [ruby-core:40518] and [ruby-core:40541] in detail.
+
+Mon Dec 26 22:31:07 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_core.h (struct rb_iseq_struct), compile.c (iseq_set_arguments),
+	  iseq.c (rb_iseq_parameters), vm_insnhelper.c
+	  (vm_callee_setup_arg_complex): support Method#parameters for keyword
+	  arguments.  The provisional spec is what Benoit Daloze proposed.
+	  [ruby-core:40541]
+
+	* test/ruby/test_keyword.rb: add a test for above.
+
+Mon Dec 26 22:15:27 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_core.h (struct rb_iseq_struct), compile.c (iseq_set_arguments,
+	  iseq_compile_each), vm_insnhelper.c (vm_callee_setup_arg_complex):
+	  implement keyword arguments.  See [ruby-core:40290]
+	  The feature is promised to be included in 2.0, but the detail spec
+	  is still under discussion; this commit is a springboard for further
+	  discussion.  Please try it and give us feedback.
+	  This commit includes fixes for some problems reported by Benoit
+	  Daloze <eregontp AT gmail.com> [ruby-core:40518] and Marc-Andre
+	  Lafortune <ruby-core-mailing-list AT marc-andre.ca>
+	  [ruby-core:41772].
+
+	* iseq.c (iseq_free, prepare_iseq_build): bookkeeping.
+
+	* test/ruby/test_keyword.rb: add tests for keyword arguments.
+
+	* test/ripper/dummyparser.rb (class DummyParser): temporal fix for
+	  ripper test.
+
+Mon Dec 26 22:00:17 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* node.h, node.c, parse.y: implement a parser part for keyword
+	  arguments.
+	  This is a preparation for keyword argument (see [ruby-core:40290]).
+
+	* gc.c (gc_mark_children): bookkeeping.
+
+Mon Dec 26 21:03:18 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* node.h, parse.y (new_args_gen), compile.c (iseq_set_arguments): use
+	  struct rb_args_info instead of NODEs.
+	  This is a preparation for keyword argument (see [ruby-core:40290]).
+
+	* node.c (dump_node), gc.c (gc_mark_children, obj_free): bookkeeping.
+
+Mon Dec 26 20:59:51 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* node.h, parse.y (lambda, f_larglist): remove NEW_LAMBDA hack.
+	  This is a preparation for keyword argument (see [ruby-core:40290]).
+
+Mon Dec 26 22:01:19 2011  Hiroshi Shirosaki <h.shirosaki@gmail.com>
+
+	* io.c (rb_sys_fail_path): move the definition.
+	  Move above for using it in set_binary_mode_with_seek_cur().
+
+	* io.c (set_binary_mode_with_seek_cur): fix improper seek cursor.
+	  Seeking file cursor with setting binary mode has possibility to
+	  cause infinite loop. Fixed the bug and refined error handling.
+	  Introduced at r34043.
+
+	  And cleanups as below.
+	  Remove unnecessary parentheses of `fptr`.
+	  Use return value of setmode().
+
+	* test/ruby/test_io_m17n.rb
+	  (TestIO_M17N#test_seek_with_setting_binmode): add a test for above.
+	  [ruby-core:41671] [Bug #5714]
+
+Mon Dec 26 17:01:14 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (LIBRUBY_A): depends on main.o since r33774.
+	  [ruby-core:41786] [Bug #5796]
+
+Mon Dec 26 13:07:08 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (TestIO#test_autoclose): Tempfile.new doesn't
+	  accept the block argument.
+
+Mon Dec 26 13:06:52 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb: Avoid zombie processes on "--separate" option
+	  added at r34121.
+
+Mon Dec 26 04:01:23 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_cipher.c: Update and complete documentation.
+
+Sun Dec 25 23:16:11 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/test_parallel.rb (test_separate): Test for "--separate"
+	  option (r34121)
+
+Sun Dec 25 22:39:49 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb (_run_parallel):
+	  New option "--separate" for test/unit; when running tests with this
+	  option, a job process will be restarted after one test file has done.
+	  This means all test files will run with separated process.
+
+	* lib/test/unit/parallel.rb: Fix for above. Now parallel.rb puts
+	  "ready!" for first ready, "ready" for afters.
+
+Sun Dec 25 00:02:15 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* configure.in: change --with-ntver to --with-winnt-ver to be more
+	  descriptive in the context. [ruby-core:41794]
+
+Sat Dec 24 23:25:15 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* configure.in: add --with-ntver option to match win32/configure.bat
+	  functionality. Set 0x0501 as default. [ruby-core:35010]
+	  [ruby-core:35035]
+
+Sat Dec 24 12:38:53 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_call): get rid of optimization-out by clang.
+
+	* proc.c (rb_proc_call, rb_proc_call_with_block): ditto.
+
+Sat Dec 24 10:56:32 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_readline): check if outstream
+	  is closed to get rid of a bug of readline 6.  [ruby-dev:45043]
+	  [Bug #5803]
+
+Sat Dec 24 06:59:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/readline/test_readline.rb (test_line_buffer__point): use
+	  lambda not to exit entire method by "return".  or "next" for
+	  proc.  [ruby-dev:45042] [Bug #5802]
+
+Sat Dec 24 01:20:39 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_eval.c (send_internal): PASS_PASSED_BLOCK_TH must be placed
+	  just before calling rb_call0.
+
+	* bootstraptest/test_flow.rb: add a test for above.
+
+Sat Dec 24 00:55:16 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tempfile.rb (Tempfile#initialize): warn if a block is given.
+
+Fri Dec 23 16:14:30 2011  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c (readline_attempted_completion_function):
+	  in Readline module with GNU Readline 6 case, Readline module
+	  resets completion_append_character to " ", after it executes
+	  completion. So, Readline module stores
+	  completion_append_character, and Readline module always sets it
+	  after Readline module executes completion. [ruby-dev:43456]
+	  [Feature #4635]
+
+Fri Dec 23 15:59:05 2011  TAKAO Kouji  <kouji@takao7.net>
+
+	* ext/readline/readline.c (Init_readline): libedit check
+	  rl_getc_function only when rl_initialize() is called, and
+	  using_history() call rl_initialize(). This assignment should be
+	  placed before using_history(). [ruby-core:40641] [Bug #5539]
+
+Fri Dec 23 10:14:47 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/thread/test_queue.rb (test_thr_kill): show the number of loop
+	  run when the test failed.
+
+Fri Dec 23 09:23:48 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/test_pty.rb (test_pty_check_default): call PTY.check until
+	  "cat" command is finished.
+
+Fri Dec 23 06:03:00 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* common.mk: add "check succeeded" message.
+
+	* README, README.ja: follow above change.
+
+Fri Dec 23 06:00:39 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/bigdecimal/bigdecimal.h: add satisfy cc-mode comment.
+	* util.c: ditto.
+
+Fri Dec 23 00:08:25 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/test_pty.rb (test_pty_check_default): "cat" may not terminated
+	  in the 0.1 second.
+
+Thu Dec 22 23:37:25 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_thread.rb (test_condvar_timed_wait): don't test the
+	  maximum sleep time.  Ruby is not a real-time system.
+
+Thu Dec 22 22:37:45 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread_pthread.c (ping_signal_thread_list): remove return value.
+	* thread_pthread.c (check_signal_thread_list): add a new function to
+	  check if signal thread list is empty.
+	* thread_pthread.c (thread_timer): check signal thread list after
+	  timer_thread_function(). main thread might be added into signal thread
+	  list during timer_thread_function().
+
+Thu Dec 22 00:40:24 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/bigdecimal/bigdecimal.c (VpMult, VpCtoV, VpSqrt): remove assigned
+	  but unused variables.
+
+Wed Dec 21 18:28:22 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (newline.c, miniprelude.c): revert r33949 because the change
+	  broke mswin build, and the changer said no reason about the change.
+	  [ruby-dev:45016] [Bug #5783]
+
+Wed Dec 21 12:35:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_s_allocate): follow
+	  Allocation Framework.  [Bug #5775]
+
+Wed Dec 21 02:25:36 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/emitter.c: fixing clang warnings. Thanks Joey!
+
+Wed Dec 21 01:06:00 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/bigdecimal/README: Update redmine.ruby-lang.org to bugs.ruby-lang.org
+	* ext/socket/ancdata.c: ditto
+	* test/-ext-/wait_for_single_fd/test_wait_for_single_fd.rb: ditto
+	* test/syck/test_yaml.rb: ditto
+	* doc/ChangeLog-1.9.3: ditto
+
+Tue Dec 20 23:50:12 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* PStore content update perf optimization. Patch by Masaki Matsushita.
+	  See #5248.
+
+	* lib/pstore.rb (save_data):
+
+	  * Delete inadequate Marshal check.
+
+	  * Deferred file truncation: when writing the new content, truncate
+	    the saved file to the data size after writing the data, instead of
+	    truncating whole bytes before writing data.
+
+	  * Deferred MD5 calculation: when comparing MD5 hash to check the
+	    content modification, calculate MD5 hash of new data iif the
+	    content length is differ from the old one.
+
+	  * Compare content size with String#bytesize instead of String#size.
+
+Tue Dec 20 21:00:30 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: uses to_integer instead.
+	* test/date/test_switch_hitter.rb: added a test.
+
+Tue Dec 20 15:04:18 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* Make sure to clear $! when ignoring an exception
+
+	* ext/openssl/ossl.c (ossl_pem_passwd_cb0, ossl_verify_cb):
+	  pem_passwd_cb and verify_cb ignores the exception raised in a
+	  callback proc so it should clear $! for subsequent execution.
+
+	  That's said, both subsequent processes for pem_passwd_cb and
+	  verify_cb raises another exception before leaking $! to Ruby world.
+	  We cannot test this fix in Ruby land.
+
+	* test/openssl/test_pkey_rsa.rb
+	  (test_read_private_key_pem_pw_exception): Test for pem_passwd_cb +
+	  exception.
+
+Tue Dec 20 11:49:13 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/date/test_date_base.rb (test_jd): tests for
+	  [ruby-dev:45008].
+
+Tue Dec 20 10:20:48 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/date_core.c (wholenum): fix the type of the return value.
+
+Tue Dec 20 05:03:24 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* README.ja:  Update redmine.ruby-lang.org to bugs.ruby-lang.org
+	* README:  ditto
+	* common.mk:  ditto
+	* man/erb.1:  ditto
+	* man/irb.1:  ditto
+	* man/ri.1:  ditto
+	* man/ruby.1:  ditto
+	* sparc.c:  ditto
+	* tool/install-sh:  ditto
+
+Tue Dec 20 02:15:18 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c:  [ruby-dev:45008].
+
+Sun Dec 18 18:52:37 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* vm.c (vm_define_method): improve guard of iseq from GC.  Fix
+	  failure or segmentation fault in test_singleton_method(TestGc)
+	  on sparc Solaris10 compiled with Oracle Solaris Studio 12.2.
+	  [Bug #5762] [ruby-dev:45000] [Bug #4178]
+
+Sun Dec 18 14:34:31 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (Init_bigdecimal): does not follow
+	  allocation framework right now.  [ruby-core:41710] [Bug #5773]
+
+Sun Dec 18 12:42:48 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: BigDecimals can be restored
+	  from YAML.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: BigDecimals can be dumped
+	  to YAML.
+	* test/psych/test_numeric.rb: tests for BigDecimal serialization
+
+Sun Dec 18 12:03:13 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: Strings that look like dates
+	  should be treated as strings and not dates.
+
+	* test/psych/test_scalar_scanner.rb: corresponding tests.
+
+Sun Dec 18 09:43:21 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/thread/test_queue.rb (test_thr_kill): extend timeout.
+	  this test takes a long time at slow machine.
+
+Sun Dec 18 09:36:51 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/ruby/envutil.rb (invoke_ruby): remove :timeout option before
+	  pass it to Kernel#spawn.
+
+Fri Dec 16 17:18:38 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README, README.ja: 'make check' is preferable to 'make test'.
+
+Thu Dec 15 23:16:13 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* error.c (builtin_type_name): don't return pointer to the buffer of
+	  temporary String object.
+
+Thu Dec 15 17:56:58 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_type): make typed data.
+
+Thu Dec 15 17:40:28 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_check_type): fix typo.
+
+Thu Dec 15 14:48:35 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/strscan/strscan.c: use typed data with
+	  onig_region_memsize().
+
+Thu Dec 15 14:33:33 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_check_typeddata): refine error message with
+	  including expected struct name.
+
+Thu Dec 15 13:15:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regcomp.c (onig_region_memsize): implemented for memsize_of().
+
+	* ext/objspace/objspace.c (memsize_of): use it.
+
+Thu Dec 15 10:44:54 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_reject_bang, rb_ary_delete_if): update rdoc.
+	  documentation from Thomas Leitner <t_leitner AT gmx.at> in
+	  [ruby-core:41616].  [Bug #5752]
+
+Thu Dec 15 10:10:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_require.rb (test_race_exception): get rid of
+	  not-guaranteed timing issue.  [ruby-core:41655] [Bug #5754]
+
+Wed Dec 14 21:58:42 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io_m17n.rb
+	  (TestIO_M17N#test_{read_with_binmode_and_get[cs]}): only for Windows.
+
+Wed Dec 14 19:57:23 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk,Makefile.in,win32/Makefile.sub (ECHO1): move platform
+	  specific hack from common.mk to Makefile.in (and win32/Makefile.sub).
+	  [Bug #5711]
+
+	* lib/mkmf.rb: we can generate Makefile as we like.
+
+Wed Dec 14 19:22:33 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_fd_is_text): new function.
+
+	* win32/win32.c (init_stdhandle): set default mode of stdin as binmode.
+
+	* io.c (set_binary_mode_with_seek_cur): new function to replace
+	  SET_BINARY_MODE_WITH_SEEK_CUR macro.  now returns previous mode of the
+	  fd and take care of LF in rbuf.
+
+	* io.c (do_writeconv): set text mode when needed.
+
+	* io.c (io_read): need to change the mode of the IO to binmode
+	  temporally when the length for IO#read, because IO#read with length
+	  must behave so.
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#test_{read_with_length,
+	  read_with_length_binmode,get[cs]_and_read_with_binmode,
+	  read_with_binmode_and_get[cs],read_write_with_binmode}): tests for
+	  above changes.
+
+	all patches are written by Hiroshi Shirosaki. [ruby-core:41496]
+	[Feature #5714]
+
+Wed Dec 14 15:28:31 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (str_encode): about the extension of :fallback
+	  option since 1.9.3.
+
+Wed Dec 14 12:19:59 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_unlock): release loading barrier and then remove it
+	  from loading_table if it is not in-use.  [Bug #5754]
+
+	* thread.c (rb_barrier_release, rb_barrier_destroy): return
+	  whether any other threads are waiting on it.
+
+Wed Dec 14 11:23:45 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (ubf_select): call rb_thread_wakeup_timer_thread()
+	  only when it is not timer_thread. [Bug #5757] [ruby-dev:44985]
+	  patched by Tomoyuki Chikanaga.
+
+Wed Dec 14 10:20:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_lock): delete the loading barrier if it has been
+	  destroyed.
+
+	* thread.c (rb_barrier_wait): return nil for recursive lock
+	  instead of false, to distinguish it from destroyed barrier.
+
+Wed Dec 14 01:24:55 2011  okkez  <okkez000@gmail.com>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): fix memory
+	  leak. [ruby-dev:44904] [Bug #5688]
+
+Wed Dec 14 00:01:15 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): point method name line.  [ruby-core:40936]
+	  [Bug #5614]
+
+Tue Dec 13 23:43:48 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* error.c (name_err_mesg_to_str): clear rb_thread_t::errinfo when
+	  ignore exception under rb_protect(). [ruby-core:41612] [Bug #5755]
+
+	* test/ruby/test_exception.rb (test_exception_in_name_error_to_str):
+	  add a corresponding test.
+
+Tue Dec 13 16:13:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_unlock): all threads requiring one file should
+	  share same loading barrier, so it must be kept alive while those
+	  are waiting on it.  [ruby-core:41618] [Bug #5754]
+
+Tue Dec 13 07:30:14 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/webrick/httpresponse.rb (setup_header): 1xx responses
+	  are allowed to have Keep-Alive connections.
+
+	* test/webrick/test_httpresponse.rb: corresponding test.
+
+Tue Dec 13 07:13:28 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/webrick/httpresponse.rb (setup_header): 204 and 304 responses
+	  are allowed to have a Keep-Alive connection. [ruby-core:41581]
+
+	* test/webrick/test_httpresponse.rb: corresponding test.
+
+Tue Dec 13 06:29:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_magic_comment): should pass the proper value.
+	  [ruby-dev:44984][Bug #5753]
+
+Tue Dec 13 05:50:07 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): splat single
+	  argument if optional arguments are defined not only mandatory or
+	  post arguments.  [ruby-core:41557] [Bug #5730]
+
+Mon Dec 12 22:35:39 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* parse.y (stmt_or_begin): changed the error message for BEGIN not
+	  at toplevel.  [ruby-dev:44963] [Bug #5738]
+
+Mon Dec 12 17:29:01 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* README: Fixed SupportedPlatforms URL in the README.
+	  patched by eMxyzptlk. https://github.com/ruby/ruby/pull/62
+
+Mon Dec 12 17:26:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_feature_p): lazy assigned load_path searched in
+	  loading_table were not expanded, but all features, pushed to
+	  loading table, are expanded.  a patch by Yura Sokolov
+	  <funny.falcon AT gmail.com> in [ruby-core:41545]. [Bug #5727]
+
+Mon Dec 12 15:41:03 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_truncate): fix typo.  patched by
+	  Nick Howard <ndh AT baroquebobcat.com>.
+	  https://github.com/ruby/ruby/pull/65
+
+Sun Dec 11 12:19:17 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: includes the sequence number of UID in a error
+	  message.  suggested by art lussos.
+	  [ruby-core:41413] [Feature #5692]
+
+Sun Dec 11 11:42:10 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/syslog/syslog.c: fix a typo. [ruby-core:41585] [Bug #5740]
+
+Sun Dec 11 10:48:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (exit_initialize): deal with true and false as well as
+	  Kernel#exit.  [ruby-dev:44951] [Bug #5728]
+
+Sun Dec 11 10:37:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_check_to_int): new function to convert a VALUE to
+	  an Integer if possible, but returns nil instead of raising an
+	  exception otherwise.
+
+Sun Dec 11 10:34:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_exit_status_code): extract from rb_f_exit_bang and
+	  rb_f_exit.  assume 0 to be success in Kernel#exit! too.
+
+Fri Dec  9 19:24:31 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/iso-8859-16-tbl.rb: add ISO-8859-16 converter.
+
+	* enc/trans/single_byte.trans: ditto.
+
+Fri Dec  9 14:28:40 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_path_convert): don't convert it when the path string is
+	  ascii only. [ruby-core:41556] [Bug #5733]
+	  tests are contributed by nobu.
+
+Fri Dec  9 08:00:15 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* include/ruby/win32.h: undef stat to silence mingw-w64 stat
+	  redefinition warnings (GCC 4.6.3).
+
+Thu Dec  8 23:38:24 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* variable.c (set_const_visibility): clear inline-cache when constant's
+	  visibility is modified. [ruby-dev:44929]
+
+	* test/ruby/test_module.rb (test_private_constants_clear_inlinecache):
+	  add test for it.
+
+Thu Dec  8 23:26:11 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): should sort after map, not before
+	  it.  in this case there is no difference, but we should write better
+	  code.  this bad smell was caught by nagachika.
+
+Thu Dec  8 22:31:13 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile): need to sort the array of current
+	  srcs before comparing to the sorted old srcs.
+	  fixed the problem that the configuring stage of exts were always
+	  run, introduced at r33801.
+
+Thu Dec  8 13:26:24 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rexml/test_order.rb (OrderTester#test_more_ordering): use
+	  Zlib::GzipReader.open instead of Zlib::GzipReader.new with File.new.
+	  fixed a test error on Windows introduced at r33946.
+
+Thu Dec  8 13:11:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_process.rb (TestProcess#test_sete[gu]id): silently
+	  skip if not implemented such functions (such as, on Windows).
+	  fixed test errors on Windows introduced at r33953.
+
+Thu Dec  8 12:57:50 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: forgotten to define HAVE_SOCKETPAIR for
+	  windows.
+	  fixed test errors on Windows introduced at r33947.
+
+Thu Dec  8 12:11:06 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_WERROR_FLAG): append all warning flags which
+	  are enabled to compile, so that printf format modifiers properly
+	  fail.  [ruby-core:41351] [Bug #5679]
+
+Thu Dec  8 07:20:15 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/re.rdoc:  Document difference between match and =~, options with
+	  Regexp.new and global variables.  Patch by Sylvain Daubert.
+	  [Ruby 1.9 - Bug #5709]
+
+Thu Dec  8 06:53:10 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/re.rdoc:  Fix example code to match documentation.  Patch by
+	  Jarno Lamberg.  [Ruby 1.9 - Bug #5624]
+
+Wed Dec  7 19:04:22 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rpath): fix typo in the help string.  a patch from
+	  Yuji Yamano <yyamano AT kt.rim.or.jp> in [ruby-list:48568].
+
+Wed Dec  7 18:55:56 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_set_top_stack, vm_set_eval_stack): check for stack
+	  overflow with stack_max before push new frame. [ruby-core:41520]
+	  [Bug #5720]
+
+	* vm.c (vm_set_main_stack): no stack overflow chances after
+	  vm_set_eval_stack().
+
+Wed Dec  7 09:58:15 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/bigdecimal/bigdecimal.c:  Document +@, -@, hash, INFINITY, Nan.
+	  Patch by Sylvain Daubert.  [Ruby 1.9 - Feature #5622]
+
+Wed Dec  7 09:48:00 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* io.c (Init_IO):  Mention io/console methods.  [Ruby 1.9 - Bug #5602]
+	* ext/io/console/console.c:  Mention that io/console must be required
+	  similar to lib/time.rb
+
+Wed Dec  7 08:04:31 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb (module Psych): parse and load methods take
+	  an optional file name that is used when raising Psych::SyntaxError
+	  exceptions
+	* ext/psych/lib/psych/syntax_error.rb (module Psych): allow nil file
+	  names and handle nil file names in the exception message
+	* test/psych/test_exception.rb (module Psych): Tests for changes.
+
+Tue Dec  6 18:26:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/dbm.c: use db_version() instead of DB_VERSION_STRING to
+	  detect runtime Berkeley DB version.
+	  use dpversion instead of _QDBM_VERSION to detect runtime QDBM
+	  version.
+	  [ruby-dev:44948]
+
+Tue Dec  6 12:30:41 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: detect gdbm_version in libgdbm.
+
+	* ext/dbm/dbm.c: make DBM::VERSION more informative for gdbm, qdbm and
+	  Berkeley DB 1.x.  [ruby-dev:44944]
+
+Tue Dec  6 07:26:37 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* range.c:  Improve documentation for Range.  Patch by Chris Zetter.
+	  [Ruby 1.9 - Bug #5656]
+
+Mon Dec  5 19:08:04 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regparse.c (PFETCH_READY): separate gcc specific trick.
+
+Mon Dec  5 19:01:59 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_seteuid_m): fix argument.
+
+	* test/ruby/test_process.rb (test_geteuid): fix typo.
+
+	* test/ruby/test_process.rb (test_getegid, test_set[eg]uid): add.
+
+Mon Dec  5 18:56:55 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (big_rshift), compile.c (validate_label,
+	  iseq_build_from_ary_exception), cont.c (cont_capture), dir.c
+	  (dir_open_dir), gc.c (objspace_each_objects), io.c (pipe_open)
+	  (rb_io_advise), parse.y (parser_compile_string)
+	  (rb_parser_compile_file), proc.c (binding_free), process.c
+	  (rb_proc_exec_n, rb_seteuid_core, proc_setegid, rb_setegid_core)
+	  (p_uid_exchange, p_gid_exchange), regparse.c (strdup_with_null),
+	  signal.c (sig_dfl), vm.c (rb_iseq_eval, rb_iseq_eval_main),
+	  vm_insnhelper.c (vm_expandarray): suppress
+	  unused-but-set-variable warnings.
+
+	* class.c (rb_obj_methods), compile.c (iseq_compile_each),
+	  iseq.c(iseq_load, rb_iseq_parameters), pack.c (pack_pack),
+	  regcomp.c (is_not_included, update_string_node_case_fold),
+	  transcode.c (rb_econv_open0, make_replacement),
+	  vm_eval.c (raise_method_missing): remove unused variable.
+
+	* signal.c (reserved_signal_p): static.
+
+Mon Dec  5 14:27:23 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/{subst.h,win32.h}, ext/socket/rubysocket.h: revert
+	  r33876.  [ruby-core:41475] [Bug #5706]
+
+	* ext/socket/extconf.rb: the alternative hack for [Bug #5675].
+
+Mon Dec  5 10:18:45 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/zlib/zlib.c (rb_gzreader_initialize): revert a part of r33937.
+	  1st, to change the mode of an IO is very sensitive problem, so
+	  the maintainer of this library should judge it.
+	  2nd, usually Zlib::GzipReader.new is not called directly.  #initialize
+	  is called via .open, and in the method the I/O is opened in binary
+	  mode, so there is no problem without changing the mode in #initialize.
+
+Sun Dec  4 22:53:12 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tempfile.rb: don't use lock directory.  [ruby-dev:39197]
+
+Sun Dec  4 22:34:43 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tempfile.rb (Tempfile::MAX_TRY): remove unused constant.
+
+Sun Dec  4 12:11:28 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/pp.rb: fix rdoc.
+
+Sun Dec  4 12:03:16 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#methods): Kernel#methods receives
+	  zero or one argument. [ruby-core:37118] [Bug #4882]
+
+Sun Dec  4 10:15:00 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* ext/zlib/zlib.c (rb_gzreader_initialize): use binary mode by default
+	  under Windows. Patch by Hiroshi Shirosaki. [ruby-core:40706]
+	  [Feature #5562]
+
+	* include/ruby/encoding.h (void rb_econv_binmode): define NEWLINE
+	  decorator.
+
+	* io.c (rb_cloexec_fcntl_dupfd): Introduce NEED_READCONV and
+	  NEED_WRITECONV to replace universal newline decorator by CRLF only
+	  when required to improve file reading and writing under Windows.
+	  Patch by Hiroshi Shirosaki. [ruby-core:40706] [Feature #5562]
+	* io.c (do_writeconv): adjust binary mode if required.
+	* io.c (read_all, appendline, swallow, rb_io_getline_1): ditto.
+	* io.c (io_getc, rb_io_each_codepoint, rb_io_ungetc): ditto.
+	* io.c (rb_io_binmode, rb_io_ascii8bit_binmode): ditto.
+	* io.c (rb_io_extract_modeenc, rb_sysopen): ditto.
+	* io.c (pipe_open, prep_stdio, io_encoding_set): ditto.
+	* io.c (rb_io_s_pipe, copy_stream_body): ditto.
+
+	* test/ruby/test_io_m17n.rb (EOT): add test for pipe and stdin in
+	  binary mode.
+
+	* win32/win32.c (init_stdhandle): remove O_BINARY from stdhandle
+	  initialization.
+	* win32/win32.c (rb_w32_write): use FTEXT mode accordingly.
+
+Sat Dec  3 20:49:16 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* variable.c (set_const_visibility): print a warning when no argument
+	  is passwd to Module#private_constant.  [ruby-list:48558]
+
+	* vm_method.c (set_method_visibility): ditto for
+	  Module#private_class_method.
+
+Sat Dec  3 20:43:14 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* variable.c (set_const_visibility): Module#private_constant has
+	  changed the visibility of only the first argument.  Now it changes
+	  all of them.  [ruby-list:48558]
+
+	* test/ruby/test_module.rb: add a test for above.
+
+Sat Dec  3 07:17:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (CFLAGS): append ARCH_FLAG.
+
+	* configure.in (ARCH_FLAG): exclude from CFLAGS.
+
+	* configure.in (UNIVERSAL_INTS): include short int.  fix for
+	  test/mkmf.
+
+Fri Dec  2 15:48:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpAllocReal): reduce extra frac.
+
+Fri Dec  2 15:41:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check whether -pie or -Wl,-pie is valid as
+	  LDFLAGS.  [ruby-core:41438] [Bug#5697]
+
+	* configure.in: use $linker_flag for LDFLAGS option which is not
+	  limited to particular platforms.
+
+Thu Dec  1 23:21:58 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread_pthread.c (thread_timer): call prctl(PR_SET_NAME) only if
+	  PR_SET_NAME is available.
+
+Thu Dec  1 22:31:16 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (linux_get_maxfd): change local variable name.
+
+Thu Dec  1 16:59:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/extconf.rb: add arguments for macro calls.
+	  [ruby-core:41370] [Bug#5681]
+
+Thu Dec  1 16:20:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#try_func): fix broken patch at r33834.
+
+Thu Dec  1 14:43:17 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.h (Real): suppress false warning from
+	  clang.  [ruby-core:41418] [Bug#5693]
+
+Thu Dec  1 10:31:55 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LDFLAGS): -fstack-protector is always needed to
+	  link static library created with it.  [ruby-core:41387]
+	  [Bug#5686]
+
+Thu Dec  1 07:03:51 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: add sys/prctl.h test.
+	* thread_pthread.c (thread_timer): call prctl(PR_SET_NAME) to change
+	  thread name. It may help to debug.
+
+Wed Nov 30 23:35:45 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* variable.c (rb_path2class): don't raise NameError when the middle
+	  constant of the path is not defined but defined on toplevel.
+	  [ruby-core:41410] [Bug #5691]
+
+Wed Nov 30 20:02:02 2011  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* transcode.c: Simplified rb_econv_binmode, avoided a warning on cygwin.
+
+Wed Nov 30 08:57:07 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mkmf.rb:  Use MakeMakefile's rm_f to avoid conflict with Rake or
+	  FileUtils.
+	* test/ruby/test_module.rb:  Hide MakeMakefile's inclusion in Object
+
+Wed Nov 30 09:12:43 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rdoc/encoding.rb (RDoc::Encoding.read_file): fixup newline chars
+	  on Windows.
+	  see https://github.com/rdoc/rdoc/issues/87
+
+	* test/rdoc/test_rdoc_markup_pre_process.rb
+	  (TestRDocMarkupPreProcess#test_include_file,
+	  TestRDocMarkupPreProcess#test_include_file_encoding_incompatible):
+	  follow above change.
+
+Wed Nov 30 09:09:37 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (parse): parse method can take an option file
+	  name for use in exception messages.
+	* test/psych/test_parser.rb: corresponding tests.
+
+Tue Nov 29 09:07:59 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mkmf.rb:  Fix indentations of constants at end of module.
+	  Document some constants.
+
+Tue Nov 29 09:58:23 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_write_error2): suppress unused variable warning.
+
+Tue Nov 29 07:45:26 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mkmf.rb:  Wrap comments to 78 columns and clean up formatting.
+
+Tue Nov 29 05:54:18 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/mkmf.rb:  Wrap mkmf.rb in module MakeMakefile to clean up Object
+	  documentation.  [Ruby 1.9 - Feature #5658]
+	* ext/extmk.rb:  Use MakeMakefile::CONFIG instead of Object::CONFIG
+	* test/mkmf/base.rb:  ditto
+
+Tue Nov 29 00:08:57 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (INSTRUBY_ARGS): added --mantype to apply mdoc2man.rb
+	  to man pages.  Fixes #5598.
+	  (do-install-nodoc, do-install-local, do-install-man,
+	  dont-install-nodoc, dont-install-local, dont-install-man):
+	  No longer needs --mantype.
+
+	  Reported by Rainer Orth <ro AT cebitec.uni-bielefeld.de>,
+	  patch by George Koehler <xkernigh AT netscape.net>.
+
+Mon Nov 28 22:26:31 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rake/test_rake_directory_task.rb
+	  (TestRakeDirectoryTask#test_directory_win32): shouldn't create any
+	  file/directory on root directory.  create on @tempdir (= Dir.pwd).
+	  see https://github.com/jimweirich/rake/issues/91
+
+Mon Nov 28 12:57:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_write_error2): fwrite() returns ssize_t.
+
+Mon Nov 28 12:47:19 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (nodetype, nodeline): static.  these functions are for
+	  debugging, and not intend to be public.
+
+Mon Nov 28 12:37:54 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gc.c (initial_params): static.  it seems to be forgotten at r33501.
+
+Mon Nov 28 12:32:24 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h, win32/win32.c (GetCurrentThreadHandle): remove
+	  unused old API.
+
+Mon Nov 28 12:29:20 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/mkexports.rb (Exports#initialize): remove old symbol name.
+
+Mon Nov 28 12:15:28 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/mkexports.rb (Exports#read_substitution): need to read
+	  from subst.h too.  [Bug #5675]
+
+Mon Nov 28 11:46:35 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_flush): release GVL during fsync() on Windows.
+
+Mon Nov 28 11:00:25 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/subst.h: typo of r33876.
+
+Mon Nov 28 10:36:00 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/subst.h: moved Windows specific substitutions from
+	  win32.h.
+
+	* ext/socket/rubysocket.h: include ruby/subst.h.  [Bug #5675]
+
+Mon Nov 28 10:20:58 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{Makeilfe.sub,win32.c} (FILE_COUNT, FILE_READPTR): move the
+	  definitions from config.h to win32.c.  I dared to have left such
+	  macros, for other future compiler support.
+	  [ruby-core:41313] [Bug #5674]
+
+Mon Nov 28 09:28:30 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_uchmod): typo.  [Bug#5671] [ruby-dev:44898]
+
+	* test/ruby/test_file.rb (TestFile#test_chmod_m17n): test of above bug.
+
+Sun Nov 27 21:25:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: added -fno-strict-overflow. it suppress annoying
+	  -Wstrict-overflow warning.
+
+Sun Nov 27 20:58:02 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_write_error2): get rid of warning on linux. fwrite
+	  of glibc is tagged __attribute__ ((__warn_unused_result__))
+	  if _FORTIFY_SOURCE != 0.
+	* vm_dump.c (rb_vm_bugreport): ditto.
+
+Sun Nov 27 19:09:02 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (stack_protector): disable on mingw.  [Bug#5676]
+
+	* Makefile.in (DLDFLAGS): also needs -fstack-protector.
+	  [Bug#5676]
+
+Sun Nov 27 14:13:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: add -fstack-protector into XLDFLAGS as well as
+	  XCFLAGS if stack-protector is used.
+
+Sun Nov 27 13:09:25 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: workaround to avoid MacOS X build error.
+	  Maybe autoconf 2.61 is slightly buggy. [ruby-core:41316]
+
+Sun Nov 27 04:57:11 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (--no-undefined): r33840 breaks FreeBSD and DragonFly
+	  with gcc 4.4 or later. Their environ is in /usr/libexec/ld-elf.so.1,
+	  so it will be false negative.
+
+Sun Nov 27 04:55:45 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP::SSL_IVNAMES): rerefix 33701.
+	  SSL_ATTRIBUTES stores names for set_params, they are symbol.
+	  SSL_IVNAMES stores instance variable names.
+
+Sun Nov 27 00:16:07 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_body): use 0666 for permission argument for open.
+	  [ruby-core:40865]
+
+Sat Nov 26 23:01:38 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_engine.rb: remove side effect of generic engine
+	  load by explicitly loading software-based "openssl" engine for
+	  all tests.
+
+Sat Nov 26 20:41:48 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP.get_response): enable use_ssl
+	  if given URI object is https.
+	  patched by Mark Ferlatte [ruby-core:40665] [Bug #5545]
+
+	* lib/net/http.rb (Net::HTTP.post_form): ditto.
+
+Sat Nov 26 20:01:18 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP::SSL_ATTRIBUTES): refix 33701.
+	  store instance variable symbol names.
+
+Sat Nov 26 15:40:25 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (script): should be ./configure
+
+Sat Nov 26 15:39:18 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (before_script): wrong name, sorry.
+
+Sat Nov 26 15:31:34 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml (before-script): autoconf required.
+
+Sat Nov 26 15:24:05 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml: Travis enable.
+
+Sat Nov 26 10:47:50 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/extconf.rb: remove checks for available functions.
+	* ext/openssl/missing.h: ditto.
+	  Thanks, Tim Mooney for reporting this!
+	  [Bug #5432] [ruby-core:40088]
+
+Sat Nov 26 10:22:28 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: add comment on where to find implementation
+	  of OpenSSL::SSL::SSLSocket#session.
+
+Sat Nov 26 05:00:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--no-undefined): RUBY_TRY_CFLAGS does nothing for
+	  linker flags.  use RUBY_TRY_LDFLAGS.
+
+Fri Nov 25 11:37:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (ioctl_narg_len, linux_iocparm_len): reinstantiate linux
+	  specific narg length calculation.
+	* test/ruby/test_io.rb (test_ioctl_linux2): add new test for old and
+	  unstructured ioctl.
+
+Fri Nov 25 10:39:14 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* Makefile.in (EXTLDFLAGS): export it.
+	* configure.in: add --no-undefined if --enable-shared is specified.
+	  Gentoo enabled this option long time. Also, export EXTLDFALGS.
+
+Fri Nov 25 08:48:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: turn on PIE if --enable-shared is not specified.
+
+Fri Nov 25 08:05:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: add -fstack-protector. It help to protect us from
+	  stack smashing attack.
+
+Fri Nov 25 08:03:28 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: add -D_FORTIFY_SOURCE=2. It provide some compile
+	  time and runtime check for security.
+
+Fri Nov 25 08:00:23 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/mkmf.rb: get rid of warnings of mkmf.rb if -Wmissing-declarations
+	  and/or -Wold-style-definition warnings if specified.
+	  Patch by Nikolai Weibull. Thank you! [Bug #5459] [ruby-core:40200]
+
+Fri Nov 25 07:46:09 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: add -Wall always.
+
+Thu Nov 24 20:02:40 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/test_engine.rb: use IO#reopen to restore stderr.
+
+Thu Nov 24 19:59:56 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_reopen): re-initialize buffering mode for stdout and
+	  stderr.
+
+Thu Nov 24 11:12:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_io_fsync,rb_io_fdatasync): release GVL during fsync().
+	  fsync() and fdatasync() may take a long time on slow disks and/or
+	  if there is much dirty data.
+	  Patch by Eric Wong. [Feature #5665] [ruby-core:41247]
+
+Thu Nov 24 10:05:02 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_engine.rb: Suppress output from 'openssl'
+	  engine's RC4 cipher.
+	  [Bug #5633] [ruby-core:41026]
+
+Thu Nov 24 08:05:02 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey_dsa.c: remove redundant colon from error
+	  message.
+	* ext/openssl/ossl_ssl.c: ditto.
+	* ext/openssl/ossl_pkey_rsa: ditto.
+	  patched by Eric Hodel [Bug #5604] [ruby-core:40896]
+
+Wed Nov 23 20:03:43 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (ioctl_narg_len): don't use _IOC_SIZE macro on Linux.
+	  On Linux some constants for ioctl(2) doesn't include the size of
+	  its return value and 16bit value; for example FIONREAD 0x541B.
+	  Moreover the manual, ioctl_list(2),  says "Note  that  the  size
+	  bits  are  very unreliable: in lots of cases they are wrong,
+	  either because of buggy macros using sizeof(sizeof(struct)),
+	  or because of legacy values."
+	  So we shouldn't use it.
+
+Tue Nov 22 18:07:32 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (_pioinfo): need to declare _pioinfo() before using
+	  _osfhnd and other macros which uses _pioinfo() internally.
+
+Tue Nov 22 17:49:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (_pioinfo): make an inline function.
+
+Tue Nov 22 11:26:08 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+Tue Nov 22 11:33:58 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (dupfd): argument of _osfhnd and so on should not
+	  have side effect.
+
+Tue Nov 22 11:26:08 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bignum.c (rb_big_divide): refix of r33536. Don't change behavior of Bignum#/.
+	  [ruby-core:40429] [Bug #5490]
+
+Tue Nov 22 10:46:57 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (ruby_float_step): improve floating point calculations.
+	  [ruby-core:35753] [Bug #4576]
+
+	* numeric.c (ruby_float_step): correct the error of floating point
+	  numbers on the excluding case.
+	  patched by Masahiro Tanaka [ruby-core:39608]
+
+	* numeric.c (ruby_float_step): use the end value when the current
+	  value is greater than or equal to the end value.
+	  patched by Akira Tanaka [ruby-core:39612]
+
+Tue Nov 22 06:59:21 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb (test_fcntl_dupfd): there is no known platform
+	  which don't have F_DUPFD.  [ruby-dev:44874]
+
+Tue Nov 22 04:46:22 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: remove autoload from psych
+	* ext/psych/lib/psych/json.rb: ditto
+
+Tue Nov 22 00:44:59 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb (test_fcntl_dupfd): the argument of F_DUPFD is
+	  minimum file descriptor.
+
+Tue Nov 22 00:25:17 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (linux_get_maxfd): get rid of a warning.
+
+Mon Nov 21 23:39:14 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (linux_get_maxfd): new function to find maximum fd on Linux.
+	  (rb_close_before_exec): use linux_get_maxfd.
+
+Mon Nov 21 06:16:24 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c (fiber_switch): ignore fiber context switch
+	  because destination fiber is same as current fiber.
+	  With out this, it may segv on FreeBSD 9.
+	  patched by Koichi Sasada.
+
+Sun Nov 20 23:22:42 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile, extmake): regenerate makefiles
+	  if globbed source file list is changed.
+
+	* lib/mkmf.rb (create_makefile): store ORIG_SRCS.
+
+Sun Nov 20 22:43:03 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/unicode.c (PROPERTY_NAME_MAX_SIZE): +1.
+	  reported by Ken Takata. [ruby-dev:44894][Bug #5652]
+
+Sun Nov 20 11:01:28 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/set.rb (SortedSet.setup): remove old_init after initialize
+	  method is redefined.  The remove before redefinition makes the
+	  warning prevention fragile.  [ruby-dev:44892]
+
+Sun Nov 20 04:01:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (enc/unicode/name2ctype.h): remove duplicated
+	  ifdefs.
+
+Sat Nov 19 19:31:47 2011  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (TIME_COPY_GMT): copy vtm.utc_offset and vtm.zone too.
+	  patch by Tomoyuki Chikanaga.
+	  [ruby-dev:44827] [Bug #5586]
+
+Sat Nov 19 16:36:57 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/http/test_http.rb: remove temporally files in ensure clause.
+
+Sat Nov 19 08:18:41 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/http/test_http.rb: remove temporally files.
+
+Fri Nov 18 17:18:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_raw, console_set_raw)
+	  (console_getch): optional parameters.  [EXPERIMENTAL]
+
+Fri Nov 18 16:12:11 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_cooked, console_set_cooked):
+	  new methods to reset cooked mode.  [EXPERIMENTAL]
+
+Fri Nov 18 13:20:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/unit/assertions.rb (MINI_DIR): quick dirty hack to get rid of
+	  warnings when using assert/assert_respond_to.
+
+Fri Nov 18 13:03:38 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_cloexec_open): set O_NOINHERIT instead of O_CLOEXEC if it is
+	  available (for Windows).
+
+	* win32/win32.c (fcntl): on F_DUPFD, determine the inheritance of the
+	  new handle by O_NOINHERIT flag of original fd.
+
+Fri Nov 18 08:00:41 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 2.8.1 (r6750)
+	* test/minitest/*: ditto
+	* configure.in: Improved gcc-llvm error message to help people migrate.
+
+Thu Nov 17 20:43:34 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: revert a part of the patch in [ruby-dev:41531].
+	  don't use db.h with other headers.  [ruby-dev:44884].
+
+Thu Nov 17 20:23:03 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* benchmark/bm_io_select[23].rb: use Process::RLIMIT_NOFILE only when
+	  it is defined.  if it is not defined, assume 64 as the max of fds.
+
+Thu Nov 17 10:36:46 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb (load_file): make sure opened yaml files are
+	  also closed. [ruby-core:41088]
+
+Wed Nov 16 18:13:52 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (LIBRUBY_A): check if generated linked library is
+	  valid for extconf.
+
+Wed Nov 16 13:51:40 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (rb_big2ulong): need to calc in unsigned long, because
+	  the range of VALUE is larger than it on LLP64 platform, such as Win64.
+	  this change fixes the failures of test/-ext-/num2int.
+
+Wed Nov 16 12:02:47 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/webrick/test_cgi.rb (TestWEBrickCGI#start_cgi_server): there are
+	  no guarantee of existence of RbConfig::CONFIG['LIBPATHENV'].
+	  it only exists in Unix-like environments.
+
+	* test/webrick/test_filehandler.rb
+	  (WEBrick::TestFileHandler#test_script_disclosure): ditto.
+
+Wed Nov 16 11:34:20 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (argf_next_argv): wrong timing of setting ecflags.
+	  fixed the failure of TestArgf#test_textmode introduced at r33662.
+
+Wed Nov 16 10:45:00 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/-test-/num2int/num2int.c: remove an unnecessary and wrong decl
+	  of rb_stdout.  it's declared in ruby.h correctly.
+
+Wed Nov 16 10:26:41 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (rb_big2ull): add a cast to get rid of a VC++ warning.
+
+Wed Nov 16 09:39:27 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/minitest/unit.rb (assert_raises): experimental fix to run
+	  correctly on chkbuild over 64bit linux. call exception_details only
+	  when the detail is really needed to avoid create needless inspect
+	  under ulimit-ed environment.
+
+Wed Nov 16 06:34:30 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_thread.rb (test_condvar_timed_wait): use
+	  assert_operator.
+
+Tue Nov 15 21:56:25 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_sleep.rb (test_sleep_5sec): 0.1sec tolerance is too
+	  small for busy environment.
+
+Tue Nov 15 20:08:55 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c, thread.c, ext/pty/pty.c, ext/fiddle/closure.c: use
+	  __linux__ macro for consistency.
+
+Tue Nov 15 14:45:15 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/ruby.h(NUM2LONG, NUM2INT, NUM2SHORT, NUM2LL,
+	  INT2NUM, UINT2NUM, LONG2NUM, ULONG2NUM, NUM2CHR): wrap by
+	  macros.
+
+Tue Nov 15 13:38:14 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/defines.h (FLUSH_REGISTER_WINDOWS): move sparc asm code
+	  to a separate file sparc.c for preventing inlining optimization.
+	  Patched by Jurij Smakov. [Bug #5244] [ruby-core:40685]
+	* sparc.c (rb_sparc_flush_register_windows): ditto.
+	* configure.in: ditto.
+
+Tue Nov 15 13:11:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/ruby.h: get rid of gcc specific rb_long2int(),
+	  NUM2LONG(), NUM2INT(), NUM2SHORT(), NUM2LL(), INT2NUM(),
+	  UINT2NUM(), LONG2NUM(), ULONG2NUM() and NUM2CHR()
+	  implementation. Because 1) They don't make any better code
+	  at all. 2) Inline function have a better debugger supoort.
+
+Tue Nov 15 09:58:25 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (TestIO#test_fcntl_dupfd): fix OpenBSD test
+	  failure. [ruby-dev:44872]
+
+Tue Nov 15 09:50:21 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (print_indent_tree): fix double printing of ENCLOSE_OPTION
+	  children bug. patched by Suraj Kurapati. [ruby-core:40964]
+
+Tue Nov 15 01:53:48 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/ruby/test_io.rb (test_fcntl_dupfd): fix test error on
+	  SnowLeopard. Pointed out by CHIKANAGA Tomoyuki. [ruby-dev:44866]
+
+Mon Nov 14 22:06:02 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_pkey.c (ossl_pkey_new_from_file): set close-on-exec
+	  flag.
+
+	* ext/openssl/ossl_x509cert.c (rb_fd_fix_cloexec): ditto.
+
+Mon Nov 14 14:54:17 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bignum.c (rb_big2ull): fix 32bit platform breakage. we must
+	  not assume sizeof(VALUE) == sizeof(LONG_LONG).
+	* test/-ext-/num2int/test_num2int.rb (class TestNum2int):
+	  fix false assumption on 32bit platform.
+
+Mon Nov 14 14:52:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* numeric.c (rb_fix2ushort): fix typo. use num rb_num2ushort()
+	  instead of num2uint().
+
+Sun Nov 13 10:31:03 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/ruby.h: add #ifdef comment.
+
+Sun Nov 13 10:28:18 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/ruby.h: add NUM2SHORT(), NUM2USHORT() macros.
+	* numeric.c: ditto.
+
+	* test/-ext-/num2int/test_num2int.rb: add testcases for NUM2SHORT().
+	* ext/-test-/num2int/num2int.c: ditto.
+
+Sun Nov 13 10:23:48 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bignum.c (rb_big2ull): fix off-by-twice bug of NUM2ULL.
+	* test/-ext-/num2int/test_num2int.rb (class TestNum2int):
+	  fix a testcase too.
+
+Sun Nov 13 10:22:44 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/-ext-/num2int/test_num2int.rb (class TestNum2int):
+	  add FIXNUM tests.
+
+Sun Nov 13 09:57:29 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* numeric.c (check_uint): fix off-by-one bug of NUM2UINT.
+	* bignum.c (rb_big2ulong): fix off-by-one bug of NUM2ULONG.
+
+	* test/-ext-/num2int/test_num2int.rb: add a testcase for NUM2INT()
+	  NUM2UINT(), NUM2LONG(), NUM2ULONG(), NUM2LL and NUM2ULL().
+	* ext/-test-/num2int/depend: ditto.
+	* ext/-test-/num2int/extconf.rb: ditto.
+	* ext/-test-/num2int/num2int.c: ditto.
+
+Sun Nov 13 23:47:29 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dbm/extconf.rb: use convertible_int.
+
+Sun Nov 13 23:45:57 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (checking_for): should not modify the result.
+
+	* lib/mkmf.rb (have_struct_member): accept compiler options.
+
+	* lib/mkmf.rb (convertible_int): add restricted support of struct
+	  member, and TYPEOF_ macro.
+
+Sun Nov 13 23:21:24 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/gdbm/gdbm.c (fgdbm_reorganize): set close-on-exec flag after
+	  gdbm_reorganize().  gdbm_reorganize() opens a new database internally.
+
+Sun Nov 13 19:57:18 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: rollback for each headers for each libraries.
+
+Sun Nov 13 16:24:48 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: treat libc as a choice for a library which
+	  provide ndbm API.
+
+Sun Nov 13 15:40:43 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: duplicate $libs and $defs when save them.
+
+Sun Nov 13 12:43:48 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: rollback $libs and $defs when db detection is
+	  failed.  It fixes -lgdbm -lqdbm when the system has qdbm and gdbm
+	  without gdbm_compat.
+
+Sat Nov 12 21:14:51 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/webrick/test_cgi.rb (class TestWEBrickCGI): respect
+	  RbConfig::CONFIG["LIBPATHENV"]. [Bug #5135] [ruby-core:38653]
+	* test/webrick/test_filehandler.rb (class WEBrick): ditto.
+
+Sat Nov 12 20:57:29 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (test_fcntl_dupfd): skip if Fcntl::DUPFD
+	  is not defined. Pointed out by CHIKANAGA Tomoyuki. Thanks.
+
+Sat Nov 12 17:26:10 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (do_ioctl, ioctl_narg_len, setup_narg, rb_ioctl): use
+	  ioctl_req_t.
+
+Sat Nov 12 17:01:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dbm/extconf.rb (headers.db_check): reduce duplicated code.
+
+Sat Nov 12 15:59:42 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: dbm_clearerr should be available in all ndbm
+	  implementation.  If it is not available, it is caused by
+	  header/library mismatch such that Berkeley DB header & gdbm library.
+
+	* ext/dbm/dbm.c (fdbm_store): use dbm_clearerr() unconditionally.
+	  gdbm 1.9 provides it as a real function instead of a empty macro.
+
+Sat Nov 12 13:35:33 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bootstraptest/runner.rb: don't suppress SIGINT.
+	  [Feature #5612] [ruby-dev:44856]
+
+Sat Nov 12 11:20:36 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (fcntl_narg_len): introduce narg calculation for fcntl instead
+	  of hard coded 256.
+	* io.c (setup_narg): ditto.
+
+Sat Nov 12 11:19:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (test_fcntl_dupfd): add another fcntl test.
+
+Sat Nov 12 11:18:17 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_io.rb (test_fcntl_lock_freebsd): add a testcase
+	  of fcntl lock for freebsd.
+
+Sat Nov 12 11:16:32 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (ioctl_narg_len): Linux doesn't have IOCPARM_LEN macro, but
+	  has _IOC_SIZE. support it.
+
+Sat Nov 12 11:13:18 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_ioctl): don't expose our sanity check value to ruby script.
+	  It may change string value meaning if the value is string.
+	  (e.g. MacOS X has F_GETPATH ioctl)
+	* io.c (rb_fcntl): ditto.
+
+Sat Nov 12 11:06:02 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (ioctl_req_t): Type of req argument of ioctl() depend on platform.
+	  Moreover almost all linux ioctl can't be represented by 32bit integer
+	  (i.e. MSB is 1). We need wrap ioctl argument type.
+	  [Bug #5429] [ruby-dev:44589]
+	* io.c (struct ioctl_arg): ditto.
+	* io.c (rb_ioctl): ditto.
+	* test/ruby/test_io.rb (test_ioctl_linux): add a testcase for ioctl
+
+Sat Nov 12 11:00:42 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (struct io_cntl_arg): remove io_p member.
+	* io.c (nogvl_fcntl, do_fcntl, rb_fcntl): separated from ioctl functions.
+	* io.c (nogvl_io_cntl): remove fcntl depended logic.
+	* io.c (io_cntl): ditto.
+	* io.c (rb_io_ctl): ditto.
+	* io.c (rb_io_ioctl): ditto.
+
+Sat Nov 12 10:59:49 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (setup_narg): fix off by one bug.
+
+Sat Nov 12 10:56:43 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (+setup_narg): factor out length calculation logic.
+	* io.c (rb_io_ctl): ditto.
+
+Sat Nov 12 10:52:17 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (+ioctl_narg_len) new helper function.
+	* io.c (rb_io_ctl): don't use ioctl specific length check
+	  if caller is fcntl.
+
+Fri Nov 11 23:00:46 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: db_prefix is not required now.
+
+Fri Nov 11 21:13:30 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/gdbm/gdbm.c (fgdbm_initialize): use GDBM_CLOEXEC if available.
+
+Fri Nov 11 21:00:05 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: fix dbm_pagfno and dbm_dirfno detection with
+	  Berkeley DB.  Macro definitions needs arguments to detect correctly.
+	  SIZEOF_DSIZE needs -DDB_DBM_HSEARCH because db.h defines datum type
+	  only if DB_DBM_HSEARCH is defined.
+
+Fri Nov 11 18:41:57 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_seteuid): separate an internal wrapper function
+	  from the method implementation.
+
+Fri Nov 11 17:21:15 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_library, find_library, have_func): allow
+	  arguments of function to be checked.
+
+Fri Nov 11 17:09:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (HAVE_BACKTRACE): fallback to 0.
+
+	* vm_dump.c (rb_vm_bugreport): show "Other runtime information"
+	  header only when available.
+
+	* vm_dump.c (rb_vm_bugreport): get rid of modifying the content of
+	  VM directly.
+
+	* vm_dump.c (rb_vm_bugreport): check if vm is non-null.
+	  Pointed out by Ikegami Daisuke <ikegami.da@gmail.com>.
+	  Thank you.
+
+Fri Nov 11 12:36:37 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (pipe_open): Remove fflush(stdin). it's no effect.
+	  Pointed out by Ikegami Daisuke <ikegami.da@gmail.com>.
+	  Thank you.
+
+Fri Nov 11 07:33:30 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb (Net::HTTP::SSL_ATTRIBUTES):  Use symbol keys instead
+	  of string keys to avoid duplicating parameters in
+	  OpenSSL::SSL:SSLContext#set_params.
+
+Thu Nov 10 15:02:37 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_library, have_func, have_var, have_header):
+	  add compiler option parameter.
+
+Thu Nov 10 07:45:16 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/lib/openssl/ssl.rb (class OpenSSL::SSL::SSLContext):
+	  Document #set_params.
+
+Wed Nov  9 11:36:53 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (gvl_yield): don't prevent concurrent sched_yield().
+	  [Bug #5130] [ruby-core:38647]
+
+Wed Nov  9 23:20:22 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_update_max_fd): fstat(2) can fail with other than
+	  EBADF.  [ruby-dev:44837] [Bug #5593].  Cf.
+	  http://pubs.opengroup.org/onlinepubs/9699919799/functions/fstat.html
+
+	* io.c (rb_sysopen): max fd is updated in rb_sysopen_internal()
+	  already.
+
+Wed Nov  9 22:13:38 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file.rb (TestFile#test_utime_with_minus_time_segv):
+	  fixed previous commit.
+
+Wed Nov  9 19:53:45 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file.rb (TestFile#test_utime_with_minus_time_segv):
+	  add test for r33685.
+
+Wed Nov  9 19:00:44 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_fiber.rb: add tests for r33684 (Fiber#resume).
+
+Wed Nov  9 16:40:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (unixtime_to_filetime): should check the return value
+	  of localtime().  reported by snowjail at gmail.com.
+	  [ruby-dev:44838] [Bug #5596]
+
+Thu Nov  9 13:17:25 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_m_transfer, rb_fiber_resume): prohibit using
+	  "resume" after "transfer" method are used.  You should not mix
+	  "resume" fiber and "transfer" fiber.
+	  [Bug #5526]
+
+	* NEWS: add information about this change.
+
+Wed Nov  9 11:40:37 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/Doxyfile.tmpl (INCLUDE_PATH): add srcdir and include.
+	  [ruby-core:40843] [Bug #5597]
+
+Wed Nov  9 11:02:54 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (do_select): fix cast, tv_sec is time_t.
+
+Wed Nov  9 10:32:20 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: should not use test -e for portability.
+	  [ruby-core:40841] [Bug #5594]
+
+Wed Nov  9 04:52:16 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/tree_builder.rb: dump complex numbers,
+	  rationals, etc with reference ids.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ditto
+	* ext/psych/lib/psych/visitors/to_ruby.rb: loading complex numbers,
+	  rationals, etc with reference ids.
+	* test/psych/test_object_references.rb: corresponding tests
+
+Tue Nov  8 23:34:37 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dbm/dbm.c (fdbm_fetch, fdbm_key, fdbm_delete, fdbm_store)
+	  (fdbm_has_key, fdbm_has_value): get rid of overflow.
+
+	* ext/gdbm/gdbm.c (rb_gdbm_fetch2, rb_gdbm_nextkey)
+	  (rb_gdbm_delete, fgdbm_store, fgdbm_has_key): ditto.
+
+	* ext/dbm/dbm.c (fdbm_delete_if): hide intermediate objects.
+
+	* ext/gdbm/gdbm.c (fgdbm_delete_if): ditto.
+
+	* ext/dbm/extconf.rb: check size of datum.dsize to get rid of
+	  overflow.
+
+Tue Nov  8 23:30:21 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (PATH_MAX): define if not defined. [ruby-core:40840]
+
+Tue Nov  8 23:26:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c (rb_thread_critical): fix type.
+
+	* ext/tk/tcltklib.c (eventloop_sleep, lib_eventloop_core): int is
+	  enough for micro seconds.  may need to check overflow in the
+	  setter though.
+
+	* ext/tk/tcltklib.c (RSTRING_LENINT): check overflow if necessary.
+
+	* ext/tk/tcltklib.c (RbTk_ALLOC_N): wrapper for ckalloc() which
+	  takes an int.
+
+	* ext/tk/tcltklib.c (ip_ruby_cmd_receiver_get, tcltklib_compile_info):
+	  get rid overflow.
+
+	* ext/tk/tcltklib.c (tcltklib_compile_info): constified.
+
+Tue Nov  8 20:50:45 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/dbm/test_dbm.rb: split tests for read only database.
+
+	* test/gdbm/test_gdbm.rb: ditto.
+
+Tue Nov  8 18:59:07 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (MasterDevice): define only when used.
+	  (SlaveDevice): ditto.
+	  (deviceNo): ditto.
+
+Tue Nov  8 17:59:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_long2int): define as a macro always, so
+	  that cpp conditionals can tell if it is provided.
+
+Tue Nov  8 17:30:50 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (cpp_command): remove multiple -arch flags since cpp
+	  cannot work.
+
+Tue Nov  8 14:50:55 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_fwrite): call rb_w32_write_console() only if FMODE_TTY is
+	  set.  this is the one of the reason of IO writing slowness of Windows
+	  in 1.9.3 or later.
+
+Tue Nov  8 11:01:04 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/pty/pty.c (get_device_once): FreeBSD 8 supported O_CLOEXEC flag
+	  for posix_openpt, but FreeBSD 9's posix_openpt doesn't support
+	  O_CLOEXEC and fails if specified.
+
+Tue Nov  8 02:36:45 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/encoding.h (ECONV_NEWLINE_DECORATOR_READ_MASK,
+	  ECONV_NEWLINE_DECORATOR_WRITE_MASK): new macro.
+
+	* io.c (rb_io_extract_modeenc, pipe_open, prep_stdio, argf_next_argv):
+	  set TEXTMODE_NEWLINE_DECORATOR_ON_WRITE for textmode on creating IO
+	  if the flag is available.
+
+	* io.c (make_writeconv): drop decorators for reading.
+
+	* io.c (make_readconv): drop decorators for writing.
+
+	* io.c (do_writeconv): existing writeconv is not the condition to raise
+	  ArgumentError.  should check textmode or not.
+
+	* test/ruby/test_io_m17n.rb
+	  (TestIO_M17N#test_{cr,lf,crlf}_decorator_on_stdout): test above
+	  changes.
+
+Mon Nov  7 22:03:47 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/gdbm/gdbm.c (fgdbm_initialize): set close-on-exec flag.
+
+Mon Nov  7 20:31:52 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: make sure strings that look
+	  like base 60 numbers are serialized as quoted strings.
+	* test/psych/test_string.rb: test for change.
+
+Mon Nov  7 20:26:37 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* test/psych/test_yamlstore.rb: make test case inherit from MiniTest,
+	  load psych/helper so that psych is loaded.
+
+Mon Nov  7 20:18:29 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* test/psych/test_yamldbm.rb: Test case should inherit from MiniTest,
+	  load psych/helper so that psych and friends are loaded.
+
+Mon Nov  7 20:15:44 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/extconf.rb: check dbm_pagfno() and dbm_dirfno().
+
+	* ext/dbm/dbm.c: use above to set close-on-exec flag.
+
+Mon Nov  7 20:05:16 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (io_fflush): remove fsync().
+
+	* io.c (rb_io_flush, rb_io_rewind): fsync() here.
+
+	these changes reduces fsync() calls to improve performance.
+	first reported at [ruby-list:48515] by ak7 at mail.goo.ne.jp .
+	[Bug #5585]
+
+Mon Nov  7 19:43:10 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_close_before_exec): use F_MAXFD if available.
+	  F_MAXFD is available on NetBSD since NetBSD 2.0.
+
+Mon Nov  7 19:25:16 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io_m17n.rb
+	  (TestIO_M17N#test_default_stdout_stderr_mode): new test for
+	  r33627-33629.  see [backport #5565]
+
+Mon Nov  7 01:14:22 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/debug.rb: add help for 'pp' and 'r[estart]'. patch
+	  from Sho Hashimoto. [Bug #5093] [ruby-dev:44222]
+
+Sun Nov  6 14:49:58 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (rsock_recvmsg): declared.
+
+	* ext/socket/ancdata.c (rsock_recvmsg): extracted from
+	  nogvl_recvmsg_func.
+	  (nogvl_recvmsg_func): use rsock_recvmsg.
+
+	* ext/socket/unixsocket.c (recvmsg_blocking): use rsock_recvmsg.
+
+Sun Nov  6 03:22:36 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_engine.rb: add test for engine cipher. RC4 is used
+	  because AES is not supported by the "openssl" engine currently.
+
+Sun Nov  6 00:11:52 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/test/unit.rb (Test::Unit::Options#non_options): options[:ruby]
+	  should be an array.  This fixes
+	  "./ruby test/runner.rb test/testunit/test_parallel.rb"
+	  [ruby-dev:44782]
+
+Sat Nov  5 20:30:30 2011  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* insns.def: Some fixes and tweaks to English explanations
+
+Sat Nov  5 19:11:50 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_cloexec_fcntl_dupfd): don't clear try_dupfd_cloexec if
+	  fcntl(F_DUPFD) failed as fcntl(F_DUPFD_CLOEXEC).
+
+Sat Nov  5 18:05:12 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_socketpair0): refactored.
+
+Sat Nov  5 17:55:52 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (rsock_socket0): don't clear try_sock_cloexec if
+	  SOCK_CLOEXEC is not a reason for EINVAL.
+
+Sat Nov  5 16:27:52 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/pathname/lib/pathname.rb, ext/tk/lib/multi-tk.rb,
+	  ext/tk/sample/demos-en/widget, lib/benchmark.rb, lib/irb/cmd/fork.rb,
+	  lib/mkmf.rb, lib/net/ftp.rb, lib/net/smtp.rb, lib/open3.rb,
+	  lib/pstore.rb, lib/rexml/element.rb, lib/rexml/light/node.rb,
+	  lib/rinda/tuplespace.rb, lib/rss/maker/base.rb,
+	  lib/rss/maker/entry.rb, lib/scanf.rb, lib/set.rb, lib/shell.rb,
+	  lib/shell/command-processor.rb, lib/shell/process-controller.rb,
+	  lib/shell/system-command.rb, lib/uri/common.rb: remove unused block
+	  arguments to avoid creating Proc objects.
+
+Sat Nov  5 15:45:04 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (rsock_socket0): extract single socket() call with
+	  CLOEXEC handling from rsock_socket.
+
+Sat Nov  5 13:49:40 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/pathname.rb (Pathname#find): return an enumerator if
+	  no block is given.
+
+	* test/pathname/test_pathname.rb: add tests for above.
+
+	[ruby-dev:44797] [Feature #5572]
+
+Sat Nov  5 11:18:12 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_socketpair0): don't clear
+	  try_sock_cloexec if SOCK_CLOEXEC is not a reason for EINVAL.
+
+Fri Nov  4 14:08:19 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey_rsa.c (rsa_generate): [SECURITY] Set RSA
+	  exponent value correctly.  Awful bug.  This bug caused exponent of
+	  generated key to be always '1'.  By default, and regardless of e
+	  given as a parameter.
+
+	  !!! Keys generated by this code (trunk after 2011-09-01) must be
+	  re-generated !!! (ruby_1_9_3 is safe)
+
+	* test/openssl/test_pkey_rsa.rb: Add tests for default exponent and
+	  specifying exponent by a parameter.
+
+Fri Nov  4 01:31:25 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_engine.rb: add first tests for builtin "openssl"
+	  engine.
+
+Fri Nov  4 08:41:26 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/extconf.rb:
+	* ext/openssl/ossl_engine.c: add some missing OpenSSL engines.
+	  Thanks, Yui Naruse, for providing the patch!
+	  [Bug #5548] [ruby-core:40670]
+
+Fri Nov  4 04:54:10 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/configure.bat: disable delayed expansion of enironment variable.
+	  [Bug #5517] [ruby-core:40531]
+
+Fri Nov  4 03:45:22 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (make_writeconv): fixed typo of previous commit.
+
+Fri Nov  4 01:56:30 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (make_writeconv): unversal_newline converter is for reading.
+	  so, if the io is text mode and has ECONV_UNIVERSAL_NEWLINE_DECORATOR
+	  flag, use crlf_newline converter for writing.
+	  this change fixes the problem about the luck of CR up Kernel.p and
+	  Kernel.puts to stdout/stderr on Windows.
+
+Fri Nov  4 01:04:48 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (Init_readline): like r18313, libedit's
+	  replace_history_entry may use offset instead of which.
+	  so introduce history_replace_offset_func and initialize it.
+
+	* ext/readline/readline.c (hist_set): use history_replace_offset_func.
+
+Fri Nov  4 00:53:35 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (Init_readline): fix wrong condition.
+
+Thu Nov  3 23:53:04 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* encoding.c (rb_locale_charmap): ignore calling nl_langinfo_codeset()
+	  on Windows except cygwin. [experimental]
+
+Thu Nov  3 22:45:09 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_socketpair0): extracted from
+	  rsock_socketpair to set close-on-exec flag for each socketpair()
+	  call.
+
+Thu Nov  3 22:12:41 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/socket/init.c (rsock_socket): set close-on-exec flag when
+	  SOCK_CLOEXEC is not available.
+
+Thu Nov  3 08:36:00 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_engine.rb: call Engine::cleanup on exit.
+	  Patch provided by Yui Naruse, thanks!
+	  [Bug #5547] [ruby-core:40669]
+
+Wed Nov  2 21:36:00 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (nucomp_rationalize): fix function.  [ruby-core:40667]
+	  [Bug #5546]
+
+Wed Nov  2 08:16:45 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/utils.rb: fix fcntl call.
+
+	* lib/drb/unix.rb: ditto.
+
+Wed Nov  2 00:43:59 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/psych/test_yamldbm.rb: avoid platform dependency.
+	  patch by Naohisa Goto. [ruby-dev:44763] [Bug #5535]
+	* test/syck/test_yamldbm.rb: ditto.
+
+Wed Nov  2 00:14:15 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_marshal.rb: renamed methods duplicated with those
+	  of marshaltestlib.rb.
+
+Tue Nov  1 22:08:27 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: reject llvm-gcc.
+
+Tue Nov  1 21:39:00 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_cloexec_pipe): remove workaround of r33587.
+	  The bug of NetBSD is fixed on Mon Oct 31 21:31:29 UTC 2011.
+	  http://gnats.netbsd.org/cgi-bin/query-pr-single.pl?number=45545
+
+Tue Nov  1 19:49:08 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_reopen): call rb_fd_fix_cloexec instead of
+	  rb_maygvl_fd_fix_cloexec.
+
+Tue Nov  1 19:00:30 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_reopen): call rb_maygvl_fd_fix_cloexec after freopen().
+
+Tue Nov  1 17:17:26 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (file_expand_path): reset coderange after expanding path.
+
+Tue Nov  1 14:55:29 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (nogvl_io_cntl): rb_cloexec_fcntl_dupfd's 2nd argument is int.
+
+	* process.c (move_fds_to_avoid_crash): ditto.
+
+Tue Nov  1 13:14:33 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): support 'll' prefix.
+
+	* vsnprintf.c (__sfeof): rename to avoid the collision with NetBSD's
+	  one.
+
+	* vsnprintf.c (__sferror): ditto.
+
+	* vsnprintf.c (__sclearerr): ditto.
+
+	* vsnprintf.c (__sfileno): ditto.
+
+Tue Nov  1 12:36:16 2011  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_maygvl_fd_fix_cloexec): change the visibility for
+	  ext/socket.
+
+Tue Nov  1 12:00:53 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_maygvl_fd_fix_cloexec): renamed from fd_set_cloexec.
+
+	* internal.h (rb_maygvl_fd_fix_cloexec): declared.
+
+	* ext/socket/init.c (cloexec_accept): use rb_maygvl_fd_fix_cloexec.
+	  (rsock_s_accept_nonblock): use rb_update_max_fd.
+	  (rsock_s_accept): use rb_update_max_fd.
+
+Tue Nov  1 08:24:40 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (cloexec_accept): new function to use accept4 if
+	  available.
+	  (rsock_s_accept_nonblock): use cloexec_accept.
+	  (accept_blocking): ditto.
+
+	* ext/socket/extconf.rb: check accept4.
+
+Tue Nov  1 07:31:55 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (nogvl_recvmsg_func): use MSG_CMSG_CLOEXEC if
+	  available.
+
+	* ext/socket/unixsocket.c (recvmsg_blocking): ditto.
+
+Tue Nov  1 05:59:41 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_socketpair): use SOCK_CLOEXEC if
+	  available.
+
+Tue Nov  1 02:56:17 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (load_file_internal): convert the encoding of load path if
+	  needed by platform.  calling open() was replaced by rb_cloexec_open()
+	  at r33549, but the function expected UTF-8 pathname on Windows.
+	  (open() expected "locale" pathname.)
+	  reported by taco via IRC.
+
+	* ruby.c (load_file): change the type of the 2nd parameter to pass its
+	  encoding to load_file_internal().
+
+	* ruby.c (process_options, rb_load_file): follow above change.
+	  NOTE: we should pass encoding information to rb_load_file().
+
+Mon Oct 31 23:49:38 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_socketpair): extracted from
+	  rsock_sock_s_socketpair.
+
+Mon Oct 31 23:31:53 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (rsock_socket): use SOCK_CLOEXEC if available.
+
+Mon Oct 31 21:47:44 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_cloexec_pipe): NetBSD 6.0 will support pipe2(2),
+	  but its return value is -1 or larger than 0.
+
+Mon Oct 31 22:04:54 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dbm/dbm.c (fdbm_initialize): use O_CLOEXEC if available.
+
+Mon Oct 31 21:47:48 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_fd_fix_cloexec): renamed from
+	  rb_fd_set_cloexec.
+
+	* io.c: follow the above renaming.
+
+	* ext/pty/pty.c: ditto.
+
+	* ext/socket/init.c: ditto.
+
+	* ext/socket/socket.c: ditto.
+
+	* ext/socket/ancdata.c: ditto.
+
+	* ext/socket/unixsocket.c: ditto.
+
+Mon Oct 31 21:02:43 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS): retry IO.select for premature wakeup.
+
+Mon Oct 31 20:14:22 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (fd_set_cloexec): clear CLOEXEC flag for standard file
+	  descriptors.
+	  (rb_cloexec_dup): use rb_cloexec_fcntl_dupfd.
+	  (rb_cloexec_fcntl_dupfd): use F_DUPFD_CLOEXEC if available.
+
+Mon Oct 31 19:14:11 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/resolv/test_dns.rb: don't check maximum slept time.
+	  ruby doesn't guarantee the maximum time because it is not a
+	  realtime application.
+
+Mon Oct 31 13:10:06 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (setfl): extract from fcntl().
+
+	* win32/win32.c (dupfd): new function to support F_DUPFD. based on a
+	  patch written by akr.
+
+	* win32/win32.c (fcntl): use above functions.
+
+	* include/ruby/win32.h (F_DUPFD): define. [experimental]
+
+	* include/ruby/win32.h (F_SETFL): change the value to correspond with
+	  other platforms.
+
+Mon Oct 31 12:37:50 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (get_device_once): use O_CLOEXEC for posix_openpt if
+	  available.
+
+Mon Oct 31 12:05:24 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_cloexec_dup2): check oldfd == newfd at first.
+	  pointed by KOSAKI Motohiro.  [ruby-dev:44713]
+
+Mon Oct 31 10:50:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_cloexec_fcntl_dupfd): this function needs F_DUPFD.
+
+	* io.c (nogvl_io_cntl): use rb_cloexec_fcntl_dupfd() only if the
+	  platform has F_DUPFD.
+
+Mon Oct 31 00:50:00 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* configure.in: check -fno-omit-frame-pointer acceptance and usage
+	  under MinGW. [ruby-core:39957] [Bug #5407]
+
+Mon Oct 31 00:16:11 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_cloexec_fcntl_dupfd): declared.
+
+	* io.c (rb_cloexec_fcntl_dupfd): new function.
+	  (nogvl_io_cntl): use rb_cloexec_fcntl_dupfd.
+
+	* process.c (move_fds_to_avoid_crash): use rb_cloexec_fcntl_dupfd.
+
+Sun Oct 30 22:46:46 2011  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: check pipe2.
+
+	* io.c (rb_cloexec_pipe): use pipe2 if available.
+
+Sun Oct 30 22:32:44 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c (fill_standard_fds): use fstat() instead of fcntl(F_GETFD)
+	  for MinGW.  reported by Luis Lavena.  [ruby-core:40526] [Bug #5516]
+
+Sun Oct 30 21:12:47 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_cloexec_pipe): declared.
+
+	* io.c (rb_cloexec_pipe): new function.
+	  (rb_pipe): use rb_cloexec_pipe.
+
+	* thread_pthread.c (rb_thread_create_timer_thread): use
+	  rb_cloexec_pipe.
+
+Sun Oct 30 20:06:07 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_cloexec_dup): refine control flow.
+	  (rb_cloexec_dup2): ditto.
+
+Sun Oct 30 18:45:50 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c (fill_standard_fds): new function to open closed standard
+	  file descriptors.
+	  (ruby_sysinit): call fill_standard_fds.
+
+Sun Oct 30 10:50:36 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (install_recursive, bin-comm): split mere
+	  string not path name.  [ruby-core:40462] [Bug #5492]
+
+Sun Oct 30 10:47:20 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_cloexec_dup, rb_cloexec_dup2): CLOEXEC has been set if
+	  dup3 succeeded.
+
+Sun Oct 30 09:58:48 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_cloexec_dup): don't allocate standard file descriptors.
+
+Sun Oct 30 08:29:51 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_cloexec_dup2): don't set CLOEXEC for standard file
+	  descriptors.
+
+Sun Oct 30 07:47:10 2011  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: check dup3.
+
+	* io.c (rb_cloexec_dup2): use dup3 if available.
+
+Sat Oct 29 22:06:37 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_cloexec_dup2): declared.
+
+	* io.c (rb_cloexec_dup2): new function.
+	  (io_reopen): use rb_cloexec_dup2.
+
+Sat Oct 20 21:08:18 2011  Tajima Akio <artonx@yahoo.co.jp>
+
+	* win32/Makefile.sub (CONFIG_H): have stdint.h if VC2010.
+	  [Bug #5243]
+
+Sat Oct 29 20:59:08 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_cloexec_dup): use F_DUPFD_CLOEXEC if available.
+
+Sat Oct 29 20:00:26 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_cloexec_dup): declared.
+
+	* io.c (rb_cloexec_dup): new function.
+	  (ruby_dup): use rb_cloexec_dup.
+
+	* ext/pty/pty.c (pty_getpty): use rb_cloexec_dup.
+
+	* ext/openssl/ossl_bio.c (ossl_obj2bio): ditto.
+
+Sat Oct 29 16:11:34 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/sdbm/_sdbm.c (sdbm_prep): use O_CLOEXEC if available.
+
+Sat Oct 29 14:26:56 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_cloexec_open): use O_CLOEXEC if available.
+
+Sat Oct 29 12:57:15 2011  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (ruby_setsid): use rb_cloexec_open.
+	  (rb_daemon): ditto.
+
+	* ruby.c (load_file_internal): ditto.
+
+	* file.c (rb_file_s_truncate): ditto.
+	  (file_load_ok): ditto.
+
+	* random.c (fill_random_seed): ditto.
+
+	* ext/pty/pty.c (chfunc): ditto.
+	  (get_device_once): ditto.
+
+	* ext/io/console/console.c (console_dev): ditto.
+
+Sat Oct 29 10:40:19 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_cloexec_open): declared.
+
+	* io.c (fd_set_cloexec): extracted from rb_fd_set_cloexec.
+	  (rb_cloexec_open): new function.
+	  (sysopen_func): use rb_cloexec_open.
+	  (rb_sysopen_internal): use rb_update_max_fd instead of
+	  rb_fd_set_cloexec.
+
+Sat Oct 29 09:05:07 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.h: no Structured Exception Handling like macros.
+	  [ruby-core:40432] [Bug #5491]
+
+Fri Oct 28 22:05:34 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/sdbm/_sdbm.c: RCS $Id$ removed.
+
+Thu Oct 27 18:58:00 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (parser_nextc): set encoding for the buffer of ripper.
+
+Fri Oct 28 06:06:08 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/sdbm/_sdbm.c (sdbm_prep): set FD_CLOEXEC flags for file
+	  descriptors.
+	  (fd_set_cloexec): new function.
+
+Fri Oct 28 03:01:27 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* vm_insnhelper.c (vm_call_cfunc): adding back useless hack.  For some
+	  reason, this fixes CFP errors on OS X 10.7.
+
+Fri Oct 28 00:09:31 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/sdbm/_sdbm.c (sdbm_prep): refactored for less nesting.
+
+Thu Oct 27 18:28:18 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_DEFINE_IF): revert r33534 partially to get
+	  rid of AS_ECHO which is not available in autoconf 2.61.
+	  [ruby-dev:44702]
+
+Thu Oct 27 16:10:46 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_divide): raise ZeroDivisionError if divisor is
+	  zero, as well as Fixnum.  [ruby-core:40429] [Bug #5490]
+
+Thu Oct 27 14:56:22 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_FUNC_ATTRIBUTE): unset temporary variable.
+
+	* configure.in (RUBY_STACK_GROW_DIRECTION): substitute CPU name as
+	  shell variable name.  based on the patch by The Written Word Inc. at
+	  [ruby-core:40421].  [Bug #5488]
+
+Thu Oct 27 09:57:56 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (SIZE_MAX): define SIZE_MAX if not defined.
+	  patched by The Written Word Inc. [ruby-core:40422] [Bug #5489]
+
+Thu Oct 27 08:47:38 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/psych/parser.c: remove unused variable.
+
+Thu Oct 27 08:38:41 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/extconf.rb: add -Wall flag by default when compiler is
+	  GCC.
+
+Wed Oct 26 15:24:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_join): honor input encodings than ASCII-8BIT.
+	  [ruby-core:40338] [Bug #5483]
+
+Tue Oct 25 21:52:31 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h: use "__sparc" instead of "sparc" and
+	  "__sparc__".
+
+	* dln.c: ditto.
+
+	  [ruby-dev:44694]
+
+Tue Oct 25 06:34:39 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* re.c (match_aref):  Use <code> around indexing examples to prevent
+	  hyperlinks.  [ruby-talk:389396]
+
+Mon Oct 24 23:55:31 2011  Tanaka Akira  <akr@fsij.org>
+
+	* complex.c: use "__sun" instead of "__sun__" to detect SunOS.
+
+	* math.c: ditto.
+
+	* hash.c: ditto.
+
+	* atomic.h: ditto.
+
+	* ext/io/wait/wait.c: ditto.
+
+	[ruby-dev:44693]
+
+Mon Oct 24 22:45:37 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: use "__sun" instead of "sun" to detect SunOS.
+
+	* dln.c: ditto.
+
+	* cont.c: ditto.
+
+	* ext/sdbm/_sdbm.c: ditto.
+
+	[ruby-dev:44693]
+
+Mon Oct 24 22:38:08 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pty/pty.c (get_device_once): delay rb_fd_set_cloexec() until
+	  grantpt() on Solaris.  grantpt() doesn't work with CLOEXEC on
+	  Solaris 10.
+	  reported by Naohisa GOTO.  [ruby-dev:44688] [Bug #5475]
+
+Mon Oct 24 08:18:14 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (copy_stream_fallback_body): check nil for EOF of read method.
+	  patch by Eric Wong.  [ruby-core:39134] [Bug #5237]
+
+Sun Oct 23 18:21:23 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/tk/MANUAL_tcltklib.eng: fix typo.
+
+Sun Oct 23 18:03:31 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (rb_infinity, rb_nan): aggregated member initializers
+	  need braces.
+
+Sun Oct 23 16:43:43 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/io/wait/wait.c: ioctl(2) is declared in unistd.h on Solaris.
+
+Sun Oct 23 16:33:35 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/tk/MANUAL_tcltklib.eng: fix typo. reported by Mimura-san.
+	  [ruby-dev:44683] [Bug #5471]
+
+Sun Oct 23 08:01:29 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_fd_set_cloexec): set close-on-exec flag only if F_GETFD is
+	  defined.  reported by Luis Lavena.  [ruby-core:40281] [Bug #5470]
+
+Sat Oct 22 19:48:50 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/test_ssl.rb (test_multibyte_read_write): start server
+	  for each length to avoid race condition.
+
+Sat Oct 22 18:49:24 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_fd_set_cloexec): declared.
+
+	* io.c (rb_fd_set_cloexec): new function.
+	  (ruby_dup): call rb_fd_set_cloexec to set close-on-exec flag.
+	  (rb_sysopen_internal): ditto.
+	  (rb_pipe): ditto.
+	  (io_reopen): ditto.
+	  (io_cntl): ditto.
+
+	* process.c (rb_f_exec): change the default :close_others option to
+	  true.
+	  (rb_f_system): ditto.
+	  (move_fds_to_avoid_crash): call rb_fd_set_cloexec to set
+	  close-on-exec flag.
+	  (ruby_setsid): ditto.
+	  (rb_daemon): ditto.
+
+	* thread_pthread.c (rb_thread_create_timer_thread): call
+	  rb_fd_set_cloexec to set close-on-exec flag.
+
+	* ruby.c (load_file_internal): ditto.
+
+	* file.c (rb_file_s_truncate): ditto.
+	  (file_load_ok): ditto.
+
+	* random.c (fill_random_seed): ditto.
+
+	* ext/pty/pty.c (chfunc): ditto.
+	  (get_device_once): ditto.
+
+	* ext/openssl/ossl_bio.c (ossl_obj2bio): ditto.
+
+	* ext/socket/init.c (rsock_socket): ditto.
+	  (rsock_s_accept_nonblock): ditto.
+	  (rsock_s_accept): ditto.
+
+	* ext/socket/socket.c (rsock_sock_s_socketpair): ditto.
+
+	* ext/socket/ancdata.c (discard_cmsg): ditto.
+	  (make_io_for_unix_rights): ditto.
+
+	* ext/socket/unixsocket.c (unix_recv_io): ditto.
+
+	* ext/io/console/console.c (console_dev): ditto.
+
+	[ruby-core:38140] [Feature #5041]
+
+Sat Oct 22 17:46:27 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: fix a exception name in previous patch.
+
+Sat Oct 22 17:43:33 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: make timeout configurable for DNS query.
+	  patch by Eric Wong.  [ruby-core:38533] [Feature #5100]
+
+Sat Oct 22 02:07:48 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* numeric.c (rb_infinity, rb_nan): use union to prevent bus error
+	  caused by misalignment.  [Bug #5469] [ruby-dev:44657]
+
+	* include/ruby/missing.h (INFINITY, NAN): ditto
+
+Fri Oct 21 22:02:17 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (initial_params): pack in a struct.
+
+	* gc.c (rb_gc_set_params): set parameters always.
+	  [ruby-dev:44648] [Bug #5467]
+
+Fri Oct 21 12:10:20 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* atomic.h: change Solaris checking macro because atomic_ops can work
+	  not only with Sun Studio but also with Fujitsu C Compiler.
+
+Fri Oct 21 02:11:00 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ns_spki.c: Complete documentation.
+	* test/openssl/test_ns_spki.rb: Integrate SPKI#to_text.
+
+Thu Oct 20 22:47:28 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (socklist_insert, socklist_lookup, socklist_delete):
+	  new functions to wrap of st_insert(), st_lookup() and st_delete() to
+	  socklist.
+	  allocating socklist is deferred until it is really needed.
+
+	* win32/win32.c (exit_handler): delete socklist only if it is
+	  initialized.
+
+	* win32/win32.c (rb_w32_sysinit, StartSockets): refactoring: move
+	  initialization of select_mutex to StartSockets().
+
+	* win32/win32.c (exit_handler): refactoring: delete select_mutex only
+	  if winsock is used.
+
+Thu Oct 20 22:38:53 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkcs5.c: add note on timing attacks and general
+	  documentation.
+
+Thu Oct 20 21:19:15 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* vm_eval.c (check_funcall): set array elements one-by-one to fix
+	  compile error with Fujitsu C Compiler 5.6 on Solaris 10 on Sparc.
+	  [Bug #5464] [ruby-dev:44632]
+
+Thu Oct 20 13:09:35 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/defines.h (flush_register_windows): use software
+	  trap on Debian Sparc 32-bit userspace.  [Bug #5244]
+
+Thu Oct 20 12:28:22 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkcs5.rb: add RFC 6070 tests for PBKDF2 with
+	  HMAC-SHA1
+
+Thu Oct 20 11:42:23 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (mmprepare): fix for fragmental size.
+
+	* util.c (mmswap_, mmrot3_): portability improvement.
+
+Thu Oct 20 05:58:02 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_ns_spki.c (Init_ossl_ns_spki):  Stub documentation
+	  for Netscape SPKI.
+
+Thu Oct 20 05:13:39 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 2.6.2 (r6712)
+	* test/minitest/*: ditto
+
+Thu Oct 20 06:55:32 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* lib/openssl/buffering.rb: Force multi-byte strings to be treated as
+	  binary data.
+	* test/openssl/test_ssl.rb: Add test for it.
+
+	Thanks to Niklas Baumstark for reporting the issue!
+
+	[Ruby 1.9 - Bug #5233] [ruby-core:39120]
+
+Wed Oct 19 17:06:54 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* version.h (RUBY_VERSION): finally declare start of 2.0 work!
+
+Wed Oct 19 11:48:44 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* error.c (Init_Exception):  Document $! and $@.  Provide
+	  recommendations for creating exceptions for a library.
+
+Wed Oct 19 11:25:46 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* error.c (Init_Exception):  Add hierarchy of Exception subclasses.
+	  Based on patch by Sylvain Daubert.  [Ruby 1.9 - Bug #5438]
+
+Wed Oct 19 11:04:47 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* enum.c: Reformat block args to a single standard, { |args| ... }.
+	  Patch by b t.  [Ruby 1.9 - Bug #5393]
+
+Wed Oct 19 12:11:26 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: Remove set, but unused variables.
+	  ext/openssl/ossl_pkey.c: ditto
+
+	* ext/openssl/ossl_pkey_dh.c: Make functions passed to
+	  rb_thread_blocking_region return VALUE instead of void.
+	  ext/openssl/ossl_pkey_dsa.c: ditto
+	  ext/openssl/ossl_pkey_rsa.c: ditto
+
+Tue Oct 18 23:28:53 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (identhash): share with type_numhash.
+
+	* st.c (st_hashtype_num): rename from type_numhash.
+
+Tue Oct 18 23:07:30 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (ruby_current_thread): probeprofiler has been removed
+	  long ago.
+
+Tue Oct 18 23:05:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/ripper/eventids2.c (ripper_init_eventids2): separate
+	  initializations of IDs and objects.
+
+	* ext/ripper/tools/generate.rb (generate_eventids1): ditto.
+
+	* parse.y (Init_ripper, InitVM_ripper): fix inversed roles.
+
+Sun Oct 16 19:46:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.gemspec (files): fixed typo, and
+	  removed nonexistent file.
+
+	* ext/bigdecimal/bigdecimal.gemspec (homepage): added.
+
+	* ext/io/console/io-console.gemspec (homepage): ditto.
+
+Fri Oct 14 12:13:57 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (pty_check): should return nil until the child
+	  terminates or stops.  [ruby-dev:44600] [Bug #2642]
+
+Fri Oct 14 11:19:37 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_ary_rotate): export.
+
+Fri Oct 14 05:58:05 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* atomic.h (ATOMIC_INC, ATOMIC_DEC): return old values.
+	  [ruby-dev:44596] [Bug #5439]
+
+	* signal.c (ruby_atomic_exchange): no needs to define on the
+	  platforms where atomic.h is available.
+
+Thu Oct 13 19:29:40 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* atomic.h (ATOMIC_*): use atomic_ops(3C) when SunStudio on Solaris.
+	  [ruby-dev:44596] [Bug #5439]
+
+Thu Oct 13 18:13:04 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* atomic.h(ATOMIC_SET): add cast to void to prevent misuse.
+	  [ruby-dev:44596] [Bug #5439]
+
+Thu Oct 13 18:04:27 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_finalize_deferred, rb_objspace_call_finalizer):
+	  should use ATOMIC_EXCHANGE() to check the previous value.
+	  [ruby-dev:44596] [Bug #5439]
+
+Wed Oct 12 23:39:58 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* test/openssl/test_ssl.rb: Move duplicated tests for SSL::Session to
+	  test_ssl_session.rb
+
+Tue Oct 11 08:49:40 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c (rb_ary_initialize):  Improve explanation of Array.new
+	  parameters.  Patch by Alvaro Pereyra Rabanal.  [Ruby 1.9 - Bug #5425]
+	* array.c (rb_ary_s_try_convert):  Fix typo (try => tries)
+	* array.c (rb_ary_rindex):  Add spacing for block.
+	* array.c (rb_ary_uniq_bang):  Describe block
+	* array.c (rb_ary_uniq):  ditto
+
+Tue Oct 11 07:55:38 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c:  Add a description to Array, minor cleanups.  Patch by
+	  Andrea Singh.  [Ruby 1.9 - Bug #5412]
+
+Tue Oct 11 06:09:52 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/pp.rb:  Move PP documentation to top of class PP.  Patch by
+	  Sylvain Daubert.  [Ruby 1.9 - Bug #5430]
+
+Tue Oct 11 06:06:29 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/coverage/coverage.c (Init_coverage):  Change list format and
+	  describe Coverage.result output.  Patch by Sylvain Daubert.
+	  [Ruby 1.9 - Bug #5428]
+
+Tue Oct 11 05:53:23 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c (Init_Object):  Add reference to BasicObject, brief
+	  explanation of constant lookup.  Based on patch by Alvaro Pereyra
+	  Rabanal.
+	  [Ruby 1.9 - Bug #5426]
+
+Sun Oct  9 11:06:52 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/psych/test_yamldbm.rb: don't run test if the system
+	  don't support yaml/dbm.
+
+	* test/syck/test_yamldbm.rb: ditto.
+
+Sat Oct  8 08:54:56 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* enum.c (group_by):  Improve group_by description.  Patch by b t.
+	  [#5411]
+
+Sat Oct  8 03:17:51 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/shell.rb:  Document some methods of Shell.  Patch by Carol
+	  Nichols.  [Ruby 1.9 - Bug #5417]
+
+Fri Oct  7 17:54:28 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (assert_send, assert_not_send):
+	  parenthesize non-empty arguments.
+
+Fri Oct  7 06:35:50 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c:  Use + for arguments described in documentation to allow
+	  rdoc -C2 to work better.  Remove <code> from method references to
+	  allow cross-references in HTML documentation.
+
+Thu Oct  6 18:46:23 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* vm_eval.c (make_no_method_exception): fix typo.
+
+	* vm_insnhelper.c, vm_insnhelper.h: ditto.
+
+Thu Oct  6 16:29:30 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (make_no_method_execption): extract from
+	  raise_method_missing().
+
+	* vm_eval.c (send_internal): remove inadvertent symbol creation
+	  from public_send.  based on a patch by Jeremy Evans <code AT
+	  jeremyevans.net> in [ruby-core:38576]. [Feature #5112]
+
+	* vm_insnhelper.c (vm_call_method): remove inadvertent symbol
+	  creation from send and __send__, too.
+
+Thu Oct  6 14:59:11 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/time.rb:  Clean up Time documentation.  Patch by Jake Goulding.
+	  [Ruby 1.9 - Bug #5416]
+
+Thu Oct  6 10:00:54 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* enum.c (group_by):  Improve documentation based on patch by b t.
+
+Thu Oct  6 09:56:30 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* enum.c:  Clean up wording in Enumerable documentation.  Patch by b t.
+	  [Ruby 1.9 - Bug #5411]
+
+Thu Oct  6 09:17:18 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* time.c (Init_Time):  Remove editorial comments from Time
+	  documentation, fix link.
+
+Thu Oct  6 09:14:20 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* time.c (Init_Time):  Improve Time documentation.  Patch by Shane
+	  Emmons.  [Ruby 1.9 - Bug #5404]
+	* lib/time.rb:  Improve time.rb documentation including Time.strptime.
+	  Patch by Shane Emmons.  [Ruby 1.9 - Bug #5402]
+
+Thu Oct  6 08:54:05 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* random.c:  Improve documentation of Random.  Patch by Gregory
+	  Parkhurst.  [Ruby 1.9 - Bug #5410]
+
+Thu Oct  6 01:44:51 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* cont.c (cont_mark): mark original Thread object from saved_thread.
+	  [ruby-dev:44571] [Bug #5386]
+
+Wed Oct  5 16:33:04 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_cfunc): remove useless hack.
+
+Wed Oct  5 05:56:39 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* hash.c (Init_Hash):  Improve Hash documentation.  Patch by Alvaro
+	  Pereyra Rabanal.  [Ruby 1.9 - Bug #5405]
+
+Wed Oct  5 05:47:59 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* random.c (Init_Random):  Add a top-level comment for Random.  Patch
+	  by Brett Bim.  [Ruby 1.9 - Bug #5403]
+
+Wed Oct  5 02:50:27 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/syntax_error.rb: Add file, line, offset, and
+	  message attributes during parse failure.
+	* ext/psych/parser.c: Update parser to raise exception with correct
+	  values.
+	* test/psych/test_exception.rb: corresponding tests.
+
+Wed Oct  5 01:52:16 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/parser.c (parse): Use context_mark for indicating error
+	  line and column.
+
+Wed Oct  5 01:22:08 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: use normal begin / rescue
+	  since postfix rescue cannot receive the exception class. Thanks
+	  nagachika!
+
+Tue Oct  4 21:10:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (class_alloc): allocate extra memory after containing
+	  object setup to get rid of rare-but-potential memory leak.
+
+	* gc.c (gc_mark_children): skip marking extended members if ptr is
+	  NULL.
+
+Tue Oct  4 16:17:50 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/time.rb (Time.strptime): use Time.at if d[:seconds] is set.
+	  Reported by Christopher Eberz. [ruby-core:39903] Bug #5399
+
+Tue Oct  4 11:44:10 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (rb_gc_set_params): ruby_verbose can be Qnil, so use RTEST.
+
+Tue Oct  4 08:33:41 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/etc/etc.c:  Document Etc, Etc.sysconfdir, Etc.systmpdir.  Patch
+	  by mathew murphy.  [Ruby 1.9 - Bug #5396]
+
+Tue Oct  4 08:21:51 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/shellwords.rb:  Update toplevel comment with an example.  Patch
+	  by Samnang Chhun.  [Ruby 1.9 - Bug #5388]
+
+Tue Oct  4 08:15:50 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* proc.c (proc_call):  Update documentation to match argument handling
+	  of proc/Proc.new/lambda/->()
+
+Tue Oct  4 07:59:16 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* proc.c (proc_call):  Fix documentation of Proc#call vs Proc#===.
+	  [Ruby 1.9 - Bug #5349]
+
+Tue Oct  4 07:43:18 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c (rb_ary_initialize):  Make Array.new description match
+	  call-seq.  Patch by Henry Maddocks.  [Ruby 1.9 - Bug #5344]
+
+Tue Oct  4 07:35:23 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c (rb_ary_initialize):  Add output for examples.  Patch by
+	  Jonathan Mukai.  [Ruby 1.9 - Bug #5216]
+
+Tue Oct  4 07:30:50 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* array.c (rb_ary_s_create):  Add example results for Array::[].  Patch
+	  by Jonathan Mukai.  [Ruby 1.9 - Bug #5215]
+
+Tue Oct  4 07:15:17 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 1.8.11.  Move Deprecate into the
+	  Gem namespace.
+
+Tue Oct  4 06:43:47 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: update psych version.
+	* ext/psych/psych.gemspec: generate new gemspec for new version.
+
+Tue Oct  4 06:29:55 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: calling `yaml` rather than `to_yaml`.
+	* ext/psych/lib/psych/nodes/node.rb: Rename `to_yaml` to just `yaml`
+	  in order to avoid YAML::ENGINE switching from replacing this method.
+	* test/psych/helper.rb: fix tests for method name change.
+	* test/psych/test_document.rb: ditto
+	* test/psych/visitors/test_emitter.rb: ditto
+
+Tue Oct  4 06:20:19 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: Match values against the
+	  floating point spec defined in YAML to avoid erroneous parses.
+	* test/psych/test_numeric.rb: corresponding test.
+
+Tue Oct  4 05:59:24 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: ToRuby visitor can be
+	  constructed with a ScalarScanner.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ScalarScanner can be
+	  passed to the YAMLTree visitor.
+
+Tue Oct  4 05:47:23 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: Define Regexp::NOENCODING
+	  for 1.9.2 backwards compatibility.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: Fix Date string
+	  generation for 1.9.2 backwards compatibility.
+
+Mon Oct  3 23:56:39 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* gc.c (rb_gc_set_params): output GC parameter change messages only
+	  if -w/-v options are specified. these messages are output to stderr,
+	  not to stdout. [ruby-core:39795] [Bug #5380]
+
+	* test/ruby/test_gc.rb (test_gc_parameter): add test for it.
+
+Sun Oct  2 20:05:32 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (rb_thread_mark), cont.c (cont_mark): revert r33369 and r33371
+	  that may cause SEGV in certain environments.
+
+Sun Oct  2 12:14:06 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/psych/test_yamldbm.rb: add test case.
+	* test/syck/test_yamldbm.rb: ditto.
+
+Sun Oct  2 11:28:09 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/yaml/store.rb: make initialize method signature match the
+	  superclass signature.
+
+Sun Oct  2 10:44:01 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* io.c: fix documentation of ARGF.lineno=.
+
+Sat Oct  1 20:03:19 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_framework): try as Objective-C.
+	  https://twitter.com/nagachika/status/120294447660539904
+
+Sun Oct  2 08:43:25 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (rb_thread_mark), cont.c (cont_mark): self pointer should not
+	  be marked by itself. Patch by Koichi Sasada.
+	  [ruby-dev:44567] [Bug #5386]
+
+Sun Oct  2 00:42:14 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (rb_thread_mark): rb_thread_t needs self to be marked.
+	  [ruby-dev:44566] [Bug #5386]
+
+Sat Oct  1 09:48:53 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* gc.c (add_heap_slots, init_heap): reset heaps_inc zero when
+	  heap slots are expanded by environment variable RUBY_HEAP_MIN_SLOTS.
+	  [ruby-core:39777] [Bug #5380]
+
+	* test/ruby/test_gc.rb (test_gc_parameter): add test for it.
+
+	* test/ruby/envutil.rb (assert_normal_exit): add :child_env option to
+	  enable pass environment variables to child process.
+
+Thu Sep 29 13:17:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_join_1): should not copy the encoding of non-string
+	  element after string element.  [ruby-core:39776] [Bug #5379]
+
+Thu Sep 29 11:53:56 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (slot_sweep, rb_gc_finalize_deferred)
+	  (rb_objspace_call_finalizer, rb_gc): run finalizers
+	  sequentially.  [ruby-dev:44562]
+
+Thu Sep 29 20:37:38 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/gdbm/gdbm.c (rb_gdbm_fatal): adjust argument type.
+
+Thu Sep 29 20:10:42 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (is_id_value, is_live_object): extract from id2ref().
+
+	* gc.c (run_finalizer): use object instead of object id.
+
+Thu Sep 29 20:07:36 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* use RB_TYPE_P which is optimized for constant types, instead of
+	  comparison with TYPE.
+
+Wed Sep 28 09:20:37 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (pthread_np.h): needs pthread.h to be included
+	  previously on OpenBSD.  a patch by George Koehler <xkernigh AT
+	  netscape.net> at [ruby-core:39752].  [Bug #5376]
+
+Wed Sep 28 04:41:35 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/psych/test_yamlstore.rb: use tmpdir for tmpfile.
+	* test/syck/test_yamlstore.rb: ditto.
+
+Wed Sep 28 04:10:46 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/bigdecimal/README: update report to.
+
+Tue Sep 28 04:05:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal_en.html: removed because this file isn't
+	  maintained now.
+
+	* ext/bigdecimal/bigdecimal_ja.html: ditto.
+
+Tue Sep 27 09:55:40 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c: make native_fd_select().
+	* thread.c (do_select): remove #ifdef _WIN32. Instead, use
+	  native_fd_select() always.
+
+Tue Sep 27 09:44:59 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (do_select): remove cygwin specific hack. It's layer
+	  violation and too large hack.
+	* thread.c (cmp_tv, subtract_tv): removed.
+
+Tue Sep 27 03:50:19 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/rexml/test_sax.rb: add require 'rexml/document'.
+
+Tue Sep 27 03:32:27 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/psych/test_yamldbm.rb: fix #setup and #teardown.
+	  [Bug #5370] [ruby-core:39730]
+	* test/syck/test_yamldbm.rb: ditto.
+
+Mon Sep 26 11:27:38 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httputils.rb: Add MIME Type definition of .js and .svg.
+	  patched by Hal Brodigan. [ruby-core:39704] [Bug #5365]
+
+Mon Sep 26 09:20:44 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: remove DJGPP support. It's not longer supported
+	  since ruby 1.9.0.
+
+Mon Sep 26 09:07:46 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/defines.h: remove NextStep, OpenStep, Rhapsody
+	  support. Last activity of their OSes are 7 years ago.
+	* configure.in: ditto.
+	* dir.c: ditto.
+	* ext/tk/extconf.rb: ditto.
+
+Mon Sep 26 09:02:49 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: remove a code for human68k. it's no longer
+	  supported since r19677.
+
+Sun Sep 25 23:43:32 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: fix int_ossl_asn1_decode0_cons when being
+	  fed arbitrary string values.
+	  Clearly distinguish between the cases "universal, infinite and
+	  not a SEQUENCE or SET" and "universal SEQUENCE or SET, possibly
+	  infinite". Raise error for universal tags that are not infinite.
+	* test/openssl/test_asn1.rb: add a test for this.
+
+	  Thanks to Hiroshi Yoshida for reporting this bug.
+	  [Bug #5363] [ruby-dev:44542]
+
+Sun Sep 25 20:57:18 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/syck/test/yamldbm.rb: add test for Syck::DBM.
+	* test/psych/test_yamldbm.rb: add test for Psych::DBM.
+	* test/psych/test_yamlstore.rb: add test for Psych::PStore.
+
+Sun Sep 25 20:54:10 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/yaml/dbm/dbm.rb: fix #update, add #key for using instead #index.
+	  [Bug #5305][ruby-dev:44485]
+
+Sun Sep 25 16:54:33 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (require_enc): reject only loading from untrusted
+	  load paths.  [ruby-dev:44541] [Bug #5279]
+
+	* transcode.c (load_transcoder_entry): ditto.
+
+Sun Sep 25 16:45:05 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: ignore all warnings from an arbitrary
+	  header in /usr/local/include.
+
+Sun Sep 25 03:43:03 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enum.c (slice_before_i): use rb_attr_get to suppress wrong warning
+	  for internal instance variable slicebefore_initial_state.
+
+Fri Sep 23 14:20:14 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_asn1.c: remove unused variable.
+
+Fri Sep 23 13:46:59 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_ssl_session.rb: execute test_session_exts_read
+	  only for OpenSSL versions >= 0.9.8k. Thanks, Eric Wong, for
+	  reporting this.
+	  [Bug #4961] [ruby-core:37726]
+
+Fri Sep 23 11:59:08 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_ssl_session.rb: ensure server calls callbacks in
+	  test_ctx_server_session_cb. Thanks to Eric Wong for the patch.
+	  [Bug #5336] [ruby-core:39619]
+
+Thu Sep 22 02:53:19 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_cfunc): suppress a warning.  note that
+	  `volatile type *var' doesn't make var itself volatile.
+
+Thu Sep 22 01:52:48 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* thread_pthread.c (ubf_select): activate timer thread when interrupt
+	  blocking thread.
+	  A patch created by Koichi Sasada. [ruby-core:39634] [Bug #5343]
+	  to cover race condition, timer thread periodically send SIGVTARLM to
+	  threads in signal thread list. so you should activate timer thread
+	  when interrupt a thread.
+
+Wed Sep 21 16:55:26 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/io/wait/test_io_wait.rb (TestIOWait#setup): of course, the
+	  behavior of mingw is just same with mswin.
+
+Tue Sep 20 18:08:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_get_cvar_base): reduce duplicated checks and
+	  move a warning outside the loop.
+
+Mon Sep 19 18:55:51 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/fileutils.rb (module FileUtils): improve performance of
+	  FileUtils.compare_stream. a patch by Masaki Matsushita.
+	  [Feature #5337] [ruby-core:39622]
+
+Mon Sep 19 18:42:58 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/-ext-/old_thread_select/test_old_thread_select.rb:
+	  select() with timeout may return early in old Linux kernels
+	  with 250 Hz tickrate and no dynticks, so skip everything older
+	  than 2.6.32 (which has long term support).
+	  And, Make the timing assertions consistently use assert_operator with
+	  timing difference in error message
+	  Patch by Eric Wong. [Bug #5335] [ruby-core:39618]
+
+Mon Sep 19 09:28:06 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* test/openssl/test_ssl.rb (class OpenSSL):  Test
+	  OpenSSL::SSL::SSLSocket#session and #session=.
+
+Mon Sep 19 07:54:17 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_clone): singleton class should be attached
+	  singleton object to.  a patch by Satoshi Shiba <shiba AT rvm.jp>
+	  at [ruby-dev:44460].  [Bug #5274]
+
+Sat Sep 17 23:34:10 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_data_type): inherit the core type in ripper so
+	  that checks in core would work.  [ruby-core:39591] [Bug #5331]
+
+Sat Sep 17 12:44:04 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/find.rb (Find.find): add documentation that Find.find
+	  without block returns an enumerator.
+
+Thu Sep 15 11:39:43 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (mark_entry, mark_key, mark_keyvalue): adjust callback
+	  argument types.
+
+Thu Sep 15 01:44:10 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/tk/*: Change encoding from EUC-JP to UTF-8
+
+Wed Sep 14 11:43:37 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_fd_rcopy): added an argument guard.
+	  Patch by NAKAMURA Usaku. [Bug #5306] [ruby-core:39435]
+
+Tue Sep 13 20:21:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pstore.rb, test/test_pstore.rb: suppress warnings with -v.
+
+	* lib/pstore.rb (PStore): always open in binary mode even if
+	  default encodings are set.  [Bug #5311] [ruby-core:39503]
+
+Tue Sep 13 05:37:15 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* io.c (Init_IO): update BINARY comment.  it should not change the
+	  encoding of the result to ASCII-8BIT.  [ruby-talk:387719]
+
+Mon Sep 12 19:55:00 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_select): fix to ignore an argument
+	  modification of rb_thread_fd_select().
+	  based on a patch by Eric Wong. [Bug #5306] [ruby-core:39435]
+	* thread.c (rb_fd_rcopy): New. for reverse fd copy.
+
+	* test/-ext-/old_thread_select/test_old_thread_select.rb
+	(test_old_select_false_positive): test for bug5306.
+
+	* ext/-test-/old_thread_select/old_thread_select.c (fdset2array):
+	  New. convert fdsets to array.
+	* ext/-test-/old_thread_select/old_thread_select.c (old_thread_select):
+	  return 'read', 'write', 'except' argument of rb_thread_select()
+	  to ruby script.
+
+Mon Sep 12 13:38:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT, README.EXT.ja (2.2.2), parse.y (rb_check_id): add
+	  documents for rb_check_id().
+
+Mon Sep 12 12:53:39 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rake/file_list.rb (Rake::FileList#egrep): there is no need to
+	  open files in binary mode.
+	  see more details in https://github.com/jimweirich/rake/issues/74
+
+Mon Sep 12 12:42:36 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_exception.rb (TestException#test_exit_success_p):
+	  assert also the cases when exiting with true and false.
+
+	* lib/test/unit/assertions.rb (assert_send): make arguments in
+	  the default message clearer.
+
+Sun Sep 11 05:23:14 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Deal with subclasses of Matrix [redmine #5307]
+
+Sat Sep 10 13:38:20 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* dir.c (dir_s_aref):
+	* dir.c (dir_entries): Two small documentation fixes.
+	  A patch from Aaron Lerch. [Bug #5302] [ruby-core:39404]
+
+Sat Sep 10 08:30:03 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (GC_PROFILE_MORE_DETAIL, CALC_EXACT_MALLOC_SIZE):
+	  define macros only if they are not defined.
+	  fixes: [Ruby 1.9 - Feature #5291]
+
+Sat Sep 10 08:25:47 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (bv_decls): parse.y relies on $$ = $1 before action
+	  routines.  a patch from Michael Edgar.  [Bug #5303]
+	  [ruby-core:39429]
+
+Sat Sep 10 01:37:55 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* sample/drb/dhasenc.rb: coding cookie of Emacs is coding,
+	  not encoding.
+
+	* sample/mine.rb: ditto.
+
+Fri Sep  9 21:56:40 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_sqrt): Fix comment.
+	  BigDecimal#sqrt requires argument. Reported by Makoto Kishimoto.
+	  Thanks for your contribution. [Bug #5267] [ruby-dev:44452]
+
+Fri Sep  9 11:00:55 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/rubygems/test_gem_commands_help_command.rb: Add one
+	  `require` because if run test-all with test/unit parallel
+	  running, sometimes this test fails by some constants not found.
+	  The error reason is some worker doesn't require the file needed by
+	  this test. This issue is related to [ruby-core:36168].
+
+Fri Sep  9 10:22:03 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_select): fix a typo to initialize efds
+	  properly.  [Bug #5299] [ruby-core:39380]
+
+Fri Sep  9 02:02:09 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* template/yarvarch.ja:
+	  Change encoding from Shift_JIS to UTF-8
+
+Thu Sep  9 01:14:00 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* sample/drb/README.rd.ja:
+	* sample/drb/dhasenc.rb:
+	* sample/mine.rb:
+	  Change encoding from EUC-JP to UTF-8
+
+Thu Sep  8 21:03:22 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: import nkf 2.1.2 (be9c280)
+	  Bump version number/release date only.
+
+Thu Sep  8 12:43:18 2011  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (Init_GC): defined GC::Profiler.raw_data. based on the
+	  patch by Eric Hodel. [ruby-core:37857] [Bug #4991]
+
+Thu Sep  8 09:02:53 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (id2ref): objects which are unmarked but not in sweep_slots
+	  are not dead.
+
+Thu Sep  8 07:44:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_declare_transcoder, load_transcoder_entry): no
+	  longer need to limit the length of transcoder library name.
+
+Thu Sep  8 07:36:36 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/syck/lib/syck/types.rb: use toplevel Syck.
+	  for the case someone define Syck::Syck (or YAML::Syck).
+
+Thu Sep  8 07:33:12 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (id2ref): unmarked object is already dead while lazy
+	  sweeping, and to it cannot come back since other objects
+	  referred from it might have been freed already.
+
+Wed Sep  8 03:48:00 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/readline/README.ja:
+	  Change encoding from EUC-JP to UTF-8
+
+Wed Sep  8 02:59:00 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/rexml/test_encoding.rb:
+	  Add require 'require 'rexml/document'
+
+Wed Sep  8 02:53:00 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/nkf/nkf-utf8/nkf.c:
+	  Change encoding from ISO-2022 to UTF-8
+
+Wed Sep  7 23:41:24 2011  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb, test/rexml/test_comment.rb:
+	  allow a single hyphen in comment. [Bug #5278] [ruby-core:39289]
+	  Reported by Thomas Fritzsche. Thanks!!!
+
+Wed Sep  7 17:27:18 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/yaml.rb: explicitly specify ::Object to avoid the collision with
+	  Syck::Object.
+
+Tue Sep  6 21:06:49 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb (_run_suites): Now reports are written the
+	  following order: Skip, Failure, Error. [Feature #5282]
+
+	* test_sorting.rb: test for above.
+
+	* test4test_sorting.rb: Ditto.
+
+	* lib/test/unit.rb (run): Put RUBY_DESCRIPTION before quitting.
+	  [Feature #5282]
+
+Tue Sep  6 21:13:47 2011  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* win32/Makefile.sub (INSNS): change command line option -Ks to -Ku
+	  for generate *.inc. because insns.def encoding has been changed SJIS
+	  to UTF-8. if $BASERUBY is 1.9, -Ks cause an error. [Feature #5128]
+	  (same as r33194)
+
+Tue Sep  6 15:55:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (load_transcoder_entry): concatenate paths directly.
+
+	* encoding.c (load_encoding): predefined encoding names are safe.
+	  [ruby-dev:44469] [Bug #5279]
+
+	* transcode.c (load_transcoder_entry): ditto.
+
+Tue Sep  6 12:07:10 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c: enabled econv newline option.
+
+Tue Sep  6 06:44:57 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c (dbl2ival): Fix Float#divmod and #round for 32 bit
+	  platform. part 1 of [bug #5276]
+
+Tue Sep  6 06:44:25 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c (flo_round): Fix criteria for 32 bits platform
+	  part 2 of [bug #5276]
+
+Tue Sep  6 05:37:11 2011  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/rinda/test_rinda.rb (test_core_03_notify): Fixed test failures
+	  [ruby-dev:44430] [Ruby 1.9 - Bug #372]
+
+Mon Sep  5 20:59:30 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* insns.def: change encoding pragma for emacs (shift_jis to utf-8).
+
+Mon Sep  5 19:32:15 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* Makefile.in (INSNS): change command line option -Ks to -Ku for
+	  generate *.inc. because insns.def encoding has been changed SJIS to
+	  UTF-8. if $BASERUBY is 1.9, -Ks cause an error. [Feature #5128]
+
+Mon Sep  5 18:10:56 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_econv_binmode): newline decorators are
+	  exclusive.
+
+Mon Sep  5 15:03:37 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rubygems/test_gem_security.rb
+	  (test_class_build_self_signed_cert): reset opt[:trust_dir] to apply
+	  temporary Gem.user_home.
+
+Mon Sep  5 10:04:35 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* README.ja, README.EXT.ja: resolve conflicts. [ruby-dev:44459]
+
+Mon Sep  5 05:13:22 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c (flo_round): Make Float#round round big values [bug
+	  #5272]
+
+Mon Sep  5 04:28:25 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c (int_round): Integer#round always returns an Integer [Bug
+	  #5271]
+
+Sun Sep  4 22:28:50 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (default_port, default_imap_port,
+	  default_tls_port, default_ssl_port, default_imaps_port):
+	  added methods for consistency with Net::POP.
+	  based on the patch by art lussos.  [ruby-core:38997] [Bug #5198]
+
+Sun Sep  4 21:19:19 2011  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* Change encoding from EUC-JP to UTF-8. [Feature #5128]
+
+Sun Sep  4 00:47:39 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/ruby/test_fiber.rb (TestFiber#test_no_valid_cfp):
+	  add a test. Unlike TestThread#test_no_valid_cfp,
+	  this test succeeds even if win32ole is required (see r33153).
+
+Sun Sep  4 00:11:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_const_set): show the previous definition
+	  location.  [EXPERIMENTAL]
+
+Sat Sep  3 23:56:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (sizeof_struct_dirent_too_small): check if struct
+	  dirent.d_name is too small.
+
+	* configure.in (RUBY_MINGW32): take tool prefix from CC.
+
+Sat Sep  3 23:52:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_next_argv): open in default text mode.
+	  [ruby-core:39234] [Bug #5268]
+
+Sat Sep  3 18:40:57 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* lib/thread.rb (SizedQueue#max=): raise ArgumentError if max is not
+	  positive number. patch by Masaki Matsushita.
+	  [ruby-dev:44449] [Bug #5259]
+
+	* test/thread/test_queue.rb (test_sized_queue_initialize,
+	  test_sized_queue_assign_max): add tests for it.
+
+Fri Sep  2 21:11:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (validate_enc_binmode, prep_stdio): default to text mode on
+	  dosish platforms.  [ruby-core:38822] [Bug #5164]
+
+	* transcode.c (rb_econv_prepare_options): keep default ecflags
+	  unchanged if no options.
+
+Fri Sep  2 14:36:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_const_defined_class): search
+	  ancestors only when global scope.  [ruby-core:39227] [Bug #5264]
+
+Fri Sep  2 09:58:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_tokadd_string, parser_yylex): ignore a backslash
+	  which prefixes an non-ascii character, which has no escape
+	  syntax.  [ruby-core:39222] [Ruby 1.9 - Bug #5262]
+
+Fri Sep  2 04:05:25 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: emit strings tagged as
+	  ascii-8bit as binary in YAML.
+	* test/psych/test_string.rb: corresponding test.
+
+Fri Sep  2 01:07:14 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_round): substitute machine dependent magic number.
+
+Thu Sep  1 17:31:22 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defineclass), vm_insnhelper.c (vm_get_cvar_base): see
+	  also inherited constants for classes without superclass and
+	  modules.  [ruby-core:37698] [Bug #3423]
+
+Thu Sep  1 16:18:44 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* Release GVL while OpenSSL's public key generation.
+
+	    t = Thread.new { print "."; sleep 0.1 }
+	    key = OpenSSL::PKey::RSA.new(2048)
+	    #=> Thread t works in parallel with public key generation if
+		OS/machine allows it.
+
+	  This works with OpenSSL >= 0.9.8. From this version, it has new
+	  public key generation function which allows us to interrupt the
+	  execution while pkey generation iterations.
+
+	* ext/openssl/extconf.rb: Check existence of OpenSSL's new public key
+	  generation function. (DH_generate_parameters_ex,
+	  DSA_generate_parameters_ex and RSA_generate_key_ex.
+
+	* ext/openssl/ossl_pkey.{h,c} (ossl_generate_cb_2,
+	  ossl_generate_cb_stop): Added new callback function for OpenSSL pkey
+	  generation which handles Thread interruption by Ruby.
+	  ossl_generate_cb_stop is the unblock function(ubf) for Ruby which
+	  sets a stop flag. New pkey generation callback ossl_generate_cb_2
+	  checks the stop flag at each iterations of OpenSSL and interrupts
+	  pkey generation when the flag is set.
+
+	* ext/openssl/ossl_pkey_dsa.c (dsa_generate): Call
+	  rb_thread_blocking_region with the above unblock function to release
+	  GVL while pkey generation.
+
+	* ext/openssl/ossl_pkey_rsa.c (rsa_generate): ditto.
+
+	* ext/openssl/ossl_pkey_dh.c (dh_generate): ditto.
+
+	* test/openssl/test_pkey_{dh,dsa,rsa}.rb: Test it.
+
+Thu Sep  1 14:06:54 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_thread.rb (TestThread#test_no_valid_cfp): skip when
+	  win32ole is required.  in such case, win32ole redefines
+	  Thread#initialize, and the block argument becomes to be not the top
+	  of the thread, then this testcase always fails.
+
+Thu Sep  1 10:20:50 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#test_{default_mode_on_dosish,
+	  default_mode_on_unix,text_mode,binary_mode}): sorry for wrong test
+	  committed in r33144.  I'd misunderstood the spec of ruby's universal
+	  newline.
+
+Thu Sep  1 09:27:57 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* variable.c (rb_autoloading_value): Fix the order of definitions.
+	  It is used by autoload_defined_p.
+
+Wed Aug 31 17:28:23 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* variable.c (rb_autoload): There was a chance to run GC (from
+	  rb_str_new2()) before finishing autoload_data_i construction. It
+	  caused SEGV at rb_gc_mark() at autoload_i_mark.
+
+	* variable.c (rb_autoload_load): Move RB_GC_GUARD() to proper
+	  position based on suggestion by CHIKANAGA Tomoyuki at
+	 http://d.hatena.ne.jp/nagachika/20110826/ruby_trunk_changes_33070_33078
+
+	* variable.c (autoload_defined_p): Fix incompatible autoload behavior
+	  that causes Rails crash. Class definition instruction defined in
+	  'defineclass' in insns.def always invokes rb_autoload_load for a
+	  constant. It's invoked for every class definition regardless of
+	  existence of autoload definition. rb_autoload_load checks if a
+	  constant is defined as autoloaded, but new thread-safe autoload
+	  returned different value if the constant is under autoloading.
+
+Wed Aug 31 17:20:56 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* Re-apply r33078, thread-safe autoload which is reverted at r33093.
+
+Wed Aug 31 16:28:04 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io_m17n.rb (TestIO_M17N#test_{default_mode_on_dosish,
+	  default_mode_on_unix,text_mode,binary_mode}): tests for [Bug #5164].
+
+Wed Aug 31 15:54:11 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Merge json gem v1.5.4 (3dab4c5a6a97fac03dac).
+
+Wed Aug 31 13:09:41 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c (flo_round): Avoid overflow by optimizing for trivial
+	  cases [Bug #5227]
+
+Wed Aug 31 00:50:01 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_select_with_thread): and my typo. we all must
+	  be more careful.
+
+Wed Aug 31 00:48:38 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (rb_thread_select): critical typo in r33117.
+
+Wed Aug 31 00:30:49 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/-ext-/old_thread_select/test_old_thread_select.rb
+	  (TestOldThreadSelect#test_old_select_read_timeout): if the machine
+	  is fast enough, the time used by code around IO.select may be smaller
+	  than Time implement threshold.
+
+Wed Aug 31 00:04:38 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/-test-/old_thread_select/old_thread_select.c (old_thread_select):
+	  typo.
+
+	* test/-ext-/old_thread_select/test_old_thread_select.rb
+	  (TestOldThreadSelect#test_old_select_signal_safe): use SIGINT instead
+	  of SIGUSR1 because the former is general and the latter is platform
+	  dependent.
+
+Tue Aug 30 23:59:36 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/intern.h (rb_w32_fd_copy): implement
+	  for rb_thread_select() in thread.c.  the use of rb_fd_copy() is
+	  introduced in r33117.
+	  [Bug #5251] [ruby-core:39195]
+
+	* thread.c (rb_thread_select): must call rb_fd_init() before using
+	  rb_fdset_t.  see the implementations of rb_fd_init()s if you want to
+	  know the reason.
+
+Tue Aug 30 22:34:45 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/dl/test_callback.rb (test_callback_with_string): prevents
+	  temporary string from GC.
+
+Tue Aug 30 22:25:38 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_cfunc): revert r33112.  RB_GC_GUARD macro
+	  protect a VALUE from GC.  It's not for general anti-optimizing
+	  purpose.
+
+Tue Aug 30 11:06:19 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: Merge json gem 1.5.4+ (2149f4185c598fb97db1).
+	  [Bug #5173] [ruby-core:38866]
+
+Tue Aug 30 09:57:50 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/thread.rb (Queue#pop): fix a race against Thread.wakeup.
+	  Patch by Masaki Matsushita <glass.saga at gmail dot com>
+	  [Bug #5195] [ruby-dev:44400]
+
+Tue Aug 30 09:48:07 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c (fiber_entry): fix stack allocation failure on Debian
+	  GNU/kFreeBSD.
+	  Patch by Lucas Nussbaum <lucas at lucas-nussbaum dot net>.
+	  [Bug #5241] [ruby-core:39147]
+
+Tue Aug 30 09:28:01 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_select): rewrite by using
+	  rb_thread_fd_select(). old one is EINTR unsafe.
+	  Patch by Eric Wong. [Bug #5229] [ruby-core:39102]
+
+	* test/-ext-/old_thread_select/test_old_thread_select.rb:
+	  a testcase for rb_thread_select().
+	* ext/-test-/old_thread_select/old_thread_select.c: ditto.
+	* ext/-test-/old_thread_select/depend: ditto.
+	* ext/-test-/old_thread_select/extconf.rb: ditto.
+
+Tue Aug 30 09:08:22 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: fix a build failure on GNU Hurd.
+	  Patch by Samuel Thibault <sthibault at debian dot org>. Thank you!
+	  [Bug #5250] [ruby-core:39185]
+
+Sun Aug 29 23:22:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* test/ruby/test_numeric.rb (test_num2long): modify a test against the
+	  change by r33108.
+
+Sun Aug 29 09:58:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (bit_coerce): A Fixnum and a Bignum are only permitted for
+	 bitwise arithmetic with a Fixnum.  #1792
+
+	* test/ruby/test_fixnum.rb: add tests for the above change.
+
+	* bignum.c (bit_coerce): A Fixnum and a Bignum are only permitted for
+	 bitwise arithmetic with a Bignum.  #1792
+
+	* test/ruby/test_bignum.rb: add tests for the above change.
+
+Sun Aug 28 15:38:17 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/date/date_parse.c (date_zone_to_diff): keep a temporary string
+	  stored in variable while the contents buffer is being used.
+
+	* ext/date/date_parse.c (date_zone_to_diff): get rid of out of bounds
+	  memory read. [ruby-dev:44409] [Bug #5213]
+
+Sun Aug 28 05:29:50 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 2.5.1 (r6596)
+	* test/minitest/*: ditto
+
+Sat Aug 27 20:46:05 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (rb_vm_rewrite_dfp_in_errinfo): change return type
+	  to suppress a warning.
+
+	* vm_core.h: ditto.
+
+Sat Aug 27 19:04:06 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* internal.h (rb_strftime_timespec): moved from time.c and define only
+	  if ruby/encoding.h is included.
+
+	* internal.h (rb_strftime): ditto.
+
+Sat Aug 27 18:53:51 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* proc.c (proc_new): force to rewrite errinfo when calling Proc.new in ensure.
+	  [Bug #5234] [ruby-core:39125]
+	  This code will be removed after changing throw mechanism (see r33064).
+
+	* vm.c (rb_vm_rewrite_dfp_in_errinfo): new function.
+
+	* vm.c (vm_make_env_each): changed accordingly.
+
+	* vm_core.h: ditto.
+
+	* bootstraptest/test_flow.rb: add tests for above.
+
+Sat Aug 27 18:44:06 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* internal.h (rb_strftime_timespec): move to time.c because it depends
+	  encoding.h.
+
+Sat Aug 27 18:17:58 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): get enc argument to specify
+	  the encoding of the format. On Windows (at least Japanese Windows),
+	  Time#strftime("%Z") includes non ASCII in locale encoding (CP932).
+	  So convert locale to default internal. [ruby-core:39092] [Bug #5226]
+
+	* strftime.c (rb_strftime): ditto.
+
+	* strftime.c (rb_strftime_timespec): ditto.
+
+	* internal.h (rb_strftime_timespec): follow above.
+
+	* time.c (rb_strftime_alloc): ditto.
+
+	* time.c (strftimev): ditto.
+
+	* time.c (time_strftime): ditto.
+
+	* time.c (time_to_s): the resulted string of Time#to_s is always
+	  ascii only, so this should be US-ASCII.
+
+	* time.c (time_asctime): ditto.
+
+Sat Aug 27 11:18:12 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* Revert r33078. It caused a Rails application NoMethodError.
+
+	 /home/nahi/git/emptyApp/ruby/1.9.1/gems/rack-mount-0.6.14/lib/rack/mount/utils.rb:157: warning: toplevel constant ScanError referenced by Regin::Parser::ScanError
+	 /home/nahi/git/emptyApp/ruby/1.9.1/gems/rack-mount-0.6.14/lib/rack/mount/vendor/regin/regin/parser.rb:17:in `parse_regexp': undefined method `scan_str' for #<Regin::Parser:0x00000002344548> (NoMethodError)
+
+Sat Aug 27 08:44:58 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Import RDoc 3.9.4.  Typo and grammar fixes by Luke Gruber.
+	  [Ruby 1.9 - Bug #5203]
+
+Sat Aug 27 07:53:34 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/open-uri.rb:  Fix indentation of OpenURI::OpenRead#open.  Use ++
+	  instead of `' for method arguments in open-uri.rb
+
+Sat Aug 27 07:22:07 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/pathname/lib/pathname.rb:  Fix typos and grammar mistakes.  Patch
+	  by Luke Gruber.  [#5203]
+	* ext/pty/lib/expect.rb:  ditto
+	* lib/mathn.rb:  ditto
+	* lib/net/http.rb:  ditto
+	* lib/open-uri.rb:  ditto
+	* lib/ostruct.rb:  ditto
+	* lib/tempfile.rb:  ditto
+	* lib/thread.rb:  ditto
+	* lib/weakref.rb:  ditto
+	* sample/webrick/httpproxy.rb:  ditto
+
+Sat Aug 27 04:03:18 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_data_to_ary): fix type of variable
+	  (long -> unsigned long) to suppress a warning.
+
+Sat Aug 27 04:02:11 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: add a decl. of rb_autoloading_value().
+
+Fri Aug 26 19:12:08 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* variable.c: Make autoload thread-safe. See #921.
+
+	  What's the problem?
+	    autoload is thread unsafe. When we define a constant to be
+	    autoloaded, we expect the constant construction is invariant. But
+	    current autoload implementation allows other threads to access the
+	    constant while the first thread is loading a file.
+
+	  What's happening inside?
+	    The current implementation uses Qundef as a marker of autoload in
+	    Constant table. Once the first thread find Qundef as a value at
+	    constant lookup, it starts loading a defined feature. Generally a
+	    loaded file overrides the Qundef in Constant table by module/class
+	    declaration at very beginning lines of the file, so other threads
+	    can see the new Module/Class object before feature loading is
+	    finished. It breaks invariant construction.
+
+	  How to solve?
+	    To ensure invariant constant construction, we need to override
+	    Qundef with defined Object after the feature loading. For keeping
+	    Qundef in Constant table, I expanded autoload_data struct in
+	    Module to have a slot for keeping the defined object while feature
+	    loading. And changed Module's constant lookup/update logic a
+	    little so that the slot is only visible from the thread which
+	    invokes feature loading. (== the first thread which accessed the
+	    autoload constant)
+
+	  Evaluation?
+	    All test passes (bootstrap test, test-all and RubySpec) and added
+	    8 tests for threading behavior. Extra logics are executed only
+	    when Qundef is found, so no perf drop should happen except
+	    autoloading.
+
+	* variable.c (rb_autoload): Prepare new autoload_data struct.
+
+	* variable.c (rb_autoload_load): Load feature and update Constant
+	  table after feature loading is finished.
+
+	* variable.c (rb_const_get_0): When the fetched constant is under
+	  autoloading, it returns the object only for the thread which starts
+	  autoloading.
+
+	* variable.c (rb_const_defined_0): Ditto.
+
+	* variable.c (rb_const_set): When the specified constant is under
+	  autoloading, it sets the object only for the thread which starts
+	  autoloading. Otherwise, simply overrides Qundef with constant
+	  override warning.
+
+	* vm_insnhelper.c (vm_get_ev_const): Apply same change as
+	  rb_const_get_0 in variable.c.
+
+	* test/ruby/test_autoload.rb: Added tests for threading behavior.
+
+Fri Aug 26 10:10:37 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 1.8.10.  Fixes security issue in
+	  creating ruby-format gemspecs.  Fixes Gem.dir not being at the front
+	  of Gem.path to fix uninstall and cleanup commands.  Fixes gem
+	  uninstall stopping on the first missing gem.
+
+Fri Aug 26 08:21:10 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* time.c (strftimev): Make Time#to_s default to US-ASCII encoding but
+	  respect Encoding.default_internal. [ruby-core:39092]
+	* test/ruby/test_time.rb (class TestTime): Corresponding test.
+
+Thu Aug 25 09:43:16 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/lib/openssl/bn.rb:  Hide copyright info from RDoc.
+	* ext/openssl/lib/openssl/digest.rb:  ditto
+	* ext/openssl/lib/openssl/x509.rb:  ditto
+	* ext/openssl/lib/openssl/cipher.rb:  ditto
+
+Thu Aug 25 09:25:48 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_digest.c:  Document OpenSSL::Digest::digest and add
+	  an example to OpenSSL::Digest.  Patch by Sylvain Daubert.
+	  [Ruby 1.9 - Bug #5166]
+	* ext/openssl/lib/openssl/digest.rb (module OpenSSL):  ditto
+
+Thu Aug 25 08:19:43 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_make_env_each): work around to solve Bug #2729.
+	  fixes: Bug #2729
+	  a patch from Kazuki Tsujimoto <kazuki@callcc.net>
+	  This problem is caused by changing dfp (dynamic env pointer)
+	  from saved dfp.  Saved dfp is pointed env in VM stack.  However,
+	  the dfp can be moved because VM copies env from VM stack to
+	  the heap.  At this copying, dfp was also changed.  To solve this
+	  problem, I'll try to change throw mechanism (not save target dfp,
+	  but save target cfp).
+
+	* bootstraptest/test_flow.rb: add a test for above.
+
+Thu Aug 25 07:57:33 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c (int_round): Fix Integer#round [ruby-core:39096]
+
+Thu Aug 25 07:00:00 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h, vm_insnhelper.c, vm.c, vm_method.c, insns.def:
+	  Manage a redefinition of special methods for each classes.
+	  A patch from Joel Gouly <joel.gouly@gmail.com>.  Thanks!
+
+Thu Aug 25 06:51:08 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: Fixing psych version number.
+	* ext/psych/psych.gemspec: updating the gemspec.
+
+Thu Aug 25 06:11:35 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/nodes/node.rb: default `to_yaml` encoding to be
+	  UTF-8.
+	* test/psych/test_encoding.rb: test yaml dump encoding.
+
+Thu Aug 25 01:24:33 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/fileutils/test_fileutils.rb (test_chmod_symbol_mode): Solaris
+	  seems to behave the same as FreeBSD.
+
+Thu Aug 25 01:11:36 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/ruby/test_rubyoptions.rb (test_script_from_stdin): slave pty
+	  should be manipulated because master pty may not be a tty on some
+	  environment (e.g. Solaris). [Bug:#5222] [ruby-dev:44420]
+
+Wed Aug 24 15:13:56 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.h, iseq.c, compile.c: Change the line number data structure
+	  to solve an issue reported at [ruby-dev:44413] [Ruby 1.9 - Bug #5217].
+	  Before this fix, each instruction has an information including
+	  line number (iseq::iseq_insn_info_table).  Instead of this data
+	  structure, recording only line number changing places
+	  (iseq::iseq_line_info_table).
+	  The order of entries in iseq_line_info_table is ascending order of
+	  iseq_line_info_table_entry::position.  You can get a line number
+	  by an iseq and a program counter with this data structure.
+	  This fix reduces memory consumption of iseq (bytecode).
+	  On my measurement, a rails application consumes 21.8MB for
+	  iseq with this fix on the 32bit CPU.  Without this fix, it
+	  consumes 24.7MB for iseq [ruby-dev:44415].
+
+	* proc.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_method.c: ditto.
+
+	* vm.c (rb_vm_get_sourceline): change to use rb_iseq_line_no().
+
+Wed Aug 24 09:49:10 2011  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (defined): fix to checking class variable.
+	  A patch by Magnus Holm <judofyr@gmail.com>.  Thanks!
+
+	* test/ruby/test_variable.rb: add a test for above.
+
+Wed Aug 24 08:53:06 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 3.9.3.  Fixes RDoc with `ruby -Ku`.  Allows
+	  HTTPS image paths to be turned into <img> tags.  Prevents special
+	  markup inside <tt> from being processed.
+
+Wed Aug 24 07:57:43 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 1.8.9.  Fixes uninstalling multiple
+	  gems and gem cleanup.
+
+Wed Aug 24 06:45:20 2011  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 2.5.0 (r6557)
+	* test/minitest/*: ditto
+
+Wed Aug 24 00:38:22 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (update_coverage): skip coverage count up if the current
+	  line is out of the way.  rb_sourceline() is unreliable when source
+	  code is big.  [ruby-dev:44413]
+
+	* test/coverage/test_coverage.rb: add a test for above.
+
+Tue Aug 23 15:23:56 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* load.c (rb_f_require):  Improve documentation of Kernel#require.
+	  [Ruby 1.9 - Bug #5210]
+
+Tue Aug 23 11:27:26 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_read_header): Ensure that each section of
+	  gzip header is readable to avoid SEGV.
+
+	* test/zlib/test_zlib.rb (test_corrupted_header): Test it.
+
+Mon Aug 22 23:43:33 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* sprintf.c (rb_str_format): add RB_GC_GUARD to prevent temporary
+	  strings from GC.
+
+Sun Aug 21 17:49:53 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* iseq.c (iseq_s_disasm): remove variable which is no longer used
+	  since r33013.
+
+Sun Aug 21 14:20:58 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: use LD_LIBRARY_PATH_64 on 64-bit Solaris.
+
+Sat Aug 20 13:19:52 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* iseq.c (iseq_s_disasm): fix a bug that may cause SEGV.
+
+	* test/ruby/test_method.rb (test_body): add a test for the above change.
+
+Sat Aug 20 10:43:24 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_read): return new string if nil
+	  is explicitly given as a buffer ([Bug #5207]), otherwise set the
+	  encoding.  also removed dead code.
+
+Fri Aug 19 14:25:51 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_spawn_v, proc_spawn): should not wait the
+	  spawned process.
+
+	* process.c (proc_spawn_v): fix missing argument, and try with
+	  /bin/sh only if failed with ENOEXEC.
+
+Fri Aug 19 14:12:57 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (idle): raises a Net::IMAP::Error when the
+	  connection is closed.  based on the patch by Hugo Barauna.
+	  [Bug #5190] [ruby-core:38930]
+
+Fri Aug 19 13:18:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* configure.in: defines _DARWIN_UNLIMITED_SELECT if the target_os
+	  is darwin.
+
+Fri Aug 19 13:14:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* thread.c: add a description for the behavior of select(2) on
+	  Mac OS X 10.7 (Lion).
+
+Fri Aug 19 11:28:58 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (msg_att): accepts extra space before ')'.
+	  based on the patch by art lussos.  [Bug #5163] [ruby-core:38820]
+
+Wed Aug 17 23:01:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (cannot_be_coerced_into_BigDecimal):
+	  remove duplication.
+
+Wed Aug 17 15:27:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (cannot_be_coerced_into_BigDecimal):
+	  add a new function for raising error when an object cannot coerce
+	  into BigDecimal.  [Bug #5172]
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimalValueWithPrec): use
+	  cannot_be_coerced_into_BigDecimal function.
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_log): ditto.
+
+	* test/bigdecimal/test_bigdecimal.rb: test for the above changes.
+
+	* test/bigdecimal/testbase.rb (under_gc_stress): add a new utility
+	  method to run tests under the condition of GC.stress = true.
+
+Wed Aug 17 10:16:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* rational.c (nurat_coerce): Rational#coerce should converts itself
+	  into Complex if the argument is a Complex with non-zero imaginary
+	  part.  [Bug #5020] [ruby-dev:44088]
+
+	* test/ruby/test_rational.rb (test_coerce): test for the above change.
+
+Wed Aug 17 06:33:19 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_x509cert.c: Add class documentation for
+	  OpenSSL::X509::Certificate.
+
+Wed Aug 17 04:54:25 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_pkey.c: corrected docs, OpenSSL::PKey::DH does
+	  *not* support #sign/verify.
+
+Tue Aug 16 18:56:54 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (ruby_threadptr_data_type): rename to hide.
+	  [ruby-core:38972]
+
+Tue Aug 16 18:52:08 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/mkexports.rb (Exports::Mswin#each_export): exclude Init_
+	  and _threadptr_ functions, as well as mingw.
+
+Tue Aug 16 09:31:44 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/dl:  Add documentation.  Patch by Vincent Batts.
+	  [Ruby 1.9 - Bug #5192]
+
+Tue Aug 16 08:48:26 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/.document (fiddle):  Remove duplicate entry
+	* ext/fiddle:  Complete documentation of Fiddle.  Patch by Vincent
+	  Batts.  [#5192]
+
+Tue Aug 16 08:00:15 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/socket:  Make Socket documentation appear.  Add documentation for
+	  Socket, TCPServer, SOCKSSocket.  Patch by Sylvain Daubert.
+	  [Ruby 1.9 - Feature #5182]
+
+Mon Aug 15 09:58:55 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_ssl.c: Support disabling OpenSSL compression.
+
+	* test/openssl/test_ssl.rb: Add a test for it.
+	  Thanks to Eric Wong for the patch.
+	  [Ruby 1.9 - Feature #5183] [ruby-core:38911]
+
+Sun Aug 14 05:57:01 2011  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_socket.rb (test_connect_timeout): added a test
+	  based on a patch by Eric Wong.  [ruby-core:38910]
+
+Sat Aug 13 22:17:27 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb: do not make the entries related to sitedir and
+	  verdordir if disabled by --without options.  [ruby-core:38922]
+	  [Bug #5187]
+
+Sat Aug 13 17:03:22 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: [ruby-core:38861]
+
+Sat Aug 13 09:39:07 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* test/date/test_*.rb: added tests.
+
+Sat Aug 13 09:36:19 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/date/date_parse.c (parse_ddd_cb): fix r32896. RB_GC_GUARD
+	  insertion position was mistaken. [ruby-dev:44337] [Bug #5152]
+
+Sat Aug 13 09:26:24 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/core_ext.rb: Make Kernel#y private.
+	  [ruby-core:38913]
+
+	* test/psych/test_yaml.rb: corresponding test.
+
+Sat Aug 13 09:05:16 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (date_strftime_alloc): followed the change
+	  of r32885.
+
+	* doc/NEWS-1.9.3: followed the above change.
+
+Sat Aug 13 08:55:38 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: Only consider strings
+	  with fewer than 2 dots to be numbers. [ruby-core:38915]
+
+Sat Aug 13 08:47:20 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: [ruby-core:38855].
+
+Sat Aug 13 03:41:37 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/uri/common.rb:  Fix documentation of URI::Parser.new.  Patch by
+	  Steve Klabnik.  [Ruby 1.9 - Bug #5177]
+
+Sat Aug 13 02:19:57 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/digest/digest.c:  Add documentation for the Digest module.  Patch
+	  by Sylvain Daubert.  [Ruby 1.9 - Bug #5167]
+
+Sat Aug 13 01:56:11 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake:  Update to Rake 0.9.2.2.  Prevent pollution of toplevel
+	  namespace by Commands.  Remove unused variable and debugging
+	  statement in tests.
+
+Fri Aug 12 11:39:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: Describe "no" configure option for site_ruby
+	  and vendor_ruby. Patch by Vit Ondruch. [Bug #5187][ruby-core:38921]
+
+Fri Aug 12 09:00:24 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Import RubyGems 1.8.8.  Fixes encoding of YAML gemspec
+	  from gems.  Github Issue #149
+
+Fri Aug 12 08:17:46 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): use SOMAXCONN for
+	  listen backlog.
+
+	* ext/socket/unixsocket.c (rsock_init_unixsock): ditto.
+
+	* ext/socket/lib/socket.rb (Addrinfo#listen): ditto.
+	  (Socket.tcp_server_sockets_port0): ditto.
+
+	* ext/socket/mkconstants.rb: define SOMAXCONN as 5 if not available.
+
+	[ruby-core:38493]
+
+Fri Aug 12 03:24:35 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Import RDoc 3.9.2.  Fixes TIDYLINK for HTML output.
+
+Thu Aug 11 15:37:42 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* variable.c (autoload_delete): An autoload entry is still in a
+	  RCLASS_IV_TBL, not in a RCLASS_CONST_TBL, so take back the table
+	  changed in r29600. And an autoload entry keeps not a
+	  rb_const_entry_t but a NODE so remove rb_const_entry_t thing added
+	  in r29602.
+
+Thu Aug 11 15:07:36 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (link_command): use LIBRUBYARG in rbconfig for
+	  unbundled extensions.  [ruby-core:38802] [Bug #5147]
+
+	* lib/mkmf.rb (init_mkmf): revert r32902.  [ruby-core:38903]
+
+Wed Aug 10 23:03:55 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb: fix argument check in the previous commit.
+
+Wed Aug 10 22:12:28 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (Socket.tcp): add :connect_timeout option.
+	  (Addrinfo#connect_from): add :timeout option.
+	  (Addrinfo#connect): ditto.
+	  (Addrinfo#connect_to): ditto.
+	  [ruby-core:38538]
+
+Wed Aug 10 21:27:19 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/net/pop.rb: fix typo in document.
+
+	* lib/net/http.rb: ditto.
+
+	* lib/net/imap.rb: ditto.
+
+Wed Aug 10 19:30:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* complex.c (nucomp_rationalize): calls rationalize of real part if
+	  imaginary part is exactly zero.  The patch is made by Marc-Andre
+	  Lafortune.  fixes [Bug #5178] [ruby-core:38885]
+
+	* test/ruby/test_complex.rb (test_rationalize): add a test for the
+	  above change.
+
+	* complex.c (nucomp_to_r): fix RDoc comment.  The patch is made by
+	  Marc-Andre Lafortune.
+
+Wed Aug 10 14:11:07 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (init_mkmf): set $LIBRUBYARG regardless of shared
+	  option.  [ruby-core:38802] [Bug #5147]
+
+Wed Aug 10 02:53:27 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb: come back autoload. OpenSSL constant is used
+	  some places, so it leads mistakes like HTTP.start.
+
+Tue Aug  9 22:57:45 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/date/date_parse.c (date_zone_to_diff): add RB_GC_GUARD.
+	  [ruby-dev:44337] [Bug #5152]
+
+	* ext/date/data_parse.c (parse_ddd_cb): ditto.
+
+Tue Aug  9 14:25:47 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/fiddle/conversions.c (generic_to_value): ffi_arg and ffi_sarg
+	  should be used to handle shorter return value. fix [Bug #3861]
+	  [ruby-core:32504]
+
+	* ext/fiddle/closure.c (callback): ditto
+
+	* ext/fiddle/conversions.h (fiddle_generic): ditto
+
+	* ext/fiddle/conversions.c (value_to_generic): char, short and int
+	  are strictly distinguished on big-endian CPU, e.g. sparc64.
+
+Tue Aug  9 11:21:08 2011  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_lazy_sweep): if sweep target slots are not found, we
+	  try heap_increment() because it might be able to expand the
+	  heap. [Bug #5127] [ruby-dev:44285]
+
+	* gc.c (gc_clear_mark_on_sweep_slots): if a sweeping was
+	  interrupted, we expand the heap if at all possible.
+
+Tue Aug  9 12:20:33 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/fiddle/helper.rb (libc_so, libm_so): Solaris support added.
+	  [ruby-core:38853] [Bug #5168]
+
+	* test/dl/test_base.rb (libc_so, libm_so): on Solaris, remove libc
+	  and libm version numbers for detecting default libc and libm.
+
+Tue Aug  9 09:18:04 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (gzfile_wrap):  Document encoding options.
+
+	* ext/zlib/zlib.c (rb_gzwriter_s_open): ditto
+
+	* ext/zlib/zlib.c (rb_gzreader_s_open): ditto
+
+Sun Aug  7 23:31:32 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* time.c (rb_strftime_alloc): raise ERANGE if width is too large.
+	  Patch by Nobuyoshi Nakada. [Bug #4457] [ruby-dev:43285]
+
+	* test/ruby/test_time.rb (class TestTime): add a test for the
+	  above change.
+
+Sun Aug  7 22:51:45 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/openssl/ossl_asn1.c (decode_eoc): remove unused variables.
+	  Patch by Eric Wong. [Feature #5157] [ruby-core:38798]
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_decode): ditto.
+
+	* ext/openssl/ossl_pkey.c (ossl_pkey_new_from_data): ditto.
+
+Sun Aug  7 22:37:08 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: add -Wunused-variable to default CFLAGS.
+	  Patch by Eric Wong. [Feature #5157] [ruby-core:38798]
+
+Sun Aug  7 15:37:35 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/digest/sha2/sha2ossl.c: use original SHA384_Final on DragonFly.
+
+Sun Aug  7 14:08:16 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/objspace/objspace.c: fix typos in a document.
+
+Sun Aug  7 07:14:57 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* cont.c (HAVE_GETCONTEXT): see getcontext(3) because DragonFly BSD
+	  x64 port doesn't have it.
+
+Sun Aug  7 00:42:55 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/tk/lib/tk/wm.rb (Tk::Wm.command): Add the missing receiver
+	  before calling epath. patched by flori
+	 https://github.com/flori/ruby/commit/aa9474d32e5f2c57f8b0e2e0c528a03f06a4d433
+
+Sat Aug  6 07:06:34 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* marshal.c (w_object):  Fix exception message when _dump_data is not
+	  defined on a T_DATA object.
+
+Fri Aug  5 22:16:20 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* numeric.c (rb_infinity, rb_nan): use WORDS_BIGENDIAN to get endian.
+	  fix [Bug #5160] [ruby-dev:44356]
+
+Fri Aug  5 17:14:11 2011  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/test_syslog.rb (TestSyslog#test_log): Do not be too
+	  specific about the log line format.  Fixes #5081.
+
+Fri Aug  5 15:57:10 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* complex.c (f_signbit): fix compile error in gcc4 on Solaris with
+	  CFLAGS="-std=gnu99".  [ruby-dev:44355] fix [Bug #5159]
+
+	* math.c: ditto.
+
+Fri Aug  5 15:55:33 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_object.rb: tests that respond_to? returns false.
+
+Fri Aug  5 13:32:43 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb, lib/xmlrpc/server.rb: should use
+	  String#bytesize instead of String#size.
+
+Fri Aug  5 12:18:20 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (check_funcall): try respond_to? first if redefined.
+	  [Bug #5158]
+
+Fri Aug  5 09:48:22 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Import RubyGems 1.8.7:
+	  Added missing require for `gem uninstall --format-executable`.
+
+	  The correct name of the executable being uninstalled is now displayed
+	  with --format-executable.
+
+	  Fixed `gem unpack uninstalled_gem` default version picker.
+
+	  RubyGems no longer claims a nonexistent gem can be uninstalled.
+
+	  `gem which` no longer claims directories are requirable files.
+
+	  `gem cleanup` continues cleaning up gems if one can't be uninstalled
+	  due to permissions.  Issue #82.
+
+	  Gem repository directories are no longer created world-writable.
+	  Patch by Sakuro OZAWA.  [Ruby 1.9 - Bug #4930]
+
+Fri Aug  5 07:00:31 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/io/console/test_io_console.rb (test_noctty): daemon() on
+	  Fedora Rawhide seems not to detach the controlling terminal,
+	  when the argument noclose is non-zero.  ref: [Bug #5135]
+
+Thu Aug  4 23:48:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* thread_pthread.c (native_cond_signal): retry to call pthread_cond_signal
+	  and pthread_cond_broadcast if they return EAGAIN in
+	  native_cond_signal and native_cond_broadcast, respectively.
+	  It is for the pthread implementation of Mac OS X 10.7 (Lion).
+	  fixes #5155.  [ruby-dev:44342].
+
+	* thread_pthread.c (native_cond_broadcast): ditto.
+
+	* thread_pthread.c (struct cached_thread_entry): stop using
+	  pthread_cond_t and its functions directly.
+
+	* thread_pthread.c (register_cached_thread_and_wait): ditto.
+
+	* thread_pthread.c (use_cached_thread): ditto.
+
+Thu Aug  4 20:29:41 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: when Solaris cc, use $(CC) to link shared libs.
+
+Thu Aug  4 20:19:11 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* error.c (report_bug): use a small message buffer instead of BUFSIZ.
+	  It is needed for avoiding nested SIGSEGV on Linux.
+	  Note: BUFSIZ is not proper buffer size. It's unrelated with maximum
+	  filename length. :-/
+	  [Bug #5139] [ruby-dev:44315]
+
+Thu Aug  4 16:08:45 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (gem): install all gemspecs under lib and ext.
+
+	* tool/rbinstall.rb (Gem::Specification): may not be defined when
+	  cross-compiling and BASERUBY is 1.8.
+
+Thu Aug  4 11:30:36 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/missing.h: define __syscall on OpenBSD as r32702.
+
+Thu Aug  4 03:02:54 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* tool/rbinstall.rb: use rubygems to load gemspecs, copy actual
+	  gemspecs on install rather than generate fake ones for all gems.
+
+Thu Aug  4 02:45:10 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* configure.in: set CXX variable to the C++ compiler that matches the
+	  C compiler specified by CC variable (e.g. use g++-4.2 for gcc-4.2).
+
+Thu Aug  4 02:21:10 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (link_command): use static library only for bundled
+	  extensions.  [Bug #5147]
+
+Thu Aug  4 02:02:10 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/psych.gemspec: installing psych as a gem.
+
+Wed Aug  3 16:01:35 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* util.c, include/ruby/util.h (ruby_add_suffix): remove the function.
+	  [Bug #5153] [ruby-core:38736]
+
+	* io.c (argf_next_argv): remove the call of above function.
+
+	* ext/-test-/add_suffix, test/-ext-/test_add_suffix.rb: remove the test
+	  extension module because this is only for testing ruby_add_suffix().
+
+	* LEGAL: remove the mention about a part of util.c, because now we
+	  removed the part.
+
+	* io.c (argf_next_argv): now the new filename is not guaranteed to
+	  use, so should check the return value of rename(2).
+
+	* test/ruby/test_argf.rb (TestArgf#test_inplace_rename_impossible):
+	  now we expect same result with other platforms on no_safe_rename
+	  platforms (=Windows).
+
+Wed Aug  3 09:18:08 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/xmlrpc/webrick_testing.rb (WEBrick_Testing#start_server):
+	  Like r32795, bind address should be specified.
+
+Wed Aug  3 07:46:30 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_find): mistakenly remained !.  [Bug #5150]
+
+Wed Aug  3 00:11:08 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/prettyprint.rb: update document.  [ruby-core:36776]
+
+Tue Aug  2 22:04:46 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (init_heap): allocate sigaltstack after heaps are allocated.
+	  [ruby-dev:44315] [Bug #5139]
+
+	* vm.c (thread_free): use free because objspace is not ready.
+
+	* vm.c (th_init): use malloc because objspace is not ready.
+
+Tue Aug  2 20:10:16 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* test/testunit/test_parallel.rb: pass "--ruby" option to
+	  test/testunit/tests_for_parallel/runner.rb. [Bug #5132] [ruby-dev:44303]
+
+Tue Aug  2 15:53:37 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* encoding.c (str_to_encoding): rename from to_encoding and
+	  use str_to_encindex.
+
+	* encoding.c (str_to_encindex): split from to_encoding.
+
+	* encoding.c (rb_to_encoding): use str_to_encoding.
+
+	* encoding.c (rb_obj_encoding): don't bypass rb_encoding*.
+	  If it uses rb_encoding*, it bypass encindex. If it uses encindex,
+	  it doesn't bypass.
+
+	* encoding.c (enc_find): add shortcut for encoding object, use
+	  str_to_encindex, and avoid bypass rb_encoding*.
+
+Tue Aug  2 12:03:16 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (recursive_hash): hash value of emptied hash should be
+	  equal to an empty hash.  [ruby-core:38650]
+
+Tue Aug  2 11:42:15 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_enc_symname2_p): :! is valid symbol.  [Bug #5136]
+
+Tue Aug  2 07:33:29 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/net/http/test_http.rb (TestNetHTTP_version_1_1_methods#test_timeout_during_HTTP_session):
+	  If you connect to localhost, you should listen localhost.
+
+	* test/net/http/test_https.rb (TestNetHTTPS#test_timeout_during_SSL_handshake):
+	  ditto.
+
+Tue Aug  2 06:18:15 2011  Luis Lavena  <luislavena@gmail.com>
+
+	* lib/rubygems/installer.rb (class Gem): Correct path check on Windows
+	  Possible fix for [Ruby 1.9 - Bug #5111]
+	* test/rubygems/test_gem_installer.rb (load Gem): ditto
+
+Mon Aug  1 20:12:03 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_process.rb (TestProcess#windows?): new method.
+
+	* test/ruby/test_process.rb (TestProcess#*): use above method.
+
+	* test/ruby/test_process.rb (TestProcess#test_execopts_redirect):
+	  windows doesn't support FD_CLOEXEC.
+
+Mon Aug  1 15:45:23 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rake/test_rake_functional.rb:  Don't assume the binary name of
+	  ruby is "ruby".  [Ruby 1.9 - Bug #5114]
+	* test/rake/helper.rb:  ditto
+
+Mon Aug  1 15:31:14 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/io/console/test_io_console.rb (TestIO_Console#test_sync):
+	  Skip when PTY allocation failed (that's not our fault).
+
+Mon Aug  1 15:04:12 2011  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* test/xmlrpc/test_webrick_server.rb (Test_Webrick#setup_http_server):
+	  XMLRPC::Client.new3(), when called without host: argument, tries
+	  to  connect to  a host  where "localhost"  resolves to.   On the
+	  other  hand  a  WEBrick::HTTPServer.new(), when  called  without
+	  BindAddress:  argument, tries  to listen  all the  address where
+	  getaddrinfo(AF_UNSPEC) resolves to.   This is a mismatch because
+	  "localhost" might not resolve to one of those listening sockets.
+	  We  would  better explicitly  specify  "localhost"  here and  if
+	  failed, just skip the whole test.
+
+Mon Aug  1 14:24:56 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc.rb:  Import RDoc 3.9.1.  Fixes bugs in the RDoc::Markup
+	  parser.
+
+Mon Aug  1 12:00:35 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (concatstrings): don't use initial ASCII-8BIT string.
+	  [ruby-core:38635] [Bug #5126]
+
+Sun Jul 31 22:57:16 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* enc/Makefile.in (ECHO1): Same as the recent fix in common.mk.
+	  ":" in a make variable replacement cause a syntax error with
+	  /usr/ccs/bin/make on Solaris.  Uses $(NULLCMD) instead.
+
+Sun Jul 31 21:16:02 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* complex.c (f_signbit): gcc4 on Solaris DOES have signbit but does
+	  not have it on header.
+
+	* math.c: ditto.
+
+Sun Jul 31 21:09:04 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (node_name.inc): Use $(Q) for consistency.
+
+	* Makefile.in (INSNS): ditto.
+
+Sun Jul 31 21:19:51 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* lib/mkmf.rb (configuration:ECHO1): Same as the recent fix in
+	  common.mk.
+	  ":" in a make variable replacement cause a syntax error with
+	  /usr/ccs/bin/make on Solaris.  Uses $(NULLCMD) instead.
+
+Sun Jul 31 20:39:12 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* common.mk (ECHO1): nmake does not allow parenthesis in make variable
+	  replacement.
+
+Sun Jul 31 23:06:57 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (check_env): print debug messages to stderr.
+	[Feature #4871] [ruby-dev:43743]
+
+Sun Jul 31 22:50:23 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (vm_make_env_each): don't save prev env value.
+	It is no longer used. [Feature #4871] [ruby-dev:43743]
+
+	* vm.c (check_env): changed accordingly.
+
+Sun Jul 31 20:21:36 2011  "Yuki Sonoda (Yugui)"  <yugui@yugui.jp>
+
+	* common.mk (ECHO1): ":" in a make variable replacement cause a syntax
+	  error with /usr/ccs/bin/make on Solaris.  Uses $(NULLCMD) instead.
+
+	* configure.in (NULLCMD): new check.
+
+	* Makefile.in (NULLCMD): Reflects checking in configure.
+
+	* win32/Makefile.sub (NULLCMD): new assignment.
+
+Sun Jul 31 18:58:59 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_pipe): pipe on cygwin can succeed half but fail
+	  half.
+
+Sun Jul 31 11:31:07 2011  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c: check if cfp is valid. [Bug #5083] [ruby-dev:44208]
+
+Sun Jul 31 09:18:28 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 3.9.  Fixed `ri []`, stopdoc creating an
+	  object reference, nodoc for class aliases, verbatim === lines.
+
+Sun Jul 31 01:29:08 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_each_byte): remove unused variable e.
+
+Sat Jul 31 01:23:45 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* test/bigdecimal/test_bigdecimal.rb (test_version): removed.
+
+Sat Jul 30 23:19:09 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/default_gems: separate from tool/rbinstall.rb.
+
+Sat Jul 30 23:14:44 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_each_byte): rbuf can be refreshed during yield.
+	  [Bug #5119]
+
+Sat Jul 30 22:35:50 2011  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* strftime.c (NEEDS): avoid SEGV due to integer overflow in
+	  sparc-solaris2.10 and i686-linux.  fix [Bug #4456] [ruby-dev:43284]
+
+Sat Jul 30 17:26:26 2011  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_variant.rb: use skip method to skip the test.
+
+	* test/win32ole/test_win32ole_variant_outarg.rb: ditto.
+
+Sat Jul 30 14:27:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_version): version 1.1.0.
+
+	* ext/bigdecimal/bigdecimal.gemspec: turn into a default gem.
+
+	* tool/rbinstall.rb: ditto.
+
+Sat Jul 30 11:21:55 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_core.h (ALT_STACK_SIZE): use MINSIGSTKSZ*2 instead of SIGSTKSZ*2.
+	  [ruby-core:38607]
+
+Sat Jul 30 10:39:14 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm.c (th_init): preallocate alternative stack.
+	  NoMemoryError is better than rb_bug, of course.
+	  Patch by Eric Wong. [ruby-core:38572][ruby-core:38594].
+
+	* signal.c (rb_register_sigaltstack): ditto.
+
+	* vm_core.h: moved ALT_STACK_SIZE definition from signal.c.
+	* vm.c (thread_free): use xfree() instead of free().
+
+Sat Jul 30 07:20:49 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb (udp_server_sockets): unused variable
+	  removed.
+	  patch by Jeremy Evans.  [ruby-core:38600]
+
+Fri Jul 29 23:56:32 2011  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb: call OpenSSL::Random.seed at the
+	  SecureRandom.random_bytes call.
+	  based on the patch by Masahiro Tomita.  [ruby-dev:44270]
+
+Fri Jul 29 23:53:48 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_set_len): new function to set array length.
+
+	* vm_eval.c (method_missing): set the length of argv array, to mark
+	  arguments.
+
+	* vm_eval.c (rb_apply): get rid of too large alloca.
+
+Fri Jul 29 20:48:39 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: fix typos.
+
+Fri Jul 29 20:28:56 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: use whitespaces as a separator.
+
+Fri Jul 29 18:59:07 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: add documents for constants.
+	  patch by Eric Hodel.  [ruby-core:37853] [Bug #4989]
+
+Fri Jul 29 16:00:43 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (enable_pthread): use -pthread on OpenBSD without
+	  explicit option. patched by Jeremy Evans. [ruby-core:38572]
+
+Thu Jul 28 23:36:28 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* ext/fiddle/closure.c (callback): use rb_ary_tmp_new() instead of
+	  xmalloc() to allocate an array for arguments of callback procedure,
+	  to prevent arguments from being swept by GC. [ruby-core:38546]
+	  [Bug #4929]
+
+Thu Jul 28 22:36:06 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_initialize): Avoid possible
+	  SEGV from AES encryption/decryption. Processing data by
+	  Cipher#update without initializing key (meaningless usage of Cipher
+	  object since we don't offer a way to export a key) could cause SEGV.
+
+	  In OpenSSL, the EVP which has EVP_CIPH_RAND_KEY flag (such as DES3)
+	  allows uninitialized key, but other EVPs (such as AES) does not
+	  allow it. Calling EVP_CipherUpdate() without initializing key causes
+	  SEGV so we set the data filled with "\0" as the key by default. See
+	  #2768.
+
+	* test/openssl/test_cipher.rb: test it.
+
+Thu Jul 28 14:25:08 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rubygems/user_interaction.rb (Gem::StreamUI#tty?): typo.
+
+Thu Jul 28 12:32:53 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/callback/mkcallback.rb (gencallback): use PTR2NUM.
+
+	* ext/dl/cptr.c (rb_dlptr_aref, rb_dlptr_aset): check NULL pointer
+	  dereference.
+
+	* ext/dl/cptr.c (rb_dlptr_s_to_ptr): use rb_check_funcall.
+
+	* ext/dl/cptr.c (rb_dlptr_s_to_ptr): fix wrapping condition.
+
+Thu Jul 28 04:53:31 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/delegate.rb:  Move file-level documentation to the appropriate
+	  classes.
+
+Thu Jul 28 02:15:04 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cfunc.c (dlcfunc_mark), ext/dl/cptr.c (dlptr_mark):
+	  workaround to mark wrapped object.  this is not a true fix,
+	  because [Bug #4929] is caused by the interface design of DL.
+
+Thu Jul 28 00:28:15 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb: add OpenBSD case.
+	  patched by Jeremy Evans [ruby-core:38530] see #5097
+
+	* test/ruby/test_process.rb: ditto.
+
+Wed Jul 27 22:46:59 2011  CHIKANAGA Tomoyuki  <nagachika00@gmail.com>
+
+	* test/rinda/test_rinda.rb (test_remote_array_and_hash):
+	  add local variables to protect objects from GC. [ruby-dev:44253]
+	  [Bug #5104]
+
+Wed Jul 27 17:55:54 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/missing.h: define __syscall if the platform has
+	  __syscall in the library but doesn't define it in headers
+	  for example Mac OS X.
+
+Wed Jul 27 15:39:14 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* object.c:  Add usage documentation for BasicObject.  Based on patch
+	  by Thomas Sawyer.  [Ruby 1.9 - Bug #5067]
+
+Wed Jul 27 12:24:17 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/uninstaller.rb:  Add missing require and update
+	  messaging to avoid confusion with uninstall --format-executable.
+	  [Ruby 1.9 - Bug #4062]
+
+Wed Jul 27 09:34:24 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 1.8.6.1.
+
+Wed Jul 27 09:27:59 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_pkcs12.rb: Add test and intermediate certificates.
+	  [ Ruby 1.9 - Feature #3793 ] [ruby-core:32088]
+
+Wed Jul 27 01:05:32 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.c (rb_print_undef_str): new function to raise
+	  NameError for undefined method.
+
+	* load.c (rb_mod_autoload_p), object.c (rb_mod_const_get),
+	  variable.c (rb_f_untrace_var, set_const_visibility), vm_method.c
+	  (rb_mod_{remove,undef,alias}_method, set_method_visibility):
+	  remove inadvertent symbol creation.  based on the first patch by
+	  Jeremy Evans at [ruby-core:38447].  [Feature #5089]
+
+	* vm_method.c (obj_respond_to): fix the respond_to_missing? override
+	  case.  based on the patch by Jeremy Evans at [ruby-core:38417].
+	  [Feature #5072]
+
+	* parse.y (rb_check_id): make the given name a symbol or a string.
+	  based on the second patch by Jeremy Evans at [ruby-core:38447]
+
+Wed Jul 27 00:50:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb (Rational#to_d):
+	  zero or negative precision is error.  fixes #5098.
+	  [ruby-dev:44210]
+
+	* test/bigdecimal/test_bigdecimal_util.rb: add test for the above
+	  change.
+
+Wed Jul 27 00:48:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb (Float#to_d): modified for
+	  specifying precision.  fixes #5098.  [ruby-dev:44210]
+
+	* test/bigdecimal/test_bigdecimal_util.rb: add test for the above
+	  change.
+
+Wed Jul 27 00:45:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb (Integer#to_d): added
+	  for symmetry to BigDecimal() function with an Integer.
+	  fixes #5098.  [ruby-dev:44210]
+
+	* test/bigdecimal/test_bigdecimal_util.rb: add test for the above
+	  change.
+
+Wed Jul 27 00:30:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb (BigDecimal#to_d): added
+	  for adapting other Numeric subclasses.  [ruby-dev:44245]
+
+	* test/bigdecimal/test_bigdecimal_util.rb: test for the above change.
+
+Wed Jul 27 00:27:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bigdecimal/bigdecimal.c (VpDup) a new function for duplicating
+	  a BigDecimal.
+
+	* bigdecimal/bigdecimal.c (BigDecimal_new): support generating a new
+	  BigDecimal from another BigDecimal using BigDecimal global function
+	  or constructor.  [ruby-dev:44245]
+
+Tue Jul 26 23:33:24 2011  Igor Zubkov  <igor.zubkov@gmail.com>
+
+	* array.c: Fix typo. https://github.com/ruby/ruby/pull/36
+
+Mon Jul 25 23:51:01 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c: pre-allocate the unlinked_method_entry_list_entry struct to
+	  avoid memory allocation during GC.  based on a patch from Eric Wong.
+	  [ruby-core:38498]
+
+Mon Jul 25 23:39:33 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rake/test_rake_directory_task.rb (TestRakeDirectoryTask#
+	  test_directory_win32): fixed wrong test.
+
+Mon Jul 25 22:36:11 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (struct METHOD), gc.c (gc_marks), vm_method.c
+	  (rb_gc_mark_unlinked_live_method_entries): fix SEGV bug.
+	  rb_method_entry_t was free'd even when the method is still on the
+	  stack if it is BMETHOD (i.e., Method#call).  This is because
+	  rb_method_entry_t is embedded in struct METHOD.  This commit
+	  separates them and marks the live method entries.
+	  See [ruby-core:38449] in detail.  fix [Bug #5047] [ruby-core:38171]
+
+Mon Jul 25 22:14:37 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb: Fix possible HTTP header formatting failure by
+	  'Basic' header. Long username caused the base64 String truncation in
+	  HTTP header which is not allowed. See #5046.
+
+	* test/xmlrpc/test_webrick_server.rb: test it.
+
+Mon Jul 25 15:04:33 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/lib/openssl.rb: End of transition period introduced by
+	  [ruby-dev:38018]. From the next version of 1.9.3, you should use
+	    require "openssl"
+	  instead of
+	    require "openssl/ssl"
+	  and
+	    require "openssl/x509"
+
+Mon Jul 25 13:46:38 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/x509.rb: Cosmetic change: move definition
+	  introduced in r30152 to x509-internal.rb.
+
+Mon Jul 25 13:09:42 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_shutdown): Avoid randomly generated
+	  SSLError from SSLSocket just after invoking SSLSocket#close.
+	  OpenSSL's SSL_shutdown could try to send alert packet and it might
+	  set SSLerr(global error stack) as the result. It causes the next
+	  SSL read/write operation to fail by unrelated reason.
+
+	  By design, we're ignoring any error at SSL_shutdown() so we clear
+	  global error stack after SSL_shutdown is called. See #5039.
+
+Sun Jul 24 20:29:53 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: refine the recvmsg test.
+
+Sun Jul 24 20:02:31 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: fix the recvmsg test.
+
+Sun Jul 24 08:42:51 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test recvmsg allocates file descriptors for
+	  fd passing even with MSG_PEEK.
+
+	* ext/socket/ancdata.c: use the above test result.
+
+Sun Jul 24 01:04:50 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/specification.rb:  Restore behavior of
+	  Gem::Specification#loaded.  [Ruby 1.9 - Bug #5032]
+
+Sun Jul 24 00:05:00 2011  Jeremy Evans  <merch-redmine@jeremyevans.net>
+
+	* error.c (rb_name_error_str): new function to raise NameError
+	  with the name string but not ID.
+
+	* object.c, proc.c, variable.c: more removal of inadvertent symbol
+	  creation.  [Feature #5079]
+
+Sat Jul 23 21:14:00 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* lib/cmath.rb (cbrt): should return a real number if possible.
+
+Sat Jul 23 20:12:52 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/rake/test_rake_functional.rb (setup): Use __FILE__ for the base
+	  directory. Current directory is not the top source directory when
+	  the building process runs on other than there.
+
+	* test/rake/test_rake_rake_test_loader.rb: ditto.
+
+	* test/rake/test_rake_task_argument_parsing.rb
+	  (test_terminal_width_using_hardcoded_80): hardcoded 80 is used
+	  when app.unix? is false.
+
+Sat Jul 23 20:11:50 2011  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: an issue that is same as  [ruby-dev:44071].
+	* ext/date/date_strftime.c: identical to [ruby-dev:44112].
+
+Sat Jul 23 19:12:53 2011  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_err_in_callback.rb (test_err_in_callback):
+	  skip test if ADODB.connection is not available.
+
+Sat Jul 23 15:37:04 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (rb_enc_symname_type): :$a!, @a! and so on are not
+	  valid symbols, so they should be inspected with quotes.
+
+Sat Jul 23 17:06:25 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_update_max_fd): validate fd.
+
+	* ext/socket/rubysocket.h (rsock_discard_cmsg_resource): add
+	  msg_peek_p argument for the declaration.
+
+	* ext/socket/ancdata.c (discard_cmsg): add msg_peek_p argument.
+	  assume FreeBSD, NetBSD and MacOS X doesn't generate passed fd
+	  when MSG_PEEK.
+	  (rsock_discard_cmsg_resource): add msg_peek_p argument.
+	  (bsock_recvmsg_internal): call rsock_discard_cmsg_resource with
+	  msg_peek_p argument.
+
+	* ext/socket/unixsocket.c (unix_recv_io): call
+	  rsock_discard_cmsg_resource with msg_peek_p argument.
+
+Sat Jul 23 14:38:28 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rake*: Remove dependencies on flexmock and session gems.
+	  [Ruby 1.9 - Bug #4987]
+
+Sat Jul 23 12:19:04 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_check_id): take care of attrset ID created
+	  implicitly by local ID.  [Bug #5084]
+
+	* parse.y (rb_check_id): conversion condition was inverse.
+	  [Bug #5084]
+
+Fri Jul 22 21:46:54 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_insnhelper.c (vm_call_cfunc): added volatile for a workaround
+	  of cfp consistency error problem on OS X 10.7 (Lion). It's
+	  suspected llvm optimization bug.
+	  [Bug #5074] [ruby-dev:44185]
+
+Fri Jul 22 21:18:20 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (WFKV_): unroll the loop of regexp.
+
+	* lib/uri/generic.rb (URI.decode_www_form_component): ditto.
+
+Fri Jul 22 21:06:39 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_{const,cvar}_defined, rb_obj_ivar_defined):
+	  avoid inadvertent symbol creation in reflection methods.  based
+	  on a patch by Jeremy Evans at [ruby-core:38367].  [Feature #5072]
+
+	* vm_method.c (rb_mod_method_defined)
+	  (rb_mod_{public,private,protected}_method_defined)
+	  (obj_respond_to): ditto.
+
+	* parse.y (rb_check_id): new function returns already interned ID
+	  or 0.
+
+Fri Jul 22 20:44:49 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_is_global_id, rb_is_attrset_id): add missing
+	  predicates.
+
+Fri Jul 22 20:24:38 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_object.rb (TestObject#test_respond_to_missing):
+	  2nd argument of respond_to_missing? is not optional.
+
+Fri Jul 22 19:05:47 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_enc_symname2_p): get rid of potential out-of-bound
+	  access.
+
+Fri Jul 22 13:55:59 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Net::HTTP#finish is used to manually close
+	  connections.  [Ruby 1.9 - Bug #5045]
+
+Fri Jul 22 13:51:29 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/readline/readline.c:  Add examples for Readline.completion_proc=.
+	  [Ruby 1.9 - Bug #5057]
+
+Fri Jul 22 13:03:12 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_hmac.c: Revert checking return type of
+	  HMAC_Init_ex as it is not compatible with OpenSSL < 1.0.0.
+
+Fri Jul 22 12:10:21 2011  Eric Hodel  <drbrain@segment7.net>
+
+	* tool/rbinstall.rb (default gems):  Install executables into the fake
+	  gem dir for Gem.bin_path.  [#4485]
+
+Fri Jul 22 11:20:20 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_digest.c: Check return value of EVP_DigestInit_ex.
+	* ext/openssl/ossl_hmac.c: Check return value of HMAC_Init_ex.
+	  Thanks, Jared Jennings, for the patch.
+	  [ Ruby 1.9 - Bug #4944 ] [ruby-core:37670]
+
+Fri Jul 22 09:09:43 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* ext/openssl/ossl_engine.c: Avoid double free of ENGINE reference.
+	* test/openssl/test_engine.rb: Add a test for it.
+	  Thanks to Ippei Obayashi for providing the patch.
+	  [ Ruby 1.9 - Bug #5062 ] [ruby-dev:44173]
+
+Fri Jul 22 06:37:13 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/csv.rb: Do not modify CSV.generate's argument [ruby-core:38356]
+
+Thu Jul 21 20:59:59 2011  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (discard_cmsg): workaround for MacOS X Lion.
+
+Thu Jul 21 20:02:11 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* thread.c (set_trace_func, thread_set_trace_func_m): reset tracing
+	  state when set_trace_func hook is removed.  This is workaround patch
+	  to force to reset tracing state that is broken by continuation call.
+	  a patch from James M. Lawrence.  [Feature #4347] [ruby-core:34998]
+
+	* test/ruby/test_continuation.rb (class TestContinuation): add a test
+	  for above.  a patch from James M. Lawrence.
+
+Thu Jul 21 19:27:19 2011  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* node.c (dump_node): add today's knowledge.  "init arguments (m)" and
+	  "init arguments (p)" of compile.c indicates a Ruby code that
+	  evaluates multiple assignments that is in method or block
+	  parameters:  def foo((m1,m2), (m3,m4), *r, (p1,p2), (p3,p4)); end
+	  The former (init arguments (m)) evaluates the multiple assignments
+	  before rest argument, that are (m1,m2) and (m3,m4).  The letter
+	  (init arguments (p)) does ones after rest argument, that are
+	  (p1,p2) and (p3, p4).
+
+Thu Jul 21 18:11:07 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enum.c (enum_inject): remove empty line to notify rdoc
+	  Enumerable#reduce is alias. patched by milki@github.
+	  https://github.com/ruby/ruby/pull/26
+
+Thu Jul 21 17:30:21 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_delete_at_m): use simple array literal in rdoc.
+	  patched by samuel tonini. [ruby-core:38310] [Bug #5066]
+
+Thu Jul 21 17:14:21 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/httprequest.rb (WEBrick::HTTPRequest#each):
+	  Allow HTTP/0.9 request which doesn't has any header or body.
+	  patched by Felix Jodoin. [ruby-core:38040] [Bug #5022]
+
+Wed Jul 20 23:02:18 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* io.c (rb_update_max_fd): remove parentheses. they are not in
+	  macro.
+
+Wed Jul 20 22:22:23 2011  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_update_max_fd): declaration moved from
+	  internal.h.
+
+	* file.c: ditto.
+
+	* io.c: call rb_update_max_fd for each new fds.
+
+	* process.c: ditto.
+
+	* random.c: ditto.
+
+	* ruby.c: ditto.
+
+	* ext/io/console/console.c: ditto.
+
+	* ext/openssl/ossl_bio.c: ditto.
+
+	* ext/pty/pty.c: ditto.
+
+	* ext/socket/init.c: ditto.
+
+	* ext/socket/socket.c: ditto.
+
+	* ext/socket/ancdata.c: ditto.
+
+	* ext/socket/unixsocket.c: ditto.
+
+Wed Jul 20 15:16:22 2011  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/dl/handle.c (dlhandle_sym): clear previous error with dlerror()
+	  before calling dlsym(). [ruby-dev:44091] [Bug #5021]
+
+Wed Jul 20 07:16:26 2011  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: mention Kernel#warn.  [ruby-core:38119] [Feature #5029]
+
+Tue Jul 19 16:40:45 2011  TAKANO Mitsuhiro (takano32)  <tak@no32.tk>
+
+	* cont.c (cont_save_thread): fix missing semicolon.
+
+Tue Jul 19 16:25:15 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (UPDATE_MAXFD): removed.
+
+Tue Jul 19 16:07:45 2011  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_update_max_fd): new function.
+
+	* internal.h (rb_update_max_fd): declare rb_update_max_fd.
+
+	* thread_pthread.c (rb_thread_create_timer_thread): update max fd when
+	  timer thread pipe is created.
+
+Mon Jul 18 13:36:47 2011  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: define a new BadAlias error class.
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: raise an exception when
+	  deserializing an alias that does not exist.
+
+	* test/psych/test_merge_keys.rb: corresponding test.
+
+Mon Jul 18 00:00:46 2011  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/curses/curses.c: added the new class Curses::Pad, which
+	  supports scrolling.  patch by Eric Hodel.  [Feature #4896]
+	  [ruby-core:37206]
+
+Sun Jul 17 16:26:40 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_check_trusted): new function to check an object is
+	  trusted.
+
+	* struct.c (rb_struct_modify), time.c (time_modify): check by the
+	  above function to show proper class names.  [Bug #5036]
+
+Sun Jul 17 15:30:04 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_warn_m): accept multiple args in like puts.  rdoc
+	  patch by Erik Price at [ruby-core:38119].  [Feature #5029]
+
+Sun Jul 17 07:56:31 2011  Martin Bosslet  <Martin.Bosslet@googlemail.com>
+
+	* test/openssl/test_ssl_session.rb: add PEM SSL session without TLS
+	  extensions. Use this as the default for the tests to ensure
+	  compatibility with OpenSSL 0.9.7.
+	  [ Ruby 1.9 - Bug #4961 ] [ruby-core:37726]
+
+Sat Jul 16 17:29:20 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_UNIVERSAL_ARCH): restore arch flag.
+	  Bug #4977
+
+Sat Jul 16 06:27:51 2011  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/uri/common.rb (module): Remove optional parser argument to
+	  Kernel#URI
+	  [ruby-core:38061]
+
+	* lib/uri/generic.rb (module): ditto
+
+Sat Jul 16 03:19:45 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (is_socket, is_console): add prototypes to fix compile
+	  problem with gcc introduced at r32549.
+	  reported by Jon Forums. [Bug #5030] [ruby-core:38079]
+
+Sat Jul 16 00:55:38 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* time.c (time_dup): used rb_obj_class() instead of CLASS_OF().
+	  The patch is made by Kazuki Tsujimoto. [Bug #5012] [ruby-dev:44071]
+
+	* test/ruby/test_time.rb (TestTime#test_getlocal_dont_share_eigenclass):
+	  added a new test for eigenclass of time object.
+
+Fri Jul 15 19:11:00 2011  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (bigsub_int): add RB_GC_GUARD.  This patch is made by
+	  Makoto Kishimoto.  fixes #4223 [ruby-dev:42907]
+
+	* bignum.c (bigadd_int): ditto.
+
+Fri Jul 15 14:27:53 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_io_cancelable_p): renamed
+	  from rb_w32_has_cancel_io().  now it takes a parameter as fd to check
+	  the fd is console or not, because we cannot cancel console input even
+	  if we have cancel_io function.
+
+	* io.c (WAIT_FD_IN_WIN32): call above function instead of the old one,
+	  so now we can kill the thread which calls STDIN.gets.
+	  the problem was reported by ko1 via IRC.
+
+Fri Jul 15 09:10:41 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/digest/sha2/sha2.c (SHA256_Update, SHA512_Update): avoid Bus
+	  Error caused by unalignment access on Sparc-Solaris (and possibly on
+	  other similar environment.) This patch just do memcpy always instead
+	  of checking architecture. I see no perf drop on my 64bit env. For
+	  more details, see #4320.
+
+	* test/digest/test_digest.rb: add test for unalignment access.
+
+Fri Jul 15 01:51:25 2011  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regint.h (PLATFORM_UNALIGNED_WORD_ACCESS): Power PC does not
+	  allow unaligned word access.
+
+	* st.c (UNALIGNED_WORD_ACCESS): x86_64 allows unaligned word
+	  access as well as i386.
+
+Thu Jul 14 12:19:34 2011  Hiroshi Nakamura  <nahi@ruby-lang.org>
+
+	* ext/openssl/ossl.c (ossl_verify_cb): trap the exception from
+	  verify callback of SSLContext and X509Store and make the
+	  verification fail normally. Raising exception directly from callback
+	  causes orphan resources in OpenSSL stack. Patched by Ippei Obayashi.
+	  See #4445.
+
+	* test/openssl/test_ssl.rb
+	  (test_exception_in_verify_callback_is_ignored): test it.
+
+Tue Jul 12 23:41:49 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* NEWS: add a description of Signal.trap change.
+
+Tue Jul 12 20:02:35 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (reserved_signal_p): reverted a part of r32523.
+	  chikanaga noticed trap(:CHLD) has some realworld usecase.
+	* test/ruby/test_signal.rb (TestSignal#test_reserved_signal):
+	  ditto.
+
+Tue Jul 12 17:12:45 2011  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* vm_method.c (rb_add_method): should not call method_added hook
+	  for undef operation. [Bug #5015]
+
+Tue Jul 12 16:58:44 2011  Shota Fukumori  <sorah@tubusu.net>
+
+	* lib/test/unit.rb(Test::Unit::Options#process_args): Fix bug.
+	  Fix process_args didn't return `@option` after r30939.
+
+Tue Jul 12 14:07:46 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (install_sighandler): fixed a race.
+
+Tue Jul 12 13:49:32 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (sig_trap): don't permit to change a signal handler which
+	  the interpreter reserved.
+	* signal.c (reserved_signal_p): ditto.
+	  [Bug #2616] [ruby-core:27625]
+
+	* test/ruby/test_signal.rb (TestSignal#test_reserved_signal):
+	  added a test for reserved signal.
+
+Tue Jul 12 11:58:28 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.mak: support x86-amd64 cross compile environment.
+
+Mon Jul 11 23:22:28 2011  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* time.c: can't compile time.c on AIX due to missing declaration for
+	  ffs(). It is declared in strings.h on AIX.
+
+Mon Jul 11 15:54:24 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c: removed signal() macro. It's no longer used.
+
+Mon Jul 11 15:02:24 2011  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (rb_num2ull): use FIX2LONG instead of FIX2ULONG. see
+	  rb_num2ulong(). fixed the problem of ObjectSpace._id2ref of IL32LLP64
+	  platforms, introduced at r32433.
+
+Mon Jul 11 05:38:05 2011  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* thread_pthread.c (get_stack): need to adjust stack addr for
+	  [Bug #1813] on AIX.
+
+Mon Jul 11 01:16:27 2011  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): removed
+	  rb_disable_interrupt()/rb_enable_interrupt().
+	* vm_core.h: ditto.
+	* process.c (static void before_exec): ditto.
+	* process.c (static void after_exec): ditto.
+	  [Bug #4765] [ruby-dev:43571]
+
+	* eval_intern.h: removed rb_trap_restore_mask().
+	* vm_eval.c (rb_throw_obj): ditto.
+	* eval.c (setup_exception): ditto.
+
+	* signal.c: removed trap_last_mask.
+	* signal.c (trap_restore_mask): removed.
+	* signal.c (init_sigchld): comment clarification why signal block
+	  is needed. and removed trap_last_mask operation.
+	* signal.c (trap_ensure): removed trap_last_mask operation.
+
+	* signal.c (rb_disable_interrupt, rb_enable_interrupt): made
+	  static and removed sigdelset(SIGVTALRM) and sigdelset(SIGSEGV).
+
+	* process.c (rb_syswait): removed implicit signal handler change.
+
+Sun Jul 10 23:49:12 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* docs/NEWS-1.9.3: moved from NEWS.
+
+	* docs/ChangeLog-1.9.3: merged ChangeLog for 1.9.3.
+
+	* NEWS: NEWS for 1.9.4 that describes changes since 1.9.3
+
+	* ChangeLog: new ChangeLog for 1.9.4.
+
+Sun Jul 10 23:30:52 2011  Yuki Sonoda (Yugui)  <yugui@yugui.jp>
+
+	* version.h (RUBY_VERSION): ruby_1_9_3 branch was forked.
+
+For the changes before 1.9.3, see doc/ChangeLog-1.9.3
+For the changes before 1.8.0, see doc/ChangeLog-1.8.0
+
+Local variables:
+coding: us-ascii
+add-log-time-format: (lambda ()
+  (let* ((time (current-time))
+	 (system-time-locale "C")
+	 (diff (+ (cadr time) 32400))
+	 (lo (% diff 65536))
+	 (hi (+ (car time) (/ diff 65536))))
+  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
+indent-tabs-mode: t
+tab-width: 8
+change-log-indent-text: 2
+end:
+vim: tabstop=8 shiftwidth=2
diff --git a/doc/ChangeLog-2.1.0 b/doc/ChangeLog-2.1.0
new file mode 100644
index 0000000000..76edfd3ce7
--- /dev/null
+++ b/doc/ChangeLog-2.1.0
@@ -0,0 +1,18060 @@
+Fri Dec 20 17:52:50 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c: check definition of
+	  GLOBAL_METHOD_CACHE_SIZE and GLOBAL_METHOD_CACHE_MASK.
+
+Fri Dec 20 17:03:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: rename OBJ_WRITE and OBJ_WRITTEN into
+	  RB_OBJ_WRITE and RB_OBJ_WRITTEN.
+
+	* array.c, class.c, compile.c, hash.c, internal.h, iseq.c,
+	  proc.c, process.c, re.c, string.c, variable.c, vm.c,
+	  vm_eval.c, vm_insnhelper.c, vm_insnhelper.h,
+	  vm_method.c: catch up this change.
+
+Fri Dec 20 16:01:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: add a comment for WB interfaces.
+
+Fri Dec 20 16:00:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: DLDFLAGS is defined in --with-opt-dir handler, so
+	  ${DLDFLAGS=} does not work now.  use RUBY_APPEND_OPTIONS instead.
+	  [ruby-dev:47855] [Bug #9256]
+
+Fri Dec 20 14:19:12 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (AC_ARG_WITH): use withval directly.
+	  fix failure on FreeBSD.
+	  http://fb32.rubyci.org/~chkbuild/ruby-trunk/log/20131217T070301Z.diff.html.gz
+
+Fri Dec 20 14:00:01 2013  Aman Gupta <ruby@tmm1.net>
+
+	* include/ruby/ruby.h (struct RClass): add super, remove iv_index_tbl.
+	  since RCLASS_SUPER() is commonly used inside while loops, we move it
+	  back inside struct RClass to improve cache hits. this provides a
+	  small improvement (1%) in hotspots like rb_obj_is_kind_of()
+	* internal.h (struct rb_classext_struct): remove super, add
+	  iv_index_table
+	* internal.h (RCLASS_SUPER): update for new location
+	* internal.h (RCLASS_SET_SUPER): ditto
+	* internal.h (RCLASS_IV_INDEX_TBL): ditto
+	* object.c (rb_class_get_superclass): ditto
+	* include/ruby/backward/classext.h (RCLASS_SUPER): ditto
+
+Fri Dec 20 07:07:35 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 03d6ae7.  Changes include:
+
+	  * Fixed typos.
+
+	  * Relaxed Gem.ruby test for ruby packagers that do not use `ruby`.
+
+	* test/rubygems:  ditto.
+
+Thu Dec 19 14:03:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_get_freeobj): improve hot path performance.
+
+	* gc.c (heap_get_freeobj_from_next_freepage): replace with
+	  heap_get_freepage(). It returns freeobj instead of freepage.
+	  This is not on hot path.
+
+Thu Dec 19 12:05:17 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master af60443.  Changes include:
+
+	  * Improved speed of `gem install --ignore-dependencies`.
+
+	  * Open read-write for exclusive flock.  [ruby-trunk - Bug #9257]
+
+	  * Remove specification before install to prevent infinite loop.
+
+Thu Dec 19 11:23:49 2013  Aman Gupta <ruby@tmm1.net>
+
+	* vm_insnhelper.c (vm_call_iseq_setup_normal): simple for loop
+	  condition optimization. this area shows up as a hotspot in VM
+	  profiles.
+
+Thu Dec 19 10:50:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (newobj_of): don't need to RBASIC_SET_CLASS() which includes WB
+	  here because created obj is always YOUNG/INFANT.
+
+Thu Dec 19 10:48:37 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: check GC::OPTS availability
+	  for not MRI 2.1.0.
+
+Thu Dec 19 03:10:30 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (heap_get_freeobj): remove redundant assignment. heap->freelist
+	  is set after the while() loop already.
+
+Thu Dec 19 01:54:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/runner.rb: fix commit miss on r44278.
+
+Thu Dec 19 00:26:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect_body): lazy_sweep setting should work
+	  without USE_RGENGC.
+
+Wed Dec 18 23:31:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_dump_major_reason): fix this function because major_reason
+	  can be OR of multiple reasons.
+
+	* gc.c (gc_profile_dump_on): ditto.
+
+Wed Dec 18 17:03:00 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_record_get): should return an empty array
+	  when profiling is active.
+
+Wed Dec 18 16:49:40 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_clear, gc_profile_enable): remove rest_sweep().
+
+	* gc.c: check objspace->profile.current_record before inserting
+	  profiling record by new macro gc_prof_enabled().
+
+Wed Dec 18 14:32:06 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_exec.h (VM_DEBUG_STACKOVERFLOW): added.
+	  disable stack overflow check for every stack pushing as default.
+
+	* vm_exec.c (vm_stack_overflow_for_insn): ditto.
+
+Wed Dec 18 10:00:22 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master d8f12e2.  This increases the
+	  speed of `gem install --ignore-dependencies` which helps bundler
+	  tests.
+	* test/rubygems:  ditto.
+
+Wed Dec 18 09:00:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb (test_expand_heap): allow +/-1 diff.
+
+Tue Dec 17 23:44:15 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/ruby/test_io.rb: fix duplicated test name.
+
+Tue Dec 17 20:15:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_reject): revert to deprecated behavior, with
+	  warnings, due to compatibility for HashWithDifferentAccess.
+	  [ruby-core:59154] [Bug #9223]
+
+Tue Dec 17 17:30:56 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Import version 2.1.1 from
+	  https://github.com/knu/ruby-electric.el.
+
+	  * ruby-electric-delete-backward-char: Enable support for number
+	    prefix.
+
+	  * ruby-electric-curlies: Fix electric operation after an open
+	    curly.
+
+Tue Dec 17 16:19:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_trace.c (rb_postponed_job_flush): isolate exceptions in
+	  postponed jobs and restore outer ones.  based on a patch by
+	  tarui.  [ruby-core:58652] [Bug #9168]
+
+Tue Dec 17 10:48:04 2013  Aman Gupta <ruby@tmm1.net>
+
+	* configure.in (RUBY_DTRACE_POSTPROCESS): Fix compatibility with
+	  systemtap on linux. stap requires `dtrace -G` post-processing, but
+	  the dtrace compatibility wrapper is very strict about probes.d
+	  syntax.
+
+Tue Dec 17 05:18:17 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 1c5f4b3.  Allows rubygems
+	  repackagers to disable backward-compatible shared gem directory
+	  behavior.
+	* test/rubygems:  ditto.
+
+Tue Dec 17 05:14:35 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS (RDoc):  Update version number so I don't have to change it
+	  for the final release.
+
+Mon Dec 16 19:19:19 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_markable_object_p): should check special_const_p
+	  first (by is_markable_object()).
+
+Mon Dec 16 19:12:54 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (reachable_object_from_root_i): use
+	  compare_by_identity hash to avoid hash modify problem
+	  during iteration.
+	  [Bug #9252]
+
+	* ext/objspace/objspace.c (reachable_objects_from_root): ditto.
+
+Mon Dec 16 18:16:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_verify_internal_consistency): should not use
+	  rb_objspace_each_objects() because it call rest_sweep().
+
+Mon Dec 16 18:07:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_markable_object_p): fix last commit (build error).
+
+Mon Dec 16 18:04:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_markable_object_p): it should be live objects.
+
+Mon Dec 16 18:00:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_each_objects): should not clear dont_lazy_sweep
+	  flag in nested case.
+
+Mon Dec 16 16:40:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_method_entry_make): fix WB miss.
+	  Note that rb_method_entry_t::klass is not constified.
+	  We may  constify this field.
+
+	* test/ruby/test_alias.rb: add a test.
+
+Mon Dec 16 14:14:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: use gc_verify_internal_consistency() instead of
+	  gc_check_before_marks_i() for check consistency
+	  on RGENGC_CHECK_MODE >= 2.
+
+Mon Dec 16 14:01:48 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c (make_clock_result): add :second as a unit for
+	  Process.clock_gettime.
+
+Mon Dec 16 13:10:54 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: introduce GC.verify_internal_consistency method to verify GC
+	  internal data structure.
+
+	  Now this method only checks generation (old/young) consistency.
+
+Mon Dec 16 11:49:26 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (gc_info_decode): Fix build errors when compiled with
+	  RGENGC_ESTIMATE_OLDMALLOC=0
+	* gc.c (objspace_malloc_increase): ditto
+
+Sun Dec 15 13:38:29 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (reachable_object_from_root_i):
+	  reachable objects should not include categories and
+	  category_objects because it is noisy information.
+
+	  In fact, objects created after calling
+	  ObjectSpace.reachable_objects_from_root should not be included
+	  as a returning hash objects. Currently, mswin64 platform has a
+	  problem because of this behavior. Should we trace new objects?
+
+Sun Dec 15 07:09:28 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc master 263a9e5.  This improves the
+	  accessibility of the search box.
+
+Sat Dec 14 17:39:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_arg_complex): count post
+	  arguments as mandatory arguments.  [ruby-core:57706] [Bug #8993]
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): ditto.
+
+Sat Dec 14 16:26:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rubylibprefix): replace exec_prefix as well as
+	  bindir and libdir.  a patch by kimuraw (Wataru Kimura) at
+	  [ruby-dev:47852].  [Bug #9160]
+
+Sat Dec 14 14:42:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/logger.rb (lock_shift_log): no need to rotate the log file
+	  if it has been rotated by another process.  based on the patch
+	  by no6v (Nobuhiro IMAI) in [ruby-core:58620].  [Bug #9133]
+
+Sat Dec 14 13:01:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew_from_me): method by respond_to_missing? should be
+	  owned by the original class.
+
+Sat Dec 14 11:55:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/scanf.rb (IO#scanf): fix mistaken use of rescue modifier.
+	  a patch by Mon_Ouie at [ruby-core:52813].  [Bug #7940]
+
+Sat Dec 14 11:44:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_qsort): fix potential stack overflow on a large
+	  machine.  based on the patch by Conrad Irwin <conrad.irwin AT
+	  gmail.com> at [ruby-core:51816].  [Bug #7772]
+
+Sat Dec 14 11:25:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_defined): support nested class path as
+	  well as const_get.  [Feature #7414]
+
+Sat Dec 14 01:31:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_rescue2): reuse tags pushed for body proc to protect
+	  rescue proc too.
+
+Sat Dec 14 01:15:51 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (wmap_final_func): Bugfix. Should update *value to new pointer.
+
+Sat Dec 14 01:05:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb: Don't test $! in "ensure" clause because
+	  it may be set before the body.
+	  Reported by ko1 and mrkn.  [ruby-core:59088] [Bug #9247]
+
+	* lib/cgi/core.rb: Ditto.
+
+	* lib/drb/ssl.rb: Ditto.
+
+Sat Dec 14 00:34:31 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* internal.h (ruby_sized_xrealloc2): fix typo introduced in r44117,
+	  which cause compile error on Solaris.
+
+Sat Dec 14 00:22:16 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c: (exec_recursive): use rb_catch_protect() instead of
+	  rb_catch_obj() and PUSH_TAG(), and reduce pushing tags and
+	  machine stack usage.
+
+Sat Dec 14 00:18:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew_from_me): achieve the original defined_class from
+	  prepended iclass, to fix inherited owner.
+
+	* proc.c (method_owner): return the defined class, but not the
+	  class which the method object is created from.
+
+Fri Dec 13 22:29:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (method_owner): return the class where alias is defined, not
+	  the class original method is defined.
+
+	* vm_method.c (rb_method_entry_make, rb_alias): store the originally
+	  defined class in me.  [Bug #7993] [Bug #7842] [Bug #9236]
+
+	* vm_method.c (rb_method_entry_get_without_cache): cache included
+	  module but not iclass.
+
+Fri Dec 13 16:27:17 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (gc_info_decode): Use :major_by=>:nofree as fallback reason
+	  when other trigger conditions are present.
+
+Fri Dec 13 13:25:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c: add Exception#backtrace_locations.
+	  Now, there are no setter and independent from Exception#backtrace.
+	  [Feature #8960]
+
+	* eval.c (setup_exception): set backtrace locations for `bt_location'
+	  special attribute.
+
+	* vm_backtrace.c (rb_backtrace_to_location_ary): added.
+
+	* internal.h: ditto.
+
+	* test/ruby/test_backtrace.rb: add a test for
+	  Exception#backtrace_locations.
+
+Fri Dec 13 12:01:07 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect_body): use rb_bug() and explicit error message
+	  instead of using assert().
+	  [Bug #9222]
+
+Fri Dec 13 11:52:41 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c: fix comment to remove the word "shady".
+
+	* variable.c: ditto.
+
+Fri Dec 13 11:33:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename *shady* func/macros.
+	  * RVALUE_RAW_SHADY()   -> RVALUE_WB_PROTECTED_RAW()
+	  * RVALUE_SHADY()       -> RVALUE_RAW_SHADY()
+	  * rgengc_check_shady() -> rgengc_check_relation().
+	  And fix some messages using "shady" to "non-WB-protected".
+
+Fri Dec 13 10:04:23 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/request_set/lockfile.rb:  Import RubyGems master a8d0669
+	  with a 1.8.7 compatibility fix.
+	* test/rubygems/test_gem_request_set_lockfile.rb:  ditto.
+
+Fri Dec 13 09:50:49 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master ddac51f.  Changes:
+
+	  * Allow override for the shared gem installation directory for
+	    rubygems packagers.
+
+	  * Lock gem cache files for read and write to improve thread safety.
+
+	  * Use io/console when available.
+
+	  * Minor cleanup.
+
+	* test/rubygems:  ditto.
+
+Fri Dec 13 08:15:31 2013  Aman Gupta <ruby@tmm1.net>
+
+	* class.c (include_modules_at): use RCLASS_M_TBL_WRAPPER for
+	  equality checks. this avoids an unnecessary deference inside a tight
+	  loop, fixing a performance regression from r43973.
+	* object.c (rb_obj_is_kind_of): ditto.
+	* object.c (rb_class_inherited_p): ditto.
+
+Wed Dec 13 02:00:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpSetPTR): fix for limitation of the resulting
+	  precision.
+	  [ruby-core:50269] [Bug #7458]
+
+	* test/bigdecimal/test_bigdecimal.rb (test_limit): add tests for the above
+	  change.
+
+Wed Dec 13 01:56:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpAddAbs): put out a conditional branch from
+	  the inside of while-loop.
+
+	* ext/bigdecimal/bigdecimal.c (VpSubAbs): ditto.
+
+Wed Dec 13 01:53:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VPrint): be a static function, support another
+	  dump formats, and add more information of the given bigdecimal.
+
+	* ext/bigdecimal/bigdecimal.h: ditto.
+
+Wed Dec 11 16:45:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_raise_jump): call c_return hook immediately after
+	  popping `raise' frame.
+	  Patches by deivid (David Rodriguez). [Bug #8886]
+
+	* test/ruby/test_settracefunc.rb: catch up this fix.
+
+Wed Dec 11 16:01:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_reject): return a plain hash, without copying
+	  the class, default value, instance variables, and taintedness.
+	  they had been copied just by accident.
+	  [ruby-core:59045] [Bug #9223]
+
+Wed Dec 11 15:36:15 2013  Aman Gupta <ruby@tmm1.net>
+
+	* compile.c (iseq_specialized_instruction): emit opt_aset instruction
+	  to optimize Hash#[]= and Array#[]= when called with Fixnum argument.
+	  [Bug #9227] [ruby-core:58956]
+
+Wed Dec 11 04:54:03 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master ec8ed22.  Notable changes
+	  include:
+
+	  * Renamed extension_install_dir to extension_dir (backwards
+	    compatible).
+
+	  * Fixed creation of gem.deps.rb.lock file from
+	    TestGemRequestSet#test_install_from_gemdeps_install_dir
+
+	  * Fixed a typo and some documentation.
+
+	* test/rubygems:  ditto.
+
+Wed Dec 11 03:18:08 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* insns.def: Fix optimization bug of Float#/ [Bug #9238]
+
+Tue Dec 10 23:58:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_strptime.c (date__strptime_internal): unset
+	  case-insensitive flag for [:alpha:], which already implies both
+	  cases, to get rid of backtrack explosion.  [ruby-core:58984]
+	  [Bug #9221]
+
+Tue Dec 10 23:44:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_hash): add salt to differentiate false and empty
+	  array.  [ruby-core:58993] [Bug #9231]
+
+	* hash.c (rb_any_hash, rb_hash_hash): ditto.
+
+Tue Dec 10 18:16:09 2013  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* man/ruby.1: [DOC] Use www.ruby-toolbox.com instead of RAA.
+
+Tue Dec 10 17:21:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (wmap_finalize, wmap_aset_update): use simple malloced array
+	  instead of T_ARRAY, to reduce GC pressure.
+
+Tue Dec 10 15:56:48 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (reflist_add): revert changes from r44109. it is unnecessary
+	  after r44113
+	* gc.c (allrefs_i): fix whitespace
+	* gc.c (allrefs_roots_i): fix whitespace
+
+Tue Dec 10 15:46:03 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (allrefs_add): push obj only if allrefs table doesn't have
+	  obj.
+
+	* gc.c (allrefs_roots_i): ditto.
+
+Tue Dec 10 15:28:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RGENGC_CHECK_MODE): separate checkers to different modes.
+	  * 2: enable generational bits check (for debugging)
+	  * 3: enable livness check
+	  * 4: show all references
+
+Tue Dec 10 15:15:37 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_marks_check): disable GC during checking and
+	  restore malloc_increase info.
+
+Tue Dec 10 14:41:53 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (reflist_add): return 0 if reference already exists
+	* gc.c (allrefs_add): return 1 on newly added references
+	* gc.c (allrefs_i): follow references to construct complete object
+	  graph. before this patch, RGENGC_CHECK could fail to verify some WB
+	  miss issues. [Bug #9226] [ruby-core:58959]
+
+Tue Dec 10 11:20:56 2013  Aman Gupta <ruby@tmm1.net>
+
+	* ext/objspace/objspace_dump.c (dump_object): include fstring flag on
+	  strings. include gc flags (old, remembered, wb_protected) on all objects.
+	* ext/objspace/objspace_dump.c (Init_objspace_dump): initialize lazy
+	  IDs before first use.
+	* gc.c (rb_obj_gc_flags): new function to retrieve object flags
+	* internal.h (RB_OBJ_GC_FLAGS_MAX): maximum flags allowed for one obj
+	* test/objspace/test_objspace.rb (test_dump_flags): test for above
+	* test/objspace/test_objspace.rb (test_trace_object_allocations):
+	  resolve name before dump (for rb_class_path_cached)
+
+Tue Dec 10 07:48:29 2013  Aman Gupta <ruby@tmm1.net>
+
+	* vm_method.c (rb_clear_method_cache_by_class): fire
+	  ruby::method-cache-clear probe on global or klass-level method cache
+	  clear [Bug #9190]
+	* probes.d (provider ruby): new dtrace probe
+	* doc/dtrace_probes.rdoc: docs for new probe
+	* test/dtrace/test_method_cache.rb: test for new probe
+
+Tue Dec 10 06:14:11 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/.document:  Remove curses from documentable directories.
+
+Tue Dec 10 04:55:36 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/lib/openssl/digest.rb: Deprecate OpenSSL::Digest::Digest
+	  [Fixes GH-446] https://github.com/ruby/ruby/pull/446
+
+Tue Dec 10 00:41:42 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/thread/thread.c: [DOC] add call-seq alias for Queue#enq, #<<, etc.
+
+	* ext/thread/thread.c (Init_thread): use rb_define_alias instead of
+	  rb_alias to document alias.
+
+Mon Dec  9 20:00:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* internal.h (RCLASS_SERIAL): Add RCLASS_SERIAL as a convenience
+	  accessor for RCLASS_EXT(klass)->class_serial.
+
+	* class.c, vm_insnhelper.c, vm_method.c: Use RCLASS_SERIAL
+
+Mon Dec  9 19:50:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* compile.c, insns.def, test/ruby/test_rubyvm.rb, vm.c, vm_core.h,
+	  vm_insnhelper.c, vm_insnhelper.h, vm_method.c: Rename method_serial
+	  to global_method_state and constant_serial to global_constant_state
+	  after discussion with ko1.
+
+Mon Dec  9 18:50:43 2013  Aman Gupta <ruby@tmm1.net>
+
+	* hash.c (rb_hash_replace): fix segv on `{}.replace({})` introduced
+	  in r44060 [Bug #9230] [ruby-core:58991]
+	* test/ruby/test_hash.rb: regression test for above
+
+Mon Dec  9 18:10:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_stat): renamed from ruby_vm_stat.
+	  Should not use ruby_ prefix here.
+
+Mon Dec  9 16:13:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (wmap_size): add ObjectSpace::WeakMap#size and #length.
+
+Mon Dec  9 15:26:17 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/test_curses.rb: removed.
+
+Mon Dec  9 13:36:55 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/curses, sample/curses: removed curses.
+
+	* NEWS: added an entry for the above change.
+
+Mon Dec  9 12:26:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/objspace/object_tracing.c (newobj_i): use cached class path
+	  only to get rid object allocation during NEWOBJ hook.
+	  [ruby-core:58853] [Bug #9212]
+
+	* variable.c (rb_class_path_cached): returns cached class path
+	  only, without searching and allocating new class path string.
+
+Mon Dec  9 11:14:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_parse.c (parse_time): unset case-insensitive flag
+	  for [:alpha:], which already implies both cases, to get rid of
+	  backtrack explosion.  [ruby-core:58876] [Bug #9221]
+
+Mon Dec  9 08:40:40 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master bf37240.  Fixes useless
+	  error message with `gem install -g` with no gem dependencies file.
+	* test/rubygems:  ditto.
+
+Mon Dec  9 04:52:25 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS:  Update RubyGems entry with notable features.
+
+Mon Dec  9 04:43:54 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/.document:  Add syslog/lib and thread/thread.c to documentable
+	  items.  [ruby-trunk - Bug #9228]
+
+Mon Dec  9 04:28:50 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 096db36.  Changes include
+	  support for PATH in Gemfile.lock and a typo fix from Akira Matsuda.
+	* test/rubygems:  ditto.
+
+Mon Dec  9 02:10:32 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/responses.rb:
+	  Add `HTTPIMUsed`, as it is also supported by rack/rails.
+	  RFC - http://tools.ietf.org/html/rfc3229
+	  by Vipul A M <vipulnsward@gmail.com>
+	  https://github.com/ruby/ruby/pull/447 fix GH-447
+
+Sun Dec  8 20:47:35 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_get_kwargs): when values is non-null, remove
+	  extracted keywords from the rest keyword argument.
+
+Sun Dec  8 20:26:54 2013  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* common.mk (ruby.imp): avoid circular dependency on AIX
+
+Sun Dec  8 20:21:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bigdecimal.c (BigDecimal_coerce): convert a Float to a BigDecimal instead
+	  of converting the receiver to a Float.  The reason is there are BigDecimal
+	  instances with precisions that is smaller than the Float's precision.
+	  [ruby-core:58756] [Bug #9192]
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests for the above change.
+
+Sun Dec  8 18:28:20 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* NEWS: [DOC] update NEWS about GC.
+
+Sun Dec  8 17:52:24 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* object.c: [DOC] document Module#singleton_class?.
+
+Sun Dec  8 16:19:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_get_kwargs): if optional is negative, unknown
+	  keywords are allowed.
+
+	* vm_insnhelper.c (vm_callee_setup_keyword_arg): check unknown
+	  keywords.
+
+Sun Dec  8 14:55:12 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* array.c (rb_ary_shuffle_bang, rb_ary_sample): rename local variables.
+
+Sun Dec  8 13:59:38 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* array.c (rb_ary_shuffle_bang, rb_ary_sample): check
+	  unknown keywords.
+
+	* test/ruby/test_array.rb (test_shuffle, test_sample): tests for
+	  the above.
+
+Sun Dec  8 13:01:11 2013  Aman Gupta <ruby@tmm1.net>
+
+	* vm.c (ruby_vm_stat): add RubyVM.stat() for access to internal cache
+	  counters. this methods behaves like GC.stat, accepting an optional
+	  hash or symbol argument. [Bug #9190] [ruby-core:58750]
+	* test/ruby/test_rubyvm.rb: test for new method
+
+Sun Dec  8 11:59:40 2013  Aman Gupta <ruby@tmm1.net>
+
+	* hash.c (rb_hash_replace): add a write barrier to fix GC mark miss on
+	  hashes using Hash#replace [Bug #9226] [ruby-core:58948]
+
+Sun Dec  8 11:21:00 2013  Aman Gupta <ruby@tmm1.net>
+
+	* include/ruby/ruby.h: add RGENGC_WB_PROTECTED_NODE_CREF setting
+	  In a large app, this reduces the size of
+	  remembered_shady_object_count by 80%. [Bug #9225] [ruby-core:58947]
+	* gc.c (rb_node_newnode): add FL_WB_PROTECTED flag to NODE_CREF
+	* class.c (rewrite_cref_stack): insert OBJ_WRITE for NODE_CREF
+	* iseq.c (set_relation): ditto
+	* iseq.c (rb_iseq_clone): ditto
+	* vm_eval.c (rb_yield_refine_block): ditto
+	* vm_insnhelper.c (vm_cref_push): ditto
+	* vm_insnhelper.h (COPY_CREF): ditto
+
+Sun Dec  8 10:45:05 2013  Aman Gupta <ruby@tmm1.net>
+
+	* hash.c (hash_aset_str): revert r43870 due to performance issue
+	  [Bug #9188] [ruby-core:58730]
+	* parse.y (assoc): convert literal string hash keys to fstrings
+	* test/ruby/test_hash.rb (class TestHash): expand test
+
+Sun Dec  8 10:22:38 2013  Aman Gupta <ruby@tmm1.net>
+
+	* parse.y (register_symid_str): use fstrings in symbol table
+	  [Bug #9171] [ruby-core:58656]
+	* parse.y (rb_id2str): ditto
+	* string.c (rb_fstring): create frozen_strings on first usage. this
+	  allows rb_fstring() calls from the parser (before cString is created)
+	* string.c (fstring_set_class_i): set klass on fstrings generated
+	  before cString was defined
+	* string.c (Init_String): convert frozen_strings table to String
+	  objects after boot
+	* ext/-test-/symbol/type.c (bug_sym_id2str): expose rb_id2str()
+	* test/-ext-/symbol/test_type.rb (module Test_Symbol): verify symbol
+	  table entries are fstrings
+
+Sun Dec  8 10:24:20 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Update version for upcoming ruby 2.1.0 RC.
+
+Sun Dec  8 10:21:36 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 14749ce.  This fixes bugs
+	  handling of gem dependencies lockfiles (Gemfile.lock).
+
+	* test/rubygems:  ditto.
+
+Sun Dec  8 09:40:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* array.c (rb_ary_or): use RHASH_TBL_RAW instead of RHASH_TBL
+
+	* process.c (rb_execarg_fixup): use RHASH_TBL_RAW and insert write
+	  barriers where appropriate
+
+	* vm.c (kwmerge_i): use RHASH_TBL_RAW
+
+	* vm.c (HASH_ASET): use rb_hash_aset instead of calling directly into
+	  st_insert
+
+Sat Dec  7 11:15:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_reject): copy unrejected elements only to new hash,
+	  so that the change on the original receiver can affect.
+	  [ruby-core:58914] [Bug #9223]
+
+Sat Dec  7 08:25:00 2013  Richo Healey  <richo@psych0tik.net>
+
+	* test/ruby/test_struct.rb: Add regression test for question marks and
+	  bangs in struct members. [Closes GH-468]
+
+Fri Dec  6 19:33:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_extract_keywords, rb_get_kwargs): move from
+	  vm_insnhelper.c.
+
+Fri Dec  6 19:18:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change oldmalloc meaning.
+	  Increase oldmalloc_increase with malloc_increase
+	  instead of using obj_memsize_of().
+
+	  This change will avoid the danger of memory full without major GC.
+
+Fri Dec  6 19:08:48 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (atomic_sub_nounderflow): not 0 but val itself.
+
+Fri Dec  6 18:37:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_alloc, Init_heap): initialize
+	  oldmalloc_increase_limit at Init_heap.
+
+	  rb_objspace_alloc() is not called on some platforms.
+
+Fri Dec  6 18:33:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect_body): bug fix.
+	  initialize after recording.
+
+Fri Dec  6 17:49:46 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (atomic_sub_nounderflow): added to simplify atomic sub with
+	  care about underflow.
+
+	* gc.c (objspace_malloc_increase): use it.
+
+Fri Dec  6 17:10:44 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (rb_get_kwargs): get keyword argument values from an
+	  option hash, not only checking keys.
+
+	* dir.c (dir_initialize): use rb_get_kwargs.
+
+	* gc.c (gc_start_internal): ditto.
+
+Fri Dec  6 16:47:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-mode.el (ruby-brace-to-do-end): split single line block.
+
+	* misc/ruby-mode.el (ruby-do-end-to-brace): shrink single line block
+	  to one line.
+
+Fri Dec  6 16:16:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_start_internal): do not use rb_gc_start() and rb_gc().
+
+Fri Dec  6 15:24:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_start_internal, rb_gc): do not need
+	  heap_pages_free_unused_pages() here.
+	  It was done in after_sweep().
+
+	* gc.c (rb_gc): The reason is now GPR_FLAG_CAPI.
+
+Fri Dec  6 14:05:19 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (gc_start_internal): GC.start() now accepts two optional
+	  keyword arguments. These can be used to disable full_mark (minor
+	  mark only) or disable immediate_sweep (use lazy sweep). These new
+	  options are useful for benchmarking GC behavior, or performing minor
+	  GC out-of-band.
+	* test/ruby/test_gc.rb (class TestGc): tests for new options.
+
+Fri Dec  6 11:51:28 2013  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/erb.rb: [DOC] fix broken link, Use rubygems.org and www.ruby-toolbox.com instead of RAA.
+	  [Bug #9197]
+
+Fri Dec  6 10:50:54 2013  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/webrick/httprequest.rb: [DOC] Fix broken link of CGI specification by @udzura [fix GH-466]
+
+Thu Dec  6 01:27:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValueWithPrec):
+	  treat 0.0 and -0.0 of floating-point numbers specially for an optimization
+	  and to correctly propagate its signbit to the result.
+	  [Bug #9214] [ruby-core:58858]
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests case for the above change.
+
+	* test/bigdecimal/test_bigdecimal_util.rb: ditto.
+
+Thu Dec  5 22:18:01 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): strip destdir part from prefix to get
+	  rid of duplication.  a patch by arton at [ruby-core:58859].
+	  [ruby-core:58856] [Bug #9213]
+
+Thu Dec  5 21:53:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_or): lhs elements are preferred, so should not
+	  replace with rhs elements.
+
+	* test/ruby/test_array.rb (test_OR_in_order): import the test failed
+	  by r43969 from rubyspec/core/array/union_spec.rb.
+
+Thu Dec  5 21:05:42 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_info_decode): fix to avoid syntax error on VS2012.
+
+Thu Dec  5 19:35:35 2013  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* st.c: tweaked comment
+
+Thu Dec  5 19:21:10 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (struct rb_objspace): rename internal last_collection_flags to
+	  latest_gc_info
+	* gc.c (gc_latest_collection_info): add GC.latest_gc_info() with similar
+	  behavior to GC.stat()
+	* gc.c (rb_gc_latest_gc_info): new c-api for above
+	* gc.c (gc_stat_internal): remove :last_collection_flags from GC.stat
+	* gc.c (gc_profile_decode_flags): remove GC::Profiler.decode_flags
+	* include/ruby/intern.h (rb_gc_latest_gc_info): export new c-api
+	* test/ruby/test_gc.rb (class TestGc): test for new behavior
+	* NEWS: note about new api
+
+	* gc.c (gc_stat_internal): raise TypeError on wrong type
+	* gc.c (gc_stat): fix error message
+
+Thu Dec  5 18:18:08 2013  Aman Gupta <ruby@tmm1.net>
+
+	* ext/objspace/gc_hook.c: remove this file
+	* ext/-test-/tracepoint/gc_hook.c: new filename for above
+	* ext/objspace/objspace.c: remove ObjectSpace.after_gc_start_hook=
+	* test/objspace/test_objspace.rb: remove test
+	* test/-ext-/tracepoint/test_tracepoint.rb: add above test for
+	  tracepoint re-entry
+
+Thu Dec  5 17:44:53 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change function names vm_ prefix to objspace_ prefix.
+	  They are objspace_ functionality.
+
+Thu Dec  5 16:11:04 2013  Aman Gupta <ruby@tmm1.net>
+
+	* include/ruby/intern.h: add rb_gc_stat() for access to GC.stat
+	  variables from c-api
+	* gc.c (rb_gc_stat): new c-api method. accepts either VALUE hash like
+	  GC.stat, or VALUE symbol key and returns size_t directly. the second
+	  form is useful to avoid allocations, i.e. for usage inside
+	  INTERNAL_EVENT_GC tracepoints.
+	* gc.c (gc_stat): add GC.stat(:key) to return single value instead of hash
+	* gc.c (gc_stat_internal): helper method to retrieve single or all stat values
+	* test/ruby/test_gc.rb (class TestGc): test for new behavior
+	* NEWS: note about this new api
+
+Thu Dec  5 14:40:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash): revert r43981 and bail out to the outermost frame
+	  when recursion is detected.
+
+Thu Dec  5 13:47:15 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (vm_malloc_size): added.
+	  return malloc_usable_size() if possible.
+
+	* gc.c (MALLOC_ALLOCATED_SIZE): add new setting macro to enable
+	  GC.allocated_size.
+	  If platform supports `malloc_usable_size()' (or similar one),
+	  GC.allocated_size can be implemented with this function.
+	  Default is 0.
+
+	* gc.c (vm_xmalloc, vm_xrealloc, vm_xfree): use vm_malloc_size()
+	  to detect collect allocated size.
+
+	* gc.c (vm_malloc_increase): refactoring.
+
+Thu Dec  5 13:19:03 2013  Aman Gupta <ruby@tmm1.net>
+
+	* include/ruby/ruby.h: remove INTERNAL_EVENT_GC_END and replace with
+	  two new events: GC_END_MARK and GC_END_SWEEP
+	* gc.c (gc_after_sweep): emit GC_END_SWEEP after lazy sweep is done
+	* gc.c (gc_marks_body): emit GC_END_MARK at end of minor/major mark
+	* ext/-test-/tracepoint/tracepoint.c (struct tracepoint_track): tests
+	  for new events.
+	* test/-ext-/tracepoint/test_tracepoint.rb (class TestTracepointObj):
+	  ditto.
+	* NEWS: remove ObjectSpace.after_gc_*_hook. These are only a sample,
+	  and will be removed before ruby 2.1.
+	* ext/objspace/gc_hook.c: remove ObjectSpace.after_gc_end_hook=
+
+Thu Dec  5 10:47:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby_atomic.h (ATOMIC_PTR_EXCHANGE): atomic exchange function for
+	  a generic pointer.
+
+Thu Dec  5 10:47:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (finalize_deferred): flush all deferred finalizers while other
+	  finalizers can get ready to run newly by lazy sweep.
+	  [ruby-core:58833] [Bug #9205]
+
+Thu Dec  5 09:07:59 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (ruby_gc_set_params): Accept safe_level argument so GC tuning
+	  settings can be applied before rb_safe_level() is available.
+	* internal.h (rb_gc_set_params): ditto.
+	* ruby.c (process_options): Apply GC tuning early during boot process
+	  so boot-time allocations can benefit. This also benefits any code
+	  loaded in via `ruby -r`.
+
+Wed Dec  4 13:02:13 2013  Aman Gupta <ruby@tmm1.net>
+
+	* vm_trace.c (rb_suppress_tracing): Fix initialization of stack
+	  allocated rb_trace_arg_t structure. Without this patch, sometimes
+	  INTERNAL_EVENT_GC would be skipped accidentally inside
+	  rb_threadptr_exec_event_hooks_orig().
+
+Wed Dec  4 12:57:24 2013  Aman Gupta <ruby@tmm1.net>
+
+	* string.c (fstr_update_callback): Improve implementation in r43968
+	  based on feedback from @nagachika. In the existing case, we can
+	  return ST_STOP to prevent any hash modification. In the !existing
+	  case, set both key and value to the fstr.
+
+Wed Dec  4 12:47:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#method_missing): ignore the target if not
+	  set, and delegate to global methods.  [ruby-core:58572] [Bug #9155]
+
+	* lib/delegate.rb (Delegator#respond_to_missing): ditto.
+
+	* lib/delegate.rb (SimpleDelegator#__getobj__): yield and return if
+	  not delegated but a block is given, like as Hash#fetch.
+
+	* lib/delegate.rb (DelegateClass#__getobj__): ditto.
+
+Tue Dec  3 23:48:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check malloc_size() availability.
+
+	* gc.c: use malloc_size() with malloc/malloc.h if available.
+
+Tue Dec  3 23:06:20 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* object.c (rb_obj_clone): don't copy FL_WB_PROTECTED of a
+	  original object.
+
+Tue Dec  3 22:32:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_recursive): make similar (recursive) constructs
+	  return same hash value.  execute recursively, and rewind to the
+	  topmost frame with an object which .eql? to the recursive
+	  object, if recursion is detected.
+
+	* hash.c (rb_hash): detect recursion for all `hash' methods.  each
+	  `hash' methods no longer need to use rb_exec_recursive().
+
+Tue Dec  3 21:53:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_catch_protect): new function similar to
+	  rb_catch_obj(), but protect from all global jumps like as
+	  rb_load_protect(), rb_protect(), etc.
+
+Tue Dec  3 20:18:46 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* object.c (rb_obj_clone): Protect FL_PROMOTED and FL_WB_PROTECTED
+	  flags of a destination object.
+
+Tue Dec  3 20:16:38 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_hash_rehash): use hash_alloc() instead of rb_hash_new(),
+	  to hide temporary object from ObjectSpace.  [Bug #9187]
+
+Tue Dec  3 17:11:47 2013  Aman Gupta <ruby@tmm1.net>
+
+	* load.c (features_index_add_single): Move loaded_features_index array values off
+	  the ruby heap. [Bug #9201] [ruby-core:58805]
+	* load.c (loaded_features_index_clear_i): Clean up off-heap array structure.
+	* vm.c (rb_vm_mark): Remove unnecessary mark_tbl for loaded_features_index.
+	  This improves minor GC time by 15% in a large application.
+
+Tue Dec  3 17:01:45 2013  Aman Gupta <ruby@tmm1.net>
+
+	* include/ruby/ruby.h (struct RClass): Add wrapper struct around
+	  RClass->m_tbl with serial. This prevents double marking method
+	  tables, since many classes/modules can share the same method table.
+	  This improves minor mark time in a large application by 30%.
+	* internal.h (struct method_table_wrapper): Define new
+	  wrapper struct with additional serial.
+	* internal.h (RCLASS_M_TBL_INIT): New macro for initializing method
+	  table wrapper and st_table.
+	* method.h (void rb_sweep_method_entry): Rename rb_free_m_table to
+	  rb_free_m_tbl for consistency
+	* .gdbinit (define rb_method_entry): Update rb_method_entry gdb helper
+	  for new method table structure.
+	* class.c: Use RCLASS_M_TBL_WRAPPER and
+	  RCLASS_M_TBL_INIT macros.
+	* class.c (rb_include_class_new): Share WRAPPER between module and
+	  iclass, so serial can prevent double marking.
+	* eval.c (rb_prepend_module): ditto.
+	* eval.c (rb_using_refinement): ditto.
+	* gc.c: Mark and free new wrapper struct.
+	* gc.c (obj_memsize_of): Count size of additional wrapper struct.
+
+Tue Dec  3 14:05:49 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_uniq_bang): remove duplicate code.
+
+Tue Dec  3 13:40:42 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (ary_add_hash): set and return values because string keys
+	  will be frozen. [ruby-core:58809] [Bug #9202]
+
+	* array.c (rb_ary_uniq_bang): ditto.
+
+	* array.c (rb_ary_or): ditto.
+
+	* array.c (rb_ary_uniq): ditto.
+
+	* test/ruby/test_array.rb: tests for above.
+
+	The patch is from normalperson (Eric Wong).
+
+Tue Dec  3 12:20:21 2013  Aman Gupta <ruby@tmm1.net>
+
+	* string.c (rb_fstring): Use st_update instead of st_lookup +
+	  st_insert.
+	* string.c (fstr_update_callback): New callback for st_update.
+
+Tue Dec  3 12:17:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/constant.rb (RDoc::Constant#documented?): workaround for
+	  NoMethodError when the original of alias is not found.
+
+Tue Dec  3 10:43:58 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/lib/openssl/buffering.rb:  Return ASCII-8BIT strings from
+	  SSLSocket methods.  [ruby-trunk - Bug #9028]
+	* test/openssl/test_ssl.rb:  Test for the above.
+
+Tue Dec  3 09:42:27 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc master 900de99.  Changes include:
+
+	  Fixed documentation display of constants
+
+	  Fixed handling of unknown parsers
+
+	* test/rdoc:  ditto.
+
+Mon Dec  2 22:30:10 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (getenv): fixed test failures introduced by r43950.
+	  [ruby-core:58774] [Bug #9195] reported by phasis68 (Heesob Park).
+
+Mon Dec  2 21:49:19 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_rehash): make temporary st_table under the control
+	  of GC. [Bug #9187]
+
+	* test/ruby/test_hash.rb: add a test for above.
+
+Mon Dec  2 17:23:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* variable.c (rb_mod_constants): when calling Module#constants with
+	  inherit=false, there is no need to use a hashtable to deduplicate
+	  constant names. [Feature #9196] [ruby-core:58786]
+
+Mon Dec  2 14:16:52 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/smtp.rb (Net::SMTP#critical):  Always return a
+	  Net::SMTP::Response.  Patch by Pawel Veselov.
+	  [ruby-trunk - Bug #9125]
+	* test/net/smtp/test_smtp.rb:  Test for the above.
+
+Mon Dec  2 05:52:33 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master baa965b.  Notable changes:
+
+	  Copy directories to lib/ when installing extensions.  This completes
+	  the fix for [ruby-trunk - Bug #9106]
+
+	* test/rubygems:  ditto.
+
+Mon Dec  2 02:03:47 2013  Shota Fukumori  <her@sorah.jp>
+
+	* test/ruby/test_case.rb (test_nomethoderror):
+	  Add test related to r43913, r43914
+
+Mon Dec  2 00:53:01 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (getenv): use ANSI codepage version of getenv() for miniruby
+	  on Windows.
+	  [ruby-core:58732] [Bug #9189] reported by phasis68 (Heesob Park).
+
+Sun Dec  1 22:14:27 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributors.rdoc: [DOC] Import contributors from redmine wiki
+	  Many wiki pages have become outdated and spam-ridden, we will import
+	  these to trunk and begin maintaining them in ruby-trunk. This will
+	  also allow new contributors to easily contribute patches to update
+	  these pages, where previously a redmine account with wiki access was
+	  required. Another bonus is having a contributors file to show thanks
+	  to all of the people who have submitted a patch to Ruby.
+
+Sun Dec  1 18:03:26 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/maintainers.rdoc: [DOC] Current maintainers of Ruby
+
+Sun Dec  1 17:17:36 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Current branch maintainers
+
+Sun Dec  1 17:16:36 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Reporting other (ruby-lang.org) issues
+
+Sun Dec  1 17:15:51 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Current platform maintainers
+
+Sun Dec  1 17:14:55 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Reporting downstream distro issues
+
+Sun Dec  1 14:37:20 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_to_a): specify array capa.
+
+Sun Dec  1 14:15:36 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_rehash): fix to free new st_table when exception
+	  is raised in do_hash(). [Bug #9187]
+
+Sun Dec  1 11:57:59 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/lib/openssl/buffering.rb: Fix warning in copyright
+
+Sun Dec  1 08:27:28 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 66e5c39.  Notable changes:
+
+	  Implement gem.deps.rb (Gemfile) .lock support
+
+	  Fixed `gem uninstall` for a relative directory in GEM_HOME.
+
+	* test/rubygems:  ditto.
+
+Sun Dec  1 06:00:49 2013  Aman Gupta <ruby@tmm1.net>
+
+	* test/ruby/test_gc.rb (test_gc_reason): Force minor GC by consuming
+	  free slots to fix test.
+
+Sat Nov 30 21:22:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_initialize): check unknown keywords.  [ruby-dev:47152]
+	  [Bug #8060]
+
+Sat Nov 30 18:05:38 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (hash2named_arg): correct declaration to fix
+	  build failure.  a patch by phasis68 (Heesob Park) at
+	  [ruby-core:58710].  [Bug #9184]
+
+Sat Nov 30 17:46:35 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): determine exit status and signal to terminate
+	  before finalization, to get rid of access destroyed T_DATA exception
+	  object.  [ruby-core:58643] [Bug #9167]
+
+Sat Nov 30 16:25:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_with_index): should not store local variable
+	  address to memoize the arguments.  it is invalidated after the return.
+	  [ruby-core:58692] [Bug #9178]
+
+Sat Nov 30 13:28:13 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* siphash.c (sip_hash24): fix for aligned word access little endian
+	  platforms.  [ruby-core:58658] [Bug #9172]
+
+Sat Nov 30 13:21:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_yield_block): implement non-nil block argument.
+
+Fri Nov 29 20:59:39 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* vm_dump.c (rb_vmdebug_debug_print_pre): Bugfix. Get PC directly.
+	  PC is cached into local stack and cfp->pc is incorrect at next of
+	  branch or jump.
+	* vm_exec.h (DEBUG_ENTER_INSN): catch up this change.
+	* vm_core.h: update signature of rb_vmdebug_debug_print_pre.
+
+Fri Nov 29 20:43:57 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* compile.c: Bugsfix for dump_disasm_list.
+	  rb_inspect denies a hidden object. So, insert wrapper that creates
+	  the unhidden one.
+	  adjust->label is null sometimes.
+	  insn_data_line_no makes no sense at all.
+
+Fri Nov 29 18:06:45 2013  Shota Fukumori  <her@sorah.jp>
+
+	* test/ruby/test_case.rb (test_method_missing): Test for r43913.
+
+Fri Nov 29 17:53:22 2013  Shota Fukumori  <her@sorah.jp>
+
+	* vm_insnhelper.c (check_match): Fix SEGV with VM_CHECKMATCH_TYPE_CASE
+	  and class of `pattern` has `method_missing`
+	  [Bug #8872] [ruby-core:58606]
+
+Fri Nov 29 17:06:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_yield_block): yield block with rb_block_call_func
+	  arguments.
+
+	* range.c (range_each): use rb_yield_block.
+
+	* include/ruby/ruby.h (RB_BLOCK_CALL_FUNC_ARGLIST): constify argv.
+
+	* enum.c (rb_enum_values_pack): ditto.
+
+	* vm_eval.c (rb_block_call, rb_check_block_call): ditto.
+
+	* include/ruby/ruby.h (RB_BLOCK_CALL_FUNC_ARGLIST): for declaration
+	  argument list of rb_block_call_func.
+
+Fri Nov 29 11:26:43 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_block_call_func): add blockarg.  block
+	  function can take block argument, e.g., proc {|&blockarg| ...}.
+
+Thu Nov 28 21:43:48 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/dtrace_probes.rdoc: [DOC] Import dtrace probes doc from wiki
+
+Thu Nov 28 21:17:32 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Add heading above ChangeLog tips to
+	  setup entry for commits, its not required. Actually easier if
+	  contributors don't include a ChangeLog entry.
+
+Thu Nov 28 21:16:18 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Add coding style heading for patch
+	  rules
+
+Thu Nov 28 21:15:45 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Add notes about deciding what to patch
+
+Thu Nov 28 19:43:45 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* benchmark/bm_hash_flatten.rb: added. r43896 is about 4 times faster
+	  than 2.0.0p353.
+
+	* benchmark/bm_hash_keys.rb: added. r43896 is about 5 times faster
+	  than 2.0.0p353.
+
+	* benchmark/bm_hash_values.rb: added. r43896 is about 5 times faster
+	  than 2.0.0p353.
+
+Thu Nov 28 19:29:04 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/contributing.rdoc: [DOC] Add notes about slideshow proposals
+	  from wiki page: HowToRequestFeatures
+
+Thu Nov 28 17:34:42 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* st.c: add st_values() and st_values_check().
+
+	* include/ruby/st.h: add prototypes for above.
+
+	* hash.c (rb_hash_values): use st_values_check() for performance
+	  improvement if VALUE and st_data_t are compatible.
+
+Thu Nov 28 17:14:14 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* st.c (st_keys): fix not to use Qundef in st.c.
+
+	* include/ruby/st.h: define modified prototype.
+
+	* hash.c (rb_hash_keys): use modified st_keys().
+
+Thu Nov 28 16:34:43 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c: Expose details about last garbage collection via GC.stat.
+	* gc.c (gc_stat): Add :last_collection_flags for reason/trigger/type of
+	  last GC run.
+	* gc.c (gc_prof_sweep_timer_stop): Record HAVE_FINALIZE GPR even
+	  without GC_PROFILE_MORE_DETAIL.
+	* gc.c (gc_profile_flags): Add GC::Profiler.decode_flags to make sense
+	  of GC.stat[:last_collection_flags]
+	* test/ruby/test_gc.rb (class TestGc): Test for above.
+
+Thu Nov 28 16:15:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_dup2): extract from rb_cloexec_dup2() and
+	  redirect_dup2().
+
+Tue Nov 28 14:40:00 2013  Akira Matsuda  <ronnie@dio.jp>
+
+	* lib/drb/ssl.rb:  [Doc] Fix typo
+
+Thu Nov 28 13:56:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (Doxyfile): tool/file2lastrev.rb needs running with
+	  BASERUBY since r43617.  [ruby-dev:47823] [Bug #9169]
+
+Thu Nov 28 09:18:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (rb_fstring): fstrings should be ELTS_SHARED.
+	  If we resurrect dying objects (non-marked, but not swept yet),
+	  pointing shared string can be collected.
+	  To avoid such issue, fstrings (recorded to fstring_table)
+	  should not be ELTS_SHARED (should not have a shared string).
+
+Thu Nov 28 01:35:08 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* st.c (st_keys): fix to use st_index_t for size of hash.
+
+Thu Nov 28 00:36:52 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* st.c (st_keys): define st_keys(). it writes each key to buffer.
+
+	* hash.c (rb_hash_keys): use st_keys() for performance improvement
+	  if st_data_t and VALUE are compatible.
+
+	* include/ruby/st.h: define macro ST_DATA_COMPATIBLE_P() to predicate
+	  whether st_data_t and passed type are compatible.
+
+	* configure.in: check existence of builtin function to use in
+	  ST_DATA_COMPATIBLE_P().
+
+Thu Nov 28 00:07:28 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* ruby_atomic.h: remove duplicate definitions between ATOMIC_XXX
+	  and ATOMIC_SIZE_XXX.
+
+Wed Nov 27 23:55:50 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* ruby_atomic.h: define ATOMIC_SIZE_CAS() with
+	  __atomic_compare_exchange_n() and refactoring.
+
+Tue Nov 27 21:43:00 2013  Akira Matsuda  <ronnie@dio.jp>
+
+	* lib/irb/notifier.rb:  [Doc] Fix typo
+	* ext/json/lib/json/common.rb:  Ditto.
+
+Tue Nov 27 18:04:57 2013  Akira Matsuda  <ronnie@dio.jp>
+
+	* lib/irb/notifier.rb:  Fix typo
+
+Wed Nov 27 17:54:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_stacked_objects): check only when check_mode > 0.
+
+Wed Nov 27 16:07:19 2013  Aman Gupta <ruby@tmm1.net>
+
+	* test/ruby/test_gc.rb (class TestGc): Fix warning in
+	  test_expand_heap.
+
+Wed Nov 27 15:55:52 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (Init_GC): Add new GC::INTERNAL_CONSTANTS for information about
+	  GC heap/page/slot sizing.
+	* test/ruby/test_gc.rb (class TestGc): test for above.
+
+Wed Nov 27 15:21:17 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (gc_page_sweep): Fix compile warning from last commit.
+	* hash.c (hash_aset_str): Re-use existing variable to avoid
+	  unnecessary pointer dereferencing.
+
+Wed Nov 27 15:12:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_page_sweep): disable debug print.
+
+Wed Nov 27 15:05:59 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): add new information heap_eden_page_length and
+	  heap_tomb_page_length.
+
+	* test/ruby/test_gc.rb: fix to use GC.stat[:heap_eden_page_length]
+	  instead of GC.stat[:heap_length].
+	  This test expects `heap_eden_page_length' (used pages size).
+
+Wed Nov 27 15:02:53 2013  Aman Gupta <ruby@tmm1.net>
+
+	* test/ruby/test_eval.rb (class TestEval): Use assert_same instead of
+	  assert_equal.
+	* test/ruby/test_hash.rb (class TestHash): ditto.
+	* test/ruby/test_iseq.rb (class TestISeq): ditto.
+
+Wed Nov 27 14:50:02 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rinda/ring.rb:  Announce RingServer for the same process.
+	  [ruby-trunk - Bug #9163]
+	* test/rinda/test_rinda.rb:  Tests for the above.
+
+Wed Nov 27 14:37:33 2013  Aman Gupta <ruby@tmm1.net>
+
+	* test/ruby/test_eval.rb (class TestEval): Add test for shared eval
+	  filenames via rb_fstring().
+	* test/ruby/test_iseq.rb (class TestISeq): Add test for shared
+	  iseq labels via rb_fstring(). [Bug #9159]
+
+Wed Nov 27 14:24:55 2013  Aman Gupta <ruby@tmm1.net>
+
+	* hash.c (hash_aset_str): Use rb_fstring() to de-duplicate hash string
+	  keys. Patch by Eric Wong. [Bug #8998] [ruby-core:57727]
+	* test/ruby/test_hash.rb (class TestHash): test for above.
+
+Wed Nov 27 10:39:39 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c: Rename rb_heap_t members:
+	    used -> page_length
+	    limit -> total_slots
+
+Wed Nov 27 08:24:49 2013  Aman Gupta <ruby@tmm1.net>
+
+	* compile.c: Use rb_fstring() to de-duplicate string literals in code.
+	  [ruby-core:58599] [Bug #9159] [ruby-core:54405]
+	* iseq.c (prepare_iseq_build): De-duplicate iseq labels and source
+	  locations.
+	* re.c (rb_reg_initialize): Use rb_fstring() for regex string.
+	* string.c (rb_fstring): Handle non-string and already-fstr arguments.
+	* vm_eval.c (eval_string_with_cref): De-duplicate eval source
+	  filename.
+
+Wed Nov 27 07:13:54 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: psych version 2.0.2
+	* ext/psych/psych.gemspec: ditto
+
+Wed Nov 27 06:40:18 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: fix support for negative
+	  years.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ditto
+	* test/psych/test_date_time.rb: test for change.
+	  Fixes: https://github.com/tenderlove/psych/issues/168
+
+Wed Nov 27 04:46:55 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: fix regexp for matching TIME
+	  strings.
+	* test/psych/test_date_time.rb: test for change.
+	  Fixes: https://github.com/tenderlove/psych/issues/171
+
+Wed Nov 27 02:26:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_new4): copy the original capacity so that memsize of
+	  frozen shared string returns correct size.
+
+Wed Nov 27 02:20:13 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_hash): should not ignore the rest of recursive
+	  constructs.
+
+	* hash.c (rb_hash_hash): ditto.
+
+	* range.c (range_hash): ditto.
+
+	* struct.c (rb_struct_hash): ditto.
+
+	* test/-ext-/test_recursion.rb (TestRecursion): separate from
+	  test/ruby/test_thread.rb.
+
+Tue Nov 26 22:43:36 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash): cut off if recursion detected to get rid of stack
+	  overflow.  [ruby-core:58567] [Bug #9151]
+
+Tue Nov 26 20:02:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_settracefunc.rb: add tests for a_call/a_return
+	  by Brandur <brandur@mutelight.org> [Feature #9120]
+
+Tue Nov 26 19:29:52 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: add useful config "set breakpoint pending on"
+	  for run.gdb.
+
+Tue Nov 26 19:17:47 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/object_tracing.c (newobj_i): skip class_path if class
+	  is frozen.
+
+	  rb_class_path() can modify frozen classes (and causes errors).
+	  This patch is temporary. We need no-modification/no-allocation
+	  class path function.
+
+Tue Nov 26 18:12:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: skip "exception check" and "reentrant check (only normal
+	  events)  for internal events.
+
+	  Reentrant check for internal events are remaining.
+
+Tue Nov 26 17:38:16 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: prohibit to specify normal events and internal events
+	  simultaneously.
+	  I will introduce special care for internal events later.
+
+	* ext/-test-/tracepoint/tracepoint.c: test this behavior.
+
+	* test/-ext-/tracepoint/test_tracepoint.rb: ditto.
+
+Tue Nov 26 16:30:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_readlink): fix buffer overflow on a long symlink. since
+	  rb_str_modify_expand() expands from its length but not its capacity,
+	  need to set the length properly for each expansion.
+	  [ruby-core:58592] [Bug #9157]
+
+Tue Nov 26 14:23:17 2013  Aman Gupta <ruby@tmm1.net>
+
+	* ext/objspace/objspace_dump.c (dump_append_string_value): Escape
+	  control characters for strict json parsers.
+	* ext/objspace/objspace_dump.c (objspace_dump): Document File/IO
+	  output option.
+
+Tue Nov 26 11:43:19 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* ruby_atomic.h: use __atomic builtin functions supported by GCC.
+	  __sync family are legacy functions now and it is recommended
+	  that new code use the __atomic functions.
+	  http://gcc.gnu.org/onlinedocs/gcc/_005f_005fatomic-Builtins.html
+
+	* configure.in: check existence of __atomic functions.
+
+Tue Nov 26 10:57:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.gemspec: revert Gem::Specification#date
+	  for snapshot/release tarballs.
+
+Tue Nov 26 06:42:50 2013  Aman Gupta <ruby@tmm1.net>
+
+	* NEWS: Add ObjectSpace.after_gc_{start,end}_hook=
+	* ext/objspace/objspace_dump.c: [DOC] catch up dump/dump_all to r43679
+
+Tue Nov 26 04:12:10 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 612f85a.  Notable changes:
+
+	  Fixed installation and activation of git: and path: gems via
+	  Gem.use_gemdeps
+
+	  Improved documentation coverage
+
+	* test/rubygems:  ditto.
+
+Mon Nov 25 22:23:03 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/xmlrpc.rb: [DOC] Fix link to xmlrpc4r site [Bug #9148]
+	  Patch by Giorgos Tsiftsis
+
+Mon Nov 25 19:48:10 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/uri/common.rb: [DOC] typo fixes by @vipulnsward [Fixes GH-456]
+	  https://github.com/ruby/ruby/pull/456
+	* lib/uri/generic.rb: [DOC] ditto.
+
+Mon Nov 25 14:34:42 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/bigdecimal/bigdecimal.gemspec: bump BigDecimal to 1.2.3 for
+	  proper release date in RubyGems
+
+Mon Nov 25 14:25:08 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/bigdecimal/bigdecimal.gemspec: Remove Gem::Specification#date
+	  We should rely on rubygems to create the date the gem was released
+	  for each version.
+
+Mon Nov 25 06:53:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: do not use ruby_sized_xrealloc() and ruby_sized_xfree()
+	  if HAVE_MALLOC_USABLE_SIZE (or _WIN32) is defined.
+
+	  We don't need these function if malloc_usable_size() is available.
+
+	* gc.c: catch up this change.
+
+	* gc.c: define HAVE_MALLOC_USABLE_SIZE on _WIN32.
+
+	* array.c (ary_resize_capa): do not use ruby_sized_xfree() with
+	  local variable to avoid "unused local variable" warning.
+	  This change only has few impact.
+
+	* string.c (rb_str_resize): ditto.
+
+Mon Nov 25 05:05:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/-ext-/tracepoint/test_tracepoint.rb: catch up GC.stat changes
+	  at r43835.
+
+Mon Nov 25 04:45:59 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: continue to change OLDSPACE -> OLDMALLOC.
+	  RGENGC_ESTIMATE_OLDSPACE -> RGENGC_ESTIMATE_OLDMALLOC.
+
+	* gc.c: add a new major GC reason GPR_FLAG_MAJOR_BY_OLDMALLOC.
+
+Mon Nov 25 04:16:09 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change terminology "..._num" to "..._slots" about slot operation.
+	  * final_num -> final_slots
+	  * objspace_live_num() -> objspace_live_slots()
+	  * objspace_limit_num() -> objspace_limit_slots()
+	  * objspace_free_num() -> objspace_free_slots()
+
+Mon Nov 25 04:03:12 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): add internal information.
+	  * heap_swept_slot
+	  * malloc_increase
+	  * malloc_limit
+	  * remembered_shady_object
+	  * remembered_shady_object_limit
+	  * old_object
+	  * old_object_limit
+	  * oldmalloc_increase
+	  * oldmalloc_limit
+
+	* gc.c (gc_stat): rename names.
+	  * heap_live_num -> heap_live_slot
+	  * heap_free_num -> heap_free_slot
+	  * heap_final_slot -> heap_final_slot
+
+	  Quote from RDoc of GC.stat():
+	    "The contents of the hash are implementation specific and may
+	    be changed in the future."
+
+	* test/ruby/test_gc.rb: catch up this change.
+
+Mon Nov 25 03:59:45 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb: catch up last commit.
+	  Now RUBY_GC_OLDSPACE_LIMIT(...) is RUBY_GC_OLDMALLOC_LIMIT(...).
+
+Mon Nov 25 03:10:46 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change terminology OLDSPACE -> OLDMALLOC.
+	  (oldspace -> oldmalloc for variable names)
+
+	  OLDSPACE is confusing because it is not includes slots.
+	  To more clearly, rename such as (oldspace_limit -> oldmalloc_limit).
+	  It is clear that it measures (estimates) malloc()'ed size.
+
+Mon Nov 25 00:50:03 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* internal.h: use __builtin_bswap16() if possible.
+
+	* configure.in: check existence of __builtin_bswap16().
+
+Sun Nov 24 22:24:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigxor_int): Apply BIGLO for long in a BDIGIT expression.
+	  (bigor_int): Ditto.
+	  (bigand_int): Ditto.
+
+Sun Nov 24 18:13:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h (SIZEOF_ACTUAL_BDIGIT): Defined.
+
+	* include/ruby/ruby.h (RBIGNUM_EMBED_LEN_MAX): Use
+	  SIZEOF_ACTUAL_BDIGIT instead of SIZEOF_BDIGITS.
+	  SIZEOF_BDIGITS can be different to sizeof(BDIGIT).
+
+Sun Nov 24 13:49:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h: Don't use int128_t for Bignum.
+	  It's not always faster.
+
+	* bignum.c: Ditto.
+
+Sun Nov 24 10:18:15 2013  Aman Gupta <ruby@tmm1.net>
+
+	* NEWS: Add details about new debugging features and APIs.
+
+Sun Nov 24 09:37:20 2013  Andrew Vit  <andrew@avit.ca>
+
+	* lib/csv.rb: Optimize header hashes by freezing string keys.
+	  [ruby-core:58510]
+
+Sun Nov 24 09:18:06 2013  Aman Gupta <ruby@tmm1.net>
+
+	* ext/objspace/objspace_dump.c (dump_object): Use PRIuSIZE to print
+	  size_t for better win32 compatibility.
+	* test/objspace/test_objspace.rb (test_dump_all): Hold reference to
+	  test string to avoid failure due to GC. Reduce size of failure message
+	  using grep(/TEST STRING/).
+
+Sun Nov 24 08:38:00 2013  Kyle Stevens  <kstevens715@gmail.com>
+
+	* lib/csv.rb: If skip_lines is set to a String, convert it to a Regexp
+	  to prevent the alternative, which is that each line in the CSV gets
+	  converted to a Regexp when calling skip_lines#match.
+
+Sun Nov 24 01:03:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_power): Use FIX2LONG instead
+	  of FIX2INT to avoid conversion error.
+
+Sun Nov 24 00:44:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RBIGNUM_EMBED_LEN_MAX): define by macros
+	  defined in defines.h, instead of complex and repeated expression.
+
+Sat Nov 23 22:22:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (RBIGNUM_EMBED_LEN_MAX): Limit the value to
+	  less than 8.
+
+Sat Nov 23 19:52:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (BigMath.E): Use BigMath.exp.
+	  [Feature #6857] [ruby-core:47130]
+
+Sat Nov 23 19:46:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): Optimize the
+	  calculation algorithm to reduce the number of divisions.
+	  This optimization was proposed by Rafal Michalski.
+	  [Feature #6857] [ruby-core:47130]
+
+Sat Nov 23 19:20:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_div2): The signature was
+	  changed to allow us to pass arguments directly.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_div3): Added for the role of
+	  the old BigDecimal_div2.
+
+Sat Nov 23 12:31:00 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix global variable name.
+	  Now we have following environments (and related variable names).
+
+	  * RUBY_GC_HEAP_INIT_SLOTS
+	  * RUBY_GC_HEAP_FREE_SLOTS
+	  * RUBY_GC_HEAP_GROWTH_FACTOR (new from 2.1)
+	  * RUBY_GC_HEAP_GROWTH_MAX_SLOTS (new from 2.1)
+
+	  * obsolete
+	    * RUBY_FREE_MIN       -> RUBY_GC_HEAP_FREE_SLOTS (from 2.1)
+	    * RUBY_HEAP_MIN_SLOTS -> RUBY_GC_HEAP_INIT_SLOTS (from 2.1)
+
+	  * RUBY_GC_MALLOC_LIMIT
+	  * RUBY_GC_MALLOC_LIMIT_MAX (new from 2.1)
+	  * RUBY_GC_MALLOC_LIMIT_GROWTH_FACTOR (new from 2.1)
+
+	  * RUBY_GC_OLDSPACE_LIMIT (new from 2.1)
+	  * RUBY_GC_OLDSPACE_LIMIT_MAX (new from 2.1)
+	  * RUBY_GC_OLDSPACE_LIMIT_GROWTH_FACTOR (new from 2.1)
+
+	* test/ruby/test_gc.rb: catch up this change.
+
+Sat Nov 23 09:45:49 2013  Aman Gupta <ruby@tmm1.net>
+
+	* marshal.c (w_object): Use HASH_PROC_DEFAULT directly from internal.h
+
+Sat Nov 23 08:43:23 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c: Rename heap_pages_swept_num to heap_pages_swept_slots to
+	  clarify meaning (number of slots, not pages).
+
+Sat Nov 23 08:23:23 2013  Aman Gupta <ruby@tmm1.net>
+
+	* lib/set.rb (class SortedSet): Fix source_location for methods
+	  defined via eval.
+
+Sat Nov 23 03:44:03 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master dcce4ff.  Important changes
+	  in this commit:
+
+	  Remove automatic detection of gem dependencies files.  This prevents a
+	  security hole as described in [ruby-core:58490]
+
+	  Fixed bugs for installing git gems.
+
+	* test/rubygems:  ditto.
+
+Fri Nov 22 22:30:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_power):
+	  Round the result value only if the precision is given.
+
+Fri Nov 22 17:20:50 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* transcode.c (str_transcode0): don't scrub invalid chars if
+	  str.encode doesn't have explicit invalid: :replace.
+	  workaround fix for see #8995
+
+Fri Nov 22 17:11:26 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* include/ruby/intern.h, internal.h: Expose rb_gc_count().
+
+Fri Nov 22 17:07:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.gemspec: version 1.2.2.
+
+Fri Nov 22 17:04:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_data_type):
+	  Use RUBY_TYPED_FREE_IMMEDIATELY only if it is available.
+
+Fri Nov 22 16:49:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_power): Round the result value.
+	  [Bug #8818] [ruby-core:56802]
+
+	* test/bigdecimal/test_bigdecimal.rb: Add a test for the above fix.
+
+Fri Nov 22 16:25:43 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_set_increment): accept minimum additional page number.
+
+	* gc.c (gc_after_sweep): allocate pages to allocate at least
+	   RUBY_HEAP_MIN_SLOTS.
+	   [Bug #9137]
+
+Fri Nov 22 16:19:52 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* include/ruby/intern.h (rb_gc_set_params): Deprecate
+	  rb_gc_set_params because it's only used in ruby internal.
+
+	* internal.h (ruby_gc_set_params): Declare rb_gc_set_params's
+	  alias function.
+
+	* gc.c: ditto.
+
+	* ruby.c: use ruby_gc_set_params.
+
+Fri Nov 22 14:55:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): Insert rb_thread_check_ints.
+
+Fri Nov 22 14:35:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): Fix the inserting points
+	  of RB_GC_GUARDs.
+
+Fri Nov 22 14:31:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c: Fix indentation.
+
+Fri Nov 22 14:03:00 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf: merge nkf 2.1.3 2a2f2c5.
+
+Fri Nov 22 12:43:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strtod): ignore too long fraction part, which does not
+	  affect the result.
+
+Fri Nov 22 12:17:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/buffering.rb (OpenSSL::Buffering#initialize):
+	  initialize of a module should pass arguments to super.
+
+Fri Nov 22 12:02:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_settracefunc.rb: Ignore events from other threads.
+
+Fri Nov 22 10:35:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (ruby_vm_destruct): do not use ruby_xfree() after freeing
+	  objspace.
+
+	* gc.c (ruby_mimfree): added. It is similar to ruby_mimmalloc().
+
+	* internal.h: ditto.
+
+Fri Nov 22 09:42:35 2013  Zachary Scott  <e@zzak.io>
+
+	* test/digest/test_digest.rb: Reverse order of assert_equal
+	  Reported by @splattael
+
+Fri Nov 22 09:03:16 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c: fix build failure on FreeBSD introduced by r43763.
+	  malloc_usable_size() is defined by malloc_np.h on FreeBSD.
+
+	* configure.in: check malloc.h and malloc_np.h.
+
+Fri Nov 22 08:27:13 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 50a8210.  Important changes
+	  in this commit:
+
+	  RubyGems now automatically checks for gem.deps.rb or Gemfile when
+	  running ruby executables.  This behavior is similar to `bundle exec
+	  rake`.  This change may be reverted before Ruby 2.1.0 if too many bugs
+	  are found.
+
+	* test/rubygems:  ditto.
+
+Thu Nov 21 22:33:59 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: RGENGC_CHECK_MODE should be 0.
+
+Thu Nov 21 21:40:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (VpAlloc): Fix the expr to adjust the size
+	  of the digit array.
+
+Thu Nov 21 21:36:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_sqrt): Fix the precision of
+	  the result BigDecimal of sqrt.
+	  [Bug #5266] [ruby-dev:44450]
+
+	* test/bigdecimal/test_bigdecimal.rb: add tests for the above changes.
+
+Thu Nov 21 18:49:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (vm_xrealloc, vm_xfree): use malloc_usable_size() to obtain old
+	  size if available.
+
+Thu Nov 21 18:47:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (SimpleDelegator#__getobj__): target object must be set.
+
+	* lib/delegate.rb (DelegateClass#__getobj__): ditto.
+
+Thu Nov 21 18:28:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#initialize): use class method to get rid
+	  of warnings when $VERBOSE.
+
+Thu Nov 21 17:43:29 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename initial_xxx variables to gc_params.xxx.
+	  They are not only used initial values.
+
+	  Chikanaga-san: Congratulations on RubyPrize!
+
+Thu Nov 21 17:16:00 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: enable "RGENGC_ESTIMATE_OLDSPACE" option as default.
+	  Without this option, some application consumes huge memory.
+	  (and there are only a few performance down)
+
+	  Introduced new environment variables:
+	  * RUBY_GC_HEAP_OLDSPACE (default 16MB)
+	  * RUBY_GC_HEAP_OLDSPACE_MAX (default 128 MB)
+	  * RUBY_GC_HEAP_OLDSPACE_GROWTH_FACTOR (default 1.2)
+
+	* gc.c (initial_malloc_limit): rename to initial_malloc_limit_min.
+
+Thu Nov 21 16:51:34 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/digest/bubblebabble/bubblebabble.c: Teach RDoc digest/bubblebabble
+
+Thu Nov 21 16:50:16 2013  Zachary Scott  <e@zzak.io>
+
+	* test/digest/test_digest.rb: Add more tests for digest/bubblebabble
+
+Thu Nov 21 16:32:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator#method_missing): try private methods defined in
+	  Kernel after the target.  [Fixes GH-449]
+
+Thu Nov 21 16:25:08 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/uri/test_generic.rb (URI#test_merge): Test uri + URI(path)
+	  in addition to uri + path.
+
+Thu Nov 21 15:36:08 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/lib/openssl/buffering.rb: [DOC] Fix HEREDOC comment for
+	  OpenSSL::Buffering which breaks overview because of RDoc bug
+
+Thu Nov 21 14:46:57 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval_intern.h (SAVE_ROOT_JMPBUF): workaround for the failure of
+	  test/ruby/test_exception.rb on Windows.
+	  wrap by __try and __exception statements on mswin to raise SIGSEGV
+	  when EXCEPTION_STACK_OVERFLOW is occurred, because MSVCRT doesn't
+	  handle the exception.
+	  however, (1) mingw-gcc doesn't support __try and __exception
+	  statements, and (2) we cannot retry SystemStackError after this
+	  change yet (maybe crashed) because SEH and longjmp() are too
+	  uncongenial.
+
+	* signal.c (check_stack_overflow, CHECK_STACK_OVERFLOW): now defined on
+	  Windows, too.
+
+	* thread_win32.c (ruby_stack_overflowed_p): ditto.
+
+Thu Nov 21 14:18:24 2013  Zachary Scott  <e@zzak.io>
+
+	* object.c: [DOC] Clarify Object#dup vs #clone [Bug #9128]
+	  Moving existing doc for this comparison to separate section of #dup
+	  Adding examples to document behavior of #dup with Module#extend.
+	  Based on a patch by stevegoobermanhill
+
+Thu Nov 21 14:06:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_marks_check): do not dump all refs.
+
+	* gc.c (allrefs_dump_i): fix output format.
+
+Thu Nov 21 13:43:07 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change RGENGC_CHECK_MODE (>= 2) logic.
+	  Basically, make an object graph of all of living objects before and
+	  after marking and check status.
+
+	  [Before marking: check WB sanity]
+	  If there is a non-old object `obj' pointed from old object
+	  (`parent') then `parent' or `obj' should be remembered.
+
+	  [After marking: check marking miss]
+	  Traversible objects with the object graph should be marked.
+	  (However, this alert about objects pointed by machine context
+	   can be false positive. We only display alert.)
+
+	  [Implementation memo]
+	  objspace_allrefs() creates an object graph.
+	  The object graph is represented by st_table, key is object (VALUE)
+	  and value is referring objects. Referring objects are stored by
+	  "struct reflist".
+
+	* gc.c (init_mark_stack): do not use push_mark_stack_chunk() at init.
+	  This pre-allocation causes failure on is_mark_stack_empty()
+	  without any pushing.
+
+Thu Nov 21 13:40:20 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/observer.rb: [DOC] Clarify default observer method.
+	  By @edward [Fixes GH-450] https://github.com/ruby/ruby/pull/450
+
+Thu Nov 21 13:32:53 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_engine.c: [DOC] Documentation for OpenSSL::Engine
+	  This patch is based off work by @vbatts in GH-436 completing the
+	  documentation for this class and its methods.
+	  https://github.com/ruby/ruby/pull/436
+
+Thu Nov 21 10:45:22 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/lib/openssl/buffering.rb: Remove unused arguments from
+	  OpenSSL::Buffering.new [Fixes GH-445]
+
+Thu Nov 21 10:30:47 2013  Zachary Scott  <e@zzak.io>
+
+	* test/digest/test_digest.rb: Add test for Digest::SHA256.bubblebabble
+
+Wed Nov 20 20:54:01 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* tool/instruction.rb : fix typo.
+
+Wed Nov 20 19:45:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (rand_init): Make it possible to specify arbitrary array
+	  for init_genrand().
+
+Wed Nov 20 17:34:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (rb_gc_mark_symbols): set global_symbols.minor_marked only
+	  when full_mark is 0.
+	  rb_gc_mark_symbols() (with full_mark == 1) can be called by other
+	  than GC (such as rb_objspace_reachable_objects_from_root()).
+
+Wed Nov 20 11:46:38 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: merge JSON 1.8.1.
+	  https://github.com/nurse/json/compare/002ac2771ce32776b32ccd2d06e5604de6c36dcd...e09ffc0d7da25d0393873936c118c188c78dbac3
+	* Remove Rubinius exception since transcoding should be working now.
+	* Fix https://github.com/flori/json/issues/162 reported by Marc-Andre
+	  Lafortune <github_rocks@marc-andre.ca>. Thanks!
+	* Applied patches by Yui NARUSE <naruse@airemix.jp> to suppress
+	  warning with -Wchar-subscripts and better validate UTF-8 strings.
+	* Applied patch by ginriki@github to remove unnecessary if.
+	* Add load/dump interface to JSON::GenericObject to make
+	  serialize :some_attribute, JSON::GenericObject
+	  work in Rails active models for convenient
+	  SomeModel#some_attribute.foo.bar access to serialised JSON data.
+
+Wed Nov 20 01:39:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/constant.rb (RDoc::Constant#documented?): workaround for
+	  NoMethodError when the original of alias is not found.
+
+Tue Nov 19 23:38:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--with-os-version-style): option to transform target
+	  OS version string.
+
+Tue Nov 19 21:27:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/http/utils.rb (spawn_server): Specify zero for port to
+	  avoid reusing an allocated port.
+
+	* test/net/http/test_http.rb: Don't specify port here.
+
+	* test/net/http/test_https.rb: Ditto.
+
+Tue Nov 19 18:52:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_is_swept_object): use heap_page::before_sweep flag.
+
+Tue Nov 19 18:49:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_reachable_objects_from_root): do major marking.
+
+Tue Nov 19 18:45:40 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_resurrect): added.
+	  rb_fstring() used rb_gc_mark() to avoid freeing used string.
+	  However, rb_gc_mark() set mark bit *and* pushes mark_stack.
+	  rb_gc_resurrect() does only set mark bit if it is before sweeping.
+
+	* string.c (rb_fstring): use rb_gc_resurrect.
+
+	* internal.h: add decl.
+
+Tue Nov 19 09:47:02 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc master a1195ce.  Changes include:
+
+	  Improved accessibility of the main sidebar navigation.
+
+	  Fixed handling of regexp options in HTML source highlighting.
+
+	* test/rdoc:  ditto.
+
+Tue Nov 19 09:33:52 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 6a3d9f9.  Changes include:
+
+	  Compatibly renamed Gem::DependencyResolver to Gem::Resolver.
+
+	  Added support for git gems in gem.deps.rb and Gemfile.
+
+	  Fixed resolver bugs.
+
+	* test/rubygems: ditto.
+
+	* lib/rubygems/LICENSE.txt:  Updated to license from RubyGems trunk.
+	  [ruby-trunk - Bug #9086]
+
+	* lib/rubygems/commands/which_command.rb:  RubyGems now indicates
+	  failure when any file is missing.  [ruby-trunk - Bug #9004]
+
+	* lib/rubygems/ext/builder:  Extensions are now installed into the
+	  extension install directory and the first directory in the require
+	  path from the gem.  This allows backwards compatibility with msgpack
+	  and other gems that calculate full require paths.
+	  [ruby-trunk - Bug #9106]
+
+
+Tue Nov 19 07:21:56 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (LOCALTIME_OVERFLOW_PROBLEM): Define it for cross
+	  compiling.
+	  [ruby-core:58391] [Bug #9119] Reported by Luis Lavena.
+	  Analyzed by Heesob Park.
+
+Tue Nov 19 05:55:05 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/rubygems_hook.rb:  Remove debugging puts committed by
+	  accident.
+
+Mon Nov 18 22:47:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (TH_PUSH_TAG, TH_EXEC_TAG): refine stack overflow
+	  detection.  chain local tag after setjmp() successed on it, because
+	  calling setjmp() also can overflow the stack.
+	  [ruby-dev:47804] [Bug #9109]
+
+	* vm_eval.c (rb_catch_obj): now th->tag points previous tag until
+	  TH_EXEC_TAG().
+
+	* thread_pthread.c (ruby_init_stack): set stack_start properly by
+	  get_main_stack() if possible.
+
+Mon Nov 18 22:45:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_jump.c (rb_exec_end_proc): unlink and free procs data before
+	  calling for each procs.  [Bug #9110]
+
+Sun Nov 17 06:33:32 2013  Shota Fukumori  <her@sorah.jp>
+
+	* configure.in: Use $LIBS for base of $SOLIBS, also in darwin.
+	  By this fix, environment that libgmp is located in $LIBS can build
+	  ruby.
+
+Sun Nov 17 01:56:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): Show error
+	  message instead of error number.
+	  (thread_create_core): Ditto.
+
+	* cont.c (fiber_machine_stack_alloc): Ditto.
+
+Sat Nov 16 18:28:08 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/ultralightparser.rb
+	  (REXML::Parsers::UltraLightParser#parse): Fix wrong :start_doctype
+	  position.
+	  [Bug #9061] [ruby-dev:47778]
+	  Patch by Ippei Obayashi. Thanks!!!
+
+	* test/rexml/parser/test_ultra_light.rb: Add a test for this case.
+
+Sat Nov 16 02:13:56 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* cont.c : Introduce ensure rollback mechanism. Please see below.
+
+	* internal.h (ruby_register_rollback_func_for_ensure): catch up above change.
+	  Add rollback mechanism API.
+
+	* vm_core.h (typedef struct rb_vm_struct): catch up above change.
+	  Introduce ensure-rollback relation table.
+
+	* vm_core.h (typedef struct rb_thread_struct): catch up above change.
+	  Introduce ensure stack.
+
+	* eval.c (rb_ensure): catch up above change.
+	  Introduce ensure stack.
+
+	* hash.c : New function for rollback ensure, and register it to
+	  ensure-rollback relation table. [ruby-dev:47803] [Bug #9105]
+
+	Ensure Rollback Mechanism:
+	A rollback's function is a function to rollback a state before ensure's
+	function execution.
+	When the jump of callcc is across the scope of rb_ensure,
+	ensure's functions and rollback's functions are executed appropriately
+	for keeping consistency.
+
+	Current API is unstable, and only internal use.
+
+	ruby_register_rollback_func_for_ensure(ensure_func,rollback_func)
+	This API create relation ensure's function to rollback's function.
+	By registered rollback's function, it is executed When jumping into
+	corresponding rb_ensure scope.
+
+Sat Nov 16 00:18:36 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* eval_jump.c (rb_exec_end_proc): fix double free or corruption error
+	  when reentering by callcc. [ruby-core:58329] [Bug #9110]
+
+	* test/ruby/test_beginendblock.rb: test for above.
+
+Fri Nov 15 01:06:04 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/objspace/objspace_dump.c (dump_output): allow IO object as
+	  output, and use Tempfile.create and return open file instead of
+	  mkstemp() and path name for :file output.
+	  [ruby-core:58266] [Bug #9102]
+
+	* test/objspace/test_objspace.rb (TestObjSpace#dump_my_heap_please):
+	  remove temporary output file.
+
+Thu Nov 14 23:39:00 2013  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb: [DOC] remove example of
+	  Rational#to_d without argument. [Bug #8958]
+
+Thu Nov 14 20:24:15 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ruby_atomic.h (ATOMIC_SIZE_CAS): fix compile error on Solaris
+	  since r43460.
+
+Thu Nov 14 19:53:00 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/test_cipher.rb (test_aes_gcm_wrong_tag): Don't use
+	  String#succ because it can make modified (wrong) auth_tag longer
+	  than 16 bytes.  The longer auth_tag makes that
+	  EVP_CIPHER_CTX_ctrl (and internally aes_gcm_ctrl) fail.
+	  [ruby-core:55143] [Bug #8439] reported by Vit Ondruch.
+
+Thu Nov 14 11:33:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (foreach_safe_i, hash_foreach_iter): deal with error detected
+	  by ST_CHECK.
+
+	* st.c (st_foreach_check): call with non-error argument in normal case.
+
+Thu Nov 14 02:37:14 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/thread/thread.c: [DOC] This patch accomplishes the following:
+
+	  - Teach RDoc about ConditionVariable
+	  - Teach RDoc about Queue
+	  - Teach RDoc about SizedQueue
+	  - Use fully-qualified namespace for Document-method
+	    This is necessary to separate definitions between classes
+	  - Fix rdoc bug in call_seq vs. call-seq
+	  - Correct doc for SizedQueue#pop patch by @jackdanger [Bug #8988]
+
+Thu Nov 14 01:11:54 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb: [DOC] +precision+ is required
+
+Wed Nov 13 19:21:36 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb: [DOC] Document the required
+	  +precision+ argument for Rational#to_d [Bug #8958]
+
+Wed Nov 13 19:02:05 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/digest/*: [DOC] Fix several typos and broken http links.
+	  Improved examples for Digest overview and fixed a broken example in
+	  Digest::HMAC overview. This patch also adds a description of
+	  Digest::SHA256.bubblebabble to the Digest overview.
+
+	  Patched by @stomar [Bug #9027]
+
+Wed Nov 13 18:32:12 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_config.c: [DOC] Document the following:
+
+	  - OpenSSL::ConfigError
+	  - OpenSSL::Config::DEFAULT_CONFIG_FILE
+
+	  Patched by @vbatts via GH-436
+	  https://github.com/ruby/ruby/pull/436
+
+Wed Nov 13 18:03:00 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_asn1.c: [DOC] Document parts of
+	  OpenSSL::ASN1::ObjectId included a fix for the class overview, which
+	  previously showed the documentation for Constructive due to missing
+	  ObjectId overview. This patch also includes a note for Primitive.
+
+	  Based on a patch by @vbatts via GH-436
+	  https://github.com/ruby/ruby/pull/436
+
+Wed Nov 13 17:19:36 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/lib/openssl/config.rb: In #parse use +string+ for +str+
+
+Wed Nov 13 17:09:45 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/lib/openssl/*.rb: [DOC] Document the following:
+
+	  - Integer#to_bn
+	  - OpenSSL::Buffering module
+	  - Deprecated OpenSSL::Digest::Digest compatibility class
+	  - OpenSSL::Config
+
+	  These changes were based on a patch by @vbatts via GH-436
+	  https://github.com/ruby/ruby/pull/436
+
+Wed Nov 13 10:55:43 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/regexp.rdoc: [DOC] Fix typo in Special global variables section.
+	  Reported by Alex Johnson on ruby-doc.org
+
+Wed Nov 13 10:43:19 2013  Zachary Scott  <e@zzak.io>
+
+	* hash.c: [DOC] Adds an example for Hash#store
+
+Wed Nov 13 09:03:40 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/regexp.rdoc: [DOC] add note about Bug #4044 as suggested by
+	  duerst-san in [ruby-core:43612] [Fixes GH-443] Patched by @rosenfeld
+	  https://github.com/ruby/ruby/pull/443
+
+Tue Nov 12 10:15:14 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/insure_session.rb:  Remove unused test file.
+
+Tue Nov 12 09:16:24 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master b9213d7.  Changes include:
+
+	  Fixed tests on Windows (I hope) by forcing platform for
+	  platform-dependent tests.
+
+	  Fixed File.exists? warnings.
+
+	  Improved testing infrastructure.
+
+	* test/rubygems:  ditto.
+
+	* test/rdoc/test_rdoc_rubygems_hook.rb:  Switch to util_spec like
+	  RubyGems.
+
+Mon Nov 11 18:31:12 2013  Aman Gupta <ruby@tmm1.net>
+
+	* internal.h: move common string/hash flags to include file.
+	* ext/objspace/objspace_dump.c: remove flags shared above.
+	* hash.c: ditto.
+	* string.c: ditto.
+
+Mon Nov 11 04:36:14 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/specification.rb:  Include 2.2.0.preview.2 when checking
+	  if extensions should be built.  Fixes a ruby-ci failure.
+	* test/rubygems/test_gem_specification.rb:  Test for the above.
+
+Mon Nov 11 03:15:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (symbol2event_flag): add secret feature.
+	  add a_call/a_return events.
+	  a_call is call | b_call | c_call, and same as a_return.
+
+Mon Nov 11 02:51:17 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems: Update to RubyGems master 4bdc4f2.  Important changes
+	  in this commit:
+
+	  RubyGems now chooses the test server port reliably.  Patch by akr.
+
+	  Partial implementation of bundler's Gemfile format.
+
+	  Refactorings to improve the new resolver.
+
+	  Fixes bugs in the resolver.
+
+	* test/rubygems:  Tests for the above.
+
+Mon Nov 11 01:02:06 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/timeout.rb: [DOC] Add note about change from #8730 [Fixes GH-440]
+	* NEWS: [DOC] Improve grammar on change to Timeout
+	  Patched by @srawlins in https://github.com/ruby/ruby/pull/440
+
+Sun Nov 10 23:47:05 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gc.c (rb_gcdebug_print_obj_condition): catch up recent changes
+	  to compile on GC_DEBUG.
+
+Sun Nov 10 22:16:19 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (exc_cause): captured previous exception.
+
+	* eval.c (make_exception): capture previous exception automagically.
+	  [Feature #8257]
+
+Sun Nov 10 08:37:20 2013  Zachary Scott  <e@zzak.io>
+
+	* thread.c: [DOC] Remove duplicate reference
+
+Sun Nov 10 08:09:29 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/drb/drb.rb: [DOC] promote better windows-safe filename regular
+	  expression in DRb Logger example. Reported by Chris Pheonix
+	  [Bug #9074]
+
+Sun Nov 10 08:03:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_define_finalizer, rb_undefine_finalizer): rename and export
+	  finalizer functions.
+
+Sun Nov 10 07:41:22 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/weakref.rb: [DOC] fix typos by @xaviershay [Fixes GH-439]
+	  https://github.com/ruby/ruby/pull/439
+
+Sun Nov 10 06:14:39 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): emit opt_str_freeze if the #freeze
+	  method is called on a static string literal with no arguments.
+
+	* defs/id.def (firstline): add freeze so idFreeze is available
+
+	* insns.def (opt_str_freeze): add opt_str_freeze instruction which
+	  pushes a frozen string literal without allocating a new object if
+	  String#freeze is not overridden
+
+	* string.c (Init_String): define String#freeze
+
+	* vm.c (vm_init_redefined_flag): define BOP_FREEZE on String class as
+	  a basic operation
+
+	* vm_insnhelper.h: ditto
+
+	[Feature #8992] [ruby-core:57705]
+
+Sun Nov 10 01:34:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (vm_malloc_increase): sweep immediately on GC due to malloc().
+	  To reduce memory usage, sweep as soon as possible.
+	  This behavior is same as Ruby 2.0.0 and before.
+
+Sun Nov 10 00:39:26 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: output version description and GC::OPTS.
+
+Sun Nov 10 00:36:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (should_be_callable): allow private call since rb_eval_cmd
+	  calls even private methods.
+
+Sun Nov 10 00:33:17 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/racc/rdoc/grammar.en.rdoc: [DOC] fix typo by Tsuyoshi Sawada
+	  [Bug #9077]
+
+Sat Nov  9 22:35:35 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (Gem::Specification.load): obtain spec date from
+	  VCS for the case using git, RUBY_RELEASE_DATE is the last resort.
+	  probably fixes [Bug #9085].
+
+Sat Nov  9 20:56:12 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* ext/objspace/object_tracing.c: use declarations in internal.h.
+
+	* ext/objspace/objspace.c: ditto
+
+Sat Nov  9 20:32:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/objspace/test_objspace.rb (test_dump_all): Make the test string
+	  shorter to be an embedded string on 32bit environment as well as
+	  64bit environment.
+
+Sat Nov  9 15:00:16 2013  Zachary Scott  <e@zzak.io>
+
+	* io.c: [DOC] ARGF.gets may return nil [Bug #9029] patch by znz
+
+Sat Nov  9 14:54:52 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/*: [DOC] document various constants @steveklabnik [Bug #8812]
+
+Sat Nov  9 14:50:09 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/rss.rb: [DOC] document Time#w3cdtf by @steveklabnik
+	  [Bug #8821]
+
+Sat Nov  9 14:29:04 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/dl/cfunc.c: [DOC] fix typo in example [Bug #8944]
+	  Patched by Heesob Park
+
+Sat Nov  9 13:59:58 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/test/unit/assertions.rb: [DOC] better example for assert_send()
+	  Patch by Andrew Grimm [Bug #8975]
+
+Sat Nov  9 12:45:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* insns.def: unify ic_constant_serial and ic_class_serial into one field
+	  ic_serial. This is possible because these fields are only ever used
+	  exclusively with each other.
+
+	* insns.def: ditto
+	* vm_core.h: ditto
+	* vm_insnhelper.c: ditto
+
+Sat Nov  9 12:31:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* class.c: unify names of vm state version counters to 'serial'.
+	  This includes renaming 'vm_state_version_t' to 'rb_serial_t',
+	  'method_state' to 'method_serial', 'seq' to 'class_serial',
+	  'vmstat' to 'constant_serial', etc.
+
+	* insns.def: ditto
+	* internal.h: ditto
+	* vm.c: ditto
+	* vm_core.h: ditto
+	* vm_insnhelper.c: ditto
+	* vm_insnhelper.h: ditto
+	* vm_method.c: ditto
+
+Sat Nov  9 09:22:29 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_page_sweep, rgengc_rememberset_mark): Refactoring.
+	  Get bitmaps directly.
+
+Sat Nov  9 09:16:36 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (RVALUE_PROMOTE_INFANT): Refactoring. Remove duplicated nonsense
+	  code.
+
+Sat Nov  9 09:04:48 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_marks_test): Bugfix. Fix a struct member name for build
+	  with RGENGC_CHECK_MODE.
+
+Sat Nov  9 08:58:23 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c : Add GC_PROFILE_DETAIL_MEMORY option.
+	  If GC_PROFILE_MORE_DETAIL && GC_PROFILE_DETAIL_MEMORY,
+	  maxrss, minflt and majflt are added to each profile record.
+
+Sat Nov  9 07:41:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (rb_vm_backtrace_object, rb_gc_count): make prototype
+	  declarations, not old-K&R style.
+
+Sat Nov  9 06:11:14 2013  vo.x (Vit Ondruch)  <vondruch@redhat.com>
+
+	* tool/rbinstall.rb (Gem::Specification#collect): make stable
+	  Gem::Specification.files in default .gemspecs the different order of
+	  "files" in .gemspec files makes them different therefore possibly
+	  conflicting in multilib scenario.  patch by vo.x (Vit Ondruch) at
+	  [ruby-core:57544] [Bug #8623].
+
+Sat Nov  9 01:59:18 2013  Aman Gupta <ruby@tmm1.net>
+
+	* ext/objspace/objspace_dump.c: Add experimental methods to
+	  dump objectspace as json: ObjectSpace.dump_all and
+	  ObjectSpace.dump(obj). These methods are useful for debugging
+	  reference leaks and memory growth in large ruby applications.
+	  [Bug #9026] [ruby-core:57893] [Fixes GH-423]
+	* test/objspace/test_objspace.rb: tests for above.
+
+Sat Nov  9 00:26:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (GetLastError): already defined in windows.h on nowadays
+	  cygwin, and caused the confliction with the system provided
+	  definition on cygwin64.  by @kou1okada [Fixes GH-433].
+
+Fri Nov  8 18:35:31 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* lib/open3.rb: receive arguments as keyword arguments.
+
+Fri Nov  8 13:19:26 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* io.c (rb_io_open_with_args): use RARRAY_CONST_PTR().
+
+	* io.c (rb_scan_open_args): use const qualifier for above.
+
+	* io.c (rb_open_file): ditto.
+
+	* io.c (rb_io_open_with_args): ditto.
+
+Fri Nov  8 11:35:06 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* dir.c, pack.c, ruby.c, struct.c, vm_eval.c: use RARRAY_CONST_PTR().
+
+Fri Nov  8 10:58:02 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* compile.c (iseq_build_from_ary_exception): use RARRAY_CONST_PTR().
+
+	* compile.c (iseq_build_from_ary_body): ditto.
+
+Fri Nov  8 10:49:34 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* enumerator.c (append_method): use RARRAY_CONST_PTR().
+
+	* enumerator.c (lazy_init_iterator): ditto.
+
+Fri Nov  8 02:44:29 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (vm_malloc_increase): check GVL before gc_rest_sweep().
+	  vm_malloc_increase() can be called without GVL.
+	  However, gc_rest_sweep() assumes acquiring GVL.
+	  To avoid this problem, check GVL before gc_rest_sweep().
+	  [Bug #9090]
+
+	  This workaround introduces possibility to set malloc_limit as
+	  wrong value (*1). However, this may be rare case. So I commit it.
+
+	  *1: Without rest_sweep() here, gc_rest_sweep() can decrease
+	      malloc_increase due to ruby_sized_xfree().
+
+Fri Nov  8 02:50:25 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/securerandom.rb: [DOC] specify arguments passed to ::random_bytes
+	  By @chastell [Fixes GH-412] https://github.com/ruby/ruby/pull/412
+
+Fri Nov  8 02:43:01 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/objspace/object_tracing.c: [DOC] trace_object_allocations_stop
+	  By @srawlins [Fixes GH-421] https://github.com/ruby/ruby/pull/421
+
+Fri Nov  8 02:34:20 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/net/ftp.rb: [DOC] Document Net::FTP.mdtm and .set_socket and fix
+	  spelling typo, based on patch by @artfuldodger [Fixes GH-426]
+	  https://github.com/ruby/ruby/pull/426
+
+Fri Nov  8 02:14:37 2013  Zachary Scott  <e@zzak.io>
+
+	* array.c: [DOC] Add note about negative indices in Array overview
+	  By @ckaenzig [Fixes GH-427] https://github.com/ruby/ruby/pull/427
+
+Fri Nov  8 02:09:12 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/csv.rb: [DOC] Fix typo in CSV.parse_line by @funky-bibimbap
+	  [Fixes GH-430] https://github.com/ruby/ruby/pull/430
+
+Fri Nov  8 01:01:54 2013  Zachary Scott  <e@zzak.io>
+
+	* golf_prelude.rb: syntax formatting for whitespace [Fixes GH-425]
+	  Patch by @edward https://github.com/ruby/ruby/pull/425
+
+Thu Nov  7 19:36:09 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: modify malloc_limit strategy.
+
+	  * fix default values:
+	    GC_MALLOC_LIMIT_GROWTH_FACTOR
+	    GC_MALLOC_LIMIT: 8MB -> 16MB
+	    GC_MALLOC_LIMIT_MAX: 384MB -> 32MB
+
+	  * algorithm of malloc_limit increment.
+	    if (malloc_increase < malloc_limit) {
+	      next_malloc_limit = malloc_limit * factor
+	      if (malloc_limit > malloc_limit_max) {
+	        malloc_limit = malloc_increase
+	      }
+	    }
+	    This algorithm change malloc_limit from
+	    16MB -> 32MB slowly.
+	    If malloc_limit exceeds malloc_limit_max, then
+	    increase with malloc_increase.
+
+Thu Nov  7 11:06:05 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_shuffle_bang): use RARRAY_PTR_USE() without WB
+	  because there are not new relations.
+
+Thu Nov  7 10:34:12 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_sample): use rb_ary_dup().
+
+Thu Nov  7 09:39:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks_orig): errinfo should not
+	  be propagated to trace blocks so that no argument raise does not
+	  throw internal objects.  [ruby-dev:47793] [Bug #9088]
+
+Wed Nov  6 21:30:55 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_before_sweep): Change algorithm of malloc_limit to
+	  conservative for closing to memory consumption of ruby 2.0.
+
+	* gc.c (GC_MALLOC_LIMIT, GC_MALLOC_LIMIT_GROWTH_FACTOR):
+	  Adjust parameters for new algorithm.
+
+Wed Nov  6 21:16:51 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_shift_m): use RARRAY_PTR_USE() without WB because
+	  there are not new relations.
+
+Wed Nov  6 21:05:20 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_reverse): use RARRAY_PTR_USE().
+
+Wed Nov  6 19:30:44 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* common.mk (help): add texts about gcbench.
+
+Wed Nov  6 16:32:32 2013  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/open3.rb: tweaked grammar in comments
+
+Wed Nov  6 11:46:36 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_sample): use RARRAY_AREF() and RARRAY_PTR_USE()
+	  instead of RARRAY_PTR().
+
+Wed Nov  6 10:37:07 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_and): defer hash creation and some refactoring.
+
+Wed Nov  6 09:14:31 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm1_gc_short_lived.rb: added.
+	  These GC benchmarks do not reflect practical applications.
+	  They are only for tuning.
+
+	* benchmark/bm_vm1_gc_short_with_complex_long.rb: added.
+
+	* benchmark/bm_vm1_gc_short_with_long.rb: added.
+
+	* benchmark/bm_vm1_gc_short_with_symbol.rb: added.
+
+	* benchmark/bm_vm1_gc_wb_ary.rb: added.
+
+	* benchmark/bm_vm1_gc_wb_obj.rb: added.
+
+	* benchmark/bm_vm_thread_queue.rb: added.
+	  This benchmark is added to know how fast C version of thread.so.
+
+Wed Nov  6 09:13:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: define RGENGC_ESTIMATE_OLDSPACE == 0 if USE_RGENGC is 0.
+
+Wed Nov  6 07:13:18 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (Init_GC): add GC::OPTS to show options.
+
+Wed Nov  6 07:12:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: add some options to make quiet.
+
+Wed Nov  6 04:14:25 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: process merge keys before
+	  reviving objects. Fixes GH psych #168
+	* test/psych/test_merge_keys.rb: test for change
+	  https://github.com/tenderlove/psych/issues/168
+
+Tue Nov  5 21:21:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_thread.rb (test_thread_join_in_trap):
+	  Run the test in a different process.
+
+Tue Nov  5 20:14:32 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (is_live_object): A hidden object may be a live object.
+	  [ruby-dev:47788] [Bug #9072]
+
+Tue Nov  5 13:37:19 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add support to estimate increase of oldspace memory usage.
+	  This is another approach to solve an issue discussed at r43530.
+	  This feature is disabled as default.
+
+	  This feature measures an increment of memory consumption by oldgen
+	  objects. It measures memory consumption for each objects when
+	  the object is promoted. However, measurement of memory consumption
+	  is not accurate now. So that this measurement is `estimation'.
+
+	  To implement this feature, move memsize_of() function from
+	  ext/objspace/objspace.c and expose rb_obj_memsize_of().
+
+	  Some memsize() functions for T_DATA (T_TYPEDDATA) have problem to
+	  measure memory size, so that we ignores T_DATA objects now.
+	  For example, some functions skip NULL check for pointer.
+
+	  The macro RGENGC_ESTIMATE_OLDSPACE enables/disables this feature,
+	  and turned off as default.
+
+	  We need to compare 3gen GC and this feature carefully.
+	  (it is possible to enable both feature)
+	  We need a help to compare them.
+
+	* internal.h: expose rb_obj_memsize_of().
+
+	* ext/objspace/objspace.c: use rb_obj_memsize_of() function.
+
+	* cont.c (fiber_memsize): fix to check NULL.
+
+	* variable.c (autoload_memsize): ditto.
+
+	* vm.c (vm_memsize): ditto.
+
+Tue Nov  5 04:03:07 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (GC_MALLOC_LIMIT_MAX): fix default value 512MB -> 384MB.
+	  512MB is huge.
+
+Tue Nov  5 03:31:23 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add 3gen GC patch, but disabled as default.
+
+	  RGenGC is designed as 2 generational GC, young and old generation.
+	  Young objects will be promoted to old objects after one GC.
+	  Old objects are not collect until major (full) GC.
+
+	  The issue of this approach is some objects can promote as old
+	  objects accidentally and not freed until major GC.
+	  Major GC is not frequently so short-lived but accidentally becoming
+	  old objects are not freed.
+
+	  For example, the program "loop{Array.new(1_000_000)}" consumes huge
+	  memories because short lived objects (an array which has 1M
+	  elements) are promoted while GC and they are not freed before major
+	  GC.
+
+	  To solve this problem, generational GC with more generations
+	  technique is known. This patch implements three generations gen GC.
+
+	  At first, newly created objects are "Infant" objects.
+	  After surviving one GC, "Infant" objects are promoted to "Young"
+	  objects.
+	  "Young" objects are promoted to "Old" objects after surviving
+	  next GC.
+	  "Infant" and "Young" objects are collected if it is not marked
+	  while minor GC. So that this technique solves this problem.
+
+	  Representation of generations:
+	  * Infant: !FL_PROMOTED and !oldgen_bitmap [00]
+	  * Young :  FL_PROMOTED and !oldgen_bitmap [10]
+	  * Old   :  FL_PROMOTED and  oldgen_bitmap [11]
+
+	  The macro "RGENGC_THREEGEN" enables/disables this feature, and
+	  turned off as default because there are several problems.
+	  (1) Failed sometimes (Heisenbugs).
+	  (2) Performance down.
+	      Especially on write barriers. We need to detect Young or Old
+	      object by oldgen_bitmap. It is slower than checking flags.
+
+	  To evaluate this feature on more applications, I commit this patch.
+	  Reports are very welcome.
+
+	  This patch includes some refactoring (renaming names, etc).
+
+	* include/ruby/ruby.h: catch up 3gen GC.
+
+	* .gdbinit: fix to show a prompt "[PROMOTED]" for promoted objects.
+
+Tue Nov  5 00:05:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* node.h: catch up comments for last commit.
+
+Tue Nov  5 00:02:00 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: rename FL_OLDGEN to FL_PROMOTED.
+	  This flag represents that "this object is promoted at least once."
+
+	* gc.c, debug.c, object.c: catch up this change.
+
+Mon Nov  4 22:20:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc: Don't use fixed ports: 8070 and 8071.
+
+Mon Nov  4 15:25:52 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc/webrick_testing.rb (start_server): Initialize the server
+	  at main thread to fail early.
+
+Mon Nov  4 10:08:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (TH_EXEC_TAG, TH_JUMP_TAG): get rid of undefined
+	  behavior of setjmp() in rhs of assignment expression.
+	  [ISO/IEC 9899:1999] 7.13.1.1
+
+Sun Nov  3 23:06:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* sample/test.rb: Make temporary file names unique.
+
+Sun Nov  3 20:41:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc: Wrap definitions by TestXMLRPC module.
+
+Sun Nov  3 20:23:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc/webrick_testing.rb (stop_server): Don't try to shutdown
+	  the server if the server is not started.
+
+Sun Nov  3 09:35:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_feature_p): deal with default loadable suffixes.
+
+	* load.c (load_lock): initialize statically linked extensions.
+
+	* load.c (search_required, rb_require_safe): deal with statically
+	  linked extensions.
+
+	* load.c (ruby_init_ext): defer initialization of statically linked
+	  extensions until required actually.  [Bug #8883]
+
+Sat Nov  2 15:14:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/logger.rb (Logger::LogDevice::LogDeviceMutex#lock_shift_log):
+	  open file can't be removed or renamed on Windows.  [ruby-dev:47790]
+	  [Bug #9046]
+
+	* test/logger/test_logger.rb (TestLogDevice#run_children): don't use
+	  fork.
+
+Sat Nov  2 07:08:43 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/logger.rb: Inter-process locking for log rotation
+	  Current implementation fails log rotation on multi process env.
+	  by sonots <sonots@gmail.com>
+	  https://github.com/ruby/ruby/pull/428 fix GH-428 [Bug #9046]
+
+Fri Nov  1 23:24:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (wmap_mark_map): mark live objects only, but delete zombies.
+	  [ruby-dev:47787] [Bug #9069]
+
+Fri Nov  1 22:45:54 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (struct heap_page, gc_page_sweep, gc_sweep): Refactoring for
+	  performance. Add before_sweep condition to heap_page structure.
+
+	* gc.c (rb_gc_force_recycle): Use before_sweep member.
+
+	* gc.c (heap_is_before_sweep, is_before_sweep): Remove. They have not
+	  already been used.
+
+Fri Nov  1 22:20:28 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (make_deferred): Refactoring. Collect codes which should be
+	  atomic.
+
+	* gc.c (make_io_deferred, obj_free, rb_objspace_call_finalizer,
+	  gc_page_sweep): Correspond to the above.
+
+Fri Nov  1 21:40:35 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (typedef struct rb_objspace): Refactoring. Move some members
+	  into profile member.
+
+	* gc.c (newobj_of): Correspond to the above.
+
+	* gc.c (finalize_list): Ditto.
+
+	* gc.c (objspace_live_num): Ditto.
+
+	* gc.c (gc_page_sweep): Ditto.
+
+	* gc.c (rb_gc_force_recycle): Ditto.
+
+	* gc.c (garbage_collect_body): Ditto.
+
+	* gc.c (rb_gc_count): Ditto.
+
+	* gc.c (gc_stat): Ditto.
+
+	* gc.c (gc_prof_set_heap_info): Ditto.
+
+	* gc.c (gc_profile_dump_on): Ditto.
+
+Fri Nov  1 20:53:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_scrub): fix typo, should yield invalid byte
+	  sequence to be scrubbed.  reported by znz at IRC.
+
+Fri Nov  1 17:25:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (is_live_object): finalizer may not run because of lazy-sweep.
+	  [ruby-dev:47786] [Bug #9069]
+
+Fri Nov  1 16:55:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_scrub): export with fixed length arguments, and
+	  allow nil as replacement string instead of omitting.
+
+Fri Nov  1 06:20:44 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_struct): reduce rb_mutex_t size by 8 bytes
+	  on 64bit platform. Patch by Eric Wong. [Feature #9068][ruby-core:58114]
+
+Fri Nov  1 01:08:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: print HWM (high water mark) if possible.
+
+Thu Oct 31 21:48:31 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/streamparser.rb: Add dependency file require.
+	  [Bug #9062] [ruby-dev:47779]
+	  Reported by Ippei Obayashi. Thanks!!!
+
+Thu Oct 31 14:09:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_method_entry_make): fix to pass an ISeq value.
+	  OBJ_WRITTEN() accepts only VALUE.
+
+Wed Oct 30 19:07:57 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el (ruby-brace-to-do-end)
+	  (ruby-do-end-to-brace, ruby-toggle-block): Remove functions that
+	  are already in the latest released version of Emacs (24.3).
+	  [Bug #7565]
+
+Wed Oct 30 12:44:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.status): add missing variables,
+	  PLATFORM_DIR and THREAD_MODEL.
+
+Wed Oct 30 12:20:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (v2w): Normalize a rational value to an integer if possible.
+	  [ruby-core:58070] [Bug #9059] reported by Isaac Schwabacher.
+
+Wed Oct 30 12:08:41 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_uniq_bang): use rb_ary_modify_check() instead of
+	  rb_ary_modify() because the array will be unshared soon.
+
+Wed Oct 30 03:25:10 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: make less garbage when
+	  testing if a string is binary.
+
+Wed Oct 30 03:08:24 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: string subclasses should
+	  not be considered to be binary.  Fixes Psych / GH 166
+	  https://github.com/tenderlove/psych/issues/166
+
+	* test/psych/test_string.rb: test for fix
+
+Tue Oct 29 23:01:18 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_zip): some refactoring.
+
+Tue Oct 29 22:11:37 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_uniq_bang): use st_foreach() instead of for loop.
+
+Tue Oct 29 20:01:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* add RUBY_TYPED_FREE_IMMEDIATELY to data types which only use
+	  safe functions during garbage collection such as xfree().
+
+	  On default, T_DATA objects are freed at same points as finalizers.
+	  This approach protects issues such as reported by [ruby-dev:35578].
+	  However, freeing T_DATA objects immediately helps heap usage.
+
+	  Most of T_DATA (in other words, most of dfree functions) are safe.
+	  However, we turned off RUBY_TYPED_FREE_IMMEDIATELY by default
+	  for safety.
+
+	* cont.c: ditto.
+
+	* dir.c: ditto.
+
+	* encoding.c: ditto.
+
+	* enumerator.c: ditto.
+
+	* error.c: ditto.
+
+	* file.c: ditto.
+
+	* gc.c: ditto.
+
+	* io.c: ditto.
+
+	* iseq.c: ditto.
+
+	* marshal.c: ditto.
+
+	* parse.y: ditto.
+
+	* proc.c: ditto.
+
+	* process.c: ditto.
+
+	* random.c: ditto.
+
+	* thread.c: ditto.
+
+	* time.c: ditto.
+
+	* transcode.c: ditto.
+
+	* variable.c: ditto.
+
+	* vm.c: ditto.
+
+	* vm_backtrace.c: ditto.
+
+	* vm_trace.c: ditto.
+
+	* ext/bigdecimal/bigdecimal.c: ditto.
+
+	* ext/objspace/objspace.c: ditto.
+
+	* ext/stringio/stringio.c: ditto.
+
+	* ext/strscan/strscan.c: ditto.
+
+Tue Oct 29 19:48:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: fix typo (FL_WB_PROTECT -> FL_WB_PROTECTED).
+
+Tue Oct 29 18:45:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (tp_free): removed because empty free function.
+	  Use RUBY_TYPED_NEVER_FREE instead.
+
+Tue Oct 29 18:37:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: introduce new flags for T_TYPEDDATA.
+	  * RUBY_TYPED_FREE_IMMEDIATELY: free the data given by DATA_PTR()
+	    with dfree function immediately.  Otherwise (default), the data
+	    freed at finalization point.
+	  * RUBY_TYPED_WB_PROTECTED: make this object with FL_WB_PROTECT
+	    (not shady).
+
+	* gc.c (obj_free): support RUBY_TYPED_FREE_IMMEDIATELY.
+
+Tue Oct 29 16:49:03 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (vm_malloc_increase): decrease it more carefully.
+
+Tue Oct 29 16:24:52 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_page_resurrect): return a page in tomb heap even if
+	  freelist is NULL.
+
+Tue Oct 29 15:46:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby_atomic.h (ATOMIC_SIZE_CAS): new macro, compare and swap size_t.
+
+Tue Oct 29 12:08:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/readline/readline.c (readline_getc): Consider
+	  NULL as input.
+
+Tue Oct 29 11:10:08 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (gc_profile_total_time): fix off-by-one error in
+	  GC::Profiler.total_time.
+	* test/ruby/test_gc.rb (class TestGc): test for above.
+
+Tue Oct 29 09:53:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* insns.def, vm.c, vm_insnhelper.c, vm_insnhelper.h, vm_method.c: split
+	  ruby_vm_global_state_version into two separate counters - one for the
+	  global method state and one for the global constant state. This means
+	  changes to constants do not affect method caches, and changes to
+	  methods do not affect constant caches. In particular, this means
+	  inclusions of modules containing constants no longer globally
+	  invalidate the method cache.
+
+	* class.c, eval.c, include/ruby/intern.h, insns.def, vm.c, vm_method.c:
+	  rename rb_clear_cache_by_class to rb_clear_method_cache_by_class
+
+	* class.c, include/ruby/intern.h, variable.c, vm_method.c: add
+	  rb_clear_constant_cache
+
+	* compile.c, vm_core.h, vm_insnhelper.c: rename vmstat field in
+	  rb_call_info_struct to method_state
+
+	* vm_method.c: rename vmstat field in struct cache_entry to method_state
+
+Mon Oct 28 23:26:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/readline/test_readline.rb (teardown): Clear Readline.input and
+	  Readline.output.
+
+Mon Oct 28 21:35:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/file/depend, ext/-test-/postponed_job/depend,
+	  ext/-test-/tracepoint/depend: New files for dependencies.
+
+Mon Oct 28 15:32:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/depend (ossl.o): work around of dependency of
+	  thread_native.h, which depends on headers by THREAD_MODEL.
+	  [ruby-dev:47777]
+
+	* ext/openssl/extconf.rb: need THREAD_MODEL.
+
+Mon Oct 28 14:57:01 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (ruby_init_ext): share feature names between frame name and
+	  provided features.
+
+Mon Oct 28 14:41:27 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Import ruby-electric.el 2.1 from
+	  https://github.com/knu/ruby-electric.el.
+
+	  * Hitting the newline-and-indent key within a comment fires
+	    comment-indent-new-line.
+
+	  * Introduce a new feature
+	    `ruby-electric-autoindent-on-closing-char`.
+
+	  * Fix fallback behavior of ruby-electric-space/return that
+	    caused error with auto-complete.
+
+Mon Oct 28 13:17:17 2013  Or Cohen  <orc@fewbytes.com>
+
+	* error.c (name_err_to_s): remove no longer needed overriding, since
+	  r30455 which made exc_to_s almost same.  Fixes [GH-413].
+
+Mon Oct 28 12:42:11 2013  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk, ext/objspace/depend, ext/coverage/depend,
+	  ext/-test-/debug/depend, ext/date/depend: Update dependencies.
+
+Mon Oct 28 09:29:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* vm.c: vm_clear_all_cache is not necessary now we use a 64 bit counter
+	  for global state version.
+
+	* vm_insnhelper.h: ruby_vm_global_state_version overflow is unnecessary
+
+Mon Oct 28 07:47:32 2013  Aman Gupta <ruby@tmm1.net>
+
+	* vm_backtrace.c (rb_profile_frame_classpath): do not use rb_inspect
+	  directly, since it might have a custom implementation or show ivars.
+
+Mon Oct 28 04:10:41 2013  Aman Gupta <ruby@tmm1.net>
+
+	* vm_backtrace.c (rb_profile_frame_classpath): handle singleton
+	  methods defined directly on an object.
+	* test/-ext-/debug/test_profile_frames.rb: test for above.
+
+Mon Oct 28 00:52:36 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (new_struct): fix warning message, class name and encoding.
+
+Sun Oct 27 20:53:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/readline/readline.c: Include ruby/thread.h for
+	  rb_thread_call_without_gvl2.
+	  (readline_rl_instream, readline_rl_outstream): Record FILE
+	  structures allocated by this extension.
+	  (getc_body): New function extracted from readline_getc.
+	  (getc_func): New function.
+	  (readline_getc): Use rb_thread_call_without_gvl2 to invoke getc_func.
+	  [ruby-dev:47033] [Bug #8749]
+	  (clear_rl_instream, clear_rl_outstream): Close FILE structure
+	  allocated by this extension reliably.  [ruby-core:57951] [Bug #9040]
+	  (readline_readline): Use clear_rl_instream and clear_rl_outstream.
+	  (readline_s_set_input): Set readline_rl_instream.
+	  (readline_s_set_output): Set readline_rl_outstream.
+	  (Init_readline): Don't call readline_s_set_input because
+	  readline_getc doesn't block other threads for any FILE structure now.
+
+	  [ruby-dev:47033] [Bug #8749] reported by Nobuhiro IMAI.
+	  [ruby-core:57951] [Bug #9040] reported by Eamonn Webster.
+
+Sat Oct 26 19:31:28 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gc.c: catch up recent changes to compile on GC_DEBUG,
+	  RGENGC_CHECK_MODE.
+
+Sat Oct 26 19:08:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_initialize_copy): disallow to modify after
+	  initialized.
+
+Sat Oct 26 17:48:54 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (meta_add_field): : Re-implemented.
+	   [ruby-core:58017] [Bug #9051] patch by Eamonn Webster.
+
+Sat Oct 26 14:35:09 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_dump_on): use "Page" terminology.
+
+Sat Oct 26 13:25:45 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_sweep, gc_heap_lazy_sweep): fix measurement code.
+	  We only need one sweep time measurement without lazy sweep.
+
+Sat Oct 26 11:59:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* addr2line.c: Include ELF header after system headers (especially
+	  sys/types.h) to avoid compilation failure,
+	  "usr/include/sh3/elf_machdep.h:4:2: error: #error Define _BYTE_ORDER!",
+	  on NetBSD/sh3 (dreamcast, hpcsh, landisk, mmeye).
+
+Sat Oct 26 11:35:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: tuning parameters.
+
+	* gc.c (GC_MALLOC_LIMIT): change default value to 16MB.
+
+	* gc.c (GC_MALLOC_LIMIT_GROWTH_FACTOR): change default value to 2.0.
+
+	* gc.c (gc_before_sweep): change decrease ratio of `malloc_limit'
+	  from 1/4 to 1/10.
+
+Sat Oct 26 11:30:07 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (vm_malloc_increase): do gc_rest_sweep() before GC.
+	  gc_rest_sweep() can reduce malloc_increase, so try it before GC.
+	  Otherwise, malloc_increase can be less than malloc_limit at
+	  gc_before_sweep(). This means that re-calculation of malloc_limit
+	  may be wrong value.
+
+Sat Oct 26 06:35:41 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_before_heap_sweep): Restructure code to mean clearly.
+	  heap->freelist is connected to end of list.
+
+Sat Oct 26 04:01:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_before_heap_sweep): fix freelist management.
+	  After rb_gc_force_recycle() for a object belonging to heap->freelist,
+	  `heap->using_page->freelist' is not null.
+
+Thu Oct 24 21:57:24 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* parse.y: Remove +(binary) and -(binary) special cases
+	  [Feature #9048]
+
+Thu Oct 24 12:45:53 2013  Zachary Scott  <e@zzak.io>
+
+	* object.c: [DOC] Document first argument also takes string for:
+
+	  rb_mod_const_get, rb_mod_const_set, rb_mod_const_defined
+
+	  Also added note about NameError exception for invalid constant name
+
+Thu Oct 24 12:23:58 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_terminate_all): add a comment why we need
+	  state check and call terminate_i again.
+
+Thu Oct 24 12:15:02 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_thread_terminate_all): add a comment why infinite
+	  sleep is safe.
+
+Thu Oct 24 07:41:42 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c: add new initial_growth_max tuning parameter.
+	  [ruby-core:57928] [Bug #9035]
+	* gc.c (heap_set_increment): when initial_growth_max is set,
+	  do not grow number of slots by more than growth_max at a time.
+	* gc.c (rb_gc_set_params): load optional new tuning value from
+	  RUBY_HEAP_SLOTS_GROWTH_MAX environment variable.
+	* test/ruby/test_gc.rb (class TestGc): test for above.
+
+Thu Oct 24 01:34:12 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (rb_infinity_float): suppress overflow in
+	  constant arithmetic warnings.  [ruby-core:57981] [Bug #9044]
+
+Thu Oct 24 00:11:24 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb: raise NoMethodError with a #name and #args.
+	  Raise RuntimeError when modifying frozen instances
+	  instead of TypeError.
+	  (OpenStruct#each_pair): Return an enumerator with size
+	  (OpenStruct#delete): Use the converted argument.
+	  Patches by Kenichi Kamiya. [Fixes GH-383]
+
+	* test/ostruct/test_ostruct.rb: Added tests for above.
+
+Thu Oct 24 00:10:22 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c: Add Array#to_h [Feature #7292]
+
+	* enum.c: Add Enumerable#to_h
+
+Wed Oct 23 23:48:28 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c: Rename free_min to min_free_slots and free_min_page to
+	  max_free_slots. The algorithm for heap growth is:
+	    if (swept_slots < min_free_slots) pages++
+	    if (swept_slots > max_free_slots) pages--
+
+Wed Oct 23 22:51:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): VC 2013 supports C99 mathematics
+	  functions.  [ruby-core:57981] [Bug #9044]
+
+Wed Oct 23 19:13:18 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: move increment from heap to heap_pages.
+	  Share `increment' information with heaps.
+
+	* gc.c: change ratio of heap_pages_free_min_page
+	  to 0.80.
+	  This change means slow down page freeing speed.
+
+Wed Oct 23 17:52:03 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_pages_free_unused_pages): cast to (int) for size_t
+	  variable `i'.
+
+Wed Oct 23 17:39:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: introduce tomb heap.
+	  Tomb heap is where zombie objects and ghost (freed slot) lived in.
+	  Separate from other heaps (now there is only eden heap) at sweeping
+	  helps freeing pages more efficiently.
+	  Before this patch, even if there is an empty page at former phase
+	  of sweeping, we can't free it.
+
+	  Algorithm:
+	    (1) Sweeping all pages in a heap and move empty pages from the
+	        heap to tomb_heap.
+	    (2) Check all existing pages and free a page
+	        if all slots of this page are empty and
+	           there is enough empty slots (checking by swept_num)
+
+	  To introduce this patch, there are several tuning of GC parameters.
+
+Wed Oct 23 14:20:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_prof_sweep_timer_stop): catch up recent changes
+	  to compile on GC_PROFILE_MORE_DETAIL=1.
+
+Wed Oct 23 11:43:27 2013  Zachary Scott  <e@zzak.io>
+
+	* file.c: [DOC] fix rdoc format of File#expand_path from r43386
+
+Tue Oct 22 21:58:28 2013  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* vm_core.h (enum): avoid syntax error.
+
+	* method.h: ditto.
+
+	* internal.h: ditto.
+
+Tue Oct 22 19:53:16 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (Init_heap): move logics from heap_pages_init() and remove
+	  heap_pages_init().
+
+Tue Oct 22 19:19:05 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: allow multiple heaps.
+	  Now, objects are managed by page. And a set of pages is called heap.
+	  This commit supports multiple heaps in the object space.
+
+	  * Functions heap_* and rb_heap_t manages heap data structure.
+	  * Functions heap_page_* and struct heap_page manage page data
+	    structure.
+	  * Functions heap_pages_* and struct rb_objspace_t::heap_pages
+	    maintains all pages.
+	    For example, pages are allocated from the heap_pages.
+
+	  See https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/GC_design
+	  and https://bugs.ruby-lang.org/attachments/4015/data-heap_structure_with_multiple_heaps.png
+	  for more details.
+
+	  Now, there is only one heap called `eden', which is a space for all
+	  new generated objects.
+
+Tue Oct 22 18:26:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb (object_address_group): Use Kernel#to_s to obtain the class
+	  name and object address.
+	  This fix a problem caused by %p in C generates variable length
+	  address.
+	  Reported by ko1 via IRC.
+
+Tue Oct 22 16:57:48 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* file.c (File#expand_path): [DOC] improve documentation of File#expand_path.
+	  Based on patch by Prathamesh Sonpatki. [ruby-core:57734] [Bug #9002]
+
+Tue Oct 22 15:59:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): don't skip current directories if FNM_DOTMATCH
+	  is given.  [ruby-core:53108] [Bug #8006]
+
+Tue Oct 22 14:53:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: exterminate Zombies.
+	  There is a bug that T_ZOMBIE objects are not collected.
+	  Because there is a pass to miss finalizer postponed job
+	  with multi-threading. This patch solve this issue.
+
+	* vm_trace.c (rb_postponed_job_register_one): set
+	  RUBY_VM_SET_POSTPONED_JOB_INTERRUPT(th) if another same job
+	  is registered.
+	  There is a possibility to remain a postponed job without
+	  interrupt flag.
+
+	* vm_trace.c (rb_postponed_job_register_one): check interrupt
+	  carefully.
+
+	* vm_trace.c (rb_postponed_job_register_one): use additional space
+	  to avoid buffer full.
+
+	* gc.c (gc_finalize_deferred_register): check failure.
+
+	* thread.c (rb_threadptr_execute_interrupts): check
+	  `postponed_job_interrupt' immediately.  There is a possibility
+	  to miss this flag.
+
+Tue Oct 22 12:11:16 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if the given CFLAGS and LDFLAGS are working, and
+	  bail out early if not.
+
+Tue Oct 22 00:06:57 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_exists_p): warn deprecated name.  [Bug #9041]
+
+Mon Oct 21 23:57:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (load_encoding): should preserve outer errinfo, so that
+	  expected exception may not be lost.  [ruby-core:57949] [Bug #9038]
+
+Sun Oct 20 15:41:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_reopen): create a new, temporary FD via rb_sysopen and
+	  call rb_cloexec_dup2 on it to atomically replace the file fptr->fd
+	  points to.  This leaves no possible window where fptr->fd is invalid
+	  to userspace (even for any threads running w/o GVL).  based on the
+	  patch by Eric Wong <normalperson@yhbt.net> at [ruby-core:57943].
+	  [Bug #9036]
+
+Sun Oct 20 15:29:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_syserr_fail_path_in): new function split from
+	  rb_sys_fail_path_in to raise SystemCallError without errno.
+
+	* internal.h (rb_syserr_fail_path): like rb_sys_fail_path but without
+	  errno.
+
+Sun Oct 20 13:58:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_obj_wb_unprotect, rb_obj_written),
+	  (rb_obj_write): suppress unused-parameter warnings.
+
+Sun Oct 20 10:32:48 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update RubyGems to master 0886307.  This commit
+	  improves documentation and should bring ruby above 75% documented on
+	  rubyci.
+
+Sun Oct 20 09:30:56 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 3de7e0f.  Changes:
+
+	  Only attempt to build extensions for newly-installed gems.  This
+	  prevents compilation attempts at gem activation time for gems that
+	  already have extensions built.
+
+	  Fix crash in the dependency resolver for dependencies that cannot be
+	  resolved.
+
+	* test/rubygems:  ditto.
+
+Sun Oct 20 05:24:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_class2name): should return real class name, not
+	  singleton class or iclass.
+
+Sun Oct 20 04:18:48 2013  Aman Gupta <ruby@tmm1.net>
+
+	* variable.c (rb_class2name): call rb_tmp_class_path() directly to
+	  avoid extra rb_str_dup() from rb_class_name().
+
+Sat Oct 19 19:59:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (code_page): use simple array instead of st_table.
+
+	* encoding.c (rb_locale_encindex): defer initialization of win32 code
+	  page table until encoding db loaded.
+
+Sat Oct 19 08:25:05 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix rb_objspace_t.
+	  * make "struct heap" and move most of variables
+	    in rb_objspace_t::heap.
+	  * rename rb_objspace_t::heap::sorted to
+	    rb_objspace_t::heap_sorted_pages
+	    and make a macro heap_sorted_pages.
+	  * rename rb_objspace_t::heap::range to
+	    rb_objspace_t::heap_range and rename macros
+	    lomem/himem to heap_lomem/heap_himem.
+
+Sat Oct 19 07:14:40 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 42543b6.  Changes:
+
+	  Fix `gem update` for gems with multiple platforms.
+
+	* test/rubygems:  ditto.
+
+Sat Oct 19 06:55:52 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 0a3814b.  Changes:
+
+	  Fixed extension directory in Gem::Specification#require_paths.
+
+	  Allow installation of gems when $HOME is nonexistent or unwritable.
+
+	  Use proper API in InstallCommand.
+
+	  Improve support for path option in gem dependency files.
+
+	  Remove warnings.
+
+	* test/rubygems:  ditto.
+
+Fri Oct 18 15:23:34 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change terminology of heap.
+	  Change "slot" to "page". "Slot" is a space of RVALUE.
+	  1. "Heap" consists of a set of "heap_page"s (pages).
+	  2. Each "heap_page" has "heap_page_body".
+	  3. "heap_page_body" has RVALUE (a.k.a. "slot") spaces.
+	  4. "sorted" is a sorted array of "heap_page"s, sorted
+	     by address of heap_page_body (for "is_pointer_to_heap").
+
+	  See https://bugs.ruby-lang.org/attachments/4008/data-heap_structure.png.
+
+Fri Oct 18 09:40:43 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master cee6788.  Changes:
+
+	  Fix test failure on vc10-x64 Server on rubyci.org due to attempting
+	  to File.chmod where it is not supported.
+
+	  Continuing work on improved gem dependencies file (Gemfile) support.
+
+	* test:  ditto.
+
+Fri Oct 18 06:02:49 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master f738c67.  Changes:
+
+	  Fixed test bug for ruby with ENABLE_SHARED = no
+
+	* test/rubygems:  ditto.
+
+Fri Oct 18 00:57:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tsort.rb (TSort.tsort): Extracted from TSort#tsort.
+	  (TSort.tsort_each): Extracted from TSort#tsort_each.
+	  (TSort.strongly_connected_components): Extracted from
+	  TSort#strongly_connected_components.
+	  (TSort.each_strongly_connected_component): Extracted from
+	  TSort#each_strongly_connected_component.
+
+Thu Oct 17 18:50:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (CALC_EXACT_MALLOC_SIZE_CHECK_OLD_SIZE): introduced.
+	  This macro enable checker compare with allocated memory and
+	  declared old_size of sized_xfree and sized_xrealloc.
+
+Thu Oct 17 18:45:41 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (STR_HEAP_SIZE): includes TERM_LEN(str).
+
+	* string.c (rb_str_memsize): use STR_HEAP_SIZE().
+
+Thu Oct 17 17:43:00 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): set ci->me to 0 when the
+	  original method of a refined method is undef to avoid SEGV.
+
+	* vm_method.c (rb_method_entry_without_refinements): return 0 when
+	  the original method of a refined method is undef to avoid SEGV.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Thu Oct 17 17:38:36 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, internal.h: rename ruby_xsizefree/realloc to
+	  rb_sized_free/realloc.
+
+	* array.c: catch up these changes.
+
+	* string.c: ditto.
+
+Thu Oct 17 17:32:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c, string.c: use ruby_xsizedfree() and ruby_xsizedrealloc().
+
+	* internal.h (SIZED_REALLOC_N): define a macro as REALLOC_N().
+
+Thu Oct 17 17:11:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (console_emulator_p): check by comparison between
+	  module handle of WriteConsoleW and kernel32.dll.
+
+	* configure.in, win32/Makefile.sub, win32/setup.mak: no longer need
+	  psapi.lib.
+
+Thu Oct 17 16:53:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, internal.h: add new internal memory management functions.
+	  * void *ruby_xsizedrealloc(void *ptr, size_t new_size, size_t old_size)
+	  * void ruby_xsizedfree(void *x, size_t size)
+	  These functions accept additional size parameter to calculate more
+	  accurate malloc_increase parameter which control GC timing.
+	  [Feature #8985]
+
+Thu Oct 17 14:21:34 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (rb_file_expand_path_internal): fix memory leaks at
+	  a non-absolute home exception.
+
+Thu Oct 17 14:06:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/object_tracing.c (newobj_i): fix memory leak.
+	  There is possibility to remain info due to missing FREEOBJ event.
+	  FREEOBJ events are skipped while suppress_tracing state, for example,
+	  during trace events are invoking.
+
+Thu Oct 17 12:30:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tsort.rb (TSort.each_strongly_connected_component_from):
+	  Extracted from TSort#each_strongly_connected_component_from.
+
+Thu Oct 17 11:07:06 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 941c21a.  Changes:
+
+	  Restored method bundler wants to remove for compatibility.
+
+	  Improvements to Gemfile compatibility.
+
+	* test/rubygems:  ditto.
+
+Thu Oct 17 08:08:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/object_tracing.c (newobj_i): add workaround.
+	  some bugs hits this check.
+
+	* ext/objspace/object_tracing.c (object_allocations_reporter_i): cast as pointer.
+
+Thu Oct 17 07:36:53 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 2abce58.  Changes:
+
+	  Fixed documentation generation when sdoc and json are installed as
+	  gems.
+
+	  Added some missing documentation.
+
+Thu Oct 17 07:10:26 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/curses/curses.c: [DOC] Cleaned up formatting consistency of rdoc
+	  comments for Curses, including period spacing and column width.
+
+	  This patch also fixed some typos. Thanks to @postmodern for the patch!
+	  [Fixes GH-420] https://github.com/ruby/ruby/pull/420
+
+Thu Oct 17 06:58:42 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/date/date_core.c: [DOC] plural grammar fixed by @scott113341
+	  Contributed via documenting-ruby.org: documenting-ruby/ruby#16
+	  https://github.com/documenting-ruby/ruby/pull/16
+
+Thu Oct 17 05:52:31 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/io/nonblock/nonblock.c: [DOC] Document io/nonblock by reprah
+	  [Fixes GH-418] https://github.com/ruby/ruby/pull/418 based on the
+	  original discussion from documenting-ruby/ruby#18
+
+Thu Oct 17 05:40:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (objspace_each_objects): do not skip empty RVALUEs.
+
+Thu Oct 17 05:31:31 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c (rb_bug_reporter_add): return simply 0 if failed.
+	  Please check return value.
+
+Thu Oct 17 05:17:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/object_tracing.c: add new method
+	  ObjectSpace.trace_object_allocations_debug_start for GC debugging.
+	  If you encounter the BUG "... is T_NONE" (and so on) on your
+	  application, please try this method at the beginning of your app.
+
+Wed Oct 16 22:35:27 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/io/nonblock/nonblock.c: use rb_cIO instead of VALUE
+
+Wed Oct 16 17:45:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/runner.rb: check nil before calling `signal?'
+	  for a process status.
+
+Wed Oct 16 17:37:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c, internal.h (rb_bug_reporter_add): add a new C-API.
+	  rb_bug_reporter_add() allows to register a function which
+	  is called at rb_bug() called.
+
+	* ext/-test-/bug_reporter/bug_reporter.c: add a test for this C-API.
+
+	* ext/-test-/bug_reporter/extconf.rb: ditto.
+
+	* test/-ext-/bug_reporter/test_bug_reporter.rb: ditto.
+
+Wed Oct 16 15:14:21 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: add a line into NEWS for last commit.
+
+Wed Oct 16 15:09:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c: add a new method `reachable_objects_from_root'.
+	  ObjectSpace.reachable_objects_from_root returns all objects referred
+	  from root (called "root objects").
+	  This feature is for deep object analysis.
+
+	* test/objspace/test_objspace.rb: add a test.
+
+Wed Oct 16 15:00:21 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master b955554.  Changes:
+
+	  Fixed NameError for Gem::Ext due to re-entering file lookup in
+	  RubyGems' overridden require.  Bug by Koichi Sasada.
+
+	  Fixed possible circular require warning in tests.
+
+	  Used existing constant for `gem install -g` dependency file list.
+
+	* test/rubygems:  ditto.
+
+Wed Oct 16 09:42:42 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 278d00d.  Changes:
+
+	  Fixes building extensions without a "clean" make rule
+
+	  Adds gem dependency file autodetection to "gem install -g"
+
+	* test/rubygems:  Tests for the above.
+
+Wed Oct 16 09:12:23 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master commit 2a74263.  This fixes
+	  several bugs in RubyGems 2.2.0.preview.1.
+
+	* test/rubygems:  ditto.
+
+Wed Oct 16 07:25:02 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (gc_mark_roots): rename roots to be categories
+	  instead of function names.
+
+Tue Oct 15 19:18:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.h (rb_objspace_reachable_objects_from_root): added.
+	  This API provides information which objects are root objects.
+	  `category' shows what kind of root objects.
+
+	* gc.c (gc_mark_roots): separate from gc_marks_body().
+
+Tue Oct 15 17:47:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: Fix a typo.  MacOS X doesn't have ENOTSUPP.
+
+Mon Oct 14 12:32:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): load statically linked extensions before
+	  rubygems, because of ext/thread.
+
+	* ruby.c (process_options): use gem_prelude instead of requiring
+	  rubygems directly when --enable=gems is given.
+
+	* Makefile.in (DEFAULT_PRELUDES): always use gem_prelude regardless of
+	  --disable-rubygems.
+
+Mon Oct 14 11:07:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_framework): should append framework options to
+	  $LIBS, not $LDFLAGS.  The former is propagated to exts.mk when
+	  enable-static-linked-ext.
+
+	* lib/mkmf.rb (create_makefile): ranlib on static library, not DLLIB.
+
+Sun Oct 13 23:53:40 2013  Andrew Grimm  <andrew.j.grimm@gmail.com>
+
+	* vsnprintf.c: Fix spelling from compliment to complement.
+	  Patch by @agrimm.
+
+	* include/ruby/ruby.h: ditto
+
+Sun Oct 13 20:59:27 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (Init_BareVM): initialize defined_module_hash here,
+	  Init_top_self() is too late to register core classes/modules.
+
+	* compile.c (compile_array_): no hash to merge if it is empty.
+
+	* vm.c (m_core_hash_merge_kwd): just check keys if only one argument
+	  is given, without merging.
+
+Sat Oct 12 06:35:01 2013-10-11  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake:  Update to rake 10.1.0
+	* bin/rake:  ditto.
+	* test/rake:  ditto.
+
+	* NEWS:  Update NEWS to include rake 10.1.0 and links to release notes.
+
+Sat Oct 12 03:26:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c, variable.c, gc.c (rb_class_tbl): removed.
+
+	* vm.c, vm_core.h (rb_vm_add_root_module): added to register as a
+	  defined root module or class.
+	  This guard helps mark miss from defined classes/modules they are
+	  only referred from C's global variables in C-exts.
+	  Basically, it is extension's bug.
+	  Register to hash object VM has.
+	  Marking a hash objects allows generational GC supports.
+
+	* gc.c (RGENGC_PRINT_TICK): disable (revert).
+
+Sat Oct 12 03:24:49 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_gc_mark_unlinked_live_method_entries):
+	  revert last commit to introduce debug prints.
+
+Fri Oct 11 21:05:19 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h, parse.y: use `full_mark' instead of `full_marking'.
+
+Fri Oct 11 20:58:16 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: use terminology `full_mark' instead of `minor_gc'
+	  in mark functions.
+
+Fri Oct 11 20:46:09 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: use __GNUC__ instead of __GCC__.
+
+Fri Oct 11 20:35:59 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, parse.y: support generational Symbol related marking.
+	  Each symbols has String objects respectively to represent
+	  Symbols.
+	  These objects are marked only when:
+	  * full marking
+	  * new symbols are added
+	  This hack reduce symbols (related strings) marking time.
+	  For example, on my Linux environment, the following code
+	    "20_000_000.times{''}"
+	  with 40k symbols (similar symbol number on Rails 3.2.14 app,
+	  @jugyo tells me) boosts, from 7.3sec to 4.2sec.
+
+	* internal.h: change prototype of rb_gc_mark_symbols().
+
+Fri Oct 11 19:27:22 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Import ruby-electric.el 2.0.1 which fixes
+	  a bug and a flaw with auto-end introduced in the revamp.
+
+	  * ruby-forward-sexp is inappropriate here because it moves the
+	    cursor past the keyword.
+
+	  * Fix a reversed looking-back check in
+	    ruby-electric--block-beg-keyword-at-point-p.
+
+	  * Do not add end again if space or return is hit repeatedly
+	    after a block beginning keyword.
+
+Fri Oct 11 18:12:47 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/gc_hook.c: prohibit reentrant.
+
+Fri Oct 11 18:11:34 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_postponed_job_flush): fix bit operation.
+
+Fri Oct 11 17:33:24 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Import ruby-electric.el 2.0 from
+	  https://github.com/knu/ruby-electric.el which integrates changes
+	  from another fork by @qoobaa.
+
+	  * Allow ruby-electric-mode to be disabled by introducing a
+	    dedicated key map.  Electric key bindings are now defined in
+	    ruby-electric-mode-map instead of overwriting ruby-mode-map.
+
+	  * Add ruby-electric-mode-hook.
+
+	  * Use a remap in binding ruby-electric-delete-backward-char.
+
+	  * Totally revamp electric keywords and then introduce electric
+	    return.  Modifier keywords are now properly detected making
+	    use of ruby-mode's indentation level calculator, and
+
+	  * block-mid keywords (then, else, elsif, when, rescue and
+	    ensure) also become electric with automatic reindentation.
+
+	  * Add standardized comments for ELPA integration.
+
+	  * Fix interaction with smartparens-mode by disabling its end
+	    keyword completion, since ruby-electric has become more clever
+	    at it.
+
+	  * The custom variable `ruby-electric-keywords` is changed to
+	    `ruby-electric-keywords-alist`, allowing user to fine-grained
+	    configuration.
+
+Fri Oct 11 16:53:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_postponed_job_flush): simplify.
+
+Fri Oct 11 03:36:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_threadptr_execute_interrupts): flush postponed job only
+	  once at last.
+
+	* vm_trace.c (rb_postponed_job_flush): defer calling postponed jobs
+	  registered while flushing to get rid of infinite reentrance of
+	  ObjectSpace.after_gc_start_hook.  [ruby-dev:47400] [Bug #8492]
+
+Thu Oct 10 23:04:00 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_or): remove unused variables.
+
+Thu Oct 10 23:01:16 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_or): use rb_hash_keys().
+
+Thu Oct 10 21:36:16 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_compact_bang): use ary_resize_smaller().
+
+Thu Oct 10 17:25:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_exec): support :b_return event for "lambda{return}.call".
+	  [Bug #8622]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Thu Oct 10 13:52:37 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (postponed_job): use preallocated buffer.
+	  Pre-allocate MAX_POSTPONED_JOB (1024) sized buffer
+	  and use it.
+	  If rb_postponed_job_register() cause overflow, simply it
+	  fails and returns 0.
+	  And maybe rb_postponed_job_register() is signal safe.
+
+	* vm_core.h: change data structure.
+
+Thu Oct 10 11:11:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (Init_VM): hide also the singleton class of frozen-core, not
+	  only frozen-core itself.
+
+Thu Oct 10 06:02:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_rand.rb: fix r43224. local variable `e' is
+	  no longer available.
+
+Thu Oct 10 00:02:35 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* numeric.c (fix_aref): avoid a possible undefined behavior.
+	  1L << 63 on 64-bit platform is undefined, at least, according to
+	  ISO/IEC 9899 (C99) 6.5.7.
+
+Wed Oct  9 23:57:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (id_for_attr): avoid inadvertent symbol creation.
+
+Wed Oct  9 18:03:01 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_attr): preserve encoding of the attribute ID in
+	  error message.
+
+Wed Oct  9 17:40:16 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_fstring): because of lazy sweep, str may be unmarked
+	  already and swept at next time, so mark it for the time being.
+	  [ruby-core:57756]
+
+Wed Oct  9 13:53:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compar.c (cmp_eq): fail if recursion.  [ruby-core:57736] [Bug #9003]
+
+	* thread.c (rb_exec_recursive_paired_outer): new function which is
+	  combination of paired and outer variants.
+
+Wed Oct  9 09:18:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/debug.h,
+	  vm_backtrace.c (rb_profile_frame_full_label): add new C API
+	  rb_profile_frame_full_label() which returns label with
+	  qualified method name.
+	  Note that in future version of Ruby label() may return
+	  same return value of full_label().
+
+	* ext/-test-/debug/profile_frames.c,
+	  test/-ext-/debug/test_profile_frames.rb: fix a test for this change.
+
+
+Wed Oct  9 00:55:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (load_lock): display backtrace to $stderr at circular
+	  require.
+
+	* vm_backtrace.c (rb_backtrace_print_to): new function to print
+	  backtrace to the given output.
+
+Tue Oct  8 21:03:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c, include/ruby/debug.h: add new APIs
+	  * VALUE rb_profile_frame_method_name(VALUE frame)
+	  * VALUE rb_profile_frame_qualified_method_name(VALUE frame)
+
+	* iseq.c (rb_iseq_klass), internal.h: add new internal function
+	  rb_iseq_method_name().
+
+	* ext/-test-/debug/profile_frames.c (profile_frames),
+	  test/-ext-/debug/test_profile_frames.rb: add a test.
+
+Tue Oct  8 16:11:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_uniq): use rb_hash_values(), as well as the case no
+	  block is given.
+
+	* internal.h: define rb_hash_values() as internal API.
+
+Tue Oct  8 13:53:21 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_uniq): use rb_hash_keys().
+
+	* internal.h: define rb_hash_keys() as internal API.
+
+	* hash.c (rb_hash_keys): ditto.
+
+Tue Oct  8 10:56:39 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* cont.c: disable FIBER_USE_NATIVE on GNU/Hurd because it doesn't
+	  support a combination getcontext() and threads. Patch by
+	  Gabriele Giacone (1o5g4r8o@gmail.com). [Bug #8990][ruby-core:57685]
+
+Tue Oct  8 05:58:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.strptime): Time.strptime('0', '%s') returns local
+	  time Time object as Ruby 2.0 and before.
+
+Tue Oct  8 05:40:37 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* .travis.yml:  Rebuild Travis CI's "ruby-head" version on successful
+	  build.  Patch by Konstantin Haase.  [Fixes GH-417]
+	  https://github.com/ruby/ruby/pull/417
+
+Tue Oct  8 04:28:25 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-mode.el: Use preceding-char/following-char
+	  (returning 0 at BOF/EOF) instead of char-before/char-after
+	  (returning nil at BOF/EOF) to avoid error from char-syntax when
+	  at BOF/EOF.
+
+Tue Oct  8 04:12:45 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el (ruby-mode-set-encoding): Add a missing
+	  else clause to unbreak with `cp932`, etc.
+
+	* misc/ruby-mode.el (ruby-mode-set-encoding): Ditto.
+
+Tue Oct  8 03:57:34 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el (ruby-mode-set-encoding): Use
+	  `default-buffer-file-coding-system` if the :prefer-utf-8
+	  property is not available.
+
+	* misc/ruby-mode.el (ruby-mode-set-encoding): Ditto.
+
+	* misc/ruby-additional.el (ruby-encoding-map): Override the
+	  default value.
+
+Tue Oct  8 03:19:19 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el (ruby-mode-set-encoding): Add support
+	  for `prefer-utf-8` which was introduced in Emacs trunk.
+
+	* misc/ruby-additional.el (ruby-encoding-map): Add a mapping from
+	  `japanese-cp932` to `cp932` to fix the problem where saving a
+	  source file written in Shift_JIS twice would end up having
+	  `coding: japanese-cp932` which Ruby could not recognize.
+
+	* misc/ruby-additional.el (ruby-mode-set-encoding): Add support
+	  for encodings mapped to nil in `ruby-encoding-map`.
+
+	* misc/ruby-additional.el (ruby-encoding-map): Map `us-ascii` and
+	  `utf-8` to nil by default, meaning they need not be explicitly
+	  declared in magic comment.
+
+	* misc/ruby-additional.el (ruby-encoding-map): Add type
+	  declaration for better customize UI.
+
+	* misc/ruby-mode.el: Ditto for the above.
+
+Tue Oct  8 00:14:53 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el: Add a standard header and footer,
+	  including (provide 'ruby-additional).
+
+Mon Oct  7 22:52:45 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-space-can-be-expanded-p):
+	  Return nil to avoid "end" insertion when in smartparens-mode
+	  that is configured to insert "end" for the same keyword.
+
+	* misc/ruby-electric.el (ruby-electric-keywords): New custom
+	  variable to replace `ruby-electric-simple-keywords-re` with.
+
+Mon Oct  7 22:52:16 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el: Use preceding-char/following-char
+	  (returning 0 at BOF/EOF) instead of char-before/char-after
+	  (returning nil at BOF/EOF) to avoid error from char-syntax when
+	  at BOF/EOF.
+
+Mon Oct  7 22:45:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (FIBER_USE_NATIVE): split long conditions.
+
+Mon Oct  7 20:29:31 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/time.rb: [DOC] typo in Time.rb overview by @srt32 [Fixes GH-416]
+	  https://github.com/ruby/ruby/pull/416
+
+Mon Oct  7 20:07:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.strptime): Use :offset.
+	  Patch by Felipe Contreras.  [ruby-core:57694]
+
+Mon Oct  7 16:47:27 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/-ext-/debug/test_profile_frames.rb: rename class C to
+	  something long name because one test depends on absence of
+	  class ::C.
+
+Mon Oct  7 16:33:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/-test-/debug/profile_frames.c:
+	  test/-ext-/debug/test_profile_frames.rb: add a test for new C-APIs.
+
+Mon Oct  7 16:12:36 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/debug.h: add backtrace collecting APIs for profiler.
+	  * int rb_profile_frames(int start, int limit, VALUE *buff, int *lines);
+	    Collect information of frame information.
+
+	  * VALUE rb_profile_frame_path(VALUE frame);
+	  * VALUE rb_profile_frame_absolute_path(VALUE frame);
+	  * VALUE rb_profile_frame_label(VALUE frame);
+	  * VALUE rb_profile_frame_base_label(VALUE frame);
+	  * VALUE rb_profile_frame_first_lineno(VALUE frame);
+	  * VALUE rb_profile_frame_classpath(VALUE frame);
+	  * VALUE rb_profile_frame_singleton_method_p(VALUE frame);
+	    Get information about each frame.
+
+	  These APIs are designed for profilers, for example, no object allocation,
+	  and enough information for profilers.
+	  In this version, this API collects only Ruby level frames.
+	  This issue will be fixed after Ruby 2.1.
+
+	* vm_backtrace.c: implement above APIs.
+
+	* iseq.c (rb_iseq_klass): return local_iseq's class.
+
+Mon Oct  7 14:26:01 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: catch up last commit.
+	  Type of return value of rb_iseq_first_lineno() is now VALUE.
+
+	* vm_insnhelper.c (argument_error): ditto.
+
+	* vm_method.c (rb_method_entry_make): ditto.
+
+Mon Oct  7 14:07:45 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c, internal.h: change to public (but internal) functions
+	  * VALUE rb_iseq_path(VALUE iseqval);
+	  * VALUE rb_iseq_absolute_path(VALUE iseqval);
+	  * VALUE rb_iseq_label(VALUE iseqval);
+	  * VALUE rb_iseq_base_label(VALUE iseqval);
+	  * VALUE rb_iseq_first_lineno(VALUE iseqval);
+	  And new (temporary) function:
+	  * VALUE rb_iseq_klass(VALUE iseqval);
+
+	* iseq.c. vm_core.h (int rb_iseq_first_lineno): remove
+	  function `int rb_iseq_first_lineno(const rb_iseq_t *iseq)'.
+	  Use `VALUE rb_iseq_first_lineno(VALUE iseqval)' instead.
+
+	* proc.c. vm_insnhelper.c, vm_method.c: catch up this change.
+
+Sun Oct  6 08:37:39 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/webrick.rb: [DOC] fix grammar in WEBrick overview [Fixes GH-413]
+	  Based on patch by @chastell https://github.com/ruby/ruby/pull/413
+
+Sat Oct  5 11:21:01 2013  Aaron Pfeifer  <aaron.pfeifer@gmail.com>
+
+	* thread.c (terminate_atfork_i): fix locking mutexes not unlocked in
+	  forks when not tracked in thread.  [ruby-core:55102] [Bug #8433]
+
+Fri Oct  4 19:54:09 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/dbm/dbm.c: [DOC] Fix wrong constant name in DBM by @edward
+	  [Fixes GH-409] https://github.com/ruby/ruby/pull/409
+
+Fri Oct  4 19:49:42 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c: rename heap.free_num as heap.swept_num to clarify meaning and
+	  avoid confusion with objspace_free_num().
+
+Fri Oct  4 19:02:01 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (objspace_free_num): new method for available/free slots on
+	  heap. [ruby-core:57633] [Bug #8983]
+	* gc.c (gc_stat): change heap_free_num definition to use new method.
+	* test/ruby/test_gc.rb: test for above.
+
+Fri Oct  4 18:53:42 2013  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c: add rb_objspace.limit to keep accurate count of total heap
+	  slots [ruby-core:57633] [Bug #8983]
+
+Fri Oct  4 09:32:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/csv.rb (CSV.foreach): support enumerator.  based on a patch by
+	  Hanmac (Hans Mackowiak) at [ruby-core:57643].  [ruby-core:57283]
+	  [Feature #8929]
+
+Thu Oct  3 18:20:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (console_emulator_p, constat_handle): disable built-in
+	  console colorizing when console-emulator-like DLL is injected.
+	  [Feature #8201]
+
+Thu Oct  3 18:01:44 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: define gc_profile_record::allocated_size if
+	  CALC_EXACT_MALLOC_SIZE is true.
+
+Thu Oct  3 13:42:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (yes-test-sample): use RUNRUBY instead of MINIRUBY to set
+	  runtime library path and run the built ruby.  [Bug #8971]
+
+Thu Oct  3 00:17:15 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-additional.el: Properly quote the body.  An unquoted
+	  body given to eval-after-load is evaluated immediately!
+
+Wed Oct  2 21:38:30 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/socket/ifaddr.c (rsock_getifaddrs): fix possible memory leak.
+	  When a system had no interface, this function used xmalloc for root
+	  but did not return any reference to it.  This patch fixes it by
+	  immediately returning an empty array if no interface is found.
+	  Coverity Scan found this bug.
+
+Wed Oct  2 21:37:04 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* random.c (make_seed_value): a local array declaration was accessed
+	  out of scope.  Coverity Scan found this bug.
+
+Wed Oct  2 18:52:40 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: relax GC condition due to malloc_limit.
+
+	* gc.c (GC_MALLOC_LIMIT_MAX): change default value
+	  (256MB -> 512MB) and permit zero to ignore max value.
+
+	* gc.c (vm_malloc_increase, vm_xrealloc): do not cause GC on realloc.
+
+	* gc.c (gc_before_sweep): change debug messages.
+
+Wed Oct  2 16:26:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_close_read): duplex IO should wait its child process
+	  even after close_read.
+
+Wed Oct  2 15:39:13 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_core.h: use __has_attribute() instead of __clang__major__ because
+	  clang says "Note that marketing version numbers should not be used
+	  to check for language features, as different vendors use different
+	  numbering schemes. Instead, use the Feature Checking Macros."
+	  http://clang.llvm.org/docs/LanguageExtensions.html
+
+Wed Oct  2 14:19:57 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_close_write): detach tied IO for writing before closing
+	  to get rid of race condition.  [ruby-list:49598]
+
+	* io.c (rb_io_close_read): keep fptr in write_io to be discarded, to
+	  fix freed pointer access when it is in use by other threads, and get
+	  rid of potential memory/fd leak.
+
+Tue Oct  1 23:44:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* vm_core.h: use __attribute__((unused)) in UNINITIALIZED_VAR on clang
+	  4.0+ instead of just on 4.2. Clang has supported the unused attribute
+	  since before version 4, so this should be safe.
+
+Tue Oct  1 22:03:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#unlink): finalizer is no longer needed
+	  after unlinking.  patched by by normalperson (Eric Wong) at
+	  [ruby-core:56521] [Bug #8768]
+
+Tue Oct  1 20:54:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (stat_new_0): constify.
+
+	* file.c (rb_stat_new): constify and export.  based on a patch by
+	  Hanmac (Hans Mackowiak) at [ruby-core:53225].  [Feature #8050]
+
+Tue Oct  1 16:03:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ruby_safe_level_4_warning): needed by extension
+	  libraries which check safe level 4.  [ruby-dev:47517] [Bug #8652]
+
+Mon Sep 30 23:14:36 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/objspace/objspace.c: [DOC] Cleaned up many rdoc formatting
+	  issues and several duplicate grammar bugs.
+
+Mon Sep 30 23:01:01 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/objspace/object_tracing.c: [DOC] Adjust rdoc formatting and fix
+	  small grammar typo
+
+Mon Sep 30 17:28:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/object_tracing.c: [DOC] add some notes for
+	  ObjectSpace::trace_object_allocations.
+
+Mon Sep 30 16:46:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/object_tracing.c: add new 3 methods to control tracing.
+	  * ObjectSpace::trace_object_allocations_start
+	  * ObjectSpace::trace_object_allocations_stop
+	  * ObjectSpace::trace_object_allocations_clear
+	  And some refactoring.
+
+	* test/objspace/test_objspace.rb: add a test for new methods.
+
+	* NEWS: add a description for new methods.
+
+Mon Sep 30 11:18:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_disable): do rest_sweep() before disable GC.
+	  This fix may solve a failure of
+	  TestTracepointObj#test_tracks_objspace_events
+	  [test/-ext-/tracepoint/test_tracepoint.rb:43].
+
+Mon Sep 30 10:40:20 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_undef): raise a NameError if the original method
+	  of a refined method is not defined.
+
+	* vm_insnhelper.c (rb_method_entry_eq): added NULL check to avoid SEGV.
+
+	* test/ruby/test_refinement.rb: related test.
+
+Sun Sep 29 23:45:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_id_attrset, intern_str): allow junk attrset ID for
+	  Struct.
+
+	* parse.y (rb_id_attrset): fix inconsistency with literals, allow
+	  ID_ATTRSET and return it itself, but ID_JUNK cannot make ID_ATTRSET.
+	  and raise a NameError instead of rb_bug() for invalid argument.
+
+Sun Sep 29 18:45:05 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_insnhelper.c (vm_callee_setup_arg_complex, vm_yield_setup_block_args):
+	  clear keyword arguments to prevent GC bug which occurs
+	  while marking VM stack.
+	  [ruby-dev:47729] [Bug #8964]
+
+	* test/ruby/test_keyword.rb: tests for the above.
+
+Sat Sep 28 23:25:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (math_log, math_log2, math_log10): fix for Bignum argument.
+	  numbits should be add only when right shifted.
+
+Sat Sep 28 14:30:29 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/dl/test_base.rb: {libc, libm} detection now handle GNU/Hurd
+	  correctly. Patch by Gabriele Giacone (1o5g4r8o@gmail.com).
+	  [Bug #8937][ruby-core:57311]
+	* test/fiddle/helper.rb: ditto.
+
+Sat Sep 28 00:19:41 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/curses/extconf.rb: check the size of chtype.
+
+	* ext/curses/curses.c (NUM2CH, CH2NUM): use proper macros for
+	  the size of chtype.
+
+	[ruby-core:56090] [Bug #8659]
+
+Fri Sep 27 18:33:23 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add two GC tuning environment variables.
+	  RUBY_GC_MALLOC_LIMIT_MAX and RUBY_GC_MALLOC_LIMIT_GROWTH_FACTOR.
+	  See r43067 for details.
+
+	* gc.c (rb_gc_set_params): refactoring. And change verbose notation.
+	  Mostly duplicated functions get_envparam_int/double is not cool.
+	  Please rewrite it.
+
+	* test/ruby/test_gc.rb: fix a test for this change.
+
+Fri Sep 27 17:44:41 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (GC_MALLOC_LIMIT): 8,000,000 -> 8 * 1,024 * 1,024.
+
+Fri Sep 27 17:19:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_before_sweep): cast to size_t to suppress warnings.
+
+Fri Sep 27 17:07:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add some fine-grained profiling codes to tuning marking phase.
+	  If you enable RGENGC_PRINT_TICK to 1, then profiling results by RDTSC
+	  (on x86/amd64 environment) are printed at last.
+	  Thanks Yoshii-san.
+
+Fri Sep 27 16:32:27 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: simplify threshold of GC caused by malloc_increase.
+	  Now, malloc_limit is increased/decreased by mysterious logic.
+	  This fix simplify malloc_limit increase/decrease logic such as:
+	    if (malloc_increase > malloc_limit) /* so many malloc */
+	      malloc_limit += malloc_limit * (GC_MALLOC_LIMIT_FACTOR-1);
+	    else
+	      malloc_limit -= malloc_limit * (GC_MALLOC_LIMIT_FACTOR-1)/4;
+	  Default value of GC_MALLOC_LIMIT_FACTOR is 1.8.
+	  malloc_limit is bounded by GC_MALLOC_LIMIT_MAX (256MB by default).
+	  This logic runs at gc_before_sweep(). So there are no effect from
+	  caused by lazy sweep. And we can remove malloc_increase2.
+
+	* gc.c (HEAP_MIN_SLOTS, FREE_MIN, HEAP_GROWTH_FACTOR): rename to
+	  GC_HEAP_MIN_SLOTS, GC_FREE_MIN, GC_HEAP_GROWTH_FACTOR respectively.
+	  Check them by `#ifndef' so you can specify these values outside gc.c.
+
+	* gc.c (ruby_gc_params_t): add initial_malloc_limit_factor and
+	  initial_malloc_limit_max.
+
+	* gc.c (vm_malloc_prepare, vm_xrealloc): use vm_malloc_increase to
+	  add and check malloc_increase.
+
+Fri Sep 27 01:05:00 2013  Zachary Scott  <e@zzak.io>
+
+	* re.c: [DOC] arguments of Regexp::union receive #to_regexp [Bug #8205]
+
+Fri Sep 27 00:39:27 2013  Zachary Scott  <e@zzak.io>
+
+	* struct.c: [DOC] grammar of ArgumentError in Struct.new [Bug #8936]
+	  Patch by Prathamesh Sonpatki
+
+Thu Sep 26 22:11:56 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/bigdecimal/bigdecimal.c: [DOC] several fixes by @chastell
+	  This includes fixing the capitalization of Infinity, return value of
+	  example "BigDecimal.new('NaN') == 0.0", and code style in example.
+	  [Fixes GH-398] https://github.com/ruby/ruby/pull/398
+
+Thu Sep 26 22:08:11 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/observer.rb: [DOC] syntax improvement in example by @chastell
+	  [Fixes GH-400] https://github.com/ruby/ruby/pull/400
+
+Thu Sep 26 22:03:15 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/digest/digest.c: [DOC] typo in overview by @chastell
+	  [Fixes GH-399] https://github.com/ruby/ruby/pull/399
+
+Thu Sep 26 22:00:42 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl.c: [DOC] typo in example by @zoranzaric
+	  [Fixes GH-401] https://github.com/ruby/ruby/pull/401
+
+Thu Sep 26 21:07:49 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-delete-backward-char): Add
+	  support for smartparens-mode.
+
+	* misc/ruby-electric.el (ruby-electric-cua-replace-region-maybe)
+	  (ruby-electric-cua-delete-region-maybe): New functions that
+	  combine `ruby-electric-cua-*-region` with
+	  `ruby-electric-cua-*-region-p`, using a slightly better way to
+	  detect if it is in cua-mode.
+
+Thu Sep 26 16:51:00 2013  Shota Fukumori  <her@sorah.jp>
+
+	* insns.def (opt_regexpmatch2): Check String#=~ hasn't overridden
+	  before calling rb_reg_match().
+
+	* test/ruby/test_string.rb: Test for above.
+
+	* vm.c (vm_init_redefined_flag): Add BOP flag for String#=~
+
+	[ruby-core:57385] [Bug #8953]
+
+Thu Sep 26 16:43:42 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Avoid use of the interactive function
+	  `self-insert-command` which fires `post-self-insert-hook` and
+	  `post-command-hook`, to make the ruby-electric commands work
+	  nicely with those minor modes that make use of them to do
+	  similar input assistance, such as electric-pair-mode,
+	  autopair-mode and smartparens-mode.
+
+Thu Sep 26 16:24:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* insns.def (opt_regexpmatch1): check Regexp#=~ is not defined before
+	  calling rb_reg_match()
+
+	* test/ruby/test_regexp.rb: add test
+
+	* vm.c (ruby_vm_redefined_flag): change type to short[]
+
+	* vm.c (vm_redefinition_check_flag): return REGEXP_REDEFINED_OP_FLAG if
+	  klass == rb_cRegexp
+
+	* vm.c (vm_init_redefined_flag): setup BOP flag for Regexp#=~
+
+	* vm_insnhelper.h: add REGEXP_REDEFINED_OP_FLAG
+
+	[ruby-core:57385] [Bug #8953]
+
+Thu Sep 26 14:46:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (mark_locations_array): disable AddressSanitizer.  based on a
+	  patch by halfie (Ruby Guy) at [ruby-core:57372].
+	  [ruby-core:56155] [Bug #8680]
+
+Wed Sep 25 17:41:29 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* README.EXT, README.EXT.ja: remove description of RARRAY_PTR()
+	  and add a caution of accessing internal data structure directly.
+	  Also add a description of rb_ary_store().
+	  [Bug #8399]
+
+Wed Sep 25 17:12:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: rename RARRAY_RAWPTR() to RARRAY_CONST_PTR().
+	  RARRAY_RAWPTR(ary) returns (const VALUE *) type pointer and
+	  usecase of this macro is not acquire raw pointer, but acquire
+	  read-only pointer. So we rename to better name.
+	  RSTRUCT_RAWPTR() is also renamed to RSTRUCT_CONST_PTR()
+	  (I expect that nobody use it).
+
+	* array.c, compile.c, cont.c, enumerator.c, gc.c, proc.c, random.c,
+	  string.c, struct.c, thread.c, vm_eval.c, vm_insnhelper.c:
+	  catch up this change.
+
+Wed Sep 25 16:58:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (rb_float_value, rb_float_new): move inline functions
+	  from ruby/ruby.h.
+
+	* numeric.c (rb_float_value, rb_float_new): define external functions
+	  for extension libraries.
+
+Wed Sep 25 15:37:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/rdoc/test_rdoc_generator_darkfish.rb: add a guard for windows.
+
+Wed Sep 25 09:53:11 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Fix CVE-2013-4363.  Miscellaneous minor improvements.
+
+	* test/rubygems:  Tests for the above.
+
+Tue Sep 24 17:38:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_inspect): get rid of out-of-bound access.
+
+	* string.c (rb_str_inspect): when a UTF-16/32 string doesn't have a
+	  BOM, inspect as a dummy encoding string.
+
+Tue Sep 24 17:15:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/encdb.c (ENC_DUMMY_UNICODE): make BOM-encodings dummy.
+
+	* encoding.c (enc_autoload): keep dummy encodings dummy.
+
+Tue Sep 24 16:41:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry#write): data size
+	  is in bytes, not chars.  terminators should be placed automatically.
+
+Tue Sep 24 16:39:36 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry#each_value): encode
+	  name.
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry#each_key): ditto.
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry#export_string):
+	  encode to locale encoding if default internal is not set.
+
+Tue Sep 24 16:35:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry::API#EnumKey):
+	  size of the name is in WCHARs, not in bytes.
+
+Tue Sep 24 14:07:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* gc.c (free_method_cache_entry_i): unused function
+
+	* gc.c (rb_free_mc_table): ditto
+
+	* internal.h (method_cache_entry_t): unused struct
+
+	* vm_method.c (verify_method_cache): remove unused variable
+
+	* vm_method.c (rb_method_entry): ditto
+
+Tue Sep 24 14:01:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* class.c (class_alloc): remove mc_tbl
+
+	* gc.c (obj_free): ditto
+
+	* internal.h (struct rb_classext_struct): ditto
+
+	* method.h (rb_method_entry): remove ent param
+
+	* vm_method.c: restore the global method cache. Per class cache tables
+	  turned out to be far too slow.
+
+	  [ruby-core:57289] [Bug #8930]
+
+Tue Sep 24 12:51:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry::API): need
+	  Constants.
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry::API#EnumValue):
+	  size of the name is in WCHARs, not in bytes.
+
+Mon Sep 23 22:16:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/encdb.c, enc/utf_16_32.h (ENC_DUMMY_UNICODE): Unicode with BOM
+	  must be based on big endian variants, so that actual encodings would
+	  work.  [ruby-core:57318] [Bug #8940]
+
+Mon Sep 23 12:11:26 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (env_each_pair): do not call rb_assoc_new() if
+	  it isn't needed.
+
+Mon Sep 23 10:42:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_module.rb (TestModule#test_include_toplevel): test
+	  for top level main.include.  based on a part of the patch by
+	  kyrylo at [GH-395].
+
+Mon Sep 23 05:07:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_ary_cat): move from internal.h, since it
+	  is described in README.EXT.
+
+Sun Sep 22 20:55:20 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_insnhelper.c (vm_make_proc_with_iseq): fix bug message.
+	  This is follow up to changes in r42637.
+
+Sun Sep 22 20:35:38 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/-test-/tracepoint/tracepoint.c (Init_tracepoint): prevent from GC.
+
+Sun Sep 22 19:00:28 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* benchmark/bm_app_answer.rb: revert r42990, benchmark scripts should
+	  be self-contained and avoid dependencies, especially such small one.
+	  See https://github.com/ruby/ruby/pull/393#issuecomment-24861301.
+
+Sat Sep 21 20:11:06 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_fork_internal): remove cloexec setting on pipes
+	  created by rb_cloexec_pipe.  patch by normalperson (Eric Wong) at
+	  [ruby-core:56523].  [Bug #8769]
+
+Sat Sep 21 01:04:25 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/benchmark.rb: [DOC] grammar of Benchmark#bm [Bug #8888]
+	  Patch by Prathamesh Sonpatki
+
+Sat Sep 21 00:50:02 2013  Zachary Scott  <e@zzak.io>
+
+	* enumerator.c: [DOC] Enumerator#each arguments documentation [GH-388]
+	  Patch by @kachick  https://github.com/ruby/ruby/pull/388
+
+Sat Sep 21 00:49:16 2013  Zachary Scott  <e@zzak.io>
+
+	* enum.c: [DOC] Enumerable#to_a accepts arguments [GH-388]
+	  Patch by @kachick https://github.com/ruby/ruby/pull/388
+
+Sat Sep 21 00:47:44 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_conv_enc_opts): make sure to scan coderange to get
+	  rid of unnecessary conversion.
+
+Sat Sep 21 00:21:08 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/lib/openssl/ssl.rb: [DOC] Document OpenSSL::SSLServer
+	  Based on a patch by Rafal Lisowski [Bug #8758]
+
+Fri Sep 20 23:54:03 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/gserver.rb: [DOC] correct gserver.rb license [Bug #8913]
+
+Fri Sep 20 23:48:34 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/psych/yaml/yaml.h: [DOC] merge upstream typo fix by @GreenGeorge
+	  https://github.com/tenderlove/psych/pull/161
+
+Fri Sep 20 23:37:40 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/securerandom.rb: [DOC] SecureRandom.hex length argument
+	  [Fixes GH-394] Patch by @avdi https://github.com/ruby/ruby/pull/394
+
+Fri Sep 20 23:34:48 2013  Zachary Scott  <e@zzak.io>
+
+	* benchmark/bm_app_answer.rb: removed duplicate code [Fixes GH-393]
+	  Patch by @gouravtiwari https://github.com/ruby/ruby/pull/393
+
+Fri Sep 20 23:24:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (btest, btest-ruby, test-knownbug): add $(RUN_OPTS) to
+	  ruby to be run, so that tests are runnable before making exts.
+
+	* common.mk (test-sample): ditto, and use $(MINIRUBY) as rubytest.rb
+	  does not need extension libraries.
+
+	* tool/rubytest.rb: pass $(RUN_OPTS) to testing ruby using --run-opt.
+
+Fri Sep 20 15:01:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (intern_str): sigil only names are junk, at least one
+	  identifier character is needed.  [ruby-dev:47723] [Bug #8928]
+
+	* parse.y (rb_enc_symname_type): fix out of bound access.
+
+Fri Sep 20 14:14:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/printf/printf.c (printf_test_call): Fix an end of buffer
+	  argument.
+
+Thu Sep 19 16:59:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lambda): adjust position to the beginning of the block.
+
+Thu Sep 19 16:25:06 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): initialize cp so that size is 0 in the
+	  commented case.  fix an accidental bug at r16716.
+
+Thu Sep 19 14:33:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: add a news for r42974.
+
+Thu Sep 19 14:12:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: make Symbol objects frozen.
+	  [Feature #8906]
+	  I want to freeze this good day, too.
+
+	* test/ruby/test_eval.rb: catch up this change.
+
+	* test/ruby/test_symbol.rb: add a test to check frozen symbols.
+
+Thu Sep 19 09:11:33 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS:  Update for RDoc 4.1.0.preview.1 and RubyGems 2.2.0.preview.1
+
+Thu Sep 19 08:59:41 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc/markdown/literals_1_9.rb:  Fix trailing whitespace.
+
+	  Previously kpeg (which generates this file) added trailing
+	  whitespace, but this bug is now fixed.
+
+	* lib/rdoc/markdown.rb:  ditto.
+
+Thu Sep 19 08:33:14 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 4.1.0.preview.1
+
+	  RDoc 4.1.0 contains a number of enhancements including a new default
+	  style and accessibility support.  You can see the changelog here:
+
+	  https://github.com/rdoc/rdoc/blob/v4.1.0.preview.1/History.rdoc
+
+	* test/rdoc:  ditto.
+
+Thu Sep 19 07:16:26 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: updating Psych version
+
+	* ext/psych/psych.gemspec: ditto
+
+Thu Sep 19 06:39:40 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/dependency_resolver.rb:  Switch the iterative resolver
+	  algorithm from recursive to iterative to avoid possible
+	  SystemStackError.
+
+Thu Sep 19 06:29:30 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.2.0.preview.1
+
+	  This brings several new features to RubyGems summarized here:
+
+	  https://github.com/rubygems/rubygems/blob/v2.2.0.preview.1/History.txt
+
+	* test/rubygems:  ditto.
+
+Wed Sep 18 23:14:58 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_enumerate_lines): make String#each_line and
+	  #lines not raise invalid byte sequence error when it is called
+	  with an argument. The patch also causes performance improvement.
+	  [ruby-dev:47549] [Bug #8698]
+
+	* test/ruby/test_m17n_comb.rb (test_str_each_line): remove
+	  assertions which check that String#each_line and #lines will
+	  raise an error if the receiver includes invalid byte sequence.
+
+Wed Sep 18 16:32:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew_from_me): allocate structs after allocated wrapper
+	  object successfully, to get rid of potential memory leak.
+
+Tue Sep 17 15:54:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/shell/command-processor.rb (Shell::CommandProcessor#find_system_command):
+	  return executable file only, should ignore directories and
+	  unexecutable files.  [ruby-core:57235] [Bug #8918]
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert_throw):
+	  assertion for throw.  MiniTest::Assertions#assert_throws discards
+	  the caught value.
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert_nothing_thrown):
+	  returns the result of the given block.
+
+Tue Sep 17 12:55:58 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/regexp.rdoc:  [DOC] Replace paragraphs in verbatim sections with
+	  plain paragraphs to improve readability as ri and HTML.
+
+Mon Sep 16 07:32:35 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: removed meaningless lines.
+	* rational.c: ditto.
+
+Mon Sep 16 00:44:23 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* ext/socket/mkconstants.rb: define MSG_FASTOPEN.
+	  [ruby-core:57138] [Feature #8897]
+
+Sun Sep 15 13:31:23 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* rational.c (nurat_div): reverted r28844, r28886 and r28887.
+	  REASON: Nobuyoshi Nakada  <nobu@ruby-lang.org>'s commits are buggy.
+	  So Rational#/ may produce exact number with inexact number.
+	  Moreover, without reducing.
+	  REALLY NONSENSE COMMITS.
+	  A bug report by me [ruby-dev:44710] is also caused by this behavior.
+	  Kenta Murata  <mrkn@mrkn.jp> patched it up.
+	  But he did not fix the origin.
+	  Today, the bug is still alive in ruby 1.9.3 and 2.0.0.
+
+Sat Sep 14 06:08:10 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* dir.c (dir_s_glob):  [DOC] Improve wording and layout.
+
+	* dir.c (file_s_fnmatch):  ditto.
+
+	* dir.c (Init_Dir):  [DOC] Document File::Constants::FNM_XXX
+	  constants.  (These won't show up in RDoc until a new RDoc is
+	  imported.)
+
+Thu Sep 12 14:58:58 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic.find_proxy): return nil if
+	  http_proxy environment variable is empty string.
+	  [ruby-core:57140] [Bug #8898]
+
+Fri Sep 13 10:40:28 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.1.3
+
+	  Fixed installing platform gems
+
+	  Restored concurrent requires
+
+	  Fixed installing gems with extensions with --install-dir
+
+	  Fixed `gem fetch -v` to install the latest version
+
+	  Fixed installing gems with "./" in their files entries
+
+	* test/rubygems/test_gem_package.rb:  Tests for the above.
+
+	* NEWS:  Updated for RubyGems 2.1.3
+
+Thu Sep 12 22:40:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIGNEDNESS): macro to check signedness of a
+	  type.
+
+	* configure.in (size_t): must be unsigned.
+	  [ruby-core:57149] [Feature #8890]
+
+Thu Sep 12 22:37:08 2013  Anton Ovchinnikov  <revolver112@gmail.com>
+
+	* ext/bigdecimal/bigdecimal.c, ext/digest/md5/md5.c,
+	  ext/json/fbuffer/fbuffer.h, ext/json/generator/generator.c:
+	  Eliminate less-than-zero checks for unsigned variables.
+	  According to section 4.1.5 of C89 standard, size_t is an unsigned
+	  type.  These checks were found with 'cppcheck' static analysis tool.
+	  [ruby-core:57117] [Feature #8890]
+
+Thu Sep 12 21:35:46 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* Makefile.in (libruby-static.a): change LDFLAGS order. LDFLAGS may
+	  include library path that should be specified before LIBS.
+	  [ruby-dev:47707] [Bug #8901]
+
+Thu Sep 12 20:07:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (MAXEXP, MAXFRACT): calculate depending on constants in
+	  float.h.
+
+	* vsnprintf.c (BSD_vfprintf): limit length for cvt() to get rid of
+	  buffer overflow.  [ruby-core:57023] [Bug #8864]
+
+	* vsnprintf.c (exponent): make expbuf size more precise.
+
+Wed Sep 11 17:30:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUNRUBY): append -- only after runruby.rb, not
+	  cross-compiling baseruby, so that $(RUN_OPT) can be command line
+	  options.  [ruby-dev:47703] [Bug #8893]
+
+Wed Sep 11 07:55:17 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (rb_mutex_unlock): Mutex#unlock no longer raise
+	  an exception even if uses on trap. [Bug #8891]
+
+Tue Sep 10 14:37:01 2013  Shota Fukumori  <sorah@tubusu.net>
+
+	* vm_backtrace.c (vm_backtrace_to_ary): Ignore the second argument if
+	  it is nil. [Bug #8884] [ruby-core:57094]
+
+	* test/ruby/test_backtrace.rb (test_caller_with_nil_length):
+	  Test for above.
+
+Tue Sep 10 12:39:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (method_entry_i): should exclude refined methods from
+	  instance method list.  [ruby-core:57080] [Bug #8881]
+
+Tue Sep 10 12:05:04 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* io.c (rb_f_printf): [DOC] add missing parenthesis in rdoc.
+
+Tue Sep 10 10:08:00 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: Update RubyGems note.
+
+Tue Sep 10 09:51:22 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.1.0.  Fixes CVE-2013-4287.
+
+	  See http://rubygems.rubyforge.org/rubygems-update/CVE-2013-4287_txt.html
+	  for CVE information.
+
+	  See http://rubygems.rubyforge.org/rubygems-update/History_txt.html#label-2.1.0+%2F+2013-09-09
+	  for release notes.
+
+	* test/rubygems:  Tests for the above.
+
+Mon Sep  9 21:31:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: Remove spaces between SI prefix and unit to follow
+	  SI brochure.
+	  http://www.bipm.org/en/si/si_brochure/
+	  https://www.nmij.jp/library/units/si/
+
+	* time.c: Ditto.
+
+	* ext/socket/ancdata.c: Ditto.
+
+Mon Sep  9 16:55:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_add_refined_method_entry): clear cache in the
+	  refined class since refining a method entry is modifying the class.
+	  [ruby-core:57079] [Bug #8880]
+
+Mon Sep  9 09:14:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (Gem::Specification#initialize): default date to
+	  RUBY_RELEASE_DATE.  [ruby-core:57072] [Bug #8878]
+
+	* tool/rbinstall.rb (Gem::Specification#to_ruby): add date.
+
+Sun Sep  8 16:01:54 2013  Tanaka Akira  <akr@fsij.org>
+
+	* rational.c (f_gcd): Relax the condition to use GMP.
+
+Sun Sep  8 13:56:38 2013  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (folevariant_initialize): check type of
+	  element of array.
+
+	* test/win32ole/test_win32ole_variant.rb (test_s_new_ary): ditto.
+
+Sat Sep  7 21:33:10 2013  Tanaka Akira  <akr@fsij.org>
+
+	* math.c (math_log): Test the sign for bignums.
+	  (math_log2): Ditto.
+	  (math_log10): Ditto.
+
+Sat Sep  7 20:25:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* math.c (math_log): Support bignums bigger than 2**1024.
+	  (math_log2): Ditto.
+	  (math_log10): Ditto.
+
+Sat Sep  7 15:36:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* vm_eval.c (vm_call0): fix prototype, the id parameter should be of
+	  type ID, not VALUE
+
+	* vm_insnhelper.c (check_match): the rb_funcall family of functions
+	  does not care about refinements. We need to use
+	  rb_method_entry_with_refinements instead to call === with
+	  refinements. Thanks to Jon Conley for reporting this bug.
+	  [ruby-core:57051] [Bug #8872]
+
+	* test/ruby/test_refinement.rb: add test
+
+Sat Sep  7 13:49:40 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* variable.c (classname): the name of class that has
+	  non class id should not be nil. This bug was introduced
+	  in r36577.
+
+	* test/thread/test_cv.rb: test for change.
+
+Sat Sep  7 13:29:22 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/find.rb (Find.find): respect the encodings of arguments.
+	  [ruby-dev:47530] [Feature #8657]
+
+	* test/test_find.rb: add tests.
+
+Sat Sep  7 10:40:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (TCP_FASTOPEN): Defined for TCP fast open.
+	  [ruby-core:57048] [Feature #8871] patch by Masaki Matsushita.
+
+Fri Sep  6 23:53:31 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* common.mk: use RUNRUBY instead of MINIRUBY because MINIRUBY can't
+	  require extension libraries. The patch is from nobu
+	  (Nobuyoshi Nakada).
+
+	* ext/thread/extconf.rb: for build ext/thread/thread.c.
+
+	* include/ruby/intern.h: ditto.
+
+	* thread.c: ditto.
+
+	* lib/thread.rb: removed and replaced by ext/thread/thread.c.
+
+	* ext/thread/thread.c: Queue, SizedQueue and ConditionVariable
+	  implementations in C. This patch is based on patches from panaggio
+	  (Ricardo Panaggio) and funny_falcon (Yura Sokolov) and  ko1
+	  (Koichi Sasada). [ruby-core:31513] [Feature #3620]
+
+	* test/thread/test_queue.rb (test_queue_thread_raise): add a test for
+	  ensuring that killed thread should be removed from waiting threads.
+	  It is based on a code by ko1 (Koichi Sasada). [ruby-core:45950]
+
+Fri Sep  6 22:47:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Define ac_cv_func_clock_getres to yes for mingw*.
+
+Fri Sep  6 21:04:10 2013  Tanaka Akira  <akr@fsij.org>
+
+	* rational.c: Include gmp.h if GMP is used.
+	  (GMP_GCD_DIGITS): New macro.
+	  (rb_gcd_gmp): New function.
+	  (f_gcd_normal): Renamed from f_gcd.
+	  (rb_gcd_normal): New function.
+	  (f_gcd): Invoke rb_gcd_gmp or f_gcd_normal.
+
+	* internal.h (rb_gcd_normal): Declared.
+	  (rb_gcd_gmp): Ditto.
+
+	* ext/-test-/rational: New directory.
+
+	* test/-ext-/rational: New directory.
+
+Fri Sep  6 14:23:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (clock_getres): required as well as clock_gettime().
+	  [ruby-dev:47699] [Bug #8869]
+
+Fri Sep  6 11:45:27 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_econv_append): new function to append a string data
+	  with converting its encoding.  split from rb_econv_substr_append.
+
+Fri Sep  6 02:37:22 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: use double quotes when
+	  strings start with special characters.
+	  https://github.com/tenderlove/psych/issues/157
+
+	* test/psych/test_string.rb: test for change.
+
+Fri Sep  6 00:05:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rewrite_cref_stack): remove recursion.
+
+Thu Sep  5 18:05:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* string.c (fstring_cmp): take string encoding into account when
+	  comparing fstrings [ruby-core:57037] [Bug #8866]
+
+	* test/ruby/test_string.rb: add test
+
+Thu Sep  5 17:25:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_fstring, rb_str_free): use st_data_t instead of VALUE.
+
+	* string.c (rb_fstring): get rid of duplicating already frozen object.
+
+Thu Sep  5 14:01:22 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/optparse.rb:  The Integer acceptable now allows binary and
+	  hexadecimal numbers per the documentation.  [ruby-trunk - Bug #8865]
+
+	  DecimalInteger, OctalInteger, DecimalNumeric now validate their input
+	  before converting to a number.  [ruby-trunk - Bug #8865]
+
+	* test/optparse/test_acceptable.rb:  Tests for the above, tests for all
+	  numeric acceptables for existing behavior.
+
+Thu Sep  5 13:49:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* include/ruby/ruby.h: add RSTRING_FSTR flag
+
+	* internal.h: add rb_fstring() prototype
+
+	* string.c (rb_fstring): deduplicate frozen string literals
+
+	* string.c (rb_str_free): delete fstrings from frozen_strings table when
+	  they are GC'd
+
+	* string.c (Init_String): initialize frozen_strings table
+
+Thu Sep  5 12:48:00 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* configure.in (with_gmp): set with_gmp no if it is empty.
+
+Thu Sep  5 10:41:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_getivar): use class sequence to check class
+	  identity, instead of pointer + vm state
+
+	* vm_insnhelper.c (vm_setivar): ditto
+
+Thu Sep  5 08:20:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (GMP_DIV_DIGITS): New macro.
+	  (bary_divmod_gmp): New function.
+	  (rb_big_divrem_gmp): Ditto.
+	  (bary_divmod_branch): Ditto.
+	  (bary_divmod): Use bary_divmod_branch.
+	  (bigdivrem): Ditto.
+
+	* internal.h (rb_big_divrem_gmp): Declared.
+
+Thu Sep  5 06:22:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_divmod_normal): Reduce temporary array allocations.
+
+Thu Sep  5 02:17:06 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_divrem_normal): Add GC guards.
+
+Thu Sep  5 00:38:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_divrem_normal): New function.
+
+	* internal.h (rb_big_divrem_normal): Declared.
+
+	* ext/-test-/bignum/div.c: New file.
+
+	* test/-ext-/bignum/test_div.rb: New file.
+
+Thu Sep  5 00:08:44 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_normal): Removed.
+	  (bary_divmod_normal): New function.
+	  (bary_divmod): Use bary_divmod_normal.
+	  (bigdivrem): Use bary_divmod_normal.
+
+Wed Sep  4 23:02:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Useless declaration removed.
+
+Wed Sep  4 22:56:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (NUM_STEP_GET_INF): split from NUM_STEP_SCAN_ARGS(), since
+	  inf is not used in num_step_size().
+
+Wed Sep  4 20:22:43 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_normal): Add assertions.
+
+Wed Sep  4 19:18:40 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (vm_state_version_t): prefer LONG_LONG to uint64_t.
+
+Wed Sep  4 16:28:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (vm_state_version_t): use uint64_t when it is larger than
+	  LONG_LONG, and fallback to unsigned long.
+
+Wed Sep  4 15:37:05 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf8_mac-tbl.rb: fix r42789.
+	  Fix conversion table and logic. [ruby-dev:47680]
+
+Wed Sep  4 14:08:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* class.c, compile.c, eval.c, gc.h, insns.def, internal.h, method.h,
+	  variable.c, vm.c, vm_core.c, vm_insnhelper.c, vm_insnhelper.h,
+	  vm_method.c: Implement class hierarchy method cache invalidation.
+
+	  [ruby-core:55053] [Feature #8426] [GH-387]
+
+Wed Sep  4 11:13:40 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_gsub): use BEG(0) for whole matched position not
+	  return value from rb_reg_search(), for \K matching.
+	  [ruby-dev:47694] [Bug #8856]
+
+Wed Sep  4 11:11:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (SOLIBS): LIBRUBY_SO also needs linking with gmp, to
+	  run worker processes in test-all on non-ELF platforms.
+
+Tue Sep  3 23:01:41 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/parser/test_tree.rb
+	  (TestTreeParser::TestInvalid#test_unmatched_close_tag):
+	  Compute expected value from test value.
+
+Tue Sep  3 22:59:58 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/treeparser.rb (REXML::Parsers::TreeParser#parse):
+	  Add source information to parse exception on no close tag error.
+	  [Bug #8844]  [ruby-dev:47672]
+	  Patch by Ippei Obayashi. Thanks!!!
+	* test/rexml/parser/test_tree.rb: Add a test for the above case.
+
+Tue Sep  3 22:57:57 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/parser/test_tree.rb: Fix test name to describe test
+	  content.
+
+Tue Sep  3 22:54:46 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/treeparser.rb (REXML::Parsers::TreeParser#parse):
+	  Remove needless nested parse exception information.
+	  [Bug #8844]  [ruby-dev:47672]
+	  Reported by Ippei Obayashi. Thanks!!!
+	* test/rexml/parser/test_tree.rb: Add a test for the above case.
+
+Tue Sep  3 22:03:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_enc_str_new_cstr): new function to create a string from
+	  the C-string pointer with the specified encoding.
+
+Tue Sep  3 21:41:37 2013  Akira Matsuda  <ronnie@dio.jp>
+
+	* eval.c (Init_eval): Make Module#include and Module#prepend public
+	  [Feature #8846]
+
+	* test/ruby/test_module.rb (class TestModule): Test for above
+
+Tue Sep  3 21:35:19 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (sys/dyntune.h): for gettune().
+
+	* thread_pthread.c (hpux_attr_getstackaddr): fix missing *.
+	  [ruby-core:56983] [Feature #8793]
+
+Tue Sep  3 20:12:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (GMP_STR2BIG_DIGITS): New macro.
+	  (str2big_gmp): New function.
+	  (rb_cstr_to_inum): Use str2big_gmp for big bignums.
+	  (rb_str2big_gmp): New function.
+
+	* internal.h (rb_str2big_gmp): Declared.
+
+Tue Sep  3 19:44:40 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry#values): added.
+	  [Feature #7763] [ruby-core:51783]
+
+Tue Sep  3 18:26:00 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/inf-ruby.el (inf-ruby-keys, run-ruby): Add magic autoload
+	  comments.
+
+	* misc/rdoc-mode.el (rdoc-mode): Ditto.
+
+	* misc/ruby-electric.el (ruby-electric-mode): Ditto.
+
+	* misc/ruby-style.el (ruby-style-c-mode): Ditto.
+
+Tue Sep  3 17:06:15 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb
+	  (TestRubyOptions::SEGVTest::ExpectedStderr): the URL was changed at
+	  r42800.
+
+Tue Sep  3 14:48:25 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/thread.rb: [DOC] CV#wait typo by @avdi [Fixes GH-386]
+	  https://github.com/ruby/ruby/pull/386
+
+Tue Sep  3 14:37:53 2013  Zachary Scott  <e@zzak.io>
+
+	* error.c: [DOC] Update bug tracker url by @ScotterC [Fixes GH-390]
+	  https://github.com/ruby/ruby/pull/390
+
+Tue Sep  3 12:45:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_str2big_poweroftwo): New function.
+	  (rb_str2big_normal): Ditto.
+	  (rb_str2big_karatsuba): Ditto.
+
+	* internal.h (rb_str2big_poweroftwo): Declared.
+	  (rb_str2big_normal): Ditto.
+	  (rb_str2big_karatsuba): Ditto.
+
+	* ext/-test-/bignum/str2big.c: New file.
+
+	* test/-ext-/bignum/test_str2big.rb: New file.
+
+	* ext/-test-/bignum/depend: Add the dependency for str2big.c.
+
+Tue Sep  3 12:09:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): Support times() based monotonic clock.
+	  (rb_clock_getres): Ditto.
+
+Tue Sep  3 12:03:02 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (str2big_scan_digits): Extracted from rb_cstr_to_inum.
+
+Tue Sep  3 11:23:57 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_select_with_thread): rounding up the fraction of
+	  tv_usec instead of rounding down.
+	  this change is an experiment to get rid of failures on vc10-x64 CI.
+
+Tue Sep  3 11:00:28 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (do_select): constify timeout.
+
+	* win32/win32.c (rb_w32_select_with_thread): constify 10ms wait and
+	  0ms wait structs.
+
+Tue Sep  3 10:03:42 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/openssl/test_pair.rb
+	  (OpenSSL::TestPair#test_write_nonblock_no_exceptions): on some CIs
+	  such as Debian 6.0, Ubuntu 10.04, CentOS and vc10-x64 (maybe depend
+	  on OpenSSL version), writing to SSLSocket after SSL_ERROR_WANT_WRITE
+	  causes SSL_ERROR_SSL "bad write retry".
+
+Tue Sep  3 08:20:46 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/utf8_mac-tbl.rb: update conversion table to recent OS X.
+	  Previous table is used on Mac OS X 10.1 or prior.
+	  This table is used on 10.2 or later. [ruby-dev:47680]
+
+Tue Sep  3 07:49:25 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* numeric.c (NUM_STEP_SCAN_ARGS): On second thought, keep
+	  Numeric#step backward compatible in that it raises TypeError
+	  when nil is given as second argument.
+
+	* test/ruby/test_float.rb (TestFloat#test_num2dbl): Revert.
+
+	* test/ruby/test_numeric.rb (TestNumeric#test_step): Fix test
+	  cases for the above change.
+
+Tue Sep  3 07:39:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bytes_2comp): Define it only for little endian
+	  environment.
+
+Tue Sep  3 07:31:29 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* numeric.c (NUM_STEP_SCAN_ARGS): Numeric#step should raise
+	  TypeError if a non-numeric parameter is given.
+
+	* test/ruby/test_float.rb (TestFloat#test_num2dbl): Allow nil as
+	  step, as with the keyword argument.
+
+	* test/ruby/test_numeric.rb (TestNumeric#test_step): Add tests for
+	  nil as step or limit.
+
+Tue Sep  3 07:28:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (bit_length): Add casts to fix compilation error with
+	  clang 3.0 -Werror,-Wshorten-64-to-32.
+	  [ruby-dev:47687] reported by SASADA Koichi.
+
+Tue Sep  3 03:17:26 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_search_super_method): use ci->argc instead of
+	  ci->orig_argc. ci->argc can be changed by splat arguments.
+	  [ruby-list:49575]
+	  This fix should be applied to Ruby 2.0.0 series.
+
+	* test/ruby/test_super.rb: add a test for above.
+
+Mon Sep  2 23:46:29 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* numeric.c (num_step): Default the limit argument to infinity and
+	  allow it to be omitted.  Keyword arguments (by: and to:) are
+	  introduced for ease of use. [Feature #8838] [ruby-dev:47662]
+	  [ruby-dev:42194]
+
+	* numeric.c (num_step): Optimize for infinite loop.
+
+Mon Sep  2 22:55:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (ISDIGIT): Unused macro removed.
+
+Mon Sep  2 22:49:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (str2big_poweroftwo): Extracted from rb_cstr_to_inum.
+	  (str2big_normal): Ditto.
+	  (str2big_karatsuba): Ditto.
+
+Mon Sep  2 14:39:29 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ruby.c (Process#setproctitle): [DOC] Fix and improve rdoc.
+
+	* ruby.c (Process#argv0): [DOC] Improve rdoc.
+
+Mon Sep  2 14:15:00 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* NEWS: fix description of number literal suffixes.
+
+Mon Sep  2 14:01:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* test/rake/test_rake_rules.rb: add space after string literal to
+	  prevent conflict with string options syntax "foo"opts
+
+	* test/rss/rss-assertions.rb: ditto
+
+Mon Sep  2 12:28:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_bignum.rb (test_interrupt_during_to_s): Disable it
+	  when GMP is used.
+
+Mon Sep  2 07:02:10 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (Init_Bignum): Define Bignum::GMP_VERSION when GMP is used.
+
+Mon Sep  2 01:46:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_generic): Reduce arguments.
+	  (big2str_gmp): Ditto.
+	  (rb_big2str1): Follow the above change.
+
+Mon Sep  2 00:08:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (get_mach_timebase_info): Extracted from rb_clock_gettime.
+	  (rb_clock_gettime): Use get_mach_timebase_info.
+	  (rb_clock_getres): Support MACH_ABSOLUTE_TIME_BASED_CLOCK_MONOTONIC.
+
+Sun Sep  1 23:30:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (GMP_BIG2STR_DIGITS): New constant.
+	  (big2str_gmp): New function.
+	  (rb_big2str1): Use big2str_gmp for big bignums.
+
+	* internal.h (rb_big2str_gmp): Declared.
+
+	* ext/-test-/bignum/big2str.c (big2str_gmp): New method.
+
+Sun Sep  1 22:37:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_gmp): Use mpz_init and mpz_clear instead of
+	  mpz_inits and mpz_clears.
+	  Older GMP don't have them.
+
+Sun Sep  1 21:17:54 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/http/test_http.rb (test_bind_to_local_port): Choose an open
+	  port more reliably.
+
+Sun Sep  1 20:32:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_base_poweroftwo): Renamed from
+	  big2str_base_powerof2.
+	  (rb_big2str_poweroftwo): New function for test.
+	  (big2str_generic): Extracted from rb_big2str1.
+	  (rb_big2str_generic): New function for test.
+
+	* internal.h (rb_big2str_poweroftwo): Declared.
+	  (rb_big2str_generic): Ditto.
+
+	* ext/-test-/bignum/big2str.c: New file.
+
+	* test/-ext-/bignum/test_big2str.rb: New file.
+
+Sun Sep  1 15:21:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_2bdigits): Renamed from big2str_orig.
+
+Sun Sep  1 13:02:24 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Remove BITSPERDIG >= INT_MAX test.  The static assertion,
+	  SIZEOF_BDIGITS <= sizeof(BDIGIT) is enough.
+
+Sun Sep  1 11:38:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (maxpow_in_bdigit): Removed.
+
+Sun Sep  1 10:30:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_fix_bit_length): Moved from bignum.c.
+
+Sun Sep  1 09:55:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (bit_length): Moved from bignum.c.
+	  (nlz_int): Ditto.
+	  (nlz_long): Ditto.
+	  (nlz_long_long): Ditto.
+	  (nlz_int128): Ditto.
+
+Sun Sep  1 03:32:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bit_length): Renamed from bitsize.
+
+Sun Sep  1 00:07:09 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_bit_length): New method.
+	  (rb_fix_bit_length): Ditto.
+	  [ruby-core:56247] [Feature #8700]
+
+Sat Aug 31 22:18:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_getres): New method.
+	  (timetick2dblnum_reciprocal): New function.
+
+	* configure.in: Check clock_getres.
+
+	[ruby-core:56780] [Feature #8809] accepted as a CRuby feature at
+	DevelopersMeeting20130831Japan
+	https://bugs.ruby-lang.org/projects/ruby/wiki/DevelopersMeeting20130831Japan
+
+Sat Aug 31 21:02:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Use GMP to accelerate big Bignum multiplication.
+	  (bary_mul_gmp): New function.
+	  (bary_mul): Use bary_mul_gmp.
+	  (bigsq): Use different threshold with GMP.
+
+	* configure.in: Detect GMP.
+
+	[ruby-core:56658] [Feature #8796]
+
+Sat Aug 31 15:03:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* compile.c (NODE_MATCH3): pass CALL_INFO to opt_regexpmatch2
+
+	* insns.def (opt_regexpmatch2): use CALL_SIMPLE_METHOD to call =~ if
+	  the receiver is not a T_STRING [Bug #8847] [ruby-core:56916]
+
+Sat Aug 31 14:07:11 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (random_bytes): Use Process.clock_gettime.
+
+Sat Aug 31 00:25:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_{ascii8bit,utf8,usascii}_encindex): get
+	  rid of conflict with macros defined in internal.h.
+
+Fri Aug 30 22:37:57 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (native_thread_init_stack): wait the creator thread
+	  to fill machine stack info, if get_stack_of() is available.
+
+	* thread_pthread.c (native_thread_create): fill the created thread
+	  stack info after starting, if get_stack_of() is available.
+
+	* thread_pthread.c (native_thread_create): define attr only if it is
+	  used, and merge pthread_create() calls.
+
+	* thread_pthread.c (get_main_stack): separate function to get stack of
+	  main thread.
+
+Thu Aug 29 18:05:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* struct.c (rb_struct_define_without_accessor_under): added.
+	  This function is similar to rb_define_class_under() against
+	  rb_define_class().
+
+	* include/ruby/intern.h: add a declaration of this function.
+
+Thu Aug 29 17:03:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): a method entry refers the based
+	  class/module, so should search superclass from the origin i-class
+	  where the entry belongs to, to get rid of infinite loop when zsuper
+	  in a prepended class/module.  [ruby-core:54105] [Bug #8238]
+
+Thu Aug 29 05:35:58 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/zlib/zlib.c (zstream_run):  Fix handling of deflate streams that
+	  need a dictionary but are being decompressed by Zlib::Inflate.inflate
+	  (which has no option to set a dictionary).  Now Zlib::NeedDict is
+	  raised instead of crashing.  [ruby-trunk - Bug #8829]
+	* test/zlib/test_zlib.rb (TestZlibInflate):  Test for the above.
+
+Thu Aug 29 02:40:45 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: invalid floats should be
+	  treated as strings.
+	  https://github.com/tenderlove/psych/issues/156
+
+	* test/psych/test_string.rb: test for change
+
+Wed Aug 28 17:20:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (hpux_attr_getstackaddr): basic support for the
+	  get_stack() under HP-UX.  based on the patch by michal@rokos.cz
+	  (Michal Rokos) at [ruby-core:56645].  [Feature #8793]
+
+Wed Aug 28 11:24:20 2013  Michal Rokos  <michal@rokos.cz>
+
+	* configure.in (sys/pstat.h): fix missing header check for
+	  missing/setproctitle.c on HP-UX.  [ruby-core:56644] [Bug #8792]
+
+Wed Aug 28 04:54:33 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read):  Replace duplicate
+	  wait_writable with wait_readable.
+
+Tue Aug 27 17:18:40 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout#timeout): skip rescue clause only when no
+	  exception class is given.
+
+Tue Aug 27 17:02:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (copy_stream_body): should write in binary mode.  based on a
+	  patch by godfat (Lin Jen-Shin) at [ruby-core:56556].
+	  [ruby-core:56518] [Bug #8767]
+
+Tue Aug 27 17:02:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (copy_stream_body): move common open flags.
+
+Tue Aug 27 16:56:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_size): use rb_check_funcall() instead of
+	  respond_to? and call.
+
+	* enumerator.c (enumerator_each): ensure that argument array size
+	  does not overflow at appending.
+
+Tue Aug 27 16:46:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_index, rb_ary_rindex): use optimized equality to
+	  improve performance.  [Feature #8820]
+
+	* vm_insnhelper.c (rb_equal_opt): optimized equality function.
+
+Tue Aug 27 16:11:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (opt_eq_func): use RBASIC_CLASS() instead of HEAP_CLASS_OF().
+
+	* insns.def (opt_plus, opt_minus, opt_mult, opt_div, opt_mod, opt_lt),
+	  (opt_gt, opt_ltlt, opt_aref, opt_aset, opt_length, opt_size),
+	  (opt_empty_p, opt_succ): ditto.
+
+Tue Aug 27 16:08:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_check_funcall, rb_check_funcall_with_hook): constify
+	  argv.
+
+Tue Aug 27 13:03:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/stringio/stringio.c (strio_read_nonblock): declare local
+	  variables at the first of function.
+
+Tue Aug 27 11:51:37 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enumerator.c: Allow Enumerator size argument to be any callable.
+	  Patch by Avdi Grimm. [bug #8641] [ruby-core:56032] [fix GH-362]
+
+	* test/ruby/test_enumerator.rb: Test for above
+
+Tue Aug 27 11:46:31 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_clear): do rest_sweep() before clearing
+	  profile.current_record.
+
+Tue Aug 27 07:35:05 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* io.c (io_read_nonblock): support non-blocking reads without raising
+	  exceptions. As in: `io.read_nonblock(size, exception: false)`
+	  [ruby-core:38666] [Feature #5138]
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read_internal): ditto
+	* ext/stringio/stringio.c (strio_sysread): ditto
+	* io.c (rb_io_write_nonblock): support non-blocking writes without
+	  raising an exception.
+	* ext/openssl/ossl_ssl.c (ossl_ssl_write_internal): ditto
+	* test/openssl/test_pair.rb (class OpenSSL): tests
+	* test/ruby/test_io.rb (class TestIO): ditto
+	* test/socket/test_nonblock.rb (class TestSocketNonblock): ditto
+	* test/stringio/test_stringio.rb (class TestStringIO): ditto
+
+Tue Aug 27 05:24:34 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Import RubyGems 2.1.0 Release Candidate
+	* test/rubygems:  ditto.
+
+Mon Aug 26 16:24:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_nextc): warn carriage return in middle of line.
+	  [ruby-core:56240] [Feature #8699]
+
+Mon Aug 26 15:27:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout#timeout): should not be caught by rescue
+	  clause.  [Bug #8730]
+
+Mon Aug 26 14:44:26 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_splice): use RARRAY_PTR_USE() without WB because
+	  there are not new relations.
+
+	* enum.c (enum_sort_by): ditto.
+
+	* struct.c (setup_struct): use RARRAY_RAWPTR().
+
+	* vm_eval.c (yield_under): ditto.
+
+	* ext/pathname/pathname.c (path_entries): use RARRAY_AREF().
+
+	* ext/pathname/pathname.c (path_s_glob): ditto.
+
+Mon Aug 26 13:11:10 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* array.c (ary_ensure_room_for_push): fix typo in r42658.
+
+Mon Aug 26 12:37:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/sizes.c.tmpl: generate automatically by extracting
+	  RUBY_CHECK_SIZEOF from configure.in.
+
+Mon Aug 26 10:16:59 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* process.c (gcd_timetick_int): Renamed from gcd_timtick_int.
+
+Sun Aug 25 21:02:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* sizes.c (Init_sizes): Define the size of clock_t.
+
+Sun Aug 25 01:47:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BARY_SHORT_MUL): Renamed from BARY_MUL1.
+	  (bary_short_mul): Renamed from bary_mul1.
+
+Sat Aug 24 10:35:09 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): The emulated clock names changed.
+
+Fri Aug 23 22:22:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): Add a cast to fix compile error by
+	  -Werror,-Wshorten-64-to-32.
+
+Fri Aug 23 22:12:13 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_intern): no symbol cache while initialization.
+
+Fri Aug 23 22:07:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (clock_t): needs time.h.
+
+Fri Aug 23 21:37:28 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (reduce_factors): New function.
+	  (timetick2dblnum): Use reduce_factors.
+	  (timetick2integer): Ditto.
+	  (make_clock_result): Follow the above change.
+	  (rb_clock_gettime): Ditto.
+
+Fri Aug 23 21:00:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (timetick_int_t): Renamed from timetick_giga_count_t.
+	  (gcd_timtick_int): Renamed from gcd_ul and make the arguments
+	  timetick_giga_count_t.
+	  (reduce_fraction): Make the arguments timetick_int_t.
+	  (timetick2integer): Ditto.
+	  (make_clock_result): Ditto.
+	  (timetick2dblnum): Fix the return type.
+	  (rb_clock_gettime): Use timetick_int_t.
+
+Fri Aug 23 20:50:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (gcd_ul): New function.
+	  (reduce_fraction): Ditto.
+	  (reduce_fraction): Ditto.
+	  (timetick2dblnum): Ditto.
+	  (timetick2integer): Ditto.
+	  (make_clock_result): Use timetick2dblnum and timetick2integer.
+	  (rb_clock_gettime): Follow the make_clock_result change.
+
+Fri Aug 23 18:39:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_make_shared): shared ary as shady. Need more effort to
+	  make it normal object.
+
+	* array.c (rb_ary_modify): use RARRAY_PTR_USE() without WB because
+	  there are not new relations.
+
+	* array.c (ary_ensure_room_for_unshift): use RARRAY_RAWPTR() because
+	  there are not new relations.
+
+Fri Aug 23 11:25:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c: introduce ARY_SHARED_OCCUPIED(shared).
+
+Fri Aug 23 11:07:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): now SIZEOF_CLOCK_T is needed for
+	  unsigned_clock_t.
+
+Thu Aug 22 22:01:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): Strip "s" from unit names.
+
+Thu Aug 22 20:14:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (unsigned_clock_t): Defined.
+	  (rb_clock_gettime): Consider clock_t overflow for
+	  ISO_C_CLOCK_BASED_CLOCK_PROCESS_CPUTIME_ID.
+
+	* configure.in: Check the size of clock_t.
+
+Thu Aug 22 16:22:48 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (build_postexe_iseq): fix to setup the local table.
+
+Thu Aug 22 15:42:43 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (rb_iseq_compile_node): accept NODE_IFUNC to support
+	  custom compilation.
+
+	* compile.c (NODE_POSTEXE): compile to
+	  "ONCE{ VMFrozenCore::core#set_postexe{...} }" with a new custom
+	  compiler `build_postexe_iseq()'.
+
+	* vm.c (m_core_set_postexe): remove parameters (passed by a block).
+
+Thu Aug 22 06:54:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): Change emulation symbols for
+	  Process.clock_gettime.
+
+Thu Aug 22 06:24:54 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (make_clock_result): Extracted from rb_clock_gettime.
+
+Wed Aug 21 22:30:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): clock() based CLOCK_PROCESS_CPUTIME_ID
+	  emulation implemented.
+
+Wed Aug 21 21:02:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_proc_times): Use RB_GC_GUARD to guard objects from GC.
+
+Wed Aug 21 20:33:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (get_clk_tck): Extracted from rb_proc_times.
+	  (rb_clock_gettime): times() based CLOCK_PROCESS_CPUTIME_ID emulation
+	  is implemented.
+
+Wed Aug 21 19:31:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: POSIX_GETTIMEOFDAY_CLOCK_REALTIME is renamed to
+	  SUS_GETTIMEOFDAY_CLOCK_REALTIME.
+
+Wed Aug 21 19:17:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): CLOCK_PROCESS_CPUTIME_ID emulation
+	  using getrusage is implemented.
+
+Wed Aug 21 17:34:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (getrusage_time): Fallback clock_gettime to getrusage when
+	  clock_gettime fails.
+	  Reported by Eric Saxby.  [ruby-core:56762] [Bug #8805]
+
+Wed Aug 21 02:32:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: fix regexp's once option behavior.
+	  fix [ruby-trunk - Bug #6701]
+
+	* insns.def: remove `onceinlinecache' and introduce `once' instruction.
+	  `once' doesn't use `setinlinecache' insn any more.
+
+	* vm_core.h: `union iseq_inline_storage_entry' to store once data.
+
+	* compile.c: catch up above changes.
+
+	* iseq.c: ditto.
+
+	* vm.c, vm_insnhelper.c: ditto. fix `m_core_set_postexe()' which
+	  is depend on `onceinlinecache' insn.
+
+	* test/ruby/test_regexp.rb: add tests.
+
+	* iseq.c: ISEQ_MINOR_VERSION to 1 (should increment major?)
+
+Wed Aug 21 02:30:15 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gcdebug_print_obj_condition): add printing information.
+
+Tue Aug 20 13:38:00 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/gdbm/test_gdbm.rb: skip TestGDBM#test_s_open_lock on Solaris.
+	  On Solaris (and platforms which do not have flock and have lockf),
+	  with GDBM 1.10, gdbm_open(3) blocks when opening already locked
+	  gdbm file. [Bug #8790] [ruby-dev:47631]
+
+Tue Aug 20 02:32:52 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/test/: [DOC] Document Test::Unit, hide most submodules and
+	  classes from rdoc. Since lib/test is only present as a compatibility
+	  layer with the legacy test suite many test/unit users will be using
+	  minitest or the test/unit gem instead. It is recommended to use one
+	  of these alternatives for writing new tests.
+
+	  This patch was based on a patch submitted by Steve Klabnik.
+	  [ruby-core:56694] [Bug #8778]
+
+Tue Aug 20 02:10:19 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/rss.rb: [DOC] Document for constants by Steve Klabnik
+	  [ruby-core:56705] [Bug #8798]
+
+Tue Aug 20 02:01:10 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/xmlparser.rb: [DOC] Hide legacy constant from rdoc
+	  Patch by Steve Klabnik [ruby-core:56708] [Bug #8799]
+
+Tue Aug 20 01:52:05 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/socket/unixserver.c: [DOC] Document #accept
+	* ext/socket/tcpserver.c: ditto
+	* ext/socket/udpsocket.c: [DOC] Fix indentation of documentation
+	* ext/socket/socket.c: ditto
+	  Patches by David Rodr'iguez [ruby-core:56734] [Bug #8802]
+
+Tue Aug 20 01:19:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Define ac_cv_func_clock_gettime to yes for mingw*.
+
+Mon Aug 19 21:31:35 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h: Fix a compilation error with
+	  i586-mingw32msvc-gcc of gcc-mingw32 package on Debian squeeze.
+	  ruby/missing.h should be included before include/ruby/win32.h
+	  because struct timespec, used in the clock_gettime declaration in
+	  include/ruby/win32.h, is defined in ruby/missing.h instead of
+	  system headers.
+
+Mon Aug 19 20:55:12 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix around GC_DEBUG.
+
+	* gc.c (RVALUE::line): should be VALUE. On some environment
+	  (such as mswin64), `int' introduces alignment mismatch.
+
+	* gc.c (newobj_of): add an assertion to check VALUE alignment.
+
+	* gc.c (aligned_malloc): `&' is low priority than `=='.
+
+	* gc.c: define GC_DEBUG everytime and use it as value 0 or 1.
+
+Mon Aug 19 17:43:44 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_fiber.rb: collect garbage fibers immediately.
+
+Mon Aug 19 17:41:49 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/profile_test_all.rb: add `failed?' information.
+
+Mon Aug 19 17:00:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (retry_fork): retry with GC if ENOMEM occurred, to free
+	  swap/kernel space.
+
+Mon Aug 19 13:28:47 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (CLOCK_MONOTONIC): typo.
+
+	* win32/win32.c: removed duplicated declarations.
+
+Mon Aug 19 13:03:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (clock_gettime): should not overwrite cache variable
+	  with different condition.  otherwise -lrt is not linked and the link
+	  fails, after reconfig.
+
+Mon Aug 19 12:56:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (Init_process): Add constants: CLOCK_REALTIME_ALARM and
+	  CLOCK_BOOTTIME_ALARM.
+
+Sun Aug 18 20:17:41 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* variable.c, vm_method.c: remove dead code.
+
+	* test/ruby/test_fiber.rb, test/ruby/test_thread.rb:
+	  change accordingly.
+
+Sun Aug 18 19:32:26 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* error.c, file.c, gc.c, hash.c, thread.c, variable.c, vm_eval.c, bin/erb:
+	  $SAFE=4 is obsolete.
+
+Sun Aug 18 14:30:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): Rename POSIX_TIME_CLOCK_REALTIME to
+	  ISO_C_TIME_CLOCK_REALTIME.
+
+Sun Aug 18 14:22:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Revert r42604.  It causes linking librt on systems
+	  with newer glibc uselessly.
+
+Sun Aug 18 13:18:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (Init_process): Add constants: CLOCK_REALTIME_COARSE,
+	  CLOCK_MONOTONIC_COARSE and CLOCK_BOOTTIME.
+
+Sun Aug 18 12:41:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (clock_gettime): need to check with -lrt prior to check
+	  for the function only.  otherwise -lrt is not linked and the link
+	  fails, when ac_cv_func_clock_gettime is cached as yes.
+
+Sun Aug 18 10:05:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2str1): Make an expression more explicit.
+
+Sun Aug 18 03:18:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2str1): Use power_level instead of bitsize(xn).
+
+Sun Aug 18 00:44:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BIGDIVREM_EXTRA_WORDS): Redefine to 1.
+	  (bigdivrem_num_extra_words): Removed.
+	  (bigdivrem_normal): Simplified.
+	  (big2str_karatsuba): Ditto.
+
+Sat Aug 17 23:25:19 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/ruby/test_time.rb: use the in_timezone() helper
+	  and define it at the top with other helpers.
+
+Sat Aug 17 22:20:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_mload): ignore auxiliary data, offset and zone, if
+	  invalid.  [ruby-core:56648] [Bug #8795]
+
+Sat Aug 17 20:11:49 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* process.c: [DOC] MACH_ABSOLUTE_TIME_CLOCK_MONOTONIC is an
+	  available emulation for a monotonic clock on Darwin.
+	  https://developer.apple.com/library/mac/qa/qa1398/_index.html
+
+Fri Aug 16 18:12:05 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/profile_test_all.rb: fix typo.
+
+Fri Aug 16 18:09:20 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/profile_test_all.rb: remove space characters from test names.
+
+Fri Aug 16 17:32:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/profile_test_all.rb: refactoring memory profiling tool for
+	  test-all.
+	  Add profiling targets /proc/meminfo and /proc/self/status.
+
+	* test/runner.rb: accept other than 'true'.
+
+Fri Aug 16 11:23:35 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_file_size, rb_file_flock): improve performance of Windows.
+
+	* file.c (rb_file_truncate): removed unnecessary #ifdef.
+
+	* test/test_file.rb (TestFile#test_truncate_size): added an assertion
+	  for File#size.
+
+Fri Aug 16 10:07:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_single1): Renamed from bigdivrem_single.  Add
+	  x_higher_bdigit argument.
+	  (bigdivrem_single): Just call bigdivrem_single1.
+	  (bigdivrem_restoring): Use bigdivrem_single1 to avoid memmove.
+
+Fri Aug 16 09:17:00 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_small_rshift): Specify the higher BDIGIT instead of
+	  sign bit.
+	  (big_shift3): Follow the above change.
+
+Fri Aug 16 02:20:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_toom3): Reduce a branch.
+
+Fri Aug 16 02:14:09 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c (rb_clock_gettime): add CLOCK_MONOTONIC support on OS X.
+	  http://developer.apple.com/library/mac/qa/qa1398/_index.html
+	  [Feature #8658]
+
+Fri Aug 16 01:37:43 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_single): Use shift when y is a power of two.
+
+Fri Aug 16 01:09:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_restoring): Use bigdivrem_single if non-topmost
+	  BDIGITs of y are zero.
+
+Fri Aug 16 00:33:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2str1): Truncate topmost zeros of x.
+
+Fri Aug 16 00:00:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_divmod): Simplify an expression.
+
+Thu Aug 15 23:26:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_normal): Remove a local variable.
+
+Thu Aug 15 23:08:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_karatsuba): Use bigdivrem_restoring directly to
+	  reduce working buffer and memory copy.
+	  (rb_big2str1): Allocate working buffer for big2str_karatsuba here.
+
+Thu Aug 15 20:51:29 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c, internal.h (rb_io_flush_raw): new function to select calling
+	  fsync() (on Windows).
+
+	* io.c (rb_io_flush_raw): use above function.
+
+	* file.c (rb_file_truncate): use above function.
+
+	* test/ruby/test_file.rb (TestFile#test_truncate_size): test for
+	  above changes.
+
+Thu Aug 15 18:39:31 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (clock_gettime): improve precision when freq is less
+	  than and nearly equals 10**9.
+
+Thu Aug 15 17:43:15 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_lazy_sweep): remove heap_increment() here because heap_inc
+	  may be 0.
+
+Thu Aug 15 16:59:56 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_rewind): remove fsync() for Windows to improve the
+	  performance.
+
+Thu Aug 15 16:30:23 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb (TestFileUtils#test_rmdir):
+	  FileUtils.rmdir ignores Errno::ENOTEMPTY, so, in such cases, this
+	  assertion is nonsense.
+
+Thu Aug 15 15:49:35 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): [DOC] FreeBSD 7.1 supports
+	  CLOCK_THREAD_CPUTIME_ID.
+	  http://www.freebsd.org/releases/7.1R/relnotes.html
+
+Thu Aug 15 14:30:23 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h, win32/Makefile.sub, win32/win32.c
+	  (clock_gettime): [experimental] emulates clock_gettime(2) of posix.
+
+Thu Aug 15 02:32:40 2013  Zachary Scott  <e@zzak.io>
+
+	* hash.c (rb_hash_aset): [DOC] Document key dup patch by @kachick
+	  [Fixes GH-382] https://github.com/ruby/ruby/pull/382
+
+Wed Aug 14 14:28:39 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* proc.c (rb_mod_define_method): now they return the symbols of the
+	  defined methods, not the methods/procs themselves.
+	  [ruby-dev:42151] [Feature #3753]
+
+	* NEWS: documents about above change and def-expr (see r42337).
+
+	* test/ruby/test_module.rb: tests about above change.
+
+Wed Aug 14 00:51:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_restoring): xn argument removed.
+	  (bigdivrem_normal): Follow the above change.
+
+Wed Aug 14 00:18:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_div_struct): Remove xn and j field.  Add zn field.
+	  (bigdivrem1): Follow the above change.
+	  (bigdivrem_restoring): Ditto.
+
+Tue Aug 13 23:38:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_div_struct): ynzero field removed.
+	  (bigdivrem1): Follow the above change.
+	  (bigdivrem_restoring): Ditto.
+
+Tue Aug 13 23:01:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_restoring): Extracted from bigdivrem_normal.
+
+Tue Aug 13 22:12:59 2013  Kenichi Kamiya  <kachick1@gmail.com>
+
+	* random.c (rb_random_ulong_limited): coerce before check negative.
+	  [Fixes GH-379]
+
+Tue Aug 13 21:52:15 2013  Kenichi Kamiya  <kachick1@gmail.com>
+
+	* object.c (Init_Object): undef Module#prepend_features on Class, as
+	  well as Module#append_features.  [Fixes GH-376]
+
+	* test_class.rb: Added test for above. And ensure type checking
+	  on similar methods as module_function.
+
+Tue Aug 13 08:52:18 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/syntax/literals.rdoc: [DOC] String literal concat by @cknadler
+	  [Fixes GH-380] https://github.com/ruby/ruby/pull/380
+
+Mon Aug 12 23:07:21 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_marks_test): inhibit gc for st's operation.
+
+Mon Aug 12 15:59:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_whole_match_p): treat CR in middle of a line as a
+	  mere whitespace.
+
+Mon Aug 12 15:16:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (rb_prepend_module): make T_ICLASS object shady because
+	  this T_ICLASS object seems to share method table with other class
+	  objects. It was causes WB miss.
+	  TODO: need to know the data structure.
+
+	* test/ruby/test_module.rb: add a test for WB miss.
+
+Mon Aug 12 13:47:54 2013  Zachary Scott  <e@zzak.io>
+
+	* process.c: [DOC] RDoc formatting of Process.clock_gettime
+
+Mon Aug 12 13:29:09 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/yaml/dbm.rb: [DOC] Document call-seq for YAML::DBM
+
+Mon Aug 12 12:57:26 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/dbm/extconf.rb: [DOC] Hide from RDoc
+	  Some libraries might want to document extconf.rb so RDoc treats it
+	  like any other ruby program. However, DBM users shouldn't care about
+	  these methods.
+
+Mon Aug 12 12:53:39 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/dbm/dbm.c: [DOC] Reformat headings of DBM class
+
+Mon Aug 12 12:46:31 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/yaml.rb, lib/yaml/: [DOC] Document YAML::DBM#key and add
+	  references to similar methods with more detail. This patch brings
+	  lib/yaml to 100% documentation coverage.
+
+Mon Aug 12 02:51:32 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_s_set_input): on OS X with editline,
+	  Readline.readline doesn't work because readline_get doesn't use
+	  rl_getc. The difference is introduced by r42402 [ruby-dev:47509]
+	  [Bug #8644]. Before it rb_io_stdio_file set ifp->stdio_file.
+	  Therefore add manually setting the value.
+
+	* ext/readline/readline.c (readline_s_set_output): ditto.
+
+Sun Aug 11 23:27:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_str_encode_ospath): OS path encoding on Mac OS X is also
+	  fixed.
+
+Sun Aug 11 22:57:24 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_require.rb (assert_require_nonascii_path): OS path
+	  encoding on Windows is fixed, so encoding of __FILE__ should be it.
+	  [ruby-core:56498] [Bug #8764]
+
+Sun Aug 11 19:11:45 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/parser/test_sax2.rb: Expand abbreviated class name.
+
+Sun Aug 11 19:06:03 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/sax2listener.rb (REXML::SAX2Listener#notationdecl): Fix
+	  wrong number of arguments in the template listener.
+	  [Bug #8731] [ruby-dev:47582]
+	  Reported by Ippei Obayashi.
+	* test/rexml/parser/test_sax2.rb: Add tests for parsing notation
+	  declarations with SAX2 API.
+
+Sun Aug 11 18:44:04 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/sax2listener.rb (REXML::SAX2Listener#elementdecl): Fix wrong
+	  examples. [Bug #8731] [ruby-dev:47582]
+	  Reported by Ippei Obayashi.
+
+Sun Aug 11 18:42:13 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/sax2parser.rb
+	  (REXML::Parsers::SAX2Parser#handle_entitydecl): Extract.
+
+Sun Aug 11 18:40:25 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/sax2parser.rb (REXML::Parsers::SAX2Parser#parse):
+	  Fix wrong "%" position in parameter entity declaration event argument.
+	* test/rexml/parser/test_sax2.rb: Add tests for the above case.
+
+Sun Aug 11 18:08:40 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/sax2parser.rb (REXML::Parsers::SAX2Parser#parse):
+	  Support NDATA in external ID entity declaration.
+	* test/rexml/parser/test_sax2.rb: Add tests for the above case.
+
+Sun Aug 11 18:07:39 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb
+	  (REXML::Parsers::BaseParser#pull_event): Support optional NDATA
+	  in external ID entity declaration.
+
+Sun Aug 11 17:54:07 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS (REXML::Parsers::SAX2Parser): Add about this change.
+	* lib/rexml/parsers/sax2parser.rb (REXML::Parsers::SAX2Parser#parse):
+	  Fix wrong number of arguments. Document says "an array of the
+	  entity declaration" but it passes two or more arguments.
+	  This is a bug but it break backward compatibility.
+	  Reported by Ippei Obayashi. [Bug #8731] [ruby-dev:47582]
+	* lib/rexml/sax2listener.rb (REXML::SAX2Listener#entitydecl): ditto.
+	  The listener template accepted two arguments.
+	* test/rexml/parser/test_sax2.rb: Add tests for external ID case.
+
+Sun Aug 11 17:41:41 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/parser/test_sax2.rb: Add SAX2 API test.
+
+Sun Aug 11 15:10:40 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_enc_symname_type): allow ID_ATTRSET for ID_INSTANCE,
+	  ID_GLOBAL, ID_CLASS, ID_JUNK too.  [Bug #8756]
+
+Sun Aug 11 13:17:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* include/ruby/encoding.h: Reduce ENCODING_INLINE_MAX to 127 as this
+	  should be sufficient to represent all the encodings Ruby supports.
+
+Sun Aug 11 11:54:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_clock_gettime): New method.
+	  This is accepted in the meeting:
+	  https://bugs.ruby-lang.org/projects/ruby/wiki/DevelopersMeeting20130809
+	  This method is accepted as a CRuby feature.
+	  I.e. Other Ruby implementations don't need to implement it.
+	  [ruby-core:56087] [Feature #8658]
+
+Sun Aug 11 10:40:48 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/time.rb: [DOC] Correcting rdoc visibility of time.rb constants
+	  Reported by Tanaka Akira [ruby-core:56517]
+
+Sun Aug 11 04:48:14 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_str_normalize_ospath):
+	  HFS Plus (Mac OS Extended) uses a variant of Normal Form D in which
+	  U+2000 through U+2FFF, U+F900 through U+FAFF, and U+2F800 through
+	  U+2FAFF are not decomposed (this avoids problems with round trip
+	  conversions from old Mac text encodings).
+	  http://developer.apple.com/library/mac/qa/qa1173/_index.html
+	  Therefore fix r42457 to exclude the range.
+
+Sun Aug 11 03:26:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bitsize): Fix a conditional expression.
+
+Sun Aug 11 02:44:03 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/time.rb: [DOC] Document constants by @markijbema [Fixes GH-377]
+	  https://github.com/ruby/ruby/pull/377
+
+Sun Aug 11 01:28:52 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Revert r42458.
+	  It removes the HAVE_CLOCK_GETTIME from config.h.
+	  http://www.rubyist.net/~akr/chkbuild/debian/ruby-trunk/log/20130809T044800Z.diff.html.gz
+
+Sat Aug 10 13:53:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_id_attrset): allow other than ID_ATTRSET.
+
+	* parse.y (intern_str): ditto.  try stem ID for ID_INSTANCE,
+	  ID_GLOBAL, ID_CLASS, ID_JUNK too.  [Bug #8756]
+
+Sat Aug 10 12:49:50 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb
+	  (REXML::Parsers::BaseParser::CDATA_END): Use "\A" instead of "^".
+	  It is not an used constant but I fix it. (Or should I remove it?)
+
+Sat Aug 10 12:47:19 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser):
+	  Fix wrong constant name. "]>" pattern match is the same but
+	  it is used for "<!DOCTYPE" end mark not "<![CDATA[" end mark.
+
+Sat Aug 10 12:43:15 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser):
+	  Use "\A" instead of "^" in document type declaration patterns
+	  because they are used as the head match in content not the head
+	  match in line. They don't cause any problems in the current code
+	  but it should be fixed.
+
+Sat Aug 10 12:39:00 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/parse/test_document_type_declaration.rb: Add tests for
+	  parsing document type declaration.
+
+Sat Aug 10 12:00:45 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser::SYSTEM):
+	  Fix loose "head" match regular expression. It doesn't cause any
+	  problem in the current code but it should be fixed because readers
+	  may confuse it.
+	  Patch by Ippei Obayashi. Thanks!!!
+
+Sat Aug 10 11:58:24 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/parse/test_notation_declaration.rb (#test_system_public):
+	  Add a test for PUBLIC notation and SYSTEM notation order case.
+
+Sat Aug 10 11:31:35 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/baseparser.rb (REXML::Parsers::BaseParser::PUBLIC):
+	  Fix loose "head" match regular expression.
+	  [Bug #8701] [ruby-dev:47551]
+	  Patch by Ippei Obayashi. Thanks!!!
+	* test/rexml/parse/test_notation_declaration.rb (#test_system_public):
+	  Add a test for the above case.
+
+Sat Aug 10 09:20:21 2013  Zachary Scott  <e@zzak.io>
+
+	* NEWS: [DOC] typo in example reported by @moretea
+	  https://github.com/ruby/ruby/commit/a39e724#commitcomment-3831489
+
+Sat Aug 10 09:19:04 2013  Zachary Scott  <e@zzak.io>
+
+	* proc.c: [DOC] rdoc code formatting
+
+Sat Aug 10 09:12:01 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_id_attrset): check if the argument is valid type as an
+	  attribute.
+
+Sat Aug 10 05:44:08 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/trackback.rb: [DOC] Hide RSS::Trackback from rdoc
+	  Patch by Steve Klabnik [Bug #8755] [ruby-core:56456]
+
+Sat Aug 10 04:52:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_div_struct): Use size_t.
+	  (bigdivrem1): Ditto.
+	  (bigdivrem_num_extra_words): Ditto.
+	  (bigdivrem_single): Ditto.
+	  (bigdivrem_normal): Ditto.
+	  (bary_divmod): Ditto.
+
+Fri Aug  9 23:47:15 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rexmlparser.rb: Remove needless REXML version check.
+	  Both RSS Parser and REXML are bundled in Ruby. RSS Parser can
+	  always use the latest REXML. [Bug #8754] [ruby-core:56454]
+	  Patch by Steve Klabnik. Thanks!!!
+
+Fri Aug  9 22:51:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (XLDFLAGS, LIBRUBYARG_STATIC): CoreFoundation framework
+	  option is now needed always, regardless enable-shared.
+	  [ruby-core:56467] [Bug #8759]
+
+Fri Aug  9 22:20:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file_internal): use rb_parser_compile_string_path and
+	  rb_parser_compile_file_path, String path name versions.  [Bug #8753]
+
+Fri Aug  9 07:16:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/io/console/console.c: delete redefinition of rb_cloexec_open.
+	  drop support for 1.8 and 1.9 from the next release of io-console gem.
+
+Fri Aug  9 19:13:54 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: update about new methods for Binding.
+
+Fri Aug  9 18:48:09 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: add Binding#local_variable_get/set/defined?
+	  to access local variables which a binding contains.
+	  Most part of implementation by nobu.
+
+	* test/ruby/test_proc.rb: add a tests for above.
+
+	* vm.c, vm_core.h (rb_binding_add_dynavars): add a new function
+	  to add a new environment to create space for new local variables.
+
+Fri Aug  9 14:02:01 2013  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/make-snapshot: Fix order of priority for option parameter.
+
+Fri Aug  9 12:06:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_str_normalize_ospath): normalize to Normalization Form C
+	  using CFString.
+
+Fri Aug  9 10:53:57 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* time.c (get_timeval, get_new_timeval): use rb_obj_class()
+	  instead of CLASS_OF() because CLASS_OF() may return
+	  a singleton class.
+
+Fri Aug  9 10:42:11 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_insnhelper.c (vm_invoke_block): returning from lambda proc
+	  now always exits from the Proc. [ruby-core:56193] [Feature #8693]
+
+	* NEWS, test/ruby/test_lambda.rb: ditto. Patch by nobu.
+
+Fri Aug  9 00:10:32 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_zip_func): fix non-single argument.  fix
+	  out-of-bound access and pack multiple yielded values.
+	  [ruby-core:56383] [Bug #8735]
+
+Thu Aug  8 23:01:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_singleton_p): new method Module#singleton_class? to
+	  return whether the receiver is a singleton class or not.
+	  [ruby-core:51087] [Feature #7609]
+
+Thu Aug  8 21:56:44 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_overflow_p): Avoid signed integer overflow.
+	  (rb_time_new): Fix overflow condition.
+
+Thu Aug  8 19:58:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (rb_threadptr_pending_interrupt_check_mask):
+	  use RARRAY_RAWPTR() instead of RARRAY_PTR() because
+	  there is no new reference.
+
+Thu Aug  8 19:56:52 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (rb_str_format_m): use RARRAY_RAWPTR() instead of
+	  RARRAY_PTR() because there is no new reference.
+
+Thu Aug  8 19:55:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: define USE_RGENGC_LOGGING_WB_UNPROTECT.
+
+Thu Aug  8 16:44:25 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: add old macro name `RUBY_EVENT_SWITCH'.
+	  This macro name is obsolete because it is renamed to
+	  RUBY_INTERNAL_EVENT_SWITCH, but it has compatibility problem
+	  using this macro name like ruby-prof.
+	  I want to remove this macro after ruby 2.1.
+
+Thu Aug  8 15:37:53 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/coverage/test_coverage.rb (TestCoverage#test_big_code): use `1'
+	  instead of `p' to get rid of a side effect.
+	  Kernel#p without any argument seems to do nothing, but flushes stdout.
+	  and, if stdout is redirected to file, fsync() will be called on
+	  Windows.  so, when running test-all on Windows with redirection, such
+	  as CI environment, this test took a lot of time.
+
+Thu Aug  8 14:54:18 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* NEWS: add description of incompatibility introduced by r42396.
+	  [ruby-core:56329] [Bug #8722]
+
+Thu Aug  8 14:50:36 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (mini): portable target to build miniruby
+
+	* common.mk (bisect): run git-bisect with miniruby
+
+	* common.mk (bisect-ruby): run git-bisect with ruby
+
+	* tool/bisect.sh: script for git-bisect
+
+Thu Aug  8 12:11:43 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/webrick/test_httpresponse.rb (test_send_body_*_chunked): these
+	  expectations assumes that the IOs are binmode.  fixed test failures
+	  introduced at r42427 on Windows.
+
+Thu Aug  8 10:27:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_last): revert r42400.  [Bug #8739]
+
+Thu Aug  8 10:26:25 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_str_normalize_ospath): extract and move from dir.c.
+
+Thu Aug  8 05:59:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* test/openssl/test_ssl.rb: Fix test for CVE-2013-4073.
+	  Patch by Antonio Terceiro. [Bug #8750] [ruby-core:56437]
+
+Thu Aug  8 03:37:38 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/webrick/httpresponse.rb:  Allow #body to be an IO-like object
+	  that responds to #readpartial and #read.
+	  [ruby-trunk - Feature #8155]
+	* NEWS:  NEWS for above
+	* test/webrick/test_httpresponse.rb:  Tests for above.
+
+Wed Aug  7 23:06:26 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ruby.c (Process.argv0): New method to return the original value
+	  of $0. [Feature #8696]
+
+Wed Aug  7 23:05:55 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ruby.c (Process.setproctitle): New method to change the title of
+	  the running process that is shown in ps(1). [Feature #8696]
+
+Wed Aug  7 20:05:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_odd_p): Check the bignum length.
+	  (rb_big_even_p): Ditto.
+
+Wed Aug  7 19:29:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (dbl2big): A condition simplified.
+
+Wed Aug  7 16:34:30 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/webrick/test_cgi.rb (TestWEBrickCGI#{start_cgi_server,test_cgi}):
+	  mswin is not only mswin32 but also mswin64. [Bug #8746]
+
+Wed Aug  7 16:19:12 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (rb_fiber_start): use RARRAY_RAWPTR() instead of
+	  RARRAY_PTR() because there is no new reference.
+
+	* proc.c (curry): ditto.
+
+	* proc.c (rb_proc_call): remove line break.
+
+Wed Aug  7 13:20:12 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* random.c (random_load): use RARRAY_RAWPTR() instead of
+	  RARRAY_PTR() because there is no new reference.
+
+Wed Aug  7 12:58:23 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (thread_start_func_2): use RARRAY_RAWPTR() instead of
+	  RARRAY_PTR() because there is no new reference.
+
+Wed Aug  7 09:00:24 2013  Zachary Scott  <e@zzak.io>
+
+	* string.c: [DOC] Description of rb_str_equal [Fixes GH-375]
+	  Based on a patch by @markijbema
+	  https://github.com/ruby/ruby/pull/375
+
+Wed Aug  7 08:30:38 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_hmac.c: [DOC] Documentation for OpenSSL::HMAC
+	  based on a patch by @repah documenting-ruby/ruby#14
+	  https://github.com/documenting-ruby/ruby/pull/14
+
+Wed Aug  7 07:46:23 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/utils.rb: [DOC] RSS::Utils by Steve Klabnik [Bug #8745]
+
+Wed Aug  7 07:38:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (nlz16): Removed.
+	  (nlz32): Ditto.
+	  (nlz64): Ditto.
+	  (nlz128): Ditto.
+	  (nlz_int): New function.
+	  (nlz_long): New function.
+	  (nlz_long_long): New function.
+	  (nlz_int128): New function.
+	  (nlz): Follow above changes.
+	  (bitsize): Follow above changes.
+
+Tue Aug  6 22:38:15 2013  Zachary Scott  <e@zzak.io>
+
+	* time.c: [DOC] Typo in Time overview by @sparr [Fixes GH-374]
+	  https://github.com/ruby/ruby/pull/374
+
+Tue Aug  6 22:35:32 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/1.0.rb: [DOC] Document RSS10 by Steve Klabnik [Bug #8740]
+
+Tue Aug  6 22:14:11 2013  Kouji Takao  <kouji.takao@gmail.com>
+
+	* ext/readline/readline.c (readline_s_delete_text): remove
+	  checking "$SAFE == 4".
+
+	* ext/readline/readline.c: fix rdoc, remove "Raises SecurityError"
+	  and add "Raises NotImplementedError".
+
+Tue Aug  6 22:04:38 2013  Kouji Takao  <kouji.takao@gmail.com>
+
+	* ext/readline/readline.c, test/readline/test_readline.rb: fix
+	  indent.
+
+Tue Aug  6 21:59:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_last): return nil for empty range, or in the case the
+	  predecessor is smaller than the begin.  [Bug #8739]
+
+Tue Aug  6 21:48:31 2013  Kouji Takao  <kouji.takao@gmail.com>
+
+	* ext/readline/readline.c (readline_s_set_point, Init_readline):
+	  add Readline.point=(pos). Patched by naruse.  [ruby-dev:47535]
+	  [Feature #8675]
+
+Tue Aug  6 21:14:11 2013  Kouji Takao  <kouji.takao@gmail.com>
+
+	* ext/readline/readline.c (Init_readline, readline_s_set_output)
+	(clear_rl_outstream, readline_s_set_input, clear_rl_instream)
+	(readline_readline): fix causing SEGV if closed IO object that is
+	  set Readline.input or Readline.output. Patched by akr
+	  [ruby-dev:47509] [Bug #8644]
+
+Tue Aug  6 17:56:40 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_push_frame): change type of stack_max to size_t.
+
+Tue Aug  6 17:42:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_last): exclude the last number of the exclusive range
+	  if the end is Numeric.  [ruby-dev:47587] [Bug #8739]
+
+Tue Aug  6 17:42:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_conv_from_wchar): converted string to CP_UTF8
+	  should have UTF-8 encoding.  otherwise no conversion takes place
+	  later.
+
+Tue Aug  6 17:21:38 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_push_frame): fix stack overflow check codes.
+	  Stack overflow check should be done *after* pushing a stack frame.
+	  However, some stack overflow checking codes checked *before*
+	  pushing a stack frame with iseq->stack_max.
+	  To solve this problem, add a new parameter `stack_max' to specify
+	  a possible consuming stack size.
+
+	* vm_core.h (CHECK_VM_STACK_OVERFLOW0): add to share the stack overflow
+	  checking code.
+
+	* insns.def: catch up this change.
+
+	* vm.c, vm_eval.c: ditto.
+
+	* test/ruby/test_exception.rb: add a stack overflow test.
+	  This code is reported by nobu.
+
+Tue Aug  6 17:02:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_conv_from_wchar): use WideCharToMultiByte(),
+	  as like as mbstr_to_wstr(), in the first step of the conversion from
+	  WCHAR.
+
+Tue Aug  6 16:14:32 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_eval.c (eval_string_with_cref): copy cref to limit the scope of
+	  refinements in the eval string. [ruby-core:56329] [Bug #8722]
+
+	* test/ruby/test_refinement.rb: related test.
+
+Tue Aug  6 12:23:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_realloc): Use VALGRIND_MAKE_MEM_UNDEFINED to
+	  declare undefined memory area.
+	  (bignew_1): Ditto.
+
+	* internal.h (VALGRIND_MAKE_MEM_DEFINED): Moved from gc.c
+	  (VALGRIND_MAKE_MEM_UNDEFINED): Ditto.
+
+Tue Aug  6 01:40:37 2013  Zachary Scott  <e@zzak.io>
+
+	* process.c: [DOC] Document caveats of command form of Process.spawn
+	  with regard to the shell and OS. Patched by Steve Klabnik [Bug #8550]
+
+Tue Aug  6 01:28:35 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/0.9.rb: [DOC] Typo in example [Bug #8732]
+
+Tue Aug  6 01:22:37 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/2.0.rb: [DOC] Document RSS::Rss by Steve Klabnik #8740
+	* lib/rss/atom.rb: [DOC] Typo in rdoc by Steve Klabnik
+
+Mon Aug  5 23:47:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Rename local variables.
+
+Mon Aug  5 22:23:59 2013  Zachary Scott  <e@zzak.io>
+
+	* vm_trace.c: [DOC] Fix TracePoint return values in examples
+	  Based on a patch by @sho-h [Fixes GH-373]
+	  https://github.com/ruby/ruby/pull/373
+
+Mon Aug  5 17:38:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write_console): use MultiByteToWideChar() for
+	  the last step of conversion to WCHAR, to get rid of warnings from
+	  rb_enc_find() in miniruby.  [ruby-dev:47584] [Bug #8733]
+
+	* win32/win32.c (wstr_to_mbstr, mbstr_to_wstr): fix wrong trimming.
+	  WideCharToMultiByte() and MultiByteToWideChar() do not count
+	  NUL-terminator in the size for conversion result, unless the input
+	  length is -1.
+
+Mon Aug  5 11:51:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* include/ruby/encoding.h: document which user flags are used by
+	  ENCODING_MASK for better greppability
+
+Mon Aug  5 10:01:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* object.c (rb_class_inherited_p): allow iclasses to be tested for
+	  inheritance. [Bug #8686] [ruby-core:56174]
+
+	* test/ruby/test_method.rb: add test
+
+Mon Aug  5 06:13:48 2013  Zachary Scott  <e@zzak.io>
+
+	* enumerator.c: [DOC] Remove reference to Enumerator::Lazy#cycle
+	  Patch by @kachick [Fixes GH-372]
+	  https://github.com/ruby/ruby/pull/372
+
+Mon Aug  5 03:57:16 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rss/0.9.rb: [DOC] Document RSS09 by Steve Klabnik [Bug #8732]
+
+Mon Aug  5 03:35:11 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rexml/attribute.rb: [DOC] Update example for #namespace
+	  Patch by Ippei Obayashi [Bug #8685] [ruby-core:56173]
+
+Sun Aug  4 21:08:29 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_zip): performance implement by using
+	  ALLOCA_N() to allocate tmp buffer.
+
+Sun Aug  4 07:14:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* README.EXT, README.EXT.ja: Mention rb_integer_pack and
+	  rb_integer_unpack.
+
+Sun Aug  4 01:54:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BARY_TRUNC): New macro.
+	  (bary_cmp): Use BARY_TRUNC.
+	  (bary_mul_toom3): Ditto.
+	  (bary_divmod): Ditto.
+	  (abs2twocomp): Ditto.
+	  (bigfixize): Ditto.
+	  (rb_cstr_to_inum): Ditto.
+	  (big2str_karatsuba): Ditto.
+	  (bigdivrem): Ditto.
+
+Sun Aug  4 00:57:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_karatsuba): Don't allocate new temporary buffer
+	  if the buffer is enough for current invocation.
+
+Sun Aug  4 00:22:34 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary2bdigitdbl): New function.
+	  (bdigitdbl2bary): Ditto.
+	  (bary_mul_single): Use bdigitdbl2bary.
+	  (power_cache_get_power): Ditto.
+	  (bary_divmod): Use bary2bdigitdbl.
+	  (big2str_orig): Ditto.
+	  (bigdivrem): Ditto.
+
+Sat Aug  3 22:47:11 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: The branch condition of selecting multiplication
+	  algorithms should check smaller argument because Karatsuba and Toom3
+	  is effective only if both arguments are big.
+	  (bary_mul_toom3_branch): Compare the smaller argument to
+	  TOOM3_MUL_DIGITS.
+	  (bary_mul): Compare the smaller argument to KARATSUBA_MUL_DIGITS.
+
+Sat Aug  3 22:23:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_orig): Receive the number to stringize as
+	  BDIGIT array and size.
+	  (big2str_karatsuba): Receive the number to stringize as BDIGIT array
+	  and size.  Use an temporary array of BDIGIT.
+	  (rb_big2str1): Follow the above change.
+
+Sat Aug  3 13:30:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (MAX_BASE36_POWER_TABLE_ENTRIES): Renamed from
+	  MAX_BIG2STR_TABLE_ENTRIES.
+	  (base36_power_cache): Renamed from big2str_power_cache.
+	  (base36_numdigits_cache): Renamed from big2str_numdigits_cache.
+
+Sat Aug  3 10:33:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_set_integer_literal): use rb_rational_raw1() for
+	  integral rational because no reduction is needed with 1.
+
+Sat Aug  3 09:46:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (setup_passwd, setup_group): set proper encodings to
+	  string members.
+
+Sat Aug  3 09:30:57 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (rb_struct_define_under): new function to define Struct
+	  under the given namespace, not under Struct.  [Feature #8264]
+
+	* ext/etc/etc.c: use rb_struct_define_under.
+
+Sat Aug  3 06:55:29 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y (value_expr_gen): now NODE_DEFN and NODE_DEFS are not void
+	  value expressions.  get rid of wrong warning with -w, and make to
+	  pass tests with chkbuild.  ref. [Feature #3753]
+
+Sat Aug  3 04:23:48 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/refinements.rdoc:  Remove mention of instance_eval and
+	  module_eval from scope section per:
+	  http://twitter.com/shugomaeda/status/363219951336693761
+
+Sat Aug  3 02:22:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_orig): Refactored.
+
+Sat Aug  3 01:20:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigadd_core): Removed.
+	  (bigadd): Use bary_add instead of bigadd_core.
+
+Sat Aug  3 00:52:43 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2str1): Simplify power_level calculation.
+
+Sat Aug  3 00:34:20 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_zip): use rb_ary_new2() to create buffer
+	  if rb_block_arity() > 1.
+
+Sat Aug  3 00:12:00 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* NEWS: Add the description that IO#seek supports SEEK_DATA
+	  and SEEK_HOLE.
+
+Fri Aug  2 23:57:57 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* vm.c (m_core_define_method, m_core_define_singleton_method): now
+	  the value of def-expr is the Symbol of the name of the method, not
+	  nil.
+	  ref. [ruby-dev:42151] [Feature #3753]
+
+	* test/ruby/test_syntax.rb (TestSyntax#test_value_of_def): test for
+	  above changes.
+
+Fri Aug  2 23:54:11 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_zip): performance improvement by avoiding
+	  array creation if rb_block_arity() > 1.
+
+Fri Aug  2 23:50:53 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (power_cache_get_power): Apply bigtrunc to the result of
+	  bigsq.
+	  (big2str_karatsuba): Fix number of leading zero characters.
+
+Fri Aug  2 23:48:36 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): calculate denominator directly as powers of
+	  ten, not parsing string.
+
+	* parse.y (parser_number_literal_suffix): return bit set of found
+	  suffixes.
+
+	* parse.y (parser_set_number_literal, parser_set_integer_literal):
+	  split from parser_number_literal_suffix to set yylval.
+
+	* parse.y (parser_yylex): parse rational number literal with decimal
+	  point precisely.
+
+	* parse.y (simple_numeric): integrate numeric literals and simplify
+	  numeric rules.
+
+	* ext/ripper/eventids2.c (ripper_init_eventids2): ripper support for
+	  new literals, tRATIONAL and tIMAGINARY.
+
+Fri Aug  2 18:33:28 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_karatsuba): Reduce power_level more than one at
+	  recursion, if possible.
+	  (rb_big2str1): Follow the above change.
+
+Fri Aug  2 12:25:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul): Swap x and y for bary_mul1 if x is longer than y.
+	  [ruby-dev:47565] [Bug #8719] Reported by Narihiro Nakamura.
+
+Fri Aug  2 10:39:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* parse.y (negate_lit): add T_RATIONAL and T_COMPLEX to the switch
+	  statement, and call rb_bug() if an unknown type is passed to
+	  negate_lit(). [ruby-core:56316] [Bug #8717]
+
+	* bootstraptest/test_literal_suffix.rb (assert_equal): add test
+
+Fri Aug  2 09:14:47 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/syntax/refinements.rdoc:  Improve description of where you may
+	  activate refinements.
+
+Fri Aug  2 07:45:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_orig): Remove len argument.
+	  (big2str_karatsuba): Ditto.
+	  (rb_big2str1): Follow above change.
+
+Thu Aug  2 02:32:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: Add the description of number literal suffixes.
+
+Thu Aug  2 00:02:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bootstraptest/test_literal_suffix.rb: add two test cases to
+	  examine that "1if true" and "1rescue nil" are recognized as 1.
+
+Thu Aug  1 23:45:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* rational.c (rb_flt_rationalize_with_prec): new public C function
+	  to rationalize a Float instance with a precision.
+
+	* rational.c (rb_flt_rationalize): new public C function to
+	 rationalize a Float instance.  A precision is calculated from
+	 the given float number.
+
+	* include/ruby/intern.h: Add rb_flt_rationalize_with_prec and
+	  rb_flt_rationalize.
+
+	* parse.y: implement number literal suffixes, 'r' and 'i'.
+	  [ruby-core:55096] [Feature #8430]
+
+	* bootstraptest/test_literal_suffix.rb: add tests for parser to scan
+	  number literals with the above tsuffixes.
+
+Thu Aug  1 23:55:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2str1): Remove a local variable.
+
+Thu Aug  1 23:33:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_cstr_to_inum): Use power_cache_get_power.
+
+Thu Aug  1 21:02:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2str1): Raise an error for too big number.
+
+Thu Aug  1 20:46:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (power_cache_get_power): Hide cached Bignum objects.
+
+Thu Aug  1 19:15:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2str1): Remove non-trim mode.
+	  (rb_big2str0): Non-trim mode implemented here.
+	  (big2str_find_n1): Change the result type to long again.
+	  (big2str_base_powerof2): Don't take arguments: len and trim.
+	  (rb_big2str): Follow above change.
+
+Thu Aug  1 12:37:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_alloc): New function to allocate the result string.
+	  It is called after actual length is calculated.
+	  (big2str_struct): Add fields: negative, result and ptr.
+	  (big2str_orig): Write out the result via b2s->ptr.
+	  (big2str_orig): Ditto.
+	  (rb_big2str1): Don't allocate the result string at beginning.
+
+Thu Aug  1 07:36:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_orig): Use temporary buffer when trim mode.
+
+Thu Aug  1 06:28:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_orig): Simplified because RBIGNUM_LEN(x) <= 2 now.
+	  (big2str_struct): Two fields added: hbase2, hbase2_numdigits.
+	  (rb_big2str1): Initialize above fields.
+
+Thu Aug  1 04:06:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/options.rb (RDoc#finish): include root path in include
+	  paths, to work in another directory than the source directory.
+	  [ruby-core:56282] [Bug #8712]
+
+	* test/test_rdoc_markup_pre_process.rb (TestRDocMarkupPreProcess#setup):
+	  fix input_file_name, as the test script is not pre-processed.
+
+Thu Aug  1 01:45:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_karatsuba): Fix a condition of power_level.
+
+Thu Aug  1 01:09:02 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (LOG2_KARATSUBA_BIG2STR_DIGITS): Removed.
+	  (KARATSUBA_BIG2STR_DIGITS): Removed.
+	  (big2str_numdigits_cache): New variable.
+	  (power_cache_get_power): Merged with power_cache_get_power0.
+	  This function returns maxpow_in_bdigit_dbl(base)**(2**power_level).
+	  (rb_big2str1): use power_cache_get_power.
+
+Wed Jul 31 23:59:28 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_find_n1): Change the return type to size_t.
+	  (big2str_orig): Ditto.
+	  (big2str_karatsuba): Ditto.
+	  (rb_big2str1): Follow the above changes.
+
+Wed Jul 31 23:19:06 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (power_cache_get_power): Change numdigits_ret to size_t *.
+	  (big2str_orig): Change len argument to size_t.
+	  (big2str_karatsuba): Ditto.
+	  (rb_big2str1): Follow the above changes.
+
+Wed Jul 31 22:59:47 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/parse/test_notation_declaration.rb: Change class
+	  name to follow file name change.
+
+Wed Jul 31 22:57:50 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Rename to ...
+	* test/rexml/parse/test_notation_declaration.rb: ... this.
+
+Wed Jul 31 22:54:39 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_mixin.rb: Remove duplicated tests.
+
+Wed Jul 31 22:52:55 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Fix typos in expected
+	  value.
+	  pubilc ->
+	  public
+	     ^^
+
+Wed Jul 31 22:50:51 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Add tests that focus
+	  system literal in external ID system notation declaration.
+
+Wed Jul 31 22:36:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_cmp): Extracted from rb_big_cmp.
+	  (power_cache_get_power): Change n1 argument (number of digits) to
+	  power_level which is just passed to power_cache_get_power0.
+	  (big2str_karatsuba): Ditto.
+	  (rb_big2str1): Calculate the initial power_level.
+
+Wed Jul 31 22:04:36 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Fix a typo.
+	  Extern  ID ->
+	  ExternalID
+	        ^^
+
+Wed Jul 31 22:01:36 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Add tests that focus
+	  public ID in external ID notation declaration.
+
+Wed Jul 31 22:01:24 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* parse.y: fix build error with bison-3.0.
+
+Wed Jul 31 21:58:53 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Split test patterns.
+
+Wed Jul 31 21:42:33 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Group tests.
+
+Wed Jul 31 21:37:51 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_mixin.rb (TestNotationDecl#test_name):
+	  Move to ...
+	* test/rexml/test_notationdecl_parsetest.rb
+	  (TestNotationDecl#test_name): ... here.
+
+Wed Jul 31 21:37:47 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+Wed Jul 31 21:31:49 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: Remove setup because it
+	  doesn't share anything with other tests.
+
+Wed Jul 31 21:24:55 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_attributes_mixin.rb: Remove a needless shebang.
+	* test/rexml/test_notationdecl_mixin.rb: ditto.
+	* test/rexml/test_doctype.rb: ditto.
+	* test/rexml/test_xml_declaration.rb: ditto.
+	* test/rexml/test_changing_encoding.rb: ditto.
+
+Wed Jul 31 21:20:08 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_notationdecl_parsetest.rb: remove a needless shebang.
+
+Wed Jul 31 20:11:01 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_rindex): fix bug introduced in r42269.
+	  "".rindex("") should return 0.
+	  (str_rindex): ditto.
+
+Wed Jul 31 19:55:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (MAX_BIG2STR_TABLE_ENTRIES): Use SIZEOF_SIZE_T.
+	  (power_cache_get_power0): Add rb_bug call for too bit i argument.
+	  (power_cache_get_power): Simplified.
+
+Wed Jul 31 18:32:25 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/uri/common.rb (URI.decode_www_form_component): Use String#b.
+
+Wed Jul 31 18:24:02 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refine, mod_using, top_using): don't show
+	  warnings because Refinements are no longer experimental.
+	  [ruby-core:55993] [Feature #8632]
+
+	* test/ruby/test_refinement.rb: related test.
+
+	* NEWS: fixes for the above change.
+
+Wed Jul 31 17:55:55 2013  Shota Fukumori  <her@sorah.jp>
+
+	* lib/uri/common.rb (URI.decode_www_form_component):
+	  Don't raise error when str includes multibyte characters.
+
+Wed Jul 31 17:45:39 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_rindex): performance improvement by using
+	  memrchr(3).
+
+Wed Jul 31 16:43:30 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_rindex): refactoring and avoid to call str_nth() if
+	  pos == 0.
+
+Wed Jul 31 14:41:36 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb: [DOC] Add a couple of notes on Hash as storage.
+	  ref. [Feature #6589]
+
+Wed Jul 31 14:38:52 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb: [DOC] Fix example result.  Hash is now ordered.
+
+Wed Jul 31 14:38:10 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb: [DOC] Use the term "sorted" instead of "ordered"
+	  when mentioning SortSet.
+
+Wed Jul 31 12:18:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_struct): New structure.
+	  (big2str_orig): Use big2str_struct.
+	  (big2str_karatsuba): Ditto.
+	  (rb_big2str1): Ditto.
+
+Wed Jul 31 12:02:16 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rubygems.rb: [DOC] typo in url patch by @Red54 [Fixes #369]
+	  https://github.com/ruby/ruby/pull/369
+
+Wed Jul 31 07:09:07 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Import RubyGems from master as of commit 523551c
+	* test/rubygems:  ditto.
+
+Tue Jul 30 22:21:54 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* test/ruby/test_hash.rb: add a test for enumeration order of Hash.
+
+Tue Jul 30 18:52:27 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#intersect?, Set#disjoint?): Add new methods for
+	  testing if two sets have any element in common.
+	  [ruby-core:45641] [Feature #6588] Based on the code by marcandre.
+
+Tue Jul 30 17:16:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (ruby__sfvextra): add QUOTE flag to escape unprintable
+	  characters.
+
+Tue Jul 30 11:00:52 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/curses/extconf.rb: [DOC] nodoc to reduce Object pollution
+
+Tue Jul 30 08:19:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* sizes.c (Init_sizes): Define sizes only if the type actually exists.
+
+Mon Jul 29 22:55:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sizes.c (Init_sizes): define RbConfig::SIZEOF.  [Feature #8568]
+
+Mon Jul 29 22:25:20 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/curses/curses.c: [DOC] Update location of samples
+	* samples/curses/*: Move Curses samples and refactor from mixin
+	  The samples are included in rdoc for module and use of mixin is
+	  confusing
+
+Mon Jul 29 22:16:11 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (LOG2_KARATSUBA_BIG2STR_DIGITS): Renamed from
+	  LOG2_KARATSUBA_DIGITS.
+	  (KARATSUBA_BIG2STR_DIGITS): Renamed from KARATSUBA_DIGITS.
+
+Mon Jul 29 22:04:45 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_compare_by_id): add function prototype.
+
+Mon Jul 29 21:53:41 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_compare_by_id): don't call rb_hash_rehash()
+	  if self.compare_by_identity? == true.
+
+Mon Jul 29 21:29:48 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_assoc): performance improvement by replacing
+	  compare function in RHASH(hash)->ntbl->type temporarily like r42224.
+	  it falls back to rb_hash_foreach() if st_lookup() doesn't find the key.
+
+	* test/ruby/test_hash.rb: add a test for above.
+
+Mon Jul 29 21:15:30 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/ruby/test_lazy_enumerator.rb
+	  (TestLazyEnumerator#test_initialize): Make sure
+	  Enumerator::Lazy#initialize raises error if the object is
+	  frozen.  The check was performed by rb_ivar_set() before
+	  rb_check_frozen() was added to enumerator_init().
+
+Mon Jul 29 21:06:42 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enumerator.c (enumerator_init): Add a frozenness check to
+	  prevent a frozen Enumerator object from being reinitialized with
+	  a different enumerable object.  This is the least we should do,
+	  and more fixes will follow. [Fixes GH-368] Patch by Kenichi
+	  Kamiya.
+
+	* enumerator.c (generator_init): Ditto.
+
+Mon Jul 29 20:14:24 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_assoc): revert r42224. table->type->compare is
+	  called only if hashes are matched.
+
+	* test/ruby/test_hash.rb: add a test to check using #== to compare.
+
+Mon Jul 29 17:00:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (yycompile): store file name as String to keep the encoding.
+
+	* parse.y (rb_parser_compile_string_path, rb_parser_compile_file_path):
+	  new functions to pass file name as a String.
+
+	* parse.y (gettable_gen): return a copy of the original file name, not
+	  a copy in filesystem encoding.
+
+	* vm_eval.c (eval_string_with_cref): use Qundef instead of "(eval)".
+
+Mon Jul 29 16:53:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_initialize_copy): copy st_table type even if empty.
+	  [ruby-core:56256] [Bug #8703]
+
+Mon Jul 29 16:34:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_initialize_copy): clear old table before copy new
+	  table.
+
+Mon Jul 29 16:34:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_assoc): aggregate object can be initialized only
+	  with link time constants.
+
+Mon Jul 29 14:54:44 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_assoc): performance improvement by replacing
+	  compare function in RHASH(hash)->ntbl->type temporarily.
+
+Mon Jul 29 14:52:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (xsystem): expand environment variable in all macros not
+	  expanded with RbConfig.  [Bug #8702]
+
+	* test/mkmf/test_framework.rb (create_framework): replace all $@ not
+	  only once.
+
+Mon Jul 29 06:54:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pipe): use enum for compile time constants,
+	  instead of const int for debugging.
+
+Mon Jul 29 00:11:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Specialized implementation added for
+	  nx == 2 && ny == 2
+
+Sun Jul 28 20:28:41 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* io.c (io_getpartial): use rb_str_locktmp_ensure().
+	  [ruby-core:56121] [Bug #8669]
+
+	* io.c (rb_io_sysread): ditto.
+
+	* test/ruby/test_io.rb: add tests for above.
+
+Sun Jul 28 20:10:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): should make static libraries for extensions
+	  to be statically linked.  [Bug #7948]
+
+Sun Jul 28 17:38:32 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c: add internal API rb_str_locktmp_ensure().
+
+	* io.c (io_fread): use rb_str_locktmp_ensure().
+	  [ruby-core:56121] [Bug #8669]
+
+	* test/ruby/test_io.rb: add a test for above.
+
+Sun Jul 28 13:04:39 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* io.c (interpret_seek_whence): support SEEK_DATA and SEEK_HOLE.
+	  These are whences for lseek(2) supported by Linux since version 3.1.
+	  [ruby-core:56123] [Feature #8671]
+
+	* test/ruby/test_io.rb: Add tests for above.
+
+Sun Jul 28 12:41:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (absint_numwords_generic): The char_bit variable changed
+	  to static constant.
+
+Sun Jul 28 12:03:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Constify bary_* functions.
+
+Sun Jul 28 11:12:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_absint_size): Declaration moved from
+	  internal.h to calculate required buffer size to pack integers.
+	  (rb_absint_numwords): Ditto.
+	  (rb_absint_singlebit_p): Ditto.
+	  [ruby-core:42813] [Feature #6065]
+
+Sun Jul 28 10:54:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pipe): fix pipe name formatting.  as "%x" may
+	  not contain '0' at all, fill at fixed position instead.
+
+Sun Jul 28 00:35:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_size): Return the bignum "bytewise" size.
+	  [ruby-core:55578] [Feature #8553]
+	  This is accepted by matz on DevelopersMeeting20130727Japan.
+
+Sun Jul 28 00:07:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_integer_pack): Declaration moved from
+	  internal.h.
+	  (rb_integer_unpack): Ditto.
+	  [ruby-core:42813] [Feature #6065]
+
+Fri Jul 26 23:18:13 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS: Add a new feature that REXML::Parsers::StreamParser
+	  supports "entity" event.
+
+Fri Jul 26 23:14:31 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/parsers/streamparser.rb
+	  (REXML::Parsers::StreamParser#parse): Add "entity" event support to
+	  listener. [Bug #8689] [ruby-dev:47542]
+	  Reported by Ippei Obayashi.
+	* test/rexml/test_stream.rb (StreamTester#entity): Add a test for
+	  the above case.
+
+Fri Jul 26 23:05:27 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): separate numeric literal from succeeding
+	  token, and treat 'e' as floating point number only if followed by
+	  exponent part.
+
+Fri Jul 26 22:14:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_exec.h (CHECK_VM_STACK_OVERFLOW_FOR_INSN): surround with
+	  do/while (0), and remove unnecessary casts.
+
+Fri Jul 26 20:12:07 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/syslog/lib/syslog/logger.rb (Syslog::Logger): Add facility
+	  to Syslog::Logger. [Fixes GH-305] patch by Max Shytikov
+	  https://github.com/ruby/ruby/pull/305
+
+Fri Jul 26 19:25:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_exec.h, tool/instruction.rb: not an error, but a BUG if stack
+	  overflow checking failed just before/after the beginning of an
+	  instruction. It should be treated as a BUG.
+	  Please tell us if your code cause BUG with this problem.
+	  This check will removed soon (for performance).
+
+Fri Jul 26 18:30:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_memcpy): cast to int to suppress a warning.
+
+Fri Jul 26 18:21:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_memcpy): try to enable optimization.
+	  At least on my environments, I don't see any errors
+	  with many trials. Please tell us if you find any GC bugs.
+
+Fri Jul 26 17:49:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (fix_string_encoding): fix target encoding.  the
+	  parameter `encoding' is not the target encoding but the original
+	  encoding.
+
+Fri Jul 26 14:05:19 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/fiddle/*: [DOC] More doc on dlopen and RTLD_DEFAULT from r42184
+
+Fri Jul 26 13:08:53 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/fiddle/lib/fiddle.rb: [DOC] Document Fiddle.dlopen(nil)
+	* ext/fiddle/handle.c: [DOC] Document Fiddle::Handle.new(nil)
+
+Fri Jul 26 13:04:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_load_internal): use rb_load_file_str() to keep path
+	  encoding.
+
+	* load.c (rb_require_safe): search in OS path encoding for Windows.
+
+	* ruby.c (rb_load_file_str): load file with keeping path encoding.
+
+	* win32/file.c (rb_file_load_ok): use WCHAR type API assuming incoming
+	  path is encoded in UTF-8.  [ruby-core:56136] [Bug #8676]
+
+	* file.c (rb_str_encode_ospath): simplify using rb_str_conv_enc().
+
+	* win32/file.c (fix_string_encoding): simplify with rb_str_conv_enc().
+
+	* win32/file.c (convert_mb_to_wchar): use bare pointer instead of
+	  VALUE, and remove useless argument.
+
+Fri Jul 26 11:42:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* rational.c (f_round_common): Rational is expected to be returned by
+	  Rational#*, but mathn.rb breaks that assumption.  [ruby-core:56177]
+	  [Bug #8687]
+
+Fri Jul 26 01:37:45 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h: check defined(USE_RGENGC_LOGGING_WB_UNPROTECT)
+
+Fri Jul 26 01:21:41 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_file_expand_path_internal): fix r42160; skip '~'.
+
+Thu Jul 25 17:53:18 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#connect): disable Nagle's algorithm on
+	  HTTP connection. [ruby-core:56158] [Feature #8681]
+
+Thu Jul 25 17:49:42 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_to_s): convert closing parenthesis to the target encoding
+	  if it is ASCII incompatible encoding. [ruby-core:56063] [Bug #8650]
+
+Thu Jul 25 17:21:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (is_obj_encoding): new macro to check if obj is an
+	  Encoding.  obj can be any type while is_data_encoding expects T_DATA
+	  only.
+
+Thu Jul 25 17:17:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_expand_path_internal): should clear coderange after
+	  copying user name as binary data.
+
+Thu Jul 25 16:17:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* encoding.c (check_encoding): Check T_DATA or not.
+	  is_data_encoding(obj) assumes that `obj' is T_DATA.
+
+Thu Jul 25 13:06:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_s_home): use rb_home_dir_of and rb_default_home_dir.
+
+	* file.c (rb_home_dir_of): split from rb_home_dir() for the home
+	  directry of the given user, and the user name is a VALUE, not a bare
+	  pointer.  should raise if the user does not exist.
+
+	* file.c (rb_default_home_dir): split from rb_home_dir() for the home
+	  directry of the current user.
+
+Thu Jul 25 12:32:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/openssl/ossl.c: support additional three thread synchronization
+	  functions. [ruby-trunk - Bug #8386]
+
+Thu Jul 25 07:15:58 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Import RubyGems from master as of commit 4ff70cc
+	* test/rubygems:  ditto.
+
+Wed Jul 24 20:57:44 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_set_arguments): use RARRAY_RAWPTR() instead of
+	  RARRAY_PTR() because there is no new reference.
+
+	* compile.c (iseq_set_exception_table): ditto.
+
+Wed Jul 24 19:49:54 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (find_proxy): raise BadURIError if the URI is
+	  a relative URI. [Bug #8645]
+
+Wed Jul 24 18:56:06 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_expandarray): use RARRAY_RAWPTR() instead of
+	  RARRAY_PTR() because there is no new reference.
+
+	* vm_insnhelper.c (vm_caller_setup_args): ditto.
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): ditto.
+
+Wed Jul 24 18:40:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c, gc.c: move ary_unprotect_logging() into
+	  rb_gc_unprotect_logging() which is general version
+
+	* include/ruby/ruby.h: add USE_RGENGC_LOGGING_WB_UNPROTECT
+	  to enable.
+
+Wed Jul 24 17:37:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_expand_path_internal): preserve the file name
+	  encoding in an exception message.
+
+Wed Jul 24 08:04:49 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/-ext-/tracepoint/test_tracepoint.rb: add GC on/off to count
+	  GC events strictly.
+
+Tue Jul 23 23:19:24 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/extconf.rb (CRYPTO_THREADID): check exist or not.
+
+	* ext/openssl/ossl.c (ossl_thread_id): use rb_nativethread_self()
+	  implemented at r42137 to allow threads which doesn't associated with
+	  Ruby thread to use openssl functions.
+
+	* ext/openssl/ossl.c (Init_ossl_locks): If CRYPTO_THREADID is defined
+	  (OpenSSL 1.0.0 or later has it) use CRYPTO_THREADID_set_callback()
+	  instead of CRYPTO_set_id_callback() because its argument is
+	  unsigned long; it may cause id collision on mswin64
+	  whose sizeof(unsigned long) < sizeof(void*).
+	  http://www.openssl.org/docs/crypto/threads.html
+
+	* ext/openssl/ossl.c (ossl_threadid_func): defined for above.
+
+Tue Jul 23 20:47:36 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Move functions.
+
+Tue Jul 23 20:14:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_divmod): Add special cases for x < y easily detected
+	  and nx == 2 && ny == 2.
+
+Tue Jul 23 19:48:38 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_(pthread|win32).h: rename rb_thread_cond_t to
+	  rb_nativethread_cond_t.
+
+	* thread.c, thread_pthread.c, thread_win32.c, vm_core.h: catch up
+	  renaming.
+
+Tue Jul 23 19:44:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_native.h: add rb_nativethread_self() which returns
+	  current running native thread identifier.
+
+	* thread_[pthread|win32].c: implement rb_nativethread_self().
+
+Tue Jul 23 19:34:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.h, thread_win32.h: rename rb_thread_id_t to
+	  rb_nativethread_id_t.
+
+	* thread_pthread.c, vm_core.h: use rb_nativethread_id_t.
+
+Tue Jul 23 18:56:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/openssl/ossl.c: use system native (system provided)
+	  thread locking APIs added by last commit.
+	  This patch fixes [Bug #8386].
+	  "rb_mutex_*" APIs control only "Ruby" threads.
+	  Not for native threads.
+
+Tue Jul 23 18:44:15 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_native.h: added.
+	  Move native thread related lines from vm_core.h.
+	  And declare several functions "rb_nativethread_lock_*",
+	  manipulate locking.
+
+	* common.mk: add thread_native.h.
+
+	* thread.c: add functions "rb_nativethread_lock_*".
+
+	* thread.c, thread_[pthread,win32].[ch]: rename rb_thread_lock_t
+	  to rb_nativethread_lock_t to make it clear that this lock is for
+	  native threads, not for ruby threads.
+
+Tue Jul 23 16:14:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_before_sweep): fix spacing.
+
+Tue Jul 23 15:57:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_get_freeobj): clear slot->freelist here.
+	  This means that this slot doesn't have any free objects.
+	  And store this slot with objspace->heap.using_slot.
+
+	* gc.c (gc_before_sweep): restore objspace->freelist
+	  into objspace->heap.using_slot->freelist.
+	  This means that using_slot has free objects which are
+	  pointed from objspace->freelist.
+
+	* gc.c (gc_slot_sweep): do not need to clear slot->freelist.
+
+Tue Jul 23 09:34:49 2013  Zachary Scott  <e@zzak.io>
+
+	* sample/drb/README*.rdoc: [DOC] migrate DRb sample READMEs to rdoc
+
+Tue Jul 23 09:28:05 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/drb/invokemethod.rb: [DOC] nodoc InvokeMethod18Mixin
+
+Tue Jul 23 08:44:37 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_asn1.c (asn1time_to_time):  Implement YYMMDDhhmmZ
+	  format for ASN.1 UTCTime.  [ruby-trunk - Bug #8664]
+	* test/openssl/test_asn1.rb:  Test for the above.
+
+Tue Jul 23 08:11:32 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rexml/streamlistener.rb: [DOC] Fix examples in
+	  REXML::StreamListener#entitydecl patch by Ippei Obayashi [Bug #8665]
+
+Tue Jul 23 07:44:59 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Import RubyGems from master as of commit b165260
+	* test/rubygems:  ditto.
+
+Tue Jul 23 07:14:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mulsub_1xN): New function.
+	  (bary_mul_toom3): Use bary_mulsub_1xN.
+
+Tue Jul 23 03:32:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (KARATSUBA_BALANCED): New macro.
+	  (TOOM3_BALANCED): Ditto.
+	  (bary_mul_balance_with_mulfunc): Use KARATSUBA_BALANCED and
+	  TOOM3_BALANCED.
+	  (rb_big_mul_balance): Relax a condition.
+	  (rb_big_mul_karatsuba): Use KARATSUBA_BALANCED.
+	  (rb_big_mul_toom3): Use TOOM3_BALANCED.
+	  (bary_mul_karatsuba_branch): Use KARATSUBA_BALANCED.
+	  (bary_mul_toom3_branch): Use TOOM3_BALANCED.
+
+Tue Jul 23 01:34:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem_mulsub): Extracted from bigdivrem1.
+	  (bigdivrem1): Use bary_add.
+
+Mon Jul 22 18:39:52 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_enumerate_chars): specify array capa
+	  with str_strlen().
+
+	* string.c (rb_str_enumerate_codepoints): ditto.
+
+Mon Jul 22 18:01:33 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_enumerate_chars): specify array capa.
+
+Mon Jul 22 17:24:14 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_each_char_size): performance improvement by
+	  using rb_str_length().
+
+Mon Jul 22 16:32:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (eval_string_with_cref): check by Check_TypedStruct
+	  instead of rb_obj_is_kind_of.
+
+Mon Jul 22 13:19:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_resize_capa): use RARRAY_RAWPTR() because
+	  this code creates no new references.
+
+Mon Jul 22 12:58:18 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_memfill): added.
+
+	* array.c (rb_ary_initialize): use ary_memfill().
+
+	* array.c (rb_ary_fill): ditto.
+
+	* array.c (rb_ary_slice_bang): use RARRAY_RAWPTR() because
+	  this code creates no new references.
+
+Mon Jul 22 10:09:46 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_slot_sweep): need to add empty RVALUE as freeobj.
+
+Mon Jul 22 09:48:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (eval_string_with_cref): use the given file name unless
+	  eval even if scope is given.  additional fix for [Bug #8436].
+	  based on the patch by srawlins at [ruby-core:56099] [Bug #8662].
+
+Mon Jul 22 09:24:19 2013  Kouji Takao  <kouji@takao7.net>
+
+	* ext/readline/readline.c (Init_readline): added
+	  Readline.delete_text. [ruby-dev:45789] [Feature #6626]
+	* ext/readline/extconf.rb: check for rl_delete_text() in Readline library.
+
+	  Thanks, Nobuyoshi Nakada, for the patch.
+
+Mon Jul 22 03:15:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_parse.c (rfc2822_cb): check if wday is given, since it
+	  can be omitted.
+
+Mon Jul 22 00:15:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_sq_fast): Refine expressions.
+
+Sun Jul 21 21:08:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul): Use simple multiplication if yl is small.
+	  (rb_cstr_to_inum): Invoke bigsq instead of bigmul0.
+	  (bigsq): Re-implemented.
+	  (bigmul0): Invoke bigsq if two arguments are identical.
+
+Sun Jul 21 09:58:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_toom3): New function based on bigmul1_toom3.
+	  (bary_mul_toom3_branch): Call bary_mul_toom3.
+	  (rb_big_mul_toom3): Ditto.
+	  (bigmul1_toom3): Removed.
+	  (big_real_len): Ditto.
+	  (big_split): Ditto.
+	  (big_split3): Ditto.
+
+Sun Jul 21 08:12:16 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* proc.c (proc_to_s): use PRIsVALUE to preserve the result encoding.
+
+Sun Jul 21 03:36:18 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* hash.c (rb_hash_flatten): use NUM2INT to raise TypeError on 32bit
+	  platform. it's introduced by r42039
+
+Sun Jul 21 01:07:45 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* common.mk (help): Fix environment variable name and argument.
+	  Actually it can also be a directory or any argument for
+	  test/unit runner. [Fixes GH-363]
+
+Sat Jul 20 22:44:50 2013  Zachary Scott  <e@zzak.io>
+
+	* common.mk: Document running a single test [Fixes GH-363]
+	  Patch by Avdi Grimm https://github.com/ruby/ruby/pull/363
+
+Sat Jul 20 22:39:56 2013  Zachary Scott  <e@zzak.io>
+
+	* sample/*: whitespace patch by Sergio Campama [Fixes GH-364]
+	  https://github.com/ruby/ruby/pull/364
+
+Sat Jul 20 22:33:13 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/regexp.rdoc: [DOC] Fix typo in example [Fixes GH-365]
+	  Patch by Juanito Fatas https://github.com/ruby/ruby/pull/365
+
+Sat Jul 20 17:46:03 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_succ): add missing case NEIGHBOR_WRAPPED.
+	  r42078 caused buggy behavior like "\xFF".b -> "\x01\xFF".b
+
+Sat Jul 20 15:22:38 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_resize): use simple memcpy because there are no new
+	  references.
+
+Sat Jul 20 15:02:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* safe.c (ruby_safe_level_4_warning): define for old extension
+	  libraries.  [Bug #8652]
+
+Sat Jul 20 14:38:00 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_make_shared): make shared array shady.
+	  Making non-shady shared array causes SEGV (see rubyci).
+	  It seems a bug around shared array.
+
+Sat Jul 20 12:14:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (enc_succ_char, enc_pred_char): consider wchar case.
+	  [ruby-core:56071] [Bug #8653]
+
+	* string.c (rb_str_succ): do not replace with invalid char.
+
+	* encoding.c (rb_enc_code_to_mbclen): add new function which returns
+	  mbclen from codepoint like as rb_enc_codelen() but 0 for invalid
+	  char.
+
+	* include/ruby/encoding.h (rb_enc_code_to_mbclen): declaration and
+	  shortcut macro.
+
+Fri Jul 19 21:59:12 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: declare type_name() at the beginning of file.
+
+Fri Jul 19 21:35:09 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c: reduce shady operations.
+
+	* array.c (rb_ary_modify, ary_make_partial, rb_ary_splice,
+	  rb_ary_replace, rb_ary_eql, rb_ary_compact_bang):
+	  use RARRAY_RAWPTR() instead of RARRAY_PTR().
+
+	* array.c (rb_ary_shift): use RARRAY_PTR_USE() without WB because
+	  there are not new relations.
+
+	* array.c (ary_ensure_room_for_unshift): ditto.
+
+	* array.c (rb_ary_sort_bang): ditto.
+
+	* array.c (rb_ary_delete_at): ditto.
+
+	* array.c (rb_ary_reverse_m): use RARRAY_RAWPTR() because
+	  there are not new relations.
+
+Fri Jul 19 20:58:20 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c: reduce shade operations.
+
+	* array.c (rb_ary_modify): use RARRAY_RAWPTR().
+
+	* array.c (ary_make_substitution, rb_ary_s_create, ary_make_partial,
+	  rb_ary_splice, rb_ary_resize, rb_ary_rotate_m, rb_ary_times):
+	  use ary_memcpy().
+
+Fri Jul 19 19:55:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_mem_clear): added. This operation doesn't need WB
+	  because this operation creates a reference to Qnil.
+
+	* array.c (ary_make_shared, rb_ary_store, rb_ary_shift_m,
+	  rb_ary_splice, rb_ary_resize, rb_ary_fill): use ary_mem_clear()
+	  instead of rb_mem_clear().
+
+	* array.c (ary_make_shared): use RARRAY_RAWPTR() instead of RARRAY_PTR().
+
+Fri Jul 19 19:18:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c: fix commit miss.
+	  RGENGC_UNPROTECT_LOGGING should be 0.
+
+Fri Jul 19 19:15:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_resurrect): use RARRAY_RAWPTR() because there is no
+	  writing.
+
+	* array.c (rb_ary_new_from_values): use ary_memcpy().
+
+Fri Jul 19 19:07:31 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_memcpy): add a function to copy VALUEs into ary
+	  with write barrier. If ary is promoted, use write barrier correctly.
+
+	* array.c (rb_ary_cat, rb_ary_unshift_m, rb_ary_dup,
+	  rb_ary_sort_bang, rb_ary_replace, rb_ary_plus): use ary_memcpy().
+
+Fri Jul 19 15:32:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_store): use RARRAY_PTR_USE() instead of RARRAY_PTR().
+	  Clearing memory space doesn't need WBs.
+
+Fri Jul 19 15:19:37 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_ensure_room_for_push): use RARRAY_RAWPTR() instead of
+	  RARRAY_PTR. In this code, there are no "write" operation.
+
+	* array.c (rb_ary_equal): ditto.
+
+	* array.c (recursive_equal): ditto.
+
+Fri Jul 19 15:09:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, internal.h (rb_gc_writebarrier_remember_promoted): add a new
+	  function to remember an specified object. This api is only
+	  experimental (strongly depend on WB/rgengc strategy).
+
+Fri Jul 19 14:56:00 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_unprotect_logging): use (void *) for first parameter
+	  because VALUE is not defined before including ruby/ruby.h.
+
+Fri Jul 19 14:19:48 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/pathname/pathname.c (path_inspect): use PRIsVALUE to preserve
+	  the result encoding.
+
+Fri Jul 19 12:35:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_tcp.rb (test_initialize_failure): Use EADDRNOTAVAIL
+	  to test an error message generated by bind() failure.
+
+Fri Jul 19 11:27:38 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/racc/parser.rb: [DOC] Capitalize "Ruby" in documentation
+	  Patch by Dave Worth https://github.com/ruby/ruby/pull/341
+
+Fri Jul 19 11:26:28 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/psych/lib/psych*: [DOC] Capitalize "Ruby" in documentation
+	  Patch by Dave Worth https://github.com/ruby/ruby/pull/341
+
+Fri Jul 19 11:25:12 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rdoc/*: [DOC] Capitalize "Ruby" in documentation
+	  Patch by Dave Worth https://github.com/ruby/ruby/pull/341
+
+Fri Jul 19 11:23:55 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rubygems*: [DOC] Capitalize "Ruby" in documentation
+	  Patch by Dave Worth https://github.com/ruby/ruby/pull/341
+
+Fri Jul 19 11:16:54 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#to_set): Define Set#to_set so that aSet.to_set
+	  returns self. [Fixes GH-359]
+
+Fri Jul 19 11:10:23 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rake/*: [DOC] Capitalize "Ruby" in documentation
+	  Patch by Dave Worth https://github.com/ruby/ruby/pull/341
+
+Fri Jul 19 01:04:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/bignum/intpack.c: Renamed from ext/-test-/bignum/pack.c.
+	  (Init_intpack): Renamed from Init_pack.
+	  Reported by Naohisa Goto.  [ruby-dev:47526] [Bug #8655]
+
+Fri Jul 19 00:54:27 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/ruby/test_array.rb (test_count): add a test case for #count
+	  with an argument. See Bug #8654.
+
+Thu Jul 18 23:45:06 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_eql): compare RARRAY_PTR() for performance
+	  improvement in case of that self and other are shared.
+
+Thu Jul 18 22:46:42 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/cgi.rb: [DOC] Capitalize "Ruby" in documentation [Fixes GH-341]
+	  Patch by Dave Worth https://github.com/ruby/ruby/pull/341
+	* lib/webrick.rb: ditto
+	* lib/scanf.rb: ditto
+	* lib/xmlrpc/config.rb: ditto
+	* lib/resolv.rb: ditto
+	* lib/e2mmap.rb: ditto
+	* lib/fileutils.rb: ditto
+	* lib/mkmf.rb: ditto
+	* lib/cgi/session.rb: ditto
+	* lib/yaml.rb: ditto
+	* lib/erb.rb: ditto
+	* lib/irb.rb: ditto
+	* lib/tracer.rb: ditto
+	* lib/net/http.rb: ditto
+	* ext/syslog/lib/syslog/logger.rb: ditto
+	* sample/pty/expect_sample.rb: ditto
+
+Thu Jul 18 21:30:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_sq_fast): Specialize the last iteration of the
+	  outer loop.
+	  (bigfixize): A condition simplified.
+
+Thu Jul 18 21:15:41 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_equal): compare RARRAY_PTR() for performance
+	  improvement in case of that self and other are shared.
+
+Thu Jul 18 20:44:51 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_fill): use memfill().
+
+Thu Jul 18 20:35:14 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* array.c (rb_ary_count): check length to avoid SEGV
+	  while iterating. Remove other pointer loop when arg is given.
+
+	* test/ruby/test_array.rb (test_count): add test for bug.
+	  [ruby-core:56072] [Bug #8654]
+
+Thu Jul 18 18:14:36 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_count): iterate items appropriately.
+	  [Bug #8654]
+
+Thu Jul 18 17:35:41 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_flatten): performance improvement by not using
+	  rb_hash_to_a() to avoid array creation with rb_assoc_new().
+
+Thu Jul 18 16:16:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c: add logging feature for RGenGC's write barrier unprotect
+	  event.
+
+Thu Jul 18 15:45:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): make only
+	  rb_set_safe_level(4) an error always but make rb_secure(4) an error
+	  only in the core.  [ruby-dev:47517] [Bug #8652]
+
+Thu Jul 18 15:42:01 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: fix spell miss.
+
+Thu Jul 18 15:11:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ruby_safe_level_4): get rid of special
+	  character.  [ruby-dev:47512] [misc #8646]
+
+Thu Jul 18 14:51:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ary_alloc): slim setup process.
+
+Thu Jul 18 14:37:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (str_alloc): no need to clear RString (already cleared).
+
+Thu Jul 18 12:57:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BDIGITS_ZERO): Defined.
+	  (bary_pack): Use BDIGITS_ZERO.
+	  (bary_unpack): Ditto.
+	  (bary_mul_single): Ditto.
+	  (bary_mul_normal): Ditto.
+	  (bary_sq_fast): Ditto.
+	  (bary_mul_balance_with_mulfunc): Ditto.
+	  (bary_mul_precheck): Ditto.
+	  (bary_mul_toom3_branch): Ditto.
+	  (rb_cstr_to_inum): Ditto.
+	  (big_shift3): Ditto.
+	  (bigmul1_toom3): Ditto.
+	  (bary_divmod): Ditto.
+
+Thu Jul 18 06:30:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename gc related functions with prefix "gc_".
+	  * before_gc_sweep() -> gc_before_sweep().
+	  * after_gc_sweep()  -> gc_after_sweep().
+	  * lazy_sweep()      -> gc_lazy_sweep().
+	  * rest_sweep()      -> gc_rest_sweep().
+	  * slot_sweep()      -> gc_slot_sweep().
+
+	* gc.c: rename a heap management function with prefix "heap_".
+	  * get_freeobj() -> heap_get_freeobj().
+
+	* gc.c: rename markable_object_p() to is_markable_object().
+
+Wed Jul 17 22:57:40 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (delete_if_i): use ST_DELETE.
+
+Wed Jul 17 22:34:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: An static assertion for relation of SIZEOF_LONG and
+	  SIZEOF_BDIGITS is added.
+	  (bary_mul_precheck): Reduce comparisons.
+	  (bary_mul): Invoke bary_sq_fast or bary_mul1 if the bignum size is
+	  small.
+	  (bigfixize): Resize the argument bignum here.
+	  (bignorm): Don't call bigtrunc after bigfixize.
+
+Wed Jul 17 22:13:26 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_replace): performance improvement by using
+	  st_copy().
+
+Wed Jul 17 17:19:54 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename heap management functions with prefix "heap_".
+	  * allocate_sorted_array() -> heap_allocate_sorted_array().
+	  * slot_add_freeobj()      -> heap_slot_add_freeobj().
+	  * assign_heap_slot()      -> heap_assign_slot().
+	  * add_heap_slots()        -> heap_add_slots().
+	  * init_heap()             -> heap_init().
+	  * set_heap_increment()    -> heap_set_increment().
+
+	* gc.c (initial_expand_heap): inlined in rb_gc_set_params().
+
+Wed Jul 17 17:12:23 2013  Matthew M. Boedicker  <matthewm@boedicker.org>
+
+	* hash.c (env_fetch): Add key name to message on ENV.fetch KeyError,
+	  as well as Hash#fetch.  [ruby-core:56062] [Feature #8649]
+
+Wed Jul 17 15:59:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: catch up last changes for debugging/checking mode.
+
+Wed Jul 17 15:50:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_free): free slot itself.
+
+	* gc.c (objspace_each_objects): fix condition.
+	  Use slot->body instead of slot.
+
+	* gc.c (count_objects): use "slot" variable.
+
+Wed Jul 17 15:21:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (unlink_heap_slot): fix memory leak.
+	  free slot itself at free_heap_slot().
+
+	  Reproduce-able code is here:
+	    N1 = 100_000; N2 = 1_000_000
+	    N1.times{ary = []; N2.times{ary << ''}}
+	  Maybe this problem is remaining in Ruby 2.0.0.
+
+	* gc.c (unlink_heap_slot): remove not working code.
+
+Wed Jul 17 14:31:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: re-design the heap structure.
+
+	  (1) The heap is consists of a set of slots.
+	  (2) Each "slot" has a "slot_body".
+	      slot::start and slot::limit specify RVALUE beginning address
+	      and number of RVALUE in a "slot_body".
+	  (3) "slot_body" contains a pointer to slot (slot_body::header::slot)
+	      and an array of RVALUE.
+	  (4) heap::sorted is an array of "slots", sorted by an address of
+	      slot::body.
+
+	  See https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/GC_design
+	  for more details (figure).
+
+	* gc.c: Avoid "heaps" terminology. It is ambiguous.
+
+Wed Jul 17 13:29:16 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix heaps_header and heaps_slot to reduce memory consumption.
+	  (1) move heaps_header::start and limit to  heaps_slot.
+	  (2) remove heaps_header::end which can be calculated by start+limit.
+
+	* gc.c: catch up above change.
+
+Wed Jul 17 12:30:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/st.h (st_strcasecmp): Macro defined for compatibility.
+	  (st_strncasecmp): Ditto.
+
+Wed Jul 17 11:57:45 2013  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/util.rb (CGI::Util#escape, unescape): Avoid use of regexp
+	  special global variable. [Feature #8648] Thanks to fotos.
+
+Wed Jul 17 11:57:10 2013  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/erb.rb (ERB::Util#url_encode): Avoid use of regexp special global
+	  variable. [Feature #8648] Thanks to fotos.
+
+Wed Jul 17 08:12:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* st.c (st_locale_insensitive_strcasecmp): Renamed from st_strcasecmp.
+	  (st_locale_insensitive_strncasecmp): Renamed from st_strncasecmp.
+
+	* include/ruby/st.h: Follow above changes.
+
+	* include/ruby/ruby.h: Ditto.
+
+Wed Jul 17 00:14:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigmul1_toom3): Use bigdivrem_single instead of bigdivrem.
+	  (big_three): Removed.
+	  (Init_Bignum): Don't initialize big_three.
+
+Tue Jul 16 21:46:03 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* configure.in: revert r42008. strcasecmp() uses the current locale.
+
+	* include/ruby/ruby.h: ditto.
+
+	* st.c (st_strcasecmp): ditto.
+
+Tue Jul 16 21:07:04 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* configure.in: check strcasecmp().
+
+	* include/ruby/ruby.h: use strcasecmp() as st_strcasecmp() if it
+	  exists.
+
+	* st.c (st_strcasecmp): define the function only if strcasecmp()
+	  doesn't exist.
+
+Tue Jul 16 20:21:28 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigsq): Renamed from bigsqr.
+
+Tue Jul 16 19:42:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (USHORT): Unused macro removed.
+
+Tue Jul 16 19:18:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: slim a path of newobj_of().
+
+	* gc.c (objspace): add a new field objspace::freelist, which contains
+	  available RVALUEs.
+
+	* gc.c (newobj_of): simply call new function `get_freeobj()'.
+	  get_freeobj() returns objspace::freelist. If  objspace::freelist
+	  is not available, refill objspace::freelist with a slot pointed by
+	  objspace::heap::free_slots.
+
+	* gc.c (before_gc_sweep): clear objspace::freelist.
+
+	* gc.c (slot_sweep): clear slot::freelist.
+
+	* gc.c (heaps_prepare_freeslot): renamed to heaps_prepare_freeslot.
+
+	* gc.c (unlink_free_heap_slot): remove unused function.
+
+	* gc.c (rb_free_const_table): remove unused function.
+
+Tue Jul 16 19:05:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_shift3): Big shift width is not a problem for right
+	  shift.
+
+Tue Jul 16 18:50:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_count): [DOC] fix typo.  Array#count uses ==, not
+	  ===.  a question at asakusa.rb ML.
+
+Tue Jul 16 18:35:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_karatsuba): Avoid duplicate calculation when
+	  squaring.
+	  (bary_mul_toom3_branch): Ditto.
+
+Tue Jul 16 17:43:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (link_free_heap_slot): removed.
+
+	* gc.c (slot_sweep): use `heaps_add_freeslot' instead of
+	  `link_free_heap_slot'.
+
+	* gc.c (assign_heap_slot): use local variable `slot' instead of
+	  `heaps'.
+
+Tue Jul 16 17:21:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (assign_heap_slot): refactoring variable names.
+
+	* gc.c (slot_add_freeobj): added.
+
+	* gc.c (heaps_add_freeslot): added.
+
+	* gc.c (finalize_list, rb_gc_force_recycle, slot_sweep): use
+	  `slot_add_freeobj' instead of modifying linked list directly.
+
+Tue Jul 16 16:30:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (lazy_sweep): refactoring.
+
+Tue Jul 16 13:32:06 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_set_index): since r41967, old terminator is dealt
+	  with in str_fill_term().  should not consider it here because this
+	  function is called before any encoding is set.
+
+Tue Jul 16 11:12:03 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* proc.c (rb_block_arity): raise ArgumentError if no block given.
+
+Tue Jul 16 08:15:22 2013  Zachary Scott  <e@zzak.io>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb: [DOC] document top-level
+	  classes from BigDecimal utils native extensions
+
+Tue Jul 16 03:23:03 2013  Zachary Scott  <e@zzak.io>
+
+	* numeric.c: [DOC] improve rdoc formatting for parameters and links
+
+Mon Jul 15 14:40:00 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h (rb_big2str0): Deprecated.
+
+	* bignum.c (rb_big2str1): Renamed from rb_big2str0.
+	  (rb_big2str0): Deprecated wrapper for rb_big2str1.
+	  (rb_big2str): Invoke rb_big2str1 instead of rb_big2str0.
+
+Mon Jul 15 14:13:02 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* struct.c (rb_struct_each_pair): use rb_yield_values(2, key, value)
+	  instead of rb_yield(rb_assoc_new(key, value)) if rb_block_arity()
+	  is greater than 1.
+
+Mon Jul 15 13:46:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Add static assertions.
+
+Mon Jul 15 13:36:02 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_each_pair): performance improvement by using
+	  rb_block_arity().
+
+Mon Jul 15 13:15:37 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* proc.c (rb_block_arity): create internal API rb_block_arity().
+	  it returns arity of given block.
+
+Mon Jul 15 13:07:27 2013  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* lib/prime.rb (Prime::EratosthenesGenerator,
+	  Prime::EratosthenesSieve): New implementation by
+	  robertjlooby <robertjlooby AT gmail.com>.
+
+	* test/test_prime.rb: updated with new method name
+
+Mon Jul 15 11:32:46 2013  Zachary Scott  <e@zzak.io>
+
+	* numeric.c (rb_cNumeric): [DOC] Added comment for Numeric to fix doc
+
+Mon Jul 15 11:24:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (maxpow_in_bdigit_dbl): Useless #if removed.
+
+Mon Jul 15 11:10:46 2013  Zachary Scott  <e@zzak.io>
+
+	* bignum.c (rb_big_coerce): [DOC] Add docs for Bignum#coerce
+	  Based on patch by Juanito Fatas [Fixes GH-360]
+	  https://github.com/ruby/ruby/pull/360
+
+Mon Jul 15 10:56:01 2013  Zachary Scott  <e@zzak.io>
+
+	* thread.c (mutex_sleep): [DOC] Awake thread will reacquire lock
+	  By Tim Abdulla [Fixes GH-342] https://github.com/ruby/ruby/pull/342
+
+Mon Jul 15 10:45:09 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (nlz16): Use __builtin_clz if possible.
+	  (nlz32): Use __builtin_clz or __builtin_clzl if possible.
+	  (nlz64): Use __builtin_clzl or __builtin_clzll if possible.
+	  (nlz128): Use __builtin_clzll if possible.
+
+	* configure.in: Check __builtin_clz, __builtin_clzl and
+	  __builtin_clzll.
+
+Mon Jul 15 09:39:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (power_cache_get_power): Use bitsize instead of ceil_log2.
+	  (ones): Removed.
+	  (next_pow2): Removed.
+	  (floor_log2): Removed.
+	  (ceil_log2): Removed.
+
+	* configure.in (__builtin_popcountl): Don't check.
+
+Mon Jul 15 02:47:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* localeinit.c (rb_locale_charmap, Init_enc_set_filesystem_encoding):
+	  move from encoding.c.
+
+	* miniinit.c (rb_locale_charmap, Init_enc_set_filesystem_encoding):
+	  define miniruby specific functions only.
+
+Mon Jul 15 02:32:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_init): no longer needs NO_PRESERVED_ENCODING.
+
+	* encoding.c (enc_inspect): defer loading autoloaded encoding.
+
+	* encoding.c (enc_check_encoding): use is_data_encoding() to check
+	  type consistently.
+
+	* encoding.c (must_encoding): return rb_encoding* instead of encoding
+	  index.
+
+	* encoding.c (enc_check_encoding): use is_data_encoding() to check
+	  type consistently.
+
+	* encoding.c (must_encoding): return rb_encoding* instead of encoding
+	  index.
+
+Mon Jul 15 02:21:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_fill_term): consider old terminator length, and should
+	  not use rb_enc_ascget since it depends on the current encoding which
+	  may not be compatible with the new terminator.  [Bug #8634]
+
+	* encoding.c (enc_inspect): use PRIsVALUE to preserve the result
+	  encoding.
+
+Sun Jul 14 23:21:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Check __builtin_popcountl, __builtin_bswap32 and
+	  __builtin_bswap64.
+
+	* internal.h (swap32): Use the configure result for the condition to
+	  use __builtin_bswap32.
+	  (swap64): Use the configure result for the condition to use
+	  __builtin_bswap64.
+
+	* bignum.c (ones): Use the configure result for the condition to use
+	  __builtin_popcountl.
+	  (bary_unpack_internal): Use appropriate types for swap argument.
+
+Sun Jul 14 22:21:11 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_subb): Support xn < yn.
+	  (bigsub_core): Removed.
+	  (bigsub): Don't compare before subtraction.  Just subtract and
+	  get the two's complement if the subtraction causes a borrow.
+
+Sun Jul 14 00:36:03 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (DIGSPERLONG): Unused macro removed.
+	  (DIGSPERLL): Ditto.
+
+Sun Jul 14 00:32:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_aref): Less scan when the number is negative.
+
+Sun Jul 14 00:17:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_shift): Avoid signed integer overflow.
+
+Sun Jul 14 00:14:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_precheck): Use bary_small_lshift or
+	  bary_mul_normal if xl is 1.
+
+Sat Jul 13 22:58:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_shift3): New function.
+	  big_lshift and big_rshift are merged.
+	  (big_shift2): New function.
+	  (big_lshift): Use big_shift3.
+	  (big_rshift): Ditto.
+	  (check_shiftdown): Removed.
+	  (rb_big_lshift): Use big_shift2 and big_shift3.
+	  (rb_big_rshift): Ditto.
+	  (big_lshift): Removed.
+	  (big_rshift): Ditto.
+
+Sat Jul 13 15:51:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_small_lshift): Use size_t instead of long.
+	  (bary_small_rshift): Ditto.
+
+Sat Jul 13 15:33:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_small_lshift): Functions moved to remove
+	  declaration.
+	  (bary_small_rshift): Ditto.
+
+Sat Jul 13 12:27:34 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_associate_index): fill new terminator length, not
+	  old one.
+
+Sat Jul 13 12:24:24 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32: move from ext/dl and ext/fiddle.  since ext/extmk.rb
+	  builds extensions in alphabetical order, compiled?('fiddle') under
+	  ext/dl makes no sense.
+
+Sat Jul 13 09:26:09 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (biglsh_bang): Removed.
+	  (bigrsh_bang): Ditto.
+	  (bigmul1_toom3): Use bary_small_lshift and bary_small_rshift.
+
+Sat Jul 13 01:04:43 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/rubygems/psych_additions.rb: Ignore Psych docs here
+
+Fri Jul 12 18:10:46 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/fiddle/win32/lib/win32/registry.rb
+	  (Win32::Registry::API#make_wstr): same as r41922.
+
+Fri Jul 12 16:28:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_associate_index): refill the terminator if it
+	  becomes longer than before.  [ruby-dev:47500] [Bug #8624]
+
+	* string.c (str_null_char, str_fill_term): get rid of out of bound
+	  access.
+
+	* string.c (rb_str_fill_terminator): add a parameter for the length of
+	  new terminator.
+
+Fri Jul 12 11:26:25 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_reject_bang): do not call rb_hash_foreach() if RHash
+	  has ntbl and it is empty.
+
+Fri Jul 12 11:17:41 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (recursive_hash): use RHASH_SIZE() to check hash size.
+
+Fri Jul 12 00:20:00 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_size): use RHASH_SIZE().
+
+Fri Jul 12 00:08:24 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_values): set array capa to RHASH_SIZE().
+
+Thu Jul 11 23:54:45 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_keys): set array capa to RHASH_SIZE().
+
+Thu Jul 11 21:30:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pow): undef pow to get rid of infinite
+	  recursive call.  re-fix [Bug #8495].  [ruby-core:55923] [Bug #8621]
+
+Thu Jul 11 20:18:13 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/registry.rb (Win32::Registry::API#make_wstr):
+	  remove workaround to append WCHAR terminator.
+
+	* transcode.c (str_encode_associate): fill terminator after conversion.
+
+	* string.c (rb_enc_str_new, rb_str_set_len, rb_str_resize): fill
+	  minimum length of the encoding as the terminator.
+
+	* string.c (str_buf_cat, rb_str_buf_append, rb_str_splice_0): ditto.
+
+	* string.c (str_make_independent_expand, rb_str_modify_expand): make
+	  the capacity enough for multi-byte terminator.
+
+	* string.c (rb_string_value_cstr): fill minimum length of the encoding
+	  as the terminator.
+
+	* string.c (rb_string_value_cstr): check null char in char, not in
+	  byte.
+
+Thu Jul 11 14:48:35 2013  Zachary Scott  <e@zzak.io>
+
+	* array.c: Replace confusing example for #reverse_each in overview
+	  Patch by Earl St Sauver [Fixes documenting-ruby/ruby-12]
+	  https://github.com/documenting-ruby/ruby/pull/12
+
+Thu Jul 11 14:22:37 2013  Zachary Scott  <e@zzak.io>
+
+	* test/drb/ut_eq.rb: Use localhost for drb tests [Bug #7311]
+	  Patch by Vit Ondruch [ruby-core:49101]
+	* test/drb/ut_array.rb: ditto
+	* test/drb/ut_array_drbssl.rb: ditto
+
+Thu Jul 11 13:48:03 2013  Zachary Scott  <e@zzak.io>
+
+	* sprintf.c: Fix typo patch by @hynkle [Fixes GH-357]
+	  https://github.com/ruby/ruby/pull/357
+
+Thu Jul 11 13:00:34 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/securerandom.rb: Refactor conditions by Rafal Chmiel
+	  [Fixes GH-326] https://github.com/ruby/ruby/pull/326
+
+Thu Jul 11 12:04:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Don't use toom3 after once karatsuba is chosen.
+	  (mulfunc_t): New type.
+	  (bary_mul_toom3_start): Renamed from bary_mul.
+	  (bary_mul_karatsuba_start): Renamed from bary_mul.
+	  (bary_mul_balance_with_mulfunc): Renamed from bary_mul_balance and
+	  new argument, mulfunc, is added.
+	  (rb_big_mul_balance): Invoke bary_mul_balance_with_mulfunc with
+	  bary_mul_toom3_start.
+	  (bary_mul_karatsuba): Invoke bary_mul_karatsuba_start instead of
+	  bary_mul.
+	  (bary_mul_precheck): Extracted from bary_mul.
+	  (bary_mul_karatsuba_branch): Extracted from bary_mul.
+	  (bary_mul_karatsuba_start): New function to call bary_mul_precheck
+	  and bary_mul_karatsuba_branch.
+	  (bary_mul_toom3_branch): Extracted from bary_mul.
+	  (bary_mul_toom3_start): New function to call bary_mul_precheck and
+	  bary_mul_toom3_branch.
+	  (bary_mul): Just call bary_mul_toom3_start.
+	  Arguments for work memory are removed.
+	  (rb_cstr_to_inum): Follow the bary_mul change.
+	  (bigmul0): Ditto.
+
+Thu Jul 11 10:46:38 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/probes_to_wiki.rb: fix usage comment.  use Enumerable#grep
+	  which yields each elements to reduce unnecessary array.
+
+Thu Jul 11 10:09:18 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c (rb_daemon): daemon(3) is implemented with fork(2).
+	  Therefore it needs rb_thread_atfork(). (and revert r41903)
+
+Thu Jul 11 03:22:10 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* tool/probes_to_wiki.rb: adding a script to convert probes.d to wiki
+	  format for easy wiki updates.
+
+Thu Jul 11 00:54:07 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* man/ri.1: Incorrect use of .Dd macro [Bug #8620] by Tristan Hill
+
+Thu Jul 11 00:48:29 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/delegate.rb: Add example for __setobj__ and __getobj__
+	  [Bug #8615] Patch by Caleb Thompson
+
+Wed Jul 10 23:29:22 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/logger.rb: Use :call-seq: for method signature rdoc
+
+Wed Jul 10 23:23:18 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/logger.rb (#add): Remove incorrect rdoc for return value
+	  [Bug #8567] Reported by Tim Pease.
+
+Wed Jul 10 23:12:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_subpos): make public function.
+
+Wed Jul 10 22:44:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Add a static assertion for RBIGNUM_EMBED_LEN_MAX.
+
+Wed Jul 10 22:31:25 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* string.c (rb_str_index): cache single byte flag and some
+	  cosmetic changes.
+
+Wed Jul 10 22:03:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_2comp): Don't use bary_plus_one.
+	  (bary_add_one): Replaced by the implementation of bary_plus_one.
+
+Wed Jul 10 20:48:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (sizeof_bdigit_dbl): check sizeof(BDIGIT_DBL).
+
+	* internal.h (STATIC_ASSERT): move from enum.c.
+
+Wed Jul 10 20:08:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (SIZEOF_BDIGIT_DBL): Add a ifdef guard for test.
+
+Wed Jul 10 14:18:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (fork_daemon): kill the other threads all and abandon the
+	  kept mutexes.
+
+Wed Jul 10 11:35:36 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/net/http/test_http.rb (TestNetHTTP_v1_2#test_get,
+	  TestNetHTTP_v1_2_chunked#test_get): shouldn't check
+	  HttpResponse#decode_content if Zlib is not available.
+	  ko1 complained via IRC.
+
+Wed Jul 10 10:20:07 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/rbinstall.rb: always require rubygems to stabilize rubygems
+	  related status like whether Gem::Specification is defined or not.
+
+	* tool/rbinstall.rb (Gem::Specification.unresolved_deps): define stub.
+
+Wed Jul 10 08:21:15 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Import RubyGems 2.1
+	* test/rubygems:  Ditto.
+
+Wed Jul 10 07:34:34 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/ext/ext_conf_builder.rb:  Remove siteconf file after
+	  building the gem.
+	* test/rubygems/test_gem_ext_ext_conf_builder.rb:  Test for the above.
+
+	* lib/rubygems/psych_tree.rb (module Gem):  Add backward compatibility
+	  for r41148
+
+	* test/rubygems/test_gem_package.rb:  Add backward compatibility for
+	  double-slash elimination.
+
+Wed Jul 10 06:22:27 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_parse.c (date_zone_to_diff): [ruby-core:55831].
+
+Wed Jul 10 00:41:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul): x*1 is x.
+
+Tue Jul  9 22:24:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul1): No need to invoke MEMZERO at last.
+	  (bary_mul_single): Invoke MEMZERO here.
+
+Tue Jul  9 21:40:01 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_text.rb: Add missing tests for Text#<<.
+	  Reported by nagachika. Thanks!!!
+
+Tue Jul  9 18:02:38 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/fileutils.rb (FileUtils#chown_R): Do not skip traversal even
+	  if user and group are both nil, to be consistent with #chown and
+	  other commands.
+
+Tue Jul  9 17:58:26 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/fileutils/test_fileutils.rb
+	  (TestFileUtils#assert_output_lines): New utility assertion
+	  method for testing verbose output.
+
+Tue Jul  9 17:43:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_tracer.rb: catch up recent rubygems changes.
+
+Tue Jul  9 16:58:30 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/{dl,fiddle}/win32/lib/win32/registry.rb: hope that the final
+	  resolution to fix the failure of test-all.  and includes Win64
+	  support (fixed a potential bug).
+
+Tue Jul  9 15:57:20 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* object.c: Fix rdoc for Kernel#<=>. [Fixes GH-352]
+
+Tue Jul  9 15:53:51 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/fileutils.rb (FileUtils#mode_to_s): Define mode_to_s() also
+	  as singleton method, or FileUtils.chmod fails in verbose mode.
+
+Tue Jul  9 15:16:02 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/fileutils/fileasserts.rb
+	  (Test::Unit::FileAssertions#assert_not_symlink): Add a missing
+	  optional argument "message".
+
+Tue Jul  9 15:03:24 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/fileutils.rb (FileUtils#chown, FileUtils#chown_R): If user
+	  and group are both nil, print ":".
+
+Tue Jul  9 12:47:08 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* io.c (appendline): use READ_CHAR_PENDING_XXX macros and
+	  RSTRING_END().
+
+	* io.c (rb_io_getline_1): rewrite nested if statement into one
+	  statement.
+
+Tue Jul  9 11:04:35 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/{dl,fiddle}/win32/lib/win32/registry.rb (Win32::Registry#check):
+	  should report the position of the error.
+
+	* ext/{dl,fiddle}/win32/lib/win32/registry.rb
+	  (Win32::Registry#QueryValue): workaround for test-all crash.
+
+Tue Jul  9 10:27:56 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/{dl,fiddle}/win32/lib/win32/registry.rb
+	  (Win32::Registry.expand_environ): use suitable encoding for the
+	  string.
+
+	* ext/{dl,fiddle}/win32/lib/win32/registry.rb (Win32::Registry#read):
+	  should return REG_SZ, REG_EXPAND_SZ and REG_MULTI_SZ values with
+	  the expected encoding -- assumed as the same encoding of name.
+
+Tue Jul  9 10:02:45 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/{dl,fiddle}/win32/lib/win32/registry.rb
+	  (Win32::Registry::Error#initialize): use suitable encoding for the
+	  string.
+
+Tue Jul  9 09:46:53 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/registry.rb (Win32::Registry.expand_environ):
+	  use suitable encoding for the string.  fixed a test-all error of
+	  r41838.
+
+	* ext/fiddle/win32/lib/win32/registry.rb: same changes of r41838 and
+	  this revision of dl's win32/registry.rb.
+
+Tue Jul  9 07:39:45 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.0.4.  See
+	  https://github.com/rubygems/rubygems/blob/2.0/History.txt for changes
+
+Tue Jul  9 01:47:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (biglsh_bang): Don't shift a BDIGIT with BITSPERDIG bits.
+	  (bigrsh_bang): Ditto.
+
+Tue Jul  9 01:17:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigrsh_bang): Fix bignum digits overrun.
+
+Tue Jul  9 00:46:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (biglsh_bang): Fix bignum digits under-run.
+
+Mon Jul  8 23:36:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/win32/lib/win32/registry.rb (Error, API): use WCHAR
+	  interfaces.  c.f. [Bug #8508]
+
+Mon Jul  8 23:13:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_pow): move from win32.h and disable strict
+	  ANSI mode macro to let _controlfp() stuff defined.
+	  [ruby-core:55312] [Bug #8495]
+
+	* numeric.c (finite): add declaration for strict ANSI.
+	  [ruby-core:55312] [Bug #8495]
+
+	* thread_win32.c (w32_thread_start_func, thread_start_func_1),
+	  (timer_thread_func): use __stdcall instead of _stdcall which is
+	  unavailable in strict ANSI mode.  [ruby-core:55312] [Bug #8495]
+
+	* win32/win32.c (gettimeofday): use __cdecl instead of _cdecl.
+
+Mon Jul  8 22:41:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul): Arguments for work memory added.
+	  (bary_mul_balance): Ditto.
+	  (bary_mul_karatsuba): Ditto.
+
+Mon Jul  8 22:03:30 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_sq_fast): New function for testing.
+	  (rb_big_mul_toom3): Ditto.
+
+	* internal.h (rb_big_sq_fast): Declared.
+	  (rb_big_mul_toom3): Ditto.
+
+Mon Jul  8 21:59:34 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_balance): Initialize a local variable to suppress
+	  a warning.
+
+Mon Jul  8 20:55:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_balance): Reduce work memory.
+
+Mon Jul  8 08:26:15 2013  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* test/openssl/test_pkey_ec.rb: Skip tests for "Oakley" curves as
+	  they are not suitable for ECDSA.
+	  [ruby-core:54881] [Bug #8384]
+
+Mon Jul  8 08:03:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul): Add a RB_GC_GUARD.
+
+Sun Jul  7 23:56:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_karatsuba): Unreachable code removed.  Remove
+	  several branches.
+
+Sun Jul  7 22:59:06 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_big_mul_normal): Declared.
+	  (rb_big_mul_balance): Ditto.
+	  (rb_big_mul_karatsuba): Ditto.
+
+	* bignum.c (rb_big_mul_normal): New function for tests.
+	  (rb_big_mul_balance): Ditto.
+	  (rb_big_mul_karatsuba): Ditto.
+
+Sun Jul  7 19:21:30 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Reorder functions to decrease forward reference.
+
+Sun Jul  7 14:41:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: (bigsub_core): Use bary_sub.
+	  (bary_sub): Returns a borrow flag.  Use bary_subb.
+	  (bary_subb): New function for actually calculating subtraction with
+	  borrow.
+	  (bary_sub_one): New function.
+	  (bigadd_core): Use bary_add.
+	  (bary_add): Returns a carry flag.  Use bary_addc.
+	  (bary_addc): New function for actually calculating addition with
+	  carry.
+	  (bary_add_one): New function.
+	  (bary_muladd_1xN): Extracted from bary_mul_normal.
+	  (bigmul1_normal): Removed.
+	  (bary_mul_karatsuba): New function.
+	  (bary_mul1): Invoke rb_thread_check_ints after bary_mul_normal.
+	  (bary_mul): Remove most and least significant zeros before actual
+	  multiplication.  Use bary_sq_fast, bary_mul_balance,
+	  bary_mul_karatsuba and bigmul1_toom3 as bigmul0.
+	  (bigmul1_balance): Removed.
+	  (bigmul1_karatsuba): Removed.
+	  (bigsqr_fast): Removed.
+	  (bary_sparse_p): Extracted from big_sparse_p.
+	  (big_sparse_p): Removed.
+	  (bigmul0): Use bary_mul.
+
+Sun Jul  7 11:54:33 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS: Add REXML::Text#<< related updates.
+
+Sun Jul  7 11:49:19 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/text.rb (REXML::Text#<<): Support appending in not
+	  "raw" mode. [Bug #8602] [ruby-dev:47482]
+	  Reported by Ippei Obayashi. Thanks!!!
+
+Sun Jul  7 11:43:13 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/text.rb (REXML::Text#<<): Support method chain use by "<<"
+	  like other objects.
+
+Sun Jul  7 11:34:18 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/text.rb (REXML::Text#clear_cache): Extract common
+	  cache clear code.
+
+Sun Jul  7 11:01:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_DTRACE_POSTPROCESS): dtrace version SUN D 1.11
+	  introduces a check in the dtrace compiler to ensure that probes
+	  actually exist. If there are no probes, then the -G step will
+	  fail. As this test is only being used to determine whether -G is
+	  necessary (for instance, on OSX it is not), adding a real probe to
+	  the conftest allows it to succeed on newer versions of dtrace.
+	  Patch by Eric Saxby <sax AT livinginthepast.org> at
+	  [ruby-core:55826].  [Fixes GH-351], [Bug #8606].
+
+Sun Jul  7 10:07:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_sq_fast): Extracted from bigsqr_fast and
+	  ensure not to access zds[2*xn].
+	  (bigsqr_fast): Allocate the result bignum with 2*xn words.
+
+Sat Jul  6 07:37:43 2013  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/ossl_pkey_ec.c: Ensure compatibility to builds of
+	  OpenSSL with OPENSSL_NO_EC2M defined, but OPENSSL_NO_EC not
+	  defined.
+	* test/openssl/test_pkey_ec.rb: Iterate over built-in curves
+	  (and assert their non-emptiness!) instead of hard-coding them, as
+	  this may cause problems with respect to the different availability
+	  of individual curves in individual OpenSSL builds.
+	  [ruby-core:54881] [Bug #8384]
+
+	  Thanks to Vit Ondruch for providing the patch!
+
+Sat Jul  6 07:12:39 2013  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* test/openssl/test_x509crl.rb: Remove unused variable.
+	  [ruby-core:53501] [Bug #8114]
+
+	  Thanks, Vipul Amler, for pointing this out!
+
+Sat Jul  6 06:37:10 2013  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/ossl.c: Provide CRYPTO_set_locking_callback() and
+	  CRYPTO_set_id_callback() callback functions ossl_thread_id and
+	  ossl_lock_callback to ensure the OpenSSL extension is usable in
+	  multi-threaded environments.
+	  [ruby-core:54900] [Bug #8386]
+
+	  Thanks, Dirkjan Bussink, for the patch!
+
+Sat Jul  6 06:06:16 2013  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* lib/openssl/ssl.rb: Fix SSL client connection crash for SAN marked
+	  critical.
+	  The patch for CVE-2013-4073 caused SSL crash when a SSL server returns
+	  the certificate that has critical SAN value.  X509 extension could
+	  include 2 or 3 elements in it:
+
+	  [id, criticality, octet_string] if critical,
+	  [id, octet_string] if not.
+
+	  Making sure to pick the last element of X509 extension and use it as
+	  SAN value.
+	  [ruby-core:55685] [Bug #8575]
+
+	  Thank you @nahi for providing the patch!
+
+Sat Jul  6 04:49:38 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: register time objects so
+	  they are referenced as ids during output.
+	* test/psych/test_date_time.rb: corresponding test.
+
+Fri Jul  5 20:46:39 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_unicode_escape.rb (TestUnicodeEscape#test_basic): this
+	  assertion doesn't seems to be checking the unicode string on command
+	  line, but seems to be checking how to treat the unicode string from
+	  stdin.  so, should escape '\' before 'u'.  this fixes a test failure
+	  on Windows.
+
+Fri Jul  5 19:05:40 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/fileutils.rb (FileUtils#chown, FileUtils#chown_R): Fix the
+	  wrong output message when user is nil, which should be "chown
+	  :group file" instead of "chown group file".
+
+Fri Jul  5 16:21:56 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/ruby/test_regexp.rb
+	  (TestRegexp#test_options_in_look_behind)
+	  (TestRegexp#assert_match_at): Add tests for another problem
+	  fixed in Onigmo 5.13.5.  Previously Onigmo did not allow option
+	  enclosures in look-behind, which makes it impossible to
+	  interpolate a regexp into another in the middle of a look-behind
+	  pattern.  cf. https://github.com/k-takata/Onigmo/pull/17
+
+	* test/ruby/test_regexp.rb
+	  (TestRegexp#test_options_in_look_behind)
+	  (TestRegexp#assert_match_at): Parse regexps in run time rather
+	  than in compile time.
+
+Fri Jul  5 12:14:40 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_notfound): after
+	  r41710, the path of command uses backslash as the separator on
+	  Windows.
+
+Fri Jul  5 11:29:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (assert_raise_with_message): move from
+	  test/fileutils/test_fileutils.rb.  this is still experimental and
+	  the interface may be changed.
+
+Fri Jul  5 11:08:00 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (w32_spawn): r41710 made that if the command starts with
+	  a quote and includes slash, removed the top quote and NOT removed the
+	  last quote.
+	  this fixes test failures on test/ruby/test_process.rb and
+	  test/webrick.
+
+Fri Jul  5 09:53:15 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+	* lib/mkmf.rb (CONFIG['CPPOUTFILE']): fix r41769; CONFIG['CPPOUTFILE']
+	  may be nil.
+
+Fri Jul  5 05:39:53 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BARY_MUL1): Renamed from BARY_MUL.
+	  (bary_mul1): Renamed from bary_mul.
+	  (bary_mul): Renamed from bary_mul2.
+
+Fri Jul  5 04:58:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_balance): Extracted from bigmul1_balance and
+	  use bary_mul2 and bary_add to decrease allocations.
+
+Fri Jul  5 02:14:00 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/fileutils.rb (FileUtils#symbolic_modes_to_i): Fix the wrong
+	  character class [+-=], which happened to match all desired
+	  characters but also match undesired characters.
+
+	* lib/fileutils.rb (FileUtils.chmod{,_R}): Enhance the symbolic
+	  mode parser to support the permission symbols u/g/o and multiple
+	  actions as defined in SUS, so that chmod("g=o+w", file) works as
+	  expected.  Invalid symbolic modes are now rejected with
+	  ArgumentError.
+
+Fri Jul  5 00:25:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (have_framework): allow header file to check.
+	  [ruby-core:55745] [Bug #8593]
+
+Thu Jul  4 22:31:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* object.c (rb_obj_equal): Fixed an rb_obj_equal documentation typo
+	  where "a" was used instead of "obj".
+	  Fixes GH-349. Patch by @adnandoric
+
+Thu Jul  4 20:39:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* tool/make-snapshot: Exit with EXIT_FAILURE when it fails.
+
+Thu Jul  4 20:20:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (maxpow_in_bdigit_dbl): Use tables if available.
+	  (maxpow_in_bdigit): Ditto.
+	  (U16): New macro.
+	  (U32): Ditto.
+	  (U64): Ditto.
+	  (U128): Ditto.
+	  (maxpow16_exp): New table.
+	  (maxpow16_num): New table.
+	  (maxpow32_exp): New table.
+	  (maxpow32_num): New table.
+	  (maxpow64_exp): New table.
+	  (maxpow64_num): New table.
+	  (maxpow128_exp): New table.
+	  (maxpow128_num): New table.
+
+Thu Jul  4 18:25:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_cstr_to_inum): Avoid temporary buffer allocation except
+	  very big base non-power-of-2 numbers.
+
+Thu Jul  4 15:51:56 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_succ): use ONIGENC_MBCLEN_CHARFOUND_P correctly.
+
+	* string.c (rb_str_dump): ditto.
+
+Thu Jul  4 10:04:11 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (): Merge Onigmo 5.13.5 23b523076d6f1161.
+
+	  * [bug]  (thanks Akinori MUSHA and Ippei Obayashi)
+	           Fix a renumbering bug in condition regexp with a named
+	           capture. [Bug #8583]
+	  * [spec] (thanks Akinori MUSHA)
+	    Allow ENCLOSE_OPTION in look-behind.
+
+Thu Jul  4 00:36:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (SIGNED_INTEGER_MAX): suppress warning C4146 on VC6.
+	  seems a logical ORed expression becomes unsigned.
+
+Thu Jul  4 00:13:01 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby_atomic.h (rb_w32_atomic_cas): call InterlockedCompareExchange
+	  directly.
+
+	* ruby_atomic.h (ATOMIC_CAS): fix missing function call.
+
+Wed Jul  3 23:47:35 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby_atomic.h (ATOMIC_CAS): suppress C4022 and C4047 warnings in
+	  VC6.  only InterlockedCompareExchange is declared using PVOID.
+
+Wed Jul  3 22:29:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (ruby_digit36_to_number_table): Declared.
+
+	* util.c (ruby_digit36_to_number_table): Moved from scan_digits.
+
+	* bignum.c (conv_digit): Use ruby_digit36_to_number_table.
+
+	* pack.c (hex2num): Ditto.
+
+Wed Jul  3 18:12:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (install_dirs): revert DESTDIR prefix by r39841, since
+	  it is fixed by r41648.  [ruby-core:55760] [Bug #8115]
+
+Wed Jul  3 14:15:25 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (do_stat): use rb_w32_ustati64() in win32.c to get rid of
+	  mysterious behavior of FindFirstFile() Windows API which treat "<"
+	  and ">" like as wildcard characters.  [ruby-core:55764] [Bug #8597]
+
+Wed Jul  3 12:06:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (maxpow_in_bdigit): Renamed from calc_hbase and return
+	  maxpow.
+
+Tue Jul  2 23:47:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (roomof): Cast to long.
+	  (rb_ull2big): Fix bignew arguments.
+
+Tue Jul  2 21:17:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_cstr_to_inum): Merge two temporary buffers.
+
+Tue Jul  2 20:25:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_cstr_to_inum): Use BDIGIT_DBL to collect adjacent digits.
+	  (BDIGIT_DBL_MAX): New macro.
+	  (maxpow_in_bdigit_dbl): New function.
+
+Tue Jul  2 17:23:33 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* doc/syntax/refinements.rdoc: add description of Module#using and
+	  refinement inheritance by module inclusion.
+
+Tue Jul  2 17:22:44 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h: add EUC-JP and Windows-31J.
+
+	* re.c (rb_char_to_option_kcode): use built-in encoding indexes in
+	  internal.h.
+
+	* internal.h: add UTF8-MAC.
+
+	* dir.c (rb_utf8mac_encoding): use built-in encoding indexes in
+	  internal.h.
+
+	* internal.h: add UTF-{16,32} dummy encodings.
+
+	* string.c (rb_str_inspect, str_scrub0): use built-in encoding indexes
+	  in internal.h.
+
+	* internal.h: add UTF-{16,32}{BE,LE}.
+
+	* io.c (io_strip_bom): use built-in encoding indexes in internal.h.
+
+	* internal.h (rb_{ascii8bit,utf8,usascii}_encindex): use built-in
+	  encoding indexes for optimization.
+
+	* encoding.c (enc_inspect, rb_locale_encindex),
+	  (enc_set_filesystem_encoding, rb_filesystem_encindex): use built-in
+	  encoding indexes directly.
+
+	* encoding.c (rb_enc_set_index, rb_enc_associate_index): validate
+	  argument encoding index.
+
+	* include/ruby/encoding.h (ENCODING_SET): use rb_enc_set_index()
+	  instead of setting inlined bits directly.
+
+	* encoding.c (rb_enc_init): register preserved indexes.
+
+	* internal.h (ruby_preserved_encindex): move from encoding.c.
+
+Tue Jul  2 11:14:36 2013  Shota Fukumori  <sorah@cookpad.com>
+
+	* lib/mkmf.rb (try_config): Fix to not replace $LDFLAGS with $libs
+	  (1.9.3 behavior) [ruby-core:55752] [Bug #8595]
+
+Tue Jul  2 00:39:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): Don't try mismatched
+	  address family if already failed.
+
+Mon Jul  1 23:07:38 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* template/encdb.h.tmpl: define encoding index macros to use the index
+	  statically from C source.
+
+Mon Jul  1 22:57:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul2): New function.
+	  (rb_cstr_to_inum): Use a better algorithm to compose the result
+	  if input length is very long.
+
+Mon Jul  1 20:22:00 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* ext/bigdecimal/bigdecimal.h (RB_UNUSED_VAR, UNREACHABLE):
+	  import macros from ruby.h for 1.9.3.
+	  [Bug #8588] [ruby-core:55730]
+
+	* ext/bigdecimal/bigdecimal.gemspec: Bump version to 1.2.1.
+
+Mon Jul  1 20:03:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): Use an address
+	  family for local address which is different to the remote
+	  address if no other choice.
+
+Mon Jul  1 15:05:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/csv.rb (CSV#<<): use StringIO#set_encoding instead of creating
+	  new StringIO instance with String#force_encoding, forcing encoding
+	  discards the cached coderange bits and can make further operations
+	  very slow.  [ruby-core:55714] [Bug #8585]
+
+	* ext/stringio/stringio.c (strio_write): keep coderange of
+	  ptr->string.
+
+	* string.c (rb_enc_cr_str_buf_cat, rb_str_append): consider an empty
+	  string 7bit-clean and should not discard cached coderange of string
+	  to be appended.
+
+Mon Jul  1 12:56:41 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_using_module): activate refinements in the ancestors of
+	  the argument module to support refinement inheritance by
+	  Module#include.  [ruby-core:55671] [Feature #8571]
+
+	* test/ruby/test_refinement.rb: related test.
+
+Mon Jul  1 12:02:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_cstr_to_inum): Skip leading zeros.
+
+Mon Jul  1 00:59:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (nlz16): New function.
+	  (nlz32): Ditto.
+	  (nlz64): Ditto.
+	  (nlz128): Ditto.
+	  (nlz): Redefined using an above function.
+	  (bitsize): New macro.
+	  (rb_cstr_to_inum): Use bitsize instead of nlz.
+
+Sun Jun 30 22:40:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* lib/prime.rb: Corrected a few comments. Patch by @Nullset14.
+	  Fixes GH-346.
+
+Sun Jun 30 21:53:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_cstr_to_inum): Use rb_integer_unpack if base is a power
+	  of 2.
+
+Sun Jun 30 10:59:23 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (join_argv): use backslash instead of slash in program
+	  path, otherwise cannot invoke "./c\u{1ee7}a.exe" for some reason.
+	  [ruby-core:24309] [Bug #1771]
+
+	* io.c (spawnv, spawn): use UTF-8 spawn family.  [Bug #1771]
+
+	* process.c (proc_exec_sh, proc_spawn_cmd, proc_spawn_sh): ditto.
+
+	* win32/win32.c (translate_char, join_argv, has_redirection): make
+	  codepage aware.
+
+	* win32/win32.c (rb_w32_udln_find_exe_r, rb_w32_udln_find_file_r):
+	  codepage independent versions.
+
+	* win32/win32.c (w32_spawn): extract codepage aware code from
+	  rb_w32_spawn().
+
+	* win32/win32.c (rb_w32_uspawn): add UTF-8 version function.
+
+	* win32/win32.c (w32_aspawn_flags): extract codepage aware code from
+	  rb_w32_aspawn_flags().
+
+	* win32/win32.c (rb_w32_uaspawn_flags, rb_w32_uaspawn_flags): add
+	  UTF-8 version functions.
+
+	* win32/win32.c (w32_getenv): extract codepage aware code from
+	  rb_w32_ugetenv() and rb_w32_getenv().
+
+	* win32/win32.c (w32_stati64): extract codepage aware code from
+	  rb_w32_ustati64() and rb_w32_stati64().
+
+	* dln.h (DLN_FIND_EXTRA_ARG, DLN_FIND_EXTRA_ARG_DECL): allow extra
+	  arguments to dln_find_{exe,file}_r().
+
+	* dln_find.c (dln_find_exe_r, dln_find_file_r): add extract arguments.
+
+	* process.c (EXPORT_STR, EXPORT_DUP): convert to default process
+	  encoding if defined.
+
+	* process.c (check_exec_env_i): convert environment variables too.
+
+	* process.c (rb_exec_fillarg): convert program path and arguments too.
+
+Sun Jun 30 01:57:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_rshift): Use abs2twocomp and twocomp2abs_bang.
+
+Sun Jun 30 00:14:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (RBIGNUM_SET_NEGATIVE_SIGN): New macro.
+	  (RBIGNUM_SET_POSITIVE_SIGN): Ditto.
+	  (rb_big_neg): Inline get2comp to avoid double negation.
+
+Sat Jun 29 23:26:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_neg): Extracted from bary_2comp.
+	  (bary_plus_one): Extracted from bary_2comp.
+	  (bary_2comp): Use bary_neg and bary_plus_one.
+	  (big_extend_carry): Extracted from get2comp.
+	  (get2comp): Use big_extend_carry.
+	  (rb_integer_unpack): Use big_extend_carry.
+	  (rb_big_neg): Use bary_neg.
+
+Sat Jun 29 22:31:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_2comp): Simplified.
+
+Sat Jun 29 09:33:53 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigor_int): Return -1 if y == -1.
+
+Sat Jun 29 09:07:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigor_int): Use RB_GC_GUARD.
+	  (bigxor_int): Take xn and hibitsx arguments.  Use twocomp2abs_bang.
+	  (rb_big_xor): Use abs2twocomp and twocomp2abs_bang.
+
+Sat Jun 29 08:19:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigand_int): Don't apply bitwise and for BDIGIT and long.
+	  (bigor_int): Take xn and hibitsx arguments.  Use twocomp2abs_bang.
+	  (rb_big_or): Use abs2twocomp and twocomp2abs_bang.
+
+Fri Jun 29 01:08:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* numeric.c (fix_mul): remove FIT_SQRT_LONG test as it was causing
+	  fix_mul to return an incorrect result for -2147483648*-2147483648
+	  on 64 bit platforms
+
+	* test/ruby/test_integer_comb.rb (class TestIntegerComb): add test case
+
+Fri Jun 28 12:26:53 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_and): Allocate new bignum with same size to shorter
+	  argument if it's high bits are zero.
+
+Fri Jun 28 12:14:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): Don't use local
+	  addresses which address family is different to remote address.
+
+Fri Jun 28 08:06:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigand_int): Add arguments, xn and hibitsx.
+	  Use twocomp2abs_bang.
+
+Thu Jun 27 23:58:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (abs2twocomp_bang): Removed.
+	  (abs2twocomp): Take n_ret argument to return actual length.
+	  (rb_big_and): Follow above change.
+
+Thu Jun 27 22:52:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (get2comp): Use bary_2comp.
+	  (abs2twocomp_bang): New function.
+	  (abs2twocomp): New function.
+	  (twocomp2abs_bang): New function.
+	  (rb_big_and): Use abs2twocomp and twocomp2abs_bang.
+
+Thu Jun 27 20:03:13 2013  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (verify_certificate_identity): fix
+	  hostname verification. Patched by nahi.
+
+	* test/openssl/test_ssl.rb (test_verify_certificate_identity): test for
+	  above.
+
+
+Thu Jun 27 00:23:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_pow): Retry if y is a Bignum and it is
+	  representable as a Fixnum.
+	  Use rb_absint_numwords.
+
+Wed Jun 26 23:53:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_save_rounding_mode): fix typo.
+	  Fixes GH-343. Patch by @jgarber.
+
+Wed Jun 26 23:22:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (rb_enumeratorize_with_size): use strict definition
+	  rb_enumerator_size_func.
+
+Wed Jun 26 23:11:14 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* gc.c (is_before_sweep): Add a missing space before a parenthesis.
+	* gc.c (rb_gc_force_recycle): Add a missing space around a parenthesis.
+
+Wed Jun 26 22:44:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_enumeratorize_with_size): cast for
+	  backward compatibility.
+
+	* include/ruby/intern.h (rb_enumerator_size_func): define strict
+	  function declaration for rb_enumeratorize_with_size().
+
+Wed Jun 26 21:01:22 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_io.rb (TestIO#test_write_32bit_boundary): skip if
+	  writing a file is slow.
+	  [ruby-core:55541] [Bug #8519]
+
+Wed Jun 26 16:42:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: should use expanded values for header directories
+	  unless extmk.  patch by vo.x (Vit Ondruch) at [ruby-core:55653]
+	  [Bug #8115], rhbz#921650.
+
+Wed Jun 26 12:48:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigxor_int): Fix a buffer over read.
+
+Wed Jun 26 12:13:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigand_int): Consider negative values.
+	  (bigor_int): The allocated bignum should have enough size
+	  to store long.
+	  This fixes (bignum fits in a BDIGIT) | (fixnum bigger than BDIGIT)
+	  on platforms which SIZEOF_BDIGITS < SIZEOF_LONG,
+	  such as LP64 with 32bit BDIGIT (no int128).
+
+Wed Jun 26 12:08:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_udp.rb: Close sockets explicitly.
+	  Don't use fixed port number.
+
+Wed Jun 26 07:27:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigand_int): Fix a buffer over read.
+
+Wed Jun 26 06:48:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigadd_int): Fix a buffer over read.
+
+Wed Jun 26 01:18:13 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (is_before_sweep): Add new helper function that check the object
+	  is before sweep?
+	* gc.c (rb_gc_force_recycle): Have to clear mark bit if object's slot
+	  already ready to minor sweep.
+
+Wed Jun 26 01:17:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigsub_int): Fix a buffer over read.
+
+Tue Jun 25 22:45:43 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_absint_singlebit_p): Use POW2_P.
+	  (bary_pack): Ditto.
+	  (rb_big2str0): Ditto.
+	  (POW2_P): Moved to top.
+
+Tue Jun 25 22:28:07 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/rubygems/ext/builder.rb (Gem::Ext::Builder.make): Pass
+	  DESTDIR via command line to override what's in MAKEFLAGS.  This
+	  fixes an installation problem under a package building
+	  environment where DESTDIR is specified in the (parent) command
+	  line. [Fixes GH-327]
+
+Tue Jun 25 21:43:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2dbl): Use (BDIGIT)1 instead of 1UL.
+	  (bary_mul_normal): Remove a useless cast.
+
+Tue Jun 25 21:26:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): Fix for the cases when
+	  the argument x is not a BigDecimal.
+	  This change is based on the patch made by Heesob Park and Garth Snyder.
+	  [Bug #6862] [ruby-core:47145]
+	  [Fixes GH-332] https://github.com/ruby/ruby/pull/332
+
+Tue Jun 25 20:36:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2ulong): "check" argument removed.
+	  (rb_big2ulong): Follow above change.
+	  (rb_big2long): Ditto.
+	  (rb_big_rshift): Ditto.
+	  (rb_big_aref): Ditto.
+
+Tue Jun 25 20:08:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big2ulong_pack): Use rb_integer_pack.
+	  (rb_big_aref): Call big2ulong with TRUE for "check" argument.
+	  It should be non-effective.
+
+Tue Jun 25 19:07:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (LSHIFTX): Revert r41611.
+	  The redundant expression suppresses a warning, C4293, by Visual
+	  Studio.
+	  http://ruby-mswin.cloudapp.net/vc10-x64/ruby-trunk/log/20130625T072854Z.log.html.gz#miniruby
+
+Tue Jun 25 19:03:00 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2ulong): Add a cast.
+	  (big2ull): Add a specialized code for SIZEOF_LONG_LONG <=
+	  SIZEOF_BDIGITS.
+
+Tue Jun 25 12:42:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_unpack_single_bdigit): Use "1 + ~u" instead of
+	  "-u" to suppress warning (C4146) by Visual Studio.
+	  Reported by ko1 via IRC.
+
+Tue Jun 25 12:28:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2ulong): Add code specialized for SIZEOF_LONG <=
+	  SIZEOF_BDIGITS.
+	  This prevents shift width warning from "num <<= BITSPERDIG".
+
+Tue Jun 25 12:23:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix oldgen/remembered_shady counting algorithm.
+
+	* gc.c (rgengc_check_shady): increment
+	  `objspace->rgengc.remembered_shady_object_count' here.
+
+	* gc.c (rgengc_remember): return FALSE if obj is already remembered.
+
+	* gc.c (rgengc_rememberset_mark): make it void.
+
+	* gc.c (gc_mark_children): fix to double counting oldgen_object_count
+	  at minor GC.
+
+Tue Jun 25 12:07:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (MSB): Removed.
+	  (BDIGIT_MSB): Defined using BIGRAD_HALF.
+	  (bary_2comp): Apply BIGLO after possible over flow of BDIGIT.
+	  (get2comp): Ditto.
+	  (bary_unpack_internal): Use BDIGIT_MSB.
+	  Apply BIGLO after possible over flow of BDIGIT.
+	  (rb_integer_unpack): Use BDIGIT_MSB.
+	  (calc_hbase): Use BDIGMAX.
+	  (big2dbl): Use BDIGMAX.
+	  Apply BIGLO after possible over flow of BDIGIT.
+	  (rb_big_neg): Apply BIGLO after possible over flow of BDIGIT.
+	  (biglsh_bang): Ditto.
+	  (bigrsh_bang): Ditto.
+	  (bary_divmod): Use BDIGIT_MSB.
+	  (bigdivrem): Ditto.
+	  (bigxor_int): Apply BIGLO after possible over flow of BDIGIT.
+
+	* marshal.c (shortlen): Use SIZEOF_BDIGITS instead of sizeof(BDIGIT).
+
+	* ext/openssl/ossl_bn.c (ossl_bn_initialize): Use SIZEOF_BDIGITS
+	  instead of sizeof(BDIGIT).
+
+Tue Jun 25 11:40:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (big2ulong): suppress shorten-64-to-32 warning.  BDIGIT can
+	  be bigger than long now.
+
+	* bignum.c (LSHIFTX): remove redundant never-true expression.
+
+Tue Jun 25 00:55:54 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (typedef struct rb_objspace): Change members for monitor objects.
+	* gc.c (gc_marks_test): Check all WriteBarrier Errors and track them in obj-tree.
+	* gc.c (rgengc_check_shady): Ditto.
+	* gc.c (gc_marks): Move 2 function calls to gc_marks_test for test initialize.
+
+Mon Jun 24 23:30:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_unpack_single_bdigit): Refine code to filling
+	  higher bits and use BIGLO.
+
+Mon Jun 24 22:26:31 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/rinda/test_rinda.rb (RingIPv6#prepare_ipv6):
+	  ifindex() function may not be implemented on Windows. We use another
+	  check for the case.
+
+Mon Jun 24 22:11:37 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/gdbm/test_gdbm.rb (TestGDBM#test_s_open_nolock):
+	  skip a failing test on Windows because flock() implementation is
+	  different from Unix.
+
+Mon Jun 24 22:06:14 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/rubygems/test_gem_installer.rb (test_install_extension_flat):
+	  use ruby in build directory in case ruby is not installed.
+	  [ruby-core:53265] [Bug #8058]
+
+Mon Jun 24 22:04:02 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): fix conversion from Bignum to
+	  pointer. sizeof(DLSTACK_TYPE) is larger than sizeof(long) on
+	  Windows x64 and higher bits over sizeof(long) of DLSTACK_TYPE was
+	  zero even if a pointer value was over 32 bits which causes SEGV on
+	  DL::TestCPtr#test_to_ptr_io. Adding a cast solves the bug.
+
+Mon Jun 24 22:04:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* eval_error.c (warn_printf): use rb_vsprintf instead so ruby specific
+	  extensions like PRIsVALUE can be used in format strings
+	* eval_error.c (error_print): use warn_print_str (alias for
+	  rb_write_error_str) to print a string value instead of using
+	  RSTRING_PTR and RSTRING_LEN manually
+	* eval.c (setup_exception): use PRIsVALUE instead of %s and RSTRING_PTR
+
+Mon Jun 24 20:31:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* compile.c (make_name_for_block): use PRIsVALUE in format string
+	  instead of %s and RSTRING_PTR to protect objects from being garbage
+	  collected too soon
+	* encoding.c (str_to_encindex): ditto
+	* hash.c (rb_hash_fetch_m): ditto
+	* io.c (rb_io_reopen): ditto
+	* parse.y (reg_fragment_check_gen): ditto
+	* parse.y (reg_compile_gen): ditto
+	* parse.y (ripper_assert_Qundef): ditto
+	* re.c (rb_reg_raise): ditto
+	* ruby.c (set_option_encoding_once): ditto
+	* vm_eval.c (rb_throw_obj): ditto
+
+Mon Jun 24 07:57:18 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (after_gc_sweep): Have to record malloc info before reset.
+	* gc.c (gc_prof_timer_start): Pick out part of new record creation as gc_prof_setup_new_record.
+	* gc.c (gc_prof_set_malloc_info): Move point of recording allocation size to front of mark.
+
+Mon Jun 24 02:53:09 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c: Return value in Array overview example found by @PragTob
+	  [Fixes GH-336] https://github.com/ruby/ruby/pull/336
+
+Mon Jun 24 02:45:51 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c (rb_ary_zip): typo by @PragTob [Fixes GH-337]
+	  https://github.com/ruby/ruby/pull/337
+
+Mon Jun 24 02:42:01 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* win32/README.win32: grammar typo by @blankenshipz [Fixes GH-334]
+	  https://github.com/ruby/ruby/pull/334
+
+Mon Jun 24 00:59:35 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BIGUP): Use LSHIFTX and avoid cast to consider the type
+	  of x is bigger than BDIGIT_DBL.
+	  (big2ulong): Use unsigned long to store the result.
+	  (big2ull): Use unsigned LONG_LONG to store the result.
+	  (bigand_int): Use long for num to avoid data loss.
+	  (bigor_int): Ditto.
+	  (bigxor_int): Ditto.
+
+Sun Jun 23 23:05:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h (BDIGIT): Define it only if it is not defined
+	  yet.  This eases tests and debug.
+	  (SIZEOF_BDIGITS): Ditto.
+	  (BDIGIT_DBL): Ditto.
+	  (BDIGIT_DBL_SIGNED): Ditto.
+	  (PRI_BDIGIT_PREFIX): Ditto.
+	  (PRI_BDIGIT_DBL_PREFIX): Ditto.
+	  (PRIdBDIGIT): Define it only if PRI_BDIGIT_PREFIX is defined.
+	  (PRIiBDIGIT): Ditto.
+	  (PRIoBDIGIT): Ditto.
+	  (PRIuBDIGIT): Ditto.
+	  (PRIxBDIGIT): Ditto.
+	  (PRIXBDIGIT): Ditto.
+	  (PRIdBDIGIT_DBL): Ditto.
+	  (PRIiBDIGIT_DBL): Ditto.
+	  (PRIoBDIGIT_DBL): Ditto.
+	  (PRIuBDIGIT_DBL): Ditto.
+	  (PRIxBDIGIT_DBL): Ditto.
+	  (PRIXBDIGIT_DBL): Ditto.
+
+	* include/ruby/ruby.h (RBIGNUM_EMBED_LEN_MAX): Define it only if it is
+	  not defined yet.
+
+Sun Jun 23 17:29:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_unpack_single_bdigit): Use a cast.
+
+Sun Jun 23 15:38:07 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_thread.rb: rescue resource limitation errors.
+
+Sun Jun 23 08:19:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_unpack_single_bdigit): Extracted from
+	  bary_unpack_internal.
+
+Sun Jun 23 07:41:52 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_unpack_internal): Suppress warnings (C4146) on Visual Studio.
+	  Reported by ko1 via IRC.
+
+Sun Jun 23 06:49:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, gc.c: rename macros and functions:
+	  OBJ_WB_GIVEUP() -> OBJ_WB_UNPROTECT(),
+	  rb_obj_wb_giveup() -> rb_obj_wb_unprotect(),
+	  rb_gc_giveup_promoted_writebarrier() ->
+	  rb_gc_writebarrier_unprotect_promoted(),
+
+	* class.c, eval.c, hash.c: use OBJ_WB_UNPROTECT().
+
+Sun Jun 23 05:41:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (rb_include_class_new), eval.c (rb_using_refinement):
+	  make classes/modules (who share method table) shady.
+	  If module `a' and `b' shares method table m_tbl and new method
+	  with iseq is added, then write barrier is applied only `a' or `b'.
+	  To avoid this issue, shade such classes/modules.
+
+	* vm_method.c (rb_method_entry_make): add write barriers.
+
+Sun Jun 23 01:27:54 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bytes_zero_p): Removed.
+	  (bary_pack): Don't call bytes_zero_p.
+
+Sun Jun 23 00:51:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bytes_zero_p): Extracted from bary_pack.
+	  (bary_pack): Use bytes_zero_p.
+
+Sun Jun 23 00:16:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (MSB): New macro.
+	  (bary_unpack_internal): Use MSB.
+	  (bary_divmod): Ditto.
+	  (bigdivrem): Ditto.
+
+Sat Jun 22 23:45:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_swap): New function.
+	  (bary_pack): Use bary_swap.
+	  (bary_unpack_internal): Ditto.
+
+Sat Jun 22 23:18:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bytes_2comp): Renamed from quad_buf_complement.
+	  (bary_pack): Use bytes_2comp.
+	  (rb_quad_pack): Use rb_integer_pack.
+	  (rb_quad_unpack): Use rb_integer_unpack.
+
+Sat Jun 22 21:46:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_unpack): Don't allocate a Bignum if possible.
+
+Sat Jun 22 21:03:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_unpack): Remove specialized unpackers for integers.
+
+Sat Jun 22 20:36:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_unpack_internal): Specialized unpacker implemented.
+	  (bary_unpack): Support INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION.
+	  (rb_integer_unpack): Support INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION.
+
+Sat Jun 22 18:53:10 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_pack): Support
+	  INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION flag.
+	  Fix byte order and word order handling in code specialized for
+	  wordsize % SIZEOF_BDIGITS == 0.
+
+	* internal.h (INTEGER_PACK_FORCE_GENERIC_IMPLEMENTATION): Defined.
+
+Sat Jun 22 15:41:25 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rgengc_check_shady): add new WB miss checking
+	  on RGENGC_CHECK_MODE >= 2.
+
+	    (1) Save bitmaps before marking
+	    (2) Run full marking
+	    (3) On each traceable object,
+	      (a) object        was not oldgen (== newly or shady object) &&
+	      (b) parent object was     oldgen &&
+	      (c) parent object was not remembered &&
+	      (d) object        was not remembered
+	      then, it should be WB miss.
+
+	  This idea of this checker is by Masaya Tarui <tarui@ruby-lang.org>.
+
+Sat Jun 22 15:25:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/etc/etc.c (setup_passwd): revert r41560, unnecessary
+
+Sat Jun 22 14:39:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/etc/etc.c (Init_etc): omit 'passwd' from definition of Etc::Passwd
+	  if HAVE_STRUCT_PASSWD_PW_PASSWD is not defined to prevent mismatch of
+	  fields and values in setup_passwd
+
+Sat Jun 22 14:35:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/dl/cfunc.c (rb_dlcfunc_call): Use rb_big_pack instead of
+	  rb_big2ulong_pack and rb_big2ull.
+
+	* include/ruby/intern.h (rb_big2ulong_pack): Deprecated.
+
+Sat Jun 22 14:31:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/etc/etc.c (setup_passwd): pass 0 as VALUE to rb_struct_new to
+	  prevent segfault if the compiler passes it as a 32 bit integer on
+	  a 64 bit ruby
+
+Sat Jun 22 13:47:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_pack): MEMZERO can be used even if nails is not zero.
+
+Sat Jun 22 13:43:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_getpwnam): use PRIsVALUE in format string instead
+	  of %s and RSTRING_PTR
+
+	* ext/etc/etc.c (etc_getgrnam): ditto
+
+Sat Jun 22 13:07:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (CLEAR_LOWBITS): Rewritten without RSHIFTX.
+	  (RSHIFTX): Removed.
+
+Sat Jun 22 10:38:03 2013  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (num2i32): Removed.
+	  (pack_pack): Don't use num2i32.
+
+Sat Jun 22 09:55:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (LSHIFTX): Defined to suppress a warning.
+	  (RSHIFTX): Ditto.
+	  (CLEAR_LOWBITS): Use LSHIFTX and RSHIFTX.
+	  (FILL_LOWBITS): Use LSHIFTX.
+	  Reported by ko1 via IRC.
+
+Sat Jun 22 09:11:33 2013  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 4.7.5 (r8724)
+	* test/minitest/*: ditto
+
+Sat Jun 22 07:20:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_prof_set_heap_info, after_gc_sweep): call
+	  gc_prof_set_heap_info() just after sweeping to calculate
+	  live object number correctly.
+	  (live object number = total generated number (before marking) -
+	                        total freed number (after sweeping))
+
+	* gc.c (gc_marks): record `oldgen_object_count' into current profile`
+	  record directly.
+
+	* gc.c (rgengc_rememberset_mark): same for remembered_normal_objects
+	  and remembered_shady_objects.
+
+Sat Jun 22 06:46:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace::profile): rename rb_objspace::profile::record to
+	  records (because it points a set of records) and add a field
+	  rb_objspace::profile::current_record to point a current profiling
+	  record.
+
+	* gc.c: use above fields.
+
+Sat Jun 22 06:05:36 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_giveup_promoted_writebarrier): remove `rest_sweep()'
+	  because all of remembered objects are called for gc_mark_children().
+
+Sat Jun 22 05:08:03 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rgengc_rememberset_mark): call gc_mark_children() for
+	  remembered objects directly instead of pushing on the mark stack.
+
+Sat Jun 22 04:48:53 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h (OBJ_WRITE): cast to (VALUE *) for second
+	  parameter `slot'. You don't need to write a cast (VALUE *) any more.
+
+	* class.c, compile.c, hash.c, iseq.c, proc.c, re.c, variable.c,
+	  vm.c, vm_method.c: remove cast expressions for OBJ_WRITE().
+
+Sat Jun 22 04:37:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (slot_sweep_body): rename to slot_sweep().
+	  No need to separate major/minor GC.
+
+	* gc.c (gc_setup_mark_bits): remove gc_clear_mark_bits() and unify to
+	  this function.
+
+Sat Jun 22 04:20:21 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (check_bitmap_consistency): add to check flag and bitmap consistency.
+	  Use this function in several places.
+
+Sat Jun 22 02:18:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_pack): Specialized packers implemented.
+	  (HOST_BIGENDIAN_P): New macro.
+	  (ALIGNOF): New macro.
+	  (CLEAR_LOWBITS): New macro.
+	  (FILL_LOWBITS): New macro.
+	  (swap_bdigit): New macro.
+	  (bary_2comp): Returns an int.
+
+	* internal.h (swap16): Moved from pack.c
+	  (swap32): Ditto.
+	  (swap64): Ditto.
+
+Fri Jun 21 21:29:49 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (typedef enum): Introduce flags of major gc reason.
+	* gc.c (garbage_collect_body): Ditto.
+	* gc.c (gc_profile_flags): Ditto.
+	* gc.c (gc_profile_dump_on): Ditto.
+
+Fri Jun 21 21:11:53 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (allocate_sorted_heaps): remove unused variable `add'.
+
+Fri Jun 21 20:50:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: constify RArray::as::ary and RArray::heap::ptr.
+	  Use RARRAY_ASET() or RARRAY_PTR_USE() to modify Array objects.
+
+	* array.c, gc.c: catch up above changes.
+
+Fri Jun 21 20:32:13 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (eval_string_with_cref): fix WB miss.
+
+Fri Jun 21 20:15:49 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: support write barrier protection for T_STRUCT.
+	  Introduce the following C APIs:
+	    * RSTRUCT_RAWPTR(st) returns pointer (do WB on your risk).
+	      The type of returned pointer is (const VALUE *).
+	    * RSTRUCT_GET(st, idx) returns idx-th value of struct.
+	    * RSTRUCT_SET(st, idx, v) set idx-th value by v with WB.
+	  And
+	    * RSTRUCT_PTR(st) returns pointer with shady operation.
+	      The type of returned pointer is (VALUE *).
+
+	* struct.c, re.c, gc.c, marshal.c: rewrite with above APIs.
+
+Fri Jun 21 19:38:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BDIGMAX): Use BIGRAD.
+	  (BIGLO): Use BDIGMAX.
+	  (bigdivrem1): Ditto.
+	  (bigor_int): Ditto.
+	  (rb_big_or): Ditto.
+
+Fri Jun 21 19:18:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_pack): Move the implementation for 'c' directive after
+	  pack_integer label.
+
+Fri Jun 21 19:11:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, re.c: support write barrier for T_REGEXP.
+
+	   Note: T_MATCH object is also easy to support write barriers.
+	         However, most of T_MATCH objects are short-lived objects.
+	         So I skipped to support non-shady T_MATCH.
+
+Fri Jun 21 18:56:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigsub_int): Use bdigit_roomof.
+	  (bigadd_int): Ditto.
+	  (bigand_int): Ditto.
+	  (bigor_int): Ditto.
+	  (bigxor_int): Ditto.
+
+Fri Jun 21 17:56:25 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: fix summary of benchmark result notation.
+
+Fri Jun 21 16:38:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/openssl/ossl_x509attr.c: change OSSL_X509ATTR_IS_SINGLE and
+	  OSSL_X509ATTR_SET_SINGLE macros to use ->value.set rather than
+	  ->set to fix compile failure
+
+Fri Jun 21 15:26:45 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_sweep): profile sweep time correctly when LAZY_SWEEP is
+	  disabled.
+
+	* gc.c (gc_marks_test): store oldgen count and shady count
+	  before test marking and restore them after marking.
+
+Fri Jun 21 15:07:42 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: enable lazy sweep (commit miss).
+
+Fri Jun 21 14:31:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (ruby_setenv): refine error message so include the variable
+	  name.
+
+Fri Jun 21 14:15:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix to use total_allocated_object_num and heaps_used
+	  at the GC time for profiler.
+
+Fri Jun 21 12:35:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: RGENGC_CHECK_MODE should be 0.
+
+Fri Jun 21 11:18:25 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_marks_body): fix to get `th' in this function.
+
+Fri Jun 21 10:21:44 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heaps_header/heaps_slot): embed bitmaps into heaps_slot.
+	  no need to maintain allocation/free bitmaps.
+
+Fri Jun 21 09:22:16 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (slot_sweep_body): add counters at a time.
+
+	* gc.c (gc_profile_dump_on): fix line break position.
+
+Fri Jun 21 08:14:00 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c: refactoring bitmaps. introduce bits_t type and some Consts.
+
+Fri Jun 21 08:04:32 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix to support USE_RGENGC == 0 (disable RGenGC).
+	  If USE_RGENGC==0, it caused compilation error.
+
+Fri Jun 21 08:08:11 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (lazy_sweep): Use is_lazy_sweeping()
+	* gc.c (rest_sweep): Ditto.
+	* gc.c (gc_prepare_free_objects): Ditto.
+
+Fri Jun 21 07:34:47 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_record::oldgen_objects): added.
+
+	* gc.c (gc_profile_dump_on): print the following information:
+	  * Living object counts
+	  * Free object counts
+	  If RGENGC_PROFILE > 0 then
+	  * Oldgen object counts
+	  * Remembered normal object counts
+	  * Remembered shady object counts
+
+Fri Jun 21 06:43:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_ull2big): Refactored.
+	  (rb_uint2big): Useless code removed.
+
+Fri Jun 21 05:37:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_prof_sweep_timer_stop): accumulate sweep time only when
+	  record->gc_time > 0.
+
+Fri Jun 21 00:37:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/bigdecimal: Workaround fix for bigdecimal test failures caused
+	  by [ruby-dev:47413] [Feature #8509]
+
+	* ext/bigdecimal/bigdecimal.h (BDIGIT): Make it independent from the
+	  definition for bignum.c.
+	  (SIZEOF_BDIGITS): Ditto.
+	  (BDIGIT_DBL): Ditto.
+	  (BDIGIT_DBL_SIGNED): Ditto.
+	  (PRI_BDIGIT_PREFIX): Undefine the definition.
+	  (PRI_BDIGIT_DBL_PREFIX): Ditto.
+
+	* ext/bigdecimal/bigdecimal.c (RBIGNUM_ZERO_P): Use rb_bigzero_p.
+	  (bigzero_p): Removed.
+	  (is_even): Use rb_big_pack.
+
+Thu Jun 20 22:52:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigmul1_toom3): Don't call bignorm twice.
+
+Thu Jun 20 22:49:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bignorm): Don't call bigtrunc if the result is a fixnum.
+
+Thu Jun 20 22:29:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_uint2big): Refactored.
+
+Thu Jun 20 22:24:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (dump_bignum): Use SIZEOF_BDIGITS.
+
+Thu Jun 20 22:22:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2ulong): Change the return type to unsigned long.
+	  (rb_big2ulong_pack): Follow the above change.
+	  (rb_big2long): Ditto.
+	  (rb_big_lshift): Ditto.
+	  (rb_big_rshift): Ditto.
+	  (rb_big_aref): Ditto.
+
+Thu Jun 20 22:02:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_unpack_internal): Return -2 when negative overflow.
+	  (bary_unpack): Set the overflowed bit if an extra BDIGIT exists.
+	  (rb_integer_unpack): Set the overflowed bit.
+
+Thu Jun 20 21:17:19 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rgengc_rememberset_mark): record
+	  (1) normal objects count in remember set
+	  (2) shady objects count in remember set
+	  each GC timing.
+
+	* gc.c (gc_profile_record_get): enable to access above information
+	  and REMOVING_OBJECTS, EMPTY_OBJECTS.
+
+Thu Jun 20 18:29:26 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: Do not use GC::Profiler::disable because
+	  GC::Profiler::disable prohibit to access profiling data. It should
+	  be spec bug.
+
+	  Skip GC::Profiler::report if RUBY_VERSION < '2.0.0'
+
+Thu Jun 20 17:59:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: stop GC::Profiler before output results.
+	  Generating GC::Profiler result under profiling causes infinite loop.
+
+Thu Jun 20 17:24:24 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/gcbench.rb: don't use __dir__ to make compatible
+	  with ruby 1.9.3.
+
+Thu Jun 20 16:57:19 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_aobench.rb: use attr_accessor/reader instead of
+	  defining methods.
+
+Thu Jun 20 16:46:46 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_aobench.rb: added.
+
+	* benchmark/gc/aobench.rb: added.
+
+Thu Jun 20 16:28:33 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_so_binary_trees.rb: disable `puts' method
+	  and change iteration parameter to increase execution time.
+
+	* benchmark/gc/binarytree.rb: added.
+
+Thu Jun 20 16:06:37 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/pentomino.rb: added.
+	  Simply load pentomino puzzle in the benchmark/ directory.
+
+Thu Jun 20 15:32:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc/redblack.rb: import red black tree benchmark from
+	  https://github.com/jruby/rubybench/blob/master/time/bench_red_black.rb
+
+	* benchmark/gc/ring.rb: add a benchmark. This benchmark create many
+	  old objects.
+
+Thu Jun 20 15:14:00 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/gc: create a directory to store GC related benchmark.
+
+	* benchmark/gc/gcbench.rb: moved from tool/gcbench.rb.
+
+	* benchmark/gc/hash(1|2).rb: ditto.
+
+	* benchmark/gc/rdoc.rb: ditto.
+
+	* benchmark/gc/null.rb: added.
+
+	* common.mk: fix rule.
+
+Thu Jun 20 14:09:54 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/hashbench1.rb: fix parameter too. Increase temporary objects.
+
+Thu Jun 20 14:01:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/hashbench1.rb: fix parameters.
+
+Thu Jun 20 14:00:34 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: remove dependency from ruby.
+
+Thu Jun 20 13:14:06 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_check_backtrace): evaluate RARRAY_AREF only once.
+	  the first argument of RB_TYPE_P is expanded twice for non-immediate
+	  types.
+
+Thu Jun 20 08:09:29 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/gcbench.rb: Summary in one line.
+
+	* common.mk: separate gcbench-hash to gcbench-hash1 and gcbench-hash2.
+
+Thu Jun 20 08:07:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BIGSIZE): New macro.
+	  (bigfixize): Use BIGSIZE.
+	  (big2ulong): Ditto.
+	  (check_shiftdown): Ditto.
+	  (rb_big_aref): Ditto.
+
+Thu Jun 20 07:46:48 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (rb_gc_writebarrier): give up rescan A and register B directly
+	  if A has huge number of children.
+
+Thu Jun 20 07:30:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: add new rules `gcbench-rdoc', `gcbench-hash'.
+
+	* tool/gcbench.rb: separate GC bench framework and process.
+
+	* tool/hashbench1.rb, tool/hashbench2.rb: add two types GC bench.
+	  hashbench1: many temporal objects (GC by newobj)
+	  hashbench2: hash size becomes bigger and bigger (GC by malloc)
+	  Two benches are executed by `gcbench-hash' rule.
+
+	* tool/rdocbench.rb: separated.
+
+Thu Jun 20 06:25:39 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/rdocbench.rb: add summary.
+
+Thu Jun 20 06:18:01 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_total_time): check objspace->profile.next_index > 0.
+
+Thu Jun 20 05:47:41 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_prof_sweep_timer_start): fix merge miss.
+
+	* gc.c (GC_PROFILE_MORE_DETAIL): set it 0.
+
+Thu Jun 20 05:38:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: Accumulate sweep time to GC time.
+	  Now [GC time] is [mark time] + [sweep time] + [misc].
+	  ([GC time] >= [mark time] + [sweep time])
+
+	* gc.c (gc_prof_sweep_slot_timer_start/stop): rename to
+	  gc_prof_sweep_timer_start/stop and locate at lazy_sweep().
+
+	* gc.c (elapsed_time_from): add a utility function.
+
+Thu Jun 20 05:08:53 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_marks): fix wrong option. FALSE means major/full GC.
+	  It should be TRUE (minor marking).
+
+Thu Jun 20 02:44:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (waitpid): should not return 0 but wait until exit
+	  unless WNOHANG is given.  waiting huge process may return while
+	  active, for some reason.
+
+Thu Jun 20 01:34:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bdigit_roomof): Use SIZEOF_BDIGITS.
+	  (bigfixize): Refine an ifdef condition.
+	  (rb_absint_size): Use bdigit_roomof.
+	  (rb_absint_singlebit_p): Ditto.
+	  (rb_integer_pack): Ditto.
+	  (integer_pack_fill_dd): Use BITSPERDIG.
+	  (integer_unpack_push_bits): Use BITSPERDIG, BIGLO and BIGDN.
+
+Thu Jun 20 01:07:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (MARKED_IN_BITMAP, FL_TEST2): return boolean value since always
+	  used as boolean value.
+
+	* gc.c (MARK_IN_BITMAP, CLEAR_IN_BITMAP): evaluate bits once.
+
+Thu Jun 20 00:05:07 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RVALUE_PROMOTED): fix type.
+
+Wed Jun 19 23:39:01 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_marks_test): rewrite checking code.
+	  When RGENGC_CHECK_MODE >= 2, all minor marking, run normal minor
+	  marking *and* major/full marking. After that, compare the results
+	  and shows BUG if a object living with major/full marking but dead
+	  with minor marking.
+	  After detecting bugs, print references information.
+	  (RGENGC_CHECK_MODE == 2, show references to dead object)
+	  (RGENGC_CHECK_MODE == 3, show all references)
+
+Wed Jun 19 23:51:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigfixize): Use rb_absint_size.
+	  (check_shiftdown): Ditto.
+	  (big2ulong): Use bdigit_roomof.
+
+Wed Jun 19 23:32:23 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RVALUE_PROMOTED): check consistency between oldgen flag and
+	  oldgen bitmap if RGENGC_CHECK_MODE > 0.
+
+Wed Jun 19 23:29:29 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_force_recycle): clear oldgen bitmap, too.
+
+Wed Jun 19 21:02:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_uint2big): Consider environments BDIGIT is bigger than
+	  long.
+	  (big2ulong): Ditto.
+	  (rb_big_aref): Ditto.
+	  (rb_big_pack): Just call rb_integer_pack.
+	  (rb_big_unpack): Just call rb_integer_unpack.
+
+Wed Jun 19 20:51:21 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* gc.c (gc_stress_get): GC.stress can be Fixnum.
+
+Wed Jun 19 19:31:30 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (DIGSPERLONG): Don't define if BDIGIT is bigger than long.
+	  (DIGSPERLL): Don't define if BDIGIT is bigger than LONG_LONG
+	  (rb_absint_size): Consider environments BDIGIT is bigger than long.
+	  Use BIGLO and BIGDN.
+	  (rb_absint_singlebit_p): Ditto.
+	  (rb_integer_pack): Ditto.
+	  (bigsub_int): Consider environments BDIGIT is bigger than long.
+	  Use SIZEOF_BDIGITS instead of sizeof(BDIGIT).
+	  (bigadd_int): Ditto.
+	  (bigand_int): Ditto.
+	  (bigor_int): Ditto.
+	  (bigxor_int): Ditto.
+
+Wed Jun 19 15:14:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h (struct rb_data_type_struct), gc.c: add
+	  rb_data_type_struct::flags. Now, this flags is passed
+	  at T_DATA object creation. You can specify FL_WB_PROTECTED
+	  on this flag.
+
+	* iseq.c: making non-shady iseq objects.
+
+	* class.c, compile.c, proc.c, vm.c: add WB for iseq objects.
+
+	* vm_core.h, iseq.h: constify fields to detect WB insertion.
+
+Wed Jun 19 15:11:13 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_mark_children): show more info for broken object.
+
+Wed Jun 19 14:04:41 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/envutil.rb (EnvUtil#rubybin): remove unnecessary
+	  unless expression.
+
+Wed Jun 19 07:47:48 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect_body): use FIX2INT for ruby_gc_stress.
+
+Wed Jun 19 07:44:31 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace::gc_stress): int -> VALUE to store Fixnum object.
+
+Wed Jun 19 07:25:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (make_deferred): clear flags to T_ZOMBIE.
+
+	* gc.c (slot_sweep_body): fix indent.
+
+Wed Jun 19 07:18:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_aref): Apply BIGLO to ~xds[i] for environment which
+	  BDIGIT is 16bit.
+
+Wed Jun 19 07:09:26 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rgengc_remember): fix output level.
+
+	* gc.c (rgengc_rememberset_mark): fix to output clear count.
+	  (shady_object_count + clear_count = count of remembered objects)
+
+Wed Jun 19 07:06:21 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rgengc_remember): check T_NONE and T_ZOMBIE
+	  if RGENGC_CHECK_MODE > 0.
+
+Wed Jun 19 07:02:19 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RGENGC_CHECK_MODE): add new check mode `3'.
+	  In this mode, show all references if there is
+	  a miss-corrected object.
+
+Wed Jun 19 06:31:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stress_set): add special option of GC.stress.
+	  `GC.stress=(flag)' accepts integer to control behavior of GC.
+	  See code for details. Of course, this feature is only for MRI.
+
+	  You can debug RGenGC (WB) using `GC.stress = 1'.
+	  Using this option, do minor marking at all possible places.
+
+	  GC::STRESS_MINOR_MARK = 1 and GC::STRESS_LAZY_SWEEP = 2
+	  seem good to add.
+
+Wed Jun 19 06:29:31 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (kwmerge_i): add WB.
+
+Wed Jun 19 06:26:49 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c: `st_update()' also has same issue of last fix.
+	  write barriers at callback function are too early.
+	  All write barriers are executed after `st_update()'
+
+Wed Jun 19 04:33:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* variable.c (rb_const_set): fix WB miss.
+
+	  WBs had located before creating reference between a klass
+	  and constant value. It causes GC bug.
+
+	    # pseudo code:
+	    WB(klass, value); # WB and remember klass
+	    st_insert(klass->const_table, const_id, value);
+
+	  `st_insert()' can cause GC before inserting `value' and
+	  forget `klass' from the remember set. After that, relationship
+	  between `klass' and `value' are created with constant table.
+	  Now, `value' can be young (shady) object and `klass' can be old
+	  object, without remembering `klass' object.
+	  At the next GC, old `klass' object will be skipped and
+	  young (shady) `value' will be miss-collected. -> GC bug
+
+	  Lesson: The place of a WB is important.
+
+Tue Jun 18 22:01:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): ensure methods of type
+	  VM_METHOD_TYPE_ATTR_SET are called with 1 argument
+
+	* test/ruby/test_module.rb
+	  (TestModule#test_attr_writer_with_no_arguments): add test
+	  [ruby-core:55543] [Bug #8540]
+
+Tue Jun 18 22:36:23 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_profile_record_flag): fix typo.
+
+Tue Jun 18 22:08:53 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/objspace/object_tracing.c: Return for ::allocation_generation
+
+Tue Jun 18 22:04:35 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/objspace/object_tracing.c: Document object_tracing methods.
+
+Tue Jun 18 21:58:17 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* gc.c: Rename rb_mObSpace -> rb_mObjSpace
+
+Tue Jun 18 20:55:05 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/objspace/objspace.c: Document ObjectSpace::InternalObjectWrapper.
+
+Tue Jun 18 20:39:04 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/objspace/object_tracing.c: Teach rdoc object_tracing.c [Bug #8537]
+
+Tue Jun 18 20:29:47 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/.document: add object_tracing.c to document file
+
+Tue Jun 18 20:20:27 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/objspace/objspace.c: rdoc on require to overview from r41355
+
+Tue Jun 18 18:39:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Check __int128.
+
+	* include/ruby/defines.h (BDIGIT_DBL): Use uint128_t if it is available.
+	  (BDIGIT): Use uint64_t if uint128_t is available.
+	  (SIZEOF_BDIGITS): Defined for above case.
+	  (BDIGIT_DBL_SIGNED): Ditto.
+	  (PRI_BDIGIT_PREFIX): Ditto.
+
+	* include/ruby/ruby.h (PRI_64_PREFIX): Defined.
+
+	* bignum.c (rb_big_pow): Don't use BITSPERDIG for the condition which
+	  rb_big_pow returns Float or Bignum.
+
+	[ruby-dev:47413] [Feature #8509]
+
+Tue Jun 18 16:43:44 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_heredoc_restore): clear lex_strterm always to get
+	  rid of marking recycled node.  this bug is revealed by r41372 with
+	  GC.stress=true.
+
+Tue Jun 18 12:53:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (nlz): Cast the result explicitly.
+	  (big2dbl): Don't assign BDIGIT values to int variable.
+
+Tue Jun 18 12:25:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_xor): Non-effective code removed.
+
+Tue Jun 18 11:26:05 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): add `generated_normal_object_count_types' for
+	  RGENGC_PROFILE >= 2.
+
+Tue Jun 18 11:02:18 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_maybe): check to skip T_NONE.
+
+	* gc.c (markable_object_p): do not need to check (flags == 0) here.
+
+Tue Jun 18 10:17:37 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* variable.c (rb_autoload): fix WB miss.
+
+Tue Jun 18 04:20:18 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_children): don't need to care about T_ZOMBIE here.
+
+Mon Jun 17 22:16:02 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/ruby/test_proc.rb (TestProc#test_block_given_method_to_proc):
+	  run test for r41359.
+
+Mon Jun 17 21:42:18 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* include/ruby/ruby.h, vm_eval.c (rb_funcall_with_block):
+	  new function to invoke a method with a block passed
+	  as an argument.
+
+	* string.c (sym_call): use the above function to avoid
+	  a block sharing. [ruby-dev:47438] [Bug #8531]
+
+	* vm_insnhelper.c (vm_yield_with_cfunc): don't set block
+	  in the frame.
+
+	* test/ruby/test_symbol.rb (TestSymbol#test_block_given_to_proc):
+	  run related tests.
+
+Mon Jun 17 21:33:27 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* include/ruby/intern.h, proc.c (rb_method_call_with_block):
+	  new function to invoke a Method object with a block passed
+	  as an argument.
+
+	* proc.c (bmcall): use the above function to avoid a block sharing.
+	  [ruby-core:54626] [Bug #8341]
+
+	* test/ruby/test_proc.rb (TestProc#test_block_persist_between_calls):
+	  run related tests.
+
+Mon Jun 17 20:53:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* loadpath.c (RUBY_REVISION): Defined to suppress revision.h
+	  inclusion actually.  r41352 removes the dependency.
+
+Mon Jun 17 18:15:57 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/objspace/objspace.c: let rdoc know about objspace methods.
+	  Specify 'objspace' should be required. See #8537.
+
+Mon Jun 17 17:44:31 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* gc.c (ObjectSpace): is a module not a class.
+
+	* ext/objspace/objspace.c: try to include overview in rdoc,
+	  see #8537.
+
+Mon Jun 17 17:38:24 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* gc.c: fix example of ObjectSpace.define_finalizer in overview
+
+Mon Jun 17 16:59:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tkutil/tkutil.c: use rb_sprintf(), rb_id2str(), and
+	  rb_intern_str() instead of rb_intern() and RSTRING_PTR() with
+	  RB_GC_GUARD(), to prevent temporary objects from GC.
+	  [ruby-core:39000] [Bug #5199]
+
+Mon Jun 17 14:27:54 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm_backtrace.c: Update rdoc for Backtrace#label with @_ko1
+
+Mon Jun 17 13:04:01 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* tool/ifchange (until): Fix the condition, although harmless in
+	  this case.
+
+Mon Jun 17 11:50:29 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_maybe): added. check `is_pointer_to_heap()' and
+	  type is not T_ZOMBIE.
+
+	* gc.c: use `gc_mark_maybe()'. T_ZOMBIE objects should not be pushed
+	  to the mark stack.
+
+Mon Jun 17 07:56:24 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_small_lshift): Renamed from bdigs_small_lshift.
+	  (bary_small_rshift): Renamed from bdigs_small_rshift.
+
+Mon Jun 17 07:38:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (absint_numwords_bytes): Removed.
+	  (rb_absint_numwords): Don't call absint_numwords_bytes.
+
+Sun Jun 16 23:14:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (BARY_ADD): New macro.
+	  (BARY_SUB): Ditto.
+	  (BARY_MUL): Ditto.
+	  (BARY_DIVMOD): Ditto.
+	  (BARY_ZERO_P): Ditto.
+	  (absint_numwords_generic): Use these macros.
+
+Sun Jun 16 21:41:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_2comp): Extracted from get2comp.
+	  (integer_unpack_num_bdigits): Extracted from
+	  rb_integer_unpack_internal.
+	  (bary_unpack_internal): Renamed from bary_unpack and support
+	  INTEGER_PACK_2COMP.
+	  (bary_unpack): New function to validate arguments and invoke
+	  bary_unpack_internal.
+	  (rb_integer_unpack_internal): Removed.
+	  (rb_integer_unpack): Invoke bary_unpack_internal.
+	  (rb_integer_unpack_2comp): Removed.
+
+	* internal.h (rb_integer_unpack_2comp): Removed.
+
+	* pack.c: Follow the above change.
+
+Sun Jun 16 18:41:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (INTEGER_PACK_2COMP): Defined.
+	  (rb_integer_pack_2comp): Removed.
+
+	* bignum.c (bary_pack): Support INTEGER_PACK_2COMP.
+	  (rb_integer_pack): Invoke bary_pack directly.
+	  (rb_integer_pack_2comp): Removed.
+	  (rb_integer_pack_internal): Ditto.
+	  (absint_numwords_generic): Follow the above change.
+
+	* pack.c (pack_pack): Ditto.
+
+	* sprintf.c (rb_str_format): Ditto.
+
+Sun Jun 16 17:48:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (absint_numwords_generic): rb_funcall invocations removed.
+
+Sun Jun 16 16:04:38 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/config_files.rb: use URI.read to allow it runs with Ruby 1.8.5.
+
+Sun Jun 16 14:32:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_pack) Extracted from rb_integer_pack_internal.
+	  (absint_numwords_generic): Use bary_pack.
+
+Sun Jun 16 11:01:57 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* NEWS (XMLRPC::Client#http): Add.
+	  [ruby-core:55197] [Feature #8461]
+
+Sun Jun 16 10:38:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_add): New function.
+	  (bary_zero_p): Extracted from bigzero_p.
+	  (absint_numwords_generic): Use bary_zero_p and bary_add.
+	  (bary_mul): Fix an argument for bary_mul_single.
+	  (bary_divmod): Use size_t for arguments.
+
+Sun Jun 16 08:55:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Use a BDIGIT variable to store the return
+	  value of bigdivrem_single.
+
+Sun Jun 16 08:43:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_divmod): New function.
+	  (absint_numwords_generic): Use bary_divmod.
+	  (bigdivrem_num_extra_words): Extracted from bigdivrem.
+	  (bigdivrem_single): Ditto.
+	  (bigdivrem_normal): Ditto.
+	  (BIGDIVREM_EXTRA_WORDS): Defined.
+
+Sun Jun 16 05:51:51 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c: Fixup around GC by MALLOC.
+	  Add allocate size to malloc_increase before GC
+	  for updating limit in after_gc_sweep.
+	  Reset malloc_increase into garbage_collect()
+	  for preventing GC again soon.
+
+Sun Jun 16 05:15:36 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c: Add some columns to more detail profile.
+	  new columns: Allocated size, Prepare Time, Removing Objects, Empty Objects
+
+Sun Jun 16 02:04:40 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_prof_timer_stop): Merge function codes of GC_PROFILE_MORE_DETAIL and !GC_PROFILE_MORE_DETAIL.
+	* gc.c (gc_prof_mark_timer_start): Ditto.
+	* gc.c (gc_prof_mark_timer_stop): Ditto.
+	* gc.c (gc_prof_sweep_slot_timer_start): Ditto.
+	* gc.c (gc_prof_sweep_slot_timer_stop): Ditto.
+	* gc.c (gc_prof_set_malloc_info): Ditto.
+	* gc.c (gc_prof_set_heap_info): Ditto.
+
+Sat Jun 15 23:50:24 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_sub): New function.
+	  (absint_numwords_generic): Use bary_sub.
+	  (bigsub_core): Skip unnecessary copy.
+
+Sat Jun 15 22:05:30 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul): New function.
+	  (absint_numwords_generic): Use bary_mul.
+	  (bary_mul_single): Extracted from bigmul1_single.
+	  (bary_mul_normal): Extracted from bigmul1_normal.
+
+Sat Jun 15 20:13:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_unpack): Extracted from rb_integer_unpack_internal.
+	  (absint_numwords_generic): Use bary_unpack.
+	  (roomof): Defined.
+	  (bdigit_roomof): Defined.
+	  (BARY_ARGS): Defined.
+	  (bary_unpack): Declared.
+
+Sat Jun 15 19:35:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (absint_numwords_bytes): Make it static.
+	  (absint_numwords_small): Ditto.
+	  (absint_numwords_generic): Ditto.
+
+Sat Jun 15 17:14:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigmul1_normal): Shrink the result Bignum length.
+
+Sat Jun 15 10:19:42 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/bigdecimal/bigdecimal.c: Update overview formatting of headers
+
+Sat Jun 15 10:19:06 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/bigdecimal/bigdecimal.gemspec: Update authors
+
+Sat Jun 15 10:02:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bdigs_small_rshift): Extracted from big_rshift.
+	  (bigdivrem): Use bdigs_small_rshift.
+
+Sat Jun 15 08:37:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (eval_string_with_cref): propagate absolute path from the
+	  binding if it is given explicitly.  patch by Gat (Dawid Janczak) at
+	  [ruby-core:55123]. [Bug #8436]
+
+Sat Jun 15 02:40:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bdigs_small_lshift): Extracted from big_lshift.
+	  (bigdivrem): Use bdigs_small_lshift.
+
+Fri Jun 14 20:47:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Reduce number of digits before bignew() for div.
+
+Fri Jun 14 20:12:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Use bignew when ny == 1.
+
+Fri Jun 14 18:52:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (rb_iseq_compile_node): fix location of a `trace'
+	  instruction (b_return event).
+	  [ruby-core:55305] [ruby-trunk - Bug #8489]
+	  (need a backport to 2.0.0?)
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Fri Jun 14 18:18:07 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c, include/ruby/ruby.h: add write barriers for T_CLASS,
+	  T_MODULE, T_ICLASS.
+
+	* constant.h: constify rb_const_entry_t::value and file to detect
+	  assignment.
+
+	* variable.c, internal.h (rb_st_insert_id_and_value, rb_st_copy):
+	  added. update table with write barrier.
+
+	* method.h: constify some variables to detect assignment.
+
+	* object.c (init_copy): add WBs.
+
+	* variable.c: ditto.
+
+	* vm_method.c (rb_add_method): ditto.
+
+Fri Jun 14 14:33:47 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* NEWS: add a note for Module#using.
+
+Fri Jun 14 13:40:27 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* .travis.yml (before_script): update config files.
+
+	* common.mk ($(srcdir)/tool/config.{guess,sub}): use get-config_files.
+
+	* tool/config_files.rb: split get-config_files.
+
+	* common.mk (update-config_files): rule to download config files.
+
+	* tool/config.guess, tool/config.sub: remove and download from the
+	  upstream.
+
+	* tool/config_files.rb: download config files from GNU.
+
+Fri Jun 14 12:21:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): suppress warnings
+	  "left-hand operand of comma expression has no effect", on gcc 4.4.
+
+Fri Jun 14 09:48:48 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* NEWS: add notes for $SAFE.
+
+	* doc/security.rdoc: remove the description of $SAFE=4.
+
+Fri Jun 14 00:14:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Zero test condition simplified.
+
+Thu Jun 13 23:43:11 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/bigdecimal/*: improve documentation, nodoc samples with @mrkn
+
+Thu Jun 13 23:02:14 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client#http): Add reader for raw
+	  Net::HTTP. [ruby-core:55197] [Feature #8461]
+	  Reported by Herwin Weststrate. Thanks!!!
+
+Thu Jun 13 22:44:52 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client#parse_set_cookies): Support
+	  multiple names in a response. [ruby-core:41711] [Bug #5774]
+	  Reported by Roman Riha. Thanks!!!
+	* test/xmlrpc/test_client.rb (XMLRPC::ClientTest#test_cookie_override):
+	  Add a test of the above case.
+
+Thu Jun 13 22:35:50 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client#parse_set_cookies): Use
+	  guard style.
+
+Thu Jun 13 22:12:32 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils#rmdir): fix traversal loop, not trying
+	  remove same directory only.
+
+Thu Jun 13 21:30:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (opt-dir), tool/ifchange: get rid of "alternate value"
+	  expansion for legacy sh.  [ruby-dev:47420] [Bug #8524]
+
+Thu Jun 13 21:24:09 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Refactored to use ALLOCV_N for temporary
+	  buffers.
+
+Thu Jun 13 18:54:11 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* bignum.c (integer_unpack_num_bdigits_generic): reorder terms (but not
+	  changed the intention of the expression) because VC++ reports a
+	  warning for it.  reported by ko1 via IRC.
+
+Thu Jun 13 18:53:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_thread.rb (test_thread_local_security): Don't create
+	  an unused thread.
+
+Thu Jun 13 18:34:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bigdivrem): Use nlz.
+
+Thu Jun 13 14:51:06 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RUBY_SAFE_LEVEL_CHECK): check constant safe
+	  level at compile time.
+
+Thu Jun 13 14:39:08 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/-ext-/test_printf.rb, test/rss/test_parser.rb,
+	  test/ruby/test_array.rb, test/ruby/test_hash.rb,
+	  test/ruby/test_m17n.rb, test/ruby/test_marshal.rb,
+	  test/ruby/test_object.rb, test/ruby/test_string.rb: don't use
+	  untrusted?, untrust, and trust to avoid warnings in case $VERBOSE is
+	  true.
+
+Thu Jun 13 10:47:16 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* bootstraptest/test_autoload.rb, bootstraptest/test_method.rb:
+	  remove tests for $SAFE=4.
+
+	* lib/pp.rb: use taint instead of untrust to avoid warnings when
+	  $VERBOSE is set to true.
+
+Thu Jun 13 06:12:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_unpack_num_bdigits_small): Fix a compile error on
+	  clang -Werror,-Wshorten-64-to-32
+	  Reported by Eric Hodel.  [ruby-core:55467] [Bug #8522]
+
+Thu Jun 13 05:32:13 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/socket/extconf.rb:  Enable RFC 3542 IPV6 socket options for OS X
+	  10.7+.  [ruby-trunk - Bug #8517]
+
+Thu Jun 13 00:17:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_unpack_2comp): New function.
+	  (rb_integer_unpack_internal): Extracted from rb_integer_unpack and
+	  nlp_bits_ret argument added.
+	  (integer_unpack_num_bdigits_small): nlp_bits_ret argument added to
+	  return number of leading padding bits.
+	  (integer_unpack_num_bdigits_generic): Ditto.
+
+	* internal.h (rb_integer_unpack_2comp): Declared.
+
+	* pack.c (pack_unpack): Use rb_integer_unpack_2comp and
+	  rb_integer_unpack.
+
+Wed Jun 12 23:27:03 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (mod_using): new method Module#using, which activates
+	  refinements of the specified module only in the current class or
+	  module definition.  [ruby-core:55273] [Feature #8481]
+
+	* test/ruby/test_refinement.rb: related test.
+
+Wed Jun 12 22:58:48 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* safe.c (rb_set_safe_level, safe_setter): raise an ArgumentError
+	  when $SAFE is set to 4.  $SAFE=4 is now obsolete.
+	  [ruby-core:55222] [Feature #8468]
+
+	* object.c (rb_obj_untrusted, rb_obj_untrust, rb_obj_trust):
+	  Kernel#untrusted?, untrust, and trust are now deprecated.
+	  Their behavior is same as tainted?, taint, and untaint,
+	  respectively.
+
+	* include/ruby/ruby.h (OBJ_UNTRUSTED, OBJ_UNTRUST): OBJ_UNTRUSTED()
+	  and OBJ_UNTRUST() are aliases of OBJ_TAINTED() and OBJ_TAINT(),
+	  respectively.
+
+	* array.c, class.c, debug.c, dir.c, encoding.c, error.c, eval.c,
+	  ext/curses/curses.c, ext/dbm/dbm.c, ext/dl/cfunc.c,
+	  ext/dl/cptr.c, ext/dl/dl.c, ext/etc/etc.c, ext/fiddle/fiddle.c,
+	  ext/fiddle/pointer.c, ext/gdbm/gdbm.c, ext/readline/readline.c,
+	  ext/sdbm/init.c, ext/socket/ancdata.c, ext/socket/basicsocket.c,
+	  ext/socket/socket.c, ext/socket/udpsocket.c,
+	  ext/stringio/stringio.c, ext/syslog/syslog.c, ext/tk/tcltklib.c,
+	  ext/win32ole/win32ole.c, file.c, gc.c, hash.c, io.c, iseq.c,
+	  load.c, marshal.c, object.c, proc.c, process.c, random.c, re.c,
+	  safe.c, string.c, thread.c, transcode.c, variable.c,
+	  vm_insnhelper.c, vm_method.c, vm_trace.c: remove code for
+	  $SAFE=4.
+
+	* test/dl/test_dl2.rb, test/erb/test_erb.rb,
+	  test/readline/test_readline.rb,
+	  test/readline/test_readline_history.rb, test/ruby/test_alias.rb,
+	  test/ruby/test_array.rb, test/ruby/test_dir.rb,
+	  test/ruby/test_encoding.rb, test/ruby/test_env.rb,
+	  test/ruby/test_eval.rb, test/ruby/test_exception.rb,
+	  test/ruby/test_file_exhaustive.rb, test/ruby/test_hash.rb,
+	  test/ruby/test_io.rb, test/ruby/test_method.rb,
+	  test/ruby/test_module.rb, test/ruby/test_object.rb,
+	  test/ruby/test_pack.rb, test/ruby/test_rand.rb,
+	  test/ruby/test_regexp.rb, test/ruby/test_settracefunc.rb,
+	  test/ruby/test_struct.rb, test/ruby/test_thread.rb,
+	  test/ruby/test_time.rb: remove tests for $SAFE=4.
+
+Wed Jun 12 22:18:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_unpack_num_bdigits_generic): Rewritten without
+	  rb_funcall.
+	  (integer_unpack_num_bdigits_bytes): Removed.
+	  (rb_integer_unpack): integer_unpack_num_bdigits_bytes invocation
+	  removed.
+
+Wed Jun 12 20:18:03 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/xmlrpc/client.rb (XMLRPC::Client#parse_set_cookies): Extract.
+
+Wed Jun 12 18:19:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (validate_integer_pack_format): supported_flags argument
+	  added and validate given flags.
+	  (rb_integer_pack_internal): Specify supported_flags.
+	  (rb_integer_unpack): Ditto.
+
+Wed Jun 12 16:41:38 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): remove duplicated assertions.
+	  ARY_HEAP_PTR() implies ary not to be embedded.  [ruby-dev:47419]
+	  [Bug #8518]
+
+Wed Jun 12 12:44:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_getc): fix 7bit coderange condition, check if ascii read
+	  data instead of read length. [ruby-core:55444] [Bug #8516]
+
+Wed Jun 12 12:35:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_pack): Use rb_integer_pack_2comp.
+
+Wed Jun 12 12:07:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* sprintf.c (rb_str_format): Fix a dynamic format string.
+
+Wed Jun 12 12:04:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_uniq_bang): must not be modified once frozen even in
+	  a callback method.
+
+Wed Jun 12 12:03:43 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): must not be modified once frozen even in
+	  a callback method.
+
+Wed Jun 12 12:00:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (FL_SET_EMBED): shared object is frozen even when get
+	  unshared.
+
+	* array.c (rb_ary_modify): ARY_SET_CAPA needs unshared array.
+
+Wed Jun 12 07:32:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (rand_int): Use rb_big_uminus.
+
+Wed Jun 12 07:12:54 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* struct.c:  Improve documentation: replace "instance variable" with
+	  "member", recommend the use of a block to customize structs, note
+	  that member accessors are created, general cleanup.
+
+Wed Jun 12 06:35:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (INTEGER_PACK_NEGATIVE): Defined.
+	  (rb_integer_unpack): sign argument removed.
+
+	* bignum.c (rb_integer_unpack): sign argument removed.
+	  Non-negative integers generated by default.
+	  INTEGER_PACK_NEGATIVE flag is used to generate non-positive integers.
+
+	* pack.c (pack_unpack): Follow the above change.
+
+	* random.c (int_pair_to_real_inclusive): Ditto.
+	  (make_seed_value): Ditto.
+	  (mt_state): Ditto.
+	  (limited_big_rand): Ditto.
+
+	* marshal.c (r_object0): Ditto.
+
+Wed Jun 12 00:07:46 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/xmlrpc/test_client.rb (XMLRPC::ClientTest#test_cookie_simple):
+	  Add a test for the extracted method.
+
+Tue Jun 11 23:56:24 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/xmlrpc/test_client.rb (XMLRPC::ClientTest::Fake::HTTP#started):
+	  Add a missing empty line.
+
+Tue Jun 11 23:37:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (validate_integer_pack_format): Don't require a word order
+	  flag if numwords is 1 or less.
+	  (absint_numwords_generic): Don't specify a word order for
+	  rb_integer_pack.
+
+	* hash.c (rb_hash): Ditto.
+
+	* time.c (v2w_bignum): Ditto.
+
+Tue Jun 11 23:01:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (validate_integer_pack_format): Refine error messages.
+
+Tue Jun 11 22:25:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (validate_integer_pack_format): numwords argument added.
+	  Move a varidation from rb_integer_pack_internal and rb_integer_unpack.
+	  (rb_integer_pack_internal): Follow above change.
+	  (rb_integer_unpack): Ditto.
+
+Tue Jun 11 20:52:43 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_pack_internal): Renamed from rb_integer_pack
+	  and overflow_2comp argument added.
+	  (rb_integer_pack): Just call rb_integer_pack_internal.
+	  (rb_integer_pack_2comp): New function.
+
+	* internal.h (rb_integer_pack_2comp): Declared.
+
+	* sprintf.c (rb_str_format): Use rb_integer_pack and
+	  rb_integer_pack_2comp to format binary/octal/hexadecimal integers.
+	  (ruby_digitmap): Declared.
+	  (remove_sign_bits): Removed.
+	  (BITSPERDIG): Ditto.
+	  (EXTENDSIGN): Ditto.
+
+Tue Jun 11 16:15:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_shrink_capa): shrink the capacity so it fits just with
+	  the length.
+
+	* array.c (ary_make_shared): release never used elements from frozen
+	  array to be shared.  [ruby-dev:47416] [Bug #8510]
+
+Tue Jun 11 12:49:01 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/re.rdoc: Rename to doc/regexp.rdoc
+	* re.c: Update rdoc include for rename of file
+
+Tue Jun 11 07:13:13 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* eval_error.c (error_print): keep that errat is non-shady object.
+	  and guard errat from GC.
+
+Tue Jun 11 05:04:25 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/racc/cparse/cparse.c: use rb_ary_entry() and
+	  rb_ary_subseq() instead of RARRAY_PTR.
+	  Based on a patch by Dirkjan Bussink. See Bug #8399.
+
+Mon Jun 10 23:51:51 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* array.c (rb_ary_new_from_values): fix a typo. pointed out by
+	  nagachika.
+	  http://d.hatena.ne.jp/nagachika/20130610/ruby_trunk_changes_41199_41220
+
+Mon Jun 10 21:51:03 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* ext/socket/raddrinfo.c (nogvl_getaddrinfo): Fix indent.
+
+Mon Jun 10 21:49:43 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* ext/socket/raddrinfo.c (nogvl_getaddrinfo): Add missing return
+	  value assignment.
+
+Mon Jun 10 20:58:11 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (nogvl_getaddrinfo): work around for Ubuntu
+	  13.04's getaddrinfo issue with mdns4. [ruby-list:49420]
+
+Mon Jun 10 19:34:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_pack): Returns sign instead of words.
+	  (absint_numwords_generic): Follow the above change.
+	  (big2str_base_powerof2): Follow the above change.
+
+	* internal.h: Ditto.
+
+	* hash.c (rb_hash): Ditto.
+
+	* pack.c (pack_pack): Ditto.
+
+	* random.c (int_pair_to_real_inclusive): Ditto.
+	  (rand_init): Ditto.
+	  (random_load): Ditto.
+	  (limited_big_rand): Ditto.
+
+	* time.c (v2w_bignum): Ditto.
+
+Mon Jun 10 17:20:01 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rgengc_remember): permit promoted object.
+	  (rb_gc_writebarrier -> remember)
+
+Mon Jun 10 17:14:01 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RVALUE_PROMOTE): fix parameter name (`x' to `obj')
+	  and make it inline function (like RVALUE_PROMOTE).
+
+Mon Jun 10 16:22:50 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_new_from_values): add assertion
+	  (ary should be young object).
+
+Mon Jun 10 16:05:59 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (wmap_mark): check allocation of `w->obj2wmap'.
+	  (no-allocation `w->obj2wmap' will be NULL pointer reference)
+
+Mon Jun 10 15:36:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.c (error_print): use checking functions instead of
+	  catching exceptions.
+
+	* eval_error.c (error_print): restore errinfo for the case new
+	  exception raised while printing the message.  [ruby-core:55365]
+	  [Bug #8501]
+
+	* eval_error.c (error_print): reduce calling setjmp.
+
+Mon Jun 10 12:10:06 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_unpack_num_bdigits_small: Extracted from
+	  rb_integer_unpack.
+	  (integer_unpack_num_bdigits_generic): Ditto.
+	  (integer_unpack_num_bdigits_bytes): New function.
+	  (rb_integer_unpack): Use above functions.
+	  Return a Bignum for INTEGER_PACK_FORCE_BIGNUM even when the result
+	  is zero.
+
+Mon Jun 10 05:38:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (absint_numwords_small): New function.
+	  (absint_numwords_generic): Use absint_numwords_small if possible.
+
+Mon Jun 10 01:07:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (absint_numwords_bytes): New function.
+	  (absint_numwords_generic): Extracted from rb_absint_numwords.
+	  (rb_absint_numwords): Use absint_numwords_bytes if possible.
+
+Sun Jun  9 21:33:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_absint_numwords): Return (size_t)-1 when overflow.
+	  Refine variable names.
+	  (rb_absint_size): Refine variable names.
+
+	* internal.h (rb_absint_size): Refine an argument name.
+	  (rb_absint_numwords): Ditto.
+
+Sun Jun  9 16:51:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_absint_numwords): Renamed from rb_absint_size_in_word.
+
+	* internal.h (rb_absint_numwords): Follow the above change.
+
+	* pack.c (pack_pack): Ditto.
+
+	* random.c (rand_init): Ditto.
+	  (limited_big_rand): Ditto.
+
+Sun Jun  9 14:41:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_pack): numwords_allocated argument removed.
+
+	* internal.h (rb_integer_pack): Follow the above change.
+
+	* hash.c (rb_hash): Ditto.
+
+	* time.c (v2w_bignum): Ditto.
+
+	* pack.c (pack_pack): Ditto.
+
+	* random.c (int_pair_to_real_inclusive): Ditto.
+	  (rand_init): Ditto.
+	  (random_load): Ditto.
+	  (limited_big_rand): Ditto.
+
+Sun Jun  9 09:34:44 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big2str_base_powerof2): New function.
+	  (rb_big2str0): Use big2str_base_powerof2 if base is 2, 4, 8, 16 or 32.
+
+Sun Jun  9 00:59:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* hash.c (rb_hash): Use rb_integer_pack to obtain least significant
+	  long integer.
+
+Sat Jun  8 23:56:00 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num_to_uint): Use rb_absint_size instead of
+	  RBIGNUM_LEN.
+
+Sat Jun  8 22:53:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (r_object0): Use rb_integer_unpack.
+
+Sat Jun  8 22:18:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (v2w): Use rb_absint_size instead of RBIGNUM_LEN.
+
+Sat Jun  8 21:47:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (v2w_bignum): Simplified using rb_integer_pack.
+	  (rb_big_abs_find_maxbit): Removed.
+
+Sat Jun  8 21:03:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_absint_singlebit_p): New function.
+
+	* internal.h (rb_absint_singlebit_p): Declared.
+
+	* time.c (v2w_bignum): Use rb_absint_singlebit_p instead of
+	  rb_big_abs_find_minbit.
+	  (rb_big_abs_find_minbit): Removed.
+
+Sat Jun  8 20:24:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (rb_big_abs_find_maxbit): Use rb_absint_size.
+	  (bdigit_find_maxbit): Removed.
+
+Sat Jun  8 19:47:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* class.c (include_modules_at): invalidate method cache if included
+	  module contains constants
+
+	* test/ruby/test_module.rb: add test
+
+Sat Jun  8 19:31:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* random.c (limited_big_rand): declare rnd, lim and mask as uint32_t
+	  to avoid 64 bit to 32 bit shorten warnings.
+
+Sat Jun  8 19:23:53 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/Makefile.sub: r41163 changed win32/win32.c and configure.in
+	  but it didn't treat about mswin32/mswin64, so fix it.
+	  NOTE: this needs a review by usa whether additional condition is
+	  required or not.
+
+Sat Jun  8 19:06:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c: Unused RBignum internal accessing macros removed.
+
+Sat Jun  8 19:04:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (limited_big_rand): The argument, limit, is changed to
+	  VALUE.  Use rb_integer_pack and rb_integer_unpack.
+
+Sat Jun  8 17:15:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (make_seed_value): Fix the length given for
+	  rb_integer_unpack.
+
+Sat Jun  8 16:38:02 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_unpack): Don't use rb_funcall if possible.
+
+	* random.c: Use uint32_t for elements of seed.
+	  (make_seed_value): Use rb_integer_unpack.
+
+Sat Jun  8 15:58:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (rand_init): Add a cast to fix clang compile error:
+	  random.c:410:32: error: implicit conversion loses integer precision:
+	  'size_t' (aka 'unsigned long') to 'int' [-Werror,-Wshorten-64-to-32]
+	  This cast doesn't cause a problem because len is not bigger than
+	  MT_MAX_STATE.
+
+Sat Jun  8 15:30:03 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (rand_init): Use rb_integer_pack.
+	  (roomof): Removed.
+
+Sat Jun  8 14:58:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (INTEGER_PACK_FORCE_BIGNUM): New flag constant.
+
+	* bignum.c (rb_integer_unpack): Support INTEGER_PACK_FORCE_BIGNUM.
+
+	* random.c (int_pair_to_real_inclusive): Use
+	  INTEGER_PACK_FORCE_BIGNUM to use rb_big_mul instead of rb_funcall.
+
+Sat Jun  8 14:17:01 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check for NET_LUID.  header macro varies across
+	  compiler versions.
+
+	* win32/win32.c: use configured macro.
+
+Sat Jun  8 11:59:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (int_pair_to_real_inclusive): Use rb_funcall instead of
+	  rb_big_mul because rb_integer_unpack can return a Fixnum.
+
+Sat Jun  8 11:17:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (int_pair_to_real_inclusive): Use rb_integer_pack.
+
+Sat Jun  8 09:49:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (int_pair_to_real_inclusive): Use rb_integer_unpack.
+
+Sat Jun  8 08:12:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (random_load): Use rb_integer_pack.
+
+Sat Jun  8 06:15:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (numberof): Removed.
+
+Sat Jun  8 06:00:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c: include internal.h.
+	  (mt_state): Use rb_integer_unpack.
+
+Sat Jun  8 00:55:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_pack_loop_setup): word_num_nailbytes_ret argument
+	  removed.
+	  (rb_integer_pack): Follow the above change.
+	  (rb_integer_unpack): Follow the above change.
+
+Sat Jun  8 00:37:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (validate_integer_pack_format): Renamed from
+	  validate_integer_format.
+	  (integer_pack_loop_setup): Renamed from integer_format_loop_setup.
+	  (integer_pack_fill_dd): Renamed from int_export_fill_dd.
+	  (integer_pack_take_lowbits): Renamed from int_export_take_lowbits.
+	  (integer_unpack_push_bits): Renamed from int_import_push_bits.
+
+Fri Jun  7 23:58:06 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_pack): Arguments changed.  Use flags to
+	  specify word order and byte order.
+	  (rb_integer_unpack): Ditto.
+	  (validate_integer_format): Follow the above change.
+	  (integer_format_loop_setup): Ditto.
+
+	* pack.c: Ditto.
+
+	* internal.h: Ditto.
+	  (INTEGER_PACK_MSWORD_FIRST): Defined.
+	  (INTEGER_PACK_LSWORD_FIRST): Ditto.
+	  (INTEGER_PACK_MSBYTE_FIRST): Ditto.
+	  (INTEGER_PACK_LSBYTE_FIRST): Ditto.
+	  (INTEGER_PACK_NATIVE_BYTE_ORDER): Ditto.
+	  (INTEGER_PACK_LITTLE_ENDIAN): Ditto.
+	  (INTEGER_PACK_BIG_ENDIAN): Ditto.
+
+Fri Jun  7 22:10:50 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/specification.rb (Gem::Specification#to_yaml):
+	  use Gem::NoAliasYAMLTree.create instead of Gem::NoAliasYAMLTree.new
+	  to suppress deprecated warnings.
+
+Fri Jun  7 21:39:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_integer_pack): Renamed from rb_int_export.
+	  (rb_integer_unpack): Renamed from rb_int_import.
+
+	* internal.h, pack.c: Follow the above change.
+
+Fri Jun  7 21:05:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (integer_format_loop_setup): Extracted from rb_int_export
+	  and rb_int_import.
+
+Fri Jun  7 19:48:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (validate_integer_format): Extracted from rb_int_export and
+	  rb_int_import.
+
+Fri Jun  7 19:23:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_absint_size): Use numberof.
+	  (rb_int_export): Ditto.
+
+Fri Jun  7 18:58:56 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (numberof): Gathered from various files.
+
+	* array.c, math.c, thread_pthread.c, iseq.c, enum.c, string.c, io.c,
+	  load.c, compile.c, struct.c, eval.c, gc.c, parse.y, process.c,
+	  error.c, ruby.c: Remove the definitions of numberof.
+
+Fri Jun  7 18:24:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_absint_size): Declare a variable, i, just before used
+	  to suppress a warning.
+	  (rb_int_export): Ditto.
+
+Fri Jun  7 17:41:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* bignum.c (rb_absint_size): explicit cast to BDIGIT to avoid implicit
+	  64 bit to 32 bit shortening warning
+	* bignum.c (rb_int_export): ditto
+	* bignum.c (int_import_push_bits): ditto
+
+Fri Jun  7 17:31:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* internal.h (RCLASS_SUPER): use descriptive variable name
+	* internal.h (RCLASS_SET_SUPER): ditto
+
+Fri Jun  7 13:25:27 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/fbuffer/fbuffer.h (fbuffer_append_str): change the place of
+	  RB_GC_GUARD. it should be after the object is used.
+
+Fri Jun  7 13:22:43 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (before_gc_sweep): noinline can also avoid the segv instead of
+	  -O0 of r41084. this way is expected less slow.
+
+Fri Jun  7 11:45:42 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* rational.c (numeric_quo): move num_quo in numeric.c to numeric_quo
+	  in rational.c to refer canonicalization state for mathn support.
+	  [ruby-core:41575] [Bug #5736]
+
+	* numeric.c (num_quo): ditto.
+
+	* test/test_mathn.rb: add a test for the change at r41109.
+
+Fri Jun  7 11:41:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: revert r41106.  size_t may not be unsigned
+
+	* bignum.c (rb_absint_size_in_word, rb_int_export, rb_int_import): use
+	  NUM2SIZET() and SIZET2NUM() already defined in ruby/ruby.h.
+
+Fri Jun  7 11:28:37 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c: use oldgen bitmap as initial mark bitmap when major gc.
+	  so can skip oldgen bitmap check around mark & sweep.
+	* gc.c (slot_sweep_body): change scan algorithm for performance:
+	  from object's pointer base to bitmap one.
+
+Fri Jun  7 11:25:56 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c: introduce oldgen bitmap for preparing performance tuning.
+
+Fri Jun  7 11:20:57 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (MARKED_IN_BITMAP, MARK_IN_BITMAP, CLEAR_IN_BITMAP): bring
+	  bitmap macros in one place, and introduce BITMAP_BIT.
+
+Fri Jun  7 11:18:35 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* array.c (ary_new): change order of allocation in order
+	  to remove FL_OLDGEN operation.
+
+Fri Jun  7 11:16:28 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* tool/rdocbench.rb: add gc total time information.
+
+Fri Jun  7 10:12:01 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove "Sunny" terminology.
+	  "Sunny" doesn't mean antonym of "Shady" (questionable, doubtful, etc).
+	  Instead of "Sunny", use "non-shady" or "normal".
+
+Fri Jun  7 09:29:33 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* bignum.c (rb_int_import): explicitly casting BDIGIT_DBL to BDIGIT
+	  to prevent warning.
+
+Fri Jun  7 07:29:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_int_export): countp argument is split into
+	  wordcount_allocated and wordcount.
+
+	* bignum.c (rb_int_export): Follow the above change.
+
+	* pack.c (pack_pack): Ditto.
+
+Fri Jun  7 07:17:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: describe a compatibility issue of Numeric#quo
+	  introduced at r41109.
+
+Fri Jun  7 07:15:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: fix style.
+
+Fri Jun  7 06:48:17 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* numeric.c: remove unused ID id_to_r introduced in r41109.
+
+Fri Jun  7 06:15:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_int_import): New function.
+	  (int_import_push_bits): Ditto.
+
+	* internal.h (rb_int_import): Declared.
+
+	* pack.c (pack_unpack): Use rb_int_import for BER compressed integer.
+
+Thu Jun  6 22:24:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (num_quo): Use to_r method to convert the receiver to
+	  rational.  [ruby-core:41575] [Bug #5736]
+
+	* test/ruby/test_numeric.rb: add a test for the above change.
+
+Thu Jun  6 20:40:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Invoke RUBY_REPLACE_TYPE for size_t.
+	  Don't invoke RUBY_CHECK_PRINTF_PREFIX for size_t to avoid conflict
+	  with RUBY_REPLACE_TYPE.
+
+	* internal.h (rb_absint_size): Declared.
+	  (rb_absint_size_in_word): Ditto.
+	  (rb_int_export): Ditto.
+
+	* bignum.c (rb_absint_size): New function.
+	  (rb_absint_size_in_word): Ditto.
+	  (int_export_fill_dd): Ditto.
+	  (int_export_take_lowbits): Ditto.
+	  (rb_int_export): Ditto.
+
+	* pack.c (pack_pack): Use rb_int_export for BER compressed integer.
+
+Thu Jun  6 19:31:33 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: fixed coding error [ruby-core:55337].
+	  reported by Riley Lynch.
+
+Thu Jun  6 14:16:37 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* ext/objspace/object_tracing.c: rename allocation_info to
+	  lookup_allocation_info. At times I confused "struct
+	  allocation_info" with "function allocation_info".
+
+Thu Jun  6 13:57:06 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* ext/objspace/object_tracing.c: allocation_info function isn't
+	  called by any other file.
+
+Thu Jun  6 09:41:00 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* numeric.c (num_quo): should return a Float for a Float argument.
+	  [ruby-dev:44710] [Bug #5515]
+
+	* test/ruby/test_fixnum.rb: Add an assertion for the above change.
+
+	* test/ruby/test_bignum.rb: ditto.
+
+Thu Jun  6 00:59:44 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (gc_mark): get rid of pushing useless objects.
+	* gc.c (rgengc_rememberset_mark): bypass gc_mark() in order to push
+	  sunny old object at minor gc.
+	* gc.c (gc_mark_children): move sunny old check to gc_mark().
+	* gc.c (rgengc_check_shady): remove DEMOTE that already unnecessary.
+	* gc.c (rb_gc_writebarrier): ditto.
+
+	  change sunny old check point in order to save mark stack and
+	  remove unnatural rest_sweep & demote.
+
+Thu Jun  6 00:52:42 2013  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* gc.c (rgengc_rememberset_mark): change scan algorithm for performance:
+	  from object's pointer base to bitmap one.
+
+Thu Jun  6 00:30:04 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/win32.c (NET_LUID): define it on MinGW32.
+	  mingw-w64 has NET_LUID but mingw32 (mingw.org) still doesn't have
+	  NET_LUID. reported by taco on IRC
+
+Thu Jun  6 00:05:08 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c (String#b): Allow code range scan to happen later so
+	  ascii_only? on a result string returns the correct value.
+	  [ruby-core:55315] [Bug #8496]
+
+Wed Jun  5 22:40:42 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (capability_response): should ignore trailing
+	  spaces.  Thanks, Peter Kovacs.  [ruby-core:55024] [Bug #8415]
+
+	* test/net/imap/test_imap_response_parser.rb: related test.
+
+Wed Jun  5 21:17:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (big_fdiv): Use nlz() instead of bdigbitsize().
+	  (bdigbitsize): Removed.
+
+Wed Jun  5 20:32:00 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* include/ruby/ruby.h: fix alignment in comment.
+
+Wed Jun  5 20:05:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (int_pair_to_real_inclusive): Add a cast to BDIGIT.
+	  (random_load): Fix shift width for fixnums.
+	  Re-implement bignum extraction without ifdefs.
+
+Wed Jun  5 15:26:10 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (before_gc_sweep): don't optimize it to avoid segv on Ubuntu
+	  10.04 gcc 4.4.
+	  http://u32.rubyci.org/~chkbuild/ruby-trunk/log/20130527T190301Z.diff.html.gz
+
+Wed Jun  5 09:46:46 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb (TestFileUtils#test_mkdir): add
+	  EACCES for Windows.
+
+Wed Jun  5 08:13:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_pow): Don't need to multiply SIZEOF_BDIGITS.
+	  Use nlz instead of bitlength_bdigit.
+	  (bitlength_bdigit): Removed.
+
+Wed Jun  5 07:14:18 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (d_lite_cmp, d_lite_equal): simplified.
+
+Wed Jun  5 07:07:01 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: fixed a bug [ruby-core:55295].  reported
+	  by Riley Lynch.
+
+Wed Jun  5 06:44:08 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.0.3
+
+	* test/rubygems:  Tests for the above.
+
+	* NEWS:  Added RubyGems 2.0.3 note.
+
+Wed Jun  5 06:35:15 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/marshal.rdoc:  Add description of Marshal format.
+
+Wed Jun  5 01:16:09 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* array.c (Array#+): fix documentation example.
+	  Patch by Logan Serman. [Fixes GH-324]
+
+Wed Jun  5 00:21:54 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/irb/lc/ja/help-message: update help messages.
+	  following r41028. [ruby-dev:46707] [Feature #7510]
+
+Wed Jun  5 00:09:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* marshal.c (r_object0): Generalize a round up expression.
+	  Use BDIGIT instead of int.
+
+Tue Jun  4 23:44:02 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* object.c (rb_Hash): fix docs. patched by Stefan Sch"ussler.
+	  [ruby-core:55299] [Bug #8487]
+
+Tue Jun  4 23:16:49 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/irb/completion.rb: Use %w literal construction for long lists.
+	  Patch by Dave Goodchild. [Fixes GH-299]
+
+Tue Jun  4 23:08:42 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/objspace/objspace.c: improve wording and remove duplicated comment.
+	  Based on a patch by Dave Goodchild. [Fixes GH-299]
+
+Tue Jun  4 18:41:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bitlength_bdigit): Fix an off-by-one error.
+
+Tue Jun  4 15:30:00 2013  Kenta Murata  <mrkn@cookpad.com>
+
+	* ext/bigdecimal/lib/bigdecimal/util.rb (Float#to_d): fix the number
+	  of figures.  Patch by Vipul A M <vipulnsward@gmail.com>.
+	  https://github.com/ruby/ruby/pull/323  fix GH-323
+
+	* test/bigdecimal/test_bigdecimal_util.rb: fix for the above change.
+
+Tue Jun  4 00:44:27 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/fileutils/test_fileutils.rb (TestFileUtils#test_mkdir): add
+	  EEXIST for Linux. (suggested by nurse)
+
+Mon Jun  3 23:58:19 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/fileutils.rb (FileUtils.rmdir): use remove_tailing_slash.
+	* test/fileutils/test_fileutils.rb: test for above.
+
+Mon Jun  3 23:47:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bitlength_bdigit): New function.
+	  (rb_big_pow): Use bitlength_bdigit instead of ffs.
+
+Mon Jun  3 23:11:19 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/fileutils.rb: fix behavior when mkdir/mkdir_p accepted "/".
+	* test/fileutils/test_fileutils.rb: add test for above change.
+	  Patched by Mitsunori Komatsu. [GH-319]
+
+Mon Jun  3 19:02:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (is_hfs): use the file descriptor instead of a path.
+
+Mon Jun  3 07:15:17 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: removes AC_CHECK_FUNCS(readdir_r). readdir_r()
+	  is only used from dir.c and it doesn't need readdir_r().
+	* configure.in (SIZEOF_STRUCT_DIRENT_TOO_SMALL): removed. It is
+	  only used for readdir_r.
+	* dir.c: removes NAME_MAX_FOR_STRUCT_DIRENT. It is not right way
+	  to detect maximum length of path len. POSIX require to use
+	  fpathconf(). IOW, it might have lead to make a vulnerability
+	  using stack smashing. Moreover, readdir() works enough for our
+	  usage.
+	* dir.c (READDIR): removes an implementation which uses
+	  readdir_r() and parenthesize in a macro body correctly.
+	* dir.c (dir_read): removes IF_HAVE_READDIR_R(DEFINE_STRUCT_DIRENT
+	  entry), it is used only for readdir_r().
+	* dir.c (dir_each): ditto.
+	* dir.c (glob_helper): ditto.
+
+	* dir.c (READDIR): removes entry and dp argument.
+	* dir.c (dir_read): adjust for the above change.
+	* dir.c (dir_each): ditto.
+	* dir.c (glob_helper): ditto.
+
+Mon Jun  3 03:40:29 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): partially revert r41019.
+	  The code is not useless.
+
+Mon Jun  3 01:25:25 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/socket/test_sockopt.rb: change test name. follow r41037.
+
+Mon Jun  3 01:08:43 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/rinda/test_rinda.rb: rename functions introduced in r41009.
+
+Sun Jun  2 23:33:42 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* enc/trans/japanese_euc.trans, test/ruby/test_transcode.rb,
+	  tool/transcode-tblgen.rb: change EUC-JP-2004 to EUC-JIS-2004.
+	  This is follow up to changes in r41024.
+
+Sun Jun  2 22:44:42 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/option.c: rename functions introduced in r41009
+	  s/ip/ipv4/g because they are ipv4 functions.
+	  (there's a policy that the name "ip" is for methods which supports
+	   both ipv4 and ipv6)
+
+Sun Jun  2 16:15:29 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln_find.c (dln_find_exe, dln_find_file): remove deprecated
+	  non-reentrant functions.
+
+Sun Jun  2 15:04:35 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/cgi/util.rb, lib/erb.rb: Use String#b [Feature #8394] by znz
+
+Sun Jun  2 14:10:21 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/irb/lc/help-message: Apply english updates for irb --help #7510
+
+Sun Jun  2 12:03:58 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* range.c: Fix rdoc on Range#bsearch [Bug #8242] [ruby-core:54143]
+
+Sun Jun  2 02:08:37 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/euc_jp.c: fix typo: the name of EUC-JIS-2004.
+
+Sat Jun  1 23:17:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* vm_eval.c (rb_mod_module_eval): mention in docs that arguments passed
+	  to the method are passed to the block
+
+Sat Jun  1 17:58:13 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#freeze, taint, untaint): Save a "self" by
+	  utilizing super returning self, and add tests while at it.
+
+Sat Jun  1 17:24:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_arguments): not a simple single argument if any
+	  keyword arguments exist.  [ruby-core:55203] [Bug #8463]
+
+	* vm_insnhelper.c (vm_yield_setup_block_args): split single parameter
+	  if any keyword arguments exist, and then extract keyword arguments.
+	  [ruby-core:55203] [Bug #8463]
+
+Sat Jun  1 11:16:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_exc_new_cstr): rename from rb_exc_new2.
+
+	* error.c (rb_exc_new_str): rename from rb_exc_new3.
+
+Sat Jun  1 10:13:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_new[2-5], rb_{tainted,usascii}_str_new2),
+	  (rb_str_buf_new2): remove old interfaces.
+
+Sat Jun  1 08:00:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/zlib/zlib.c (gzfile_read, gzfile_read_all, gzfile_getc),
+	  (gzreader_gets): check EOF.  [ruby-core:55220] [Bug #8467]
+
+Sat Jun  1 07:32:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Use BDIGIT type for hbase.
+
+Sat Jun  1 02:37:35 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/option.c (sockopt_s_byte): constructor of the sockopt
+	  whose value's is byte.
+
+	* ext/socket/option.c (sockopt_byte): getter for above.
+
+	* ext/socket/option.c (inspect_byte): inspect for above.
+
+	* ext/socket/option.c (sockopt_s_ip_multicast_loop): constructor of
+	  the sockopt whose optname is IP_MULTICAST_LOOP.
+
+	* ext/socket/option.c (sockopt_ip_multicast_loop): getter for above.
+
+	* ext/socket/option.c (sockopt_s_ip_multicast_ttl): constructor of
+	  the sockopt whose optname is IP_MULTICAST_TTL.
+
+	* ext/socket/option.c (sockopt_ip_multicast_ttl): getter for above.
+
+	* ext/socket/option.c (sockopt_inspect): use above.
+
+Sat Jun 01 01:50:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_power): use rb_dbl2big
+	  to convert a double value to a Bignum.
+
+Sat Jun  1 00:19:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (calc_hbase): Make hbase the maximum power of base
+	  representable in BDIGIT.
+
+Fri May 31 23:56:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (calc_hbase): Extracted from rb_big2str0.
+
+Fri May 31 23:22:24 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c: Don't hard code SIZEOF_BDIGITS for log_base(hbase).
+	  (big2str_orig): hbase_numdigits argument added.
+	  (big2str_karatsuba): Ditto.
+	  (rb_big2str0): Calculate hbase_numdigits.
+
+Fri May 31 17:57:21 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* process.c: Improve Process::exec documentation
+
+Fri May 31 17:26:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_funcallv): add better names of rb_funcall2.
+
+	* vm_eval.c (rb_funcallv_public): ditto for rb_funcall3.
+
+Fri May 31 17:04:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_new_capa): add better names of rb_ary_new2.
+
+	* array.c (rb_ary_new_from_args): ditto for rb_ary_new3.
+
+	* array.c (rb_ary_new_from_values): ditto for rb_ary_new4.
+
+Fri May 31 16:35:44 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (HAVE_ATTRIBUTE_FUNCTION_ALIAS): define to tell if
+	  alias attribute is available.
+
+Fri May 31 16:03:23 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* object.c, proc.c: s/call_seq/call-seq in rdoc. [Fixes GH-322]
+
+Fri May 31 15:56:36 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/openssl/ossl_ssl.c: Add missing paren in rdoc [Fixes GH-321]
+
+Fri May 31 11:58:24 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (set_visibility): extract from rb_mod_public(),
+	  rb_mod_protected() and rb_mod_private().
+
+Thu May 30 19:47:42 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_insnhelper.c (vm_callee_setup_keyword_arg,
+	  vm_callee_setup_arg_complex): consider a hash argument for keyword
+	  only when the number of arguments is more than the expected
+	  mandatory parameters.  [ruby-core:53199] [ruby-trunk - Bug #8040]
+
+	* test/ruby/test_keyword.rb: update a test for above.
+
+Thu May 30 17:55:04 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* process.c: RDoc on Process.spawn
+
+Thu May 30 00:08:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_enable): rest_sweep() to finish last GC.
+	  Profiling record is allocated at first of marking phase.
+	  Enable at lazy sweeping may cause an error (SEGV).
+
+Wed May 29 10:33:27 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c: fix WB bug.
+	  (1) Hash's key also needs WB.
+	  (2) callback parameter *key and *value of st_update() is not a
+	      storage of st_table itself (only local variable). So that
+	      OBJ_WRITE() is not suitable, especially for `!existing'.
+	      OBJ_WRITTEN() is used instead of OBJ_WRITE().
+
+Tue May 28 12:31:21 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/object_tracing.c: fix a bug reported at
+	  "[ruby-core:55182] [ruby-trunk - Bug #8456][Open] Sugfault in Ruby Head"
+	  Care about the case TracePoint#path #=> `nil'.
+
+	* ext/objspace/object_tracing.c: add two new methods:
+	  * ObjectSpace.allocation_class_path(o)
+	  * ObjectSpace.allocation_method_id(o)
+	  They are not useful for Object.new because they are always
+	  "Class" and :new.
+	  To trace more useful information, we need to maintain call-tree
+	  using call/return hooks, which is implemented by
+	  ll-prof <http://sunagae.net/wiki/doku.php?id=software:llprof>
+
+	* test/objspace/test_objspace.rb: add a test.
+
+Tue May 28 11:30:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): leave makefiles untouched if the content is
+	  not changed, to get rid of unnecessary re-linking.
+
+Tue May 28 03:11:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/gc_hook.c, ext/objspace/objspace.c: add new methods to
+	  hook GC invocation.
+	  * ObjectSpace.after_gc_start_hook=(proc)
+	  * ObjectSpace.after_gc_end_hook=(proc)
+
+	  Note that hooks are not kicked immediately. Procs are kicked
+	  at postponed_job.
+
+	  This feature is a sample of new internal event and
+	  rb_postponed_job API.
+
+Tue May 28 02:56:15 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): remove wrong rest_sweep().
+
+Tue May 28 02:44:23 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect_body): fix GC_ENABLE_LAZY_SWEEP condition.
+
+	* gc.c (GC_NOTIFY): move debug print location and use stderr instead
+	  of stdout.
+
+Tue May 28 02:07:21 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_postponed_job_register_one): fix iteration bug.
+
+	* ext/-test-/postponed_job/postponed_job.c,
+	  test/-ext-/postponed_job/test_postponed_job.rb: add a test.
+
+Tue May 28 00:34:23 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, gc.c: add new internal event
+	  RUBY_INTERNAL_EVENT_GC_END. This event invokes at the end of
+	  after_sweep().
+	  Time chart with lazy sweep is:
+	    (1) Kick RUBY_INTERNAL_EVENT_GC_START
+	    (2) [gc_marks()]
+	    (3) [lazy_sweep()]
+	    (4) [... run Ruby program (mutator) with lazy_sweep() ...]
+	    (5) [after_sweep()]
+	    (6) Kick RUBY_INTERNAL_EVENT_GC_END
+	    (7) [... run Ruby program (mutator), and go to (1) ...]
+	  Time chart without lazy sweep (GC.start, etc) is:
+	    (1) Kick RUBY_INTERNAL_EVENT_GC_START
+	    (2) [gc_marks()]
+	    (3) [gc_sweep()]
+	    (4) [after_sweep()]
+	    (5) Kick RUBY_INTERNAL_EVENT_GC_END
+	    (6) [... run Ruby program (mutator), and go to (1) ...]
+
+	* ext/-test-/tracepoint/tracepoint.c,
+	  test/-ext-/tracepoint/test_tracepoint.rb: modify a test.
+
+Tue May 28 00:18:57 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_postponed_job_flush): remove a wrong comment.
+
+Mon May 27 22:09:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (RHASH_SIZE): Add a cast to suppress a
+	  warning, comparison between signed and unsigned integer
+	  expressions [-Wsign-compare], on ILP32.
+
+Mon May 27 19:25:47 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: rename RUBY_INTERNAL_EVENT_FREE to
+	  RUBY_INTERNAL_EVENT_FREEOBJ.
+
+	* ext/-test-/tracepoint/tracepoint.c,
+	  ext/objspace/object_tracing.c,
+	  gc.c, vm_trace.c: catch up this change.
+
+Mon May 27 18:57:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c: support ObjectSpace.trace_object_allocations.
+	  Read the following test to know HOWTO.
+	  This feature is a sample of RUBY_INTERNAL_EVENT.
+
+	* test/objspace/test_objspace.rb: add a test.
+
+	* ext/objspace/object_tracing.c: ditto.
+
+	* gc.c (rb_gc_count): add. This function returns GC count.
+
+	* internal.h: add decl. of rb_gc_count(). Same as `GC.count'.
+
+Mon May 27 17:33:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (install_recursive): add maxdepth option.
+
+	* tool/rbinstall.rb (bin-comm): limit depth of bindir and reject empty
+	  files.  [ruby-core:55101] [Bug #8432]
+
+Mon May 27 16:16:18 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_postponed_job_flush, rb_postponed_job_register): use
+	  ruby_xmalloc/xfree. It is safe during GC.
+
+Mon May 27 09:24:03 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* test/-ext-/postponed_job/test_postponed_job.rb: fix typo and class name.
+
+Mon May 27 09:05:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, gc.c, vm_trace.c: add internal events.
+	  * RUBY_INTERNAL_EVENT_NEWOBJ: object created.
+	  * RUBY_INTERNAL_EVENT_FREE: object freed.
+	  * RUBY_INTERNAL_EVENT_GC_START: GC started.
+	  And rename `RUBY_EVENT_SWITCH' to `RUBY_INTERNAL_EVENT_SWITCH'.
+
+	  Internal events can not invoke any Ruby program because the tracing
+	  timing may be critical (under huge restriction).
+	  These events can be hooked only by C-extensions.
+	  We recommend to use rb_postponed_job_register() API to call Ruby
+	  program safely.
+
+	  This change is mostly written by Aman Gupta (tmm1).
+	  https://bugs.ruby-lang.org/issues/8107#note-12
+	  [Feature #8107]
+
+	* include/ruby/debug.h, vm_trace.c: added two new APIs.
+	  * rb_tracearg_event_flag() returns rb_event_flag_t of this event.
+	  * rb_tracearg_object() returns created/freed object.
+
+	* ext/-test-/tracepoint/extconf.rb,
+	  ext/-test-/tracepoint/tracepoint.c,
+	  test/-ext-/tracepoint/test_tracepoint.rb: add a test.
+
+Mon May 27 08:38:21 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/-test-/postponed_job/postponed_job.c: fix `init' function name.
+
+Mon May 27 06:22:41 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/debug.h, vm_trace.c: add rb_postponed_job API.
+	  Postponed jobs are registered with this API. Registered jobs
+	  are invoked at `ruby-running-safe-point' as soon as possible.
+	  This timing is completely same as finalizer timing.
+
+	  There are two APIs:
+	  * rb_postponed_job_register(flags, func, data): register a
+	    postponed job with data. flags are reserved.
+	  * rb_postponed_job_register_one(flags, func, data): same as
+	    `rb_postponed_job_register', but only one `func' job is
+	    registered (skip if `func' is already registered).
+
+	  This change is mostly written by Aman Gupta (tmm1).
+	  https://bugs.ruby-lang.org/issues/8107#note-15
+	  [Feature #8107]
+
+	* gc.c: use postponed job API for finalizer.
+
+	* common.mk: add dependency from vm_trace.c to debug.h.
+
+	* ext/-test-/postponed_job/extconf.rb, postponed_job.c,
+	  test/-ext-/postponed_job/test_postponed_job.rb: add a test.
+
+	* thread.c: implement postponed API.
+
+	* vm_core.h: ditto.
+
+Mon May 27 02:26:02 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): collect promote_operation_count and
+	  types (RGENGC_PROFILE >= 2).
+
+Mon May 27 01:40:58 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): collect shade_operation_count,
+	  remembered_sunny_object_count and remembered_shady_object_count
+	  for each types when RGENGC_PROFILE >= 2.
+	  They are informative for optimization.
+
+Mon May 27 01:15:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c (rb_hash_tbl_raw), internal.h: added.
+	  Returns st_table without shading hash.
+
+	* array.c: use rb_hash_tbl_raw() for read-only purpose.
+
+	* compile.c (iseq_compile_each): ditto.
+
+	* gc.c (count_objects): ditto.
+
+	* insns.def: ditto.
+
+	* process.c: ditto.
+
+	* thread.c (clear_coverage): ditto.
+
+	* vm_insnhelper.c: ditto.
+
+Mon May 27 00:31:09 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/make-snapshot: use ENV["AUTOCONF"] instead of directly using
+	  literal "autoconf".
+
+Sun May 26 21:31:46 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c, include/ruby/ruby.h: support WB protected hash.
+	  * constify RHash::ifnone and make new macro RHASH_SET_IFNONE().
+	  * insert write barrier for st_update().
+
+	* include/ruby/intern.h: declare rb_hash_set_ifnone(hash, ifnone).
+
+	* marshal.c (r_object0): use RHASH_SET_IFNONE().
+
+	* ext/openssl/ossl_x509name.c (Init_ossl_x509name): ditto.
+
+Sat May 25 23:22:38 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/fiddle/test_c_struct_entry.rb,
+	  test/fiddle/test_c_union_entity.rb,
+	  test/fiddle/test_cparser.rb, test/fiddle/test_func.rb,
+	  test/fiddle/test_handle.rb, test/fiddle/test_import.rb,
+	  test/fiddle/test_pointer.rb: don't run test if the system
+	  don't support fiddle.
+
+Sat May 25 21:29:34 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/pty/pty.c (get_device_once): FreeBSD 10-current and 9-stable
+	  added O_CLOEXEC support to posix_openpt, so assume FreeBSD 9.2 or
+	  later supports it.
+	  http://www.freebsd.org/cgi/query-pr.cgi?pr=162374
+
+Sat May 25 18:46:23 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* proc.c (rb_method_entry_min_max_arity): fix missing break in switch.
+	  This was introduced in r38236, which is not intentional apparently.
+	  This has caused no actual harm because VM_METHOD_TYPE_OPTIMIZED is
+	  not used except for OPTIMIZED_METHOD_TYPE_SEND, but may do in
+	  future.  Coverity Scan found this inadequacy.
+
+Sat May 25 18:08:06 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* dir.c (bracket): fix copy-paste error.  When the first and last
+	  characters of fnmatch range have different length, fnmatch may
+	  have wrongly matched a path that does not really match.
+	  Coverity Scan found this bug.
+
+Sat May 25 17:06:25 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (after_gc_sweep): reduce full GC timing.
+
+Sat May 25 11:28:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (set_const_visibility): return without clearing method
+	  cache if no arguments.
+
+	* vm_method.c (set_method_visibility): ditto.
+
+Sat May 25 11:27:32 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (set_method_visibility): quote unprintable method name.
+
+Sat May 25 11:24:24 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_frame_callee): returns the called name of the current
+	  frame, not the previous frame.
+
+	* eval.c (prev_frame_callee, prev_frame_func): rename and make static,
+	  as these are used by rb_f_method_name() and rb_f_callee_name() only.
+
+	* variable.c (set_const_visibility): use the called name.
+
+Sat May 25 08:58:23 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_quote_unprintable): check if argument is a string.
+
+Fri May 24 19:32:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* variable.c (set_const_visibility): use rb_frame_this_func() instead
+	  of rb_frame_callee() for getting the name of the called method
+
+	* test/ruby/test_module.rb: add test for private_constant with no args
+
+Fri May 24 18:53:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: do major/full GC when:
+	  * number of oldgen object is bigger than twice of
+	    number of oldgen object at last full GC.
+	  * number of remembered shady object is bigger than twice of
+	    number of remembered shady object at last full GC.
+	  * number of oldgen object and remembered shady object is bigger
+	    than half of total object space.
+	  (please fix my English!)
+
+Fri May 24 17:07:00 2013  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* intern.h: remove dangling rb_class_init_copy declaration
+	  [ruby-core:55120] [Bug #8434]
+
+Fri May 24 16:31:23 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/strscan/strscan.c (strscan_aref): raise error if given
+	  name reference is not found.
+
+Fri May 24 15:48:18 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (after_gc_sweep, garbage_collect_body): do major GC (full GC)
+	  before extending heaps.
+	  TODO: do major GC when there are many old (promoted) objects.
+
+	* gc.c (after_gc_sweep): remove TODO comments.
+
+Fri May 24 11:04:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (LIBRUBY_RPATHFLAGS): do not append -L option with
+	  runtime library directory if cross compiling, but only -R option.
+	  runtime path makes no sense on the host system.  [ruby-dev:47363]
+	  [Bug #8443]
+
+Fri May 24 02:57:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* object.c (rb_obj_clone): should not propagate OLDGEN status.
+	  This propagation had caused WB miss for class.
+
+Thu May 23 17:35:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (loaded_feature_path): fix invalid read by index underflow.
+	  the beginning of name is also a boundary as well as just after '/'.
+
+Thu May 23 17:21:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_profile_dump_on): revert r40898.  ok to show the record
+	  accumulating while lazy_sweep().
+
+Wed May 22 16:50:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_profile_dump_on): use size_t to get rid of overflow and
+	  show the header when next_index > 0, instead of next_index != 1.
+
+Wed May 22 15:18:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (setup_overlapped): check the error code in addition
+	  to the result of SetFilePointer() to determine if an error occurred,
+	  because INVALID_SET_FILE_POINTER is a valid value.
+	  [ruby-core:55098] [Bug #8431]
+
+	* win32/win32.c (setup_overlapped, finish_overlapped): extract from
+	  rb_w32_read() and rb_w32_write().
+
+Wed May 22 14:19:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_prepare_free_objects, rest_sweep, lazy_sweep): fix position
+	  of `during_gc' setting.
+
+Wed May 22 07:36:08 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect): all GC is start from garbage_collect()
+	  (or garbage_collect_body()). `garbage_collect()' accept additional
+	  two parameters `full_mark' and `immediate_sweep'.
+	  If `full_mark' is TRUE, then force it full gc (major gc), otherwise,
+	  it depends on status of object space. Now, it will be minor gc.
+	  If `immediate_sweep' is TRUE, then disable lazy sweep.
+	  To allocate free memory, `full_mark' and `immediate_sweep' should be
+	  TRUE. Otherwise, they should be FALSE.
+
+	* gc.c (gc_prepare_free_objects): use `garbage_collect_body()'.
+
+	* gc.c (slot_sweep, before_gc_sweep, after_gc_sweep): add logging code.
+
+Tue May 21 22:47:06 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/strscan/strscan.c (strscan_aref): support named captures.
+	  patched by Konstantin Haase [ruby-core:54664] [Feature #8343]
+
+Tue May 21 21:48:44 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/ruby/test_dir_m17n.rb (TestDir_M17N#test_entries_compose):
+	  Use #each instead of #map just for iteration.
+
+Tue May 21 19:57:22 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest.rb (Digest::Class.file): Take optional
+	  arguments that are passed to the constructor of the digest
+	  class.
+
+Tue May 21 17:21:12 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove gc_profile_record::is_marked. always true.
+
+Tue May 21 17:13:40 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix to collect additional information for GC::Profiler.
+	  * major/minor GC
+	  * trigger reason of GC
+
+	* gc.c (gc_profile_dump_on): change reporting format with
+	  added information.
+
+	* gc.c (gc_profile_record_get): return added information by
+	  :GC_FLAGS => array.
+
+Tue May 21 16:45:31 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: GC::Profiler's sweeping time is accumulated all slot
+	  sweeping time. At lazy GC, GC::Profiler makes new record entry
+	  for each lazy_sweep(). In this change, accumulating all
+	  slot_sweep() time.
+	  And change indentation.
+
+Tue May 21 16:29:09 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk (rdoc-bench): add a benchmark rule
+	  using RDoc. Generate all rdoc related files
+	  (same as `make rdoc') in temporary directory
+	  and remove them. Execution time, GC::Profiler
+	  and results of GC.stat are printed.
+
+	* tool/rdocbench.rb: added for `rdoc-bench'.
+
+Tue May 21 16:25:05 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_profile_dump_on): `count' should be (int) because it
+	  can be negative number.
+	  And use pointer for `record' (don't copy).
+
+Tue May 21 03:11:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_each): compose HFS file names from
+	  UTF8-MAC. [ruby-core:48745] [Bug #7267]
+
+Tue May 21 03:08:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_separately): require envutil in the
+	  child process too.
+
+Tue May 21 03:07:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_conv_enc_opts): should infect.
+
+Mon May 20 22:24:45 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#delete_if, Set#keep_if): Avoid blockless call of
+	  proc, which is not portable to JRuby.  Replace &method() with
+	  faster and simpler literal blocks while at it.
+
+Mon May 20 22:00:31 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/e2mmap.rb: Format of E2MM documentation
+
+Mon May 20 21:41:15 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/extmk.rb: nodoc this file
+
+Mon May 20 20:43:32 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/cmath.rb: Remove duplicate RDoc heading from overview
+
+Mon May 20 20:36:19 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/securerandom.rb: Update position of overview for RDoc
+
+Mon May 20 19:33:55 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* math.c: improve and fix documentation of sin, tan and log
+
+Mon May 20 19:31:49 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/logger.rb (Logger::Application): show namespace in documentation
+
+Mon May 20 11:50:12 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/pp.rb: Revert part of r40834 and nodoc PP::ObjectMixin
+	  [ruby-core:55068]
+
+Mon May 20 10:40:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/htmlutils.rb (WEBrick::HTMLUtils#escape): replace HTML
+	  meta chars even in non-ascii string.  [Bug #8425] [ruby-core:55052]
+
+	* lib/webrick/httputils.rb (WEBrick::HTTPUtils#{_escape,_unescape}):
+	  fix %-escape encodings.  [Bug #8425] [ruby-core:55052]
+
+	* lib/webrick/httpservlet/filehandler.rb (set_dir_list): revert r20152
+	  partially and fix misuse of bytesize and regexp repetition operator.
+
+Mon May 20 08:03:51 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/profiler.rb: Document Profiler__ methods
+
+Mon May 20 08:02:13 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/tempfile.rb: nodoc Tempfile#inspect
+
+Mon May 20 07:48:24 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/stringio/stringio.c: Correct position of method rdoc
+
+Mon May 20 07:27:41 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* math.c: RDoc formatting of Math core docs with domains and codomains
+	  Patch by @eLobato [Fixes GH-309]
+
+Mon May 20 05:58:12 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/bigdecimal/bigdecimal.c: Formatting for BigMath [Fixes GH-306]
+	  Based on a patch by @eLobato.
+	* ext/bigdecimal/lib/bigdecimal/math.rb: ditto
+
+Mon May 20 04:56:59 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/forwardable.rb: Forwardable examples in overview were broken
+	  Based on patch by @joem [Fixes GH-303] [Bug #8392]
+
+Mon May 20 03:35:26 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/optparse.rb: nodoc OptionParser::Version and SPLAT_PROC
+
+Mon May 20 03:16:52 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/pp.rb: Document PP::ObjectMixin [Fixes GH-312]
+
+Sun May 19 23:52:22 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/webrick/test_htmlutils.rb: add test for WEBrick::HTMLUtils.
+
+Sun May 19 23:12:07 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* encoding.c: document fix, change default script encoding.
+	  patched by @windwiny [Fixes GH-310]
+
+Sun May 19 17:29:07 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#delete_if, Set#keep_if): Add comments.
+
+Sun May 19 11:37:36 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/fiddle/extconf.rb: ignore rc version of libffi to fix build failure.
+
+Sun May 19 10:38:50 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-delete-backward-char): Use
+	  delete-char instead of delete-backward-char, which is an
+	  interactive function.
+
+Sun May 19 03:59:29 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_scrub0): added for refactoring.
+
+Sun May 19 03:48:26 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.decode_www_form): scrub string if decoded
+	  bytes are invalid for the encoding.
+
+Sun May 19 02:46:32 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#delete_if, Set#keep_if): Make Set#delete_if and
+	  Set#keep_if more space and time efficient by avoiding to_a.
+
+Sun May 19 02:33:09 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-setup-keymap): Make
+	  backquotes electric as well.  It was listed in
+	  ruby-electric-expand-delimiters-list but not activated.
+
+	* misc/ruby-electric.el (ruby-electric-delete-backward-char):
+	  Introduce electric DEL that deletes what the previous electric
+	  command has input.
+
+	* misc/ruby-electric.el (ruby-electric-matching-char): Make
+	  electric quotes work again at the end of buffer.
+
+Sun May 19 01:39:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (setjmp-type): check if setjmpex() is really available.
+	  workaround for i686-w64-mingw32 which declares it but lacks its
+	  definition.
+
+	* include/ruby/defines.h: include setjmpex.h only if also setjmpex()
+	  is available.
+
+Sat May 18 23:57:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (setjmp-type): use setjmpex() on w64-mingw32 to get rid
+	  of -Wclobbered warnings.
+
+	* include/ruby/defines.h: include setjmpex.h here becase setjmp.h is
+	  included from win32.h via intrin.h, winnt.h, and so on.
+
+Sat May 18 20:28:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (INTEGER2NUM): Make less comparisons.
+
+Sat May 18 20:15:28 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_scrub_bang): add String#scrub!. [Feature #8414]
+
+Sat May 18 16:59:52 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (INTEGER2NUM): Renamed from INTEGER2VALUE.
+
+Sat May 18 16:57:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (INTEGER2VALUE): Suppress a warning:
+	  comparison between signed and unsigned integer expressions
+
+Sat May 18 16:38:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): forward anonymous and first keyword
+	  rest argument one.  [ruby-core:55033] [Bug #8416].
+
+Sat May 18 15:49:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_vm_tag): move jmpbuf between tag and prev so ensure to
+	  be accessible.
+
+Sat May 18 11:05:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (inspect_enumerator): use VALUE instead of mere char*
+	  by using rb_sprintf() and rb_id2str().
+
+	* enumerator.c (append_method): extract from inspect_enumerator().
+
+Sat May 18 09:00:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb (INTEGER2VALUE): Use LONG2FIX if possible.
+
+Sat May 18 00:38:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: Convert integer constants bigger than int
+	  correctly.
+
+Fri May 17 22:02:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ifaddr.c: Use unsigned LONG_LONG to represent flags
+	  because SunOS 5.11 (OpenIndiana) defines ifa_flags as uint64_t.
+
+Fri May 17 21:47:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* cont.c: Typo in constant MAX_MACHINE_STACK_CACHE from '..MAHINE..'
+	  patch by @schmurfy [Fixes GH-307]
+
+Fri May 17 19:18:24 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-matching-char): Do not put
+	  a closing quote when the quote typed does not start a string, as
+	  in $', ?\' or ?\".
+
+Fri May 17 18:06:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Consider error messages to find out version option of
+	  C compiler.
+	  The C compiler of Sun Studio C emits "Warning: Option -qversion
+	  passed to ld, if ld is invoked, ignored otherwise" and exit
+	  successfully.
+
+Fri May 17 17:34:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_guarded_ptr): unoptimize on other compilers than gcc and
+	  msvc.
+
+Fri May 17 11:06:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_intern.h (TH_PUSH_TAG): ensure jmpbuf to be accessible before
+	  pushing tag to get rid of unaccessible tag by stack overflow.
+
+Thu May 16 17:15:32 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_eval.c (rb_catch_obj): add volatile to tag to prevent crash
+	  experimentally.
+	  http://www.rubyist.net/~akr/chkbuild/debian/ruby-trunk/log/20130515T133500Z.log.html.gz
+
+Thu May 16 16:19:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (verconf.in): no longer used.
+
+	* win32/Makefile.sub (config.status): fix typo.
+
+	* configure.in, template/verconf.h.in (RUBY_EXEC_PREFIX): fix for
+	  default prefix.
+
+Thu May 16 13:12:27 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/verconf.h.in: generate verconf.h from the template and
+	  rbconfig.rb.
+
+Thu May 16 05:47:18 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: fix syntax error.
+	  Thanks @spastorino! [ruby-core:55011]
+
+Thu May 16 03:05:45 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_node_newnode): use newobj_of() instead of rb_newobj().
+
+Thu May 16 02:03:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/depend: Add a dependency for ifaddr.o.
+
+Thu May 16 01:44:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (verconf.h): $< cannot be used in explicit rules with
+	  nmake.
+
+	* win32/Makefile.sub (CONFIG_H): create verconf.in instead of
+	  verconf.h.
+
+Thu May 16 01:25:07 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: only emit warnings when
+	  -w is enabled.
+
+Wed May 15 18:58:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (newobj): rename to `newobj_of' and accept additional
+	  three parameters v1, v2, v3. newobj_of() do OBJSETUP() and
+	  fill values with v1, v2, v3.
+
+	* gc.c (rb_data_object_alloc, rb_data_typed_object_alloc):
+	  use newobj_of().
+
+Wed May 15 17:55:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_PLATFORM): move to config.h as needed by
+	  version.c.
+
+Wed May 15 17:04:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add an additional RGENGC_PROFILE mode (2).
+	  Profiling result can be check by GC.stat.
+
+	* gc.c (type_name): separate from obj_type_name().
+
+Wed May 15 16:58:24 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: save configured load path values into verconf.in.
+
+	* common.mk (verconf.h): create from verconf.in with shvar_to_cpp.rb.
+
+	* tool/shvar_to_cpp.rb: turn shell variables into C macros.
+	  [Bug #7959]
+
+	* loadpath.c: split load path staffs from version.c.
+
+	* dmyloadpath.c: miniruby has no builtin load paths, so verconf.h is
+	  not needed.
+
+Wed May 15 03:56:09 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: adding backwards
+	  compatible YAMLTree.new method
+
+Wed May 15 02:22:16 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: Adding Psych.safe_load for loading a user
+	  defined, restricted subset of Ruby object types.
+	* ext/psych/lib/psych/class_loader.rb: A class loader for
+	  encapsulating the logic for which objects are allowed to be
+	  deserialized.
+	* ext/psych/lib/psych/deprecated.rb: Changes to use the class loader
+	* ext/psych/lib/psych/exception.rb: ditto
+	* ext/psych/lib/psych/json/stream.rb: ditto
+	* ext/psych/lib/psych/nodes/node.rb: ditto
+	* ext/psych/lib/psych/scalar_scanner.rb: ditto
+	* ext/psych/lib/psych/stream.rb: ditto
+	* ext/psych/lib/psych/streaming.rb: ditto
+	* ext/psych/lib/psych/visitors/json_tree.rb: ditto
+	* ext/psych/lib/psych/visitors/to_ruby.rb: ditto
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ditto
+	* ext/psych/psych_to_ruby.c: ditto
+	* test/psych/helper.rb: ditto
+	* test/psych/test_safe_load.rb: tests for restricted subset.
+	* test/psych/test_scalar_scanner.rb: ditto
+	* test/psych/visitors/test_to_ruby.rb: ditto
+	* test/psych/visitors/test_yaml_tree.rb: ditto
+
+Wed May 15 02:06:35 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* test/psych/helper.rb: envutil is not available outside Ruby, so
+	  port the functions from envutil to the test helper.
+
+	* test/psych/test_deprecated.rb: ditto
+
+	* test/psych/test_encoding.rb: ditto
+
+Wed May 15 00:42:54 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* signal.c: need to include unistd.h for write(2).
+	  unistd.h is now included via ruby/defines.h, but should explicitly
+	  include here.  (suggested by kosaki)
+
+Tue May 14 23:43:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/.document: Add ifaddr.c.
+
+Tue May 14 23:24:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/extconf.rb: check for if_nametoindex() for
+	  i686-w64-mingw32, and check for declarations of if_indextoname() and
+	  if_nametoindex().
+
+	* ext/socket/ifaddr.c (ifaddr_ifindex): not-implement unless
+	  if_nametoindex() is available.
+
+	* ext/socket/rubysocket.h: declare if_indextoname() and
+	  if_nametoindex() if available but not declared.
+
+Tue May 14 19:58:17 2013  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/dl/lib/dl/func.rb (DL::Function#call): check tainted when
+	  $SAFE > 0.
+	* ext/fiddle/function.c (function_call): check tainted when $SAFE > 0.
+	* test/fiddle/test_func.rb (module Fiddle): add test for above.
+
+
+Tue May 14 14:51:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (INTPTR_MAX, INTPTR_MIN, UINTPTR_MAX): split
+	  from intptr_t and uintptr_t, since VC9 defines the latter only in
+	  crtdefs.h.
+
+Tue May 14 12:21:28 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (NET_LUID): mingw may have NET_LUID and not defined
+	  _IFDEF_.
+
+Tue May 14 03:33:17 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (rb_str_new_frozen): remove debug print.
+
+Tue May 14 03:22:51 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: enable to generate write barrier protected
+	  arrays (T_ARRAY).
+
+Tue May 14 03:21:42 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: enable to generate write barrier protected
+	  strings (T_STRING).
+
+Tue May 14 03:19:59 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: enable to generate write barrier protected
+	  objects (T_OBJECT).
+
+Tue May 14 03:17:15 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: enable to generate write barrier protected
+	  objects for numeric types (Float, Complex, Rational, Bignum).
+
+Tue May 14 03:10:59 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: enable RGENGC (USE_RGENGC)
+	  but no type creates write protected (sunny) objects
+	  (RGENGC_WB_PROTECTED_* == 0).
+
+Tue May 14 02:47:30 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: support RGENGC. [ruby-trunk - Feature #8339]
+	  See this ticket about RGENGC.
+
+	* gc.c: Add several flags:
+	  * RGENGC_DEBUG: if >0, then prints debug information.
+	  * RGENGC_CHECK_MODE: if >0, add assertions.
+	  * RGENGC_PROFILE: if >0, add profiling features.
+	    check GC.stat and GC::Profiler.
+
+	* include/ruby/ruby.h: disable RGENGC by default (USE_RGENGC == 0).
+
+	* array.c: add write barriers for T_ARRAY and generate sunny objects.
+
+	* include/ruby/ruby.h (RARRAY_PTR_USE): added. Use this macro if
+	  you want to access raw pointers. If you modify the contents which
+	  pointer pointed, then you need to care write barrier.
+
+	* bignum.c, marshal.c, random.c: generate T_BIGNUM sunny objects.
+
+	* complex.c, include/ruby/ruby.h: add write barriers for T_COMPLEX
+	  and generate sunny objects.
+
+	* rational.c (nurat_s_new_internal), include/ruby/ruby.h: add write
+	  barriers for T_RATIONAL and generate sunny objects.
+
+	* internal.h: add write barriers for RBasic::klass.
+
+	* numeric.c (rb_float_new_in_heap): generate sunny T_FLOAT objects.
+
+	* object.c (rb_class_allocate_instance), range.c:
+	  generate sunny T_OBJECT objects.
+
+	* string.c: add write barriers for T_STRING and generate sunny objects.
+
+	* variable.c: add write barriers for ivars.
+
+	* vm_insnhelper.c (vm_setivar): ditto.
+
+	* include/ruby/ruby.h, debug.c: use two flags
+	  FL_WB_PROTECTED and FL_OLDGEN.
+
+	* node.h (NODE_FL_CREF_PUSHED_BY_EVAL, NODE_FL_CREF_OMOD_SHARED):
+	  move flag bits.
+
+Tue May 14 01:54:48 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove rb_objspace_t::marked_num.
+	  We can use `objspace_live_num()' instead of removed `marked_num'
+	  if it is after `after_gc_sweep()' function call.
+
+	* gc.c (after_gc_sweep): use objspace_live_num() instead of removed
+	  rb_objspace_t::marked_num.
+
+	* gc.c (gc_mark_ptr, gc_marks): remove rb_objspace_t::marked_num code.
+
+	* gc.c (gc_prepare_free_objects): do not call set_heaps_increment()
+	  with checking objspace->heap.marked_num. At this point, we only
+	  need to check availability of free-cell.
+
+	* gc.c (lazy_sweep): call after_gc_sweep() if there are no sweep_able entry.
+
+	* gc.c (rest_sweep, gc_prepare_free_objects): remove after_gc_sweep() call.
+
+Tue May 14 01:50:41 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: disable GC_PROFILE_MORE_DETAIL (fix last commit).
+
+	* gc.c (gc_prof_set_malloc_info): fix "objspace->heap.live_num" to
+	  "objspace_live_num(objspace)". There is no such member variable.
+
+Tue May 14 01:25:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: refactoring GC::Profiler.
+
+	* gc.c (gc_prof_sweep_timer_start/stop): removed because
+	  they doesn't support lazy sweep.
+
+	* gc.c (gc_prof_sweep_slot_timer_start/stop): added.
+	  redefine `sweeping time' to accumulated time of all of
+	  slot_sweep().
+
+	* gc.c (rb_objspace_t::profile::count): renamed to
+	  rb_objspace_t::profile::next_index. `counter' seems ambiguous.
+	  increment it when next record is acquired.
+
+Tue May 14 00:48:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: constify RRational::(num,den) and
+	  RComplex::(real,imag).
+	  Add macro to set these values:
+	    * RRATIONAL_SET_NUM()
+	    * RRATIONAL_SET_DEN()
+	    * RCOMPLEX_SET_REAL()
+	    * RCOMPLEX_SET_IMAG()
+	  This change is a part of RGENGC branch [ruby-trunk - Feature #8339].
+
+	  TODO: API design. RRATIONAL_SET(rat,num,den) is enough?
+	  TODO: Setting constify variable with cast has same issue of r40691.
+
+	* complex.c, rational.c: use above macros.
+
+Mon May 13 21:49:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Check socketpair again.
+	  It is required on Unix.
+
+Mon May 13 21:20:32 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (getipaddrs): use alternative interface name if
+	  available, because if_nametoindex() requires them.
+
+Mon May 13 20:23:24 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (getipaddrs): [experimental]
+	  emulate getipaddrs(3) on Unix.
+
+	* win32/Makefile.sub, configure.in (LIBS): need iphlpapi.lib for above
+	  function.
+
+	* include/ruby/win32.h (socketpair): rb_w32_socketpair() doesn't
+	  substitute for any function, so use non-prefixed name.
+
+	* ext/socket/extconf.rb (socketpair); follow above change.
+
+Mon May 13 20:11:06 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (prepare_iseq_build): remove additional line break.
+
+Mon May 13 19:29:54 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: constify RBasic::klass and add
+	  RBASIC_CLASS(obj) macro which returns a class of `obj'.
+	  This change is a part of RGENGC branch [ruby-trunk - Feature #8339].
+
+	* object.c: add new function rb_obj_reveal().
+	  This function reveal internal (hidden) object by rb_obj_hide().
+	  Note that do not change class before and after hiding.
+	  Only permitted example is:
+	    klass = RBASIC_CLASS(obj);
+	    rb_obj_hide(obj);
+	    ....
+	    rb_obj_reveal(obj, klass);
+
+	  TODO: API design. rb_obj_reveal() should be replaced with others.
+
+	  TODO: modify constified variables using cast may be harmful for
+	        compiler's analysis and optimization.
+	        Any idea to prohibit inserting RBasic::klass directly?
+	        If rename RBasic::klass and force to use RBASIC_CLASS(obj),
+	        then all codes such as `RBASIC(obj)->klass' will be
+	        compilation error. Is it acceptable? (We have similar
+	        experience at Ruby 1.9,
+	        for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)".
+
+	* internal.h: add some macros.
+	  * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal
+	    object.
+	  * RBASIC_SET_CLASS(obj, cls) set RBasic::klass.
+	  * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS
+	    without write barrier (planned).
+	  * RCLASS_SET_SUPER(a, b) set super class of a.
+
+	* array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c,
+	  file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c,
+	  parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c,
+	  string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c:
+	  Use above macros and functions to access RBasic::klass.
+
+	* ext/coverage/coverage.c, ext/readline/readline.c,
+	  ext/socket/ancdata.c, ext/socket/init.c,
+	* ext/zlib/zlib.c: ditto.
+
+Mon May 13 18:44:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* *.c, parse.y, insns.def: use RARRAY_AREF/ASET macro
+	  instead of using RARRAY_PTR().
+
+Mon May 13 16:53:53 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: add new utility macros to access
+	  Array's element.
+	  * RARRAY_AREF(a, i) returns i-th element of an array `a'
+	  * RARRAY_ASET(a, i, v) set i-th element of `a' to `v'
+	  This change is a part of RGENGC branch [ruby-trunk - Feature #8339].
+
+Mon May 13 15:31:10 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* object.c (rb_obj_setup): added.
+
+	* include/ruby/ruby.h (OBJSETUP): use rb_obj_setup() instead of
+	  a macro.
+
+Mon May 13 15:24:16 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_data_object_alloc): check klass only if klass is not 0.
+	  klass==0 means internal object.
+
+Mon May 13 14:57:28 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_data_object_alloc, rb_data_typed_object_alloc):
+	  use NEWOBJ_OF() instead of NEWOBJ().
+
+Mon May 13 14:51:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_obj_singleton_method): new method Kernel#singleton_method
+	  which returns a Method object of the singleton method.
+	  non-singleton method causes NameError, but not aliased or zsuper
+	  method, right now.
+	  [ruby-core:54914] [Feature #8391]
+
+	* vm_method.c (rb_method_entry_at): return the method entry for id at
+	  klass, without ancestors.
+
+	* class.c (rb_singleton_class_get): get the singleton class if exists,
+	  or nil.
+
+Mon May 13 10:20:59 2013  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* ext/openssl/ossl_ssl.c: Disabled OpenSSL::SSL::SSLSocket if
+	  defined(OPENSSL_NO_SOCK).
+
+	  This fixes a linkage error on platforms which do not have socket.
+	  OpenSSL itself is still useful as a set of cryptographic functions
+	  even on such platforms.
+
+Mon May 13 10:30:04 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* hash.c: Hash[] and {} are not equivalent by @eam [Fixes GH-301]
+
+Mon May 13 10:04:22 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* random.c: Document Random::DEFAULT by @eLobato [Fixes GH-304]
+
+Sun May 12 21:12:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (OFFT2NUM): RUBY_REPLACE_TYPE also defines macro
+	  to convert int type to VALUE if found.
+
+Wed May  8 13:46:52 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/intern.h (rb_iv_set, rb_iv_get): removed. Because
+	  ruby.h has a declaration for that.
+
+Wed May  8 13:49:06 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/intern.h (rb_uint2big, rb_int2big, rb_uint2inum)
+	(rb_int2inum, rb_ll2inum, rb_ull2inum): removed because ruby.h
+	  has a declaration for these.
+
+Sun May 12 17:52:23 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: removes 'ac_cv_func_fseeko=yes' form MinGW
+	  specific definitions.
+
+Sun May 12 17:25:46 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (rb_file_s_truncate): use correct type. chsize takes
+	  a long.
+
+Sun May 12 17:18:46 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c: move '#define HAVE_SPAWNV 1' to win32/Makefile.sub.
+	* win32/Makefile.sub: see above.
+
+Sun May 12 17:13:32 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: removes AC_CHECK_FUNCS(setitimer) because it's
+	  unused.
+
+Sun May 12 17:08:16 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: removes AC_CHECK_FUNCS(pause) because it's unused.
+
+Sun May 12 17:05:18 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (rb_f_kill): fixes typo. s/HAS_KILLPG/HAVE_KILLPG/.
+
+Sun May 12 17:03:27 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: abort if gettimeofday doesn't exist.
+
+Sun May 12 16:31:27 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: adds RUBY_REPLACE_TYPE(off_t) for creating
+	  NUM2OFFT.
+	* file.c (rb_file_truncate): use correct type. chsize() take
+	  a long.
+	* include/ruby/ruby.h (NUM2OFFT): use a definition created by
+	  a configure script by default.
+
+Sun May 12 16:03:41 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: removes AC_CHECK_FUNC(fseeko, fseeko64, ftello,
+	  ftello64). They are not used from anywhere.
+
+	* win32/win32.c (fseeko): removes.
+	* win32/win32.c (rb_w32_ftello): removes.
+	* include/ruby/win32.h: removes declarations of rb_w32_ftello and
+	  rb_w32_fseeko.
+	* win32/Makefile.sub: removes '#define HAVE_FTELLO 1'.
+
+Sun May 12 15:51:47 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: remove AC_CHECK_FUNC(close). It is not used from
+	  anywhere.
+
+Sun May 12 15:50:45 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: adds comments for setjmp check.
+
+Sun May 12 15:38:09 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: move clock_gettime() check into regular place.
+
+Wed May  8 13:45:53 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: add getenv() declaration check.
+	* dln_find.c: add HAVE_DECL_GETENV test.
+
+Sun May 12 15:33:18 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: sorts AC_CHECK_FUNCS()s as alphabetical order.
+
+Wed May  8 13:41:57 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* bignum.c: remove redundant decl for big_lshift() big_rshift().
+
+Sun May 12 16:06:43 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/rubysocket.h (rsock_inspect_sockaddr): as r40646
+	  check HAVE_TYPE_STRUCT_SOCKADDR_DL.
+
+Sat May 11 23:01:58 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/rubysocket.h (HAVE_TYPE_STRUCT_SOCKADDR_DL):
+	  MSVC has struct sockaddr_dl, but its content is broken.
+	  http://ruby-mswin.cloudapp.net/vc10-x64/ruby-trunk/log/20130511T103938Z.log.html.gz
+
+Sat May 11 22:07:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/rinda/test_rinda.rb: Socket.getifaddrs may returns an interface
+	  which #addr method returns nil for venet0 in OpenVZ.
+
+Sat May 11 21:56:34 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (rsock_inspect_sockaddr): Add casts to
+	  suppress warnings.
+
+Sat May 11 17:28:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: New method, Socket.getifaddrs, implemented.
+	  [ruby-core:54777] [Feature #8368]
+
+Sat May 11 00:47:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* gc.h (SET_MACHINE_STACK_END): Add !defined(_ILP32) to a defining
+	  condition to avoid compilation error on x32.
+	  https://sites.google.com/site/x32abi/
+
+Fri May 10 23:56:34 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_peek_variable_name): treat invalid global, class,
+	  and instance variable names as mere strings rather than errors.
+	  [ruby-core:54885] [Bug #8375]
+
+Fri May 10 20:22:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Move library checks into "Checks for libraries." part.
+
+Fri May 10 19:32:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Reformat arguments of AC_CHECK_HEADERS and
+	  AC_CHECK_FUNCS to track modifications easily.
+
+Fri May 10 12:01:36 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Don't link librt if clock_gettime is available in
+	  the main C library.
+	  glibc 2.17 moves clock_* from librt to the main C library.
+	  http://sourceware.org/ml/libc-announce/2012/msg00001.html
+
+Thu May  9 22:00:35 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): controls_num should
+	  not be negative.
+
+Thu May  9 21:09:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* file.c, ext/etc/etc.c, ext/socket/unixsocket.c,
+	  ext/openssl/ossl.h, ext/openssl/openssl_missing.c: Use
+	  HAVE_AGGREGATE_MEMBER instead of HAVE_ST_MEMBER.
+
+Thu May  9 20:43:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): Always set
+	  controls_num to raise NotImplementedError appropriately.
+	  (bsock_recvmsg_internal): Raise NotImplementedError if
+	  :scm_rights=>true is given on platforms which don't have
+	  4.4BSD style control message.
+
+Thu May  9 12:06:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h, ext/socket/unixsocket.c,
+	  ext/socket/ancdata.c: Use HAVE_STRUCT_MSGHDR_MSG_CONTROL instead
+	  of HAVE_ST_MSG_CONTROL.
+
+Thu May  9 11:30:02 2013  Zachary Scott	 <zachary@zacharyscott.net>
+
+	* string.c: Add call-seq alias for String#=== [Bug #8381]
+
+Thu May  9 11:14:18 2013  Zachary Scott	 <zachary@zacharyscott.net>
+
+	* doc/contributing.rdoc: Add guide for contributing to CRuby
+
+Thu May  9 04:55:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Check socket library again.  shutdown() is used in
+	  io.c.
+
+Thu May  9 01:52:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Don't check socketpair.  socketpair is not used in
+	  ruby command itself.
+
+Thu May  9 01:05:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mod_included_modules): should not include non-modules.
+	  [ruby-core:53158] [Bug #8025]
+
+Wed May  8 22:46:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mod_included_modules): should not include the original
+	  module itself.  [ruby-core:53158] [Bug #8025]
+
+Wed May  8 17:43:55 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_ext_int_to_encs): ignore internal encoding if external
+	  encoding is ASCII-8BIT. [Bug #8342]
+
+Wed May  8 13:49:38 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/generator/generator.c (isArrayOrObject): cast char to
+	  unsigned char. [Bug #8378]
+
+Wed May  8 13:46:10 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json/generator/depend: fix dependencies [Bug #8379]
+
+	* ext/json/parser/depend: ditto.
+
+Wed May  8 13:07:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): fail if $, @, @@ are not followed by a valid
+	  name character.  [ruby-core:54846] [Bug #8375].
+
+Wed May  8 13:06:31 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ISGRAPH): add missing macro.
+
+Wed May  8 06:42:56 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/socket.c (socket_s_ip_address_list): fix wrongly filled
+	  sin6_scope_id on KAME introduced by r40593 for OpenIndiana.
+	  KAME uses fe80:<scope_id>::<interface id> for link-local address
+	  internally.
+	  Setting sin6_scope_id causes it leaked.
+	  see also comments of sockaddr_obj().
+
+Tue May  7 22:12:34 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/readline/readline.c (insert_ignore_escape): Add a cast to
+	  unsigned char * before dereference.
+	  This suppress a warning on Cygwin.
+
+Tue May  7 12:15:24 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): Add a cast to
+	  suppress warning.
+	  Bionic defines socklen_t as int.
+	  Bionic defines msg_controllen as unsigned int (__kernel_size_t)
+	  instead of socklen_t as POSIX.
+
+Tue May  7 12:12:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (ancillary_inspect): Don't call
+	  anc_inspect_ipv6_pktinfo if !HAVE_TYPE_STRUCT_IN6_PKTINFO.
+	  anc_inspect_ipv6_pktinfo is not defined in the case.
+
+Tue May  7 12:10:52 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (socket_s_ip_address_list): Cast EXTRA_SPACE as
+	  int.  This suppress a warning.
+
+Tue May  7 12:09:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Set close_fds false for Cygwin.
+	  Cygwin doesn't support fd passing.
+	  This enables socket extension library cross-compilable by default.
+
+Tue May  7 12:07:35 2013  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (swap32): Don't redefine it if it is already defined.
+	  Bionic defines it.
+	  (swap64): Ditto.
+
+Mon May  6 20:50:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (socket_s_ip_address_list): Fill sin6_scope_id
+	  if getifaddrs() returns an IPv6 link local address which
+	  sin6_scope_id is zero, such as on OpenIndiana SunOS 5.11.
+
+Sun May  5 18:56:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defined): use vm_search_superclass() like as normal super
+	  call.  based on a patch <https://gist.github.com/wanabe/5520026> by
+	  wanabe.
+
+	* vm_insnhelper.c (vm_search_superclass): return error but not raise
+	  exceptions.
+
+	* vm_insnhelper.c (vm_search_super_method): check the result of
+	  vm_search_superclass and raise exceptions on error.
+
+Sun May  5 16:29:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defined): get method entry from the method top level
+	  frame, not block frame.  [ruby-core:54769] [Bug #8367]
+
+Sun May  5 13:28:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/ruby.pc.in (Cflags): use rubyarchhdrdir for multiarch.
+	  [Bug #7874]
+
+Sat May  4 07:20:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/security.rdoc: Add note about reporting security vulns
+
+Sat May  4 04:13:27 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/defines.h (RUBY_ATTR_ALLOC_SIZE): New for
+	  attribute((alloc_size(params))).
+
+	* include/ruby/defines.h (xmalloc, xmalloc2, xcalloc)
+	(xrealloc, xrealloc2): Annotated by RUBY_ATTR_ALLOC_SIZE.
+	* include/ruby/ruby.h (rb_alloc_tmp_buffer): ditto.
+
+Fri May  3 19:32:13 2013  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/util.rb: All class methods modulized.
+	  We can use these methods like a function when "include CGI::Util".
+	  [Feature #8354]
+
+Fri May  3 14:09:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Make default_ipv6 true for Cygwin.
+	  Cygwin supports IPv6 since Cygwin 1.7.1 (2009-12).
+	  http://cygwin.com/ml/cygwin-announce/2009-12/msg00027.html
+
+Fri May  3 13:35:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/{getaddrinfo,getnameinfo}.c: define socklen_t if not
+	  defined, e.g., older VC.
+
+Fri May  3 13:29:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (INTPTR_MAX, INTPTR_MIN, UINTPTR_MAX): also
+	  should be defined when defining intptr_t and uintptr_t.
+	  bigdecimal.c requires the former two now.
+
+Fri May  3 13:22:12 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (poll_child_status): fix build error on older mingw.
+
+Fri May  3 00:15:58 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* common.mk: remove timestamps in distclean-ext realclean-ext.
+
+Thu May  2 23:23:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_is_kind_of): skip prepending modules.
+	  [ruby-core:54742] [Bug #8357]
+
+	* object.c (rb_class_inherited_p): ditto.
+	  [ruby-core:54736] [Bug #8357]
+
+Thu May  2 22:11:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bin/irb: remove dead code from sample/irb.rb.
+
+Thu May  2 17:32:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (copy_ivar_i): get rid of overwriting already copied
+	  instance variables.  c.f. [Bug #8276]
+
+Thu May  2 16:55:43 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (id_locals): use cached ID.
+
+	* vm.c (ruby_thread_init): ditto.
+
+	* defs/id.def: add more predefined IDs used in core.
+
+Thu May  2 13:42:42 2013  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	* lib/minitest/*: Imported minitest 4.7.4 (r8483)
+	* test/minitest/*: ditto
+
+Thu May  2 11:32:22 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (poll_child_status): [experimental] set the cause of
+	  a child's death to status if its exitcode seems to be an error.
+
+	* test/ruby/test_process.rb (TestProcess#test_no_curdir): maybe now
+	  we can test it.
+
+	* test/ruby/test_thread.rb (TestThread#test_thread_timer_and_interrupt):
+	  ditto.
+
+Thu May  2 11:24:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/yaml.rb: nodoc EngineManager, add History doc #8344
+
+Wed May  1 21:11:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (localtime_with_gmtoff_zone): musl libc may return NULL for
+	  tm_zone.
+
+Wed May  1 18:59:36 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* enum.c (Enumerable#chunk): fix grammar of error message
+	  for symbols beginning with an underscore [Bug #8351]
+
+Wed May  1 16:47:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/curses/extconf.rb (curses_version): try once for each tests, a
+	  function or a variable.  fallback to variable for old SVR4.
+
+Wed May  1 16:17:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): extensions not to be installed should not
+	  make static libraries, but make dynamic libraries always.
+
+Wed May  1 12:20:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/rake/version.rb: Fix RDoc warning with  :include: [Bug #8347]
+
+Wed May  1 11:40:25 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/id.def (predefined): add "idProc".
+
+	* eval.c (frame_func_id): use predefined IDs.
+
+	* proc.c (mnew, mproc, mlambda): use predefined IDs.
+
+	* vm.c (rb_vm_control_frame_id_and_class): ditto.
+
+	* vm.c (Init_VM): ditto.
+
+Tue Apr 30 23:18:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/benchmark.rb: Update Benchmark results on newer CPU
+
+Tue Apr 30 12:31:40 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mproc, mlambda): use frozen core methods instead of plain
+	  global methods, so that methods cannot be overridden.
+	  [ruby-core:54687] [Bug #8345]
+
+	* vm.c (Init_VM): define proc and lambda on the frozen core object.
+
+	* include/ruby/intern.h (rb_block_lambda): add declaration instead of
+	  deprecated rb_f_lambda.
+
+Mon Apr 29 17:02:30 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.h: Bionic libc doesn't have locale.
+	  [Feature #8338]
+
+
+Mon Apr 29 06:58:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_initialize): no need of alloca for
+	  small fixed size array.
+
+	* ext/openssl/ossl_bn.c (ossl_bn_initialize): check overflow first,
+	  and use alloca for small size input.
+
+Mon Apr 29 00:40:13 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/yaml.rb: Clarify documentation about YAML being always Psych.
+	  Give a tip about using Syck. See #8344.
+
+Sun Apr 28 23:34:01 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/yaml.rb: Use another trick to define the YAML module.
+	  https://twitter.com/n0kada/status/328342207511801856
+
+Sun Apr 28 23:19:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/pp.rb: Update PP module overview by @geopet
+
+Sun Apr 28 22:04:37 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_initialize): fix buffer overflow on
+	  x64 Windows and memory leak when initializing with integer.
+	  [ruby-core:54615] [Bug #8337]
+
+Sun Apr 28 12:38:04 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT: correct method name to be used.  [Bug #7982]
+
+	* README.EXT.ja: add notes too.
+
+Sun Apr 28 10:35:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* object.c: With feedback from Steve Klabnik, reverted a change to
+	  #untrusted? and #tainted?. Also adjusted grammar for $SAFE levels
+
+Sun Apr 28 10:10:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/yaml.rb: Disable setting YAML const twice [ruby-core:54642]
+
+Sun Apr 28 09:50:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* object.c: Documentation for taint and trust [Bug #8162]
+
+Sun Apr 28 09:40:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* README.EXT: Copy note from r40505 for rb_sprintf() [Bug #7982]
+
+Sun Apr 28 08:28:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/curses/curses.c: Update Curses::Window example for nicer output
+	  Patch by Michal Suchanek [Bug #8121] [ruby-core:53520]
+
+Sun Apr 28 08:10:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* README.EXT: Update note from r40504, by Jeremy Evans [Bug #7982]
+
+Sun Apr 28 08:02:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* README.EXT: Add note to warn use of %i in Exceptions [Bug #7982]
+
+Sun Apr 28 02:41:05 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Fix a typo.  Should check endgrent() instead of
+	  endgrnam().
+
+Sun Apr 28 00:35:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (obj2gid): Don't call endgrent() if not exist.
+	  Bionic (Android's libc) don't have endgrent().
+
+	* configure.in: Check endgrnam function.
+
+Sat Apr 27 23:53:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* lib/yaml.rb: add security warning to YAML documentation
+
+Sat Apr 27 23:25:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/yaml.rb: Documentation for YAML module [Bug #8213]
+
+Sat Apr 27 20:19:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* thread_pthread.c (ruby_init_stack): Add STACK_GROW_DIR_DETECTION.
+	  This fixes a compilation failure while cross-compiling for Tensilica
+	  Xtensa Processor.
+
+Sat Apr 27 19:32:44 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* thread.c: fix typos and documentation
+
+Sat Apr 27 19:04:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* sparc.c: Use __asm__ instead of asm for gcc.
+	  gcc doesn't provide asm keyword if -ansi option is given.
+	  http://gcc.gnu.org/onlinedocs/gcc/Alternate-Keywords.html
+
+Sat Apr 27 17:22:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Redundant test removed.
+
+Sat Apr 27 16:00:10 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb (test_recvmsg_with_msg_peek_creates_fds):
+	  Extracted.
+
+Sat Apr 27 15:50:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (SIGNED_INTEGER_TYPE_P): New macro.
+	  (SIGNED_INTEGER_MAX): Ditto.
+	  (SIGNED_INTEGER_MIN): Ditto.
+	  (UNSIGNED_INTEGER_MAX): Ditto.
+	  (TIMET_MAX): Use SIGNED_INTEGER_MAX and UNSIGNED_INTEGER_MAX.
+	  (TIMET_MIN): Use SIGNED_INTEGER_MIN.
+
+	* thread.c (TIMEVAL_SEC_MAX): Use SIGNED_INTEGER_MAX.
+	  (TIMEVAL_SEC_MIN): Use SIGNED_INTEGER_MIN.
+
+Sat Apr 27 10:52:52 2013  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (TIMEVAL_SEC_MAX, TIMEVAL_SEC_MIN): Consider environments,
+	  sizeof(time_t) is smaller than sizeof(tv_sec), such as
+	  OpenBSD 5.2 (amd64).
+
+Fri Apr 26 23:34:59 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/text.rb (REXML::Text.normalize): Fix a bug that all
+	  entity filters are ignored. [ruby-dev:47278] [Bug #8302]
+	  Patch by Ippei Obayashi. Thanks!!!
+	* test/rexml/test_entity.rb (EntityTester#test_entity_filter): Add
+	  a test of the above change.
+
+Fri Apr 26 22:53:55 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/element.rb (REXML::Attributes#to_a): Support
+	  namespaced attributes. [ruby-dev:47277] [Bug #8301]
+	  Patch by Ippei Obayashi. Thanks!!!
+	* test/rexml/test_attributes.rb
+	  (AttributesTester#test_to_a_with_namespaces): Add a test of the
+	  above change.
+
+Fri Apr 26 21:48:29 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/atom.rb (RSS::Atom::Entry): Fix indent of document comment.
+
+Fri Apr 26 21:21:17 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/maker.rb (RSS::Maker): Fix indent of document comment.
+
+Fri Apr 26 18:41:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Use a block of enable_config() for
+	  --{enable,disable}-close-fds-by-recvmsg-with-peek configure option
+
+Fri Apr 26 18:08:08 2013  Tanaka Akira  <akr@fsij.org>
+
+	* dir.c (dir_set_pos): Fix a compilation error when seekdir() is not
+	  exist.
+
+Fri Apr 26 17:41:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* thread_pthread.c (ruby_init_stack): Add STACK_GROW_DIR_DETECTION.
+	  This fixes a compilation failure while cross-compiling for ARM.
+
+Fri Apr 26 14:35:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/rss/atom.rb: Documentation for RSS::Atom based on a patch by
+	  Michael Denomy
+	* lib/rss/maker.rb: Documentation for RSS::Maker also by @mdenomy
+
+Fri Apr 26 12:41:22 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: Test linkability of curses_version at first.
+
+	* ext/socket/extconf.rb: Test the behavior of fd passing with MSG_PEEK
+	  only if recvmsg(), msg_control member, AF_UNIX and SCM_RIGHTS are
+	  available.
+
+Fri Apr 26 00:07:52 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* lib/rinda/ring.rb (Rinda::RingServer#initialize): accept array
+	  arguments of address to specify multicast interface.
+
+	* lib/rinda/ring.rb (Rinda::RingServer#make_socket): add optional
+	  arguments for multicast interface.
+
+	* test/rinda/test_rinda.rb
+	  (TestRingFinger#test_ring_server_ipv4_multicast,
+	   TestRingFinger#test_ring_server_ipv6_multicast): add tests for
+	  above change.
+
+	* test/rinda/test_rinda.rb
+	  (TestRingServer#test_make_socket_ipv4_multicast,
+	   TestRingServer#test_make_socket_ipv6_multicast): change bound
+	  interface address because multicast address is not allowed on Linux
+	  or Windows.
+	  [ruby-core:53692] [Bug #8159]
+
+Thu Apr 25 23:45:02 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* lib/rinda/ring.rb (Rinda::RingServer#initialize): add a socket
+	  to @sockets in make_socket() to close sockets on shutdown even if
+	  make_socket() is called after initialize.
+
+	* lib/rinda/ring.rb (Rinda::RingServer#make_socket): ditto.
+
+Thu Apr 25 23:39:42 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/rinda/test_rinda.rb (TupleSpaceProxyTest#test_take_bug_8215):
+	  use KILL on Windows since TERM doen't work and ruby process remains
+	  after test-all on Windows.
+
+Thu Apr 25 23:16:28 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/curses/extconf.rb: Implement
+	  --with-curses-version={function,variable} configure option for
+	  cross-compiling.
+
+Thu Apr 25 18:15:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Don't use WIDE getaddrinfo by default.
+
+Thu Apr 25 17:56:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Remove obsolete options: ---with-ipv6-lib and
+	  --with-ipv6-libdir.
+
+Thu Apr 25 17:43:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Implement
+	  --{enable,disable}-close-fds-by-recvmsg-with-peek configure option
+	  for cross-compiling.
+	  Make --{enable,disable}-wide-getaddrinfo configure option
+	  cross-compiling friendly.
+
+Thu Apr 25 16:11:06 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_ext_int_to_encs, parse_mode_enc): bom-prefixed name is
+	  not a real encoding name, just a fallback.  so the proper conversion
+	  should take place even if if the internal encoding is equal to the
+	  bom-prefixed name, unless actual encoding is equal to the internal
+	  encoding.  [ruby-core:54563] [Bug #8323]
+
+	* io.c (io_set_encoding_by_bom): reset extenal encoding if no BOM
+	  found.  [ruby-core:54569]
+
+Thu Apr 25 14:35:01 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_initialize): allow Fixnum and Bignum.
+	  [ruby-core:53986] [Feature #8217]
+
+Thu Apr 25 14:26:32 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI.decode_www_form): follow current URL Standard.
+	  It gets encoding argument to specify the character encoding.
+	  It now allows loose percent encoded strings, but denies ;-separator.
+	  [ruby-core:53475] [Bug #8103]
+
+	* lib/uri/common.rb (URI.decode_www_form): follow current URL Standard.
+	  It gets encoding argument to convert before percent encode.
+	  Now UTF-16 strings aren't converted to UTF-8 before percent encode
+	  by default.
+
+Wed Apr 25 14:26:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* benchmark/bm_hash_shift.rb: add benchmark for Hash#shift
+
+	* hash.c (rb_hash_shift): use st_shift if hash is not being iterated to
+	  delete element without iterating the whole hash.
+
+	* hash.c (shift_i): remove function
+
+	* include/ruby/st.h (st_shift): add st_shift function
+
+	* st.c (st_shift): ditto
+
+	[Bug #8312] [ruby-core:54524] Patch by funny-falcon
+
+Thu Apr 25 12:03:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Extract C programs as toplevel constants.
+
+Thu Apr 25 02:23:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_RM_RECURSIVE): this hack is needed by only
+	  autoconf 2.69 or earlier on darwin.
+
+Thu Apr 25 01:22:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tracer.rb (get_line): simply read by File.readlines.
+
+	* lib/debug.rb (script_lines): get source lines from SCRIPT_LINES__ or
+	  read from the file.
+
+	* lib/debug.rb (display_list): use script_lines instead of recursion.
+	  [Bug #8318]
+
+	* lib/debug.rb (line_at): use script_lines same as display_list.
+
+	* lib/debug.rb (display_list): Fix debug listing when called from the
+	  same file it has been required.  patch by Dario Bertini <berdario AT
+	  gmail.com> [Bug #8318] [fix GH-280]
+
+Wed Apr 24 21:51:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Check mblen().
+	  mblen() is optional in uClibc.
+
+	* eval_intern.h (CharNext): Don't use mblen() is not available.
+
+Wed Apr 24 15:55:06 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_fd_fix_cloexec): use rb_update_max_fd().
+
+Wed Apr 24 14:08:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* numeric.c: Fix wiki link on Float imprecision in overview, patched
+	  by Makoto Kishimoto [Bug #8304] [ruby-dev:47280]
+
+Wed Apr 24 14:03:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): disallow $- without following identifier
+	  character.  [ruby-talk:406969]
+
+	* parse.y (is_special_global_name): mere $- is not a valid global
+	  variable name.
+
+Wed Apr 24 13:54:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* string.c: Document String#setbyte return value by @gjmurakami-10gen
+	  [Fixes GH-294]
+
+Wed Apr 24 13:45:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* class.c: Example of Object#methods by @windwiny [Fixes GH-293]
+	* ruby.c: Document return values of Kernel #sub, #gsub, and #chop
+
+Wed Apr 24 12:54:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/socket/lib/socket.rb: Doc typos by @vipulnsward [Fixes GH-292]
+
+
+Wed Apr 24 12:54:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/socket/lib/socket.rb: Doc typos by @vipulnsward [Fixes GH-292]
+
+Wed Apr 24 12:27:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c: Fix documentation for Array#index and #replace aliases
+	  Based on a patch by @phiggins [Fixes GH-282]
+
+Tue Apr 23 21:14:38 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_inspect): refix r40413, on Ruby 1.9 usual character
+	  escape uses hex/Unicode escapes, so fix to use Unicode escape on
+	  Unicode strings and hex on others. [ruby-core:54458] [Bug #8290]
+
+Tue Apr 23 20:10:02 2013  Tanaka Akira  <akr@fsij.org>
+
+	* missing/isnan.c (isnan): Don't define if isnan() macro is defined.
+	  This fixes a compilation failure on uClibc based Gentoo system.
+
+Tue Apr 23 17:40:40 2013  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/rexml/document.rb, lib/rexml/element.rb,
+	  lib/rexml/formatters/pretty.rb: remove opinionated
+	  language in documentation. [Bug #8309],
+	  reported by Charles Beckmann
+
+Tue Apr 23 14:04:44 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (getacl_response): parse the mailbox of an ACL
+	  response correctly.  [ruby-core:54365] [Bug #8281]
+
+Tue Apr 23 11:58:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_scrub): fix for UTF-32.  strlen() on strings
+	  contain NUL returns wrong result, use sizeof operator instead.
+	  [ruby-dev:45975] [Feature #6752]
+
+Tue Apr 23 10:26:50 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/ruby/test_module.rb
+	  (TestModule#test_const_get_invalid_name)
+	  (test_const_defined_invalid_name): Fix expected values.
+
+Tue Apr 23 09:51:26 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c (rb_str_inspect): NUL should not be represented as "\0"
+	  when octal digits may follow.  [ruby-core:54458] [Bug #8290]
+
+Mon Apr 22 22:54:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* insns.def (opt_mod): Use % operator if both operands are positive for
+	  a significant performance improvement. Thanks to @samsaffron.
+
+Mon Apr 22 17:09:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): copy all instance variables not only generic
+	  ivars, before calling post proc.  [ruby-core:51163] [Bug #7627]
+
+Mon Apr 22 10:25:21 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* util.c (ruby_hdtoa): revert r29729.
+	  If you want ruby to behave as before on x86, specify to use SSE like
+	  -msse2 -mfpmath=sse for gcc.
+
+Sun Apr 21 23:19:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* configure.in: Revert using sigsetjmp by default due to performance
+	  problems on some systems (eg. older Linux)
+
+Sun Apr 21 21:35:00 2013  Charlie Somerville  <charlie@charliesomerville.com>
+
+	* configure.in: Use sigsetjmp by default so jumping out of signal
+	  handlers properly restores the signal mask and SS_ONSTACK flag.
+	  [ruby-core:54175] [Bug #8254]
+
+	* configure.in: Manually check for presence of sigsetjmp. It is not a
+	  function on some systems, so AC_CHECK_FUNCS cannot be used.
+
+Sun Apr 21 08:00:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/csv/test_features.rb, test/logger/test_logger.rb
+	  test/mkmf/test_have_macro.rb, test/net/http/test_http.rb,
+	  test/openssl/test_config.rb, test/psych/test_encoding.rb,
+	  test/psych/test_exception.rb, test/psych/test_psych.rb,
+	  test/psych/test_tainted.rb, test/readline/test_readline.rb,
+	  test/rexml/test_contrib.rb, test/ruby/test_autoload.rb,
+	  test/ruby/test_beginendblock.rb, test/ruby/test_exception.rb,
+	  test/ruby/test_file.rb, test/ruby/test_io.rb,
+	  test/ruby/test_marshal.rb, test/ruby/test_process.rb,
+	  test/ruby/test_require.rb, test/ruby/test_rubyoptions.rb,
+	  test/syslog/test_syslog_logger.rb, test/webrick/test_httpauth.rb,
+	  test/zlib/test_zlib.rb: Use Tempfile.create.
+
+Sun Apr 21 00:15:36 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tempfile.rb (Tempfile.create): Close when the block exits.
+
+Sat Apr 20 23:38:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/httpauth/htpasswd.rb: Use Tempfile.create to avoid
+	  unintentional unlink() by the finalizer.
+	  lib/webrick/httpauth/htdigest.rb: Ditto.
+
+Sat Apr 20 22:47:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tempfile.rb (Tempfile.create): New method.
+	  The method name is proposed by Shugo Maeda.  [ruby-dev:47220]
+	  [ruby-core:41478] [Feature #5707]
+
+Sat Apr 20 14:22:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): dump no ivars to the original by marshal_dump.
+	  [ruby-core:54334] [Bug #8276]
+
+	* marshal.c (r_object0): copy all ivars of marshal_dump data to the
+	  result object instead.  [ruby-core:51163] [Bug #7627]
+
+Sat Apr 20 02:33:27 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_scrub): add ruby method String#scrub which verify and
+	  fix invalid byte sequence. [ruby-dev:45975] [Feature #6752]
+
+	* string.c (str_compat_and_valid): check given string is compatible
+	  and valid with given encoding.
+
+	* transcode.c (str_transcode0): If invalid: :replace is specified for
+	  String#encode, replace invalid byte sequence even if the destination
+	  encoding equals to the source encoding.
+
+Fri Apr 19 21:55:40 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* README.EXT.ja (Data_Wrap_Struct): Remove a description about
+	  orphan argument. Oh, I renamed the argument name without
+	  changing description at r36180... Sorry....
+	  Patch by Makoto Kishimoto. Thanks!!! [ruby-dev:47269] [Bug #8292]
+	* README.EXT.ja (Data_Make_Struct): Add a sample code that describes
+	  how it works.
+	  Patch by Makoto Kishimoto. Thanks!!! [ruby-dev:47269] [Bug #8292]
+
+Fri Apr 19 17:54:57 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (body_type_msg): should accept
+	  message/delivery-status with extra data.
+	  [ruby-core:53741] [Bug #8167]
+
+	* test/net/imap/test_imap_response_parser.rb: related test.
+
+Fri Apr 19 13:03:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): do not dump encoding which is dumped with
+	  marshal_dump data.  [ruby-core:54334] [Bug #8276]
+
+Fri Apr 19 11:36:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (stack_protector): control use of -fstack-protector.
+
+	* configure.in (debugflags): let -fstack-protector precede and disable
+	  debugflags, because they can't work together on SmartOS.  [Bug #8268]
+
+Fri Apr 19 07:43:52 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/openssl/test_cipher.rb: Correct a typo
+	  by jgls <joerg@joergleis.com>
+	  https://github.com/ruby/ruby/pull/291 fix GH-291
+
+Thu Apr 18 16:58:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_mod_public_method): fix visibility on anonymous
+	  module. set visibility of singleton method, not method in base
+	  class.  [ruby-core:54404] [Bug #8284]
+
+Thu Apr 18 16:20:51 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): should skip dot directories only for recursion,
+	  but should not if matching to the given pattern.  [ruby-core:54387]
+	  [Bug #8283]
+
+Thu Apr 18 16:20:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_unpack): increase buffer size to fix buffer overflow,
+	  and fix garbage just after unpacking without missing paddings.
+	  [Bug #8286]
+
+Thu Apr 18 13:35:54 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_unpack): output characters even if the input doesn't
+	  have paddings. [Bug #8286]
+
+Thu Apr 18 08:20:48 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (clean-ext): remove timestamps.
+
+Wed Apr 17 22:07:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (SOCKLEN_MAX): Expression simplified.
+
+Wed Apr 17 20:09:19 2013  Aman Gupta  <ruby@tmm1.net>
+
+	* compile.c (iseq_add_mark_object): Use new rb_iseq_add_mark_object().
+
+	* insns.def (setinlinecache): Ditto.
+
+	* iseq.c (rb_iseq_add_mark_object): New function to allocate
+	  iseq->mark_ary on demand. [Bug #8142]
+
+	* iseq.h (rb_iseq_add_mark_object): Ditto.
+
+	* iseq.c (prepare_iseq_build): Avoid allocating mark_ary until needed.
+
+	* iseq.c (rb_iseq_build_for_ruby2cext): Ditto.
+
+Wed Apr 17 20:00:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (SOCKLEN_MAX): Defined.
+
+	* ext/socket/raddrinfo.c (ext/socket/raddrinfo.c): Reject too long
+	  Linux abstract socket name.
+
+Wed Apr 17 19:45:27 2013  Aman Gupta  <tmm1@ruby-lang.org>
+
+	* iseq.c (iseq_location_setup): re-use existing string when iseq has
+	  the same path and absolute_path. [Bug #8149]
+
+Wed Apr 17 11:38:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/test/unit/assertions.rb (Test::Unit::Assertions#assert):
+	  UNASSIGNED is not a valid message.
+
+Wed Apr 17 10:58:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (sleep_timeval): get rid of overflow on Windows where
+	  timeval.tv_sec is not time_t but mere long.
+
+Tue Apr 16 23:07:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/unixsocket.c (unix_send_io): Suppress a warning by clang.
+	  (unix_recv_io): Ditto.
+
+Tue Apr 16 12:27:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/sdbm/init.c: Fix comment indentation, by windwiny [Fixes GH-277]
+
+Tue Apr 16 12:25:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/socket/option.c: Document synonymous methods, by windwiny [GH-277]
+	* ext/stringio/stringio.c: ditto
+	* ext/io/wait/wait.c: ditto
+	* ext/gdbm/gdbm.c: ditto
+	* ext/dl/cfunc.c: ditto
+	* ext/zlib/zlib.c: ditto
+	* ext/win32ole/win32ole.c: ditto
+	* ext/dbm/dbm.c: ditto
+	* ext/json/generator/generator.c: ditto
+	* ext/date/date_core.c: ditto
+
+Tue Apr 16 11:23:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/openssl/*: Document synonymous methods, by windwiny [GH-277]
+
+Mon Apr 15 22:21:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/fiddle/depend: New file.
+
+Mon Apr 15 22:01:02 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-insert): Check
+	  ruby-electric-is-last-command-char-expandable-punct-p here.
+
+	* misc/ruby-electric.el (ruby-electric-closing-char): New
+	  interactive function bound to closing characters.  Typing one of
+	  those closing characters right after the matching counterpart
+	  cancels the effect of automatic closing.  For example, typing
+	  "{" followed by "}" simply makes "{}" instead of "{ } }".
+
+Mon Apr 15 12:54:42 2013  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* ext/openssl/ossl_ssl.c: Correct shutdown behavior w.r.t GC.
+
+	* test/openssl/test_ssl.rb: Add tests to verify correct behavior.
+
+	[Bug #8240] Patch provided by Shugo Maeda. Thanks!
+
+Mon Apr 15 10:23:39 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/coverage/depend: fix id.h place as r40283.
+
+	* ext/coverage/extconf.rb: add topdir and topsrcdir to VPATH.
+
+Sun Apr 14 19:46:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/debug/depend: New file.
+
+	* ext/-test-/exception/depend: Ditto.
+
+	* ext/-test-/printf/depend: Ditto.
+
+	* ext/-test-/string/depend: Ditto.
+
+	* ext/coverage/depend: Ditto.
+
+	* ext/io/console/depend: Ditto.
+
+	* ext/io/nonblock/depend: Ditto.
+
+	* ext/io/wait/depend: Ditto.
+
+	* ext/openssl/depend: Ditto.
+
+	* ext/pathname/depend: Ditto.
+
+	* ext/psych/depend: Ditto.
+
+	* ext/zlib/depend: Ditto.
+
+Sun Apr 14 02:46:50 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#create_makefile): remove {$(VPATH)} other
+	  than nmake.
+
+	* ext/ripper/depend: use VPATH expecting removed by above.
+
+Sat Apr 13 23:06:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (timestamp_file): gather timestamp files in one
+	  directory from each extension directories.
+
+Sat Apr 13 21:09:02 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#create_makefile): output new macro
+	  disthdrdir to specify the path of id.h, parse.h and etc.
+
+	* ext/ripper/depend: use above macro.
+
+Sat Apr 13 20:28:08 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Merge Onigmo 5.13.4 f22cf2e566712cace60d17f84d63119d7c5764ee.
+	  [bug] fix problem with optimization of \z (Issue #16) [Bug #8210]
+
+Sat Apr 13 18:56:15 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/ripper/depend: parse.h and id.h may be created on topdir.
+
+Sat Apr 13 12:08:16 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Vector#cross_product, patch by Luis Ezcurdia
+	  [fix GH-276] [rubyspec:81eec89a124]
+
+Sat Apr 13 10:20:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (rb_struct_define_without_accessor, rb_struct_define),
+	  (rb_struct_s_def): hide member names array.
+
+	* struct.c (anonymous_struct, new_struct, setup_struct): split
+	  make_struct() for each purpose.
+
+Sat Apr 13 09:34:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/mkmf.rb: Add ruby/ruby.h, ruby/missing.h, ruby/intern.h,
+	  ruby/st.h and ruby/subst.h for ruby_headers in generated Makefile.
+
+	* ext/-test-/old_thread_select/depend: Update dependencies.
+
+	* ext/-test-/wait_for_single_fd/depend: Ditto.
+
+	* ext/bigdecimal/depend: Ditto.
+
+	* ext/curses/depend: Ditto.
+
+	* ext/digest/bubblebabble/depend: Ditto.
+
+	* ext/digest/depend: Ditto.
+
+	* ext/digest/md5/depend: Ditto.
+
+	* ext/digest/rmd160/depend: Ditto.
+
+	* ext/digest/sha1/depend: Ditto.
+
+	* ext/digest/sha2/depend: Ditto.
+
+	* ext/dl/callback/depend: Ditto.
+
+	* ext/dl/depend: Ditto.
+
+	* ext/etc/depend: Ditto.
+
+	* ext/nkf/depend: Ditto.
+
+	* ext/objspace/depend: Ditto.
+
+	* ext/pty/depend: Ditto.
+
+	* ext/readline/depend: Ditto.
+
+	* ext/ripper/depend: Ditto.
+
+	* ext/sdbm/depend: Ditto.
+
+	* ext/socket/depend: Ditto.
+
+	* ext/stringio/depend: Ditto.
+
+	* ext/strscan/depend: Ditto.
+
+	* ext/syslog/depend: Ditto.
+
+	* ext/-test-/num2int/depend: Removed.
+
+	* ext/dbm/depend: Ditto.
+
+	* ext/fcntl/depend: Ditto.
+
+	* ext/gdbm/depend: Ditto.
+
+	* ext/racc/cparse/depend: Ditto.
+
+Sat Apr 13 00:15:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (Init_etc): move Passwd and Group under Etc namespace
+	  as primary names.
+
+Fri Apr 12 21:06:55 2013  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: pack.o depends on internal.h.
+
+Fri Apr 12 20:59:24 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (ones): Use __builtin_popcountl if available.
+
+	* internal.h (GCC_VERSION_SINCE): Macro moved from pack.c.
+
+	* pack.c: Include internal.h for GCC_VERSION_SINCE.
+
+Fri Apr 12 18:29:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: version.o depends on $(srcdir)/include/ruby/version.h
+	  instead of {$(VPATH)}version.h to avoid confusion by VPATH between
+	  top level version.h and include/ruby/version.h for build in-place.
+	  [ruby-dev:47249] [Bug #8256]
+
+Fri Apr 12 15:21:24 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_keyword_arg): non-symbol key is not
+	  a keyword argument, keep it as a positional argument.
+
+Fri Apr 12 11:58:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c: Document synonymous methods, by windwiny [GH-277]
+	* bignum.c: ditto
+	* complex.c: ditto
+	* dir.c: ditto
+	* encoding.c: ditto
+	* enumerator.c: ditto
+	* numeric.c: ditto
+	* proc.c: ditto
+	* re.c: ditto
+	* string.c: ditto
+
+Thu Apr 11 23:41:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: Add dependencies for include/ruby.h
+
+	* tool/update-deps: Use "make -p all miniruby ruby golf" to extract
+	  dependencies in makefiles.
+
+Thu Apr 11 23:21:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: Use "make -p all golf" to extract dependencies in
+	  makefiles.
+
+Thu Apr 11 21:02:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: Dependency updated.
+
+	* tool/update-deps: Rewritten.
+
+Thu Apr 11 19:59:48 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: partially revert r40183, which breaks building on
+	  other than source directory. (its commit log also says the same
+	  thing, but such failure is not reproducible on my environment
+	  and the commit breaks build on my environment)
+
+Thu Apr 11 16:10:01 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/fiddle/closure.c (USE_FFI_CLOSURE_ALLOC): define 0 on
+	  Mac OS X and Linux [Bug #3371]
+
+Thu Apr 11 13:19:22 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/drb/drbtest.rb (Drb{Core,Ary}#teardown): retry Process.kill
+	  if it fails with Errno::EPERM on Windows (workaround).
+	  [ruby-dev:47245] [Bug #8251]
+
+Thu Apr 11 11:11:38 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* dir.c: Fix a typo.
+
+Thu Apr 11 10:39:34 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/fiddle/closure.c (USE_FFI_CLOSURE_ALLOC): add missing case:
+	  RUBY_LIBFFI_MODVERSION is not defined (usually on Windows).
+
+Thu Apr 11 09:27:04 2013  Konstantin Haase  <me@rkh.im>
+
+	* dir.c (file_s_fnmatch): Document File::FNM_EXTGLOB flag.
+
+Thu Apr 11 09:17:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* README: Fix typo by Benjamin Winkler [Fixes GH-281]
+
+Thu Apr 11 06:15:51 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regint.h: fix typo: _M_AMD86 -> _M_AMD64.
+
+	* siphash.c: ditto.
+
+	* st.c: ditto.
+
+Thu Apr 11 06:09:57 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/fiddle/extconf.rb: define RUBY_LIBFFI_MODVERSION macro.
+
+	* ext/fiddle/closure.c (USE_FFI_CLOSURE_ALLOC): define 0 or 1
+	  with platform and libffi's version. [Bug #3371]
+
+Thu Apr 11 05:30:43 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/mkmf.rb (pkg_config): Add optional argument "option".
+	  If it is given, it returns the result of
+	  `pkg-config --<option> <pkgname>`.
+
+Thu Apr 11 03:33:05 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/fiddle/closure.c (initialize): check mprotect's return value.
+	  If mprotect is failed because of PaX or something, its function call
+	  will cause SEGV.
+	  http://c5664.rubyci.org/~chkbuild/ruby-trunk/log/20130401T210301Z.diff.html.gz
+
+Wed Apr 10 17:39:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/bigdecimal/bigdecimal.c (VpCtoV): Initialize a local variable
+	  even when overflow.
+
+Wed Apr 10 12:32:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_ll2big): Don't overflow on signed integer negation.
+
+	* ext/bigdecimal/bigdecimal.c (MUL_OVERFLOW_SIGNED_VALUE_P): New
+	  macro.
+	  (AddExponent): Don't overflow on signed integer multiplication.
+	  (VpCtoV): Don't overflow on signed integer arithmetic.
+	  (VpCtoV): Don't overflow on signed integer arithmetic.
+
+Wed Apr 10 06:32:12 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (MUL_OVERFLOW_INT_P): New macro.
+
+	* sprintf.c (GETNUM): Don't overflow on signed integer multiplication.
+
+Tue Apr  9 20:38:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (MUL_OVERFLOW_SIGNED_INTEGER_P): New macro.
+	  (MUL_OVERFLOW_FIXNUM_P): Ditto.
+	  (MUL_OVERFLOW_LONG_P): Ditto.
+
+	* array.c (rb_ary_product): Don't overflow on signed integer
+	  multiplication.
+
+	* numeric.c (fix_mul): Ditto.
+	  (int_pow): Ditto.
+
+	* rational.c (f_imul): Ditto.
+
+	* insns.def (opt_mult): Ditto.
+
+	* thread.c (sleep_timeval): Don't overflow on signed integer addition.
+
+	* bignum.c (rb_int2big): Don't overflow on signed integer negation.
+	  (rb_big2ulong): Ditto.
+	  (rb_big2long): Ditto.
+	  (rb_big2ull): Ditto.
+	  (rb_big2ll): Ditto.
+
+Tue Apr  9 19:45:44 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: Support multiple fields with same field
+	  name (like Set-Cookie).
+	  (OpenURI::Meta#metas): New accessor to obtain fields as a Hash from
+	  field name (string) to field values (array of strings).
+	  [ruby-core:37734] [Bug #4964] reported by ren li.
+
+Tue Apr  9 15:26:12 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): append keyword hash to argument array
+	  to splat if needed.  [ruby-core:54094] [Bug #8236]
+
+Tue Apr  9 10:02:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (timestamp_file): gather timestamp files in one
+	  directory from each extension directories, with considering
+	  target_prefix.
+
+Tue Apr  9 04:57:59 JST 2013  Charles Oliver Nutter  <headius@headius.com>
+
+	* error.c: Capture EAGAIN, EWOULDBLOCK, EINPROGRESS exceptions and
+	  export them for use in WaitReadable/Writable exceptions.
+	* io.c: Create versions of EAGAIN, EWOULDBLOCK, EINPROGRESS that
+	  include WaitReadable and WaitWritable. Add rb_readwrite_sys_fail
+	  for nonblocking failures using those exceptions. Use that
+	  function in io_getpartial and io_write_nonblock instead of
+	  rb_mod_sys_fail
+	* ext/openssl/ossl_ssl.c: Add new SSLError subclasses that include
+	  WaitReadable and WaitWritable. Use those classes for
+	  write_would_block and read_would_block instead of rb_mod_sys_fail.
+	* ext/socket/ancdata.c: Use rb_readwrite_sys_fail instead of
+	  rb_mod_sys_fail in bsock_sendmsg_internal and
+	  bsock_recvmsg_internal.
+	* ext/socket/init.c: Use rb_readwrite_sys_fail instead of
+	  rb_mod_sys_fail in rsock_s_recvfrom_nonblock and
+	  rsock_s_connect_nonblock.
+	* ext/socket/socket.c: Use rb_readwrite_sys_fail instead of
+	  rb_mod_sys_fail in sock_connect_nonblock.
+	* include/ruby/ruby.h: Export rb_readwrite_sys_fail for use instead
+	  of rb_mod_sys_fail. Introduce new constants RB_IO_WAIT_READABLE and
+	  RB_IO_WAIT_WRITABLE for first arg to rb_readwrite_sys_fail.
+
+Tue Apr  9 02:44:32 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/extconf.rb: $defs needs -D or -U.  nothing is added
+	  otherwize.
+
+	* ext/socket/extconf.rb: check struct in_addr6, which is defined in
+	  VC6 instead of in6_addr.
+
+	* ext/socket/option.c (optname_to_sym): fix macro name.
+
+	* ext/socket/constants.c (rsock_cmsg_type_arg): fix macro name.
+
+Mon Apr  8 23:57:21 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (id_for_setter): extract common code from const, class
+	  variable, instance variable setters.
+
+Mon Apr  8 23:55:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/depend (ENCOBJS, TRANSOBJS): use explicit path to ruby.h for
+	  nmake.
+
+	* ext/depend (ENCOBJS, TRANSOBJS): fix header dependency, VPATH has
+	  $(srcdir)/include/ruby but not $(srcdir)/include, so cannot find out
+	  ruby/ruby.h.  use ruby.h instead and ../ruby for include/ruby.h.
+
+Mon Apr  8 20:30:37 2013  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* ext/depend (ENCOBJS, TRANSOBJS): Add missing dependencies.
+
+Mon Apr  8 17:19:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (fole_missing): should check actual argument
+	  count before accessing.
+
+Mon Apr  8 16:03:55 2013  Yuki Yugui Sonoda  <yugui@google.com>
+
+	Fixes a build failure of ext/ripper/ripper.c on building out of place.
+	* common.mk (id.h, id.c): Always generated in $(srcdir).
+	(ext/ripper/ripper.c): Passes $(PATH_SEPARATOR) too to the sub make.
+
+Mon Apr  8 12:05:02 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* object.c (rb_obj_ivar_set): call to_str for string only once.
+	  to_str was called from rb_is_const_name and rb_to_id before.
+
+	* object.c (rb_mod_const_set): ditto.
+
+	* object.c (rb_mod_cvar_set): ditto.
+
+Sun Apr  7 13:56:16 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_require.rb (TestRequire#test_require_nonascii_path):
+	  RUBY_PLATFORM should escape as Regexp,
+	  because RUBY_PLATFORM may contain '.'.
+
+Sun Apr  7 10:44:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h: Simplify the logic to include sys/select.h.
+	  This fixes a compilation error on Haiku (gcc2 and gcc4).
+
+	* configure.in: Use shared linker as $(CC) for Haiku.
+	  This fixes a build error on Haiku (gcc2).
+
+Sun Apr  7 10:41:30 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (MDNSOneShot#sender): Delete an unused variable.
+
+Sun Apr  7 03:24:36 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: use more generic type:
+	  * u_char -> unsigned char
+	  * u_short -> unsigned short
+	  * u_int -> unsigned int
+	  * u_long -> unsigned long
+	  * quad_t -> int64_t
+	  * u_quad_t -> uint64_t
+
+	* addr2line.c (imax): inline is defined by configure.
+
+Sun Apr  7 01:40:39 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-hash): New electric
+	  function that expands a hash sign inside a string or regexp to
+	  "#{}".
+
+	* misc/ruby-electric.el (ruby-electric-curlies): Do not insert
+	  spaces inside when the curly brace is a delimiter of %r, %w,
+	  etc.
+
+	* misc/ruby-electric.el (ruby-electric-curlies): Insert another
+	  space before a closing curly brace when
+	  ruby-electric-newline-before-closing-bracket is nil.
+
+Sun Apr  7 01:01:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* strftime.c (rb_strftime_with_timespec): Test yday range.
+	  [ruby-core:44088] [Bug #6247] reported by Ruby Submit.
+
+Sat Apr  6 23:46:54 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in (AC_CHECK_HEADERS): atomic.h for Solaris atomic_ops.
+
+	* ruby_atomic.h: Skip using Solaris10 atomic_ops on Solaris 9 or
+	  earlier if atomic.h is not available. [ruby-dev:47229] [Bug #8228]
+
+Sat Apr  6 23:40:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: Support LOC resources.
+	  [ruby-core:23361] [Feature #1436] by JB Smith.
+
+Sat Apr  6 23:38:09 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* addr2line.c: quad_t and u_quad_t is not available on Solaris.
+	  __inline is not available with old compilers on Solaris.
+	  [ruby-dev:47229] [Bug #8227]
+
+Sat Apr  6 23:31:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: Add one-shot multicast DNS support.
+	  [ruby-core:53387] [Feature #8089] by Eric Hodel.
+
+Sat Apr  6 22:12:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS.fetch_resource): New method to obtain
+	  full result.
+	  [ruby-dev:43587] [Feature #4788] proposed by Makoto Kishimoto.
+
+Sat Apr  6 20:17:51 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_sys_fail_raddrinfo): Renamed from
+	  rsock_sys_fail_addrinfo.
+	  (rsock_sys_fail_raddrinfo_or_sockaddr): Renamed from
+	  rsock_sys_fail_addrinfo_or_sockaddr.
+
+	* ext/socket/rubysocket.h: Follow the above change.
+
+Sat Apr  6 19:24:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_sys_fail_sockaddr): Takes struct sockaddr
+	  and socklen_t instead of String object.
+	  (rsock_sys_fail_addrinfo_or_sockaddr): Follow the above change.
+
+	* ext/socket/rubysocket.h (rsock_sys_fail_sockaddr): Follow the above
+	  change.
+
+Sat Apr  6 14:28:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (SockAddrStringValueWithAddrinfo): New macro.
+	  (rsock_sockaddr_string_value_with_addrinfo): New declaration.
+	  (rsock_addrinfo_inspect_sockaddr): Ditto.
+	  (rsock_sys_fail_addrinfo): Ditto.
+	  (rsock_sys_fail_sockaddr_or_addrinfo): Ditto.
+
+	* ext/socket/raddrinfo.c (rsock_addrinfo_inspect_sockaddr): Renamed
+	  from addrinfo_inspect_sockaddr and exported.
+	  (rsock_sockaddr_string_value_with_addrinfo): New function to obtain
+	  string and possibly addrinfo object.
+
+	* ext/socket/socket.c (rsock_sys_fail_sockaddr): Don't use
+	  rsock_sys_fail_host_port which is IP dependent.  Invoke
+	  rsock_sys_fail_addrinfo.
+	  (rsock_sys_fail_addrinfo): New function using
+	  rsock_addrinfo_inspect_sockaddr.
+	  (rsock_sys_fail_addrinfo_or_sockaddr): New function.
+	  (sock_connect): Use SockAddrStringValueWithAddrinfo and
+	  rsock_sys_fail_addrinfo_or_sockaddr.
+	  (sock_connect_nonblock): Ditto.
+	  (sock_bind): Ditto.
+
+Sat Apr  6 13:34:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_sys_fail_sockaddr): Delete 2nd argument.
+
+	* ext/socket/rubysocket.h (rsock_sys_fail_sockaddr): Follow above
+	  change.
+
+Sat Apr  6 13:13:39 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_sys_fail_path): Use rb_str_inspect only
+	  for String to avoid SEGV.
+
+Sat Apr  6 12:40:16 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (rsock_sys_fail_host_port): Wrap by NORETURN.
+	  (rsock_sys_fail_path): Ditto.
+	  (rsock_sys_fail_sockaddr): Ditto.
+
+Sat Apr  6 11:49:35 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (rsock_sys_fail_path): Use rb_str_inspect if the
+	  path contains a NUL.
+
+Sat Apr  6 11:39:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: Improve socket exception message to show socket address.
+	  [ruby-core:45617] [Feature #6583] proposed Eric Hodel.
+
+	* ext/socket/rubysocket.h (rsock_sys_fail_host_port): Declared.
+	  (rsock_sys_fail_path): Ditto.
+	  (rsock_sys_fail_sockaddr): Ditto.
+
+	* ext/socket/udpsocket.c (udp_connect): Use rsock_sys_fail_host_port.
+	  (udp_bind): Ditto.
+	  (udp_send): Ditto.
+
+	* ext/socket/init.c (rsock_init_sock): Specify a string for rb_sys_fail
+	  argument.
+	  (make_fd_nonblock): Ditto.
+	  (rsock_s_accept): Ditto.
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): Use
+	  rsock_sys_fail_host_port.
+
+	* ext/socket/socket.c (rsock_sys_fail_host_port): Defined.
+	  (rsock_sys_fail_path): Ditto.
+	  (rsock_sys_fail_sockaddr): Ditto.
+	  (setup_domain_and_type): Use rsock_sys_fail_sockaddr.
+	  (sock_connect_nonblock): Ditto.
+	  (sock_bind): Ditto.
+	  (sock_gethostname): Specify a string for rb_sys_fail argument.
+	  (socket_s_ip_address_list): Ditto.
+
+	* ext/socket/basicsocket.c (bsock_shutdown): Specify a string for
+	  rb_sys_fail argument.
+	  (bsock_setsockopt): Use rsock_sys_fail_path.
+	  (bsock_getsockopt): Ditto.
+	  (bsock_getpeereid): Refine the argument for rb_sys_fail.
+
+	* ext/socket/unixsocket.c (rsock_init_unixsock): Use
+	  rsock_sys_fail_path.
+	  (unix_path): Ditto.
+	  (unix_send_io): Ditto.
+	  (unix_recv_io): Ditto.
+	  (unix_addr): Ditto.
+	  (unix_peeraddr): Ditto.
+
+Sat Apr  6 11:23:18 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_require.rb (TestRequire#test_require_nonascii_path):
+	  fix load path for encoding to run the test as stand-alone.
+
+Sat Apr  6 09:54:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (NATINT_LEN): fix definition order, must be after
+	  NATINT_PACK.
+
+Sat Apr  6 03:11:07 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: fix symbol keys in coder
+	  emission.  Thanks @tjwallace
+	* test/psych/test_coder.rb: test for change
+
+Sat Apr  6 02:54:08 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/exception.rb: there should be only one exception
+	  base class. Fixes tenderlove/psych #125
+	* ext/psych/lib/psych.rb: require the correct exception class
+	* ext/psych/lib/psych/syntax_error.rb: ditto
+	* ext/psych/lib/psych/visitors/to_ruby.rb: ditto
+
+Sat Apr  6 02:30:28 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (new_defined): remove all extra parentheses, and return
+	  "nil" for defined? with empty expression.
+	  [ruby-core:54024] [Bug #8224]
+
+Sat Apr  6 02:06:04 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: correctly register
+	  self-referential strings. Fixes tenderlove/psych #135
+
+	* test/psych/test_string.rb: appropriate test.
+
+Sat Apr  6 01:21:56 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/init.c (cloexec_accept): Fix a compile error on
+	  Debian GNU/kFreeBSD.  Consider HAVE_ACCEPT4 is defined
+	  but SOCK_CLOEXEC is not defined.
+
+Sat Apr  6 00:19:30 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* load.c (features_index_add): use rb_str_subseq() to specify C string
+	  position properly to fix require non ascii path.
+	  [ruby-core:53733] [Bug #8165]
+
+	* test/ruby/test_require.rb (TestRequire#test_require_nonascii_path):
+	  a test for the above.
+
+Fri Apr  5 20:41:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h (HAVE_TRUE_LONG_LONG): Defined to distinguish
+	  availability of long long and availability of 64bit integer type.
+
+	* pack.c: Use HAVE_TRUE_LONG_LONG to distinguish q! and Q! support.
+
+Fri Apr  5 20:19:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* addr2line.c: Include ruby/missing.h to fix compile error on Debian.
+
+Fri Apr  5 19:39:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): fix of defined? with empty
+	  expression.  [ruby-core:53999] [Bug #8220]
+
+Fri Apr  5 13:22:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/curses/curses.c (Init_curses): fix implementation function,
+	  crmode should be same as cbreak.  [ruby-core:54013] [Bug #8222]
+
+Fri Apr  5 12:06:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/curses/hello.rb: Typo in Curses example by Drew Blas
+	  [Fixes GH-273]
+
+Thu Apr  4 23:45:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (bind_random_port): Rescue EACCES for SunOS.
+	  bind() on SunOS for port 2049 (nfs) and 4045 (lockd) causes
+	  EACCES with unprivileged process.  cf. PRIV_SYS_NFS in privileges(5)
+	  [ruby-core:48064] [Bug #7183] reported by Frank Meier.
+
+Thu Apr  4 23:24:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Remove condition for bcc.
+
+Thu Apr  4 22:53:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (FIX2LONG): Parenthesize the macro body.
+
+Thu Apr  4 22:32:32 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_strftime): Describe %L and %N truncates digits under
+	  the specified length.
+	  [ruby-core:52130] [Bug #7829]
+
+Thu Apr  4 22:08:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* object.c (rb_mod_cvar_set): Reverted "avoid inadvertent
+	  symbol creation" to avoid SEGV by
+	  Class.new.class_variable_set(1, 2).
+
+Thu Apr  4 20:07:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_write): New method.
+	  (path_binwrite): Ditto.
+	  [ruby-core:49468] [Feature #7378]
+
+Thu Apr  4 16:51:29 2013  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* thread_pthread.c: Fixes wrong scopes of #if USE_SLEEPY_TIMER_THREAD
+	  .. #endif sections.  This fixes a build error on NativeClient.
+
+Wed Apr  3 17:25:31 2013  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* thread_pthread.c (ruby_init_stack): Avoid using uninitialized value.
+	  stackaddr and size are not set if get_stack() fails.
+
+Thu Apr  4 16:55:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (make_struct): avoid inadvertent symbol creation.
+	  (rb_struct_aref): ditto.
+	  (rb_struct_aset): ditto.
+
+Thu Apr  4 16:54:40 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_set): avoid inadvertent symbol creation.
+	  (rb_obj_ivar_set): ditto.
+	  (rb_mod_cvar_set): ditto.
+
+Thu Apr  4 15:46:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_inject): avoid inadvertent symbol creation.
+
+Thu Apr  4 14:37:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_aref): avoid inadvertent symbol creation.
+	  (rb_thread_variable_get): ditto.
+	  (rb_thread_key_p): ditto.
+	  (rb_thread_variable_p): ditto.
+
+Thu Apr  4 11:33:57 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_to_i): Use bn2hex to speed up.
+	  In general, binary to/from decimal needs extra cost.
+
+Thu Apr  4 07:24:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Specify arguments to test functions.
+
+Thu Apr  4 03:25:09 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_initialize): fix can't create from bn.
+
+Wed Apr  3 22:09:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Test functions and libraries after headers.
+
+Wed Apr  3 21:23:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	 * io.c (rb_io_seek_m): Accept :CUR, :END, :SET as "whence" argument.
+	   (interpret_seek_whence): New function.
+	   [ruby-dev:45818] [Feature #6643]
+
+Wed Apr  3 20:52:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: Describe the behavior which Ruby invokes a commandline
+	  directly without shell if the commandline is simple enough.
+	  [ruby-core:50459] [Bug #7489]
+
+Wed Apr  3 20:27:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/extmk.rb (extmake): Invoke Logging::log_close in a ensure
+	  clause.
+
+Wed Apr  3 18:53:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/extmk.rb (extmake): Use Logging.open to switch stdout and
+	  stderr.  Delay Logging::log_close until the failure message is
+	  written.  Write the failure message only if log file is opened.
+
+	* lib/mkmf.rb (Logging.log_opened?): New method.
+
+	  [ruby-dev:47215] [Bug #8209]
+
+Wed Apr  3 17:11:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (constat_apply): pass through unknown sequence which
+	  starts with ESC but is not followed by a bracket.  [ruby-core:53879]
+	  [Bug #8201]
+
+Wed Apr  3 16:35:32 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_eq): hide intermediate Bignums not just freeing
+	  memory.  [ruby-core:53893] [Bug #8204]
+
+	* object.c (rb_obj_hide): hide an object by clearing klass.
+
+	* bignum.c (rb_big_eq): test as Fixnum if possible and get rid of zero
+	  length Bignum.  [ruby-core:53893] [Bug #8204]
+
+Tue Apr  2 23:56:03 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb (SecureRandom.random_bytes): Use
+	  OpenSSL::Random.random_add instead of OpenSSL::Random.seed and
+	  specify 0.0 as the entropy.
+	  [ruby-core:47308] [Bug #6928]
+
+Tue Apr  2 20:24:52 2013  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c: Support Q! and q! for long long.
+	  (natstr): Moved to toplevel.  Add q and Q if there is long long type.
+	  (endstr): Moved to toplevel.
+	  (NATINT_PACK): Consider long long.
+	  (NATINT_LEN_Q): New macro.
+	  (pack_pack): Support Q! and q!.
+	  (pack_unpack): Ditto.
+	  [ruby-dev:43970] [Feature #3946]
+
+Tue Apr  2 19:24:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/num2int/num2int.c: Define utility methods
+	  as module methods of Num2int.
+
+	* test/-ext-/num2int/test_num2int.rb: Follow the above change.
+
+Tue Apr  2 18:49:01 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb: Don't use Array#to_s.
+	   [ruby-core:52058] [Bug #7811] fixed by zzak (Zachary Scott).
+
+Tue Apr  2 17:38:20 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_to_s): suppress duplicated charclass warning.
+	  Regexp#to_s suppress extra its whole regexp options by calling
+	  onig_new with its source, but it doesn't call rb_reg_preprocess.
+	  Therefore its Unicode escapes (\u{XXXX}) are given as is,
+	  and it may cause duplicated charclass warning for example
+	  "[\u{33}]" (3 is duplicated) or "[\u{a}\u{b}]" (u is duplicated).
+	  [ruby-core:53649] [Bug #8151]
+
+Tue Apr  2 16:00:06 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c (rb_print_backtrace): separate to ease showing C backtrace.
+
+	* internal.h (rb_print_backtrace): ditto.
+
+Tue Apr  2 15:22:09 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/envutil.rb (assert_separately): stop_auto_run of
+	  Test::Unit::Runner to prevent auto runner use ARGV.
+
+	* test/ruby/envutil.rb (assert_separately): add $: to separate process.
+
+	* test/ruby/envutil.rb (assert_separately): fail if stderr is not
+	  empty and ignore_stderr is false.
+
+Tue Apr  2 06:46:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/num2int/num2int.c: Rename utility methods
+	  to global functions to ease manual experiments.
+
+	* test/-ext-/num2int/test_num2int.rb: Follow the above change.
+
+Mon Apr  1 22:26:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/zlib/zlib.c (rb_gzfile_set_mtime): Use NUM2UINT.
+	  The old logic doesn't work well on LP64 platforms as:
+	  .. -2**63-1 => error,
+	  -2**63 .. -2**62-1 => success,
+	  -2**62 .. -2**31-1 => error,
+	  -2**31 .. 2**31-1 => success,
+	  2**31 .. 2**62-1 => error,
+	  2**62 .. 2**64-1 => success,
+	  2**64 ..  => error.
+
+Mon Apr  1 22:08:02 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/zlib/zlib.c (Zlib::Inflate.new):
+	  Fix documentation syntax and naming errors.
+	  Based on patch by Robin Dupret. Fix GH-271.
+
+Mon Apr  1 21:22:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/-ext-/num2int/test_num2int.rb: Test small bignums.
+
+Mon Apr  1 21:10:56 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num2ulong_internal): Don't cast a negative double value
+	  into unsigned long, which is undefined behavior.
+	  (rb_num2ull): Don't cast a value bigger than LLONG_MAX into
+	  long long, which is undefined behavior.
+
+Mon Apr  1 20:57:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/num2int/num2int.c: Return string for result, instead of
+	  printing.
+
+	* test/-ext-/num2int/test_num2int.rb: updated to follow above change.
+
+Mon Apr  1 20:08:07 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num2long): Don't use SIGNED_VALUE uselessly.
+	  (check_int): Ditto.
+	  (check_short): Ditto.
+	  (rb_num2fix): Ditto.
+	  (rb_num2ulong_internal): Add a cast.
+
+Mon Apr  1 18:41:35 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: skip autoconf 2.64 and 2.66, 2.67 seems short-lived
+	  but stick on it for Debian Squeeze.
+
+Mon Apr  1 14:22:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check clang version by predefined macro values.
+	  [Bug #8192]
+
+Mon Apr  1 12:05:15 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (check_uint): Take the 1st argument as unsigned long,
+	  instead of VALUE.  Refine the validity test conditions.
+	  (check_ushort): Ditto.
+
+Mon Apr  1 07:15:03 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* configure.in: use quadrigraph to put '[' or ']'. [Bug #8192]
+
+Mon Apr  1 04:16:41 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: kick old clang. [ruby-dev:47204] [Bug #8192]
+
+Mon Apr  1 01:12:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (FIX2ULONG): Make it consistent with NUM2ULONG.
+
+	* ext/-test-/num2int/num2int.c: Add utility methods for FIX2XXX tests.
+
+	* test/-ext-/num2int/test_num2int.rb: Add tests for FIX2XXX.
+
+Sun Mar 31 17:17:56 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_mod_define_method): consider visibility in define_method.
+	  patch by mashiro <mail AT mashiro.org>.  fix GH-268.
+
+Sun Mar 31 15:40:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/configure.bat: try to fix option arguments split by commas and
+	  equals here.  this batch file no longer run with old command.com.
+
+	* tool/mkconfig.rb: no hacks for cmd.exe.
+
+Sun Mar 31 13:47:04 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num2ulong_internal): New function similar to
+	  rb_num2ulong but integer wrap around flag is also returned.
+	  (rb_num2ulong): Use rb_num2ulong_internal.
+	  (rb_num2uint): Use rb_num2ulong_internal and the wrap around flag is
+	  used instead of negative_int_p(val).
+	  (rb_num2ushort): ditto.
+
+Sun Mar 31 06:27:17 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (HAVE_METACLASS_P): should check FL_SINGLETON flag before get
+	  instance variable to get rid of wrong warning about __attached__.
+	  [ruby-core:53839] [Bug #8188]
+
+Sat Mar 30 14:11:28 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* bcc32: removed. agreed at
+	  http://bugs.ruby-lang.org/projects/ruby/wiki/DevelopersMeeting20130223Japan
+
+Sat Mar 30 03:58:00 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/file.c (code_page): use cp1252 instead of cp20127 as US-ASCII.
+	  fix [ruby-core:53079] [Bug #7996]
+	  reported and patched by mmeltner (Michael Meltner).
+
+Sat Mar 30 03:49:21 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (wrename): use MoveFileExW instead of MoveFileW,
+	  because the latter fails on cross device file move of some
+	  environments.
+	  fix [ruby-core:53492] [Bug #8109]
+	  reported by mitchellh (Mitchell Hashimoto).
+
+Fri Mar 29 22:09:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_mutex_synchronize_m): yield no block params.  patch by
+	  splattael (Peter Suschlik) in [ruby-core:53773] [Bug #8097].
+	  fix GH-266.
+
+Fri Mar 29 16:51:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_next_argv): set init flag if succeeded to forward, after
+	  skipping.
+
+	* io.c (argf_block_call_i, argf_block_call): no more forwarding if
+	  forwarded after skipping.  [ruby-list:49185]
+
+	* io.c (argf_close): deal with init flag.
+
+	* io.c (argf_block_call_i, argf_block_call): forward next file if
+	  skipped while iteration, to get rid of IOError.  [ruby-list:49185]
+
+Fri Mar 29 11:09:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (configuration): not include all CFLAGS in CXXFLAGS, to
+	  use different set than C for C++.  [ruby-core:45273] [Bug #6504]
+
+Fri Mar 29 10:24:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h: undef POSIX compliant names on AIX, which are no
+	  longer needed.  patch suggested by edelsohn (David Edelsohn) in
+	  [ruby-core:53815].  [Bug #8174]
+
+Fri Mar 29 06:39:42 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num2ull): Cast double to unsigned LONG_LONG via
+	  LONG_LONG instead of double to unsigned LONG_LONG directly.
+	  This is a challenge to fix a test_num2ull(TestNum2int)
+	  failure (NUM2ULL(-1.0) should be "18446744073709551615" but was "0")
+	  on Mac OS X with 32bit clang.
+	  http://a.mrkn.jp/~mrkn/chkbuild/mountain_lion/ruby-trunk-m32-o0/log/20130328T191100Z.diff.html.gz
+
+Fri Mar 29 00:54:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MAIN_DOES_NOTHING): ensure symbols for tests to be
+	  preserved.  [ruby-core:53745] [Bug #8169]
+
+Thu Mar 28 23:11:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: Test Windows platform by detecting LoadError when
+	  require 'win32/resolv' suggested by Nobuyoshi Nakada [ruby-core:53389].
+	  [ruby-core:53388] [Feature #8090] Reported by Charles Nutter.
+
+Thu Mar 28 23:10:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h: rename SVR3,4 member names as POSIX compliant,
+	  to get rid of conflict on AIX.  [ruby-core:53765] [Bug #8174]
+
+Thu Mar 28 18:22:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/-ext-/num2int/test_num2int.rb: extract
+	  assert_num2i_success_internal and assert_num2i_error_internal and
+	  provide assertion messages as "NUM2XXX(NNN)".
+
+Thu Mar 28 07:05:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h: Delete redundant inclusions caused by
+	  AC_INCLUDES_DEFAULT in defines.h.
+
+	* include/ruby/defines.h: Ditto.
+
+	* include/ruby/ruby.h: Ditto.
+
+	* include/ruby/st.h: Ditto.
+
+Thu Mar 28 06:51:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/defines.h: Fix a compilation error on NetBSD,
+	  "type of formal parameter 1 is incomplete" for the rb_thread_wait_for
+	  invocation in rb_file_flock, by including header files as
+	  AC_INCLUDES_DEFAULT of autoconf.
+
+Wed Mar 27 22:09:14 2013  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (LONG_MIN_MINUS_ONE_IS_LESS_THAN): New macro.
+	  (LLONG_MIN_MINUS_ONE_IS_LESS_THAN): Ditto.
+	  (rb_num2long): Use LONG_MIN_MINUS_ONE_IS_LESS_THAN.
+	  (rb_num2ulong): Ditto.
+	  (rb_num2ll): Use LLONG_MIN_MINUS_ONE_IS_LESS_THAN.
+	  (rb_num2ull): Ditto.
+
+	* test/-ext-/num2int/test_num2int.rb (assert_num2i_success): Test the
+	  value converted into a Float if Float can represent the value
+	  exactly.
+	  (assert_num2i_error): Ditto.
+
+Wed Mar 27 20:59:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/-ext-/num2int/test_num2int.rb (assert_num2i_success): New
+	  utility method.
+	  (assert_num2i_error): Ditto.
+
+Wed Mar 27 20:37:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (num_exact): Use to_r method only if to_int method is
+	  available.
+	  [ruby-core:53764] [Bug #8173] Reported by Hiro Asari.
+
+Wed Mar 27 12:07:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/-ext-/num2int/test_num2int.rb (test_num2ll): test LLONG_MIN,
+	  not LONG_MIN.
+
+Wed Mar 27 12:02:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (TIMET_MAX_PLUS_ONE): definition simplified.
+
+Wed Mar 27 06:39:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MAIN_DOES_NOTHING): force to refer symbols for tests
+	  to be preserved.  [ruby-core:53745] [Bug #8169]
+
+Wed Mar 27 05:15:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_REPLACE_TYPE): define SIGNEDNESS_OF_type same as
+	  check_signedness of mkmf.rb.
+
+	* internal.h (TIMET_MAX, TIMET_MIN, TIMET_MAX_PLUS_ONE): use
+	  SIGNEDNESS_OF_TIME_T.
+
+Wed Mar 27 00:28:45 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (TIMET_MAX_PLUS_ONE): Defined.
+
+	* thread.c (double2timeval): Saturate out-of-range values.
+
+Tue Mar 26 23:41:18 2013  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: Define TIMET_MAX and TIMET_MIN here.
+
+	* time.c: Remove TIMET_MAX and TIMET_MIN definitions.
+
+	* thread.c: Ditto.
+
+	* thread_pthread.c: Remove TIMET_MAX definition.
+
+	* thread_win32.c: Ditto.
+
+Tue Mar 26 22:31:10 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sockaddr_len): return the shortest length for
+	  unknown socket address.
+
+Tue Mar 26 22:14:46 2013  Tanaka Akira  <akr@fsij.org>
+
+	* thread.c (double2timeval): convert the infinity to TIME_MAX to avoid
+	  SEGV by Thread.new {}.join(Float::INFINITY) on
+	  Debian GNU/Linux (amd64).
+
+Mon Mar 25 07:09:20 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rinda/tuplespace.rb:  Only return tuple entry once on move,
+	  either through port or regular return, not both.  This results in a
+	  120% speedup when combined with #8125.  Patch by Joel VanderWerf.
+	  [ruby-trunk - Feature #8119]
+
+Mon Mar 25 06:59:01 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rinda/test_rinda.rb:  Skip IPv6 tests if no IPv6 addresses
+	  exist.  Skip fork-dependent test if fork is not available.
+	  [ruby-trunk - Bug #8159]
+
+Sun Mar 24 10:38:24 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* addr2line.c (putce): suppress unused return value warning.
+
+Mon Mar 25 02:01:03 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* proc.c (bm_free): need to clean up the mark flag of a free and
+	  unlinked method entry. [Bug #8100] [ruby-core:53439]
+
+Sun Mar 24 22:13:51 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_rpartition): revert r39903, and convert byte offset
+	  to char offset; the return value of rb_reg_search is byte offset,
+	  but other than it of rb_str_rpartition expects char offset.
+	  [Bug #8138] [ruby-dev:47183]
+
+Sun Mar 24 18:29:46 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* string.c (rb_str_rpartition): Fix String#rpartition(/re/)
+	  against a multibyte string. [Bug #8138] [ruby-dev:47183]
+
+Sun Mar 24 13:42:24 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (GC_ENABLE_LAZY_SWEEP): new macro to switch lazy sweeping
+	  for debugging. [Feature #8024] [ruby-dev:47135]
+
+Sun Mar 24 12:55:47 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: We have no chance to expand the heap when lazy sweeping is
+	  restricted. So collecting is often invoked if there is not
+	  enough free space in the heap. Try to expand heap when this is
+	  the case.
+
+Sun Mar 24 11:03:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_require.rb: Remove temporally files in the tests.
+
+	* test/ruby/test_rubyoptions.rb: Ditto.
+
+	* test/logger/test_logger.rb: Ditto.
+
+	* test/psych/test_psych.rb: Ditto.
+
+	* test/readline/test_readline.rb: Ditto.
+
+	* test/syslog/test_syslog_logger.rb: Ditto.
+
+	* test/webrick/test_httpauth.rb: Ditto.
+
+	* test/zlib/test_zlib.rb: Ditto.
+
+Sun Mar 24 05:36:29 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rinda/ring.rb:  Added documentation for multicast support.
+
+	* NEWS:  Point to above documentation.
+
+Sun Mar 24 05:32:39 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rinda/test_rinda.rb:  Restore tests commented out while fixing
+	  test slowdown bug before r39895.
+
+Sun Mar 24 05:03:36 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rinda/ring.rb:  Add multicast support to Rinda::RingFinger and
+	  Rinda::RingServer.  [ruby-trunk - Bug #8073]
+	* test/rinda/test_rinda.rb:  Test for the above.
+
+	* NEWS: Update with Rinda multicast support
+
+Sun Mar 24 04:13:27 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rinda/test_rinda.rb:  Fixed test failures in r39890 and r39891
+	  due to stopping DRb service.
+
+Sun Mar 24 03:34:02 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rinda/rinda.rb:  Fixed loss of tuple when remote is alive but the
+	  call stack was unwound.  Patch by Joel VanderWerf.
+	  [ruby-trunk - Bug #8125]
+	* test/rinda/test_rinda.rb:  Test for the above.
+
+Sun Mar 24 02:14:53 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/mkmf/test_have_macro.rb: remove temporally files in the tests.
+
+Sat Mar 23 23:50:04 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (kprintf): added from FreeBSD libstand's printf.
+	  this is consided as async signal safe function.
+
+	* addr2line.c (rb_dump_backtrace_with_lines): use kfprintf.
+	  [Bug #8144] [ruby-core:53632]
+
+Sat Mar 23 23:28:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_divide): Use Qnil and NIL_P
+	  instead of (VALUE)0 as a return value.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_div): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_divremain): ditto.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_remainder): ditto.
+
+Sat Mar 23 17:39:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (check_funcall_respond_to): preserve passed_block, which
+	  is modified in vm_call0_body() via vm_call0(), and caused a bug of
+	  rb_check_funcall() by false negative result of rb_block_given_p().
+	  re-fix [ruby-core:53650] [Bug #8153].
+	  [ruby-core:53653] [Bug #8154]
+
+Fri Mar 22 17:48:34 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/forwardable.rb (Forwardable::FILE_REGEXP): create regexp object
+	  outside sources for eval, to reduce allocations in def_delegators
+	  wrappers.  //o option does not make each regexps shared.  patch by
+	  tmm1 (Aman Gupta) in [ruby-core:53620] [Bug #8143].
+
+Fri Mar 22 17:38:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_feature_p), vm_core.h (rb_vm_struct): turn
+	  loaded_features_index into st_table.  patches by tmm1 (Aman Gupta)
+	  in [ruby-core:53251] and [ruby-core:53274] [Bug #8048]
+
+Fri Mar 22 10:29:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: Fix style.
+
+Fri Mar 22 05:30:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ambiguous_operator): refine warning message, since this
+	  warning is shown after literal too.
+
+Fri Mar 22 04:51:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_keyword_arg): should check required
+	  keyword arguments even if rest hash is defined.  [ruby-core:53608]
+	  [Bug #8139]
+
+Fri Mar 22 01:00:17 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* process.c (rb_execarg_addopt, run_exec_pgroup): use rb_pid_t
+	  instead of pid_t.
+
+	* ext/pty/pty.c (raise_from_check, pty_check): ditto.
+
+Fri Mar 22 00:04:15 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (rb_dump_backtrace_with_lines): output line at once.
+
+Thu Mar 21 23:17:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (ruby_kill): get rid of deadlock on signal 0.
+	  [ruby-dev:47182] [Bug #8137]
+
+Thu Mar 21 22:39:46 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* marshal.c (marshal_dump, marshal_load): workaround for segv on
+	  Intel Solaris compiled with Oracle SolarisStudio 12.3.
+	  Partly revert r38174. [ruby-core:52042] [Bug #7805]
+
+Thu Mar 21 16:48:06 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (simple_re_meta): escape all closing characters, not only
+	  round parenthesis.  [ruby-core:53578] [Bug #8133]
+
+Thu Mar 21 13:50:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (UNINITIALIZED_VAR): suppress warnings by clang 4.2.
+	  [ruby-core:51742] [Bug #7756]
+
+Thu Mar 21 07:34:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/date/date_core.c: Typo in Date::MONTHNAMES by Matt Gauger
+	  [GH fixes #261]
+
+Wed Mar 20 22:53:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (find_library): fix to format message.
+	  [ruby-core:53568] [Bug #8130]
+
+Wed Mar 20 22:52:52 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (install_dirs, with_destdir): prefix with DESTDIR
+	  directories to install only unless bundled extension libraries.
+	  [ruby-core:53502] [Bug #8115]
+
+Wed Mar 20 17:47:53 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/win32ole/test_err_in_callback.rb (TestErrInCallBack#setup):
+	  allow using different root for source and build directories.
+	  this may fixes a minor problem of r39834.
+
+Wed Mar 20 16:40:48 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/ruby/test_signal.rb (test_hup_me): skip if HUP isn't supported.
+	  On Windows this test causes ArgumentError.
+
+Wed Mar 20 16:24:12 2013  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* test/rubygems/test_gem_installer.rb (test_install_extension_flat):
+	  use ruby in build directory in case ruby is not installed.
+	  [ruby-core:53265] [Bug #8058]
+
+Wed Mar 20 15:22:07 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/win32ole/test_err_in_callback.rb (TestErrInCallBack#setup): use
+	  relative path to get rid of "too long commandline" error.
+
+Wed Mar 20 04:27:42 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/rinda/test_rinda.rb: remove unused variables.
+	  patched by Vipul A M <vipulnsward@gmail.com>
+
+Wed Mar 20 04:15:32 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* ext/bigdecimal/bigdecimal.c: fixed typo.
+	  patched by Vipul A M <vipulnsward@gmail.com>
+
+Sat Mar 16 03:40:49 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_signal.rb (test_hup_me): added a few comments.
+
+Sat Mar 16 03:39:38 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (ruby_kill): added a few comments.
+
+Sat Mar 16 03:36:56 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (ruby_kill): release GVL while waiting signal delivered.
+
+Tue Mar 19 19:50:48 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby_kill (internal.h, thread.c): use rb_pid_t instead of pid_t.
+	  this fixes the build failure of mswin introduced at r39819.
+
+Tue Mar 19 17:09:30 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_conv_enc_opts): convert with one converter, instead
+	  of re-creating converters for each buffer expansion.
+
+Tue Mar 19 17:06:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): compose HFS file names from UTF8-MAC.
+	  [ruby-core:48745] [Bug #7267]
+
+Sat Mar 16 01:44:29 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* internal.h: added a declaration of ruby_kill().
+	* thread.c (ruby_kill): helper function of kill().
+
+	* signal.c (rb_f_kill): use ruby_kill() instead of kill().
+	* signal.c (rb_f_kill): call rb_thread_execute_interrupts()
+	  to ensure that make SignalException if sent a signal
+	  to myself. [Bug #7951] [ruby-core:52864]
+
+	* vm_core.h (typedef struct rb_thread_struct): added
+	  th->interrupt_cond.
+	* thread.c (rb_threadptr_interrupt_common): added to
+	  initialization of th->interrupt_cond.
+	* thread.c (thread_create_core): ditto.
+
+	* test/ruby/test_signal.rb (TestSignal#test_hup_me): test for
+	  the above.
+
+Sat Mar 16 00:42:39 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (linux_iocparm_len): enable only exist _IOC_SIZE().
+	  Because musl libc doesn't have it. [Bug #8051] [ruby-core:53229]
+
+Tue Mar 19 10:05:04 2013  Shota Fukumori  <her@sorah.jp>
+
+	* ext/objspace/objspace.c: Fix typo in doc. Patch by Sho Hashimoto.
+	  [Bug #8116] [ruby-dev:47177]
+
+Tue Mar 19 02:13:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* configure.in: set ac_cv_prog_cxx if CXX is supplied.
+
+Tue Mar 19 01:18:00 2013  Kenta Murata  <mrkn@mrkn.jp>
+
+	* configure.in: Fix c++ compiler auto-selection not only for
+	  Darwin 11.x, but also the other versions of Darwin.
+
+Tue Mar 19 00:26:22 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: Improve accuracy of objspace_live_num() and
+	  allocated/freed counters. patched by tmm1(Aman Gupta).
+	  [Bug #8092] [ruby-core:53392]
+
+Mon Mar 18 21:42:48 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: Avoid unnecessary heap growth. patched by tmm1(Aman Gupta).
+	  [Bug #8093] [ruby-core:53393]
+
+Mon Mar 18 17:58:36 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: Fix unlimited memory growth with large values of
+	  RUBY_FREE_MIN. patched by tmm1(Aman Gupta).
+	  [Bug #8095] [ruby-core:53405]
+
+Mon Mar 18 14:46:19 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/win32ole/test_err_in_callback.rb
+	  (TestErrInCallBack#test_err_in_callback): shouldn't create a file in
+	  the top of build directory.
+
+Mon Mar 18 13:29:52 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c (backtrace): on darwin use custom backtrace() to trace
+	  beyond _sigtramp. darwin's backtrace can't trace beyond signal
+	  trampoline with sigaltstack.
+
+	* configure.in: check execinfo.h on darwin.
+
+Mon Mar 18 11:03:23 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_exec.h (END_INSN): revert r39517 because the segv seems fixed by
+	  r39806.
+
+Mon Mar 18 10:41:06 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_exec.c: Correct predefined macro name. This typo is introduced by
+	  r36534 and should be backported to ruby_2_0_0.
+
+Mon Mar 18 03:18:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c: Typo in Array#delete by Timo Sand [GH fixes #258]
+
+Mon Mar 18 01:14:56 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (io_fillbuf): show fd number on failure to debug.
+	  http://c5632.rubyci.org/~chkbuild/ruby-trunk/log/20130316T050302Z.diff.html.gz
+
+Sun Mar 17 02:38:21 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/date/date_core.c: include sys/time.h for avoiding implicit
+	  declaration of gettimeofday().
+
+Sun Mar 17 00:55:31 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/missing.h: removed __linux__. it's unnecessary.
+
+Fri Mar 15 14:57:16 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c: disabled _FORTIFY_SOURCE for avoid to hit glibc bug.
+	  [Bug #8080] [ruby-core:53349]
+	* test/ruby/test_io.rb (TestIO#test_io_select_with_many_files):
+	  test for the above.
+
+Wed Mar 13 15:16:35 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/missing.h (__syscall): moved to...
+	* io.c: here. because __syscall() is only used from io.c.
+
+	* include/ruby/missing.h: move "#include <sys/type.h>" to ....
+	* include/ruby/intern.h: here. because it was introduced for
+	  fixing NFDBITS issue. [ruby-core:05179].
+
+Wed Mar 13 14:38:53 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/missing.h (struct timespec): include <sys/time.h>
+
+Wed Mar 13 13:54:45 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: check struct timeval exist or not.
+	* include/ruby/missing.h (struct timeval): check HAVE_STRUCT_TIMEVAL
+	  properly. and don't include sys/time.h if struct timeval exist.
+
+	* file.c: include sys/time.h explicitly.
+	* random.c: ditto.
+	* thread_pthread.c: ditto.
+	* time.c: ditto.
+	* ext/date/date_strftime.c: ditto.
+
+Fri Mar 15 14:45:02 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in (_FORTIFY_SOURCE): added a few comments.
+
+Fri Mar 15 14:17:55 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (numberof): renamed from ARRAY_SIZE() because
+	  other all files use numberof().
+
+Say Mar 15 01:33:00 2013  Charles Oliver Nutter <headius@headius.com>
+
+	* test/ruby/test_lazy_enumerator.rb (TestLazyEnumerator#test_drop_while):
+	  Modify while condition to show dropping remains off after first false
+	  value. This change was made in 39711.
+
+Fri Mar 15 23:06:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (GetTimeval): check if already initialized instance.
+
+	* time.c (GetNewTimeval): check if newly created instance.
+
+	* time.c (time_init_0, time_init_1, time_init_copy, time_mload): must
+	  be newly created instance.  [ruby-core:53436] [Bug #8099]
+
+Fri Mar 15 14:51:33 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_sys_fail_path_with_func): share same function, and path
+	  may be nil.
+
+Fri Mar 15 08:24:51 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_sys_fail_path): define & use rb_sys_fail_path0 like r39752
+
+Fri Mar 15 04:08:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* proc.c: Typo in Proc.arity found by Jack Nagel [Bug #8094]
+
+Thu Mar 14 16:59:09 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_function_name_string): macro for function name
+	  string predefined identifier, __func__ in C99, or __FUNCTION__ in
+	  gcc.
+
+	* file.c (rb_sys_fail_path): use RUBY_FUNCTION_NAME_STRING.
+
+Thu Mar 14 14:12:34 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_sys_fail_path): use rb_sys_fail_path0 only on GCC.
+	  __func__ is C99 feature.
+
+Thu Mar 14 12:59:59 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_sys_fail_path0): add to append the name of called function
+	  to ease debugging for example blow umask_spec failure.
+	  http://fbsd.rubyci.org/~chkbuild/ruby-trunk/log/20130309T010202Z.diff.html.gz
+
+	* file.c (rb_sys_fail_path): use rb_sys_fail_path0.
+
+Thu Mar 14 12:53:15 2013  Luis Lavena  <luislavena@gmail.com>
+
+	* win32/file.c (get_user_from_path):  add internal function that retrieves
+	  username from supplied path (refactored).
+	* win32/file.c (rb_file_expand_path_internal):  refactor expansion of user
+	  home to use get_user_from_path and cover dir_string corner cases.
+	  [ruby-core:53168] [Bug #8034]
+
+Thu Mar 14 11:53:01 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* NEWS: describe RUBY_HEAP_SLOTS_GROWTH_FACTOR.
+
+Thu Mar 14 10:01:12 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/globals.rdoc:  $? is thread-local
+
+Wed Mar 13 23:25:59 2013  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: allow to tune growth of heap by environment variable
+	  RUBY_HEAP_SLOTS_GROWTH_FACTOR. patched by tmm1(Aman Gupta).
+	  [Feature #8015] [ruby-core:53131]
+
+Wed Mar 13 19:43:46 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* doc/irb/irb.rd.ja: fix typo
+
+	* ext/tk/MANUAL_tcltklib.eng: fix typos
+
+	* ext/tk/sample/tktextframe.rb (Tk#component_delegates): fix typo
+
+Wed Mar 13 15:13:04 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_obj_singleton_methods): collect methods from the origin
+	  class.  [ruby-core:53207] [Bug #8044]
+
+Wed Mar 13 14:51:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_export_method): directly override the flag of method
+	  defined in prepending class too, not adding zsuper entry.
+	  [ruby-core:53106] [Bug #8005]
+
+Wed Mar 13 13:06:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rm, shvar_to_cpp, unexpand_shvar): local is not
+	  available on old shells.
+
+	* configure.in (shvar_to_cpp): escape quotes for old shells.
+	  [Bug #7959] [Bug #8071]
+
+Wed Mar 13 11:11:07 2013  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* object.c (Init_Object): remove Module#used, which has been
+	  introduced in Ruby 2.0 by mistake.  [Bug #7916] [ruby-core:52719]
+
+Wed Mar 13 05:49:29 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/irb.rb:  Fix typo
+
+Tue Mar 12 22:20:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_arguments, iseq_compile_each): support required
+	  keyword arguments.   [ruby-core:51454] [Feature #7701]
+
+	* iseq.c (rb_iseq_parameters): ditto.
+
+	* parse.y (f_kw, f_block_kw): ditto.  this syntax is still
+	  experimental, the notation may change.
+
+	* vm_core.h (rb_iseq_struct): ditto.
+
+	* vm_insnhelper.c (vm_callee_setup_keyword_arg): ditto.
+
+Tue Mar 12 17:02:53 2013  TAKANO Mitsuhiro <tak@no32.tk>
+
+	* date_core.c: clearly specify operator precedence.
+
+Tue Mar 12 17:00:45 2013  TAKANO Mitsuhiro <tak@no32.tk>
+
+	* insns.def: fix condition.
+
+Tue Mar 12 16:48:19 2013  TAKANO Mitsuhiro <tak@no32.tk>
+
+	* rational.c: fix dangling if, else-if and else.
+
+Tue Mar 12 06:27:59 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/setup_command.rb:  Don't delete non-rubygems
+	  files when installing RubyGems.
+	* test/rubygems/test_gem_commands_setup_command.rb:  Test for the
+	  above.
+
+	* lib/rubygems/ext/ext_conf_builder.rb:  Use full path to siteconf.rb
+	  in case the extconf.rb changes directories (like memcached does).
+
+	* lib/rubygems/package.rb:  Remove double slash from path.
+	* test/rubygems/test_gem_package.rb:  Test for the above.
+	* test/rubygems/test_gem_package_old.rb:  ditto.
+
+	* lib/rubygems/source.rb:  Revert automatic HTTPS upgrade
+	* lib/rubygems/spec_fetcher.rb:  ditto.
+	* test/rubygems/test_gem_remote_fetcher.rb:  ditto.
+	* test/rubygems/test_gem_source.rb:  ditto.
+	* test/rubygems/test_gem_spec_fetcher.rb:  ditto.
+
+Tue Mar 12 02:25:19 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/smtp.rb:  Added Net::SMTP#rset method to implement the SMTP
+	  RSET command.  [ruby-trunk - Feature #5373]
+	* NEWS:  ditto.
+	* test/net/smtp/test_smtp.rb:  Test for the above.
+
+Mon Mar 11 22:44:57 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv-replace.rb (TCPSocket#initialize): resolve the 3rd
+	  argument only if non-nil value is given.
+	  [ruby-dev:47150] [ruby-trunk - Bug #8054] reported and analyzed by
+	  mrkn.
+
+Mon Mar 11 19:22:54 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/mkmf/base.rb: class name conflict.
+
+Mon Mar 11 18:45:09 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enumerator.c (enumerator_with_index): try to convert given offset to
+	  integer. fix bug introduced in r39594.
+
+Mon Mar 11 17:27:57 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/envutil.rb (EnvUtil.with_default_external): add for
+	  changing Encoding.default_external without warnings.
+
+	* test/ruby/envutil.rb (EnvUtil.with_default_internal): ditto.
+
+	* test/ruby/test_io_m17n.rb: use above with_default_external.
+
+Mon Mar 11 16:57:00 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (extract_binmode): raise error even if binmode and textmode
+	  don't conflict. [Bug #5918] [ruby-core:42199]
+
+Mon Mar 11 12:25:12 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Merge Onigmo d4bad41e16e3eccd97ccce6f1f96712e557c4518.
+	  fix lookbehind assertion fails with /m mode enabled. [Bug #8023]
+	  fix \Z matches where it shouldn't. [Bug #8001]
+
+Mon Mar 11 11:53:35 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#dir_config, MakeMakefile#_libdir_basename):
+	  defer use of instance variable until needed.  [Bug #8074]
+
+Thu Mar  7 10:42:28 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/thread.rb (Queue#clear): return self.
+	  Patch by Cubing Cube. Thank you! [Bug #7947] [ruby-dev:47098]
+	* lib/thread.rb (Queue#push): ditto.
+	* lib/thread.rb (SizedQueue#push): ditto.
+	* test/thread/test_queue.rb: add tests for the above.
+
+Thu Mar  7 10:40:49 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* tool/change_maker.rb (#diff2index): check Encoding::BINARY.
+	  BASERUBY may still be 1.8.x.
+
+Thu Mar  7 08:47:42 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* NEWS (Mutex#owned?): no longer experimental.
+
+Sun Mar 10 23:38:15 2013  Luis Lavena  <luislavena@gmail.com>
+
+	* win32/file.c (rb_file_expand_path_internal):  Expand home directory when
+	  used as second parameter (dir_string).  [ruby-core:53168] [Bug #8034]
+	* test/ruby/test_file_exhaustive.rb: add test to verify.
+
+Sun Mar 10 23:27:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
+	  it is impossible to predict which file will be installed to where,
+	  by the arguments, so use intermediate destination directory always.
+	  [Bug #7698]
+
+Sun Mar 10 17:00:22 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: edited rdoc.
+	* rational.c: ditto.
+
+Sun Mar 10 15:02:39 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (setup_communication_pipe): remove unused function.
+	  it was unintentionally added r39683.
+
+Wed Mar  6 00:30:40 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* tool/gen_ruby_tapset.rb: add tapset generator.
+
+Wed Mar  6 03:27:43 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* probes.d (symbol-create): change argument name `string' to
+	  `str'. `string' is a keyword for systemtap.
+
+Tue Mar  5 22:23:01 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* probes.d: added argument name
+
+Thu Mar  7 01:17:00 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/thread/test_queue.rb (TestQueue#test_thr_kill): reduce
+	  iterations from 2000 to 250. When running on uniprocessor
+	  systems, every th.kill needs TIME_QUANTUM_USEC time (i.e.
+	  100msec on posix systems). Because, "r.read 1" is 3 steps
+	  operations that 1) release GVL 2) read 3) acquire gvl and
+	  (1) invoke context switch to main thread. and then, main
+	  thread's th.kill resume (1), but not (2). Thus read interrupt
+	  need TIME_QUANTUM_USEC. Then maximum iteration is 30sec/100msec
+	  = 300.
+
+Thu Mar  7 00:14:51 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (rb_update_max_fd): use ATOMIC_CAS because this function
+	  is used from timer thread too.
+
+Wed Mar  6 23:30:21 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (ARRAY_SIZE): new.
+	* thread_pthread.c (gvl_acquire_common): use low priority
+	  notification for avoiding timer thread interval confusion.
+	  If we use timer_thread_pipe[1], every gvl_yield() request
+	  one more gvl_yield(). It lead to thread starvation.
+	  [Bug #7999] [ruby-core:53095]
+	* thread_pthread.c (rb_reserved_fd_p): adds timer_thread_pipe_low
+	  to reserved fds.
+
+Wed Mar  6 22:36:19 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (rb_thread_wakeup_timer_thread_fd): add fd
+	  argument and remove hardcoded dependency of timer_thread_pipe[1].
+	* thread_pthread.c (consume_communication_pipe): add fd argument.
+	* thread_pthread.c (close_communication_pipe): ditto.
+
+	* thread_pthread.c (timer_thread_sleep): adjust the above changes.
+
+	* thread_pthread.c (setup_communication_pipe_internal): factor
+	  out pipe initialize logic.
+
+Wed Mar  6 22:56:14 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (ubf_select): add to small comments why we
+	  need to call rb_thread_wakeup_timer_thread().
+
+Wed Mar  6 21:42:24 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): factor out
+	  creating communication pipe logic into separate function.
+	* thread_pthread.c (setup_communication_pipe): new helper function.
+	* thread_pthread.c (set_nonblock): moves a definition before
+	  setup_communication_pipe.
+
+Sun Mar  3 02:42:29 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (consume_communication_pipe): retry when
+	  read returned CCP_READ_BUFF_SIZE.
+
+Wed Mar  6 21:31:35 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (timer_thread_sleep): use poll() instead of
+	  select(). select doesn't work if timer_thread_pipe[0] is
+	  greater than FD_SETSIZE.
+	* thread_pthread.c (USE_SLEEPY_TIMER_THREAD): add a dependency
+	  against poll.
+
+Wed Mar  6 21:00:23 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (USE_SLEEPY_TIMER_THREAD): use more accurate
+	  ifdef conditions.
+
+Sun Mar  3 02:30:36 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (set_nonblock): new helper function for set
+	  O_NONBLOCK.
+	* thread_pthread.c (rb_thread_create_timer_thread): set O_NONBLOCK
+	  to timer_thread_pipe[0] too.
+
+Sun Mar 10 09:12:51 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* complex.c: described syntax of string form.
+	* rational.c: ditto.
+
+Sat Mar  9 11:58:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_extended): check for prepended object.
+	  [ruby-core:53206] [Bug #8043]
+
+Sat Mar  9 08:36:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (features_index_add_single, rb_feature_p): store single index
+	  as Fixnum to reduce the number of arrays for the indexes.  based on
+	  the patch by tmm1 (Aman Gupta) in [ruby-core:53216] [Bug #8048].
+
+Sat Mar  9 00:25:57 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_object0): load prepended objects.  treat the class of
+	  extended object in the included modules as prepended singleton
+	  class. [ruby-core:53202] [Bug #8041]
+
+Fri Mar  8 19:44:00 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* man/rake.1, man/ruby.1: Use the Pa macro to make URLs stand out.
+
+Fri Mar  8 13:20:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/pathname/pathname.c (path_f_pathname): rdoc for Pathname()
+
+Fri Mar  8 12:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* man/rake.1: Document ENVIRONMENT variables on RAKE(1) manpage
+
+Fri Mar  8 10:44:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/webrick/httpproxy.rb: Fix typos in HTTPProxyServer [Bug #8013]
+	  Patch by Nobuhiro IMAI [ruby-core:53127]
+
+Fri Mar  8 03:16:15 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* class.c (rb_mod_ancestors): Include singleton_class in ancestors
+	  list [Feature #8035]
+
+	* test/ruby/test_module.rb (class): test for above
+
+	* test/ruby/marshaltestlib.rb (module): adapt test
+
+	* NEWS: list change
+
+Thu Mar  7 14:21:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): pass keyword arguments to zsuper,
+	  with current values.  [ruby-core:53114] [Bug #8008]
+
+Thu Mar  7 12:53:47 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/setup_command.rb:  Install .pem files.
+	* test/rubygems/test_gem_commands_setup_command.rb:  Test for the
+	  above.
+
+	* lib/rubygems/spec_fetcher.rb:  Test HTTPS upgrade with URI::HTTPS,
+	  not URI::HTTP.  Fixes bug in automatic HTTPS upgrade.
+	* test/rubygems/test_gem_spec_fetcher.rb:  Test for the above.
+
+	* lib/rubygems.rb:  Version 2.0.2
+
+	* lib/rubygems/test_utilities.rb:  Ensure scheme and uri class match.
+
+Thu Mar  7 10:39:04 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (gem): Gem.ensure_gem_subdirectories now has mode
+	  option since r39607.  refix of r38870.
+
+Wed Mar  6 13:14:28 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/test_gem_spec_fetcher.rb:  Removed unused variable.
+
+Wed Mar  6 08:10:15 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/test_require.rb:  Fix tests when 'a.rb' exists.
+	  [ruby-trunk - Bug #7749]
+
+Wed Mar  6 08:00:59 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Allow specification of directory permissions.
+	  [ruby-trunk - Bug #7713]
+	* test/rubygems/test_gem.rb:  Test for the above.
+
+Wed Mar  6 07:40:21 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/query_command.rb:  Only fetch remote specs when
+	  showing details.  [ruby-trunk - Bug #8019]  RubyGems bug #487
+	* lib/rubygems/remote_fetcher.rb:  ditto.
+	* lib/rubygems/security/policy.rb:  ditto.
+	* test/rubygems/test_gem_commands_query_command.rb:  Test for the
+	  above.
+
+	* lib/rubygems/security.rb:  Make OpenSSL optional for RubyGems.
+	* lib/rubygems/commands/cert_command.rb:  ditto.
+
+	* lib/rubygems/config_file.rb:  Display file with YAML error, not
+	  ~/.gemrc
+
+	* lib/rubygems/remote_fetcher.rb:  Only create gem subdirectories when
+	  installing gems.
+	* lib/rubygems/dependency_resolver.rb:  ditto.
+	* lib/rubygems/test_utilities.rb:  ditto.
+	* test/rubygems/test_gem_commands_fetch_command.rb:  Test for the
+	  above.
+
+	* lib/rubygems/spec_fetcher.rb:  Only try to upgrade
+	  http://rubygems.org to HTTPS
+	* test/rubygems/test_gem_spec_fetcher.rb:  Test for the above.
+
+	* lib/rubygems.rb:  Update win_platform? check for JRuby compatibility.
+
+	* test/rubygems/test_gem_installer.rb:  Update for Ruby 1.9.2
+	  compatibility
+
+Wed Mar  6 01:19:28 2013  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* enumerator.c (enumerator_with_index, lazy_take): use INT2FIX(0)
+	  instead of INT2NUM(0).
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_exp): ditto.
+
+	* ext/fiddle/function.c (function_call): ditto.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_initialize): ditto.
+
+	* process.c (proc_getsid): ditto.
+
+	* transcode.c (econv_finish): ditto.
+
+Tue Mar  5 21:36:43 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_prepend_module): check redefinition of built-in optimized
+	  methods.  [ruby-dev:47124] [Bug #7983]
+
+	* vm.c (rb_vm_check_redefinition_by_prepend): ditto.
+
+Tue Mar  5 20:29:25 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew): revert r39224.  [ruby-core:53038] [Bug #7988]
+
+Tue Mar  5 20:23:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_check_arity): make a static inline
+	  function so it can be used as an expression and argc would be
+	  evaluated only once.
+
+Tue Mar  5 12:30:55 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Bump version to 2.0.1 for upcoming bugfix release
+
+	* lib/rubygems/ext/ext_conf_builder.rb:  Restore ruby 1.8 compatibility
+	  for [Bug #7698]
+	* test/rubygems/test_gem_installer.rb:  Ditto.
+
+	* lib/rubygems/package.rb:  Restore ruby 1.8 compatibility.
+
+	* test/rubygems/test_gem_dependency_installer.rb:  Fix warnings
+
+Tue Mar  5 12:24:23 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* enumerator.c (enumerator_with_index):  Restore handling of a nil memo
+	  from r39594.
+
+Tue Mar  5 10:40:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/objspace/objspace.c (count_nodes): count also newly added nodes,
+	  and fix key for unknown node.  patch by tmm1 (Aman Gupta) in
+	  [ruby-core:53130] [Bug #8014]
+
+Tue Mar  5 10:20:16 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (enumerator_with_index_i): allow Bignum as offset, to
+	  get rid of conversion exception and integer overflow.
+	  [ruby-dev:47131] [Bug #8010]
+
+	* numeric.c (rb_int_succ, rb_int_pred): shortcut optimization for
+	  Bignum.
+
+Tue Mar  5 10:02:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
+	  clear DESTDIR so RUBYARCHDIR and RUBYLIBDIR are not be overridden.
+	  [Bug #7698]
+
+Mon Mar  4 15:33:40 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
+	  fix for unusual cases again.  install to a temporary directory once
+	  and move installed files to the destination directory, if it is same
+	  as the current directory.  [Bug #7698]
+
+Mon Mar  4 14:13:36 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (miniruby, ruby): move MAINLIBC because linker arguments
+	  must appear after object files with newer versions of gcc.  patch by
+	  tmm1 (Aman Gupta) in [ruby-core:53121] [Bug #8009]
+
+Mon Mar  4 10:23:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* encoding.c: Typo in Encoding overview by Tom Wardrop [GH fixes #255]
+
+Sun Mar  3 12:35:08 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#libpath_env): set runtime library path for
+	  the case rpath is disabled.
+
+Sun Mar  3 12:17:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/ext/ext_conf_builder.rb
+	  (Gem::Ext::ExtConfBuilder.hack_for_obsolete_style_gems): remove
+	  circular dependencies in install-so too.  [ruby-core:52882]
+	  [Bug #7698]
+
+Sun Mar  3 07:33:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/socket/tcpserver.c: Grammar for TCPServer.new from r39554
+
+Sun Mar  3 01:17:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/ext/ext_conf_builder.rb
+	  (Gem::Ext::ExtConfBuilder.hack_for_obsolete_style_gems): remove
+	  circular dependencies for old style gems which locate extconf.rb on
+	  the toplevel.  [ruby-core:53059] [ruby-trunk - Bug #7698]
+
+	* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
+	  use RUBYOPT instead of -r option, and revert some tests.  [Bug #7698]
+
+	* lib/rubygems/ext/ext_conf_builder.rb (Gem::Ext::ExtConfBuilder.build):
+	  revert use of temporary directory for build, to work some buggy
+	  extconf.rb which cannot build outside the source directory.
+	  [ruby-core:53056] [Bug #7698]
+
+Sun Mar  3 00:04:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (CPPFLAGS), lib/mkmf.rb (MakeMakefile#create_makefile):
+	  define RUBY_EXPORT for static-linked-ext mswin.  [Bug #7960]
+
+Sat Mar  2 22:49:47 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (ENCOBJS, EXTOBJS, config.h): definitions for
+	  static-linked-ext.  [Bug #7960]
+
+Sat Mar  2 17:34:19 2013  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/utils.rb: use Socket.tcp_server_sockets to create server
+	  sockets.
+	  fix [Bug #7100] https://bugs.ruby-lang.org/issues/7100
+	  reported by sho-h (Sho Hashimoto).
+
+Sat Mar  2 02:45:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c: typo in comment patch by Nami-Doc [Github fixes #253]
+
+Sat Mar  2 01:33:17 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Merge Onigmo 0fe387da2fee089254f6b04990541c731a26757f
+	  v5.13.3 [Bug#7972] [Bug#7974]
+
+Fri Mar  1 11:09:06 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/fileutils.rb:  Revert r34669 which altered the way
+	  metaprogramming in FileUtils occurred.  [ruby-trunk - Bug #7958]
+
+	* test/fileutils/visibility_tests.rb:  Refactored tests of FileUtils
+	  options modules to expose bug found in #7958
+	* test/fileutils/test_dryrun.rb:  ditto.
+	* test/fileutils/test_nowrite.rb:  ditto.
+	* test/fileutils/test_verbose.rb:  ditto.
+
+Fri Mar  1 09:18:00 2013  Zachary Scott <zachary@zacharyscott.net>
+
+	* lib/psych.rb: specify in rdoc what object is returned in parser
+	  By Adam Stankiewicz [Github tenderlove/psych#133]
+
+Fri Mar  1 07:21:41 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/ext/builder.rb:  Fix incompatibilities when installing
+	  extensions.  Patch by Nobu.
+	  [ruby-trunk - Bug #7698] [ruby-trunk - Bug #7971]
+	* lib/rubygems/ext/ext_conf_builder.rb:  ditto.
+	* lib/rubygems/installer.rb:  ditto.
+	* test/rubygems/test_gem_ext_ext_conf_builder.rb:  Test for the above.
+	* test/rubygems/test_gem_installer.rb:  ditto.
+
+	* lib/rubygems/commands/sources_command.rb:  Prefer HTTPS over HTTP.
+	* lib/rubygems/defaults.rb:  ditto
+	* lib/rubygems/dependency_resolver.rb:  Ditto.
+	* lib/rubygems/source.rb:  ditto.
+	* lib/rubygems/spec_fetcher.rb:  ditto.
+	* lib/rubygems/specification.rb:  ditto.
+	* lib/rubygems/test_utilities.rb:  ditto.
+	* test/rubygems/test_gem.rb:  Test for the above.
+	* test/rubygems/test_gem_commands_sources_command.rb:  ditto.
+	* test/rubygems/test_gem_dependency_resolver_api_set.rb:  ditto.
+	* test/rubygems/test_gem_remote_fetcher.rb:  ditto.
+	* test/rubygems/test_gem_source.rb:  ditto.
+	* test/rubygems/test_gem_spec_fetcher.rb:  ditto.
+
+Fri Mar 1 03:25:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/psych/lib/psych.rb: rdoc for Psych overview by Adam Stankiewicz
+	  [Github tenderlove/psych#134]
+
+Thu Feb 28 22:57:48 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): remove redundant trace(line)
+	  instruction. for example, at the following script
+	    def m()
+	      p:xyzzy
+	      1
+	      2
+	    end
+	  compiler ignores `1' because there is no effect. However,
+	  `trace(line)' instruction remains in bytecode.
+	  This modification removes such redundant trace(line) instruction.
+
+	* test/ruby/test_iseq.rb: add a test.
+
+Thu Feb 28 22:23:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (inspect_sockaddr): don't show that Unix
+	  domain socket filename is bigger than sizeof(sun_path).
+	  This limit is not rigid on some platforms such as Darwin and SunOS.
+
+Thu Feb 28 21:33:01 2013  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* configure.in(AC_DISABLE_OPTION_CHECKING): avoid warning "WARNING:
+	  Unrecognized options: --with-PACKAGE".
+
+Thu Feb 28 20:22:04 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_data_to_ary): fix condition.
+	  r34303 introduces a bug to avoid all line information from
+	  a result of ISeq#to_a. This is a regression problem from 2.0.0p0.
+
+	* test/ruby/test_iseq.rb: add a test of lines after ISeq#to_a.
+
+Thu Feb 28 08:20:33 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/available_set.rb:  Undent for style
+
+	* lib/rubygems/dependency_installer.rb:  Pick latest prerelease gem to
+	  install.  Fixes RubyGems bug #468.
+	* test/rubygems/test_gem_dependency_installer.rb:  Test for the above.
+
+	* lib/rubygems/dependency_installer.rb:  Don't display "Done installing
+	  documentation" if documentation will not be installed.
+	* lib/rubygems/rdoc.rb:  ditto
+
+	* lib/rubygems/dependency_list.rb:  Use Array#concat for Ruby 1.x
+	  performance.
+
+	* lib/rubygems/installer.rb:  Use formatted program name when comparing
+	  executables.  RubyGems pull request #471
+	* test/rubygems/test_gem_installer.rb:  Test for the above.
+
+	* lib/rubygems/package.rb:  Use more explicit feature check to work
+	  around JRuby bug #552
+
+	* lib/rubygems/ssl_certs/GeoTrust_Global_CA.pem:  Added GeoTrust root
+	  certificate.
+
+	* test/rubygems/test_gem_source_list.rb:  Use "example" instead of real
+	  hostname
+
+Thu Feb 28 05:57:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c: rdoc formatting for Thread, ThreadGroup, and ThreadError
+
+Thu Feb 28 02:42:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm.c: Typo in overview for example of Thread#status returning false
+	  Reported by Lee Jarvis
+
+Wed Feb 27 22:54:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/rubysocket.h (union_sockaddr): make it longer for SunOS
+	  and Darwin.
+
+Wed Feb 27 21:14:34 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/security.rb (REXML::Security): create.
+	* lib/rexml/rexml.rb: move entity_expansion_limit and
+	  entity_expansion_text_limit accessors to ...
+	* lib/rexml/security.rb: ... here.
+	* lib/rexml/document.rb: use REXML::Security.
+	* lib/rexml/text.rb: use REXML::Security.
+	* test/rexml/test_document.rb: use REXML::Security.
+
+Wed Feb 27 19:53:32 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* vm.c (Thread): fix typos in overview
+
+Wed Feb 27 13:21:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm.c (Thread): Typo in overview, swap setting and getting
+
+Wed Feb 27 13:02:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm.c (Thread): Documentation overview of Thread class
+
+Wed Feb 27 12:57:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c (rb_thread_wakeup): rdoc formatting
+
+Wed Feb 27 12:53:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c (rb_thread_group): rdoc formatting
+
+Wed Feb 27 12:33:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/ostruct.rb: Typo in OpenStruct overview [Github Fixes #251]
+	  Patch by Chun-wei Kuo
+
+Wed Feb 27 12:13:32 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_exec.h (END_INSN): llvm-gcc may optimize out reg_cfp and cause
+	  Stack/cfp consistency error when the instruction doesn't use reg_cfp.
+	  Usually instructions use PUSH() but for example trace doesn't.
+	  This hack cause speed down but you shouldn't use llvm-gcc, use clang.
+	  [Bug #7938]
+
+Wed Feb 27 10:23:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c (thread_raise_m): rdoc formatting
+
+Tue Feb 26 23:32:44 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb: move entity_expansion_limit accessor to ...
+	* lib/rexml/rexml.rb: ... here for consistency.
+	* lib/rexml/document.rb (REXML::Document.entity_expansion_limit):
+	  deprecated.
+	* lib/rexml/document.rb (REXML::Document.entity_expansion_limit=):
+	  deprecated.
+
+Tue Feb 26 23:26:13 2013  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/document.rb: move entity_expansion_text_limit accessor to ...
+	* lib/rexml/rexml.rb: ... here to make rexml/text independent from
+	  REXML::Document. It causes circular require.
+	* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit):
+	  deprecated.
+	* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit=):
+	  deprecated.
+	* lib/rexml/text.rb: add missing require "rexml/rexml" for
+	  REXML.entity_expansion_text_limit.
+	  Reported by Robert Ulejczyk. Thanks!!! [ruby-core:52895] [Bug #7961]
+
+Tue Feb 26 15:12:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb: reconstruct comma separated list values.  a
+	  command line to Windows batch file is split not only by spaces
+	  and equal signs but also by commas and semicolons.
+
+Tue Feb 26 15:04:19 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (unexpand_shvar): get rid of non-portable shell
+	  behavior on OpenBSD, so no extra quotes.  [Bug #7959]
+
+Tue Feb 26 10:24:49 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (IS_LABEL_POSSIBLE): allow labels for keyword arguments just
+	  after method definition without a parenthesis.  [ruby-core:52820]
+	  [Bug #7942]
+
+Tue Feb 26 04:50:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* error.c: clarify reason for sleep in SignalException example
+
+Tue Feb 26 03:47:00 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* error.c: clarify a document of SignalException. Process.kill()
+	  doesn't have any guarantee when signal will be delivered.
+	  [Bug #7951] [ruby-core:52864]
+
+Mon Feb 25 23:51:04 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/version.h: bump RUBY_API_VERSION same as RUBY_VERSION.
+
+Mon Feb 25 21:03:34 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_byte_substr): don't set coderange if it's not known.
+	  [Bug #7954] [ruby-dev:47108]
+
+Mon Feb 25 16:47:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (realclean-local): miniprelude.c is made by srcs, so it
+	  should not removed by distclean but by realclean.  [Bug #6807]
+
+Mon Feb 25 16:30:30 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/config_file.rb:  Lazily load .gem/credentials to only
+	  check permissions when necessary.  RubyGems bug #465
+	* test/rubygems/test_gem_config_file.rb:  Test for the above.
+
+	* test/rubygems/test_gem_commands_push_command.rb:  Remove duplicated
+	  test.
+
+Mon Feb 25 15:47:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (ARFLAGS): VisualC++ linker does not allow spaces between
+	  output option and the output file name.  [Bug #7950]
+
+	* enc/depend (RANLIB): set default command to do nothing, or make the
+	  entire line a label on Windows.
+
+Mon Feb 25 14:41:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#init_mkmf): default libdirname to libdir.
+
+	* tool/rbinstall.rb: ditto.
+
+Mon Feb 25 13:12:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (setup): find Setup file from target_os 1. by
+	  suffix (e.g. Setup.nacl, Setup.atheos), 2. by "platform"
+	  option (e.g. Setup.nt, Setup.emx), and 3. default Setup.  And
+	  Setup.dj had been removed.
+
+Mon Feb 25 12:48:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c: Document Thread::new, clean up ::fork and mention calling
+	  super if subclassing Thread
+
+Mon Feb 25 12:38:50 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: don't test ss_family and ss_len member of
+	  struct sockaddr_storage.  They are not used now except SunOS
+	  specific code.
+
+Mon Feb 25 11:03:38 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in (unexpand_shvar): Use the numeric comparison
+	  operator instead of '==' which is a ksh extension. [Bug #7941]
+
+Mon Feb 25 02:37:56 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: define and use union_sockaddr instead of struct
+	  sockaddr_storage for less casts.
+
+	* ext/socket/rubysocket.h (union_sockaddr): defined.
+
+	* ext/socket/socket.c (sock_accept): use union_sockaddr.
+	  (sock_accept_nonblock): ditto.
+	  (sock_sysaccept): ditto.
+	  (sock_s_getnameinfo): ditto.
+
+	* ext/socket/basicsocket.c (bsock_getsockname): ditto.
+	  (bsock_getpeername): ditto.
+	  (bsock_local_address): ditto.
+	  (bsock_remote_address): ditto.
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): ditto.
+
+	* ext/socket/init.c (recvfrom_arg): ditto.
+	  (recvfrom_blocking): ditto.
+	  (rsock_s_recvfrom): ditto.
+	  (rsock_s_recvfrom_nonblock): ditto.
+	  (rsock_getfamily): ditto.
+
+	* ext/socket/raddrinfo.c (rb_addrinfo_t): ditto.
+	  (ai_get_afamily): ditto.
+	  (inspect_sockaddr): ditto.
+	  (addrinfo_mdump): ditto.
+	  (addrinfo_mload): ditto.
+	  (addrinfo_getnameinfo): ditto.
+	  (addrinfo_ip_port): ditto.
+	  (extract_in_addr): ditto.
+	  (addrinfo_ipv6_to_ipv4): ditto.
+	  (addrinfo_unix_path): ditto.
+
+	* ext/socket/tcpserver.c (tcp_accept): ditto.
+	  (tcp_accept_nonblock): ditto.
+	  (tcp_sysaccept): ditto.
+
+	* ext/socket/ipsocket.c (ip_addr): ditto.
+	  (ip_peeraddr): ditto.
+	  (ip_s_getaddress): ditto.
+
+Sun Feb 24 21:15:05 2013  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c: [ruby-core:52303]
+
+Sun Feb 24 15:33:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_random_ulong_limited): limit is inclusive, but generic
+	  rand method should return a number less than it, so increase for the
+	  difference.  [ruby-core:52779] [Bug #7935]
+
+Sun Feb 24 15:32:36 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_random_ulong_limited): limit is inclusive, but generic
+	  rand method should return a number less than it, so increase for the
+	  difference.  [ruby-core:52779] [Bug #7935]
+
+Sun Feb 24 15:14:43 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Removed duplicate Accept-Encoding in Net::HTTP#get.
+	  [ruby-trunk - Bug #7924]
+	* test/net/http/test_http.rb:  Test for the above.
+
+Wed Feb 20 14:28:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c: Document ThreadGroup::Default
+
+Wed Feb 20 14:23:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c: Grammar for #backtrace_locations and ::handle_interrupt
+
+Sun Feb 24 13:35:57 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): block level control frame does not
+	  have method entry, so obtain the method entry from method top-level
+	  control frame to be compared with refined method entry.
+	  [ruby-core:52750] [Bug #7925]
+
+Wed Feb 20 13:23:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* object.c: Document methods receiving string and convert to symbol
+	  Patch by Stefan Rusterholz
+	* vm_eval.c: ditto
+	* vm_method.c: ditto
+
+Wed Feb 20 07:20:56 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (sigsegv): suppress unused result warning. Because
+	  write(2) is marked __warn_unused_result__ on Linux glibc.
+
+Sun Feb 24 07:50:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_arguments): no keyword check if any keyword rest
+	  argument exists, even unnamed.  [ruby-core:52744] [Bug #7922]
+
+Sat Feb 23 16:51:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c: Documentation for Thread#backtrace_locations
+
+Sat Feb 23 16:05:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm.c: Typo in ObjectSpace::WeakMap overview
+
+Sat Feb 23 16:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* thread.c: Improved rdoc for ::handle_interrupt, ::pending_interrupt?
+	  and #pending_interrupt?
+
+Sat Feb 23 12:26:43 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el (ruby-electric-curlies)
+	  (ruby-electric-matching-char, ruby-electric-bar): Avoid electric
+	  insertion when there is a prefix argument.
+
+	* misc/ruby-electric.el (ruby-electric-insert)
+	  (ruby-electric-cua-replace-region-p)
+	  (ruby-electric-cua-replace-region): Avoid electric insertion and
+	  fall back when cua-mode is enabled and a region is active.
+
+Sat Feb 23 12:35:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c: Document #<=> return values and formatting
+	* bignum.c: ditto
+	* file.c: ditto
+	* object.c: ditto
+	* numeric.c: ditto
+	* rational.c: ditto
+	* string.c: ditto
+	* time.c: ditto
+
+Sat Feb 23 10:50:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* array.c (rb_ary_diff, rb_ary_and, rb_ary_or): Document return order
+	  [RubySpec #7803]
+
+Sat Feb 23 10:17:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* object.c (rb_obj_comp): Documenting Object#<=> return values
+	  Patch by Stefan Rusterholz
+
+Sat Feb 23 09:48:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (file_s_fnmatch, fnmatch_brace): encoding-incompatible pattern
+	  and string do not match, instead of exception.  [ruby-dev:47069]
+	  [Bug #7911]
+
+Sat Feb 23 08:57:46 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* doc/NEWS-*: Update NEWS from their respective branches
+
+Sat Feb 23 08:14:43 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* NEWS: many additions for Ruby 2.0.0
+
+	* object.c: Add doc for Module.prepended
+
+Sat Feb 23 07:52:53 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/ruby.pc.in: reorder library flags which may refer library
+	  names.  [Bug #7913]
+
+Fri Feb 22 23:46:20 2013  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit):
+	  fix a typo in comment in r39384.
+
+Fri Feb 22 18:31:46 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/rexml/document.rb (REXML::Document.entity_expansion_text_limit):
+	  new attribute to read/write entity expansion text limit.  the default
+	  limit is 10Kb.
+
+	* lib/rexml/text.rb (REXML::Text.unnormalize): check above attribute.
+
+Fri Feb 22 17:36:23 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/test_rbconfig.rb (TestRbConfig): fix r39372.
+	  It must see RbConfig::CONFIG instead of CONFIG.
+
+Fri Feb 22 14:55:41 2013  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* signal.c (ruby_abort): fix typo in r39354 [Bug #5014]
+
+Fri Feb 22 12:46:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rb_random_ulong_limited): fix error message for negative
+	  value.  [ruby-dev:47061] [Bug #7903]
+
+Fri Feb 22 11:36:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/test_rbconfig.rb (TestRbConfig): skip user defined values by
+	  configuration options.  [Bug #7902]
+
+Fri Feb 22 11:33:42 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#init_mkmf): adjust default library path
+	  for multiarch.  [Bug #7874]
+
+Fri Feb 22 11:10:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* enum.c (Enumerable#chunk: Improved examples, grammar, and formatting
+	  Patch by Dan Bernier and Rich Bruchal of newhaven.rb
+	  [Github documenting-ruby/ruby#8]
+
+Fri Feb 22 11:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* numeric.c: Examples and formatting for Numeric and Float
+	  Based on a patch by Zach Morek and Oren K of newhaven.rb
+	  [Github documenting-ruby/ruby#5]
+
+Fri Feb 22 07:04:41 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/installer.rb (build_extensions):  Create extension
+	  install destination before building extension.  Patch by Kenta Murata.
+	  [ruby-trunk - Bug #7897]
+	* test/rubygems/test_gem_installer.rb:  Test for the above.
+
+Fri Feb 22 06:30:57 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/globals.rdoc:  Document what setting $DEBUG does.
+
+	* doc/globals.rdoc:  Added pointer to $-d for full documentation.
+
+Fri Feb 22 06:27:07 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/globals.rdoc:  Document what setting $VERBOSE does.  [Bug #7899]
+
+	* doc/globals.rdoc:  Added pointer to $-w and $-v for full
+	  documentation.
+
+Fri Feb 22 02:33:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/abbrev.rb: Add words parameter to Abbrev::abbrev
+	  Patch by Devin Weaver [Github documenting-ruby/ruby#7]
+
+Thu Feb 21 17:28:14 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: add interaction when only ChangeLog is modified.
+
+Thu Feb 21 16:34:46 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (check_stack_overflow): extract duplicated code and get rid
+	  of declaration-after-statement.  [Bug #5014]
+
+Thu Feb 21 14:14:13 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* signal.c (sigsegv): avoid to use async signal unsafe functions
+	  when nested sigsegv is happen.
+	  [Bug #5014] [ruby-dev:44082]
+
+Thu Feb 21 13:47:59 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (rb_group_member): added an error check. SUS says,
+	  getgroups(small_value) may return EINVAL.
+
+Thu Feb 21 13:37:07 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* process.c (RB_MAX_GROUPS): moved to
+	* internal.h (RB_MAX_GROUPS): here.
+
+	* file.c (rb_group_member): use RB_MAX_GROUPS instead of
+	  RUBY_GROUP_MAX. They are the same.
+
+Thu Feb 21 13:15:40 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (access_internal): removed.
+	* file.c (rb_file_readable_real): use access() instead of
+	  access_internal().
+	* file.c (rb_file_writable_real): ditto.
+	* file.c (rb_file_executable_real): ditto.
+
+Thu Feb 21 13:04:59 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (eaccess): use access() when not using setuid nor setgid.
+	  This is minor optimization.
+
+Thu Feb 21 12:56:19 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (rb_group_member): get rid of NGROUPS dependency.
+	  [Bug #7886] [ruby-core:52537]
+
+Thu Feb 21 12:45:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): try two levels upper for stripping
+	  libdir name.  [Bug #7874]
+
+	* configure.in (libdir_basename): expand with multiarch in configure,
+	  not to defer the expansion till ruby.pc.in and mkmf.rb.  [Bug #7874]
+
+	* configure.in (libdir_basename): also -rpath and -install_name flags
+	  are affected when libruby directory changes.  [Bug #7874]
+
+Wed Feb 20 19:27:02 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (HAVE_RB_SCAN_ARGS_OPTIONAL_HASH): for
+	  rb_scan_args() optional hash feature.  [Bug #7861]
+
+Wed Feb 20 18:02:26 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (target_os): do not strip -gnu suffix on Linux if
+	  --target is given explicitly.  [Bug #7874]
+
+	* configure.in (libdirname): adjust library path name which libruby
+	  files will be installed.  [Bug #7874]
+
+	* tool/rbinstall.rb (libdir): ditto.
+
+Wed Feb 20 13:37:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/pty/pty.c: Documentation for the PTY module
+
+Wed Feb 20 12:18:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* object.c: Document Data class [Bug #7890] [ruby-core:52549]
+	  Patch by Matthew Mongeau
+
+Wed Feb 20 11:50:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/mutex_m.rb: Add rdoc for Mutex_m module
+
+Wed Feb 20 09:34:43 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/update_command.rb:  Create the installer after
+	  options are processed.  [ruby-trunk - Bug #7779]
+	* test/rubygems/test_gem_commands_update_command.rb:  Test for the
+	  above.
+
+Wed Feb 20 07:51:19 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/installer.rb:  Use gsub instead of gsub! to avoid
+	  altering @bin_dir.  Fixes tests on windows.  [ruby-trunk - Bug #7885]
+
+Tue Feb 19 20:50:00 2013  Kenta MURATA  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.gemspec: bump to 1.2.0.
+	  [ruby-core:51777] [Bug #7761]
+
+Tue Feb 19 13:07:25 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/syslog/syslog.c (Init_syslog): Define inspect as a singleton
+	  method and remove it as an instance method. [Bug #6502]
+
+Tue Feb 19 12:30:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* object.c: rdoc formatting for Kernel#Array()
+	* array.c: Add rdoc for Array() method to Creating Arrays section
+
+Tue Feb 19 10:35:52 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl.c (class OpenSSL):  Use only inner parenthesis in
+	  create_extension examples.
+
+Tue Feb 19 10:27:12 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl.c (class OpenSSL):  Fixed ExtensionFactory example.
+	  Patch by Richard Bradley.  [ruby-trunk - Bug #7551]
+
+Tue Feb 19 08:32:11 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (vm_call0_body): check interrupts after method dispatch
+	  from C methods. [Bug #7878]
+
+Tue Feb 19 08:14:40 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/installer.rb:  Fixed placement of executables with
+	  --user-install.  [ruby-trunk - Bug #7779]
+	* test/rubygems/test_gem_installer.rb:  Test for above.
+
+Tue Feb 19 06:04:06 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump: FreeBSD ports' libexecinfo's backtrace(3) can't trace
+	  beyond signal trampoline, and as described in r38342 it can't
+	  trace on -O because it see stack frame pointers.
+	  libunwind unw_backtrace see dwarf information in the binary
+	  and it works with -O (without frame pointers).
+
+	* configure.in: remove r38342's hack and check libunwind.
+
+Tue Feb 19 04:26:29 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: check whether backtrace(3) works well or not.
+
+	* vm_dump.c: set HAVE_BACKTRACE 0 if BROKEN_BACKTRACE.
+
+Mon Feb 18 16:30:18 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (IPAddr#in6_addr): Fix a typo with the closing
+	  parenthesis.
+
+Mon Feb 18 12:32:24 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (IPAddr#in6_addr): Fix the parser so that it can
+	  recognize IPv6 addresses with only one edge 16-bit piece
+	  compressed, like [::2:3:4:5:6:7:8] or [1:2:3:4:5:6:7::].
+	  [Bug #7477]
+
+Mon Feb 18 10:09:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (unexpand_shvar): regularize a shell variable by
+	  unexpanding shell variables in it.
+
+Sun Feb 17 20:55:44 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compar.c (rb_invcmp): compare by inversed comparison, with preventing
+	  from infinite recursion.  [ruby-core:52305] [Bug #7870]
+
+	* string.c (rb_str_cmp_m), time.c (time_cmp): get rid of infinite
+	  recursion.
+
+Sun Feb 17 17:23:22 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb: remove extra topdir in VPATH, which was in
+	  win32/Makefile.sub for some reason and moved from there.
+	  [ruby-dev:46998] [Bug #7864]
+
+Sun Feb 17 01:19:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* ext/psych/lib/psych/y.rb: Document Kernel#y by Adam Stankiewicz
+	  [Github tenderlove/psych#127]
+
+Sun Feb 17 00:52:14 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/mkconfig.rb: remove prefix from rubyarchdir.
+	  r39267 expands variables, it changes expansion timing,
+	  breaks RbConfig::CONFIG["includedir"] and building
+	  extension libraries with installed ruby.
+
+Sat Feb 16 20:51:17 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (ENV_IN_HEAP_P): fix off-by-one error.
+
+Sat Feb 16 20:47:16 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in (LIBRUBY_DLDFLAGS): Fix a bug where --with-opt-dir
+	  options given were not reflected to LIBRUBY_DLDFLAGS on many
+	  platforms including Linux and other GNU-based systems, NetBSD,
+	  AIX and BeOS.
+
+Sat Feb 16 20:43:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c (rsock_recvmsg): ignore truncated part of
+	  socket address returned from recvmsg().
+
+	* ext/socket/init.c (recvfrom_blocking): ignore truncated part of
+	  socket address returned from recvfrom().
+	  (rsock_s_recvfrom_nonblock): ditto.
+
+Sat Feb 16 20:05:26 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/ruby/test_thread.rb: fixed typo
+	  patched by Hiroki Matsue via https://github.com/ruby/ruby/pull/248
+
+Sat Feb 16 16:08:35 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_thread_mark): mark a working Proc of bmethod
+	  (a method defined by define_method) even if the method was removed.
+	  We could not trace working Proc object which represents the body
+	  of bmethod if the method was removed (alias/undef/overridden).
+	  Simply, it was mark miss.
+	  This patch by Kazuki Tsujimoto. [Bug #7825]
+
+	  NOTE: We can brush up this marking because we do not need to mark
+	  `me' on each living control frame. We need to mark `me's
+	  only if `me' was free'ed. This is future work after Ruby 2.0.0.
+
+	* test/ruby/test_method.rb: add a test.
+
+Sat Feb 16 15:45:56 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_binding_new_with_cfp): create binding object even if
+	  the frame is IFUNC. But return a ruby-level binding to keep
+	  compatibility.
+	  This patch fix degradation introduced from r39067.
+	  [Bug #7774] [ruby-dev:46960]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Sat Feb 16 13:40:13 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (shvar_to_cpp): do not substitute exec_prefix itself
+	  with RUBY_EXEC_PREFIX, which cause recursive definition.
+	  [ruby-core:52296] [Bug #7860]
+
+Sat Feb 16 13:13:04 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/io-console.gemspec: bump to 0.4.2.  now explicitly
+	  requires ruby 1.9.3 or later.  [Bug #7847]
+
+	* ext/io/console/console.c (console_dev): compatibility with ruby 1.8.
+
+	* ext/io/console/console.c (rawmode_opt, console_dev): compatibility
+	  with ruby 1.9.  [ruby-core:52220] [Bug #7847]
+
+Sat Feb 16 12:45:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: unexpand arch sitearch and exec_prefix values, so
+	  directly specified bindir, libdir, rubyprefix, etc can be properly
+	  substituted.  [ruby-core:52296] [Bug #7860]
+
+Sat Feb 16 12:15:20 2013  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* parse.y: add dtrace probe for symbol create.
+
+	* probes.d: ditto
+
+Sat Feb 16 09:27:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: don't test sys/feature_tests.h which is not
+	  used now.
+	  It was included in r7901 as "bug of gcc 3.0 on Solaris 8 ?".
+
+Sat Feb 16 09:24:37 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: reorder header tests to consider inclusion
+	  order in rubysocket.h.
+
+Sat Feb 16 08:42:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in, ext/socket/extconf.rb: test netinet/in_systm.h in
+	  ext/socket/extconf.rb instead of configure.in.
+
+	  Originally, netinet/in_systm.h is included for NextStep, OpenStep,
+	  and Rhapsody.  [ruby-core:1596]
+
+Sat Feb 16 07:55:40 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: don't test xti.h here.
+
+	* ext/socket/extconf.rb: test xti.h here.
+
+	  Originally, xti.h is included for IRIX [ruby-core:14447].
+
+Sat Feb 16 07:16:49 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test struct sockaddr_un and its member,
+	  sun_len.
+
+	* ext/socket/sockport.h (INIT_SOCKADDR_UN): new macro defined.
+
+	* ext/socket/socket.c (sock_s_pack_sockaddr_un): use INIT_SOCKADDR_UN.
+
+	* ext/socket/unixsocket.c (rsock_init_unixsock): ditto.
+
+	* ext/socket/raddrinfo.c (init_unix_addrinfo): ditto.
+	  (addrinfo_mload): ditto.
+
+Sat Feb 16 07:05:59 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/sockport.h (INIT_SOCKADDR_IN): don't need family
+	  argument.  it is always AF_INET.
+
+	* ext/socket/raddrinfo.c (make_inetaddr): follow INIT_SOCKADDR_IN
+	  change.
+	  (addrinfo_ipv6_to_ipv4): ditto.
+
+Sat Feb 16 04:21:07 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/extconf.rb: workaround for mswin/mingw build problem.
+	  sendmsg emulation in win32/win32.c is not enough.
+
+Sat Feb 16 00:19:20 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: use all all tested available headers for
+	  have_func.
+
+Fri Feb 15 22:21:37 2013  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in: Fix a bug introduced in r38342 that the cflagspat
+	  substitution is messed up by the way CFLAGS and optflags are
+	  modified, which affected FreeBSD and NetBSD/amd64 when
+	  configured to use libexecinfo.  This bug resulted in CFLAGS and
+	  CXXFLAGS in RbConfig::CONFIG having warnflags expanded in them,
+	  forcing third-party C/C++ extensions to follow what warnflags
+	  demands, like ANSI/ISO-C90 conformance.  ref [Bug #7101]
+
+Fri Feb 15 20:29:11 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/sockport.h (SET_SIN_LEN): defined for strict-aliasing
+	  rule.
+	  (INIT_SOCKADDR_IN): ditto.
+
+	* ext/socket/raddrinfo.c (make_inetaddr): use INIT_SOCKADDR_IN.
+	  (addrinfo_ipv6_to_ipv4): ditto.
+
+Fri Feb 15 18:24:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#try_run): bail out explicitly if cross
+	  compiling, because it cannot work of course.
+
+Fri Feb 15 12:34:58 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test struct sockaddr_storage directly.
+
+	* ext/socket/rubysocket.h: use HAVE_TYPE_STRUCT_SOCKADDR_STORAGE.
+
+Fri Feb 15 12:26:13 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/getaddrinfo.c (GET_AI): don't cast 1st argument for
+	  INIT_SOCKADDR.
+
+Fri Feb 15 08:12:11 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/sockport.h (SET_SS_LEN): removed.
+	  (SET_SIN_LEN): removed.
+	  (INIT_SOCKADDR): new macro.
+
+	* ext/socket/ancdata.c (extract_ipv6_pktinfo): use INIT_SOCKADDR.
+
+	* ext/socket/raddrinfo.c (make_inetaddr): use INIT_SOCKADDR.
+	  (addrinfo_ipv6_to_ipv4): ditto.
+
+	* ext/socket/getaddrinfo.c (GET_AI): use INIT_SOCKADDR.
+
+Fri Feb 15 07:49:27 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc.rb:  Update to release version of 4.0.0
+
+	* lib/rubygems.rb:  Update to release version of 2.0.0
+
+Fri Feb 15 07:07:27 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/sockport.h (SA_LEN): removed because unused now.
+	  (SS_LEN): ditto.
+	  (SIN_LEN): ditto.
+
+Thu Feb 14 10:45:31 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_process.rb (test_setsid): Added a workaround for
+	  MacOS X. Patch by nagachika. [Bug #7826] [ruby-core:52126]
+
+Fri Feb 15 00:15:31 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/sockport.h (VALIDATE_SOCKLEN): new macro to validate
+	  sa_len member of 4.4BSD socket address.
+
+	* ext/socket/getnameinfo.c (getnameinfo): use VALIDATE_SOCKLEN,
+	  instead of SA_LEN.
+
+	* ext/socket/socket.c (sock_s_getnameinfo): use VALIDATE_SOCKLEN
+	  instead of SS_LEN.
+
+Thu Feb 14 22:25:54 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sockaddr_len): extracted from sockaddr_obj.
+	  (sockaddr_obj): add an argument to length of socket address.
+	  (socket_s_ip_address_list): call sockaddr_obj with actual socket
+	  address length if given, use sockaddr_len otherwise.
+
+Thu Feb 14 20:11:23 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: always operate length of socket address companion with
+	  socket address.
+
+	* ext/socket/rubysocket.h (rsock_make_ipaddr): add an argument for
+	  socket address length.
+	  (rsock_ipaddr): ditto.
+
+	* ext/socket/ipsocket.c (ip_addr): pass length to rsock_ipaddr.
+	  (ip_peeraddr): ditto.
+	  (ip_s_getaddress): pass length to rsock_make_ipaddr.
+
+	* ext/socket/socket.c (make_addrinfo): pass length to rsock_ipaddr.
+	  (sock_s_getnameinfo): pass actual address length to rb_getnameinfo.
+	  (sock_s_unpack_sockaddr_in): pass length to rsock_make_ipaddr.
+
+	* ext/socket/init.c (rsock_s_recvfrom): pass length to rsock_ipaddr.
+	  (rsock_s_recvfrom_nonblock): ditto.
+
+	* ext/socket/tcpsocket.c (tcp_sockaddr): pass length to
+	  rsock_make_ipaddr.
+
+	* ext/socket/raddrinfo.c (make_ipaddr0): add an argument for socket
+	  address length.  pass the length to rb_getnameinfo.
+	  (rsock_ipaddr): ditto.
+	  (rsock_make_ipaddr): add an argument for socket address length.
+	  pass the length to make_ipaddr0.
+	  (make_inetaddr): pass length to make_ipaddr0.
+	  a local variable renamed.
+	  (host_str): a local variable renamed.
+	  (port_str): ditto.
+
+Thu Feb 14 14:31:43 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Removed OpenSSL dependency from Net::HTTP.
+
+	* test/net/http/test_http.rb:  Remove Zlib dependency from tests.
+	* test/net/http/test_http_request.rb:  ditto.
+
+Thu Feb 14 11:08:15 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (include_modules_at): detect cyclic prepend with original
+	  method table.  [ruby-core:52205] [Bug #7841]
+
+Thu Feb 14 10:30:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c: call method_removed hook on called class, not on
+	  prepending iclass.  [ruby-core:52207] [Bug #7843]
+
+Thu Feb 14 10:05:57 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http:  Do not handle Content-Encoding when the user sets
+	  Accept-Encoding.  This allows users to handle Content-Encoding for
+	  themselves.  This restores backwards-compatibility with Ruby 1.x.
+	  [ruby-trunk - Bug #7831]
+	* lib/net/http/generic_request.rb:  ditto.
+	* lib/net/http/response.rb:  ditto
+	* test/net/http/test_http.rb:  Test for the above.
+	* test/net/http/test_http_request.rb:  ditto.
+	* test/net/http/test_httpresponse.rb:  ditto.
+
+Thu Feb 14 08:18:47 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: don't define HAVE_SA_LEN and HAVE_SA_LEN.
+	  use HAVE_STRUCT_SOCKADDR_SA_LEN and HAVE_STRUCT_SOCKADDR_IN_SIN_LEN
+	  instead.
+
+Wed Feb 13 20:59:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: don't define socklen_t here, just test.
+
+	* ext/socket/rubysocket.h: define socklen_t if not available.
+
+Wed Feb 13 18:37:50 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew): skip prepending modules and return the method bound
+	  on the given class.  [ruby-core:52160] [Bug #7836]
+
+Wed Feb 13 18:11:59 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (method_original_name): new methods Method#original_name and
+	  UnboundMethod#original_name.  [ruby-core:52048] [Bug #7806]
+	  [EXPERIMENTAL]
+
+	* proc.c (method_inspect): show the given name primarily, and
+	  original_id if aliased.  [ruby-core:52048] [Bug #7806]
+
+Wed Feb 13 17:56:39 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (warnflags): disable -Werror by default unless
+	  development.  [ruby-core:52131] [Bug #7830]
+
+Wed Feb 13 06:05:52 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Return BINARY strings from Gem.gzip and Gem.gunzip.
+	  Fixes intermittent test failures.  RubyGems issue #450 by Jeremey
+	  Kemper.
+	* test/rubygems/test_gem.rb:  Test for the above.
+
+Wed Feb 13 05:49:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test functions just after struct members.
+
+Tue Feb 12 12:02:35 2013  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/json: merge JSON 1.7.7.
+	  This includes security fix. [CVE-2013-0269]
+	  https://github.com/flori/json/commit/d0a62f3ced7560daba2ad546d83f0479a5ae2cf2
+	  https://groups.google.com/d/topic/rubyonrails-security/4_YvCpLzL58/discussion
+
+Mon Feb 11 23:08:48 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: enable rb_cv_page_size_log test for MirOS BSD.
+
+Mon Feb 11 20:06:38 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: use -pthread on mirbsd*.
+
+Mon Feb 11 16:07:09 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: add SOLIBS and LIBRUBY_SO definition for mirbsd*.
+
+Mon Feb 11 13:17:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rubysitearchprefix): sitearchdir and vendorarchdir
+	  should use sitearch, not arch.  [ruby-dev:46964] [Bug #7823]
+
+	* win32/Makefile.sub (config.status): site and vendor directories
+	  should use sitearch, not arch.  [ruby-dev:46964] [Bug #7823]
+
+Mon Feb 11 12:31:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: move OS specific header/function knowledge before
+	  automatic header tests.
+
+Mon Feb 11 11:04:29 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: move the test for -march=i486 just after
+	  RUBY_UNIVERSAL_ARCH/RUBY_DEFAULT_ARCH.
+
+Sun Feb 10 23:42:26 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test structure members just after types test.
+
+Sun Feb 10 20:58:17 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test types just after headers test.
+
+Sun Feb  10 16:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/rake/doc/MIT-LICENSE: Add license file from upstream
+	* lib/rake/doc/README.rdoc: Link to license file from Rake README
+	* lib/rake/version.rb: Include README rdoc for Rake module overview
+
+Sun Feb  10 15:26:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/rake/doc/*: Sync Rake rdoc files from upstream
+
+Sun Feb 10 15:50:02 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm_exec.h (DISPATCH_ARCH_DEPEND_WAY): use __asm__ __volatile__
+	  instead of asm volatile.
+
+Sun Feb 10 15:50:02 2013  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* gc.h (SET_MACHINE_STACK_END): use __volatile__ instead of volatile.
+
+Sun Feb 10 14:25:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/rake/, lib/rake/doc/: Move Rake rdoc files to lib/rake
+
+Sun Feb 10 12:10:25 2013  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: test headers at first.
+
+Sun Feb  10 12:00:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* doc/rake/*: Removed stale Rake static files
+
+Sun Feb  10 09:10:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* lib/pp.rb, lib/prettyprint.rb: Documentation for PP and PrettyPrint
+	  Based on a patch by Vincent Batts [ruby-core:51253] [Bug #7656]
+
+Sat Feb  9 21:11:21 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: move header files check to the beginning of
+	  "header and library section".
+	  test rlim_t with sys/types.h and sys/time.h for MirOS BSD.
+	  sys/types.h and sys/time.h is guarded by #ifdef and the above
+	  move is required for this change.
+
+Sat Feb  9 17:45:58 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, version.c: prevent duplicated load paths by empty
+	  version string, it does not work right now.
+
+Sat Feb  9 17:38:41 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: fix arch parameters in help message.  [Bug #7804]
+
+Sat Feb  9 13:13:00 2013  Zachary Scott  <zachary@zacharyscott.net>
+
+	* vm_trace.c: Note about TracePoint events set, and comment on
+	  Kernel#set_trace_func to prefer new TracePoint API
+
+Sat Feb  9 10:07:47 2013  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* BSDL: update copyright notice for 2013.
+
+Sat Feb  9 09:24:38 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/package/old.rb:  Fix behavior only on ruby 1.8.
+
+	* lib/rubygems/package.rb:  Include checksums.yaml.gz signatures for
+	  verification.
+	* test/rubygems/test_gem_package.rb:  Test for the above.
+
+Sat Feb  9 01:23:24 2013  Tanaka Akira  <akr@fsij.org>
+
+	* test/fiddle/helper.rb: specify libc and libm locations for MirOS BSD.
+
+	* test/dl/test_base.rb: ditto.
+
+Fri Feb  8 23:25:33 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: change CFLAGS temporally to test
+	  ARCH_FLAG="-march=i486".
+
+Fri Feb  8 21:19:41 2013  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: don't define ARCH_FLAG="-march=i486" if it causes
+	  compilation problem.
+
+For the changes before 2.0.0, see doc/ChangeLog-2.0.0
+For the changes before 1.9.3, see doc/ChangeLog-1.9.3
+For the changes before 1.8.0, see doc/ChangeLog-1.8.0
+
+Local variables:
+coding: us-ascii
+add-log-time-format: (lambda ()
+  (let* ((time (current-time))
+	 (system-time-locale "C")
+	 (diff (+ (cadr time) 32400))
+	 (lo (% diff 65536))
+	 (hi (+ (car time) (/ diff 65536))))
+  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
+indent-tabs-mode: t
+tab-width: 8
+change-log-indent-text: 2
+end:
+vim: tabstop=8 shiftwidth=2
diff --git a/doc/ChangeLog-2.2.0 b/doc/ChangeLog-2.2.0
new file mode 100644
index 0000000000..5a7dbf826d
--- /dev/null
+++ b/doc/ChangeLog-2.2.0
@@ -0,0 +1,12157 @@
+Thu Dec 25 16:01:19 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_expand_path_internal): drop characters ignored
+	  by filesystem on Mac OS X.
+
+Thu Dec 25 15:36:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (replace_real_basename): get the real name and replace the
+	  base name with it by getattrlist(2) if available.
+	  suggested by Matthew Draper at [ruby-core:67116].  [Bug #10015]
+
+	* dir.c (glob_helper): get the real name of the whole path, not
+	  only the last name.
+
+Thu Dec 25 13:59:17 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (NET_LUID): include also ifdef.h as a workaround of
+	  a bug in mingw-w64 header.  [ruby-core:67103] [Bug #10640]
+
+Thu Dec 25 12:47:44 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (gettable_gen): warn possible reference to a local
+	  variable defined in a past scope.
+
+Thu Dec 25 10:09:14 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_dev): id_console is not a
+	  constant name, use rb_const_remove() to get rid of NameError.
+
+Thu Dec 25 09:18:55 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/resolv/test_dns.rb (TestResolvDNS#test_query_ipv4_address):
+	  set timeout for recvfrom because if client thread is crashed, it
+	  waits infinity.
+
+Thu Dec 25 08:42:11 2014  Eric Wong  <e@80x24.org>
+
+	* lib/uri/generic.rb (split_userinfo): fstring for 1-byte split
+	  (set_port): reduce bytecode size
+	  (check_path): reduce garbage via opt_str_freeze
+	  (query=): ditto
+	  (fragment=): ditto
+	  [misc #10628]
+	* lib/uri/rfc3986_parser.rb (regexp): cache as attr
+	  (initialize): setup and freeze regexp attr once
+	  (split): reduce bytecode size, use opt_str_freeze
+	  (parse): minor bytecode and garbage reduction
+	  (default_regexp): rename for initialize
+
+Wed Dec 24 20:38:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_make_pattern): restrict searching case-insensitive
+	  name from the filesystem to only last part, for the performance.
+	  [ruby-core:63591] [Bug #10015]
+
+Wed Dec 24 18:21:27 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove unused rb_objspace_t::rgengc::old_objects_at_gc_start.
+
+Wed Dec 24 13:25:22 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: require view_changesets permission.
+
+Wed Dec 24 13:00:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/downloader.rb (Downloader.download): fix the exception to
+	  re-raise.  initialize methods of subclasses of Exception may
+	  have different parameters.  [ruby-core:67086] [Bug #10639]
+
+Wed Dec 24 12:16:19 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.h (rb_data_type_t): revert r48647 and revise parent member.
+	  [ruby-core:66969] [Bug #10621]
+
+Wed Dec 24 05:40:52 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/downloader.rb: use config.guess in gcc repo.
+
+Wed Dec 24 11:50:19 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* hash.c (rb_hash_delete): return Qnil if there are no corresponding
+	  entry. [Bug #10623]
+
+	* hash.c (rb_hash_delete_entry): try delete and return Qundef if there
+	  are no corresponding entry.
+
+	* internal.h: add rb_hash_delete_entry()'s declaration.
+
+	* symbol.c: use rb_hash_delete_entry().
+
+	* thread.c: use rb_hash_delete_entry().
+
+	* ext/-test-/hash/delete.c: use rb_hash_delete_entry().
+
+Wed Dec 24 09:35:11 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/fiddle/extconf.rb: remove ffitarget.h generated by configure on
+	  mswin, because it's not normal file (cygwin's symlink) and have
+	  system attribute.
+
+Wed Dec 24 05:40:52 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/downloader.rb: support ruby 1.8.
+
+Wed Dec 24 02:44:06 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/response.rb (Net::HTTPResponse): require one or more
+	  spaces [Bug #10591].
+	  by leriksen <leif.eriksen.au@gmail.com>
+	  https://github.com/ruby/ruby/pull/782 fix GH-782
+	  NOTE: graph.facebook.com returns without SP Reason-Phrase.
+
+Wed Dec 24 02:12:22 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): VCS#branch_list expects glob a
+	  pattern string but not a regexp.  based on the patch by Vit
+	  Ondruch.  in [ruby-core:67064].  [Bug #10636]
+
+	* tool/vcs.rb (VCS::SVN#branch_list): strip newlines.
+
+	* tool/vcs.rb (VCS::GIT.get_revisions): retrieve modified time
+	  from toplevel log too.
+
+	* tool/vcs.rb (VCS::GIT#branch_list): yield for each lines.
+
+Wed Dec 24 00:23:13 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/extlibs.rb (do_extract): the pipe should be binmode.
+
+Wed Dec 24 00:21:44 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* Makefile.in, common.mk: move common-srcs to Makefile.in because
+	  it breaks build on mswin.
+
+Wed Dec 24 00:04:45 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/open-uri.rb (OpenURI.open_http): accept multiple certs path in
+	  ssl_ca_certs.
+
+	* tool/downloader.rb: use certs of rubygems for downloading gems.
+
+Tue Dec 23 22:39:11 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/extlibs: libffi-3.2.1 and patch for mswin.
+
+Tue Dec 23 22:04:38 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#query=): don't escape [\]^
+	  on both rfc2396 and rfc3986. [Bug #10619]
+
+Tue Dec 23 16:03:35 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry::Error#initialize):
+	  try en_US message if the default message cannot be encoded to
+	  locale.  [ruby-core:65295] [Bug #10300]
+
+Tue Dec 23 11:42:14 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_update_long): update huge
+	  data gradually not to exceed INT_MAX.  workaround of OpenSSL API
+	  limitation.  [ruby-core:67043] [Bug #10633]
+
+Mon Dec 22 21:30:16 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_event.rb: some tests are
+	  executed on standard Windows OS without ADO.
+
+Mon Dec 22 14:08:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (received_signal): fix condition to define.
+	  [ruby-core:67032] [Bug #10629]
+
+Sun Dec 21 10:51:51 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_event.rb: test_s_new_exception is
+	  executed on standard Windows OS without ADO.
+
+Sun Dec 21 08:35:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (jemalloc): fix option argument, should use
+	  `$withval` but not `yes` always.  [ruby-core:66994] [Bug #10625]
+
+	* configure.in (jemalloc): defer adding the liner option to get
+	  rid of linking contest against jemalloc, so that it works
+	  without runtime dynamic load path.
+
+Sat Dec 20 17:49:03 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb (Dir.mktmpdir): Accept nil again, as Ruby 2.1.
+	  [ruby-core:66943] [Bug #10616] Fixed by Alex Slynko.
+
+Sat Dec 20 11:22:58 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/depend, ext/fiddle/extconf.rb: try to build bundled
+	  libffi if existing.
+
+Sat Dec 20 05:21:00 2014  Eric Wong  <e@80x24.org>
+
+	* test/test_weakref.rb (test_repeated_object_leak): increase timeout
+	  [Bug #10618]
+
+Fri Dec 19 22:33:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: Use $(hdrdir) if possible.
+
+Fri Dec 19 22:10:00 2014  Kenta Murata  <mrkn@cookpad.com>
+
+	* ext/bigdecimal/depend: Fix dependencies to make bigdecimal
+	  installable by rubygems.
+
+	* ext/bigdecimal/bigdecimal.gemspec: version 1.2.6.
+
+Fri Dec 19 20:00:19 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRIsVALUE): put a space after string
+	  literals not to be confused with C++11 string literal suffix.
+	  https://github.com/ruby/ruby/commit/a9f3eb7#commitcomment-9040169
+
+Fri Dec 19 15:36:02 2014  Simon Genier  <simon.genier@shopify.com>
+
+	* hash.c (hash_equal): prefer true than the result of implicit
+	  conversion from int returned by rb_eql() to VALUE.  [Fix GH-789]
+
+Thu Dec 18 17:45:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_binary_elf): get rid of -e option of cat
+	  which is not available on BusyBox, use tr instead.
+	  [ruby-core:64824] [Bug #10210]
+
+Thu Dec 18 14:25:17 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* signal.c (ruby_signal): since SIGKILL is not supported by MSVCRT,
+	  should be treated before calling signal(3).
+	  [Bug #10615]
+
+Wed Dec 17 12:20:56 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (rb_method_for_self_aref, rb_method_for_self_aset):
+	  move from iseq.c to build from node instead of arrays.
+
+Wed Dec 17 10:50:09 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/gdbm/test_gdbm.rb: Added test for each_key called without a block.
+	  Patch by @joeyates [fix GH-783]
+
+Wed Dec 17 10:18:42 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): check
+	  iseq->compile_data->option->specialized_instruction for opt_* insn.
+
+	* test/ruby/test_iseq.rb: check no specialized_instructions option.
+
+Wed Dec 17 09:48:57 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (iseq_compile_each): only emit opt_str_freeze,
+	  opt_aref_with, and opt_aset_with insn when no block is given
+	  [Bug #10557] [ruby-core:66595]
+	* test/ruby/test_optimization.rb (test_block_given_aset_aref):
+	  new test for bug thanks to Bartosz Kopinski.
+	  (test_string_freeze): additional assertion for object_id
+
+Wed Dec 17 01:06:47 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32/lib/Win32API.rb (Win32API#call): need to splat.  hmm, when
+	  was this broken?
+
+Tue Dec 16 15:18:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_method_for_self_aref, rb_method_for_self_aset): call
+	  accessor functions directly, not to be affected by [] and []=
+	  methods.  [ruby-core:66846] [Bug #10601]
+
+	* struct.c (define_aref_method, define_aset_method): ditto.
+
+	* vm_insnhelper.c (rb_vm_opt_struct_aref, rb_vm_opt_struct_aset):
+	  direct accessors of Struct.
+
+Tue Dec 16 12:01:29 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking):
+	  extend timeout seconds to 60 for RGENGC_CHECK_MODE > 0 environment.
+
+Tue Dec 16 08:53:12 2014  Eric Wong  <e@80x24.org>
+
+	* tool/vcs.rb: fix Ruby 1.8 compatibility harder
+
+Tue Dec 16 07:37:18 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (GC_HEAP_FREE_SLOTS): move definition to match use order
+	  (RUBY_GC_HEAP_GROWTH_SLOTS): s/factor/number of slots/
+
+	* man/ruby.1: add section for GC environment variables
+	  [Feature #10197]
+
+Tue Dec 16 05:41:46 2014  Eric Wong  <e@80x24.org>
+
+	* tool/vcs.rb: fix Ruby 1.8 compatibility
+
+Mon Dec 15 17:51:28 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c: ObjectSpace.memsize_of(obj) returns
+	  with sizeof(RVALUE). [Bug #8984]
+
+	* gc.c (obj_memsize_of): ditto.
+
+	* NEWS: add a NEWS entry.
+
+	* test/objspace/test_objspace.rb: catch up this fix.
+
+	* test/ruby/test_file_exhaustive.rb: ditto.
+
+Mon Dec 15 16:19:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_enc_str_coderange): dummy wchar, non-endianness
+	  encoding string cannot be ascii only.
+	  [ruby-core:66835] [Bug #10598]
+
+Sun Dec 14 20:11:42 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): restore current_arg so that circular
+	  reference after a method definition is also warned.
+	  [ruby-core:61299] [Bug #9593]
+
+Sat Dec 13 20:41:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_trace.c (rb_postponed_job_flush): mask signal trap interrupt
+	  too to defer handling after finalizers finished.
+	  [ruby-core:66825] [Bug #10595]
+
+Sat Dec 13 18:33:25 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/openssl/test_pkey_ec.rb: ignored tests with old OpenSSL.
+
+Sat Dec 13 18:01:57 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (check_stack_overflow): check sp also on i386/x86_64
+	  FreeBSD.
+
+Sat Dec 13 09:58:41 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (define_final0): avoid duplicate blocks
+	  [Bug #10537]
+	* test/test_weakref.rb (test_repeated_object_leak): new test
+
+Sat Dec 13 04:59:20 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bin/erb (ERB::Main#run): get rid of shadowing outer local
+	  variables.  [ruby-core:65772] [Feature #10395]
+
+Fri Dec 12 21:56:44 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* gems/bundled_gems: Upgrade to test-unit 3.0.8. assert_throw and
+	  assert_nothing_thrown in test-unit 3.0.7 were broken by
+	  UncaughtThrowError change introduced in Ruby 2.2.0
+	  preview2. These assertions in test-unit 3.0.8 work well with
+	  UncaughtThrowError in Ruby 2.2.0 preview2.
+
+Fri Dec 12 19:48:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bin/erb (ARGV.switch, ERB::Main#run): allow variables to be set
+	  from the command line.  [ruby-core:65772] [Feature #10395]
+
+Fri Dec 12 19:31:44 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/erb.rb (ERB#lineno): accessor for line number to eval.
+
+	* lib/erb.rb (ERB#location=): setter of file name and line number.
+
+Fri Dec 12 13:09:13 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_latest_gc_info): return :state field to show current
+	  GC state (none/marking/sweeping).
+	  [Feature #10590]
+
+Fri Dec 12 10:49:18 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_crypt): check arguments more strictly.
+	  * crypt() is not for wide char strings
+	  * salt bytes should not be NUL
+
+Fri Dec 12 08:16:01 2014  Matt Hoyle  <matt@deployable.co>
+
+	* io.c (io_read) Fix spelling in docco for read.  [Fix GH-781]
+	  try > tries
+
+Thu Dec 11 19:06:01 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (class_alloc): Start from age == 2.
+	  Class and Module objects can be living long life.
+
+	* iseq.c: Same for ISeq objects.
+
+	* gc.c (RVALUE_AGE_RESET): added.
+
+	* gc.c (newobj_of): allow to generate (age != 0) objects.
+
+	* gc.c (rb_copy_wb_protected_attribute): reset age for wb unprotected
+	  objects.
+
+	* include/ruby/ruby.h: add RUBY_TYPED_PROMOTED1 as an unrecommended
+	  flag.
+
+Thu Dec 11 05:37:52 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Remove useless loop and block capture.
+	  See [#10354]
+
+Thu Dec 11 04:27:24 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: introduce new field
+	  rb_thread_t::local_storage_recursive_hash_for_trace to store
+	  recursive hash to avoid creating new recursive (nested) hashes
+	  for each trace events.
+	  [Bug #10511]
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks_orig): use it.
+
+	* cont.c: catch up this fix.
+
+	* vm.c (rb_thread_mark): ditto.
+
+Wed Dec 10 13:39:27 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (define_aref_method, define_aset_method): use iseq
+	  VALUE instead of rb_iseq_t to prevent from GC, as RB_GC_GUARD
+	  makes sense only for local variables.  [Feature #10575]
+
+Wed Dec 10 09:38:40 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (exec_recursive): use the same last method name as
+	  recursive_push in the error message when recursive_pop failed.
+	  [ruby-core:66742] [Bug #10579]
+
+Wed Dec 10 02:48:46 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/net/http/test_https.rb
+	  (TestNetHTTPS#test_certificate_verify_failure): on Windows,
+	  Errno::ECONNRESET will be raised when the verify is failure at the
+	  client side, and it'll be eaten by WEBrick.
+
+	* test/open-uri/test_ssl.rb (TestOpenURISSL#test_validation_failure):
+	  ditto.
+
+Wed Dec 10 00:42:13 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (rb_method_for_self_aref, rb_method_for_self_aset):
+	  new methods to generate bytecode for struct.c
+	  [Feature #10575]
+	* struct.c (rb_struct_ref, rb_struct_set): remove
+	  (define_aref_method, define_aset_method): new functions
+	  (setup_struct): use new functions
+	* test/ruby/test_struct.rb: add test for struct >10 members
+	* benchmark/bm_vm2_struct_big_aref_hi.rb: new benchmark
+	* benchmark/bm_vm2_struct_big_aref_lo.rb: ditto
+	* benchmark/bm_vm2_struct_big_aset.rb: ditto
+	* benchmark/bm_vm2_struct_small_aref.rb: ditto
+	* benchmark/bm_vm2_struct_small_aset.rb: ditto
+
+Tue Dec  9 20:24:41 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* string.c: [DOC] Add missing documentation around String#chomp.
+	  Patch by @stderr [ci skip][fix GH-780]
+
+Tue Dec  9 18:20:02 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c: [DOC] Revise documentation by Marcus Stollsteimer at
+	  [ruby-core:66368].  [Bug #10526]
+
+	  * #inspect: be more specific about generated string, remove
+	    obsolete example.
+	  * #nil?: use code examples instead of different call-seq's.
+	  * #tap: clarify what is yielded.
+	  * Integer(): be more specific about to_int and to_i, remove
+	    reference to Ruby 1.8.
+	  * Array(): fix error.
+	  * Class: fix variable name style and indentation in example.
+	  * improve consistency, fix typos and formatting.
+
+Tue Dec  9 12:48:32 2014  Josef Simanek  <josef.simanek@gmail.com>
+
+	* vm_eval.c (rb_eval_string_wrap): [DOC] Fix `rb_eval_string_wrap`
+	  documentation.  It is referencing `require` instead of `load`.
+	  The former does not have the optional argument.  [Fix GH-779]
+
+Tue Dec  9 10:16:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_frame_last_func): return the most recent frame method
+	  name.
+
+	* thread.c (recursive_list_access): use the last method name,
+	  instead of the current method name which can be unset in some
+	  cases, not to use a symbol by the invalid ID.
+	  [ruby-core:66742] [Bug #10579]
+
+Sun Dec  7 19:36:12 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/socket/basicsocket.c, ext/socket/sockssocket.c:
+	  remove code for $SAFE=4.
+
+Sun Dec  7 10:20:55 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc:  Update to RDoc 4.2.0.
+	* test/rdoc:  ditto.
+
+Sun Dec  7 09:52:30 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.4.5.
+	* test/rubygems:  ditto.
+
+Sat Dec  6 10:05:08 2014  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: Fix undefined variable usage & refactor/DRY
+	  code.  Patch by @aledovsky. [Fixes GH-770]
+
+	* test/net/test_imap.rb: related test.
+
+Sat Dec  6 10:09:44 2014  Eric Wong  <e@80x24.org>
+
+	* thread.c (do_select): rename parameters to avoid shadowing
+
+Sat Dec  6 09:22:45 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rake:  Update to rake 10.4.2
+	* test/rake: ditto.
+
+Sat Dec  6 06:48:03 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (rb_iseq_build_from_ary): remove misc handling
+
+Sat Dec  6 06:14:23 2014  Vit Ondruch  <vondruch@redhat.com>
+
+	* configure.in (RUBY_LIB_VERSION): Fix --with-ruby-version
+	  configuration option.  get rid of quoting in config.status.
+
+	* template/verconf.h.tmpl: quote RUBY_LIB_VERSION here.
+	  [ruby-core:66724] [Bug #10572]
+
+Sat Dec  6 04:33:52 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/pp.rb (File::Stat#pretty_print): some platforms (such as Windows)
+	  does not have major/minor parts of device.
+
+Fri Dec  5 22:43:04 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/psych/lib/psych.rb: bump version to 2.0.8
+	* ext/psych/psych.gemspec: ditto.
+	* ext/psych/psych_emitter.c: ditto.
+	* ext/psych/psych_parser.c: ditto.
+
+Fri Dec  5 17:09:09 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/option.c (inet_ntop): link aliased inet_ntop in
+	  libruby on mswin not rb_w32_inet_ntop which fails to link for
+	  unknown reason.
+
+Fri Dec  5 11:09:54 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (prepare_iseq_build): remove unused block_opt param
+	  (rb_iseq_new_with_bopt_and_opt): remove
+	  (rb_iseq_new_with_opt): inline removed function
+	  (rb_iseq_new_with_bopt): remove
+	  (iseq_load): adjust prepare_iseq_build call
+	  [Feature #10565]
+
+Fri Dec  5 09:46:05 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): fix lex_state after tLABEL_END, should
+	  be EXPR_LABELARG to be followed by "paren with arg".
+	  [ruby-core:66705] [Feature #4935]
+
+Fri Dec  5 02:27:47 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: as all extension objects including initializations
+	  of ext and enc should be linked to libruby if enable-shared,
+	  EXTOBJS should not be linked to main programs.
+	  [ruby-core:66675] [Bug #10566]
+
+Thu Dec  4 07:06:02 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (rb_iseq_build_from_exception): entry->sp is unsigned
+	  (iseq_build_callinfo_from_hash): account for kw_arg
+	  (iseq_build_from_ary_body): update for r35459
+	  (CHECK_STRING, CHECK_INTEGER): remove unused checks
+	  (int_param): new function for checking new `params' hash
+	  (iseq_build_kw): new function for loading rb_iseq_param_keyword
+	  (rb_iseq_build_from_ary): account for `misc' entry and general
+	   structure changes
+	  [Feature #8543]
+	* iseq.c (CHECK_HASH): new macro (for `misc' and `param' entries)
+	  (iseq_load): account for `misc' and `params' hashes
+	  (iseq_data_to_ary): add final opt to arg_opt_labels,
+	   fix kw support, account for unsigned entry->sp
+	* ext/-test-/iseq_load/iseq_load.c: new ext for test
+	* ext/-test-/iseq_load/extconf.rb: ditto
+	* test/-ext-/iseq_load/test_iseq_load.rb: new test
+
+Thu Dec  4 06:56:57 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_free): avoid segfault on incomplete iseq
+	* test/ruby/test_syntax.rb (test_invalid_next): new test
+	  for syntax error, not segfault
+
+Thu Dec  4 04:20:34 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (ruby_require_internal): ignore error detail, just return
+	  an error.
+
+Wed Dec  3 17:13:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (load_encoding): use rb_require_internal instead of
+	  calling rb_require_safe with protection.
+
+Wed Dec  3 16:47:35 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (register_init_ext, ruby_init_ext): defer running the
+	  registered initialization function until required, not to enable
+	  extensions which have global effects just by loading, e.g.,
+	  mathn/complex and mathn/rational.  fix `make test` with
+	  --with-static-linked-ext.
+
+	* enc/encinit.c.erb (Init_enc): initialize encdb and transdb
+	  directly.
+
+Wed Dec  3 14:51:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (ruby_require_internal): separate from rb_require_safe,
+	  not to raise exceptions.
+
+	* ruby.c (process_options): remove unnatural encoding search.
+
+Wed Dec  3 14:34:07 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (setup_fake_str): fake string does not share another
+	  string, but just should not free.
+
+Wed Dec  3 11:14:14 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (w32_spawn): `v2` is used not only for `shell` but also
+	  `cmd`, so must not free before using `cmd`.
+	  [ruby-core:66648] [Bug #10563]
+
+Wed Dec  3 09:48:57 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/racc/cparse/cparse.c (cparse_params_type): use typed data.
+
+Tue Dec  2 21:33:56 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: use typed data for com_hash.
+
+Tue Dec  2 15:30:30 2014  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* array.c (rb_ary_plus): in documentation, added note about
+	  inefficiency of repeated += operations.
+
+Tue Dec  2 07:20:21 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_data_to_ary): keep hidden variables
+	  Thanks to wanabe [ruby-core:66566]
+
+Tue Dec  2 06:46:57 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: bumping version
+
+	* ext/psych/psych.gemspec: ditto
+
+Tue Dec  2 06:34:08 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: support objects that are
+	  marshalable, but inherit from basic object.
+	  Thanks Sean Griffin <sean@thoughtbot.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: ditto
+
+	* test/psych/test_marshalable.rb: test for fix
+
+Tue Dec  2 06:32:02 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ripper_flush_string_content): preserve the dispatched
+	  results at tSTRING_CONTENT.  [ruby-dev:48714] [Bug #10437]
+
+	* parse.y (regexp_contents): check in ripper only if the whole
+	  content is a single regexp without interpolation.
+	  [ruby-dev:48714] [Bug #10437]
+
+Tue Dec  2 06:30:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_region_copy): new function to try with GC if copy
+	  failed and return the error.
+
+Tue Dec  2 04:43:08 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (CHECK_REGION_COPIED): onig_region_copy() can fail when
+	  memory exhausted but returns nothing, so check by if allocated.
+
+Tue Dec  2 02:53:00 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (magic_comment_encoding): enable in ripper, since the
+	  encoding is necessary to parse non-default encoding scripts.
+
+Tue Dec  2 02:30:25 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic.build):
+	  use hostname= to detect and wrap IPv6 hosts.
+	  Build is accepting URI components and users may not expect
+	  that a host component needs to be wrapped with square brackets
+	  since it's not providing a URI.
+	  Note: initialize with arg_check => true does not wrap IPv6 hosts.
+	  by Joe Rafaniello <jrafanie@redhat.com>
+	  https://github.com/ruby/ruby/pull/765 fix GH-765
+
+	* test/uri/test_generic.rb: Add more tests
+
+Mon Dec  1 20:01:12 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: use typed data for WIN32OLE.
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_event.c: ditto.
+
+Mon Dec  1 17:20:42 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub (verconf.h): so depends on verconf.mk, which
+	  is rebuilt by setup.mak.
+
+Mon Dec  1 11:05:46 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* proc.c: fix grammar by @BenMorganIO [fix GH-764][ci skip]
+
+Mon Dec  1 10:49:53 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* man/rake.1: Update latest man content by @aledovsky
+	  [fix GH-771][ci skip]
+
+Mon Dec  1 10:42:31 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/syntax/refinements.rdoc: refinements are no longer experimental.
+	  patch by @gaurish [fix GH-775][ci skip]
+
+Sun Nov 30 20:05:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (regexp_contents): fix a typo.  pointed out by wanabe.
+	  [ruby-dev:48741] [Bug #10543]
+
+Sun Nov 30 18:55:32 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking):
+	  Don't raise Interrupt.
+
+Sun Nov 30 17:11:05 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking): Use
+	  assert_separately.
+
+Sun Nov 30 00:02:52 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options, ruby_script): transcode script name and
+	  program name to locale encoding as well as argv.
+	  [ruby-dev:48752] [Bug #10555]
+
+	* ruby.c (translit_char_bin): should not use code page dependent
+	  CharNext on UTF-8 string.  [ruby-dev:48752] [Bug #10555]
+
+Sat Nov 29 16:53:14 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_set_argv): convert argv from UTF-8.
+
+	* win32/win32.c (rb_w32_sysinit, cmdglob, w32_cmdvector): convert
+	  wide char command line to UTF-8 argv, and glob in UTF-8 so that
+	  metacharacters would match multibyte characters.
+	  [ruby-dev:48752] [Bug #10555]
+
+Sat Nov 29 10:49:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (win32_direct_conv, rb_w32_readdir): convert UTF-8
+	  and filesystem code page by using Win32 API directly.
+
+Sat Nov 29 09:37:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (syserr_initialize): simplify message building and get
+	  rid of potential invalid byte sequence.
+
+Sat Nov 29 06:09:44 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval_error.c (error_print): respect the encoding of the message.
+
+	* io.c (rb_write_error_str): use rb_w32_write_console() on Windows
+	  if stderr is a tty.
+
+Fri Nov 28 05:10:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--with-setup): add option to select ext/Setup file.
+
+Fri Nov 28 05:02:29 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dmyenc.c (Init_enc): separate from dmyext.c for statically
+	  linked extension excluding encoding libraries.
+
+Thu Nov 27 21:58:30 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (miniprelude.c): It does not depend on prelude.rb now.
+
+Thu Nov 27 21:49:49 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: List up files built always in the source directory
+	  and source files built always in the build directory.
+
+Thu Nov 27 21:24:55 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add obsolete callcc.
+
+Thu Nov 27 19:59:49 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): remove duplicated line event.
+	  [Bug #10449]
+
+	* test/ruby/test_settracefunc.rb: add and fix tests.
+
+Thu Nov 27 19:04:50 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_args.c: fix backtrace location for keyword related exceptions.
+
+	  For example, the following program
+	    def foo(k1: 1); end # line 1
+	    foo(k2: 2)          # line 2
+	  causes "unknown keyword: k2 (ArgumentError)".
+
+	  Before this patch, the backtrace location is only line 2.
+	  However, error should be located at line 1 (over line 2 in
+	  stack trace). This patch fix this problem.
+
+	* class.c (rb_keyword_error_new): separate exception creation logic
+	  from rb_keyword_error(), to use in vm_args.c.
+
+	* vm_insnhelper.c (rb_arg_error_new): rename to rb_arity_error_new().
+
+	* vm_args.c (argument_arity_error): rename to argument_arity_error().
+
+	* vm_args.c (argument_kw_error): added to fix backtrace.
+
+	* test/ruby/test_keyword.rb: add tests.
+
+Thu Nov 27 17:31:58 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (prelude.c): no longer depends on miniruby, since not
+	  depending on rbconfig.rb.
+
+Thu Nov 27 17:12:14 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (miniprelude.c): miniruby needs no preludes.
+
+Thu Nov 27 17:10:19 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/prelude.c.tmpl: no code if no prelude code is given.
+
+Thu Nov 27 13:11:00 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (objspace_malloc_increase): enable lazy sweep on GC by malloc()
+	  (malloc_increase) to make GC incrementally.
+
+	  This change can increase memory consumption. Report us if you find
+	  any problem.
+
+Thu Nov 27 12:46:38 2014  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_s_mkutc): [DOC] Time.utc's 10 arguments form
+	  doesn't examine wday, yday, isdst and tz.
+	  (time_s_mktime): [DOC] Time.mktime's 10 arguments form
+	  doesn't examine wday, yday and tz.
+	  Suggested by naruse.
+
+Thu Nov 27 11:45:33 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_data_to_ary): add missing GC guard
+
+Thu Nov 27 10:51:59 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: add rb_thread_t::local_storage_recursive_hash
+	  to speed up Thread#[:__recursive_key__] access.
+	  [Bug #10511]
+
+	* thread.c (threadptr_local_aref): add fast path for
+	  :__recursive_data__.
+
+	* thread.c (threadptr_recursive_hash, threadptr_recursive_hash_set):
+	  add special accessor for recursive hash.
+
+	* cont.c: store/restore local_storage_recursive_hash.
+
+	* vm.c: init and mark local_storage_recursive_hash.
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks_orig): clear and restore
+	  local_storage_recursive_hash directly.
+
+Thu Nov 27 07:11:00 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (iseq_calc_param_size): hoist out of iseq_set_arguments
+
+Wed Nov 26 22:28:12 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_get_kwargs, rb_extract_keywords): export
+	  keyword argument functions.
+
+Wed Nov 26 21:18:40 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/inlinetest.rb: removed unused test helper.
+
+Wed Nov 26 20:47:28 2014  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* ext/continuation/continuation.c (Init_continuation): obsolete callcc.
+	  first step of [Feature #10548].
+
+Wed Nov 26 19:57:54 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb (test_latest_gc_info): do test separately
+	  to avoid mysterious behavior.
+
+Wed Nov 26 19:54:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (constat_reset): do nothing on non-standard
+	  console emulators.  [ruby-core:66471] [Bug #10546]
+
+Wed Nov 26 19:44:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tsort.rb: Returns an enumerator if no block is given.
+	  [ruby-core:66270] [Feature #10508] Proposed by Andrey Savchenko.
+
+Wed Nov 26 17:25:45 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_label, f_kw, formal_argument_gen): ignore invalid
+	  formal argument in keyword argument definition.
+	  [ruby-dev:48742] [Bug #10545]
+
+Wed Nov 26 15:32:06 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_set_sequence): use "nop" insn instead of
+	  "jump to next insn".
+	  https://bugs.ruby-lang.org/issues/8543#change-50085
+
+Wed Nov 26 11:01:35 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_s_compile_file): close IO when done
+
+Wed Nov 26 06:06:23 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: encdb.h and transdb.h depends on $(PREP).
+	  So prebuild files for them in tarball are useless.
+
+Wed Nov 26 02:08:44 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/make-snapshot: Don't generate enc/trans/newline.c in tarball.
+
+Wed Nov 26 00:41:44 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (prereq): Don't depends on prelude.c and golf_prelude.c.
+	  Since they depend on $(PREP) which is miniruby, they are rebuilt
+	  after miniruby is built, even if tarball contains them.
+
+Wed Nov 26 00:20:48 2014  Tanaka Akira  <akr@fsij.org>
+
+	* template/prelude.c.tmpl: Don't expand RbConfig::Config[...].
+	  It is not used now.
+
+	* common.mk: prelude.c and golf_prelude.c doesn't depend on rbconfig.
+
+Tue Nov 25 17:07:06 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: add an "Implementation changes" section.
+
+Tue Nov 25 16:09:28 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/net/http.rb:  Do not attempt SSL session resumption when the
+	  session is expired.  [Bug #10533]
+
+Tue Nov 25 15:59:46 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rake:  Update to rake 10.4.0
+	* test/rake:  ditto.
+	* NEWS:  ditto.
+
+	* test/lib/minitest/unit.rb:  Add compatibility shim for minitest 5.
+	  This only provides minitest 5 unit test naming compatibility.
+
+Tue Nov 25 15:26:33 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/vcs.rb (get_revisions): use Time.new instead of Time.mktime
+	  which does not accept UTC offset, and offset manually for older
+	  versions than 1.9.
+
+Tue Nov 25 12:14:43 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (Init_process): initialize static IDs before constant
+	  definitions.  [ruby-core:66445]
+
+Tue Nov 25 10:32:23 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (iseq_build_callinfo_from_hash): hoist out
+	  (iseq_build_from_ary_body): shorten callinfo case
+
+Mon Nov 24 23:03:21 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* gems/bundled_gems: Update to test-unit 3.0.7.
+
+Mon Nov 24 12:44:35 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (check_exec_redirect_fd, check_exec_redirect),
+	  (rb_execarg_addopt): get rid of inadvertent ID pindown.
+
+Mon Nov 24 02:03:40 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_setter): preserve encoding of global variable
+	  name in error message.
+
+Mon Nov 24 02:03:30 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_insn_operand_intern): preserve encoding of method
+	  name in CALL_INFO at disassembling.
+
+Mon Nov 24 02:02:59 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (must_respond_to): preserve encodings of variable name and
+	  class name in warning message.
+
+Sun Nov 23 10:46:23 2014  Eric Wong  <e@80x24.org>
+
+	* internal.h (struct rb_execarg): 160 => 144 bytes on x86-64
+	* ruby.c (struct load_file_arg): 48 => 40 bytes on x86-64
+	* vm_args.c (struct args_info): ditto
+
+Sun Nov 23 07:46:54 2014  Andy Maloney  <asmaloney@gmail.com>
+
+	* io.c (rb_io_sysread): Remove redundant assignment of 'n'.
+	  [Fix GH-767]
+
+Sat Nov 22 09:48:33 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/make-snapshot: Specify PWD macro for make.
+	  PWD environment variable may not exist.
+
+Fri Nov 21 11:58:58 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* lib/resolv.rb: fall back if canonicalization fails.
+	  Thanks Vit Ondruch for the patch!  [ruby-core:65836]
+
+	* test/resolv/test_dns.rb: test for patch
+
+Sat Nov 22 01:11:53 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_get, rb_mod_const_defined): ditto.
+
+	* variable.c (rb_const_missing, rb_mod_const_missing): call
+	  const_missing without new ID to get rid of inadvertent ID
+	  creation.
+
+Fri Nov 21 19:32:57 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (ext/ripper/ripper.c): revert about srcdir and top_srcdir.
+
+	* common.mk (ext/ripper/ripper.c): use $(PWD) for Unix,
+	  $(MAKEDIR) for Windows.
+
+Fri Nov 21 18:12:37 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: Refactored.
+
+Fri Nov 21 14:25:40 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (top_srcdir): added because lacking this macro
+	  causes build error at r48526.
+
+Fri Nov 21 12:00:58 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps (in_makefile): Use FILES_NEED_VPATH and
+	  FILES_CONFUSING.
+
+	* ext/objspace/extconf.rb: Add VPATH for id.h
+
+Fri Nov 21 09:10:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ext/ripper/ripper.c): fix dependencies for the case
+	  to make ripper.y and id.h under the build directory.
+
+Fri Nov 21 08:42:21 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (ext/ripper/ripper.c): id.h in VPATH may exist in the build
+	  directory.
+
+	* common.mk (ext/ripper/ripper.c): $(RM) was not defined.
+
+Fri Nov 21 00:36:09 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/x509.rb
+	  (OpenSSL::X509::Name::RFC2253DN::StringChar): get rid of a false
+	  positive assertion in ripper's test.
+
+Fri Nov 21 00:29:51 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/net/imap.rb (Net::IMAP::ResponseParser::BEG_REGEXP): no need to
+	  use embed string.
+
+Fri Nov 21 00:19:17 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/uri/mailto.rb (URI::MailTo::EMAIL_REGEXP): should escape `#`.
+
+Thu Nov 20 23:17:11 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: Insert all dependencies found by compiler.
+
+Thu Nov 20 15:51:01 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/depend (nkf.o): add nkf.c as dependency.
+	  bsdmake tries to make nkf.o with nkf-utf8/nkf.c without this.
+
+Thu Nov 20 08:54:56 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (struct parser_params): reduce size by reordering
+	  members and an unused member.
+
+Thu Nov 20 02:44:27 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Vector#independent? and associated class method
+	  patch by gogo tanaka [#10451]
+
+Thu Nov 20 02:32:34 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Vector#angle_with
+	  Patch by Egunov Dmitriy [#10442]
+
+Thu Nov 20 02:10:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ripper_flush_string_content, parser_parse_string):
+	  preserve parsed string content.  [ruby-dev:48714] [Bug #10437]
+
+	* parse.y (ripper_new_yylval): abstract function to create ripper
+	  wrapper, and make it able to hold another object.
+
+Thu Nov 20 01:00:59 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_named_capture_assign_gen): use predefined ID
+	  instead of rb_intern.
+
+Thu Nov 20 00:54:57 2014  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (ruby_init_setproctitle): Declare here.
+
+Thu Nov 20 00:26:37 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_here_document): append byte sequence directly to
+	  the delayed content instead of creating an intermediate string
+	  object.
+
+Wed Nov 19 21:11:01 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (ext/json/parser/parser.c): don't touch parse.c,
+	  ruby repo is a downstream.
+
+Wed Nov 19 20:38:11 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (ext/ripper/ripper.c): use $(PWD) to get
+	  <build-directory>/ext/ripper.
+
+Wed Nov 19 18:12:17 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/downloader.rb (RubyGems.download): Don't download gem if the
+	  version is already downloaded. A gem file is versioned and
+	  it must be identical if the version is the same.
+
+Wed Nov 19 17:59:25 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (ext/ripper/ripper.c): pass build directory as VPATH.
+
+	* ext/ripper/depend (.y.c): use VPATH for y.tab.c.
+
+Wed Nov 19 10:07:57 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/prelude.c.tmpl (Prelude#translate): strip VPATH prefix
+	  from prelude names, so that srcdir differences do not make the
+	  generated code different.
+
+Wed Nov 19 07:45:11 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write): should set the error of
+	  GetOverlappedResult()'s, not WriteFile()'s (it's always
+	  ERROR_IO_PENDING, of course).
+
+Tue Nov 18 14:16:47 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_record): add information for debug print.
+
+Wed Nov 19 04:49:07 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* debug.c (set_debug_option): need the declaration.
+
+	* debug.c (set_debug_option): use the same macro with the implementation
+	  at win32/win32.c.
+
+Wed Nov 19 04:16:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (symbol_list): fix the node type of literal symbol list
+	  with no interpolation.  [ruby-core:66343]
+
+Wed Nov 19 00:26:15 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: Sort dependencies.
+
+Wed Nov 19 00:24:18 2014  Tanaka Akira  <akr@fsij.org>
+
+	* enc/encdb.c: Include internal.h.
+
+Tue Nov 18 23:23:45 2014  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: Gather declarations in non-header files.
+
+Tue Nov 18 23:45:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* debug.c (SET_WHEN): Don't declare debug variables here.
+	  ruby_initial_gc_stress_ptr is changed int* to VALUE* at r41406.
+
+	* internal.h (ruby_initial_gc_stress_ptr): Declared.
+	  (ruby_enable_coredump): Ditto.
+
+Tue Nov 18 18:06:43 2014  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/io.h (FMODE_WSPLIT): Removed.  The write() system call
+	  is not required to split.  It was useful to avoid whole process
+	  blocking in Ruby 1.8 but not useful since write() is invoked without
+	  GVL.
+	  (FMODE_WSPLIT_INITIALIZED): Ditto.
+
+	* io.c (wsplit_p): Removed.
+	  (io_writable_length): Removed.
+	  (rb_fcntl): Don't update the removed flags.
+
+Tue Nov 18 03:23:06 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (check_setter_id): show the original argument instead
+	  of nil on TypeError.
+
+Tue Nov 18 03:20:19 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.h (is_{local,global,instance,attrset,const,class,junk}_sym):
+	  fix ID type names.
+
+Mon Nov 17 20:17:59 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_event.c: use typed data.
+
+Mon Nov 17 12:54:56 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems/*, test/rubygems/*: Update to RubyGems 2.4.4
+	  master (2f6e42e).
+
+Mon Nov 17 06:13:06 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_binding): use the original iseq on a binding from
+	  proc from method object to get the location.
+
+Sun Nov 16 19:38:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_current_receiver): new function to return the
+	  receiver in the current control frame.  [Feature #10195]
+
+Sun Nov 16 19:11:04 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout::ExitException#exception): rescue
+	  UncaughtThrowError which is specific for throw, instead of
+	  ArgumentError.
+
+Sun Nov 16 18:22:18 2014  Eric Wong  <e@80x24.org>
+
+	* tool/update-deps: warning to disable ccache
+
+Sun Nov 16 13:11:35 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk (CCAN_LIST_INCLUDES): Unused variable removed.
+	  (ENCODING_H_INCLUDES): Ditto.
+	  (PROBES_H_INCLUDES): Ditto.
+	  (VM_CORE_H_INCLUDES): Ditto.
+
+Sun Nov 16 11:07:25 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_io.rb (test_readpartial_locktmp):
+	  remove unnecessary begin/end
+
+Sun Nov 16 00:45:23 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: Specify dependencies for generated C sources.
+
+Sat Nov 15 23:10:45 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: No need to declare dependencies which
+	  will be detected by inference rules.
+
+Sat Nov 15 20:34:23 2014  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: Include ruby.h and ruby/encoding.h to be
+	  includable without prior inclusion.
+
+Sat Nov 15 20:46:44 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb
+	  (REXMLTests::TestDocument::EntityExpansionLimitTest):
+	  Group tests by general entity and parameter entity.
+
+Sat Nov 15 20:43:31 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb
+	  (REXMLTests::TestDocument::EntityExpansionLimitTest): Define
+	  test XML in each test method because (1) each XML in used only
+	  one test and (2) related data and code should be close.
+
+Sat Nov 15 20:39:06 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb
+	  (REXMLTests::TestDocument::EntityExpansionLimitTest): Use
+	  one test method for one test.
+
+Sat Nov 15 20:16:59 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb
+	  (REXMLTests::TestDocument::EntityExpansionLimitTest): Use
+	  setup and teardown instead of ensure in test.
+
+Sat Nov 15 20:11:34 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb (REXMLTests::TestDocument): Group
+	  entity expansion limit related tests.
+
+Sat Nov 15 20:09:00 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb (REXMLTests::TestDocument::BomTest):
+	  Fix wrong parent class. It doesn't need inherit tests in
+	  TestDocument class.
+
+Sat Nov 15 19:48:59 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (depend_rules): support comments during a line.
+
+	* lib/mkmf.rb (depend_rules): apply RULE_SUBST even if the dependency
+	  file contains path.
+
+Sat Nov 15 19:25:46 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: Remove comments in Dependency lines.
+	  Notified by usa.
+
+	* enc/depend: Ditto.
+
+	* ext/**/depend: Ditto.
+
+Sat Nov 15 16:28:05 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_throw_obj): throw UncaughtThrowError instead of
+	  ArgumentError.  [Feature #10480]
+
+Sat Nov 15 14:13:38 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: Extend to fix dependencies.
+
+	* common.mk: Dependencies updated by tool/update-deps.
+
+	* enc/depend: Ditto.
+
+	* ext/**/depend: Ditto.
+
+Fri Nov 14 17:36:48 2014  Tanaka Akira  <akr@fsij.org>
+
+	* tool/update-deps: Support GNU Make 4.0.
+
+Fri Nov 14 16:59:53 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (default_proc_for_compat_func): check arguments number and
+	  type, and get rid of reentering this default proc.
+
+Fri Nov 14 16:33:06 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): support compatible layer for
+	  GC.stat(symbol) type access.
+
+Fri Nov 14 16:19:08 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): add compatible layer.
+	  From Ruby 2.2, keys of GC.stat are changed [Feature #9924].
+	  To provide compatible layer, GC.stat add a default_proc
+	  (if default_proc of given Hash object is not set).
+
+	  At first use of this compatible layer of interpreter process,
+	  show a warning message like that:
+	  program: GC.stat[:total_allocated_object]
+	  warning message: "warning: GC.stat keys were changed from Ruby
+	  2.1. In this case, you refer to obsolete `total_allocated_object'
+	  (new key is `total_allocated_objects').
+	  Please check <https://bugs.ruby-lang.org/issues/9924>
+	  for more information."
+
+	  Please correct my English message :)
+
+	* hash.c (rb_hash_set_default_proc): export (in internal).
+
+	* internal.h: ditto.
+
+Fri Nov 14 10:41:25 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: guard by #if/#endif with GC_ENABLE_INCREMENTAL_MARK
+	  to hide unused codes.
+
+	* gc.c: similar to GC_ENABLE_LAZY_SWEEP.
+
+Fri Nov 14 10:23:35 2014  Eric Wong  <e@80x24.org>
+
+	* vm_eval.c (rb_eval_cmd): use pre-defined idCall
+
+Fri Nov 14 09:25:44 2014  Eric Wong  <e@80x24.org>
+
+	* vm_eval.c (rb_yield_splat): add missing GC guard
+	  [Bug #10509]
+
+Fri Nov 14 08:12:40 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gc.c: fix build error caused by implicit conversion with clang.
+
+Fri Nov 14 06:54:06 2014  Eric Wong  <e@80x24.org>
+
+	* insns.def (opt_succ): remove Time#succ optimization
+	  [Feature #10501]
+
+Fri Nov 14 05:29:46 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: refactoring.
+
+	* gc.c: use GC_ENABLE_INCREMENTAL_MARK instead of USE_RINCGC.
+
+	* gc.c (gc_start): check FORCE_MAJOR_GC.
+
+Fri Nov 14 04:51:18 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: Tuning RincGC parameters.
+
+	  Before this patch, gc_marks_step() marks (white -> grey) fixed
+	  number objects. However, this strategy does not fit practical
+	  cases, for example too slow to make progress.
+
+	  This patch changes this strategy how many objects the
+	  gc_marks_step() should handle.
+
+	  We can estimate how many times gc_marks_step() is called during
+	  this major marking (== C) with the free slot number in pooled
+	  pages. We also can estimate the living object number (== L)
+	  using last marked_slots value. We can solve this problem (how
+	  many objects should be process in gc_marks_step()) by L/C.
+
+	* gc.c (rb_objspace_t): add rb_objspace_t::rincgc::pooled_slots and
+	  step_slots.
+
+Fri Nov 14 01:26:47 2014  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (search_response): parse MODSEQ in SEARCH
+	  responses properly.  [ruby-core:64203] [Bug #10112]
+
+Fri Nov 14 01:03:17 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/envutil.rb: Moved from test/ruby/.
+
+	* test/lib/find_executable.rb: Ditto.
+
+	* test/lib/memory_status.rb: Ditto.
+
+	* test/lib/test/unit.rb: require envutil.
+
+	* test/: Don't require envutil in test files.
+
+Thu Nov 13 21:59:58 2014  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* lib/rexml/document.rb: add REXML::Document#document.
+	  reported by Tomas Hoger <thoger@redhat.com> and patched by nahi.
+
+Thu Nov 13 21:51:56 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/monitor/test_monitor.rb: Use assert_join_threads.
+
+Thu Nov 13 21:45:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl: Don't specify port number.
+
+Thu Nov 13 21:22:35 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/ssl_server.rb: Unused file removed.
+	  It is not used since r16111, Ruby 1.9.1.
+
+Thu Nov 13 18:50:14 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_ensure): remove obsolete prot_tag comment.  patch by
+	  Jack Danger at [ruby-core:66238].  [misc #10502]
+
+Thu Nov 13 18:10:38 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_vm_make_proc_lambda): similar to rb_vm_make_proc() with
+	  is_lambda argument.
+
+Thu Nov 13 12:11:18 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_fstat{,i64}): speed up.  adjusting
+	  timestamps in this function is to get rid of the side effect of
+	  ENV["TZ"].  then, if ENV["TZ"] is not set, no need to adjust.
+	  this change makes File#stat about 60% faster.
+
+Thu Nov 13 11:56:12 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c: include "internal.h" for STATIC_ASSERT.
+
+Thu Nov 13 03:56:38 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (struct heap_page): trivial packing
+	  304 => 296 bytes on x86-64
+
+Wed Nov 12 22:50:12 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gems/bundled_gems: Update to test-unit 3.0.6 and minitest 5.4.3.
+
+Wed Nov 12 22:30:52 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems/*: Update to RubyGems 2.4.3 master (7b1f684).
+
+Wed Nov 12 00:26:37 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_object.rb: Specify an exception class for rescue clause.
+
+	* test/dbm/test_dbm.rb: Ditto.
+
+	* test/gdbm/test_gdbm.rb: Ditto.
+
+	* test/sdbm/test_sdbm.rb: Ditto.
+
+Tue Nov 11 23:43:51 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/fileutils/test_fileutils.rb: Use assert_join_threads.
+
+Tue Nov 11 22:51:14 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/resolv/test_dns.rb: Use assert_join_threads.
+
+Tue Nov 11 22:33:08 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/pop/test_pop.rb: Use assert_join_threads.
+
+Tue Nov 11 22:07:20 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/http/test_https_proxy.rb: Use assert_join_threads.
+
+Tue Nov 11 18:09:11 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/prelude.c.tmpl: move from tool/compile_prelude.rb and
+	  expand by generic_erb.rb.
+
+Tue Nov 11 13:01:31 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rubygems/commands/contents_command.rb (files_in_default_gem):
+	  remove useless sort. show_files will sort the result and
+	  another branch, files_in_gem, doesn't sort.
+	  it should be removed for consistency.
+
+	* test/rubygems/test_gem_commands_contents_command.rb
+	  (test_execute_default_gem): adjust the sort algorithm with
+	  Gem::Commands::ContentsCommand#show_files, which sort items
+	  as array of [prefix, basename] not strings.
+
+Tue Nov 11 10:37:09 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (sym_equal): use rb_obj_equal().
+	  rb_obj_equal() is specially optimized in
+	  opt_eq_func()@vm_insnhelper.c.
+
+	  This fix is made from this discussion:
+	  https://www.omniref.com/ruby/2.1.4/symbols/Symbol/%3D%3D#line=8361.
+
+Tue Nov 11 09:38:55 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rdoc/known_classes.rb: reverted regression changes of
+	  rdoc known class.
+
+Tue Nov 11 00:21:50 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/imap/test_imap.rb: Don't ignore exceptions on server
+	  threads.
+
+Mon Nov 10 23:34:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/ftp/test_ftp.rb (create_ftp_server): Don't ignore
+	  exceptions on server thread.
+	  Delete read_timeout method call to fix NoMethodError.
+
+Mon Nov 10 20:20:53 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read): retry with reduced length if cannot to
+	  write any data but no error occurs.
+
+Mon Nov 10 20:04:16 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/server.rb: Setup shutdown pipe in listen method.
+
+Mon Nov 10 19:37:09 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/win32/win32.h (rb_w32_set_nonblock): new
+	  function to support nonblock-mode of pipes.
+
+	* win32/win32.c (rb_w32_read): nonblock-mode pipe returns ERROR_NO_DATA
+	  if there is no data, but also returns it if remote-end is closed.
+
+	* win32/win32.c (rb_w32_write): if cannot to write any data, it may be
+	  blocking.
+
+	* io.c (rb_io_set_nonblock): use rb_w32_set_nonblock for Windows.
+
+	* ext/io/nonblock/nonblock.c (rb_io_nonblock_set): use ruby's API when
+	  setting nonblock-mode.
+
+	* test/ruby/test_io.rb: test nonblock pipes on Windows.
+
+Mon Nov 10 17:24:34 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_getlogin): set login name encoding properly.
+	  [ruby-core:66163] [Bug #10493]
+
+Mon Nov 10 16:20:42 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems/*.rb: Update to RubyGems master(3e36528).
+
+Mon Nov 10 16:09:43 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (rb_f_kill): [DOC] mention known signal list.
+	  [ruby-core:66162] [Bug #10492]
+
+Mon Nov 10 14:17:58 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rdoc/*.rb: Update to RDoc 4.2.0.alpha(579a11c)
+
+Mon Nov 10 12:44:39 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/server.rb: Less instance variables.
+
+Mon Nov 10 12:19:43 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/server.rb (shutdown): Use close() on @shutdown_pipe_w to
+	  notify readability on the read side of the pipe.
+	  write_nonblock() is not usable for pipe on Windows.
+	  (cleanup_shutdown_pipe): Rescue IOError for @shutdown_pipe_w.close.
+
+Mon Nov 10 07:31:59 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/server.rb (initialize): Initialize shutdown pipe here
+	  to avoid race condition.
+	  (cleanup_shutdown_pipe): New private method.
+	  (cleanup_listener): Extracted from shutdown method.
+	  Call this method from start method to avoid race condition.
+
+Mon Nov 10 05:57:53 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/webrick/webrick.cgi: Don't use debug mode.
+
+	* test/webrick/webrick_long_filename.cgi: Ditto.
+
+Sun Nov  9 23:25:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call_super): allow bound proc method to call super
+	  method.
+
+	* vm_insnhelper.c (vm_yield_with_cfunc): push defined class and
+	  bound proc method entry to the control frame.
+
+Sun Nov  9 22:46:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/open-uri: Test server log in server thread.
+
+	* test/webrick: Ditto.
+
+Sun Nov  9 22:28:34 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/httpstatus.rb: require webrick/accesslog for AccessLog.
+
+Sun Nov  9 21:03:59 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/webrick: Fix the argument order of assert_equal.
+
+Sun Nov  9 20:29:01 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/webrick: Store log in an array.
+
+	* test/net/http: Ditto.
+
+	* test/open-uri: Ditto.
+
+Sun Nov  9 18:35:36 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc: Refine log test.
+
+Sun Nov  9 18:33:33 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/normalize.rb: Replaced if-else by
+	  case in self.normalized? in parallel to r48309.
+
+Sun Nov  9 18:07:00 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc: Use assert_join_threads.
+
+Sun Nov  9 14:06:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/xmlrpc: Test webrick error log is empty.
+
+Sun Nov  9 13:47:02 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (is_incremental_marking): use #if/#else because
+	  rb_objspace_t::flags::during_incremental_marking is not defined
+	  when GC_ENABLE_INCREMENTAL_MARK is 0.
+
+	* gc.c (will_be_incremental_marking, is_full_marking): similar fix.
+
+Sun Nov  9 12:16:22 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/securerandom.rb (SecureRandom.gen_random): separate
+	  implementation details and select at the load time.
+
+Sun Nov  9 12:09:38 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry::API#Enum{Value,Key):
+	  ditto.
+
+Sun Nov  9 11:48:40 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/http: Examine webrick log.
+
+Sun Nov  9 11:45:19 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gems/bundled_gems: upgraded to test-unit 3.0.5.
+
+Sun Nov  9 11:40:50 2014  Tanaka Akira  <akr@fsij.org>
+
+	* defs/known_errors.def: More errors for FreeBSD.
+
+Sun Nov  9 11:25:11 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gems/bundled_gems: upgraded to power_assert 0.2.0.
+
+Sun Nov  9 10:31:03 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/securerandom.rb: set the script encoding to make a string
+	  literal in SecureRandom::Kernel32.last_error_message single byte
+	  encoding so msg[] works in bytes, since FormatMessage() returns
+	  the size in TCHARs, not in characters.
+
+Sun Nov  9 09:50:22 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/webrick: Refine log tests.
+
+Sun Nov  9 08:58:05 2014  Tanaka Akira  <akr@fsij.org>
+
+	* defs/known_errors.def (EHWPOISON): New errno symbol.
+	  It is defined by glibc-2.16.
+
+Sun Nov  9 05:00:23 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/securerandom.rb (initialize): call the special method for Win32
+	  before checking `/dev/urandom` because we know windows doesn't have it.
+
+Sun Nov  9 04:01:46 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/securerandom.rb (SecureRandom::AdvApi32): split from `initialize`.
+	  thanks @zzak to remember it.
+
+Sun Nov  9 02:05:33 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/net/http/utils.rb: Don't connect to spawned server
+	  before actual test.
+
+Sun Nov  9 01:51:50 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* object.c (Module#const_defined?): [DOC] Revise the documentation.
+	  Patch by Xavier Noria.
+	  [Fixes GH-754] https://github.com/ruby/ruby/pull/754
+
+Sun Nov  9 00:37:44 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/webrick: Examine log and use assert_join_threads.
+
+Fri Nov  7 00:00:12 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/unicode_norm_gen.tmpl: expand kompatible_table so that
+	  recursive expansion is not needed at runtime.
+
+Thu Nov  6 23:58:40 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack): escape unprintable characters and preserve
+	  the encoding of warning message.
+
+Thu Nov  6 23:55:18 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_printable): QUOTE() should not raise an exception
+	  even on invalid byte sequence.
+
+Thu Nov  6 21:44:36 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/test_unicode_normalize.rb: Rename TestNormalize to
+	  TestUnicodeNormalize.
+	  Define constants under TestUnicodeNormalize.
+
+Thu Nov  6 21:22:59 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/open-uri/test_open-uri.rb: Check empty webrick log.
+
+Thu Nov  6 19:27:34 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (unescape_nonascii): cast -1 for the case char is unsigned char.
+	  If char is signed char, for example gcc for ARM or ppc64, it caused
+	  infinite loop.
+	  http://kmuto.jp/build-ruby/arm/ruby-trunk/log/20141106T013005Z.fail.html.gz
+	  http://rubyci.blob.core.windows.net/f19p8/ruby-trunk/log/20141106T090217Z.fail.html.gz
+
+Thu Nov  6 09:53:18 2014  Eric Wong  <e@80x24.org>
+
+	* lib/uri/rfc2396_parser.rb (initialize): reduce bytecode size
+	  2088 => 1332 bytes on 32-bit x86
+
+Thu Nov  6 08:49:49 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/normalize.rb: Comment clarification.
+
+Wed Nov  5 23:43:24 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* compile.c (compile_data_alloc): add padding when strict alignment
+	  is required for memory access. Currently, the padding is enabled
+	  only when the CPU is 32-bit SPARC and the compiler is GCC.
+	  [Bug #9681] [ruby-core:61715]
+
+	* compile.c (STRICT_ALIGNMENT): defined if strict alignment is required
+
+	* compile.c (ALIGNMENT_SIZE, ALIGNMENT_SIZE_MASK, PADDING_SIZE_MAX):
+	  new macros for alignment word size, bit mask, max size of padding.
+
+	* compile.c (calc_padding): new function to calculate padding size.
+
+Wed Nov  5 23:24:45 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/open-uri/test_open-uri.rb: Don't ignore webrick's log.
+
+Wed Nov  5 19:20:08 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/server.rb: Stop listener loop properly.
+	  [ruby-core:66085] [Bug #10478] Fixed by Charles Nutter.
+
+Wed Nov  5 17:20:29 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_disable_super, rb_enable_super): warn
+	  as deprecated at build time, instead of ignoring silently or
+	  warning at runtime only.
+
+Wed Nov  5 16:55:52 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_f_lambda): remove deprecated function, which has been
+	  unavailable from extension libraries.
+
+Wed Nov  5 16:26:58 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_clear_cache): mark as deprecated, not
+	  only warnings at runtime.
+
+Wed Nov  5 15:05:12 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_{,l,r}strip_bang): rb_str_subseq() will not
+	  NUL-terminate the result string, in the future, so it will not
+	  be needed in other cases.
+
+Wed Nov  5 14:11:30 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (lib/unicode_normalize/tables.rb): do nothing unless
+	  BASERUBY is available.  MINIRUBY cannot load extension libraries,
+	  so cannot update Unicode data.
+
+Wed Nov  5 12:13:54 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_lstrip, rb_str_strip): reduce memory copy by
+	  copying necessary part only.
+
+	* string.c (rb_str_strip_bang, rb_str_strip): ditto.
+
+Wed Nov  5 12:13:48 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_lstrip, rb_str_strip): reduce memory copy by
+	  copying necessary part only.
+
+	* string.c (rb_str_strip_bang, rb_str_strip): ditto.
+
+Wed Nov  5 10:54:19 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_lstrip_bang, rb_str_rstrip_bang): terminate
+	  wchar strings with wchar 0.
+
+Tue Nov  4 21:23:22 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32/lib/Win32API.rb: Fiddle::Importer is defined in
+	  fiddle/import.rb and it's not loaded implicitly.
+
+	* ext/win32/lib/Win32API.rb (Win32API#initialize): `import` is a string.
+
+	* ext/win32/lib/Win32API.rb (Win32API#initialize):
+	  Fiddle::Importer::CALL_TYPE_TO_ABI is private constant.
+
+Tue Nov  4 21:20:07 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/securerandom.rb (SecureRandom.random_bytes): use fiddle directly
+	  instead of using Win32API.
+
+Tue Nov  4 21:04:30 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_request_set_lockfile.rb
+	  (test_relative_path_from): driveletter support.
+
+Tue Nov  4 16:23:57 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_remote_fetcher.rb (RemoteFetcherCleanup):
+	  close all pooled connections for each tests to fix leaked file
+	  descriptors.
+
+Tue Nov  4 12:51:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (PRIsVALUE), vsnprintf.c (BSD_vfprintf): add
+	  RUBY_PRI_VALUE_MARK to reduce danger of accidental conflict with
+	  plain "%i".  binary incompatible with extension libraries using
+	  PRIsVALUE and built for 2.1 and earlier.  [EXPERIMENTAL]
+
+Tue Nov  4 12:33:44 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* object.c: fix document of Kernel#String by @suzukaze
+	  [fix GH-743][ci skip]
+
+Tue Nov  4 12:21:45 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/csv.rb: added documentation for skip_blanks option by @decasia
+	  [fix GH-744][ci skip]
+
+Tue Nov  4 12:09:18 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/open3.rb: fix code formatting in documentation by @JoshCheek
+	  [fix GH-747][ci skip]
+
+Tue Nov  4 08:57:37 2014  Eric Wong  <e@80x24.org>
+
+	* encoding.c (enc_memsize): remove unnecessary function
+	  [ruby-core:65304]
+
+Mon Nov  3 18:09:39 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/utils.rb: The default of :ignore_listener_error is
+	  changed to false.
+
+Mon Nov  3 14:42:37 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_free): resolve memory leak.
+
+Mon Nov  3 13:49:18 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_memsize): catch up recent changes.
+
+Mon Nov  3 13:38:28 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: change semantics of opt_num and opt_table.
+	  `opt_num' was the number of optional parameters + 1.
+	  `opt_table' has "opt_num" entries.
+
+	  Change them to:
+	  `opt_num' is the number of optional parameters.
+	  `opt_table' has "opt_num + 1" entries.
+
+	  This change simplify parameter fitting logics.
+
+	* compile.c: catch up this change.
+
+	* iseq.c: ditto.
+
+	* proc.c: ditto.
+
+	* vm_args.c: ditto.
+
+Mon Nov  3 11:47:44 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* NEWS: added period into Matrix section. [ci skip]
+	  [misc #10446][ruby-core:65987]
+
+Mon Nov  3 09:43:30 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* vm_args.c: fixed build error with clang
+
+Mon Nov  3 09:32:46 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/rfc3986_parser.rb (URI::RFC3986_Parser::RFC3986_URI):
+	  allow '[' and ']' for URI input (and escape). [Bug #10402]
+
+	* lib/uri/generic.rb (URI#query=): escape '[', '\', and ']'.
+
+Mon Nov  3 07:49:34 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: change iseq parameter data structure.
+	  https://bugs.ruby-lang.org/issues/10440#change-49694
+
+	  * change terminology `arg' to `param'.
+	  * move rb_iseq_t::arg_* to rb_iseq_t::param.
+	  * move rb_iseq_t::arg_size to rb_iseq_t::param::size.
+	  * move rb_iseq_t::argc to rb_iseq_t::param::lead_num.
+	  * move rb_iseq_t::arg_opts to rb_iseq_t::param::opt_num.
+	  * move rb_iseq_t::arg_rest to rb_iseq_t::param::rest_start.
+	  * move rb_iseq_t::arg_post_num to rb_iseq_t::param::post_num.
+	  * move rb_iseq_t::arg_post_start to rb_iseq_t::param::post_start.
+	  * move rb_iseq_t::arg_block to rb_iseq_t::param::block_start.
+	  * move rb_iseq_t::arg_keyword* to rb_iseq_t::param::keyword.
+	    rb_iseq_t::param::keyword is allocated only when keyword
+	    parameters are available.
+	  * introduce rb_iseq_t::param::flags to represent parameter
+	    availability. For example, rb_iseq_t::param::flags::has_kw
+	    represents that this iseq has keyword parameters and
+	    rb_iseq_t::param::keyword is allocated.
+	    We don't need to compare with -1 to check availability.
+	  * remove rb_iseq_t::arg_simple.
+
+	* compile.c: catch up this change.
+
+	* iseq.c: ditto.
+
+	* proc.c: ditto.
+
+	* vm.c, vm_args.c, vm_dump.c, vm_insnhelper.c: ditto.
+
+	* iseq.c (iseq_data_to_ary): support keyword argument.
+
+Mon Nov  3 03:39:04 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_method.rb: r48239 makes this test green.
+
+Mon Nov 03 03:02:38 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* rewrite method/block parameter fitting logic to optimize
+	  keyword arguments/parameters and a splat argument.
+	  [Feature #10440] (Details are described in this ticket)
+
+	  Most of complex part is moved to vm_args.c.
+
+	  Now, ISeq#to_a does not catch up new instruction format.
+
+	* vm_core.h: change iseq data structures.
+
+	  * introduce rb_call_info_kw_arg_t to represent keyword arguments.
+	  * add rb_call_info_t::kw_arg.
+	  * rename rb_iseq_t::arg_post_len to rb_iseq_t::arg_post_num.
+	  * rename rb_iseq_t::arg_keywords to arg_keyword_num.
+	  * rename rb_iseq_t::arg_keyword to rb_iseq_t::arg_keyword_bits.
+	    to represent keyword bitmap parameter index.
+	    This bitmap parameter shows that which keyword parameters are given
+	    or not given (0 for given).
+	    It is referred by `checkkeyword' instruction described bellow.
+	  * rename rb_iseq_t::arg_keyword_check to rb_iseq_t::arg_keyword_rest
+	    to represent keyword rest parameter index.
+	  * add rb_iseq_t::arg_keyword_default_values to represent default
+	    keyword values.
+	  * rename VM_CALL_ARGS_SKIP_SETUP to VM_CALL_ARGS_SIMPLE
+	    to represent
+	      (ci->flag & (SPLAT|BLOCKARG)) &&
+	      ci->blockiseq == NULL &&
+	      ci->kw_arg == NULL.
+
+	* vm_insnhelper.c, vm_args.c: rewrite with refactoring.
+
+	  * rewrite splat argument code.
+	  * rewrite keyword arguments/parameters code.
+	  * merge method and block parameter fitting code into one code base.
+
+	* vm.c, vm_eval.c: catch up these changes.
+
+	* compile.c (new_callinfo): callinfo requires kw_arg parameter.
+
+	* compile.c (compile_array_): check the last argument Hash object or
+	  not. If Hash object and all keys are Symbol literals, they are
+	  compiled to keyword arguments.
+
+	* insns.def (checkkeyword): add new instruction.
+	  This instruction check the availability of corresponding keyword.
+
+	  For example, a method "def foo k1: 'v1'; end" is compiled to the
+	  following instructions.
+
+	    0000 checkkeyword     2, 0  # check k1 is given.
+	    0003 branchif         9     # if given, jump to address #9
+	    0005 putstring        "v1"
+	    0007 setlocal_OP__WC__0 3   # k1 = 'v1'
+	    0009 trace            8
+	    0011 putnil
+	    0012 trace            16
+	    0014 leave
+
+	* insns.def (opt_send_simple): removed and add new instruction
+	  "opt_send_without_block".
+
+	* parse.y (new_args_tail_gen): reorder variables.
+	  Before this patch, a method "def foo(k1: 1, kr1:, k2: 2, **krest, &b)"
+	  has parameter variables "k1, kr1, k2, &b, internal_id, krest",
+	  but this patch reorders to "kr1, k1, k2, internal_id, krest, &b".
+	  (locate a block variable at last)
+
+	* parse.y (vtable_pop): added.
+	  This function remove latest `n' variables from vtable.
+
+	* iseq.c: catch up iseq data changes.
+
+	* proc.c: ditto.
+
+	* class.c (keyword_error): export as rb_keyword_error().
+
+	* common.mk: depend vm_args.c for vm.o.
+
+	* hash.c (rb_hash_has_key): export.
+
+	* internal.h: ditto.
+
+Mon Nov  3 02:35:32 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* sample/simple-bench.rb: added to measure performance of simple
+	  lines.
+
+Mon Nov  3 02:33:43 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info): show T_STRING more details.
+
+Sun Nov  2 01:30:32 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/test_ssl.rb: Don't ignore errors on listener threads,
+	  as much as possible.
+
+	* test/openssl/test_ssl_session.rb: Ditto.
+
+	* test/openssl/test_partial_record_read.rb: Ditto.
+
+Sat Nov  1 23:11:05 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/utils.rb (start_server): Don't close sockets before
+	  threads finished.
+
+Sat Nov  1 22:06:24 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/test_ssl_session.rb (test_ctx_client_session_cb): Don't
+	  ignore errors of SSL accept.
+	  (test_ctx_server_session_cb): Ditto.
+
+	* test/openssl/utils.rb (server_loop): Add ignore_ssl_accept_error
+	  argument.
+	  (start_server): Refine threads waits.
+
+	* test/ruby/envutil.rb (assert_join_threads): Show a thread before
+	  backtrace.
+
+Sat Nov  1 20:40:18 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl/utils.rb (start_server, server_loop): Use a
+	  pipe to stop server instead of shutdown/close a listening socket.
+
+Sat Nov  1 19:24:59 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_join_threads): New assertion to
+	  join multiple threads without exceptions.
+
+Sat Nov  1 17:09:32 2014  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (bary_mul_balance_with_mulfunc): Fix free work area
+	  location.
+	  [ruby-dev:48723] [Bug #10464]
+	  [ruby-core:66044] [Bug #10465]
+	  Reported by Kohji Nishihama.
+
+Sat Nov  1 15:45:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parse_atmark): mere atmark and two atmarks without
+	  succeeding identifiers are invalid as instance/class variable
+	  names.
+
+Sat Nov  1 06:31:41 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_variant.c: use typed data.
+
+Fri Oct 31 13:55:28 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/dl/*: remove DL as it is replaced by Fiddle.
+	  [Feature #5458] Thanks to Jonan Scheffler <jonanscheffler@gmail.com>
+	  for this patch
+
+	* test/dl/*: ditto.
+
+Fri Oct 31 15:26:02 2014  Charles Oliver Nutter  <headius@headius.com>
+
+	* test/openssl/test_ssl.rb: Add certificate verification chain
+	  test from JRuby community.
+
+Fri Oct 31 18:58:02 2014  Charles Oliver Nutter  <headius@headius.com>
+
+	* test/psych/test_emitter.rb: Fix line_width test...initial value
+	  is impl-specific and attr assignment always returns LHS.
+
+Fri Oct 31 22:19:30 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest.rb (Digest()): This function should now be
+	  thread-safe.  If you have a problem with regard to on-demand
+	  loading under a multi-threaded environment, preload "digest/*"
+	  modules on boot or use this method instead of directly
+	  referencing Digest::*. [Bug #9494]
+	  cf. https://github.com/aws/aws-sdk-ruby/issues/525
+
+Fri Oct 31 21:33:17 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/digest/test_digest.rb: Drop #!.  This no longer runs
+	  stand-alone because it depends on ruby/envutil.
+
+Fri Oct 31 17:22:19 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/setup.mak: $(APPEND) with some arguments insert a space before
+	  the arguments, so it causes error if the arguments are expected to be
+	  a macro definition.  this fix resolve the build error introduced at
+	  r48210.
+
+Fri Oct 31 16:47:35 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/syslog/lib/syslog/logger.rb (Syslog::Logger::VERSION): Bump
+	  the VERSION to 2.1.0. [ruby-core:64483] [Bug #10159]
+
+Fri Oct 31 16:33:46 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/net/imap.rb (Net::IMAP#fetch): [DOC] Document that
+	  Net::IMAP#fetch will return nil instead of an empty array.
+
+Fri Oct 31 12:54:43 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (.unicode-tables.time): needs Unicode files always,
+	  and should update after downloading these files.
+	  [ruby-core:66026] [Bug #10461]
+
+Fri Oct 31 10:16:42 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: reverted r48199, it's unrelated configuration.
+
+Fri Oct 31 09:58:14 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_unicode_normalize.rb: added unicode version number to
+	  test data location.
+
+Fri Oct 31 09:56:41 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .gitignore: ignored unicode data with version directories.
+
+Fri Oct 31 09:35:30 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: tweak build scripts for unicode_normalize.rb.
+
+Thu Oct 30 18:47:04 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk: fix for the case ALWAYS_UPDATE_UNICODE=no
+
+Thu Oct 30 13:23:23 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: changed Unicode data download location
+	  from latest Unicode version to Unicode 7.0.0.
+
+Thu Oct 30 11:16:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_arg{_complex,}): try conversion
+	  by to_ary for a lambda, as well as a proc.
+	  [ruby-core:65887] [Bug #9605]
+
+Wed Oct 29 21:13:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (gettable_gen): warn circular argument reference, for
+	  transition from 2.1 and earlier.  [ruby-core:65990] [Bug #10314]
+
+Wed Oct 29 20:41:01 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_params): remove unused member `cur_mid`.
+	  this has been taken over by `in_def` since 1.6.
+
+Wed Oct 29 14:44:27 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_f_load): path name needs to be transcoded to OS path
+	  encoding.  [ruby-list:49994]
+
+Wed Oct 29 11:48:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (__builtin_setjmp): disable with gcc/clang earlier
+	  than 4.3 on Mac OS X.  [ruby-core:65174] [Bug #10272]
+
+Wed Oct 29 11:43:43 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Generalize Vector#cross_product to arbitrary
+	  dimensions
+	  based on a patch by gogo tanaka [#10074]
+
+Wed Oct 29 11:43:11 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Matrix#adjucate
+	  patch by gogo tanaka [#10056]
+
+Wed Oct 29 11:42:33 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add aliases for Vector#cross & dot
+	  patch by gogo tanaka [#10352]
+
+Wed Oct 29 10:00:18 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gems/bundled_gems: Update latest version of bundled gems.
+
+Tue Oct 28 16:52:07 2014  Eric Wong  <e@80x24.org>
+
+	* signal.c (install_sighandler): remove rb_disable_interrupt and
+	  rb_enable_interrupt calls
+	  (init_sigchld): ditto
+	  (Init_signal): disable and enable interrupt once around all
+	  install_sighandler and init_sigchld to reduce syscalls at start
+	  [Feature #9345] [ruby-core:59480]
+
+Tue Oct 28 16:22:41 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options, load_file_internal2): should not
+	  require other files when dump option is given.
+	  [ruby-dev:48712] [Bug #10435]
+
+Tue Oct 28 14:51:38 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: remove apple-gcc4.2 from CC candidates.
+
+Mon Oct 27 20:13:37 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rexml/entity.rb: keep the entity size within the limitation.
+	  reported by Willis Vandevanter <will@silentrobots.com> and
+	  patched by nahi.
+
+Mon Oct 27 17:17:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (method_proc): the receiver of binding from method should
+	  be same as the receiver of the method.
+	  [ruby-core:65917] [Bug #10432]
+
+Mon Oct 27 16:26:37 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_super.rb: add a test to check block passing.
+
+Mon Oct 27 15:59:26 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: is_incremental_marking(), will_be_incremental_marking():
+	  use `&&' with GC_ENABLE_INCREMENTAL_MARK instead of using
+	  #if/#else/#endif.
+
+Mon Oct 27 13:40:11 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_sweep_rest): sweep rest pages regardless of whether
+	  lazy sweep is enabled or not.  based on the patch by Masahiro
+	  Ide at [ruby-dev:48706].  [Bug #10431]
+
+Mon Oct 27 11:18:32 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking):
+	  reduce garbage during forks
+
+Sun Oct 25 12:26:26 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* template/insns.inc.tmpl, insns_info.inc.tmpl, known_errors.inc.tmpl,
+	  minsns.inc.tmpl: fixed path of generating script.
+
+Sun Oct 26 12:24:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/ripper/lib/ripper/sexp.rb (Ripper.sexp, Ripper.sexp_raw):
+	  return nil on error.  [ruby-dev:48678] [Bug #10405]
+
+Sun Oct 25 11:24:24 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* string.c: improved comment.
+
+Sun Oct 26 07:40:11 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_val2variant, ole_invoke): refactoring.
+	  use ole_variant2variant to convert WIN32OLE_VARIANT object to
+	  VARIANT object.
+
+	* ext/win32ole/win32ole_variant.c: refactoring. add
+	  ole_variant2variant.
+	* ext/win32ole/win32ole_variant.h: ditto.
+
+Sat Oct 25 22:28:17 2014  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (io_binwrite_string): Test writev() failure.
+
+Sat Oct 25 20:19:19 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* test/test-unicode_normalize.rb: added test_us_ascii.
+
+Sat Oct 25 20:09:09 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/normalize.rb: added US_ASCII
+	  as trivially supported encoding (is always normalized,
+	  and may appear mixed in with UTF-8 or other Unicode
+	  encodings).
+
+Sat Oct 25 20:01:01 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* test/test-unicode_normalize.rb: added
+	  test_raise_exception_for_non_unicode_encoding.
+
+Sat Oct 25 19:30:30 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* test/test-unicode_normalize.rb: removed unused function.
+
+Sat Oct 25 18:41:41 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* template/unicode_norm_gen.tmpl: Adjusted name of generating file.
+
+Fri Oct 24 22:49:42 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): dispatch newline and space at fluent
+	  interface, so that the following identifier does not include the
+	  space.  [ruby-dev:48684] [Bug #10411]
+
+Fri Oct 24 20:41:36 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* signal.c (check_reserved_signal_): fix write count since r47991.
+
+Thu Oct 23 21:42:54 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_delete): now does not call the block given to
+	  the current method.  [ruby-core:65861] [Bug #10413]
+
+Thu Oct 23 19:13:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_make): warn redefinition only for
+	  already defined methods, but not for undefined methods.
+	  [ruby-dev:48691] [Bug #10421]
+
+Thu Oct 23 17:19:04 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/source.rb (REXML::IOSource#encoding_updated): Fix a
+	  bug that can't parse XML correctly when
+	  Encoding.default_internal is different with XML
+	  encoding. REXML::Source converts XML encoding on read. So IO
+	  should not convert XML encoding.
+	  Based on patch by NAKAMURA Usaku.
+	  [ruby-dev:48686] [Bug #10418]
+
+	* test/rexml/test_encoding.rb
+	  (REXMLTests::EncodingTester#test_parse_utf16_with_utf8_default_internal):
+	  Add the for the above case.
+
+Thu Oct 23 16:29:02 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_encoding.rb
+	  (REXMLTests::EncodingTester#test_parse_utf16): Use meaningful
+	  test name. "ticket" in the old test name means the ticket in
+	  REXML's issue tracker. The REXML's issue tracker was gone. So
+	  "ticket" is meaningless.
+
+	* test/rexml/data/ticket_110_utf16.xml: Rename to ...
+	* test/rexml/data/utf16.xml: ... this.
+
+Thu Oct 23 16:18:11 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_encoding.rb
+	  (REXMLTests::EncodingTester#test_ticket_110): Fix expected and
+	  actual order.
+	  Patch by NAKAMURA Usaku. Thanks!!!
+
+Thu Oct 23 10:47:16 2014  Eric Wong  <e@80x24.org>
+
+	* insns.def (getlocal,setlocal): add comment to def/opt_operand.def
+
+Thu Oct 23 10:22:41 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking):
+	  use IO#wait_readable instead of timeout
+
+Thu Oct 23 10:22:24 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking):
+	  ensure exit! during fork failure
+
+Thu Oct 23 10:21:21 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* NEWS: Added String#unicode_normalize(|!|d?) [ci skip]
+
+Thu Oct 23 03:41:51 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (unknown_keyword_error): delete expected keywords
+	  directly from raw table, so that the given block is not called.
+	  [ruby-core:65837] [Bug #10413]
+
+Thu Oct 23 02:33:01 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (update-unicode): invert dependency to run every times.
+	  [ruby-core:65842] [Bug #10415]
+
+	* common.mk (after-update): update files under source tree.
+	  [ruby-core:65840] [Bug #10414]
+
+Wed Oct 22 22:38:59 2014  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (DEFAULT_PARAMS): override
+	  options even if OpenSSL::SSL::OP_NO_SSLv3 is not defined.
+	  this is pointed out by Stephen Touset. [ruby-core:65711] [Bug #9424]
+
+Wed Oct 22 21:31:56 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (prelude.c): add dependency to LIB_SRCS because
+	  enc/prelude.rb requires lib/unicode_normalize.rb, and it's also
+	  requires lib/unicode_normalize/tables.rb (=LIB_SRCS).
+
+Wed Oct 22 21:09:51 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* configure.in (nacl_cv_cpu_nick): fix typo in PNaCl.
+	  (XCFLAGS) Add -isystem flag to pnacl and nacl-newlib
+	  (CXX): added
+
+	* nacl/GNUmakefile.in (CXX): Added
+	  (PPROGRAM): Use clang++ instead of clang because libnacl_io
+	  depends on c++ std lib.
+
+Wed Oct 22 21:07:32 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* common.mk (build-ext): avoid trying to build dynamic libraries
+	  if configured --with-static-linked-ext.
+
+Wed Oct 22 20:33:33 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Fixed grammar in comment [ci skip]
+
+Wed Oct 22 19:18:18 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/unicode_norm_gen.rb: Fixed escaping of backslash and
+	  double quote ('\\\&' -> "\\\\\\\&"; double quoted string
+	  is needed to make \& mean last match; double double
+	  backslashes are needed because of two layers of escaping).
+
+Wed Oct 22 18:13:29 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/test-unicode_normalize.rb: as often said, ruby is sometimes built
+	  at non-srcdir.
+
+Wed Oct 22 18:12:12 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* test/test-unicode_normalize.rb: Adjusted path for test
+	  data file (now ../enc/unicode/data/NormalizationTest.txt).
+
+Wed Oct 22 18:07:07 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* test/test-unicode_normalize.rb: Removed explicit require,
+	  changed method names, adjusted copyright.
+
+Wed Oct 22 18:00:00 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* test/test-unicode_normalize.rb: Importing from
+	  https://github.com/duerst/eprun/blob/master/test/test_normalize.rb.
+	  (removing trailing whitespace, fixing EOLs and adding EOL property)
+
+Wed Oct 22 08:21:09 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* nacl/pepper_main.c (Instance_DidCreate): mount devfs and rebind fd 0
+	  .. 2 so that stderr goes to the console of the browser.
+
+Wed Oct 22 03:47:43 2014  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/etc/etc.c (etc_nprocessors_affin): maximum "n" should be 16384.
+
+Wed Oct 22 03:37:00 2014  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/etc/etc.c (etc_nprocessors_affin): minor spell fix.
+
+Wed Oct 22 03:33:58 2014  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/etc/etc.c (etc_nprocessors_affin): optimize memory usage a
+	  bit. Typical rubyist never use 8k cpus machine.
+
+Wed Oct 22 00:01:09 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* configure.in (XCFLAGS): Add include path for NaCl libraries.
+	  (XLDFLAGS): ditto.
+	  (NACL_LIB_PATH): new substitution
+
+	* nacl/nacl-config.rb: support NACL_LIB_PATH
+
+	* nacl/package.rb: ditto.
+
+	* nacl/pepper_main.c: replace old implementations with nacl_io.
+
+	* nacl/GNUmakefile.in: link nacl_io to pepper_ruby
+
+	* ruby.c (rb_load_file): remove __attribute__((weak)) because the old
+	  override hack was replaced with nacl_io.
+
+	* file.c (rb_file_load_ok): ditto.
+
+Tue Oct 21 17:32:32 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Adding explicit creation of directory
+	  enc/unicode/data because git doesn't handle empty
+	  directories. [patch by Masahiro Ide, filed with r48073]
+
+Tue Oct 21 17:12:12 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/tables.rb: Committing to make version
+	  update easier and more predictable, and reducing compilation
+	  time.
+
+Tue Oct 21 15:56:56 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/normalize.rb: Added comment to point to
+	  relevant portion of Unicode standard for Hangul (de)composition
+	  identifiers and algorithm.
+
+Tue Oct 21 11:49:16 2014  Andreas Schwab  <schwab@suse.de>
+
+	* gc.c (mark_current_machine_context) [__mc68000__]: Update stack
+	  marking.
+	  (rb_gc_mark_machine_stack) [__mc68000__]: Also handle it here.
+
+Mon Oct 20 23:59:38 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* enc/prelude.rb: we sometimes run ruby without library path (especially
+	  for test), so should permit to run ruby if unicode_normalize.rb is
+	  missing.
+
+Mon Oct 20 23:57:58 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (lib/unicode_normalize/tables.rb): enable running (n)make
+	  in non-srcdir.
+
+Mon Oct 20 23:58:04 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/downloader.rb: add -a option to always download regardless
+	  existing files.
+
+Mon Oct 20 23:18:18 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb: no need to convert path separator for COPY because it's
+	  ruby -run cp and it can treat '/' on any platform.
+
+Mon Oct 20 19:54:54 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* config.mk: Added missing data files as targets for
+	  prerequisite update_unicode.
+
+Mon Oct 20 19:06:06 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: revert r48046. The s in sIndex
+	  is not hungarian notation. The variable name sIndex is
+	  directly taken from the relevant part of the Unicode
+	  Standard, where it is written SIndex and stands for
+	  'syllable index'. See pp. 144/145 of
+	  http://www.unicode.org/versions/Unicode7.0.0/ch03.pdf.
+
+Mon Oct 20 12:46:46 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: removing unnecessary 'self'.
+
+Mon Oct 20 12:37:37 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: change method names
+	  in commented-out code. Followup to r48027.
+
+Mon Oct 20 02:23:27 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_ivar_get), vm_insnhelper.c (vm_getivar): improve
+	  instance variable retrieval performance by checking ruby_verbose
+	  before call of rb_warning and evaluation of its argument.
+	  [ruby-core:65786] [Feature #10396]
+
+Sun Oct 19 23:31:29 2014  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* lib/unicode_normalize.rb: (unicode_normalize!): change method name.
+	  catch up the method name change at r48014. [Feature #10084]
+
+Sun Oct 19 20:05:58 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/tk/lib/tkextlib/tile/treeview.rb: fix syntax error.
+
+Sun Oct 19 18:39:39 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* enc/prelude.rb: Added automatic loading of
+	  lib/unicode_normalize.rb. This makes sure that all
+	  the methods that are available on String are
+	  available without explicit require.
+
+Sun Oct 19 18:35:35 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/normalize.rb: Added a missing
+	  file extension in require statement.
+
+Sun Oct 19 18:13:13 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Added a rule to generate
+	  lib/unicode_normalize/tables.rb. This rule still
+	  needs to be integrated into the overall make process.
+
+Sun Oct 19 17:53:53 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: Changed to dynamic
+	  loading of actual normalization code and tables.
+
+Sun Oct 19 17:37:37 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: Small documentation fix.
+
+Sun Oct 19 17:26:26 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: Added documentation.
+
+Sun Oct 19 11:09:09 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/unicode_norm_gen.rb, lib/unicode_normalize.rb:
+	  File name change from lib/unicode_normalize/normalize_tables.rb
+	  to lib/unicode_normalize/tables.rb.
+
+Sun Oct 19 10:12:12 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: Changing method names, see
+	  https://bugs.ruby-lang.org/issues/10084#note-7
+
+Sun Oct 19 10:10:10 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: Changing module name.
+
+Sun Oct 19 10:08:08 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: Changing require statement,
+	  adjusting copyright.
+
+Sun Oct 19 10:04:04 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize.rb: Importing from
+	  https://github.com/duerst/eprun/blob/master/lib/string_normalize.rb.
+	  (removing trailing whitespace, fixing EOLs and adding EOL property)
+
+Sun Oct 19 09:56:56 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/unicode_norm_gen.rb: Changed module name.
+
+Sun Oct 19 09:48:48 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/normalize.rb: Changed module name,
+	  adjusted copyright.
+
+Sun Oct 19 09:38:38 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/normalize.rb: Importing from
+	  https://github.com/duerst/eprun/blob/master/lib/normalize.rb.
+
+Sat Oct 18 20:40:52 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_core.h, proc.c, vm_backtrace.c, vm_trace.c:
+	  remove rb_binding_new_with_cfp, and use rb_vm_make_binding instead.
+
+Sat Oct 18 20:38:48 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_core.h, vm.c, proc.c: fix GC mark miss on bindings.
+	  [ruby-dev:48616] [Bug #10368]
+
+	* test/ruby/test_eval.rb: add a test code.
+
+Fri Oct 17 22:47:11 2014  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_unpack): Add casts for char references for 'u'.
+	  Fix line ending recognition algorithm.
+
+Fri Oct 17 21:49:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (pack_unpack): Add casts for char references for 'b' and 'h'.
+
+Fri Oct 17 17:50:10 2014  Tanaka Akira  <akr@fsij.org>
+
+	* Avoid undefined behaviors found by gcc -fsanitize=undefined.
+	  gcc (Debian 4.9.1-16) 4.9.1
+
+	* string.c (rb_str_sum): Avoid undefined behavior.
+
+Fri Oct 17 17:43:50 2014  Tanaka Akira  <akr@fsij.org>
+
+	* Avoid undefined behaviors found by gcc -fsanitize=undefined.
+	  gcc (Debian 4.9.1-16) 4.9.1
+
+	* include/ruby/ruby.h (INT2FIX): Avoid undefined behavior.
+
+	* node.h (nd_set_line): Ditto.
+
+	* pack.c (encodes): Ditto.
+	  (pack_unpack): Ditto.
+
+	* regint.h (BIT_STATUS_AT): Ditto.
+	  (BS_BIT): Ditto.
+
+	* time.c (time_mdump): Ditto.
+	  (time_mload): Ditto.
+
+	* vm_core.h (VM_FRAME_MAGIC_MASK): Ditto.
+
+	* vm_trace.c (recalc_add_ruby_vm_event_flags): Ditto.
+
+Fri Oct 17 15:06:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (unescape_nonascii): make dynamically compiled US-ASCII
+	  regexps ASCII-8BIT encoding if binary (hexadecimal, control,
+	  meta) escapes are contained, as well as literal regexps.
+	  [ruby-dev:48626] [Bug #10382]
+
+Fri Oct 17 03:05:08 2014  Eric Wong  <e@80x24.org>
+
+	* test/-ext-/bug_reporter/test_bug_reporter.rb
+	  (test_bug_reporter_add): revert r47972
+	* test/ruby/test_rubyoptions.rb (test_segv_test): revert r47971
+	  [ruby-core:65764]
+
+Thu Oct 16 23:17:40 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (rb_sigaltstack_size): double default size to get rid
+	  of heap corruption by alternate stack overflow in SEGV handler.
+	  typically happened at fprintf() in control_frame_dump().
+
+Thu Oct 16 22:43:12 2014  Tanaka Akira  <akr@fsij.org>
+
+	* vm_backtrace.c (id2str): Fix a variable name.
+	  [ruby-dev:48642] [Bug #10389]
+
+Thu Oct 16 20:01:26 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_pattern_list): Show number of matched
+	  patterns and characters.
+
+Thu Oct 16 16:26:09 2014  Eric Wong  <e@80x24.org>
+
+	* cont.c (fiber_store): fix WIN32 fibers
+	  [ruby-core:65745] [ruby-core:65758]
+
+Thu Oct 16 15:05:07 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_here_document): do not append already appended
+	  and disposed code fragment.  [ruby-dev:48647] [Bug #10392]
+
+Thu Oct 16 10:35:33 2014  Eric Wong  <e@80x24.org>
+
+	* test/-ext-/bug_reporter/test_bug_reporter.rb
+	  (test_bug_reporter_add): fix race
+
+Thu Oct 16 10:09:02 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_rubyoptions.rb (test_segv_test): fix race
+
+Thu Oct 16 09:17:48 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cont.c (rb_fiber_t): fix compile error caused by move to
+	  vm_core.h at r47964.  [Feature #10341]
+
+Thu Oct 16 08:58:11 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking):
+	  avoid killing wrong parent
+
+Thu Oct 16 08:40:04 2014  Eric Wong  <e@80x24.org>
+
+	* cont.c (fiber_store): restore references to next_fib (fix typo)
+
+Thu Oct 16 08:26:08 2014  Eric Wong  <e@80x24.org>
+
+	* cont.c (fiber_store): remove references to nextfib
+	  fix build when FIBER_USE_NATIVE is 0
+
+Thu Oct 16 06:51:35 2014  Knut Franke  <Knut.Franke@gmx.de>
+
+	* vm_core.h: declare rb_fiber_t typedef
+	  (rb_thread_t): fiber and root_fiber become rb_fiber_t * (from VALUE)
+	* vm.c (rb_thread_mark): use rb_fiber_mark_self
+	* cont.c (rb_fiber_t): prev becomes rb_fiber_t * (from VALUE)
+	  (cont_mark, cont_free): simplify conditions
+	  (rb_fiber_mark_self): new function
+	  (fiber_mark): use rb_fiber_mark_self
+	  (cont_save_thread, cont_restore_thread): inline
+	  (cont_restore_thread): simplify
+	  (fiber_setcontext): simplify conditions
+	  (rb_cont_call): remove dereference
+	  (fiber_t_alloc): update for rb_fiber_t->prev type change
+	  (rb_fiber_start): ditto
+	  (fiber_current): extract from rb_fiber_current
+	  (return_fiber): move, simplify type checks
+	  (rb_fiber_current): use fiber_current
+	  (fiber_store): simplify type checks
+	  (fiber_switch): ditto, simplify call to fiber_setcontext,
+	   use fiber_current
+	  (rb_fiber_transfer): update for type changes
+	  (rb_fiber_terminate): move, use fiber_switch
+	  (rb_fiber_resume): update for type changes
+	  (rb_fiber_reset_root_local_storage): ditto
+	  (rb_fiber_yield): use rb_fiber_switch instead of rb_fiber_transfer
+	  (rb_fiber_m_transfer): ditto
+	  [ruby-core:65518] [Feature #10341]
+
+Thu Oct 16 06:25:29 2014  Knut Franke  <Knut.Franke@gmx.de>
+
+	* cont.c (rb_context_t): comment on saved_thread
+	  (cont_save_thread): sparse copy
+	  (cont_init): copy extra fields
+	  (fiber_init): use current thread VM stack size
+	  [ruby-core:65518] [Feature #10341]
+
+Thu Oct 16 06:13:09 2014  Knut Franke  <Knut.Franke@gmx.de>
+
+	* cont.c (cont_capture): remove unnecessary variable
+	  [ruby-core:65518] [Feature #10341]
+
+Thu Oct 16 05:02:31 2014  Knut Franke  <Knut.Franke@gmx.de>
+
+	* cont.c (fiber_store, fiber_switch): simplify
+	  [ruby-core:65518] [Feature #10341]
+
+Thu Oct 16 04:28:41 2014  Knut Franke  <Knut.Franke@gmx.de>
+
+	* cont.c (rb_fiber_t): remove prev_fiber/next_fiber
+	  (fiber_link_join, fiber_link_remove): remove functions
+	  (fiber_free, fiber_init, root_fiber_alloc):
+	   remove references to removed fields and functions
+	  [ruby-core:65518] [Feature #10341]
+
+Wed Oct 15 22:08:37 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/etc/etc.c (etc_nprocessors_affin): Test CPU_ALLOC availability.
+	  CentOS 5 don't have CPU_ALLOC().
+
+Wed Oct 15 18:26:19 2014  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/etc/etc.c (etc_nprocessors_affinity): use sched_getaffinity
+	  for getting precious number of available cpus.
+
+	* ext/etc/etc.c (etc_nprocessors): use etc_nprocessors_affinity if
+	  possible.
+
+	  [Feature #10267] etc-nprocessors-kosaki2.patch
+
+Wed Oct 15 17:53:28 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_pattern_list) Renamed from
+	  assert_regexp_list.
+	  Show multiline string in multi lines.
+
+	* test/-ext-/bug_reporter/test_bug_reporter.rb: Use
+	  assert_pattern_list.
+
+Wed Oct 15 12:26:58 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (assert_regexp_list): New assertion method.
+
+	* test/ruby/test_rubyoptions.rb: Use assert_regexp_list.
+
+Wed Oct 15 07:21:09 2014  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c: min(n) drops elements bigger than the n-th maximum element.
+	  (struct nmin_data): New field to record the n-th maximum element, limit
+	  (nmin_filter): Update limit field.
+	  (nmin_i): Drop too big elements.
+	  (nmin_run): Initialize limit field.
+
+Wed Oct 15 07:00:14 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_optimization.rb (test_string_size): new test
+
+Wed Oct 15 06:51:13 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_optimization.rb (test_string_eq_neq): new test
+	  (test_string_ltlt): ditto
+
+Wed Oct 15 06:50:29 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_optimization.rb (test_hash_aset_with):
+	  assert assignment
+
+Wed Oct 15 04:56:27 2014  Zachary Scott  <e@zzak.io>
+
+	* gc.c (rb_obj_id): [DOC] Fix typo, clean up sentence, and wrap cols
+
+Wed Oct 15 04:53:30 2014  Zachary Scott  <e@zzak.io>
+
+	* error.c: [DOC] Fix case of type in exception message by @tricknotes
+	  [Fixes GH-740] https://github.com/ruby/ruby/pull/740
+
+	* object.c: ditto
+
+Tue Oct 14 21:39:16 2014  Vit Ondruch  <vondruch@redhat.com>
+
+	* tool/rbinstall.rb (gem): Fix permissions of bundled gems
+	  specification files.  [ruby-core:65700] [Bug #10383]
+
+Tue Oct 14 19:15:31 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_record.c: use typed data.
+
+Tue Oct 14 16:23:12 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.c (global_symbols): make ids two-dimensional array of
+	  strings and symbols, for write-barrier.
+
+	* symbol.c (global_symbols): make IDs immortal always, instead
+	  of treating dynamic symbols as IDs.
+
+	* iseq.c, marshal.c, string.c: use rb_str_intern instead of
+	  rb_str_dynamic_intern.
+
+	* symbol.c (rb_str_intern): rename rb_str_dynamic_intern.
+
+Tue Oct 14 10:19:10 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_optimization.rb (test_string_freeze): new test
+	  (test_hash_aref_with): ditto
+	  (test_hash_aset_with): ditto
+
+Tue Oct 14 01:27:54 2014  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (nmin_run): max(n) and max_by(n) returns an array in
+	  descending order.
+	  [ruby-core:65452] Suggested by David Grayson.
+
+Mon Oct 13 20:44:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (update-gems): chdir to the target directory and then
+	  add the tool directory to load paths, for older BASERUBY.
+	  [Bug #10372][ruby-core:65630]
+
+Mon Oct 13 17:53:01 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/xmlrpc/parser.rb: added new parser class using libxml-ruby gem.
+	  [Feature #9379][ruby-core:59633]
+	* lib/xmlrpc/config.rb: ditto.
+
+Mon Oct 13 16:32:56 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/find.rb (Find.find): Call to_path for arguments to obtain
+	  strings.
+	  [ruby-core:63713] [Bug #10035] Reported by Herwin.
+
+Mon Oct 13 15:42:25 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* common.mk: use relative load path for bundled_gems directory.
+	  [Bug #10372][ruby-core:65630]
+
+Mon Oct 13 08:44:06 2014  Zachary Scott  <e@zzak.io>
+
+	* README.EXT: [DOC] fix example missing typedef with patch by
+	  @steveklabnik [Fixes GH-739] https://github.com/ruby/ruby/pull/739
+
+	* README.EXT.ja: ditto.
+
+Mon Oct 13 06:52:09 2014  Eric Wong  <e@80x24.org>
+
+	* array.c (ary_recycle_hash): add RB_GC_GUARD
+	  (rb_ary_diff): remove volatile
+	  [Bug #10369]
+
+Mon Oct 13 03:20:23 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/date/date_core.c: [DOC] Clean up whitespace, examples, and typos
+	  in date_core based on a patch by @vipulnsward [Fixes GH-724]
+	  https://github.com/ruby/ruby/pull/724
+
+Mon Oct 13 02:39:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (remove_duplicate_keys): should not simply eliminate all
+	  value nodes, which may have side effects.
+	  [ruby-core:65625] [Bug #10315]
+
+Sun Oct 12 10:39:16 2014  Zachary Scott  <e@zzak.io>
+
+	* vm.c: [DOC] fix typo by @yui-knk [Fixes GH-738]
+	  https://github.com/ruby/ruby/pull/738
+
+Sun Oct 12 09:24:15 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/test/unit.rb: Hide skips by default.
+
+Sun Oct 12 01:37:11 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* file.c: include sys/time.h only if HAVE_SYS_TIME_H
+
+Sat Oct 11 22:29:40 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* file.c (HAVE_UTIMENSAT): disabled for NativeClient.
+	  Fixes build error.
+
+Sat Oct 11 22:11:58 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* ext/extmk.rb: generates the rule for extinit.$(OBJEXT).
+	  extinit.$(OBJEXT) used to be generated by the builtin rule, thus
+	  didn't accept custom $(CC) and caused linkage error for cross
+	  compiling.
+
+Sat Oct 11 18:46:50 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* include/ruby/intern.h (rb_fd_select): declare struct timeval, or the
+	  struct gets local to the function in C99.
+
+	* file.c (#include): add nacl/stat.h for PNaCl.
+	  (utimes): added a declaration for PNaCl.
+	  (stat_atimespec): stat::st_atimensec is long long but
+	  timespec::tv_nsec is long in PNaCl.
+	  (stat_mtimespec, stat_ctimespec): ditto.
+	  (rb_group_member): disable getgroups unless HAVE_GETGROUPS.
+	  (eaccess): unify the fallback to generic defined(USE_GETEUID).
+
+	* io.c: include sys/time.h for struct timeval.
+	  (rb_close_before_exec): nothing we can do if F_GETFD is not
+	  available.
+	  (ioctl): pnacl newlib actually doesn't have ioctl.
+
+	* process.c (maxgroups): it is used iff
+	   defined(_SC_NGROUPS_MAX) || defined(NGROUPS_MAX) but not
+	   defined(HAVE_GETGROUPS) || defined(HAVE_SETGROUPS).
+	  (obj2gid): fail unless the object is a Fixnum if getgrnam is not
+	  available.
+	  (disable_child_handler_fork_child): sigaction is not available in
+	  PNaCl newlib.
+
+	* configure.in (warnflags, strict_warnflags): avoid -ansi for strlcpy.
+	  (rb_cv_gcc_atomic_builtins): also check
+	  __atomic_or_etch because it is used in ruby_atomic.h.
+	  (rb_cv_gcc_sync_builtins): ditto.
+	  (HAVE_GETGRNAM): added.
+
+Sat Oct 11 15:32:08 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (iseq_build_from_ary_exception): move RB_GC_GUARD
+	  (iseq_build_from_ary_body): use PRIsVALUE instead of RB_GC_GUARD
+
+Sat Oct 11 14:57:08 2014  Eric Wong  <e@80x24.org>
+
+	* string.c (rb_str_intern): remove unnecessary RB_GC_GUARD
+
+Sat Oct 11 13:47:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (remove_duplicate_keys): remove duplicate literal keys,
+	  i.e., symbols and strings.  [ruby-core:65368] [Bug #10315]
+
+	* vm.c (kwmerge_i): override existing keys by new keys.
+	  [ruby-core:65368] [Bug #10315]
+
+	* parse.y (assocs): concatenate splatted literal hashes.  the
+	  former key has precedence even if duplicated literal keys
+	  follow.  [ruby-core:65368] [Bug #10315]
+
+Sat Oct 11 12:27:03 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* configure.in (RUBY_NACL): automatically locate pnacl-clang.
+	  (RUBY_PLATFORM): pnacl instead of le32-nacl.
+
+Sat Oct 11 11:27:14 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* io.c: fix issues in the last two commits. don't disable cloexec for
+	  platforms other than NativeClient.
+
+	* ChangeLog: ditto. add entries for the last two commits.
+
+Sat Oct 11 11:12:00 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* signal.c (install_signalhandler, init_sigchld): allow failure because it
+	  always fails with ENOSYS on NaCl.
+
+Sat Oct 11 11:11:53 2014  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* configure.in (RUBY_NACL and others): merge patch from naclports. Supports PNaCl.
+
+	* dln.c: ditto. replace the old hacky dynamic loading over HTTP with nacl_io.
+
+	* file.c: ditto. tentatively use access(2) instead of eaccess.
+	  (rb_file_load_ok): weaken with attribute but not by postprocess.
+
+	* io.c: ditto.
+	  (socket.h): now NaCl has socket.h
+	  (flock): disable here instead of nacl/ioctl.h
+
+	* nacl/GNUmakefile.in: ditto.
+	  (CC, LD, NM, AR, AS, RANLIB, OBJDUMP, OBJCOPY):
+	  respect path to them if they are absolute.
+	  This helps naclports to build ruby in their source tree.
+	  (PROGRAM_NMF, .SUFFIXES): support .pnexe for PNaCl.
+	  (ruby.o, file.o): move the hack to attributes in ruby.c and file.c
+
+	* nacl/ioctl.h: ditto. removed. move the hack to io.c.
+
+	* nacl/nacl-config.rb: ditto. support arm, pnacl and others.
+
+	* nacl/pepper_main.c: ditto. support build in a naclports tree.
+
+	* ruby.c (rb_load_file): ditto. weaken with attribute but not by postprocess.
+
+Sat Oct 11 09:32:00 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/socket/unixsocket.c: [DOC] Fix example to render in HTML
+	  properly, with a patch by @eval [Fixes GH-733]
+	  https://github.com/ruby/ruby/pull/733
+
+Sat Oct 11 04:14:41 2014  Kir Shatrov  <shatrov@me.com>
+
+	* lib/open-uri.rb (OpenURI::Options): add :open_timeout default
+	* (def OpenURI.open_http): check :open_timeout option
+	* (module OpenURI): rdoc for :open_timeout
+	* test/open-uri/test_open-uri.rb (test_open_timeout): new test
+	  [Feature #10361]
+
+Fri Oct 10 11:27:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/io.h (rb_io_mode_flags, rb_io_modenum_flags):
+	  deprecate old macros for compatibility for ruby 1.8 and older.
+
+Thu Oct  9 23:31:47 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* bignum.c (absint_numwords_generic): set an array element after
+	  definition of a variable to fix compile error with older version
+	  of fcc (Fujitsu C Compiler) 5.6 on Solaris 10 on Sparc.
+	  [Bug #10350] [ruby-dev:48608]
+
+Thu Oct  9 16:15:26 2014  Eric Wong  <e@80x24.org>
+
+	* ext/-test-/st/foreach/extconf.rb: new file
+	* ext/-test-/st/foreach/foreach.c: ditto
+	* test/-ext-/st/test_foreach.rb: ditto
+	  [Feature #10321]
+
+Thu Oct  9 12:40:28 2014  Eric Wong  <e@80x24.org>
+
+	* benchmark/bm_hash_aref_sym*.rb: force static symbols
+
+Thu Oct  9 12:38:28 2014  Eric Wong  <e@80x24.org>
+
+	* hash.c (rb_any_hash): remove unnecessary dsym check
+
+Thu Oct  9 07:20:30 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* missing/setproctitle.c: Avoid invalidating argv[1], argv[2],
+	  etc. until the first call to Process.setproctitle, because
+	  the ps command of AIX refers to the argv array.
+	  [Bug #10090]
+
+Thu Oct  9 00:53:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_s_aref): fix rdoc.  `Dir.glob` allows an array but
+	  `Dir[]` not.  the former accepts an optional parameter `flags`,
+	  while the latter accepts arbitrary number of arguments but no
+	  `flags`.  [ruby-core:65265] [Bug #10294]
+
+Wed Oct  8 21:44:10 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_variable.c: use typed data.
+
+Wed Oct  8 16:36:47 2014  gogo tanaka  <mail@tanakakazuki.com>
+
+	* test/ruby/test_syntax.rb: added syntax tests of underscore
+	  arguments.  [Feature #10340][ruby-core:65496]
+
+Wed Oct  8 07:42:39 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI#inspect): remove Object id.
+	  URI is considered that it doesn't require id.
+
+Wed Oct  8 05:22:42 2014  Eric Wong  <e@80x24.org>
+
+	* ext/etc/etc.c (etc_systmpdir): set default tmplen correctly
+	  Fixup r47826
+
+Wed Oct  8 05:16:32 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add @- and @+ for Matrix and Vector.
+	  patch by gogo tanaka [#10068] [#10069]
+
+Wed Oct  8 04:58:48 2014  John Bachir  <j@jjb.cc>
+
+	* bootstraptest/test_io.rb (assert_finish):
+	  normalize rescue for Timeout::Error
+	* lib/net/ftp.rb (Net#read_timeout): ditto for doc
+	* lib/resolv.rb (Resolv::ResolvTimeout): ditto for subclass
+	* lib/webrick/httprequest.rb (_read_data): ditto for rescue
+	* sample/timeout.rb (p timeout): ditto for call
+	* test/drb/drbtest.rb (test_06_timeout): ditto
+	* test/ruby/test_readpartial.rb (test_open_pipe): ditto
+	* test/thread/test_queue.rb (test_queue_thread_raise): ditto
+	* thread.c (rb_thread_s_handle_interrupt): ditto for doc
+	  [ruby-core:65481] [misc #10339]
+
+Wed Oct  8 04:38:29 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/ruby/test_process.rb (TestProcess#test_setsid): AIX
+	  does not allow Process::getsid(pid) when pid is in a
+	  different session.
+
+Wed Oct  8 04:33:04 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/ruby/test_rubyoptions.rb (TestRubyOptions#test_encoding):
+	  On AIX, locale_charmap is ISO-8859-1 with LANG=C.  This means
+	  the source encoding of stdin is ISO-8859-1, so "invalid
+	  multibyte char" error does not occur.
+
+Wed Oct  8 04:30:29 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Matrix#laplace_expansion.
+	  patch by gogo tanaka [#10073]
+
+Wed Oct  8 04:29:21 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Vector.basis.
+	  Based on patch by gogo tanaka [#10072]
+
+Tue Oct  7 23:40:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (rb_f_kill): get rid of deadlock as unhandled and
+	  discarded signals do not make interrupt_cond signaled.
+	  based on the patch by Kazuki Tsujimoto at [ruby-dev:48606].
+	  [Bug #9820]
+
+Tue Oct  7 22:43:44 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_method.c: use typed data.
+
+Tue Oct  7 21:47:05 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_param.c: refactoring.
+
+Tue Oct  7 21:40:17 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_method.c: refactoring. add
+	  olemethod_data_get_struct to wrap Data_Get_Struct.
+	* ext/win32ole/win32ole_method.h: ditto.
+
+	* ext/win32ole/win32ole_param.c (oleparam_ole_param):
+	  call olemethod_data_get_struct instead of Data_Get_Struct.
+
+Tue Oct  7 11:17:08 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_systmpdir): try user temporary directory by
+	  confstr() on Mac OS X.
+	  c.f. http://www.opensource.apple.com/source/ruby/ruby-104/patches/ext_etc_etc.c.diff
+
+Tue Oct  7 10:48:17 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (struct stat.st_size): prefer off_t over int, long,
+	  and so on.  inspired by
+	  http://www.opensource.apple.com/source/ruby/ruby-104/patches/config.h.ed
+
+Tue Oct  7 10:37:39 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_UNIVERSAL_ARCH): fix missing quoting
+	  brackets.  incorporated from
+	  http://www.opensource.apple.com/source/ruby/ruby-104/patches/configure.diff
+
+Mon Oct  6 23:34:42 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_param.c: use typed data.
+
+Mon Oct  6 22:37:09 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.h (struct RSymbol): move from internal.h.
+
+Mon Oct  6 21:43:03 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* error.c: update exception tree. [DOC]
+	  reported by @hemge via twitter.
+
+Mon Oct  6 18:43:03 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* configure.in: Fix typo. [Bug #9914]
+
+Mon Oct  6 16:23:30 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (rb_f_kill): should not ignore signal unless the
+	  default handler is registered.  [ruby-dev:48592] [Bug #9820]
+
+Mon Oct  6 16:07:11 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_string.rb (test_LSHIFT_neary_long_max): enable
+	  only on platforms where string size range is smaller than memory
+	  space.  this test does not make sense but just wastes memory and
+	  time on other platforms, as it is hardly possible that a string
+	  size becomes neary LONG_MAX if long size equals pointer size.
+	  [ruby-core:65410] [Bug #10325]
+
+Mon Oct  6 11:21:21 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/unicode_norm_gen.rb: Adding/tweaking comments.
+
+Mon Oct  6 10:57:57 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/unicode_norm_gen.rb: Adjusted directory paths.
+
+Mon Oct  6 10:27:27 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/unicode_norm_gen.rb: Data generation script imported from
+	  https://github.com/duerst/eprun/blob/master/lib/generate.rb
+
+Mon Oct  6 10:15:15 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: Adjust example in documentation for
+	  Downloader.download.
+
+Mon Oct  6 10:07:07 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize: New folder for Unicode normalization
+	  functionality
+
+Sun Oct  5 11:04:13 2014  Luiz Angelo Daros de Luca  <luizluca@gmail.com>
+
+	* ext/digest/{md5,rmd160,sha1,sha2}/extconf.rb: configure OpenSSL
+	  only if bundled libraries is not used, so that OpenSSL is not
+	  linked unnecessarily.  [ruby-core:65404] [Bug #10324]
+
+Sun Oct  5 10:39:11 2014  Luiz Angelo Daros de Luca  <luizluca@gmail.com>
+
+	* ext/digest/rmd160/extconf.rb: fix transform function name to
+	  check.  [ruby-core:65091] [Bug #10252]
+
+Sun Oct  5 05:46:00 2014  Eric Wong  <e@80x24.org>
+
+	* ext/zlib/zlib.c (zstream_mark, zstream_free): update signature
+	  (gzfile_mark, gzfile_free): ditto
+	  (zstream_memsize): new function for rb_data_type->dsize
+	  (gzfile_memsize): ditto
+	  (zstream_data_type, gzfile_data_type): new data types
+	  (zstream_new): Data_Make_Struct => TypedData_Make_Struct
+	  (gzfile_new): ditto
+	  (get_zstream, get_gzfile): Data_Get_Struct => TypedData_Get_Struct
+	  (rb_zstream_flush_next_in): ditto
+	  (rb_zstream_flush_next_out): ditto
+	  (rb_zstream_avail_out): ditto
+	  (rb_zstream_avail_in): ditto
+	  (rb_zstream_closed_p): ditto
+	  (rb_deflate_initialize): ditto
+	  (rb_deflate_init_copy): ditto
+	  (rb_inflate_initialize): ditto
+	  (gzfile_ensure_close): ditto
+	  (rb_gzfile_closed_p): ditto
+	  (rb_gzfile_path): ditto
+	  (rb_gzwriter_initialize): ditto
+	  (rb_gzreader_initialize): ditto
+	  (rb_gzreader_unused): ditto
+	  [ruby-core:65377] [Feature #10319]
+
+Sat Oct  4 16:24:41 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/test_syslog.rb (TestSyslog#test_log): In AIX, each output
+	  line of LOG_PERROR to stderr has an additional empty line appended,
+	  so skip that line.
+
+Sat Oct  4 16:05:49 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/socket/test_unix.rb (TestSocket_UNIXSocket#test_too_long_path):
+	  sockaddr_un.sun_path in AIX is defined as char[1024],
+	  so "a" * 300 is not too long. "a" * 3000 would be enough.
+
+Sat Oct  4 09:12:03 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/win32ole/sample/example*.rb: Add wait input to quit for examples
+	  with patch provided by @windwiny [Fixes GH-705]
+	  https://github.com/ruby/ruby/pull/705
+
+Sat Oct  4 09:08:18 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/win32ole/win32ole.c: [DOC] Fix typo in :nodoc: reported by
+	  @windwiny to [Fix GH-705] https://github.com/ruby/ruby/pull/705
+
+	* ext/pty/pty.c: ditto
+
+Sat Oct  4 08:59:45 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_rand.c: [DOC] Add call signature for pseudo_bytes
+	  and random_bytes, wrap lines at 80 chars, and remove useless
+	  comments.
+
+Sat Oct  4 08:49:34 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_rand.c: [DOC] Add rdoc for method descriptions
+	  By @vipulnsward [Fixes GH-657] https://github.com/ruby/ruby/pull/657
+
+Sat Oct  4 08:23:48 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_rand.c: Use rb_define_module_function instead of
+	  macro. [Fixes GH-686] https://github.com/ruby/ruby/pull/686
+
+Sat Oct  4 06:04:56 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_method.c(olemethod_set_member): remove
+	  redundant NULL check.
+	* ext/win32ole/win32ole_type.c(oletype_set_member): ditto.
+
+Sat Oct  4 00:25:04 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: removed needless preparation for gcc.
+
+Fri Oct  3 23:41:20 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: enabled gcc build with osx on travis.
+
+Fri Oct  3 23:22:23 2014  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* template/fake.rb.in: fix make install failure due to MSYS path
+	  with mingw on MSYS environment.
+	  [ruby-core:64965] [Bug #10230]
+
+Fri Oct  3 21:02:32 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_io.rb (TestIO#test_advise): avoid to infinite loop.
+
+Fri Oct  3 19:26:01 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: enabled test results of linux.
+
+Fri Oct  3 18:52:16 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_io.rb (TestIO#test_advise): added workaround of fadvise(2)
+	  with tmpfs and old linux kernel. [ruby-core:65355][Bug #10313]
+
+Fri Oct  3 18:22:45 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: Disabled to generate document on travis.
+	  Reduce test running time.
+
+Fri Oct  3 12:42:15 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add hstack & vstack methods.
+	  Based on a patch by creasywuqiong. [Fix GH-344]
+
+Fri Oct  3 12:37:48 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Fix Matrix.rows copy bug.
+	  Patch by Arron Mabrey. [Fix GH-707]
+
+Fri Oct  3 06:06:28 2014  Eric Wong  <e@80x24.org>
+
+	* st.c (next_pow2): new function (from old bignum.c)
+	  (new_size): use next_pow2 function
+
+Fri Oct  3 05:58:58 2014  Eric Wong  <e@80x24.org>
+
+	* vm_trace.c (rb_tp_t): pack 56 => 48 bytes on 64-bit
+
+Thu Oct  2 18:41:45 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_nprocessors): Windows support.
+	  see [Feature #10267]
+
+Thu Oct  2 12:21:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/etc/etc.c (etc_nprocessors): New method.
+	  Accepted by matz at RubyKaigi 2014.
+	  [ruby-core:65142] [Feature #10267]
+
+Thu Oct  2 07:56:49 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (rb_iseq_line_trace_each): explicit cast
+	  Fix https://travis-ci.org/ruby/ruby/jobs/36814282
+
+Thu Oct  2 05:40:05 2014  Eric Wong  <e@80x24.org>
+
+	* ruby.h: set rb_event_flag_t to uint32_t
+	  [ruby-core:65315] [misc #10249]
+
+Thu Oct  2 05:32:17 2014  Eric Wong  <e@80x24.org>
+
+	* io.c (fptr_finalize): free memory before GC sweep
+	  [ruby-core:65269] [Feature #10295]
+
+Thu Oct  2 05:27:24 2014  Eric Wong  <e@80x24.org>
+
+	* marshal.c (w_class): check dump_arg->compat_tbl before lookup
+	  (w_object): lazy init ->compat_tbl before insert
+	  (obj_alloc_by_class): ditto
+	  (clear_dump_arg): free only non-NULL ->compat_tbl
+	  (clear_load_arg): ditto for ->compat_tbl
+	  (marshal_dump): ->compat_tbl defaults to zero
+	  (marshal_load): ditto for ->compat_tbl
+	  (r_entry0): check l->compat_tbl before lookup
+	  (r_fixup_compat): ditto
+	  [ruby-core:65305] [Feature #10302]
+
+Wed Oct  1 21:14:34 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_type.c: use typed data.
+
+Wed Oct  1 18:15:42 2014  Nolan Evans  <nolane@gmail.com>
+
+	* compile.c: remove commented out code.
+
+Wed Oct  1 17:38:53 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/fileutils/test_fileutils.rb: AIX does not allow
+	  a sticky bit on a regular file.
+
+Wed Oct  1 17:31:41 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* NEWS:  Add RubyGems update.
+
+Wed Oct  1 17:28:58 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.4.2.
+	* test/rubygems:  ditto.
+
+Tue Sep 30 22:25:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_data_type): separate ripper data type for from
+	  parser.
+
+Tue Sep 30 18:46:31 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_typelib.c: use typed data.
+
+Tue Sep 30 09:51:46 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* common.mk: fixed broken reference of update-config_files task
+
+Mon Sep 29 22:54:51 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (append_wstr): set expanded length, not length of
+	  appended string.  fix "probable buffer overflow" bug.
+	  [ruby-core:65317] [Bug #10304]
+
+	* string.c (str_make_independent_expand): drop NOFREE flag after
+	  reallocation, static buffer is not pointed anymore.
+	  [ruby-core:65317] [Bug #10304]
+
+Sun Sep 28 23:59:17 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gc.c (rb_gcdebug_print_obj_condition): use RVALUE_REMEMBERED
+	  because GET_HEAP_REMEMBERSET_BITS is obsoleted.
+
+Sun Sep 28 11:14:14 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Created new target update-unicode to download
+	  some Unicode data files.
+
+Fri Sep 26 15:03:19 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* lib/uri/rfc3986_parser.rb: raise URI::InvalidURIError when
+	  uri doesn't respond to #to_str. [ruby-core:64453] [Bug #10150]
+
+	* test/uri/test_parser.rb: test for above.
+
+Sat Sep 27 10:31:48 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gems/bundled_gems: upgraded to power_assert 0.1.4.
+
+Fri Sep 26 12:52:36 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_write): ASCII-8BIT StringIO
+	  should be writable any encoding strings, without conversion.
+	  [ruby-core:65240] [Bug #10285]
+
+Fri Sep 26 05:21:01 2014  Eric Wong  <e@80x24.org>
+
+	* object.c (rb_class_real): do not dereference 0 VALUE
+
+	* test/ruby/test_module.rb (test_inspect_segfault):
+	  Test case and bug report by Thomas Stratmann.
+	  [ruby-core:65214] [Bug #10282]
+
+Fri Sep 26 05:12:10 2014  Eric Wong  <e@80x24.org>
+
+	* man/ruby.1: document stack size env variables
+	  [Feature #10197]
+
+Thu Sep 25 19:37:34 2014  Eric Wong  <e@80x24.org>
+
+	* io.c (free_io_buffer): new function for a common pattern
+	  (clear_readconv): use free_io_buffer
+	  (rb_io_fptr_finalize): ditto
+
+Thu Sep 25 07:51:07 2014  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/matrix.rb: Fix docs. Patched by Ben Woodall. [GH-726]
+
+Wed Sep 24 19:04:04 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/data: New directory for downloaded Unicode
+	  data files.
+
+Wed Sep 24 18:59:59 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: Adjusting example for
+	  Downloader.download to implementation changes in r47693.
+
+Wed Sep 24 18:06:06 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: Removing unused method
+	  Downloader.download_if_modified_since.
+	  (if ever used, just replace with Downloader.download)
+
+Wed Sep 24 17:59:59 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: Fixing raise after return.
+
+Wed Sep 24 17:55:55 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: Made Unicode data file location available
+	  via :unicode Symbol.
+
+Wed Sep 24 10:45:45 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: Small fix to documentation comment.
+
+Tue Sep 23 22:00:20 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parse_ident): just after a label, new expression should
+	  start, cannot be a modifier.  [ruby-core:65211] [Bug #10279]
+
+Tue Sep 23 16:07:07 2014  Martin Duerst <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: added Downloader.download_if_modified_since
+	  to reduce downloads of large files that change only rarely.
+	  [ruby-core:65164] [CommonRuby - Feature #10084]
+
+Tue Sep 23 11:55:09 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: added rubyspec into travis tasks and eliminate to stdout.
+
+Mon Sep 22 20:00:29 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_variant.rb
+	  (test_conversion_time2date_with_msec): test by using only
+	  assert_in_delta  to avoid to fail when converting Time object with
+	  999999999 nanoseconds into VT_DATE Variant.
+
+Mon Sep 22 19:49:12 2014  Zachary Scott  <e@zzak.io>
+
+	* doc/syntax/methods.rdoc: [DOC] [] and []= methods by @process
+	  [Fixes GH-662] https://github.com/ruby/ruby/pull/662
+
+Mon Sep 22 18:21:35 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: Only osx build is enabled. linux builds is random failure
+	  and test results of major linux is covered by rubyci.
+
+Mon Sep 22 12:10:29 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_time_tz.rb: Fix test error with tzdata-2014g.
+	  [ruby-core:65058] [Bug #10245] Reported by Vit Ondruch.
+
+Mon Sep 22 09:28:43 2014  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c ({send,recv}msg_args_struct): 24 => 16 bytes
+	* ext/socket/init.c (connect_arg): ditto
+	* ext/socket/raddrinfo.c (getnameinfo_arg): 56 => 48 bytes
+	  (reductions only for 64-bit systems)
+
+Mon Sep 22 02:04:25 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/drb/drb.rb: Support graceful shutdown.
+	  (DRbTCPSocket#initialize): Create a pipe for shutdown notification.
+	  (DRbTCPSocket#close): Invoke close_shutdown_pipe.
+	  (DRbTCPSocket#close_shutdown_pipe): New private method.
+	  (DRbTCPSocket#accept): Use accept_or_shutdown.
+	  (DRbTCPSocket#accept_or_shutdown): New private method which returns
+	  nil on shutdown.
+	  (DRbServer#stop_service): Use shutdown instead of Thread#kill.
+	  (DRbServer#run): Break infinite loop when main_loop returns nil.
+	  (DRbServer#main_loop): @protocol.accept may return nil.
+
+	* lib/drb/ssl.rb: Follow above change.
+
+	* lib/drb/unix.rb: Ditto.
+
+Sun Sep 21 13:54:36 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* time.c: raise exception when minutes of utc_offset is out of 00-59.
+	  patch is from Kenichi Kamiya.
+	  [ruby-dev:47539] [Bug #8679]
+
+	* test/ruby/test_time.rb: test for above.
+	  patch is from Kenichi Kamiya.
+
+Sun Sep 21 19:04:08 2014  Narihiro Nakamura <authornari@gmail.com>
+
+	* st.c (do_hash_bin): unused macro.
+
+Sun Sep 21 18:45:01 2014  Narihiro Nakamura <authornari@gmail.com>
+
+	* parse.y (parser_class_nest): unused variable after YARV
+	  merged (r11439).
+
+Sun Sep 21 18:14:03 2014  Narihiro Nakamura <authornari@gmail.com>
+
+	* st.c (numberof): unused. internal.h has same macro.
+
+	* node.c (F_CUSTOM2): unused.
+
+Sun Sep 21 14:11:23 2014  Tanaka Akira  <akr@fsij.org>
+
+	* thread_pthread.c (native_set_thread_name): New function to
+	  set thread name visible with ps command on GNU/Linux.
+	  Ex. ps -o %c -L
+
+	* thread.c (thread_start_func_2): Call native_set_thread_name at
+	  beginning.
+	  (rb_thread_inspect_msg): Extract from rb_thread_inspect.
+
+Sun Sep 21 12:49:11 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.c (rb_iseq_defined_string): trim redundant semi-colon
+
+Sun Sep 21 12:19:29 2014  Eric Wong  <e@80x24.org>
+
+	* file.c (rb_find_file_ext_safe): clear tmp buffer on failure
+	  (rb_find_file_safe): ditto
+
+Sat Sep 20 04:42:18 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* lib/csv.rb: avoid unnecessary object allocations.
+	  patch is from Andrew Vit. [ruby-core:63215] [Feature #9952]
+
+Sun Sep 21 12:10:18 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rexml/**/*.rb: removed commented-out code.
+
+Sat Sep 20 03:46:58 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* array.c: use rb_equal_opt() for performance improvement.
+	  [ruby-core:64954] [Feature #10227]
+
+Sun Sep 21 11:16:56 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (rbtime2vtdate, vtdate2rbtime): fix
+	  the bug in conversion of milliseconds. [Bug #10258]
+
+	* test/win32ole/test_win32ole_variant.rb
+	  (test_conversion_dbl2date_with_msec,
+	  test_conversion_time2date_with_msec): use assert_in_delta instead
+	  of assert_equal to treat an acceptable error range.
+
+Sun Sep 21 11:03:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (ruby_signal): although "EINVAL from sigaction(2) is
+	  not a bug", but even it is a failure.  pointed at toRuby/guRuby
+	  in RubyHiroba.
+
+Sat Sep 20 03:00:26 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* lib/tempfile.rb: define parameters appropriately and some
+	  refactoring.
+
+	* lib/tmpdir.rb: ditto.
+
+Sat Sep 20 23:58:21 2014  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_chunk): Deprecate the state management.
+	  (enum_slice_before): Ditto.
+
+Sat Sep 20 15:39:11 2014  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_slice_when): New method: Enumerable#slice_when.
+	  (slicewhen_i): New function.
+	  (slicewhen_ii): New function.
+
+	* enumerator.c (InitVM_Enumerator): New method:
+	  Enumerator::Lazy#slice_when.
+
+	  [ruby-core:62499] [Feature #9826]
+
+Sat Sep 20 11:55:19 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: added new configurations for osx on travis ci.
+	  [fix GH-723]
+	* test/ruby/test_object.rb: tweaked to memory leak limit for osx build.
+
+Sat Sep 20 10:48:41 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assoc): allow quoted ID as a key of a hash literal.
+	  [ruby-core:34453] [Feature #4276]
+
+Sat Sep 20 10:23:00 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_arguments): store local variable IDs in
+	  temporary list as Symbols.  previously these are stored as
+	  Fixnums to prevent from GC, but IDs of dynamic symbols can
+	  exceed Fixnum range and cause RangeError at inverting from
+	  Fixnum.  [ruby-dev:48564] [Bug #10266]
+
+Sat Sep 20 10:02:51 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/lib/openssl/x509.rb (OpenSSL::X509::Name#pretty_print):
+	  New method.
+	  (OpenSSL::X509::Certificate#pretty_print): Ditto.
+
+	* ext/openssl/lib/openssl/bn.rb (OpenSSL::BN#pretty_print): Ditto.
+
+Sat Sep 20 07:55:57 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (eval_string_with_cref): fix super from eval with
+	  scope.  set klass in the current control frame to the class of
+	  the receiver in the context to be evaluated, this class/module
+	  must match the actual receiver to call super.
+	  [ruby-core:65122] [Bug #10263]
+
+Fri Sep 19 20:06:00 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.c (rb_str_dynamic_intern): check if the stem ID of
+	  attrset ID is already registered as a static ID.
+	  [ruby-dev:48559] [Bug #10259]
+
+Fri Sep 19 15:48:09 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (VCSUP): nothing to do if this worktree is not
+	  under any VCS (it means that the worktree may be from the release
+	  package).
+
+Fri Sep 19 10:47:03 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_math.rb (TestMath#assert_infinity): Float#finite?
+	  returns true also for NaN, so use Float#infinite? instead.
+	  [ruby-core:65117] [Feature #10261]
+
+Fri Sep 19 05:36:16 2014  Eric Wong  <e@80x24.org>
+
+	* NEWS: note --with-jemalloc option [ci skip]
+
+Thu Sep 18 16:26:27 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_commands_setup_command.rb: @ui uses StringIO
+	  as its streams, and Encoding.default_external does not effect to
+	  StringIOs already exist.  so, we need to set external_encoding of
+	  @ui.outs directly.  this problem (test failure) does not appear in
+	  the environments default_external is us-ascii or utf-8.
+
+Thu Sep 18 15:02:15 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_specification.rb: '/' is not always the root.
+
+Thu Sep 18 09:36:37 2014  Scott Francis  <scott.francis@shopify.com>
+
+	* vm_method.c (Init_Method): make global method cache size
+	  configurable by environment variable
+	  "RUBY_GLOBAL_METHOD_CACHE_SIZE"  [Fix GH-719]
+
+Thu Sep 18 07:03:36 2014  Eric Wong  <e@80x24.org>
+
+	* test/-ext-/string/test_modify_expand.rb: increase limit
+	  for {je,tc}malloc [Bug #10236]
+
+Thu Sep 18 06:41:18 2014  Eric Wong  <e@80x24.org>
+
+	* ext/zlib/zlib.c (struct gzfile): pack (288 => 272 bytes) on 64-bit
+
+Thu Sep 18 05:44:05 2014  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_connect): refactor for blocking
+	  (wait_connectable): clear error before wait
+	  [Bug #9356]
+
+Wed Sep 17 23:12:36 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/rfc3986_parser.rb: specify a regexp for :OPAQUE; generic.rb
+	  assumes it is present, and will refuse all values otherwise.
+	  by Matthew Draper <matthew@trebex.net>
+	  https://github.com/ruby/ruby/pull/718 fix GH-718
+
+Wed Sep 17 16:22:58 2014  Eric Wong  <e@80x24.org>
+
+	* ext/zlib/zlib.c (zlib_mem_alloc): check overflow
+
+Wed Sep 17 11:33:35 2014  Laurent Arnoud  <laurent@spkdev.net>
+
+	* test/fiddle/test_import.rb (Fiddle::TestImport#test_sizeof):
+	  added test for long long [fix GH-716]
+
+Wed Sep 17 11:09:21 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/rbinstall.rb: fixed invalid options with latest rubygems.
+	  https://github.com/rubygems/rubygems/issues/1013
+
+Tue Sep 16 19:19:00 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* benchmark/bm_app_aobench.rb: update outdated links to the
+	  original program.  [ruby-dev:48550] [Feature #10247]
+
+Tue Sep 16 01:06:40 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* reg*.c: Merge Onigmo 5.15.0 38a870960aa7370051a3544
+
+Mon Sep 15 16:21:10 2014  Eric Wong  <e@80x24.org>
+
+	* io.c (struct io_advise_struct): 32 => 24 bytes on 64-bit
+	* io.c (struct io_internal_writev_struct): 24 => 16 bytes on 64-bit
+	* process.c (struct waitpid_arg): ditto
+
+Mon Sep 15 10:29:25 2014  Natalie Weizenbaum  <nweiz@google.com>
+
+	* ext/pathname/lib/pathname.rb (SAME_PATHS):
+	  Pathname#relative_path_from uses String#casecmp to compare strings
+	  on case-insensitive filesystem platforms (e.g., Windows). This can
+	  return nil for strings with different encodings, and the code
+	  previously assumed that it always returned a Fixnum.  [Fix GH-713]
+
+Mon Sep 15 09:43:18 2014  Sho Hashimoto  <sho.hsmt@gmail.com>
+
+	* ext/fiddle/lib/fiddle/import.rb (Fiddle::Importer#sizeof): fix typo,
+	  SIZEOF_LONG_LON.  [Fix GH-714]
+
+Mon Sep 15 08:13:40 2014  Matthew Draper  <matthew@trebex.net>
+
+	* sprintf.c (rb_str_format): rational 'f' format works for more
+	  values.  [fix GH-717]
+
+Sun Sep 14 16:57:27 2014  Eric Wong  <e@80x24.org>
+
+	* template/vm.inc.tmpl: "insns.c" => "insns.def"
+	* tool/instruction.rb: typo fix
+
+Sun Sep 14 12:29:52 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_tracer.rb: fixed testcase for rubygems update.
+
+Sun Sep 14 12:29:02 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems:  Update to RubyGems 2.4.1 master(713ab65)
+	  Complete history at:
+	  https://github.com/rubygems/rubygems/blob/master/History.txt#L3-L216
+
+	* test/rubygems:  ditto.
+
+Sun Sep 14 11:03:24 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: update version
+	* ext/psych/psych.gemspec: ditto
+
+Sun Sep 14 08:43:37 2014  Eric Wong  <e@80x24.org>
+
+	* ccan/container_of/container_of.h (container_of_or_null): added
+	  [ccan 7ec5b8e06b2fd5fa98b1fcde1158c286d2d429d8] (David Gibson)
+
+Sun Sep 14 08:41:44 2014  Eric Wong  <e@80x24.org>
+
+	* ccan/list/list.h (list_del_init, list_node_init): new functions
+	  for multiple list_del() calls
+	  [ccan ec8654d94d3c5c47aa5f82698f7e8048c79765b1] (Rusty Russell)
+
+Sat Sep 13 22:19:26 2014  Bernard Potocki  <bernard.potocki@imanel.org>
+
+	* hash.c (rb_hash_aset): fix misleading example which may suggest
+	  that Hash.store will return self instead of value - Hash#store
+	  is returning value and update itself, as well as Hash#[]=.
+	  [Fix GH-715]
+
+Sat Sep 13 15:16:31 2014  Eric Wong  <e@80x24.org>
+
+	* class.c: use ALLOC(rb_subclass_entry_t)
+
+Sat Sep 13 14:14:00 2014  Eric Wong  <e@80x24.org>
+
+	* process.c (free_exec_arg): remove
+	  (memsize_exec_arg): ptr is never NULL
+	  (exec_arg_data_type): use RUBY_TYPED_DEFAULT_FREE
+
+	* variable.c (autoload_i_free): remove
+	  (autoload_data_i_type): use RUBY_TYPED_DEFAULT_FREE
+	  (autoload_memsize): ptr is never NULL
+
+	* vm_backtrace.c (location_free): remove
+	  (location_mark): ptr is never NULL
+	  (location_data_type): use RUBY_TYPED_DEFAULT_FREE
+	  (backtrace_mark): ditto
+	  (backtrace_free): ditto
+
+Sat Sep 13 13:43:07 2014  Eric Wong  <e@80x24.org>
+
+	* doc/NEWS-2.0.0: fix typo for default RUBY_FIBER_MACHINE_STACK_SIZE
+	  [ci skip]
+
+Sat Sep 13 11:16:58 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_event.c(ev_advise, ole_event_free,
+	  fev_s_allocate, fev_unadvise): avoid segmentation fault when COM
+	  server freed before calling Unadvise from WIN32OLE_EVENT object.
+	* ext/win32ole/win32ole.c: ditto.
+
+Sat Sep 13 09:47:44 2014  Eric Wong  <e@80x24.org>
+
+	* man/ruby.1: use https for *.ruby-lang.org links
+
+Sat Sep 13 06:31:23 2014  Eric Wong  <e@80x24.org>
+
+	* vm.c (thread_alloc): remove needless volatile
+
+Sat Sep 13 06:13:55 2014  Eric Wong  <e@80x24.org>
+
+	* proc.c (proc_free): remove, use RUBY_TYPED_DEFAULT_FREE
+	  (proc_mark, proc_memsize): remove needless branching
+
+	* vm.c (env_free): remove, use RUBY_TYPED_DEFAULT_FREE
+	  (env_mark, env_memsize): remove needless branching
+
+Sat Sep 13 05:52:15 2014  Eric Wong  <e@80x24.org>
+
+	* proc.c (rb_proc_alloc): inline and move to vm.c
+	  (rb_proc_wrap): new wrapper function used by rb_proc_alloc
+	  (proc_dup): simplify alloc + copy + wrap operation
+	  [ruby-core:64994]
+
+	* vm.c (rb_proc_alloc): new inline function
+	  (rb_vm_make_proc): call rb_proc_alloc
+
+	* vm_core.h: remove rb_proc_alloc, add rb_proc_wrap
+
+	* benchmark/bm_vm2_newlambda.rb: short test to show difference
+
+Sat Sep 13 04:40:04 2014  Eric Wong  <e@80x24.org>
+
+	* process.c (Init_process): subclass Thread as Process::Waiter
+	  (rb_detach_process): use Process::Waiter instead of singleton class
+	  Thanks to headius and nobu. [Bug #10231]
+
+	* test/ruby/test_process.rb (test_process_detach): new test
+
+	* inits.c (rb_call_inits): call Init_Thread before Init_process to
+	  ensure Process::Waiter may be a subclass of Thread
+
+Fri Sep 12 18:14:28 2014  Eric Wong  <e@80x24.org>
+
+	* vm.c (env_alloc): inline to avoid extra zeroing
+	  tiny speedup [ruby-core:64980]
+
+Fri Sep 12 17:13:29 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_make, remove_method): ditto.
+
+Fri Sep 12 14:39:55 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gems/bundled_gems: Upgraded to test-unit-3.0.1 and minitest-5.4.1
+
+Fri Sep 12 06:55:40 2014  Eric Wong  <e@80x24.org>
+
+	* string.c (Init_frozen_strings): use st_init_table_with_size
+
+Fri Sep 12 06:15:37 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (sym_find): remove Symbol.find because we have Symbol GC now.
+	  https://bugs.ruby-lang.org/projects/ruby/wiki/DevelopersMeeting20140904Japan
+	  If you still want this, request again on Redmine. [Feature #7854]
+	  https://bugs.ruby-lang.org/issues/7854
+
+	* ext/-test-/symbol/init.c (sym_find): moved from string.c for tests.
+
+Fri Sep 12 04:24:03 2014  Eric Wong  <e@80x24.org>
+
+	* insns.def (once): define and use fake RUNNING_THREAD_ONCE_DONE
+	  pointer to indicate is->once.running_thread is done.
+
+	* vm_core.h (iseq_inline_storage_entry): remove done field,
+	  allowing the union to be reduced from 24=>16 bytes on 64-bit
+	  [Feature #10187]
+
+Thu Sep 11 20:10:00 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_thread_mark): use rb_gc_mark_values() to mark VM stack.
+
+Thu Sep 11 19:50:57 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_register_special_exception): make new function to
+	  make and register special exceptions.
+
+	* vm.c (rb_vm_mark): do not need to mark special exceptions
+	  because they are registered by rb_gc_register_mark_object().
+
+	* eval.c (Init_eval): use rb_vm_register_special_exception().
+
+	* gc.c (Init_GC): ditto.
+
+	* proc.c (Init_Proc): ditto.
+
+	* thread.c (Init_Thread): ditto.
+
+Thu Sep 11 19:32:30 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_mark_values): added.
+	  This function is similar to rb_gc_mark_locations(), but not
+	  conservative.
+
+	* internal.h: ditto.
+
+	* vm.c (env_mark): use rb_gc_mark_values() because env values should
+	  be Ruby VALUEs.
+
+Thu Sep 11 19:16:39 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_ptr): rename to gc_mark_set.
+
+	* gc.c (gc_mark): add gc_mark_ptr() to skip is_markable_object()
+	  check. gc_mark_maybe() can use gc_mark_ptr() directly because
+	  passed pointer is checked by is_pointer_to_heap().
+
+Thu Sep 11 18:40:16 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .gitignore: ignored temporary files and coverage results.
+
+Thu Sep 11 18:15:30 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .gitignore: ignored only simplecov.
+	* coverage/README: Added coverage docs.
+
+Thu Sep 11 17:25:31 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, gc.c: add new internal events
+	  RUBY_INTERNAL_EVENT_GC_ENTER and
+	  RUBY_INTERNAL_EVENT_GC_EXIT.
+
+	  When invoking GC process, GC_ENTER event is called.
+	  When exiting from GC process, GC_EXIT event is called.
+
+	  Incremental GC (incremental marking and lazy sweep) can call
+	  these events many times.
+
+	  For example (minor marking):
+	  (1) GC_ENTER
+	  - (2) GC_START (minor GC)
+	    (minor marking)
+	  - (3) GC_END_MARK
+	    (start lazy sweep)
+	  (4) GC_EXIT
+	    (ruby process)
+	  (5) GC_ENTER
+	    (lazy sweep)
+	  (6) GC_EXIT
+	    (ruby process)
+	    (... repeat (5), (6))
+	  (7) GC_ENTER
+	      (finish lazy sweep)
+	    - (8) GC_END_SWEEP
+	  (9) GC_EXIT
+
+	  2nd example (incremental major marking):
+	  (1) GC_ENTER
+	  - (2) GC_START (minor GC)
+	    (start incremental marking)
+	  (3) GC_EXIT
+	    (ruby process)
+	  (4) GC_ENTER
+	    (incremental marking)
+	  (5) GC_EXIT
+	     (ruby process)
+	     (... repeat (4), (5))
+	  (6) GC_ENTER
+	    (finish incremental marking)
+	  - (7) GC_END_MARK
+	    (start lazy sweep)
+	  (8) GC_EXIT
+	    (ruby process)
+	  (9) GC_ENTER
+	    (lazy sweep)
+	  (10) GC_EXIT
+	    (ruby process)
+	    (... repeat (9), (10))
+	  (11) GC_ENTER
+	    (finish lazy marking)
+	  - (12) GC_STOP_SWEEP
+	  (13) GC_EXIT
+
+	  These internal events enable to measure GC pause time completely.
+
+Thu Sep 11 17:04:54 2014  Eric Wong  <e@80x24.org>
+
+	* lib/benchmark.rb: remove CLOCK_MONOTONIC_RAW support
+	  Thanks to Vit Ondruch for reporting the issue on ARM.
+	  [Bug #10202]
+
+Thu Sep 11 14:31:57 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: freeze nil/true/false.
+	  [Feature #8923]
+
+	* gc.c (should_be_finalizable): check frozen after checkin FL_ABLE.
+
+	* object.c (rb_obj_taint): check
+	  OBJ_TAINTABLE(obj).
+
+	* object.c (rb_obj_freeze): remove immediate_frozen_tbl
+	  because all of immediate values are frozen. YAY!
+
+	* object.c (rb_obj_frozen_p): ditto.
+
+	* test/ruby/test_eval.rb: skip instance_variable_set for
+	  frozen objects.
+
+	* test/ruby/test_weakmap.rb: check ArgumentError instead of
+	  RuntimeError.
+
+Thu Sep 11 10:03:16 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rdoc.rb, lib/rdoc, test/rdoc: Update to RDoc 4.2.0.alpha(21b241a)
+
+Wed Sep 10 17:52:25 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (rb_vm_addr2insn): rename to rb_vm_insn_addr2insn
+	  to clear what address.
+
+Wed Sep 10 16:22:26 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb: fix condition.
+
+Wed Sep 10 15:29:46 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_call_info_t): ci->flag becomes 32-bit unsigned int
+	  ci->index becomes a 32-bit signed int (from signed long).
+	  Reorder for better packing on 64-bit, giving an 8 byte reduction
+	  from 104 to 96 bytes for each ci.
+	  [Feature #10187]
+
+	* compile.c (new_callinfo, setup_args, iseq_compile_each,
+	  iseq_build_from_ary_body): adjust for type changes
+
+	* vm_insnhelper.c (vm_getivar): ditto
+
+Wed Sep 10 15:07:35 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (rb_iseq_translate_threaded_code):
+	  modify in-place w/o copy
+	  (rb_vm_addr2insn): new function for debug
+	  (rb_iseq_original_iseq): ditto
+	  (iseq_set_sequence): assign iseq_encoded directly
+	  [Feature #10185]
+
+	* vm_core (rb_iseq_t): move original ->iseq to bottom
+
+	* iseq.c (iseq_free, iseq_free): adjust for new layout
+	  (rb_iseq_disasm): use original iseq for dump
+	  (iseq_data_to_ary): ditto
+	  (rb_iseq_line_trace_each): ditto
+	  (rb_iseq_build_for_ruby2cext): use iseq_encoded directly
+
+	* vm_dump.c (rb_vmdebug_debug_print_pre): use original iseq
+
+Wed Sep 10 15:00:11 2014  Eric Wong  <e@80x24.org>
+
+	* time.c (time_mark): remove NULL check
+	  (time_memsize): ditto
+	  (time_free): remove, use RUBY_TYPED_DEFAULT_FREE instead
+	  [Feature #10219]
+
+Wed Sep 10 14:14:57 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (encs enc trans libencs libenc libtrans): force to run
+	  enk.mk because common.mk does not know the dependency, but enk.mk
+	  knows. [ruby-dev:48530] [Bug #10220]
+
+Wed Sep 10 11:59:10 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat): update rdoc.
+
+Wed Sep 10 11:52:08 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): rename:
+	  * malloc_increase -> malloc_increase_bytes
+	  * malloc_limit -> malloc_increase_bytes_limit
+	  * oldmalloc_increase -> oldmalloc_increase_bytes
+	  * oldmalloc_limit -> oldmalloc_increase_bytes_limit
+	  ref: [Feature #9924]
+
+Wed Sep 10 11:45:40 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): rename `heap_used' to `heap_allocated_pages'.
+	  ref: [Feature #9924]
+
+	* test/ruby/test_gc.rb: add constraints test for gc stat information.
+
+Wed Sep 10 11:31:16 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): rename:
+	  * remembered_shady_object -> remembered_wb_unprotected_objects
+	  * remembered_shady_object_limit -> remembered_wb_unprotected_objects_limit
+	  * old_object -> old_objects
+	  * old_object_limit -> old_objects_limit
+	  ref: [Feature #9924]
+
+Wed Sep 10 11:12:25 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): support:
+	  * total_allocated_pages
+	  * total_freed_pages
+	  ref: [Feature #9924]
+
+Wed Sep 10 10:48:04 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb: catch up last fix.
+
+Wed Sep 10 10:36:08 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (objspace_total_slot): rename objspace_available_slots.
+
+	* gc.c (objspace_live_slot, objspace_free_slot): rename
+	  ..._slot() to ..._slots().
+
+	* gc.c (objspace_free_slot): should subtract heap_pages_final_slots.
+
+	* gc.c (gc_stat_internal):
+	  * add `heap_available_slots' field
+	  * rename heap_live_slot to heap_live_slots
+	  * rename heap_free_slot to heap_free_slots
+	  ref: [Feature #9924]
+
+Wed Sep 10 07:22:53 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: refactoring for RGENGC_PROFILE > 0.
+
+	  * rename rb_objspace_t::profile::..._count
+	    to rb_objspace_t::profile::total_..._count
+	  * rename promote_infant_types to promote_types
+
+	* gc.c (gc_remember_unprotected): count remembered shady objects here.
+
+Wed Sep 10 03:12:12 2014  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* gc.c (init_mark_stack): MEMZERO() receive type as 2nd argument instead
+	  of size.
+	  Coverity Scan found this bug.
+
+Tue Sep  9 21:55:39 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/sample/excel2.rb: remove some commented-out code.
+	  rotate graph more slowly to see graph clearly.
+
+Tue Sep  9 19:52:33 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename rb_objspace_t::marked_objects to marked_slots.
+
+	* gc.c (gc_marks_start): should be clear first.
+
+	* gc.c (gc_marks_start): remembered shady objects are also marked.
+
+	* gc.c (gc_stat_internal): add heap_marked_slots.
+
+Tue Sep  9 18:58:48 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename
+	  * total_allocated_object_num -> total_allocated_objects
+	  * total_allocated_object_num_at_gc_start -> total_allocated_objects_at_gc_start
+	  * total_freed_object_num -> total_freed_objects
+
+	* gc.c (gc_stat_internal):
+	  * rename total_allocated_object -> total_allocated_objects
+	  * rename total_freed_object -> total_freed_objects
+
+Tue Sep  9 18:51:36 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): fix symbol names
+	  * heap_final_slot -> heap_final_slots
+	  * heap_swept_slot -> heap_swept_slots
+
+Tue Sep  9 18:18:07 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_t::heap_pages): rename field names:
+	  * used -> allocated_pages
+	  * increment -> allocatable_pages
+	  * length -> sorted_length
+	  And remove unused `limit' field.
+
+	* gc.c: rename macros:
+	  * heap_pages_used -> heap_allocated_pages
+	  * heap_pages_length -> heap_pages_sorted_length
+	  * heap_pages_increment -> heap_allocatable_pages
+
+	* gc.c (gc_stat_internal): fix symbol names
+	  * heap_used -> heap_allocated_pages
+	  * heap_eden_page_length -> heap_eden_pages
+	  * heap_tomb_page_length -> heap_tomb_pages
+	  * heap_increment -> heap_allocatable_pages
+	  * heap_length -> heap_sorted_length
+
+	  ref: [Feature #9924]
+	  https://docs.google.com/spreadsheets/d/11Ua4uBr6o0k-nORrZLEIIUkHJ9JRzRR0NyZfrhEEnc8/edit?usp=sharing
+	  Yellow color fields in this table are changed.
+
+	* test/ruby/test_gc.rb: catch up this change.
+
+Tue Sep  9 14:56:03 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: continue layout changing.
+
+	  newobj_of() also touch:
+	  (4) increment total_allocated_object_num
+	  (5) check hook_events
+
+	  And gather fields related to marking phase.
+
+Tue Sep  9 14:21:50 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix layout of rb_objspace_t to improve cache locality.
+
+	  newobj_of() accesses:
+	  (1) rb_objspace_t::flags
+	  (2) rb_objspace_t::eden_heap::freelist
+	  (3) and rb_objspace_t::eden_heap::free_pages if freelist is NULL.
+
+Tue Sep  9 14:09:36 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: move rb_objspace_t::flags::gc_stressful after during_gc
+	  to make accessing both parameters easy.
+
+	* gc.c (heap_get_freeobj): add LIKELY() hint.
+
+	* gc.c (heap_get_freeobj_from_next_freepage): ditto.
+
+	* gc.c (newobj_of): check both parameters at once for exceptional
+	  case.
+
+Tue Sep  9 13:51:32 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add rb_objspace_t::flags::gc_stressful and
+	  ruby_gc_stressful macro.
+	  Rename objspace->gc_stress to objspace->gc_stress_mode.
+
+	  If objspace->gc_stress_mode is true (!nil and !false) then
+	  ruby_gc_stressful becomes TRUE.
+
+	  ruby_gc_stressful will speedup newobj_of() slightly.
+
+	* gc.c: initialize ruby_gc_stress(full|_mode) by gc_params.gc_stress
+	  even if ENABLE_VM_OBJSPACE is false.
+
+Tue Sep  9 13:05:50 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove ruby_disable_gc_stress and add ruby_disable_gc
+	  to speed-up newobj_of().
+
+	* gc.c (ready_to_gc): check ruby_disable_gc.
+
+	* signal.c: use ruby_disable_gc.
+
+Tue Sep  9 12:11:41 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename gc_stat entries and check stat transition.
+
+Tue Sep  9 12:06:03 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_sweep_rest): remove wrong modification of during_gc flag.
+
+Tue Sep  9 11:39:41 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: pack boolean values into rb_objspace_t::flags with bit fields
+	  to improve cache locality.
+
+Tue Sep  9 11:11:05 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_object.rb: extend timeout.
+
+Tue Sep  9 09:02:07 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#pkg_config): append --cflags to also
+	  $CXXFLAGS, as they are often used by C++ compiler.
+	  [ruby-core:54532] [Bug #8315]
+
+Tue Sep  9 07:03:22 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c: remove needless SYM2ID <-> ID2SYM conversions
+	  [misc #10207]
+
+Tue Sep  9 05:48:42 2014  Eric Wong  <e@80x24.org>
+
+	* symbol.c (rb_intern_cstr_without_pindown): check dsymbol on return
+	  This is not a complete fix for bug 10206, but seems to reduce
+	  that crash and also looks correct.
+
+Tue Sep  9 04:36:24 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_env_t): use flexible array
+	  This reduces allocations and speeds up the lambda calculus
+	  fizzbuzz (bm_app_lc_fizzbuzz.rb) benchmark [ruby-core:64858]
+	* proc.c (get_local_variable_ptr): deconst to adjust for flex array
+	* vm.c (env_mark, env_free, env_memsize): remove check for env->env
+	* vm.c (env_alloc): single allocation for flex array
+	* vm.c (vm_make_env_each): adjust env_alloc call
+
+Mon Sep  8 16:08:22 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_lc_fizzbuzz.rb: should skip output on benchmark.
+
+Mon Sep  8 16:04:02 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_lc_fizzbuzz.rb: `answer.to_a' does not return
+	  a string, but an array.
+
+Mon Sep  8 13:18:37 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_lc_fizzbuzz.rb: added.
+
+	  This program is described closely in "Understanding Computation"
+	  chapter 6 by Tom Stuart. <http://computationbook.com/>
+
+	  Japanese translation will be published soon.
+	  <http://www.oreilly.co.jp/books/9784873116976/>
+
+Mon Sep  8 12:01:39 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add incremental GC algorithm. [Feature #10137]
+
+	  Please refer this ticket for details.
+
+	  This change also introduces the following changes.
+
+	  * Remove RGENGC_AGE2_PROMOTION and introduce object age (0 to 3).
+	    Age can be count with FL_PROMOTE0 and FL_PROMOTE1 flags in
+	    RBasic::flags (2 bit). Age == 3 objects become old objects.
+	  * WB_PROTECTED flag in RBasic to WB_UNPROTECTED bitmap.
+	  * LONG_LIVED bitmap to represent living objects while minor GCs
+	    It specifies (1) Old objects and (2) remembered shady objects.
+	  * Introduce rb_objspace_t::marked_objects which counts marked
+	    objects in current marking phase. marking count is needed to
+	    introduce incremental marking.
+	  * rename mark related function and sweep related function to
+	    gc_(marks|sweep)_(start|finish|step|rest|continue).
+	  * rename rgengc_report() to gc_report().
+	  * Add obj_info() function to get cstr of object details.
+	  * Add MEASURE_LINE() macro to measure execution time of specific line.
+	  * and many small fixes.
+
+	* include/ruby/ruby.h: add flag USE_RINCGC.
+	  Now USE_RINCGC can be set only with USE_RGENGC.
+
+	* include/ruby/ruby.h: introduce FL_PROMOTED0 and add FL_PROMOTED1
+	  to count object age.
+
+	* include/ruby/ruby.h: rewrite write barriers for incremental marking.
+
+	* debug.c: catch up flag name changes.
+
+	* internal.h: add rb_gc_writebarrier_remember() instead of
+	  rb_gc_writebarrier_remember_promoted().
+
+	* array.c (ary_memcpy0): use rb_gc_writebarrier_remember().
+
+	* array.c (rb_ary_modify): ditto.
+
+	* hash.c (rb_hash_keys): ditto.
+
+	* hash.c (rb_hash_values): ditto.
+
+	* object.c (init_copy): use rb_copy_wb_protected_attribute() because
+	  FL_WB_PROTECTED is moved from RBasic::flags.
+
+	* test/objspace/test_objspace.rb: catch up ObjectSpace.dump() changes.
+
+Sun Sep  7 12:47:06 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: PTHREAD_CANCEL_DISABLE is not defined on Android.
+
+Sat Sep  6 20:59:06 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (has_privilege): The gid zero is not a privilege.
+
+Sat Sep  6 20:19:16 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (struct child_handler_disabler_state): cancelstate field
+	  added.
+	  (disable_child_handler_before_fork): Record cancelstate.
+	  (disable_child_handler_fork_parent): Restore cancelstate.
+
+Sat Sep  6 19:27:10 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (struct child_handler_disabler_state): Defined.
+
+Sat Sep  6 18:31:32 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rake.rb, lib/rake/*, test/rake/*: Update latest rake master(e47d023)
+
+Sat Sep  6 16:38:08 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_variant.c (ole_val2variant_err,
+	  ole_val2variantdata, Init_win32ole_variant): support VT_ERROR
+	  variant with error code. add WIN32OLE_VARIANT::NoParam.
+	* test/win32ole/test_win32ole_variant.rb(test_c_noparam,
+	  test_vt_error_noparam): ditto.
+	* ext/win32ole/win32ole.c: ditto.
+
+Sat Sep  6 11:08:52 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (arg_ambiguous_gen): fix warning message, "even" does
+	  not mean the number of spaces here.  state the place to put a
+	  space and the operator.  [ruby-core:64790] [Bug #10204]
+
+Sat Sep  6 08:44:40 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/rdoc/generator/template/darkfish/js/jquery.js: Backport
+	  rdoc/rdoc@74f60fcb04fee1778fe2694d1a0ea6513f8e67b7
+
+Sat Sep  6 08:10:44 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_io.rb (test_readpartial_locktmp): use IO#nonblock=
+	  Old fcntl invocation may drop necessary flags on some platforms.
+
+Sat Sep  6 07:46:51 2014  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_io.rb (test_readpartial_locktmp): avoid EBADF
+	  [ruby-core:64773] [ruby-core:64775]
+
+Sat Sep  6 01:34:31 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_f_exec): Call before_exec_async_signal_safe and
+	  after_exec_async_signal_safe around rb_exec_async_signal_safe.
+	  (rb_exec_async_signal_safe): Don't call
+	  before_exec_async_signal_safe and after_exec_async_signal_safe.
+	  (rb_exec_without_timer_thread): Call before_exec and
+	  after_exec.
+	  (disable_child_handler_fork_child): Make SIGPIPE handler SIG_DFL.
+
+Sat Sep  6 00:49:41 2014  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (ruby_signal): Don't set SA_SIGINFO for SIG_IGN and
+	  SIG_DFL.
+
+Fri Sep  5 21:45:33 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (disable_child_handler_before_fork): New function.
+	  (disable_child_handler_fork_parent): Ditto.
+	  (disable_child_handler_fork_child): Ditto.
+	  (retry_fork_async_signal_safe): Call above functions to disable
+	  signal handlers in child process.
+
+Fri Sep  5 21:02:54 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (handle_fork_error): Make try_gc_p argument volatile to
+	  suppress "clobbered" warning.
+
+Fri Sep  5 20:48:06 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (handle_fork_error): Don't need state_p argument.
+
+Fri Sep  5 20:35:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (has_privilege): Fix a return value.
+
+Fri Sep  5 19:00:40 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/shellwords.rb: proofreading documentation.
+	  [Bug #10155][ruby-core:64471]
+
+Fri Sep  5 18:34:33 2014  Laurent Arnoud  <laurent@spkdev.net>
+
+	* test/csv/test_row.rb: Added some missing tests in CSV.
+	  [fix GH-710]
+	* test/csv/test_table.rb: ditto.
+
+Fri Sep  5 12:57:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (has_privilege): Refine uid/gid check.
+
+Fri Sep  5 12:40:55 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Check sys/id.h, getuidx and getgidx for AIX.
+
+	* process.c (getresuid): Defined for AIX.
+	  (getresgid): Ditto
+	  AIX don't have getresuid/getresgid but getuidx/getgidx.
+
+Fri Sep  5 12:28:21 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (has_privilege): Fix assignments.
+
+Fri Sep  5 11:10:13 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rdoc/generator/pot/po.rb: fixed broken tests for trailing whitespace.
+	* test/rdoc/test_rdoc_generator_pot.rb: ditto.
+	* test/rdoc/test_rdoc_generator_pot_po.rb: ditto.
+
+Fri Sep  5 10:41:07 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/rdoc/test_rdoc_rdoc.rb (TestRDocRDoc#test_parse_file_encoding):
+	  typofix.
+
+Fri Sep  5 10:39:14 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rdoc.rb, lib/rdoc, test/rdoc: Update to RDoc 4.2.0.alpha(313287)
+
+Fri Sep  5 06:04:22 2014  Eric Wong  <e@80x24.org>
+
+	* vm.c: remove unused USE_THREAD_RECYCLE [misc #10198]
+
+Fri Sep  5 00:29:08 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (dirfd): Check function.
+
+	* dir.c (dir_fileno): New method.
+	  [ruby-dev:48265] [Feature #9880]
+
+Thu Sep  4 23:39:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (has_privilege): New function.
+	  (retry_fork_async_signal_safe): Don't use vfork() for privileged
+	  process.
+
+	* configure.in (getresuid): Check function.
+	  (getresgid): Ditto.
+
+Thu Sep  4 20:22:14 2014  Laurent Arnoud  <laurent@spkdev.net>
+
+	* test/pathname/test_pathname.rb: added testcase for Pathname#mountpoint?.
+	  [fix GH-709]
+
+Thu Sep  4 20:09:21 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* symbian/*: removed Symbian support.
+	  [Feature #10199][ruby-core:64725]
+	* dln.c: ditto.
+	* include/ruby/defines.h: ditto.
+	* thread_pthread.c: ditto.
+	* vm.c: ditto.
+
+Thu Sep  4 17:44:40 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* dir.c (glob_helper): use #ifdef instead of #if.
+	  gcc's -Wundef option shows warning for undefined macro.
+
+	* numeric.c (flo_is_finite_p): ditto.
+
+	* vm_dump.c (rb_vmdebug_thread_dump_state): ditto.
+
+	* vm_core.h: define VM_DEBUG_VERIFY_METHOD_CACHE to 0.
+
+Thu Sep  4 03:57:46 2014  Eric Wong  <e@80x24.org>
+
+	* man/ruby.1: spelling fix ("bellow" => "below") [ci-skip]
+
+Thu Sep  4 03:52:16 2014  Eric Wong  <e@80x24.org>
+
+	* man/ruby.1: add trailing slash to URLs [ci-skip]
+
+Wed Sep  3 19:10:28 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (before_fork_ruby): Renamed from before_fork.
+	  (after_fork_ruby): Renamed from after_fork.
+
+Wed Sep  3 18:56:05 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (forked_child): Removed.
+
+Wed Sep  3 16:56:07 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_copy_ivar): allocate no memory for empty
+	  instance variables.  [ruby-core:64700] [Bug #10191]
+
+Wed Sep  3 12:05:17 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (retry_fork_async_signal_safe): Use vfork() if available.
+	  vfork() is still faster than fork() especially when the parent
+	  process uses big memory.
+
+	  ruby -rbenchmark -e 'a = "a" * 1_000_000_000; puts Benchmark.measure { system("true") }'
+	  fork:   0.000000   0.010000   0.010000 (  0.014968)
+	  vfork:  0.000000   0.000000   0.000000 (  0.000912)
+	  on Debian sid.
+
+Wed Sep  3 11:33:08 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/openssl/test_pkey_rsa.rb (OpenSSL#test_sign_verify_memory_leak):
+	  added timeout into testcase for low performance environment.
+	  [Bug #9984][ruby-core:63367]
+
+Wed Sep  3 07:50:15 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Use AC_FUNC_FORK.
+
+	* io.c: Use HAVE_WORKING_FORK instead of HAVE_FORK.
+
+	* process.c: Ditto.
+
+Wed Sep  3 00:12:44 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (retry_fork_async_signal_safe): Don't return on in child
+	  process.
+
+Tue Sep  2 23:47:35 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (retry_fork_async_signal_safe): Specialized version of
+	  retry_fork respect to rb_fork_async_signal_safe.
+	  (retry_fork_ruby): Specialized version of retry_fork respect to
+	  rb_fork_ruby.
+	  (rb_fork_ruby): Removed.
+
+Tue Sep  2 23:26:26 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (send_child_error): Simplified.
+	  (recv_child_error): Ditto.
+
+Tue Sep  2 22:56:25 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_fork_async_signal_safe): Inline rb_fork_internal.
+	  (rb_fork_ruby): Ditto.
+	  (rb_fork_internal): Removed.
+	  (chfunc_protect): Removed.
+
+Tue Sep  2 22:43:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb (test_new_with_block): Set autoclose to avoid
+	  EBADF.
+
+Tue Sep  2 22:01:51 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* Makefile.in (update-coverage): Remove a never executed line.
+
+Tue Sep  2 19:48:26 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (handle_fork_error): Extracted from retry_fork.
+
+Tue Sep  2 17:02:53 2014  Vit Ondruch  <v.ondruch@tiscali.cz>
+
+	* tool/rbinstall.rb: fixed error of local installation.
+	  [Bug #10192][ruby-core:64702]
+
+Tue Sep  2 16:58:03 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/runner.rb: reporting test coverage for test-all with COVERAGE env.
+	  [Feature #10189][ruby-core:64681][fix GH-708]
+	* Makefile.in: added task for coverage report.
+	* common.mk: added definition of forked simplecov url.
+	* .gitignore: ignored coverage directory.
+
+Mon Sep  1 20:11:02 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (rbtime2vtdate): try to convert millisecond
+	  of Time object to millisecond of VT_DATE VARIANT.
+	* test/win32ole/test_win32ole_variant.rb
+	  (test_conversion_time2date_with_msec): ditto.
+
+Sun Aug 31 16:58:49 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/benchmark.rb: Fix a syntax error.
+
+Sun Aug 31 08:46:44 2014  Eric Wong  <e@80x24.org>
+
+	* ext/zlib/zlib.c (gzfile_reset): preserve ZSTREAM_FLAG_GZFILE
+	  [Bug #10101]
+
+	* test/zlib/test_zlib.rb (test_rewind): test each_byte
+
+Sat Aug 30 19:22:47 2014  Eric Wong  <e@80x24.org>
+
+	* symbol.c (rb_sym2id): do not return garbage object
+
+Sat Aug 30 06:39:48 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: fix NameError dumping and
+	  loading. Fixes GH #85. Thanks @brentdax for the patch!
+	* test/psych/test_exception.rb: test for fix
+
+Sat Aug 30 06:23:40 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/scalar_scanner.rb: fix loading strings that
+	  look like integers but have a newline. Fixes GH #189
+	* test/psych/test_string.rb: test for fix
+
+Sat Aug 30 06:10:39 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: merge keys with a hash
+	  should merge the hash in to the parent.
+	* test/psych/test_merge_keys.rb: test for change. Fixes GH #202
+
+Sat Aug 30 06:00:26 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: quoted "<<" strings
+	  should not be treated as merge keys.
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: hashes with keys
+	  containing "<<" should roundtrip.
+	* test/psych/test_merge_keys.rb: test for change. Fixes GH #203
+
+Fri Aug 29 17:56:44 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/net/imap/test_imap_response_parser.rb: removed needless code.
+
+Fri Aug 29 17:36:58 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/rinda/test_rinda.rb: removed useless assignment variables.
+	* test/rss/rss-assertions.rb: ditto.
+	* test/rss/test_maker_itunes.rb: ditto.
+
+Fri Aug 29 16:18:26 2014  Eric Wong  <e@80x24.org>
+
+	* string.c: revert part of r47311, add rb_vm_fstring_table(),
+	  remove vm_core.h dependency. [ruby-core:64627]
+
+Fri Aug 29 15:17:13 2014  Eric Wong  <e@80x24.org>
+
+	* string.c: remove static frozen_strings
+	* string.c (Init_frozen_strings): new function
+	* string.c (rb_fstring): remove check for frozen strings,
+	  use per-VM table
+	* string.c (rb_str_free): use per-VM table
+	* string.c (Init_String): use per-VM table
+	* vm_core.h (rb_vm_t): add frozen_strings table
+	* internal.h (Init_frozen_strings): new function prototype
+	* eval.c (ruby_setup): call Init_frozen_strings
+	  [Feature #10182]
+
+Wed Aug 27 23:10:24 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* lib/tempfile.rb: remove "require 'thread'". its features are no
+	  longer used.
+
+Wed Aug 27 21:19:40 2014  gogo tanaka  <mail@tanakakazuki.com>
+
+	* lib/drb/acl.rb: Removed meaningless #to_s methods in interpolation.
+	  [Feature #10174][ruby-core:64584]
+	* lib/erb.rb: ditto.
+	* lib/observer.rb: ditto.
+	* lib/rake/invocation_chain.rb: ditto.
+	* lib/rubygems/command_manager.rb: ditto.
+	* lib/rubygems/config_file.rb: ditto.
+	* lib/uri/common.rb: ditto.
+
+Wed Aug 27 21:08:20 2014  gogo tanaka  <mail@tanakakazuki.com>
+
+	* lib/drb/drb.rb: use attr_reader instead of Module#attr.
+	  [Feature #10172][ruby-core:64582]
+	* lib/irb/ruby-token.rb: ditto.
+	* lib/net/telnet.rb: ditto.
+	* lib/rdoc/ruby_token.rb: ditto.
+	* lib/thwait.rb: ditto.
+
+Wed Aug 27 19:52:33 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (vtdate2rbtime): try to convert millisecond
+	  of VT_DATE VARIANT to nsec of Time object.
+	* test/win32ole/test_win32ole_variant.rb
+	  (test_conversion_dbl2date_with_msec): ditto.
+
+Wed Aug 27 09:57:29 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_complex.rb: removed unreachable code.
+	* test/ruby/test_rational.rb: ditto.
+
+Wed Aug 27 07:59:17 2014  Eric Wong  <e@80x24.org>
+
+	* compile.c (iseq_set_sequence): check for multiplication overflow
+
+Tue Aug 26 22:07:42 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_complex.rb: cherry-picked working assertions from r47251.
+	* test/ruby/test_rational.rb: cherry-picked working assertions from r47263.
+
+Tue Aug 26 21:07:56 2014  gogo tanaka  <mail@tanakakazuki.com>
+
+	* lib/mathn.rb (Fixnum#**, Bignum#**, Float#**, Rational#**):
+	  remove as these are now built-in.  [ruby-core:63973] [Bug #10086]
+
+Tue Aug 26 20:46:55 2014  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (rb_time_unmagnify_to_float): Avoid double rounding.
+	  Reported by Tsuyoshi Sawada.
+	  https://bugs.ruby-lang.org/issues/10135#note-1
+
+Tue Aug 26 17:12:47 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_close): ignore only "closed stream" IOError and
+	  NoMethodError, do not swallow other exceptions at the end of
+	  block.  [ruby-core:64463] [Bug #10153]
+
+Tue Aug 26 13:46:33 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* template/fake.rb.in: fix failed to make install when @srcdir@ is
+	  absolute path.
+
+Tue Aug 26 13:43:50 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_range.rb: added workaround for VERBOSE message.
+
+Tue Aug 26 12:38:02 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_mathn.rb: added workaround for VERBOSE messages.
+
+Tue Aug 26 11:44:04 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/shell/process-controller.rb: removed commented-out code.
+
+Tue Aug 26 11:39:01 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/thwait.rb (ThreadsWait): removed needless constant.
+
+Tue Aug 26 09:27:10 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/mathn.rb: mathn library is deprecated on ruby 2.2.
+	  [Feature #10169][ruby-core:64553]
+
+Tue Aug 26 09:25:03 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/mathn.rb: removed commented-out code.
+
+Mon Aug 25 20:15:50 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c(fole_s_connect, fole_initialize): raise a
+	  security error with the tainted string object.
+	* ext/win32ole/win32ole_event.c(ev_advise): ditto.
+	* test/win32ole/test_win32ole.rb(test_s_new_exc_svr_tainted,
+	  test_s_new_exc_host_tainted): ditto.
+	* test/win32ole/test_win32ole_event.rb(test_s_new_exc_tainted): ditto.
+
+Mon Aug 25 12:56:54 2014  Ivan Korunkov  <ivankorunkov@ya.ru>
+
+	* lib/logger.rb (format_datetime): use "%6N" to show microsecond.
+	  [Fix GH-704]
+
+Mon Aug 25 11:02:07 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_iseq_location_t): change first_lineno type to VALUE
+	* iseq.c (rb_iseq_build_for_ruby2cext): update based on argument
+
+Sun Aug 24 16:14:46 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/e2mmap.rb: remove needless instance variables.
+	* lib/irb.rb: ditto.
+	* lib/irb/**/*.rb: ditto.
+	* lib/shell.rb: ditto.
+
+Sun Aug 24 12:44:26 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_complex.rb: removed needless conditions.
+	* test/ruby/test_rational.rb: ditto.
+
+Sun Aug 24 11:47:39 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_rational.rb: fixed indent.
+
+Sun Aug 24 11:44:11 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_rational.rb: remove commented-out code.
+
+Sun Aug 24 11:09:29 2014  Eric Wong  <e@80x24.org>
+
+	* lib/benchmark.rb (measure): reduce allocations as in r47260
+
+Sun Aug 24 10:35:54 2014  Pete Higgins  <pete@peterhiggins.org>
+
+	* lib/benchmark.rb (module Benchmark): define BENCHMARK_CLOCK
+	  (realtime): use Process.clock_gettime(BENCHMARK_CLOCK)
+	  Reduces allocations to improve performance [Feature #10165]
+
+	* test/benchmark/test_benchmark.rb (test_realtime_output): new test
+
+Fri Aug 22 20:23:54 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (rb_fstring): fix condition (easy to cause infinite loop!).
+
+Fri Aug 22 20:07:43 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (rb_fstring, fstr_update_callback): simply delete garbage
+	  key first.
+
+	  Garbage keys can be swept by lazy sweeping invoked by creating new
+	  fstring.  So that simply do:
+	  (1) delete garbage key and return `fstr_update_callback' immediately
+	  (2) try again `fstr_update_callback()' to create a new fstr.
+
+	  This bug can be cause memory corruption, reported by
+	  http://u64.rubyci.org/~chkbuild/ruby-trunk/log/20140821T220302Z.fail.html.gz
+
+Fri Aug 22 19:30:39 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_complex.rb: removed commented-out code.
+
+Fri Aug 22 19:25:28 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_complex.rb: fixed broken tests. Math is not
+	  compatible CMath now.
+
+Fri Aug 22 15:36:09 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (Doxyfile): revert r43888, not to require preinstalled
+	  ruby.  [ruby-core:64488] [Bug #10161]
+
+Fri Aug 22 12:32:15 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/shell.rb: removed commented-out code.
+	* lib/shell/builtin-command.rb: ditto.
+	* lib/shell/command-processor.rb: ditto.
+
+Fri Aug 22 12:21:46 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/complex.rb: removed deprecated library.
+	* lib/rational.rb: ditto.
+
+Fri Aug 22 11:38:49 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/prettyprint.rb: removed PrettyPrint#first?
+	  because it is obsoleted method since Ruby 1.8.2
+
+Thu Aug 21 17:10:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (constat_attr): manage reverse video internally
+	  since Windows console window does not manage it.  based on the
+	  patch by white leaf in [ruby-dev:48483].  [Bug #10158]
+
+Thu Aug 21 14:45:41 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/e2mmap.rb: removed commented-out code.
+
+Thu Aug 21 13:23:34 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rinda/ring.rb: split executable code into sample directory.
+	* sample/rinda-ring.rb: ditto.
+
+Thu Aug 21 13:21:45 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/drb/acl.rb: split executable code into sample directory.
+	* sample/drb/acl.rb: ditto.
+
+Thu Aug 21 12:55:35 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .gitignore: ignored temporary file for Changelog.
+	  http://mkosaki.blog46.fc2.com/blog-entry-1284.html
+
+Thu Aug 21 12:40:22 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/net/imap.rb: split executable code into sample directory.
+	* sample/net-imap.rb: ditto.
+
+Thu Aug 21 12:23:56 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/net/imap.rb: removed commented-out code.
+
+Wed Aug 20 17:27:02 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_TRY_CFLAGS, RUBY_TRY_LDFLAGS),
+	  (RUBY_CHECK_BUILTIN_SETJMP, RUBY_SETJMP_TYPE),
+	  (RUBY_STACK_GROW_DIRECTION): quote defun names, for some
+	  versions of autoconf possibly.  [ruby-core:64473] [Bug #10156]
+
+Tue Aug 19 22:28:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): fix condition to round.
+	  [ruby-core:64454] [Bug #10151]
+
+Tue Aug 19 22:22:45 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/euckr-tbl.rb (EUCKR_TO_UCS_TBL): add missing euro and
+	  registered signs.  [ruby-core:64452] [Bug #10149]
+
+Tue Aug 19 13:59:43 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compar.c (rb_cmperr): preserve encodings of arguments in the
+	  message.
+
+Tue Aug 19 10:13:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/thread/thread.c (get_array): check instance variables are
+	  initialized properly.  [ruby-core:63826][Bug #10062]
+
+Mon Aug 18 17:06:27 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): support rational 'f' format.
+	  [ruby-core:64382] [Bug #10136]
+
+Mon Aug 18 08:03:46 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* spec/default.mspec: use 2.2 definition.
+
+Sun Aug 17 19:41:40 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gc.c (obj_memsize_of): don't calculate memsize of T_NODE
+	  when called from check_gen_consistency. It fixes segmentation
+	  fault on RGENGC_CHECK_MODE >= 1 introduced by r47188.
+
+Sun Aug 17 17:08:12 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): use RHASH_SIZE instead of
+	  calling Hash#length method.
+
+Sat Aug 16 19:32:06 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_event.c (evs_length): use RARRAY_LEN instead
+	  of calling Array#length method.
+
+Sat Aug 16 10:20:17 2014  Eric Wong  <e@80x24.org>
+
+	* time.c (time_timespec): fix tv_nsec overflow
+	  [Bug #10144]
+
+Fri Aug 15 20:34:17 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE_EVENT src from
+	  win32ole.c.
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_event.c: ditto.
+	* ext/win32ole/win32ole_event.h: ditto.
+	* ext/win32ole/depend: ditto.
+
+Fri Aug 15 19:38:00 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (rb_iseq_clone): Should not insert write barrier from
+	  non-RVALUE data (to non-RVALUE data, of course).
+
+	  Ruby 2.1 also has a same problem.
+
+Fri Aug 15 19:34:33 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (setup_fake_str): fake strings should not set class by
+	  RBASIC_SET_CLASS() because it insert write barriers to fake
+	  (non-RVALUE) structure.
+
+	  It can cause unexpected behaviour.
+
+	  Ruby 2.1 also have a same problem (setup_fake_str() in parse.y).
+
+	* symbol.c (setup_fake_str): ditto.
+
+Fri Aug 15 19:27:25 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_tmp_new_fill): added.
+	  This function creates internal use only array (which is completely
+	  hided by ObjectSpace.each_object) with filling nil.
+
+	  Otherwise, it can be includes strange VALUEs.
+
+	* internal.h: added.
+
+	* node.h: use rb_ary_tmp_new_fill() for MEMO.
+
+Fri Aug 15 10:13:37 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/gserver.rb: removed unmaintained code.
+	  [ruby-core:40313][Feature #5480]
+	* lib/xmlrpc/httpserver.rb: ditto.
+
+Fri Aug 15 09:22:12 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_thread_struct): reorder to pack
+	* cont.c (rb_context_struct, rb_fiber_struct): ditto
+	  On x86-64, these changes reduces:
+	  rb_thread_struct to 1000 bytes (from 1016)
+	  rb_context_struct to 1288 bytes (from 1312)
+	  rb_fiber_struct to 2272 bytes (from 2304)
+
+Fri Aug 15 09:06:31 2014  Eric Wong  <e@80x24.org>
+
+	* thread_pthread.h: define RB_NATIVETHREAD_LOCK_INIT and
+	  RB_NATIVETHREAD_COND_INIT macros
+
+	* thread_pthread.c (native_mutex_lock, native_mutex_unlock,
+	  native_mutex_trylock, native_mutex_initialize,
+	  native_mutex_destroy, native_cond_wait):
+	  use rb_nativethread_lock_t instead of pthread_mutex_t
+	  [Feature #10134]
+
+	* thread_pthread.c (native_mutex_debug): make argument type-agnostic
+	  to avoid later cast.
+
+	* thread_pthread.c (register_cached_thread_and_wait):
+	  replace PTHREAD_COND_INITIALIZER with RB_NATIVETHREAD_COND_INIT,
+	  use native_mutex_{lock,unlock}
+
+	* thread_pthread.c (use_cached_thread):
+	  use native_mutex_{lock,unlock}
+
+	* thread_pthread.c (native_sleep):
+	  use rb_nativethread_lock_t to match th->interrupt_lock,
+	  use native_mutex_{lock,unlock}
+
+	* thread_pthread.c (timer_thread_lock): use rb_nativethread_lock_t type
+
+Fri Aug 15 08:10:29 2014  Eric Wong  <e@80x24.org>
+
+	* cont.c (cont_mark): fix typo in unused path [ci skip]
+
+Fri Aug 15 06:00:56 2014  Eric Wong  <e@80x24.org>
+
+	* vm.c (rb_thread_mark): update comment about marking `me'
+	  [ruby-core:64340] [ruby-core:64341]
+
+Fri Aug 15 05:53:59 2014  Eric Wong  <e@80x24.org>
+
+	* README.EXT: preliminary documentation for RB_GC_GUARD
+	  [Bug #10100] [ruby-core:60741]
+
+Thu Aug 14 00:26:19 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE_RECORD src from
+	  win32ole.c
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_record.c: ditto.
+	* ext/win32ole/win32ole_record.h: ditto.
+	* ext/win32ole/depend: ditto.
+
+Wed Aug 13 21:41:04 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE_VARIANT src from
+	  win32ole.c.
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_variant.c: ditto.
+	* ext/win32ole/win32ole_variant.c: ditto.
+	* ext/win32ole/depend: ditto.
+
+Wed Aug 13 20:09:37 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: remove unused variable.
+
+Wed Aug 13 19:31:27 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_search_handler_method, rescue_callback,
+	  folerecord_inspect): use PRIsVALUE in rb_sprintf.
+
+	* ext/win32ole/win32ole_param.c (foleparam_inspect): ditto.
+
+	* ext/win32ole/win32ole_variable.c (folevariable_inspect): use
+	  PRIsVALUE in rb_sprintf, use rb_inspect.
+
+Wed Aug 13 11:54:41 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/irb/completion.rb: reverted r47163.
+	  because another case can't be show completion target.
+
+Wed Aug 13 11:17:00 2014  Shimpei Makimoto  <github@makimoto.org>
+
+	* lib/irb.rb: Prevent irb from crashing when exception with
+	  nil backtrace is raised.
+	  [fix GH-434][ruby-core:58078][Bug #9063]
+	* test/irb/test_raise_no_backtrace_exception.rb: ditto.
+
+Wed Aug 13 11:08:55 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/irb/completion.rb: fixed broken completion list with
+	  String including spaces. Contributed from @dunric. [fix GH-465]
+
+Wed Aug 13 00:07:01 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE_PARAM src from win32ole.c
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_param.c: ditto.
+	* ext/win32ole/win32ole_param.h: ditto.
+	* ext/win32ole/depend: ditto.
+
+Tue Aug 12 23:17:47 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: refactoring. move some methods
+	  into win32ole_type.c / win32ole_method.c
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_method.c: ditto.
+	* ext/win32ole/win32ole_method.h: ditto.
+	* ext/win32ole/win32ole_type.h: ditto.
+	* ext/win32ole/win32ole_type.h: ditto.
+
+Tue Aug 12 22:59:48 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: refactoring. move
+	  ole_typelib_from_itypeinfo into win32ole_typelib.c.
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_typelib.h: ditto.
+	* ext/win32ole/win32ole_typelib.h: ditto.
+
+Tue Aug 12 21:49:40 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_create_dcom): use the converted
+	  result if the argument can be converted to a string, to get rid
+	  of invalid access. Thanks to nobu. [ruby-dev:48467] [Bug #10127]
+
+Tue Aug 12 14:22:58 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* configure.in: ignored working directory same as prefix value.
+	  [ruby-core:54999] [Bug #8409]
+
+Tue Aug 12 13:34:25 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/fileutils.rb: enable to remove with non-owner directory.
+	  [ruby-dev:45976] [Bug #6756]
+	* test/fileutils/test_fileutils.rb: add testcase for #6756.
+
+Tue Aug 12 12:57:28 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* vm_exec.c: improve performance in ppc64 arch.
+	  [ruby-core:63437] [Feature #9997]
+
+Tue Aug 12 12:14:52 2014  Akira Matsuda <ronnie@dio.jp>
+
+	* lib/fileutils.rb: fix typo.
+	  [ruby-dev:47831] [Bug #9180]
+
+Tue Aug 12 10:10:42 2014  Eric Wong  <e@80x24.org>
+
+	* vm_method.c (release_method_definition): use rb_free_method_entry
+
+Tue Aug 12 06:16:09 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE_METHOD src from
+	  win32ole.c.
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_method.c: ditto.
+	* ext/win32ole/win32ole_method.h: ditto.
+	* ext/win32ole/depend: ditto.
+
+Mon Aug 11 22:19:15 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_variable.c (folevariable_inspect): refactoring.
+
+Mon Aug 11 20:47:27 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE_VARIABLE src from
+	  win32ole.c
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_variable.c: ditto.
+	* ext/win32ole/win32ole_variable.h: ditto.
+	* ext/win32ole/depend: ditto.
+
+Mon Aug 11 16:17:21 2014  Tony Miller  <mcfiredrill@gmail.com>
+
+	* dir.c (rb_dir_exists_p): [DOC] Document that Dir.exists? is
+	  deprecated.  [ruby-core:64135] [Bug #10102]
+
+Mon Aug 11 11:26:33 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/tempfile.rb: start rdoc parsing inside singleton class
+	  definition to include the document there.
+	  [ruby-core:64157] [Bug #10105]
+
+Sun Aug 10 12:22:43 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_type.c: refactoring.
+
+Sun Aug 10 10:34:00 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/cgi/session/pstore.rb: separated sample code.
+	* lib/open3.rb: ditto.
+
+Sun Aug 10 10:03:24 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/irb/ext/multi-irb.rb: removed commented-out code.
+
+Sat Aug  9 11:02:07 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/irb.rb: removed commented-out code.
+	* lib/irb/**/*.rb: ditto.
+
+Sat Aug  9 10:35:30 2014  Laurent Arnoud  <laurent@spkdev.net>
+
+	* lib/cmath.rb: fixed indent. [fix GH-696]
+	* lib/drb/ssl.rb: ditto.
+	* lib/irb/**/*.rb: ditto.
+
+Sat Aug  9 10:28:03 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/minitest/test_minitest_unit.rb: removed obsoleted condition
+	  for Ruby 1.8.
+	* test/ruby/test_time_tz.rb: ditto.
+
+Sat Aug  9 10:18:00 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/cgi/test_cgi_core.rb: removed obsoleted condition for Ruby 1.8.
+	* test/cgi/test_cgi_header.rb: ditto.
+	* test/cgi/test_cgi_multipart.rb: ditto.
+	* test/cgi/test_cgi_tag_helper.rb: ditto.
+
+Sat Aug  9 00:34:37 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE_TYPE src from win32ole.c.
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_type.c: ditto.
+	* ext/win32ole/win32ole_type.h: ditto.
+	* ext/win32ole/depend: ditto.
+
+Fri Aug  8 01:53:37 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* lib/securerandom.rb: use OpenSSL::BN for performance improvement.
+
+	* benchmark/bm_securerandom.rb: benchmark script.
+
+Fri Aug  8 17:19:57 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/open-uri.rb: remove needless condition for old ruby version.
+	* test/open-uri/test_open-uri.rb: ditto.
+
+Fri Aug  8 16:40:59 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/irb/init.rb: removed unreachable code.
+
+Fri Aug  8 16:34:22 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/drb/drb.rb: removed unreachable code.
+
+Fri Aug  8 14:33:49 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/webrick/httpproxy.rb: remove needless condition
+	  for old ruby version.
+
+Fri Aug  8 01:07:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): fix invalid char in eval, should raise
+	  an syntax error too, as well as directly coded.
+	  [ruby-core:64243] [Bug #10117]
+
+Thu Aug  7 23:25:29 2014  Masaki Matsushita <glass.saga@gmail.com>
+
+	* lib/open3.rb: avoid unnecessary write if stdin_data is empty.
+
+Thu Aug  7 21:42:49 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole_typelib.c (foletypelib_version): return
+	  version string.
+
+	* test/win32ole/test_win32ole_typelib.rb (test_version): ditto.
+
+Thu Aug  7 15:13:13 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/cgi.rb: remove needless condition for old ruby version.
+
+Thu Aug  7 06:04:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yyerror): preserve source code encoding in
+	  syntax error messages.  [ruby-core:64228] [Bug #10114]
+
+Wed Aug  6 20:56:02 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate src of WIN32OLE_TYPELIB from
+	  win32ole.c
+
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_typelib.c: ditto.
+	* ext/win32ole/win32ole_typelib.h: ditto.
+	* ext/win32ole/depend: ditto.
+
+Wed Aug  6 20:44:07 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* enum.c (enum_one): [DOC] Move enum.one? documentation before the
+	  relevant method.  Submitted by @vipulnsward. [Fixes GH-687]
+	  https://github.com/ruby/ruby/pull/687
+
+Wed Aug  6 20:25:47 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#replace): Check if an object given is enumerable
+	  before clearing self.  Reported by yui-knk. [GH-675]
+	  https://github.com/ruby/ruby/pull/675
+
+Wed Aug  6 20:07:26 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (olerecord_ivar_set): remove rb_str_subseq.
+
+Wed Aug  6 19:09:27 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set): Implement Set#clone by splitting
+	  initialize_copy into initialize_dup and initialize_clone.
+	  Submitted by yui-knk. [Fixes GH-661]
+	  https://github.com/ruby/ruby/pull/661
+
+Wed Aug  6 18:42:58 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate src of WIN32OLERuntimeError
+	  from win32ole.c.
+
+	* ext/win32ole/win32ole.h: ditto
+	* ext/win32ole/depend: ditto.
+	* ext/win32ole/win32ole_error.c: ditto.
+	* ext/win32ole/win32ole_error.h: ditto.
+
+Wed Aug  6 04:33:58 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP.proxy_uri): use initializer instead
+	  of parser to handle IPv6 address. [Bug #9129]
+
+Wed Aug  6 04:16:05 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/requests.rb (Net::HTTP::Options::RESPONSE_HAS_BODY):
+	  OPTIONS requests may have response bodies. [Feature #8429]
+	  http://tools.ietf.org/html/rfc7231#section-4.3.7
+
+Wed Aug  6 03:18:04 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/generic_request.rb (Net::HTTP::GenericRequest#exec):
+	  handle req['host'] in update_uri.
+
+	* lib/net/http/generic_request.rb
+	  (Net::HTTP::GenericRequest#update_uri):
+	  use req['host'] if it is explicitly set. Even if URI is given,
+	  it is already used for the initial value of req['host'].
+	  Therefore overwritten value should be respected. [Bug #10054]
+
+Wed Aug  6 03:17:34 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/generic_request.rb
+	  (Net::HTTP::GenericRequest#update_uri):
+	  handle scheme, host, and port to reflect connection to @uri.
+
+	* lib/net/http.rb (Net::HTTP#begin_transport): move trivial handling
+	  to Net::HTTP::GenericRequest#update_uri.
+
+
+Wed Aug  6 02:16:43 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/generic_request.rb
+	  (Net::HTTP::GenericRequest#initialize):
+	  optimize object allocation.
+
+Wed Aug  6 01:16:47 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#path_query): remove a private method.
+
+Wed Aug  6 01:15:47 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#normalize!): use String#empty?
+
+	* lib/uri/generic.rb (URI::Generic#path_query): optimized.
+
+	* lib/uri/generic.rb (URI::Generic#to_s): optimized.
+
+Wed Aug  6 00:15:10 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/http.rb (URI::HTTP#request_uri): optimized.
+	  decrease object allocation, and ensure always create at least one new
+	  object for return value.
+
+Wed Aug  6 03:41:21 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: backwards compatibility for
+	  hashes emitted by Syck. Github #198
+	* test/psych/test_hash.rb: test for change.
+
+Tue Aug  5 19:27:59 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): skip VariantClear when
+	  argument is VT_RECORD variant.
+
+Tue Aug  5 15:52:51 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gems/bundled_gems: Upgrade to test-unit-3.0.0 and minitest-5.4.0.
+
+Mon Aug  4 21:50:09 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_record.rb: add for WIN32OLE_RECORD
+	  test(need .NET Framework 3.5 to run test).
+
+Mon Aug  4 19:49:34 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): call rb_hash_foreach instead
+	  of rb_block_call.
+
+	* ext/win32ole/win32ole.c: add comment for rdoc of WIN32OLE_VARIANT
+	  class.
+
+Mon Aug  4 09:12:47 2014  Eric Wong  <e@80x24.org>
+
+	* variable.c: cleanup to use rb_const_lookup
+	  [Feature #10107]
+
+	* vm_insnhelper.c: ditto
+
+Sun Aug  3 10:55:07 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (rb_check_symbol_cstr): ditto.
+
+Sun Aug  3 10:43:08 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): unusable super class should cause
+	  method missing when BasicObject is refined but not been using.
+	  [ruby-core:64166] [Bug #10106]
+
+Sat Aug  2 23:47:45 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: separate WIN32OLE::VARIANT src file
+	  from win32ole.c
+	* ext/win32ole/win32ole.h: ditto.
+	* ext/win32ole/win32ole_variant_m.c: ditto.
+	* ext/win32ole/win32ole_variant_m.h: ditto.
+	* ext/win32ole/depend: ditto.
+	* ext/.document: ditto.
+
+Sat Aug  2 14:34:58 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add comments for rdoc.
+
+Sat Aug  2 10:26:57 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_itself): new method Object#itself.  based on the
+	  patch by Rafael Franca in [ruby-core:64156].
+	  [EXPERIMENTAL] this method may be renamed due to compatibilities.
+	  [ruby-core:44704] [Feature #6373]
+
+Fri Aug  1 22:30:40 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (folerecord_initialize): accept
+	  only 2 arguments. The 2nd argument should be WIN32OLE object or
+	  WIN32OLE_RECORD object.
+
+Fri Aug  1 20:17:33 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_variant2val): call
+	  folerecord_s_allocate instead of WIN32OLE_RECORD.new.
+
+Fri Aug  1 18:39:57 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/date/test_date.rb: remove commented-out code.
+	* test/date/test_date_arith.rb: ditto.
+	* test/date/test_date_attr.rb: ditto.
+	* test/date/test_date_parse.rb: ditto.
+
+Fri Aug  1 16:35:32 2014  Evan Miller  <evan@squareup.com>
+
+	* numeric.c (flodivmod): all results are NaN if divisor is NaN.
+	  [fix GH-692]
+
+Thu Aug 01 07:28:12 2014  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c: [DOC] Add description of
+	  `BigDecimal.new` exceptions.  Patched by @joker1007 and
+	  @prathamesh-sonpatki [Fixes GH-690]
+	  https://github.com/ruby/ruby/pull/690
+
+Thu Jul 31 22:20:12 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_RECORD#inspect.
+
+Thu Jul 31 20:35:32 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add
+	  WIN32OLE_RECORD#ole_instance_variable_set and
+	  WIN32OLE_RECORD#ole_instance_variable_get
+
+Wed Jul 30 23:28:10 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* sprintf.c (rb_str_format): like r47006, get rid of
+	  function calls in RSTRING_PTR().
+
+Wed Jul 30 22:10:29 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* process.c (rlimit_resource_type, rlimit_resource_value):
+	  get rid of inadvertent dynamic symbol pin-down.
+
+	* re.c (match_backref_number): ditto.
+
+	* signal.c (esignal_init, rb_f_kill, trap_signm): ditto.
+
+	* transcode.c (econv_opts): ditto.
+
+	* vm_trace.c (symbol2event_flag): ditto.
+
+Wed Jul 30 21:29:39 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): pass WIN32OLE_RECORD variant
+	  by reference when invoke OLE methods at first.
+
+	* ext/win32ole/win32ole.c (olerecord_set_ivar): release
+	  IRecordInfo interface before setting another IRecordInfo interface.
+
+Wed Jul 30 13:17:35 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove unused macros.
+
+Tue Jul 29 22:21:37 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (hash2olerec): ignore WIN32OLE_RECORD
+	  instance variable if the variable is nil.
+
+Tue Jul 29 19:43:27 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (folerecord_method_missing): refactoring.
+	  divide functionality of folerecord_method_missing into
+	  olerecord_ivar_set and olerecord_ivar_get.
+
+Mon Jul 28 20:20:08 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win23ole.c (folerecord_method_missing): support
+	  setter of member of WIN32OLE_RECORD object.
+
+Mon Jul 28 06:37:19 2014  Zachary Scott  <e@zzak.io>
+
+	* vm_eval.c: [DOC] Fix rdoc formatting of patch from [Bug #9551]
+
+Mon Jul 28 06:34:43 2014  Zachary Scott  <e@zzak.io>
+
+	* vm_eval.c: [DOC] [Bug #9551] Improve clarity of Kernel::catch
+	  documentation, patch by Jesse Sielaff.
+
+Mon Jul 28 06:24:54 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/uri/common.rb: [DOC] [Bug #9563] Recommend using URI.escape
+	  before parsing a uri to avoid invalid characters. Reported by
+	  Evgeniy Serykh.
+
+Mon Jul 28 05:55:56 2014  Zachary Scott  <e@zzak.io>
+
+	* time.c: [DOC] Clarify %Y in strftime, which can accept any digits
+	  and will output at least 4 digits as the year. Reported by Yury
+	  Trofimenko [Bug #10049]
+
+	* lib/time.rb: ditto
+
+Mon Jul 28 05:32:06 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/uri/common.rb: [DOC] [Bug #10075] Clarify how URI.join arguments
+	  are handled by RFC3986, originally reported by John Feminella.
+
+Mon Jul 28 05:21:41 2014  Zachary Scott  <e@zzak.io>
+
+	* file.c: [DOC] Clarify how File.file? handles symbolic links. Also
+	  cleaned up the rdoc style for this method, more to follow.
+	  Originally reported by Michael Renner [Bug #10067]
+
+Mon Jul 28 05:12:22 2014  Zachary Scott  <e@zzak.io>
+
+	* time.c: [DOC] Remove dead link and old bug report, which hasn't been
+	  reproduced in a few years. Reported by Federico Builes [Bug #10071]
+
+Mon Jul 28 04:39:58 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/zlib/zlib.c: [DOC] Remove default value of Zlib constants, as
+	  they may change in the implementation without notice. Patched by
+	  @robin850 [Fixes GH-682] https://github.com/ruby/ruby/pull/682
+
+Mon Jul 28 04:35:35 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_hmac.c: Fix NO_HMAC warning [Fixes GH-665]
+	  Patched by @vipulnsward https://github.com/ruby/ruby/pull/665
+
+Sun Jul 27 19:49:36 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/cgi/core.rb: remove unused variables.
+	* lib/erb.rb: ditto.
+	* lib/mkmf.rb: ditto.
+	* lib/net/http/response.rb: ditto.
+	* lib/optparse/version.rb: ditto.
+	* lib/prime.rb: ditto.
+	* lib/racc/parser.rb: ditto.
+	* lib/rexml/document.rb: ditto.
+	* lib/rexml/dtd/dtd.rb: ditto.
+	* lib/rexml/element.rb: ditto.
+	* lib/rexml/functions.rb: ditto.
+	* lib/rexml/parsers/xpathparser.rb: ditto.
+
+Sun Jul 27 05:11:21 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/irb.rb: [DOC] PROMPT_I cannot be nil, patch by @hgillane
+	  Fixes documenting-ruby/ruby#37
+	  https://github.com/documenting-ruby/ruby/pull/37
+
+Sun Jul 27 02:41:50 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/shell/command-processor.rb: remove unused variable.
+	* lib/shell/system-command.rb: ditto.
+	* lib/tmpdir.rb: ditto.
+	* lib/uri/generic.rb: ditto.
+
+Sun Jul 27 02:08:31 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/weakref.rb: split executable code into sample directory.
+	* sample/weakref.rb: ditto.
+
+Sun Jul 27 02:06:55 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/delegate.rb: split executable code into sample directory.
+	* sample/delegate.rb: ditto.
+
+Sun Jul 27 01:46:34 2014  Zachary Scott  <e@zzak.io>
+
+	* proc.c (method_super_method): [DOC] Method#super_method
+
+Sun Jul 27 01:22:39 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (method_super_method): new method Method#super_method,
+	  which returns a method object of the method to be called by
+	  `super` in the receiver method object.
+	  [ruby-core:62202] [Feature #9781]
+
+Sat Jul 26 17:22:14 2014  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* ext/objspace/objspace_dump.c (dump_append): avoid fflush.
+
+	  because dump_append_string_value() iterates over each chars,
+	  fflush()-ing here effectively issues system calls on every single
+	  bytes exist in a ruby process.
+
+Sat Jul 26 16:55:18 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.h (struct iseq_compile_data_storage): reduce overhead
+	  to 16 bytes (from 32) on 64-bit
+
+Sat Jul 26 16:28:06 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (struct rb_iseq_struct): reduce to 280 bytes
+	  (from 288 bytes) on 64-bit
+
+Sat Jul 26 06:44:43 2014  Eric Wong  <e@80x24.org>
+
+	* parse.y (struct parse_params): shrink to 320 to 304 bytes on 64-bit
+
+Sat Jul 26 05:58:35 2014  Eric Wong  <e@80x24.org>
+
+	* include/ruby/ruby.h (ZALLOC, ZALLOC_N): implement
+	  (Data_Make_Struct, TypedData_Make_Struct):
+	  ZALLOC replaces ALLOC+memset
+	  [ruby-core:63951][Feature #10082]
+	* compile.c (iseq_seq_sequence): ZALLOC_N replaces ALLOC_N+MEMZERO
+	* cont.c (fiber_t_alloc): ZALLOC replaces ALLOC+MEMZERO
+	* io.c (rb_io_reopen): ditto
+	* iseq.c (prepare_iseq_build): ditto
+	* parse.y (new_args_tail_gen, parser_new, ripper_s_allocate): ditto
+	* re.c (match_alloc): ditto
+	* variable.c (rb_const_set): ditto
+	* ext/socket/raddrinfo.c (get_addrinfo): ditto
+	* ext/strscan/strscan.c (strscan_s_allocate): ditto
+	* gc.c (rb_objspace_alloc): calloc replaces malloc+MEMZERO
+
+Sat Jul 26 05:54:54 2014  Eric Wong  <e@80x24.org>
+
+	* symbol.c (dsymbol_check): remove unneeded semi-colon
+
+Fri Jul 25 14:07:27 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change objspace::rgengc::parent_object_is_old (boolean)
+	  to objspace::rgengc::parent_object (VALUE).
+	  Use Qfalse or RVALUE pointer instead of FALSE and TRUE.
+
+	* gc.c (gc_marks_body): should clear parent_object just before
+	  gc_mark_roots() because there are no parents objects
+	  for root objects.
+
+Fri Jul 25 13:45:39 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_writebarrier_remember_promoted): should remember only
+	  OLD objects on RGENGC_AGE2_PROMOTION.
+
+Fri Jul 25 13:42:02 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_stacked_objects): fix error message.
+
+Fri Jul 25 13:18:00 2014  Will Farrington  <wfarrington@digitalocean.com>
+
+	* ext/socket/socket.c (sock_gethostname): Use NI_MAXHOST to support
+	  hostnames longer than 64 characters if the system supports it.
+	  [fixes GH-683]
+
+Fri Jul 25 12:21:11 2014  Santiago Pastorino  <santiago@wyeworks.com>
+
+	* compile.c (defined_expr): make the condition if the receiver
+	  is explicit or implicit cleaner.  [fix GH-681]
+
+Fri Jul 25 03:53:52 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* doc/keywords.rdoc:  [DOC] Describe each keyword.
+
+Thu Jul 24 22:40:24 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (Init_win32ole): add WIN32OLE_RECORD#initialize
+	  method.
+
+	* ext/win32ole/win32ole.c (ole_val2variant): convert WIN32OLE_RECORD
+	  object to VT_RECORD variant.
+
+Thu Jul 24 20:10:59 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix major GC flags.
+	  * add GPR_FLAG_MAJOR_BY_FORCE, which indicates
+	    major GC by METHOD, CAPI and so on (see GC_BY).
+	  * remove GPR_FLAG_MAJOR_BY_RESCAN because not used.
+	  * remove GPR_FLAG_MAJOR_BY_STRESS, use FORCE instead.
+
+	* test/ruby/test_gc.rb: catch up.
+
+Thu Jul 24 15:55:02 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/io.h (struct rb_io_buffer_t): PACKED_STRUCT should not
+	  be used for platform-specific optimization. PACKED_STRUCT_UNALIGNED
+	  should be used. [ruby-core:63988] [Bug #10088]
+
+Thu Jul 24 04:42:13 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/benchmark.rb: split executable code into sample directory.
+	* sample/benchmark.rb: ditto.
+
+Thu Jul 24 04:36:49 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/tempfile.rb: split executable code into sample directory.
+	* sample/tempfile.rb: ditto.
+
+Thu Jul 24 04:29:36 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/pstore.rb: split executable code into sample directory.
+	* sample/pstore.rb: ditto.
+
+Wed Jul 23 23:50:11 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/defines.h, siphash.c, st.c (UNALIGNED_WORD_ACCESS):
+	  add PowerPC64 too, which is capable to access unaligned words.
+	  patched by Gustavo Frederico Temple Pedrosa in [ruby-core:63937].
+	  [Feature #10081]
+
+	* regint.h (PLATFORM_UNALIGNED_WORD_ACCESS): ditto.
+
+Wed Jul 23 04:04:38 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/drb/extserv.rb: remove duplicate code with sample directory.
+	  contributed from @vipulnsward. [fix GH-679]
+
+Tue Jul 22 12:56:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_count): fix wrong single-byte optimization.
+	  7bit ascii can be a trailing byte in Shift_JIS.
+	  [ruby-dev:48442] [Bug #10078]
+
+Tue Jul 22 01:48:38 2014  Eric Wong  <e@80x24.org>
+
+	* include/ruby/io.h (rb_io_buffer_t): fix packing on gcc
+	  r46892 caused packing to be a no-op on gcc (4.7.2-5, Debian)
+	  [Bug #10079][ruby-core:63912]
+
+Mon Jul 21 15:55:42 2014  fuji70  <fujifuji70@gmail.com>
+
+	* lib/optparse.rb (getopts): print default values and descriptions
+	  in the help message.  [fix GH-676]
+
+Sun Jul 20 14:26:27 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_proc_t): reduce to 64 bytes from 72 on 64-bit
+
+Sun Jul 20 13:50:34 2014  Eric Wong  <e@80x24.org>
+
+	* transcode.c (rb_econv_t): reduce to 184 bytes from 200 on 64-bit
+
+Sun Jul 20 12:44:23 2014  Eric Wong  <e@80x24.org>
+
+	* include/ruby/io.h (rb_io_buffer_t): pack structure
+	  Reduces rb_io_t from 200 to 192 bytes, allowing rb_io_t to
+	  occupy one less cache line.
+	  [Feature #10050]
+
+Sun Jul 20 12:41:53 2014  Eric Wong  <e@80x24.org>
+
+	* include/ruby/io.h (rb_io_t): shrink to 200 bytes from 216 on 64-bit
+	  This puts us within 8 bytes of being three cache lines instead of
+	  four lines on x86-64.  This breaks the ABI.
+	  [Feature #10050]
+
+Sun Jul 20 12:36:46 2014  Eric Wong  <e@80x24.org>
+
+	* include/ruby/oniguruma.h (struct re_pattern_buffer): shrink to 448
+	  bytes from 464 bytes on 64-bit.  This breaks the ABI.
+	  [Feature #10034]
+
+Sun Jul 20 01:06:06 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/openssl/ossl.c: use encryptor instead of encrypter in doc.
+	  contributed from @vipulnsward. [fix GH-663]
+
+Sun Jul 20 00:32:44 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_initialize): [DOC] fix rdoc of append mode.  it does
+	  not move the pointer at open.  [ruby-core:63747] [Bug #10039]
+
+Sat Jul 19 12:40:50 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): allow to access private attribute
+	  reader in op_assign.  [ruby-core:63817] [Bug #10060]
+
+Sat Jul 19 11:56:36 2014  Grey Baker  <greysteil@gmail.com>
+
+	* lib/time.rb (Time#apply_offset): Guards against a `nil` return
+	  value from `Time.month_days` when offsetting date.  Out of range
+	  values are then caught when `Time.utc` is called (as usual).
+
+	  Previously a `nil` return value from `Time.month_days` would
+	  have the `<` operator called on it, and raise `NoMethodError`.
+	  [fix GH-667]
+
+	* lib/rdoc/parser/changelog.rb (RDoc#parse_entries): fix dirty hack.
+
+Sat Jul 19 06:19:01 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: refactoring.
+
+Fri Jul 18 22:34:41 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (folevariant_initialize): WIN32OLE_VARIANT
+	  does not support VT_RECORD. VT_RECORD should be supported in
+	  WIN32OLE_RECORD.
+
+	* test/win32ole/test_win32ole_variant.rb (test_s_new_vt_record_exc):
+	  ditto.
+
+Fri Jul 18 19:54:03 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (folevariant_initialize): remove unnecessary
+	  code.
+
+Fri Jul 18 19:11:03 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix/eigenvalue_decomposition: Style fix
+	  Patch by Gogo Tanaka [#10058]
+
+Fri Jul 18 19:03:53 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Avoid using `and`.
+	  Patch by gogo tanaka [#10058]
+
+Fri Jul 18 17:41:54 2014  GoGo tanaka  <qlli.illb@gmail.com>
+
+	* test/matrix/test_matrix.rb: Add tests for Matrix class.
+	  [Feature #10057][ruby-core:63809]
+
+Fri Jul 18 10:14:42 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/fileutils.rb: added missing options of FileUtils.touch by @Domon.
+	  [fix GH-669]
+
+Thu Jul 17 19:57:27 2014  Herwin  <herwin@quarantainenet.nl>
+
+	* ext/thread/thread.c (rb_szqueue_push): add optional parameter,
+	  non_block defaulted to false.  [ruby-core:63794] [Feature #10052]
+
+Wed Jul 16 23:01:43 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_variant2val): support array of
+	  VT_RECORD variant.
+
+Wed Jul 16 20:21:49 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* vm_core.h (struct rb_iseq_struct): stack_max is changed to int
+	  because all calculations related to stack_max in compile.c
+	  (iseq_set_sequence) and vm_insnhelper.c (vm_push_frame) are
+	  conducted by using int. This partly reverts r23945.
+	* vm_insnhelper.c (vm_push_frame): ditto. This reverts r42401.
+
+Wed Jul 16 19:55:32 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* vm_core.h (struct rb_iseq_struct): temporal workaround of [Bug 10037].
+	  Add padding on big-endian 64-bit architecture (e.g. sparc64).
+
+Wed Jul 16 19:32:23 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fole_record_method_missing): call
+	  rb_hash_fetch instead of rb_hash_aref.
+
+Wed Jul 16 18:08:47 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (rb_iseq_defined_string): use rb_gc_mark_object() instead of
+	  marking from vm_mark().
+
+	* vm.c (rb_vm_mark): ditto.
+
+Wed Jul 16 18:03:50 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_roots): call rb_vm_mark directly.
+
+	* vm.c: remove mark function for RubyVM object because
+	  RubyVM object marked manually.
+
+Wed Jul 16 12:25:39 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c: Merge Onigmo 5.14.1 25a8a69fc05ae3b56a09.
+	  this includes Support for Unicode 7.0 [Bug #9092].
+
+Tue Jul 15 23:59:27 2014  Jared Jennings  <jared.jennings.ctr@us.af.mil>
+
+	* ext/digest: make built-in digest function implementations
+	  indicate success or failure of init and final functions.
+	  [ruby-core:61614] [Bug #9659]
+
+	* ext/digest/digest.c: expect digest init and finish functions to
+	  indicate success or failure; raise exception on failure.
+	  [ruby-core:61614] [Bug #9659]
+
+Tue Jul 15 20:31:40 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: modify document for WIN32OLE_RECORD.
+
+Tue Jul 15 12:42:23 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* defs/default_gems: change version definition file of rake.
+
+Tue Jul 15 12:00:03 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rake.rb, lib/rake/*.rb: Upgrade to rake-10.3.2
+	  [fix GH-668]
+	* test/rake/*.rb: ditto.
+
+Mon Jul 14 19:14:51 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: modify WIN32OLE class document and
+	  add comment for constants of WIN32OLE.
+
+Mon Jul 14 16:38:45 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (struct rb_iseq_struct): reduce to 288 bytes
+	  on x86-64 (from 296 bytes)
+
+Mon Jul 14 16:07:25 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.h (struct iseq_catch_table_entry): shrink to 32 bytes
+	  on x86-64 (from 48 bytes)
+
+Mon Jul 14 16:04:41 2014  Eric Wong  <e@80x24.org>
+
+	* iseq.h (struct iseq_catch_table): new flexible array struct
+		 (iseq_catch_table_bytes): allocated size function
+	* vm_core.h (struct rb_iseq_struct): update catch_table member
+	  This reduces the struct from 304 to 296 bytes on x86-64.
+	* compile.c (iseq_set_exception_table): update for struct changes
+	* iseq.c (iseq_free): ditto
+	* iseq.c (iseq_memsize): ditto
+	* iseq.c (rb_iseq_disasm): ditto
+	* iseq.c (iseq_data_to_ary): ditto
+	* iseq.c (rb_iseq_build_for_ruby2cext): ditto (untested)
+	* vm.c (vm_exec): ditto
+	* vm_core.h (struct rb_iseq_struct): ditto
+	* vm_insnhelper.c (vm_throw): ditto
+
+Sun Jul 13 17:49:52 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/openssl/ossl_cipher.c: Fix call to ciphers class method and
+	  spell out `encryption` by @vipulnsward [fix GH-664]
+
+Sun Jul 13 17:31:51 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/gdbm/gdbm.c: fix wrong arguments in GetDBM2 macro.
+	* ext/sdbm/init.c: ditto.
+
+Sun Jul 13 17:25:50 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/dbm/dbm.c: fix wrong arguments in GetDBM2 macro by @v2e4lisp.
+	  [fix GH-655]
+
+Sun Jul 13 16:44:56 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (struct rb_call_info_struct): improve packing
+	  This reduces the struct from 112 to 104 bytes on x86-64.
+
+Sun Jul 13 15:53:25 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (struct rb_iseq_struct): stack_max is uint32_t
+	  This reduces the struct from 312 to 304 bytes on x86-64.
+
+Sun Jul 13 10:56:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_broken_backtrace): exit with failure
+	  normally, no needs to abort.  [ruby-core:63678] [Bug #10008]
+
+Sat Jul 12 15:10:22 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (fole_record_method_missing): correct
+	  fields Hash key.
+
+Sat Jul 12 04:17:40 2014  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/net/smtp.rb (Net::SMTP#data): enable buffering while
+	  'data' send for optimizing Net::SMTP#send_message.
+	  [ruby-dev:48329] [misc #9981]
+	  patch by Masahiro Tomita.
+
+Sat Jul 12 01:13:45 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/ruby/envutil.rb (assert_no_memory_leak): On Solaris 9 or later,
+	  if possible, execute child ruby with environment variables
+	  LD_PRELOAD=libumem.so UMEM_OPTIONS="backend=mmap". With these
+	  variables, freed memory is immediately returned to the OS.
+	  [Bug #10020] [ruby-dev:48391]
+
+Fri Jul 11 20:49:10 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c: add WIN32OLE_RECORD class to support
+	  VT_RECORD OLE variables.
+
+Fri Jul 11 17:15:08 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/abbrev.rb: remove executable.
+
+Fri Jul 11 16:45:39 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/fileutils.rb: handle ENOENT error with symlink targeted to
+	  non-exists file. [ruby-dev:45933] [Bug #6716]
+
+Fri Jul 11 15:59:42 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* array.c: Clarify documentation for Array#insert.
+	  [ruby-core:62934] [Bug #9901]
+
+Fri Jul 11 15:39:36 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* io.c: Improve Documentation by @dapplebeforedawn.
+	  [fix GH-658] [ruby-core:63579] [Bug #10012]
+
+Fri Jul 11 14:19:14 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Fix sign for cross_product [#9499]
+
+Fri Jul 11 11:11:50 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/prepare_so_k_nucleotide.rb: use require_relative.
+
+	* benchmark/prepare_so_reverse_complement.rb: ditto.
+
+Fri Jul 11 10:09:03 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (encodes): fix buffer overrun by tail_lf.  Thanks to
+	  Mamoru Tasaka and Tomas Hoger.  [ruby-core:63604] [Bug #10019]
+
+Thu Jul 10 23:51:36 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* hash.c (ruby_setenv): Fix TestEnv#test_aset failure on Solaris 9.
+	  When name contains '=', ruby_setenv raises Errno::EINVAL.
+	  That is the same behavior as Solaris 10.
+	  NULL check for malloc return value is also added.
+
+Thu Jul 10 15:02:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_keyword_arg): adjust VM stack
+	  pointer to get rid of overwriting splat arguments by arguments
+	  for `to_hash` conversion.  [ruby-core:63593] [Bug #10016]
+
+Thu Jul 10 01:09:57 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* symbol.c: remove rb_gc_mark_symbols().
+
+	  fstrings referred by static symbols and pinned dynamic symbols
+	  are registered by rb_gc_register_mark_object().
+
+	  fstrings referred by dynamic symbols (not pinned symbols)
+	  are referred from global_symbols.dsymbol_fstr_hash (Hash object).
+
+	  Note that fstrings referred from dynamic symbols must live logger
+	  than symbol objects themselves because rb_gc_free_dsymbol() uses
+	  fstrings to remove from symbol tables.
+	  This is why we can not mark fstrings from dynamic symbols.
+
+	  This technique reduces root objects for GC marking.
+
+	* gc.c (gc_mark_roots): ditto.
+
+	* internal.h: ditto.
+
+Thu Jul 10 00:24:18 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* common.mk (DTRACE_DEPENDENT_OBJS): fix build failure on Solaris
+	  introduced in r46768. Object files containing dtrace probes should
+	  be listed in DTRACE_DEPENDENT_OBJS.
+
+Wed Jul  9 17:07:28 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.c, symbol.h: Symbol class implementation and internals,
+	  split from parse.y.
+
+Wed Jul  9 14:45:39 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: change Symbol <-> ID relationship to avoid
+	  exposing IDs from collectable symbols.
+	  [Bug #10014]
+
+	  Now, rb_check_id() returns 0 if corresponding symbol is
+	  pinned dynamic symbol.
+
+	  There is remaining intern_cstr_without_pindown(), it can return
+	  IDs from collectable symbols. We must be careful to use it
+	  (only used in parse.y).  I think it should be removed if
+	  it does not have impact for performance.
+
+	* parse.y:
+	  add:
+	  * STATIC_SYM2ID()
+	  * STATIC_ID2SYM()
+	  rename:
+	  * rb_pin_dynamic_symbol() -> dsymbol_pindown()
+
+	* internal.h:
+	  remove:
+	  * rb_check_id_without_pindown()
+	  * rb_sym2id_without_pindown()
+	  add:
+	  * rb_check_symbol()
+	  * rb_check_symbol_cstr()
+
+	* load.c: use rb_check_id() or rb_check_id_cstr().
+
+	* object.c: ditto.
+
+	* struct.c: ditto.
+
+	* thread.c: ditto.
+
+	* vm_method.c: ditto.
+
+	* string.c (sym_find): use only rb_check_symbol().
+
+	* sprintf.c (rb_str_format): use rb_check_symbol_cstr().
+
+Wed Jul  9 12:21:55 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (symbols_i): delete garbage symbols for Symbol.all_symbols.
+
+Wed Jul  9 05:49:08 2014  Eric Wong  <e@80x24.org>
+
+	* thread_pthread.h (struct rb_global_vm_lock_struct):
+	  do not expose pthread type for lock
+
+Wed Jul  9 05:41:40 2014  Eric Wong  <e@80x24.org>
+
+	* thread_pthread.h: remove unneeded semaphore.h include
+
+Wed Jul  9 00:12:28 2014  Keiju Ishitsuka  <keiju@ishitsuka.com>
+
+	* lib/irb/ruby-lex.rb: fix counting indent in identify_string_dvar.
+
+Tue Jul  8 16:58:02 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/memory_status.rb (Memory::PSCMD): use ps command which
+	  outputs expected result.  [ruby-dev:48370] [Bug #10010]
+
+Tue Jul  8 14:45:17 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (dsymbol_alloc): set global_symbols.minor_marked to 0.
+
+	* parse.y (dsymbol_check): set RSYMBOL(sym)->fstr to 0
+	  because we should not touch fstr after that.
+
+	* parse.y (rb_gc_free_dsymbol): skip deleting str and sym
+	  from tables if fstr == 0.
+
+Mon Jul  7 14:31:52 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: remove global_symbols::pinned_dsym
+	  (and ::pinned_dsym_minor_marked).
+
+	  Mark pinned dsymbols by rb_gc_register_mark_object() because
+	  they are immortal.
+
+	* parse.y (rb_gc_free_dsymbol): rename parameter name `ptr' to `sym'.
+
+Mon Jul  7 12:45:51 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: revert miss-commit.
+
+Mon Jul  7 12:40:59 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: need to use updated (re-created) symbols.
+
+Mon Jul  7 11:02:55 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/mkconfig.rb: remove not to require rbconfig/obsolete.rb.
+
+	* lib/rbconfig/obsolete.rb: removed.
+
+Mon Jul  7 10:52:03 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: do not use rb_gc_resurrect(), but create a new dynamic
+	  symbol for garbage dynamic symbol.
+
+	* common.mk: use gc.h by parse.y.
+
+Mon Jul  7 02:18:42 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (fstr_update_callback): do not use rb_gc_resurrect()
+	  any more.
+
+	  Make new frozen string and replace with garbage frozen string.
+
+	* common.mk: use gc.h from string.c.
+
+Mon Jul  7 00:36:13 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename is_dying_object() to is_garbage_object().
+
+	* gc.h: rb_objspace_garbage_object_p() as an exported function.
+
+Sun Jul  6 21:30:35 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (is_dying_object): fix missed condition.
+
+	* gc.c (is_live_object): move frequent path first.
+
+Sun Jul  6 21:00:11 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename is_dead_object() to is_dying_object().
+	  This function is not opposite against is_live_object()
+	  because is_dying_object() does *not* check object type.
+
+	* gc.c (is_dying_object): change condition.
+
+	* gc.c (is_live_object): use T_NONE instead of 0.
+
+	* gc.c (rb_objspace_dying_object_p): added.
+
+Sun Jul  6 13:37:27 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_register_mark_object): change data structure.
+	  From single array, to array of arrays. Each array only has 1024
+	  entries.
+
+	* vm.c (Init_vm_objects): change default capa from 1 to 128.
+
+Sat Jul  5 05:05:53 2014  Vipul A M  <vipulnsward@gmail.com>
+
+	* lib/irb/locale.rb (IRB::Locale#modifier): fix wrong attr_reader
+	  `modifieer` => `modifier` from irb locale.  [fix GH-656]
+
+Fri Jul  4 20:45:26 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: rename symbols::sym_id to symbols::str_id.
+	  This table is not {Symbol => ID} table, but
+	  {String => ID} table.
+
+	* parse.y (lookup_sym_id): also rename lookup_sym_id() to
+	  lookup_str_id() because key is not Symbol, but String.
+
+Fri Jul  4 18:42:04 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (must_be_dynamic_symbol): fix missed-condition.
+
+Fri Jul  4 18:38:11 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (rb_pin_dynamic_symbol): should be `static' function.
+
+Fri Jul  4 18:03:35 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y (must_be_dynamic_symbol): refactoring.
+	  * add `inline'.
+	  * use UNLIKELY().
+	  * check only DYNAMIC_SYM_P(), otherwise it is a bug.
+	  * lookup_id_str() is not needed in second condition.
+
+Fri Jul  4 11:53:56 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y: remove unused code
+	  surrounded by `#if ENABLE_SELECTOR_NAMESPACE'
+
+Fri Jul  4 10:08:24 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/rubygems/test_gem_package.rb: avoid tempfile leaks using Tempfile#close!
+	* test/rubygems/test_gem_request_set.rb: ditto.
+	* test/rubygems/test_gem_request_set_gem_dependency_api.rb: ditto.
+
+Fri Jul  4 04:42:05 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/response.rb (Net::Inflater#inflate_adapter):
+	  prevent automatic encoding conversion.
+
+Fri Jul  4 04:39:52 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/response.rb (Net::HTTPResponse.each_response_header):
+	  raise first exception even if inflate_body_io.finish raises error.
+	  when begin block raises error, finish usually raises error too.
+
+Fri Jul  4 02:56:04 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#query=): remove validation, just
+	  escape. [Feature #2542]
+
+	* lib/uri/generic.rb (URI::Generic#fragment=): ditto.
+
+	* lib/uri/generic.rb (URI::Generic#check_query): removed.
+
+	* lib/uri/generic.rb (URI::Generic#set_query): ditto.
+
+	* lib/uri/generic.rb (URI::Generic#check_fragment): ditto.
+
+	* lib/uri/generic.rb (URI::Generic#set_fragment): ditto.
+
+Thu Jul  3 12:40:22 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (--with-static-linked-ext): fix for extensions to
+	  be linked statically.
+
+	* Makefile.in, common.mk: use ENCSTATIC for enc directory.
+
+	* ext/extmk.rb: supply dependencies of statically linked extension
+	  libraries.
+
+Wed Jul  2 15:45:49 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_heap_lazy_sweep): simplify logic.
+
+	* gc.c (gc_page_sweep): return TRUE if empty slots are available.
+
+Wed Jul  2 09:48:42 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* logger.rb: removed unmaintained code.
+	  [Feature #9860][ruby-core:62724]
+	* test/logger/test_application.rb: ditto.
+
+Wed Jul  2 03:20:00 2014  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* node.c (dump_node): handle nd_value == (NODE *)-1 to mean this
+	  keyword argument is required
+
+Wed Jul  2 02:57:27 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (rb_vm_env_local_variables): returns array of local
+	  variable name symbols in the environment by envval.
+
+	* proc.c (bind_local_variables): use rb_vm_env_local_variables.
+
+Wed Jul  2 02:23:52 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (bind_receiver): new method to return the bound receiver
+	  of the binding object.  [ruby-dev:47613] [Feature #8779]
+
+Wed Jul  2 02:14:37 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (bind_local_variables): update env from envval for each
+	  iterations.  [ruby-dev:48351] [Bug #10001]
+
+Tue Jul  1 23:46:34 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* NEWS: [DOC] mention about Binding#local_variables, introduced at
+	  r44392 (see [Feature #8773]).
+
+Tue Jul  1 23:30:51 2014  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* numeric.c (num_step_scan_args): table argument of rb_get_kwargs() is
+	  array of IDs, not Symbols. [ruby-dev:48353] [Bug #9811]
+
+Tue Jul  1 16:18:22 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* ext/digest/lib/digest/hmac.rb, test/digest/test_digest_hmac.rb:
+	  Digest::HMAC is finally removed as previously noticed.
+	  [fix GH-648]
+
+Tue Jul  1 11:13:43 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/date/lib/date/format.rb: removed empty file by @vipulnsward.
+	* ext/date/lib/date.rb: removed needless require.
+	  [fix GH-647]
+
+Mon Jun 30 16:42:52 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_stat_internal): return size_t value instead of VALUE
+	  and remove `out' parameter.
+
+	* gc.c: add braces for `if' statements.
+
+	* gc.c (gc_stat_internal): fix comment.
+
+Mon Jun 30 15:07:34 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: support `USE_RGENGC == 0'.
+
+	* test/ruby/test_gc.rb: ditto.
+
+Mon Jun 30 11:36:04 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* file.c: [DOC] document File.join returns a string.
+	  Contributed by @dapplebeforedawn. [fix GH-646]
+
+Sat Jun 28 22:57:01 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/pathname/pathname.c (path_birthtime): Windows support.
+	  see [Feature #9857] [ruby-dev:48339]
+
+Sat Jun 28 22:44:16 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (path_birthtime): New method,
+	  Pathname#birthtime.
+	  Proposed by Kazuhiro NISHIYAMA.  [ruby-dev:48232] [Feature #9857]
+
+Sat Jun 28 20:29:03 2014  Simon Baird  <simon.baird@gmail.com>
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb (BigMath#PI): change error
+	  message about zero or negative precision for clarity and
+	  consistency with other methods.  [GH-644]
+
+Sat Jun 28 15:32:57 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/utils.rb (create_listeners): Close socket objects.
+
+Sat Jun 28 13:58:48 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (setup_exception): should not overwrite SystemStackError
+	  backtrace if set already.  [ruby-core:63377] [Feature #6216]
+
+	* eval.c (setup_exception): get rid of method calls before raising
+	  stack overflow, not to cause stack overflow again.
+
+	* defs/id.def: add IDs for backtraces.
+
+Sat Jun 28 04:08:22 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/mailto.rb: update to latest specs, RFC 6068 and HTML5.
+
+	* lib/uri/mailto.rb (HEADER_PATTERN): removed.
+
+	* lib/uri/mailto.rb (HEADER_REGEXP): use RFC 6068 hfields.
+
+	* lib/uri/mailto.rb (EMAIL_REGEXP): use HTML5 email regexp.
+
+	* lib/uri/mailto.rb (URI::MailTo.build): support multiple to addresses.
+
+	* lib/uri/mailto.rb (URI::MailTo#initialize): Support multiple to
+	  addresses. Don't check with regexp, only split.
+
+	* lib/uri/mailto.rb (URI::MailTo#check_to): verify by matching
+	  URI path-rootless and HTML5 email regexp with unescaped one.
+
+	* lib/uri/mailto.rb (URI::MailTo#check_headers): verify only by
+	  HEADER_REGEXP.
+
+	* lib/uri/mailto.rb (URI::MailTo#set_headers): don't check by
+	  HEADER_REGEXP, only split it.
+
+Sat Jun 28 00:35:10 2014  Lauri Tirkkonen  <lotheac@iki.fi>
+
+	* tool/mkconfig.rb: fix empty RbConfig::CONFIG["prefix"] when
+	  configured --with-rubyarchprefix, remove prefix from rubyarchdir
+	  after expansion for the case it does not start with '$(prefix)'.
+	  [fix GH-643]
+
+Fri Jun 27 15:20:12 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems/test_case.rb: rescue Gem::LoadError in Gem::TestCase.
+	  because it's effected by removing minitest from stdlib.
+
+Fri Jun 27 12:29:37 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems/specification.rb: fixed broken condition caused
+	  by removing YAML::ENGINE.
+	* lib/rubygems/package/old.rb: ditto.
+
+Fri Jun 27 05:33:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_shift): fix memory leak on Windows, free environment
+	  strings block always.  [ruby-dev:48332] [Bug #9983]
+
+Fri Jun 27 03:41:53 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (GETASTER): should not use the numbered argument to be
+	  formatted, raise ArgumentError instead.
+	  [ruby-dev:48330] [Bug #9982]
+
+Thu Jun 26 18:18:28 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/with_different_ofs.rb: move into test library directory.
+	* test/csv/base.rb: fix require path for with_different_ofs.rb.
+	* test/digest/test_digest_extend.rb: ditto.
+
+Thu Jun 26 18:06:50 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/profile_test_all.rb: move into test library directory.
+	* test/runner.rb: fix require path for profile_test_all.rb.
+
+Thu Jun 26 17:57:57 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/webrick/httpproxy.rb: remove useless assigned variables.
+	* lib/webrick/httpservlet/cgihandler.rb: ditto.
+	* lib/webrick/httpservlet/erbhandler.rb: ditto.
+	* lib/webrick/server.rb: ditto.
+
+Thu Jun 26 08:28:01 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_select): fix memory leak and crash on Windows, make
+	  keys array first instead of iterating on environ directly.
+	  [ruby-dev:48325] [Bug #9978]
+
+Thu Jun 26 02:45:04 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_error.c (error_print): put a newline after an anonymous
+	  exception class name.
+
+Wed Jun 25 22:31:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (ruby_setenv): fix memory leak on Windows, free
+	  environment strings block after check for the size.
+	  [ruby-dev:48323] [Bug #9977]
+
+Wed Jun 25 15:44:12 2014  Eric Wong  <e@80x24.org>
+
+	* ccan/container_of/container_of.h (container_off_var):
+	  avoid warning with -Wcast-qual
+	  [ccan ba5ad771af4aa9e085498de6c3c665c52694460f (Rusty Russell)]
+
+Wed Jun 25 10:19:59 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_aset, env_has_key, env_assoc, env_has_value),
+	  (env_rassoc, env_key): prohibit tainted strings if $SAFE is
+	  non-zero.  [Bug #9976]
+
+Tue Jun 24 14:46:17 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/gserver.rb: remove redundant use of to_s in interpolation.
+	* lib/logger.rb: ditto.
+	* lib/optparse.rb: ditto.
+	* lib/rbconfig/obsolete.rb: ditto.
+	* lib/resolv.rb: ditto.
+	* lib/webrick/httpresponse.rb: ditto.
+
+Tue Jun 24 10:50:06 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_path_str_new): make PATH environment variable
+	  string, to be frozen.
+
+Tue Jun 24 10:40:52 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/make-snapshot: download bundle gems when package making.
+	  [Feature #9852][ruby-core:62676]
+	* gems/bundled_gems: listed bundled gems for Ruby 2.2.
+
+Tue Jun 24 10:20:35 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/downloader.rb: make Downloader class to general download utility.
+	  It can be used without config.guess and config.sub.
+	* tool/get-config_files: ditto.
+	* tool/make-snapshot: ditto.
+
+Tue Jun 24 06:17:52 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* eval.c (setup_exception): "mesg == sysstack_error" and
+	  sysstack_error_p(mesg) are duplicated.
+	  r46502 seems to want to use latter.
+
+Tue Jun 24 06:15:36 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c: fix format specifiers for VALUE and
+	  Tcl_Interp*.  [ruby-core:63283] [Bug #9972]
+
+Tue Jun 24 05:40:41 2014  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* nacl/nacl-config.rb: Use File.exist? instead of executable?
+	  for irt_core. Recent nacl_sdk has non-executable irt_core.
+	  Patch by Shinichiro Hamaji.
+	  [Fixes GH-529] https://github.com/ruby/ruby/pull/529
+
+Mon Jun 23 18:44:45 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/config_files.rb: rename class ConfigFiles to Downloader.
+	* tool/get-config_files: ditto.
+	* tool/make-snapshot: ditto.
+
+Mon Jun 23 18:03:13 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/rbinstall.rb: support to install bundle gems.
+
+Mon Jun 23 17:33:11 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/net/imap.rb (Net::IMAP#fetch): [DOC] Describe how a range in
+	  +set+ is interpreted, and mention -1 which can be used for '*'.
+
+Mon Jun 23 16:22:50 2014  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* include/ruby/ruby.h (struct RHash): no longer. [Feature #9889]
+
+	* include/ruby/ruby.h (RHASH): ditto.
+
+	* include/ruby/ruby.h (RHASH_ITER_LEV): deprecated.  Will be deleted later.
+
+	* include/ruby/ruby.h (RHASH_IFNONE): ditto.
+
+	* internal.h (struct RHash): moved here.
+
+	* internal.h (RHASH): ditto.
+
+	* hash.c (rb_hash_iter_lev): do not use this.
+
+	* hash.c (rb_hash_ifnone): ditto.
+
+Mon Jun 23 13:30:11 2014  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* include/ruby/ruby.h (struct RComplex): no longer. [Feature #9888]
+
+	* include/ruby/ruby.h (RCOMPLEX): ditto.
+
+	* include/ruby/ruby.h (RCOMPLEX_SET_REAL): deprecated.  Will be deleted later.
+
+	* include/ruby/ruby.h (RCOMPLEX_SET_IMAG): ditto.
+
+	* internal.h (struct RFloat): moved here.
+
+	* internal.h (RCOMPLEX): ditto.
+
+	* complex.c (rb_complex_set_real): do not use this.
+
+	* complex.c (rb_complex_set_imag): ditto.
+
+Mon Jun 23 13:10:15 2014  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* include/ruby/ruby.h (struct RFloat): no longer.  [Feature #9863]
+
+	* include/ruby/ruby.h (RFLOAT): ditto.
+
+	* internal.h (struct RFloat): moved here.
+
+	* internal.h (RFLOAT): ditto.
+
+Mon Jun 23 12:01:42 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (check_port): allow strings for port= as
+	  described in rdoc.
+
+	* lib/uri/rfc3986_parser.rb (regexp): implementation detail of above.
+
+Mon Jun 23 11:35:01 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (setup_exception): set backtrace in system stack error
+	  other than the pre-allocated sysstack_error.  [Feature #6216]
+
+	* proc.c (Init_Proc): freeze the pre-allocated sysstack_error.
+
+	* vm_insnhelper.c (vm_stackoverflow): raise new instance for each
+	  times without calling any methods to keep the backtrace with no
+	  further stack overflow.
+
+	* object.c (rb_obj_copy_ivar): extract function to copy instance
+	  variables only for T_OBJECT from init_copy.
+
+Mon Jun 23 11:11:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (check_stack_overflow): drop the last tag too close to
+	  the fault page, to get rid of stack overflow deadlock.
+	  [Bug #9971]
+
+Sun Jun 22 09:11:15 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb: remove registry.
+	  'registry' is not used and RFC3986 doesn't use it.
+
+Sun Jun 22 09:10:09 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/rfc3986_parser.rb: raise exception when given a URI string
+	  has non ASCII in order to keep the regexp compiled for US-ASCII.
+
+Sun Jun 22 09:05:42 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/common.rb (URI::REGEXP): move to lib/uri/rfc2396_parser.rb.
+
+	* lib/uri/common.rb (URI::Parser): ditto.
+
+	* lib/uri/common.rb (URI.split): use RFC3986_Parser. [Feature #2542]
+
+	* lib/uri/common.rb (URI.parse): ditto.
+
+	* lib/uri/common.rb (URI.join): ditto.
+
+	* lib/uri/common.rb (URI.extract): deprecated.
+
+	* lib/uri/common.rb (URI.regexp): ditto.
+
+	* lib/uri/rfc2396_parser.rb: added.
+
+	* lib/uri/rfc3986_parser.rb: added.
+
+Sun Jun 22 09:04:50 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/mailto.rb (initialize): as previous commit, fix arg_check
+
+Sun Jun 22 09:01:47 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/ftp.rb (initialize): argument checking flag is arg_check,
+	  but arg[-1] is fragment.
+
+	* lib/uri/ftp.rb (initialize): explicitly specify arguments.
+
+Sat Jun 21 12:50:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/extconf.rb: supply 0 to fill RUBY_LIBFFI_MODVERSION
+	  with 3-digit.  libffi 3.1 returns just 2-digit.
+	  [ruby-core:62920] [Bug #9897]
+
+Sat Jun 21 07:06:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_find): [DOC] never accepted a symbol.
+	  [ruby-dev:48308] [Bug #9966]
+
+Fri Jun 20 17:15:43 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/lib/tracepointchecker.rb: add to check TracePoint healthiness.
+
+	* test/runner.rb: use it.
+
+Fri Jun 20 07:26:44 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_settracefunc.rb: rewrite tests with
+	  assert_consistent_call_return().
+
+	  assert_consistent_call_return() is also modified to check
+	  consistency.
+
+Fri Jun 20 07:07:28 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (rb_iseq_compile_node): put start label of block after
+	  trace (b_call).
+	  [Bug #9964]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+	  added assert_consistent_call_return() method check call/return
+	  consistency.
+
+Fri Jun 20 05:26:27 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (rb_catch_protect): fix same problem of [Bug #9961].
+
+	* vm_eval.c (rb_iterate): ditto.
+
+Thu Jun 19 21:41:30 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_rewind_cfp): add new function to rewind specified cfp
+	  with invoking RUBY_EVENT_C_RETURN.
+	  [Bug #9961]
+
+	* vm_core.h: ditto.
+
+	* eval.c (rb_protect): use it.
+
+	* eval.c (rb_rescue2): ditto.
+
+	* vm_eval.c (rb_iterate): ditto.
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+	* vm_core.h (rb_vm_rewind_cfp): add the prototype declaration.
+
+Thu Jun 19 19:47:21 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (invoke_block_from_c): move call/return event timing for
+	  bmethod. It can invoke inconsistent call event if this call raises
+	  argument error.
+	  [Bug #9959]
+
+	* vm_insnhelper.c (vm_call_bmethod_body): ditto.
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Thu Jun 19 18:14:47 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: add VM_FRAME_MAGIC_RESCUE to recognize normal block or
+	  rescue clause.
+
+	* vm.c (vm_exec): use VM_FRAME_MAGIC_RESCUE on at rescue/ensure.
+
+	* test/ruby/test_settracefunc.rb: should not invoke b_return at rescue
+	  clause.
+	  [Bug #9957]
+
+	* vm_dump.c (control_frame_dump): check VM_FRAME_MAGIC_RESCUE.
+
+	* vm_dump.c (vm_stack_dump_each): ditto.
+
+Thu Jun 19 13:39:11 2014  Arne Brasseur  <arne@arnebrasseur.net>
+
+	* proc.c (rb_method_curry): Implement Method#curry, which delegates
+	  to to_proc.curry.  [ruby-core:62212] [Feature #9783]
+
+Tue Jun 17 16:41:49 2014  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (gets, readline): read lines without LF properly.
+	  [ruby-core:63205] [Bug #9949]
+
+	* test/net/ftp/test_buffered_socket.rb: related test.
+
+Tue Jun 17 12:35:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (extract_raise_opts): pass unknown options to the
+	  exception, so that exception class can receive a hash argument.
+	  [ruby-core:63203] [Feature #8257]
+
+Tue Jun 17 12:24:57 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_memsize_of): memsize_of(T_ZOMBIE) returns 0, not a rb_bug.
+	  ObjectSpace.count_objects_size() uses memsize_of(T_ZOMBIE).
+
+	  This bug introduced at r46348.
+
+Mon Jun 16 19:00:11 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/runner.rb: failure message should be passed as an argument.
+
+Mon Jun 16 18:42:57 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/runner.rb: capture TracePoint stat before setup and compare
+	  it after teardown.
+
+Mon Jun 16 14:33:56 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (open): use UTF-8 version function to support
+	  non-ascii path properly.  [ruby-core:63185] [Bug #9946]
+
+Sat Jun 14 10:54:08 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rcombinate0): remove recursion, by looping with indexes
+	  stored in `p`.
+
+	* array.c (rpermute0): remove recursion, by looping with indexes
+	  stored in `p`.
+
+	* array.c (permute0): remove recursion, by looping with indexes
+	  stored in `p`.  [ruby-core:63103] [Bug #9932]
+
+Sat Jun 14 10:52:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_resize): update capa only when buffer get
+	  reallocated.
+	  http://d.hatena.ne.jp/nagachika/20140613/ruby_trunk_changes_46413_46420#r46413
+
+Sat Jun 14 08:28:59 2014  Zachary Scott  <e@zzak.io>
+
+	* man/rake.1: [DOC] Update links for Rake, patch by @hsbt [Bug #9904]
+	  [Fixes GH-628] https://github.com/ruby/ruby/pull/628
+
+Fri Jun 13 17:58:58 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: add new method TracePoint.stat to debug
+	  TracePoint mechanism.
+
+	  Ruby users should not use this method. So I don't note this method
+	  in the NEWS file.
+
+	* test/runner.rb: detect zombie active TracePoints with
+	  TracePoint.stat.
+
+Fri Jun 13 17:46:31 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c: clear and restore recursive checking thread local data
+	  to avoid unexpected throw from TracePoint.
+	  [Bug #9940]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+	* thread.c: added
+	  * rb_threadptr_reset_recursive_data(rb_thread_t *th);
+	  * rb_threadptr_restore_recursive_data(rb_thread_t *th, VALUE old);
+
+	* vm_core.h: ditto.
+
+Fri Jun 13 17:33:14 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_combination): iterate on a shared copy, and use
+	  array of indexes instead of array of chosen objects.
+	  [ruby-core:63149] [Bug #9939]
+
+	* array.c (yield_indexed_values): extract from permute0(),
+	  rpermute0(), and rcombinate0().
+
+Fri Jun 13 13:42:58 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_permutation): `p` is the array of size `r`, as
+	  commented at permute0().  since `n >= r` here, buffer overflow
+	  never happened, just reduce unnecessary allocation though.
+
+Thu Jun 12 20:32:28 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_resize): should consider the capacity instead
+	  of the old length, as pointed out by nagachika.
+
+Thu Jun 12 18:31:01 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/net/http/responses.rb: added Net::HTTPPermanentRedirect(308)
+	  Contributed by @yorkie [fix GH-638]
+
+Thu Jun 12 13:27:38 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_freeze): shrink the buffer before freezing, as
+	  pointed out by Eric Wong at [ruby-core:63119].
+
+Thu Jun 12 13:09:03 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (expand_path): shrink expanded path which no longer needs
+	  rooms to append.  [ruby-core:63114] [Bug #9934]
+
+Wed Jun 11 17:37:48 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_scalar_pthread_t): pthread_t is not required
+	  to be a scalar type.
+
+	* thread.c (fill_thread_id_string, thread_id_str): dump pthread_t
+	  in hexadecimal form if it is not a scalar type, assume it can be
+	  represented in a pointer form otherwise.  based on the patch by
+	  Rei Odaira at [ruby-core:62867].  [ruby-core:62857] [Bug #9884]
+
+	* thread_pthread.c (Init_native_thread, thread_start_func_1),
+	  (native_thread_create): set thread_id_str if needed.
+
+	* vm_core.h (rb_thread_t): add thread_id_string if needed.
+
+Wed Jun 11 01:53:22 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: invoke GC before memory allocation (xmalloc/xrealloc)
+	  when GC.stress = true.
+	  [Bug #9859]
+
+	* test/ruby/test_gc.rb: add a test.
+
+Tue Jun 10 13:20:14 2014  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/core.rb: Provide a mechanism to specify the
+	  max_multipart_length of multipart data.
+	  [Feature #8370] patch by Leif Eriksen <leif.eriksen.au@gmail.com>
+
+Tue Jun 10 10:57:07 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/csv.rb (CSV#<<): honor explicitly given encoding.  based on
+	  the patch by DAISUKE TANIWAKI <daisuketaniwaki AT gmail.com> at
+	  [ruby-core:62113].  [Bug #9766]
+
+Mon Jun  9 20:40:48 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change full GC timing to keep lower memory usage.
+
+	  Extend heap only at
+	  (1) after major GC
+	  or
+	  (2) after several (two times, at current) minor GC
+
+	  Details in https://bugs.ruby-lang.org/issues/9607#note-9
+	  [Bug #9607]
+
+Mon Jun  9 16:01:41 2014  Masahiro Ide  <imasahiro9@gmail.com>
+
+	* gc.c (gcdebug_sentinel): fix typo, "sentinel" not "sential".
+	  [fix GH-634]
+
+Mon Jun  9 00:04:25 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (posix_fadvise): disable use of posix_fadvise
+	  itself on 32-bit AIX.  [ruby-core:62968] [Bug #9914]
+
+Sun Jun  8 23:28:00 2014    <kanemoto@ruby-lang.org>
+
+	* io.c (rb_io_advise): AIX currently does not support a 32-bit call to
+	  posix_fadvise() if _LARGE_FILES is defined. Patch by Rei Odaira.
+	  [ruby-core:62968] [Bug #9914]
+
+Sun Jun  8 04:52:40 2014  Jun Hiroe  <Jun.Hiroe@gmail.com>
+
+	* string.c (rb_str_slice_bang): [DOC] update return value against
+	  a fixnum, which has changed because of M17N.  [fix GH-631]
+
+Sat Jun  7 22:13:42 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* numeric.c (do_coerce): Add a warning when an exception is raised
+	  or an invalid value is returned in #coerce called by
+	  numeric comparison operators and the exception
+	  thrown by the caller has no information on the failure.
+	  In the next release such exception should not be rescued or
+	  should be the cause of the caller exception. nil is accepted
+	  as the "no possible coercion" return value. See #7688.
+
+	* test/ruby/test_numeric.rb: Add corresponding test.
+
+Sat Jun  7 18:15:33 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* numeric.c (bit_coerce): remove constant parameter `err'
+	  (always TRUE) of bit_coerce().
+
+Sat Jun  7 16:01:57 2014  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* cont.c (rb_fiber_struct): keep context.uc_stack.ss_sp and context.uc_stack.ss_size
+	  for later use. Patch by Rei Odaira. [ruby-core:62945] [Bug #9905]
+
+Sat Jun  7 12:51:51 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (read_all): truncate the buffer before appending read data,
+	  instead of truncating before reading.
+	  [ruby-core:55951] [Bug #8625]
+
+Sat Jun  7 12:28:53 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/digest.c (rb_digest_instance_equal): no need to call
+	  `to_s` twice.  [Bug #9913]
+
+Sat Jun  7 11:35:01 2014  Tanaka Akira  <akr@fsij.org>
+
+	* object.c (rb_mod_initialize_clone): Override Kernel#initialize_clone
+	  to avoid an exception on Class.new.freeze.clone.to_s.
+	  Reported by Andrew Grimm.  [ruby-core:41858] [Bug #5828]
+
+Sat Jun  7 06:03:11 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/digest/digest.c (rb_digest_instance_equal):
+	  fix #== for non-string arguments. [ruby-core:62967] [Bug #9913]
+
+	* test/digest/test_digest.rb: add test for above.
+
+Fri Jun  6 22:19:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (private_recv_p): check by node type, instead of a
+	  magic number.
+
+	* node.h (NODE_PRIVATE_RECV), parse.y (attrset_gen): remove
+
+Fri Jun  6 17:07:08 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each), parse.y (new_attr_op_assign_gen):
+	  allow op assign to a private attribute.
+	  [ruby-core:62949] [Bug #9907]
+
+Fri Jun  6 13:39:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_setstrbuf, io_read): should not shorten the given buffer until
+	  read succeeds.  [ruby-core:55951] [Bug #8625]
+
+Fri Jun  6 07:41:41 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: dump empty symbols with a
+	  tag so that they can be parsed on input. [Bug #9873] [ruby-core:62825]
+	* test/psych/test_symbol.rb: test for change
+
+Thu Jun  5 16:08:39 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_page_sweep): refactoring.
+
+	* gc.c (gc_page_sweep): should not set, but add final_slots into
+	  sweep_page->final_slots.
+
+Thu Jun  5 14:36:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (jemalloc): check for the header regardless drop-in
+	  libjemalloc is found, for `malloc_conf` declaration.
+
+	* version.c (ruby_show_version): show `malloc_conf` if set.
+	  [Feature #9113]
+
+	* configure.in (with-jemalloc): also check for header, for ABIs
+	  which JEMALLOC_MANGLE is needed, i.e., Mach-O and PE-COFF
+	  platforms.  [ruby-core:62939] [Feature #9113]
+
+	* include/ruby/missing.h: include alternative malloc header to
+	  replace memory management functions.
+
+	* dln.c, io.c, parse.y, st.c: undef malloc family before
+	  re-definition to suppress warnings.
+
+Thu Jun  5 12:52:18 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* man/ruby.1: remove rubyforge entry.
+
+Thu Jun  5 12:45:32 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* misc/README: use github link instead of rubyforge.
+
+Thu Jun  5 10:03:29 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_free): check also FL_PROMOTED bit by RVALUE_OLD_P().
+
+Thu Jun  5 03:45:28 2014  Eric Wong  <e@80x24.org>
+
+	* configure.in: add --with-jemalloc option
+	  [ruby-core:62912]
+
+Wed Jun  4 22:28:14 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: introduce RZombie to manage zombie objects.
+	  Rewrite finalizing logics with this type.
+
+	* gc.c (gc_verify_internal_consistency): verify zombie (finalizing)
+	  objects count.
+
+Wed Jun  4 22:09:53 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (match_aref, rb_reg_regsub): consider encoding of captured
+	  names, encoding-incompatible should not match.
+	  [ruby-dev:48278] [Bug #9903]
+
+Wed Jun  4 21:23:52 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (match_aref): should not ignore name after NUL byte.
+	  [ruby-dev:48275] [Bug #9902]
+
+Wed Jun  4 04:08:37 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (core_hash_merge_kwd): should return the result hash, which
+	  may be converted from and differ from the given argument.
+	  [ruby-core:62921] [Bug #9898]
+
+Tue Jun  3 23:32:34 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ruby.c (load_file_internal2): Extracted from load_file_internal.
+	  (load_file_internal): Invoke load_file_internal2 using rb_protect.
+	  Close an opened FD if load_file_internal2 raises an exception.
+
+Tue Jun  3 19:11:45 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_free): should not rest_sweep() here.
+	  Some data structures are already freed.
+
+Tue Jun  3 18:43:51 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb: allocate more objects to invoke GC by newobj.
+	  GC allows extending pages depends on heap_increment.
+
+Tue Jun  3 18:01:27 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_call_finalizer_at_exit): add
+	  gc_verify_internal_consistency() when RGENGC_CHECK_MODE >= 2.
+
+Tue Jun  3 17:54:21 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change the counting method for young objects.
+	  clear counter at the beginning of every GC and
+	  count promoted (infant->young) objects.
+
+	  Some promotions (infant->young) are transition of promoting to old
+	  objects. We should not count such promotions.
+
+	  With this technique, we don't need to check young objects
+	  at obj_free().
+
+Tue Jun  3 16:38:19 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add verifying counters code in gc_verify_internal_consistency().
+	  gc_verify_internal_consistency() counts all
+	  - live objects
+	  - young objects (if age2 promotion)
+	  - old objects
+	  in all pages and compares with objspace managing counters.
+
+	* gc.c (gc_after_sweep): do gc_verify_internal_consistency()
+	  when RGENGC_CHECK_MODE >= 2.
+
+Tue Jun  3 13:14:04 2014  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (body_type_1part): Gmail IMAP reports a body
+	  type as "MIXED" followed immediately by params
+	  [ruby-core:62864] [Bug #9885]
+	  Patch by @rayners (David Raynes). [Fixes GH-622]
+	  https://github.com/ruby/ruby/pull/622
+
+Tue Jun  3 13:18:24 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (objspace_live_slot): live slot count should not include final
+	  slot (contains T_ZOMBIE) count.
+
+Tue Jun  3 13:03:21 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_free): fix spacing.
+
+Tue Jun  3 12:59:32 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (check_gen_consistency): fix error message.
+
+Tue Jun  3 12:40:23 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: count old/young objects more correctly.
+
+	* gc.c (RVALUE_DEMOTE_FROM_OLD): decrement old object count.
+
+	* gc.c (RVALUE_DEMOTE_FROM_YOUNG): decrement young object count.
+
+	* gc.c (rb_gc_resurrect): increment old object count.
+
+	* gc.c (gc_marks_body): should not add old object count.
+	  This code is completely my misunderstanding.
+
+	* gc.c (rb_gc_force_recycle): decrement young or old object count
+	  correctly.
+
+Tue Jun  3 12:26:47 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/memory_status.rb: add $LOAD_PATH to load test/unit
+	  correctly for fiddle/import unavailable environments.
+
+Tue Jun  3 09:45:13 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/openssl/test_ssl.rb (OpenSSL::TestSSL#test_verify_result):
+	  shouldn't use same server for respective tests, because the 1st
+	  test sometimes kills the server main loop silently.
+	  [Bug #9881] [ruby-dev:48266]
+
+Tue Jun  3 01:34:59 2014  Zachary Scott  <e@zzak.io>
+
+	* README.EXT: [DOC] Add rb_call_super when subclassing from @robin850
+	  [Fixes GH-623] https://github.com/ruby/ruby/pull/623
+
+Mon Jun  2 17:14:49 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (ruby_vm_destruct): remove useless call of
+	  rb_gc_force_recycle().
+
+	  At this line, a VM object is already freed
+	  (is changed to T_NONE) by rb_gc_call_finalizer_at_exit().
+
+Mon Jun  2 15:50:24 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_using_refinement): add write-barriers for
+	  cref->nd_refinements.
+
+Mon Jun  2 12:26:08 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* CONTRIBUTING.md: added contributing guide for github. [fix GH-625]
+
+Mon Jun  2 07:30:33 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/envutil.rb (default_warning): New method.
+
+	* test/ruby/test_autoload.rb: Use EnvUtil.default_warning.
+
+Mon Jun  2 07:05:59 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h: constify `rb_encoding` arguments.
+
+	* include/ruby/oniguruma.h: constify `OnigEncoding` arguments.
+
+Sun Jun  1 12:05:10 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/drb: Wrap tests definitions by DRbTests module.  This makes
+	  several tests (ACLEntryTest, TestBug4409, etc.) easier to understand
+	  that they are tests for DRb.
+
+Sun Jun  1 11:36:25 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rinda/ring.rb (RingFinger#make_socket): Close the socket on
+	  exception.
+
+Sun Jun  1 06:55:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (intern_str): dynamic attrset ID is registered by
+	  `rb_id_attrset()` already, so no further registration is needed.
+	  [ruby-core:62861]
+
+Sun Jun  1 04:52:47 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/English.rb: [DOC] $LOADED_FEATURES moved to load.c [Fixes GH-620]
+	  Patch submitted by @leafac in https://github.com/ruby/ruby/pull/620
+	* doc/globals.rdoc: Added $LOADED_FEATURES to list
+
+Sat May 31 22:30:14 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/leakchecker.rb: Leak checker extracted from
+	  test/lib/minitest/unit.rb.
+
+Sat May 31 21:15:43 2014  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* thread.c (rb_thread_atfork_internal): My compiler complains
+	  about this variable being used before initialized. I looked at
+	  the code and expanded the macro and turned out it was actually
+	  USED for pointer arithmetic, not dereferenced.  So this was
+	  never a serious bug.  But is annoying indeed to see warnings
+	  every time.  I added `=0` and all went healthy.
+
+	* configure.in: Also, I found that the problematic macro expansion
+	  only happens when we lack __typeof__ C extension, which shall
+	  not be the case of my compiler.  I added AC_C_TYPEOF to kick ass.
+
+Sat May 31 16:32:50 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/ipaddr.rb: extracted inline tests into test dir.
+	* test/test_ipaddr.rb: ditto.
+
+Sat May 31 16:29:21 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/digest/lib/digest/hmac.rb: extracted inline tests into test dir.
+	* test/digest/test_digest_hmac.rb: ditto.
+
+Sat May 31 16:02:03 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_syslog.rb: remove executable.
+
+Sat May 31 08:58:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/unicode.c (init_case_fold_table): no longer need to
+	  initialize tables at runtime.
+
+	* enc/unicode/case-folding.rb (lookup_hash): make perfect hash to
+	  lookup case unfolding table 3.
+
+	* enc/unicode/case-folding.rb (lookup_hash): make perfect hash to
+	  lookup case unfolding table 2.
+
+	* enc/unicode/case-folding.rb (lookup_hash): make perfect hash to
+	  lookup case unfolding table 1.
+
+	* enc/unicode/case-folding.rb (lookup_hash): make perfect hash to
+	  lookup case folding table.
+
+	* enc/unicode/case-folding.rb (print_table): merge non-locale and
+	  locale tables, and reduce initializing loops.
+
+	* enc/unicode/case-folding.rb (CaseFolding): modularize, and add
+	  --output-file option.
+
+	* enc/unicode/case-folding.rb: script to convert CaseFolding.txt,
+	  translated from CaseFolding.py.
+
+Sat May 31 08:31:41 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Check Tempfile leaks for each test method
+	  again.
+
+Sat May 31 03:50:50 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/delegate.rb: [DOC] Document raise in Delegator class
+	  Patch by @lucasmazza. [Fixes GH-621]
+	  https://github.com/ruby/ruby/pull/621
+
+Fri May 30 21:23:26 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/server.rb: Use a pipe to detect server shutdown.
+	  shutdown() or close() for listening socket is not a reliable.
+	  Actually, both doesn't work (doesn't wake up select()) on
+	  DragonFly BSD 3.6.2.
+
+	* test/webrick/utils.rb: :ShutdownSocketWithoutClose is not required
+	  now to immediate server shutdown detection.
+	  This fixes fd leaks.
+
+	* test/net/http/utils.rb: Ditto.
+
+Fri May 30 20:58:37 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb (check_fd_leak): Sort the inspected
+	  objects list for a FD.
+
+Fri May 30 18:06:55 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb (check_fd_leak): Try GC to delete leaked
+	  FDs.
+
+Fri May 30 12:05:59 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/lib/test/unit/parallel.rb (_run_suite): orig_stdout may be nil
+	  though I don't know the reason.
+
+Fri May 30 11:33:35 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_substr): need to reset code range for shared
+	  string too, not only copied string.
+	  [ruby-core:62842] [Bug #9882]
+
+Fri May 30 10:22:21 2014  Mark Lorenz  <mlorenz@covermymeds.com>
+
+	* lib/erb.rb (result): [DOC] no longer accepts a Proc, as
+	  Kernel.eval does not.  [fix GH-619]
+
+Fri May 30 07:25:46 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/dir: Dir#fileno implemented.
+
+	* test/lib/minitest/unit.rb (find_fds): Don't return the fd used to
+	  scan /proc/$$/fd.
+
+Fri May 30 04:48:00 2014  Eric Wong  <e@80x24.org>
+
+	* parse.y (rb_gc_mark_parser): remove, empty since r8758
+	* internal.h: ditto, not usable from extensions since 2.0.0
+	* gc.c (gc_mark_roots): remove checkpoint for parser
+
+Thu May 29 23:27:50 2014  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* signal.c (ruby_signal): should return either `old.sa_sigaction`
+	  or `old.sa_handler`, depending on whether `SA_SIGINFO` is set in
+	  `old.sa_flags`, because they may not be a union.
+	  [ruby-core:62836] [Bug #9878]
+
+Thu May 29 23:11:20 2014  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (pipe_open): Close pipes when rb_execarg_fixup() raises
+	  an exception.
+	  (rb_execarg_fixup_v): New function.
+
+Thu May 29 22:18:57 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb (capture_subprocess_io): Close fds.
+
+Thu May 29 19:47:08 2014  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_s_pipe): Close pipes if io_encoding_set() raises an
+	  exception.
+	  (io_encoding_set_v): New function.
+
+Thu May 29 19:42:49 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/csv.rb (CSV.open): Close the opened file when an exception
+	  occur.
+
+Thu May 29 19:31:10 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (SSLServer#accept): Close a socket
+	  if any exception occur.
+
+Thu May 29 05:05:29 2014  Eric Wong  <e@80x24.org>
+
+	* include/ruby/ruby.h: Hide Symbol internals.
+	  (struct RSymbol): moved to internal.h
+	  (RSYMBOL): ditto
+
+Thu May 29 00:28:56 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/unixsocket.c (rsock_init_unixsock): Open a socket
+	  after path length check.
+	  This fixes a fd leak by TestSocket_UNIXSocket#test_too_long_path.
+
+Wed May 28 23:04:35 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb (test_flush_in_finalizer1): Use
+	  ObjectSpace.each_object to close files.
+	  GC.start is not reliable.
+
+Wed May 28 19:00:31 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/net/imap.rb (Net::IMAP#initialize): Close the opened socket when
+	  any exception occur.
+	  This fixes a fd leak by IMAPTest#test_imaps_post_connection_check
+	  which start_tls_session() raises an exception.
+
+Wed May 28 18:06:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_close): Fix sync_close to work
+	  when SSL is not started.
+	  This fix the fd leak by test_https_proxy_authentication in
+	  test/net/http/test_https_proxy.rb.
+
+Wed May 28 10:29:28 2014  Eric Wong  <e@80x24.org>
+
+	* vm.c (rb_vm_living_threads_foreach): remove function
+	  [ruby-core:62745]
+	* thread.c (terminate_i): remove
+	* thread.c (terminate_all): implement (inlines old terminate_i)
+	* thread.c (rb_thread_terminate_all): use terminate_all
+	* thread.c (rb_thread_fd_close_i): remove
+	* thread.c (rb_thread_fd_close): iterate inline
+	* thread.c (thread_list_i): remove
+	* thread.c (rb_thread_list): iterate inline
+	* thread.c (rb_thread_atfork_internal): iterate inline
+	* thread.c (terminate_atfork_i): update types to remove casts
+	* thread.c (terminate_atfork_before_exec_i): ditto
+	* thread.c (struct thgroup_list_params): remove definition
+	* thread.c (thgroup_list_i): remove
+	* thread.c (thgroup_list): iterate inline
+	* thread.c (check_deadlock_i): remove
+	* thread.c (debug_deadlock_check): implement (inlines check_deadlock_i)
+	* thread.c (debug_i): remove
+	* thread.c (rb_check_deadlock): iterate inline
+	* vm.c (vm_mark_each_thread_func): remove
+	* vm.c (rb_vm_mark): iterate inline
+	* vm_core.h (rb_vm_living_threads_remove): remove
+	* vm_trace.c (clear_trace_func_i): remove
+	* vm_trace.c (rb_clear_trace_func): iterate inline
+
+Wed May 28 09:30:51 2014  Eric Wong  <e@80x24.org>
+
+	* signal.c (signal_exec): ignore immediate cmd for SIG_IGN
+	* signal.c (trap_handler): set cmd to true for SIG_IGN
+	* signal.c (trap): handle nil and true values for oldcmd
+	  [Bug #9835]
+
+Wed May 28 01:02:54 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tempfile.rb (Tempfile#inspect): Show "(closed)" if the tempfile
+	  is closed.
+
+Wed May 28 00:38:09 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Use Tempfile#close! instead of
+	  Tempfile#unlink to close file descriptors.
+
+	* test/openssl/test_config.rb: Ditto.
+
+	* test/ruby/test_io.rb: Ditto.
+
+Wed May 28 00:06:18 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/net/ftp.rb (transfercmd): Close TCP server socket even if an
+	  exception occur.
+
+Tue May 27 23:50:07 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/cgi/core.rb: Use Tempfile#close! instead of Tempfile#unlink
+	  to close file descriptors.
+
+Tue May 27 23:06:46 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_number_literal_suffix): refine error message for
+	  extra dot and digits.
+
+Tue May 27 22:44:20 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/rexml: Avoid fd leaks.
+
+Tue May 27 22:24:25 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb: Indent.
+
+Tue May 27 22:15:29 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_document.rb: Wrap by REXMLTests module.
+
+Tue May 27 22:11:10 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_encoding_2.rb: Remove a needless file.
+
+Tue May 27 22:10:30 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_contrib.rb: Indent.
+
+Tue May 27 21:28:16 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/ifaddr.c (ifaddr_inspect_flags): support IFF_SIMPLEX.
+
+Tue May 27 21:03:03 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/: Use REXMLTests as wrapping module for REXML tests.
+	  I avoid using the same module for library in test because
+	  it provides "include REXML" environment in test. Normally,
+	  users don't use REXML on "include REXML" environment. So I
+	  don't want to write tests on "include REXML" environment.
+
+Tue May 27 20:59:37 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_comment.rb: Remove needless REXML module wrapping.
+
+Tue May 27 20:56:49 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/openssl/test_pkcs7.rb: Fix inverted expected and actual values.
+
+Tue May 27 20:26:06 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_io.rb (test_flush_in_finalizer1): some opened fds are
+	  remain before GC, so unlink the tempfile is failed.
+
+Tue May 27 19:07:26 2014  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_autoclose_p): Don't raise on frozen IO.
+
+	* test/lib/minitest/unit.rb: IO#autoclose? may raise IOError.
+
+Tue May 27 19:01:49 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/openssl/test_pair.rb: Modify TestSSL#test_read_and_write
+	  to handle partial sysreads. [Bug #7398][ruby-core:49563]
+	* test/openssl/test_ssl.rb: ditto.
+
+Tue May 27 18:46:23 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/openssl/test_pkcs7.rb: Add tests for PKCS7#type= and add_data.
+	  [Feature #7399][ruby-core:49565]
+
+Tue May 27 17:45:09 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/openssl/test_hmac.rb (test_binary_update): Added Test for
+	  HMAC signing with UTF-8 String. [Bug #7512][ruby-core:50559]
+
+Tue May 27 17:10:14 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/runner.rb: fixed randomly test failure.
+	  [Bug #6573][ruby-core:45563]
+
+Tue May 27 16:58:12 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_timeout.rb (test_timeout): inverted test condition.
+	  [Bug #8523]
+
+Tue May 27 12:24:22 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Show leaked file descriptors.
+
+Tue May 27 11:12:56 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_fileno, rb_io_inspect): non-modification does not
+	  error on frozen IO.  [ruby-dev:48241] [Bug #9865]
+
+Tue May 27 00:00:21 2014  yui-knk  <spiketeika@gmail.com>
+
+	* insns.def (defineclass): fix typo in the instruction comment.
+	  [fix GH-618]
+
+Mon May 26 16:33:15 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file.rb: skip the test of atime on Windows, because
+	  Windows delays updating atime about 1 hour.
+	  see more details:
+	    http://msdn.microsoft.com/en-us/library/windows/desktop/ms724290%28v=vs.85%29.aspx
+
+Mon May 26 12:25:36 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optionparser.rb, lib/optparse.rb (OptParse): aliases.
+	  [ruby-core:62751] [Feature #9864]
+
+Mon May 26 07:59:34 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Show Finished threads line-by-line.
+
+Mon May 26 01:39:02 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/csv.rb: Reject nil as data source for CSV.new, patch by @Peeja.
+	  [Fixes GH-580] https://github.com/ruby/ruby/pull/580
+
+Mon May 26 01:07:51 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Show leaked threads and tempfiles
+	  line-by-line.
+
+Sun May 25 23:02:06 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb (MiniTest::Assertions#diff): Remove
+	  tempfiles.
+
+Sun May 25 22:42:27 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Check tempfile leak for each test class.
+
+Sun May 25 20:31:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func_{getcontext,setcontext}): do not
+	  disable ucontext.h entirely, but disable use of functions only.
+	  `ucontext_t` is necessary in the signal handler now.
+
+Sun May 25 20:00:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (check_stack_overflow): revert part of r46100, the
+	  previous condition was correct, and fix compilation error on
+	  other architecture linux.  [ruby-core:62746] [Bug #9862]
+
+Sun May 25 17:09:13 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Less ObjectSpace.each_object(Tempfile)
+	  invocation.
+
+Sun May 25 16:54:06 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Use Thread.list instead of
+	  ObjectSpace.each_object(Thread).
+
+Sun May 25 15:53:54 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/rinda/ring.rb (Rinda::RingServer#shutdown): Join the killed
+	  threads.
+
+Sun May 25 15:26:17 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/utils.rb: Override the inspect method of the thread
+	  used in WEBrick::Utils::TimeoutHandler.
+
+Sun May 25 14:22:30 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/openssl: Join threads.
+
+Sun May 25 12:46:47 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_bug_context): new function to report bug with
+	  context.
+
+	* vm_dump.c (rb_vm_bugreport): accepts `ucontext_t` argument to
+	  dump machine registers.  based on [GH-584].
+
+	* signal.c (sigbus, sigsegv): dump machine registers if available.
+
+Sun May 25 12:32:42 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Sort leaked threads and tempfiles.
+
+Sun May 25 12:15:30 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (check_stack_overflow): fix condition to use ucontext
+	  register, mcontext_t dereference, and its member names, on Mac
+	  OS X.
+
+Sun May 25 11:58:26 2014  Zachary Scott  <e@zzak.io>
+
+	* enumerator.c: [DOC] Fix example to show Enumerator#peek behavior
+	  Patch by Erik Hollembeak [Bug #9814]
+
+Sun May 25 11:56:33 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vsnprintf.c (BSD_vfprintf): fix string width when precision is
+	  given.  as the result of `memchr` is NULL or its offset from the
+	  start cannot exceed the size, the comparison was always false.
+	  [ruby-core:62737] [Bug #9861]
+
+Sun May 25 11:32:42 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/yaml.rb: Remove Psych::EngineManager [Bug #8344]
+	* test/psych/*: ditto.
+
+Sun May 25 10:34:15 2014  Zachary Scott  <e@zzak.io>
+
+	* doc/regexp.rdoc: [DOC] Clarify whitespace matching by @allolex
+	  [Fixes GH-606] https://github.com/ruby/ruby/pull/606
+
+Sun May 25 10:19:34 2014  Zachary Scott  <e@zzak.io>
+
+	* enum.c: [DOC] Use #find in example to clarify alias by @rachellogie
+	  Patch submitted via documenting-ruby/ruby#34
+
+Sun May 25 10:16:43 2014  Zachary Scott  <e@zzak.io>
+
+	* cont.c: [DOC] Fix rdoc in example for Fiber#transfer by @majjoha
+	  Patch submitted via documenting-ruby/ruby#33
+
+Sun May 25 10:01:11 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/irb.rb: [DOC] Fixed syntax error in example by @jasdeepsingh.
+	  Patch submitted via documenting-ruby/ruby#32
+
+Sun May 25 09:58:02 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/fileutils/test_fileutils.rb (test_chown_R): Add tests for
+	  chown_R. [Feature #9383][ruby-core:59641]
+
+Sun May 25 09:57:09 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/fileutils/test_fileutils.rb: Added recursively chown tests.
+	  [Feature #9303][ruby-core:59325]
+
+Sun May 25 09:41:56 2014  Zachary Scott  <e@zzak.io>
+
+	* class.c: [DOC] Fixed grammar and examples of instance_methods.
+	  By @alex-frost via documenting-ruby/ruby#31
+
+Sun May 25 09:40:44 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb: Show leaked threads and tempfiles.
+
+Sun May 25 08:54:38 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/openssl/test_partial_record_read.rb: Testing read_nonblock on
+	  a partial TLS record results in IO::WaitReadable by @mohamedhafez.
+	  [fix GH-547]
+
+Sun May 25 08:43:16 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/logger.rb: refactored to include Logger::Period.
+
+Sun May 25 06:50:19 2014  Zachary Scott  <e@zzak.io>
+
+	* vm_eval.c: [DOC] Improve instance_eval description when given a
+	  block or String arguments. By @nathanl via documenting-ruby/ruby#28
+
+Sun May 25 06:29:39 2014  Zachary Scott  <e@zzak.io>
+
+	* array.c: [DOC] Clarify default argument for Array.new.
+	  By @Elffers [Fixes GH-610]
+
+Sat May 24 22:37:20 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* array.c: [DOC] Add more documents to shuffle! and shuffle.
+	  Contributed by @JuanitoFatas [ci skip][fix GH-612]
+
+Sat May 24 22:28:55 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/lib/minitest/.document: removed unused configuration.
+
+Sat May 24 19:08:47 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/spec.rb: Unused file removed.
+
+	* test/lib/minitest/autorun.rb: Don't require minitest/spec.
+
+	* test/lib/minitest/benchmark.rb: Ditto.
+
+Sat May 24 18:45:30 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/benchmark/test_benchmark.rb: Use test/unit.
+
+Sat May 24 16:20:59 2014  Eric Wong  <e@80x24.org>
+
+	* process.c (proc_getgroups, proc_setgroups): use ALLOCV_N
+	  [Bug #9856]
+
+Sat May 24 15:49:39 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/unit.rb (parallelize_me!): Removed.
+	  This fixes the line-by-line structure of the test result in verbose
+	  mode.  [ruby-core:54905]
+
+	* test/lib/minitest/parallel_each.rb: Removed.
+
+	* test/minitest/test_minitest_mock.rb: Don't call parallelize_me!.
+
+	* test/minitest/test_minitest_spec.rb: Ditto.
+
+	* test/minitest/test_minitest_unit.rb: Ditto.
+	  Tests for parallel feature removed.
+
+Sat May 24 15:29:10 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest/hell.rb: Unused file removed.
+
+	* test/lib/minitest/pride.rb: Ditto.
+
+Sat May 24 15:05:32 2014  yui-knk  <spiketeika@gmail.com>
+
+	* enumerator.c (yielder_yield_push): Insert a break after the
+	  method return value.  [fix GH-617]
+
+Sat May 24 14:59:12 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/minitest: Remove comments not appropriate now.
+
+	* test/minitest: Ditto.
+
+Sat May 24 14:02:04 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* NEWS: added minitest changes.
+
+Sat May 24 13:42:46 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/lib/test/unit/test-unit.gemspec: removed needless gemspec file.
+
+Sat May 24 09:39:06 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* defs/default_gems: removed minitest entry.
+
+Sat May 24 06:17:33 2014  Ryan Davis  <ryand-ruby@zenspider.com>
+
+	*  lib/minitest: minitest 4.7.5 removed. Need to support proper
+	  gem packaging / installation before minitest 5 can be added.
+
+Sat May 24 05:54:06 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: More constants
+
+Sat May 24 00:25:34 2014  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* NEWS: add information of incompatibility about Prime.prime?
+	* lib/prime.rb: fix docs.
+
+Fri May 23 21:36:28 2014  Josh Goebel  <dreamer3@gmail.com>
+
+	* net/protocol.rb (using_each_crlf_line): fix SMTP dot-stuffing
+	  for messages not ending with a new-line.
+	  [ruby-core:61441] [Bug #9627] [fix GH-616]
+
+Fri May 23 03:48:08 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (rb_free_m_tbl): mark function as static
+
+	* method.h (rb_free_m_tbl): remove prototype
+
+Thu May 22 22:58:27 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: More TCP option constants.
+	  Describe Linux and glibc versions.
+
+Thu May 22 20:38:10 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (stat_birthtime): add birthtime support [Feature #9647]
+
+	* file.c (rb_stat_birthtime): add File::Stat.birthtime
+
+	* file.c (rb_file_s_birthtime): add File.birthtime
+
+	* file.c (rb_file_birthtime): add File#birthtime
+
+	* configure.in: check struct stat.st_birthtimespec.
+
+Thu May 22 19:38:14 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c: remove IO::Statfs because of reject. [Feature #9772]
+
+Thu May 22 14:02:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/jis/props.kwd: constify character property tables of JIS
+	  based encodings by perfect hash.
+
+	* enc/euc_jp.c, enc/shift_jis.c: use character property functions.
+
+Wed May 21 12:21:10 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c: Fix compilation error on Android.
+	  Bionic doesn't define TCP state constants.
+
+Wed May 21 11:42:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: workaround for Info.plist to get rid of `dsymutil`
+	  crash by wrong files in parent directories.
+	  [ruby-core:62594] [Bug #9840]
+
+Tue May 20 20:57:34 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_dir.rb (test_glob): added testcase of double
+	  slash path.
+
+Tue May 20 04:58:54 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/extconf.rb: Don't check fields of struct tcp_info if the
+	  structure is not available.
+
+Mon May 19 23:13:33 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_tcp_info): Permit longer data.  (glibc
+	  2.7 adds tcpi_rcv_rtt, tcpi_rcv_space and tcpi_total_retrans to
+	  struct tcp_info.)
+
+Mon May 19 20:49:07 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c (inspect_tcp_info): New function to inspect
+	  struct tcp_info.
+	  (sockopt_inspect): Use inspect_tcp_info.
+
+	* ext/socket/extconf.rb: Check tcp_info related things.
+
+	* ext/socket/rubysocket.h: Include netinet/tcp_fsm.h if available.
+
+Mon May 19 19:36:39 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/dbm/test_dbm.rb: Use Etc.uname.
+
+	* test/gdbm/test_gdbm.rb: Ditto.
+
+Mon May 19 16:54:22 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/etc/etc.c (etc_uname): add support for Windows using
+	  GetVersionExW(), GetSystemInfo(), and GetComputerNameExW() with
+	  `ComputerNameDnsHostname`.  [Feature #9842]
+
+Mon May 19 16:29:48 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_pat_search): advance by byte offset but not by char
+	  offset.  [ruby-core:62669] [Bug #9849]
+
+Mon May 19 14:06:18 2014  Shota Fukumori  <her@sorah.jp>
+
+	* bin/testrb: Removed. Forgot to remove in r45971.
+	  [Feature #9711] [ruby-core:62620]
+
+Sun May 18 16:42:08 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_m17n_comb.rb (test_str_crypt): Use Etc.confstr to
+	  detect the glibc version.
+	  libc.so is not an executable on Debian GNU/kFreeBSD 7.0 (wheezy).
+
+Sun May 18 12:15:54 2014  Jonathan Mukai-Heidt  <johnnymukai@gmail.com>
+
+	* io.c (argf_each_line, argf_inplace_mode_set): [DOC] Update ARGF
+	  documentation examples.  `ARGF.lines` has been deprecated in
+	  favor of `ARGF.each_line`.  [Fixes GH-615]
+
+Sun May 18 11:59:25 2014  Tanaka Akira  <akr@fsij.org>
+
+	* missing/nextafter.c: Include ruby/missing.h.
+
+Sun May 18 11:09:28 2014  Tanaka Akira  <akr@fsij.org>
+
+	* win32/Makefile.sub: Add nextafter.obj to MISSING.
+
+Sun May 18 10:46:04 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/etc/etc.c: Etc.sysconf, Etc.confstr and IO#pathconf implemented.
+
+	* ext/etc/extconf.rb: Check sysconf(), confstr() and fpathconf().
+
+	* ext/etc/mkconstants.rb: New file.
+
+	   [ruby-core:62600] [Feature #9842]
+
+Sun May 18 09:58:17 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/etc/etc.c: Etc.uname method implemented.
+
+	* ext/etc/extconf.rb: Check uname() function.
+
+	  [ruby-core:62139] [Feature #9770]
+
+Sun May 18 09:16:33 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Check nextafter() availability.
+
+	* include/ruby/missing.h (nextafter): New optional declaration.
+
+	* missing/nextafter.c: New file.
+
+	* numeric.c: Float#next_float and Float#prev_float implemented.
+
+	   [ruby-core:62562] [Feature #9834]
+
+Sun May 18 09:02:17 2014  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c: Enumerable#slice_after implemented.
+
+	* enumerator.c: Enumerator::Lazy#slice_after implemented.
+
+	  Requested by Tsuyoshi Sawada.  [ruby-core:58123] [Feature #9071]
+
+Sun May 18 08:22:25 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (io_setstrbuf): always check if the buffer is modifiable.
+	  [ruby-core:62643] [Bug #9847]
+
+Sun May 18 01:21:23 2014  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h: Hide Rational internal.
+	  (RRational): Moved to internal.h
+	  (RRATIONAL): Ditto.
+	  (RRATIONAL_SET_NUM): Moved to rational.c.
+	  (RRATIONAL_SET_DEN): Ditto.
+
+	* rational.c (rb_rational_num): New function.
+	  (rb_rational_den): Ditto.
+
+	* include/ruby/intern.h (rb_rational_num): Declared.
+	  (rb_rational_den): Ditto.
+
+	* ext/bigdecimal/bigdecimal.c: Follow the above change.
+
+	* ext/date/date_core.c: Ditto.
+
+	  [ruby-core:60665] [Feature #9513]
+
+Sat May 17 17:04:32 2014  Shota Fukumori  <her@sorah.jp>
+
+	* NEWS: Add news about removal of lib/test/**/*.rb.
+
+Sat May 17 16:57:33 2014  Shota Fukumori  <her@sorah.jp>
+
+	* lib/test: Removed because ruby's test cases now independent to
+	  lib/test by r45970. [Feature #9711] [ruby-core:62620]
+
+	  I'm still considering about the future of lib/minitest, lib/test.
+	  (bundling gems?)
+
+Sat May 17 15:06:40 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/runner.rb: remove dependency test-unit and minitest
+	  from stdlib when running with test-all.
+	  [Feature #9711][ruby-core:61890]
+	* test/testunit/*.rb: ditto.
+	* test/lib: ditto.
+
+Sat May 17 11:02:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): try match PLAIN as well as ALPHA, which are
+	  separated by previous commits.  [ruby-core:61552] [Bug #9648]
+
+	* dir.c (glob_make_pattern): set PLAIN for non-magical path to
+	  skip parts which not need to glob.
+	  [ruby-core:61552] [Bug #9648]
+
+	* dir.c (has_magic): return ALPHA at alphabetical name regardless
+	  FNM_CASEFOLD flag.
+
+	* dir.c (glob_helper): fix conditions for ALPHA.
+	  [ruby-core:61552] [Bug #9648]
+
+Sat May 17 01:49:27 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): reduce matching at non-magical path on
+	  Windows.
+
+Sat May 17 01:49:23 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_pattern_type): separate names with alphabet but no
+	  magical from plain.
+
+	* dir.c (glob_helper): match plain names as-is to treat super-root
+	  same as the root.  [ruby-core:61552] [Bug #9648]
+
+Fri May 16 17:38:22 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_marks, gc_marks_body): increase the counter of young objects
+	  at the major GC because AGE2Promotion changes all old objects into
+	  young objects at major GC.
+
+Fri May 16 17:26:24 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_before_sweep): heap_pages_swept_slots should contains
+	  heap_pages_increment.
+
+	  For example, GC by exceeding malloc_limit can remain
+	  heap_pages_increment.
+
+Thu May 15 21:18:43 2014  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* configure.in: enable SSE2 on mingw. target='i386-pc-mingw32'.
+	  [ruby-core:62095] [Bug #8358]
+
+Thu May 15 21:04:06 2014  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* lib/test/unit/parallel.rb: fix test-all parallel failure if a test
+	  is skipped after raise.
+	  DL::TestFunc#test_sinf is skipped after raise on mingw ruby.
+	  But it causes Marshal.load failure due to undefined class/module
+	  DL::DLError when doing test-all parallel and test-all doesn't
+	  complete. We create new MiniTest::Skip object to avoid Marshal.load
+	  failure.
+	  [ruby-core:62133] [Bug #9767]
+
+	* test/testunit/test_parallel.rb (TestParallel): add a test.
+
+	* test/testunit/tests_for_parallel/ptest_forth.rb: ditto.
+
+Thu May 15 18:57:23 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (garbage_collect_body): move gc_heap_prepare_minimum_pages()
+	  from gc_sweep().
+
+Thu May 15 18:51:25 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_extend_pages): calculate next growing heap size.
+
+	* gc.c (heap_set_increment): accept addition pages instead of
+	  minimum pages.
+
+	* gc.c (gc_after_sweep): use heap_etend_pages().
+
+	* gc.c (gc_heap_prepare_minimum_pages): add only 1 page.
+
+	* gc.c (heap_ready_to_gc): add only 1 page.
+
+Thu May 15 18:42:49 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: introduce macros to remove magic number.
+
+	  GC_HEAP_FREE_SLOTS_MIN_RATIO = 0.3: guarantee minimum empty slots
+	                                      ratio after sweep.
+	  GC_HEAP_FREE_SLOTS_MAX_RATIO = 0.8: allow to free pages 0.2 (= 1-0.8)
+	                                      of current existing slots.
+
+Thu May 15 17:32:51 2014  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* thread_win32.c (rb_w32_stack_overflow_handler): use Structured
+	  Exception Handling by AddVectoredExceptionHandler() for machine
+	  stack overflow on mingw.
+	  This would be equivalent to the handling using __try and __except
+	  on mswin introduced by r43748.
+
+Wed May 14 19:31:03 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/openssl/depend: remove dependency from internal headers.
+	  [Feature #9612]
+
+	* ext/openssl/ossl.c (ossl_fips_mode_set): ditto.
+
+	* ext/coverage/depend: ditto.
+
+	* include/ruby/thread_native.h: added.
+
+	  This header file only provides wrapper functions to control
+	  native threads. These wrapper functions are used by MRI
+	  implementation.
+
+	* vm_core.h: use include/ruby/thread_native.h.
+
+	* thread.c: ditto.
+
+	* thread_pthread.h: ditto.
+
+	* thread_win32.h: ditto.
+
+	* thread_native.h: removed.
+
+Wed May 14 18:03:28 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: enable RGENGC_AGE2_PROMOTION.
+
+Wed May 14 18:02:30 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rgengc_rememberset_mark): promote remembered object earlier.
+
+Mon May 12 23:57:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_atan2_inf_c99): check whether runtime atan2
+	  handles Inf as C99. [ruby-core:62536] [Bug #9831]
+
+Mon May 12 20:33:01 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Invoke AC_REPLACE_FUNCS for each function.
+
+Mon May 12 19:52:11 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: count young object correctly and show it in GC.stat
+	  on RGENGC_AGE2_PROMOTION.
+
+	* gc.c (RVALUE_PROMOTE_YOUNG): decrement young object count on
+	  YOUNG->OLD.
+
+	* gc.c (obj_free): decrement young object count when young object
+	  freed.
+
+	* gc.c (gc_marks): should not clear young object count.
+
+	* gc.c (gc_stat_internal): GC.stat :young_object information.
+
+Mon May 12 01:30:59 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ifaddr.c (IS_IFADDRS): Unused macro removed.
+
+	* ext/strscan/strscan.c (BUFSIZE): Ditto.
+
+	* ext/zlib/zlib.c (OBJ_IS_FREED): Ditto.
+
+Sun May 11 22:27:18 2014  Tanaka Akira  <akr@fsij.org>
+
+	* compile.c (BUFSIZE): Unused macro removed.
+
+	* vm.c (BUFSIZE): Ditto.
+
+	* pack.c (INT64toNUM): Ditto.
+	  (UINT64toNUM): Ditto.
+	  (BYTEWIDTH): Ditto.
+
+	* time.c (lshift): Ditto.
+	  (UINT64toNUM): Ditto.
+	  (id_lshift): Unused variable removed.
+
+Sun May 11 21:23:27 2014  Tanaka Akira  <akr@fsij.org>
+
+	* pack.c (swaps): Unused macro removed.
+	  (swapi): Ditto.
+	  (swapl): Ditto.
+	  (swapll): Ditto.
+
+Sun May 11 08:02:49 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_vm_t): list_head and counter for living_threads
+	  (rb_thread_t): vmlt_node for living_threads linkage
+	  (rb_vm_living_threads_init): new function wrapper
+	  (rb_vm_living_threads_insert): ditto
+	  (rb_vm_living_threads_remove): ditto
+	* vm.c (rb_vm_living_threads_foreach): new function wrapper
+	* thread.c (terminate_i, thread_start_func_2, thread_create_core,
+	  thread_fd_close_i, thread_fd_close): update to use new APIs
+	* vm.c (vm_mark_each_thread_func, rb_vm_mark, ruby_vm_destruct,
+	  vm_memsize, vm_init2, Init_VM): ditto
+	* vm_trace.c (clear_trace_func_i, rb_clear_trace_func): ditto
+	* benchmark/bm_vm_thread_close.rb: added to show improvement
+	* ccan/build_assert/build_assert.h: added as a dependency of list.h
+	* ccan/check_type/check_type.h: ditto
+	* ccan/container_of/container_of.h: ditto
+	* ccan/licenses/BSD-MIT: ditto
+	* ccan/licenses/CC0: ditto
+	* ccan/str/str.h: ditto (stripped of unused macros)
+	* ccan/list/list.h: ditto
+	* common.mk: add CCAN_LIST_INCLUDES
+	  [ruby-core:61871][Feature #9632 (part 1)]
+
+Sun May 11 01:10:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (rb_f_kill): directly enqueue an ignored signal to self,
+	  except for SIGSEGV and SIGBUS.  [ruby-dev:48203] [Bug #9820]
+
+Sat May 10 22:37:56 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (push_glob): match in UTF-8 on Mac OS X.
+	  [ruby-dev:48213] [Bug #9825]
+
+Sat May 10 13:32:18 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): stop if forked in a sub-thread,
+	  the thread has become the main thread.
+	  [ruby-core:62070] [Bug #9751]
+
+Sat May 10 09:32:19 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* man/ruby.1: remove deadlink. [ruby-core:62145][Bug #9773]
+
+Sat May 10 08:47:36 2014  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (trap): Return "SYSTEM_DEFAULT" if SIG_DFL is set.
+
+Fri May  9 14:27:05 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_SETJMP_TYPE): check for setjmp type after
+	  CCDLFLAGS is appended to CFLAGS, since __builtin_setjmp can be
+	  affected.  [ruby-core:62469] [Bug #9818]
+
+Fri May  9 03:59:06 2014  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/delegate.rb: Fix example of using delegator.
+	  patched from Andrey Koleshko. [Fixes GH-505]
+
+Fri May  9 03:42:43 2014  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/shell.rb: add documentation in lib/shell.rb
+	  patched from reprah. [Fixes GH-516]
+
+Fri May  9 03:28:04 2014  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/fileutils.rb: show fileutils require at top.
+	  patched from Richard Schneeman. [Fixes GH-604]
+
+Fri May  9 03:07:09 2014  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* lib/prime.rb (Prime#prime?): negative numbers can't be primes
+	  by definition. reported by Ivan Kataitsev. [Bug #7395]
+	* test/test_prime.rb: add test.
+
+Thu May  8 14:34:29 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mod_init_copy): always clear instance variable,
+	  constant and method tables first, regardless the source tables.
+	  [ruby-dev:48182] [Bug #9813]
+
+Thu May  8 10:53:14 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: OpenBSD needs to include sys/param.h before include
+	  sys/mount.h. [ruby-dev:48167]
+
+Thu May  8 10:17:04 2014  Karsten Sperling  <karsten@sperling.co.nz>
+
+	* lib/webrick/httpserver.rb (WEBrick::HTTPServer#run): stop
+	  handling requests on shutdown, even if the socket is readable
+	  and IO.select() returns true.  [Fixes GH-607]
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#start): IO.select()
+	  raises ENOTSOCK on shutdown on Windows.
+
+Wed May  7 21:45:00 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLServer#accept):
+	  Consider Socket#accept as well as TCPServer#accept.
+	  Reported by Sam Stelfox.  [ruby-core:62064] [Bug #9750]
+
+Wed May  7 17:24:07 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_step_scan_args): check keyword arguments and fail
+	  if they conflict with positional arguments.
+	  [ruby-dev:48177] [Bug #9811]
+
+Wed May  7 12:06:14 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: remove debug output and output results into
+	  specified file.
+
+Wed May  7 11:55:40 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: add '--rawdata-output=[FILE] option to output
+	  raw results into FILE.
+
+Wed May  7 11:25:41 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_f_local_variables): exclude variables hidden by
+	  shadowing.  [ruby-core:60501] [Bug #9486]
+
+	* vm.c (collect_local_variables_in_iseq): ditto.
+
+Tue May  6 23:29:05 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (new_bv_gen): no duplicated names, if already added in
+	  shadowing_lvar().
+
+	* parse.y (local_tbl_gen): remove local variables duplicated with
+	  arguments.
+	  [ruby-core:60501] [Bug #9486]
+
+Tue May  6 18:48:50 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.make_time): Adjust the time zone of "now".
+
+Tue May  6 18:33:12 2014  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* io.c (io_{read,write}_nonblock): use rb_get_kwargs instead of
+	  rb_hash_aref.
+
+Tue May  6 18:03:05 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.make_time): Argument validation code moved from
+	  Time.parse and Time.strptime.
+
+Tue May  6 17:27:06 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.parse): [DOC] Fix an example in the documentation
+	  to use EST.
+	  Reported by Marcus Stollsteimer.
+	  [ruby-core:60778] [Bug #9521] and [ruby-core:61718] [Bug #9682]
+
+Tue May  6 04:31:48 2014  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* file.c (rb_f_test): removed meaningless "case 'a'".
+
+Tue May  6 01:28:14 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb (OpenURI.open_uri): Call StringIO#close only if
+	  the StringIO object is not closed yet.
+	  Reported by Jordi Massaguer Pla.  [ruby-core:42538] [Bug #6010]
+
+Tue May  6 01:08:01 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: define File::NULL if not defined and /dev/null
+	  is available to run benchmark driver on ruby 1.9.2.
+
+Mon May  5 23:53:24 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/mkconstants.rb: Add IP_TRANSPARENT.
+	  IP_TRANSPARENT is provieded since glibc-2.12.
+	  Reported by Eliezer Croitoru.  [ruby-core:50372] [Bug #7476]
+
+Mon May  5 22:29:47 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (check_exec_redirect): Open the file in write mode for
+	  redirect from [:out, :err].
+	  Proposed and implemented by Yusuke Endoh.
+	  [ruby-dev:41430] [Feature #3348]
+
+Mon May  5 21:52:35 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/lib/pathname.rb (cleanpath_aggressive): make all
+	  separators File::SEPARATOR from File::ALT_SEPARATOR.
+	  Reported by Daniel Rikowski.
+	  Fixed by Nobuyoshi Nakada.  [Bug #9618]
+
+	* ext/pathname/lib/pathname.rb (cleanpath_conservative): ditto.
+
+Mon May  5 21:48:04 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/lib/pathname.rb (Pathname#/): Aliased to Pathname#+.
+	  Suggested by Alexey Muranov.  [ruby-core:61432] [Feature #9625]
+
+Mon May  5 17:26:09 2014  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* math.c (rb_math_sqrt): omitted exporting an unused function,
+	  anyway.
+	* internal.h: follows the above change.
+
+Mon May  5 11:44:03 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.strptime): Raise ArgumentError if Date._strptime
+	  doesn't extract date information.
+	  Reported by tadayoshi funaba.  [ruby-core:62349]
+
+Mon May  5 01:12:27 2014  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (rt_rewrite_frags): a new feature (not a
+	  bug fix) of strptime.  applies offset even if the given date is
+	  not local time (%s and %Q).  This is an exceptional feature and
+	  I do NOT recommend to use this at all.  Thank you git community.
+
+Sun May  4 20:51:32 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.force_zone!): Use usual local time if it has
+	  expected offset from UTC.
+
+Sun May  4 17:58:12 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.force_zone!): New private method.
+	  (Time.make_time): Use Time.force_zone!.
+	  (Time.strptime): Ditto.
+	  (Time.rfc2822): Ditto.
+	  (Time.xmlschema): Ditto.
+
+	* lib/rss/rss.rb (Time.w3cdtf): Use Time.force_zone!.
+
+Sun May  4 10:22:59 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (math_atan2): return values like as expected by C99 if
+	  both two arguments are infinity.  based on the patch by cremno
+	  phobia <cremno AT mail.ru> in [ruby-core:62310].  [Feature #9799]
+
+Sun May  4 03:46:42 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.httpdate): Always return a UTC Time object.
+
+Sun May  4 03:26:39 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.make_time): Refactored.
+
+Sun May  4 02:53:17 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.rfc2822): Fix year completion.
+	  Produce fixed-offset time object if appropriate.
+	  (Time.xmlschema): Produce fixed-offset time object if appropriate.
+
+Sat May  3 23:52:20 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (make_time): Produce fixed-offset time object if
+	  appropriate.
+	  (Time.strptime): Use d[:zone] instead of d[:offset].
+
+	* lib/rss/rss.rb (Time.w3cdtf): Produce fixed-offset time object if
+	  appropriate.
+
+Sat May  3 20:21:38 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (Time.strptime): Use d[:offset] if d[:seconds] is not
+	  given.
+	  Reported by tadayoshi funaba.  [ruby-core:62322]
+
+Sat May  3 04:04:16 2014  Eric Wong  <e@80x24.org>
+
+	* complex.c (parse_comp): replace ALLOCA_N with ALLOCV_N/ALLOCV_END
+	  [Bug #9608]
+	* rational.c (read_digits): ditto
+
+Sat May  3 00:06:30 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* file.c (HAVE_STRUCT_STATVFS_F_BASETYPE): File::Statfs#fstypename
+	  is supported on AIX, HP-UX, and Solaris, by using the value of
+	  struct statvfs.f_basetype.
+
+	* configure.in (HAVE_STRUCT_STATVFS_F_BASETYPE): check struct
+	  statvfs.f_basetype which is available on AIX, HP-UX, and Solaris.
+
+Fri May  2 21:04:02 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (next_id): no reason to set ID_STATIC_SYM here, as ID
+	  returned by rb_intern3 can be a dynamic symbol and the static
+	  symbol flag is set otherwise.  [Bug #9787]
+
+Fri May  2 11:32:51 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_io.rb (test_seek, test_seek_symwhence): defer
+	  File::Statfs#type call which may not be implemented, to mitigate
+	  errors on platforms where SEEK_DATA is available but f_type in
+	  struct statfs is not.  [ruby-dev:48154] [Bug #9789]
+
+Fri May  2 10:37:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_id_attrset): turn dynamically interned Symbol into
+	  an ID, since rb_str_dynamic_intern returns a Symbol but not an
+	  ID.  [ruby-core:62226] [Bug #9787]
+
+Thu May  1 22:19:34 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* file.c: Change AND condition to nested condition.
+
+Thu May  1 00:36:26 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* file.c (FSTATFS): check availability of struct statfs and
+	  struct statvfs in addition to fstatfs(2) and fstatvfs(2).
+	  This fixes error in Solaris. [Bug #9788] [ruby-dev:48145]
+
+Wed Apr 30 19:46:23 2014  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c (gc_after_sweep): suppress unnecessary expanding heap.
+	  Tomb heap pages are freed pages here, so expanding heap is
+	  not required.
+
+Wed Apr 30 17:58:40 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (invoke_block_from_c): add VM_FRAME_FLAG_BMETHOD to record
+	  it is bmethod frame.
+
+	* vm.c (vm_exec): invoke RUBY_EVENT_RETURN event if rollbacked frame
+	  is VM_FRAME_FLAG_BMETHOD.
+	  [Bug #9759]
+
+	* test/ruby/test_settracefunc.rb: add a test for TracePoint/set_trace_func.
+
+	* vm_core.h: rename rb_thread_t::passed_me to
+	  rb_thread_t::passed_bmethod_me to clarify the usage.
+
+	* vm_insnhelper.c (vm_call_bmethod_body): use renamed member.
+
+Wed Apr 30 17:06:49 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_id_attrset): pin down dynamic symbol only. it is
+	  possible that attrset ID can be registered as a static symbol
+	  after the corresponding attrget ID has been registered as a
+	  dynamic, and then the latter may be collected.
+	  [ruby-core:62226] [Bug #9787]
+
+Tue Apr 29 14:17:57 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb: Rescue LoadError on etc.so for miniruby.
+	  Revert r45707, r45711, r45717.
+
+Tue Apr 29 12:50:02 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/fileutils.rb: Don't need to define fu_get_gid and fu_get_gid in
+	  rescue LoadError on 'etc'.
+
+Tue Apr 29 10:21:38 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* parse.y (symbols_i): like r45492, call rb_gc_resurrect().
+
+Tue Apr 29 04:29:05 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* file.c (HAVE_STRUCT_STATFS_T_F_FSTYPENAME): Add new macro for
+	  statfs_t.
+	* file.c (HAVE_STRUCT_STATFS_T_F_TYPE): ditto.
+
+	* file.c (rb_io_statfs): check FSTATFS macro only instead of
+	  HAVE_FSTATFS and HAVE_FSTATVFS.
+
+	* file.c (statfs_type): use new macro.
+	* file.c (statfs_fstypename): ditto.
+	* file.c (statfs_inspect): ditto.
+
+Tue Apr 29 00:20:26 2014  Rajarshi Das  <rajarshid@cybage.com>
+
+	* bootstraptest/test_literal.rb: fix typo of "dynamic".  [ci skip]
+
+	* regexp.rdoc: fix typo of "organized".  [ci skip]
+
+	* lib/session.rb: fix typo of "recognized".  [ci skip]
+
+Mon Apr 28 21:40:27 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (HAVE_STRUCT_STATFS_F_TYPE): check struct statfs.f_type
+	  to support OpenBSD.
+
+	* file.c (statfs_type): use above macro to switch.
+
+	* file.c (statfs_inspect): ditto.
+
+Mon Apr 28 18:06:08 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: check struct statvfs and struct statvfs.f_fstypename.
+
+	* configure.in: on NetBSD fstatfs is obsoleted.
+
+	* file.c: support NetBSD for File::Statfs.
+
+Mon Apr 28 17:42:42 2014  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: This argument must be a pointer.
+
+Mon Apr 28 17:40:15 2014  Narihiro Nakamura  <authornari@gmail.com>
+
+	* gc.c: Fix typos. These are undefined variables.
+
+Sun Apr 27 19:39:42 2014  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_strptime.c (date__strptime_internal): do not
+	  overwrite century.
+
+Sat Apr 26 11:50:08 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_enum.rb (test_flat_map): Added test for flat_map.
+	  Contribute from @igaiga. [fix GH-598]
+
+Sat Apr 26 10:55:33 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_array_): make copy a first hash not to modify
+	  the argument itself.  keyword splat should be non-destructive.
+	  [ruby-core:62161] [Bug #9776]
+
+Sat Apr 26 08:05:36 2014  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_process.rb (test_rlimit_nofile): Don't limit
+	  RLIMIT_NOFILE too small.
+	  This fix sporadic "[ASYNC BUG] thread_timer: select" on GNU/Linux.
+
+Fri Apr 25 22:54:34 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* lib/fileutils.rb (rmdir): rescue Errno::EEXIST in addition to
+	  ENOTEMPTY (and ENOENT), because SUSv3 describes that "If the
+	  directory is not an empty directory, rmdir() shall fail and set
+	  errno to [EEXIST] or [ENOTEMPTY]" and Solaris uses EEXIST.
+	  [Bug #9571] [ruby-dev:48017]
+
+Fri Apr 25 19:16:30 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/utils.rb: Don't rescue LoadError for 'etc' extension.
+
+Fri Apr 25 14:55:59 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_func___builtin_unreachable): try with an
+	  external variable not only by a warning, which might not be
+	  shown due to the optimization.  [ruby-core:61647] [Bug #9665]
+
+Fri Apr 25 13:11:49 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: NetBSD's ksh, used by configure, needs escapes.
+
+Fri Apr 25 12:51:08 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: correct pthread_setname_np's prototype on NetBSD.
+	  [Bug #9586]
+
+Thu Apr 24 23:17:25 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/fileutils.rb (fu_get_uid, fu_get_gid): Etc.getpwnam/getgrnam may
+	  returns nil.
+
+	* lib/webrick/utils.rb (su): ditto.
+
+Thu Apr 24 22:55:22 2014  Tanaka Akira  <akr@fsij.org>
+
+	* bootstraptest/test_io.rb: Add etc.so to $" before require 'tmpdir'.
+
+Thu Apr 24 21:09:55 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* man/ruby.1: fix broken link.
+
+Thu Apr 24 20:53:02 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb: Don't need to rescue LoadError for etc.so.
+
+Thu Apr 24 17:39:53 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (statfs_inspect): suppress warnings.
+	  assume those values won't be larger than LONG_LONG_MAX.
+
+Thu Apr 24 11:53:28 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_initialize): SYMID on a value
+	  other than Symbol is an undefined behavior.  fix up r31699.
+	  [ruby-core:62142] [Bug #9771]
+
+Thu Apr 24 11:21:37 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (rb_sym2id, rb_sym2id_without_pindown): return 0 for
+	  non-symbol values, for the time being.
+
+Thu Apr 24 05:50:13 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* parse.y (dsym_node_gen): like r45492, call rb_gc_resurrect().
+
+Wed Apr 23 20:36:22 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/etc/extconf.rb: Build ext/etc unconditionally.
+
+Wed Apr 23 14:10:50 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (statfs_fsid): remove statfs.f_fsid because it doesn't return
+	  meaningful value portably. http://togetter.com/li/658517
+
+Wed Apr 23 11:03:41 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_write): use rb_str_append to
+	  reuse coderange bits other than ASCII-8BIT, and keep
+	  taintedness.  [ruby-dev:48118] [Bug #9769]
+
+Wed Apr 23 00:43:00 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (ustatfs): implementation of
+	  statfs(2) clone. [EXPERIMENTAL]
+
+	* file.c (rb_io_statfs): use above function.
+
+	* configure.in, win32/Makefile.sub (struct statfs): available.
+
+Tue Apr 22 23:56:24 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_io_stafs): use statfs(2) if fstatfs(2) is unavailable.
+
+	* configure.in (fstatfs): check it.
+
+Tue Apr 22 22:15:51 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_io_statfs): need to define even if the system doesn't have
+	  fstatfs(2).
+
+	* test/ruby/test_file.rb (TestFile#test_statfs): skip if IO#stafs is not
+	  implemented.
+
+Tue Apr 22 19:32:48 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c: newly added a class File::Statfs. (experimental)
+
+Tue Apr 22 08:22:33 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (objspace_malloc_increase): don't cause GC by malloc_increase
+	  when memop type is MEMOP_TYPE_REALLOC.
+
+	  GC at realloc is not well maintained.
+	  We need a time to make it safe.
+	  [ruby-dev:48117]
+
+Tue Apr 22 06:54:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (objspace_malloc_increase): run full mark if 0x04 bit is
+	  set in ruby_gc_stress.  [ruby-core:62103] [Feature #9761]
+
+	* gc.c (objspace_malloc_increase): run GC after realloc not only
+	  malloc and calloc by GC.stress.  [ruby-core:62103] [Feature #9761]
+
+Mon Apr 21 19:12:20 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_coderange_scan_restartable): coderange is always
+	  ENC_CODERANGE_VALID if the string is ASCII-8BIT and already has a non
+	  ASCII character.
+
+Mon Apr 21 19:02:44 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (coderange_scan): remove useless condition `p < e` after
+	  search_nonascii.
+
+	* string.c (rb_str_coderange_scan_restartable): ditto.
+
+Mon Apr 21 18:55:21 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/-test-/string/coderange.c: add Bug::String.new#coderange_scan
+	  to explicitly scan coderange.
+
+Mon Apr 21 18:19:35 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (coderange_scan): remove unused logic.
+
+	* string.c (rb_str_coderange_scan_restartable): ditto.
+
+Mon Apr 21 14:11:48 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_putc): fix for non-ascii
+	  encoding, like as IO#putc.  [ruby-dev:48114] [Bug #9765]
+
+Sun Apr 20 12:57:15 2014  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* st.c (st_foreach_check): change start point of search at check
+	  from top to current. [ruby-dev:48047] [Bug #9646]
+
+Sun Apr 20 08:41:33 2014  Andrew DeMaria  <ademariad@gmail.com>
+
+	* lib/mkmf.rb (link_command, libpathflag, create_makefile): prefer
+	  user specified `$LIBPATH` than `$DEFLIBPATH`.  [ruby-core:62100]
+	  [ruby-trunk - Bug #9760]
+
+Sun Apr 20 06:01:18 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (rb_gc_writebarrier): drop special case for big hash/array
+	  [Bug #9518]
+
+Sat Apr 19 15:38:29 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): flush cmdarg flags inside left-paren in a
+	  command argument, to allow parenthesed do-block as an argument
+	  without arguments parentheses.  [ruby-core:61950] [Bug #9726]
+
+Sat Apr 19 10:07:24 2014  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (struct RBignum): Use size_t for len.
+
+	* include/ruby/intern.h (rb_big_new): Use size_t instead of long to
+	  specify the size of bignum.
+	  (rb_big_resize): Ditto.
+
+	* bignum.c: Follow above changes.
+
+	* rational.c: Follow above changes.
+
+	* marshal.c: Follow above changes.
+
+Sat Apr 19 00:32:07 2014  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_num2long): Returns a long.
+	  (rb_num2ulong): Returns a unsigned long.
+
+	* bignum.c (rb_big2long): Returns a long.
+	  (rb_big2ulong): Returns a unsigned long.
+
+	* include/ruby/intern.h: Follow above changes.
+
+	* include/ruby/ruby.h: Follow above changes.
+	  (rb_num2long_inline): No need to cast.
+	  (rb_num2ulong_inline): Ditto.
+
+Sat Apr 19 00:17:20 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (SHARABLE_SUBSTRING_P): predicate if substring can be
+	  shared with the original string.  true if just at the end of the
+	  original string, for the time being.  all substring will be able to
+	  be shared in the future.
+
+Fri Apr 18 21:48:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_new_frozen): consider the shared string at
+	  middle.
+
+	* string.c (rb_str_subseq, rb_str_substr, str_byte_substr): share
+	  middle of a string.
+
+Fri Apr 18 15:40:05 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c: use uintptr_t instead of VALUE because they are not ruby
+	  object.
+
+Fri Apr 18 14:51:42 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c: check str_strlen's argument, and add comment or
+	  use NULL if simply it uses str's enc.
+
+Fri Apr 18 14:32:40 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_strlen): use enc_strlen if the coderange is known.
+
+Fri Apr 18 14:21:21 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (enc_strlen): move UTF-8 optimization from str_strlen to
+	  enc_strlen.
+
+Fri Apr 18 08:50:18 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_getcwd_malloc): check if getcwd allocates
+	  buffer if NULL is given [ruby-core:62072] [Bug #9752]
+
+Thu Apr 17 16:28:10 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* prelude.rb: [DOC] Update Thread::exclusive docs by @stevenharman.
+
+Thu Apr 17 10:03:53 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils#copy_entry): update rdoc about
+	  preserve option and permissions, following r31123.
+	  [ruby-core:62065] [Bug #9748]
+
+Wed Apr 16 23:47:36 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (dont_recycle_block_arg): fix condition to recycle block
+	  argument.  lambda with rest can get internal array directly.
+	  [ruby-core:62060] [Bug #9749]
+
+Wed Apr 16 09:51:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey.c (ossl_pkey_verify): as EVP_VerifyFinal()
+	  finalizes only a copy of the digest context, the context must be
+	  cleaned up after initialization by EVP_MD_CTX_cleanup() or a
+	  memory leak will occur.  [ruby-core:62038] [Bug #9743]
+
+Tue Apr 15 19:36:42 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/win32.h (rb_w32_cmdvector): removed.
+
+	* win32/win32.c (rb_w32_sysinit): use WCHAR version of GetCommandLine()
+	  internally.
+
+	* win32/win32.c (w32_cmdvector): renamed from rb_w32_cmdvector.  use
+	  WCHAR* instead of char* internally.
+
+	these changes are expected to not changing the behavior yet.
+
+Tue Apr 15 19:26:05 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/extmk.rb: Re-generate extmk.mk and dummy makefiles only if
+	  really required.
+	  This fixes a problem to run multiple test-all concurrently as:
+	  make test-all & make test-all & make test-all & ...
+
+Tue Apr 15 12:49:53 2014  Sam Rawlins  <sam.rawlins@gmail.com>
+
+	* enum.c (enum_each_slice, enum_each_cons): make more efficient by
+	  allocating less and recycling block argument arrays if possible.
+	  [Fixes GH-596]
+
+Mon Apr 14 18:44:45 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): get base addrs in fill_lines to use it
+	  with dladdr_fbases introduced at r45563.
+	  it didn't get before if the executable is not pie.
+
+Mon Apr 14 18:05:48 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (main_exe_path): support FreeBSD.
+	  At least sh, csh, tcsh, bash, and zsh sets realpath of the main
+	  executable for dladdr, but gdb doesn't.
+
+Mon Apr 14 17:20:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (umethod_bind): use the ancestor iclass instead of new
+	  iclass to get rid of infinite recursion, if the defined module
+	  is already included.  [ruby-core:62014] [Bug #9721]
+
+Sun Apr 13 12:46:58 2014  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (SIZEOF_BDIGIT): Renamed from SIZEOF_BDIGITS.
+
+	* internal.h: Ditto.
+
+	* marshal.c: Ditto.
+
+	* rational.c: Ditto.
+
+Sun Apr 13 10:18:09 2014  Tanaka Akira  <akr@fsij.org>
+
+	* common.mk: Unused target, $(MKMAIN_CMD), removed.
+
+	* Makefile.in (MKMAIN_CMD): Unused macro removed.
+
+	* win32/Makefile.sub (MKMAIN_CMD): Ditto.
+
+Sat Apr 12 22:11:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_to_proc), proc.c (rb_block_clear_env_self): clear
+	  caller's self which is useless, so that it can get collected.
+	  [Fixes GH-592]
+
+Sat Apr 12 09:26:48 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* ext/openssl/ossl_ocsp.c:  [DOC] Document OpenSSL::OCSP.
+
+Fri Apr 11 18:52:38 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (ARY_SET): added.
+
+	  ARY_SET() is same functionality of RARRAY_ASET(), but
+	  it has an assertion (`ary' doesn't have shared array).
+
+Fri Apr 11 16:54:26 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c: make shared arrays WB-protected objects.
+
+	  Shared arrays were WB-unprotected object because
+	  sharing array can modify shared array's buffer
+	  if it occupied shared array.
+
+	  [sharing array (ary)] -> [shared array (shared)] -> <buff>
+	                |                                       A
+	                +---------------------------------------+
+	             write `buff' with WB(ary, &buff[i], obj)
+	             -> if `ary' and `shared' are old, then only `ary'
+	                will be remembered.
+	             -> traverse from `ary'. But `shared' is old, so
+	                that written `obj' is not marked.
+
+	  It cause WB miss so that shared arrays were WB-unprotected.
+	  (WB-unprotected objects are marked everytime if it is living)
+
+	  This patch insert WB() for `shared' if it is needed.
+
+Fri Apr 11 15:05:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_method_call_with_block, umethod_bind): call with
+	  IClass including the module for a module instance method.
+	  [ruby-core:61936] [Bug #9721]
+
+	* vm_insnhelper.c (vm_search_super_method): allow bound
+	  UnboundMethod case.
+
+Fri Apr 11 12:02:30 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (rb_dump_backtrace_with_lines): set base address
+	  which is retrieved from dladdr to dladdr_fbases, to skip already
+	  parsed objects.
+
+Fri Apr 11 12:44:50 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_reject): may be turned into a shared array during
+	  the given block.  [ruby-dev:48101] [Bug #9727]
+
+Thu Apr 10 23:41:21 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/net/ftp.rb (Net::FTP#login): [DOC] The default password for
+	  anonymous login was changed to "anonymous@" in r25313.
+
+Thu Apr 10 19:22:58 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_array.rb: remove useless `assert'.
+
+Thu Apr 10 19:11:11 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c (rb_ary_modify): remember shared array owner if a shared
+	  array owner is promoted and a shared array is not promoted.
+
+	  Now, shared array is WB-unprotected so that shared arrays are not
+	  promoted.  All objects referred from shared array should be marked
+	  correctly.
+
+	  [ruby-core:61919] [ruby-trunk - Bug #9718]
+
+	* test/ruby/test_array.rb: add a test for above.
+
+Thu Apr 10 18:57:12 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_verify_internal_consistency): move lines and enable
+	  allrefs_dump() on RGENGC_CHECK_MODE >= 4.
+
+Thu Apr 10 15:01:06 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (append_obj): clear allocated memory.
+
+	* addr2line.c (rb_dump_backtrace_with_lines): free `base_addrs'.
+
+Thu Apr 10 14:40:18 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_gc_writebarrier_unprotect_promoted): disable to dump debug
+	  message when RGENGC_CHECK_MODE == 0.
+
+Thu Apr 10 08:13:47 2014  Tanaka Akira  <akr@fsij.org>
+
+	* signal.c (check_stack_overflow): Don't use ucontext_t if ucontext.h
+	  is not available.
+	  Fixes build on Android (x86).
+
+Wed Apr  9 23:22:44 2014  Tanaka Akira  <akr@fsij.org>
+
+	* gc.c (mark_current_machine_context): Call SET_STACK_END.
+	  This reverts a hunk of r40703 by ko1.
+	  This fixes [ruby-dev:48098] [Bug #9717].
+
+Wed Apr  9 21:02:04 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (OBJ2UID1): Defined even if getpwnam_r is not usable.
+	  (OBJ2GID1): Defined even if getgrnam_r is not usable.
+	  This fixes compilation error on Android.
+
+Wed Apr  9 15:16:59 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_enc_default_internal): fix rdoc.  `__FILE__` is
+	  in filesystem encoding but not `default_internal`.
+	  [ruby-core:61894] [Bug #9713]
+
+Wed Apr  9 14:43:00 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb: more long timeout.
+	  This test failed under RGENGC_CHECK_MODE >= 2.
+
+Wed Apr  9 13:07:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: get rid of __builtin_setjmp/__builtin_longjmp on
+	  x64-mingw, which causes SEGV with callcc.
+	  [ruby-core:61887] [Bug #9710]
+
+Wed Apr  9 12:44:54 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_buf_cat): should round up the capacity by 4KiB,
+	  but not number of rooms.  [ruby-core:61886] [Bug #9709]
+
+Tue Apr  8 22:55:32 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/mkmf.rb (MakeMakefile#dir_config): [DOC] Improve
+	  documentation.
+
+Tue Apr  8 22:31:44 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename `RGENGC_THREEGEN' to `RGENGC_AGE2_PROMOTION'.
+
+	* gc.c (rgengc_rememberset_mark): don't promote, but remain in
+	  remember set for infant objects.
+
+	* gc.c (RVALUE_PROMOTE_INFANT, RVALUE_PROMOTE_YOUNG): count numbers
+	  in these functions.
+
+Mon Apr  7 21:11:49 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/socket.c (sock_s_getnameinfo): Save errno for EAI_SYSTEM.
+	  Reported by Saravana kumar.  [ruby-core:61820] [Bug #9697]
+	  Fixed by Heesob Park.  [ruby-core:61868]
+
+Mon Apr  7 07:20:23 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/xmlrpc/client.rb (do_rpc): don't check body length.
+	  If HTTP content-encoding is used, the length may be different.
+	  [Bug #8182] [ruby-core:53811]
+
+Mon Apr  7 02:39:48 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Matrix#cofactor [fix GH-568]
+	  Add first_minor [fix GH-568]
+	  Handle empty diagonal matrix case [fix GH-576]
+	  Patches by gogotanaka
+
+Sun Apr  6 08:52:50 2014  Bugra Barin  <bugrabarin@hotmail.com>
+
+	* dln.c (dln_load): use wchar version to load a library in
+	  non-ascii path on Windows.  based on the patch by Bugra Barin
+	  <bugrabarin AT hotmail.com> in [ruby-core:61845].  [Bug #9699]
+
+Sat Apr  5 19:36:33 2014  Tadayoshi Funaba  <tadf@dotrb.org>
+
+	* ext/date/date_core.c (d_lite_cmp): should compare with #<.
+
+Sat Apr  5 00:31:21 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/extconf.rb (rl_hook_func_t): check pointer type.
+	  [ruby-dev:48089] [Bug #9702]
+
+Fri Apr  4 07:13:44 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func___builtin_setjmp): should not skip
+	  flags restoration in RUBY_WERROR_FLAG by `break`.
+	  [ruby-dev:48086] [Bug #9698]
+
+Wed Apr  2 21:50:06 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func___builtin_setjmp): __builtin_longjmp()
+	  in Apple LLVM 5.1 (LLVM 3.4svn) uses `void**`, not `jmp_buf`.
+	  [Bug #9692]
+
+Wed Apr  2 20:57:15 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, gc.h (rb_objspace_each_objects_without_setup):
+	  Add a new (hidden) C-API to iterate objspace snapshot.
+
+	  This API is not safe to call any C-APIs in a given callback
+	  function. Be careful to use this C-API.
+
+Wed Apr  2 17:43:17 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func___builtin_setjmp): gcc 4.9 disallows a
+	  variable as the second argument of __builtin_longjmp().
+	  [ruby-core:61800] [Bug #9692]
+
+Wed Apr  2 15:12:18 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* common.mk: Use redmine-2.x url for DeveloperHowto wiki.
+	  [ruby-core:60657] [Bug #9511]
+
+Wed Apr  2 11:46:29 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/pathname/lib/pathname.rb (Pathname#join): Fix error with
+	  empty args. Reported by ko1 via IRC.
+
+	* test/pathname/test_pathname.rb (TestPathname#test_join): Add the
+	  test for above case.
+
+Tue Apr  1 11:39:57 2014  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb: Symbol HeaderConverter: strip leading/trailing space.
+	  Reported by Skye Shaw
+	  [Fixes GH-575]
+
+Tue Apr  1 11:34:04 2014  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb: Don't attempt to convert nil headers.
+	  Reported by Skye Shaw
+
+Tue Apr  1 17:29:35 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/config_files.rb (ConfigFiles.download): show failed URI.
+	  [ruby-core:61792] [Bug #9690]
+
+Tue Apr  1 12:06:49 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (rb_dump_backtrace_with_lines): don't depend hard coded
+	  symbol '_start'.
+
+	* addr2line.c (fill_lines): instead of above, get a dynamic symbol
+	  in the main executable and use it to know the base address.
+
+	* addr2line.c (follow_debuglink0): use obj_info_t instead of
+	  line_info_t to handle object related data.
+
+	* addr2line.c (main_exe_path): defined for Linux.
+
+Tue Apr  1 08:58:39 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* parse.y (rb_str_dynamic_intern): set mark bit if dynamic symbol
+	  is before sweeping.
+
+Tue Apr  1 07:37:00 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): use dynsym, which is used for dynamic
+	  linking and always exists, if there's no symtab.
+
+Tue Apr  1 07:27:15 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c (rb_print_backtrace): current implementation
+	  uses dladdr to get the path of objects.
+
+Mon Mar 31 23:57:45 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/extconf.rb: fix typo, `$defs` not `$DEFS`.
+	  [ruby-core:61756] [Bug #9578]
+
+Mon Mar 31 17:23:50 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/memory_status.rb: require envutil before accessing EnvUtil
+	  module.  reported by ko1 via twitter.
+
+Mon Mar 31 10:28:01 2014  Eric Wong  <e@80x24.org>
+
+	* st.c (st_init_table_with_size): update comment
+	  [Feature #9425]
+
+Sun Mar 30 23:39:26 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_accept, open_ifs_socket, socketpair_internal):
+	  reset inherit flag of socket to avoid unintentional inheritance of
+	  socket.  note that the return value of SetHandleInformation() is not
+	  verified intentionally because old Windows may return an error.
+	  [Bug #9688] [ruby-core:61754]
+
+Sat Mar 29 13:04:22 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_before_sweep): cap `malloc_limit' to
+	  gc_params.malloc_limit_max.  It can grow and grow with such case:
+	    `loop{"a" * (1024 ** 2)}'
+	  [Bug #9687]
+
+	  This issue is pointed by Tim Robertson.
+	  http://www.omniref.com/blog/blog/2014/03/27/ruby-garbage-collection-still-not-ready-for-production/
+
+Fri Mar 28 19:32:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (not_a_member): extract name error and use same error
+	  messages.  based on the patch by Marcus Stollsteimer <sto.mar AT
+	  web.de> at [ruby-core:61721].  [Bug #9684]
+
+Fri Mar 28 09:21:54 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/psych/psych.gemspec: update gemspec for psych-2.0.5
+
+Fri Mar 28 09:11:06 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/psych/lib/psych.rb: Merge psych-2.0.5. bump version to
+	  libyaml-0.1.6 for CVE-2014-2525.
+	* ext/psych/yaml/config.h: ditto.
+	* ext/psych/yaml/scanner.c: ditto.
+	* ext/psych/yaml/yaml_private.h: ditto.
+
+Thu Mar 27 18:58:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (match_regexp): set regexp for MatchData from string.
+
+	* re.c (rb_backref_set_string): create MatchData from string and
+	  set backref.
+
+	* string.c (rb_pat_search, rb_str_sub, rb_str_sub_bang, str_gsub),
+	  (scan_once, rb_str_scan, rb_str_partition): use rb_str_index
+	  instead of rb_reg_search() when pattern is a String.  based on
+	  the patch by Sam Rawlins <sam.rawlins@gmail.com> [Fixes GH-579]
+
+Thu Mar 27 11:58:55 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): check shdr[i].sh_type because even if
+	  .symtab section exists, the section's type can be SHT_NOBITS and
+	  actual data doesn't exist in the file.
+	  revert r45441.
+
+Wed Mar 26 14:57:35 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* parse.y: inline must be static (for mswin).
+	  fixed build error introduced at r45426.
+
+Wed Mar 26 14:33:00 2014  Narihiro Nakamura  <authornari@gmail.com>
+
+	* internal.h (USE_SYMBOL_GC): enable Symbol GC by default (USE_SYMBOL_GC == 1).
+
+Tue Mar 25 22:57:11 2014  Narihiro Nakamura  <authornari@gmail.com>
+
+	* parse.y: support Symbol GC. [ruby-trunk Feature #9634]
+	  See this ticket about Symbol GC.
+
+	* include/ruby/ruby.h:
+	  Declare few functions.
+	  * rb_sym2id: almost same as old SYM2ID but support dynamic symbols.
+	  * rb_id2sym: almost same as old ID2SYM but support dynamic symbols.
+	  * rb_sym2str: almost same as `rb_id2str(SYM2ID(sym))` but not
+	    pin down a dynamic symbol.
+	  Declare a new struct.
+	  * struct RSymbol: represents a dynamic symbol as object in
+	    Ruby's heaps.
+	  Add few macros.
+	  * STATIC_SYM_P: check a static symbol.
+	  * DYNAMIC_SYM_P: check a dynamic symbol.
+	  * RSYMBOL: cast to RSymbol
+
+	* gc.c: declare RSymbol. support T_SYMBOL.
+
+	* internal.h: Declare few functions.
+	  * rb_gc_free_dsymbol: free up a dynamic symbol. GC call this
+	    function at a sweep phase.
+	  * rb_str_dynamic_intern: convert a string to a dynamic symbol.
+	  * rb_check_id_without_pindown: not pinning function.
+	  * rb_sym2id_without_pindown: ditto.
+	  * rb_check_id_cstr_without_pindown: ditto.
+
+	* string.c (Init_String): String#intern and String#to_sym use
+	  rb_str_dynamic_intern.
+
+	* template/id.h.tmpl: use LSB of ID as a flag for determining a
+	  static symbol, so we shift left other ruby_id_types.
+
+	* string.c: use rb_sym2str instead `rb_id2str(SYM2ID(sym))` to
+	  avoid pinning.
+
+	* load.c: use xx_without_pindown function at creating temporary ID
+	  to avoid pinning.
+
+	* object.c: ditto.
+
+	* sprintf.c: ditto.
+
+	* struct.c: ditto.
+
+	* thread.c: ditto.
+
+	* variable.c: ditto.
+
+	* vm_method.c: ditto.
+
+Wed Mar 26 13:25:54 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): loop reverse order not to overwrite
+	  the basis of base addresses comparison.
+
+	* addr2line.c: use uintptr_t instead of intptr_t for pointers.
+
+	* addr2line.c (rb_dump_backtrace_with_lines): don't use syms.
+
+	* vm_dump.c (rb_print_backtrace): ditto.
+
+	* addr2line.h: ditto.
+
+Wed Mar 26 11:20:50 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object): internal objects are not dumpable.
+	  [ruby-core:61677] [Bug #9674]
+
+	* ext/thread/thread.c (undumpable): ConditionVariable and Queue
+	  are not dumpable.  [ruby-core:61677] [Bug #9674]
+
+Wed Mar 26 10:36:39 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (follow_debuglink): show message if it closes opened
+	  (and maybe used) elf binary.
+
+Wed Mar 26 10:34:25 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_line): pass and use offset instead of
+	  curobj_baseaddr.
+
+Wed Mar 26 09:07:48 2014  Yutaka Kanemoto  <kanemoto@ruby-lang.org>
+
+	* configure.in: add --disable-pie. [Feature #9673]
+
+Wed Mar 26 08:47:04 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): don't run fill_lines multiple times.
+
+Wed Mar 26 08:45:00 2014  Sam Rawlins  <sam.rawlins@gmail.com>
+
+	* internal.h: add prototype for rb_reg_search0
+
+	* re.c: rename rb_reg_search to rb_reg_search0, add set_backref_str
+	  argument to allow callers to indicate that they don't require the
+	  backref string to be allocated.
+
+	* string.c: don't allocate backref str if replacement string is provided
+
+	[GH-578] [Bug #9676] [ruby-core:61682]
+
+Wed Mar 26 08:29:43 2014  mo khan  <mo@mokhan.ca>
+
+	* lib/rubygems.rb: fix spelling of Jim Weirich.  [Fixes GH-577]
+
+Wed Mar 26 01:55:45 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (rb_dump_backtrace_with_lines): a function to get must
+	  be a function in the main executable, whose absolute path is not
+	  available by dladdr, and ruby get it by /proc/self/exe on Linux.
+
+Wed Mar 26 01:34:50 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): skip if path is NULL.
+
+Tue Mar 25 23:57:17 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): only a newline after label should be
+	  significant.  [ruby-core:61658] [Bug #9669]
+
+Tue Mar 25 23:32:25 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/pathname/lib/pathname.rb (Pathname#join): remove unnecessary
+	  unshift.
+
+	* test/pathname/test_pathname.rb (TestPathname#test_join): add tests.
+
+Tue Mar 25 16:47:36 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lex_state_e, parser_params, f_arglist, parser_yylex):
+	  separate EXPR_LABELARG from EXPR_BEG and let newline significant,
+	  so that required keyword argument can place at the end of
+	  argument list without parentheses.  [ruby-core:61658] [Bug #9669]
+
+Mon Mar 24 22:19:56 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (ripper_initialize): filename can not be modified.
+
+Mon Mar 24 15:19:47 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (rb_dump_backtrace_with_lines): fetch path of the
+	  executable from /proc/self/exe on Linux.
+
+Mon Mar 24 14:14:37 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gc.c: Fix up default GC params by @csfrancis [fix GH-556]
+
+Mon Mar 24 13:13:36 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (parse_debug_line_cu): explicitly specify signed char
+	  because DWARF's line_Base is signed char and char maybe unsigned.
+	  patched by Rei Odaira. [ruby-dev:48068] [Bug #9654]
+
+Sun Mar 23 11:03:50 2014  Kohei Suzuki  <eagletmt@gmail.com>
+
+	* vm_method.c (rb_method_entry_get_without_cache): me->klass is 0
+	  for a method aliased in a module.  [ruby-core:61636] [Bug #9663]
+
+Sun Mar 23 08:12:27 2014  Eric Wong  <e@80x24.org>
+
+	* st.c (hash_pos): use bitwise AND to avoid slow modulo op
+	  (new_size): power-of-two sizes for hash_pos change
+	  (st_numhash): adjust for common keys due to lack of prime modulo
+	  [Feature #9425]
+	* hash.c (rb_any_hash): right shift for symbols
+	* benchmark/bm_hash_aref_miss.rb: added to show improvement
+	* benchmark/bm_hash_aref_sym_long.rb: ditto
+	* benchmark/bm_hash_aref_str.rb: ditto
+	* benchmark/bm_hash_aref_sym.rb: ditto
+	* benchmark/bm_hash_ident_num.rb: added to prevent regression
+	* benchmark/bm_hash_ident_obj.rb: ditto
+	* benchmark/bm_hash_ident_str.rb: ditto
+	* benchmark/bm_hash_ident_sym.rb: ditto
+
+Sat Mar 22 22:56:45 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): compare the file names of object in which
+	  symbols exist. [Bug #9654] [ruby-dev:48058]
+
+Sat Mar 22 06:46:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi/util.rb (escape_html, unescape_html): make synonyms
+	  aliases instead of wrapper methods.
+
+	* lib/cgi/util.rb (escape_element, unescape_element): ditto.
+	  [Fixes GH-573]
+
+Fri Mar 21 21:57:34 2014  Akinori MUSHA  <knu@iDaemons.org>
+
+	* configure.in: Fix a build problem with clang and --with-opt-dir.
+	  If ruby is configured with --with-opt-dir=dir when using clang
+	  as compiler, a warning `clang: warning: argument unused during
+	  compilation: '-I dir'` is emitted almost every time clang
+	  compiles a file.  Unfortunately, RUBY_CHECK_PRINTF_PREFIX takes
+	  any output from the compiler as fatal error, and the check thus
+	  fails due to the warning.  This is an attempt to fix the problem
+	  by adding a flag -Qunused-arguments to CFLAGS locally in the
+	  function to suppress the warning. [ruby-dev:48062] [Bug #9658]
+	  [Fixes GH-571] https://github.com/ruby/ruby/pull/571
+
+Fri Mar 21 16:31:56 2014  Zachary Scott  <e@zzak.io>
+
+	* gc.c: [DOC] Fix call-seq for GC.start by @jasonrclark [Fixes GH-572]
+	  https://github.com/ruby/ruby/pull/572
+
+Thu Mar 20 11:37:28 2014  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb: Fixed a broken regular expression that was causing
+	  CSV to miss escaping some special meaning characters when used
+	  in parsing.
+	  Reported by David Unric
+	  [ruby-core:54986] [Bug #8405]
+
+Thu Mar 20 16:53:07 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (objspace_malloc_increase): should not invoke
+	  garbage_collect_with_gvl() here on non-ruby threads.
+
+	  Should just ignore the malloc_increase.
+
+	  This issue is pointed by Eric Wong [ruby-core:61519].
+
+Thu Mar 20 13:05:16 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* struct.c (rb_struct_alloc): use RARRAY_CONST_PTR() instead of
+	  RARRAY_PTR().
+
+Thu Mar 20 12:59:39 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/intern.h (rb_obj_call_init, rb_class_new_instance):
+	  constify a parameter (VALUE *).
+	  I believe this incompatibility doesn't break any code.
+	  However, if you have trouble, please tell us.
+
+	* eval.c, object.c: ditto.
+
+Thu Mar 20 12:31:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_get_without_cache): get rid of
+	  infinite recursion at aliases in a subclass and a superclass.
+	  return actually defined class for other than singleton class.
+	  [ruby-core:60431] [Bug #9475]
+
+Wed Mar 19 17:13:06 2014  Eric Wong  <e@80x24.org>
+
+	* time.c (time_mload): freeze and preserve marshal-loaded time zone
+	* test/ruby/test_time.rb: add test for GC on loaded object
+	  [Bug #9652]
+
+Tue Mar 18 23:20:12 2014  Shota Fukumori  <her@sorah.jp>
+
+	* vm_eval.c (eval_string_with_cref): Unify to use NIL_P.
+
+Tue Mar 18 22:03:41 2014  Shota Fukumori  <her@sorah.jp>
+
+	* vm_eval.c (eval_string_with_cref): Use file path even if scope is
+	  given. Related to [ruby-core:56099] [Bug #8662] and r42103.
+
+Mon Mar 17 13:17:47 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* enumerator.c (enumerator_block_call): use RARRAY_CONST_PTR()
+	  instead of RARRAY_PTR().
+
+	* io.c (rb_io_s_popen): ditto.
+
+	* numeric.c (num_step_size): ditto.
+
+	* vm_eval.c (rb_apply): ditto.
+
+	* vm_eval.c (rb_eval_cmd): ditto.
+
+Mon Mar 17 10:11:59 2014  Eric Wong  <e@80x24.org>
+
+	* variable.c (rb_const_set): delete existing entry on redefinition
+	  [Bug #9645]
+	* test/ruby/test_const.rb (test_redefinition): test for leak
+
+Sun Mar 16 21:33:01 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/time.rb: [DOC] Fix timezone in example of Time.parse [Bug #9521]
+	  Based on patch by @stomar
+
+Sun Mar 16 13:21:40 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DLDFLAGS): insert a space between option and its
+	  argument for non-GCC compilers.  [ruby-core:61429] [Bug #9624]
+
+Sun Mar 16 08:05:06 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (objspace_xcalloc): fix GC accounting
+
+Sun Mar 16 06:33:35 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): return address is just after calling
+	  address. Therefore noreturn function with tail call's return
+	  address may be in another function.
+
+Sun Mar 16 05:51:55 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/gserver.rb: [DOC] Fixed typo in example by @stomar [Bug #9543]
+
+Sat Mar 15 18:54:03 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/.document: remove refinement from documentable directories.
+
+Sat Mar 15 11:02:58 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DLDFLAGS): check for each options to control
+	  symbol resolution.  [ruby-core:61429] [Bug #9624]
+
+Sat Mar 15 07:02:35 2014  Eric Wong  <e@80x24.org>
+
+	* st.c (st_update): remove unnecessary assignment
+
+Fri Mar 14 14:58:38 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c (fill_lines): fetch symbol names from ELF binary's
+	  symbol table if it is built with cc -g and not stripped.
+	  Now ruby can show static symbols on Linux though glibc's
+	  backtrace_symbols(3) don't show them.
+
+	* addr2line.c (rb_dump_backtrace_with_lines): use dladdr(3) to
+	  detect what object file declares the symbol because
+	  dl_iterate_phdr can't detect the main executable file
+	  and codes on the stack.
+	  NOTE: signal trampolines sometimes on the user stack. (FreeBSD)
+
+	* addr2line.c (rb_dump_backtrace_with_lines): stop showing
+	  backtrace if the function's name is main.
+	  NOTE: FreeBSD's backtrace (libexecinfo) shows _start and
+	  an additional address. Why it doesn't remove them on dladdr phase
+	  is, dladdr may fail to detect the main function but detect
+	  as _start function. Therefore it must be after scanning
+	  the symbol table and getting correct name.
+
+
+Fri Mar 14 12:07:46 2014  Zachary Scott  <e@zzak.io>
+
+	* doc/syntax/literals.rdoc: [DOC] Single quote strings allows escape
+	  of backslash as well, patch by @idupree [Fixes GH-553]
+	  https://github.com/ruby/ruby/pull/553
+
+Fri Mar 14 01:18:24 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (invoke_block_from_c): add splattable argument.
+
+	* vm.c (vm_invoke_proc): disallow to splat when directly invoked.
+
+	* vm_insnhelper.c (vm_callee_setup_arg_complex, vm_callee_setup_arg):
+	  relax arity check of yielded lambda.  [ruby-core:61340] [Bug #9605]
+
+	* test/ruby/test_yield.rb (TestRubyYieldGen#emu_bind_params): no
+	  longer raise ArgumentError when splatting to lambda.
+
+Thu Mar 13 23:51:02 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/-test-/win32/dln/libdlntest.c (dlntest_ordinal): no need to
+	  specify export in the source file because .def file do it.
+	  get rid of warning on linking.
+
+Wed Mar 12 11:19:03 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_callee_setup_arg): disable fastpath if splat
+	  argument, since argc may differ for each calls.
+	  [ruby-core:61422] [Bug #9622]
+
+	* vm_insnhelper.c (vm_callee_setup_arg): turn a macro into an
+	  inline function.
+
+Wed Mar 12 07:26:05 2014  Eric Wong  <e@80x24.org>
+
+	* insns.def (opt_regexpmatch2): respect redefined match op
+	  Thanks to Sam Rawlins for the fix.
+	* test/ruby/test_string.rb: test based on Tsuyoshi Sawada's report
+	  [Bug #9581]
+
+Tue Mar 11 22:31:25 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/.document: add objspace/objspace_dump.c to document file.
+
+Tue Mar 11 22:22:38 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* test/objspace/test_objspace.rb (TestObjSpace#test_dump_uninitialized_file):
+	  remove dependency on json library.
+
+Tue Mar 11 10:55:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* README.EXT{,.ja} (Appendix B): update contents of `ruby_options`
+	  and replace `ruby_run` with `ruby_run_node`.  based on the patch
+	  by Kaneko Yuichiro at [ruby-dev:48030] [Bug #9619].
+
+Tue Mar 11 06:54:00 2014  Scott Francis  <scott.francis@shopify.com>
+
+	* ext/objspace/objspace_dump.c: Check fptr before trying to dump RFILE
+	  object fd. [GH-562]
+
+	* test/objspace/test_objspace.rb: add test
+
+Tue Mar 11 02:04:36 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): show vm maps on FreeBSD.
+
+	* vm_dump.c (procstat_vm): copied from FreeBSD.
+	  http://svnweb.freebsd.org/base/head/usr.bin/procstat/procstat_vm.c?revision=261780
+
+Mon Mar 10 12:14:26 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: always check dladdr(1).
+
+	* addr2line.c (fill_lines): show the line number in C backtrace if
+	  ruby is built without --enable-shared (PIE) on Linux.
+	  patch is originally by Shinichiro Hamaji
+	  https://twitter.com/shinh/status/441957774264504321
+	  NOTE: ld doesn't insert __executable_start for PIE.
+	  dladdr(3)'s argument must be a function pointer.
+
+Mon Mar 10 10:51:17 2014  ksss  <co000ri@gmail.com>
+
+	* test/ruby/test_enumerator.rb (test_iterators): fix test for hash
+	  iterators.  [Fixes GH-558]
+
+Sun Mar  9 14:14:49 2014  Eric Wong  <e@80x24.org>
+
+	* class.c (rb_class_subclass_add): use xmalloc
+	* class.c (rb_module_add_to_subclasses_list): ditto
+	* class.c (rb_class_remove_from_super_subclasses): use xfree
+	* class.c (rb_class_remove_from_module_subclasses): ditto
+	  [Bug #9616]
+
+Sun Mar  9 13:51:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/function.c (function_call): fix memory leak when an
+	  exception occurs at argument conversion or the function call.
+
+Sun Mar  9 06:42:40 2014  Eric Wong  <e@80x24.org>
+
+	* variable.c (struct global_variable): shrink by 8 bytes on 64-bit
+
+Sat Mar  8 17:42:51 2014  Eric Wong  <e@80x24.org>
+
+	* vm.c (add_opt_method): cleanup to use rb_method_entry_at
+
+Sat Mar  8 13:46:40 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/dl/cptr.c (dlptr_free), ext/dl/handle.c (dlhandle_free),
+	  ext/fiddle/handle.c (fiddle_handle_free),
+	  ext/fiddle/pointer.c (fiddle_ptr_free): fix memory leak.
+	  based on the patch Heesob Park at [ruby-dev:48021] [Bug #9599].
+
+Sat Mar  8 13:30:39 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (obj2uid, obj2gid): now getpwnam_r() and getgrnam_r()
+	  may need larger buffers than sysconf values, so retry with
+	  expanding the buffer when ERANGE is returned.
+	  [ruby-core:61325] [Bug #9600]
+
+Fri Mar  7 19:29:13 2014  Eric Wong  <e@80x24.org>
+
+	* vm_eval.c (vm_call0_body): use RARRAY_CONST_PTR
+	  (check_funcall_exec): ditto
+	  [ruby-core:61360]
+
+Fri Mar  7 19:14:11 2014  Eric Wong  <e@80x24.org>
+
+	* vm_eval.c (vm_call0_body): fix RB_GC_GUARD location
+	  (check_funcall_exec): ditto
+	  [Bug #9609]
+
+Fri Mar  7 14:48:17 2014  Narihiro Nakamura  <authornari@gmail.com>
+
+	* parse.y (ENC_SINGLE): Unused macro removed.
+
+Fri Mar  7 12:06:19 2014  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* test/openssl/test_ssl.rb: Reuse TLS default options from
+	  OpenSSL::SSL::SSLContext::DEFAULT_PARAMS.
+
+Thu Mar  6 15:15:24 2014  Zachary Scott  <e@zzak.io>
+
+	* doc/syntax/assignment.rdoc: [DOC] Fix assignment directions
+	  By @idupree [Fixes GH-555] https://github.com/ruby/ruby/pull/555
+
+Thu Mar  6 15:07:18 2014  Zachary Scott  <e@zzak.io>
+
+	* doc/syntax/methods.rdoc: [DOC] Fix example for block arguments
+	  By @idupree [Fixes GH-554] https://github.com/ruby/ruby/pull/554
+
+Thu Mar  6 10:33:31 2014  Martin Bosslet  <Martin.Bosslet@gmail.com>
+
+	* lib/openssl/ssl.rb: Explicitly whitelist the default
+	  SSL/TLS ciphers. Forbid SSLv2 and SSLv3, disable
+	  compression by default.
+	  Reported by Jeff Hodges.
+	  [ruby-core:59829] [Bug #9424]
+
+Wed Mar  5 15:56:18 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_arg_asgn): define optional arguments as argument
+	  variables in the rhs default expressions.
+	  [ruby-core:61299] [Bug #9593]
+
+Wed Mar  5 11:58:30 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl.c (ossl_make_error): check NULL for unknown
+	  error reasons with old OpenSSL, and insert a colon iff formatted
+	  message is not empty.
+
+Wed Mar  5 00:42:00 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* ext/pathname/lib/pathname.rb (Pathname#find): add "ignore_error"
+	  keyword argument defaulted to true as well as Find#find.
+
+Tue Mar  4 23:00:18 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_eval.rb (TestEval#make_test_binding): renamed.
+	  it's not test method.
+
+Tue Mar  4 20:50:59 2014  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* st.c (st_foreach): fix type of hash. not st_data_t but st_index_t.
+
+Tue Mar  4 19:41:40 2014  Tanaka Akira  <akr@fsij.org>
+
+	* Makefile.in: ".DEFAULT" target removed because it is not for
+	  specifying default target.
+
+Tue Mar  4 00:25:35 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/find.rb (Find#find): should pass ignore_error option to enumerators.
+
+Mon Mar  3 13:27:35 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/test_find.rb (TestFind#test_unsearchable_dir): ruby cannot make
+	  directory unreachable by owner on Windows.
+
+Mon Mar  3 08:10:04 2014  Eric Wong  <e@80x24.org>
+
+	* vm_method.c (rb_method_entry_get_without_cache): disable GMC
+	  writing if GMC is disabled.
+	  [ruby-core:61218]
+
+Mon Mar  3 07:47:17 2014  Eric Wong  <e@80x24.org>
+
+	* README.EXT: wrap GetDBM with do/while(0)
+	* README.EXT.ja: ditto
+	* ext/dbm/dbm.c: ditto, likewise for GetDBM2
+	* ext/gdbm/gdbm.c: ditto
+	* ext/sdbm/init.c: ditto
+	  [ruby-core:61217]
+
+Mon Mar  3 07:17:31 2014  Zachary Scott  <e@zzak.io>
+
+	* NEWS: [DOC] Update doc regarding filesystem load when flushing IO
+
+Mon Mar  3 04:37:50 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_fsync): need to fsync even if on Windows.  fixed mistake
+	  of r45254 and r45256.
+
+Mon Mar  3 04:21:34 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/win32ole: get rid of warnings (unused variable).
+
+Mon Mar  3 02:53:53 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (rb_io_flush_raw): [EXPERIMENTAL] remove force syncing for Win32
+	  to speed up IO.  this may break some tests, and they'll be fixed
+	  later.
+	  [ruby-core:58570] [Bug #9153]
+
+Mon Mar  3 00:17:43 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_backtrace.rb: get rid of warnings.  unused variable,
+	  shadowing.
+
+Sun Mar  2 11:15:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/find.rb (Find#find): add "ignore_error" keyword argument
+	  defaulted to true.  [ruby-core:51025] [Feature #7596]
+
+Sun Mar  2 11:13:30 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/extconf.rb (rl_hook_func_t): define as Function for
+	  very old readline versions.  [ruby-core:61209] [Bug #9578]
+
+Sun Mar  2 10:47:58 2014  Eric Wong  <e@80x24.org>
+
+	* load.c (ruby_init_ext): make idempotent to suppress warnings
+
+Sat Mar  1 19:51:42 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb (Open3.capture3): Ignore Errno::EPIPE for writing
+	  stdin_data.
+	  (Open3.capture2): Ditto.
+	  (Open3.capture2e): Ditto.
+
+Sat Mar  1 19:06:47 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (ruby_gc_set_params): simplify condition
+
+Sat Mar  1 16:18:40 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/readline/readline.c (Init_readline): Use rl_hook_func_t instead
+	  of Function to support readline-6.3.  (rl_hook_func_t is available
+	  since readline-4.2.)
+	  Reported by Dmitry Medvinsky.  [ruby-core:61141] [Bug #9578]
+
+Sat Mar  1 16:05:58 2014  Eric Wong  <e@80x24.org>
+
+	* gc.c (ruby_gc_set_params): fix building without RGenGC
+
+Sat Mar  1 11:08:00 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: support dumping Encoding
+	  objects.
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: support loading Encoding
+	  objects.
+
+	* test/psych/test_encoding.rb: add test
+
+	* ext/psych/lib/psych.rb: add version
+
+Sat Mar  1 10:52:34 2014  Zachary Scott  <e@zzak.io>
+
+	* README.EXT.ja: [DOC] Fix typo "macro macro" @utenmiki [Fixes GH-551]
+	  https://github.com/ruby/ruby/pull/551
+
+Fri Feb 28 11:16:55 2014  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* numeric.c: Fix Numeric#step with 0 unit [Bug #9575]
+
+Thu Feb 27 17:59:01 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/optparse.rb: [DOC] Add example of generating help with optparse.
+	  Patch by @joelmccracken documenting-ruby/ruby#19
+	  https://github.com/documenting-ruby/ruby/pull/19
+
+Thu Feb 27 12:10:09 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (ruby_num_interval_step_size): check signs and get rid
+	  of implementation dependent behavior of negative division.
+	  [ruby-core:61106] [Bug #9570]
+
+Thu Feb 27 03:55:45 2014  Zachary Scott  <e@zzak.io>
+
+	* thread.c: [DOC] Typo in comment for _FORTIFY_SOURCE [Fixes GH-548]
+	  Patch by @qnet-herwin https://github.com/ruby/ruby/pull/548
+
+Wed Feb 26 18:43:43 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_pages_free_unused_pages): check tomb page availability
+	  at first.
+	  And return immediately if we don't touch sorted list any more.
+
+Wed Feb 26 14:10:44 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (setup_exception): preserve exception class name encoding
+	  in debug mode messages.
+
+	* eval.c (setup_exception): preserve errinfo across calling #to_s
+	  method on the exception.  [ruby-core:61091] [Bug #9568]
+
+Wed Feb 26 01:29:27 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (sym_find): Add Symbol.find(str), which returns whether given
+	  string is defined as symbol or not. [Feature #7854]
+
+Tue Feb 25 22:52:02 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/dl/dl.c (rb_dl_realloc): use NUM2SIZET instead of NUM2INT.
+
+	* ext/fiddle/fiddle.c (rb_fiddle_realloc): ditto.
+
+Tue Feb 25 22:49:30 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/dl/dl.c (rb_dl_malloc): use NUM2SIZET instead of NUM2INT.
+	  Coverity Scan found this bug.
+
+	* ext/fiddle/fiddle.c (rb_fiddle_malloc): ditto.
+
+Tue Feb 25 12:06:13 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/Makefile.sub: define PACKED_STRUCT.
+
+Mon Feb 24 21:41:56 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* README.md, README.ja.md: removed (wrong) mode setting for emacs.
+
+Mon Feb 24 20:05:41 2014  Eric Wong  <e@80x24.org>
+
+	* configure.in: define PACKED_STRUCT_UNALIGNED for x86*
+	* timev.h (struct vtm): use PACKED_STRUCT_UNALIGNED
+	* time.c (struct time_object): ditto
+	  [Bug #9558] non-x86 cannot safely access unaligned addresses
+
+Mon Feb 24 18:10:02 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/fiddle/test_function.rb: remove unused variables.
+	* test/fileutils/test_fileutils.rb: ditto.
+	* test/io/console/test_io_console.rb: ditto.
+
+Mon Feb 24 12:37:51 2014  Eric Wong  <e@80x24.org>
+
+	* configure.in: use -Wno-packed-bitfield-compat for GCC 4.4+
+	  use __attribute__((packed)) if available
+	* timev.h: shrink and pack struct vtm
+	* time.c: pack struct time_object and adjust/introduce helpers
+	  [ruby-core:60794]
+
+Sun Feb 23 17:55:50 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/xmltokens.rb: Add missing non ASCII valid characters
+	  to element name characters. Now, REXML name tokens exactly
+	  match "[5] Name" in the XML spec and "[4] NCName" in the
+	  Namespaces in XML spec. See comment about the details.
+	  [Bug #9539]  [ruby-core:60901]
+	  Reported by Mario Barcala. Thanks!!!
+
+	* test/rexml/xpath/test_node.rb: Add tests for the above case.
+
+Sun Feb 23 12:18:54 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (inet_pton): use rb_w32_inet_pton, instead of
+	  inet_pton directly, which is unavailable on older version Windows.
+
+	* include/ruby/win32.h, win32/win32.c (rb_w32_inet_pton): add a
+	  wrapper function for inet_pton minimum supported client is
+	  Vista, as well as inet_ntop.
+
+Sun Feb 23 11:33:25 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/net/imap/test_imap.rb: remove unused variables.
+	* test/net/imap/test_imap_response_parser.rb: ditto.
+	* test/net/pop/test_pop.rb: ditto.
+
+Sun Feb 23 02:19:51 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (bind_random_port): Rescue EPERM for FreeBSD which
+	  security.mac.portacl.port_high is changed.
+	  See mac_portacl(4) for details.
+	  Reported by Jakub Szafranski.  [ruby-core:60917] [Bug #9544]
+
+Sat Feb 22 23:17:01 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/xpath_parser.rb: Fix indent.
+
+Sat Feb 22 23:15:35 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/xpath/test_attribute.rb: Simplify.
+
+Sat Feb 22 20:28:47 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: more friendly.
+
+Sat Feb 22 20:24:43 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/test_xpath*.rb: Move to ...
+	* test/rexml/xpath/*.rb: ... here.
+
+Sat Feb 22 20:04:41 2014  Kouhei Sutou  <kou@cozmixng.org>
+
+	* test/rexml/listener.rb: Untabify.
+
+Sat Feb 22 19:07:31 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/io/console/console.c (console_dev): need read access for conout$
+	  because some functions need it.  [Bug#9554]
+
+Sat Feb 22 18:40:58 2014  Eric Wong  <e@80x24.org>
+
+	* .gitignore: ignore benchmark files
+
+Sat Feb 22 01:22:24 2014  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* bignum.c (bary_mul_precheck): fix a copy-paste error.
+	  Coverity Scan found this bug.
+
+Sat Feb 22 00:58:51 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/socket/raddrinfo.c (rb_getaddrinfo): second argument of
+	  MEMZERO is type.  Coverity Scan found this bug.
+
+Fri Feb 21 23:47:24 2014  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* ext/socket/init.c (wait_connectable): break if the socket is
+	  writable to avoid infinite loops on FreeBSD and other platforms
+	  which conforms to SUSv3.  This problem cannot be reproduced with
+	  loopback interfaces, so it's hard to write test code.
+	  rsock_connect() and wait_connectable() are overly complicated, so
+	  they should be refactored, but I commit this fix as a workaround
+	  for the release of Ruby 1.9.3 scheduled on Feb 24.
+	  [ruby-core:60940] [Bug #9547]
+
+Fri Feb 21 23:03:39 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: added to handle redmine tickets.
+
+Fri Feb 21 20:42:01 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_mod_init_copy): do nothing if copying self.
+	  [ruby-dev:47989] [Bug #9535]
+
+	* hash.c (rb_hash_initialize_copy): ditto.
+
+Fri Feb 21 16:45:54 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/logger.rb (next_rotate_time, previous_period_end): consider
+	  DST change.
+
+	* lib/logger.rb (Logger::LogDevice#check_shift_log): compare the
+	  current time with the time for the next rotation to fix rotation
+	  miss when date changed between the comparison and log writing.
+	  based on the patch by megayu <yuhg2310 AT gmail.com>.
+	  [Fixes GH-539]
+
+Fri Feb 21 10:39:33 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	*  test/monitor/test_monitor.rb: remove unused variables.
+	*  test/resolv/test_dns.rb: ditto.
+	*  test/rexml/test_functions.rb: ditto.
+	*  test/rss/test_setup_maker_itunes.rb: ditto.
+
+Fri Feb 21 09:48:56 2014  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): only retry on error
+	  (bsock_recvmsg_internal): ditto
+	* test/socket/test_unix.rb: test above for infinite loop
+
+Fri Feb 21 08:27:19 2014  Eric Wong  <e@80x24.org>
+
+	* include/ruby/ruby.h (RB_GC_GUARD):
+	  use rb_gc_guarded_ptr_val on non-GCC/MSC
+	* gc.c (rb_gc_guarded_ptr_val): rename and adjust argument.
+	  RB_GC_GUARD should be robust enough for any compiler.
+	  [ruby-core:60816] [Bug #7805]
+
+Thu Feb 20 22:21:26 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (numeric_getaddrinfo): Use xcalloc.
+	  Suggested by Eric Wong.
+	  https://bugs.ruby-lang.org/issues/9525#note-14
+
+Thu Feb 20 11:21:13 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_flatten): fix behavior of flatten(-1).
+	  [ruby-dev:47988] [Bug #9533]
+
+	* test/ruby/test_array.rb: test for above.
+
+Wed Feb 19 18:57:02 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: Bypass getaddrinfo() if node and serv are numeric.
+	  Reporeted by Naotoshi Seo.  [ruby-core:60801] [Bug #9525]
+
+	* ext/socket/extconf.rb: Detect struct sockaddr_in6.sin6_len.
+
+	* ext/socket/sockport.h (SET_SIN6_LEN): New macro.
+	  (INIT_SOCKADDR_IN6): Ditto.
+
+	* ext/socket/rubysocket.h (struct rb_addrinfo): Add
+	  allocated_by_malloc field.
+
+	* ext/socket/raddrinfo.c (numeric_getaddrinfo): New function.
+	  (rb_getaddrinfo): Call numeric_getaddrinfo at first.
+	  (rb_freeaddrinfo): Free struct addrinfo properly when it is
+	  allocated by numeric_getaddrinfo.
+
+Wed Feb 19 18:31:48 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: Wrap struct addrinfo by struct rb_addrinfo.
+
+Wed Feb 19 17:47:01 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ipsocket.c (ip_s_getaddress): Don't access freed memory.
+
+Wed Feb 19 11:39:41 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: it must see rb_cv_broken_memmem not rb_cv_func_memmem.
+
+Tue Feb 18 23:18:41 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/socket/test_socket.rb: unix socket is required by test case.
+
+Tue Feb 18 20:48:38 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/socket/test_addrinfo.rb: remove unused variables.
+	* test/socket/test_nonblock.rb: ditto.
+	* test/socket/test_socket.rb: ditto.
+	* test/socket/test_unix.rb: ditto.
+	* test/testunit/test_parallel.rb: ditto.
+	* test/webrick/test_filehandler.rb: ditto.
+	* test/xmlrpc/test_features.rb: ditto.
+	* test/zlib/test_zlib.rb: ditto.
+
+Tue Feb 18 14:27:18 2014  Shota Fukumori  <her@sorah.jp>
+
+	* lib/test/unit.rb: Requires minitest < 5.0.0 if Gem is available.
+
+Tue Feb 18 14:24:07 2014  Shota Fukumori  <her@sorah.jp>
+
+	* lib/test/unit/test-unit.gemspec: Add minitest < 5.0.0 dependency
+
+	* tool/rbinstall.rb: Add empty implementations for `add_dependency`,
+	  `add_runtime_dependency`, `add_development_dependency` for
+	  Gem::Specification.
+
+Tue Feb 18 12:06:39 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (FILE_COUNT): Removed.  (win32.c defines it in itself.)
+	  (FILE_READPTR): Ditto.
+
+Tue Feb 18 09:35:44 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test/psych/test_string.rb: remove unused variables.
+	* test/test/psych/test_yaml.rb: ditto.
+
+Mon Feb 17 21:31:31 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb: ignore warning messages for running with -w
+	  option such as chkbuild.
+
+Mon Feb 17 20:00:27 2014  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: Move BDIGIT and related definitions from
+	  include/ruby/defines.h.
+
+Mon Feb 17 17:41:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (marshal_dump, marshal_load): do not recycle wrapper
+	  objects, to prevent from segfault with continuation.
+	  [ruby-dev:47970] [Bug #9523]
+
+Mon Feb 17 15:43:59 2014  Zachary Scott  <e@zzak.io>
+
+	* doc/keywords.rdoc: [DOC] Add keywords doc by documenting-ruby/ruby#29
+	  https://github.com/documenting-ruby/ruby/pull/29
+
+Mon Feb 17 12:31:31 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (get_envparam_double): fix a warning message.
+
+Mon Feb 17 12:09:52 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: introduce new environment variable
+	  "RUBY_GC_HEAP_OLDOBJECT_LIMIT_FACTOR" to control major/minor GC
+	  frequency.
+
+	  Do full GC when the number of old objects is more than R * N
+	  where R is this factor and
+                N is the number of old objects just after last full GC.
+
+	* test/ruby/test_gc.rb: add a test.
+
+Mon Feb 17 11:28:40 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_pty.rb: ignore warnings to unused variables.
+
+Mon Feb 17 11:27:36 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_find.rb: remove unused variables.
+
+Sun Feb 17 02:12:00 2014  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_initialize): Insert GC guard.
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_global_new): ditto.
+
+Sun Feb 16 15:53:36 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_securerandom.rb: File.exists? is deprecated. use File.exist?
+
+Sun Feb 16 15:05:00 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/pathname/test_pathname.rb: File.exists? is deprecated. use File.exist?
+
+Sun Feb 16 15:00:28 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/net/ftp/test_ftp.rb: remove unused variables.
+	* test/logger/test_logger.rb: ditto.
+
+Sun Feb 16 14:52:46 2014  Eric Wong  <e@80x24.org>
+
+	* dir.c (dir_s_glob): RB_GC_GUARD instead of volatile
+
+Sun Feb 16 14:33:52 2014  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h (RBIGNUM_SIGN): Defined for compatibility.
+	  (RBIGNUM_POSITIVE_P): Ditto.
+	  (RBIGNUM_NEGATIVE_P): Ditto.
+
+Sun Feb 16 12:46:47 2014  Eric Wong  <e@80x24.org>
+
+	* io.c (rb_f_backquote): trade volatile for manual recycle
+	  rb_gc_force_recycle ensures object is visible until recycle
+
+Sun Feb 16 11:55:14 2014  Eric Wong  <e@80x24.org>
+
+	* marshal.c (marshal_dump): use rb_gc_force_recycle for GC-safety
+	  (marshal_load): ditto
+	  [ruby-core:60730] [Bug #7805]
+
+Sun Feb 16 08:11:23 2014  Zachary Scott  <e@zzak.io>
+
+	* README.EXT.ja: [DOC] Fix typo by @utenmiki [Fixes GH-534]
+	  https://github.com/ruby/ruby/pull/534
+
+Sun Feb 16 07:48:20 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/bigdecimal/bigdecimal.c (BIGNUM_ZERO_P): Unused macro removed.
+
+Sun Feb 16 06:12:23 2014  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h: Rename macro names: RBIGNUM_FOO to BIGNUM_FOO.
+	  (BIGNUM_EMBED_LEN_NUMBITS): Renamed from RBIGNUM_EMBED_LEN_NUMBITS.
+	  (BIGNUM_EMBED_LEN_MAX): Renamed from RBIGNUM_EMBED_LEN_MAX.
+	  (BIGNUM_SIGN_BIT): Renamed from RBIGNUM_SIGN_BIT.
+	  (BIGNUM_SIGN): Renamed from RBIGNUM_SIGN.
+	  (BIGNUM_SET_SIGN): Renamed from RBIGNUM_SET_SIGN.
+	  (BIGNUM_POSITIVE_P): Renamed from RBIGNUM_POSITIVE_P.
+	  (BIGNUM_NEGATIVE_P): Renamed from RBIGNUM_NEGATIVE_P.
+	  (BIGNUM_EMBED_FLAG): Renamed from RBIGNUM_EMBED_FLAG.
+	  (BIGNUM_EMBED_LEN_MASK): Renamed from RBIGNUM_EMBED_LEN_MASK.
+	  (BIGNUM_EMBED_LEN_SHIFT): Renamed from RBIGNUM_EMBED_LEN_SHIFT.
+	  (BIGNUM_LEN): Renamed from RBIGNUM_LEN.
+	  (RBIGNUM_DIGITS): Renamed from RBIGNUM_DIGITS.
+	  (BIGNUM_LENINT): Renamed from RBIGNUM_LENINT.
+
+	* bignum.c: Follow the above change.
+
+	* gc.c: Ditto.
+
+	* marshal.c: Ditto.
+
+	* math.c: Ditto.
+
+	* numeric.c: Ditto.
+
+	* random.c: Ditto.
+
+	* rational.c: Ditto.
+
+	* sprintf.c: Ditto.
+
+	* ext/-test-/bignum/bigzero.c: Ditto.
+
+	* ext/-test-/bignum/intpack.c: Ditto.
+
+	* ext/bigdecimal/bigdecimal.c: Ditto.
+
+Sat Feb 15 20:48:49 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in (FILE_READEND): Don't detect it because it is not used.
+
+Sat Feb 15 13:22:28 2014  Eric Wong  <e@80x24.org>
+
+	* probes_helper.h (RUBY_DTRACE_HOOK): correct type for _id
+
+Sat Feb 15 11:47:47 2014  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_cmp): Avoid bignum allocation for comparison
+	  between bignum and fixnum.
+
+Sat Feb 15 10:55:12 2014  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* ext/-test-/win32/fd_setsize/depend: fix wrong dependencies.
+	  [ruby-dev:47253]
+
+Sat Feb 15 00:38:54 2014  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c: Enumerable#{min,min_by,max,max_by} extended to take an
+	  optional argument.
+	  (nmin_cmp): New function.
+	  (nmin_block_cmp): Ditto
+	  (nmin_filter): Ditto.
+	  (nmin_i): Ditto.
+	  (nmin_run): Ditto.
+	  (enum_min): Call nmin_run if the optional argument is given.
+	  (nmin_max): Ditto.
+	  (nmin_min_by): Ditto.
+	  (nmin_max_by): Ditto.
+
+	* range.c: Range#{min,max} extended to take an optional argument.
+	  (range_min): Call range_first if the optional argument is given.
+	  (range_max): Call rb_call_super if the optional argument is given.
+
+	[ruby-core:57111] [Feature #8887]
+
+Sat Feb 15 00:27:46 2014  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/ruby.h,
+	  internal.h,
+	  ext/-test-/bignum/bigzero.c: Hide a Bignum definition.
+	  [ruby-core:42891] [Feature #6083]
+
+Sat Feb 15 00:13:14 2014  Tanaka Akira  <akr@fsij.org>
+
+	* include/ruby/intern.h,
+	  include/ruby/io.h,
+	  include/ruby/ruby.h,
+	  include/ruby/win32.h,
+	  include/ruby/backward/rubysig.h,
+	  bignum.c,
+	  gc.c,
+	  io.c,
+	  process.c,
+	  safe.c,
+	  struct.c,
+	  thread.c,
+	  ext/socket/rubysocket.h,
+	  ext/-test-/old_thread_select: Remove deprecated definitions
+	  [ruby-core:60581] [Feature #9502]
+
+Fri Feb 14 18:38:46 2014  Eric Wong  <e@80x24.org>
+
+	* string.c (rb_str_format_m): trade volatile for RB_GC_GUARD
+	  RB_GC_GUARD meaning is clear and has better code generation.
+	  [ruby-core:60688]
+
+Thu Feb 13 23:30:30 2014  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): should check ci->me->flag of
+	  a refining method in case the method is private.
+	  [ruby-core:60111] [Bug #9452]
+
+	* vm_method.c (make_method_entry_refined): set me->flag of a refined
+	  method entry to NOEX_PUBLIC in case the original method is private
+	  and it is refined as a public method.  The original flag is stored
+	  in me->def->body.orig_me, so it's OK to make a refined method
+	  entry public.  [ruby-core:60111] [Bug #9452]
+
+	* test/ruby/test_refinement.rb: related tests.
+
+Thu Feb 13 18:38:15 2014  Eric Wong  <e@80x24.org>
+
+	* re.c (rb_reg_raise): remove volatile
+	  Unnecessary since r41597
+
+Thu Feb 13 18:28:51 2014  Eric Wong  <e@80x24.org>
+
+	* re.c (rb_reg_regcomp): remove volatile
+	  Unnecessary since r13261
+
+Thu Feb 13 16:54:32 2014  Zachary Scott  <e@zzak.io>
+
+	* test/ruby/test_array.rb: Ensure flatten! is used for test_flatten
+	  Patch by @ksss [Fixes GH-530] https://github.com/ruby/ruby/pull/530
+
+Thu Feb 13 15:43:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (IDSET_ATTRSET_FOR_INTERN): fix off-by-one bug.
+
+	* parse.y (rb_enc_symname_type): junk ID succeeded by '=' is also
+	  attrset ID.  [ruby-core:60668] [Bug #8756]
+
+Thu Feb 13 11:06:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if pthread_setname_np is available.
+
+	* thread_pthread.c: pthread_setname_np is not available on old
+	  Darwins.  [ruby-core:60524] [Bug #9492]
+
+Thu Feb 13 00:56:59 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* configure.in: revert r44922. I should have used AC_CHECK_FUNCS()
+	  to just define a symbol if the function is available.
+
+Thu Feb 13 00:20:58 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* configure.in: use AC_CHECK_FUNC instead of AC_CHECK_FUNCS
+	  if available.
+
+Thu Feb 13 00:15:10 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* configure.in: fix to undefine HAVE_MEMMEM correctly if it is broken.
+
+Tue Feb 11 23:54:40 2014  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_cmp): Specialize a comparison to zero.
+
+	* ext/bigdecimal/bigdecimal.c (is_negative): Use rb_big_cmp instead of
+	  RBIGNUM_NEGATIVE_P.
+	  (BigMath_s_log): Ditto.
+
+Tue Feb 11 22:59:10 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/openssl/ossl_bn.c (ossl_bn_initialize): Use rb_integer_pack.
+	  Fix SEGV by OpenSSL::BN.new(1 << (2**34)).
+
+Tue Feb 11 17:00:38 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/tk/README.tcltklib: [DOC] Fix typo by @xta [Fixes GH-532]
+
+Sun Feb  9 13:59:29 2014  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Fix compilation error.
+	  https://bugs.ruby-lang.org/issues/8358#note-16
+
+Sun Feb  9 05:20:24 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (rb_cv_gnu_qsort_r): use compile error "conflicting
+	  types for 'qsort_r'" instead of AC_RUN_IFELSE.
+
+Sun Feb  9 04:07:34 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/yaml.rb: [DOC] Add links to syck and psych on github [Bug #9501]
+	  Based on a patch by Giorgos Tsiftsis
+
+Sun Feb  9 02:13:53 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (OBJ_TAINTABLE, OBJ_TAINT, OBJ_INFECT),
+	  marshal.c (r_entry0): all Numerics never be tainted now.
+	  [ruby-core:57346] [Bug #8945]
+
+Sat Feb  8 23:40:35 2014  Vit Ondruch  <vondruch@redhat.com>
+
+	* configure.in: add quoting brackets and append wildcard for the
+	  rest after target_cpu, to properly detect platform for SSE2
+	  instructions.  [ruby-core:60576] [Bug #8358]
+
+Sat Feb  8 21:44:07 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* configure.in: check qsort_r(3) and whether it is GNU version.
+	  BSD version has different prototype.
+
+	* util.h: use qsort_r() as ruby_qsort() if it is GNU version.
+
+	* util.c: define ruby_qsort() if needed.
+
+Sat Feb  8 16:34:36 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Message::MessageDecoder#get_labels):
+	  Make it iterative.
+
+Sat Feb  8 15:54:12 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, gc.h (rb_objspace_marked_object_p): added.
+	  This function *ONLY* works just after marking phase,
+	  before any sweeping.
+	  This function is highly depending current GC implementation
+	  and can be removed future version.
+
+Sat Feb  8 15:41:37 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: Don't set CLOEXEC flag explicitly.  (Ruby set it by
+	  default.)
+
+Sat Feb  8 15:27:02 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Message::MessageDecoder): Raise
+	  DecodeError if no data before the limit.
+	  Reported by Will Bryant.  [ruby-core:60557] [Bug #9498]
+
+Sat Feb  8 15:11:21 2014  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (SMALLBUF): Unused macro removed.
+
+Fri Feb  7 23:37:49 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb: Ignore name servers which cause EAFNOSUPPORT on
+	  socket creation.
+	  Reported by Bjoern Rennhak.  [ruby-core:60442] [Bug #9477]
+
+Fri Feb  7 21:58:48 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/open-uri.rb: [DOC] use lower case version of core classes, same
+	  as commit r44878, based on patch by Jonathan Jackson [Bug #9483]
+
+Fri Feb  7 21:54:53 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/ripper/lib/ripper/lexer.rb: [DOC] use lower case version of core
+	  classes when referring to return value, since we aren't directly
+	  talking about the class. Patch by Jonathan Jackson [Bug #9483]
+
+Fri Feb 7 05:28:38 2014  Eric Wong  <e@80x24.org>
+
+	* constant.h: reduce rb_const_entry_t size on 64-bit
+	  Patch by Adam Avilla [ruby-core:60542] [Feature #9496]
+
+Thu Feb  6 15:27:46 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (get_envparam_int): correct warning messages.
+
+	* gc.c (get_envparam_double): ditto.
+
+Thu Feb  6 15:17:30 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (get_envparam_int): don't accept a value equals to lowerbound
+	  (changed by last commit) because "" or "foo" (not a number) strings
+	  are parsed as 0. They should be rejected.
+
+	* gc.c (get_envparam_double): ditto.
+
+Thu Feb  6 09:00:35 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (ruby_gc_set_params): if RUBY_GC_OLDMALLOC_LIMIT is provided,
+	  then set objspace->rgengc.oldmalloc_increase_limit.
+	  Without this fix, the env variable RUBY_GC_OLDMALLOC_LIMIT
+	  does not work.
+
+	* gc.c (get_envparam_int): accept a value equals to lowerbound.
+
+	* gc.c (get_envparam_double): ditto.
+
+Thu Feb 6 08:23:28 2014  Eric Wong  <e@80x24.org>
+
+	* ext/thread/thread.c (rb_szqueue_max_set): use correct queue and
+	  limit wakeups.  [Bug #9343][ruby-core:60517]
+	* test/thread/test_queue.rb (test_sized_queue_assign_max):
+	  test for bug
+
+Thu Feb  6 07:18:01 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.2.2.  Complete history at:
+
+	  http://rubygems.rubyforge.org/rubygems-update/History_txt.html#label-2.2.2+%2F+2014-02-05
+
+	* test/rubygems:  ditto.
+
+Wed Feb  5 20:56:32 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (to_be_skipped_id): ignore anonymous attributes.
+
+	* pack.c (Init_pack): use anonymous ID so that associated objects
+	  do not appear in the packed result.
+
+	* parse.y (rb_make_internal_id): return an anonymous ID for
+	  internal use.
+
+Wed Feb  5 14:41:56 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vsnprintf.c: remove duplicated def of `UNINITIALIZED_VAR()'.
+
+Wed Feb  5 14:40:16 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace_dump.c (dump_object): use STR_SHARED_P()
+	  instead of removed STR_NOCAPA_P() macro.
+
+Wed Feb  5 13:51:33 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h, vm_core.h: move LIKELY/UNLIKELY/UNINITIALIZED_VAR()
+	  macros from vm_core.h to internal.h.
+
+	* string.c: remove dependency to "vm_core.h".
+
+	* common.mk: ditto.
+
+Wed Feb  5 13:29:01 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (rb_str_free): use FL_TEST(str, STR_SHARED) directly
+	  because str is not embed.
+
+	* string.c (str_replace): remove `FL_SET(str, STR_SHARED)' line
+	  because STR_SET_SHARED() set STR_SHARED.
+
+Wed Feb  5 13:18:08 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: remove macros STR_NOCAPA and STR_NOCAPA_P().
+
+	* string.c (rb_str_resize): remove `STR_SET_NOEMBED(str)' because
+	  str_make_independent_expand() set NOEMBED flag.
+
+	* string.c (rb_str_resize): remove `STR_NOCAPA_P(str)' check because
+	  `str' is independent (not shared).
+
+Wed Feb  5 12:54:25 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: refactoring, especially about string flags.
+
+	* string.c (STR_UNSET_NOCAPA): removed.
+	  Use FL_UNSET() with STR_SHARED.
+
+	* string.c (rb_str_capacity): check STR_SHARED directly
+	  because it is not a embed string.
+
+	* string.c (rb_str_modify_expand): ditto.
+
+	* string.c (rb_str_shared_replace): use STR_SET_SHARED().
+
+	* string.c (str_make_independent_expand): remove STR_UNSET_NOCAPA()
+	  because `str' is not shared string.
+
+Wed Feb  5 12:11:04 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (RESIZE_CAPA): should not resize shared string.
+
+Wed Feb  5 11:46:42 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_children): STR_ASSOC is no longer available.
+	  Reported by @nagachika.
+	  http://d.hatena.ne.jp/nagachika/20140204
+
+Wed Feb  5 11:27:22 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c (rb_str_new_frozen): refactoring code.
+	  * Move code from str_new_frozen_with_klass() (and remove it)
+	  * `aux.shared' should not be 0 for STR_SHARED strings.
+
+Wed Feb  5 04:23:41 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: New release of psych.
+	* ext/psych/psych.gemspec: ditto
+
+Wed Feb  5 04:16:41 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/yaml/emitter.c: merge libyaml 0.1.5
+	* ext/psych/yaml/loader.c: ditto
+	* ext/psych/yaml/parser.c: ditto
+	* ext/psych/yaml/reader.c: ditto
+	* ext/psych/yaml/scanner.c: ditto
+	* ext/psych/yaml/writer.c: ditto
+	* ext/psych/yaml/yaml_private.h: ditto
+
+Tue Feb  4 19:10:29 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: use long allocator names instead of numbered
+	  allocator names.
+	  * rb_str_new2 -> rb_str_new_cstr
+	  * rb_str_new4 -> rb_str_new_frozen
+	  * rb_str_new5 -> rb_str_new_with_class
+	  * str_new3 -> str_new_shared
+	  * str_new4 -> str_new_frozen_with_klass
+
+Tue Feb  4 17:20:03 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): return the filename with actual cases on
+	  the filesystem if it is case-insensitive.  [ruby-core:42469]
+	  [Feature #5994]
+
+Tue Feb  4 16:16:58 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: use STR_SHARED instead of ELTS_SHARED.
+	  (same value, but more clear meaning)
+
+Tue Feb  4 16:09:14 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: remove STR_ASSOC related code.
+	  By r44804, string objects can not have STR_ASSOC flag.
+
+	* internal.h: ditto.
+
+	* ext/objspace/objspace_dump.c (dump_object): ditto.
+
+Tue Feb  4 14:07:20 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (str_associate, str_associated): keep associated objects
+	  in an instance variables, instead of in the internal structure.
+
+	* string.c (rb_str_associate, rb_str_associated): deprecate.
+
+Tue Feb  4 12:55:31 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_modify_expand): enable capacity and disable
+	  association with packed objects when setting capa, so that
+	  pack("p") string fails to unpack properly after modified.
+
+Tue Feb  4 12:45:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_make_pattern): all alphabets are magic characters on
+	  case-insensitive filesystems.  [ruby-core:42469] [Feature #5994]
+
+Tue Feb  4 09:47:57 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems 2.2.2 prerelease to check fixes to
+	  CI.
+	* test/rubygems:  ditto.
+
+Mon Feb  3 12:04:47 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* error.c: [DOC] Exception#cause may return nil. [ci skip]
+
+Sun Feb 2 05:48:42 2014  Eric Wong <e@80x24.org>
+
+	* io.c (rb_io_syswrite): add RB_GC_GUARD
+	  [Bug #9472][ruby-core:60407]
+
+Sat Feb  1 15:09:16 2014  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_typedesc2val): add VT_RECORD case.
+
+Sat Feb  1 06:38:51 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/drb/drb.rb: [DOC] Add note about start_service for each process
+	  Based on a patch by @rosenfeld [Fixes GH-514] [ci skip]
+	  https://github.com/ruby/ruby/pull/514
+
+Sat Feb  1 06:30:20 2014  Zachary Scott  <e@zzak.io>
+
+	* error.c: [DOC] Document Exception#cause by @jasonrclark [ci skip]
+	  [Fixes GH-519] https://github.com/ruby/ruby/pull/519
+
+Sat Feb  1 06:10:49 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/securerandom.rb: [DOC] Add note on require for examples
+	  Based on a patch by @schneems [Fixes GH-518] [ci skip]
+	  https://github.com/ruby/ruby/pull/518
+
+Sat Feb  1 06:04:56 2014  Zachary Scott  <e@zzak.io>
+
+	* numeric.c: [DOC] Fix typo in example for #step [ci skip]
+	  Patch by @ksss [Fixes GH-522] https://github.com/ruby/ruby/pull/522
+
+Fri Jan 31 17:01:47 2014  Eric Wong <e@80x24.org>
+
+	* ext/socket/init.c (rsock_socket0): split out SOCK_CLOEXEC version
+	* ext/socket/socket.c (rsock_socketpair0): ditto
+	  [ruby-core:60377]
+
+Fri Jan 31 03:48:40 2014  Eric Wong <e@80x24.org>
+
+	* benchmark/driver: avoid large alloc in driver process
+	  [ruby-core:59869] [Bug #9430]
+
+Thu Jan 30 14:45:49 2014  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* configure.in: use $@ instead of $(.TARGET) because .TARGET is not
+	  supported by GNU make.
+
+Thu Jan 30 08:26:21 2014  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* ext/fiddle/closure.c: use sizeof(*pcl) for correct sizeof value.
+	  [ruby-core:57599] [Bug #8978].
+
+Wed Jan 29 20:08:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (nogvl_copy_stream_sendfile): check socket on other than
+	  linux, as sendfile(2) on non-socket fd works only on linux.
+	  [Feature #9427]
+
+Wed Jan 29 18:09:48 2014  Eric Wong <e@80x24.org>
+
+	* io.c (nogvl_copy_stream_sendfile): remove socket check
+	  [ruby-core:59856][Feature #9427]
+
+Wed Jan 29 04:29:54 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/openssl/ossl_ssl.c: pass read_nonblock options to underlying IO
+	  when SSL session has not been started.
+
+	* test/openssl/test_ssl.rb: test for change.
+
+Wed Jan 29 03:49:36 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/fiddle/closure.c: use sizeof(*pcl) for correct sizeof value.
+	  [ruby-core:57599] [Bug #8978].  Thanks mame!
+
+Wed Jan 29 03:36:42 2014  Eric Wong <e@80x24.org>
+
+	* doc/contributing.rdoc: allow/encourage other git hosts
+	  [ruby-core:59807][misc #9421]
+
+Tue Jan 28 23:36:01 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: Avoid redundant fcntl/fstat syscalls for cloexec
+	  sockets.
+	  Patch by Eric Wong.  [ruby-core:59429] [Feature #9330]
+
+Tue Jan 28 20:51:07 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (READ_FROM_CHILD): Apply the last hunk of
+	  0001-process.c-avoid-EINTR-from-Process.spawn.patch written by
+	  Eric Wong in [Bug #8770].
+
+Tue Jan 28 16:31:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack, ruby_stack_overflowed_p):
+	  place get_stack above others to get stack boundary information.
+	  [ruby-core:60113] [Bug #9454]
+
+Tue Jan 28 15:27:36 2014  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c: rlimit is only available on Linux.
+	  At least r44712 breaks FreeBSD.
+	  [ruby-core:60113] [Bug #9454]
+
+Tue Jan 28 15:17:59 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/set.rb: [DOC] Add examples for Set#intersect? and Set#disjoint?
+	  Patch by xavier nayrac [Bug #9331] [ci skip]
+
+Tue Jan 28 15:12:22 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/zlib/zlib.c (rb_zlib_adler32): [DOC] Add example for adler32
+	  Patch by Vajrasky Kok [Bug #9307] [ci skip]
+
+Tue Jan 28 08:56:00 2014  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* compile.c (iseq_build_from_ary_body): Use :blockptr instead of :block
+	  as hash key when loading serialized instruction sequences from arrays.
+	  [Bug #9455] [ruby-core:60146]
+
+Mon Jan 27 21:52:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c: get current main thread stack size, which may
+	  be expanded than allocated size at initialization, by rlimit().
+	  [ruby-core:60113] [Bug #9454]
+
+Sat Jan 25 22:17:02 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* README.ja.md, README.md: update the controller address of
+	  mailing lists.
+
+Sat Jan 25 14:50:42 2014  Eric Wong  <normalperson@yhbt.net>
+
+	* process.c (send_child_error): retry write on EINTR to fix
+	  occasional Errno::EINTR from Process.spawn.
+
+	* process.c (recv_child_error): retry read on EINTR to fix
+	  occasional Errno::EINTR from Process.spawn.
+
+Sat Jan 25 14:21:06 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): result of assignment should be
+	  its rhs instead of returned value from a method.
+	  [ruby-core:60071] [Bug #9448]
+
+Sat Jan 25 11:16:19 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_extract_keywords): treat nil keyword_hash same as 0,
+	  for the case rb_scan_args returns nil if no keyword hash.
+
+Fri Jan 24 15:13:20 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/racc/rdoc/grammar.en.rdoc: [DOC] Correct grammar and typos
+	  Patch by Giorgos Tsiftsis [Bug #9429] [ci skip]
+
+Thu Jan 23 20:20:17 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/envutil.rb: try to wait a bit (0.1sec) when ruby process
+	  exits by signals because some SEGV tests fail because of not enough
+	  error output.
+
+Thu Jan 23 20:06:27 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_settracefunc.rb: check the target thread.
+
+Thu Jan 23 19:59:16 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_settracefunc.rb: check the target thread.
+
+Thu Jan 23 14:26:44 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/fileutils.rb: [DOC] Fix typo in options_of() example [Bug #9392]
+	  Patch by Giorgos Tsiftsis
+
+Thu Jan 23 13:56:16 2014  Zachary Scott  <e@zzak.io>
+
+	* README -> README.md: [DOC] Format README with Markdown [Bug #9255]
+	* README.ja -> README.ja.md: ditto
+
+Wed Jan 22 15:59:39 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/thread/thread.c (Init_thread): ConditionVariable and Queue
+	  are not able to copy.  [ruby-core:59961] [Bug #9440]
+
+Tue Jan 21 20:14:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): fix for platforms
+	  where PTHREAD_STACK_MIN is a dynamic value and not a compile-time
+	  constant.  [ruby-dev:47911] [Bug #9436]
+
+Tue Jan 21 17:55:09 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/uri/common.rb: [DOC] Use static w3.org uri [ci skip]
+	  Patch by @ykzts [Fix GH-484] https://github.com/ruby/ruby/pull/484
+
+Tue Jan 21 16:43:22 2014  Zachary Scott  <e@zzak.io>
+
+	* enum.c: [DOC] Add simple example of Enumerable#zip [ci skip]
+	  Patch by @nruth on documenting-ruby/ruby#22
+	  https://github.com/documenting-ruby/ruby/pull/22
+
+Tue Jan 21 16:26:44 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): expand timer
+	  thread stack size to get rid of segfault on FreeBSD/powerpc64.
+	  based on the patch by Steve Wills at [ruby-core:59923].
+	  [ruby-core:56590] [Bug #8783]
+
+Tue Jan 21 04:31:23 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c: Use "int" for IP_MULTICAST_LOOP and
+	  IP_MULTICAST_TTL on Mac OS X and Windows.
+
+Tue Jan 21 00:39:15 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c: Use "byte" as default argument for
+	  IP_MULTICAST_LOOP and IP_MULTICAST_TTL socket option to follow
+	  the original multicast implementation.
+
+Mon Jan 20 20:20:27 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c: Use preprocessor macros to avoid repeated
+	  conditionals.
+
+Mon Jan 20 13:55:03 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/rubygems/version.rb: [DOC] Use gender-neutral pronouns [ci skip]
+	* lib/rubygems/security.rb: ditto
+
+Sun Jan 19 06:38:48 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* compar.c (cmp_equal): warn for this release and still rescue
+	  standard exceptions for a nicer transition. See #7688.
+	  Partly reverts r44502.
+
+	* test/ruby/test_comparable.rb: adapt assertion to match new behavior.
+
+Sun Jan 19 06:27:18 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/ruby/test_comparable.rb: specify behavior for the different
+	  kind of exceptions rescued (or not) by Comparable#==.
+
+Sat Jan 18 23:12:19 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket: Avoid unnecessary ppoll/select on Linux.
+	  Patch by Eric Wong.  [ruby-core:57950] [Bug #9039]
+
+Sat Jan 18 22:57:44 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Resource::TXT#data): Return concatenated
+	  string.
+	  Patch by Ryan Brunner.  [ruby-core:58220] [Bug #9093]
+
+Sat Jan 18 22:35:15 2014  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_update_max_fd): Return immediately if the given fd is small
+	  enough.
+
+Sat Jan 18 22:25:53 2014  Tanaka Akira  <akr@fsij.org>
+
+	* io.c: Test O_CLOEXEC only once.
+	  Patch by Eric Wong.  [ruby-core:59419] [Feature #9328]
+
+Sat Jan 18 21:24:49 2014  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/option.c: IP_MULTICAST_LOOP and IP_MULTICAST_TTL socket
+	  option takes a byte on OpenBSD.
+	  Fixed by Jeremy Evans.  [ruby-core:59496] [Bug #9350]
+
+Sat Jan 18 21:19:04 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: Make proxy disabling working again.
+	  Fixed by Christophe Philemotte.   [ruby-core:59650] [Bug #9385]
+
+Fri Jan 17 20:05:02 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/delegate.rb (Delegator): keep source information methods
+	  which start and end with '__'.  [ruby-core:59718] [Bug #9403]
+
+Fri Jan 17 17:58:04 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_mod_s_constants): return its own constants for other
+	  than Module itself.  [ruby-core:59763] [Bug #9413]
+
+Tue Jan 16 00:17:00 2014  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.gemspec: bigdecimal version 1.2.5.
+
+Wed Jan 15 20:30:31 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* io.c (io_binwrite): use writev(2) to avoid double write if available.
+
+	* configure.in: check writev(2)
+
+Wed Jan 15 14:04:33 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (get_encoding): respect BOM on pseudo encodings.
+	  [ruby-dev:47895] [Bug #9415]
+
+Wed Jan 15 14:03:47 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (get_actual_encoding): get actual encoding according to
+	  the BOM if exists.
+
+	* string.c (rb_str_inspect): use according encoding, instead of
+	  pseudo encodings, UTF-{16,32}.  [ruby-core:59757] [Bug #8940]
+
+Tue Jan 14 21:07:22 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* ext/thread/thread.c (rb_szqueue_clear): notify SZQUEUE_WAITERS
+	  on SizedQueue#clear. [ruby-core:59462] [Bug #9342]
+
+	* test/thread/test_queue.rb: add test. the patch is from
+	  Justin Collins.
+
+Tue Jan 14 15:58:43 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (CLASS_NAME): macro to wrap
+	  depending on PRIsVALUE for 1.9.  [Backport #9406]
+
+	* ext/bigdecimal/bigdecimal.c (DECIMAL_SIZE_OF_BITS): fallback
+	  definition for 2.1 or older.  [ruby-core:59750] [Backport #9406]
+
+Tue Jan 14 11:28:44 2014  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* vm_exec.c (cfp): Fixes a SEGV issue in r44554.
+	  r11 can be broken by subroutine and sometimes causes SEGV at
+	  runtime.  Use r13 instead.
+
+Tue Jan 14 02:20:00 2014  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_divide): Add an additional
+	  digit for the quotient to be compatible with bigdecimal 1.2.1 and
+	  the former.  [ruby-core:59365] [#9316] [#9305]
+
+	* test/bigdecimal/test_bigdecimal.rb: tests for the above change.
+
+	* ext/bigdecimal/bigdecimal.gemspec: bigdecimal version 1.2.4.
+
+Mon Jan 13 14:55:31 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/xmlrpc/client.rb: [DOC] Remove note about SSL package on RAA
+	  Since RAA has been deprecated, and the SSL package has been replaced
+	  with net/https this statement is entirely false and should be
+	  deleted. [Bug #9152]
+
+Mon Jan 13 14:47:07 2014  Zachary Scott  <e@zzak.io>
+
+	* lib/net/smtp.rb: [DOC] Remove dead link to RAA by Giorgos Tsiftsis
+	  Fixes the following bugs: [Bug #9152] [Bug #9268] [Bug #9394]
+	* lib/open-uri.rb: ditto
+
+Mon Jan 13 14:25:55 2014  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_pkey_dh.c: Fixed typo by Sandor Szuecs [Bug #9243]
+
+Mon Jan 13 12:03:00 2014  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* ext/thread/thread.c (rb_szqueue_push): check GET_SZQUEUE_WAITERS
+	  instead of GET_QUEUE_WAITERS to prevent deadlock. Patch by Eric Wong.
+	  [Bug #9302] [ruby-core:59324]
+
+	* test/thread/test_queue.rb: add test
+
+Sun Jan 12 16:41:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseq_load): keep type_map to get rid of memory leak.
+	  based on a patch by Eric Wong at [ruby-core:59699].  [Bug #9399]
+
+Sun Jan 12 09:21:35 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/util.h (DECIMAL_SIZE_OF_BITS): a preprocessor
+	  constant macro to approximate decimal representation size of n-bits
+	  integer.
+
+	* iseq.c (register_label): use DECIMAL_SIZE_OF_BITS for better
+	  approximation.
+
+	* ext/bigdecimal/bigdecimal.c (BigMath_s_log): ditto.
+
+	* common.mk (iseq.o), ext/bigdecimal/depend (bigdecimal.o): add
+	  dependency to ruby/util.h for DECIMAL_SIZE_OF_BITS.
+
+Fri Jan 10 16:27:20 2014  Yuki Yugui Sonoda  <yugui@google.com>
+
+	* vm_exec.c (cfp): Avoid generating invalid binary for
+	  NativeClient.
+	  r15 on x86_64 is reserved by NativeClient.  So r15 to cfp used to
+	  generate invalid binary under some combinations of compiler
+	  optimization flags.
+
+Fri Jan 10 18:01:41 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_super_method): allow bound method from a
+	  module, yet another method transplanting.
+
+Fri Jan 10 13:34:04 2014  Aman Gupta <ruby@tmm1.net>
+
+	* insns.def (opt_aref_with): new instruction to optimize Hash#[],
+	  removing any allocation overhead when used with a string literal
+	  key. Patch by normalperson (Eric Wong). [ruby-core:59640] [Bug #9382]
+	* insns.def (opt_aset_with): new instruction to optimize Hash#[]=
+	* compile.c (iseq_compile_each): compiler shortcuts for new
+	  instructions
+	* hash.c (static VALUE rb_hash_compare_by_id_p): fix documentation for
+	  Hash#compare_by_identity to reflect frozen string sharing
+	* test/ruby/test_hash.rb (class TestHash): test for new behavior
+
+Fri Jan 10 06:23:21 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* range.c (Range#size): [DOC] improve description and add examples.
+	  Patch by @skade. [Fixes GH-501]
+
+Fri Jan 10 00:47:52 2014  Josef Stribny  <strzibny@gmail.com>
+
+	* ext/tk/extconf.rb: fix to pass arrays instead of strings to
+	  libpathflag.  patch at [ruby-core:59665].  [Bug #9386]
+
+Thu Jan  9 20:49:22 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: -mstackrealign is necessary for -msse2 working.
+	  [ruby-core:54716] [Bug #8349]
+
+	* configure.in: use SSE2 instructions to drop unexpected precisions on
+	  other than mingw.  [ruby-core:59472] [Bug #8358]
+
+Thu Jan  9 20:31:10 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (rb_objid_hash): should return `long'.  brushup r44534.
+
+	* object.c (rb_obj_hash): follow above change.
+
+Thu Jan  9 19:12:37 2014  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_pop_cfunc_frame): added.  It cares c_return event.
+	  The patch base by drkaes (Stefan Kaes).
+	  [Bug #9321]
+
+	* variable.c (rb_mod_const_missing): use rb_vm_pop_cfunc_frame()
+	  instead of rb_frame_pop().
+
+	* vm_eval.c (raise_method_missing): ditto.
+
+	* vm_eval.c (rb_iterate): ditto.
+
+	* internal.h (rb_vm_pop_cfunc_frame): add decl.
+
+	* test/ruby/test_settracefunc.rb: add tests.
+	  provided by drkaes (Stefan Kaes).
+
+	* vm.c, eval.c, include/ruby/intern.h (rb_frame_pop):
+	  move definition of rb_frame_pop() and deprecate it.
+	  It doesn't care about `return' events.
+
+Thu Jan  9 17:40:28 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (rb_any_hash): should treat the return value of rb_objid_hash()
+	  as `long', because ruby assumes the hash value of the object id of
+	  an object is `long'.
+	  this fixes test failures on mswin64 introduced at r44525.
+
+Thu Jan  9 09:55:20 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: dumping strings with
+	  quotes should not have changed.  [ruby-core:59316] [Bug #9300]
+
+	* ext/psych/lib/psych.rb: fixed missing require.
+
+	* test/psych/test_string.rb: test
+
+Thu Jan  9 09:51:00 2014  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: anonymous structs
+	  should be able to roundtrip.  Thanks @splattael!
+
+	* test/psych/test_object_references.rb: test for change
+
+Wed Jan  8 22:53:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_search_super_method): when super called in a
+	  bound UnboundMethod generated from a module, no superclass is
+	  found since the current defined class is the module, then call
+	  method_missing in that case.  [ruby-core:59619] [Bug #9377]
+
+Wed Jan  8 15:55:21 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_objid_hash): return hash value from object ID with a
+	  salt, extract from rb_any_hash().
+
+	* object.c (rb_obj_hash): return same value as rb_any_hash().
+	  fix r44125.  [ruby-core:59638] [Bug #9381]
+
+Wed Jan  8 13:12:41 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout::ExitException.catch): pass arguments
+	  for new instance.
+
+	* lib/timeout.rb (Timeout::ExitException#exception): fallback to
+	  Timeout::Error if couldn't throw.  [ruby-dev:47872] [Bug #9380]
+
+	* lib/timeout.rb (Timeout#timeout): initialize ExitException with
+	  message for the fallback case.
+
+Tue Jan  7 12:43:06 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout#timeout): should not rescue ordinarily
+	  raised ExitException, which should not be thrown.
+
+	* lib/timeout.rb (Timeout::ExitException.catch): set @thread only if
+	  it ought to be caught.
+
+	* lib/timeout.rb (Timeout#timeout): when a custom exception is given,
+	  no instance is needed to be caught, so defer creating new instance
+	  until it is raised.  [ruby-core:59511] [Bug #9354]
+
+Tue Jan  7 10:16:02 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems:  Update to RubyGems master 21e409d / RubyGems 2.2.1.
+
+	  See http://rubygems.rubyforge.org/rubygems-update/History_txt.html
+	  for a list of bug fixes.
+
+	* test/rubygems:  ditto.
+
+Tue Jan  7 10:10:46 2014  Eric Wong <e@80x24.org>
+
+	* ext/json/generator/depend: add build dependencies for json extension
+	  [Bug #9374] [ruby-core:59609]
+	* ext/json/parser/depend: ditto
+
+Tue Jan  7 04:35:46 2014  Aman Gupta <ruby@tmm1.net>
+
+	* array.c (ary_add_hash): Fix consistency issue between Array#uniq and
+	  Array#uniq! [Bug #9340] [ruby-core:59457]
+	* test/ruby/test_array.rb (class TestArray): regression test for above.
+
+Mon Jan  6 21:28:48 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* .gitignore: ignore *-fake.rb generated even when CROSS_COMPILING = no
+	  since r42862.
+
+Sun Jan  5 20:14:14 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* compar.c (cmp_equal): remove error hiding in Comparable#==.
+	  Comparable#== no longer rescues exceptions silently.
+	  This was the cause of quite a couple bugs. See #7688. [EXPERIMENTAL]
+
+	* test/ruby/test_comparable.rb: adapt assertion to match new behavior.
+
+	* lib/rdoc/method_attr.rb: fix bugs discovered by this change.
+
+	* test/rdoc/test_rdoc_normal_class.rb: fix bugs in tests.
+
+Sat Jan  4 22:44:00 2014  Charlie Somerville  <charliesome@ruby-lang.org>
+
+	* struct.c (rb_struct_set): return assigned value from setter method
+	  rather than struct object. [Bug #9353] [ruby-core:59509]
+
+	* test/ruby/test_struct.rb (test_setter_method_returns_value): add test
+
+Sat Jan  4 21:44:31 2014  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_gc.rb (TestGc#test_latest_gc_info): use
+	  GC.stat(:key) instead of GC.stat.
+
+Sat Jan  4 19:15:29 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (rsock_syserr_fail_host_port): use format flags,
+	  '+' to inspect, ' ' to quote unprintables.
+
+	* ext/socket/socket.c (rsock_syserr_fail_path): ditto.
+
+	* ext/socket/socket.c (rsock_syserr_fail_raddrinfo): ditto.
+
+	* ext/socket/socket.c (rsock_syserr_fail_host_port): add errno
+	  argument version and use rb_syserr_fail_str() instead of
+	  rb_sys_fail_str() with restoring errno.
+
+	* ext/socket/socket.c (rsock_syserr_fail_path): ditto, and
+	  rb_syserr_fail().
+
+	* ext/socket/socket.c (rsock_sys_fail_sockaddr): ditto, use
+	  rsock_syserr_fail_raddrinfo().
+
+	* ext/socket/socket.c (rsock_sys_fail_raddrinfo): ditto.
+
+	* ext/socket/socket.c (setup_domain_and_type): ditto.
+
+Sat Jan  4 17:18:58 2014  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm.c (RubyVM::OPTS): get rid of a garbage character.
+
+Sat Jan  4 10:17:54 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rinda/ring.rb (Rinda::RingFinger#make_socket):  Use
+	  ipv4_multicast_ttl option for portability.
+
+Sat Jan  4 10:15:47 2014  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rinda/ring.rb (Rinda::RingFinger#make_socket):  Use
+	  ipv4_multicast_loop option for portability.  Patch by Jeremy Evans.
+	  [ruby-trunk - Bug #9351]
+
+Fri Jan  3 19:09:00 2014  Eric Wong  <normalperson@yhbt.net>
+
+	* ext/socket/socket.c (rsock_sys_fail_host_port): save and restore errno
+	  before calling rb_sys_fail_str to prevent [BUG] errno == 0.
+	  Patch by Eric Wong. [ruby-core:59498] [Bug #9352]
+
+	* ext/socket/socket.c (rsock_sys_fail_path): ditto
+	* ext/socket/socket.c (rsock_sys_fail_sockaddr): ditto
+	* ext/socket/socket.c (rsock_sys_fail_raddrinfo): ditto
+	* ext/socket/socket.c (rsock_sys_fail_raddrinfo_or_sockaddr): ditto
+
+Fri Jan  3 10:43:57 2014  Aman Gupta <ruby@tmm1.net>
+
+	* test/net/imap/cacert.pem: generate new CA cert, since the last one
+	  expired. [Bug #9341] [ruby-core:59459]
+	* test/net/imap/server.crt: new server cert signed with updated CA.
+	* test/net/imap/Makefile: add `make regen_certs` to automate this
+	  process.
+
+Fri Jan  3 00:09:54 2014  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/bigdecimal: update class method call style from :: to .
+	  in documentation and usage.
+
+	* ext/bigdecimal/lib/bigdecimal/math.rb: [DOC] fix examples values.
+	  Computations were made using ruby 2.0.0p247 to ensure
+	  no effect of the recent BigDecimal bug.
+
+	* ext/bigdecimal/sample/nlsolve.rb: fix indent.
+
+Thu Jan  2 16:07:21 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* io.c (io_fwrite): freeze converted str.
+
+Thu Jan  2 04:15:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_longjmp): remove an extra modifier from the forward
+	  declaration to match the actual definition.  [ruby-core:59451]
+	  [Bug #9338]
+
+Thu Jan  2 01:23:30 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* vm_eval.c (method_missing): use ALLOCV_N() instead of
+	  ALLOCA_N() and rb_ary_tmp_new().
+
+Thu Jan  2 00:53:16 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* array.c (rb_ary_zip): use ALLOCV_N() instead of ALLOCA_N().
+
+Thu Jan  2 00:04:29 2014  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* hash.c (rb_hash_keys): make rb_hash_keys() static.
+	  it is no longer used from array.c since r43969.
+	  the patch is from normalperson (Eric Wong).
+	  [ruby-core:59449] [Feature #9336]
+
+	* internal.h: remove definition of rb_hash_keys().
+
+Wed Jan  1 18:19:35 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: reset LDFLAGS and DLDFLAGS for opt-dir again after
+	  LIBPATHFLAG and RPATHFLAG are set.  [ruby-dev:47868] [Bug #9317]
+
+Wed Jan  1 11:12:29 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: use SSE2 instructions for drop unexpected
+	  precisions.  [ruby-core:54738] [Bug #8358]
+
+Tue Dec 31 23:49:07 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (rb_f_raise): add cause: optional keyword argument.
+	  [ruby-core:58610] [Feature #8257] [EXPERIMENTAL]
+
+Tue Dec 31 21:44:17 2013  Akio Tajima <artonx@yahoo.co.jp>
+
+	* win32/Makefile.sub: remove HAVE_FSEEKO because fseeko removed from win32/win32.c
+	  Fixed [Bug #9333].
+
+Tue Dec 31 21:02:27 2013  Masaki Matsushita  <glass.saga@gmail.com>
+
+	* io.c (io_fwrite): allocate frozen str only when str is not converted.
+
+Tue Dec 31 15:44:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (RbConfig): expand RUBY_SO_NAME for extensions
+	  backward compatibility.  [ruby-core:59426] [Bug #9329]
+
+Mon Dec 30 23:33:07 2013  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* variable.c: [DOC] adding extra example in docs.
+	  patched by Steve Klabnik. [Bug #9210]
+
+Mon Dec 30 18:34:18 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (must_encindex, rb_enc_from_index, rb_obj_encoding): mask
+	  encoding index and ignore dummy flags.  [ruby-core:59354] [Bug #9314]
+
+Mon Dec 30 16:11:52 2013  WATANABE Hirofumi  <eban@ruby-lang.org>
+
+	* tool/make-snapshot: needs CXXFLAGS.  [ruby-core:59393][Bug #9320]
+
+Sun Dec 29 18:36:54 2013  Shota Fukumori  <her@sorah.jp>
+
+	* lib/mkmf.rb (configuration): Make CXXFLAGS customizable.
+	  Patch by Kohei Suzuki (eagletmt). [Fixes GH-492]
+
+Sun Dec 29 12:11:11 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew_from_me): keep iclass as-is, to make inheritance
+	  chain consistent.  [ruby-core:59358] [Bug #9315]
+
+	* proc.c (method_owner): return the original defined_class from
+	  prepended iclass, instead.
+
+Sun Dec 29 08:47:24 2013  Lucas Allan Amorim  <lucas.allan@gmail.com>
+
+	* test/ruby/test_sprintf.rb (test_hash): Added tests for sprintf
+	  with a hash as parameter.  [Fixes GH-491]
+
+Sun Dec 29 07:27:51 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* compar.c (cmp_eq_recursive): Fix the return value, the value for
+	  failed #<=> should be nil. It was raising a NoMethodError for
+	  the test case TestComparable#test_no_cmp (undefined method `>'
+	  for false:FalseClass). Yet one more reason for #7688.
+
+Sat Dec 28 22:21:59 2013  Benoit Daloze  <eregontp@gmail.com>
+
+	* object.c (Kernel#<=>) surround Comparable operators with <code> tags.
+	  The #== method was hidden in ri/rdoc's output and was highlighting
+	  the line instead.
+
+Sat Dec 28 17:24:00 2013  DV Suresh  <e@dvsuresh.me>
+
+	* benchmark/bm_so_meteor_contest.rb: [DOC] Fix a few typos
+	* ext/fiddle/lib/fiddle/import.rb: ditto
+	* ext/psych/lib/psych.rb: ditto
+	* ext/psych/lib/psych/nodes/sequence.rb: ditto
+	* ext/tk/lib/multi-tk.rb: ditto
+	* ext/tk/lib/tcltk.rb: ditto
+
+Sat Dec 28 00:42:37 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (local_push_gen, local_pop_gen): save cmdarg_stack to
+	  isolate command argument state from outer scope.
+	  [ruby-core:59342] [Bug #9308]
+
+Fri Dec 27 13:25:03 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/{setup.mak,Makefile.sub}: update fake.rb like
+	  template/fake.rb.in.
+
+Thu Dec 26 16:10:41 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (fake.rb): should depend on version.h because
+	  if RUBY_VERSION is updated, fake.rb need to say the new version
+	  to avoid install error in rbconfig.rb.
+
+Thu Dec 26 14:25:03 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (HASH_REJECT_COPY_EXTRA_STATES): turn off the old
+	  behavior, copying extra states by accident.
+
+Thu Dec 26 10:49:14 2013  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* version.h (RUBY_VERSION): 2.2.0 development has started.
+
+Thu Dec 26 10:27:53 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/merger.rb (tag): support 2.1.1 semi-automatic tagging and 2.2.0
+	  explicit tagging.
+
+Thu Dec 26 06:35:25 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c: Having any mandatory keyword argument increases min arity
+	  [#9299]
+
+Thu Dec 26 06:27:08 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* proc.c: Having optional keyword arguments makes maximum arity +1,
+	  not unlimited [#8072]
+
+Thu Dec 26 01:09:57 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/release.sh: make symbolic links.
+
+Thu Dec 26 00:45:33 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/make-snapshot: support new version scheme.
+
+Wed Dec 25 22:44:14 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_set_arguments): set arg_keyword_check from
+	  nd_cflag, which is set by parser.  internal ID is used for
+	  unnamed keyword rest argument, which should be separated from no
+	  keyword check.
+
+	* iseq.c (rb_iseq_parameters): if no keyword check, keyword rest is
+	  present.
+
+	* parse.y (new_args_tail_gen): set keywords check to nd_cflag, which
+	  equals to that keyword rest is not present.
+
+Wed Dec 25 22:32:19 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/abbrev.rb: [DOC] rdoc format patch by Giorgos Tsiftsis [Bug #9146]
+
+Wed Dec 25 20:30:10 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_parameters): push argument type symbol only for
+	  unnamed rest keywords argument.
+
+Wed Dec 25 20:28:48 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_iseq_min_max_arity): maximum argument is unlimited if
+	  having rest keywords argument.  [ruby-core:53298] [Bug #8072]
+
+Wed Dec 25 18:29:22 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (argument_error): insert dummy frame to make
+	  a backtrace object intead of modify backtrace string array.
+	  [Bug #9295]
+
+	* test/ruby/test_backtrace.rb: add a test for this patch.
+	  fix test to compare a result of Exception#backtrace with
+	  a result of Exception#backtrace_locations.
+
+Wed Dec 25 13:00:54 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: let mingw do something black-magic, and check if
+	  _gmtime64_s() is available actually.
+
+	* win32/win32.c (gmtime_s, localtime_s): use _gmtime64_s() and
+	  _localtime64_s() if available, not depending on very confusing
+	  mingw variants macros.  based on the patch by phasis68 (Heesob
+	  Park) at [ruby-core:58764].  [ruby-core:58391] [Bug #9119]
+
+Wed Dec 25 12:33:41 2013  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* sample/trick2013/: added the award-winning entries of TRICK 2013.
+	  See https://github.com/tric/trick2013 for the contest outline.
+	  (Matz has approved the attachment.)
+
+Wed Dec 25 10:42:02 2013  Yamashita Yuu  <yamashita@geishatokyo.com>
+
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): Declare a constant
+	  `OP_MSIE_SSLV2_RSA_PADDING` only if the macro is defined.  The
+	  `SSL_OP_MSIE_SSLV2_RSA_PADDING` has been removed from latest
+	  snapshot of OpenSSL 1.0.1.  [Fixes GH-488]
+
+Wed Dec 25 01:03:00 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (bind_local_variables): allowing binding to list its
+	  local variables.  patch by Jack Danger Canty <jackdanger AT
+	  squareup.com> at [ruby-core:56543].  [Feature #8773]
+
+Tue Dec 24 23:20:38 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fileutils/fileasserts.rb (assert_ownership_user): new
+	  assertion for user ownership.
+
+	* test/fileutils/test_fileutils.rb (test_chown_error),
+	  (test_chown_without_permission, test_chown_with_root):
+	  based on the patch by vajrasky (Vajrasky Kok) at
+	  [ruby-core:59298].  [Feature #9292]
+
+Tue Dec 24 16:28:05 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_mod_define_method): consider visibility only if self
+	  in the caller is same as the receiver, otherwise make public as
+	  well as old behavior.  [ruby-core:57747] [Bug #9005]
+	  [ruby-core:58497] [Bug #9141]
+
+	* vm.c (rb_vm_cref_in_context): return ruby level cref if self is
+	  same.
+
+Tue Dec 24 14:13:14 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* README.EXT: add a refer to URL.
+
+Tue Dec 24 13:48:45 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* README.EXT: add a document about RGenGC.
+	  Reviewed by havenwood.
+	  [misc #8962]
+
+	* README.EXT.ja: ditto.
+
+Tue Dec 24 12:11:43 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h (RARRAY_ASET): try to avoid compiler warning.
+	  [Bug #9287]
+
+Tue Dec 24 05:04:56 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb (setup): should not call
+	  Process.groups for Windows.  get rid of many errors introduced by
+	  r44364.
+
+Mon Dec 23 18:37:16 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fileutils/fileasserts.rb (assert_ownership_group): new
+	  assertion for group ownership.
+
+	* test/fileutils/test_fileutils.rb (test_chown{,_verbose,_noop}):
+	  based on the patch by vajrasky (Vajrasky Kok) at
+	  [ruby-core:59281].  [Feature #9286]
+
+Mon Dec 23 15:53:45 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (HAS_EXTRA_STATES): warn extra states only when something
+	  differ.  [ruby-core:59254] [Bug #9275]
+
+Mon Dec 23 12:42:13 2013  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* array.c: Have to_h raise on elements that are not key-value pairs
+	  [#9239]
+
+	* enum.c: ditto
+
+Mon Dec 23 05:01:55 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/syntax/methods.rdoc: [DOC] Added example for underscore
+	  conventions in method names. Also added doc to clarify encoding
+	  character set support for Ruby programs and elaborated on defining
+	  predicate and bang methods. Based on a patch by @gaurish
+	  [Fixes GH-477] https://github.com/ruby/ruby/pull/477
+
+Mon Dec 23 03:18:09 2013  Zachary Scott  <e@zzak.io>
+
+	* doc/ChangeLog-1.9.3: [DOC] Fix typos by @dvsuresh
+	  [Fixes GH-485] https://github.com/ruby/ruby/pull/485
+	* ext/openssl/ossl_config.c: ditto
+	* lib/rss/utils.rb, lib/time.rb: ditto
+	* test/ruby/envutil.rb: ditto
+
+Sun Dec 22 23:56:56 2013  Zachary Scott  <e@zzak.io>
+
+	* lib/{rake,rdoc,rss}/*, test/rexml/test_listener.rb: [DOC] Fix typos
+	  by @dvsuresh [Fixes GH-486] https://github.com/ruby/ruby/pull/486
+
+Sun Dec 22 11:08:47 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* test/rubygems/test_gem_ext_builder.rb:  Fix warning due to ambiguous
+	  expression.
+
+Sun Dec 22 11:05:53 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems/commands/install_command.rb:  Restore gem install
+	  --ignore-dependencies for remote gems
+	* test/rubygems/test_gem_commands_install_command.rb:  Test for the
+	  above.
+
+Sun Dec 22 10:23:40 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rdoc.rb:  Set RDoc to release version.
+
+Sun Dec 22 10:19:07 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb:  Set RubyGems to release version.
+
+Sun Dec 22 10:16:08 2013  Eric Hodel  <drbrain@segment7.net>
+
+	* lib/rubygems.rb (module Gem):  Fix comment for
+	  Gem::load_path_insert_index.
+
+Sun Dec 22 04:07:55 2013  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (ruby_gc_set_params): don't show obsolete warnings for
+	  RUBY_FREE_MIN/RUBY_HEAP_MIN_SLOTS if
+	  RUBY_GC_HEAP_FREE_SLOTS/RUBY_GC_HEAP_INIT_SLOTS are given.
+	  [Bug #9276]
+
+Sat Dec 21 13:00:20 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c: remove unnecessary the source path from EEXIST error
+	  messages and show the destination path only.  [ruby-core:59202]
+	  [Feature #9263]
+
+Sat Dec 21 12:37:19 2013  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::Hosts#lazy_initialize): should not
+	  consider encodings in hosts file.  [ruby-core:59239] [Bug #9273]
+
+	* lib/resolv.rb (Resolv::Config.parse_resolv_conf): ditto.
+
+Sat Dec 21 05:43:27 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/make-snapshot: replace RUBY_PATCHLEVEL_STR in version.h to
+	  show users that this ruby is a preview/rc version.
+
+Sat Dec 21 05:03:49 2013  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (fake.rb): fixed wrong RUBY_PLATFORM, to correctly
+	  install win32.h.
+	  [ruby-core:58801][Bug #9199] reported by arton.
+
+For the changes before 2.1.0, see doc/ChangeLog-2.1.0
+For the changes before 2.0.0, see doc/ChangeLog-2.0.0
+For the changes before 1.9.3, see doc/ChangeLog-1.9.3
+For the changes before 1.8.0, see doc/ChangeLog-1.8.0
+
+Local variables:
+coding: us-ascii
+add-log-time-format: (lambda ()
+  (let* ((time (current-time))
+	 (system-time-locale "C")
+	 (diff (+ (cadr time) 32400))
+	 (lo (% diff 65536))
+	 (hi (+ (car time) (/ diff 65536))))
+  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
+indent-tabs-mode: t
+tab-width: 8
+change-log-indent-text: 2
+end:
+vim: tabstop=8 shiftwidth=2
diff --git a/doc/ChangeLog-2.3.0 b/doc/ChangeLog-2.3.0
new file mode 100644
index 0000000000..7f3c4e672a
--- /dev/null
+++ b/doc/ChangeLog-2.3.0
@@ -0,0 +1,12187 @@
+Thu Dec 24 23:01:57 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: added entry for CGI.escapeHTML optimization.
+
+Thu Dec 24 18:43:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_compile_error_with_enc, rb_compile_error),
+	  (rb_compile_bug): deprecate internal functions.
+
+	* parse.y (parser_yyerror): construct exception message with
+	  source code and caret.
+
+Thu Dec 24 17:25:42 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (append_compile_error), parse.y (compile_error):
+	  preserve encoding of source file name in exceptions.
+
+	* error.c (rb_compile_error_str, rb_compile_bug_str): add.
+
+Thu Dec 24 16:17:47 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (fake.rb): $(arch)-fake.rb must depend on miniruby because
+	  it may depend on miniruby.
+
+Thu Dec 24 16:13:05 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (ripper.c): r50045 wrongly replace $(PWD) with ../..
+	  It is the top of build directory, not topsrcdir.
+
+Thu Dec 24 15:02:42 2015  sorah (Shota Fukumori)  <her@sorah.jp>
+
+	* tool/vcs.rb (IO.popen): Refactor. Avoid assigning in condition.
+
+Thu Dec 24 15:01:38 2015  sorah (Shota Fukumori)  <her@sorah.jp>
+
+	* tool/file2lastrev.rb: Fix ArgumentError to work on Ruby 1.8.7.
+
+Thu Dec 24 14:44:08 2015  sorah (Shota Fukumori)  <her@sorah.jp>
+
+	* tool/vcs.rb (IO.popen): Enable on Ruby 1.9 where chdir option is not
+	  supported on IO.popen
+
+	* tool/vcs.rb (IO.popen): Fix NoMethodError. I guess r49705 was not
+	  tested... :/
+
+Thu Dec 24 14:57:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* NEWS: rename
+	  "Implementation changes" section to
+	  "Supported platform changes" section.
+
+	* NEWS: add "Implementation improvements" and add several entries.
+
+	* NEWS: add NEWS entries by Eric Wong. [ruby-core:72450]
+
+Thu Dec 24 00:26:05 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/extconf.rb: fix gem build failure on Windows.
+	  only win32_vk.inc is included in the gem and no dependencies for
+	  the header, so that gperf will not be mandatory.
+	  [ruby-core:72453] [Bug #11866]
+
+	* ext/io/console/io-console.gemspec: include depend file and
+	  win32_vk header.
+
+Wed Dec 23 23:58:44 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_readwrite_syserr_fail): works with the given errno than
+	  thread local errno.
+
+Wed Dec 23 17:57:45 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c, io.c, util.c: prefer rb_syserr_fail with saved errno
+	  over setting errno then call rb_sys_fail, not to be clobbered
+	  potentially and to reduce thread local errno accesses.
+
+Wed Dec 23 11:58:52 2015  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* string.c: Fix document. Default value of the first
+	  argument of `String#split` is not `$;` but `nil`.
+	  When `nil` is passed as first argument, `$;` is used.
+	  [ci skip] [Bug #11729] [ruby-dev:49378]
+
+Wed Dec 23 07:15:17 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_init_sock): reject reserved FDs
+	  [ruby-core:72445] [Bug #11862]
+
+Wed Dec 23 02:59:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (__STDC_WANT_LIB_EXT1__): necessary to use memset_s
+	  in strict C99 mode.
+
+Wed Dec 23 02:34:36 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regexec.c (match_at): move direct threaded VM code to get rid of
+	  mixed declarations and code, and enable it only for gcc since it
+	  depends on a gcc extension.
+
+Wed Dec 23 02:23:19 2015  Yuki Nishijima  <mail@yukinishijima.net>
+
+	* gems/bundled_gems: Upgrade the did_you_mean gem to 1.0.0
+
+	* NEWS: Add news about the did_you_mean gem
+
+Wed Dec 23 02:18:57 2015  Jake Worth  <jakeworth82@gmail.com>
+
+	* doc/contributing.rdoc: [DOC] remove an extra word "here".
+	  [Fix GH-1169]
+
+Wed Dec 23 01:58:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regexec.c (USE_DIRECT_THREADED_VM): enable direct threaded VM by
+	  the default.
+
+Tue Dec 22 22:15:53 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (internal_object_p): should not expose singleton classes
+	  without a metaclass.  based on patches by ko1 and shugo.
+	  [Bug #11740]
+
+	* class.c (rb_singleton_class_object_p): added.
+
+Tue Dec 22 22:15:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/io-console.gemspec: bump up to 0.4.4.
+
+Tue Dec 22 22:11:06 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: Added news entry of Psych 2.0.17
+
+Tue Dec 22 22:09:01 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: Added news entry of RDoc 4.2.1
+
+Tue Dec 22 21:20:00 2015  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.gemspec: bump version to 1.2.8.
+
+Tue Dec 22 21:08:05 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rdoc.rb: bump version to 4.2.1. It contains following fixes.
+	  https://github.com/rdoc/rdoc/pull/340
+	  https://github.com/rdoc/rdoc/pull/341
+	  https://github.com/rdoc/rdoc/pull/367
+	  https://github.com/rdoc/rdoc/pull/368
+	* lib/rdoc/*: ditto.
+	* test/rdoc/*: ditto.
+
+Tue Dec 22 20:25:33 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/lib/psych.rb: bump version to 2.0.17
+	* ext/psych/psych.gemspec: ditto.
+
+Tue Dec 22 20:14:47 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: move vm_callee_setup_block_arg() (and related
+	  functions) to the latter location.
+	  This moving recovers performance a little.
+	  [Bug #11829]
+
+Tue Dec 22 15:21:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_compat_and_valid): as scrub does nothing for dummy
+	  encoding string now, incompatible encoding is not a matter.
+
+Tue Dec 22 14:31:28 2015  Toru Iwase  <tietew@tietew.net>
+
+	* ext/cgi/escape/escape.c (optimized_escape_html): CGI.escapeHTML
+	  should return unfrozen new string.
+	  [ruby-core:72426] [Bug #11858]
+
+Tue Dec 22 05:39:58 2015  Takashi Kokubun  <takashikkbn@gmail.com>
+
+	* ext/cgi/escape/escape.c (preserve_original_state): Preserve
+	  original state for tainted and frozen.  [Fix GH-1166]
+	  [ruby-dev:49451] [Bug #11855]
+
+Tue Dec 22 03:57:20 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_init_sock): check FD after validating
+	* test/socket/test_basicsocket.rb (test_for_fd): new
+	  [ruby-core:72418] [Bug #11854]
+
+Mon Dec 21 21:29:45 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* variable.c (struct ivar_update): rename "extended" to "iv_extended"
+	  to avoid name conflict with /usr/include/floatingpoint.h on
+	  Solaris. [Bug #11853] [ruby-dev:49448]
+
+Mon Dec 21 12:15:32 2015  Kimura Wataru  <kimuraw@i.nifty.jp>
+
+	* test/ruby/test_io.rb: handled rlimit value same as r52277
+	  [Bug #11852][ruby-dev:49446]
+
+Mon Dec 21 10:21:22 2015  Ilya Vassilevsky <vassilevsky@gmail.com>
+
+	* lib/net/http.rb (open_timeout): update default value in RDoc
+	  [ruby-core:72413]
+
+Mon Dec 21 10:18:46 2015  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* vm_backtrace.c (rb_profile_frames): ignore ifunc frames as it
+	  did before.  [ruby-core:72409] [Bug #11851]
+
+Mon Dec 21 09:33:17 2015  Karol Bucek  <kares@users.noreply.github.com>
+
+	* ext/openssl/lib/openssl/ssl.rb (OpenSSL::SSL::SSLSocket): fix
+	  NotImplementedError typo.  [Fix GH-1165]
+
+Sun Dec 20 20:54:51 2015  Takashi Kokubun  <takashikkbn@gmail.com>
+
+	* cgi/escape/escape.c: Optimize CGI.escapeHTML for
+	  ASCII-compatible encodings.  [Fix GH-1164]
+
+Sun Dec 20 15:36:46 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/erb.rb: revert r53123. It breaks compatibility like thor and
+	  rspec-rails.
+	  We should try with Ruby 2.4 or 3.0.
+	  [Bug #11842][ruby-core:72374]
+	* lib/rdoc/erb_partial.rb: ditto.
+	* template/verconf.h.tmpl: ditto.
+
+Sun Dec 20 11:43:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): allow here documents in labeled
+	  argument.  [ruby-core:72396] [Bug #11849]
+
+Sun Dec 20 11:14:11 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_mod_define_method): should check Symbol or not.
+	  [Bug #11850]
+
+	* test/ruby/test_method.rb: add a test.
+
+Sun Dec 20 11:01:57 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_mod_define_method): fix notation.
+
+Sun Dec 20 10:54:15 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (proc_new): fix notation.
+
+Sun Dec 20 00:29:00 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_proc_get_iseq): proc made from symbol does not have
+	  iseq.  fix infinite loop.  [ruby-core:72381] [Bug #11845]
+
+Sat Dec 19 20:06:10 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* enc/windows_1250.c: Should not use C++ style comments (C99 feature).
+	  [Bug #11843]
+
+Sat Dec 19 17:17:04 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler#initialize):
+	  use WEBrick::Utils::TimeoutHandler::Thread, which is ignored by
+	  LeakChecker#find_threads, instead of ::Thread to get rid of
+	  thread leak checker.  since this TimeoutHandler is resident
+	  during tests because of Singleton, it waits for the next timeout
+	  if it has any schedules.  in the case of nested timeouts, inner
+	  timeout does not cancel outer timeouts and then those schedules
+	  still remain.
+
+Sat Dec 19 14:28:01 2015  Jake Worth  <jakeworth82@gmail.com>
+
+	* ext/zlib/zlib.c (Init_zlib): [DOC] Fix double-word typo and
+	  grammatical error.  [Fix GH-1162]
+
+Sat Dec 19 14:23:59 2015  Jake Worth  <jakeworth82@gmail.com>
+
+	* lib/csv.rb (CSV#initialize): [DOC] Fix double-word typo.
+	  [Fix GH-1161]
+
+Sat Dec 19 10:33:33 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_2.c, enc/windows_1250.c: separate Windows-1250
+	  from ISO-8859-2 to fix 0x80..0x9e range (from Kimihito Matsui)
+
+Fri Dec 18 21:26:54 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler): To prevent
+	  potential deadlocks, Queue is used to tell update of @timeout_info
+	  instead of sleep and wakeup. [Bug #11742] [ruby-dev:49387]
+
+Fri Dec 18 17:24:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (ibf_load_object_string): use fstring if frozen string.
+
+Fri Dec 18 16:54:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_set_encoding): add StringIO's own
+	  encoding and separate it from the buffer string to override the
+	  encoding of string when reading.  [ruby-core:72189] [Bug #11827]
+	  note that setting the encoding of its buffer string directly
+	  without StringIO#set_encoding may cause unpredictable behavior.
+
+Fri Dec 18 16:50:35 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (ibf_load_setup): check tainted string argument.
+
+Fri Dec 18 16:12:13 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: define USE_LAZY_LOAD if it is not defined.
+
+Fri Dec 18 15:40:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_unget_bytes): extract from
+	  strio_ungetbyte to share with strio_ungetc.
+
+Fri Dec 18 12:39:42 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/syntax/*.rdoc: separated modifier at sentence.
+	  [ci skip][fix GH-1121] Patch by @clandry94
+
+Fri Dec 18 12:09:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_ungetbyte): pad with \000 when
+	  the current position is after the end.
+
+Fri Dec 18 11:24:48 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_method_entry_make, check_override_opt_method):
+	  should check whether a newly created method overrides an optimize
+	  method in case the method is defined in a prepended module of a
+	  built-in class.
+	  [ruby-core:72226] [Bug #11836]
+
+Fri Dec 18 11:09:38 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* vm.c (vm_exec): call RUBY_DTRACE_CMETHOD_RETURN_HOOK instead of
+	  RUBY_DTRACE_METHOD_RETURN_HOOK.
+
+Fri Dec 18 10:24:44 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/irb/ruby-lex.rb: fixed parse error for striped heredocument syntax.
+	  [fix GH-1127] Patch by @koic
+
+Fri Dec 18 09:44:47 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/pty/pty.c: fix double words typo.
+	  [ci skip][fix GH-1157] Patch by @jwworth
+
+Fri Dec 18 09:42:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/utf8tbl.c: fix a typo.
+	  [ci skip][fix GH-1159] Patch by @akshay-vishnoi
+	* ext/nkf/nkf-utf8/utf8tbl.h: ditto.
+
+Fri Dec 18 07:39:01 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm.c (rb_vm_check_redefinition_opt_method): should check the real
+	  class instead of the origin iclass.
+	  [ruby-core:72188] [Bug #11826]
+
+Thu Dec 17 22:13:10 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_args.c (vm_caller_setup_arg_block): remove code for ifunc
+	  because it was made unnecessary by r52138.
+
+Thu Dec 17 16:13:10 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* proc.c (rb_block_arity): should not call GetProcPtr() for symbols.
+	  [ruby-core:72205] [Bug #11830]
+
+Thu Dec 17 14:16:29 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_scrub): the result should be infected by the
+	  original string.
+
+Thu Dec 17 13:35:27 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_econv_substr_append, econv_primitive_convert):
+	  the result should be infected by the original string.
+
+Thu Dec 17 09:46:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (reg_names_iter): should consider encoding of regexp.
+	  [ruby-core:72185] [Bug #11825]
+
+Thu Dec 17 03:52:10 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_make_env_each): should not compare with Qfalse and FALSE.
+	  Pointed at http://d.hatena.ne.jp/nagachika/20151216/ruby_trunk_changes_53128_53163
+
+Thu Dec 17 03:15:25 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method_each_type): should not set fastpath
+	  with keyword arguments for VM_METHOD_TYPE_ATTRSET type methods.
+
+	  Normally, we can not use keyword arguments for this kind of methods,
+	  (obj.foo = 1), but we can set alias names for them.
+	  [Bug #11657]
+
+	* test/ruby/test_keyword.rb: add a test for this fix.
+
+Wed Dec 16 20:32:43 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/fiddle/handle.c: check tainted string arguments.
+	  Patch provided by tenderlove and nobu.
+
+	* test/fiddle/test_handle.rb (class TestHandle): add test for above.
+
+
+Wed Dec 16 19:30:56 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm.c (vm_make_proc_from_block): should convert a Symbol to a Proc.
+	  [ruby-core:72083] [Bug #11811]
+
+Wed Dec 16 16:17:34 2015  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_io.rb: fix spelling errors
+
+Wed Dec 16 16:04:49 2015  Eric Wong  <e@80x24.org>
+
+	* NEWS: note IO#advise change [ruby-core:72168]
+
+Wed Dec 16 15:35:13 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: fix mark miss for proc given as passed block.
+	  [Bug #11750]
+
+	* vm.c (vm_make_proc_from_block): should return a Proc object
+	  if block is given. Previous implementation returns
+	  a Proc object only when corresponding Proc object is not
+	  available.
+
+	* vm.c (vm_make_env_each): ditto.
+
+	* test/ruby/test_proc.rb: add a test for this bug.
+
+Wed Dec 16 12:24:59 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* test_struct.rb: Test that initialize is overridable [#11708]
+
+Wed Dec 16 10:49:51 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (block_command, block_call): fix `&.` calls after
+	  block_call.  [Feature #11537]
+
+Wed Dec 16 00:53:45 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler): Acquire
+	  TimeoutMutex only when accessing @timeout_info for avoiding
+	  potential deadlock. [Bug #11742] [ruby-dev:49387]
+
+Wed Dec 16 00:39:27 2015  Jake Worth  <jakeworth82@gmail.com>
+
+	* doc/extension.rdoc: [DOC] fix double-word typo.  [Fix GH-1153]
+
+Wed Dec 16 00:25:41 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler#initialize):
+	  TimeoutMutex should be acquired when accessing @timeout_info.
+	  To avoid deadlock, interrupt() calls are delayed.
+	  Due to the mutex, it is safe to treat ary without ary.dup.
+	  [Bug #11742] [ruby-dev:49387]
+
+Tue Dec 15 23:13:10 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* gc.c: Delete excess semicolon after RUBY_ALIAS_FUNCTION().
+	  Suppress "syntax error:  empty declaration" warnings by
+	  Oracle Solaris Studio 12.x on Solaris. [Bug #11821]
+
+	* hash.c: ditto, after NOINSERT_UPDATE_CALLBACK().
+
+Tue Dec 15 18:04:04 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* NEWS: added news about EBCDIC encoding
+
+Tue Dec 15 17:57:57 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/ebcdic.h, enc/trans/ebcdic.trans,
+	  test/ruby/test_transcode.rb: Fixed encoding name
+	  to the correct one in the IANA registry (IBM037)
+	  and added an alias (ebcdic-cp-us)
+
+Tue Dec 15 16:19:26 2015  Takashi Kokubun  <takashikkbn@gmail.com>
+
+	* lib/erb.rb: Render erb with array buffer for function call optimization.
+	  [fix GH-1143]
+	* lib/rdoc/erb_partial.rb: ditto.
+	* template/verconf.h.tmpl: ditto.
+
+Tue Dec 15 13:50:05 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_oct): [DOC] mention radix indicators.
+	  [ruby-core:71310] [Bug #11648]
+
+Tue Dec 15 12:20:30 2015  Takashi Kokubun  <takashikkbn@gmail.com>
+
+	* lib/erb.rb: Simplify regexp to optimize erb scanner.
+	  [fix GH-1144]
+
+Tue Dec 15 11:56:24 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/uri/common.rb: make code block for rdoc.
+	  [ci skip][fix GH-1152] Patch by @Tonkpils
+
+Tue Dec 15 11:55:08 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/zlib/zlib.c: fix a typo.
+	  [ci skip][fix GH-1149] Patch by @crismali
+
+Tue Dec 15 09:14:14 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* tool/transcode_tablegen.rb: detailed documentation
+	  for transcode_tblgen function [ci skip]
+
+Mon Dec 14 22:11:11 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/ebcdic.h: new dummy encoding EBCDIC-US
+	* enc/trans/ebcdic.trans: transcodings between EBCDIC-US
+	  and iso-8859-1 [with code from Andrea Ribuoli]
+	* test/ruby/test_transcode.rb: tests for above
+	* tool/transcode_tablegen.rb: additional argument for
+	  method transcode_tblgen
+
+Mon Dec 14 17:04:14 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/socket/lib/socket.rb: use safe navigation operator.
+	  [fix GH-1142] Patch by @mlarraz
+	* lib/drb/extservm.rb: ditto.
+	* lib/net/http.rb: ditto.
+	* lib/net/http/response.rb: ditto.
+	* lib/scanf.rb: ditto.
+	* lib/uri/generic.rb: ditto.
+
+Mon Dec 14 17:03:05 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* bootstraptest/runner.rb: use safe navigation operator.
+	  [fix GH-1142] Patch by @mlarraz
+	* test/openssl/test_pair.rb: ditto.
+	* test/ruby/test_econv.rb: ditto.
+	* test/ruby/test_settracefunc.rb: ditto.
+	* test/thread/test_queue.rb: ditto.
+
+Mon Dec 14 14:33:35 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/xmlrpc.rb: added documentation for parser details.
+	  [ci skip][fix GH-1124] Patch by @jrafanie
+
+Mon Dec 14 11:46:52 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* enum.c: fix a typo in documentation.
+	  [ci skip][fix GH-1140] Patch by @jutaz
+	* io.c: ditto.
+	* iseq.c: ditto.
+	* numeric.c: ditto.
+	* process.c: ditto.
+	* string.c: ditto.
+	* vm_trace.c: ditto.
+
+Mon Dec 14 11:41:59 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/cgi.rb: fix a typo in documentation.
+	  [ci skip][fix GH-1140] Patch by @jutaz
+
+Mon Dec 14 11:31:00 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* compile.c: fix typos.
+	  [ci skip][fix GH-1140] Patch by @jutaz
+	* dir.c: ditto.
+	* gc.c: ditto.
+	* io.c: ditto.
+	* node.h: ditto.
+	* thread_pthread.c: ditto.
+	* vm_insnhelper.c: ditto.
+	* vsnprintf.c: ditto.
+
+Mon Dec 14 11:27:01 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* enc/iso_2022_jp.h: fix typos.
+	  [ci skip][fix GH-1140] Patch by @jutaz
+	* enc/utf_16_32.h: ditto.
+	* enc/utf_7.h: ditto.
+
+Mon Dec 14 11:25:57 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* benchmark/bm_app_aobench.rb: fix typos.
+	  [ci skip][fix GH-1140] Patch by @jutaz
+	* benchmark/bm_vm_thread_pipe.rb: ditto.
+
+Sun Dec 13 23:46:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (trace_lex_state): trace lex_state changes if yydebug is
+	  set, and send the messages to rb_stdout.
+
+	* parse.y (rb_parser_printf): store YYPRINTF messages per lines
+	  so that lex_state traces do not mix.
+
+	* tool/ytab.sed: add parser argument to yy_stack_print too.
+
+Sun Dec 13 20:41:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (build_lex_state_name, trace_lex_state): lex_state is
+	  now bit flags and can be set 2 bits or more.
+
+Sun Dec 13 20:26:30 2015  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* test/ruby/test_syntax.rb: fix typo in test
+
+Sun Dec 13 20:12:14 2015  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* parse.y (parse_percent): Allow %-literals in labeled arg as
+	  r51624 did for parentheses.
+	  Fixes [ruby-core:72084] [Bug #11812].
+
+Sun Dec 13 20:02:15 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: fix a typo
+
+Sun Dec 13 19:54:26 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/lib/envutil.rb: move envutil's assertions under Test::Unit::Assertion.
+	* test/lib/test/unit/assertions.rb: ditto.
+
+Sun Dec 13 19:24:20 2015  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* parse.y (lex_state_name): Make it return the correct names.
+	  Add new names to follow r51617; Indices ffs(2) returns are 1-origin.
+
+Sun Dec 13 18:40:45 2015  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* parse.y: debug output of lex_state transition if PARSER_DEBUG
+
+Sun Dec 13 18:49:25 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (parse_mode_enc): preserve encoding of mode string in
+	  warnings.
+
+	* io.c (io_encname_bom_p): check BOM prefix only, not including
+	  UTF prefix.
+
+	* io.c (parse_mode_enc): warn BOM with non-UTF encoding.
+
+	* io.c (parse_mode_enc): fix buffer overflow.
+
+Sun Dec 13 18:35:57 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/function.c (initialize): check all arguments first.
+	  reported by Marcin 'Icewall' Noga of Cisco Talos.
+
+	* ext/fiddle/conversions.h (PTR2NUM): use signed integer to make
+	  Fixnum for negative values.
+
+Sun Dec 13 18:33:41 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* pack.c (pack_pack): always check index range against the
+	  receiver array length, which can be shortened by elements
+	  conversion.  reported by Marcin 'Icewall' Noga of Cisco Talos.
+
+Sun Dec 13 18:28:52 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/psych/psych_emitter.c (start_document): should not exceed
+	  tags array range.
+
+	* ext/psych/psych_emitter.c (start_document): ensure string before
+	  encoding conversion.
+
+Sun Dec 13 18:26:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tkutil/tkutil.c (tk_hash_kv): check types of array
+	  argument.  reported by Marcin 'Icewall' Noga of Cisco Talos.
+
+	* ext/tk/tkutil/tkutil.c (cbsubst_table_setup): check length of
+	  argument arrays for each access, as callback methods can modify
+	  them.  reported by Marcin 'Icewall' Noga of Cisco Talos.
+
+	* ext/tk/tkutil/tkutil.c (cbsubst_table_setup): check types of
+	  argument elements.  reported by Marcin 'Icewall' Noga of Cisco
+	  Talos.
+
+Sun Dec 13 18:19:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (ole_vstr2wc): check argument type, vstr
+	  must be a String here.  reported by Marcin 'Icewall' Noga of
+	  Cisco Talos.
+
+Sun Dec 13 16:41:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (BOOTSTRAPRUBY): make BASERUBY full path before
+	  building ruby to get rid of unexpectedly invoking built ruby.
+	  [ruby-core:72065] [Bug #11807]
+
+	* configure.in (BOOTSTRAPRUBY): use MINIRUBY but not BASERUBY
+	  unless cross compiling.  [ruby-core:72065] [Bug #11807]
+
+Sun Dec 13 14:17:19 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* test/test_shellwords.rb (TestShellwords): Add many more tests
+	  for handling whitespace characters and frozenness.
+
+Sun Dec 13 14:16:09 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb (Shellwords#shellsplit): Document that this
+	  method does not treat shell metacharacters as such.
+
+Sun Dec 13 12:17:43 2015  Eric Wong  <e@80x24.org>
+
+	* lib/shellwords.rb (shellescape): duplicate frozen literal
+	* test/test_shellwords.rb (test_stringification): new test
+
+Sun Dec 13 11:47:35 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* object.c (rb_inspect): check the default internal encoding as
+	  String#inspect do.
+	  [ruby-dev:49415] [Bug #11787]
+
+Sun Dec 13 11:38:12 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb: Turn on frozen-string-literal after fixing
+	  shellsplit.
+
+Sun Dec 13 10:44:44 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* KNOWNBUGS.rb: Fixed typo, made more explicit [ci skip]
+
+Sun Dec 13 10:26:47 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ssl_npn_select_cb_common): fix parsing
+	  protocol list.
+	  The protocol list from OpenSSL is not null-terminated.
+	  patched by Kazuki Yamaguchi [Bug #11810] [ruby-core:72082]
+
+Sun Dec 13 06:40:30 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb: Have OpenStruct#dig raise if argument is not a
+	  symbol
+	  nor a string. See [#11762]
+
+Sun Dec 13 00:05:42 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method_missing): method_missing should
+	  not be refined.
+	  [ruby-core:72080] [Bug #11809]
+
+Sat Dec 12 23:00:17 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Merge nkf 2.1.4.
+
+Sat Dec 12 18:52:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_dig): raise TypeError if an element does not
+	  have #dig method.  [ruby-core:71798] [Bug #11762]
+
+Sat Dec 12 17:59:07 2015  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* test/ruby/test_regexp.rb: Add test cases for `$KCODE` and `$=` warning
+	  [Misc #11770][ruby-dev:49398]
+
+Sat Dec 12 17:11:57 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* doc/NEWS-0.2.2: add description about incompatible change in Hash
+	  duplicated key overriding policy. [Bug #10315] [Bug #11501]
+
+Sat Dec 12 07:44:38 2015  Eric Wong  <e@80x24.org>
+
+	* io.c (do_io_advise): do not raise on ENOSYS
+	* test/ruby/test_io.rb (test_advise): do not skip on Errno::ENOSYS
+	  (test_advise_pipe): ditto
+	  [ruby-core:72066] [Feature #11806]
+
+Sat Dec 12 07:05:29 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/windows_1252.c: separate from ISO-8859-1 to fix 0x80..0x9e
+	  range.  [ruby-core:64049] [Bug #10097]
+
+Fri Dec 11 23:33:40 2015  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* sample/trick2015/: added the award-winning entries of TRICK 2015.
+	  See https://github.com/tric/trick2015 for the contest outline.
+
+Fri Dec 11 17:59:05 2015  Eric Wong  <e@80x24.org>
+
+	* insns.def (opt_case_dispatch): avoid converting Infinity
+	* test/ruby/test_optimization.rb (test_opt_case_dispatch_inf): new
+	  [ruby-dev:49423] [Bug #11804]'
+
+Fri Dec 11 16:48:57 2015  Eric Wong  <e@80x24.org>
+
+	* hash.c (rb_num_hash_start): avoid pathological behavior
+	  [ruby-core:72028] [Feature #11405]
+
+Fri Dec 11 11:58:46 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: Mentioned rubygems-2.5.1
+
+Fri Dec 11 11:52:39 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems: Update to RubyGems 2.5.1
+	* test/rubygems: ditto.
+
+Fri Dec 11 11:38:14 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each, ibf_load_object_hash): rehash
+	  case-dispatch hash to reduce collisions.
+	  http://d.hatena.ne.jp/ku-ma-me/20151210
+
+Fri Dec 11 03:44:43 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* object.c (rb_inspect): dump inspected result with rb_str_escape()
+	  instead of raising Encoding::CompatibilityError. [Feature #11801]
+
+	* string.c (rb_str_escape): added to dump given string like
+	  rb_str_inspect without quotes and always dump in US-ASCII
+	  like rb_str_dump.
+
+Thu Dec 10 14:59:59 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb (test_expand_heap): relax condition (1->2).
+
+Thu Dec 10 14:15:59 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* cont.c: fix a double word typo.
+	  [Bug #11313][ruby-core:69749]
+
+Thu Dec 10 14:13:34 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/tk/lib/multi-tk.rb: fix typos.
+	  [Bug #11764][ruby-core:71800]
+
+Thu Dec 10 11:33:34 2015  Eric Wong  <e@80x24.org>
+
+	* compile.c (iseq_compile_each): reduce needless rb_str_dup
+	  [ruby-core:72018] <5668DB6E.8000101@ruby-lang.org>
+
+Thu Dec 10 09:32:51 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb, lib/shellwords.rb: disable frozen-string-literal.
+	  [ruby-core:72011] [Bug #11800]
+
+Thu Dec 10 06:33:39 2015  Eric Wong  <e@80x24.org>
+
+	* marshal.c (memsize_dump_arg): remove NULL check
+	  (memsize_load_arg): ditto
+
+Thu Dec 10 05:53:18 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb: rbconfig must not be frozen-string-literal to
+	  expand CONFIG hash.  [ruby-core:72006] [Bug #11798]
+
+Thu Dec 10 05:03:51 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ifaddr.c (ifaddr_mark): remove empty function
+	  (ifaddr_type): pass zero to rb_data_type_t.function.dmark
+
+Thu Dec 10 04:49:16 2015  Eric Wong  <e@80x24.org>
+
+	* proc.c (bm_free): remove, use default free
+	  (method_data_type): use RUBY_TYPED_DEFAULT_FREE
+
+Thu Dec 10 02:01:41 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): do not add debug information
+	  without --debug or --debug=frozen-string-literal option
+	  because String#dup slows down with debug information.
+	  [Feature #11725]
+
+	* NEWS: apply about it.
+
+	* test/ruby/test_rubyoptions.rb: catch up this fix with refactoring.
+
+Thu Dec 10 00:06:56 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c: rename methods
+	  RubyVM::InstructionSequence#to_binary_format -> #to_binary
+	  RubyVM::InstructionSequence.from_binary_format -> .load_from_binary
+	  RubyVM::InstructionSequence.from_binary_format_extra_data ->
+	                             .load_from_binary_extra_data
+
+	* iseq.c: fix document of iseq.to_binary.
+	  [Fix GH-1134]
+
+	* sample/iseq_loader.rb: catch up this change.
+
+	* test/lib/iseq_loader_checker.rb: ditto.
+
+Wed Dec  9 17:02:03 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* regparse.h (SET_NTYPE): get rid of breaking strict aliasing.
+	  patch by Zarko Todorovski in [ruby-core:71953].  [Bug #11790]
+
+Wed Dec  9 16:10:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_cref_in_context): Module#define_method in non-class
+	  expression should be public.
+	  [Bug #11754]
+
+	* test/ruby/test_method.rb: add a test.
+
+Wed Dec  9 14:45:27 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_stacked_objects): fix typo.
+	  reported by XIE Zhibang. [Bug #11763]
+
+Wed Dec  9 14:37:51 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* doc/syntax/refinements.rdoc: remove outdated description.
+
+Wed Dec  9 09:58:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/**/*.c (*_memsize): same as r52986 for extensions.
+
+Wed Dec  9 09:46:19 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* .gitignore: ignored ISeq binary format.
+
+Wed Dec  9 09:34:41 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* *.c (*_memsize): do not check ptr.
+	  NULL checking is finished Before call of memsize functions.
+	  See r52979.
+
+Wed Dec  9 09:25:29 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/net/smtp/test_response.rb: use Test::Unit. We should use Test::Unit
+	  without rubygems and rdoc.
+
+Wed Dec  9 06:26:23 2015  Colin Kelley <colindkelley@gmail.com>
+
+	* lib/uri/generic.rb: enable frozen_string_literal
+	  (split_userinfo): remove explicit .freeze for string literals
+	  (check_path): ditto
+	  (query): ditto
+	  (fragment): ditto
+	  (to_s): ditto
+	  [ruby-core:71910] [Bug #11759]
+
+Wed Dec  9 06:25:47 2015  Eric Wong  <e@80x24.org>
+
+	* test/uri/test_generic.rb (to_s): new test
+	  [ruby-core:71820]
+
+Wed Dec  9 02:18:52 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (ibf_dump_memsize): should check NULL.
+
+Wed Dec  9 01:46:35 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_init): now accepts new option parameter `encoding'.
+	  [Feature #11785]
+
+Wed Dec  9 00:52:37 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_stat_wr, rb_stat_ww): call get_stat only once and
+	  reduce checking struct.  patch by Yuki Kurihara in
+	  [ruby-core:71949].  [Misc #11789]
+
+Wed Dec  9 00:24:33 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_ibf_dump): dump extra data just string length.
+
+	* sample/iseq_loader.rb: add using
+	  RubyVM::InstructionSequence.from_binary_format_extra_data method
+	  (commented out).
+
+Mon Dec  9 00:21:19 2015  Yuki Nishijima  <mail@yukinishijima.net>
+
+	* gems/bundled_gems: Upgrade the did_you_mean gem to 1.0.0.rc1
+
+Wed Dec  9 00:17:49 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (ibf_load_setup): cast to int.
+
+Wed Dec  9 00:13:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (ibf_setup_load): rename to ibf_load_setup().
+
+	* compile.c (iseq_load_setup): check binary format.
+
+Tue Dec  8 23:30:07 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/io/console/test_io_console.rb (run_pty): Avoid waiting twice
+	  for a process. Fix Errno::ECHILD in TestIO_Console#test_close and
+	  TestIO_Console#test_sync.
+
+Tue Dec  8 23:05:47 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_ibf_dump): fix for clang type checker.
+
+Tue Dec  8 23:04:02 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_s_load): fix mysterious bug.
+
+Tue Dec  8 22:31:58 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* introduce new ISeq binary format serializer/de-serializer
+	  and a pre-compilation/runtime loader sample.
+	  [Feature #11788]
+
+	* iseq.c: add new methods:
+	  * RubyVM::InstructionSequence#to_binary_format(extra_data = nil)
+	  * RubyVM::InstructionSequence.from_binary_format(binary)
+	  * RubyVM::InstructionSequence.from_binary_format_extra_data(binary)
+
+	* compile.c: implement body of this new feature.
+
+	* load.c (rb_load_internal0), iseq.c (rb_iseq_load_iseq):
+	  call RubyVM::InstructionSequence.load_iseq(fname) with
+	  loading script name if this method is defined.
+
+	  We can return any ISeq object as a result value.
+	  Otherwise loading will be continue as usual.
+
+	  This interface is not matured and is not extensible.
+	  So that we don't guarantee the future compatibility of this method.
+	  Basically, you shouldn't use this method.
+
+	* iseq.h: move ISEQ_MAJOR/MINOR_VERSION (and some definitions)
+	  from iseq.c.
+
+	* encoding.c (rb_data_is_encoding), internal.h: added.
+
+	* vm_core.h: add several supports for lazy load.
+	  * add USE_LAZY_LOAD macro to specify enable or disable of
+	    this feature.
+	  * add several fields to rb_iseq_t.
+	  * introduce new macro rb_iseq_check().
+
+	* insns.def: some check for lazy loading feature.
+
+	* vm_insnhelper.c: ditto.
+
+	* proc.c: ditto.
+
+	* vm.c: ditto.
+
+	* test/lib/iseq_loader_checker.rb: enabled iff suitable
+	  environment variables are provided.
+
+	* test/runner.rb: enable lib/iseq_loader_checker.rb.
+
+	* sample/iseq_loader.rb: add sample compiler and loader.
+
+	    $ ruby sample/iseq_loader.rb [dir]
+
+	  will compile all ruby scripts in [dir].
+	  With default setting, this compile creates *.rb.yarb files
+	  in same directory of target .rb scripts.
+
+	    $ ruby -r sample/iseq_loader.rb [app]
+
+	  will run with enable to load compiled binary data.
+
+Tue Dec  8 21:21:16 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: mention about Enumerator::Lazy#grep_v.
+	  [ruby-core:71845] [Feature #11773]
+
+Tue Dec  8 17:36:36 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: removed unused variable
+
+Tue Dec  8 16:23:40 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_grep_v): add Enumerator::Lazy#grep_v as well
+	  as Enumerable, to enumerate lazily.
+	  [ruby-core:71845] [Feature #11773]
+
+Tue Dec  8 14:27:07 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (name_err_local_variables): new method
+	  NameError#local_variables for internal use only.
+	  [Feature #11777]
+
+Tue Dec  8 14:20:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_objivar): skip internal instance variables in
+	  T_OBJECT too.
+
+Tue Dec  8 12:58:04 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/io/console/test_io_console.rb (test_getpass): s.getpass
+	  should be tested. Narrow ensure block. This reverts r52911.
+	  [Bug #11780] [ruby-dev:49412]
+
+Tue Dec  8 10:40:21 2015  Eric Wong  <e@80x24.org>
+
+	* benchmark/bm_vm2_case_lit.rb: new benchmark
+	* compile.c (case_when_optimizable_literal): add nil/true/false
+	* insns.def (opt_case_dispatch): ditto
+	* vm.c (vm_redefinition_check_flag): ditto
+	* vm.c (vm_init_redefined_flag): ditto
+	* vm_core.h: ditto
+	* object.c (InitVM_Object): define === explicitly for nil/true/false
+	* test/ruby/test_case.rb (test_deoptimize_nil): new test
+	* test/ruby/test_optimization.rb (test_opt_case_dispatch): update
+	  (test_eqq): new test
+	  [ruby-core:71923] [Feature #11769]
+	  Original patch by Aaron Patterson <tenderlove@ruby-lang.org>
+
+Tue Dec  8 10:19:02 2015  Jake Worth  <jakeworth82@gmail.com>
+
+	* lib/optparse.rb: fix double word typo in the document.
+	  [Misc #10608] [Fix GH-1126]
+
+Tue Dec  8 09:03:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_core.c (d_lite_lshift): should check the argument
+	  before negation.
+
+Tue Dec  8 08:56:16 2015  Eric Wong  <e@80x24.org>
+
+	* insns.def (opt_case_dispatch): check Float#=== redefinition
+	* test/ruby/test_optimization.rb (test_opt_case_dispatch): new
+	  [ruby-core:71920] [Bug #11784]
+
+Tue Dec  8 03:56:05 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/lib/iseq_loader_checker.rb: add iseq dumper/loader checker.
+	  If you enable this checker (remove `#' in test/runner.rb),
+	  you can see comparison results between an original iseq disassembed
+	  result and dumped and loaded iseq disassembed result.
+
+	  There are several bugs around there, because of inexact stack depth
+	  calculation. Now, I leave these bugs because they are not critical
+	  and difficult to solve completely.
+
+	* test/runner.rb: require test/lib/iseq_loader_checker.rb but
+	  disabled at default (commented out).
+
+Tue Dec  8 03:45:47 2015  Eric Wong  <e@80x24.org>
+
+	* doc/extension.rdoc: warn about kwargs performance in C
+	  [Feature #11339] [ci skip]
+
+Tue Dec  8 03:44:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_load): disable peephole optimization option
+	  because apply it multiple times change the sequence.
+	  (iseq != peephole_optimize(load(iseq.to_a)))
+
+Tue Dec  8 03:43:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (rb_iseq_build_from_ary): do not allocate table
+	  if table_size is 0.
+
+Tue Dec  8 03:30:34 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/unixsocket.c (unix_send_io): document args
+	  (unix_recv_io): ditto
+	* test/socket/test_unix.rb (test_fd_passing_class_mode): added
+
+Tue Dec 08 02:21:35 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_translate): at the end of constructing an iseq,
+	  call RubyVM::InstructionSequence.translate(iseq) if this method
+	  is defined. If the return value is also an object of
+	  RubyVM::InstructionSequence, then use it instead of created one.
+
+	  For example, this method is useful to test iseq dumper/loader
+	  such as RubyVM::InstructionSequence#to_a and rb_iseq_load().
+
+	  Because this method is for such internal experimental usage,
+	  the interface is not matured. For example, this interface has
+	  no extensibility. Two or more translators can not run
+	  simultaneously.
+
+	  So that we don't guarantee future compatibility of this method.
+	  Basically, do not use this method.
+
+Tue Dec  8 01:57:13 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/psych/*: update psych to 2.0.16
+	* test/psych/*: ditto
+
+Mon Dec  7 23:45:20 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: introduce String#+@ and String#-@ to control
+	  String mutability.
+	  [Feature #11782]
+
+Mon Dec  7 23:39:49 2015  Ben Miller  <bjmllr@gmail.com>
+
+	* parse.y: add heredoc <<~ syntax.  [Feature #9098]
+
+Mon Dec  7 23:06:16 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* prelude.rb (IO#read_nonblock): [DOC] add missing options to
+	  call-seq.  [ruby-core:71627] [Bug #11730]
+
+Mon Dec  7 15:50:50 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* .gitignore: added cygruby*.def for Cygwin
+
+Sun Dec  6 19:52:31 2015  Eric Wong  <e@80x24.org>
+
+	* include/ruby/intern.h (rb_autoload): deprecate
+	* internal.h (rb_autoload_str): declare
+	* load.c (rb_mod_autoload): use rb_autoload_str
+	* variable.c (rb_autoload): become compatibility wrapper
+	  (rb_autoload_str): hoisted out from old rb_autoload
+	  [ruby-core:71369] [Feature #11664]
+
+Sun Dec  6 18:25:22 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/webrick/test_cgi.rb (TestWEBrickCGI#test_cgi): gave up the test
+	  of binary path info test on Windows because the test had passed
+	  occasionally as the comment said.
+
+Sun Dec  6 15:25:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_getpass): add IO#getpass
+	  method.
+
+Sun Dec  6 08:39:05 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/json/json.gemspec: bump version to json 1.8.3. CRuby already contained
+	  upstream changes.
+
+	  https://github.com/ruby/ruby/commit/4d059bf9f5f10f3d3088de49fc87e5555db7770d
+	  https://github.com/flori/json/commit/d4c99de78905d96c3f301f48b2c789943bb3f098
+
+	* ext/json/lib/json/version.rb: ditto.
+
+Sat Dec  5 17:48:25 2015  Lars Kanis  <lars@greiz-reinsdorf.de>
+
+	* tool/fake.rb: Fix cross build when srcdir is an absolute path.
+
+	* Makefile.in: PREP dependency is needed when cross build too, not
+	  "-r$(arch)-fake" to be  used before created.  [Fix GH-1125]
+
+Sat Dec  5 17:26:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_str_new, env_path_str_new): make default string
+	  UTF-8 for the case conversion is not possible.  [Bug #8822]
+
+	* hash.c (get_env_cstr): convert non-ASCII string to UTF-8 string.
+
+	* hash.c (ruby_setenv): use wide char version to put environment
+	  variable to deal with non-ASCII value.
+
+Sat Dec  5 09:56:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby_atomic.h (ATOMIC_CAS): old value to be swapped should be
+	  same as the destination.  immediate value may need type
+	  promotion.
+
+	* ruby_atomic.h (ATOMIC_SIZE_CAS): fix the argument order of
+	  InterlockedCompareExchange64.  new value and then old value is
+	  the last.
+
+Sat Dec  5 09:23:34 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (fill_random_seed): fix the size to be filled, not the
+	  size of element, but the whole size of array.
+
+Sat Dec  5 06:03:54 2015  Eric Wong  <e@80x24.org>
+
+	* vm.c (ruby_vm_verbose_ptr): make static
+	  (ruby_vm_debug_ptr): ditto
+
+Sat Dec  5 00:56:29 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/ruby.h (RB_OBJ_TAINT): follow-up of r52881.
+	  Turn into void expression not to use unexpected result.
+	  Fix "operands have incompatible types" error with
+	  Oracle Solaris Studio 12.x on Solaris.
+
+Fri Dec  4 19:52:52 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_13.c: Added three missing lower/upper-case
+	  character pairs (from Kimihito Matsui)
+
+Fri Dec  4 18:57:57 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_4.c: Added missing lower/upper-case character
+	  pair (U+014A and U+014B, LATIN CAPITAL/SMALL LETTER ENG)
+	  (from Kimihito Matsui)
+
+Fri Dec  4 16:48:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_obj_as_string): fstring should not be infected.
+	  re-apply r52872 and fix a typo.
+	  TODO: other frozen strings also may not be.
+
+Fri Dec  4 15:21:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems: Update to RubyGems 2.5.0+ HEAD(fdab4c4).
+	  this version includes #1396, #1397, #1398, #1399
+	* test/rubygems: ditto.
+
+Fri Dec  4 11:22:40 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_setname): name must be ascii-compatible, as
+	  pthread APIs do not accept legacy wide char strings.
+
+Thu Dec  3 15:39:21 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/scanf.rb: fixed double words typo.
+	  [ci skip][fix GH-1123] Patch by @jwworth
+
+Thu Dec  3 15:37:56 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/ruby/test_mixed_unicode_escapes.rb: fixed typo.
+	  [fix GH-1122] Patch by @davydovanton
+	* test/ruby/test_object.rb: ditto.
+	* test/socket/test_tcp.rb: ditto.
+
+Thu Dec  3 15:33:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): fix wrong shifting position in
+	  Rational conversion when not at the beginning of the result.
+	  [ruby-core:71806] [Bug #11766]
+
+Thu Dec  3 14:22:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (range_to_s): should be infected by the receiver.
+	  str2 infects by appending.  [ruby-core:71811] [Bug #11767]
+
+Thu Dec  3 11:57:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: separate SET_CURRENT_THREAD_NAME, which can set
+	  the name of current thread only, and SET_ANOTHER_THREAD_NAME,
+	  which can set the name of other threads.
+
+	* thread.c (rb_thread_setname): use SET_ANOTHER_THREAD_NAME.  OS X
+	  is not possible to set another thread name.
+
+	* thread_pthread.c (native_set_thread_name, thread_timer): use
+	  SET_CURRENT_THREAD_NAME.
+
+Wed Dec 02 22:57:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, iseq.h: remove rb_iseq_t::variable_body.
+	  Fields in rb_iseq_t::variable_body are contained by
+	  rb_iseq_t::body::mark_ary (hidden Array object).
+
+	  Index 0 to 2 of mark_ary are reserved by these objects.
+
+	* iseq.c: catch up this fix.
+
+	* compile.c (rb_iseq_original_iseq): trivial rewrite.
+
+Wed Dec  2 17:19:02 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.h: introduce ISEQ_ORIGINAL_ISEQ() and
+	  ISEQ_ORIGINAL_ISEQ_ALLOC() macro.
+
+	* compile.c: use them to access original iseq buffer.
+
+	* iseq.c: ditto.
+
+	* vm_core.h: rename iseq field to support this fix.
+
+Wed Dec  2 17:10:32 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.h: introduce ISEQ_FLIP_CNT_INCREMENT() macro.
+
+	* compile.c (iseq_compile_each): use it.
+
+	* vm_core.h: rename flip_cnt field to support this fix.
+
+Wed Dec  2 17:05:15 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.h: introduce ISEQ_COVERAGE() and ISEQ_COVERAGE_SET() macro.
+
+	* compile.c: use them.
+
+	* iseq.c: ditto.
+
+	* iseq.c (rb_iseq_coverage): added.
+
+	* thread.c (update_coverage): use rb_iseq_coverage().
+
+	* vm_core.h: rename coverage field name to support this fix.
+
+Wed Dec  2 17:00:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_name, rb_enc_name_list_i, rb_enc_aliases_enc_i):
+	  make fstring instead of making each copies.
+
+Wed Dec  2 16:32:08 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.h: introduce ISEQ_COMPILE_DATA() macro.
+
+	* compile.c, iseq.c: use ISEQ_COMPILE_DATA().
+
+	* vm_core.h: rename compile_data field to support this fix.
+
+Wed Dec  2 16:27:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_m_loader): defer finding encoding object not to
+	  be infected by marshal source.  [ruby-core:71793] [Bug #11760]
+
+	* marshal.c (r_object0): enable compatible loader on USERDEF
+	  class.  the loader function is called with the class itself,
+	  instead of an allocated object, and the loaded data.
+
+	* marshal.c (compat_allocator_table): initialize
+	  compat_allocator_tbl on demand.
+
+	* object.c (rb_undefined_alloc): extract from rb_obj_alloc.
+
+Wed Dec  2 15:12:43 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* configure.in: Fixed double negative comments.
+	  [Bug #11698][ruby-core:71506]
+
+Wed Dec  2 14:55:01 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test/ruby/test_string.rb: removed non executing tests.
+	  [Misc #11757][ruby-dev:49397]
+
+Wed Dec  2 11:23:06 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/csv.rb: enable frozen_string_literal.
+	  [fix GH-1116] Patch by @marshall-lee
+
+Wed Dec  2 10:36:25 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c: Fix double word typo.
+	  [ci skip][fix GH-1120] Patch by @jwworth
+
+Wed Dec  2 07:43:51 2015  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_ssl.c (mSSLExtConfig): make static
+	  (eSSLError): ditto
+	  (ID_callback_state): ditto
+	  (ossl_ssl_ex_vcb_idx): ditto
+	  (ossl_ssl_ex_store_p): ditto
+	  (ossl_ssl_ex_ptr_idx): ditto
+	* ext/openssl/ossl_ssl.h: remove extern declarations for
+	  mSSLExtConfig and eSSLError
+
+Wed Dec  2 07:41:08 2015  Eric Wong  <e@80x24.org>
+
+	* missing/explicit_bzero.c (explicit_bzero): fixup r52839
+	  for compilers with "weak" attribute
+
+Wed Dec  2 06:47:25 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* missing/explicit_bzero.c: add ruby_explicit_bzero_hook_unused
+	  for preventing optimization. Inspired from OpenBSD.
+
+Tue Dec  1 23:36:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_setname): allow to reset thread name.
+
+Tue Dec  1 23:14:04 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_setname): check the argument if valid
+	  string.  [ruby-core:71774] [Bug #11756]
+
+Tue Dec  1 17:13:41 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_string_value_cstr): should not raise on frozen
+	  string.
+
+Tue Dec  1 09:35:29 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* missing/explicit_bzero.c: add a few comment.
+
+Tue Dec  1 09:31:19 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* missing/explicit_bzero.c: add disabling optimization on gcc.
+
+Tue Dec  1 07:50:33 2015  Eric Wong  <e@80x24.org>
+
+	* missing/explicit_bzero.c: new file. define explicit_bzero.
+	  Fixup r52806
+
+Thu Oct 22 12:54:43 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread_pthread.c (register_ubf_list): renamed from
+	  add_signal_thread_list.
+	* thread_pthread.c (unregister_ubf_list): renamed
+	  from remove_signal_thread_list.
+	* thread_pthread.c (ubf_wakeup_all_threads): renamed from
+	  ping_signal_thread_list.
+	* thread_pthread.c (ubf_wakeup_thread): renamed from
+	  ubf_select_each.
+	* thread_pthread.c (ubf_threads_empty): renamed from
+	  check_signal_thread_list().
+	* thread_pthread.c (ubf_list_lock): renamed from
+	  signal_thread_list_lock.
+
+	* thread_pthread.c (register_ubf_list): large simplification
+	  by using ccan/list.h.
+	  bonus: removed malloc() and exit(EXIT_FAILURE).
+	* thread_pthread.c (unregister_ubf_list): ditto.
+	* thread_pthread.c (ubf_threads_empty): ditto.
+	* thread_pthread.c (ubf_wakeup_all_threads): ditto.
+
+	* thread_pthread.c (print_signal_list): removed.
+
+Thu Oct 22 08:03:49 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_rand.rb (TestRand#test_default_seed): add
+	  srand case.
+
+Thu Oct 22 06:33:38 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c (InitVM_Random): move Random::DEFAULT initialization
+	  bits to Init_Random_default.
+	* random.c (Init_Random_default): renamed from Init_RandomSeed2.
+	* random.c (Init_RandomSeedCore): renamed from Init_RandomSeed.
+
+Thu Oct 22 06:20:48 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c (Init_RandomSeed): move all Random::DEFAULT
+	  construction bits to Init_RandomSeed2. Random::DEFAULT
+	  and Ruby internal hashes are no longer shared their seed.
+	* random.c (Init_RandomSeed2): ditto. And, kill evil
+	  rb_obj_reveal() stuff.
+
+	* random.c (init_hashseed): add MT argument.
+	* random.c: (init_siphash): ditto.
+
+	* test/ruby/test_rand.rb (TestRand#test_default_seed): new
+	  test for Random::DEFAULT::seed.
+
+Thu Oct 22 05:23:48 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c (init_hashseed, init_siphash): extract initialize
+	  functions.
+
+Thu Oct 22 01:01:34 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* configure.in: sort AC_CHECK_HEADERS() by alphabetical order.
+
+Thu Oct 22 00:19:07 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c (init_randomseed): remove "initial" argument. It never
+	  be used from outside of this function.
+
+Thu Oct 22 00:12:33 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/securerandom.rb (SecureRandom::gen_random): use /dev/urandom
+	  for initialize OpenSSL's rand.
+
+Wed Oct 21 12:10:04 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ext/openssl/ossl_rand.c (ossl_rand_bytes): RAND_bytes could
+	  be return -1 as an error. Therefore, added error handling.
+	* ext/openssl/ossl_pkey_dsa.c (dsa_generate): ditto.
+
+Wed Oct 21 09:04:09 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/util.h: remove a warning suppression C4723
+	  (potential divide by zero) for VisualC++. It's meaningless.
+	  Before r26197, there is ruby_div0() in this place and it
+	  actually made divide by zero. But now it's just garbage.
+
+Wed Oct 21 08:23:36 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c: random_raw_seed don't use GRND_NONBLOCK. GRND_NONBLOCK
+	  mean the result might not have an enough cryptic strength and
+	  easy predictable. That's no good for SecureRandom.
+
+Sun Oct 18 17:26:53 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* common.mk: add a rule for explicit_bzero.o.
+	* configure.in: detect explicit_bzero and memset_s.
+	* include/ruby/missing.h: add explicit_bzero.
+	* random.c (init_randomseed): use explicit_bzero() instead of
+	  memset(). memset could be eliminated by compiler optimization.
+
+Mon Nov 30 18:46:44 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/readline/extconf.rb: call dir_config("libedit")
+	  if --enable-libedit is specified. [Bug #11751]
+	  patched by John Hein
+
+Mon Nov 30 08:44:29 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c: remove spurious #define for globals
+	  [ruby-core:71735] [Feature #11749]
+
+Sun Nov 29 09:13:03 2015  Conor Landry  <clandry94@gmail.com>
+
+	* NEWS: [DOC] Various grammar corrections and clarifications to
+	  increase readability.  [Fix GH-1115]
+
+Sat Nov 28 19:33:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_here_document): store dispatched result of
+	  on_tstring_content at the last fragment of a here document.
+
+Fri Nov 27 19:19:44 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (connect): detect closed connection and reconnect
+	  If the server closes a keep-alive http connection, the client socket
+	  reaches EOF. To avoid an EOFError, detect the closed connection and
+	  reconnect.
+	  Added test to ensure HTTP#post succeeds even if the
+	  keep-alive-connection has been closed by the server.
+	  by Kristian Hanekamp <kris.hanekamp@gmail.com>
+	  https://github.com/ruby/ruby/pull/1089 fix GH-1089
+
+Thu Nov 26 21:36:40 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): enable tail call
+	  optimization for specialized indexers.
+
+	* compile.c (iseq_compile_each): blockiseq should be NULL, but not
+	  Qnil.
+
+Thu Nov 26 17:22:53 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c, regenc.c, regexec.c, regint.h, enc/unicode.c:
+	  Merge Onigmo 58fa099ed1a34367de67fb3d06dd48d076839692
+	  + https://github.com/k-takata/Onigmo/pull/52
+
+Thu Nov 26 09:50:02 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test/coverage/test_coverage.rb: Added test-case for Coverage.restart.
+	  [Misc #11732][ruby-dev:49379]
+
+Thu Nov 26 09:46:36 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test/coverage/test_coverage.rb: Added test-case for Coverage.peek_result
+	  without Coverage.start. [Misc #11726][ruby-core:71622]
+
+Thu Nov 26 07:22:55 2015  Eric Wong  <e@80x24.org>
+
+	* test/openssl/test_ssl.rb (test_copy_stream): new test
+
+Wed Nov 25 21:23:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (copy_stream_body): try to_io conversion before read,
+	  readpartial, and write methods.  [ruby-dev:49008] [Bug #11199]
+
+Wed Nov 25 10:55:21 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* io.c (argf_getpartial): should not resize str if the second
+	  argument is not given.
+	  [ruby-core:71668] [Bug #11738]
+
+Tue Nov 24 23:56:25 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: On Solaris, it is safe to define _LARGEFILE_SOURCE
+	  when _FILE_OFFSET_BITS=64 is defined (= when 32-bit compile).
+
+Tue Nov 24 10:00:10 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems/installer.rb: Fix two double-word typos.
+	  [ci skip][fix GH-1108] Patch by @jwworth
+
+Tue Nov 24 09:17:02 2015  Alexander von Gluck IV  <kallisti5@unixzen.com>
+
+	* beos: Drop support for BeOS now that Haiku is stable.
+	  [Fix GH-1112]
+
+Tue Nov 24 09:16:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (rb_gc_for_fd): move to export, as referred by
+	  ext/socket.
+
+Tue Nov 24 09:04:29 2015  David Rodriguez  <deivid.rodriguez@gmail.com>
+
+	* file.c: [DOC] add a missing period to File docs, to terminate
+	  the sentence and separate from the next sentence.  [Fix GH-1111]
+
+Tue Nov 24 08:30:06 2015  JuanitoFatas  <katehuang0320@gmail.com>
+
+	* NEWS: Fix the issue number of `Struct#dig`, which should be
+	  [Feature #11688].  [Fix GH-1110]
+
+Tue Nov 24 07:56:54 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_s_accept): handle ENOMEM
+
+Tue Nov 24 07:50:15 2015  Eric Wong  <e@80x24.org>
+
+	* dir.c (dir_initialize): use rb_gc_for_fd for ENOMEM
+	* ext/socket/init.c (rsock_socket): ditto
+	* ext/socket/socket.c (rsock_socketpair): ditto
+	* internal.h (rb_gc_for_fd): prototype
+	* io.c (rb_gc_for_fd): remove static
+	  [ruby-core:71623] [Feature #11727]
+
+Tue Nov 24 06:46:27 2015  Eric Wong  <e@80x24.org>
+
+	* io.c (rb_gc_for_fd): new helper function
+	  (ruby_dup): use rb_gc_for_fd
+	  (rb_sysopen): ditto
+	  (rb_fdopen): ditto
+	  (rb_pipe): ditto
+	  [ruby-core:71623] [Feature #11727]
+
+Tue Nov 24 05:13:35 2015  Eric Wong  <e@80x24.org>
+
+	* ext/fiddle/function.c (struct nogvl_ffi_call_args):
+	  new struct for GVL release
+	  (nogvl_ffi_call): new function
+	  (function_call): adjust for GVL release
+	  [ruby-core:71642] [Feature #11607]
+	* ext/fiddle/closure.c (struct callback_args):
+	  new struct for GVL acquire
+	  (with_gvl_callback): adjusted original callback function
+	  (callback): wrapper for conditional GVL acquire
+	* ext/fiddle/depend: add dependencies
+	* ext/fiddle/extconf.rb: include top_srcdir for internal.h
+	* internal.h (ruby_thread_has_gvl_p): expose for fiddle
+	* vm_core.h (ruby_thread_has_gvl_p): moved to internal.h
+	* test/fiddle/test_function.rb (test_nogvl_poll): new test
+
+Mon Nov 23 19:53:12 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: On Solaris, with gcc, "-std=iso9899:1999"
+	  in $ansi_options is often also needed in CPPFLAGS,
+	  because some feature definitions vary depending on such
+	  standards options.
+
+Mon Nov 23 12:54:39 2015  Hamish Morrison  <hamishm53@gmail.com>
+
+	* configure.in: remove obsolete workarounds for Haiku.
+
+	* dln.c, file.c, io.c: remove obsolete Haiku workarounds.
+
+	* thread_pthread.c: add stack bounds detection for Haiku.
+
+	* signal.c: get stack pointer from signal context on Haiku.
+	  [ruby-core:67923] [Bug #10811] [Fix GH-1109]
+
+Mon Nov 23 11:44:11 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: bump version to minitest-5.8.3
+
+Mon Nov 23 08:55:00 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: fix wrong reference for r52714
+
+Sun Nov 22 22:23:37 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* gc.c (rb_raw_obj_info): fix compile errors when USE_RGENGC
+	  is 0.
+
+Sun Nov 22 21:58:09 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* lib/cmath.rb: methods which has suffix '!' are now deprecated.
+	  Re-apply r52469 made by Kazuki Tanaka, with fixing bug about
+	  mathn.rb compatibility. [ruby-core:68528] [Feature #10974]
+
+Sun Nov 22 19:36:51 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/openssl/ossl.c: fix brew command for installation of openssl.
+	  [ci skip][fix GH-1107] Patch by @arthurnn
+
+Sun Nov 22 17:59:50 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: On Solaris, add -D_XOPEN_SOURCE=n only when both
+	  AC_TRY_CPP and AC_TRY_COMPILE pass, because some options
+	  (e.g. -std=iso9899:1999) are not set when running C preprocessor
+	  or building ext.
+
+Sun Nov 22 16:53:34 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): remove unreachable code
+	  chunk after jump/leave.
+
+	* parse.y: move dead code elimination of logical operation to
+	  compile.c.  not to warn logical operation of literal constants.
+
+Sun Nov 22 16:37:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): eliminate always/never
+	  branches after a literal object and when the value is used after
+	  the branch.
+
+Sun Nov 22 01:23:43 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: Add -D_XOPEN_SOURCE=500 (or 600 or 700) on Solaris
+	  if available, mainly for enabling some features in sockets.
+
+Sun Nov 22 00:17:22 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/socket/test_socket.rb (test/socket/test_socket.rb): skip
+	  the test when Socket::SO_TIMESTAMP is not defined. Fix error
+	  on Solaris 10. [Bug #11728] [ruby-dev:49377]
+
+Sat Nov 21 18:57:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (need_argument): move frozen-string-literal-debug option
+	  from --enable to --debug.  [Feature #11725]
+
+	* ruby.c (proc_options): fix pointer overrun. do not advance argv
+	  until it is valid.
+
+Sat Nov 21 13:59:09 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/digest/sha1/extconf.rb: OpenSSL's struct name for SHA1 is
+	  SHA_CTX. http://openssl.org/docs/man0.9.8/crypto/SHA1.html
+
+Sat Nov 21 13:31:52 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/digest/*/*.[ch]: include ruby.h before digest.h to avoid
+	  including ext/digest/extconf.h. [Bug #3231]
+	  https://msdn.microsoft.com/library/36k2cdd4.aspx
+
+	* ext/digest/*/extconf.rb: remove ext/digest from include search path
+	  to avoid confusion of cl.exe.
+
+	* ext/digest/*/*.[ch]: explicitly specify def.h's path.
+
+Sat Nov 21 13:05:16 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl.h: LibreSSL doesn't have and need e_os2.h.
+
+Sat Nov 21 09:18:10 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_sync.c: reduce the specification of Queue#close.
+	  * Queue#close accepts no arguments.
+	  * deq'ing on closed queue returns nil, always.
+	  [Feature #10600]
+
+	* test/thread/test_queue.rb: catch up this fix.
+
+Sat Nov 21 08:44:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): add debug information to NODE_STR
+	  strings as default.
+	  [Feature #11725]
+
+	* insns.def (freezestring): add new instruction to support adding
+	  debug information for dynamically constructed strings.
+
+	* compile.c (iseq_compile_each): support adding debug information
+	  for NODE_DSTR with freezestring instruction.
+
+	* error.c (rb_error_frozen): change the debug information ID name
+	  id_debug_created_info and this field should have a 2 element array
+	  containing path and line information.
+
+	* defs/id.def: ditto.
+
+	* test/ruby/test_rubyoptions.rb: catch up this fix.
+
+	* test/ruby/test_iseq.rb: now frozen strings are not same.
+
+Sat Nov 21 04:34:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.c (rb_str_intern): should not freeze the receiver itself
+	  unexpectedly.  [ruby-core:71611] [Bug #11721]
+
+Fri Nov 20 23:15:18 2015  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/logger.rb: expose logger mutex
+	  [fix GH-541] Patch by @arthurnn
+
+Fri Nov 20 15:05:28 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* Added missing reference of GitHub
+
+Fri Nov 20 14:57:01 2015  Trevor Rowe  <trevorrowe@gmail.com>
+
+	* lib/net/http.rb: Fixed regression for Net::HTTP::PUT with "Expect-100"
+	  header. [fix GH-949]
+	* test/net/http/test_http.rb: added test.
+
+Fri Nov 20 14:39:56 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/http.rb: set hostname before call ossl_ssl_set_session.
+	  [Bug #11401][ruby-core:70152][fix GH-964] Patch by @mkarnebeek
+
+Fri Nov 20 12:53:19 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* array.c: clarify docs for take_while/drop_while samples.
+	  [ci skip][fix GH-1028] Patch by @leriksen
+
+Fri Nov 20 12:48:04 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/socket/socket.c: remove mention of :UNIX in getaddrinfo().
+	  It's typically not a support option.
+	  [ci skip][fix GH-990] Patch by @eam
+
+Fri Nov 20 12:44:06 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rss/syndication.rb: Add nodoc marker for #validate_sy_updatePeriod.
+	  [ci skip][fix GH-1105] Patch by @davydovanton
+
+Fri Nov 20 09:05:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_cref_replace_with_duplicated_cref): added.
+
+	  CREFs should not be shared by methods between `using'.
+	  [Bug #11247]
+
+	* vm_insnhelper.c (vm_cref_replace_with_duplicated_cref): ditto.
+
+	* vm.c (vm_cref_dup): should copy refinements correctly.
+
+	* eval.c: use rb_vm_cref_replace_with_duplicated_cref().
+
+	* eval_intern.h: add a decl. of
+	  rb_vm_cref_replace_with_duplicated_cref().
+
+	* vm_eval.c (eval_string_with_cref): do not need to pass
+	  scope's CREF because VM can find out CREF from stack frames.
+
+	* test/ruby/test_refinement.rb: add a test.
+
+Fri Nov 20 06:52:53 2015  Eric Wong  <e@80x24.org>
+
+	* .gitattributes: new file for git users
+	  [ruby-core:71578] [Feature #11713]
+
+Thu Nov 19 22:35:31 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/ancdata.c: Check buffer full and ignore MSG_TRUNC flag.
+	  buffer fullness is more robust to detect the message is too big for
+	  the buffer.
+	  AIX 7.1 recvmsg doesn't set MSG_TRUNC for rflags when MSG_PEEK is
+	  given.
+
+Thu Nov 19 21:55:11 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_start): force to invoke GC by GC.start
+	  even if it is GC.disable'd.
+
+	* test/ruby/test_gc.rb: add a test.
+
+Thu Nov 19 20:08:59 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: trivial performance improvements.
+
+	    name    modified
+	    vm1_gc_short_lived*                1.015
+	    vm1_gc_short_with_complex_long*    1.014
+	    vm1_gc_short_with_long*            1.000
+	    vm1_gc_short_with_symbol*          1.016
+	    vm1_gc_wb_ary*                     1.002
+	    vm1_gc_wb_ary_promoted*            0.996
+	    vm1_gc_wb_obj*                     1.045
+	    vm1_gc_wb_obj_promoted*            1.014
+	    vm3_gc                             1.021
+
+	* gc.c (gc_writebarrier_generational): reorder parameters to optimize
+	  register passing function call.
+
+	* gc.c (gc_writebarrier_incremental): ditto.
+
+	* gc.c (rb_gc_writebarrier): remove LIKELY().
+	  LIKELY() seems to move related functions not better places.
+
+Thu Nov 19 19:45:05 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_prog_init): [DOC] ARGV does not contain the name of
+	  the executable.  [ruby-core:71561] [Bug #11711]
+
+Thu Nov 19 15:53:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* signal.c: should also clear ruby_disable_gc.
+	  [Bug #11692]
+
+Thu Nov 19 15:31:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_compile_each): T_IMEMO/iseq objects should be
+	  wrap with ISeq wrappers. [Bug #11676]
+
+Thu Nov 19 15:16:12 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems: Update to RubyGems 2.5.0+ HEAD(c6b4946).
+	  this version includes #1114, #1314, #1322, #1375, #1383, #1387
+	* test/rubygems: ditto.
+
+Thu Nov 19 14:14:37 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (finish_overlapped_socket): return value of this
+	  function should be only 0 or SOCKET_ERROR.
+
+Thu Nov 19 14:12:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_tailcall_optimize): apply tail call optimization
+	  before conversion to specialized instructions.  when looking
+	  back from `leave` instruction, `send` instructions have been
+	  translated already.
+
+Thu Nov 19 13:57:58 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (finish_overlapped_socket): ignore EMSGSIZE when input,
+	  because POSIX platforms just do so.  fixes test errors revealed by
+	  r52647.
+
+Thu Nov 19 02:52:30 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): stretch the buffer size
+	  only when vmaxdatlen is nil.
+
+Thu Nov 19 02:20:11 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/socket/test_socket.rb (test_udp_recvmsg_truncation): rflags is
+	  nil on Solaris 10 which have no HAVE_STRUCT_MSGHDR_MSG_CONTROL.
+	  Reported by Naohisa Goto. [ruby-core:71557] [Bug #11709]
+
+Thu Nov 19 01:48:05 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in: add -static-libgcc for mingw automatically if available.
+
+Thu Nov 19 00:53:26 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/extmk.rb (--extflags): new option to pass EXTLDFLAGS to children,
+	  especially exts.mk.
+
+	* common.mk (EXTMK_ARGS): use above option.
+
+Wed Nov 18 22:50:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_class_clear_method_cache): should clear all
+	  RCLASS_CALLABLE_M_TBLs of all sub-classes (T_ICLASS).
+
+	  RCLASS_CALLABLE_M_TBL() caches complemented method entries.
+	  It should be cleared when the modules are cleared.
+	  On previous version clears only for direct children.
+	  It is enough for normal modules because corresponding T_ICLASSes
+	  are direct children.
+
+	  However, refinements create complex data structure. So that
+	  we need to clear all children (and descendants).
+	  [ruby-core:71423] [Bug #11672]
+
+	* vm_method.c (rb_clear_method_cache_by_class): rb_mKernel
+	  doesn't call rb_class_clear_method_cache, so that
+	  clear child T_ICLASSes.
+
+	* test/ruby/test_refinement.rb: enable disabled test.
+
+Wed Nov 18 21:09:08 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (prepare_callable_method_entry): use
+	  RCLASS_CALLABLE_M_TBL() instead of accessing a filed directly.
+
+Wed Nov 18 17:08:18 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: introduce the following field and macros.
+
+	  * rb_method_definition_t::complemented_count to count shared method
+	    entries because of complemented method entries and separate from
+	    alias_count.
+
+	    Shared `def' only by complemented method entries should not prevent
+	    method re-definition warning.
+
+	  * METHOD_ENTRY_COMPLEMENTED(me) to represent complemented method entry.
+	  * METHOD_ENTRY_COMPLEMENTED_SET(me) to check it as  complemented me.
+
+	* vm_insnhelper.c (aliased_callable_method_entry): should also
+	  check me->def->complemented_count.
+
+	* vm_method.c (method_definition_addref_complement): add to count
+	  complemented method entries number.
+
+	* vm_method.c (rb_method_definition_release): release `def' iff
+	  alias_count == 0 and complemented_count == 0.
+
+	* test/ruby/test_module.rb: add a test.
+
+Wed Nov 18 17:06:19 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_raw_obj_info): fix trivial issues.
+
+	  * support SPECIAL_CONSTs.
+	  * fix IMEMO/ment outputs.
+
+Wed Nov 18 11:32:15 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): eliminate always/never
+	  branches after a literal object.  this sequence typically
+	  appears by defined? operator for a method call on a local
+	  variable.
+
+Wed Nov 18 10:33:06 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): stretch the buffer size
+	  when EMSGSIZE occurs on non HAVE_STRUCT_MSGHDR_MSG_CONTROL platforms
+	  (such as, Windows).  fixes a test error revealed by r52625.
+
+Wed Nov 18 10:12:36 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): use 4096 as
+	  default size to match pre-r52610, which also maps to a common
+	  page size.
+
+Wed Nov 18 10:05:25 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* doc/syntax/refinements.rdoc: update documentation to reflect
+	  recent changes.
+	  [ci skip] [ruby-core:71466] [Misc #11681] Patch by James Adam
+
+Wed Nov 18 09:50:21 2015  Naotoshi Seo  <sonots@gmail.com>
+
+	* test/logger/test_logdevice.rb: Fix tests of logger to make it work on
+	  windows (windows can not remove opened file) [Bug #11702]
+
+Wed Nov 18 06:59:52 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): grow buffer
+	  on unspecified maxdatlen
+	  [ruby-core:71517] [Bug #11701]
+	* ext/socket/lib/socket.rb (Socket#recvmsg): nil default for dlen
+	  (Socket#recvmsg_nonblock): ditto
+	* test/socket/test_socket.rb (test_recvmsg_udp_no_arg): new test
+
+Tue Nov 17 19:50:06 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (fstat): declare for mingw.
+
+Tue Nov 17 19:02:59 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in (BASERUBY): use Kernel#print instead of Kernel#p because
+	  the baseruby may output CRLF as end of line.
+
+Tue Nov 17 15:34:34 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+	* NEWS: Added update from Unicode 7.0.0 to 8.0.0 [ci skip]
+
+Tue Nov 17 15:30:30 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+	* enc/unicode/casefold.h, name2ctype.h: Change Unicode
+	  Version for regular expressions from 7.0.0 to
+	  8.0.0 (with help from Kimihito Matsui) [Feature #11563]
+
+Tue Nov 17 14:36:00 2015  Kenichi Kamiya  <kachick1@gmail.com>
+
+	* lib/ostruct.rb (dig): Implement OpenStruct#dig
+	  [Feature #11688]
+
+Tue Nov 17 14:04:14 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/lib/socket.rb (Socket#recvmsg{,_nonblock}): default values
+	  of clen must be nil.
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): handle nil of clen.
+	  fixes test errors introduced at r52602.
+
+Tue Nov 17 13:43:46 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/lib/socket.rb: UNIXSocket is not always exists.  fixes
+	  install error on Windows, introduced at r52601.
+
+Tue Nov 17 11:27:23 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/lib/socket.rb (Socket#recvfrom_nonblock):
+	  UDPSocket#recvfrom_nonblock):
+	  update doc for `exception: false` and destination buffer
+	  [ruby-core:69542] [Feature #11229]
+	  [ruby-core:69543] [Feature #11242]
+
+Tue Nov 17 11:25:05 2015  Eric Turner  <ericturnerdev@gmail.com>
+
+	* array.c (rb_ary_dig), hash.c (rb_hash_dig): [DOC] Update
+	  comments describing dig methods.  [Fix GH-1103]
+
+	* struct.c (rb_struct_dig): [DOC] add rdoc.
+
+Tue Nov 17 11:22:22 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+	* NEWS: Small grammatical fix [ci skip]
+
+Tue Nov 17 10:12:30 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/lib/socket.rb (Socket.accept_loop): avoid exceptions
+	  (Socket.udp_server_recv): ditto
+
+Tue Nov 17 09:59:00 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): avoid arg parsing
+	  [ruby-core:71439] [Feature #11339]
+	  (rsock_bsock_sendmsg): make private, adjust for above
+	  (rsock_bsock_sendmsg_nonblock): ditto
+	* ext/socket/rubysocket.h: adjust prototypes
+	  (rsock_opt_false_p): remove
+	* ext/socket/basicsocket.c (rsock_init_basicsocket):
+	  define private methods
+	* ext/socket/lib/socket.rb (BasicSocket#sendmsg): new wrapper
+	  (BasicSocket#sendmsg_nonblock): ditto
+
+Tue Nov 17 09:45:18 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): avoid arg parsing
+	  (rsock_bsock_recvmsg): adjust for above change
+	  (rsock_bsock_recvmsg_nonblock): ditto
+	  [ruby-core:71439] [Feature #11339]
+	* ext/socket/rubysocket.h: adjust prototypes for above
+	* ext/socket/basicsocket.c (rsock_init_basicsocket):
+	  adjust private methods
+	* ext/socket/lib/socket.rb (BasicSocket#recvmsg): wrapper method
+	  (BasicSocket#recvmsg_nonblock): ditto
+
+Tue Nov 17 08:36:34 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_s_accept_nonblock): avoid parsing args
+	  [ruby-core:71439] [Feature #11339]
+	* ext/socket/rubysocket.h: adjust prototype
+	* ext/socket/socket.c (sock_accept_nonblock): make private
+	* ext/socket/tcpserver.c (tcp_accept_nonblock): ditto
+	* ext/socket/unixserver.c (unix_accept_nonblock): ditto
+	* ext/socket/lib/socket.rb (Socket#accept_nonblock):
+	  implement as wrapper, move RDoc
+	  (TCPServer#accept_nonblock): ditto
+	  (UNIXServer#accept_nonblock): ditto
+
+Tue Nov 17 08:25:57 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/socket.c (sock_connect_nonblock):
+	  avoid argument parsing in C.
+	  [ruby-core:71439] [Feature #11339]
+	* ext/socket/lib/socket.rb (Socket#connect_nonblock):
+	  new wrapper for private method, move RDoc
+
+Tue Nov 17 08:16:09 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_s_recvfrom_nonblock):
+	  avoid arg parsing with C API
+	  [ruby-core:71439] [Feature #11339]
+	* ext/socket/basicsocket.c (bsock_recv_nonblock):
+	  adjust for above change, make private
+	* ext/socket/socket.c (sock_recvfrom_nonblock): ditto
+	* ext/socket/udpsocket.c (udp_recvfrom_nonblock): ditto
+	* ext/socket/lib/socket.rb (BasicSocket#recv_nonblock):
+	  new wrapper for private method, move RDoc
+	  (Socket#recvfrom_nonblock): ditto
+	  (UDPSocket#recvfrom_nonblock): ditto
+
+Mon Nov 16 21:27:54 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/dtrace/helper.rb (Dtrace::TestCase#trap_probe): dtrace buffer
+	  size is set as 8m on Solaris (default 4m). [Bug #11697]
+
+Mon Nov 16 20:03:14 2015  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/logger.rb: Add Logger#reopen
+
+Mon Nov 16 18:21:52 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_dig): dig in nested structs too.
+
+	* struct.c (rb_struct_dig): new method Struct#dig.
+	  [Feature #11688]
+
+Mon Nov 16 17:41:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): optimize tail calls on aref
+	  and aset specialized instructions.
+
+	* compile.c (iseq_peephole_optimize): optimize replaced leave
+	  instruction copied to jump instruction too.
+
+Mon Nov 16 16:39:38 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb: Enable frozen_string_literal.
+
+	* lib/set.rb: Move << out of the begin block that ensures pop.
+
+Mon Nov 16 16:28:30 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Hash#flatten!, #add?, #delete?, #collect!, #reject!,
+	  #select!, #^, #classify): Micro-optimize some methods for
+	  performance and readability.
+
+Mon Nov 16 16:17:58 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: fixed accidentally commit.
+
+Mon Nov 16 16:10:51 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* mkconfig.rb: Add some high-level documentation.
+	  [ci skip][fix GH-1081] Patch by @ulfalizer
+
+Mon Nov 16 15:59:14 2015  yui-knk  <spiketeika@gmail.com>
+
+	* proc.c: Add call-seq of `Method#super_method`
+	  [ci skip][fix GH-1094]
+
+Mon Nov 16 15:58:39 2015  Kenichi Kamiya  <kachick1@gmail.com>
+
+	* struct.c: Standardize a method signature of Struct#[]=.
+	  [ci skip][fix GH-1095]
+
+Mon Nov 16 15:42:36 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (#>=, #>, #<=, #<): Make use of Hash#>=, #>, #<, and
+	  #<= when comparing against an instance of the same kind.
+
+Mon Nov 16 15:37:11 2015  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/logger.rb: Support symbol and string log level setting
+
+Mon Nov 16 15:33:11 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* tool/rbinstall.rb: fix wrong permission for gem specification without
+	  zlib runtime. [Bug #11685][ruby-dev:49343]
+
+Mon Nov 16 12:11:11 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/webrick/httpauth/basicauth.rb: fix a typo.
+	  [ci skip][fix GH-1099] Patch by @jwworth
+	* lib/webrick/httpauth/digestauth.rb: ditto.
+
+Sun Nov 15 18:28:43 2015  Kenichi Kamiya  <kachick1@gmail.com>
+
+	* vm_method.c (set_method_visibility): should fail if the receiver
+	  is frozen.  [ruby-core:71489] [Bug #11687]
+
+Sat Nov 14 22:15:07 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/lib/socket.rb: Specify frozen_string_literal: true.
+
+Sat Nov 14 21:44:56 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb: Use "<<" to reduce string allocation.
+
+Sat Nov 14 17:45:49 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tsort.rb: Specify frozen_string_literal: true.
+
+Sat Nov 14 17:25:15 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv-replace.rb: Specify frozen_string_literal: true.
+
+Sat Nov 14 17:00:13 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb: Specify frozen_string_literal: true.
+
+Sat Nov 14 16:43:02 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb: Specify frozen_string_literal: true.
+
+Sat Nov 14 05:04:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* node.h: remove old comments.
+
+Sat Nov 14 04:55:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* refactoring CREF related code.
+
+	* eval_intern.h: remove unused setter functions.
+	  CREF_CLASS_SET()
+	  CREF_NEXT_SET()
+	  CREF_SCOPE_VISI_COPY()
+
+	* eval_intern.h: rename flags:
+	  * NODE_FL_CREF_PUSHED_BY_EVAL_ -> CREF_FL_PUSHED_BY_EVAL
+	  * NODE_FL_CREF_OMOD_SHARED_ -> CREF_FL_OMOD_SHARED
+	  and use IMEMO_FL_USER1/2.
+
+	* vm.c (vm_cref_new): accept push_by_eval parameter.
+
+	* vm.c (vm_cref_new_use_prev): added for rb_vm_rewrite_cref().
+
+	* vm_insnhelper.c (vm_cref_push): accept pushed_by_eval parameter.
+
+	* vm_insnhelper.h: remove unused macros:
+	  COPY_CREF_OMOD() and COPY_CREF().
+
+	* vm_eval.c, insns.def: catch up this fix.
+
+Sat Nov 14 02:58:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_define_method): refactoring.
+	  * get CREF in this function.
+	  * cbase is no longer needed (CREF_CLASS(cref) is enough).
+
+	* compile.c: RubyVM::FrozenCore.define_method only accept 2 args.
+
+Sat Nov 14 02:34:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_define_method): do not use current CREF immediately,
+	  but check CREF in environment or methods. Methods defined in methods
+	  should be public.
+	  [Bug #11571]
+
+	* vm_method.c (rb_scope_module_func_check): check CREF in env or me.
+	  if CREF is contained by `me', then return FALSE.
+
+	* test/ruby/test_method.rb: add a test.
+
+Sat Nov 14 02:19:16 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: constify rb_cref_t::scope_visi;
+
+	* eval_intern.h (CREF_SCOPE_VISI_COPY): catch up this fix.
+
+	* vm_method.c: ditto.
+
+Sat Nov 14 01:53:52 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* pack.c (pack_unpack, AVOID_CC_BUG): Very ugly workaround for
+	  optimization bug of Oracle Solaris Studio 12.4 on Solaris
+	  with -xO4 optimization option. [Bug #11684]
+
+Fri Nov 13 23:00:23 2015  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* configure.in: unset LD_PRELOAD on mingw. msys2 child processes
+	  crash at make test-all with LD_PRELOAD.
+	  [ruby-core:71461] [Bug #11680]
+
+Fri Nov 13 14:00:43 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/openssl/ossl_pkey.c: Merge ruby/openssl@b9ea8ef [Bug #10735]
+
+Fri Nov 13 13:09:16 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c: Merge ruby/openssl@81e1a30
+
+	* test/openssl/test_ssl.rb: ditto
+
+Fri Nov 13 13:05:37 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* prelude.rb (Thread.exclusive): warn as deprecated.
+
+Fri Nov 13 10:36:39 2015  Victor Nawothnig  <Victor.Nawothnig@gmail.com>
+
+	* parse.y (new_unless): optimize constant condition for `unless`
+	  as well as `if`.  [Fix GH-1092]
+
+Fri Nov 13 10:08:41 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/psych.gemspec: bump version to 2.0.15
+
+Thu Nov 12 18:44:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_magic_comment): should match exactly.
+	  [ruby-core:71460] [Bug #11679]
+
+Thu Nov 12 16:16:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/prelude.c.tmpl: enable tail call optimization.
+
+Thu Nov 12 14:17:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): ANDDOT at the head of the line denote
+	  line continuation from previous one to support fluent interface,
+	  as well as single dot.
+
+Thu Nov 12 13:49:50 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems: Update to RubyGems 2.5.0+ HEAD(db78980).
+	  this version includes #1367 , #1373 , #1375
+	* test/rubygems: ditto.
+
+Thu Nov 12 10:53:41 2015  Eric Wong  <e@80x24.org>
+
+	* benchmark/bm_io_nonblock_noex2.rb: new benchmark based
+	  on bm_io_nonblock_noex.rb
+	* io.c (io_read_nonblock): move documentation to prelude.rb
+	  (io_write_nonblock): ditto
+	  (Init_io): private, internal methods for prelude.rb use only
+	* prelude.rb (IO#read_nonblock): wrapper + documentation
+	  (IO#write_nonblock): ditto
+	  [ruby-core:71439] [Feature #11339]
+
+Wed Nov 11 18:30:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): look up the key, then get default
+	  value and raise KeyError if the returned value is nil.
+	  [ruby-dev:49338] [Ruby trunk - Bug #11677]
+
+Wed Nov 11 17:38:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (local_var_list_add): skip internal local variable
+	  name by its type but not if it has a name.  internal local
+	  variable names are just unique per frame, not globally.
+	  [ruby-core:71437] [Bug #11674]
+
+Wed Nov 11 14:14:33 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* transcode.c: fix a typo
+	[ci skip][fix GH-1091] Patch by @jwworth
+
+Wed Nov 11 11:58:38 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (initialize): Connections are in passive mode per
+	  default now.  The default mode can be changed by
+	  Net::FTP.default_passive=.
+	  [ruby-core:71146] [Feature #11612]
+
+	* lib/net/ftp.rb (default_passive=, default_passive): new methods.
+
+Wed Nov 11 09:03:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): respect default value of a hash.  no
+	  longer raises KeyError unless the default value of the hash is
+	  nil.  [ruby-core:71354] [Bug #11661]
+
+Tue Nov 10 20:35:12 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: Remove indicator for "frozen_string_literal: true".
+
+	* lib/pp.rb: Ditto.
+
+	* lib/prettyprint.rb: Ditto.
+
+	* lib/resolv.rb: Ditto.
+
+	* lib/securerandom.rb: Ditto.
+
+	* lib/tmpdir.rb: Ditto.
+
+	* lib/unicode_normalize/tables.rb: Ditto.
+
+	* test/net/ftp/test_buffered_socket.rb: Ditto.
+
+	* test/net/ftp/test_mlsx_entry.rb: Ditto.
+
+	* test/open-uri/test_open-uri.rb: Ditto.
+
+	* test/open-uri/test_ssl.rb: Ditto.
+
+	* test/pathname/test_pathname.rb: Ditto.
+
+	* test/test_pp.rb: Ditto.
+
+	* test/test_prettyprint.rb: Ditto.
+
+	* tool/transcode-tblgen.rb: Ditto.
+
+	* ext/pathname/lib/pathname.rb: Ditto.
+
+Tue Nov 10 18:42:24 2015  Aleksandrs Ledovskis  <aleksandrs@ledovskis.lv>
+
+	* defs/id.def, parse.y: Switch internal token name to reflect
+	  current form of safe-call operator.  [Fix GH-1090]
+
+Tue Nov 10 18:25:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_to_proc): use rb_func_proc_new to make light
+	  weight proc.  [Feature #11653]
+
+Tue Nov 10 18:23:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (cfunc_proc_t): add room for me.
+
+	* proc.c (cfunc_proc_new): generalise for cfunc proc without env.
+
+	* proc.c (rb_func_proc_new, rb_func_lambda_new): new functions to
+	  make proc/lambda without env from cfunc.
+
+Tue Nov 10 17:32:35 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* bootstraptest/test_fork.rb ([ruby-dev:37934]): :NPROC (RLIMIT_NPROC)
+	  is not supported on some platforms (e.g. Solaris 10).
+
+Tue Nov 10 16:57:14 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_to_proc): new method Hash#to_proc.
+	  [Feature #11653]
+
+Tue Nov 10 14:34:09 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (rb_time_timespec_new): swap utc and localtime
+	  to generate gmt flag by INT_MAX - gmtoff.
+
+Tue Nov 10 14:01:59 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_{le,lt,ge,gt}): new methods, Hash#<=, Hash#<,
+	  Hash#>=, Hash#>, to test if all elements of a hash are also
+	  included in another hash, and vice versa.
+	  [ruby-core:68561] [Feature #10984]
+
+Tue Nov 10 11:25:29 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (rb_timespec_now): added.  [Feature #11558]
+
+	* time.c (rb_time_timespec_new): added.  [Feature #11558]
+
+Tue Nov 10 06:17:17 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (rb_autoload_load): allow recursive calls
+	  [ruby-core:71345] [Bug #11658]
+	* test/ruby/test_autoload.rb (test_autoload_while_autoloading):
+	  new test by: Hiroshi Shirosaki <h.shirosaki@gmail.com>
+	  [ruby-core:71390]
+
+Tue Nov 10 00:36:46 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Message::MessageEncoder#put_labels):
+	  Prevent overflow of pointer to labels.
+	  Patch by Hannes Georg.  [ruby-core:71248] [Bug #11632]
+
+Tue Nov 10 00:25:41 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gems/bundled_gems: update to power_assert 0.2.6.
+
+Mon Nov  9 21:48:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_check_funcall_default): split from
+	  rb_check_funcall to return the given fallback value.
+
+	* object.c (rb_obj_dig): use rb_check_funcall_default so that tail
+	  call optimization will be possible.  [Feature #11643]
+
+Mon Nov  9 21:27:23 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_dig): new method Array#dig.
+
+	* hash.c (rb_hash_dig): new method Hash#dig.
+
+	* object.c (rb_obj_dig): dig in nested arrays/hashes.
+	  [Feature #11643]
+
+Mon Nov  9 18:00:47 2015  Yuki Nishijima  <mail@yukinishijima.net>
+
+	* gems/bundled_gems: Upgrade the did_you_mean gem to 1.0.0.beta3
+
+Mon Nov  9 17:38:14 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/runner.rb: use official repository for coverage tool.
+	* Makefile.in: ditto.
+	* common.mk: ditto.
+	* .gitignore: ignored third party repositories.
+
+Mon Nov  9 17:29:09 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): Dynamic string literals should be
+	  frozen.
+	  [ruby-core:57574] [Feature #8976]
+
+Mon Nov  9 15:56:07 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* common.mk: Use ruby organization url for simplecov repository.
+
+Sun Nov  8 16:24:09 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* NEWS: describe addition of File::TMPFILE
+
+Sun Nov  8 15:19:17 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* file.c: Add O_TMPFILE.
+
+Sun Nov  8 14:24:43 2015  windwiny  <windwiny.ubt@gmail.com>
+
+	* method.h (METHOD_ENTRY_{VISI,BASIC,FLAGS}_SET): suppress
+	  shift-op-parentheses warnings.  [Fix GH-1082]
+
+Sun Nov  8 14:01:22 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/psych_emitter.c: backport 5bd7744 from tenderlove/psych.
+	  support backward compatibility of Ruby 2.0
+
+Sun Nov  8 10:55:10 2015  Anton Davydov  <antondavydov.o@gmail.com>
+
+	* io.c (rb_io_gets_m): Update IO#gets doc for characters more than
+	  1 byte.  [Fix GH-1085]
+
+Sun Nov  8 10:37:58 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/uri/ftp.rb: fix a typo.
+	  [fix GH-1084][ci skip] Patch by @windwiny
+
+Sun Nov  8 08:10:31 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (exec_hooks_precheck): check need_clean everytime
+	  to clean-up unused hooks.
+
+	* vm_trace.c (list->need_clean): use as boolean value.
+
+Sun Nov  8 01:31:27 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#initialize):
+	  default value of Net::HTTP#open_timeout is now 60 (was nil).
+
+Sat Nov  7 12:18:05 2015  Eric Wong  <e@80x24.org>
+
+	* string.c (id_to_s): remove redundant variable
+	  (rb_obj_as_string): trade id_to_s for idTo_s
+	  (rb_str_equal): replace rb_intern(...) with pre-defined ID
+	  (rb_str_cmp_m): ditto
+	  (rb_str_match): ditto
+	  (str_upto_each): ditto
+	  (rb_str_sum): ditto
+	  (Init_String): remove id_to_s initialization
+
+Sat Nov  7 11:40:05 2015  Eric Wong  <e@80x24.org>
+
+	* thread.c (rb_cThreadShield): make static
+
+Sat Nov  7 09:51:38 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_threadptr_exec_event_hooks_orig):
+	  maintain trace_running counter on internal events.
+
+	  This patch is made by Takashi Kokubun <takashikkbn@gmail.com>.
+	  [Bug #11603] https://github.com/ruby/ruby/pull/1059
+
+Sat Nov  7 03:32:27 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h (RSTRUCT_PTR): need a close parenthesis.
+
+Sat Nov  7 01:32:06 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* dir.c (dir_fileno, dirfd): support of Dir#fileno on Solaris 10.
+	  Solaris 10 does not have dirfd, but the file descriptor of a
+	  directory is stored in the d_fd or dd_fd member in the DIR struct.
+	  Note that Solaris 11 has dirfd(3C).
+
+	* configure.in: checks for DIR.d_fd and DIR.dd_fd on Solaris 10.
+
+Fri Nov  6 23:13:53 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* array.c: clarifies Array#reject! documentation.
+	  [fix GH-894][ci skip] Patch by @GxSplinter
+
+Fri Nov  6 20:18:25 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/runner.rb: extracted test helper.
+	* test/lib/zombie_hunter.rb: ditto.
+
+Fri Nov  6 18:07:47 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/ruby.h (rb_array_const_ptr, rb_struct_const_ptr):
+	  Suppress pointer type mismatch warnings occurred with old version
+	  of Fujitsu C Compiler (fcc) on Solaris 10. The warnings cause
+	  failure of TestMkmf::TestConvertible. [Bug #11644] [ruby-dev:49326]
+	* include/ruby/ruby.h (FIX_CONST_VALUE_PTR): macro for the above,
+	  only effective with fcc.
+
+Fri Nov  6 12:39:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/id.def (token_ops), parse.y (parser_yylex): change DOTQ
+	  from ".?" to "&.".  [ruby-core:71363] [Feature #11537]
+
+Fri Nov  6 09:01:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (kwd_append): fix segv after invalid keyword argument,
+	  preceding keyword list is NULL when syntax error is there.
+	  [ruby-core:71356] [Bug #11663]
+
+Fri Nov  6 06:59:37 2015  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_autoload: hoist out ruby_impl_require
+
+Thu Nov  5 13:03:58 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/id.def (token_ops): gather associations between IDs,
+	  operators, and parser tokens.
+
+Thu Nov  5 10:17:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (make_addrinfo): use RARRAY_ASET for
+	  write-barrier.
+
+	* ext/tk/tcltklib.c ({call,eval,invoke}_queue_handler): ditto.
+
+	* ext/tk/tkutil/tkutil.c (ary2list, ary2list2): ditto.
+
+Thu Nov  5 10:09:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/**/*.c: prefer RARRAY_AREF to indexing RARRAY_CONST_PTR.
+	  pointed out by hanmac.
+	  https://github.com/ruby/ruby/commit/3553a86#commitcomment-14187670
+
+Wed Nov  4 17:33:24 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/debug.rb: Add documentation for #thread_list_all.
+	  [Misc #11580][ci skip]
+
+Wed Nov  4 15:45:59 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* class.c: fix documentation for rb_define_class{_id}_under.
+	  [fix GH-991][ci skip] Patch by @kachick
+
+Wed Nov  4 15:40:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* method.h: fix typo. Patch by @davydovanton
+	  [fix GH-1076][ci skip]
+
+Wed Nov  4 15:39:32 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* hash.c: use correct grammar. Patch by @tveastman
+	  [fix GH-1079][ci skip]
+
+Wed Nov  4 11:38:23 2015  Jake Worth  <jakeworth82@gmail.com>
+
+	* process.c (proc_getsid): [DOC] Fix double word 'for' and typo.
+	  [Fix GH-1080]
+
+Wed Nov  4 06:01:52 2015  Eric Wong  <e@80x24.org>
+
+	* include/ruby/ruby.h (struct RObject): hide iv_index_tbl type
+	  [ruby-core:71306] [Feature #11647]
+
+Tue Nov  3 06:48:58 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (find_class_path): remove cast for rb_class_ivar_set
+	  (rb_ivar_set): ditto
+	  (rb_cvar_set): ditto
+
+Tue Nov  3 06:18:21 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (rb_global_tbl): convert to id_table
+
+Tue Nov  3 01:58:46 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* parse.y (NO_QCALL): fix type mismatch of operands that causes
+	  compile error with Oracle Solaris Studio on Solaris.
+	  [Bug #11645] [ruby-dev:49327]
+
+Sun Nov  1 17:14:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* id_table.c (mix_id_table_insert): do not touch list during
+	  list->hash transition because GC can run during transition.
+
+Sun Nov  1 11:07:31 2015  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_memsize): account for rb_call_cache entries
+
+Sun Nov  1 09:12:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_yylex): ':' separated by a comment and a newline
+	  is not valid as symbol.
+
+Sat Oct 31 20:15:48 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/openssl/test_pair.rb: skipped tests if openssl doesn't support
+	  ECDH cipher.
+
+Sat Oct 31 14:58:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* man/ruby.1 (SYNOPSIS): remove extraneous space for -F option as
+	  it does not allow spaces before its argument.
+	  [ruby-core:71283] [Bug #11641]
+
+Sat Oct 31 14:58:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* man/ruby.1 (SYNOPSIS): remove extraneous space for -F option as
+	  it does not allow spaces before its argument.
+	  [ruby-core:71283] [Bug #11641]
+
+Sat Oct 31 10:22:49 2015  yui-knk  <spiketeika@gmail.com>
+
+	* eval_error.c (undef_mesg_for): fix typo.  Before this commit
+	  `ArgumentError: malformed format string - %$` was raised when
+	  `NameError#message` is called.  [ruby-core:71282] [Bug #11640]
+	  [Fix GH-1077]
+
+Fri Oct 30 21:12:45 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gems/bundled_gems: update to power_assert 0.2.5.
+
+Fri Oct 30 19:29:52 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (newobj_slowpath): do not need to use flags hack (commit miss).
+
+Fri Oct 30 19:08:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_get_freeobj_from_next_freepage): not so UNLIKELY.
+
+Fri Oct 30 18:09:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (newobj_slowpath): reduce 1 parameter to use only registers
+	  for performance.
+
+	  On my laptop, 'N.times{x = []}' (where N = 29_000_000) is
+	  1.86 sec -> 1.74 sec.
+
+Fri Oct 30 12:53:21 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test/ruby/test_call.rb: added test for safe navigation operator.
+	  [fix GH-1066]
+
+Fri Oct 30 12:47:34 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: fix wrong commit name.
+
+Fri Oct 30 12:36:16 2015  yui-knk  <spiketeika@gmail.com>
+
+	* vm_method.c: added documentation of protected/private methods.
+	  [fix GH-1072]
+	* test/ruby/test_module.rb: added testcase for method_defined?
+	  [fix GH-1071]
+
+Fri Oct 30 12:06:59 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_class_ivar_set): rename as class specific ivar
+	  setter, and st_table is no longer involved.
+
+Fri Oct 30 11:36:33 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (generic_ivar_remove): adjust type, set valp
+	  (rb_obj_remove_instance_variable): simplify call
+	* test/ruby/test_object.rb (test_remove_instance_variable):
+	  expand for implementation details
+
+Fri Oct 30 10:37:56 2015  Eric Wong  <e@80x24.org>
+
+	* internal.h (rb_st_insert_id_and_value): update prototype
+	* variable.c (rb_st_insert_id_and_value): reduce args
+	  (find_class_path): adjust call for less args
+	  (rb_ivar_set): ditto
+	  (rb_cvar_set): ditto
+	* class.c (rb_singleton_class_attached): ditto
+
+Fri Oct 30 09:57:22 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: update latest gems.
+	  test-unit-3.1.5 and minitest-5.8.2
+
+Fri Oct 30 09:54:05 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems: Update to RubyGems HEAD(60d7972).
+	  this version contains pull requests number of #1343, #1356, #1357, #1363
+	  at https://github.com/rubygems/rubygems/pulls
+	* test/rubygems: ditto.
+
+Fri Oct 30 07:38:29 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (getinlinecache/setinlinecache): compare ic->ic_cref and
+	  current cref only when cached CREF list includes singleton class.
+
+	  Singleton classes have own namespaces, so that we need to check
+	  cref as a key (#10943).
+
+	  However, if current CREF list does not include singleton class,
+	  no need to check CREF because it should be same name space.
+
+	* vm_insnhelper.c (vm_get_const_key_cref): add a function returns
+	  CREF only when it includes singleton class.
+
+	* vm_core.h: constify iseq_inline_cache_entry::ic_cref.
+
+Fri Oct 30 06:43:50 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_env_cref): make it inline for performance.
+
+	* vm_insnhelper.c (rb_vm_get_cref): use NULL instead of 0.
+
+Fri Oct 30 06:20:40 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def: nobody set ic->ic_value.value to Qundef.
+
+Fri Oct 30 06:15:50 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: add ifndef guard for VM_CHECK_MODE.
+
+Fri Oct 30 06:13:10 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_check_frame_detail): should require me for
+	  VM_FRAME_FLAG_BMETHOD type frame.
+
+Thu Oct 29 18:42:30 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_ptr): specify NOINLINE so that gc_mark() can return
+	  immediately when obj is not a markable object.
+
+Thu Oct 29 18:05:22 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* encoding.c (rb_enc_check_str): add for performance.
+	  This function only accepts T_STRING (and T_REGEXP).
+
+	  This patch improves performance of a tiny_segmenter benchmark
+	  (num=2) 2.54sec -> 2.42sec on my machine.
+	  https://github.com/chezou/TinySegmenter.jl/blob/master/benchmark/benchmark.rb
+
+	* encoding.c: add ENC_DEBUG and ENC_ASSERT() macros.
+
+	* internal.h: add a decl. of rb_enc_check_str().
+
+	* string.c (rb_str_plus): use rb_enc_check_str().
+
+	* string.c (rb_str_subpat_set): ditto.
+
+Thu Oct 29 17:16:40 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: export rb_wb_(un)protected_newobj_of()
+	  because some extensions include internal.h.
+
+Thu Oct 29 16:42:19 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_imemo_new): should not pass FL_WB_PROTECTED flag.
+
+	* gc.c (rb_wb_protected_newobj_of): add more assertions.
+
+	* gc.c (rb_wb_unprotected_newobj_of): ditto.
+
+Thu Oct 29 16:20:26 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: introduce rb_wb_unprotected_newobj_of() and
+	  rb_wb_protected_newobj_of(), pass the WB_PROTECTED
+	  information explicitly.
+
+	* internal.h: use introduced functions by NEWOBJ_OF().
+	  `flag' is immediate value, so that C compilers can
+	  solve them at compile time.
+
+	* include/ruby/ruby.h: add a comment about that.
+
+Thu Oct 29 14:52:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add rb_objspace::flags::has_hook to represent hook availability.
+
+	* gc.c: add gc_event_hook_available_p(objspace) to check that flag.
+
+	* gc.c (newobj_of): use gc_event_hook_available_p() instead of
+	  checking gc_event_hook_needed_p(objspace, RUBY_INTERNAL_EVENT_NEWOBJ).
+	  for performance.
+
+	* gc.c (newobj_init): add UNLIKELY() for FL_WB_PROTECTED flag.
+
+	* gc.c (newobj_init): change parameters order (trivial change).
+
+Thu Oct 29 14:45:15 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_thread_struct): move forward declarations before
+	  used.
+
+Thu Oct 29 14:07:54 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_ptr): remove debug code for #11244.
+
+Thu Oct 29 10:08:33 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (struct autoload_state): usable as wait-queue head
+	  (struct autoload_data_i): remove 2 words of overhead
+	  (autoload_i_mark): remove marking for thread
+	  (autoload_reset): adjust for struct changes
+	  (rb_autoload): ditto
+	  (rb_autoloading_value): ditto
+	  (rb_autoload_load): ditto
+	  (const_update): ditto
+
+Thu Oct 29 08:48:05 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (struct autoload_data_i): add waitq_head
+	  (struct autoload_state): new struct
+	  (autoload_require): save result in autoload_state for use
+	    in autoload_reset
+	  (autoload_reset): wake up other waiters with open-coded
+	    wait-queues
+	  (rb_autoload_load): add ensure autoload_const_set happens
+	    atomically w.r.t. autoload-triggered "require"
+	  [ruby-core:70075] [ruby-core:71239] [Bug #11384]
+
+Wed Oct 29 00:39:50 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/rubygems/test_gem_commands_server_command.rb
+	  (test_handle_options_port): change port from http to discard.
+	  Solaris does not include "http 80/tcp" in its default
+	  /etc/inet/services. AFAIK, discard (9/tcp) is older than http
+	  and it is expected that all OS can resolve the service name.
+	  [Bug #10004] [ruby-core:63518]
+
+Wed Oct 28 23:52:48 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* probes_helper.h (RUBY_DTRACE_HOOK): add RB_GC_GUARD, though paranoic.
+
+Wed Oct 28 15:36:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_name_err_new): store the receiver directly.
+
+	* error.c (name_err_receiver): return directly stored receiver.
+	  [Feature #10881]
+
+	* error.c (name_err_mesg_to_str): quote the name if unprintable.
+
+	* object.c (check_setter_id): use rb_check_id to convert names.
+
+	* variable.c (uninitialized_constant): use NameError::message to
+	  keep the receiver of uninitialized constant.  [Feature #10881]
+
+	* error.c (rb_name_err_new): new function to create NameError
+	  exception instance.  [Feature #10881]
+
+Wed Oct 28 13:29:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (new_attr_op_assign): fix op_assign type, which is
+	  already an ID since r52284.  [Feature #11537]
+
+Tue Oct 27 23:14:14 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/id.def: enable anonymous IDs not to expose internal IDs for
+	  frozen-string-literal-debug by Marshal.dump.
+
+Tue Oct 27 17:06:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/id.def: move internal IDs for frozen-string-literal-debug.
+
+Tue Oct 27 16:41:05 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_error_frozen_object): use rb_attr_get instead of
+	  rb_ivar_get to get rid of warnings for string objects created
+	  when frozen-string-literal-debug is disabled.
+
+Tue Oct 27 16:18:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/logger.rb (Logger::Period#previous_period_end): as weekly
+	  rotation shifts the log file on Sundays, the end date of the
+	  previous period should be Saturdays.  fix r45072.
+	  [ruby-dev:49314] [Bug #11622]
+
+Tue Oct 27 16:12:37 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c (rb_print_backtrace): our addr2line doesn't work on sparc.
+	  http://rubyci.s3.amazonaws.com/unstable11s/ruby-trunk/log/20151027T043311Z.log.html.gz
+
+Tue Oct 27 12:00:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/logger.rb (Logger::Period#next_rotate_time): get rid of
+	  adding to mday not to exceed the days of the month.
+	  [ruby-core:71185] [Bug #11620]
+
+Mon Oct 26 22:43:03 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test/ruby/test_module.rb (test_method_defined): Add test cases
+	  for `public/protected/private _method_defined?`
+	  These methods accept string as argument, so add string argument
+	  cases.  [Fix GH-1067]
+
+Mon Oct 26 22:23:30 2015  SimonDKnight  <simondknight@hotmail.com>
+
+	* lib/racc/rdoc/grammar.en.rdoc: Grammatical errors fixed.
+	  [Fix GH-1070]
+
+Mon Oct 26 18:36:43 2015  Shota Fukumori (sora_h) <her@sorah.jp>
+
+	* vm_method.c(rb_method_entry_make):
+	  [DOC] [ci skip] Remove a needless space from comment
+	  [Fixes GH-1069] Patch by @yui-knk
+
+Mon Oct 26 17:30:13 2015  Ryan Hosford  <tad.hosford@gmail.com>
+
+	* lib/fileutils.rb: rename tailing to trailing.
+	  [Misc #11548]
+
+Mon Oct 26 17:11:53 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (call_op, call_op2): fix values on ripper.  [Feature #11537]
+
+Mon Oct 26 12:55:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (call_op2): separate from call_op and also allow "::",
+	  while dot_or_colon should not allow ".?".  [Feature #11537]
+
+Mon Oct 26 01:03:23 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* thread_pthread.c: fix compile errors when
+	  USE_SLEEPY_TIMER_THREAD is disabled.
+
+Sun Oct 25 10:12:05 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.c (op_tbl): add DOTQ for ripper.  [Feature #11537]
+
+Sat Oct 24 22:51:18 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* configure.in: fixed build failure of Haiku.
+	  [fix GH-984] Patch by @kallisti5
+	* ext/socket/getaddrinfo.c: ditto.
+	* ext/socket/getnameinfo.c: ditto.
+	* ext/socket/rubysocket.h: ditto.
+
+Sat Oct 24 21:16:53 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fileutils/test_fileutils.rb (test_uptodate): relax error
+	  message format.  [Feature #9025], [ruby-core:71178] [Bug #11617]
+
+Sat Oct 24 21:06:43 2015  Shota Fukumori (sora_h)  <her@sorah.jp>
+
+	* lib/mkmf.rb: Revert r45640 because it may lead to link
+	  with different libruby. [Bug #9760]
+
+Sat Oct 24 15:42:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/test_method.rb: relax error message format.
+
+	* test/ruby/test_arity.rb (err_mess): ditto.
+	  [Feature #9025], [ruby-core:71178] [Bug #11617]
+
+Sat Oct 24 12:47:47 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* vm_insnhelper.c: improved error message for "wrong number
+	  of arguments", distinguishing given and expected argument
+	  numbers clearly. [Feature #9025]
+
+Sat Oct 24 11:57:59 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c: remove the typedef redeclaration of
+	  vm_call_handler.
+
+Sat Oct 24 07:29:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/forwardable.rb (def_instance_delegator, def_single_delegator):
+	  match backtraces against ::Forwardable in case the target class
+	  is a subclass of BasicObject and does not include Kernel.
+	  [ruby-core:71176] [Bug #11616]
+
+Sat Oct 24 04:10:13 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (make_compile_option_value): include frozen_string_literal*
+	  in a made option value.
+
+	* vm_opts.h: forgot to add OPT_FROZEN_STRING_LITERAL_DEBUG
+	  at last commit.
+
+Sat Oct 24 03:58:02 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ruby.c: introduce --enable-frozen-string-literal-debug option.
+	  If this option is enabled, the modify error will be:
+	    can't modify frozen String (RuntimeError) =>
+	    can't modify frozen String, created at test.rb:3 (RuntimeError)
+
+	* iseq.h: add compile option frozen_string_literal_debug.
+
+	* compile.c: catch up this fix.
+
+	* error.c (rb_error_frozen): ditto.
+
+	* iseq.c (set_compile_option_from_hash): ditto.
+
+	* test/ruby/test_rubyoptions.rb: add a test for this fix.
+
+Sat Oct 24 02:02:24 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: introduce new call handler for simple ISeqs.
+
+	  vm_call_iseq_setup_normal_0start() is simple, however it has
+	  some loops/conditions depends on ISeq::param.size and
+	  ISeq::local_size (in vm_push_frame(), inlined into this function).
+
+	  There are many simple methods which has a few parameters and local
+	  variables. So that this patch introduces several special functions
+	  generated in vm_call_iseq_optimized.inc by
+	  tool/mk_call_iseq_optimized.rb.
+
+	  This script makes
+	    vm_call_iseq_setup_normal_0start_Xparams_Ylocals()
+	  where X is 0 to 3 and Y is 1 to 6 (as current setting).
+	  In this case, X * Y = 24 functions are created.
+
+	  These functions creates fast method dispatch by inlining
+	  vm_push_frame() with immediate params/locals sizes.
+
+	  On my laptop, we can have the following results.
+
+	  vm2_method*       1.083 (8.3% faster)
+	  vm2_poly_method*  0.961 (3.4% slower)
+
+	  It shows 8.3% faster for inner loop method dispatch (hit inline
+	  cache), but 3.4% slower when inline cache miss because we need
+	  to find a suitable call handler.
+
+	* common.mk: add a rule for vm_call_iseq_optimized.inc.
+
+	* tool/mk_call_iseq_optimized.rb: added.
+
+	* vm.c: include vm_call_iseq_optimized.inc.
+
+Sat Oct 24 01:58:50 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: define vm_call_handler.
+
+Sat Oct 24 01:56:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, vm_insnhelper.h: move definition of VMDEBUG
+	  from vm_insnhelper.h to vm_core.h.
+
+Sat Oct 24 01:51:01 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* NEWS: [DOC] In the new safe call syntax, arguments are evaluated
+	  only if a call is made.
+
+	* doc/syntax/calling_methods.rdoc: Fix a typo.
+
+Sat Oct 24 00:38:34 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/forwardable (def_instance_delegator, def_single_delegator):
+	  rescue ::Exception instead of Exception in case Exception is
+	  defined under the target class.
+	  [ruby-core:71175] [Ruby trunk - Bug #11615]
+
+Fri Oct 23 21:10:37 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (name_err_mesg_to_str): separate class names from the
+	  receiver description.
+
+	* vm_eval.c (make_no_method_exception, raise_method_missing): add
+	  format specifiers for class names.
+
+Fri Oct 23 18:10:32 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* .gitignore: ignored environmental wrapper files.
+
+Fri Oct 23 17:55:29 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/irb.rb: Ignored assignment of STDOUT.sync = true
+	  when irb.rb loaded. It's affected to IDE such as Jetbrain.
+	  [fix GH-864] Patch by @os97673
+
+Fri Oct 23 16:35:08 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/ipaddr.rb, test/test_ipaddr.rb: Reject invalid address contained
+	  EOL string. Patch by @kachick [fix GH-942][Bug #11513]
+
+Fri Oct 23 16:03:26 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* file.c: fix indent style. [fix GH-977]
+	* test/ruby/test_string.rb: indent. [fix GH-975]
+	  [ci skip] These patches are contributed from @yui-knk
+
+Fri Oct 23 15:46:09 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* string.c: Added method signature to include hash. It's inconsistency
+	  with `gsub` method signature.
+	  [ci skip][fix GH-1023] Patch by @danielevans
+
+Fri Oct 23 15:25:51 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: remove an empty comment line and -*-.
+
+Fri Oct 23 15:20:02 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (gettextfile, getbinaryfile): use the safe
+	  navigation operator.
+
+Fri Oct 23 13:51:33 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test_call.rb (test_safe_call): Add test cases for safe
+	  navigation operator assignment.  [Fix GH-1064]
+	  Validate:
+	  * can assign an attribute which is `nil`
+	  * can "or assign" an attribute which is `nil`
+
+Fri Oct 23 11:58:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): optimize lengthy safe
+	  navigation method chain.  [Feature #11537]
+
+Fri Oct 23 10:58:41 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/matrix/eigenvalue_decomposition.rb (tridiagonalize): fix
+	  indentation to avoid a warning when the command line option -w of
+	  ruby is specified.
+
+	* lib/matrix/eigenvalue_decomposition.rb (hessenberg_to_real_schur):
+	  change the name of a block parameter to avoid a warning when the
+	  command line option -w of ruby is specified.
+
+Fri Oct 23 10:49:36 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): support safe navigation of simple
+	  attribute assignment.  [Feature #11537]
+
+	* parse.y (mlhs_node, lhs, attrset_gen): ditto.  keep mid
+	  non-attrset as the sign of safe navigation.
+
+Fri Oct 23 07:17:11 2015  Eric Wong  <e@80x24.org>
+
+	* test/io/wait/test_io_wait.rb (test_wait_eof): test return value
+
+Fri Oct 23 00:32:02 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ssl_npn_select_cb): explicitly raise error
+	  in ext/openssl instead of OpenSSL itself because LibreSSL
+	  silently truncate the selected protocol name by casting the length
+	  from int to unsigned char. [Bug #11369]
+	  Patch by Jeremy Evans <merch-redmine@jeremyevans.net>
+
+Fri Oct 23 00:49:45 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/un.rb (help): change the name of a block parameter to avoid
+	  a warning when the command line option -w of ruby is specified.
+
+Fri Oct 23 00:22:20 2015  Josef Simanek  <josef.simanek@gmail.com>
+
+	* string.c (rb_str_tr): [DOC] Escape backslash in String#tr
+	  documentation.  [Fix GH-1063]
+
+Fri Oct 23 00:19:04 2015  yui-knk  <spiketeika@gmail.com>
+
+	* array.c (rb_ary_collect): [DOC] Fix space of code example of
+	  Array#map.  [Fix GH-1062]
+
+Thu Oct 22 18:52:53 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* vm_eval.c (rb_f_loop): When a loop is stopped by a StopIteration
+	  exception, return what the enumerator has returned instead of
+	  nil. [ruby-core:71133] [Feature #11498]
+
+Thu Oct 22 18:25:10 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (idle): add a new argument timeout for keep-alive.
+	  [ruby-core:63693] [Bug #10031]
+
+Thu Oct 22 15:30:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): peephole optimization for
+	  branchnil jumps.
+
+	* compile.c (iseq_compile_each): generate save navigation operator
+	  code.
+
+	* insns.def (branchnil): new opcode to pop the tos and branch if
+	  it is nil.
+
+	* parse.y (NEW_QCALL, call_op, parser_yylex): parse token '.?'.
+	  [Feature #11537]
+
+Thu Oct 22 13:16:19 2015  Guilherme Reis Campos  <guilhermekbsa@gmail.com>
+
+	* dir.c (ruby_brace_expand): glob brace expansion edge case fix.
+	  When there are closing braces '}' before a open brace '{' it
+	  must be ignored and considered as literal.
+	  [ruby-core:71138] [Bug #11609]
+
+Thu Oct 22 13:13:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (argf_next_argv): check ARGV element type, and try
+	  conversion if necessary.  [ruby-core:71140] [Bug #11610]
+
+Thu Oct 22 11:11:16 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/net/ftp/test_ftp.rb: add tests for getbinaryfile and
+	  gettextfile.
+
+Wed Oct 21 18:34:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_magic_comment): allow a sole magic comment without
+	  indicators, neither other non-space comments.  [Feature #8976]
+
+Tue Oct 20 12:17:56 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Add basic argument checking to Prime.prime?
+	  [Bug #11606]
+
+Tue Oct 20 12:17:50 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Optimize Integer#prime?
+	  Patch by Nick Slocum [Bug #10354]
+
+Tue Oct 20 08:12:47 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* configure.in: pthread_getattr_np is broken on AIX.
+	  More specifically, the stack address and size returned are
+	  not correct.
+
+Tue Oct 20 05:54:46 2015  Eric Wong  <e@80x24.org>
+
+	* ext/fiddle/closure.c (callback): static function
+
+Mon Oct 19 10:33:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/init.c (rsock_raise_socket_error): get rid of a glibc
+	  bug.  [ruby-core:71100] [Bug #11600]
+
+Mon Oct 19 01:26:26 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (rb_file_identical_p): not necessary to compare the paths after
+	  comparing the file indexes on Windows.  designate by kosaki.
+
+Sun Oct 18 21:17:27 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: Specify frozen_string_literal: true.
+
+Sun Oct 18 14:37:56 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c (fill_random_bytes_urandom): add a comment why using
+	  O_NONBLOCK and O_NOCTTY.
+
+Sun Oct 18 13:24:17 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* random.c (fill_random_bytes_syscall): use ATOMIC_SET() for
+	  updating try_syscall.
+
+Sun Oct 18 13:03:52 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/backward/util.h: Good-by Borland-C.
+
+Sun Oct 18 13:03:09 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* common.mk: add a comment how to use "make test-all"
+
+Sun Oct 18 12:59:22 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* common.mk: add comments how to use "make benchmark"
+
+Sun Oct 18 12:58:15 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/securerandom.rb: Specify frozen_string_literal: true.
+
+Sun Oct 18 11:22:52 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* dln.c: remove defined(__WATCOMC__).
+
+Sun Oct 18 11:16:33 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* lib/mkmf.rb: Good-by Borland-C.
+
+Sun Oct 18 11:04:36 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* numeric.c: Good-by Borland-C.
+	* include/ruby/backward/rubyio.h: ditto.
+	* include/ruby/backward/st.h: ditto.
+	* include/ruby/backward/util.h: ditto.
+	* include/ruby/backward/rubysig.h: ditto.
+	* include/ruby/backward/classext.h: ditto.
+	* dln.c: ditto.
+	* gc.c: ditto.
+	* win32/resource.rb: ditto.
+	* win32/dir.h: ditto.
+	* ext/tk/tcltklib.c: ditto.
+	* NEWS: announce that Borland-C is no longer supported.
+
+Sun Oct 18 10:54:52 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* dln.c: simplify #ifdef. _WIN32 and __CYGWIN__ are exclusive.
+	  see include/ruby/defines.h
+	* gc.c: ditto.
+	* ext/sdbm/_sdbm.c: ditto.
+
+Sun Oct 18 10:42:19 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ruby.c (open_load_file): add a comment.
+
+Sun Oct 18 10:12:46 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (rb_file_identical_p): simplify ifdefs
+
+Sun Oct 18 10:01:40 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ChangeLog: Good-bye OS/2.
+	* common.mk: ditto.
+	* configure.in: ditto.
+	* dln_find.c: ditto.
+	* ext/Setup.emx: ditto.
+	* ext/extmk.rb: ditto.
+	* ext/socket/extconf.rb: ditto.
+	* ext/zlib/extconf.rb: ditto.
+	* file.c: ditto.
+	* include/ruby/defines.h: ditto.
+	* io.c: ditto.
+	* lib/mkmf.rb: ditto.
+	* missing/os2.c: ditto.
+	* process.c: ditto.
+	* ruby.c: ditto.
+	* NEWS: announce OS/2 is no longer supported.
+
+Sun Oct 18 08:50:15 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* include/ruby/defines.h (DOSISH): add comments.
+
+Sun Oct 18 08:26:51 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* io.c (fptr_finalize): don't release gvl if fptr is not writable.
+	  writable fd may block on close(2) when it's on NFS. But readonly
+	  fd doesn't. [Bug #11559]
+	  result: make benchmark OPTS="-p bm_require_t -e ruby-trunk -e ruby-2.2.2"
+	    build-ruby:             0.171
+	    ruby 2.3.0dev(r52151):  0.659
+	    ruby 2.2.0p95 (r50295): 0.834
+
+Sun Oct 18 09:32:58 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* file.c (ruby_is_fd_loadable): this should be fail if st_mode is
+	  not regular file nor FIFO.
+
+Sun Oct 18 09:20:17 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ruby.c (open_load_file): use rb_thread_wait_fd() instead of reopen.
+
+Sun Oct 18 05:11:22 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* ruby.c (open_load_file): reset O_NONBLOCK after open.
+	  Even if S_ISREG() is true, the file may be file on FUSE filesystem
+	  or something. We can't assume O_NONBLOCK is safe.
+	  Moreover, we should wait if the path is point to FIFO. That's
+	  FIFO semantics. GVL should be transparent from ruby script.
+	  Thus, just reopen without O_NONBLOCK for filling the requirements.
+	  [Bug #11060][Bug #11559]
+
+	* ruby.c (loadopen_func): new for the above.
+
+	* file.c (ruby_is_fd_loadable): new. for checks loadable file type
+	  of not.
+	* file.c (rb_file_load_ok): use ruby_is_fd_loadble()
+	* internal.h: add ruby_is_fd_loadble()
+
+	* common.mk: now, ruby.o depend on thread.h.
+
+	* test/ruby/test_require.rb
+	(TestRequire#test_loading_fifo_threading_success): new test.
+	  This test successful case that loading from FIFO.
+
+	* test/ruby/test_require.rb
+	(TestRequire#test_loading_fifo_threading_raise): rename from
+	  test_loading_fifo_threading. You shouldn't rescue an exception
+	  if you test raise or not.
+	  Moreover, this case should be caught IOError because load(FIFO)
+	  should be blocked until given any input.
+
+Sat Oct 17 13:55:32 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_expand_path_internal): concatenate converted
+	  string to the result instead of making converted string and
+	  append it.
+
+	* string.c (rb_str_cat_conv_enc_opts): from rb_str_conv_enc_opts,
+	  separate function to concatenate with transcoding.
+
+Sat Oct 17 13:19:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (load_file): unify each preparations and clean-ups by
+	  merging load_file_internal and load_file_internal2, and remove
+	  nested rb_protect and rb_ensure.
+
+Sat Oct 17 05:28:32 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/ruby/test_symbol.rb (test_symbol_fstr_leak): add a warm-up
+	  code and check RSS to avoid false positive on AIX and false
+	  negative on Mac OS X. [Bug #10686]
+
+Fri Oct 16 15:54:37 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_load_ok): open in non-blocking mode withoout
+	  releasing GVL.  don't care about others than regular files and
+	  directories.  [ruby-dev:49272] [Bug #11559]
+
+	* ruby.c (load_file_internal): ditto.
+
+Thu Oct 15 23:56:03 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_sym_to_proc): make void env.
+
+Thu Oct 15 13:37:23 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_sym_to_proc): move from string.c and create a Proc
+	  with no environments.  [ruby-core:71088] [Bug #11594]
+
+Thu Oct 15 01:57:03 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* test/objspace/test_objspace.rb
+	  (test_trace_object_allocations_start_stop_clear): clear object
+	  allocation table first to get rid of erroneous detection for obj3.
+	  [ruby-dev:49095] [Bug #11271]
+
+Thu Oct 15 01:53:38 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/ostruct/test_ostruct.rb: Add tests for OpenStruct#respond_to.
+	  Patch by @jeremy in [GH-1041]: https://github.com/ruby/ruby/pull/1041
+
+Thu Oct 15 01:49:25 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/ostruct.rb: Finish defining OpenStruct attributes lazily.
+	  Patch by @sferik in [GH-1037]: https://github.com/ruby/ruby/pull/1037
+	  This commit is an addendum to https://github.com/ruby/ruby/pull/1033.
+	  It:
+	  1. lazily defines attribute accessors for copied and marshaled objects,
+	  2. returns nil when an attribute reader is not defined, and
+	  3. defines respond_to_missing? to maintain the same respond_to? behavior
+
+Wed Oct 14 16:56:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check for libunwind.h, which is not available in
+	  very old OS X SDK.  [ruby-core:71080] [Bug #11591]
+
+Wed Oct 14 14:11:42 2015  Brian Black  <bblack@veracode.com>
+
+	* iseq.c (rb_insn_operand_intern): change kw in callinfo disasm from the
+	  number of keyword arguments to an ordered list of the keywords used.
+	  [Feature #11589]
+
+Wed Oct 14 13:58:44 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_nextc): send a warning to ripper, not to STDERR
+	  always.
+
+	* parse.y (rb_warn1, rb_warning1): move argument conversions to
+	  callers.  PRIsVALUE is not valid in String#%.
+
+Wed Oct 14 13:37:23 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/racc/rdoc/grammar.en.rdoc: fix spell error.
+	  [fix GH-1053][ci skip] Patch by @Matrixbirds
+
+Tue Oct 13 22:06:50 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (rsock_fd_family): Check sa_len.
+
+Tue Oct 13 12:14:10 2015  Craig Davison  <craig65535@gmail.com>
+
+	* ext/socket/rsock_addrinfo (rsock_addrinfo): specify address
+	  family.  [Fix GH-1052]
+
+	* ext/socket/udpsocket.c (udp_connect, udp_bind, udp_send):
+	  address family by the receiver.
+
+Sun Oct 11 07:09:19 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_push_frame): initialize other than sp (and ep)
+	  first for performance.
+
+Sun Oct 11 06:21:50 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c, internal.h (rb_yield_1): added for performance which
+	  doesn't check Qundef.
+
+	* numeric.c (int_dotimes): use rb_yield_1.
+
+Sun Oct 11 06:19:49 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_iseq_setup_normal): setup sp first
+	  for performance.
+
+Sun Oct 11 05:29:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (invoke_block_from_c): split this function into several
+	  functions.
+
+	* vm_insnhelper.c (vm_yield_callee_setup_arg): remove this function
+	  because it is only delegation function.
+
+Sun Oct 11 03:48:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (newobj_of_slowpass): fix typo (pass -> path).
+	  Pointed out by Yukihiro Matsumoto  <matz@ruby-lang.org>.
+
+	* gc.c (newobj_of_...): `of' is unnecessary.
+
+Sat Oct 10 19:04:42 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/udpsocket.c (udp_connect, udp_bind): get open files
+	  inside ensure functions.
+
+Sat Oct 10 18:35:12 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method0): do not propagate enable_fastpath,
+	  but pass dummy CC to prevent wrong caching.
+
+Sat Oct 10 15:28:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* import a github pull request
+	  https://github.com/ruby/ruby/pull/1050
+	  by Kazuho Oku <kazuho@natadeco.co>.
+
+	  This pull request has the following commits.
+
+	* gc.c: reduce # of args to 6 (max. of register args on x86-64) so
+	  that the `newobj_of_slowpass` can be called via TCO.
+
+	* gc.c (newobj_of), string.c (str_duplicate): for performance,
+	  the hot functions must be inlined.
+
+	* gc.c: for performance, preceding arguments of `.*newobj_of.*`
+	  must be same, so that the arg registers can be reused in case of
+	  TCO.
+
+Sat Oct 10 08:52:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/udpsocket.c (udp_connect, udp_bind, udp_send): fix
+	  memory leaks at closed socket.
+
+Fri Oct  9 17:29:07 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (parse257): refactor.
+
+Fri Oct  9 16:42:26 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb: use frozen_string_literal: true.
+
+	* test/net/imap/test_imap.rb: ditto.
+
+	* test/net/imap/test_imap_response_parser.rb: ditto.
+
+Fri Oct  9 15:52:28 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb: use frozen_string_literal: true.
+
+	* test/net/ftp/test_buffered_socket.rb: ditto.
+
+	* test/net/ftp/test_ftp.rb: ditto.
+
+	* test/net/ftp/test_mlsx_entry.rb: ditto.
+
+Fri Oct  9 14:12:35 2015  Shota Fukumori (sora_h)  <her@sorah.jp>
+
+	* ext/openssl/lib/openssl/ssl.rb: Revert r52082 because it was
+	  dropping TLS v1.1 support too. Supporting only TLS v1.2 is too
+	  early, because many popular websites still don't support it.
+
+	  For instance, Servers where aws-sdk connects to still don't support
+	  TLS v1.2 and it became broken.
+
+	  We should consider more carefully about this.
+
+	  [Fix GH-873] [Feature #11524]
+
+Fri Oct  9 12:52:08 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): Dynamic string literals (e.g.,
+	  "#{x}") should not be frozen because they don't literally
+	  represent strings.
+	  https://twitter.com/shugomaeda/status/651937650027401216
+	  https://twitter.com/yukihiro_matz/status/651942882312482817
+	  https://twitter.com/yukihiro_matz/status/651980835181096960
+
+Fri Oct  9 06:52:49 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* benchmark/prepare_require.rb: skip file creation if it already
+	  exist. Suggested by ko1.
+
+Fri Oct  9 06:18:04 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+	* benchmark/bm_require.rb: new benchmark for require.
+	* benchmark/bm_require_thread.rb: new benchmark for conflicting
+	  require vs thread. like [Bug #11559]
+	* prepare_require.rb: new file for preparing above tests.
+	* prepare_require.rb: ditto.
+
+Thu Oct  8 14:10:45 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb: Default to TLSv1.2 and drop TLS v1
+	  Patch provided by @claudijd [Fixes GH-873] [Feature #11524]:
+	  https://github.com/ruby/ruby/pull/873
+
+Wed Oct  7 22:55:02 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/minitest/metametameta.rb (with_output): restore output to
+	  fix mixing test result output in worker responses.
+
+Wed Oct  7 21:32:51 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_duplicate): move from rb_str_resurrect to short
+	  circuit initialization.
+
+Wed Oct  7 20:43:14 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_resurrect): fix resurrection of short enough to
+	  be embedded but not embedded string.
+
+Wed Oct  7 20:17:29 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (newobj_of): divide fast path and slow path
+	  to avoid register savings for fast path.
+
+	  This idea is given by Kazuho Oku <kazuho@natadeco.co>.
+
+Wed Oct  7 17:30:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_times): optimize for the argument 0 and 1.
+
+Wed Oct  7 01:20:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.h, gc.c: introduce new debug function rb_obj_info_dump(VALUE obj)
+	  which prints the result of rb_raw_obj_info(..., obj).
+
+Wed Oct  7 01:16:11 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_args.c: remove an unused field args_info::calling.
+
+Tue Oct  6 23:43:10 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_method_entry_min_max_arity): should support
+	  OPTIMIZED_METHOD_TYPE_CALL.
+
+Tue Oct  6 21:29:08 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/tmpdir.rb (Dir.tmpdir): return duplicated string to be
+	  modify safely even when $SAFE > 0.
+
+Tue Oct  6 19:24:38 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_method0): use switch() for visibilities
+	  (for readability).
+
+Tue Oct  6 19:23:58 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (Init_Proc): Proc#call and others should be public.
+
+Tue Oct  6 18:51:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: IMEMO_FL_USER3 and IMEMO_FL_USER4 is not needed any more.
+
+Tue Oct  6 18:47:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: remove METHOD_ENTRY_SAFE(me) and related code
+	  because $SAFE = 3 and 4 is not available.
+	  Now, $SAFE is not checked on method dispatch at all.
+
+	* vm_eval.c, vm_insnhelper.c, vm_method.c: ditto.
+
+Tue Oct  6 13:56:14 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h: turn function macros into inline functions,
+	  for debuggers.
+
+	* include/ruby/ruby.h: turn constant macros into enums, for
+	  debuggers.
+
+Tue Oct  6 13:48:05 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* method.h: typo fix. Patch by @davydovanton [fix GH-1032][ci skip]
+
+Tue Oct  6 06:54:34 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (rb_iseq_free): free iseq::variable_body to avoid memory
+	  leak.
+
+Tue Oct  6 06:32:52 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: enable optimization of Proc#call.
+	  [Feature #11569]
+
+	* NEWS: write about this optimization and incompatibilities.
+
+	* test/ruby/test_backtrace.rb: catch up this fix.
+
+Tue Oct  6 04:41:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: solve goto spaghetti.
+
+	  Change all goto statement across blocks to tail call functions.
+
+Tue Oct  6 02:29:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_resurrect): optimize by short circuit to copy
+	  hidden string without checking length, encoding and so on.
+
+Mon Oct  5 23:08:17 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* test/ruby/test_thread.rb (test_handle_interrupt_blocking): check if
+	  exception handling was postponed until sleep.
+
+Mon Oct  5 22:25:49 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb: Use frozen_string_literal: true.
+
+	* lib/prettyprint.rb: Ditto.
+
+	* lib/resolv.rb: Ditto.
+
+	* lib/tmpdir.rb: Ditto.
+
+	* test/test_pp.rb: Ditto.
+
+	* test/test_prettyprint.rb: Ditto.
+
+	* tool/transcode-tblgen.rb: Ditto.
+
+Mon Oct  5 20:39:32 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/ruby/test_thread.rb: fix potential race condition.
+	  The thread could have a "sleep" status because it tries
+	  to acquire the mutex, but does not have it yet.
+
+Mon Oct  5 15:39:30 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* numeric.c: [DOC] Overview for Numeric class by Joe Corcoran
+	  This patch was created at ROSSConf Berlin 2015 [Bug #11555]
+
+Mon Oct  5 15:34:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_new): link ep to calling block.
+	  [ruby-core:70980] [Bug #11566]
+
+Mon Oct  5 00:53:51 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (rb_dir_getwd): make ASCII-8BIT if filesystem encoding is
+	  US-ASCII, like as Dir.glob.
+
+Sun Oct  4 23:39:09 2015  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (nmin_filter): Fix limit value.
+	  patch by Helder Pereira.
+	  [Bug #11471] [ruby-core:70477]
+
+Sun Oct  4 15:11:48 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/euc_jp.c (mbc_case_fold): check given string is valid or not,
+	  and if invalid, return 1. [Bug #11486]
+
+Sun Oct  4 10:09:57 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (rb_dir_getwd): normalize OS path to UTF-8 on OS X.
+
+Sun Oct  4 00:09:45 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/ruby-runner.c.in: wrapper to set dynamic loading path
+	  environment variable.  /bin/sh on Mac OS X 10.11 (El Capitan)
+	  clears DYLD_LIBRARY_PATH.
+
+	  it must:
+	  - do nothing even if current directory is not present
+	  - do not set other environment variables, e.g. PWD, SHLVL, etc
+	  - do not open other FDs, e.g. pipes for timer thread
+
+Sun Oct 2 10:59:00 2015  schneems  <richard.schneeman@gmail.com>
+
+	* ext/pathname/lib/pathname.rb: freeze string literals for
+	  reduced object allocation.
+	  [Feature #11375] [ruby-core:70043]
+
+Fri Oct  2 09:20:20 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk, lib/unicode_normalize/tables.rb: Change Unicode
+	  Version for character normalization data from 7.0.0 to
+	  8.0.0.
+
+Fri Oct  2 00:18:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_mark): block.ep of Proc from Symbol is now NULL.
+	  [ruby-core:70961] [Bug #11560]
+
+Wed Sep 30 15:47:13 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_args.c (vm_caller_setup_arg_block): bypass Symbol#to_proc
+	  call to optimize symbol block passing.
+
+Wed Sep 30 01:34:34 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_free): fix memory leak at syntax error when
+	  warn-indent is enabled.
+
+Tue Sep 29 22:27:50 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* parse.y: fix minor typo. [ci skip][fix GH-1038].
+	  Patch by @ltratt.
+
+Tue Sep 29 16:53:53 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* tool/unicode_norm_gen.tmpl, lib/unicode_normalize/tables.rb:
+	  get rid of many .freeze commands by using frozen_string_literal
+	  pragma.
+
+Tue Sep 29 16:37:29 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_dstr_fragments): fix performance by omitting
+	  the first empty string only for keeping literal encoding if
+	  other literals are too.  [ruby-core:70930] [Bug #11556]
+
+	* string.c (rb_str_append_literal): append but keep encoding non
+	  US-ASCII.
+
+Mon Sep 28 17:40:17 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (mtime): use usec instead of fractions to parse
+	  decimal fractions of a second correctly when the number of digits
+	  is not 6.
+
+Mon Sep 28 16:07:08 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (mtime): parse decimal fractions of a second as
+	  specified in RFC 3659.
+
+Mon Sep 28 10:31:12 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/test_forwardable.rb: Write basic tests for lib/forwardable.
+	  [fix GH-1035] Patch by @kachick
+
+Sun Sep 27 23:32:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_define_class, rb_define_class_id_under): refine
+	  error messages.
+
+	* class.c (rb_define_module, rb_define_module_id_under): ditto,
+	  and make consistent with class.
+
+Sun Sep 27 18:44:43 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: removed duplicated message.
+
+Sun Sep 27 15:46:58 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): add an option to enable/disable
+	  frozen-string-literal.  [Feature #8976]
+
+	* compile.c (iseq_compile_each): override compile option by option
+	  given by pragma.
+
+	* iseq.c (rb_iseq_make_compile_option): extract a function to
+	  overwrite rb_compile_option_t.
+
+	* parse.y (parser_set_compile_option_flag): introduce pragma to
+	  override compile options.
+
+	* parse.y (magic_comments): new pragma "frozen-string-literal".
+	  [Feature #8976]
+
+Sun Sep 27 08:16:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (delete_field): do not raise NameError for
+	  existing keys.  [Fix GH-1033]
+
+Sun Sep 27 00:34:31 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* lib/ostruct.rb: Move method definitions for getter/setter to be lazy
+	  Patch by @sferik in [GH-1033]: https://github.com/ruby/ruby/pull/1033
+
+Fri Sep 25 10:07:25 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/http.rb: removed unused variable. It's removed at r13648.
+	  [fix GH-1022] Patch by @nkondratyev
+
+Fri Sep 25 09:48:27 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: upgrade to minitest-5.8.1
+
+Fri Sep 25 09:47:12 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* id_table.c: fix typo. [ci skip][fix GH-1031] Patch @davydovanton
+
+Fri Sep 25 07:54:05 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/gdbm/test_gdbm.rb (TestGDBM#test_s_open_lock): skip
+	  this test on AIX. The issue is the same as on Solaris.
+	  [ruby-dev:47631]
+
+Thu Sep 24 17:25:09 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (paren_args): fix separator token at `foo::bar()` in
+	  ripper.
+
+Thu Sep 24 00:00:17 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* complex.c: ruby/config.h must be included before math.h
+	  because it defines _LARGE_FILES on AIX and _LARGE_FILES
+	  must be defined before sys/types.h is included from math.h.
+	  [Bug #11483]
+
+Wed Sep 23 22:22:38 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/openssl/ossl_pkcs12*: Remove svn commit id macro
+
+Wed Sep 23 01:11:28 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/openssl/*: Remove svn commit id macros to make sync easier
+
+Tue Sep 22 04:20:01 2015  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* test/drb/test_drb.rb: Run Rinda/DRb tests on localhost. [Fix GH-1027]
+	  patch by voxik.
+
+	* test/rinda/test_rinda.rb: ditto
+
+Mon Sep 21 20:53:39 2015  tbpgr  <tbpgr@tbpgr.jp>
+
+	* test/win32ole/test_win32ole_event.rb: fix typo.
+	  swbemsink_avairable? => swbemsink_available?  [Fix GH-1025]
+
+Sun Sep 20 10:07:35 2015  Anton Davydov  <antondavydov.o@gmail.com>
+
+	* cont.c (rb_callcc): [DOC] append continuations example across
+	  methods.  [Fix GH-1026]
+
+Sun Sep 20 03:20:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (rb_iseq_free): free rb_iseq_t::body::cc_entries.
+
+Sun Sep 20 02:46:34 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: split rb_call_info_t into several structs.
+	  * rb_call_info (ci) has compiled fixed information.
+	  * if ci->flag & VM_CALL_KWARG, then rb_call_info is
+	    also rb_call_info_with_kwarg. This technique reduce one word
+	    for major rb_call_info data.
+	  * rb_calling_info has temporary data (argc, blockptr, recv).
+	    for each method dispatch. This data is allocated only on
+	    machine stack.
+	  * rb_call_cache is for inline method cache.
+
+	  Before this patch, only rb_call_info_t data is passed.
+	  After this patch, above three structs are passed.
+
+	  This patch improves:
+	  * data locality (rb_call_info is now read-only data).
+	  * reduce memory consumption (rb_call_info_with_kwarg,
+	    rb_calling_info).
+
+	* compile.c: use above data.
+
+	* insns.def: ditto.
+
+	* iseq.c: ditto.
+
+	* vm_args.c: ditto.
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_insnhelper.h: ditto.
+
+	* iseq.h: add iseq_compile_data::ci_index and
+	  iseq_compile_data::ci_kw_index.
+
+	* tool/instruction.rb: introduce TS_CALLCACHE operand type.
+
+Sun Sep 20 02:18:10 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/envutil.rb: mkfifo command based File.mkfifo method
+	  definition removed.
+
+Fri Sep 18 20:11:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_s_mkfifo): implement File.mkfifo.
+	  [Feature #11536]
+
+Fri Sep 18 16:56:19 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* NEWS: add Net::FTP#mlst and Net::FTP#mlsd.
+
+Fri Sep 18 07:39:22 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/objspace/objspace_dump.c (obj_type): add IMEMO types to the heap
+	  dump information.
+
+Thu Sep 17 22:33:07 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* common.mk: fix command error with outside builddir.
+
+Thu Sep 17 17:42:09 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* common.mk: separated test for test-framework from test-all task.
+	  They should be invoke at first before tests of test-all.
+
+Thu Sep 17 12:05:54 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_dir.rb (TestDir#test_fileno): s/?x/"x"/. Don't
+	  use tricky code, please.
+
+Wed Sep 16 20:49:56 2015  Masaki Suketa  <masaki.suketa@nifty.ne.jp>
+
+	* encindex.h: fix typo of last #endif comment. [ci skip]
+
+Wed Sep 16 20:39:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (set_const_visibility): fail if the class/module is
+	  frozen.  [ruby-core:70828] [Bug #11532]
+
+Wed Sep 16 17:16:43 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (ENABLE_VM_OBJSPACE): enable per-VM object space on
+	  Windows by default, as rb_w32_sysinit() no longer depends on
+	  ruby_xmalloc.
+
+Wed Sep 16 15:08:17 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* doc/syntax/literals.rdoc (Strings): [DOC] Revise the character
+	  literal part.
+
+Wed Sep 16 14:55:33 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* doc/syntax/literals.rdoc (Strings): [DOC] Document the full list
+	  of supported escape sequences in string literals.
+
+Wed Sep 16 14:49:58 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_setbyte): keep the code range as possible.
+
+Wed Sep 16 13:23:48 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* doc/syntax/literals.rdoc (Strings): mention about ?a literal.
+
+Wed Sep 16 12:06:53 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* dir.c (glob_helper): check pathtype once again by lstat(2) if
+	  dp->d_type is DT_UNKNOWN. XFS may return DT_UNKNOWN.
+
+Wed Sep 16 03:49:19 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* test/ruby/test_thread.rb (TestThread#test_mutex_synchronize):
+	  insert waste loop for invoking preemptive thread context switch.
+	  [Bug #11496]
+
+Tue Sep 15 19:38:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_objspace_alloc, rb_objspace_free): define always
+	  regardless ENABLE_VM_OBJSPACE, and free heap pages.
+
+Tue Sep 15 15:15:41 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_sysinit, rb_w32_readdir): compare by
+	  encoding index to get rid of encoding initialization before VM
+	  object space allocation.
+
+	* dir.c (fundamental_encoding_p, push_glob): compare by encoding
+	  index immediately.
+
+	* enc/{ascii,us_ascii,utf_8}.c: set encoding indexes of
+	  fundamental built-in encodings so that usable as well as
+	  allocated rb_encoding before rb_enc_init().
+
+	* encindex.h: separate encoding index constants from internal.h.
+
+Tue Sep 15 13:13:13 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang, rb_ary_sort): [DOC] correct block
+	  return values, which may be a negative or positive integer, not
+	  only -1 or +1.
+
+Tue Sep 15 12:49:10 2015  Jason Barnabe  <jason.barnabe@gmail.com>
+
+	* array.c (rb_ary_sort_bang, rb_ary_sort): [DOC] Correct
+	  description of array sort block return values.  And also fix up
+	  the grammar a bit.  [Fix GH-1020]
+
+Tue Sep 15 12:44:32 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_qsort): use BSD-style qsort_r if available.
+
+Mon Sep 14 19:26:34 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (parse_mlsx_entry): parse pathnames including
+	  space correctly.
+
+Mon Sep 14 11:12:10 2015  Anton Davydov  <antondavydov.o@gmail.com>
+
+	* lib/racc/rdoc/grammar.en.rdoc: [DOC] fix typo, "convertion" to
+	  "conversion".  [Fix GH-1016]
+
+Sun Sep 13 11:03:13 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h: prefix RUBY or RB to global symbols to get
+	  rid of name conflicts with other headers.
+
+	* include/ruby/encoding.h, include/ruby/intern.h: ditto.
+
+Sun Sep 13 09:38:51 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (size, modify, create, type, unique, perm, lang,
+	  media_type, charset): new methods to return standard facts.
+
+Sat Sep 12 19:43:49 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_call_iseq_setup_normal): do not clear local
+	  variables here. vm_push_frame() clears.
+
+	* vm_insnhelper.c (vm_call_iseq_setup_tailcall): ditto.
+
+	* vm_insnhelper.c (vm_push_frame): move check code to
+	  vm_check_frame().
+
+	  Reorder initialization timing to reuse same values (sp).
+
+	* compile.c (rb_iseq_compile_node): use
+	  iseq_set_exception_local_table() for ISEQ_TYPE_DEFINED_GUARD.
+
+Sat Sep 12 23:06:51 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (file?, directory?, appendable?, creatable?,
+	  deletable?, enterable?, renamable?, listable?, directory_makable?,
+	  purgeable?, readable?, writable?): new methods.
+
+Sat Sep 12 21:27:22 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (FACT_PARSERS): support system dependent facts
+	  UNIX.mode, UNIX.owner, UNIX.group, UNIX.ctime, and UNIX.atime.
+
+Sat Sep 12 19:08:58 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_dup2): should return the new fd on
+	  success, while msvcrt returns 0 wrongly.
+
+Sat Sep 12 18:14:11 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (parse_mlsx_entry, mlst) raise an FTPProtoError
+	  when parsing failed.
+
+Sat Sep 12 18:00:35 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (TIME_PARSER): use "Z" instead of "+00:00" to
+	  get UTC time.  Thanks, Wilson Bilkovich.
+
+Sat Sep 12 17:55:24 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (mlst, mlsd): support new commands MLST and MLSD
+	  specified in RFC 3659.
+
+Sat Sep 12 16:14:31 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* file.c: access()/eaccess() wrapping methods check more than just uid.
+	  [fix GH-1007][ci skip] Patch by @eam
+
+Sat Sep 12 16:07:01 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* README.md: improve markdown rendering for readability.
+	  [fix GH-1015][ci skip] Patch by @Matrixbirds
+
+Sat Sep 12 14:30:03 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (save_redirect_fd): make saved FDs close-on-exec not
+	  to be inherited.
+
+	* process.c (run_exec_dup2): restore close-on-exec flags too.
+
+	* win32/win32.c (fcntl): implement F_GETFD, F_SETFD, and
+	  F_DUPFD_CLOEXEC.
+
+Sat Sep 12 05:35:24 2015  Eric Wong  <e@80x24.org>
+
+	* rational.c (string_to_r_strict): preserve encoding in exception
+
+Fri Sep 11 20:23:35 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove rb_call_info_t::aux.opt_pc.
+
+	* vm_insnhelper.c: introduce shortcut functions for opt_pc == 0
+	  because opt_pc is always 0 on shortcut function.
+
+Fri Sep 11 17:49:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c: disable ISeq.load. It enabled accidentally at r51794.
+
+Fri Sep 11 11:15:12 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (size, mdtm, system): parse responses according to
+	  RFC 959 and 3659, where reply codes must be followed by SP.
+
+	* lib/net/ftp.rb (system): remove LF from the return value.
+
+Thu Sep 10 22:48:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (literal_concat_gen, evstr2dstr_gen): keep literal
+	  encoding beginning with an interpolation same as the source file
+	  encoding.  [ruby-core:70703] [Bug #11519]
+
+Thu Sep 10 22:15:51 2015  Joe Rafaniello  <jrafanie@redhat.com>
+
+	* process.c (rb_f_spawn): Be more specific regarding "other
+	  values" by having "non-zero positive integers" Add nil, the
+	  default value, as a possible value and what it means.
+
+	  Try to use more consistent language.
+	  [Fix GH-1008]
+
+Thu Sep 10 15:16:02 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (getmultiline): refactor.
+
+Thu Sep 10 12:17:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_build_from_ary_body): register cdhash to the
+	  iseq constant body instead of compile time mark array, not to
+	  get GCed.  [ruby-core:70708] [Feature #8543]
+
+Wed Sep  9 18:16:14 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/rubygems/stub_specification.rb (Gem::StubSpecification#data):
+	  should not change the value of $. when `require`ing gems.
+	  this fixed test failures introduced by r51813.
+
+Wed Sep  9 16:55:45 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (usage, enable_option, disable_option, process_options): new
+	  option `--disable-did_you_mean`.
+
+	* gem_prelude.rb: now requires did_you_mean gem by default if available.
+
+Wed Sep  9 13:38:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/extlibs.rb (do_patch): let "patch" command change the
+	  working directory and open the patch file there, instead of
+	  spawn options, so that proper error message will be shown by the
+	  command not just "chdir" or "open".
+
+Wed Sep  9 11:33:05 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk (update-gems): use BASERUBY instead of RUNRUBY.
+
+Wed Sep  9 11:08:59 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* lib/delegate.rb: Remove backtrace cleaning for delegated methods
+	  This patch was provided by Rafael Franca and greatly improves
+	  performance when an exception is raised. [Bug #11461]
+
+Wed Sep  9 10:05:41 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/rubygems/test_config.rb: fix broken tests for Windows platform.
+
+Wed Sep  9 07:46:32 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems: Update to RubyGems HEAD(fe61e4c112).
+	  this version contains new feature that warn invalid SPDX license
+	  identifiers. https://github.com/rubygems/rubygems/pull/1249
+	  and #1032, #1023, #1332, #1328, #1306, #1321, #1324
+	* test/rubygems: ditto.
+
+Tue Sep  8 23:17:36 2015  Yuki Nishijima  <mail@yukinishijima.net>
+
+	* gems/bundled_gems: Upgrade the did_you_mean gem to 1.0.0.beta2.
+
+Tue Sep  8 23:09:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_s_popen): do not wait the child process during being
+	  killed.  [ruby-core:70671] [Bug #11510]
+
+Tue Sep  8 22:18:04 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gems/bundled_gems: revert because ruby trunk never be able to install
+	  the did_you_mean gem.  retry after enough test.
+
+Tue Sep  8 21:48:22 2015  Yuki Nishijima  <mail@yukinishijima.net>
+
+	* gems/bundled_gems: Automatically install the did_you_mean gem
+	  as a bundled gem.  [Feature #11252]
+
+Tue Sep  8 17:17:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove rb_call_info_t::blockiseq.
+
+	* insns.def (send, invokesuper): pass blockiseq explicitly.
+
+	* compile.c: catch up this fix.
+
+	* iseq.c: ditto.
+
+	* vm_args.c: ditto.
+
+	* iseq.c (ISEQ_MINOR_VERSION): 2->3 because instruction spec was
+	  changed.
+
+Tue Sep  8 15:01:19 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (list): fetch all the lines before yielding a block
+	  to allow other commands in the block.  [Feature #11454]
+	  Patched by Srikanth Shreenivas.
+
+Tue Sep  8 12:05:00 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_read_reparse_point): return correct required
+	  buffer size for IO_REPARSE_TAG_MOUNT_POINT.
+
+Tue Sep  8 00:14:43 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_execarg_parent_start1): raise with the target path
+	  name when open() failed.
+
+Mon Sep  7 23:45:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_exec_fail): raise with the target directory name
+	  when chdir() failed.  pointed out by sorah.
+
+Mon Sep  7 22:05:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (insert): should use plain strdup() instead of
+	  ruby_strdup() at startup time, and plain free()ed in cmdglob().
+
+Mon Sep  7 16:49:30 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_vm_struct): define objspace always regardless
+	  ENABLE_VM_OBJSPACE.
+
+Mon Sep  7 15:54:58 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby_atomic.h (ATOMIC_VALUE_CAS): fix typo.
+	  TODO: make arguments of all CAS macros consistent.
+
+Sun Sep  6 16:07:22 2015  Eric Wong  <e@80x24.org>
+
+	* ccan/list/list.h: suppress unused argument warnings
+	  [ccan commit 6aaca17e07588997417a73fac19dcf0ff17ed81b]
+
+Sat Sep  5 11:39:52 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rss/rss.rb (Time#w3cdtf): fix zero-trimmed width of fraction
+	  digits.  [ruby-core:70667] [Bug #11509]
+
+Sat Sep  5 08:28:58 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_equal, rb_hash_eql): [DOC] the orders of each
+	  hashes are not compared.  [Bug #11508]
+
+Fri Sep  4 23:26:22 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h: fix macro name for VC runtime version,
+	  RT_VER is only in Makefile.
+
+Fri Sep  4 17:46:17 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/contributing.rdoc: fix configuration option.
+	  [ci skip] [fix GH-1009]
+
+Fri Sep  4 04:46:54 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_memsize): functions for wrapper object should have
+	  iseqw_ prefix.
+
+Thu Sep  3 21:12:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi/session.rb (create_new_id): use SHA512 instead of MD5.
+	  pointed out by SARWAR JAHAN.
+
+Thu Sep  3 20:29:18 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_raw_obj_info): iseq->body->location.first_lineno is Fixnum.
+
+Thu Sep  3 17:54:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (raise_method_missing): "names" should be singular.
+	  pointed out by Filip Bartuzi.
+
+Thu Sep  3 17:50:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_raw_obj_info): should support IMEMO/iseq.
+
+Thu Sep  3 10:07:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (raise_method_missing): refine error messages when a
+	  symbol is not given.  [Fix GH-1013]
+
+Wed Sep  2 18:49:55 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/*: merge psych master(8737e5b). It contains following fixes.
+	  https://github.com/tenderlove/psych/pull/242
+	  https://github.com/tenderlove/psych/pull/246 [ruby-list:50219]
+	* test/psych/*: ditto.
+
+Wed Sep  2 18:04:13 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h (GET_PC_COUNT): remove unused macro.
+
+Wed Sep  2 17:18:37 2015  Chris Schneider  <chris@christopher-schneider.com>
+
+	* process.c (proc_detach): [DOC] fix typo "intent" as "intend" in
+	  rdoc.  [Fix GH-1011]
+
+Wed Sep  2 16:58:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_realpath_internal): use filesystem encoding if the
+	  argument is in ASCII encodings.
+
+	* win32/file.c (rb_readlink): needs the result encoding.
+
+Tue Sep  1 18:37:15 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/thread/test_queue.rb: catch up last commit.
+
+Tue Sep  1 18:16:32 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_sync.c (queue_do_close): ignore multiple close to allow
+	  multiple producers.
+	  https://bugs.ruby-lang.org/issues/10600#note-14
+
+Tue Sep  1 18:06:26 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_tools.c: rename thread_tools.c to thread_sync.c.
+
+Mon Aug 31 17:04:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (move_refined_method): should insert a write barrier
+	  from an original class to a created (cloned) method entry.
+
+	* test/ruby/test_refinement.rb: add a test.
+
+Sun Aug 30 02:42:22 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_method_tab): Only add SSLv3 support
+	  if the SSL library supports it.  Thanks Kurt Roeckx <kurt@roeckx.be>
+	  [Bug #11376]
+
+	* ext/openssl/extconf.rb: check for SSLv3 support in the SSL
+	  implementation.
+
+	* test/openssl/test_ssl.rb (class OpenSSL): Skip tests that need SSLv3
+	  if there is no support.
+
+Fri Aug 28 16:05:09 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rdoc/*: Update rdoc master(cfffed5)
+	  https://github.com/rdoc/rdoc/pull/337
+	  https://github.com/rdoc/rdoc/pull/367
+
+Fri Aug 28 10:16:20 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (hook_before_rewind): prevent kicking :return event while
+	  finishing vm_exec func because invoke_block_from_c() kick a :return
+	  event for bmethods.
+	  [Bug #11492]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Thu Aug 27 18:05:42 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/webrick/server.rb: use IO::NULL instead of '/dev/null'
+	* test/ruby/test_string.rb: ditto.
+
+Thu Aug 27 15:24:57 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c (iseq_set_sequence): rename variable names
+	  to make it readable.
+
+Thu Aug 27 07:45:34 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_tools.c: add Queue#close(exception=false) and
+	  SizedQueue#close(exception=false).
+	  [Feature #10600]
+
+	  Trying to deq from a closed empty queue return nil
+	  if exception parameter equals to false (default).
+
+	  If exception parameter is truthy, it raises
+	  ClosedQueueError (< StopIteration).
+	  ClosedQueueError inherits StopIteration so that you can write:
+
+	    loop{ e = q.deq; (using e) }
+
+	  Trying to close a closed queue raises ClosedQueueError.
+
+	  Blocking threads to wait deq for Queue and SizedQueue will be
+	  restarted immediately by returning nil (exception=false) or
+	  raising a ClosedQueueError (exception=true).
+
+	  Blocking threads to wait enq for SizedQueue will be
+	  restarted by raising a ClosedQueueError immediately.
+
+	  The above specification is not proposed specification, so that
+	  we need to continue discussion to conclude specification this
+	  method.
+
+	* test/thread/test_queue.rb: add tests originally written by
+	  John Anderson and modify detailed behavior.
+
+Wed Aug 26 10:52:02 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_memsearch_wchar, rb_memsearch_qchar): test matching
+	  till the end of string.  [ruby-core:70592] [Bug #11488]
+
+	* test/ruby/test_m17n.rb (test_include?, test_index): add tests by
+	  Tom Stuart.
+
+Wed Aug 26 09:26:00 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* id_table.c (list_table_extend, hash_table_extend): remove C99
+	  features.  [ruby-dev:49239] [Bug #11487]
+
+Tue Aug 25 06:34:43 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (w32_symlink): implement symlink().
+
+Mon Aug 24 16:01:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (rb_locale_encindex): find encoding index without
+	  making a string object every time.  [ruby-core:58160] [Bug #9080]
+
+Sat Aug 22 15:43:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (check_funcall_failed, check_funcall_missing): cache
+	  results of respond_to? and respond_to_missing?, and search a
+	  public method only for compatibility with rb_respond_to.
+
+Sat Aug 22 08:23:32 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/thread/thread.c: move definitions of Queue, SizedQueue
+	  and ConditionalVariables to thread_tools.c. In other words,
+	  such classes are built-in.
+	  [Feature #8919]
+
+	  At first, I planned to embed only a Queue class.
+	  However, rubygems requires 'thread.rb' (rubygems are
+	  required at first, when launch MRI without --disable-gems).
+	  So most of people require 'thread.rb' as an embedded library.
+
+	  Now, ext/thread/thread.c is empty, only for a dummy for
+	  compatibility.
+
+	* thread.c: move a definition of Mutex class to thread_tools.c.
+
+	  And define Mutex class under Thread (so now Mutex is Thread::Mutex).
+	  Because other thread related classes are also defined under Thread.
+	  We remain ::Mutex as Thread::Mutex. Only an inspect result is changed.
+
+	* common.mk: add dependency from thread.o to thread_tools.c.
+
+Sat Aug 22 05:31:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_opts.h, iseq.c, iseq.h: add compile option to force frozen
+	  string literals.
+	  [Feature #11473]
+
+	  This addition is not specification change, but to try frozen
+	  string literal world discussed on [Feature #11473].
+
+	  You can try frozen string literal world using this magical line:
+
+	    RubyVM::InstructionSequence.compile_option =
+	      {frozen_string_literal: true}
+
+	  Note that this is a global compilation option, so that you need to
+	  compile another script like that:
+
+	    p 'foo'.frozen? #=> false
+	    RubyVM::InstructionSequence.compile_option =
+	      {frozen_string_literal: true}
+	    p 'foo'.frozen? #=> false, because this line is already compiled.
+	    p eval("'foo'.frozen?") #=> true
+
+	  Details:
+	    * String literals are deduped by rb_fstring().
+	    * Dynamic string literals ("...#{xyz}...") is now only frozen,
+	      not deduped. Maybe you have other ideas.
+
+	  Now, please do not use this option on your productions :)
+	  Of course, current specification can be changed.
+
+	* compile.c: ditto.
+
+	* test/ruby/test_iseq.rb: add a test.
+
+Sat Aug 22 02:53:12 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/psych/*: update to Psych 2.0.14
+
+	* test/psych/*: ditto
+
+Fri Aug 21 19:58:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c: add a new method ObjectSpace.count_symbols.
+	  [Feature #11158]
+
+	* symbol.c (rb_sym_immortal_count): added to count immortal symbols.
+
+	* symbol.h: ditto.
+
+	* test/objspace/test_objspace.rb: add a test for this method.
+
+	* NEWS: describe about this method.
+
+Fri Aug 21 19:48:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/Makefile.sub ($(LIBRUBY_SO)): needs additional libraries
+	  for extension libraries to link statically.
+	  [ruby-core:70499] [Feature #9018]
+
+Fri Aug 21 18:49:22 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h, cont.c, vm_trace.c: add a new event
+	  fiber_switch. We need more discussion about this feature
+	  so that I don't write it on NEWS.
+	  [Feature #11348]
+
+	* test/ruby/test_settracefunc.rb: add tests.
+
+Fri Aug 21 17:32:42 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_invoke_block): we should not expect ci->argc is
+	  stable after invoking a block. [Bug #11451]
+
+	* test/ruby/test_yield.rb: add a test. This test script is given by
+	  Alex Dowad.
+
+Fri Aug 21 06:35:50 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* test/openssl/test_ssl_session.rb: Fix tests so that they take in to
+	  account OpenSSL installations that have SSLv3 disabled by default.
+	  Thanks Jeremy Evans <code@jeremyevans.net> for the patches.
+	  [Bug #11366] [Bug #11367]
+
+Thu Aug 20 22:19:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (basic_obj_respond_to): call respond_to_missing?
+	  only when redefined.  [ruby-core:70460] [Bug #11465]
+
+Thu Aug 20 14:13:27 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (check_funcall_respond_to): share the behavior with
+	  rb_obj_respond_to.  [ruby-core:70460] [Bug #11465]
+
+	* vm_method.c (vm_respond_to): extract from rb_obj_respond_to and
+	  merge r39881.
+
+Thu Aug 20 08:53:09 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_obj_respond_to): reuse found method entry
+	  instead of searching same entry repeatedly.
+
+Thu Aug 20 08:31:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (replace_real_basename), win32/win32.c (opendir_internal):
+	  check reparse point tags and treat supported tags only as
+	  symbolic links.  [ruby-core:70454] [Bug #11462]
+
+Wed Aug 19 23:59:28 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): add OP_ALL to
+	  existing options rather than just setting it.  Some vendors apply
+	  custom patches to their versions of OpenSSL that set default values
+	  for options.  This commit respects the custom patches they've
+	  applied.
+
+	* test/openssl/test_ssl.rb (class OpenSSL): check that OP_ALL has been
+	  added to the options.
+
+Wed Aug 19 23:55:29 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_f_spawn): [DOC] elaborate environment variable
+	  values.  [ruby-core:70456] [Bug #11463]
+
+Wed Aug 19 23:48:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (winnt_lstat): check reparse point tags and treat
+	  supported tags only as symbolic links.
+	  [ruby-core:70454] [Bug #11462]
+
+Tue Aug 18 20:05:49 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (reserve_stack): ensure the memory is really
+	  allocated. [Bug #11457]
+
+Tue Aug 18 17:19:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (IS_BEG): include labeled argument state, which was
+	  EXPR_LABELARG.  [ruby-dev:49221] [Bug #11456]
+
+Tue Aug 18 16:16:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RClass): define only in C, `__attribute__`
+	  between `struct` and the name can't compile with g++.
+	  [ruby-core:70297] [Bug #11426]
+
+Mon Aug 17 20:56:36 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y: fix syntax error at do-block after a conditional
+	  operator.  separate label-allowed and after-a-label states from
+	  others as bit flags.  [ruby-dev:48790] [Bug #10653]
+
+Mon Aug 17 11:57:36 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_each_codepoint): raise an exception at incomplete
+	  character before EOF when conversion takes place.  [Bug #11444]
+
+Sun Aug 16 17:33:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: update latest version of bundled gems.
+	  It includes minitest-5.8.0 and test-unit 3.1.3.
+
+Sun Aug 16 17:24:10 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gc.c (gc_mark_children): check if RCLASS_EXT is valid
+	  before marking. This fixes the following test failure
+	  introduced in r51126:
+
+	    make test-all TESTOPTS='--gc-stress ruby/test_refinement.rb'
+
+Sat Aug 15 10:51:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (API#SetValue): data size should
+	  be in bytes, not in chars.  [ruby-core:70365] [Bug #11439]
+
+Sat Aug 15 10:15:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_each_codepoint): read more data when read partially.
+	  [ruby-core:70379] [Bug #11444]
+
+Sat Aug 15 04:33:39 2015  Eric Wong  <e@80x24.org>
+
+	* hash.c (any_hash): skip rb_objid_hash for static syms
+	  (rb_num_hash_start): extract from rb_ident_hash
+	  (rb_objid_hash): call rb_num_hash_start
+	  (rb_ident_hash): ditto
+	  [ruby-core:70181] [Feature #11405]
+
+Sat Aug 15 04:16:13 2015  Eric Wong  <e@80x24.org>
+
+	* iseq.c (rb_iseq_mark): reduce NULL checks
+
+Fri Aug 14 18:50:57 2015  Eric Wong  <e@80x24.org>
+
+	* method.h (METHOD_ENTRY_VISI_SET): cast visi to int
+	  (METHOD_ENTRY_FLAGS_SET): ditto
+
+Fri Aug 14 18:43:11 2015  Eric Wong  <e@80x24.org>
+
+	* process.c (close_unless_reserved): add extra check
+	  (dup2_with_divert): remove
+	  (redirect_dup2): use dup2 without divert
+	  (before_exec_non_async_signal_safe): adjust call + comment
+	  (rb_f_exec): stop timer thread for all OSes
+	  (rb_exec_without_timer_thread): remove
+	* eval.c (ruby_cleanup): adjust call
+	* thread.c (rb_thread_stop_timer_thread): always close pipes
+	* thread_pthread.c (struct timer_thread_pipe): add writing field,
+	  mark owner_process volatile for signal handlers
+	  (rb_thread_wakeup_timer_thread_fd): check valid FD
+	  (rb_thread_wakeup_timer_thread): set writing flag to prevent close
+	  (rb_thread_wakeup_timer_thread_low): ditto
+	  (CLOSE_INVALIDATE): new macro
+	  (close_invalidate): new function
+	  (close_communication_pipe): removed
+	  (setup_communication_pipe_internal): make errors non-fatal
+	  (setup_communication_pipe): ditto
+	  (thread_timer): close reading ends inside timer thread
+	  (rb_thread_create_timer_thread): make errors non-fatal
+	  (native_stop_timer_thread): close write ends only, always,
+	   wait for signal handlers to finish
+	  (rb_divert_reserved_fd): remove
+	* thread_win32.c (native_stop_timer_thread): adjust (untested)
+	  (rb_divert_reserved_fd): remove
+	* vm_core.h: adjust prototype
+	  [ruby-core:70386] [Bug #11336]
+
+Fri Aug 14 18:40:43 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (API#SetValue): add terminator
+	  size, not 1 byte.  [ruby-core:70365] [Bug #11439]
+
+Thu Aug 13 22:49:42 2015  Juanito Fatas  <katehuang0320@gmail.com>
+
+	* lib/timeout.rb (Timeout#timeout): freeze a string message to
+	  reduce string allocations.  [Fix GH-996]
+
+Thu Aug 13 17:42:34 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (rb_call_info_kw_arg_bytes): move the definition
+	  to iseq.h because this function is shared with iseq.c and compile.c.
+
+Thu Aug 13 14:36:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_num_to_dbl): move from num2dbl_with_to_f in math.c.
+
+Thu Aug 13 09:01:25 2015  Eric Wong  <e@80x24.org>
+
+	* load.c (features_index_add): avoid repeat calculation
+
+Wed Aug 12 21:57:31 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* id_table.c: IMPL() macro accept op as _opname instead of opname
+	  because jemalloc seems to replace the word `free' to `je_free'.
+
+Wed Aug 12 21:51:11 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* id_table.c (mix_id_table_insert): fix memory leak.
+
+Wed Aug 12 21:17:38 2015  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_memsize): reimplement for wrapper
+	  (param_keyword_size): extracted from iseq_memsize
+	  (iseqw_mark): new mark function
+	  (iseqw_data_type): new data type
+	  (iseqw_new): wrap as iseqw_data_type
+	  (iseqw_check): adjust for wrapper
+	  (Init_ISeq): remove iseqw_iseq_key initialization
+	* test/objspace/test_objspace.rb: new test
+	  [ruby-core:70344] [Feature #11435]
+
+Wed Aug 12 21:15:27 2015  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_call_info_kw_arg_bytes): extract from compile.c
+	* compile.c (iseq_build_callinfo_from_hash): use above function
+
+Wed Aug 12 18:00:17 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (move_refined_method): same as the last commit.
+
+Wed Aug 12 17:57:53 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c, gc.c vm.c: use ID_TABLE_* instead of ST_*
+	  (such as ST_CONTINUE) for enum rb_id_table_iterator_result.
+
+Wed Aug 12 17:05:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* id_table.h: introduce ID key table.
+	  [Feature #11420]
+
+	  This table only manage ID->VALUE table to reduce overhead of st.
+
+	  Some functions prefixed rb_id_table_* are provided.
+
+	* id_table.c: implement rb_id_table_*.
+
+	  There are several algorithms to implement it.
+
+	  Now, there are roughly 4 types:
+
+	    * st
+	    * array
+	    * hash (implemented by  Yura Sokolov)
+	    * mix of array and hash
+
+	  The macro ID_TABLE_IMPL can choose implementation.
+	  You can see detailes about them at the head of id_table.c.
+
+	  At the default, I choose 34 (mix of list and hash).
+	  This is not final decision.
+	  Please report your suitable parameters or
+	  your data structure.
+
+	  * symbol.c: introduce rb_id_serial_t and rb_id_to_serial()
+	    to represent ID by serial number.
+
+	  * internal.h: use id_table for method tables.
+
+	  * class.c, gc.c, marshal.c, vm.c, vm_method.c: ditto.
+
+Wed Aug 12 05:19:11 2015  Eric Wong  <e@80x24.org>
+
+	* parse.y (rb_parser_compile_cstr): remove volatile arg
+	  (rb_parser_compile_string): ditto
+	  (rb_parser_compile_file): ditto
+	  (rb_parser_compile_string_path): ditto
+	  (rb_parser_compile_file_path): ditto
+	  [ruby-core:70323] [Misc #11431]
+
+Tue Aug 11 22:59:57 2015  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (Init_Numeric): Fix document for Float::MIN and
+	  Float::EPSILON.
+
+Tue Aug 11 15:22:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ALLOCV_N): check integer overflow, as well
+	  as ruby_xmalloc2.  pointed out by Paul <pawlkt AT gmail.com>.
+
+Tue Aug 11 14:57:09 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_repeated_permutation): fix buffer size, ALLOCV_N
+	  already multiplies element size.
+
+Tue Aug 11 12:13:20 2015  Jeremy Evans  <merch-redmine@jeremyevans.net>
+
+	* test/openssl/test_ssl.rb: Fix LocalJumpErrors being raised
+	  in OpenSSL tests. [ruby-core:70020][Bug #11368]
+
+Tue Aug 11 11:54:13 2015  Alexey Lipnyagov  <liptonshmidt@gmail.com>
+
+	* string.c: Fix documentation for String#slice
+	  [ruby-core:70298][Bug #11427]
+
+Tue Aug 11 11:53:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (superclass): make superclass rule optional and allow
+	  any contents without a terminator.  [EXPERIMENTAL]
+
+Tue Aug 11 10:58:42 2015  Juanito Fatas  <juanitofatas@gmail.com>
+
+	* string.c: [DOC] Make #end_with? example doc symmetry
+	  with #start_with? [fix GH-992][ci skip]
+
+Tue Aug 11 10:51:19 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/ruby/test_array.rb: Add test for `Array#flatten` with level 1
+	  [fix GH-986] Patch @yui-knk
+
+Tue Aug 11 10:48:16 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* enum.c: added doc for Enumerable#zip
+	  [fix GH-985] Patch by @yui-knk
+	* test/ruby/test_enum.rb: added tests for Enumerable#zip
+	  [fix GH-985] Patch @yui-knk
+
+Tue Aug 11 10:33:26 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* vm_method.c: typo fix [fix GH-993][ci skip] Patch by @0x0dea
+	* test/ruby/test_refinement.rb: ditto.
+
+Sun Aug  9 14:15:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (Init_vm_objects, rb_vm_fstring_table): use frozen_strings
+	  table in rb_vm_t.  [ruby-core:70274] [Bug #11423]
+
+Sat Aug  8 03:59:51 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* object.c: [DOC] Improve grammar for Module#===
+	  Patch by @SkyBirdSoar in documenting-ruby/ruby#52:
+	  https://github.com/documenting-ruby/ruby/pull/52
+
+Sat Aug  8 03:39:33 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* hash.c: [DOC] Improve description of symbol key syntax
+	  Patch by Raphael Das Gupta in documenting-ruby/ruby#51:
+	  https://github.com/documenting-ruby/ruby/pull/51
+
+Fri Aug  7 21:04:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_params): turn in_def and in_single into bit
+	  flags and reduce the size by 2-words.
+
+	* parse.y (parser_params): remove redundant prefixes.
+
+	* parse.y (yylex): non-pure parser has not been supported since
+	  merger of ripper.  change argument types from void pointers.
+
+Fri Aug  7 17:07:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (method_super_method): uncallable method entry does not
+	  have the defined class, use the owner instead.
+	  [ruby-core:70254] [Bug #11419]
+
+	* test/ruby/test_method.rb (test_super_method_unbound): add test
+	  by Akira Matsuda.
+
+Thu Aug  6 10:49:57 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.c (rb_alloc_tmp_buffer): round up the size and check the
+	  range.
+
+	* ruby_atomic.h (ATOMIC_VALUE_EXCHANGE, ATOMIC_VALUE_CAS): add
+	  atomic operations for VALUE.
+
+Thu Aug  6 08:15:49 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): move
+	  SSLSocket#sysclose to Ruby.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_close): ditto
+
+Thu Aug  6 07:57:21 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): move nonblock
+	  enable to SSLSocket#initialize and remove Nonblock module.
+
+Thu Aug  6 07:53:47 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): move
+	  OpenSSL::SSL::SSLSocket#initialize to Ruby.
+
+	* ext/openssl/ossl_ssl.c: ditto
+
+Thu Aug  6 02:25:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.c (rb_alloc_tmp_buffer): use NODE_ALLOCA to mark locations
+	  like as builtin alloca.  [ruby-core:70251] [Bug #11418]
+
+Wed Aug  5 14:37:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_econv_open0): rb_econv_t::source_encoding_name
+	  and rb_econv_t::destination_encoding_name should refer static
+	  strings always or NULL.  [ruby-core:70247] [Bug #11416]
+
+Tue Aug  4 16:53:43 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): extract callback
+	  lookup to private Ruby methods. This means we can keep the default
+	  DH callback logic hidden from consumers.  Also, since the SSLSocket
+	  always has a context, we can remove conditionals about that
+	  instance.
+
+	* ext/openssl/ossl_ssl.c: move callback lookup methods to private Ruby
+	  methods.
+
+Tue Aug  4 16:40:26 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_module.rb: should not expect a method table ordering.
+	  [Feature #11414]
+
+Tue Aug 04 15:30:04 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_block_clear_env_self): clear by Qfalse instead of Qnil.
+	  [Bug #11409]
+
+	* test/ruby/test_eval.rb: add tests for this issue,
+	  written by @0x0dea.
+	  https://github.com/ruby/ruby/pull/988
+
+Tue Aug  4 12:12:14 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c: wrap long lines
+
+Tue Aug  4 09:32:30 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* proc.c: Removing duplicate doc [fix GH-987][ci skip]
+	  Patch by @ronakjangir47
+
+Tue Aug  4 09:21:58 2015  Richard Schneeman  <richard.schneeman+foo@gmail.com>
+
+	* doc/contributing.rdoc: fixed wrong instructions with OS X
+	  [fix GH-989][ci skip] Patch by @schneems
+
+Mon Aug  3 10:08:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_memsearch): should match only char boundaries in wide
+	  character encodings.  [ruby-core:70220] [Bug #11413]
+
+Sun Aug  2 07:01:17 2015  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/lib/openssl/buffering.rb (gets):
+	  avoid comparing fixnum with nil
+	* test/openssl/test_pair.rb: test gets with limit when EOF is hit
+	  Thanks to Bar Hofesh <bar.hofesh@safe-t.com> for the bug report
+	  and testing.
+	  [ruby-core:70149] [Bug #11400]
+
+Sat Aug  1 17:13:15 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/net/http/response.rb (Net::HTTPResponse::Inflater#finish):
+	  fix a bug that empty gzipped response body causes Zlib::BufError.
+	  [ruby-core:68846] [Bug #11058]
+
+	* test/net/http/test_httpresponse.rb: tests for the above.
+
+Sat Aug  1 17:05:18 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/net/http/response.rb (Net::HTTPResponse#inflater):
+	  fix TypeError. An exception object might be nil.
+	  [ruby-core:68846] [Bug #11058]
+
+Sat Aug  1 09:09:46 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): Implement
+	  SSLContext#options and options= using SSL_CTX_set_options and
+	  SSL_CTX_get_options. This reduces the number of ivars we need and
+	  simplifies `ossl_sslctx_setup`.
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): Default `options`
+	  to SSL_OP_ALL
+
+Sat Aug  1 06:54:36 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): OpenSSL declares these
+	  constants as longs, so we should follow that and use LONG2NUM.
+	  http://git.io/vOqxD
+
+Sat Aug  1 04:06:29 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_call_tmp_dh_callback): change callback
+	  to return the Ruby dh (or ecdh) object that the caller cares about
+	  instead of doing rb_iv_get / set to communicate.  This means we can
+	  remove an rb_iv_get call, and only use the set calls for their
+	  intended purpose (to prevent the object from being GC'd).
+
+	* ext/openssl/ossl_ssl.c (ossl_tmp_dh_callback): ditto
+	* ext/openssl/ossl_ssl.c (ossl_call_tmp_ecdh_callback): ditto
+	* ext/openssl/ossl_ssl.c (ossl_tmp_ecdh_callback): ditto
+
+Sat Aug  1 03:49:31 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_call_tmp_dh_callback): Similarly to the
+	  tmp_ecdh_callback, the SSLSocket instance always holds a reference
+	  to the SSLContext object (it's always set in `initialize`).  The
+	  SSLContext holds a reference to the tmp_dh_callback.  Ask the
+	  context for the callback instead of storing the callback in two
+	  places.
+
+Sat Aug  1 03:43:10 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_call_tmp_dh_callback): create an array
+	  and use `rb_apply` to clean up calls to `rb_protect`.
+
+	* ext/openssl/ossl_ssl.c (ossl_tmp_dh_callback): ditto
+
+Sat Aug  1 03:27:12 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_call_tmp_ecdh_callback): The SSL socket
+	  always holds a reference to the SSLContext object, which will have
+	  the callback object.  Ask the context for the callback instead of
+	  storing the callback in two places.
+
+Sat Aug  1 03:14:07 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): convert
+	  `tmp_dh_callback` to Ruby, and call it when setting up an SSL
+	  connection.  This allows us to move the "default" behavior to the
+	  reader method.
+
+	* ext/openssl/ossl_ssl.c: call the tmp_dh_callback instead of
+	  accessing the SSLContext's internals.
+
+Fri Jul 31 23:34:27 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* .travis.yml: update libssl before running tests.
+	  Thanks to Chris Sinjakli <chris@sinjakli.co.uk> for figuring out the
+	  travis settings!
+
+Fri Jul 31 21:34:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_require_internal): use rb_load_internal0 not to raise
+	  a exception to be caught.
+
+Thu Jul 30 13:19:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_const_get_0): warn deprecated constant reference.
+
+	* variable.c (rb_mod_deprecate_constant): mark constants to be
+	  warned as deprecated.  [Feature #11398]
+
+Thu Jul 30 11:53:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_s_handle_interrupt): make identity hash, to
+	  compare masking classes just by their IDs.
+
+Thu Jul 30 11:52:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_load_internal0): stop separating exits at loading
+	  from exits from execution.  TAG_FATAL is the only case that
+	  `errinfo` is a Fixnum, and should continue to exit by JUMP_TAG
+	  but not raising as an ordinary exception.
+	  [ruby-core:70169] [Bug #11404]
+
+Thu Jul 30 10:42:27 2015  Alex Dowad  <alexinbeijing@gmail.com>
+
+	* load.c (rb_load_internal0): extra check before returning
+	  TAG_RAISE when a non-local transfer of control happens while
+	  loading and parsing a Ruby source file.
+	  [ruby-core:70169] [Bug #11404]
+
+Thu Jul 30 08:48:42 2015  Eric Wong  <e@80x24.org>
+
+	* st.c (find_entry): constify st_table*
+	  (find_packed_index_from): ditto
+	  (find_packed_index): ditto
+	  (get_keys): ditto
+	  (get_values): ditto
+
+Thu Jul 30 04:29:25 2015  Eric Wong  <e@80x24.org>
+
+	* benchmark/bm_hash_aref_dsym.rb: new benchmark
+	* benchmark/bm_hash_aref_dsym_long.rb: ditto
+	* benchmark/bm_hash_aref_fix.rb: ditto
+
+Wed Jul 29 21:38:41 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (any_hash), symbol.c (dsymbol_alloc): fix dynamic symbol
+	  hash value by restricting in Fixnum range, that is `long`.
+
+Wed Jul 29 17:25:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_obj_hash): move in order to share with rb_any_hash.
+
+Wed Jul 29 16:00:22 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_buf_cat): consider empty non-embed string case,
+	  not to loop infinitely.  [ruby-core:70074] [Bug #11383]
+
+Wed Jul 29 15:25:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (send_internal): set method_missing_reason before
+	  invoking overriding method_missing method so that the default
+	  method_missing can achieve it properly.
+	  [ruby-core:68515] [Bug #10969]
+
+Wed Jul 29 14:54:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_sym_hash): return same value as rb_any_hash() of
+	  Symbol.  [Bug #9381]
+
+	* hash.c (rb_any_hash): fix Float hash.  rb_dbl_hash() returns a
+	  Fixnum, but not a long.  [Bug #9381]
+
+Wed Jul 29 11:07:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (LIKELY, UNLIKELY): make a boolean to enforce 1 or 0.
+
+Wed Jul 29 10:44:43 2015  Alex Dowad  <alexinbeijing@gmail.com>
+
+	* gc.c: document argument passed to finalizer proc.
+	  [fix GH-976][ci skip] Patch by @alexdowad
+
+Wed Jul 29 10:36:58 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* io.c (rb_io_extract_modeenc): add option parameter `flags'
+	  to append extra oflags to normal mode.
+	  [Feature #11253] [ruby-core:69539]
+
+Wed Jul 29 04:54:47 2015  Eric Wong  <e@80x24.org>
+
+	* test/rubygems/test_gem_remote_fetcher.rb: pre-generate test key
+	  [ruby-core:70151] [Bug #11397]
+
+Tue Jul 28 10:32:09 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (struct RClass): moved from ruby/ruby.h to hide the
+	  internals.
+
+Tue Jul 28 08:48:29 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_TYPE_ATTRIBUTE): attribute declaration for
+	  types.
+
+Tue Jul 28 07:23:03 2015  Eric Wong  <e@80x24.org>
+
+	* symbol.h (struct RSymbol): add hashval field
+	* symbol.c (dsymbol_alloc): setup hashval field once
+	* hash.c (rb_any_hash): return RSymbol->hashval directly
+	* common.mk: hash.o depends on symbol.h
+	  Thanks to Bruno Escherl <bruno@escherl.net> for the bug report
+	  [ruby-core:70129] [Bug #11396]
+
+Tue Jul 28 03:26:15 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): raise a more
+	  helpful exception when verifying the peer connection and an
+	  anonymous cipher has been selected. [ruby-core:68330] [Bug #10910]
+	  Thanks to Chris Sinjakli <chris@sinjakli.co.uk> for the patch.
+
+	* test/openssl/test_ssl.rb (class OpenSSL): test for change
+
+Mon Jul 27 13:24:11 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* template/id.h.tmpl (ID2ATTRSET): remove an unused macro.
+
+Mon Jul 27 12:21:15 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/openssl/test_ssl.rb: run tests on non-Unix platforms.
+
+Sun Jul 26 19:21:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_alpn_protocol): fix condition
+	  to compile, needs ALPN to be available.  [Feature #9390]
+
+Sun Jul 26 11:29:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (default_handler, Init_signal): discard SIGSYS, ENOSYS
+	  should raise a SystemCallError always instead.
+
+Sun Jul 26 10:26:35 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_call_servername_cb): set the ssl context
+	  object returned by the servername callback on to the socket as an
+	  instance variable.  If the callback allocated a new context object
+	  and didn't keep a reference to it, it could be GC'd out from under
+	  the socket object.
+
+	* test/openssl/test_ssl.rb (class OpenSSL): test for change.
+
+Sun Jul 26 10:07:26 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* test/openssl/test_ssl.rb (class OpenSSL): add test coverage around
+	  OpenSSL::SSL::SSLContext#servername_cb
+
+Sun Jul 26 09:10:32 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: update latest version of bundled power_assert.
+
+Sun Jul 26 08:49:28 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/rubygems/test_gem_remote_fetcher.rb: backport rubygems upstream
+	  change for OpenSSL key length. see detail to
+	  https://github.com/rubygems/rubygems/pull/1290
+
+Sun Jul 26 08:33:03 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/pkey.rb: implement DEFAULT_512 and
+	  DEFAULT_1024 constants in Ruby.
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): Ask PKey for the
+	  default DH callback since it already must check whether openssl has
+	  been compiled with DH support.
+
+	* ext/openssl/ossl_pkey_dh.c (OSSL_PKEY_BN): Remove C definitions of
+	  DEFAULT_512 and DEFAULT_1024
+
+	* ext/openssl/ossl_pkey_dh.c (Init_ossl_dh): ditto
+
+	* test/openssl/test_pkey_dh.rb (class OpenSSL): add test to ensure the
+	  Ruby definitions are the same as the C definitions were.
+
+Sun Jul 26 08:14:59 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): support
+	  specifically setting the tmp_dh_callback to nil.
+
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): ditto
+
+	* test/openssl/test_pair.rb (module OpenSSL): add a test
+
+Sun Jul 26 07:47:14 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): move the default
+	  tmp_dh_callback Ruby code and set it as a default in `initialize`.
+
+	* ext/openssl/ossl_pkey_dh.c (static unsigned char DEFAULT_DH_512_GEN):
+	  move this constant to Ruby.
+
+	* ext/openssl/ossl_pkey_dh.c (static unsigned char DEFAULT_DH_1024_GEN):
+	  ditto
+
+	* ext/openssl/ossl_pkey_dh.c (Init_ossl_dh): ditto
+
+	* ext/openssl/ossl_ssl.c (ossl_tmp_dh_callback): ditto
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): tmp_dh_callback should
+	  always be set, so we can remove this conditional
+
+Sun Jul 26 06:22:24 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* test/openssl/test_pair.rb: add a test ensuring that the default DH
+	  callback is used when no DH callback is specified.
+
+Sun Jul 26 04:08:27 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): add missing
+	  instance variables to squash warnings with alpn.
+
+Sun Jul 26 03:42:19 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb (module OpenSSL): move
+	  OpenSSL::SSL::SSLContext#initialize implementation to pure Ruby.
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_initialize): ditto
+
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): ditto
+
+Sat Jul 25 21:03:45 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (fill_random_bytes_syscall): get rid of blocking when
+	  no entropy is available.  based on the patch by mame in
+	  [ruby-core:70114].  [Bug #11395]
+
+Sat Jul 25 11:05:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_replace_shared_without_enc): fill the terminator
+	  of embedded strings in wide char encodings.
+
+Sat Jul 25 06:38:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: size should be unsigned.
+	  * rb_call_info_t::index
+	  * rb_iseq_constant_body::stack_max
+	  * rb_iseq_constant_body::local_size
+	  * rb_iseq_constant_body::param::size
+	  * rb_iseq_constant_body::local_table_size
+	  * rb_iseq_constant_body::is_size
+	  * rb_iseq_constant_body::callinfo_size
+
+	* iseq.h: same for iseq_catch_table::size.
+
+	* compile.c: catch up these fix.
+
+	* iseq.c: ditto.
+
+	* proc.c: ditto.
+
+	* vm.c: ditto.
+
+	* vm_args.c: ditto.
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+Sat Jul 25 06:00:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_iseq_constant_body::line_info_table.
+
+	* iseq.c: catch up this fix.
+
+Sat Jul 25 05:56:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_iseq_constant_body::param::opt_table and
+	  rb_iseq_constant_body::param::keyword.
+
+	* compile.c: catch up this fix.
+
+Sat Jul 25 04:47:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_iseq_constant_body::catch_table.
+
+	* compile.c (iseq_set_exception_table): catch up this fix.
+
+	* iseq.c: ditto.
+
+	* vm.c (vm_exec): ditto.
+
+Fri Jul 24 21:29:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* st.c (EQUAL, st_delete_safe): fix arguments order to compare
+	  function, searching key is the first and stored key is the
+	  second always.
+
+Fri Jul 24 21:27:29 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (fstr_update_callback): fstring must not be a shared
+	  string, or the content without RSTRING_FSTR may be freed.
+	  [ruby-dev:49188] [Bug #11386]
+
+Fri Jul 24 20:09:43 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/rinda/test_rinda.rb (RingIPv6#prepare_ipv6): prevent to use
+	  IPv6 loopback interface for
+	  Rinda::TestRingFinger#test_make_socket_ipv6_multicast and
+	  Rinda::TestRingFinger#test_make_socket_ipv6_multicast_hops.
+	  The tests are skipped if there are no IPv6 devices other than the
+	  loopback device. [Bug #11394] [ruby-dev:49199]
+
+	* test/rinda/test_rinda.rb (test_make_socket_ipv6_multicast): ditto
+	  for Rinda::TestRingServer#test_make_socket_ipv6_multicast.
+
+	* test/rinda/test_rinda.rb (test_ring_server_ipv6_multicast): ditto
+	  for Rinda::TestRingServer#test_ring_server_ipv6_multicast.
+
+Fri Jul 24 16:35:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (fstr_update_callback): pool bare strings only.
+
+	* string.c (rb_fstring): return the original string with sharing a
+	  fstring if it has extra attributes, not the fstring itself.
+	  [ruby-dev:49188] [Bug #11386]
+
+Fri Jul 24 16:35:34 2015  yui-knk  <spiketeika@gmail.com>
+
+	* file.c (rb_file_s_extname): [DOC] add an example.
+
+	* test/ruby/test_path.rb (test_extname): add tests.  [Fix GH-978]
+	  * path starts with dot ('.a.rb')
+	  * path includes dir name ('a/b/d/test.rb')
+	  * path includes dir name and dir name starts with dot
+	    ('.a/b/d/test.rb')
+
+Thu Jul 23 18:50:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_iseq_constant_body::local_table and
+	  rb_iseq_param_keyword::table and
+	  rb_iseq_param_keyword::default_values.
+
+	* compile.c: catch up this fix.
+
+	* iseq.c: ditto.
+
+Thu Jul 23 17:30:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_iseq_constant_body::iseq_encoded and
+	  rb_control_frame_t::pc.
+
+	* compile.c (rb_iseq_translate_threaded_code): catch up this fix.
+
+	* iseq.c: ditto.
+
+	* vm_exec.c (vm_exec_core): ditto.
+
+Thu Jul 23 10:25:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h: add raw FL macros, which assume always the
+	  argument object is not a special constant.
+
+	* internal.h (STR_EMBED_P, STR_SHARED_P): valid only for T_STRING.
+
+	* string.c: deal with taint flags directly across String instances.
+
+Thu Jul 23 09:05:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lambda_body): pop cmdarg stack for lookahead
+	  token.  [ruby-core:70067] [Bug #11380]
+
+Thu Jul 23 04:03:03 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c: fix tests by not setting the instance
+	  variable on the frozen ssl instance.
+
+Thu Jul 23 03:32:26 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c: add ECDH callback support. [Feature #11356]
+
+	* test/openssl/test_pair.rb: test for ECDH callback support
+
+Thu Jul 23 03:29:49 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/openssl/ossl_ssl.c: add ALPN support. [Feature #9390]
+
+	* ext/openssl/extconf.rb: detect ALPN support in OpenSSL
+
+	* test/openssl/test_ssl.rb: test for ALPN
+
+Wed Jul 22 23:44:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_reverse): reversed string is not a substring,
+	  and should not set coderange of the original string.
+	  [ruby-dev:49189] [Bug #11387]
+
+Wed Jul 22 20:17:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: modify layout of rb_iseq_constant_body.
+
+	  Move frequent accessing fields to upper part.
+
+Wed Jul 22 19:57:47 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove unused declaration of
+	  iseq_compile_data_ensure_node_stack.
+
+Wed Jul 22 19:52:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: separate rb_iseq_body into rb_iseq_constant_body and
+	  rb_iseq_variable_body (rb_iseq_t::variable_body).
+
+	  rb_iseq_variable_body can be modified after compilation.
+
+	* compile.c: use rb_iseq_t::variable_body.
+
+	* iseq.c: ditto.
+
+	* thread.c: ditto.
+
+Wed Jul 22 17:50:35 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/matrix/eigenvalue_decomposition.rb: refine code style.
+	  [fix GH-959][ci skip] Patch by @bogdanvlviv
+
+Wed Jul 22 15:48:47 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/ruby/test_range.rb: Add test case for Range#end with
+	  exclude_end true case. [fix GH-968] Patch by @yui-knk
+
+Wed Jul 22 09:45:31 2015  Maksim Sitnikov  <sitnikovme@undev.ru>
+
+	* numeric.c (num_coerce): [DOC] fix doc for Numeric#coerce,
+	  missing '+'.  [Fix GH-974]
+
+Wed Jul 22 07:24:18 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* make rb_iseq_t T_IMEMO object (type is imemo_iseq).
+
+	  All contents of previous rb_iseq_t is in rb_iseq_t::body.
+	  Remove rb_iseq_t::self because rb_iseq_t is an object.
+
+	  RubyVM::InstructionSequence is wrapper object points T_IMEMO/iseq.
+	  So RubyVM::ISeq.of(something) method returns different wrapper
+	  objects but they point the same T_IMEMO/iseq object.
+
+	  This patch is big, but most of difference is replacement of
+	  iseq->xxx to iseq->body->xxx.
+
+	  (previous) rb_iseq_t::compile_data is also located to
+	  rb_iseq_t::compile_data.
+	  It was moved from rb_iseq_body::compile_data.
+
+	  Now rb_iseq_t has empty two pointers.
+	  I will split rb_iseq_body data into static data and dynamic data.
+
+	* compile.c: rename some functions/macros.
+	  Now, we don't need to separate iseq and iseqval (only VALUE).
+
+	* eval.c (ruby_exec_internal): `n' is rb_iseq_t (T_IMEMO/iseq).
+
+	* ext/objspace/objspace.c (count_imemo_objects): count T_IMEMO/iseq.
+
+	* gc.c: check T_IMEMO/iseq.
+
+	* internal.h: add imemo_type::imemo_iseq.
+
+	* iseq.c: define RubyVM::InstructionSequence as T_OBJECT.
+	  Methods are implemented by functions named iseqw_....
+
+	* load.c (rb_load_internal0): rb_iseq_new_top() returns
+	  rb_iseq_t (T_IMEMO/iesq).
+
+	* method.h (rb_add_method_iseq): accept rb_iseq_t (T_IMEMO/iseq).
+
+	* vm_core.h (GetISeqPtr): removed because it is not T_DATA now.
+
+	* vm_core.h (struct rb_iseq_body): remove padding for
+	  [Bug #10037][ruby-core:63721].
+
+Wed Jul 22 07:15:33 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (total_i): no need to skip singleton classes.
+
+Wed Jul 22 06:37:54 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_call_info_t::kw_arg,
+	  rb_control_frame_t::iseq and rb_control_frame_t::block_iseq.
+
+	* iseq.c (iseq_free): catch up this fix.
+
+	* vm.c: ditto.
+
+	* vm_dump.c: ditto.
+
+Wed Jul 22 06:25:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_call_info_t::blockiseq and rb_block_t::iseq.
+
+	* vm.c, vm_insnhelper.c: catch up this fix.
+
+	* iseq.c (iseq_data_to_ary): constify the first iseq parameter.
+
+	* vm_insnhelper.c (vm_make_proc_with_iseq): ditto.
+
+Wed Jul 22 06:17:35 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: constify rb_method_iseq_t::iseqptr.
+
+	* proc.c (rb_method_entry_min_max_arity): catch up this fix.
+
+	* vm_insnhelper.c (def_iseq_ptr): constify.
+
+Wed Jul 22 03:37:39 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (internal_object_p): Now a singleton classes appear by
+	  ObjectSpace.each_object. [Bug #11360]
+
+	* test/ruby/test_objectspace.rb: add a test about it.
+
+Tue Jul 21 21:21:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (do_select): replace switch and goto with a loop to
+	  suppress maybe-uninitialized warnings by gcc6.
+
+	* thread.c (set_unblock_function, rb_wait_for_single_fd): ditto.
+
+Tue Jul 21 20:32:33 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/ruby/test_range.rb: Add test for Range#hash
+	  [fix GH-969] Patch by @yui-knk
+
+Tue Jul 21 19:43:20 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c: constify the first parameter (iseq).
+	  * iseq_add_mark_object()
+	  * iseq_add_mark_object_compile_time()
+
+	* iseq.c, iseq.h (rb_iseq_add_mark_object): ditto.
+
+Tue Jul 21 16:18:48 2015  Eric Wong  <e@80x24.org>
+
+	* test/socket/test_nonblock.rb: increase buffer sizes
+	  to OpenBSD limits. Thanks to Jeremy Evans <code@jeremyevans.net>
+	  [ruby-core:70058]
+
+Tue Jul 21 16:08:53 2015  Eric Wong  <e@80x24.org>
+
+	* load.c (ruby_dln_librefs): make static
+
+Tue Jul 21 13:36:54 2015  yuuji.yaginuma  <yuuji.yaginuma@gmail.com>
+
+	* lib/optparse.rb (complete): [DOC] fix typo.  [Fix GH-973]
+
+Tue Jul 21 05:20:21 2015  Eric Wong  <e@80x24.org>
+
+	* io.c (nogvl_wait_for_single_fd): new function for Linux
+	  (maygvl_copy_stream_wait_read): Linux-specific version
+	  (nogvl_copy_stream_wait_write): use nogvl_wait_for_single_fd
+	  [ruby-core:70051] [Feature #11377]
+
+Mon Jul 20 15:04:30 2015  Eric Wong  <e@80x24.org>
+
+	* parse.y (parser_initialize): avoid redundant zero-ing
+
+Mon Jul 20 12:12:05 2015  Eric Wong  <e@80x24.org>
+
+	* parse.y (struct parser_params): pack: 88 => 256 bytes on 64-bit
+	  [ruby-core:70034] [Feature #11371]
+
+Sun Jul 19 14:29:18 2015  windwiny  <windwiny.ubt@gmail.com>
+
+	* ext/pty/pty.c: [DOC] fix example typo, an old name at move from
+	  PTY.open.  [Fix GH-972]
+
+Sat Jul 18 21:29:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (vm_check_ints_blocking): gather common statements at
+	  the end, and prefer LIKELY for Visual C optimization.
+
+Sat Jul 18 20:44:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_load_internal0): do not raise any exceptions but
+	  return the result tag state.
+
+	* load.c (rb_load_protect): reduce nested EXEC_TAGs.
+
+Sat Jul 18 19:52:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (run_finalizer): set and restore safe level here to reduce
+	  nested EXEC_TAGs.
+
+Sat Jul 18 18:45:22 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ruby_cleanup): error_handle() returns exit status to the
+	  system, not internal error state, do not convert the exit status
+	  again.
+
+Sat Jul 18 10:29:03 2015  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_process.rb: test thread+sigs work after failed exec
+
+Sat Jul 18 07:20:18 2015  Jeremy Evans  <code@jeremyevans.net>
+
+	* test/socket/test_nonblock: use smaller buffer for sendmsg
+	  [ruby-core:70016] [Bug #11364]
+
+Sat Jul 18 07:04:24 2015  Eric Wong  <e@80x24.org>
+
+	* signal.c (trap_handler): cleanup to use RSTRING_GETMEM + memcmp
+
+Sat Jul 18 02:53:06 2015  Eric Wong  <e@80x24.org>
+
+	* io.c (argf_read_nonblock): support `exception: false'
+	  (io_nonblock_eof): new function
+	  (io_read_nonblock): use io_nonblock_eof
+	  (argf_getpartial): accept kwargs hash for `exception: false'
+	* test/ruby/test_argf.rb (test_read_nonblock): new test
+	* NEWS: add item for ARGF.read_nonblock
+	  [ruby-core:70000] [Feature #11358]
+
+Fri Jul 17 23:51:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_eval_cmd): $SAFE=4 has been deprecated.
+
+Fri Jul 17 22:18:09 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): use enum ruby_tag_type names.
+
+	* vm_core.h (ruby_tag_type): move from eval_intern.h for compiling
+	  break/next/redo/return.
+
+Fri Jul 17 15:39:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h (ENC_CODERANGE_CLEAN_P): predicate that
+	  tells if the coderange is clean, that is 7bit or valid, and no
+	  needs to scrub.
+
+	* re.c (rb_reg_expr_str): use ENC_CODERANGE_CLEAN_P.
+
+	* string.c (enc_strlen, rb_enc_cr_str_buf_cat, rb_str_scrub):
+	  ditto.
+
+	* string.c (rb_str_enumerate_chars): ditto, and suppress a warning
+	  by gcc6.
+
+Fri Jul 17 15:36:52 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test/ruby/test_range.rb (test_first_last): Add test for
+	  `Range.new`.  [Fix GH-971]
+
+Fri Jul 17 15:36:40 2015  yui-knk  <spiketeika@gmail.com>
+
+	* test/ruby/test_range.rb (test_first_last): Add assertions to
+	  test of `Range#last` with exclude_end true case.  [Fix GH-970]
+
+Fri Jul 17 09:59:14 2015  Eric Wong  <e@80x24.org>
+
+	* thread.c (rb_thread_alone): simplify
+
+Fri Jul 17 09:58:32 2015  Eric Wong  <e@80x24.org>
+
+	* lib/rinda/tuplespace.rb: remove enumerator require
+	* test/pathname/test_pathname.rb: ditto
+
+Fri Jul 17 05:33:58 2015  Eric Wong  <e@80x24.org>
+
+	* iseq.c (rb_iseq_compile_with_option): reuse result of previous
+	  GET_THREAD() call
+	* thread.c (thread_create_core): ditto
+	  (rb_mutex_trylock): ditto
+	  (rb_mutex_lock): ditto
+	* process.c (rb_waitpid): avoid multiple eval from RUBY_VM_CHECK_INTS
+	* thread.c (rb_thread_check_ints): ditto
+
+Thu Jul 16 19:12:30 2015  Eric Wong  <e@80x24.org>
+
+	* thread.c (mutex_alloc): remove needless volatile
+
+Thu Jul 16 22:05:29 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: constify rb_iseq_t::parent_iseq.
+
+	  rb_iseq_t::local_iseq is not constant data because
+	  local_iseq::flip_cnt can be modified (commented).
+
+	* compile.c: catch up this fix.
+
+	* iseq.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+Thu Jul 16 21:47:47 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* process.c (redirect_dup2): when the new FD of dup2() conflicts
+	  with one of the timer thread FDs, the internal FD is diverted.
+	  [Bug #11336] [ruby-core:69886] [Bug #11350] [ruby-core:69961]
+
+	* process.c (dup2_with_divert): new function for the above purpose.
+
+	* thread_pthread.c (rb_divert_reserved_fd): new function for
+	  diverting reserved FD. If the given FD is the same as one of the
+	  reserved FDs, the reserved FD number is internally changed.
+	  It returns -1 when error. Otherwise, returns 0. It also returns
+	  0 if there is no need to change reserved FD number.
+
+	* thread_win32.c (rb_divert_reserved_fd): always returns 0 because
+	  of no reserved FDs.
+
+	* internal.h (rb_divert_reserved_fd): prototype declaration.
+	  It is Ruby internal use only.
+
+Thu Jul 16 21:47:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (rb_iseq_disasm): rename rb_iseq_t *iseqdat to iseq
+	  and VALUE *iseq to code.
+
+	* iseq.c (rb_iseq_disasm_insn): ditto.
+
+Thu Jul 16 14:34:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (REWIND_CFP): keep the arguments region inside the valid
+	  value stack.  [ruby-core:69969] [Bug #11352]
+
+Thu Jul 16 11:38:21 2015  Eric Wong  <e@80x24.org>
+
+	* process.c (close_unless_reserved): declare type of `fd' arg
+
+Thu Jul 16 08:47:29 2015  Eric Wong  <e@80x24.org>
+
+	* load.c (rb_construct_expanded_load_path): fstring expanded path
+	  (get_loaded_features_index): fstring feature path
+	  (rb_provide_feature): ditto
+	  [ruby-core:69871] [Feature #11331]
+
+Thu Jul 16 02:56:14 2015  Eric Wong  <e@80x24.org>
+
+	* thread.c (thread_initialize): avoid RSTRING_PTR and NUMT2INT
+
+Thu Jul 16 01:00:46 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/ruby/test_process.rb (test_exec_close_reserved_fd): test for
+	  [Bug #11353]
+
+Thu Jul 16 00:35:42 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* .gitignore: ignore version.i.
+
+Wed Jul 15 23:40:32 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/ruby/test_process.rb (test_deadlock_by_signal_at_forking):
+	  use RUBY (= EnvUtil.rubybin)
+
+Wed Jul 15 23:01:22 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* process.c (redirect_close, parent_redirect_close): should not close
+	  reserved FD. It should be closed in the exec system call due to the
+	  O_CLOEXEC or FD_CLOEXEC flag. [Bug #11353] [ruby-core:69977]
+
+	* process.c (close_unless_reserved): new function to close FD unless
+	  it is reserved for internal communication.
+
+	* thread_pthread.c (rb_reserved_fd_p): should check owner_process pid
+	  to avoid false positive in forked child process.
+
+Wed Jul 15 18:31:18 2015  Eric Wong  <e@80x24.org>
+
+	* proc.c (proc_mark): remove redundant check
+	* vm.c (env_mark): ditto
+
+Wed Jul 15 17:27:40 2015  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_mark): remove check for data pointer
+	* proc.c (binding_mark): ditto
+	* vm.c (rb_thread_mark): ditto
+	* vm_trace.c (tp_mark): ditto
+
+Wed Jul 15 16:55:04 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_autoload): drop dummy encoding flag from
+	  the loaded encoding index.  this flag is used only in this
+	  source.
+
+Wed Jul 15 14:39:29 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_make_env_each): add comments about env layout.
+	  Do not use `i' to specify `new_ep'.
+
+	* vm.c (rb_proc_create, rb_vm_make_proc_lambda): envval is not used.
+
+Wed Jul 15 08:59:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.h (RUBY_MARK_UNLESS_NULL): evaluate the argument only once
+	  to get rid of inadvertent side effects.
+
+Wed Jul 15 02:53:11 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, vm.c: remove rb_proc_t::envval because we can know it via
+	  rb_proc_t::block::ep.
+
+	  rb_vm_proc_envval(const rb_proc_t *proc) returns an Env object which
+	  the Proc object use.
+
+	* proc.c: catch up this fix.
+
+	* vm_dump.c (rb_vmdebug_proc_dump_raw): ditto.
+
+Wed Jul 15 02:27:22 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h, vm.c: remove rb_env_t::prev_envval because we can know it
+	  via env->ep.
+
+	  rb_vm_env_prev_envval(env) returns prev_envval via env->ep.
+
+	* vm_core.h (rb_vm_env_local_variables): change parameter type
+	  from VALUE (T_DATA/env) to `const rb_env_t *' to make same as
+	  rb_vm_env_prev_envval().
+
+	* proc.c: catch up these changes.
+
+	* vm_dump.c: ditto.
+
+	* vm.c: rename macros.
+
+	  * ENV_IN_HEAP_P() to VM_EP_IN_HEAP_P() because it uses ep.
+	  * ENV_VAL() to VM_ENV_EP_ENVVAL() because it is too short.
+
+Wed Jul 15 01:09:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: refactoring Proc/Env related code.
+
+	* vm_core.h: remove blockprocval field from rb_proc_t and rb_binding_t.
+	  Instead of this field, mark given block in Proc at rb_env_t::env.
+
+	* vm.c (vm_make_env_each): make an Env object with this layout.
+	  And also simplify parameters.
+
+	* proc.c: catch up this fix.
+
+	* vm_core.h: remove rb_env_t::local_size because it is not used.
+
+	* vm_dump.c (rb_vmdebug_env_dump_raw): catch up this fix.
+
+	* vm_core.h (rb_vm_make_env_object): remove rb_vm_make_env_object()
+	  because it is only referred from vm.c.
+
+	* vm_eval.c (eval_string_with_cref): catch up this fix.
+
+Wed Jul 15 00:03:36 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* gc.c (__has_feature): move into internal.h.
+
+	* internal.h (__has_feature): ditto.
+
+	* internal.h (__has_extension): new macro.
+
+	* internal.h (STATIC_ASSERT): use _Static_assert with
+	  clang. [ruby-core:69931] [Bug #11343]
+
+Wed Jul 15 00:00:00 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* random.c (fill_random_bytes_syscall): fix compile error with
+	  clang. [ruby-core:69931] [Bug #11343]
+
+Tue Jul 14 11:22:42 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler#register):
+	  notify the handler thread of new timeout registration.
+
+	* lib/webrick/utils.rb (WEBrick::Utils::TimeoutHandler#initialize):
+	  make sleep intervals adaptive than fixed period intervals.
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#start): flush
+	  shutdown pipe.
+
+	* lib/webrick/server.rb (WEBrick::GenericServer#stop): request the
+	  server to stop immediately by sending data via shutdown pipe.
+
+Mon Jul 13 23:58:08 2015  Stefano Tortarolo  <stefano.tortarolo@gmail.com>
+
+	* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#do_CONNECT):
+	  fix typos in debugger statements.  [Fix GH-967]
+
+Mon Jul 13 19:11:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (timeout): warn as deprecated for a long time.
+
+Mon Jul 13 01:37:27 2015  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/openssl/ossl.c: [DOC] Backport ruby/openssl@dbb3fdb [Bug #11345]
+	  Thanks to Tomoya Chiba for the report and help with patch.
+
+Sun Jul 12 09:20:02 2015  Shota Fukumori  <her@sorah.jp>
+
+	* ext/socket/basicsocket.c: [DOC] typo (Errno::AGAIN -> Errno::EAGAIN)
+
+	* ext/socket/socket.c: ditto
+
+	* ext/socket/tcpserver.c: ditto
+
+	* ext/socket/udpsocket.c: ditto
+
+	* ext/socket/unixserver.c: ditto
+
+	* io.c: ditto
+
+Sun Jul 12 06:42:23 2015  ksss  <co000ri@gmail.com>
+
+	* test/stringio/test_stringio.rb (test_sysread): add a test for
+	  StringIO#sysread.  [Fix GH-966]
+
+Sat Jul 11 21:16:34 2015  ksss  <co000ri@gmail.com>
+
+	* ext/stringio/stringio.c (Init_stringio): [DOC] Fix an example,
+	  StringIO#puts should be set "\n" at last.  [Fix GH-965]
+
+Sat Jul 11 12:45:51 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout#timeout): remove regexp with wrong line
+	  number and fix caller depth.
+
+Fri Jul 10 22:05:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (ExitException): removed internal exception class
+	  and use Timeout::Error instead, as using throw/catch to isolate
+	  each timeouts now.  [ruby-dev:49179] [Bug #11344]
+
+Fri Jul 10 17:41:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (is_case_sensitive): get attributes by the file descriptor
+	  of open directory, instead of using mount point name.
+
+Fri Jul 10 10:46:02 2015  ksss  <co000ri@gmail.com>
+
+	* ext/stringio/stringio.c (writable): remove unnecessary check for
+	  deprecated safe level 4.  [Fix GH-963]
+
+Thu Jul  9 15:07:12 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (waitpid): return immediately if interrupted.
+	  reported by <takkanm AT gmail.com> [ruby-dev:49176] [Bug #11340]
+
+Thu Jul  9 13:03:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_search_super_method): use CI_SET_FASTPATH().
+
+Thu Jul  9 11:07:06 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove rb_call_info_t::klass because
+	  rb_callable_method_entry_t has information about defined class.
+
+	* vm_insnhelper.c (vm_search_method): don't set ci->klass because
+	  it is removed.
+
+	* vm_insnhelper.c (rb_equal_opt): ditto.
+
+	* vm_insnhelper.c (vm_search_superclass): removed because it is too
+	  simple to write code directly.
+
+	* vm_insnhelper.c (vm_defined): don't use vm_search_superclass().
+	  This fix avoid searching current callable `me' twice.
+
+	* vm_insnhelper.c (vm_search_super_method): ditto.
+
+Thu Jul  9 10:03:10 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/http/responses.rb: Added 308 status to CODE_TO_OBJ list.
+	  [fix GH-961] Patch by @billinghamj
+
+Thu Jul  9 09:34:14 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (rb_control_frame_t): fix comments (layout index).
+
+Thu Jul  9 09:25:50 2015  Zachary Scott  <e@zzak.io>
+
+	* parse.y: Improve duplicate key warning with patch by @andremedeiros
+	  [Fix GH-938] https://github.com/ruby/ruby/pull/938 [Bug #11327]
+
+Wed Jul  8 07:43:01 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/csv.rb: typo fix [ci skip][fix GH-958] Patch by @henrik
+
+Wed Jul  8 04:42:27 2015  Eric Wong  <e@80x24.org>
+
+	* iseq.c (iseq_data_to_ary): dump kw_arg as symbol
+	* test/-ext-/iseq_load/test_iseq_load.rb: test kw_arg roundtrip
+	  [ruby-core:69891] [Bug #11338]
+
+Tue Jul  7 18:18:41 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* random.c (fill_random_bytes_syscall): fix compile error.
+
+Tue Jul  7 16:47:30 2015  Eric Wong  <e@80x24.org>
+
+	* compile.c (COMPILE_ERROR): reduce GET_THREAD() calls
+
+Tue Jul  7 16:39:04 2015  Eric Wong  <e@80x24.org>
+
+	* random.c (fill_random_bytes_syscall): return -1 for error
+	* random.c (fill_random_bytes): try urandom on syscall failure
+
+Tue Jul  7 15:02:18 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_str_normalize_ospath): skip invalid byte sequence not
+	  to loop infinitely.  this case usually does not happen as the
+	  input name should come from real file systems.
+
+Tue Jul  7 14:40:08 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c: remove debug flag introduced accidentally.
+
+Tue Jul  7 12:05:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (cont_free): remove mysterious fflush()
+	  introduced at r19890, maybe accidentally.
+
+Tue Jul  7 11:45:14 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_method_call): because data->me should be non-NULL,
+	  do not check data->me
+
+	* proc.c (method_inspect): ditto.
+
+Tue Jul  7 11:37:25 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove rb_iseq_t::orig because rb_iseq_clone()
+	  no longer exists.
+
+	* iseq.c: don't use rb_iseq_t::orig.
+
+Tue Jul 07 11:25:57 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c, internal.h (rb_iseq_clone): removed because we don't need to
+	  clone iseq any more.
+
+	* class.c (clone_method): share iseq between cloned methods. All of
+	  method dependent information are able to refer from method entry.
+
+Tue Jul  7 04:42:25 2015  Eric Wong  <e@80x24.org>
+
+	* string.c (Init_String): use rb_str_freeze for String#freeze
+	  to resize internal buffer
+	  [ruby-core:69870] [Feature #11330]
+
+Tue Jul  7 04:12:32 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_define_method): remove an unused local variable.
+
+Tue Jul  7 03:57:28 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove rb_iseq_t::defined_method_id because it is not
+	  needed.
+
+	* eval.c (frame_func_id): simplify. rb_callable_method_entry_t
+	  has enough information.
+
+	* eval.c (frame_called_id): ditto.
+
+	* iseq.c (prepare_iseq_build): catch up this fix.
+
+	* proc.c (rb_mod_define_method): ditto.
+
+	* vm.c (vm_define_method): ditto.
+
+Tue Jul  7 03:47:26 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove a useless declaration.
+
+Tue Jul  7 03:33:20 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: remove rb_iseq_t::klass to reduce dynamic data.
+
+	* internal.h, iseq.c (rb_iseq_klass): remove it because
+	  rb_iseq_t::klass is removed.
+
+	* vm_insnhelper.c (vm_super_outside): do not see cfp->iseq, but
+	  check callable method entry on a frame.
+	  This fix simplify the logic to search super class.
+
+	* test/ruby/test_method.rb: support super() from Proc.
+	  Now, [Bug #4881] and [Bug #3136] was solved.
+
+	* proc.c (rb_mod_define_method): catch up this change.
+
+	* vm.c (vm_define_method): ditto.
+
+	* vm_backtrace.c (rb_profile_frames): now, each `frame' objects
+	  are rb_callable_method_entry_t data or iseq VALUEs.
+
+	  This fix introduce minor compatibility issue that
+	  rb_profile_frame_label() always returns
+	  rb_profile_frame_base_label().
+
+	* test/-ext-/debug/test_profile_frames.rb: catch up this change.
+
+Tue Jul  7 01:52:14 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* cont.c (fiber_init): initialize control frame correctly.
+	  This fix does not affect any ordinal execution, but
+	  affects debug prints.
+
+Mon Jul  6 17:59:05 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_search_super_method): do not skip calling
+	  same methods in super.
+	  [Bug #3351]
+
+	* test/ruby/test_super.rb: fix a test.
+
+Mon Jul  6 17:59:11 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/tk/tcltklib.c: removed deprecated safe level.
+
+Mon Jul  6 17:16:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h, proc.c (rb_method_entry_location): make it static
+	  and remove prefix `rb_' because it is used only in proc.c.
+
+Mon Jul  6 16:42:10 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/lib/memory_status.rb: removed redundant path.
+
+Mon Jul  6 01:18:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit/parallel.rb: make @@project_dir one level
+	  upper as this file had moved one level deeper.
+
+Sun Jul  5 23:54:10 2015  mizokami  <suzunatsu@yahoo.com>
+
+	* lib/optparse.rb: [DOC] Fix typo.
+
+Sun Jul  5 18:25:37 2015  Eric Wong  <e@80x24.org>
+
+	* gc.c (gc_profile_record_get): fix spelling error in keys
+
+Sun Jul  5 14:49:01 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* README.md: fix a typo pointed out by raoulvdberge.
+	  https://github.com/ruby/ruby/pull/953#commitcomment-11998186
+
+Sun Jul  5 12:56:20 2015  Irvi Firqotul Aini  <viarc7@gmail.com>
+
+	* README.md: Added link HowToReport bugs.
+
+Sun Jul  5 10:51:48 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang): the original array may not be
+	  embedded even if a substitution array is embedded, as it is
+	  embedded when the original array is short enough but not
+	  embedded.  [ruby-dev:49166] [Bug #11332]
+
+Sun Jul  5 09:31:40 2015  Eric Wong  <e@80x24.org>
+
+	* test/ruby/test_process.rb: test for fd=3 usability in child
+
+Sat Jul  4 19:43:31 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* Add test case for empty array and first method with args.
+	  Patch by @yui-knk [fix GH-955]
+
+Sat Jul  4 19:39:08 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* Add test for `Enumerable#sort` with block. Patch by @yui-knk
+	  [fix GH-954]
+
+Sat Jul  4 14:38:43 2015  Eric Wong  <e@80x24.org>
+
+	* enum.c (zip_ary): remove volatile, use RB_GC_GUARD
+	  (zip_i): ditto
+
+Sat Jul  4 10:42:57 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/test_case.rb (teardown): do not delete features
+	  loaded from the original load paths, the same libraries should
+	  be loaded again when the same features are required.
+	  [ruby-dev:49031] [Bug #11222]
+
+Sat Jul  4 09:38:52 2015  Eric Wong  <e@80x24.org>
+
+	* vm.c (rb_vm_mark): reduce branches for always-set VM fields
+	  (rb_vm_add_root_module): ditto
+
+Fri Jul 03 20:05:10 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: introduce rb_callable_method_entry_t to remove
+	  rb_control_frame_t::klass.
+	  [Bug #11278], [Bug #11279]
+
+	  rb_method_entry_t data belong to modules/classes.
+	  rb_method_entry_t::owner points defined module or class.
+
+	    module M
+	      def foo; end
+	    end
+
+	  In this case, owner is M.
+
+	  rb_callable_method_entry_t data belong to only classes.
+	  For modules, MRI creates corresponding T_ICLASS internally.
+	  rb_callable_method_entry_t can also belong to T_ICLASS.
+
+	  rb_callable_method_entry_t::defined_class points T_CLASS or
+	  T_ICLASS.
+	  rb_method_entry_t data for classes (not for modules) are also
+	  rb_callable_method_entry_t data because it is completely same data.
+	  In this case, rb_method_entry_t::owner == rb_method_entry_t::defined_class.
+
+	  For example, there are classes C and D, and includes M,
+
+	    class C; include M; end
+	    class D; include M; end
+
+	  then, two T_ICLASS objects for C's super class and D's super class
+	  will be created.
+
+	  When C.new.foo is called, then M#foo is searched and
+	  rb_callable_method_t data is used by VM to invoke M#foo.
+
+	  rb_method_entry_t data is only one for M#foo.
+	  However, rb_callable_method_entry_t data are two (and can be more).
+	  It is proportional to the number of including (and prepending)
+	  classes (the number of T_ICLASS which point to the module).
+
+	  Now, created rb_callable_method_entry_t are collected when
+	  the original module M was modified. We can think it is a cache.
+
+	  We need to select what kind of method entry data is needed.
+	  To operate definition, then you need to use rb_method_entry_t.
+
+	  You can access them by the following functions.
+
+	  * rb_method_entry(VALUE klass, ID id);
+	  * rb_method_entry_with_refinements(VALUE klass, ID id);
+	  * rb_method_entry_without_refinements(VALUE klass, ID id);
+	  * rb_resolve_refined_method(VALUE refinements, const rb_method_entry_t *me);
+
+	  To invoke methods, then you need to use rb_callable_method_entry_t
+	  which you can get by the following APIs corresponding to the
+	  above listed functions.
+
+	  * rb_callable_method_entry(VALUE klass, ID id);
+	  * rb_callable_method_entry_with_refinements(VALUE klass, ID id);
+	  * rb_callable_method_entry_without_refinements(VALUE klass, ID id);
+	  * rb_resolve_refined_method_callable(VALUE refinements, const rb_callable_method_entry_t *me);
+
+	  VM pushes rb_callable_method_entry_t, so that rb_vm_frame_method_entry()
+	  returns rb_callable_method_entry_t.
+	  You can check a super class of current method by
+	  rb_callable_method_entry_t::defined_class.
+
+	* method.h: renamed from rb_method_entry_t::klass to
+	  rb_method_entry_t::owner.
+
+	* internal.h: add rb_classext_struct::callable_m_tbl to cache
+	  rb_callable_method_entry_t data.
+
+	  We need to consider about this field again because it is only
+	  active for T_ICLASS.
+
+	* class.c (method_entry_i): ditto.
+
+	* class.c (rb_define_attr): rb_method_entry() does not takes
+	  defined_class_ptr.
+
+	* gc.c (mark_method_entry): mark RCLASS_CALLABLE_M_TBL() for T_ICLASS.
+
+	* cont.c (fiber_init): rb_control_frame_t::klass is removed.
+
+	* proc.c: fix `struct METHOD' data structure because
+	  rb_callable_method_t has all information.
+
+	* vm_core.h: remove several fields.
+	  * rb_control_frame_t::klass.
+	  * rb_block_t::klass.
+
+	  And catch up changes.
+
+	* eval.c: catch up changes.
+
+	* gc.c: ditto.
+
+	* insns.def: ditto.
+
+	* vm.c: ditto.
+
+	* vm_args.c: ditto.
+
+	* vm_backtrace.c: ditto.
+
+	* vm_dump.c: ditto.
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_method.c: ditto.
+
+Fri Jul  3 14:30:18 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c: some mingw compilers need a tweek for the
+	  declarations of _wfreopen_s.  [Bug #11320]
+
+Fri Jul  3 12:25:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (rb_econv_set_replacement): target encoding name can
+	  be empty now.  [ruby-core:69841] [Bug #11324]
+
+Fri Jul  3 07:21:06 2015  Eric Wong  <e@80x24.org>
+
+	* benchmark/bm_io_nonblock_noex.rb: new benchmark
+	* ext/openssl/ossl_ssl.c (no_exception_p): new function
+	  (ossl_start_ssl): adjust for no_exception_p
+	  (ossl_ssl_connect): adjust ossl_start_ssl call
+	  (ossl_ssl_connect_nonblock): ditto
+	  (ossl_ssl_accept): ditto
+	  (ossl_ssl_accept_nonblock): ditto
+	  (ossl_ssl_read_internal): adjust for no_exception_p
+	  (ossl_ssl_write_internal): ditto
+	  (ossl_ssl_write): adjust ossl_write_internal call
+	  (ossl_ssl_write_nonblock): ditto
+	* ext/stringio/stringio.c (strio_read_nonblock):
+	  delay exception check
+	* io.c (no_exception_p): new function
+	  (io_getpartial): call no_exception_p
+	  (io_readpartial): adjust for io_getpartial
+	  (get_kwargs_exception): remove
+	  (io_read_nonblock): adjust for io_getpartial,
+	  check no_exception_p on EOF
+	  (io_write_nonblock): call no_exception_p
+	  (rb_io_write_nonblock): do not check `exception: false'
+	  (argf_getpartial): adjust for io_getpartial
+	  [ruby-core:69778] [Feature #11318]
+
+Fri Jul  3 07:13:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (replace_real_basename): Win32 API does not set errno, get
+	  the last error by GetLastError() and map to errno.  [Bug #10015]
+
+Thu Jul  2 21:32:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (replace_real_basename): show warnings at errors.
+	  [Bug #10015]
+
+Thu Jul  2 18:39:20 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove `#define RGENGC_OBJ_INFO 1' line introduced to
+	  debug Bug #11244.
+
+Thu Jul  2 18:34:26 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_raw_obj_info): separated from rb_obj_info().
+	  Fill internal object information into passed buffer.
+
+	* gc.h: declare rb_raw_obj_info().
+
+Thu Jul  2 16:15:04 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (replace_real_basename): update path type by the target
+	  attributes if possible, to improve the performance.  [Bug #10015]
+
+Thu Jul  2 14:45:53 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* st.c: get rid of VC++'s warnings of C4700 (uninitialized local
+	  variable used).  I think that these are wrong, but should shut them
+	  up.
+
+Thu Jul  2 14:15:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems.rb (Gem.load_path_insert_index): search
+	  @gem_prelude_index first.
+
+	* lib/rubygems/test_case.rb (Gem::TestCase#setup): keep already
+	  expanded paths to preserve instance variables.
+
+Thu Jul  2 14:12:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): also copy initial load path marks at
+	  setting load paths encoding.
+
+Thu Jul  2 12:26:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_server.rb (process_based_port): use
+	  dynamically chosen port numbers to get rid of conflicts.
+
+Thu Jul  2 11:58:59 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/rubygems/test_gem_specification.rb: skip tests which the
+	  platform does not permit the filename of its test file.
+
+Thu Jul  2 11:36:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rubygems/test_gem_resolver_git_specification.rb: require
+	  rubygems/installer.rb before Gem::TestCase#setup runs, otherwise
+	  as Gem::TestCase#teardown restores $LOADED_FEATURES to the state
+	  at that time, the requiring the file in GitSpecification#install
+	  method causes a lot of constant redefinitions.
+
+Thu Jul  2 10:43:36 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/rubysocket.h: flags for common socket families
+	  (rsock_getfamily): update signature
+	* include/ruby/io.h: comment socket FMODE flags
+	* ext/socket/init.c (rsock_getfamily): memoize family
+	* ext/socket/basicsocket.c: adjust rsock_getfamily calls
+	* ext/socket/ancdata.c: ditto
+	  [ruby-core:69713] [Feature #11298]
+
+Thu Jul  2 10:30:01 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems/resolver.rb: fixed NameError of Gem::Util::NULL_DEVICE.
+
+Thu Jul  2 09:51:44 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems/resolver.rb: fix error of null device reference with DOSISH
+	  platform.
+
+Thu Jul  2 06:49:44 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems: Update to RubyGems HEAD(c202db2).
+	  this version contains many enhancements see http://git.io/vtNwF
+	* test/rubygems: ditto.
+
+Wed Jul  1 23:50:34 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/net/http/test_httpresponse.rb
+	(HTTPResponseTest#test_read_body_content_encoding_deflate_uppercase):
+	fix a failure without zlib.
+
+Wed Jul  1 10:54:56 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* Add test for Enumerable#none? [fix GH-950] Patch by @yui-knk
+
+Wed Jul  1 09:30:36 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (struct_set_members): hide internal back_members
+	  object, and members object does not need to be duped as it
+	  should be frozen and hidden.
+
+Wed Jul  1 09:28:47 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (struct_member_pos): revert r51080 to fix other
+	  implicit conversions but cast the return value to fix the
+	  previous implicit conversion.
+
+Wed Jul  1 08:47:24 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* struct.c (struct_member_pos): avoid implicit conversion loses
+	  integer precision: 'long' to 'int'.
+
+Wed Jul  1 05:57:03 2015  Eric Wong  <e@80x24.org>
+
+	* vm_method.c (rb_add_method_iseq): add RB_GC_GUARD
+	* class.c (clone_method): remove RB_GC_GUARD
+	* struct.c (define_aref_method): ditto
+	  (define_aset_method): ditto
+	* vm.c (vm_define_method):
+	* iseq.c (rb_iseq_clone): add RB_GC_GUARD
+
+Wed Jul  1 05:43:58 2015  Eric Wong  <e@80x24.org>
+
+	* struct.c (AREF_HASH_THRESHOLD): new macro
+	  (id_back_members): new ID
+	  (struct_member_pos_ideal): new function
+	  (struct_member_pos_probe): ditto
+	  (struct_set_members): ditto
+	  (struct_member_pos): ditto
+	  (rb_struct_getmember): use struct_member_pos for O(1) access
+	  (rb_struct_aref_sym): ditto
+	  (rb_struct_aset_sym): ditto
+	  (setup_struct): call struct_set_members
+	  (struct_define_without_accessor): ditto
+	  (Init_Struct): initialize __members_back__
+	  [ruby-core:66851] [ruby-core:69705] [ruby-core:69821]
+
+Tue Jun 30 23:12:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_reopen): FilePathValue() ensures the path
+	  NUL-terminated and frozen, so it is unnecessary to make it shared.
+
+Tue Jun 30 23:11:53 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (check_dirname): ensure path name NUL-terminated for
+	  SHARABLE_MIDDLE_SUBSTRING.
+
+	* io.c (rb_sysopen): ditto.
+
+Tue Jun 30 18:38:16 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/file.c (rb_freopen): need to terminate by NUL.
+
+Tue Jun 30 17:28:25 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_reopen): freopen(3) with OS encoding path.
+	  [ruby-core:69780] [Bug #11320]
+
+	* win32/file.c (rb_freopen): wrapper of wchar version freopen(3).
+	  use _wfreopen_s() if available.
+
+Tue Jun 30 08:24:08 2015  Eric Wong  <e@80x24.org>
+
+	* io.c (rb_io_oflags_modestr): handle O_TRUNC correctly
+	* test/ruby/test_io.rb (test_reopen_stdio): new test
+	  Patch-by: cremno phobia <cremno@mail.ru>
+	  [ruby-core:69779] [Bug #11319]
+
+Tue Jun 30 02:47:02 2015  Eric Wong  <e@80x24.org>
+
+	* include/ruby/st.h (struct st_table): hide struct list_head
+	* st.c (struct st_table_entry): adjust struct
+	  (head, tail): remove shortcut macros
+	  (st_head): new wrapper function
+	  (st_init_table_with_size): adjust to new struct and API
+	  (st_clear): ditto
+	  (add_direct): ditto
+	  (unpack_entries): ditto
+	  (rehash): ditto
+	  (st_copy): ditto
+	  (remove_entry): ditto
+	  (st_shift): ditto
+	  (st_foreach_check): ditto
+	  (st_foreach): ditto
+	  (get_keys): ditto
+	  (get_values): ditto
+	  (st_values_check): ditto
+	  (st_reverse_foreach_check): ditto (unused)
+	  (st_reverse_foreach): ditto (unused)
+	  [ruby-core:69726] [Misc #10278]
+
+Mon Jun 29 17:38:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defineclass): do not quote unprintable characters at
+	  raising an exception.
+
+Mon Jun 29 16:01:24 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http/response.rb (inflater): CONTENT_ENCODING can be upper
+	  case. [ruby-core:69670] [Bug #11285] patched by Andy Chu
+
+Mon Jun 29 14:50:08 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (add_activated_refinement): should not include the original
+	  class.
+
+Mon Jun 29 12:09:10 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* README.md: tweak styles. [fix GH-945][ci skip] Patch by @bryndyment
+
+Mon Jun 29 07:23:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/sizes.c.tmpl: extract RUBY_DEFINT to define sizes of
+	  types checked by configure.in, and fix size of intptr_t in
+	  universal binary.
+
+Mon Jun 29 02:10:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (defineclass): preserve encoding of name in error
+	  messages for super class mismatch.
+
+	* insns.def (defineclass): preserve encoding of name in error
+	  messages for non-class super.
+
+	* insns.def (defineclass): preserve encoding of name in error
+	  messages when already defined but type mismatch.
+
+Sun Jun 28 12:07:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_define_class_id_under): raise TypeError exception
+	  same as ruby level class definition when superclass mismatch.
+
+Sun Jun 14 19:02:03 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/net/ftp.rb (makeport): close the TCPServer
+	  when sending the port fails.
+
+	* test/net/ftp/test_ftp.rb: test for above.
+
+Fri Jun 26 12:48:37 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (load_transcoder_entry): fix transcoder loading race
+	  condition, by waiting in require.  [ruby-dev:49106] [Bug #11277]
+
+Fri Jun 26 07:53:56 2015  Eric Wong  <e@80x24.org>
+
+	* enum.c (enum_minmax): simplify return value creation
+	* test/ruby/test_enum.rb: test behavior on empty
+	* hash.c (rb_hash_fetch_m): remove unnecessary volatile since r41597
+	  (env_reject_bang): trade volatile for GC guard
+	  (env_select): ditto
+	  (env_select_bang): ditto
+	  (env_keep_if): ditto
+	  (rb_env_clear): ditto
+
+Thu Jun 25 21:24:28 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/-ext-/popen_deadlock/test_popen_deadlock.rb: test [Bug #11265]
+
+	* ext/-test-/popen_deadlock/infinite_loop_dlsym.c: new ext to call
+	  dlsym(3) infinitely without GVL, used in the above test.
+
+	* ext/-test-/popen_deadlock/extconf.rb: extconf.rb for the above
+	  ext. Currently, only enabled on Solaris (main target) and Linux
+	  (as a reference platform and for debugging the ext).
+
+Thu Jun 25 19:24:25 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: not to use vfork on Solaris to avoid deadlock
+	  occurred in vfork(2) with multi-threading and dynamic linker
+	  on Solaris. [Bug #11265] [ruby-dev:49089]
+
+Thu Jun 25 18:25:41 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/lib/envutil.rb (Test::Unit::Assertions#assert_no_memory_leak):
+	  NO_MEMORY_LEAK_ENVS is moved to Memory::NO_MEMORY_LEAK_ENVS
+	  to reduce child executions during test-all on Solaris.
+
+	* test/lib/memory_status.rb (Memory::NO_MEMORY_LEAK_ENVS): ditto.
+
+Thu Jun 25 17:32:33 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_method_entry_create): need to call
+	  method_definition_reset() if def is given.
+
+	  Actually, `me' is a new object, so we don't need to call it.
+	  It is just to make sure.
+
+	* vm_method.c (method_definition_reset): remove duplicated insertion.
+
+	* vm_method.c (rb_method_entry_clone): assign dst->def here,
+	  not in method_definition_reset().
+
+Thu Jun 25 16:44:54 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c: make a rb_method_definition_t data (def) *after* making
+	  a rb_method_entry_t data (me).
+
+	  Normally, `me' points `def'. Some Ruby objects pointed from `def'
+	  and objects are marked by `me' (mark_method_entry() in gc.c).
+	  However, `def' is built before making a `me', then nobody can mark
+	  objects pointed from `def' before making (and pointing from) `me'.
+
+	  I hope this patch solve #11244.
+
+	* vm_method.c: remove `rb_' prefix from some static functions.
+
+	* method.h (rb_method_entry_create): constify
+
+	* gc.c (mark_method_entry): add checking `def' and
+	  `def->body.iseq.iseqptr' availability because they can be NULL.
+
+Thu Jun 25 14:14:16 2015  takiy33  <takiy33@gmail.com>
+
+	* test/test_prime.rb (test_eratosthenes_works_fine_after_timeout):
+	  use spaces instead of TABs in ruby codes.  [Fix GH-944]
+
+Thu Jun 25 07:08:35 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info, method_type_name): show method type name in a string
+	  instead of a number.
+
+Thu Jun 25 06:49:25 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info): show more details for T_IMEMO/imemo_ment.
+
+Thu Jun 25 06:40:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_method_definition_reset): need a WB for
+	  VM_METHOD_TYPE_ATTRSET.
+
+Thu Jun 25 03:33:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RGENGC_OBJ_INFO, obj_info): add a macro to enable/disable
+	  rich obj_info() output.
+
+	  At the default, the value of RGENGC_OBJ_INFO is
+	  (RGENGC_DEBUG | RGENGC_CHECK_MODE).
+
+	* gc.c (RGENGC_OBJ_INFO): force enable it to debug #11244.
+
+	* gc.c (gc_mark_ptr): print more details with obj_info().
+
+	* gc.c (gc_mark_children): remove useless debug prints.
+
+Thu Jun 25 02:40:33 2015  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_read_internal):
+	  do not process kwargs in blocking mode
+	* test/openssl/test_ssl.rb: test sysread
+
+Wed Jun 24 16:54:11 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_children): add additional debug code for #11244.
+
+Wed Jun 24 16:05:42 2015  Eric Wong  <e@80x24.org>
+
+	* string.c (rb_str_justify): use RB_GC_GUARD
+
+Wed Jun 24 14:25:17 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_ptr): add a check code for #11244.
+
+	  It should be removed later. But we can remain this check
+	  because it is only a branch.
+
+Wed Jun 24 12:49:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_fstring_cstr): new function to make a fstring from
+	  a string literal.
+
+	* internal.h (rb_fstring_lit): new macro to make a fstring from a
+	  string literal.
+
+	* include/ruby/intern.h (rb_strlen_lit): new macro to get the
+	  length of a string literal, borrowed from mruby/mruby@e4afd53.
+
+Wed Jun 24 12:21:16 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* re.c: Update documentation for Regexp class.
+	  [fix GH-937][ci skip] Patch by @davydovanton
+
+Wed Jun 24 09:23:03 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (generic_ivar_set): remove FL_ABLE check
+	  (gen_ivar_copy): ditto
+	  [ruby-core:69715]
+
+Wed Jun 24 08:28:15 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): reduce stack use
+	  [ruby-core:69595] [Feature #11263]
+
+Tue Jun 23 14:32:42 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (name_err_receiver): raise ArgumentError if no receiver
+	  is available on this exception object.  [Feature #10881]
+
+Tue Jun 23 09:48:34 2015  Eric Wong  <e@80x24.org>
+
+	* dir.c (check_dirname): avoid volatile, use return value
+	  (dir_s_chroot, dir_s_mkdir, dir_s_rmdir): adjust callers
+
+Tue Jun 23 06:37:10 2015  Eric Wong  <e@80x24.org>
+
+	* struct.c (struct_ivar_get): cache member definition in a subclass
+	  Thanks to Sokolov Yura aka funny_falcon <funny.falcon@gmail.com>
+	  in https://bugs.ruby-lang.org/issues/10585
+
+Tue Jun 23 04:58:06 2015  Eric Wong  <e@80x24.org>
+
+	* benchmark/bm_vm2_struct_big_href_hi.rb: new benchmark
+	* benchmark/bm_vm2_struct_big_href_lo.rb: ditto
+	* benchmark/bm_vm2_struct_big_hset.rb: ditto
+	* benchmark/bm_vm2_struct_small_href.rb: ditto
+	* benchmark/bm_vm2_struct_small_hset.rb: ditto
+	  Thanks to Sokolov Yura aka funny_falcon <funny.falcon@gmail.com>
+	  in https://bugs.ruby-lang.org/issues/10585
+
+Mon Jun 22 18:08:48 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/lib/test/unit.rb (Test::Unit::Parallel#start_watchdog): removed
+	  because it has been meaningless since r36385. [Bug #11288]
+
+	* test/lib/test/unit.rb (Test::Unit::Parallel#_run_parallel): delete
+	  lines related to the removed start_watchdog method
+
+Sun Jun 21 23:52:46 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: mention about Array#bsearch_index and Hash#fetch_values.
+
+Sun Jun 21 23:46:27 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: add a reference to a ticket.
+
+Sun Jun 21 20:28:09 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (roomof): extract from type_roomof, and move from
+	  bignum.c.
+
+Sun Jun 21 18:32:37 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby_atomic.h (ATOMIC_PTR_CAS): define by generic CAS macro, not
+	  via size_t, to suppress a warning by mingw gcc.
+
+Sun Jun 21 05:31:41 2015  Shota Fukumori  <her@sorah.jp>
+
+	* ext/objspace/objspace_dump.c(dump_object): Return empty JSON object when
+	  passed object is a special const, instead of SEGV.
+	  Based patch by Kohei Suzuki (eagletmt). [ruby-core:69692] [Bug #11291]
+
+	* test/objspace/test_objspace.rb(test_dump_special_consts): Test for above fix.
+
+
+Sat Jun 20 03:56:58 2015  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* enc/make_encmake.rb: the list of encoding extension libraries must
+	  not include encinit.c itself.  It caused "undefined reference to
+	  Init_encinit".
+
+Sat Jun 20 02:03:53 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* process.c (rb_execarg_parent_start1): new macro ALWAYS_NEED_ENVP
+	  to generate envp_str anytime on Solaris 10 (or earlier version
+	  of Solaris) to avoid calling execv() which is async-signal unsafe
+	  on Solaris 10. [Bug #11265] [ruby-dev:49089]
+
+	* process.c (exec_with_sh, proc_exec_cmd): On Solaris 10,
+	  because ALWAYS_NEED_ENVP is 1 and envp_str is always generated,
+	  execv() in exec_with_sh() and proc_exec_cmd() are never called.
+	  To guarantee this, execv() is replaced by a macro to print
+	  out error message on Solaris 10.
+
+	* process.c (proc_exec_sh): Because proc_exec_sh() may be called
+	  by rb_proc_exec() with envp_str = Qfalse, execl() is replaced
+	  by a macro that calls execle() with "extern char **environ"
+	  traditional global variable on Solaris 10.
+	  TODO: This may be unsafe and should be changed in the future.
+	  Although rb_proc_exec() is not used from inside current version
+	  of ruby, it may be called by third-party extensions.
+
+Sat Jun 20 01:10:13 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: mention about $SAFE.
+
+Fri Jun 19 14:53:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_mod_define_method): now requires a block direct to
+	  this method call.  [ruby-core:69655] [Bug #11283]
+
+Fri Jun 19 13:54:43 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (rb_mod_define_method): get rid of inadvertent ID
+	  creations at error.
+
+Fri Jun 19 07:58:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: if no with-ext option is given, default to
+	  enable everything.  [ruby-dev:49108] [Bug #11280]
+
+Fri Jun 19 06:30:07 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* bootstraptest/test_method.rb: remove a test because $SAFE=2 was
+	  obsolete.
+
+	  Please check btest, too.
+
+Thu Jun 18 23:51:51 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* bin/erb: $SAFE=3 is obsolete.
+
+Thu Jun 18 23:45:11 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* safe.c: removed needless doc related $SAFE=2
+
+Thu Jun 18 23:38:07 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (rb_thread_safe_level): fix document.  $SAFE=3 is obsolete.
+	  [ci skip]
+
+Thu Jun 18 23:25:51 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: configure intersection of with-ext and not
+	  without-ext, as withouts is no longer true by default if
+	  with-ext option is given.  [ruby-dev:49108] [Bug #11280]
+
+Thu Jun 18 23:20:46 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* include/ruby/ruby.h: $SAFE=2 is now obsolete.
+
+	* dir.c, ext/fiddle/handle.c, ext/socket/basicsocket.c, file.c
+	  gc.c, io.c, process.c, safe.c, signal.c, win32/file.c:
+	  removed code for $SAFE=2
+
+	* test/erb/test_erb.rb, test/fiddle/test_handle.rb
+	  test/ruby/test_env.rb: removed tests for $SAFE=2.
+
+Thu Jun 18 22:50:07 2015  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* enc/make_encmake.rb: added --transes and --no-transes options.
+
+Thu Jun 18 18:24:12 2015  Eric Wong  <e@80x24.org>
+
+	* test/socket/test_unix.rb: replace sleep with select
+
+Thu Jun 18 17:59:06 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (rb_vm_control_frame_id_and_class): remove useless codes.
+	  `me' knows ID and owner class.
+
+Thu Jun 18 16:58:35 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: constify rb_method_alias_struct::original_me and
+	  rb_method_refined_struct::orig_me.
+
+	* class.c (move_refined_method): use RB_OBJ_WRITE() for
+	  me->def->body.refined.orig_me.
+
+Thu Jun 18 14:35:28 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (count_imemo_objects): support imemo_ment.
+
+Thu Jun 18 13:32:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (name_err_receiver): add NameError#receiver method.
+	  [Feature #10881]
+
+Thu Jun 18 10:00:06 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* safe.c: removed needless doc related $SAFE=3
+
+Thu Jun 18 09:59:23 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* safe.c: rename old method name for $SAFE=3
+
+Thu Jun 18 06:02:42 2015  Eric Wong  <e@80x24.org>
+
+	* compile.c (get_exception_sym2type, iseq_build_from_ary_body):
+	  rely on %+PRIsVALUE instead of calling rb_inspect directly
+
+Wed Jun 17 20:59:25 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ENC_MK): needs fake.rb if cross compilation.
+	  [ruby-dev:49098] [Bug #11272]
+
+Wed Jun 17 20:23:29 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/rbconfig/sizeof/extconf.rb: Check __float80.
+
+Wed Jun 17 15:15:53 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* safe.c (safe_setter): of course, don't have to warn the limitation of
+	  $SAFE=3 after it's removed.
+
+Wed Jun 17 14:29:43 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* include/ruby/ruby.h: $SAFE=3 is now obsolete.
+
+	* ext/socket/init.c, ext/socket/socket.c, ext/socket/tcpsocket.c
+	  ext/socket/udpsocket.c, gc.c, object.c, re.c, safe.c: removed code
+	  for $SAFE=3
+
+	* bootstraptest/test_method.rb, test/erb/test_erb.rb, test/ruby/test_dir.rb
+	  test/ruby/test_file.rb, test/ruby/test_method.rb, test/ruby/test_regexp.rb
+	  test/ruby/test_thread.rb: remove tests for $SAFE=3
+
+Wed Jun 17 12:13:33 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: added contributor name.
+
+Wed Jun 17 10:57:28 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/csv.rb: accept to use Range object for row selection.
+	  contributed by Mitsutaka Mimura.
+	  [Feature #11267][ruby-dev:49091]
+
+Wed Jun 17 09:50:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/servlet.rb (documentation_search, root_search):
+	  requires json for JSON.dump and fix sporadic failures due to
+	  the loading order.
+
+Tue Jun 16 19:19:53 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/rbconfig/sizeof/extconf.rb: Check several types defined in C99
+	  and x86_64 ABI.
+
+	* template/sizes.c.tmpl: Relax a pattern for types.
+
+Tue Jun 16 17:37:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/objspace/test_objspace.rb: relax pattern because uncollectible
+	  flag and marked flag can be false at major GC.
+	  [Bug #10852]
+
+Tue Jun 16 04:50:44 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/basicsocket.c (bsock_recv): document outbuf
+	* ext/socket/unixsocket.c (unix_recvfrom): ditto
+	* ext/socket/init.c (rsock_strbuf, recvfrom_locktmp): new functions
+	  (rsock_s_recvfrom): support destination buffer as 3rd arg
+	  (rsock_s_recvfrom_nonblock): ditto
+	* string.c (rb_str_locktmp_ensure): export for internal ext
+	* test/socket/test_nonblock.rb: test recv_nonblock
+	* test/socket/test_unix.rb: test recv
+	  [ruby-core:69543] [Feature #11242]
+
+Tue Jun 16 04:38:02 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal,
+	  bsock_recvmsg_internal):
+	  support "exception: false" kwarg
+	* ext/socket/init.c (rsock_s_recvfrom_nonblock):
+	  ditto
+	* ext/socket/init.c (rsock_s_recvfrom_nonblock): use rsock_opt_false_p
+	* ext/socket/socket.c (sock_connect_nonblock): ditto
+	* ext/socket/rubysocket.h (rsock_opt_false_p): new function
+	* ext/socket/basicsocket.c (bsock_recv_nonblock): update rdoc
+	* ext/socket/udpsocket.c (udp_recvfrom_nonblock): ditto
+	* test/socket/test_nonblock.rb: new tests
+	  [ruby-core:69542] [Feature #11229]
+
+Mon Jun 15 14:33:02 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb: Make Set#each and SortedSet#each generate a sized
+	  enumerator. [GH-931] by kachick (Kenichi Kamiya)
+
+	* test/test_set.rb: Import tests from Set into SortedSet. [GH-931]
+	  by kachick (Kenichi Kamiya)
+
+Mon Jun 15 02:26:34 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#connect): use connect_nonblock and
+	  io/wait to eliminate timeout use. fix GH-899
+
+Sat Jun 13 07:21:18 2015  KOSAKI Motohiro  <kosaki.motohiro@gmail.com>
+
+	* thread.c (thread_start_func_2): don't interrupt when last thread
+	  exit unless main thread is already exited. Otherwise main thread
+	  could be wrongly interrupted when it uses rb_thread_call_without_gvl().
+	  Patch by Takehiro Kubo. [Bug #11237][ruby-dev:49044][GH-898]
+
+	* test/-ext-/gvl/test_last_thread.rb: new test for the above fix.
+
+	* ext/-test-/gvl/call_without_gvl/call_without_gvl.c: new ext for
+	  the above test.
+	* ext/-test-/gvl/call_without_gvl/extconf.rb: ditto.
+
+Mon Jun 15 00:14:33 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/lib/pathname.rb (descend): Blockless form supported.
+	  (ascend): Ditto.
+	  [ruby-core:68820] [Feature #11052] Patch by Piotr Szotkowski.
+
+Sun Jun 14 20:09:25 2015  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (time_getlocaltime): [DOC] Add examples of valid utc_offset
+	  formats.
+	  [ruby-core:68306] [Misc #10905] Patch by Charles Korn.
+
+Sun Jun 14 18:49:56 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/socket/raddrinfo.c (parse_numeric_port): Detect
+	  port overflow.
+	  (numeric_getaddrinfo): Use parse_numeric_port.
+	  numeric_getaddrinfo fails if port is too big now.
+	  This makes rb_getaddrinfo invokes the real getaddrinfo()
+	  on such condition.
+	  This change is related to [ruby-core:69355] [Bug #11179].
+
+Sun Jun 14 17:26:03 2015  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_chunk_while): New method Enumerable#chunk_while.
+	  [ruby-core:67738] [Feature #10769] proposed by Tsuyoshi Sawada.
+
+Sun Jun 14 17:20:40 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_load_ok): try opening file without gvl not to
+	  lock entire process.  [Bug #11060]
+
+Sun Jun 14 10:43:50 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/runruby.rb: just remove the lines of RUBY_VERSION check and raise
+	  instead of replacing the check to `true`, for getting rid of a
+	  warning `possibly useless use of true in void context`.
+
+Sun Jun 14 10:13:55 2015  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/source.rb (REXML::IOSource#scan): Fix a typo in
+	  document. [fix GH-934]
+	  Patch by Anton Davydov. Thanks!!!
+
+Sun Jun 14 10:09:48 2015  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/attlistdecl.rb (REXML::AttlistDecl): Fix a typo in
+	  document. [fix GH-934]
+	  Patch by Anton Davydov. Thanks!!!
+
+Sun Jun 14 06:24:57 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/ruby/test_io.rb: add test for IO.binread fd leak.
+	  See r50881.
+
+Sun Jun 14 05:23:51 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* io.c (rb_io_s_binread): close fd if seek offset is invalid.
+
+Sun Jun 14 04:40:32 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/lib/leakchecker.rb (check): refactor.
+
+Sun Jun 14 04:34:14 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/lib/leakchecker.rb: Return false for no leaks.
+	  Otherwise the GC could run for nothing.
+
+Sun Jun 14 04:15:40 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* lib/delegate.rb: [DOC] Update SimpleDelegator example. [ci skip]
+
+Sat Jun 13 20:28:14 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_stat_ino): get inode from the interval of struct st.
+
+	* win32/win32.c (stati64_set_inode): get nFileIndexHigh/Low, and set it
+	  to the interval of struct st as inode.
+	  [Feature #11216]
+
+	* win32/win32.c (stati64_set_inode_handle): call stati64_set_inode.
+
+	* win32/win32.c (rb_w32_fstati64): call stati64_set_inode_handle.
+
+	* win32/win32.c (stati64_handle): call stati64_set_inode.
+
+Sat Jun 13 19:44:53 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/io/console/depend (.list.chksum): revert a part of r50859, because
+	  it was not mentioned at its commit log and it caused a build error on
+	  Windows ($(MAKE) is already quoted).
+
+Sat Jun 13 17:35:11 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_core.h (rb_thread_t): add th->name.
+
+	* vm.c (th_init): initialize th->name.
+
+	* thread.c (Init_Thread): add Thread.name and name=. [Feature #11251]
+
+	* thread.c (rb_thread_inspect): show thread's name if set.
+
+	* thread.c (rb_thread_getname): defined.
+
+	* thread.c (rb_thread_setname): ditto.
+
+Sat Jun 13 11:39:43 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/tempfile.rb: Fix typo. [fix GH-933] Patch by @Zorbash
+
+Sat Jun 13 11:38:00 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/openssl/ossl_ocsp.c: fix documentation on ocsp response cert status.
+	  [fix GH-932] Patch by @chrisholmes
+
+Sat Jun 13 11:35:19 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.gemspec: Fix require paths for released gem.
+	  [fix GH-929] Patch by @voxik
+	* ext/io/console/io-console.gemspec: ditto.
+
+Sat Jun 13 00:45:08 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Return sized enumerators.
+	  Patch by Kenichi Kamiya [GH-931]
+
+Sat Jun 13 00:45:06 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Fix with_object with no block given
+
+Sat Jun 13 00:44:59 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Have with_index accept an offset parameter.
+	  Based on patch by T Yamada. [#11007]
+
+Fri Jun 12 22:21:12 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/test_extlibs.rb (TestExtLibs::check_existence): fix
+	  error. [Bug #11255] [ruby-dev:49079]
+
+Fri Jun 12 21:17:46 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* pack.c (pack_{un,}pack): new template character `j` and `J`, pointer
+	  with signed and unsigned integers.
+
+	* NEWS: mention about this feature.
+	  [Feature #11215] [ruby-dev:49015]
+
+Fri Jun 12 21:01:44 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* file.c (File::SHARE_DELETE): new flag to be able to delete opened file
+	  on Windows.
+
+	* include/win32/win32.c (O_SHARE_DELETE): new pseudo file mode flag.
+
+	* win32/win32.c (rb_w32_{w,}open): support above flag.  [EXPERIMENTAL]
+
+	* NEWS: mention about this feature.
+	  [Feature #11218] [ruby-dev:49022]
+
+Fri Jun 12 18:21:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: added missing commit message.
+
+Fri Jun 12 18:20:37 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* sample/exyacc.rb: Fix some typos. [fix GH-927] Patch by @davydovanton
+	* sample/list.rb: ditto.
+	* sample/trick2013/kinaba/remarks.markdown: ditto.
+
+Fri Jun 12 17:34:14 2015  Wojciech Mach  <wojtek@wojtekmach.pl>
+
+	* hash.c (rb_hash_fetch_values): add `Hash#fetch_values`.
+	  [Feature #10017] [Fix GH-776]
+
+Fri Jun 12 16:28:17 2015  Radan Skoric  <radan.skoric@gmail.com>
+
+	* array.c (rb_ary_bsearch_index): Implement Array#bsearch_index
+	  method, which is similar to bsearch and returns the index or
+	  nil.  [Feature #10730]
+
+Thu Jun 11 19:11:22 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/zlib/zlib.c: Fix indentation for rdoc.
+	  [Bug #11221][ruby-core:69465]
+
+Thu Jun 11 16:23:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h (METHOD_ENTRY_BASIC_SET): fix last commit (unbalanced parens).
+
+Thu Jun 11 15:14:16 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: define SET_THREAD_NAME if it has pthread_set_name_np
+	  for FreeBSD, and don't define it if both pthread_setname_np
+	  and pthread_set_name_np don't exist.
+
+	* thread_pthread.c (SET_THREAD_NAME): don't define if they don't exist.
+
+	* thread_pthread.c (native_set_thread_name): run if SET_THREAD_NAME
+	  is defined.
+
+Thu Jun 11 15:53:31 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h (METHOD_ENTRY_BASIC_SET): should clear last bit.
+
+Thu Jun 11 14:34:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb: bump version to 2.4.7 and 2.4.8. these versions fixed
+	  CVE-2015-3900.
+	* lib/rubygems/remote_fetcher.rb: ditto.
+	* test/rubygems/test_gem_remote_fetcher.rb: added testcase for CVE-2015-3900
+
+Thu Jun 11 14:18:51 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb: bump version to 2.4.6. It's missing change at r49774.
+
+Thu Jun 11 13:50:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (ary_ensure_room_for_push): check if array size will
+	  exceed maximum size to get rid of buffer overflow.
+	  [ruby-dev:49043] [Bug #11235]
+
+	* array.c (ary_ensure_room_for_unshift, rb_ary_splice): ditto.
+
+Thu Jun 11 13:17:34 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/test_cmath.rb (TestCMath#test_trigonometric_functions): should
+	  not compare float values (in complex values) by `==`.
+
+Thu Jun 11 08:52:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: embed rb_method_entry_t::attr::flags (5 bits) into
+	  rb_method_entry_t::flags to make one word spare space.
+
+	  Add some macros to access these flags.
+
+	* vm_method.c: use these macros.
+
+	* internal.h: define IMEMO_FL_USHIFT and IMEMO_FL_USER[0-4]
+	  for T_IMEMO local flags.
+
+Thu Jun 11 08:27:06 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c: use VM_ASSERT instead of assert().
+
+	* vm_args.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_method.c: ditto.
+
+Thu Jun 11 08:16:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: define VM_ASSERT() for assertion
+	  enabled only when (VM_CHECK_MODE > 0).
+
+	* vm_insnhelper.c: move definition VM_CHECK_MODE
+	  from vm_insnhelper.c to vm_core.h.
+
+	* vm.c: remove <assert.h>
+
+Thu Jun 11 06:46:07 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (check_frame): check type of cref_or_me first.
+
+Thu Jun 11 04:34:39 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* test/test_cmath.rb: Add some assertions.
+
+Thu Jun 11 00:34:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in: expanded macro result may have spaces
+	  between tokens.  [ruby-dev:49047] [Bug #11243]
+
+Wed Jun 10 22:27:32 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* lib/mkmf.rb: revert r50804 because of build failure when specifying
+	  LDFLAGS during configure, observed on Solaris with GCC 4.6.
+	  [Bug #11245]
+
+Wed Jun 10 21:59:51 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* README.ja.md: fix markup miss.
+
+Wed Jun 10 11:06:25 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* template/fake.rb.in: turn into erb template from autoconf
+	  template to fake more accurately.
+
+	* common.mk (fake.rb): needs preprocessed file now.
+
+	* version.c (Init_version): add dummy expression to
+	  RUBY_ENGINE_VERSION.
+
+Tue Jun  9 12:31:25 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Simplify and optimize EratosthenesSieve
+
+Tue Jun  9 11:45:00 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Simplify and optimize EratosthenesSieve
+	  based on patch by Ajay Kumar. [Fixes GH-921]
+
+Mon Jun  8 05:09:58 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info): print method id for T_IMEMO/ment.
+
+Sun Jun  7 07:05:43 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* Move test cases from test/ruby/test_complex.rb to test/test_cmath.rb
+
+Sat Jun  6 18:23:41 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: back to share rb_method_definition_t by
+	  rb_method_entry_t.
+
+	  r50728 changed sharing `def's to isolating `def's
+	  on alias and so on. However, this change conflicts
+	  future improvement plan. So I change back to sharing approach.
+
+	* method.h: move rb_method_definition_t::flags to
+	  rb_method_entry_t::attr::flags.
+
+	  rb_method_entry_t::attr is union with VALUE because this field
+	  should have same size of VALUE. rb_method_entry_t is T_IMEMO).
+
+	  And also add the following access macros to it's fields.
+
+	  * METHOD_ENTRY_VISI(me)
+	  * METHOD_ENTRY_BASIC(me)
+	  * METHOD_ENTRY_SAFE(me)
+
+	* vm_method.c (rb_method_definition_addref): added instead of
+	  rb_method_definition_clone().
+
+	  Do not create new definition, but increment alias_count.
+
+	* class.c (clone_method): catch up this fix.
+
+	* class.c (method_entry_i): ditto.
+
+	* proc.c (mnew_internal): ditto.
+
+	* proc.c (mnew_missing): ditto.
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_method.c: ditto.
+
+Sat Jun  6 15:59:38 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c: ins_methods_push() needs rb_method_visibility_t type on
+	  2nd arg.
+
+Sat Jun  6 15:05:47 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (ins_methods_push): Change type and name of parameters
+	  to make more clear.
+
+Sat Jun  6 08:52:13 2015  Eric Wong  <e@80x24.org>
+
+	* test/socket/test_nonblock.rb: try to avoid EMSGSIZE
+	  [ruby-core:69466]
+
+Sat Jun  6 07:58:30 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove struct mark_tbl_arg and pass objspace directly
+	  to avoid indirect access overhead.
+
+Sat Jun  6 07:08:45 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): drop redundant assignment
+
+Sat Jun  6 01:00:06 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): all arguments are
+	  parsed even on systems without HAVE_STRUCT_MSGHDR_MSG_CONTROL
+	  to prevent SEGV caused by passing Qnil to RARRAY_LENINT and
+	  to preserve behavior before r50776.
+	  [Bug #11224] [ruby-core:69468] [Bug #11225] [ruby-core:69469]
+
+Fri Jun  5 22:37:42 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (ins_methods_push): change 3rd parameter's type
+	  from long to rb_method_visibility_t.
+
+	* class.c (ins_methods_i): catch up this fix.
+
+	* class.c (method_entry_i): cast to st_data_t instead of `long'.
+
+Fri Jun  5 20:37:10 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: move definition of rb_cref_t to method.h.
+
+	* eval_intern.h: move definition of rb_scope_visibility_t
+	  to method.h.
+
+	* method.h: change rb_cref_t::scope_visi from VALUE to
+	  rb_scope_visibility_t.
+	  [Bug #11219]
+
+	* vm.c (vm_cref_new): accept rb_method_visibility_t directly.
+
+	* vm_insnhelper.c (rb_vm_rewrite_cref): don't use 0,
+	  but METHOD_VISI_UNDEF.
+
+	* vm_method.c (rb_scope_visibility_set): don't need to use cast.
+
+	* vm_method.c (rb_scope_module_func_set): ditto.
+
+Fri Jun  5 17:27:30 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): avoid msg_control
+	  ptr if msg_controllen is zero to fix portability problems.
+
+Fri Jun  5 09:17:45 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): fix build error
+	  from r50776
+
+Fri Jun  5 07:05:58 2015  Eric Wong  <e@80x24.org>
+
+	* io.c (sym_wait_readable, sym_wait_writable): declare
+	  (io_getpartial): use sym_wait_readable
+	  (io_write_nonblock): use sym_wait_writable
+	  (Init_IO): initialize sym_wait_*able
+
+Fri Jun  5 06:43:00 2015  Eric Wong  <e@80x24.org>
+
+	* doc/extension.rdoc: note rb_get_kwargs changes keywords_hash
+	  [ruby-core:68507]
+
+Fri Jun  5 05:50:29 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal): use rb_scan_args
+	  [ruby-core:69439] [Feature #11207]
+
+Fri Jun  5 02:20:06 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/lib/envutil.rb (assert_no_memory_leak): change default value
+	  of limit from 1.5 to 2.0. It is ad-hoc fix to solve test failure
+	  in travis-ci.
+
+Fri Jun  5 01:17:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_defined): no need to use cast.
+
+Fri Jun  5 01:14:02 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_defined): show additional messages on rb_bug().
+
+Fri Jun  5 01:06:02 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_add_method_iseq): use intermediate struct to
+	  avoid initializing struct with variables.
+	  [Bug #11217]
+
+	* method.h: add a comment about it.
+
+Fri Jun  5 00:55:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: constify rb_method_refined_t::orig_me.
+
+	  Also constify the following functions.
+
+	  * rb_resolve_refined_method()
+	  * rb_method_entry_with_refinements()
+	  * rb_method_entry_without_refinements()
+	  * rb_method_entry_copy()'s parameter.
+
+	* class.c: catch up this fix.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_method.c: ditto.
+
+Thu Jun  4 12:47:54 SGT 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* array.c: Revert r50763. because "reentered" is not typo.
+
+Thu Jun  4 11:12:29 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c: fix a typo. Patch by @davydovanton
+	  [fix GH-923]
+	* include/ruby/st.h: ditto.
+	* include/ruby/util.h: ditto.
+
+Thu Jun  4 10:54:30 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* array.c: fix a typo. Patch by @manish-shrivastava
+	  [fix GH-922]
+
+Thu Jun  4 09:52:02 2015  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/lib/openssl/ssl.rb: use io/nonblock instead of fcntl
+	  [ruby-core:69382] [Feature #11190]
+
+Thu Jun  4 07:22:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: introduce rb_method_refined_t for refined method entry.
+
+	* class.c (move_refined_method): catch up this fix.
+
+	* gc.c (mark_method_entry): ditto.
+
+	* vm_eval.c (vm_call0_body): ditto.
+
+	* vm_insnhelper.c (vm_call_method): ditto.
+
+	* vm_method.c: ditto.
+
+Thu Jun  4 07:12:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dirent_match): match short names only when FNM_SHORTNAME
+	  flag is given, for the backward compatibility, and the new
+	  behavior is often dangerous.  [ruby-core:69435] [Bug #11206]
+
+Thu Jun  4 05:44:01 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (special_generic_ivar): remove flag
+	  (givar_i, rb_mark_generic_ivar_tbl): remove functions
+	  (rb_free_generic_ivar, rb_ivar_lookup, rb_ivar_delete,
+	   generic_ivar_set, rb_ivar_set, rb_ivar_defined,
+	   rb_copy_generic_ivar, rb_ivar_foreach, rb_ivar_count,
+	   rb_obj_remove_instance_variable):
+	   adjust for lack of ivar support in special constants
+	* test/ruby/test_variable.rb: test ivars for special consts
+	* internal.h: remove rb_mark_generic_ivar_tbl decl
+	* gc.c (gc_mark_roots): remove rb_mark_generic_ivar_tbl call
+	  [ruby-core:69441] [Feature #11208]
+
+Thu Jun  4 05:13:34 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (def_iseq_ptr): `iseqval' is not available any more.
+
+Thu Jun  4 04:50:12 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (method_entry_i): mtbl should not have `me' as NULL.
+
+Thu Jun  4 04:28:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (clone_method): do not use me->klass, but use explicitly
+	  passed argument.
+
+Thu Jun  4 04:10:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (rb_vm_rewrite_cref_stack): rename to rb_vm_rewrite_cref().
+
+	* class.c (clone_method): use renamed name.
+
+	* vm_insnhelper.c (rb_vm_rewrite_cref): do not use `node' in variable
+	  names.
+
+Wed Jun  3 23:03:50 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c: rename `rb_frame_...' to `rb_scope_...'.
+
+	* eval_intern.h: move decl. of rb_scope_visibility_set() to method.h.
+
+	* load.c: catch up this fix.
+
+Wed Jun  3 21:14:20 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/rbconfig/sizeof/extconf.rb: Check C99 standard integer types.
+
+Wed Jun  3 21:00:47 2015  Tanaka Akira  <akr@fsij.org>
+
+	* configure.in: Don't check __int128.
+
+	* ext/rbconfig/sizeof/extconf.rb: Check __int128.
+
+	* ext/rbconfig/sizeof/depend: sizes.c depends on
+	  ext/rbconfig/sizeof/extconf.rb.
+
+	* template/sizes.c.tmpl: Detect check_sizeof.
+
+Wed Jun  3 20:07:07 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (clone_method): remove redundant check for me->def != NULL.
+	  Now, all `me` have `me->def`.
+
+	* proc.c (rb_method_entry_location): ditto.
+
+	* vm.c (rb_vm_check_redefinition_opt_method): ditto.
+
+	* vm.c (add_opt_method): ditto.
+
+	* vm_eval.c (vm_call0_body): ditto.
+
+Wed Jun  3 19:24:12 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: rename enum missing_reason to enum method_missing_reason.
+
+	* vm_core.h: use enum method_missing_reason for
+	  rb_thread_t::method_missing_reason.
+
+	* vm_eval.c: catch up this fix.
+
+	* vm_insnhelper.c: ditto.
+
+Wed Jun  3 16:17:21 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* vm.c: eagerly allocate `loading_table`.  This eliminates the need to
+	  do NULL checks when looking up the `loading_table` hash.
+	  https://github.com/ruby/ruby/pull/918
+
+	* load.c: remove various NULL checks
+
+Wed Jun  3 11:47:15 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: change fields order to gather frequent access fields.
+
+	* vm_insnhelper.c (vm_call_method): add LIKELY().
+
+Wed Jun 03 10:35:45 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: split rb_method_definition_t::flag to several flags.
+
+	  `flag' contains several categories of attributes and it makes us
+	  confusion (at least, I had confused).
+
+	  * rb_method_visibility_t (flags::visi)
+	    * NOEX_UNDEF     -> METHOD_VISI_UNDEF     = 0
+	    * NOEX_PUBLIC    -> METHOD_VISI_PUBLIC    = 1
+	    * NOEX_PRIVATE   -> METHOD_VISI_PRIVATE   = 2
+	    * NOEX_PROTECTED -> METHOD_VISI_PROTECTED = 3
+	  * NOEX_SAFE(flag)  -> safe (flags::safe, 3 bits)
+	  * NOEX_BASIC       -> basic (flags::basic, 1 bit)
+	  * NOEX_MODFUNC     -> rb_scope_visibility_t in CREF
+	  * NOEX_SUPER       -> MISSING_SUPER (enum missing_reason)
+	  * NOEX_VCALL       -> MISSING_VCALL (enum missing_reason)
+	  * NOEX_RESPONDS    -> BOUND_RESPONDS (macro)
+
+	  Now, NOEX_NOREDEF is not supported (I'm not sure it is needed).
+
+	  Background:
+	    I did not know what "NOEX" stands for.
+	    I asked Matz (who made this name) and his answer was "Nothing".
+	    "At first, it meant NO EXport (private), but the original
+	    meaning was gone."
+	    This is why I remove the mysterious word "NOEX" from MRI.
+
+	* vm_core.h: introduce `enum missing_reason' to represent
+	  method_missing (NoMethodError) reason.
+
+	* eval_intern.h: introduce rb_scope_visibility_t to represent
+	  scope visibility.
+	  It has 3 method visibilities (public/private/protected)
+	  and `module_function`.
+
+Wed Jun  3 08:06:30 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gem/bundled_gems: updated to test-unit 3.1.1 and minitest 5.7.0.
+
+Wed Jun  3 04:48:05 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_defined): check respond_to_missing?
+	  at defined?(func()).
+	  [Bug #11212]
+
+	* test/ruby/test_defined.rb: add a test for this fix.
+
+Wed Jun  3 04:34:39 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (vm_defined): skip respond_to_missing? when
+	  a method is available.
+	  [Bug #11211]
+
+	* test/ruby/test_defined.rb: add a test for this fix.
+
+Wed Jun  3 04:14:13 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (defined), vm_insnhelper.c (vm_defined):
+	  move instruction body to the vm_defined() function.
+
+Wed Jun  3 02:29:25 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* test/ruby/test_module.rb: Do not assume class variable order.
+	  Patch by @enebo.
+
+Wed Jun  3 01:10:38 2015  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* vm_method.c (rb_method_definition_set): remove a double assignment.
+	  Coverity Scan found this bug.
+
+Wed Jun  3 00:04:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_alias): rename parameter names.
+
+Tue Jun  2 23:27:18 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: remove rb_method_iseq_t::iseqval.
+	  While making a r50728, iseqval is needed (to mark correctly),
+	  but now just iseqptr is enough.
+
+	* class.c: catch up this fix.
+
+	* gc.c: ditto.
+
+	* proc.c: ditto.
+
+	* vm_method.c: ditto.
+
+Tue Jun  2 21:29:28 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (proc_curry): remove a debug line.
+
+Tue Jun  2 12:43:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: make rb_method_entry_t a VALUE.
+	  Motivation and new data structure are described in [Bug #11203].
+
+	  This patch also solve the following issues.
+
+	  * [Bug #11200] Memory leak of method entries
+	  * [Bug #11046] __callee__ returns incorrect method name in orphan
+	                 proc
+
+	* test/ruby/test_method.rb: add a test for [Bug #11046].
+
+	* vm_core.h: remove rb_control_frame_t::me. me is located at value
+	  stack.
+
+	* vm_core.h, gc.c, vm_method.c: remove unlinked_method... codes
+	  because method entries are simple VALUEs.
+
+	* method.h: Now, all method entries has own independent method
+	  definitions. Strictly speaking, this change is not essential,
+	  but for future changes.
+
+	  * rb_method_entry_t::flag is move to rb_method_definition_t::flag.
+	  * rb_method_definition_t::alias_count is now
+	    rb_method_definition_t::alias_count_ptr, a pointer to the counter.
+
+	* vm_core.h, vm_insnhelper.c (rb_vm_frame_method_entry) added to
+	  search the current method entry from value stack.
+
+	* vm_insnhelper.c (VM_CHECK_MODE): introduced to enable/disable
+	  assertions.
+
+Tue Jun  2 10:46:36 2015  Eric Wong  <e@80x24.org>
+
+	* test/socket/test_nonblock.rb: new test for sendmsg_nonblock
+
+Tue Jun  2 09:04:14 2015  Eric Wong  <e@80x24.org>
+
+	* lib/benchmark.rb: just use Process::CLOCK_MONOTONIC
+	  [ruby-core:69390]
+
+Mon Jun  1 22:01:27 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (pkg_config): split --libs if --libs-only-l option
+	  is not available.  patch in [ruby-core:69428] by Hans Mackowiak.
+	  [ruby-core:69421] [Bug #11201]
+
+Mon Jun  1 21:18:24 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_children): remove a garbage character
+	  introduced at the last commit.
+
+Mon Jun  1 19:24:42 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_method_entry_make): do not show warning message
+	  when method_entry is an alias.
+
+Mon Jun  1 15:47:16 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: move class related definitions.
+
+Mon Jun  1 15:43:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c: remove needless include pragma for method.h.
+
+	* struct.c: ditto.
+
+	* vm_method.c: ditto.
+
+Mon Jun  1 05:42:00 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h, vm_method.c: move macros to functions.
+	  * SCOPE_TEST(f)  -> rb_frame_visibility_test(flag).
+	  * SCOPE_CHECK(f) -> rb_frame_visibility_check(flag).
+	  * SCOPE_SET(f)   -> rb_frame_visibility_set(flag).
+
+	* load.c (load_ext): use it.
+
+Mon Jun  1 04:47:37 2015  Zachary Scott  <e@zzak.io>
+
+	* ext/date/date_core.c: [DOC] Add comparison of Time and DateTime
+	  Patch provided by @pixeltrix
+
+Mon Jun  1 04:22:09 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (VM_FRAME_MAGIC_DUMMY): introduce new frame type to
+	  recognize dummy frame.
+
+	* vm.c (th_init): use new frame type.
+
+	* vm_args.c (raise_argument_error): ditto.
+
+Mon Jun  1 04:15:42 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (rb_class_has_methods): added to reduce dependency
+	  to internal class data structure.
+
+	* internal.h: ditto.
+
+	* hash.c (has_extra_methods): use added function.
+
+Mon Jun  1 04:11:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c , gc.h (rb_obj_info): export obj_info(VALUE) for debugging.
+
+Mon Jun  1 03:52:55 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_gc.rb: increase timeout seconds for GC stressful
+	  debugging.
+
+Sun May 31 04:18:06 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: fix typo of comments.
+
+Sun May 31 03:36:42 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: add VM_METHOD_TYPE_ALIAS rb_method_definition_t::type
+	  to fix [Bug #11173].
+
+	  Now, inter class/method alias creates new method entry
+	  VM_METHOD_TYPE_ALIAS, which has an original method entry.
+
+	* vm_insnhelper.c (find_defined_class_by_owner): added.
+	  Search corresponding defined_class from owner class/module.
+
+	* vm_method.c (rb_method_entry_get_without_cache): return me->klass
+	  directly for defined_class.
+
+	  Now, no need to check me->klass any more.
+
+	* vm_method.c (method_entry_set0): separated from method_entry_set().
+
+	* vm_method.c (rb_alias): make method entry has VM_METHOD_TYPE_ALIAS.
+
+	* vm_method.c (release_method_definition): support VM_METHOD_TYPE_ALIAS.
+
+	* vm_method.c (rb_hash_method_definition): ditto.
+
+	* vm_method.c (rb_method_definition_eq): ditto.
+
+	* vm_method.c (release_method_definition): ditto.
+
+	* vm_insnhelper.c (vm_call_method): ditto.
+
+	* vm_insnhelper.c (vm_method_cfunc_entry): ditto.
+
+	* vm_eval.c (vm_call0_body): ditto.
+
+	* gc.c (mark_method_entry): ditto.
+
+	* proc.c (method_def_iseq): ditto.
+
+	* proc.c (method_cref): ditto.
+
+	* proc.c (rb_method_entry_min_max_arity): ditto.
+
+	* test/ruby/test_alias.rb: add tests.
+
+	* test/ruby/test_module.rb: fix a test to catch up current behavior.
+
+Sun May 31 03:34:25 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (rb_unlink_method_entry): make it static.
+
+Sun May 31 03:26:58 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h, vm_method.c (rb_free_method_entry): constify a parameter.
+
+	* vm_core.h: remove useless declaration about rb_unlink_method_entry().
+
+Sat May 30 18:05:02 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/tempfile.rb (Remover#call): fixed wrong condition.  introduced at
+	  r50682.
+
+Sat May 30 16:12:35 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c: use RB_GC_GUARD instead of volatile
+	  [ruby-core:69419] [Feature #11198]
+
+Sat May 30 15:59:10 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/tempfile.rb (Tempfile#initialize): initialize @unlinked to fix
+	  test failures introduced at r50682.  I hope that check the results of
+	  tests before committing, at least the tests about the changed feature.
+
+Sat May 30 11:02:55 2015  Martin Englund <martin@englund.nu>
+
+	* doc/dtrace_probes.rdoc: change lingering dtrace probe documentation
+	  from function- to method-
+
+Sat May 30 10:26:09 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* lib/tempfile.rb: refactoring.
+	  * use warn instead of STDERR.print
+	  * remove @tmpname and use @tmpfile.path
+	  * introduce @unlinked flag
+	  * Remover takes only @tmpfile
+	  * mode will be modified just before file reopen
+
+Sat May 30 09:02:51 2015  Eric Wong  <e@80x24.org>
+
+	* include/ruby/intern.h (rb_generic_ivar_table): deprecate
+	* internal.h (rb_attr_delete): declare
+	* marshal.c (has_ivars): use rb_ivar_foreach
+	  (w_ivar): ditto
+	  (w_object): update for new interface
+	* time.c (time_mload): use rb_attr_delete
+	* variable.c (generic_ivar_delete): implement
+	  (rb_ivar_delete): ditto
+	  (rb_attr_delete): ditto
+	  [ruby-core:69323] [Feature #11170]
+
+Sat May 30 09:14:28 2015  Scott Francis  <scott.francis@shopify.com>
+
+	* cont.c (cont_free): check if ruby_current_thread is still valid.
+	  [Fix GH-914]
+
+Sat May 30 08:36:04 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (static int special_generic_ivar): move
+	  (rb_generic_ivar_table): rewrite for compatibility
+	  (gen_ivtbl_bytes): new function
+	  (generic_ivar_get): update to use ivar index
+	  (generic_ivar_update): ditto
+	  (generic_ivar_set): ditto
+	  (generic_ivar_defined): ditto
+	  (generic_ivar_remove): ditto
+	  (rb_mark_generic_ivar): ditto
+	  (givar_i): ditto
+	  (rb_free_generic_ivar): ditto
+	  (rb_mark_generic_ivar_tbl): ditto
+	  (rb_generic_ivar_memsize): ditto
+	  (rb_copy_generic_ivar): ditto
+	  (rb_ivar_set): ditto
+	  (rb_ivar_foreach): ditto
+	  (rb_ivar_count): ditto
+	  (givar_mark_i): remove
+	  (gen_ivtbl_mark): new function
+	  (gen_ivar_each): ditto
+	  (iv_index_tbl_extend): update for struct ivar_update
+	  (iv_index_tbl_newsize): ditto
+	  [ruby-core:69323] [Feature #11170]
+
+Sat May 30 08:10:46 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (iv_index_tbl_make): extract from rb_ivar_set
+	  (iv_index_tbl_extend): ditto
+	  (iv_index_tbl_newsize): ditto
+	  (rb_ivar_set): use extracted functions
+	  [ruby-core:69323] (Part 1)
+
+Fri May 29 17:39:14 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* tool/make_hgraph.rb: added.
+
+Fri May 29 14:39:00 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): out of range NTH_REF is always
+	  nil.
+
+	* parse.y (parse_numvar): check overflow of NTH_REF and range.
+	  [ruby-core:69393] [Bug #11192]
+
+	* util.c (ruby_scan_digits): make public and add length parameter.
+
+Fri May 29 11:18:58 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_sendmsg_internal,
+	  bsock_recvmsg_internal):
+	  avoid redundant fcntl on Linux
+	  [ruby-core:69154] [Feature #11145]
+	* ext/socket/init.c (rsock_s_recvfrom_nonblock): ditto
+	* ext/socket/rubysocket.h (MSG_DONTWAIT_RELIABLE): new macro
+
+Fri May 29 10:30:34 2015  Eric Wong  <e@80x24.org>
+
+	* lib/net/resolv.rb (request): use monotonic clock
+	* lib/net/http.rb (begin_transport, end_transport): ditto
+	  [ruby-core:69384] [Feature #11124]
+
+Fri May 29 04:37:38 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c: add two methods to debug internals.
+
+	  * ObjectSpace.internal_class_of: return RBASIC_CLASS(obj).
+	  * ObjectSpace.internal_super_of: return RCLASS_SUPER(cls).
+
+	* NEWS: add information about both methods.
+
+	* test/objspace/test_objspace.rb: add tests for both methods.
+
+Thu May 28 06:55:53 2015  Anton Davydov  <antondavydov.o@gmail.com>
+
+	* ext/tk/sample/figmemo_sample.rb (open_file),
+	  ext/tk/sample/tktextio.rb (TkTextIO): fix typo in messages.
+	  [Fix GH-916]
+
+Wed May 27 09:50:51 2015  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_asn1.c (ossl_asn1_traverse, ossl_asn1_decode,
+	  ossl_asn1_decode_all): use RB_GC_GUARD instead of volatile
+	  [ruby-core:69371] [Bug #11185]
+
+Wed May 27 09:27:30 2015  Eric Wong  <e@80x24.org>
+
+	* lib/drb/drb.rb (set_sockopt): remove redundant fcntl call
+	* lib/drb/unix.rb (set_sockopt): ditto
+	  [ruby-core:69128] [Feature #11137]
+
+Tue May 26 22:10:43 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (rb_alias): should resolve refined methods.
+	  [ruby-core:69360] [Bug #11182]
+
+Tue May 26 21:35:13 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* include/ruby/defines.h (RUBY_ATTR_ALLOC_SIZE): fix condition.
+
+Mon May 25 22:35:58 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (rel): after the relations is changed,
+	  @changesets is no longer right.
+
+Mon May 25 11:27:14 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/win32.c (setup_overlapped): seek to the file end only when
+	  writing (mode:a), not reading (mode:a+, read).
+
+Mon May 25 00:27:37 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* numeric.c (Numeric#negative?): [DOC] Fix call-seq.
+	  Patch by @yui-knk. [Fixes GH-908]
+
+Sun May 24 01:35:22 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* debug.c (ruby_debug_print_id): use rb_id2name() for fprintf().
+
+Sat May 23 18:38:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* man/*.1: updated dates in man pages.
+	  [ruby-dev:48988] [Bug #11171]
+
+Sat May 23 03:10:58 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write_console): should return the count of
+	  actually eaten characters, include escape sequences.
+
+Fri May 22 22:36:14 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Remove obsolete Prime.new
+	  patch by Ajay Kumar. [Fixes GH-891]
+
+Fri May 22 21:13:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_sym_count): move `rb_sym_all_symbols`
+	  to a symbol.c specific section.  a part of patch by Lourens
+	  Naude.
+
+Fri May 22 20:56:33 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* complex.c (f_complex_polar): simple bug reproduced only when y is
+	  a float but x is not a float.
+
+Fri May 22 19:42:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_spawn_process): do not discard global escape
+	  status.  [ruby-core:69304] [Bug #11166]
+
+	* process.c (rb_execarg_spawn): extract the start procedure in a
+	  parent process with ensuring the end procedure.
+
+Fri May 22 16:48:32 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: added news for net-telnet and rake
+
+Thu May 21 20:27:07 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (RUBY_DTRACE_POSTPROCESS): cmp -b is GNU extension.
+	  darwin uses GNU cmp, and FreeBSD and Solaris are not.
+	  Note that accidentally equals to expected result.
+
+Thu May 21 18:00:19 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c: constify.
+
+	* iseq.h: ditto.
+
+	* method.h: ditto.
+
+	* proc.c: ditto.
+
+	* vm_method.c: ditto.
+
+Thu May 21 17:44:53 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: fix issues caused by binding created from Method#to_proc.
+	  [Bug #11163]
+
+	* vm.c (vm_cref_new_toplevel): export as rb_vm_cref_new_toplevel().
+
+	* test/ruby/test_method.rb: add some assertions.
+
+Thu May 21 17:29:26 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/matrix.rb: added documentation for Matrix#empty and Matrix#/
+	  [Feature #10070][ruby-dev:48433] Patch by @gogotanaka
+
+Thu May 21 17:02:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: rename functions.
+	  * method_get_def() -> method_def()
+	  * method_get_iseq() -> method_def_iseq()
+	  * method_get_cref() -> method_cref()
+
+Thu May 21 16:52:44 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (rb_method_get_iseq): rename to rb_method_iseq.
+
+	* iseq.c: catch up this fix.
+
+	* iseq.h: commit ditto.
+
+Thu May 21 15:41:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/telnet.rb: gemify net-telnet.
+	  [Feature #11083]
+	* gems/bundled_gems: added net-telnet to bundled gems.
+
+Thu May 21 15:37:32 2015  Zachary Scott  <e@zzak.io>
+
+	* vm_method.c: Remove private attribute warning [Bug #10967]
+	  Patch by @spastorino [Fixes GH-849]
+	  https://github.com/ruby/ruby/pull/849
+
+	* test/ruby/test_module.rb: Update test for changes
+
+Thu May 21 10:59:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (method_proc): rename to method_to_proc.
+
+Thu May 21 10:51:54 2015  Jake Worth  <jakeworth82@gmail.com>
+
+	* io.c (rb_f_select): [DOC] Fixes for grammar and style.
+	  [Fix GH-906]
+
+Thu May 21 08:25:19 2015  Eric Wong  <e@80x24.org>
+
+	* variable.c (Init_var_tables): init generic_iv_tbl
+	  (rb_generic_ivar_table, generic_ivar_get, generic_ivar_set,
+	   generic_ivar_defined, generic_ivar_remove,
+	   rb_mark_generic_ivar, givar_i, rb_mark_generic_ivar_tbl,
+	   rb_free_generic_ivar, rb_copy_generic_ivar, rb_ivar_foreach,
+	   rb_ivar_count): remove checks for uninitialize generic_iv_tbl
+	[ruby-core:69155] [Feature #11146]
+
+Thu May 21 04:11:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (exception_type2symbol): show correct bug message.
+
+Wed May 20 23:19:05 2015  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* lib/base64.rb: Fix rdoc-formatting for padding argument.
+	  [fix GH-905][ci skip] Patch by @davydovanton
+
+Wed May 20 13:16:23 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DEPRECATED_BY): deprecate warning with preferable
+	  alternative.
+
+	* configure.in (RUBY_FUNC_ATTRIBUTE): allow attribute arguments in
+	  the macro.
+
+Wed May 20 11:23:24 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c (rb_print_backtrace): return value of libexec's backtrace
+	  is size_t, so simply cast as int.
+
+Tue May 19 18:54:41 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_compile_with_option): check source type, must be
+	  an IO or a String.  [ruby-core:69219] [Bug #11159]
+
+Tue May 19 17:15:03 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/benchmark.rb: Update Benchmark documentation and formatting.
+	  [fix GH-903][ci skip] Patch by @davydovanton
+
+Tue May 19 13:10:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_positive_p): should false on Bignum 0.
+	  http://twitter.com/rafaelfranca/status/600509783427391488
+	  [ruby-core:69173] [Feature #11151]
+
+Tue May 19 11:22:28 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/rfc2396_parser.rb (initialize_pattern):
+	  URI::Generic.build should accept port as a string.
+	  pattern[:PORT] is not defined for long.
+	  by Dave Slutzkin <daveslutzkin@fastmail.fm>
+	  https://github.com/ruby/ruby/pull/804 fix GH-804
+
+Tue May 19 11:18:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_data_typed_object_alloc),
+	  (rb_data_object_alloc): add old names for source level backward
+	  compatibilities.
+
+	* gc.c (rb_data_object_alloc, rb_data_typed_object_alloc): add
+	  aliases for binary level backward compatibilities.
+
+Tue May 19 09:54:44 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (Data_Make_Struct0): needs function pointer
+	  casts to fix function overloading in C++.
+
+Tue May 19 09:43:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (Data_Make_Struct0, TypedData_Make_Struct0):
+	  explicit cast from void* is necessary as implicit cast is
+	  disallowed in C++.
+
+Mon May 18 15:31:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_f_notimplement): should not respond to
+	  not-implemented methods.  as the address inside a DLL and the
+	  imported address are different on Windows, use an exported
+	  variable to share the same address.
+
+Mon May 18 13:55:01 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/monitor.rb (mon_try_enter, mon_enter): should reset @mon_count
+	  just in case the previous owner thread dies without mon_exit.
+	  [fix GH-874] Patch by @chrisberkhout
+
+Sun May 17 17:21:29 2015  Eric Wong  <e@80x24.org>
+
+	* lib/webrick/utils.rb (set_non_blocking): use IO#nonblock=
+	* (set_close_on_exec): use IO#close_on_exec=
+	  [Feature #11136]
+
+Sun May 17 15:01:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_positive_p, num_negative_p): add methods
+	  Numeric#positive? and Numeric#negative?.
+	  [ruby-core:69173] [Feature #11151]
+
+	* numeric.c (flo_positive_p, flo_negative_p): specialized
+	  functions for Float.
+
+	* complex.c (Init_Complex): Complex do not have positive? and
+	  negative? methods
+
+Sun May 17 14:57:42 2015  Eric Wong  <e@80x24.org>
+
+	* lib/webrick/server.rb (accept_client): avoid redundant fcntl call
+	  [Feature #11137]
+
+Sun May 17 12:13:33 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (cloexec_accept): support nonblock flag and
+	  use SOCK_NONBLOCK if possible
+	* ext/socket/init.c (rsock_s_accept_nonblock):
+	  update cloexec_accept call
+	* ext/socket/init.c (accept_blocking): ditto for blocking
+	* test/socket/test_nonblock.rb: check nonblock? on accepted socket
+	  [Feature #11138]
+
+Sun May 17 03:58:59 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* load.c (loaded_feature_path): stop returning false negatives for
+	  filenames which are trailing substrings of file extensions.  For
+	  example, 'b', which a trailing substring of ".rb" should not return
+	  false. [Bug #11155][ruby-core:69206]
+
+	* test/ruby/test_autoload.rb: test for fix
+
+Sat May 16 21:41:24 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* string.c: added documentation for character sequence \' with String#sub
+	  [Bug #11132][ruby-core:69121][fix GH-900][ci skip] Patch by @shishir127
+
+Sat May 16 21:38:05 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* enum.c: fix a sample code. Patch by @eagletmt
+	  [fix GH-901][ci skip]
+
+Sat May 16 21:17:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_data_object_wrap, rb_data_typed_object_wrap): rename
+	  alloc as wrap.  these functions do not allocate data pointers
+	  but just wrap the given pointers.
+
+Sat May 16 19:59:24 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_accept): simplified.
+
+Fri May 15 18:28:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_assoc, rb_ary_rassoc): [DOC] the result when key
+	  was found is the existing element, not a new array.  reported by
+	  Giau Nguyen <giaunv AT nustechnology.com>.
+
+Fri May 15 08:24:28 2015  Takeyuki FUJIOKA  <xibbar@ruby-lang.org>
+
+	* lib/cgi/cookie.rb: Implement HttpOnly flag for cookies.
+	  [fix GH-887] Patch by @martinpovolny
+
+Fri May 15 06:54:19 2015  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* variable.c: Change autoload to call `require` through Ruby rather
+	  than directly calling `rb_require_safe`.  This allows things like
+	  RubyGems to intercept file loading done though `autoload`.
+	  [Feature #11140]
+
+	* test/ruby/test_autoload.rb: Test for change.
+
+Wed Apr  8 19:18:02 2015  Shota Fukumori (sora_h)  <her@sorah.jp>
+
+	* enum.c (enum_grep_v, grep_i, grep_iter_i, Init_enum):
+	  Implement Enumerable#grep_v. [Feature #11049]
+
+Thu May 14 15:54:13 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/pathname/lib/pathname.rb: Remove condition of RUBY_VERSION <= 1.9.
+	  [Feature #11082]
+
+Wed May 13 17:10:37 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* enum.c (enum_to_a): revert r50457.
+	  it requires recursion check.
+	  then, it doesn't make performance improvement.
+	  [Bug #11130] [Feature #9118]
+
+Wed May 13 11:13:40 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parse_gvar): separate message for gvar without
+	  non-space characters from message for invalid identifiers.
+
+Tue May 12 22:18:27 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* enum.c (enum_to_a): fix incompatibility introduced in r50457.
+	  [Bug #11130]
+
+	* test/ruby/test_enum.rb: test for above.
+
+Tue May 12 17:08:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* method.h: remove unused declaration.
+
+Mon May 11 10:58:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: Update minitest-5.6.1 and power_assert-0.2.3.
+
+Mon May 11 00:20:31 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (Data_Make_Struct, TypedData_Make_Struct):
+	  allocate wrapper data object before allocating DATA_PTR to get
+	  rid of possible memory leak when the former failed.
+
+Sun May 10 21:32:45 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_mark_children): call dmark function for non-NULL
+	  pointers only, so that DATA_PTR can be NULL safely now.
+
+Sun May 10 16:23:58 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* proc.c (proc_binding): fix segmentation fault on marking phase.
+	  envptr of newenvval should not be NULL.
+
+	  You can reproduce by
+	   make test-all TESTS='--gc-stress -n test_to_proc_binding ruby/test_method.rb'
+
+Sun May 10 12:41:18 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* ext/zlib/zlib.c (rb_gzreader_external_encoding):
+	  define GzipReader#external_encoding.
+	  [Bug #10900]
+
+	* test/zlib/test_zlib.rb: test for above.
+
+Sun May 10 11:57:48 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* ext/win32ole/win32ole_variant.c: fix typo "indicies".
+	  the patch is from davydovanton <antondavydov.o at gmail.com>.
+	  [fix GH-892]
+
+	* lib/rubygems/indexer.rb: ditto.
+
+	* test/rubygems/test_gem_indexer.rb: ditto.
+
+Sun May 10 11:44:37 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* string.c (rb_str_crypt): Raise ArgumentError when
+	  string passed to String#crypt contains null.
+	  the patch is from jrusnack <jrusnack at redhat.com>.
+	  [Bug #10988] [fix GH-853]
+
+	* test/ruby/test_string.rb: test for above.
+
+Sun May 10 11:23:03 2015  Masaki Matsushita <glass.saga@gmail.com>
+
+	* enum.c (enum_to_a): Use size to set array capa when possible.
+	  the patch is from HonoreDB <aweiner at mdsol.com>.
+	  [fix GH-444]
+
+Sat May  9 06:48:36 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/ancdata.c (bsock_recvmsg_internal): GC guard
+	  [Bug #11123]
+
+Fri May  8 22:57:24 2015  takiy33  <takiy33@gmail.com>
+
+	* test/matrix/test_matrix.rb (test_determinant): refactor test on
+	  Matrix#determinant, by merging with test_det for an alias method
+	  det.  [Fix GH-897]
+
+Fri May  8 15:43:11 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (body_ext_mpart): should work even if body-fld-dsp
+	  is omitted.  [ruby-core:69093] [Bug #11128]
+
+Fri May  8 15:05:57 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/syntax/control_expressions.rdoc: fix a missing "a"
+	  [fix GH-888][ci skip] Patch by @riffraff
+
+Fri May  8 12:11:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_method_call_status): resolve refined method entry
+	  to check if undefined.  [ruby-core:69064] [Bug #11117]
+
+Thu May  7 22:22:59 2015  Sho Hashimoto  <sho-h@ruby-lang.org>
+
+	* proc.c: [DOC] fix Binding#local_variable_set example. [ci skip]
+
+Thu May  7 11:32:57 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (rbconfig.rb): add rule to make.
+
+Thu May  7 05:14:39 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/lib/socket.rb (connect_nonblock): use IO#wait_writable
+	* lib/drb/drb.rb (DRB::DRbTCPSocket#alive?): use IO#wait_readable
+	* lib/webrick/httpserver.rb (run): ditto
+	* lib/resolv.rb (request): ditto for single socket case
+	  [ruby-core:68943] [Feature #11081]
+
+Wed May  6 22:49:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_method_call_status): undefined refined method is
+	  not callable unless using.  [ruby-core:69064] [Bug #11117]
+
+Sun May  3 22:40:06 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* ext/-test-/file/fs.c: need to include sys/statvfs.h
+	  to use statvfs().
+
+	* ext/-test-/file/extconf.rb: check the existence of
+	  sys/statvfs.h
+
+Sun May  3 21:59:48 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/yaml.rb: fix typo. [ci skip][fix GH-890]
+	  Patch by @miketheman
+
+Sun May  3 10:02:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (linear_object_p, range_include): test if covered for
+	  linear objects.  [ruby-core:69052] [Bug #11113]
+
+Fri May  1 13:30:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln.c (dln_load): check if a different libruby is loaded by the
+	  extension library, and then bail out to get rid of very
+	  frequently reported stale bug reports.
+
+Thu Apr 30 19:51:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): revert r46873 and r46875, not to
+	  allow to execute private readers by pretending op assign.
+	  [ruby-core:68984] [Bug #11096]
+
+Thu Apr 30 17:02:33 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* rational.c: Added documentation for rational literal.
+	  [Bug #11075][fix GH-885][ci skip] Patch by @shishir127
+
+Thu Apr 30 16:39:44 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/ipsocket.c (init_inetsock_internal): preserve errno
+	  before other library calls and use rb_syserr_fail.
+	  [ruby-core:68531] [Bug #10975]
+
+Thu Apr 30 16:22:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lambda): push and reset cmdarg_stack in lambda body.
+	  [ruby-core:69017] [Bug #11107]
+
+Sun Apr 26 07:36:48 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/utf_8.c (code_to_mbclen, code_to_mbc): reject values larger
+	  than UTF-8 max codepoints.  [Feature #11094]
+
+Sat Apr 25 14:26:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_buf_cat): expand later so that the buffer can be
+	  larger for further use.  [Bug #11080] [Bug #11095]
+
+Fri Apr 24 23:48:45 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Import version 2.2.3 from
+	  https://github.com/knu/ruby-electric.el.
+
+Fri Apr 24 10:40:02 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_{getc,putc}): removed.  they are needed for old
+	  ruby (before 1.8), but not now.
+
+Fri Apr 24 08:40:13 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/win32.c: remove bcc related code.
+
+	* include/ruby/win32.h: ditto.
+
+Fri Apr 24 08:21:07 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/win32.c (rb_acrt_lowio_lock_fh): wrap _pioinfo(i)->lock.
+
+	* win32/win32.c (rb_acrt_lowio_unlock_fh): ditto.
+
+Fri Apr 24 06:47:19 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/win32.c (_filbuf): msvc14 doesn't have it, use _fgetc_nolock.
+
+	* win32/win32.c (_flsbuf): msvc14 doesn't have it, use _fputc_nolock.
+
+	* win32/win32.c (vcruntime_file): define vcruntime_file on msvc14
+	  because it doesn't export FILE's internal structure.
+
+	* win32/win32.c (FILE_COUNT): added to abstract FILE->_cnt.
+
+	* win32/win32.c (FILE_READPTR): added to abstract FILE->_ptr.
+
+	* win32/win32.c (FILE_FILENO): added to abstract FILE->_file.
+
+	* win32/win32.c (init_stdhandle): use FILE_FILENO.
+
+	* win32/win32.c (rb_w32_getc): use FILE_COUNT and FILE_READPTR.
+
+	* win32/win32.c (rb_w32_putc): ditto.
+
+Fri Apr 24 06:37:07 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/win32.c (dupfd): use _set_osfhnd.
+
+	* win32/win32.c (rb_w32_wopen): use _set_osflags.
+
+Thu Apr 24 05:38:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_roots): fox to work PRINT_ROOT_TICKS.
+
+Fri Apr 24 04:49:05 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* win32/Makefile.sub: MSVC14 have struct timespec.
+
+	* win32/rtname.cmd: support vcruntime140.dll.
+
+	* time.c (localtime_with_gmtoff_zone): MSVC14 doesn't have tzname and
+	  daylight but have _tzname and _daylight.
+
+Thu Apr 23 11:35:55 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_obj_instance_eval, rb_obj_instance_exec): allow
+	  symbols to just instance_eval/exec, except for definition of
+	  singletons.  [ruby-core:68961] [Bug #11086]
+
+Thu Apr 23 10:01:36 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/delegate.rb: fix a typo.
+	  [fix GH-881][ci skip] Patch by @Zorbash
+
+Wed Apr 22 18:36:50 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/imap.rb (move, uid_move): support the MOVE command defined
+	  in RFC6851.  Patch by ojab ojab.
+	  [ruby-core:68960] [Feature #11077]
+
+Tue Apr 22 12:42:12 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* test/ruby/test_object.rb: add tests for Kernel#String and Kernel#Array.
+	  [fix GH-879][fix GH-880] Patch by @yui-knk
+
+Tue Apr 21 20:46:02 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/ruby/test_object.rb: renamed tests to explicitly class name.
+	  [fix GH-877] Patch by @yui-knk
+
+Tue Apr 21 05:31:00 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/lib/socket.rb (connect_internal): avoid common exceptions
+	  from connect_nonblock. [ruby-core:68909]
+
+Mon Apr 20 23:46:53 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_wreadlink): follow the official format of
+	  REPARSE_DATA_BUFFER structure.
+
+Mon Apr 20 20:23:04 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk ($(arch)-fake.rb): revert r50354 because bsdmake seems not
+	  to handle such substitution.
+
+	* tool/expand-config.rb: convert path separators here.
+
+Mon Apr 20 16:52:20 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/fake.rb: don't fake libdir.  use libdirname instead.
+
+Mon Apr 20 16:49:52 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/mkmf.rb (MakeMakefile#configuration): DESTDIR should never affect
+	  top_srcdir and builddir.
+
+Mon Apr 20 16:18:17 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk ($(arch)-fake.rb): fix the path separator up for Windows.
+
+Mon Apr 20 15:02:47 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_wreadlink): fixed a bug that a junction misses
+	  its drive letter.
+
+Mon Apr 20 12:54:56 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/openssl/*: use license instead of licence.
+	  [fix GH-876][ci skip] Patch by @davydovanton
+	* lib/net/https.rb: ditto.
+
+Mon Apr 20 12:42:40 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* benchmark/bm_so_meteor_contest.rb: fix a typo.
+	  [fix GH-876][ci skip] Patch by @davydovanton
+	* tool/bisect.sh: ditto.
+	* tool/update-deps: ditto.
+
+Mon Apr 20 11:10:46 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/socket.c (sock_connect_nonblock): do not raise EISCONN
+	  [ruby-core:68926] [Feature #11072]
+	* test/socket/test_nonblock.rb: check non-EISCONN on 2nd connect
+
+Sun Apr 19 12:19:17 2015  Chad Brewbaker  <crb002@gmail.com>
+
+	* ext/{etc,openssl,tk}: Adding parens and comparisons around
+	  assignments to get rid of Wparentheses warnings.  [Fix GH-875]
+
+Sun Apr 19 10:42:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (get_env_cstr): environment variables must be ASCII
+	  compatible, as dummy encodings and wide char encodings are
+	  unsupported now.
+
+Sat Apr 18 15:18:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/parser/parser.rl: raise with messages in UTF-8
+	  encoding.  [ruby-core:67386] [Bug #10705]
+
+Fri Apr 17 11:58:34 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (STR_SET_EMBED): clear NOFREE flag at embedding as
+	  embedded strings no longer refer static strings.
+	  [ruby-core:68436] [Bug #10942]
+
+Thu Apr 16 05:15:50 2015  Eric Wong  <e@80x24.org>
+
+	* NEWS: note IO#wait_readable change in r50263
+
+Thu Apr 16 05:09:36 2015  Eric Wong  <e@80x24.org>
+
+	* lib/net/protocol.rb (rbuf_fill): use IO#wait_*able
+	* lib/net/http/generic_request.rb (wait_for_continue): ditto
+	  [ruby-core:68891] [Feature #11056]
+
+Wed Apr 15 18:43:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_tracepoint_new): fix documentation.
+	  Commented by @emilsoman.
+
+Wed Apr 15 17:36:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_trace.c (rb_tracepoint_new): Add documentation for
+	  rb_tracepoint_new C level API [ci skip]
+	  Provided by @emilsoman. [fix GH-869]
+
+Wed Apr 15 10:37:40 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/contributors.rdoc: fix a typo. Patch by @davydovanton
+	  [fix GH-872][ci skip]
+	* doc/syntax/methods.rdoc: ditto.
+	* ext/digest/sha2/sha2.c: ditto.
+	* ext/socket/ipsocket.c: ditto.
+	* ext/tk/*: ditto.
+
+Wed Apr 15 07:51:23 2015  Yuki Yugui Sonoda  <yugui@yugui.jp>
+
+	* doc/extension.ja.rdoc: Added description of TypedData_XXX.
+	  Deprecated the old DATA_XXX.
+	  Reviewed by ko1 and nobu.
+	  Fixes [ruby-dev:40881] #3064
+
+	* doc/extension.rdoc: ditto.
+
+Wed Apr 15 07:34:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (reserve_stack): keep sp safe zone to get rid
+	  of crash by -fstack-check.  [ruby-core:68740] [Bug #11030]
+
+Tue Apr 14 16:03:49 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/merger.rb (versionup): should also increment revision when
+	  changing teeny.
+
+Tue Apr 14 11:24:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/-test-/printf/printf.c (uint_to_str): renamed to get rid of
+	  conflict on cygwin.  [ruby-core:68877] [Bug #11065]
+
+Tue Apr 14 08:59:04 2015  Zachary Scott  <e@zzak.io>
+
+	* gc.c: [DOC] Improve documentation for ObjectSpace.count_objects
+	  with regards to `:TOTAL` key, with patch by @schneems [Fixes GH-871]
+	  https://github.com/ruby/ruby/pull/871 [Bug #11067]
+
+Mon Apr 13 22:44:07 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/envutil.rb (File.mkfifo): Defined using mkfifo command.
+
+	* test/ruby/test_io.rb: Use File.mkfifo.
+
+	* test/ruby/test_file_exhaustive.rb: Ditto.
+
+	* test/ruby/test_process.rb: Ditto.
+
+Mon Apr 13 21:20:20 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/ssl.rb: stricter hostname verification
+	  following RFC 6125. with the patch provided by Tony Arcieri and
+	  Hiroshi Nakamura [ruby-core:61545] [Bug #9644]
+	* test/openssl/test_ssl.rb: add tests for above.
+
+Sun Apr 12 18:40:04 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/json/json.gemspec: bump version to 1.8.2.
+	* ext/json/lib/json/version.rb: ditto.
+
+Sun Apr 12 18:12:07 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/json/json.gemspec, lib/rdoc/rdoc.gemspec: added gemspec directly.
+	* defs/default_gems, tool/rbinstall.rb: removed default_gems definition.
+	  it make simple installation for default gems.
+
+Sun Apr 12 17:35:17 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/json/*, test/json/*: Reverted r50231.  Because it's not
+	  working with cross-compile environment.
+
+Sun Apr 12 15:34:59 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (arg): fix segfault by null caused by syntax error.
+	  [ruby-core:68851] [Bug #10957]
+
+Sun Apr 12 15:11:16 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems/test_case.rb: use explicitly exception class and reverted
+	  to require JSON library for rubygems tests with Ruby 2.2.0 or earlier.
+
+Sun Apr 12 15:10:18 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/wait/wait.c (io_wait_readable): simply returns that IO is
+	  readable without blocking, but no longer returns EOF.
+
+	* ext/io/wait/wait.c (io_ready_p, io_wait_readable): try polling
+	  first and check FIONREAD optionally to see if EOF.
+	  [ruby-core:36805] [Feature #4849]
+
+Sun Apr 12 14:53:23 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems/test_case.rb: fixed json load error for rubygems tests.
+
+Sun Apr 12 14:13:28 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gc.c: Document require name for ObjectSpace methods.
+	  [ci skip][fix GH-860] Patch by @schneems
+
+Sun Apr 12 13:54:05 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_io.rb: New test that open(fifo) doesn't block other
+	  threads.
+
+Sun Apr 12 13:52:18 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/coverage/coverage.c: Remove extension from require argument.
+	  [ci skip][fix GH-870] Patch by @yui-knk
+
+Sun Apr 12 10:29:14 2015  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_connect_nonblock):
+	  support `exception: false'
+	* (get_no_exception): move function location
+	* ext/socket/socket.c (sock_connect_nonblock):
+	  support `exception: false'
+	* test/openssl/test_pair.rb (test_connect_accept_nonblock_no_exception):
+	  test `exception: false' on connect,
+	  rename from `test_accept_nonblock_no_exception'
+	* test/socket/test_nonblock.rb (test_connect_nonblock_no_exception):
+	  new test
+
+Sun Apr 12 09:57:16 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Test a block device on GNU/Linux.
+
+Sun Apr 12 09:24:03 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Test a file not owned.
+
+Sun Apr 12 09:05:44 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/fiddle/depend: Fix "Wrong mix of special targets" error with
+	  OpenBSD make.
+
+Sun Apr 12 09:04:37 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file_exhaustive.rb: Windows doesn't support Unix file
+	  modes.
+
+Sun Apr 12 08:56:44 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/-test-/file/fs.c: OpenBSD needs sys/param.h before sys/mount.h.
+
+Sun Apr 12 08:52:01 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Test suid, sgid and sticky file.
+
+Sat Apr 11 23:48:30 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* template/fake.rb.in: Don't assign baseruby, Because it's affect to
+	  Makefile of native gem like json on bundled gems.
+
+Sat Apr 11 23:33:22 2015  Tanaka Akira  <akr@fsij.org>
+
+	* file.c (rb_f_test): Consider nsec for "=", "<" and ">" for "test"
+	  method.
+
+Sat Apr 11 23:26:05 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/rbinstall.rb: support destdir for native extension gem.
+
+Sat Apr 11 21:02:06 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_file_exhaustive.rb
+	  (TestFileExhaustive#test_stat_socket_p): r50226 accidentally missed
+	  the guard for non-unix environments.
+
+Sat Apr 11 20:14:21 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/json/*, test/json/*, defs/default_gems: Gemify JSON library.
+	  [fix GH-867][Feature #11057]
+	* test/ruby/test_extlibs.rb: removed json gem from existence extensions.
+	* gems/bundled_gems: added json gem into bundled gem.
+	* lib/rdoc/rubygems_hook.rb: ignored no json environment.
+	* lib/rubygems/test_case.rb, test/rubygems/*: ditto.
+	* lib/rdoc/test_case.rb, test/rdoc/*: ditto.
+
+Sat Apr 11 15:56:58 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Create sample files lazily.
+
+Sat Apr 11 14:03:47 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Test character device using
+	  /dev/null.
+
+Sat Apr 11 10:59:58 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (append_cppflags, append_cflags, append_ldflags):
+	  utility methods to append compiler options.
+
+Sat Apr 11 08:22:24 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rdoc/text.rb: removed duplicated code.
+
+Sat Apr 11 04:46:42 2015  Eric Wong  <e@80x24.org>
+
+	* lib/net/protocol.rb (rbuf_fill): avoid exception with read_nonblock
+	  [ruby-core:68787] [Feature #11044]
+
+Fri Apr 10 23:57:44 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (need_normalization): use getattrlist() if fgetattrlist()
+	  is unavailable, on OSX 10.5.  [ruby-core:68829] [Bug #11054]
+
+Fri Apr 10 22:29:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_compile): pass the given werror flag to try_do
+	  to check if stderr is empty.
+
+	* lib/mkmf.rb (try_cflags, try_ldflags): default werror to true.
+
+	* win32/Makefile.sub (WERRORFLAG): remove useless option.  VC does
+	  not make warnings of unknown command option an error.
+
+Fri Apr 10 19:34:24 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Test socket.
+
+Fri Apr 10 19:38:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/objspace/test_objspace.rb: remove debug prints.
+
+Fri Apr 10 19:35:51 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c: add ObjectSpace.count_imemo_objects method
+	  to count imemo objects for each type.
+
+	* test/objspace/test_objspace.rb: add a test.
+
+	* NEWS: describe about this addition.
+
+Fri Apr 10 19:34:24 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Test anonymous pipe.
+
+Fri Apr 10 19:31:56 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_file_exhaustive.rb: Test named pipe.
+
+Fri Apr 10 19:10:34 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/objspace/objspace.c (setup_hash): unify common routine.
+
+Fri Apr 10 18:29:49 2015  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (rb_execarg_parent_start1): Handle EINTR.
+
+Fri Apr 10 17:27:58 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (vm_exec): check other events when RETURN is thrown.
+	  [Bug #10724]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Fri Apr 10 11:44:09 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/extconf.rb: check warnings.
+
+	* lib/mkmf.rb (try_cflags): pass options to try_compile.
+
+	* win32/Makefile.sub (WERRORFLAG): make unknown command line
+	  options an error.
+
+Fri Apr 10 08:00:17 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_process.rb: unfortunately, windows is not POSIX...
+	  cygwin has mkfifo command, but it does not affect system-wide.
+
+Fri Apr 10 02:18:53 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_process.rb: Use mkfifo command instead of mknod
+	  command to create a named pipe.  mkfifo command is defined by POSIX.
+
+Fri Apr 10 01:29:05 2015  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: Release GVL when opening a file in spawn() to avoid whole
+	  process blocking when opening a named pipe.
+	  (open_func): New function.
+	  (rb_execarg_parent_start1): Extracted from rb_execarg_parent_start and
+	  use rb_thread_call_without_gvl2 to release GVL when opening a file.
+	  (rb_execarg_parent_start): Invoke rb_execarg_parent_start1 via
+	  rb_protect and invoke rb_execarg_parent_end when error.
+
+Thu Apr  9 22:19:19 2015  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (redirect_open): Removed.
+
+Thu Apr  9 21:38:20 2015  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg_parent_end): Declared.
+
+	* process.c: "spawn" opens files in the parent process.
+	  (check_exec_redirect): Add an placeholder for fd in parameters
+	  for fd_open.
+	  (check_exec_fds_1): Delete fd_open condition.
+	  (check_exec_fds): Don't call check_exec_fds_1 with fd_open.
+	  (rb_execarg_parent_start): Open files specified as "spawn" options
+	  and add "dup2" options.
+	  (rb_execarg_parent_end): New function to close opened fds.
+	  (run_exec_open): Removed.
+	  (rb_execarg_run_options): Don't call run_exec_open.
+	  (rb_spawn_internal): Call rb_execarg_parent_end.
+
+	* io.c (pipe_open): Call rb_execarg_parent_end.
+
+	* ext/pty/pty.c (establishShell): Call rb_execarg_parent_end.
+
+Thu Apr  9 20:52:31 2015  Tanaka Akira  <akr@fsij.org>
+
+	* internal.h (rb_execarg_parent_start): Renamed from rb_execarg_fixup.
+
+	* process.c: Follows the above change.
+
+	* io.c: Ditto.
+
+	* ext/pty/pty.c: Ditto.
+
+Thu Apr  9 20:35:12 2015  Tanaka Akira  <akr@fsij.org>
+
+	* process.c (fd_clear_cloexec): Extracted from run_exec_dup2.
+
+Thu Apr  9 09:26:47 2015  Eric Wong  <e@80x24.org>
+
+	* ccan/list/list.h: sync with ccan upstream
+	  [ccan commit c2fbfe5282ba264f3485586e7efa8a5967f2d386]
+
+Thu Apr  9 08:24:03 2015  Masahiro Tomita  <tommy@tmtm.org>
+
+	* ext/socket/raddrinfo.c (addrinfo_mload): fix memory leak of
+	  addrinfo.  [ruby-dev:48923] [Bug #11051]
+
+Wed Apr  8 17:45:02 2015  Shannon Skipper  <shannonskipper@gmail.com>
+
+	* version.c (Init_version): the version of the engine or
+	  interpreter.  [Fix GH-858]
+
+Wed Apr  8 16:15:30 2015  Kenta Murata  <mrkn@cookpad.com>
+
+	* bigdecimal: conform to ruby's license.  [ruby-core:68466] [Bug #10952]
+
+Wed Apr  8 14:57:06 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_wreadlink): should treat junctions like as
+	  symlinks.
+
+Wed Apr  8 14:03:47 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_symbol.rb: fix syntax error.
+
+Wed Apr  8 13:01:06 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_any_hash): Symbols are compared by the identities
+	  always.  [ruby-core:68767] [Bug #11035]
+
+Tue Apr  7 10:22:51 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* internal.h: fix typo. Patch by @sferik [fix GH-865]
+
+Mon Apr  6 22:52:35 2015  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c: Enumerable#chunk and Enumerable#slice_before no longer takes
+	  the initial_state argument.  [Feature #10958]
+
+Mon Apr  6 16:09:58 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_args.c: protect value stack from calling other methods
+	  during complex parameter setting process (splat, kw, and so on).
+	  [Bug #11027]
+
+	* vm_core.h: remove rb_thread_t::mark_stack_len.
+	  With this modification, we don't need to use th->mark_stack_len.
+
+	* test/ruby/test_keyword.rb: add a test.
+
+	* cont.c (cont_capture): catch up this fix.
+
+	* vm.c (rb_thread_mark): ditto.
+
+Mon Apr  6 11:26:42 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/downloader.rb (http_options): prevent content auto decoding
+	  because this is a downloader.
+
+Sun Apr  5 09:55:18 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* doc/contributing.rdoc: update Maintainers list.
+
+Sun Apr  5 09:11:00 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/rbinstall.rb: fix bin script permission of bundled gems.
+
+Sun Apr  5 08:46:08 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/rbinstall.rb: support --program-suffix option.
+
+Sat Apr  4 21:31:18 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rake/*: Gemify rake [fix GH-862][Feature #11025]
+	* test/rake/*: ditto.
+	* tool/rbinstall.rb: ditto.
+
+Sat Apr  4 11:30:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_setbyte): check the argument first not to
+	  discard shared string and code range unnecessarily until
+	  actually changing the contents.  pointed out by headius.
+
+Sat Apr  4 08:16:43 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (edit_path): use path which is absolute ftp url
+	  on using ftp_proxy.
+
+Fri Apr  3 11:43:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call0_cfunc): update invoker arguments.
+
+	* vm_insnhelper.c (vm_call_cfunc_latter): ditto.
+
+	* vm_insnhelper.c (rb_vm_call_cfunc_push_frame): ditto, and prefix
+	  with rb_.
+
+Thu Apr  2 16:26:59 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, tool/mkconfig.rb: check the running ruby version in
+	  rbconfig.rb with the program version, as RUBY_VERSION has never
+	  been affected by --with-ruby-version option.
+	  [ruby-core:68639] [Bug #11002]
+
+	* configure.in (LIBRUBY_DLDFLAGS): compatibility_version must be
+	  valid version numbers, not an arbitrary string.
+
+Wed Apr  1 11:09:15 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (push_glob): remove indirect links of arguments for
+	  trampoline.
+
+Wed Apr  1 09:59:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils#mv): show the exact target path in
+	  the error message instead of the destination parent directory
+	  name.  patched by Joao Britto <jabcalves AT gmail.com> at
+	  [ruby-core:68706].  [Bug #11021]
+
+Tue Mar 31 15:25:07 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/ChangeLog-0.06_to_0.52: added archived Changelog.
+	  [ruby-list:50105]
+	* doc/ChangeLog-0.50_to_0.60: ditto.
+	* doc/ChangeLog-0.60_to_1.1: ditto.
+
+Mon Mar 30 22:02:55 2015  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* README.EXT.ja: add redirect [ruby-core:68631]
+
+Mon Mar 30 14:42:41 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (fileattr_to_unixmode, winnt_lstat): deal with
+	  symbolic link than directory, and set executable bits.
+
+Mon Mar 30 11:27:54 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (copy_stream_body): use the arguments without conversion if
+	  having read, readpartial, and write methods, than conversion by
+	  to_path method.  [ruby-core:68676] [Bug #11015]
+
+Sun Mar 29 21:08:37 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* gc.c (objspace_allrefs_destruct_i): fix a typo.
+	  [Bug #11013]
+
+Sun Mar 29 11:51:32 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_binding): replicate env from method object, and
+	  allocate the local variable area for the iseq local table.
+	  [ruby-core:68673] [Bug #11012]
+
+Sat Mar 28 09:19:41 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/date/extconf.rb: try_cflags("-std=iso9899:1999") [Bug #10906]
+	  ruby itself (including numeric.c) is built with strict compile
+	  options including -std=iso9899:1999, but ext/date is not.
+	  By the way -std=iso9899:1999 is not only a warning option but also
+	  changes behavior like MACRO definitions for example INFINITY.
+	  gcc on Solaris affect this.
+
+Fri Mar 27 16:34:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk: revert using BOOTSTRAPRUBY for enc.mk, as
+	  enc/depend uses CONFIG.  [ruby-core:68647] [Bug #11004]
+
+Thu Mar 26 10:05:13 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/test_observer.rb: add tests for Observable module.
+	  [fix GH-859] Patch by @brightbits
+
+Thu Mar 26 06:35:10 2015  Eric Wong  <e@80x24.org>
+
+	* README.EXT: add redirect [ruby-core:68631]
+
+Wed Mar 25 16:46:49 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/socket/extconf.rb: Solaris 11 has struct tcp_info.tcpi_ca_state,
+	  but it is a dummy.
+
+	* ext/socket/option.c: Solaris 11 doesn't have u_intN_t.
+
+	* ext/socket/option.c: Solaris 11 needs inspect_tcpi_msec.
+
+	* ext/socket/raddrinfo.c: Solaris 11 has AF_PACKET but doesn't have
+	  related macros.
+
+Wed Mar 25 17:03:08 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/-test-/file/fs.c (get_fsname): try magic number only if
+	  f_type is included.  [ruby-dev:48913] [Bug #11000]
+
+Wed Mar 25 11:20:40 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* benchmark/bm_hash_aref_flo.rb: make more realistic data.
+	  [ruby-core:68632] [[Bug #10999]
+
+Wed Mar 25 10:39:06 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .document: removed needless entries.
+
+Wed Mar 25 08:07:28 2015  Eric Wong  <e@80x24.org>
+
+	* doc/extension.rdoc: fix spelling of filename
+	* doc/extension.ja.rdoc: ditto.
+
+Tue Mar 25 06:55:43 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* complex.c (k_inexact_p, k_exact_zero_p): use k_exact_zero_p macro
+	  to remove k_inexact_p macro.
+
+	* complex.c (k_exact_one_p): remove unused macro k_exact_one_p.
+
+Tue Mar 24 22:23:33 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/extension.rdoc: move from toplevel document and added extname.
+	* doc/extension.ja.rdoc: ditto.
+
+Tue Mar 24 22:06:58 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/standard_library.rdoc: strip.
+
+Tue Mar 24 22:06:27 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/standard_library.rdoc: move Thread to Extensions.
+
+Tue Mar 24 21:59:10 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/contributing.rdoc: update Maintainers list.
+
+Tue Mar 24 19:10:24 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_writebarrier_generational): fix messages for rb_bug().
+	  Remove `rb_' prefix.
+
+	* gc.c (gc_writebarrier_incremental): ditto.
+
+Tue Mar 24 17:34:01 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* README.ja.md: should be chunibyo.
+
+Tue Mar 24 17:30:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/-test-/file/fs.c (get_fsname): return filesystem name by
+	  statfs/statvfs.  [ruby-core:68624] [Bug #10998]
+
+Tue Mar 24 16:46:02 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: now doesn't required spaces just after
+	  `!`.
+
+Mon Mar 23 23:18:27 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_close): don't raise on double close for consistent to
+	  IO#close.  [Feature #10950]
+
+Mon Mar 23 21:22:07 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (rb_readlink): move from file.c for better buffer
+	  allocation and the result encoding.
+
+	* win32/win32.c (wreadlink, rb_w32_ureadlink): implement readlink().
+
+Mon Mar 23 14:40:45 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (winnt_stat): stat with following symbolic links.
+
+	* win32/win32.c (winnt_lstat): rename old winnt_stat, which does
+	  not follow symbolic links.
+
+Mon Mar 23 01:44:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (r_symreal): register symbol names as strings first so
+	  that r_symlink always returns valid names.
+	  [ruby-core:68587] [Bug #10991]
+
+	* marshal.c (r_ivar, r_object0): now need to intern symbol names.
+
+	* marshal.c (r_object0): compare with symbol names.
+
+Sun Mar 22 22:07:40 2015  Kouhei Sutou  <kou@cozmixng.org>
+
+	* doc/etc.rd.ja: Fix wrong coding for Emacs.
+
+Sun Mar 22 09:53:15 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): add default CONFIGURE name to
+	  follow r50039.
+
+	* tool/make-snapshot (package): substitute configuration variables
+	  in Makefile.in instead of passing by the command line, and make
+	  temporary Makefile instead of a pipe.
+
+Sun Mar 22 08:09:47 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (ext/ripper/ripper.c, ext/rbconfig/sizeof/sizes.c):
+	  strip autogenerated dependencies which have invalid syntax in
+	  other than nmake.
+
+Sat Mar 21 15:01:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_set_winsize): use handle for
+	  writing.  GetConsoleScreenBufferInfo seems failing on a handle
+	  for reading.
+
+	* io.c: [DOC] update the example of IO#winsize to use $stdout
+	  instead of $stdin, which does not work on Windows.  a patch by
+	  Jan Lelis <mail AT janlelis.de> at [ruby-core:68574].
+	  [Bug #10986]
+
+Fri Mar 20 18:41:03 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (respond_to_missing_p): check if the receiver responds to
+	  the given method by respond_to_missing?.
+
+	* proc.c (mnew_missing): create Method object for method_missing.
+	  [ruby-core:68564] [Bug #10985]
+
+Fri Mar 20 17:43:18 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: enabled email notification.
+
+Fri Mar 20 17:39:52 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: removed Ruby 1.9.3 build on Travis CI
+
+Fri Mar 20 12:38:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info): obj_info() can receive internal objects.
+
+	* gc.c (check_rvalue_consistency): obj_info() returns const char *.
+
+Fri Mar 20 12:14:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info): show class name and T_DATA type_name.
+
+Thu Mar 19 22:12:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_copy_wb_protected_attribute): `dest' can be WB unprotected.
+
+Thu Mar 19 21:25:25 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_copy_wb_protected_attribute): demote `dest' object.
+
+Thu Mar 19 16:18:00 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c, internal.h: export rb_gc_verify_internal_consistency().
+
+Thu Mar 19 16:15:24 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info): show allocation site if GC_DEBUG is not 0.
+
+Thu Mar 19 16:12:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (newobj_of): fix RGENGC_OLD_NEWOBJ_CHECK logics.
+	  * skip on incremental marking because not sure what happen :p
+	  * rb_gc_writebarrier_remember() is enough to mark children.
+
+Thu Mar 19 16:08:42 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_copy_wb_protected_attribute): need demote for old objects.
+
+Thu Mar 19 10:31:00 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (fill_random_bytes): release the handle in the static
+	  variable, not a local variable.
+
+Thu Mar 19 06:30:35 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* object.c (rb_obj_clone): do not touch age (FL_PROMOTED[01]) because
+	  rb_obj_alloc() can return old object in debug.
+
+Thu Mar 19 06:29:28 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/objspace/test_objspace.rb: flag name was changed
+	  (long_lived -> uncollectible).
+
+Thu Mar 19 05:30:13 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (iseq_mark): skip some marking if iseq->orig is available.
+
+	* iseq.c (rb_iseq_clone): need WB for iseq1->klass = iseq0->klass
+	  (done in MEMCPY).
+
+Thu Mar 19 04:55:53 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h (IMEMO_DEBUG): added.
+
+	* internal.h: remove unused FL_IMEMO_MARK_V[0-3].
+
+	* gc.c (rb_imemo_new_debug): added.
+
+	* gc.c (obj_info): show imemo type name.
+
+Thu Mar 19 04:52:26 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RGENGC_OLD_NEWOBJ_CHECK): add check mechanism.
+
+	  If RGENGC_OLD_NEWOBJ_CHECK > 0, then create old objects
+	  (not new objects) periodically.
+
+	  Create one old objects per RGENGC_OLD_NEWOBJ_CHECK objects are
+	  created.
+
+Thu Mar 19 04:46:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* enum.c (enum_sort_by): add WBs.
+
+Thu Mar 19 03:37:52 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (check_rvalue_consistency): refactoring.
+	  * not inline on RGENGC_CHECK_MODE > 0.
+	  * check SPECIAL_CONST_P(obj) first.
+	  * add a check that remembered_bit is only TRUE when old (age == 3).
+
+	* gc.c (RVALUE_DEMOTE): should clear RVALUE_REMEMBERED bit.
+
+	  remembered_bit should be TRUE only for old (age == 3) objects.
+
+	  Actually there are no effect because demoted objects will be
+	  uncollectible WB unprotected objects (marked at the beginning of
+	  every minor GC).
+
+Thu Mar 19 02:52:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename terminologies.
+	  * long_lived -> uncollectible:
+	    because this bitmap does not mean "long lived objects in past",
+	    but means "prohibit collection these objects until next major GC".
+
+	    Uncollectible objects consist of two types objects, one is old
+	    objects (WB protected objects which age == 3) and another is
+	    uncollectible WB unprotected objects which are referred from old
+	    objects
+
+	  * remembered_wb_unprotected_objects ->
+	    uncollectible_wb_unprotected_objects:
+	    because uncollectible objects does not mean remembered objects.
+
+Wed Mar 18 17:21:12 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_writebarrier_generational): add an alternative write
+	  barrier (WB) implementation.
+	  When finding reference from [Old obj] to [New obj] by WB, current
+	  implementation marks [Old obj] as remembered old objects and marks
+	  children of [Old obj] at the beginning of marking.
+
+	  Added (but disabled) code changes current behaviour. This fix promote
+	  [New obj] to old and marks as a remembered old object. We can assume
+	  "new objects referred from old objects are maybe long-lived old
+	  objects".
+
+	  Disadvantage of added algorithm is we may promote unwilling
+	  short-lived objects. For example, consider many new objects push and
+	  pop to an old stack object. All of new objects (short-lived objects)
+	  promote to old objects unexpectedly.
+
+	  To compare these behaviour, I add this new code (but disabled it).
+
+Wed Mar 18 17:14:39 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (RVALUE_PROMOTE_RAW): rename to RVALUE_OLD_LONG_LIVED_SET()
+	  to make clear.
+
+Wed Mar 18 17:10:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (check_rvalue_consistency): do not need to check is_sweeping().
+
+Wed Mar 18 14:13:22 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm1_gc_wb_ary.rb: run GC to finish "marking" phase.
+
+	* benchmark/bm_vm1_gc_wb_obj.rb: ditto.
+
+	* benchmark/bm_vm1_gc_wb_ary_promoted.rb: add parameter
+	  full_mark: false to invoke only minor GC.
+
+	* benchmark/bm_vm1_gc_wb_obj_promoted.rb: ditto.
+
+Wed Mar 18 12:07:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: add a comment about RSTRING_FSTR.
+
+Wed Mar 18 12:01:53 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_any_hash): use same hash values with Float#hash so
+	  that -0.0 and +0.0 will be identical.
+	  [ruby-core:68541] [Bug #10979]
+
+Wed Mar 18 05:34:32 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* string.c: introduce STR_FAKESTR to show string is FAKESTR or not.
+
+	* string.c (STR_SET_SHARED): ignore FAKESTR because only Ruby objects
+	  can use write barrier.
+
+Tue Mar 17 18:59:16 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: use rb_gc_writebrrier() simply.
+	  For incremental GC, we need to get a pointer to the objspace.
+	  We can share this pointer for the following WB process.
+	  And considering icache hit ratio, process in the GC.
+
+	* gc.c (rb_gc_writebarrier): added.
+
+	* gc.c (gc_writebarrier_generational, gc_writebarrier_incremental):
+	  make them NOINLINE because inlining them into rb_gc_writebarrier()
+	  makes a prologue code of rb_gc_writebarrier() longer (storing callee
+	  save registers).
+
+	  This patch improve the performance of WB on micro-benchmarks.
+
+	  name                   ruby 2.1   trunk   modified
+	  vm1_gc_wb_ary*            0.511   0.632      0.532
+	  vm1_gc_wb_ary_promoted*   0.578   0.701      0.674
+	  vm1_gc_wb_obj*            0.419   0.575      0.492
+	  vm1_gc_wb_obj_promoted*   0.537   0.664      0.618
+	                                                 (sec)
+
+Tue Mar 17 18:51:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm1_gc_wb_ary(_promoted).rb: separate fastpath and
+	  slowpath for WB.
+
+	  Before this change bm_vm1_gc_wb_ary.rb tried to check the performance
+	  for WB slowpath (making a reference from oldobj to newobj). However,
+	  from Ruby 2.2, 3 GCs are needed to promote new objects because
+	  only 3 age objects are promoted objects.
+
+	  To compare fastpath and slowpath, introduce new "promoted" version
+	  benchmark.
+
+	  bm_vm1_gc_wb_ary.rb is for fastpath and
+	  bm_vm1_gc_wb_ary_promoted.rb is for slowpath.
+
+	* benchmark/bm_vm1_gc_wb_obj(_promoted).rb: ditto.
+
+Tue Mar 17 17:23:11 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): distinguish not-yet-stated and DT_UNKNOWN
+	  by readdir, and traverse recursively for the former.  Linux
+	  readdir returns DT_UNKNOWN on some filesystems, e.g., smbfs,
+	  iso9660.
+
+Mon Mar 16 17:43:21 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/webrick/server.rb: Fix regression bug in WEBrick's
+	  :DoNotReverseLookup config option implementation.
+	  [fix GH-731] Patch by @vais
+	* test/webrick/test_do_not_reverse_lookup.rb: ditto.
+
+Sat Mar 14 20:05:23 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* math.c (math_gamma): optimization for passed small integer.
+
+Sat Mar 14 18:07:23 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* enum.c: [DOC] Fixes Enumerable#member? documentation
+	  [fix GH-756][ci skip] Patch by @shamanime
+
+Sat Mar 14 12:23:53 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): use d_type to reduce lstat system calls.
+
+	* win32/dir.h (struct direct): add d_type instead of d_isdir and
+	  d_isrep.  SYMLINKD is unreliable, since the target can be
+	  replaced after a link was created.
+
+	* win32/win32.c (readdir_internal): set d_type.
+
+Sat Mar 14 02:14:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): empty parentheses at cmdarg can be null.
+	  [ruby-core:68477] [Bug #10957]
+
+Fri Mar 13 15:04:36 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_close): don't raise on double
+	  close for consistent to IO#close.
+
+Fri Mar 13 15:03:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_close_read, rb_io_close_write): don't raise after
+	  close same as IO#close.
+
+Fri Mar 13 12:29:07 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/readline/test_readline.rb: Restore environment variables:
+	  COLUMNS LINES
+
+Fri Mar 13 11:37:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/extconf.rb: needs --enable-shared when linked to
+	  libruby or fiddle.so.  since --with-static-linked-ext does no
+	  longer link extensions to ruby program with --enable-shared, the
+	  only combination needs --enable-static is --disable-shared and
+	  --with-static-linked-ext.  [ruby-dev:48901] [Bug #10960]
+
+Fri Mar 13 07:02:20 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_s_accept_nonblock): use rb_hash_lookup2
+	* ext/openssl/ossl_ssl.c (get_no_exception): new function
+	  (ossl_ssl_accept_nonblock): use get_no_exception
+	  (ossl_ssl_read_internal): ditto
+	  (ossl_ssl_write_nonblock): ditto
+	  [ruby-core:68511]
+
+Fri Mar 13 07:01:38 2015  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_ssl.c: predefine wait_*able symbols
+
+Thu Mar 12 22:59:53 2015  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/leakchecker.rb: Check environment variables.
+
+Thu Mar 12 05:54:27 2015  Eric Wong  <e@80x24.org>
+
+	* ext/socket/init.c (rsock_s_accept_nonblock):
+	  support exception: false
+	  [ruby-core:66385] [Feature #10532]
+	* ext/socket/init.c (rsock_init_socket_init): define new symbols
+	* ext/socket/rubysocket.h: adjust prototype
+	* ext/socket/socket.c (sock_accept_nonblock): support exception: false
+	* ext/openssl/ossl_ssl.c (ossl_ssl_accept_nonblock): ditto
+	* ext/socket/socket.c (Init_socket): adjust accept_nonblock definition
+	* ext/openssl/ossl_ssl.c (Init_ossl_ssl): ditto
+	* ext/socket/tcpserver.c (rsock_init_tcpserver): ditto
+	* ext/socket/unixserver.c (rsock_init_unixserver): ditto
+	* ext/socket/tcpserver.c (tcp_accept_nonblock): adjust
+	  rsock_s_accept_nonblock call
+	* ext/socket/unixserver.c (unix_accept_nonblock): ditto
+	* ext/openssl/ossl_ssl.c (ossl_start_ssl): support no_exception
+	* ext/openssl/ossl_ssl.c (ossl_ssl_connect): adjust ossl_start_ssl call
+	* ext/openssl/ossl_ssl.c (ossl_ssl_connect_nonblock): ditto
+	* ext/openssl/ossl_ssl.c (ossl_ssl_accept): ditto
+	* test/socket/test_nonblock.rb (test_accept_nonblock): test for
+	  "exception :false"
+	* test/socket/test_tcp.rb (test_accept_nonblock): new test
+	* test/socket/test_unix.rb (test_accept_nonblock): ditto
+	* test/openssl/test_pair.rb (test_accept_nonblock_no_exception): ditto
+
+Thu Mar 12 10:53:06 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h, node.h: move a definition of `struct rb_global_entry'
+	  and related functions from node.h to internal.h.
+
+	* variable.c: remove unused include pragma.
+
+	* common.mk: remove unused dependency.
+
+Thu Mar 12 10:32:39 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk: remove unused dependency.
+
+Thu Mar 12 08:20:14 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* load.c: removed unused header file "node.h".
+
+	* method.h: ditto.
+
+	* symbol.c: ditto.
+
+Thu Mar 12 08:14:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: RGENGC_CHECK_MODE should be 0.
+
+Thu Mar 12 07:44:17 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: use T_IMEMO to represent `struct MEMO' value.
+	  memo->v1 and memo->v2 is WB protected values.
+	  So use MEMO_V1/V2_SET() macros to set these values.
+	  memo->u3 is ambiguous (sometimes a VALUE, sometimes an integer
+	  value), so use gc_mark_maybe() in gc.c to mark it.
+
+	  Rename NEW_MEMO() to MEMO_NEW().
+
+	  Move MEMO_FOR and NEW_MEMO_FOR macros from node.h.
+
+	  Export a rb_imemo_new() function for ext/ripper.
+
+	* node.h: remove NODE_MEMO.
+
+	* enum.c: catch up these change.
+
+	* enumerator.c: ditto.
+
+	* load.c: ditto.
+
+	* ext/objspace/objspace.c (count_nodes): ditto.
+
+	* gc.c (gc_mark_children): mark imemo_memo type.
+
+	* parse.y (new_args_gen): use T_IMEMO.
+	  (I'm not sure it is working correctly...)
+
+Wed Mar 11 22:36:34 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (frame_called_id): it should use vm_ifunc type.
+
+	* eval.c (frame_func_id): ditto.
+
+Wed Mar 11 22:27:05 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* node.h: remove NODE_IFUNC, NEW_IFUNC.
+
+	* internal.h: use T_IMEMO for IFUNC.
+
+	  rename `struct IFUNC' to `struct vm_ifunc' and move the definition
+	  from vm_insnhelper.h. Add imemo_ifunc.
+
+	* gc.c (gc_mark_children): mark imemo_ifunc type T_IMEMO object.
+
+	* compile.c: catch up these changes.
+
+	* proc.c: ditto.
+
+	* vm_core.h (RUBY_VM_IFUNC_P): ditto.
+
+	* vm_eval.c (rb_iterate): ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* ext/objspace/objspace.c: ditto.
+
+Wed Mar 11 21:53:43 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h, eval_intern.h: move CREF accessors.
+
+	  List IMEMO supported types in internal.h.
+
+Wed Mar 11 21:45:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h: use T_IMEMO to create THROW_DATA.
+
+	  Add THROW_DATA_NEW().
+
+	* internal.h: move definition of `struct THROW_DATA'
+	  from vm_insnhelper.h to internal.h.
+
+	  Rename `THROW_DATA' to `vm_throw_data'.
+
+	* eval_intern.h (THROW_DATA_P): move to internal.h.
+	  THROW_DATA is no longer T_NODE, so check T_IMEMO.
+
+	* gc.c (gc_mark_children): mark THROW_DATA.
+
+	* vm.c: catch up these changes.
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+Wed Mar 11 21:21:56 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: use T_IMEMO to create SVAR.
+
+	* internal.h, vm_insnhelper.h: move definition `struct SVAR'
+	  from vm_insnhelper.h to internal.h. And rename it to struct vm_svar.
+
+	  new imemo_type imemo_svar is added.
+
+	* gc.c (gc_mark_children): mark imemo_svar.
+
+	* node.c (rb_gc_mark_node): remove useless marking.
+
+Wed Mar 11 19:35:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* include/ruby/ruby.h: introduce new type T_IMEMO.
+	  T_IMEMO is Internal Memo type, internal use only.
+	  T_IMEMO has same purpose of NODE_MEMO.
+
+	  To insert T_IMEMO, type numbers are modified a little.
+
+	* internal.h: define struct RIMemo. Each RIMemo objects
+	  has imemo_type. We can observe it by the imemo_type() function.
+
+	* gc.c (rb_imemo_new): added.
+
+	* node.h: remove NODE_CREF and NEW_CREF().
+
+	* node.c (rb_gc_mark_node): ditto.
+
+	* vm.c (vm_cref_new): use rb_imem_new().
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: use RIMemo objects for CREF.
+
+	* ext/objspace/objspace.c: support T_IMEMO.
+
+Wed Mar 11 17:03:20 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: fix memory leak by prepend method.
+
+	  It is easy to reproduce with such script:
+
+	    module M; def bar; end; end
+	    loop{
+	      Class.new do
+	        def foo; end
+	        prepend M
+	      end
+	    }
+
+	* gc.c (obj_free): free T_ICLASS::m_tbl if it is created by prepend.
+	  To recognize it, check RICLASS_IS_ORIGIN flag.
+
+	* gc.c (gc_mark_children): T_ICLASS objects only need to mark
+	  T_ICLASS::m_tbl if RICLASS_IS_ORIGIN is set.
+
+	* gc.c (obj_memsize_of): count T_ICLASS if RICLASS_IS_ORIGIN is set.
+
+	* internal.h (RCLASS_SET_ORIGIN): add to set RCLASS_SET_ORIGIN.
+
+	  TODO: The word `origin' seems not good name. We need to invent
+	  another good name.
+
+	* class.c: use RCLASS_SET_ORIGIN().
+
+	* class.c (class_alloc): zero clear rb_classext_t.
+
+Wed Mar 11 13:28:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check also procstat_getvmmap, which is not
+	  available on FreeBSD 9.  [ruby-core:68468] [Bug #10954]
+
+	* vm_dump.c (procstat_vm): use kinfo_getvmmap instead if
+	  procstat_getvmmap is not available.
+
+Wed Mar 11 09:15:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: define struct MEMO.
+
+	* enum.c: use MEMO.
+
+	* enumerator.c: ditto.
+
+	* load.c: ditto.
+
+	* node.h: return (struct MEMO *) pointer.
+
+Wed Mar 11 06:26:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h (THROW_DATA_STATE): return int, not VALUE.
+
+	* vm_insnhelper.h (THROW_DATA_STATE_SET): accept int value.
+
+Wed Mar 11 05:06:46 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_eval.c (rb_catch_protect): use THROW_DATA_VAL().
+
+Wed Mar 11 04:56:04 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h: define struct IFUNC.
+
+	* vm_eval.c (rb_iterate): use it.
+
+	* vm_insnhelper.c (vm_yield_with_cfunc): ditto.
+
+Wed Mar 11 03:52:12 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h (THROW_DATA_P): use RB_TYPE_P() instead of
+	  BUILTIN_TYPE().
+
+	* thread.c (thread_join): use THROW_DATA_P().
+
+Wed Mar 11 03:48:01 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c: use RUBY_VM_IFUNC_P() to recognize IFUNC or not.
+
+	* vm.c: ditto.
+
+	* vm_dump.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+	* vm_core.h (RUBY_VM_IFUNC_P): use RB_TYPE_P() instead of
+	  BUILTIN_TYPE().
+
+Wed Mar 11 03:21:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h: define struct THROW_DATA to represent
+	  throwing data. Also define accessor functions.
+
+	* eval_intern.h: move related changes into vm_insnhelper.h.
+	  Now these MACROs (functions) are only used in vm*.c.
+
+	  There is only THROW_DATA_P(err) to check this data type or not.
+
+	* vm.c: catch up these changes.
+
+	* vm_eval.c: ditto.
+
+	* vm_insnhelper.c: ditto.
+
+Wed Mar 11 00:57:00 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/rubygems/test_gem_security_trust_dir.rb: The return value of
+	  File::Stat#mode is OS dependent. In AIX, 0200000 is set.
+
+Tue Mar 10 20:03:41 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/webrick/server.rb: Invoke setup_shutdown_pipe in start method
+	  instead of listen method.
+	  [ruby-core:68476] [Bug #10956] Reported by Shintaro Kojima.
+
+Tue Mar 10 17:27:27 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c (thread_join): Fixnum (except TAG_FATAL) and
+	  NODE should not be reached here.
+
+Mon Mar  9 21:42:10 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (ep_cref): rename to lep_cref() because it should be
+	  local ep.
+
+Mon Mar  9 16:34:36 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_close): don't raise on double
+	  close for consistent to IO#close.
+
+Mon Mar 09 06:44:48 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h: define struct SVAR for SVAR.
+	  This data type is also same layout of NODE (NODE_IF).
+
+	* vm_insnhelper.c: catch up this change.
+
+Mon Mar  9 06:43:21 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (lep_svar_set): add WBs.
+
+Mon Mar  9 06:19:06 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: define rb_cref_t and change to use it.
+
+	  rb_cref_t is data type of CREF. Now, the body is still NODE.
+	  It is easy to understand what is CREF and what is pure NODE.
+
+Mon Mar  9 06:00:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h (COPY_CREF_OMOD): fix translation miss.
+
+Mon Mar  9 04:47:58 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: define CREF accessor macros.
+	  * CREF_CLASS(cref)
+	  * CREF_NEXT(cref)
+	  * CREF_VISI(cref)
+	  * CREF_VISI_SET(cref, v)
+	  * CREF_REFINEMENTS(cref)
+	  * CREF_PUSHED_BY_EVAL(cref)
+	  * CREF_PUSHED_BY_EVAL_SET(cref)
+	  * CREF_OMOD_SHARED(cref)
+	  * CREF_OMOD_SHARED_SET(cref)
+	  * CREF_OMOD_SHARED_UNSET(cref)
+
+	  This is process to change CREF data type from NODE.
+
+Sun Mar  8 22:50:57 2015  Tanaka Akira  <akr@fsij.org>
+
+	* ext/zlib/zlib.c (rb_gzfile_close): Don't raise on double
+	  close for consistent to IO#close.
+
+Sun Mar  8 16:57:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): match patterns against legacy short names
+	  too, not only ordinary names.  [ruby-core:67954] [Bug #10819]
+
+	* win32/dir.h (struct direct): add short name members.
+
+	* win32/win32.c (opendir_internal, readdir_internal): ditto.
+
+Sat Mar  7 09:36:05 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* array.c: document that first element is kept when using
+	  Array#uniq and #uniq! [fix GH-845][ci skip]
+	  Patch by @riffraff
+
+Sat Mar  7 09:28:02 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* thread.c: Mutex#owned? is no longer experimental since 2.1.0
+	  [fix GH-839][ci skip] Patch by @takiy33
+
+Sat Mar  7 09:18:42 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/merger.rb: Added documentation to version method.
+	  [fix GH-847][ci skip] Patch by @magikid
+
+Fri Mar  6 22:50:36 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (rb_prepend_module): need a WB for klass -> origin.
+
+Fri Mar  6 20:18:38 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* fix namespace issue on singleton class expressions. [Bug #10943]
+
+	* vm_core.h, method.h: remove rb_iseq_t::cref_stack. CREF is stored
+	  to rb_method_definition_t::body.iseq_body.cref.
+
+	* vm_insnhelper.c: modify SVAR usage.
+	  When calling ISEQ type method, push CREF information onto method
+	  frame, SVAR located place. Before this fix, SVAR is simply nil.
+	  After this patch, CREF (or NULL == Qfalse for not iseq methods)
+	  is stored at the method invocation.
+
+	  When SVAR is required, then put NODE_IF onto SVAR location,
+	  and NDOE_IF::nd_reserved points CREF itself.
+
+	* vm.c (vm_cref_new, vm_cref_dump, vm_cref_new_toplevel): added.
+
+	* vm_insnhelper.c (vm_push_frame): accept CREF.
+
+	* method.h, vm_method.c (rb_add_method_iseq): added. This function
+	  accepts iseq and CREF.
+
+	* class.c (clone_method): use rb_add_method_iseq().
+
+	* gc.c (mark_method_entry): mark method_entry::body.iseq_body.cref.
+
+	* iseq.c: remove CREF related codes.
+
+	* insns.def (getinlinecache/setinlinecache): CREF should be cache key
+	  because a different CREF has a different namespace.
+
+	* node.c (rb_gc_mark_node): mark NODE_IF::nd_reserved for SVAR.
+
+	* proc.c: catch up changes.
+
+	* struct.c: ditto.
+
+	* insns.def: ditto.
+
+	* vm_args.c (raise_argument_error): ditto.
+
+	* vm_eval.c: ditto.
+
+	* test/ruby/test_class.rb: add a test.
+
+Fri Mar  6 18:19:13 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/webrick/test_filehandler.rb: on vboxsf (on VirtualBox
+	  on Windows 7), file name and permissions are strange (can access
+	  by short file name and so on).
+
+	  Simply skip on such tests on such FS. To detect strange FS, this
+	  patch use a part of code `File.executable?(__FILE__)`.
+	  Please correct them if there are better ways.
+
+Fri Mar  6 17:31:29 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_beginendblock.rb: do not change directory.
+
+	  Run system command in the directory mounted by vboxsf on Windows 7
+	  and get warning like that "warning: Insecure world writable dir...".
+
+Fri Mar  6 10:31:00 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call_super): search next super class from the
+	  original class, to get rid of infinite recursion with
+	  prepending.  a patch by Seiei Higa <hanachin AT gmail.com> at
+	  [ruby-core:68434].  [ruby-core:68093] [Bug #10847]
+
+Fri Mar  6 08:45:26 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/matrix.rb: Add Vector#round. Patch by Jordan Stephens.
+	  [Fixes GH-802]
+
+Fri Mar  6 07:33:03 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (obj_info): show node name too.
+
+Fri Mar  6 07:00:44 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* internal.h: remove struct method_table_wrapper.
+	  struct method_table_wrapper was introduced to avoid duplicate marking
+	  for method tables.
+
+	  For example, `module M1; def foo; end; end` make one method table
+	  (mtbl) contains a method `foo`. M1 (T_MODULE) points mtbl.
+	  Classes C1 and C2 includes M1, then two T_ICLASS objects are created
+	  and they points mtbl too. In this case, three objects (one T_MODULE
+	  and two T_ICLASS objects) points same mtbl. On marking phase, these
+	  three objects mark same mtbl. To avoid such duplication, struct
+	  method_table_wrapper was introduced.
+
+	  However, created two T_ICLASS objects have same or shorter lifetime
+	  than M1 (T_MODULE) object. So that we only need to mark mtbl from M1,
+	  not from T_ICLASS objects. This patch tries marking only from M1.
+	  In other words, original module (M1) has responsibility to mark mtbl.
+	  Because of no duplicate marking, we don't need method_table_wrapper
+	  any more.
+
+	  Note that one `Module#prepend` call creates two T_ICLASS objects.
+	  One for referring to a prepending Module object, same as
+	  `Module#include`. We don't need to care this T_ICLASS.
+	  One for moving original mtbl from a prepending class. We need to
+	  mark such mtbl from this T_ICLASS object. To mark the mtbl,
+	  we need to use `RCLASS_ORIGIN(klass)` on marking from a prepended
+	  class `klass`.
+
+	* class.c: ditto.
+
+	* eval.c (rb_using_refinement): ditto.
+
+	* gc.c: ditto.
+
+	* include/ruby/ruby.h: define m_tbl directly. The definition of
+	  struct RClass should be moved to (srcdir)/internal.h.
+
+	* method.h: remove decl of rb_free_m_tbl_wrapper().
+
+	* object.c: use RCLASS_M_TBL() directly.
+
+Fri Mar  6 02:50:12 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* dir.c (replace_real_basename): need to check the return value of
+	  GLOB_REALLOC().
+
+Fri Mar  6 02:26:03 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* dir.c (replace_real_basename): shouldn't create Ruby object before
+	  the object system is loaded.
+	  [ruby-core:68430] [Bug #10941]
+
+Wed Mar  5 16:58:43 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* hash.c: [DOC] #delete method actually returns nil, if the key
+	  is not found. [fix GH-844][ci skip] Patch by @ivdma
+
+Wed Mar  5 12:22:23 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* math.c: refactoring: remove unnecessary variable d0 to unify code
+	  appearance.
+
+Thu Mar  5 11:50:54 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_eval.c (eval_string_with_cref): A binding should keep
+	  refinements activation information and the refinements should be
+	  activated in subsequent eval calls with the binding.
+	  [ruby-core:67945] [Bug #10818]
+
+Thu Mar  5 11:16:55 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_refinement.rb: There is no need anymore to suppress
+	  warnings.
+
+Thu Mar  5 08:31:02 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* random.c (random_raw_seed): Avoid calling fill_random_bytes()
+	  if the requested size is 0. AIX returns -1 for 0-byte read from
+	  /dev/urandom, while other UNIX returns 0. With this change,
+	  Random.raw_seed(0) consistently returns "" in any UNIX.
+
+Wed Mar  4 12:43:32 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* test/ruby/test_math.rb (assert_float_and_int): Refactor test cases
+	  by introducing assert_float_and_int. [misc #10810]
+
+Wed Mar  4 11:52:30 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.c (Init_sym): make dsym_fstrs a hash compared by identity
+	  as the keys are unique fstrings, to get rid of running hash and
+	  compare methods and causing new object allocation during garbage
+	  collection phase.  [ruby-dev:48891] [Bug #10933]
+
+Wed Mar  4 10:16:57 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* enum.c: Fix typo in slice_after's exception message.
+	  [fix GH-842][ci skip] Patch by @jsyeo
+
+Wed Mar  4 10:15:37 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/syntax/methods.rdoc: add some missing spaces and
+	  fix a grammatical error in method docs.
+	  [fix GH-843][ci skip] Patch by @nikolas
+
+Wed Mar  4 02:13:06 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (backport_command_string): pick up only
+	  when the revision exists in trunk.
+
+Wed Mar  4 00:44:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_io_blocking_region): assigned variables
+	  inside EXEC_TAG() should be volatile.
+
+	* thread.c (rb_thread_s_handle_interrupt): ditto.
+
+	* thread.c (exec_recursive): ditto.
+
+Wed Mar  4 00:29:18 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: now can specify shorten form of commands.
+
+Tue Mar  3 23:41:42 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (Readline.readline): drop untreated control
+	  characters.
+
+Tue Mar  3 22:25:24 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/lib/envutil.rb (EnvUtil.invoke_ruby): need to rescue because
+	  Signal.signame may raise exception.
+
+Tue Mar  3 16:57:39 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: use 'b' instead of 's' for showing
+	  Backport options for merger.rb.
+
+Tue Mar  3 16:55:23 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: show selected ticket number at prompt.
+
+Tue Mar  3 14:47:30 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* math.c (num2dbl_with_to_f): direct casting from Rational to double.
+	  [Feature #10909]
+
+	* test/ruby/test_math.rb: add tests for the above change.
+
+Tue Mar  3 07:52:20 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/ruby/test_symbol.rb: avoid a false positive in AIX.
+
+Tue Mar  3 00:59:39 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* configure.in: set PRELOADENV in Solaris to avoid "wrong ELF class"
+	  error. [Bug #10926] [ruby-dev:48888]
+	* configure.in: set LIBPATHENV for 32-bit compile in Solaris
+	  in addition to 64-bit.
+
+Mon Mar  2 15:36:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: do not check _setjmp unless _longjmp is available,
+	  so that configure results will not be changed by cache.
+
+Mon Mar  2 14:44:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_REPLACE_TYPE): restore unsigned type from
+	  cached variable only if the target type is not available.
+
+Mon Mar  2 13:04:27 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (sig_signame): return nil if the argument is a valid
+	  signal number.
+
+Mon Mar  2 12:05:04 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/net/ftp/test_ftp.rb (create_ftp_server): set SO_OOBINLINE
+	  for receiving OOB data which is sent with MSG_OOB flag in
+	  portable way. [Bug #10915] [ruby-dev:48885]
+	* test/net/ftp/test_ftp.rb (test_abort, test_status): use gets
+	  for receiving OOB data in portable way.
+
+Mon Mar  2 11:43:07 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_REPLACE_TYPE): restore unsigned type from
+	  cached variable.
+
+Mon Mar  2 06:01:41 2015  Eric Wong  <e@80x24.org>
+
+	* ext/io/wait/wait.c (io_nread): wrap return value with INT2FIX
+	  Thanks to Yura Sokolov <funny.falcon@gmail.com>
+	  [ruby-core:68369] [Bug#10923]
+	* test/io/wait/test_io_wait.rb (test_nread_buffered):
+	  fix broken test
+
+Sun Mar  1 20:21:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_REPLACE_TYPE): restore convertible type from
+	  cached variable, so that configured results will be stable.
+
+Sun Mar  1 18:10:34 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_broken_memmem): check before adding the
+	  result HAVE_MEMMEM macro to confdefs.h, so that configured
+	  results will be stable.
+
+Sun Mar  1 11:17:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (id2ref): prohibit from accessing internal objects.
+	  [ruby-core:68348] [Bug #10918]
+
+Sun Mar  1 09:06:11 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (strptime): Support %s.%N.
+	  [ruby-core:68301] [Bug #10904] Patch by Sadayuki Furuhashi.
+
+Sat Feb 28 17:18:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enum.c (enum_each_slice, enum_each_cons): limit elements size by
+	  the enumerator size.  suggested by Hans Mackowiak <hanmac AT
+	  gmx.de> at [ruby-core:68335]
+
+Sat Feb 28 15:44:20 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): get rid of making new strings
+	  inside signal context.
+
+	* variable.c (rb_tmp_class_path): defer making temporary class
+	  path string.
+
+	* variable.c (rb_search_class_path): search class path or return
+	  Qnil or Qfalse if unnamed, not creating a temporary path.
+
+Sat Feb 28 15:02:02 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_tmp_class_path): preserve name encoding of an
+	  anonymous instance of module/class subclass.
+
+Sat Feb 28 08:24:30 2015  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* ext/pty/pty.c: AIX supports autopush.
+	  Patch by Perry Smith [ruby-core:58539] [Bug #9144]
+
+Fri Feb 27 22:00:05 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems: Update to RubyGems 2.4.6 and HEAD(800f2e6).
+	  Fixed #1159, #1171, #1173 on rubygems/rubygems
+	* test/rubygems: ditto.
+
+Fri Feb 27 20:55:42 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rake: Update to rake (9237e74), typo fix and remove needless
+	  private syntax.
+	* test/rake: ditto.
+
+Fri Feb 27 17:06:44 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: define vm_svar_index.
+
+	* vm_insnhelper.c, vm.c, compile.c: use vm_svar_index names.
+
+	* iseq.h: remove DEFAULT_SPECIAL_VAR_COUNT.
+	  use VM_SVAR_FLIPFLOP_START instead.
+
+Fri Feb 27 13:57:48 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (setup_narg): wipe away expanded part of buffer to get rid
+	  of revealing uncleaned data.  reported by Dongkwan Kim <dkay AT
+	  kaist.ac.kr>.
+
+Wed Feb 25 22:25:07 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* spec/default.mspec: use default configuration file name.
+	  https://github.com/ruby/rubyspec/commit/cc69f337b06362e5607ffa3e3ad40ef7494960cf
+
+Wed Feb 25 22:21:56 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* spec/default.mspec: remove specific version number.
+	  https://github.com/ruby/rubyspec/commit/7a909e925c1baa9c700bd44af9241aef6e596714
+
+Wed Feb 25 22:04:04 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/win32/Win32API.rb (initialize): accept both a string and an array
+	  for the arguments of the imported function.
+	  reported by Aaron Stone [ruby-core:68208] [Bug #10876] [Fixes GH-835]
+
+Wed Feb 25 18:12:11 2015  Eric Wong  <e@80x24.org>
+
+	* signal.c (sighandler): preserve errno
+	  Patch by Steven Stewart-Gallus <sstewartgallus00@mylangara.bc.ca>
+	  [ruby-core:68172] [Bug #10866]
+
+Wed Feb 25 15:59:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (push_pattern, push_glob): make globbed file names same
+	  encoding to the given pattern.
+
+Wed Feb 25 15:27:16 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: support 2.1+ versioning scheme.
+
+Tue Feb 25 08:49:12 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* lib/cmath.rb (log): raise ArgumentError when more than 2 arguments
+	  are passed. [ruby-core:66143] [Bug #10487]
+
+Tue Feb 25 02:15:17 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* test/ruby/test_math.rb: Use assert_infinity instead of assert_equal(1.0/0, ...).
+
+	* test/ruby/test_math.rb: Add tests for overriding Integer#to_f.
+	  [ruby-core:67919] [Misc #10809]
+
+Tue Feb 24 22:58:48 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (nucomp_mul): calculate as rotation in complex plane
+	  if matrix calculation resulted in NaN.
+
+Tue Feb 24 21:45:39 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* test/ruby/test_math.rb(test_cbrt): Add an assertion for Math.cbrt(1.0/0)
+	  and move #test_cbrt to more proper place.
+
+Tue Feb 24 19:09:25 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c (lep_svar_place, lep_svar_get): do not create
+	  additional T_NODE object (svars holder) when only getting
+	  svars.
+
+Tue Feb 24 11:49:48 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_zone_name): should be US-ASCII only if all 7-bits,
+	  otherwise locale encoding.  [ruby-core:68230] [Bug #10887]
+
+Tue Feb 24 09:47:07 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (chompped_length): enable smart chomp for all non-dummy
+	  encoding strings, not only default_rs.
+	  [ruby-core:68258] [Bug #10893]
+
+Mon Feb 23 23:19:42 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/vcs.rb (IO.popen): support :chdir option.
+
+	* tool/vcs.rb (VCS::GIT.get_revisions): use :chdir option instead
+	  of -C option which is not supported by older git.
+	  [ruby-dev:48880] [Bug #10890]
+
+Mon Feb 23 15:26:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_split_m): raise ArgumentError at broken string
+	  not RegexpError, as Regexp is not involved in.
+	  [ruby-core:68229] [Bug #10886]
+
+Mon Feb 23 07:25:29 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* time.c: Zone encoding should be US-ASCII if all 7-bits. Fix r46907.
+
+	* test/ruby/test_time.rb, test/ruby/test_time_tz.rb: Update tests.
+
+Sun Feb 22 18:33:42 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* common.mk: use ruby organization for rubyspec.
+
+Sun Feb 22 15:56:06 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_insnhelper.c (rb_vm_rewrite_cref_stack): copy nd_refinements
+	  of original crefs. It fixes segmentation fault when calling
+	  refined method in duplicate module. [ruby-dev:48878] [Bug #10885]
+
+	* vm_core.h, class.c: change accordingly.
+
+	* test/ruby/test_refinement.rb: add a test for above.
+
+Sun Feb 22 10:43:37 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_objspace_call_finalizer): control GC execution during
+	  force firnalizations at the end of interpreter process.
+	  [Bug #10768]
+
+	  1) Prohibit incremental GC while running Ruby-level finalizers
+	     to avoid any danger.
+	  2) Prohibit GC while invoking T_DATA/T_FILE data structure
+	     because these operations break object relations consistency.
+
+	  This patch can introduce another memory consuming issue because
+	  Ruby-level finalizers can run after (2), GC is disabled.
+	  However, basically object consistency was broken at (2) as I
+	  described above. So that running Ruby-level finalizers contains
+	  danger originally. Because of this point, I need to suggest to
+	  remove these 3 lines (invoking remaining finalizers). And add a
+	  rule to add that finalizers should not add new finalizers, or
+	  say there is no guarantee to invoke finalizers that added by
+	  another finalizer.
+
+Sun Feb 22 04:07:05 2015  Zachary Scott  <e@zzak.io>
+
+	* ext/openssl/ossl_asn1.c: [DOC] RDoc formatting fixes for
+	  ASN1::ObjectId with patch from @vbatts [Fixes GH-834]
+
+	* ext/openssl/ossl_bn.c: ditto.
+
+Sat Feb 21 19:51:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (match_aref): RMatch::regexp is Qnil after matching by a
+	  string since r45451.  [ruby-core:68209] [Bug #10877]
+
+Sat Feb 21 16:18:42 2015  Stefan Schuler  <mail@stefanschuessler.de>
+
+	* compar.c (Init_Comparable): [DOC] Replace camelcase variable name.
+	  [Fix GH-833]
+
+Fri Feb 20 17:27:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_file_identical_p): fix handle leak, ensure to close
+	  the handle of the first argument.
+
+Fri Feb 20 17:19:23 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (different_device_p): compare by volume serial
+	  numbers, not by path names.  [ruby-core:68162] [Bug #10865]
+
+Thu Feb 19 01:58:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (rb_file_expand_path_internal): neither the drive
+	  of base directory nor the current drive are involved in the
+	  result if different than the drive of path.
+	  [ruby-core:68130] [Bug #10858]
+
+Wed Feb 18 10:48:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (wrename): return EXDEV if moving a directory to
+	  another drive, since MoveFileExW does not set proper error code.
+	  [ruby-core:68162] [Bug #10865]
+
+Wed Feb 18 03:13:52 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: bump psych version.
+	* ext/psych/psych.gemspec: ditto
+	* ext/psych/yaml/scanner.c: add latest libyaml change.
+	* test/psych/helper.rb: support newer minitest
+	* test/psych/test_to_yaml_properties.rb: ditto
+
+Tue Feb 17 11:47:17 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::DNS::Resource#==, #hash): elements
+	  returned by Kernel#instance_variables are Symbols now.
+	  [ruby-core:68128] [Bug #10857]
+
+Tue Feb 17 10:59:10 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/syntax/calling_methods.rdoc: Fix documentation for "calling_methods"
+	  Patch by @sos4nt [fix GH-830][ci skip]
+
+Tue Feb 17 10:53:29 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (rb_file_expand_path_internal): do not make invalid
+	  (or ADS) path if the path has a drive letter, the result also
+	  should have be under it.  [ruby-core:68130] [Bug #10858]
+
+Tue Feb 17 10:47:20 2015  Iain Beeston  <iain.beeston@gmail.com>
+
+	* hash.c: Added docs to explain that #include? and #member? do not
+	  check member equality
+	* lib/set.rb: ditto
+
+Mon Feb 16 20:58:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_massign): optimization for special case,
+	  assignments by aset or attrset.
+	  http://kokizzu.blogspot.jp/2015/02/c-java-hhvm-ruby-nodejsrhinojscspidermo.html
+	  http://www.atdot.net/~ko1/diary/201502.html#d16
+
+Sun Feb 15 10:41:23 2015  Sho Hashimoto  <sho-h@ruby-lang.org>
+
+	* doc/standard_library.rdoc: [DOC] delete removed libraries.
+	  [misc #10843] [ci skip]
+
+Sat Feb 14 12:20:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (rand_random_number): add a method to return a random
+	  number like SecureRandom to Random::Formatter.
+
+	* lib/securerandom.rb (random_bytes): move to Random::Formatter,
+	  the base method of the module.
+
+Sat Feb 14 12:01:32 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (random_raw_seed): extract platform dependent random
+	  seed initialization function as a new method Random.raw_seed.
+
+	* lib/securerandom.rb (SecureRandom): use Random.raw_seed.
+
+Sat Feb 14 00:49:37 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/coverage/coverage.c: Add Coverage.peek_result.  Allows you to
+	  capture coverage information without stopping the coverage tool.
+	  [ruby-core:67940] [Feature #10816]
+
+	* test/coverage/test_coverage.rb: test for change.
+
+Fri Feb 13 21:52:05 2015  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* string.c (str_discard): does not free for STR_NOFREE string.
+	  [Bug #10853][ruby-core:68110]
+
+	* bootstraptest/test_string.rb: test for above.
+
+Fri Feb 13 21:16:00 2015  Yusuke Endoh  <mame@tsg.ne.jp>
+
+	* lib/base64.rb: make urlsafe mode user-friendly.
+
+	* lib/base64.rb (Base64.urlsafe_encode64): a new option "padding" to
+	  suppress the padding character ("=").
+
+	* lib/base64.rb (Base64.urlsafe_decode64): now it accepts not only
+	  correctly-padded input but also unpadded input.
+	  [Feature #10740][ruby-core:67570]
+
+	* test/base64/test_base64.rb: Test for above
+
+Fri Feb 13 14:19:06 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/json: merge upstream from flori/json
+	  change usage of TypedData. [Feature #10739][ruby-core:67564]
+
+Thu Feb 12 18:34:01 2015  multisnow  <infinity.blick.winkel@gmail.com>
+
+	* ext/openssl/extconf.rb: check RAND_edg to support libressl.
+
+	* ext/openssl/ossl_rand.c (ossl_rand_egd): define only if RAND_edg
+	  is available.  [Fix GH-829]
+
+Thu Feb 12 10:46:14 2015  Eric Hodel  <drbrain@segment7.net>
+
+	* proc.c (proc_call):  Improve Proc#call documentation.  Patch by
+	  Hsing-Hui Hsu.  [fix GH-761]
+
+Thu Feb 12 04:33:02 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* compar.c (cmp_equal): no more error hiding for Comparable#==.
+	  It now behaves as other Comparable methods. See #7688.
+
+	* test/ruby/test_comparable.rb: update related test.
+
+Thu Feb 12 03:28:05 2015  Eric Wong  <e@80x24.org>
+
+	* lib/set.rb (initialize): internal hash defaults to false
+
+	* lib/set.rb (include?): use Hash#[] for optimized dispatch.
+	  Patch by Ismael Abreu <ismaelga@gmail.com>
+	  [ruby-core:67664] [Misc #10754]
+
+Wed Feb 11 11:09:52 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/digest/digest_conf.rb (digest_conf): check for CommonDigest.
+
+	* ext/digest/*/*cc.h: for Apple CommonCrypto/CommonDigest.h.
+
+	* ext/digest/digest.h (DEFINE_FINISH_FUNC_FROM_FINAL): macro for
+	  finish functions, by inverting arguments order.
+
+	* ext/digest/digest_conf.rb (digest_conf): extract common
+	  configurations.
+
+Wed Feb 11 11:01:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/generator/generator.c (generate_json): get rid of
+	  unnecessary recursive calls which can cause infinite recursion.
+	  T_STRING may not have rb_cString.
+
+Wed Feb 11 07:53:35 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_word.rb: use skip method to skip test.
+
+Tue Feb 10 11:38:28 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_method): stop method search when a method
+	  is not found in a refinement, to support undef in refinements.
+	  [ruby-core:66741] [Bug #10578]
+
+Tue Feb 10 11:19:11 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (chdir, delete, gettextfile, mdtm, mkdir, nlst,
+	  putbinaryfile, puttextfile, rename, rmdir, size): support
+	  Pathname. Patch by Joe Rafaniello. [fix GH-828]
+
+Mon Feb  9 16:36:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): get rid of loading unbundled and
+	  unexpected libraries.  [ruby-core:67977] [Bug #10822]
+
+Sun Feb  8 20:09:37 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/net/http/header.rb: pass header names as symbols.
+	  Patch by @DamirSvrtan [fix GH-805]
+	* test/net/http/test_httpheader.rb: added test.
+
+Sun Feb  8 13:04:25 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/getaddrinfo.c (get_addr): reject too long hostname to
+	  get rid of GHOST vulnerability on very old platforms.
+
+	* ext/socket/raddrinfo.c (make_hostent_internal): ditto, paranoic
+	  check for the canonical name.
+
+Sun Feb  8 12:48:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/win32/registry.rb (Win32::Registry::API): use wide
+	  versions of RegDeleteValue and RegDeleteKey.
+	  [ruby-core:67958] [Bug #10820]
+
+Sat Feb  7 22:13:08 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_record.rb: remove test using .NET
+	  Framework 3.5 because it is not included in Windows 8/8.1.
+
+Sat Feb  7 19:25:25 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (has_magic): always get long path name on Windows even if
+	  no tilde is there.  [ruby-core:68011] [Bug #10819]
+
+	* dir.c (replace_real_basename): FindFirstFile ignore redirection
+	  character, check if exists before call it.  cf. [Bug #8597]
+
+Sat Feb  7 13:30:11 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* test/win32ole/test_win32ole_record.rb
+	  (test_ole_instance_variable_get): correct VT_RECORD type and
+	  instance variables.
+
+Fri Feb  6 17:47:05 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: register nodes when
+	  dumping objects with custom coders. [ruby-core:66215] [Bug #10496]
+
+	* test/psych/test_coder.rb: test for fix
+
+Fri Feb  6 16:58:31 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: fix support for regular
+	  expressions with newlines. tenderlove/psych#222
+
+	* test/psych/test_yaml.rb: test for change.
+
+Fri Feb  6 10:31:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_call_info_kw_arg_struct): make keywords a symbols
+	  list to get rid of inadvertent creation by variable keyword
+	  arguments.  [ruby-core:68031] [Bug #10831]
+
+Thu Feb  5 22:42:34 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems:  Update to RubyGems HEAD(5c3b6f3).
+	  Fixed #1156, #1142, #1115, #1142, #1139 on rubygems/rubygems
+	* test/rubygems:  ditto.
+
+Thu Feb  5 13:41:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (send_internal), vm_insnhelper.c (vm_call_opt_send):
+	  convert String method name into a Symbol, as method_missing
+	  method expects its first argument to be a Symbol.  [Bug #10828]
+
+	* vm_insnhelper.c (ci_missing_reason): return the reason of method
+	  missing in call info.
+
+	* vm_insnhelper.c (vm_call_opt_send): re-apply r49500 with the
+	  proper missing reason.  [Bug #10828]
+
+Thu Feb  5 10:31:46 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* class.c (rb_obj_singleton_methods): should use RTEST() to convert
+	  VALUE to int.
+
+Thu Feb  5 03:59:33 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* vm_insnhelper.c: Fix symbol leak with +send+ [Bug #10828]
+
+Wed Feb  4 20:26:54 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (Init_win32ole): should not use atexit to
+	  free allocated hash table to avoid error on Cygwin.
+
+Wed Feb  4 15:34:25 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* class.c (method_entry_i, class_instance_method_list,
+	  rb_obj_singleton_methods): should not include methods of
+	  superclasses if recur is false. [ruby-dev:48854] [Bug #10826]
+
+Wed Feb  4 16:32:40 2015  Matt Hoyle  <matt@deployable.co>
+
+	* ext/bigdecimal/bigdecimal.c (VpSetPTR): fix a typo, 'expoennt'
+	  to 'exponent'.  [ruby-core:67980] [Bug #10823] [Fix GH-825]
+
+Wed Feb  4 15:55:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/sdbm/_sdbm.c: include ruby/ruby.h for PRIdPTRDIFF when a
+	  macro `DEBUG` is defined.  based on the patch by Owen Rodley in
+	  [ruby-core:67987]. [Bug #10825]
+
+Wed Feb  4 11:12:43 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (probes.stamp): rebuild dtrace dependent objects
+	  only when `dtrace -G` modifies its input files.
+
+Tue Feb  3 19:27:16 2015  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* common.mk (ruby-glommed.o): dependency on $(OBJ) should be written
+	  in common.mk (in which OBJ is defined) because of Makefile include
+	  and parse order. This partly reverts r49419.
+	  [ruby-dev:48849] [Bug #10808]
+
+	* Makefile.in (ruby-glommed.o): ditto.
+
+	* Makefile.in (ruby-glommed.o): remove excess $(DTRACE_OBJ) because
+	  it is included in $(OBJS) since r49451.
+
+	* Makefile.in (probes.o): should depend on $(DTRACE_DEPENDENT_OBJS)
+
+Tue Feb  3 17:15:45 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_cppflags, try_cflags, try_ldflags): get rid of
+	  interference by modifying global variables in have_devel? method.
+	  [ruby-core:67962] [Bug #10821]
+
+Tue Feb  3 15:23:58 2015  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (remove_method): When remove refined
+	  method, raise a NameError if the method is not
+	  defined in refined class.
+
+	  But if the method is defined in refined class,
+	  it should keep refined method and remove original
+	  method.
+
+	  Patch by Seiei Higa. [ruby-core:67722] [Bug #10765]
+
+Tue Feb  3 14:04:47 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): obtain real name with FindFirstFile API
+	  instead of matching all entries, on Windows.
+	  [ruby-core:67954] [Bug #10819]
+
+Tue Feb  3 12:26:35 2015  Katsuhiko Nishimra  <ktns.87@gmail.com>
+
+	* lib/mkmf.rb (configuration): set the default cxxflags, which is
+	  referred from the default CXXFLAGS, for extension libraries.
+	  [Fix GH-823]
+
+Tue Feb  3 05:52:00 2015  Aman Gupta <ruby@tmm1.net>
+
+	* gc.c (rb_objspace_free): cause rb_bug if lazy sweep is in progress
+	  during rb_objspace_free. Adds extra protection for r46340.
+	  Patch by Vicent Marti. [Bug #10768] [ruby-core:67734]
+	* gc.c (rb_objspace_call_finalizer): Ensure GC is completed after
+	  finalizers have run. We already call gc_rest() before invoking
+	  finalizers, but finalizer can allocate new objects and start new GC
+	  cycle, so we call gc_rest() again after finalizers are complete.
+
+Mon Feb  2 10:51:34 2015  Ari Pollak  <ajp@aripollak.com>
+
+	* doc/security.rdoc (Symbols): update about Symbol GC.  Symbols
+	  explicitly converted from Strings now can be collected, but
+	  reflection/metaprogramming still can cause memory flooding.
+	  [Fix GH-725]
+
+Sun Feb  1 13:46:52 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (bin-comm): drop batch file installation.
+	  Windows 95 support has not been supported already.
+	  [Feature #10806]
+
+Sat Jan 31 12:06:23 2015  Scott Francis  <scott.francis@shopify.com>
+
+	* thread_pthread.c (reserve_stack): fix intermittent SIGBUS on
+	  Linux, by reserving the stack virtual address space at process
+	  start up so that it will not clash with the heap space.
+	  [Fix GH-822]
+
+Fri Jan 30 17:28:29 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* math.c (num2dbl_with_to_f): make faster when Bignum passed by
+	  direct conversion using rb_big2dbl().  [Feature #10800]
+
+Thu Jan 29 23:30:00 2015  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c (rb_rational_num): add fallback function
+	  for rubies lower than 2.2.0.
+
+	* ext/bigdecimal/bigdecimal.c (rb_rational_den): ditto.
+
+	* ext/bigdecimal/extconf.rb: check the existences of struct RRational,
+	  rb_rational_num, and rb_rational_den.
+
+	* ext/bigdecimal/bigdecimal.bundle: bump version.
+
+Thu Jan 29 20:28:25 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/make-snapshot: removed md5 digest with package information
+
+Thu Jan 29 10:41:52 2015  Kazuki Tanaka  <gogotanaka@ruby-lang.org>
+
+	* math.c (Get_Double): direct casting from Fixnum to double.
+	  [Feature #10785]
+
+Thu Jan 29 02:34:27 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: fix parsing hashes with
+	  instance variables when it is referenced multiple times.
+	* ext/psych/lib/psych.rb: bump version
+	* ext/psych/psych.gemspec: bump version
+	* test/psych/test_hash.rb: test for fix
+
+Thu Jan 29 02:06:09 2015  Eric Wong  <e@80x24.org>
+
+	* thread.c (struct join_arg): restructure and make smaller
+	  (thread_join_sleep): avoid timeofday() call if forever
+	  (thread_join): pass join_arg.delay directly
+	  (rb_thread_inspect_msg): remove, inline into rb_thread_inspect
+	  (rb_thread_inspect): reduce branching and string creation
+	* thread_pthread.c (native_set_thread_name): create string directly
+	  to avoid reparsing. [Misc #10723]
+
+Wed Jan 28 21:32:24 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* thread.c: Improve documentation for Thread#value
+	  [Bug #10694][ruby-core:67324][ci skip]
+
+Tue Jan 27 16:04:19 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: added `!` command.
+
+Tue Jan 27 15:58:23 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: added history feature for platforms which
+	  lack readline.
+
+Mon Jan 26 22:09:35 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .gitignore: ignored temporary file with git.
+
+Mon Jan 26 18:06:03 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Import version 2.2.2 from
+	  https://github.com/knu/ruby-electric.el.
+
+Mon Jan 26 11:37:49 2015  Dave Stevens  <dave@crowdlab.com>
+
+	* string.c (str_make_independent_expand): terminate String when
+	  moved from heap to embedded.  [Fix GH-821].
+
+Sun Jan 25 12:04:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (flatten): no need to call to_ary method on elements
+	  beyond the given level.  [ruby-core:67637] [Bug #10748]
+
+Sun Jan 25 00:42:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/win32/libffi.mk.tmpl: assemble without directory prefix.
+	  workaround of a bug of VC12 ml, by unak at [ruby-core:67792].
+	  [ruby-core:67789] [Bug #10780]
+
+Sat Jan 24 19:56:25 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dln_find.c (dln_find_1): search regular files only.  based on
+	  the patch by Alex Coomans in [ruby-core:67766].  [Bug #10776]
+
+Fri Jan 23 23:57:05 2015  Misumi Rize  <r@ayase-e.li>
+
+	* vm_insnhelper.c (vm_throw_start): search the target to break
+	  from a block with nested rescue, from the nested blocks.
+	  [ruby-core:67765] [Bug #10775] [Fix GH-820]
+
+Fri Jan 23 20:00:59 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_object, marshal_dump): use identity tables for
+	  arbitrary VALUE keys, because of performance of FLONUM.
+	  [Bug #10761]
+
+	* marshal.c (obj_alloc_by_klass, marshal_load): ditto.
+
+Fri Jan 23 17:12:33 2015  Eric Wong  <e@80x24.org>
+
+	* benchmark/bm_marshal_dump_flo.rb: new benchmark for [Bug #10761]
+
+Thu Jan 22 18:03:19 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (find_svn_log): use double quotes instead
+	  of single quotes because cmd.exe doesn't handle them.
+
+	* tool/redmine-backporter.rb (done): the 2nd element of matched data
+	  is the offset of the end of matched string, not length.
+
+Thu Jan 22 16:45:24 2015  Eric Wong  <e@80x24.org>
+
+	* st.c (st_numhash): mix float value for flonum
+	* hash.c (rb_any_hash): ditto
+	* benchmark/bm_hash_aref_flo.rb: new benchmark
+	* benchmark/bm_hash_ident_flo.rb: ditto
+	  [Bug #10761]
+
+Wed Jan 21 22:33:51 2015  Akinori MUSHA  <knu@iDaemons.org>
+
+	* misc/ruby-electric.el: Import version 2.2.1 from
+	  https://github.com/knu/ruby-electric.el.  Improve compatibility
+	  with and optimize for Emacs 24.4.
+
+Wed Jan 21 09:40:52 2015  Zachary Scott  <e@zzak.io>
+
+	* file.c: Document other cases of missing birthtime on OS with patch
+	  provided by @sho-h similar to GH-817. [ci skip] [DOC]
+
+Wed Jan 21 09:09:19 2015  Tanaka Akira  <akr@fsij.org>
+
+	* NEWS: References to tickets added.
+	  [ruby-core:67701] [Bug #10760] Suggested by Zachary Scott.
+
+Tue Jan 20 22:59:54 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/vcs.rb: fix the exception given remote-url of svn.
+
+Tue Jan 20 12:58:33 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: now can change the page of `ls`.
+
+Tue Jan 20 12:28:37 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (readline): fallback to normal gets on
+	  Windows because IO.console.getch is not always do as expected.
+
+Tue Jan 20 11:31:07 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: update usage.
+
+Tue Jan 20 11:23:47 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (mygets): to support Backspace
+	  implement gets by itself.
+
+Tue Jan 20 02:54:11 2015  Zachary Scott  <e@zzak.io>
+
+	* file.c: NotImplementedError is raised if birthtime is unavailable.
+	  Patch by @y-yagi san and [Fixes GH-817] [ci skip] [DOC]
+
+	* ext/pathname/pathname.c: ditto.
+
+Mon Jan 19 22:08:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (include_modules_at): allow to prepend each module up to
+	  once for each classe.  [EXPERIMENTAL]
+
+Sun Jan 18 18:32:20 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* math.c (math_atan2): revive documentation before r49220.
+	  http://d.hatena.ne.jp/nagachika/20150112/ruby_trunk_changes_49213_49226
+
+Sun Jan 18 15:57:32 2015  Seiei Higa  <hanachin@gmail.com>
+
+	* vm_method.c (check_definition): Module#public_method_defined?,
+	  Module#private_method_defined?, Module#protected_method_defined?
+	  should not use refinements. [ruby-core:67656] [Bug #10753]
+
+Sun Jan 18 15:50:39 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_dump.c (rb_vm_bugreport): check by configured result instead
+	  of system name for old FreeBSD.  based on a patch by Steve Wills
+	  at [ruby-core:67655].  [Bug #10752]
+
+Sun Jan 18 12:56:49 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (VPATH, NEWLINE_C), common.mk (common-srcs): make
+	  and use newline.c under enc/trans directory, not toplevel.  no
+	  longer search enc directory implicitly.
+
+	* configure.in, enc/Makefile.in (BUILTIN_ENCS, BUILTIN_TRANSES):
+	  prefix respective directory names to builtin encodings and
+	  transcoder source names.
+
+Sun Jan 18 11:49:46 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_invoke): avoid SEGV when VT_RECORD
+	  variable is passed by reference. [ruby-dev:48803] [Bug #10697]
+
+Sat Jan 17 23:59:15 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb (visit_String):
+	  anchors like `\Z` are not valid inside character class.  use
+	  negative-lookahead instead.
+	  Fixes: https://github.com/tenderlove/psych/issues/221
+
+Sat Jan 17 23:42:27 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: get rid of pattern substitution, which is not
+	  supported by ash, and ash on NetBSD parses whole source first
+	  and fails to start.  [ruby-dev:48823] [Bug #10750]
+
+Fri Jan 16 18:42:58 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: support adding related revisions
+	  to issues.
+
+Fri Jan 16 17:20:33 2015  Koichi Sasada  <ko1@atdot.net>
+
+	vm.c, vm_core.h: constify VM_CF_LEP, VM_CF_PREV_EP, VM_CF_BLOCK_PTR
+	and rb_vm_control_frame_block_ptr.
+
+Fri Jan 16 15:41:21 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: bump version
+
+	* ext/psych/psych.gemspec: ditto
+
+Fri Jan 16 15:37:49 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: only dump hash
+	  subclasses. Thanks Joe Eli McIlvain <joe.eli.mac@gmail.com>
+
+	* test/psych/test_hash.rb: test for change
+
+Fri Jan 16 15:35:21 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych.rb: bump version
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: fix line width wrapping
+	  for long strings.  Thanks Jakub Jirutka <jakub@jirutka.cz>
+
+	* test/psych/test_string.rb: test for change
+
+Fri Jan 16 11:44:44 2015  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* eval_intern.h, vm.c, vm_eval.c, vm_insnhelper.c:
+	  change throw mechanism (not save target ep, but save target cfp).
+	  It fixes `unexpected break' bug that occurs when
+	  TracePoint#binding is called.
+	  [ruby-dev:48797] [Bug #10689]
+
+	* test/ruby/test_settracefunc.rb: add a test.
+
+Thu Jan 15 23:55:15 2015  Tanaka Akira  <akr@fsij.org>
+
+	* io.c (rb_io_close_m): Don't raise when the IO object is closed.
+	  [ruby-core:67444] [Feature #10718]
+
+Thu Jan 15 21:34:57 2015  Seiei Higa  <hanachin@gmail.com>
+
+	* proc.c (rb_obj_singleton_method): Kernel#singleton_method should
+	  not use refinements, as well as Kernel#method.
+	  [ruby-core:67603] [Bug #10744]
+
+Thu Jan 15 10:45:04 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_select_bang, ary_reject_bang): linear
+	  performance.  [ruby-core:67418] [Feature #10714]
+
+Wed Jan 14 18:06:06 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* lib/uri/mailto.rb: raising URI::InvalidComponentError instead
+	  of failing with undefined method `split' for nil:NilClass for
+	  mailto: URIs without opaque part. [Bug #10738]
+	* test/uri/testuri.rb: Test for above
+
+Wed Jan 14 16:45:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/downloader.rb (RubyGems.download): verify downloaded gem
+	  packages.  LowSecurity to allow untrusted certificates now.
+
+Wed Jan 14 15:43:48 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/readline/readline.c (readline_s_refresh_line): initialize
+	  before rl_refresh_line(), as some function make the internal
+	  state non-clean but rl_refresh_line() does not re-initialize it.
+	  [ruby-core:43957] [Bug #6232]
+
+Tue Jan 13 21:59:24 2015  Michal Papis  <mpapis@gmail.com>
+
+	* tool/rbinstall.rb (gem): fix changing permissions of installed
+	  bundled gems.  [Fix GH-812]
+
+Tue Jan 13 21:57:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (distclean-local): remove autom4te.cache generated by
+	  autoconf.
+
+	* common.mk (realclean-local): remove id sources and dummy header
+	  for dtrace.  [ruby-core:67562] [Bug #10737]
+
+Tue Jan 13 21:08:22 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/json, test/json: merge JSON HEAD(259dee6)
+	  separate implementation of Typed_Data macro.
+	  https://github.com/flori/json/compare/v1.8.1...v1.8.2
+
+Tue Jan 13 14:16:35 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/net/http/test_http.rb: get rid of accessing DNS actually
+	  for some servers returning wrong results.
+	  [ruby-core:67454] [Bug #10721]
+
+Mon Jan 12 23:21:57 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* gems/bundled_gems: update test-unit to 3.0.9.
+
+Mon Jan 12 18:35:44 2015  Eric Wong  <e@80x24.org>
+
+	* numeric.c (bit_coerce): use original value for error message
+	  [ruby-core:67405] [Bug #10711]
+	* test/ruby/test_numeric.rb (test_coerce): check error message
+
+Mon Jan 12 18:01:24 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rdoc/text.rb (expand_tabs): get rid of infinite loop with
+	  CR.  should check if substitution occurred too.
+	  [ruby-dev:48813] [Bug #10732]
+
+Mon Jan 12 16:45:46 2015  Seiei Higa  <hanachin@gmail.com>
+
+	* vm_method.c (rb_alias): raise a NameError when creating alias to
+	  a refined method if the original method of the refined method is
+	  not defined.  [ruby-core:67523] [Bug #10731]
+
+Mon Jan 12 13:53:17 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* math.c (math_atan2): improve documentation.
+	  [Feature #10323][ruby-core:65400][ci skip]
+
+Mon Jan 12 13:50:49 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/bigdecimal/bigdecimal.c: fixes documentation like labeled lists,
+	  code examples etc. [ruby-core:66730][Bug #10576][ci skip]
+
+Mon Jan 12 13:36:44 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/optparse.rb: improvements for OptionParser documentation.
+	  [misc #10608][ruby-core:66901][ci skip]
+
+Mon Jan 12 13:33:52 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* array.c (rb_ary_each): documented return value.
+	  [misc #10469][ruby-core:66063]
+
+Sun Jan 11 15:11:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/webrick/test_utils.rb (test_create_listeners): use
+	  dynamically chosen port number, not hardcoded port number.
+	  [ruby-core:67508]
+
+Sat Jan 10 12:57:12 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/zlib/zlib.c: fix document of method signatures.
+	  [Bug #10668][ruby-core:67186][ci skip]
+
+Sat Jan 10 12:32:44 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* eval_error.c (error_print): pos and len parameters of rb_str_substr()
+	  are counted by characters, not bytes.  use rb_str_subseq() instead.
+	  [Bug #10727] [ruby-core:67473]
+
+Sat Jan 10 10:58:55 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* complex.c: removed commented-out code.
+
+Sat Jan 10 10:57:19 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* rational.c: removed commented-out code.
+	  [Feature #10376][ruby-core:65643]
+
+Sat Jan 10 10:12:15 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_select_bang): keep the array consistent by
+	  removing unselected values soon.  [ruby-dev:48805] [Bug #10722]
+
+Fri Jan  9 23:20:04 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/rubygems:  Update to RubyGems HEAD(e53c54a).
+	* test/rubygems:  ditto.
+
+Fri Jan  9 11:13:01 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assocs, assoc): eliminate splatting empty literal
+	  hashes.  [ruby-core:67446] [Bug #10719]
+
+	* compile.c (compile_array_): support splatted hash in hash type.
+
+Fri Jan  9 10:57:09 2015  Vit Ondruch  <vondruch@redhat.com>
+
+	* configure.in (RUBY_SETJMP_TYPE): Remove superfluous semicolon
+	  which causes a syntax error with autoconf 2.63.
+	  [ruby-core:67429] [Bug #10716]
+
+Fri Jan  9 07:23:32 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: correctly quote non-ascii
+	  letters. Thanks @jirutka for the patch.
+
+	* test/psych/test_string.rb: test for change
+
+Fri Jan  9 07:13:55 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: call `allocate` on hash
+	  subclasses.  Fixes github.com/tenderlove/psych/issues/196
+
+	* test/psych/test_hash.rb: test for change
+
+Fri Jan  9 06:58:43 2015  Aaron Patterson <aaron@tenderlovemaking.com>
+
+	* ext/psych/lib/psych/visitors/to_ruby.rb: revive hashes with ivars
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb: dump hashes with ivars.
+	  Fixes github.com/psych/issues/43
+
+	* test/psych/test_hash.rb: test for change
+
+Thu Jan  8 17:05:00 2015  Seiei Higa  <hanachin@gmail.com>
+
+	* vm_method.c (rb_method_entry): if no super class, no original
+	  method entry.  [ruby-core:67389] [Bug #10707]
+
+Thu Jan  8 16:31:43 2015  Seiei Higa  <hanachin@gmail.com>
+
+	* vm_method.c (rb_export_method): bail out if the original method
+	  is undefined when the method is refined.
+	  [ruby-core:67387] [Bug #10706]
+
+Thu Jan  8 12:53:44 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (glob_helper): match in case-folding only if the directory
+	  resides on a case-insensitive file system, on OSX.
+	  [ruby-core:67364] [Bug #10700]
+
+Thu Jan  8 11:39:18 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* .travis.yml: Remove redundant configuration option.
+	  [fix GH-809] Patch by @gxworld
+
+Thu Jan  8 07:17:14 2015  Eric Wong  <e@80x24.org>
+
+	* lib/resolv.rb: consider ENETUNREACH as ResolvTimeout
+	  [ruby-core:67411] [Bug #10712]
+
+Thu Jan  8 00:13:52 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open3.rb: Open3 properly passes non-keyword hash args to spawn.
+	  Fixed by Josh Cheek.  [Fix GH-808]
+	  Related to [ruby-core:67347] [Bug #10699]
+
+Wed Jan  7 19:19:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (dir_initialize): workaround of opendir failure at symlink
+	  directories on Windows via CIFS.
+
+Wed Jan  7 18:52:50 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (need_normalization): not only HFS+, CIFS (SMB) is also
+	  decomposed.  [Bug #10704]
+
+	* dir.c (NORMALIZE_UTF8PATH): Unicode decomposition seems to
+	  perform in an upper layer than file systems on OSX, as all path
+	  names are always decomposed regardless of file system types.
+
+Tue Jan  6 21:41:04 2015  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (timelocalw): Set tm_isdst field -1 if vtm->isdst is
+	  VTM_ISDST_INITVAL.  This bug is introduced at packing struct
+	  vtm (r45155).
+	  [ruby-core:67345] [Bug #10698] Reported by Boris Ruf.
+
+Tue Jan  6 03:10:54 2015  Koichi Sasada  <ko1@atdot.net>
+
+	* test/fiddle/test_handle.rb: fix syntax.
+
+Tue Jan  6 00:16:10 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/fiddle/test_handle.rb (test_NEXT): use -test-/dln/empty
+	  which is always a shared object and is not used by others.
+	  [ruby-dev:48629] [Bug #10384]
+
+Mon Jan  5 14:58:01 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_io.rb: added timeout for AIX environment.
+	  [ruby-core:62983][Bug #9917]
+
+Sun Jan  4 22:33:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit.rb (ExcludesOption): add "excludes" support
+	  to test suite, for alternative implementations and platforms.
+	  [Feature #10682]
+
+Sun Jan  4 22:32:42 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit.rb (Test::Unit): reorder modules and merge
+	  each modules.
+
+	* test/lib/test/unit.rb (Test::Unit): split the large class into
+	  each modules.
+
+Sun Jan  4 21:32:52 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_label): return tLABEL value as it is.
+	  [ruby-core:67315] [Bug #10693]
+
+Sun Jan  4 14:02:37 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit/parallel.rb (run): expand the file name to be
+	  loaded, so that relative paths work in parallel mode.
+
+Sun Jan  4 13:36:56 2015  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* test/test_tempfile.rb: use assert_file for more descriptive message.
+	  following r49131.
+
+Sun Jan  4 13:05:09 2015  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* string.c: improve docs for String#<=>. [ruby-core:65399][Feature #10322]
+	  Patch by gogo tanaka.
+
+Sun Jan  4 12:42:24 2015  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* array.c: improve docs for Array#reject. [ruby-core:65324][misc #10307]
+	  Patched by Nebu Pookins.
+
+Sun Jan  4 12:24:11 2015  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* string.c: improve docs for String#strip and variations.
+	  [ruby-core:66081][Bug #10476]
+
+Sun Jan  4 09:21:04 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/drb/drb.rb: removed unused argument. Patch by @vipulnsward
+	  [fix GH-515]
+
+Sun Jan  4 09:18:31 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/tempfile.rb: provide default basename parameter.
+	  [fix GH-523] Patch by @dissolved
+	* test/test_tempfile.rb: ditto.
+
+Sun Jan  4 00:43:41 2015  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* README.ja.md: add guidance of mailing list and bugs.r-l.o.
+
+Sat Jan  3 23:56:28 2015  Ayumu AIZAWA  <ayumu.aizawa@gmail.com>
+
+	* hash.c: fix docs for Hash#invert. [ruby-core:66917] [Bug #10612]
+
+Sat Jan  3 19:52:12 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_str_new_literal): define on all
+	  platforms, not only gcc.
+
+Sat Jan  3 18:53:28 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/lib/fiddle/cparser.rb (parse_ctype): limit split word
+	  number as the rest are not used.
+
+Sat Jan  3 18:19:50 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/fiddle/lib/fiddle/cparser.rb: r49110 broke Fiddle::Import with
+	  type_alias.
+	* test/fiddle/test_cparser.rb: added type_alias test for parse_ctype
+	  and parse_struct_signature.
+
+Sat Jan  3 11:50:16 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/openssl/ossl.h: avoid to build failure of Windows environment.
+	* ext/openssl/ossl_ssl_session.c: ditto.
+
+Sat Jan  3 11:27:46 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* array.c: Improve performance of Array#shift. use shared instead of
+	  MEMMOVE if with arguments. Patch by @ksss [fix GH-537]
+	* test/ruby/test_array.rb: ditto.
+	* benchmark/bm_array_shift.rb: Added benchmark of GH-537 issue.
+
+Sat Jan  3 10:38:52 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/net/http.rb: More descriptive error message when net/http fails
+	  to connect to a server. Patch by @xaviershay [fix GH-700]
+	* test/net/http/test_http.rb: ditto.
+
+Sat Jan  3 10:14:51 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/openssl/ossl.h: Make `SSL_SESSION_cmp` use `CRYPTO_memcmp`
+	  [fix GH-591] Patch by @PiPeep
+	* ext/openssl/ossl_ssl_session.c: ditto.
+
+Sat Jan  3 09:54:32 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/fiddle/lib/fiddle/cparser.rb: Support for Fiddle::CParser
+	  to handle rich signatures including parameter names and function
+	  pointer types. Patch by @theryan [fix GH-590]
+	* test/fiddle/test_cparser.rb: ditto.
+
+Sat Jan  3 09:01:43 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* NEWS: added compatibility entry of r49101.
+
+Fri Jan  2 21:06:59 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/net/http.rb (Net::HTTP#send_request): there is no response body
+	  with HEAD request. Patch by @rodrigosaito [fix GH-520]
+
+Fri Jan  2 21:04:36 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/net/http/test_http.rb (_test_send_request__HEAD): Added
+	  failing test for send_request with HEAD method.
+
+Fri Jan  2 18:41:20 2015  Tanaka Akira  <akr@fsij.org>
+
+	* eval.c (ruby_init): Print ruby_setup() error only in debug mode.
+	  Unsuppressable error message is not a good idea.
+	  Note that the message is printed sometimes with following
+	  code (highly timing dependent, though):
+	  pid = spawn("ruby -e ''"); Process.kill(:TERM, pid)
+
+Fri Jan  2 16:18:44 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/ruby/test_module.rb: Refactor invalid testcase.
+	  [fix GH-472][ruby-core:59035][Bug #9240]
+
+Fri Jan  2 15:53:00 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/webrick/utils.rb: removed unused argument variable.
+	  [fix GH-356] Patch by @vipulnsward
+	* lib/webrick/server.rb: ditto.
+	* lib/webrick/ssl.rb: ditto.
+	* test/webrick/test_utils.rb: added test for WEBrick::Utils#create_listeners.
+
+Fri Jan  2 15:35:53 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/securerandom.rb: improve syntax and grammar of documentation.
+	  [fix GH-796][ci skip] Patch by @Erol
+
+Fri Jan  2 15:10:01 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* test/openssl/test_ssl_session.rb (OpenSSL#test_ctx_client_session_cb):
+	  fix test failure with OpenSSL disabled SSLv3 protocol.
+	  [ruby-core:63772] [Bug #10046]
+
+Fri Jan  2 09:08:31 2015  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Label::Str#==): Check class equality.
+	  (Resolv::DNS::Name#initialize): Normalize labels as
+	  Resolv::DNS::Label::Str objects.
+
+Thu Jan  1 21:41:49 2015  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* doc/regexp.rdoc: fix regexp docs for whitespace character.
+	  [ruby-dev:48765] [Bug #10624]
+
+Thu Jan  1 17:50:52 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_rubyoptions.rb: try locale encoding name.
+	  [ruby-core:67109] [Bug #10643]
+
+Thu Jan  1 11:07:12 2015  Eric Wong  <e@80x24.org>
+
+	* symbol.c (rb_gc_free_dsymbol): delete from global fstr hash
+	* test/ruby/test_symbol.rb (test_symbol_fstr_leak): test for bug
+	  [ruby-core:67268] [Bug #10686]
+
+Thu Jan  1 09:14:21 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_args.c (keyword_hash_p): fix non-symbol keys hash.
+	  rb_extract_keywords() returns 0 not Qnil when no symbol keys is
+	  included.  [ruby-core:67264] [Bug #10685]
+
+Wed Dec 31 17:48:43 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Label::Str#initialize): Set encoding
+	  ASCII-8BIT before downcase.  case insensitivity of DNS labels doesn't
+	  apply non-ASCII characters.  [RFC 4343]
+
+Wed Dec 31 16:48:44 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (gettable_gen): disable warnings of possible reference
+	  to a local variable defined in a past scope.
+	  [ruby-core:67162] [Bug #10661]
+
+Wed Dec 31 13:43:48 2014  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS::Name#==): Compare an array of Label:Str
+	  objects.  Label#Str#== is case-insensitive.
+
+Tue Dec 30 16:16:12 2014  Ben Miller  <bmiller@rackspace.com>
+
+	* lib/resolv.rb (Resolv::DNS::Name#==): DNS is case-insensitive, so the
+	  comparison should be case-insensitive as well.
+	  [ruby-core:66498] [Bug #10550]
+
+Tue Dec 30 16:03:45 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/resolv.rb (Resolv::DNS::Name): names with different dots
+	  should be different.
+
+Tue Dec 30 13:16:56 2014  Martin Duerst  <bernhard+git@lsmod.de>
+
+	* lib/uri/common.rb: Initialize HTML5ASCIIINCOMPAT to empty Array
+	  to avoid error during bootstrap when encodings are not yet defined.
+	  [Bug #10678]
+
+Tue Dec 30 09:29:26 2014  Bernhard M. Wiedemann  <bernhard+git@lsmod.de>
+
+	* ext/dbm/dbm.c (Init_dbm): [DOC] as UNIX permissions are octal
+	  numbers, needs to be prefixed by 0.  [Fix GH-800]
+
+Tue Dec 30 08:57:39 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* lib/unicode_normalize.rb: typo fix. [ci skip]
+	  [ruby-dev:48794][misc #10675]
+
+Mon Dec 29 19:38:01 2014  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* version.h (RUBY_VERSION): 2.3.0 development has started.
+
+Mon Dec 29 18:58:46 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (unescape_nonascii): append escape sequence as-is not
+	  unescaped character, to get rid of unexpected meta-character.
+	  [ruby-core:67193] [Bug #10670]
+
+Mon Dec 29 14:27:33 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/syntax/literals.rdoc (Symbols): now Symbols created by
+	  interpolation can be garbage collected.  patch by Yihang Ho in
+	  [ruby-core:67194].  [Bug #10671]
+
+Mon Dec 29 11:18:17 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_variable_get): fix dynamic symbol keys.
+	  rb_check_id() returns non-zero only for static symbols, whereas
+	  thread local variable keys can be dynamic symbols.
+	  [ruby-core:67185] [Bug #10667]
+
+Mon Dec 29 10:37:27 2014  Thiago Lewin  <thiago_lewin@yahoo.com.br>
+
+	* io.c (rb_f_select): [DOC] workaround for YARD doc.  [Fix GH-799]
+
+	* process.c (proc_detach): [DOC] fix missing closing parenthesis.
+	  [Fix GH-799]
+
+Mon Dec 29 07:27:23 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/json, test/json: merge JSON HEAD(17fe8e7)
+	  https://github.com/flori/json/compare/v1.8.1...17fe8e7
+
+Sun Dec 28 23:49:37 2014  Michal Papis  <mpapis@gmail.com>
+
+	* rbinstall.rb: fix target location for installing bundled gems.
+	  install to the prepared directory instead of default Gem.dir,
+	  not to be affected GEM_HOME environment variable.  [Fix GH-798]
+
+Sun Dec 28 18:19:28 2014  Eric Wong  <e@80x24.org>
+
+	* test/-ext-/iseq_load/test_iseq_load.rb
+	  (test_next_in_block_in_block): test using ISeq#eval
+	  (test_break_ensure): ditto
+	  [ruby-core:66988]
+
+Sun Dec 28 16:25:12 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (EXTOBJS): override to add resource files
+	  always.  [ruby-core:67153] [Bug #10657]
+
+Sun Dec 28 13:54:26 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (f_kwrest, new_args_tail_gen): unnamed rest keyword and
+	  keywords bits arguments should be unique.  since internal IDs
+	  depend on the local variable index in the current scope, new ID
+	  should be made before popping those vtables.
+	  [ruby-core:67157] [Bug #10659]
+
+Sat Dec 27 20:12:55 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/generator/generator.c (JSON_Generator_State_type): add
+	  #ifdef for backward compatibility.
+
+	* ext/json/parser/parser.rl (JSON_Parser_type): ditto.
+
+	* ext/json/generator/generator.h (ZALLOC): add fallback definition.
+
+	* ext/json/parser/parser.h (ZALLOC): ditto.
+
+Sat Dec 27 16:54:05 2014  Tanaka Akira  <akr@fsij.org>
+
+	* process.c: Unused code removed.
+	  It seems waitpid() is universally available on POSIX platforms.
+
+Sat Dec 27 15:08:27 2014  Eric Wong  <e@80x24.org>
+
+	* vm_core.h (rb_vm_living_threads_insert): preserve order
+	  [Bug #10660] [ruby-core:67154] [ruby-core:67159]
+
+Sat Dec 27 13:08:20 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/socket/socket.c: improved document for YARD doc.
+	  [fix GH-795][ci skip] Patch by @tlewin
+
+Sat Dec 27 10:11:21 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* ext/tk/lib/tkextlib/tcllib/plotchart.rb: fix to invoke correct function
+	  of tcllib. Patch by @zalt50 [fix GH-787]
+
+Sat Dec 27 10:03:41 2014  SHIBATA Hiroshi  <shibata.hiroshi@gmail.com>
+
+	* tool/make-snapshot: show sha1 digest when making packages.
+	  it's request from https://github.com/ruby/www.ruby-lang.org/issues/921
+	  [fix GH-794]
+
+Fri Dec 26 15:32:16 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c (ip_invoke_core): remove probably duplicate
+	  dead code.
+
+Fri Dec 26 15:28:27 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tcltklib.c (ip_ruby_cmd_receiver_const_get): simply use
+	  rb_path2class() to get a class/module from its name.
+
+Fri Dec 26 15:20:54 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* marshal.c (w_long): append at once by w_nbyte() instead of
+	  appending byte by byte.
+
+Fri Dec 26 15:13:13 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/parser/parser.rl (unescape_unicode): check if valid
+	  before bit-or assignments.
+	  reported by Denis Denisov <denji0k AT gmail.com>.
+
+	* ext/nkf/nkf-utf8/nkf.c (nkf_iconv_t): fix a missing semicolon.
+	  reported by Denis Denisov <denji0k AT gmail.com>.
+
+	* process.c (rb_spawn_process): get rid of usage of uninitialized
+	  variable.
+	  reported by Denis Denisov <denji0k AT gmail.com>.
+
+	* regexec.c (match_at): ditto.
+
+	* ext/win32ole/win32ole.c (ole_wc2mb_alloc, ole_vstr2wc, ole_mb2wc):
+	  ditto.
+
+	* dir.c (ruby_glob0): no need to check never-NULL pointer.
+	  reported by Denis Denisov <denji0k AT gmail.com>.
+
+	* win32/file.c (rb_file_expand_path_internal): ditto.
+
+	* win32/file.c (code_page_i): handle realloc failure.
+	  reported by Denis Denisov <denji0k AT gmail.com>.
+
+	* win32/stub.c (stub_sysinit): ditto.
+
+	* fix printf format conversion specifiers.
+	  reported by Denis Denisov <denji0k AT gmail.com>.
+
+Fri Dec 26 01:41:40 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/rbinstall.rb: skip installing bundle gems if zlib is unavailable.
+	  [Bug #10647] [ruby-dev:48787]
+
+Fri Dec 26 01:24:42 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/downloader.rb: support old versions of ruby.
+
+	* tool/downloader.rb: now can download gems by http if openssl is not
+	  available (this may be danger!)
+
+Fri Dec 26 00:13:48 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_extlibs.rb: check existence of extension libraries
+	  which not depend on outer libraries. (experimental)
+
+Thu Dec 25 21:58:15 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/fiddle/extconf.rb: make PIC objects if it will be linked as
+	  a shared object eventually.  [ruby-core:67128]
+
+Thu Dec 25 19:01:13 2014  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/fiddle/win32/libffi-3.2.1-mswin.patch: support mswin32.
+
+Thu Dec 25 17:30:40 2014  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* gc.c (wmap_final_func): fix memory size shortage when realloc wmap.
+	  Fix SEGV during finalize of WeakRef on Solaris (though the SEGV
+	  could occur on all OS/platforms). [ruby-dev:48779] [Bug #10646]
+
+Thu Dec 25 17:27:06 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (NET_LUID): include winsock2.h instead of windows.h.
+	  patch by Jon Forums in [ruby-core:67125].  [Bug #10640]
+
+Thu Dec 25 16:14:10 2014  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_dev): send the given arguments
+	  to the opened console.  as a special case, do nothing if :close
+	  is given.
+
+	* test/lib/leakchecker.rb (LeakChecker#check_fd_leak): close if
+	  console.
+
+For the changes before 2.2.0, see doc/ChangeLog-2.2.0
+For the changes before 2.1.0, see doc/ChangeLog-2.1.0
+For the changes before 2.0.0, see doc/ChangeLog-2.0.0
+For the changes before 1.9.3, see doc/ChangeLog-1.9.3
+For the changes before 1.8.0, see doc/ChangeLog-1.8.0
+
+Local variables:
+coding: us-ascii
+add-log-time-format: (lambda ()
+  (let* ((time (current-time))
+	 (system-time-locale "C")
+	 (diff (+ (cadr time) 32400))
+	 (lo (% diff 65536))
+	 (hi (+ (car time) (/ diff 65536))))
+  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
+indent-tabs-mode: t
+tab-width: 8
+change-log-indent-text: 2
+end:
+vim: tabstop=8 shiftwidth=2
diff --git a/doc/ChangeLog-2.4.0 b/doc/ChangeLog-2.4.0
new file mode 100644
index 0000000000..96b5ecb077
--- /dev/null
+++ b/doc/ChangeLog-2.4.0
@@ -0,0 +1,9492 @@
+Mon Nov  7 00:32:45 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/ruby/test_optimization.rb (test_tailcall_interrupted_by_sigint):
+	  ignore on Windows because we cannot handle signaled status of child
+	  ruby process on this platform.
+
+Sun Nov  6 23:36:07 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_optimization.rb (test_tailcall_interrupted_by_sigint):
+	  send SIGKILL if the child process doesn't die within 1 second.
+
+Sun Nov  6 21:54:28 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/vcs.rb (export_changelog): generate ChangeLog file from
+	  vcs's log. [Feature #12283]
+
+	* tool/make-snapshot (package): overwrite ChangeLog by generated
+	  ChangeLog.
+
+Sun Nov  6 21:36:33 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: Added entry for RDoc 5.
+
+Sun Nov  6 21:35:37 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: ordered entries and adjust styles.
+
+Sun Nov  6 18:23:31 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: Added entry for [Feature #12160][ruby-core:74239]
+
+Sun Nov  6 18:18:25 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* NEWS: Added entry for r56624.
+
+Sun Nov  6 15:53:03 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* configure.in (-Wno-maybe-uninitialized): gcc 6 also shows the same
+	  warnings as described in r49410.
+
+Sun Nov  6 14:42:50 2016  takkanm  <takkanm@gmail.com>
+
+	* eval.c (rb_mod_include, rb_mod_prepend): check if arguments are
+	  given, as well as Kernel#extend.  [ruby-dev:49854] [Bug #12887]
+	  [Fix GH-1470]
+
+Sun Nov  6 11:59:05 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/rinda/test_rinda.rb (test_make_socket_ipv6_multicast,
+	  test_make_socket_ipv6_multicast_hops): skip if IPv6 multicast
+	  address is not available.
+
+Sun Nov  6 11:53:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb.rb (Binding#irb): new method like Binding#pry.
+
+Sun Nov  6 11:48:55 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/net/smtp/test_smtp.rb (test_tls_connect, test_tls_connect):
+	  use Socket.tcp_server_sockets in case localhost is resolved to ::1.
+
+Sun Nov  6 11:49:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/irb.rb (IRB::Irb#run): split from IRB.start.
+
+Sun Nov  6 11:45:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/e2mmap.rb (E2MM.def_exception): remove old definition if
+	  exists so that works multiple times.
+
+Sun Nov  6 10:03:00 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_optimization.rb (test_tailcall_interrupted_by_sigint):
+	  don't use sleep to run faster.
+
+Sun Nov  6 09:58:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_exec_fillarg): honor the given path environment
+	  variable.  [ruby-core:53103] [Bug #8004]
+
+Sun Nov  6 01:52:31 2016  Akira Matsuda  <ronnie@dio.jp>
+
+	* lib/erb.rb: Alias regist_scanner to register_scanner
+
+Sun Nov  6 01:39:36 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* compile.c (tailcallable_p): disable tail call optimization for
+	  toplevel, eval, and load to avoid SEGV when interrupted by SIGINT.
+	  [ruby-core:76327] [Bug #12576]
+
+Sun Nov  6 01:28:41 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.h: rename REG_(CFP|PC|SP|EP) to VM_REG_....
+	  [Bug #12527]
+
+	* vm_exec.c: ditto.
+
+Sun Nov  6 00:25:11 2016  Sho Hashimoto  <sho.hsmt@gmail.com>
+
+	* enumerator.c (InitVM_Enumerator): Enumerator::Lazy#chunk_while
+	  defined.  [Fix GH-1186]
+	  https://github.com/ruby/ruby/pull/1186
+
+Sun Nov  6 00:22:51 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_eval.c (vm_call0_cfunc_with_frame): pass method id to
+	  dtrace hook. follow up r56592.
+
+Sun Nov  6 00:06:38 2016  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::LOC::Coord.create): fixed.
+	  [ruby-core:72567] [Bug #11912] fixed by Eric Wong and Kazuhiro
+	  NISHIYAMA.
+
+Sun Nov  6 00:03:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (-Wimplicit-fallthrough): gcc7 introduces case
+	  fall through warnings but it is too noisy.
+	  https://gcc.gnu.org/bugzilla/show_bug.cgi?id=7652
+
+Sat Nov  5 23:58:32 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* numeric.c (rb_int_round): cast to SIGNED_VALUE to suppress
+	  warnings by -Wsign-compare.
+
+Sat Nov  5 23:48:27 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* file.c (rb_home_dir_of): convert given username into filesystem
+	  encoding. [ruby-core:76682] [Bug #12652]
+	  patched by Davis Mosans
+
+Sat Nov  5 23:46:03 2016  Tanaka Akira  <akr@fsij.org>
+
+	* time.c (vtm_add_offset): Fix yday on last day of year.
+	  [ruby-core:72878] [Bug #11994] Fixed by Andrew White.
+
+Sat Nov  5 23:30:41 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP.post): new convenience method to send
+	  a POST request.  [ruby-core:75484] [Feature #12375]
+
+Sat Nov  5 23:03:54 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/net/http.rb (transport_request): other than HTTPContinue
+	  in 1xx (HTTPInformation) also needs to continue. [Bug #12890]
+
+Sat Nov  5 22:51:06 2016  Tanaka Akira  <akr@fsij.org>
+
+	* lib/time.rb (make_time): "now" argument as nil works again.
+	  This is broken since Ruby 2.2.
+	  Mathieu Jobin pointed a problem.
+	  https://github.com/ruby/ruby/commit/e4b05d91eb0d48fd172abf015c493bb42d755d07#commitcomment-17421387
+
+Sat Nov  5 22:50:13 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/ipaddr.rb (IPAddr#==): If coercion fails, return false
+	  instead of passing through the exception. [ruby-core:77451]
+	  [Bug #12799]
+
+	* lib/ipaddr.rb (IPAddr#<=>): If coercion fails, return nil
+	  instead of passing through the exception. [ruby-core:77451]
+	  [Bug #12799]
+
+Sat Nov  5 22:11:33 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_trace.c (tracepoint_attr_callee_id, rb_tracearg_callee_id):
+	  add TracePoint#callee_id. [ruby-core:77241] [Feature #12747]
+
+	* cont.c, eval.c, gc.c, include/ruby/intern.h, insns.def, thread.c,
+	  vm.c, vm_backtrace.c, vm_core.h, vm_eval.c, vm_insnhelper.c, vm_trace.c: ditto.
+
+	* test/ruby/test_settracefunc.rb: tests for above.
+
+Sat Nov  5 22:09:48 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* eval.c, method.h, proc.c, vm.c, vm_eval.c, vm_insnhelper.c, vm_method.c:
+	  TracePoint#method_id should return method_id, not callee_id.
+	  [ruby-core:77241] [Feature #12747]
+
+	* test/ruby/test_settracefunc.rb: change accordingly.
+
+Sat Nov  5 18:49:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_round, int_round): support round-to-nearest-even
+	  semantics of IEEE 754 to match sprintf behavior, and add `half:`
+	  optional keyword argument for the old behavior.
+	  [ruby-core:76273] [Bug #12548]
+
+Sat Nov  5 18:17:54 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/set.rb (Set#compare_by_identity, Set#compare_by_identity?):
+	  New methods. [Feature #12210]
+
+Sat Nov  5 18:17:08 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rdoc/*, test/rdoc/*: Update rdoc-5.0.0
+	  Release note: https://github.com/rdoc/rdoc/blob/b825775647f62c5b525e9780a28ff2fbb1d5bf6f/History.rdoc#500--2016-11-05
+
+Sat Nov  5 17:29:06 2016  Tanaka Akira  <akr@fsij.org>
+
+	* lib/resolv.rb (Resolv::DNS#extract_resources): Use each_resource
+	  instead of each_answer.
+	  [ruby-core:75461] [Bug #12372] reported by Rafael Fernandez Lopez.
+
+Sat Nov  5 17:18:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/-test-/file/fs.c (get_atime_p): Updating of file access times
+	  is enabled or not.
+
+Sat Nov  5 16:28:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (extract_getline_opts): extract chomp option.
+	  [Feature #12553]
+
+Sat Nov  5 15:58:24 2016  Sho Hashimoto  <sho-h@netlab.jp>
+
+	* tool/mkconfig.rb: [DOC] add rbconfig documentation.
+
+Sat Nov  5 15:42:52 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/smtp.rb (tlsconnect): support timeout for TLS handshake.
+	  [ruby-core:76893] [Bug #12678]
+
+	* lib/net/protocol.rb (ssl_socket_connect): new method to implement
+	  timeout for TLS handshake.
+
+	* lib/net/http.rb (connect): use Net::Protocol#ssl_socket_connect.
+
+Sat Nov  5 14:17:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (brace_body, do_body): since cmdarg_stack is saved in
+	  VALUE val, should restore from the same member.  on big-endian
+	  platforms where VALUE is larger than int, it restored 0 in the
+	  upper word.  [ruby-core:77920] [Bug #12900]
+
+Sat Nov  5 13:52:52 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* lib/shellwords.rb (Shellwords#shellsplit): Fix the handling of
+	  the backslash in double quotes to conform to the standard.
+	  [ruby-core:63807] [Bug #10055]
+
+Sat Nov  5 12:14:31 2016  Tanaka Akira  <akr@fsij.org>
+
+	* ext/pathname/pathname.c (Pathname#empty?): New method.
+	  [ruby-core:76404] [Feature #12596] Proposed by John Backus.
+
+Sat Nov  5 11:53:02 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_refinement.rb (test_refine_alias_in_subclass):
+	  add a test to check that alias in subclasses can be refined.
+	  [ruby-core:69374] [Bug #11186]
+
+Sat Nov  5 11:20:57 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* cont.c (cont_new): disable optimization if clang's version is
+	  3.8.0.  [ruby-core:77894] [Bug #12893]
+
+Sat Nov  5 10:07:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_file.rb (TestFile#test_stat): fix noatime case.
+	  [ruby-core:77943] [Bug #12903]
+
+Fri Nov  4 17:52:44 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (heap_page_resurrect): do not return tomb_pages when
+	  page->freelist == NULL.
+	  [Bug #12670]
+
+Fri Nov  4 16:31:45 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_dtoa): round to even, instead of rounding to
+	  nearest.  [ruby-core:77864] [Bug #12889]
+
+Fri Nov  4 15:31:00 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: Add compiler version message into rbconfig
+	  as RbConfig::CONFIG['CC_VERSION_MESSAGE']. [Feature #12896]
+
+Fri Nov  4 13:03:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (rb_int_digits): Fix exception message
+	  [ruby-core:77502] [Bug #12815]
+
+Fri Nov  4 10:51:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pathname/pathname.c (path_hash): fix unnormalized Fixnum
+	  value bug on mingw/mswin.
+
+Fri Nov  4 10:35:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils::Entry_#copy): ensure that the
+	  source entry exists first, to fix error for non-existent entry.
+	  [ruby-core:77885] [Bug #12892]
+
+Thu Nov  3 21:45:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ext/bigdecimal/bigdecimal.c: Import changes from ruby/bigdecimal
+	  repository.
+
+Thu Nov  3 15:01:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c: include sys/sysmacros.h for ArchLinux which deprecated
+	  use of major() and minor() in sys/types.h.
+
+Wed Nov  2 08:45:49 2016  Dmitry Gritsay  <unseductable@gmail.com>
+
+	* doc/extension.rdoc: set ANYARGS as arguments for func1 and func2.
+
+	* doc/extension.rdoc: mention the exception object which is passed
+	  to func2 as the second argument.  [Fix GH-1471]
+
+Wed Nov  2 07:34:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (rb_num_coerce_bit): enable bit operations with
+	  coercing by non-integer object.  [ruby-core:77783] [Bug #12875]
+
+Tue Nov  1 01:31:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ac_cv_func_round): round(3) in x86_64-w64-mingw32
+	  is not accurate in an edge case.  [ruby-core:77794] [Bug #12878]
+
+Mon Oct 31 17:36:04 2016  Code Ahss  <aycabta@gmail.com>
+
+	* test/test_forwardable.rb (TestForwardable#test_on_private_method):
+	  add test for [Bug #12782] [Bug #12840]
+
+Mon Oct 31 11:08:51 2016  Samuel Williams  <samuel.williams@oriontransfer.co.nz>
+
+	* lib/uri/generic.rb (URI::Generic#merge): merge merge0.
+	  [GH-1469]
+
+Sun Oct 30 15:32:43 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update
+	  rubygems to 2.6.8.
+	  Release note of 2.6.8: https://github.com/rubygems/rubygems/commit/9fb8880976f5ab998912898b091d88aa10eb1d4a
+
+Sun Oct 30 06:39:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1254.c: Fix typo. Reported by k-takata at
+	  https://github.com/k-takata/Onigmo/commit/ceb59cc. Thanks!
+
+Sat Oct 29 19:59:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (cond0): !-operator is a method call, no warning for
+	  literal in condition.  [ruby-core:77801] [Bug #12881]
+
+Sat Oct 29 10:09:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): turn flip-flop in a not-operator
+	  into a boolean value.  fix up r56315
+
+Sat Oct 29 09:39:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (vm_call0_body): follow the original class, not to
+	  loop the prepended module.  [ruby-core:77784] [Bug #12876]
+
+Sat Oct 29 00:14:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/trans/windows-1255-tbl.rb: update mapping from 0xCA to
+	  U+05BA.  [Feature #12877]
+
+Fri Oct 28 23:22:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend: extract transcode_tblgen method calls for libraries
+	  loaded by dynamically generated names, in single_byte.trans.
+
+Fri Oct 28 16:05:03 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* insns.def (opt_case_dispatch): extract float value only if the
+	  Float method is not redefined.
+
+Fri Oct 28 15:58:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (RB_OBJ_BUILTIN_TYPE): special-const safe
+	  BUILTIN_TYPE.
+
+Fri Oct 28 15:20:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (id_finite_p, id_infinite_p, id_rationalize, id_PI):
+	  initialize static IDs.
+
+	* complex.c (FINITE_TYPE_P): extract predicate.
+
+	* complex.c (rb_complex_finite_p, rb_complex_infinite_p): use
+	  dedicated predicates instead of switch by TYPE.
+
+Thu Oct 27 23:28:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_integer_type_p): turn into macro to help
+	  clang based on old gcc to eliminate CSE.
+
+Thu Oct 27 16:33:47 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (rel): check the exception and show right
+	  message.
+
+Thu Oct 27 14:57:33 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/socket/lib/socket.rb (UDPSocket#recvfrom_nonblock): [DOC] Remove
+	  a false statement "If _maxlen_ is omitted, its default value is
+	  65536." maxlen, the first parameter, cannot be omitted as the method
+	  signature indicates. This hasn't changed ever since it was first
+	  implemented.
+
+Thu Oct 27 09:42:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_convert_to_integer): convert a fixable float to a
+	  fixnum directly without the conversion method, as well as bignum
+	  case.
+
+	* object.c (rb_convert_to_integer): should not drop the converted
+	  string.
+
+Wed Oct 26 17:09:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_intern): use prefixed version macro
+	  RUBY_CONST_ID_CACHE.
+
+Wed Oct 26 11:28:25 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* NEWS: added several entries which (if I remember correctly) are
+	  added in a year.
+
+Tue Oct 25 12:54:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (mnew_internal): follow the original class, not to loop
+	  the prepended module.  [ruby-core:77591] [Bug #12832]
+
+Mon Oct 24 10:52:17 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit.rb (Test::Unit::Parallel#deal): update the
+	  job status after a task finished, to show idling workers.
+
+Mon Oct 24 10:51:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (Init_Complex): undefine methods inherited from
+	  Comparable, because Complex does not have <=> method.
+	  [Bug #12866]
+
+	* class.c (rb_undef_methods_from): undefine methods defined in
+	  super from klass.
+
+Mon Oct 24 10:19:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* complex.c (Init_Complex): undefine Complex#clamp, which does not
+	  work like other Comparable methods, because Complex does not
+	  have <=> method.  patched by Tim Peters <zomg.tim AT gmail.com>
+	  in [ruby-core:77720].  [Bug #12866]
+
+Sun Oct 23 11:41:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/encoding.h: include "ruby/ruby.h" explicitly for
+	  enum ruby_fl_type and VALUE.
+
+Sat Oct 22 23:33:55 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_compile_gen): always append error message to the
+	  error buffer.
+
+Sat Oct 22 22:33:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_funcall1): check recursion by inverse pair, to
+	  fix fake infinite recursion.  [ruby-core:77713] [Bug #12864]
+
+Sat Oct 22 18:52:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_compact_bang): should return nil if no elements
+	  is deleted.  [ruby-core:77709] [Bug #12863]
+
+Sat Oct 22 10:28:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DLDFLAGS): fallback to LDFLAGS.
+	  [ruby-core:72444] [Bug #11863]
+
+	* configure.in (LIBRUBY_DLDFLAGS): fallback to DLDFLAGS.
+
+	* configure.in (RUBY_APPEND_OPTION, RUBY_PREPEND_OPTION): expand
+	  the option to be appended/prepended when matching, as well as
+	  RUBY_APPEND_OPTIONS and  RUBY_PREPEND_OPTIONS.
+
+Sat Oct 22 09:52:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (setup_args): duplicate splatting array if more
+	  arguments present to obey left-to-right execution order.
+	  [ruby-core:77701] [Bug# 12860]
+
+Fri Oct 21 16:44:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (do_opendir): retry after GC when the limit for open file
+	  descriptors reached.
+
+Fri Oct 21 16:06:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (open_load_file): retry after GC when the limit for open
+	  file descriptors reached.
+
+Fri Oct 21 11:34:17 2016  Pascal Schmid  <Lechindianer@users.noreply.github.com>
+
+	* string.c (rb_str_sub, rb_str_gsub): [DOC] 'backlash' should read
+	  'backslash'.  [Fix GH-1461]
+
+Thu Oct 20 17:02:56 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/uri/common.rb: added documentation for deprecated method.
+	  [Misc #11960][ruby-core:72733][ci skip]
+
+Thu Oct 20 16:57:23 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/digest/digest.c: Add documentation for Digest.
+	  [Feature #10452][ruby-core:66001][ci skip]
+	  * remove HMAC from list of digest algorithms,
+	  * add MD5 in list of digest algorithms,
+	  * add information about writing a C digest implementation using Digest::Base,
+	  * add documentation for Digest::Base public methods.
+	* ext/digest/md5/md5init.c: add examples for MD5.
+	* ext/digest/rmd160/rmd160init.c: add examples for Digest::RMD160.
+	* ext/digest/sha1/sha1init.c: add examples for Digest::SHA1.
+
+Thu Oct 20 16:19:51 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/open-uri.rb: Improved documentation grammar for
+	  open-uri#open option. [Misc #11329][ruby-core:69868][ci skip]
+
+Wed Oct 19 21:48:40 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: mention refinements with Kernel#send and
+	  BasicObject#__send__.  [Feature #11476]
+
+Wed Oct 19 17:02:15 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_call_opt_send): enable refinements with
+	  Kernel#send and BasicObject#__send__.  [Feature #11476]
+
+Wed Oct 19 14:22:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* basictest/runner.rb: do not clobber the option by --run-opt with
+	  RUBYOPT.  reported by Allen Hewes <rallenh AT hotmail.com>.
+
+Wed Oct 19 00:09:06 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* addr2line.c (parse_debug_line_cu): boundary checks for
+	  compressed debug sections.  [ruby-dev:49840] [Bug #12850]
+
+Tue Oct 18 16:36:40 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (DLDFLAGS): append --compress-debug-sections=zlib
+	  if available, which reduces the size of LIBRUBY_SO by half or
+	  more.
+
+Mon Oct 17 16:20:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/configure.bat: add option to enable/disable to install
+	  static ruby library.  defaulted to "no".  [Feature #12845]
+
+	* configure.in (install-static-library): add option to enable/
+	  disable to install static ruby library.  defaulted to "no" if
+	  enable-shared.  [Feature #12845]
+
+	* tool/rbinstall.rb (local-arch-lib): respect the option.
+
+Sun Oct 16 15:09:06 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1254.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for Windows-1254.
+
+Sat Oct 15 14:17:05 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_args.c (refine_sym_proc_call): search and call method with
+	  refinements.
+
+	* vm_args.c (vm_caller_setup_arg_block): enable refinements when
+	  enabled in the caller.  [Feature #9451]
+
+Sat Oct 15 00:54:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (proc_exec_cmd): use UTF-8 version aspawn.
+	  [ruby-dev:49838] [Bug #12841]
+
+Fri Oct 14 22:26:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse/kwargs.rb (OptionParser#define_by_keywords):
+	  [EXPERIMENTAL] extract command line option definitions from the
+	  information of keyword arguments.
+
+Fri Oct 14 18:27:18 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* object.c: Improve documentation for Float conversion.
+	  [ruby-core:71661][Bug #11736][ci skip]
+
+Fri Oct 14 18:00:20 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/logger.rb: Improve Logger.new option documentation.
+	  [Feature #12803][ruby-core:77467]
+
+Fri Oct 14 17:20:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (make_switch, parse_in_order): unify underscores
+	  to hyphens.
+
+Fri Oct 14 10:48:37 2016  Eric Wong  <e@80x24.org>
+
+	* lib/webrick/utils.rb (TimeoutHandler): use monotonic clock
+	  (watch): ditto
+
+Thu Oct 13 19:41:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (copy_stream_body): use IO to write to copy to duplex IO.
+	  http://twitter.com/knu/status/786505317974585344
+
+Thu Oct 13 17:05:57 2016  Dwain Faithfull  <dwfaithfull@gmail.com>
+
+	* hash.c (rb_hash_compact, rb_hash_compact_bang): Removes nil
+	  values from the original hash, to port Active Support behavior.
+	  [Feature #11818]
+
+Thu Oct 13 11:35:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_sort_bang, rb_ary_sort, rb_ary_sort_by_bang):
+	  [DOC] describe that sort may not be stable.
+
+	* enum.c (enum_sort, enum_sort_by): ditto.
+
+Thu Oct 13 11:31:40 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/rexml/xpath/test_text.rb (test_ancestors): Array#sort may
+	  not be stable.  [ruby-core:76088] [Bug #12509]
+
+	* test/rss/test_maker_{0.9,1.0,2.0}.rb (test_items): ditto.
+
+Thu Oct 13 10:37:31 2016  Eric Wong  <e@80x24.org>
+
+	* doc/extension.rdoc: wording fix
+
+Thu Oct 13 00:39:08 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/test_array.rb (test_sort_bang_with_freeze): make a
+	  clone to copy a <=> singleton method, instead of dup.  which
+	  element will be called is not predictable.
+	  [ruby-core:76088] [Bug #12509]
+
+Thu Oct 13 00:21:27 2016  Simon Soriano  <simon0191@gmail.com>
+
+	* doc/extension.rdoc: Replace "You can defined hooked variables"
+	  with "You can define hooked variables". [Fix GH-1460]
+
+Wed Oct 12 21:05:50 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (open_load_file): revert r56385.  it introduced incompatibility
+	  about `DATA.binmode?`.
+
+Wed Oct 12 15:24:53 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* tool/downloader.rb: Removed verification of gem certification.
+	  Because signed gem is not working on rubygems ecosystem.
+	* tool/gem-unpack.rb: ditto.
+
+Tue Oct 11 22:08:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (prep_io): fix typo of struct member name.
+	  [ruby-core:77550] [Bug #12829]
+
+Tue Oct 11 16:45:24 2016  Tanaka Akira  <akr@fsij.org>
+
+	* lib/uri/generic.rb (URI.find_proxy): Add an optional argument, env.
+
+Tue Oct 11 16:38:32 2016  Tanaka Akira  <akr@fsij.org>
+
+	* lib/pp.rb (String#pretty_print): Defined to print a string as
+	  multiple lines.
+	  [ruby-core:76800] [Feature#12664] proposed by Petr Chalupa.
+
+Mon Oct 10 15:22:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (open_load_file): bind the open fd to an IO instance
+	  before waiting FIFO, not to leak the fd if interrupted.
+
+Mon Oct 10 12:40:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (open_load_file): compare with EXEEXT instead of hard
+	  coded name, and do not match with mere EXEEXT.
+
+	* ruby.c (open_load_file): open in binary mode if available, as
+	  parser deals with EOLs.
+
+	* io.c (prep_io): reduce isatty call (and its system call) on
+	  Cygwin.
+
+Sun Oct  9 23:05:53 2016  Prathamesh Sonpatki  <csonpatki@gmail.com>
+
+	* array.c, class.c: Fixed documentation where Fixnum was referred
+	  directly to use Integer, as Fixnum and Bignum are now unified
+	  into Integer and direct usage is deprecated.  [Fix GH-1459]
+
+Sun Oct  9 11:37:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (VM_ASSERT): stringify expr here before expansion in
+	  RUBY_ASSERT_WHEN.
+
+Sat Oct  8 10:54:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/-test-/memory_status/memory_status.c (read_status): use
+	  Win32 GetProcessMemoryInfo API.
+
+	* ext/-test-/memory_status/memory_status.c: get memory sizes by
+	  mach task_info system call.
+
+Sat Oct  8 09:06:55 2016  Aurelien Jacobs  <aurel@gnuage.org>
+
+	* lib/logger.rb (Logger::Period#next_rotate_time): fix monthly log
+	  rotate when DST is applied during a month of 31 days.
+	  [Fix GH-1458]
+
+Fri Oct  7 20:21:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_prof_setup_new_record): fix the condition to get
+	  rusage.
+
+	* gc.c (gc_profile_dump_major_reason): remove undefined flags.
+
+Fri Oct  7 19:18:33 2016  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/*.c, ext/win32ole/win32ole.h: use RB_INT2FIX instead of
+	  INT2FIX, and so on.
+
+Fri Oct  7 14:18:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* basictest/test.rb: Adjust spaces in class declarations
+	  with inheritance. [fix GH-1227] Patch by @adrfer
+	* lib/irb/*: ditto.
+	* lib/prime.rb: ditto.
+	* lib/shell/builtin-command.rb: ditto.
+	* object.c: ditto.
+	* sample/*.rb: ditto.
+	* test/-ext-/method/test_arity.rb: ditto.
+
+Thu Oct  6 17:29:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_require_safe): SyntaxError created by the parser just
+	  has the message and needs to set up the backtrace.
+	  [ruby-core:77491] [Bug #12811]
+
+	* load.c (rb_load_internal0): load/require is not the main
+	  script.
+
+Thu Oct  6 13:35:15 2016  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enum.c: [DOC] Improve doc [ci-skip]
+
+Thu Oct  6 11:38:14 2016  Chris Jones  <chris@christopherjones.us>
+
+	* ext/readline/extconf.rb: Update error message with correct words.
+	  [fix GH-1453][ci skip] Patch by @magikid
+
+Thu Oct  6 11:01:49 2016  Tieg Zaharia  <tieg.zaharia@gmail.com>
+
+	* lib/uri/common.rb (WEB_ENCODINGS_): Update URI::WEB_ENCODINGS_
+	  hash, and fix documented command to grab it.  [Fix GH-1430]
+
+Thu Oct  6 10:37:13 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* enum.c: Add reduce/inject alias note.
+	  [fix GH-1400][ci skip] Patch by @getaaron
+
+Thu Oct  6 10:30:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (update_coverage): check coverage values, and ignore
+	  non-fixnum values.
+
+Thu Oct  6 09:19:21 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (fptr_finalize): use dedicated macro RB_INTEGER_TYPE_P.
+
+Wed Oct  5 18:02:44 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/http/response.rb: Improve document readability.
+	  [fix GH-1411][ci skip] Patch by @stz-seongheon
+
+Wed Oct  5 15:43:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.c (dump_node): flatten statements in NODE_BLOCK.
+
+Wed Oct  5 14:27:36 2016  Byron Bowerman  <me@bm5k.com>
+
+	* range.c: Add docs for max/min behavior with exclusive range.
+	  [fix GH-1433][ci skip] Patch by @BM5k
+
+Wed Oct  5 12:57:21 2016  Richard Schneeman  <richard.schneeman+foo@gmail.com>
+
+	* ext/socket/*.c: Add proper require for example to work.
+	  [fix GH-1378][ci skip] Patch by @schneems
+
+Wed Oct  5 11:47:19 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* io.c: Fixed equivalent ruby code with core implemention.
+	  [fix GH-1429][ci skip] Patch by @sos4nt
+
+Wed Oct  5 11:36:21 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/delegate.rb: Added missing spaces and Removed needless spaces.
+	  [fix GH-1454][ci skip] Patch by @bogdanvlviv
+
+Wed Oct  5 03:24:55 2016  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* enum.c: Make Enumerable#chunk with no block return
+	  an Enumerator [#2172]
+
+Wed Oct  5 01:19:45 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* internal.h (ST2FIX): new macro to convert st_index_t to Fixnum.
+	  a hash value of Object might be Bignum, but it causes many troubles
+	  especially the Object is used as a key of a hash.  so I've gave up
+	  to do so.
+
+	* array.c (rb_ary_hash): use above macro.
+
+	* bignum.c (rb_big_hash): ditto.
+
+	* hash.c (rb_obj_hash, rb_hash_hash): ditto.
+
+	* numeric.c (rb_dbl_hash): ditto.
+
+	* proc.c (proc_hash): ditto.
+
+	* re.c (rb_reg_hash, match_hash): ditto.
+
+	* string.c (rb_str_hash_m): ditto.
+
+Tue Oct  4 12:59:44 2016  Koichi ITO  <koic.ito@gmail.com>
+
+	* array.c (rb_ary_dig): [DOC] update an example of error message
+	  by Array#dig, because of Integer Unification.  [Fix GH-1455]
+
+Tue Oct  4 09:55:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (BigDecimal_div2): need more room
+	  for precision to round.  [ruby-core:77475] [Bug #12805]
+
+Tue Oct  4 06:05:46 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (onig_print_compiled_byte_code): make the shown address
+	  look relative.
+
+Tue Oct  4 05:53:11 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regexec.c (OPCODE_EXEC_HOOK): op is p-1 because p is already
+	  incremented.
+
+	* regexec.c (OPCODE_EXEC_HOOK): use the exact end address.
+
+	* regexec.c (match_at): don't call OPCODE_EXEC_HOOK in CASE()
+	  when it comes from goto fail.
+
+Mon Oct  3 21:24:41 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: mention Readline.quoting_detection_proc{,=}.
+
+Mon Oct  3 21:19:46 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/logger.rb: [DOC] Fix default value of shift_age.
+
+Mon Oct  3 17:20:05 2016  George Brocklehurst  <ruby@georgebrock.com>
+
+	* ext/readline/readline.c (readline_s_set_quoting_detection_proc):
+	  support rl_char_is_quoted_p.  [Feature #12659]
+
+	* ext/readline/readline.c (readline_s_get_quoting_detection_proc):
+	  ditto.
+
+Sun Oct  2 08:22:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_INT2FIX, RB_LONG2FIX): prefix RB to
+	  global symbols to get rid of name conflicts with other headers.
+
+Sun Oct  2 07:51:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_hash_m): hash values may be negative.
+
+Sun Oct  2 02:04:12 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* string.c (rb_str_hash_m): st_index_t is not guaranteed as the same
+	  size with int, and of course also not guaranteed the value can be
+	  Fixnum.
+
+Sun Oct  2 02:03:06 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* numeric.c (rb_fix2str): detect unnormalized Fixnum value.
+
+Sat Oct  1 23:08:47 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/date/date_parse.c (date_zone_to_diff): it's nonsence and really
+	  harm that to use unary minus operator with unsigned value.
+	  get rid of test failures introduced at r56312.
+
+Sat Oct  1 22:17:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): move numeric literal range
+	  optimization from fixup_nodes() in parse.y.
+
+Sat Oct  1 19:19:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_flip_flop): simplify generated code.
+
+	* compile.c (compile_branch_condition): flip-flop can appear only
+	  in conditional expressions.
+
+Sat Oct  1 02:02:02 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (poll_child_status): rb_w32_wait_events_blocking() sets
+	  errno internally, then should not set it here.
+
+Fri Sep 30 19:06:21 2016  Anton Davydov  <mail@davydovanton.com>
+
+	* lib/uri/http.rb: Documentation and code style improvements.
+	* test/uri/test_http.rb: Added test for coverage.
+	  [fix GH-1427][ruby-core:77255][Misc #12756]
+
+Fri Sep 30 18:43:20 2016  Jason Yeo  <jason@jasonyeo.me>
+
+	* doc/syntax/control_expressions.rdoc: Add missing 'as'
+	  [ci skip][fix GH-1448] Patch by @jsyeo
+
+Thu Sep 29 23:38:04 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/yaml/store.rb (YAML::Store#dump): use table argument instead
+	  of @table directly.
+
+Thu Sep 29 22:22:22 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/yaml/store.rb (YAML::Store#initialize): Fix arguments.
+	  [ruby-dev:49821] [Bug #12800]
+
+	* test/yaml/test_store.rb: Add tests from test/test_pstore.rb.
+
+	* test/yaml/test_store.rb (YAMLStoreTest#test_with_options): Add options test.
+
+	* lib/yaml/store.rb (YAML::Store#dump): Revert to to_yaml.
+
+	* lib/yaml/store.rb (YAML::Store#empty_marshal_data): Use to_yaml with options.
+
+	* lib/yaml/store.rb (YAML::Store#empty_marshal_checksum): Use CHECKSUM_ALGO.
+
+Thu Sep 29 19:34:23 2016  Pete Higgins  <pete@peterhiggins.org>
+
+	* thread_sync.c (rb_queue_pop, rb_szqueue_push, rb_szqueue_pop):
+	  Document exception types, ThreadError, raised by Queue and
+	  SizedQueue.  [Fix GH-1451]
+
+Thu Sep 29 19:21:02 2016  Pete Higgins  <pete@peterhiggins.org>
+
+	* thread_sync.c (Init_thread_sync): Remove confusing doc comments,
+	  which are picked up by rdoc unexpectedly, from Queue and
+	  SizedQueue.  [Fix GH-1450]
+
+Wed Sep 28 23:14:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pstore.rb (PStore::CHECKSUM_ALGO): find available hashing
+	  algorithm for checksum.  MD5 is not available in FIPS mode.
+	  [Feature #6943]
+
+Wed Sep 28 13:00:25 2016  ksss  <co000ri@gmail.com>
+
+	* signal.c (sig_signame): [DOC] Add documentation in the case of
+	  return nil.  [Fix GH-1449]
+
+Wed Sep 28 10:43:27 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* .travis.yml: delete comments. They were outdated.
+
+Wed Sep 28 09:57:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update
+	  rubygems to 2.6.7.
+	  Release note of 2.6.7: https://github.com/rubygems/rubygems/commit/60f35bd1d2359fc30301d2d4cd72bc6833e8d12a
+
+Wed Sep 28 00:21:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_warning_s_warn): the argument must be an
+	  ASCII-compatible string.  [ruby-core:77430] [Bug #12793]
+
+Tue Sep 27 23:22:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (symbol, dsym, parser_set_number_literal): set state to
+	  ENDARG, so that `do` after a literal should be `do_block` and
+	  bound to the outer method.  [ruby-core:72482] [Bug #11873]
+
+	* parse.y (parse_ident): revert r56198.
+
+	* parse.y (warn_balanced): the state of symbol and numeric
+	  literals is now EXPR_ENDARG, do not exclude it.
+
+Tue Sep 27 22:59:42 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* NEWS: news about Warning.warn.
+
+Tue Sep 27 18:10:18 2016  Jeremy Evans <code@jeremyevans.net>
+
+	* error.c: This makes all warnings raised call Warning.warn, which
+	  by default does the same thing it does currently
+	  (rb_write_error_str).  You can override Warning.warn to change
+	  the behavior. [ruby-core:75016] [Feature #12299]
+
+Tue Sep 27 17:35:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (iseqw_s_compile_file): deal with syntax error as well as
+	  compile, and should not abort when rescued.
+
+Tue Sep 27 12:07:17 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/cgi/cookie.rb (parse): don't allow , as a separator. [Bug #12791]
+
+	* lib/webrick/cookie.rb (parse): ditto.
+
+Mon Sep 26 21:37:21 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* man/erb.1, man/irb.1, man/ri.1, man/ruby.1: Remove Ns before
+	  punctuation.
+
+Mon Sep 26 16:23:49 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/stringio/stringio.c (strio_seek): Avoid signed integer overflow.
+	  It's not harmful in practice here, but is still undefined behavior.
+
+	* ext/stringio/stringio.c (strio_extend): Check that the new length does
+	  not exceed LONG_MAX. This fixes the invalid write on the overflow.
+
+	* test/stringio/test_stringio.rb (test_write_integer_overflow): Add a
+	  test case for the above fix in strio_extend().
+
+Mon Sep 26 15:43:34 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* eval_intern.h (TH_PUSH_TAG): Initialize struct rb_vm_tag::tag with
+	  Qundef rather than 0 which is equal to Qfalse. Since Kernel#throw(obj)
+	  searches a tag with rb_vm_tag::tag == obj, throw(false) can
+	  accidentally find an unrelated tag which is not created by
+	  Kernel#catch.  [ruby-core:77229] [Bug #12743]
+
+	* test/ruby/test_exception.rb (test_throw_false): Add a test case for
+	  this.
+
+Mon Sep 26 14:36:12 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/tempfile.rb: provide default basename parameter for
+	  Tempfile.create. [Feature #11965] Patch by Yuki Kurihara
+	* test/test_tempfile.rb: ditto.
+
+Mon Sep 26 14:10:54 2016  Ary Borenszweig  <ary@esperanto.org.ar>
+
+	* string.c (lstrip_offset): add a fast path in the case of single
+	  byte optimizable strings, as well as rstrip_offset.
+	  [ruby-core:77392] [Feature #12788]
+
+Mon Sep 26 12:00:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gems/bundled_gems: update to minitest-5.9.1.
+
+Mon Sep 26 11:50:03 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* tool/rbinstall.rb (gem): Don't install gemspec if no files will be
+	  installed for the gem. This happens when the extension library is not
+	  compiled.
+
+Mon Sep 26 11:06:47 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* string.c (enc_strlen, rb_enc_strlen_cr): Avoid signed integer
+	  overflow. The result type of a pointer subtraction may have the same
+	  size as long. This fixes String#size returning an negative value on
+	  i686-linux environment:
+
+	    str = "\x00" * ((1<<31)-2))
+	    str.slice!(-3, 3)
+	    str.force_encoding("UTF-32BE")
+	    str << 1234
+	    p str.size
+
+Sun Sep 25 22:48:06 2016  namusyaka  <namusyaka@gmail.com>
+
+	* lib/erb.rb (ERB::Compiler::TrimScanner#stag): The :stag accessor
+	  has already been available because it is defined in parent
+	  Scanner class.  [Fix GH-1445]
+
+Sun Sep 25 19:21:26 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* node.c (dump_array): remove duplicate header of NODE_ARRAY/
+	  NODE_VALUES.
+
+Sat Sep 24 22:26:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extract_makefile, extmake, configuration): store
+	  extra libraries to be installed.
+
+	* tool/rbinstall.rb (ext-arch): install extra libraries.
+
+	* ext/zlib/extconf.rb: install zlib if built.
+
+Sat Sep 24 14:24:55 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* spec/README: update URL.
+
+Sat Sep 24 11:28:22 2016  Koichi ITO  <koic.ito@gmail.com>
+
+	* fix typos, "a" before "Integer" to "an".  [Fix GH-1438]
+
+Sat Sep 24 10:19:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/misc/test_ruby_mode.rb (assert_indent): since write-region
+	  in Emacs 25.1 no longer displays the "Wrote file" message, shows
+	  the explicit message to check if successfully finished.
+	  [ruby-core:77355] [Bug #12785]
+
+	* test/misc/test_ruby_mode.rb (EXPR_SAVE): use a numeric prefix
+	  argument of 0, not to make the previous version into a backup
+	  file, instead of backup-inhibited.
+
+	* test/misc/test_ruby_mode.rb (run_emacs): always save the buffer
+	  if modified.
+
+Fri Sep 23 23:57:40 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_method.c (make_method_entry_refined): fix wrong use of ||.
+
+Fri Sep 23 21:38:58 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* tool/mdoc2man.rb (Mdoc2Man#parse_macro): Add support for the
+	  `Lk` macro.
+
+	* tool/mdoc2man.rb (Mdoc2Man#shift_arg): A new function to extract
+	  a possibly quoted argument.
+
+Fri Sep 23 20:36:05 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_refine): refine modules as well.
+	  [ruby-core:76199] [Feature #12534]
+
+Fri Sep 23 20:19:09 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* man/ruby.1: Update the paragraphs in "Rich Libraries" which
+	  included a dead link.
+
+Fri Sep 23 20:05:27 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* man/ruby.1, man/erb.1, man/irb.1, man/ri.1: Use the `Lk` macro
+	  for hyperlinks, which has been available since mdocml 1.5.1
+	  (released more than seven years ago).
+
+Fri Sep 23 15:47:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/forwardable.rb (_delegator_method): allow private methods to
+	  be delegated, with warnings.  [ruby-core:77341] [Bug #12782]
+
+Fri Sep 23 12:15:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): enable tail call
+	  optimization inside a conditional block.
+
+Thu Sep 22 13:00:04 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* gc.c (gc_start_internal): [DOC] methods without arguments
+	  like r56194. [Bug #12777]
+
+Wed Sep 21 17:43:53 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c (InitVM_process): Support CLOCK_MONOTONIC_RAW_APPROX,
+	  CLOCK_UPTIME_RAW, and CLOCK_UPTIME_RAW_APPROX which are introduced
+	  by macOS 10.12.
+
+Wed Sep 21 13:47:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (brace_body, do_body): preserve cmdarg_stack so that
+	  `do` after cmdarg in a block should be `do_block` and bound to
+	  the outer method.  [ruby-core:72482] [Bug #11873]
+
+	* parse.y: `do` after cmdarg in parentheses should be `do_block`
+	  and bound to the outer method.  [ruby-core:72482] [Bug #11873]
+
+	* parse.y (brace_body, do_body): extract block bodies.
+
+Tue Sep 20 23:02:50 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (gc_start_internal): [DOC] add ObjectSpace.garbage_collect
+	  and fix GC#garbage_collect.  [Bug #12777]
+
+Tue Sep 20 21:24:01 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* doc/extension.ja.rdoc: translate r56189 (rb_gc_adjust_memory_usage).
+
+Tue Sep 20 17:52:28 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: Update rake-11.3.0
+
+Tue Sep 20 16:52:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (rb_gc_adjust_memory_usage): notify memory usage to the GC
+	  engine by extension libraries, to trigger GC.  [Feature #12690]
+
+Mon Sep 19 17:05:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (Init_Numeric), bignum.c (Init_Bignum): deprecate
+	  Fixnum and Bignum.  this may be reverted after previews.
+	  [Feature #12739]
+
+Mon Sep 19 10:36:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_init_yielder): directly call stored functions.
+	  [Feature #6183]
+
+	* enumerator.c (lazy_add_method): create lazy enumerator which
+	  uses lazy_init_yielder().
+
+Sun Sep 18 22:48:54 2016  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rss/rss.rb (RSS::BaseModel): Remove needless codes.
+	  [Bug #12773][ruby-dev:49813]
+	  Reported by Kazuhiro NISHIYAMA. Thanks!!!
+
+Sun Sep 18 19:23:47 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* lib/sync.rb: Fix NameError when error.
+
+Sat Sep 17 22:46:02 2016  Koichi ITO  <koic.ito@gmail.com>
+
+	* vm_insnhelper.c (vm_throw_start): Remove too much ";"
+	  [fix GH-1432]
+
+Sat Sep 17 19:19:06 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* test/openssl/test_ssl.rb (test_ctx_options): Fix test failure on
+	  Ubuntu 16.04. The fix in r56147 was incomplete. This is a cherry-pick
+	  of the commit b039f3e268c2 at ruby/openssl.
+
+Sat Sep 17 10:49:35 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* doc/extension.ja.rdoc: translate r56130 (rb_check_arity).
+
+Sat Sep 17 02:40:52 2016  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* ext/objspace/objspace_dump.c: Fix stream processing support for heap
+	  dumps. Full heap dumps should use JSON lines (http://jsonlines.org)
+	  so that we can process very large heaps without loading the entire
+	  heap dump in to memory at once.
+
+Fri Sep 16 22:10:31 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/openssl/ossl_ssl.c (ssl_npn_select_cb_common): Fix compile error
+	  with old version of fcc (Fujitsu C Compiler) on Solaris 10.
+	  [Bug #12769] [ruby-dev:49809]
+
+Fri Sep 16 21:28:25 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* internal.h (MAYBE_UNUSED): The fallback definition is needed.
+	  Fix compile error with some non-GCC compilers such as
+	  Oracle Solaris Studio 12.3 on Solaris 10.
+	  [Bug #12767] [ruby-dev:49807]
+
+	* internal.h (WARN_UNUSED_RESULT): ditto.
+
+Fri Sep 16 21:19:51 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* hash.c (each_pair_i_fast): Fix compile error with old version of
+	  fcc on Solaris 10. [Bug #12768] [ruby-dev:49808]
+
+Fri Sep 16 19:46:05 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (config.h): fixed compile error with VC introduced
+	  by previous commit.
+
+Fri Sep 16 14:54:34 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* internal.h (WARN_UNUSED_RESULT): moved to configure.in, to
+	  actually check its availability rather to check GCC's version.
+
+	* configure.in (WARN_UNUSED_RESULT): moved to here.
+
+	* configure.in (RUBY_FUNC_ATTRIBUTE): change function declaration
+	  to return int rather than void, because it makes no sense for a
+	  warn_unused_result attributed function to return void.
+
+	  Funny thing however is that it also makes no sense for noreturn
+	  attributed function to return int.  So there is a fundamental
+	  conflict between them.  While I tested this, I confirmed both
+	  GCC 6 and Clang 3.8 prefers int over void to correctly detect
+	  necessary attributes under this setup.  Maybe subject to change
+	  in future.
+
+	* internal.h (UNINITIALIZED_VAR): renamed to MAYBE_UNUSED, then
+	  moved to configure.in for the same reason we move
+	  WARN_UNUSED_RESULT.
+
+	* configure.in (MAYBE_UNUSED): moved to here.
+
+	* internal.h (__has_attribute): deleted, because it has no use now.
+
+	* string.c (rb_str_enumerate_lines): refactor macro rename.
+
+	* string.c (rb_str_enumerate_bytes): ditto.
+
+	* string.c (rb_str_enumerate_chars): ditto.
+
+	* string.c (rb_str_enumerate_codepoints): ditto.
+
+	* thread.c (do_select): ditto.
+
+	* vm_backtrace.c (rb_debug_inspector_open): ditto.
+
+	* vsnprintf.c (BSD_vfprintf): ditto.
+
+Fri Sep 16 14:35:55 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* ChangeLog (add-log-time-format): Not exactly sure when but
+	  recently (25.x maybe), emacs changed its API to take optional
+	  two arguments.  We have to follow that.
+
+Fri Sep 16 06:43:25 2016  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* lib/uri/generic.rb (def check_password): don't include bad password
+	  in URI exception output
+
+	* test/uri/test_generic.rb (def test_set_component): test for behavior
+
+Thu Sep 15 21:40:03 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* doc/extension.ja.rdoc: Fix file name.
+
+Thu Sep 15 00:59:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* benchmark/: add vnmakarov's benchmarks around hash [ruby-core:75265]
+
+Wed Sep 14 07:02:31 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* hash.c (each_pair_i_fast): use rb_yield_values2 to avoid var args.
+
+Tue Sep 13 21:32:54 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* string.c (STR_HEAP_SIZE, RESIZE_CAPA_TERM, str_new0, rb_str_buf_new,
+	  str_shared_replace, rb_str_init, str_make_independent_expand,
+	  rb_str_resize): Avoid overflow by casting the length to size_t. size_t
+	  should be able to represent LONG_MAX+termlen.
+
+	* string.c (rb_str_modify_expand): Check that the new length is in the
+	  range of long before resizing. Also refactor to use RESIZE_CAPA_TERM
+	  macro.
+
+	* string.c (str_buf_cat): Fix so that it does not create a negative
+	  length String. Also fix the condition for 'string sizes too big', the
+	  total length can be up to LONG_MAX.
+
+	* string.c (rb_str_plus): Check the resulting String length does not
+	  exceed LONG_MAX.
+
+	* string.c (rb_str_dump): Fix integer overflow. The dump result will be
+	  longer then the original String.
+
+Tue Sep 13 21:30:53 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* gc.c (heap_extend_pages, get_envparam_size, ruby_malloc_size_overflow,
+	  gc_profile_dump_on): Use PRIuSIZE instead of PRIdSIZE as the passed
+	  value is size_t, not ssize_t.
+
+	* iseq.c (get_line_info, rb_iseq_disasm_insn): Ditto.
+
+	* sprintf.c (rb_str_format): Ditto.
+
+	* thread_win32.c (native_thread_create): Ditto.
+
+	* vm.c (get_param): Ditto.
+
+	* ext/objspace/objspace_dump.c (dump_append_string_content,
+	  dump_object): Ditto.
+
+	* ext/socket/raddrinfo.c (host_str, port_str): Ditto.
+
+Tue Sep 13 21:27:35 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* string.c (STR_EMBEDDABLE_P): Renamed from STR_EMBEDABLE_P(). And use
+	  it in more places.
+
+Tue Sep 13 21:23:51 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* README.md: Drop support for BeOS now that Haiku is stable since
+	  2.3.0.
+
+	* README.ja.md: ditto.
+
+Tue Sep 13 21:11:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (STR_EMBEDABLE_P): extract the predicate macro to tell
+	  if the given length is capable in an embedded string, and fix
+	  possible integer overflow.
+
+Tue Sep 13 18:37:08 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_exception.rb: fix thread issues.
+	  * use Queue instead of a local variable for synchronization.
+	  * join created thread to solve leaking threads warning.
+
+Tue Sep 13 16:07:26 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* string.c (rb_str_set_len): The buffer overflow check is wrong. The
+	  space for termlen is allocated outside the capacity returned by
+	  rb_str_capacity(). This fixes r41920 ("string.c: multi-byte
+	  terminator", 2013-07-11).  [ruby-core:77257] [Bug #12757]
+
+	* test/-ext-/string/test_set_len.rb (test_capacity_equals_to_new_size):
+	  Test for this change. Applying only the test will trigger [BUG].
+
+Tue Sep 13 06:03:34 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (benchmark): fix lib path.
+
+Tue Sep 13 00:39:47 2016  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* class.c (singleton_class_of): Copy superclass serial number to
+	  singleton class.  This improves singleton class IMC hit rates.
+	  [Feature #12364] [ruby-core:75425]
+
+Mon Sep 12 13:46:23 2016  Anton Davydov  <mail@davydovanton.com>
+
+	* lib/uri/mailto.rb: Removed needless `return` and use `.` instead of `::`
+	  with class method.
+	* test/uri/test_mailto.rb: Added tests for coverage.
+	  [fix GH-1426][Misc #12750][ruby-core:77242]
+
+Sun Sep 11 21:30:26 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: News about Module.used_modules.
+
+Sun Sep 11 19:06:49 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write_console): should set written length as the
+	  return value.  [Bug #12748] [ruby-dev:49786]
+
+Fri Sep  9 22:43:29 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* gems/bundled_gems: sort lines.
+
+Fri Sep  9 17:59:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_threadptr_raise): set cause from the called thread,
+	  but not from the thread to be interrupted.
+	  [ruby-core:77222] [Bug #12741]
+
+Fri Sep  9 13:50:05 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/extension.rdoc, doc/extension.ja.rdoc: fix file name.
+	  pointed out by @takkanm in the RubyKaigi talk.
+
+Fri Sep  9 13:14:53 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* News: Announcing update to Unicode version 9.0.0 [ci skip]
+
+Fri Sep  9 10:10:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_const_search): warn with the actual class/module
+	  name which defines the deprecated constant.
+
+	* variable.c (rb_const_search): raise with the actual class/module
+	  name which defines the private constant.
+
+Thu Sep  8 17:47:18 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* array.c (flatten): use rb_obj_class instead of rb_class_of
+	  because rb_class_of may return a singleton class.
+	  [ruby-dev:49781] [Bug #12738]
+
+Thu Sep  8 17:40:15 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (gem): use the bindir of each gemspec instead
+	  of hardcoded 'bin', since rdoc 5.0.0 overrides it.
+
+Thu Sep  8 16:47:03 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_s_used_modules): rename Module.used_refinements to
+	  Module.used_modules.  [Feature #7418] [ruby-core:49805]
+
+Thu Sep  8 14:21:48 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/psych.gemspec, lib/rdoc/rdoc.gemspec: Use file list instead of
+	  git output. It shows warning message when invoke `make install`
+	  [Bug #12736][ruby-dev:49778]
+
+Thu Sep  8 13:41:46 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* insns.def (setclassvariable, setconstant): warn when self is a
+	  refinement.  [Bug #10103] [ruby-core:64143]
+
+Thu Sep  8 11:29:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* hash.c (rb_hash_transform_values, rb_hash_transform_values_bang):
+	  Rename map_v to transform_values.
+	  [Feature #12512] [ruby-core:76095]
+
+	* test/ruby/test_hash.rb: ditto.
+
+Thu Sep  8 10:08:35 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* {ext,test}/openssl: Import Ruby/OpenSSL 2.0.0.beta.2. The full commit
+	  history since v2.0.0.beta.1 can be found at:
+	  https://github.com/ruby/openssl/compare/v2.0.0.beta.1...v2.0.0.beta.2
+
+Thu Sep  8 07:23:34 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rdoc/*, test/rdoc/*: Update rdoc-5.0.0.beta2
+	  Fixed ri parse defect with left-hand matched classes.
+	  https://github.com/rdoc/rdoc/pull/420
+
+Thu Sep  8 01:12:47 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* eval.c (rb_mod_s_used_refinements): new method
+	  Module.used_refinements.  based on the patch by Charlie
+	  Somerville.  [Feature #7418] [ruby-core:49805]
+
+Wed Sep  7 17:50:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/util.h (setenv): remove POSIX-noncompliant
+	  definition with 2 arguments.
+
+Wed Sep  7 17:35:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* unicode/8.0.0/casefold.h, name2ctype.h, unicode/data/8.0.0:
+	  removing directories/files related to Unicode version 8.0.0
+
+Wed Sep  7 17:21:55 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/timeout.rb (Timeout#timeout): add custom error message
+	  argument.  [Feature #11650]
+
+Wed Sep  7 17:13:05 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Updated Unicode version to 9.0.0 [Feature #12513]
+
+	* unicode/9.0.0/casefold.h, name2ctype.h, unicode/data/9.0.0:
+	  new directories/files for Unicode version 9.0.0
+
+Wed Sep  7 16:00:45 2016  Tanaka Akira  <akr@fsij.org>
+
+	* lib/open-uri.rb: Allow http to https redirection.
+	  Note that https to http is still forbidden.
+	  [ruby-core:20485] [Feature #859] by Roman Shterenzon.
+
+Wed Sep  7 14:56:59 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/csv.rb (CSV::{Row,Table}#{each,delete_if}): returns an enumerator
+	  if no block is given. [ruby-core:75346] [Feature #12347]
+
+	* test/csv/test_row.rb: add test for above.
+
+	* test/csv/test_table.rb: ditto.
+
+Wed Sep  7 14:50:01 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* gems/bundled_gems: update to power_assert 0.3.1.
+
+Wed Sep  7 12:16:09 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/*, test/psych/*: Update psych-2.1.1
+	  This version fixed following pull requests.
+	  https://github.com/tenderlove/psych/pull/284
+	  https://github.com/tenderlove/psych/pull/276
+
+Wed Sep  7 11:51:06 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rdoc/*, test/rdoc/*: Update rdoc-5.0.0.beta1
+	  This version is mostly same as r56072. It contains to remove code
+	  for Ruby 1.8
+
+Tue Sep  6 09:23:06 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rdoc/rdoc.gemspec: partly reverted for default gem installer.
+	  upstream configuration is not working on ruby core repository.
+
+Mon Sep  5 19:35:22 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rdoc/*, test/rdoc/*: Update rdoc/rdoc master(f191513)
+	  https://github.com/rdoc/rdoc/blob/master/History.rdoc#423--2016--
+	  https://github.com/rdoc/rdoc/blob/master/History.rdoc#422--2016-02-09
+
+Sun Sep  4 00:17:55 2016  Sho Hashimoto  <sho-h@ruby-lang.org>
+
+	* proc.c: [DOC] fix Object#define_singleton_method and
+	  main.define_method return value. [ci skip]
+
+Sat Sep  3 11:28:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread_pthread.c (ruby_init_stack): check stack bounds even if
+	  get_main_stack succeeded, on the "co-routine" case.
+	  https://github.com/ruby/ruby/commit/53953ee#commitcomment-18887413
+
+Fri Sep  2 16:06:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (MEMO_V1_SET, MEMO_V2_SET): fix typos.  use the macro
+	  parameter, not the local variable.
+
+Fri Sep  2 00:55:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (timestamp_file): move extmk.rb specific tricks
+	  from lib/mkmf.rb.  keep RUBYCOMMONDIR prefix not to conflict
+	  with a timestamp file in the toplevel.
+
+Thu Sep  1 14:24:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (gems): move dirty hacks for bundled gems from
+	  mkmf.rb.
+
+	* lib/mkmf.rb (create_makefile): yield all configuration strings.
+
+Wed Aug 31 17:39:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (create_makefile): make gem.build_complete file
+	  under TARGET_SO_DIR and install it only when the gem build
+	  succeeded.  [ruby-core:77057] [Bug #12681]
+
+Wed Aug 31 15:36:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: move TARGET_SO_DIR stuffs to mkmf.rb.
+
+	* lib/mkmf.rb (create_makefile): create target shared object files
+	  under $(TARGET_SO_DIR) which is $sodir if it is defined with
+	  $extout.  [ruby-core:77058] [Bug #12681]
+
+Wed Aug 31 01:56:55 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* doc/extension.ja.rdoc: [DOC] Fix a typo. [ci skip]
+
+Wed Aug 31 00:52:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: make the gems target directory under the expanded
+	  name.  [ruby-core:77102] [Bug #12714]
+
+Tue Aug 30 15:27:27 2016  Kouhei Yanagita  <yanagi@shakenbu.org>
+
+	* ext/json/lib/json/add/ostruct.rb (OpenStruct.json_create):
+	  Correct documentation, fix the name of values.  [Fix GH-1421]
+
+Tue Aug 30 14:53:34 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* io.c (nogvl_fsync, nogvl_fdatasync): on Windows, just ignore if the
+	  fd is associated to non-disk device.  if call fsync and/or fdatasync
+	  with such fds, it causes Errno::EBADF exception and the behavior is
+	  incompatible with ruby 2.1 and earlier unintentionally introduced.
+
+Tue Aug 30 03:38:35 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_dump.c (backtrace): use rip in the saved context for the case
+	  the SIGSEGV is received when the process is in userland.
+	  Note that ip in the stack should be used if the signal is received
+	  when it is in kernel (when it is calling syscall) [Bug #12711]
+
+Sat Aug 27 10:26:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* array.c (rb_ary_concat_multi): take multiple arguments.  based
+	  on the patch by Satoru Horie.  [Feature #12333]
+
+	* string.c (rb_str_concat_multi, rb_str_prepend_multi): ditto.
+
+Thu Aug 25 00:42:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (append_wstr): remove a codepage argument, and use
+	  INVALID_CODE_PAGE for conversion by econv.
+
+	* win32/file.c (append_wstr): exclude the terminator from the
+	  result length when input len == -1.
+
+Wed Aug 24 22:41:30 2016  Kouhei Sutou  <kou@cozmixng.org>
+
+	* gc.c (gc_reset_malloc_info): Remove too much ";".
+
+Wed Aug 24 20:07:57 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/defines.h (ALWAYS_INLINE): Add alternative definition.
+	  Fix compile error with compilers that do not have force inline
+	  attribute, including old version of fcc on Solaris 10.
+	  [ruby-dev:49773] [Bug #12701]
+
+Wed Aug 24 16:56:26 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .gdbinit: follow r55766's VM change.
+
+Wed Aug 24 12:57:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_initialize, rb_class_initialize): [DOC] these
+	  methods do not invoke module_eval/class_eval, just eval the
+	  given block under the new module/class but sharing the context
+	  with the surrounding scope like those methods.
+	  [ruby-core:77023] [Bug #12696]
+
+Tue Aug 23 10:34:40 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/psych/test_psych.rb (test_load_file_with_fallback): fix
+	  Tempfile leak.  https://github.com/tenderlove/psych/pull/288
+
+Tue Aug 23 10:15:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_fs_setter): check and convert $; value at
+	  assignment.
+
+Tue Aug 23 02:09:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_split_m): show $; name in error message when it
+	  is a wrong object.
+
+Mon Aug 22 16:29:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/csv.rb (CSV#shift): store partial quoted strings in an array
+	  and join at last, to improve performance with very long quoted
+	  lines.  [ruby-core:76987] [Bug #12691]
+
+Mon Aug 22 14:35:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* man/irb.1: remove useless -width option.
+	  [ruby-dev:49767] [Bug #12692]
+
+Mon Aug 22 09:02:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (Init_ISeq): undefine allocator of InstructionSequence,
+	  to get rid of segfaults at method call on uninitialized object.
+
+Sat Aug 21 05:47:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (enum_sort): prevent wasteful array duplication.
+
+Sat Aug 20 11:20:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* rubystub.c: generalize win32/stub.c.
+
+Fri Aug 19 11:39:06 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (primary): allow parenthesised statement as a method
+	  argument.  [Feature #12686]
+
+Fri Aug 19 09:12:45 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_set_main_stack): TOPLEVEL_BINDING must be built.
+	  http://www.viva64.com/en/b/0414/#ID0EQ1CI  [ruby-core:76973]
+
+Fri Aug 19 01:00:53 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* proc.c (mnew_missing): Remove an unused argument.
+	  After r51126 rid is not used.
+
+Thu Aug 18 09:26:52 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: bump to test-unit-3.2.1
+
+Thu Aug 18 02:36:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb: skip gems which failed to build extensions.
+	  [ruby-dev:49764] [Bug #12683]
+
+Wed Aug 17 23:35:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gems/bundled_gems (tk): bump up to 0.1.1.
+
+Wed Aug 17 23:14:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: build gem extensions into separate directories
+
+	* tool/rbinstall.rb: install pre-built gem extension files gem
+	  extension directories.  [ruby-core:76931] [Bug #12681]
+
+Tue Aug 16 21:04:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (UNICODE_HDR_DIR): separate unicode header files from
+	  unicode data files.  [ruby-core:76879] [Bug #12677]
+
+Tue Aug 16 11:17:51 2016  Koichi ITO  <koic.ito@gmail.com>
+
+	* lib/net/http/header.rb: Fix typo. [ci skip][fix GH-1407]
+	* transcode.c: ditto.
+
+Tue Aug 16 11:02:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/make-snapshot (package): save generated header files from
+	  unicode data.  [ruby-core:76879] [Bug #12677]
+
+Mon Aug 15 20:31:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.c (dump_array): show nd_alen field in NODE_ARRAY only in
+	  the first node.  it is nd_end in the rest nodes.
+
+Mon Aug 15 16:41:32 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* appveyor.yml: Update libressl version to 2.3.7.
+
+Mon Aug 15 11:46:50 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb (extmake): extension libraries in gems cannot link
+	  statically.
+
+Sun Aug 14 22:35:40 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* id_table.c (hash_table_extend): should not shrink the table than
+	  the previous capacity.  [ruby-core:76534] [Bug #12614]
+
+Sun Aug 14 18:51:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gems/bundled_gems: add gemified tk 0.1.0.
+	  this needs `extract-gems` to build.
+
+Sun Aug 14 14:54:14 2016  Kouhei Sutou  <kou@cozmixng.org>
+
+	* object.c (InitVM_Object): Update referenced document path.
+
+Sat Aug 13 23:08:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_funcall0, num_funcall1): get rid of infinite
+	  recursion in fallback methods of Numeric.
+
+Sat Aug 13 11:10:08 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (command_asgn, arg): fix syntax errors with chained
+	  assignment with op assign.  [Bug #12669]
+
+Sat Aug 13 10:52:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (stmt, arg): rescue modifier in command op assignment
+	  should be limited to rhs only.  [ruby-core:75621] [Bug #12402]
+
+Sat Aug 13 07:51:40 2016  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_val2variant): fix integer conversion in
+	  cygwin64.
+
+Fri Aug 12 21:05:19 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/webrick/config.rb (WEBrick::Config::General):
+	  disable reverse lookup by default. [ruby-core:45514] [Feature #6559]
+	  Socket.do_not_reverse_lookup is true by default but WEBrick
+	  overwrote it.
+	  patch by Eric Hodel [ruby-core:45527]
+
+Fri Aug 12 12:50:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_syntax_error_append): fix newline in syntax error
+	  message to the beginning, not after file name and line number.
+	  [Feature #11951]
+
+Thu Aug 11 16:24:23 2016  Ferdinand Niedermann  <nerdinand@nerdinand.com>
+
+	* compar.c (cmp_clamp): Introduce Comparable#clamp.  [Feature #10594]
+
+Thu Aug 11 03:16:59 2016  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/prime.rb: Optimize prime?
+	  Adapted from patch by Jabari Zakiya [#12665]
+
+	* test/test_prime.rb: Improve test
+
+Wed Aug 10 22:37:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (command_rhs, arg_rhs): introduce new rules to reduce
+	  repeated rules with rescue modifier.
+
+Wed Aug 10 17:26:43 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (command_asgn): rescue modifier in command assignment
+	  should be limited to rhs only.  [ruby-core:75621] [Bug #12402]
+
+Wed Aug 10 15:35:03 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/resolv/resolv.c: needs windows.h for iphlpapi.h on
+	  cygwin.  [ruby-core:76791] [Bug #12663]
+
+	* ext/win32/resolv/resolv.c (w32error_make_error): use
+	  Win32::Resolv::Error, an alias of Win32::Registry::Error.
+
+Tue Aug  9 17:50:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* hash.c (rb_hash_map_v, rb_hash_map_v_bang): implement Hash#map_v and
+	  Hash#map_v! [Feature #12512] [ruby-core:76095]
+
+	* test/ruby/test_hash.rb: add tests for above change.
+
+Tue Aug  9 16:09:03 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_getivar): use always_inline because
+	  gcc7 doesn't inline this without always_inline.
+
+Tue Aug  9 15:41:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/tk: Tk is removed from stdlib.  [Feature #8539]
+	  https://github.com/ruby/tk is the new upstream.
+
+Tue Aug  9 00:12:31 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* doc/maintainers.rdoc: Remove moved file section.
+
+Mon Aug  8 20:56:46 2016  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/sample/excel1.rb, ext/win32ole/sample/excel2.rb,
+	  ext/win32ole/sample/excel3.rb, ext/win32ole/sample/ie.rb,
+	  ext/win32ole/sample/ienavi.rb, ext/win32ole/sample/ienavi2.rb: use
+	  true instead of deprecated TRUE. [ci skip]
+
+Mon Aug  8 12:51:12 2016  Zarko Todorovski <zarko@ca.ibm.com>
+
+	* internal.h (RBASIC_CLEAR_CLASS): Reroute ANSI C's strict
+	  aliasing rule.
+	  [ruby-core:74427][Bug #12191][ruby-core:76747][Bug #12657]
+
+Sun Aug  7 18:08:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (InitVM_Object): deprecate toplevel constants TRUE,
+	  FALSE, and NIL.  [Feature #12574]
+
+Sun Aug  7 06:48:21 2016  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_write_internal):
+	  avoid undefined behavior
+	* test/openssl/test_pair.rb (test_write_zero): new test
+	  [ruby-core:76751] [Bug #12660]
+
+Sat Aug  6 09:35:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* id_table.h (rb_id_table_iterator_result): add dummy sentinel
+	  member because C standard prohibits a trailing comma.
+
+Sat Aug  6 00:39:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_enc_str_new): make string for an environment
+	  variable name or value.
+
+	* hash.c (env_name_new): make environment value string with the
+	  encoding for its name.
+
+Fri Aug  5 23:18:35 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* hash.c (env_str_new): taint the string.  get rid of a test failure
+	  introduced at r55811.
+
+Fri Aug  5 17:04:02 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (w32_getenv): call rb_w32_getenv and rb_w32_ugetenv via
+	  this pointer without further comparisons.
+
+Thu Aug  4 11:54:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (env_assoc): the encoding of the value should be the
+	  locale, as well as other methods, [], fetch, values, etc.
+
+Wed Aug  3 21:31:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_fragment_enc_error): compile_error is different
+	  between parser and ripper.  [ruby-core:76397] [Bug #12651]
+
+Wed Aug  3 17:15:06 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_obj_clone2): restrict freeze option to true other
+	  than false which only has the effect.  [Feature #12300]
+
+Wed Aug  3 10:47:07 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: introduce VM_FRAME_RUBYFRAME_P()
+	  and VM_FRAME_CFRAME_P().
+	  Most of case, RUBY_VM_NORMAL_ISEQ_P() is no
+	  longer needed.
+
+	* vm_core.h: introduce rb_obj_is_iseq().
+
+	* cont.c, vm.c: VM_FRAME_MAGIC_DUMMY with
+	  VM_FRAME_FLAG_CFRAME.
+
+Wed Aug  3 09:25:16 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: rename macros and make them inline functions.
+
+	  * rename VM_FRAME_TYPE_FINISH_P() to VM_FRAME_FINISHED_P().
+	  * rename VM_FRAME_TYPE_BMETHOD_P() to VM_FRAME_BMETHOD_P().
+
+Wed Aug 03 09:15:02 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: introduce VM_FRAME_FLAG_CFRAME to represent cfp->iseq
+	  type.
+
+Tue Aug  2 21:42:40 2016  Chia-sheng Chen  <qitar888@gmail.com>
+
+	* math.c (tanh): make faster by the extract form if three
+	  hyperbolic functions are unavailable.  [Feature #12647]
+
+Tue Aug  2 12:37:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/option.c, ext/socket/rubysocket.h (inet_ntop): share
+	  the fallback definition.  [ruby-core:76646] [Bug #12645]
+
+Tue Aug  2 04:07:29 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (set_pioinfo_extra): use more reliable way to search
+	  the position of pioinfo of VC14, and also support debug library of it.
+	  patched by davispuh AT gmail.com
+	  [ruby-core:76644] [Bug #12644]
+	  this fixes also [Bug #12631]
+
+Mon Aug  1 21:39:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: [EXPERIMENTAL] build extension libraries in
+	  extracted gems.
+
+Mon Aug  1 16:07:18 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* include/ruby/ruby.h (struct RStruct): no longer.
+
+	* internal.h (struct RStruct): moved here.
+
+	* struct.c (rb_struct_ptr): a compensation function for the lack
+	  of RSTRUCT_PTR.  But now that we have RSTRUCT_GET/SET, that must
+	  not be used anyway.  I mark this deprecated.  Dont use it.
+
+Mon Aug  1 14:50:06 2016  Jeremy Evans <code@jeremyevans.net>
+
+	* object.c (rb_obj_clone2): Allow Object#clone to take freeze:
+	  false keyword argument to not freeze the clone.
+	  [ruby-core:75017][Feature #12300]
+
+	* test/ruby/test_object.rb (TestObject): test for it.
+
+Mon Aug  1 12:16:19 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/json/*, test/json/json_parser_test.rb: Update json-2.0.2.
+
+Sun Jul 31 16:17:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/resolv/resolv.c (get_dns_server_list): [Win32] get DNS
+	  servers only for connected network devices by GetNetworkParams
+	  API.  [Bug #12604]
+
+Sat Jul 30 12:13:26 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c (String#downcase), NEWS: Mentioned that case mapping for all
+	  of ISO-8859-1~16 is now supported. [ci skip]
+
+Sat Jul 30 12:00:01 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_2.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-2, by Yushiro Ishii.
+
+Fri Jul 29 20:57:12 2016  chuanshuo  <lilijreey@126.com>
+
+	* *.c: rename rb_funcall2 to rb_funcallv, except for extensions
+	  which are/will be/may be gems.  [Fix GH-1406]
+
+Fri Jul 29 10:51:34 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* proc.c (env_write): remove unused function.
+
+Fri Jul 29 10:49:52 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (VM_LOCAL_P): should return an integer value.
+	  reported at
+	  http://d.hatena.ne.jp/nagachika/20160728/ruby_trunk_changes_55764_55770
+
+Fri Jul 29 04:23:08 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h (VM_ENV_LOCAL_P): return truthy (0 or not) value.
+
+	* vm.c (rb_vm_make_proc_lambda): use VM_ENV_ESCAPED_P() macro.
+
+Fri Jul 29 03:49:04 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, internal.h: remove RubyVM::Env class and all of env objects
+	  are imemo objects (imemo_env).
+
+	* NEWS: describe this change. I believe nobody touch these objects
+	  because there are no method defined.
+
+	* vm_core.h: remove the following definitions.
+	  * rb_cEnv decl.
+	  * GetEnvPtr() because Env is no longer T_DATA object.
+
+	* vm_core.h (rb_env_t): fix layout for imemo values.
+
+	* vm_core.h (vm_assert_env): added.
+
+	* vm_core.h (vm_env_new): added.
+
+Thu Jul 28 19:53:21 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_core.h: revisit the structure of frame, block and env.
+	  [Bug #12628]
+
+	  This patch introduce many changes.
+
+	  * Introduce concept of "Block Handler (BH)" to represent
+	    passed blocks.
+
+	  * move rb_control_frame_t::flag to ep[0] (as a special local
+	    variable). This flags represents not only frame type, but also
+	    env flags such as escaped.
+
+	  * rename `rb_block_t` to `struct rb_block`.
+
+	  * Make Proc, Binding and RubyVM::Env objects wb-protected.
+
+	  Check [Bug #12628] for more details.
+
+Thu Jul 28 15:05:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (ruby_fl_type): use __extension__ to get rid
+	  of pedantic warning against RUBY_FL_USER19.
+	  https://github.com/skylightio/skylight-ruby/issues/64
+
+	* include/ruby/ruby.h (rb_mul_size_overflow): ditto for use of
+	  int128.
+
+Wed Jul 27 10:32:59 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1253.c: Remove dead code found by Coverity Scan.
+
+Tue Jul 26 22:43:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (run_finalizer): make saved running finalizer state
+	  volatile to ensure not to be clobbered by longjmp.
+
+Tue Jul 26 19:26:00 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_insnhelper.c: introduce rb_vm_pop_frame() and use it
+	  instead of setting rb_thread_t::cfp directly.
+
+	* vm_insnhelper.c (vm_pop_frame): return the result of
+	  finish frame or not.
+
+Tue Jul 26 19:06:39 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (rb_raw_obj_info): support to show Proc obj.
+
+Tue Jul 26 18:55:55 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark): add `inline' explicitly.
+	  I expected to inline this function implicitly at the loop
+	  (ex: marking T_ARRAY objects) but sometimes it remains as
+	  normal call.
+
+Tue Jul 26 16:33:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1257.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for Windows-1257, by Sho Koike.
+
+Tue Jul 26 16:19:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1250.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for Windows-1250, by Sho Koike.
+
+	* ChangeLog: Fixed order of previous two entries.
+
+Tue Jul 26 15:54:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1253.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for Windows-1253, by Takumi Koyama.
+
+Tue Jul 26 15:30:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1251.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for Windows-1251, by Shunsuke Sato.
+
+Tue Jul 26 13:04:59 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Add explicit skip test for
+	  availability of Unicode data files.
+
+Mon Jul 25 21:33:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* range.c (check_step_domain): check step argument domain by <=>
+	  method, instead of < and >.
+
+Mon Jul 25 21:11:32 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* doc/maintainers.rdoc: fix filenames.
+
+Mon Jul 25 16:59:00 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* debug.c (ruby_debug_printf): use rb_raw_obj_info()
+	  instead of rb_inspect() because it is more robust way
+	  to see object internal.
+
+Sun Jul 24 16:33:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* regenc.h/c, include/ruby/oniguruma.h, enc/ascii.c, big5.c, cp949.c,
+	  emacs_mule.c, euc_jp.c, euc_kr.c, euc_tw.c, gb18030.c, gbk.c,
+	  iso_8859_1|2|3|4|5|6|7|8|9|10|11|13|14|15|16.c, koi8_r.c, koi8_u.c,
+	  shift_jis.c, unicode.c, us_ascii.c, utf_16|32be|le.c, utf_8.c,
+	  windows_1250|51|52|53|54|57.c, windows_31j.c, unicode.c:
+	  Remove conditional compilation macro ONIG_CASE_MAPPING. [Feature #12386].
+
+Sun Jul 24 12:53:42 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/maintainers.rdoc: xmlrpc is bundled gem from Ruby 2.4.
+
+Sun Jul 24 12:07:39 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/maintainers.rdoc: Update OpenSSL maintainer.
+
+Sat Jul 23 22:43:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (Check_Type): inline check for the object type.
+
+Sat Jul 23 04:06:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RTEST, NIL_P): use RUBY prefixed name in
+	  macros.
+
+Sat Jul 23 01:41:29 2016  Eric Wong  <e@80x24.org>
+
+	* lib/webrick/httpservlet/cgihandler.rb (do_GET): delete HTTP_PROXY
+	* test/webrick/test_cgi.rb (test_cgi_env): new test
+	* test/webrick/webrick.cgi (do_GET): new endpoint to dump env
+	  [ruby-core:76511] [Bug #12610]
+
+Fri Jul 22 19:55:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm.c (vm_set_main_stack): remove unnecessary check.  toplevel
+	  binding must be initialized.  [Bug #12611] (N1)
+
+	* win32/win32.c (w32_symlink): fix return type.  [Bug #12611] (N3)
+
+	* string.c (rb_str_split_m): simplify the condition.
+	  [Bug #12611](N4)
+
+Fri Jul 22 17:13:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c (String#dump): Change escaping of non-ASCII characters in
+	  UTF-8 to use upper-case four-digit hexadecimal escapes without braces
+	  where possible [Feature #12419].
+
+	* test/ruby/test_string.rb (test_dump): Add tests for above.
+
+Fri Jul 22 10:35:35 2016  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/rexml/attribute.rb (REXML::Attribute#to_string): Fix wrong
+	  entry reference name of double quote.
+	  [Bug #12609][ruby-core:76509]
+	  Patch by Joseph Marrero. Thanks!!!
+
+Fri Jul 22 10:32:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* template/unicode_norm_gen.tmpl: Remove
+	  UnicodeNormalize::UNICODE_VERSION at origin [Feature #12546].
+
+Fri Jul 22 09:23:51 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* LEGAL: Added entries for files under the USD license.
+	  [Bug #12598][ruby-core:76428][ci skip]
+
+Fri Jul 22 09:19:57 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* LEGAL: Added entry for `lib/rdoc/generator/template/darkfish/css/fonts.css`
+	  [Misc #12550][ruby-core:76255][ci skip]
+
+Fri Jul 22 06:28:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (run_finalizer): push and exec tag just once, instead of
+	  protecting for each finalizer.
+
+	* gc.c (gc_start_internal, rb_gc_start): set finalizing flag
+	  whenever calling deferred finalizers not to recurse.
+
+Thu Jul 21 22:26:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* missing/strl{cat,cpy}.c: Update latest upstream files.
+	  [Misc #12205][ruby-core:74487]
+	* LEGAL: Update license for missing/strl{cat,cpy}.c.
+
+Thu Jul 21 21:53:30 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* LEGAL: added file list with Public domain license.
+	  [ruby-core:76254][Bug #12549]
+
+Wed Jul 20 17:44:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enumerator.c (lazy_uniq): new method Enumerator::Lazy#uniq.
+	  [Feature #11090]
+
+	* enum.c (enum_uniq): new method Enumerable#uniq.
+	  [Feature #11090]
+
+Wed Jul 20 17:35:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (rb_hash_add_new_element): add new element or do nothing
+	  if it is contained already.
+
+	* array.c (ary_add_hash, ary_add_hash_by): use
+	  rb_hash_add_new_element.
+
+Tue Jul 19 18:21:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* lib/unicode_normalize/tables.rb: Remove
+	  UnicodeNormalize::UNICODE_VERSION (#12546).
+
+Tue Jul 19 15:38:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_local_constants_i): exclude private constants
+	  when excluding inherited constants too.  [Bug #12345]
+
+Sun Jul 17 23:42:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (num_finite_p, num_infinite_p): Add Numeric#finite? and
+	  Numeric#infinite?  [Feature #12039] [ruby-core:73618]
+
+	* complex.c (rb_complex_finite_p): Add Complex#finite?
+
+	* complex.c (rb_complex_infinite_p): Add Complex#infinite?
+
+	* test/ruby/test_bignum.rb: Add test for Integer#finite? and
+	  Integer#infinite?
+
+	* test/ruby/test_fixnum.rb: ditto.
+
+	* test/ruby/test_rational.rb: Add test for Rational#finite? and
+	  Rational#infinite?
+
+	* test/ruby/test_complex.rb: Add test for Complex#finite? and
+	  Complex#infinite?
+
+Sun Jul 17 20:59:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, enc/depend (casefold.h, name2ctype.h): move to
+	  unicode data directory per version.
+
+Sat Jul 16 06:26:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, enc/Makefile.in: moved timestamp files for
+	  directories under the specific directory, to get rid of match
+	  with files under the source directory.
+
+Fri Jul 15 22:05:13 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c (str_buf_cat): Fix potential integer overflow of capa.
+	  In addition, termlen is used instead of +1.
+
+Fri Jul 15 21:30:38 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c (str_buf_cat): Fix capa size for embed string.
+	  Fix bug in r55547. [Bug #12536]
+
+Fri Jul 15 18:13:15 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: update latest gems.
+
+Fri Jul 15 17:08:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_strtod): do not underflow only by preceding zeros,
+	  which may be canceled out by the exponent.
+	  http://twitter.com/kazuho/status/753829998767714305
+
+Fri Jul 15 09:53:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/unicode/case-folding.rb, tool/enc-unicode.rb: check if
+	  Unicode versions are consistent with each other.
+
+Fri Jul 15 08:25:15 2016  Jeremy Evans <code@jeremyevans.net>
+
+	* string.c (STR_BUF_MIN_SIZE): reduce from 128 to 127
+	  [ruby-core:76371] [Feature #12025]
+	* string.c (rb_str_buf_new): adjust for above reduction
+
+Thu Jul 14 17:26:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (enc/unicode/name2ctype.h): remove stale recipe,
+	  which did not support Unicode age properties.
+
+	* common.mk (enc/unicode/name2ctype.h): update by --header option
+	  of tool/enc-unicode.rb.  enc/unicode/name2ctype.kwd file has not
+	  been used.
+
+	* common.mk (enc/unicode/name2ctype.kwd): rule to create from
+	  Unicode data files, used only when the target does not exist.
+
+Thu Jul 14 13:10:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/json/lib/json/ext: remove stale directory.  bundled
+	  extension libraries are placed under the directory for each
+	  architectures, but not mixed with plain text script libraries.
+
+Thu Jul 14 12:48:47 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/json/**/*.rb: merge original files from upstream repository.
+	  It only fixes styles of frozen string literal.
+
+Wed Jul 13 22:23:03 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/json/json_common_interface_test.rb: use assert_raise instead of
+	  assert_raises.
+
+Wed Jul 13 22:14:23 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/lib/test/unit.rb: added test files with `_test` suffix for json
+	  upstream.
+	* test/json: merge original test files from json upstream.
+
+Wed Jul 13 18:09:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_9.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-9, by Kazuki Iijima.
+
+	* enc/iso_8859_9.c: Exclude dotless i/I with dot from case-insensitive
+	  matching because they are not a case pair.
+
+	* test/ruby/enc/test_iso_8859.rb: Make test coverage for ISO-8859-9
+	  a bit more complete.
+
+Wed Jul 13 17:21:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1252.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for Windows-1252, by Serina Tai.
+
+	* test/ruby/enc/test_case_comprehensive.rb: Fix order of encodings.
+
+Wed Jul 13 16:19:14 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_7.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-7, by Kosuke Kurihara.
+
+	* test/ruby/enc/test_case_comprehensive.rb: Fix order of encodings.
+
+Wed Jul 13 16:08:08 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_mark_roots): should mark the VM object itself to mark
+	  singleton class of the VM object.
+	  Before this patch, we only set mark bit for the VM object and
+	  invoke mark function separately.
+	  [Bug #12583]
+
+	* test/ruby/test_gc.rb: add a test.
+
+Wed Jul 13 15:59:59 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* math.c (_USE_MATH_DEFINES): it must be set before including internal.h
+	  because internal.h includes ruby.h, ruby.h includes win32.h, and
+	  win32.h includes system's math.h.
+	  this change is to get rid of a compiler warning (redefinition of
+	  a macro) introduced at r55641.
+
+Wed Jul 13 15:19:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_1.c, enc/iso_8859_4.c: Avoid setting modification flag if
+	  there is no modification.
+
+Wed Jul 13 14:40:04 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_5.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-5, by Masaru Onodera.
+
+	* test/ruby/enc/test_case_comprehensive.rb: Fix order of encodings.
+
+Wed Jul 13 14:28:33 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* math.c (rb_math_sqrt): r55646 must use f_signbit.
+
+Wed Jul 13 14:22:50 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c (Init_ISeq): undef ISeq.translate and ISeq.load_iseq
+	  to prevent calling super classes' methods.
+
+	  Without this patch, you can write workaround like:
+
+	    class << RubyVM::InstructionSequence
+	      def translate; end
+	      undef translate
+	    end
+
+	* test/ruby/test_iseq.rb: add a test.
+
+Wed Jul 13 14:16:03 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_method.c (method_entry_get_without_cache): check
+	  undefined method even if ruby_running is FALSE.
+
+	  We haven't call "undef"ed methods before ruby_running.
+	  So that this issue does not make troubles.
+
+Wed Jul 13 14:15:22 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/windows_1254.c: Adjust variable/macro names.
+
+Wed Jul 13 13:19:12 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_9.c, enc/windows_1254.c: Split Windows-1254 from
+	  ISO-8859-9 to be able to implement different case conversions.
+
+Wed Jul 13 13:08:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_7.c, enc/windows_1253.c: Split Windows-1253 from
+	  ISO-8859-7 to be able to implement different case conversions.
+
+Wed Jul 13 10:50:12 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_13.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-13, by Kanon Shindo.
+
+Wed Jul 13 10:31:39 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_13.c, enc/windows_1257.c: Split Windows-1257 from
+	  ISO-8859-13 to be able to implement different case conversions.
+
+Wed Jul 13 09:02:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_3.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-3, by Takuya Miyamoto.
+
+	* test/ruby/enc/test_case_comprehensive.rb: Extend special treatment
+	  for Turkic.
+
+	* enc/iso_8859_3.c: Exclude dotless i/I with dot from case-insensitive
+	  matching because they are not a case pair.
+
+Wed Jul 13 08:40:21 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_iso_8859.rb: Exclude dotless i/I with dot from
+	  case-insensitive matching because they are not a case pair.
+
+Tue Jul 12 23:13:43 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (rb_math_sqrt): [EXPERIMENTAL] move Complex sqrt support
+	  from mathn.rb.
+
+Tue Jul 12 01:25:09 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* configure.in, lib/mkmf.rb, win32/Makefile.sub (CSRCFLAG): make the
+	  compiler option replacable in Makefile.
+
+	* win32/Makefile.sub (OUTFLAG, COUTFLAG): ditto.
+
+	* win32/Makeile.sub, win32/setup.mak (CC): should not append `-nologo`
+	  option forcely.
+
+Mon Jul 11 18:05:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_1.c: Moved test for lowercase characters without
+	  uppercase equivalent.
+
+Mon Jul 11 17:49:25 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_4.c, enc/iso_8859_10.c, enc/iso_8859_14.c,
+	  enc/iso_8859_15.c, enc/iso_8859_16.c: Replace case-by-case code with
+	  lookup in ENC_ISO_8859_xx_TO_LOWER_CASE table.
+
+Mon Jul 11 16:00:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_each, strio_readlines): convert
+	  arguments just once before reading, instead of conversions for
+	  each lines, as r55603.
+
+Sun Jul 10 19:53:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_10.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-10, by Toya Hosokawa.
+
+Sun Jul 10 19:33:47 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Changed testing logic to
+	  catch unintended modifications of characters that do not have a case
+	  equivalent in the respective encoding.
+	* enc/iso_8859_1.c, enc/iso_8859_15.c: Fixed unintended modifications of
+	  micro sign and y with diaeresis.
+
+Sun Jul 10 17:05:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_4.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-4, by Kotaro Yoshida.
+
+Sun Jul 10 16:17:47 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Fixed a comment
+	  (message belongs to last commit). [ci skip]
+
+Sun Jul 10 14:27:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_dtoa): [EXPERIMENTAL] adjust the case that the
+	  Float value is close to the exact but unrepresentable middle
+	  value of two values in the given precision, as r55604.
+
+Sun Jul 10 08:57:20 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* thread.c: Fixed implicit conversion error with Apple clang-800.0.31
+
+Sat Jul  9 12:43:09 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/getoptlong.rb: use false instead of FALSE.
+
+Fri Jul  8 21:49:28 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* thread.c (rb_wait_for_single_fd): Clean up fds.revents every time
+	  before calling ppoll(2). [Bug #12575] [ruby-dev:49725]
+
+Fri Jul  8 14:16:48 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_args.c (vm_caller_setup_arg_block): call rb_sym_to_proc()
+	  directly to reduce method dispatch overhead.
+
+Fri Jul  8 08:43:31 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* io.c (rb_io_s_read): add description of pipes to the documentation
+	  of IO.read.
+
+Fri Jul  8 03:54:22 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_args.c (setup_parameters_complex): don't raise ArgumentError
+	  if an array is given for instance_exec with optional argument.
+	  [ruby-core:76300] [Bug #12568]
+	  https://github.com/rails/rails/pull/25699
+
+Fri Jul  8 00:47:36 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_eval.c (yield_under): change prototype to get argc/argv.
+
+	* vm_eval.c (specific_eval): change for above.
+
+	* vm_eval.c (rb_obj_instance_exec): avoid object allocation.
+
+	* vm_eval.c (rb_mod_module_exec): ditto.
+
+Thu Jul  7 20:08:37 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm_args.c (vm_caller_setup_arg_block): disable symbol block
+	  argument optimization when tail call optimization is enabled,
+	  in order to avoid SEGV.  [ruby-core:76288] [Bug #12565]
+
+Thu Jul  7 16:37:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_round): [EXPERIMENTAL] adjust the case that the
+	  receiver is close to the exact but unrepresentable middle value
+	  of two values in the given precision.
+	  http://d.hatena.ne.jp/hnw/20160702
+
+Thu Jul  7 16:31:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_io_s_foreach, rb_io_s_readlines): convert arguments
+	  just once before reading, instead of conversions for each lines.
+
+Wed Jul  6 19:54:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_14.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-14, by Yutaro Tada.
+
+Wed Jul  6 19:24:48 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_1.c, enc/iso_8859_15.c, enc/iso_8859_16.c:
+	  Align indenting to onigmo convention.
+
+Wed Jul  6 18:59:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_15.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-15, by Maho Harada.
+
+Wed Jul  6 18:34:21 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_16.c, test/ruby/enc/test_case_comprehensive.rb:
+	  Implement non-ASCII case conversion for ISO-8859-16, by Satoshi Kayama.
+
+Wed Jul  6 14:44:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (create_makefile): store $headers in LOCAL_HDRS for
+	  depend files.
+
+	* ext/digest/digest_conf.rb (digest_conf): add implementation
+	  specific headers to $header.
+
+	* ext/digest/{md5,rmd160,sha1,sha2}/depend: add LOCAL_HDRS to the
+	  dependencies.
+
+Wed Jul  6 08:59:35 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/http/generic_request.rb (write_header): A Request-Line must
+	  not contain CR or LF.
+
+Wed Jul  6 07:11:27 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/ftp.rb (putline): raise an ArgumentError when
+	  CR or LF is included in a line.
+
+Tue Jul  5 20:49:30 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/json/*, test/json/*: Update json-2.0.1.
+	  Changes of 2.0.0: https://github.com/flori/json/blob/f679ebd0c69a94e3e70a897ac9a229f5779c2ee1/CHANGES.md#2015-09-11-200
+	  Changes of 2.0.1: https://github.com/flori/json/blob/f679ebd0c69a94e3e70a897ac9a229f5779c2ee1/CHANGES.md#2016-07-01-201
+	  [Feature #12542][ruby-dev:49706][fix GH-1395]
+
+Tue Jul  5 19:39:49 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c (rb_str_change_terminator_length): New function to change
+	  termlen and resize heap for the terminator. This is split from
+	  rb_str_fill_terminator (str_fill_term) because filling terminator
+	  and changing terminator length are different things. [Bug #12536]
+
+	* internal.h: declaration for rb_str_change_terminator_length.
+
+	* string.c (str_fill_term): Simplify only to zero-fill the terminator.
+	  For non-shared strings, it assumes that (capa + termlen) bytes of
+	  heap is allocated. This partially reverts r55557.
+
+	* encoding.c (rb_enc_associate_index): rb_str_change_terminator_length
+	  is used, and it should be called whenever the termlen is changed.
+
+	* string.c (str_capacity): New static function to return capacity
+	  of a string with the given termlen, because the termlen may
+	  sometimes be different from TERM_LEN(str) especially during
+	  changing termlen or filling terminator with specific termlen.
+
+	* string.c (rb_str_capacity): Use str_capacity.
+
+Tue Jul  5 11:07:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* pack.c (pack_pack): use union instead of bare variable to ease
+	  optimizations and avoid assigning x87 floating point number.
+	  [ruby-core:74496] [Bug #12209]
+
+	* pack.c (pack_unpack): ditto.
+
+Mon Jul  4 13:56:34 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* process.c: define sig_t if not exist.
+	  at least Solaris 10 and 11 doesn't have sig_t.
+
+Mon Jul  4 13:08:48 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* random.c (random_ulong_limited): avoid left shift count >= width of
+	  type on 32bit environment.
+
+Sun Jul  3 18:51:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb, test_regex_casefold.rb,
+	  test/test_unicode_normalize.rb: Replace UNICODE_VERSION from
+	  UnicodeNormalize with RbConfig::CONFIG['UNICODE_VERSION'] from
+	  feature 12460.
+
+Sun Jul  3 06:04:09 2016  Eric Wong  <e@80x24.org>
+
+	* process.c (disable_child_handler_fork_child): simplify
+	  [ruby-core:75781] [Misc #12439]
+
+Sun Jul  3 05:25:46 2016  Eric Wong  <e@80x24.org>
+
+	* tool/asm_parse.rb: add description
+	* tool/change_maker.rb: ditto
+	* tool/downloader.rb: ditto
+	* tool/eval.rb: ditto
+	* tool/expand-config.rb: ditto
+	* tool/extlibs.rb: ditto
+	* tool/fake.rb: ditto
+	* tool/file2lastrev.rb: ditto
+	* tool/gem-unpack.rb: ditto
+	* tool/gen_dummy_probes.rb: ditto
+	* tool/gen_ruby_tapset.rb: ditto
+	* tool/generic_erb.rb: ditto
+	* tool/id2token.rb: ditto
+	* tool/ifchange: ditto
+	* tool/insns2vm.rb: ditto
+	* tool/instruction.rb: ditto
+	* tool/jisx0208.rb: ditto
+	* tool/merger.rb: ditto
+	* tool/mkrunnable.rb: ditto
+	* tool/node_name.rb: ditto
+	* tool/parse.rb: ditto
+	* tool/rbinstall.rb: ditto
+	* tool/rbuninstall.rb: ditto
+	* tool/rmdirs: ditto
+	* tool/runruby.rb: ditto
+	* tool/strip-rdoc.rb: ditto
+	* tool/vcs.rb: ditto
+	* tool/vtlh.rb: ditto
+	* tool/ytab.sed: ditto
+	* tool/enc-unicode.rb: fix typo
+	* tool/mk_call_iseq_optimized.rb: ditto
+	* tool/update-deps: ditto
+	  [ruby-core:76215] [Bug #12539]
+	  by Noah Gibbs <the.codefolio.guy@gmail.com>
+
+Sat Jul  2 18:04:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/mailto.rb (initialize): RFC3986_Parser#split sets opaque
+	  only if the URI has path-rootless, not path-empty.
+	  [ruby-core:76055] [Bug #12498]
+	  patched by Chris Heisterkamp <cheister@squareup.com>
+
+Sat Jul  2 04:26:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (noname_disable_map): don't optimize out group 0
+	  Ruby's Regexp doesn't allow normal numbered groups if the regexp
+	  has named groups. On such case it optimizes out related NT_ENCLOSE.
+	  But even on the case it can use \g<0>.
+	  This fix not to remove NT_ENCLOSE whose regnum is 0.
+	  [ruby-core:75828] [Bug #12454]
+
+Sat Jul  2 03:09:27 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c: Partially reverts r55547 and r55555.
+	  ChangeLog about the reverted changes are also deleted in this file.
+	  [Bug #12536] [ruby-dev:49699] [ruby-dev:49702]
+
+Sat Jul  2 02:22:22 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c (str_fill_term): When termlen increases, re-allocation
+	  of memory for termlen should always be needed.
+	  In this fix, if possible, decrease capa instead of realloc.
+	  [Bug #12536] [ruby-dev:49699]
+
+Fri Jul  1 20:20:20 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c: Specify termlen as far as possible.
+	  Additional fix for [Bug #12536] [ruby-dev:49699].
+
+	* string.c (str_new_static): Specify termlen from the given encoding
+	  when creating a new String object is needed.
+
+	* string.c (rb_tainted_str_new_with_enc): New function to create a
+	  tainted String object with the given encoding. This means that
+	  the termlen is correctly specified. Currently static function.
+	  The function name might be renamed to rb_tainted_enc_str_new
+	  or rb_enc_tainted_str_new.
+
+	* string.c (rb_external_str_new_with_enc): Use encoding by using the
+	  above rb_tainted_str_new_with_enc().
+
+Fri Jul  1 19:38:57 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* test/fiddle/test_pointer.rb (test_to_str, test_to_s, test_aref_aset):
+	  Attempt to use independent strings for destructive tests that
+	  directly modify values on memory by using Fiddle::Pointer.
+	  [Bug #12537] [ruby-dev:49700]
+
+Fri Jul  1 18:20:00 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .gdbinit (rb_ps_thread): show the detail of cfunc in ruby level
+	  backtrace.
+
+Fri Jul  1 13:26:39 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c (rb_str_subseq, str_substr): When RSTRING_EMBED_LEN_MAX
+	  is used, TERM_LEN(str) should be considered with it because
+	  embedded strings are also processed by TERM_FILL.
+	  Additional fix for [Bug #12536] [ruby-dev:49699].
+
+Fri Jul  1 12:11:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .gdbinit (rb_count_objects): added gdb version of count_objects().
+
+Fri Jul  1 04:32:52 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* .gdbinit (rb_ps_thread): show ruby level backtrace.
+	  Usually you can call `rb_ps` to show ruby level backtraces
+	  for all living threads.
+	  Note that it can call with core file like `gcore <pid>`
+	  and `gdb ruby core.<pid>`.
+
+Thu Jun 30 19:15:13 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* string.c: Fix memory corruptions when using UTF-16/32 strings.
+	  [Bug #12536] [ruby-dev:49699]
+
+	* string.c (rb_str_new_with_class): Use TERM_LEN of the "obj".
+
+	* string.c (rb_str_plus, rb_str_justify): Use str_new0 which is aware
+	  of termlen.
+
+	* string.c (str_shared_replace): Copy +termlen bytes instead of +1.
+
+	* string.c (rb_str_times): termlen should not be included in capa.
+
+	* string.c (RESIZE_CAPA_TERM): When using RSTRING_EMBED_LEN_MAX,
+	  termlen should be counted with it because embedded strings are
+	  also processed by TERM_FILL.
+
+	* string.c (rb_str_capacity, str_shared_replace, str_buf_cat): ditto.
+
+	* string.c (rb_str_drop_bytes, rb_str_setbyte, str_byte_substr): ditto.
+
+Wed Jun 29 22:24:37 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/lib/psych_jars.rb: removed needless file required to JRuby.
+
+Wed Jun 29 22:21:38 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c: The "reuse" behavior of d2i_ functions does
+	  not work well with OpenSSL 1.0.0t. So avoid it.
+
+Wed Jun 29 15:18:28 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (opt_succ): optimize like r55515. (but this argument is
+	  constant)
+
+Wed Jun 29 12:41:58 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* test/ruby/test_refinement.rb: skip
+	  test_prepend_after_refine_wb_miss on ARM or MIPS.
+	  [ruby-core:76031] [Bug #12491]
+
+Wed Jun 29 08:45:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (passed_block): convert passed block symbol to proc.
+	  based on the patch by Daisuke Sato in [ruby-dev:49695].
+	  [Bug #12531]
+
+Wed Jun 29 03:34:41 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bignum.c (rb_big2ulong): the old logic seems to try to avoid
+	  calculating `-(long)(num-1)-1` if `num` is not LONG_MIN. (Note that
+	  `-LONG_MIN` may be larger than LONG_MAX) But C compilers can
+	  optimize it into single NEG instruction.
+	  Therefore those two conditions can be single if-body.
+
+	* bignum.c (rb_big2long): ditto.
+
+	* bignum.c (rb_big2ull): ditto.
+
+	* bignum.c (rb_big2ll): ditto.
+
+Tue Jun 28 22:55:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/pstore.rb (PStore::CHECKSUM_ALGO): extract the algorithm for
+	  checksum, instead of qualified names for each times.
+
+Tue Jun 28 22:29:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bootstraptest/runner.rb: do not use safe navigation operator.
+	  this runner may run on older ruby.  partially revert r53110
+	  (GH-1142 patched by @mlarraz).
+
+Tue Jun 28 22:09:09 2016  Akio Tajima <artonx@yahoo.co.jp>
+
+	* lib/fileutils.rb: rescue Errno:EACCES for chown.
+	  [Bug #12520]
+
+Tue Jun 28 18:38:09 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* ext/digest/md5/md5ossl.h: Remove excess semicolons.
+	  Suppress warning on Solaris with Oracle Solaris Studio 12.
+	  [ruby-dev:49692] [Bug #12524]
+
+	* ext/digest/md5/md5cc.h: ditto.
+	* ext/digest/sha1/sha1cc.h: ditto.
+	* ext/digest/sha1/sha1ossl.h: ditto.
+	* ext/digest/sha2/sha2cc.h: ditto.
+	* ext/digest/sha2/sha2ossl.h: ditto.
+	* ext/openssl/ossl_pkey_rsa.c: ditto.
+
+Tue Jun 28 15:56:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/ruby/enc/test_case_comprehensive.rb: noting to test if
+	  Unicode data files are available.
+	  [ruby-core:76160] [Bug #12433]
+
+	* test/test_unicode_normalize.rb: ditto.
+
+Tue Jun 28 15:20:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/net/http.rb (Net::HTTP#proxy_uri): cache the case no proxy
+	  is used.
+
+Tue Jun 28 09:56:29 2016  Stefan Schussler  <mail@stefanschuessler.de>
+
+	* object.c (rb_mod_eqq): [DOC] Fix typo in RDoc.  [Fix GH-1393]
+
+Tue Jun 28 02:41:32 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (opt_plus): use `- 1` instead of `& (~1)` to allow
+	  compilers to use x86 LEA instruction (3 operand).
+	  Even if 3 operand LEA's latency is 3 cycle after SandyBridge,
+	  it reduces code size and can be faster because of super scalar.
+
+	* insns.def (opt_plus): calculate and use rb_int2big.
+	  On positive Fixnum overflow, `recv - 1 + obj` doesn't carry
+	  because recv's msb and obj's msb are 0, and resulted msb is 1.
+	  Therefore simply rshift and cast as signed long works fine.
+	  On negative Fixnum overflow, it will carry because both arguments'
+	  msb are 1, and resulted msb is also 1.
+	  In this case it needs to restore carried sign bit after rshift.
+
+Mon Jun 27 16:58:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb (FileUtils#install): accecpt symbolic mode, as
+	  well as chmod.
+
+	* lib/fileutils.rb (FileUtils#install): add owner and group
+	  options.
+
+Mon Jun 27 08:56:55 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (ADD_TRACE): ignore trace instruction on non-positive
+	  line.
+
+	* parse.y (coverage): get rid of ArgumentError when the starting
+	  line number is not positive.  [ruby-core:76141] [Bug #12517]
+
+Sun Jun 26 10:20:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32/lib/Win32API.rb (Win32API#initialize): Cygwin
+	  2.5.2-1 (perhaps) seems to no longer append ".dll" suffix
+	  implicitly.
+
+	* ext/win32/lib/win32/resolv.rb (Win32::Resolv): ditto.  Fix the
+	  error reported by yamataka AT u08.itscom.net in
+	  [ruby-list:50339], and pointed out and patched by cerberus AT
+	  m3.kcn.ne.jp in [ruby-list:50341].
+
+Sat Jun 25 10:07:52 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* test/openssl/test_ocsp.rb: Ignore errors caused by bugs that exist in
+	  LibreSSL >= 2.3.1.
+
+Sat Jun 25 02:33:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (vm_respond_to): try method_missing if respond_to?
+	  is undefined, as if it is the default definition.
+	  [ruby-core:75377] [Bug #12353]
+
+Fri Jun 24 17:04:21 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/psych/*, test/psych/*: Update psych 2.1.0
+	  This version fixed [Bug #11988][ruby-core:72850]
+
+Fri Jun 24 13:12:41 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*:
+	  Update rubygems 2.6.5 and 2.6.6.
+	  Release note of 2.6.5: https://github.com/rubygems/rubygems/commit/656f5d94dc888d78d0d00f3598a4fa37391aac80
+	  Release note of 2.6.6: https://github.com/rubygems/rubygems/commit/ccb9c3300c063f5b5656669972d24a10ef8afbf5
+
+Fri Jun 24 09:17:15 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (lib/unicode_normalize/tables.rb): should not depend
+	  on Unicode data files unless ALWAYS_UPDATE_UNICODE=yes, to get
+	  rid of downloading Unicode data unnecessary.  [ruby-dev:49681]
+
+	* common.mk (enc/unicode/casefold.h): update Unicode files in a
+	  sub-make, not to let the header depend on the files always.
+
+	* enc/unicode/case-folding.rb: if gperf is not usable, assume the
+	  existing file is OK.
+
+Tue Jun 21 19:44:54 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_regex_casefold.rb: Add Windows-1251, KOI8-R, and
+	  KOI8-U to encodings; definitely removed EUC-JP.
+
+Tue Jun 21 19:32:23 2016  Mark St.Godard  <markstgodard@gmail.com>
+
+	* lib/webrick/httprequest.rb (setup_forwarded_info): Use the first
+	  value in X-Forwarded-Proto, if header contains multiple comma
+	  separated values.  Some middlewares may add these values to the
+	  list, not replacing.  [Fix GH-1386]
+
+Tue Jun 21 17:17:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_io.rb: Skip test_readpartial_with_not_empty_buffer,
+	  test_read_buffer_error, test_read_unlocktmp_ensure,
+	  test_readpartial_unlocktmp_ensure, and
+	  test_sysread_unlocktmp_ensure on cygwin,
+	  because these tests repeatedly hang. This makes test_io.rb
+	  complete in finite time on cygwin.
+
+	* ChangeLog: Fix test_in.rb -> test_io.rb (two instances).
+
+Tue Jun 21 16:38:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_casemap): do not put code with side effects
+	  inside RSTRING_PTR() macro which evaluates the argument multiple
+	  times.
+
+Tue Jun 21 16:13:45 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_casemap): fix memory leak.
+
+Tue Jun 21 16:12:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_casemap): int is too small for string size.
+
+Tue Jun 21 15:42:22 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_io.rb: Skip test_read_buffer_error on cygwin,
+	  because this test repeatedly hangs.
+
+Tue Jun 21 15:35:14 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* LEGAL: Added filenames and copyrights for some files in
+	  enc/trans/JIS.
+
+Tue Jun 21 00:56:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (get_special_folder): fix calling convention of
+	  SHGetPathFromIDListEx, which should be WINAPI.  pointed out by
+	  @arton at http://twitter.com/arton/status/744884064277016576
+
+Tue Jun 21 00:22:02 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, include/ruby/defines.h (RUBY_USE_SETJMPEX):
+	  include setjmpex.h only when using setjmpex() for RUBY_SETJMP.
+	  the header of mingw32 overrides setjmp() by setjmpex().
+
+Mon Jun 20 18:39:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_io.rb: Skip test_open_fifo_does_not_block_other_threads
+	  on cygwin. Fifos seem to work okay in cygwin, but this test repeatedly
+	  hangs.
+
+Mon Jun 20 13:35:06 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* vm.c (invoke_bmethod, invoke_block_from_c_0): revert r52104
+	  partially to avoid "self has wrong type to call super in this
+	  context" errors.
+	  [ruby-core:72724] [Bug #11954]
+
+Mon Jun 20 12:53:38 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* test/rubygems/test_gem_remote_fetcher.rb: OpenSSL::PKey::DH#priv_key=
+	  is not defined when ext/openssl is built with OpenSSL 1.1.0.
+	  https://github.com/rubygems/rubygems/pull/1648
+	  [ruby-core:75225] [Feature #12324]
+
+Sun Jun 19 21:25:43 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c: Add OCSP::SingleResponse that represents an
+	  OCSP SingleResponse structure. Also add two new methods #responses
+	  and #find_response to OCSP::BasicResponse. A BasicResponse has one or
+	  more SingleResponse. We have OCSP::BasicResponse#status that returns
+	  them as an array of arrays, each containing the content of a
+	  SingleResponse, but this is not useful. When validating an OCSP
+	  response, we need to look into the each SingleResponse and check their
+	  validity but it is not simple. For example, when validating for a
+	  certificate 'cert', the code would be like:
+
+	    # certid_target is an OpenSSL::OCSP::CertificateId for cert
+	    basic = res.basic
+	    result = basic.status.any? do |ary|
+	      ary[0].cmp(certid_target) &&
+	        ary[4] <= Time.now && (!ary[5] || Time.now <= ary[5])
+	    end
+
+	  Adding OCSP::SingleResponse at the same time allows exposing
+	  OCSP_check_validity(). With this, the code above can be rewritten as:
+
+	    basic = res.basic
+	    single = basic.find_response(certid_target)
+	    result = single.check_validity
+
+	* test/openssl/test_ocsp.rb: Test this.
+
+Sun Jun 19 18:40:19 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspbres_add_status): Allow specifying
+	  the times (thisUpdate, nextUpdate and revocationTime) with Time
+	  objects. Currently they accepts only relative seconds from the current
+	  time. This is inconvenience, especially for revocationTime. When
+	  Integer is passed, they are still treated as relative times. Since the
+	  type check is currently done with rb_Integer(), this is a slightly
+	  incompatible change. Hope no one passes a relative time as String or
+	  Time object...
+	  Also, allow passing nil as nextUpdate. It is optional.
+
+	* ext/openssl/ruby_missing.h: Define RB_INTEGER_TYPE_P() if not defined.
+	  openssl gem will be released before Ruby 2.4.0.
+
+Sun Jun 19 18:39:38 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c: Implement OCSP::{CertificateId,Request,
+	  BasicResponse,Response}#initialize_copy.
+	  [ruby-core:75504] [Bug #12381]
+
+	* test/openssl/test_ocsp.rb: Test them.
+
+Sun Jun 19 18:29:50 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkey_dh.c, ext/openssl/ossl_pkey_dsa.c,
+	  ext/openssl/ossl_pkey_ec.c, ext/openssl/ossl_pkey_rsa.c: Implement
+	  initialize_copy method for OpenSSL::PKey::*.
+	  [ruby-core:75504] [Bug #12381]
+
+	* test/openssl/test_pkey_dh.rb, test/openssl/test_pkey_dsa.rb,
+	  test/openssl/test_pkey_ec.rb, test/openssl/test_pkey_rsa.rb: Test they
+	  actually copy the OpenSSL objects, and modifications to cloned object
+	  don't affect the original object.
+
+Sun Jun 19 16:55:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/test_dir_m17n.rb: Skip tests with non-UTF-8 encodings
+	  on cygwin. Cygwin can use the Unicode PUA (private use area) to store
+	  bytes from non-UTF-8 filenames (see
+	  https://cygwin.com/cygwin-ug-net/using-specialnames.html#pathnames-specialchars),
+	  but we are not supporting this. [Bug #12443]
+
+Sun Jun 19 15:01:18 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* localeinit.c: Fix filesystem encoding for cygwin to UTF-8 (see
+	  https://cygwin.com/cygwin-ug-net/using-specialnames.html#pathnames-unusual)
+
+Sun Jun 19 14:31:07 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkey.h, ext/openssl/ossl_pkey_dh.c,
+	  ext/openssl/ossl_pkey_dsa.c, ext/openssl/ossl_pkey_rsa.c: A few days
+	  ago, OpenSSL changed {DH,DSA,RSA}_get0_*() to take const BIGNUM **.
+	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=fd809cfdbd6e32b6b67b68c59f6d55fbed7a9327
+	  [ruby-core:75225] [Feature #12324]
+
+Sun Jun 19 11:19:43 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_path_to_class): consider the string length
+	  instead of a terminator.
+
+	* variable.c (rb_path_to_class): search the constant at once
+	  instead of checking if defined and then getting it.
+
+Sat Jun 18 14:01:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/rubygems/test_gem_installer.rb: Fixed broken test with extension
+	  build. https://github.com/rubygems/rubygems/pull/1645
+
+Sat Jun 18 13:59:54 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems
+	  HEAD(2c6d256). It contains to update vendored Molinillo to 0.5.0.
+	  https://github.com/rubygems/rubygems/pull/1638
+
+Sat Jun 18 10:13:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (build-ext), ext/extmk.rb: use variable EXTENCS
+	  different than ENCOBJS, to get rid of circular dependency.
+	  build libencs when linking encodings statically.
+	  [ruby-core:75618] [Bug #12401]
+
+Sat Jun 18 08:52:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_getline): fix pointer index
+	  overflow.  reported by Guido Vranken <guido AT guidovranken.nl>.
+
+Thu Jun 16 16:35:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (Init_class_hierarchy): prevent rb_cObject which is the
+	  class tree root, from GC.  [ruby-dev:49666] [Bug #12492]
+
+Thu Jun 16 12:17:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (tr_trans): adjust buffer size by processed and rest
+	  lengths, instead of doubling repeatedly.
+
+Thu Jun 16 11:15:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (tr_trans): consider terminator length and fix heap
+	  overflow.  reported by Guido Vranken <guido AT guidovranken.nl>.
+
+Thu Jun 16 00:02:32 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_verify, ossl_ocspbres_verify):
+	  Use ossl_clear_error() so that they don't print warnings to stderr and
+	  leak errors in the OpenSSL error queue. Also, check the return value
+	  of OCSP_*_verify() correctly. They can return -1 on verification
+	  failure.
+
+Wed Jun 15 19:52:23 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspreq_sign, ossl_ocspbres_sign): Allow
+	  specifying hash algorithm used in signing. They are hard coded to use
+	  SHA-1.
+	  Based on a patch provided by Tim Shirley <tidoublemy@gmail.com>.
+	  [ruby-core:70915] [Feature #11552] [GH ruby/openssl#28]
+
+	* test/openssl/test_ocsp.rb: Test sign-verify works.
+
+Wed Jun 15 01:46:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c: [DOC] fix rdoc directive, and an example of negative
+	  value.  [ruby-core:76025] [Bug #12487]
+
+Wed Jun 15 01:44:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb: provide Unicode Version information as
+	  RbConfig::CONFIG['UNICODE_VERSION'].
+	  [ruby-core:75845] [Feature #12460]
+
+Wed Jun 15 00:01:18 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* test/ruby/enc/test_case_comprehensive.rb
+	  (TestComprehensiveCaseFold::read_data): use \A and \z instead of
+	  ^ and $ in regexp.
+
+Tue Jun 14 23:43:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/backward.h (rb_cFixnum, rb_cBignum): remove the
+	  backward compatibility macros, to fail incompatible extension
+	  libraries early.  [Bug #12427]
+
+Tue Jun 14 22:22:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): limit the result string
+	  size by the format length, to get rid of unlimited memory use.
+
+Tue Jun 14 22:11:11 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspcid_get_issuer_name_hash,
+	  ossl_ocspcid_get_issuer_key_hash, ossl_ocspcid_get_hash_algorithm):
+	  Add accessor methods OCSP::CertificateId#issuer_name_hash,
+	  #issuer_key_hash, #hash_algorithm.
+	  Based on a patch provided by Paul Kehrer <paul.l.kehrer@gmail.com>.
+	  [ruby-core:48062] [Feature #7181]
+
+	* test/openssl/test_ocsp.rb: Test these new methods.
+
+Tue Jun 14 22:07:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_strftime.c (date_strftime_with_tmx): reject too
+	  large precision to get rid of buffer overflow.
+	  reported by Guido Vranken <guido AT guidovranken.nl>.
+
+Tue Jun 14 21:40:42 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ocsp.c (ossl_ocspbres_to_der, ossl_ocspcid_to_der):
+	  Implement #to_der methods for OCSP::BasicResponse and
+	  OCSP::CertificateId.
+
+	  (ossl_ocspreq_initialize, ossl_ocspres_initialize): Use GetOCSP*()
+	  instead of raw DATA_PTR().
+
+	  (ossl_ocspbres_initialize, ossl_ocspcid_initialize): Allow
+	  initializing from DER string.
+
+	  (Init_ossl_ocsp): Define new #to_der methods.
+
+	* test/openssl/test_ocsp.rb: Test these changes. Also add missing tests
+	  for OCSP::{Response,Request}#to_der.
+
+Tue Jun 14 21:35:00 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/openssl_missing.h (DH_set0_pqg, RSA_set0_key):
+	  DH_set0_pqg() allows 'q' to be NULL. Fix a typo in RSA_set0_key().
+	  Fixes r55285.  [ruby-core:75225] [Feature #12324]
+
+Mon Jun 14 10:19:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* NEWS: describe Integer#digits.
+
+Mon Jun 13 21:09:40 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread.c (debug_deadlock_check): show thread lock dependency and
+	  backtrace [Feature #8214] [ruby-dev:47217]
+
+	* thread.c (thread_status_name): show "sleep_forever" instead of
+	  "sleep" if called from inspect.
+
+Mon Jun 13 20:50:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (reg_named_capture_assign_iter): remove named capture
+	  conflict warnings.  it is just annoying rather than useful.
+	  [ruby-core:75416] [Bug #12359]
+
+Mon Jun 13 20:04:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (rb_int_digits, rb_fix_digits, rb_int_digits_bigbase):
+	  Add Integer#digits to extract columns in place-value notation
+	  [Feature #12447] [ruby-core:75799]
+
+	* test/ruby/test_integer.rb: Add tests for the above change.
+
+	* test/ruby/test_bignum.rb: ditto.
+
+Mon Jun 13 20:34:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RUBY_INTEGER_UNIFICATION): macro to tell if
+	  Integer is integrated.  [ruby-core:75718][Bug #12427]
+
+	* include/ruby/backward.h, internal.h (rb_cFixnum, rb_cBignum):
+	  fallback to rb_cInteger.
+
+	* bignum.c, numeric.c, ext/json/generator/generator.{c,h}: use the
+	  macro.
+
+Mon Jun 13 16:58:53 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Add tests for full Unicode
+	  swapcase.
+
+Sun Jun 12 14:48:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Add tests for ASCII-only
+	  swapcase; store calculated values in hashes.
+
+Sun Jun 12 14:05:45 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_get_auth_tag,
+	  ossl_cipher_set_auth_tag): Check if the cipher flags retrieved by
+	  EVP_CIPHER_CTX_flags() includes EVP_CIPH_FLAG_AEAD_CIPHER to see if
+	  the cipher supports AEAD. AES-GCM was the only supported in OpenSSL
+	  1.0.1.
+
+	  (Init_ossl_cipher): Fix doc; OpenSSL::Cipher::AES.new(128, :GCM) can't
+	  work.
+
+	* ext/openssl/openssl_missing.h: Define EVP_CTRL_AEAD_{GET,SET}_TAG if
+	  missing. They are added in OpenSSL 1.1.0, and have the same value as
+	  EVP_CTRL_GCM_{GET,SET}_TAG and EVP_CTRL_CCM_{GET,SET}_TAG.
+
+Sun Jun 12 13:47:42 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* test/openssl/test_engine.rb (test_openssl_engine_builtin,
+	  test_openssl_engine_by_id_string): Skip test if 'openssl' engine is
+	  already loaded. And test the number increased by Engine.load{_by_id,},
+	  not the total count of loaded engines. Previously, we called
+	  OpenSSL::Engine.cleanup every time running a test case, but we no
+	  longer can do it.
+	  [ruby-core:75225] [Feature #12324]
+
+Sun Jun 12 09:24:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (append_fspath): normalize directory name to be appended
+	  on OS X.  [ruby-core:75957] [Ruby trunk Bug#12483]
+	  https://github.com/rails/rails/issues/25303#issuecomment-224834804
+
+Sat Jun 11 23:07:32 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* lib/forwardable.rb (_delegator_method): get rid of a warning which
+	  causes test failures introduced at r55376.
+
+Sat Jun 11 18:37:58 2016  Marcus Stollsteimer  <sto.mar@web.de>
+
+	* ext/json/lib/*.rb: Removed some comments. Because these are unnecessary
+	  class description. [ci skip][Bug #12255][ruby-core:74835]
+
+Sat Jun 11 15:19:38 2016  takiy33  <takiy33@users.noreply.github.com>
+
+	* lib/net/smtp.rb: [DOC] Remove dead link on documentation for
+	  Japanese of SMTP.  [Fix GH-1380]
+
+Sat Jun 11 15:02:45 2016  Grant Hutchins  <github@nertzy.com>
+
+	* string.c (rb_str_oct): [DOC] fix typo, hornored -> honored.
+	  [Fix GH-1379]
+
+Sat Jun 11 14:04:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/objspace/objspace_dump.c: generate valid JSON for dump_all.
+
+Sat Jun 11 13:52:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/forwardable.rb (_delegator_method): remove __send__ call if
+	  possible, so that more optimizations will be enabled.
+
+Sat Jun 11 11:24:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/iso_8859.h (SHARP_s): name frequently used codepoint.
+
+Sat Jun 11 09:58:45 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_1.c: Revert to older version of code.
+
+Sat Jun 11 09:46:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/iso_8859_1.c: Implement non-ASCII case mapping.
+
+	* test/ruby/enc/test_case_comprehensive.rb: Tests for above.
+
+	* string.c: Add iso-8859-1 to supported encodings.
+
+Sat Jun 11 09:31:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/forwardable.rb (_delegator_method): leave the backtrace
+	  untouched during accessor.  forwardable.rb does not appear in
+	  the backtrace during delegated method because of tail-call
+	  optimization.
+
+Sat Jun 11 01:38:31 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/defines.h (GCC_VERSION_SINCE): Fix logic error by
+	  adding parentheses. Fix failures of TestMkmf::TestConvertible
+	  with GCC 3.4.3 on Solaris 10. [Bug #12479] [ruby-dev:49660]
+
+Fri Jun 10 21:54:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/forwardable.rb (Forwardable._delegator_method): extract
+	  method generator and deal with non-module objects.
+	  [ruby-dev:49656] [Bug #12478]
+
+Fri Jun 10 17:35:11 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Special-case :ascii option in rb_str_capitalize_bang and
+	  rb_str_swapcase_bang.
+
+Fri Jun 10 17:12:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Special-case :ascii option in rb_str_upcase_bang (retry).
+
+Fri Jun 10 14:48:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* hash.c (get_env_cstr): ensure NUL-terminated.
+	  [ruby-dev:49655] [Bug #12475]
+
+	* string.c (rb_str_fill_terminator): return the pointer to the
+	  NUL-terminated content.
+
+Thu Jun  9 21:42:00 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_asn1.c (asn1integer_to_num): Use
+	  ASN1_ENUMERATED_to_BN() to convert an ASN1_ENUMERATED to a BN.
+	  Starting from OpenSSL 1.1.0, ASN1_INTEGER_to_BN() rejects
+	  non-ASN1_INTEGER objects. The format of INTEGER and ENUMERATED are
+	  almost identical so they behaved in the same way in OpenSSL <= 1.0.2.
+	  [ruby-core:75225] [Feature #12324]
+
+	* test/openssl/test_asn1.rb (test_decode_enumerated): Test that it
+	  works.
+
+Thu Jun  9 21:10:04 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* tool/ifchange: fix timestamp error when target without
+	  directory.
+
+Thu Jun  9 19:46:22 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ssl.c: Add define guards for OPENSSL_NO_EC.
+	  SSL_CTX_set_ecdh_auto() is defined even when ECDH is disabled in
+	  OpenSSL's configuration. This fixes r55214.
+
+	* test/openssl/test_pair.rb (test_ecdh_curves): Skip if the OpenSSL does
+	  not support ECDH.
+
+	* test/openssl/utils.rb (start_server): Ignore error in
+	  SSLContext#ecdh_curves=.
+
+Thu Jun  9 18:12:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* Makefile.in (un-runnable): fail with proper error message.
+	  [ruby-core:75905] [Bug #12472]
+
+Thu Jun  9 15:32:17 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (RBCONFIG): use ifchange tool to see if the content is
+	  changed and update the timestamp file.
+
+	* tool/mkconfig.rb: remove ifchange features.
+
+	* tool/ifchange: make target directory if it does not exist with
+	  its parent directories.
+
+	* win32/ifchange.bat: drop support for command.com used on old
+	  systems.
+
+Thu Jun  9 15:03:35 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Check for CRYPTO_malloc() and SSL_new().
+	  OpenSSL_add_all_digests() and SSL_library_init() are deprecated and
+	  converted to macros in OpenSSL 1.1.0.
+	  [ruby-core:75225] [Feature #12324]
+
+Wed Jun  8 23:09:51 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* string.c (rb_str_ascii_casemap): fix compile error.
+
+Wed Jun  8 22:22:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Revert previous commit (possibility of endless loop).
+
+Wed Jun  8 21:57:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Special-case :ascii option in rb_str_upcase_bang.
+
+Wed Jun  8 21:28:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: New static function rb_str_ascii_casemap; special-casing
+	  :ascii option in rb_str_upcase_bang and rb_str_downcase_bang.
+
+	* regenc.c: Fix a bug (wrong use of unnecessary slack at end of string).
+
+	* regenc.h -> include/ruby/oniguruma.h: Move declaration of
+	  onigenc_ascii_only_case_map so that it is visible in string.c.
+
+Wed Jun  8 20:33:44 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/intern.h: Remove excess semicolons in PUREFUNC().
+	  Fix failure of TestMkmf::TestConvertible on Solaris with
+	  Oracle Solaris Studio 12. [ruby-dev:49651] [Bug #12470]
+	* internal.h: ditto.
+
+Wed Jun  8 16:03:09 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* lib/net/smtp.rb (getok, get_response): raise an ArgumentError when
+	  CR or LF is included in a line, because they are not allowed in
+	  RFC5321.  Thanks, Jeremy Daer.
+
+Tue Jun  7 21:27:25 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* test/rubygems/*_{cert,cert_32}.pem: Regenerate test certificates for
+	  OpenSSL 1.1.0. This is already in upstream.
+	  https://github.com/rubygems/rubygems/commit/9be5c53939440a61c4bba73cfffbeb5cfadf72be
+	  [ruby-core:75225] [Feature #12324]
+
+Tue Jun  7 21:27:17 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* test/open-uri/test_ssl.rb: Regenerate test certificates. The test CA
+	  certificate was incorrectly generated. A CA certificate must have the
+	  basic constraints extension with cA bit set to TRUE. OpenSSL <= 1.0.2
+	  allowed the error when the certificate is in the trusted store but
+	  OpenSSL 1.1.0 no longer does.
+	  [ruby-core:75225] [Feature #12324]
+
+Tue Jun  7 21:20:38 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* test/openssl/test_x509name.rb: Don't register OID for 'emailAddress'
+	  and 'serialNumber'. A recent change in OpenSSL made OBJ_create()
+	  reject an already existing OID. They were needed to run tests with
+	  OpenSSL 0.9.6 which is now unsupported.
+	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=52832e470f5fe8c222249ae5b539aeb3c74cdb25
+	  [ruby-core:75225] [Feature #12324]
+
+	* test/openssl/test_ssl_session.rb (test_server_session): Duplicate
+	  SSL::Session before re-adding to the session store. OpenSSL 1.1.0
+	  starts rejecting SSL_SESSION once removed by SSL_CTX_remove_session().
+	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=7c2d4fee2547650102cd16d23f8125b76112ae75
+
+	* test/openssl/test_pkey_ec.rb (setup): Remove X25519 from @keys. X25519
+	  is new in OpenSSL 1.1.0 but this is for key agreement and not for
+	  signing.
+
+	* test/openssl/test_pair.rb, test/openssl/test_ssl.rb,
+	  test/openssl/utils.rb: Set security level to 0 when using aNULL cipher
+	  suites.
+
+	* test/openssl/utils.rb: Use 1024 bits DSA key for client certificates.
+
+	* test/openssl/test_engine.rb: Run each test in separate process.
+	  We can no longer cleanup engines explicitly as ENGINE_cleanup() was
+	  removed.
+	  https://git.openssl.org/gitweb/?p=openssl.git;a=commit;h=6d4fb1d59e61aacefa25edc4fe5acfe1ac93f743
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_cleanup): Add a note to the
+	  RDoc for Engine.cleanup.
+
+	* ext/openssl/lib/openssl/digest.rb: Don't define constants for DSS,
+	  DSS1 and SHA(-0) when using with OpenSSL 1.1.0. They are removed.
+
+	* test/openssl/test_digest.rb, test/openssl/test_pkey_dsa.rb,
+	  test/openssl/test_pkey_dsa.rb, test/openssl/test_ssl.rb,
+	  test/openssl/test_x509cert.rb, test/openssl/test_x509req.rb: Don't
+	  test unsupported hash functions.
+
+Tue Jun  7 17:49:52 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive: Change test for encodings
+	  without any non-ASCII case conversions from ASCII-only test
+	  to full test.
+
+Tue Jun  7 17:18:39 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c (rb_str_upcase_bang, rb_str_capitalize_bang,
+	  rb_str_swapcase_bang): Switch to use primitive.
+
+Tue Jun  7 16:44:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c (rb_str_downcase_bang): Switch to use primitive except if
+	  conversion can be done ASCII-only.
+
+Tue Jun  7 16:13:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive: Add regression tests for
+	  current ASCII-only case conversion.
+
+Tue Jun  7 15:28:38 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive: Fix method name
+	  (generate_casefold_tests -> generate_case_mapping_tests).
+
+Tue Jun  7 15:05:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* regenc.h/c: Rename onigenc_not_support_case_map to
+	  onigenc_ascii_only_case_map.
+
+	* regenc.h: Add definition of onigenc_single_byte_ascii_only_case_map.
+
+	* enc/iso_8859_X.c, windows_125X.c, ascii.c, us-ascii.c, koi8_x.c:
+	  Replace onigenc_not_support_case_map by
+	  onigenc_single_byte_ascii_only_case_map.
+
+	* enc/big5.c, cp949.c, emacs_mule.c, euc_X.c, gbX.c, shift_jis.c,
+	  windows_31j.c: Replace onigenc_not_support_case_map by
+	  onigenc_ascii_only_case_map.
+
+Tue Jun  7 14:57:09 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Check for SSL_CTX_set_min_proto_version()
+	  macro added in OpenSSL 1.1.0. Version-specific methods, such as
+	  TLSv1_method(), are deprecated in OpenSSL 1.1.0. We need to use
+	  version-flexible methods (TLS_*method() or SSLv23_*method()) and
+	  disable other protocol versions as necessary.
+	  [ruby-core:75225] [Feature #12324]
+
+	* ext/openssl/ossl_ssl.c: Use SSL_CTX_set_{min,max}_proto_version() to
+	  fix the protocol version.
+
+Tue Jun  7 12:55:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* regenc.c (onigenc_not_support_case_map): Move to end of file;
+	  (onigenc_single_byte_ascii_only_case_map): Add new function.
+
+Tue Jun  7 09:26:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* regenc.c (onigenc_not_support_case_map): Rewrite to work correctly
+	  in ASCII range.
+
+Mon Jun  6 23:00:00 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* appveyor.yml: Update libressl version to 2.3.5.
+
+Mon Jun  6 18:37:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_throw_start): check if the iseq is symbol
+	  proc, class definition should not be a symbol proc.
+	  [ruby-core:75856] [Bug #12462]
+
+Mon Jun  6 18:36:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Added UTF-16BE/LE and UTF-32BE/LE to supported encodings
+	  for Unicode case mapping.
+
+	* test/ruby/enc/test_case_comprehensive.rb: Tests for above
+	  functionality; fixed an encoding issue in assertion error message.
+
+Mon Jun  6 17:29:35 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Speed up testing for small
+	  encodings by preselecting codepoints.
+
+Mon Jun  6 17:10:50 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_free): Use EVP_CIPHER_CTX_free()
+	  to free EVP_CIPHER_CTX allocated by EVP_CIPHER_CTX_new().
+	  [ruby-core:75225] [Feature #12324]
+
+Mon Jun  6 13:37:08 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c Change rb_str_casemap to use encoding primitive
+	  case_map instead of directly calling onigenc_unicode_case_map.
+
+Mon Jun  6 13:16:46 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_mapping.rb:
+	  Remove :lithuanian guard for Unicode case mapping.
+
+Mon Jun  6 10:39:56 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/openssl_missing.h: Include ruby/config.h. r55285 added
+	  some inline functions but VC does not recognize 'inline' keyword.
+
+Mon Jun  6 09:25:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (thread_start_func_2): report raised exception if
+	  report_on_exception flag is set.  [Feature #6647]
+
+Mon Jun  6 01:36:24 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Check existence of SSL_is_server(). This
+	  function was introduced in OpenSSL 1.0.2.
+	  [ruby-core:75225] [Feature #12324]
+
+	* ext/openssl/openssl_missing.h: Implement SSL_is_server() if missing.
+
+	* ext/openssl/ossl_ssl.c (ssl_info_cb): Use SSL_is_server() to see if
+	  the SSL is server. The state machine in OpenSSL was rewritten and
+	  SSL_get_state() no longer returns SSL_ST_ACCEPT.
+
+	  (ossl_ssl_cipher_to_ary, ossl_sslctx_session_get_cb): Add some
+	  `const`s to suppress warning.
+
+Mon Jun  6 01:18:10 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_asn1.c (decode_bool): Do the same thing as
+	  d2i_ASN1_BOOLEAN() does by ourselves. This function is removed in
+	  OpenSSL 1.1.0.
+	  [ruby-core:75225] [Feature #12324]
+
+Mon Jun  6 00:34:16 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Check existence of accessor functions that
+	  don't exist in OpenSSL 0.9.8. OpenSSL 1.1.0 made most of its
+	  structures opaque and requires use of these accessor functions.
+	  [ruby-core:75225] [Feature #12324]
+
+	* ext/openssl/openssl_missing.[ch]: Implement them if missing.
+
+	* ext/openssl/ossl*.c: Use these accessor functions.
+
+	* test/openssl/test_hmac.rb: Add missing test for HMAC#reset.
+
+Mon Jun  6 00:00:13 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/openssl_missing.[ch]: Implement EVP_PKEY_get0_*() and
+	  {RSA,DSA,EC_KEY,DH}_get0_*() functions.
+	  OpenSSL 1.1.0 makes EVP_PKEY/RSA/DSA/DH opaque. We used to provide
+	  setter methods for each parameter of each PKey type, for example
+	  PKey::RSA#e=, but this is no longer possible because the new API
+	  RSA_set0_key() requires the 'n' at the same time. This commit adds
+	  deprecation warning to them and adds PKey::*#set_* methods as direct
+	  wrapper for those new APIs. For example, 'rsa.e = 3' now needs to be
+	  rewritten as 'rsa.set_key(rsa.n, 3, rsa.d)'.
+	  [ruby-core:75225] [Feature #12324]
+
+	* ext/openssl/ossl_pkey*.[ch]: Use the new accessor functions. Implement
+	  RSA#set_{key,factors,crt_params}, DSA#set_{key,pqg}, DH#set_{key,pqg}.
+	  Emit a warning with rb_warning() when old setter methods are used.
+
+	* test/drb/ut_array_drbssl.rb, test/drb/ut_drb_drbssl.rb,
+	  test/rubygems/test_gem_remote_fetcher.rb: Don't set a priv_key for DH
+	  object that are used in tmp_dh_callback. Generating a new key pair
+	  every time should be fine - actually the private exponent is ignored
+	  in OpenSSL >= 1.0.2f/1.0.1r even if we explicitly set.
+	  https://www.openssl.org/news/secadv/20160128.txt
+
+Sun Jun  5 22:06:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* configure.in: Fix the timing to detect the appropriate C++ compiler
+	  in OS X.
+
+Sun Jun  5 21:42:24 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Check absence of CRYPTO_lock() to see if the
+	  OpenSSL has the new threading API. In OpenSSL <= 1.0.2, an application
+	  had to set locking callbacks to use OpenSSL in a multi-threaded
+	  environment. OpenSSL 1.1.0 now finds pthreads or Windows threads so we
+	  don't need to do something special.
+	  [ruby-core:75225] [Feature #12324]
+
+	  Also check existence of *_up_ref(). Some structures in OpenSSL have
+	  a reference counter. We used to increment it with CRYPTO_add() which
+	  is a part of the old API.
+
+	* ext/openssl/openssl_missing.h: Implement *_up_ref() if missing.
+
+	* ext/openssl/ossl.c: Don't set locking callbacks if unneeded.
+
+	* ext/openssl/ossl_pkey.c, ext/openssl/ossl_ssl.c,
+	  ext/openssl/ossl_x509cert.c, ext/openssl/ossl_x509crl.c,
+	  ext/openssl/ossl_x509store.c: Use *_up_ref() instead of CRYPTO_add().
+
+Sun Jun  5 21:38:13 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Check if RAND_pseudo_bytes() is usable. It is
+	  marked as deprecated in OpenSSL 1.1.0.
+	  [ruby-core:75225] [Feature #12324]
+
+	* ext/openssl/ossl_rand.c: Disable Random.pseudo_bytes if
+	  RAND_pseudo_bytes() is unavailable.
+
+	* test/openssl/test_random.rb: Don't test Random.pseudo_bytes if not
+	  defined.
+
+Sun Jun  5 19:06:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* NEWS: Add news about Unicode-wide case mapping for
+	  String/Symbol#upcase/downcase/swapcase/capitalize(!).
+
+Sun Jun  5 15:24:33 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb:
+	  Remove :lithuanian guard for Unicode case mapping.
+
+Sun Jun  5 14:46:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Remove :lithuanian guard for Unicode case mapping.
+
+Sat Jun  4 10:54:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/crypt.h (struct crypt_data): remove unnecessary member
+	  "initialized".
+
+	* missing/crypt.c (des_setkey_r): nothing to be initialized in
+	  crypt_data.
+
+	* configure.in (struct crypt_data): check for "initialized" in
+	  struct crypt_data, which may be only in glibc, and isn't on AIX
+	  at least.
+
+Sat Jun  4 10:38:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c: unify MAX_PATH, _MAX_PATH, and MAXPATHLEN to
+	  PATH_MAX, except for MAX_PATH in get_special_folder for an API
+	  limit.
+
+Fri Jun  3 21:27:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): rb_str_conv_enc() never set encoding
+	  of the source string, but returns the string itself if the
+	  conversion failed.  then the instance variable does not need to
+	  be set again.
+
+Fri Jun  3 18:04:37 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (process_options): need to duplicate path before passing it to
+	  rb_str_conv_enc() because the function might call rb_enc_associate()
+	  internally.  this fixes test failures on Windows introduced at r55260.
+
+Fri Jun  3 17:44:25 2016  Reiner Herrmann  <reiner@reiner-h.de>
+
+	* lib/mkmf.rb (create_makefile): sort lists of source and object
+	  files in generated Makefile, unless given by extconf.rb.
+	  [Fix GH-1367]
+
+Thu Jun  2 21:18:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/win32.c (get_special_folder): use SHGetPathFromIDListEx if
+	  available instead of old SHGetPathFromIDListW, to check the
+	  buffer size.
+
+Thu Jun  2 17:05:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (ruby_init_loadpath_safe): remove MAXPATHLEN restriction
+	  for Windows 10.
+
+Thu Jun  2 16:51:35 2016  Koichi ITO  <koic.ito@gmail.com>
+
+	* misc/ruby-mode.el (ruby-here-doc-beg-re),
+	  (ruby-here-doc-beg-match, ruby-parse-partial): Support for
+	  `squiggly heredoc' syntax in ruby-mode.  [Fix GH-1372]
+
+Thu Jun  2 10:24:48 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Raise ArgumentError when invalid string is detected in
+	  case mapping methods.
+
+	* enc/unicode.c: Check for invalid string and signal with negative
+	  length value.
+
+	* test/ruby/enc/test_case_mapping.rb: Add tests for above.
+
+	* test/ruby/test_m17n_comb.rb: Add a message to clarify test failure.
+
+Wed Jun  1 21:41:05 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Check existence of ASN1_TIME_adj(). The old
+	  ASN1_TIME_set() is not Year 2038 ready on sizeof(time_t) == 4
+	  environment. This function was added in OpenSSL 1.0.0.
+	  [ruby-core:45552] [Bug #6571]
+
+	* ext/openssl/ossl_asn1.c (ossl_time_split): Added. Split the argument
+	  (Time) into the number of days elapsed since the epoch and the
+	  remainder seconds to conform to ASN1_TIME_adj().
+	  (obj_to_asn1utime, obj_to_asn1gtime): Use ossl_time_split() and
+	  ASN1_*TIME_adj().
+
+	* ext/openssl/ossl_asn1.h: Add the function prototype for
+	  ossl_time_split().
+
+	* ext/openssl/ossl_x509.[ch]: Add ossl_x509_time_adjust(). Similarly to
+	  obj_to_asn1*time(), use X509_time_adj_ex() instead of X509_time_adj().
+
+	* ext/openssl/ossl_x509cert.c, ext/openssl/ossl_x509crl.c,
+	  ext/openssl/ossl_x509revoked.c: Use ossl_x509_time_adjust().
+
+Wed Jun  1 15:58:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: revert r55237.  replace crypt, not crypt_r, and
+	  check if crypt is broken more.
+
+	* missing/crypt.c: move crypt_r.c
+
+	* string.c (rb_str_crypt): use crypt_r if provided by the system.
+
+Wed Jun  1 14:07:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/crypt_r.c (a64toi): initialize statically and fix out of
+	  bounds access when salt is not 7bit clean.
+
+Wed Jun  1 11:34:59 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (MISSING): fixed build error introduced at r55237.
+
+Wed Jun  1 09:48:06 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_crypt): use reentrant crypt_r.
+
+Wed Jun  1 09:37:26 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* missing/crypt.c (des_setkey): void function never returns any value.
+
+Wed Jun  1 09:16:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* crypt.h: separate header file from missing/crypt.c.
+
+	* missing/crypt.c (crypt_r, setkey_r, encrypt_r): add reentrant
+	  versions.
+
+Wed Jun  1 02:25:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* missing/crypt.c: fix size macros to use configured values
+	  for platforms long is larger than 32bit.
+	  [ruby-core:75792] [Bug #12446]
+
+Tue May 31 17:28:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser::Completion.candidate): get rid of
+	  nil as key names.  [ruby-core:75773] [Bug #12438]
+
+	* lib/optparse.rb (OptionParser#make_switch): char class option
+	  cannot be NoArgument, default to RequiredArgument.
+
+Tue May 31 00:30:11 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (host_str, port_str): Use StringValueCStr
+	  instead of (Safe)StringValue, to detect NUL byte in the string.
+
+Mon May 30 22:02:01 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_asn1.c (time_to_time_t): Use NUM2TIMET() instead of
+	  NUM2LONG(). time_t may be larger than long.
+	  [ruby-core:45552] [Bug #6571]
+
+Mon May 30 21:15:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Document current behavior for other case mapping methods
+	  on String. [ci skip]
+
+Mon May 30 20:00:25 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Document current situation for String#downcase. [ci skip]
+
+Mon May 30 18:29:28 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_s_alloc): Enable the automatic
+	  curve selection for ECDH by calling SSL_CTX_set_ecdh_auto(). With
+	  this a TLS server automatically selects a curve which both the client
+	  and the server support to use in ECDH. This changes the default
+	  behavior but users can still disable ECDH by excluding 'ECDH' cipher
+	  suites from the cipher list (with SSLContext#ciphers=). This commit
+	  also deprecate #tmp_ecdh_callback=. It was added in Ruby 2.3.0. It
+	  wraps SSL_CTX_set_tmp_ecdh_callback() which will be removed in OpenSSL
+	  1.1.0. Its callback receives two values 'is_export' and 'keylength'
+	  but both are completely useless for determining a curve to use in
+	  ECDH. The automatic curve selection was introduced to replace this.
+
+	  (ossl_sslctx_setup): Deprecate SSLContext#tmp_ecdh_callback=. Emit a
+	  warning if this is in use.
+
+	  (ossl_sslctx_set_ecdh_curves): Add SSLContext#ecdh_curves=. Wrap
+	  SSL_CTX_set1_curves_list(). If it is not available, this falls back
+	  to SSL_CTX_set_tmp_ecdh().
+
+	  (Init_ossl_ssl): Define SSLContext#ecdh_curves=.
+
+	* ext/openssl/extconf.rb: Check the existence of EC_curve_nist2nid(),
+	  SSL_CTX_set1_curves_list(), SSL_CTX_set_ecdh_auto() and
+	  SSL_CTX_set_tmp_ecdh_callback().
+
+	* ext/openssl/openssl_missing.[ch]: Implement EC_curve_nist2nid() if
+	  missing.
+
+	* test/openssl/test_pair.rb (test_ecdh_callback): Use
+	  EnvUtil.suppress_warning to suppress deprecated warning.
+
+	  (test_ecdh_curves): Test that SSLContext#ecdh_curves= works.
+
+	* test/openssl/utils.rb (start_server): Use SSLContext#ecdh_curves=.
+
+Mon May 30 16:28:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/raddrinfo.c (host_str, port_str): use RSTRING_LEN
+	  instead of strlen, since RSTRING_PTR StringValueCStr may not be
+	  NUL-terminated when SHARABLE_MIDDLE_SUBSTRING=1.  reported by
+	  @tmtms, http://twitter.com/tmtms/status/736910516229005312
+
+Mon May 30 16:20:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_fill_term): return new pointer reallocated by
+	  filling terminator.
+
+Mon May 30 14:54:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (enc_subseq): share the return value and
+	  the buffer as possible.
+
+Mon May 30 14:50:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (str_substr, rb_str_aref): refactor not to create
+	  unnecessary empty string.
+
+	* string.c (str_byte_substr, str_byte_aref): ditto.
+
+Mon May 30 00:09:37 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/-test-/auto_ext.rb: fixed a heedless bug introduced at r55198.
+	  this change will make RubyCI green.
+
+Sun May 29 22:58:19 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regexec.c (ONIGENC_IS_MBC_ASCII_WORD): redefine optimized one.
+	  WORD of Ruby's ascii compatible encoding is always [a-zA-Z0-9_].
+
+Sun May 29 22:44:19 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regexec.c (match_at): make compilers optimize harder.
+
+Sun May 29 12:08:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/-test-/auto_ext.rb (auto_ext): utility method to create
+	  extension libraries for tests.
+
+Sat May 28 20:40:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* misc/ruby-additional.el (ruby-insert-heredoc-code-block): insert
+	  here document code block for assert_separately mainly.
+
+Sat May 28 20:34:19 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/test_unicode_normalize.rb: Add test to check for availability of
+	  Unicode data file; refactoring; fix an error with tests for destructive
+	  method (unicode_normalize!).
+
+Sat May 28 19:08:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Add error messages to tests
+	  for data file availability; refactoring.
+
+Sat May 28 14:00:10 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/lib/openssl/ssl.rb (SSLSocket): Move the implementation of
+	  SSLSocket#initialize to C. Initialize the SSL (OpenSSL object) in it.
+	  Currently this is delayed until ossl_ssl_setup(), which is called from
+	  SSLSocket#accept or #connect. Say we call SSLSocket#hostname= with an
+	  illegal value. We expect an exception to be raised in #hostname= but
+	  actually we get it in the later SSLSocket#connect. Because the SSL is
+	  not ready at #hostname=, the actual call of SSL_set_tlsext_host_name()
+	  is also delayed.
+	  This also fixes: [ruby-dev:49376] [Bug #11724]
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_initialize): Added. Almost the same
+	  as the Ruby version but this instantiate the SSL object at the same
+	  time.
+
+	  (ossl_ssl_setup): Adjust to the changes. Just set the underlying IO to
+	  the SSL.
+
+	  (ssl_started): Added. Make use of SSL_get_fd(). This returns -1 if not
+	  yet set by SSL_set_fd().
+
+	  (ossl_ssl_data_get_struct): Removed. Now GetSSL() checks that the SSL
+	  exists.
+
+	  (ossl_ssl_set_session): Don't call ossl_ssl_setup() here as now the
+	  SSL is already instantiated in #initialize.
+
+	  (ossl_ssl_shutdown, ossl_start_ssl, ossl_ssl_read_internal,
+	   ossl_ssl_write_internal, ossl_ssl_stop, ossl_ssl_get_cert,
+	   ossl_ssl_get_peer_cert, ossl_ssl_get_peer_cert_chain,
+	   ossl_ssl_get_version, ossl_ssl_get_cipher, ossl_ssl_get_state,
+	   ossl_ssl_pending, ossl_ssl_session_reused,
+	   ossl_ssl_get_verify_result, ossl_ssl_get_client_ca_list,
+	   ossl_ssl_npn_protocol, ossl_ssl_alpn_protocol, ossl_ssl_tmp_key): Use
+	  GetSSL() instead of ossl_ssl_data_get_struct(). Use ssl_started().
+
+	  (Init_ossl_ssl): Add method declarations of SSLSocket#{initialize,
+	  hostname=}.
+
+	* ext/openssl/ossl_ssl.h (GetSSL): Check that the SSL is not NULL. It
+	  should not be NULL because we now set it in #initialize.
+
+	* ext/openssl/ossl_ssl_session.c (ossl_ssl_session_initialize): No need
+	  to check if the SSL is NULL.
+
+Sat May 28 10:47:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: Update latest releases, power_assert-0.3.0,
+	  test-unit 3.1.9, minitest 5.9.0, did_you_mean 1.0.1
+
+Sat May 28 10:45:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* addr2line.c: drop support for ATARI ST platform. It was
+	  discontinued more than two decades ago. [fix GH-1350] Patch by
+	  @cremno
+	* include/ruby/ruby.h: ditto.
+	* io.c: ditto.
+
+Sat May 28 10:39:47 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Converted exception for
+	  unavailable Unicode data files to failed assertion.
+
+Sat May 28 10:26:18 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/cgi/util.rb: added missing quote.
+	  [fix GH-1363][ci skip] Patch by @dwaller
+
+Fri May 27 17:38:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_local_constants_i): exclude hidden constants.
+	  [ruby-core:75575] [Bug #12389]
+
+Fri May 27 17:09:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (str_transcode0): scrub in the given encoding when
+	  the source encoding is given, not in the encoding of the
+	  receiver.  [ruby-core:75732] [Bug #12431]
+
+Fri May 27 15:07:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args): remove nul padding which
+	  caused syntax error if fmt is not a string literal.
+
+	* include/ruby/ruby.h (rb_scan_args_verify): suppress array-bounds
+	  warnings by old clang.
+
+	* include/ruby/ruby.h (rb_scan_args0): make extractor macros
+	  inline functions, which do not validate the format and are
+	  unnecessary to be expanded.
+
+Fri May 27 01:00:36 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* symbol.c (is_identchar): use ISDIGIT instead of rb_enc_isalnum.
+	  Though rb_enc_isalnum is encoding aware function, its argument here
+	  is *m, which is a single byte. Therefore ISDIGIT is faster.
+
+	* symbol.c (is_special_global_name): ditto.
+
+	* symbol.c (rb_enc_symname_type): ditto.
+
+Fri May 27 00:39:40 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args): add nul padding here to
+	  apply to all references.
+
+	* include/ruby/ruby.h (rb_scan_args_verify): move length mismatch
+	  check outside conditional operators.
+
+Thu May 26 14:21:10 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkey_dh.c (ossl_dh_compute_key): Check that the DH
+	  has 'p' (the prime) before calling DH_size(). We can create a DH with
+	  no parameter but DH_size() does not check and dereferences NULL.
+	  [ruby-core:75720] [Bug #12428]
+
+	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_sign): Ditto. DSA_size() does
+	  not check dsa->q.
+
+	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_public_encrypt,
+	  ossl_rsa_public_decrypt, ossl_rsa_private_encrypt,
+	  ossl_rsa_private_decrypt): Ditto. RSA_size() does not check rsa->n.
+
+Thu May 26 14:13:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args_count): verify length with
+	  counting variables together.
+
+Thu May 26 09:45:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_comprehensive.rb: Add set of comprehensive
+	  (across most Unicode characters; later across most character encodings)
+	  tests for case mapping.
+
+Thu May 26 05:00:13 2016  Benoit Daloze  <eregontp@gmail.com>
+
+	* class.c (rb_define_class): Fix documentation.
+
+Wed May 25 20:50:12 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (unescape_nonascii): scan hex up to only 3 characters.
+	  [Bug #12420] [Bug #12423]
+
+Wed May 25 19:07:19 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Handle DOTLESS_i by hand because it isn't involved in folding.
+
+Wed May 25 18:30:53 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (fetch_token_in_cc): raise error if given octal escaped
+	  character is too big. [Bug #12420] [Bug #12423]
+
+Wed May 25 17:45:15 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl, test/openssl: Drop OpenSSL < 0.9.8 support.
+
+Wed May 25 17:43:30 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/openssl_missing.h, ext/openssl/ossl.h: Remove
+	  unnecessary 'extern "C"' blocks. We don't use C++ and these headers
+	  are local to ext/openssl, so there is no need to enclose with it.
+
+Wed May 25 17:42:58 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: Remove check of OPENSSL_FIPS macro. This is
+	  unneeded because we can check the macro directly in source code,
+	  just as we already do for OPENSSL_NO_* macros.
+
+	* ext/openssl/ossl.c: Replace occurrences of HAVE_OPENSSL_FIPS with
+	  OPENSSL_FIPS.
+
+Wed May 25 17:13:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_scan_args): merge code for n_trail.
+
+Wed May 25 17:11:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args_validate): move failed
+	  condition to the terminal.  [ruby-core:75714] [Bug #12426]
+
+Wed May 25 13:13:37 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c: remove condition for debug output because prelude
+	  doesn't use regexp now.
+
+Wed May 25 13:10:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regcomp.c (compile_length_tree): return error code immediately
+	  if compile_length_tree raised error [Bug #12418]
+
+Wed May 25 08:01:39 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Fix flag error for switch from titlecase to lowercase.
+
+	* test/ruby/enc/test_case_mapping.rb: Tests for above error.
+
+Wed May 25 01:13:55 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkey_ec.c (ec_key_new_from_group): Create a new
+	  EC_KEY on given EC group. Extracted from ossl_ec_key_initialize().
+	  (ossl_ec_key_s_generate): Added. Create a new EC instance and
+	  generate a random private and public key.
+	  (ossl_ec_key_initialize): Use ec_key_new_from_group().
+	  (Init_ossl_ec): Define the new method EC.generate. This change is
+	  for consistency with other PKey types.  [ruby-core:45541] [Bug #6567]
+
+	* test/openssl/test_pkey_ec.rb: Test that EC.generate works.
+
+Wed May 25 00:37:16 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_generate_key): Fix up RDoc.
+	  (Init_ossl_ec): Rename EC#generate_key to EC#generate_key!. Make the
+	  old name an alias of #generate_key!. This change is for consistency
+	  with other PKey types.  [ruby-core:45541] [Bug #6567]
+
+	* test/openssl/test_pkey_ec.rb: Use EC#generate_key! instead of
+	  EC#generate_key.
+
+Wed May 25 00:23:05 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args_set): check the arity after
+	  adjusting argc for an option hash, for optimization in simpler
+	  cases.
+
+Wed May 25 00:21:52 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (XCFLAGS): merge flags only for ruby itself from
+	  ruby_cflags.
+
+Tue May 24 22:04:15 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_cipher.c (ossl_cipher_set_key, ossl_cipher_set_iv):
+	  Reject too long values as well as too short ones. Currently they
+	  just truncate the input but this would hide bugs and lead to
+	  unexpected encryption/decryption results.
+
+	* test/openssl/test_cipher.rb: Test that Cipher#key= and #iv= reject
+	  Strings with invalid length.
+
+Tue May 24 21:32:21 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_x509ext.c (ossl_x509ext_set_value): Use
+	  ASN1_OCTET_STRING_set() instead of M_ASN1_OCTET_STRING_set(). Macros
+	  prefixed by "M_" are discouraged to be used from outside OpenSSL
+	  library[1].
+	  (ossl_x509ext_get_value): Likewise, use ASN1_STRING_print() instead
+	  of M_ASN1_OCTET_STRING_print().
+	  [1] https://git.openssl.org/gitweb/?p=openssl.git;a=blob;f=CHANGES;h=bf61913d7b01212b4d8b2f3c13d71d645914f67c;hb=b6079a7835f61daa9fb2cbf9addfa86049523933#l878
+
+	* ext/openssl/ossl.h: Include openssl/asn1.h instead of
+	  openssl/asn1_mac.h. It just includes openssl/asn1.h and defines some
+	  additional "M_" macros.
+
+Tue May 24 18:52:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args_verify): verify the format to
+	  scan if no invalid chars and variable argument length matching,
+	  at the compile time if possible.
+
+Tue May 24 17:18:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/defines.h (ERRORFUNC, WARNINGFUNC): add fallback
+	  definitions.
+
+Tue May 24 16:37:43 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (ruby_cflags): separate from optflags [Bug #12409]
+	  -fexcess-precision=standard and -fp-model precise are set to this now.
+
+	* configure.in (cflags): use ruby_cflags.
+
+Tue May 24 16:20:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ERRORFUNC, WARNINGFUNC): __error__ and __warning__
+	  attributes take a parenthesized string literal.
+
+Tue May 24 12:35:56 2016  URABE Shyouhei  <shyouhei@ruby-lang.org>
+
+	* common.mk: sort lines, and add missing dependencies suggested
+	  by tool/update-deps
+
+Mon May 23 21:33:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/missing.h (isfinite): move from numeric.c.
+
+Mon May 23 21:09:06 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl.c (Init_openssl): Avoid reference to unset global
+	  variable. ossl_raise() may be called before dOSSL is set. Since
+	  global variables default to 0 and the default value of dOSSL set in
+	  Init_openssl() is also Qfalse, there is no real issue but confusing.
+	  Patch by Bertram Scharpf <software@bertram-scharpf.de>
+	  [ruby-core:58264] [Bug #9101]
+
+Mon May 23 20:32:16 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_asn1.c, ext/openssl/ossl_bn.c,
+	  ext/openssl/ossl_cipher.c, ext/openssl/ossl_digest.c
+	  ext/openssl/ossl_engine.c, ext/openssl/ossl_ns_spki.c
+	  ext/openssl/ossl_pkcs12.c, ext/openssl/ossl_pkcs7.c
+	  ext/openssl/ossl_pkey.c, ext/openssl/ossl_pkey_ec.c
+	  ext/openssl/ossl_rand.c, ext/openssl/ossl_ssl.c
+	  ext/openssl/ossl_x509attr.c, ext/openssl/ossl_x509cert.c
+	  ext/openssl/ossl_x509ext.c, ext/openssl/ossl_x509store.c: Use
+	  StringValueCStr() where NUL-terminated string is expected.
+
+Mon May 23 20:20:12 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (isfinite): get rid of a warning on
+	  cygwin. [Bug #12417][ruby-core:75691]
+
+Mon May 23 19:41:27 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_rand.c (ossl_rand_egd, ossl_rand_egd_bytes):
+	  RAND_egd{_bytes,}() return -1 on failure, not 0.
+	  Patch by cremno phobia <cremno@mail.ru>
+	  [ruby-core:63795] [Bug #10053]
+	  (ossl_pseudo_bytes): Similar, RAND_pseudo_bytes() may return 0 or
+	  -1 on failure.
+
+Mon May 23 15:52:07 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (isfinite): isfinite does not always
+	  exist.  fixed build error on Windows introduced at r55123.
+
+Mon May 23 13:19:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args0): make compile error if the
+	  format is wrong or does not match with the variable argument
+	  length if possible.
+
+Mon May 23 12:47:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args0): raise fatal error if
+	  variable argument length does not match, it is a bug in the code
+	  which uses rb_scan_args, not a runtime error.
+
+Mon May 23 12:30:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/bigdecimal/bigdecimal.c (GetVpValueWithPrec): consider
+	  non-finite float values not to raise FloatDomainError.
+	  [ruby-core:75682] [Bug #12414]
+
+Mon May 23 12:21:18 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_fill): suppress warnings: 'item' may be used
+	  uninitialized in this function
+
+Mon May 23 07:41:49 2016  Eric Wong  <e@80x24.org>
+
+	* dir.c (dir_close): update RDoc for 2.3 #close change
+	  [ruby-core:75679] [Bug #12413]
+
+Sun May 22 20:01:21 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/timeridconv.rb: use finalizer trick instead of thread.
+
+	* test/drb/ut_timerholder.rb: ditto.
+
+Sun May 22 17:25:18 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_options.rb: adjust test class name
+	  to match file name
+
+Sun May 22 17:24:07 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* rename test/ruby/enc/test_casing_options.rb to test_case_options.rb
+	  for consistency
+
+Sun May 22 17:06:55 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* transcode.c (enc_arg, str_transcode_enc_args, econv_args):
+	  remove volatile, and add GC guards in callers.
+	  [ruby-core:75664] [Bug #12411]
+
+Sun May 22 16:27:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/-test-/integer/core_ext.c: move testutil/integer.c.
+
+	* test/lib/-test-/integer.rb: extract implementation details from
+	  test/unit/assertions.rb.  [Bug #12408]
+
+Sun May 22 14:57:43 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* include/ruby/oniguruma.h: Extend OnigEncodingTypeDefine to define a
+	  new encoding primitive 'case_map' for case mapping
+
+	* enc/utf-8.c, utf_16be/le.c, utf_32be/le.c:
+	  add onigenc_unicode_case_map as case_map primitive
+
+	* enc/ascii.c, big5.c, cp949.c, emacs_mule.c, euc_jp/kr/tw.c, gb18030.c,
+	  gbk.c, iso_8859_1/2/3/4/5/6/7/8/9/10/11/13/14/15/16.c, koi8_r/u.c,
+	  shift_jis.c, us_ascii.c, windows_1250/1251/1252.c:
+	  add onigenc_not_support_case_map as case_map primitive
+
+Sun May 22 14:45:45 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* regenc.h/c: Define new function onigenc_not_support_case_map
+
+Sun May 22 12:14:06 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args): use original rb_scan_args
+	  when fmt is dynamic.
+
+Sun May 22 11:41:12 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* class.c (rb_scan_args): moved to bottom of the file to make the
+	  effect of `#undef rb_scan_args` the minimum.
+
+	* include/ruby/ruby.h (rb_scan_args): overwrite only if GCC and
+	  optimized. Visual C++ 14 or later can compile it but make it
+	  conservative.
+
+Sat May 21 22:45:50 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args): don't use ALWAYS_INLINE with
+	  `inline`.  if gcc needs this duplication, do in ALWAYS_INLINE macro.
+
+Sat May 21 21:11:56 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_scan_args): use __VA_ARGS__ instead of
+	  va_arg to allow compilers optimize more aggressive.
+	  https://gustedt.wordpress.com/2011/07/10/avoid-writing-va_arg-functions/
+	  rb_scan_args is now expected to be statically resolved.
+
+Sun May 22 02:41:52 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/zlib/zlib.c: remove hacky macro introduced at r30437.
+
+	* ext/zlib/zlib.c (gzfile_make_header): cast as long (instead of int).
+
+	* ext/zlib/zlib.c (gzfile_make_footer): ditto.
+
+Sat May 21 21:07:18 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (ALWAYS_INLINE): force compilers the function inlined.
+
+Sat May 21 16:16:03 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_stop): Don't free the SSL struct
+	  here. Since some methods such as SSLSocket#connect releases GVL,
+	  there is a chance of use after free if we free the SSL from another
+	  thread. SSLSocket#stop was documented as "prepares it for another
+	  connection" so this is a slightly incompatible change. However when
+	  this sentence was added (r30090, Add toplevel documentation for
+	  OpenSSL, 2010-12-06), it didn't actually. The current behavior is
+	  from r40304 (Correct shutdown behavior w.r.t GC., 2013-04-15).
+	  [ruby-core:74978] [Bug #12292]
+
+	* ext/openssl/lib/openssl/ssl.rb (sysclose): Update doc.
+
+	* test/openssl/test_ssl.rb: Test this.
+
+Sat May 21 14:41:14 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl.c: [DOC] Fix SSL client example. The variable name
+	  was wrong. Patch by Andreas Tiefenthaler <at@an-ti.eu> (@pxlpnk).
+	  [GH ruby/openssl#32]
+
+Sat May 21 14:25:38 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkey_ec.c: rename PKey::EC#private_key? and
+	  #public_key? to #private? and #public? for consistency with other
+	  PKey types. Old names remain as alias.  [ruby-core:45541] [Bug #6567]
+
+	* test/openssl/test_pkey_ec.rb (test_check_key): check private? and
+	  public? works correctly.
+
+Sat May 21 12:40:36 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/lib/openssl/buffering.rb (read_nonblock, readpartial):
+	  Remove impossible EOFError raise. Patch by Zach Anker
+	  <zanker@squareup.com>.  [GH ruby/openssl#23]
+
+Sat May 21 11:18:42 2016  Evgeni Golov  <evgeni@golov.de>
+
+	* lib/ipaddr.rb: [DOC] fix documentation of IN6MASK to mention
+	  IPv6.  [Fix GH-1349]
+
+Sat May 21 11:12:53 2016  Dan Martinez  <dfm@razorwind.org>
+
+	* io.c (Init_IO): [DOC] define dummy ARGF instead of ARGF.class to
+	  re-enable the generation of ARGF documentation.  [Fix GH-1358]
+
+Sat May 21 11:07:29 2016  0x01f7  <souk.0x01f7@gmail.com>
+
+	* doc/syntax/methods.rdoc (Method Names): add proper closing tag.
+	  [Fix GH-1356]
+
+Sat May 21 09:26:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SETJMP): fix missing macro definition
+	  for the configured result.  fix up r55021.
+
+Sat May 21 00:36:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_execarg_commandline): build command line string
+	  from argument vector in rb_execarg.
+	  [ruby-core:75611] [Bug #12398]
+
+Fri May 20 23:25:42 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl.c (ossl_pem_passwd_value): Added. Convert the
+	  argument to String with StringValue() and validate the length is in
+	  4..PEM_BUFSIZE. PEM_BUFSIZE is a macro defined in OpenSSL headers.
+	  (ossl_pem_passwd_cb): When reading/writing encrypted PEM format, we
+	  used to pass the password to PEM_def_callback() directly but it was
+	  problematic. It is not NUL character safe. And surprisingly, it
+	  silently truncates the password to 1024 bytes.  [GH ruby/openssl#51]
+
+	* ext/openssl/ossl.h: Add function prototype declaration of newly
+	  added ossl_pem_passwd_value().
+
+	* ext/openssl/ossl_pkey.c (ossl_pkey_new_from_data): Use
+	  ossl_pem_passwd_value() to validate the password String.
+
+	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_initialize, ossl_dsa_export):
+	  ditto.
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_initialize,
+	  ossl_ec_key_to_string): ditto.
+
+	* ext/openssl/ossl_pkey_rsa.c (ossl_rsa_initialize, ossl_rsa_export):
+	  ditto.
+
+	* test/openssl/test_pkey_{dsa,ec,rsa}.rb: test this.
+
+Fri May 20 23:45:53 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* id_table.c (list_id_table_init): When unaligned word access is
+	  prohibited and sizeof(VALUE) is 8 (64-bit machines),
+	  capa should always be even number for 8-byte word alignment
+	  of the values of a table. This code assumes that sizeof(ID) is 4,
+	  sizeof(VALUE) is 8, and xmalloc() returns 8-byte aligned memory.
+	  This fixes bus error on 64-bit SPARC Solaris 10.
+	  [Bug #12406][ruby-dev:49631]
+
+Fri May 20 22:30:09 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* symbol.h (rb_id2sym): Use HAVE_BUILTIN___BUILTIN_CONSTANT_P
+
+Fri May 20 22:19:00 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* ext/date/date_core.c: [DOC] fix typos.
+	  [fix GH-1360] patched by @soundasleep
+
+Fri May 20 21:26:58 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* include/ruby/defines.h (RB_GNUC_EXTENSION, RB_GNUC_EXTENSION_BLOCK):
+	  macros for skipping __extension__ on non-GCC compilers.
+	* eval_error.c (warn_print): use RB_GNUC_EXTENSION_BLOCK instead of
+	  __extension__ because __extension__ is a GNU extension.
+	  Fix compile error on Solaris 10 with Oracle Solaris Studio 12.x.
+	  [Bug #12397] [ruby-dev:49629].
+	* internal.h (rb_fstring_cstr, rb_fstring_enc_cstr): ditto
+	* include/ruby/encoding.h (rb_enc_str_new, rb_enc_str_new_cstr): ditto
+	* include/ruby/intern.h (rb_str_new, rb_str_new_cstr,
+	  rb_usascii_str_new, rb_utf8_str_new, rb_tainted_str_new_cstr,
+	  rb_usascii_str_new_cstr, rb_utf8_str_new_cstr,
+	  rb_external_str_new_cstr, rb_locale_str_new_cstr,
+	  rb_str_buf_new_cstr, rb_str_cat_cstr, rb_exc_new_cstr): ditto
+
+Fri May 20 21:17:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole.c (fole_missing): make substring or dup to
+	  share the content if possible.
+
+Fri May 20 19:48:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (NEW_PARTIAL_MEMO_FOR): shrink buffer array not to
+	  mark non-VALUE fields.  fix check_rvalue_consistency abort with
+	  RGENGC_CHECK_MODE=2.
+
+	* internal.h (NEW_CMP_OPT_MEMO): exclude struct cmp_opt_data from
+	  the valid array range.
+
+	* enum.c (slicewhen_i): exclude inverted too.
+
+Thu May 19 21:21:57 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* re.c (rb_reg_match_m_p): [DOC] fix return value in rdoc.
+
+	* test/ruby/test_regexp.rb (TestRegexp#test_match_p): add some
+	  tests from document.
+
+Thu May 19 13:22:44 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl.c (Init_openssl): register an ex_data index for
+	  X509_STORE and X509_STORE_CTX respectively. Since they don't share
+	  the ex_data index registry, we can't use the same index.
+	  (ossl_verify_cb): use the correct index.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_verify_callback): ditto.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509store_set_vfy_cb): ditto.
+	  (ossl_x509stctx_verify): ditto.
+
+	* ext/openssl/ossl.h (void ossl_clear_error): add extern declarations
+	  of ossl_store_{ctx_,}ex_verify_cb_idx.
+
+	* ext/openssl/openssl_missing.c: remove X509_STORE_set_ex_data and
+	  X509_STORE_get_ex_data.
+
+	* ext/openssl/openssl_missing.h: implement X509_STORE_get_ex_data,
+	  X509_STORE_set_ex_data and X509_STORE_get_ex_new_index as macros.
+
+Thu May 19 13:11:35 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_x509attr.c (ossl_x509attr_set_value): check that the
+	  argument is an OpenSSL::ASN1::Data before converting to ASN1_TYPE.
+	  This fixes SEGV on OpenSSL::X509::Attribute#value=(non-asn1-value).
+
+	* test/openssl/test_x509attr.rb: add tests for OpenSSL::X509::Attribute.
+
+Thu May 19 12:10:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_match_m_p): fix match against empty string.
+	  rb_str_offset returns the end when the position exceeds the
+	  length.  fix the range parameter of onig_search.
+	  [ruby-core:75604] [Bug #12394]
+
+Thu May 19 11:37:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (rb_reg_match_m_p): should return false if no match, as the
+	  document says.  [Feature #8110]
+
+Thu May 19 00:17:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (reg_names_iter): specify capacify
+
+Wed May 18 21:29:59 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* thread.c (recursive_list_access): a object id may be a Bignum.  so,
+	  the list must be a objhash, instead of a identhash.
+	  this fixes many test errors on mswin64 CI.
+
+Wed May 18 19:33:54 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_match_m_p): Introduce Regexp#match?, which returns
+	  bool and doesn't save backref.
+
+Wed May 18 16:52:03 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkcs12.c (ossl_pkcs12_initialize): pop errors
+	  leaked by PKCS12_parse(). This is a bug in OpenSSL, which exists
+	  in the versions before the version 1.0.0t, 1.0.1p, 1.0.2d.
+
+Wed May 18 16:04:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/downloader.rb (Downloader::RubyGems.download): verify gems
+	  only if RubyGems is 2.4 or later.  old RubyGems fails to verify
+	  almost all of bundled gems.
+
+Wed May 18 14:52:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_modify_expand): check integer overflow.
+	  [ruby-core:75592] [Bug #12390]
+
+Wed May 18 13:11:44 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (match_ary_subseq): get subseq of match array without creating
+	  temporary array.
+
+	* re.c (match_ary_aref): get element(s) of match array without creating
+	  temporary array.
+
+	* re.c (match_aref): Use match_ary_subseq with handling irregulars.
+
+	* re.c (match_values_at): Use match_ary_aref.
+
+Wed May 18 13:03:07 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_x509cert.c (ossl_x509_verify): X509_verify()
+	  family may put errors on 0 return (0 means verification failure).
+	  Clear OpenSSL error queue before return to Ruby. Since the queue is
+	  thread global, remaining errors in the queue can cause an unexpected
+	  error in the next OpenSSL operation.  [ruby-core:48284] [Bug #7215]
+
+	* ext/openssl/ossl_x509crl.c (ossl_x509crl_verify): ditto.
+
+	* ext/openssl/ossl_x509req.c (ossl_x509req_verify): ditto.
+
+	* ext/openssl/ossl_x509store.c (ossl_x509stctx_verify): ditto.
+
+	* ext/openssl/ossl_pkey_dh.c (dh_generate): clear the OpenSSL error
+	  queue before re-raising exception.
+
+	* ext/openssl/ossl_pkey_dsa.c (dsa_generate): ditto.
+
+	* ext/openssl/ossl_pkey_rsa.c (rsa_generate): ditto.
+
+	* ext/openssl/ossl_ssl.c (ossl_start_ssl): ditto.
+
+	* test/openssl: check that OpenSSL.errors is empty every time after
+	  running a test case.
+
+Wed May 18 12:07:42 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl.c (ossl_clear_error): Extracted from
+	  ossl_make_error(). This prints errors in the OpenSSL error queue if
+	  OpenSSL.debug is true, and clears the queue.
+	  (ossl_make_error): use ossl_clear_error().
+
+	* ext/openssl/ossl.h: add prototype declaration of ossl_make_error().
+	  (OSSL_BIO_reset) use ossl_clear_error() to clear the queue. Clearing
+	  silently makes debugging difficult.
+
+	* ext/openssl/ossl_engine.c (ossl_engine_s_by_id): ditto.
+
+	* ext/openssl/ossl_ns_spki.c (ossl_spki_initialize): ditto.
+
+	* ext/openssl/ossl_pkcs7.c (ossl_pkcs7_verify): ditto.
+
+	* ext/openssl/ossl_pkey_dsa.c (ossl_dsa_initialize): ditto.
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_key_initialize): ditto.
+	  (ossl_ec_group_initialize): ditto.
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_shutdown): ditto.
+
+Wed May 18 11:53:49 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_pkey_ec.c (ossl_ec_point_mul): Validate the
+	  arguments before passing to EC_POINT(s)_mul(). Add description of this
+	  method.  [ruby-core:65152] [Bug #10268]
+
+	* test/openssl/test_pkey_ec.rb (test_ec_point_mul): Test that
+	  OpenSSL::PKey::EC::Point#mul works.
+
+Wed May 18 11:19:59 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/ossl_bn.c (try_convert_to_bnptr): Extracted from
+	  GetBNPtr(). This doesn't raise exception but returns NULL on error.
+	  (GetBNPtr): Raise TypeError if conversion fails.
+	  (ossl_bn_eq): Implement BN#==.
+	  (ossl_bn_eql): #eql? should not raise TypeError even if the argument
+	  is not compatible with BN.
+	  (ossl_bn_hash): Implement BN#hash.
+
+	* ext/openssl/ossl_bn.c (Init_ossl_bn): Define #== and #hash.
+
+	* test/openssl/test_bn.rb: Test BN#eql?, #== and #hash
+
+Wed May 18 10:17:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_INTEGER_TYPE_P): new macro and
+	  underlying inline function to check if the object is an
+	  Integer (Fixnum or Bignum).
+
+Wed May 18 09:52:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (enum_sum, hash_sum, hash_sum_i, enum_sum_i, sum_iter):
+	  Optimize for hashes when each method isn't redefined.
+
+Wed May 18 09:14:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (enum_sum, int_range_sum): Extract int_range_sum from
+	  enum_sum.
+
+Wed May 18 03:16:06 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (match_values_at): fix regression at r55036.
+	  MatchData#values_at accepts Range.
+
+Wed May 18 02:02:58 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (match_aref): remove useless condition and call rb_fix2int.
+	  rb_reg_nth_match handles negative index.
+
+Wed May 18 01:57:43 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (match_values_at): MatchData#values_at supports named captures
+	  [Feature #9179]
+
+	* re.c (namev_to_backref_number): separated.
+
+Wed May 18 00:05:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (enum_sum): Optimize for a range from int to int.
+
+	* test/ruby/test_enum.rb (test_range_sum): Move from test_range.rb,
+	  and add assertions for some conditions.
+
+	* test/ruby/test_enum.rb (test_hash_sum): Move from test_hash.rb.
+
+	* test/ruby/test_hash.rb, test/ruby/test_range.rb: Remove test_sum.
+
+Tue May 17 23:08:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (enum_sum): [DOC] Write documentation.
+
+Tue May 17 22:53:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (enum_sum): Implement Enumerable#sum.
+
+	* test/ruby/test_enum.rb (test_sum): Test sum for Enumerable.
+
+	* test/ruby/test_hash.rb (test_sum): Test sum for Hash.
+
+	* test/ruby/test_range.rb (test_sum): Test sum for Range.
+
+Tue May 17 22:11:41 2016  Tanaka Akira  <akr@fsij.org>
+
+	* object.c, numeric.c, enum.c, ext/-test-/bignum/mul.c,
+	  lib/rexml/quickpath.rb, lib/rexml/text.rb, lib/rexml/xpath_parser.rb,
+	  lib/rubygems/specification.rb, lib/uri/generic.rb,
+	  bootstraptest/test_eval.rb, basictest/test.rb,
+	  test/-ext-/bignum/test_big2str.rb, test/-ext-/bignum/test_div.rb,
+	  test/-ext-/bignum/test_mul.rb, test/-ext-/bignum/test_str2big.rb,
+	  test/csv/test_data_converters.rb, test/date/test_date.rb,
+	  test/json/test_json_generate.rb, test/minitest/test_minitest_mock.rb,
+	  test/openssl/test_cipher.rb, test/rexml/test_jaxen.rb,
+	  test/ruby/test_array.rb, test/ruby/test_basicinstructions.rb,
+	  test/ruby/test_bignum.rb, test/ruby/test_case.rb,
+	  test/ruby/test_class.rb, test/ruby/test_complex.rb,
+	  test/ruby/test_enum.rb, test/ruby/test_eval.rb,
+	  test/ruby/test_iseq.rb, test/ruby/test_literal.rb,
+	  test/ruby/test_math.rb, test/ruby/test_module.rb,
+	  test/ruby/test_numeric.rb, test/ruby/test_range.rb,
+	  test/ruby/test_rational.rb, test/ruby/test_refinement.rb,
+	  test/ruby/test_rubyvm.rb, test/ruby/test_struct.rb,
+	  test/ruby/test_variable.rb, test/rubygems/test_gem_specification.rb,
+	  test/thread/test_queue.rb: Use Integer instead of Fixnum and Bignum.
+
+Tue May 17 15:26:10 2016  Tanaka Akira  <akr@fsij.org>
+
+	* [Feature #12005] Unify Fixnum and Bignum into Integer
+
+	* include/ruby/ruby.h (rb_class_of): Return rb_cInteger for fixnums.
+
+	* insns.def (INTEGER_REDEFINED_OP_FLAG): Unified from
+	  FIXNUM_REDEFINED_OP_FLAG and BIGNUM_REDEFINED_OP_FLAG.
+
+	* vm_core.h: Ditto.
+
+	* vm_insnhelper.c (opt_eq_func): Use INTEGER_REDEFINED_OP_FLAG instead
+	  of FIXNUM_REDEFINED_OP_FLAG.
+
+	* vm.c (vm_redefinition_check_flag): Use rb_cInteger instead of
+	  rb_cFixnum and rb_cBignum.
+	  (C): Use Integer instead of Fixnum and Bignum.
+
+	* numeric.c (fix_succ): Removed.
+	  (Init_Numeric): Define Fixnum as Integer.
+
+	* bignum.c (bignew): Use rb_cInteger instead of rb_cBignum.
+	  (rb_int_coerce): replaced from rb_big_coerce and return fixnums
+	  as-is.
+	  (Init_Bignum): Define Bignum as Integer.
+	  Don't define ===.
+
+	* error.c (builtin_class_name): Return "Integer" for fixnums.
+
+	* sprintf.c (ruby__sfvextra): Use rb_cInteger instead of rb_cFixnum.
+
+	* ext/-test-/testutil: New directory to test.
+	  Currently it provides utilities for fixnum and bignum.
+
+	* ext/json/generator/generator.c: Define mInteger_to_json.
+
+	* lib/mathn.rb (Fixnum#/): Redefinition removed.
+
+Tue May 17 11:58:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_BUILTIN_SETJMP): declare t as NORETURN
+	  to suppress warnings by -Wsuggest-attribute=noreturn.
+	  [ruby-core:75510] [Bug #12383]
+
+Tue May 17 10:40:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SETJMP): needs the header and proper
+	  arguments for builtin setjmp functions.
+
+Mon May 16 20:00:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.h: Additional uses of ONIG_CASE_MAPPING compilation switch
+
+Mon May 16 19:46:33 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* include/ruby/oniguruma.h: Introducing ONIG_CASE_MAPPING compilation
+	  switch
+
+	* include/ruby/oniguruma.h, enc/unicode.h: Using ONIG_CASE_MAPPING
+	  compilation switch
+
+Mon May 16 19:29:31 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: Update xmlrpc-0.1.1. xmlrpc-0.1.0 didn't allow
+	  to install on 2.4.0dev.
+
+Mon May 16 13:28:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (FUNC_STDCALL, FUNC_CDECL, FUNC_FASTCALL): set
+	  macro names explicitly to the old names, which are accidentally
+	  changed at r54985, for backward compatibilities.
+	  fiddle also depends on these names to fallback to ANSI names.
+	  [ruby-core:75494] [Bug #12377]
+
+Mon May 16 11:39:02 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/xmlrpc.rb, lib/xmlrpc/*, test/xmlrpc: XMLRPC is bundled gem
+	  on Ruby 2.4. It is extracted to https://github.com/ruby/xmlrpc
+	  [Feature #12160][ruby-core:74239]
+	* gems/bundled_gems: ditto.
+
+Mon May 16 06:06:21 2016  Eric Wong  <e@80x24.org>
+
+	* proc.c: fix RDoc of Proc#===/call/yield/[]
+	  [Bug #12332]
+
+Sun May 15 20:55:31 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/timeridconv.rb: don't use keeper thread. [Bug #12342]
+
+	* test/drb/ut_timerholder.rb: ditto.
+
+Sun May 15 16:15:25 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* array.c (rb_ary_entry): extract rb_ary_elt to organize if-conditions
+	  and check whether is embedded at once.
+
+Sun May 15 10:57:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_get_ev_const): warn deprecated constant even
+	  in the class context.  [ruby-core:75505] [Bug #12382]
+
+Sun May 15 03:13:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* iseq.h (struct iseq_compile_data): use struct rb_id_table
+	  instead of st_table.
+
+	* iseq.c (prepare_iseq_build): don't allocate ivar_cache_table
+	  until it has at least one element.
+
+	* iseq.c (compile_data_free): free ivar_cache_table only if it
+	  is allocated.
+
+	* compile.c (get_ivar_ic_value): allocate if the table is not
+	  allocated yet.
+
+Sat May 14 09:04:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (pkg_config): use xsystem consistently to set up
+	  library path environment variable as well as latter pkg-config
+	  calls.  [ruby-dev:49619] [Bug #12379]
+
+Sat May 14 00:16:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (make_seed_value): append leading-zero-guard and get
+	  rid of making a local copy of the seed.
+
+Fri May 13 08:46:42 2016  cremno  <cremno@mail.ru>
+
+	* NEWS: drop FreeBSD < 4 support.  [Fix GH-1339]
+	  The most recent version affected by this is 3.5 and was released
+	  in 2000.
+	  https://www.freebsd.org/releases/3.5R/announce.html
+	  https://en.wikipedia.org/wiki/History_of_FreeBSD#Version_history
+
+Fri May 13 03:12:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/defines.h (GCC_VERSION_SINCE): moved from internal.h.
+
+Fri May 13 03:11:20 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (__builtin_constant_p): check.
+
+	* internal.h: Use HAVE_BUILTIN___BUILTIN_CONSTANT_P
+
+Fri May 13 03:10:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: use alternative keyword
+	  to avoid macros conflicts with them.
+
+Thu May 12 01:54:08 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/mkmf.rb (try_func): get rid of conflict of declarations of
+	  main().  checking local symbol reference does not make sense.
+
+Thu May 12 00:18:19 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/Makefile.sub (HAVE_QSORT_S): use qsort_s only for Visual Studio
+	  2012 or later, because VS2010 seems to causes a SEGV in
+	  test/ruby/test_enum.rb.
+
+Wed May 11 23:59:47 2016  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_getivar): describe fast-path explicit
+	  (compiler friendly). [Bug #12274].
+
+Wed May 11 21:30:07 2016  Masaya Tarui  <tarui@ruby-lang.org>
+
+	* compile.c (iseq_compile_each): share InlineCache during same
+	  instance variable accesses. Reducing memory consumption,
+	  rising cache hit rate and rising branch prediction hit rate
+	  are expected. A part of [Bug #12274].
+
+	* iseq.h (struct iseq_compile_data): introduce instance
+	  variable IC table for sharing.
+
+	* iseq.c (prepare_iseq_build, compile_data_free):
+	  construct/destruct above table.
+
+Wed May 11 17:18:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* util.c (ruby_qsort): use qsort_s if available, for Microsoft
+	  Visual Studio 2005 (msvcr80.dll) and mingw.
+
+Wed May 11 10:33:26 2016  Marcus Stollsteimer  <sto.mar@web.de>
+
+	* COPYING: Remove trailing-whitespaces.
+	  [ci skip][fix GH-1348]
+
+Tue May 10 21:05:45 2016  Benoit Daloze  <eregontp@gmail.com>
+
+	* insns.def (defineclass): Also raise an error when redeclaring the
+	  superclass of a class as Object and it has another superclass.
+	  [Bug #12367] [ruby-core:75446]
+
+	* test/ruby/test_class.rb: test for above.
+
+Tue May 10 14:57:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (obj_random_bytes): base on bytes method instead of
+	  rand method, not to call toplevel rand method.
+
+Tue May 10 13:07:28 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (-fexcess-precision=standard): before r54895 -std=c99
+	  is specified and it implied -fexcess-precision=standard.
+	  Now with -std=gnu99, it should be explicitly specified.
+	  https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html
+
+Mon May  9 10:51:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_atfork_internal): move th to an argument.
+
+	* thread.c (rb_thread_atfork): do not repeat GET_THREAD().
+
+Mon May  9 10:46:36 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_atfork, rb_thread_atfork_before_exec): do
+	  nothing unless working fork is available.
+
+	* thread_sync.c (rb_mutex_abandon_all): define only if working
+	  fork is available.
+
+	* thread_sync.c (rb_mutex_abandon_keeping_mutexes): ditto.
+
+	* thread_sync.c (rb_mutex_abandon_locking_mutex): ditto.
+
+	* thread_win32.c (gvl_init): never used.
+
+Mon May  9 07:18:06 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/{defines,ruby}.h: need to define function attributes
+	  alternatives in defines.h instead of ruby.h, because they are used
+	  in oniguruma.h and the header used without including ruby.h at
+	  encoding library sources.
+
+Mon May  9 06:30:12 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* include/ruby/ruby.h (CONSTFUNC, PUREFUNC): fixed build errors on
+	  non-gcc build environments introduced at r54952.
+
+Mon May  9 02:51:51 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (rb_gc_unprotect_logging): throw rb_memerror when it cannot
+	  allocate memory. This is pointed out by Facebook's Infer.
+
+	* gc.c (gc_prof_setup_new_record): ditto.
+
+	* regparse.c (parse_regexp): ditto.
+
+	* util.c (MALLOC): use xmalloc and xfree like above.
+
+Mon May  9 02:39:16 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: check function attribute const and pure,
+	  and define CONSTFUNC and PUREFUNC if available.
+	  Note that I don't add those options as default because
+	  it still shows many false-positive (it seems not to consider
+	  longjmp).
+
+	* vm_eval.c (stack_check): get rb_thread_t* as an argument
+	  to avoid duplicate call of GET_THREAD().
+
+Sun May  8 21:01:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/openssl/extconf.rb: assume it doesn't have SSLv2 related
+	  functions when OPENSSL_NO_SSL2 is defined.
+	  Usually openssl's header and the library (libssl) have the same
+	  set of functions, but on some environment the library has functions
+	  whose headers doesn't declare. (openssl/opensslconf.h and libssl.so
+	  aren't be synchronized)
+	  To detect such case explicitly check feature macro and remove
+	  related functions.
+
+Sun May  8 18:51:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_home_dir_of): return the default home path if the
+	  user name is the current user name, on platforms where struct
+	  pwd is not supported.  a temporary measure against
+	  [Bug #12226].
+
+Sun May  8 08:51:38 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: add -Wsuggest-attribute=format and suppress warnings.
+
+Sun May  8 08:31:03 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: add -Wsuggest-attribute=noreturn and suppress warnings.
+
+Sun May  8 08:19:16 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: add -Werror=implicit-int to avoid missing type of
+	  function declaration.
+
+Sat May  7 22:22:37 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/webrick/ssl.rb: Accept string value for SSLCertName. It is used
+	  to invoke ssl server with command line.
+	  [fix GH-1329] Patch by @kerlin
+	* test/webrick/test_ssl_server.rb: Added test for GH-1329
+
+Sat May  7 21:55:12 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/webrick/test_ssl_server.rb: Added basic test for `webrick/ssl`
+
+Sat May  7 16:22:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (int_pair_to_real_inclusive): optimize to multiply
+	  without Bignum.
+
+Sat May  7 07:58:02 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* process.c (rb_exec_getargs): honor the expected argument types
+	  over the conversion method.  the basic language functionality
+	  should be robust.  [ruby-core:75388] [Bug #12355]
+
+Fri May  6 08:16:26 2016  David Silva  <david.silva@digital.cabinet-office.gov.uk>
+
+	* enum.c (enum_find): [DOC] add more examples to the documentation
+	  of Enumerable#detect, to show that it equals to Enumerable#find.
+	  [Fix GH-1340]
+
+Thu May  5 18:08:31 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* test/ruby/test_complexrational.rb: Remove duplicated raise.
+
+Thu May  5 14:41:05 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* string.c (rb_str_sub): Fix a special match variable name.
+	  [ci skip]
+
+Thu May  5 12:22:17 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_eval_cmd, rb_catch_obj): use TH_JUMP_TAG with the
+	  same rb_thread_t used for TH_PUSH_TAG, instead of JUMP_TAG with
+	  the current thread global variable.
+
+Thu May  5 10:49:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* random.c (fill_random_bytes_syscall): use arc4random_buf if
+	  available.
+
+Wed May  4 23:13:58 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* numeric.c (fix_plus): Remove rb_nucomp_add prototype
+	  declaration.
+
+	* numeric.c (fix_mul): Remove rb_nucomp_mul prototype
+	  declaration.
+
+	* internal.h (rb_nucomp_add, rb_nucomp_mul): add prototype
+	  declarations.
+
+Wed May  4 18:38:00 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/net/http/header.rb (Net::HTTPHeader#{each_header,each_name,
+	  each_capitalized_name,each_value,each_capitalized}): Return
+	  sized enumerators.
+
+	* test/net/http/test_httpheader.rb: add test for above.
+
+Wed May  4 17:53:15 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* lib/set.rb (Set#{delete_if,keep_if,collect!,reject!,select!,classify,divide},
+	  SortedSet#{delete_if,keep_if}): Return sized enumerators.
+
+	* test/test_set.rb: add test for above.
+
+Tue May  3 23:25:48 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* numeric.c: [DOC] Update result of 123456789 ** -2.
+	  [ruby-dev:49606] [Bug #12339]
+
+Tue May  3 23:13:16 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* internal.h (RCOMPLEX_SET_IMAG): undef RCOMPLEX_SET_IMAG
+	  instead of duplicated undef RCOMPLEX_SET_REAL.
+
+Tue May  3 22:55:07 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* complex.c (rb_complex_set_imag): Fix to properly set imag
+	  of complex.
+
+Tue May  3 22:19:55 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (warnflags): use -std=gnu99 instead of
+	  -std=iso9899:1999. [Feature #12336]
+
+Tue May  3 22:10:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (count_utf8_lead_bytes_with_word): Use __builtin_popcount
+	  only if it can use SSE 4.2 POPCNT whose latency is 3 cycle.
+
+	* internal.h (rb_popcount64): use __builtin_popcountll because now
+	  it is in fast path.
+
+Tue May  3 14:19:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (new_if_gen): set newline flag to NODE_IF to trace all
+	  if/elsif statements.  [ruby-core:67720] [Bug #10763]
+
+Tue May  3 05:35:54 2016  Eric Wong  <e@80x24.org>
+
+	* process.c (disable_child_handler_fork_child):
+	  initialize handler for SIGPIPE for !POSIX_SIGNAL
+
+Mon May  2 23:03:42 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c, include/ruby/win32.h (rb_w32_utruncate): implements new
+	  truncate alternative which accepts UTF-8 path.
+
+	* file.c (truncate): use above function.
+	  [Bug #12340]
+
+Mon May  2 20:59:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (str_coderange): to avoid function call when the string already
+	  has coderange information.
+
+	* re.c (rb_reg_prepare_enc): add shortcut path when the regexp has
+	  the same encoding of given string.
+
+	* re.c (rb_reg_prepare_re): avoid duplicated allocation of
+	  onig_errmsg_buffer.
+
+
+Mon May  2 12:34:52 2016  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_refinement.rb (test_inspect): Use Integer instead of
+	  Fixnum.
+
+Mon May  2 06:58:38 2016  Tanaka Akira  <akr@fsij.org>
+
+	* complex.c: Don't refer rb_cFixnum and rb_cBignum.
+	  (k_fixnum_p): Use FIXNUM_P.
+	  (k_bignum_p): Use RB_TYPE_P.
+
+Mon May  2 01:27:59 2016  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_numeric.rb (test_step): Use Integer::FIXNUM_MAX.
+
+Mon May  2 01:15:01 2016  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_module.rb (test_name): Use Integer instead of Fixnum.
+
+Mon May  2 01:00:04 2016  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/test/unit/assertions.rb (assert_fixnum): Defined.
+	  (assert_bignum): Defined.
+
+	* test/ruby/test_bignum.rb: Use assert_bignum.
+
+	* test/ruby/test_integer_comb.rb: Use assert_fixnum and assert_bignum.
+
+	* test/ruby/test_optimization.rb: Ditto.
+
+Mon May  2 00:41:53 2016  Tanaka Akira  <akr@fsij.org>
+
+	* vm_trace.c (recalc_remove_ruby_vm_event_flags): Add a cast to
+	  avoid signed integer overflow.
+
+Mon May  2 00:06:04 2016  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/envutil.rb: Define Integer::{FIXNUM_MIN,FIXNUM_MAX}.
+
+	* test/ruby/test_bignum.rb: Use Integer::{FIXNUM_MIN,FIXNUM_MAX}.
+
+	* test/ruby/test_bignum.rb: Ditto.
+
+	* test/ruby/test_integer_comb.rb: Ditto.
+
+	* test/ruby/test_marshal.rb: Ditto.
+
+	* test/ruby/test_optimization.rb: Ditto.
+
+Sun May  1 23:59:59 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* array.c (rb_ary_sum): fix for mathn
+
+	* test/ruby/test_array.rb (test_sum): ditto.
+
+Sun May  1 23:51:54 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* test/lib/test/unit.rb (Options#non_options): fixed wrong regexp.
+	  if both positives and negatives were specified, positives had to
+	  be specified from the beginning.
+
+Sun May  1 21:00:07 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c: drop Win2K support.
+
+Sun May  1 20:39:47 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* cont.c, hash.c, random.c, win32/win32.c: cleanup some Win9x/ME/NT4
+	  support leftovers.
+	  [fix GH-1328] patched by @cremno
+
+Sun May  1 07:30:44 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (search_nonascii): use nlz on big endian environments.
+
+	* internal.h (nlz_intptr): defined.
+
+Sun May  1 00:03:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (__builtin_ctz): check.
+
+	* configure.in (__builtin_ctzll): check.
+
+	* internal.h (rb_popcount32): defined for ntz_int32.
+	  it can use __builtin_popcount but this function is not used on
+	  GCC environment because it uses __builtin_ctz.
+	  When another function uses this, using __builtin_popcount
+	  should be re-considered.
+
+	* internal.h (rb_popcount64): ditto.
+
+	* internal.h (ntz_int32): defined for ntz_intptr.
+
+	* internal.h (ntz_int64): defined for ntz_intptr.
+
+	* internal.h (ntz_intptr): defined as ntz for uintptr_t.
+
+	* string.c (search_nonascii): unroll and use ntz.
+
+Sat Apr 30 21:54:13 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (Init_Numeric): Gather Fixnum method definitions.
+
+Sat Apr 30 21:28:14 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_div): Define Integer#/.
+
+	* bignum.c (rb_big_div): Don't define Bignum#/.
+
+	* lib/mathn.rb (Integer#/): Replace Integer#/ instead of Bignum#/.
+
+Sat Apr 30 21:11:08 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_plus): Define Integer#+.
+
+	* bignum.c (rb_big_plus): Don't define Bignum#+.
+
+Sat Apr 30 21:01:20 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_minus): Define Integer#-.
+
+	* bignum.c (rb_big_minus): Don't define Bignum#-.
+
+Sat Apr 30 20:53:33 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_mul): Define Integer#*.
+
+	* bignum.c (rb_big_mul): Don't define Bignum#*.
+
+Sat Apr 30 20:30:44 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_modulo): Define Integer#%.
+
+	* bignum.c (rb_big_modulo): Don't define Bignum#%.
+
+Sat Apr 30 20:17:08 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_equal): Define Integer#==.
+
+	* bignum.c (rb_big_eq): Don't define Bignum#==.
+
+Sat Apr 30 19:41:15 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_gt): Define Integer#>.
+
+	* bignum.c (rb_big_gt): Don't define Bignum#>.
+	  Renamed from big_gt.
+
+	* internal.h (rb_big_gt): Declared.
+
+Sat Apr 30 19:24:40 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_ge): Define Integer#>=.
+
+	* bignum.c (rb_big_ge): Don't define Bignum#>=.
+	  Renamed from big_ge.
+
+	* internal.h (rb_big_ge): Declared.
+
+Sat Apr 30 19:20:40 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/standard_library.rdoc: Remove obsoleted classes and modules.
+
+Sat Apr 30 19:09:23 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_lt): Define Integer#<.
+
+	* bignum.c (rb_big_lt): Don't define Bignum#<.
+	  Renamed from big_lt.
+
+	* internal.h (rb_big_lt): Declared.
+
+Sat Apr 30 18:44:05 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_le): Define Integer#<=.
+
+	* bignum.c (rb_big_le): Don't define Bignum#<=.
+	  Renamed from big_le.
+
+	* internal.h (rb_big_le): Declared.
+
+Sat Apr 30 18:11:44 2016  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (Init_Bignum): Define Integer::GMP_VERSION.
+
+Sat Apr 30 16:58:18 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_remainder): Define Integer#remainder.
+
+	* bignum.c (rb_big_remainder): Don't define Bignum#remainder.
+
+	* internal.h (rb_big_remainder): Declared.
+
+Sat Apr 30 15:29:24 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_uminus): {Fixnum,Bignum}#-@ is unified into
+	  Integer.
+
+	* bignum.c (rb_big_uminus): Don't define Bignum#-@.
+
+Sat Apr 30 14:42:20 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_idiv): {Fixnum,Bignum}#div is unified into
+	  Integer.
+
+	* bignum.c (rb_big_idiv): Don't define Bignum#div.
+
+Sat Apr 30 14:25:55 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_modulo): {Fixnum,Bignum}#modulo is unified into
+	  Integer.
+
+	* bignum.c (rb_big_modulo): Don't define Bignum#modulo.
+
+Sat Apr 30 14:04:30 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_divmod): {Fixnum,Bignum}#divmod is unified into
+	  Integer.
+
+	* bignum.c (rb_big_divmod): Don't define Bignum#divmod.
+
+Sat Apr 30 13:20:00 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_fdiv): {Fixnum,Bignum}#fdiv is unified into
+	  Integer.
+
+	* bignum.c (rb_big_fdiv): Don't define Bignum#fdiv.
+
+Sat Apr 30 12:25:43 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_pow): {Fixnum,Bignum}#** is unified into
+	  Integer.
+
+	* bignum.c (rb_big_pow): Don't define Bignum#**.
+
+Sat Apr 30 12:28:59 2016  Tanaka Akira  <akr@fsij.org>
+
+	* bignum.c (rb_big_comp): Renamed from rb_big_neg.
+
+	* numeric.c (fix_comp): Renamed from fix_rev.
+
+Sat Apr 30 12:25:43 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_comp): {Fixnum,Bignum}#~ is unified into
+	  Integer.
+
+	* bignum.c (rb_big_neg): Don't define Bignum#~.
+
+	* internal.h (rb_big_neg): Declared.
+
+Sat Apr 30 12:07:42 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_and): {Fixnum,Bignum}#& is unified into
+	  Integer.
+
+	* bignum.c (rb_big_and): Don't define Bignum#&.
+
+Sat Apr 30 11:56:15 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/thread: removed dummy extension library.  thread_sync.c
+	  provides "thread.rb" already.
+
+Sat Apr 30 11:53:48 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_or): {Fixnum,Bignum}#| is unified into
+	  Integer.
+
+	* bignum.c (rb_big_or): Don't define Bignum#|.
+
+Sat Apr 30 11:18:47 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* vm_trace.c: Fix typos. [ci skip]
+
+Sat Apr 30 10:09:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/pty/pty.c (establishShell): honor USER environment variable
+	  and login name over uid, one uid can be shared by some login
+	  names.
+
+Fri Apr 29 22:40:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/maintainers.rdoc (ext/io/nonblock): still maintained, as
+	  well as ext/io/wait, which is the origin.
+
+Fri Apr 29 21:18:12 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* doc/maintainers.rdoc (ext/win32): of course, it's still maintained.
+
+Fri Apr 29 21:03:10 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* gems/bundled_gems: Update latest gems, test-unit-3.1.8 and rake-11.1.2.
+
+Fri Apr 29 20:43:02 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/maintainers.rdoc: Update latest maintainers list on Ruby 2.4
+
+Fri Apr 29 19:52:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/maintainers.rdoc: Removed deprecated entries. These are already deleted.
+
+Fri Apr 29 19:48:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/maintainers.rdoc: Removed Ruby 1.8 entries. It's not necessary to
+	  Ruby 2.4 or later.
+
+Thu Apr 28 17:03:17 2016  Nicholas Maccharoli  <nmaccharoli@gmail.com>
+
+	* hash.c (rb_hash_update_{block,func}_callback): dry up hash
+	  update callback code.  [Fix GH-1338]
+
+Thu Apr 28 16:52:05 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_prepare_enc): use rb_enc_asciicompat(enc) instead of
+	  rb_enc_str_asciicompat_p(str) to avoid useless rb_enc_get(str) call.
+
+Thu Apr 28 16:33:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb: [DOC] fix example code.  base on the code by
+	  Semyon Gaivoronskiy in [ruby-core:75224].  [Bug #12323]
+
+Thu Apr 28 09:33:03 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.4.
+	  Please see entries of 2.6.4 on
+	  https://github.com/rubygems/rubygems/blob/master/History.txt
+
+Thu Apr 28 04:49:07 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* configure.in (rb_cv_lgamma_r_pm0): check if lgamma_r(+0.0)
+	  returns positive infinity, in addition to lgamma_r(-0.0).
+	  AIX returns an incorrect result of negative infinity.
+
+	* math.c (ruby_lgamma_r): handle +0.0, in addition to -0.0.
+
+Thu Apr 28 01:11:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c: define _DEFAULT_SOURCE because glibc 2.20 deprecates
+	  _BSD_SOURCE.
+	  https://sourceware.org/glibc/wiki/Release/2.20
+
+Thu Apr 28 00:27:55 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_xor): {Fixnum,Bignum}#^ is unified into
+	  Integer.
+
+	* bignum.c (rb_big_xor): Don't define Bignum#^.
+
+Wed Apr 27 20:53:59 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_aref): {Fixnum,Bignum}#[] is unified into
+	  Integer.
+
+	* bignum.c (rb_big_aref): Don't define Bignum#<<.
+
+	* internal.h (rb_big_aref): Declared.
+
+Wed Apr 27 16:10:35 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/instruction.rb: fix to follow current implementation.
+
+Wed Apr 27 15:47:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/stringio/stringio.c (strio_s_new): warn if a block is given,
+	  as well as IO.new.
+
+Wed Apr 27 14:29:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (ruby_only_for_internal_use): raise fatal error when
+	  deprecated function only for internal use is called, not just a
+	  warning.
+
+Tue Apr 26 23:42:30 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (rel): should not raise exceptions even if
+	  the user input is wrong.  only reports the error and continue process.
+
+Tue Apr 26 23:35:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (process_options): convert -e script to the encoding
+	  given by a command line option on Windows.  assume it is the
+	  expected encoding.  [ruby-dev:49461] [Bug #11900]
+
+Tue Apr 26 21:11:02 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_lshift): {Fixnum,Bignum}#<< is unified into
+	  Integer.
+
+	* bignum.c (rb_big_lshift): Don't define Bignum#<<.
+
+Tue Apr 26 20:59:40 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_rshift): {Fixnum,Bignum}#>> is unified into
+	  Integer.
+
+	* bignum.c (rb_big_rshift): Don't define Bignum#>>.
+
+Tue Apr 26 20:46:16 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_size): {Fixnum,Bignum}#size is unified into Integer.
+
+	* bignum.c (rb_big_size_m): Don't define Bignum#size.
+
+	* internal.h (rb_big_size_m): Declared.
+
+Tue Apr 26 20:09:08 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (rb_int_bit_length): {Fixnum,Bignum}#bit_length is
+	  unified into Integer.
+
+	* bignum.c (rb_big_bit_length): Don't define Bignum#bit_length.
+
+	* internal.h (rb_big_bit_length): Declared.
+
+Tue Apr 26 19:56:16 2016  Tanaka Akira  <akr@fsij.org>
+
+	* numeric.c (int_abs): Integer#{abs,magnitude} moved from
+	  Fixnum and Bignum.
+
+	* bignum.c (rb_big_abs): Don't define Bignum#{abs,magnitude}.
+
+	* internal.h (rb_big_abs): Declared.
+
+Mon Apr 25 14:39:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/rbconfig/sizeof/extconf.rb: just check the existence of each
+	  types, to reduce configuration time, especially cross-compiling.
+
+	* template/sizes.c.tmpl: calculate sizes of checked types at
+	  compilation time.
+
+Mon Apr 25 11:27:27 2016  Marcus Stollsteimer  <sto.mar@web.de>
+
+	* doc/extension.rdoc: Improvements to english grammars.
+	  [Bug #12246][ruby-core:74792][ci skip]
+
+Mon Apr 25 11:17:50 2016  Marcus Stollsteimer  <sto.mar@web.de>
+
+	* encoding.c: Fix return value of `Encoding::ISO8859_1.name`
+	  [Bug #12313][ruby-core:75147][ci skip]
+	* ext/bigdecimal/bigdecimal.c: Fix code sample of `BigDecimal.new`
+
+Sun Apr 24 23:29:16 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* configure.in: add missing -lm for AIX.
+
+Sun Apr 24 18:33:58 2016  Kazuki Tsujimoto  <kazuki@callcc.net>
+
+	* vm_insnhelper.c (INLINE): disable r54738 if __NO_INLINE__ is defined.
+	  It caused "undefined reference to `vm_getivar'".
+
+Sun Apr 24 09:32:12 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* test/ruby/test_array.rb: Add test cases for Array#sum with
+	  non-numeric objects.
+
+Sun Apr 24 04:21:27 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_insnhelper.c (INLINE): define as `inline` when it is optimized.
+	  define as `static inline` when it is not optimized to keep
+	  the symbol generated.
+
+	* vm_insnhelper.c (vm_getivar): use `INLINE` to force inline
+	  so that a compiler inlines it into vm_getinstancevariable
+	  and optimizes out is_attr and related branches.
+
+	* vm_insnhelper.c (vm_getivar): use `inline` to recommend inline.
+	  Without this vm1_ivar_set is degraded.
+
+	  benchmark results:
+	  minimum results in each 5 measurements.
+	  Execution time (sec)
+	  name    ruby 2.4.0dev (2016-04-23 trunk 54727) [x86_64-linux]   ruby 2.4.0dev (2016-04-23 trunk 54733) [x86_64-linux]       built-ruby
+	  loop_whileloop 0.641                                                   0.642 0.646
+	  vm1_ivar* 1.002                                                   0.999 0.831
+	  vm1_ivar_set* 0.369                                                   1.106 0.362
+
+	  Speedup ratio: compare with the result of `ruby 2.4.0dev (2016-04-23
+	  trunk 54727) [x86_64-linux]' (greater is better)
+	  name    ruby 2.4.0dev (2016-04-23 trunk 54733) [x86_64-linux]
+	  built-ruby
+	  loop_whileloop
+	  0.998        0.991
+	  vm1_ivar*
+	  1.003        1.205
+	  vm1_ivar_set*
+	  0.334        1.018
+
+Sat Apr 23 18:01:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* vm_insnhelper.c (vm_getivar): specify inline instead of static inline.
+	  vm_getivar is called by vm_call_ivar and vm_getinstancevariable.
+	  At least with GCC 4.8 and 5.3 on Linux, they are inlining it into
+	  vm_call_ivar but not vm_getinstancevariable.
+	  By `inline`, they correctly inline it and gains performance.
+
+	  Speedup ratio: compare with the result of `ruby 2.4.0dev (2016-04-23
+	  trunk 54727) [x86_64-linux]' (greater is better)
+	  name    built-ruby
+	  loop_whileloop       1.001
+	  vm1_ivar*            1.189
+	  vm1_ivar_set*        1.024
+
+	  Note the `inline`'s meaning is different between old GCC
+	  and C99. Old GCC's inline means C99's extern inline.
+	  https://gcc.gnu.org/onlinedocs/gcc/Inline.html
+	  Since Ruby specify -std=iso9899:1999, it works like C99.
+
+Sat Apr 23 16:11:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_mul_size_overflow): use UNLIKELY
+	  by user side to improve generality.
+
+Sat Apr 23 16:10:02 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_LIKELY): use prefix in ruby.h.
+
+	* intern.h (LIKELY): define with RB_LIKELY.
+
+Sat Apr 23 13:27:25 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* NEWS: Add descriptions for Time#to_time updates.
+	  [Bug #12271]
+
+Sat Apr 23 13:21:24 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* NEWS: Add descriptions for DateTime#to_time updates.
+	  [Bug #12189]
+
+Sat Apr 23 11:21:27 2016  Marcus Stollsteimer  <sto.mar@web.de>
+
+	* ext/date/date_core.c (Init_date_core): [DOC] Convert DateTime
+	  documentation to RDoc from Markdown.
+	  [ruby-core:75136] [Bug #12311]
+
+Sat Apr 23 09:03:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c: cygwin does not use w32_cmdvector, command line can be
+	  other than UTF-8.  [ruby-dev:49519] [Bug #12184]
+
+Sat Apr 23 01:00:03 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* configure.in: don't use the system-provided round(3) on AIX.
+	  In AIX, round(0.49999999999999994) returns 1.0.
+	  Use round() in numeric.c instead.
+
+Fri Apr 22 21:00:44 2016  Tanaka Akira  <akr@fsij.org>
+
+	* test/ruby/test_time_tz.rb: Tests depends on Europe/Moscow removed
+	  to avoid test failures due to the tzdata change.
+	  https://github.com/eggert/tz/commit/8ee11a301cf173afb0c76e0315b9f9ec8ebb9d95
+	  Found by naruse.
+
+Fri Apr 22 20:18:40 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_mul_size_overflow): added to handle
+	  mul overflow efficiently.
+
+	* include/ruby/ruby.h (rb_alloc_tmp_buffer2): use rb_mul_size_overflow
+	  and avoid division where it can define DSIZE_T.
+
+	* gc.c (xmalloc2_size): moved from ruby.h and use rb_mul_size_overflow.
+
+Fri Apr 22 20:34:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* time.c (time_asctime): [DOC] add ctime example, not only
+	  asctime.  [ruby-core:75126] [Bug #12310]
+
+Fri Apr 22 18:44:32 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* variable.c: use uint32_t instead of long to avoid confusion about
+	  the type of ivtbl->numiv.
+
+Fri Apr 22 15:09:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval_jump.c (exec_end_procs_chain): restore previous error info
+	  for each end procs.  [ruby-core:75038] [Bug #12302]
+
+Fri Apr 22 15:04:56 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: the fullpath of merger.rb is too long to
+	  copy&paste on Windows.  show shorter name instead on the platform.
+	  I'm sure that the user of this command on Windows is only me.
+
+Fri Apr 22 14:52:04 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/merger.rb: remove temporary file.
+
+Fri Apr 22 11:27:03 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/http.rb: Improve documentation for SSL requests via GET method.
+	  [fix GH-1325][ci skip] Patch by @jsyeo
+
+Fri Apr 22 10:51:13 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/webrick/ssl.rb: Support to add SSLCiphers option.
+	  [fix GH-1321] Patch by @rhadoo
+
+Fri Apr 22 10:43:19 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* file.c, win32/file.c: Removed obsoleted safe level checks.
+	  [fix GH-1327] Patch by @cremno
+
+Fri Apr 22 10:01:48 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* benchmark/bm_so_meteor_contest.rb: fix a typo.
+	  [fix GH-1330][ci skip] Patch by @sachin21
+
+Fri Apr 22 04:57:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (rb_alloc_tmp_buffer_with_count): added like xmalloc2 to
+	  avoid duplicated check of size.
+
+	* gc.c (ruby_xmalloc2): added to keep separate layers.
+
+	* include/ruby/ruby.h (rb_alloc_tmp_buffer2): added to check
+	  the size more statically.
+
+Fri Apr 22 04:54:40 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (LIKELY): moved from internal.h.
+
+	* include/ruby/ruby.h (UNLIKELY): ditto.
+
+Thu Apr 21 01:44:19 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* gc.c (objspace_malloc_prepare): remove size check because it is
+	  used by objspace_xmalloc and objspace_xcalloc.
+	  objspace_xmalloc introduces its own check in this commit.
+	  objspace_xcalloc checks with xmalloc2_size (ruby_xmalloc2_size).
+
+	* gc.c (objspace_xmalloc0): common xmalloc function.
+
+	* gc.c (objspace_xmalloc): introduce its own size check.
+
+	* gc.c (objspace_xmalloc2): separated from ruby_xmalloc2 to clarify
+	  the layer who has the responsibility to check the size.
+
+	* gc.c (objspace_xrealloc): remove duplicated size check.
+
+	* gc.c (ruby_xmalloc2): use objspace_xmalloc2.
+
+	* include/ruby/ruby.h (ruby_xmalloc2_size): follow the size limit
+	  as SSIZE_MAX. Note that ISO C says size_t is unsigned integer.
+
+Thu Apr 21 12:14:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if succeeded in creating config.h.
+
+	* tool/ifchange: ignore failures when TEST_COLORS unmatched. just
+	  use the default value if expected name is not contained in it.
+	  [ruby-core:75046] [Bug #12303]
+
+Wed Apr 20 17:33:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/cgi/escape/escape.c (cgiesc_unescape): define unescape
+	  method instead of _unescape, and should pass the optional
+	  argument to the super method.
+
+	* lib/cgi/util.rb (CGI::Util#_unescape): remove intermediate
+	  method.
+
+Wed Apr 20 15:52:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (syntax_error_initialize): move the default message,
+	  "compile error", from parse.y.  the default parameter should
+	  belong to the class definition.
+
+	* parse.y (yycompile0): use the default parameter.
+
+Wed Apr 20 10:25:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (append_compile_error): use rb_syntax_error_append.
+
+	* error.c (rb_syntax_error_append): append messages into a
+	  SyntaxError exception instance.
+
+	* parse.y (yycompile0): make new SyntaxError instance in main
+	  mode, otherwise error_buffer should be a SyntaxError if error
+	  has occurred.
+
+Tue Apr 19 17:42:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (err_vcatf): rename, and separate appending message from
+	  creating a string buffer.
+
+	* error.c (rb_syntax_error_append): merge rb_error_vsprintf and
+	  rb_compile_err_append.
+
+	* parse.y (parser_compile_error): use rb_syntax_error_append.
+
+Tue Apr 19 13:46:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (append_compile_error, compile_bug): pass iseq and get
+	  error info and file from it, not by the thread error info.
+
+	* error.c (rb_report_bug_valist): take va_list instead of variadic
+	  arguments, and just report the bug but not abort.
+
+Tue Apr 19 13:18:12 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/time.rb: revert r54167 because it would break
+	  backward compatibilities, and it is documented that
+	  Time.parse does not take into account time zone
+	  abbreations other than ones described in RFC 822
+
+Tue Apr 19 13:12:03 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* ChangeLog: Fix dates of previous commits
+
+Tue Apr 19 12:45:03 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* ChangeLog: Add descriptions for logger updates
+	* NEWS: Add descriptions for logger updates
+
+Tue Apr 19 12:45:02 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/logger.rb: Add shift_period_suffix option
+
+Tue Apr 19 12:45:01 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/logger.rb: Allow specifying logger parameters in constructor
+	  such as level, progname, datetime_format, formatter.
+
+Mon Apr 18 16:07:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_peephole_optimize): should not replace the
+	  current target INSN, not to follow the replaced dangling link in
+	  the caller.  [ruby-core:74993] [Bug #11816]
+
+Mon Apr 18 12:56:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_truncate): add an optional parameter, digits, as
+	  well as Float#round.  [Feature #12245]
+
+	* numeric.c (int_truncate): add an optional parameter, digits, as
+	  well as Integer#round.  [Feature #12245]
+
+Sun Apr 17 04:18:56 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: revisions are strings.
+
+Sat Apr 16 14:26:49 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* ext/date/date_core.c : remove not used f_getlocal macro.
+	  After r54553 f_getlocal macro is not used.
+
+Sat Apr 16 14:15:24 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* ext/date/date_core.c : remove not used f_utc6 macro.
+	  After r54169 f_utc6 macro is not used.
+
+Sat Apr 16 10:00:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* struct.c (struct_make_members_list, rb_struct_s_def): member
+	  names should be unique. [ruby-core:74971] [Bug #12291]
+
+	* struct.c (struct_make_members_list): extract making member name
+	  list from char* va_list, with creating symbols without
+	  intermediate IDs.
+
+Sat Apr 16 01:33:27 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: sort revisions.
+
+Sat Apr 16 01:16:02 2016  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_sum): Don't yield same element twice.
+	  Found by nagachika.
+
+Sat Apr 16 01:03:32 2016  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_sum): Fix SEGV by [1/2r, 1].sum.
+
+Fri Apr 15 23:52:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* rational.c (rb_rational_plus): rename from rb_rational_add
+	  to be aligned with rb_fix_plus.
+
+	* array.c (rb_ary_sum): ditto.
+
+	* internal.h: ditto.
+
+Fri Apr 15 23:42:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* rational.c (rb_rational_add): rename from nurat_add.
+
+	* array.c (rb_ary_sum): use rb_rational_add directly.
+
+	* test/ruby/test_array.rb (test_sum): add assertions for an array of
+	  Rational values.
+
+Fri Apr 15 22:31:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* array.c (rb_ary_sum): apply the precision compensated algorithm
+	  for an array in which Rational and Float values are mixed.
+
+	* test/ruby/test_array.rb (test_sum): add assertions for the above
+	  change.
+
+Fri Apr 15 22:30:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_setname): defer setting native thread name
+	  set in initialize until the native thread is created.
+	  [ruby-core:74963] [Bug #12290]
+
+Fri Apr 15 20:27:16 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/irb/ext/save-history.rb: Fix NoMethodError when method is not defined.
+
+Fri Apr 15 15:38:58 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* common.mk (benchmark): order options for built-ruby and compare-ruby.
+
+Fri Apr 15 14:14:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* test/ruby/test_array.rb (test_sum): add assertions for Rational and
+	  Complex numbers.
+
+Fri Apr 15 10:07:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (console_key_pressed_p): raise the same
+	  exception, "unknown virtual key code", for names with nul chars.
+	  though console_win32_vk() considers the length and can deal with
+	  nul chars, rb_sprintf() raised at PRIsVALUE previously, so quote
+	  it if it is unprintable.
+
+Fri Apr 15 09:02:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/console/console.c (rb_sym2str): fallback definition for
+	  older ruby.  [ruby-core:74953] [Bug #12284]
+
+Thu Apr 14 21:46:36 2016  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_sum): Support the optional argument, init, and
+	  block.
+
+Thu Apr 14 19:02:41 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/irb/ext/save-history.rb: suppress warning: method redefined;
+	  discarding old save_history=.
+
+Thu Apr 14 14:58:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tkutil/tkutil.c (tk_hash_kv): the third argument can be
+	  nil not only an Array.  reported by @windwiny at
+	  https://github.com/ruby/ruby/commit/cdaa94e#commitcomment-17096618
+
+Thu Apr 14 14:28:55 2016  cremno phobia  <cremno@mail.ru>
+
+	* cont.c (fiber_initialize_machine_stack_context): fix wrong
+	  _MSC_VER check, should be decimal but not hexadecimal.
+	  [ruby-core:74936] [Bug #12279]
+
+Wed Apr 13 22:51:38 2016  Tanaka Akira  <akr@fsij.org>
+
+	* array.c (rb_ary_sum): Array#sum is implemented.
+	  Kahan's compensated summation algorithm for precise sum of float
+	  numbers is moved from ary_inject_op in enum.c.
+
+	* enum.c (ary_inject_op): Don't specialize for float numbers.
+
+	  [ruby-core:74569] [Feature #12217] proposed by mrkn.
+
+Wed Apr 13 15:56:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flo_ceil): add an optional parameter, digits, as
+	  well as Float#round.  [Feature #12245]
+
+	* numeric.c (flo_floor): add an optional parameter, digits, as
+	  well as Float#round.  [Feature #12245]
+
+	* numeric.c (int_ceil): add an optional parameter, digits, as
+	  well as Integer#round.  [Feature #12245]
+
+	* numeric.c (int_floor): add an optional parameter, digits, as
+	  well as Integer#round.  [Feature #12245]
+
+Wed Apr 13 14:47:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (Init_File): add alias File.empty? to File.zero?.
+	  [Feature #9969]
+
+Wed Apr 13 14:36:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (assign_in_cond): allow multiple assignment in
+	  conditional expression.  [Feature #10617]
+
+Wed Apr 13 14:11:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_big_size): add wrapper function of BIGSIZE and
+	  rename the method function with _m suffix.
+
+	* numeric.c (int_round_zero_p): extracted from rb_int_round.
+	  optimize for Bignum, and convert VALUE returned by Numeric#size
+	  to long.
+
+Wed Apr 13 12:00:08 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_basicinstructions.rb: add a test to check access
+	  instance variables on special const objects.
+
+	  All of such objects are frozen, so that we can not set instance
+	  variables for them. But we can read instance variables and return
+	  default value (nil).
+
+Tue Apr 12 20:40:35 2016  Kaneko Yuichiro  <spiketeika@gmail.com>
+
+	* ext/date/date_core.c (time_to_time): should preserve timezone
+	  info.  [ruby-core:74889] [Bug #12271]
+
+Tue Apr 12 11:51:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (new_label_body): initialize bit fields, since
+	  compile_data_alloc does not clear the memory.  [Bug #12082]
+
+Mon Apr 11 20:18:43 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_backtrace.c (frame2klass): filter only for imemo_ment.
+	  T_IMEMO/imemo_iseq can be passed here.
+
+Mon Apr 11 17:43:04 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_optimize): disable tail call optimization in
+	  rescued, rescue, and ensure blocks.
+	  [ruby-core:73871] [Bug #12082]
+
+Mon Apr 11 06:54:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (apply2files): apply to a VALUE vector instead of a
+	  temporary array.
+
+Sun Apr 10 20:54:16 2016  Joe Swatosh  <joe.swatosh@gmail.com>
+
+	* ext/win32/lib/win32/registry.rb (DeleteValue, DeleteKey): fix
+	  API names.  [ruby-core:74863] [Bug #12264]
+
+Sun Apr 10 17:47:42 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (rb_realpath_internal): no argument conversions since
+	  this internal function does not need to_path and encoding
+	  conversions, not to be affected by the default internal
+	  encoding.
+
+Sat Apr  9 10:03:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* load.c (rb_f_load): raise with the original path name before
+	  encoding conversion.
+
+Sat Apr  9 02:05:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* bignum.c (rb_cstr_parse_inum): [EXPERIMENTAL] new function to
+	  parse integer in C-string with length.  the name and the
+	  arguments may be changed in the future.
+
+	* bignum.c (rb_str_to_inum): preserve encoding of the argument in
+	  error messages, and no longer needs to copy non-terminated
+	  strings.
+
+	* bignum.c (rb_str2big_{poweroftwo,normal,karatsuba,gmp}): ditto.
+
+Thu Apr  7 19:04:03 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/regexp.rdoc (comments): [DOC] terminators cannot appear in
+	  comments.  [ruby-core:74838] [Bug #12256]
+
+Thu Apr  7 11:24:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/tk/tkutil/tkutil.c (cbsubst_initialize): fix out-of-bound
+	  access when no arguments given.  `p Tk::Event.new` crashed.
+
+Fri Apr  1 01:26:00 2016  Benoit Daloze  <eregontp@gmail.com>
+
+	* ext/coverage/coverage.c: Fully reset coverage to not persist global state.
+	  It was returning old file coverages as empty arrays to the user.
+	  [ruby-core:74596] [Bug #12220]
+
+	* ext/coverage/coverage.c (rb_coverages): remove unused static state.
+
+	* thread.c: Moved and renamed coverage_clear_result_i to reset_coverage_i.
+
+	* test/coverage/test_coverage.rb: improve precision of tests.
+
+Wed Apr  6 22:41:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_lgamma_r_m0): fix the condition for
+	  lgamma_r(-0.0).  [Bug #12249]
+
+Wed Apr  6 17:38:42 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* tool/downloader.rb (RubyGems.download): follow the change of the
+	  rubygems ssl_certs directory tree introduced by previous commit.
+
+Wed Apr  6 15:00:27 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.3.
+	  Please see entries of 2.6.3 on
+	  https://github.com/rubygems/rubygems/blob/master/History.txt
+
+Wed Apr  6 14:13:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (rb_cv_lgamma_r_m0): check if lgamma_r(-0.0)
+	  returns negative infinity.  [Bug #12249]
+
+	* math.c (ruby_lgamma_r): define by the configured result.
+
+Wed Apr  6 10:56:15 2016  Anton Davydov  <antondavydov.o@gmail.com>
+
+	* lib/logger.rb (Logger#level=): remove unnecessary local
+	  variable.
+
+	* lib/logger.rb (Logger#initialize, Logger#reopen): [DOC] mention
+	  the default values.  cherrypicked from [GH-1319].
+
+Wed Apr  6 10:17:53 2016  cremno phobia  <cremno@mail.ru>
+
+	* math.c (ruby_lgamma_r): missing/lgamma_r.c is used on Windows,
+	  since msvcrt does not provide it.
+
+	* missing/lgamma_r.c (lgamma_r): fix lgamma(-0.0).
+	  [ruby-core:74823] [Bug #12249]
+
+Wed Apr  6 01:22:55 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* math.c (ruby_lgamma_r): mswin's lgamma_r also seems to be wrong.
+	  cf. [Bug #12249]
+
+Wed Apr  6 00:53:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* math.c (ruby_lgamma_r): fix lgamma(-0.0) on mingw and OSX.
+
+	* math.c (ruby_tgamma): fix tgamma(-0.0) on mingw.
+	  [ruby-core:74817] [Bug #12249]
+
+Tue Apr  5 14:50:28 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c (mime_putc): fix typo.
+	  [Bug #12202] [ruby-core:74802]
+
+Tue Apr  5 00:06:44 2016  Aeris  <aeris@imirhil.fr>
+
+	* ext/openssl/ossl_ssl.c (ossl_ssl_tmp_key): Access to ephemeral
+	  TLS session key in case of forward secrecy cipher.  Only
+	  available since OpenSSL 1.0.2.  [Fix GH-1318]
+
+	* ext/openssl/extconf.rb: Check for SSL_get_server_tmp_key.
+
+Mon Apr  4 23:37:05 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (rb_vm_struct): make at_exit a single linked list but
+	  not RArray, not to mark the registered functions by the write
+	  barrier.  based on the patches by Evan Phoenix.
+	  [ruby-core:73908] [Bug #12095]
+
+Mon Apr  4 17:43:45 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change default value of
+	  RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO 0.3 -> 0.2
+	  RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO 0.8 -> 0.65
+
+	  These values are same as Ruby 2.0.0.
+
+	  This change cause GC counts.
+	  However, generational GC reduced each (minor) GC time and
+	  increase memory locality. So that not so big impact on my
+	  benchmarking results.
+	  (surprisingly, this fix speed up programs on some cases)
+
+	  You can change these values by environment variables
+	  if you feel wrong.
+
+Mon Apr  4 17:36:52 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (get_envparam_double): take an upper_bound.
+
+	  And also take an accept_zero flag which allow to accept zero
+	  even if lower_bound is set.
+
+	* gc.c (ruby_gc_set_params): fix parameters.
+
+	  RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO set 0.9 as *lower_bound*, so that
+	  it should be upper_bound.
+	  Set RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO as lower bound.
+
+	  Also set lower/upper bound of RUBY_GC_HEAP_FREE_SLOTS_GOAL_RATIO to
+	  RUBY_GC_HEAP_FREE_SLOTS_MIN/MAX_RATIO.
+
+Mon Apr  4 16:41:32 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (Init_VM): should pass tokens.
+
+Sun Apr  3 09:34:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (flodivmod): round division if it is a finite number
+	  and module is required.
+
+	* numeric.c (dbl2ival): do not round here.
+
+	* numeric.c (flo_ceil): use dbl2ival.
+
+	* numeric.c (flo_round): round explicitly.
+
+Sat Apr  2 15:24:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/intern.h (rb_check_arity): returns argc.
+
+Fri Apr  1 20:58:33 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Data generation to implement
+	  swapcase functionality for titlecase characters. Swapcase isn't defined
+	  by Unicode, because the purpose/usage of swapcase is unclear anyway.
+	  The implementation follows a proposal from Nobu, swapping the case of
+	  each component of a titlecase character individually.
+	  This means that the titlecase characters have to be decomposed.
+
+	* enc/unicode.c: Code using the above data.
+
+	* test/ruby/enc/test_case_mapping.rb: Tests for the above.
+
+Fri Apr  1 14:55:28 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* configure.in (AC_CONFIG_FILES): $srcdir/.git can be a file pointing
+	  the real git_dir, such as when the git working tree is a "linked
+	  working tree" (a working tree created by git-worktree). So use
+	  git-rev-parse --git-dir to check if $srcdir is the top-level of a git
+	  repository, not just checking if the $srcdir/.git directory does exist
+	  or not.  [ruby-core:74759] [Bug #12239]
+
+	* tool/change_maker.rb: use tool/vcs.rb to detect VCS. This used to have
+	  its own VCS detection code, while we have tool/vcs.rb.
+
+	* tool/vcs.rb (detect): remove code duplication
+
+Fri Apr  1 04:50:44 2016  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_s_alloc):
+	  enable SSL_MODE_ACCEPT_MOVING_WRITE_BUFFER by default
+	  [Bug #12126]
+
+Fri Apr  1 01:13:55 2016  Benoit Daloze  <eregontp@gmail.com>
+
+	* thread.c (update_coverage): Do not track coverage in loaded files
+	  after Coverage.result. Avoids out-of-bounds access. [Bug #12237]
+
+	* ext/coverage/coverage.c (coverage_clear_result_i): document.
+
+Thu Mar 31 19:16:16 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: need to set initial value of GC_HEAP_FREE_SLOTS_GOAL_RATIO.
+
+Thu Mar 31 17:50:27 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: change additional allocation policy.
+
+	  Introduce new environment variable
+	  RUBY_GC_HEAP_FREE_SLOTS_GOAL_RATIO (goal_ratio) to calculate the
+	  ratio of additional memory.
+
+	  Before this change, we add pages with the following formula
+	  (when free_slots < total_pages * RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO):
+	    next_pages = total_pages * RUBY_GC_HEAP_GROWTH_FACTOR
+
+	  This addition can allocate too much.
+
+	  With this change, we increase pages to satisfy the following formula:
+	    next_free_slots = next_total_slots * goal_ratio
+	  where
+	    next_free_slots = free_slots + adding_slots
+	    next_total_slots = total_slots + adding_slots.
+
+	  If you want to prepare many free slots, increase this ratio.
+
+	  If this variable is 0, then simply multiply
+	  RUBY_GC_HEAP_GROWTH_FACTOR.
+
+	* gc.c (get_envparam_double): enable to accept 0.
+
+Thu Mar 31 17:48:25 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_marks_finish): fix syntax error.
+
+Thu Mar 31 16:49:36 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: simplify allocate/free detecting logic at the end of marking.
+
+	  Before this change, heap_pages_min_slots are calculated at the
+	  beginning sweeping phase. And this value is used at the end of
+	  *next* marking phase.
+
+	  To simplify it, we use this value at the end of this marking phase.
+	  It means that we don't need to store this value as global state.
+
+	  Also heap_pages_max_slots is calculated at the begging of sweeping
+	  phase and used at the end of sweeping phase.
+	  To simplify this logic, we introduced new global value
+	  heap_pages_freeable_pages it means extra pages count we can free.
+	  gc_sweep_step() checks this value and moves empty pages to tomb_heap
+	  not more than this value.
+
+	  Because of this fix, heap_pages_swept_slots is no longer needed.
+
+	* gc.c (rb_objspace_t::heap_pages): restruct the objspace global
+	  status.
+
+	  remove the following fields
+	    * swept_slots (and heap_pages_swept_slots)
+	    * min_free_slots (and heap_pages_min_free_slots)
+	    * max_free_slots (and heap_pages_max_free_slots)
+	  And add the following filed.
+	    * freeable_pages (and heap_pages_freeable_pages)
+
+	* gc.c (heap_pages_free_unused_pages): unlink tomb heap pages
+	  because tomb heap should have only freeable pages.
+
+	* gc.c (heap_extend_pages): add parameters for future extension.
+
+Thu Mar 31 16:43:02 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: add GC parameters to configure the following values:
+	  * RUBY_GC_HEAP_FREE_SLOTS_MIN_RATIO:
+	    allocate additional pages when free slots is lower than
+	    the value (total_slots * (this ratio)).
+	  * RUBY_GC_HEAP_FREE_SLOTS_MAX_RATIO:
+	    allow to free pages when free slots is greater than
+	    the value (total_slots * (this ratio)).
+
+	  Before this change, these values are hard coded.
+
+	* gc.c (ruby_gc_params_t): ditto.
+
+	* gc.c (ruby_gc_set_params): ditto.
+
+Thu Mar 31 15:59:17 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_verify_heap_page): check the number of zombies.
+
+	* gc.c (gc_verify_heap_pages): check also tomb heap.
+
+Thu Mar 31 15:48:18 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c (gc_page_sweep): return free slots count.
+
+	* gc.c (gc_sweep_step): use returned free slots count.
+
+	* gc.c (gc_sweep_step): change variable name `next'
+	  to `next_sweep_page'.
+
+Thu Mar 31 11:33:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_core.c (d_lite_strftime, dt_lite_strftime): [DOC]
+	  fix indent not to be a big sole verbatim.
+
+Thu Mar 31 11:18:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_core.c (Init_date_core): [DOC] fix misplaced doc
+	  of DateTime.  [ruby-core:74729] [Bug #12233]
+
+Thu Mar 31 03:41:02 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Merge upstream 69f7e74dde.
+	  fix indent.
+
+Wed Mar 30 16:33:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* extension.rdoc, extension.ja.rdoc: [DOC] Fix some errors.
+	  Renamed files, wrong method names or argument types; the example
+	  GetDBM macro is now updated to the current version of the actual
+	  code.  patch by Marcus Stollsteimer in [ruby-core:74690].
+	  [Bug #12228]
+
+Wed Mar 30 09:46:01 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/open-uri.rb: Use `userinfo` for authenticated proxy.
+	  [fix GH-1148] Patch by @SokichiFujita
+	* test/open-uri/test_open-uri.rb: ditto.
+	  [fix GH-1309] Patch by @jdamick
+
+Wed Mar 30 01:56:06 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* ext/nkf/nkf-utf8/nkf.c: Merge upstream 4f3edf80a0.
+	  patched by Anton Sivakov [Bug #12201] [Bug #12202]
+
+Wed Mar 30 01:54:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/redmine-backporter.rb: add given revision to current changesets
+	  on associating the revision to the related ticket.
+
+Wed Mar 30 01:53:17 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: update revision.h before merge.
+
+Tue Mar 29 19:33:54 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* addr2line.c: define toupper for its use. fix r54391.
+
+Tue Mar 29 19:23:46 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* include/ruby/ruby.h (rb_isupper, rb_islower, rb_isalpha, rb_isdigit,
+	  rb_isalnum, rb_isxdigit, rb_isblank, rb_isspace, rb_isblank,
+	  rb_iscntrl, rb_isprint, rb_ispunct, rb_isgraph,
+	  rb_tolower, rb_toupper): use inline function to avoid function call.
+
+	* include/ruby/ruby.h (rb_isascii): use inline function to clarify
+	  the logic.
+
+Tue Mar 29 18:56:55 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/redmine-backporter.rb (backport): show merger.rb's path.
+
+	* tool/redmine-backporter.rb (show): show current issue again if no
+	  ticket number is given.
+
+	* tool/redmine-backporter.rb (rel): show error message if current
+	  bugs.ruby-lang.org doesn't support the API.
+
+Tue Mar 29 18:54:34 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* tool/merger.rb: support to backport header as backport identifier.
+	  Now you can specify by 'merge revision(s) 49254: [Backport #10738]'.
+
+Tue Mar 29 16:53:44 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Tweaked handling of 6
+	  special cases in CaseUnfold_11_Table.
+
+	* enc/unicode.c: Adjustments for above.
+
+	* test/ruby/enc/test_case_mapping.rb: Tests for the above: Some tests in
+	  test_titlecase activated; test_greek added. A test in test_cherokee fixed.
+
+Tue Mar 29 13:31:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Cleaned up some comments.
+
+Tue Mar 29 13:24:56 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Removing data for idempotent
+	  titlecasing.
+
+	* enc/unicode.c: Adjust code to data removal.
+
+Tue Mar 29 12:45:18 2016  Laurent Arnoud  <laurent@spkdev.net>
+
+	* lib/webrick/httpresponse.rb: Move error_body to method. It allow to
+	  override the body more easily. [fix GH-1307]
+	* test/webrick/test_httpresponse.rb: ditto.
+
+Tue Mar 29 06:40:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (rb_compile_err_append): rb_thread_t::base_block is no
+	  longer used.
+
+	* iseq.c (rb_iseq_compile_with_option): ditto, no protection is
+	  needed.
+
+Tue Mar 29 06:39:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (struct parser_params): move parse_in_eval flag from
+	  rb_thread_t.
+
+	* parse.y (rb_parser_set_context): set parsing context, not only
+	  mild error flag.
+
+	* iseq.c (rb_iseq_compile_with_option): the parser now refers no
+	  thread local states to be restored.
+
+	* vm_eval.c (eval_string_with_cref): ditto.
+
+Mon Mar 28 21:24:02 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* numeric.c (int_pos_p): fix typos.
+
+Mon Mar 28 14:54:49 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Refactoring in preparation for data reduction for
+	  titlecase.
+
+Mon Mar 28 14:36:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Minor refactoring for I WITH DOT ABOVE.
+
+Mon Mar 28 14:26:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Removed code now covered by data from table.
+
+Mon Mar 28 11:49:21 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Adding comments. [ci skip]
+
+Mon Mar 28 11:30:23 2016  Shinichi Maeshima  <netwillnet@gmail.com>
+
+	* lib/rubygems.rb: Fix `Gem.find_spec_for_exe` picks oldest gem.
+	  https://github.com/travis-ci/travis-ci/issues/5798
+	  https://github.com/rubygems/rubygems/pull/1566
+	* test/rubygems/test_gem.rb: ditto.
+
+Mon Mar 28 11:26:31 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.2.
+	  Please see entries of 2.6.2 on
+	  https://github.com/rubygems/rubygems/blob/master/History.txt
+
+Mon Mar 28 11:02:31 2016  Hiroshi Shirosaki  <h.shirosaki@gmail.com>
+
+	* lib/rubygems/test_case.rb: Fix test on Windows for inconsistent temp path.
+	  https://github.com/rubygems/rubygems/pull/1554
+	  [Bug #12193][ruby-core:74431]
+
+Mon Mar 28 08:19:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): refactor floating point format of
+	  Rational by using generic Integer functions.
+
+	* sprintf.c (rb_str_format): fix buffer overflow, length must be
+	  greater than precision.  reported by William Bowling <will AT
+	  wbowling.info>.
+
+Sun Mar 27 12:13:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): convert Rational to floating point
+	  format by using generic Integer functions, not by methods which
+	  can be overwritten.
+
+Sat Mar 26 10:55:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (rb_int2str): conversion function to String for
+	  generic Integer.
+
+	* numeric.c (rb_int_round): rounding function for generic
+	  Integers.
+
+	* numeric.c (rb_int_{uminus,plus,minus,mul,idiv,modulo}): basic
+	  arithmetic functions for generic Integers.
+
+	* numeric.c (FIXNUM_{POSITIVE,NEGATIVE,ZERO}_P): predict macros
+	  only for Fixnum.
+
+Sat Mar 26 06:34:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* localeinit.c (rb_locale_charmap_index): fix prototype.
+	  patched by Andreas Schwab [Bug #12218]
+
+Fri Mar 25 16:40:48 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_mapping.rb: Additional tests title case;
+	  some not yet activated.
+
+Fri Mar 25 13:38:11 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* ext/openssl/extconf.rb: check SSL_CTX_set_next_proto_select_cb
+	  function rather than OPENSSL_NPN_NEGOTIATED macro. it exists
+	  even if it is disabled by OpenSSL configuration.
+	  [ruby-core:74384] [Bug #12182]
+
+	* ext/openssl/ossl_ssl.c: update #ifdef(s) as above.
+
+	* test/openssl/test_ssl.rb: skip NPN tests if NPN is disabled.
+
+Fri Mar 25 11:08:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/uri/http.rb (URI::HTTP#initialize): [DOC] fix example,
+	  missing mandatory arguments.  [ruby-core:74540] [Bug #12215]
+
+Fri Mar 25 01:50:58 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* thread_pthread.c (reserve_stack): fix reserving position where
+	  the stack growing bottom to top. [Bug #12118]
+
+Fri Mar 25 01:10:42 2016  Sebastian Schuberth  <sschuberth@gmail.com>
+
+	* lib/mkmf.rb (find_executable0): On Windows, it is actually valid
+	  to surround individual PATH directory entries with double
+	  quotes. Remove these before joining the path as otherwise the
+	  literal quotes would become part of the path, resulting in the
+	  executable not to be found.  [Fix GH-1305]
+
+Thu Mar 24 22:38:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (FMT, FMTV): remove recursive-assignments to get rid
+	  of undefined behavior.  [ruby-core:74532] [Bug #12213]
+
+Thu Mar 24 17:44:02 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (FMT_PADDING): extract format for padding.
+
+	* strftime.c (FMT_PRECISION): extract precision formula.
+
+	* strftime.c (FMTV): append formatted string to expand the result.
+
+Thu Mar 24 14:20:21 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (STRFTIME): deal with case conversion flags for
+	  recursive formats.
+
+Thu Mar 24 12:43:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/date/date_core.c (dt_lite_iso8601): strftimev() always
+	  returns a String, so append them directly.
+
+	* ext/date/date_core.c (d_lite_jisx0301, iso8601_timediv),
+	  (dt_lite_jisx0301): format by the format string in local buffer
+	  to prevent intermediate strings from GC.
+
+	* ext/date/date_core.c (mk_inspect_raw, mk_inspect): inspect by
+	  "%+"PRIsVALUE, to prevent intermediate strings from GC.
+
+Thu Mar 24 11:43:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): remove unnecessary
+	  check, as `s` equals to `endp` when recursed STRFTIME resized
+	  the capacity same as the size.
+
+Wed Mar 23 21:48:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (ary_inject_op): put subtract operation out of if-clause.
+
+Wed Mar 23 21:38:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* enum.c (ary_inject_op): Use Kahan's compensated summation algorithm
+	  for summing up float values.
+
+Wed Mar 23 20:56:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* strftime.c (rb_strftime_with_timespec): append formatted results
+	  to the given string with expanding, and also deal with NUL chars.
+
+	* strftime.c (rb_strftime, rb_strftime_timespec): return formatted
+	  string, not the length put in the given buffer.
+
+	* time.c (rb_strftime_alloc): no longer needs to retry with
+	  reallocating buffers.
+
+	* time.c (time_strftime): no longer needs to split by NUL chars.
+
+Wed Mar 23 14:23:54 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/rdoc/ri/driver.rb (interactive): rescue NotFoundError raised in
+	  expand_name. (display_name rescues NotFoundError by itself,
+	  the original logic looks buggy...)
+
+Wed Mar 23 11:44:53 2016  cremno  <cremno@mail.ru>
+
+	* marshal.c (r_long): cast to `signed char`, which is used
+	  already, instead of SIGN_EXTEND_CHAR.
+
+	* parse.y: SIGN_EXTEND_CHAR is no longer used.  [Fix GH-1302]
+
+Wed Mar 23 11:38:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cygwin/GNUmakefile.in (MSYS2_ARG_CONV_EXCL_PARAM):
+	  * add missing parentheses and remove double quotes.
+	  * rename to get rid of recursive references.
+	  * as --excludes-dir option is for a path name, its argument
+	    should be converted.
+	  [ruby-dev:49526] [Bug #12199]
+
+Wed Mar 23 10:39:38 2016  Koichi ITO  <koic.ito@gmail.com>
+
+	* variable.c: Added documentation about order of `Module#constants`
+	  [ci skip][Bug #12121][ruby-dev:49505][fix GH-1301]
+
+Tue Mar 22 21:08:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* include/ruby/oniguruma.h: Additional flag for characters that are titlecase.
+
+	* enc/unicode/case-folding.rb, casefold.h: Using above flag in data.
+
+	* enc/unicode.c: Marking capitalized character as unmodified if it is
+	  already titlecase.
+
+	* test/ruby/enc/test_case_mapping.rb: Tests for above functionality.
+
+Tue Mar 22 14:18:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (lambda_body, parser_yylex): warn mismatched indentation
+	  of lambda block.
+
+Tue Mar 22 11:36:49 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (wmul): wrong condition.
+	  fixed many test failures on 32bit and LLP64 platforms.
+
+Tue Mar 22 10:31:34 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (wdiv, wmod): wdivmod0() assumes the 3rd and the 4th arguments
+	  are valid pointers.
+	  maybe checking them in wdivmod0() is better manner, but I guess that
+	  passing real dummy pointers may be faster than checking and branching
+	  in wdivmod0().
+	  this commit fixes SEGV on 32bit and LLP64 platforms.
+
+Tue Mar 22 10:24:04 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* time.c (divmodv): void function never returns any value.
+
+Tue Mar 22 10:11:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit.rb (Test::Unit::StatusLine#failed): print
+	  failed messages only if replacing mode, otherwise defer them
+	  until the end, to get rid of interleaving failures with progress
+	  messages.  refix r54195.
+
+Tue Mar 22 03:45:03 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (MUL_OVERFLOW_FIXWV_P): defined for FIXWV.
+
+	* time.c (wmul): use MUL_OVERFLOW_FIXWV_P and only switch.
+
+	* time.c (wmul): use mul which has Fixnum optimization.
+
+	* time.c (rb_time_magnify): If WIDEVALUE_IS_WIDER, wmul() has the same
+	  optimized logic, else mul() has also the similar logic for Fixnum.
+
+	* time.c (rb_time_unmagnify): almost ditto.
+
+Tue Mar 22 03:10:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (divmodv): add the case both arguments are Fixnum.
+
+	* time.c (wquo): use quo which has Fixnum optimization.
+
+	* time.c (wdivmod0): added for WIDEVALUE_IS_WIDER.
+
+	* time.c (wdivmod): use wdivmod0 and divmodv.
+	  divmodv has Fixnum optimization.
+
+	* time.c (wdiv): use wdivmod0 and div to avoid the use of divmodv which
+	  calls id_quo whose return value is array.
+
+	* time.c (wmod): use wdivmod0 and mod to avoid the use of divmodv which
+	  calls id_quo whose return value is array.
+
+Mon Mar 21 22:32:50 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* internal.h (rb_fix_divmod_fix): like r54213, use FIX2NUM only if
+	  x == FIXNUM_MIN && y == -1. This must be a rare case and it is
+	  expected compiler to handle well.
+
+Mon Mar 21 22:15:11 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (mod): Add Fixnum case.
+
+	* time.c (quo): c can be Fixnum except a == FIXNUM_MIN && b == -1.
+	  Such case can be optimized out because quo()'s argument is constant.
+
+Mon Mar 21 22:09:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (rb_fix_mul_fix): multiply converted values, not
+	  object VALUEs.
+
+Mon Mar 21 20:18:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk (TEST_EXCLUDES, EXCLUDE_TESTFRAMEWORK): use full spell
+	  long option.
+
+	* cygwin/GNUmakefile.in (MSYS2_ARG_CONV_EXCL): suppress path name
+	  conversions by msys2.  [ruby-dev:49525] [Bug #12199]
+
+Mon Mar 21 19:09:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (enc_succ_alnum_char): try to skip an invalid character
+	  gap between GREEK CAPITAL RHO and SIGMA.
+	  [ruby-core:74478] [Bug #12204]
+
+Mon Mar 21 18:55:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.c (rb_gc_mark_node): NODE_MATCH2 can have nd_args, u3,
+	  since r54100.
+
+Sun Mar 20 21:17:13 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* internal.h (rb_int128t2big): declare only when HAVE_INT128_T.
+	  fixed a compile error with VC++ introduced at r54203.
+
+Sun Mar 20 20:10:14 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* internal.h (DLONG): defined if long is 32bit (and LONG_LONG is 64bit;
+	  but LONG_LONG is always defined as 64bit), or there's int128_t.
+
+	* internal.h (DL2NUM): defined if DLONG is defined.
+
+	* internal.h (rb_fix_mul_fix): defined for `Fixnum * Fixnum`.
+
+	* insns.def (opt_mul): use rb_fix_mul_fix().
+
+	* numeric.c (fix_mul): ditto.
+
+	* time.c (mul): ditto.
+
+Sun Mar 20 18:53:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_gt, fix_ge, fix_lt, fix_le): optimize comparisons
+	  Fixnum against Bignum by rb_big_cmp in inversed order without
+	  new Bignum instance.
+
+Sun Mar 20 18:44:52 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (add): remove FIXABLE() which is in LONG2NUM().
+
+	* time.c (sub): ditto.
+
+	* time.c (mul): ditto.
+
+Sun Mar 20 04:46:02 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bignum.c (rb_big_cmp): reduce the code.
+
+	* bignum.c (rb_big_eq): If normalized bignum is still bignum,
+	  it must be larger than fixnum.
+
+Sat Mar 20 00:58:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* include/ruby/intern.h (rb_big_odd_p, rb_big_even_p): move to
+	  internal.h so that they are exported only for ruby itself.
+
+	* internal.h (rb_big_odd_p, rb_big_even_p): ditto.
+
+Sat Mar 19 21:56:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit.rb (Test::Unit::StatusLine#failed): defer
+	  failed messages until the end in verbose mode, to get rid of
+	  interleaving failures with progress messages.
+
+Sat Mar 19 21:53:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (fix_cmp): invert the result as the comparison is
+	  inverted.
+
+Sat Mar 19 18:32:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (int_to_f): raise NotImplementedError when a receiver
+	  class is unknown.
+
+	* test/-ext-/integer/test_my_integer.rb (test_my_integer_to_f): modify
+	  a test for the above change.
+
+Sat Mar 19 18:21:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (Bignum#<=>): remove it because they are unified with
+	  Integer#<=>.
+
+	* numeric.c (Integer#<=>, Fixnum#<=>): move <=> method from Fixnum to
+	  Integer.
+
+	* numeric.c (int_cmp): add this method for Integer#<=>.
+
+	* test/-ext-/integer/test_my_integer.rb (test_my_integer_cmp): add a
+	  test to examine Integer#<=> for unknown subclasses.
+
+Sat Mar 19 14:46:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_compile_with_option): make the parser in mild
+	  error.
+
+	* load.c (rb_load_internal0): ditto.
+
+	* parse.y (yycompile0): return the error message within the error
+	  to be raised.  [Feature #11951]
+
+	* parse.y (parser_compile_error): accumulate error messages in the
+	  error_buffer.
+
+Sat Mar 19 03:57:13 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* time.c (LOCALTIME): organize #ifdefs.
+
+	* time.c (GMTIME): define only ifndef HAVE_STRUCT_TM_TM_GMTOFF.
+
+Sat Mar 19 03:53:31 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (rb_cv_member_struct_tm_tm_gmtoff): For Linux (glibc)
+	  define _BSD_SOURCE for time.h to define struct tm.tm_gmtoff.
+
+	* time.c: define _BSD_SOURCE at the top.
+
+Sat Mar 19 03:00:50 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/-ext-/time/test_new.rb (test_timespec_new): change a gmtoff
+	  test to a better one that does not depend on whether the current
+	  time is in summer time or not.
+
+Fri Mar 19 00:00:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (rb_big_to_f, Bignum#to_f): removed them because they are
+	  unified with int_to_f and Integer#to_f.
+
+	* numeric.c (int_to_f): treat Bignum values directly.
+
+Fri Mar 18 23:41:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (int_to_f, fix_to_f): rename fix_to_f to int_to_f, and add
+	  treatment for subclasses which don't have definitions of to_f method.
+
+	* numeric.c (Integer#to_f, Fixnum#to_f): move to_f method from Fixnum
+	  to Integer.
+
+	* ext/-test-/integer/my_integer.c: define helper class for testing
+	  to_f method for a subclass of Integer.
+
+	* ext/-test-/integer/extconf.rb: ditto.
+
+	* ext/-test-/integer/init.c: ditto.
+
+	* test/-ext-/integer/test_my_integer.rb: examine to_f method for a
+	  subclass of Integer.
+
+Fri Mar 18 22:32:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* include/ruby/intern.h (rb_big_hash): Move to internal.h.
+
+	* internal.h: ditto.
+
+Fri Mar 18 22:10:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (Bignum#eql?): remove its definition because it is unified
+	  with Numeric#eql?.
+
+	* numeric.c (num_eql): treat Bignum values directly.
+
+Fri Mar 18 21:57:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (rb_big_to_s, Bignum#to_s): remove its definition because
+	  it is unified with Integer#to_s.
+
+	* numeric.c (int_to_s): treat Bignum values directly.
+
+Fri Mar 18 21:30:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (int_to_s): Move from fix_to_s.
+
+	* numeric.c (Integer#to_s): Move from Fixnum#to_s.
+
+Fri Mar 18 16:22:24 2016  Victor Nawothnig  <Victor.Nawothnig@gmail.com>
+
+	* parse.y (parse_numvar): NTH_REF must be less than a half of
+	  INT_MAX, as it is left-shifted to be ORed with back-ref flag.
+	  [ruby-core:74444] [Bug#12192] [Fix GH-1296]
+
+Fri Mar 18 12:25:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* gc.c (tick): fix missing close parenthesis.  [Fix GH-1291]
+
+Fri Mar 18 10:24:12 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* ext/date/date_core.c (datetime_to_time): preserve timezone info
+	  [Bug #12189] [Fix GH-1295]
+
+Fri Mar 18 10:17:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (rb_big_hash): make it public function to be available in
+	  other source files, and remove documentation comment for Bignum#hash.
+
+	* bignum.c (Bignum#hash): remove its definition because it is unified
+	  with Object#hash.
+
+	* include/ruby/intern.h (rb_big_hash): add a prototype declaration.
+
+	* hash.c (any_hash): treat Bignum values directly.
+
+Fri Mar 18 02:35:12 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/time.rb (parse, strptime): Fix Time.parse/strptime does not
+	  have compatibility with DateTime.parse/strptime in terms of parsing
+	  timezone [Bug #12190] [Fix GH-1297]
+
+Fri Mar 18 02:17:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (fix_zero_p, fix_even_p, fix_odd_p): remove needless
+	  functions.
+
+Fri Mar 18 02:15:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* numeric.c (int_even_p): treat Fixnum and Bignum values directly.
+
+Fri Mar 18 02:07:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (Bignum#even?, Bignum#odd?): remove definitions
+	  because they are unified with Integer#even? and Integer#odd?.
+
+	* numeric.c (Fixnum#zero?, Fixnum#even?, Fixnum#odd?): remove
+	  definitions because they are unified with Numeric#zero?,
+	  Integer#even?, and Integer#odd?.
+
+	* numeric.c (num_zero_p, int_odd_p): treat Fixnum and
+	  Bignum values directly.
+
+	* test/ruby/test_integer.rb (test_odd_p_even_p): remove meaningless
+	  test case.
+
+Fri Mar 18 01:51:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* bignum.c (rb_big_even_p, rb_big_odd_p): make them public functions
+	  to be available in other source files.
+
+	* include/ruby/intern.h (rb_big_even_p, rb_big_odd_p): add prototype
+	  declarations.
+
+Fri Mar 18 00:25:56 2016  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (ary_inject_op): Implement the specialized code for sum of
+	  float numbers.
+
+Fri Mar 18 00:15:05 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* numeric.c (num_step): use rb_equal for zero check. rb_num_coerce_cmp
+	  created an object which caused extra overhead.
+
+Thu Mar 17 22:21:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/ruby.h (RB_GC_GUARD_PTR): remove intermediate
+	  macro, and expand for each RB_GC_GUARD.  [Fix GH-1293]
+
+Thu Mar 17 22:08:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (iseq_specialized_instruction): move specialization
+	  for opt_newarray_max/min from translation phase.
+
+Thu Mar 17 21:52:09 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* array.c, enum.c: make rdoc format consistent.
+
+Thu Mar 17 21:47:57 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* NEWS: add Array#max, #min, and the optimization.  [Feature #12172]
+
+Thu Mar 17 21:45:02 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* compile.c (NODE_CALL): add optimization shortcut for Array#max/min.
+	  Now `[x, y].max` is optimized so that a temporal array object is not
+	  created in some condition.  [Feature #12172]
+
+	* insns.def (opt_newarray_max, opt_newarray_min): added.
+
+Thu Mar 17 21:35:52 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* array.c (rb_ary_max, rb_ary_min): implement Array#max and min with
+	  arguments.  replace super call with rb_nmin_run.
+
+	* enum.c (nmin_run): exported (as rb_nmin_run).
+
+	* internal.h: added a prototype for rb_nmin_run.
+
+Thu Mar 17 21:24:52 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* array.c (rb_ary_max, rb_ary_min): implement a block by itself instead
+	  of delegating Enumerable#max/min.
+
+Thu Mar 17 21:09:34 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* array.c (rb_ary_max, rb_ary_min): Array#max and Array#min added.
+	  [Feature #12172]
+
+	* internal.h (OPTIMIZED_CMP): moved from enum.c so that array.c can
+	  use it.
+
+	* test/ruby/test_array.rb (test_max, test_min): tests for Array#max
+	  and Array#min.
+
+	* test/ruby/test_enum.rb (test_max, test_min): revised a bit to test
+	  Enumerable#max and #min explicitly.
+
+Thu Mar 17 21:02:42 2016  Yusuke Endoh  <mame@ruby-lang.org>
+
+	* internal.c: struct cmp_opt_data added for refactoring out a data
+	  structure for CMP_OPTIMIZABLE
+
+	* array.c (struct ary_sort_data): use struct cmp_opt_data.
+
+	* enum.c (struct min_t, max_t, min_max_t): use struct cmp_opt_data.
+
+Thu Mar 17 20:55:21 2016  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (ary_inject_op): Extracted from enum_inject.
+
+Thu Mar 17 18:39:04 2016  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_inject): Implement the specialized code for sum of
+	  integers including Bignums.
+
+	* internal.h (rb_fix_plus): Declared to be usable from enum_inject.
+
+	* numeric.c (rb_fix_plus): Defined.
+
+Thu Mar 17 17:20:28 2016  Anton Davydov  <antondavydov.o@gmail.com>
+
+	* thread_sync.c: [DOC] Update documentation for Queue class
+	  description.  [Fix GH-1292]
+
+Thu Mar 17 17:14:51 2016  Dinar Valeev  <dvaleev@suse.com>
+
+	* gc.c (tick): Use __builtin_ppc_get_timebase for POWER arch.
+	  [Fix GH-1291]
+
+Thu Mar 17 11:51:48 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/securerandom.rb (gen_random): to avoid blocking on Windows.
+	  On Windows OpenSSL RAND_bytes (underlying implementation is
+	  RAND_poll in crypto/rand/rand_win.c) may be blocked at
+	  NetStatisticsGet.
+	  https://wiki.openssl.org/index.php/Random_Numbers#Windows_Issues
+	  Instead of this, use Random.raw_seed directly (whose implementation
+	  CryptGenRandom is one of the source of
+	  entropy of RAND_poll on Windows).
+	  https://wiki.openssl.org/index.php/Random_Numbers
+	  Note: CryptGenRandom function is PRNG and doesn't check its entropy,
+	  so it won't block. [Bug #12139]
+	  https://msdn.microsoft.com/ja-jp/library/windows/desktop/aa379942.aspx
+	  https://tools.ietf.org/html/rfc4086#section-7.1.3
+	  https://eprint.iacr.org/2007/419.pdf
+	  http://www.cs.huji.ac.il/~dolev/pubs/thesis/msc-thesis-leo.pdf
+
+Thu Mar 17 12:09:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Fixed two macro definitions.
+	* test/ruby/enc/test_case_mapping.rb: Test cases that detected
+	  the above bugs.
+
+Thu Mar 17 11:36:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/option.c (inspect_tcpi_msec): more accurate condition
+	  for TCPI msec member inspection function.
+	  [ruby-core:74388] [Bug #12185]
+
+Thu Mar 17 08:13:43 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/-ext-/time/test_new.rb (test_timespec_new): Time#gmtoff values
+	  are the same only when both or neither of the Time objects are in
+	  summer time (daylight-saving time).
+
+Thu Mar 17 07:17:36 2016  Eric Hodel  <drbrain@segment7.net>
+
+	* marshal.c (r_object0):  raise ArgumentError when linking to undefined
+	  object.
+
+Thu Mar 17 00:45:00 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* test/ruby/test_bignum.rb: Make sure to use Bignum values in the tests.
+
+Wed Mar 16 23:10:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/keywords (alias, undef): symbol literals are allowed.
+
+	* parse.y (parse_percent): should parse symbol literals for alias
+	  and undef.  [ruby-dev:47681] [Bug #8851]
+
+Wed Mar 16 21:39:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* enc/trans/JIS: update Unicode's notice. [Bug #11844]
+
+Wed Mar 16 20:03:35 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* proc.c (proc_binding): proc from symbol can not make a binding.
+	  [ruby-core:74100] [Bug #12137]
+
+Wed Mar 16 18:42:45 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_mapping.rb: Fixed and activated a test for Cherokee.
+
+Wed Mar 16 17:58:56 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_mapping.rb: Fixed a logical error.
+
+Wed Mar 16 17:57:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_mapping.rb: Adding tests for Cherokee.
+	  One test not yet working.
+	  (with Kimihito Matsui)
+
+Wed Mar 16 15:44:05 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_mapping.rb: Adding tests for actual Unicode
+	  case mapping. Fixing some aliasing issues.
+	  (with Kimihito Matsui)
+
+Tue Mar 15 21:38:28 2016  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_inject): Consider redefinition of Fixnum#+.
+	  [ruby-dev:49510] [Bug#12178] Reported by usa.
+
+Tue Mar 15 20:32:57 2016  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_inject): Implement the specialized code for :+ operator
+	  for Fixnums.
+
+Tue Mar 15 20:21:01 2016  Tanaka Akira  <akr@fsij.org>
+
+	* enum.c (enum_inject): Implement the specialized code for self is an
+	  array and a symbol operator is given.
+
+Tue Mar 15 16:29:51 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Eliminating common code.
+	  (with Kimihito Matsui)
+
+Tue Mar 15 16:17:09 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Expansion of some code repetition in preparation for
+	  elimination of common code pieces.
+	  (with Kimihito Matsui)
+
+Tue Mar 15 13:49:23 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Additional macros and code to use mapping data in
+	  CaseMappingSpecials array.
+	  (with Kimihito Matsui)
+
+Tue Mar 15 13:41:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h (rb_gc_mark_global_tbl): should be private,
+	  but was accidentally exported.
+
+Tue Mar 15 12:51:06 2016  Marcus Stollsteimer  <sto.mar@web.de>
+
+	* doc/extension.ja.rdoc: Fix RDoc markup in doc/extension*.rdoc.
+	  [ci skip][Bug #12143][ruby-core:74143]
+	* doc/extension.rdoc: ditto.
+
+Tue Mar 15 09:56:41 2016  Ian Lunderskov  <ian.lunderskov@gmail.com>
+
+	* time.c: Minor typo in Time#dst? documentation.
+	  [ci skip][fix GH-1290]
+
+Tue Mar 15 04:36:41 2016  Charles Oliver Nutter  <headius@headius.com>
+
+	* test/ruby/test_rubyoptions.rb (test_disable): add tests for
+	  --disable-gems and --disable-did_you_mean.
+
+Tue Mar 15 03:35:04 2016  Eric Hodel  <drbrain@segment7.net>
+
+	* marshal.c (r_object0): Fix Marshal crash for corrupt extended object.
+
+Tue Mar 15 01:22:27 2016  Charles Oliver Nutter  <headius@headius.com>
+
+	* test/ruby/test_rubyoptions.rb: make version matching support
+	  JRuby's version output.
+
+Mon Mar 14 19:05:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* bignum.c (big2str_2bdigits): reduce div instruction.
+
+Mon Mar 14 18:39:53 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* include/ruby/oniguruma.h, enc/unicode.c: Adjusting flag assignments
+	  and macros to work with unified CaseMappingSpecials array.
+	  (with Kimihito Matsui)
+
+Mon Mar 14 16:53:37 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_named_capture_assign): optimize named capture
+	  assignments, by replacing repeating global variable accesses
+	  with `dup`, and by returning the matched result instead of
+	  re-getting it from the MatchData.
+
+	* parse.y (reg_named_capture_assign_gen): build just assignment
+	  nodes for the optimization.
+
+Mon Mar 14 16:02:59 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* file.c (ruby_is_fd_loadable): now return -1 if loadable but
+	  may block.
+
+	* ruby.c (open_load_file): wait to read by the result of
+	  ruby_is_fd_loadable, without fstat.
+
+Mon Mar 14 13:38:38 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* numeric.c (fix2str): improve r54092 like rb_int2big().
+
+Mon Mar 14 10:02:23 2016  Eric Wong  <e@80x24.org>
+
+	* ext/openssl/ossl_ssl.c (ossl_sslctx_setup): document as MT-unsafe
+	  [ruby-core:73803] [Bug #12069]
+
+Sun Mar 13 09:43:23 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/win32.h (O_SHARE_DELETE): change to fit Fixnum
+	  limit.  [ruby-core:74285] [Bug #12171]
+
+Sun Mar 13 09:15:45 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (rb_fix2str): fix edge case, accidentally generated
+	  wrong Fixnum from LONG_MIN.
+
+Sat Mar 12 09:50:27 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_f_catch): [DOC] fix malformed RDoc syntax, "+...+"
+	  cannot enclose non-identifier characters.
+	  a patch by Sebastian S in [ruby-core:74278].  [Bug#12170]
+
+Sat Mar 12 02:44:48 2016  Tanaka Akira  <akr@fsij.org>
+
+	* test/lib/test/unit.rb: describe !/REGEXP/ in the help message.
+
+Fri Mar 11 17:03:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/test/unit.rb (Options#non_options): make regexp name
+	  options prefixed with "!" negative filters.
+
+	* common.mk (TEST_EXCLUDES): use negative filter to exclude memory
+	  leak tests.  -x option excludes test files, not test methods.
+
+Fri Mar 11 16:11:27 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Streamlining approach to
+	  case mapping data not available from case folding by unifying all
+	  three cases (special title, special upper, special lower).
+	* enc/unicode.c: Adjust macro names for above (macros are currently inactive).
+	  (with Kimihito Matsui)
+
+Thu Mar 10 17:34:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (prepare_iseq_build): enable coverage by coverage_enabled
+	  option, not by parse_in_eval flag in the thread context.
+
+	* iseq.h (rb_compile_option_struct): add coverage_enabled flag.
+
+	* parse.y (yycompile0): set coverage_enabled flag if coverage
+	  array is made.
+
+Thu Mar 10 15:19:54 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* node.c (dump_option): nd_compile_option is a hidden hash object,
+	  cannot call inspect on it.
+
+Thu Mar 10 09:49:54 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/socket/test_socket.rb (test_udp_recvmsg_truncation):
+	  AIX does not set the MSG_TRUNC flag for a message partially read
+	  by recvmsg(2) with the MSG_PEEK flag set.
+
+Wed Mar  9 16:48:45 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: fix my last commit (syntax error).
+
+Wed Mar  9 16:41:44 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: fix output messages.
+
+	* benchmark/memory_wrapper.rb: use respond_to? because
+	  member? does not work well.
+
+Wed Mar  9 16:20:25 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: support memory usage benchmark.
+	  use `--measure-target=[target]'.
+	  Now, we can use the following targets:
+	    * real (default): real time which returns process time in sec.
+	    * peak: peak memory usage (physical memory) in bytes.
+	    * size: last memory usage (physical memory) in bytes.
+
+	* benchmark/memory_wrapper.rb: ditto.
+
+Wed Mar  9 15:04:22 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm3_gc_old_full.rb: add GC.start benchmark.
+
+	* benchmark/bm_vm3_gc_old_immediate.rb: ditto.
+
+	* benchmark/bm_vm3_gc_old_lazy.rb: ditto.
+
+Wed Mar  9 14:59:11 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/driver.rb: exit benchmarking if a benchmark process
+	  receives signals.
+
+Wed Mar  9 13:22:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* test/lib/memory_status.rb: make Memory::Status independent of
+	  MiniTest::Skip.
+
+	* test/lib/test/unit/assertions.rb (assert_no_memory_leak): skip
+	  if Memory::Status is not available.
+
+Wed Mar  9 09:19:55 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/io/wait/test_io_wait.rb (test_wait_readwrite_timeout):
+	  select(2) in AIX returns "readable" for the write-side fd
+	  of a pipe, so it is not possible to use a pipe to test
+	  the read-write timeout of IO#wait on AIX.
+
+Wed Mar  9 03:35:22 2016  Charles Oliver Nutter  <headius@headius.com>
+
+	* test/ruby/test_require.rb (test_require_with_loaded_features_pop):
+	  Only remove PATH so threads don't accidentally double-pop.
+
+Wed Mar  9 00:29:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_method.c (rb_alias): the original name should be properly
+	  available method_added method, set the name before calling the
+	  hook.
+
+Wed Mar  9 00:07:03 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/logger.rb (Logger::LogDevice#initialize): define using
+	  keyword arguments.
+
+Tue Mar  8 23:37:07 2016  Charles Oliver Nutter  <headius@headius.com>
+
+	* test/ruby/test_array.rb: split out the test for no stack error
+	  on large input for test_permutation, test_repeated_permutation,
+	  and test_repeated_combination, and make them all timeout:30.
+
+Tue Mar  8 17:20:21 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* intern.h (rb_divmod): assume compilers `/` and `%` comply C99
+	  and reduce branching. If a compiler doesn't comply, add #ifdefs.
+
+	* intern.h (rb_div): added for Ruby's behavior.
+
+	* intern.h (rb_mod): added for Ruby's behavior.
+
+	* insns.def (opt_div): use rb_div.
+
+	* insns.def (opt_mod): use rb_mod.
+
+	* numeric.c (fixdivmod): removed.
+
+	* numeric.c (fix_divide): use rb_div.
+
+	* numeric.c (fix_mod): use rb_mod.
+
+	* numeric.c (fix_divmod): use rb_divmod.
+
+Tue Mar  8 17:53:09 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (opt_mod): show its method name on ZeroDivisionError.
+	  [Bug #12158]
+
+Tue Mar  8 17:33:38 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write_console): now no need to check
+	  ERROR_CALL_NOT_IMPLEMENTED because it is for old Win9X.
+
+Tue Mar  8 16:54:29 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (rb_w32_write_console): stop the VT100 emulation if the
+	  console supports it natively.
+
+Tue Mar  8 08:13:01 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/net/imap/test_imap.rb (test_idle_timeout): Because of the
+	  timeout specified in "imap.idle(0.2)", there is no guarantee that
+	  the server thread has done all the work before the client thread
+	  performs the assertions. It depends on the thread scheduling.
+	  Add checks to avoid false positives (on AIX, particularly).
+
+Tue Mar  8 00:42:22 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* ruby.c (warn_cr_in_shebang): meaningless check on DOSISH platforms.
+	  fixed a test failure introduced at r53998.
+
+Tue Mar  8 00:27:53 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/tk/lib/tkextlib/tcllib/tablelist_tile.rb: fix method name typo.
+	  [ruby-core:72513] [Bug #11893] The patch provided by Akira Matsuda.
+
+
+Tue Mar  8 00:25:08 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/tk/lib/tkextlib/tcllib/toolbar.rb: fix method name typo.
+	  [ruby-core:72511] [Bug #11891] The patch provided by Akira Matsuda.
+
+
+Tue Mar  8 00:21:58 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/tk/lib/tkextlib/blt/tree.rb: fix method name typo.
+	  [ruby-core:72510] [Bug #11890] The patch provided by Akira Matsuda.
+
+
+Tue Mar  8 00:11:47 2016  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* ext/tk/lib/tk/menubar.rb: fix a typo in font name. [ruby-core:72505]
+	  [Bug #11886] The patch provided by Akira Matsuda.
+
+	* ext/tk/sample/*.rb: ditto.
+
+Mon Mar  7 13:32:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* class.c (rb_define_class, rb_define_class_id_under): raise
+	  ArgumentError if super is 0, deprecated behavior which has been
+	  warned long time.
+
+Mon Mar  7 13:28:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* internal.h: move function declarations for class internals from
+	  include/ruby/intern.h.
+
+Mon Mar  7 10:58:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/win32ole/win32ole_event.c (rescue_callback): use
+	  rb_write_error_str instead of rb_write_error, to respect
+	  the encoding and prevent the message from GC.
+
+	* internal.h (rb_write_error_str): export.
+
+Mon Mar  7 01:38:41 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/ruby/test_process.rb (test_execopts_gid): Skip a test
+	  that is known to fail on AIX. AIX allows setgid to
+	  a supplementary group, but Ruby does not allow the "-e"
+	  option when setgid'ed, so the test does not work as intended.
+
+Sun Mar  6 22:43:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* io.c (rb_obj_display): [DOC] fix output of Array, as Array#to_s
+	  is same as Array#inspect since 1.9.
+
+Sat Mar  5 09:50:58 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/socket/test_addrinfo.rb (test_ipv6_address_predicates):
+	  IN6_IS_ADDR_V4COMPAT and IN6_IS_ADDR_V4MAPPED are broken
+	  on AIX, so skip related tests.
+
+Sat Mar  5 09:17:54 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/rinda/test_rinda.rb (test_make_socket_ipv4_multicast):
+	  The fifth argument to getsockopt(2) should be modified to
+	  indicate the actual size of the value on return,
+	  but not in AIX. This is a know bug. Skip related tests.
+	* test/rinda/test_rinda.rb (test_ring_server_ipv4_multicast):
+	  ditto.
+	* test/rinda/test_rinda.rb (test_make_socket_unicast): ditto.
+	* test/socket/test_basicsocket.rb (test_getsockopt): ditto.
+	* test/socket/test_sockopt.rb (test_bool): ditto.
+
+Sat Mar  5 07:36:27 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/-ext-/float/test_nextafter.rb: In AIX,
+	  nextafter(+0.0,-0.0)=+0.0, and nextafter(-0.0,+0.0)=-0.0,
+	  but they should return -0.0 and +0.0, respectively. This is
+	  a known bug in nextafter(3) on AIX, so skip related tests.
+
+Sat Mar  5 07:14:10 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* test/zlib/test_zlib.rb (test_adler32_combine, test_crc32_combine):
+	  Skip two tests on AIX because zconf.h in zlib does not correctly
+	  recognize _LARGE_FILES in AIX. The problem was already reported
+	  to zlib, and skip these tests until it is fixed.
+
+Sat Mar  5 03:07:40 2016  Rei Odaira  <Rei.Odaira@gmail.com>
+
+	* thread_pthread.c (getstack): __pi_stacksize returned by
+	  pthread_getthrds_np() is wrong on AIX. Use
+	  __pi_stackend - __pi_stackaddr instead.
+
+Fri Mar  4 19:19:42 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: use 2 bits with unsigned int for rb_objspace::flags::mode
+	  because it always returns 0 to 2 (non-negative value).
+
+Fri Mar  4 18:42:08 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename "enum gc_stat" to "enum gc_mode"
+	  because there is a same name (no related) function gc_stat().
+
+	  Also gc_stat_* are renamed to gc_mode_*,
+	  gc_stat_transition() to gc_mode_transition(),
+	  rb_objspace::flags::stat is renamed to rb_objspace::flags::mode.
+
+	  Change rb_objspace::flags::mode from 2 bits to 3 bits because VC++
+	  returns negative enum value with 2 bits.
+
+	* gc.c (gc_mode): add a macro to access rb_objspace::flags::mode
+	  with verification code (verification is enabled only on
+	  RGENGC_CHECK_MODE > 0).
+
+	* gc.c (gc_mode_set): same macro for setter.
+
+Fri Mar  4 09:28:18 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.6.1.
+	  Please see entries of 2.6.0 and 2.6.1 on
+	  https://github.com/rubygems/rubygems/blob/master/History.txt
+	  [fix GH-1270] Patch by @segiddins
+
+Thu Mar  3 14:09:00 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (modifiable?, new_ostruct_member!, table!):
+	  rename methods for internal use with suffixes and make private,
+	  [ruby-core:71069] [Bug #11587]
+
+Wed Mar  2 16:28:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (method_missing): call by found method entry and get
+	  rid of searching the same method entry twice.
+
+	* vm_eval.c (vm_call0_body): calling method_missing method is
+	  method_missing().
+
+Wed Mar  2 15:13:33 2016  herwinw  <herwin@quarantainenet.nl>
+
+	* lib/xmlrpc.rb: Removed broken parser named XMLScanStreamParser.
+	  It's not works with current Ruby version.
+	  [fix GH-1271][ruby-core:59588][Bug #9369]
+	* lib/xmlrpc/config.rb: ditto.
+	* lib/xmlrpc/parser.rb: ditto.
+
+Wed Mar  2 15:08:33 2016  herwinw  <herwin@quarantainenet.nl>
+
+	* lib/xmlrpc.rb: Removed broken parser named XMLTreeParser.
+	  Required gem of its parser didn't compile on newer Ruby versions.
+	  [fix GH-1271][ruby-core:59590][Bug #9370]
+	* lib/xmlrpc/config.rb: ditto.
+	* lib/xmlrpc/parser.rb: ditto.
+
+Tue Mar  1 11:25:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/fileutils.rb: use keyword arguments instead of option
+	  hashes.
+
+Mon Feb 29 16:50:20 2016  hanachin  <hanachin@gmail.com>
+
+	* array.c (rb_ary_push_m): [DOC] Remove trailing comma from
+	  Array#push example, as other Array examples doesn't put trailing
+	  comma.  [Fix GH-1279]
+
+Mon Feb 29 16:31:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* common.mk, tool/mkconfig.rb: set cross_compiling option from
+	  Makefile, but not from rbconfig.rb, which is just going to be
+	  created by this command.
+
+Sun Feb 28 23:13:49 2016  C.J. Collier  <cjcollier@linuxfoundation.org>
+
+	* configure.in: Add summary to end of configure output.
+	  [Fix GH-1275]
+
+Sun Feb 28 20:23:36 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/drb/drb.rb (error_print): Add verbose failure messages and
+	  avoid infamous DRb::DRbConnError. [Feature #12101]
+
+Sun Feb 28 13:40:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (nometh_err_initialize): add private_call? parameter.
+
+	* error.c (nometh_err_private_call_p): add private_call? method,
+	  to tell if the exception raised in private form FCALL or VCALL.
+	  [Feature #12043]
+
+	* vm_eval.c (make_no_method_exception): append private_call?
+	  argument.
+
+	* vm_insnhelper.c (ci_missing_reason): copy FCALL flag.
+
+Sun Feb 28 10:19:47 2016  Ryan T. Hosford  <tad.hosford@gmail.com>
+
+	* array.c (rb_ary_and): clarify that set intersection returns the
+	  unique elements common to both arrays.
+
+	* array.c (rb_ary_or): clarify that union preserves the order from
+	  the given arrays.
+
+Sat Feb 27 17:05:29 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Reducing size of TitleCase
+	  table by eliminating duplicates.
+	  (with Kimihito Matsui)
+
+Fri Feb 26 14:40:48 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (num_step_scan_args): comparison String with Numeric
+	  should raise TypeError. it is an invalid type, but not a
+	  mismatch the number of arguments.  [ruby-core:62430] [Bug #9810]
+
+Fri Feb 26 14:39:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* doc/extension.rdoc, doc/extension.ja.rdoc: add editor local
+	  variables, with commenting out by :enddoc: directives which are
+	  just ignored unless code object mode.  [Bug #12111]
+
+Fri Feb 26 12:25:56 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/extension.ja.rdoc: removed rendering error caused by editor specific
+	  configuration on http://docs.ruby-lang.org/en/trunk/extension_rdoc.html .
+	  [Bug #12111][ruby-core:73990]
+
+Fri Feb 26 11:21:41 2016  herwinw  <herwin@quarantainenet.nl>
+
+	* lib/xmlrpc.rb: Removed references to NQXML. It's obsoleted parser.
+	  [fix GH-1245][ruby-core:59593][Feature #9371]
+	* lib/xmlrpc/config.rb: ditto.
+	* lib/xmlrpc/parser.rb: ditto.
+
+Fri Feb 26 11:10:19 2016  Rick Salevsky  <rsalevsky@suse.com>
+
+	* lib/tmpdir.rb: Unify to coding-style for method definition.
+	  [fix GH-1252]
+
+Fri Feb 26 11:02:04 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* README.md: update markdown syntax for anchor tag.
+	  [fix GH-1265] Patch by @lukBarros
+
+Fri Feb 26 10:52:29 2016  Alex Boyd  <alex@opengroove.org>
+
+	* lib/irb.rb: avoid to needless truncation when using back_trace_limit option.
+	  [fix GH-1205][ruby-core:72773][Bug #11969]
+
+Fri Feb 26 08:11:58 2016  Aaron Patterson <tenderlove@ruby-lang.org>
+
+	* gem_prelude.rb: Reduce system calls by activating the `did_you_mean`
+	  gem before requiring the gem.  Activating the gem puts the gem on
+	  the load path, where simply requiring the file will search every gem
+	  that's installed until it can find a gem that contains the
+	  `did_you_mean` file.
+
+Thu Feb 25 19:04:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb: Adding possibility for debugging output
+	  for TitleCase table in casefold.h.
+	  (with Kimihito Matsui)
+
+Wed Feb 24 22:31:13 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* include/ruby/oniguruma.h: Rearranging flag assignments and making
+	  space for titlecase indices; adding additional macros to add or
+	  extract titlecase index; adding comments for better documentation.
+	* enc/unicode.c: Moving some macros to include/ruby/oniguruma.h;
+	  activating use of titlecase indices.
+	  (with Kimihito Matsui)
+
+Wed Feb 24 21:03:04 2016  Tanaka Akira  <akr@fsij.org>
+
+	* random.c (limited_rand): Add a specialized path for when the limit fits
+	  in 32 bit.
+
+Tue Feb 23 21:52:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Outputting actual titlecase
+	  data (new table, with indices from other tables).
+	* enc/unicode.c: Ignoring titlecase data indices for the moment.
+	  (with Kimihito Matsui)
+
+Tue Feb 23 15:21:14 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Reading casing data from
+	  SpecialCasing.txt.
+	  (with Kimihito Matsui)
+
+Mon Feb 22 18:33:55 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Adding flag for title-case,
+	  not yet operational.
+	  (with Kimihito Matsui)
+
+Mon Feb 22 18:17:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Fixed bug that avoided inclusion
+	  of compatibility characters in upper-/lower-case mappings.
+	  (with Kimihito Matsui)
+
+Sun Feb 21 13:57:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cgi/escape/escape.c: Optimize CGI.unescape performance by C ext
+	  for ASCII-compatible encodings.  [Fix GH-1250]
+
+Sun Feb 21 13:56:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* cgi/escape/escape.c: Optimize CGI.unescapeHTML performance by C
+	  ext for ASCII-compatible encodings.  [Fix GH-1242]
+
+Sat Feb 20 15:38:16 2016  Eric Wong  <e@80x24.org>
+
+	* doc/extension.rdoc: update paths for defs/ directory
+
+Sat Feb 20 14:44:15 2016  Lucas Buchala  <lucasbuchala@gmail.com>
+
+	* vm_eval.c (rb_mod_module_eval): [DOC] Fix documentation
+	  signature for Module#module_eval.  [Fix GH-1258]
+
+Sat Feb 20 14:40:44 2016  Adam O'Connor  <northband@gmail.com>
+
+	* README.md: a few grammatical changes to the main Ruby README.md.
+	  [Fix GH-1259]
+
+Sat Feb 20 13:04:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (push_pattern, push_glob): deal with read paths as UTF-8
+	  to stat later, on Windows as well as OS X.
+	  [ruby-core:73868] [Bug #12081]
+
+Sat Feb 20 01:53:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_mod_const_get): make error message at uninterned
+	  string consistent with symbols.  [ruby-dev:49498] [Bug #12089]
+
+Fri Feb 19 23:37:52 2016  Masahiro Tomita  <tommy@tmtm.org>
+
+	* lib/find.rb (Find#find): raise with the given path name if it
+	  does not exist.  [ruby-dev:49497] [Bug #12087]
+
+Fri Feb 19 12:44:57 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Activated use of case mapping data in CaseUnfold_11 array.
+	  (with Kimihito Matsui)
+
+Fri Feb 19 11:08:32 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/extmk.rb: add cygwin case, nothing excluded.
+	  [ruby-core:73806] [Bug#12071]
+
+Thu Feb 18 21:32:15 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* man/irb.1: fix output in EXAMPLES.
+
+Thu Feb 18 21:05:47 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (sym_match_m): delegate to String#match but not
+	  String#=~.  [ruby-core:72864] [Bug #11991]
+
+Thu Feb 18 14:15:38 2016  Shota Fukumori  <her@sorah.jp>
+
+	* re.c: Add MatchData#named_captures
+	  [Feature #11999] [ruby-core:72897]
+
+	* test/ruby/test_regexp.rb(test_match_data_named_captures): Test for above.
+
+	* NEWS: News about MatchData#named_captures.
+
+Wed Feb 17 21:41:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* defs/id.def (predefined): add idLASTLINE and idBACKREF for $_
+	  and $~ respectively.
+
+	* parse.y: use idLASTLINE and idBACKREF instead of rb_intern.
+
+Wed Feb 17 20:23:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_str_init): fix segfault and memory leak, consider
+	  wide char encoding terminator.
+
+Wed Feb 17 12:14:59 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (rb_str_init): introduce String.new(capacity: size)
+	  [Feature #12024]
+
+Tue Feb 16 19:10:08 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, casefold.h: Used only first element
+	  (rather than all) of target in CaseUnfold_11 array.
+	  (with Kimihito Matsui)
+
+Tue Feb 16 18:24:38 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* numeric.c (compare_with_zero): fix variable name, rb_cmperr
+	  requires VALUEs but not an ID.
+
+Tue Feb 16 17:34:18 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* dir.c (rb_dir_s_empty_p): add Dir.empty? method, which tells the
+	  argument is the name of an empty directory.  [Feature #10121]
+
+Tue Feb 16 09:51:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (without_destdir): just strip a drive letter
+	  which is prepended by with_destdir.
+	  pointed out by @DavidEGrayson.
+	  https://github.com/ruby/ruby/commit/0e5f9ae#commitcomment-16101763
+
+Tue Feb 16 04:42:13 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (opt_plus): simply use LONG2NUM() instead of wrongly
+	  complex overflow case.
+
+	* insns.def (opt_sub): ditto.
+
+Tue Feb 16 02:49:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/rbinstall.rb (without_destdir): compare with the destdir
+	  after stripping a drive letter, on dosish platforms.
+	  pointed out by @DavidEGrayson.
+	  https://github.com/ruby/ruby/commit/d0cf23b#commitcomment-16100407
+
+Mon Feb 15 15:44:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parse_ident): allow keyword arguments just after a
+	  method where the same name local variable is defined.
+	  [ruby-core:73816] [Bug#12073]
+
+Mon Feb 15 14:43:28 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb: Added debugging option
+	  (with Kimihito Matsui)
+
+Sun Feb 14 17:31:50 2016  Lars Kanis  <lars@greiz-reinsdorf.de>
+
+	* lib/mkmf.rb (with_{cpp,c,ld}flags): copy caller strings not to
+	  be modified, in append_{cpp,c,ld}flags respectively.
+	  [Fix GH-1246]
+
+Sun Feb 14 16:18:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (setup_exception): set the cause only if it is explicitly
+	  given or not set yet.  [Bug #12068]
+
+Sat Feb 13 21:44:58 2016  Tanaka Akira  <akr@fsij.org>
+
+	* hash.c (rb_hash_invert): [DOC] more examples.
+
+Sat Feb 13 17:30:49 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#find_proxy): support CIDR in
+	  no_proxy.  [ruby-core:73769] [Feature#12062]
+
+Sat Feb 13 17:11:58 2016  Fabian Wiesel  <fabian.wiesel@sap.com>
+
+	* lib/uri/generic.rb (find_proxy): exclude white-spaces and allow
+	  for a leading dot in the domain name in no_proxy.
+	  [ruby-core:54542] [Feature #8317]
+
+Fri Feb 12 12:20:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* error.c (name_err_initialize, nometh_err_initialize): [DOC] fix
+	  argument positions.  optional parameters except for the message
+	  are placed at the last.
+
+Fri Feb 12 11:49:49 2016  Anthony Dmitriyev  <antstorm@gmail.com>
+
+	* net/ftp.rb: add NullSocket#closed? to fix closing not opened
+	  connection.  [Fix GH-1232]
+
+Fri Feb 12 11:17:38 2016  Bogdan  <bogdanvlviv@gmail.com>
+
+	* re.c (rb_reg_initialize_m): [DOC] fix missing right bracket.
+	  [Fix GH-1243]
+
+Thu Feb 11 14:57:58 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (RUBY_CHECK_SIZEOF, RUBY_DEFINT): fix for types
+	  which are conditionally available depending on architectures
+	  when universal binary, e.g., __int128.
+
+Thu Feb 11 06:26:18 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (RUBY_DEFINT): use Parameter Expansion.
+
+Thu Feb 11 05:33:24 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in (int128_t): don't check HAVE_XXX (for example
+	  HAVE___INT128) because RUBY_CHECK_SIZEOF() don't define it for
+	  config.h and use of $ac_cv_sizeof___int128 alternates the check.
+	  (and don't need to define because users shouldn't know that)
+
+Wed Feb 10 12:03:41 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in (ARFLAGS): check if deterministic mode flag is
+	  effective, which is on by default on Ubuntu.
+
+Tue Feb  9 16:36:23 2016  Naotoshi Seo  <sonots@gmail.com>
+
+	* lib/logger.rb: Remove block from Logger.add as it's not needed
+	  patch provided by Daniel Lobato Garcia [fix GH-1240] [Bug #12054]
+
+Tue Feb  9 14:32:23 2016  Zachary Scott  <zzak@ruby-lang.org>
+
+	* ext/zlib/zlib.c: Document mtime header behavior with patch by @schneems
+	  Fixes [GH-1129]: https://github.com/ruby/ruby/pull/1129
+
+Tue Feb  9 13:52:49 2016  Zachary Scott  <zzak@ruby-lang.org>
+
+	* re.c: Remove deprecated kcode argument from Regexp.new and compile
+	  patch provided by Dylan Pulliam [Bug #11495]
+
+Mon Feb  8 21:26:19 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb, enc/unicode/casefold.h: Flags for
+	  upper/lower conversion added (titlecase and SpecialCasing still missing)
+	  (with Kimihito Matsui)
+
+Mon Feb  8 20:43:57 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c, enc/unicode.c: Disassociating ONIGENC_CASE_FOLD flag from
+	  ONIGENC_CASE_DOWNCASE.
+	  (with Kimihito Matsui)
+
+Mon Feb  8 13:00:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Shortened macros for enc/unicode/casefold.h to
+	  single-letter; use flags in casefold.h for logic.
+
+	* enc/unicode/case-folding.rb: Added flag for case folding.
+	  Changed parameter passing.
+
+	* enc/unicode/casefold.h: New flags added.
+	  (with Kimihito Matsui)
+
+Mon Feb  8 10:30:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (feature_option): raise a runtime error if ambiguous
+	  feature name is given, in the future.  [Bug #12050]
+
+Mon Feb  8 09:43:57 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Removed enc/unicode/casefold.h from automatic build because
+	  some CI systems don't have gperf. Creation of enc/unicode/casefold.h
+	  is now possible with make unicode-up. This is intended as a temporary measure.
+
+Sun Feb  7 22:10:08 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Added two more precondition files for enc/unicode/casefold.h
+
+	* enc/unicode.c: Added shortening macros for enc/unicode/casefold.h
+
+	* enc/unicode/case-folding.rb: Fixed file encoding for CaseFolding.txt
+	  to ASCII-8BIT (should fix some ci errors). Clarified usage. Created
+	  class MapItem. Partially implemented class CaseMapping.
+	  (with Kimihito Matsui)
+
+Sun Feb  7 14:12:32 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb: Fixing parameter passing.
+	  (with Kimihito Matsui)
+
+Sun Feb  7 11:44:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode/case-folding.rb: New classes CaseMapping/CaseMappingDummy
+	  to pass as parameters; not yet implemented or used.
+	  (with Kimihito Matsui)
+
+Sun Feb  7 11:16:00 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: using new option in recipe for enc/unicode/casefold.h
+
+	* enc/unicode/case-folding.rb: Correctly specify argument to new option.
+	  (with Kimihito Matsui)
+
+Sun Feb  7 10:43:27 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	(this commit message applies to the previous commit)
+	* common.mk: explicit recipe for enc/unicode/casefold.h
+
+	* enc/unicode/case-folding.rb: Adding -m option to prepare for using
+	  multiple data files.
+	  (with Kimihito Matsui)
+
+Sat Feb  6 22:30:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/cgi/util.rb (escapeHTML, unescapeHTML): consider
+	  ASCII-incompatible encodings.  [Fix GH-1239]
+
+Sat Feb  6 15:18:28 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_regex_casefold.rb: Added data-based testing for
+	  String#downcase :fold.
+
+	* enc/unicode.c: Fixed a range error (lowest non-ASCII character affected
+	  by case operations is U+00B5, MICRO SIGN)
+
+	* test/ruby/enc/test_case_mapping.rb: Explicit test for case folding of
+	  MICRO SIGN to Greek mu.
+	  (with Kimihito Matsui)
+
+Sat Feb  6 14:51:23 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_regex_casefold.rb: Tests for three case folding
+	  primitives (mbc_case_fold, get_case_fold_codes_by_str,
+	  apply_all_case_fold) in the various encodings. Currently only known
+	  good encodings are tested to avoid test failures. For bug hunting,
+	  start by adding more encodings with
+	  generate_test_casefold encoding
+	  (with Kimihito Matsui)
+
+Sat Feb  6 14:37:16 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c, test/ruby/enc/test_case_mapping.rb: Implemented :fold
+	  option for String#downcase by using case folding data from
+	  regular expression engine, and added a few simple tests.
+	  (with Kimihito Matsui)
+
+Fri Feb  5 20:08:59 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_case_mapping.rb: added tests for :ascii option.
+	  (with Kimihito Matsui)
+
+Fri Feb  5 12:22:20 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* insns.def (opt_mult): Use int128_t for overflow detection.
+
+	* bignum.c (rb_uint128t2big): added for opt_mult.
+
+	* bignum.c (rb_uint128t2big): added for rb_uint128t2big..
+
+	* configure.in: define int128_t, uint128_t and related MACROs.
+	  Initially introduced by r41379 but reverted by r50749.
+
+Thu Feb  4 21:05:17 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Activated :ascii flag for ASCII-only case conversion
+	  (with Kimihito Matsui)
+
+Thu Feb  4 17:38:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* re.c (reg_set_source): make source string frozen without
+	  copying.
+
+	* re.c (rb_reg_initialize_m): refactor initialization with
+	  encoding.
+
+Thu Feb  4 15:35:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* string.c (rb_fstring_enc_new, rb_fstring_enc_cstr): functions to
+	  make fstring with encoding.
+
+Thu Feb  4 14:42:29 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Added Unicode data file SpecialCasing.txt to be additionally
+	  downloaded (with Kimihito Matsui)
+
+Thu Feb  4 12:39:08 2016  joker1007  <kakyoin.hierophant@gmail.com>
+
+	* cgi/escape/escape.c: Optimize CGI.escape performance by C ext
+	  for ASCII-compatible encodings.  [Fix GH-1238]
+
+Thu Feb  4 11:53:56 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Introduce two variables (UNICODE_DATA_DIR and
+	  UNICODE_SRC_DATA_DIR) to eliminate repetitions.
+
+Wed Feb  3 12:13:20 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* string.c (str_new_frozen): if the given string is embeddedable
+	  but not embedded, embed a new copied string. [Bug #11946]
+
+Wed Feb  3 08:25:38 2016  boshan  <boshan@subsplash.com>
+
+	* ext/openssl/ossl_pkey.c (Init_ossl_pkey): [DOC] Fix typo
+	  "encrypted" to "decrypted".  [Fix GH-1235]
+
+Wed Feb  3 08:21:32 2016  Seiei Miyagi  <hanachin@gmail.com>
+
+	* ext/ripper/lib/ripper/lexer.rb (on_heredoc_dedent): Fix
+	  Ripper.lex error in dedenting squiggly heredoc.  heredoc tree is
+	  also an array of Elem in the outer tree.  [Fix GH-1234]
+
+Wed Feb  3 02:33:39 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c (rb_reg_prepare_enc): use already compiled US-ASCII regexp
+	  if given string is ASCII only.
+	  121.2s to 113.9s on my x86_64-freebsd10.2 Intel Core i5 661
+
+Tue Feb  2 13:02:03 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* re.c: Introduce RREGEXP_PTR.
+	  patch by dbussink.
+	  partially merge https://github.com/ruby/ruby/pull/497
+
+	* include/ruby/ruby.h: ditto.
+
+	* gc.c: ditto.
+
+	* ext/strscan/strscan.c: ditto.
+
+	* parse.y: ditto.
+
+	* string.c: ditto.
+
+Tue Feb  2 09:08:27 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems/specification.rb: `coding` is effective only first
+	  line except shebang.
+
+	* lib/rubygems/package.rb, lib/rubygems/package/*: ditto.
+
+Mon Feb  1 21:41:58 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/rubygems.rb, lib/rubygems/*, test/rubygems/*: Update rubygems-2.5.2.
+	  It supports to enable frozen string literal and add `--norc` option for
+	  disable to `.gemrc` configuration.
+	  See 2.5.2 release notes for other fixes and enhancements.
+	  https://github.com/rubygems/rubygems/blob/a8aa3bac723f045c52471c7b9328310a048561e0/History.txt#L3
+
+Sun Jan 31 12:33:13 2016  Dan Kreiger  <dan@dankreiger.com>
+
+	* test/drb/ut_large.rb (multiply, avg, median): add additional
+	  math operations to DRbLarge.  [Fix GH-1086]
+
+Sun Jan 31 12:19:15 2016  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_file_exhaustive.rb (test_lstat): Add lacking test
+	  for File#lstat.  [Fix GH-1231]
+
+Sun Jan 31 12:15:33 2016  Prayag Verma  <prayag.verma@gmail.com>
+
+	* doc/standard_library.rdoc: fix typo  [Fix GH-1230]
+	  Spelling mistakes -
+	  outputing > outputting
+	  publich > publish
+
+Sat Jan 30 15:18:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_eval.c (rb_check_funcall_with_hook): also should call the
+	  given hook before returning Qundef when overridden respond_to?
+	  method returned false.  [ruby-core:73556] [Bug #12030]
+
+Fri Jan 29 17:40:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* win32/file.c (rb_readlink): drop garbage after the substitute
+	  name, as rb_w32_read_reparse_point returns the expected buffer
+	  size but "\??\" prefix is dropped from the result.
+
+	* win32/win32.c (w32_readlink): ditto, including NUL-terminator.
+
+Fri Jan 29 17:07:27 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* win32/win32.c (fileattr_to_unixmode, rb_w32_reparse_symlink_p): volume
+	  mount point should be treated as directory, not symlink.
+	  [ruby-core:72483] [Bug #11874]
+
+	* win32/win32.c (rb_w32_read_reparse_point): check the reparse point is
+	  a volume mount point or not.
+
+	* win32/file.c (rb_readlink): follow above change (but this pass won't
+	  be used).
+
+Fri Jan 29 16:17:07 2016  Lucas Buchala  <lucasbuchala@gmail.com>
+
+	* enum.c (enum_take_while, enum_drop_while): rename block
+	  parameter to obj, since they are generic objects.  [Fix GH-1226]
+
+Fri Jan 29 14:15:26 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/erb.rb (ERB::Compiler#detect_magic_comment): allow
+	  frozen-string-literal in comment as well as encoding.
+
+	* lib/erb.rb (ERB#def_method): insert def line just before the
+	  first non-comment and non-empty line, not to leave duplicated
+	  and stale magic comments.
+
+Fri Jan 29 11:13:33 2016  Jeremy Evans  <code@jeremyevans.net>
+
+	* lib/erb.rb (ERB#set_eoutvar): explicitly make mutable string as
+	  a buffer to make ERB work with --enable-frozen-string-literal.
+	  [ruby-core:73561] [Bug #12031]
+
+Fri Jan 29 10:44:56 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/http/header.rb: Warn nil variable on HTTP Header.
+	  It caused to NoMethodError. [fix GH-952][fix GH-641] Patch by @teosz
+	* test/net/http/test_httpheader.rb: Added test for nil HTTP Header.
+
+Thu Jan 28 17:31:43 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/socket.c (sock_gethostname): support unlimited size
+	  hostname.
+
+Wed Jan 27 21:03:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/-ext-/string/test_capacity.rb: Added missing library.
+
+Wed Jan 27 18:53:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Fixed bit mask in macro OnigCodePointCount
+
+Wed Jan 27 17:54:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Protect code point count by macro, in order to
+	  be able to use the remaining bits for flags.
+	  (with Kimihito Matsui)
+
+Wed Jan 27 16:34:35 2016  boshan  <boshan@subsplash.com>
+
+	* lib/tempfile.rb (Tempfile#initialize): [DOC] the first parameter
+	  `basename` is optional and defaulted to an empty string since
+	  [GH-523].  [Fix GH-1225]
+
+Wed Jan 27 16:25:54 2016  Koichi ITO  <koic.ito@gmail.com>
+
+	* array.c (rb_ary_dig): [DOC] fix the exception class to be raised
+	  when intermediate object does not have dig method.  TypeError
+	  will be raised now.  [Fix GH-1224]
+
+Tue Jan 26 19:36:15 2016  Aggelos Avgerinos  <evaggelos.avgerinos@gmail.com>
+
+	* array.c (permute0, rpermute0): [DOC] Substitute indexes ->
+	  indices in documentation for consistency.  [Fix GH-1222]
+
+Tue Jan 26 15:21:37 2016  Eric Wong  <e@80x24.org>
+
+	* compile.c (caller_location): use rb_fstring_cstr for "<compiled>"
+	  (it is converted to fstring anyways inside rb_iseq_new_with_opt)
+	* iseq.c (iseqw_s_compile): ditto
+	* iseq.c (rb_iseq_new_main): use rb_fstring_cstr for "<main>"
+	* vm.c (Init_VM): ditto, share with with above
+	* iseq.c (iseqw_s_compile_file): rb_fstring before rb_io_t->pathv
+	  share "<main>" with above
+	* vm.c (rb_binding_add_dynavars): fstring "<temp>" immediately
+
+Tue Jan 26 15:14:01 2016  Kazuki Yamaguchi  <k@rhe.jp>
+
+	* compile.c (iseq_peephole_optimize): don't apply tailcall
+	  optimization to send/invokesuper instructions with blockiseq.
+	  This is a follow-up to the changes in r51903; blockiseq is now
+	  the third operand of send/invokesuper instructions.
+	  [ruby-core:73413] [Bug #12018]
+
+Tue Jan 26 14:26:46 2016  Eric Wong  <e@80x24.org>
+
+	* signal.c (sig_list): use fstring for hash key
+	* test/ruby/test_signal.rb (test_signal_list_dedupe_keys): added
+
+Tue Jan 26 13:08:34 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* signal.c (rb_f_kill): should immediately deliver reserved
+	  signals SIGILL and SIGFPE, not only SIGSEGV and SIGBUS.
+
+Tue Jan 26 07:57:28 2016  Joseph Tibbertsma  <josephtibbertsma@gmail.com>
+
+	* gc.c (RVALUE_PAGE_WB_UNPROTECTED): fix a typo of argument name.
+	  [Fix GH-1221]
+
+Mon Jan 25 17:26:54 2016  Eric Wong  <e@80x24.org>
+
+	* ruby_assert.h (RUBY_ASSERT_WHEN): fix reference to macro name
+	* vm_core.h: include ruby_assert.h before using
+	  [ruby-core:73371]
+
+Mon Jan 25 15:55:30 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.c (sym_check_asciionly): more informative error message
+	  with the encoding name and the inspected content.
+	  [ruby-core:73398] [Feature #12016]
+
+Mon Jan 25 09:38:26 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/ruby/test_string.rb: added testcase for next!, succ and succ!
+	  [fix GH-1213] Patch by @K0mAtoru
+
+Mon Jan 25 09:32:25 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/webrick/httpservlet/filehandler.rb: fix documentation for namespace.
+	  [fix GH-1219][ci skip] Patch by @leafac
+
+Sun Jan 24 19:34:23 2016  Eric Wong  <e@80x24.org>
+
+	* vm_insnhelper.c (vm_check_if_namespace): tiny size reduction
+
+Sun Jan 24 18:12:36 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Simplifying Unicode data file download logic to make
+	  it more reliable (including additional fix not in r53633) [Bug #12007]
+
+Sun Jan 24 16:54:11 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/io/wait/wait.c (io_wait_readwrite): [EXPERIMENTAL] allow to
+	  wait for multiple modes, readable and writable, at once.  the
+	  arguments may change in the future.  [Feature #12013]
+
+Sat Jan 23 22:30:59 2016  K0mA  <mctj1218@gmail.com>
+
+	* test/ruby/test_array.rb (test_keep_if): Add test for
+	  Array#keep_if separate from Array#select!  [Fix GH-1218]
+
+Sat Jan 23 20:54:26 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* common.mk: revert r53633. It broke rubyci and travis.
+	  https://travis-ci.org/ruby/ruby/builds/104259623
+
+Sat Jan 23 20:10:29 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* range.c (range_eqq): revert r51585 because rb_call_super() is
+	  called in range_include() and thus r51585 doesn't work when the
+	  receiver Range object consists of non linear objects such as Date
+	  objects.
+	  [ruby-core:72908] [Bug #12003]
+
+Sat Jan 23 18:37:37 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* ChangeLog: Fixing wrong time on previous commit, and adding
+	  previous commit message to svn [ci skip]
+
+Sat Jan 23 18:30:30 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Simplifying Unicode data file download logic to make
+	  it more reliable [Bug #12007]
+
+Sat Jan 23 16:29:42 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* tool/downloader.rb: Fixed a logical error, improved documentation
+
+Sat Jan 23 11:42:43 2016  Peter Suschlik  <ps@neopoly.de>
+
+	* README.md: Use SVG Travis badge over PNG for better quality and
+	  device support.  [Fix GH-1214] [Fix GH-1216]
+
+Sat Jan 23 11:29:16 2016  Pascal Betz  <pascal.betz@simplificator.com>
+
+	* lib/csv.rb: Update documentation of CSV header converter for
+	  r45498, [GH-575].  [Fix GH-1215]
+
+Fri Jan 22 17:36:46 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* vm_core.h (VM_ASSERT): use RUBY_ASSERT instead of rb_bug.
+
+	* error.c (rb_assert_failure): assertion with stack dump.
+
+	* ruby_assert.h (RUBY_ASSERT): new header for the assertion.
+
+Fri Jan 22 00:25:57 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* regparse.c (fetch_name_with_level): allow non word characters
+	  at the first character.  [Feature #11949]
+
+	* regparse.c (fetch_name): ditto.
+
+Thu Jan 21 17:34:01 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* marshal.c (r_object0): honor Marshal.load post proc
+	  value for TYPE_LINK.  by Hiroshi Nakamura <nahi@ruby-lang.org>
+	  https://github.com/ruby/ruby/pull/1204 fix GH-1204
+
+Thu Jan 21 16:37:50 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* Makefile.in (update-rubyspec): fix r53208 like r53451.
+
+Wed Jan 20 20:58:25 2016  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* common.mk, Makefile.in: update-config_files is only for Unix
+	  platforms.
+
+Wed Jan 20 17:13:39 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/extlibs.rb: add --cache option to change cache directory.
+
+Tue Jan 19 17:03:40 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* common.mk: Added Unicode data file CaseFolding.txt to be additionally
+	  downloaded (with Kimihito Matsui)
+
+Tue Jan 19 10:09:58 2016  Sho Hashimoto  <sho-h@ruby-lang.org>
+
+	* lib/shell.rb (Shell.debug_output_exclusive_unlock): remove
+	  because Mutex#exclusive_unlock was already deleted. [fix GH-1185]
+
+Tue Jan 19 09:38:27 2016  Nick Cox  <nick@nickcox.me>
+
+	* vm_method.c: fix grammar in respond_to? warning.
+	  [fix GH-1047]
+
+Mon Jan 18 14:37:07 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_here_document): an escaped newline is not an
+	  actual newline, and the rest part should not be dedented.
+	  [ruby-core:72855] [Bug #11989]
+
+Mon Jan 18 12:04:34 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/ruby/test_string.rb: Added extra testcase for test_rstrip_bang
+	  and test_lstrip_bang. [fix GH-1178] Patch by @Matrixbirds
+
+Mon Jan 18 11:47:27 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* string.c: fix a typo. [fix GH-1202][ci skip] Patch by @sunboshan
+
+Sun Jan 17 21:15:30 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* configure.in: improve ICC (Intel C Compiler) support.
+
+	* configure.in (CXX): The name of icc's c++ compiler is `icpc`.
+
+	* configure.in (warnings): Add `-diag-disable=2259` to suppress
+	  noisy warnings: "non-pointer conversion from "..." to "..." may
+	  lose significant bits".
+
+	* configure.in (optflags): Add `-fp-model precise` like -fno-fast-math.
+
+	* lib/mkmf.rb: icc supports -Werror=division-by-zero
+	  and -Werror=deprecated-declarations, but doesn't support
+	  -Wdivision-by-zero and -Wdeprecated-declarations.
+
+Sun Jan 17 20:40:10 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Any kind of option is now taking the new code path for
+	  upcase/downcase/capitalize/swapcase. :lithuanian can be used for
+	  testing if no specific option is desired.
+	* test/ruby/enc/test_case_mapping.rb: Adjusted to above.
+	  (with Kimihito Matsui)
+
+Sun Jan 17 20:10:10 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Fixed a logical error and some comments.
+	* test/ruby/enc/test_case_mapping.rb: Made tests more general.
+	  (with Kimihito Matsui)
+
+Sun Jan 17 17:41:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Removed artificial expansion for Turkic,
+	  added hand-coded support for Turkic, fixed logic for swapcase.
+	* string.c: Made use of new case mapping code possible from upcase,
+	  capitalize, and swapcase (with :lithuanian as a guard).
+	* test/ruby/enc/test_case_mapping.rb: Adjusted for above.
+	  (with Kimihito Matsui)
+
+Sun Jan 17 15:30:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/option.c (sockopt_bool): relax boolean size to be one
+	  too not only sizeof(int).  Winsock getsockopt() returns a single
+	  byte as a boolean socket option.  [ruby-core:72730] [Bug #11958]
+
+Sun Jan 17 14:43:01 2016  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_env.rb: [Fix GH-1201]
+	  * Extract test code for ENV#keep_if from ENV#select_bang
+	  * Add a test case for ENV#select_bang,keep_if
+
+Sun Jan 17 14:42:25 2016  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_env.rb: [Fix GH-1201]
+	  * Extract test code for ENV#delete_if from ENV#reject_bang
+	  * Add a test case for ENV#reject_bang,delete_if
+
+Sun Jan 17 14:40:22 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ext/socket/option.c (check_size): extract a macro to check
+	  binary data size, with a consistent message.
+
+	* ext/socket/option.c (sockopt_byte): fix error message,
+	  sizeof(int) differs from sizeof(unsigned char) in general.
+
+Sat Jan 16 21:16:21 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (xstring): reset heredoc indent after dedenting,
+	  so that following string literal would not be dedented.
+	  [ruby-core:72857] [Bug #11990]
+
+Sat Jan 16 17:24:24 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* enc/unicode.c: Artificial mapping to test buffer expansion code.
+	* string.c: Fixed buffer expansion logic.
+	* test/ruby/enc/test_case_mapping.rb: Tests for above.
+	  (with Kimihito Matsui)
+
+Sat Jan 16 16:47:14 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ext/openssl/lib/openssl/pkey.rb: Added 2048 bit DH parameter.
+	* test/openssl/test_pkey_dh.rb: ditto.
+
+Sat Jan 16 10:51:19 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* enc/unicode.c: fix implicit conversion error with clang. fixup r53548.
+	* string.c: ditto.
+
+Sat Jan 16 10:31:00 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* common.mk: test-sample was changed to test-basic.
+	  [Feature #11982][ruby-core:72823]
+	* basictest/runner.rb: ditto. rename from tool/rubytest.rb.
+	* basictest/test.rb: ditto. rename from sample/test.rb.
+	* defs/gmake.mk: picked from r53540
+	* sample/test.rb: backward compatibility for chkbuild.
+
+Sat Jan 16 10:23:23 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c, enc/unicode.c: New code path as a preparation for Unicode-wide
+	  case mapping. The code path is currently guarded by the :lithuanian
+	  option to avoid accidental problems in daily use.
+	* test/ruby/enc/test_case_mapping.rb: Test for above.
+	* string.c: function 'check_case_options': fixed logical errors
+	  (with Kimihito Matsui)
+
+Fri Jan 15 20:20:20 2016  Naohisa Goto  <ngotogenome@gmail.com>
+
+	* regint.h (PLATFORM_UNALIGNED_WORD_ACCESS): The value of
+	  UNALIGNED_WORD_ACCESS should be used to determine whether
+	  unaligned word access is allowed or not. After this commit,
+	  ./configure CPPFLAGS="-DUNALIGNED_WORD_ACCESS=0" disables
+	  unaligned word access even on platforms that support the feature.
+
+Fri Jan 15 16:12:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (string1): reset heredoc indent for each string literal
+	  so that concatenated string would not be dedented.
+	  [ruby-core:72857] [Bug #11990]
+
+Thu Jan 14 20:01:00 2016  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* lib/uri/generic.rb (URI::Generic#to_s): change encoding to
+	  UTF-8 as Ruby 2.2/ by Koichi ITO <koic.ito@gmail.com>
+	  https://github.com/ruby/ruby/pull/1188 fix GH-1188
+
+Thu Jan 14 17:36:16 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_f_global_variables): add matched back references
+	  only, as well as defined? operator.
+
+Thu Jan 14 16:12:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* sprintf.c (rb_str_format): format exact number more exactly.
+
+Thu Jan 14 15:08:43 2016  Tony Arcieri  <bascule@gmail.com>
+
+	* Remove 512-bit DH group. It's affected by LogJam Attack.
+	  https://weakdh.org/
+	  [fix GH-1196][Bug #11968][ruby-core:72766]
+
+Thu Jan 14 11:44:29 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* variable.c (rb_f_global_variables): add $1..$9 only if $~ is
+	  set.  fix the condition removed at r14014.
+
+Wed Jan 13 17:21:45 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* .travis.yml: removed commented-out code.
+
+Wed Jan 13 17:14:54 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* .travis.yml: removed osx code. follow up with r53517
+
+Wed Jan 13 16:56:19 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* iseq.c (rb_iseq_mark): mark parent iseq to prevent dynamically
+	  generated iseq by eval from GC.  [ruby-core:72620] [Bug #11928]
+
+Wed Jan 13 03:42:58 2016  Eric Wong  <e@80x24.org>
+
+	* class.c (Init_class_hierarchy): resolve name for rb_cObject ASAP
+	* object.c (rb_mod_const_set): move name resolution to rb_const_set
+	* variable.c (rb_const_set): do class resolution here
+	  [ruby-core:72807] [Bug #11977]
+
+Wed Jan 13 00:37:12 2016  Satoshi Ohmori  <sachin21dev@gmail.com>
+
+	* man/ruby.1: fix double word typo.  [Fix GH-1194]
+
+Tue Jan 12 21:01:09 2016  Benoit Daloze  <eregontp@gmail.com>
+
+	* common.mk: update URL and name for the Ruby spec suite.
+
+Tue Jan 12 19:52:19 2016  sorah (Shota Fukumori)  <her@sorah.jp>
+
+	* lib/forwardable.rb: Convert given accessors to String.
+
+	  r53381 changed to accept only Symbol or String for accessors, but
+	  there are several rubygems that pass classes (e.g. Array,
+	  Hash, ...) as accessors. Prior r53381, it was accepted because Class#to_s
+	  returns its class name. After r53381 given accessors are checked
+	  with define_method, but it accepts only Symbol or String, otherwise
+	  raises TypeError.
+
+	      def_delegator Foo, :some_method
+
+	  This change is to revert unexpected incompatibility. But this behavior
+	  may change in the future.
+
+Mon Jan 12 18:41:41 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: made a variable name more grammatically correct
+
+Mon Jan 12 18:34:34 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: minor grammar fix [ci skip]
+
+Mon Jan 12 16:09:09 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* test/ruby/enc/test_casing_options.rb: Tests for option
+	  parsing/checking for upcase/downcase/capitalize/swapcase
+	  (see r53503; with Kimihito Matsui)
+
+Mon Jan 12 16:03:03 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* string.c: Added option parsing/checking for upcase/downcase/
+	  capitalize/swapcase (with Kimihito Matsui)
+
+Mon Jan 11 21:28:28 2016  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* include/ruby/oniguruma.h: Added flags needed for upcase/downcase
+	  Unicode addition (with Kimihito Matsui)
+
+Mon Jan 11 09:50:24 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: check if the API version number is consistent with
+	  the program version number.
+
+Sun Jan 10 20:57:25 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (compile_massign_lhs): when index ends with splat,
+	  append rhs value to it like POSTARG, since VM_CALL_ARGS_SPLAT
+	  splats the last argument only.  [ruby-core:72777] [Bug #11970]
+
+Sun Jan 10 15:45:10 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/missing.h (explicit_bzero_by_memset_s): remove
+	  inline implementation by memset_s, which needs a macro before
+	  including headers and can cause problems in extension libraries
+	  by the order of the macro and headers.
+
+Sun Jan 10 13:41:36 2016  Eric Wong  <e@80x24.org>
+
+	* io.c (rb_deferr): remove long obsolete global
+
+Sun Jan 10 09:14:42 2016  Eric Wong  <e@80x24.org>
+
+	* ext/psych/lib/psych/visitors/yaml_tree.rb (visit_String):
+	  eliminate chomp
+	* lib/net/http.rb (connect): eliminate delete
+	* lib/net/http/header.rb (basic_encode): ditto
+	* lib/net/imap.rb (authenticate): eliminate gsub
+	  (self.encode_utf7): shorten delete arg
+	* lib/net/smtp.rb (base64_encode): eliminate gsub
+	* lib/open-uri.rb (OpenURI.open_http): eliminate delete
+	* lib/rss/rss.rb: ditto
+	* lib/securerandom.rb (base64): ditto
+	  (urlsafe_base64): eliminate delete!
+	* lib/webrick/httpauth/digestauth.rb (split_param_value):
+	  eliminate chop
+	* lib/webrick/httpproxy.rb (do_CONNECT): eliminate delete
+	  (setup_upstream_proxy_authentication): ditto
+	  [ruby-core:72666] [Feature #11938]
+
+Sat Jan  9 23:19:14 2016  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_hash.rb (test_try_convert): Add test for
+	  Hash.try_convert.  [Fix GH-1190]
+
+Sat Jan  9 23:15:25 2016  Jon Moss  <maclover7@users.noreply.github.com>
+
+	* ext/openssl/ossl.c: Add missing variables to documentation
+	  examples.  [Fix GH-1189]
+
+Sat Jan  9 18:25:57 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* symbol.h (is_attrset_id): ASET is an attrset ID.  fix
+	  unexpected safe call instead of an ordinary ASET.
+
+Sat Jan  9 10:44:33 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in, win32/setup.mak: extract RUBY_PROGRAM_VERSION from
+	  RUBY_VERSION in version.h instead of RUBY_API_VERSION numbers in
+	  include/ruby/version.h, and cut it into version numbers.
+
+Sat Jan  9 07:13:33 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename PAGE_* to HEAP_PAGE_* because PAGE_SIZE is used
+	  in Mac OS X.
+
+	* test/ruby/test_gc.rb: catch up this fix.
+
+Sat Jan  9 05:45:40 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: PAGE_BITMAP_PLANES (the number of bitmap) is 4, not 3.
+
+Sat Jan  9 05:42:57 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename constant names HEAP_* to PAGE_*.
+
+	  Keys of GC::INTERNAL_CONSTANTS are also renamed.
+
+	* test/ruby/test_gc.rb: catch up this fix.
+
+Fri Jan  8 22:30:06 2016  Akinori MUSHA  <knu@iDaemons.org>
+
+	* doc/regexp.rdoc: [DOC] Elaborate on the \G anchor. [ci skip]
+
+Fri Jan  8 19:49:27 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove heap_page::body. Instead of this field,
+	  heap_page::start field works well.
+
+Fri Jan  8 19:31:52 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: rename rb_heap_t::page_length to rb_heap_t::total_pages.
+
+	  `page_length' is not clear (we may understand with length of
+	  a page).
+
+Fri Jan  8 17:07:14 2016  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c: remove heap_page::heap. This field is only used to recognize
+	  whether a page is in a tomb or not. Instead of this field,
+	  heap_page::flags::in_tomb (1 bit field) is added.
+
+	  Also type of heap_page::(total|free|final)_slots are changed from
+	  int to short. 2B is enough for them.
+
+Fri Jan  8 09:33:59 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* iseq.c (rb_iseq_compile_with_option): move variable initialization
+	  code to avoid maybe-uninitialized warnings by gcc 4.8.
+
+Fri Jan  8 00:03:22 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enum.c (enum_min, enum_max): do the same optimization as r53454.
+
+Thu Jan  7 22:32:21 2016  Kenta Murata  <mrkn@mrkn.jp>
+
+	* ruby.h: undef HAVE_BUILTIN___BUILTIN_CHOOSE_EXPR_CONSTANT_P
+	  and HAVE_BUILTIN___BUILTIN_TYPES_COMPATIBLE_P on C++.
+	  [ruby-core:72736] [Bug #11962]
+
+Thu Jan  7 22:02:21 2016  Shugo Maeda  <shugo@ruby-lang.org>
+
+	* enum.c (enum_minmax): optimize object comparison in
+	  Enumerable#minmax.
+
+Thu Jan  7 14:49:12 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* thread.c (rb_thread_pending_interrupt_p): no pending interrupt
+	  before initialization.
+
+	* thread.c (thread_raise_m, rb_thread_kill): uninitialized thread
+	  cannot interrupt.  [ruby-core:72732] [Bug #11959]
+
+Thu Jan  7 11:34:14 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* include/ruby/backward.h (ruby_show_copyright_to_die): for source
+	  code backward compatibility.
+
+	* ruby.c (process_options): return Qtrue to exit the process
+	  successfully.
+
+	* version.c (ruby_show_copyright): no longer exit.
+
+Wed Jan  6 17:22:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/optparse.rb (OptionParser#order!): add `into` optional
+	  keyword argument to store the results.  [Feature #11191]
+
+Tue Jan  5 21:44:37 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* ChangeLog: fix wrong class name.
+
+Tue Jan  5 21:43:50 2016  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_string.rb(test_chr): added test for String#chr
+	  [fix GH-1179]
+
+Tue Jan  5 21:32:26 2016  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_numeric.rb (test_nonzero_p): added test for Numeric#nonzero?
+	  [fix GH-1187]
+
+Tue Jan  5 11:47:23 2016  Damir Gaynetdinov  <damir.gaynetdinov@gmail.com>
+
+	* doc/marshal.rdoc: Clarify object references example, that the
+	  reference is same object.  [Fix GH-1156]
+
+Tue Jan  5 05:06:51 2016  Eric Wong  <e@80x24.org>
+
+	* ext/stringio/stringio.c (strio_binmode): implement to set encoding
+	* test/stringio/test_stringio.rb (test_binmode): new test
+	  [ruby-core:72699] [Bug #11945]
+
+Mon Jan  4 15:44:37 2016  Sho Hashimoto  <sho-h@ruby-lang.org>
+
+	* variable.c (rb_mod_deprecate_constant): [DOC] added
+	  documentation for Module#deprecate_constant. [ci skip]
+
+Mon Jan  4 15:36:38 2016  Sho Hashimoto  <sho-h@ruby-lang.org>
+
+	* thread_sync.c: [DOC] remove SizedQueue#close argument.
+	  [ci skip]
+
+Mon Jan  4 10:14:24 2016  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* test/coverage/test_coverage.rb: ignored test when enabled to coverage.
+	  It lead to crash with `make test-all`.
+
+Mon Jan  4 08:10:44 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* insns.def (opt_case_dispatch): Move a comment to the
+	  appropriate position.
+	  [ci skip]
+
+Sun Jan  3 23:55:13 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/rubygems/security.rb (DIGEST_ALGORITHM, KEY_ALGORITHM):
+	  should check same name as the used constants.
+	  [ruby-core:72674] [Bug #11940]
+
+Sun Jan  3 19:22:01 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* aclocal.m4: add fallback file for non-aclocal environments.
+	  [ruby-core:72683] [Bug #11942]
+
+Sun Jan  3 13:56:49 2016  Yuichiro Kaneko  <yui-knk@ruby-lang.org>
+
+	* eval_error.c (rb_print_undef): Use `rb_method_visibility_t`
+	  instead of `int`.
+	* eval_intern.h (rb_print_undef): ditto
+	* proc.c (mnew_internal): ditto
+	* vm_method.c (rb_export_method): ditto
+	  [Misc #11649] [ruby-core:71311] [fix GH-1078]
+
+Sun Jan  3 12:12:09 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* acinclude.m4: rename aclocal.m4, which should be generated by
+	  aclocal.  [ruby-core:72675] [Bug #11941]
+
+Sat Jan  2 21:07:55 2016  Eric Wong  <e@80x24.org>
+
+	* thread_sync.c (queue_do_pop): avoid cast with Qfalse
+	  (rb_szqueue_push): ditto, use queue_sleep wrapper
+
+Sat Jan  2 16:16:14 2016  Masatoshi SEKI  <m_seki@mva.biglobe.ne.jp>
+
+	* lib/erb.rb: Allow ERB subclass to add token easily.
+	  [Feature #11936]
+
+	* test/erb/test_erb.rb: ditto.
+
+Sat Jan  2 14:44:31 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (regexp): set_yylval_num sets u1, should use nd_tag
+	  instead of nd_state.  [ruby-core:72638] [Bug #11932]
+
+Sat Jan  2 02:27:22 2016  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb: Fix case of frozen object with initializer.
+	  Bug revealed by RubySpec [ruby-core:72639]
+
+Fri Jan  1 22:01:52 2016  Kazuhiro NISHIYAMA  <zn@mbf.nifty.com>
+
+	* NEWS: mention CSV's liberal_parsing option.
+
+Fri Jan  1 19:38:23 2016  okkez  <okkez000@gmail.com>
+
+	* doc/NEWS-2.3.0: fix double words typo.
+	  [ci skip][fix GH-1183]
+
+Fri Jan  1 15:28:56 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (remove_unreachable_chunk): decrease count of
+	  call_info in removed instructions.  fix up r53402.
+
+Fri Jan  1 12:05:53 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* compile.c (remove_unreachable_chunk): remove unreferred label
+	  to optimize away unreachable chunk.
+
+Fri Jan  1 11:42:57 2016  James Edward Gray II  <james@graysoftinc.com>
+
+	* lib/csv.rb (CSV): Add a liberal_parsing option.
+	  Patch by Braden Anderson. [#11839]
+	* test/csv/test_features.rb:  test liberal_parsing
+
+Fri Jan  1 10:27:28 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* tool/mkconfig.rb (RbConfig): prefix SDKROOT to oldincludedir
+	  not includedir, the latter is outside the ruby installation.
+	  [ruby-core:72496] [Bug #11881]
+
+Fri Jan  1 08:53:02 2016  Yuki Kurihara  <co000ri@gmail.com>
+
+	* test/ruby/test_lazy_enumerator.rb (test_take_bad_arg): Add test
+	  code in case of Enumerator::Lazy#take called with negative number.
+	  [ruby-dev:49467] [Bug #11933]
+
+Fri Jan  1 05:06:20 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (parser_here_document): update indent for each line in
+	  indented here document with single-quotes.
+	  [ruby-core:72479] [Bug #11871]
+
+Fri Jan  1 03:26:44 2016  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (freeze): define deferred accessors before
+	  freezing to get rid of an error when just reading frozen
+	  OpenStruct.
+
+Thu Dec 31 14:36:45 2015  Marc-Andre Lafortune  <ruby-core@marc-andre.ca>
+
+	* lib/ostruct.rb: Fix new_ostruct_member to correctly avoid
+	  redefinition [#11901]
+
+Thu Dec 31 02:45:12 2015  NARUSE, Yui  <naruse@ruby-lang.org>
+
+	* test/ruby/test_module.rb (test_classpath): r53376 may change
+	  the order of m.constants.
+	  `make TESTS='-v ruby/test_class.rb ruby/test_module.rb' test-all`
+	  may fail after that.
+	  http://rubyci.s3.amazonaws.com/tk2-243-31075/ruby-trunk/log/20151230T164202Z.log.html.gz
+
+Thu Dec 31 02:20:00 2015  Benoit Daloze  <eregontp@gmail.com>
+
+	* common.mk (help): Fix typo.
+
+Wed Dec 30 20:53:09 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/net/http/responses.rb: Added new response class for 451 status code.
+	* lib/net/http.rb: documentation for HTTPUnavailableForLegalReasons
+
+Wed Dec 30 20:45:45 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* lib/webrick/httpstatus.rb: Added HTTP 451 Status Code.
+	  [fix GH-1167] Patch by @MuhammetDilmac
+	  https://tools.ietf.org/html/draft-tbray-http-legally-restricted-status-00
+
+Wed Dec 30 20:25:52 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/syntax/calling_methods.rdoc: fix old operator for safe navigation
+	  operator. [ci skip][fix GH-1182] Patch by @dougo
+
+Wed Dec 30 16:43:23 2015  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_string.rb (test_ord): Add test for String#ord.
+	  [Fix GH-1181]
+
+Wed Dec 30 11:28:57 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/forwardable.rb (def_instance_delegator): adjust backtrace of
+	  method body by tail call optimization.  adjusting the delegated
+	  target is still done by deleting backtrace.
+
+	* lib/forwardable.rb (def_single_delegator): ditto.
+
+Wed Dec 30 11:18:42 2015  Elliot Winkler  <elliot.winkler@gmail.com>
+
+	* lib/forwardable.rb (def_instance_delegator) fix delegating to
+	  'args' and 'block', clashing with local variables in generated
+	  methods.  [ruby-core:72579] [Bug #11916]
+
+	* lib/forwardable.rb (def_single_delegator): ditto.
+
+Wed Dec 30 09:58:56 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* object.c (rb_class_inherited_p): search the corresponding
+	  ancestor to prepended module from prepending class itself.
+	  [ruby-core:72493] [Bug #11878]
+
+Wed Dec 30 09:20:00 2015  Yuki Kurihara  <co000ri@gmail.com>
+
+	* test/stringio/test_io.rb (test_flag): add assertion for error when
+	  text and binary mode are mixed.
+	  [ruby-dev:49465] [Feature #11921]
+
+Wed Dec 30 08:43:59 2015  Yuki Kurihara  <co000ri@gmail.com>
+
+	* test/stringio/test_stringio.rb (test_initialize): add test for
+	  StringIO#initialize.  [ruby-core:72585] [Feature #11920]
+
+Wed Dec 30 05:19:24 2015  Eric Wong  <e@80x24.org>
+
+	* class.c (struct clone_const_arg): adjust for id_table
+	  (clone_const): ditto
+	  (clone_const_i): ditto
+	  (rb_mod_init_copy): ditto
+	  (rb_singleton_class_clone_and_attach): ditto
+	  (rb_include_class_new): ditto
+	  (include_modules_at): ditto
+	* constant.h (rb_free_const_table): ditto
+	* gc.c (free_const_entry_i): ditto
+	  (rb_free_const_table): ditto
+	  (obj_memsize_of): ditto
+	  (mark_const_entry_i): ditto
+	  (mark_const_tbl): ditto
+	* internal.h (struct rb_classext_struct): ditto
+	* object.c (rb_mod_const_set): resolve class name on assignment
+	* variable.c (const_update): replace with const_tbl_update
+	  (const_tbl_update): new function
+	  (fc_i): adjust for id_table
+	  (find_class_path): ditto
+	  (autoload_const_set): st_update => const_tbl_update
+	  (rb_const_remove): adjust for id_table
+	  (sv_i): ditto
+	  (rb_local_constants_i): ditto
+	  (rb_local_constants): ditto
+	  (rb_mod_const_at): ditto
+	  (rb_mod_const_set): ditto
+	  (rb_const_lookup): ditto
+	  [ruby-core:72112] [Feature #11614]
+
+Wed Dec 30 04:10:13 2015  CHIKANAGA Tomoyuki  <nagachika@ruby-lang.org>
+
+	* thread_pthread.c (rb_thread_create_timer_thread): destroy attr even
+	  if pthread_create() failed.
+
+Wed Dec 30 02:55:09 2015  Eric Wong  <e@80x24.org>
+
+	* thread_pthread.c (setup_communication_pipe): delay setting owner
+	  (rb_thread_create_timer_thread): until thread creation succeeds
+	  [ruby-core:72590] [Bug #11922]
+
+Tue Dec 29 19:12:46 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* ruby.c (proc_options): -W command line option should be able to
+	  override -w in RUBYOPT environment variable.
+
+Tue Dec 29 17:54:16 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* eval.c (ignored_block): warn if a block is given to `using`,
+	  which is probably for `Module.new`.
+
+Tue Dec 29 12:48:34 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* lib/ostruct.rb (OpenStruct): make respond_to? working on
+	  just-allocated objects for workaround of Psych.
+	  [ruby-core:72501] [Bug #11884]
+
+Tue Dec 29 10:35:00 2015  Kenta Murata  <mrkn@mrkn.jp>
+
+	* test/mkmf/test_have_func.rb (test_have_func):
+	  Add assertion to examine the existence of HAVE_RUBY_INIT.
+
+	* test/mkmf/test_have_func.rb (test_not_have_func):
+	  Add assertion to examine the absence of HAVE_RUBY_INIT.
+
+Tue Dec 29 06:50:42 2015  Eric Wong  <e@80x24.org>
+
+	* thread_sync.c: static classes
+
+Tue Dec 29 05:30:30 2015  Eric Wong  <e@80x24.org>
+
+	* lib/resolv.rb (Resolv::IPv6.create): avoid modifying frozen
+	* test/resolv/test_dns.rb (test_ipv6_create): test for above
+	  [Bug #11910] [ruby-core:72559]
+
+Mon Dec 28 14:55:57 2015  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_string.rb (TestString#test_rstrip_bang): Add test
+	  for String#rstrip!.  [Fix GH-1176]
+
+Mon Dec 28 09:18:53 2015  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_string.rb (TestString#test_lstrip_bang): Add test
+	  for String#lstrip!.  [Fix GH-1176]
+
+Sun Dec 27 23:32:26 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_variant2val): refactoring.
+
+Sun Dec 27 21:14:42 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c (rb_execarg_parent_start1): need to convert the encoding to
+	  ospath's one.
+
+Sun Dec 27 20:54:22 2015  NAKAMURA Usaku  <usa@ruby-lang.org>
+
+	* process.c: use rb_w32_uchdir() instead of plain chdir() on Windows.
+	  reported by naruse via twitter.
+
+	* process.c (rb_execarg_addopt): need to convert the encoding to
+	  ospath's one.
+
+Sun Dec 27 20:00:31 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* enc/x_emoji.h: fix dead-link.
+
+Sun Dec 27 19:55:55 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
+
+	* doc/NEWS-2.3.0: fix a typo.
+
+Sun Dec 27 18:08:15 2015  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* string.c (rb_str_lstrip_bang, rb_str_rstrip_bang): [DOC] Fix
+	  ruby-doc comments for String#rstrip! and lstrip!.  It looks like
+	  dropped bang.  [Fix GH-1175]
+
+Sun Dec 27 15:14:20 2015  Eric Wong  <e@80x24.org>
+
+	* io.c (io_getpartial): remove unused kwarg from template
+	* test/ruby/test_io.rb (test_readpartial_bad_args): new
+	  [Bug #11885]
+
+Sun Dec 27 11:50:53 2015  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_string.rb (test_rstrip, test_lstrip): Add tests
+	  for String#lstrip and rstrip.  The test cases are used from
+	  string.c ruby-doc comments.  [Fix GH-1174]
+
+Sun Dec 27 11:47:46 2015  Kuniaki IGARASHI  <igaiga@gmail.com>
+
+	* test/ruby/test_string.rb (test_insert): Add test for
+	  String#insert.  The test cases are written in string.c
+	  comments as a reference.  [Fix GH-1173]
+
+Sun Dec 27 11:03:33 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* parse.y (show_bitstack): trace stack_type value if yydebug.
+
+Sun Dec 27 10:03:14 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* enc/depend (enc, trans): fix version dependency, shared object
+	  files depend on the RUBY_SO_NAME value for runtime link.
+
+Sun Dec 27 09:47:20 2015  Masaki Suketa <masaki.suketa@nifty.ne.jp>
+
+	* ext/win32ole/win32ole.c (ole_vstr2wc, ole_variant2val): fix blank
+	  string conversion.
+	  [Bug #11880]
+	  Thanks Akio Tajima for the patch!
+
+Sun Dec 27 09:34:53 2015  craft4coder  <yooobuntu@163.com>
+
+	* doc/extension.rdoc: [DOC] `nul` should be uppercase.
+	  change 'nul' => 'NUL'.  [Fix GH-1172]
+
+Sat Dec 26 18:29:01 2015  Kouhei Sutou  <kou@cozmixng.org>
+
+	* lib/xmlrpc/client.rb: Support SSL options in async methods of
+	  XMLRPC::Client.
+	  [Bug #11489]
+	  Reported by Aleksandar Kostadinov. Thanks!!!
+
+Sat Dec 26 11:26:38 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* miniinit.c (Init_enc): add some common aliases of built-in
+	  encodings.  [ruby-core:72481] [Bug #11872]
+
+Fri Dec 25 22:43:26 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* configure.in: extract RUBY_RELEASE_DAY at generating Makefile.
+
+	* version.h (RUBY_RELEASE_DATE): construct from RUBY_RELEASE_YEAR,
+	  RUBY_RELEASE_MONTH, and RUBY_RELEASE_DAY.
+
+Fri Dec 25 21:33:06 2015  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* version.h (RUBY_VERSION): 2.4.0 development has started.
+
+Fri Dec 25 14:12:12 2015  Martin Duerst  <duerst@it.aoyama.ac.jp>
+
+	* doc/ChangeLog-2.3.0, ext/tk/lib/tkextlib/SUPPORT_STATUS,
+	  include/ruby/version.h: minor grammar fixes [ci skip]
+
+Fri Dec 25 08:23:22 2015  Tadashi Saito  <tad.a.digger@gmail.com>
+
+	* compile.c, cont.c, doc, man: fix common misspelling.
+	  [ruby-core:72466] [Bug #11870]
+
+
+For the changes before 2.3.0, see doc/ChangeLog-2.3.0
+For the changes before 2.2.0, see doc/ChangeLog-2.2.0
+For the changes before 2.1.0, see doc/ChangeLog-2.1.0
+For the changes before 2.0.0, see doc/ChangeLog-2.0.0
+For the changes before 1.9.3, see doc/ChangeLog-1.9.3
+For the changes before 1.8.0, see doc/ChangeLog-1.8.0
+
+Local variables:
+coding: us-ascii
+add-log-time-format: (lambda (&optional x y)
+  (let* ((time (or x (current-time)))
+	 (system-time-locale "C")
+	 (diff (+ (cadr time) 32400))
+	 (lo (% diff 65536))
+	 (hi (+ (car time) (/ diff 65536))))
+  (format-time-string "%a %b %e %H:%M:%S %Y" (list hi lo) t)))
+indent-tabs-mode: t
+tab-width: 8
+change-log-indent-text: 2
+end:
+vim: tabstop=8 shiftwidth=2
diff --git a/doc/ChangeLog-2016 b/doc/ChangeLog-2016
new file mode 100644
index 0000000000..c708428a93
--- /dev/null
+++ b/doc/ChangeLog-2016
@@ -0,0 +1,5 @@
+------------------------------------------------------------------------
+r56645 | naruse | 2016-11-07 00:56:27 +0900 (Mon, 07 Nov 2016) | 1 line
+
+Obsolete ChangeLog [Feature #12283]
+------------------------------------------------------------------------
diff --git a/doc/ChangeLog-YARV b/doc/ChangeLog-YARV
new file mode 100644
index 0000000000..a8b999dff2
--- /dev/null
+++ b/doc/ChangeLog-YARV
@@ -0,0 +1,6917 @@
+# $Id: ChangeLog 590 2006-12-31 09:02:34Z ko1 $
+#
+# YARV ChangeLog
+#  from Mon, 03 May 2004 01:24:19 +0900
+#
+
+Sun Dec 31 18:01:50 2006  Koichi Sasada  <ko1@atdot.net>
+
+	* bin/* : ruby/trunk/bin 11437
+
+
+Sun Dec 31 17:42:05 2006  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : remove old Kernel#funcall definition
+
+
+2006-12-30(Sat) 07:59:26 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* catch up ruby/trunk 11437
+
+	* eval_intern.h : reorder tag initialization
+
+	* eval.c : fix to support __send!, funcall and prohibit funcall as
+	send
+
+	* eval_error.h, eval_jump.h, eval_safe.h : fix prototypes
+
+	* eval_method.h, vm.c : check re-definition at rb_add_method()
+
+	* yarvcore.h : fix typo
+
+	* compile.c : fix white spaces
+
+	* lib/delegate.rb : fix to support __send, ...
+
+	* lib/getoptlong.rb : fix to work on YARV
+
+	* lib/rss/parser.rb : use __send! instead of __send__
+
+	* sample/test.rb : comment out codes which use |&b| type block parameter
+
+	* ext/ripper/extconf.rb : turn off
+
+	* test/ripper/test_files.rb, test_parser_events.rb,
+	test_scanner_events.rb : fix to check it has ripper module
+
+	* vm_dump.c : remove showing file path length limitation
+
+	* yarvtest/test_eval.rb : use __send! instead of __send__
+
+
+2006-12-19(Tue) 11:46:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* doc/* : added
+
+	* ext/openssl : added
+
+	* ext/ripper : added
+
+	* test/openssl : added
+
+	* test/ripper : added
+
+	* misc : added
+
+	* rb/ -> tool/ : renamed
+
+	* common.mk : fixed for above change
+
+	* ruby_doc/* : move to topdir
+
+	* sample/* : added
+
+	* test2.rb : removed
+
+
+2006-12-15(Fri) 09:42:46 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : remove obsolete codes
+
+	* insns.def : fix a comment of getconstant
+
+
+2006-12-13(Wed) 16:26:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c, compile.c, compile.h, debug.c, debug.h,
+	insnhelper.h, insns.def, iseq.c, thread.c, thread_pthread.ci,
+	thread_pthread.h, thread_win32.ci, thread_win32.h, vm.c, vm.h,
+	vm_dump.c, vm_evalbody.ci,  vm_opts.h.base,  yarv.h,
+	yarv_version.h, yarvcore.c, yarvcore.h :
+	add a header includes copyright
+
+
+2006-12-12(Tue) 13:13:32 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb : add PREFETCH() statement
+
+	* vm.h : ditto
+
+	* yarvcore.h : fix LIKELY(x) and
+	remove main_thread_val field from yarv_vm_t
+
+	* yarvcore.c : ditto
+
+	* thread.c : support fork
+
+	* eval_thread.c : ditto
+
+	* process.c : ditto
+
+	* signal.c : ditto
+
+	* test/ruby/test_signal.rb :
+
+	* thread_pthread.ci : rename timer thread functions
+
+	* thread_win32.ci : ditto
+
+
+2006-11-10(Fri) 21:29:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix to compile arguments
+
+	* insns.def : fix to duplicate first array value on concatarray
+	instruction
+
+	* yarvtest/test_bin.rb : add a test for above change
+
+	* sample/test.rb : fix to catch up Ruby HEAD (fix to remove test about
+	module duplicate)
+
+
+2006-11-10(Fri) 12:49:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_macro.def : fix to inherit visibility on
+	NODE_SUPER method invocation
+
+
+2006-11-10(Fri) 09:13:46 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c : revert module duplicate inclusion
+
+	* parse.y : catch up current Ruby HEAD
+
+	* node.h : ditto
+
+	* compile.c : ditto
+
+	* gc.c : ditto
+
+	* iseq.c : ditto
+
+	* eval_thread.c : define Continuation (null class)
+
+	* vm_dump.c : fix to output backtrae to stderr
+
+	* yarvtest/test_block.rb : remove unsupported test
+
+	* yarvtest/test_class.rb : add a test about super
+
+	* yarvtest/test_syntax.rb : add a test about case/when
+
+
+2006-11-09(Thu) 10:22:59 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* call_cfunc.h -> call_cfunc.ci : renamed
+
+	* vm_evalbody.h, vm_evalbody.ci : ditto
+
+	* thread_pthread.h, thread_pthread.ci : separate declaration and
+	implementation
+
+	* thread_win32.h, thread_win32.ci : ditto
+
+	* thread.c : use *.ci instead of *.c as implementation
+
+	* vm.c : ditto
+
+	* common.mk : fix rules for above changes
+
+
+2006-11-08(Wed) 17:23:23 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_dump.c : show C level backtrace (pointer only) with
+	backtrace() function (glibc feature)
+
+	* configure.in : ditto
+
+	* yarvcore.c : add NSDR method (show C level backtrace)
+
+	* error.c : fix indent
+
+
+2006-11-07(Tue) 13:17:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (rb_set_errinfo) : added
+
+	* ruby.h : ditto
+
+	* version.h : fix version number
+
+	* lib/webrick/utils.rb : fix to remove Thread.critical
+
+	* ext/dbm, dl, gdbm, iconv, io, pty, sdbm : added
+
+	* test/dbm, gdbm, io, logger, net, readline, sdbm, soap,
+	  webrick, win32ole, wsdl, xsd : added
+
+
+2006-11-06(Mon) 22:32:18 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c : import Ruby HEAD
+
+	* ext/socket/extconf.rb : ditto
+
+	* ext/socket/socket.c : ditto
+
+	* gc.c : ditto
+
+	* lib/date.rb : ditto
+
+	* lib/net/imap.rb : ditto
+
+	* lib/rss/0.9.rb : ditto
+
+	* lib/set.rb : ditto
+
+	* lib/soap/mapping/rubytypeFactory.rb : ditto
+
+	* lib/soap/mimemessage.rb : ditto
+
+	* lib/soap/property.rb : ditto
+
+	* lib/webrick/httprequest.rb : ditto
+
+	* lib/webrick/httputils.rb : ditto
+
+	* lib/xmlrpc/create.rb : ditto
+
+	* lib/xsd/codegen/gensupport.rb : ditto
+
+	* object.c : ditto
+
+	* ruby.h : ditto
+
+	* string.c : ditto
+
+	* version.h : ditto
+
+	* rb/ir.rb : fix to use "diffs" directory
+
+	* vm_dump.c : add "const"
+
+
+2006-11-06(Mon) 16:36:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_proc.c : remove "static" from external global variables
+
+	* eval_thread.c : ditto
+
+	* array.c : fix indent
+
+	* insns.def : add a suitable cast
+
+	* vm_macro.def : allow scalar value on splat arguments
+
+	* yarvtest/test_block.rb : fix to synchronize Ruby HEAD
+
+	* rb/insns2vm.rb : remove String#each for 1.9
+
+	* template/vm.inc.tmpl : ditto (remove String#each_with_index)
+
+
+2006-11-06(Mon) 13:22:34 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* iseq.c : fixed GC debugging outputs
+
+	* rb/parse.rb : fixed output format
+
+
+2006-11-04(Sat) 09:46:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix to duplicate "#{'foo'}" string
+
+	* yarvtest/test_bin.rb : add a test for above
+
+	* ext/readline/readline.c : import Ruby HEAD
+
+	* keywords : ditto
+
+	* lex.c : ditto
+
+	* parse.y : ditto
+
+	* lib/mkmf.rb : ditto
+
+	* test/ruby/test_hash.rb : fix to current specification
+
+	* test/ruby/test_string.rb : ditto
+
+
+2006-11-03(Fri) 20:58:36 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/nkf/nkf-utf8/utf8tbl.h : missed to add
+
+	* configure.in : import ruby HEAD
+
+	* test/ruby/test_array.rb : ditto
+
+	* test/ruby/test_assignment.rb : ditto
+
+	* test/ruby/test_clone.rb : ditto
+
+	* test/socket/test_socket.rb : ditto
+
+	* test/socket/test_unix.rb : ditto
+
+	* test/strscan/test_stringscanner.rb : ditto
+
+	* test/testunit/collector/test_dir.rb : ditto
+
+
+2006-11-03(Fri) 20:22:24 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c : import current ruby HEAD and apply API changes
+	This version has some known bugs
+
+	* bignum.c : ditto
+
+	* blockinlining.c : ditto
+
+	* class.c : ditto
+
+	* compile.c : ditto
+
+	* dir.c : ditto
+
+	* dln.c : ditto
+
+	* enum.c : ditto
+
+	* enumerator.c : ditto
+
+	* error.c : ditto
+
+	* eval.c : ditto
+
+	* eval_error.h : ditto
+
+	* eval_jump.h : ditto
+
+	* eval_load.c : ditto
+
+	* eval_proc.c : ditto
+
+	* ext/*
+
+	* file.c : ditto
+
+	* gc.c : ditto
+
+	* hash.c : ditto
+
+	* insns.def : ditto
+
+	* instruby.rb : ditto
+
+	* intern.h : ditto
+
+	* io.c : ditto
+
+	* iseq.c : ditto
+
+	* lib/*
+
+	* marshal.c : ditto
+
+	* math.c : ditto
+
+	* missing/vsnprintf.c : ditto
+
+	* mkconfig.rb : ditto
+
+	* node.h : ditto
+
+	* numeric.c : ditto
+
+	* object.c : ditto
+
+	* oniguruma.h : ditto
+
+	* pack.c : ditto
+
+	* parse.y : ditto
+
+	* prec.c : ditto
+
+	* process.c : ditto
+
+	* random.c : ditto
+
+	* range.c : ditto
+
+	* rb/ir.rb : ditto
+
+	* re.c : ditto
+
+	* regcomp.c : ditto
+
+	* regerror.c : ditto
+
+	* regexec.c : ditto
+
+	* regint.h : ditto
+
+	* regparse.c : ditto
+
+	* regparse.h : ditto
+
+	* ruby.c : ditto
+
+	* ruby.h : ditto
+
+	* rubytest.rb : ditto
+
+	* runruby.rb : ditto
+
+	* sample/test.rb : ditto
+
+	* signal.c : ditto
+
+	* sprintf.c : ditto
+
+	* st.c : ditto
+
+	* st.h : ditto
+
+	* string.c : ditto
+
+	* struct.c : ditto
+
+	* test/*
+
+	* thread.c : ditto
+
+	* time.c : ditto
+
+	* util.c : ditto
+
+	* variable.c : ditto
+
+	* version.h : ditto
+
+	* vm.c : ditto
+
+	* vm_dump.c : ditto
+
+	* vm_macro.def : ditto
+
+	* win32/*
+
+
+2006-10-31(Tue) 22:47:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y : fix NEWHEAP bugs (import HEAD)
+
+	* ruby.c, intern.h, yarvcore.c (rb_load_file) : change to
+	return parsed node pointer
+
+	* rb/ir.rb : add check mode
+
+
+2006-09-01(Fri) 22:05:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix a bug of peephole optimization and enable
+	regexp optimization
+
+
+2006-08-21(Mon) 05:27:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/mathn.rb : remove "remove_method :gcd2"
+
+	* opt_insn_unif.def : unset opt setting
+
+	* opt_operand.def : ditto
+
+
+2006-08-18(Fri) 17:55:31 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add dependency of yarvcore.h to thread.o
+
+	* gc.c : change comment line
+
+	* thread.c : remove some line break
+
+	* yarvcore.c : reoder initialize sequence to mark main thread
+
+
+2006-08-18(Fri) 16:51:34 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : add a support for cache values per thread
+
+	* yarvcore.c : ditto
+
+	* gc.c : ditto
+
+	* thread.c : move a expression after acquiring lock
+
+	* compile.c : add a cast to remove warning
+
+
+2006-08-18(Fri) 02:07:45 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix to return rhs value on ATTRASGIN
+
+	* insns.def (setn) : add insn setn
+
+	* yarvtest/test_bin.rb : add tests for above
+
+
+2006-08-17(Thu) 22:46:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : clear callee_id ([yarv-dev:1073])
+
+
+2006-08-17(Thu) 22:14:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.h : fix error message
+
+
+2006-08-17(Thu) 12:23:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : change initialize routine order ([yarv-dev:1067])
+
+	* yarvcore.c (Init_yarv) : init th->machine_stack_start
+
+	* thread_pthread.h : add malloc value check ([yarv-dev:1066])
+
+	* insns.def (opt_eq) : fix typo ([yarv-dev:1072])
+
+	* yarvtest/test_opts.rb : add a test for above
+
+	* yarvtest/test_class.rb : add a test for last commit
+
+
+2006-08-17(Thu) 11:02:16 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (clone_method) : check undef-ed method ([yarv-dev:1068])
+
+
+2006-08-15(Tue) 15:07:43 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix opt_plus routine ([yarv-dev-en:149])
+
+	* yarvtest/test_opts.rb : add tests for above
+
+
+2006-08-06(Sun) 06:24:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : fix build rule (build only ruby binary when benchmark)
+
+	* yarvcore.[ch] : fix and add yarv_iseq_new_with_* API
+
+	* blockinlining.c : ditto
+
+	* compile.c : ditto
+
+	* compile.h : ditto
+
+	* iseq.c : ditto
+
+	* eval_method.h : check redefinition for specialized instruction
+
+	* insnhelper.h : ditto
+
+	* insns.def : ditto
+
+	* vm.c : ditto
+
+	* vm.h : ditto
+
+	* numeric.c : add Fixnum#succ
+
+	* thread.c : remove duplicated method Thread#current
+
+	* yarvcore.c : remove duplicated method Proc#clone
+
+	* yarvtest/test_opts.rb : added
+
+
+2006-07-20(Thu) 04:10:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix [yarv-dev:1041] problem (raise TypeError)
+
+	* eval.c : rb_funcall2 send as NOEX_PRIVATE and check scope
+
+
+2006-07-20(Thu) 03:38:46 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : fix [yarv-dev:1040] bug
+
+
+2006-07-18(Tue) 18:45:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* some files : set property "svn:eol-style" as native
+
+
+2006-07-18(Tue) 18:35:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+        * gc.h : fix a static function name
+
+        * vm.c : remove Japanese comments
+
+        * yarvcore.c : add a comment
+
+	* some files : set property "svn:eol-style" as native
+
+
+2006-07-18(Tue) 16:48:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : remove unused code
+
+	* compile.c : add checking value
+
+	* iseq.c : ditto
+
+	* yarvcore.c : fix yarv_th_eval prototype declaration
+
+	* yarvtest/yarvtest.rb : use compile instead of parse method
+
+
+2006-07-12(Wed) 15:18:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarv_version.h : 0.4.1
+
+	* Changes : ditto
+
+
+2006-07-12(Wed) 13:38:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+        * eval.c : fix indent
+
+        * gc.h : fix syntax bug
+
+        * thread_pthread.h : vanish warnning message
+
+        * iseq.c : ditto
+
+        * compile.c : ditto
+
+        * thread.c : ditto
+
+        * vm.c : ditto
+
+        * yarvcore.c : prohibit tail call optimization to mark
+	iseq object
+
+        * yarvcore.h : add some allocator function declaration
+
+        * yarvtest/test_eval.rb : remove output
+
+
+2006-07-12(Wed) 05:01:23 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : undef alloc funcs
+
+	* eval_proc.c : ditto (use factory faction)
+
+	* thread.c : ditto
+
+	* vm.c : ditto
+
+	* iseq.c : fix compile option creation
+
+	* rb/allload.rb : use compile_file method
+
+	* rb/compile.rb : ditto
+
+	* rb/parse.rb : ditto
+
+	* template/insnstbl.html : hide mail addr
+
+
+2006-07-11(Tue) 21:34:29 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_dir.rb: new test test_JVN_13947696.
+
+
+2006-07-11(Tue) 21:26:41 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_alias.rb: new test test_JVN_83768862.
+
+
+2006-07-11(Tue) 11:33:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix compile error on C90
+
+
+2006-07-11(Tue) 10:40:23 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* disasm.c : removed
+
+	* iseq.c : added
+
+	* common.mk : ditto
+
+	* blockinlining.c : Get*Val => Get*Ptr
+
+	* eval.c : ditto
+
+	* yarvcore.c : ditto
+
+	* eval_proc.c : ditto
+
+	* vm_dump.c : ditto
+
+	* vm_macro.def : ditto
+
+	* signal.c : ditto
+
+	* vm.c : ditto
+
+	* thread.c : ditto
+
+	* compile.c : rename local variable insnobj => iobj
+
+	* compile.c : support yarv_compile_option_t
+
+	* gc.h : added
+
+	* insns.def : use OPT_CHECKED_RUN instead of IGNORE_OPTIMIZE
+
+	* rb/compile.rb : use compile option
+
+	* template/optinsn.inc.tmpl : fix function name
+
+	* vm_opts.h.base : change macros
+
+	* rb/insns2vm.rb : ditto
+
+	* yarv.h : fix yarvcore_eval_parsed parameter type
+
+	* yarvcore.c : fix some interfaces (functions)
+
+	* yarvcore.h : add a type yarv_compile_option_t
+
+
+2006-07-06(Thu) 13:45:20 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/yasm.rb : pass builder object if block arity == 1
+
+
+2006-07-05(Wed) 11:23:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/yasm.rb : fix method name
+
+	* vm.c (th_set_top_stack) : check toplevel or not
+
+
+2006-07-04(Tue) 20:05:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/compile.rb : added
+
+	* yarvtest/yarvtest.rb : disable load/store test
+
+
+2006-07-04(Tue) 18:17:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix some bugs about load iseq data
+
+	* disasm.c : ditto (store)
+
+	* eval.c (rb_f_local_variables) : fix bugs
+
+	* insns.def : fix otp_ltlt condition bug
+
+	* vm.c : ditto
+
+	* yarvcore.c : rename some functions
+
+	* yarvtest/yarvtest.rb : add iseq load/store tests
+	(to enable this, remove comment)
+
+
+2006-07-03(Mon) 01:54:23 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_thread.c : add parameter "th" to thread_set_raised
+
+	* yarvcore.h : ditto
+
+	* eval_intern.h : ditto
+
+	* eval.c : ditto
+
+	* eval_error.h : declare with ANSI style
+
+	* disasm.c : rename iseq_iseq2simpledata() to iseq_data_to_ary
+
+	* lib/yasm.rb : rename Instruction#to_simpledata to
+	Instruction#to_a
+
+	* yarvcore.c : ditto
+
+	* vm.c : fix bug (Proc.new{|*args| p args}.call(1) #=> 1)
+
+	* yarvtest/test_proc.rb : add a tests for above
+
+
+2006-06-21(Wed) 09:19:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : remove yarv_iseq_t#catch_table_ary and
+	add yarv_iseq_t#compile_data#catch_table_ary
+
+	* compile.h : ditto
+
+	* yarvcore.c : ditto
+
+	* yarvcore.h : ditto
+
+	* eval_thread.c : remove unused code
+
+	* thread.c : add rb_gc_mark_threads() (from eval_thread.c)
+
+
+2006-05-31(Wed) 21:26:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+        * parse.y : prohibit tail call optimization to mark vparsr
+	object
+
+
+2006-05-25(Thu) 15:37:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c : support NEW_ATTRASGN node
+
+	* class.c : skip undefined method to collect ([yarv-dev:999])
+
+	* yarvtest/test_class.rb : add a test for above
+
+	* compile.c : fix opt_regexpmatch1 condition
+
+	* lib/monitor.rb : fix [yarv-dev:1009]
+
+	* rb/insns2vm.rb : fix typo
+
+	* thread.c : prohibit unlock by not mutex owner thread
+
+	* vm_opts.h.base : change default option
+
+
+2006-05-18(Thu) 16:00:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* intern.h : fix prototype declarations for last re.c change
+
+
+2006-05-18(Thu) 12:12:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/runruby.rb : added
+
+	* thread.c (rb_thread_alone) : check if vm->living_threads
+	is available
+
+
+2006-05-18(Thu) 12:05:35 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* signal.c : not mask SIGSEGV
+
+	* thread.c : fix debug output on Win32
+
+	* thread.c, thread_pthread.h : add some debug prints
+
+	* yarvcore.c : mark machine registers on thread_mark
+
+
+2006-05-17(Wed) 18:09:20 +900  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* dir.c (sys_warning): should not call a vararg function
+	  rb_sys_warning() indirectly.  [ruby-core:07886]
+
+
+2006-05-17(Wed) 16:41:41 +900  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* re.c (rb_reg_initialize): should not allow modifying literal
+	  regexps.  frozen check moved from rb_reg_initialize_m as well.
+
+	* re.c (rb_reg_initialize): should not modify untainted objects in
+	  safe levels higher than 3.
+
+	* re.c (rb_memcmp): type change from char* to const void*.
+
+	* dir.c (dir_close): should not close untainted dir stream.
+
+	* dir.c (GetDIR): add tainted/frozen check for each dir operation.
+
+
+2006-05-07(Sun) 21:06:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c : remove Mutex#unlock_and_stop and add Mutex#sleep
+
+	* lib/monitor.rb : ditto
+
+	* lib/thread.rb : ditto
+
+	* thread_pthread.h : fix stack size
+
+	* thread_win32.h : fix sleep
+
+	* yarvcore.h : disable to use get/setcontext
+
+	* lib/webrick/server.rb : add experimental implementation
+	using thraeds pool
+
+
+2006-05-05(Fri) 13:59:00 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_signal.rb : disable a test
+
+	* thread.c : do trylock before lock on mutex_lock
+
+	* thread_win32.h : use CriticalSection instead of Mutex
+
+
+2006-05-05(Fri) 03:03:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : vtune rule make run test.rb
+
+	* disasm.c : fix syntax errors (on VC)
+
+	* yarvcore.c : ditto
+
+	* lib/thread.rb : Mutex#synchronize is defined here
+
+	* lib/*.rb : ditto
+
+	* signal.c : separate pthread or not
+
+	* thread.c : support lightweight wakeup
+
+	* thread_pthread.h : ditto
+
+	* thread_win32.h : ditto
+
+	* yarvcore.h : ditto
+
+	* yarvtest/test_thread.rb : restore last change
+
+
+2006-05-04(Thu) 18:11:43 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_thread.c : remove rb_thread_interrupt
+
+	* intern.h : ditto
+
+	* signal.c : change signal transfer route
+
+	* thread.c : ditto
+
+	* thread_pthread.h : ditto
+
+	* thread_win32.h : ditto
+
+	* yarv.h : support GET_VM()
+
+	* yarvcore.h : change yarv_thread_t/yarv_vm_t structure
+
+	* yarvtest/test_thread.rb : decrease threads to test
+
+
+2006-05-04(Thu) 00:26:18 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.h : experimental support of thread cache
+
+
+2006-04-25(Tue) 22:30:14 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : remove struct yarv_cmethod_info, add
+	data structure for profiling and extend yarv_control_frame_t
+
+	* vm.c : make pop_frame() and apply above change
+
+	* eval.c : ditto
+
+	* vm_dump.c : ditto
+
+	* vm_macro.def : ditto
+
+	* insns.def (leave): use pop_frame() instead of
+	POP_CONTROL_STACK_FRAME() macro
+
+	* insnhelper.h : remove some macros
+
+	* yarvcore.c : change th_set_top_stack() prototype
+
+
+2006-04-18(Tue) 18:37:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, disasm.c : support export/import exception
+	information
+
+	* yarvcore.h : change "struct catch_table_entry" member variable
+	order
+
+
+2006-04-13(Thu) 17:11:30 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* bignum.c : import ruby 1.9 HEAD (Ruby 1.9.0 2006-04-08)
+
+	* dir.c : ditto
+
+	* enumerator.c : ditto
+
+	* ext/.document : ditto
+
+	* ext/extmk.rb : ditto
+
+	* ext/nkf/lib/kconv.rb : ditto
+
+	* ext/nkf/nkf-utf8/nkf.c : ditto
+
+	* ext/nkf/nkf-utf8/utf8tbl.c : ditto
+
+	* ext/nkf/nkf.c : ditto
+
+	* ext/nkf/test.rb : ditto
+
+	* ext/socket/.cvsignore : ditto
+
+	* ext/win32ole/sample/excel2.rb : ditto
+
+	* ext/win32ole/tests/testOLEMETHOD.rb : ditto
+
+	* ext/win32ole/tests/testOLEPARAM.rb : ditto
+
+	* ext/win32ole/tests/testOLETYPE.rb : ditto
+
+	* ext/win32ole/tests/testOLETYPELIB.rb : ditto
+
+	* ext/win32ole/tests/testOLEVARIABLE.rb : ditto
+
+	* ext/win32ole/tests/testOLEVARIANT.rb : ditto
+
+	* ext/win32ole/tests/testWIN32OLE.rb : ditto
+
+	* ext/win32ole/tests/testall.rb : ditto
+
+	* ext/win32ole/win32ole.c : ditto
+
+	* gc.c : ditto
+
+	* instruby.rb : ditto
+
+	* io.c : ditto
+
+	* lib/delegate.rb : ditto
+
+	* lib/fileutils.rb : ditto
+
+	* lib/find.rb : ditto
+
+	* lib/irb/ruby-lex.rb : ditto
+
+	* lib/mkmf.rb : ditto
+
+	* lib/net/http.rb : ditto
+
+	* lib/open-uri.rb : ditto
+
+	* lib/pathname.rb : ditto
+
+	* lib/rational.rb : ditto
+
+	* lib/rdoc/parsers/parse_rb.rb : ditto
+
+	* lib/rdoc/ri/ri_paths.rb : ditto
+
+	* lib/resolv.rb : ditto
+
+	* lib/test/unit/collector/objectspace.rb : ditto
+
+	* lib/webrick/httpservlet/cgihandler.rb : ditto
+
+	* math.c : ditto
+
+	* mkconfig.rb : ditto
+
+	* object.c : ditto
+
+	* oniguruma.h : ditto
+
+	* pack.c : ditto
+
+	* parse.y : ditto
+
+	* re.c : ditto
+
+	* re.h : ditto
+
+	* regcomp.c : ditto
+
+	* regerror.c : ditto
+
+	* regparse.c : ditto
+
+	* ruby.h : ditto
+
+	* rubytest.rb : ditto
+
+	* runruby.rb : ditto
+
+	* string.c : ditto
+
+	* test/digest/test_digest.rb : ditto
+
+	* test/pathname/test_pathname.rb : ditto
+
+	* test/ruby/envutil.rb : ditto
+
+	* test/ruby/test_float.rb : ditto
+
+	* test/ruby/test_pack.rb : ditto
+
+	* time.c : ditto
+
+	* util.c : ditto
+
+	* version.h : ditto
+
+	* win32/mkexports.rb : ditto
+
+	* win32/resource.rb : ditto
+
+	* win32/win32.c : ditto
+
+
+2006-04-11(Tue) 11:26:53 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/yasm.rb : move to lib/yasm.rb
+
+
+2006-04-09(Sun) 03:04:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : change to accept method iseq object when loading from
+	simple data
+
+	* yarvcore.c : add a debug output
+
+	* rb/yasm.rb : change some interfaces
+
+
+2006-04-07(Fri) 20:25:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix miss about range of catch "next"
+
+	* eval.c : add braces
+
+
+2006-04-07(Fri) 11:09:43 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : fix some make rules
+
+	* insns.def : rename some instructions name
+
+	* rb/insns2vm.rb : change some operand type name
+
+	* vm_evalbody.h : ditto
+
+	* template/insns.inc.tmpl : add YARV_MAX_INSTRUCTION_SIZE macro
+
+	* compile.c, disasm.c, yarvcore.c : support load/store iseq from/to simple
+	data structure such as array, literals, and so on
+
+	* rb/yasm.rb : supported
+
+	* vm.c : change interface of eval_define_method
+
+	* yarvcore.h : remove unused externals
+
+
+2006-03-08(Wed) 10:31:29 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/delegate.rb (DelegateClass): do not delegate #send and
+	  #funcall.
+
+
+2006-02-27(Mon) 22:39:17 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/thread.rb: last commit causes busy loop, revert it. [yarv-dev:990]
+
+	* lib/thread.rb: non_block=true wrongly caused ThreadError.
+
+
+2006-02-27(Mon) 21:33:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : fix to display command line
+
+	* compile.c : fix comparison between a pointer and 0
+
+	* debug.c : fix to output stder
+
+	* disasm.c : add debug function
+
+	* vm_dump.c : ditto
+
+	* eval_proc.c : fix to skip class definition
+
+	* ruby.h : fix T_VALUE to T_VALUES
+
+	* gc.c : ditto
+
+	* node.h : fix prototypes
+
+	* vm.c : add VM_DEBUG macro
+
+	* vm.c : fix compile error on VC++
+
+	* vm.c : fix to inherit last lfp[0] on th_set_finish_env
+
+	* vm.c : fix to add one svar location for any frame
+
+	* vm_macro.def : ditto
+
+	* yarvcore.h : add YARV_CLASS_SPECIAL_P() and YARV_BLOCK_PTR_P()
+
+	* rdoc/ : removed
+
+	* insns.def : fix to propagete throw state
+
+
+2006-02-27(Mon) 13:54:47 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/syslog: imported from Ruby CVS trunk HEAD.
+
+	* ext/racc: ditto.
+
+
+2006-02-27(Mon) 12:47:10 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* parse.y: follow coding style change.
+
+
+2006-02-27(Mon) 11:53:07 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/README: imported from Ruby CVS trunk HEAD.
+
+	* lib/gserver.rb: ditto.
+
+	* lib/readbytes.rb: ditto.
+
+	* lib/parsearg.rb: ditto.
+
+	* lib/racc: ditto.
+
+	* lib/rinda: ditto.
+
+
+2006-02-27(Mon) 11:27:19 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/thread.rb (Queue#pop): faster code. [yarv-dev:973]
+
+	* lib/thread.rb (Queue#pop): avoid to push same thread in to
+	  @waiting.
+
+
+2006-02-23(Thu) 23:32:53 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/open3.rb: imported from Ruby CVS trunk HEAD (rev 1.12).
+
+
+2006-02-23(Thu) 15:10:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : support rb_frame_self()
+
+	* eval_intern.h (th_get_ruby_level_cfp) : return 0 if no cfp
+
+	* eval_load.c : comment out scope set
+
+	* yarvcore.c : fix to initialize/free process of iseq
+
+	* vm.c (th_invoke_proc) : fix to set special cref always
+
+	* yarvtest/test_proc.rb : add a test for above
+
+
+2006-02-22(Wed) 23:33:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add rule "runruby"
+
+	* eval_thread.c : remove obsolete comment
+
+	* eval.c : remove unused functions
+
+	* signal.c : ditto
+
+	* gc.c : add rb_register_mark_object() and use it
+
+	* eval_load.c : ditto
+
+	* eval_proc.c : ditto
+
+	* ext/etc/etc.c : ditto
+
+	* ext/win32ole/win32ole.c : ditto
+
+	* ruby.h : ditto
+
+	* yarvcore.h : ditto
+
+	* thread.c : add rb_thread_run_parallel()
+
+	* yarvcore.c : change bootstrap
+
+
+2006-02-22(Wed) 19:27:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* ext/win32ole/.cvsignore : removed
+
+	* ext/win32ole/.document : ditto
+
+
+2006-02-22(Wed) 18:17:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : set Binding as YARVCore::VM::Binding
+
+
+2006-02-22(Wed) 12:54:45 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* ChangeLog : remove needless line
+
+
+2006-02-22(Wed) 12:49:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rubysig.h : remove CHECK_INTS
+
+	* eval.c : ditto
+
+	* eval_load.c : ditto
+
+	* ext/readline/readline.c : ditto
+
+	* thread.c : ditto
+
+	* win32/win32.c : ditto
+
+	* yarv_version.h : 0.4.0
+
+	* Changes : ditto
+
+
+2006-02-22(Wed) 11:36:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test.rb : removed
+
+
+2006-02-22(Wed) 11:12:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* README : renewed
+
+	* version.c : fixed version message
+
+	* yarvext/ : removed
+
+
+2006-02-22(Wed) 10:33:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/.document : imported from Ruby 1.9 HEAD
+
+	* .document : ditto
+
+	* ext/.document : ditto
+
+	* lib/ftools.rb : ditto
+
+	* lib/rdoc/ : ditto
+
+	* eval_thread.c : remove unused functions
+
+	* process.c : ditto
+
+	* rb/insns2vm.rb : compare modified date of vm_opts.h and
+	vm_opts.h.base
+
+	* ruby.h : rename RValue to RValues
+
+	* gc.c : ditto
+
+	* vm.c : ditto
+
+
+2006-02-22(Wed) 06:32:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* configure.in : remove last commit
+
+
+2006-02-22(Wed) 06:18:53 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* configure.in : add default program prefix "-yarv"
+
+
+2006-02-22(Wed) 06:11:36 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : change default rule (same as HEAD)
+
+	* configure : removed
+
+	* eval.c : remove last commit
+
+	* vm.c : fix stack traverse
+
+	* yarvcore.c : initialize top of control frame
+
+	* version.c : 2.0
+
+	* version.h : ditto
+
+
+2006-02-22(Wed) 04:50:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : change to rewind C level control frame
+
+	* vm.c : change to initialize cfp#proc and fix comparison of
+	cfp and limit_cfp
+
+	* yarvcore.c : remove last commit
+
+
+2006-02-22(Wed) 03:25:56 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : initialize each stack of thread
+
+
+2006-02-22(Wed) 00:02:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c : fix synchornize return value ([yarv-dev:957])
+	and some synchornization error
+
+	* thread_pthread.h : add debug helper function
+
+
+2006-02-21(Tue) 20:54:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : fix place of rb_thread_terminate_all()
+
+	* eval_thread.c : remove unused functions
+
+	* yarv.h : remove GET_VM()
+
+	* eval_jump.h : ditto
+
+	* insns.def : ditto
+
+	* vm_dump.c :
+
+	* intern.h : change rb_thread_signal_raise/exit interface
+
+	* signal.c : ditto
+
+	* thread.c : ditto
+
+	* test/ruby/test_beginendblock.rb : use block with IO.popen
+
+	* thread_pthread.h : fix interrupt process
+
+	* thread_win32.h : ditto
+
+	* yarvcore.c : fix thread free process
+
+	* yarvcore.h : remove yarv_vm_t#thread_critical, etc
+
+
+2006-02-21(Tue) 12:42:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_thread.c : remove unused function rb_thread_schedule()
+
+	* thread.c : rename yarv_thread_schedule to rb_thread_schedule()
+
+	* thread.c, eval.c : fix to terminate all thread and barrier at
+	eval.c#ruby_cleanup()
+
+	* thread_win32.h : remove native_thread_cleanup()
+
+	* thread_pthread.h : ditto
+
+	* yarvcore.c : ditto
+
+	* yarvtest/test_thread.rb : separete assersions to tests
+
+
+2006-02-21(Tue) 02:13:33 +900  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* parse.y (f_arglist): should set command_start = Qtrue for
+	  command body.  [ruby-talk:180648]
+
+
+2006-02-20(Mon) 20:41:07 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c : fix to synchronize signal_thread_list access
+	and fix typo
+
+
+2006-02-20(Mon) 17:54:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_proc.c : remove unused Binding functions and
+	set is_lambda of Proc used define_method
+
+	* yarvcore.c : support Proc#dup/clone, Binding#dup/clone
+
+	* sample/test.rb : remove unsupport features (Proc as Binding)
+
+
+2006-02-20(Mon) 16:28:59 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add a dependency to vm.c on eval_intern.h
+
+	* eval_intern.h : fix to initialize tag->tag
+
+	* yarvtest/test_jump.rb : add tests for above
+
+	* eval_jump.h : use local variable
+
+
+2006-02-20(Mon) 15:13:24 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm3_thread_create_join.rb : added
+
+	* test/yaml/test_yaml.rb : imported from Ruby CVS trunk HEAD
+
+
+2006-02-20(Mon) 14:49:46 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/yaml.rb: imported from Ruby CVS trunk HEAD.
+
+	* lib/yaml: ditto.
+
+	* ext/syck: ditto.
+
+
+2006-02-20(Mon) 13:58:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support block parameter which is NODE_ATTRASGN
+
+	* yarvtest/test_block.rb : add tests for above
+
+	* compile.c : fix NODE_DASGN_CURR level check
+
+	* compile.c : fix "||=" (at firtst, check "defined? val")
+
+	* compile.c : fix NODE_MATCH3 (permute receiver and argument)
+
+	* yarvtest/test_bin.rb : add tests for above
+
+	* eval.c : add rb_each()
+
+	* test/ruby/test_signal.rb : increment a timeout value
+
+	* thread.c, yarvcore.h : fix "join" flow
+
+	* thread_pthread.h : ditto
+
+	* thread_win32.h : ditto
+
+	* yarvtest/test_thread.rb : add a test for above
+
+	* vm.h, vm.c, vm_dump.c, insns.def : add FRAME_MAGIC_LAMBDA and
+	support return from lambda (especially retrun from method defined
+	by "define_method")
+
+	* yarvtest/test_method.rb : add a test for above
+
+	* yarvcore.c : remove unused functions
+
+
+2006-02-20(Mon) 11:22:31 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_eval.rb: now Object#funcall is defined.
+
+
+2006-02-20(Mon) 11:04:32 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/irb/lc/ja/CVS: removed.
+
+
+2006-02-20(Mon) 10:55:59 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/mutex_m.rb: imported from Ruby CVS trunk HEAD.
+
+	* lib/observer.rb: ditto.
+
+	* lib/wsdl: ditto.
+
+	* lib/monitor.rb: ditto (removing Thread.critical=).
+
+	* lib/xsd: ditto.
+
+	* lib/soap: ditto.
+
+	* lib/drb.rb: ditto.
+
+	* lib/drb: ditto.
+
+
+2006-02-20(Mon) 10:49:31 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* yarvcore.c (Init_yarvcore): fix typo (duo -> dup).
+
+
+2006-02-19(Sun) 01:27:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : "return" from lambda{} break block
+
+	* eval.c : Unsupport Proc as Binding
+
+	* test/ruby/test_eval.rb : apply above changes
+
+	* yarvcore.c : remove unused function yarv_yield_values()
+
+
+2006-02-18(Sat) 03:19:36 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, insns.def : fix passing value when thread killed
+
+	* yarvtest/test_thread.rb : add tests for above
+
+
+2006-02-19(Sun) 01:19:42 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/thread.rb (SizedQueue): didn't work.  This patch was
+	  contributed by yukimizake. [yarv-dev:916]
+
+
+2006-02-18(Sat) 03:19:36 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, insns.def : fix passing value when thread killed
+
+	* yarvtest/test_thread.rb : add tests for above
+
+
+2006-02-18(Sat) 02:40:18 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm.c, vm_macro.def : change BMETHOD algorithm
+	([yarv-dev:914])
+
+	* yarvtest/test_class.rb : add a test for above
+
+
+2006-02-17(Fri) 23:59:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, yarv.h : change th_invoke_proc() interface
+
+	* eval_proc.c : ditto
+
+	* signal.c : ditto
+
+	* thread.c : ditto
+
+	* yarvcore.c : ditto
+
+	* vm_macro.def : ditto and fix NODE_BMETHOD call
+
+	* vm.c : change name ("th_set_env()" to "push_frame()") and
+	change interface
+
+	* insns.def : ditto
+
+	* eval.c : remove proc_jump_error()
+
+	* benchmark/bm_app_answer.rb : added
+
+	* vm_opts.h.base : add optimize option
+
+
+2006-02-17(Fri) 13:37:57 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, ruby.h : add rb_errinfo()
+
+	* eval_error.h (error_pos) : fix process order
+
+	* bin/erb : imported from ruby 1.9
+
+	* bin/irb : ditto
+
+	* bin/rdoc : ditto
+
+	* bin/ri : ditto
+
+	* bin/testrb : ditto
+
+	* ext/curses/.cvsignore : ditto
+
+	* ext/curses/curses.c : ditto
+
+	* ext/curses/depend : ditto
+
+	* ext/curses/extconf.rb : ditto
+
+	* ext/curses/hello.rb : ditto
+
+	* ext/curses/mouse.rb : ditto
+
+	* ext/curses/rain.rb : ditto
+
+	* ext/curses/view.rb : ditto
+
+	* ext/curses/view2.rb : ditto
+
+	* ext/fcntl/.cvsignore : ditto
+
+	* ext/fcntl/depend : ditto
+
+	* ext/fcntl/extconf.rb : ditto
+
+	* ext/fcntl/fcntl.c : ditto
+
+	* ext/readline/README : ditto
+
+	* ext/readline/README.ja : ditto
+
+	* ext/readline/depend : ditto
+
+	* ext/readline/extconf.rb : ditto
+
+	* ext/readline/readline.c : ditto
+
+	* ext/win32ole/.document : ditto
+
+	* ext/zlib/doc/zlib.rd : ditto
+
+	* ext/zlib/extconf.rb : ditto
+
+	* ext/zlib/zlib.c : ditto
+
+	* lib/cgi/.document : ditto
+
+	* lib/cgi/session.rb : ditto
+
+	* lib/cgi/session/pstore.rb : ditto
+
+	* lib/shell/builtin-command.rb : ditto
+
+	* lib/shell/command-processor.rb : ditto
+
+	* lib/shell/error.rb : ditto
+
+	* lib/shell/filter.rb : ditto
+
+	* lib/shell/process-controller.rb : ditto
+
+	* lib/shell/system-command.rb : ditto
+
+	* lib/shell/version.rb : ditto
+
+	* lib/xmlrpc/.document : ditto
+
+	* lib/xmlrpc/README.rdoc : ditto
+
+	* lib/xmlrpc/README.txt : ditto
+
+	* lib/xmlrpc/base64.rb : ditto
+
+	* lib/xmlrpc/client.rb : ditto
+
+	* lib/xmlrpc/config.rb : ditto
+
+	* lib/xmlrpc/create.rb : ditto
+
+	* lib/xmlrpc/datetime.rb : ditto
+
+	* lib/xmlrpc/httpserver.rb : ditto
+
+	* lib/xmlrpc/marshal.rb : ditto
+
+	* lib/xmlrpc/parser.rb : ditto
+
+	* lib/xmlrpc/server.rb : ditto
+
+	* lib/xmlrpc/utils.rb : ditto
+
+	* rdoc/README : ditto
+
+	* rdoc/code_objects.rb : ditto
+
+	* rdoc/diagram.rb : ditto
+
+	* rdoc/dot/dot.rb : ditto
+
+	* rdoc/generators/chm_generator.rb : ditto
+
+	* rdoc/generators/html_generator.rb : ditto
+
+	* rdoc/generators/ri_generator.rb : ditto
+
+	* rdoc/generators/template/chm/chm.rb : ditto
+
+	* rdoc/generators/template/html/hefss.rb : ditto
+
+	* rdoc/generators/template/html/html.rb : ditto
+
+	* rdoc/generators/template/html/kilmer.rb : ditto
+
+	* rdoc/generators/template/html/old_html.rb : ditto
+
+	* rdoc/generators/template/html/one_page_html.rb : ditto
+
+	* rdoc/generators/template/xml/rdf.rb : ditto
+
+	* rdoc/generators/template/xml/xml.rb : ditto
+
+	* rdoc/generators/xml_generator.rb : ditto
+
+	* rdoc/markup/sample/rdoc2latex.rb : ditto
+
+	* rdoc/markup/sample/sample.rb : ditto
+
+	* rdoc/markup/simple_markup.rb : ditto
+
+	* rdoc/markup/simple_markup/fragments.rb : ditto
+
+	* rdoc/markup/simple_markup/inline.rb : ditto
+
+	* rdoc/markup/simple_markup/lines.rb : ditto
+
+	* rdoc/markup/simple_markup/preprocess.rb : ditto
+
+	* rdoc/markup/simple_markup/to_flow.rb : ditto
+
+	* rdoc/markup/simple_markup/to_html.rb : ditto
+
+	* rdoc/markup/simple_markup/to_latex.rb : ditto
+
+	* rdoc/markup/test/AllTests.rb : ditto
+
+	* rdoc/markup/test/TestInline.rb : ditto
+
+	* rdoc/markup/test/TestParse.rb : ditto
+
+	* rdoc/options.rb : ditto
+
+	* rdoc/parsers/parse_c.rb : ditto
+
+	* rdoc/parsers/parse_f95.rb : ditto
+
+	* rdoc/parsers/parse_rb.rb : ditto
+
+	* rdoc/parsers/parse_simple.rb : ditto
+
+	* rdoc/parsers/parserfactory.rb : ditto
+
+	* rdoc/rdoc.rb : ditto
+
+	* rdoc/ri/ri_cache.rb : ditto
+
+	* rdoc/ri/ri_descriptions.rb : ditto
+
+	* rdoc/ri/ri_display.rb : ditto
+
+	* rdoc/ri/ri_driver.rb : ditto
+
+	* rdoc/ri/ri_formatter.rb : ditto
+
+	* rdoc/ri/ri_options.rb : ditto
+
+	* rdoc/ri/ri_paths.rb : ditto
+
+	* rdoc/ri/ri_reader.rb : ditto
+
+	* rdoc/ri/ri_util.rb : ditto
+
+	* rdoc/ri/ri_writer.rb : ditto
+
+	* rdoc/template.rb : ditto
+
+	* rdoc/tokenstream.rb : ditto
+
+	* rdoc/usage.rb : ditto
+
+	* test/xmlrpc/data/bug_bool.expected : ditto
+
+	* test/xmlrpc/data/bug_bool.xml : ditto
+
+	* test/xmlrpc/data/bug_cdata.expected : ditto
+
+	* test/xmlrpc/data/bug_cdata.xml : ditto
+
+	* test/xmlrpc/data/bug_covert.expected : ditto
+
+	* test/xmlrpc/data/bug_covert.xml : ditto
+
+	* test/xmlrpc/data/datetime_iso8601.xml : ditto
+
+	* test/xmlrpc/data/fault.xml : ditto
+
+	* test/xmlrpc/data/value.expected : ditto
+
+	* test/xmlrpc/data/value.xml : ditto
+
+	* test/xmlrpc/data/xml1.expected : ditto
+
+	* test/xmlrpc/data/xml1.xml : ditto
+
+	* test/xmlrpc/test_datetime.rb : ditto
+
+	* test/xmlrpc/test_features.rb : ditto
+
+	* test/xmlrpc/test_marshal.rb : ditto
+
+	* test/xmlrpc/test_parser.rb : ditto
+
+	* test/xmlrpc/test_webrick_server.rb : ditto
+
+	* test/xmlrpc/webrick_testing.rb : ditto
+
+	* test/zlib/test_zlib.rb : ditto
+
+
+2006-02-17(Fri) 09:41:35 +900  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* thread.c (sleep_timeval): sleep should always sleep for
+	  specified amount of time.  [ruby-talk:180067]
+
+
+2006-02-17(Fri) 02:20:32 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_safe.h, ruby.h : remove ruby_safe_level and add
+	rb_safe_level() and rb_set_safe_level_force()
+
+	* eval.c : use above functions
+
+	* eval_jump.h : ditto
+
+	* eval_load.c : ditto
+
+	* eval_method.h : ditto
+
+	* eval_proc.c : ditto
+
+	* eval_thread.c : ditto
+
+	* gc.c : ditto
+
+	* signal.c : ditto
+
+	* variable.c : ditto
+
+	* ext/win32ole/win32ole.c : ditto
+
+	* vm.c (th_invoke_proc) : save and restore safe level
+
+	* yarvtest/test_proc.rb : add tests for above
+
+	* thread.c : remove unused functions
+
+
+2006-02-17(Fri) 01:08:23 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : remove a setspecial second unused operand
+
+	* eval_load.c : remove unused variable th
+
+	* eval_proc.c, yarvcore.c : remove some functions from eval_proc.c
+	and move to yarvcore.c
+
+	* insns.def : fix to delete warnings
+
+	* sample/test.rb : comment out Proc#clone tests
+
+	* version.c : add constant RUBY_VM_DATE
+
+	* vm.c : fix some functions
+
+
+2006-02-16(Thu) 22:58:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm.c : use th_yield_setup_args at yield and Proc#call
+
+
+2006-02-16(Thu) 19:51:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix analysis of block parameter
+
+	* disasm.c : remove rb_bug() (temporarily)
+
+	* insns.def, vm.c : fix passing block parameter
+
+	* sample/test.rb : add "Proc = YARVCore::VM::Proc"
+
+	* test/ruby/test_readpartial.rb : disable on mswin32
+
+	* test/socket/test_tcp.rb : ditto
+
+	* thread.c : fix syntax error (for non GCC)
+
+
+2006-02-15(Wed) 22:34:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.h : move rb_clear_cache_by_id position
+
+	* thread.c : fix Thread#kill
+
+	* test/ruby/test_readpartial.rb : enable tests except cygwin version
+
+	* test/ruby/test_signal.rb : ditto and enable timeout
+
+
+2006-02-15(Wed) 22:13:29 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/runit: forgot to commit.
+
+
+2006-02-15(Wed) 22:12:25 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/weakref.rb: do not use Thread.critical=.
+
+	* lib/singleton.rb: ditto.
+
+	* lib/timeout.rb: ditto.
+
+	* lib/thread.rb: ditto.
+
+	* test/inlinetest.rb: forgot to commit.
+
+
+2006-02-15(Wed) 21:34:17 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/test_pp.rb: imported from Ruby CVS trunk HEAD.
+
+	* test/test_shellwords.rb: ditto.
+
+	* test/test_set.rb: ditto.
+
+	* test/test_time.rb: ditto.
+
+	* test/test_ipaddr.rb: ditto.
+
+	* test/test_prettyprint.rb: ditto.
+
+	* test/test_tsort.rb: ditto.
+
+	* test/strscan: ditto.
+
+	* test/testunit: ditto.
+
+
+2006-02-15(Wed) 20:03:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.h : duplicate NODE_METHOD at make an alias
+
+	* yarvtest/test_method.rb : add a test for above
+
+
+2006-02-15(Wed) 19:48:59 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/rss: imported from Ruby CVS trunk HEAD.
+
+
+2006-02-15(Wed) 19:47:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c, vm.c : remove methoddef, singletonmethoddef
+	instructions and make new insn definemethod
+
+	* yarvcore.c : set toplevel visibility to private
+
+
+2006-02-15(Wed) 17:39:16 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h :
+
+	* eval_jump.h, vm.c : localjump_error() and jump_tag_but_local_jump()
+	move to th_localjump_error and th_jump_tag_but_local_jump at vm.c
+
+	* eval.c : ditto
+
+	* eval_load.c : ditto
+
+	* insns.def : ditto
+
+	* vm.c : ditto
+
+	* vm.c (th_make_jump_tag_but_local_jump) : added
+
+	* opt_insn_unif.def : fix indnet (revert change)
+
+	* opt_operand.def : ditto
+
+	* rb/insns2vm.rb : fix error message
+
+	* thread.c : raise exception at join if illegal local jump
+
+
+2006-02-15(Wed) 14:21:45 +900  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* ChangeLog: add local variables line to support Emacs.
+
+	* eval.c (rb_obj_instance_exec): add new method from 1.9.
+
+	* eval.c (rb_mod_module_exec): ditto.
+
+	* eval.c (yield_under_i): should not pass self as an argument to
+	  the block for instance_eval.  [ruby-core:07364]
+
+	* eval.c (rb_obj_instance_eval): should be no singleton classes for
+	  true, false, and nil.  [ruby-dev:28186]
+
+
+2006-02-14(Tue) 19:30:20 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c : fix indent
+
+	* eval.c : fix block_given
+
+	* gc.c : add STACK_START and use it as a substitute for
+	rb_gc_stack_start
+
+	* vm.c : fix to raise error if th_yield doesn't have block given
+
+	* yarvcore.c : fix to skip iseq mark array at ObjectSpace.each_object
+
+
+2006-02-14(Tue) 18:15:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* configure.in : enable pthread by default
+
+	* ascii.c : import ruby 1.9 HEAD
+
+	* bignum.c : ditto
+
+	* compar.c : ditto
+
+	* configure : ditto
+
+	* defines.h : ditto
+
+	* dln.c : ditto
+
+	* dln.h : ditto
+
+	* enum.c : ditto
+
+	* enumerator.c : ditto
+
+	* euc_jp.c : ditto
+
+	* ext/win32ole/tests/testWIN32OLE.rb : ditto
+
+	* ext/win32ole/win32ole.c : ditto
+
+	* file.c : ditto
+
+	* hash.c : ditto
+
+	* io.c : ditto
+
+	* lex.c : ditto
+
+	* lib/irb/init.rb : ditto
+
+	* lib/rexml/document.rb : ditto
+
+	* main.c : ditto
+
+	* marshal.c : ditto
+
+	* math.c : ditto
+
+	* missing.h : ditto
+
+	* object.c : ditto
+
+	* oniguruma.h : ditto
+
+	* pack.c : ditto
+
+	* process.c : ditto
+
+	* random.c : ditto
+
+	* range.c : ditto
+
+	* rb/ir.rb : ditto
+
+	* re.c : ditto
+
+	* regcomp.c : ditto
+
+	* regenc.c : ditto
+
+	* regenc.h : ditto
+
+	* regerror.c : ditto
+
+	* regexec.c : ditto
+
+	* regint.h : ditto
+
+	* regparse.c : ditto
+
+	* regparse.h : ditto
+
+	* ruby.c : ditto
+
+	* ruby.h : ditto
+
+	* rubyio.h : ditto
+
+	* sjis.c : ditto
+
+	* sprintf.c : ditto
+
+	* st.c : ditto
+
+	* st.h : ditto
+
+	* struct.c : ditto
+
+	* test/ruby/envutil.rb : ditto
+
+	* test/ruby/test_struct.rb : ditto
+
+	* time.c : ditto
+
+	* utf8.c : ditto
+
+	* util.c : ditto
+
+	* util.h : ditto
+
+	* version.h : ditto
+
+	* win32/Makefile.sub : ditto
+
+	* win32/win32.c : ditto
+
+
+2006-02-14(Tue) 16:40:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, eval_proc.c : fix rb_proc_arity
+
+	* eval.c : declare funcall same as send (temporarily)
+
+	* lib/thread.rb : added
+
+	* test/pathname/test_pathname.rb : imported from ruby 1.9
+
+	* test/scanf/data.txt : ditto
+
+	* test/scanf/test_scanf.rb : ditto
+
+	* test/scanf/test_scanfblocks.rb : ditto
+
+	* test/scanf/test_scanfio.rb : ditto
+
+	* test/socket/test_socket.rb : ditto
+
+	* test/socket/test_tcp.rb : ditto
+
+	* test/socket/test_udp.rb : ditto
+
+	* test/socket/test_unix.rb : ditto
+
+	* test/stringio/test_stringio.rb : ditto
+
+	* test/uri/test_common.rb : ditto
+
+	* test/uri/test_ftp.rb : ditto
+
+	* test/uri/test_generic.rb : ditto
+
+	* test/uri/test_http.rb : ditto
+
+	* test/uri/test_ldap.rb : ditto
+
+	* test/uri/test_mailto.rb : ditto
+
+
+2006-02-14(Tue) 15:59:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c : Change Thread.critical warnning message
+
+	* lib/webrick.rb : imported from ruby 1.9
+
+	* lib/webrick/accesslog.rb : ditto
+
+	* lib/webrick/cgi.rb : ditto
+
+	* lib/webrick/compat.rb : ditto
+
+	* lib/webrick/config.rb : ditto
+
+	* lib/webrick/cookie.rb : ditto
+
+	* lib/webrick/htmlutils.rb : ditto
+
+	* lib/webrick/httpauth.rb : ditto
+
+	* lib/webrick/httpauth/authenticator.rb : ditto
+
+	* lib/webrick/httpauth/basicauth.rb : ditto
+
+	* lib/webrick/httpauth/digestauth.rb : ditto
+
+	* lib/webrick/httpauth/htdigest.rb : ditto
+
+	* lib/webrick/httpauth/htgroup.rb : ditto
+
+	* lib/webrick/httpauth/htpasswd.rb : ditto
+
+	* lib/webrick/httpauth/userdb.rb : ditto
+
+	* lib/webrick/httpproxy.rb : ditto
+
+	* lib/webrick/httprequest.rb : ditto
+
+	* lib/webrick/httpresponse.rb : ditto
+
+	* lib/webrick/https.rb : ditto
+
+	* lib/webrick/httpserver.rb : ditto
+
+	* lib/webrick/httpservlet.rb : ditto
+
+	* lib/webrick/httpservlet/abstract.rb : ditto
+
+	* lib/webrick/httpservlet/cgi_runner.rb : ditto
+
+	* lib/webrick/httpservlet/cgihandler.rb : ditto
+
+	* lib/webrick/httpservlet/erbhandler.rb : ditto
+
+	* lib/webrick/httpservlet/filehandler.rb : ditto
+
+	* lib/webrick/httpservlet/prochandler.rb : ditto
+
+	* lib/webrick/httpstatus.rb : ditto
+
+	* lib/webrick/httputils.rb : ditto
+
+	* lib/webrick/httpversion.rb : ditto
+
+	* lib/webrick/log.rb : ditto
+
+	* lib/webrick/server.rb : ditto
+
+	* lib/webrick/ssl.rb : ditto
+
+	* lib/webrick/utils.rb : ditto
+
+	* lib/webrick/version.rb : ditto
+
+
+2006-02-14(Tue) 14:55:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : support "defined?($1)", ...
+
+	* yarvtest/test_syntax.rb : add a test for above
+
+	* rb/makedocs.rb : fix template directory path
+
+	* vm.c : fix to handle break from proc
+
+
+2006-02-14(Tue) 12:42:59 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : fix rb_iterate hook
+
+	* yarvtest/test_block.rb : add a tests for above
+
+	* vm.c : remove unused comment
+
+
+2006-02-14(Tue) 12:01:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : fix to check passed block at block_given_p
+
+	* eval_proc.c : fix to pass block at Method#call
+
+	* runruby.rb : fix to apply ruby
+
+	* test/runner.rb : GC.stress (comment out)
+
+	* vm.c : fix indnet
+
+
+2006-02-14(Tue) 08:04:33 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/tempfile.rb: use Mutex instead of Thread.critical.
+
+	* lib/rss/dublincore.rb: |x,| -> |x,_| to avoid YARV bug (tmp).
+
+	* lib/rexml: imported from ruby CVS trunk HEAD.
+
+	* test/digest: ditto.
+
+	* test/fileutils: ditto.
+
+	* test/ostruct: ditto.
+
+	* test/erb: ditto.
+
+	* test/optparse: ditto.
+
+	* test/ruby/test_signal.rb: turn off a test to avoid unknown error
+	  (tmp).
+
+
+2006-02-14(Tue) 07:52:03 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/digest: imported from ruby CVS trunk HEAD.
+
+	* test/fileutils: ditto.
+
+	* test/ostruct: ditto.
+
+	* test/erb: ditto.
+
+	* test/optparse: ditto.
+
+
+2006-02-14(Tue) 06:26:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, parse.y : support BEGIN{} (remove local scope)
+
+	* test/ruby/beginmainend.rb : fix to apply YARV's specification
+
+	* test/ruby/test_beginendblock.rb : enable BEGIN{} test
+
+	* signal.c : exit at double segv
+
+	* insns.def (preexe) : remove instruction "preexe"
+
+
+2006-02-14(Tue) 05:53:56 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* eval.c (ruby_cleanup): th->errinfo contains a NODE while
+	  break'ing, check it before referring klass.
+
+
+2006-02-14(Tue) 05:45:07 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix stack calc of send
+
+	* sample/test.rb : remove SEGV causing code
+
+
+2006-02-14(Tue) 02:24:21 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_module.rb: list order is not a matter.
+
+	* test/csv: imported from ruby CVS trunk HEAD.
+
+
+2006-02-14(Tue) 02:06:25 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_beginendblock.rb: unlock tests.
+
+	* test/ruby/beginmainend.rb: new file (imported from ruby CVS
+	  trunk HEAD).
+
+	* test/ruby/endblockwarn.rb: new file (imported from ruby CVS
+	  trunk HEAD).
+
+	* test/ruby/test_file.rb: new file (imported from ruby CVS trunk
+	  HEAD).
+
+
+2006-02-14(Tue) 01:42:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* error.c : fix include file position
+
+	* test/ruby/test_signal.rb : skip test_exit_action on cygwin
+
+
+2006-02-14(Tue) 01:36:57 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_class.rb: new file (imported from rubicon).
+
+
+2006-02-14(Tue) 01:32:23 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_module.rb: ignore PP mixins.
+
+
+2006-02-14(Tue) 01:24:56 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_lambda.rb: removed (->(){...} syntax is
+	  obsolete).
+
+
+2006-02-14(Tue) 01:20:54 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_module.rb: import many tests from rubicon.
+
+
+2006-02-14(Tue) 01:06:57 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix to avoid stack consistency error
+
+	* yarvtest/test_exception.rb : add a test for above
+
+
+2006-02-14(Tue) 00:42:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, vm_macro.def : rename VM_CALL_SUPER to VM_CALL_SUPER_BIT
+
+	* insns.def (send) : set a flag of super as fcall
+
+	* yarvtest/test_class.rb : add a test for above
+
+
+2006-02-14(Tue) 00:31:24 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_eval.rb: fix typo.
+
+	* test/ruby/test_signal.rb: unlock tests.
+
+
+2006-02-13(Mon) 23:53:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm_macro.def : fix NODE_ZSUPER dispatch and
+	fix error message when super without suitable method ([yarv-dev:846])
+
+	* yarvcore.h : add VM_CALL_SUPER definition
+
+	* yarvtest/test_method.rb : add a test of Module#private_class_method
+
+
+2006-02-13(Mon) 22:49:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : traverse all iseq to find super method ([yarv-dev:859])
+
+	* yarvtest/test_class.rb : add a test for above
+
+	* yarvcore.c : add clear iseq->defined_method_id
+
+	* signal.c : fix to prohibit double segv handler kicked
+
+
+2006-02-13(Mon) 22:09:12 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_DECL, NODE_CLASS with NODE_CLON3 prefix
+
+	* yarvtest/test_class.rb : add tests for above
+
+
+2006-02-13(Mon) 21:20:57 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix indent
+
+	* compile.c : fix to prohibit "redo" from eval expression
+
+
+2006-02-13(Mon) 20:36:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : fix constant search bug ([yarv-dev:788])
+
+	* yarvtest/test_class.rb : add a test of [yarv-dev:788]
+
+
+2006-02-13(Mon) 18:09:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/ruby/test_clone.rb : enable tests with Class#clone
+
+	* test/ruby/test_marshal.rb : ditto
+
+
+2006-02-13(Mon) 17:42:37 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c : support Class#clone
+
+	* compile.c, insns.def : remove popcref
+
+	* yarvcore.h, vm.c, insns.def : remove yarv_thread_t#cref_stack
+
+	* eval.c, eval_intern.h, eval_load.c : ditto
+
+	* yarvtest/test_class.rb : add tests for singleton class
+
+	* gc.c : remove "FRAME *" unused variable
+
+	* insnhelper.h : fix COPY_CREF
+
+	* rb/mklog.rb : add default message
+
+	* vm_macro.def : support NODE_ZSUPER as method type
+
+
+2006-02-13(Mon) 00:11:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c : refoctoring with CFLAGS+=-Wunused
+
+	* eval.c : ditto
+
+	* eval_intern.h : ditto
+
+	* eval_load.c : ditto
+
+	* eval_method.h : ditto
+
+	* eval_proc.c : ditto
+
+	* eval_thread.c : ditto
+
+	* insns.def : ditto
+
+	* parse.y : ditto
+
+	* thread.c : ditto
+
+	* vm.c : ditto
+
+
+2006-02-13(Mon) 02:32:34 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_const.rb: show better message.
+
+	* test/ruby/test_eval.rb: ditto.
+
+	* test/ruby/test_module.rb: new file.
+
+
+2006-02-12(Sun) 22:22:35 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c : revert last commit
+
+	* ascii.c : ditto
+
+	* bignum.c : ditto
+
+	* class.c : ditto
+
+	* compar.c : ditto
+
+	* defines.h : ditto
+
+	* dir.c : ditto
+
+	* dln.c : ditto
+
+	* dln.h : ditto
+
+	* enum.c : ditto
+
+	* enumerator.c : ditto
+
+	* error.c : ditto
+
+	* euc_jp.c : ditto
+
+	* file.c : ditto
+
+	* gc.c : ditto
+
+	* hash.c : ditto
+
+	* intern.h : ditto
+
+	* io.c : ditto
+
+	* lex.c : ditto
+
+	* main.c : ditto
+
+	* marshal.c : ditto
+
+	* math.c : ditto
+
+	* missing.h : ditto
+
+	* node.h : ditto
+
+	* numeric.c : ditto
+
+	* object.c : ditto
+
+	* oniguruma.h : ditto
+
+	* pack.c : ditto
+
+	* prec.c : ditto
+
+	* process.c : ditto
+
+	* random.c : ditto
+
+	* range.c : ditto
+
+	* rb/mklog.rb : ditto
+
+	* re.c : ditto
+
+	* regcomp.c : ditto
+
+	* regenc.c : ditto
+
+	* regenc.h : ditto
+
+	* regerror.c : ditto
+
+	* regex.h : ditto
+
+	* regexec.c : ditto
+
+	* regint.h : ditto
+
+	* regparse.c : ditto
+
+	* regparse.h : ditto
+
+	* ruby.c : ditto
+
+	* ruby.h : ditto
+
+	* rubyio.h : ditto
+
+	* rubysig.h : ditto
+
+	* signal.c : ditto
+
+	* sjis.c : ditto
+
+	* sprintf.c : ditto
+
+	* st.c : ditto
+
+	* st.h : ditto
+
+	* string.c : ditto
+
+	* struct.c : ditto
+
+	* time.c : ditto
+
+	* utf8.c : ditto
+
+	* util.c : ditto
+
+	* util.h : ditto
+
+	* variable.c : ditto
+
+	* version.c : ditto
+
+
+2006-02-12(Sun) 21:33:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c : fix to ruby's indent
+
+	* ascii.c : ditto
+
+	* bignum.c : ditto
+
+	* blockinlining.c : ditto
+
+	* call_cfunc.h : ditto
+
+	* class.c : ditto
+
+	* compar.c : ditto
+
+	* compile.c : ditto
+
+	* compile.h : ditto
+
+	* debug.c : ditto
+
+	* debug.h : ditto
+
+	* defines.h : ditto
+
+	* dir.c : ditto
+
+	* disasm.c : ditto
+
+	* dln.c : ditto
+
+	* dln.h : ditto
+
+	* enum.c : ditto
+
+	* enumerator.c : ditto
+
+	* error.c : ditto
+
+	* euc_jp.c : ditto
+
+	* eval.c : ditto
+
+	* eval_error.h : ditto
+
+	* eval_intern.h : ditto
+
+	* eval_jump.h : ditto
+
+	* eval_load.c : ditto
+
+	* eval_method.h : ditto
+
+	* eval_proc.c : ditto
+
+	* eval_safe.h : ditto
+
+	* eval_thread.c : ditto
+
+	* file.c : ditto
+
+	* gc.c : ditto
+
+	* hash.c : ditto
+
+	* insnhelper.h : ditto
+
+	* insns.def : ditto
+
+	* intern.h : ditto
+
+	* io.c : ditto
+
+	* lex.c : ditto
+
+	* main.c : ditto
+
+	* marshal.c : ditto
+
+	* math.c : ditto
+
+	* missing.h : ditto
+
+	* node.h : ditto
+
+	* numeric.c : ditto
+
+	* object.c : ditto
+
+	* oniguruma.h : ditto
+
+	* opt_insn_unif.def : ditto
+
+	* opt_operand.def : ditto
+
+	* pack.c : ditto
+
+	* prec.c : ditto
+
+	* process.c : ditto
+
+	* random.c : ditto
+
+	* range.c : ditto
+
+	* re.c : ditto
+
+	* re.h : ditto
+
+	* regcomp.c : ditto
+
+	* regenc.c : ditto
+
+	* regenc.h : ditto
+
+	* regerror.c : ditto
+
+	* regex.h : ditto
+
+	* regexec.c : ditto
+
+	* regint.h : ditto
+
+	* regparse.c : ditto
+
+	* regparse.h : ditto
+
+	* ruby.c : ditto
+
+	* ruby.h : ditto
+
+	* rubyio.h : ditto
+
+	* rubysig.h : ditto
+
+	* signal.c : ditto
+
+	* sjis.c : ditto
+
+	* sprintf.c : ditto
+
+	* st.c : ditto
+
+	* st.h : ditto
+
+	* string.c : ditto
+
+	* struct.c : ditto
+
+	* test.rb : ditto
+
+	* thread.c : ditto
+
+	* thread_pthread.h : ditto
+
+	* thread_win32.h : ditto
+
+	* time.c : ditto
+
+	* utf8.c : ditto
+
+	* util.c : ditto
+
+	* util.h : ditto
+
+	* variable.c : ditto
+
+	* version.c : ditto
+
+	* vm.c : ditto
+
+	* vm.h : ditto
+
+	* vm_dump.c : ditto
+
+	* vm_evalbody.h : ditto
+
+	* vm_macro.def : ditto
+
+	* yarv.h : ditto
+
+	* yarv_version.h : ditto
+
+	* yarvcore.c : ditto
+
+	* yarvcore.h : ditto
+
+
+2006-02-12(Sun) 15:53:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/abbrev.rb : added
+
+	* lib/base64.rb : ditto
+
+	* lib/cgi-lib.rb : ditto
+
+	* lib/csv.rb : ditto
+
+	* lib/date2.rb : ditto
+
+	* lib/eregex.rb : ditto
+
+	* lib/ipaddr.rb : ditto
+
+	* lib/irb.rb : ditto
+
+	* lib/irb/cmd/chws.rb : ditto
+
+	* lib/irb/cmd/fork.rb : ditto
+
+	* lib/irb/cmd/help.rb : ditto
+
+	* lib/irb/cmd/load.rb : ditto
+
+	* lib/irb/cmd/nop.rb : ditto
+
+	* lib/irb/cmd/pushws.rb : ditto
+
+	* lib/irb/cmd/subirb.rb : ditto
+
+	* lib/irb/completion.rb : ditto
+
+	* lib/irb/context.rb : ditto
+
+	* lib/irb/ext/change-ws.rb : ditto
+
+	* lib/irb/ext/history.rb : ditto
+
+	* lib/irb/ext/loader.rb : ditto
+
+	* lib/irb/ext/math-mode.rb : ditto
+
+	* lib/irb/ext/multi-irb.rb : ditto
+
+	* lib/irb/ext/save-history.rb : ditto
+
+	* lib/irb/ext/tracer.rb : ditto
+
+	* lib/irb/ext/use-loader.rb : ditto
+
+	* lib/irb/ext/workspaces.rb : ditto
+
+	* lib/irb/extend-command.rb : ditto
+
+	* lib/irb/frame.rb : ditto
+
+	* lib/irb/help.rb : ditto
+
+	* lib/irb/init.rb : ditto
+
+	* lib/irb/input-method.rb : ditto
+
+	* lib/irb/lc/error.rb : ditto
+
+	* lib/irb/lc/help-message : ditto
+
+	* lib/irb/lc/ja/CVS/Entries : ditto
+
+	* lib/irb/lc/ja/CVS/Repository : ditto
+
+	* lib/irb/lc/ja/CVS/Root : ditto
+
+	* lib/irb/lc/ja/error.rb : ditto
+
+	* lib/irb/lc/ja/help-message : ditto
+
+	* lib/irb/locale.rb : ditto
+
+	* lib/irb/notifier.rb : ditto
+
+	* lib/irb/output-method.rb : ditto
+
+	* lib/irb/ruby-lex.rb : ditto
+
+	* lib/irb/ruby-token.rb : ditto
+
+	* lib/irb/slex.rb : ditto
+
+	* lib/irb/version.rb : ditto
+
+	* lib/irb/workspace.rb : ditto
+
+	* lib/irb/ws-for-case-2.rb : ditto
+
+	* lib/irb/xmp.rb : ditto
+
+	* lib/jcode.rb : ditto
+
+	* lib/logger.rb : ditto
+
+	* lib/mailread.rb : ditto
+
+	* lib/mathn.rb : ditto
+
+	* lib/parsedate.rb : ditto
+
+	* lib/pathname.rb : ditto
+
+	* lib/ping.rb : ditto
+
+	* lib/pstore.rb : ditto
+
+	* lib/resolv-replace.rb : ditto
+
+	* lib/resolv.rb : ditto
+
+	* lib/rss.rb : ditto
+
+	* lib/rss/0.9.rb : ditto
+
+	* lib/rss/1.0.rb : ditto
+
+	* lib/rss/2.0.rb : ditto
+
+	* lib/rss/content.rb : ditto
+
+	* lib/rss/converter.rb : ditto
+
+	* lib/rss/dublincore.rb : ditto
+
+	* lib/rss/image.rb : ditto
+
+	* lib/rss/maker.rb : ditto
+
+	* lib/rss/maker/0.9.rb : ditto
+
+	* lib/rss/maker/1.0.rb : ditto
+
+	* lib/rss/maker/2.0.rb : ditto
+
+	* lib/rss/maker/base.rb : ditto
+
+	* lib/rss/maker/content.rb : ditto
+
+	* lib/rss/maker/dublincore.rb : ditto
+
+	* lib/rss/maker/image.rb : ditto
+
+	* lib/rss/maker/syndication.rb : ditto
+
+	* lib/rss/maker/taxonomy.rb : ditto
+
+	* lib/rss/maker/trackback.rb : ditto
+
+	* lib/rss/parser.rb : ditto
+
+	* lib/rss/rexmlparser.rb : ditto
+
+	* lib/rss/rss.rb : ditto
+
+	* lib/rss/syndication.rb : ditto
+
+	* lib/rss/taxonomy.rb : ditto
+
+	* lib/rss/trackback.rb : ditto
+
+	* lib/rss/utils.rb : ditto
+
+	* lib/rss/xml-stylesheet.rb : ditto
+
+	* lib/rss/xmlparser.rb : ditto
+
+	* lib/rss/xmlscanner.rb : ditto
+
+	* lib/rubyunit.rb : ditto
+
+	* lib/scanf.rb : ditto
+
+	* lib/shell.rb : ditto
+
+	* lib/singleton.rb : ditto
+
+	* lib/tsort.rb : ditto
+
+	* lib/weakref.rb : ditto
+
+	* eval_jump.c : removed
+
+
+2006-02-12(Sun) 15:39:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* parse.y : fix to remove including env.h
+
+	* yarvtest/test_exception.rb : fix syntax (add 'end')
+
+
+2006-02-12(Sun) 15:14:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* env.h : removed
+
+	* common.mk : remove env.h dependency
+
+	* compile.c, eval_intern.h : remove include env.h
+
+	* vm.c : ditto
+
+	* ruby.h, gc.c, error.c : remove T_SCOPE, T_VARMAP
+
+	* parse.y, eval.c : use rb_parse_in_eval() instead of ruby_in_eval
+
+	* yarvcore.c, yarvcore.h : add a prase_in_eval member to yarv_thread_t
+
+	* insns.def : add push value to throw instruction
+	for stack consistency
+
+	* yarvtest/test_exception.rb : add a test for above
+
+	* test/ruby/test_gc.rb : fix typo
+
+
+2006-02-12(Sun) 05:05:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, eval_intern.h, eval_load.c, eval_proc.c, node.h,
+	insnhelper.h, insns.def, vm.c, yarvcore.c, yarvcore.h :
+	change cref data structure and unify ruby_class and ruby_cbase
+	and some refoctoring
+
+
+2006-02-11(Sat) 23:41:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (methoddef) : fix method declaration in method
+
+	* thread.c : Thread.critical to show warning (no effect)
+
+
+2006-02-11(Sat) 20:20:18 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix [yarv-dev:831]
+
+	* yarvtest/test_class.rb : add a test for above
+
+
+2006-02-11(Sat) 14:29:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/mklog.rb : use svk
+
+	* error.c : remove newline
+
+	* eval.c (rb_block_call) : added
+
+	* eval_thread.c : remove some unused functions, comments
+
+	* thread.c : add comments (move from eval_thread.c) and support Mutex
+
+	* thread.c (rb_thread_select) : supported
+
+	* thread_pthread.h (native_mutex_trylock) : added (macro)
+
+	* thread_win32.h (native_mutex_trylock) : added
+
+	* yarvcore.c : remove unused code
+
+	* array.c : import ruby 1.9
+
+	* compar.c : ditto
+
+	* dln.c : ditto
+
+	* enum.c : ditto
+
+	* enumerator.c : ditto
+
+	* ext/digest/digest.c : ditto
+
+	* ext/digest/digest.h : ditto
+
+	* ext/digest/sha2/sha2.c : ditto
+
+	* ext/etc/etc.c : ditto
+
+	* ext/win32ole/win32ole.c : ditto
+
+	* hash.c : ditto
+
+	* intern.h : ditto
+
+	* io.c : ditto
+
+	* main.c : ditto
+
+	* missing.h : ditto
+
+	* missing/flock.c : ditto
+
+	* missing/isinf.c : ditto
+
+	* missing/vsnprintf.c : ditto
+
+	* lib/cgi.rb : ditto
+
+	* lib/complex.rb : ditto
+
+	* lib/delegate.rb : ditto
+
+	* lib/erb.rb : ditto
+
+	* lib/fileutils.rb : ditto
+
+	* lib/matrix.rb : ditto
+
+	* lib/mkmf.rb : ditto
+
+	* lib/optparse.rb : ditto
+
+	* lib/ostruct.rb : ditto
+
+	* lib/pp.rb : ditto
+
+	* lib/timeout.rb : ditto
+
+	* lib/tmpdir.rb : ditto
+
+	* lib/test/unit/autorunner.rb : ditto
+
+	* node.h : ditto
+
+	* object.c : ditto
+
+	* parse.y : ditto
+
+	* ruby.c : ditto
+
+	* sample/test.rb : ditto
+
+	* sprintf.c : ditto
+
+	* st.c : ditto
+
+	* test/ruby/test_whileuntil.rb : ditto
+
+	* test/runner.rb : ditto
+
+	* time.c : ditto
+
+	* lib/net/.document : added
+
+	* lib/net/ftp.rb : ditto
+
+	* lib/net/http.rb : ditto
+
+	* lib/net/https.rb : ditto
+
+	* lib/net/imap.rb : ditto
+
+	* lib/net/pop.rb : ditto
+
+	* lib/net/protocol.rb : ditto
+
+	* lib/net/smtp.rb : ditto
+
+	* lib/net/telnet.rb : ditto
+
+	* lib/open-uri.rb : ditto
+
+
+2006-02-10(Fri) 08:07:34 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def, yarvcore.h : support defined?(private_method) and
+	defined?(protected_method) (separate DEFINE_METHOD / DEFINE_FUNC)
+
+	* yarvtest/test_syntax.rb : add a test for above
+
+	* compile.c (iseq_compile_each) : fix NODE_RETURN bug
+	(double ensure invoke)
+
+	* yarvtest/test_flow.rb : add a test for above
+
+	* eval.c (get_errinfo) : fix to search $!
+
+	* yarvtest/test_exception.rb : add tests for above
+
+	* eval_safe.h : support $SAFE
+
+	* ext/socket/socket.c : import ruby 1.9
+
+	* gc.c (gc_mark_children) : fix making T_VALUE
+
+	* test/ruby/test_gc.rb : use GC.stress
+
+	* signal.c (sighandler) : send interrupt signal if thread blocked
+
+	* test/ruby/test_proc.rb : remove assert false
+
+	* test/ruby/test_readpartial.rb : change fail message
+
+	* test/ruby/test_signal.rb : remove assert false
+
+	* thread.c (thread_start_func_2) : set local_lfp/local_svar
+	at thread creation
+
+	* thread_pthread.h : export native_thread_interrupt
+
+	* thread_win32.h : export native_thread_interrupt
+
+	* version.h : import ruby 1.9
+
+	* vm.c (lfp_svar), yarvcore.h : fix to use Thread local svar
+
+	* yarvtest/test_thread.rb : add a test for above
+
+	* win32/Makefile.sub : import ruby 1.9
+
+	* win32/dir.h : ditto
+
+	* win32/setup.mak : ditto
+
+	* win32/win32.c : ditto
+
+	* yarvtest/yarvtest.rb : fix to remove using ARGV
+
+
+2006-02-10(Fri) 01:04:58 +0900  Yukihiro Matsumoto  <matz@ruby-lang.org>
+
+	* gc.c (rb_gc_call_finalizer_at_exit): turn on during_gc while
+	  invoking finalizers.
+
+	* gc.c (rb_gc_finalize_deferred): ditto.
+
+
+2006-02-08(Wed) 23:17:44 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_proc.rb: method names were wrongly duplicated.
+
+
+2006-02-08(Wed) 21:30:01 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* ext/nkf: added (imported from ruby CVS trunk HEAD).
+
+	* ext/nkf/depend: new file (rev 1.5).
+
+	* ext/nkf/extconf.rb: new file (rev 1.2).
+
+	* ext/nkf/nkf.c: new file (rev 1.12).
+
+	* ext/nkf/test.rb: new file (rev 1.7).
+
+	* ext/nkf/nkf-utf8/nkf.c: new file (rev 1.17).
+
+	* ext/nkf/nkf-utf8/config.h: new file (rev 1.4).
+
+	* ext/nkf/nkf-utf8/utf8tbl.c: new file (rev 1.6).
+
+	* ext/nkf/lib/kconv.rb: new file (rev 1.13).
+
+	* test/nkf: added (imported from ruby CVS trunk HEAD).
+
+	* test/nkf/test_kconv.rb: new file (rev 1.1).
+
+	* test/nkf/test_nkf.rb: new file (rev 1.1).
+
+
+2006-02-08(Wed) 21:07:36 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/find.rb: new file (imported from ruby CVS trunk HEAD,
+	  rev 1.15).
+
+
+2006-02-07(Tue) 17:58:18 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : support BEGIN{} and add preexe instruction
+
+	* insns.def : fix getspecial/setspecial instructions
+	to catch up svar change
+
+	* test/ruby/test_system.rb : remove stopper
+
+	* thread.c (rb_thread_fd_writable) : add a debug output
+
+	* thread.c (rb_thread_wait_fd) : add a debug output
+
+	* vm.c (lfp_svar) : refactoring and fix some problems
+
+	* vm_dump.c (yarv_bug) : add branch
+
+	* yarv.h : remove unused declarations
+
+	* yarvcore.c (vm_free) : VM object should not free by GC
+
+
+2006-02-07(Tue) 14:42:25 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, eval_load.c : remove rb_thread_start_1()
+
+	* eval.c : fix some prototypes and indents
+
+	* eval_thread.c, thread.c : move some functions
+	from eval_thread.c to thread.c
+
+	* signal.c (sighandler) : add line braek in error message
+
+	* yarvcore.c, yarvcore.h, thread.c : support ThreadGroup
+
+	* ruby.h, gc.c, vm.c : make new basic type RValue and T_VALUE.
+	RValue includes three values in itself.  RValue is used as
+	svar
+
+
+2006-02-06(Mon) 23:51:41 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_hash.rb: import many tests from rubicon.
+
+
+2006-02-04(Sat) 18:36:41 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_array.rb: import many tests from rubicon.
+
+
+2006-02-04(Sat) 17:47:44 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_signal.rb (test_exit_action): lib/timeout.rb is
+	  not implemented yet.
+
+
+2006-02-04(Sat) 17:42:31 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_readpartial.rb: lib/timeout.rb is not implemented
+	  yet.
+
+
+2006-02-04(Sat) 16:22:38 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_pipe.rb: remove useless require.
+
+	* test/ruby/test_signal.rb: turn off the test case which causes
+	  segmentation fault (tmp).
+
+
+2006-02-04(Sat) 08:19:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add dependency to yarvcore.h on signal.o
+
+	* compile.c (iseq_compile_each) : fix [yarv-dev:795] problem
+	(prohibit "break", "next" jump from eval)
+
+	* eval.c : fix indent
+
+	* eval_thread.c, thread.c : remove some functions and move to thread.c
+
+	* insns.def, vm.c : fix [yarv-dev:799] and [yarv-dev:800]
+
+	* yarvtest/test_class.rb : add a test for above
+
+	* test/ruby/test_gc.rb : remove GC.debug_flag control
+
+	* test/ruby/test_readpartial.rb : disable
+
+	* test/ruby/test_signal.rb : disable
+
+	* thread.c : fix thread_debug() and many bugs
+
+	* thread.c (yarv_thread_s_new) : move living_threads setting
+
+	* thread.c (yarv_thread_join) : fix
+
+	* thread_pthread.h : add type native_thread_data_t (dummy)
+	and support interrupt blocking thread
+
+	* thread_pthread.h (native_thread_apply_priority) : added
+
+	* thread_win32.h : add type native_thread_data_t (dummy)
+	and support interrupt blocking thread
+
+	* yarvcore.h : use win32 thread system on cygwin and fix
+	some struct members
+
+	* yarvtest/test_thread.rb : added
+
+
+2006-02-03(Fri) 00:08:09 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_string.rb: import many tests from rubicon.
+
+
+2006-02-02(Thu) 23:20:13 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/envutil.rb: new file (imported from ruby trunk HEAD).
+
+	* test/ruby/marshaltestlib.rb: ditto.
+
+	* test/ruby/test_array.rb: ditto.
+
+	* test/ruby/test_beginendblock.rb: ditto.
+
+	* test/ruby/test_clone.rb: ditto.
+
+	* test/ruby/test_dir.rb: ditto.
+
+	* test/ruby/test_env.rb: ditto.
+
+	* test/ruby/test_file.rb: ditto.
+
+	* test/ruby/test_float.rb: ditto.
+
+	* test/ruby/test_fnmatch.rb: ditto.
+
+	* test/ruby/test_hash.rb: ditto.
+
+	* test/ruby/test_io.rb: ditto.
+
+	* test/ruby/test_marshal.rb: ditto.
+
+	* test/ruby/test_math.rb: ditto.
+
+	* test/ruby/test_pack.rb: ditto.
+
+	* test/ruby/test_path.rb: ditto.
+
+	* test/ruby/test_pipe.rb: ditto.
+
+	* test/ruby/test_rand.rb: ditto.
+
+	* test/ruby/test_range.rb: ditto.
+
+	* test/ruby/test_readpartial.rb: ditto.
+
+	* test/ruby/test_regexp.rb: ditto.
+
+	* test/ruby/test_settracefunc.rb: ditto.
+
+	* test/ruby/test_signal.rb: ditto.
+
+	* test/ruby/test_sprintf.rb: ditto.
+
+	* test/ruby/test_string.rb: ditto.
+
+	* test/ruby/test_stringchar.rb: ditto.
+
+	* test/ruby/test_struct.rb: ditto.
+
+	* test/ruby/test_symbol.rb: ditto.
+
+	* test/ruby/test_system.rb: ditto.
+
+	* test/ruby/test_time.rb: ditto.
+
+	* test/ruby/ut_eof.rb: ditto.
+
+
+2006-02-02(Thu) 22:53:44 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_proc.rb: test [yarv-dev:777].
+
+
+2006-02-01(Wed) 03:51:39 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* gc.c : add GC.debug_flag= method
+
+	* insns.def : support method definition in method
+
+	* yarvtest/test_method.rb : add tests for above
+
+
+2006-01-29(Sun) 11:40:26 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_proc.c (proc_alloc) : fix [yarv-dev:777]
+
+	* yarvtest/test_proc.rb : add a test for above
+
+	* insns.def : fix [yarv-dev:782] and add YARV_CHECK_INTS()
+
+	* yarvtest/test_class.rb : add a test for above
+
+	* thread_win32.h : fix [yarv-dev-en:23]
+
+	* vm.c (th_call0) : add YARV_CHECK_INTS()
+
+
+2006-01-09(Mon) 11:56:34 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* yarvcore.h: add prototype (remove warning).
+
+	* vm.c (th_invoke_proc): make save variables volatile.
+
+	* eval.c (eval): initialize local variables (remove warnings).
+
+	* eval_thread.c (rb_exec_recursive): ditto.
+
+	* yarvcore.c (thread_mark): ditto.
+
+	* vm.c (th_invoke_proc): ditto.
+
+	* eval.c: remove useless prototypes.
+
+
+2006-01-09(Mon) 10:25:12 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* eval_thread.c: rb_thread_join is required to build ruby on
+	  Linux.
+
+	* compile.c: unify coding style.
+
+	* yarvcore.c: ditto.
+
+
+2006-01-06(Fri) 09:21:34 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* vm.c: coding style change only.
+
+
+2006-01-04(Wed) 14:12:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (ruby_init), eval_intern.h : use POP_TAG_INIT() at bootstrap
+
+	* eval_thread.c : remove unused functions and comments
+
+	* intern.h : expose rb_make_exception()
+
+	* signal.c : support signal
+
+	* thread.c (yarv_thread_execute_interrupts) : added
+
+	* thread_pthread.h (thread_timer) : set interrupt_flag of
+	current running threads
+
+	* vm.c (th_invoke_proc) : jump with JUMP_TAG() if some exception
+	occurs
+
+	* yarv.h : add yarv_set_current_running_thread_raw() for bootstrap
+
+	* yarvcore.c : add yarv_segv() and segv() method for test
+
+	* yarvcore.c (Init_yarvcore) : set yarv_thread_t#running_thread
+
+	* yarvcore.h : fix yarv_thread_t members
+
+
+2006-01-03(Tue) 22:25:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* disasm.c (insn_operand_intern) : fix to add child iseq
+
+	* eval.c, gc.c : remove obsolete static variables (ruby_scope,
+	ruby_dyna_vars, ruby_frame)
+
+	* eval.c (rb_mod_s_constants) : use ruby_cref()
+
+	* eval.c (eval) : use th_restore_klass()
+
+	* eval_proc.c (rb_f_binding) : use th_store_klass()
+
+	* insns.def (concatarray) : fix insn ([expr, *nil] => [expr])
+
+	* vm.c (th_set_env), insnhelper.h : remove macro
+
+	* vm.c (eval_get_cvar_base) : use get_cref
+
+	* vm.c (th_make_proc) : use th_store_klass()
+
+	* vm_macro.def (macro_eval_invoke_func) : fix option args size
+
+	* vm_macro.def (macro_eval_invoke_func) : raise stack overflow error
+
+	* yarvcore.h : add yarv_stored_klass_t type
+
+	* yarvcore.c : fix mark functions around yarv_stored_klass_t
+
+
+2006-01-01(Sun) 05:14:26 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* lib/benchmark.rb: new file (imported from original ruby, rev
+	  1.10).
+
+
+2006-01-01(Sun) 03:51:10 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* yarvcore.c: add prototype.
+
+	* re.c: remove warning: long -> unsigned long.
+
+	* debug.c: adjust coding style.
+
+	* yarv.h: ditto.
+
+
+2006-01-01(Sun) 03:43:33 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* variable.c: add prototype.
+
+	* eval.c: ditto.
+
+	* eval_load.c: ditto.
+
+
+2006-01-01(Sun) 02:41:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add address analyse to vtune rule
+
+	* rb/vtlh.rb : added for above
+
+	* rb/insns2vm.rb, template/vm.inc.tmpl : insert #line directive
+	to reference above
+
+	* vm_macro.def (macro_eval_invoke_cfunc) : fix indent
+
+	* yarvtest/test_method.rb : fix indent, spacing
+	and add a test for alias
+
+
+2005-12-31(Sat) 12:42:05 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add Intel VTune rule (make vtune)
+
+	* eval.c, yarvcore.h : fix to remove yarv_thread_t#local_*
+
+	* parse.y (top_local_init_gen) : fix a problem ([yarv-dev:765])
+
+	* yarvtest/test_eval.rb : add a test for above
+
+	* vm.c (thread_eval) :remove unused function
+
+	* yarvcore.c (Init_yarvcore) : remove YARVCore::Thread::eval method
+
+	* yarvcore.c (thread_eval) : remove unused function
+
+
+2005-12-31(Sat) 06:05:00 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (eval_search_super_klass) : pass block to method missing
+
+	* vm_macro.def (macro_eval_invoke_method) : ditto
+
+	* yarvtest/test_method.rb : add a test for above
+
+
+2005-12-31(Sat) 03:11:14 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c (eval), eval_proc.c (rb_f_binding) : save klass, etc to
+	binding and use it at eval
+
+	* eval_intern.h : ditto
+
+	* yarvtest/test_eval.rb : add tests for above
+
+	* yarvcore.c (th_get_special_cref) : added
+
+	* yarvcore.h : add a prototype of above
+
+	* vm.c (th_get_cref) : refactoring
+
+	* vm.c (eval_get_ev_const) : fix SEGV at A::B (A is not class/module)
+	([yarv-dev:758])
+
+	* yarvtest/test_bin.rb : add a test for above
+
+	* rb/mklog.rb : use external diff command and show function name
+
+
+2005-12-30(Fri) 19:07:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, yarvcore.h, eval.c, eval_proc.c : support
+	Ruby's Binding
+
+	* yarvcore.c : support TOPLEVEL_BINDING
+
+	* yarvtest/test_eval.rb : add tests for above
+
+
+2005-12-30(Fri) 13:12:28 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_eval.rb: more tests for
+	  module_eval/instance_eval.
+
+
+2005-12-30(Fri) 05:06:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add dependency (yarvcore.h) for gc.c
+
+	* eval.c, eval_intern.h, eval_load.c, eval_method.h,
+	insns.def, insnhelper.h, vm.c, yarvcore.c, yarvcore.h :
+	re-write class reference
+
+	* yarvtest/test_eval.rb : added
+
+	* yarvtest/test_proc.rb :
+
+
+2005-12-29(Thu) 12:27:12 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, yarvcore.h :
+	remvoe needless yarv_iseq_t#rewind_frame_size
+
+
+2005-12-29(Thu) 11:17:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add dependency to test-all rule
+
+	* eval.c (rb_sourceline), vm.c (th_get_sourceline) :
+	fix to skip process if iseq is ifunc
+
+	* test/ruby/test_lambda.rb : assert(fail, ...) instead of assert_fail
+
+	* test/ruby/test_proc.rb : ditto
+
+	* vm_dump.c : fix stack dump (iseq name)
+
+	* vm_macro.def : store proc (block proc) to cfp#proc for GC mark
+
+	* yarvcore.c : mark above on thread_mark
+
+	* eval.c (exec_under) : replace block#self ([yarv-dev:751])
+
+
+2005-12-29(Thu) 01:56:46 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : fix setting of Proc cref ([yarv-dev:741])
+
+	* yarvcore.c : fix indent
+
+
+2005-12-29(Thu) 00:17:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* disasm.c : show (block) local variable simple (not as symbol)
+
+	* gc.c : fix syntax error
+
+
+2005-12-28(Wed) 23:35:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* class.c (method_entry) : fixed for undefed method ([yarv-dev:743])
+
+	* compile.c : fix errinfo dvar id (#$!)
+	and fix NODE_ERRINFO compilation
+
+	* eval_proc.c, yarvcore.c : support YARVCore::VM::Proc.new
+
+	* insns.def : remove useless TODO comments
+
+	* insns.def : fix to use strict array conversion on
+	checkarrayinclude
+
+	* insns.def : fix defined?(yield) ([yarv-dev:744])
+
+	* yarvcore.h : change yarv_iseq_t layout
+
+
+2005-12-28(Wed) 16:49:55 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_eval.rb: add TODO comment.
+
+	* test/ruby/test_iterator.rb: rename YARVCore::VM::Proc -> Proc
+	  (tmp).
+
+	* test/ruby/test_lambda.rb: use assert_fail.
+
+	* test/ruby/test_proc.rb: ditto.
+
+
+2005-12-28(Wed) 16:28:35 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_clone.rb: removed (tmp).
+
+	* test/ruby/test_eval.rb: define missing method Object#funcall
+	  (tmp).
+
+	* test/ruby/test_lambda.rb: turn off tests for "->".
+
+	* test/ruby/test_proc.rb: turn off tests for |&b|.
+
+	* test/ruby/test_proc.rb: turn off tests for $SAFE setter.
+
+
+2005-12-28(Wed) 15:31:46 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix calculation of stack_max
+
+	* eval.c (rb_iter) : fix block/retry handling
+
+	* yarvtest/test_flow.rb : add tests for above
+
+	* insns.def : fix block passing on super (super(&nil))
+
+	* vm_macro.def, insns.def : fix convert method of object to array
+
+	* yarvtest/test_method.rb : fix a test for above
+
+	* vm.c : fix backtrace generate algorithm
+
+
+2005-12-28(Wed) 10:36:45 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, compile.h : refactoring (remove self passing, etc)
+
+	* disasm.c : support showing ID of method/dynamic local variables
+
+	* rb/allload.rb : add verbose version (it's enable by $DEBUG)
+
+	* template/insns.inc.tmpl, template/insns_info.inc.tmpl,
+	template/minsns.inc.tmpl, template/opt_sc.inc.tmpl,
+	template/optinsn.inc.tmpl, template/optunifs.inc.tmpl,
+	template/vmtc.inc.tmpl : fix a comment
+
+	* variable.c (mod_av_set) : fix to clear inline cache ([yarv-dev:720])
+
+	* eval_method.h : fix to clear inline method cache
+
+	* vm.c, rb/insns2vm.rb, template/insns_info.inc.tmpl, compile.c,
+	insns.def, vm_evalbody.h, vm_macro.def :
+	fix operands types (ulong -> num_t, ...)
+
+	* vm_macro.def : fix to check SPECIAL_CONST_P() at splat array
+	([yarv-dev:722])
+
+	* yarvcore.c : fix to throw syntax error
+
+	* yarvcore.h, eval.c, eval_error.h, eval_jump.h :
+	add yarv_vm_t#exit_code to fix problem at cleanup ([yarv-dev:723])
+
+	* insns.def : fix to invoke zsuper in method defined by define_method
+	([yarv-dev:704])
+
+	* yarvtest/test_class.rb : add tests for above
+
+	* yarvtest/test_method.rb : fix comments
+
+
+2005-12-27(Tue) 01:52:07 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c, intern.h, insns.def : expose rb_ary_replace and use it
+	in insns.def
+
+	* eval.c : fix to use SCOPE_* to NOEX_*
+
+	* eval_intern.h : remove SCOPE_*
+	and fix SCOPE_TEST() and SCOPE_SET(f)
+
+	* eval_load.c : save and store klass and visibility
+	at require and load
+
+	* eval_method.h : fix undefed method node ([yarv-dev-en:8])
+
+	* eval_proc.c : fix define_method ([yarv-dev:704])
+
+	* insnhelper.h, vm.h : remove GET_VM_STATE_VERSION(),
+	INC_VM_STATE_VERSION() and move these to vm.h
+
+	* insns.def : supportintg visibility
+
+	* node.h : remove NOEX_RECV
+
+	* variable.c, vm.c : add rb_vm_change_state() and use it in
+	remove_const
+
+	* vm.c, insns.def, yarvcore.h, yarvcore.c : add eval_push_cref(),
+	eval_pop_cref() and th_cref_init to manage current visibility
+
+	* yarv.h : add a prototype of rb_vm_change_state()
+
+	* yarvcore.h, insns.def : add defined_method_id and support
+	super in define_method scope
+
+	* yarvtest/test_class.rb : add tests for above
+
+
+2005-12-26(Mon) 20:44:38 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_basicinstructions.rb: new file.
+
+
+2005-12-26(Mon) 08:40:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c (eval_get_ev_const) : fix to skip nil
+
+
+2005-12-26(Mon) 08:27:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.h : fix GET_CVAR_EV_KLASS [yarv-dev:703]
+
+
+2005-12-26(Mon) 07:51:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : add emptstack insn for all NODE_RETURN
+	and optimize it if it's not needed
+
+	* yarvtest/test_flow.rb : add a test for above
+
+
+2005-12-26(Mon) 07:08:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, gc.c : add "gc_debug_flag" to debug gc
+
+	* insns.def : add emptstack
+
+	* compile.c, rb/insns2vm.rb, template/insns_info.inc.tmpl :
+	change interface of insn_stack_increase
+
+	* compile.c : fix return from ensure in method [yarv-dev:702]
+
+	* yarvtest/test_flow.rb : add tests for above
+
+
+2005-12-26(Mon) 02:15:02 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/ruby/test_alias.rb: do not use unimplemented defined?.
+
+
+2005-12-26(Mon) 02:00:11 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* test/runner.rb: new file.
+
+	* test/ruby/test_alias.rb: new file.
+
+	* test/ruby/test_clone.rb: new file.
+
+	* test/ruby/test_eval.rb: new file.
+
+	* test/ruby/test_iterator.rb: new file.
+
+	* test/ruby/test_lambda.rb: new file.
+
+	* test/ruby/test_proc.rb: new file.
+
+	* test/ruby/test_super.rb: new file.
+
+	* test/ruby/test_assignment.rb: new file.
+
+	* test/ruby/test_bignum.rb: new file.
+
+	* test/ruby/test_call.rb: new file.
+
+	* test/ruby/test_case.rb: new file.
+
+	* test/ruby/test_condition.rb: new file.
+
+	* test/ruby/test_const.rb: new file.
+
+	* test/ruby/test_defined.rb: new file.
+
+	* test/ruby/test_exception.rb: new file.
+
+	* test/ruby/test_gc.rb: new file.
+
+	* test/ruby/test_ifunless.rb: new file.
+
+	* test/ruby/test_method.rb: new file.
+
+	* test/ruby/test_trace.rb: new file.
+
+	* test/ruby/test_variable.rb: new file.
+
+	* test/ruby/test_whileuntil.rb: new file.
+
+
+2005-12-25(Sun) 07:40:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c, compile.c : fix block inlining
+
+	* rb/insns2vm.rb : fix to support tracing stack depth
+	with operands unification
+
+	* vm_dump.c : fix to print Qundef on stack dump
+
+
+2005-12-25(Sun) 01:45:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c, rb/insns2vm.rb, template/insns_info.inc.tmpl :
+	trace stack depth at compile time
+	and use it  as cont_sp for exception handling
+
+	* yarvtest/test_exception.rb : add tests for above
+
+	* yarvtest/test_flow.rb : ditto
+
+	* Merry Xmas :)
+
+
+2005-12-24(Sat) 19:34:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, compile.h : fix ADD_CATCH_ENTRY and add LABEL#sp
+
+	* eval_jump.h : fix catch to remove illegal error
+
+
+2005-12-24(Sat) 09:05:23 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_method.h : change data structure for RClass#m_tbl
+
+	* class.c, eval.c, eval_proc.c : fix for above changes
+
+	* node.h, gc.c : change NODE_FBODY, NODE_METHOD members
+	for above changes
+
+	* insns.def : support private/protected visibility
+
+	* vm_macro.def : ditto
+
+	* vm.c : ditto
+
+	* thread.c : fix typo
+
+	* thread_pthread.h : fix typo
+
+	* thread_win32.h : fix typo
+
+	* eval.c, yarvcore.h : add yarv_thread_t#method_missing_reason
+	to pass method_missing reason and use it to build error message
+
+	* compile.c : use ADD_CALL instead of ADD_SEND for
+	NODE_X(D)STR, NODE_CONST (func)
+
+
+2005-12-22(Thu) 02:45:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarv_version.h, Changes : 0.3.3
+
+
+2005-12-20(Tue) 04:04:45 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix self::Const access
+
+	* yarvtest/test_bin.rb : add a test for above
+
+
+2005-12-20(Tue) 01:52:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix to expand VALUES value
+
+	* yarvtest/test_massign.rb : add a test for above
+
+
+2005-12-20(Tue) 01:32:35 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, insnhelper.h : fix cvar in singleton method/class
+
+	* yarvtest/test_bin.rb : add tests for above
+
+
+2005-12-20(Tue) 01:03:34 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, yarvcore.h : support all defined?() syntax
+
+	* compile.c : fix NODE_COLON2
+
+	* yarvtest/test_bin.rb : add or fix tests for above
+
+	* win32/* : update all
+
+
+2005-12-17(Sat) 10:46:08 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* vm_macro.def: fix printf type mismatch for LP64 system (again).
+
+	* parse.y: introduce descriptive macro for special values of
+	  lvtbl->dvars.
+
+
+2005-12-17(Sat) 09:39:27 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* vm_macro.def (macro_eval_invoke_method): fix printf type mismatch
+	  for LP64 system.
+
+
+2005-12-14(Wed) 03:49:40 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : change rescue/ensure iseq name
+
+	* eval.c, intern.h : fix a prototype
+
+	* insns.def, yarvcore.h : add trace_function
+
+	* vm.c : fix deadly bug (illegal pointer cast)
+
+	* vm_dump.c : remove unused local variables
+
+	* vm_macro.def : add parameter size check
+
+	* yarvtest/test_bin.rb : comment out 2 assertions
+
+
+2005-12-13(Tue) 03:55:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_proc.c : fix indent
+
+	* insns.def : fix getspecial instruction to return nil
+	if no entry
+
+	* yarvtest/test_syntax.rb : add a test for above
+
+	* lib/un.rb : added
+
+	* template/*.tmpl : fix typo
+
+
+2005-12-13(Mon) 01:38:17 +0900  Minero Aoki  <aamine@loveruby.net>
+
+	* yarv.h: add prototypes.
+
+	* intern.h: ditto.
+
+	* eval.c: ditto.
+
+	* debug.c: ditto.
+
+	* thread_pthread.h: fix printf type mismatch for LP64 system
+	  (Linux/AMD64).
+
+	* variable.c: ditto.
+
+	* object.c: ditto.
+
+	* gc.c: ditto.
+
+	* process.c: ditto.
+
+	* error.c: ditto.
+
+	* vm.c: ditto.
+
+	* vm.h: ditto.
+
+	* vm_dump.c: ditto.
+
+	* disasm.c: ditto.
+
+	* marshal.c: ditto.
+
+	* eval_thread.c: ditto.
+
+
+2005-12-11(Sun) 22:00:34 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : call "inherited" method when a class is inherited
+
+	* yarvcore.h : fix yarv_iseq_t field layout
+
+	* common.mk : add dependence on yarvcore.h to eval*.o files
+
+	* compile.c : fix NODE_POSTEXE logic
+
+	* insnhelper.h : use GC_GUARDED_PTR_REF instead of magic number
+
+	* eval_proc.c : fix indent
+
+	* configure : re-autoconf
+
+
+2005-12-10(Sat) 03:57:20 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : fix blockinlining.o build rule
+
+	* insns.def : remove logic for zsuper
+
+	* template/optinsn.inc.tmpl :
+
+	* vm.c : remove thread_yield_light_prepare, thread_yield_light_invoke
+
+	* compile.c : support NODE_ZSUPER with optargs, restarg
+
+	* yarvtest/test_class.rb : add tests for above
+
+
+2005-12-09(Fri) 01:13:37 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c, numeric.c, range.c : add prototype of
+	block inlining function
+
+	* blockinlining.c, vm_opts.h.base : add block inlining flag
+
+	* common.mk, debug.h, debug.c : add debug_breakpoint() for gdb
+
+	* compile.c : fix to use size_t on compile_data_alloc(),
+	fix illegal cast, fix to set arg_simple at compiling block,
+
+	* compile.c, vm.c : fix NODE_NEXT, NODE_BREAK logic
+
+	* yarvtest/test_flow.rb : add a test for above
+
+	* yarvcore.c, yarvcore.h, compile.c, eval.c : remove
+	yarv_iseq_t#root_iseq and add yarv_iseq_t#local_iseq and fix
+	to use this member field
+
+	* eval_method.h : fix indent
+
+	* gc.c : fix indent
+
+	* insns.def, compile.c : remove "zsuper" instruction (use "super"
+	instead).  This is because NODE_ZSUPER represent with only "super"
+	instruction
+
+	* yarvcore.c : add proc_arity
+
+
+2005-12-05(Mon) 03:58:30 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* array.c, blockinlining.c : support block inlining for Array#each
+
+	* disasm.c : fix catch table format
+
+	* insns.def : fix stack consistency error message
+
+	* vm.c : fix to skip pushing value at "next"
+
+	* yarvcore.h : move definition of
+	"struct iseq_compile_data_ensure_node_stack" to compile.c
+
+	* compile.c : fix ensure catch table creation
+
+	* yarvtest/test_flow.rb : add tests for above
+
+
+2005-12-03(Sat) 22:27:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c, compile.c, yarvcore.c, yarvcore.h,
+	numeric.c, range.c : collect block inlining logic to blockinlining.c
+
+
+2005-12-03(Sat) 20:24:07 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* blockinlining.c, common.mk : add blockinlining.c
+
+	* yarvcore.c, yarvcore.h, blockinlining.c, compile.c, compile.h,
+	gc.c, node.h, numeric.c, range.c :
+	support block inlining for Integer#times, Range#each
+
+	* compile.c : fix to set block redo/next point at last,
+	and fix NODE_OP_ASGN1
+
+	* compile.c, vm.c : add specialized instruction "opt_le"
+
+	* disasm.c : fix to show block, and to show catch type as string
+	and change node_name logic
+
+	* eval_thread.c : fix function type declaration
+
+	* insns.def : add instruction "putundef", "opt_checkenv"
+	to support block inlining and add stack check routine
+
+	* lib/cgi.rb : add global variable $CGI_DONTINPUT
+
+	* opt_operand.def : add some operand unification rules
+
+	* rb/insns2vm.rb : fix operand unification logic for BLOCKISEQ
+
+	* vm.c : fix exception handling routine (collect stack operations)
+
+	* vm_macro.def : fix macro_eval_invoke_bmethod
+
+	* yarvsubst.c : removed
+
+	* yarvtest/test_syn.rb : rename to yarvtest/test_syntax.rb
+
+	* yarvtest/yarvtest.rb : remove tempfile explicitly
+
+
+2005-11-30(Wed) 01:13:57 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add vm_opts.h rule
+
+	* vm.c, insns.def : fix proc creation under class and block
+	environment
+
+
+2005-11-29(Tue) 16:39:07 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, eval_proc.c, vm.c, vm_macro.def :
+	support define_method and invoke NODE_BMETHOD method
+
+
+2005-11-29(Tue) 13:18:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : add iseq_add_mark_object, iseq_add_mark_object_compile_time
+	and use it to mark objects on iseq
+
+	* compile.h, compile.c : remove cast on NEW_CHILD_ISEQVAL, NEW_ISEQVAL
+	and interface
+
+	* compile.c, disasm.c, insns.def, vm_macro.def, rb/insns2vm.rb :
+	add BLOCKISEQ parameter type
+
+	* gc.c : fix garbage_collect to return true if only allocate memory
+
+	* vm.c : fix insertion order of proc/env
+
+	* vm_evalbody.h : add typedef yarv_iseq_t *BLOCKISEQ
+
+	* yarvcore.c, yarvcore.c : add idTimes
+
+	* yarvcore.c : fix proc_mark, env_mark around iseq mark
+
+
+2005-11-28(Mon) 09:02:57 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def, vm_evalbody.h : support super
+	with splat argument and block (and zsuper with block)
+
+	* yarvtest/test_class.rb : add tests for above
+
+	* compile.c, yarvcore.h, yarvcore.c, insns.def, time.c, string.c :
+	add opt_succ insn
+
+	* eval_method.h : fix indent
+
+	* eval_thread.c : apply cast to vanish a warning
+
+	* lib/tempfile.rb, lib/tmpdir.rb : added
+
+	* vm.c : eval_method_missing added
+
+	* vm_macro.def : refactoring
+
+
+2005-11-21(Mon) 21:21:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, compile.h, yarvcore.c : remove "iseqobj"
+	variables and rename to "iseq"
+
+
+2005-11-21(Mon) 07:31:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix block parameter error
+
+	* ext/* : added
+
+	* lib/optparse* : added
+
+	* benchmark/bm_so_sieve.rb : fix parameter
+
+
+
+2005-11-21(Mon) 03:47:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : optimize condition in literal
+
+	* thread_win32.h : fix win32 thread function prototype
+
+
+2005-11-20(Sun) 17:58:24 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix NODE_AND/OR bug
+
+	* eval.c : support rb_frame_this_func()
+
+
+2005-11-20(Sun) 12:32:31 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, yarvcore.c, yarvcore.h : support NODE_OPT_N
+
+	* compile.h : add macro ADD_CALL
+
+	* debug.c : add debug_v() and change to use only printf
+	on debug_id()
+
+	* sample/test.rb :
+
+	* vm.c : fix make_proc_from_block
+
+
+2005-11-19(Sat) 14:55:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* import ruby 1.9.0 (2005-11-18)
+
+
+2005-11-19(Sat) 06:08:37 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/test : added
+
+
+2005-11-19(Sat) 05:48:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : useless jump elimination (if/unless destination)
+
+	* eval.c : rb_iter_break support,
+	 fix rb_iterate (clear errinfo if break)
+
+	* eval_proc.c : support rb_node_arity (YARV_METHOD_NODE)
+
+	* insns.def : change variable name
+
+	* vm.c : fix th_invoke_yield and add th_iter_break()
+
+	* vm_dump.c : fix yarv_bug()
+
+	* yarvcore.c : fix proc_mark to check IFUNC node and add
+	global ruby method SDR() for debug
+
+	* yarvtest/test_syn.rb : add a test for all condition combination
+
+
+2005-11-15(Tue) 05:52:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/forwardable.rb : added
+
+	* common.mk : remove "vm.o : CFLAGS += -fno-crossjumping" rule
+
+	* compile.c, yarvcore.h, insns.def : add FCALL/VCALL flag
+
+	* compile.c, insns.def : add onceinlinecache instruction
+
+	* eval.c : support $!, $@, raise (== raise $!)
+
+	* opt_operand.def : add some unification rule (send flags)
+
+	* vm.c : fix return process
+
+	* vm_macro.def : fix option prameters
+
+	* yarvtest/test_method.rb : add tests for above
+
+
+2005-11-15(Tue) 00:42:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : support rb_frame_pop() and rb_frame_callee(),
+	add rb_sourcefile(), rb_sourceline(),
+
+
+	* compile.c : support postposition while/until,
+	fix block parameter index
+
+	* yarvtest/test_syn.rb : add tests for above
+
+	* yarvcore.c : fix env_mark
+
+	* vm.h, yarvcore.h : move vm.h#cmethod_info to
+	 yarvcore.h#yarv_cmethod_info
+
+	* vm.c : add th_get_sourceline()
+
+	* eval_intern.h : fix PASS_PASSED_BLOCK()
+
+	* eval_load.c : fix re-enter require (temporalily)
+
+	* insns.def : permit re-open class when superclass is same
+
+
+2005-11-11(Fri) 01:20:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : add "allload" rule
+
+	* compile.c, yarvcore.h, insns.def, vm_macro.def, disasm.c :
+	change arg_rest, arg_block offset (1)
+
+	* insns.def : add postexe instruction
+
+	* insns.def, vm.c : support rest block parameter
+
+	* yarvtest/test_block.rb : add tests for above
+
+	* rb/allload.rb : get path from ARGV
+
+	* vm_opts.h.base : set default off
+
+
+2005-11-01(Tue) 08:28:19 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/other-lang/eval.rb : fix path
+
+	* lib/English.rb, lib/cgi.rb, lib/complex.rb, lib/delegate.rb :
+	added
+
+
+2005-11-01(Tue) 08:18:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : push and pop values after checkincludearray for
+	stack caching
+
+
+2005-10-31(Mon) 15:37:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_app_mandelbrot.rb : added
+
+	* benchmark/bm_app_factorial.rb : fixed parameter
+
+	* benchmark/bm_so_count_words.rb, benchmark/run_rite.rb : use
+	real file
+
+	* common.mk : add "ext" rule, add some dependencies and add option
+	to bench-each rule (renamed from bench-item)
+
+	* compile.c : fix get_root_iseq_object (check iseq type),
+	support splat case/when. support //o (regexp)
+
+	* eval.c : support *_eval, fix rb_obj_call_init to pass block
+
+	* eval_jump.h : support throw/catch
+
+	* eval_load.c : save klass_nest_stack when require
+
+	* eval_method.h : fix ruby_cbase()
+
+	* insnhelper.h : GET_EV_KLASS checks toplevel or not
+
+	* insns.def, yarvcore.c : fix singleton method definition and fix
+	super class's method
+
+	* lib/shellwords.rb : use String() instead of String.new()
+
+	* vm.c : check class iseq or not when making Proc and
+	add eval_search_super_klass function
+
+	* vm.h : CMETHOD_INFO_P to yarvcore.h
+
+	* vm_macro.def : splat if object type is T_ARRAY
+
+	* vm_opts.h, vm_opts.h.base : rename to vm_opts.h.base
+	insns2vm.rb will copy it to build directory
+
+	* yarvcore.c : add Proc#[]
+
+	* yarvcore.h : change INITIAL_ISEQ_COMPILE_DATA_STORAGE_BUFF_SIZE
+	to 512
+
+	* yarvtest/test_* : invalidate splat non array code (like: "*1")
+
+	* yarvtest/yarvtest.rb : use tempfile instead of popen
+
+
+2005-10-28(Fri) 09:11:53 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvtest/test_method.rb : fix test
+
+
+2005-10-28(Fri) 08:43:29 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run_rite.rb : add -I options to run benchmark
+
+	* common.mk : pass options to some rules with RUNOPT
+	and add -I options
+
+	* compile.c : fix massign with constant
+
+	* yarvtest/test_massign.rb : add tests for above
+
+	* eval_load.c : fix load_wait()
+
+	* eval_method.h : support ruby_cbase()
+
+	* lib/*.rb : add or modify libraries to run on yarv
+	* parse.y : change to ANSI C style
+
+	* vm.c : fix making proc process under cfunc/ifunc environment
+
+	* vm_macro.def : fix block pass
+
+	* yarvtest/test_method.rb : add tests for above
+
+	* yarvcore.c : add yarv_obj_is_proc()
+
+	* eval.c : fix rb_obj_is_proc to use yarv_obj_is_proc()
+
+
+2005-10-27(Thu) 11:50:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* some files : import from ruby 1.9.0 (2005-10-12)
+
+
+2005-10-16(Sun) 14:50:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c, yarvcore.h, yarvcore.c : add insns "bitblt" and "answer"
+
+
+2005-10-11(Tue) 17:01:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarv_version.h, Changes : 0.3.2
+
+
+2005-10-11(Tue) 13:35:25 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : add YARV_CHECK_INTS()
+
+	* thread.c, thread_pthread.h, thread_win32.h : kick timer thread
+	when another thread kicked
+
+	* vm.c : remove debug print
+
+	* vm_opts.h : add OPT_CALL_THREADED_CODE
+
+	* yarvtest/yarvtest.rb : remove "\r" from answer
+
+
+2005-10-07(Fri) 09:36:36 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : add member variable "interrupt_flag" to yarv_thread_t
+
+
+2005-10-05(Wed) 21:20:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eva.c, eval_thread.c, ruby.h, eval_error.h, eval_jump.h,
+	eval_load.c, thread.c, error.c, compile.h : remove ruby_errinfo
+
+	* thread_win32.h, thread_pthread.h : set stack size to 4KB
+
+	* vm.c : fix making env routine
+
+	* vm_dump.c, vm.h : support frame type "EVAL" and fix magic number
+
+	* yarvcore.c : fix some mark/free routine
+
+
+2005-10-05(Wed) 09:08:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, eval_intern.h, vm.c, eval_jump.h, yarvcore.h :
+	re-define PUSH/POP/EXEC/JUMP_TAG to use thread local tag
+
+	* inits.c, yarvcore.c : fix boostrap
+
+
+2005-10-03(Mon) 22:28:24 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix NODE_COLON2 bugs
+
+	* compile.h : fix debug routine
+
+	* disasm.c : add space between insn and operand
+
+	* insns.def : add comment of classdef, singletonclassdef
+
+	* vm.c, yarv.h : fix invoke_light routine
+
+	* yarvcore.c : fix to mark each threads
+
+
+2005-10-02(Sun) 05:55:34 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_pthread.h : add "system_working" global variable
+
+
+2005-10-02(Sun) 01:23:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c : add raw gets (for test), and fix indent
+
+
+2005-10-01(Sat) 23:06:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread_win32.h, common.mk : add thread_win32.h
+
+	* thread.c : support _WIN32 thread
+
+	* thread.c, thread_pthread.h : fix some interface
+
+	* eval_thread.c : remove debug print
+
+	* gc.c : fix stack region
+
+	* win32/Makefile.sub : add -MD flag to LDFLAGS
+
+	* yarvcore.c : fix mark and sweep debug print
+
+	* yarvcore.h : fix VM#living_threads data type to st_table
+
+
+2005-10-01(Sat) 00:25:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, yarvcore.h : rename GIL (Global Interpreter Lock) to
+	GVL (Global VM Lock)
+
+	* thread_pthread.h : fix pthread mutex initialize
+
+
+2005-09-30(Fri) 20:11:19 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c : support join with timeout
+
+	* yarvcore.h : use GET_VM()
+
+
+2005-09-30(Fri) 14:59:29 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* thread.c, common.mk : add thread.c
+
+	* thread.c, gc.c, eval_thread.c, yarvcore.c, yarvcore.h :
+	support native thread (on pthread)
+
+	* insns.def : add YARV_CHECK_INTS() check
+
+	* yarv.h : add GET_VM() macro
+
+
+2005-09-29(Thu) 22:43:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_intern.h, eval_thread.c : move thread_status to eval_intern.h
+
+	* yarvcore.c : fix thread/vm value
+
+	* yarvcore.h : add some parameter to yarv_thread_t
+
+
+2005-09-29(Thu) 01:52:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, yarvcore.h : add line number on last end instruction
+
+	* vm.c : fix line no detection
+
+
+2005-09-28(Wed) 00:02:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk, eval_load.c, eval.c, eval_intern.h : add eval_load.c
+
+	* disasm.c : fix around block local variables
+
+	* eval_proc.c : fix typo
+
+
+2005-09-27(Tue) 16:45:20 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : remove debug print
+
+
+2005-09-27(Tue) 16:41:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : support Kernel.local_variables
+
+	* parse.y, yarvcore.c : move some functions
+	(rb_(backref|lastline)_(get|set)) from parse.y to yarvcore.c
+
+	* yarvcore.h : fix typo of YARV_PREVIOUS_CONTROL_FRAME
+
+
+2005-09-26(Mon) 18:51:29 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c, compile.c, parse.y, vm.c, yarvcore.h :
+	 eval() works with binding (Env)
+
+	* vm.c : add th_set_eval_stack
+
+	* yarvtest/test_syn.rb : remove an assert "defined?(local_var)"
+
+
+2005-09-25(Sun) 19:30:59 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_vm2_send.rb : added
+
+	* common.mk : add rule "bench-item"
+
+	* eval_intern.h : add PASS_PASSED_BLOCK()
+
+	* eval_proc.c : support some functions
+
+	* rb/mklog.rb : added
+
+	* vm.c : fix prototype style and coding style
+
+	* yarv.h : add some prototypes of functions
+
+	* yarvcore.c, yarvcore.h, eval.c : yarv_thread_t#ifuncnode -> passed_block,
+	and add yarv_proc_t#safe_level
+
+
+2005-09-25(Sun) 11:01:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* some files : import from ruby 1.9.0 (2005-09-25)
+
+	* eval*, vm.c, vm_macro.def : remove frame, scope, ...
+
+	* yarvcore.c : remove yarv_block_given_p()
+
+	* yarvcore.h, insnhelper.h : move some macro from insnhelper.h to yarvcore.h
+	to use these in eval.c
+
+
+2005-09-24(Sat) 15:51:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval* : remove dependency to ruby_dyna_vars and ruby_class
+
+
+2005-09-23(Fri) 20:39:14 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval_*.[ch] : split eval.c to some files
+
+	* *.[ch] : import ruby 1.9.0 (2004-09-23)
+
+	* parse.y : remove dependency to ruby_dyna_vars and ruby_scope
+
+
+2005-09-15(Thu) 16:51:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, yarvcore.h : fix "for" scope
+
+	* yarvtest/test_block.rb : add tests for above
+
+
+2005-09-14(Wed) 06:11:43 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, vm_evalbody.h, vm.h, vm_dump.c,
+	compile.c, yarvcore.c : use #ifdef insted of #if for recognize
+	vm options
+
+	* vm_opts.h : fix default options
+
+
+2005-09-10(Sat) 14:10:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_opts.h : added
+
+	* yarvcore.h, rb/insns2vm.h : use vm_opts.h
+
+
+2005-09-10(Sat) 04:53:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, insns.def, compile.c : add DEFINED_YIELD
+
+	* yarvtest/test_yield.rb : add test_1_ary_and_1_params
+
+	* insns.def : fix splat and svalue
+
+	* vm.c : fix to perform with proc with ifunc (incomplete)
+
+	* sample/test.rb : added (comment out unsupported features)
+
+	* common.mk : add rule "runtest"
+
+
+2005-09-09(Fri) 19:32:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c : add splatarray
+
+	* yarvtest/test_massign.rb : add tests for above
+
+
+2005-08-31(Wed) 22:55:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c (yarvcore_eval_parsed): fix to return value
+
+	* yarv_version.h, Changes : 0.3.1
+
+
+2005-08-20(Sat) 10:19:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/ir.rb : add some check
+
+	* import today's ruby HEAD
+
+
+2005-08-18(Thu) 23:29:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : fix object file extension
+
+	* rb/ir.rb : added (import ruby script)
+
+	* rb/diff.rb : removed
+
+	* import today's ruby HEAD
+
+
+2005-08-18(Thu) 12:59:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk : rule test -> test2, test1 -> test
+
+	* compile.c : fix when clause bug and splat arugment
+
+
+2005-08-17(Wed) 05:22:31 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix block local parameter setting routine and support
+	massign in block parameter initialze
+
+	* yarvtest/test_yield.rb : add tests for above
+
+	* insns.def, compile.c : support array concat (ex: "[x, *y]")
+
+	* yarvtest/test_bin.rb : add tests for above
+
+
+2005-08-16(Tue) 19:51:19 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support nested massign
+
+	* yarvtest/test_massign.rb : add tests for above
+
+
+2005-08-16(Tue) 10:25:29 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : support rb_yield_0 with 0 args
+
+
+2005-08-16(Tue) 09:09:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* lib/fileutils.rb : imported
+
+	* insns.def : fix yield argument (same as last commit)
+
+	* yarvtest/test_yield.rb : add tests for above
+
+
+2005-08-16(Tue) 08:29:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : fix to support rb_yield_0 with multiple values
+
+	* common.mk : add parse, run1p ruelse
+
+	* compile.c : support yield with ARGSCAT/SPLAT
+
+	* vm.c, insns.def : fix yield arguments to do compatible behaviour
+
+	* yarvtest/test_yield.rb : added for above
+
+
+2005-08-16(Tue) 06:00:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix to set klass_nest_stack on singleton
+	method definition
+
+	* yarvtest/test_method.rb : add a test for above
+
+
+2005-08-16(Tue) 05:34:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test1.rb : added.  gdb and run1 rule run this script
+
+	* compile.c : fix error handled variable access
+
+	* yarvtest/test_exception.rb : add tests for above
+
+
+2005-08-16(Tue) 04:26:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* base ruby : ruby 1.9.0 (2005-08-15)
+
+
+2005-08-16(Tue) 03:54:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* common.mk, Makefile.in : move some rules to common.mk
+
+	* rb/diff.rb : added
+
+	* yarvtest/yarvtest.rb : fix to compare output last value
+
+
+2005-08-15(Mon) 18:27:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* Changes : 0.3.0
+
+
+2005-08-15(Mon) 17:56:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : fix to add prototype
+
+	* all files : propset svn:eol-style native
+
+
+2005-08-15(Mon) 10:48:53 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* eval.c : support rb_load
+
+
+2005-08-15(Mon) 09:42:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : define SDR()
+
+	* vm_dump.c : stack_dump_raw() -> vm_stack_dump_raw()
+
+	* yarvtest/yarvtest.rb : add rite test scheme
+
+	* benchmark/run_rite.rb : added
+
+	* yarvcore.c, inits.c : add Init_vm()
+
+	* yarv.h : add some prototype declarations, GET_THREAD()
+
+	* eval.c : remove unused functions
+
+	* eval.c : support Kernel.eval, some schemes (same as evalc.patch)
+
+
+2005-08-15(Mon) 00:53:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarv_version.h : move configurations to yarvcore.h
+
+	* yarvcore.c : remove VALUE yarv_get_current_running_thread() and
+	add yarv_thread_t *yarv_get_current_running_thread(), ...
+
+	* yarvcore.h : yarv_thread_t#vm -> vm_value
+
+	* compile.c : fix "break from nested classes"
+
+	* yarvext/extconf.rb : use have_func instead of defined?(YARV_PACHED)
+
+	* depend : fix pass
+
+	* eval.c : change to kick VM
+
+	* version.c : fix to show yarv version
+
+	* common.mk : fix dependent
+
+	* inits.c : fix to kick Init_yarvcore
+
+
+2005-08-14(Sun) 02:05:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* README : add description
+
+	* yarvext/depend : move to topdir/depend
+
+2005-08-14(Sun) 01:50:43 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* merge yarv to ruby (prepare)
+
+	* make yarvext/ to build as extension
+
+
+2005-08-13(Sat) 09:36:26 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* evalc.patch, insns.def, compile.c : fix to support current
+	ruby HEAD.
+
+	* 0.2.3
+
+
+2005-08-08(Mon) 19:13:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h, Changes : 0.2.2
+
+
+2005-08-08(Mon) 17:17:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.h, vm.c, insns.def, yarvcore.h, yarvcore.c :
+	remove yarv_iseq_t#iseq_dt and add yarv_iseq_t#encoded.
+	use yarv_iseq_t#encoded anytime
+
+	* vm_evalbody.h, vm.h, extconf.rb, version.h :
+	support call threaded code (incomplete)
+
+
+2005-08-01(Mon) 05:26:12 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : support yield with multiple values
+
+	* compile.c : fix dynavars
+
+	* yarvcore.h : fix to mark defined method
+
+
+2005-07-31(Sun) 23:27:24 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, vm.c, insns.def : fix search object path
+
+	* compile.c : fix "for" statement
+
+	* vm_macro.def : fix rest, opt arguments
+
+
+2005-07-31(Sun) 14:52:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm_macro.def : fix block parameter
+
+	* compile.c : fix to unuse compile_data->in_ensure
+
+	* insns.def : add orphan check when return
+
+
+2005-07-31(Sun) 03:25:05 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, compile.c, yarvcore.h, insns.def :
+	support jump from rescue/ensure/class/module
+
+	* test/test_flow.rb : add tests for above fix
+
+
+2005-07-30(Sat) 04:44:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : struct iseq_compile_data_ensure_node_stack is added
+
+	* compile.c : insert ensure clause before break/next/redo
+
+	* vm.c : fix return/break handling
+
+	* yarv.h, vm.c : fix lightweight yield
+
+	* vm.c, insns.def, vm_macro.def : change arguments of th_set_env (add sp)
+
+	* test/test_flow.rb : added
+
+	* test/yarvtest.rb : add ae_flow
+
+	* compile.c, vm_macro.def : add tail-call/tail-recursion optimization
+	(experimental)
+
+
+2005-07-29(Fri) 20:14:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : make_name_for_block and make_name_with_str
+	are added
+
+	* insns.def : fix if unmatched size arg size to yield
+
+	* test/test_block.rb : add test for above fix
+
+	* vm.c : add th_backtrace_each and fix backtrace notation
+
+	* yarvcore.c : set top level iseq name to "<main>"
+
+
+2005-07-29(Fri) 13:20:19 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : fix yarv_iseq_t to pass VC (cl)
+
+	* vm_dump.c : ditto
+
+	* compile.h : ditto
+
+	* insnhelper.h : ditto
+
+	* vm_evalbody.h : include 'math.h'
+
+	* insns.def, vm.c : raise error when yield without block
+
+	* vm.c : implement thread_backtrace
+
+	* vm.c, yarvsubst.c, yarv.h : implement thread_yield_light_prepare and
+	thread_yield_light_invoke
+
+	* yarvcore.c : Integer#times uses yarv specific version
+
+
+2005-07-28(Thu) 21:35:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : add another mark function for thread/stack
+
+	* vm_evalbody.h : fix register allocation for x86_64
+
+	* vm.h : use asm for tc on x86_64
+
+
+2005-07-28(Thu) 20:17:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : add mark/free message to debug gc
+
+	* insnhelper.h, insns.def, vm_macro.def : remove and
+	add new RESTORE_REGS
+
+	* vm_evalbody.h : fix register allocation
+
+
+2005-07-28(Thu) 02:00:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, etc : change VM stack structure. re-write all
+	vm functions to do it
+
+	* vm_macro.def : added
+
+
+2005-07-08(Fri) 01:36:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : don't use fmod on AMD64
+
+
+2005-07-08(Fri) 00:14:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* Changes : added
+
+
+2005-07-07(Thu) 23:54:37 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h : 0.3.0
+
+
+2005-07-07(Thu) 23:52:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* 0.2.1 : released
+
+
+2005-07-07(Thu) 23:50:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h : 0.2.1
+
+
+2005-07-07(Thu) 23:47:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb, extconf.rb : add --[enable|disable]-opt-unify-all-combination
+	and --disable-opts
+
+	* vm.h : DISPATCH_ARCH_DEPEND_WAY is only enabled on GCC 3.x
+
+
+2005-07-06(Wed) 13:20:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* depend, rb/eval.rb : add ITEMS option to benchmark rule
+
+	* benchmark/* : changed
+
+	* benchmark/other-lang/* : added
+
+
+2005-07-04(Mon) 04:02:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h. yarvcore.c : add idDIV, idMOD, idEq, idLength
+
+	* compile.c, insns.def : add specialized insn for above method id
+
+	* test/test_bin.rb : add tests for above
+
+
+2005-07-03(Sun) 20:31:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, yarvcore.h : remove cYarvThrowObject (unused)
+
+	* yarvcore.c, yarvcore.h, insns.def :
+	thread_object#stack_mark_poinetr
+
+	* depend, rb/eval.rb : BOPT, TOPT -> OPT
+
+
+2005-07-03(Sun) 13:53:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, compile.h : INSN_OBJECT, LABEL_OBJECT -> INSN, LABEL,
+	ISEQ_LINK_ELEMENT, ISEQ_LINK_ANCHOR -> LINK_ELEMENT, LINK_ANCHOR,
+	and some fixes
+
+	* tmpl/optinsn.inc.tmpl : ditto
+
+	* yarvcore.c, yarvcore.h : remove label_object, insn_object
+	prepare_iseq_build, cleanup_iseq_build are added
+
+	* insns.def : remove unused variable from send
+
+
+2005-07-02(Sat) 04:19:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : add GC protect for opt_aset
+
+
+2005-07-02(Sat) 03:49:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb : add option -fno-reorder-blocks to vm.asm rule
+
+	* insns.def : fix opt_aset bugs
+
+	* test/test_bin.rb : add tests for aset, aref
+
+
+2005-07-02(Sat) 03:05:12 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb : fix output
+
+	* vm_evalbody.h : add register for x86_64
+
+	* rb/asm_parse.rb : fix to shor size and length
+
+
+2005-07-02(Sat) 02:56:31 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : move specialized instruction point (new_insn_send)
+
+	* insns.def : add opt_aref, opt_aset
+
+
+2005-07-01(Fri) 11:04:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.h : fix to pass VALUE type to new_insn_body
+
+	* insnhelper.h : add cast
+
+	* compile.c : fix getdynamic argument (0 == Qfalse -> I2F(0))
+
+
+2005-06-30(Thu) 23:34:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/eval.rb : add and fix some rules
+
+	* rb/insns2vm.rb : generate all
+
+	* benchmark/run.rb : add -r (ruby only) option
+
+
+2005-06-30(Thu) 23:25:23 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* tmpl/vmtc.inc.tmpl : add const prefix
+
+	* /rb/asm_parse.rb, extconf.rb : added and make assembler analised output
+
+	* opt_operand.def : add send operands unification
+
+	* insnhelper.h : add HEAP_CLASS_OF(obj)
+
+	* insns.def : fix opt_plus, opt_ltlt
+
+	* vm_evalbody.h : move _tag
+
+	* benchmark/run.rb : fix file select
+
+
+2005-06-30(Thu) 06:07:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb : add collect-usage-analysis option
+
+	* opt_operand.def, opt_insn_unif.def : add some rules
+
+
+2005-06-29(Wed) 23:28:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, extconf.rb, vm.h, compile.c :
+	DISPATCH_DIRECT_THREADED_CODE, DISPATCH_THREADED_CODE
+	 -> OPT_DIRECT_THREADED_CODE,  OPT_INDIRECT_THREADED_CODE.
+	if at least one of then is defined, OPT_THREADED_CODE is defined
+
+	* benchmark/* : fix name and parameters
+
+	* rb/eval.rb : added for YARV evaluation
+
+
+2005-06-29(Wed) 16:16:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb : fix output format
+
+	* call_cfunc.inc -> call_cfunc.h
+
+	* vm.h : add sign by asm statement
+
+
+2005-06-28(Tue) 22:28:40 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : fix method search
+
+
+2005-06-28(Tue) 22:26:34 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb : fix options
+
+
+2005-06-28(Tue) 21:50:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb : fix output format
+
+
+2005-06-28(Tue) 21:34:54 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* depend : add option TOPT to test rules
+
+	* benchmark/run.rb : fix output format
+
+
+2005-06-28(Tue) 21:15:54 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix opt_case_dispatch instruction
+
+	* benchmark/run.rb : output all usertimes when exit benchmark
+
+
+2005-06-28(Tue) 20:35:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb, compile.c, tmpl/optinsn.inc.tmpl, vm.c :
+	change extconf options
+
+2005-06-28(Tue) 13:20:59 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb : add -y, --yarv-only option
+
+	* depend : add BOPT to tbench rule
+
+
+2005-06-27(Mon) 23:31:12 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* depend : add gdb rule
+
+	* vm.h : use inline assembler for x86 (to support gcc 3.4.x)
+
+
+2005-06-27(Mon) 20:04:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, compile.c, disasm.c : remove unused variables
+
+	* vm.h, insnhelper.h, debug.h : fix to reduce warning
+
+	* vm.c, vm_dump.c : move VM state dump (debug) functions to vm_dump.c
+
+	* depend : adde reconf rule
+
+	* insnhelper.h :
+
+	* vm_evalbody.inc : rename to vm_evalbody.h
+
+
+2005-06-27(Mon) 16:50:31 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns2vm.rb : fix generating unif insn
+
+	* compile.c : add useless pop/swap insn elimination with stack caching
+
+	* depend : remove compiled.o dependency
+
+
+2005-06-26(Sun) 14:06:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb : use tmpfile instead of popen
+
+	* rb/insns2vm.rb : fix generating insn unification logic
+
+	* opt_insn_unif.def : add some unification rules
+
+	* compile.c : add verify_list function and fix unification logic
+
+
+2005-06-22(Wed) 12:58:26 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, yarvcore.c, insns.def,  compile.c : add mult optimization
+
+	* test/test_bin.rb : add test_fact
+
+
+2005-06-21(Tue) 22:34:07 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, compile.[ch], tmpl/optinsn.inc.tmpl, rb/insns2vm.rb :
+	change data structure (don't use Ruby's array to represent a
+	instruction sequence)
+
+	* disasm.c : add separator
+
+
+2005-06-14(Tue) 07:48:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support "for" statement
+
+	* test/test_block.rb : add test for above
+
+	* yarvcore.[ch] : add global id idEach
+
+
+2005-06-08(Wed) 22:30:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : add if/unless(L1) jump (L2) :L1 => unless/if(L2)
+	optimize (condition reversal) and fix typo
+
+
+2005-06-07(Tue) 08:29:41 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : fix to remove compiler warning
+
+	* version.h : 0.2.1
+
+
+2005-06-07(Tue) 08:16:22 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : iseq_link_element changed to double linked list
+
+	* disasm.c : support dump struct iseq_link_element
+
+	* compile.c : use double linked list instead of array
+	for intermediate representation
+
+
+2005-06-06(Mon) 15:38:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, yarvcore.c : add link structure to insn and label object
+
+	* compile.h, compile.c : remove some variables in function top scope
+	of iseq_compile_each and some optimization (now working)
+
+
+2005-06-04(Sat) 16:12:59 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix previous commit
+
+
+2005-06-04(Sat) 15:56:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix stack caching (after jump state)
+
+
+2005-06-04(Sat) 09:12:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix some point for previous commit
+
+
+2005-06-04(Sat) 07:31:21 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : optimize case/when statement
+	(dispatch on constant time)
+
+	* yarvcore.h, disasm.c, rb/insns2vm.rb : fixed for above
+	(CDHASH)
+
+	* test/test_syn.rb : add test for above
+
+
+2005-06-04(Sat) 03:41:29 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, yarvcore.c : add some temporary variable test
+	(it'll be vanished)
+
+	* compile.c : NODE_CASE optimize (use topn instead of dup/swap)
+
+
+2005-06-03(Fri) 00:54:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : apply flow optimization for while/until statement
+
+
+2005-03-04(Fri) 19:34:32 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb : fix category (comment)
+
+	* depend : remove space between target name and colon
+
+
+2005-03-04(Fri) 15:55:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* tmpl/yarvarch.ja : fix typo
+
+
+2005-03-04(Fri) 13:30:19 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* depend : add a rule for jitcompile.o
+
+	* vm.h : fix a macro argument
+
+	* version.h : 0.2.0
+
+
+2005-03-03(Thu) 08:35:14 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb : remove vm_evalbody.inc call_cfunc.inc from clean target
+
+
+2005-03-03(Thu) 00:54:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* tmpl/insns.inc.tmpl : fixed typo
+
+	* insns.def : store th->pc to current pc
+
+
+2005-03-03(Thu) 00:31:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* tmpl/yarvarch.ja, doc/yarv.rb : write current architecture of yarv
+
+
+2005-03-01(Tue) 13:50:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c (yarvcore_eval_parsed) : added
+	(separeted from yarvcore_eval)
+
+	* yarvcore.c, compile.c : iseq_translate_direct_threaded_code
+	is moved to compile.c
+
+	* depend : add rule for yasmdata.rb
+
+	* rb/yasm.rb : support top-level and method-level assemble
+
+
+2005-02-26(Sat) 08:09:57 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb, compile.c, vm.h : change type long to OFFSET
+
+	* tmpl/yasmdata.rb.tmpl : added
+
+	* rb/insns2vm.rb : add yasmdata_rb method
+
+	* rb/yasm.rb : fix some interface (incomplete)
+
+	* compile.c : iseq_setup added
+
+	* yarvcore.c : YARVCore::InstructionSequence::Instruction#make added
+
+
+2005-02-24(Thu) 07:45:37 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/yasm.rb : added
+
+
+2005-02-24(Thu) 01:13:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : remove useless statements
+
+
+2005-02-24(Thu) 00:46:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb (InsnInfo) : add @is_sc attr and remove
+	is_sc method
+
+	* compile.c : fix NODE_CASE/NODE_WHEN bug (cond at 'when'
+	must not be popped)
+
+	* compile.c : support NODE_OP_ASGN1 to &&= and ||=
+
+	* test/test_bin.rb : add tests for above
+
+
+2005-02-23(Wed) 09:17:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, yarvcore.c : thread_svar added and fix svar location
+
+
+2005-02-21(Mon) 08:38:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h : make type "struct iseq_compile_data"
+
+	* yarvcore.h : iseq_object#insn_info_ary to iseq_object#insn_info_tbl
+
+
+2005-02-21(Mon) 05:24:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c (compile_string) : remove null check of node
+
+
+2005-02-19(Sat) 03:52:45 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h : 0.1.1
+
+
+2005-02-18(Fri) 20:57:18 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, yarvcore.c : add idLTLT, idMethodMissing
+
+	* compile.c : suopport lval (or others) block parameter
+
+	* test/test_block.rb : add tests for above
+
+	* insns.def (send) : support method_missing
+
+	* test/test_method.rb : add tests for above
+
+	* insns.def : opt_ltlt and
+
+
+2005-02-18(Fri) 08:54:40 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/runc.rb : added
+
+	* benchmark/contrib/pentomino.rb : added opt_ltlt
+	and Float, String plus specialization
+
+
+2005-02-18(Fri) 07:49:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : remove debug print
+
+	* rb/aotcompile.rb : skip if yarvcore.so is not created
+
+
+2005-02-18(Fri) 06:46:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix block passing
+	and block argument
+
+
+2005-02-18(Fri) 05:52:41 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : thread_get_ev_const, thread_get_ev_defined is added
+	(separated from insns.def)
+
+	* insnhelper.h : GET_EV_KLASS(klass) is added
+	(separated from insns.def)
+
+	* yarvcore.h, insns.def, compile.c : support defined? expression (limited)
+
+	* test/test_syn.rb : tests for above is added
+
+	* compile.c, insns.def : support block passed method dispatch
+
+	* test/test_method.rb : tests for above is added
+
+	* compile.h : CALL_ARGS_SPLAT is removed
+
+
+2005-02-16(Wed) 13:32:37 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* disasm.c : fix ID to String method
+
+	* compile.c : NODE_SUPER, NODE_ZSUPER check 'popped'
+	and NODE_RETURN check outer type
+	and NODE_DREGX_ONCE supported (temporarily)
+
+	* test/test_syn.rb : add a test
+
+	* test/test_jump.rb : add a test
+
+
+2005-02-16(Wed) 06:07:41 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.[hc] : use Symbol instead of Fixnum to represent ID
+
+	* rb/insns2vm.rb : add attr_reader :insns, :insn_map
+
+	* vm.h, rb/insns2vm.rb : END_INSN have one arg
+
+	* jitcompile.c : jit compiler framework (experimental)
+
+	* rb/aotcompile.rb : refactoring
+
+	* compiled.c : add constant pool
+
+	* vm_evalbody.inc, call_cfunc.inc, vm.c : separeted from vm.c
+
+	* insns.def : fix return val
+
+	* depend : add rules for compiled.o
+
+
+2005-02-14(Mon) 13:09:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insnhelper.h, yarvcore.h: move YARV_METHOD_NODE to yarvcore.h
+
+	* yarvcore.h : add 2 members jit_compiled and iseq_orig
+	to struct iseq_object
+
+	* yarvcore.c : add yarv_jitcompile and global function jitcompile
+
+	* insns.def : insn opt_call_native_compiled added
+
+	* jitcompile.c : added
+
+
+2005-02-12(Sat) 05:38:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (putstring) : fixed to duplicate string object
+
+	* rb/insns2vm.rb, tmpl/optunifs.inc.tmpl, compile.c : support
+	instructions unification (aka super instruction)
+
+	* opt_insn_unif.def : added for above
+
+	* benchmark/bm_unif1.rb : added to measure efficiency of unification
+
+	* depend : fixed for above
+
+	* extconf.rb : add option --(enable|disalbe)-opt-insns-unification
+
+
+2005-02-11(Fri) 12:14:39 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, vm.c, insns.def : permit to access svar from
+	cfunc environment
+
+	* test/test_method.rb : add tests for above
+
+
+2005-02-09(Wed) 19:31:06 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* ite.rb : added (ruby -rite [script file])
+
+
+2005-02-09(Wed) 02:25:43 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.[hc] : add member compile_data (hash) to iseq_object
+
+	* compile.c, yarvcore.h : check label is already set
+
+	* compile.c, extconf.rb : support __goto__ and __label__ statement
+
+
+2005-01-25(Tue) 12:49:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_block.rb : add break test to test_times
+
+
+2005-01-25(Tue) 03:34:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb : check ruby version if yarv patch is applied or not
+
+	* evalc.patch : fixed for rb_call_super and above check
+
+
+2005-01-25(Tue) 03:21:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb : refactoring (mainly, make InsnsDef::InsnInfo
+	to represent each instruction information)
+
+	* depend, rb/makedocs.rb : fixed for above
+
+	* yarvcore.c (thread_call_super) : added
+
+	* vm.c (thread_call_super) : added
+
+	* vm.h : add struct cmethod_info
+
+	* insns.def, vm.c : use cmethod_info to represent C method info
+
+	* insns.def : use iseq_object#klass_nest_stack
+	to search super/zsuper's class
+
+	* prosym.rb : removed
+
+	* ToDo : write todo things on wiki
+
+
+
+2005-01-18(Tue) 23:44:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb : check ENV['RUBY'] to use ruby binary
+
+
+2005-01-10(Mon) 08:44:40 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h : 0.1.0
+
+
+2005-01-09(Sun) 22:01:29 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* repository : svn propset svn:eol-style native *.c *.h tmpl/*.tmpl
+
+
+2005-01-09(Sun) 21:48:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : FREE_UNLESS_NULL, MARK_UNLESS_NULL macros are added
+
+	* yarvcore.c : some insn/label methods are added
+
+	* yarvcore.h : add structure menber "insns_ary" to iseq_object
+
+	* vm.c, insns.def (thread_eval_body) : return values with throw
+
+	* prosym.rb : added
+
+	* insns.def : add YARV_AOT_COMPILED and some procedure
+
+	* depend : add compiled.c
+
+	* compiled.c : added to build compiled Ruby program (C source)
+	by AOT compiler
+
+	* rb/aotcompile.rb : AOT compiler
+
+	* aotct.rb, rb/aotctest.rb : test and benchmark AOT compiler
+
+	* rb/allload.rb : added
+
+
+2005-01-09(Sun) 08:30:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c (yarv_yield_values) : added
+
+	* vm.c (thread_call0) : change interface. substitute rb_call0 in
+	yarv environment
+
+	* yarvcore.c (yarv_call0) : fix for above
+
+	* yarvcore.c (yarv_call0_cfunc) : removed
+
+	* yarvcore.c : change passing items for yarv_setup
+
+	* evalc.patch : fix for above
+
+	* benchmark/bm_lists.rb : fix (unsupport block passing)
+
+	* benchmark/run.rb : use full path to ruby
+
+	* insns.def (yield): raise error if argc > expected argc
+
+
+2005-01-08(Sat) 16:07:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb : add descrioptions
+
+	* compile.c : fix bugs (getinlinecache operands)
+
+	* yarvcore.c : initial value of yarvGlobalStateVersion
+	to 1
+
+
+2005-01-08(Sat) 14:39:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, vm.c, evalc.patch : support making backtrace
+	(incompatible with current ruby interpreter)
+
+
+2005-01-08(Sat) 11:25:46 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* evalc.patch : commit for previous commit change
+
+	* yarvcore.h, compile.c, insns.def : MC to IC (inline cache),
+	and changed to using IC by set/getinlinecache
+
+
+2005-01-08(Sat) 10:04:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : add global variable sym[IC]FUNC
+
+	* yarvcore.c (yarv_iterate, yarv_call0_cfunc) : added
+	(each called from rb_iterate, rb_call0 with NODE_CFUNC)
+
+	* vm.c (stack_dump_raw) : fixed to prints more detail
+
+	* vm.c (stack_dump_th, stack_dump_thobj) : added to
+	dumps thread_object states (for VALUE, struct pointer)
+
+	* vm.c (thread_dump_regs) : added
+
+	* vm.c (thread_call0, thread_call0_cfunc, thread_invoke_yield,
+	thread_invoke_yield_cfunc), insns.def (yield, send) :
+	fixed, added to support IFUNC
+
+	* vm.c, yarvcore.c, insns.def : change type purpose
+	thread_object#block_ptr (it holds IFUNC block information,
+	so this type was changed to 'NODE *')
+
+	* vm.c (stack_dump_each) : fixed for above
+
+	* test/test_block.rb (test_ifunc) : test for above
+
+	* vm.c (get_block_objec, thread_make_env_object) : fixed bugs
+
+	* test/test_bin.rb (test_xstr) : remove `ls` test
+
+
+2005-01-06(Thu) 21:35:18 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarv : trying to support NODE_IFUNC (rb_iterate)
+
+
+2005-01-05(Wed) 06:50:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, insns.def, disasm.c, rb/insns2vm.rb, compile.[ch] :
+	support inline method cache
+
+	* extconf.rb : add -*-inline-method-cache (default: enable)
+
+	* test/test_method.rb : add a test for above
+
+	* benchmark/bm_poly_method.rb : added
+
+	* yarvcore.c : add option string
+
+
+2005-01-04(Tue) 17:15:41 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c : add compile_array and duparray insn
+	to optimize only literal array creation
+
+	* benchmark/bm_array.rb : added
+
+
+2005-01-04(Tue) 10:02:40 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* README : fix version
+
+
+2005-01-04(Tue) 09:57:25 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* ToDo : reflect current status
+
+
+2005-01-04(Tue) 09:43:54 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_VALUES, NODE_ARGSCAT, NODE_SPLAT
+
+	* test/test_massign.rb : add tests for above
+
+	* benchmark/bm_swap.rb : added
+
+
+2005-01-04(Tue) 06:25:45 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.h : COMPILE_ERROR break contol (instead of return)
+
+	* compile.c : support NODE_MASGN
+
+	* insns.def : change expandarray for massign and add topn insn
+
+	* test/test_massign.rb : added
+
+
+2005-01-03(Mon) 21:20:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : store block when create proc
+
+	* test/test_proc.rb : add a test for above change
+
+	* yarvcore.c : add global function "once"
+
+
+2005-01-02(Sun) 00:40:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_super.rb : fix bug (remove infinite loop)
+
+
+2005-01-01(Sat) 23:45:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_z?super.rb : added
+
+
+2005-01-01(Sat) 23:37:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bmx_so_object.rb : rename to benchmark/bm_so_object.rb
+
+
+2005-01-01(Sat) 23:19:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_OP_ASGN2, NODE_OP_ASGN_AND, NODE_OP_ASGN_OR,
+	NODE_SUPER, NODE_ZSUPER, NODE_MATCH
+
+	* insns.def : support super, zsuper (currently, super can't
+	handle with block)
+
+	* test/test_bin.rb : add test for op_asgin2, op_assgin_and/or
+
+	* test/test_class.rb : add test for super, zsuper
+
+
+2005-01-01(Sat) 20:39:29 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_MATCH
+
+	* yarvcore.c : fix yarv_svar bug (fix condition boundary)
+
+	* insnhelper.h : save cfp/lfp/dfp vars to thread_object (th)
+
+
+2005-01-01(Sat) 20:03:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h : 0.0.1
+
+	* yarvcore.h : add idIntern declaration
+
+	* insns.def : add getspecial, setspecial.
+	implement getclassvariable, setclassvariable.
+	store lfp before reg match (opt_regexpmatch1)
+
+	* compile.c : support ditto, flipflop
+
+	* yarvcore.c : support svar
+
+	* test/test_syn.rb : add test for flipflop
+
+	* test/test_bin.rb : add test for dsym, cvar, backref
+
+
+2005-01-01(Sat) 09:09:32 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : add getspecial insn
+
+	* compile.c : support NODE_NTH_REF, NODE_BACK_REF
+
+
+2005-01-01(Sat) 06:53:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c : support alias, undef
+
+	* test/test_method.rb : test for above
+
+	* rb/insns2vm.rb : fix enbug
+
+
+2005-01-01(Sat) 06:00:32 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_jump.rb : add test (next with value)
+
+	* yarvcore.h, yarvcore.c, compile.c, compile.h :
+	raise compile error exception instead of rb_bug
+
+	* yarvcore.c, evalc.patch : support "require"
+
+	* test.rb : restore $" after evaluation with ruby
+
+	* rb/insns2vm.rb : remove unnecessary each
+
+
+2004-12-17(Fri) 18:56:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix newhash
+
+
+2004-12-15(Wed) 13:29:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : add version string
+
+	* compile.c : fix rescure clause bug
+
+
+2004-12-14(Tue) 22:46:30 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : add reput insn
+
+	* vm.h : show stack cache registers when stack dump
+
+	* rb/insns2vm.rb, compile.c : fix stack caching bugs
+
+
+2004-12-14(Tue) 00:51:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns2vm.rb, compile.c, tmpl/opt_sc.inc.tmpl : fix bugs
+
+	* rb/mixc-asm.rb : added
+
+
+2004-12-14(Tue) 00:17:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, yarvcore.c, compile.c : fix SC bugs
+	(SC state management)
+
+	* extconf.rb : add option -[enable|disable]-opt-stack-caching
+
+	* insns2vm.rb : accept CPPFLAGS options
+
+	* vm.c : support restrore register for pc
+
+
+2004-12-13(Mon) 16:53:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb : add macro INSN_IS_SC()
+
+
+2004-12-11(Sat) 10:51:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, compile.c : support singleton method definition
+
+	* test/test_method.rb : add test for above
+
+
+2004-12-11(Sat) 03:17:54 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/*.rb : modify
+
+	* extconf.rb : add $cleanfiles
+
+
+2004-12-08(Wed) 13:01:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : change to disable stack caching
+
+
+2004-12-07(Tue) 19:37:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.rb : add default after
+
+	* insns.def : fix to work on stack caching
+
+
+2004-12-07(Tue) 15:07:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* depend : add some dependency to *.inc files
+
+	* vm.c : add "register" and asm("regname") descriptor
+
+	* rb/insns2vm.rb, compile.c : add stack caching support
+
+	* tmpl/opt_sc.inc.tmpl : added to above change
+
+	* rb/makedocs.rb : fix file path
+
+	* extconf.rb : fix option selection
+
+
+2004-12-06(Mon) 11:20:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* extconf.rb : add vm.asm target if compiler is gcc
+
+
+2004-12-06(Mon) 09:56:24 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.h : rename method_frame's member block to block_ptr
+
+	* extconf.rb : add "-fno-crossjumping" option when compiler
+	is gcc
+
+	* opt_operand.def : add unification insn send
+
+	* rb/insns2vm.rb : define symbol instead of declare const
+	variable (for more optmize on VC)
+
+	* insns.def : move enter point in send
+
+
+2004-12-06(Mon) 04:53:51 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, opt_operand.def, rb/insns2vm.rb, depend :
+	support operand unification
+
+
+2004-12-05(Sun) 03:16:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, insns.def : speed up throw/catch scheme
+
+
+2004-12-05(Sun) 01:47:05 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : fix catch handler bugs
+
+	* test/test_jump.rb : test_complex_jump added
+
+
+2004-12-03(Fri) 20:39:05 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/contrib/mcq.rb : added
+	(from URABE Syouhei)
+
+
+2004-12-03(Fri) 20:35:28 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : support break in rb_yield block
+
+
+2004-12-03(Fri) 14:26:35 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support block local variable in current
+	ruby specification (patche from Kent Sibilev)
+
+	* insns.def : support attr_* (patch from Kent Sibilev)
+
+
+2004-12-02(Thu) 21:04:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* opt_operand.def : added
+
+
+2004-12-02(Thu) 13:20:41 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c, vm.h, vm.c, insns.def, insnhelper.h, yarvutil.rb :
+	add usage analisys framework
+
+	* disasm.c : insn_operand_intern to separate function
+
+	* benchmark/run.rb : run each benchmark on another process
+
+
+2004-12-01(Wed) 10:26:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : yield check block is given
+
+	* benchmark/bm_lists.rb : rename to bmx_lists.rb
+	(because it's not work ... bug?)
+
+	* insns.def : opt_* support other type calc
+
+
+2004-11-30(Tue) 16:14:54 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_so_array.rb : added
+
+	* benchmark/bm_so_matrix.rb : added
+
+
+2004-11-30(Tue) 14:11:30 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/getrev.rb : added
+
+	* yarvcore.c : add YARVCore::REV, YARVCore::DATE constant
+
+
+2004-11-30(Tue) 13:05:42 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_OP_ASGN1 (incomplete)
+
+	* insns.def : add dupn
+
+
+2004-11-30(Tue) 08:52:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h : 0.0.0.f
+
+
+2004-11-30(Tue) 08:43:59 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_class.rb : add test_initialize and test_to_s
+
+	* yarvsubst.c : use rb_funcall instead of yarv_funcall
+
+	* evalc.patch : fix ruby's patch
+
+	* benchmark/bm_so_*.rb : change naming rule. "bm_so_*" from
+	language shootout
+
+	* depend : tbench target item is ITEM env val (default: bmx_temp)
+
+	* vm.c : show raw address if environment is in heap at dumping stack trace
+
+	* vm.c : thread_call0 added
+
+	* vm.c : fix thread_yield_light_invoke
+
+	* yarv.h, yarvcore.c : remove yarv_funcall
+
+
+2004-11-29(Mon) 11:37:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test/test_proc.rb : add test test_nestproc
+
+	* yarvsubst.c : comment out yarv_Array_each
+
+	* insns.def : restore lfp/dfp after call_cfunc
+
+	* vm.c : fix stack dump routine
+
+	* vm.c : implement thread_funcall (temporarily)
+
+	* yarv.h : add IS_YARV_WORKING(), SET_YARV_START(), SET_YARV_STOP()
+
+	* yarvcore.c : remove check with yarv_in_work
+
+	* evalc.patch : added
+
+
+2004-11-27(Sat) 00:19:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.c : free -> ruby_xfree
+
+
+2004-11-26(Fri) 02:11:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm,c : fix bug
+
+
+2004-11-22(Mon) 11:19:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/bm_ackermann.rb, bm_proc.rb, bm_simpleiter.rb,
+	bm_so_exception.rb, bm_wc.rb, wc.input added
+
+
+2004-11-22(Mon) 02:31:56 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_proc.rb : add some test
+
+	* yarvcore.c, vm.c : support yield in C method (as rb_yield)
+
+	* vm.c (thread_yield_light_(prepare|invoke)) : support lightweight
+	yield
+
+	* yarv.h : added
+
+	* yarvcore.c, yarv.h : support yarv_is_working, yarv_block_given_p,
+	yarv_yield, yarv_funcall (only dummy function)
+
+	* vm.c : thread_eval_body changed return value
+
+	* yarvsubst.c : added and add yarv_Integer_times, yarv_Array_each
+
+	* yarvcore.h : block_ptr is added to struct thread_object
+
+	* insns.def : pass block when C method call
+
+	* insnhelper.h : add GET_ISEQOBJ(cfp) macro
+
+
+2004-11-21(Sun) 07:25:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c : support Proc#call
+
+	* test/test_proc.rb : added
+
+
+2004-11-19(Fri) 18:04:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def, vm.c : support creating Proc object
+
+
+2004-11-15(Mon) 14:19:27 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def (send) : use clear_local_size to specify
+	clear local table vars.
+
+	* insns.def : block represent data shares lfp, dfp with frame data
+
+
+2004-11-13(Sat) 18:19:41 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, insns.def : add VM_CALL_ARGS_SPLAT_BIT and
+	VM_CALL_ARGS_BLOCKARG_BIT
+
+	* compile.c, compile.h : add ADD_SEND, ADD_SEND_R
+
+
+2004-11-10(Wed) 08:26:25 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* add "vm_" prefix to (block_object, proc_object, env_object)
+
+
+2004-11-03(Wed) 15:52:14 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, yarvcore.c, disasm.c, compile.c, insns.def, vm.c :
+	fix to move x86_64 (illegal cast, etc)
+
+
+2004-11-01(Mon) 04:45:54 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, compile.c, debug.c, version.h :
+	redesgin gc debug scheme (GC_CHECK())
+
+	* yarvcore.c : mark iseqobj->current_block on GC
+
+	* insns.def, compile.c : last "throw" in ensure/rescue block
+	use operand throwobj and before this insn, use "getdynamic 0, 1"
+
+	* benchmark/bm_temp.rb : move to benchmark bmx_temp.rb
+
+	* depend : change some targets
+
+
+2004-10-25(Mon) 19:57:58 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : push exception iseq to iseqobj->iseq_mark_ary
+	to mark for GC
+
+
+2004-10-10(Sun) 16:25:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : remove $_, $' area from method local frame
+	and provide that's special method local variables pointer(LFP[-1])
+
+	* disasm.c : change environment showing format
+
+	* yarvcore.(h|c) : add YarvProc, YarvEnv
+
+	* yarvcore.h : add arg_block field to iseq_object
+	and init -1 as default value
+
+
+2004-09-30(Thu) 19:50:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : support passing splat argument
+
+	* compile.c, insns.def : support rest argument
+
+	* compile.c, insns.def : support optional argument initializer
+
+	* test/test_method.rb : add tests for above
+
+
+2004-09-29(Wed) 10:50:03 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix rescue clause popped
+
+	* benchmark/bm_random.rb : move to benchmark/bmx_random.rb
+
+
+2004-09-29(Wed) 01:25:35 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* many many files: change stack frame design
+
+
+2004-09-16(Thu) 08:51:37 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, yarvcore.h : support 'return' from method
+	in ensure clause
+
+
+2004-09-13(Mon) 21:56:40 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support inline cache constant access
+	on NODE_COLON2, NODE_COLON3
+
+	* depend : add 'vtest' rule(verbose test)
+
+
+2004-09-13(Mon) 10:58:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, yarvcore.h : support redo/next/break in
+	while/until
+
+
+2004-09-13(Mon) 08:50:19 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_jump.rb : added(correctly)
+
+	* benchamark/bm_(ensure|rescue|simplereturn).rb added
+
+
+2004-09-12(Sun) 23:30:20 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_jump.rb : added
+
+	* insns.def, compile.c : add 'putnil' insn
+
+	* compile.c : use '===' when rescue check
+
+	* insns.def : remove 'rescuecheck' insn
+
+	* compile.c : support retry in begin/rescue clause
+
+	* ToDo : added
+
+
+2004-09-08(Wed) 12:34:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvcore.h, yarvcore.c : add idThrow*
+
+	* insns.def, compile.c, vm.c : support retry, break,
+	next, redo, return(incomplete)
+
+
+2004-09-03(Fri) 13:40:08 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : add nop after rescue body
+
+	* insns.def, vm.c : support stack rewind when thrown
+
+
+2004-09-01(Wed) 17:31:01 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_exception.rb : added
+
+
+2004-09-01(Wed) 13:15:14 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.c, insns.def : implementing exception handling
+
+
+2004-09-01(Wed) 00:18:54 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : add 'throw' insn
+
+	* compile.c : support 'rescue' and 'ensure' clause
+
+	* yarvcore.c, yarvcore.h : add 'catch_table' to iseq_struct
+
+
+2004-08-30(Mon) 19:06:12 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.h : NEW_ISEQOBJ don't pass self as parent
+
+	* compile.c : use NEW_CHILD_ISEQOBJ explicitly
+
+
+2004-08-29(Sun) 21:09:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : trying to implement rescue/ensure
+
+	* insns.def : fix yield bug(lfp, dfp link)
+
+
+2004-08-28(Sat) 13:52:15 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix dvar bug
+
+	* test/test_block.rb : add test
+
+	* insns.def, insnhelper.h : remove unused source code
+
+
+2004-08-28(Sat) 08:51:26 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_DASGN
+
+	* test/test_block.rb : add test
+
+
+2004-08-28(Sat) 08:13:04 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : support access to instance variable
+
+	* test/test_class.rb : add test of instance variable
+
+	* benchmark/bm_block.rb : added
+
+
+2004-08-28(Sat) 07:48:43 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_block.rb : fix block parameter name
+
+
+2004-08-28(Sat) 07:27:52 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c, insns.def : support method call with block
+	and yield and add some functions
+
+	* compile.c, insns.def : support dynavars accessor
+
+	* test/test_block.rb : added
+
+	* vm.c : fix block parameter stack dump
+
+
+2004-08-27(Fri) 23:56:47 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c(iseq_compile) : remove parameter iseqtype
+	(this information can access via self)
+
+2004-08-27(Fri) 17:13:35 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_bin.rb : add test(absolute path constant)
+
+	* yarvcore.h, compile.c(iseq_compile) : change parameter
+
+	* insns.def(classdef) : fix bug
+
+
+2004-08-27(Fri) 04:53:13 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : support setconstant, getconstant, classdef,
+	moduledef
+
+	* vm.h : fix debug levels and so on
+
+	* vm.h : foo_WORD -> foo_WC
+
+	* test/test_class.rb : added
+
+
+2004-08-25(Wed) 17:51:50 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix getconstant/setconstant/classdef
+
+
+2004-08-25(Wed) 14:27:10 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* debug.[ch] : added
+
+	* compile.c, disasm.c : use debug interface
+
+	* compile.c : support some nodes
+
+	* compile.c, rb/insns2vm.rb : remove TS_CPATH
+
+	* insns.def : modify classdef/moduledef/singletonclassdef
+	and add popcref
+
+	* and others...
+
+
+2004-08-18(Wed) 20:16:45 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix case/when statement with empty else clause
+
+	* insns.def : enable compile
+
+	* yarvcore.h : add class search path scheme
+
+	* test/test_syn.rb : add switch/case test case
+
+	* tmpl/yarvarch.ja : update documents
+
+
+2004-05-22(Sat) 01:30:44 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvutil.rb : add eval_in_wrap
+
+	* test/test_*.rb : change to use eval_in_wrap
+
+
+2004-05-20(Thu) 02:50:32 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* support global variables
+
+	* benchmark/bm_*.rb : add some benchmarks
+
+	* compile.c : support NODE_ATTRASGN
+
+	* compile.c : add debugi(...)
+
+
+2004-05-19(Wed) 23:19:38 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* test/test_method.rb : added
+
+
+2004-05-19(Wed) 22:56:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : fix typo
+
+	* benchmark/run.rb : sort benchmark order by filename
+
+	* extconf.rb : use --enable/disable-xxx
+
+	* version.h : ditto(don't touch to change yarv options)
+
+
+2004-05-19(Wed) 21:18:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* yarvutil.rb : added
+
+	* test.rb, test/*, benchmark/run.rb : use yarvutil.rb
+
+	* version.h : USE_OPTIMIZED_REGEXP_MATCH added
+
+	* yarvcore.h : add idEqTilde
+
+	* yarvcore.c(yarvcore_parse, yarvcore_eval) : require file and line
+	parameter
+
+	* test/test_bin.rb : add regexp test
+
+	* benchmark/bm_regexp.rb : added
+
+
+2004-05-19(Wed) 13:57:31 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : add compile_dstr(self, node)
+
+	* compile.c : support NODE_MATCH2, NODE_MATCH3, NODE_DXSTR
+
+	* insns.def : add toregexp
+
+
+2004-05-18(Tue) 10:12:20 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_XDSTR
+
+	* test/test_bin.rb : add test for above change
+
+
+2004-05-18(Tue) 09:46:33 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def(send) : store regs before call_cfunc
+
+
+2004-05-18(Tue) 08:55:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : support NODE_DSTR, NODE_EVSTR
+
+	* compile.c : support NODE_XSTR
+
+	* insns.def : add tostring operation
+
+	* rb/makedocs.rb : fix directory path
+
+	* depend : add tbench rule
+
+	* yarvcore.h : add 'exten ID idBackquote'
+
+
+2004-05-18(Tue) 00:09:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h : add USE_OPTIMIZED_BASIC_OPERATION
+
+	* yarvcore.h(struct thread_object) : add 'VALUE stat_insn_usage'
+
+
+2004-05-17(Mon) 11:28:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* version.h, insns.def, yarvcore.c : add FAKE_INLINE_METHOD_CACHE
+
+
+2004-05-17(Mon) 09:05:53 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* compile.c : fix generating opt_* insn process
+
+
+2004-05-17(Mon) 08:58:49 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/(bm_tarai.rb, bm_fib.rb) : added
+
+
+2004-05-17(Mon) 08:20:12 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/(bm_tak.rb, bm_reccount.rb) : added
+
+	* insns.def : test method cache(incomplete)
+
+	* insns.def : add expandarray insn
+
+	* yarvcore.c(iseq_init) : add parameter 'parent'
+
+
+2004-05-17(Mon) 01:49:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* benchmark/run.rb, bm_factorial.rb, bm_whileloop.rb : added
+
+	* insns.def(send) : set id to ruby_frame->orig_func
+
+	* check behavior on mswin32 and cygwin
+
+	* insns.def(send) : check stack overflow
+
+
+2004-05-16(Sun) 08:00:55 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* change frame structure(debugging)
+
+
+2004-05-14(Fri) 15:06:02 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns2vm.rb : support file name arguments
+
+
+2004-05-14(Fri) 04:33:09 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* insns.def : support (easy) constant
+
+
+2004-05-12(Wed) 01:51:48 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* rb/insns2vm.b : set directory prefix
+
+	* disasm.c : fix bug
+
+
+2004-05-12(Wed) 00:00:17 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.h, compiler.h, version.h : move *DEBUG defs to version.h
+
+
+2004-05-11(Tue) 23:00:11 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* vm.h, version.h, yarvcore.h : move gcc ver check to version.h
+	and include version.h from yarvcore.h
+
+
+2004-05-11(Tue) 19:16:26 +0900  Koichi Sasada  <ko1@atdot.net>
+
+	* 0.0.0.d : imported
+
+
+Local variables:
+add-log-time-format: (lambda ()
+  (let* ((time (current-time))
+	 (diff (+ (cadr time) 32400))
+	 (lo (% diff 65536))
+	 (hi  (+ (car time) (/ diff 65536))))
+  (format-time-string "%Y-%m-%d(%a) %H:%M:%S +900"  (list hi lo) t)))
+indent-tabs-mode: t
+tab-width: 8
+end:
diff --git a/doc/NEWS b/doc/NEWS
deleted file mode 100644
index b4445fa59f..0000000000
--- a/doc/NEWS
+++ /dev/null
@@ -1,837 +0,0 @@
-= command line options
-
-: -W option
-
-  new option to specify warning level. -W0 to shut up warnings, -W1 for normal level,
-  -W2 for verbose level.  -w equals to -W2.
-
-= language syntax
-
-: arbitrary delimited string array
-
-  %W(...) notation, word list literal like %w(...) with the
-  exception that #{} interpolation is allowed.
-
-: arbitrary delimited symbol literl
-
-  :"foo", :"foo#{bar}", etc.
-
-: expression interpolation in strings
-
-  Now arbitrary statements are allowed inside #{} interpolation
-  without escapes.  In other hand, they can no longer access to
-  variables defined in eval.
-
-: negative number literals
-
-  Digits preceded minus sign is a literal integer.
-
-: array expansion
-
-  Fixed with the following behavior:
-
-    a = *[1]
-    p a #=> [1]
-
-  Now 1-element array in rhs is expanded properly.
-
-    a = *[1]
-    p a #=> 1
-
-: break and next
-
-  Extended to take an optional expression, which is used as a value
-  for termination.
-
-: direct assignment to Foo::Bar is allowed
-
-  also, you can define "class Foo::Bar; end".
-
-= language core
-
-: $stdin, $stdout, $stderr
-
-  can be assignable again.  the original stdio are preserved as STDIN,
-  STDOUT, STDERR.
-
-: $VERBOSE now has 3 levels
-
-  nil - silence, false - medium (default), true - verbose
-
-: allocation framework
-
-  any instance of class can be allocated by class.allocate,
-  (except for a few classes).
-
-: comparison of exception classes in a rescue clause
-
-  changed to use Module#=== for comparing $! with the exception
-  class specified in each rescue clause.
-
-  as the previous behavior was to use kind_of?, the effect is limited
-  to the SystemCallError case.  SystemCallError.=== has been newly
-  defined to return true when the two have the same errno.  With this
-  change, SystemCallError's with the same errno, such as Errno::EAGAIN
-  and Errno::EWOULDBLOCK, can both be rescued by listing just one of
-  them.
-
-: constants lookup
-
-  improved at the performance of searching by using an internal hash
-  table.
-
-  calls const_missing method of the class/module, if constant is not
-  found in the look up path.
-
-: expression parenthesis in the first argument
-
-  altered to get the following code (note the space after p):
-
-    p ("xx"*2).to_i
-
-  Interpreted as:
-
-    p (("xx"*2).to_i)
-
-  Instead of:
-
-    (p("xx"*2)).to_i
-
-: implicit comparison in conditional expressions
-
-  Obsoleted except when it is used in -e.
-
-  : between Range and $.
-      Use explicit comparison instead.
-
-  : between Regexp and $_
-      Use the unary method ~/re/ instead.
-
-: to_str
-
-  added to get objects which define to_str() treated as String's.
-
-  now almost all the built-in methods try each argument with to_str()
-  when they expect it to be a String.
-
-    foo = Object.new
-    class <<foo
-      def to_str
-        "foo"
-      end
-    end
-    p File.open(foo)
-    => -:7:in `open': wrong argument type Object (expected String) (TypeError)
-       ruby 1.6.4 (2001-04-19) [i586-linux]
-    => -:7:in `open': No such file or directory - "foo" (Errno::ENOENT)
-       ruby 1.7.0 (2001-05-02) [i586-linux]
-
-: multiple assignment behavior
-
-  Fixed so that "*a = nil" results in "a == []".
-
-= changes in core class library
-
-: open
-
-  Extended so that when the third argument is permission flags it
-  calls open(2) instead of fopen(3).
-
-: sprintf
-
-  new format specifier "%p" is available.
-
-: lambda and proc
-
-  Proc object returns from these methods has the following attributes:
-
-  * strict argument number check
-  * break and return terminates the proc execution.
-
-: warn(message)
-
-  a method to give warnings.
-
-: abort()
-
-  takes optional terminate message argument.
-
-: Object#initialize_copy
-
-  copy constructor for clone and dup.
-
-: Object#instance_variable_set, Object#instance_variable_get
-
-  added.
-
-: Object#singleton_method_removed
-: Object#singleton_method_undefined
-
-  Added.
-
-: Array#transpose
-
-  added.
-
-: Array#fetch(index [, default])
-
-  Added.  If a default value isn't given, raises index error if index
-  is out of range.
-
-: Array#insert(n, other, ...)
-
-  Added. [ruby-talk:14289]
-
-  This is much the same as (({ary[n,0] = [other,...]})) except
-  returing self.
-
-    ary = [0,1,2,3]
-    ary[2, 0] = [4, 5, 6]
-    p ary
-
-    ary = [0,1,2,3]
-    ary.insert(2, 4, 5, 6)
-    p ary
-
-: Array#sort!
-
-  Changed to always return self without checking whether the sequence
-  of the elements was modified or not.
-
-  Beware that this behavior is not guaranteed to continue in the
-  future.  Do not rely on its return value. [ruby-dev:12506]
-
-: Array#filter
-
-  Previously deprecated, now removed.  Use Array#collect!.
-
-: Array#pack, String#unpack
-
-  Allows comment in template strings.
-
-: Array#pack, String#unpack
-
-  New templates 'q' and 'Q' for 64bit integer (signed and unsigned respectively).
-
-: Array#new
-
-  Now takes block to fill initial values.  E.g.
-
-	Array.new(10) { |i| i + 1 }
-	=> [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
-
-: Array#fill
-
-  Takes block to get the values to fill.
-
-: Array#fetch
-
-  Takes block to get the default value.
-
-: Array#zip
-
-  added.
-
-: Hash#update
-
-  Takes block to resolve key conflict.
-
-: Hash#merge and Hash#merge!
-
-  update hash.  Hash#merge! is a synonym of Hash#update.
-
-: String#split
-
-  if "sep" argument is a string, regular expression meta characters
-  are escaped internally.
-
-: String#rstrip
-
-  chop off NULs at the end of strings.
-
-: String#to_i
-
-  Now accepts optional base argument.
-
-	"101".to_i(10) => 101
-	"101".to_i(2)  => 5
-	"101".to_i(8)  => 65
-	"101".to_i(16) => 257
-
-  A base argument of 0 guesses at the base.
-
-	"101".to_i(0)   => 101
-	"0b101".to_i(0) => 5
-	"0101".to_i(0)  => 65
-	"0x101".to_i(0) => 257
-
-: String#[regexp, nth]
-
-  Extended to accepts optional second argument.
-
-  It tries match between self and REGEXP, then returns the
-  content of the NTH regexp register.
-
-: String#casecmp
-
-  Added.  This is a case insensitive version of String#<=>.
-
-: String#chomp
-
-  If $/ == "\n", chops off last newlines (any of \n, \r, \r\n).
-
-: String#eql?
-
-  Changed to be always case sensitive.
-
-: String#insert(n, other)
-
-  Added.
-
-  This is much the same as (({str[n, 0] = other})) except returing
-  self.
-
-: String#lstrip, rstrip, lstrip!, rstrip!
-
-  Added.  These strip only left or right part of a string.
-
-: String#match
-
-  Added.
-
-: String/Array methods
-
-  Returns an instance of receivers class.
-
-: String.new
-
-  The first argument becomes optional.
-
-: Symbol#intern
-
-  Added.
-
-: Symbol.all_symbols
-
-  Added. [ruby-dev:12921]
-
-: IO
-
-  64bit off_t support by Janathan Baker.
-
-: IO#read
-: IO#sysread
-
-  takes optinal second argument for read buffer.
-
-: IO::sysopen
-
-  New method to get a raw file descriptor.
-
-: IO#sysseek
-
-  Added.
-
-: IO#fsync
-
-  new method that copies all in-memory parts of a file to disk and
-  waits until the device reports that all parts are on stable storage.
-  Implemented with fsync(2) or equivalent.
-
-: IO.open
-
-  Made public.  Can only associate an IO object with a file number
-  like IO.new and IO.for_fd, but can take a block.
-
-: IO.for_fd
-
-  Added as a synonym for IO.new.
-
-: IO.read
-
-  Added.  Like IO.readlines, except it returns the entire file as a
-  string.  [ruby-talk:9460]
-
-: File#fnmatch, File::Constants::FNM_*
-
-  Added.  Refer to the fnmatch(3) manpage for details.
-
-  Localism is FNM_DOTMATCH which has the opposite meaning of the
-  commonly known FNM_PERIOD, which does not exist in Ruby.
-
-  e.g.
-
-    # exclude files matching "*.bak" case-insensitively.
-    files.reject! {|fn| File.fnmatch?("*.bak", fn, File::FNM_CASEFOLD) }
-
-: File.lchmod
-: File.lchown
-
-  Added.
-
-: File.open, IO.open
-
-  File mode can be specified by flags like open(2),
-  e.g. File::open(path, File::CREAT|File::WRONLY).
-
-: Regexp#options
-
-  Added.
-
-: Regexp.last_match(n)
-
-  Extended to take an optional argument.
-
-: MatchData#captures
-
-  added.
-
-: Dir#path
-
-  Added.
-
-: Dir.chdir
-
-  Extended to take a block.
-
-: Dir.glob
-
-  Made to support meta-character escaping by a backslash.  Wildcards
-  and spaces may now be escaped using a backslash.
-
-: Dir.open
-
-  Changed to return what the block returns when a block is given, just
-  as File.open does. (It always returned (({nil})) in 1.6 and
-  prior)
-
-: Dir.chdir
-
-  Changed to warn only when invoked from multiple threads or no block
-  is given. [ruby-dev:13823]
-
-    Dir.chdir('foo') {
-      Dir.chdir('bar') {   # previously warned
-        puts Dir.pwd
-      }
-    }
-
-: Dir#pos=
-
-  Returns the new position instead of self.
-
-: Dir::glob
-
-  Now accepts optional FNM_* flags via the second argument, whereas
-  Dir::[] doesn't.
-
-    Dir.glob("makefile", File::FNM_CASEFOLD) #=> ['Makefile', 'makefile']
-
-: Class#inherited
-
-  Method is called when Class is inherited by another class.
-
-	class A; end
-	def A.inherited(by)
-          puts "A inherited by #{by.inspect}"
-        end
-        class B < A; end
-
-        Prints out "A inherited by B"
-
-: Module#include?
-
-  Added. [ruby-dev:13941] 
-
-: Module#included
-
-  Added.  This is a hook called after Module#append_feature.
-
-: Module#method_removed
-: Module#method_undefined
-
-  Added.
-
-: Module.new, Class.new
-
-  Extended to take block.
-
-: Time
-
-  Extended to accept a negative time_t. (Only when the platform
-  supports it)
-
-    p Time.at(-1)
-    => Thu Jan 01 08:59:59 JST 1970
-
-: Time#to_a
-: Time#zone
-
-  Made to return "UTC" under gmtime.  It used to return a platform
-  dependent value, typically "GMT", in 1.6 and prior.
-
-: Marshal to use marshal_dump and marshal_load
-
-  if a dumping object responds to 'marshal_dump', Marshal.dump calls
-  it, and dumps object returned.  Marshal.load allocates a new instance
-  using "allocate", then calls its "marshal_load" with dumped data.
-  Marshal format version is now 4.8 (was 4.6 in 1.6.8).
-
-: Marshal
-
-  Fixed not to dump anonymous classes/modules.
-
-  Fixed with loading modules.
-
-: Thread#group
-
-  new method to get belonging  ThreadGroup.
-
-: Thread#terminate
-
-  synonym of Thread#exit
-
-: Thread#join
-
-  Optional argument limits maximum time to wait the thread in second.
-  And returns nil if timed out.
-
-: ThreagGroup#enclose
-
-  prohibits thread movement from/to enclosed groups.
-
-: Range#step([step=1])
-
-  Added.
-
-: SystemCallError
-
-  SystemCallError's "===" match (used in rescue also) is now based on its errno.
-
-: Interrupt
-
-  Made a subclass of SignalException. (It was a subclass of
-  Exception in 1.6 and prior)
-
-: NameError and NoMethodError
-
-  Moved and now NoMethodError < NameError < StandardError.
-
-: NoMethodError
-
-  Added. [ruby-dev:12763]
-
-: NotImplementError
-
-  Finally obsoleted.  Use NotImplementedError.
-
-: SystemCallError.===
-
-  Added. (See the "Comparison of exception classes in a rescue clause"
-  paragraph above) [ruby-dev:12670]
-
-: SystemExit#status
-
-  Added.
-
-: Proc#==
-
-  Added.
-
-: Method#==
-
-  Added.
-
-: UnboundMethod is no longer subclass of Method
-
-  class hierarchy changed. 
-
-: Enumerable#all?
-: Enumerable#any?
-: Enumerable#inject
-: Enumerable#sort_by
-
-  Added.
-
-: Math.acos(x)
-: Math.asin(x)
-: Math.atan(x)
-: Math.cosh(x)
-: Math.hypot(x,y)
-: Math.sinh(x)
-: Math.tanh(x)
-
-  Added.
-
-: Process.abort
-: Process.exit
-
-  synonym of Kernel#abort, and Kernel#exit respectively.
-
-: Process::detach(pid)
-
-  new method to detach child process.  child process will be "wait"ed
-  automagically.
-
-: Process.times
-
-  Moved from Time.times. (Time.times still remains but emits a
-  warning)
-
-: Process.waitall
-
-  Added.
-
-: Process::Status
-
-  Added.  (({$?})) is now an instance of this class.
-
-: Process::UID, Process::GID, Process::Sys,
-
-  Added.
-
-: Signal
-
-  Added. This module has module functions Signal.trap and Signal.list.
-
-= changes in bundled libraries
-
-: lib/cgi.rb
-
-  cgi[name] returns CGI::QueryExtension::Value that wraps string
-  value, no longer array.
-
-: lib/timeout
-
-  timeout "function" wrapped in Timeout module.
-
-: TCPServer#accept, UNIXServer#accept, Socket#accept
-
-  New methods to return an accepted socket fd.
-
-: Date and DateTime
-
-  lib/date.rb now provides both Date and DateTime.
-
-  Some methods have been renamed.  But the old names are still alive.
-
-  Some new methods have been added (Date::parse, Date#strftime, etc.).
-
-  Date#mjd now returns the chronological modified Julian day number.
-
-  All facilities about tjd have been removed.
-
-: Curses
-
-  Updated.  New methods and constants for using the mouse, character
-  attributes, colors and key codes have been added.
-
-: Net::HTTP
-
-  New version of Net::HTTP has introduced seriously incompatible
-  changes. For details, see document embedded in net/http.rb itself.
-
-: Socket.pack_sockaddr_in, Socket.unpack_sockaddr_in
-
-  Added.  Utility for direct Socket access.
-
-: Socket.pack_sockaddr_un, Socket.unpack_sockaddr_un
-
-  Added.  Utility for direct Socket access.
-
-: TCPServer#listen, UNIXServer#listen
-
-  Added.
-
-: TCPSocket.new
-: TCPSocket.open
-
-  Extended to take an address and a port number for the local side in
-  optional 3rd and 4th arguments.
-
-= newly bundled library
-
-: ext/bigdecimal
-
-  variable precision decimal number 
-
-: ext/dl
-
-  an interface to the dynamic linker.
-
-: ext/enumerator
-
-  a helper module for the Enumerable interface.
-
-: ext/io/wait
-
-  IO wait methods.
-
-: ext/iconv
-
-  wrapper library of (({iconv})).
-
-: ext/openssl
-
-  OpenSSL for Ruby
-
-: ext/racc/cparse
-
-  Racc runtime library in C. (Racc is a parser generator for ruby)
-
-: ext/stringio
-
-  Pseudo (({IO})) class from/to (({String})).
-
-: ext/strscan
-
-  Fast string scanner library.
-
-: ext/syck
-
-  fast YAML parser.
-
-: lib/abbrev
-
-  creates an abbreviation table from a list
-
-: lib/benchmark
-
-  Ruby scripts benchmarker
-
-: lib/cgi/session/pstore
-
-  cgi/session back-end using pstore 
-
-: lib/csv
-
-  reads/writes CSV files.
-
-: lib/date/format
-
-  strftime for Date class
-
-: lib/drb
-
-  dRuby or distributed Ruby
-
-: lib/fileutils
-
-  file utility library.
-
-: lib/generator
-
-  converts an internal iterator to an external iterator
-
-: lib/gserver
-
-  generic server used by xmlrpc
-
-: lib/ipaddr
-
-  manipulates IP address.
-
-: lib/multi-tk
-
-  to allow safe Tk, etc.
-
-: lib/open-uri
-
-  easy-to-use wrapper for net/http and net/ftp
-
-: lib/optparse
-
-  command line options utility library
-
-: lib/pathname
-
-  handles pathname in OO manner.
-
-: lib/pp
-
-  prettyprinter for Ruby objects
-
-: lib/prettyprint
-
-  implements prettyprint algorithm.
-
-: lib/profiler
-
-  library to implement -r "profile"
-
-: lib/racc/parser
-
-  RACC parser generator runtime in Ruby.
-
-: lib/scanf
-
-  scan string and retrieve object with format
-
-: lib/set
-
-  Set class
-
-: lib/runit
-
-  RubyUnit compatible layer for test/unit
-
-: lib/test/unit
-
-  unit testing framework for Ruby
-
-: lib/tmpdir
-
-  get temporary directory path.
-
-: lib/tsort
-
-  topological sorting library.
-
-: lib/rexml
-
-  REXML XML library
-
-: lib/webrick
-
-  generic internet server kit
-
-: lib/xmlrpc
-
-  simple RPC via XML
-
-: lib/un
-
-  used like 'ruby -run -e cp -- -p foo bar'.  neat, isn't it?
-
-: lib/win32/registry
-
-  win32/registry is registry accessor
-
-: lib/yaml
-
-  YAML Ain't Mark-up Language
-
-= removed libraries
-
-: lib/ftplib
-
-  use net/ftp instead.
-
-: lib/telnet
-
-  use net/telnet instead.
-
-= new port
-
-: WindowsCE port
-: Win32 BCC
-
-= interpreter implementation
-
-: garbage collector
-
-  faster, but uses more memory for the worst case.
-
-: string concatenation
-
-  faster by avoiding too frequent realloc(3).
diff --git a/doc/NEWS-1.8.7 b/doc/NEWS-1.8.7
new file mode 100644
index 0000000000..5da39ff265
--- /dev/null
+++ b/doc/NEWS-1.8.7
@@ -0,0 +1,669 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 1.8.7
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file.
+
+== Changes since the 1.8.6 release
+
+=== Configuration changes
+
+* default C flags
+
+  Some C compiler flags may be added by default depending on your
+  environment.  Specify optflags=.. and warnflags=.. as necessary to
+  override them.
+
+* vendor_ruby directory
+
+  A new library directory named `vendor_ruby' is introduced in
+  addition to `site_ruby'.  The idea is to separate libraries
+  installed by the package system (`vendor') from manually (`site')
+  installed libraries preventing the former from getting overwritten
+  by the latter, while preserving the user option to override vendor
+  libraries with site libraries. (`site_ruby' takes precedence over
+  `vendor_ruby')
+
+  If you are a package maintainer, make each library package configure
+  the library passing the `--vendor' option to `extconf.rb' so that
+  the library files will get installed under `vendor_ruby'.
+
+  You can change the directory locations using configure options such
+  as `--with-sitedir=DIR' and `--with-vendordir=DIR'.
+
+=== Global constants
+
+* new constants
+
+  * RUBY_COPYRIGHT
+  * RUBY_DESCRIPTION
+
+=== Library updates (outstanding ones only)
+
+* new library
+
+  * securerandom
+
+* builtin classes
+
+  * Array#flatten
+  * Array#flatten!
+
+    Takes an optional argument that determines the level of recursion
+    to flatten.
+
+  * Array#eql?
+  * Array#hash
+  * Array#==
+  * Array#<=>
+
+    Handle recursive data properly.
+
+  * Array#index
+  * Array#rindex
+
+    Use a given block if no argument is given.
+
+  * Array#collect!
+  * Array#map!
+  * Array#each
+  * Array#each_index
+  * Array#reverse_each
+  * Array#reject
+  * Array#reject!
+  * Array#delete_if
+  * Array#select
+
+    Return an enumerator if no block is given.
+
+    Note that #map and #collect still return an array unlike Ruby 1.9
+    to keep compatibility.
+
+  * Array#pop
+  * Array#shift
+
+    Take an optional argument specifying the number of elements to
+    remove.
+
+  * Array#choice
+  * Array#combination
+  * Array#cycle
+  * Array#drop
+  * Array#drop_while
+  * Array#permutation
+  * Array#product
+  * Array#shuffle
+  * Array#shuffle!
+  * Array#take,
+  * Array#take_while
+
+    New methods.
+
+  * Binding#eval
+
+    New method.
+
+  * Dir#each
+  * Dir#foreach
+
+    Return an enumerator if no block is given.
+
+  * Enumerable::Enumerator
+
+    New class for various enumeration defined by the enumerator library.
+
+  * Enumerable#each_slice
+  * Enumerable#each_cons
+  * Object#to_enum
+  * Object#enum_for
+
+    New methods for various enumeration defined by the enumerator library.
+
+  * Enumerable#count
+  * Enumerable#cycle
+  * Enumerable#drop
+  * Enumerable#drop_while
+  * Enumerable#find_index
+  * Enumerable#first
+  * Enumerable#group_by
+  * Enumerable#max_by
+  * Enumerable#min_by
+  * Enumerable#minmax
+  * Enumerable#minmax_by
+  * Enumerable#none?
+  * Enumerable#one?
+  * Enumerable#take
+  * Enumerable#take_while
+
+    New methods.
+
+  * Enumerable#find
+  * Enumerable#find_all
+  * Enumerable#partition
+  * Enumerable#reject
+  * Enumerable#select
+  * Enumerable#sort_by
+
+    Return an enumerator if no block is given.
+
+    Note that #map and #collect still return an array unlike Ruby 1.9
+    to keep compatibility.
+
+  * Enumerable#inject
+
+    Accepts a binary operator instead of a block.
+
+  * Enumerable#reduce
+
+    New alias to #inject.
+
+  * Enumerable#to_a
+
+    Can take optional arguments and pass them to #each.
+
+  * Hash#eql?
+  * Hash#hash
+  * Hash#==
+
+    Handle recursive data properly.
+
+  * Hash#delete_if
+  * Hash#each
+  * Hash#each_key
+  * Hash#each_pair
+  * Hash#each_value
+  * Hash#reject!
+  * Hash#select
+  * ENV.delete_if
+  * ENV.each
+  * ENV.each_key
+  * ENV.each_pair
+  * ENV.each_value
+  * ENV.reject!
+  * ENV.select
+
+    Return an enumerator if no block is given.
+
+  * GC.stress
+  * GC.stress=
+
+    New methods.
+
+  * Integer#ord
+  * Integer#odd?
+  * Integer#even?
+  * Integer#pred
+
+    New methods.
+
+  * Integer#downto
+  * Integer#times
+  * Integer#upto
+
+    Return an enumerator if no block is given.
+
+  * IO#each
+  * IO#each_line
+  * IO#each_byte
+  * IO.foreach
+  * ARGF.each
+  * ARGF.each_line
+  * ARGF.each_byte
+
+    Return an enumerator if no block is given.
+
+  * IO#bytes
+  * IO#chars
+  * IO#each_char
+  * IO#getbyte
+  * IO#lines
+  * IO#readbyte
+  * ARGF.bytes
+  * ARGF.chars
+  * ARGF.each_char
+  * ARGF.getbyte
+  * ARGF.lines
+  * ARGF.readbyte
+
+    New methods.
+
+  * Method#name
+  * Method#owner
+  * Method#receiver
+  * UnboundMethod#name
+  * UnboundMethod#owner
+
+    New methods.
+
+  * Module#class_exec
+  * Module#module_exec
+
+    New methods.
+
+  * Numeric#step
+
+    Return an enumerator if no block is given.
+
+  * Object#instance_exec
+  * Object#tap
+
+    New methods.
+
+  * ObjectSpace.each_object
+
+    Return an enumerator if no block is given.
+
+  * Process.exec implemented.
+
+  * Range#each
+  * Range#step
+
+    Return an enumerator if no block is given.
+
+  * Regexp.union accepts an array of patterns.
+
+  * String#bytes
+
+    New method
+
+  * String#bytesize
+
+    New method, returning the size in bytes. (alias length and size)
+
+  * String#chars
+  * String#each_char
+  * String#lines
+  * String#partition
+  * String#rpartition
+  * String#start_with?
+  * String#end_with?
+
+    New methods.  These are $KCODE aware unlike #index, #rindex and
+    #include?.
+
+  * String#each_byte
+  * String#each
+  * String#each_line
+  * String#gsub(pattern)
+
+    Return an enumerator if no block is given.
+
+  * String#upto
+
+    An optional second argument is added to specify if the last value
+    should be included.
+
+  * StopIteration
+
+    New exception class that causes Kernel#loop to stop iteration when
+    raised.
+
+  * Struct#each
+  * Struct#each_pair
+
+    Return an enumerator if no block is given.
+
+  * Symbol#to_proc
+
+    New method.
+
+  * __method__
+
+    New global function that returns the name of the current method as
+    a Symbol.
+
+* enumerator
+
+  * Enumerator is now a built-in module.  The #next and #rewind
+    methods are implemented using the "generator" library.  Use with
+    care and be aware of the performance loss.
+
+* ipaddr
+
+  * New methods
+    * IPAddr#<=>
+    * IPAddr#succ
+
+      IPAddr objects are now comparable and enumerable having these
+      methods.  This also means that it is possible to have a Range
+      object between two IPAddr objects.
+
+    * IPAddr#to_range
+
+      A new method to create a Range object for the (network) address.
+
+  * Type coercion support
+    * IPAddr#&
+    * IPAddr#|
+    * IPAddr#==
+    * IPAddr#include?
+
+      These methods now accept a string or an integer instead of an
+      IPAddr object as the argument.
+
+* net/smtp
+
+  * Support SSL/TLS.
+
+* openssl
+
+  * New classes
+    * OpenSSL::PKey::EC
+    * OpenSSL::PKey::EC::Group
+    * OpenSSL::PKey::EC::Point
+    * OpenSSL::PKey::PKCS5
+    * OpenSSL::SSL::Session
+
+  * Documentation!
+
+  * Various new methods (see documentation).
+
+  * Remove redundant module namespace in Cipher, Digest, PKCS7, PKCS12.
+    Compatibility classes are provided which will be removed in Ruby 1.9.
+
+* shellwords
+
+  * Add methods for escaping shell-unsafe characters:
+    * Shellwords.join
+    * Shellwords.escape
+    * Array#shelljoin
+    * String#shellescape
+
+  * Add shorthand methods:
+    * Shellwords.split (alias shellwords)
+    * String#shellsplit
+
+* stringio
+
+  * StringIO#getbyte
+  * StringIO#readbyte
+
+    New methods. (aliases for compatibility with 1.9)
+
+  * StringIO#each_char
+  * StringIO#chars
+
+    New methods.
+
+  * StringIO#each
+  * StringIO#each_line
+  * StringIO#each_byte
+
+    Return an enumerator if no block is given.
+
+* tempfile
+
+  * Tempfile.open and Tempfile.new now accept a suffix for the
+    temporary file to be created.  To specify a suffix, pass an array
+    of [basename, suffix] as the first argument.
+
+      Tempfile.open(['image', 'jpg']) { |tempfile| ... }
+
+* tmpdir
+
+  * New method:
+
+    * Dir.mktmpdir
+
+* uri
+
+  * added LDAPS scheme.
+  * Change for RFC3986:
+    * FTP
+      * URI('ftp://example.com/foo').path #=> 'foo'
+      * URI('ftp://example.com/%2Ffoo').path #=> '/foo'
+      * URI::FTP.build([nil, 'example.com', nil, '/foo', 'i']).to_s #=> 'ftp://example.com/%2Ffoo;type=i'
+    * URI merge
+      * URI('http://a/b/c/d;p?q').merge('?y') == URI('http://a/b/c/d;p?y')
+      * URI('http://a/b/c/d;p?q').merge('/./g') == URI('http://a/g')
+      * URI('http://a/b/c/d;p?q').merge('/../g') == URI('http://a/g')
+      * URI('http://a/b/c/d;p?q').merge('../../../g') == URI('http://a/g')
+      * URI('http://a/b/c/d;p?q').merge('../../../../g') == URI('http://a/g')
+
+* rss
+
+  * 0.1.6 -> 0.2.4
+
+  * Fix image module URI
+
+  * Atom support
+
+  * ITunes module support
+
+  * Slash module support
+
+  * content:encoded with RSS 2.0 support
+
+=== Interpreter Implementation
+
+* passing a block to a Proc [experimental]
+
+  This implementation in current shape is known to be buggy/broken,
+  especially with nested block invocation.  Take this as an
+  experimental feature.
+
+* stack trace
+
+  On non-SystemStackError exception, full stack trace is shown.
+
+=== Compatibility issues (excluding feature bug fixes)
+
+* String#slice! had some unintentional bugs and they have been fixed
+  because either they disagreed with documentation or their respective
+  behavior of #slice.  Unfortunately, this causes some
+  incompatibilities in the following (somewhat rare) cases.
+
+  * #slice! no longer expands the array when an out-of-boundary value
+    is given.
+
+      # Ruby 1.8.6
+      a = [1,2]
+      a.slice!(4,0)   #=> nil
+      a               #=> [1,2,nil,nil]
+
+      # Ruby 1.8.7
+      a = [1,2]
+      a.slice!(4,0)   #=> nil
+      a               #=> [1,2]
+
+  * #slice! no longer raises an exception but returns nil when a
+    negative length or out-of-boundary negative position is given.
+
+      # Ruby 1.8.6
+      a = [1,2]
+      a.slice!(1,-1)  #=> (raises IndexError)
+      a.slice!(-5,1)  #=> (raises IndexError)
+
+      # Ruby 1.8.7
+      a = [1,2]
+      a.slice!(1,-1)  #=> nil
+      a.slice!(-5,1)  #=> nil
+
+* String#to_i, String#hex and String#oct no longer accept a sequence
+  of underscores (`__') as part of a number.
+
+    # Ruby 1.8.6
+    '1__0'.to_i     #=> 10
+    '1__0'.to_i(2)  #=> 2  # 0b10
+    '1__0'.oct      #=> 8  # 010
+    '1__0'.hex      #=> 16 # 0x10
+
+    # Ruby 1.8.7
+    '1__0'.to_i     #=> 1
+    '1__0'.to_i(2)  #=> 1
+    '1__0'.oct      #=> 1
+    '1__0'.hex      #=> 1
+
+  The old behavior was inconsistent with Ruby syntax and considered as
+  a bug.
+
+* date
+
+  * Date.parse
+
+    '##.##.##' (where each '#' is a digit) is now taken as 'YY.MM.DD'
+    instead of 'MM.DD.YY'.  While the change may confuse you, you can
+    always use Date.strptime() when you know what you are dealing
+    with.
+
+* REXML
+
+  * REXML::Document.entity_expansion_limit=
+
+    New method to set the entity expansion limit. By default the limit is
+    set to 10000.  See the following URL for details.
+
+    https://www.ruby-lang.org/en/news/2008/08/23/dos-vulnerability-in-rexml/
+
+* stringio
+
+  * StringIO#each_byte
+
+    The return value changed from nil to self.  This is what the
+    document says and the same as each_line() does.
+
+* tempfile
+
+  * The file name format has changed.  No dots are included by default
+    in temporary file names any more.  See above for how to specify a
+    suffix.
+
+* uri
+
+  * See above for details.
+
+== Changes since the 1.8.5 release
+
+=== New platforms/build tools support
+
+* IA64 HP-UX
+
+* Visual C++ 8 SP1
+
+* autoconf 2.6x
+
+=== Global constants
+
+* RUBY_PATCHLEVEL
+
+  New constant since 1.8.5-p1.
+
+=== Library updates (outstanding ones only)
+
+* builtin classes
+
+  * New method: Kernel#instance_variable_defined?
+
+  * New method: Module#class_variable_defined?
+
+  * New feature: Dir::glob() can now take an array of glob patterns.
+
+* date
+
+  * Updated based on date2 4.0.3.
+
+* digest
+
+  * New internal APIs for C and Ruby.
+
+  * Support for autoloading.
+
+      require 'digest'
+
+      # autoloads digest/md5
+      md = Digest::MD5.digest("string")
+
+  * New digest class methods: file
+
+  * New digest instance methods: clone, reset, new, inspect,
+    digest_length (alias size or length), block_length()
+
+  * New library: digest/bubblebabble
+
+  * New function: Digest(name)
+
+* fileutils
+
+  * New option for FileUtils.cp_r(): :remove_destination
+
+* nkf
+
+  * Updated based on nkf as of 2007-01-28.
+
+* thread
+
+  * Replaced with much faster mutex implementation in C.  The former
+    implementation, which is slow but considered to be stable, is
+    available with a configure option `--disable-fastthread'.
+
+* tk
+
+  * Updated Tile extension support based on Tile 0.7.8.
+
+  * Support --without-X11 configure option for non-X11 versions of
+    Tcl/Tk (e.g. Tcl/Tk Aqua).
+
+  * New sample script: irbtkw.rbw -- IRB on Ruby/Tk. It has no trouble
+    about STDIN blocking on Windows.
+
+* webrick
+
+  * New method: WEBrick::Cookie.parse_set_cookies()
+
+=== Compatibility issues (excluding feature bug fixes)
+
+* builtin classes
+
+  * String#intern now raises SecurityError when $SAFE level is greater
+    than zero.
+
+* date
+
+  * Time#to_date and Time#to_datetime are added as private methods.
+    They cause name conflict error in ActiveSupport 1.4.1 and prior,
+    which comes with Rails 1.2.2 and prior.  Updating ActiveSupport
+    and/or Rails to the latest versions fixes the problem.
+
+* digest
+
+  * The constructor does no longer take an initial string to feed.
+    The following examples show how to migrate:
+
+      # Before
+      md = Digest::MD5.new("string")
+      # After (works with any version)
+      md = Digest::MD5.new.update("string")
+
+      # Before
+      hd = Digest::MD5.new("string").hexdigest
+      # After (works with any version)
+      hd = Digest::MD5.hexdigest("string")
+
+  * Digest::Base#==
+
+    * self == string
+
+      Automatic detection between binary digest values and hexadecimal
+      digest values has been dropped.  It is always assumed that a
+      hexadecimal digest value is given for comparison.
+
+    * self == md
+
+      Digest objects are compared by the resulting digest values, not
+      by the exact vector states.
+
+* fileutils
+
+  * A minor implementation change breaks Rake <=0.7.1.
+    Updating Rake to 0.7.2 or higher fixes the problem.
+
+* tk
+
+  * Tk::X_Scrollable (Y_Scrollable) is renamed to Tk::XScrollable
+    (YScrollable). Tk::X_Scrollable (Y_Scrollable) is still available,
+    but it is an alias name.
diff --git a/doc/NEWS-1.9.1 b/doc/NEWS-1.9.1
new file mode 100644
index 0000000000..fb11026d60
--- /dev/null
+++ b/doc/NEWS-1.9.1
@@ -0,0 +1,429 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 1.9.1
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file.
+
+== Changes since the 1.8.7 release
+See doc/NEWS-1.8.7 for changes between 1.8.6 and 1.8.7.
+
+=== Compatibility issues
+
+* language core
+
+    * New syntax and semantics
+          o Block arguments are always local
+          o New semantics for block arguments
+          o defined? and local variables
+          o Parser expects that your source code has only valid byte
+            sequence in some character encoding.  Use magic comments
+            to tell the parser which encoding you use.
+          o New semantics for constant definition in instance_eval
+            or in module_eval.
+
+    * Deprecated syntax
+          o colon (:) instead of "then" in if/unless or case expression.
+          o retry in a loop or an iterator.
+
+* builtin classes and objects
+
+    * Kernel and Object
+          o Kernel#methods and #singleton_methods used to return an
+            array of strings but now they return an array of symbols.
+    * Class and Module
+          o Module#attr works as Module#attr_reader by default.
+	    Optional boolean argument is obsolete.
+          o Module#instance_methods, #private_instance_methods and
+            #public_instance_methods used to return an array of
+            strings but now they return an array of symbols.
+          o Extra subclassing check when binding UnboundMethods
+
+    * Exceptions
+          o Exceptions are equal to each other if they belong to
+            the same class and have the same message and backtrace.
+          o SystemStackError used to be a subclass of StandardError
+            but not it is a direct subclass of Exception.
+          o SecurityError: ditto
+          o Removed Exception#to_str [Ruby2]
+
+    * Enumerable and Enumerator
+          o Enumerable::Enumerator, compatibility alias of Enumerator,
+            is removed.
+          o Enumerable#{map,collect} called without a block returns
+            an enumerator.
+          o Even more builtin and bundled libraries have been made to
+            return an enumerator when called without a block.
+    * Array
+          o Array#nitems was removed (use count {|i| !i.nil?})
+          o Array#choice was removed (use sample)
+          o Array#[m,n] = nil places nil in the array.
+    * Hash
+          o Hash#to_s is equivalent to Hash#inspect
+          o Semantics for Hash#each and Hash#each_pair
+          o Hash#select returns a hash
+          o Hash#key is the new name for #index which has been
+            deprecated.
+          o Hash preserves order.  It enumerates its elements in the
+            order in which the keys are inserted.
+          o Most of the changes in Hash apply to hash like interfaces
+            such as ENV and *DBM.
+    * IO operations
+          o Many methods used to act byte-wise but now some of those act
+            character-wise. You can use alternate byte-wise methods.
+          o IO#getc
+          o Non-blocking IO
+          o Kernel#open takes "t" for newline conversion
+          o Kernel#open takes encoding specified
+          o IO automatically converts byte sequence from a character
+            encodings into another if specified.
+          o StringIO#readpartial
+          o IO.try_convert
+          o IO.binread
+          o IO.copy_stream
+          o IO#binmode?
+          o IO#close_on_exec= and IO#close_on_exec?
+          o Limit input in IO#gets, IO#readline, IO#readlines,
+            IO#each_line, IO#lines, IO.foreach, IO.readlines,
+            StringIO#gets, StringIO#readline, StringIO#each,
+            StringIO#readlines
+          o IO#ungetc, StringIO#ungetc
+          o IO#ungetbyte, StringIO#ungetbyte
+          o IO#internal_encoding, IO#external_encoding,
+            IO#set_encoding
+          o IO.pipe takes encoding option
+          o Directive %u behaves like %d for negative values in
+            printf-style formatting.
+    * File and Dir operations
+          o #to_path is called as necessary in File.path, File.chmod,
+            File.lchmod, File.chown, File.lchown, File.utime,
+            File.unlink, etc..
+          o File.world_readable?
+          o File.world_writable?
+          o Dir.[], Dir.glob
+          o Dir.exist?
+          o Dir.exists?
+    * File::Stat
+          o File::Stat#world_readable?
+          o File::Stat#world_writable?
+    * String
+          o No longer an Enumerable: use each_line/lines for line
+            oriented operation
+          o Encoding-awareness
+              o Character-wise semantics in many methods instead of
+                byte-wise.
+                  o String#[]: Indexing a String with an integer returns a
+                    single character String instead of an integer.
+                  o String#[]=: No longer takes an integer as right
+                    side value.  Note that "str[i] = ?c" because of
+                    the following change.
+                  o ?c is evaluated to a single character string
+                    instead of an integer.
+    * Regexp
+          o Encoding-awareness
+          o Regexp matches only with strings which is encoded in a
+            compatible character encoding to the regexp's.
+          o Regexp#kcode is removed. use Regexp#encoding.
+    * Symbols: restriction on literal symbols
+    * Numeric
+          o Numeric#div always rounds as Integer#div has done.
+          o Numeric#fdiv: ditto.
+    * Integer
+          o Integer(nil) raises TypeError
+    * Fixnum
+          o Fixnum#id2name removed
+          o Fixnum#to_sym removed
+    * Struct
+          o Struct#inspect
+    * Time
+          o New format in Time#to_s
+          o Timezone information preserved on Marshal.dump/load
+    * $SAFE and bound methods
+          o New trusted/untrusted model in addition to
+            tainted/untainted model.
+
+    * Deprecation
+          o $= (global flag for case-sensitiveness on string matching)
+          o Kernel#to_a
+          o Kernel#getc, #gsub, #sub
+          o Kernel#callcc and Continuation now become 'continuation'
+            bundled library.
+          o Object#type
+          o Removed Array and Hash #indices, #indexes
+          o Hash#index
+          o ENV.index
+          o Process::Status#to_int
+          o Numeric#rdiv
+          o Precision is removed.  Don't cry, it will be redesigned
+            and come back in future version.
+          o Symbol#to_int and Symbol#to_i
+          o $KCODE is no longer effective.  Use Encoding related
+            features of each class.
+          o VERSION and friends
+
+* bundled libraries
+
+    * Pathname
+          o No longer has #to_str nor #=~.
+    * time and date
+          o Time.parse and Date.parse interprets slashed numerical dates
+	    as "dd/mm/yyyy".
+    * Readline
+          o If Readline uses libedit, Readline::HISTORY[0] returns the
+            first of the history.
+    * Continuation
+          o as above
+
+    * Deprecation
+          o Complex#image: use Complex#imag
+          o All SSL-related class methods in Net::SMTP
+          o Prime#cache, Prime#primes, Prime#primes_so_far
+          o mailread library: use tmail gem.
+          o cgi-lib library: use cgi.
+          o date2 library: use date.
+          o eregex library
+          o finalize library: use ObjectSpace.define_finalizer if you
+            really need a finalizer. really?
+          o ftools library: use fileutils.
+          o generator library: use Enumerator.
+          o importenv library and Env library
+          o jcode library: use multilingualization support of String
+          o parsedate library
+          o ping library
+          o readbytes library
+          o getopts library and parsearg library: use optparse or
+            getoptlong.
+          o soap, wsdl and xsd libraries: use soap4r gem.
+          o Win32API library: use dl.
+          o dl library: Reimplemented and API changed. use the new
+	    version of dl or ffi gem.
+          o rubyunit library and runit library: use minitest or
+            test/unit. Or use anything you love through RubyGems.
+          o test/unit is reimplemented on top of minitest.  This is
+            not fully compatible with the original.
+
+=== Language core changes
+
+    * New syntax and semantics
+          o Magic comments to declare in which encoding your source
+            code is written
+          o New literal hash syntax and new syntax for hash style
+	    arguments
+          o New syntax for lambdas
+          o .() and calling Procs without #call/#[]
+          o Block in block arguments
+          o Block local variables
+          o Mandatory arguments after optional arguments allowed
+          o Multiple splats allowed
+          o #[] can take splatted arguments, hash style arguments
+            and a block.
+          o New directives in printf-style formatted strings (%).
+          o Newlines allowed before ternary colon operator (:) and
+            method call dot operator (.)
+          o Negative operators such as !, != and !~ are now
+            overloadable
+          o Encoding.default_external and default_internal
+          o __ENCODING__: New pseudo variable to hold the current
+            script's encoding
+
+=== Library updates
+
+* builtin classes and objects
+    * Kernel and Object
+          o BasicObject
+          o Object#=~ returns nil instead of false by default.
+          o Kernel#define_singleton_method
+	  o Kernel#load can load a library from the highest versions
+	    of gems by default.
+    * Class and Module
+          o Module#const_defined?, #const_get and #method_defined?
+            take an optional parameter.
+          o #class_variable_{set,get} are public.
+          o Class of singleton classes
+
+    * Errno::EXXX
+          o All of those are always defined.  Errno::EXXX will be
+            defined as an alias to Errno::NOERROR if your platform
+            does not have one.
+
+    * Binding#eval
+    * Blocks and Procs
+          o Arity of blocks without arguments
+          o proc is now a synonym of Proc.new
+          o Proc#yield
+          o Passing blocks to #[]
+          o Proc#lambda?
+          o Proc#curry
+    * Fiber: coroutines/micro-threads
+    * Thread
+          o Thread.critical and Thread.critical= removed
+          o Thread#exit!, Thread#kill! and Thread#terminate! removed.
+
+    * Enumerable and Enumerator
+          o Enumerator#enum_cons and Enumerator#enum_slice are
+            removed.  Use #each_cons and #each_slice without a block.
+          o Enumerable#each_with_index can take optional arguments
+            and passes them to #each.
+          o Enumerable#each_with_object
+          o Enumerator#with_object
+          o Enumerator.new { ... }
+    * Array
+          o Array#delete returns a deleted element rather than a given
+            object
+          o Array#to_s is equivalent to Array#inspect
+          o Array.try_convert
+          o Array#pack('m0') complies with RFC 4648.
+    * Hash
+          o preserving item insertion order
+          o Hash#default_proc=
+          o Hash#_compare_by_identity and Hash#compare_by_identity?
+          o Hash.try_convert
+          o Hash#assoc
+          o Hash#rassoc
+          o Hash#flatten
+    * Range
+          o Range#cover?
+          o Range#include? iterates over elements and compares the
+            given value with each element unless the range is numeric.
+            Use Range#cover? for the old behavior, i.e. comparison
+            with boundary values.
+          o Range#min, Range#max
+
+    * File and Dir operations
+          o New methods
+    * Process
+          o Process.spawn
+          o Process.daemon
+    * String
+          o String#clear
+          o String#ord
+          o String#getbyte, String#setbyte
+          o String#chars and String#each_char act as character-wise.
+          o String#codepoints, String#each_codepoint
+          o String#unpack with a block
+          o String#hash
+          o String.try_convert
+          o String#encoding
+	  o String#force_encoding, String#encode and String#encode!
+          o String#ascii_only?
+          o String#valid_encoding?
+	  o String#match
+    * Symbol
+          o Zero-length symbols allowed
+          o Symbol#intern
+          o Symbol#encoding
+          o Symbol methods similar to those in String
+    * Regexp
+          o Regexp#=== matches symbols
+          o Regexp.try_convert
+          o Regexp#match
+          o Regexp#fixed_encoding?
+          o Regexp#encoding
+          o Regexp#named_captures
+          o Regexp#names
+    * MatchData
+          o MatchData#names
+          o MatchData#regexp
+    * Encoding
+    * Encoding::Converter
+          o supports conversion between many encodings
+    * Numeric
+          o Numeric#upto, #downto, #times, #step
+          o Numeric#real?, Complex#real?
+          o Numeric#magnitude
+          o Numeric#round
+    * Float
+          o Float#round
+    * Integer
+          o Integer#round
+    * Rational / Complex
+          o They are in the core library now
+    * Math
+          o Math#log takes an optional argument.
+          o Math#log2
+          o Math#cbrt, Math#lgamma, Math#gamma
+    * Time
+          o Time.times removed.  Use Process.times.
+          o Time#sunday?
+          o Time#monday?
+          o Time#tuesday?
+          o Time#wednesday?
+          o Time#thursday?
+          o Time#friday?
+          o Time#saturday?
+          o Time#tv_nsec and Time#nsec
+    * Misc. new methods
+          o RUBY_ENGINE to distinguish between Ruby processor implementation
+          o public_method
+          o public_send
+          o GC.count
+          o ObjectSpace.count_objects
+          o Method#hash, Proc#hash
+          o Method#source_location, UnboundMethod#source_location and
+	    Proc#source_location
+          o __callee__
+          o Elements in $LOAD_PATH and $LOADED_FEATURES are expanded
+
+* bundled libraries
+    * RubyGems
+          o Package management system for Ruby.
+          o Integrated with Ruby's library loader.
+    * Rake
+          o Ruby make. A simple ruby build program with capabilities
+            similar to make.
+    * minitest
+          o Our new testing library which is faster, cleaner and easier
+	    to read than the old test/unit.
+          o You can introduce the old test/unit as testunit gem through
+            RubyGems if you want.
+    * CMath
+          o Complex number version of Math
+    * Prime
+          o Extracted from Mathn and improved. You can easily enumerate
+            prime numbers.
+          o Prime.new is obsolete. Use its class methods.
+    * ripper
+          o Ruby script parser
+    * Readline
+          o Readline.vi_editing_mode?
+          o Readline.emacs_editing_mode?
+          o Readline::HISTORY.clear
+    * Tk
+          o TkXXX widget classes are removed and redefined as aliases of
+            Tk::XXX classes.
+    * RDoc
+          o Updated to version 2.2.2.  See:
+            http://rubyforge.org/frs/shownotes.php?group_id=627&release_id=26434
+    * json
+          o JSON (JavaScript Object Notation) encoder/decoder
+
+* commandline options
+    * -E, --encoding
+    * -U
+    * --enable-gems, --disable-gems
+    * --enable-rubyopt, --disable-rubyopt
+    * long options are allowed in RUBYOPT environment variable.
+
+=== Implementation changes
+
+* Memory Diet
+    * Object Compaction - Object, Array, String, Hash, Struct, Class,
+      Module
+    * st_table compaction (inlining small tables)
+* YARV
+    * Ruby codes are compiled into opcodes before executed.
+    * Native thread
+
+* Platform supports
+    * Support levels
+          (0) Supported
+          (1) Best effort
+          (2) Perhaps
+          (3) Not supported
+    * Dropped
+          o No longer supports djgpp, bcc32, human68k, MacOS 9 or earlier,
+            VMS nor Windows CE.
diff --git a/doc/NEWS-1.9.2 b/doc/NEWS-1.9.2
new file mode 100644
index 0000000000..fedb1f6633
--- /dev/null
+++ b/doc/NEWS-1.9.2
@@ -0,0 +1,509 @@
+# -*- rdoc -*-
+= NEWS for Ruby 1.9.2
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file.
+
+== Changes since the 1.9.1 release
+=== Library updates (outstanding ones only)
+
+* builtin classes
+
+  * Array
+    * new methods:
+      * Array#keep_if
+      * Array#repeated_combination
+      * Array#repeated_permutation
+      * Array#rotate
+      * Array#rotate!
+      * Array#select!
+      * Array#sort_by!
+
+    * extended method:
+      * Array#{uniq,uniq!,product} can take a block.
+
+  * Complex
+    * new method:
+      * Complex#rationalize
+
+  * Dir
+    * new method:
+      * Dir.home
+
+  * Encoding
+    * new encodings:
+      * Big5
+      * Big5-UAO
+      * ISO-2022-JP-KDDI
+      * SJIS-DoCoMo
+      * SJIS-KDDI
+      * SJIS-SoftBank
+      * UTF8-DoCoMo
+      * UTF8-KDDI
+      * UTF8-SoftBank
+
+    * new method:
+      * ascii_compatible?
+
+  * Enumerable
+    * new methods:
+      * Enumerable#chunk
+      * Enumerable#collect_concat
+      * Enumerable#each_entry
+      * Enumerable#flat_map
+      * Enumerable#slice_before
+
+  * Enumerator
+    * new methods:
+      * Enumerator#peek
+      * Enumerator#next_values
+      * Enumerator#peek_values
+      * Enumerator#feed
+      * StopIteration#result
+
+    * extended method:
+      * #with_index accepts an optional argument that specifies the
+        index number to start with, defaulted to 0.
+
+    * incompatible changes:
+      * #rewind now calls the "rewind" method of the enclosed object
+        if defined.
+      * #next doesn't clear the position at end.
+
+  * ENV
+    * Uses locale's encoding
+    * ENV.[]= raises Errno::{EINVAL,ENOMEM} etc. on failure.
+    * new methods:
+      * ENV.keep_if
+      * ENV.select!
+
+  * Float
+    * new constants:
+      * Float::INFINITY
+      * Float::NAN
+    * new method:
+      * Float#rationalize
+
+  * File
+    * new methods:
+      * File.realpath
+      * File.realdirpath
+
+  * GC::Profiler
+    * new method:
+      * GC::Profiler.total_time
+
+  * Hash
+    * new methods:
+      * Hash#keep_if
+      * Hash#select!
+
+  * IO
+    * new methods:
+      * IO#autoclose=
+      * IO#autoclose?
+      * IO#fdatasync
+      * IO#codepoints
+      * IO#each_codepoint
+
+    * extended method:
+      * IO.pipe can take a block.
+
+    * new modules:
+      * IO::WaitReadable
+      * IO::WaitWritable
+        They are used to extend non-blocking exceptions.
+
+  * Integer
+    * new method:
+      * Integer#rationalize
+
+  * Kernel
+    * new methods:
+      * Kernel#respond_to_missing?
+      * Kernel#singleton_class
+      * Kernel#require_relative
+
+    * extended method:
+      * Kernel#respond_to? can be used to detect methods not implemented.
+        For example, Process.respond_to?(:fork) returns false on Windows.
+
+    * incompatible changes:
+      * Kernel#instance_eval yields the receiver.
+
+      * Kernel#exec
+      * Kernel#spawn
+      * Kernel#system
+      * Kernel#` (`...`)
+        ..etc.
+
+        On Windows, the current directory is no longer implicitly
+        prepended to the default command search path, just like other
+        platforms.
+
+  * MatchData
+    * new method:
+      * MatchData#==
+
+  * Method
+    * new method:
+      * Method#parameters
+
+  * NilClass
+    * new method:
+      * NilClass#rationalize
+
+  * Object
+    * extended methods:
+      * Float() supports hexadecimal floating point format.
+      * printf() supports %a/%A format.
+
+  * Proc
+    * new method:
+      * Proc#parameters
+    * extended method:
+      * Proc#source_location returns location even if receiver is a method
+        defined by attr_reader / attr_writer / attr_accessor.
+
+  * Process
+    * extended method:
+      * Process.spawn accepts [:child, FD] for a redirect target.
+
+  * Random (new class to generate pseudo-random numbers)
+
+  * Rational
+    * new method:
+      * Rational#rationalize
+
+  * String
+    * extended methods:
+      * string[regexp, name] is supported for named capture.
+
+  * Thread
+    * new methods:
+      * Thread#add_trace_func
+      * Thread#set_trace_func
+
+  * Time
+    * extended features:
+      * time_t restriction is removed to represent before 1901 and after 2038.
+        Proleptic Gregorian calendar is used for old dates.
+      * Time.new have optional arguments to specify date with time offset.
+      * Time#getlocal, Time#localtime have optional time offset argument.
+
+    * new methods:
+      * Time#to_r
+      * Time#subsec
+      * Time#round
+
+    * incompatible change:
+      * The year argument of Time.{utc,gm,local,mktime} is now interpreted as
+        the value itself.  For example, Time.utc(99) means the year 99 AD,
+        not 1999 AD.
+
+  * UnboundMethod
+    * new method:
+      * UnboundMethod#parameters
+
+
+* digest
+  * new methods:
+    * Digest::Class.base64digest
+    * Digest::Instance#base64digest
+    * Digest::Instance#base64digest!
+
+  * Digest::HMAC (digest/hmac) has been marked as deprecated because
+    it was unintentional for the experimental library to be included
+    in the final release of 1.9.1.  Please use OpenSSL::HMAC instead.
+
+* rss
+
+  * 0.2.4 -> 0.2.7.
+
+  * RSS::Maker.make
+    * raise an exception not returns nil for invalid feed making.
+    * requires block.
+
+  * RSS::Maker.[]
+    * new method to return maker class.
+
+  * RSS::Maker.supported?(version)
+    * new method to check whether given version is supported.
+
+  * RSS::Maker: item.guid.permanent_link?
+    * new alias of item.guid.isPermaLink
+  * RSS::Maker: item.guid.permanent_link=
+    * new alias of item.guid.isPermaLink=
+
+* DL
+  * Now uses libffi as a backend if avaiable.
+    It means DL works fine on more platforms.
+
+* Fiddle
+  * A lightweight wrapper for libffi.
+
+* YAML
+  * Now uses libyaml as a backend if available.
+    It means YAML library supports new standards for YAML.
+
+* Psych
+  * A wrapper for libyaml.
+
+* JSON
+  * Update to JSON 1.1.9
+
+* REXML
+
+  * REXML::Document.entity_expansion_limit=
+
+    New method to set the entity expansion limit. By default the limit is
+    set to 10000.  See the following URL for details.
+
+    https://www.ruby-lang.org/en/news/2008/08/23/dos-vulnerability-in-rexml/
+
+* RDoc
+
+  * Updated to RDoc 2.5.8
+
+* RubyGems
+
+  * Updated to RubyGems 1.3.7
+
+* logger
+
+  * imported upstream version (logger/1.2.7)
+    * do not raise an exception even if log writing failed.
+    * do not raise ShiftingError if an aged file already exists.
+      (no ShiftingError will be raised from 1.2.7, just warn() instead)
+
+* matrix
+  * API change to adhere strictly to mathematical definitions:
+    * Matrices must now be rectangular.
+    * trace, regular?, singular? are defined only for square matrices
+    * support for empty matrices
+    * all integer matrices now have the right determinant (also an integer)
+
+  * Matrix and Vector include Enumerable.
+
+  * new methods:
+    * Matrix.build
+    * Matrix.empty
+    * Matrix#conj
+    * Matrix#conjugate
+    * Matrix#each
+    * Matrix#each_with_index
+    * Matrix#empty?
+    * Matrix#imag
+    * Matrix#imaginary
+    * Matrix#real
+    * Matrix#real?
+    * Matrix#rect
+    * Matrix#rectangular
+
+* net/http
+  * merged net/https.
+
+* open3
+  * new methods:
+    * Open3.popen2
+    * Open3.popen2e
+    * Open3.capture3
+    * Open3.capture2
+    * Open3.capture2e
+    * Open3.pipeline_rw
+    * Open3.pipeline_r
+    * Open3.pipeline_w
+    * Open3.pipeline_start
+    * Open3.pipeline
+
+* pty
+  * new methods:
+    * PTY.open
+    * PTY.check
+  * deprecated methods:
+    * protect_signal
+    * reset_signal
+
+* openssl
+  * new methods:
+    * OpenSSL::Buffering#read_nonblock
+    * OpenSSL::Buffering#write_nonblock
+    * OpenSSL::SSL::SSLSocket#connect_nonblock
+    * OpenSSL::SSL::SSLSocket#accept_nonblock
+
+* scanf
+  * support %a/%A format.
+
+
+* socket
+
+  * incompatible changes:
+    * Socket#{recvfrom,recvfrom_nonblock,accept,accept_nonblock,sysaccept}
+      returns a sender address as Addrinfo object instead of a binary sockaddr string.
+      Addrinfo#to_s returns the old binary sockaddr string.
+    * BasicSocket#getsockopt returns Socket::Option object instead of a binary string.
+      Socket::Option#to_s returns the old binary string.
+    * Socket.do_not_reverse_lookup is turned on by default now.
+
+  * new class:
+    * Addrinfo
+    * Socket::Option
+    * Socket::AncillaryData
+
+  * new methods:
+    * Socket.ip_address_list
+    * Socket.tcp
+    * Socket.tcp_server_loop
+    * Socket.tcp_server_sockets
+    * Socket.udp_server_sockets
+    * Socket.udp_server_loop_on
+    * Socket.udp_server_loop
+    * Socket.unix
+    * Socket.unix_server_loop
+    * Socket.unix_server_socket
+    * Socket.accept_loop
+    * Socket#ipv6only!
+    * BasicSocket#local_address
+    * BasicSocket#remote_address
+    * BasicSocket#connect_address
+    * BasicSocket#sendmsg
+    * BasicSocket#sendmsg_nonblock
+    * BasicSocket#recvmsg
+    * BasicSocket#recvmsg_nonblock
+    * BasicSocket#getpeereid
+
+  * extended methods:
+    * Socket.new's 3rd argument is optional now.
+    * Socket.pair's 3rd argument is optional now.
+    * Socket.pair and UNIXSocket.pair can take a block.
+    * BasicSocket#send, UDPSocket#send, Socket.getnameinfo, Socket#bind, and
+      Socket#{connect,connect_nonblock} accepts an Addrinfo object as sockaddr.
+    * BasicSocket#getsockopt accepts a Socket::Option object.
+    * Socket.getaddrinfo and IPSocket#{addr,peeraddr} accept an optional
+      argument to turn reverse lookup on/off.
+
+    * constant names can be accepted as well as constant values.
+      i.e. Socket.new(:PF_INET, :SOCK_STREAM, 0)
+      The constant names can be specified without the prefix.
+      i.e. Socket.new(:INET, :STREAM, 0)
+      * protocol/address family
+      * socket type
+      * socket option protocol level
+      * socket option name
+      * shutdown's argument
+
+* stringio
+  * new methods:
+    * StringIO#read_nonblock
+    * StringIO#write_nonblock
+* pathname
+  * new methods:
+    * Pathname#binread
+    * Pathname#realdirpath
+    * Pathname#each_child
+
+  * extended methods:
+    * Pathname#realpath and Pathname#realdirpath takes optional basedir
+      argument.
+
+* Readline
+  * new methods:
+    * Readline.set_screen_size
+    * Readline.get_screen_size
+
+  * extended methods:
+    * Readline.completion_proc= accepts nil.
+      nil means to use default completion proc.
+
+* set
+  * new methods:
+    * Set#keep_if
+    * Set#select!
+
+* time
+  * incompatible changes:
+    * Time.parse raises ArgumentError when no date information.
+
+* thread
+  * extended method:
+    * ConditionVariable#wait takes timeout argument.
+
+* securerandom
+  * new methods:
+    * SecureRandom.urlsafe_base64
+
+* URI
+  * new methods:
+    * URI.encode_www_form
+    * URI.decode_www_form
+    * URI.encode_www_form_component
+    * URI.decode_www_form_component
+  * Obsoleted methods:
+    * URI.decode
+    * URI.encode
+    * URI.escape
+    * URI.unescape
+
+* etc
+  * new methods:
+    * Etc::Passwd.each
+    * Etc::Group.each
+
+* zlib
+  * new methods:
+    * Zlib::GzipFile#path
+    * Zlib.#adler32_combine
+    * Zlib.#crc32_combine
+
+* rbconfig
+  * new methods:
+    * RbConfig.ruby
+
+=== Language changes
+
+* Regexp properties (\p{}) names now ignore underscores, spaces, and case, so
+  \p{ol chiki} is the same as \p{Ol_Chiki}
+* Regexps now support Unicode 5.2 (new characters and scripts)
+* \d, \s, and \w are now ASCII only; use POSIX bracket classes and \p{} for
+  Unicode semantics
+* $: no longer includes the current directory, use require_relative
+* Symbol with an invalid encoding is forbidden to exist.
+
+=== Compilation options
+
+* --program-prefix and --program-suffix no longer act on the shared object
+  names nor paths to libraries.
+
+  use --with-rubylibprefix='${libruby}/${RUBY_INSTALL_NAME}' and
+  --with-soname='${RUBY_INSTALL_NAME}' for the same result as Ruby 1.9.1.
+
+* --with-arch is added for universal binary, instead of
+  --enable-fat-binary option.
+
+=== Compatibility issues (excluding feature bug fixes)
+
+  * Enumerator#rewind
+  * Socket#recvfrom
+  * Socket#recvfrom_nonblock
+  * Socket#accept
+  * Socket#accept_nonblock
+  * Socket#sysaccept
+  * BasicSocket#getsockopt
+  * Time.utc
+  * Time.gm
+  * Time.local
+  * Time.mktime
+  * Time.parse
+  * --program-prefix and --program-suffix
+  * --enable-fat-binary
+  * $:
+
+    See above.
+
+  * Digest::HMAC
+
+    Deprecated.  See above.
diff --git a/doc/NEWS-1.9.3 b/doc/NEWS-1.9.3
new file mode 100644
index 0000000000..484660f420
--- /dev/null
+++ b/doc/NEWS-1.9.3
@@ -0,0 +1,341 @@
+# -*- rdoc -*-
+= NEWS for Ruby 1.9.3
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file.
+
+== Changes since the 1.9.2 release
+=== License
+
+* Ruby's License is changed from a dual license with GPLv2
+  to a dual license with 2-clause BSDL.
+
+=== Known platform dependent issues
+==== OS X Lion
+
+* You have to configure ruby with '--with-gcc=gcc-4.2' if you're using
+  Xcode 4.1, or, if you're using Xcode 4.2, you have to configure ruby
+  with '--with-gcc=clang'.
+
+=== C API updates
+
+* rb_scan_args() is enhanced with support for option hash argument
+  extraction.
+
+* ruby_vm_at_exit() added.  This enables extension libs to hook a VM
+  termination.
+
+* rb_reserved_fd_p() added.  If you want to close all file descriptors,
+  check using this API.  [ruby-core:37759]
+
+=== Library updates (outstanding ones only)
+
+* builtin classes
+
+  * ARGF
+    * new methods:
+      * ARGF.print
+      * ARGF.printf
+      * ARGF.putc
+      * ARGF.puts
+      * ARGF.read_nonblock
+      * ARGF.to_write_io
+      * ARGF.write
+
+  * Array
+    * extended method:
+      * Array#pack supports endian modifiers
+
+  * Bignum
+    * Multiplication algorithm for Bignums with a large number of digits over
+      150 BDIGITs is changed in order to reduce its calculation time.
+      Now such large Bignums are multiplied by using Toom-3 algorithm.
+
+  * Encoding
+    * new encodings:
+      * CP950
+      * CP951
+      * UTF-16
+      * UTF-32
+    * change alias:
+      * SJIS is Windows-31J
+
+  * File
+    * new constant:
+      * File::NULL
+        name of NULL device.
+      * File::DIRECT
+        name of O_DIRECT.
+
+  * IO
+    * extended method:
+      * IO#putc supports multibyte characters
+    * new methods:
+      * IO#advise
+      * IO.write(name, string, [offset] )
+        Write `string` to file `name`.
+        Opposite with File.read.
+      * IO.binwrite(name, string, [offset] )
+        binary version of IO.write.
+
+  * Kernel
+    * move #__id__ to BasicObject.
+    * extended method:
+      * Kernel#rand supports range argument
+
+  * Module
+    * new methods:
+      * Module#private_constant
+      * Module#public_constant
+
+  * Random
+    * extended method:
+      * Random.rand supports range argument
+
+  * String
+    * extended method:
+      * String#unpack supports endian modifiers
+    * new method:
+      * String#prepend
+      * String#byteslice
+
+  * Time
+    * extended method:
+      * Time#strftime supports %:z and %::z.
+
+  * Process
+    * Process#maxgroups and Process#maxgroups= now raise NotImplementedError if
+      the platform don't support supplementary groups concept.
+
+* bigdecimal
+
+  * BigDecimal#power and BigDecimal#** support non-integral exponent.
+
+  * Kernel.BigDecimal and BigDecimal.new now accept instances of Integer,
+    Rational, Float, and BigDecimal.  If you pass a Rational or a Float to
+    them, you must specify the precision to produce the digits of a BigDecimal.
+
+  * The behavior of BigDecimal#coerce with a Rational is changed.  It uses
+    the precision of the receiver BigDecimal to produce the digits of a
+    BigDecimal from the given Rational.
+
+* bigdecimal/util
+
+  * BigDecimal#to_d and Integer#to_d are added.
+
+  * Float#to_d accepts a precision.
+
+  * Rational#to_d raises ArgumentError when passing zero or negative
+    precision.
+
+  * Rational#to_d
+
+    * Zero and an implicit precision is deprecated.
+      This feature is removed at the next release of bigdecimal.
+
+    * A negative precision isn't supported.
+      Be careful it is an incompatible change.
+
+* date
+
+  * Accepts flonum explicitly with limitations.
+    * If the given offset is flonum, DateTime assumes its precision is
+      at most second.
+
+      DateTime.new(2001,2,3,0,0,0,3.0/24) ==
+      DateTime.new(2001,2,3,0,0,0,'+03:00')
+        #=> true
+
+    * If the given operand for -/+ is flonum, DateTime assumes its
+      precision is at most nanosecond.
+
+      DateTime.new(2001,2,3) + 0.5 == DateTime.new(2001,2,3,12)
+        #=> true
+
+    * Precision of offset is always at most second.
+
+      Rational('0.5') == Rational('0.500001')  #=> false
+      DateTime.new(2001,2,3,0,0,0,Rational('0.5')) ==
+      DateTime.new(2001,2,3,0,0,0,Rational('0.500001'))
+        #=> true
+
+  * Ignores long offset and far reform day (with warning).
+
+    * Now accepts only:
+
+      -1<=offset<=1 (-24:00..+24:00)
+      2298874<=start<=2426355 or -/+oo
+        (proleptic Gregorian/Julian mean -/+oo)
+
+  * A method strftime cannot produce huge output (same as Time's one).
+
+    * Even though Date/DateTime can handle far dates, the following causes
+      an exception.
+
+      DateTime.new(1<<10000).strftime('%Y')  # Errno::ERANGE
+
+  * Changed the format of inspect.
+  * Changed the format of marshal (but, can load old dumps).
+
+* io/console
+  * new methods:
+    * IO#noecho {|io| }
+    * IO#echo=
+    * IO#echo?
+    * IO#raw {|io| }
+    * IO#raw!
+    * IO#getch
+    * IO#winsize
+    * IO.console
+
+* json
+  * updated to v1.5.4.
+
+* matrix
+  * new classes:
+    * Matrix::EigenvalueDecomposition
+    * Matrix::LUPDecomposition
+  * new methods:
+    * Matrix#diagonal?
+    * Matrix#eigen
+    * Matrix#eigensystem
+    * Matrix#hermitian?
+    * Matrix#lower_triangular?
+    * Matrix#lup
+    * Matrix#lup_decomposition
+    * Matrix#normal?
+    * Matrix#orthogonal?
+    * Matrix#permutation?
+    * Matrix#round
+    * Matrix#symmetric?
+    * Matrix#unitary?
+    * Matrix#upper_triangular?
+    * Matrix#zero?
+    * Vector#magnitude, #norm
+    * Vector#normalize
+  * extended methods:
+    * Matrix#each and #each_with_index can iterate on a subset of the elements
+    * Matrix#find_index returns [row, column] and can iterate on a subset
+      of the elements
+    * Matrix#** implements Numeric exponents (using the eigensystem)
+    * Matrix.zero can build rectangular matrices
+
+* minitest
+  * Minitest has been updated to version 2.2.2.
+  * For full details, see https://github.com/seattlerb/minitest/blob/master/History.txt
+
+* net/http
+  * SNI (Server Name Indication) supported for HTTPS.
+
+  * Allow to configure to wait server returning '100 continue' response
+    before sending HTTP request body. Set Net::HTTP#continue_timeout AND pass
+    'expect' => '100-continue' to a extra HTTP header.
+
+    For example, the following code sends HTTP header and waits for getting
+    '100 continue' response before sending HTTP request body. When 0.5 [sec]
+    timeout occurs or the server send '100 continue', the client sends HTTP
+    request body.
+      http.continue_timeout = 0.5
+      http.request_post('/continue', 'body=BODY', 'expect' => '100-continue')
+
+  * new method:
+    * Net::HTTPRequest#set_form): Added to support
+      both application/x-www-form-urlencoded and multipart/form-data.
+
+* objspace
+  * new method:
+    * ObjectSpace::memsize_of_all
+
+* openssl
+  * PKey::RSA and PKey::DSA now use the generic X.509 encoding scheme
+    (e.g. used in a X.509 certificate's Subject Public Key Info) when
+    exporting public keys to DER or PEM. Backward compatibility is
+    ensured by (already existing) fallbacks during creation.
+  * OpenSSL::ASN1::Constructive#new and OpenSSL::ASN1::Primitive#new
+    (and the constructors of their sub-classes) will no longer force
+    tagging to be set to :EXPLICIT when tag and/or tag_class are passed
+    as parameters. tagging must be set explicitly.
+  * Support for infinite length encodings via infinite_length attribute.
+  * OpenSSL::PKey.read( file | string [, pwd] ) allows to read arbitrary
+    public/private keys in DER-/PEM-encoded form with an optional password
+    for encrypted PEM encodings.
+  * Add new method OpenSSL::X509::Name#hash_old as a wrapper of
+    X509_NAME_hash_old() defined from OpenSSL 1.0.0. It returns OpenSSL 0.9.8
+    compatible hash value.
+
+* optparse
+  * support for bash/zsh completion.
+
+* Rake
+  * Rake has been upgraded from 0.8.7 to 0.9.2.2.  For full release notes see
+    https://github.com/jimweirich/rake/blob/master/CHANGES
+
+* RDoc
+  * RDoc has been upgraded to version 3.9.4.  For full release notes see
+    http://docs.seattlerb.org/rdoc/History_txt.html
+
+* rexml
+  * Support Ruby native encoding mechanism and iconv dependency is dropped.
+
+* RubyGems
+  * RubyGems has been upgraded to version 1.8.10. For full release notes see
+    http://rubygems.rubyforge.org/rubygems-update/History_txt.html
+
+* stringio
+  * extended method:
+    * StringIO#set_encoding can get 2nd argument and optional hash.
+
+* test/unit
+  * New arguments:
+    * -j N, --jobs=N: Allow run N testcases at once.
+    * --jobs-status: Show status of jobs when parallel running.
+    * --no-retry: Don't retry testcases which failed when parallel running.
+    * --ruby=RUBY: path to ruby for job(worker) process. optional.
+    * --hide-skip: Hide skip messages. You'll see the number of skips at end of
+      test result.
+
+* uri
+  * new methods:
+    * URI::Generic#hostname
+    * URI::Generic#hostname=
+
+* webrick
+  * new method:
+    * WEBrick::HTTPRequest#continue for generating '100 continue' response.
+  * new logging directive:
+    * %{remote}p for remote (client) port number.
+
+* yaml
+  * The default YAML engine is now Psych. You may downgrade to syck by setting
+    YAML::ENGINE.yamler = 'syck'.
+
+* zlib
+  * new methods:
+    * Zlib.deflate
+    * Zlib.inflate
+
+* FileUtils
+  * extended method:
+    * FileUtils#chmod supports symbolic mode argument.
+
+=== Language changes
+
+* Regexps now support Unicode 6.0. (new characters and scripts)
+
+* [experimental] Regexps now support Age property.
+  Unlike Perl, current implementation takes interpretation of the
+  interpretation of UTS #18.
+  http://www.unicode.org/reports/tr18/
+
+* Turning on/off indentation warnings with directives.
+  ("# -*- warn-indent: true -*-" / "# -*- warn-indent: false -*-")
+
+=== Compatibility issues (excluding feature bug fixes)
+
+  * Rational#to_d
+
+    See above.
diff --git a/doc/NEWS-2.0.0 b/doc/NEWS-2.0.0
new file mode 100644
index 0000000000..414789dcd1
--- /dev/null
+++ b/doc/NEWS-2.0.0
@@ -0,0 +1,530 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 2.0.0
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file.
+
+== Changes since the 1.9.3 release
+
+=== Language changes
+
+* Added keyword arguments.
+
+* Added %i and %I for symbol list creation (similar to %w and %W).
+
+* Default source encoding is changed to UTF-8. (was US-ASCII)
+
+* No warning for unused variables starting with '_'
+
+=== Core classes updates (outstanding ones only)
+
+* ARGF
+  * added method:
+    * added ARGF#codepoints and ARGF#each_codepoint, like the corresponding
+      methods for IO.
+
+* Array
+  * added method:
+    * added Array#bsearch for binary search.
+  * incompatible changes:
+    * random parameter of Array#shuffle! and Array#sample now
+      will be called with one argument, maximum value.
+    * when given Range arguments, Array#values_at now returns nil for each
+      value that is out-of-range.
+
+* Enumerable
+  * added method:
+    * added Enumerable#lazy method for lazy enumeration.
+
+* Enumerator
+  * added method:
+    * added Enumerator#size for lazy size evaluation.
+  * extended method:
+    * Enumerator.new accept an argument for lazy size evaluation.
+  * new class Enumerator::Lazy for lazy enumeration
+
+* ENV
+  * aliased method:
+    * ENV.to_h is a new alias for ENV.to_hash
+
+* Fiber
+  * incompatible changes:
+    * Fiber#resume cannot resume a fiber which invokes "Fiber#transfer".
+
+* File
+  * extended method:
+    * File.fnmatch? now expands braces in the pattern if
+      File::FNM_EXTGLOB option is given.
+
+* GC
+  * improvements:
+    * introduced the bitmap marking which suppresses to copy a memory page
+      with Copy-on-Write.
+    * introduced the non-recursive marking which avoids unexpected stack overflow.
+
+* GC::Profiler
+  * added method:
+    * added GC::Profiler.raw_data which returns raw profile data for GC.
+
+* Hash
+  * added method:
+    * added Hash#to_h as explicit conversion method, like Array#to_a.
+  * extended method:
+    * Hash#default_proc= can be passed nil to clear the default proc.
+
+* IO
+  * deprecated methods:
+    * IO#lines, #bytes, #chars and #codepoints are deprecated.
+
+* Kernel
+  * added method:
+    * added Kernel#Hash conversion method like Array() or Float().
+    * added Kernel#__dir__ which returns the absolute path of the
+      directory of the file from which this method is called.
+    * added Kernel#caller_locations which returns an array of
+      frame information objects.
+  * extended method:
+    * Kernel#warn accepts multiple args in like puts.
+    * Kernel#caller accepts second optional argument `n' which specify
+      required caller size.
+    * Kernel#to_enum and enum_for accept a block for lazy size evaluation.
+  * incompatible changes:
+    * system() and exec() close non-standard file descriptors
+      (The default of :close_others option is changed to true by default.)
+    * respond_to? against a protected method now returns false unless
+      the second argument is true.
+    * __callee__ has returned to the original behavior, and now
+      returns the called name but not the original name in an
+      aliased method.
+    * Kernel#inspect does not call #to_s anymore
+      (it used to call redefined #to_s).
+
+* LoadError
+  * added method:
+    * added LoadError#path method to return the file name that could not be
+      loaded.
+
+* Module
+  * added method:
+    * added Module#prepend which is similar to Module#include,
+      however a method in the prepended module overrides the
+      corresponding method in the prepending module.
+    * added Module.prepended and Module.prepend_features, similar
+      to included and append_features.
+    * added Module#refine, which extends a class or module locally.
+      [experimental]
+  * extended method:
+    * Module#define_method accepts a UnboundMethod from a Module.
+    * Module#const_get accepts a qualified constant string, e.g.
+      Object.const_get("Foo::Bar::Baz")
+
+* Mutex
+  * added method:
+    * added Mutex#owned? which returns the mutex is held by current
+      thread or not. [experimental]
+  * incompatible changes:
+    * Mutex#lock, Mutex#unlock, Mutex#try_lock, Mutex#synchronize
+      and Mutex#sleep are no longer allowed to be used from trap handler
+      and raise a ThreadError in such case.
+    * Mutex#sleep may spurious wakeup. Check after wakeup.
+
+* NilClass
+  * added method:
+    * added nil.to_h which returns {}
+
+* ObjectSpace::WeakMap
+  * new low level class to hold weak references to objects.
+
+* Proc
+  * incompatible change:
+    * removed Proc#== and #eql? so two procs are == only when they are
+      the same object.
+
+* Process
+  * added method:
+    * added getsid for getting session id (unix only).
+
+* Range
+  * added method:
+    * added Range#size for lazy size evaluation.
+    * added Range#bsearch for binary search.
+
+* RubyVM (MRI specific)
+  * added RubyVM::InstructionSequence.of to get the instruction sequence
+    from a method or a block.
+  * added RubyVM::InstructionSequence#path, #absolute_path, #label,
+    #base_label and #first_lineno to retrieve information from where
+    the instruction sequence was defined.
+  * added Environment variables to specify stack usage:
+    * RUBY_THREAD_VM_STACK_SIZE: vm stack size used at thread creation.
+      default: 128KB (32bit CPU) or 256KB (64bit CPU).
+    * RUBY_THREAD_MACHINE_STACK_SIZE: machine stack size used at thread
+      creation. default: 512KB or 1024KB.
+    * RUBY_FIBER_VM_STACK_SIZE: vm stack size used at fiber creation.
+      default: 64KB or 128KB.
+    * RUBY_FIBER_MACHINE_STACK_SIZE: machine stack size used at fiber
+      creation. default: 256KB or 512KB.
+    These variables are checked only at launched time.
+  * added constant DEFAULT_PARAMS to get above default parameters.
+
+* Signal
+  * added method:
+    * added Signal.signame which returns signal name
+
+  * incompatible changes:
+    * Signal.trap raises ArgumentError when :SEGV, :BUS, :ILL, :FPE, :VTALRM
+      are specified.
+
+* String
+  * added method:
+    * added String#b returning a copied string whose encoding is ASCII-8BIT.
+  * change return value:
+    * String#lines now returns an array instead of an enumerator.
+    * String#chars now returns an array instead of an enumerator.
+    * String#codepoints now returns an array instead of an enumerator.
+    * String#bytes now returns an array instead of an enumerator.
+
+* Struct
+  * added method:
+    * added Struct#to_h returning values with keys corresponding to the
+      instance variable names.
+
+* Thread
+  * added method:
+    * added Thread#thread_variable_get for getting thread local variables
+      (these are different than Fiber local variables).
+    * added Thread#thread_variable_set for setting thread local variables.
+    * added Thread#thread_variables for getting a list of the thread local
+      variable keys.
+    * added Thread#thread_variable? for testing to see if a particular thread
+      variable has been set.
+    * added Thread.handle_interrupt as well as instance and singleton methods
+      pending_interrupt? for asynchronous handling of exceptions
+    * added Thread#backtrace_locations which returns similar information of
+      Kernel#caller_locations.
+  * new class Thread::Backtrace::Location to hold backtrace location
+    information. These are returned by Thread#backtrace_locations and
+    Kernel#caller_locations.
+  * incompatible changes:
+    * Thread#join and Thread#value now raises a ThreadError if target thread
+      is the current or main thread.
+
+* Time
+  * change return value:
+    * Time#to_s now returns US-ASCII encoding instead of BINARY.
+
+* TracePoint
+  * new class. This class is replacement of set_trace_func.
+    Easy to use and efficient implementation.
+
+* toplevel
+  * added method:
+    * added main.define_method which defines a global function.
+    * added main.using, which imports refinements into the current file or
+      eval string. [experimental]
+
+=== Core classes compatibility issues (excluding feature bug fixes)
+
+* Array#values_at
+
+  See above.
+
+* String#lines
+* String#chars
+* String#codepoints
+* String#bytes
+
+  These methods no longer return an Enumerator, although passing a
+  block is still supported for backwards compatibility.
+
+  Code like str.lines.with_index(1) { |line, lineno| ... } no longer
+  works because str.lines returns an array.  Replace lines with
+  each_line in such cases.
+
+* IO#lines
+* IO#chars
+* IO#codepoints
+* IO#bytes
+* ARGF#lines
+* ARGF#chars
+* ARGF#bytes
+* StringIO#lines
+* StringIO#chars
+* StringIO#codepoints
+* StringIO#bytes
+* Zlib::GzipReader#lines
+* Zlib::GzipReader#bytes
+
+  These methods are deprecated in favor of each_line, each_byte,
+  each_char and each_codepoint.
+
+* Proc#==
+* Proc#eql?
+
+  These methods were removed. Two procs are == only when they are
+  the same object.
+
+* Fixnum
+* Bignum
+* Float
+
+  Fixnums, Bignums and Floats are frozen.
+
+* Signal.trap
+
+  See above.
+
+* Merge Onigmo.
+  https://github.com/k-takata/Onigmo
+
+* The :close_others option is true by default for system() and exec().
+  Also, the close-on-exec flag is set by default for all new file descriptors.
+  This means file descriptors doesn't inherit to spawned process unless
+  explicitly requested such as system(..., fd=>fd).
+
+* Kernel#respond_to? against a protected method now returns false
+  unless the second argument is true.
+
+* Kernel#respond_to_missing?
+* Kernel#initialize_clone
+* Kernel#initialize_dup
+
+  These methods are now private.
+
+* Thread#join, Thread#value
+
+  See above.
+
+* Mutex#lock, Mutex#unlock, Mutex#try_lock, Mutex#synchronize and Mutex#sleep
+
+  See above.
+
+=== Stdlib updates (outstanding ones only)
+
+* cgi
+  * Add HTML5 tag maker.
+  * CGI#header has been renamed to CGI#http_header and
+    aliased to CGI#header.
+  * When HTML5 tagmaker called, overwrite CGI#header,
+    CGI#header function is to create a <header> element.
+
+* CSV
+  * Removed CSV::dump and CSV::load to protect users from dangerous
+    serialization vulnerability
+
+* iconv
+  * Iconv has been removed. Use String#encode instead.
+
+* io/console
+  * new methods:
+    * added IO#cooked which sets the terminal to cooked mode within the given block.
+    * added IO#cooked! which sets the terminal to cooked.
+  * extended method:
+    * IO#raw, IO#raw!, and IO#getch accept keyword arguments, :min and :time.
+
+* io/wait
+  * new features:
+    * added IO#wait_writable method.
+    * added IO#wait_readable method as alias of IO#wait.
+
+* json
+  * updated to 1.7.7.
+
+* net/http
+  * new features:
+    * Proxies are now automatically detected from the http_proxy environment
+      variable.  See Net::HTTP::new for details.
+    * gzip and deflate compression are now requested for all requests by
+      default.  See Net::HTTP for details.
+    * SSL sessions are now reused across connections for a single instance.
+      This speeds up connection by using a previously negotiated session.
+    * Requests may be created from a URI which sets the request_uri and host
+      header of the request (but does not change the host connected to).
+    * Responses contain the URI requested which allows easier implementation of
+      redirect following.
+  * new methods:
+    * Net::HTTP#local_host
+    * Net::HTTP#local_host=
+    * Net::HTTP#local_port
+    * Net::HTTP#local_port=
+  * extended method:
+    * Net::HTTP#connect uses local_host and local_port if specified.
+
+* net/imap
+  * new methods:
+    * Net::IMAP.default_port
+    * Net::IMAP.default_imap_port
+    * Net::IMAP.default_tls_port
+    * Net::IMAP.default_ssl_port
+    * Net::IMAP.default_imaps_port
+
+* objspace
+  * new method:
+    * ObjectSpace.reachable_objects_from(obj)
+
+* openssl
+  * Consistently raise an error when trying to encode nil values. All instances
+    of OpenSSL::ASN1::Primitive now raise TypeError when calling to_der on an
+    instance whose value is nil. All instances of OpenSSL::ASN1::Constructive
+    raise NoMethodError in the same case. Constructing such values is still
+    permitted.
+  * TLS 1.1 & 1.2 support by setting OpenSSL::SSL::SSLContext#ssl_version to
+    :TLSv1_2, :TLSv1_2_server, :TLSv1_2_client or :TLSv1_1, :TLSv1_1_server
+    :TLSv1_1_client. The version being effectively used can be queried
+    with OpenSSL::SSL#ssl_version. Furthermore, it is also possible to
+    blacklist the new TLS versions with OpenSSL::SSL:OP_NO_TLSv1_1 and
+    OpenSSL::SSL::OP_NO_TLSv1_2.
+  * Added OpenSSL::SSL::SSLContext#renegotiation_cb. A user-defined callback
+    may be set which gets called whenever a new handshake is negotiated. This
+    also allows to programmatically decline (client) renegotiation attempts.
+  * Support for "0/n" splitting of records as BEAST mitigation via
+    OpenSSL::SSL::OP_DONT_INSERT_EMPTY_FRAGMENTS.
+  * The default options for OpenSSL::SSL::SSLContext have changed to
+    OpenSSL::SSL::OP_ALL & ~OpenSSL::SSL::OP_DONT_INSERT_EMPTY_FRAGMENTS
+    instead of OpenSSL::SSL::OP_ALL only. This enables the countermeasure for
+    the BEAST attack by default.
+  * OpenSSL requires passwords for decrypting PEM-encoded files to be at least
+    four characters long. This led to awkward situations where an export with
+    a password with fewer than four characters was possible, but accessing the
+    file afterwards failed. OpenSSL::PKey::RSA, OpenSSL::PKey::DSA and
+    OpenSSL::PKey::EC therefore now enforce the same check when exporting a
+    private key to PEM with a password - it has to be at least four characters
+    long.
+  * SSL/TLS support for the Next Protocol Negotiation extension. Supported
+    with OpenSSL 1.0.1 and higher.
+  * OpenSSL::OPENSSL_FIPS allows client applications to detect whether OpenSSL
+    is FIPS-enabled. OpenSSL.fips_mode= allows turning on and off FIPS mode
+    manually in order to adapt to situations where FIPS mode would be an
+    explicit requirement.
+  * Authenticated Encryption with Associated Data (AEAD) is supported via
+    Cipher#auth_data= and Cipher#auth_tag/Cipher#auth_tag=.
+    Currently (OpenSSL 1.0.1c), only GCM mode is supported.
+
+* ostruct
+  * new methods:
+    * OpenStruct#[], []=
+    * OpenStruct#each_pair
+    * OpenStruct#eql?
+    * OpenStruct#hash
+    * OpenStruct#to_h converts the struct to a hash.
+  * extended method:
+    * OpenStruct.new also accepts an OpenStruct / Struct.
+
+* pathname
+  * extended method:
+    * Pathname#find returns an enumerator if no block is given.
+
+* rake
+  * rake has been updated to version 0.9.5.
+
+    This version is backwards-compatible with previous rake versions and
+    contains many bug fixes.
+
+    See
+    http://rake.rubyforge.org/doc/release_notes/rake-0_9_5_rdoc.html for a list
+    of changes in rake 0.9.3, 0.9.4 and 0.9.5.
+
+* RDoc
+  * RDoc has been updated to version 4.0
+
+    This version is largely backwards-compatible with previous rdoc versions.
+    The most notable change is an update to the ri data format (ri data must
+    be regenerated for gems shared across rdoc versions).  Further API changes
+    are internal and won't affect most users.
+
+    Notable changes include:
+
+    * Page support for ri.  Try `ri ruby:` for a list of pages in ruby or
+      `ri ruby:syntax/literals` for the syntax documentation for literals.
+
+      This also works for gems such as `ri rspec:README` for the rspec gem's
+      README file.
+    * Markdown support.  See ri RDoc::Markdown for details.
+
+    See https://github.com/rdoc/rdoc/blob/master/History.rdoc for a full list
+    of changes in rdoc 4.0.
+
+* resolv
+  * new methods:
+    * Resolv::DNS#timeouts=
+    * Resolv::DNS::Config#timeouts=
+
+* rexml
+  * REXML::Document#write supports Hash arguments.
+  * REXML::Document#write supports new :encoding option. It changes
+    XML document encoding. Without :encoding option, encoding in
+    XML declaration is used for XML document encoding.
+
+* RubyGems
+  * Updated to 2.0.0
+
+    RubyGems 2.0.0 features the following improvements:
+
+    * Improved support for default gems shipping with ruby 2.0.0+
+    * A gem can have arbitrary metadata through Gem::Specification#metadata
+    * `gem search` now defaults to --remote and is anchored like gem list.
+    * Added --document to replace --rdoc and --ri.  Use --no-document to
+      disable documentation, --document=rdoc to only generate rdoc.
+    * Only ri-format documentation is generated by default.
+    * `gem server` uses RDoc::Servlet from RDoc 4.0 to generate HTML
+      documentation.
+
+    For an expanded list of updates and bug fixes see:
+    https://github.com/rubygems/rubygems/blob/master/History.txt
+
+* shellwords
+  * Shellwords#shellescape now stringifies the given object using to_s.
+  * Shellwords#shelljoin accepts non-string objects in the given
+    array, each of which is stringified using to_s.
+
+* stringio
+  * deprecated methods:
+    * StringIO#lines, #bytes, #chars and #codepoints are deprecated.
+
+* syslog
+  * Added Syslog::Logger which provides a Logger API atop Syslog.
+  * Syslog::Priority, Syslog::Level, Syslog::Option and Syslog::Macros
+    are introduced for easy detection of available constants on a
+    running system.
+
+* tmpdir
+  * incompatible changes:
+    * Dir.mktmpdir uses FileUtils.remove_entry instead of
+      FileUtils.remove_entry_secure.  This means that applications should not
+      change the permission of the created temporary directory to make
+      writable from other users.
+
+* yaml
+  * Syck has been removed.  YAML now completely depends on libyaml being
+    installed.
+  * libyaml is now bundled with ruby, for cases where the library is not
+    installed locally.
+
+* zlib
+  * Added streaming support for Zlib::Inflate and Zlib::Deflate.  This allows
+    processing of a stream without the use of large amounts of memory.
+  * Added support for the new deflate strategies Zlib::RLE and Zlib::FIXED.
+  * Zlib streams are now processed without the GVL.  This allows gzip, zlib and
+    deflate streams to be processed in parallel.
+  * deprecated methods:
+    * Zlib::GzipReader#lines and #bytes are deprecated.
+
+=== Stdlib compatibility issues (excluding feature bug fixes)
+
+* OpenStruct new methods can conflict with custom attributes named
+  "each_pair", "eql?", "hash" or "to_h".
+
+* Dir.mktmpdir in lib/tmpdir.rb
+
+  See above.
+
+=== C API updates
+
+* NUM2SHORT() and NUM2USHORT() added. They are similar to NUM2INT, but short.
+
+* rb_newobj_of() and NEWOBJ_OF() added. They create a new object of a given class.
diff --git a/doc/NEWS-2.1.0 b/doc/NEWS-2.1.0
new file mode 100644
index 0000000000..5d4152b8dc
--- /dev/null
+++ b/doc/NEWS-2.1.0
@@ -0,0 +1,376 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 2.1.0
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file.
+
+== Changes since the 2.0.0 release
+
+=== Language changes
+
+* Now the default values of keyword arguments can be omitted.  Those
+  "required keyword arguments" need giving explicitly at the call time.
+
+* Added suffixes for integer and float literals: 'r', 'i', and 'ri'.
+  * "42r" and "3.14r" are evaluated as Rational(42, 1) and 3.14.rationalize,
+    respectively.  But exponential form with 'r' suffix like "6.022e+23r" is
+    not accepted because it is misleading.
+  * "42i" and "3.14i" are evaluated as Complex(0, 42) and Complex(0, 3.14),
+    respectively.
+  * "42ri" and "3.14ri" are evaluated as Complex(0, 42r) and Complex(0, 3.14r),
+    respectively.
+
+* def-expr now returns the symbol of its name instead of nil.
+
+=== Core classes updates (outstanding ones only)
+
+* Array
+  * New methods
+    * Array#to_h converts an array of key-value pairs into a Hash.
+
+* Binding
+  * New methods
+    * Binding#local_variable_get(symbol)
+    * Binding#local_variable_set(symbol, obj)
+    * Binding#local_variable_defined?(symbol)
+
+* Enumerable
+  * New methods
+    * Enumerable#to_h converts a list of key-value pairs into a Hash.
+
+* Exception
+  * New methods
+    * Exception#cause provides the previous exception which has been caught
+      at where raising the new exception.
+
+* GC
+  * improvements:
+    * introduced the generational GC a.k.a RGenGC.
+  * added environment variables:
+    * RUBY_GC_HEAP_INIT_SLOTS
+    * RUBY_GC_HEAP_FREE_SLOTS
+    * RUBY_GC_HEAP_GROWTH_FACTOR
+    * RUBY_GC_HEAP_GROWTH_MAX_SLOTS
+    * RUBY_GC_MALLOC_LIMIT_MAX
+    * RUBY_GC_MALLOC_LIMIT_GROWTH_FACTOR
+    * RUBY_GC_OLDMALLOC_LIMIT
+    * RUBY_GC_OLDMALLOC_LIMIT_MAX
+    * RUBY_GC_OLDMALLOC_LIMIT_GROWTH_FACTOR
+  * obsoleted environment variables:
+    * RUBY_FREE_MIN (Use RUBY_GC_HEAP_FREE_SLOTS instead)
+    * RUBY_HEAP_MIN_SLOTS (Use RUBY_GC_HEAP_INIT_SLOTS instead)
+
+* Integer
+  * New methods
+    * Fixnum#bit_length
+    * Bignum#bit_length
+  * Bignum performance improvement
+    * Use GMP if available.
+      GMP is used only for several operations:
+      multiplication, division, radix conversion, GCD
+
+* IO
+  * extended methods:
+    * IO#seek supports SEEK_DATA and SEEK_HOLE as whence.
+    * IO#seek accepts symbols (:CUR, :END, :SET, :DATA, :HOLE) for 2nd argument.
+    * IO#read_nonblock accepts optional `exception: false` to return symbols
+    * IO#write_nonblock accepts optional `exception: false` to return symbols
+
+* Kernel
+  * New methods:
+    * Kernel#singleton_method
+
+* Module
+  * New methods:
+    * Module#using, which activates refinements of the specified module only
+      in the current class or module definition.
+    * Module#singleton_class? returns true if the receiver is a singleton class
+      or false if it is an ordinary class or module.
+  * extended methods:
+    * Module#refine is no longer experimental.
+    * Module#include and Module#prepend are now public methods.
+
+* Mutex
+  * misc
+    * Mutex#owned? is no longer experimental.
+
+* Numeric
+  * extended methods:
+    * Numeric#step allows the limit argument to be omitted, in which
+      case an infinite sequence of numbers is generated.  Keyword
+      arguments `to` and `by` are introduced for ease of use.
+      `by` can be 0, in which case the same value will be generated
+      indefinitely.
+
+* Process
+  * New methods:
+    * alternative methods to $0/$0=:
+      * Process.argv0() returns the original value of $0.
+      * Process.setproctitle() sets the process title without affecting $0.
+    * Process.clock_gettime
+    * Process.clock_getres
+
+* String
+  * "literal".freeze is now optimized to return the same object
+  * New methods:
+    * String#scrub and String#scrub! verify and fix invalid byte sequence.
+      If you want to use this function with older Ruby,
+      consider to use string-scrub.gem.
+
+* Symbol
+  * All symbols are now frozen.
+
+* pack/unpack (Array/String)
+  * Q! and q! directives for long long type if platform has the type.
+
+* toplevel
+  * extended methods:
+    * main.using is no longer experimental. The method activates refinements
+      in the ancestors of the argument module to support refinement
+      inheritance by Module#include.
+
+=== Core classes compatibility issues (excluding feature bug fixes)
+
+* Hash
+  * incompatible changes:
+    * Hash#reject will return plain Hash object in the future versions, that
+      is the original object's subclass, instance variables, default value,
+      and taintedness will be no longer copied, so now warnings are emitted
+      when called with such Hash.
+
+* IO
+  * incompatible changes:
+    * open ignore internal encoding if external encoding is ASCII-8BIT.
+
+* Kernel#eval, Kernel#instance_eval, and Module#module_eval.
+  * Copies the scope information of the original environment, which means
+    that private, protected, public, and module_function without arguments
+    do not affect the environment outside the eval string.
+    For example, `class Foo; eval "private"; def foo; end; end' doesn't make
+    Foo#foo private.
+
+* Kernel#untrusted?, untrust, and trust
+  * These methods are deprecated and their behavior is same as tainted?,
+    taint, and untaint, respectively.  If $VERBOSE is true, they show warnings.
+
+* Module#ancestors
+  * The ancestors of a singleton class now include singleton classes,
+    in particular itself.
+
+* Module#define_method and Object#define_singleton_method
+  * Now they return the symbols of the defined methods, not the methods/procs
+    themselves.
+
+* Numeric#quo
+  * Raises TypeError instead of ArgumentError if the receiver doesn't have
+    to_r method.
+
+* Proc
+  * Returning from lambda proc now always exits from the Proc, not from the
+    method where the lambda is created.  Returning from non-lambda proc exits
+    from the method, same as the former behavior.
+
+String
+  * If invalid: :replace is specified for String#encode, replace
+    invalid byte sequence even if the destination encoding equals to
+    the source encoding.
+
+=== Stdlib updates (outstanding ones only)
+
+* CGI::Util
+  * All class methods modulized.
+
+* Digest
+  * extended methods:
+    * Digest::Class.file takes optional arguments for its constructor
+
+* Matrix
+  * Added Vector#cross_product.
+
+* Net::SMTP
+  * Added Net::SMTP#rset to implement the RSET command
+
+* objspace
+  * new method:
+    * ObjectSpace.trace_object_allocations
+    * ObjectSpace.trace_object_allocations_start
+    * ObjectSpace.trace_object_allocations_stop
+    * ObjectSpace.trace_object_allocations_clear
+    * ObjectSpace.allocation_sourcefile
+    * ObjectSpace.allocation_sourceline
+    * ObjectSpace.allocation_class_path
+    * ObjectSpace.allocation_method_id
+    * ObjectSpace.allocation_generation
+    * ObjectSpace.reachable_objects_from_root
+    * ObjectSpace.dump
+    * ObjectSpace.dump_all
+
+* OpenSSL::BN
+  * extended methods:
+    * OpenSSL::BN.new allows Fixnum/Bignum argument.
+
+* open-uri
+  * Support multiple fields with same field name (like Set-Cookie).
+
+* Pathname
+  * New methods:
+    * Pathname#write
+    * Pathname#binwrite
+
+* rake
+  * Updated to 10.1.0.  Major changes include removal of the class namespace,
+    Rake::DSL to hold the rake DSL methods and removal of support for legacy
+    rake features.
+
+    For a complete list of changes since rake 0.9.6 see:
+
+    http://rake.rubyforge.org/doc/release_notes/rake-10_1_0_rdoc.html
+
+    http://rake.rubyforge.org/doc/release_notes/rake-10_0_3_rdoc.html
+
+* RbConfig
+  * New constants:
+    * RbConfig::SIZEOF is added to provide the size of C types.
+
+* RDoc
+  * Updated to 4.1.0.  Major enhancements include a modified default template
+  * and accessibility enhancements.
+
+    For a list of minor enhancements and bug fixes see:
+    https://github.com/rdoc/rdoc/blob/v4.1.0.preview.1/History.rdoc
+
+* Resolv
+  * New methods:
+    * Resolv::DNS.fetch_resource
+  * One-shot multicast DNS support
+  * Support LOC resources
+
+* REXML::Parsers::SAX2Parser
+  * Fixes wrong number of arguments of entitydecl event. Document of the event
+    says "an array of the entity declaration" but implementation passes two
+    or more arguments. It is an implementation bug but it breaks backward
+    compatibility.
+
+* REXML::Parsers::StreamParser
+  * Supports "entity" event.
+
+* REXML::Text
+  * REXML::Text#<< supports method chain like 'text << "XXX" << "YYY"'.
+  * REXML::Text#<< supports not "raw" mode.
+
+* Rinda::RingServer, Rinda::RingFinger
+  * Rinda now supports multicast sockets.  See Rinda::RingServer and
+    Rinda::RingFinger for details.
+
+* RubyGems
+  * Updated to 2.2.0.  Notable new features include:
+
+    * Gemfile or gem.deps.rb support including Gem.file.lock (experimental)
+    * Improved, iterative resolver (compared to RubyGems 2.1 and earlier)
+    * Support for a sharing a GEM_HOME across ruby platforms and versions
+
+    For a complete list of enhancements and bug fixes see:
+    https://github.com/rubygems/rubygems/tree/master/History.txt
+
+* Set
+  * New methods:
+    * Set#intersect?
+    * Set#disjoint?
+
+* Socket
+  * New methods:
+    * Socket.getifaddrs
+
+* StringScanner
+  * extended methods:
+    * StringScanner#[] supports named captures.
+
+* Syslog::Logger
+  * Added facility.
+
+* Tempfile
+  * New methods:
+    * Tempfile.create
+
+* Timeout
+  * The exception to terminate the given block can no longer be rescued
+    inside the block, by default, unless the exception class is given
+    explicitly.
+
+* TSort
+  * New methods:
+    * TSort.tsort
+    * TSort.tsort_each
+    * TSort.strongly_connected_components
+    * TSort.each_strongly_connected_component
+    * TSort.each_strongly_connected_component_from
+
+* WEBrick
+  * The body of a response may now be a StringIO or other IO-like that responds
+    to #readpartial and #read.
+
+* XMLRPC::Client
+  * New methods:
+    * XMLRPC::Client#http. It returns Net::HTTP for the client. Normally,
+      it is not needed. It is useful when you want to change minor HTTP client
+      options. You can change major HTTP client options by XMLRPC::Client
+      methods. You should use XMLRPC::Client methods for changing major
+      HTTP client options instead of XMLRPC::Client#http.
+
+=== Stdlib compatibility issues (excluding feature bug fixes)
+
+* Set
+  * incompatible changes:
+    * Set#to_set now returns self instead of generating a copy.
+
+* URI
+  * incompatible changes:
+    * URI.decode_www_form follows current WHATWG URL Standard.
+      It gets encoding argument to specify the character encoding.
+      It now allows loose percent encoded strings, but denies ;-separator.
+    * URI.encode_www_form follows current WHATWG URL Standard.
+      It gets encoding argument to convert before percent encode.
+      UTF-16 strings aren't converted to UTF-8 before percent encode by default.
+
+* curses
+  * Removed.
+    curses is now available as a gem.
+    See https://rubygems.org/gems/curses for details.
+
+=== Built-in global variables compatibility issues
+
+* $SAFE
+  * $SAFE=4 is obsolete.  If $SAFE is set to 4 or larger, an ArgumentError
+    is raised.
+
+=== C API updates
+
+* rb_gc_set_params() is deprecated. This is only used in Ruby internal.
+
+* rb_gc_count() added. This returns the number of times GC occurred.
+
+* rb_gc_stat() added. This allows access to specific GC.stat() values from C
+  without any allocation overhead.
+
+* rb_gc_latest_gc_info() added. This allows access to GC.latest_gc_info().
+
+* rb_postponed_job_register() added. Takes a function callback which is invoked
+  when the VM is in a consistent state, i.e. to perform work from a C signal
+  handler.
+
+* rb_profile_frames() added. Provides low-cost access to the current ruby stack
+  for callstack profiling.
+
+* rb_tracepoint_new() supports new internal events accessible only from C:
+  * RUBY_INTERNAL_EVENT_NEWOBJ
+  * RUBY_INTERNAL_EVENT_FREEOBJ
+  * RUBY_INTERNAL_EVENT_GC_START
+  * RUBY_INTERNAL_EVENT_GC_END_MARK
+  * RUBY_INTERNAL_EVENT_GC_END_SWEEP
+  * Note that you *can not* specify "internal events" with normal events
+    (such as RUBY_EVENT_CALL, RUBY_EVENT_RETURN) simultaneously.
diff --git a/doc/NEWS-2.2.0 b/doc/NEWS-2.2.0
new file mode 100644
index 0000000000..5564c606ae
--- /dev/null
+++ b/doc/NEWS-2.2.0
@@ -0,0 +1,361 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 2.2.0
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file.
+
+== Changes since the 2.1.0 release
+
+=== Language changes
+
+* nil/true/false
+  * nil/true/false objects are frozen. [Feature #8923]
+
+* Hash literal
+  * Symbol key followed by a colon can be quoted. [Feature #4276]
+
+* default argument
+  fixed a very longstanding bug that an optional argument was not
+  accessible in its default value expression.  [Bug #9593]
+
+=== Core classes updates (outstanding ones only)
+
+* Binding
+  * New methods:
+    * Binding#local_variables
+    * Binding#receiver
+
+* Dir
+  * New methods:
+    * Dir#fileno
+
+* Enumerable
+  * New methods:
+    * Enumerable#slice_after
+    * Enumerable#slice_when
+  * Extended methods:
+    * min, min_by, max and max_by supports optional argument to return
+      multiple elements.
+
+* Float
+  * New methods:
+    * Float#next_float
+    * Float#prev_float
+
+* File
+  * New methods:
+    * File.birthtime
+    * File#birthtime
+
+* File::Stat
+  * New methods:
+    * File::Stat#birthtime
+
+* GC
+  * GC.latest_gc_info returns :state to represent current GC status.
+  * Improvements
+    * Introduce incremental marking for major GC. [Feature #10137]
+
+* IO
+  * Improvements
+    * IO#read_nonblock and IO#write_nonblock for pipes on Windows are supported.
+
+* Kernel
+  * New methods:
+    * Kernel#itself
+  * Improvements
+    * Kernel#throw raises UncaughtThrowError, subclass of ArgumentError when
+      there is no corresponding catch block, instead of ArgumentError.
+
+* Process
+  * Extended method:
+    * Process execution methods such as Process.spawn opens the file in write
+      mode for redirect from [:out, :err].
+
+* String
+  * New methods:
+    * String#unicode_normalize
+    * String#unicode_normalize!
+    * String#unicode_normalized?
+
+* Symbol
+  * Improvements
+    * Most symbols which are returned by String#to_sym and
+      String#intern are GC-able.
+
+* Method
+  * New methods:
+    * Method#curry([arity]) returns a curried Proc.
+    * Method#super_method returns a Method of superclass, which would be called
+      when super is used.
+
+=== Core classes compatibility issues (excluding feature bug fixes)
+
+* Enumerable
+  * Enumerable#slice_before's state management deprecated.
+  * Enumerable#chunk's state management deprecated.
+
+* GC
+  * incompatible changes:
+    * Rename GC.stat entries. [Feature #9924]
+      See https://docs.google.com/spreadsheets/d/11Ua4uBr6o0k-nORrZLEIIUkHJ9JRzRR0NyZfrhEEnc8/edit?usp=sharing
+
+* Hash
+  * incompatible changes:
+    * Change overriding policy for duplicated key. [Bug #10315]
+      { **hash1, **hash2 } contains values of hash2 for duplicated keys.
+
+* IO
+  * incompatible changes:
+    * When flushing file IO with IO#flush, you cannot assume that the metadata
+      of the file is updated immediately.  On some platforms (especially
+      Windows), it is delayed until the filesystem load is decreased.
+      Use IO#fsync if you want to guarantee updating metadata.
+
+* Math
+  * incompatible changes:
+    * Math.log now raises Math::DomainError instead of returning NaN if the
+      base is less than 0, and returns NaN instead of -infinity if both of
+      two arguments are 0.
+    * Math.atan2 now returns values like as expected by C99 if both two
+      arguments are infinity.
+
+* Proc
+  * incompatible changes:
+    * ArgumentError is no longer raised when lambda Proc is passed as a
+      block, and the number of yielded arguments does not match the formal
+      arguments of the lambda, if just an array is yielded and its length
+      matches.
+
+* Process
+  * Process execution methods such as Process.spawn opens the file in write
+    mode for redirect from [:out, :err].
+    Before Ruby 2.2, it was opened in read mode.
+
+=== Stdlib updates (outstanding ones only)
+
+* Continuation
+  * callcc is obsolete. use Fiber instead.
+
+* Digest
+
+  * Digest() should now be thread-safe.  If you have a problem with
+    regard to on-demand loading under a multi-threaded environment,
+    preload "digest/*" modules on boot or use this method instead of
+    directly referencing Digest::*.
+  * Digest::HMAC has been removed just as previously noticed.
+
+* DL
+  * DL has been removed from stdlib. Please use Fiddle instead!
+
+* Etc
+  * New methods:
+    * Etc.uname
+    * Etc.sysconf
+    * Etc.confstr
+    * IO#pathconf
+    * Etc.nprocessors
+
+* Find, Pathname
+  * Extended methods:
+    * find method accepts "ignore_error" keyword argument.
+
+* Matrix
+  * New methods:
+    * Matrix#first_minor(row, column) returns the submatrix obtained
+      by deleting the specified row and column.
+    * Matrix#cofactor(row, column) returns the (row, column) cofactor
+      which is obtained by multiplying the first minor by (-1)**(row + column).
+    * Matrix#adjugate returns the adjugate of the matrix.
+    * hstack and vstack are new instance and class methods to stack matrices
+      horizontally and vertically.
+    * Matrix#laplace_expansion(row_or_column: num) returns the laplace_expansion
+      along the +num+ -th row or column.
+    * Vector.basis(size:, index:) returns the specified basis vector.
+    * Unary - and + added for Vector and Matrix.
+    * Vector#cross_product generalized to arbitrary dimensions.
+    * Vector#dot and #cross are aliases for #inner_product and #cross_product.
+    * Vector#angle_with returns the angle with its argument
+    * New instance and class method independent? to test linear independence.
+
+* Pathname
+  * Pathname#/ is aliased to Pathname#+.
+  * New methods:
+    * Pathname#birthtime
+
+* Rake
+  * Updated to Rake 10.4.0.  For full release notes see:
+
+    http://docs.seattlerb.org/rake/History_rdoc.html#label-10.4.0
+
+* RubyGems
+  * Updated to RubyGems 2.4.2.  For full release notes see:
+
+    http://docs.seattlerb.org/rubygems/History_txt.html#label-2.4.2+%2F+2014-10-01
+
+* TSort
+  * TSort.tsort_each, TSort.each_strongly_connected_component and
+    TSort.each_strongly_connected_component_from returns an enumerator if
+    no block given.
+
+* XMLRPC
+  * Added new parser class named LibXMLStreamParser.
+
+=== Stdlib compatibility issues (excluding feature bug fixes)
+
+* lib/mathn.rb
+  * Show deprecated warning [Feature #10169]
+
+* ext/date/lib/date/format.rb
+  * Removed because it's empty file.
+
+* Digest
+  * Digest::HMAC has finally ceased to exist.  Use OpenSSL::HMAC or an external gem instead.
+
+* time.rb
+  * Time.parse, Time.strptime, Time.rfc2822, Time.xmlschema may produce
+    fixed-offset Time objects.
+    It is happen when usual localtime doesn't preserve the offset from UTC.
+  * Time.httpdate produces always UTC Time object.
+  * Time.strptime raises ArgumentError when no date information.
+
+* lib/rational.rb
+  * Removed because it is deprecated from 2009.
+
+* lib/complex.rb
+  * Removed because it is deprecated from 2009.
+
+* lib/prettyprint.rb
+  * Removed PrettyPrint#first?
+
+* lib/minitest/*.rb
+  * Removed because it conflicts to minitest 5. [Feature #9711]
+
+* lib/test/**/*.rb
+  * Removed because it conflicts to minitest 5, and it was just an wrapper
+    of minitest 4. [Feature #9711]
+
+* lib/uri
+  * support RFC 3986. [Feature #2542]
+
+* GServer
+  * GServer is extracted to gserver gem. It's unmaintain code.
+
+* Logger
+  * Logger::Application is extracted to logger-application gem. It's unmaintain code.
+
+* ObjectSpace (after requiring "objspace")
+  * ObjectSpace.memsize_of(obj) returns a size includes sizeof(RVALUE).
+    [Bug #8984]
+
+* Prime
+  * incompatible changes:
+    * Prime.prime? now returns false for negative numbers. This method
+      should not be used to know the number is composite or not. [Bug #7395]
+
+* Psych
+  * Removed Psych::EngineManager [Bug #8344]
+
+=== Built-in global variables compatibility issues
+
+=== C API updates
+
+* Deprecated APIs removed.  [Feature #9502]
+
+  Check_SafeStr -> SafeStringValue
+  rb_check_safe_str -> SafeStringValue
+  rb_quad_pack -> rb_integer_pack
+  rb_quad_unpack -> rb_integer_unpack
+  rb_read_check : access struct FILE internal. no replacement.
+  rb_struct_iv_get : internal function. no replacement.
+  struct rb_blocking_region_buffer : internal type. no replacement.
+  rb_thread_blocking_region_begin -> rb_thread_call_without_gvl family
+  rb_thread_blocking_region_end -> rb_thread_call_without_gvl family
+  TRAP_BEG -> rb_thread_call_without_gvl family
+  TRAP_END -> rb_thread_call_without_gvl family
+  rb_thread_select -> rb_thread_fd_select
+  struct rb_exec_arg : internal type. no replacement.
+  rb_exec : internal function. no replacement.
+  rb_exec_arg_addopt : internal function. no replacement.
+  rb_exec_arg_fixup : internal function. no replacement.
+  rb_exec_arg_init : internal function. no replacement.
+  rb_exec_err : internal function. no replacement.
+  rb_fork : internal function. no replacement.
+  rb_fork_err : internal function. no replacement.
+  rb_proc_exec_n : internal function. no replacement.
+  rb_run_exec_options : internal function. no replacement.
+  rb_run_exec_options_err : internal function. no replacement.
+  rb_thread_blocking_region -> rb_thread_call_without_gvl family
+  rb_thread_polling -> rb_thread_wait_for
+  rb_big2str0 : internal function. no replacement.
+  rb_big2ulong_pack -> rb_integer_pack
+  rb_gc_set_params : internal function. no replacement.
+  rb_io_mode_flags -> rb_io_modestr_fmode
+  rb_io_modenum_flags -> rb_io_oflags_fmode
+
+* struct RBignum is hidden.  [Feature #6083]
+  Use rb_integer_pack and rb_integer_unpack instead.
+
+* struct RRational is hidden.  [Feature #9513]
+  Use rb_rational_num and rb_rational_den instead.
+
+* rb_big_new and rb_big_resize takes a size_t instead of long.
+
+* rb_num2long returns a long instead of SIGNED_VALUE.
+
+* rb_num2ulong returns an unsigned long instead of VALUE.
+
+* st hash table uses power-of-two sizes for speed [Feature #9425].
+  Lookups are 10-25% faster if using appropriate hash functions.
+  However, weaknesses in hash distribution can no longer be masked
+  by prime number-sized tables, so extensions may need to tweak
+  hash functions to ensure good distribution.
+
+* rb_sym2str() added. This is almost same as `rb_id2str(SYM2ID(sym))`
+  but not pinning a dynamic symbol.
+
+* rb_str_cat_cstr() added. This is same as `rb_str_cat2()`.
+
+* `rb_str_substr()` and `rb_str_subseq()` will share middle of a string,
+  but not only the end of a string, in the future.  Therefore, result
+  strings may not be NUL-terminated, `StringValueCStr()` is needed
+  calling to obtain a NUL-terminated C string.
+
+* rb_tracepoint_new() supports new internal events accessible only from C:
+  * RUBY_INTERNAL_EVENT_GC_ENTER
+  * RUBY_INTERNAL_EVENT_GC_EXIT
+    r47528
+
+* rb_hash_delete() now does not call the block given to the current method.
+
+* rb_extract_keywords() and rb_get_kwargs() exported.  See README.EXT
+  for details.
+
+=== Build system updates
+
+* jemalloc is optionally supported via `./configure --with-jemalloc`
+  jemalloc may be suitable when system malloc is slow or prone
+  to fragmentation.  [Feature #9113]
+
+=== Implementation changes
+
+* GC
+    * Most symbols which are returned by String#to_sym and
+      String#intern are GC-able [Feature #9634]
+    * Introduce incremental marking for major GC. [Feature #10137]
+    * Enable lazy sweep on GC caused by malloc().
+
+* VM
+    * Use frozen string literals for Hash#[] and Hash#[]=
+    * Fast keyword arguments passing [Feature #10440]
+    * Allow to receive huge splatted array by a rest argument
+      [Feature #10440]
+
+* Process
+    * Process creation methods, such as spawn(), uses vfork() system call.
+      vfork() is faster than fork() when the parent process uses huge memory.
diff --git a/doc/NEWS-2.3.0 b/doc/NEWS-2.3.0
new file mode 100644
index 0000000000..489aba4a89
--- /dev/null
+++ b/doc/NEWS-2.3.0
@@ -0,0 +1,404 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 2.3.0
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file or Redmine
+(e.g. <tt>https://bugs.ruby-lang.org/issues/$FEATURE_OR_BUG_NUMBER</tt>)
+
+== Changes since the 2.2.0 release
+
+=== Language changes
+
+* frozen-string-literal pragma:
+
+  * new pragma, frozen-string-literal has been experimentally introduced.
+    [Feature #8976]
+  * besides, --enable/--disable=frozen-string-literal options also have
+    been introduced. [Feature #8976]
+  * command line options --debug or --debug=frozen-string-literal enable
+    additional debugging mode which shows created location with at frozen
+    object error (RuntimeError).
+    [Feature #11725]
+
+* safe navigation operator:
+
+  * new method call syntax, `object&.foo', method #foo is called on
+    `object' if it is not nil.
+    this is similar to `try!' in Active Support, except:
+    * method name is syntactically required
+          obj.try! {} # valid
+          obj&. {}    # syntax error
+    * arguments are evaluated only if a call is made:
+          obj.try!(:foo, bar())  # bar() is always evaluated
+          obj&.foo(bar())        # bar() is conditionally evaluated
+    * attribute assignment is valid
+          obj&.attr += 1
+    [Feature #11537]
+
+* the did_you_mean gem:
+
+  * When a NameError or NoMethodError occurs because of a typo in the name,
+    the did_you_mean gem automatically suggests other names similar to the
+    method name.
+
+      "Yuki".starts_with?("Y")
+      # => NoMethodError: undefined method `starts_with?' for "Yuki":String
+      #    Did you mean?  start_with?
+
+* indented here document:
+
+  * new string literal, here document starts with `<<~`.
+    refer doc/syntax/literals.rdoc for more details.
+    [Feature #9098]
+
+=== Core classes updates (outstanding ones only)
+
+* ARGF
+
+  * ARGF.read_nonblock supports `exception: false' like IO#read_nonblock.
+    [Feature #11358]
+
+* Array
+
+  * Array#bsearch_index [Feature #10730]
+  * Array#dig [Feature #11643]
+
+* Comparable
+
+  * Comparable#== no longer rescues exceptions [Feature #7688]
+
+* Encoding
+
+  * new Encoding::IBM037 (alias ebcdic-cp-us; dummy)
+
+* Enumerable
+
+  * Enumerable#grep_v is added as inverse version of Enumerable#grep.
+    [Feature #11049]
+  * Enumerable#chunk_while [Feature #10769]
+
+* Enumerator::Lazy
+
+  * Enumerator::Lazy#grep_v [Feature #11773]
+
+* File
+
+  * File.mkfifo [Feature #11536]
+  * Add File::TMPFILE corresponding to O_TMPFILE
+
+* Hash
+
+  * Hash#fetch_values [Feature #10017]
+  * Hash#dig [Feature #11643]
+  * Hash#<=, Hash#<, Hash#>=, Hash#> [Feature #10984]
+  * Hash#to_proc [Feature #11653]
+
+* IO
+
+  * new mode flag File::SHARE_DELETE is available.
+    this flag means to permit deleting opened file on Windows, but currently
+    this affect only files opened as binary.  [Feature #11218]
+
+  * new option parameter `flags' is added.
+    this parameter is bitwise-ORed to oflags generated by normal mode argument.
+    [Feature #11253]
+
+  * IO#advise no longer raises Errno::ENOSYS in cases where it was
+    detected at build time but not available at runtime.  [Feature #11806]
+
+* Kernel
+
+  * Kernel#loop, when stopped by a StopIteration exception, returns
+    what the enumerator has returned instead of nil. [Feature #11498]
+
+* Module
+  * Module#deprecate_constant [Feature #11398]
+
+* NameError
+  * NameError#receiver is added to take the receiver object. [Feature #10881]
+
+* Numeric
+
+  * Numeric#positive? and Numeric#negative? are added, which return
+    true when the receiver is positive and negative respectively.
+    [Feature #11151]
+
+* Proc
+
+  * Proc#call (and also #[], #===, #yield) are optimized.
+    Backtrace doesn't show each method (show block lines directly).
+    TracePoint also ignores these calls. [Feature #11569]
+
+* Queue (Thread::Queue)
+
+  * Queue#close is added to notice a termination. [Feature #10600]
+
+* Regexp/String: Updated Unicode version from 7.0.0 to 8.0.0
+
+* RubyVM::InstructionSequence
+  * add the following methods as a primitive tool of iseq loader.
+    See sample/iseq_loader.rb for usage.
+    Note that loader does not have verifier so it is easy to cause
+    critical problem by loading modified/broken binary data.
+    See [Feature #11788] for more details. (experimental feature)
+    * RubyVM::InstructionSequence#to_binary(extra_data = nil)
+    * RubyVM::InstructionSequence.load_from_binary(binary)
+    * RubyVM::InstructionSequence.load_from_binary_extra_data(binary)
+
+* String
+
+  * String#+@ and String#-@ are added to get mutable/frozen strings.
+    [Feature #11782]
+
+  * String.new now accepts new option parameter `encoding'.
+    [Feature #11785]
+
+* Struct
+  * Struct#dig [Feature #11688]
+
+* Thread
+  * Thread#name, Thread#name= are added to handle thread names [Feature #11251]
+
+=== Core classes compatibility issues (excluding feature bug fixes)
+
+* Array
+  * Array#select!, Array#keep_if, Array#reject!, and Array#delete_if
+    no longer changes the receiver array instantly every time the
+    block is called.  [Feature #10714]
+
+  * Array#flatten and Array#flatten! no longer try to call #to_ary
+    method on elements beyond the given level.  [Bug #10748]
+
+  * Array#inspect doesn't raise error even if its content returns
+    a string which is not compatible with Encoding.default_external
+    as inspected result. [Feature #11801]
+
+* Enumerable
+  * Enumerable#chunk and Enumerable#slice_before no longer takes the
+    initial_state argument.  [Feature #10958]
+    Use a local variable instead to maintain a state.
+
+* File::Stat
+  * On Windows File::Stat#ino always returned 0, but now returns
+    BY_HANDLE_FILE_INFORMATION.nFileIndexHigh/Low.  [Feature #11216]
+
+* Hash
+  * Hash#inspect doesn't raise error even if its content returns
+    a string which is not compatible with Encoding.default_external
+    as inspected result. [Feature #11801]
+
+* IO
+  * IO#close doesn't raise when the IO object is closed.  [Feature #10718]
+  * IO#each_codepoint raises an exception at incomplete character
+    before EOF when conversion takes place.  [Bug #11444]
+
+* Module
+  * Module#define_method and Object.define_singleton_method now
+    require method body, Proc, Method, or a block, and raise
+    ArgumentError if no block is given directly.  [Bug #11283]
+
+* pack/unpack (Array/String)
+  * j and J directives for pointer width integer type.  [Feature #11215]
+
+
+=== Stdlib updates (outstanding ones only)
+
+* Logger
+
+  * Logger#level= now supports symbol and string levels such as :debug, :info,
+    :warn, :error, :fatal (case insensitive) [Feature #11695]
+  * Logger#reopen is added to reopen a log device. [Feature #11696]
+
+* io/wait
+  * IO#wait_readable no longer checks FIONREAD, it may be used for
+    non-bytestream IO such as listen sockets.
+
+* Net::FTP
+  * Net::FTP#mlst is added.
+  * Net::FTP#mlsd is added.
+
+* nkf
+  * Merge nkf 2.1.4.
+
+* ObjectSpace (objspace)
+  * ObjectSpace.count_symbols is added.
+  * ObjectSpace.count_imemo_objects is added.
+  * ObjectSpace.internal_class_of is added.
+  * ObjectSpace.internal_super_of is added.
+
+* OpenSSL
+  * OpenSSL::SSL::SSLSocket#accept_nonblock and
+    OpenSSL::SSL::SSLSocket#connect_nonblock supports `exception: false`.
+    [Feature #10532]
+
+* Pathname
+  * Pathname#descend and Pathname#ascend supported blockless form.
+    [Feature #11052]
+
+* Socket
+  * Socket#connect_nonblock, Socket#accept_nonblock,
+    TCPServer#accept_nonblock, UNIXServer#accept_nonblock,
+    BasicSocket#recv_nonblock, BasicSocket#recvmsg_nonblock,
+    BasicSocket#sendmsg_nonblock all support `exception: false` to return
+    :wait_readable or :wait_writable symbols instead of raising
+    IO::WaitReadable or IO::WaitWritable exceptions
+    [Feature #10532] [Feature #11229]
+  * BasicSocket#recv and BasicSocket#recv_nonblock allow an output
+    String buffer argument like IO#read and IO#read_nonblock to reduce
+    GC overhead [Feature #11242]
+
+* StringIO
+  * In read-only mode, StringIO#set_encoding no longer sets the encoding
+    of its buffer string.  Setting the encoding of the string directly
+    without StringIO#set_encoding may cause unpredictable behavior now.
+    [Bug #11827]
+
+* timeout
+  * Object#timeout is now warned as deprecated when called.
+
+=== Stdlib compatibility issues (excluding feature bug fixes)
+
+* ext/coverage/coverage.c
+  * Coverage.peek_result: new method to allow coverage to be captured without
+    stopping the coverage tool.  [Feature #10816]
+
+* Fiddle
+  * Fiddle::Function#call releases the GVL.  [Feature #11607]
+
+* io-console
+  * Update to io-console 0.4.5, and change the license to BSD 2-clause
+    "Simplified" License.
+
+* lib/base64.rb
+  * Base64.urlsafe_encode64: added a "padding" option to suppress
+    the padding character ("=").  [Feature #10740]
+  * Base64.urlsafe_decode64: now it accepts not only correctly-padded
+    input but also unpadded input.  [Feature #10740]
+
+* lib/drb/drb.rb
+  * removed unused argument. https://github.com/ruby/ruby/pull/515
+
+* lib/matrix.rb
+  * Add Vector#round. https://github.com/ruby/ruby/pull/802
+
+* lib/webrick/utils.rb
+  * removed unused argument. https://github.com/ruby/ruby/pull/356
+
+* Net::FTP
+  * Connections are in passive mode per default now.  The default mode can
+    be changed by Net::FTP.default_passive=.  [Feature #11612]
+
+* Net::HTTP
+  * default value of Net::HTTP#open_timeout is now 60 (was nil).
+
+* Net::Telnet
+  * Net::Telnet is extracted to net-telnet gem. It's unmaintain code.
+    [Feature #11083]
+
+* Psych
+  * Updated to Psych 2.0.17
+
+* Rake
+  * Rake is removed from stdlib.  [Feature #11025]
+
+* RDoc
+  * Updated to RDoc 4.2.1.  For full release notes see:
+
+    https://github.com/rdoc/rdoc/blob/master/History.rdoc#421--2015-12-22
+
+* RubyGems
+  * Updated to RubyGems 2.5.1.  For full release notes see:
+
+    http://docs.seattlerb.org/rubygems/History_txt.html#label-2.5.0+-2F+2015-11-03
+    and
+    http://docs.seattlerb.org/rubygems/History_txt.html#label-2.5.1+-2F+2015-12-10
+
+=== Built-in global variables compatibility issues
+
+* $SAFE
+  * $SAFE=2 and $SAFE=3 are obsolete.  If $SAFE is set to 2 or larger,
+    an ArgumentError is raised.  [Feature #5455]
+
+=== C API updates
+
+* rb_define_class_id_under() now raises a TypeError exception when the
+  class is already defined but its superclass does not match the given
+  superclass, as well as definitions in ruby level.
+
+* rb_timespec_now() is added to fetch current datetime as struct timespec.
+  [Feature #11558]
+
+* rb_time_timespec_new() is added to create a time object with epoch,
+  nanosecond, and UTC/localtime/time offset arguments.  [Feature #11558]
+
+* rb_autoload() deprecated, use rb_funcall() instead.  [Feature #11664]
+
+* rb_compile_error_with_enc(), rb_compile_error(), and rb_compile_bug()
+  deprecated.  these functions are exposed but only for internal use.
+  external libraries should not use them.
+
+=== Supported platform changes
+
+* OS/2 is no longer supported
+
+* BeOS is no longer supported
+
+* Borland-C is no longer supported
+
+* Haiku now stable and best effort
+
+=== Implementation improvements
+
+* Optimize Proc#call to eliminate method frame construction.
+  [Feature #11569]
+
+* Reconsidering method entry data structure.
+  [Bug #11278]
+
+* Introducing new table data structure for ID keys tables used by
+  method table and so on. New table structure is simple and fast
+  than st_table. [Feature #11420]
+
+* Machine code level tuning for object allocation and method calling
+  code. r52099, r52254
+
+* RubyVM::InstructionSequence is extended for future improvement.
+  [Feature #11788]
+
+* Case dispatch is now optimized for all special constant literals
+  including nil, true, and false.  Previously, only literal strings,
+  symbols, integers and floats compiled to optimized case dispatch.
+  [Feature #11769]
+
+* Instance variables on non-pure Ruby classes (T_DATA, T_FILE,
+  etc..) is less expensive to store than before. [Feature #11170]
+
+* All accesses to members of big Struct objects are performed in
+  constant-time.  Previously, Struct elements beyond the first 10
+  elements used a linear scan. [Feature #10585]
+
+* The Set class got several speed up.
+  [Misc #10754], [r52591]
+
+* Socket and I/O-related improvements
+
+  * Calling overhead of most of new keyword-using I/O methods in
+    [Feature #11229] is reduced by avoiding the inefficient C API
+    to parse keywords.  [Feature #11339]
+
+  * The standard library is updated to use the improved
+    exception-free non-blocking I/O from [Feature #11229].
+    This has the additional benefit of quieter $DEBUG output in
+    addition to reducing expensive exceptions. [Feature #11044]
+
+  * (Linux-only) waiting on a single FD anywhere in the stdlib no longer
+    uses select(2), making it immune to slowdowns with high-numbered FDs.
+    [Feature #11081] [Feature #11377]
+
+* CGI.escapeHTML is optimized with C extension.
+  https://github.com/ruby/ruby/pull/1164
diff --git a/doc/NEWS-2.4.0 b/doc/NEWS-2.4.0
new file mode 100644
index 0000000000..28e855cde1
--- /dev/null
+++ b/doc/NEWS-2.4.0
@@ -0,0 +1,397 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 2.4.0
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file or Redmine
+(e.g. <tt>https://bugs.ruby-lang.org/issues/$FEATURE_OR_BUG_NUMBER</tt>)
+
+== Changes since the 2.3.0 release
+
+=== Language changes
+
+* Multiple assignment in conditional expression is now allowed.
+  [Feature #10617]
+
+* Refinements is enabled at method by Symbol#to_proc.  [Feature #9451]
+
+* Refinements is enabled with Kernel#send and BasicObject#__send__.
+  [Feature #11476]
+
+* Rescue modifier now applicable to method arguments.
+  [Feature #12686]
+
+* Toplevel return is now allowed.  [Feature #4840]
+
+=== Core classes updates (outstanding ones only)
+
+* Array
+
+  * Array#concat [Feature #12333]
+    Now takes multiple arguments.
+
+  * Array#max and Array#min.  [Feature #12172]
+    This may cause a tiny incompatibility: if you redefine
+    Enumerable#max and call max to an Array, your redefinition will be
+    now ignored.  You should also redefine Array#max.
+
+  * Array#pack [Feature #12754]
+    Now takes optional argument `buffer:' to reuse already allocated buffer.
+
+  * Array#sum  [Feature #12217]
+    This is different from Enumerable#sum in that Array#sum doesn't depend on
+    the definition of each method.
+
+* Comparable
+
+  * Comparable#clamp.  [Feature #10594]
+
+* Dir
+
+  * Dir.empty?.  [Feature #10121]
+
+* Enumerable
+
+  * Enumerable#chunk called without a block now return an Enumerator
+    [Feature #2172]
+  * Enumerable#sum  [Feature #12217]
+  * Enumerable#uniq  [Feature #11090]
+
+* Enumerator::Lazy
+
+  * Enumerator::Lazy#chunk_while  [GH-1186]
+  * Enumerator::Lazy#uniq  [Feature #11090]
+
+* File
+
+  * File.empty?.  [Feature #9969]
+
+* Float
+
+  * Float#ceil, Float#floor, and Float#truncate now take an optional
+    digits, as well as Float#round.  [Feature #12245]
+
+  * Float#round now takes an optional keyword argument, half option, and
+    the default behavior is round-up.  [Bug #12548] [Bug #12958]
+    half option can be one of :even, :up, and :down.  [Feature #12953]
+
+* Hash
+
+  * Hash#compact and Hash#compact!  [Feature #11818]
+  * Hash#transform_values and Hash#transform_values!  [Feature #12512]
+
+* Integer
+
+  * Fixnum and Bignum are unified into Integer  [Feature #12005]
+
+  * Integer#ceil, Integer#floor, and Integer#truncate now take an optional
+    digits, as well as Integer#round.  [Feature #12245]
+
+  * Integer#digits for extracting columns of place-value notation [Feature #12447]
+
+  * Integer#round now takes an optional keyword argument, half option, and the
+    default behavior is round-up now.  [Bug #12548] [Bug #12958]
+    half option can be one of :even, :up, and :down.  [Feature #12953]
+
+* IO
+
+  * IO#gets, IO#readline, IO#each_line, IO#readlines, IO.foreach now takes
+    an optional keyword argument, chomp flag.  [Feature #12553]
+
+* Kernel
+
+  * Kernel#clone now takes an optional keyword argument, freeze flag.
+    [Feature #12300]
+
+* MatchData
+
+  * MatchData#named_captures [Feature #11999]
+  * MatchData#values_at supports named captures [Feature #9179]
+
+* Module
+
+  * Module#refine accepts a module as the argument now. [Feature #12534]
+  * Module.used_modules [Feature #7418]
+
+* Numeric
+
+  * Numeric#finite?, Numeric#infinite? [Feature #12039]
+
+* Process
+
+  * Support CLOCK_MONOTONIC_RAW_APPROX, CLOCK_UPTIME_RAW, and
+    CLOCK_UPTIME_RAW_APPROX which are introduced by macOS 10.12.
+
+* Rational
+
+  * Rational#round now takes an optional keyword argument, half option, and
+    the default behavior is round-up now.  [Bug #12548] [Bug #12958]
+    half option can be one of :even, :up, and :down.  [Feature #12953]
+
+* Regexp
+
+  * meta character \X matches Unicode 9.0 characters with some workarounds
+    for UTR #51 Unicode Emoji, Version 4.0 emoji zwj sequences.
+
+  * Regexp#match? [Feature #8110]
+    This returns bool and doesn't save backref.
+
+  * Update to Onigmo 6.0.0.
+
+* Regexp/String: Update Unicode version from 8.0.0 to 9.0.0 [Feature #12513]
+
+* RubyVM::Env
+
+  * RubyVM::Env was removed.
+
+* String
+
+  * String#casecmp? [Feature #12786]
+
+  * String#concat, String#prepend [Feature #12333]
+    Now takes multiple arguments.
+
+  * String#each_line, String#lines now takes an optional keyword argument,
+    chomp flag.  [Feature #12553]
+
+  * String#match? [Feature #12898]
+
+  * String#unpack1 [Feature #12752]
+
+  * String#upcase, String#downcase, String#capitalize, String#swapcase and
+    their bang variants work for all of Unicode, and are no longer limited
+    to ASCII. Supported encodings are UTF-8, UTF-16BE/LE, UTF-32BE/LE, and
+    ISO-8859-1~16. Variations are available with options. See the documentation
+    of String#downcase for details. [Feature #10085]
+
+  * String.new(capacity: size) [Feature #12024]
+
+* StringIO
+
+  * StringIO#gets, StringIO#readline, StringIO#each_line, StringIO#readlines now takes
+    an optional keyword argument, chomp flag.  [Feature #12553]
+
+* Symbol
+
+  * Symbol#casecmp? [Feature #12786]
+
+  * Symbol#match now returns MatchData.  [Bug #11991]
+
+  * Symbol#match? [Feature #12898]
+
+  * Symbol#upcase, Symbol#downcase, Symbol#capitalize, and Symbol#swapcase now
+    work for all of Unicode. See the documentation of String#downcase
+    for details. [Feature #10085]
+
+* Thread
+
+  * Thread#report_on_exception and Thread.report_on_exception
+    [Feature #6647]
+
+* TracePoint
+
+  * TracePoint#callee_id [Feature #12747]
+
+* Warning
+
+  * New module named Warning is introduced.  By default it has only
+    one singleton method, named warn.  This makes it possible for
+    3rd-party libraries to control the way warnings are handled.
+    [Feature #12299]
+
+=== Stdlib updates (outstanding ones only)
+
+* CGI
+
+  * Don't allow , as a separator [Bug #12791]
+
+* CSV
+
+  * Add a liberal_parsing option. [Feature #11839]
+
+* IPAddr
+
+  * IPAddr#== and IPAddr#<=> no longer raise an exception if coercion fails.
+    [Bug #12799]
+
+* IRB
+
+  * Binding#irb: Start a REPL session like `binding.pry` at r56624.
+
+* Logger
+
+  * Allow specifying logger parameters in constructor such
+    as level, progname, datetime_format, formatter. [Feature #12224]
+  * Add shift_period_suffix option. [Feature #10772]
+
+* Net::HTTP
+
+  * New method: Net::HTTP.post [Feature #12375]
+
+* Net::FTP
+
+  * Support TLS (RFC 4217).
+  * Support hash style options for Net::FTP.new.
+  * Add a new optional argument pathname to Net::FTP#status.
+    Contributed by soleboxy. [GH-1478] [Feature #12965]
+
+* OpenSSL
+
+  * Includes Ruby/OpenSSL 2.0. OpenSSL has been extracted as a Gem and is
+    maintained at a separate repository now: https://github.com/ruby/openssl.
+    It still remains as a 'default gem'.  [Feature #9612]
+    Refer to ext/openssl/History.md for the full release note.
+
+* optparse
+
+  * Add an into option. [Feature #11191]
+
+* pathname
+
+  * New method: Pathname#empty? [Feature #12596]
+
+* Readline
+
+  * Readline.quoting_detection_proc and Readline.quoting_detection_proc=
+    [Feature #12659]
+
+* REXML
+
+  * REXML::Element#[]: If String or Symbol is specified, attribute
+    value is returned. Otherwise, Nth child is returned. This is
+    backward compatible change.
+
+* set
+
+  * New methods: Set#compare_by_identity and Set#compare_by_identity?.
+    [Feature #12210]
+
+* WEBrick
+
+  * Don't allow , as a separator [Bug #12791]
+
+=== Compatibility issues (excluding feature bug fixes)
+
+* Array#sum and Enumerable#sum are implemented.  [Feature #12217]
+  Ruby itself has no compatibility problem because Ruby didn't have sum method
+  for arrays before Ruby 2.4.
+  However many third party gems, activesupport, facets, simple_stats, etc,
+  defines sum method.  These implementations are mostly compatible but
+  there are subtle differences.
+  Ruby's sum method should be mostly compatible but it is impossible to
+  be perfectly compatible with all of them.
+
+* Fixnum and Bignum are unified into Integer  [Feature #12005]
+  Fixnum class and Bignum class is removed.
+  Integer class is changed from abstract class to concrete class.
+  For example, 0 is an instance of Integer: 0.class returns Integer.
+  The constants Fixnum and Bignum is bound to Integer.
+  So obj.kind_of?(Fixnum) works as obj.kind_of?(Integer).
+  At C-level, Fixnum object and Bignum object should be distinguished by
+  FIXNUM_P(obj) and RB_TYPE_P(obj, T_BIGNUM).
+  RUBY_INTEGER_UNIFICATION can be used to detect this feature at C-level.
+  0.class == Integer can be used to detect this feature at Ruby-level.
+  The C-level constants, rb_cFixnum and rb_cBignum, are removed.
+  They can cause compilation failure.
+
+* String/Symbol#upcase/downcase/swapcase/capitalize(!) now work for all of
+  Unicode, not only for ASCII. [Feature #10085]
+  No change is needed if the data is in ASCII anyway or if the limitation
+  to ASCII was only tolerated while waiting for a more extensive implementation.
+  A change (using the :ascii option) is needed in cases where Unicode data
+  is processed, but the operation has to be limited to ASCII only.
+  A good example of this are internationalized domain names.
+
+* TRUE / FALSE / NIL
+  These constants are now obsoleted. [Feature #12574]
+  Use true / false / nil resp. instead.
+
+=== Stdlib compatibility issues (excluding feature bug fixes)
+
+* DateTime
+
+  * DateTime#to_time now preserves timezone.  [Bug #12189]
+
+* Psych
+
+  * Update to Psych 2.2.2
+
+* RDoc
+
+  * Update to RDoc 5.0.0
+
+* RubyGems
+
+  * Update to RubyGems 2.6.8
+
+* shellwords
+
+  * Shellwords.shellwords (shellsplit) treats the backslash as escape
+    character only when followed by one of the following characters:
+    $ ` " \ <newline>
+    [Bug #10055]
+
+* Time
+
+  * Time#to_time now preserves timezone.  [Bug #12271]
+
+* thread
+
+  * the extension library is removed.  Till 2.0 it was a pure ruby script
+    "thread.rb", which has precedence over "thread.so", and has been provided
+    in $LOADED_FEATURES since 2.1.
+
+* Tk
+
+  * Tk is removed from stdlib.  [Feature #8539]
+    https://github.com/ruby/tk is the new upstream.
+
+* XMLRPC
+
+  * XMLRPC is removed from stdlib, and bundled as gem. [Feature #12160][ruby-core:74239]
+    https://github.com/ruby/xmlrpc is the new upstream.
+
+* Zlib
+
+  * Zlib.gzip and Zlib.gunzip [Feature #13020]
+
+=== C API updates
+
+* ruby_show_version() will no longer exits the process, if
+  RUBY_SHOW_COPYRIGHT_TO_DIE is set to 0.  This will be the default in
+  the future.
+
+* rb_gc_adjust_memory_usage() [Feature #12690]
+
+=== Supported platform changes
+
+* FreeBSD < 4 is no longer supported
+
+=== Implementation improvements
+
+* In some condition, `[x, y].max` and `[x, y].min` are optimized
+  so that a temporal array is not created.  The concrete condition is
+  an implementation detail: currently, the array literal must have no
+  splat, must have at least one expression but literal, the length must
+  be <= 0x100, and Array#max and min must not be redefined.  It will work
+  in most casual and real-life use case where it is written with intent
+  to `Math.max(x, y)`.
+
+* Thread deadlock detection now shows their backtrace and dependency. [Feature #8214]
+
+* st_table (st.c) internal data structure is improved. [Feature #12142]
+
+* Rational is extensively optimized. [Feature #12484]
+
+=== Miscellaneous changes
+
+* ChangeLog is removed from the repository.
+  It is generated from commit messages in Subversion by `make dist`.
+  Also note that now people should follow Git style commit message.
+  The template is written at
+  [Short (50 chars or less) summary of changes](https://git-scm.com/book/ch5-2.html).
+  [Feature #12283]
diff --git a/doc/NEWS-2.5.0 b/doc/NEWS-2.5.0
new file mode 100644
index 0000000000..e7613d5caa
--- /dev/null
+++ b/doc/NEWS-2.5.0
@@ -0,0 +1,567 @@
+# -*- rdoc -*-
+
+= NEWS for Ruby 2.5.0
+
+This document is a list of user visible feature changes made between
+releases except for bug fixes.
+
+Note that each entry is kept so brief that no reason behind or
+reference information is supplied with.  For a full list of changes
+with all sufficient information, see the ChangeLog file or Redmine
+(e.g. <tt>https://bugs.ruby-lang.org/issues/$FEATURE_OR_BUG_NUMBER</tt>)
+
+== Changes since the 2.4.0 release
+
+=== Language changes
+
+* Top-level constant look-up is removed.  [Feature #11547]
+
+* rescue/else/ensure are allowed inside do/end blocks.  [Feature #12906]
+
+* refinements take place in string interpolations.  [Feature #13812]
+
+=== Core classes updates (outstanding ones only)
+
+* Array
+
+  * New methods:
+
+    * Array#append  [Feature #12746]
+    * Array#prepend  [Feature #12746]
+
+* Data
+
+  * Is deprecated. It was a base class for C extensions, and it's not
+    necessary to expose in Ruby level. [Feature #3072]
+
+* Exception
+
+  * New methods:
+
+    * Exception#full_message to retrieve a String expression of an exception,
+      formatted in the same way in which Ruby prints out an uncaught
+      exception.  [Feature #14141] [experimental]
+
+* Dir
+
+  * Dir.glob provides new optional keyword argument, +:base+ .  [Feature #13056]
+  * Dir.chdir (without block arg), Dir.open, Dir.new, Dir.mkdir, Dir.rmdir,
+    Dir.empty? releases GVL
+
+  * New methods:
+
+    * Dir.children  [Feature #11302]
+    * Dir.each_child  [Feature #11302]
+
+* Enumerable
+
+  * Enumerable#any?, Enumerable#all?, Enumerable#none? and Enumerable#one?
+    accept a pattern argument. [Feature #11286]
+
+* File
+
+  * File.open accepts +:newline+ option to imply text mode. [Bug #13350]
+  * File#path raises an IOError for files opened with
+    File::Constants::TMPFILE option. [Feature #13568]
+  * File.stat, File.exist? and other <code>rb_stat()</code>-using methods
+    release GVL. [Bug #13941]
+  * File.rename releases GVL. [Feature #13951]
+  * File::Stat#atime, File::Stat#mtime and File::Stat#ctime support fractional
+    second timestamps on Windows 8 and later.  [Feature #13726]
+  * File::Stat#ino and File.indentical? support ReFS 128bit ino on Windows 8.1
+    and later.  [Feature #13731]
+  * File.readable?, File.readable_real?, File.writable?, File.writable_real?,
+    File.executable?, File.executable_real?, File.mkfifo, File.readlink,
+    File.truncate, File#truncate, File.chmod, File.lchmod, File.chown,
+    File.lchown, File.unlink, File.utime, File.lstat release GVL
+
+  * New method:
+
+    * File.lutime  [Feature #4052]
+
+* Hash
+
+  * New methods:
+
+    * Hash#transform_keys  [Feature #13583]
+    * Hash#transform_keys!  [Feature #13583]
+    * Hash#slice  [Feature #8499]
+
+* IO
+
+  * IO.copy_stream tries copy offload with copy_file_range(2) [Feature #13867]
+
+  * New methods:
+
+    * IO#pread  [Feature #4532]
+    * IO#pwrite  [Feature #4532]
+    * IO#write accepts multiple arguments  [Feature #9323]
+
+* IOError
+
+  * IO#close might raise an error with message "stream closed",
+    but it is refined to "stream closed in another thread". The new message
+    is more clear for user. [Bug #13405]
+
+* Integer
+
+  * Integer#round, Integer#floor, Integer#ceil and Integer#truncate always
+    return an Integer. [Bug #13420]
+  * Integer#pow accepts modulo argument for calculating modular
+    exponentiation.  [Feature #12508] [Feature #11003]
+
+  * New methods:
+
+    * Integer#allbits?, Integer#anybits?, Integer#nobits? [Feature #12753]
+    * Integer.sqrt  [Feature #13219]
+
+* Kernel
+
+  * Kernel#yield_self   [Feature #6721]
+  * Kernel#pp  [Feature #14123]
+  * Kernel#warn(..., uplevel:n)  [Feature #12882]
+
+* Method
+
+  * New methods:
+
+    * Method#=== that invokes Method#call, as same as Proc#=== [Feature #14142]
+
+* Module
+
+  * Module#attr, Module#attr_accessor, Module#attr_reader and Module#attr_writer
+    become public. [Feature #14132]
+  * Module#define_method, Module#alias_method, Module#undef_method and
+    Module#remove_method become public. [Feature #14133]
+
+* Numeric
+
+  * Numeric#step no longer hides errors from coerce method when
+    given a step value which cannot be compared with #> to 0. [Feature #7688]
+  * Numerical comparison operators (<,<=,>=,>) no longer hide exceptions
+    from #coerce method internally. Return nil in #coerce if the coercion is
+    impossible.  [Feature #7688]
+
+* Process
+
+  * Precision of Process.times is improved if getrusage(2) exists. [Feature #11952]
+
+  * New method:
+
+    * Process.last_status as an alias of $? [Feature #14043]
+
+* Range
+  * Range#initialize no longer hides exceptions when comparing begin and
+    end with #<=> and raise a "bad value for range" ArgumentError
+    but instead lets the exception from the #<=> call go through. [Feature #7688]
+
+* Regexp
+
+  * Update to Onigmo 6.1.3-669ac9997619954c298da971fcfacccf36909d05.
+
+    * Support absence operator https://github.com/k-takata/Onigmo/issues/82
+
+  * Support new 5 emoji-related Unicode character properties
+
+* RubyVM::InstructionSequence
+
+  * New method:
+
+    * RubyVM::InstructionSequence#each_child
+    * RubyVM::InstructionSequence#trace_points
+
+* String
+
+  * <code>String#-@</code> deduplicates unfrozen strings.  Already-frozen
+    strings remain unchanged for compatibility.  [Feature #13077]
+  * <code>-"literal"</code> (<code>String#-@</code>) optimized to return the same object
+    (same as <code>"literal".freeze</code> in Ruby 2.1+) [Feature #13295]
+  * String#casecmp and String#casecmp? return nil for non-string arguments
+    instead of raising a TypeError. [Bug #13312]
+  * String#start_with? accepts a regexp [Feature #13712]
+
+  * New methods:
+
+    * String#delete_prefix, String#delete_prefix! [Feature #12694]
+    * String#delete_suffix, String#delete_suffix! [Feature #13665]
+    * String#each_grapheme_cluster and String#grapheme_clusters to
+      enumerate grapheme clusters [Feature #13780]
+    * String#undump to unescape String#dump'ed string [Feature #12275]
+
+* Struct
+
+  * Struct.new takes `keyword_init: true` option to initialize members
+    with keyword arguments. [Feature #11925]
+
+* Regexp/String: Update Unicode version from 9.0.0 to 10.0.0 [Feature #13685]
+
+* Thread
+
+  * Description set by Thread#name= is now visible on Windows 10.
+
+  * New method:
+    * Thread#fetch  [Feature #13009]
+
+  * The default of Thread.report_on_exception is now true,
+    showing unhandled exceptions terminating threads on $stderr. [Feature #14143]
+
+* Time
+
+  * Time.at receives 3rd argument which specifies the unit of 2nd argument. [Feature #13919]
+
+* KeyError
+
+  * New methods:
+
+    * KeyError#receiver [Feature #12063]
+    * KeyError#key      [Feature #12063]
+
+* FrozenError
+
+  * New exception class. [Feature #13224]
+
+=== Stdlib updates (outstanding ones only)
+
+* BigDecimal
+
+  * Update to BigDecimal 1.3.4
+
+  * The following features are added:
+
+    * BigDecimal::VERSION
+
+  * The following features have been deprecated,
+    and are planned to be removed in the version 1.4.0:
+
+    * BigDecimal.new
+
+    * BigDecimal.ver
+
+  * BigDecimal#clone and #dup now do not make a new instance,
+    but returns the receiver itself.
+
+* Coverage
+
+  * Support branch coverage and method coverage measurement. [Feature #13901]
+    Branch coverage tells you which branches are executed, and which not.
+    Method coverage tells you which methods are invoked, and which not.
+    By running a test suite with this new feature, you can know which branches
+    and methods are executed by a test, and evaluate total coverage of a test
+    suite more strictly.
+
+    You can specify the measuring target by an option to `Coverage.start`:
+
+        Coverage.start(lines: true, branches: true, methods: true)
+
+    After some Ruby files are loaded, you can use `Coverage.result` to get
+    the coverage result:
+
+        Coverage.result
+        #=> { "/path/to/file.rb"=>
+        #     { :lines => [1, 2, 0, nil, ...],
+        #       :branches =>
+        #         { [:if, 0, 2, 1, 6, 4] =>
+        #             { [:then, 1, 3, 2, 3, 8] => 0,
+        #               [:else, 2, 5, 2, 5, 8] => 2
+        #             }
+        #         },
+        #       :methods => {
+        #          [Object, :foo, 1, 0, 7, 3] => 2
+        #       }
+        #     }
+        #   }
+
+    The result type of line coverage is not changed; it is just an array that
+    contains numbers, which means the count that each line was executed,
+    or `nil`s, which means that the line is not relevant.
+
+    The result type of branch coverage is:
+
+        { (jump base) => { (jump target) => (counter) } }
+
+    where jump base and targets have the format
+
+        [type, unique-id, start lineno, start column, end lineno, end column]
+
+    For example, `[:if, 0, 2, 1, 6, 4]` reads an `if` statement that ranges from
+    line 2 and column 1, to line 6 and column 4.  `[:then, 1, 3, 2, 3, 8]` reads
+    a `then` clause that ranges from line 3 and column 2, to line 3 and column 8.
+    Note that lineno starts from 1, and that columnno starts from 0.  So, the
+    above example shows a branch from the `if` to the `then` was never executed,
+    and a branch from the `if` to the `else` was executed twice.
+
+    The result type of method coverage is:
+
+        { (method key) => (counter) }
+
+    where method key has the format
+
+        [class, method-name, start lineno, start column, end lineno, end column]
+
+    For example, `[Object, :foo, 1, 0, 7, 3]` reads `Object#foo` that ranges from
+    line 1 and column 0, to line 7 and column 3.  The above example shows this
+    `Object#foo` was invoked twice.
+
+    Note: To keep compatibility, passing no option to `Coverage.start` will measure
+    only line coverage, and `Coverage.result` will return the old format:
+
+        Coverage.result
+        #=> { "/path/to/file.rb"=> [1, 2, 0, nil, ...] }
+
+* DRb
+
+  * ACL::ACLEntry.new no longer suppresses IPAddr::InvalidPrefixError.
+
+* ERB
+
+  * Add ERB#result_with_hash to render a template with local variables passed
+    with a Hash object. [Feature #8631]
+
+  * Default template file encoding is changed from ASCII-8BIT to UTF-8 in erb
+    command. [Bug #14095]
+
+  * Carriage returns are changed to be trimmed properly if trim_mode is specified
+    and used. Duplicated newlines will be removed on Windows. [Bug #5339] [Bug #11464]
+
+* IPAddr
+
+  * IPAddr no longer accepts invalid address mask. [Bug #13399]
+  * IPAddr#ipv4_compat and IPAddr#ipv4_compat? are marked for deprecation. [Bug #13769]
+
+  * New methods:
+
+    * IPAddr#prefix
+    * IPAddr#loopback?
+    * IPAddr#private? [Feature #11666]
+    * IPAddr#link_local? [Feature #10912]
+
+* IRB
+
+  * Print backtrace and error message in reverse order [Feature #8661] [experimental]
+  * `binding.irb` automatically requires irb and runs [Bug #13099] [experimental]
+  * `binding.irb` on its start shows source around the line where it was called [Feature #14124]
+
+* Matrix
+
+  * New methods:
+
+    * Matrix.combine and Matrix#combine [Feature #10903]
+    * Matrix#hadamard_product and Matrix#entrywise_product
+
+* Net::HTTP
+
+  * Net::HTTP.new supports no_proxy parameter [Feature #11195]
+  * Net::HTTP#min_version and Net::HTTP#max_version [Feature #9450]
+  * Add more HTTP status classes
+  * Net::HTTP::STATUS_CODES is added as HTTP Status Code Repository [Misc #12935]
+  * Net::HTTP#proxy_user and Net::HTTP#proxy_pass reflect http_proxy environment
+    variable if the system's environment variable is multiuser safe. [Bug #12921]
+
+* open-uri
+  * URI.open method defined as an alias to open-uri's Kernel.open.
+    open-uri's Kernel.open will be deprecated in future.
+
+* OpenSSL
+
+  * Updated Ruby/OpenSSL from version 2.0 to 2.1. Changes are noted in
+    "Version 2.1.0" section in ext/openssl/History.md.
+
+* Pathname
+
+  * New method:
+
+    * Pathname#glob [Feature #7360]
+
+* Psych
+
+  * Update to Psych 3.0.2.
+
+    * Convert fallback option to a keyword argument
+      https://github.com/ruby/psych/pull/342
+    * Add :symbolize_names option to Psych.load, Psych.safe_load like JSON.parse
+      https://github.com/ruby/psych/pull/333, https://github.com/ruby/psych/pull/337
+    * Add Psych::Handler#event_location
+      https://github.com/ruby/psych/pull/326
+    * Make frozen string literal = true
+      https://github.com/ruby/psych/pull/320
+    * Preserve time zone offset when deserializing times
+      https://github.com/ruby/psych/pull/316
+    * Remove deprecated method aliases for syck gem
+      https://github.com/ruby/psych/pull/312
+
+* RbConfig
+
+  * RbConfig::LIMITS is added to provide the limits of C types.
+    This is available when rbconfig/sizeof is loaded.
+
+* Ripper
+
+  * Ripper::EXPR_BEG and so on for Ripper#state.
+
+  * New method:
+
+    * Ripper#state to tell the state of scanner. [Feature #13686]
+
+* RDoc
+
+  * Update to RDoc 6.0.1.
+
+    * Replace IRB based lexer with Ripper.
+      * https://github.com/ruby/rdoc/pull/512
+      * This much improves the speed of generating documents.
+      * It also facilitates supporting new syntax in the future.
+    * Support many new syntaxes of Ruby from the past few years.
+    * Use "frozen_string_literal: true".
+      Performance survey: https://gist.github.com/aycabta/abdfaa75ea8a6877eeb734e942e73800
+    * Support did_you_mean.
+
+* Rubygems
+
+  * Update to Rubygems 2.7.3.
+    * http://blog.rubygems.org/2017/11/28/2.7.3-released.html
+    * http://blog.rubygems.org/2017/11/08/2.7.2-released.html
+    * http://blog.rubygems.org/2017/11/03/2.7.1-released.html
+    * http://blog.rubygems.org/2017/11/01/2.7.0-released.html
+    * http://blog.rubygems.org/2017/10/09/2.6.14-released.html
+    * http://blog.rubygems.org/2017/08/27/2.6.13-released.html
+
+* SecureRandom
+
+  * New method:
+
+    * SecureRandom.alphanumeric
+
+* Set
+
+  * New methods:
+
+    * Set#to_s as alias to #inspect [Feature #13676]
+    * Set#=== as alias to #include? [Feature #13801]
+    * Set#reset [Feature #6589]
+
+* StringIO
+
+  * StringIO#write accepts multiple arguments
+
+* StringScanner
+
+  * New methods:
+
+    * StringScanner#size, StringScanner#captures, StringScanner#values_at  [Feature #836]
+
+* URI
+
+  * Relative path operations no longer collapse consecutive slashes to a single slash. [Bug #8352]
+
+* WEBrick
+
+  * Add Server Name Indication (SNI) support [Feature #13729]
+  * support Proc objects as body responses [Feature #855]
+  * released as a RubyGem [Feature #13173]
+  * avoid unintended behavior from Kernel#open [Misc #14216]
+
+* Zlib
+
+  * Zlib::GzipWriter#write accepts multiple arguments
+
+=== Compatibility issues (excluding feature bug fixes)
+
+* Socket
+
+  * BasicSocket#read_nonblock and BasicSocket#write_nonblock no
+    longer set the O_NONBLOCK file description flag as side effect
+    (on Linux only) [Feature #13362]
+
+* Random
+
+  * Random.raw_seed renamed to become Random.urandom.  It is now
+    applicable to non-seeding purposes due to [Bug #9569].
+
+* Socket
+
+  * Socket::Ifaddr#vhid is added [Feature #13803]
+
+* ConditionVariable, Queue and SizedQueue reimplemented for speed.
+  They no longer subclass Struct. [Feature #13552]
+
+=== Stdlib compatibility issues (excluding feature bug fixes)
+
+* Gemification
+
+  * Promote following standard libraries to default gems.
+    * cmath
+    * csv
+    * date
+    * dbm
+    * etc
+    * fcntl
+    * fiddle
+    * fileutils
+    * gdbm
+    * ipaddr
+    * scanf
+    * sdbm
+    * stringio
+    * strscan
+    * webrick
+    * zlib
+
+* Logger
+
+  * Logger.new("| command") had been working to open a command
+    unintentionally. It was prohibited, and now Logger#initialize
+    treats a String argument only as a filename, as its specification. [Bug #14212]
+
+* Net::HTTP
+
+  * Net::HTTP#start now passes :ENV to p_addr by default. [Bug #13351]
+    To avoid this, pass nil explicitly.
+
+* mathn.rb
+
+  * Removed from stdlib. [Feature #10169]
+
+* Rubygems
+
+  * Removed "ubygems.rb" file from stdlib. It's needless since Ruby 1.9.
+
+=== C API updates
+
+=== Supported platform changes
+
+* Drop support of NaCl platform
+
+  * https://bugs.chromium.org/p/chromium/issues/detail?id=239656#c160
+
+=== Implementation improvements
+
+* (This might not be a "user visible feature change" but) Hash class's
+  hash function is now SipHash13. [Feature #13017]
+
+* SecureRandom now prefers OS-provided sources than OpenSSL. [Bug #9569]
+
+* Mutex rewritten to be smaller and faster [Feature #13517]
+
+* Performance of block passing using block parameters is improved by
+  lazy Proc allocation [Feature #14045]
+
+* Dynamic instrumentation for TracePoint hooks instead of using "trace"
+  instruction to avoid overhead [Feature #14104]
+
+* ERB now generates code from a template twice as fast as Ruby 2.4
+
+=== Miscellaneous changes
+
+* Print backtrace and error message in reverse order if $stderr is unchanged
+  and a tty. [Feature #8661] [experimental]
+
+* Print error message in bold/underlined text if $stderr is unchanged and a
+  tty. [Feature #14140] [experimental]
+
+* configure option --with-ext now mandates its arguments.  So for
+  instance if you run ./configure --with-ext=openssl,+ then the
+  openssl library is guaranteed compiled, otherwise the build fails
+  abnormally.
+
+  Note however to always add the ",+" at the end of the argument.
+  Otherwise nothing but openssl are built.  [Feature #13302]
diff --git a/doc/contributing.rdoc b/doc/contributing.rdoc
new file mode 100644
index 0000000000..b92dc7427e
--- /dev/null
+++ b/doc/contributing.rdoc
@@ -0,0 +1,459 @@
+= Contributing to Ruby
+
+Ruby has a vast and friendly community with hundreds of people contributing to
+a thriving open-source ecosystem. This guide is designed to cover ways for
+participating in the development of CRuby.
+
+There are plenty of ways for you to help even if you're not ready to write
+code or documentation.  You can help by reporting issues, testing patches, and
+trying out beta releases with your applications.
+
+== How To Report
+
+If you've encountered a bug in Ruby please report it to the redmine issue
+tracker available at {bugs.ruby-lang.org}[https://bugs.ruby-lang.org/].  Do not
+report security vulnerabilities here, there is a {separate
+channel}[rdoc-label:label-Reporting+Security+Issues] for them.
+
+There are a few simple steps you should follow in order to receive feedback
+on your ticket.
+
+* If you haven't already,
+  {sign up for an account}[https://bugs.ruby-lang.org/account/register] on the
+  bug tracker.
+* Try the latest version.
+
+  If you aren't already using the latest version, try installing a newer
+  stable release. See
+  {Downloading Ruby}[https://www.ruby-lang.org/en/downloads/].
+* Look to see if anyone already reported your issue, try
+  {searching on redmine}[https://bugs.ruby-lang.org/projects/ruby-trunk/issues]
+  for your problem.
+* If you can't find a ticket addressing your issue,
+  {create a new one}[https://bugs.ruby-lang.org/projects/ruby-trunk/issues/new].
+* Choose the target version, usually current. Bugs will be first fixed in the
+  current release and then {backported}[rdoc-label:label-Backport+Requests].
+* Fill in the Ruby version you're using when experiencing this issue
+  (<code>ruby -v</code>).
+* Attach any logs or reproducible programs to provide additional information.
+  Reproducible scripts should be as small as possible.
+* Briefly describe your problem.  A 2-3 sentence description will help give a
+  quick response.
+* Pick a category, such as core for common problems, or lib for a standard
+  library.
+* Check the {Maintainers
+  list}[https://bugs.ruby-lang.org/projects/ruby/wiki/Maintainers] and assign
+  the ticket if there is an active maintainer for the library or feature.
+* If the ticket doesn't have any replies after 10 days, you can send a
+  reminder.
+* Please reply to feedback requests. If a bug report doesn't get any feedback,
+  it'll eventually get rejected.
+
+=== Reporting to downstream distributions
+
+You can report downstream issues for the following distributions via their bug tracker:
+
+* {debian}[https://bugs.debian.org/cgi-bin/pkgreport.cgi?src=ruby-defaults]
+* {freebsd}[http://www.freebsd.org/cgi/query-pr-summary.cgi?text=ruby]
+* {redhat}[https://bugzilla.redhat.com/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&bug_status=MODIFIED]
+* {macports}[https://trac.macports.org/query?status=assigned&status=new&status=reopened&port=~ruby]
+* etc (add your distribution bug tracker here)
+
+=== Platform Maintainers
+
+For platform specific bugs in Ruby, you can assign your ticket to the current
+maintainer for a specific platform.
+
+The current active platform maintainers are as follows:
+
+[mswin64 (Microsoft Windows)]
+  NAKAMURA Usaku (usa)
+[mingw32 (Minimalist GNU for Windows)]
+  Nobuyoshi Nakada (nobu)
+[AIX]
+  Yutaka Kanemoto (kanemoto)
+[FreeBSD]
+  Akinori MUSHA (knu)
+[Solaris]
+  Naohisa Goto (ngoto)
+[RHEL, CentOS]
+  KOSAKI Motohiro (kosaki)
+[macOS]
+  Kenta Murata (mrkn)
+[cygwin, bcc32, djgpp, wince, ...]
+  none. (Maintainer WANTED)
+
+== Reporting Security Issues
+
+Security vulnerabilities receive special treatment since they may negatively
+affect many users. There is a private mailing list that all security issues
+should be reported to and will be handled discretely. Email the
+mailto:security@ruby-lang.org list and the problem will be published after
+fixes have been released. You can also encrypt the issue using {the PGP public
+key}[https://www.ruby-lang.org/security.asc] for the list.
+
+== Reporting Other Issues
+
+If you're having an issue with the website, or maybe the mailing list, you can
+contact the webmaster to help resolve the problem.
+
+The current webmaster is:
+
+* Hiroshi SHIBATA (hsbt)
+
+You can also report issues with the ruby-lang.org website on the issue tracker:
+
+* {issue tracker}[https://github.com/ruby/www.ruby-lang.org/issues]
+
+== Resolve Existing Issues
+
+As a next step beyond reporting issues you can help the core team resolve
+existing issues. If you check the Everyone's Issues list in GitHub Issues,
+you will find a lot of issues already requiring attention. What can you do for
+these? Quite a bit, actually:
+
+When a bug report goes for a while without any feedback, it goes to the bug
+graveyard which is unfortunate. If you check the {issues
+list}[https://bugs.ruby-lang.org/projects/ruby-trunk/issues] you will find lots
+of delinquent bugs that require attention.
+
+You can help by verifying the existing tickets, try to reproduce the reported
+issue on your own and comment if you still experience the bug. Some issues
+lack attention because of too much ambiguity, to help you can narrow down the
+problem and provide more specific details or instructions to reproduce the
+bug. You might also try contributing a failing test in the form of a patch,
+which we will cover later in this guide.
+
+It may also help to try out patches other contributors have submitted to
+redmine, if gone without notice. In this case the +patch+ command is your
+friend, see <code>man patch</code> for more information. Basically this would
+go something like this:
+
+  cd path/to/ruby/trunk
+  patch -p0 < path/to/patch
+
+You will then be prompted to apply the patch with the associated files. After
+building ruby again, you should try to run the tests and verify if the change
+actually worked or fixed the bug. It's important to provide valuable feedback
+on the patch that can help reach the overall goal, try to answer some of these
+questions:
+
+* What do you like about this change?
+* What would you do differently?
+* Are there any other edge cases not tested?
+* Is there any documentation that would be affected by this change?
+
+If you can answer some or all of these questions, you're on the right track.
+If your comment simply says "+1", then odds are that other reviewers aren't
+going to take it too seriously. Show that you took the time to review the
+patch.
+
+== How To Request Features
+
+If there's a new feature that you want to see added to Ruby, you will need to
+write a convincing proposal and patch to implement the feature.
+
+For new features in CRuby, use the {'Feature'
+tracker}[https://bugs.ruby-lang.org/projects/ruby-trunk/issues?set_filter=1&tracker_id=2]
+on ruby-trunk. For non-CRuby dependent features, features that would apply to
+alternate Ruby implementations such as JRuby and Rubinius, use the {CommonRuby
+tracker}[https://bugs.ruby-lang.org/projects/common-ruby].
+
+When writing a proposal be sure to check for previous discussions on the
+topic and have a solid use case. You will need to be persuasive and convince
+Matz on your new feature. You should also consider the potential compatibility
+issues that this new feature might raise.
+
+Consider making your feature into a gem, and if there are enough people who
+benefit from your feature it could help persuade ruby-core. Although feature
+requests can seem like an alluring way to contribute to Ruby, often these
+discussions can lead nowhere and exhaust time and energy that could be better
+spent fixing bugs. Choose your battles.
+
+A good template for a feature proposal should look something like this:
+
+[Abstract]
+  Summary of your feature
+[Background]
+  Describe current behavior and why it is problem. Related work, such as
+  solutions in other language helps us to understand the problem.
+[Proposal]
+  Describe your proposal in details
+[Details]
+  If it has complicated feature, describe it
+[Usecase]
+  How would your feature be used? Who will benefit from it?
+[Discussion]
+  Discuss about this proposal. A list of pros and cons will help start
+  discussion.
+[Limitation]
+  Limitation of your proposal
+[Another alternative proposal]
+  If there are alternative proposals, show them.
+[See also]
+  Links to the other related resources
+
+=== Slideshow
+
+At the Ruby Developer Meeting in Japan, committers discuss Feature Proposals together in Tokyo. We will judge proposals and then accept, reject, or give feedback for them.
+If you have a stalled proposal, making a slide to submit is good way to get feedback.
+
+Slides should be:
+
+* One-page slide
+* Include a corresponding ticket number
+* MUST include a figure and/or short example code
+* SHOULD have less sentence in natural language (try to write less than 140 characters)
+* It is RECOMMENDED to itemize: motivation/use case, proposal, pros/cons, corner case
+* PDF or Image (Web browsers can show it)
+
+Please note:
+
+* Even if the proposal is generally acceptable, it won't be accepted without writing corner cases in the ticket
+* Slide's example: DevelopersMeeting20130727Japan
+
+== Backport Requests
+
+When a new version of Ruby is released, it starts at patch level 0 (p0), and
+bugs will be fixed first on the trunk branch. If it's determined that a bug
+exists in a previous version of Ruby that is still in the bug fix stage of
+maintenance, then a patch will be backported. After the maintenance stage of a
+particular Ruby version ends, it goes into "security fix only" mode which
+means only security related vulnerabilities will be backported. Versions in
+End-of-life (EOL) will not receive any updates and it is recommended you
+upgrade as soon as possible.
+
+If a major security issue is found or after a certain amount of time since the
+last patch level release, a new patch-level release will be made.
+
+When submitting a backport request please confirm the bug has been fixed in
+newer versions and exists in maintenance mode versions. There is a backport
+tracker for each major version still in maintenance where you can request a
+particular revision merged in the affected version of Ruby.
+
+Each major version of Ruby has a release manager that should be assigned to
+handle backport requests. You can find the list of release managers on the
+{wiki}[https://bugs.ruby-lang.org/projects/ruby/wiki/ReleaseEngineering].
+
+=== Branches
+
+Status and maintainers of branches are listed on the
+{wiki}[https://bugs.ruby-lang.org/projects/ruby/wiki/ReleaseEngineering].
+
+== Running tests
+
+In order to help resolve existing issues and contributing patches to Ruby you
+need to be able to run the test suite.
+
+CRuby uses subversion for source control, you can find installation
+instructions and lots of great info to learn subversion on the
+{svnbook.red-bean.com}[http://svnbook.red-bean.com/]. For other resources see
+the {ruby-core documentation on
+ruby-lang.org}[https://www.ruby-lang.org/en/community/ruby-core/].
+
+This guide will use git for contributing.  The {git
+homepage}[https://git-scm.com/] has installation instructions with links to
+documentation for learning more about git. There is a mirror of the subversion
+repository on {github}[https://github.com/ruby/ruby].
+
+Install the prerequisite dependencies for building the CRuby interpreter to
+run tests.
+
+* C compiler
+* autoconf
+* bison
+* gperf
+* ruby - Ruby itself is prerequisite in order to build Ruby from source. It
+  can be 1.8.
+
+You should also have access to development headers for the following
+libraries, but these are not required:
+
+* Tcl/Tk
+* NDBM/QDBM
+* GDBM
+* OpenSSL
+* readline/editline(libedit)
+* zlib
+* libffi
+* libyaml
+* libexecinfo (FreeBSD)
+
+Now let's build CRuby:
+
+* Checkout the CRuby source code:
+
+    git clone https://github.com/ruby/ruby.git ruby-trunk
+
+* Generate the configuration files and build:
+
+    cd ruby-trunk
+    autoconf
+    mkdir build && cd build # its good practice to build outside of source dir
+    mkdir ~/.rubies # we will install to .rubies/ruby-trunk in our home dir
+    ../configure --prefix="${HOME}/.rubies/ruby-trunk"
+    make up && make install
+
+After adding Ruby to your PATH, you should be ready to run the test suite:
+
+    make test
+
+You can also use +test-all+ to run all of the tests with the RUNRUBY
+interpreter just built. Use TESTS or RUNRUBYOPT to pass parameters, such as:
+
+    make test-all TESTS=-v
+
+This is also how you can run a specific test from our build dir:
+
+    make test-all TESTS=drb/test_drb.rb
+
+You can run +test+ and +test-all+ at once by +check+ .
+
+    make check
+
+For older versions of Ruby you will need to run the build setup again after
+checking out the associated branch in git, for example if you wanted to
+checkout 1.9.3:
+
+    git clone https://github.com/ruby/ruby.git --branch ruby_1_9_3
+
+Once you checked out the source code, you can update the local copy by:
+
+    make up
+
+Or, update, build, install and check, by just:
+
+    make love
+
+== Contributing Documentation
+
+If you're interested in contributing documentation directly to CRuby there is
+some information available at
+{Contributing}[https://github.com/ruby/ruby#contributing].
+
+There is also the {Ruby Reference
+Manual}[https://github.com/rurema/doctree/wiki] in Japanese.
+
+== Contributing A Patch
+
+=== Deciding what to patch
+
+Before you submit a patch, there are a few things you should know:
+
+* Pay attention to the maintenance policy for stable and maintained versions of Ruby.
+* Released versions in security mode will not merge feature changes.
+* Search for previous discussions on ruby-core to verify the maintenance policy
+* Patches must be distributed under Ruby's license.
+* This license may change in the future, you must join the discussion if you don't agree to the change
+
+To improve the chance your patch will be accepted please follow these simple rules:
+
+* Bug fixes should be committed on trunk first
+* Format of the patch file must be a unified diff (ie: diff -pu, svn diff, or git diff)
+* Don't introduce cosmetic changes
+* Follow the original coding style of the code
+* Don't mix different changes in one commit
+
+First thing you should do is check out the code if you haven't already:
+
+  git clone https://github.com/ruby/ruby.git ruby-trunk
+
+Now create a dedicated branch:
+
+  cd ruby-trunk
+  git checkout -b my_new_branch
+
+The name of your branch doesn't really matter because it will only exist on
+your local computer and won't be part of the official Ruby repository. It will
+be used to create patches based on the differences between your branch and
+trunk, or edge Ruby.
+
+=== Coding style
+
+Here are some general rules to follow when writing Ruby and C code for CRuby:
+
+* Indent 4 spaces for C without tabs (old codes might use tabs for eight-space indentation,
+  but newer codes recommend to use spaces only)
+* Indent 2 space tabs for Ruby
+* Do not use TABs in ruby codes
+* ANSI C style for 1.9+ for function declarations
+* Follow C90 (not C99) Standard
+* PascalStyle for class/module names.
+* UNDERSCORE_SEPARATED_UPPER_CASE for other constants.
+* Capitalize words.
+* ABBRs should be all upper case.
+* Do as others do
+
+=== ChangeLog
+
+Although not required, if you wish to add a ChangeLog entry for your change
+please note:
+
+You can use the following template for the ChangeLog entry on your commit:
+
+  Thu Jan  1 00:00:00 2004  Your Name  <yourmail@example.com>
+
+  	* filename (function): short description of this commit.
+  	  This should include your intention of this change.
+  	  [bug:#number] [mailinglist:number]
+
+  	* filename2 (function2): additional description for this file/function.
+
+This follows {GNU Coding Standards for Change
+Logs}[http://www.gnu.org/prep/standards/html_node/Change-Logs.html#Change-Logs],
+some other requirements and tips:
+
+* Timestamps must be in JST (+09:00) in the style as above.
+* Two spaces between the timestamp and your name. Two spaces between
+  your name and your mail address.
+* One blank line between the timestamp and the description.
+* Indent the description with TAB. 2nd line should begin with TAB+2SP.
+* Write a entry (*) for each change.
+* Refer to redmine issue or discussion on the mailing list.
+* For GitHub issues, use [GH-#] (such as [Fixes GH-234]
+* One blank line between entries.
+* Do as other committers do.
+
+You can generate the ChangeLog entry by running <code>make change</code>
+
+When you're ready to commit, copy your ChangeLog entry into the commit message,
+keeping the same formatting and select your files:
+
+    git commit ChangeLog path/to/files
+
+In the likely event that your branch becomes outdated, you will have to update
+your working branch:
+
+    git fetch origin
+    git rebase remotes/origin/master
+
+Now that you've got some code you want to contribute, let's get set up to
+generate a patch. Start by forking the github mirror, check the {github docs on
+forking}[https://help.github.com/articles/fork-a-repo] if you get stuck here.
+You will only need a github account if you intend to host your repository
+on github.
+
+Next copy the writable url for your fork and add it as a git remote, replace
+"my_username" with your github account name:
+
+    git remote add my_fork git@github.com:my_username/ruby.git
+    # Now we can push our branch to our fork
+    git push my_fork my_new_branch
+
+In order to generate a patch that you can upload to the bug tracker, we can use
+the github interface to review our changes just visit
+https://github.com/my_username/ruby/compare/trunk...my_new_branch
+
+Next, you can simply add '.patch' to the end of this URL and it will generate
+the patch for you, save the file to your computer and upload it to the bug
+tracker. Alternatively you can submit a pull request, but for the best chances
+to receive feedback add it is recommended you add it to redmine.
+
+Since git is a distributed system, you are welcome to host your git repository
+on any {publicly accessible hosting
+site}[https://git.wiki.kernel.org/index.php/GitHosting], including {hosting your
+own}[https://www.kernel.org/pub/software/scm/git/docs/user-manual.html#public-repositories]
+You may use the {'git format-patch'}[https://git-scm.com/docs/git-format-patch]
+command to generate patch files to upload to redmine.  You may also use
+the {'git request-pull'}[https://git-scm.com/docs/git-request-pull] command for
+formatting pull request messages to redmine.
diff --git a/doc/contributors.rdoc b/doc/contributors.rdoc
new file mode 100644
index 0000000000..7c3722032b
--- /dev/null
+++ b/doc/contributors.rdoc
@@ -0,0 +1,793 @@
+= Contributors to Ruby
+
+The following list might be incomplete. Feel free to add your name if your
+patch was accepted into Ruby.
+
+== A
+
+Ayumu AIZAWA (ayumin)
+* committer
+
+AKIYOSHI, Masamichi (akiyoshi)
+* committer
+* He had maintained the VMS support on 2003-2004.
+
+Muhammad Ali
+* wrote rdoc for Fiber
+
+Minero Aoki (aamine)
+* committer
+* He is the maintainer of:
+  * fileutils
+  * net/http, net/https
+  * net/pop
+  * net/smtp
+  * racc
+  * ripper
+  * strscan
+
+Wakou Aoyama (wakou)
+* committer
+* He was the maintainer of some standard libraries.
+
+Koji Arai
+* committer
+
+arton
+* He is the distributor of ActiveScriptRuby and experimental 1.9.0-x installers for win32.
+* Wrote patches for win32ole, gc.c, tmpdir.rb
+
+Sergey Avseyev
+* Added IO#pread and IO#pwrite.
+
+== B
+
+Daniel Berger
+* a patch for irb
+* documentation
+* He wrote forwardable.rb
+
+David Black (dblack)
+* committer
+* He is the maintainer of scanf
+
+Ken Bloom
+* a patch for REXML.
+
+Oliver M. Bolzer
+* a patch for soap
+
+Alexey Borzenkov
+* a patch for mkmf.rb
+
+Evan Brodie
+* a patch for documentation of Float#round
+
+Richard Brown
+* a patch for configure.in
+
+Dirkjan Bussink
+* a patch for date.rb
+
+Daniel Bovensiepen
+* documentation
+* a patch for irb
+
+== C
+
+Brian Candler
+* a patch for configure.in, net/telnet
+
+keith cascio
+* a patch for optparse.rb
+
+Frederick Cheung
+* a patch for test/ruby/test_symbol.rb
+
+Christoph
+* patches for set.rb
+
+Sean Chittenden
+* patches for net/http, cgi
+
+William D. Clinger
+* ruby_strtod is based on his paper.
+
+== D
+
+Ryan Davis (ryan)
+* committer
+* He wrote and is the maintainer of miniunit
+
+Guy Decoux (ts)
+* committer
+
+Zach Dennis
+
+Martin Duerst (duerst)
+* committer
+* M17N
+
+Paul Duncan
+* patches for rdoc
+
+Alexander Dymo
+* a patch for lib/benchmark.rb
+
+== E
+
+Yusuke Endoh (mame)
+* committer
+* He wrote and is the maintainer of base64 library (1.9)
+* did much upon YARV compiler.
+
+erlercw
+* wrote Integer::gcd2
+
+== F
+
+Frank S.Fejes
+* a patch for net/pop
+
+Fundakowski Feldman
+* a patch for process.c
+
+Mauricio Fernandez
+* patches for parse.y
+
+David Flanagan (davidflanagan)
+* committer
+* M17N
+
+Takeyuki Fujioka (xibbar)
+* committer
+* He is the maintainer of cgi/*
+
+FUKUMOTO, Atsushi
+* a patch for tracer.rb
+
+Shota Fukumori (sorah)
+* committer
+* #4415 parallel unit/test
+
+Tadayoshi Funaba (tadf)
+* committer
+* He wrote and is the maintainer of
+  * date
+  * parsedate (1.8)
+* He ported rational.rb and complex.rb, which 1.8 contains, into rational.c and complex.c of 1.9.
+
+== G
+
+David M. Gay
+* ruby_strtod
+
+Florian Gilcher
+* documentation
+
+GOTOU, Kentaro (gotoken)
+* committer
+* He wrote benchmark.rb
+* He is the maintainer of:
+  * benchmark.rb
+  * open3
+
+GOTOU, Yuuzou (gotoyuzo)
+* committer
+
+James Edward Gray II (jeg2)
+* committer
+* He wrote the faster implementation of CSV and is the maintainer of csv.
+* Wrote documentation for rdoc
+
+== H
+
+Phil Hagelberg
+* patch for ruby-mode.el's documentation.
+
+Kirk Haines (wyhaines)
+* committer
+* the maintainer of ruby_1_8_6 branch
+
+Shinichiro Hamaji
+* fixed memory leaks (marshal.c, string.c)
+
+Shin-ichiro HARA
+* the developer and the sysop of ruby-{dev,list,core,talk} archive.
+* a patch for numeric.c
+
+Chris Heath (traumdeutung)
+* a patch for proc.c
+
+HIROKAWA Hisashi
+* fixed socket/socket.c
+
+Daniel Hob
+* He wrote:
+  * SMTP-TLS support for net/smtp.
+  * POP3S support
+
+Eric Hodel (drbrain)
+* committer
+* He is the maintainer of:
+  * rdoc
+  * ri
+  * rubygems
+
+Erik Hollensbe
+* a patch for delegate.rb
+
+Johan Holmberg
+* a patch for dir.c
+* documentation
+
+Erik Huelsmann
+
+Dae San Hwang
+* built a continuous integration environment on OpenSolaris.
+
+== I
+
+Nobuhiro IMAI
+* a patch for logger.rb
+
+"incorporate"
+* a patch for sprintf.c
+
+Keiju Ishitsuka (keiju)
+* committer
+* He wrote and is the maintainer of:
+  * cmath.rb (1.9)
+  * complex.rb (1.8)
+  * e2mmap.rb
+  * forwardable.rb
+  * irb
+  * mathn
+  * matrix.rb
+  * mutex_m.rb
+  * rational.rb (1.8)
+  * sync.rb
+  * shell/*
+  * thwait.rb
+  * tracer.rb
+
+== J
+
+Curtis Jackson
+* missing/dup2.c
+
+Alan Johnson
+* a patch for net/ftp
+
+Lyle Johnson
+* patches for nkf, bigdecimal, numeric.c
+
+== K
+
+Yoshihiro Kambayashi
+* a patch for enc/trans/single_byte.trans.
+* He wrote supports for some encodings.
+
+Yutaka Kanemoto
+* patches for common.mk, AIX AF_INET6 support
+
+Motoyuki Kasahara
+* He wrote getoptlong.rb
+
+Masahiro Kawato
+* a patch for shellwords.rb
+
+Wataru Kimura
+* a patch for configure.in
+
+Michael Klishin
+* patch for make help.
+
+Noritada Kobayashi
+* a patch for optparse.rb
+
+Shigeo Kobayashi (shigek)
+* committer
+* He is the maintainer of bigdecimal
+
+KONISHI, Hiromasa (H_Konishi)
+* committer
+* He had maintained the bcc32 support in 2004.
+
+Kornelius "murphy" Kalnbach
+* documentation
+
+K.Kosako (kosako)
+* committer
+* He wrote Oniguruma.
+
+Takehiro Kubo
+* patches for dl 64bit support.
+
+== L
+
+Marc-Andre Lafortune (marcandre)
+* committer
+* patches for hash.c, array.c, thread.c, enumc, string.c, range.c and rdoc documentation.
+
+Hongli Lai
+* improved pstore.rb
+* patch for tool/file2lastrev.rb.
+
+raspberry lemon
+* a patch for webrick/httpproxy.rb.
+
+Christian Loew
+* a patch for fileutils.rb
+
+== M
+
+Shugo Maeda (shugo)
+* committer
+* A system administrator of ruby-lang.org servers.
+* He wrote and is the maintainer of:
+  * monitor.rb
+  * net/ftp
+  * net/imap
+
+Stephan Maka (mathew)
+* documentation
+
+Yukihiro Matsumoto (matz)
+* Matz -- the founder, language designer of Ruby.
+* committer
+* Ruby itself, most of Ruby.
+* He is the maintainer of:
+  * singleton
+  * timeout
+  * gdbm
+  * sdbm
+
+Konrad Meyer
+* documentation
+
+Mib Software
+* missing/vsnprintf.c
+
+Todd C. Miller
+* missing/strlcat.c
+* missing/strlcpy.c
+
+MIYASAKA, Masaru
+* a patch for cgi.rb
+
+Stefan Monnier
+* regex.c was fixed with based on his Emacs21 patch.
+
+Marcel Moolenaar
+* patches for eval.c and gc.c.
+
+moonwolf
+* a patch for REXML, xmlrpc
+
+Hiroshi Moriyama
+* a patch for yaml.
+
+Kyosuke Morohashi
+* a patch for gem_prelude.rb
+
+Kenta Murata
+* patches for json, bignum.c
+
+Akinori MUSHA (knu)
+* committer
+* He wrote and is the maintainer of:
+  * abbrev.rb
+  * generator (1.8)
+  * enumerator (1.8)
+  * set
+  * ipaddr.rb
+  * digest/*
+  * syslog
+* He is the branch maintainer of ruby_1_8, the release manager of 1.8 series.
+
+== N
+
+Hidetoshi NAGAI (nagai)
+* committer
+* He is the maintainer of tk/*
+
+Nobuyoshi Nakada (nobu)
+* committer
+* a.k.a. the "patch monster"
+* He wrote and is the maintainer of:
+  * optparse
+  * stringio
+  * io/wait
+  * iconv
+
+Satoshi Nakagawa
+* patches for util.c
+
+Narihiro Nakamura (nari)
+* committer
+* a.k.a. authorNari
+* working at GC
+
+NAKAMURA, Hiroshi (nahi)
+* committer
+* He is the maintainer of:
+  * csv.rb (1.8)
+  * logger.rb
+  * soap/* (1.8)
+  * wsdl/* (1.8)
+  * xsd/* (1.8)
+
+NAKAMURA, Usaku (usa)
+* committer
+* a.k.a. unak
+* He is the maintainer of mswin32 and mswin64 support.
+
+NARUSE, Yui (naruse)
+* committer
+* a.k.a. "nurse"
+* Did much upon m17n.
+* He is the maintainer of:
+  * json
+  * nkf
+
+Christian Neukirchen
+* a patch for webrick/httputils
+
+Michael Neumann (mneumann)
+* committer
+* He is the maintainer of
+  * xmlrpc (1.8)
+  * gserver (1.8)
+
+NISHIO Hirokazu
+* wrote a patch for CVE-2010-0541
+
+Kazuhiro NISHIYAMA (kazu)
+* committer
+* a.k.a. znz
+
+Go Noguchi
+
+Martin Nordholts
+* misc/rdebug.el
+
+nmu
+* a patch for socket
+
+== O
+
+okkez
+* He is a sysop of the Ruby Reference Manual Renewal Project.
+* fixed ipaddr.rb, ext/etc
+
+Haruhiko Okumura
+* some of missing/* is based on his book:
+  * missing/erf.c
+  * missing/lgamma_r.c
+  * missing/tgamma.c
+
+OMAE, jun
+* a patch for debug.rb
+
+Eugene Ossintsev
+* documentation
+
+== P
+
+Heesob Park
+* a patch for win32/win32.c.
+
+pegacorn
+* a patch for instruby.rb
+
+== Q
+
+== R
+
+Gaston Ramos
+* documentation
+
+The Regents of the University of California
+* missing/crypt.c
+* missing/vsnprintf.c
+
+Sam Roberts
+* patch for socket
+* documentation
+
+Michal Rokos (michal)
+* committer
+* He was the maintainer of DJGPP support.
+
+rubikitch
+* a patch for io.c
+
+Marcus Rueckert
+* a patch for mkconfig.rb.
+
+Run Paint Run Run
+* patch for enc/unicode.c
+* documentation
+
+Sean Russell (ser)
+* committer
+* He wrote and is the maintainer of REXML.
+
+== S
+
+Kazuo Saito (ksaito)
+* committer
+* M17N
+
+Tadashi Saito
+* patches for test/ruby/test_math.rb, thread_*.c, bignum.c
+* working upon BigDecimal.
+* did much upon documentation
+
+Masahiro Sakai
+* a patch for io.c
+
+Laurent Sansonetti
+* a patch for tool/ytab.sed
+
+Jeff Saracco
+* documentation
+
+Koichi Sasada (ko1)
+* committer
+* He wrote YARV.
+
+Hugh Sasse
+* a patch for net/http
+* documentation
+
+Charlie Savage
+* a patch for win32/Makefile.sub
+
+Michael Scholz
+* a patch for ruby-mode.el
+
+Arthur Schreiber
+* patch for net/http and rdoc.
+
+Masatoshi SEKI (seki)
+* committer
+* He wrote and is the maintainer of:
+  * drb/*
+  * erb
+  * rinda
+
+Roman Shterenzon
+* a patch for open-uri.
+
+Kent Sibilev
+
+Gavin Sinclair (gsinclair)
+* committer
+
+John W. Small
+* He wrote gserver.rb
+
+Yuki Sonoda (yugui)
+* committer
+* She is the maintainer of man/* manual pages and is the release manager of 1.9 series.
+* She wrote prime.rb.
+* A developer and a sysop of redmine.ruby-lang.org.
+
+SOUMA, Yutaka
+* a patch for pack.c.
+
+Tatsuki Sugiura
+* WebDAV support for net/http
+
+Masaki Suketa (suke)
+* committer
+* He is the maintainer of win32ole
+
+sheepman
+* patches for ruby.c, thread.c, stringio, enum.c, webrick, net/http
+
+Siena. (siena)
+* committer
+
+Kirill A. Shutemov
+* a patch for parse.y
+
+Darren Smith
+* a patch for golf_prelude.rb
+
+Richard M. Stallman
+* missing/alloca.c
+
+Robin Stocker
+* documentation
+
+Joshua Stowers
+* a patch for array.c
+
+Marcus Stollsteimer (stomar)
+* committer
+* a maintainer of www.ruby-lang.org
+* patches for cgi (HTML5 tag maker), numeric.c, bigdecimal, ostruct.rb, prime.rb, and others
+* documentation
+
+Adam Strzelecki
+* a patch for compile.c
+
+Masashi Sumi
+* improved net/pop.rb
+
+Eric Sunshine
+* NeXT OpenStep, Rhapsody support
+
+Kouhei Sutou (kou)
+* committer
+* He wrote and is the maintainer of rss/*
+
+David Symonds
+* documentation
+
+== T
+
+TAKANO Mitsuhiro (takano32)
+* committer
+* He is the maintainer of IA-64 support.
+* BigDecimal
+
+TAKAO, Kouji (kouji)
+* committer
+* He is the maintainer of readline.
+
+Nathaniel Talbott (ntalbott)
+* committer
+* He was the maintainer of test/unit, runit, rubyunit.
+
+TANAKA, Akira (akr)
+* committer
+* Did much upon m17n.
+* And he is the maintainer of:
+  * open-uri
+  * pathname
+  * pp
+  * resolv-replace
+  * resolv
+  * time
+  * tsort
+
+Takaaki Tateishi (ttate)
+* committer
+* He was the maintainer of dl
+
+Technorama Ltd. (technoroma)
+* committer
+* openssl
+
+Andrew Thompson
+* a patch for socket.c IRIX support.
+
+Dave Thomas (dave)
+* committer
+* a.k.a. the Pragmatic Programmer.
+* He wrote rdoc.
+
+Tietew
+* patches for win32 support
+
+Masahiro Tomita
+* a patch for cgi.rb
+
+Jakub Travnik
+* a patch for eval.c
+
+Tom Truscott
+* missing/crypt.c
+
+== U
+
+UEDA, Satoshi
+* a patch for uri
+
+Takaaki Uematsu (uema2)
+* committer
+* He was the maintainer of WinCE support.
+
+UENO, Katsuhiro (katsu)
+* committer
+* He is the maintainer of zlib
+
+Hajimu UMEMOTO
+* He wrote ipaddr.rb
+
+URABE, Shyouhei (shyouhei)
+* committer
+* a.k.a. mput.
+* He is the branch maintainer of ruby_1_8_6 and ruby_1_8_7
+* and is the release manager of 1.8.x-pXXX.
+
+== V
+
+Joel VanderWerf
+* a patch for numeric.c
+
+Peter Vanbroekhoven
+
+Corinna Vinschen
+
+== W
+
+wanabe (wanabe)
+* committer
+* fixed YARV and Oniguruma.
+
+Chun Wang
+* a patch for time.rb
+
+WATANABE, Hirofumi (eban)
+* committer
+* He is the maintainer of
+  * ftools (1.8)
+  * tmpdir
+  * un
+  * Win32API
+
+WATANABE, Tetsuya
+* a patch for ruby.c
+
+William Webber (wew)
+* committer
+
+Jim Weirich (jim)
+* committer
+* He wrote Rake.
+
+Nathan Weizenbaum
+* fixed misc/ruby-mode.el.
+
+why the lukky stiff (why)
+* committer
+* He is the maintainer of syck
+
+Caley Woods
+* documentation
+
+Gary Wright
+* documentation
+
+== X
+
+== Y
+
+Akira Yamada (akira)
+* committer
+* He is the maintainer of ruby related packages at Debian project.
+
+Keita Yamaguchi
+* patches for enum.c, parse.y
+* documentation
+
+Hirokazu Yamamoto (ocean)
+* committer
+
+Hirotaka Yoshioka
+* a patch for improving SEGV handling
+
+== Z
+
+Aristarkh A Zagorodnikov
+* a patch for io.c
+
+Alexander Zavorine
+* committer
+* He is the maintainer for Symbian OS.
+
+Chiyuan Zhang
+* a patch for misc/ruby-mode.el.
+
+Dee Zsombor (zunda)
+* a patch for thread_pthread.c
+
+Dan Zwell
+* a patch for net/pop
+
+
diff --git a/doc/dtrace_probes.rdoc b/doc/dtrace_probes.rdoc
new file mode 100644
index 0000000000..d2cdd56902
--- /dev/null
+++ b/doc/dtrace_probes.rdoc
@@ -0,0 +1,178 @@
+= DTrace Probes
+
+A list of DTrace probes and their functionality. "Module" and "Function" cannot
+be defined in user defined probes (known as USDT), so they will not be
+specified. Probe definitions are in the format of:
+
+  provider:module:function:name(arguments)
+
+Since module and function cannot be specified, they will be blank. An example
+probe definition for Ruby would then be:
+
+  ruby:::method-entry(class name, method name, file name, line number)
+
+Where "ruby" is the provider name, module and function names are blank, the
+probe name is "method-entry", and the probe takes four arguments:
+
+* class name
+* method name
+* file name
+* line number
+
+== Probes List
+
+=== Stability
+
+Before we list the specific probes, let's talk about stability. Probe stability
+is declared in the probes.d file at the bottom on the #pragma D attributes
+lines. Here is a description of each of the stability declarations.
+
+[Provider name stability]
+  The provider name of "ruby" has been declared as stable. It is unlikely that
+  we will change the provider name from "ruby" to something else.
+
+[Module and Function stability]
+  Since we are not allowed to provide values for the module and function name,
+  the values we have provided (no value) is declared as stable.
+
+[Probe name stability]
+  The probe names are likely to change in the future, so they are marked as
+  "Evolving". Consumers should not depend on these names to be stable.
+
+[Probe argument stability]
+  The parameters passed to the probes are likely to change in the future, so
+  they are marked as "Evolving". Consumers should not depend on these to be
+  stable.
+
+=== Declared probes
+
+Probes are defined in the probes.d file. Here are the declared probes along
+with when they are fired and the arguments they take:
+
+[ruby:::method-entry(classname, methodname, filename, lineno);]
+  This probe is fired just before a method is entered.
+
+    classname name of the class (a string)
+    methodname name of the method about to be executed (a string)
+    filename the file name where the method is _being called_ (a string)
+    lineno the line number where the method is _being called_ (an int)
+
+[ruby:::method-return(classname, methodname, filename, lineno);]
+  This probe is fired just after a method has returned. The arguments are the
+  same as "ruby:::method-entry".
+
+[ruby:::cmethod-entry(classname, methodname, filename, lineno);]
+  This probe is fired just before a C method is entered. The arguments are the
+  same as "ruby:::method-entry".
+
+[ruby:::cmethod-return(classname, methodname, filename, lineno);]
+  This probe is fired just before a C method returns. The arguments are the
+  same as "ruby:::method-entry".
+
+[ruby:::require-entry(requiredfile, filename, lineno);]
+  This probe is fired on calls to rb_require_safe (when a file is required).
+
+    requiredfile is the name of the file to be required (string).
+    filename is the file that called "require" (string).
+    lineno is the line number where the call to require was made (int).
+
+[ruby:::require-return(requiredfile, filename, lineno);]
+  This probe is fired just before rb_require_safe (when a file is required)
+  returns. The arguments are the same as "ruby:::require-entry". This probe
+  will not fire if there was an exception during file require.
+
+[ruby:::find-require-entry(requiredfile, filename, lineno);]
+  This probe is fired right before search_required is called. search_required
+  determines whether the file has already been required by searching loaded
+  features ($"), and if not, figures out which file must be loaded.
+
+    requiredfile is the file to be required (string).
+    filename is the file that called "require" (string).
+    lineno is the line number where the call to require was made (int).
+
+[ruby:::find-require-return(requiredfile, filename, lineno);]
+  This probe is fired right after search_required returns. See the
+  documentation for "ruby:::find-require-entry" for more details. Arguments for
+  this probe are the same as "ruby:::find-require-entry".
+
+[ruby:::load-entry(loadedfile, filename, lineno);]
+  This probe is fired when calls to "load" are made. The arguments are the same
+  as "ruby:::require-entry".
+
+[ruby:::load-return(loadedfile, filename, lineno);]
+  This probe is fired when "load" returns. The arguments are the same as
+  "ruby:::load-entry".
+
+[ruby:::raise(classname, filename, lineno);]
+  This probe is fired when an exception is raised.
+
+    classname is the class name of the raised exception (string)
+    filename the name of the file where the exception was raised (string)
+    lineno the line number in the file where the exception was raised (int)
+
+[ruby:::object-create(classname, filename, lineno);]
+  This probe is fired when an object is about to be allocated.
+
+    classname the class of the allocated object (string)
+    filename the name of the file where the object is allocated (string)
+    lineno the line number in the file where the object is allocated (int)
+
+[ruby:::array-create(length, filename, lineno);]
+  This probe is fired when an Array is about to be allocated.
+
+    length the size of the array (long)
+    filename the name of the file where the array is allocated (string)
+    lineno the line number in the file where the array is allocated (int)
+
+[ruby:::hash-create(length, filename, lineno);]
+  This probe is fired when a Hash is about to be allocated.
+
+    length the size of the hash (long)
+    filename the name of the file where the hash is allocated (string)
+    lineno the line number in the file where the hash is allocated (int)
+
+[ruby:::string-create(length, filename, lineno);]
+  This probe is fired when a String is about to be allocated.
+
+    length the size of the string (long)
+    filename the name of the file where the string is allocated (string)
+    lineno the line number in the file where the string is allocated (int)
+
+[ruby:::symbol-create(str, filename, lineno);]
+  This probe is fired when a Symbol is about to be allocated.
+
+    str the contents of the symbol (string)
+    filename the name of the file where the string is allocated (string)
+    lineno the line number in the file where the string is allocated (int)
+
+[ruby:::parse-begin(sourcefile, lineno);]
+  Fired just before parsing and compiling a source file.
+
+    sourcefile the file being parsed (string)
+    lineno the line number where the source starts (int)
+
+[ruby:::parse-end(sourcefile, lineno);]
+  Fired just after parsing and compiling a source file.
+
+    sourcefile the file being parsed (string)
+    lineno the line number where the source ended (int)
+
+[ruby:::gc-mark-begin();]
+  Fired at the beginning of a mark phase.
+
+[ruby:::gc-mark-end();]
+  Fired at the end of a mark phase.
+
+[ruby:::gc-sweep-begin();]
+  Fired at the beginning of a sweep phase.
+
+[ruby:::gc-sweep-end();]
+  Fired at the end of a sweep phase.
+
+[ruby:::method-cache-clear(class, sourcefile, lineno);]
+  Fired when the method cache is cleared.
+
+    class is the classname being cleared, or "global" (string)
+    sourcefile the file being parsed (string)
+    lineno the line number where the source ended (int)
+
diff --git a/doc/etc.rd.ja b/doc/etc.rd.ja
new file mode 100644
index 0000000000..b4ff6ed04e
--- /dev/null
+++ b/doc/etc.rd.ja
@@ -0,0 +1,75 @@
+# etc.rd.ja -  -*- mode: rd; coding: utf-8; -*- created at: Fri Jul 14 00:47:15 JST 1995
+=begin
+
+= Etc(モジュール)
+
+実行しているOSからの情報を得るためのモジュール．クラスにインクルード
+して使うこともできる．
+
+== Module Function
+
+--- getlogin
+
+	自分のlogin名を返す．これが失敗した場合はgetpwuid()を用いると
+	良い．
+
+--- getpwnam(name)
+
+	/etc/passwdファイル(あるいはDBMファイルやNISデータベース)を検
+	索し，nameの名前を持つpasswdエントリを返す．戻り値はpasswd構造
+	体で以下のメンバを持つ．
+
+	  struct passwd
+	    name 	# ユーザ名(文字列)
+	    passwd	# パスワード(文字列)
+	    uid		# ユーザID(整数)
+	    gid		# グループID(整数)
+	    gecos	# gecosフィールド(文字列)
+	    dir		# ホームディレクトリ(文字列)
+	    shell	# ログインシェル(文字列)
+	    # 以降のメンバはシステムによっては提供されない．
+	    change	# パスワード変更時間(整数)
+	    quota	# クォータ(整数)
+	    age		# エージ(整数)
+	    class	# ユーザアクセスクラス(文字列)
+	    comment	# コメント(文字列)
+	    expire	# アカウント有効期限(整数)
+	  end
+
+	詳細はgetpwnam(3)を参照のこと．
+
+--- getpwuid([uid])
+
+	uidをユーザIDとするpasswdエントリを返す．戻り値はgetpwnam()と
+	同様である．引数を省略した場合にはgetuid()の値を用いる．詳細は
+	getpwuid(3)を参照のこと．
+
+--- getgrgid(gid)
+
+	/etc/groupファイル(あるいは…getpwnam参照)を検索し，gidをグルー
+	プIDとするグループエントリを返す．戻り値はgroup構造体で以下の
+	メンバを持つ．
+
+	  struct group
+	    name 	# グループ名(文字列)
+	    passwd	# グループのパスワード(文字列)
+	    gid		# グループID(整数)
+	    mem		# グループメンバ名の配列
+	  end
+
+	詳細はgetgrgid(3)を参照のこと．
+
+--- getgrnam(name)
+
+	nameという名前のグループエントリを返す．戻り値はgetgrgid()と同
+	様である．詳細はgetgrnam(3)を参照．
+
+--- group
+
+	全てのグループエントリを順にアクセスするためのイテレータ．
+
+--- passwd
+
+	全てのpasswdエントリを順にアクセスするためのイテレータ．
+
+=end
diff --git a/doc/extension.ja.rdoc b/doc/extension.ja.rdoc
new file mode 100644
index 0000000000..88f300a562
--- /dev/null
+++ b/doc/extension.ja.rdoc
@@ -0,0 +1,1833 @@
+# extension.ja.rdoc -  -*- RDoc -*- created at: Mon Aug  7 16:45:54 JST 1995
+
+= Rubyの拡張ライブラリの作り方
+
+Rubyの拡張ライブラリの作り方を説明します．
+
+== 基礎知識
+
+Cの変数には型があり，データには型がありません．ですから，た
+とえばポインタをintの変数に代入すると，その値は整数として取
+り扱われます．逆にRubyの変数には型がなく，データに型がありま
+す．この違いのため，CとRubyは相互に変換しなければ，お互いの
+データをアクセスできません．
+
+RubyのデータはVALUEというCの型で表現されます．VALUE型のデー
+タはそのデータタイプを自分で知っています．このデータタイプと
+いうのはデータ(オブジェクト)の実際の構造を意味していて，Ruby
+のクラスとはまた違ったものです．
+
+VALUEからCにとって意味のあるデータを取り出すためには
+
+1. VALUEのデータタイプを知る
+2. VALUEをCのデータに変換する
+
+の両方が必要です．(1)を忘れると間違ったデータの変換が行われ
+て，最悪プログラムがcore dumpします．
+
+=== データタイプ
+
+Rubyにはユーザが使う可能性のある以下のタイプがあります．
+
+T_NIL      :: nil
+T_OBJECT   :: 通常のオブジェクト
+T_CLASS    :: クラス
+T_MODULE   :: モジュール
+T_FLOAT    :: 浮動小数点数
+T_STRING   :: 文字列
+T_REGEXP   :: 正規表現
+T_ARRAY    :: 配列
+T_HASH     :: 連想配列
+T_STRUCT   :: (Rubyの)構造体
+T_BIGNUM   :: 多倍長整数
+T_FIXNUM   :: Fixnum(31bitまたは63bit長整数)
+T_COMPLEX  :: 複素数
+T_RATIONAL :: 有理数
+T_FILE     :: 入出力
+T_TRUE     :: 真
+T_FALSE    :: 偽
+T_DATA     :: データ
+T_SYMBOL   :: シンボル
+
+その他に内部で利用されている以下のタイプがあります．
+
+  T_ICLASS
+  T_MATCH
+  T_UNDEF
+  T_NODE
+  T_ZOMBIE
+
+ほとんどのタイプはCの構造体で実装されています．
+
+=== VALUEのデータタイプをチェックする
+
+ruby.hではTYPE()というマクロが定義されていて，VALUEのデータ
+タイプを知ることが出来ます．TYPE()マクロは上で紹介したT_XXXX
+の形式の定数を返します．VALUEのデータタイプに応じて処理する
+場合には，TYPE()の値で分岐することになります．
+
+  switch (TYPE(obj)) {
+    case T_FIXNUM:
+      /* FIXNUMの処理 */
+      break;
+    case T_STRING:
+      /* 文字列の処理 */
+      break;
+    case T_ARRAY:
+      /* 配列の処理 */
+      break;
+    default:
+      /* 例外を発生させる */
+      rb_raise(rb_eTypeError, "not valid value");
+      break;
+  }
+
+それとデータタイプをチェックして，正しくなければ例外を発生す
+る関数が用意されています．
+
+  void Check_Type(VALUE value, int type)
+
+この関数はvalueがtypeで無ければ，例外を発生させます．引数と
+して与えられたVALUEのデータタイプが正しいかどうかチェックす
+るためには，この関数を使います．
+
+FIXNUMとNILに関してはより高速な判別マクロが用意されています．
+
+  FIXNUM_P(obj)
+  NIL_P(obj)
+
+=== VALUEをCのデータに変換する
+
+データタイプがT_NIL，T_FALSE，T_TRUEである時，データはそれぞ
+れnil，false，trueです．このデータタイプのオブジェクトはひと
+つずつしか存在しません．
+
+データタイプがT_FIXNUMの時，これは31bitまたは63bitのサイズを
+持つ整数です．longのサイズが32bitのプラットフォームであれば
+31bitに，longのサイズが64bitのプラットフォームであれば63bit
+になります. FIXNUM を C の整数に変換するためにはマクロ
+「FIX2INT()」または「FIX2LONG()」を使います．これらのマクロ
+を使用する際には事前にデータタイプがFIXNUMであることを確認す
+る必要がありますが，比較的高速に変換を行うことができます．ま
+た，「FIX2LONG()」は例外を発生しませんが，「FIX2INT()」は変
+換結果がintのサイズに収まらない場合には例外を発生します．
+それから，FIXNUMに限らずRubyのデータを整数に変換する
+「NUM2INT()」および「NUM2LONG()」というマクロがあります．こ
+れらのマクロはデータタイプのチェック無しで使えます
+(整数に変換できない場合には例外が発生する)．同様にチェック無し
+で使える変換マクロはdoubleを取り出す「NUM2DBL()」があります．
+
+char* を取り出す場合， StringValue() と StringValuePtr()
+を使います．
+StringValue(var) は var が String
+であれば何もせず，そうでなければ var を var.to_str() の結果
+に置き換えるマクロ，StringValuePtr(var) は同様に var を
+String に置き換えてから var のバイト列表現に対する char* を
+返すマクロです．var の内容を直接置き換える処理が入るので，
+var は lvalue である必要があります．
+また，StringValuePtr() に類似した StringValueCStr() というマ
+クロもあります．StringValueCStr(var) は var を String に置き
+換えてから var の文字列表現に対する char* を返します．返され
+る文字列の末尾には NUL 文字が付加されます．なお，途中に NUL
+文字が含まれる場合は ArgumentError が発生します．
+一方，StringValuePtr() では，末尾に NUL 文字がある保証はなく，
+途中に NUL 文字が含まれている可能性もあります．
+
+それ以外のデータタイプは対応するCの構造体があります．対応す
+る構造体のあるVALUEはそのままキャスト(型変換)すれば構造体の
+ポインタに変換できます．
+
+構造体は「struct RXxxxx」という名前でruby.hで定義されていま
+す．例えば文字列は「struct RString」です．実際に使う可能性が
+あるのは文字列と配列くらいだと思います．
+
+ruby.hでは構造体へキャストするマクロも「RXXXXX()」(全部大文
+字にしたもの)という名前で提供されています(例: RSTRING())．た
+だし、構造体への直接のアクセスはできるだけ避け，対応する
+rb_xxxx() といった関数を使うようにして下さい．例えば，配列の
+要素へアクセスする場合は，rb_ary_entry(ary, offset)，
+rb_ary_store(ary, offset, obj) を利用するようにして下さい．
+
+構造体からデータを取り出すマクロが提供されています．文字列
+strの長さを得るためには「RSTRING_LEN(str)」とし，文字列strを
+char*として得るためには「RSTRING_PTR(str)」とします．
+
+Rubyの構造体を直接アクセスする時に気をつけなければならないこ
+とは，配列や文字列の構造体の中身は参照するだけで，直接変更し
+ないことです．直接変更した場合，オブジェクトの内容の整合性が
+とれなくなって，思わぬバグの原因になります．
+
+=== CのデータをVALUEに変換する
+
+VALUEの実際の構造は
+
+FIXNUMの場合 ::
+
+  1bit左シフトして，LSBを立てる．
+
+その他のポインタの場合 ::
+
+  そのままVALUEにキャストする．
+
+となっています．よって，LSBをチェックすればVALUEがFIXNUMかど
+うかわかるわけです(ポインタのLSBが立っていないことを仮定して
+いる)．
+
+ですから，FIXNUM以外のRubyのオブジェクトの構造体は単にVALUE
+にキャストするだけでVALUEに変換出来ます．ただし，任意の構造
+体がVALUEにキャスト出来るわけではありません．キャストするの
+はRubyの知っている構造体(ruby.hで定義されているstruct RXxxx
+のもの)だけです．
+
+FIXNUMに関しては変換マクロを経由する必要があります．Cの整数
+からVALUEに変換するマクロは以下のものがあります．必要に応じ
+て使い分けてください．
+
+INT2FIX() :: もとの整数が31bitまたは63bit以内に収まる自信
+             がある時
+INT2NUM() :: 任意の整数からVALUEへ
+
+INT2NUM()は整数がFIXNUMの範囲に収まらない場合，Bignumに変換
+してくれます(が，少し遅い)．
+
+=== Rubyのデータを操作する
+
+先程も述べた通り，Rubyの構造体をアクセスする時に内容の更新を
+行うことは勧められません．で，Rubyのデータを操作する時には
+Rubyが用意している関数を用いてください．
+
+ここではもっとも使われるであろう文字列と配列の生成/操作を行
+う関数をあげます(全部ではないです)．
+
+==== 文字列に対する関数
+
+rb_str_new(const char *ptr, long len) ::
+
+  新しいRubyの文字列を生成する．
+
+rb_str_new2(const char *ptr) ::
+rb_str_new_cstr(const char *ptr) ::
+
+  Cの文字列からRubyの文字列を生成する．この関数の機能は
+  rb_str_new(ptr, strlen(ptr))と同等である．
+
+rb_str_new_literal(const char *ptr) ::
+
+  Cのリテラル文字列からRubyの文字列を生成する．
+
+rb_tainted_str_new(const char *ptr, long len) ::
+
+  汚染マークが付加された新しいRubyの文字列を生成する．外部
+  からのデータに基づく文字列には汚染マークが付加されるべき
+  である．
+
+rb_tainted_str_new2(const char *ptr) ::
+rb_tainted_str_new_cstr(const char *ptr) ::
+
+  Cの文字列から汚染マークが付加されたRubyの文字列を生成する．
+
+rb_str_append(VALUE str1, VALUE str2) ::
+
+  Rubyの文字列str1にRubyの文字列str2を追加する．
+
+rb_sprintf(const char *format, ...) ::
+rb_vsprintf(const char *format, va_list ap) ::
+
+  Cの文字列formatと続く引数をprintf(3)のフォーマットにしたがって
+  整形し，Rubyの文字列を生成する．
+
+  注意: "%"PRIsVALUEがObject#to_s('+'フラグが指定されている
+  ときはObject#inspect)を使ったVALUEの出力に利用できる．これ
+  は"%i"と衝突するため，整数には"%d"を使用すること．
+
+rb_str_cat(VALUE str, const char *ptr, long len) ::
+
+  Rubyの文字列strにlenバイトの文字列ptrを追加する．
+
+rb_str_cat2(VALUE str, const char* ptr) ::
+rb_str_cat_cstr(VALUE str, const char* ptr) ::
+
+  Rubyの文字列strにCの文字列ptrを追加する．この関数の機能は
+  rb_str_cat(str, ptr, strlen(ptr))と同等である．
+
+rb_str_catf(VALUE str, const char* format, ...) ::
+rb_str_vcatf(VALUE str, const char* format, va_list ap) ::
+
+  Cの文字列formatと続く引数をprintf(3)のフォーマットにしたがって
+  整形し，Rubyの文字列strに追加する．この関数の機能は，それぞれ
+  rb_str_append(str, rb_sprintf(format, ...)) や
+  rb_str_append(str, rb_vsprintf(format, ap)) と同等である．
+
+rb_enc_str_new(const char *ptr, long len, rb_encoding *enc) ::
+rb_enc_str_new_cstr(const char *ptr, rb_encoding *enc) ::
+
+  指定されたエンコーディングでRubyの文字列を生成する.
+
+rb_enc_str_new_literal(const char *ptr, rb_encoding *enc) ::
+
+  Cのリテラル文字列から指定されたエンコーディングでRubyの文字列を生成する．
+
+rb_usascii_str_new(const char *ptr, long len) ::
+rb_usascii_str_new_cstr(const char *ptr) ::
+
+  エンコーディングがUS-ASCIIのRubyの文字列を生成する.
+
+rb_usascii_str_new_literal(const char *ptr) ::
+
+  Cのリテラル文字列からエンコーディングがUS-ASCIIのRubyの文字列を生成する．
+
+rb_utf8_str_new(const char *ptr, long len) ::
+rb_utf8_str_new_cstr(const char *ptr) ::
+
+  エンコーディングがUTF-8のRubyの文字列を生成する.
+
+rb_utf8_str_new_literal(const char *ptr) ::
+
+  Cのリテラル文字列からエンコーディングがUTF-8のRubyの文字列を生成する．
+
+rb_str_resize(VALUE str, long len) ::
+
+  Rubyの文字列のサイズをlenバイトに変更する．strの長さは前
+  以てセットされていなければならない．lenが元の長さよりも短
+  い時は，lenバイトを越えた部分の内容は捨てられる．lenが元
+  の長さよりも長い時は，元の長さを越えた部分の内容は保存さ
+  れないでゴミになるだろう．この関数の呼び出しによって
+  RSTRING_PTR(str)が変更されるかもしれないことに注意．
+
+rb_str_set_len(VALUE str, long len) ::
+
+  Rubyの文字列のサイズをlenバイトにセットする．strが変更可
+  能でなければ例外が発生する．RSTRING_LEN(str)とは無関係に，
+  lenバイトまでの内容は保存される．lenはstrの容量を越えてい
+  てはならない．
+
+==== 配列に対する関数
+
+rb_ary_new() ::
+
+  要素が0の配列を生成する．
+
+rb_ary_new2(long len) ::
+rb_ary_new_capa(long len) ::
+
+  要素が0の配列を生成する．len要素分の領域をあらかじめ割り
+  当てておく．
+
+rb_ary_new3(long n, ...) ::
+rb_ary_new_from_args(long n, ...) ::
+
+  引数で指定したn要素を含む配列を生成する．
+
+rb_ary_new4(long n, VALUE *elts) ::
+rb_ary_new_from_values(long n, VALUE *elts) ::
+
+  配列で与えたn要素の配列を生成する．
+
+rb_ary_to_ary(VALUE obj) ::
+
+  オブジェクトを配列に変換する.
+  Object#to_aryと同等である.
+
+他にも配列を操作する関数が多数ある. これらは
+引数aryに配列を渡さなければならない. さもないと
+コアを吐く.
+
+rb_ary_aref(int argc, const VALUE *argv, VALUE ary) ::
+
+  Array#[]と同等.
+
+rb_ary_entry(VALUE ary, long offset) ::
+
+  \ary[offset]
+
+rb_ary_store(VALUE ary, long offset, VALUE obj) ::
+
+  \ary[offset] = obj
+
+rb_ary_subseq(VALUE ary, long beg, long len) ::
+
+  ary[beg, len]
+
+rb_ary_push(VALUE ary, VALUE val) ::
+rb_ary_pop(VALUE ary) ::
+rb_ary_shift(VALUE ary) ::
+rb_ary_unshift(VALUE ary, VALUE val) ::
+
+  ary.push, ary.pop, ary.shift, ary.unshift
+
+rb_ary_cat(VALUE ary, const VALUE *ptr, long len) ::
+
+  配列aryにptrからlen個のオブジェクトを追加する．
+
+== Rubyの機能を使う
+
+原理的にRubyで書けることはCでも書けます．RubyそのものがCで記
+述されているんですから，当然といえば当然なんですけど．ここで
+はRubyの拡張に使うことが多いだろうと予測される機能を中心に紹
+介します．
+
+=== Rubyに機能を追加する
+
+Rubyで提供されている関数を使えばRubyインタプリタに新しい機能
+を追加することができます．Rubyでは以下の機能を追加する関数が
+提供されています．
+
+- クラス，モジュール
+- メソッド，特異メソッドなど
+- 定数
+
+では順に紹介します．
+
+==== クラス/モジュール定義
+
+クラスやモジュールを定義するためには，以下の関数を使います．
+
+  VALUE rb_define_class(const char *name, VALUE super)
+  VALUE rb_define_module(const char *name)
+
+これらの関数は新しく定義されたクラスやモジュールを返します．
+メソッドや定数の定義にこれらの値が必要なので，ほとんどの場合
+は戻り値を変数に格納しておく必要があるでしょう．
+
+クラスやモジュールを他のクラスの内部にネストして定義する時に
+は以下の関数を使います．
+
+  VALUE rb_define_class_under(VALUE outer, const char *name, VALUE super)
+  VALUE rb_define_module_under(VALUE outer, const char *name)
+
+==== メソッド/特異メソッド定義
+
+メソッドや特異メソッドを定義するには以下の関数を使います．
+
+  void rb_define_method(VALUE klass, const char *name,
+                        VALUE (*func)(ANYARGS), int argc)
+
+  void rb_define_singleton_method(VALUE object, const char *name,
+                                  VALUE (*func)(ANYARGS), int argc)
+
+念のため説明すると「特異メソッド」とは，その特定のオブジェク
+トに対してだけ有効なメソッドです．RubyではよくSmalltalkにお
+けるクラスメソッドとして，クラスに対する特異メソッドが使われ
+ます．
+
+これらの関数の argcという引数はCの関数へ渡される引数の数(と
+形式)を決めます．argcが0以上の時は関数に引き渡す引数の数を意
+味します．16個以上の引数は使えません(が，要りませんよね，そ
+んなに)．実際の関数には先頭の引数としてselfが与えられますの
+で，指定した数より1多い引数を持つことになります．
+
+argcが負の時は引数の数ではなく，形式を指定したことになります．
+argcが-1の時は引数を配列に入れて渡されます．argcが-2の時は引
+数はRubyの配列として渡されます．
+
+メソッドを定義する関数はまだいくつかあります. ひとつはメソッド
+名としてIDを取ります. IDについては2.2.2を参照.
+
+  void rb_define_method_id(VALUE klass, ID name,
+                           VALUE (*func)(ANYARGS), int argc)
+
+private/protectedなメソッドを定義するふたつの関数があります.
+
+  void rb_define_private_method(VALUE klass, const char *name,
+                                VALUE (*func)(ANYARGS), int argc)
+  void rb_define_protected_method(VALUE klass, const char *name,
+                                  VALUE (*func)(ANYARGS), int argc)
+
+privateメソッドとは関数形式でしか呼び出すことの出来ないメソッ
+ドです．
+
+最後に， rb_define_module関数はモジュール関数を定義します．
+モジュール関数とはモジュールの特異メソッドであり，同時に
+privateメソッドでもあるものです．例をあげるとMathモジュール
+のsqrt()などがあげられます．このメソッドは
+
+  Math.sqrt(4)
+
+という形式でも
+
+  include Math
+  sqrt(4)
+
+という形式でも使えます．モジュール関数を定義する関数は以下の
+通りです．
+
+  void rb_define_module_function(VALUE module, const char *name,
+                                 VALUE (*func)(ANYARGS), int argc)
+
+関数的メソッド(Kernelモジュールのprivate method)を定義するた
+めの関数は以下の通りです．
+
+  void rb_define_global_function(const char *name, VALUE (*func)(ANYARGS), int argc)
+
+メソッドの別名を定義するための関数は以下の通りです．
+
+  void rb_define_alias(VALUE module, const char* new, const char* old);
+
+属性の取得・設定メソッドを定義するには
+
+  void rb_define_attr(VALUE klass, const char *name, int read, int write)
+
+クラスメソッドallocateを定義したり削除したりするための関数は
+以下の通りです．
+
+  void rb_define_alloc_func(VALUE klass, VALUE (*func)(VALUE klass));
+  void rb_undef_alloc_func(VALUE klass);
+
+funcはクラスを引数として受け取って，新しく割り当てられたイン
+スタンスを返さなくてはなりません．このインスタンスは，外部リ
+ソースなどを含まない，できるだけ「空」のままにしておいたほう
+がよいでしょう．
+
+継承したクラスにある既存のメソッドをオーバーライドしているな
+ら，オーバーライドされたメソッドを呼び出すには以下の関数を使
+います．
+
+  VALUE rb_call_super(int argc, const VALUE *argv)
+
+現在のスコープのレシーバは(他に方法がなければ)，以下の関数で
+得ることができます．
+
+  VALUE rb_current_receiver(void)
+
+==== 定数定義
+
+拡張ライブラリが必要な定数はあらかじめ定義しておいた方が良い
+でしょう．定数を定義する関数は二つあります．
+
+  void rb_define_const(VALUE klass, const char *name, VALUE val)
+  void rb_define_global_const(const char *name, VALUE val)
+
+前者は特定のクラス/モジュールに属する定数を定義するもの，後
+者はグローバルな定数を定義するものです．
+
+=== Rubyの機能をCから呼び出す
+
+既に『1.5 Rubyのデータを操作する』で一部紹介したような関数を
+使えば，Rubyの機能を実現している関数を直接呼び出すことが出来
+ます．
+
+# このような関数の一覧表はいまのところありません．ソースを見
+# るしかないですね．
+
+それ以外にもRubyの機能を呼び出す方法はいくつかあります．
+
+==== Rubyのプログラムをevalする
+
+CからRubyの機能を呼び出すもっとも簡単な方法として，文字列で
+与えられたRubyのプログラムを評価する以下の関数があります．
+
+  VALUE rb_eval_string(const char *str)
+
+この評価は現在の環境で行われます．つまり，現在のローカル変数
+などを受け継ぎます．
+
+評価は例外を発生するかもしれないことに注意しましょう. より安全
+な関数もあります.
+
+  VALUE rb_eval_string_protect(const char *str, int *state)
+
+この関数はエラーが発生するとnilを返します．そして，成功時には
+*stateはゼロに，さもなくば非ゼロになります．
+
+==== IDまたはシンボル
+
+Cから文字列を経由せずにRubyのメソッドを呼び出すこともできま
+す．その前に，Rubyインタプリタ内でメソッドや変数名を指定する
+時に使われているIDについて説明しておきましょう．
+
+IDとは変数名，メソッド名を表す整数です．Rubyの中では
+
+ :識別子
+
+または
+
+ :"任意の文字列"
+
+でアクセスできます．Cからこの整数を得るためには関数
+
+  rb_intern(const char *name)
+  rb_intern_str(VALUE name)
+
+を使います．Rubyから引数として与えられたシンボル(または文字
+列)をIDに変換するには以下の関数を使います．
+
+  rb_to_id(VALUE symbol)
+  rb_check_id(volatile VALUE *name)
+  rb_check_id_cstr(const char *name, long len, rb_encoding *enc)
+
+もし引数がシンボルでも文字列でもなければ，to_strメソッドで文
+字列に変換しようとします．第二の関数はその変換結果を*nameに保
+存し,その名前が既知のシンボルでない場合は0を返します．この関
+数が0以外を返した場合は*nameは常にシンボルか文字列であり，0を
+返した場合は常に文字列です．第三の関数はRubyの文字列ではなく
+NUL終端されたCの文字列を使います．
+
+Rubyから引数として与えられたシンボル(または文字列)をシンボル
+に変換するには以下の関数を使います．
+
+  rb_to_symbol(VALUE name)
+  rb_check_symbol(volatile VALUE *namep)
+  rb_check_symbol_cstr(const char *ptr, long len, rb_encoding *enc)
+
+これらの関数は，IDの代わりにシンボルを返すことを除けば上記の
+関数と同じです．
+
+==== CからRubyのメソッドを呼び出す
+
+Cから文字列を経由せずにRubyのメソッドを呼び出すためには以下
+の関数を使います．
+
+  VALUE rb_funcall(VALUE recv, ID mid, int argc, ...)
+
+この関数はオブジェクトrecvのmidで指定されるメソッドを呼び出
+します．その他に引数の指定の仕方が違う以下の関数もあります．
+
+  VALUE rb_funcall2(VALUE recv, ID mid, int argc, VALUE *argv)
+  VALUE rb_funcallv(VALUE recv, ID mid, int argc, VALUE *argv)
+  VALUE rb_apply(VALUE recv, ID mid, VALUE args)
+
+applyには引数としてRubyの配列を与えます．
+
+==== 変数/定数を参照/更新する
+
+Cから関数を使って参照・更新できるのは，定数，インスタンス変
+数です．大域変数は一部のものはCの大域変数としてアクセスでき
+ます．ローカル変数を参照する方法は公開していません．
+
+オブジェクトのインスタンス変数を参照・更新する関数は以下の通
+りです．
+
+  VALUE rb_ivar_get(VALUE obj, ID id)
+  VALUE rb_ivar_set(VALUE obj, ID id, VALUE val)
+
+idはrb_intern()で得られるものを使ってください．
+
+定数を参照するには以下の関数を使ってください．
+
+  VALUE rb_const_get(VALUE obj, ID id)
+
+定数を新しく定義するためには『2.1.3 定数定義』で紹介さ
+れている関数を使ってください．
+
+== RubyとCとの情報共有
+
+C言語とRubyの間で情報を共有する方法について解説します．
+
+=== Cから参照できるRubyの定数
+
+以下のRubyの定数はCのレベルから参照できます．
+
+Qtrue ::
+Qfalse ::
+
+  真偽値．QfalseはC言語でも偽とみなされます(つまり0)．
+
+Qnil ::
+
+  C言語から見た「nil」．
+
+=== CとRubyで共有される大域変数
+
+CとRubyで大域変数を使って情報を共有できます．共有できる大域
+変数にはいくつかの種類があります．そのなかでもっとも良く使わ
+れると思われるのはrb_define_variable()です．
+
+  void rb_define_variable(const char *name, VALUE *var)
+
+この関数はRubyとCとで共有する大域変数を定義します．変数名が
+`$'で始まらない時には自動的に追加されます．この変数の値を変
+更すると自動的にRubyの対応する変数の値も変わります．
+
+またRuby側からは更新できない変数もあります．このread onlyの
+変数は以下の関数で定義します．
+
+  void rb_define_readonly_variable(const char *name, VALUE *var)
+
+これら変数の他にhookをつけた大域変数を定義できます．hook付き
+の大域変数は以下の関数を用いて定義します．hook付き大域変数の
+値の参照や設定はhookで行う必要があります．
+
+  void rb_define_hooked_variable(const char *name, VALUE *var,
+                                 VALUE (*getter)(), void (*setter)())
+
+この関数はCの関数によってhookのつけられた大域変数を定義しま
+す．変数が参照された時には関数getterが，変数に値がセットされ
+た時には関数setterが呼ばれる．hookを指定しない場合はgetterや
+setterに0を指定します．
+--
+getterもsetterも0ならばrb_define_variable()と同じになる．
+++
+
+getterとsetterの仕様は次の通りです．
+
+  VALUE (*getter)(ID id, VALUE *var);
+  void (*setter)(VALUE val, ID id, VALUE *var);
+
+それから，対応するCの変数を持たないRubyの大域変数を定義する
+こともできます. その変数の値はフック関数のみによって取得・設定
+されます.
+
+  void rb_define_virtual_variable(const char *name,
+                                  VALUE (*getter)(), void (*setter)())
+
+この関数によって定義されたRubyの大域変数が参照された時には
+getterが，変数に値がセットされた時にはsetterが呼ばれます．
+
+getterとsetterの仕様は以下の通りです．
+
+  (*getter)(ID id);
+  (*setter)(VALUE val, ID id);
+
+=== CのデータをRubyオブジェクトにする
+
+Cの世界で定義されたデータ(構造体)をRubyのオブジェクトとして
+取り扱いたい場合がありえます．このような場合はTypedData_XXX
+マクロ群を用いて構造体へのポインタとRubyのオブジェクトとを互
+いに変換できます．
+
+--
+古い(非Typedな)Data_XXXマクロ群は非推奨になりました．
+将来のバージョンのRubyでは古いマクロは動作しなくなる可能性
+があります．
+++
+
+==== 構造体からオブジェクトへ
+
+構造体へのポインタsvalをRubyオブジェクトに変換するには次のマ
+クロを使います。
+
+  TypedData_Wrap_Struct(klass, data_type, sval)
+
+このマクロの戻り値は生成されたオブジェクトを表すVALUE値です．
+
+klassはこのオブジェクトのクラスです．data_typeはこの構造体を
+Rubyが管理するための情報を記述したconst rb_data_type_t型への
+ポインタです．
+
+なお, klassは, Objectや他のクラスではなくData (rb_cData)とい
+う特別なクラスから派生することが推奨されます．
+Dataから派生しない場合には, 必ずrb_undef_alloc_func(klass)
+を呼び出してください．
+
+rb_data_type_tは次のように定義されています．
+
+  typedef struct rb_data_type_struct rb_data_type_t;
+
+  struct rb_data_type_struct {
+          const char *wrap_struct_name;
+          struct {
+                  void (*dmark)(void*);
+                  void (*dfree)(void*);
+                  size_t (*dsize)(const void *);
+                  void *reserved[2];
+          } function;
+          const rb_data_type_t *parent;
+          void *data;
+          VALUE flags;
+  };
+
+wrap_struct_nameはこの構造体を識別する名前です．主に統計情報
+の収集と出力に用いられます．プロセス内で一意であれば特にCや
+Rubyの識別子として有効である必要はありません．
+
+dmarkおよびdfree関数はGC実行中に呼び出されます.
+なお, GC実行中はRubyオブジェクトのアロケーションは禁止されま
+す. よって, dmarkおよびdfree関数でRubyオブジェクトのアロケー
+ションは行わないでください.
+
+dmarkはガーベージコレクタがオブジェクトへの参照をマークする
+ときに用いる関数です．この構造体がRubyのオブジェクトへの参照
+を保持するときには, dmarkではrb_gc_markなどを用いて構造体内
+のすべての参照をマークしなければなりません．
+そのような参照を含まない時には0を指定します．
+
+--
+そのような参照は勧められません．
+++
+
+dfreeはこの構造体がもう不要になった時に呼ばれる関数です．こ
+の関数がガーベージコレクタから呼ばれます．これが-1の場合は，
+単純に構造体が解放されます．
+
+dsizeは構造体が消費しているメモリのバイト数を返す関数です．
+引数として構造体へのポインタが渡されます．実装困難であれば0
+を渡しても差し支えありませんが, できるだけ指定するようにして
+ください．
+
+reservedとparentは0で埋めなければなりません．
+
+dataにはユーザー定義の任意の値を指定できます．Rubyはこの値に
+は関知しないので，好きに使ってください．
+
+flagsには次のフラグのうち当てはまるもののビット和を指定しま
+す．いずれもRubyのガーベージコレクタについての深い理解を必要
+としますので，良くわからない場合には0を指定すると良いでしょ
+う．
+
+RUBY_TYPED_FREE_IMMEDIATELY ::
+
+  このフラグを指定すると，ガーベージコレクタはこの構造体が不
+  要になった場合にはGC中に直ちにdfreeを呼び出します．
+  dfreeがRuby内部のロック(GVL)を解放する可能性がない場合はこ
+  のフラグを指定できます．
+
+  指定しない場合はdfree呼び出しは遅延され, ファイナライザと
+  同じタイミングで実行されます．
+
+RUBY_TYPED_WB_PROTECTED ::
+
+  オブジェクトの実装がライトバリアをサポートしていることを示
+  します．このフラグを指定するとRubyはそのオブジェクトに対し
+  てGCをより効率的に実行できます．
+  ただし，指定する場合はユーザーはそのオブジェクトのすべての
+  メソッドの実装に適切にライトバリアを挿入する責任があります．
+  さもなくばRubyは実行時にクラッシュする可能性があります．
+
+  ライトバリアについてはdoc/extension.ja.rdocのAppendix D
+  "世代別GC"も参照してください．
+
+
+Cの構造体の割当と対応するオブジェクトの生成を同時に行うマク
+ロとして以下のものが提供されています．
+
+  TypedData_Make_Struct(klass, type, data_type, sval)
+
+このマクロの戻り値は生成されたオブジェクトのVALUE値
+です．このマクロは以下の式のように働きます:
+
+  (sval = ZALLOC(type), TypedData_Wrap_Struct(klass, data_type, sval))
+
+klass, data_typeはData_Wrap_Structと同じ働きをします．type
+は割り当てるC構造体の型です．割り当てられた構造体は変数sval
+に代入されます．この変数の型は (type*) である必要があります．
+
+==== オブジェクトから構造体へ
+
+TypedData_Wrap_StructやTypedData_Make_Structで生成したオブジェ
+クトから構造体へのポインタを復元するには以下のマクロを用いま
+す．
+
+  TypedData_Get_Struct(obj, type, &data_type, sval)
+
+Cの構造体へのポインタは変数svalに代入されます．
+
+これらのマクロの使い方はちょっと分かりにくいので，後で説明す
+る例題を参照してください．
+
+== 例: dbmの拡張ライブラリの作成
+
+=== ディレクトリを作る
+
+  % mkdir ext/dbm
+
+Ruby 1.1からは任意のディレクトリでダイナミックライブラリを作
+ることができるようになりました．Rubyに静的にリンクする場合に
+はRubyを展開したディレクトリの下，extディレクトリの中に拡張
+ライブラリ用のディレクトリを作る必要があります．名前は適当に
+選んで構いません．
+
+=== 設計する
+
+まあ，当然なんですけど，どういう機能を実現するかどうかまず設
+計する必要があります．どんなクラスをつくるか，そのクラスには
+どんなメソッドがあるか，クラスが提供する定数などについて設計
+します．
+
+=== Cコードを書く
+
+拡張ライブラリ本体となるC言語のソースを書きます．C言語のソー
+スがひとつの時には「ライブラリ名.c」を選ぶと良いでしょう．C
+言語のソースが複数の場合には逆に「ライブラリ名.c」というファ
+イル名は避ける必要があります．オブジェクトファイルとモジュー
+ル生成時に中間的に生成される「ライブラリ名.o」というファイル
+とが衝突するからです．また，後述する mkmf ライブラリのいくつ
+かの関数がコンパイルを要するテストのために「conftest.c」とい
+うファイル名を使用することに注意してください．ソースファイル
+名として「conftest.c」を使用してはなりません．
+
+Rubyは拡張ライブラリをロードする時に「Init_ライブラリ名」と
+いう関数を自動的に実行します．dbmライブラリの場合「Init_dbm」
+です．この関数の中でクラス，モジュール，メソッド，定数などの
+定義を行います．dbm.cから一部引用します．
+
+  void
+  Init_dbm(void)
+  {
+      /* DBMクラスを定義する */
+      VALUE cDBM = rb_define_class("DBM", rb_cObject);
+      /* DBMはEnumerableモジュールをインクルードする */
+      rb_include_module(cDBM, rb_mEnumerable);
+
+      /* DBMクラスのクラスメソッドopen(): 引数はCの配列で受ける */
+      rb_define_singleton_method(cDBM, "open", fdbm_s_open, -1);
+
+      /* DBMクラスのメソッドclose(): 引数はなし */
+      rb_define_method(cDBM, "close", fdbm_close, 0);
+      /* DBMクラスのメソッド[]: 引数は1個 */
+      rb_define_method(cDBM, "[]", fdbm_fetch, 1);
+
+      /* ... */
+
+      /* DBMデータを格納するインスタンス変数名のためのID */
+      id_dbm = rb_intern("dbm");
+  }
+
+DBMライブラリはdbmのデータと対応するオブジェクトになるはずで
+すから，Cの世界のdbmをRubyの世界に取り込む必要があります．
+
+dbm.cではTypedData_Make_Structを以下のように使っています．
+
+  struct dbmdata {
+      int  di_size;
+      DBM *di_dbm;
+  };
+
+  static const rb_data_type_t dbm_type = {
+      "dbm",
+      {0, free_dbm, memsize_dbm,},
+      0, 0,
+      RUBY_TYPED_FREE_IMMEDIATELY,
+  };
+
+  obj = TypedData_Make_Struct(klass, struct dbmdata, &dbm_type, dbmp);
+
+ここではdbmdata構造体へのポインタをDataにカプセル化してい
+ます．DBM*を直接カプセル化しないのはclose()した時の処理を考
+えてのことです．
+
+Dataオブジェクトからdbmstruct構造体のポインタを取り出すため
+に以下のマクロを使っています．
+
+  #define GetDBM(obj, dbmp) do {\
+      TypedData_Get_Struct((obj), struct dbmdata, &dbm_type, (dbmp));\
+      if ((dbmp) == 0) closed_dbm();\
+      if ((dbmp)->di_dbm == 0) closed_dbm();\
+  } while (0)
+
+ちょっと複雑なマクロですが，要するにdbmdata構造体のポインタ
+の取り出しと，closeされているかどうかのチェックをまとめてい
+るだけです．
+
+DBMクラスにはたくさんメソッドがありますが，分類すると3種類の
+引数の受け方があります．ひとつは引数の数が固定のもので，例と
+してはdeleteメソッドがあります．deleteメソッドを実装している
+fdbm_delete()はこのようになっています．
+
+  static VALUE
+  fdbm_delete(VALUE obj, VALUE keystr)
+  {
+      /* ... */
+  }
+
+引数の数が固定のタイプは第1引数がself，第2引数以降がメソッド
+の引数となります．
+
+引数の数が不定のものはCの配列で受けるものとRubyの配列で受け
+るものとがあります．dbmライブラリの中で，Cの配列で受けるもの
+はDBMのクラスメソッドであるopen()です．これを実装している関
+数fdbm_s_open()はこうなっています．
+
+  static VALUE
+  fdbm_s_open(int argc, VALUE *argv, VALUE klass)
+  {
+      /* ... */
+
+      if (rb_scan_args(argc, argv, "11", &file, &vmode) == 1) {
+          mode = 0666;          /* default value */
+      }
+
+      /* ... */
+  }
+
+このタイプの関数は第1引数が与えられた引数の数，第2引数が与え
+られた引数の入っている配列になります．selfは第3引数として与
+えられます．
+
+この配列で与えられた引数を解析するための関数がopen()でも使わ
+れているrb_scan_args()です．第3引数に指定したフォーマットに従
+い，第4引数以降に指定したVALUEへの参照に値を代入してくれま
+す．
+
+引数の数をチェックするだけならrb_check_arity()が使えます．
+これは引数をリストとして扱いたいときに便利です．
+
+引数をRubyの配列として受け取るメソッドの例には
+Thread#initializeがあります．実装はこうです．
+
+  static VALUE
+  thread_initialize(VALUE thread, VALUE args)
+  {
+      /* ... */
+  }
+
+第1引数はself，第2引数はRubyの配列です．
+
+*注意事項*
+
+Rubyと共有はしないがRubyのオブジェクトを格納する可能性のある
+Cの大域変数は以下の関数を使ってRubyインタプリタに変数の存在
+を教えてあげてください．でないとGCでトラブルを起こします．
+
+  void rb_global_variable(VALUE *var)
+
+=== extconf.rbを用意する
+
+Makefileを作る場合の雛型になるextconf.rbというファイルを作り
+ます．extconf.rbはライブラリのコンパイルに必要な条件のチェッ
+クなどを行うことが目的です．まず，
+
+  require 'mkmf'
+
+をextconf.rbの先頭に置きます．extconf.rbの中では以下のRuby関
+数を使うことが出来ます．
+
+  have_library(lib, func): ライブラリの存在チェック
+  have_func(func, header): 関数の存在チェック
+  have_header(header): ヘッダファイルの存在チェック
+  create_makefile(target[, target_prefix]): Makefileの生成
+
+以下の変数を使うことができます．
+
+  $CFLAGS: コンパイル時に追加的に指定するフラグ(-Oなど)
+  $CPPFLAGS: プリプロセッサに追加的に指定するフラグ(-Iや-Dなど)
+  $LDFLAGS: リンク時に追加的に指定するフラグ(-Lなど)
+  $objs: リンクされるオブジェクトファイル名のリスト
+
+オブジェクトファイルのリストは，通常はソースファイルを検索し
+て自動的に生成されますが，makeの途中でソースを生成するような
+場合は明示的に指定する必要があります．
+
+ライブラリをコンパイルする条件が揃わず，そのライブラリをコン
+パイルしない時にはcreate_makefileを呼ばなければMakefileは生
+成されず，コンパイルも行われません．
+
+=== dependを用意する
+
+もし，ディレクトリにdependというファイルが存在すれば，
+Makefileが依存関係をチェックしてくれます．
+
+  % gcc -MM *.c > depend
+
+などで作ることが出来ます．あって損は無いでしょう．
+
+=== Makefileを生成する
+
+Makefileを実際に生成するためには
+
+  ruby extconf.rb
+
+とします．extconf.rbに require 'mkmf' の行がない場合にはエラー
+になりますので，引数を追加して
+
+  ruby -r mkmf extconf.rb
+
+としてください．
+
+site_ruby ディレクトリでなく，
+vendor_ruby ディレクトリにインストールする場合には
+以下のように --vendor オプションを加えてください．
+
+  ruby extconf.rb --vendor
+
+ディレクトリをext以下に用意した場合にはRuby全体のmakeの時に
+自動的にMakefileが生成されますので，このステップは不要です．
+
+=== makeする
+
+動的リンクライブラリを生成する場合にはその場でmakeしてくださ
+い．必要であれば make install でインストールされます．
+
+ext以下にディレクトリを用意した場合は，Rubyのディレクトリで
+makeを実行するとMakefileを生成からmake，必要によってはそのモ
+ジュールのRubyへのリンクまで自動的に実行してくれます．
+extconf.rbを書き換えるなどしてMakefileの再生成が必要な時はま
+たRubyディレクトリでmakeしてください．
+
+拡張ライブラリはmake installでRubyライブラリのディレクトリの
+下にコピーされます．もし拡張ライブラリと協調して使うRubyで記
+述されたプログラムがあり，Rubyライブラリに置きたい場合には，
+拡張ライブラリ用のディレクトリの下に lib というディレクトリ
+を作り，そこに 拡張子 .rb のファイルを置いておけば同時にイン
+ストールされます．
+
+=== デバッグ
+
+まあ，デバッグしないと動かないでしょうね．ext/Setupにディレ
+クトリ名を書くと静的にリンクするのでデバッガが使えるようにな
+ります．その分コンパイルが遅くなりますけど．
+
+=== できあがり
+
+後はこっそり使うなり，広く公開するなり，売るなり，ご自由にお
+使いください．Rubyの作者は拡張ライブラリに関して一切の権利を
+主張しません．
+
+== Appendix A. Rubyのソースコードの分類
+
+Rubyのソースはいくつかに分類することが出来ます．このうちクラ
+スライブラリの部分は基本的に拡張ライブラリと同じ作り方になっ
+ています．これらのソースは今までの説明でほとんど理解できると
+思います．
+
+=== Ruby言語のコア
+
+class.c    :: クラスとモジュール
+error.c    :: 例外クラスと例外機構
+gc.c       :: 記憶領域管理
+load.c     :: ライブラリのロード
+object.c   :: オブジェクト
+variable.c :: 変数と定数
+
+=== Rubyの構文解析器
+
+parse.y       :: 字句解析器と構文定義
+parse.c       :: 自動生成
+defs/keywords :: 予約語
+lex.c         :: 自動生成
+
+=== Rubyの評価器 (通称YARV)
+
+  compile.c
+  eval.c
+  eval_error.c
+  eval_jump.c
+  eval_safe.c
+  insns.def           : 仮想機械語の定義
+  iseq.c              : VM::ISeqの実装
+  thread.c            : スレッド管理とコンテキスト切り替え
+  thread_win32.c      : スレッド実装
+  thread_pthread.c    : 同上
+  vm.c
+  vm_dump.c
+  vm_eval.c
+  vm_exec.c
+  vm_insnhelper.c
+  vm_method.c
+
+  defs/opt_insns_unif.def  : 命令融合
+  defs/opt_operand.def     : 最適化のための定義
+
+    -> insn*.inc           : 自動生成
+    -> opt*.inc            : 自動生成
+    -> vm.inc              : 自動生成
+
+=== 正規表現エンジン (鬼車)
+
+  regex.c
+  regcomp.c
+  regenc.c
+  regerror.c
+  regexec.c
+  regparse.c
+  regsyntax.c
+
+=== ユーティリティ関数
+
+debug.c    :: Cデバッガ用のデバッグシンボル
+dln.c      :: 動的ローディング
+st.c       :: 汎用ハッシュ表
+strftime.c :: 時刻整形
+util.c     :: その他のユーティリティ
+
+=== Rubyコマンドの実装
+
+  dmyext.c
+  dmydln.c
+  dmyencoding.c
+  id.c
+  inits.c
+  main.c
+  ruby.c
+  version.c
+
+  gem_prelude.rb
+  prelude.rb
+
+=== クラスライブラリ
+
+array.c      :: Array
+bignum.c     :: Bignum
+compar.c     :: Comparable
+complex.c    :: Complex
+cont.c       :: Fiber, Continuation
+dir.c        :: Dir
+enum.c       :: Enumerable
+enumerator.c :: Enumerator
+file.c       :: File
+hash.c       :: Hash
+io.c         :: IO
+marshal.c    :: Marshal
+math.c       :: Math
+numeric.c    :: Numeric, Integer, Fixnum, Float
+pack.c       :: Array#pack, String#unpack
+proc.c       :: Binding, Proc
+process.c    :: Process
+random.c     :: 乱数
+range.c      :: Range
+rational.c   :: Rational
+re.c         :: Regexp, MatchData
+signal.c     :: Signal
+sprintf.c    :: String#sprintf
+string.c     :: String
+struct.c     :: Struct
+time.c       :: Time
+
+defs/known_errors.def :: 例外クラス Errno::*
+-> known_errors.inc   :: 自動生成
+
+=== 多言語化
+
+encoding.c  :: Encoding
+transcode.c :: Encoding::Converter
+enc/*.c     :: エンコーディングクラス群
+enc/trans/* :: コードポイント対応表
+
+=== gorubyコマンドの実装
+
+  goruby.c
+  golf_prelude.rb      : goruby固有のライブラリ
+    -> golf_prelude.c  : 自動生成
+
+== Appendix B. 拡張用関数リファレンス
+
+C言語からRubyの機能を利用するAPIは以下の通りである．
+
+=== 型
+
+VALUE ::
+
+  Rubyオブジェクトを表現する型．必要に応じてキャストして用いる．
+  組み込み型を表現するCの型はruby.hに記述してあるRで始まる構造
+  体である．VALUE型をこれらにキャストするためにRで始まる構造体
+  名を全て大文字にした名前のマクロが用意されている．
+
+=== 変数・定数
+
+Qnil ::
+
+  定数: nilオブジェクト
+
+Qtrue ::
+
+  定数: trueオブジェクト(真のデフォルト値)
+
+Qfalse ::
+
+  定数: falseオブジェクト
+
+=== Cデータのカプセル化
+
+Data_Wrap_Struct(VALUE klass, void (*mark)(), void (*free)(), void *sval) ::
+
+  Cの任意のポインタをカプセル化したRubyオブジェクトを返す．こ
+  のポインタがRubyからアクセスされなくなった時，freeで指定した
+  関数が呼ばれる．また，このポインタの指すデータが他のRubyオブ
+  ジェクトを指している場合，markに指定する関数でマークする必要
+  がある．
+
+Data_Make_Struct(klass, type, mark, free, sval) ::
+
+  type型のメモリをmallocし，変数svalに代入した後，それをカプセ
+  ル化したデータを返すマクロ．
+
+Data_Get_Struct(data, type, sval) ::
+
+  dataからtype型のポインタを取り出し変数svalに代入するマクロ．
+
+=== 型チェック
+
+  RB_TYPE_P(value, type)
+  TYPE(value)
+  FIXNUM_P(value)
+  NIL_P(value)
+  RB_INTEGER_TYPE_P(value)
+  RB_FLOAT_TYPE_P(value)
+  void Check_Type(VALUE value, int type)
+  SafeStringValue(value)
+
+=== 型変換
+
+  FIX2INT(value), INT2FIX(i)
+  FIX2LONG(value), LONG2FIX(l)
+  NUM2INT(value), INT2NUM(i)
+  NUM2UINT(value), UINT2NUM(ui)
+  NUM2LONG(value), LONG2NUM(l)
+  NUM2ULONG(value), ULONG2NUM(ul)
+  NUM2LL(value), LL2NUM(ll)
+  NUM2ULL(value), ULL2NUM(ull)
+  NUM2OFFT(value), OFFT2NUM(off)
+  NUM2SIZET(value), SIZET2NUM(size)
+  NUM2SSIZET(value), SSIZET2NUM(ssize)
+  rb_integer_pack(value, words, numwords, wordsize, nails, flags), rb_integer_unpack(words, numwords, wordsize, nails, flags)
+  NUM2DBL(value)
+  rb_float_new(f)
+  RSTRING_LEN(str)
+  RSTRING_PTR(str)
+  StringValue(value)
+  StringValuePtr(value)
+  StringValueCStr(value)
+  rb_str_new2(s)
+
+=== クラス/モジュール定義
+
+VALUE rb_define_class(const char *name, VALUE super) ::
+
+  superのサブクラスとして新しいRubyクラスを定義する．
+
+VALUE rb_define_class_under(VALUE module, const char *name, VALUE super) ::
+
+  superのサブクラスとして新しいRubyクラスを定義し，moduleの
+  定数として定義する．
+
+VALUE rb_define_module(const char *name) ::
+
+  新しいRubyモジュールを定義する．
+
+VALUE rb_define_module_under(VALUE module, const char *name) ::
+
+  新しいRubyモジュールを定義し，moduleの定数として定義する．
+
+void rb_include_module(VALUE klass, VALUE module) ::
+
+  モジュールをインクルードする．classがすでにmoduleをインク
+  ルードしている時には何もしない(多重インクルードの禁止)．
+
+void rb_extend_object(VALUE object, VALUE module) ::
+
+  オブジェクトをモジュール(で定義されているメソッド)で拡張する．
+
+=== 大域変数定義
+
+void rb_define_variable(const char *name, VALUE *var) ::
+
+  RubyとCとで共有するグローバル変数を定義する．変数名が`$'で
+  始まらない時には自動的に追加される．nameとしてRubyの識別子
+  として許されない文字(例えば` ')を含む場合にはRubyプログラ
+  ムからは見えなくなる．
+
+void rb_define_readonly_variable(const char *name, VALUE *var) ::
+
+  RubyとCとで共有するread onlyのグローバル変数を定義する．
+  read onlyであること以外はrb_define_variable()と同じ．
+
+void rb_define_virtual_variable(const char *name, VALUE (*getter)(), void (*setter)()) ::
+
+  関数によって実現されるRuby変数を定義する．変数が参照された
+  時にはgetterが，変数に値がセットされた時にはsetterが呼ばれ
+  る．
+
+void rb_define_hooked_variable(const char *name, VALUE *var, VALUE (*getter)(), void (*setter)()) ::
+
+  関数によってhookのつけられたグローバル変数を定義する．変数
+  が参照された時にはgetterが，関数に値がセットされた時には
+  setterが呼ばれる．getterやsetterに0を指定した時にはhookを
+  指定しないのと同じ事になる．
+
+void rb_global_variable(VALUE *var) ::
+
+  GCのため，Rubyプログラムからはアクセスされないが, Rubyオブ
+  ジェクトを含む大域変数をマークする．
+
+=== 定数
+
+void rb_define_const(VALUE klass, const char *name, VALUE val) ::
+
+  定数を定義する．
+
+void rb_define_global_const(const char *name, VALUE val) ::
+
+  大域定数を定義する．
+
+    rb_define_const(rb_cObject, name, val)
+
+  と同じ意味．
+
+=== メソッド定義
+
+rb_define_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc) ::
+
+  メソッドを定義する．argcはselfを除く引数の数．argcが-1の時,
+  関数には引数の数(selfを含まない)を第1引数, 引数の配列を第2
+  引数とする形式で与えられる(第3引数はself)．argcが-2の時,
+  第1引数がself, 第2引数がargs(argsは引数を含むRubyの配列)と
+  いう形式で与えられる．
+
+rb_define_private_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc) ::
+
+  privateメソッドを定義する．引数はrb_define_method()と同じ．
+
+rb_define_singleton_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc) ::
+
+  特異メソッドを定義する．引数はrb_define_method()と同じ．
+
+rb_check_arity(int argc, int min, int max) ::
+
+  引数の数であるargcがmin..maxの範囲に入っているかをチェックします．
+  もしmaxがUNLIMITED_ARGUMENTSなら，上限はチェックしません．
+  もしargcが範囲外ならArgumentErrorが発生します．
+
+rb_scan_args(int argc, VALUE *argv, const char *fmt, ...) ::
+
+  argc, argv形式で与えられた指定されたフォーマットに従って引
+  数を分解し，続くVALUEへの参照にセットします．このフォーマッ
+  トは，ABNFで記述すると以下の通りです．
+
+    scan-arg-spec  := param-arg-spec [option-hash-arg-spec] [block-arg-spec]
+
+    param-arg-spec := pre-arg-spec [post-arg-spec] / post-arg-spec /
+                      pre-opt-post-arg-spec
+    pre-arg-spec   := num-of-leading-mandatory-args [num-of-optional-args]
+    post-arg-spec  := sym-for-variable-length-args
+                      [num-of-trailing-mandatory-args]
+    pre-opt-post-arg-spec := num-of-leading-mandatory-args num-of-optional-args
+                             num-of-trailing-mandatory-args
+    option-hash-arg-spec := sym-for-option-hash-arg
+    block-arg-spec := sym-for-block-arg
+
+    num-of-leading-mandatory-args  := DIGIT ; 先頭に置かれる省略不能な引数の数
+    num-of-optional-args           := DIGIT ; 続いて置かれる省略可能な引数の数
+    sym-for-variable-length-args   := "*"   ; 続いて置かれる可変長引数を
+                                            ; Rubyの配列で取得するための指定
+    num-of-trailing-mandatory-args := DIGIT ; 終端に置かれる省略不能な引数の数
+    sym-for-option-hash-arg        := ":"   ; オプションハッシュを取得する
+                                            ; ための指定; 省略不能な引数の
+                                            ; 数よりも多くの引数が指定され，
+                                            ; 最後の引数がハッシュ（または
+                                            ; #to_hashで変換可能）の場合に
+                                            ; 取得される．最後の引数がnilの
+                                            ; 場合，可変長引数指定がなく，
+                                            ; 省略不能引数の数よりも多くの
+                                            ; 引数が指定された場合に取得される
+    sym-for-block-arg              := "&"   ; イテレータブロックを取得するための
+                                            ; 指定
+
+  フォーマットが"12"の場合，引数は最低1つで，3つ(1+2)まで許さ
+  れるという意味になります．従って，フォーマット文字列に続い
+  て3つのVALUEへの参照を置く必要があります．それらには取得した
+  変数がセットされます．変数への参照の代わりにNULLを指定する
+  こともでき，その場合は取得した引数の値は捨てられます．なお，
+  省略可能引数が省略された時の変数の値はnil(C言語のレベルでは
+  Qnil)になります．
+
+  返り値は与えられた引数の数です．オプションハッシュおよびイ
+  テレータブロックは数えません．
+
+int rb_get_kwargs(VALUE keyword_hash, const ID *table, int required, int optional, VALUE *values) ::
+
+  キーワードで指定された値をtableにしたがって取り出します．
+  tableの最初のrequired個のIDは必須キーワードを表し，続く
+  optional (optionalが負の場合は-optional-1) 個のIDは省略可能
+  キーワードです．必須キーワードがkeyword_hash中にない場合，
+  "missing keyword"ArgumentErrorが発生します．省略可能キーワー
+  ドがない場合は，values中の対応する要素にはQundefがセットされ
+  ます．keyword_hashに使用されない要素がある場合は，optionalが
+  負なら無視されますが，そうでなければ"unknown keyword"
+  ArgumentErrorが発生します．
+
+VALUE rb_extract_keywords(VALUE *original_hash) ::
+
+  original_hashで参照されるHashオブジェクトから，Symbolである
+  キーとその値を新しいHashに取り出します．original_hashの指す
+  先には，元のHashがSymbol以外のキーを含んでいた場合はそれらが
+  コピーされた別の新しいHash，そうでなければ0が保存されます．
+
+=== Rubyメソッド呼び出し
+
+VALUE rb_funcall(VALUE recv, ID mid, int narg, ...) ::
+
+  メソッド呼び出し．文字列からmidを得るためにはrb_intern()を
+  使う．
+  private/protectedなメソッドでも呼び出せる．
+
+VALUE rb_funcall2(VALUE recv, ID mid, int argc, VALUE *argv) ::
+VALUE rb_funcallv(VALUE recv, ID mid, int argc, VALUE *argv) ::
+
+  メソッド呼び出し．引数をargc, argv形式で渡す．
+  private/protectedなメソッドでも呼び出せる．
+
+VALUE rb_funcallv_public(VALUE recv, ID mid, int argc, VALUE *argv) ::
+
+  メソッド呼び出し．
+  publicなメソッドしか呼べない．
+
+VALUE rb_eval_string(const char *str) ::
+
+  文字列をRubyスクリプトとしてコンパイル・実行する．
+
+ID rb_intern(const char *name) ::
+
+  文字列に対応するIDを返す．
+
+char *rb_id2name(ID id) ::
+
+  IDに対応する文字列を返す(デバッグ用)．
+
+char *rb_class2name(VALUE klass) ::
+
+  クラスの名前を返す(デバッグ用)．クラスが名前を持たない時に
+  は, 祖先を遡って名前を持つクラスの名前を返す．
+
+int rb_respond_to(VALUE obj, ID id) ::
+
+  objがidで示されるメソッドを持つかどうかを返す．
+
+=== インスタンス変数
+
+VALUE rb_iv_get(VALUE obj, const char *name) ::
+
+  objのインスタンス変数の値を得る．`@'で始まらないインスタン
+  ス変数は Rubyプログラムからアクセスできない「隠れた」イン
+  スタンス変数になる．定数は大文字の名前を持つクラス(または
+  モジュール)のインスタンス変数として実装されている．
+
+VALUE rb_iv_set(VALUE obj, const char *name, VALUE val) ::
+
+  objのインスタンス変数をvalにセットする．
+
+=== 制御構造
+
+VALUE rb_block_call(VALUE obj, ID mid, int argc, VALUE * argv, VALUE (*func) (ANYARGS), VALUE data2) ::
+
+  funcをブロックとして設定し，objをレシーバ，argcとargvを引数
+  としてmidメソッドを呼び出す．funcは第一引数にyieldされた値，
+  第二引数にdata2を受け取る．複数の値がyieldされた場合(Cでは
+  rb_yield_values()とrb_yield_values2(), rb_yield_splat())，
+  data2はArrayとしてパックされている．第三, 第四引数のargcと
+  argvによってyieldされた値を取り出すことができる．
+
+\[OBSOLETE] VALUE rb_iterate(VALUE (*func1)(), VALUE arg1, VALUE (*func2)(), VALUE arg2) ::
+
+  func2をブロックとして設定し, func1をイテレータとして呼ぶ．
+  func1には arg1が引数として渡され, func2には第1引数にイテレー
+  タから与えられた値, 第2引数にarg2が渡される．
+
+  1.9でrb_iterateを使う場合は, func1の中でRubyレベルのメソッド
+  を呼び出さなければならない.
+  1.9でobsoleteとなった. 代わりにrb_block_callが用意された.
+
+VALUE rb_yield(VALUE val) ::
+
+  valを値としてイテレータブロックを呼び出す．
+
+VALUE rb_rescue(VALUE (*func1)(ANYARGS), VALUE arg1, VALUE (*func2)(ANYARGS), VALUE arg2) ::
+
+  関数func1をarg1を引数に呼び出す．func1の実行中に例外が発生
+  した時には func2をarg2を第一引数, 発生した例外オブジェクト
+  を第二引数として呼ぶ．戻り値は例外が発生しなかった時はfunc1
+  の戻り値, 例外が発生した時にはfunc2の戻り値である．
+
+VALUE rb_ensure(VALUE (*func1)(ANYARGS), VALUE arg1, VALUE (*func2)(ANYARGS), VALUE arg2) ::
+
+  関数func1をarg1を引数として実行し, 実行終了後(たとえ例外が
+  発生しても) func2をarg2を引数として実行する．戻り値はfunc1
+  の戻り値である(例外が発生した時は戻らない)．
+
+VALUE rb_protect(VALUE (*func) (VALUE), VALUE arg, int *state) ::
+
+  関数funcをargを引数として実行し, 例外が発生しなければその戻
+  り値を返す．例外が発生した場合は, *stateに非0をセットして
+  Qnilを返す．
+  rb_jump_tag()を呼ばずに捕捉した例外を無視する場合には，
+  rb_set_errinfo(Qnil)でエラー情報をクリアしなければならない．
+
+void rb_jump_tag(int state) ::
+
+  rb_protect()やrb_eval_string_protect()で捕捉された例外を再
+  送する．stateはそれらの関数から返された値でなければならない．
+  この関数は直接の呼び出し元に戻らない．
+
+void rb_iter_break() ::
+
+  現在の最も内側のブロックを終了する．この関数は直接の呼び出
+  し元に戻らない．
+
+void rb_iter_break_value(VALUE value) ::
+
+  現在の最も内側のブロックをvalueで終了する．ブロックは引数で
+  与えられたvalueを返す．この関数は直接の呼び出し元に戻らない．
+
+=== 例外・エラー
+
+void rb_warning(const char *fmt, ...) ::
+
+  rb_verbose時に標準エラー出力に警告情報を表示する．引数は
+  printf()と同じ．
+
+void rb_raise(rb_eRuntimeError, const char *fmt, ...) ::
+
+  RuntimeError例外を発生させる．引数はprintf()と同じ．
+
+void rb_raise(VALUE exception, const char *fmt, ...) ::
+
+  exceptionで指定した例外を発生させる．fmt以下の引数は
+  printf()と同じ．
+
+void rb_fatal(const char *fmt, ...) ::
+
+  致命的例外を発生させる．通常の例外処理は行なわれず, インター
+  プリタが終了する(ただしensureで指定されたコードは終了前に
+  実行される)．
+
+void rb_bug(const char *fmt, ...) ::
+
+  インタープリタなどプログラムのバグでしか発生するはずのない
+  状況の時呼ぶ．インタープリタはコアダンプし直ちに終了する．
+  例外処理は一切行なわれない．
+
+注意: "%"PRIsVALUEがObject#to_s('+'フラグが指定されていると
+きはObject#inspect)を使ったVALUEの出力に利用できる．これは
+"%i"と衝突するため，整数には"%d"を使用すること．
+
+=== Rubyの初期化・実行
+
+Rubyをアプリケーションに埋め込む場合には以下のインタフェース
+を使う．通常の拡張ライブラリには必要ない．
+
+void ruby_init() ::
+
+  Rubyインタプリタの初期化を行なう．
+
+void *ruby_options(int argc, char **argv) ::
+
+  Rubyインタプリタのコマンドライン引数の処理を行ない，
+  Rubyのソースコードをコンパイルする．
+  コンパイルされたソースへのポインタ，もしくは特殊値を返す.
+
+int ruby_run_node(void *n) ::
+
+  コンパイルされたコードを実行する．
+  実行に成功した場合はEXIT_SUCCESSを，エラーが起こったときはそれ以外を返す．
+
+void ruby_script(char *name) ::
+
+  Rubyのスクリプト名($0)を設定する．
+
+=== インタプリタのイベントのフック
+
+void rb_add_event_hook(rb_event_hook_func_t func, rb_event_flag_t events, VALUE data) ::
+
+  指定されたインタプリタのイベントに対するフック関数を追加します．
+  eventsは以下の値のorでなければなりません:
+
+    RUBY_EVENT_LINE
+    RUBY_EVENT_CLASS
+    RUBY_EVENT_END
+    RUBY_EVENT_CALL
+    RUBY_EVENT_RETURN
+    RUBY_EVENT_C_CALL
+    RUBY_EVENT_C_RETURN
+    RUBY_EVENT_RAISE
+    RUBY_EVENT_ALL
+
+  rb_event_hook_func_tの定義は以下の通りです:
+
+    typedef void (*rb_event_hook_func_t)(rb_event_t event, VALUE data,
+                                         VALUE self, ID id, VALUE klass)
+
+  rb_add_event_hook() の第3引数 data は，フック関数の第2引数と
+  して渡されます．これは1.8では現在のNODEへのポインタでした．以
+  下の RB_EVENT_HOOKS_HAVE_CALLBACK_DATA も参照してください．
+
+int rb_remove_event_hook(rb_event_hook_func_t func) ::
+
+  指定されたフック関数を削除します．
+
+=== メモリ使用量
+
+void rb_gc_adjust_memory_usage(ssize_t diff) ::
+
+  登録された外部のメモリ使用量を調整します．この関数で外部のライブラリが
+  どのくらいメモリを使っているのかをGCに伝えることができます．正のdiffで
+  この関数を呼び出すとメモリ使用量の増加を意味します．新しいメモリブロッ
+  クが確保されたり，ブロックがより大きなサイズで再割り当てされたりした場
+  合などです．負のdiffでこの関数を呼び出すとメモリ使用量の減少を意味しま
+  す．メモリブロックが解放されたり，メモリブロックがより小さいサイズで再
+  確保されたりした場合などです．この関数はGCを引き起こすかもしれません．
+
+=== 互換性のためのマクロ
+
+APIの互換性をチェックするために以下のマクロがデフォルトで定義されています．
+
+NORETURN_STYLE_NEW ::
+
+  NORETURN マクロが関数型マクロとして定義されていることを意味する．
+
+HAVE_RB_DEFINE_ALLOC_FUNC ::
+
+  rb_define_alloc_func() 関数が提供されていること，つまり
+  allocation framework が使われることを意味する．
+  have_func("rb_define_alloc_func", "ruby.h")
+  の結果と同じ．
+
+HAVE_RB_REG_NEW_STR ::
+
+  StringオブジェクトからRegexpオブジェクトを作る
+  rb_reg_new_str() 関数が提供されていることを意味する．
+  have_func("rb_reg_new_str", "ruby.h").
+  の結果と同じ．
+
+HAVE_RB_IO_T ::
+
+  rb_io_t 型が提供されていることを意味する．
+
+USE_SYMBOL_AS_METHOD_NAME ::
+
+  メソッド名を返すメソッド，Module#methods, \#singleton_methods
+  などがSymbolを返すことを意味する．
+
+HAVE_RUBY_*_H ::
+
+  ruby.h で定義されている．対応するヘッダが提供されていること
+  を意味する．たとえば，HAVE_RUBY_ST_H が定義されている場合は
+  単なる st.h ではなく ruby/st.h を使用する．
+
+RB_EVENT_HOOKS_HAVE_CALLBACK_DATA ::
+
+  rb_add_event_hook() がフック関数に渡す data を第3引数として
+  受け取ることを意味する．
+
+== Appendix C. extconf.rbで使える関数たち
+
+extconf.rbの中では利用可能なコンパイル条件チェックの関数は以
+下の通りである．
+
+have_macro(macro, headers) ::
+
+  ヘッダファイルheaderをインクルードしてマクロmacroが定義さ
+  れているかどうかチェックする．マクロが定義されている時true
+  を返す．
+
+have_library(lib, func) ::
+
+  関数funcを定義しているライブラリlibの存在をチェックする．
+  チェックに成功すると，-llibを$libsに追加し，trueを返す．
+
+find_library(lib, func, path...) ::
+
+  関数funcを定義しているライブラリlibの存在を -Lpath を追加
+  しながらチェックする．チェックに成功すると，-llibを$libsに
+  追加し，trueを返す．
+
+have_func(func, header) ::
+
+  ヘッダファイルheaderをインクルードして関数funcの存在をチェ
+  ックする．funcが標準ではリンクされないライブラリ内のもので
+  ある時には先にhave_libraryでそのライブラリをチェックしてお
+  く事．チェックに成功すると，プリプロセッサマクロ
+  `HAVE_{FUNC}` を定義し，trueを返す．
+
+have_var(var, header) ::
+
+  ヘッダファイルheaderをインクルードして変数varの存在をチェッ
+  クする．varが標準ではリンクされないライブラリ内のものであ
+  る時には先にhave_libraryでそのライブラリをチェックしておく
+  事．チェックに成功すると，プリプロセッサマクロ
+  `HAVE_{VAR}` を定義し，trueを返す．
+
+have_header(header) ::
+
+  ヘッダファイルの存在をチェックする．チェックに成功すると，
+  プリプロセッサマクロ `HAVE_{HEADER_H}` を定義し，trueを返す．
+  (スラッシュやドットはアンダースコアに置換される)
+
+find_header(header, path...) ::
+
+  ヘッダファイルheaderの存在を -Ipath を追加しながらチェック
+  する．チェックに成功すると，プリプロセッサマクロ
+  `HAVE_{HEADER_H}` を定義し，trueを返す．
+  (スラッシュやドットはアンダースコアに置換される)
+
+have_struct_member(type, member[, header[, opt]]) ::
+
+  ヘッダファイルheaderをインクルードして型typeが定義され，
+  なおかつメンバmemberが存在するかをチェックする．チェックに
+  成功すると，プリプロセッサマクロ `HAVE_{TYPE}_{MEMBER}` を
+  定義し，trueを返す．
+
+have_type(type, header, opt) ::
+
+  ヘッダファイルheaderをインクルードして型typeが存在するかを
+  チェックする．チェックに成功すると，プリプロセッサマクロ
+  `HAVE_TYPE_{TYPE}` を定義し，trueを返す．
+
+check_sizeof(type, header) ::
+
+  ヘッダファイルheaderをインクルードして型typeのchar単位サイ
+  ズを調べる．チェックに成功すると，プリプロセッサマクロ
+  `SIZEOF_{TYPE}` を定義し，そのサイズを返す．定義されていな
+  いときはnilを返す．
+
+create_makefile(target[, target_prefix]) ::
+
+  拡張ライブラリ用のMakefileを生成する．この関数を呼ばなけれ
+  ばそのライブラリはコンパイルされない．targetはモジュール名
+  を表す．
+
+find_executable(command, path) ::
+
+  コマンドcommandをFile::PATH_SEPARATORで区切られたパス名の
+  リストpathから探す．pathがnilまたは省略された場合は，環境
+  変数PATHの値を使用する．実行可能なコマンドが見つかった場合
+  はパスを含むファイル名，見つからなかった場合はnilを返す．
+
+with_config(withval[, default=nil]) ::
+
+  コマンドライン上の--with-<withval>で指定されたオプション値
+  を得る．
+
+enable_config(config, *defaults) ::
+disable_config(config, *defaults) ::
+
+  コマンドライン上の--enable-<config>または
+  --disable-<config>で指定された真偽値を得る．
+  --enable-<config>が指定されていた場合はtrue，
+  --disable-<config>が指定されていた場合はfalseを返す．
+  どちらも指定されていない場合は，ブロックつきで呼び出されて
+  いる場合は*defaultsをyieldした結果，ブロックなしなら
+  \*defaultsを返す．
+
+dir_config(target[, default_dir]) ::
+dir_config(target[, default_include, default_lib]) ::
+
+  コマンドライン上の--with-<target>-dir, --with-<target>-include,
+  --with-<target>-libのいずれかで指定されるディレクトリを
+  $CFLAGS や $LDFLAGS に追加する．--with-<target>-dir=/pathは
+  --with-<target>-include=/path/include --with-<target>-lib=/path/lib
+  と等価である．追加された include ディレクトリと lib ディレ
+  クトリの配列を返す． ([include_dir, lib_dir])
+
+pkg_config(pkg, option=nil) ::
+
+  pkg-configコマンドからパッケージpkgの情報を [cflags, ldflags, libs]
+  の配列として得る．$CFLAGS, $LDFLAGS, $libs にはそれぞれの値が
+  追加される．
+
+  pkg-configの実際のコマンドは，以下の順で試される．
+
+  1. コマンドラインで--with-{pkg}-config={command}オプションが
+     指定された場合: {command} {option}
+  2. {pkg}-config {option}
+  3. pkg-config {option} {pkg}
+
+  optionが指定された場合は，上記の配列の代わりにそのオプションを
+  指定して得られた出力をstripしたものを返す．
+
+== Appendix D. 世代別GC
+
+Ruby 2.1から世代別GCに対応しました．我々はこれをRGenGCと呼んでいます．
+RGenGCは，過去の拡張ライブラリに（ほぼ）互換性を保つように開発されている
+ため，拡張ライブラリ側の対応はほぼ不要です．
+
+ただし，対応をすることで性能を向上することができる可能性があります．もし
+拡張ライブラリに高い性能が必要である場合は対応を検討して下さい．
+
+とくにRARRAY_PTR()/RHASH_TBL()のようなマクロを用いてポインタに直接アクセ
+スするようなコードは書かないようにして下さい．代わりに，rb_ary_aref(),
+rb_ary_store() などの，適切な API 関数を利用するようにして下さい．
+
+そのほか，対応についての詳細は extension.rdoc の「Appendix D. Generational
+GC」を参照して下さい．
+
+:enddoc: Local variables:
+:enddoc: fill-column: 60
+:enddoc: end:
diff --git a/doc/extension.rdoc b/doc/extension.rdoc
new file mode 100644
index 0000000000..691631ea17
--- /dev/null
+++ b/doc/extension.rdoc
@@ -0,0 +1,1959 @@
+# extension.rdoc -  -*- RDoc -*- created at: Mon Aug  7 16:45:54 JST 1995
+
+= Creating Extension Libraries for Ruby
+
+This document explains how to make extension libraries for Ruby.
+
+== Basic Knowledge
+
+In C, variables have types and data do not have types.  In contrast,
+Ruby variables do not have a static type, and data themselves have
+types, so data will need to be converted between the languages.
+
+Data in Ruby are represented by the C type `VALUE'.  Each VALUE data
+has its data type.
+
+To retrieve C data from a VALUE, you need to:
+
+1. Identify the VALUE's data type
+2. Convert the VALUE into C data
+
+Converting to the wrong data type may cause serious problems.
+
+=== Data Types
+
+The Ruby interpreter has the following data types:
+
+T_NIL       :: nil
+T_OBJECT    :: ordinary object
+T_CLASS     :: class
+T_MODULE    :: module
+T_FLOAT     :: floating point number
+T_STRING    :: string
+T_REGEXP    :: regular expression
+T_ARRAY     :: array
+T_HASH      :: associative array
+T_STRUCT    :: (Ruby) structure
+T_BIGNUM    :: multi precision integer
+T_FIXNUM    :: Fixnum(31bit or 63bit integer)
+T_COMPLEX   :: complex number
+T_RATIONAL  :: rational number
+T_FILE      :: IO
+T_TRUE      :: true
+T_FALSE     :: false
+T_DATA      :: data
+T_SYMBOL    :: symbol
+
+In addition, there are several other types used internally:
+
+T_ICLASS    :: included module
+T_MATCH     :: MatchData object
+T_UNDEF     :: undefined
+T_NODE      :: syntax tree node
+T_ZOMBIE    :: object awaiting finalization
+
+Most of the types are represented by C structures.
+
+=== Check Data Type of the VALUE
+
+The macro TYPE() defined in ruby.h shows the data type of the VALUE.
+TYPE() returns the constant number T_XXXX described above.  To handle
+data types, your code will look something like this:
+
+  switch (TYPE(obj)) {
+    case T_FIXNUM:
+      /* process Fixnum */
+      break;
+    case T_STRING:
+      /* process String */
+      break;
+    case T_ARRAY:
+      /* process Array */
+      break;
+    default:
+      /* raise exception */
+      rb_raise(rb_eTypeError, "not valid value");
+      break;
+  }
+
+There is the data type check function
+
+  void Check_Type(VALUE value, int type)
+
+which raises an exception if the VALUE does not have the type
+specified.
+
+There are also faster check macros for fixnums and nil.
+
+  FIXNUM_P(obj)
+  NIL_P(obj)
+
+=== Convert VALUE into C Data
+
+The data for type T_NIL, T_FALSE, T_TRUE are nil, false, true
+respectively.  They are singletons for the data type.
+The equivalent C constants are: Qnil, Qfalse, Qtrue.
+Note that Qfalse is false in C also (i.e. 0), but not Qnil.
+
+The T_FIXNUM data is a 31bit or 63bit length fixed integer.
+This size depends on the size of long: if long is 32bit then
+T_FIXNUM is 31bit, if long is 64bit then T_FIXNUM is 63bit.
+T_FIXNUM can be converted to a C integer by using the
+FIX2INT() macro or FIX2LONG().  Though you have to check that the
+data is really FIXNUM before using them, they are faster.  FIX2LONG()
+never raises exceptions, but FIX2INT() raises RangeError if the
+result is bigger or smaller than the size of int.
+There are also NUM2INT() and NUM2LONG() which converts any Ruby
+numbers into C integers.  These macros include a type check,
+so an exception will be raised if the conversion failed.  NUM2DBL()
+can be used to retrieve the double float value in the same way.
+
+You can use the macros
+StringValue() and StringValuePtr() to get a char* from a VALUE.
+StringValue(var) replaces var's value with the result of "var.to_str()".
+StringValuePtr(var) does the same replacement and returns the char*
+representation of var.  These macros will skip the replacement if var
+is a String.  Notice that the macros take only the lvalue as their
+argument, to change the value of var in place.
+
+You can also use the macro named StringValueCStr(). This is just
+like StringValuePtr(), but always adds a NUL character at the end of
+the result. If the result contains a NUL character, this macro causes
+the ArgumentError exception.
+StringValuePtr() doesn't guarantee the existence of a NUL at the end
+of the result, and the result may contain NUL.
+
+Other data types have corresponding C structures, e.g. struct RArray
+for T_ARRAY etc. The VALUE of the type which has the corresponding
+structure can be cast to retrieve the pointer to the struct.  The
+casting macro will be of the form RXXXX for each data type; for
+instance, RARRAY(obj).  See "ruby.h".  However, we do not recommend
+to access RXXXX data directly because these data structures are complex.
+Use corresponding rb_xxx() functions to access the internal struct.
+For example, to access an entry of array, use rb_ary_entry(ary, offset)
+and rb_ary_store(ary, offset, obj).
+
+There are some accessing macros for structure members, for example
+`RSTRING_LEN(str)' to get the size of the Ruby String object.  The
+allocated region can be accessed by `RSTRING_PTR(str)'.
+
+Notice: Do not change the value of the structure directly, unless you
+are responsible for the result.  This ends up being the cause of
+interesting bugs.
+
+=== Convert C Data into VALUE
+
+To convert C data to Ruby values:
+
+FIXNUM ::
+
+  left shift 1 bit, and turn on its least significant bit (LSB).
+
+Other pointer values ::
+
+  cast to VALUE.
+
+You can determine whether a VALUE is a pointer or not by checking its LSB.
+
+Notice: Ruby does not allow arbitrary pointer values to be a VALUE.  They
+should be pointers to the structures which Ruby knows about.  The known
+structures are defined in <ruby.h>.
+
+To convert C numbers to Ruby values, use these macros:
+
+INT2FIX() :: for integers within 31bits.
+INT2NUM() :: for arbitrary sized integers.
+
+INT2NUM() converts an integer into a Bignum if it is out of the FIXNUM
+range, but is a bit slower.
+
+=== Manipulating Ruby Data
+
+As I already mentioned, it is not recommended to modify an object's
+internal structure.  To manipulate objects, use the functions supplied
+by the Ruby interpreter. Some (not all) of the useful functions are
+listed below:
+
+==== String Functions
+
+rb_str_new(const char *ptr, long len) ::
+
+  Creates a new Ruby string.
+
+rb_str_new2(const char *ptr) ::
+rb_str_new_cstr(const char *ptr) ::
+
+  Creates a new Ruby string from a C string.  This is equivalent to
+  rb_str_new(ptr, strlen(ptr)).
+
+rb_str_new_literal(const char *ptr) ::
+
+  Creates a new Ruby string from a C string literal.
+
+rb_tainted_str_new(const char *ptr, long len) ::
+
+  Creates a new tainted Ruby string.  Strings from external data
+  sources should be tainted.
+
+rb_tainted_str_new2(const char *ptr) ::
+rb_tainted_str_new_cstr(const char *ptr) ::
+
+  Creates a new tainted Ruby string from a C string.
+
+rb_sprintf(const char *format, ...) ::
+rb_vsprintf(const char *format, va_list ap) ::
+
+  Creates a new Ruby string with printf(3) format.
+
+  Note: In the format string, "%"PRIsVALUE can be used for Object#to_s
+  (or Object#inspect if '+' flag is set) output (and related argument
+  must be a VALUE).  Since it conflicts with "%i", for integers in
+  format strings, use "%d".
+
+rb_str_append(VALUE str1, VALUE str2) ::
+
+  Appends Ruby string str2 to Ruby string str1.
+
+rb_str_cat(VALUE str, const char *ptr, long len) ::
+
+  Appends len bytes of data from ptr to the Ruby string.
+
+rb_str_cat2(VALUE str, const char* ptr) ::
+rb_str_cat_cstr(VALUE str, const char* ptr) ::
+
+  Appends C string ptr to Ruby string str.  This function is
+  equivalent to rb_str_cat(str, ptr, strlen(ptr)).
+
+rb_str_catf(VALUE str, const char* format, ...) ::
+rb_str_vcatf(VALUE str, const char* format, va_list ap) ::
+
+  Appends C string format and successive arguments to Ruby string
+  str according to a printf-like format.  These functions are
+  equivalent to rb_str_append(str, rb_sprintf(format, ...)) and
+  rb_str_append(str, rb_vsprintf(format, ap)), respectively.
+
+rb_enc_str_new(const char *ptr, long len, rb_encoding *enc) ::
+rb_enc_str_new_cstr(const char *ptr, rb_encoding *enc) ::
+
+  Creates a new Ruby string with the specified encoding.
+
+rb_enc_str_new_literal(const char *ptr, rb_encoding *enc) ::
+
+  Creates a new Ruby string from a C string literal with the specified
+  encoding.
+
+rb_usascii_str_new(const char *ptr, long len) ::
+rb_usascii_str_new_cstr(const char *ptr) ::
+
+  Creates a new Ruby string with encoding US-ASCII.
+
+rb_usascii_str_new_literal(const char *ptr) ::
+
+  Creates a new Ruby string from a C string literal with encoding
+  US-ASCII.
+
+rb_utf8_str_new(const char *ptr, long len) ::
+rb_utf8_str_new_cstr(const char *ptr) ::
+
+  Creates a new Ruby string with encoding UTF-8.
+
+rb_utf8_str_new_literal(const char *ptr) ::
+
+  Creates a new Ruby string from a C string literal with encoding
+  UTF-8.
+
+rb_str_resize(VALUE str, long len) ::
+
+  Resizes a Ruby string to len bytes.  If str is not modifiable, this
+  function raises an exception.  The length of str must be set in
+  advance.  If len is less than the old length the content beyond
+  len bytes is discarded, else if len is greater than the old length
+  the content beyond the old length bytes will not be preserved but
+  will be garbage.  Note that RSTRING_PTR(str) may change by calling
+  this function.
+
+rb_str_set_len(VALUE str, long len) ::
+
+  Sets the length of a Ruby string.  If str is not modifiable, this
+  function raises an exception.  This function preserves the content
+  up to len bytes, regardless RSTRING_LEN(str).  len must not exceed
+  the capacity of str.
+
+==== Array Functions
+
+rb_ary_new() ::
+
+  Creates an array with no elements.
+
+rb_ary_new2(long len) ::
+rb_ary_new_capa(long len) ::
+
+  Creates an array with no elements, allocating internal buffer
+  for len elements.
+
+rb_ary_new3(long n, ...) ::
+rb_ary_new_from_args(long n, ...) ::
+
+  Creates an n-element array from the arguments.
+
+rb_ary_new4(long n, VALUE *elts) ::
+rb_ary_new_from_values(long n, VALUE *elts) ::
+
+  Creates an n-element array from a C array.
+
+rb_ary_to_ary(VALUE obj) ::
+
+  Converts the object into an array.
+  Equivalent to Object#to_ary.
+
+There are many functions to operate an array.  They may dump core if other
+types are given.
+
+rb_ary_aref(int argc, const VALUE *argv, VALUE ary) ::
+
+  Equivalent to Array#[].
+
+rb_ary_entry(VALUE ary, long offset) ::
+
+  \ary[offset]
+
+rb_ary_store(VALUE ary, long offset, VALUE obj) ::
+
+  \ary[offset] = obj
+
+rb_ary_subseq(VALUE ary, long beg, long len) ::
+
+  ary[beg, len]
+
+rb_ary_push(VALUE ary, VALUE val) ::
+rb_ary_pop(VALUE ary) ::
+rb_ary_shift(VALUE ary) ::
+rb_ary_unshift(VALUE ary, VALUE val) ::
+
+  ary.push, ary.pop, ary.shift, ary.unshift
+
+rb_ary_cat(VALUE ary, const VALUE *ptr, long len) ::
+
+  Appends len elements of objects from ptr to the array.
+
+== Extending Ruby with C
+
+=== Adding New Features to Ruby
+
+You can add new features (classes, methods, etc.) to the Ruby
+interpreter.  Ruby provides APIs for defining the following things:
+
+- Classes, Modules
+- Methods, Singleton Methods
+- Constants
+
+==== Class and Module Definition
+
+To define a class or module, use the functions below:
+
+  VALUE rb_define_class(const char *name, VALUE super)
+  VALUE rb_define_module(const char *name)
+
+These functions return the newly created class or module.  You may
+want to save this reference into a variable to use later.
+
+To define nested classes or modules, use the functions below:
+
+  VALUE rb_define_class_under(VALUE outer, const char *name, VALUE super)
+  VALUE rb_define_module_under(VALUE outer, const char *name)
+
+==== Method and Singleton Method Definition
+
+To define methods or singleton methods, use these functions:
+
+  void rb_define_method(VALUE klass, const char *name,
+                        VALUE (*func)(ANYARGS), int argc)
+
+  void rb_define_singleton_method(VALUE object, const char *name,
+                                  VALUE (*func)(ANYARGS), int argc)
+
+The `argc' represents the number of the arguments to the C function,
+which must be less than 17.  But I doubt you'll need that many.
+
+If `argc' is negative, it specifies the calling sequence, not number of
+the arguments.
+
+If argc is -1, the function will be called as:
+
+  VALUE func(int argc, VALUE *argv, VALUE obj)
+
+where argc is the actual number of arguments, argv is the C array of
+the arguments, and obj is the receiver.
+
+If argc is -2, the arguments are passed in a Ruby array. The function
+will be called like:
+
+  VALUE func(VALUE obj, VALUE args)
+
+where obj is the receiver, and args is the Ruby array containing
+actual arguments.
+
+There are some more functions to define methods. One takes an ID
+as the name of method to be defined. See also ID or Symbol below.
+
+  void rb_define_method_id(VALUE klass, ID name,
+                           VALUE (*func)(ANYARGS), int argc)
+
+There are two functions to define private/protected methods:
+
+  void rb_define_private_method(VALUE klass, const char *name,
+                                VALUE (*func)(ANYARGS), int argc)
+  void rb_define_protected_method(VALUE klass, const char *name,
+                                  VALUE (*func)(ANYARGS), int argc)
+
+At last, rb_define_module_function defines a module function,
+which are private AND singleton methods of the module.
+For example, sqrt is a module function defined in the Math module.
+It can be called in the following way:
+
+  Math.sqrt(4)
+
+or
+
+  include Math
+  sqrt(4)
+
+To define module functions, use:
+
+  void rb_define_module_function(VALUE module, const char *name,
+                                 VALUE (*func)(ANYARGS), int argc)
+
+In addition, function-like methods, which are private methods defined
+in the Kernel module, can be defined using:
+
+  void rb_define_global_function(const char *name, VALUE (*func)(ANYARGS), int argc)
+
+To define an alias for the method,
+
+  void rb_define_alias(VALUE module, const char* new, const char* old);
+
+To define a reader/writer for an attribute,
+
+  void rb_define_attr(VALUE klass, const char *name, int read, int write)
+
+To define and undefine the `allocate' class method,
+
+  void rb_define_alloc_func(VALUE klass, VALUE (*func)(VALUE klass));
+  void rb_undef_alloc_func(VALUE klass);
+
+func has to take the klass as the argument and return a newly
+allocated instance.  This instance should be as empty as possible,
+without any expensive (including external) resources.
+
+If you are overriding an existing method of any ancestor of your class,
+you may rely on:
+
+  VALUE rb_call_super(int argc, const VALUE *argv)
+
+To achieve the receiver of the current scope (if no other way is
+available), you can use:
+
+  VALUE rb_current_receiver(void)
+
+==== Constant Definition
+
+We have 2 functions to define constants:
+
+  void rb_define_const(VALUE klass, const char *name, VALUE val)
+  void rb_define_global_const(const char *name, VALUE val)
+
+The former is to define a constant under specified class/module.  The
+latter is to define a global constant.
+
+=== Use Ruby Features from C
+
+There are several ways to invoke Ruby's features from C code.
+
+==== Evaluate Ruby Programs in a String
+
+The easiest way to use Ruby's functionality from a C program is to
+evaluate the string as Ruby program.  This function will do the job:
+
+  VALUE rb_eval_string(const char *str)
+
+Evaluation is done under the current context, thus current local variables
+of the innermost method (which is defined by Ruby) can be accessed.
+
+Note that the evaluation can raise an exception. There is a safer
+function:
+
+  VALUE rb_eval_string_protect(const char *str, int *state)
+
+It returns nil when an error occurred. Moreover, *state is zero if str was
+successfully evaluated, or nonzero otherwise.
+
+==== ID or Symbol
+
+You can invoke methods directly, without parsing the string.  First I
+need to explain about ID.  ID is the integer number to represent
+Ruby's identifiers such as variable names.  The Ruby data type
+corresponding to ID is Symbol.  It can be accessed from Ruby in the
+form:
+
+  :Identifier
+
+or
+
+  :"any kind of string"
+
+You can get the ID value from a string within C code by using
+
+  rb_intern(const char *name)
+  rb_intern_str(VALUE name)
+
+You can retrieve ID from Ruby object (Symbol or String) given as an
+argument by using
+
+  rb_to_id(VALUE symbol)
+  rb_check_id(volatile VALUE *name)
+  rb_check_id_cstr(const char *name, long len, rb_encoding *enc)
+
+These functions try to convert the argument to a String if it was not
+a Symbol nor a String.  The second function stores the converted
+result into *name, and returns 0 if the string is not a known symbol.
+After this function returned a non-zero value, *name is always a
+Symbol or a String, otherwise it is a String if the result is 0.
+The third function takes NUL-terminated C string, not Ruby VALUE.
+
+You can retrieve Symbol from Ruby object (Symbol or String) given as
+an argument by using
+
+  rb_to_symbol(VALUE name)
+  rb_check_symbol(volatile VALUE *namep)
+  rb_check_symbol_cstr(const char *ptr, long len, rb_encoding *enc)
+
+These functions are similar to above functions except that these
+return a Symbol instead of an ID.
+
+You can convert C ID to Ruby Symbol by using
+
+  VALUE ID2SYM(ID id)
+
+and to convert Ruby Symbol object to ID, use
+
+  ID SYM2ID(VALUE symbol)
+
+==== Invoke Ruby Method from C
+
+To invoke methods directly, you can use the function below
+
+  VALUE rb_funcall(VALUE recv, ID mid, int argc, ...)
+
+This function invokes a method on the recv, with the method name
+specified by the symbol mid.
+
+==== Accessing the Variables and Constants
+
+You can access class variables and instance variables using access
+functions.  Also, global variables can be shared between both
+environments.  There's no way to access Ruby's local variables.
+
+The functions to access/modify instance variables are below:
+
+  VALUE rb_ivar_get(VALUE obj, ID id)
+  VALUE rb_ivar_set(VALUE obj, ID id, VALUE val)
+
+id must be the symbol, which can be retrieved by rb_intern().
+
+To access the constants of the class/module:
+
+  VALUE rb_const_get(VALUE obj, ID id)
+
+See also Constant Definition above.
+
+== Information Sharing Between Ruby and C
+
+=== Ruby Constants That Can Be Accessed From C
+
+As stated in section 1.3,
+the following Ruby constants can be referred from C.
+
+Qtrue ::
+Qfalse ::
+
+  Boolean values.  Qfalse is false in C also (i.e. 0).
+
+Qnil ::
+
+  Ruby nil in C scope.
+
+=== Global Variables Shared Between C and Ruby
+
+Information can be shared between the two environments using shared global
+variables.  To define them, you can use functions listed below:
+
+  void rb_define_variable(const char *name, VALUE *var)
+
+This function defines the variable which is shared by both environments.
+The value of the global variable pointed to by `var' can be accessed
+through Ruby's global variable named `name'.
+
+You can define read-only (from Ruby, of course) variables using the
+function below.
+
+  void rb_define_readonly_variable(const char *name, VALUE *var)
+
+You can define hooked variables.  The accessor functions (getter and
+setter) are called on access to the hooked variables.
+
+  void rb_define_hooked_variable(const char *name, VALUE *var,
+                                 VALUE (*getter)(), void (*setter)())
+
+If you need to supply either setter or getter, just supply 0 for the
+hook you don't need.  If both hooks are 0, rb_define_hooked_variable()
+works just like rb_define_variable().
+
+The prototypes of the getter and setter functions are as follows:
+
+  VALUE (*getter)(ID id, VALUE *var);
+  void (*setter)(VALUE val, ID id, VALUE *var);
+
+Also you can define a Ruby global variable without a corresponding C
+variable.  The value of the variable will be set/get only by hooks.
+
+  void rb_define_virtual_variable(const char *name,
+                                  VALUE (*getter)(), void (*setter)())
+
+The prototypes of the getter and setter functions are as follows:
+
+  VALUE (*getter)(ID id);
+  void (*setter)(VALUE val, ID id);
+
+=== Encapsulate C Data into a Ruby Object
+
+Sometimes you need to expose your struct in the C world as a Ruby
+object.
+In a situation like this, making use of the TypedData_XXX macro
+family, the pointer to the struct and the Ruby object can be mutually
+converted.
+
+--
+The old (non-Typed) Data_XXX macro family has been deprecated.
+In the future version of Ruby, it is possible old macros will not
+work.
+++
+
+==== C struct to Ruby object
+
+You can convert sval, a pointer to your struct, into a Ruby object
+with the next macro.
+
+  TypedData_Wrap_Struct(klass, data_type, sval)
+
+TypedData_Wrap_Struct() returns a created Ruby object as a VALUE.
+
+The klass argument is the class for the object.
+data_type is a pointer to a const rb_data_type_t which describes
+how Ruby should manage the struct.
+
+It is recommended that klass derives from a special class called
+Data (rb_cData) but not from Object or other ordinal classes.
+If it doesn't, you have to call rb_undef_alloc_func(klass).
+
+rb_data_type_t is defined like this.  Let's take a look at each
+member of the struct.
+
+  typedef struct rb_data_type_struct rb_data_type_t;
+
+  struct rb_data_type_struct {
+          const char *wrap_struct_name;
+          struct {
+                  void (*dmark)(void*);
+                  void (*dfree)(void*);
+                  size_t (*dsize)(const void *);
+                  void *reserved[2];
+          } function;
+          const rb_data_type_t *parent;
+          void *data;
+          VALUE flags;
+  };
+
+wrap_struct_name is an identifier of this instance of the struct.
+It is basically used for collecting and emitting statistics.
+So the identifier must be unique in the process, but doesn't need
+to be valid as a C or Ruby identifier.
+
+These dmark / dfree functions are invoked during GC execution.  No
+object allocations are allowed during it, so do not allocate ruby
+objects inside them.
+
+dmark is a function to mark Ruby objects referred from your struct.
+It must mark all references from your struct with rb_gc_mark or
+its family if your struct keeps such references.
+
+--
+Note that it is recommended to avoid such a reference.
+++
+
+dfree is a function to free the pointer allocation.
+If this is -1, the pointer will be just freed.
+
+dsize calculates memory consumption in bytes by the struct.
+Its parameter is a pointer to your struct.
+You can pass 0 as dsize if it is hard to implement such a function.
+But it is still recommended to avoid 0.
+
+You have to fill reserved and parent with 0.
+
+You can fill "data" with an arbitrary value for your use.
+Ruby does nothing with the member.
+
+flags is a bitwise-OR of the following flag values.
+Since they require deep understanding of garbage collector in Ruby,
+you can just set 0 to flags if you are not sure.
+
+RUBY_TYPED_FREE_IMMEDIATELY ::
+
+  This flag makes the garbage collector immediately invoke dfree()
+  during GC when it need to free your struct.
+  You can specify this flag if the dfree never unlocks Ruby's
+  internal lock (GVL).
+
+  If this flag is not set, Ruby defers invokation of dfree()
+  and invokes dfree() at the same time as finalizers.
+
+RUBY_TYPED_WB_PROTECTED ::
+
+  It shows that implementation of the object supports write barriers.
+  If this flag is set, Ruby is better able to do garbage collection
+  of the object.
+
+  When it is set, however, you are responsible for putting write
+  barriers in all implementations of methods of that object as
+  appropriate. Otherwise Ruby might crash while running.
+
+  More about write barriers can be found in "Generational GC" in
+  Appendix D.
+
+You can allocate and wrap the structure in one step.
+
+  TypedData_Make_Struct(klass, type, data_type, sval)
+
+This macro returns an allocated Data object, wrapping the pointer to
+the structure, which is also allocated.  This macro works like:
+
+  (sval = ZALLOC(type), TypedData_Wrap_Struct(klass, data_type, sval))
+
+Arguments klass and data_type work like their counterparts in
+TypedData_Wrap_Struct().  A pointer to the allocated structure will
+be assigned to sval, which should be a pointer of the type specified.
+
+==== Ruby object to C struct
+
+To retrieve the C pointer from the Data object, use the macro
+TypedData_Get_Struct().
+
+  TypedData_Get_Struct(obj, type, &data_type, sval)
+
+A pointer to the structure will be assigned to the variable sval.
+
+See the example below for details.
+
+== Example - Creating the dbm Extension
+
+OK, here's the example of making an extension library.  This is the
+extension to access DBMs.  The full source is included in the ext/
+directory in the Ruby's source tree.
+
+=== Make the Directory
+
+  % mkdir ext/dbm
+
+Make a directory for the extension library under ext directory.
+
+=== Design the Library
+
+You need to design the library features, before making it.
+
+=== Write the C Code
+
+You need to write C code for your extension library.  If your library
+has only one source file, choosing ``LIBRARY.c'' as a file name is
+preferred.  On the other hand, in case your library has multiple source
+files, avoid choosing ``LIBRARY.c'' for a file name.  It may conflict
+with an intermediate file ``LIBRARY.o'' on some platforms.
+Note that some functions in mkmf library described below generate
+a file ``conftest.c'' for checking with compilation.  You shouldn't
+choose ``conftest.c'' as a name of a source file.
+
+Ruby will execute the initializing function named ``Init_LIBRARY'' in
+the library.  For example, ``Init_dbm()'' will be executed when loading
+the library.
+
+Here's the example of an initializing function.
+
+  void
+  Init_dbm(void)
+  {
+      /* define DBM class */
+      VALUE cDBM = rb_define_class("DBM", rb_cObject);
+      /* DBM includes Enumerable module */
+      rb_include_module(cDBM, rb_mEnumerable);
+
+      /* DBM has class method open(): arguments are received as C array */
+      rb_define_singleton_method(cDBM, "open", fdbm_s_open, -1);
+
+      /* DBM instance method close(): no args */
+      rb_define_method(cDBM, "close", fdbm_close, 0);
+      /* DBM instance method []: 1 argument */
+      rb_define_method(cDBM, "[]", fdbm_fetch, 1);
+
+      /* ... */
+
+      /* ID for a instance variable to store DBM data */
+      id_dbm = rb_intern("dbm");
+  }
+
+The dbm extension wraps the dbm struct in the C environment using
+TypedData_Make_Struct.
+
+  struct dbmdata {
+      int  di_size;
+      DBM *di_dbm;
+  };
+
+  static const rb_data_type_t dbm_type = {
+      "dbm",
+      {0, free_dbm, memsize_dbm,},
+      0, 0,
+      RUBY_TYPED_FREE_IMMEDIATELY,
+  };
+
+  obj = TypedData_Make_Struct(klass, struct dbmdata, &dbm_type, dbmp);
+
+This code wraps the dbmdata structure into a Ruby object.  We avoid
+wrapping DBM* directly, because we want to cache size information.
+
+To retrieve the dbmdata structure from a Ruby object, we define the
+following macro:
+
+  #define GetDBM(obj, dbmp) do {\
+      TypedData_Get_Struct((obj), struct dbmdata, &dbm_type, (dbmp));\
+      if ((dbmp) == 0) closed_dbm();\
+      if ((dbmp)->di_dbm == 0) closed_dbm();\
+  } while (0)
+
+This sort of complicated macro does the retrieving and close checking
+for the DBM.
+
+There are three kinds of way to receive method arguments.  First,
+methods with a fixed number of arguments receive arguments like this:
+
+  static VALUE
+  fdbm_delete(VALUE obj, VALUE keystr)
+  {
+        /* ... */
+  }
+
+The first argument of the C function is the self, the rest are the
+arguments to the method.
+
+Second, methods with an arbitrary number of arguments receive
+arguments like this:
+
+  static VALUE
+  fdbm_s_open(int argc, VALUE *argv, VALUE klass)
+  {
+      /* ... */
+      if (rb_scan_args(argc, argv, "11", &file, &vmode) == 1) {
+          mode = 0666;          /* default value */
+      }
+      /* ... */
+  }
+
+The first argument is the number of method arguments, the second
+argument is the C array of the method arguments, and the third
+argument is the receiver of the method.
+
+You can use the function rb_scan_args() to check and retrieve the
+arguments.  The third argument is a string that specifies how to
+capture method arguments and assign them to the following VALUE
+references.
+
+You can just check the argument number with rb_check_arity(), this is
+handy in the case you want to treat the arguments as a list.
+
+The following is an example of a method that takes arguments by Ruby's
+array:
+
+  static VALUE
+  thread_initialize(VALUE thread, VALUE args)
+  {
+      /* ... */
+  }
+
+The first argument is the receiver, the second one is the Ruby array
+which contains the arguments to the method.
+
+<b>Notice</b>: GC should know about global variables which refer to Ruby's objects,
+but are not exported to the Ruby world.  You need to protect them by
+
+  void rb_global_variable(VALUE *var)
+
+=== Prepare extconf.rb
+
+If the file named extconf.rb exists, it will be executed to generate
+Makefile.
+
+extconf.rb is the file for checking compilation conditions etc.  You
+need to put
+
+  require 'mkmf'
+
+at the top of the file.  You can use the functions below to check
+various conditions.
+
+  have_macro(macro[, headers[, opt]]): check whether macro is defined
+  have_library(lib[, func[, headers[, opt]]]): check whether library containing function exists
+  find_library(lib[, func, *paths]): find library from paths
+  have_func(func[, headers[, opt]): check whether function exists
+  have_var(var[, headers[, opt]]): check whether variable exists
+  have_header(header[, preheaders[, opt]]): check whether header file exists
+  find_header(header, *paths): find header from paths
+  have_framework(fw): check whether framework exists (for MacOS X)
+  have_struct_member(type, member[, headers[, opt]]): check whether struct has member
+  have_type(type[, headers[, opt]]): check whether type exists
+  find_type(type, opt, *headers): check whether type exists in headers
+  have_const(const[, headers[, opt]]): check whether constant is defined
+  check_sizeof(type[, headers[, opts]]): check size of type
+  check_signedness(type[, headers[, opts]]): check signedness of type
+  convertible_int(type[, headers[, opts]]): find convertible integer type
+  find_executable(bin[, path]): find executable file path
+  create_header(header): generate configured header
+  create_makefile(target[, target_prefix]): generate Makefile
+
+See MakeMakefile for full documentation of these functions.
+
+The value of the variables below will affect the Makefile.
+
+  $CFLAGS: included in CFLAGS make variable (such as -O)
+  $CPPFLAGS: included in CPPFLAGS make variable (such as -I, -D)
+  $LDFLAGS: included in LDFLAGS make variable (such as -L)
+  $objs: list of object file names
+
+Normally, the object files list is automatically generated by searching
+source files, but you must define them explicitly if any sources will
+be generated while building.
+
+If a compilation condition is not fulfilled, you should not call
+``create_makefile''.  The Makefile will not be generated, compilation will
+not be done.
+
+=== Prepare Depend (Optional)
+
+If the file named depend exists, Makefile will include that file to
+check dependencies.  You can make this file by invoking
+
+  % gcc -MM *.c > depend
+
+It's harmless.  Prepare it.
+
+=== Generate Makefile
+
+Try generating the Makefile by:
+
+  ruby extconf.rb
+
+If the library should be installed under vendor_ruby directory
+instead of site_ruby directory, use --vendor option as follows.
+
+  ruby extconf.rb --vendor
+
+You don't need this step if you put the extension library under the ext
+directory of the ruby source tree.  In that case, compilation of the
+interpreter will do this step for you.
+
+=== Run make
+
+Type
+
+  make
+
+to compile your extension.  You don't need this step either if you have
+put the extension library under the ext directory of the ruby source tree.
+
+=== Debug
+
+You may need to rb_debug the extension.  Extensions can be linked
+statically by adding the directory name in the ext/Setup file so that
+you can inspect the extension with the debugger.
+
+=== Done! Now You Have the Extension Library
+
+You can do anything you want with your library.  The author of Ruby
+will not claim any restrictions on your code depending on the Ruby API.
+Feel free to use, modify, distribute or sell your program.
+
+== Appendix A. Ruby Source Files Overview
+
+=== Ruby Language Core
+
+class.c    :: classes and modules
+error.c    :: exception classes and exception mechanism
+gc.c       :: memory management
+load.c     :: library loading
+object.c   :: objects
+variable.c :: variables and constants
+
+=== Ruby Syntax Parser
+
+parse.y       :: grammar definition
+parse.c       :: automatically generated from parse.y
+defs/keywords :: reserved keywords
+lex.c         :: automatically generated from keywords
+
+=== Ruby Evaluator (a.k.a. YARV)
+
+  compile.c
+  eval.c
+  eval_error.c
+  eval_jump.c
+  eval_safe.c
+  insns.def           : definition of VM instructions
+  iseq.c              : implementation of VM::ISeq
+  thread.c            : thread management and context switching
+  thread_win32.c      : thread implementation
+  thread_pthread.c    : ditto
+  vm.c
+  vm_dump.c
+  vm_eval.c
+  vm_exec.c
+  vm_insnhelper.c
+  vm_method.c
+
+  defs/opt_insns_unif.def  : instruction unification
+  defs/opt_operand.def     : definitions for optimization
+
+    -> insn*.inc           : automatically generated
+    -> opt*.inc            : automatically generated
+    -> vm.inc              : automatically generated
+
+=== Regular Expression Engine (Oniguruma)
+
+  regex.c
+  regcomp.c
+  regenc.c
+  regerror.c
+  regexec.c
+  regparse.c
+  regsyntax.c
+
+=== Utility Functions
+
+debug.c    :: debug symbols for C debugger
+dln.c      :: dynamic loading
+st.c       :: general purpose hash table
+strftime.c :: formatting times
+util.c     :: misc utilities
+
+=== Ruby Interpreter Implementation
+
+  dmyext.c
+  dmydln.c
+  dmyencoding.c
+  id.c
+  inits.c
+  main.c
+  ruby.c
+  version.c
+
+  gem_prelude.rb
+  prelude.rb
+
+=== Class Library
+
+array.c      :: Array
+bignum.c     :: Bignum
+compar.c     :: Comparable
+complex.c    :: Complex
+cont.c       :: Fiber, Continuation
+dir.c        :: Dir
+enum.c       :: Enumerable
+enumerator.c :: Enumerator
+file.c       :: File
+hash.c       :: Hash
+io.c         :: IO
+marshal.c    :: Marshal
+math.c       :: Math
+numeric.c    :: Numeric, Integer, Fixnum, Float
+pack.c       :: Array#pack, String#unpack
+proc.c       :: Binding, Proc
+process.c    :: Process
+random.c     :: random number
+range.c      :: Range
+rational.c   :: Rational
+re.c         :: Regexp, MatchData
+signal.c     :: Signal
+sprintf.c    :: String#sprintf
+string.c     :: String
+struct.c     :: Struct
+time.c       :: Time
+
+defs/known_errors.def :: Errno::* exception classes
+-> known_errors.inc   :: automatically generated
+
+=== Multilingualization
+
+encoding.c  :: Encoding
+transcode.c :: Encoding::Converter
+enc/*.c     :: encoding classes
+enc/trans/* :: codepoint mapping tables
+
+=== goruby Interpreter Implementation
+
+  goruby.c
+  golf_prelude.rb     : goruby specific libraries.
+    -> golf_prelude.c : automatically generated
+
+== Appendix B. Ruby Extension API Reference
+
+=== Types
+
+VALUE ::
+
+  The type for the Ruby object.  Actual structures are defined in ruby.h,
+  such as struct RString, etc.  To refer the values in structures, use
+  casting macros like RSTRING(obj).
+
+=== Variables and Constants
+
+Qnil ::
+
+  nil object
+
+Qtrue ::
+
+  true object (default true value)
+
+Qfalse ::
+
+  false object
+
+=== C Pointer Wrapping
+
+Data_Wrap_Struct(VALUE klass, void (*mark)(), void (*free)(), void *sval) ::
+
+  Wrap a C pointer into a Ruby object.  If object has references to other
+  Ruby objects, they should be marked by using the mark function during
+  the GC process.  Otherwise, mark should be 0.  When this object is no
+  longer referred by anywhere, the pointer will be discarded by free
+  function.
+
+Data_Make_Struct(klass, type, mark, free, sval) ::
+
+  This macro allocates memory using malloc(), assigns it to the variable
+  sval, and returns the DATA encapsulating the pointer to memory region.
+
+Data_Get_Struct(data, type, sval) ::
+
+  This macro retrieves the pointer value from DATA, and assigns it to
+  the variable sval.
+
+=== Checking Data Types
+
+RB_TYPE_P(value, type) ::
+
+  Is +value+ an internal type (T_NIL, T_FIXNUM, etc.)?
+
+TYPE(value) ::
+
+  Internal type (T_NIL, T_FIXNUM, etc.)
+
+FIXNUM_P(value) ::
+
+  Is +value+ a Fixnum?
+
+NIL_P(value) ::
+
+  Is +value+ nil?
+
+RB_INTEGER_TYPE_P(value) ::
+
+  Is +value+ an Integer?
+
+RB_FLOAT_TYPE_P(value) ::
+
+  Is +value+ a Float?
+
+void Check_Type(VALUE value, int type) ::
+
+  Ensures +value+ is of the given internal +type+ or raises a TypeError
+
+SafeStringValue(value) ::
+
+  Checks that +value+ is a String and is not tainted
+
+=== Data Type Conversion
+
+FIX2INT(value), INT2FIX(i) ::
+
+  Fixnum <-> integer
+
+FIX2LONG(value), LONG2FIX(l) ::
+
+  Fixnum <-> long
+
+NUM2INT(value), INT2NUM(i) ::
+
+  Numeric <-> integer
+
+NUM2UINT(value), UINT2NUM(ui) ::
+
+  Numeric <-> unsigned integer
+
+NUM2LONG(value), LONG2NUM(l) ::
+
+  Numeric <-> long
+
+NUM2ULONG(value), ULONG2NUM(ul) ::
+
+  Numeric <-> unsigned long
+
+NUM2LL(value), LL2NUM(ll) ::
+
+  Numeric <-> long long
+
+NUM2ULL(value), ULL2NUM(ull) ::
+
+  Numeric <-> unsigned long long
+
+NUM2OFFT(value), OFFT2NUM(off) ::
+
+  Numeric <-> off_t
+
+NUM2SIZET(value), SIZET2NUM(size) ::
+
+  Numeric <-> size_t
+
+NUM2SSIZET(value), SSIZET2NUM(ssize) ::
+
+  Numeric <-> ssize_t
+
+rb_integer_pack(value, words, numwords, wordsize, nails, flags), rb_integer_unpack(words, numwords, wordsize, nails, flags) ::
+
+  Numeric <-> Arbitrary size integer buffer
+
+NUM2DBL(value) ::
+
+  Numeric -> double
+
+rb_float_new(f) ::
+
+  double -> Float
+
+RSTRING_LEN(str) ::
+
+  String -> length of String data in bytes
+
+RSTRING_PTR(str) ::
+
+  String -> pointer to String data
+  Note that the result pointer may not be NUL-terminated
+
+StringValue(value) ::
+
+  Object with \#to_str -> String
+
+StringValuePtr(value) ::
+
+  Object with \#to_str -> pointer to String data
+
+StringValueCStr(value) ::
+
+  Object with \#to_str -> pointer to String data without NUL bytes
+  It is guaranteed that the result data is NUL-terminated
+
+rb_str_new2(s) ::
+
+  char * -> String
+
+=== Defining Classes and Modules
+
+VALUE rb_define_class(const char *name, VALUE super) ::
+
+  Defines a new Ruby class as a subclass of super.
+
+VALUE rb_define_class_under(VALUE module, const char *name, VALUE super) ::
+
+  Creates a new Ruby class as a subclass of super, under the module's
+  namespace.
+
+VALUE rb_define_module(const char *name) ::
+
+  Defines a new Ruby module.
+
+VALUE rb_define_module_under(VALUE module, const char *name) ::
+
+  Defines a new Ruby module under the module's namespace.
+
+void rb_include_module(VALUE klass, VALUE module) ::
+
+  Includes module into class.  If class already includes it, just ignored.
+
+void rb_extend_object(VALUE object, VALUE module) ::
+
+  Extend the object with the module's attributes.
+
+=== Defining Global Variables
+
+void rb_define_variable(const char *name, VALUE *var) ::
+
+  Defines a global variable which is shared between C and Ruby.  If name
+  contains a character which is not allowed to be part of the symbol,
+  it can't be seen from Ruby programs.
+
+void rb_define_readonly_variable(const char *name, VALUE *var) ::
+
+  Defines a read-only global variable.  Works just like
+  rb_define_variable(), except the defined variable is read-only.
+
+void rb_define_virtual_variable(const char *name, VALUE (*getter)(), void (*setter)()) ::
+
+  Defines a virtual variable, whose behavior is defined by a pair of C
+  functions.  The getter function is called when the variable is
+  referenced.  The setter function is called when the variable is set to a
+  value.  The prototype for getter/setter functions are:
+
+    VALUE getter(ID id)
+    void setter(VALUE val, ID id)
+
+  The getter function must return the value for the access.
+
+void rb_define_hooked_variable(const char *name, VALUE *var, VALUE (*getter)(), void (*setter)()) ::
+
+  Defines hooked variable.  It's a virtual variable with a C variable.
+  The getter is called as
+
+    VALUE getter(ID id, VALUE *var)
+
+  returning a new value.  The setter is called as
+
+    void setter(VALUE val, ID id, VALUE *var)
+
+void rb_global_variable(VALUE *var) ::
+
+  GC requires C global variables which hold Ruby values to be marked.
+  rb_global_variable tells GC to protect these variables.
+
+=== Constant Definition
+
+void rb_define_const(VALUE klass, const char *name, VALUE val) ::
+
+  Defines a new constant under the class/module.
+
+void rb_define_global_const(const char *name, VALUE val) ::
+
+  Defines a global constant.  This is just the same as
+
+    rb_define_const(rb_cObject, name, val)
+
+=== Method Definition
+
+rb_define_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc) ::
+
+  Defines a method for the class.  func is the function pointer.  argc
+  is the number of arguments.  if argc is -1, the function will receive
+  3 arguments: argc, argv, and self.  if argc is -2, the function will
+  receive 2 arguments, self and args, where args is a Ruby array of
+  the method arguments.
+
+rb_define_private_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc) ::
+
+  Defines a private method for the class.  Arguments are same as
+  rb_define_method().
+
+rb_define_singleton_method(VALUE klass, const char *name, VALUE (*func)(ANYARGS), int argc) ::
+
+  Defines a singleton method.  Arguments are same as rb_define_method().
+
+rb_check_arity(int argc, int min, int max) ::
+
+  Check the number of arguments, argc is in the range of min..max.  If
+  max is UNLIMITED_ARGUMENTS, upper bound is not checked.  If argc is
+  out of bounds, an ArgumentError will be raised.
+
+rb_scan_args(int argc, VALUE *argv, const char *fmt, ...) ::
+
+  Retrieve argument from argc and argv to given VALUE references
+  according to the format string.  The format can be described in ABNF
+  as follows:
+
+    scan-arg-spec  := param-arg-spec [option-hash-arg-spec] [block-arg-spec]
+
+    param-arg-spec := pre-arg-spec [post-arg-spec] / post-arg-spec /
+                      pre-opt-post-arg-spec
+    pre-arg-spec   := num-of-leading-mandatory-args [num-of-optional-args]
+    post-arg-spec  := sym-for-variable-length-args
+                      [num-of-trailing-mandatory-args]
+    pre-opt-post-arg-spec := num-of-leading-mandatory-args num-of-optional-args
+                             num-of-trailing-mandatory-args
+    option-hash-arg-spec := sym-for-option-hash-arg
+    block-arg-spec := sym-for-block-arg
+
+    num-of-leading-mandatory-args  := DIGIT ; The number of leading
+                                            ; mandatory arguments
+    num-of-optional-args           := DIGIT ; The number of optional
+                                            ; arguments
+    sym-for-variable-length-args   := "*"   ; Indicates that variable
+                                            ; length arguments are
+                                            ; captured as a ruby array
+    num-of-trailing-mandatory-args := DIGIT ; The number of trailing
+                                            ; mandatory arguments
+    sym-for-option-hash-arg        := ":"   ; Indicates that an option
+                                            ; hash is captured if the last
+                                            ; argument is a hash or can be
+                                            ; converted to a hash with
+                                            ; #to_hash.  When the last
+                                            ; argument is nil, it is
+                                            ; captured if it is not
+                                            ; ambiguous to take it as
+                                            ; empty option hash; i.e. '*'
+                                            ; is not specified and
+                                            ; arguments are given more
+                                            ; than sufficient.
+    sym-for-block-arg              := "&"   ; Indicates that an iterator
+                                            ; block should be captured if
+                                            ; given
+
+  For example, "12" means that the method requires at least one
+  argument, and at most receives three (1+2) arguments.  So, the format
+  string must be followed by three variable references, which are to be
+  assigned to captured arguments.  For omitted arguments, variables are
+  set to Qnil.  NULL can be put in place of a variable reference, which
+  means the corresponding captured argument(s) should be just dropped.
+
+  The number of given arguments, excluding an option hash or iterator
+  block, is returned.
+
+int rb_get_kwargs(VALUE keyword_hash, const ID *table, int required, int optional, VALUE *values) ::
+
+  Retrieves argument VALUEs bound to keywords, which directed by +table+
+  into +values+, deleting retrieved entries from +keyword_hash+ along
+  the way.  First +required+ number of IDs referred by +table+ are
+  mandatory, and succeeding +optional+ (- +optional+ - 1 if
+  +optional+ is negative) number of IDs are optional.  If a
+  mandatory key is not contained in +keyword_hash+, raises "missing
+  keyword" +ArgumentError+.  If an optional key is not present in
+  +keyword_hash+, the corresponding element in +values+ is set to +Qundef+.
+  If +optional+ is negative, rest of +keyword_hash+ are ignored, otherwise
+  raises "unknown keyword" +ArgumentError+.
+
+  Be warned, handling keyword arguments in the C API is less efficient
+  than handling them in Ruby.  Consider using a Ruby wrapper method
+  around a non-keyword C function.
+  ref: https://bugs.ruby-lang.org/issues/11339
+
+VALUE rb_extract_keywords(VALUE *original_hash) ::
+
+  Extracts pairs whose key is a symbol into a new hash from a hash
+  object referred by +original_hash+.  If the original hash contains
+  non-symbol keys, then they are copied to another hash and the new hash
+  is stored through +original_hash+, else 0 is stored.
+
+=== Invoking Ruby method
+
+VALUE rb_funcall(VALUE recv, ID mid, int narg, ...) ::
+
+  Invokes a method.  To retrieve mid from a method name, use rb_intern().
+  Able to call even private/protected methods.
+
+VALUE rb_funcall2(VALUE recv, ID mid, int argc, VALUE *argv) ::
+VALUE rb_funcallv(VALUE recv, ID mid, int argc, VALUE *argv) ::
+
+  Invokes a method, passing arguments as an array of values.
+  Able to call even private/protected methods.
+
+VALUE rb_funcallv_public(VALUE recv, ID mid, int argc, VALUE *argv) ::
+
+  Invokes a method, passing arguments as an array of values.
+  Able to call only public methods.
+
+VALUE rb_eval_string(const char *str) ::
+
+  Compiles and executes the string as a Ruby program.
+
+ID rb_intern(const char *name) ::
+
+  Returns ID corresponding to the name.
+
+char *rb_id2name(ID id) ::
+
+  Returns the name corresponding ID.
+
+char *rb_class2name(VALUE klass) ::
+
+  Returns the name of the class.
+
+int rb_respond_to(VALUE obj, ID id) ::
+
+  Returns true if the object responds to the message specified by id.
+
+=== Instance Variables
+
+VALUE rb_iv_get(VALUE obj, const char *name) ::
+
+  Retrieve the value of the instance variable.  If the name is not
+  prefixed by `@', that variable shall be inaccessible from Ruby.
+
+VALUE rb_iv_set(VALUE obj, const char *name, VALUE val) ::
+
+  Sets the value of the instance variable.
+
+=== Control Structure
+
+VALUE rb_block_call(VALUE recv, ID mid, int argc, VALUE * argv, VALUE (*func) (ANYARGS), VALUE data2) ::
+
+  Calls a method on the recv, with the method name specified by the
+  symbol mid, with argc arguments in argv, supplying func as the
+  block. When func is called as the block, it will receive the value
+  from yield as the first argument, and data2 as the second argument.
+  When yielded with multiple values (in C, rb_yield_values(),
+  rb_yield_values2() and rb_yield_splat()), data2 is packed as an Array,
+  whereas yielded values can be gotten via argc/argv of the third/fourth
+  arguments.
+
+\[OBSOLETE] VALUE rb_iterate(VALUE (*func1)(), VALUE arg1, VALUE (*func2)(), VALUE arg2) ::
+
+  Calls the function func1, supplying func2 as the block.  func1 will be
+  called with the argument arg1.  func2 receives the value from yield as
+  the first argument, arg2 as the second argument.
+
+  When rb_iterate is used in 1.9, func1 has to call some Ruby-level method.
+  This function is obsolete since 1.9; use rb_block_call instead.
+
+VALUE rb_yield(VALUE val) ::
+
+  Evaluates the block with value val.
+
+VALUE rb_rescue(VALUE (*func1)(ANYARGS), VALUE arg1, VALUE (*func2)(ANYARGS), VALUE arg2) ::
+
+  Calls the function func1, with arg1 as the argument.  If an exception
+  occurs during func1, it calls func2 with arg2 as the first argument
+  and the exception object as the second argument.  The return value
+  of rb_rescue() is the return value from func1 if no exception occurs,
+  from func2 otherwise.
+
+VALUE rb_ensure(VALUE (*func1)(ANYARGS), VALUE arg1, VALUE (*func2)(ANYARGS), VALUE arg2) ::
+
+  Calls the function func1 with arg1 as the argument, then calls func2
+  with arg2 if execution terminated.  The return value from
+  rb_ensure() is that of func1 when no exception occurred.
+
+VALUE rb_protect(VALUE (*func) (VALUE), VALUE arg, int *state) ::
+
+  Calls the function func with arg as the argument.  If no exception
+  occurred during func, it returns the result of func and *state is zero.
+  Otherwise, it returns Qnil and sets *state to nonzero.  If state is
+  NULL, it is not set in both cases.
+  You have to clear the error info with rb_set_errinfo(Qnil) when
+  ignoring the caught exception.
+
+void rb_jump_tag(int state) ::
+
+  Continues the exception caught by rb_protect() and rb_eval_string_protect().
+  state must be the returned value from those functions.  This function
+  never return to the caller.
+
+void rb_iter_break() ::
+
+  Exits from the current innermost block.  This function never return to
+  the caller.
+
+void rb_iter_break_value(VALUE value) ::
+
+  Exits from the current innermost block with the value.  The block will
+  return the given argument value.  This function never return to the
+  caller.
+
+=== Exceptions and Errors
+
+void rb_warn(const char *fmt, ...) ::
+
+  Prints a warning message according to a printf-like format.
+
+void rb_warning(const char *fmt, ...) ::
+
+  Prints a warning message according to a printf-like format, if
+  $VERBOSE is true.
+
+void rb_raise(rb_eRuntimeError, const char *fmt, ...) ::
+
+  Raises RuntimeError.  The fmt is a format string just like printf().
+
+void rb_raise(VALUE exception, const char *fmt, ...) ::
+
+  Raises a class exception.  The fmt is a format string just like printf().
+
+void rb_fatal(const char *fmt, ...) ::
+
+  Raises a fatal error, terminates the interpreter.  No exception handling
+  will be done for fatal errors, but ensure blocks will be executed.
+
+void rb_bug(const char *fmt, ...) ::
+
+  Terminates the interpreter immediately.  This function should be
+  called under the situation caused by the bug in the interpreter.  No
+  exception handling nor ensure execution will be done.
+
+Note: In the format string, "%"PRIsVALUE can be used for Object#to_s
+(or Object#inspect if '+' flag is set) output (and related argument
+must be a VALUE).  Since it conflicts with "%i", for integers in
+format strings, use "%d".
+
+=== Threading
+
+As of Ruby 1.9, Ruby supports native 1:1 threading with one kernel
+thread per Ruby Thread object.  Currently, there is a GVL (Global VM Lock)
+which prevents simultaneous execution of Ruby code which may be released
+by the rb_thread_call_without_gvl and rb_thread_call_without_gvl2 functions.
+These functions are tricky-to-use and documented in thread.c; do not
+use them before reading comments in thread.c.
+
+void rb_thread_schedule(void) ::
+
+  Give the scheduler a hint to pass execution to another thread.
+
+=== Input/Output (IO) on a single file descriptor
+
+int rb_io_wait_readable(int fd) ::
+
+  Wait indefinitely for the given FD to become readable, allowing other
+  threads to be scheduled.  Returns a true value if a read may be
+  performed, false if there is an unrecoverable error.
+
+int rb_io_wait_writable(int fd) ::
+
+  Like rb_io_wait_readable, but for writability.
+
+int rb_wait_for_single_fd(int fd, int events, struct timeval *timeout) ::
+
+  Allows waiting on a single FD for one or multiple events with a
+  specified timeout.
+
+  +events+ is a mask of any combination of the following values:
+
+  * RB_WAITFD_IN - wait for readability of normal data
+  * RB_WAITFD_OUT - wait for writability
+  * RB_WAITFD_PRI - wait for readability of urgent data
+
+  Use a NULL +timeout+ to wait indefinitely.
+
+=== I/O Multiplexing
+
+Ruby supports I/O multiplexing based on the select(2) system call.
+The Linux select_tut(2) manpage
+<http://man7.org/linux/man-pages/man2/select_tut.2.html>
+provides a good overview on how to use select(2), and the Ruby API has
+analogous functions and data structures to the well-known select API.
+Understanding of select(2) is required to understand this section.
+
+typedef struct rb_fdset_t ::
+
+  The data structure which wraps the fd_set bitmap used by select(2).
+  This allows Ruby to use FD sets larger than that allowed by
+  historic limitations on modern platforms.
+
+void rb_fd_init(rb_fdset_t *) ::
+
+  Initializes the rb_fdset_t, it must be initialized before other rb_fd_*
+  operations.  Analogous to calling malloc(3) to allocate an fd_set.
+
+void rb_fd_term(rb_fdset_t *) ::
+
+  Destroys the rb_fdset_t, releasing any memory and resources it used.
+  It must be reinitialized using rb_fd_init before future use.
+  Analogous to calling free(3) to release memory for an fd_set.
+
+void rb_fd_zero(rb_fdset_t *) ::
+
+  Clears all FDs from the rb_fdset_t, analogous to FD_ZERO(3).
+
+void rb_fd_set(int fd, rb_fdset_t *) ::
+
+  Adds a given FD in the rb_fdset_t, analogous to FD_SET(3).
+
+void rb_fd_clr(int fd, rb_fdset_t *) ::
+
+  Removes a given FD from the rb_fdset_t, analogous to FD_CLR(3).
+
+int rb_fd_isset(int fd, const rb_fdset_t *) ::
+
+  Returns true if a given FD is set in the rb_fdset_t, false if not.
+  Analogous to FD_ISSET(3).
+
+int rb_thread_fd_select(int nfds, rb_fdset_t *readfds, rb_fdset_t *writefds, rb_fdset_t *exceptfds, struct timeval *timeout) ::
+
+  Analogous to the select(2) system call, but allows other Ruby
+  threads to be scheduled while waiting.
+
+  When only waiting on a single FD, favor rb_io_wait_readable,
+  rb_io_wait_writable, or rb_wait_for_single_fd functions since
+  they can be optimized for specific platforms (currently, only Linux).
+
+=== Initialize and Start the Interpreter
+
+The embedding API functions are below (not needed for extension libraries):
+
+void ruby_init() ::
+
+  Initializes the interpreter.
+
+void *ruby_options(int argc, char **argv) ::
+
+  Process command line arguments for the interpreter.
+  And compiles the Ruby source to execute.
+  It returns an opaque pointer to the compiled source
+  or an internal special value.
+
+int ruby_run_node(void *n) ::
+
+  Runs the given compiled source and exits this process.
+  It returns EXIT_SUCCESS if successfully runs the source.
+  Otherwise, it returns other value.
+
+void ruby_script(char *name) ::
+
+  Specifies the name of the script ($0).
+
+=== Hooks for the Interpreter Events
+
+void rb_add_event_hook(rb_event_hook_func_t func, rb_event_flag_t events, VALUE data) ::
+
+  Adds a hook function for the specified interpreter events.
+  events should be OR'ed value of:
+
+    RUBY_EVENT_LINE
+    RUBY_EVENT_CLASS
+    RUBY_EVENT_END
+    RUBY_EVENT_CALL
+    RUBY_EVENT_RETURN
+    RUBY_EVENT_C_CALL
+    RUBY_EVENT_C_RETURN
+    RUBY_EVENT_RAISE
+    RUBY_EVENT_ALL
+
+  The definition of rb_event_hook_func_t is below:
+
+    typedef void (*rb_event_hook_func_t)(rb_event_t event, VALUE data,
+                                         VALUE self, ID id, VALUE klass)
+
+  The third argument `data' to rb_add_event_hook() is passed to the hook
+  function as the second argument, which was the pointer to the current
+  NODE in 1.8.  See RB_EVENT_HOOKS_HAVE_CALLBACK_DATA below.
+
+int rb_remove_event_hook(rb_event_hook_func_t func) ::
+
+  Removes the specified hook function.
+
+=== Memory usage
+
+void rb_gc_adjust_memory_usage(ssize_t diff) ::
+
+  Adjusts the amount of registered external memory.  You can tell GC how
+  much memory is used by an external library by this function.  Calling
+  this function with positive diff means the memory usage is increased;
+  new memory block is allocated or a block is reallocated as larger
+  size.  Calling this function with negative diff means the memory usage
+  is decreased; a memory block is freed or a block is reallocated as
+  smaller size.  This function may trigger the GC.
+
+=== Macros for Compatibility
+
+Some macros to check API compatibilities are available by default.
+
+NORETURN_STYLE_NEW ::
+
+  Means that NORETURN macro is functional style instead of prefix.
+
+HAVE_RB_DEFINE_ALLOC_FUNC ::
+
+  Means that function rb_define_alloc_func() is provided, that means the
+  allocation framework is used.  This is same as the result of
+  have_func("rb_define_alloc_func", "ruby.h").
+
+HAVE_RB_REG_NEW_STR ::
+
+  Means that function rb_reg_new_str() is provided, that creates Regexp
+  object from String object.  This is same as the result of
+  have_func("rb_reg_new_str", "ruby.h").
+
+HAVE_RB_IO_T ::
+
+  Means that type rb_io_t is provided.
+
+USE_SYMBOL_AS_METHOD_NAME ::
+
+  Means that Symbols will be returned as method names, e.g.,
+  Module#methods, \#singleton_methods and so on.
+
+HAVE_RUBY_*_H ::
+
+  Defined in ruby.h and means corresponding header is available.  For
+  instance, when HAVE_RUBY_ST_H is defined you should use ruby/st.h not
+  mere st.h.
+
+RB_EVENT_HOOKS_HAVE_CALLBACK_DATA ::
+
+  Means that rb_add_event_hook() takes the third argument `data', to be
+  passed to the given event hook function.
+
+== Appendix C. Functions available for use in extconf.rb
+
+See documentation for {mkmf}[rdoc-ref:MakeMakefile].
+
+== Appendix D. Generational GC
+
+Ruby 2.1 introduced a generational garbage collector (called RGenGC).
+RGenGC (mostly) keeps compatibility.
+
+Generally, the use of the technique called write barriers is required in
+extension libraries for generational GC
+(https://en.wikipedia.org/wiki/Garbage_collection_%28computer_science%29).
+RGenGC works fine without write barriers in extension libraries.
+
+If your library adheres to the following tips, performance can
+be further improved. Especially, the "Don't touch pointers directly" section is
+important.
+
+=== Incompatibility
+
+You can't write RBASIC(obj)->klass field directly because it is const
+value now.
+
+Basically you should not write this field because MRI expects it to be
+an immutable field, but if you want to do it in your extension you can
+use the following functions:
+
+VALUE rb_obj_hide(VALUE obj) ::
+
+  Clear RBasic::klass field. The object will be an internal object.
+  ObjectSpace::each_object can't find this object.
+
+VALUE rb_obj_reveal(VALUE obj, VALUE klass) ::
+
+  Reset RBasic::klass to be klass.
+  We expect the `klass' is hidden class by rb_obj_hide().
+
+=== Write barriers
+
+RGenGC doesn't require write barriers to support generational GC.
+However, caring about write barrier can improve the performance of
+RGenGC. Please check the following tips.
+
+==== Don't touch pointers directly
+
+In MRI (include/ruby/ruby.h), some macros to acquire pointers to the
+internal data structures are supported such as RARRAY_PTR(),
+RSTRUCT_PTR() and so on.
+
+DO NOT USE THESE MACROS and instead use the corresponding C-APIs such as
+rb_ary_aref(), rb_ary_store() and so on.
+
+==== Consider whether to insert write barriers
+
+You don't need to care about write barriers if you only use built-in
+types.
+
+If you support T_DATA objects, you may consider using write barriers.
+
+Inserting write barriers into T_DATA objects only works with the
+following type objects: (a) long-lived objects, (b) when a huge number
+of objects are generated and \(c) container-type objects that have
+references to other objects. If your extension provides such a type of
+T_DATA objects, consider inserting write barriers.
+
+(a): short-lived objects don't become old generation objects.
+(b): only a few oldgen objects don't have performance impact.
+\(c): only a few references don't have performance impact.
+
+Inserting write barriers is a very difficult hack, it is easy to
+introduce critical bugs. And inserting write barriers has several areas
+of overhead. Basically we don't recommend you insert write barriers.
+Please carefully consider the risks.
+
+==== Combine with built-in types
+
+Please consider utilizing built-in types. Most built-in types support
+write barrier, so you can use them to avoid manually inserting write
+barriers.
+
+For example, if your T_DATA has references to other objects, then you
+can move these references to Array. A T_DATA object only has a reference
+to an array object. Or you can also use a Struct object to gather a
+T_DATA object (without any references) and an that Array contains
+references.
+
+With use of such techniques, you don't need to insert write barriers
+anymore.
+
+==== Insert write barriers
+
+\[AGAIN] Inserting write barriers is a very difficult hack, and it is
+easy to introduce critical bugs. And inserting write barriers has
+several areas of overhead. Basically we don't recommend you insert write
+barriers. Please carefully consider the risks.
+
+Before inserting write barriers, you need to know about RGenGC algorithm
+(gc.c will help you). Macros and functions to insert write barriers are
+available in include/ruby/ruby.h. An example is available in iseq.c.
+
+For a complete guide for RGenGC and write barriers, please refer to
+<https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/RGenGC>.
+
+== Appendix E. RB_GC_GUARD to protect from premature GC
+
+C Ruby currently uses conservative garbage collection, thus VALUE
+variables must remain visible on the stack or registers to ensure any
+associated data remains usable.  Optimizing C compilers are not designed
+with conservative garbage collection in mind, so they may optimize away
+the original VALUE even if the code depends on data associated with that
+VALUE.
+
+The following example illustrates the use of RB_GC_GUARD to ensure
+the contents of sptr remain valid while the second invocation of
+rb_str_new_cstr is running.
+
+  VALUE s, w;
+  const char *sptr;
+
+  s = rb_str_new_cstr("hello world!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!");
+  sptr = RSTRING_PTR(s);
+  w = rb_str_new_cstr(sptr + 6); /* Possible GC invocation */
+
+  RB_GC_GUARD(s); /* ensure s (and thus sptr) do not get GC-ed */
+
+In the above example, RB_GC_GUARD must be placed _after_ the last use of
+sptr.  Placing RB_GC_GUARD before dereferencing sptr would be of no use.
+RB_GC_GUARD is only effective on the VALUE data type, not converted C
+data types.
+
+RB_GC_GUARD would not be necessary at all in the above example if
+non-inlined function calls are made on the `s' VALUE after sptr is
+dereferenced.  Thus, in the above example, calling any un-inlined
+function on `s' such as:
+
+  rb_str_modify(s);
+
+Will ensure `s' stays on the stack or register to prevent a
+GC invocation from prematurely freeing it.
+
+Using the RB_GC_GUARD macro is preferable to using the "volatile"
+keyword in C.  RB_GC_GUARD has the following advantages:
+
+1. the intent of the macro use is clear
+
+2. RB_GC_GUARD only affects its call site, "volatile" generates some
+   extra code every time the variable is used, hurting optimization.
+
+3. "volatile" implementations may be buggy/inconsistent in some
+   compilers and architectures. RB_GC_GUARD is customizable for broken
+   systems/compilers without negatively affecting other systems.
+
+:enddoc: Local variables:
+:enddoc: fill-column: 70
+:enddoc: end:
diff --git a/doc/forwardable.rd b/doc/forwardable.rd
deleted file mode 100644
index f9c8234761..0000000000
--- a/doc/forwardable.rd
+++ /dev/null
@@ -1,84 +0,0 @@
- -- forwardable.rb
-						
-                                                $Release Version: 1.1 $
-                                                $Revision: 1.2 $
-                                                $Date: 2001/05/07 23:52:57 $
-						Original version by Tosh
-
-=begin
-
-= Forwardable
-
-A Module to define delegations for selected methods to a class.
-
-== Usage
-
-Using through extending the class.
-  
-  class Foo
-    extend Forwardable
-
-    def_delegators("@out", "printf", "print")
-    def_delegators(:@in, :gets)
-    def_delegator(:@contents, :[], "content_at")
-  end
-  f = Foo.new
-  f.printf ...
-  f.gets
-  f.content_at(1)
-
-== Methods
-
---- Forwardable#def_instance_delegators(accessor, *methods)
-
-      adding the delegations for each method of ((|methods|)) to
-      ((|accessor|)).
-
---- Forwardable#def_instance_delegator(accessor, method, ali = method)
-      
-      adding the delegation for ((|method|)) to ((|accessor|)). When
-      you give optional argument ((|ali|)), ((|ali|)) is used as the
-      name of the delegation method, instead of ((|method|)).
-
---- Forwardable#def_delegators(accessor, *methods)
-
-      the alias of ((|Forwardable#def_instance_delegators|)).
-
---- Forwardable#def_delegator(accessor, method, ali = method)
-      
-      the alias of ((|Forwardable#def_instance_delegator|)).
-
-= SingleForwardable
-
-a Module to define delegations for selected methods to an object.
-
-== Usage
-
-Using through extending the object.
-
-  g = Goo.new
-  g.extend SingleForwardable
-  g.def_delegator("@out", :puts)
-  g.puts ...
-
-== Methods
-
---- SingleForwardable#def_singleton_delegators(accessor, *methods)
-
-      adding the delegations for each method of ((|methods|)) to
-      ((|accessor|)).
-
---- SingleForwardable#def_singleton_delegator(accessor, method, ali = method)
-
-      adding the delegation for ((|method|)) to ((|accessor|)). When
-      you give optional argument ((|ali|)), ((|ali|)) is used as the
-      name of the delegation method, instead of ((|method|)).
-
---- SingleForwardable#def_delegators(accessor, *methods)
-
-      the alias of ((|SingleForwardable#def_instance_delegators|)).
-
---- SingleForwardable#def_delegator(accessor, method, ali = method)
-
-      the alias of ((|SingleForwardable#def_instance_delegator|)).
-=end
diff --git a/doc/forwardable.rd.ja b/doc/forwardable.rd.ja
index a56823dd1a..171724b2e5 100644
--- a/doc/forwardable.rd.ja
+++ b/doc/forwardable.rd.ja
@@ -1,17 +1,16 @@
   -- forwatable.rb
                                                 $Release Version: 1.1 $
-                                                $Revision: 1.1 $
-                                                $Date: 2001/07/19 05:42:06 $
+                                                $Revision$
 
 =begin
 = Forwardable
 
-���饹���Ф��᥽�åɤΰѾ���ǽ��������ޤ�.
+クラスに対しメソッドの委譲機能を定義します.
 
-== �Ȥ���
+== 使い方
+
+クラスに対してextendして使います.
 
-���饹���Ф���extend���ƻȤ��ޤ�. 
-  
   class Foo
     extend Forwardable
 
@@ -24,58 +23,58 @@
   f.gets
   f.content_at(1)
 
-== �᥽�å�
+== メソッド
 
 --- Forwardable#def_instance_delegators(accessor, *methods)
 
-      ((|methods|))���Ϥ��줿�᥽�åɤΥꥹ�Ȥ�((|accessor��|))�Ѿ�����
-      �褦�ˤ��ޤ�.
+      ((|methods|))で渡されたメソッドのリストを((|accessorに|))委譲する
+      ようにします.
 
 --- Forwardable#def_instance_delegator(accessor, method, ali = method)
 
-      ((||method|))���Ϥ��줿�᥽�åɤ�((|accessor|))�˰Ѿ�����褦�ˤ�
-      �ޤ�. ((|ali|))�������Ȥ����Ϥ��줿�Ȥ���, �᥽�å�((|ali|))���Ƥ�
-      �줿�Ȥ��ˤ�, ((|accessor|))���Ф�((|method|))��ƤӽФ��ޤ�.
+      ((||method|))で渡されたメソッドを((|accessor|))に委譲するようにし
+      ます. ((|ali|))が引数として渡されたときは, メソッド((|ali|))が呼ば
+      れたときには, ((|accessor|))に対し((|method|))を呼び出します.
 
 --- Forwardable#def_delegators(accessor, *methods)
 
-      ((|Forwardable#def_instance_delegators|))����̾�Ǥ�.
+      ((|Forwardable#def_instance_delegators|))の別名です.
 
 --- Forwardable#def_delegator(accessor, method, ali = method)
 
-      ((|Forwardable#def_instance_delegator|))����̾�Ǥ�.
+      ((|Forwardable#def_instance_delegator|))の別名です.
 
 = SingleForwardable
 
-���֥������Ȥ��Ф�, �᥽�åɤΰѾ���ǽ��������ޤ�.
+オブジェクトに対し, メソッドの委譲機能を定義します.
 
-== �Ȥ���
+== 使い方
 
-���֥������Ȥ��Ф���((|extend|))���ƻȤ��ޤ�. 
+オブジェクトに対して((|extend|))して使います.
 
   g = Goo.new
   g.extend SingleForwardable
   g.def_delegator("@out", :puts)
   g.puts ...
 
-== �᥽�å�
+== メソッド
 
 --- SingleForwardable#def_singleton_delegators(accessor, *methods)
 
-      ((|methods|))���Ϥ��줿�᥽�åɤΥꥹ�Ȥ�((|accessor|))�˰Ѿ�����
-      �褦�ˤ��ޤ�.
+      ((|methods|))で渡されたメソッドのリストを((|accessor|))に委譲する
+      ようにします.
 
 --- SingleForwardable#def_singleton_delegator(accessor, method, ali = method)
 
-      ((|method|))���Ϥ��줿�᥽�åɤ�((|accessor|))�˰Ѿ�����褦�ˤ���
-      ��. ((|ali|))�������Ȥ����Ϥ��줿�Ȥ���, �᥽�å�((|ali|))���ƤФ�
-      ���Ȥ��ˤ�, ((|accessor|))���Ф�((|method|))��ƤӽФ��ޤ�.
+      ((|method|))で渡されたメソッドを((|accessor|))に委譲するようにしま
+      す. ((|ali|))が引数として渡されたときは, メソッド((|ali|))が呼ばれ
+      たときには, ((|accessor|))に対し((|method|))を呼び出します.
 
 --- SingleForwardable#def_delegators(accessor, *methods)
 
-      ((|SingleForwardable#def_singleton_delegators|))����̾�Ǥ�.
+      ((|SingleForwardable#def_singleton_delegators|))の別名です.
 
 --- SingleForwardable#def_delegator(accessor, method, ali = method)
 
-      ((|SingleForwardable#def_singleton_delegator|))����̾�Ǥ�.
+      ((|SingleForwardable#def_singleton_delegator|))の別名です.
 =end
diff --git a/doc/globals.rdoc b/doc/globals.rdoc
new file mode 100644
index 0000000000..1e70555988
--- /dev/null
+++ b/doc/globals.rdoc
@@ -0,0 +1,70 @@
+# -*- mode: rdoc; coding: utf-8; fill-column: 74; -*-
+
+== Pre-defined variables
+
+$!::   The exception information message set by 'raise'.
+$@::   Array of backtrace of the last exception thrown.
+$&::   The string matched by the last successful match.
+$`::   The string to the left  of the last successful match.
+$'::   The string to the right of the last successful match.
+$+::   The highest group matched by the last successful match.
+$1::   The Nth group of the last successful match. May be > 1.
+$~::   The information about the last match in the current scope.
+$=::   This variable is no longer effective. Deprecated.
+$/::   The input record separator, newline by default.
+$\::   The output record separator for the print and IO#write. Default is nil.
+$,::   The output field separator for the print and Array#join.
+$;::   The default separator for String#split.
+$.::   The current input line number of the last file that was read.
+$<::   The virtual concatenation file of the files given on command line (or from $stdin if no files were given).
+$>::   The default output for print, printf. $stdout by default.
+$_::   The last input line of string by gets or readline.
+$0::   Contains the name of the script being executed. May be assignable.
+$*::   Command line arguments given for the script sans args.
+$$::   The process number of the Ruby running this script.
+$?::   The status of the last executed child process.  This value is
+       thread-local.
+$:::   Load path for scripts and binary modules by load or require.
+$"::   The array contains the module names loaded by require.
+$DEBUG::     The debug flag, which is set by the -d switch.  Enabling debug
+             output prints each exception raised to $stderr (but not its
+             backtrace).  Setting this to a true value enables debug output as
+             if -d were given on the command line.  Setting this to a false
+             value disables debug output.
+$LOADED_FEATURES:: The alias to the $".
+$FILENAME::  Current input file from $<. Same as $<.filename.
+$LOAD_PATH:: The alias to the $:.
+$stderr::    The current standard error output.
+$stdin::     The current standard input.
+$stdout::    The current standard output.
+$VERBOSE::   The verbose flag, which is set by the -w or -v switch.  Setting
+             this to a true value enables warnings as if -w or -v were given
+             on the command line.  Setting this to nil disables warnings,
+             including from Kernel#warn.
+$-0::  The alias to $/.
+$-a::  True if option -a is set. Read-only variable.
+$-d::  The alias of $DEBUG.  See $DEBUG above for further discussion.
+$-F::  The alias to $;.
+$-i::  In in-place-edit mode, this variable holds the extension, otherwise nil.
+$-I::  The alias to $:.
+$-l::  True if option -l is set. Read-only variable.
+$-p::  True if option -p is set. Read-only variable.
+$-v::  An alias of $VERBOSE.  See $VERBOSE above for further discussion.
+$-w::  An alias of $VERBOSE.  See $VERBOSE above for further discussion.
+
+== Pre-defined global constants
+
+TRUE::              The typical true value.
+FALSE::             The false itself.
+NIL::               The nil itself.
+STDIN::             The standard input. The default value for $stdin.
+STDOUT::            The standard output. The default value for $stdout.
+STDERR::            The standard error output. The default value for $stderr.
+ENV::               The hash contains current environment variables.
+ARGF::              The alias to the $<.
+ARGV::              The alias to the $*.
+DATA::              The file object of the script, pointing just after __END__.
+RUBY_VERSION::      The ruby version string (VERSION was deprecated).
+RUBY_RELEASE_DATE:: The release date string.
+RUBY_PLATFORM::     The platform identifier.
+
diff --git a/doc/images/boottime-classes.png b/doc/images/boottime-classes.png
new file mode 100644
index 0000000000..b5fab75c41
--- /dev/null
+++ b/doc/images/boottime-classes.png
diff --git a/doc/irb/irb-tools.rd.ja b/doc/irb/irb-tools.rd.ja
index 38145576dc..b997f0edea 100644
--- a/doc/irb/irb-tools.rd.ja
+++ b/doc/irb/irb-tools.rd.ja
@@ -1,71 +1,70 @@
-irb��Ϣ���ޤ����ޥ�ɤȥ饤�֥��
+irb関連おまけコマンドとライブラリ
 				$Release Version: 0.7.1 $
-			   	$Revision: 1.1 $
-			   	$Date: 2001/07/19 05:42:06 $
+			   	$Revision$
 			   	by Keiju ISHITSUKA(Nihon Rational Co.,Ltd.)
 
 =begin
 
-:���ޥ��:
+:コマンド:
 * rtags		-- ruby tags command
 
-:�ؿ��饤�֥��:
+:関数ライブラリ:
 * xmp		-- irb version of gotoken xmp-function
 
-:���饹�饤�֥��:
+:クラスライブラリ:
 * frame.rb      -- frame tracer
 * completion.rb -- irb completor
 
 = rtags
 
-rtags��emacs�ڤ�vi�Ѥ�, TAG�ե������Ĥ��륳�ޥ�ɤǤ�.
+rtagsはemacs及びvi用の, TAGファイルをつくるコマンドです.
 
-== �Ȥ���
+== 使い方
 
    rtags [-vi] file....
 
-�����ȥǥ��쥯�ȥ��emacs�Ѥ�TAGS�ե����뤬�Ǥ��ޤ�. -vi���ץ�����
-�Ĥ������ˤ�vi�Ѥ�tags�ե������������ޤ�.
+カレントディレクトリにemacs用のTAGSファイルができます. -viオプションを
+つけた時にはvi用のtagsファイルを作成します.
 
-emacs�ξ��, �̾��etags.el�����Τޤ޻Ȥ��ޤ�. ������ǽ�ʤΤ�,
+emacsの場合, 通常のetags.elがそのまま使えます. 検索可能なのは,
 
-* ���饹
-* �᥽�å�
-* �ðۥ᥽�å�
+* クラス
+* メソッド
+* 特異メソッド
 * alias
-* attr��������줿��������(�ѥ�᡼��������ܥ뤫ʸ�����ƥ��˸¤�)
-* attr_XXX��������줿��������(�ѥ�᡼��������ܥ뤫ʸ�����ƥ��˸¤�)
+* attrで宣言されたアクセサ(パラメータがシンボルか文字列リテラルに限る)
+* attr_XXXで宣言されたアクセサ(パラメータがシンボルか文字列リテラルに限る)
 
-�Ǥ�.
+です.
 
-C�ʤɤǻȤäƤ���ΤȰ㤦�Τ�, ����ץ꡼�����˴ؤ�����ʬ��,
+Cなどで使っているのと違うのは, コンプリーションに関する部分で,
 
-�ؿ�̾��,
+関数名は,
 
-  �ؿ�̾(
+  関数名(
 
-���饹��,
+クラスは,
 
-  ::���饹̾::....::���饹̾
+  ::クラス名::....::クラス名
 
-�᥽�åɤ�,
+メソッドは,
 
-  ::���饹̾::....::���饹̾#�᥽�å�̾
+  ::クラス名::....::クラス名#メソッド名
 
-�ðۥ᥽�å�(���饹�᥽�å�)��
+特異メソッド(クラスメソッド)は
 
-  ::���饹̾::....::���饹̾.�᥽�å�̾
+  ::クラス名::....::クラス名.メソッド名
 
-�ǥ���ץ꡼������Ԥʤ��Ȥ����Ǥ�.
+でコンプリーションを行なうところです.
 
 = xmp.rb
 
-���Ȥ���xmp�ξ�̸ߴ��С������Ǥ�. ����, ���˽Ť��ΤǤ��Ȥ���xmp��
-���б��Ǥ��ʤ�����, ���Ѥ�����ɤ��Ǥ��礦.
+ごとけんxmpの上位互換バージョンです. ただ, 非常に重いのでごとけんxmpで
+は対応できない時に, 使用すると良いでしょう.
 
-== �Ȥ���
+== 使い方
 
-=== �ؿ��Ȥ��ƻȤ�.
+=== 関数として使う.
 
     require "irb/xmp"
     xmp <<END
@@ -78,10 +77,10 @@ C�ʤɤǻȤäƤ���ΤȰ㤦�Τ�, ����ץ꡼�����˴ؤ�����ʬ��,
     foo
 	==>1
 
-=== XMP���󥹥��󥹤��Ѥ���.
+=== XMPインスタンスを用いる.
 
-���ξ���, XMP������ƥ����Ⱦ������ĤΤ�, �ѿ����ͤʤɤ��ݻ����Ƥ�
-�ޤ�.
+この場合は, XMPがコンテキスト情報を持つので, 変数の値などを保持してい
+ます.
 
   require "irb/xmp"
   xmp = XMP.new
@@ -100,52 +99,52 @@ C�ʤɤǻȤäƤ���ΤȰ㤦�Τ�, ����ץ꡼�����˴ؤ�����ʬ��,
   foo
       ==>1
 
-== ����ƥ����Ȥ˴ؤ���
+== コンテキストに関して
 
-XMP�᥽�åɷ��Υ���ƥ����Ȥ�, �ƤӽФ����Υ���ƥ����Ȥ�ɾ������ޤ�.
-����Ū�˥���ƥ����Ȥ���ꤹ��Ȥ��Υ���ƥ����Ȥ�ɾ�����ޤ�.
+XMPメソッド群のコンテキストは, 呼び出す前のコンテキストで評価されます.
+明示的にコンテキストを指定するとそのコンテキストで評価します.
 
-��:
+例:
 
   xmp "foo", an_binding
 
-:��:
-�ޥ������åɤˤ��б����Ƥ��ޤ���.
+:注:
+マルチスレッドには対応していません.
 
 = frame.rb
-���߼¹���Υե졼�������갷������Υ��饹�Ǥ�. 
+現在実行中のフレーム情報を取り扱うためのクラスです.
 
 * IRB::Frame.top(n = 0)
-  �夫��n���ܤΥ���ƥ����Ȥ���Ф��ޤ�. n��0���Ǿ�̤ˤʤ�ޤ�.
+  上からn番目のコンテキストを取り出します. nは0が最上位になります.
 * IRB::Frame.bottom(n = 0)
-  ������n���ܤΥ���ƥ����Ȥ���Ф��ޤ�. n��0���ǲ��̤ˤʤ�ޤ�.
+  下からn番目のコンテキストを取り出します. nは0が最下位になります.
 * IRB::Frame.sender
-  ������ˤʤäƤ��륪�֥������Ȥ���Ф��ޤ�. ������Ȥ�, ���Υ᥽��
-  �ɤ�ƤӽФ���¦��self�Τ��ȤǤ�.
+  センダになっているオブジェクトを取り出します. センダとは, そのメソッ
+  ドを呼び出した側のselfのことです.
 
-:��:
-set_trace_func���Ѥ���Ruby�μ¹Ԥ�ȥ졼�����Ƥ��ޤ�. �ޥ������åɤ�
-���б����Ƥ��ޤ���.
+:注:
+set_trace_funcを用いてRubyの実行をトレースしています. マルチスレッドに
+は対応していません.
 
 = completion.rb
-irb��completion��ǽ���󶡤����ΤǤ�. 
+irbのcompletion機能を提供するものです.
 
-== �Ȥ���
+== 使い方
 
    % irb -r irb/completion
 
-�Ȥ��뤫, ~/.irbrc ���
+とするか, ~/.irbrc 中に
 
    require "irb/completion"
 
-������Ƥ�������. irb�¹���� require "irb/completion" ���Ƥ�褤�Ǥ�.
+を入れてください. irb実行中に require "irb/completion" してもよいです.
 
-irb�¹���� (TAB) �򲡤��ȥ���ץ졼����󤷤ޤ�.
+irb実行中に (TAB) を押すとコンプレーションします.
 
-�ȥåץ�٥��(TAB)�򲡤��Ȥ��٤Ƥι�ʸ����, ���饹, �᥽�åɤθ��䤬��
-�ޤ�. ���䤬ͣ��ʤ�д������䴰���ޤ�.
+トップレベルで(TAB)を押すとすべての構文要素, クラス, メソッドの候補がで
+ます. 候補が唯一ならば完全に補完します.
 
-  irb(main):001:0> in    
+  irb(main):001:0> in
   in                    inspect               instance_eval
   include               install_alias_method  instance_of?
   initialize            install_aliases       instance_variables
@@ -154,8 +153,8 @@ irb�¹���� (TAB) �򲡤��ȥ���ץ졼����󤷤ޤ�.
   irb(main):002:0> foo = Object.new
   #<Object:0x4027146c>
 
-  ((|�ѿ�̾.|))�θ��(TAB)�򲡤���, ���Υ��֥������ȤΥ᥽�åɰ������Ǥ�
-  ��.
+  ((|変数名.|))の後に(TAB)を押すと, そのオブジェクトのメソッド一覧がでま
+  す.
 
   irb(main):003:0> foo.
   foo.==                  foo.frozen?             foo.protected_methods
@@ -169,8 +168,8 @@ irb�¹���� (TAB) �򲡤��ȥ���ץ졼����󤷤ޤ�.
   foo.dup                 foo.kind_of?            foo.to_s
   foo.eql?                foo.method              foo.type
   foo.equal?              foo.methods             foo.untaint
-  foo.extend              foo.nil?                
-  foo.freeze              foo.private_methods     
+  foo.extend              foo.nil?
+  foo.freeze              foo.private_methods
 
 =end
 
diff --git a/doc/irb/irb.rd b/doc/irb/irb.rd
deleted file mode 100644
index f56e8b60e0..0000000000
--- a/doc/irb/irb.rd
+++ /dev/null
@@ -1,392 +0,0 @@
-irb -- interactive ruby
-		$Release Version: 0.9 $
-		$Revision: 1.6 $
-		$Date: 2003/07/31 16:34:07 $
-		by Keiju ISHITSUKA(keiju@ishitsuka.com)
-		by gotoken-san who is original translater from japanese version
-
-=begin
-= What is irb?
-
-irb stands for `interactive ruby'. irb is a tool to execute interactively
-ruby expressions read from stdin. 
-
-= Invoking
-
-  % irb
-
-= Usage
-
-Use of irb is easy if you know ruby.  Executing irb, prompts are 
-displayed as follows. Then, enter expression of ruby. A input is
-executed when it is syntacticaly completed. 
-
-  dim% irb
-  irb(main):001:0> 1+2
-  3
-  irb(main):002:0> class Foo
-  irb(main):003:1>  def foo
-  irb(main):004:2>    print 1
-  irb(main):005:2>  end
-  irb(main):006:1> end
-  nil
-  irb(main):007:0> 
-
-And, Readline extesion module can be used with irb. Using Readline
-is the standard default action if Readline is installed. 
-
-= Command line option
-
-  irb.rb [options] file_name opts
-  options:
-  -f		    suppress read ~/.irbrc 
-  -m		    bc mode (fraction or matrix are available)
-  -d                set $DEBUG  to true (same as `ruby -d')
-  -Kc		    same as `ruby -Kc'
-  -r load-module    same as `ruby -r'
-  --verbose	    command input is echoed(default)
-  --noverbose	    command input isn't echoed
-  --echo	    commands are echoed immediately before execution(default)
-  --noecho	    commands aren't echoed immediately before execution
-  --inspect	    uses `inspect' for output (the default except bc mode)
-  --noinspect	    doesn't uses inspect for output
-  --readline	    uses Readline extension module
-  --noreadline	    doesn't use Readline extension module
-  --prompt prompt-mode
-  --prompt-mode prompt-mode
-		    switches prompt mode. Pre-defined prompt modes are
-		    `default', `simple', `xmp' and `inf-ruby'
-			    
-  --inf-ruby-mode   uses prompt appreciate for inf-ruby-mode on emacs. 
-		    Suppresses --readline. 
-  --simple-prompt   simple prompt mode
-  --noprompt	    no prompt
-  --tracer	    display trace for each execution of commands.
-  --back-trace-limit n
-		    displayes backtrace top n and tail n. The default
-		    value is 16. 
-  --irb_debug n	    sets internal debug level to n (It shouldn't be used)
-  -v, --version	    prints the version of irb
-
-= Configurations
-
-irb reads `~/.irbrc' when it is invoked. If `~/.irbrb' doesn't exist
-irb try to read in the order `.irbrc', `irb.rc', `_irbrc' then `$irbrc'. 
-
-The following is altanative to the command line option. To use them
-type as follows in an irb session. 
-
-  IRB.conf[:IRB_NAME]="irb"
-  IRB.conf[:MATH_MODE]=false
-  IRB.conf[:USE_TRACER]=false
-  IRB.conf[:USE_LOADER]=false
-  IRB.conf[:IGNORE_SIGINT]=true
-  IRB.conf[:IGNORE_EOF]=false
-  IRB.conf[:INSPECT_MODE]=nil
-  IRB.conf[:IRB_RC] = nil
-  IRB.conf[:BACK_TRACE_LIMIT]=16
-  IRB.conf[:USE_LOADER] = false
-  IRB.conf[:USE_READLINE] = nil
-  IRB.conf[:USE_TRACER] = false
-  IRB.conf[:IGNORE_SIGINT] = true
-  IRB.conf[:IGNORE_EOF] = false
-  IRB.conf[:PROMPT_MODE] = :DEFALUT
-  IRB.conf[:PROMPT] = {...}
-  IRB.conf[:DEBUG_LEVEL]=0
-  IRB.conf[:VERBOSE]=true
-
-== Customizing prompt
-
-To costomize the prompt you set a variable
-
-  IRB.conf[:PROMPT]
-
-For example, describe as follows in `.irbrc'. 
-
-  IRB.conf[:PROMPT][:MY_PROMPT] = { # name of prompt mode
-    :PROMPT_I => nil,		  # normal prompt
-    :PROMPT_S => nil,		  # prompt for continuated strings
-    :PROMPT_C => nil,		  # prompt for continuated statement
-    :RETURN => "    ==>%s\n"	  # format to return value
-  }
-
-Then, invoke irb with the above prompt mode by
-
-  % irb --prompt my-prompt
-
-Or add the following in `.irbrc'. 
-
-  IRB.conf[:PROMPT_MODE] = :MY_PROMPT
-
-Constants PROMPT_I, PROMPT_S and PROMPT_C specifies the format. 
-In the prompt specification, some special strings are available. 
-
-  %N	command name which is running
-  %m	to_s of main object (self)
-  %M	inspect of main object (self)
-  %l	type of string(", ', /, ]), `]' is inner %w[...]
-  %NNi	indent level. NN is degits and means as same as printf("%NNd"). 
-        It can be ommited
-  %NNn	line number. 
-  %%    %
-
-For instance, the default prompt mode is defined as follows:
-
-IRB.conf[:PROMPT_MODE][:DEFAULT] = {
-      :PROMPT_I => "%N(%m):%03n:%i> ",
-      :PROMPT_S => "%N(%m):%03n:%i%l ",
-      :PROMPT_C => "%N(%m):%03n:%i* ",
-      :RETURN => "%s\n"
-} 
-
-RETURN is used to printf. 
-
-== Configurating subirb
-
-The command line option or IRB.conf specify the default behavior of
-(sub)irb. On the other hand, each conf of in the next sction `6. Command' 
-is used to individually configurate (sub)irb. 
-
-If proc is set to IRB.conf[:IRB_RC], its subirb will be invoked after
-execution of that proc under giving the context of irb as its
-aregument. By this mechanism each subirb can be configurated. 
-
-= Command
-
-For irb commands, both simple name and `irb_'-prefixed name are prepared. 
-
---- exit, quit, irb_exit	
-    Quits (sub)irb. 
-
---- conf, irb_context
-    Displays current configuration. Modifing the configuration is
-    achieved by sending message to `conf'. 
-
---- conf.eval_history = N
-    Sets execution result history.
-    N is a integer or nil. If N > 0, the number of historys is N. 
-    If N == 0, the number of historys is unlimited. If N is nill,
-    execution result history isn't used(default).
-
---- conf.back_trace_limit
-    Sets display lines of backtrace as top n and tail n. 
-    The default value is 16.
-    
---- conf.debug_level = N
-    Sets debug level of irb. 
-
---- conf.ignore_eof = true/false
-    Whether ^D (control-d) will be ignored or not. 
-    If false is set, ^D means quit. 
-
---- conf.ignore_sigint= true/false
-    Whether ^C (control-c) will be ignored or not. 
-    If false is set, ^D means quit.  If true, 
-      during input:   cancel inputing then return to top level. 
-      during execute: abondon current execution. 
-
---- conf.inf_ruby_mode = true/false
-    Whether inf-ruby-mode or not. The default value is false.
-
---- conf.inspect_mode = true/false/nil
-    Specifies inspect mode. 
-    true:  display inspect
-    false: display to_s
-    nil:   inspect mode in non math mode, 
-           non inspect mode in math mode. 
-
---- conf.math_mode
-    Whether bc mode or not. 
-
---- conf.use_loader = true/false
-    Whether irb's own file reader method is used when load/require or not. 
-    This mode is globaly affected (irb wide). 
-
---- conf.prompt_c
-    prompt for a continuating statement (e.g, immediately after of `if')
-
---- conf.prompt_i
-    standard prompt
-
---- conf.prompt_s
-    prompt for a continuating string
-
---- conf.rc
-    Whether ~/.irbrc is read or not. 
-
---- conf.use_prompt = true/false
-    Prompting or not. 
-
---- conf.use_readline = true/false/nil
-    Whether readline is used or not. 
-    true: uses 
-    false: doen't use
-    nil: intends to use readline except for inf-ruby-mode (default)
-#
-#--- conf.verbose=T/F
-#    Whether verbose messages are display or not. 
-
---- cws, chws, irb_change_workspace [obj]
-    obj will be self. If obj is omitted, self will be home-object, or
-    the main object of first started irb.
-
---- pushws, irb_pushws, irb_push_workspace [obj]
-    same as UNIX-shell command pushd.
-
---- popws, irb_popws, irb_pop_workspace
-    same as UNIX-shell command popd
-
---- irb [obj]
-    Invoke subirb. If obj is given, obj will be self. 
-
---- jobs, irb_jobs
-    List of subirb
-
---- fg n, irb_fg n
-    Switch into specified subirb. The following is candidates of n:
-
-      irb number
-      thhread
-      irb object
-      self(obj which is specified of irb obj)
-
---- kill n, irb_kill n
-    Kill subirb. The means of n is as same as the case of irb_fg. 
-
---- souce, irb_source  path
-    This is a like UNIX-shell command source. evaluate script in path
-    on current context.
-
---- irb_load path, prev
-    irb-version of Ruby's load.
-
-= System variable
-
---- _  The latest value of evaluation (it is local)
---- __ The history of evaluation values.
-    __[line_no] return an evaluation value of line number<line_no>. If
-    line_no is a negative, return value before -<line_no> from latest
-    value.
-
-= Session Example
-
-  dim% ruby irb.rb
-  irb(main):001:0> irb                        # invoke subirb
-  irb#1(main):001:0> jobs                     # list of subirbs
-  #0->irb on main (#<Thread:0x400fb7e4> : stop)
-  #1->irb#1 on main (#<Thread:0x40125d64> : running)
-  nil
-  irb#1(main):002:0> fg 0                     # switch job
-  nil
-  irb(main):002:0> class Foo;end
-  nil
-  irb(main):003:0> irb Foo                    # invoke subirb which has the 
-					      #              context of Foo
-  irb#2(Foo):001:0> def foo                   # define Foo#foo
-  irb#2(Foo):002:1>   print 1
-  irb#2(Foo):003:1> end
-  nil
-  irb#2(Foo):004:0> fg 0                      # switch job
-  nil
-  irb(main):004:0> jobs                       # list of job
-  #0->irb on main (#<Thread:0x400fb7e4> : running)
-  #1->irb#1 on main (#<Thread:0x40125d64> : stop)
-  #2->irb#2 on Foo (#<Thread:0x4011d54c> : stop)
-  nil
-  irb(main):005:0> Foo.instance_methods       # Foo#foo is defined asurely
-  ["foo"]
-  irb(main):006:0> fg 2                       # switch job
-  nil
-  irb#2(Foo):005:0> def bar                   # define Foo#bar
-  irb#2(Foo):006:1>  print "bar"
-  irb#2(Foo):007:1> end
-  nil
-  irb#2(Foo):010:0>  Foo.instance_methods
-  ["bar", "foo"]
-  irb#2(Foo):011:0> fg 0                      
-  nil
-  irb(main):007:0> f = Foo.new
-  #<Foo:0x4010af3c>
-  irb(main):008:0> irb f                      # invoke subirb which has the
-					      #  context of f (instance of Foo)
-  irb#3(#<Foo:0x4010af3c>):001:0> jobs
-  #0->irb on main (#<Thread:0x400fb7e4> : stop)
-  #1->irb#1 on main (#<Thread:0x40125d64> : stop)
-  #2->irb#2 on Foo (#<Thread:0x4011d54c> : stop)
-  #3->irb#3 on #<Foo:0x4010af3c> (#<Thread:0x4010a1e0> : running)
-  nil
-  irb#3(#<Foo:0x4010af3c>):002:0> foo         # evaluate f.foo
-  1nil
-  irb#3(#<Foo:0x4010af3c>):003:0> bar         # evaluate f.bar
-  barnil
-  irb#3(#<Foo:0x4010af3c>):004:0> kill 1, 2, 3# kill job
-  nil
-  irb(main):009:0> jobs
-  #0->irb on main (#<Thread:0x400fb7e4> : running)
-  nil
-  irb(main):010:0> exit                       # exit
-  dim% 
-
-= Restrictions
-
-Because irb evaluates the inputs immediately after the imput is
-syntactically completed, irb gives slight different result than
-directly use ruby. Known difference is pointed out here. 
-
-
-== Declaration of the local variable
-
-The following causes an error in ruby:
-
-  eval "foo = 0"
-  foo
-  --
-  -:2: undefined local variable or method `foo' for #<Object:0x40283118> (NameError)
-  ---
-  NameError
-
-Though, the above will successfully done by irb. 
-
-  >> eval "foo = 0"
- => 0
- >> foo
- => 0
-
-Ruby evaluates a code after reading entire of code and determination
-of the scope of local variables. On the other hand, irb do
-immediately. More precisely, irb evaluate at first
-
-  evel "foo = 0" 
-
-then foo is defined on this timing. It is because of this
-incompatibility.
-
-If you'd like to detect those differences, begin...end can be used:
-
-  >> begin
-  ?>   eval "foo = 0"
-  >>   foo
-  >> end
-  NameError: undefined local variable or method `foo' for #<Object:0x4013d0f0>
-  (irb):3
-  (irb_local_binding):1:in `eval'
-
-== Here-document
-
-Implementation of Here-document is incomplete. 
-
-== Symbol
-
-Irb can not always recognize a symbol as to be Symbol. Concretely, an
-expression have completed, however Irb regard it as continuation line.
-
-=end
-
-% Begin Emacs Environment
-% Local Variables:
-% mode: text
-% comment-column: 0
-% comment-start: "%"
-% comment-end: "\n"
-% End:
-%
diff --git a/doc/irb/irb.rd.ja b/doc/irb/irb.rd.ja
index aa3c0e13aa..85b6536ee4 100644
--- a/doc/irb/irb.rd.ja
+++ b/doc/irb/irb.rd.ja
@@ -1,26 +1,25 @@
 irb -- interactive ruby
 				$Release Version: 0.9.5 $
-			   	$Revision: 1.3.2.1 $
-			   	$Date: 2005/04/19 19:24:56 $
+			   	$Revision$
 			   	by Keiju ISHITSUKA(keiju@ruby-lang.org)
 =begin
-= irb�Ȥ�?
+= irbとは?
 
-irb��interactive ruby��ά�Ǥ�. ruby�μ���ɸ�����Ϥ����ñ������/�¹Ԥ���
-����Υġ���Ǥ�.
+irbはinteractive rubyの略です. rubyの式を標準入力から簡単に入力/実行する
+ためのツールです.
 
-= ��ư
+= 起動
 
   % irb
 
-�ǹԤʤ��ޤ�. 
+で行ないます.
 
-= �Ȥ���
+= 使い方
 
-irb�λȤ�����, Ruby�����ΤäƤ���Ф����äƴ�ñ�Ǥ�. ����Ū�ˤ� irb ��
-�������ޥ�ɤ�¹Ԥ�������Ǥ�. irb��¹Ԥ����, �ʲ��Τ褦�ʥץ����
-�Ȥ�ɽ��Ƥ��ޤ�. ���, ruby�μ�������Ʋ�����. �������뤷�������Ǽ¹�
-����ޤ�.
+irbの使い方は, Rubyさえ知っていればいたって簡単です. 基本的には irb と
+いうコマンドを実行するだけです. irbを実行すると, 以下のようなプロンプ
+トが表れてきます. 後は, rubyの式を入れて下さい. 式が完結した時点で実行
+されます.
 
   dim% irb
   irb(main):001:0> 1+2
@@ -31,60 +30,59 @@ irb�λȤ�����, Ruby�����ΤäƤ���Ф����äƴ�ñ�Ǥ�. ����Ū�ˤ� irb ��
   irb(main):005:2>  end
   irb(main):006:1> end
   nil
-  irb(main):007:0> 
+  irb(main):007:0>
 
-�ޤ�, irb��Readline�⥸�塼��ˤ��б����Ƥ��ޤ�. Readline�⥸�塼�뤬
-���󥹥ȡ��뤵��Ƥ�����ˤ�, �����Ȥ��Τ�ɸ���ư��ˤʤ�ޤ�.
+また, irbはReadlineモジュールにも対応しています. Readlineモジュールが
+インストールされている時には, それを使うのが標準の動作になります.
 
-= ���ޥ�ɥ��ץ����
+= コマンドオプション
 
   irb.rb [options] file_name opts
   options:
-  -f		    ~/.irbrc ���ɤ߹��ޤʤ�.
-  -m		    bc�⡼��(ʬ��, ����η׻����Ǥ���)
-  -d                $DEBUG ��true�ˤ���(ruby -d ��Ʊ��)
-  -Kc		    ruby -Kc��Ʊ��
-  -r load-module    ruby -r ��Ʊ��.
-  --verbose	    ���줫��¹Ԥ���Ԥ�ɽ������(�ǥե����)
-  --noverbose	    ���줫��¹Ԥ���Ԥ�ɽ�����ʤ�
-  --echo	    �¹Է�̤�ɽ������(�ǥե����)
-  --noecho	    �¹Է�̤�ɽ�����ʤ�
-  --inspect	    ��̽��Ϥ�inspect���Ѥ���(bc�⡼�ɰʳ��ϥǥե����). 
-  --noinspect	    ��̽��Ϥ�inspect���Ѥ��ʤ�.
-  --readline	    readline�饤�֥������Ѥ���.
-  --noreadline	    readline�饤�֥������Ѥ��ʤ�. �ǥե���Ȥ�ư���,
-		    inf-ruby-mode�ʳ���readline�饤�֥������Ѥ��褦
-		    �Ȥ���. 
+  -f		    ~/.irbrc を読み込まない.
+  -m		    bcモード(分数, 行列の計算ができる)
+  -d                $DEBUG をtrueにする(ruby -d と同じ)
+  -Kc		    ruby -Kcと同じ
+  -r load-module    ruby -r と同じ.
+  --verbose	    これから実行する行を表示する(デフォルト)
+  --noverbose	    これから実行する行を表示しない
+  --echo	    実行結果を表示する(デフォルト)
+  --noecho	    実行結果を表示しない
+  --inspect	    結果出力にinspectを用いる(bcモード以外はデフォルト).
+  --noinspect	    結果出力にinspectを用いない.
+  --readline	    readlineライブラリを利用する.
+  --noreadline	    readlineライブラリを利用しない. デフォルトの動作は,
+		    inf-ruby-mode以外でreadlineライブラリを利用しよう
+		    とする.
   --prompt prompt-mode
   --prompt-mode prompt-mode
-		    �ץ���ץȥ⡼�ɤ����ؤ��ޤ�. �����������Ƥ����
-		    ����ץȥ⡼�ɤ�, default, simple, xmp, inf-ruby��
-		    �Ѱդ���Ƥ��ޤ�. �ǥե���Ȥ�default�ץ���ץȥ⡼
-		    �ɤˤʤäƤ��ޤ�.
-			    
-  --inf-ruby-mode   emacs��inf-ruby-mode�ѤΥץ���ץ�ɽ����Ԥʤ�. ��
-		    �˻��꤬�ʤ��¤�, readline�饤�֥��ϻȤ�ʤ��ʤ�.
+		    プロンプトモードを切替えます. 現在定義されているプ
+		    ロンプトモードは, default, simple, xmp, inf-rubyが
+		    用意されています. デフォルトはdefaultプロンプトモー
+		    ドになっています.
+
+  --inf-ruby-mode   emacsのinf-ruby-mode用のプロンプト表示を行なう. 特
+		    に指定がない限り, readlineライブラリは使わなくなる.
   --simple-prompt
-		    ���˥���ץ�ʥץ���ץȤ��Ѥ���⡼�ɤǤ�.
-  --noprompt	    �ץ���ץ�ɽ����Ԥʤ�ʤ�.
-  --tracer	    ���ޥ�ɼ¹Ի��˥ȥ졼����Ԥʤ�.
+		    非常にシンプルなプロンプトを用いるモードです.
+  --noprompt	    プロンプト表示を行なわない.
+  --tracer	    コマンド実行時にトレースを行なう.
   --back-trace-limit n
-		    �Хå��ȥ졼��ɽ����Хå��ȥ졼����Ƭ���� n, ���
-		    ����n�����Ԥʤ�. �ǥե���Ȥ�16 
-  --irb_debug n	    irb�ΥǥХå��ǥХå���٥��n�����ꤹ��(���Ѥ���
-		    ������̵��Ǥ��礦).
-  -v, --version	    irb�ΥС�������ɽ������
+		    バックトレース表示をバックトレースの頭から n, 後ろ
+		    からnだけ行なう. デフォルトは16
+  --irb_debug n	    irbのデバッグデバッグレベルをnに設定する(利用しな
+		    い方が無難でしょう).
+  -v, --version	    irbのバージョンを表示する
 
-= ����ե�����졼�����
+= コンフィギュレーション
 
-irb��ư����``~/.irbrc''���ɤ߹��ߤޤ�. �⤷¸�ߤ��ʤ�����,
-``.irbrc'', ``irb.rc'', ``_irbrc'', ``$irbrc''�ν��load���ߤޤ�.
+irb起動時に``~/.irbrc''を読み込みます. もし存在しない場合は,
+``.irbrc'', ``irb.rc'', ``_irbrc'', ``$irbrc''の順にloadを試みます.
 
-���ץ��������ꤹ�������, �ʲ��Υ��ޥ�ɤǤ�ǥե���Ȥ�ư�������
-�Ǥ��ޤ�.
+オプションを設定する代わりに, 以下のコマンドでもデフォルトの動作を設定
+できます.
 
   IRB.conf[:IRB_NAME]="irb"
-  IRB.conf[:MATH_MODE]=false
   IRB.conf[:USE_TRACER]=false
   IRB.conf[:USE_LOADER]=false
   IRB.conf[:IGNORE_SIGINT]=true
@@ -97,48 +95,48 @@ irb��ư����``~/.irbrc''���ɤ߹��ߤޤ�. �⤷¸�ߤ��ʤ�����,
   IRB.conf[:USE_TRACER] = false
   IRB.conf[:IGNORE_SIGINT] = true
   IRB.conf[:IGNORE_EOF] = false
-  IRB.conf[:PROMPT_MODE] = :DEFALUT
+  IRB.conf[:PROMPT_MODE] = :DEFAULT
   IRB.conf[:PROMPT] = {...}
   IRB.conf[:DEBUG_LEVEL]=0
   IRB.conf[:VERBOSE]=true
 
-== �ץ���ץȤ�����
+== プロンプトの設定
 
-�ץ���ץȤ򥫥����ޥ������������ˤ�, 
+プロンプトをカスタマイズしたい時には,
 
    IRB.conf[:PROMPT]
 
-���Ѥ��ޤ�. �㤨��, .irbrc����ǲ��Τ褦�ʼ��򵭽Ҥ��ޤ�:
+を用います. 例えば, .irbrcの中で下のような式を記述します:
 
-   IRB.conf[:PROMPT][:MY_PROMPT] = { # �ץ���ץȥ⡼�ɤ�̾��
-     :PROMPT_I => nil,		  # �̾�Υץ���ץ�
-     :PROMPT_N => nil,		  # ��³�ԤΥץ���ץ�
-     :PROMPT_S => nil,		  # ʸ����ʤɤη�³�ԤΥץ���ץ�
-     :PROMPT_C => nil,		  # ������³���Ƥ�����Υץ���ץ�
-     :RETURN => "    ==>%s\n"	  # �꥿������Υץ���ץ�
+   IRB.conf[:PROMPT][:MY_PROMPT] = { # プロンプトモードの名前
+     :PROMPT_I => nil,		  # 通常のプロンプト
+     :PROMPT_N => nil,		  # 継続行のプロンプト
+     :PROMPT_S => nil,		  # 文字列などの継続行のプロンプト
+     :PROMPT_C => nil,		  # 式が継続している時のプロンプト
+     :RETURN => "    ==>%s\n"	  # リターン時のプロンプト
    }
 
-�ץ���ץȥ⡼�ɤ���ꤷ�������ˤ�,
+プロンプトモードを指定したい時には,
 
   irb --prompt my-prompt
 
-�Ǥ��Υץ���ץȥ⡼�ɤǵ�ư����ޤ�. �ޤ���, .irbrc�˲����򵭽Ҥ��Ƥ�
-OK�Ǥ�.
+でそのプロンプトモードで起動されます. または, .irbrcに下式を記述しても
+OKです.
 
   IRB.conf[:PROMPT_MODE] = :MY_PROMPT
 
-PROMPT_I, PROMPT_N, PROMPT_S, PROMPT_C��, �ե����ޥåȤ���ꤷ�ޤ�.
+PROMPT_I, PROMPT_N, PROMPT_S, PROMPT_Cは, フォーマットを指定します.
 
-  %N	��ư���Ƥ��륳�ޥ��̾�����Ϥ����.
-  %m	main���֥�������(self)��to_s�ǽ��Ϥ����.
-  %M	main���֥�������(self)��inspect����ƽ��Ϥ����.
-  %l	ʸ������Υ����פ�ɽ��(", ', /, ], `]'��%w����λ�)
-  %NNi	����ǥ�ȤΥ�٥��ɽ��. NN�Ͽ���������printf��%NNd��Ʊ��. ��
-	ά��ǽ
-  %NNn	���ֹ��ɽ���ޤ�.
+  %N	起動しているコマンド名が出力される.
+  %m	mainオブジェクト(self)がto_sで出力される.
+  %M	mainオブジェクト(self)がinspectされて出力される.
+  %l	文字列中のタイプを表す(", ', /, ], `]'は%wの中の時)
+  %NNi	インデントのレベルを表す. NNは数字が入りprintfの%NNdと同じ. 省
+	略可能
+  %NNn	行番号を表します.
   %%    %
 
-�㤨��, �ǥե���ȤΥץ���ץȥ⡼�ɤ�:
+例えば, デフォルトのプロンプトモードは:
 
   IRB.conf[:PROMPT_MODE][:DEFAULT] = {
       :PROMPT_I => "%N(%m):%03n:%i> ",
@@ -146,216 +144,213 @@ PROMPT_I, PROMPT_N, PROMPT_S, PROMPT_C��, �ե����ޥåȤ���ꤷ�ޤ�.
       :PROMPT_S => "%N(%m):%03n:%i%l ",
       :PROMPT_C => "%N(%m):%03n:%i* ",
       :RETURN => "%s\n"
-  } 
+  }
 
-�ȤʤäƤ��ޤ�.
+となっています.
 
-RETURN��, ���ߤΤȤ���printf�����Ǥ�. ������ͤ��Ѥ�뤫���Τ�ޤ���.
+RETURNは, 現在のところprintf形式です. 将来仕様が変わるかも知れません.
 
-== ����irb������
+== サブirbの設定
 
-���ޥ�ɥ饤�󥪥ץ���󤪤��IRB.conf��(����)irb��ư���Υǥե���Ȥ�
-���������Τ�, `5. ���ޥ��'�ˤ���conf�Ǹ��̤�(����)irb�����꤬��
-����褦�ˤʤäƤ��ޤ�.
+コマンドラインオプションおよびIRB.confは(サブ)irb起動時のデフォルトの
+設定を決めるもので, `5. コマンド'にあるconfで個別の(サブ)irbの設定がで
+きるようになっています.
 
-IRB.conf[:IRB_RC]��proc�����ꤵ��Ƥ����, ����irb��ư������ˤ���
-proc��irb�Υ���ƥ����Ȥ�����Ȥ��ƸƤӽФ��ޤ�. ����ˤ�äƸ��̤Υ�
-��irb���Ȥ�������Ѥ��뤳�Ȥ��Ǥ���褦�ˤʤ�ޤ�.
+IRB.conf[:IRB_RC]にprocが設定されていると, サブirbを起動する時にその
+procをirbのコンテキストを引数として呼び出します. これによって個別のサ
+ブirbごとに設定を変えることができるようになります.
 
 
-= ���ޥ��
+= コマンド
 
-irb��ĥ���ޥ�ɤ�, ��ñ��̾����Ƭ��`irb_'��Ĥ���̾����ξ����������
-���ޤ�. �����, ��ñ��̾����override���줿���Τ���Ǥ�.
+irb拡張コマンドは, 簡単な名前と頭に`irb_'をつけた名前と両方定義されて
+います. これは, 簡単な名前がoverrideされた時のためです.
 
---- exit, quit, irb_exit	
-    ��λ����. 
-    ����irb�ξ��, ���Υ���irb��λ����.
+--- exit, quit, irb_exit
+    終了する.
+    サブirbの場合, そのサブirbを終了する.
 
 --- conf, irb_context
-    irb�θ��ߤ������ɽ������. ������ѹ���, conf�˥�å����������뤳
-    �Ȥˤ�äƹԤʤ���. 
+    irbの現在の設定を表示する. 設定の変更は, confにメッセージを送るこ
+    とによって行なえる.
 
 --- conf.eval_history = N
-    �¹Է�̤Υҥ��ȥ굡ǽ������.
-    nn��������nil�� nn>0 �Ǥ���Ф��ο������ҥ��ȥ�ˤ���롣nn==0�λ���
-    ̵���¤˵������롢nil���ȥҥ��ȥ굡ǽ�Ϥ���(�ǥե����). 
+    実行結果のヒストリ機能の設定.
+    nnは整数かnilで nn>0 であればその数だけヒストリにためる。nn==0の時は
+    無制限に記憶する、nilだとヒストリ機能はやめる(デフォルト).
 
 --- Conf.back_trace_limit
-    �Хå��ȥ졼��ɽ����Хå��ȥ졼����Ƭ����n, �������n�����Ԥʤ�.
-    �ǥե���Ȥ�16
-    
+    バックトレース表示をバックトレースの頭からn, 後ろからnだけ行なう.
+    デフォルトは16
+
 --- conf.debug_level = N
-    irb�ѤΥǥХå���٥������
+    irb用のデバッグレベルの設定
 
 --- conf.ignore_eof = true/false
-    ^D�����Ϥ��줿����ư������ꤹ��. true�λ���^D��̵�뤹��, false��
-    ����irb��λ����. 
+    ^Dが入力された時の動作を設定する. trueの時は^Dを無視する, falseの
+    時はirbを終了する.
 
 --- conf.ignore_sigint= true/false
-    ^C�����Ϥ��줿����ư������ꤹ��. false����, irb��λ����. true��
-    ����ư��ϰʲ��Τ褦�ˤʤ�:
-      ������: ����ޤ����Ϥ�����Τ򥭥�󥻥뤷�ȥåץ�٥�����. 
-      �¹���: �¹Ԥ���ߤ���.
+    ^Cが入力された時の動作を設定する. false時は, irbを終了する. trueの
+    時の動作は以下のようになる:
+      入力中: これまで入力したものをキャンセルしトップレベルに戻る.
+      実行中: 実行を中止する.
 
 --- conf.inf_ruby_mode = true/false
-    inf-ruby-mode�ѤΥץ���ץ�ɽ����Ԥʤ�. �ǥե���Ȥ�false.
+    inf-ruby-mode用のプロンプト表示を行なう. デフォルトはfalse.
 
 --- conf.inspect_mode = true/false/nil
-    ���󥹥ڥ��ȥ⡼�ɤ����ꤹ��.
-    true: ���󥹥ڥ��Ȥ���ɽ������.
-    false: �̾��print��ɽ������.
-    nil: �̾�⡼�ɤǤ����, inspect mode�Ȥʤ�, math�⡼�ɤλ���, non
-	 inspect mode�Ȥʤ�.   
-
---- conf.math_mode
-    ���ȤΤ�. bc�⡼��(ʬ��, ����η׻����Ǥ��ޤ�)���ɤ���?
+    インスペクトモードを設定する.
+    true: インスペクトして表示する.
+    false: 通常のprintで表示する.
+    nil: 通常モードであれば, inspect modeとなり, mathモードの時は, non
+	 inspect modeとなる.
 
 --- conf.use_loader = true/false
-    load/require����irb��file�ɤ߹��ߵ�ǽ���Ѥ���⡼�ɤΥ����å�(�ǥե�
-    ��Ȥ��Ѥ��ʤ�). ���Υ⡼�ɤ�IRB���Τ�ȿ�Ǥ����.
+    load/require時にirbのfile読み込み機能を用いるモードのスイッチ(デフォ
+    ルトは用いない). このモードはIRB全体に反映される.
 
 --- conf.prompt_c
-    if��ľ��ʤ�, �Ԥ���³���Ƥ�����Υץ���ץ�.
+    ifの直後など, 行が継続している時のプロンプト.
 
 --- conf.prompt_i
-    �̾�Υץ���ץ�.
+    通常のプロンプト.
 
 --- conf.prompt_s
-    ʸ������ʤɤ�ɽ���ץ���ץ�.
+    文字列中などを表すプロンプト.
 
 --- conf.rc
-    ~/.irbrc���ɤ߹�������ɤ���?
+    ~/.irbrcを読み込んだかどうか?
 
 --- conf.use_prompt = true/false
-    �ץ���ץ�ɽ�����뤫�ɤ���? �ǥե���ȤǤϥץ���ץȤ�ɽ������.
+    プロンプト表示するかどうか? デフォルトではプロンプトを表示する.
 
 --- conf.use_readline = true/false/nil
-    readline��Ȥ����ɤ���?
-    true: readline��Ȥ�.
-    false: readline��Ȥ�ʤ�.
-    nil: (�ǥե����)inf-ruby-mode�ʳ���readline�饤�֥������Ѥ���
-         ���Ȥ���.  
+    readlineを使うかどうか?
+    true: readlineを使う.
+    false: readlineを使わない.
+    nil: (デフォルト)inf-ruby-mode以外でreadlineライブラリを利用しよ
+         うとする.
 #
 #--- conf.verbose=T/F
-#    irb���餤�������ʥ�å���������Ϥ��뤫?
+#    irbからいろいろなメッセージを出力するか?
 
 --- cws, chws, irb_cws, irb_chws, irb_change_workspace [obj]
-    obj��self�Ȥ���. obj����ά���줿�Ȥ���, home workspace, ���ʤ��
-    irb��ư�����Ȥ���main object��self�Ȥ���.
+    objをselfとする. objが省略されたときは, home workspace, すなわち
+    irbを起動したときのmain objectをselfとする.
 
 --- pushws, irb_pushws, irb_push_workspace [obj]
-    UNIX�����륳�ޥ�ɤ�pushd��Ʊ��.
+    UNIXシェルコマンドのpushdと同様.
 
 --- popws, irb_popws, irb_pop_workspace
-    UNIX�����륳�ޥ�ɤ�popd��Ʊ��.
+    UNIXシェルコマンドのpopdと同様.
 
 --- irb [obj]
-    ����irb��Ω��������. obj�����ꤵ�줿����, ����obj��self�Ȥ���.
+    サブirbを立ちあげる. objが指定された時は, そのobjをselfとする.
 
 --- jobs, irb_jobs
-    ����irb�Υꥹ��
+    サブirbのリスト
 
 --- fg n, irb_fg n
-    ���ꤷ������irb�˥����å�����. n��, ���Τ�Τ���ꤹ��.
+    指定したサブirbにスイッチする. nは, 次のものを指定する.
 
-      irb�ֹ�
-      ����å�
-      irb���֥�������
-      self(irb obj�ǵ�ư��������obj)
+      irb番号
+      スレッド
+      irbオブジェクト
+      self(irb objで起動した時のobj)
 
 --- kill n, irb_kill n
-      ����irb��kill����. n��fg��Ʊ��.
+      サブirbをkillする. nはfgと同じ.
 
---- souce, irb_source  path
-    UNIX�����륳�ޥ�ɤ�source�Ȼ��Ƥ���. ���ߤδĶ����path��Υ�����
-    �ץȤ�ɾ������.
+--- source, irb_source  path
+    UNIXシェルコマンドのsourceと似ている. 現在の環境上でpath内のスクリ
+    プトを評価する.
 
 --- irb_load path, prev
 
-    Ruby��load��irb��.
+    Rubyのloadのirb版.
 
-= �����ƥ��ѿ�
+= システム変数
 
---- _  
-    ���η׻��μ¹Է�̤�Ф��Ƥ���(���������ѿ�).
---- __ 
-    �¹Է�̤������Ф��Ƥ���.
-    __[line_no]�ǡ����ιԤǼ¹Ԥ�����̤����뤳�Ȥ��Ǥ���. line_no�����
-    ���ˤϡ��ǿ��η�̤���-line_no���η�̤����뤳�Ȥ��Ǥ���.
+--- _
+    前の計算の実行結果を覚えている(ローカル変数).
+--- __
+    実行結果の履歴を覚えている.
+    __[line_no]で、その行で実行した結果を得ることができる. line_noが負の
+    時には、最新の結果から-line_no前の結果を得ることができる.
 
-=  ������
+=  使用例
 
-�ʲ��Τ褦�ʴ����Ǥ�.
+以下のような感じです.
 
   dim% ruby irb.rb
-  irb(main):001:0> irb                        # ����irb��Ω������
-  irb#1(main):001:0> jobs                     # ����irb�Υꥹ��
+  irb(main):001:0> irb                        # サブirbの立ちあげ
+  irb#1(main):001:0> jobs                     # サブirbのリスト
   #0->irb on main (#<Thread:0x400fb7e4> : stop)
   #1->irb#1 on main (#<Thread:0x40125d64> : running)
   nil
-  irb#1(main):002:0> fg 0                     # job�Υ����å�
+  irb#1(main):002:0> fg 0                     # jobのスイッチ
   nil
   irb(main):002:0> class Foo;end
   nil
-  irb(main):003:0> irb Foo                    # Foo�򥳥�ƥ����Ȥ���irb
-					      # Ω������
-  irb#2(Foo):001:0> def foo                   # Foo#foo�����
+  irb(main):003:0> irb Foo                    # Fooをコンテキストしてirb
+					      # 立ちあげ
+  irb#2(Foo):001:0> def foo                   # Foo#fooの定義
   irb#2(Foo):002:1>   print 1
   irb#2(Foo):003:1> end
   nil
-  irb#2(Foo):004:0> fg 0                      # job�򥹥��å�
+  irb#2(Foo):004:0> fg 0                      # jobをスイッチ
   nil
-  irb(main):004:0> jobs                       # job�Υꥹ��
+  irb(main):004:0> jobs                       # jobのリスト
   #0->irb on main (#<Thread:0x400fb7e4> : running)
   #1->irb#1 on main (#<Thread:0x40125d64> : stop)
   #2->irb#2 on Foo (#<Thread:0x4011d54c> : stop)
   nil
-  irb(main):005:0> Foo.instance_methods       # Foo#foo�������������
-					      # ��Ƥ���
+  irb(main):005:0> Foo.instance_methods       # Foo#fooがちゃんと定義さ
+					      # れている
   ["foo"]
-  irb(main):006:0> fg 2                       # job�򥹥��å�
+  irb(main):006:0> fg 2                       # jobをスイッチ
   nil
-  irb#2(Foo):005:0> def bar                   # Foo#bar�����
+  irb#2(Foo):005:0> def bar                   # Foo#barを定義
   irb#2(Foo):006:1>  print "bar"
   irb#2(Foo):007:1> end
   nil
   irb#2(Foo):010:0>  Foo.instance_methods
   ["bar", "foo"]
-  irb#2(Foo):011:0> fg 0                      
+  irb#2(Foo):011:0> fg 0
   nil
   irb(main):007:0> f = Foo.new
   #<Foo:0x4010af3c>
-  irb(main):008:0> irb f                      # Foo�Υ��󥹥��󥹤�irb��
-					      # Ω��������.
+  irb(main):008:0> irb f                      # Fooのインスタンスでirbを
+					      # 立ちあげる.
   irb#3(#<Foo:0x4010af3c>):001:0> jobs
   #0->irb on main (#<Thread:0x400fb7e4> : stop)
   #1->irb#1 on main (#<Thread:0x40125d64> : stop)
   #2->irb#2 on Foo (#<Thread:0x4011d54c> : stop)
   #3->irb#3 on #<Foo:0x4010af3c> (#<Thread:0x4010a1e0> : running)
   nil
-  irb#3(#<Foo:0x4010af3c>):002:0> foo         # f.foo�μ¹�
+  irb#3(#<Foo:0x4010af3c>):002:0> foo         # f.fooの実行
   nil
-  irb#3(#<Foo:0x4010af3c>):003:0> bar         # f.bar�μ¹�
+  irb#3(#<Foo:0x4010af3c>):003:0> bar         # f.barの実行
   barnil
-  irb#3(#<Foo:0x4010af3c>):004:0> kill 1, 2, 3# job��kill
+  irb#3(#<Foo:0x4010af3c>):004:0> kill 1, 2, 3# jobのkill
   nil
   irb(main):009:0> jobs
   #0->irb on main (#<Thread:0x400fb7e4> : running)
   nil
-  irb(main):010:0> exit                       # ��λ
-  dim% 
+  irb(main):010:0> exit                       # 終了
+  dim%
 
-= ���Ѿ������
+= 使用上の制限
 
-irb��, ɾ���Ǥ������(�����Ĥ�������)�Ǥ��༡�¹Ԥ�Ԥʤ��ޤ�. ��������
-��, ruby��ľ�ܻȤä�����, �㴳�ۤʤ�ư���Ԥʤ���礬����ޤ�.
+irbは, 評価できる時点(式が閉じた時点)での逐次実行を行ないます. したがっ
+て, rubyを直接使った時と, 若干異なる動作を行なう場合があります.
 
-�������餫�ˤʤäƤ������������������ޤ�.
+現在明らかになっている問題点を説明します.
 
-== ���������ѿ������
+== ローカル変数の宣言
 
-ruby�Ǥ�, �ʲ��Υץ������ϥ��顼�ˤʤ�ޤ�.
+rubyでは, 以下のプログラムはエラーになります.
 
   eval "foo = 0"
   foo
@@ -364,24 +359,24 @@ ruby�Ǥ�, �ʲ��Υץ������ϥ��顼�ˤʤ�ޤ�.
   ---
   NameError
 
-�Ȥ�����, irb���Ѥ����
+ところが, irbを用いると
 
   >> eval "foo = 0"
   => 0
   >> foo
   => 0
 
-�Ȥʤ�, ���顼�򵯤����ޤ���. �����, ruby���ǽ�˥�����ץ����Τ򥳥�
-�ѥ��뤷�ƥ��������ѿ�����ꤹ�뤫��Ǥ�. ������Ф�, irb�ϼ¹Բ�ǽ��
-�ʤ�(�����Ĥ���)�ȼ�ưŪ��ɾ�����Ƥ��뤫��Ǥ�. �嵭����Ǥ�, 
+となり, エラーを起こしません. これは, rubyが最初にスクリプト全体をコン
+パイルしてローカル変数を決定するからです. それに対し, irbは実行可能に
+なる(式が閉じる)と自動的に評価しているからです. 上記の例では,
 
-  evel "foo = 0" 
+  evel "foo = 0"
 
-��Ԥʤä�������ɾ����Ԥʤ�, ���λ������ѿ����������뤿��, ������ 
-�ѿ�foo���������Ƥ��뤫��Ǥ�.
+を行なった時点で評価を行ない, その時点で変数が定義されるため, 次式で
+変数fooは定義されているからです.
 
-���Τ褦��ruby��irb��ư��ΰ㤤���褷��������, begin...end�ǳ�ä�
-�Хå�Ū�˼¹Ԥ��Ʋ�����:
+このようなrubyとirbの動作の違いを解決したい場合は, begin...endで括って
+バッチ的に実行して下さい:
 
   >> begin
   ?>   eval "foo = 0"
@@ -391,14 +386,14 @@ ruby�Ǥ�, �ʲ��Υץ������ϥ��顼�ˤʤ�ޤ�.
   (irb):3
   (irb_local_binding):1:in `eval'
 
-== �ҥ��ɥ������
+== ヒアドキュメント
 
-���ߤΤȤ����ҥ��ɥ�����Ȥμ������Դ����Ǥ�. 
+現在のところヒアドキュメントの実装は不完全です.
 
-== ����ܥ�
+== シンボル
 
-����ܥ�Ǥ��뤫�ɤ�����Ƚ�Ǥ�ְ㤨�뤳�Ȥ�����ޤ�. ����Ū�ˤϼ�����λ
-���Ƥ���Τ˷�³�Ԥȸ��ʤ����Ȥ�����ޤ�.
+シンボルであるかどうかの判断を間違えることがあります. 具体的には式が完了
+しているのに継続行と見なすことがあります.
 
 =end
 
@@ -410,4 +405,3 @@ ruby�Ǥ�, �ʲ��Υץ������ϥ��顼�ˤʤ�ޤ�.
 % comment-end: "\n"
 % End:
 %
-
diff --git a/doc/keywords.rdoc b/doc/keywords.rdoc
new file mode 100644
index 0000000000..98bbd5e864
--- /dev/null
+++ b/doc/keywords.rdoc
@@ -0,0 +1,158 @@
+== Keywords
+
+The following keywords are used by Ruby.
+
+__ENCODING__::
+  The script encoding of the current file.  See Encoding.
+
+__LINE__::
+  The line number of this keyword in the current file.
+
+__FILE__::
+  The path to the current file.
+
+BEGIN::
+  Runs before any other code in the current file.  See {miscellaneous
+  syntax}[rdoc-ref:syntax/miscellaneous.rdoc]
+
+END::
+  Runs after any other code in the current file.  See {miscellaneous
+  syntax}[rdoc-ref:syntax/miscellaneous.rdoc]
+
+alias::
+  Creates an alias between two methods (and other things).  See {modules and
+  classes syntax}[rdoc-ref:syntax/modules_and_classes.rdoc]
+
+and::
+  Short-circuit Boolean and with lower precedence than <code>&&</code>
+
+begin::
+  Starts an exception handling block.  See {exceptions
+  syntax}[rdoc-ref:syntax/exceptions.rdoc]
+
+break::
+  Leaves a block early.  See {control expressions
+  syntax}[rdoc-ref:syntax/control_expressions.rdoc]
+
+case::
+  Starts a +case+ expression.  See {control expressions
+  syntax}[rdoc-ref:syntax/control_expressions.rdoc]
+
+class::
+  Creates or opens a class.  See {modules and classes
+  syntax}[rdoc-ref:syntax/modules_and_classes.rdoc]
+
+def::
+  Defines a method.  See {methods syntax}[rdoc-ref:syntax/methods.rdoc]
+
+defined?::
+  Returns a string describing its argument.  See {miscellaneous
+  syntax}[rdoc-ref:syntax/miscellaneous.rdoc]
+
+do::
+  Starts a block.
+
+else::
+  The unhandled condition in +case+, +if+ and +unless+ expressions.  See
+  {control expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+elsif::
+  An alternate condition for an +if+ expression.  See {control
+  expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+end::
+  The end of a syntax block.  Used by classes, modules, methods, exception
+  handling and control expressions.
+
+ensure::
+  Starts a section of code that is always run when an exception is raised.
+  See {exception handling}[rdoc-ref:syntax/exceptions.rdoc]
+
+false::
+  Boolean false.  See {literals}[rdoc-ref:syntax/literals.rdoc]
+
+for::
+  A loop that is similar to using the +each+ method.  See {control
+  expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+if::
+  Used for +if+ and modifier +if+ expressions.  See {control
+  expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+in::
+  Used to separate the iterable object and iterator variable in a +for+ loop.
+  See {control expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+module::
+  Creates or opens a module.  See {modules and classes
+  syntax}[rdoc-ref:syntax/modules_and_classes.rdoc]
+
+next::
+  Skips the rest of the block.  See {control
+  expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+nil::
+  A false value usually indicating "no value" or "unknown".  See
+  {literals}[rdoc-ref:syntax/literals.rdoc]
+
+not::
+  Inverts the following boolean expression.  Has a lower precedence than
+  <code>!</code>
+
+or::
+  Boolean or with lower precedence than <code>||</code>
+
+redo::
+  Restarts execution in the current block.  See {control
+  expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+rescue::
+  Starts an exception section of code in a +begin+ block.  See {exception
+  handling}[rdoc-ref:syntax/exceptions.rdoc]
+
+retry::
+  Retries an exception block.  See {exception
+  handling}[rdoc-ref:syntax/exceptions.rdoc]
+
+return::
+  Exits a method.  See {methods}[rdoc-ref:syntax/methods.rdoc]
+
+self::
+  The object the current method is attached to.  See
+  {methods}[rdoc-ref:syntax/methods.rdoc]
+
+super::
+  Calls the current method in a superclass.  See
+  {methods}[rdoc-ref:syntax/methods.rdoc]
+
+then::
+  Indicates the end of conditional blocks in control structures.  See
+  {control expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+true::
+  Boolean true.  See {literals}[rdoc-ref:syntax/literals.rdoc]
+
+undef::
+  Prevents a class or module from responding to a method call.
+  See {modules and classes}[rdoc-ref:syntax/modules_and_classes.rdoc]
+
+unless::
+  Used for +unless+ and modifier +unless+ expressions.  See {control
+  expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+until::
+  Creates a loop that executes until the condition is true.  See
+  {control expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+when::
+  A condition in a +case+ expression.  See
+  {control expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+while::
+  Creates a loop that executes while the condition is true.  See
+  {control expressions}[rdoc-ref:syntax/control_expressions.rdoc]
+
+yield::
+  Starts execution of the block sent to the current method.  See
+  {methods}[rdoc-ref:syntax/methods.rdoc]
+
diff --git a/doc/maintainers.rdoc b/doc/maintainers.rdoc
new file mode 100644
index 0000000000..e7baf933c6
--- /dev/null
+++ b/doc/maintainers.rdoc
@@ -0,0 +1,296 @@
+= Maintainers
+
+This page describes the current module, library, and extension maintainers of Ruby.
+
+== Module Maintainers
+
+A module maintainer is responsible for a certain part of Ruby.
+
+* The maintainer fixes bugs of the part. Particularly, they should fix security vulnerabilities as soon as possible.
+* They handle issues related the module on the Redmine or ML.
+* They may be discharged by the 3 months rule [ruby-core:25764].
+* They have commit right to Ruby's repository to modify their part in the repository.
+* They have "developer" role on the Redmine to modify issues.
+* They have authority to decide the feature of their part. But they should always respect discussions on ruby-core/ruby-dev.
+
+A submaintainer of a module is like a maintainer. But The submaintainer does
+not have authority to change/add a feature on his/her part. They need consensus
+on ruby-core/ruby-dev before changing/adding. Some of submaintainers have
+commit right, others don't.
+
+=== Language core features including security
+
+Yukihiro Matsumoto (matz)
+
+=== Evaluator
+
+Koichi Sasada (ko1)
+
+=== Core classes
+
+Yukihiro Matsumoto (matz)
+
+=== Documentation
+
+Zachary Scott (zzak)
+
+== Standard Library Maintainers
+
+=== Libraries
+
+[lib/English.rb]
+  _unmaintained_
+[lib/abbrev.rb]
+  Akinori MUSHA (knu)
+[lib/base64.rb]
+  Yusuke Endoh (mame)
+[lib/benchmark.rb]
+  _unmaintained_
+[lib/cgi.rb, lib/cgi/*]
+  Takeyuki Fujioka (xibbar)
+[lib/drb.rb, lib/drb/*]
+  Masatoshi SEKI (seki)
+[lib/debug.rb]
+  _unmaintained_
+[lib/delegate.rb]
+  _unmaintained_
+[lib/erb.rb]
+  Masatoshi SEKI (seki), Takashi Kokubun (k0kubun)
+[lib/find.rb]
+  Kazuki Tsujimoto (ktsj)
+[lib/getoptlong.rb]
+  _unmaintained_
+[lib/mkmf.rb]
+  _unmaintained_
+[lib/monitor.rb]
+  Shugo Maeda (shugo)
+[lib/net/ftp.rb]
+  Shugo Maeda (shugo)
+[lib/net/imap.rb]
+  Shugo Maeda (shugo)
+[lib/net/http.rb, lib/net/https.rb]
+  NARUSE, Yui (naruse)
+[lib/net/pop.rb]
+  _unmaintained_
+[lib/net/protocol.rb]
+  _unmaintained_
+[lib/net/smtp.rb]
+  _unmaintained_
+[lib/observer.rb]
+  _unmaintained_
+[lib/open-uri.rb]
+  Tanaka Akira (akr)
+[lib/open3.rb]
+  _unmaintained_
+[lib/optparse.rb, lib/optparse/*]
+  Nobuyuki Nakada (nobu)
+[lib/pp.rb]
+  Tanaka Akira (akr)
+[lib/prettyprint.rb]
+  Tanaka Akira (akr)
+[lib/profile.rb]
+  _unmaintained_
+[lib/profiler.rb]
+  _unmaintained_
+[lib/pstore.rb]
+  _unmaintained_
+[lib/racc/*]
+  Aaron Patterson (tenderlove)
+[lib/resolv-replace.rb]
+  Tanaka Akira (akr)
+[lib/resolv.rb]
+  Tanaka Akira (akr)
+[lib/rinda/*]
+  Masatoshi SEKI (seki)
+[lib/rubygems.rb, lib/ubygems.rb, lib/rubygems/*]
+  Eric Hodel (drbrain), Hiroshi SHIBATA (hsbt)
+  https://github.com/rubygems/rubygems
+[lib/set.rb]
+  Akinori MUSHA (knu)
+[lib/securerandom.rb]
+  Tanaka Akira (akr)
+[lib/shellwords.rb]
+  Akinori MUSHA (knu)
+[lib/singleton.rb]
+  Yukihiro Matsumoto (matz)
+[lib/tempfile.rb]
+  _unmaintained_
+[lib/tmpdir.rb]
+  _unmaintained_
+[lib/time.rb]
+  Tanaka Akira (akr)
+[lib/timeout.rb]
+  Yukihiro Matsumoto (matz)
+[lib/tsort.rb]
+  Tanaka Akira (akr)
+[lib/un.rb]
+  WATANABE Hirofumi (eban)
+[lib/unicode_normalize.rb, lib/unicode_normalize/*]
+  Martin J. Dürst
+[lib/uri.rb, lib/uri/*]
+  YAMADA, Akira (akira)
+[lib/weakref.rb]
+  _unmaintained_
+[lib/yaml.rb, lib/yaml/*]
+  Aaron Patterson (tenderlove), Hiroshi SHIBATA (hsbt)
+
+=== Extensions
+
+[ext/cgi]
+  Nobuyoshi Nakada (nobu)
+[ext/continuation]
+  Koichi Sasada (ko1)
+[ext/coverage]
+  Yusuke Endoh (mame)
+[ext/digest, ext/digest/*]
+  Akinori MUSHA (knu)
+[ext/fiber]
+  Koichi Sasada (ko1)
+[ext/io/nonblock]
+  Nobuyuki Nakada (nobu)
+[ext/io/wait]
+  Nobuyuki Nakada (nobu)
+[ext/nkf]
+  NARUSE, Yui (narse)
+[ext/objspace]
+  _unmaintained_
+[ext/pathname]
+  Tanaka Akira (akr)
+[ext/pty]
+  _unmaintained_
+[ext/racc]
+  Aaron Patterson (tenderlove)
+[ext/readline]
+  TAKAO Kouji (kouji)
+[ext/ripper]
+  _unmaintained_
+[ext/socket]
+  * Tanaka Akira (akr)
+  * API change needs matz's approval
+[ext/syslog]
+  Akinori MUSHA (knu)
+[ext/win32]
+  NAKAMURA Usaku (usa)
+[ext/win32ole]
+  Masaki Suketa (suke)
+
+== Default gems Maintainers
+
+=== Libraries
+
+[lib/bundler.rb, lib/bundler/*]
+  Hiroshi SHIBATA (hsbt)
+  https://github.com/bundler/bundler
+[lib/cmath.rb]
+  _unmaintained_
+  https://github.com/ruby/cmath
+[lib/csv.rb]
+  Kenta Murata (mrkn), Kouhei Sutou (kou)
+  https://github.com/ruby/csv
+[lib/e2mmap.rb]
+  Keiju ISHITSUKA (keiju)
+[lib/fileutils.rb]
+  _unmaintained_
+  https://github.com/ruby/fileutils
+[lib/forwardable.rb]
+  Keiju ISHITSUKA (keiju)
+[lib/ipaddr.rb]
+  Akinori MUSHA (knu)
+[lib/irb.rb, lib/irb/*]
+  Keiju ISHITSUKA (keiju)
+[lib/logger.rb]
+  Naotoshi Seo (sonots)
+[lib/matrix.rb]
+  Marc-Andre Lafortune (marcandre)
+[lib/mutex_m.rb]
+  Keiju ISHITSUKA (keiju)
+[lib/ostruct.rb]
+  Marc-Andre Lafortune (marcandre)
+[lib/prime.rb]
+  Yuki Sonoda (yugui)
+[lib/rdoc.rb, lib/rdoc/*]
+  Eric Hodel (drbrain), Hiroshi SHIBATA (hsbt)
+  https://github.com/ruby/rdoc
+[lib/rexml/*]
+  Kouhei Sutou (kou)
+[lib/rss.rb, lib/rss/*]
+  Kouhei Sutou (kou)
+[lib/scanf.rb]
+  David A. Black (dblack)
+  https://github.com/ruby/scanf
+[lib/shell.rb, lib/shell/*]
+  Keiju ISHITSUKA (keiju)
+[lib/sync.rb]
+  Keiju ISHITSUKA (keiju)
+[lib/thwait.rb]
+  Keiju ISHITSUKA (keiju)
+[lib/tracer.rb]
+  Keiju ISHITSUKA (keiju)
+[lib/webrick.rb, lib/webrick/*]
+  Eric Wong (normalperson)
+  https://bugs.ruby-lang.org/
+
+=== Extensions
+
+[ext/bigdecimal]
+  Kenta Murata (mrkn)
+  https://github.com/ruby/bigdecimal
+[ext/date]
+  _unmaintained_
+  https://github.com/ruby/date
+[ext/dbm]
+  _unmaintained_
+  https://github.com/ruby/dbm
+[ext/etc]
+  Ruby core team
+  https://github.com/ruby/etc
+[ext/fcntl]
+  Ruby core team
+  https://github.com/ruby/fcntl
+[ext/fiddle]
+  Aaron Patterson (tenderlove)
+  https://github.com/ruby/fiddle
+[ext/gdbm]
+  Yukihiro Matsumoto (matz)
+  https://github.com/ruby/gdbm
+[ext/io/console]
+  Nobuyuki Nakada (nobu)
+  https://github.com/ruby/io-console
+[ext/json]
+  NARUSE, Yui (naruse), Hiroshi SHIBATA (hsbt)
+  https://github.com/flori/json
+[ext/openssl]
+  Kazuki Yamaguchi (rhe)
+  https://github.com/ruby/openssl
+[ext/psych]
+  Aaron Patterson (tenderlove), Hiroshi SHIBATA(hsbt)
+  https://github.com/ruby/psych
+[ext/sdbm]
+  Yukihiro Matsumoto (matz)
+  https://github.com/ruby/sdbm
+[ext/stringio]
+  Nobuyuki Nakada (nobu)
+  https://github.com/ruby/stringio
+[ext/strscan]
+  _unmaintained_
+  https://github.com/ruby/strscan
+[ext/zlib]
+  NARUSE, Yui (naruse)
+  https://github.com/ruby/zlib
+
+== Bundled gems upstream repositories
+
+[did_you_mean]
+  https://github.com/yuki24/did_you_mean
+[minitest]
+  https://github.com/seattlerb/minitest
+[net-telnet]
+  https://github.com/ruby/net-telnet
+[power_assert]
+  https://github.com/k-tsj/power_assert
+[rake]
+  https://github.com/ruby/rake
+[test-unit]
+  https://github.com/test-unit/test-unit
+[xmlrpc]
+  https://github.com/ruby/xmlrpc
diff --git a/doc/marshal.rdoc b/doc/marshal.rdoc
new file mode 100644
index 0000000000..a51f1bf873
--- /dev/null
+++ b/doc/marshal.rdoc
@@ -0,0 +1,313 @@
+= Marshal Format
+
+The Marshal format is used to serialize ruby objects.  The format can store
+arbitrary objects through three user-defined extension mechanisms.
+
+For documentation on using Marshal to serialize and deserialize objects, see
+the Marshal module.
+
+This document calls a serialized set of objects a stream.  The Ruby
+implementation can load a set of objects from a String, an IO or an object
+that implements a +getc+ method.
+
+== Stream Format
+
+The first two bytes of the stream contain the major and minor version, each as
+a single byte encoding a digit.  The version implemented in Ruby is 4.8
+(stored as "\x04\x08") and is supported by ruby 1.8.0 and newer.
+
+Different major versions of the Marshal format are not compatible and cannot
+be understood by other major versions.  Lesser minor versions of the format
+can be understood by newer minor versions.  Format 4.7 can be loaded by a 4.8
+implementation but format 4.8 cannot be loaded by a 4.7 implementation.
+
+Following the version bytes is a stream describing the serialized object.  The
+stream contains nested objects (the same as a Ruby object) but objects in the
+stream do not necessarily have a direct mapping to the Ruby object model.
+
+Each object in the stream is described by a byte indicating its type followed
+by one or more bytes describing the object.  When "object" is mentioned below
+it means any of the types below that defines a Ruby object.
+
+=== true, false, nil
+
+These objects are each one byte long.  "T" is represents +true+, "F"
+represents +false+ and "0" represents +nil+.
+
+=== Fixnum and long
+
+"i" represents a signed 32 bit value using a packed format.  One through five
+bytes follows the type.  The value loaded will always be a Fixnum.  On
+32 bit platforms (where the precision of a Fixnum is less than 32 bits)
+loading large values will cause overflow on CRuby.
+
+The fixnum type is used to represent both ruby Fixnum objects and the sizes of
+marshaled arrays, hashes, instance variables and other types.  In the
+following sections "long" will mean the format described below, which supports
+full 32 bit precision.
+
+The first byte has the following special values:
+
+"\x00"::
+  The value of the integer is 0.  No bytes follow.
+
+"\x01"::
+  The total size of the integer is two bytes.  The following byte is a
+  positive integer in the range of 0 through 255.  Only values between 123
+  and 255 should be represented this way to save bytes.
+
+"\xff"::
+  The total size of the integer is two bytes.  The following byte is a
+  negative integer in the range of -1 through -256.
+
+"\x02"::
+  The total size of the integer is three bytes.  The following two bytes are a
+  positive little-endian integer.
+
+"\xfe"::
+  The total size of the integer is three bytes.  The following two bytes are a
+  negative little-endian integer.
+
+"\x03"::
+  The total size of the integer is four bytes.  The following three bytes are
+  a positive little-endian integer.
+
+"\xfd"::
+  The total size of the integer is two bytes.  The following three bytes are a
+  negative little-endian integer.
+
+"\x04"::
+  The total size of the integer is five bytes.  The following four bytes are a
+  positive little-endian integer.  For compatibility with 32 bit ruby,
+  only Fixnums less than 1073741824 should be represented this way.  For sizes
+  of stream objects full precision may be used.
+
+"\xfc"::
+  The total size of the integer is two bytes.  The following four bytes are a
+  negative little-endian integer.  For compatibility with 32 bit ruby,
+  only Fixnums greater than -10737341824 should be represented this way.  For
+  sizes of stream objects full precision may be used.
+
+Otherwise the first byte is a sign-extended eight-bit value with an offset.
+If the value is positive the value is determined by subtracting 5 from the
+value.  If the value is negative the value is determined by adding 5 to the
+value.
+
+There are multiple representations for many values.  CRuby always outputs the
+shortest representation possible.
+
+=== Symbols and Byte Sequence
+
+":" represents a real symbol.  A real symbol contains the data needed to
+define the symbol for the rest of the stream as future occurrences in the
+stream will instead be references (a symbol link) to this one.  The reference
+is a zero-indexed 32 bit value (so the first occurrence of <code>:hello</code>
+is 0).
+
+Following the type byte is byte sequence which consists of a long indicating
+the number of bytes in the sequence followed by that many bytes of data.  Byte
+sequences have no encoding.
+
+For example, the following stream contains the Symbol <code>:hello</code>:
+
+  "\x04\x08:\x0ahello"
+
+";" represents a Symbol link which references a previously defined Symbol.
+Following the type byte is a long containing the index in the lookup table for
+the linked (referenced) Symbol.
+
+For example, the following stream contains <code>[:hello, :hello]</code>:
+
+  "\x04\b[\a:\nhello;\x00"
+
+When a "symbol" is referenced below it may be either a real symbol or a
+symbol link.
+
+=== Object References
+
+Separate from but similar to symbol references, the stream contains only one
+copy of each object (as determined by #object_id) for all objects except
+true, false, nil, Fixnums and Symbols (which are stored separately as
+described above) a one-indexed 32 bit value will be stored and reused when the
+object is encountered again.  (The first object has an index of 1).
+
+"@" represents an object link.  Following the type byte is a long giving the
+index of the object.
+
+For example, the following stream contains an Array of the same
+<code>"hello"</code> object twice:
+
+  "\004\b[\a\"\nhello@\006"
+
+=== Instance Variables
+
+"I" indicates that instance variables follow the next object.  An object
+follows the type byte.  Following the object is a length indicating the number
+of instance variables for the object.  Following the length is a set of
+name-value pairs.  The names are symbols while the values are objects.  The
+symbols must be instance variable names (<code>:@name</code>).
+
+An Object ("o" type, described below) uses the same format for its instance
+variables as described here.
+
+For a String and Regexp (described below) a special instance variable
+<code>:E</code> is used to indicate the Encoding.
+
+=== Extended
+
+"e" indicates that the next object is extended by a module.  An object follows
+the type byte.  Following the object is a symbol that contains the name of the
+module the object is extended by.
+
+=== Array
+
+"[" represents an Array.  Following the type byte is a long indicating the
+number of objects in the array.  The given number of objects follow the
+length.
+
+=== Bignum
+
+"l" represents a Bignum which is composed of three parts:
+
+sign::
+  A single byte containing "+" for a positive value or "-" for a negative
+  value.
+length::
+  A long indicating the number of bytes of Bignum data follows, divided by
+  two.  Multiply the length by two to determine the number of bytes of data
+  that follow.
+data::
+  Bytes of Bignum data representing the number.
+
+The following ruby code will reconstruct the Bignum value from an array of
+bytes:
+
+  result = 0
+
+  bytes.each_with_index do |byte, exp|
+   result += (byte * 2 ** (exp * 8))
+  end
+
+=== Class and Module
+
+"c" represents a Class object, "m" represents a Module and "M" represents
+either a class or module (this is an old-style for compatibility).  No class
+or module content is included, this type is only a reference.  Following the
+type byte is a byte sequence which is used to look up an existing class or
+module, respectively.
+
+Instance variables are not allowed on a class or module.
+
+If no class or module exists an exception should be raised.
+
+For "c" and "m" types, the loaded object must be a class or module,
+respectively.
+
+=== Data
+
+"d" represents a Data object.  (Data objects are wrapped pointers from ruby
+extensions.)  Following the type byte is a symbol indicating the class for the
+Data object and an object that contains the state of the Data object.
+
+To dump a Data object Ruby calls _dump_data.  To load a Data object Ruby calls
+_load_data with the state of the object on a newly allocated instance.
+
+=== Float
+
+"f" represents a Float object.  Following the type byte is a byte sequence
+containing the float value.  The following values are special:
+
+"inf"::
+  Positive infinity
+
+"-inf"::
+  Negative infinity
+
+"nan"::
+  Not a Number
+
+Otherwise the byte sequence contains a C double (loadable by strtod(3)).
+Older minor versions of Marshal also stored extra mantissa bits to ensure
+portability across platforms but 4.8 does not include these.  See
+[ruby-talk:69518] for some explanation.
+
+=== Hash and Hash with Default Value
+
+"{" represents a Hash object while "}" represents a Hash with a default value
+set (<code>Hash.new 0</code>).  Following the type byte is a long indicating
+the number of key-value pairs in the Hash, the size.  Double the given number
+of objects follow the size.
+
+For a Hash with a default value, the default value follows all the pairs.
+
+=== Module and Old Module
+
+=== Object
+
+"o" represents an object that doesn't have any other special form (such as
+a user-defined or built-in format).  Following the type byte is a symbol
+containing the class name of the object.  Following the class name is a long
+indicating the number of instance variable names and values for the object.
+Double the given number of pairs of objects follow the size.
+
+The keys in the pairs must be symbols containing instance variable names.
+
+=== Regular Expression
+
+"/" represents a regular expression.  Following the type byte is a byte
+sequence containing the regular expression source.  Following the type byte is
+a byte containing the regular expression options (case-insensitive, etc.) as a
+signed 8-bit value.
+
+Regular expressions can have an encoding attached through instance variables
+(see above).  If no encoding is attached escapes for the following regexp
+specials not present in ruby 1.8 must be removed: g-m, o-q, u, y, E, F, H-L,
+N-V, X, Y.
+
+=== String
+
+'"' represents a String.  Following the type byte is a byte sequence
+containing the string content.  When dumped from ruby 1.9 an encoding instance
+variable (<code>:E</code> see above) should be included unless the encoding is
+binary.
+
+=== Struct
+
+"S" represents a Struct.  Following the type byte is a symbol containing the
+name of the struct.  Following the name is a long indicating the number of
+members in the struct.  Double the number of objects follow the member count.
+Each member is a pair containing the member's symbol and an object for the
+value of that member.
+
+If the struct name does not match a Struct subclass in the running ruby an
+exception should be raised.
+
+If there is a mismatch between the struct in the currently running ruby and
+the member count in the marshaled struct an exception should be raised.
+
+=== User Class
+
+"C" represents a subclass of a String, Regexp, Array or Hash.  Following the
+type byte is a symbol containing the name of the subclass.  Following the name
+is the wrapped object.
+
+=== User Defined
+
+"u" represents an object with a user-defined serialization format using the
++_dump+ instance method and +_load+ class method.  Following the type byte is
+a symbol containing the class name.  Following the class name is a byte
+sequence containing the user-defined representation of the object.
+
+The class method +_load+ is called on the class with a string created from the
+byte-sequence.
+
+=== User Marshal
+
+"U" represents an object with a user-defined serialization format using the
++marshal_dump+ and +marshal_load+ instance methods.  Following the type byte
+is a symbol containing the class name.  Following the class name is an object
+containing the data.
+
+Upon loading a new instance must be allocated and +marshal_load+ must be
+called on the instance with the data.
+
diff --git a/doc/pty/README.expect.ja b/doc/pty/README.expect.ja
new file mode 100644
index 0000000000..7c0456f24f
--- /dev/null
+++ b/doc/pty/README.expect.ja
@@ -0,0 +1,21 @@
+	README for expect
+		by A. Ito, 28 October, 1998
+
+  Expectライブラリは，tcl の expect パッケージと似たような機能を
+IOクラスに追加します．
+
+  追加されるメソッドの使い方は次の通りです．
+
+   IO#expect(pattern,timeout=9999999)
+
+pattern は String か Regexp のインスタンス，timeout は Fixnum
+のインスタンスです．timeout は省略できます．
+  このメソッドがブロックなしで呼ばれた場合，まずレシーバである
+IOオブジェクトから pattern にマッチするパターンが読みこまれる
+まで待ちます．パターンが得られたら，そのパターンに関する配列を
+返します．配列の最初の要素は，pattern にマッチするまでに読みこ
+まれた内容の文字列です．2番目以降の要素は，pattern の正規表現
+の中にアンカーがあった場合に，そのアンカーにマッチする部分です．
+もしタイムアウトが起きた場合は，このメソッドはnilを返します．
+  このメソッドがブロック付きで呼ばれた場合には，マッチした要素の
+配列がブロック引数として渡され，ブロックが評価されます．
diff --git a/doc/pty/README.ja b/doc/pty/README.ja
new file mode 100644
index 0000000000..2d83ffa033
--- /dev/null
+++ b/doc/pty/README.ja
@@ -0,0 +1,76 @@
+pty 拡張モジュール version 0.3 by A.ito
+
+1. はじめに
+
+この拡張モジュールは，仮想tty (pty) を通して適当なコマンドを
+実行する機能を ruby に提供します．
+
+2. インストール
+
+次のようにしてインストールしてください．
+
+(1)  ruby extconf.rb
+
+    を実行すると Makefile が生成されます．
+
+(2) make; make install を実行してください．
+
+3. 何ができるか
+
+この拡張モジュールは，PTY というモジュールを定義します．その中
+には，次のようなモジュール関数が含まれています．
+
+   getpty(command)
+   spawn(command)
+
+      この関数は，仮想ttyを確保し，指定されたコマンドをその仮想tty
+      の向こうで実行し，配列を返します．戻り値は3つの要素からなる
+      配列です．最初の要素は仮想ttyから読み出すためのIOオブジェクト，
+      2番目は書きこむためのIOオブジェクト，3番目は子プロセスのプロ
+      セスIDです．この関数がイテレータとして呼ばれた場合，これらの
+      要素はブロックパラメータとして渡され，関数自体はnilを返します．
+
+      子プロセスが終了したり停止した場合には，例外が発生します．この関
+      数がブロックパラメータ付きで呼ばれた場合には，そのブロックの中で
+      のみ例外が発生します．子プロセスをモニターしているスレッドはブロッ
+      クを抜けるときに終了します．
+
+   protect_signal
+   reset_signal
+
+      廃止予定です．
+
+   PTY.open
+
+      仮想ttyを確保し，マスター側に対応するIOオブジェクトとスレーブ側に
+      対応するFileオブジェクトの配列を返します．ブロック付きで呼び出さ
+      れた場合は，これらの要素はブロックパラメータとして渡され，ブロッ
+      クから返された結果を返します．また、このマスターIOとスレーブFile
+      は、ブロックを抜けるときにクローズ済みでなければクローズされます．
+
+   PTY.check(pid[, raise=false])
+
+      pidで指定された子プロセスの状態をチェックし，実行中であればnilを
+      返します．終了しているか停止している場合、第二引数が偽であれば、
+      対応するProcess::Statusオブジェクトを返します。真であれば
+      PTY::ChildExited例外が発生します．
+
+4. 利用について
+
+伊藤彰則が著作権を保有します．
+
+ソースプログラムまたはドキュメントに元の著作権表示が改変されずに
+表示されている場合に限り，誰でも，このソフトウェアを無償かつ著作
+権者に無断で利用・配布・改変できます．利用目的は限定されていませ
+ん．
+
+このプログラムの利用・配布その他このプログラムに関係する行為によ
+って生じたいかなる損害に対しても，作者は一切責任を負いません．
+
+5. バグ報告等
+
+バグレポートは歓迎します．
+
+	aito@ei5sun.yz.yamagata-u.ac.jp
+
+まで電子メールでバグレポートをお送りください．
diff --git a/doc/regexp.rdoc b/doc/regexp.rdoc
new file mode 100644
index 0000000000..9218a75b67
--- /dev/null
+++ b/doc/regexp.rdoc
@@ -0,0 +1,709 @@
+# -*- mode: rdoc; coding: utf-8; fill-column: 74; -*-
+
+Regular expressions (<i>regexp</i>s) are patterns which describe the
+contents of a string. They're used for testing whether a string contains a
+given pattern, or extracting the portions that match. They are created
+with the <tt>/</tt><i>pat</i><tt>/</tt> and
+<tt>%r{</tt><i>pat</i><tt>}</tt> literals or the <tt>Regexp.new</tt>
+constructor.
+
+A regexp is usually delimited with forward slashes (<tt>/</tt>). For
+example:
+
+    /hay/ =~ 'haystack'   #=> 0
+    /y/.match('haystack') #=> #<MatchData "y">
+
+If a string contains the pattern it is said to <i>match</i>. A literal
+string matches itself.
+
+Here 'haystack' does not contain the pattern 'needle', so it doesn't match:
+
+    /needle/.match('haystack') #=> nil
+
+Here 'haystack' contains the pattern 'hay', so it matches:
+
+    /hay/.match('haystack')    #=> #<MatchData "hay">
+
+Specifically, <tt>/st/</tt> requires that the string contains the letter
+_s_ followed by the letter _t_, so it matches _haystack_, also.
+
+== <tt>=~</tt> and Regexp#match
+
+Pattern matching may be achieved by using <tt>=~</tt> operator or Regexp#match
+method.
+
+=== <tt>=~</tt> operator
+
+<tt>=~</tt> is Ruby's basic pattern-matching operator.  When one operand is a
+regular expression and the other is a string then the regular expression is
+used as a pattern to match against the string.  (This operator is equivalently
+defined by Regexp and String so the order of String and Regexp do not matter.
+Other classes may have different implementations of <tt>=~</tt>.)  If a match
+is found, the operator returns index of first match in string, otherwise it
+returns +nil+.
+
+    /hay/ =~ 'haystack'   #=> 0
+    'haystack' =~ /hay/   #=> 0
+    /a/   =~ 'haystack'   #=> 1
+    /u/   =~ 'haystack'   #=> nil
+
+Using <tt>=~</tt> operator with a String and Regexp the <tt>$~</tt> global
+variable is set after a successful match.  <tt>$~</tt> holds a MatchData
+object. Regexp.last_match is equivalent to <tt>$~</tt>.
+
+=== Regexp#match method
+
+The #match method returns a MatchData object:
+
+    /st/.match('haystack')   #=> #<MatchData "st">
+
+== Metacharacters and Escapes
+
+The following are <i>metacharacters</i> <tt>(</tt>, <tt>)</tt>,
+<tt>[</tt>, <tt>]</tt>, <tt>{</tt>, <tt>}</tt>, <tt>.</tt>, <tt>?</tt>,
+<tt>+</tt>, <tt>*</tt>. They have a specific meaning when appearing in a
+pattern. To match them literally they must be backslash-escaped. To match
+a backslash literally, backslash-escape it: <tt>\\\\</tt>.
+
+    /1 \+ 2 = 3\?/.match('Does 1 + 2 = 3?') #=> #<MatchData "1 + 2 = 3?">
+    /a\\\\b/.match('a\\\\b')                    #=> #<MatchData "a\\b">
+
+Patterns behave like double-quoted strings so can contain the same
+backslash escapes.
+
+    /\s\u{6771 4eac 90fd}/.match("Go to 東京都")
+        #=> #<MatchData " 東京都">
+
+Arbitrary Ruby expressions can be embedded into patterns with the
+<tt>#{...}</tt> construct.
+
+    place = "東京都"
+    /#{place}/.match("Go to 東京都")
+        #=> #<MatchData "東京都">
+
+== Character Classes
+
+A <i>character class</i> is delimited with square brackets (<tt>[</tt>,
+<tt>]</tt>) and lists characters that may appear at that point in the
+match. <tt>/[ab]/</tt> means _a_ or _b_, as opposed to <tt>/ab/</tt> which
+means _a_ followed by _b_.
+
+    /W[aeiou]rd/.match("Word") #=> #<MatchData "Word">
+
+Within a character class the hyphen (<tt>-</tt>) is a metacharacter
+denoting an inclusive range of characters. <tt>[abcd]</tt> is equivalent
+to <tt>[a-d]</tt>. A range can be followed by another range, so
+<tt>[abcdwxyz]</tt> is equivalent to <tt>[a-dw-z]</tt>. The order in which
+ranges or individual characters appear inside a character class is
+irrelevant.
+
+    /[0-9a-f]/.match('9f') #=> #<MatchData "9">
+    /[9f]/.match('9f')     #=> #<MatchData "9">
+
+If the first character of a character class is a caret (<tt>^</tt>) the
+class is inverted: it matches any character _except_ those named.
+
+    /[^a-eg-z]/.match('f') #=> #<MatchData "f">
+
+A character class may contain another character class. By itself this
+isn't useful because <tt>[a-z[0-9]]</tt> describes the same set as
+<tt>[a-z0-9]</tt>. However, character classes also support the <tt>&&</tt>
+operator which performs set intersection on its arguments. The two can be
+combined as follows:
+
+    /[a-w&&[^c-g]z]/ # ([a-w] AND ([^c-g] OR z))
+
+This is equivalent to:
+
+    /[abh-w]/
+
+The following metacharacters also behave like character classes:
+
+* <tt>/./</tt> - Any character except a newline.
+* <tt>/./m</tt> - Any character (the +m+ modifier enables multiline mode)
+* <tt>/\w/</tt> - A word character (<tt>[a-zA-Z0-9_]</tt>)
+* <tt>/\W/</tt> - A non-word character (<tt>[^a-zA-Z0-9_]</tt>).
+  Please take a look at {Bug #4044}[https://bugs.ruby-lang.org/issues/4044] if
+  using <tt>/\W/</tt> with the <tt>/i</tt> modifier.
+* <tt>/\d/</tt> - A digit character (<tt>[0-9]</tt>)
+* <tt>/\D/</tt> - A non-digit character (<tt>[^0-9]</tt>)
+* <tt>/\h/</tt> - A hexdigit character (<tt>[0-9a-fA-F]</tt>)
+* <tt>/\H/</tt> - A non-hexdigit character (<tt>[^0-9a-fA-F]</tt>)
+* <tt>/\s/</tt> - A whitespace character: <tt>/[ \t\r\n\f\v]/</tt>
+* <tt>/\S/</tt> - A non-whitespace character: <tt>/[^ \t\r\n\f\v]/</tt>
+
+POSIX <i>bracket expressions</i> are also similar to character classes.
+They provide a portable alternative to the above, with the added benefit
+that they encompass non-ASCII characters. For instance, <tt>/\d/</tt>
+matches only the ASCII decimal digits (0-9); whereas <tt>/[[:digit:]]/</tt>
+matches any character in the Unicode _Nd_ category.
+
+* <tt>/[[:alnum:]]/</tt> - Alphabetic and numeric character
+* <tt>/[[:alpha:]]/</tt> - Alphabetic character
+* <tt>/[[:blank:]]/</tt> - Space or tab
+* <tt>/[[:cntrl:]]/</tt> - Control character
+* <tt>/[[:digit:]]/</tt> - Digit
+* <tt>/[[:graph:]]/</tt> - Non-blank character (excludes spaces, control
+  characters, and similar)
+* <tt>/[[:lower:]]/</tt> - Lowercase alphabetical character
+* <tt>/[[:print:]]/</tt> - Like [:graph:], but includes the space character
+* <tt>/[[:punct:]]/</tt> - Punctuation character
+* <tt>/[[:space:]]/</tt> - Whitespace character (<tt>[:blank:]</tt>, newline,
+  carriage return, etc.)
+* <tt>/[[:upper:]]/</tt> - Uppercase alphabetical
+* <tt>/[[:xdigit:]]/</tt> - Digit allowed in a hexadecimal number (i.e.,
+  0-9a-fA-F)
+
+Ruby also supports the following non-POSIX character classes:
+
+* <tt>/[[:word:]]/</tt> - A character in one of the following Unicode
+  general categories _Letter_, _Mark_, _Number_,
+  <i>Connector_Punctuation</i>
+* <tt>/[[:ascii:]]/</tt> - A character in the ASCII character set
+
+    # U+06F2 is "EXTENDED ARABIC-INDIC DIGIT TWO"
+    /[[:digit:]]/.match("\u06F2")    #=> #<MatchData "\u{06F2}">
+    /[[:upper:]][[:lower:]]/.match("Hello") #=> #<MatchData "He">
+    /[[:xdigit:]][[:xdigit:]]/.match("A6")  #=> #<MatchData "A6">
+
+== Repetition
+
+The constructs described so far match a single character. They can be
+followed by a repetition metacharacter to specify how many times they need
+to occur. Such metacharacters are called <i>quantifiers</i>.
+
+* <tt>*</tt> - Zero or more times
+* <tt>+</tt> - One or more times
+* <tt>?</tt> - Zero or one times (optional)
+* <tt>{</tt><i>n</i><tt>}</tt> - Exactly <i>n</i> times
+* <tt>{</tt><i>n</i><tt>,}</tt> - <i>n</i> or more times
+* <tt>{,</tt><i>m</i><tt>}</tt> - <i>m</i> or less times
+* <tt>{</tt><i>n</i><tt>,</tt><i>m</i><tt>}</tt> - At least <i>n</i> and
+  at most <i>m</i> times
+
+At least one uppercase character ('H'), at least one lowercase character
+('e'), two 'l' characters, then one 'o':
+
+    "Hello".match(/[[:upper:]]+[[:lower:]]+l{2}o/) #=> #<MatchData "Hello">
+
+Repetition is <i>greedy</i> by default: as many occurrences as possible
+are matched while still allowing the overall match to succeed. By
+contrast, <i>lazy</i> matching makes the minimal amount of matches
+necessary for overall success. A greedy metacharacter can be made lazy by
+following it with <tt>?</tt>.
+
+Both patterns below match the string. The first uses a greedy quantifier so
+'.+' matches '<a><b>'; the second uses a lazy quantifier so '.+?' matches
+'<a>':
+
+    /<.+>/.match("<a><b>")  #=> #<MatchData "<a><b>">
+    /<.+?>/.match("<a><b>") #=> #<MatchData "<a>">
+
+A quantifier followed by <tt>+</tt> matches <i>possessively</i>: once it
+has matched it does not backtrack. They behave like greedy quantifiers,
+but having matched they refuse to "give up" their match even if this
+jeopardises the overall match.
+
+== Capturing
+
+Parentheses can be used for <i>capturing</i>. The text enclosed by the
+<i>n</i><sup>th</sup> group of parentheses can be subsequently referred to
+with <i>n</i>. Within a pattern use the <i>backreference</i>
+<tt>\n</tt>; outside of the pattern use
+<tt>MatchData[</tt><i>n</i><tt>]</tt>.
+
+'at' is captured by the first group of parentheses, then referred to later
+with <tt>\1</tt>:
+
+    /[csh](..) [csh]\1 in/.match("The cat sat in the hat")
+        #=> #<MatchData "cat sat in" 1:"at">
+
+Regexp#match returns a MatchData object which makes the captured text
+available with its #[] method:
+
+    /[csh](..) [csh]\1 in/.match("The cat sat in the hat")[1] #=> 'at'
+
+Capture groups can be referred to by name when defined with the
+<tt>(?<</tt><i>name</i><tt>>)</tt> or <tt>(?'</tt><i>name</i><tt>')</tt>
+constructs.
+
+    /\$(?<dollars>\d+)\.(?<cents>\d+)/.match("$3.67")
+        #=> #<MatchData "$3.67" dollars:"3" cents:"67">
+    /\$(?<dollars>\d+)\.(?<cents>\d+)/.match("$3.67")[:dollars] #=> "3"
+
+Named groups can be backreferenced with <tt>\k<</tt><i>name</i><tt>></tt>,
+where _name_ is the group name.
+
+    /(?<vowel>[aeiou]).\k<vowel>.\k<vowel>/.match('ototomy')
+        #=> #<MatchData "ototo" vowel:"o">
+
+*Note*: A regexp can't use named backreferences and numbered
+backreferences simultaneously.
+
+When named capture groups are used with a literal regexp on the left-hand
+side of an expression and the <tt>=~</tt> operator, the captured text is
+also assigned to local variables with corresponding names.
+
+    /\$(?<dollars>\d+)\.(?<cents>\d+)/ =~ "$3.67" #=> 0
+    dollars #=> "3"
+
+== Grouping
+
+Parentheses also <i>group</i> the terms they enclose, allowing them to be
+quantified as one <i>atomic</i> whole.
+
+The pattern below matches a vowel followed by 2 word characters:
+
+    /[aeiou]\w{2}/.match("Caenorhabditis elegans") #=> #<MatchData "aen">
+
+Whereas the following pattern matches a vowel followed by a word character,
+twice, i.e. <tt>[aeiou]\w[aeiou]\w</tt>: 'enor'.
+
+    /([aeiou]\w){2}/.match("Caenorhabditis elegans")
+        #=> #<MatchData "enor" 1:"or">
+
+The <tt>(?:</tt>...<tt>)</tt> construct provides grouping without
+capturing. That is, it combines the terms it contains into an atomic whole
+without creating a backreference. This benefits performance at the slight
+expense of readability.
+
+The first group of parentheses captures 'n' and the second 'ti'. The second
+group is referred to later with the backreference <tt>\2</tt>:
+
+    /I(n)ves(ti)ga\2ons/.match("Investigations")
+        #=> #<MatchData "Investigations" 1:"n" 2:"ti">
+
+The first group of parentheses is now made non-capturing with '?:', so it
+still matches 'n', but doesn't create the backreference. Thus, the
+backreference <tt>\1</tt> now refers to 'ti'.
+
+    /I(?:n)ves(ti)ga\1ons/.match("Investigations")
+        #=> #<MatchData "Investigations" 1:"ti">
+
+=== Atomic Grouping
+
+Grouping can be made <i>atomic</i> with
+<tt>(?></tt><i>pat</i><tt>)</tt>. This causes the subexpression <i>pat</i>
+to be matched independently of the rest of the expression such that what
+it matches becomes fixed for the remainder of the match, unless the entire
+subexpression must be abandoned and subsequently revisited. In this
+way <i>pat</i> is treated as a non-divisible whole. Atomic grouping is
+typically used to optimise patterns so as to prevent the regular
+expression engine from backtracking needlessly.
+
+The <tt>"</tt> in the pattern below matches the first character of the string,
+then <tt>.*</tt> matches <i>Quote"</i>. This causes the overall match to fail,
+so the text matched by <tt>.*</tt> is backtracked by one position, which
+leaves the final character of the string available to match <tt>"</tt>
+
+          /".*"/.match('"Quote"')     #=> #<MatchData "\"Quote\"">
+
+If <tt>.*</tt> is grouped atomically, it refuses to backtrack <i>Quote"</i>,
+even though this means that the overall match fails
+
+    /"(?>.*)"/.match('"Quote"') #=> nil
+
+== Subexpression Calls
+
+The <tt>\g<</tt><i>name</i><tt>></tt> syntax matches the previous
+subexpression named _name_, which can be a group name or number, again.
+This differs from backreferences in that it re-executes the group rather
+than simply trying to re-match the same text.
+
+This pattern matches a <i>(</i> character and assigns it to the <tt>paren</tt>
+group, tries to call that the <tt>paren</tt> sub-expression again but fails,
+then matches a literal <i>)</i>:
+
+    /\A(?<paren>\(\g<paren>*\))*\z/ =~ '()'
+
+
+    /\A(?<paren>\(\g<paren>*\))*\z/ =~ '(())' #=> 0
+    # ^1
+    #      ^2
+    #           ^3
+    #                 ^4
+    #      ^5
+    #           ^6
+    #                      ^7
+    #                       ^8
+    #                       ^9
+    #                           ^10
+
+1.  Matches at the beginning of the string, i.e. before the first
+    character.
+2.  Enters a named capture group called <tt>paren</tt>
+3.  Matches a literal <i>(</i>, the first character in the string
+4.  Calls the <tt>paren</tt> group again, i.e. recurses back to the
+    second step
+5.  Re-enters the <tt>paren</tt> group
+6.  Matches a literal <i>(</i>, the second character in the
+    string
+7.  Try to call <tt>paren</tt> a third time, but fail because
+    doing so would prevent an overall successful match
+8.  Match a literal <i>)</i>, the third character in the string.
+    Marks the end of the second recursive call
+9.  Match a literal <i>)</i>, the fourth character in the string
+10. Match the end of the string
+
+== Alternation
+
+The vertical bar metacharacter (<tt>|</tt>) combines two expressions into
+a single one that matches either of the expressions. Each expression is an
+<i>alternative</i>.
+
+    /\w(and|or)\w/.match("Feliformia") #=> #<MatchData "form" 1:"or">
+    /\w(and|or)\w/.match("furandi")    #=> #<MatchData "randi" 1:"and">
+    /\w(and|or)\w/.match("dissemblance") #=> nil
+
+== Character Properties
+
+The <tt>\p{}</tt> construct matches characters with the named property,
+much like POSIX bracket classes.
+
+* <tt>/\p{Alnum}/</tt> - Alphabetic and numeric character
+* <tt>/\p{Alpha}/</tt> - Alphabetic character
+* <tt>/\p{Blank}/</tt> - Space or tab
+* <tt>/\p{Cntrl}/</tt> - Control character
+* <tt>/\p{Digit}/</tt> - Digit
+* <tt>/\p{Graph}/</tt> - Non-blank character (excludes spaces, control
+  characters, and similar)
+* <tt>/\p{Lower}/</tt> - Lowercase alphabetical character
+* <tt>/\p{Print}/</tt> - Like <tt>\p{Graph}</tt>, but includes the space character
+* <tt>/\p{Punct}/</tt> - Punctuation character
+* <tt>/\p{Space}/</tt> - Whitespace character (<tt>[:blank:]</tt>, newline,
+  carriage return, etc.)
+* <tt>/\p{Upper}/</tt> - Uppercase alphabetical
+* <tt>/\p{XDigit}/</tt> - Digit allowed in a hexadecimal number (i.e., 0-9a-fA-F)
+* <tt>/\p{Word}/</tt> - A member of one of the following Unicode general
+  category <i>Letter</i>, <i>Mark</i>, <i>Number</i>,
+  <i>Connector\_Punctuation</i>
+* <tt>/\p{ASCII}/</tt> - A character in the ASCII character set
+* <tt>/\p{Any}/</tt> - Any Unicode character (including unassigned
+  characters)
+* <tt>/\p{Assigned}/</tt> - An assigned character
+
+A Unicode character's <i>General Category</i> value can also be matched
+with <tt>\p{</tt><i>Ab</i><tt>}</tt> where <i>Ab</i> is the category's
+abbreviation as described below:
+
+* <tt>/\p{L}/</tt> - 'Letter'
+* <tt>/\p{Ll}/</tt> - 'Letter: Lowercase'
+* <tt>/\p{Lm}/</tt> - 'Letter: Mark'
+* <tt>/\p{Lo}/</tt> - 'Letter: Other'
+* <tt>/\p{Lt}/</tt> - 'Letter: Titlecase'
+* <tt>/\p{Lu}/</tt> - 'Letter: Uppercase
+* <tt>/\p{Lo}/</tt> - 'Letter: Other'
+* <tt>/\p{M}/</tt> - 'Mark'
+* <tt>/\p{Mn}/</tt> - 'Mark: Nonspacing'
+* <tt>/\p{Mc}/</tt> - 'Mark: Spacing Combining'
+* <tt>/\p{Me}/</tt> - 'Mark: Enclosing'
+* <tt>/\p{N}/</tt> - 'Number'
+* <tt>/\p{Nd}/</tt> - 'Number: Decimal Digit'
+* <tt>/\p{Nl}/</tt> - 'Number: Letter'
+* <tt>/\p{No}/</tt> - 'Number: Other'
+* <tt>/\p{P}/</tt> - 'Punctuation'
+* <tt>/\p{Pc}/</tt> - 'Punctuation: Connector'
+* <tt>/\p{Pd}/</tt> - 'Punctuation: Dash'
+* <tt>/\p{Ps}/</tt> - 'Punctuation: Open'
+* <tt>/\p{Pe}/</tt> - 'Punctuation: Close'
+* <tt>/\p{Pi}/</tt> - 'Punctuation: Initial Quote'
+* <tt>/\p{Pf}/</tt> - 'Punctuation: Final Quote'
+* <tt>/\p{Po}/</tt> - 'Punctuation: Other'
+* <tt>/\p{S}/</tt> - 'Symbol'
+* <tt>/\p{Sm}/</tt> - 'Symbol: Math'
+* <tt>/\p{Sc}/</tt> - 'Symbol: Currency'
+* <tt>/\p{Sc}/</tt> - 'Symbol: Currency'
+* <tt>/\p{Sk}/</tt> - 'Symbol: Modifier'
+* <tt>/\p{So}/</tt> - 'Symbol: Other'
+* <tt>/\p{Z}/</tt> - 'Separator'
+* <tt>/\p{Zs}/</tt> - 'Separator: Space'
+* <tt>/\p{Zl}/</tt> - 'Separator: Line'
+* <tt>/\p{Zp}/</tt> - 'Separator: Paragraph'
+* <tt>/\p{C}/</tt> - 'Other'
+* <tt>/\p{Cc}/</tt> - 'Other: Control'
+* <tt>/\p{Cf}/</tt> - 'Other: Format'
+* <tt>/\p{Cn}/</tt> - 'Other: Not Assigned'
+* <tt>/\p{Co}/</tt> - 'Other: Private Use'
+* <tt>/\p{Cs}/</tt> - 'Other: Surrogate'
+
+Lastly, <tt>\p{}</tt> matches a character's Unicode <i>script</i>. The
+following scripts are supported: <i>Arabic</i>, <i>Armenian</i>,
+<i>Balinese</i>, <i>Bengali</i>, <i>Bopomofo</i>, <i>Braille</i>,
+<i>Buginese</i>, <i>Buhid</i>, <i>Canadian_Aboriginal</i>, <i>Carian</i>,
+<i>Cham</i>, <i>Cherokee</i>, <i>Common</i>, <i>Coptic</i>,
+<i>Cuneiform</i>, <i>Cypriot</i>, <i>Cyrillic</i>, <i>Deseret</i>,
+<i>Devanagari</i>, <i>Ethiopic</i>, <i>Georgian</i>, <i>Glagolitic</i>,
+<i>Gothic</i>, <i>Greek</i>, <i>Gujarati</i>, <i>Gurmukhi</i>, <i>Han</i>,
+<i>Hangul</i>, <i>Hanunoo</i>, <i>Hebrew</i>, <i>Hiragana</i>,
+<i>Inherited</i>, <i>Kannada</i>, <i>Katakana</i>, <i>Kayah_Li</i>,
+<i>Kharoshthi</i>, <i>Khmer</i>, <i>Lao</i>, <i>Latin</i>, <i>Lepcha</i>,
+<i>Limbu</i>, <i>Linear_B</i>, <i>Lycian</i>, <i>Lydian</i>,
+<i>Malayalam</i>, <i>Mongolian</i>, <i>Myanmar</i>, <i>New_Tai_Lue</i>,
+<i>Nko</i>, <i>Ogham</i>, <i>Ol_Chiki</i>, <i>Old_Italic</i>,
+<i>Old_Persian</i>, <i>Oriya</i>, <i>Osmanya</i>, <i>Phags_Pa</i>,
+<i>Phoenician</i>, <i>Rejang</i>, <i>Runic</i>, <i>Saurashtra</i>,
+<i>Shavian</i>, <i>Sinhala</i>, <i>Sundanese</i>, <i>Syloti_Nagri</i>,
+<i>Syriac</i>, <i>Tagalog</i>, <i>Tagbanwa</i>, <i>Tai_Le</i>,
+<i>Tamil</i>, <i>Telugu</i>, <i>Thaana</i>, <i>Thai</i>, <i>Tibetan</i>,
+<i>Tifinagh</i>, <i>Ugaritic</i>, <i>Vai</i>, and <i>Yi</i>.
+
+Unicode codepoint U+06E9 is named "ARABIC PLACE OF SAJDAH" and belongs to the
+Arabic script:
+
+    /\p{Arabic}/.match("\u06E9") #=> #<MatchData "\u06E9">
+
+All character properties can be inverted by prefixing their name with a
+caret (<tt>^</tt>).
+
+Letter 'A' is not in the Unicode Ll (Letter; Lowercase) category, so this
+match succeeds:
+
+    /\p{^Ll}/.match("A") #=> #<MatchData "A">
+
+== Anchors
+
+Anchors are metacharacter that match the zero-width positions between
+characters, <i>anchoring</i> the match to a specific position.
+
+* <tt>^</tt> - Matches beginning of line
+* <tt>$</tt> - Matches end of line
+* <tt>\A</tt> - Matches beginning of string.
+* <tt>\Z</tt> - Matches end of string. If string ends with a newline,
+  it matches just before newline
+* <tt>\z</tt> - Matches end of string
+* <tt>\G</tt> - Matches first matching position:
+
+  In methods like <tt>String#gsub</tt> and <tt>String#scan</tt>, it changes on each iteration.
+  It initially matches the beginning of subject, and in each following iteration it matches where the last match finished.
+
+      "    a b c".gsub(/ /, '_')    #=> "____a_b_c"
+      "    a b c".gsub(/\G /, '_')  #=> "____a b c"
+
+  In methods like <tt>Regexp#match</tt> and <tt>String#match</tt> that take an (optional) offset, it matches where the search begins.
+
+      "hello, world".match(/,/, 3)    #=> #<MatchData ",">
+      "hello, world".match(/\G,/, 3)  #=> nil
+
+* <tt>\b</tt> - Matches word boundaries when outside brackets;
+  backspace (0x08) when inside brackets
+* <tt>\B</tt> - Matches non-word boundaries
+* <tt>(?=</tt><i>pat</i><tt>)</tt> - <i>Positive lookahead</i> assertion:
+  ensures that the following characters match <i>pat</i>, but doesn't
+  include those characters in the matched text
+* <tt>(?!</tt><i>pat</i><tt>)</tt> - <i>Negative lookahead</i> assertion:
+  ensures that the following characters do not match <i>pat</i>, but
+  doesn't include those characters in the matched text
+* <tt>(?<=</tt><i>pat</i><tt>)</tt> - <i>Positive lookbehind</i>
+  assertion: ensures that the preceding characters match <i>pat</i>, but
+  doesn't include those characters in the matched text
+* <tt>(?<!</tt><i>pat</i><tt>)</tt> - <i>Negative lookbehind</i>
+  assertion: ensures that the preceding characters do not match
+  <i>pat</i>, but doesn't include those characters in the matched text
+
+If a pattern isn't anchored it can begin at any point in the string:
+
+    /real/.match("surrealist") #=> #<MatchData "real">
+
+Anchoring the pattern to the beginning of the string forces the match to start
+there. 'real' doesn't occur at the beginning of the string, so now the match
+fails:
+
+    /\Areal/.match("surrealist") #=> nil
+
+The match below fails because although 'Demand' contains 'and', the pattern
+does not occur at a word boundary.
+
+    /\band/.match("Demand")
+
+Whereas in the following example 'and' has been anchored to a non-word
+boundary so instead of matching the first 'and' it matches from the fourth
+letter of 'demand' instead:
+
+    /\Band.+/.match("Supply and demand curve") #=> #<MatchData "and curve">
+
+The pattern below uses positive lookahead and positive lookbehind to match
+text appearing in <b></b> tags without including the tags in the match:
+
+    /(?<=<b>)\w+(?=<\/b>)/.match("Fortune favours the <b>bold</b>")
+        #=> #<MatchData "bold">
+
+== Options
+
+The end delimiter for a regexp can be followed by one or more single-letter
+options which control how the pattern can match.
+
+* <tt>/pat/i</tt> - Ignore case
+* <tt>/pat/m</tt> - Treat a newline as a character matched by <tt>.</tt>
+* <tt>/pat/x</tt> - Ignore whitespace and comments in the pattern
+* <tt>/pat/o</tt> - Perform <tt>#{}</tt> interpolation only once
+
+<tt>i</tt>, <tt>m</tt>, and <tt>x</tt> can also be applied on the
+subexpression level with the
+<tt>(?</tt><i>on</i><tt>-</tt><i>off</i><tt>)</tt> construct, which
+enables options <i>on</i>, and disables options <i>off</i> for the
+expression enclosed by the parentheses:
+
+    /a(?i:b)c/.match('aBc')   #=> #<MatchData "aBc">
+    /a(?-i:b)c/i.match('ABC') #=> nil
+
+Additionally, these options can also be toggled for the remainder of the
+pattern:
+
+    /a(?i)bc/.match('abC') #=> #<MatchData "abC">
+
+Options may also be used with <tt>Regexp.new</tt>:
+
+    Regexp.new("abc", Regexp::IGNORECASE)                     #=> /abc/i
+    Regexp.new("abc", Regexp::MULTILINE)                      #=> /abc/m
+    Regexp.new("abc # Comment", Regexp::EXTENDED)             #=> /abc # Comment/x
+    Regexp.new("abc", Regexp::IGNORECASE | Regexp::MULTILINE) #=> /abc/mi
+
+== Free-Spacing Mode and Comments
+
+As mentioned above, the <tt>x</tt> option enables <i>free-spacing</i>
+mode. Literal white space inside the pattern is ignored, and the
+octothorpe (<tt>#</tt>) character introduces a comment until the end of
+the line. This allows the components of the pattern to be organized in a
+potentially more readable fashion.
+
+A contrived pattern to match a number with optional decimal places:
+
+    float_pat = /\A
+        [[:digit:]]+ # 1 or more digits before the decimal point
+        (\.          # Decimal point
+            [[:digit:]]+ # 1 or more digits after the decimal point
+        )? # The decimal point and following digits are optional
+    \Z/x
+    float_pat.match('3.14') #=> #<MatchData "3.14" 1:".14">
+
+There are a number of strategies for matching whitespace:
+
+* Use a pattern such as <tt>\s</tt> or <tt>\p{Space}</tt>.
+* Use escaped whitespace such as <tt>\ </tt>, i.e. a space preceded by a backslash.
+* Use a character class such as <tt>[ ]</tt>.
+
+Comments can be included in a non-<tt>x</tt> pattern with the
+<tt>(?#</tt><i>comment</i><tt>)</tt> construct, where <i>comment</i> is
+arbitrary text ignored by the regexp engine.
+
+Comments in regexp literals cannot include unescaped terminator
+characters.
+
+== Encoding
+
+Regular expressions are assumed to use the source encoding. This can be
+overridden with one of the following modifiers.
+
+* <tt>/</tt><i>pat</i><tt>/u</tt> - UTF-8
+* <tt>/</tt><i>pat</i><tt>/e</tt> - EUC-JP
+* <tt>/</tt><i>pat</i><tt>/s</tt> - Windows-31J
+* <tt>/</tt><i>pat</i><tt>/n</tt> - ASCII-8BIT
+
+A regexp can be matched against a string when they either share an
+encoding, or the regexp's encoding is _US-ASCII_ and the string's encoding
+is ASCII-compatible.
+
+If a match between incompatible encodings is attempted an
+<tt>Encoding::CompatibilityError</tt> exception is raised.
+
+The <tt>Regexp#fixed_encoding?</tt> predicate indicates whether the regexp
+has a <i>fixed</i> encoding, that is one incompatible with ASCII. A
+regexp's encoding can be explicitly fixed by supplying
+<tt>Regexp::FIXEDENCODING</tt> as the second argument of
+<tt>Regexp.new</tt>:
+
+    r = Regexp.new("a".force_encoding("iso-8859-1"),Regexp::FIXEDENCODING)
+    r =~ "a\u3042"
+       # raises Encoding::CompatibilityError: incompatible encoding regexp match
+       #         (ISO-8859-1 regexp with UTF-8 string)
+
+== Special global variables
+
+Pattern matching sets some global variables :
+* <tt>$~</tt> is equivalent to Regexp.last_match;
+* <tt>$&</tt> contains the complete matched text;
+* <tt>$`</tt> contains string before match;
+* <tt>$'</tt> contains string after match;
+* <tt>$1</tt>, <tt>$2</tt> and so on contain text matching first, second, etc
+  capture group;
+* <tt>$+</tt> contains last capture group.
+
+Example:
+
+    m = /s(\w{2}).*(c)/.match('haystack') #=> #<MatchData "stac" 1:"ta" 2:"c">
+    $~                                    #=> #<MatchData "stac" 1:"ta" 2:"c">
+    Regexp.last_match                     #=> #<MatchData "stac" 1:"ta" 2:"c">
+
+    $&      #=> "stac"
+            # same as m[0]
+    $`      #=> "hay"
+            # same as m.pre_match
+    $'      #=> "k"
+            # same as m.post_match
+    $1      #=> "ta"
+            # same as m[1]
+    $2      #=> "c"
+            # same as m[2]
+    $3      #=> nil
+            # no third group in pattern
+    $+      #=> "c"
+            # same as m[-1]
+
+These global variables are thread-local and method-local variables.
+
+== Performance
+
+Certain pathological combinations of constructs can lead to abysmally bad
+performance.
+
+Consider a string of 25 <i>a</i>s, a <i>d</i>, 4 <i>a</i>s, and a
+<i>c</i>.
+
+    s = 'a' * 25 + 'd' + 'a' * 4 + 'c'
+    #=> "aaaaaaaaaaaaaaaaaaaaaaaaadaaaac"
+
+The following patterns match instantly as you would expect:
+
+    /(b|a)/ =~ s #=> 0
+    /(b|a+)/ =~ s #=> 0
+    /(b|a+)*/ =~ s #=> 0
+
+However, the following pattern takes appreciably longer:
+
+    /(b|a+)*c/ =~ s #=> 26
+
+This happens because an atom in the regexp is quantified by both an
+immediate <tt>+</tt> and an enclosing <tt>*</tt> with nothing to
+differentiate which is in control of any particular character. The
+nondeterminism that results produces super-linear performance. (Consult
+<i>Mastering Regular Expressions</i> (3rd ed.), pp 222, by
+<i>Jeffery Friedl</i>, for an in-depth analysis). This particular case
+can be fixed by use of atomic grouping, which prevents the unnecessary
+backtracking:
+
+    (start = Time.now) && /(b|a+)*c/ =~ s && (Time.now - start)
+       #=> 24.702736882
+    (start = Time.now) && /(?>b|a+)*c/ =~ s && (Time.now - start)
+       #=> 0.000166571
+
+A similar case is typified by the following example, which takes
+approximately 60 seconds to execute for me:
+
+Match a string of 29 <i>a</i>s against a pattern of 29 optional <i>a</i>s
+followed by 29 mandatory <i>a</i>s:
+
+    Regexp.new('a?' * 29 + 'a' * 29) =~ 'a' * 29
+
+The 29 optional <i>a</i>s match the string, but this prevents the 29
+mandatory <i>a</i>s that follow from matching. Ruby must then backtrack
+repeatedly so as to satisfy as many of the optional matches as it can
+while still matching the mandatory 29. It is plain to us that none of the
+optional matches can succeed, but this fact unfortunately eludes Ruby.
+
+The best way to improve performance is to significantly reduce the amount of
+backtracking needed.  For this case, instead of individually matching 29
+optional <i>a</i>s, a range of optional <i>a</i>s can be matched all at once
+with <i>a{0,29}</i>:
+
+    Regexp.new('a{0,29}' + 'a' * 29) =~ 'a' * 29
+
diff --git a/doc/security.rdoc b/doc/security.rdoc
new file mode 100644
index 0000000000..d7d6464ce1
--- /dev/null
+++ b/doc/security.rdoc
@@ -0,0 +1,152 @@
+= Ruby Security
+
+The Ruby programming language is large and complex and there are many security
+pitfalls often encountered by newcomers and experienced Rubyists alike.
+
+This document aims to discuss many of these pitfalls and provide more secure
+alternatives where applicable.
+
+Please check the full list of publicly known CVEs and how to correctly report a
+security vulnerability, at: https://www.ruby-lang.org/en/security/
+Japanese version is here: https://www.ruby-lang.org/ja/security/
+
+Security vulnerabilities should be reported via an email to
+mailto:security@ruby-lang.org ({the PGP public
+key}[https://www.ruby-lang.org/security.asc]), which is a private mailing list.
+Reported problems will be published after fixes.
+
+== <code>$SAFE</code>
+
+Ruby provides a mechanism to restrict what operations can be performed by Ruby
+code in the form of the <code>$SAFE</code> variable.
+
+However, <code>$SAFE</code> does not provide a secure environment for executing
+untrusted code.
+
+If you need to execute untrusted code, you should use an operating system level
+sandboxing mechanism. On Linux, ptrace or LXC can be used to sandbox
+potentially malicious code. Other similar mechanisms exist on every major
+operating system.
+
+== +Marshal.load+
+
+Ruby's +Marshal+ module provides methods for serializing and deserializing Ruby
+object trees to and from a binary data format.
+
+Never use +Marshal.load+ to deserialize untrusted or user supplied data.
+Because +Marshal+ can deserialize to almost any Ruby object and has full
+control over instance variables, it is possible to craft a malicious payload
+that executes code shortly after deserialization.
+
+If you need to deserialize untrusted data, you should use JSON as it is only
+capable of returning 'primitive' types such as strings, arrays, hashes, numbers
+and nil. If you need to deserialize other classes, you should handle this
+manually. Never deserialize to a user specified class.
+
+== YAML
+
+YAML is a popular human readable data serialization format used by many Ruby
+programs for configuration and database persistence of Ruby object trees.
+
+Similar to +Marshal+, it is able to deserialize into arbitrary Ruby classes.
+For example, the following YAML data will create an +ERB+ object when
+deserialized:
+
+  !ruby/object:ERB
+  src: puts `uname`
+
+Because of this, many of the security considerations applying to Marshal are
+also applicable to YAML. Do not use YAML to deserialize untrusted data.
+
+== Symbols
+
+Symbols are often seen as syntax sugar for simple strings, but they play a much
+more crucial role. The MRI Ruby implementation uses Symbols internally for
+method, variable and constant names. The reason for this is that symbols are
+simply integers with names attached to them, so they are faster to look up in
+hashtables.
+
+Starting in version 2.2, most symbols can be garbage collected; these are
+called <i>mortal</i> symbols. Most symbols you create (e.g. by calling
++to_sym+) are mortal.
+
+<i>Immortal</i> symbols on the other hand will never be garbage collected.
+They are created when modifying code:
+* defining a method (e.g. with +define_method+),
+* setting an instance variable (e.g. with +instance_variable_set+),
+* creating a variable or constant (e.g. with +const_set+)
+C extensions that have not been updated and are still calling `SYM2ID`
+will create immortal symbols.
+Bugs in 2.2.0: +send+ and +__send__+ also created immortal symbols,
+and calling methods with keyword arguments could also create some.
+
+Don't create immortal symbols from user inputs. Otherwise, this would
+allow a user to mount a denial of service attack against your application by
+flooding it with unique strings, which will cause memory to grow indefinitely
+until the Ruby process is killed or causes the system to slow to a halt.
+
+While it might not be a good idea to call these with user inputs, methods that
+used to be vulnerable such as +to_sym+, +respond_to?+,
++method+, +instance_variable_get+, +const_get+, etc. are no longer a threat.
+
+== Regular expressions
+
+Ruby's regular expression syntax has some minor differences when compared to
+other languages. In Ruby, the <code>^</code> and <code>$</code> anchors do not
+refer to the beginning and end of the string, rather the beginning and end of a
+*line*.
+
+This means that if you're using a regular expression like
+<code>/^[a-z]+$/</code> to restrict a string to only letters, an attacker can
+bypass this check by passing a string containing a letter, then a newline, then
+any string of their choosing.
+
+If you want to match the beginning and end of the entire string in Ruby, use
+the anchors +\A+ and +\z+.
+
+== +eval+
+
+Never pass untrusted or user controlled input to +eval+.
+
+Unless you are implementing a REPL like +irb+ or +pry+, +eval+ is almost
+certainly not what you want. Do not attempt to filter user input before passing
+it to +eval+ - this approach is fraught with danger and will most likely open
+your application up to a serious remote code execution vulnerability.
+
+== +send+
+
+'Global functions' in Ruby (+puts+, +exit+, etc.) are actually private instance
+methods on +Object+. This means it is possible to invoke these methods with
++send+, even if the call to +send+ has an explicit receiver.
+
+For example, the following code snippet writes "Hello world" to the terminal:
+
+  1.send(:puts, "Hello world")
+
+You should never call +send+ with user supplied input as the first parameter.
+Doing so can introduce a denial of service vulnerability:
+
+  foo.send(params[:bar]) # params[:bar] is "exit!"
+
+If an attacker can control the first two arguments to +send+, remote code
+execution is possible:
+
+  # params is { :a => "eval", :b => "...ruby code to be executed..." }
+  foo.send(params[:a], params[:b])
+
+When dispatching a method call based on user input, carefully verify that the
+method name. If possible, check it against a whitelist of safe method names.
+
+Note that the use of +public_send+ is also dangerous, as +send+ itself is
+public:
+
+  1.public_send("send", "eval", "...ruby code to be executed...")
+
+== DRb
+
+As DRb allows remote clients to invoke arbitrary methods, it is not suitable to
+expose to untrusted clients.
+
+When using DRb, try to avoid exposing it over the network if possible. If this
+isn't possible and you need to expose DRb to the world, you *must* configure an
+appropriate security policy with <code>DRb::ACL</code>.
diff --git a/doc/shell.rd b/doc/shell.rd
deleted file mode 100644
index ae6855cbd8..0000000000
--- a/doc/shell.rd
+++ /dev/null
@@ -1,348 +0,0 @@
- -- shell.rb
-				$Release Version: 0.6.0 $
-			   	$Revision: 1.2 $
-			   	$Date: 2001/05/17 10:09:49 $
-			   	by Keiju ISHITSUKA(keiju@ishitsuka.com)
-
-=begin
-
-= What's shell.rb?
-
-It realizes a wish to do execution of commands with filters and pipes
-like sh/csh by using just native facilities of ruby.
-
-= Main classes
-
-== Shell
-
-Every shell object has its own current working directory, and executes
-each command as if it stands in the directory.
-
---- Shell#cwd
---- Shell#dir
---- Shell#getwd
---- Shell#pwd
-
-      Returns the current directory
-
---- Shell#system_path
-
-      Returns the command search path in an array
-
---- Shell#umask
-
-      Returns the umask
-
-== Filter
-
-Any result of command exection is a Filter.  Filter include
-Enumerable, therefore a Filter object can use all Enumerable
-facilities.
-
-= Main methods
-
-== Command definitions
-
-In order to execute a command on your OS, you need to define it as a
-Shell method.
-
-Alternatively, you can execute any command via Shell#system even if it
-is not defined.
-
---- Shell.def_system_command(command, path = command)
-
-      Defines a command.  Registers <path> as a Shell method
-      <command>.
-
-      ex)
-      Shell.def_system_command "ls"
-        Defines ls.
-
-      Shell.def_system_command "sys_sort", "sort"
-        Defines sys_sort as sort.
-
---- Shell.undef_system_command(command)
-
-      Undefines a commmand
-
---- Shell.alias_command(ali, command, *opts) {...}
-
-      Aliases a command.
-
-      ex)
-        Shell.alias_command "lsC", "ls", "-CBF", "--show-control-chars"
-        Shell.alias_command("lsC", "ls"){|*opts| ["-CBF", "--show-control-chars", *opts]}
-
---- Shell.unalias_command(ali)
-
-      Unaliases a command.
-
---- Shell.install_system_commands(pre = "sys_")
-
-      Defines all commands in the default_system_path as Shell method,
-      all with <pre> prefixed to their names.
-
-== Creation
-
---- Shell.new
-
-      Creates a Shell object which current directory is set to the
-      process current directory.
-
---- Shell.cd(path)
-
-      Creates a Shell object which current directory is set to
-      <path>.
-
-== Process management
-
---- Shell#jobs
-
-      Returns a list of scheduled jobs.
-
---- Shell#kill sig, job
-
-      Sends a signal <sig> to <job>.
-
-== Current directory operations
-
---- Shell#cd(path, &block)
---- Shell#chdir
-
-      Changes the current directory to <path>.  If a block is given,
-      it restores the current directory when the block ends.
-
---- Shell#pushd(path = nil, &block)
---- Shell#pushdir
-
-      Pushes the current directory to the directory stack, changing
-      the current directory to <path>.  If <path> is omitted, it
-      exchanges its current directory and the top of its directory
-      stack.  If a block is given, it restores the current directory
-      when the block ends.
-
---- Shell#popd
---- Shell#popdir
-
-      Pops a directory from the directory stack, and sets the current
-      directory to it.
-
-== File and directory operations
-
---- Shell#foreach(path = nil, &block)
-
-      Same as:
-        File#foreach (when path is a file)
-        Dir#foreach (when path is a directory)
-
---- Shell#open(path, mode)
-
-      Same as:
-        File#open (when path is a file)
-        Dir#open (when path is a directory)
-
---- Shell#unlink(path)
-
-      Same as:
-        Dir#open (when path is a file)
-        Dir#unlink (when path is a directory)
-
---- Shell#test(command, file1, file2)
---- Shell#[command, file1, file2]
-
-      Same as test().
-      ex)
-          sh[?e, "foo"]
-          sh[:e, "foo"]
-          sh["e", "foo"]
-          sh[:exists?, "foo"]
-          sh["exists?", "foo"]
-
---- Shell#mkdir(*path)
-
-      Same as Dir.mkdir (with multiple directories allowed)
-
---- Shell#rmdir(*path)
-
-      Same as Dir.rmdir (with multiple directories allowed)
-
-== Command execution
-
---- System#system(command, *opts)
-
-      Executes <command> with <opts>.
-
-      ex)
-        print sh.system("ls", "-l")
-        sh.system("ls", "-l") | sh.head > STDOUT
-
---- System#rehash
-
-      Does rehash.
-
---- Shell#transact &block
-
-      Executes a block as self.
-      ex)
-        sh.transact{system("ls", "-l") | head > STDOUT}
-
---- Shell#out(dev = STDOUT, &block)
-
-      Does transact, with redirecting the result output to <dev>.
-
-== Internal commands
-
---- Shell#echo(*strings)
---- Shell#cat(*files)
---- Shell#glob(patten)
---- Shell#tee(file)
-
-      Return Filter objects, which are results of their execution.
-
---- Filter#each &block
-
-      Iterates a block for each line of it.
-
---- Filter#<(src)
-
-      Inputs from <src>, which is either a string of a file name or an
-      IO.
-
---- Filter#>(to)
-
-      Outputs to <to>, which is either a string of a file name or an
-      IO.
-
---- Filter#>>(to)
-
-      Appends the ouput to <to>, which is either a string of a file
-      name or an IO.
-
---- Filter#|(filter)
-
-      Processes a pipeline.
-
---- Filter#+(filter)
-
-      (filter1 + filter2) outputs filter1, and then outputs filter2.
-
---- Filter#to_a
---- Filter#to_s
-
-== Built-in commands
-
---- Shell#atime(file)
---- Shell#basename(file, *opt)
---- Shell#chmod(mode, *files)
---- Shell#chown(owner, group, *file)
---- Shell#ctime(file)
---- Shell#delete(*file)
---- Shell#dirname(file)
---- Shell#ftype(file)
---- Shell#join(*file)
---- Shell#link(file_from, file_to)
---- Shell#lstat(file)
---- Shell#mtime(file)
---- Shell#readlink(file)
---- Shell#rename(file_from, file_to)
---- Shell#split(file)
---- Shell#stat(file)
---- Shell#symlink(file_from, file_to)
---- Shell#truncate(file, length)
---- Shell#utime(atime, mtime, *file)
-
-      Equivalent to the class methods of File with the same names.
-
---- Shell#blockdev?(file)
---- Shell#chardev?(file)
---- Shell#directory?(file)
---- Shell#executable?(file)
---- Shell#executable_real?(file)
---- Shell#exist?(file)/Shell#exists?(file)
---- Shell#file?(file)
---- Shell#grpowned?(file)
---- Shell#owned?(file)
---- Shell#pipe?(file)
---- Shell#readable?(file)
---- Shell#readable_real?(file)
---- Shell#setgid?(file)
---- Shell#setuid?(file)
---- Shell#size(file)/Shell#size?(file)
---- Shell#socket?(file)
---- Shell#sticky?(file)
---- Shell#symlink?(file)
---- Shell#writable?(file)
---- Shell#writable_real?(file)
---- Shell#zero?(file)
-
-      Equivalent to the class methods of FileTest with the same names.
-
---- Shell#syscopy(filename_from, filename_to)
---- Shell#copy(filename_from, filename_to)
---- Shell#move(filename_from, filename_to)
---- Shell#compare(filename_from, filename_to)
---- Shell#safe_unlink(*filenames)
---- Shell#makedirs(*filenames)
---- Shell#install(filename_from, filename_to, mode)
-
-      Equivalent to the class methods of FileTools with the same
-      names.
-
-      And also, there are some aliases for convenience:
-
---- Shell#cmp	<- Shell#compare
---- Shell#mv	<- Shell#move
---- Shell#cp	<- Shell#copy
---- Shell#rm_f	<- Shell#safe_unlink
---- Shell#mkpath	<- Shell#makedirs
-
-= Samples
-
-== ex1
-
-  sh = Shell.cd("/tmp")
-  sh.mkdir "shell-test-1" unless sh.exists?("shell-test-1")
-  sh.cd("shell-test-1")
-  for dir in ["dir1", "dir3", "dir5"]
-    if !sh.exists?(dir)
-      sh.mkdir dir
-      sh.cd(dir) do
-	f = sh.open("tmpFile", "w")
-	f.print "TEST\n"
-	f.close
-      end
-      print sh.pwd
-    end
-  end
-
-== ex2
-
-  sh = Shell.cd("/tmp")
-  sh.transact do
-    mkdir "shell-test-1" unless exists?("shell-test-1")
-    cd("shell-test-1")
-    for dir in ["dir1", "dir3", "dir5"]
-      if !exists?(dir)
-	mkdir dir
-	cd(dir) do
-	  f = open("tmpFile", "w")
-	  f.print "TEST\n"
-	  f.close
-	end
-	print pwd
-      end
-    end
-  end
-
-== ex3
-
-  sh.cat("/etc/printcap") | sh.tee("tee1") > "tee2"
-  (sh.cat < "/etc/printcap") | sh.tee("tee11") > "tee12"
-  sh.cat("/etc/printcap") | sh.tee("tee1") >> "tee2"
-  (sh.cat < "/etc/printcap") | sh.tee("tee11") >> "tee12"
-
-== ex4
-
-  print sh.cat("/etc/passwd").head.collect{|l| l =~ /keiju/}
-
-=end
diff --git a/doc/shell.rd.ja b/doc/shell.rd.ja
index aab9e5c7d2..a9507fe92a 100644
--- a/doc/shell.rd.ja
+++ b/doc/shell.rd.ja
@@ -1,147 +1,146 @@
  -- shell.rb
 				$Release Version: 0.6.0 $
-			   	$Revision: 1.1 $
-			   	$Date: 2001/07/19 05:42:06 $
+			   	$Revision$
 			   	by Keiju ISHITSUKA(keiju@ishitsuka.com)
 
 =begin
 
-= ��Ū
+= 目的
 
-ruby���sh/csh�Τ褦�˥��ޥ�ɤμ¹Եڤӥե��륿��󥰤��ڤ˹Ԥ�.
-sh/csh������ʸ��ruby�ε�ǽ���Ѥ��Ƽ¸�����.
+ruby上でsh/cshのようにコマンドの実行及びフィルタリングを手軽に行う.
+sh/cshの制御文はrubyの機能を用いて実現する.
 
-= ��ʥ��饹����
+= 主なクラス一覧
 
 == Shell
 
-Shell���֥������Ȥϥ����ȥǥ��쥯�ȥ�����, ���ޥ�ɼ¹ԤϤ��������
-���Хѥ��ˤʤ�ޤ�.
+Shellオブジェクトはカレントディレクトリを持ち, コマンド実行はそこからの
+相対パスになります.
 
 --- Shell#cwd
 --- Shell#dir
 --- Shell#getwd
 --- Shell#pwd
 
-      �����ȥǥ��쥯�ȥ���֤���
+      カレントディレクトリを返す。
 
 --- Shell#system_path
 
-      ���ޥ�ɥ������ѥ���������֤���
+      コマンドサーチパスの配列を返す。
 
 --- Shell#umask
 
-      umask���֤���
+      umaskを返す。
 
 == Filter
 
-���ޥ�ɤμ¹Է�̤Ϥ��٤�Filter�Ȥ��Ƥ�����ޤ�. Enumerable��include��
-�Ƥ��ޤ�.
+コマンドの実行結果はすべてFilterとしてかえります. Enumerableをincludeし
+ています.
 
-= ��ʥ᥽�åɰ���
+= 主なメソッド一覧
 
-== ���ޥ�����
+== コマンド定義
 
-OS��Υ��ޥ�ɤ�¹Ԥ���ˤϤޤ�, Shell�Υ᥽�åɤȤ���������ޤ�.
+OS上のコマンドを実行するにはまず, Shellのメソッドとして定義します.
 
-��) ���ޥ�ɤ�������ʤ��Ȥ�ľ�ܼ¹ԤǤ���Shell#system���ޥ�ɤ⤢��ޤ�.
+注) コマンドを定義しなくとも直接実行できるShell#systemコマンドもあります.
 
 --- Shell.def_system_command(command, path = command)
 
-      Shell�Υ᥽�åɤȤ���command����Ͽ���ޤ�. 
+      Shellのメソッドとしてcommandを登録します.
 
-      ��)
+      例)
       Shell.def_system_command "ls"
-        ls �����
+        ls を定義
 
       Shell.def_system_command "sys_sort", "sort"
-        sort���ޥ�ɤ�sys_sort�Ȥ������
+        sortコマンドをsys_sortとして定義
 
 --- Shell.undef_system_command(command)
 
-      command�������ޤ�.
+      commandを削除します.
 
 --- Shell.alias_command(ali, command, *opts) {...}
 
-      command��alias�򤷤ޤ�. 
+      commandのaliasをします.
 
-      ��)
+      例)
         Shell.alias_command "lsC", "ls", "-CBF", "--show-control-chars"
         Shell.alias_command("lsC", "ls"){|*opts| ["-CBF", "--show-control-chars", *opts]}
 
 --- Shell.unalias_command(ali)
 
-      command��alias�������ޤ�.
+      commandのaliasを削除します.
 
 --- Shell.install_system_commands(pre = "sys_")
 
-      system_path��ˤ������Ƥμ¹Բ�ǽ�ե������Shell���������. �᥽��
-      ��̾�ϸ��Υե�����̾��Ƭ��pre��Ĥ�����ΤȤʤ�.
+      system_path上にある全ての実行可能ファイルをShellに定義する. メソッ
+      ド名は元のファイル名の頭にpreをつけたものとなる.
 
-== ����
+== 生成
 
 --- Shell.new
 
-      �ץ������Υ����ȥǥ��쥯�ȥ�򥫥��ȥǥ��쥯�ȥ�Ȥ���Shell��
-      �֥������Ȥ��������ޤ�.
+      プロセスのカレントディレクトリをカレントディレクトリとするShellオ
+      ブジェクトを生成します.
 
 --- Shell.cd(path)
 
-      path�򥫥��ȥǥ��쥯�ȥ�Ȥ���Shell���֥������Ȥ��������ޤ�.
+      pathをカレントディレクトリとするShellオブジェクトを生成します.
 
-== �ץ���������
+== プロセス管理
 
 --- Shell#jobs
 
-      �������塼��󥰤���Ƥ���job�ΰ������֤�.
+      スケジューリングされているjobの一覧を返す.
 
 --- Shell#kill sig, job
 
-      job�˥����ʥ�sig������
+      jobにシグナルsigを送る
 
-== �����ȥǥ��쥯�ȥ����
+== カレントディレクトリ操作
 
 --- Shell#cd(path, &block)
 --- Shell#chdir
 
-      �����ȥǥ��쥯�ȥ��path�ˤ���. ���ƥ졼���Ȥ��ƸƤФ줿�Ȥ��ˤ�
-      �֥��å��¹���Τߥ����ȥǥ��쥯�ȥ���ѹ�����.
+      カレントディレクトリをpathにする. イテレータとして呼ばれたときには
+      ブロック実行中のみカレントディレクトリを変更する.
 
 --- Shell#pushd(path = nil, &block)
 --- Shell#pushdir
 
-      �����ȥǥ��쥯�ȥ��ǥ��쥯�ȥꥹ���å��ˤĤ�, �����ȥǥ��쥯
-      �ȥ��path�ˤ���. path����ά���줿�Ȥ��ˤ�, �����ȥǥ��쥯�ȥ��
-      �ǥ��쥯�ȥꥹ���å��Υȥåפ�򴹤���. ���ƥ졼���Ȥ��ƸƤФ줿��
-      ���ˤ�, �֥��å��¹���Τ�pushd����.
+      カレントディレクトリをディレクトリスタックにつみ, カレントディレク
+      トリをpathにする. pathが省略されたときには, カレントディレクトリと
+      ディレクトリスタックのトップを交換する. イテレータとして呼ばれたと
+      きには, ブロック実行中のみpushdする.
 
 --- Shell#popd
 --- Shell#popdir
 
-      �ǥ��쥯�ȥꥹ���å�����ݥåפ�, ����򥫥��ȥǥ��쥯�ȥ�ˤ���.
+      ディレクトリスタックからポップし, それをカレントディレクトリにする.
 
-== �ե�����/�ǥ��쥯�ȥ����
+== ファイル/ディレクトリ操作
 
 --- Shell#foreach(path = nil, &block)
 
-      path���ե�����ʤ�, File#foreach
-      path���ǥ��쥯�ȥ�ʤ�, Dir#foreach
+      pathがファイルなら, File#foreach
+      pathがディレクトリなら, Dir#foreach
 
 --- Shell#open(path, mode)
 
-      path���ե�����ʤ�, File#open
-      path���ǥ��쥯�ȥ�ʤ�, Dir#open
+      pathがファイルなら, File#open
+      pathがディレクトリなら, Dir#open
 
 --- Shell#unlink(path)
 
-      path���ե�����ʤ�, File#unlink
-      path���ǥ��쥯�ȥ�ʤ�, Dir#unlink
+      pathがファイルなら, File#unlink
+      pathがディレクトリなら, Dir#unlink
 
 --- Shell#test(command, file1, file2)
 --- Shell#[command, file1, file2]
 
-      �ե�����ƥ��ȴؿ�test��Ʊ��. 
-      ��)
+      ファイルテスト関数testと同じ.
+      例)
           sh[?e, "foo"]
           sh[:e, "foo"]
           sh["e", "foo"]
@@ -150,75 +149,75 @@ OS��Υ��ޥ�ɤ�¹Ԥ���ˤϤޤ�, Shell�Υ᥽�åɤȤ���������ޤ�.
 
 --- Shell#mkdir(*path)
 
-      Dir.mkdir��Ʊ��(ʣ����)
+      Dir.mkdirと同じ(複数可)
 
 --- Shell#rmdir(*path)
 
-      Dir.rmdir��Ʊ��(ʣ����)
+      Dir.rmdirと同じ(複数可)
 
-== ���ޥ�ɼ¹�
+== コマンド実行
 
 --- System#system(command, *opts)
 
-      command��¹Ԥ���.
-      ��)
+      commandを実行する.
+      例)
         print sh.system("ls", "-l")
         sh.system("ls", "-l") | sh.head > STDOUT
 
 --- System#rehash
 
-      ��ϥå��夹��
+      リハッシュする
 
 --- Shell#transact &block
 
-      �֥��å���Ǥ�shell��self�Ȥ��Ƽ¹Ԥ���.
-      ��)
+      ブロック中ではshellをselfとして実行する.
+      例)
         sh.transact{system("ls", "-l") | head > STDOUT}
 
 --- Shell#out(dev = STDOUT, &block)
 
-      transact��ƤӽФ����η�̤�dev�˽��Ϥ���.
+      transactを呼び出しその結果をdevに出力する.
 
-== �������ޥ��
+== 内部コマンド
 
 --- Shell#echo(*strings)
 --- Shell#cat(*files)
 --- Shell#glob(patten)
 --- Shell#tee(file)
 
-      �����ϼ¹Ԥ����, ���������ƤȤ���Filter���֥������Ȥ��֤��ޤ�. 
+      これらは実行すると, それらを内容とするFilterオブジェクトを返します.
 
 --- Filter#each &block
 
-      �ե��륿�ΰ�Ԥ��Ĥ�block���Ϥ�.
+      フィルタの一行ずつをblockに渡す.
 
 --- Filter#<(src)
 
-      src��ե��륿�����ϤȤ���. src��, ʸ����ʤ�Хե������, IO�Ǥ���
-      �Ф���򤽤Τޤ����ϤȤ���.
+      srcをフィルタの入力とする. srcが, 文字列ならばファイルを, IOであれ
+      ばそれをそのまま入力とする.
 
 --- Filter#>(to)
 
-      src��ե��륿�ν��ϤȤ���. to��, ʸ����ʤ�Хե������, IO�Ǥ���
-      �Ф���򤽤Τޤ޽��ϤȤ���.
+      srcをフィルタの出力とする. toが, 文字列ならばファイルに, IOであれ
+      ばそれをそのまま出力とする.
 
 --- Filter#>>(to)
 
-      src��ե��륿���ɲä���. to��, ʸ����ʤ�Хե������, IO�Ǥ����
-      ����򤽤Τޤ޽��ϤȤ���.
+      srcをフィルタに追加する. toが, 文字列ならばファイルに, IOであれば
+      それをそのまま出力とする.
 
 --- Filter#|(filter)
 
-      �ѥ��׷��
+      パイプ結合
 
 --- Filter#+(filter)
 
-      filter1 + filter2 �� filter1�ν��Ϥθ�, filter2�ν��Ϥ�Ԥ�.
+      filter1 + filter2 は filter1の出力の後, filter2の出力を行う.
 
 --- Filter#to_a
 --- Filter#to_s
 
-== �ȹ��ߥ��ޥ��
+== 組込みコマンド
 
 --- Shell#atime(file)
 --- Shell#basename(file, *opt)
@@ -240,7 +239,7 @@ OS��Υ��ޥ�ɤ�¹Ԥ���ˤϤޤ�, Shell�Υ᥽�åɤȤ���������ޤ�.
 --- Shell#truncate(file, length)
 --- Shell#utime(atime, mtime, *file)
 
-      ������File���饹�ˤ���Ʊ̾�Υ��饹�᥽�åɤ�Ʊ���Ǥ�.
+      これらはFileクラスにある同名のクラスメソッドと同じです.
 
 --- Shell#blockdev?(file)
 --- Shell#chardev?(file)
@@ -264,7 +263,7 @@ OS��Υ��ޥ�ɤ�¹Ԥ���ˤϤޤ�, Shell�Υ᥽�åɤȤ���������ޤ�.
 --- Shell#writable_real?(file)
 --- Shell#zero?(file)
 
-      ������FileTest���饹�ˤ���Ʊ̾�Υ��饹�᥽�åɤ�Ʊ���Ǥ�.
+      これらはFileTestクラスにある同名のクラスメソッドと同じです.
 
 --- Shell#syscopy(filename_from, filename_to)
 --- Shell#copy(filename_from, filename_to)
@@ -274,9 +273,9 @@ OS��Υ��ޥ�ɤ�¹Ԥ���ˤϤޤ�, Shell�Υ᥽�åɤȤ���������ޤ�.
 --- Shell#makedirs(*filenames)
 --- Shell#install(filename_from, filename_to, mode)
 
-      ������FileTools���饹�ˤ���Ʊ̾�Υ��饹�᥽�åɤ�Ʊ���Ǥ�.
+      これらはFileToolsクラスにある同名のクラスメソッドと同じです.
 
-      ����¾, �ʲ��Τ�Τ������ꥢ������Ƥ��ޤ�.
+      その他, 以下のものがエイリアスされています.
 
 --- Shell#cmp	<- Shell#compare
 --- Shell#mv	<- Shell#move
@@ -284,7 +283,7 @@ OS��Υ��ޥ�ɤ�¹Ԥ���ˤϤޤ�, Shell�Υ᥽�åɤȤ���������ޤ�.
 --- Shell#rm_f	<- Shell#safe_unlink
 --- Shell#mkpath	<- Shell#makedirs
 
-= ����ץ�
+= サンプル
 
 == ex1
 
diff --git a/doc/signals.rdoc b/doc/signals.rdoc
new file mode 100644
index 0000000000..fdd9206afb
--- /dev/null
+++ b/doc/signals.rdoc
@@ -0,0 +1,106 @@
+= Caveats for implementing Signal.trap callbacks
+
+As with implementing signal handlers in C or most other languages,
+all code passed to Signal.trap must be reentrant.  If you are not
+familiar with reentrancy, you need to read up on it at
+{Wikipedia}[https://en.wikipedia.org/wiki/Reentrancy_(computing)] or
+elsewhere before reading the rest of this document.
+
+Most importantly, "thread-safety" does not guarantee reentrancy;
+and methods such as Mutex#lock and Mutex#synchronize which are
+commonly used for thread-safety even prevent reentrancy.
+
+== An implementation detail of the Ruby VM
+
+The Ruby VM defers Signal.trap callbacks from running until it is safe
+for its internal data structures, but it does not know when it is safe
+for data structures in YOUR code.  Ruby implements deferred signal
+handling by registering short C functions with only
+{async-signal-safe functions}[http://man7.org/linux/man-pages/man7/signal-safety.7.html] as
+signal handlers.  These short C functions only do enough tell the VM to
+run callbacks registered via Signal.trap later in the main VM loop.
+
+== Unsafe methods to call in Signal.trap blocks
+
+When in doubt, consider anything not listed as safe below as being
+unsafe.
+
+* Mutex#lock, Mutex#synchronize and any code using them are explicitly
+  unsafe.  This includes Monitor in the standard library which uses
+  Mutex to provide reentrancy.
+
+* Dir.chdir with block
+
+* any IO write operations when IO#sync is false;
+  including IO#write, IO#write_nonblock, IO#puts.
+  Pipes and sockets default to `IO#sync = true', so it is safe to
+  write to them unless IO#sync was disabled.
+
+* File#flock, as the underlying flock(2) call is not specified by POSIX
+
+== Commonly safe operations inside Signal.trap blocks
+
+* Assignment and retrieval of local, instance, and class variables
+
+* Most object allocations and initializations of common types
+  including Array, Hash, String, Struct, Time.
+
+* Common Array, Hash, String, Struct operations which do not execute a block
+  are generally safe; but beware if iteration is occurring elsewhere.
+
+* Hash#[], Hash#[]= (unless Hash.new was given an unsafe block)
+
+* Thread::Queue#push and Thread::SizedQueue#push (since Ruby 2.1)
+
+* Creating a new Thread via Thread.new/Thread.start can used to get
+  around the unusability of Mutexes inside a signal handler
+
+* Signal.trap is safe to use inside blocks passed to Signal.trap
+
+* arithmetic on Integer and Float (`+', `-', '%', '*', '/')
+
+  Additionally, signal handlers do not run between two successive
+  local variable accesses, so shortcuts such as `+=' and `-=' will
+  not trigger a data race when used on Integer and Float classes in
+  signal handlers.
+
+== System call wrapper methods which are safe inside Signal.trap
+
+Since Ruby has wrappers around many
+{async-signal-safe C functions}[http://man7.org/linux/man-pages/man7/signal-safety.7.html]
+the corresponding wrappers for many IO, File, Dir, and Socket methods
+are safe.
+
+(Incomplete list)
+
+* Dir.chdir (without block arg)
+* Dir.mkdir
+* Dir.open
+* File#truncate
+* File.link
+* File.open
+* File.readlink
+* File.rename
+* File.stat
+* File.symlink
+* File.truncate
+* File.unlink
+* File.utime
+* IO#close
+* IO#dup
+* IO#fsync
+* IO#read
+* IO#read_nonblock
+* IO#stat
+* IO#sysread
+* IO#syswrite
+* IO.select
+* IO.pipe
+* Process.clock_gettime
+* Process.exit!
+* Process.fork
+* Process.kill
+* Process.pid
+* Process.ppid
+* Process.waitpid
+...
diff --git a/doc/standard_library.rdoc b/doc/standard_library.rdoc
new file mode 100644
index 0000000000..09dc61bc52
--- /dev/null
+++ b/doc/standard_library.rdoc
@@ -0,0 +1,129 @@
+= Ruby Standard Library
+
+The Ruby Standard Library is a vast collection of classes and modules that you
+can require in your code for additional features.
+
+Below is an overview of libraries and extensions followed by a brief
+description.
+
+== Libraries
+
+Abbrev:: Calculates a set of unique abbreviations for a given set of strings
+Base64:: Support for encoding and decoding binary data using a Base64 representation
+Benchmark:: Provides methods to measure and report the time used to execute code
+CGI:: Support for the Common Gateway Interface protocol
+DEBUGGER__:: Debugging functionality for Ruby
+Delegator:: Provides three abilities to delegate method calls to an object
+DRb:: Distributed object system for Ruby
+English.rb:: Require 'English.rb' to reference global variables with less cryptic names
+ERB:: An easy to use but powerful templating system for Ruby
+Find:: This module supports top-down traversal of a set of file paths
+GetoptLong:: Parse command line options similar to the GNU C getopt_long()
+MakeMakefile:: Module used to generate a Makefile for C extensions
+Monitor:: Provides an object or module to use safely by more than one thread
+Net::FTP:: Support for the File Transfer Protocol
+Net::HTTP:: HTTP client api for Ruby
+Net::IMAP:: Ruby client api for Internet Message Access Protocol
+Net::POP3:: Ruby client library for POP3
+Net::SMTP:: Simple Mail Transfer Protocol client library for Ruby
+Observable:: Provides a mechanism for publish/subscribe pattern in Ruby
+OpenURI:: An easy-to-use wrapper for Net::HTTP, Net::HTTPS and Net::FTP
+Open3:: Provides access to stdin, stdout and stderr when running other programs
+OptionParser:: Ruby-oriented class for command-line option analysis
+PP:: Provides a PrettyPrinter for Ruby objects
+PrettyPrinter:: Implements a pretty printing algorithm for readable structure
+profile.rb:: Runs the Ruby Profiler__
+Profiler__:: Provides a way to profile your Ruby application
+PStore:: Implements a file based persistence mechanism based on a Hash
+Racc:: A LALR(1) parser generator written in Ruby.
+RbConfig:: Information of your configure and build of Ruby
+resolv-replace.rb:: Replace Socket DNS with Resolv
+Resolv::  Thread-aware DNS resolver library in Ruby
+Rinda:: The Linda distributed computing paradigm in Ruby
+Gem:: Package management framework for Ruby
+SecureRandom:: Interface for secure random number generator
+Set:: Provides a class to deal with collections of unordered, unique values
+Shellwords:: Manipulates strings with word parsing rules of UNIX Bourne shell
+Singleton:: Implementation of the Singleton pattern for Ruby
+Tempfile:: A utility class for managing temporary files
+Time:: Extends the Time class with methods for parsing and conversion
+Timeout:: Auto-terminate potentially long-running operations in Ruby
+tmpdir.rb:: Extends the Dir class to manage the OS temporary file path
+TSort:: Topological sorting using Tarjan's algorithm
+un.rb:: Utilities to replace common UNIX commands
+URI:: A Ruby module providing support for Uniform Resource Identifiers
+WeakRef:: Allows a referenced object to be garbage-collected
+YAML:: Ruby client library for the Psych YAML implementation
+
+== Extensions
+
+Coverage:: Provides coverage measurement for Ruby
+Digest:: Provides a framework for message digest libraries
+IO:: Extensions for Ruby IO class, including #wait and ::console
+NKF:: Ruby extension for Network Kanji Filter
+objspace:: Extends ObjectSpace module to add methods for internal statistics
+Pathname:: Representation of the name of a file or directory on the filesystem
+PTY:: Creates and manages pseudo terminals
+Readline:: Provides an interface for GNU Readline and Edit Line (libedit)
+Ripper:: Provides an interface for parsing Ruby programs into S-expressions
+Socket:: Access underlying OS socket implementations
+Syslog:: Ruby interface for the POSIX system logging facility
+WIN32OLE:: Provides an interface for OLE Automation in Ruby
+
+= Default gems
+
+== Libraries
+
+Bundler:: Manage your Ruby application's gem dependencies
+CMath:: Provides Trigonometric and Transcendental functions for complex numbers
+CSV:: Provides an interface to read and write CSV files and data
+E2MM:: Module for defining custom exceptions with specific messages
+FileUtils:: Several file utility methods for copying, moving, removing, etc
+Forwardable:: Provides delegation of specified methods to a designated object
+IPAddr:: Provides methods to manipulate IPv4 and IPv6 IP addresses
+IRB:: Interactive Ruby command-line tool for REPL (Read Eval Print Loop)
+Logger:: Provides a simple logging utility for outputting messages
+Matrix:: Represents a mathematical matrix.
+Mutex_m:: Mixin to extend objects to be handled like a Mutex
+OpenStruct:: Class to build custom data structures, similar to a Hash
+Prime:: Prime numbers and factorization library
+RDoc:: Produces HTML and command-line documentation for Ruby
+REXML:: An XML toolkit for Ruby
+RSS:: Family of libraries that support various formats of XML "feeds"
+Scanf:: A Ruby implementation of the C function scanf(3)
+Shell:: An idiomatic Ruby interface for common UNIX shell commands
+Synchronizer:: A module that provides a two-phase lock with a counter
+ThreadsWait:: Watches for termination of multiple threads
+Tracer:: Outputs a source level execution trace of a Ruby program
+WEBrick:: An HTTP server toolkit for Ruby
+
+== Extensions
+
+BigDecimal:: Provides arbitrary-precision floating point decimal arithmetic
+Date:: A subclass of Object includes Comparable module for handling dates
+DateTime:: Subclass of Date to handling dates, hours, minutes, seconds, offsets
+DBM:: Provides a wrapper for the UNIX-style Database Manager Library
+Etc:: Provides access to information typically stored in UNIX /etc directory
+Fcntl:: Loads constants defined in the OS fcntl.h C header file
+Fiddle:: A libffi wrapper for Ruby
+GDBM:: Ruby extension for the GNU dbm (gdbm) library
+IO::console:: Console interface
+JSON:: Implements Javascript Object Notation for Ruby
+OpenSSL:: Provides SSL, TLS and general purpose cryptography for Ruby
+Psych:: A YAML parser and emitter for Ruby
+SDBM:: Provides a simple file-based key-value store with String keys and values
+StringIO:: Pseudo I/O on String objects
+StringScanner:: Provides lexical scanning operations on a String
+Zlib:: Ruby interface for the zlib compression/decompression library
+
+= Bundled gems
+
+== Libraries
+
+DidYouMean:: "Did you mean?" experience in Ruby
+MiniTest:: A test suite with TDD, BDD, mocking and benchmarking
+Net::Telnet:: Telnet client library for Ruby
+PowerAssert:: Power Assert for Ruby.
+Rake:: Ruby build program with capabilities similar to make
+Test::Unit:: A compatibility layer for MiniTest
+XMLRPC:: Remote Procedure Call over HTTP support for Ruby
diff --git a/doc/syntax.rdoc b/doc/syntax.rdoc
new file mode 100644
index 0000000000..fe0f98ce4c
--- /dev/null
+++ b/doc/syntax.rdoc
@@ -0,0 +1,34 @@
+= Ruby Syntax
+
+The Ruby syntax is large and is split up into the following sections:
+
+Literals[rdoc-ref:syntax/literals.rdoc] ::
+  Numbers, Strings, Arrays, Hashes, etc.
+
+Assignment[rdoc-ref:syntax/assignment.rdoc] ::
+  Assignment and variables
+
+{Control Expressions}[rdoc-ref:syntax/control_expressions.rdoc] ::
+  +if+, +unless+, +while+, +until+, +for+, +break+, +next+, +redo+
+
+Methods[rdoc-ref:syntax/methods.rdoc] ::
+  Method and method argument syntax
+
+{Calling Methods}[rdoc-ref:syntax/calling_methods.rdoc] ::
+  How to call a method (or send a message to a method)
+
+{Modules and Classes}[rdoc-ref:syntax/modules_and_classes.rdoc] ::
+  Creating modules and classes including inheritance
+
+Exceptions[rdoc-ref:syntax/exceptions.rdoc] ::
+  Exception handling syntax
+
+Precedence[rdoc-ref:syntax/precedence.rdoc] ::
+  Precedence of ruby operators
+
+Refinements[rdoc-ref:syntax/refinements.rdoc] ::
+  Use and behavior of the experimental refinements feature
+
+Miscellaneous[rdoc-ref:syntax/miscellaneous.rdoc] ::
+  +alias+, +undef+, +BEGIN+, +END+
+
diff --git a/doc/syntax/assignment.rdoc b/doc/syntax/assignment.rdoc
new file mode 100644
index 0000000000..83300cbece
--- /dev/null
+++ b/doc/syntax/assignment.rdoc
@@ -0,0 +1,454 @@
+= Assignment
+
+In Ruby, assignment uses the <code>=</code> (equals sign) character.  This
+example assigns the number five to the local variable +v+:
+
+  v = 5
+
+Assignment creates a local variable if the variable was not previously
+referenced.
+
+== Local Variable Names
+
+A local variable name must start with a lowercase US-ASCII letter or a
+character with the eight bit set.  Typically local variables are US-ASCII
+compatible since the keys to type them exist on all keyboards.
+
+(Ruby programs must be written in a US-ASCII-compatible character set.  In
+such character sets if the eight bit is set it indicates an extended
+character.  Ruby allows local variables to contain such characters.)
+
+A local variable name may contain letters, numbers, an <code>_</code>
+(underscore or low line) or a character with the eighth bit set.
+
+== Local Variable Scope
+
+Once a local variable name has been assigned-to all uses of the name for the
+rest of the scope are considered local variables.
+
+Here is an example:
+
+  1.times do
+    a = 1
+    puts "local variables in the block: #{local_variables.join ", "}"
+  end
+
+  puts "no local variables outside the block" if local_variables.empty?
+
+This prints:
+
+  local variables in the block: a
+  no local variables outside the block
+
+Since the block creates a new scope, any local variables created inside it do
+not leak to the surrounding scope.
+
+Variables defined in an outer scope appear inner scope:
+
+  a = 0
+
+  1.times do
+    puts "local variables: #{local_variables.join ", "}"
+  end
+
+This prints:
+
+  local variables: a
+
+You may isolate variables in a block from the outer scope by listing them
+following a <code>;</code> in the block's arguments.  See the documentation
+for block local variables in the {calling
+methods}[rdoc-ref:syntax/calling_methods.rdoc] documentation for an example.
+
+See also Kernel#local_variables, but note that a +for+ loop does not create a
+new scope like a block does.
+
+== Local Variables and Methods
+
+In Ruby local variable names and method names are nearly identical.  If you
+have not assigned to one of these ambiguous names ruby will assume you wish to
+call a method.  Once you have assigned to the name ruby will assume you wish
+to reference a local variable.
+
+The local variable is created when the parser encounters the assignment, not
+when the assignment occurs:
+
+  a = 0 if false # does not assign to a
+
+  p local_variables # prints [:a]
+
+  p a # prints nil
+
+The similarity between method and local variable names can lead to confusing
+code, for example:
+
+  def big_calculation
+    42 # pretend this takes a long time
+  end
+
+  big_calculation = big_calculation()
+
+Now any reference to +big_calculation+ is considered a local variable and will
+be cached.  To call the method, use <code>self.big_calculation</code>.
+
+You can force a method call by using empty argument parentheses as shown above
+or by using an explicit receiver like <code>self.</code>.  Using an explicit
+receiver may raise a NameError if the method's visibility is not public.
+
+Another commonly confusing case is when using a modifier +if+:
+
+  p a if a = 0.zero?
+
+Rather than printing "true" you receive a NameError, "undefined local variable
+or method `a'".  Since ruby parses the bare +a+ left of the +if+ first and has
+not yet seen an assignment to +a+ it assumes you wish to call a method.  Ruby
+then sees the assignment to +a+ and will assume you are referencing a local
+method.
+
+The confusion comes from the out-of-order execution of the expression.  First
+the local variable is assigned-to then you attempt to call a nonexistent
+method.
+
+== Instance Variables
+
+Instance variables are shared across all methods for the same object.
+
+An instance variable must start with a <code>@</code> ("at" sign or
+commercial at).  Otherwise instance variable names follow the rules as local
+variable names.  Since the instance variable starts with an <code>@</code> the
+second character may be an upper-case letter.
+
+Here is an example of instance variable usage:
+
+  class C
+    def initialize(value)
+      @instance_variable = value
+    end
+
+    def value
+      @instance_variable
+    end
+  end
+
+  object1 = C.new "some value"
+  object2 = C.new "other value"
+
+  p object1.value # prints "some value"
+  p object2.value # prints "other value"
+
+An uninitialized instance variable has a value of +nil+.  If you run Ruby with
+warnings enabled, you will get a warning when accessing an uninitialized
+instance variable.
+
+The +value+ method has access to the value set by the +initialize+ method, but
+only for the same object.
+
+== Class Variables
+
+Class variables are shared between a class, its subclasses and its instances.
+
+A class variable must start with a <code>@@</code> (two "at" signs).  The rest
+of the name follows the same rules as instance variables.
+
+Here is an example:
+
+  class A
+    @@class_variable = 0
+
+    def value
+      @@class_variable
+    end
+
+    def update
+      @@class_variable = @@class_variable + 1
+    end
+  end
+
+  class B < A
+    def update
+      @@class_variable = @@class_variable + 2
+    end
+  end
+
+  a = A.new
+  b = B.new
+
+  puts "A value: #{a.value}"
+  puts "B value: #{b.value}"
+
+This prints:
+
+  A value: 0
+  B value: 0
+
+Continuing with the same example, we can update using objects from either
+class and the value is shared:
+
+  puts "update A"
+  a.update
+
+  puts "A value: #{a.value}"
+  puts "B value: #{b.value}"
+
+  puts "update B"
+  b.update
+
+  puts "A value: #{a.value}"
+  puts "B value: #{b.value}"
+
+  puts "update A"
+  a.update
+
+  puts "A value: #{a.value}"
+  puts "B value: #{b.value}"
+
+This prints:
+
+  update A
+  A value: 1
+  B value: 1
+  update B
+  A value: 3
+  B value: 3
+  update A
+  A value: 4
+  B value: 4
+
+Accessing an uninitialized class variable will raise a NameError exception.
+
+Note that classes have instance variables because classes are objects, so
+try not to confuse class and instance variables.
+
+== Global Variables
+
+Global variables are accessible everywhere.
+
+Global variables start with a <code>$</code> (dollar sign).  The rest of the
+name follows the same rules as instance variables.
+
+Here is an example:
+
+  $global = 0
+
+  class C
+    puts "in a class: #{$global}"
+
+    def my_method
+      puts "in a method: #{$global}"
+
+      $global = $global + 1
+      $other_global = 3
+    end
+  end
+
+  C.new.my_method
+
+  puts "at top-level, $global: #{$global}, $other_global: #{$other_global}"
+
+This prints:
+
+  in a class: 0
+  in a method: 0
+  at top-level, $global: 1, $other_global: 3
+
+An uninitialized global variable has a value of +nil+.
+
+Ruby has some special globals that behave differently depending on context
+such as the regular expression match variables or that have a side-effect when
+assigned to.  See the {global variables documentation}[rdoc-ref:globals.rdoc]
+for details.
+
+== Assignment Methods
+
+You can define methods that will behave like assignment, for example:
+
+  class C
+    def value=(value)
+      @value = value
+    end
+  end
+
+  c = C.new
+  c.value = 42
+
+Using assignment methods allows your programs to look nicer.  When assigning
+to an instance variable most people use Module#attr_accessor:
+
+  class C
+    attr_accessor :value
+  end
+
+When using method assignment you must always have a receiver.  If you do not
+have a receiver, Ruby assumes you are assigning to a local variable:
+
+  class C
+    attr_accessor :value
+
+    def my_method
+      value = 42
+
+      puts "local_variables: #{local_variables.join ", "}"
+      puts "@value: #{@value.inspect}"
+    end
+  end
+
+  C.new.my_method
+
+This prints:
+
+  local_variables: value
+  @value: nil
+
+To use the assignment method you must set the receiver:
+
+  class C
+    attr_accessor :value
+
+    def my_method
+      self.value = 42
+
+      puts "local_variables: #{local_variables.join ", "}"
+      puts "@value: #{@value.inspect}"
+    end
+  end
+
+  C.new.my_method
+
+This prints:
+
+  local_variables:
+  @value: 42
+
+== Abbreviated Assignment
+
+You can mix several of the operators and assignment.  To add 1 to an object
+you can write:
+
+  a = 1
+
+  a += 2
+
+  p a # prints 3
+
+This is equivalent to:
+
+  a = 1
+
+  a = a + 2
+
+  p a # prints 3
+
+You can use the following operators this way:  <code>+</code>, <code>-</code>,
+<code>*</code>, <code>/</code>, <code>%</code>, <code>**</code>,
+<code>&</code>, <code>|</code>, <code>^</code>, <code><<</code>,
+<code>>></code>
+
+There are also <code>||=</code> and <code>&&=</code>.  The former makes an
+assignment if the value was +nil+ or +false+ while the latter makes an
+assignment if the value was not +nil+ or +false+.
+
+Here is an example:
+
+  a ||= 0
+  a &&= 1
+
+  p a # prints 1
+
+Note that these two operators behave more like <code>a || a = 0</code> than
+<code>a = a || 0</code>.
+
+== Implicit Array Assignment
+
+You can implicitly create an array by listing multiple values when assigning:
+
+  a = 1, 2, 3
+
+  p a # prints [1, 2, 3]
+
+This implicitly creates an Array.
+
+You can use <code>*</code> or the "splat" operator or unpack an Array when
+assigning.  This is similar to multiple assignment:
+
+  a = *[1, 2, 3]
+
+  p a # prints [1, 2, 3]
+
+You can splat anywhere in the right-hand side of the assignment:
+
+  a = 1, *[2, 3]
+
+  p a # prints [1, 2, 3]
+
+== Multiple Assignment
+
+You can assign multiple values on the right-hand side to multiple variables:
+
+  a, b = 1, 2
+
+  p a: a, b: b # prints {:a=>1, :b=>2}
+
+In the following sections any place "variable" is used an assignment method,
+instance, class or global will also work:
+
+  def value=(value)
+    p assigned: value
+  end
+
+  self.value, $global = 1, 2 # prints {:assigned=>1}
+
+  p $global # prints 2
+
+You can use multiple assignment to swap two values in-place:
+
+  old_value = 1
+
+  new_value, old_value = old_value, 2
+
+  p new_value: new_value, old_value: old_value
+  # prints {:new_value=>1, :old_value=>2}
+
+If you have more values on the right hand side of the assignment than variables
+on the left hand side, the extra values are ignored:
+
+  a, b = 1, 2, 3
+
+  p a: a, b: b # prints {:a=>1, :b=>2}
+
+You can use <code>*</code> to gather extra values on the right-hand side of
+the assignment.
+
+  a, *b = 1, 2, 3
+
+  p a: a, b: b # prints {:a=>1, :b=>[2, 3]}
+
+The <code>*</code> can appear anywhere on the left-hand side:
+
+  *a, b = 1, 2, 3
+
+  p a: a, b: b # prints {:a=>[1, 2], :b=>3}
+
+But you may only use one <code>*</code> in an assignment.
+
+== Array Decomposition
+
+Like Array decomposition in {method arguments}[rdoc-ref:syntax/methods.rdoc]
+you can decompose an Array during assignment using parenthesis:
+
+  (a, b) = [1, 2]
+
+  p a: a, b: b # prints {:a=>1, :b=>2}
+
+You can decompose an Array as part of a larger multiple assignment:
+
+  a, (b, c) = 1, [2, 3]
+
+  p a: a, b: b, c: c # prints {:a=>1, :b=>2, :c=>3}
+
+Since each decomposition is considered its own multiple assignment you can use
+<code>*</code> to gather arguments in the decomposition:
+
+  a, (b, *c), *d = 1, [2, 3, 4], 5, 6
+
+  p a: a, b: b, c: c, d: d
+  # prints {:a=>1, :b=>2, :c=>[3, 4], :d=>[5, 6]}
diff --git a/doc/syntax/calling_methods.rdoc b/doc/syntax/calling_methods.rdoc
new file mode 100644
index 0000000000..b86d60ad88
--- /dev/null
+++ b/doc/syntax/calling_methods.rdoc
@@ -0,0 +1,371 @@
+= Calling Methods
+
+Calling a method sends a message to an object so it can perform some work.
+
+In ruby you send a message to an object like this:
+
+  my_method()
+
+Note that the parenthesis are optional:
+
+  my_method
+
+Except when there is difference between using and omitting parentheses, this
+document uses parenthesis when arguments are present to avoid confusion.
+
+This section only covers calling methods.  See also the {syntax documentation
+on defining methods}[rdoc-ref:syntax/methods.rdoc].
+
+== Receiver
+
++self+ is the default receiver.  If you don't specify any receiver +self+ will
+be used.  To specify a receiver use <code>.</code>:
+
+  my_object.my_method
+
+This sends the +my_method+ message to +my_object+.  Any object can be a
+receiver but depending on the method's visibility sending a message may raise a
+NoMethodError.
+
+You may also use <code>::</code> to designate a receiver, but this is rarely
+used due to the potential for confusion with <code>::</code> for namespaces.
+
+=== Safe navigation operator
+
+<code>&.</code>, called "safe navigation operator", allows to skip method call
+when receiver is +nil+. It returns +nil+ and doesn't evaluate method's arguments
+if the call is skipped.
+
+  REGEX = /(ruby) is (\w+)/i
+  "Ruby is awesome!".match(REGEX).values_at(1, 2)
+  # => ["Ruby", "awesome"]
+  "Python is fascinating!".match(REGEX).values_at(1, 2)
+  # NoMethodError: undefined method `values_at' for nil:NilClass
+  "Python is fascinating!".match(REGEX)&.values_at(1, 2)
+  # => nil
+
+This allows to easily chain methods which could return empty value. Note that
+<code>&.</code> skips only one next call, so for a longer chain it is necessary
+to add operator on each level:
+
+  "Python is fascinating!".match(REGEX)&.values_at(1, 2).join(' - ')
+  # NoMethodError: undefined method `join' for nil:NilClass
+  "Python is fascinating!".match(REGEX)&.values_at(1, 2)&.join(' - ')
+  # => nil
+
+== Arguments
+
+There are three types of arguments when sending a message, the positional
+arguments, keyword (or named) arguments and the block argument.  Each message
+sent may use one, two or all types of arguments, but the arguments must be
+supplied in this order.
+
+All arguments in ruby are passed by reference and are not lazily evaluated.
+
+Each argument is separated by a <code>,</code>:
+
+  my_method(1, '2', :three)
+
+Arguments may be an expression, a hash argument:
+
+  'key' => value
+
+or a keyword argument:
+
+  key: value
+
+Hash and keyword arguments must be contiguous and must appear after all
+positional arguments, but may be mixed:
+
+  my_method('a' => 1, b: 2, 'c' => 3)
+
+=== Positional Arguments
+
+The positional arguments for the message follow the method name:
+
+  my_method(argument1, argument2)
+
+In many cases, parenthesis are not necessary when sending a message:
+
+  my_method argument1, argument2
+
+However, parenthesis are necessary to avoid ambiguity.  This will raise a
+SyntaxError because ruby does not know which method argument3 should be sent
+to:
+
+  method_one argument1, method_two argument2, argument3
+
+If the method definition has a <code>*argument</code> extra positional
+arguments will be assigned to +argument+ in the method as an Array.
+
+If the method definition doesn't include keyword arguments the keyword or
+hash-type arguments are assigned as a single hash to the last argument:
+
+  def my_method(options)
+    p options
+  end
+
+  my_method('a' => 1, b: 2) # prints: {'a'=>1, :b=>2}
+
+If too many positional arguments are given, an ArgumentError is raised.
+
+=== Default Positional Arguments
+
+When the method defines default arguments you do not need to supply all the
+arguments to the method.  Ruby will fill in the missing arguments in-order.
+
+First we'll cover the simple case where the default arguments appear on the
+right.  Consider this method:
+
+  def my_method(a, b, c = 3, d = 4)
+    p [a, b, c, d]
+  end
+
+Here +c+ and +d+ have default values which ruby will apply for you.  If you
+send only two arguments to this method:
+
+  my_method(1, 2)
+
+You will see ruby print <code>[1, 2, 3, 4]</code>.
+
+If you send three arguments:
+
+  my_method(1, 2, 5)
+
+You will see ruby print <code>[1, 2, 5, 4]</code>
+
+Ruby fills in the missing arguments from left to right.
+
+Ruby allows default values to appear in the middle of positional arguments.
+Consider this more complicated method:
+
+  def my_method(a, b = 2, c = 3, d)
+    p [a, b, c, d]
+  end
+
+Here +b+ and +c+ have default values.  If you send only two arguments to this
+method:
+
+  my_method(1, 4)
+
+You will see ruby print <code>[1, 2, 3, 4]</code>.
+
+If you send three arguments:
+
+  my_method(1, 5, 6)
+
+You will see ruby print <code>[1, 5, 3, 6]</code>.
+
+Describing this in words gets complicated and confusing. I'll describe it
+in variables and values instead.
+
+First <code>1</code> is assigned to +a+, then <code>6</code> is assigned to
++d+.  This leaves only the arguments with default values.  Since
+<code>5</code> has not been assigned to a value yet, it is given to +b+ and
++c+ uses its default value of <code>3</code>.
+
+=== Keyword Arguments
+
+Keyword arguments follow any positional arguments and are separated by commas
+like positional arguments:
+
+  my_method(positional1, keyword1: value1, keyword2: value2)
+
+Any keyword arguments not given will use the default value from the method
+definition.  If a keyword argument is given that the method did not list an
+ArgumentError will be raised.
+
+=== Block Argument
+
+The block argument sends a closure from the calling scope to the method.
+
+The block argument is always last when sending a message to a method.  A block
+is sent to a method using <code>do ... end</code> or <code>{ ... }</code>:
+
+  my_method do
+    # ...
+  end
+
+or:
+
+  my_method {
+    # ...
+  }
+
+<code>do end</code> has lower precedence than <code>{ }</code> so:
+
+  method_1 method_2 {
+    # ...
+  }
+
+Sends the block to +method_2+ while:
+
+  method_1 method_2 do
+    # ...
+  end
+
+Sends the block to +method_1+.  Note that in the first case if parentheses are
+used the block is sent to +method_1+.
+
+A block will accept arguments from the method it was sent to.  Arguments are
+defined similar to the way a method defines arguments.  The block's arguments
+go in <code>| ... |</code> following the opening <code>do</code> or
+<code>{</code>:
+
+  my_method do |argument1, argument2|
+    # ...
+  end
+
+==== Block Local Arguments
+
+You may also declare block-local arguments to a block using <code>;</code> in
+the block arguments list.  Assigning to a block-local argument will not
+override local arguments outside the block in the caller's scope:
+
+  def my_method
+    yield self
+  end
+
+  place = "world"
+
+  my_method do |obj; place|
+    place = "block"
+    puts "hello #{obj} this is #{place}"
+  end
+
+  puts "place is: #{place}"
+
+This prints:
+
+  hello main this is block
+  place is world
+
+So the +place+ variable in the block is not the same +place+ variable as
+outside the block.  Removing <code>; place</code> from the block arguments
+gives this result:
+
+  hello main this is block
+  place is block
+
+=== Array to Arguments Conversion
+
+Given the following method:
+
+  def my_method(argument1, argument2, argument3)
+  end
+
+You can turn an Array into an argument list with <code>*</code> (or splat)
+operator:
+
+  arguments = [1, 2, 3]
+  my_method(*arguments)
+
+or:
+
+  arguments = [2, 3]
+  my_method(1, *arguments)
+
+Both are equivalent to:
+
+  my_method(1, 2, 3)
+
+If the method accepts keyword arguments, the splat operator will convert a
+hash at the end of the array into keyword arguments:
+
+  def my_method(a, b, c: 3)
+  end
+
+  arguments = [1, 2, { c: 4 }]
+  my_method(*arguments)
+
+You may also use the <code>**</code> (described next) to convert a Hash into
+keyword arguments.
+
+If the number of objects in the Array do not match the number of arguments for
+the method, an ArgumentError will be raised.
+
+If the splat operator comes first in the call, parentheses must be used to
+avoid a warning.
+
+=== Hash to Keyword Arguments Conversion
+
+Given the following method:
+
+  def my_method(first: 1, second: 2, third: 3)
+  end
+
+You can turn a Hash into keyword arguments with the <code>**</code> operator:
+
+  arguments = { first: 3, second: 4, third: 5 }
+  my_method(**arguments)
+
+or:
+
+  arguments = { first: 3, second: 4 }
+  my_method(third: 5, **arguments)
+
+Both are equivalent to:
+
+  my_method(first: 3, second: 4, third: 5)
+
+If the method definition uses <code>**</code> to gather arbitrary keyword
+arguments, they will not be gathered by <code>*</code>:
+
+  def my_method(*a, **kw)
+    p arguments: a, keywords: kw
+  end
+
+  my_method(1, 2, '3' => 4, five: 6)
+
+Prints:
+
+  {:arguments=>[1, 2, {"3"=>4}], :keywords=>{:five=>6}}
+
+Unlike the splat operator described above, the <code>**</code> operator has no
+commonly recognized name.
+
+=== Proc to Block Conversion
+
+Given a method that use a block:
+
+  def my_method
+    yield self
+  end
+
+You can convert a proc or lambda to a block argument with the <code>&</code>
+operator:
+
+  argument = proc { |a| puts "#{a.inspect} was yielded" }
+
+  my_method(&argument)
+
+If the splat operator comes first in the call, parenthesis must be used to
+avoid a warning.
+
+Unlike the splat operator described above, the <code>&</code> operator has no
+commonly recognized name.
+
+== Method Lookup
+
+When you send a message, Ruby looks up the method that matches the name of the
+message for the receiver.  Methods are stored in classes and modules so method
+lookup walks these, not the objects themselves.
+
+Here is the order of method lookup for the receiver's class or module +R+:
+
+* The prepended modules of +R+ in reverse order
+* For a matching method in +R+
+* The included modules of +R+ in reverse order
+
+If +R+ is a class with a superclass, this is repeated with +R+'s superclass
+until a method is found.
+
+Once a match is found method lookup stops.
+
+If no match is found this repeats from the beginning, but looking for
++method_missing+.  The default +method_missing+ is BasicObject#method_missing
+which raises a NameError when invoked.
+
+If refinements (an experimental feature) are active, the method lookup changes.
+See the {refinements documentation}[rdoc-ref:syntax/refinements.rdoc] for
+details.
diff --git a/doc/syntax/control_expressions.rdoc b/doc/syntax/control_expressions.rdoc
new file mode 100644
index 0000000000..65f7b431e3
--- /dev/null
+++ b/doc/syntax/control_expressions.rdoc
@@ -0,0 +1,499 @@
+= Control Expressions
+
+Ruby has a variety of ways to control execution.  All the expressions described
+here return a value.
+
+For the tests in these control expressions, +nil+ and +false+ are false-values
+and +true+ and any other object are true-values.  In this document "true" will
+mean "true-value" and "false" will mean "false-value".
+
+== +if+ Expression
+
+The simplest +if+ expression has two parts, a "test" expression and a "then"
+expression.  If the "test" expression evaluates to a true then the "then"
+expression is evaluated.
+
+Here is a simple if statement:
+
+  if true then
+    puts "the test resulted in a true-value"
+  end
+
+This will print "the test resulted in a true-value".
+
+The +then+ is optional:
+
+  if true
+    puts "the test resulted in a true-value"
+  end
+
+This document will omit the optional +then+ for all expressions as that is the
+most common usage of +if+.
+
+You may also add an +else+ expression.  If the test does not evaluate to true
+the +else+ expression will be executed:
+
+  if false
+    puts "the test resulted in a true-value"
+  else
+    puts "the test resulted in a false-value"
+  end
+
+This will print "the test resulted in a false-value".
+
+You may add an arbitrary number of extra tests to an if expression using
++elsif+.  An +elsif+ executes when all tests above the +elsif+ are false.
+
+  a = 1
+
+  if a == 0
+    puts "a is zero"
+  elsif a == 1
+    puts "a is one"
+  else
+    puts "a is some other value"
+  end
+
+This will print "a is one" as <code>1</code> is not equal to <code>0</code>.
+Since +else+ is only executed when there are no matching conditions.
+
+Once a condition matches, either the +if+ condition or any +elsif+ condition,
+the +if+ expression is complete and no further tests will be performed.
+
+Like an +if+, an +elsif+ condition may be followed by a +then+.
+
+In this example only "a is one" is printed:
+
+  a = 1
+
+  if a == 0
+    puts "a is zero"
+  elsif a == 1
+    puts "a is one"
+  elsif a >= 1
+    puts "a is greater than or equal to one"
+  else
+    puts "a is some other value"
+  end
+
+The tests for +if+ and +elsif+ may have side-effects.  The most common use of
+side-effect is to cache a value into a local variable:
+
+  if a = object.some_value
+    # do something to a
+  end
+
+The result value of an +if+ expression is the last value executed in the
+expression.
+
+== Ternary if
+
+You may also write a if-then-else expression using <code>?</code> and
+<code>:</code>.  This ternary if:
+
+  input_type = gets =~ /hello/i ? "greeting" : "other"
+
+Is the same as this +if+ expression:
+
+  input_type =
+    if gets =~ /hello/i
+      "greeting"
+    else
+      "other"
+    end
+
+While the ternary if is much shorter to write than the more verbose form, for
+readability it is recommended that the ternary if is only used for simple
+conditionals.  Also, avoid using multiple ternary conditions in the same
+expression as this can be confusing.
+
+== +unless+ Expression
+
+The +unless+ expression is the opposite of the +if+ expression.  If the value
+is false, the "then" expression is executed:
+
+  unless true
+    puts "the value is a false-value"
+  end
+
+This prints nothing as true is not a false-value.
+
+You may use an optional +then+ with +unless+ just like +if+.
+
+Note that the above +unless+ expression is the same as:
+
+  if not true
+    puts "the value is a false-value"
+  end
+
+Like an +if+ expression you may use an +else+ condition with +unless+:
+
+  unless true
+    puts "the value is false"
+  else
+    puts "the value is true"
+  end
+
+This prints "the value is true" from the +else+ condition.
+
+You may not use +elsif+ with an +unless+ expression.
+
+The result value of an +unless+ expression is the last value executed in the
+expression.
+
+== Modifier +if+ and +unless+
+
++if+ and +unless+ can also be used to modify an expression.  When used as a
+modifier the left-hand side is the "then" expression and the right-hand side
+is the "test" expression:
+
+  a = 0
+
+  a += 1 if a.zero?
+
+  p a
+
+This will print 1.
+
+  a = 0
+
+  a += 1 unless a.zero?
+
+  p a
+
+This will print 0.
+
+While the modifier and standard versions have both a "test" expression and a
+"then" expression, they are not exact transformations of each other due to
+parse order.  Here is an example that shows the difference:
+
+  p a if a = 0.zero?
+
+This raises the NameError "undefined local variable or method `a'".
+
+When ruby parses this expression it first encounters +a+ as a method call in
+the "then" expression, then later it sees the assignment to +a+ in the "test"
+expression and marks +a+ as a local variable.
+
+When running this line it first executes the "test" expression, <code>a =
+0.zero?</code>.
+
+Since the test is true it executes the "then" expression, <code>p a</code>.
+Since the +a+ in the body was recorded as a method which does not exist the
+NameError is raised.
+
+The same is true for +unless+.
+
+== +case+ Expression
+
+The +case+ expression can be used in two ways.
+
+The most common way is to compare an object against multiple patterns.  The
+patterns are matched using the +===+ method which is aliased to +==+ on
+Object.  Other classes must override it to give meaningful behavior.  See
+Module#=== and Regexp#=== for examples.
+
+Here is an example of using +case+ to compare a String against a pattern:
+
+  case "12345"
+  when /^1/
+    puts "the string starts with one"
+  else
+    puts "I don't know what the string starts with"
+  end
+
+Here the string <code>"12345"</code> is compared with <code>/^1/</code> by
+calling <code>/^1/ === "12345"</code> which returns +true+.  Like the +if+
+expression, the first +when+ that matches is executed and all other matches are
+ignored.
+
+If no matches are found, the +else+ is executed.
+
+The +else+ and +then+ are optional, this +case+ expression gives the same
+result as the one above:
+
+  case "12345"
+  when /^1/
+    puts "the string starts with one"
+  end
+
+You may place multiple conditions on the same +when+:
+
+  case "2"
+  when /^1/, "2"
+    puts "the string starts with one or is '2'"
+  end
+
+Ruby will try each condition in turn, so first <code>/^1/ === "2"</code>
+returns +false+, then <code>"2" === "2"</code> returns +true+, so "the string
+starts with one or is '2'" is printed.
+
+You may use +then+ after the +when+ condition.  This is most frequently used
+to place the body of the +when+ on a single line.
+
+  case a
+  when 1, 2 then puts "a is one or two
+  when 3    then puts "a is three"
+  else           puts "I don't know what a is"
+  end
+
+The other way to use a +case+ expression is like an if-elsif expression:
+
+  a = 2
+
+  case
+  when a == 1, a == 2
+    puts "a is one or two"
+  when a == 3
+    puts "a is three"
+  else
+    puts "I don't know what a is"
+  end
+
+Again, the +then+ and +else+ are optional.
+
+The result value of a +case+ expression is the last value executed in the
+expression.
+
+== +while+ Loop
+
+The +while+ loop executes while a condition is true:
+
+  a = 0
+
+  while a < 10 do
+    p a
+    a += 1
+  end
+
+  p a
+
+Prints the numbers 0 through 10.  The condition <code>a < 10</code> is checked
+before the loop is entered, then the body executes, then the condition is
+checked again.  When the condition results in false the loop is terminated.
+
+The +do+ keyword is optional.  The following loop is equivalent to the loop
+above:
+
+  while a < 10
+    p a
+    a += 1
+  end
+
+The result of a +while+ loop is +nil+ unless +break+ is used to supply a
+value.
+
+== +until+ Loop
+
+The +until+ loop executes while a condition is false:
+
+  a = 0
+
+  until a > 10 do
+    p a
+    a += 1
+  end
+
+  p a
+
+This prints the numbers 0 through 11.  Like a while loop the condition <code>a
+> 10</code> is checked when entering the loop and each time the loop body
+executes.  If the condition is false the loop will continue to execute.
+
+Like a +while+ loop, the +do+ is optional.
+
+Like a +while+ loop, the result of an +until+ loop is nil unless +break+ is
+used.
+
+== +for+ Loop
+
+The +for+ loop consists of +for+ followed by a variable to contain the
+iteration argument followed by +in+ and the value to iterate over using #each.
+The +do+ is optional:
+
+  for value in [1, 2, 3] do
+    puts value
+  end
+
+Prints 1, 2 and 3.
+
+Like +while+ and +until+, the +do+ is optional.
+
+The +for+ loop is similar to using #each, but does not create a new variable
+scope.
+
+The result value of a +for+ loop is the value iterated over unless +break+ is
+used.
+
+The +for+ loop is rarely used in modern ruby programs.
+
+== Modifier +while+ and +until+
+
+Like +if+ and +unless+, +while+ and +until+ can be used as modifiers:
+
+  a = 0
+
+  a += 1 while a < 10
+
+  p a # prints 10
+
++until+ used as a modifier:
+
+  a = 0
+
+  a += 1 until a > 10
+
+  p a # prints 11
+
+You can use +begin+ and +end+ to create a +while+ loop that runs the body once
+before the condition:
+
+  a = 0
+
+  begin
+    a += 1
+  end while a < 10
+
+  p a # prints 10
+
+If you don't use +rescue+ or +ensure+, Ruby optimizes away any exception
+handling overhead.
+
+== +break+ Statement
+
+Use +break+ to leave a block early.  This will stop iterating over the items in +values+ if one of them is even:
+
+  values.each do |value|
+    break if value.even?
+
+    # ...
+  end
+
+You can also terminate from a +while+ loop using +break+:
+
+  a = 0
+
+  while true do
+    p a
+    a += 1
+
+    break if a < 10
+  end
+
+  p a
+
+This prints the numbers 0 and 1.
+
++break+ accepts a value that supplies the result of the expression it is
+"breaking" out of:
+
+  result = [1, 2, 3].each do |value|
+    break value * 2 if value.even?
+  end
+
+  p result # prints 4
+
+== +next+ Statement
+
+Use +next+ to skip the rest of the current iteration:
+
+  result = [1, 2, 3].map do |value|
+    next if value.even?
+
+    value * 2
+  end
+
+  p result # prints [2, nil, 6]
+
++next+ accepts an argument that can be used as the result of the current
+block iteration:
+
+  result = [1, 2, 3].map do |value|
+    next value if value.even?
+
+    value * 2
+  end
+
+  p result # prints [2, 2, 6]
+
+== +redo+ Statement
+
+Use +redo+ to redo the current iteration:
+
+  result = []
+
+  while result.length < 10 do
+    result << result.length
+
+    redo if result.last.even?
+
+    result << result.length + 1
+  end
+
+  p result
+
+This prints [0, 1, 3, 3, 5, 5, 7, 7, 9, 9, 11]
+
+In Ruby 1.8, you could also use +retry+ where you used +redo+.  This is no
+longer true, now you will receive a SyntaxError when you use +retry+ outside
+of a +rescue+ block.  See {Exceptions}[rdoc-ref:syntax/exceptions.rdoc]
+for proper usage of +retry+.
+
+== Flip-Flop
+
+The flip-flop is a rarely seen conditional expression.  It's primary use is
+for processing text from ruby one-line programs used with <code>ruby -n</code>
+or <code>ruby -p</code>.
+
+The form of the flip-flop is an expression that indicates when the
+flip-flop turns on, <code>..</code> (or <code>...</code>), then an expression
+that indicates when the flip-flop will turn off.  While the flip-flop is on it
+will continue to evaluate to +true+, and +false+ when off.
+
+Here is an example:
+
+
+  selected = []
+
+  0.upto 10 do |value|
+    selected << value if value==2..value==8
+  end
+
+  p selected # prints [2, 3, 4, 5, 6, 7, 8]
+
+In the above example, the on condition is <code>n==2</code>.  The flip-flop
+is initially off (false) for 0 and 1, but becomes on (true) for 2 and remains
+on through 8.  After 8 it turns off and remains off for 9 and 10.
+
+The flip-flop must be used inside a conditional such as +if+, +while+,
++unless+, +until+ etc. including the modifier forms.
+
+When you use an inclusive range (<code>..</code>), the off condition is
+evaluated when the on condition changes:
+
+  selected = []
+
+  0.upto 5 do |value|
+    selected << value if value==2..value==2
+  end
+
+  p selected # prints [2]
+
+Here, both sides of the flip-flop are evaluated so the flip-flop turns on and
+off only when +value+ equals 2.  Since the flip-flop turned on in the
+iteration it returns true.
+
+When you use an exclusive range (<code>...</code>), the off condition is
+evaluated on the following iteration:
+
+  selected = []
+
+  0.upto 5 do |value|
+    selected << value if value==2...value==2
+  end
+
+  p selected # prints [2, 3, 4, 5]
+
+Here, the flip-flop turns on when +value+ equals 2, but doesn't turn off on the
+same iteration.  The off condition isn't evaluated until the following
+iteration and +value+ will never be two again.
diff --git a/doc/syntax/exceptions.rdoc b/doc/syntax/exceptions.rdoc
new file mode 100644
index 0000000000..a2e75616fb
--- /dev/null
+++ b/doc/syntax/exceptions.rdoc
@@ -0,0 +1,95 @@
+= Exception Handling
+
+Exceptions are rescued in a +begin+/+end+ block:
+
+  begin
+    # code that might raise
+  rescue
+    # handle exception
+  end
+
+If you are inside a method, you do not need to use +begin+ or +end+ unless you
+wish to limit the scope of rescued exceptions:
+
+  def my_method
+    # ...
+  rescue
+    # ...
+  end
+
+The same is true for a +class+ or +module+.
+
+You can assign the exception to a local variable by using <tt>=>
+variable_name</tt> at the end of the +rescue+ line:
+
+  begin
+    # ...
+  rescue => exception
+    warn exception.message
+    raise # re-raise the current exception
+  end
+
+By default, StandardError and its subclasses are rescued.  You can rescue a
+specific set of exception classes (and their subclasses) by listing them after
++rescue+:
+
+  begin
+    # ...
+  rescue ArgumentError, NameError
+    # handle ArgumentError or NameError
+  end
+
+You may rescue different types of exceptions in different ways:
+
+  begin
+    # ...
+  rescue ArgumentError
+    # handle ArgumentError
+  rescue NameError
+    # handle NameError
+  rescue
+    # handle any StandardError
+  end
+
+The exception is matched to the rescue section starting at the top, and matches
+only once.  If an ArgumentError is raised in the begin section, it will not be
+handled in the StandardError section.
+
+You may retry rescued exceptions:
+
+  begin
+    # ...
+  rescue
+    # do something that may change the result of the begin block
+    retry
+  end
+
+Execution will resume at the start of the begin block, so be careful not to
+create an infinite loop.
+
+Inside a rescue block is the only valid location for +retry+, all other uses
+will raise a SyntaxError.  If you wish to retry a block iteration use +redo+.
+See {Control Expressions}[rdoc-ref:syntax/control_expressions.rdoc] for
+details.
+
+To always run some code whether an exception was raised or not, use +ensure+:
+
+  begin
+    # ...
+  rescue
+    # ...
+  ensure
+    # this always runs
+  end
+
+You may also run some code when an exception is not raised:
+
+  begin
+    # ...
+  rescue
+    # ...
+  else
+    # this runs only when no exception was raised
+  ensure
+    # ...
+  end
diff --git a/doc/syntax/literals.rdoc b/doc/syntax/literals.rdoc
new file mode 100644
index 0000000000..08eefd21de
--- /dev/null
+++ b/doc/syntax/literals.rdoc
@@ -0,0 +1,396 @@
+= Literals
+
+Literals create objects you can use in your program.  Literals include:
+
+* Booleans and nil
+* Numbers
+* Strings
+* Symbols
+* Arrays
+* Hashes
+* Ranges
+* Regular Expressions
+* Procs
+
+== Booleans and nil
+
++nil+ and +false+ are both false values.  +nil+ is sometimes used to indicate
+"no value" or "unknown" but evaluates to +false+ in conditional expressions.
+
++true+ is a true value.  All objects except +nil+ and +false+ evaluate to a
+true value in conditional expressions.
+
+(There are also the constants +TRUE+, +FALSE+ and +NIL+, but the lowercase
+literal forms are preferred.)
+
+== Numbers
+
+You can write integers of any size as follows:
+
+  1234
+  1_234
+
+These numbers have the same value, 1,234.  The underscore may be used to
+enhance readability for humans.  You may place an underscore anywhere in the
+number.
+
+Floating point numbers may be written as follows:
+
+  12.34
+  1234e-2
+  1.234E1
+
+These numbers have the same value, 12.34.  You may use underscores in floating
+point numbers as well.
+
+You can use a special prefix to write numbers in decimal, hexadecimal, octal
+or binary formats.  For decimal numbers use a prefix of <tt>0d</tt>, for
+hexadecimal numbers use a prefix of <tt>0x</tt>, for octal numbers use a
+prefix of <tt>0</tt> or <tt>0o</tt>, for binary numbers use a prefix of
+<tt>0b</tt>.  The alphabetic component of the number is not case-sensitive.
+
+Examples:
+
+  0d170
+  0D170
+
+  0xaa
+  0xAa
+  0xAA
+  0Xaa
+  0XAa
+  0XaA
+
+  0252
+  0o252
+  0O252
+
+  0b10101010
+  0B10101010
+
+All these numbers have the same decimal value, 170.  Like integers and floats
+you may use an underscore for readability.
+
+=== Rational numbers
+
+Numbers suffixed by +r+ are Rational numbers.
+
+  12r         #=> (12/1)
+  12.3r       #=> (123/10)
+
+Rational numbers are exact, whereas Float numbers are inexact.
+
+  0.1r + 0.2r #=> (3/10)
+  0.1 + 0.2   #=> 0.30000000000000004
+
+=== Complex numbers
+
+Numbers suffixed by +i+ are Complex (or imaginary) numbers.
+
+  1i          #=> (0+1i)
+  1i * 1i     #=> (-1+0i)
+
+Also Rational numbers may be imaginary numbers.
+
+  12.3ri      #=> (0+(123/10)*i)
+
++i+ must be placed after +r+, the opposite is not allowed.
+
+  12.3ir      #=> syntax error
+
+== Strings
+
+The most common way of writing strings is using <tt>"</tt>:
+
+  "This is a string."
+
+The string may be many lines long.
+
+Any internal <tt>"</tt> must be escaped:
+
+  "This string has a quote: \".  As you can see, it is escaped"
+
+Double-quote strings allow escaped characters such as <tt>\n</tt> for
+newline, <tt>\t</tt> for tab, etc.  The full list of supported escape
+sequences are as follows:
+
+  \a             bell, ASCII 07h (BEL)
+  \b             backspace, ASCII 08h (BS)
+  \t             horizontal tab, ASCII 09h (TAB)
+  \n             newline (line feed), ASCII 0Ah (LF)
+  \v             vertical tab, ASCII 0Bh (VT)
+  \f             form feed, ASCII 0Ch (FF)
+  \r             carriage return, ASCII 0Dh (CR)
+  \e             escape, ASCII 1Bh (ESC)
+  \s             space, ASCII 20h (SPC)
+  \\             backslash, \
+  \nnn           octal bit pattern, where nnn is 1-3 octal digits ([0-7])
+  \xnn           hexadecimal bit pattern, where nn is 1-2 hexadecimal digits ([0-9a-fA-F])
+  \unnnn         Unicode character, where nnnn is exactly 4 hexadecimal digits ([0-9a-fA-F])
+  \u{nnnn ...}   Unicode character(s), where each nnnn is 1-6 hexadecimal digits ([0-9a-fA-F])
+  \cx or \C-x    control character, where x is an ASCII printable character
+  \M-x           meta character, where x is an ASCII printable character
+  \M-\C-x        meta control character, where x is an ASCII printable character
+  \M-\cx         same as above
+  \c\M-x         same as above
+  \c? or \C-?    delete, ASCII 7Fh (DEL)
+
+Any other character following a backslash is interpreted as the
+character itself.
+
+Double-quote strings allow interpolation of other values using
+<tt>#{...}</tt>:
+
+  "One plus one is two: #{1 + 1}"
+
+Any expression may be placed inside the interpolated section, but it's best to
+keep the expression small for readability.
+
+Interpolation may be disabled by escaping the "#" character or using
+single-quote strings:
+
+  '#{1 + 1}' #=> "\#{1 + 1}"
+
+In addition to disabling interpolation, single-quoted strings also disable all
+escape sequences except for the single-quote (<tt>\'</tt>) and backslash
+(<tt>\\\\</tt>).
+
+You may also create strings using <tt>%</tt>:
+
+  %(1 + 1 is #{1 + 1}) #=> "1 + 1 is 2"
+
+There are two different types of <tt>%</tt> strings <tt>%q(...)</tt> behaves
+like a single-quote string (no interpolation or character escaping), while
+<tt>%Q</tt> behaves as a double-quote string.  See Percent Strings below for
+more discussion of the syntax of percent strings.
+
+Adjacent string literals are automatically concatenated by the interpreter:
+
+  "con" "cat" "en" "at" "ion" #=> "concatenation"
+  "This string contains "\
+  "no newlines."              #=> "This string contains no newlines."
+
+Any combination of adjacent single-quote, double-quote, percent strings will
+be concatenated as long as a percent-string is not last.
+
+  %q{a} 'b' "c" #=> "abc"
+  "a" 'b' %q{c} #=> NameError: uninitialized constant q
+
+There is also a character literal notation to represent single
+character strings, which syntax is a question mark (<tt>?</tt>)
+followed by a single character or escape sequence that corresponds to
+a single codepoint in the script encoding:
+
+  ?a       #=> "a"
+  ?abc     #=> SyntaxError
+  ?\n      #=> "\n"
+  ?\s      #=> " "
+  ?\\      #=> "\\"
+  ?\u{41}  #=> "A"
+  ?\C-a    #=> "\x01"
+  ?\M-a    #=> "\xE1"
+  ?\M-\C-a #=> "\x81"
+  ?\C-\M-a #=> "\x81", same as above
+  ?あ      #=> "あ"
+
+=== Here Documents
+
+If you are writing a large block of text you may use a "here document" or
+"heredoc":
+
+  expected_result = <<HEREDOC
+  This would contain specially formatted text.
+
+  That might span many lines
+  HEREDOC
+
+The heredoc starts on the line following <tt><<HEREDOC</tt> and ends with the
+next line that starts with <tt>HEREDOC</tt>.  The result includes the ending
+newline.
+
+You may use any identifier with a heredoc, but all-uppercase identifiers are
+typically used.
+
+You may indent the ending identifier if you place a "-" after <tt><<</tt>:
+
+    expected_result = <<-INDENTED_HEREDOC
+  This would contain specially formatted text.
+
+  That might span many lines
+    INDENTED_HEREDOC
+
+Note that the while the closing identifier may be indented, the content is
+always treated as if it is flush left.  If you indent the content those spaces
+will appear in the output.
+
+To have indented content as well as an indented closing identifier, you can use
+a "squiggly" heredoc, which uses a "~" instead of a "-" after <tt><<</tt>:
+
+    expected_result = <<~SQUIGGLY_HEREDOC
+      This would contain specially formatted text.
+
+      That might span many lines
+    SQUIGGLY_HEREDOC
+
+The indentation of the least-indented line will be removed from each line of
+the content.  Note that empty lines and lines consisting solely of literal tabs
+and spaces will be ignored for the purposes of determining indentation, but
+escaped tabs and spaces are considered non-indentation characters.
+
+A heredoc allows interpolation and escaped characters.  You may disable
+interpolation and escaping by surrounding the opening identifier with single
+quotes:
+
+  expected_result = <<-'EXPECTED'
+  One plus one is #{1 + 1}
+  EXPECTED
+
+  p expected_result # prints: "One plus one is \#{1 + 1}\n"
+
+The identifier may also be surrounded with double quotes (which is the same as
+no quotes) or with backticks.  When surrounded by backticks the HEREDOC
+behaves like Kernel#`:
+
+  puts <<-`HEREDOC`
+  cat #{__FILE__}
+  HEREDOC
+
+To call a method on a heredoc place it after the opening identifier:
+
+  expected_result = <<-EXPECTED.chomp
+  One plus one is #{1 + 1}
+  EXPECTED
+
+You may open multiple heredocs on the same line, but this can be difficult to
+read:
+
+  puts(<<-ONE, <<-TWO)
+  content for heredoc one
+  ONE
+  content for heredoc two
+  TWO
+
+== Symbols
+
+A Symbol represents a name inside the ruby interpreter.  See Symbol for more
+details on what symbols are and when ruby creates them internally.
+
+You may reference a symbol using a colon: <tt>:my_symbol</tt>.
+
+You may also create symbols by interpolation:
+
+  :"my_symbol1"
+  :"my_symbol#{1 + 1}"
+
+Like strings, a single-quote may be used to disable interpolation:
+
+  :'my_symbol#{1 + 1}' #=> :"my_symbol\#{1 + 1}"
+
+When creating a Hash, there is a special syntax for referencing a Symbol as
+well.
+
+== Arrays
+
+An array is created using the objects between <tt>[</tt> and <tt>]</tt>:
+
+  [1, 2, 3]
+
+You may place expressions inside the array:
+
+  [1, 1 + 1, 1 + 2]
+  [1, [1 + 1, [1 + 2]]]
+
+See Array for the methods you may use with an array.
+
+== Hashes
+
+A hash is created using key-value pairs between <tt>{</tt> and <tt>}</tt>:
+
+  { "a" => 1, "b" => 2 }
+
+Both the key and value may be any object.
+
+You can create a hash using symbol keys with the following syntax:
+
+  { a: 1, b: 2 }
+
+This same syntax is used for keyword arguments for a method.
+
+Like Symbol literals, you can quote symbol keys.
+
+  { "a 1": 1, "b #{1 + 1}": 2 }
+
+is equal to
+
+  { :"a 1" => 1, :"b 2" => 2 }
+
+See Hash for the methods you may use with a hash.
+
+== Ranges
+
+A range represents an interval of values.  The range may include or exclude
+its ending value.
+
+  (1..2)  # includes its ending value
+  (1...2) # excludes its ending value
+  (1..)   # endless range, representing infinite sequence from 1 to Infinity
+
+You may create a range of any object.  See the Range documentation for details
+on the methods you need to implement.
+
+== Regular Expressions
+
+A regular expression is created using "/":
+
+  /my regular expression/
+
+The regular expression may be followed by flags which adjust the matching
+behavior of the regular expression.  The "i" flag makes the regular expression
+case-insensitive:
+
+  /my regular expression/i
+
+Interpolation may be used inside regular expressions along with escaped
+characters.  Note that a regular expression may require additional escaped
+characters than a string.
+
+See Regexp for a description of the syntax of regular expressions.
+
+== Procs
+
+A proc can be created with <tt>-></tt>:
+
+  -> { 1 + 1 }
+
+Calling the above proc will give a result of <tt>2</tt>.
+
+You can require arguments for the proc as follows:
+
+  ->(v) { 1 + v }
+
+This proc will add one to its argument.
+
+== Percent Strings
+
+Besides <tt>%(...)</tt> which creates a String, the <tt>%</tt> may create
+other types of object.  As with strings, an uppercase letter allows
+interpolation and escaped characters while a lowercase letter disables them.
+
+These are the types of percent strings in ruby:
+
+<tt>%i</tt> :: Array of Symbols
+<tt>%q</tt> :: String
+<tt>%r</tt> :: Regular Expression
+<tt>%s</tt> :: Symbol
+<tt>%w</tt> :: Array of Strings
+<tt>%x</tt> :: Backtick (capture subshell result)
+
+For the two array forms of percent string, if you wish to include a space in
+one of the array entries you must escape it with a "\\" character:
+
+  %w[one one-hundred\ one]
+  #=> ["one", "one-hundred one"]
+
+If you are using "(", "[", "{", "<" you must close it with ")", "]", "}", ">"
+respectively.  You may use most other non-alphanumeric characters for percent
+string delimiters such as "%", "|", "^", etc.
diff --git a/doc/syntax/methods.rdoc b/doc/syntax/methods.rdoc
new file mode 100644
index 0000000000..a47c1a3cbf
--- /dev/null
+++ b/doc/syntax/methods.rdoc
@@ -0,0 +1,459 @@
+= Methods
+
+Methods implement the functionality of your program.  Here is a simple method
+definition:
+
+  def one_plus_one
+    1 + 1
+  end
+
+A method definition consists of the +def+ keyword, a method name, the body of
+the method, +return+ value and the +end+ keyword.  When called the method will
+execute the body of the method.  This method returns +2+.
+
+This section only covers defining methods.  See also the {syntax documentation
+on calling methods}[rdoc-ref:syntax/calling_methods.rdoc].
+
+== Method Names
+
+Method names may be one of the operators or must start a letter or a character
+with the eight bit set. It may contain letters, numbers, an <code>_</code>
+(underscore or low line) or a character with the eight bit set. The convention
+is to use underscores to separate words in a multiword method name:
+
+  def method_name
+    puts "use underscores to separate words"
+  end
+
+Ruby programs must be written in a US-ASCII-compatible character set such as
+UTF-8, ISO-8859-1 etc. In such character sets if the eight bit is set it
+indicates an extended character. Ruby allows method names and other identifiers
+to contain such characters. Ruby programs cannot contain some characters like
+ASCII NUL (<code>\x00</code>).
+
+The following are examples of valid Ruby methods:
+
+  def hello
+    "hello"
+  end
+
+  def こんにちは
+    puts "means hello in Japanese"
+  end
+
+Typically method names are US-ASCII compatible since the keys to type them
+exist on all keyboards.
+
+Method names may end with a <code>!</code> (bang or exclamation mark), a
+<code>?</code> (question mark), or <code>=</code> (equals sign).
+
+The bang methods (<code>!</code> at the end of the method name) are called and
+executed just like any other method. However, by convention, a method with an
+exclamation point or bang is considered dangerous. In Ruby's core library the
+dangerous method implies that when a method ends with a bang (<code>!</code>),
+it indicates that unlike its non-bang equivalent, permanently modifies its
+receiver. Almost always, the Ruby core library will have a non-bang
+counterpart (method name which does NOT end with <code>!</code>) of every bang
+method (method name which does end with <code>!</code>) that does not modify
+the receiver. This convention is typically true for the Ruby core library but
+may or may not hold true for other Ruby libraries.
+
+Methods that end with a question mark by convention return boolean, but they
+may not always return just +true+ or +false+.  Often, they will return an
+object to indicate a true value (or "truthy" value).
+
+Methods that end with an equals sign indicate an assignment method.  For
+assignment methods, the return value is ignored and the arguments are returned
+instead.
+
+These are method names for the various Ruby operators.  Each of these
+operators accepts only one argument.  Following the operator is the typical
+use or name of the operator.  Creating an alternate meaning for the operator
+may lead to confusion as the user expects plus to add things, minus to
+subtract things, etc.  Additionally, you cannot alter the precedence of the
+operators.
+
+<code>+</code>   :: add
+<code>-</code>   :: subtract
+<code>*</code>   :: multiply
+<code>**</code>  :: power
+<code>/</code>   :: divide
+<code>%</code>   :: modulus division, String#%
+<code>&</code>   :: AND
+<code>^</code>   :: XOR (exclusive OR)
+<code>>></code>  :: right-shift
+<code><<</code>  :: left-shift, append
+<code>==</code>  :: equal
+<code>!=</code>  :: not equal
+<code>===</code> :: case equality.  See Object#===
+<code>=~</code>  :: pattern match.  (Not just for regular expressions)
+<code>!~</code>  :: does not match
+<code><=></code> :: comparison aka spaceship operator.  See Comparable
+<code><</code>   :: less-than
+<code><=</code>  :: less-than or equal
+<code>></code>   :: greater-than
+<code>>=</code>  :: greater-than or equal
+
+To define unary methods minus, plus, tilde and not (<code>!</code>) follow the
+operator with an <code>@</code> as in <code>+@</code> or <code>!@</code>:
+
+  class C
+    def -@
+      puts "you inverted this object"
+    end
+  end
+
+  obj = C.new
+
+  -obj # prints "you inverted this object"
+
+Unary methods accept zero arguments.
+
+Additionally, methods for element reference and assignment may be defined:
+<code>[]</code> and <code>[]=</code> respectively. Both can take one or more
+arguments, and element reference can take none.
+
+  class C
+    def [](a, b)
+      puts a + b
+    end
+
+    def []=(a, b, c)
+      puts a * b + c
+    end
+  end
+
+  obj = C.new
+
+  obj[2, 3]     # prints "5"
+  obj[2, 3] = 4 # prints "10"
+
+== Return Values
+
+By default, a method returns the last expression that was evaluated in the body
+of the method.  In the example above, the last (and only) expression evaluated
+was the simple sum <code>1 + 1</code>.  The +return+ keyword can be used to
+make it explicit that a method returns a value.
+
+  def one_plus_one
+    return 1 + 1
+  end
+
+It can also be used to make a method return before the last expression is
+evaluated.
+
+  def two_plus_two
+    return 2 + 2
+    1 + 1  # this expression is never evaluated
+  end
+
+Note that for assignment methods the return value will be ignored when using
+the assignment syntax.  Instead, the argument will be returned:
+
+  def a=(value)
+    return 1 + value
+  end
+
+  p(self.a = 5) # prints 5
+
+The actual return value will be returned when invoking the method directly:
+
+  p send(:a=, 5) # prints 6
+
+== Scope
+
+The standard syntax to define a method:
+
+  def my_method
+    # ...
+  end
+
+adds the method to a class.  You can define an instance method on a specific
+class with the +class+ keyword:
+
+  class C
+    def my_method
+      # ...
+    end
+  end
+
+A method may be defined on another object.  You may define a "class method" (a
+method that is defined on the class, not an instance of the class) like this:
+
+  class C
+    def self.my_method
+      # ...
+    end
+  end
+
+However, this is simply a special case of a greater syntactical power in Ruby,
+the ability to add methods to any object.  Classes are objects, so adding
+class methods is simply adding methods to the Class object.
+
+The syntax for adding a method to an object is as follows:
+
+  greeting = "Hello"
+
+  def greeting.broaden
+    self + ", world!"
+  end
+
+  greeting.broaden # returns "Hello, world!"
+
++self+ is a keyword referring to the current object under consideration
+by the compiler, which might make the use of +self+ in defining a class
+method above a little clearer.  Indeed, the example of adding a +hello+
+method to the class +String+ can be rewritten thus:
+
+  def String.hello
+    "Hello, world!"
+  end
+
+A method defined like this is called a "singleton method".  +broaden+ will only
+exist on the string instance +greeting+.  Other strings will not have +broaden+.
+
+== Overriding
+
+When Ruby encounters the +def+ keyword, it doesn't consider it an error if the
+method already exists: it simply redefines it.  This is called
+_overriding_.  Rather like extending core classes, this is a potentially
+dangerous ability, and should be used sparingly because it can cause unexpected
+results.  For example, consider this irb session:
+
+  >> "43".to_i
+  => 43
+  >> class String
+  >>   def to_i
+  >>     42
+  >>   end
+  >> end
+  => nil
+  >> "43".to_i
+  => 42
+
+This will effectively sabotage any code which makes use of the method
+<code>String#to_i</code> to parse numbers from strings.
+
+== Arguments
+
+A method may accept arguments.  The argument list follows the method name:
+
+  def add_one(value)
+    value + 1
+  end
+
+When called, the user of the +add_one+ method must provide an argument.  The
+argument is a local variable in the method body.  The method will then add one
+to this argument and return the value.  If given +1+ this method will
+return +2+.
+
+The parentheses around the arguments are optional:
+
+  def add_one value
+    value + 1
+  end
+
+Multiple arguments are separated by a comma:
+
+  def add_values(a, b)
+    a + b
+  end
+
+When called, the arguments must be provided in the exact order.  In other
+words, the arguments are positional.
+
+=== Default Values
+
+Arguments may have default values:
+
+  def add_values(a, b = 1)
+    a + b
+  end
+
+The default value does not need to appear first, but arguments with defaults
+must be grouped together.  This is ok:
+
+  def add_values(a = 1, b = 2, c)
+    a + b + c
+  end
+
+This will raise a SyntaxError:
+
+  def add_values(a = 1, b, c = 1)
+    a + b + c
+  end
+
+=== Array Decomposition
+
+You can decompose (unpack or extract values from) an Array using extra
+parentheses in the arguments:
+
+  def my_method((a, b))
+    p a: a, b: b
+  end
+
+  my_method([1, 2])
+
+This prints:
+
+  {:a=>1, :b=>2}
+
+If the argument has extra elements in the Array they will be ignored:
+
+  def my_method((a, b))
+    p a: a, b: b
+  end
+
+  my_method([1, 2, 3])
+
+This has the same output as above.
+
+You can use a <code>*</code> to collect the remaining arguments.  This splits
+an Array into a first element and the rest:
+
+  def my_method((a, *b))
+    p a: a, b: b
+  end
+
+  my_method([1, 2, 3])
+
+This prints:
+
+  {:a=>1, :b=>[2, 3]}
+
+The argument will be decomposed if it responds to #to_ary.  You should only
+define #to_ary if you can use your object in place of an Array.
+
+Use of the inner parentheses only uses one of the sent arguments.  If the
+argument is not an Array it will be assigned to the first argument in the
+decomposition and the remaining arguments in the decomposition will be +nil+:
+
+  def my_method(a, (b, c), d)
+    p a: a, b: b, c: c, d: d
+  end
+
+  my_method(1, 2, 3)
+
+This prints:
+
+  {:a=>1, :b=>2, :c=>nil, :d=>3}
+
+You can nest decomposition arbitrarily:
+
+  def my_method(((a, b), c))
+    # ...
+  end
+
+=== Array/Hash Argument
+
+Prefixing an argument with <code>*</code> causes any remaining arguments to be
+converted to an Array:
+
+  def gather_arguments(*arguments)
+    p arguments
+  end
+
+  gather_arguments 1, 2, 3 # prints [1, 2, 3]
+
+The array argument must be the last positional argument, it must appear before
+any keyword arguments.
+
+The array argument will capture a Hash as the last entry if a hash was sent by
+the caller after all positional arguments.
+
+  gather_arguments 1, a: 2 # prints [1, {:a=>2}]
+
+However, this only occurs if the method does not declare any keyword arguments.
+
+  def gather_arguments_keyword(*positional, keyword: nil)
+   p positional: positional, keyword: keyword
+  end
+
+  gather_arguments_keyword 1, 2, three: 3
+  #=> raises: unknown keyword: three (ArgumentError)
+
+Also, note that a bare <code>*</code> can be used to ignore arguments:
+
+  def ignore_arguments(*)
+  end
+
+=== Keyword Arguments
+
+Keyword arguments are similar to positional arguments with default values:
+
+  def add_values(first: 1, second: 2)
+    first + second
+  end
+
+Arbitrary keyword arguments will be accepted with <code>**</code>:
+
+  def gather_arguments(first: nil, **rest)
+    p first, rest
+  end
+
+  gather_arguments first: 1, second: 2, third: 3
+  # prints 1 then {:second=>2, :third=>3}
+
+When calling a method with keyword arguments the arguments may appear in any
+order.  If an unknown keyword argument is sent by the caller an ArgumentError
+is raised.
+
+When mixing keyword arguments and positional arguments, all positional
+arguments must appear before any keyword arguments.
+
+== Block Argument
+
+The block argument is indicated by <code>&</code> and must come last:
+
+  def my_method(&my_block)
+    my_block.call(self)
+  end
+
+Most frequently the block argument is used to pass a block to another method:
+
+  def each_item(&block)
+    @items.each(&block)
+  end
+
+If you are only going to call the block and will not otherwise manipulate it
+or send it to another method using <code>yield</code> without an explicit
+block parameter is preferred.  This method is equivalent to the first method
+in this section:
+
+  def my_method
+    yield self
+  end
+
+There is also a performance benefit to using yield over a calling a block
+parameter.  When a block argument is assigned to a variable a Proc object is
+created which holds the block.  When using yield this Proc object is not
+created.
+
+If you only need to use the block sometimes you can use Proc.new to create a
+proc from the block that was passed to your method.  See Proc.new for further
+details.
+
+== Exception Handling
+
+Methods have an implied exception handling block so you do not need to use
++begin+ or +end+ to handle exceptions.  This:
+
+  def my_method
+    begin
+      # code that may raise an exception
+    rescue
+      # handle exception
+    end
+  end
+
+May be written as:
+
+  def my_method
+    # code that may raise an exception
+  rescue
+    # handle exception
+  end
+
+If you wish to rescue an exception for only part of your method, use +begin+ and
++end+.  For more details see the page on {exception
+handling}[rdoc-ref:syntax/exceptions.rdoc].
diff --git a/doc/syntax/miscellaneous.rdoc b/doc/syntax/miscellaneous.rdoc
new file mode 100644
index 0000000000..d5691f8d60
--- /dev/null
+++ b/doc/syntax/miscellaneous.rdoc
@@ -0,0 +1,106 @@
+= Miscellaneous Syntax
+
+== Ending an Expression
+
+Ruby uses a newline as the end of an expression.  When ending a line with an
+operator, open parentheses, comma, etc. the expression will continue.
+
+You can end an expression with a <code>;</code> (semicolon).  Semicolons are
+most frequently used with <code>ruby -e</code>.
+
+== Indentation
+
+Ruby does not require any indentation.  Typically, ruby programs are indented
+two spaces.
+
+If you run ruby with warnings enabled and have an indentation mis-match, you
+will receive a warning.
+
+== +alias+
+
+The +alias+ keyword is most frequently used to alias methods.  When aliasing a
+method, you can use either its name or a symbol:
+
+  alias new_name old_name
+  alias :new_name :old_name
+
+For methods, Module#alias_method can often be used instead of +alias+.
+
+You can also use +alias+ to alias global variables:
+
+  $old = 0
+
+  alias $new $old
+
+  p $new # prints 0
+
+You may use +alias+ in any scope.
+
+== +undef+
+
+The +undef+ keyword prevents the current class from responding to calls to the
+named methods.
+
+  undef my_method
+
+You may use symbols instead of method names:
+
+  undef :my_method
+
+You may undef multiple methods:
+
+  undef method1, method2
+
+You may use +undef+ in any scope.  See also Module#undef_method
+
+== +defined?+
+
++defined?+ is a keyword that returns a string describing its argument:
+
+  p defined?(UNDEFINED_CONSTANT) # prints nil
+  p defined?(RUBY_VERSION)       # prints "constant"
+  p defined?(1 + 1)              # prints "method"
+
+You don't need to use parenthesis with +defined?+, but they are recommended due
+to the {low precedence}[rdoc-ref:syntax/precedence.rdoc] of +defined?+.
+
+For example, if you wish to check if an instance variable exists and that the
+instance variable is zero:
+
+  defined? @instance_variable && @instance_variable.zero?
+
+This returns <code>"expression"</code>, which is not what you want if the
+instance variable is not defined.
+
+  @instance_variable = 1
+  defined?(@instance_variable) && @instance_variable.zero?
+
+Adding parentheses when checking if the instance variable is defined is a
+better check.  This correctly returns +nil+ when the instance variable is not
+defined and +false+ when the instance variable is not zero.
+
+Using the specific reflection methods such as instance_variable_defined? for
+instance variables or const_defined? for constants is less error prone than
+using +defined?+.
+
+== +BEGIN+ and +END+
+
++BEGIN+ defines a block that is run before any other code in the current file.
+It is typically used in one-liners with <code>ruby -e</code>.  Similarly +END+
+defines a block that is run after any other code.
+
++BEGIN+ must appear at top-level and +END+ will issue a warning when you use it
+inside a method.
+
+Here is an example:
+
+  BEGIN {
+    count = 0
+  }
+
+You must use <code>{</code> and <code>}</code> you may not use +do+ and +end+.
+
+Here is an example one-liner that adds numbers from standard input or any files
+in the argument list:
+
+  ruby -ne 'BEGIN { count = 0 }; END { puts count }; count += gets.to_i'
diff --git a/doc/syntax/modules_and_classes.rdoc b/doc/syntax/modules_and_classes.rdoc
new file mode 100644
index 0000000000..dd70d4ac21
--- /dev/null
+++ b/doc/syntax/modules_and_classes.rdoc
@@ -0,0 +1,344 @@
+= Modules
+
+Modules serve two purposes in Ruby, namespacing and mix-in functionality.
+
+A namespace can be used to organize code by package or functionality that
+separates common names from interference by other packages.  For example, the
+IRB namespace provides functionality for irb that prevents a collision
+for the common name "Context".
+
+Mix-in functionality allows sharing common methods across multiple classes or
+modules.  Ruby comes with the Enumerable mix-in module which provides many
+enumeration methods based on the +each+ method and Comparable allows comparison
+of objects based on the <code><=></code> comparison method.
+
+Note that there are many similarities between modules and classes.  Besides the
+ability to mix-in a module, the description of modules below also applies to
+classes.
+
+== Module Definition
+
+A module is created using the +module+ keyword:
+
+  module MyModule
+    # ...
+  end
+
+A module may be reopened any number of times to add, change or remove
+functionality:
+
+  module MyModule
+    def my_method
+    end
+  end
+
+  module MyModule
+    alias my_alias my_method
+  end
+
+  module MyModule
+    remove_method :my_method
+  end
+
+Reopening classes is a very powerful feature of Ruby, but it is best to only
+reopen classes you own.  Reopening classes you do not own may lead to naming
+conflicts or difficult to diagnose bugs.
+
+== Nesting
+
+Modules may be nested:
+
+  module Outer
+    module Inner
+    end
+  end
+
+Many packages create a single outermost module (or class) to provide a
+namespace for their functionality.
+
+You may also define inner modules using <code>::</code> provided the outer
+modules (or classes) are already defined:
+
+  module Outer::Inner::GrandChild
+  end
+
+Note that this will raise a +NameError+ if +Outer+ and
+<code>Outer::Inner</code> are not already defined.
+
+This style has the benefit of allowing the author to reduce the amount
+of indentation.  Instead of 3 levels of indentation only one is necessary.
+However, the scope of constant lookup is different for creating a namespace
+using this syntax instead of the more verbose syntax.
+
+== Scope
+
+=== +self+
+
++self+ refers to the object that defines the current scope.  +self+ will change
+when entering a different method or when defining a new module.
+
+=== Constants
+
+Accessible constants are different depending on the module nesting (which
+syntax was used to define the module).  In the following example
+the constant <code>A::Z</code> is accessible from B as A is part of the
+nesting:
+
+  module A
+    Z = 1
+
+    module B
+      p Module.nesting #=> [A::B, A]
+      p Z #=> 1
+    end
+  end
+
+However, if you use <code>::</code> to define <code>A::B</code> without
+nesting it inside +A+, a NameError exception will be raised because the nesting
+does not include +A+:
+
+  module A
+    Z = 1
+  end
+
+  module A::B
+    p Module.nesting #=> [A::B]
+    p Z #=> raises NameError
+  end
+
+If a constant is defined at the top-level you may preceded it with
+<code>::</code> to reference it:
+
+  Z = 0
+
+  module A
+    Z = 1
+
+    module B
+      p ::Z #=> 0
+    end
+  end
+
+=== Methods
+
+For method definition documentation see the {syntax documentation for
+methods}[rdoc-ref:syntax/methods.rdoc].
+
+Class methods may be called directly.  (This is slightly confusing, but a
+method on a module is often called a "class method" instead of a "module
+method".  See also Module#module_function which can convert an instance method
+into a class method.)
+
+When a class method references a constant, it uses the same rules as referencing
+it outside the method as the scope is the same.
+
+Instance methods defined in a module are only callable when included.  These
+methods have access to the constants defined when they were included through
+the ancestors list:
+
+  module A
+    Z = 1
+
+    def z
+      Z
+    end
+  end
+
+  include A
+
+  p self.class.ancestors #=> [Object, A, Kernel, BasicObject]
+  p z #=> 1
+
+=== Visibility
+
+Ruby has three types of visibility.  The default is +public+.  A public method
+may be called from any other object.
+
+The second visibility is +protected+.  When calling a protected method the
+sender must be a subclass of the receiver or the receiver must be a subclass of
+the sender.  Otherwise a NoMethodError will be raised.
+
+Protected visibility is most frequently used to define <code>==</code> and
+other comparison methods where the author does not wish to expose an object's
+state to any caller and would like to restrict it only to inherited classes.
+
+Here is an example:
+
+  class A
+    def n(other)
+      other.m
+    end
+  end
+
+  class B < A
+    def m
+      1
+    end
+
+    protected :m
+
+  end
+
+  class C < B
+  end
+
+  a = A.new
+  b = B.new
+  c = C.new
+
+  c.n b #=> 1 -- C is a subclass of B
+  b.n b #=> 1 -- m called on defining class
+  a.n b # raises NoMethodError A is not a subclass of B
+
+The third visibility is +private+.  A private method may not be called with a
+receiver, not even +self+.  If a private method is called with a receiver a
+NoMethodError will be raised.
+
+=== +alias+ and +undef+
+
+You may also alias or undefine methods, but these operations are not
+restricted to modules or classes.  See the {miscellaneous syntax
+section}[rdoc-ref:syntax/miscellaneous.rdoc] for documentation.
+
+= Classes
+
+Every class is also a module, but unlike modules a class may not be mixed-in to
+another module (or class).  Like a module, a class can be used as a namespace.
+A class also inherits methods and constants from its superclass.
+
+== Defining a class
+
+Use the +class+ keyword to create a class:
+
+  class MyClass
+    # ...
+  end
+
+If you do not supply a superclass your new class will inherit from Object.  You
+may inherit from a different class using <code><</code> followed by a class
+name:
+
+  class MySubclass < MyClass
+    # ...
+  end
+
+There is a special class BasicObject which is designed as a blank class and
+includes a minimum of built-in methods.  You can use BasicObject to create an
+independent inheritance structure.  See the BasicObject documentation for
+further details.
+
+== Inheritance
+
+Any method defined on a class is callable from its subclass:
+
+  class A
+    Z = 1
+
+    def z
+      Z
+    end
+  end
+
+  class B < A
+  end
+
+  p B.new.z #=> 1
+
+The same is true for constants:
+
+  class A
+    Z = 1
+  end
+
+  class B < A
+    def z
+      Z
+    end
+  end
+
+  p B.new.z #=> 1
+
+You can override the functionality of a superclass method by redefining the
+method:
+
+  class A
+    def m
+      1
+    end
+  end
+
+  class B < A
+    def m
+      2
+    end
+  end
+
+  p B.new.m #=> 2
+
+If you wish to invoke the superclass functionality from a method use +super+:
+
+  class A
+    def m
+      1
+    end
+  end
+
+  class B < A
+    def m
+      2 + super
+    end
+  end
+
+  p B.new.m #=> 3
+
+When used without any arguments +super+ uses the arguments given to the
+subclass method.  To send no arguments to the superclass method use
+<code>super()</code>.  To send specific arguments to the superclass method
+provide them manually like <code>super(2)</code>.
+
++super+ may be called as many times as you like in the subclass method.
+
+= Singleton Classes
+
+The singleton class (also known as the metaclass or eigenclass) of an object is
+a class that holds methods for only that instance.  You can access the
+singleton class of an object using <code>class << object</code> like this:
+
+  class C
+  end
+
+  class << C
+    # self is the singleton class here
+  end
+
+Most frequently you'll see the singleton class accessed like this:
+
+  class C
+    class << self
+      # ...
+    end
+  end
+
+This allows definition of methods and attributes on a class (or module) without
+needing to write <code>def self.my_method</code>.
+
+Since you can open the singleton class of any object this means that this code
+block:
+
+  o = Object.new
+
+  def o.my_method
+    1 + 1
+  end
+
+is equivalent to this code block:
+
+  o = Object.new
+
+  class << o
+    def my_method
+      1 + 1
+    end
+  end
+
+Both objects will have a +my_method+ that returns +2+.
diff --git a/doc/syntax/precedence.rdoc b/doc/syntax/precedence.rdoc
new file mode 100644
index 0000000000..515626c74f
--- /dev/null
+++ b/doc/syntax/precedence.rdoc
@@ -0,0 +1,60 @@
+= Precedence
+
+From highest to lowest, this is the precedence table for ruby.  High precedence
+operations happen before low precedence operations.
+
+  !, ~, unary +
+
+  **
+
+  unary -
+
+  *, /, %
+
+  +, -
+
+  <<, >>
+
+  &
+
+  |, ^
+
+  >, >=, <, <=
+
+  <=>, ==, ===, !=, =~, !~
+
+  &&
+
+  ||
+
+  .., ...
+
+  ?, :
+
+  modifier-rescue
+
+  =, +=, -=, etc.
+
+  defined?
+
+  not
+
+  or, and
+
+  modifier-if, modifier-unless, modifier-while, modifier-until
+
+  { } blocks
+
+Unary <code>+</code> and unary <code>-</code> are for <code>+1</code>,
+<code>-1</code> or <code>-(a + b)</code>.
+
+Modifier-if, modifier-unless, etc. are for the modifier versions of those
+keywords.  For example, this is a modifier-unless expression:
+
+  a += 1 unless a.zero?
+
+<code>{ ... }</code> blocks have priority below all listed operations, but
+<code>do ... end</code> blocks have lower priority.
+
+All other words in the precedence table above are keywords.
+
diff --git a/doc/syntax/refinements.rdoc b/doc/syntax/refinements.rdoc
new file mode 100644
index 0000000000..34ca3a1983
--- /dev/null
+++ b/doc/syntax/refinements.rdoc
@@ -0,0 +1,285 @@
+= Refinements
+
+Due to Ruby's open classes you can redefine or add functionality to existing
+classes.  This is called a "monkey patch".  Unfortunately the scope of such
+changes is global.  All users of the monkey-patched class see the same
+changes.  This can cause unintended side-effects or breakage of programs.
+
+Refinements are designed to reduce the impact of monkey patching on other
+users of the monkey-patched class.  Refinements provide a way to extend a
+class locally.
+
+Here is a basic refinement:
+
+  class C
+    def foo
+      puts "C#foo"
+    end
+  end
+
+  module M
+    refine C do
+      def foo
+        puts "C#foo in M"
+      end
+    end
+  end
+
+First, a class +C+ is defined.  Next a refinement for +C+ is created using
+Module#refine. Refinements can modify both classes and modules.
+
+Module#refine creates an anonymous module that contains the changes or
+refinements to the class (+C+ in the example).  +self+ in the refine block is
+this anonymous module similar to Module#module_eval.
+
+Activate the refinement with #using:
+
+  using M
+
+  c = C.new
+
+  c.foo # prints "C#foo in M"
+
+== Scope
+
+You may activate refinements at top-level, and inside classes and modules.
+You may not activate refinements in method scope.  Refinements are activated
+until the end of the current class or module definition, or until the end of
+the current file if used at the top-level.
+
+You may activate refinements in a string passed to Kernel#eval. Refinements
+are active until the end of the eval string.
+
+Refinements are lexical in scope.  Refinements are only active within a scope
+after the call to +using+. Any code before the +using+ statement will not have the
+refinement activated.
+
+When control is transferred outside the scope, the refinement is deactivated.
+This means that if you require or load a file or call a method that is defined
+outside the current scope the refinement will be deactivated:
+
+  class C
+  end
+
+  module M
+    refine C do
+      def foo
+        puts "C#foo in M"
+      end
+    end
+  end
+
+  def call_foo(x)
+    x.foo
+  end
+
+  using M
+
+  x = C.new
+  x.foo       # prints "C#foo in M"
+  call_foo(x) #=> raises NoMethodError
+
+If a method is defined in a scope where a refinement is active, the refinement
+will be active when the method is called.  This example spans multiple files:
+
+c.rb:
+
+  class C
+  end
+
+m.rb:
+
+  require "c"
+
+  module M
+    refine C do
+      def foo
+        puts "C#foo in M"
+      end
+    end
+  end
+
+m_user.rb:
+
+  require "m"
+
+  using M
+
+  class MUser
+    def call_foo(x)
+      x.foo
+    end
+  end
+
+main.rb:
+
+  require "m_user"
+
+  x = C.new
+  m_user = MUser.new
+  m_user.call_foo(x) # prints "C#foo in M"
+  x.foo              #=> raises NoMethodError
+
+Since the refinement +M+ is active in <code>m_user.rb</code> where
+<code>MUser#call_foo</code> is defined it is also active when
+<code>main.rb</code> calls +call_foo+.
+
+Since #using is a method, refinements are only active when it is called.  Here
+are examples of where a refinement +M+ is and is not active.
+
+In a file:
+
+  # not activated here
+  using M
+  # activated here
+  class Foo
+    # activated here
+    def foo
+      # activated here
+    end
+    # activated here
+  end
+  # activated here
+
+In a class:
+
+  # not activated here
+  class Foo
+    # not activated here
+    def foo
+      # not activated here
+    end
+    using M
+    # activated here
+    def bar
+      # activated here
+    end
+    # activated here
+  end
+  # not activated here
+
+Note that the refinements in +M+ are *not* activated automatically if the class
++Foo+ is reopened later.
+
+In eval:
+
+  # not activated here
+  eval <<EOF
+    # not activated here
+    using M
+    # activated here
+  EOF
+  # not activated here
+
+When not evaluated:
+
+  # not activated here
+  if false
+    using M
+  end
+  # not activated here
+
+When defining multiple refinements in the same module inside multiple +refine+ blocks,
+all refinements from the same module are active when a refined method
+(any of the +to_json+ methods from the example below) is called:
+
+  module ToJSON
+    refine Integer do
+      def to_json
+        to_s
+      end
+    end
+
+    refine Array do
+      def to_json
+        "[" + map { |i| i.to_json }.join(",") + "]"
+      end
+    end
+
+    refine Hash do
+      def to_json
+        "{" + map { |k, v| k.to_s.dump + ":" + v.to_json }.join(",") + "}"
+      end
+    end
+  end
+
+  using ToJSON
+
+  p [{1=>2}, {3=>4}].to_json # prints "[{\"1\":2},{\"3\":4}]"
+
+
+== Method Lookup
+
+When looking up a method for an instance of class +C+ Ruby checks:
+
+* If refinements are active for +C+, in the reverse order they were activated:
+  * The prepended modules from the refinement for +C+
+  * The refinement for +C+
+  * The included modules from the refinement for +C+
+* The prepended modules of +C+
+* +C+
+* The included modules of +C+
+
+If no method was found at any point this repeats with the superclass of +C+.
+
+Note that methods in a subclass have priority over refinements in a
+superclass.  For example, if the method <code>/</code> is defined in a
+refinement for Numeric <code>1 / 2</code> invokes the original Integer#/
+because Integer is a subclass of Numeric and is searched before the refinements
+for the superclass Numeric. Since the method <code>/</code> is also present
+in child +Integer+, the method lookup does not move up to the superclass.
+
+However, if a method +foo+ is defined on Numeric in a refinement, <code>1.foo</code>
+invokes that method since +foo+ does not exist on Integer.
+
+== +super+
+
+When +super+ is invoked method lookup checks:
+
+* The included modules of the current class.  Note that the current class may
+  be a refinement.
+* If the current class is a refinement, the method lookup proceeds as in the
+  Method Lookup section above.
+* If the current class has a direct superclass, the method proceeds as in the
+  Method Lookup section above using the superclass.
+
+Note that +super+ in a method of a refinement invokes the method in the
+refined class even if there is another refinement which has been activated in
+the same context.
+
+== Indirect Method Calls
+
+When using indirect method access such as Kernel#send, Kernel#method or
+Kernel#respond_to? refinements are not honored for the caller context during
+method lookup.
+
+This behavior may be changed in the future.
+
+== Refinement inheritance by Module#include
+
+When a module X is included into a module Y, Y inherits refinements from X.
+
+For example, C inherits refinements from A and B in the following code:
+
+  module A
+    refine X do ... end
+    refine Y do ... end
+  end
+  module B
+    refine Z do ... end
+  end
+  module C
+    include A
+    include B
+  end
+
+  using C
+  # Refinements in A and B are activated here.
+
+Refinements in descendants have higher precedence than those of ancestors.
+
+== Further Reading
+
+See https://bugs.ruby-lang.org/projects/ruby-trunk/wiki/RefinementsSpec for the
+current specification for implementing refinements.  The specification also
+contains more details.
diff --git a/doc/yarvarch.en b/doc/yarvarch.en
new file mode 100644
index 0000000000..7a76e25b7e
--- /dev/null
+++ b/doc/yarvarch.en
@@ -0,0 +1,7 @@
+#title YARV: Yet another RubyVM - Software Architecture
+
+maybe writing.
+
+* YARV instruction set
+
+<%= d %>
diff --git a/doc/yarvarch.ja b/doc/yarvarch.ja
new file mode 100644
index 0000000000..2739ec6b14
--- /dev/null
+++ b/doc/yarvarch.ja
@@ -0,0 +1,454 @@
+#title  YARVアーキテクチャ
+#set author 日本 Ruby の会 ささだこういち
+
+
+- 2005-03-03(Thu) 00:31:12 +0900 いろいろと書き直し
+
+----
+
+* これは？
+
+[[YARV: Yet Another RubyVM|http://www.atdot.net/yarv]] の 設計メモです。
+
+
+YARV は、Ruby プログラムのための次の機能を提供します。
+
+- Compiler
+- VM Generator
+- VM (Virtual Machine)
+- Assembler
+- Dis-Assembler
+- (experimental) JIT Compiler
+- (experimental) AOT Compiler
+
+
+現在の YARV は Ruby インタプリタの拡張ライブラリとして実装しています。こ
+れにより、Ruby インタプリタの必要な機能（パーサ、オブジェクト管理、既存
+の拡張ライブラリ）などがほぼそのまま利用できます。
+
+ただし、いくつかのパッチを Ruby インタプリタに当てなければなりません。
+
+今後は、Ruby 本体のインタプリタ部分（eval.c）を置き換えることを目指して
+開発を継続する予定です。
+
+
+* Compiler (compile.h, compile.c)
+
+コンパイラは、Ruby インタプリタのパーサによって生成された構文木（RNode
+データによる木）を YARV 命令列に変換します。YARV 命令については後述しま
+す。
+
+とくに難しいことはしていませんが、スコープなどの開始時にローカル変数の初
+期化などを行い、あとは構文木を辿り変換していきます。
+
+変換中は Ruby の Array オブジェクトに YARV 命令オブジェクト、およびオペ
+ランドを格納していき、最後に実行できる形に変換します。コンパイラでは、コ
+ンパイル中に生成するメモリ領域の管理が問題になることがありますが、YARV
+の場合、Ruby インタプリタがすべて面倒をみてくれるのでこの部分は非常に楽
+に作ることができました（ガーベージコレクタによって自動的にメモリ管理をし
+てくれるため）。
+
+YARV 命令は、命令を示す識別子、オペランドなど、すべて 1 word （マシンで
+表現できる自然な値。C 言語ではポインタのサイズ。Ruby インタプリタ用語で
+は VALUE のサイズ）で表現されます。そのため、YARV 命令はいわゆる「バイト
+コード」ではありません。そのため、YARV の説明などでは「命令列」という用
+語を使っています。
+
+1 word であるため、メモリの利用効率は多少悪くなりますが、アクセス速度な
+どを考慮すると、本方式が一番いいと考えております。たとえばオペランドをコ
+ンスタントプールに格納し、インデックスのみをオペランドで示すことも可能で
+すが、間接アクセスになってしまうので性能に影響が出るため、却下しました。
+
+
+* VM Generator (rb/insns2vm.rb, insns.def)
+
+rb/insns2vm.rb というスクリプトは、insns.def というファイルを読み込み、
+VM のために必要なファイルを生成します。具体的には、命令を実行する部分を
+生成しますが、ほかにもコンパイルに必要な情報、最適化に必要な情報、やアセ
+ンブラ、逆アセンブラに必要な情報を示すファイルも生成します。
+
+
+** 命令記述
+
+insns.def には、各命令がどのような命令であるかを記述します。具体的には次
+の情報を記述します。
+
+- 命令の名前
+- その命令のカテゴリ、コメント（英語、日本語）
+- オペランドの名前
+- その命令実行前にスタックからポップする値
+- その命令実行後にスタックにプッシュする値
+- その命令のロジック（C 言語で記述）
+
+たとえば、スタックに self をおく putself という命令は次のように記述しま
+す。
+
+#code
+/**
+  @c put
+  @e put self.
+  @j self を置く。
+ */
+DEFINE_INSN
+putself
+()
+()
+(VALUE val)
+{
+  val = GET_SELF();
+}
+#end
+
+この場合、オペランドと、スタックからポップする値は無いことになります。命
+令終了後、self をスタックトップに置きたいわけですが、それは val という、
+スタックにプッシュする値として宣言しておいた変数に代入しておくことで、こ
+れを変換するとスタックトップに置く C プログラムが生成されます。
+
+細かいフォーマットは insns.def の冒頭を参照してください。そんなに難しく
+ないと思います。
+
+insnhelper.h というファイルに、命令ロジックを記述するために必要なマクロ
+が定義されています。また、VM の内部構造に関する定義は vm.h というファイ
+ルにあります。
+
+
+* VM (Virtual Machine, vm.h, vm.c)
+
+VM は、実際にコンパイルした結果生成される YARV 命令列を実行します。まさ
+に、この部分が YARV のキモになり、将来的には eval.c をこの VM で置き換え
+たいと考えています。
+
+現在の Ruby インタプリタで実行できるすべてのことが、この VM で実現できる
+ように作っています（現段階ではまだ完全ではありませんが、そうなるべきです）。
+
+VM は、単純なスタックマシンとして実装しています。スレッドひとつにスタッ
+クひとつを保持します。スタックの領域はヒープから取得するので、柔軟な領域
+設定が可能です。
+
+
+** レジスタ
+
+VM は 5 つの仮想的なレジスタによって制御されます。
+
+- PC (Program Counter)
+- SP (Stack Pointer)
+- CFP (Control Frame Pointer)
+- LFP (Local Frame Pointer)
+- DFP (Dynamic Frame Pointer)
+
+PC は現在実行中の命令列の位置を示します。SP はスタックトップの位置を示し
+ます。CFP、LFP、DFP はそれぞれフレームの情報を示します。詳細は後述します。
+
+
+** スタックフレーム
+
+obsolete (update soon)
+
+
+** フレームデザインについての補足
+
+Lisp の処理系などをかんがえると、わざわざブロックローカルフレームとメソ
+ッドローカルフレームのようなものを用意するのは奇異に見えるかもしれません。
+あるフレームを、入れ子構造にして、ローカル変数のアクセスはその入れ子を外
+側に辿れば必ずたどり着くことができるからです（つまり、lfp は必要ない）。
+
+しかし、Ruby ではいくつか状況が違います。まず、メソッドローカルな情報が
+あること、具体的にはブロックとself（callee からみると receiver）です。こ
+の情報をそれぞれのフレームにもたせるのは無駄です。
+
+また、Ruby2.0 からはブロックローカル変数はなくなります（ブロックローカル
+引数は残るので、構造自体はあまり変わりません）。そのため、メソッドローカ
+ル変数へのアクセスが頻発することが予想されます。
+
+このとき、メソッドローカル変数へのアクセスのたびにフレーム（スコープ）の
+リストをたどるのは無駄であると判断し、明示的にメソッドローカルスコープと
+ブロックフレームを分離し、ブロックフレームからはメソッドローカルフレーム
+が lfpレジスタによって容易にアクセスできるようにしました。
+
+
+** メソッド呼び出しについて
+
+メソッド呼び出しは、YARV 命令列で記述されたメソッドか、C で記述されたメ
+ソッドかによってディスパッチ手法が変わります。
+
+YARV 命令列であった場合、上述したスタックフレームを作成して命令を継続し
+ます。とくに VM の関数を再帰呼び出すすることは行ないません。
+
+C で記述されたメソッドだった場合、単純にその関数を呼び出します（ただし、
+バックトレースを正しく生成するためにメソッド呼び出しの情報を付加してから
+行ないます）。
+
+このため、VM 用スタックを別途用意したものの、プログラムによってはマシン
+スタックを使い切ってしまう可能性があります（C -> Ruby -> C -> ... という
+呼び出しが続いた場合）。これは、現在では避けられない仕様となっています。
+
+
+** 例外
+
+例外は、Java の JVM と同様に例外テーブルを用意することで実現します。例外
+が発生したら、当該フレームを、例外テーブルを検査します。そこで、例外が発
+生したときの PC の値に合致するエントリがあった場合、そのエントリに従って
+動作します。もしエントリが見つからなかった場合、スタックを撒き戻してまた
+同様にそのスコープの例外テーブルを検査します。
+
+また、break、return（ブロック中）、retry なども同様の仕組みで実現します。
+
+*** 例外テーブル
+
+例外テーブルエントリは具体的には次の情報が格納されています。
+
+- 対象とする PC の範囲
+- 対象とする例外の種類
+- もし対象となったときにジャンプする先（種類による）
+- もし対象となったときに起動するブロックの iseq
+
+
+*** rescue
+
+rescue 節はブロックとして実現しています。$! の値を唯一の引数として持ちま
+す。
+
+#code
+begin
+rescue A
+rescue B
+rescue C
+end
+#end
+
+は、次のような Ruby スクリプトに変換されます。
+
+#code
+{|err|
+  case err
+  when A === err
+  when B === err
+  when C === err
+  else
+    raise # yarv の命令では throw
+  end
+}
+#end
+
+
+*** ensure
+
+正常系（例外が発生しなかった場合）と異常系（例外が発生したときなど）の2
+種類の命令列が生成されます。正常系では、ただの連続したコード領域としてコ
+ンパイルされます。また、異常系ではブロックとして実装します。最後は必ず
+throw 命令で締めることになります。
+
+
+*** break, return（ブロック中）、retry
+
+break 文、ブロック中の return 文、retry 文は throw 命令としてコンパイル
+されます。どこまで戻るかは、break をフックする例外テーブルのエントリが判
+断します。
+
+
+** 定数の検索
+
+定数という名前なのに、Ruby ではコンパイル時に決定しません。というか、い
+つまでも再定義可能になっています。
+
+定数アクセスのためのRuby記述は次のようになります。
+
+#code
+Ruby表現:
+expr::ID::...::ID
+#end
+
+これは、yarv命令セットでは次のようになります。
+
+#code
+(expr)
+getconstant ID
+...
+getconstant ID
+#end
+
+
+*** 定数検索パス
+
+もし expr が nil だった場合、定数検索パスに従って定数を検索します。この
+挙動は今後 Ruby 2.0 に向けて変更される場合があります。
+
++ クラス、モジュールの動的ネスト関係（プログラムの字面上）をルートまで辿る
++ 継承関係をルート（Object）まで辿る
+
+このため、クラス、モジュールの動的ネスト関係を保存しなければなりません。
+このために、thread_object には klass_nest_stack というものを用意しました。
+これは、現在のネストの情報を保存します。
+
+メソッド定義時、その現在のネスト情報をメソッド定義時に（dupして）加える
+ことで、そのメソッドの実行時、そのネスト情報を参照することが可能になりま
+す。
+
+トップレベルでは、その情報はないことになります。
+
+クラス/モジュール定義文実行時は、現在の情報そのものを参照することになり
+ます。これは、クラススコープ突入時、その情報をクラス定義文にコピーします
+（すでにコピーされていれば、これを行いません）。
+
+これにより、動的なネスト情報を統一的に扱うことができます。
+
+
+** 最適化手法
+
+YARV では高速化を目的としているので、さまざまな最適化手法を利用していま
+す。詳細は割愛しますが、以下に述べる最適化などを行なっております。
+
+
+*** threaded code
+
+GCC の C 言語拡張である値としてのラベルを利用して direct threaded code
+を実現しています。
+
+
+*** Peephole optimization
+
+いくつかの簡単な最適化をしています。
+
+
+*** inline method cache
+
+命令列の中にメソッド検索結果を埋め込みます。
+
+
+*** inline constant cache
+
+命令列の中に定数検索結果を埋め込みます。
+
+
+*** ブロックと Proc オブジェクトの分離
+
+ブロック付きメソッド呼び出しが行なわれたときにはすぐにはブロックを Proc
+オブジェクトとして生成しません。これにより、必要ない Proc オブジェクトの
+生成を抑えています。
+
+Proc メソッドは、実際に必要になった時点で作られ、そのときに環境（スコー
+プ上に確保された変数など）をヒープに保存します。
+
+
+*** 特化命令
+
+Fixnum 同士の加算などを正直に関数呼び出しによって行なうと、コストがかか
+るので、これらのプリミティブな操作を行なうためのメソッド呼び出しは専用命
+令を用意しました。
+
+
+*** 命令融合
+
+複数の命令を 1 命令に変換します。融合命令は opt_insn_unif.def の記述によ
+り自動的に生成されます。
+
+
+*** オペランド融合
+
+複数のオペランドを含めた命令を生成します。融合命令は opt_operand.def の
+記述によって自動的に生成されます。
+
+
+*** stack caching
+
+スタックトップを仮想レジスタに保持するようにします。現在は 2 個の仮想レ
+ジスタを想定し、5状態のスタックキャッシングを行ないます。スタックキャッ
+シングする命令は自動的に生成されます。
+
+
+*** JIT Compile
+
+機械語を切り貼りします。非常に実験的なコードものしか作っておりません。ほ
+とんどのプログラムは動きません。
+
+
+*** AOT Compile
+
+YARV 命令列を C 言語に変換します。まだ十分な最適化を行なえておりませんが、
+それなりに動きます。rb/aotc.rb がコンパイラです。
+
+
+* Assembler (rb/yasm.rb)
+
+YARV 命令列のアセンブラを用意しました。使い方は rb/yasm.rb を参照してく
+ださい（まだ、例示してある生成手法のすべてをサポートしているわけではあり
+ません）。
+
+
+* Dis-Assembler (disasm.c)
+
+YARV 命令列を示すオブジェクト YARVCore::InstructionSequence には disasm
+メソッドがあります。これは、命令列を逆アセンブルした文字列を返します。
+
+
+* YARV 命令セット
+
+<%= d %>
+
+* その他
+
+** テスト
+
+test/test_* がテストケースです。一応、ミスなく動くはずです。逆にいうと、
+このテストに記述されている例ではきちんと動作するということです。
+
+
+** ベンチマーク
+
+benchmark/bm_* にベンチマークプログラムがおいてあります。
+
+
+** 今後の予定
+
+まだまだやらなければいけないこと、未実装部分がたくさんありますんでやって
+いかなければなりません。一番大きな目標は eval.c を置き換えることでしょう
+か。
+
+
+*** Verifier
+
+YARV 命令列は、ミスがあっても動かしてしまうため危険である可能性がありま
+す。そのため、スタックの利用状態をきちんと事前に検証するようなベリファイ
+アを用意しなければならないと考えています。
+
+
+*** Compiled File の構想
+
+Ruby プログラムをこの命令セットにシリアライズしたデータ構造をファイルに
+出力できるようにしたいと考えています。これを利用して一度コンパイルした命
+令列をファイルに保存しておけば、次回ロード時にはコンパイルの手間、コスト
+を省くことができます。
+
+
+**** 全体構成
+
+次のようなファイル構成を考えていますが、まだ未定です。
+
+#code
+u4 : 4 byte unsigned storage
+u2 : 2 byte unsigned storage
+u1 : 1 byte unsigned storage
+
+every storages are little endian :-)
+
+CompiledFile{
+  u4 magic;
+
+  u2 major;
+  u2 minor;
+
+  u4 character_code;
+
+  u4 constants_pool_count;
+  ConstantEntry constants_pool[constants_pool_count];
+
+  u4 block_count;
+  blockEntry blocks[block_count];
+
+  u4 method_count;
+  MethodEntry methods[method_count];
+}
+#end
+
+Java classfile のパクリ。
+
diff --git a/enc/Makefile.in b/enc/Makefile.in
new file mode 100644
index 0000000000..8385236494
--- /dev/null
+++ b/enc/Makefile.in
@@ -0,0 +1,84 @@
+V = 0
+Q1 = $(V:1=)
+Q = $(Q1:0=@)
+ECHO1 = $(V:1=@@NULLCMD@)
+ECHO = $(ECHO1:0=@echo)
+
+encsrcdir = @srcdir@
+topdir = .
+prefix = @prefix@
+exec_prefix = @exec_prefix@
+libdir = @libdir@
+top_srcdir = $(encsrcdir:/enc=)
+srcdir = $(top_srcdir)
+arch = @arch@
+EXTOUT = @EXTOUT@
+hdrdir = $(srcdir)/include
+arch_hdrdir = $(EXTOUT)/include/$(arch)
+ENCSODIR = $(EXTOUT)/$(arch)/enc
+TRANSSODIR = $(ENCSODIR)/trans
+DLEXT = @DLEXT@
+OBJEXT = @OBJEXT@
+LIBEXT = @LIBEXT@
+TIMESTAMPDIR  = $(EXTOUT)/.timestamp
+ENC_TRANS_D   = $(TIMESTAMPDIR)/.enc-trans.time
+ENC_TRANS_SO_D = $(TIMESTAMPDIR)/.enc-trans.so.time
+
+BUILTIN_ENCS	= enc/ascii.c enc/us_ascii.c\
+		  enc/unicode.c enc/utf_8.c
+
+BUILTIN_TRANSES	= enc/trans/newline.trans
+
+RUBY_SO_NAME = @RUBY_SO_NAME@
+LIBRUBY = @LIBRUBY@
+LIBRUBYARG_SHARED = @LIBRUBYARG_SHARED@
+LIBRUBYARG_STATIC = $(LIBRUBYARG_SHARED)
+
+empty =
+AR = @AR@
+CC = @CC@
+ARFLAGS = @ARFLAGS@$(empty)
+RANLIB = @RANLIB@
+OUTFLAG = @OUTFLAG@$(empty)
+COUTFLAG = @COUTFLAG@$(empty)
+CFLAGS = $(CCDLFLAGS) @CFLAGS@ @ARCH_FLAG@
+cflags = @cflags@
+optflags = @optflags@
+debugflags = @debugflags@
+warnflags = @warnflags@
+CCDLFLAGS = @CCDLFLAGS@
+INCFLAGS = -I. -I$(arch_hdrdir) -I$(hdrdir) -I$(top_srcdir)
+DEFS = @DEFS@
+CPPFLAGS = @CPPFLAGS@ -DONIG_ENC_REGISTER=rb_enc_register
+LDFLAGS = @LDFLAGS@
+LDSHARED = @LDSHARED@
+ldflags  = $(LDFLAGS)
+dldflags = @DLDFLAGS@
+extdldflags = @EXTDLDFLAGS@
+archflag = @ARCH_FLAG@
+DLDFLAGS = $(ldflags) $(dldflags) $(extdldflags) $(archflag)
+RUBY     = $(MINIRUBY)
+TARGET_NAME_ = $(@F) # BSD make seems unable to deal with @F with substitution
+TARGET_NAME = $(TARGET_NAME_:.@DLEXT@=)
+TARGET_ENTRY = @EXPORT_PREFIX@Init_$(TARGET_NAME)
+
+WORKDIRS = @WORKDIRS@
+
+NULLCMD = @NULLCMD@
+RM = @RM@
+RMDIR = @RMDIR@
+RMDIRS = @RMDIRS@
+MAKEDIRS = @MAKEDIRS@
+
+.SUFFIXES: .trans
+
+all:
+
+make-workdir:
+	$(Q)$(MAKEDIRS) $(WORKDIRS)
+
+clean:
+
+distclean: clean
+	$(Q)$(RM) enc.mk
+realclean: distclean clean-srcs
diff --git a/enc/ascii.c b/enc/ascii.c
new file mode 100644
index 0000000000..8b32c414fe
--- /dev/null
+++ b/enc/ascii.c
@@ -0,0 +1,104 @@
+/**********************************************************************
+  ascii.c -  Onigmo (Oniguruma-mod) (regular expression library)
+**********************************************************************/
+/*-
+ * Copyright (c) 2002-2006  K.Kosako  <sndgk393 AT ybb DOT ne DOT jp>
+ * Copyright (c) 2011       K.Takata  <kentkt AT csc DOT jp>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include "regenc.h"
+#ifdef RUBY
+# include "encindex.h"
+#endif
+
+#ifndef ENCINDEX_ASCII
+# define ENCINDEX_ASCII 0
+#endif
+
+OnigEncodingDefine(ascii, ASCII) = {
+  onigenc_single_byte_mbc_enc_len,
+  "ASCII-8BIT",/* name */
+  1,           /* max byte length */
+  1,           /* min byte length */
+  onigenc_is_mbc_newline_0x0a,
+  onigenc_single_byte_mbc_to_code,
+  onigenc_single_byte_code_to_mbclen,
+  onigenc_single_byte_code_to_mbc,
+  onigenc_ascii_mbc_case_fold,
+  onigenc_ascii_apply_all_case_fold,
+  onigenc_ascii_get_case_fold_codes_by_str,
+  onigenc_minimum_property_name_to_ctype,
+  onigenc_ascii_is_code_ctype,
+  onigenc_not_support_get_ctype_code_range,
+  onigenc_single_byte_left_adjust_char_head,
+  onigenc_always_true_is_allowed_reverse_match,
+  onigenc_single_byte_ascii_only_case_map,
+  ENCINDEX_ASCII,
+  ONIGENC_FLAG_NONE,
+};
+ENC_ALIAS("BINARY", "ASCII-8BIT")
+ENC_REPLICATE("IBM437", "ASCII-8BIT")
+ENC_ALIAS("CP437", "IBM437")
+ENC_REPLICATE("IBM737", "ASCII-8BIT")
+ENC_ALIAS("CP737", "IBM737")
+ENC_REPLICATE("IBM775", "ASCII-8BIT")
+ENC_ALIAS("CP775", "IBM775")
+ENC_REPLICATE("CP850", "ASCII-8BIT")
+ENC_ALIAS("IBM850", "CP850")
+ENC_REPLICATE("IBM852", "ASCII-8BIT")
+ENC_REPLICATE("CP852", "IBM852")
+ENC_REPLICATE("IBM855", "ASCII-8BIT")
+ENC_REPLICATE("CP855", "IBM855")
+ENC_REPLICATE("IBM857", "ASCII-8BIT")
+ENC_ALIAS("CP857", "IBM857")
+ENC_REPLICATE("IBM860", "ASCII-8BIT")
+ENC_ALIAS("CP860", "IBM860")
+ENC_REPLICATE("IBM861", "ASCII-8BIT")
+ENC_ALIAS("CP861", "IBM861")
+ENC_REPLICATE("IBM862", "ASCII-8BIT")
+ENC_ALIAS("CP862", "IBM862")
+ENC_REPLICATE("IBM863", "ASCII-8BIT")
+ENC_ALIAS("CP863", "IBM863")
+ENC_REPLICATE("IBM864", "ASCII-8BIT")
+ENC_ALIAS("CP864", "IBM864")
+ENC_REPLICATE("IBM865", "ASCII-8BIT")
+ENC_ALIAS("CP865", "IBM865")
+ENC_REPLICATE("IBM866", "ASCII-8BIT")
+ENC_ALIAS("CP866", "IBM866")
+ENC_REPLICATE("IBM869", "ASCII-8BIT")
+ENC_ALIAS("CP869", "IBM869")
+ENC_REPLICATE("Windows-1258", "ASCII-8BIT")
+ENC_ALIAS("CP1258", "Windows-1258")
+ENC_REPLICATE("GB1988", "ASCII-8BIT")
+ENC_REPLICATE("macCentEuro", "ASCII-8BIT")
+ENC_REPLICATE("macCroatian", "ASCII-8BIT")
+ENC_REPLICATE("macCyrillic", "ASCII-8BIT")
+ENC_REPLICATE("macGreek", "ASCII-8BIT")
+ENC_REPLICATE("macIceland", "ASCII-8BIT")
+ENC_REPLICATE("macRoman", "ASCII-8BIT")
+ENC_REPLICATE("macRomania", "ASCII-8BIT")
+ENC_REPLICATE("macThai", "ASCII-8BIT")
+ENC_REPLICATE("macTurkish", "ASCII-8BIT")
+ENC_REPLICATE("macUkraine", "ASCII-8BIT")
diff --git a/enc/big5.c b/enc/big5.c
new file mode 100644
index 0000000000..ab4fb69819
--- /dev/null
+++ b/enc/big5.c
@@ -0,0 +1,376 @@
+/**********************************************************************
+  big5.c -  Oniguruma (regular expression library)
+**********************************************************************/
+/*-
+ * Copyright (c) 2002-2007  K.Kosako  <sndgk393 AT ybb DOT ne DOT jp>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include "regenc.h"
+
+static const int EncLen_BIG5[] = {
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+  2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+  2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+  2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+  2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+  2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1
+};
+static const int EncLen_BIG5_HKSCS[] = {
+ /* LEN  0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F */
+ /* 0 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 1 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 2 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 3 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 4 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 5 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 6 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 7 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 8 */ 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* 9 */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* A */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* B */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* C */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* D */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* E */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* F */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1,
+};
+
+static const int EncLen_BIG5_UAO[] = {
+ /* LEN  0  1  2  3  4  5  6  7  8  9  A  B  C  D  E  F */
+ /* 0 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 1 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 2 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 3 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 4 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 5 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 6 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 7 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+ /* 8 */ 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* 9 */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* A */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* B */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* C */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* D */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* E */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
+ /* F */ 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 1,
+};
+
+typedef enum { FAILURE = -2, ACCEPT = -1, S0 = 0, S1 } state_t;
+#define A ACCEPT
+#define F FAILURE
+static const signed char trans[][0x100] = {
+  { /* S0   0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f */
+    /* 0 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 1 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 2 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 3 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 4 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 5 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 6 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 7 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 8 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* 9 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* a */ F, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* b */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* c */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* d */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* e */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* f */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, F
+  },
+  { /* S1   0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f */
+    /* 0 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* 1 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* 2 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* 3 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* 4 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 5 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 6 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 7 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, F,
+    /* 8 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* 9 */ F, F, F, F, F, F, F, F, F, F, F, F, F, F, F, F,
+    /* a */ F, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* b */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* c */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* d */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* e */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* f */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, F
+  },
+  { /* S2   0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f */
+    /* 0 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 1 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 2 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 3 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 4 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 5 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 6 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 7 */ A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A,
+    /* 8 */ F, F, F, F, F, F, F, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* 9 */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* a */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* b */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* c */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* d */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* e */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+    /* f */ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, F
+  }
+};
+#undef A
+#undef F
+
+static int
+big5_mbc_enc_len0(const UChar* p, const UChar* e, int tridx, const int tbl[])
+{
+  int firstbyte = *p++;
+  state_t s = trans[tridx][firstbyte];
+#define RETURN(n) \
+    return s == ACCEPT ? ONIGENC_CONSTRUCT_MBCLEN_CHARFOUND(n) : \
+                         ONIGENC_CONSTRUCT_MBCLEN_INVALID()
+  if (s < 0) RETURN(1);
+  if (p == e) return ONIGENC_CONSTRUCT_MBCLEN_NEEDMORE(tbl[firstbyte]-1);
+  s = trans[s][*p++];
+  RETURN(2);
+#undef RETURN
+}
+
+static int
+big5_mbc_enc_len(const UChar* p, const UChar* e, OnigEncoding enc ARG_UNUSED)
+{
+  return big5_mbc_enc_len0(p, e, 0, EncLen_BIG5);
+}
+
+static int
+big5_hkscs_mbc_enc_len(const UChar* p, const UChar* e, OnigEncoding enc ARG_UNUSED)
+{
+  return big5_mbc_enc_len0(p, e, 2, EncLen_BIG5_HKSCS);
+}
+
+static int
+big5_uao_mbc_enc_len(const UChar* p, const UChar* e, OnigEncoding enc ARG_UNUSED)
+{
+  return big5_mbc_enc_len0(p, e, 2, EncLen_BIG5_UAO);
+}
+
+static OnigCodePoint
+big5_mbc_to_code(const UChar* p, const UChar* end, OnigEncoding enc)
+{
+  return onigenc_mbn_mbc_to_code(enc, p, end);
+}
+
+static int
+big5_code_to_mbc(OnigCodePoint code, UChar *buf, OnigEncoding enc)
+{
+  return onigenc_mb2_code_to_mbc(enc, code, buf);
+}
+
+static int
+big5_mbc_case_fold(OnigCaseFoldType flag, const UChar** pp, const UChar* end,
+                   UChar* lower, OnigEncoding enc)
+{
+  return onigenc_mbn_mbc_case_fold(enc, flag,
+                                   pp, end, lower);
+}
+
+#if 0
+static int
+big5_is_mbc_ambiguous(OnigCaseFoldType flag,
+		      const UChar** pp, const UChar* end, OnigEncoding enc)
+{
+  return onigenc_mbn_is_mbc_ambiguous(enc, flag, pp, end);
+}
+#endif
+
+static int
+big5_is_code_ctype(OnigCodePoint code, unsigned int ctype, OnigEncoding enc)
+{
+  return onigenc_mb2_is_code_ctype(enc, code, ctype);
+}
+
+static const char BIG5_CAN_BE_TRAIL_TABLE[256] = {
+  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0
+};
+
+#define BIG5_HKSCS_P(enc) ((enc)->precise_mbc_enc_len == big5_hkscs_mbc_enc_len)
+#define BIG5_UAO_P(enc) ((enc)->precise_mbc_enc_len == big5_uao_mbc_enc_len)
+
+#define BIG5_ISMB_FIRST(byte)  ( \
+	BIG5_HKSCS_P(enc) ? EncLen_BIG5_HKSCS[byte] > 1 : \
+	EncLen_BIG5[byte] > 1 \
+	)
+#define BIG5_ISMB_TRAIL(byte)  BIG5_CAN_BE_TRAIL_TABLE[(byte)]
+
+static UChar*
+big5_left_adjust_char_head(const UChar* start, const UChar* s, const UChar* end, OnigEncoding enc)
+{
+  const UChar *p;
+  int len;
+
+  if (s <= start) return (UChar* )s;
+  p = s;
+
+  if (BIG5_ISMB_TRAIL(*p)) {
+    while (p > start) {
+      if (! BIG5_ISMB_FIRST(*--p)) {
+	p++;
+	break;
+      }
+    }
+  }
+  len = enclen(enc, p, end);
+  if (p + len > s) return (UChar* )p;
+  p += len;
+  return (UChar* )(p + ((s - p) & ~1));
+}
+
+static int
+big5_is_allowed_reverse_match(const UChar* s, const UChar* end ARG_UNUSED, OnigEncoding enc ARG_UNUSED)
+{
+  const UChar c = *s;
+
+  return (BIG5_ISMB_TRAIL(c) ? FALSE : TRUE);
+}
+
+/*
+ * Name: Big5  (preferred MIME name)
+ * MIBenum: 2026
+ * Source: Chinese for Taiwan Multi-byte set.
+ *         PCL Symbol Set Id: 18T
+ * Alias: csBig5
+ */
+OnigEncodingDefine(big5, BIG5) = {
+  big5_mbc_enc_len,
+  "Big5",     /* name */
+  2,          /* max enc length */
+  1,          /* min enc length */
+  onigenc_is_mbc_newline_0x0a,
+  big5_mbc_to_code,
+  onigenc_mb2_code_to_mbclen,
+  big5_code_to_mbc,
+  big5_mbc_case_fold,
+  onigenc_ascii_apply_all_case_fold,
+  onigenc_ascii_get_case_fold_codes_by_str,
+  onigenc_minimum_property_name_to_ctype,
+  big5_is_code_ctype,
+  onigenc_not_support_get_ctype_code_range,
+  big5_left_adjust_char_head,
+  big5_is_allowed_reverse_match,
+  onigenc_ascii_only_case_map,
+  0,
+  ONIGENC_FLAG_NONE,
+};
+
+/*
+ * Name: CP950
+ * Source: http://msdn.microsoft.com/en-us/goglobal/cc305155.aspx
+ */
+ENC_REPLICATE("CP950", "Big5")
+
+/*
+ * Name: Big5-HKSCS
+ * MIBenum: 2101
+ * Source: http://www.iana.org/assignments/charset-reg/Big5-HKSCS
+ * Source: http://www.ogcio.gov.hk/ccli/eng/hkscs/mapping_table_2008.html
+ * Alias: None
+ */
+OnigEncodingDefine(big5_hkscs, BIG5_HKSCS) = {
+  big5_hkscs_mbc_enc_len,
+  "Big5-HKSCS",     /* name */
+  2,          /* max enc length */
+  1,          /* min enc length */
+  onigenc_is_mbc_newline_0x0a,
+  big5_mbc_to_code,
+  onigenc_mb2_code_to_mbclen,
+  big5_code_to_mbc,
+  big5_mbc_case_fold,
+  onigenc_ascii_apply_all_case_fold,
+  onigenc_ascii_get_case_fold_codes_by_str,
+  onigenc_minimum_property_name_to_ctype,
+  big5_is_code_ctype,
+  onigenc_not_support_get_ctype_code_range,
+  big5_left_adjust_char_head,
+  big5_is_allowed_reverse_match,
+  onigenc_ascii_only_case_map,
+  0,
+  ONIGENC_FLAG_NONE,
+};
+ENC_ALIAS("Big5-HKSCS:2008", "Big5-HKSCS")
+
+/*
+ * Name: CP951
+ * Source: http://www.microsoft.com/hk/hkscs/default.aspx
+ * Source: http://www.microsoft.com/downloads/en/details.aspx?FamilyID=0e6f5ac8-7baa-4571-b8e8-78b3b776afd7&DisplayLang=en
+ * Source: http://blogs.msdn.com/b/shawnste/archive/2007/03/12/cp-951-hkscs.aspx
+ */
+ENC_REPLICATE("CP951", "Big5-HKSCS")
+
+/*
+ * Name: Big5-UAO [NOT registered by IANA!]
+ * Source: http://moztw.org/docs/big5/table/big5_2003-b2u.txt
+ */
+OnigEncodingDefine(big5_uao, BIG5_UAO) = {
+  big5_uao_mbc_enc_len,
+  "Big5-UAO",     /* name */
+  2,          /* max enc length */
+  1,          /* min enc length */
+  onigenc_is_mbc_newline_0x0a,
+  big5_mbc_to_code,
+  onigenc_mb2_code_to_mbclen,
+  big5_code_to_mbc,
+  big5_mbc_case_fold,
+  onigenc_ascii_apply_all_case_fold,
+  onigenc_ascii_get_case_fold_codes_by_str,
+  onigenc_minimum_property_name_to_ctype,
+  big5_is_code_ctype,
+  onigenc_not_support_get_ctype_code_range,
+  big5_left_adjust_char_head,
+  big5_is_allowed_reverse_match,
+  onigenc_ascii_only_case_map,
+  0,
+  ONIGENC_FLAG_NONE,
+};
diff --git a/enc/cp949.c b/enc/cp949.c
new file mode 100644